PROCEEDINGS OF A SYMPOSIUM,VIENNA, 29 MARCH -1 APRIL 1971

INTERNATIONAL ATOM IC ENERGY AGENCY, VIENNA, 1971

RADIATION AND RADIOISOTOPES FOR INDUSTRIAL MICROORGANISMS The following States are Members of the International Atomic Energy Agency:

AFGHANISTAN GREECE PAKISTAN ALBANIA GUATEMALA PANAMA ALGERIA HAITI PARAGUAY ARGENTINA HOLY SEE PERU AUSTRALIA HUNGARY PHILIPPINES AUSTRIA ICELAND POLAND BELGIUM INDIA PORTUGAL BOLIVIA INDONESIA ROMANIA BRAZIL IRAN SAUDI ARABIA BULGARIA IRAQ SENEGAL BURMA IRELAND SIERRA LEONE BYELORUSSIAN SOVIET ISRAEL SINGAPORE SOCIALIST REPUBLIC ITALY SOUTH AFRICA CAMEROON IVORY COAST SPAIN CANADA JAMAICA SUDAN CEYLON JAPAN SWEDEN CHILE JORDAN SWITZERLAND CHINA KENYA SYRIAN ARAB REPUBLIC COLOMBIA KHMER REPUBLIC THAILAND CONGO, DEMOCRATIC KOREA, REPUBLIC OF TUNISIA REPUBLIC OF KUWAIT TURKEY COSTA RICA LEBANON UGANDA CUBA LIBERIA UKRAINIAN SOVIET SOCIALIST CYPRUS LIBYAN ARAB REPUBLIC REPUBLIC CZECHOSLOVAK SOCIALIST LIECHTENSTEIN UNION OF SOVIET SOCIALIST REPUBLIC LUXEMBOURG REPUBLICS DENMARK MADAGASCAR UNITED ARAB REPUBLIC DOMINICAN REPUBLIC MALAYSIA UNITED KINGDOM OF GREAT ECUADOR MALI BRITAIN AND NORTHERN EL SALVADOR MEXICO IRELAND ETHIOPIA MONACO UNITED STATES OF AMERICA FINLAND MOROCCO URUGUAY FRANCE NETHERLANDS VENEZUELA GABON NEW ZEALAND VIET-NAM GERMANY, FEDERAL REPUBLIC OF NIGER YUGOSLAVIA GHANA NIGERIA ZAMBIA NORWAY

The Agency's Statute was approved on 23 October 1956 by the Conference on the Statute of the IAEA held at United Nations Headquarters, New York; it entered into force on 29 July 1957. The Headquarters of the Agency are situated in Vienna. Its principal objective is "to accelerate and enlarge the contribution of atomic energy to peace, health and prosperity throughout the world".

Printed by the IAEA, in Austria August 1971 PROCEEDINGS SERIES

RADIATION AND RADIOISOTOPES FOR INDUSTRIAL MICROORGANISMS

PROCEEDINGS OF A SYMPOSIUM ON USE OF RADIATION AND RADIOISOTOPES FOR GENETIC IMPROVEMENT OF INDUSTRIAL MICROORGANISMS HELD BY THE INTERNATIONAL ATOMIC ENERGY AGENCY IN VIENNA, 29 MARCH - 1 APRIL 1971

INTERNATIONAL ATOMIC ENERGY AGENCY VIENNA, 1971 COVER

The right-hand photograph shows a strain of Aspergillus nidulans carrying a duplicate chromosome segment. After deletions from one or other of these segments, sectors emerge. Loss of the dominant allele for green conidia gives a sector with yellow conidia. The left-hand photograph shows the same strain, grown on medium with 0.1% caffeine,which inhibits repair of spontaneous errors arising during chromosome replication.

For these photographs acknowledgment is due to Professor J.A . Roper of the Department of Genetics, University of Sheffield, Sheffield, United Kingdom.

RADIATION AND RADIOISOTOPES FOR INDUSTRIAL MICROORGANISMS IAEA, VIENNA, 1971 ST I/PU B / 287 FOREWORD

M an1 s use of microbial fermentation goes back to the beginnings of civilization, as exemplified by the ancient arts of brewing, of producing wine and spirits, and of bread-making. Duringthe last forty years, as a striking departure from the inefficient traditional fermentation procedures, man has recognized the great poten­ tial of the specific metabolic processes of certain microorganisms. This recognition has led to new developments in the microbial fermentation in­ dustries, which today produce many organic substances that have nutri­ tional, medicinal and other applications for human welfare. These advances, if properly utilized, can bring even greater benefits for the welfare and economy of the developing and developed nations of the world. The recent advances in microbiology, microbial genetics, biochemistry, molecular biology and related disciplines have elucidated the genetic basis of the biosynthetic pathways and their regulatory mechanisms and have thus provided a means of improving the efficiency of many fermentation pro­ cesses, resulting in a quantitative increase in the yields of products and in the synthesis of new products. In this work the application of ionizing radiations and radioisotopes has made a significant contribution, as is evident from the present volume, which contains the papers and discussions from the Symposium on the Use of Radiation and Radioisotopes for Genetic Improvement of Industrial Micro­ organisms, held by the IAEA in Vienna from 29 March till 1 April 1971. The symposium was attended by about 100 participants from 28 countries and five international organizations. The papers review the results of research on microbial mutagenesis and recombination, physiology, biochemistry and allied topics in relation to microbial fermentation. In addition, the book includes a brief review of the current status and future outlook of applied microbiology in certain developing countries; this may be of special interest to those responsible for international programs concerned with microbiology. EDITORIAL NOTE

The papers and discussions incorporated in the proceedings published by the International Atomic Energy Agency are edited by the Agency's edi­ torial staff to the extent considered necessary for the reader's assistance. The views expressed and the general style adopted remain, however, the responsibility of the named authors or participants. For the sake of speed of publication the present Proceedings have been printed by composition typing and photo-offset lithography. Within the lim i­ tations imposed by this method, every effort has been made to maintain a high editorial standard; in particular, the units and symbols employed are to the fullest practicable extent those standardized or recommended by the competent international scientific bodies. The affiliations of authors are those given at the time of nomination. The use in these Proceedings of particular designations of countries or territories does not imply any judgement by the Agency as to the legal status of such countries or territories, of their authorities and institutions or of the delimitation of their boundaries. The mention of specific companies or of their products or brand-names does not imply any endorsement or recommendation on the part of the International Atomic Energy Agency. CONTENTS

INTRODUCTORY PAPER

Микробиологическая генетика как теоретическая основа селекции промышленных микроорганизмов (IAEA-3M-134/20)...... 3 С. И. Алиханян D iscussion ...... 9

MECHANISMS OF MUTAGENESIS AND REPAIR PROCESSES (Sessions 1 and 2)

Molecular mechanisms of mutation (IAEA-SM -134/27) ...... 13 H . Heslot D iscussion ...... 40 Basis for radiosensitivity of some mutants of Hemophilus influenzae (IAEA-SM -134/3) ...... 43 N.K. Notani, V.R. Joshi and A. R. Go p a 1 - Ay e n g a r D iscussion ...... 51 Radioresistance of some microorganisms and their purine-5-phosphoribose-1 -pyrophosphate transferase activity (IAEA-SM -1 3 4 /8) ...... 53 G . Partsch and H. Altmann D iscussion ...... 61 U.V. mutability in gamma-ray-sensitive mutants of N eu rospora c ra s s a (IAEA-SM - 13 4 /13) ...... 63 R. D. Mehta and J. Wei j er D iscussion ...... 71 Some sites for the indirect effects of radiation on DNA constituents (IAEA-SM - 134/4) ...... 73 B.B. Singh, V.T. Srinivasan, K.P. Mishra and A. R. Go p al - Ay en gar D iscussion ...... 79

GENETICS AND PHYSIOLOGY OF INDUSTRIAL MICROORGANISMS (Session 3)

Application and importance of fungal genetics for industrial re& earch (IA E A -S M -1 3 4 /2 4 )...... 83 К . E s s e r D iscussion ...... 90 Physiological and genetical studies on yeasts of the genus Candida (IAEA-SM - 134/5) ...... 93 C . Gaillardin and H. Heslot D iscussion ...... 109 Vegetative instability in fungi: the role of chromosome aberrations (IAEA-SM-134/1) ...... 113 J. A. R o p e r D iscussion ...... 119 Instability at mitosis in Aspergillus nidulans (IAEA-SM-134/2) ...... 123 B. H. N g a Microbial genetics and the control of the pathogens in agricultural industries (IAEA-SM-134/22) ...... 129 S. G. Georgopoulos

ROLE OF PHYSICAL AND CHEMICAL MUTAGENS IN INDUSTRIAL MICROBIOLOGY RESEARCH (Session 4)

Induction of amylase-producing mutants in Aspergillus oryzae by different irrad ia tio n s (IAEA-SM- 13 4 /14) ...... 137 J . M eyrath, M. Bahn, H. E. Han and H. Altmann Mutation studies in Streptomyces aureofaciens (IAEA-SM-134/11 ) .. 157 M . Blumauerová, A .A . Ismail, Z.Hosïâlek and Z . V a n ё к D iscussion ...... : ...... 165

BIOSYNTHETIC PATHWAYS AND THEIR REGULATORY MECHANISMS (Session 5)

Regulation of amino-acid biosynthesis and industrial production of amino acids (IAEA-SM -134/31) ...... 169 R . Hü tt er D iscussion ...... 178 Regulation of amino-acid biosynthesis in Saccharomyces cerevisiae (IAEA-SM -134/26) ...... ____ 181 Huguette d e Robichon.-Szulmajster D iscussion ...... 188 Speculations on genetic loci controlling the biosynthesis of tetracy clin es (IA EA -SM -1 3 4 /3 5 )...... 189 Z . H ostàlek, M. Blum auerová, J. Cudlin and Z.Vanëk ' ' , ■ D iscussion ...... -198 Protein synthesis and production of tetracycline in S treptom yces aureofaciens (IAEA-SM- 1 3 4 /3 2 )...... 201 K . Mikulik, J. Karnetoya, A. Kremen, J. Tax and Z. Vanëk Discussion ...... 222 Selection of mutants not accumulating storage materials (IAEA-SM-1 3 4 /9) ...... 223 H.G. Schlegel and V. Oeding D iscussion ...... 230 Complexity of genetic control of biochemical processes in fungi as evidenced by studies on resistance to toxicants (IAEA-SM-134/ 28) ...... Vassiliki: Vo m v o y a nn i , S.G. Georgopoulos and A . К ap p a s D iscussion ......

MUTATION AND SELECTION OF INDUSTRIALLY USEFUL , MICROORGANISMS (Session 6 )

Techniques for the development of novel microorganisms (IAEA-SM - 134/ 21 ) ...... H.I. Adler D iscussion ...... Индуцированный мутагенез при селекции промышленных микро­ организмов (IA E A -S M -134/25)...... С. И. Алиханян D iscussion ...... Radiation improvement of Candida No. 25 for petroprotein ferm entation (IAEA-SM - 13 4 /12) ...... Shu-Hsun Ting, Chuan-Hsian Li and Ching-Song Lee D iscussion ...... Techniques utilisables pour l'amélioration génétique de la levure dans une optique industrielle (IAEA-SM-134/6) ...... P . Galzy et P. Dupuy Effect of microorganisms on the structure of uranium raw m a te ria ls (IAEA-SM -1 3 4 /15 ) ...... F . Barbie and B ranka К r a j i n с a n i с

CURRENT AND FUTURE OUTLOOK FOR APPLIED MICROBIOLOGY RESEARCH IN DEVELOPING COUNTRIES (Session 7)

Opening rem arks on the status of applied microbiology in developing countries (Short contribution) ...... J . M eyrath Microbiological and pharmaceutical activity in Ghana: a review (IA EA -SM -134/3 0 ) ...... M . С au r i e Use of isotopes and radiation in the study of microbial aspects of ruminant nutrition (IAEA-SM-134/33) (Abstract only) .. E sth er Balogh and A.A. Adegbola D iscussion ...... Utilization of microorganisms for the production of food and metabolites in the Philippines (IAEA-SM-134/29) ...... Lydia M . J o s о n Activities of UNIDO in the fermentation industries (IAEA-SM -1 3 4 /19) ...... C .S. Chiang Microbiological program activities of UNESCO (IAEA-SM-134/17) . 319 A. C. J. B urgers D iscussion ...... 322 Microbiological program activities of the IAEA (IAEA-SM-134/18) . 325 R . M u k h e r j e e

C hairm en of Sessions ...... 329 S ecretariat of the S y m p o siu m ...... 329 L ist of P articip an ts ...... 330 Author I n d e x ...... 338 INTRODUCTORY PAPER

IAEA-SM-134/20

МИКРОБИОЛОГИЧЕСКАЯ ГЕНЕТИКА КАК ТЕОРЕТИЧЕСКАЯ ОСНОВА СЕЛЕКЦИИ ПРОМЫШЛЕННЫХ МИКРООРГАНИЗМОВ

С. И. АЛИХАНЯН Всесоюзный научно-исследовательский институт генетики и селекции промышленных микроорганизмов, Москва, Союз Советских Социалистических Республик

Abstract — Аннотация

MICROBIOLOGICAL GENETICS AS A THEORETICAL BASIS FOR THE SELECTION OF INDUSTRIAL MICRO­ ORGANISMS. On the basis of concrete examples taken from the history of the development of the microbiological industry in the Soviet Union and the rest of the world, the paper develops the theme that the decisive factor in the rapid progress of microbiological synthesis on an industrial scale is the improvement of genetic methods for work with microorganisms. It is shown that by applying methods developed along traditional lines, in particular methods of producing mutagenesis by means of X-rays, ultraviolet rays or fast neutrons, a resounding success has been achieved in the selection of strains producing antibiotics. Side by side with these methods an important role has been played by biochemical mutations, particularly in the breeding of strains producing amino acids. A summary is given of the results of the initial period of work on increasing the yield of micro­ organisms, a characteristic of the period being its empiricism. At the present time, characterized by the rapid development of activities in the field of molecular genetics and also by advances in the microbiological industry, progress is being made towards gearing theoretical research to the practical needs of the micro­ biological industry. In conclusion, the paper presents a review of the latest achievements and ideas in mole­ cular genetics that are likely to be of use in the task of selection in the future.

МИКРОБИОЛОГИЧЕСКАЯ ГЕНЕТИКА КАК ТЕОРЕТИЧЕСКАЯ ОСНОВА СЕЛЕКЦИИ ПРО­ МЫШЛЕННЫХ МИКРООРГАНИЗМОВ. В докладе на конкретных примерах, взятых из истории развития советской и мировой микробиологической промышленности, развивается концепция, что решающим фактором, обусловившим быстрый прогресс промышленности микробиологического синтеза, является совершенствование генетических методов работы с микроорганизмами. Показано, что на основе применения методов, разработанных на классических объектах, в частности, мето­ дов мутагенеза с использованием рентгеновских, ультрафиолетовых лучей, быстрых ней­ тронов, был достигнут решающий успех в селекции штаммов-продуцентов антибиотиков. Наряду с этими методами, большое место занимало использование биохимических мутаций, особенно при выведении штаммов-продуцентов аминокислот. Подводится итог первому пе­ риоду работ по повышению продуктивности микроорганизмов, характерной чертой которого является эмпиризм. В настоящее время, характеризующееся бурным развитием работ по молекулярной генетике с одновременным подъемом в развитии микробиологической про­ мышленности, выдвигается задача подчинения теоретических исследований практическим потребностям микробиологической промышленности. В заключительной части доклада рас­ сматриваются самые последние достижения и идеи в молекулярной генетике, которые дол­ жны послужить основой селекционной работы в будущем.

На данном симпозиуме речь пойдет о промышленных микроорганиз­ мах, в том числе о выведении высокоактивных, селекционных штаммов. Сочетание в программе работы симпозиума генетических, биохимичес­ ких и технологических вопросов не случайно. Высокопродуктивные штам­ мы — основа организации промышленности микробиологического синтеза.

3 4 АЛИХАНЯН

Организация любого производства, основанного на микробиологичес­ ком синтезе, требует решения двух вопросов. Первый вопрос связан с физиологическими и биохимическими иссле­ дованиями. Микробиологи находят в природе новые формы микроорганиз­ мов и совместно с биохимиками открывают у них новые свойства - синте­ зировать специфические продукты. Иногда это бывают хорошо известные формы микроорганизмов, у которых обнаруживается способность синте­ зировать новые вещества, как это имело место в случае с пенициллином или стрептомицином. Однако большинство микроорганизмов-дикарей об­ ладают слабой синтезирующей способностью, вследствие чего организация на их базе промышленного производства не рентабельна и экономически не оправдана. Второй вопрос связан с организацией микробиологического производ­ ства и определяется селекцией указанных штаммов, проводимой с целью резкого повышения продуктивности дикарей-микроорганизмов. Такая схе­ ма работы была бы неполной, если бы мы не добавили к сказанному, что па­ раллельно с созданием новых высокоактивных штаммов ведется исключи­ тельно серьезная работа по изучению потребностей селекционных штаммов в питательных веществах, по поиску предшественников, по установлению оптимальных условий аэрации, по изучению температурных условий био­ синтеза, а также ведутся разработки методов подготовки и выращивания посевного материала и т.д. Во всех звеньях работ по повышению продуктивности микроорганиз­ мов решающее слово принадлежит генетике. Сегодня совершенно очевид­ но, что вся селекционная работа с микроорганизмами, все ее успехи це­ ликом и полностью связаны с применением генетических методов. Мно­ гократные попытки игнорировать генетические методы в селекции микро­ организмов ни к чему хорошему не привели. Подавляющую роль в селекции микроорганизмов на первом этапе ее развития сыграло использование генетических принципов, основанных на индуцированном мутагенезе с использованием ионизирующей радиации, точнее - рентгеновских лучей. Двадцать пять лет прошло с того времени, когда Милощ Демерец впервые использовал рентгеновские лучи (в то вре­ мя известные как очень активный мутаген) для получения первого мутан­ та P. chrysogenum, штамма Х-1612, утроившего производительность штамма-дикаря 1951-В25. Тогда это было первым триумфом практичес­ кого использования мутагенного эффекта рентгеновских лучей, который был за 20 лет до этого открыт Меллером. Утвердившись как эффектив­ ный метод селекции, экспериментальный мутагенез в селекции промыш­ ленных микроорганизмов стал главным и на какой-то промежуток време­ ни единственным методом селекции микроорганизмов. Таким образом бесспорно, что радиация явилась первым фактором, открывшим необъятные просторы в селекции промышленных микроорга­ низмов, на основе применения которого созданы сверхактивные штаммы. С применением этих штаммов удалось превратить микробиологический синтез не только в единственное средство рентабельного производства целого ряда соединений, но и во вполне конкурентоспособный метод по сравнению с химическим синтезом. Вслед за рентгеновскими лучами широ­ кое применение в селекции микроорганизмов нашел другой вид радиа­ ции - ультрафиолетовые лучи, сильный мутагенный эффект которых был открыт еще в 1931 году советским генетиком В. А. Промптовым. Первым микроорганизмом, подвергнутым действию ультрафиолетовых лучей был IAEA-SM -134/20 5

опять-таки P. chrysogenurn. В результате был получен мутант Q-176, который резко превосходил по активности исходный штамм 1951-В25 и втрое превысил по активности первый рентгеновский мутант. Получение радиомутантов Х-1612 и Q-176 произвело подлинную рево­ люцию в пенициллиновой промышленности. "Авторитет" радиации, а точ­ нее ее генетического действия, в селекции микроорганизмов резко воз­ рос. С тех пор прошло 25 лет, а рентгеновские и ультрафиолетовые лучи продолжают успешно использоваться в селекции микроорганизмов. По­ лучение штаммов Х-1612 и Q-176 не было случайностью. Это —резуль­ тат разработки генетиками проблем индуцированного мутагенеза. Вслед за рентгеновскими и ультрафиолетовыми лучами были испытаны и затем успешно использованы в селекции гамма-лучи и быстрые нейтро­ ны. Эффективность быстрых нейтронов в селекции микроорганизмов за­ метно превосходила рентгеновские и ультрафиолетовые лучи. Полезной задачей данного симпозиума является обсуждение вопроса о более широком использовании реакторов (там, где они имеются), а так­ же специальных контейнеров для обработки микроорганизмов быстрыми нейтронами и получения заметных результатов. Нельзя не отметить некоторых успехов в использовании метода ком­ бинированной обработки спор — радиацией и химическими мутагенами. Кстати, заметим, что нами впервые, в 1957 году, была показана высокая эффективность комбинации этиленимина и ультрафиолетовых лучей. По­ лученный результат намного превышал ожидаемый — суммарный эффект действия двух мутагенов. Был получен не куммулятивный, а синергид- ный эффект. Синергидный эффект, выявленный на морфологических му­ тациях, был затем повторен и в селекции продуцента эритромицина и стрептомицина. Можно рекомендовать пользоваться методом воздей­ ствия ультрафиолетовыми лучами на споры, предварительно обработан­ ные этиленимином. Попытка же добиться положительного эффекта путем комбинирова­ ния рентгеновских лучей или быстрых нейтронов с химическими вещества­ ми не дала положительных результатов. Неэффективной оказалась и комбинация ультрафиолетовых и рентгеновских лучей. Успех в селекции микроорганизмов не всегда обеспечивается приме­ нением мутагенов. Так, например, большую роль в работе с продуцента­ ми аминокислот сыграло применение биохимических мутаций (лизин, глу­ таминовая кислота,, триптофан, инозиновая кислота и др. ). Селекция микроорганизмов прошла большой путь развития. Ее поло­ жительный эффект сравним с эффектом лучших образцов селекции расте­ ний и животных, а в некоторых случаях и превосходит его. Поэтому мы вправе высоко оценить роль генетической теории, которая обеспечи­ ла эти успехи. Высоко оценивая роль учения об индуцированном мутаге­ незе в области селекции микроорганизмов, а также учения о биохимичес­ ких мутациях, мы должны особо подчеркнуть важность того первого пе­ риода селекции микроорганизмов,которому сегодня мы отмечаем 25-ле­ тие. Основной характерной чертой этого периода является эмпиризм, особенно в области применения различных мутагенов. Дело в том, что до самого последнего времени решение теоретических проблем экспери­ ментального мутагенеза производилось независимо от практических нужд. Такое соотношение теоретических проблем и практических нужд во. мно­ гом объясняется (в области микробиологии) тем, что в то время не воз­ 6 АЛИХАНЯН

никало необходимости применять мутагены в селекционных целях по про­ стой причине отсутствия микробиологической промышленности. Значи­ тельно позже, после начала активных исследований экспериментального мутагенеза, в 1943-1945 гг. вместе с развитием крупнотоннажного про­ изводства пенициллина и других антибиотиков, возникла необходимость в селекционно-генетических работах с микроорганизмами. В этот пери­ од были использованы уже имеющиеся в арсенале генетики методы: экспе­ риментальный мутагенез и позже — биохимические мутации. Практические эффекты использования этих методов были настолько велики, что селекционерам оставалось лишь приспособить микроорганиз­ мы к различным условиям обработки мутагенами физической и химичес­ кой природы и сделать некоторые обобщения, касающиеся, главным обра­ зом, особенностей индуцированной изменчивости полигенной системы. Наоборот, второй период в истории развития генетики связан с бур­ ным развитием работ по генетике микроорганизмов и молекулярной гене­ тике, совпавшим по времени с развертыванием микробиологической про­ мышленности. Такое параллельное развитие науки и развертывание про­ мышленности ставит перед нами задачу подчинения теоретических иссле­ дований практическим потребностям микробиологической промышленности. Перечислим некоторые проблемы генетики микроорганизмов и моле­ кулярной генетики, которые, по-видимому, могут найти эффективное при­ менение в практике селекционной работы с промышленными микроорга­ низмами, а также остановимся на тех вопросах генетики микроорганиз­ мов, которые сегодня еще далеки от практического разрешения, но над которыми следует подумать селекционерам. В последнее время развиваются исследования, связанные с изучением влияния соответствующего расположения генов в хромосоме на жизнедея­ тельность клеток. Как известно, репликация хромосомы начинается в определенной точке. Следовательно, есть гены, которые удваиваются раньше других. Возникает своеобразный эффект дозы генов, связанный с накапливанием вдвое большего количества продуктов таких генов. Этот факт может иметь определенное значение для регуляции последователь­ ности синтеза разных веществ в клетке. Например, оказалось: гены на хромосоме B.subtilis расположены таким образом, что прежде всего удва­ иваются столь важные гены, как гены, контролирующие синтез транспорт­ ных и рибосомных РНК. Гены аминокислотного метаболиза, напротив, концентрируются в противоположном конце хромосомы. Такой эффект, названный Суэокой-эффектом последовательности, может приобрести большое практическое значение. Так, добившись син­ хронного развития определенной культуры, можно использовать эффект дозы генов, контролирующих синтез рибосомных РНК и направить их на синтез нужных белков, ферментов и вторичных метаболитов. Другая проблема связана с обнаружением повышения частоты спон­ танных и индуцированных мутаций у штаммов с нарушенной системой ре­ комбинации и повышенной чувствительностью к ультрафиолетовым лучам. Было показано, что обнаруженное повышение спонтанного мутагенеза имеет специфический характер — возникали своеобразные "горячие точ­ ки", дававшие мутации, ранее не известные у исходного штамма. Раз­ работка путей практического использования этого эффекта откроет ог­ ромные перспективы перестройки целого ряда метаболитов и обнаруже­ ния "новых" генов, среди которых окажутся гены, контролирующие син­ тез ранее не известных продуктов. IAEA-SM-134/20 7

Интенсивные исследования ведутся по установлению родства между генами. Эти работы основаны на межвидовой трансформации и гибриди­ зации разных участков ДНК. В итоге этих исследований выделены 3 груп­ пы генов: 1. Гены с малой степенью родства. К ним, в частности, относятся гены аминокислотного обмена. Они не дают перекрестной трансформа­ ции. 2. Гены со средней степенью родства, контролирующие, главным об­ разом, морфологию клеток, спорообразование, дыхательные процессы. Они дают перекрестную трансформацию и лежат в двух участках хро­ мосомы B.subtilis. 3. Гены с высокой степенью родства. К ним относятся гены, кон­ тролирующие синтез рибосомной и транспортной РНК. Они дают гибри­ ды даже между ДНК В. subtilis и E. coli. Эти исследования, на первый взгляд кажущиеся отдаленными от прак­ тического применения, могут неожиданно обернуться эффективным при­ емом селекции микроорганизмов, если согласиться, что участки генома, контролирующие сравнительно менее важные, приспособительные функции, подвержены большей эволюционной изменчивости. В последнее время большое место в генетических исследованиях ми­ кроорганизмов занимают работы по выяснению механизмов репарации. Один из активных исследователей этого вопроса — Ханавальт — справедли­ во считает, что в процессе репарации участвует, по меньшей мере, 5 фер­ ментов. Они принимают участие в различных этапах репарации: в обра­ зовании тиминовых димеров; в "надрезании", т. е. "инцизии" той цепиДНК, в которой лежит димер; в "вырезании" - "эксцизии" димера вместе с при­ легающим участком цепи ДНК; в репаративном синтезе на месте образо­ вавшейся бреши; и в сшивке, посредством лигазы. Значение восстанови­ тельных процессов очень велико, так как они позволяют сохранить без существенных изменений генетический материал на протяжении многих поколений. Данная проблема имеет как теоретическое, так и практичес­ кое значение. Так, например,в нашем институте на основе проведенных работ были получены интересные штаммы-продуценты триптофана, зна­ чительно превосходящие по своей продуктивности известные до сих пор штаммы. Эти штаммы дают высокий уровень синтеза триптофана без уча­ стия предшественника — антраниловой кислоты. Очень важные эффекты могут иметь для промышленных микроорга­ низмов исследования по регуляции конститутивного синтеза. Как извест­ но, уровень конститутивного синтеза ферментов не зависит от состава среды, аэрации и ряда других внешних факторов. Вместе с тем, уровни конститутивного синтеза различных ферментов существенно отличаются друг от друга. В связи с этим возникает важная проблема: как генети­ чески определяется потенциальный уровень конститутивного синтеза фер­ ментов? Решение этой проблемы может иметь важное, кроме теорети­ ческого, практическое значение, так как она связана с выяснением во­ проса, каким образом можно увеличить уровень конститутивного синте­ за природных метаболитов. А.Парди, например, считает, что за счет по­ лучения сверхустойчивых к репрессии мутантов, за счет увеличения род­ ства ферментов по отношению к субстрату, т.е. увеличения его активно­ сти, а также за счет более эффективной проницаемости клеточной оболоч­ ки в отношении субстрата можно повысить уровень конститутивного син­ теза ферментов. Нет необходимости говорить о том, какой грандиозный 8 АЛИХАНЯН

эффект в производстве ферментов может произвести решение этой проб­ лемы. В плане использования результатов генетических исследований регу­ ляторных механизмов клетки представляют интерес работы по созданию высокопродуктивных штаммов - продуцентов аминокислот. Как известно, Производство ауксотрофов до сих пор являлось наиболее перспективным и единственным методом получения продуцентов аминокислот. Ныне, пос­ ле большой серии генетических работ по изучению регуляторных генов, возникает проблема использования регуляторных мутантов, т. е. мутан­ тов с нарушенной регуляторной системой, одним из способов получения которых является отбор видов среди культур, резистентных к аналогам аминокислот. Вероятно, можно считать перспективным получение штам­ мов, сочетающих ауксотрофную мутацию и мутацию, затрагивающую регу­ ляторный аппарат клетки. Первые опыты с продуцентом лизина, резис­ тентным к S-/2 аминоэтил/ цистеину-ингибитору первого общего фермен­ та синтеза аминокислот аспарагинового ряда, дали положительные резуль­ таты. Такой фермент теряет чувствительность к совместному ингибиро­ ванию лизином и треонином. В итоге увеличивается синтез лизина. Для использования этого метода в практической селекции необходимо четкое представление о регуляторных механизмах биосинтеза аминокислот. В последние 1,5-2 года повысился интерес генетиков к вопросам спо- руляции. Актуальность этой проблемы, с точки зрения промышленной микробиологии, в значительной мере определяется тем,что спорулирующие бактерии иногда выделяют в среду большое количество какого-либо мета­ болита непосредственно в процессе споруляции, а не в период предшест­ вующего роста. Недавно Фриз на основании известных фактов о том, что в предспоруляционный период или на самых ранних стадиях споруляции в культуральной жидкости особенно активно выделяются некоторые метабо­ литы, высказал предположение, что этот "выгодный" период может быть растянут на долгий срок, если использовать мутанты с нарушениями в стадии споруляции. В последнее время были выражены вполне обоснованные соображения о том, что в случае спорообразования наблюдается положительная регуля­ ция, когда появление того или иного вещества "разрешает" начать дея­ тельность группе генов. Если это справедливо, то открываются интерес­ ные практические перспективы для новых подходов к селекции промыш­ ленных микроорганизмов. В последнее время в ряде стран проводятся интенсивные исследова­ ния, связанные с вопросами производства клеточного белка на углеводо­ родах нефти. Эти исследования приняли довольно широкие масштабы, а в некоторых странах даже организованы заводы для производства белка. В большинстве описанных случаев белок представляется в виде высушен­ ной биомассы дрожжей. Во всех случаях применяются дрожжи-дикари. Нам представляется весьма перспективной идея, высказанная недавно на X Международном микробиологическом конгрессе, об изучении путей про­ изводства микробного белка микроорганизмами, продуцирующими боль­ шие количества какого-либо одного белка (например, каталазы или иного белка с нужным составом аминокислот). Решение этой проблемы, в пер­ вую очередь, связано с разработкой большинства из перечисленных выше теоретических вопросов. Нынешний период развития генетики дает нам прекрасное сочетание теоретических исследований в области генетики микроорганизмов с их IAEA-SM-134/20 9

практическим использованием. Если первый период селекции носил на себе солидный отпечаток эмпиризма, то в переживаемый нами период есть все основания надеяться на развитие теоретической базы селекции микроорганизмов. В заключение позвольте от имени советских генетиков микроорганиз­ мов пожелать участникам симпозиума успешной работы.

DISCUSSION

J. MEYRATH: There are reports in the patent literature on the successful use of hybridization in connection with baker's yeast and on improvements in the baking process as a result. S. I. ALIKHANIAN: This is true. Yeast, especially baker's yeast, is perhaps the only microorganism which is used effectively in hybridization for purposes of selection. H. HESLOT: You mentioned that you have isolated bacterial mutants excreting lysine. What amount of lysine, expressed in g/litre, do they excrete in the growth medium? Are these mutants already being used industrially? S. I. ALIKHANIAN: Yes, they are. The strains which have been isolated at our institute synthesize 40-45 mg/ml or 40-45 g/litre L-lysine. We are beginning to use these strains in the production of lysine.

MECHANISMS OF MUTAGENESIS AND REPAIR PROCESSES (Sessions 1 and 2)

Chairmen

J.A. ROPER (United Kingdom) H .I. ADLER (United States of America)

IAEA-SM-134/27

MOLECULAR MECHANISMS OF MUTATION

H. HESLOT Institut national agronomique, Paris, France

Abstract

MOLECULAR MECHANISMS OF MUTATION. The paper reviews the principal categories of mutagenic agents: analogues, hydroxylamines, nitrous acid, alkylating agents, acridines, ultra-violet rays and ionizing radiations. Their mode of action and mutagenic activities are discussed. The different kinds of repair mechanisms are reviewed and their importance is stressed. A few recommendations are given for the proper choice of mutagenic agent.

Investigations during the last decade have given us a clearer under­ standing of the nature of mutations. Molecular mechanisms have been proposed to explain the origin of both spontaneous and induced mutations. It has been shown that enzymes played a very important part in the process either as repair devices or as controlling systems at the level of DNA replication. In the present review an attempt is made to discuss the established facts and to point out the areas of uncertainties. Finally, a few suggestions are made on the choice of the proper mutagenic agent.

1. CLASSIFICATION OF MUTATIONS

As suggested by Drake [28] it is convenient to distinguish between micro- and macrolesions.

A. Microlesions

In the first class we are dealing with true point mutations involving the replacement of one base pair by another (substitutions). Freese [34-37] has proposed to divide substitutions into two subclasses which he defines as indicated in Fig. 1. A transition is therefore the replacement of a purine by another purine or of a pyrimidine by another pyrimidine in one DNA strand. Transversion is the replacement of a purine by a pyrimidine and vice versa. A second group of microlesions corresponds to the so-called frame- shift mutations, i.e. the loss or addition of one (or a few) base pairs. Ob­ viously, there is no clear limit between frameshift mutations and large deletions, but usually the first are able to revert whereas extensive deletions cannot. It is well established that the translation process, at the level of messenger RNA, is polarized. It starts at a fixed initiation point and pro­ ceeds by successive translation of groups of three bases (codons). Therefore, if one base is added or deleted a reading frameshift will result. Because codons always contain three bases, the loss or addition of three bases (or a multiple of three) re-establishes the reading frame (Fig. 2).

13 14 HESLOT

Iwiki'al bast. Fífxal bait Nani «P S*bjt'iUtr«i

*Pur»he* otK«(r purihe. TranfV-biotw Pyn> icfi’ftc. ofckcr py n'midUn«-. TVftH ffitl'oh. purine. any pyríMídine- "T*-aniV«.*Ti'oK. By п’|И» (Jt'nt ftoy purin*. "TrftHJVtrii'o»»

A------^ ------___ С------

T RANfVE RSiöNJ

FIG.l. Base-pair substitutions.

At the level of proteins, base-pair substitution can result in three possible events:

(1) No change of the corresponding amino-acid residue, when the codon has been altered to a synonymous one; (2) Replacement of one residue by another (missense) producing active or inactive protein. In the first case, the mutant cannot be detected. In the second case, the mutant may be leaky or temperature sensitive (ts). These ts mutants are very powerful tools for the study of indis­ pensable functions (conditional lethals); (3) Production of chain-terminating codons, amber (UAG), ochre (UAA) or opal (UGA). In this case, only a fragment of protein is made. These chain-terminating mutants are suppressed by specific suppressors.

Frameshift mutants have a very drastic effect upon proteins. From the point of base addition/deletion, all codons are changed and the corresponding amino-acid residues will also be altered. LAEA-SM-134/27 15

A B C AB+- CAB СЛВ CAB CAB CAB F r a m e r U if t fco t K t le-Ctr

CLbnermat s«£|4«Kce. tofck«. e*A-

AtDiTioN OF ONE ВЛГЕ-

A SC ABC ABC A-ÏC ABC MC ABC Normal Situation,

DeccTioN ср­ оке ВаГЕ

i' С ? Vfcc : fia ßC A gCA BC A BCA Be A ----\------\------»------1------1 ■■ *---- F "raw i« .skif fc to Ькв. rv^U. b" 1------CLb*orma.l Sc-ci4cn.ee. fc© tk«_ erv

Л BC t A 8C1 ABC t ЛВС t A8С ; ABC . ABC | ...... Normal sCtuat.W

A"b DIT l ON OF тнкее B«res ; Ф I ABC- ftb+ g-AB -^Cft B-t-C A ВС ABC ABC Slvoi-t abnormal Science. ^ l------;------=*- bw-t readm^ fmmi restored GLbwormal SaûuÆirvce. I------Rta^i'n^ fглmeskif ir res t 0

FÏG.2. Frameshift mutations.

B. M acrolesions

Macrolesions can be classified as deletions, duplications and re­ arrangements . Deletions are distinguished from frameshift mutants by their inability to revert. Provided that they involve the loss of a long enough DNA segment, they can be detected genetically by the decrease in recombination frequency of outside markers. In viruses, they can also be mapped with great accuracy by electron microscopy, as shown by the work performed on X bacteriophage by Davis et al. [20] and Westmorland et al. [109]. The method involves DNA extraction from the wild type and the presumed deletion mutant. The two DNAs are mixed and melted. Renaturation shows single-stranded loops extending from DNA heteroduplex. Deletions and duplications could result from errors arising during DNA replication as shown in Fig. ЗА. Here, DNA polymerase is supposed to detach from the parental strand and to re-attach, either at a later point (deletion) or at a previously replicated point (duplication). Deletions and duplications could also result from errors in a repair process as shown in Fig. 3B. Inversions, on the other hand, produce a local reversal of the polarity of the chromosome, because the inverted segment can only re-insert by respecting the polarity of strands 3 1 -* 5 1 (Fig. 4). Asa consequence, messenger 16 HESLOT

A- ERRoR. AT REPLICATION :

тттпт/^тттп "1114_lio

ÎELETION и MimTT

lili I 11111*1.1 “УГ* £ и Plí ca t i on

В . E RftoR AT RePftiR.

AG AG AG AG 1 1 1 1 1 1 1 1 И И 1 1 1 II 1 1 И 1 1 Г Г 1111 1 II 1 1 II 1i il 1 1 1 1 1 1 I I 1 1 1 1 1 1 Il 1 I TC тс TC тс

A& AG AG '1 A& 1 И И 1 1 1II 1 1 1 1 1 II 1 1 1 1 И II 1 1 1 1 LLL.i 1 1 ± 1 1 1 1 1 1 т с 1 >C

AG- ’ AG 1 1 1 1 T 1 1 1 ГТ 1 1 1 II 1 II 1 1 1 II M i l 1Л 1_1 1 1 1 1 1 1 1 1 1 U INI т с JJC G/T\ i L,^ St Lijoje. A Gr AG- TT Г Т Чттт T| I I I II I I III П I I I I______, , I I I I I I I I и Л ? д т с т с j . TC\ ■DELETION DuPl'icatÍon

FIG.3. Possible origins of deletions and duplications. (Adapted from Drake [28].)

RNA of the inverted segment will be transcribed in a direction opposite to that of the adjacent regions. In fact, such a situation exists in X and in T4 bacteriophages.

2. MUTAGENIC AGENTS

A. Analogues

Analogues are substances closely related to A, G, С or T which can be incorporated into DNA without hindering its replication. However, since the analogue differs from the normal base in certain substituents, its electronic structure is modified and one expects that occasionnai errors will occur in the specificity of bonding. The most frequently used analogues are 5-bromo uracil (BU) and 5-bromo deoxy-uridine (BUdR), which are analogues of thymine, and 2-amino purine (AP) which is an analogue of adenine. IAEA-SM-134/Í27 17

N A ------^

c А В С Ь ! ..« ______et- b с. d t F 3 l~

2. Br-to-Wx

t A B С T> E F G H

о, Ь c A t f g fC

Inversion. ' В ^ di c. _ F G- H

л. ь e ъ c " ( 9 к

R e in iw tioK.

i f ««Rna

^ A В e el с. P 6- Ц ^ ,

к '------______:------^ ----=------—Г-» > 23 ' а.Ь<=1>С.рЭк <------*ч - RNA

FIG.4. Effect of an inversion (From Drake [28]).

With regard to BU and BUdR, Freese [34-37] has attributed their mutagenic act:‘”ity to the replacement of the methyl group of thymine by a bromine atom whose electronegativity is greater, the result being that the probabi­ lity of accidental pairing between BU and G is increased. Theoretically, base- pairing errors can occur in two different ways (Fig. 5), either at incorporation or at replication. The net result is that BU and BUdR should be able to induce transitions in two directions: A-T ** G-C. The other analogue, 2-amino purine (AP), behaves in general like adenine and pairs therefore with thymine by hydrogen-bond formation. How­ ever, it can exist occasionally in an imino form and pair with cytosine. Therefore AP should be, like BU, able to induce transitions in the two directions А-T ** G-C. Such were the simple ideas proposed ten years ago by Freese [37] to explain the mechanism of mutations induced by the analogues. However, it has since been demonstrated that the situation was far more complicated and even now great uncertainties remain. 18 HESLOT

I - Replication after incorporation 2 _ Incorporation

A-T G-C

/ 4

IM i stake! A-BU A-T I i II ‘ G-B U G -C I i I I

I I I MistcKel A - T G-BU' G - С A -B U

/ \ /\

G-C A-BU A - T A -B U i I I ! ¡I

Consequence : transition Consequence : transition

A - T — ►G - С G-C ------► A -T

FIG. 5. Mutations induced by 5-bromo-uracil (BU) mistake pairing with guanine at the time of replication.

As soon as the genetic code was firmly established it became possible, on a few selected systems, to study the amino-acid substitutions induced by the analogues. Such was the A protein component of Escherichia coli tryptophan synthetase, studied by Yanofsky et al. [113]. Figure 6 shows the substitutions observed at position 210 , occupied by a gly residue in the enzyme of the wild strain. It is clear that AP induces the transition A-T -» G-C. Out of 32 reversions induced, 29 were transitions and the remaining 3 transversions possibly attributable to the spontaneous background. A similar conclusion results from the work of Stretton et al. [99] on the head protein of bacteriophage T4 . Here, the authors studied the reversion pattern of an amber mutant (UAG). Out of 10 revertants induced by AP, 9 contained a glutamine residue (UAG -» CAG). A consequence of the model pictured in Fig.5 is that it oughttobe possible to distinguish between errors of replication (A-T - G-C) and errors of in­ corporation (G-C -» A-T) by- a short exposure of the DNA of a suitable micro­ organism to BU. In the first case, as BU is already in DNA, A-T ^ G-C transitions can continue to occur at later generations. In the second case, in the absence of free BU, G-C A-T cannot occur. Strelzoff [96-98] studied the BU-induced reversion of six E^ coli auxo- trophs. Revertants of one of them continued to appear after removal of BU, whereas for the five others reversions stopped. A i»9 A&A

í p U j spO; t f l P C ; HN»iOj е и г с у

y v v y y IUu. T hf S ir Gly Ale. Vo-L A U A A C A AG-py GG A & C A G-UA

sp = fpon.taM.our 2. AP = 2, o.»nino purint. Origin. of mutant.! ЕМГ = ebhy) m ttk»nc. iul-foKo.t«_

U V = u L tv o -V io U - t HNOt „ KikyoKr bcCcl

FIG. 6. Amino-acid substitutions observed at position 210 of protein A of tryptophan synthetase. The number of independent occurrences are given in brackets. (From Yanofsky, Ito and Horn [113] .)

A similar experiment was performed by Terzaghi et al. [100] on 1ysozyme mutants of bacteriophage T4. Unfortunately, when the mutants were tested for their ability to revert under the effect of hydroxylamine (a specific inducer of G-C -> A-T transitions, see below), contradictions were found. In-vitro experiments have been made in an attempt to clarify the mecha­ nism of BU mutagenesis. For instance, Trautner et al. [104] studied the replication of poly-d-AT and of poly-d-ABU directed by E_. coli DNA poly­ merase. Two sets of assays were made:

(1) Use of poly-d-AT as template in the presence of ATP, BUTP and GTP (of high specific radioactivity). In this case, only one guanine residue was incorporated among 105 total residues. (2) Use of poly-d-ABU as template. In this case guanine is incorporated much more frequently, on an average one residue out of 10 4.

As the two templates are regularly alternating, the nearest-neighbour analysis has been applied by Trautner et al. [104]. If guanine really mispairs with BU, the preceding base should be BU. In fact, this occurred only in 41% of the cases, the other guanine residue following guanine (42%) or adenine (17%). Up to now we have no clear explanation of these discrepancies. It may be that in vitro the mispairing possibilities are wider than in vivo. 20 HESLOT

NH,

NHiOH

Effe,et of hydroxylanrune. on cytosine.

Ос(ел|к«.

HONH

O n t pairing poCU'biUby with. GLotenmc

FIG.7. Hydroxylamine mutagenesis. (From Phillips and Brown [76] .)

With regard to BU and AP mutagenesis, it can therefore be concluded that the situation is still far from clear. We shall see later that DNA poly­ merase itself plays an active role in the pairing process during replication.

B. Hydroxylamine

Freese, Bautz-Freese and Bautz [39-41 , 42-45] have shown that hydroxylamine is a potent mutagen. It reacts with both cytosine and uracil, but thymine is practically not affected. The chemistry of the reaction has been studied by Phillips, Brown and Grossman [76-78]. Cytosine gives rise to a derivative, which is apparently the mutagenic species, able to pair with adenine (Fig. 7). The reactivity of cytosine varies considerably according to the origin of DNA and its state of dénaturation; for instance Bacillus subtilis DNA is much more mutable (X 1000) when it is first denaturated [45]. T 4 £n mutants induced by hydroxylamine can be reverted by BU or AP but cannot be reverted by hydroxylamine itself, as shown by Champe et al. and Freese et al. [18,41]. A similar situation applies with mutants induced in фХ 174 [52, 103], which contains sin g le-stranded DNA. A num ber of w ork ers [13, 14, 18] have shown that chain-term inating codons, i.e. UAA, UAG and UGA, are induced from the wild type by hydro­ xylamine, but they cannot be reverted by the same mutagen. This is in IAEA-SM-134/27 21 perfect agreement with the idea that NH2OH induces only G-C -*■ A-T transi­ tions. It was also demonstrated, by use of appropriate nonsence supressors, that the transition UGA -» UAA is induced by hydroxylamine and AP, whereas UAA - UAG cannot be induced by hydroxylamine. Finally, a test of the mutagenic activity of hydroxylamine was performed in vitro by Phillips et al. [77] with the RNA polymerase of Micrococcus lysodeikticus. In this system poly-С is used as template to synthesize poly-G from GTP. If the template is treated by hydroxylamine, synthesis of poly-G stops. However, synthesis resumes when ATP is supplied into the incubation medium and A is incorporated into the polymer.

C. N itrous acid

Nitrous acid has a very strong mutagenic activity for bacteriophages [8, 104, 106], bacteria [56, 57], pneumococcus transforming DNA [64] and tobacco mosaic virus RNA [74]. In the latter case, analysis of the protein component of a number of mutants showed that they could be explained on the b asis of tran sitio n s A -* G and С -* U. Similarly, Yanofsky et al. [113 ],in studying the A component of tryptophan synthetase of E. coli, showed that HN02-induced mutants were mostly the result of transitions A-T-* G-C. Tranversions were also observed, but it was not clear whether they came from the spontaneous background or not. In Neurospora. Mailing and de Serres [70] concluded that HN0 2 -m utants resulted from the A-T -» G-C transition. On the other hand, Tessman [102] showed that nitrous acid induces deletions in phage T4. Nitrous acid acting on DNA induces deaminations, e. g. guanine -* xanthine, adenine -* hypoxanthine, and cytosine -> uracil [86- 88]. As hypoxanthine is able to pair with cytosine, this can explain the transition A-T -> G-C. Again, as uracil pairs with adenine, the transition G-C -» A-T is readily accounted for. For xanthine, the situation is much less clear. We would expect xanthine to pair with cytosine and so this deamination would bot be mutagenic. However, when xanthine is incorporated in artificial polynucleotides it cannot be used as template [72,81]. If the same situation applies in vivo, the deamination G -* X ought to be a lethal event. S chuster and V ielm etter [85, 87, 88, 106, 107] have studied compara­ tively in bacteriophage T2, as a function of pH, the rate of deamination of the bases as well as the kinetics of mutation and inactivation. By lowering the pH, the deamination rate of A and С is greatly increased (X 90) but that of G is somewhat less increased (X 35). At the same time, the mutation rate increases by X 80 and the inactivation rate by X 30. The correlation between the chemical action of HN0 2 and its biological effects is therefore satisfactory. Another effect of HN02 on DNA is the covalent binding of the two strands. This is a fairly frequent event: one cross-link for four deaminations [9, 46]. The exact chemical nature of this linkage is still unknown. May be such a reaction could explain the occurrence of the deletions mentioned by Tessman [102] in T4.

D. A lkylating agents

These compounds [49-51, 61, 65, 82] bear one or more reactive alkyl groups capable of being transferred to other molecules at positions where 22 HESLOT

1Л 01 (Л 2 LU ш ш Ш о — — '—' 0- X 1 и ° 4 in о / iO X 1 I ? \ / X СМ in -r"/ СМ о X + 2 и CN о и It о

X ÍCC о I 1 / СМ +2 X ' \/ О о сл III / \ I IAEA-SM-134/27 23 the electron density is sufficiently high. Figure 8 shows the principal cate­ gories of alkylating agents: sulphur and nitrogen mustards, epoxides, ethylene imines, sulphates and sulphonates, certain sultones and lactones, some diazoalkanes and nitroso compounds (the latter probably being indirect alkylating agents via diazo derivatives). The mutational effect of EMS (ethyl methane sulphonate) and EES (ethyl ethane sulphonate) has been studied with a number of systems: the гц mutants of T 4 [V, 38, 59], mutants indu ced in bacteriophage SI 3 [103], mutants of the E. coli tryptophan synthetase [113], and adenineless mutants of Neurospora crassa [71]. In T4, alkylating agents reverse very actively those mutants which are also reverted by hydroxylamine. This means that G-C -*■ A-T transitions are induced. However, this is not the only event because some mutants not reverted by hydroxylamine are reverted with a low efficiency by EMS or EES as well as by base analogues. Even proflavin-indiiced mutants are reversed by EMS and EES. One is therefore led to conclude that A-T - G-C transitions and frameshifts are also induced by alkylating agents. There are also indications that transversions do occur. Studies [113] on reverse mutations induced by EMS in tryptophan synthe­ tase of E_. coli showed that amino acid replacements can be explained by A-E -* G-C transitions and by A-T -*• T-А and G-C -* С-G transversions. In N. crassa, Mailing and de Serres [71] found that EMS induces pre­ dominantly transitions (mostly A-T -► G-C), but also transversions, fram e­ shifts and deletions. These results show that the overall mutagenic effects of alkylating agents are fairly complex. A special mention must be made of N-methyl-N'-nitro-N-nitrosoguanidine (NG) studied by a number of workers [1, 15-17]. In bacteria it induces transitions, transversions and deletions. It has been shown that NG muta- genize preferentially the replication point. NG is a methylating agent but can also give rise, under certain conditions, to nitrous acid. The alkylating agents react preferentially with the N7 of guanine, then to a lower degree with N3, N1 and N7 of adenine, and N1 of cytosine [61] . The alkylation of guanine in N7 is likely to be mutagenic because the product ionizes and could pair with thymine. Such an event would lead to a G-C -► A-T transition. However, this is not definitely proved and the occurrence of transversions has still to be accounted for. Alkylating agents also react with DNA by alkylating the phosphate groups as has been proved by Alexander [2], Reiner and Zamenhof [80], Stacey et al. [94] and Alexander and Stacey [3]. The triesters thus formed are unstable. Breakage can result in the adjacent covalent link between deoxyribose and phosphate. Alkylation also results in depurination of DNA because alkylated bases tend to be liberated by slow hydrolysis at the glycosidic linkage [61]. It has been postulated repeatedly [7, 38] that this could be a source of either transi­ tions or transversions, by incorporation of any base opposite the gap, in the new synthesized strand. However, some experimental facts strongly argue against this hypothesis. For instance, in bacteriophage T4, the highest mutation frequency is observed immediately after EMS treatment. If the phages are post incubated, lethality increases but there is not further in­ crease of mutations. 24 HESLOT

лт TTl 11 I I ГТГ■ Ж I I M I 1 I LL —=»" T Half-replication. Çar\(

TTTTTTTT nun llu

Sl'hjlt ftrttrwi loop

I IIIJLLUjrt

ConiKju-lnct : T -ft-* &-C tlraHSVirfion

FIG. 9. Origin of transversions.

CH, CH, NH CH.CH.CH, N ' г г г \c.HtCH*Ct

C H ,0

XCR _ i7o

Î C R _ 191

FIG.10. ICR compounds.

As transversions are undoubtedly induced by alkylating agents one must look for another explanation. As shown in Fig. 9, formation of a temporary single-strand loop at the site of replication could induce the formation of an abnormal pur-pur or pyr-pyr base pair, the source of a future tranversion. It may be that alkylating agents, by a still unknown mechanism, favour such a p ro c e ss. IAEA-SM-134/27 25

E. Acridines

Proflavine and numerous other acridines induce frameshift mutants in bacteriophages T 2 and T4. However, proflavine is not an effective mutagen in b a c te ria [63, 75]. On the other hand, a series of compounds (Fig. 10) synthesized by Creech et al. at the Institute for Cancer Research (the so-called ICR series) are mutagenic for bacteria and yeasts [4, 11]. These compounds possess an acridine, azaacridine or benzacridine ring to which is attached a side chain, often with alkylating properties. For instance, Ames and Whitfield [4] found that mutants induced in Salmonella by ICR-191 are reverted by the same compound, but insensitive to base analogues. ICR-170 efficiently reverses frameshift mutants of T4 Гц. Again Brammar et al. [11] showed that frameshift mutants affecting the A component of tryptophan synthetase were reversed by ICR-191. In the yeasts Saccharomyces cerevisiae and Schizosaccharomyces pombe a number of workers have shown that frameshift mutants could be induced by ICR-170 [78, 97]. Also in N. crassa, ICR-170 induces what appear to be frameshift mutants. They can be reversed by ICR-170, but not by nitrous acid or EMS. The consequences of a frameshift mutation at protein level are well characterized and in perfect agreement with the theory initially proposed by Brenner et al. [12] to explain the nature of mutations induced by acridines in bacteriophage T4. A base loss (-) can be nearly corrected by a base addition (+). The combination of three (-) mutants or of three (+) mutants also leads, in many cases, to a pseudo-wild type. Obviously, in the interval between a (+) and a (-) event, the amino-acid sequence will be changed. A pseudo-wild phenotype will only result if this sequence is not deleterious to the enzymatic activity. In this respect, the left end of the rnB cistron is especially tolerant. By studying the position on the genetic map of those (+ -) combinations that fail to achieve suppression, a few "barriers" have been located. Asymmetry is usually found with respect to a given "barrier": for instance, if a (-) to the left is not suppressed by a (+) to the right the (+-) combination will be effective. Several "barriers" have been studied in detail. Some contain chain- terminating codons, some correspond to unacceptable amino-acid residues. The asymmetry reported above is thus easily explained. As soon as the sequence of amino-acid residues in T4 lysozyme and in the A component of tryptophan synthetase were known, several teams of workers started to study the consequences of frameshift mutations. The technique consists in a determination of the amino-acid sequence in the affected region of the wild type end of the (+-) pseudo-wild type. Although the genetic code shows an extensive degeneracy, it has usually been possible to determine the m-RNA sequence by writing down all possible codons for each amino-acid residue concerned and looking for the frameshift which explains the difference between wild and pseudo-wild types. Figure 11 shows an example of such a determination. Similar work has been done by Brammar et al. [11] with the A component of tryptophan synthetase. L erm a n [62, 63] and Luzzati et al. [6 6 ] have studied the interaction of acridines with DNA by X-ray diffraction and other methods. They have 26 HESLOT

&c G-C a I I I I I I'l'l I I r'l I ГТ LL 1 I I I 11 1 I II I 1 1 1 1 CG CG I Break GC GC Т П I TT"I I I I I I I I I I I I 111 I 111 I I 1 1 1 с ь c g

"ГГ ПТгп mi I I I I I I I I I I I I I L-LLI 11 CG CG

GC TTTT M I I I I I I i i i I i CG CG-

G- с IP47 . _ Repair R e p a ir

G- C GC ...... тгптт т г п i 11 I ^ 1 1 1 1 11 1 1 i i i i i i m i i i i i 11 i 11 i i i i i i 11 C G CG

ADDITION OF FOUR d e l e t i o n o f F o u r BASES TO UPPER STRAND BAÍES i N UPPER STRAND

FIG .ll. Origin of ftameshift mutations by misrepair. (From Streisinger et al. [95] and Drake [28].)

WILD ТУРЕ «.+

.TUr _ Lyi- Scr_ Pro _ Str_ Leu _ Aih- flla. . ACK - ААры - A&U- CCA . UCA. COU _ AAU.GCX В л« loífc-

B a jc oAdAdi / ACX_ AApu.GUC.CAU.CAC.UUA. AUG.GCX ...... ,*T"kr _ Lyj _ Val _ Mù _ Hijf _ Lea - Ntb- Alft......

Double mutant e. x tiitv C-'1+>f)

FIG.12. Frameshift mutants in T4 lysozyme. (From Streisinger et al. [95].) IAEA-SM-134/27 27

concluded that molecules of certain acridines are capable of being inserted between two successive base pairs thus causing a partial unwinding of the double helix. The usual distance between two bases is 3.4 Â. When a mole­ cule of acridine is inserted this distance becomes 6.8 A. As thermal dénaturation of a DNA double helix does not eliminate acri­ dine binding [3 0, 79], it appears that intercalation occurs between adjacent bases rather than between base pairs. Several models have been proposed to explain the origin of frameshift mutants induced by acridines:

(1) Brenner et al. [12] thought that intercalation between successive bases could interfere with DNA replication and give rise to either loss or addition of one base pair. However, it has since been shown [2 7,, 95] that frameshift mutants corresponded very frequently to the simul­ taneous loss or gain of several base pairs. (2) Lerman [63] thought that acridine could induce unequal recombi­ nation. This model has been extensively tested in the T4 r n system but with negative results. However, in the yeast Saccharomyces cerevisiae a clear correlation has been found between recombination and acridine- induced m utations by Magni et al. [67, 6 8 ]. In fact, frameshift mutants arise spontaneously and are induced by acridines only during meiosis, but not at mitosis. (3) Streisinger et al. [95] supposed that frameshift mutations couldbe the consequence of mispairing errors during the enzymatic repair of single-strand breaks in a DNA double helix. Figure 12 shows that this model requires identical adjacent short sequences of bases to be present. It explains deletions or additions of one or several bases, as well as the correlation sometimes found between recombination and frameshift events.

Acridines would then act by stabilizing mispaired configurations. Although model (3) seems to be most appealing, difficulties remain. For instance, all acridines are not mutagenic and there is no clear corre­ lation between their mutagenic activity and their capacity to bind to DNA.

F. Ultra-violet light

Drake [24-26] has studied u. v. -induced mutations in bacteriophage T4. Approximately half of the induced mutants are reverted by acridines which shows that they are of the frameshift type. The other half are reverted by base analogues and therefore correspond to base-pair substitutions. How­ ever, as hydroxylamine is inefficient, it means that the original mutations were the result of G-C -» A-T transitions. With the A component of tryptophan synthetase in E_. coli, it could also be shown [8 ] that the G-C -* A-T transition was a common event. However, about 10% of the induced mutants were frameshifts. Smith [93] and Setlow [89, 90] have reviewed the effects of u.v. radiation on DNA. Pyrimidines appear to be much more sensitive than purines. There are two main classes of photoproducts: hydrates and dimers (Fig.13). Pyrimidines can fix a water molecule on the 5, 6 double bond to form hydrates. Thymine also gives rise to 5, 6-dihydrothymine. These compounds could be important in inducing mispairing and subsequent mutations. 28 HESLOT

tfH i 0 II C N CH*

С V T о г ¡ n e TM H V DRo H V t j r a t e -THYMÍIVE

THYMÍNE-CYT oSÍNE

d í m e r

FIG.13. Ultra-violet photoproducts.

Ultra-violet photoproducts in DNA also include pyrimidine dimers (mostly TT, but also TC and CC). of the cyclobutane type [10]. Usually they arise from a pair of adjacent pyrimidines in the same DNA strand. Other types of dimers have also been described, some of them possibly resulting in cross-linking between the strands of the DNA double helix. Ultra-violet effects have also been studied in vitro [47, 48, 108]. Usually poly-U directs the synthesis of polyphenylalanine, but if it has been first u.v. irradiated, serine is also incorporated. The overall priming ability is, however, reduced. These results can be interpreted as follows: (1) in­ activation of priming ability due to thymine dimers; (2) incorporation of ser in place of phe due to abnormal pairing of pyrimidine hydrates. Again, if one uses poly-C as a template for poly-G synthesis, u.v. i r r a ­ diation of poly-C reduces its priming ability. If one supplies ATP, it is incorporated. These results can be interpreted as above. A new light has been shed on the problem of u.v. -induced mutagenesis during the last decade when it was shown that microorganisms as well as eukaryotes were able to repair, by enzymatic mechanisms, the u.v. -induced damage in their DNA. This very important topic is treated below.

G. Ionizing radiations

Although the action of these radiations on nucleic acids has been studied for a long time by many workers [5, 21, 31-33, 60, 83, 84, 108], no complete picture of the phenomena is yet available. X-rays alter nucleic acids both directly and indirectly. In the direct effect the chemical bonds of the bases, of deoxyribose and of the sugar- phosphate linkages are ruptured. In the indirect effect, ionizing radiations produce free radicals either from water (H*, OH*) or from organic molecules. These free radicals attack the constituants of DNA. The destructive effects IAEA-SM-134Æ7 29

ad-3 A* ad-38* LINKAGE GROUP I ------О ------1 I---- 1 I------

TYPE OF EVENT LOCUS TYPE OF RECESSIVE LETHAL MUTATION POINT CHROMOSOME ad-ЗА ad-38 MUTATION DELETION

SINGLE-HIT —КТЖП — I \------» a d - 3 A "

— <— h— f77W7Ä----- od- з е *

TWO-HIT XHZZZa*4Z Z Z : ------o d - З А 1" OELETION

- C = H W 7 7 7 7 7 2 ^ ------~ а * - З В ,я OELETION

Ш 7 Т 7 7 Ш Ш Ш Ш _ ------~Kod-3A ad-38),R

OELETION

FIG.14. Events resulting in recessive lethal mutations in the ad-3 region in heterokaryotic conidia of a tv/o-component heterokaryon of Neurospora crassa« (From de Serres and Mailing [23].)

are greater for a given dose in the presence of oxygen. This seems to be due to an HO2 radical. It has been shown that one important reaction in water is the fixation of an OH* radical at the 5, 6 double bond of pyrimidines. Purines appear to be less sensitive. Under anaerobic conditions it is essentially the imidazole ring which is destroyed. In the presence of oxygen there is also a peroxidation of the 5, 6 double bond. Undamaged bases are also released from irradiated DNA as a result of chemical attack on the deoxyribose with subsequent splitting of the N-glycosidic bond [21 ]. Damage to the sugar leads, in general, to a break in the nucleo­ tide chain [21 ]. In short, the mechanisms are complex and incompletely elucidated. In the present state of knowledge it is still impossible to formulate a co­ herent molecular theory of the mutagenic action of ionizing radiations. The mutagenic effect of ionizing radiations has been the subject of ■ numerous investigations, especially in eukaryotes. It has been shown that they could induce point mutations as well as macrolesions (deletions, inver­ sions, translocations, etc.). One of the most detailed studies at the molecular level has been done in Neurospora crassa by de Serres, Mailing and Webber [22, 23, 69]. These authors have used the two closely linked genes, ad-ЗА and ad-3B, which control two successive steps in purine nucleotide biosynthesis. A mutation in either one of these two genes leads to the formation of a red 30 HESLOT

hist-2 tys-4 hist-3 oc/-3A od-3ß n ic-2 T O T A L S T R A IN S U S E D CISTRON CISTRON CISTRON OSTRON X REGION OSTRON OSTRON MUTANTS GENOTYPE ASTESTERS ЪЩ Г T ... — * '------p m - i h¡st-2* nie-?* 74-OR06-33A ------Ъ&Э, ------1 tyS - 4 * 3954-0R2-7A ----Г.....I'" I рУ-х| I I...... - 1 1 I ---Г 1 . Mst-3* 7 4 -0 R101-13Û 12-4-1059 12-7*108 12*9*65 12-9-167 12-7-104 12-9-88

од-ЗА*

{ad-ЗА od-3B)m {ad-3 A o d -3 6 nic-2)m

FIG.15. Complementation map of ad-3 and immediately adjacent regions showing the extent and type of functional inactivation in various types of ad-3 mutations induced in a dikaryon in relation to markers at other loci in the immediately adjacent regions. |xj[X] = recessive lethal damage supplementable on supplemented basal medium; ■ ■ = recessive lethal damagenon-supplementable on supplemented basal medium. Irregular ends indicate unknown limits. (From de Serres and Mailing [23] .)

pigment and to a requirement for adenine. The test system has been a heterokaryon ad-ЗА ad-313/++. The mutants recovered were shown to be: ( 1 ) point mutations resulting from single events within each locus, producing mutants of genotype ad-3AR and ad-3I3R. Such mutants increase linearly with the dose of X-rays; (2) deletions leading to a loss of either locus or both loci simultaneously. Such mutations are said to be of type ad-3A1R , ad- 3BIR or (ad-3.A ad-3B)IR, and increase as the square of the dose, implying that two independent events (breaks) are involved. These type (2) mutants show an important dose-rate effect implying the existence of a repair process (Fig. 14). As many markers are available both to the right and to the left of the ad-3A/ad-ЗВ region, the authors showed that the type (2) deletions not only covered the ad-3 region but also other adjacent loci. Figure 15 gives a precise map of these deletions. On the other hand, a sample of 33 ad-ЗВ mutants of type (1), induced by X-rays, were tested for specific revertibility. The mutagenic chemicals nitrosoguanidine (NG), orthomethyl hydroxylamine (OMHA) and ICR-170 were used for this purpose. The 33 mutants were classified as follows: IAEA-SM-134/27 31

Г 6% A-T - G-C Base-pair transitions ...... [21% G -C - A -T F ra m e s h if t...... 34%

Reverting only spontaneously 2 7 % Non-reverting ...... 0% Too leaky to te s t ...... 12%

From the above analysis it seems that point mutations induced by X-rays in this system are a consequence of two types of DNA alterations namely, damage or deletions of the bases.

3. REPAIR MECHANISMS

A close study of u.v. mutagenesis in 13. coli has shown that part of the damage to DNA could be repaired by specific enzymes [53, 54, 91, 110, 111 ]. All the strains of E. coli that have been studied so far show an equal susceptibility to the production of pyrimidine dimers in their DNA (about six dimers being formed per erg, per mm2, per chromosome), although the range of resistance varies greatly (by a factor of 1 to 2000), as shown in Table I. Three repair mechanisms are known in E. coli:

(1) P hotoreactivation

Exposure to visible light, after u.v. irradiation, results in a higher survival and a decreased mutation frequency. This effect is now very well understood. A single enzyme, coded by a single locus (phr + ), is responsible for splitting in situ the pyrimidine dimers. This enzyme requires visible light as an energy source to perform the splitting (Fig. 16-1).

TABLE I. U.V. SENSITIVITY OF E sch erich ia coli STRAINS

Ultra-violet dose for 3 l°Jo G enotype survival Characteristics Dimers per uvr-A rec-A ergs/m m 2 107 bases

+ + W ild type 500 3200

- + Excision defective 8 50

+ - Recombination deficient 3 20

- Excision defective and 0 .2 1 .3 recombination deficient 32 HESLOT

I . P HOTOR вЛС TI VATIO N *.

i I I I 9.....9 9.....9 Ç...... & ç ...... & iiw fY Pkob»r«Acfciv»h‘*« CHjyiwc. J - ....A 4 fi — * vft/biT V — * t .....^ A ...... T 3 A ...... T î .....t ?....t

2 -EVCÍSÍON- REPAIR :

I r...... r 1 « G-...... C 4.....Ф ç & ç....о ^ rT El,

З .Р о г т - replication R e p a i r ;

RifU'tab\»n

Sl'stltr fhr-Ahtitf fi,* c k û c

ï i ’ n e R

FIG. 16. Repair mechanisms.

TABLE II. MUTABILITY OF Escherichia coli STRAINS

Frequency of No. of dimers N o. of mutation to Genotype Dose produced per dim ers streptomycin E. coli chromosome excised resistance per 107 survivors

u vr-A + 20 ergs/frim2 120 Practically all 0

uvr-A 20 ergs/frim2 120 None 200 IAEA-SM-134/27 33

b o o QJ CL.

о -и >

д + . г е с - A

-й X г е с - A " £

0 ------X----- *------X------*------X------sf------> 0 2o U-o 6o

UV doS«.(trgS J v n v r J 1 )

FIG. 17. Induction of mutations by u.v. radiation in rec-A+ and rec-A~ strains.

(2) E xcision re p a ir

In this process the dimers are removed from DNA by the introduction of two single-strands breaks on either side of the dimer. The corresponding oligonucleotide, comprising the dimer and a few other bases, is released. The resulting gap is then repaired by a DNA polymerase, utilizing as template the intact strand opposite the gap (Fig. 16-2). An important point is that excision repair is not specific for pyrimidine dimers: Alterations produced by nitrogen mustard, 4-nitro-quinoline and nitrous acid are reparable by the same mechanism.

(3) P o st-rep licatio n re p a ir

Howard-Flanders et al. [54] have shown that, if the dimers are not removed by excision, the DNA strands synthesized after u.v. irradiation are discontinuous. They have a gap at the level of each dimer (Fig. 16-3). Later these gaps are closed by a post-replication repair mechanism in­ volving a recombinational process between daughter strands. E . coli mutants at the uvr-A locus lack excision ability. The same u.v. dose applied to uvr-A' cells gives a much higher mutation frequency than in uvr-A* cells. We can therefore conclude that the excision of a dimer is much less liable to cause a mutation than if it remains unchanged in the DNA (Table II). Other experiments show that the excision-repair mechanism is very accurate indeed: at the level of the locus controlling sensitivity/resistance to streptomycin it can repair about 1 0 6 dimers without causing a single mutation. When the locus rec-A+ mutates to rec-A one observes an increased sensitivity to u. v. light, X-rays and nitrosoguanidine (Table I). The rec-A~ mutants exhibit also a reduced recombination ability. 34 HESLOT

. г ' 5 '

Specific endohucle ал

S'. 3 ' .

■ S'

L í g a s e

s ', . i ' г ' . S'

• X > i^eR

FIG. 18. Enzymatic mechanism of excision-repair.

A comparison of u.v. -induced mutations in rec-A~ and rec-A+ strains indicates that no mutants can be induced in the first Ones (Fig. 17). This supports the conclusion, already expressed, that excision-repair is a very accurate process and that u.v. -induced mutations are probably due to errors in recombinational repair of gaps in daughter strands opposite unexcised pyrimidine dimers. At present we know nothing about this error inducing repair, but Witkin [110, 111] has suggested that it could possibly involve enzymatic modifications of an end base in the region of the gap as a preliminary step to stop degradation by exonucleases and prepare it for recombination. This altered base could have modified pairing specificity so that the probability of errors would be greatly increased. Until recently it was supposed that at least four enzymes were involved in excision-repair: ( 1 ) a specific endonuclease introducing a single-strand break near the pyrimidine dimer; (2) DNA polymerase to fill the gap; (3) a second specific endonuclease to remove the oligonucleotide containing the dimer; and (4) one ligase to seal t.he new synthesized strand with the old one. IAEA-SM-134/27 35

UV Dose (fR K /|Я*! )

N î t r o o e n m u s t a M ) d o j e ( ‘• y / w £ )

FIG.19. Non-specificity of repair process as shown by the survival curves of the two E. coli strains Bs_1 and B/r after treatment with u.v. radiation and nitrogen mustard. (From Hanawalt, P.C ., Haynes, R.H., Scientific American 216 2 (1967) 41. • - • Ultra-violet X ----- X Nitrogen mustard.

TABLE III. ANTIMUTATOR SUPPRESSION OF BASE-ANALOGUE MUTAGENESIS

R evertants per 10 progeny particles Antimutator M utagen a lle le j _ m 13 ± U V 363 j_ UV 183 r UV 199

None 1 6 . 8 .5 9 .2 4 .0

ts+ 5 BU 4100 1700 67 60

2 AP 14 23 830 260

None 6 .4 7 .7 0 .4 0 .1

ts_CB 87 5 BU 120 2100 0 .2 0 .5

2 AP 3 6 .9 1 .1 23

None 7 .0 4 .9 0 .5 0 .0 3

ts^CB 120 5 BU 59 44 1 .9 0 .5

2 AP ■12 6 .8 52 0 .7

Mutantsj^UV 13 andxUV 363 revert by G-С-*- А-T transitions. Mutants r^UV 183 and_r_UV 199 revert by A-T-* G-С transitions. (From Drake and Greening [29]). 36 HESLOT lAEA-SM-134/27 37

Now, the ligase has been purified over 1000-fold from E_. coli. Within a single strand, hydrogen bonded to a template strand, it joins a 5' -phosphoribosyl terminus to a juxtaposed З'-hydroxyl terminus. On the other hand, Kelly et al. [55] have shown that the well-known DNA polymerase is able to act as a repair enzyme on a u. v. -irradiated DNA containing pyrimidine dimers (Fig. 18). In this new scheme only three enzymes are needed to effect excision-repair. This review has treated at length the case of u. v. -induced damage because it is the best known regarding repair processes. However, there is frequently a close parallelism between the sensitivities of a given bacterial strain to both u.v. light and alkylating agents (Fig. 19). This means that some repair enzymes must be able to detect rather unspecific alterations in DNA, and to correct them. Obviously, the discovery of enzymatic repair mechanisms of damaged DNA gives a new dimension to the mutagenic process.

4. MUTATOR AND ANTIMUTATOR GENES

A remarkable mutator gene (mut T) has been found in E_. coli [6 , 105]. It acts in a very specific way that the tryptophan synthetase A component helped to elucidate [19, 112]. The amino-acid substitutions observed can only be explained by A-T->G-C transversions. The spontaneous mutation frequency is increased, for all genes, by afactor of X 1000. After some 1500 bacterial generations, the frequency of G-С base pairs in DNA is sufficiently increased to alter its density. The mode of action of mut T is still unknown, but it could possibly alter the functioning of DNA polymerase. Other mutator genes have been described in E. coli [92] and also one in Salmonella typhimurium [73]. Another interesting system concerns bacteriophage T 4 . Temperature- sensitive mutants of gene 43, which is the structural gene for T4 DNA poly­ merase, show an increased frequency of mutants arising from both transi­ tions and transversions. Drake and Greening [29] have recently shown that two mutants of gene 43, i.e. t£ CB 87 and ts CB 120, exhibited a strong antimutator activity. They decrease the frequency of both spontaneous and induced (BU, AP) transitions. The antimutator alleles suppress both errors of replication and errors of incorporation, as indicated by the suppression of both types of BU-induced transitions (Table III). These results can only be accounted for if DNA polymerase plays an active role in base selection during the replication process. Drake and Greening [29] assume that DNA polymerase contains binding sites for both parental-strand and progeny-strand bases. Mutations altering these binding sites could result in either an increased or a decreased ac­ curacy of replication. Kornberg [58] shares a different point of view and assumes that pairing between bases occurs first, the DNA polymerase then accepting or rejecting the new inserted residue. These findings show that a better understanding of the mutagenic process would certainly result from a close study of the behaviour of DNA polymerase. This is a very important area of investigation for the future. The discovery of the antimutator alleles of DNA polymerase raises a fundamental problem, i.e. why a more accurate type of enzyme has not been 38 HESLOT retained during the course of evolution? Probably an increase in the ac­ curacy of replication has some sort of selective disadvantage.

5. SUMMARY AND CONCLUSIONS

Table IV summarizes the main effects of chemical and physical mutagens with regard to their primary interaction with DNA and their mutagenic consequences. From a practical point of view, it is not so easy to give precise rules for the proper choice of a mutagen. Although base analogues and hydroxylamine are the most specific of all mutagens, they are rarely used in practice because they are not always very effective. Nitrous acid and alkylating agents, on the other hand, induce mutations in a very large range of organisms, but give rise to a variety of genetic lesions. The acridines of the ICR series are probably worthy of more attention, if one looks for completely blocked mutants. Ultra-violet light is quite effective and often used. Ionizing radiations remain the best source of macrolesions, especially deletions, which have the advantage of non-revertibility, but the disadvantage of possibly affecting several adjacent genes. If one wishes to induce a complete block in a biochemical pathway, acridines and ionizing radiations can be recommended. If, on the other hand, feedback mutants are looked for it will be advisable to choose a mutagen that preferentially induces missense mutations. From a more general point of view, future investigations in experi­ mental mutagenesis should be directed towards a better understanding of the eukaryotic chromosomes. They are not naked DNA, but a complex with nucleohistones and other proteins. It is to be expected that mutagenic agents also react with these protein components and that this is probably the source of significant side effects. Another important area of investigation is the mutagenic effect of altered DNA polymerases. This could possibly lead to a new technology in the field of mutagenesis.

REFERENCES

[1] ADELBERG, E .A ., MANDEL, M ., CHEN, G .C .C ., B iochem .biophys. Res .C o m m u n . J ji ( 1965) 788. [2] ALEXANDER. P., Nature Í69 ( 1952) 226. [3] ALEXANDER, P., STACEY, K.A., Ann.N.Y. Acad.Sci. 68(1958) 1225. [4] AMES, B.N.,. WHITFIELD, H .]., Cold Spring Harbor Quant.Biol. 31 (1966) 221. [5] APELGOT, S.. ECKERT, B., J.Chim.phys. (1963) 505. [6] BACON, D .F ., TREFFERS, H .P ., J.B a c te rio l. 81 (1961) 786. [7] BAUTZ, E ., FREESE, E ., Proc. n atn . A cad. S e i.. USA 46 ( 1960) 1585. [8] BAUTZ-FREESE, E ., FREESE, E ., V irology ¿3 (1961) 19. [9] BECKER, E.F., ZIMMERMAN, В.K., GEIDUSCHEK, E.P., J.molec.Biol. 8 (1964) 377. [10] BEUKERS, R .,' BERENDS, W ., B iochim . bioph y s.A cta 41 ( 1960) 550. [11] BRAMMAR, W .J., BERGER, H., YANOFSKY, C., Proc.natn. Acad. Sei. USA 58 (1967) 1499. [12] BRENNER, S ., BARNETT, L ., CRICK, F .H .C ., ORGEL, A ., J .m o le c .B io l. 3 (1 9 6 1 ) 121. [13] BRENNER, S ., STRETTTON, A .D . W ., KAPLAN, S ., N ature 206 (1965) 994. IAEA-SM-134/27 39

[14 BRENNER, S.. BARNETT, L., KATZ, E.R., CRICK, F.H .C ., Nature 213 (1967) 449. [15 CERDA-OLMEDO, E., HANAWALT, P.C ., Biochim.biophys. Acta 142 (1967) 450. [16 CERDA-OLMEDO, E., HANAWALT, P.C ., Molec. Gen.Genetics 101 (1968) 191. [17 CERDA'-OLMEDO, E., HANAWALT, P.C., GUEROLA, N .. J.molec.Biol. 33 (1968) 705. [18 CHAMPE, S.P., BENZER, S., Proc.natn. Acad.Sei. USA 48 (1962) 532. [19 COX, E.C., YANOFSKY, C .. Proc.natn. Acad.Sei. USA 58 ( 1967) 1895. [20 DAVIS, R. W., DAVIDSON, N.. Proc.natn.Acad.Sei. USA 60 (1968) 243. [21 DERTINGER, H ., JUNG, H., Molecular Radiation Biology, Springer Verlag (1970) 237 pp. [22 De SERRES, F.J., MALLING, H .V ., WEBBER, B.B., Brookhaven Symp.in Biology 20 (1967) 56. [23 De SERRES, F.J., MALLING, H .V ., Japan J.Genetics 44 Suppl. 1 (1969) 106. [24 DRAKE, J.W ., J.molec.Biol. 6 (1963) 268. [25 DRAKE, J.W ., J.Bacteriol. 91 (1966) 1775. [26 DRAKE, J.W ., J.Bacteriol. 92(1966) 144. [27 DRAKE, J.W ., Proc.natn.Acad.Sei. USA 55(1966) 506. [28 DRAKE, J.W ., The Molecular Basis of Mutation, Holden-Day (1970) 273 pp. [29 DRAKE, J.W ., GREENING, E.U., Proc.natn. Acad. Sei. USA 66 No.3 (1970) 823. [30 DRUMMOND, D .S., SIMPSON-GILDENEMEISTER, V .F.W ., PEACOKE, A.R., Biopolymers 3 (1965) 135. [31 ECKERT, B., MONIER, R., Nature (London) 1J34 (1959) 58. [32 ECKERT. B.. Nature (London) 194 (1962) 278. [33 EMERSON, P.. SCHOLES, G., THOMPSON, D .H., WARD, J.F., WEISS, J., Nature (London) 187 (1960) 319. [34 FREESE, E ., P ro c .n a tn .A c a d .S e i. USA 45 (1959) 622. [35 FREESE. E .. J .m o le c .B io l. 1 (1959) 87. [36 FREESE, E ., Brookhaven S y m p .in Biology 12 (1959) 63. [37 FREESE, E ., M olecular G enetics I . , A cadem ic Press (1963) 207. [38 FREESE, E .B ., P ro c .n a tn . A c a d .S ei. USA 47 (1961) 540. [39 FREESE, E .B .', M utat. Res. 5 ( 1968) 299. [40 FREESE, E ., BAUTZ, E ., FREESE, E .B .. P ro c .n a tn . A cad. S ei. USA 47 (1961) 845. [41 FREESE, E .. BAUTZ-FREESE, E ., BAUTZ, E ., J .m o le c .B io l. 3 (1961) 133. [42 FREESE, E .B ., FREESE. E ., P ro c .n a tn .A c a d .S ei. USA 5 2 (1964) 1289. [43 FREESE. E ., FREESE, E .B ., B iochem . 4 (1965) 2419. [44 FREESE, E ., FREESE, E .B ., GRAHAM, S ., B io chim .biophys. A cta 123 (1966) 17. [45 FREESE, E .. STACK, H .B ., Proc. n a tn . A cad . S e i. USA 48 (1962) 1796. [46 GEIDUSCHEK, E.P., Proc.natn. Acad.Sei. USA 47(1961) 950. [47 GROSSMAN, L.. Proc.natn. Acad.Sei. USA 48 (1962) 1609. [48 GROSSMAN, L., Proc.natn.Acad.Sei. USA 50 (1963) 657. [49 HESLOT, H., Abhandl.Deutsch. Akad.Wiss., Berlin (1962) 193. [50 HESLOT, H ., "The nature of mutations", The Use of induced mutations in Plant Breeding, Rep.FAO/IAEA Technical Meeting (Rome, 1964), Pergamon Press Ltd (1965) 3. [51 HESLOT, H., Manual on Mutation Breeding, Tech.Rep. Series 119, IAEA, Vienna (1970) 53. [52 HOWARD, B.D., TESSMAN, I., J.molec.Biol. 9 (1964) 364. [53 HOWARD-FLANDERS, P., Ann.Rev.Biochem. 37 (1968) 175. [54 HOWARD-FLANDERS, P., RUPP, M .D., WILKINS, B.M., Replication and Recombination of Genetic Material (PEACOCK, W .J., BROCK, R.D., Eds). Canberra (1968) 142. [55 KELLY, R .B ., ATKINSON, M .R ., HUBERMAN, J .A .. KORNBERG, A .. N ature 224 (1969) 495. [56 KAUDEWrrZ, F., Nature 83(1959) 1829. [57 KAUDEWITZ. F., Z.Naturforsch. 146(1959) 528. [58 KORNBERG, A .. Science 163 (1969) 1410. [59 KRIEG, D.R.. Genetics 48 (1963) 561. [60 LATARJET, R., ECKERT, B., APELGOT, S., REBEYROTTE, N.. J.Chim.phys. (1961) 1046. [61 LAWLEY, P.D ., Prog.Nucleic Acid Res.Mol.Biol. _5 (1966) 89. [62 LERMAN, L.S., J.molec.Biol. 3(1961) 18. [63 LERMAN, L.S., Proc.natn. Acad.Sei. USA 49 (1963) 94. [64 LITMAN, R., EPHRUSSI-TAYLOR, H., C.R.Acad.Sei.Paris 249(1959) 838. [65 LOVELESS, A ., Genetic and Allied Effects of Alkylating Agents, Pennsylvania State University Press (1966). [66 LUZZATI, V ., MASSON, F., LERMAN, L.S., J.molec.Biol. 3 (1961) 634. [67 MAGNI, G.E., Proc.natn.Acad.Sei. USA 50 (1963) 975. [68 MAGNI, G.E., vonBORSTEL. R.C., SORA. S., Mutat.Res. 1 (1964) 227. 40 HESLOT

[69] MALL ING, H.V., de SERRES,F.J., Radiat.Res. 31 (1967) 637. [70] MALL ING, H.V., de SERRES,F.J., Mutat. Res. 5 (1968) 359. [71] MALLING, H.V.. de SERRES, F.J.. Mutat.Res. 6 (1968) 181. [72] MICHELSON, A .M ., GRUNBERG-MANAGO, M ., Biochim.biophys. Acta 91 (1964) 92. [73] MIYAKE, T ., Genetics 45 (1960) 11. [74] MUNDRY, K. W., GIERER, A., Z.Vererbungslehre 89 ( 1958) 614. [75] ORGEL, L .E ., A dv. Enzym ology 27 ( 1965) 289. [76] PHILLIPS, J.H ., BROWN, D .M ., Prog.Nucleic Acid Res. M ol.Biol. 7 (1967) 349. [77] PHILLIPS, J.H ., BROWN, D.M ., ADMAN, R., GROSSMAN, L., J.m olec.Biol. 12 (1965) 816. [78] PHILLIPS, J.H ., BROWN, D.M ., GROSSMAN, L.. J.molec.Biol. 21 ( 1966) 405. [79] PRITCHARD, N .J., BLAKE, A., PEACOKE, A.R., Nature 212 (1966) 1360. [80] REINER, B ., ZAMENHOF, S ., J .b io l.C h e m . 228 (1957) 475. [81] RICHARDSON, C .C ., SCHILDKRAUT, C .L ., KORNBERG A ., Cold Spring H arbor S y m p .Q u an t.B io l. 28 (1963) 9. [82] ROSS, W .C.J., Biological Alkylating Agents, Butterworths (1962). [83] SCHOLES, G., WARD, J.F., WEISS, J., J.molec.Biol. 2 (1960) 379. [84] SCHOLES, G., WEISS, J., Abhandl.Deutsch. Akad. Wiss.Berlin 1Д1962) 94. [85] SCHUSTER, H., VIELMETTER, W., J.Chim.phys. 58 (1961) 1005. [86] SCHUSTER, H., SCHRAMM, G., Z. Naturforsch. 13B (1958) 697. [87] SCHUSTER, H., Z.Naturforsch. 15B ( 1960) 298. [88] SCHUSTER, H., Biochim. biophys.Res.Commun. 2 (1960) 320. [89] SETLOW, J.K ., in Current Topics in Radiation Research, V ol.2, North Holland (1936) 195. [90] SETLOW, J.K .. in Comprehensive Biochemistry, V ol.27 (1967) 157. [91] SETLOW, R.В., Prog.Nucleic Acid Res.Mol.Biol. 8 (1968) 257. [92] SIEGEL, E .C .. BRYSON, V ., J.B a c te rio l. 94 (1967) 38. [93] SMITH, K.C., Radiat.Res.Suppl. 6 (1966) 54. [94] STACEY, K.A., COBB, M., COUSENS, S.F., ALEXANDER, P., Ann.N. Y. Acad. Sei. 68 (1958) 682. [95] ST REIS INGER, G., OKADA, Y., EMRICH, I., NEWTON, TSUGITA, A., TERZAGHI, E., INOUYE, M. Cold Spring Harbor Symp.Quant.Biol. 31 (1966) 77. [96] STRELZOFF, E., Biochim. biophys. Res. Commun. ^ (1961) 384. [97] STRELZOFF, E., Z.Vererbungsslehre 93 ( 1962) 301. [98] STRELZOFF, E., RYAN, F.J., Biochim.biophys.Res.Commun. 1_( 1962) 471. [99] STRETTON, A .O.W ., KAPLAN, S., BRENNER, S., Cold Spring Harbor Symp.Quant.Biol. 31 (1966) 173. [100] TERZAGHI, B.E., STREISINGER, G., STAHL, F. W., Proc.natn. Acad.Sei. USA 48 (1962) 1519. [101] TESSMAN, I., Virology-9 (1959) 375. [102] TESSMAN, I., J.molec.Biol. 5(1962) 442. [103] TESSMANN, I., PODDAR, R.K., KUMAR, S., J.molec.Biol. 9( 1964) 352. [104] TRAUTNER, T .A ., SWARTS, M .N ., KORNBERG, A ., P ro c .n a tn . A c a d .S e i. USA. 48 (1962) 4 49. [105] TREFFERS, H .P ., SPINELLI, V ., BELSER, N .O ., P ro c .n a tn . A cad. S ei. USA 40 (1954) 1064. [106] VIELMETTER, W., SCHUSTER, H., Biochim.biophys.Res.Commun. 2 (1960) 324. [107] VIELMETTER, W., SCHUSTER, H., Z.Naturforsch. 15B (1960) 304. [108] WACKER, A., JACHERTS, J.D ., JACHERTS, B., Angew. Chem. Int. Edit. 1 (1962) 509. [109] WESTMORLAND, B.C., SZYBALSKI, W., RIS, H., Science 163 (1969) 1343. [110] WITKIN, E.M ., Proc.XII Int.Congress Genetics 3 (1969) 225. [111] WITKIN, E.M ., Ann.Rev.Genetics 3 (1969) 525. [112] YANOFSKY, C ., COX, E.C.. HORN, V ., Proc.natn. Acad.Sei. USA 55 (1966) 274. [113] YANOFSKY. C ., ITO, J., HORN, V., Cold Spring Harbor Symp.Quant.Biol. 31 (1966) 151.

DISCUSSION

N. K. NOTANI: Why are some thymine dimers not removed by the excision mechanism but left to be taken care of by the post-replication repair mechanism? H. HESLOT: Thymine dimers are usually excised after u.v. -irradiation. However, depending on the conditions under which irradiated cells are IAEA-SM-134/27 41 incubated, some of these dimers may be left. The most efficient method of studying post-replication repair is, of course, partial or total in­ activation of the excision system. H. ALTMANN: Have you any experimental evidence to show that the changed DNA produced by the intercalating substances of the ICR series can be repaired by the dark repair system, and does the chloride in the side group also interact with DNA constituents? H. HESLOT: I don't remember any investigation dealing with the repair of mutagenic damage induced by the ICR compounds. If any such investigation was published, it may have escaped my notice. The alkylating side-chain (-N-CH 2CH2C1) can react with DNA or the associated proteins. There is a possibility for the acridine ring to be intercalated subsequently into the DNA double helix. H. ALTMANN: In connection with Fig. 11 of your paper, I have a comment regarding your speculation on the mechanism of manifestation of mutations. Deletion of several bases on the right side of this scheme can lead to a mutation even if the loop produced is excised by the repair enzyme system. But addition of several bases could lead to a mutation only if the feature is produced during or just before the beginning of DNA synthesis, so that the repair system cannot excise the loop produced. H. HESLOT: As regards the origin of frameshift mutations by mis- repair, the schemes shown in Fig. 11 have been suggested by Streisinger et al. [95] andaré, for the time being, purely theoretical. J. MEYRATH: Is anything known about the mechanism of reactivation (repair mechanism) in dormant cells, such as spores or conidia? H. HESLOT: I don't recall at the moment whether investigations along this line have been carried out. J. MEYRATH: Can any conclusion be drawn from the shape of the survival curves regarding the mechanism of repair? H. HESLOT: Repair mechanisms vary in different organisms, and photo-reactivation is sometimes completely lacking. This point can be cleared up by plotting u.v. -survival curves under both sets of conditions (with or without visible light after treatment). MARCELLE PEYRE: Has any study been made of the action of the ICR compounds on different microorganisms belonging to the family Enterobacteriaceae, for example, Mycobacteria, Streptomyces and various fungi, such as Penicillia. H. HESLOT: So far as I know, the action of the ICR compounds has been studied on Neurospora crassa, Saccharomyces cerevisiae and Schizosaccharomyces pombe. I am unable, at the moment, to say whether any work has been carried out on Penicillia. J. A. ROPER: In the earliest studies of acridine-induced mutations it was suggested that meiosis, or some other recombination process, was necessary for mutation. Has this idea been discarded? H. HESLOT: Magni and von Borstel have shown that a clear correlation exists between meiosis and acridine-induced mutations in yeasts. This is not incompatible with the scheme proposed by Streisinger because at least some of the repair enzymes must be operative to perform re­ combination at meiosis. V. V. SUKHODOLETS: To what extent can one distinguish between the excision repair and post-replication repair mechanisms? 42 HESLOT

H. HESLOT: The two mechanisms "excision repair" and "post­ replication repair" can be most easily recognized when one or the other is inactivated by mutation. When both systems are lacking, the bacterial cell is killed by a single thymine dimer per chromosome. S. I. ALIKHANIAN: Did I correctly understand your remarks regarding the specificity of mutagenesis - that this problem concerns only the eukaryotes? H. HESLOT: To my mind, the specificity of mutagenesis concerns both prokaryotes and eukaryotes. What I wanted to say is that the close association of DNA with specific proteins in eukaryotes is likely to interfere with the induction of mutations. S. I. ALIKHANIAN: I personally think that the problem of specificity of mutagenesis, as stated, does not exist. We may speak about selective mutagenesis or controlled mutagenesis but not about specificity. I shall however dwell upon this matter later in my second paper (SM-134/25). H. HESLOT: The first step towards increased specificity is perhaps the use of nitroso-guanidine, for short periods, on synchronized bacterial cells. Achieving complete specificity is a goal for the future. J. A. ROPER: It may be worthwhile mentioning the selection of particular mutant types by techniques other than that of mutagen specificity. In particular, Professor Auerbach and colleagues have been concerned with the events occurring between production of the DNA lesion and the actual recovery of the mutant. Different plating media may select, preferentially, different classes of mutants. In some cases, this may appear at first sight as mutagen specificity. As far as I know, there has been no attempt to utilize this approach for practical purposes. H. HESLOT: I entirely agree with you, but this is what may be called false mutagenic specificity. True specificity would involve mutating, at will, one (or a few) selected gene(s). IAEA-SM-134/3

BASIS FOR RADIOSENSITIVITY OF" SOME MUTANTS OF Hemophilus influenzae

N.K. NOTANI, V.R. JOSHI, A.R. GOPAL-AYENGAR Bhabha Atomic Research Centre, Trombay, Bombay, India

Abstract

BASIS FOR RADIOSENSITIVITY OF SOME MUTANTS OF Hemophilus influenzae. Four u.v.-sensitive mutants, including one that was also sensitive to gamma radiation, of Hemophilus influenzae were isolated and characterized in regard to their ability (i) to form colonies after u.v. - and gamma-irradiation, (ii) to produce transformants from uptake of genetically marked, irradiated DNA, and (iii) to yield phage progeny when infected with phage HPt C j. Mutants N12, N17, N19 and N21 showed, respectively, a 20, 18, 2 and 20-fold greater sensitivity than the wild type. Mutants N12 and N21 are also defective in the repair of extra-cellularly irradiated transforming DNA. Mutants N17 and N21 have a lower capacity than the wild type to do the host-cell reactivation of irradiated phage. Measurement of thymine dimers in cellular DNA following u.v. -irradiation showed that mutants N12 and N21 are defective in the repair mechanism of excising thymine dimers. Mutant N17, although normal in regard to thymine-dimer excision, was found to be Slow in rejoining of the DNA breaks. Coincidentally, mutant N17 is also somewhat sensitive to gamma radiation. The basis for the sensitivity of mutant N19 is not understood at present, but it is observed to have a most unusual property of giving differential transformation for two-linked markers Sr (resistance to streptomycin) and Cr (resistance to cathomycin) following uptake of unirradiated DNA; the transformation for Cr is about l/100th that of Sr. The uptake of gamma-irradiated DNA (in buffer) by competent cells is near-normal but the integration of input DNA into the resident DNA is reduced. Furthermore, the reisolated, unintegrated irradiated input DNA is observed to have lower average molecular weight than unirradiated input DNA. The inactivation of biological activity of transforming DNA, after gamma irradiation, is thus correlated with strand breakage.

INTRODUCTION

Elucidation of the steps involved in the repair of radiation-damaged deoxyribonucleic acid (DNA) is greatly facilitated by an analysis of the mutants which are sensitive to radiations. The mutant strains are sensitive presumably because of their inability to catalyze a reaction in the sequence of repair steps. The Hemophilus influenzae transforming system was first used by Setlow et al. [1] for this purpose. With this system it is possible to study the repair of bacterial DNA and infecting phage DNA as well as that of transforming DNA and transfecting (phage) DNA. Several u.v.-sensitive mutants were isolated and characterized. The present report describes the basis for the radiation sensitivity of four of these mutants.

CHARACTERIZATION OF MUTANTS

Four u.v.-sensitive mutants which have been studied in detail were characterized in regard to their ability (i) to form colonies after u.v. irradiation, (ii) to produce transformants from uptake of genetically- marked irradiated DNA, and (iii) to yield phage progeny when infected with irradiated Hemophilus phage сд (HPi£i).

43 44 NOTANI et al.

FIG. 1. Ultra-violet inactivation of wild-type Hemophilus influenzae and three mutants N12, N17 and N21.

TABLE I. U.V. SENSITIVITY OF Hemophilus influenzae TRANSFORMING DNA (C r 25 MARKER) ON WILD TYPE AND TWO U.V.-SENSITIVE MUTANTS

VNo/N-1 for a VNo/N-1 for a dose of 1200 ergs/mm2 dose of 3000 ergs/mm2

W ild type 4 .0 12.0

N12 8 .7 2 9 .0 IAEA-SM-134/3 45

FIG.2. Ultra-violet inactivation of H P^ ; plated on wild-type Hemophilus influenzae and two u.v.-sensitive mutants.

Figure 1 shows the ability of the wild type and mutant strains N12, N17 and N21 to form colonies after u.v. irradiation. All the three mutants are about 18 to 2 0 times more sensitive than the wild type. Sensitivity of mutant N19 was about twice that of the wild type (not shown in the figure). Mutants N12 and N21 are also defective in the repair of extra- cellularly irradiated transforming DNA (Table I). The square root plot of No/N (where No is the number of transformants resulting from u.v.- irradiated DNA and N is the number after irradiation) has been found to be useful because a plot of the data in this manner gives a straight line [1 , 2 ] and therefore the relative sensitivities to u.v. radiation of dif­ ferent strains can be measured from the slopes. Under given ex­ perimental conditions, repeated determinations show little variation [1 ]. 46 NOTANI et al.

It provided, therefore, an opportunity to compare our results with those of Setlow et al. [1]. The value of JNo/N - 1 of 12 at 3000 ergs/m m 2 obtained by us for the wild type comes very close to a value of 13 at 2940 ergs/mm 2 reported by Setlow et al. [1]. The ratios of sensitivities of N12/wild type and N21 /wild type are calculated at 2.18 and 3.7 5, respectively. N21 is found to be the most sensitive for reactivation of u.v.-irradiated transforming DNA. Figure 2 shows dose-effect curves for HPi£i plated on wild type and mutants N17 and N21. It is quite clear that mutants N17 and N21 have a lower capacity than the wild type to do the host-cell reactivation. Both mutant strains N12 and N19 were similar to the wild type in this resp ect.

MOLECULAR BASIS FOR RADIOSENSITIVITY OF MUTANT STRAINS

Ultra-violet irradiation of bacterial cells produces many physico­ chemical changes in their DNA. One of these changes, namely the thymine dimers, seems to cause most of the lethal effects observed in some strains of bacteria at low doses [3]. Excision repair is a mechanism whereby pyrimidine dimers may be eliminated from DNA Г4, 5]. Some classes of mutant strains could be sensitive because of their inability to excise thymine dimers. All four u.v.-sensitive and wild-type strains were labelled with 3 H-thymidine, irradiated with u.v., incubated after irradiation for various times, and washed and fractionated into acid soluble and insoluble parts. Hydrolysis and paper chromatography of the samples was carried out as described by Carrier and Setlow [6 ]. Figure 3 gives the presence of thymidine dimers (acid insoluble) in DNA of irradiated cells as a func­ tion of post-irradiation incubation time. It is observed that the wild type consistently eliminates thymine dimers from its DNA, whereas in both mutants N12 and N21 the number of thymine dimers remains un­ changed. Both mutants N17 and N19 excise thymine dimers normally. Since mutant N17 was found to be normal in excising thymine dimers it was thought conceivable that it may be defective in joining the excision breaks. In fact, a mutant of this type has been reported [7]. Irradiated 3 H-labelled cells were incubated for different times and then lysed gently and sedimented through alkaline sucrose gradients. Figure 4 shows the sedimentation profiles of wild-type (Rd) and N17 lysates at various times after irradiation. 1 2 0 minutes after irradiation, there is a con­ siderably larger amount of slower-sedimenting material present in N17 than in the wild type. We interpret this to mean that mutant N17 is defective in the rate of repair of excision breaks. Mutant N19 has only about two-fold greater sensitivity to u.v. radia­ tion than the wild type. It was found to possess a normal system of excision of thymine dimers. The basis for its radiation sensitivity remains to be determined but it has disclosed an unusual property with unirradiated DNA. It is known that some u.v.-sensitive mutants are also recombination- defective. All the u.v.-sensitive mutants were tested for transformation with unirradiated DNA. When transformation was done with SrCr (streptomycin resistance and cathomycin resistance) DNA (in N19), Sr transformation was about half that of the wild type but transformation IAEA-SM-134/3 47

M INUTES AFTER U.V

FIG.3. Ultra-violet-induced thymine-containing dimers in the TCA-insoluble fraction of the wild type and mutants N12 and N21 of Hemophilus influenzae as a function of time of post-irradiation incubation. Cells given a dose of 100 ergs/mm г. to Cr was almost one-hundredth of the wild type (Table II). Markers Sr and Cr are linked in transformation. Since the difference is observed at the marker level, it is presumed that mutant N19 has an enzyme which selectively attacks certain base sequences. Its relevance for slight u.v. sensitivity of this mutant remains to be determined.

FATE OF GAMMA-IRRADIATED TRANSFORMING DNA

Irradiation of transforming DNA by gamma rays lowers its bio­ logical activity. Whatever may be the nature of the lesion, it does not seem to affect the ability of irradiated DNA to be taken up irreversibly by competent cells (Table III). These findings are in agreement with those of Randolph and Setlow [ 8 ]. Their additional observations have shown that irradiation of transforming DNA with X-rays induces many single and double-strand breaks and that transforming DNA is inactivated primarily because integration of DNA is prevented as a result of the production of double-strand breaks. We have extended these studies using 3 2 P-labelled DNA and find that the incorporation of the 32P label in the recipient DNA is approximately similar from both irradiated and unirradiated DNA during transformation. However, the specific biological activity is very much lower with ir­ radiated DNA (Table IV). One interpretation of this finding is that ir­ radiated DNA undergoes more than the usual degradation and the products of degradation are then rebuilt into the recipient genome. Extra degradation could be initiated by the strand breaks. In our experiments reported recently [9] it has appeared to us that single­ strand breaks may also be important for the survival of biological activity of transforming DNA. 48 NOTANI et al.

2 a о X со

FRACTION NO.

FIG.4. Wild type(Rd) and N17 cells with 3H-labelled DNA, irradiated with a dose of 100 ergs/mm2 at 254 nm and incubated for 60 and 120 min in nutrient medium. Cells were treated with lysozyme and lysed with 0.5 M sodium hydroxide which had been layered on top of 5-20% sucrose gradient (pH 12). Tubes were spun for 90 min at 30 000 rev/min in SW 65 rotor at 10°C.

TABLE II. UPTAKE OF 32P-L A B E L L E D S rC r DNA AND ITS TRANSFORMING ACTIVITY ON WILD TYPE AND MUTANT N19

Integration Percent transformation Irreversible uptake Strain (°Jo o f to ta l 32 P counts Viable centres ( 32P c p m /0 .1 m l of lysate) in p ellet) Sr C r

W ild type 8 747 72.6 178 X 10 s 0 .2 5 0 .2 6

N19 10 984 72.2 160 X 1 0 6 0 .1 3 0.0022 IAEA-SM-134/3 49

TABLE III. IRREVERSIBLE UPTAKE AND ABILITY TO TRANSFORM BY UNIRRADIATED AND GAMMA-IRRADIATED 32P-LABELLED Sr TRANSFORMING DNA

Inactivation of biological Gamma-ray exposure Irreversible uptake Transformation activity of DNA (kR) ( 32P cpm/0.1 ml) С lo) 6 ¡0

0 2403 0 .5 0 -

2 .5 2136 0 .1 3 76.3 1

5 .0 2101 0 .0 6 87.0 5

TABLE IV. EFFECT OF GAMMA-RAY EXPOSURE OF 32P-LABELLED TRANSFORMING DNA ON ITS 'INTEGRATION' INTO THE RECIPIENT GENOME AND SPECIFIC BIOLOGICAL ACTIVITY OF INTEGRATED AND UNINTEGRATED TRANSFORMING DNA

Gamma-ray exposure Incubation time 'Integration’ Specific biological activity (kR) (m in) (°¡o) (Sr colonies/ 32P cpm)

Unintegrated In teg rated

0 o 2 8 .4 0 1 .7 3 0 .1 2 4

30 7 1 .1 0 0 .6 8 0 .0 8 8

2 7 .0 9 0 .1 4 0.0 2 2 2 .5 ° 30 6 6 .6 1 0 .1 3 0 .0 2 9

о 2 4 .1 2 0.0 2 9 0 .0 1 5 о СЛ 30 6 4 .4 4 0.0 3 0 0.011

DISCUSSION

Steps in the (excision) repair of u.v.-damaged DNA appear to be the following: Incision next to thymine dimers, excision of thymine dimers, enlargement of the gap produced by excision by controlled de­ gradation, repair synthesis and sealing of the remaining backbone break by a DNA ligase (reviewed in Ref. [10]). Mutants N12 and N21 cannot excise thymine dimers and therefore should be either incisionless or excisionless. Setlow et al. [11] have reported an endonuclease isolated from Micrococcus luteus which reactivates u.v.-inactivated DNA especially when assayed in the u.v.-sensitive mutant DB112. They have postulated that the endonuclease function is involved in the repair of biological damage resulting from u.v. irradiation and that the u.v.-sensitive mutant is deficient in this step. Although we have not used a similar test to define whether mutants N12 and N21 are incisionless, we have tried to determine the sequence of steps by a genetic analysis (unpublished). 50 NOTANI et al.

The genetic analysis was done with mutants N12 and N21. Genetically- marked N12 and N21 mutant cells were u.v. irradiated, incubated after irradiation and lysed. The lysate of N12 Cr was assayed on N12 Cs . (cathomycin sensitivity) and N21 Cs^ Reciprocally, N21 Cr was assayed on N21 Cs and N12 Cs_. If N12 and N21 are mutations of different loci, repair complementation will result only when DNA from the mutant blocked in a later step is assayed on the mutant blocked in the earlier step. Repair complementation should be observed in only one direction of a cross but not in the reciprocal direction. In our preliminary ex­ periments we observed that irradiated N21 DNA is repaired when assayed on N12 but that irradiated N12 DNA assayed on N21 is not repaired. While this suggests that mutation in N21 controls a later step of repair than the one in N12, other interpretations of the data also may be possible. It should, however, be possible to confirm the classification of these mutants independently by micrococcal endonuclease. Following irradiation, mutant N17 has been found to possess slower sedimenting DNA at later times. Presumably there is some time re­ quired to do the excision which wiil generate gaps. Normally the gaps may be filled without much delay but mutant N17 appears to be slow in the rate of joining strand breaks. Coincidentally, mutant N17 is also somewhat sensitive to gamma radiation. Mutant N19, the least sensitive of the mutants studied here shows a peculiar property of yielding much lower transformation with Cr un­ irradiated transforming DNA as compared with Sr DNA. Our earlier observations have indicated that the base composition of Sr and Cr markers is different. This is yet another peculiar property of the Cr marker and suggests that N19 may have an enzyme which preferentially attacks the base sequences of Cr and prevents it at some step from effecting transformation. Studies on the fate of labelled gamma-irradiated DNA in transforma­ tion has disclosed that the loss of biological activity may be due to de­ gradation of irradiated DNA, presumably initiated by strand breaks. Degradation would lower the amount of macromolecular integration of input DNA into the recipient.

CONCLUSIONS

Ultra-violet-sensitive mutants N12 and N21, which are about 20 times more sensitive than the wild type, were found unable to carry out the excision of thymine dimers induced in their DNA by u.v. irradiation. The equally sensitive mutant N17, although normal in regard to its ex­ cision ability, was found to be slow in its rate of joining of excision breaks. Mutant N19 has the unusual property of yielding a much lower number of transformants from unirradiated Cr DNA than from Sr DNA. This may or may not have relevance for the u.v.-sensitivity of the m utant. Inactivation of gamma-irradiated DNA on transformation was found to be due to its degradation. Extra degradation is presumably initiated by DNA strand breaks. IAEA-SM-134/3 51

REFERENCES

[1] SETLOW, J.K ., BROWN, D .C., BOLING, M .E., MATTINGLY, A., GORDON, M .P., J. Bacteriol. 95 (1968) 546. [2] RUPERT, C.S., GOODGAL, S.H., Nature 185 (1960) 556. [3] SETLOW, R.B., SWENSON, P.A., CARRIER, W.L., Science 142 (1963) 1464. [4] SETLOW, R.B., CARRIER, W .L., Proc. natn. Acad. Sei. USA 51 (1964) 226. [5] BOYCE, R.P., HOWARD-FLANDERS, P., Proc. natn. Acad. Sei. USA 51 (1964) 293. [6] CARRIER, W .L., SETLOW, R.B., Methods in Enzymology (in press). [7] SETLOW, J.K ., RANDOLPH, M .L., BOLING, M .E., MATTINGLY, A., PRICE, G.. GORDON, M .P., Cold Spring Harbor Symp. Quant. Biol. 33 (1968) 209. [8] RANDOLPH, M .L., SETLOW. J.K ., Proc. natn. Acad. Sei. USA (in press). [9] JOSHI, V.R., NOTANI, N.K., D.A.E. (India) Symp. on "Basic Mechanisms in Radiation Biology and Medicine" (1971) Abst. 11. [10] WITKIN, E.M ., Annual Review of Microbiology 23 (1969) 487. [11] SETLOW, R.B., SETLOW, I.K., CARRIER, W .L., J. Bacteriol. 102 (1970) 187.

DISCUSSION

H. ALTMANN: Is there a close correlation between the shift in sedimentation profiles after ultracentrifugation in alkaline sucrose and your transformation experiments? N. K. NQTANI: Yes, there is, and this shift can satisfactorily explain the sensitivity of the mutant on a qualitative basis. H. ALTMANN: Have you investigated repair steps shorter than 60 min, since repair processes occur very rapidly? N. K. NOTANI: Although repair steps shorter than 60 min were not considered in our sedimentation studies, it is nevertheless remarkable that even at 120 min a considerable amount of more slowly sedimenting material is present in the mutant (N17) but not in the wild type. H.I. ADLER: Does Haemophilus have a good mechanism for recognizing the newly introduced transforming DNA? N.K. NOTANI: Yes. In fact, extracts of Haemophilus influenzae contain an endonuclease which is capable of recognizing and degrading foreign DNA but not the DNA of IL influenzae.

IAEA-SM-134/8

RADIORESISTANCE OF SOME MICROORGANISMS AND THEIR PURINE-5-PHOSPHORIBOSE- 1-PYROPHOSPHATE TRANSFERASE ACTIVITY*

G. PARTSCH, H. ALTMANN Reactor Centre Seibersdorf, Seibersdorf, Austria

Abstract

RADIORESISTANCE OF SOME MICROORGANISMS AND THEIR PURINE-5-PHOSPHORIBOSE-l-PYROPHOSPHATE TRANSFERASE ACTIVITY. The purine-5-phosphoribose-l-pyrophosphate-transferase activities and the radioresistance of these enzymes were determined in some fruit-juice spoiling microorganisms. The radioresistance of the organisms and their transferase activities were compared to find a possible relation between these factors.

INTRODUCTION

One of the primary affects of ionizing radiation in célls is the inhibition of the DNA biosynthesis and especially the de-novo formation of pyrimidine and purine precursors for the nucleotides. In microorganisms the nucleo­ tides are anabolized in two separate ways, either de novo by low molecular precursors as monocarbon units or by the re-utilization of preformed purine bases or nucleosides. The bases, which originate mainly from damage of the cells'own nucleic acids, combine with 5-phosphoribose-1-pyrophosphate and the nucleotides formed in this way are transferred into the common nucleotide pool where they may be used for the synthesis of new nucleic acids or coenzymes. Both the de-novo pathway and the utilization of preformed bases are connected by feedback reactions so that the de-novo synthesis is decreased by greater amounts of nucleotides produced by the salvage pathway. Two independent enzymes are responsible for the formation of purine nucleotides using preformed bases, the adenine-5-phosphoribose-1 -pyrophosphate transferase (adenine-PRPP-transferase) and the hypoxanthine-guanine- 5-phosphoribose-1 -pyrophosphate transferase (hypoxanthine-guanine-PRPP- transferase). The reaction between the purine bases and PRPP can be summarized according to the following equation:

purine base + PR PP—gnzyme ^ purine-nucleoside-monophosphate+PPj M g++

In earlier investigations it was pointed out that in mammalian cells the nucleotides for the repair processes, especially at high radiation doses, are formed along the preformed pathway. Therefore, we studied the purine- PRPP-transferases in the food-spoiling microorganisms: Pénicillium terrestre, Rhinocladiella spp. , Aspergillus versicolor, Byssochlamys fulva

* This investigation was supported by grant FG-Austria-102 from the United States Department of Agriculture.

53 54 PARTSCH and ALTMANN and Pullularia pullulans and compared their different radioresistance with these enzyme activities. Byssochlamys fulva became a dangerous factor in the food industry because its ascospores are very heat resistant and withstand the pasteurization temperature. Pullularia pullulans, also called black yeast, is one of the most radioresistant microorganisms we know [1 ,2 ].

MATERIAL AND METHODS

1. Cultivation and irradiation experiments

Pénicillium terrestre (Jensen), Rhinocladiella sp. and Aspergillus versicolor (Tirabashi) were isolated from the juice of fresh squeezed apples and were identified by the Centralbureau of Schimmel-cultures in Baarn (Netherlands). The moulds were cultivated on potato-dextrose agar for 4 days and the conidiospores harvested by washing with phosphate buffer (0. 14 JM, pH 7. 0) supplemented with 2 ppm Tween 80. Pullularia pullulans (Danish strain) was cultivated in submersed form in potato-dextrose broth at pH 4.6 for 12 hours in a Brunswick micro-fermentor. For anaerobic growth nitrogen was passed through the nutrition medium. Byssochlam.ys fulva NRRL 2614 was grown on Sabouraud agar at pH 5. 6 for 4 days and the conidiospores were harvested according to the above method. The cells were suspended in the above buffer and irradiated with a 60Co source at a dose rate of 450 krad/h. After irradiation of the cells, the cell suspensions were diluted and plated out on potato-dextrose agar or, in the case of Byssochlamys fulva, on Sabouraud agar. After incubation for 3-4 days at 30°C thè number of calories that survived were counted.

2. Determination of the purine base-PRPP-transferase-activities

The enzyme activities were determined in the cell-free extracts of the submersed cultivated cells. For this purpose one gram of each organism (wet weight) was homogenized in 2 ml of phosphate buffer with 4 g of glass beads (0. 45 - 0. 50 mm in diameter) for two minutes in a Braun homogenizer and the disrupted cells centrifuged in a preparative MSE centrifuge for one hour at 100 000 X g. The transferase was determined in 2 ß l according to the method of Cábela [3]. The labelled adenine-8 - 14C (specific activity 6.55 mCi/mM). guanine-8 - 14C (5.4 mCi/mM) and hypoxanthine-8 - 14C (4. 18 mCi/mM) were NEN chemicals. The protein concentration in the cell-free extracts was measured in each sample with the Biuret-reaction [4] for the calculation of the m /nmol transposed base per mg protein.

3. Autoradiographic study of hypoxanthine- and guanine-PRPP-transferase

Additional autoradiographic studies were done to determine whether there are two different hypoxanthine- and guanine-PRPP-transferases in Pullularia. Therefore the proteins of homogenized Pullularia (aerobically grown cells) were separated by disc electrophoresis (7. 5% acrylamid gel) according to Davis [5]. After the electrophoresis, the gels were washed for 15 min with Tris buffer (0.1 M, pH 7. 4)+0.05 M MgCl2 and then a further IAEA-SM-134/8 55

FIG.l. The survival curves of: 1. Byssochlamys fulva; 2. Aspergillus versicolor; 3. Rhinocladiella spp. ; 4. Pénicillium terrestre.

20 min with the same buffer system supplemented with 5 mg PRPP. Anion- exchange chromatography paper DEAE (Serva) was soaked with 100 iuCi 14C-guanine and another paper with 100 juCi 14C-hypoxanthine (in 2 ml of Tris buffer). The gels and the prepared papers were put together in a sandwich­ like manner and incubated for 20 min at 37°C under slight pressure. Then the papers were washed with 1 mM ammonium form ate+ 0. 1 mM guanine (unlabelled) for 10 min, to remove the remaining 14C-base, dried and in­ cubated together with an X-ray film (Kodak) for a week at -30°C in a freeze box. The radioactive zones could be detected after the development of the film [6 ].

RESULTS AND DISCUSSION

The survival curves of the investigated microorganisms are plotted on Fig. 1. The conidiospores of the mould Byssochlamys fulva were the most radiosensitive cells among the tested fruit-juice spoilers. The LD 99 was detected with a 50 krad dose when the cells were plated on Sabouraud medium at pH 5.6. The purine-nucleoside-monophosphate-formation determined by the investigated enzyme systems showed a decreasing affinity for the three purine bases in the order adenine, guanine and hypoxanthine. In contrast to Cabela [3], we obtained a much greater enzyme activity. With Byssochlamys we also carried out an irradiation series to follow the radi­ ation effects On the enzyme activities. For this purpose the submersed material was irradiated with 0 .5 , 1.0 and 2. 0 M rad in phosphate buffer (0.14 M, pH 7.0). The adenine-PRPP-transferase activity decreased continuously with increasing irradiation dose and reached approximately 50% of the control at 2. 0 Mrad (Fig. 2). The same tendency was obtained with the guanine- and hypoxanthine-PRPP-transferases (Figs 3 and 4). 56 PARTSCH and ALTMANN

FIG.2. The adenine-PRPP-transferase activities of Byssochlamys fulva. 1 = control; 2 = 0.5 Mrad; 3 = 1.0 Mrad and 4 = 2.0 Mrad.

FIG.3. The guanine-PRPP-transferase activities of Byssochlamys fulva. 1 = control; 2 = 0.5 Mrad; 3 = 1.0 Mrad and 4 = 2.0 Mrad.

Aspergillus versicolor showed an LD99 of 90 krad. We expected a transferase activity similar to that of Byssochlamys because the LD99 was not very different. But Fig. 5 demonstrates much less activity for all three enzymes. The Rhinocladiella sp. was not so sensitive as Aspergillus versicolor. Its LD99 was about 140 krad. The bases hypoxanthine and adenine were transposed to the same low extent but we obtained the highest guanine-PRPP- transferase activity of all the organisms tested (Fig. 6). The LD99 of Pénicillium terrestre was 160 krad. After Pullularia it was the most resistent mould. Figure 7 shows the measured purine-PRPP-transferase activities. In this mould the hypoxanthine- and guanine-PRPP-transferases IAEA-SM-134/S 57

FIG.4. The hypoxanthine-PRPP-transferase activities of Byssochlamys fulva. 1 = control; 2 -0 .5 Mrad; 3 = 1.0 Mrad and 4 = 2.0 Mrad.

FIG. 5. The purine-PRPP-transferase activities of Aspergillus versicolor. ------guanine-PRPP-transferase ------hypoxanthine-PRPP-transferase ------adenine-PRPP-transferase were equal but higher than the adenine-PRPP-transferase. These enzyme activities were generally lower than in other microorganisms. There is a distinct difference between the radioresistance of the previously described moulds and the survival of Pullularia. The blastospores of the yeast-like fungus Pullularia pullulans showed a different radioresistance due to growth conditions [2, 7]. Anaerobically grown cells showed an LDgg of about 900 krads and cells which grew for some hours under strong aeration, following anaerobic growth, increased in their resistance (LDgg about 1.4 Mrad). The survival 58 PARTSCH and ALTMANN

FIG.6. The purine-PRPP-transferase activities of Rhinocladiella sp. ------guanine-PRPP-transferase ------hypoxanthine-PRPP-transferase adenine-PRPP-transferase

FIG.7. The purine-PRPP-transferase activities of Pénicillium terrestre. ------guanine-PRPP-transferase ------hypoxanthine-PRPP-transferase ------adenine-PRPP-transferase. IAEA-SM-134/3 59

FIG. ¡В. The purine-PRPP-transferase activities of Pullularia pullulans grown anaerobically. 1 = control; 2 = 0 . 5 Mrad ; 3 = 1.0 Mrad and 4 = 1.5 Mrad. ------guanine-PRPP-transferase ------hypoxanthine-PRPP-transferase ------adenine-PRPP-transferase

FIG. 9. The purine-PRPP-transferase activities of Pullularia pullulans grown aerobically. 1 = control; 2 = 0 . 5 Mrad, 3 = 1.0 Mrad and 4 = 1.5 Mrad. ------guanine-PRPP-transferase ------hypoxanthine-PRPP-transferase adenine-PRPP-transferase 60 PARTSCH and ALTMANN curves are of sigmoidal character indicating strong mechanisms to repair DNA damage. In aerobically grown blastospores we detected a great activity of adenine-PRPP-transferase similar to the results in Byssochlamys. The values for the hypoxanthine-transferring enzyme reached 60% and for the guanine-transferring enzyme only 30% of the adenine-PRPP-transferase (Fig. 8 ). The enzyme activity of the adenine- and hypoxanthine-PRPP-transferases was lower in the aerobically grown blastospores than in the anaerobically grown ones. Only the guanine-PRPP-transferase activity was enhanced (Fig. 9). Enzyme activity after irradiation was determined with anaerobically and aerobically grown cells. Samples were irradiated with 0. 5, 1.0 and 1. 5 Mrad. Irradiation of the Pullularia blastospores up to 1.5 Mrad did not influence the activity of the investigated enzymes. The noted oscillations within an irradiation series are not significant deviation or inhibition effects caused by irrad ia tio n (Figs> 8 and 9). If we compare the survival curves of Byssochlamys and Pullularia with the radioresistance of their investigated enzymes, we may conclude that microorganisms which differ in their radioresistance show the same tendency in the radioresistance of their purine-PRPP-transferases. The conidiospores of Byssochlamys were growth inhibited by low gamma doses (LD9g 50 krad). The activities of the adenine-PRPP-transferase and hypo- xanthine-PRPP-transferase attained 40% of the control at 2 . 0 Mrad. The guanine-PRPP-transferase was inhibited only to 50%. In Pullularia blastos­ pores grown under aeration we determined the LD 99 with 1. 4 Mrad but their enzymes were not affected up to 1.5 Mrad. The important role of the hypoxanthine (guanine)-PRPP-transferase for the control of the purine biosynthesis in different diseases could be shown in the human [8 , 9]. In humans, animals and microorganisms most nucleotides are synthesized de novo' whereby the purine-PRPP-transferase regulates these mechanisms by feedback reactions. But it also seems that the de-novo synthesis is markedly suppressed after irradiation and es­ pecially the monocarbon transport may be more affected than the preformed pathway [10] . From that we may therefore assume that nucleotides neces­ sary for the repair of DNA damage after irradiation may be synthesized even when one or more steps in the pathway from monocarbon units are interrupted by radiation effects. There is also evidence [2] that such a difference in the radiosensitivity of these two purine nucleotide pathways also occurs in Pullularia and B.yssochlam.ys. The purine-PRPP-transferases were partly inactivated in Byssochlamys at gamma doses between 10 and 40 times the LD9g of the cells obtained in survival experiments. The SH-group content of the cells may also play an effective preventive role against radical damage. We found a high SH-group content in Pullularia pullulans by measuring the degree of p-hydroxymercuribenzoate accumu­ lation [1 1 ] which is additional evidence for these suppositions. An additional electrophoretic study was performed using also auto­ radiography to clarify whether there are two different hypoxanthine-guanine - PRPP-transferases in Pullularia. Brockman [12], Kalle [13] and Adey [14] detected separate transferases for adenine, guanine and hypoxanthine in microorganisms, whereas in mammals only one enzyme for hypoxanthine and guanine may exist [15]. It could be shown without any doubt that in Pullularia both bases were transposed by the same protein fraction. These findings do not exclude that in other microorganisms (for example in IAEA-SM-134/S 61

Rhinocladiella, see Fig. 6 , where anextremely high guanine-PRPP-transferase was detected in contrast to the hypoxanthine-PRPP-transferase) two separate enzymes exist. In summary we are able to say that the purine-PRPP-transferase activities for the bases adenine, hypoxanthine and guanine showed an apparent difference in the investigated microorganisms. A direct correlation between their radiosensitivity or radioresistance and their enzyme activities could not be concluded from these results. In the radioresistant Pullularia these enzymes were not affected by gamma doses up to 1.5 Mrad as were the enzymes in the very sensitive Byssochlam.ys fulva. This may be due to a different SH-group content. The investigation also showed that the hypo- xanthine-guanine-PRPP-transferase is a single protein complex in Pullularia.

REFERENCES

[1] SKOU, S.P., Radiation Preservation of Foodstuffs. 2nd Scand.Meeting of Food Preservation by Ionizing Radiation (1964). [2] PARTSCH, G., ALTMANN, H., in preparation. [3] CABELA, E., ALTMANN, H., EBERL, R., VI th Int.Symp.on Microtechniques 21 (1970). [4] WEICHSELBAUM, T.E., J.A m .clin.Pathol. 7 (1946) 40. [5] DAVIS, B.J., Ann.N.Y. Acad.Sei. 121 (1964) 404. [6] DER KALOUSTLAN, v . M ., BYRNE, R ., YOUNG, W .J., CHILDS. B ., B iochem . G enetics 3 (1969) 2 9 9 ; [7] ALTMANN, .H ., PARTSCH, G .. DRAXLER, H ., MERDINGER, E ., 63 rd A nn. M eeting of the Illinois State Acad.Science (1970). [8] SEEGMILLER, J .E ., ROSENBLOOM. F .M ., KELLEY, W .N ., S cien ce 155 (1967) 1682. [9] FELIX, J.S., De MARS, R., Proc.natn. Acad. Science USA 62 (1969) 536. [10] WINCHELL, H.S., VIMOKESANT, S., RALEY, R., J.nucl.M ed. 10 (1969) 624. [11] FIEDLER, W ., PARTSCH, G ., ALTMANN, H ., to be p ublished. [12] BROCKMAN, R.W., SPARKS, C., SIMPSON, M .S., SKIPPER, H.E., Biochem.Pharmac. 2 (1959) 77. [13] KALLE, G.P., GOTS, J.S., Biochim.biophys.Acta 53 (1961) 166. [14] ADEY, J.C ., GOTS, J.S., Biochim.biophys. Acta 118 (1966) 344. [15] HENDERSON, J.F., Can.J.Biochem. 47 (1969) 69.

DISC USSION

S. G. GEORGOPOULOS: In Pullularia pullulans we have more than one type of spore, and these spores, if I am not mistaken, have been reported to differ in their radioresistance. Were the enzyme activities in your work correlated with the types of spores present? G. PARTSCH: The enzyme activities in our experiments were correlated only with blastospores obtained after anaerobic growth. The aerobically grown cells, which showed enhanced radioresistance, were also blastospores but with initiated pigmentation. V. V. SUKHODOLETS: Do you think that nucleotides synthesized de novo and nucleotides synthesized from exogenous bases differ in their function as precurors of nucleic acids in the cell? G. PARTSCH: The function of nucleotides as precursors of nucleic acids or coenzymes may not depend on their origin or on the method of their formation. There are differences in the de-novo synthesis activity owing to feedback reactions of the nucleotides formed by the preformed pathway. The de-novo pathway may therefore be repressed when exogenous bases are available. 62 PARTSCH and ALTMANN

H. HESLOT: Are some of the species utilized in your work able to grow in a strictly minimal medium? If so, have you tried to obtain mutants inactivating at least some of the pyrophosphorylase? G. PARTSCH: It has been reported that Pullularia pullulans and Byssochlamys fulva are able to grow in minimal media. Byssochlamys produces its ascospores in such media at low pH, but we have never tried to obtain pyrophosphorylase mutants. IA EA -SM - 1 3 4 /13

U. V. MUTABILITY IN GAMMA-RAY-SENSITIVE MUTANTS OF Neurospora crassa

R.D. MEHTA, J.WEIJER University of Alberta, Edmonton, Alta., Canada

Abstract

U .V . MUTABILITY IN GAMMA-RAY-SENSITIVE MUTANTS OF Neurospoia ciassa. Three gamma-ray-sensitive mutants were isolated in Neurospora crassa by u.v. induction. Genetic analysis of these mutants indicated that gamma-ray sensitivity is controlled by a single nuclear gene. Tests for u.v. sensitivity revealed that the gamma-ray-sensitive mutant gs-6 a, pe is, in addition, sensitive to u.v. radiation, whereas two other mutants and their progenitor wild type are u.v. resistant. U.V. mutability tests showed that fewer caffeine-resistant mutants were induced in all the three gamma-ray-sensitive mutants as compared to the wild type when irradiated at the same u.v. dose levels. The results are discussed in the light of similar data reported in other fungi and bacteria.

INTRODUCTION

Different investigations of radiation sensitive mutants in bacteria have revealed that a relationship exists between the processes of DNA repair after irradiation, recombination and mutation induction by UV light [1, 2, 3]. For instance, bacteria mutants deficient in recombina­ tion are sen sitiv e to UV lig h t and X-rays and exhibit in addition a suppression of mutation induction. By way of extending similar studies to Eukaryotes, attempts have been made to isolate radiation (ionizing as well as non-ionizing) sensitive mutants in algae and fungi [4, 5, 6 , 7, 8 , 9, 10]. Effects of genes conferring UV sensitivity on recombination and on induced mutation have been reported in fungi [7, 10, 11, 12, 13, 14]. Chang and Tuveson [1 2 ] have shown that the induction of mutations to acriflavine or caffeine resistance was sharply reduced in a UV-sensitive mutant of Neurospora crassa. In order to determine the effects of genes which confer у ray sensitivity on UV mutation induction, у ray-sensitive mutants have been isolated in N. crassa and are reported on in this communication.

MATERIALS AND METHODS

The following strains of Neurospora crassa were used throughout the investigation: Y 8743 a_, £e^ obtained from the Fungal Genetics Stock Center (Hanover, New Hampshire) and 46-S A, a l-2 , pan-2, try p -2, obtained from Dr. S.F.H. Threlkeld, McMaster U niversity, Hamilton (Ontario). The above stra in s were maintained on so lid ifie d Vogel [15] medium supple­ mented with the appropriate amino acids (50 mg/1000 ml medium) and incubated at 25°C, For iso latio n of у ray -sen sitiv e mutants, the procedure by Lung-Ting Chang and Tuveson [12] for UV-sensitive mutants was adopted using 0.066M phosphate buffer (Na-Na) at pH 7.0 for the preparation of cOnidial suspensions (1 X 106 conidia/m l). The conidia were harvested from cultures 7 to 10 days old.

63 64 MEHTA and WEIJER

Y ray-sensitive mutants were induced by UV irradiation produced by a GE G 15T8 germicidal_larap at 25 cm from source (dose: 20 erg mm"2; dose ra te 10 erg nun 2 sec x) of a conidial suspension of strain Y 8743 a_, After irradiation the suspension was plated on minimal solidified medium (density: approximately 70 colonies per plate) supplemented with 1.5% sorbose, 0.05% fructose and 0.05% glucose. Replica platin g methods were employed using f i l t e r paper saturated with minimal medium supplemented with sorbose, fructose and glucose. The filter papers were placed on a solidified minimal medium (supplemented with sorbose, fructose and glucose) in a P etri dish and were exposed to y rays (dose: 67 Krad; dose ra te : 6.7 Krad/min *) from a 60-Cobalt source. After irra d ia tio n they were incubated at 25°C. After 3 days of incubation, absence of colonies or very limited growth of colonies on the re p lic a plates (when compared with the "master" plate) was scored and subsequently these colonies were isolated from the "master" plate as suspected y ray-sensitive mutants. Twenty suspected у ray -sen sitiv e mutants were isolated in this manner (to ta l number of surviving colonies scored: 7250). All 20 isolates were re-tested for y ray s e n s itiv ity and fin a lly 8 mutants were selected for further studies. Of these, three are described in this communication, viz■, gs -3

gs-3 a_, £i3 X 46-5 A, al-2 , pan-2, try p -2 gs-6 a, ££ X 46-5 A, al-2, pan-2, tryp-2 gs-20 a^, ge_ X 46-5 A, al-2, pan-2, tryp-2 46-5 A, al_-2, pan-2, tryp-2 x gs_-6 a_, £e 46-5 A, al-2, pan-2, tryp-2 x gs_-20 a, ge_ IAEA-SM - 1 3 4 /13 65

The reciprocal cross 46-5 A, al-2, pan-2, tryp-2 x gs-3 a_, could not be established due to male s t e r il it y of the s tra in gs-3 a_, j>e_. Approximately 7 complete tetrads resulting from a particular mating (including the reciprocal cross where possible) were analyzed for the segregation of the genetic markers al- 2 , pan-2 , try p -2 and gs_. The test for y ray sensitivity consisted of irradiating a 4 ml conidial suspension (density: 1 x 105 conidia/ml) with a dose of 134 Krad (dose rate: 6.7 Krad min”1) of у rays deliverèd by a 60-Cobalt source, y ray sensitivity was indicated by an absence of growth or extremely weak growth of the irradiated isolate after 3 days of incubation at 30°C in minimal medium supplemented with the appropriate growth requirements. To avoid discrepancies in dose received the 8 ascospore cultures (resulting from the same ascus) were irradiated simultaneously at isodose positions in the 60-Cobalt y cell. All experiments wore replicated (2) and repeated once except those involving a few asci from the cross 46-5 A, al-2, pan-2, try p-2 x gs_-3 a_, pe^. Due to the very fast growth rate of gs-3 a, pe, when compared with other g_s_ stra in s and its progenitor Y 8743 £, £c_, the recognition of y ray sensitivity was found to be less conclusive in these segregants. However, on re -te stin g these segregants, a normal segregation pattern for the marker gs-3 was obtained.

RESULTS

(i) The genetic inheritance of y ray sensitivity

To te s t for chromosomal inheritance of у ray se n s itiv ity , mutants gs-3 a, pe; gs-6 a, ge^ and gs-20 a, ge_ (all induced by UV radiation) were crossed with strain 46-5 A, al-2, pan-2, tryp-2■ Unordered tetrads were analyzed from each cross (see Methods) and the segregants were tested for mating type (A vs_. a), auxotrophic requirement (pan-2, tryp-2) and у ray sensitivity (sg). In addition, the segregation for the marker albino (al-2) was noted. For all markers a 4:4 segregation pattern was observed. The tetrad analysis of the cross gs -6 a_, £e x 46-5 k_, a l -2, pan-2, tryp-2 showed th a t out of 8 asci tested , 5 were PD's (Parental d ity p es), 3 TT's (Tetratypes) with respect to the pan-2 and tryp-2 markers located in linkage group VI. No NPD's (Non-parental ditypes) were encountered. These results are taken to indicate that the gene controlling у ray sensitivity in mutant gs -6 is probably located in linkage group VI close to the markers pan-2 and try p -2 . Due to the lim ited number of asci analyzed and the locatiön of markers (linkage group I, II, VI) no infor­ mation with regard to the linkage relationship of the mutants gs-3 and gs-20 could be established.

(ii) у ray survival characteristics of the mutants gs-3 a, pe; gs-6 a, pe and gs-20 a, pe

The survival curves of the mutants after treatment with different doses of у rays showed that gs-3 a_, p£; gs-6 a_, ge^ and gs-20 a_, ££ are more у ray sensitive than the wild type strain and progenitor: Y 8743 a_, pe (Fig. 1). When the respective survivals at the dose level of 67 Krad were compared, i t was found that gs_-20 a, ge_; gs-3 a_, and gs_-6 a, were 7, 10 and 20 times more sensitive than the wild type strain. The difference in у ray sensitivity between the wild type strain and the three mutants became greater with increasing dose. Comparison of the survival curves of the mutants with the wild type strain curves was based exclusively on the qualitative appearance of the curves. The difference between у sensitivity of the mutant gs_-20 £, ££ appeared less conspicuous 66 MEHTA and WEIJER

G A M M A RAY DOSE (in minutes)

FIG. 1. Gamma-ray-survival curves for a wild-type strain (gs+) and three gamma-ray-sensitive mutants (gs) of Neurospora crassa.

at lower doses (26 Krad)' however, the difference became more significant with increase in dose (Fig. 1). The slope of the survival curve for gs-20 a_, pe_ increased in steepness a fte r a dose of 56.6 Krad which yielded in the wild type strain a survival of 18%. Although mutant gs-3 a, was more у ray sensitive than gs-20 a_, £e over the entire dose range tested, both curves parallelled one another indicating an identical slope. A rather conspicuous response was displayed by mutant gs-6 a_, pe, the most у ray-sensitive mutant. The dose response curve yielded a slope approximately comparable with those obtained at low dose levels for gs-3 a_, ££ and gs-20 a_, pe. At higher dose levels, however, the slope of the survival curve of gs-6 a_, j>e_ decreased and hence intercepted the curve for gs-3 a_, ££ at 124 Krad thereby attaining the slope of the survival curve for wild type.

(iii) UV sensitivity

The dose response curves of strains Y 8743 а, pe_ (wild type), gs-3 a_, pe and gs-6 a_, ge_ showed characteristic shoulders at low dose levels (extending to approximately 80% survival of Y 8743 a_, £e) and a subsequent decline (Fig. 2). Mutants gs-3 a_, pe and gs-20 a_, p£ were indistinguishable IAEA-SM-134/13 67

U V - DOSE (in minutes)

FIG.2. Forward mutation rate to caffeine resistance (per 107 survivors) and survival №) of a wild-type strain (gs+) and three gamma-ray-sensitive mutants (gs) of Neurospora crassa as a function of u.v. dose.

in th e ir response to UV radiation (as measured by survival) when compared with the response of the wild type strain Y 8743 a_, £e_. Mutant gs-6 a_, ££, on the other hand, exhibited a larger degree of UV sensitivity than the wild type strain especially at higher dose levels: at a dose yielding 53% survival, the slope of the survival curve for gs-6 a_, £e increased when compared with the slope of the wild type strain for these dose levels. The survival curve for gs-20 a, ge_ lacked the shoulder characteristic for the wild type (and also for mutants gs_-3 a_, p£ and gs -6 a_, pe) survival curve at low levels of radiation, however, a decrease of the slope followed by an increase was noticed at increasingly higher doses of UV.

(iv) UV mutability

The UV-induced caffeine re s ista n t mutants of the wild type Y 8743 a_, pe accumulated exponentially and reached a plateau at a dose yielding approximately 13% survival (Figs. 2 and 3; Table 1). When conidia from mutant gs-3 £

FIG. 3. U .V . -induced forward mutation rate to caffeine resistance (per 106 survivors) of a wild-type strain (gs'1') and three gamma-ray-sensitive mutants (gs) of Neurospora crassa as a function of survival Wo).

resistance appeared to be different. In gs-6 a, pe_ the incidence of induced caffeine resistant mutations increased steadily until the point at which the wild type strain reached a plateau, whereas at higher dose levels the frequency of induced caffeine resistant mutants declined sharply. No forward mutation was observed in this mutant at a UV dose of 4068 erg mm 2. In mutant gs_-20 a_, pe UV-induced caffeine re s ista n t mutations accumulated exponentially at approximately the same rate as in the irradiated wild type strain up to a dose of 2034 erg mm 2 at which a plateau was established well before the wild type strain reached its plateau. The plateau for mutant gs-20 a, was immediately followed by an exponential increase in mutation yield at a dose of 2712 erg mm 2 which permitted a survival ra te of 2 1% in the mutant strain. Although the UV-induced mutation curve for gs-3 a, pe closely followed the curve for the wild type strain, a reduction in the overall frequency of mutations was noted. The UV induction of caffeine resistant mutants was sharply reduced in both gs-6 a, ££ and gs-20 a_, pe strains, when compared with the wild type stra in Y 8743 a_, p£ (Fig. 2). In Fig. 3 the yields of caffeine resistant mutants is plotted as a function of survival. It is evident that the wild type strain yielded more caffeine re s is ta n t mutants at a ll levels of survival as compared with the mutant y ray-sensitive strains. At survival levels greater than 13% the UV induced mutations appeared to accumulate at a faster rate in gs-3 a_, ££ as compared to wild type. Over the 50 - 5% survival rate fewer mutations were produced in gs-20 a, pe as compared to gs-6 a_, p e, however, the difference might turn out not to be significant. At a survival rate of 2% strains gs-20 a_, continued to accumulate caffeine resistant mutations, whereas at the same rate of survival gs-6 a_, ££ produced no mutations. Hence, whether the comparison between the UV mutability of the wild type strain and the gs_ mutants is established on IA EA -SM - 1 3 4 /13 69

TABLE I. INDUCTION OF MUTANTS RESISTANT TO ACRIFLAVINE AND CAFFEINE IN THREE y RAY-SENSITIVE MUTANTS (gs) AND A WILD TYPE STRAIN (gs^) OF N. crassa

Strain UV dose in minutes Survival Caffeine* Acriflavine* dose ra te : (%) re s is ta n t re s is ta n t mutations/ mutations/ 1 1 .3 erg mm 2 sec 1 107 survivors 10 7 survivors

+ 0 100.0 19 0.42 1 1 1 86.4 39 0.38 1 .5 65.5 60 0 3 21.5 316 0 4 7.9 366 0 6 1 . 1 407 0

gs-3 0 100.0 1 0 1 78.7 10 0.31 1.5 52.6 39 1.93 3 27.2 157 4.70 4 1 1 .6 294 0 6 2 .1 330 0

gs_-6 0 100.0 0 1.1 0 1 80.7 3 4.70 1.5 56.7 6 0.37 3 14.6 15 0 4 1.8 24 0 6 0.0 0 0

gs_-20 0 100.0 1 0 1 65.8 2 0 1.5 62.8 4 0 3 29.1 12 0 4 10.0 11 0 6 0 .6 47 0

* The number of mutants produced at each dose represents the total number of induced and spontaneous mutations for caffeine or for acriflavine resistance

the basis of UV dose (Fig. 2), or on incidence of survival (Fig. 3), it is evident th^t more caffeine resistant mutants were induced by UV in the wild type (gs_ ) strain than in y ray-sensitive mutants such as gs-3 a_, gs-6 a_, pe^ and gs-20 я, pe. Acriflavine mutants wore also UV induced in the wild type strain Y 8743 a, and 'die three y ray-sensitive mutants (Table 1). It is, however, apparent from the result that the forward mutation to acriflavine resistance remained extremely low in all strains tested. No acriflavine resistant mutant was recovered in strain gs_-20 a, p£ whereas the incidence of forward mutation in the wild type, gs-3 a_, ££ and gs-6 a, pe did not allow for a statistical comparison. — 70 MEHTA and WEIJER

DISCUSSION Although many UV sensitive-mutants have been reported in Neurospora crassa [12, 17, 18, 19] there is no d ire c t evidence th at DNA repair systems active after UV irradiation (as present in bacteria), are operative in Neurospora. Isolation of mutants defective for photo­ reactivation and dark repair system [17, 19] together with the overall resemblance of some UV sensitivc-mutants of Neurospora to UV sensitive- mutants of E_. coli [18] , may indicate that the same or very similar DNA repair systems exist in both organisms. Three y ray-sensitive mutants gs_-3 a, pe_; gs-6 a, pe and gs-20 a, £e were isolated in crassa (Fig. 1) and from genetical data i t is evident that each of the mutations is controlled by one but a single nuclear gene. Tests for UV sensitivity showed that mutant gs-6 a, pe is UV sensitive in addition to у ray s e n s itiv ity , whereas gs^3 a,~~pe~and gs-20 a, pe are indistinguishable in UV sensitivity when compared to the wild type strain (Fig. 2) and hence are UV resistant. Mutants sensitive to both у rays and UV, and mutants sensitive to yrays but resistant to UV have also been reported to occur in yeast [5], however, no data is available on the UV mutability of these strains. Mutants gs -3 a_, pej gs -6 a_, ££ and gs -20 a_, p£ exhibited a lower yield of UV induced forward mutation to caffeine resistan ce as compared to the wild type strain (Table I), however, the individual mutational response to the three gs_ mutants is distinctly different (Figs. 2 and 3). Similar results have been reported by Chang e£ a^., [11]. It has to be noted, however, that both: UV induced and spontaneous mutation for у ray-sensitive strains, showed significantly lower frequencies in the present study than those reported by Chang et_ aK , [11]. It is assumed that differences in genetic background of the strains used may account for this difference. In the present investigation only a very few acriflavine resistant mutants were observed after irradiation with low UV doses of the three gs mutants and the wild type control strain. The low incidence of these forward mutations for acriflavine resistance is probably due to an enhancement of the lethal effect by acriflavine over the mutagenic effects of UV radiation. A similar observation was reported for E. coli [20, 21].. On the basis of the results reported on, it is concluded that genes controlling у ray se n s itiv ity in N. crassa cause reduction in UV-induced mutability for caffeine resistance irrespective of their effect on UV s e n s itiv ity . However, the response of the three gs_ mutants for UV-induced mutability is distinctly different for each mutant (Figs. 2 and 3). The differential response is thought to be due to the differential expression of the genes in question in controlling different aspects of the DNA repair system. Averbeck et_ a^. , [14] has reported that even two alleles of the same gene controlling sensitivity to both UV and X rays in Saccharomyces, behave d ifferen tly in th e ir response to UV-induced mutations from isoleucine-valine dependence to independence. One of the alleles (r 3-2) shows the highest frequencies of induced mutations, whereas no revertants, spontaneous or induced, have been found for the other allele (r 3-1). The possibility that radiation effects, other than those of a genetic nature, . may have influenced the rate of mutation induction, cannot be ruled out. Before drawing.any definite conclusion with regard to the phenotypic expression of the three gs^ mutations reported on, a detailed characteriza­ tion of th e ir effect on thc^DNA repair mechanism is needed and presently the effect of these mutations on intra-cistronic recombination is under investigation. Rl-FF.RENCES.

[1] CLARK, A .J., MARGULIES, A.D., Proc. Nat. Acad. Sei. (Wash.) 53 (1965) 451. IAEA -SM-134/13 71

[2] IIOWARD-F LANUEKS, 1’., THERIOT, L., Genetics 53 (1966) 1137. [3] 1Ÿ1ÏKIN, E.M., Mutation Res. 8 (1969) 9. [4] ADLER, H. I ., Advances in Radiobiology, Academic Press 2_ (1966) 167. [5] COX, B.S., PARRY, J.M ., Mutation Res. 6_ (196S) 37. [6 ] DAVIES, D.R., Mutation Res. 4_ (1967) 765. [7] HOLLIDAY, R., Mutation Res. 4 (1967) 275. [S] LENNOX, J.E ., TUVESON, R.K., Radiation Res. 31_ (1967) 382. [9] SNOW, R., J. B acteriol. 94 (1967) 571. [10] SCHROEDER, A.L., Molec. Ge'n. Genetics K)7_ (1970) 291. [11] CHANG, L.T., LENNOX, J .E ., TUVESON, R.W., Mutation Res. 5_ (1968) 217. [12] CHANG, L.T., TUVESON, R.W., Genetics 56_ (1967) SOI. [13] NAKAI, S ., YAMAGUCHI, E., Japan. J. Genetics 4£ (1969) 355. [14] AVERBECK, D., et_ a l. , Molec. Gen. Genetics 107_ (1970) 117. [15] VOGEL, H .J., Microb. Gen. Bull. 13_ (1956) 42. [16] NESTERGAARD, M., MITCHELL, H.K., Ara. J. Botany S4_ (1947) 573. [17] STADLER, D.R., SMITH, D.A., Canad. J. Genet. C ytol. 1£ (1968) 916. [18] SCHROEDER, A.L., Molec. Gen. Genetics 107_ (1970) 305. [19] TUVESON, R.W., MANGAN, J . , Mutation Res. £ (1970) 455. [20] HARM, W., Mutation Res. 4 (1967) 93. [21] KITKIN, E.M., Ann. Rev. Genetics 3^ (1969) 525.

DISCUSSION

C. EMBORG: Do you have any experience in u.v. -mutability with mutants of Neurospora crassa having higher radiation resistance than the wild type? J. WEIJER: No, the replica plating method employed does not allow for the recognition of mutants having greater radiation resistance than the wild type. Although I am convinced that such mutants exist, it is difficult to recognize them by a simple test. J. A. ROPER: Have you measured the mutation rates, in these sensitive strains, towards genes other than those determining resistance to caffeine or acriflavine? I ask you this question because the former and, I believe, the latter, inhibit dark repair, and I wonder whether the sensitive mutants would interact with these particular selective media. J. WEIJER: The reason why we measured mutation towards caffeine and acriflavine resistance is that both caffeine and acriflavine are known to be inhibitors of DNA dark repair. In case mutation induction for caffeine and acriflavine resistance is affected by this inhibition of dark repair, we can expect the decline in their mutation frequency not to differ widely in repair-deficient (u.v. -sensitive) and repair-normal strains. H. HESLOT: You mentioned that some of your gamma-ray-sensitive mutants could not be genetically analysed because they gave abnormal asci. Can this be attributed to the loss of enzymes involved in re­ combination, leading to asynapsis or desynapsis at meiosis? J. WEIJER г The work of Holliday with Ustilago and of Schroeder with Neurospora also indicate meiotic sterility accompanying u.v. - sensitivity alone or u. v. -sensitivity together with X-ray sensitivity. Since chromosomal rearrangements cannot easily be ruled out as a cause, it is difficult to state that the loss of an enzyme involved in recombination is indicated.

IAEA-SM-134A

SOME SITES FOR THE INDIRECT EFFECTS OF RADIATION ON DNA CONSTITUENTS

B.B. SINGH, V .T . SRINIVASAN, K .P . MISHRA, A.R. GOPAL-AYENGAR Bhabha Atomic Research Centre, Trombay, Bombay, India

Presented by N.K. Notani.

Abstract

SOME SITES FOR THE INDIRECT EFFECTS OF RADIATION ON DNA CONSTITUENTS. Amongst the various types of radiation-induced DNA damages, single-strand and double-strand breaks and their genetic consequences have been extensively investigated in recent years. The nature of base damage, however, remains comparatively overlooked although high rates of reaction, of bases with the radiolytic products of water have been demonstrated. Whereas hydroxyl radicals and hydrogen atoms add to purine and pyrimidine rings giving rise to corresponding radicals, reaction of bases with radiation-induced electrons leads to the formation of base anions which under suitable environmental conditions become hydrolysed to yield neutral radicals. The addition of sugar and sugar phosphate residues to the bases was found to affect the yields of anions and radicals of bases and the results could not suggest the breakage of the sugar-phosphate and sugar-base bonds as was shown earlier for the direct effects of radiation. The genetic implications of these types of base damages need careful consideration.

INTRODUCTION

The major chemical changes induced in deoxyribonucleic acid (DNA) by high-energy radiations are: dénaturation of the macromolecule, breakage of internucleotide linkages and destruction of bases [1]. Such changes are mediated mainly through the primary radiolytic products of water, namely: eaq, OH and H, but it has not yet been possible to unequivocally attribute the reaction of a particular species to a specific alteration in the genetic material. Although these changes have been implicated in radiation-induced cell death and genetic abnormalities, only a few attempts have been made to study the biological consequences of each alteration individually. The double-strand breaks caused by the disruption of internucleotide linkages mainly due to OH radicals [2] have been held responsible for the loss of colony-forming ability of cells [3]. Attempts have also been made to assess the relative contributions of the radiolytic transients of water in the in­ activation of transforming principle DNA [4, 5]. In view of the observations of DeFillipes and Guild [5] and those of Collyns et al. [6 ], it has been suggested that the destruction of bases due to the indirect effects of radi­ ation accounts for 50% of the genetic damage. The destruction of bases is mediated through intermediary radical ions and radicals which have been studied by the technique of electron spin resonance (e.s.r. ). These are discussed in detail elsewhere [7]. Whereas the intermediary species formed due to the reaction of purines with OH could not be detected, the radicals formed by addition of OH at the 5-6 double bonds of pyrimidine rings have been reported [8,9]. The reaction of hydrogen

73 74 SINGH et al.

AFTER IRRADIATION AFTER 3 MIN AT- 9S С

A D E N O S I NE

AMP

FIG.l. Electron spin resonance spectra of adenine, adenosine and adenylic acid (AMP) in 2.0M NaOH.

atoms or electrons also leads to the formation of radicals due to the addition of H atoms to the purine and pyrimidine bases [7]. It has also been de­ monstrated that the sites of the OH reaction on some pyrimidine nucleosides are different from those on the corresponding bases [1 0 ] and more investi­ gations are urgently warranted to identify the site of the OH reaction on other DNA constituents. The present investigation deals with the inter­ mediary species formed during the reaction of electrons with DNA, nucleo­ sides and nucleotides. Particular emphasis has been given to the effects of sugar and sugar phosphate moeities on the nature, yield and stability of radicals and radical ions.

EXPERIMENTAL

Calf thymus DNA, bases, nucleosides and nucleotides were obtained commercially and used without purification. These were dissolved in 2. 0M NaOH at different concentrations varying from ImM to 0. 3M. IAEA-SM-134A 75

AFTER IRRADIATION AFTER 3MIN AT-9 5C

FIG.2. Electron spin resonance spectra of guanine, guanosine and guanylic acid (GMP) in 2,OMNaOH.

Polycrystalline cylindrical pellets of samples were prepared and irradiated at -196°C with 340 krads of 60Co gamma rays. After irradiation, the pellets were transferred to quartz e. s. r. tubes without being warmed or exposed to light. The photobleaching procedures and the details of e. s. r. measurements have been discussed elsewhere [1 1 ].

RESULTS

When a 2. 0M NaOH pellet is irradiated at -196°C it develops a blue colour which has been attributed to trapped electrons [12]. Its e. s. r. spectrum at -196°C consists of signals from О giving a broad single line (AH = 48 gauss) at g = 2.056, a doublet due to OH radicals and a sharp singlet of half width 12-14 gauss ascribed to trapped electrons, et [13]. The blue colour of the sample as well as the e. s. r. signal from ej can easily be re­ moved on exposure to visible light or by annealing at higher temperatures. The presence of adenine, its nucleoside and nucleotide, at concentra­ tions higher than 0.15M removes the blue colour at -196°C. The e.s.r. 76 SINGH e t a l.

AFTER IRRADIATION AFTER 3MIN AT _ 9 S С

CMP

I SOGI

FIG.3. Electron spin resonance spectra of cytosine, cytidine and cytidylic acid (CMP) in 2.0M NaOH.

spectra of adenosine and adenylic acid are similar to that of adenine (Fig.l), consisting of signals from О , OH and the adenine anion [11] . The anion signal appears at g = 2.0 0 and is a single line of half width about 20 gauss. On warming the adenosine and adenylic acid samples to -160°C the OH signal disappears leaving behind a triplet e. s. r. signal which has been attributed to the radical formed by H-atom addition to the purine ring [11]. The e .s.r. spectra of guanine, guanosine and guanylic acid at -196°C show great resemblance to each other (Fig. 2). Whereas the guanine sample is shown to exhibit mainly a single line at g = 2.0 0 from its anion [1 1 ], guanosine and guanylic acid show the presence of a triplet signal ascribed to the purinyl radical formed by H addition after hydrolysis of the anion. Cytosine, cytidine and cytidylic acid show single lines at -196°C attri­ buted to the pyrimidine anion [11] (Fig. 3). On warming to higher tempera­ tures no changes have been detected in their e.s.r. spectra except the disappearance of the signal from OH. Thymine anion exhibits a characteristic e. s.r. signal with a high field line as shown in Fig. 4. In thymidine and thymidylic acid an eight-line spectrum, attributed to the thymyl radicals, is also seen along with the signal from the anion. In thymine, however, this radical appears only at higher temperatures [11]. Whereas the thymine anion could be bleached by exposing the thymine sample to visible light [1 1 ], it gives rise to thymyl radicals in thymidine and thymidylic acid. IAEA-SM-134/1 77

THVMIOINE THYMIDYLiC ACIO

AFTER IRRAOIATION

— \ J\T

FIG.4. Electron spin resonance spectra of thymine, thymidine, thymidylic acid and DNA in 2.0M NaOH.

DNA at concentrations up to 10 mg/ml does not scavenge all the trapped electrons. Its spectrum after photobleaching the residual trapped electrons consists of a single broad line (ДН = 20 gauss) at g = 2 .00 in addition to the signals from O" and OH. On warming the DNA sample to higher temperatures thymyl radicals appear (Fig. 4). The yields of anions and radicals increase with the concentration of bases, nucleosides or nucleotides and reach a saturation value at about 0. 15M in 2.0M NaOH matrix.

DISCUSSION

The present system of sodium hydroxide is chosen because it has been shown to trap radiation-induced electrons at -196°C [12]. This system thus seems to be convenient to study electron reactions. In fact, it has earlier been demonstrated that gamma irradiation of nucleic acid bases in aqueous alkaline solutions at -196°C results in the formation of the corresponding anions which get hydrolysed at higher temperatures to give neutral H-addition radicals [11 ]. The similarity between the spectra of each base, and its corresponding nucleoside and nucleotide, indicates that the addition of sugar and sugar phosphate groups does not affect the nature of the initial damage due to the electron reaction which is restricted to the base moeity only. 78 SINGH et al.

TABLE I. EFFECT OF SUGAR AND SUGAR PHOSPHATE MOIETIES ON THE YIELDS OF BASIC RADICALS

Sensitization factora

Base Nucleoside N ucleotide

A denine -0 .1 0 -0 .2 4

Guanine 0 .0 9 -0 .2 0

T hym ine 0 .1 2 0 .4 4

Cytosine -0 .1 4 -0 .0 3

a The negative sign indicates lesser yield as compared to the corresponding base.

Sensitization factor = ' 1

G(N) - Yield of radicals after warming to -95°C in the nucleoside or nucleotide. G(B) - Yield of radicals after warming to-95”C in the base.

However, such groups affect the stability of the base anions towards optical bleachingand thermal annealing. This is clearly evident for adenine and thymine anions. In adenine, the hydrolysis of the anion does not occur at -160°C, whereas in adenosine and adenylic acid the secondary purinyl radi­ cals appear at this temperature. In thymine, only the anion is seen after irradiation at -196°C whereas in thymidine and thymidylic acid, thymyl radicals appear at the temperature of irradiation indicating partial hydro­ lysis of the anion. Furthermore, the e.s.r. signal from the thymine anion, T , could be photobleached in thymine but exposure to light of thymidine and thymidylic acid samples results in the formation of thymyl radicals. Irradiation of aqueous solutions of DNA results in the release of in­ organic phosphates predominantly due to the reaction of hydrated electrons and hydroxyl radicals. Although evidence for the formation of radicals characteristic of sugar has been obtained in adenosine, adenylic acid and Poly A irradiated in dry state [14], the similarities noticed in the spectra of bases, nucleosides and nucleotides in the present study preclude the formation of any radicals characteristic of sugar and sugar phosphate moeities. Our results therefore suggest that the radiation damage due to electron reaction is localized at the base leading to the inference that the disruption of the inter­ nucleotide linkage is mainly because of hydroxyl radical reactions. F ro m the e.s.r. studies made on irradiated dry bases, nucleosides and nucleotides, Muller [15] observed higher yields of radicals in some nucleo­ sides and nucleotides. A similar sensitization of adenine by the addition of sugar and sugar phosphate groups has been shown by Singh and Charlesby [14]. We have also reported an increase in the yields of base anions and their corresponding secondary radicals in the order: Base < nucleoside < nucleo­ tide [16]. These yields had been calculated on the basis of unit mass of the base present in each sample assuming a linearity between the yield of radi­ cals and concentration of the solute. Table I, however, shows the sensiti­ zation factors calculated according to Muller [15] at a concentration of 0.15M, which corresponds to the saturation value. It can be seen that, except in the case of thymine, no sensitization is observed. LAEA-SM-134/1 79

The problem of localization and migration of radiation damage over the DNA molecule is of considerable interest. The e. s. r. signal of DNA irradi­ ated at -196°C is a single line of about 20 gauss line width which is distinctly different from that of thymine. The e.s.r. signals from G , С and A are also single lines but the presence of G" in DNA can be ruled out on the basis of its line width which is only 14.5 gauss. Furthermore, the DNA signal is considerably narrower than that of С (ДН = 27 gauss) whereas it is comparable to that of A (AH = 22 gauss). Although one cannot arrive at a definite conclusion, the above observations can be taken to indicate that electron reaction with DNA initially occurs at adenine. Since thymyl radicals are observed on warming the DNA sample to higher temperatures, intramole­ cular migration of radiation damage from adenine to thymine is indicated [17].

CONCLUSIONS

On the basis of the nature of radicals formed in nucleic-acid bases, nucleosides and nucleotides as a result of electron reaction, it appears that the site as well as the extent of the radiation damage are unaffected by the presence of sugar and sugar phosphate groups. The damage is restricted mainly to the base moeity and no radicals characteristic of sugar or phosphate groups are observed suggesting the absence of internucleotide breakage. Some evidence is also obtained to indicate that in DNA the electron is initially localized at adenine which later migrates to thymine.

REFERENCES

[1] WEISS, J., Prog.Nucleic Acid Res. Mol.Biol. 3 (1964) 136. [2] KAPP, D.S., SMITH, K.C., Radiat.Res. 42 (1970) 34. [3] KANAZIR, D .T., Prog.Nucleic Acid Res. Mol. Biol. 9 (1969) 117. [4] EPHRUSSI-TAYLOR, H., LATARJET, R., Biochim.biophys. Acta 16 (1955) 183. [5] DeFILLIPES, F.M ., GUILD, W.R., Radiat.Res. Д ( 1959) 38. [6] COLLYNS, B., OKADA, S., SCHOLES, G., WEISS, J.J., WHEELER,C.M ., Radiat.Res. 25 (1965) 526. [7] SINGH, B.B., Proc.Symp. ' Macromolecules in Storage and Transfer of Biological Information! Published by DAE Bombay (1969) 211. [8] ORMEROD, M .G., SINGH, B.B., Int.J.radiât.Biol. 6 (1966) 533. [9] NICOLAU, C l., McMILLAN, M., NORMAN, R.O.C., Biochim.biophys Acta 174 (1969) 413. [10] NICOLAU, Cl. (personal communication). [11] SRINIVASAN, V .T ., SINGH, B.B., GOPAL-AYENGAR, A.R., Int.J.radiat.Biol. 17 (1970) 577. [12] KEVAN, L., Prog. Solid State Chemistry 2 (1965) 304. [13] HENRIKS EN, T . , R ad iat. Res. 23 ( 1964) 63. [14] SINGH, В.В., CHARLESBY, A., Int.J.radiat.Biol. 9 (1965) 157. [15] MULLER, A ., I n t.J .ra d ia t.B io l, ji (1964) 131. [16] SINGH, B.B., SRINIVASAN, V .T ., GOPAL-AYENGAR, A.R., Ill International Biophysics Congress, Cambridge, USA (1969). [17] SINGH, B.B., Adv.Biol. Med. Phys. 12 (1968) 245.

DISCUSSION

H. HOFER: Could the base alterations described in the paper be of some biological importance? Or, in other words, to what extent could they occur when irradiation is performed at room temperature? 80 SINGH et al.

N.K. NOTANI: These investigations were carried out at low temperatures only for convenience. The species could be trapped at low temperatures, and this facilitated the investigations into their nature and reactions. The same sequence of reactions would occur even at room temperature and the same species would be formed, but they would be so short-lived that detailed investigation of them would be almost impossible with a conventional e. s.r. spectrometer. The biological consequences of base destruction by ionising radiations have been emphasized by Collyns et al. (see Ref. 6 of the paper) and are at present being investigated in detail in our laboratory. H.I. ADLER: The paper discussed the role played by various free radicals in the inactivation of DNA. Do you think that the molecular species, hydrogen peroxide, plays any part in the phenomena observed? N.K. NOTANI: Since the radicals responsible for the production of hydrogen peroxide during water radiolysis are highly reactive with respect to DNA and its constituents, the contribution of hydrogen peroxide to the inactivation of DNA would not be significant. GENETICS AND PHYSIOLOGY OF INDUSTRIAL MICROORGANISMS (Session 3)

Chairman

S.I. ALIKHANLAN (Union of Soviet Socialist Republics)

IAEA-SM-134/24

APPLICATION AND IMPORTANCE OF FUNGAL GENETICS FOR INDUSTRIAL RESEARCH*

K. ESSER Lehrstuhl für Allgemeine Botanik, Ruhr-Universität Bochum, Federal Republic of Germany

Abstract

APPLICATION AND IMPORTANCE OF FUNGAL GENETICS FOR INDUSTRIAL RESEARCH. The genetic control of breeding systems, their action and interaction as regulatory mechanisms for recombination of the genetic material, the production of mutants and their fixation in the genome, and the phenomenon of senescence are discussed with respect to the breeding of strains for the fermentation industry. The strains discussed have an optimal effectiveness and may keep these characters constant over a long range of propagation.

Although the whole spectrum of fungal genetics may be comprehended under the title ot this symposium, this review will be restricted to topics that are of immediate importance for the practical uses of fungi in the fermentation processes. In this connection there seems to be no need to enumerate all the fungi that are of interest to the fermentation industries. Also excluded from the scope of this paper are the details on the mode of action of the microorganisms in producing their special metabolites of interest or in "catalysing" biochemical reactions responsible for the transformation of a given substrate into the desired product. From a geneticist's point of view one of the main problems of research involved in fungal fermentation seems to be the breeding of suitable strains with an optimal effectiveness and the possibilities to keep these characters constant over a long range of vegetative propagation (i. e. stable strains). This requires not only the consideration and the knowledge but also the practical use of several parameters which are dealt with in this paper.

I. "BREEDING SYSTEMS", THEIR ACTION AND INTERACTION

The essential prerequisites for handling a fungus in all procedures of breeding (e. g. production of mutants, recombination of desirable genes, etc. ) are not only the proper knowledge of its reproductive cycles (including all the modes of transmission of the genetic traits by sexual and other accessory means) but also of the genetic controls of those processes. The latter topic is dealt with in the considerations.of the breeding systems. Since there is much confusion in the literature over the generally adopted definitions, a brief description is given, for the benefit of non-mycologists, of the various breeding systems encountered in fungi. These are summarized in the schematic representation in Fig. 1.

* This review does not quote individual papers. The complete references can be found in: ESSER, K. and KUENEN, R., Genetics of Fungi, Springer-Verlag, Berlin-Heidelberg-New York (1967).

8 3 84 ESSER

HETEROKARYOSIS DiOECISM HOMOGENIC (self-compatible and morphological physiological INCOMPATIBILITY self-incompatible species, (Achlya) (Phycomyces) (Neurospora, Schizophyltum) fungi imperfecti)

Sterility genes may interfere with all systems Heterokaryosis and inhibit may cancel the karyogamy. effect of sterility - genes by non­ allelic comple­ mentation in perfect fungi. ¥

MONOECISM HETEROGENIC HETEROGENIC (all self-compatible INCOMPATIBILITY INCOMPATIBILITY fungi, Sordaria etc.) (sexual and vegetative phase, (vegetative phase,e.g.Podo- eg.Podospora races) spora, Neurospora, Asper­ gillus and fungi impertecti)

FIG. 1. Action and interaction of breeding system in fungi. The large square in the centre displays the main systems. Heterokaryosis occupies the right side of the figure. Representative organisms for each system are given in parentheses. The rectangles represent single individuals (except of heterogenic incompatibility in the centre part, where the rectangles represent single Podospora races). The male and femal symbols represent nuclei of different sex. Differences in the genetic equipment of nuclei are characterized by white and black. In the case of physiological dioecism and heterokaryosis, where sexual differentiation of the nuclei cannot be proved, they are symbolized by white and black circles. The thick arrows indicate the direction of karyogamy or heterokaryotization respectively. The blocked arrows indicate that karyogamy or heterokaryotization is impossible. Interactions of the different systems are depicted by thin arrows. (From Esse r, K., Molec. Gen. Genetics 110 (1971) 86-100.)

1. Monoecism and dioecism are the two basic breeding systems. They depend on the capacity of an organism to contribute one or both nuclei to karyogamy. From this simple criterion of sex it follows that a monoecious individual can function both as donor and as recipient of a nucleus. An individual which exhibits only one or the other potential is called dioecious. There are no special genes controlling monoecism since all those numerous factors which are responsible for the differentiation of an organism have to be disregarded in this connection. In contrast to the higher organism dioecism in fungi is not controlled by sex chromosomes but merely by single genetic traits.

2. Incompatibility may be defined as a genetically determined prevention of karyogamy which is not caused by sterility defects of the participating nuclei. There are two systems which show opposite genetic control:

(a) Homogenic incompatibility where karyogamy is prevented when the nuclei carry identical incompatibility factors. In the most simple case a single gene difference consisting of one allelic pair (usually called + and - or A and a) is sufficient to control the sexual reaction (Ascomycetes). This general principle holds true even in the more complicated cases, described as tetrapolar incompatibility in Holobasidiomycetes, where all together four genes with numerous allelic configurations are instrumental. IAEA-SM-134/24 85

(b) Heterogenic incompatibility where karyogamy is prevented when nuclei carry non-identical incompatibility factors. In contrast to the homogenic system there is no case known where a single gene difference is sufficient to block the sexual compatibility. In the most thoroughly analysed case of the Ascomycete Podospora anserina at least four loci are required to have different allelic configurations. Heterogenic incompatibility seems to be widespread within all groups of fungi (Esser, K. and Blaich, R., in preparation). Another difference between the homogenic and the heterogenic systems consists in the fact that, in the latter system, incompatible genes may neither coexist in the same nucleus nor in a common cytoplasm when the genes are located in different nuclei of a heterokaryon. Therefore, heterogenic incompatibility is also efficient in the vegetative phase and inhibits hyphal fusion.

3. Heterokaryosis, the association of genetically different nuclei in a common cytoplasm, is widespread and unique to the fungi. It is not a breeding system per se, like the ones previously discussed. However, in this context it needs to be mentioned that heterokaryosis by initiating the plasmogamy in some somatogamic fungi controls the sexual cycle. The most eminent function, however, is contributed to heterokaryosis in the Fungi Imperfecti, where hyphal fusion is the only means to start the parasexual cycle. Many species of industrial significance, e. g. Aspergillus spp., Actinomycetes, Streptomyces aureofaciens and other members, take recourse to the parasexual cycle which leads to the genetic recombination of characteristics. Thus, the genetic control of heterokaryosis depends on both categories of genes, those resp o n sib le for hom ogenic (e. g. m ating factors of Holobasidiomycetes) and those for heterogenic incompatibility. After these somewhat complicated definitions, one may now ask for the above-mentioned practical meaning of these breeding systems and their relationship to the statement that "they control the recombination of the genetic material". In the microbial fermentation industries, so far, strain improvement has been based almost exclusively on induced mutations and selection, and the potentialities of the far more powerful tool of recombination have been ignored. The biological significance of recombination in creating new genetic endowments cannot be over-emphasized. This is even more strengthened in the light of rapidly increasing knowledge revealing the versatility of the mechanisms of recombination in the microorganisms. In regarding the breeding systems in this respect it follows that the effect of recombination is generally zero in true breeding strains and very poor in inbreeding organisms. This means that any mechanism which prevents or decreases inbreeding enhances the efficiency of recombination by an increase of outbreeding. This is achieved in nature by the action and interaction of the various breeding systems, and their unique advantages can be exploited in the appropriate industrial research. The following are of interest:

1. Inbreeding is decreased, above all, by dioecism. Both morphological and physiological mechanisms prevent self -fertilization and require for karyogamy two individuals which differ at least by sex or mating factors. 8 6 ESSER

The same effect is achieved in many monoecious organisms by homogenic incompatibility. Because of this limit of inbreeding, wild strains that are controlled by one of these systems are, to a large extent, heterogeneous.

2. Inbreeding is increased most drastically by monoecism. There is little chance for a quick distribution of genic diversity within a species. This may lead to formation of local races with different genetic make up. This sort of genetic isolation is greatly favoured by heterogenic incompatibility, since this system effects a partial or complete block of mating between the various races within one single species. As a result of this tendency towards isolation, recombination becomes rather limited and, in practice, restricted to single races.

3. The action of heterokaryosis is not as evident as that of the sexual systems. However, it must be realized that in nature the mixing of nuclei in the vegetative hyphae increases the likelihood of outbreeding. Nuclei containing different genetic information, when brought together by hyphal fusion, may easily become incorporated into the germ line. The importance of heterokaryosis becomes most evident in the Fungi Imperfecti where it is the sole mechanism for genetic recombination. Several fungi of great industrial importance belong to the group of Fungi Imperfecti. Here a few remarks would be appropriate on the practical importance of heterokaryosis. In diploid organisms, spontaneously occurring deleterious mutations, which are mostly recessive, fail to become very effective in the vegetative cells since they are covered up by the action of their dominant "normal" alleles. In the haploid organisms (like most of the fungi of industrial interest) this is not the case and such mutational events would lead to an inviability of their carriers. However, this is prevented by heterokaryosis which mimics, to a certain extent, the "diploid phase" in creating the possibility of a coexistence of heterogeneous nuclei in a common cytoplasm. Thus, in fairly balanced heterokaryons recessive detrimental mutations will not become expressed. An enrichment of such "negative nuclei" (i. e. for the industrial goal of the microbial strains) may lead to an unpredictable behaviour of industrial fungi. In this connection notice should be taken of the fact that some water moulds (Oomycetales) that are unable to undergo heterokaryosis by hyphal fusions have been found to be diploids in their genetic composition.

4. The action of each breeding system may be partially or totally cancelled by sterility genes thereby causing deficiencies in male or female sex organs, or in both. Genes of this kind are often found in natural isolates and occur in some laboratory strains after prolonged vegetative propagation. Such defects may be overcome after heterokaryosis via non-allelic complementation which leads to fertile heterokaryons with wild phenotype. It should be realized that the control of the sexual cycle and, to a certain extent, also of the parasexual cycle and the consequent control of recombination in most fungi is not solely achieved by a single breeding system, but by the interaction of various systems. The most common interaction is the nullification of monoecism by homogenic incompatibility. This attributes to a species consisting of hermaphrodite individuals the same outbreeding value as is present in a IAEA-SM-134/24 87 dioecious species. Heterogenic incompatibility may overlap this out- breeding effect of dioecism and the homogenic incompatibility in such a way that the opportunities for recombination are restricted to particular races. This phenomenon also occurs in the Fungi Imperfecti thereby preventing heterokaryosis and mitotic recombination. What are the consequences of the theoretical considerations presented in this chapter for the industrial mycologist? Since it is his aim to make use of strains which have an utmost constant "biochemical productivity", he must be able to regenerate his strains in "case of emergency" (if they have lost their desired properties during prolonged vegetative growth) and try to re-isolate his "original fungus". This is mostly done by cultures of single vegetative spores. As many fungi do not produce these spores, this technique is not adequate since it does not eliminate hereditary alter­ ations which are extrachromosomal. The best way to regenerate weak or leaky strains is via meiotic or mitotic recombination. In analysing a series of sexual or parasexual descendants there is much more chance to pick up again the "original nucleus" out of the mass of mutated nuclei or mutated cytoplasm which have led to a heterokaryotic status of the strain. The author does not know how often genetic regenerations of this kind are followed in practice. However, from his experience, it is advisable to regenerate all fungal strains at regular intervals in order to avoid the undesirable consequences from their "degeneration" which sometimes can only be solved the hard way by eliminating the strain altogether. For an industrial enterprise this is far from being an economically feasible proposition.' Furthermore, it is not enough to keep the properties of a strain constant. On the contrary, a planned breeding program is necessary to enhance its productivity. This, however, requires primarily the close familiarity with formal genetics of reproduction, the complicated sequences of the action and interaction of the various breeding systems involved and the theoretical background for producing new strains. There are several ways to succeed: (1) Analogous to plant and animal breeding, a systematic crossing program between the various geographic races may lead (via recombinational events) to new gene combinations with enhanced selective (in this case, productive) value. (2) Production and selection of mutants using the available array of chemical and physical mutagens. This is dealt with in more detail in a later part of this paper. (3) A combined mutation, selection and recombination program for the development of improved strains of industrial microorganisms.

II. MUTATION-INDUCTION AND THEIR FIXATION IN THE GENOME

Mutational events, which lead to discontinuous and heritable alterations of the genetic material, may arise spontaneously as well through the' action of mutagenic agents. A concerted breeding of fungi in industrial research cannot exclusively rely on the slow processes of spontaneous mutation and their natural selection but needs a systematic use of mutagenic agents for rapid results. Since it is not the purpose of this paper to review special techniques for the production of mutants, only some general rules are given: 8 8 ESSER

1. Fungal structures to be treated with mutagenic agents should, preferably, not contain more than one nucleus. Most asexual spores of the conidial type, e. g. Aspergillus, Pénicillium, Neurospora (microconidia) are uninuclear. In using sexual spores, e. g. the ascospores of Neurospora and related fungi, one has to consider that these spores (though uninucleated in the beginning of their development) are multinucleated after ripening due to continuation of the mitotic divisions of the original nucleus.

2. If this condition cannot be fulfilled in a specific organism and one has to use polynucleated spores or even mycelial fragments, most survivors from the mutagenic treatment will grow out into heterokaryotic mycelia. This again may lead to a camouflage of some mutated nuclei via interallelic complementation. This difficulty may be overcome by analysing a number of descendente from these heterokaryons that have originated either from uninucleated vegetative spores, hyphal tips or sexual spores. Thus, from a single heterokaryon a broad spectrum of mutants may be obtained.

3. The selection of the mutagenic agent has been correlated to the fungal structure to which it is supposed to be applied. For example, u.v. -treatment of fungal spores which are coated with dark melanin pigments is very inefficient because of the poor u. v. pene­ tration of the cell.walls. In this case more penetrating ionizing radiations (X-rays, gamma rays) may be a better choice.

4. The utilization of selective techniques is not only less time consuming but also the most effective method to obtain mutants in a greater frequency. In contrast to the geneticist who is mainly interested in obtaining mutants with biochemical deficiencies, the industrial mycologist is pre­ dominantly searching for mutants which have acquired better biochemical properties that assure a bigger economic return. Therefore, all the selective techniques which allow only the recognition of deficiency mutants cannot be used for this purpose. The most advisable technique would be to plate the treated material on agar media containing a specific chemical reagent which allows screening of the desired mutant due to a colour reaction or a morphological peculiarity.

5. Chemical mutagenic agents have proved, within the last years, from may points of view to be much more suitable than any kind of radiation to obtain biochemical mutants. As the chromosomal deletions and rearrangements are believed to occur less frequently by the chemical mutagens, the chance to obtain point mutations is consequently higher. Furthermore, their mutagenic effect on the DNA molecules is fairly well understood in the molecular terms. The main problem in producing biochemical mutants is not to produce mutations and to select specific strains but to obtain a stable and viable mutant. It is nearly trivial to mention that all strains with altered properties selected after mutagenic treatment are not true mutants but only variants, since their phenotype may be the result of various events, e. g. nuclei with chromosomal duplications which are even unstable after several mitotic divisions and heterokaryosis, etc. Genuine mutants are only received when the mutated nucleus has passed through a meiotic IAEA-SM-134/24 89 division. By this procedure all kinds of unbalanced genomes will be eliminated and, furthermore, the mycelium of the mutant originates from one single nucleus which warrants it to be homokaryotic. The simplest method is to backcross the variant with the wild strain. This procedure will also give valuable information on the nature of the mutation by evaluation of the segregation pattern and, in turn, will enable one to distinguish between monogenic, polygenic and stable chromosomal m utatio n s. The most complete information will be obtained from those fungi which allow the isolation of ordered tetrads of spores. In this case the localization of the mutated gene could be predicted with reasonable accuracy. With polygenic mutations the method of backcrossing also allows the incorporation of each single mutated gene from the variant via recombination during the meiotic events. Thus, by installing only one mutated gene in the wild genome the chance to obtain not only stable, but also viable, strains is much enhanced. In some monoecious fungi it is difficult to distinguish between fruiting bodies originating from selfings of both partners or from crossing. This may be overcome in two ways: (1) By crossing more or less sterile mutants which form only a few fruiting bodies in the zone of contact. (2) By using spore colour-marker genes for the two partners of the mating. The fruiting bodies resulting from crosses can be distinguished by segregation of the spore colour genes whereas those from selfing contain only the one or the other spore colour. It may be recalled that a great majority of the industrially useful fungi are lacking a well-defined sexual cycle (like most Aspergillus species) and are therefore not suited for sexual propagation. In these cases the fixation of mutations may be obtained by the parasexual cycle via mitotic recombination. The use of marker genes (e. g. conidial colour) are invaluable under such circumstances. In concluding these expositions about the production of mutants it seems necessary to point, at least briefly, to the phenomenon of pleiotropy, the production of apparantly unrelated effects at the phenotypic level, caused by a single mutated gene. Strains exhibiting such a "genetic syndrome" are relatively often found in experiments screening for mutants. They exhibit mostly joint alterations in morphology and physiology (e. g. alterations in growth and hyphal morphology, loss of sex organs or conidiophores, auxotrophy, etc. ). Sometimes these evident deficiencies may be correlated with an enhancement in productivity for the desired metabolites. In the latter case it is necessary to establish its true pleiotropy and to exclude polygenic or chromosomal alterations which may not guarantee a permanent stability. This can be achieved by back- mutation tests. Usually the point mutational events involving one gene alteration can revert by back-mutation to its original wild-type characters (cf. suppressor mutation).

III. THE PHENOM ENON OP SENESCENCE

It is well known to mycologists that strains of fungi maintained in culture collections through serial transfer of mycelia become senescent. The same event occurs sometimes more rapidly after continuous mass 90 ESSER production in fermenters. Senescent strains first lose their capacity to reproduce sexually and, in the case of pathogens, their pathogenicity is lost. This is often followed by a degeneration of the mycelium which may result in a complete cessation of hyphal growth. Such senescence syndromes are often viewed as spontaneous mutational alterations of the genome. This explanation does not hold generally, since in many fungi degeneration is nullified through plasmogamy with young mycelia via sexual reproduction or heterokaryotization. In some well-analysed cases it has been demon­ strated that the senescence is determined by a mutation of the plasmone and thus is maternally inherited. Therefore in organisms with oogamous reproduction (like many Ascomycetes), senescent strains can be easily rejuvenated by the backcrosses of senescent male parents with the suitable juvenile female counterpart. In one other case, rejuvenation has been achieved after storage for several months at 4°C. However, the problem of senescence; has turned out to be somewhat more complicated and not just a simple case of extrachromosomal inheritance. It has been shown that in some fungi senescent hyphae are able to "infect" juvenile hyphae by a cytoplasmic bridge when brought in contact. This phenomenon has been explained by the fact that senescence is determined by infectious particles in the cytoplasm. These particles of so far unknown nature are supposed to be self-replicating as they are capable of multiplying in the cytoplasm of the juvenile strains and migrate through the hyphal septae. Once an infection of a juvenile strain has occurred it leads to complete senescence of the mycelium in a short time. Since senescence is known to occur in each wild strain sooner or later one must take into account the fact that this syndrome may be of an infectious nature and would be able to destroy the whole strain. As a pre­ cautionary measure we propose that regneration of the strains be practised at regular intervals by means of cross matings.

DISCUSSION

The paper discusses some points which might be of interest to an industrial mycologist for the improvement of the yield of strains used in fermentation processes. The main objection to these ideas is the argument that the empiric method (to use the strain as long as it is a good producer and then to throw it away and start with a new isolate), which is widespread in fermentation industries, is much more economic than to install a laboratory for basic genetic research. On the surface, this argument seems to be economically justifiable because small companies would not be able to afford the costs of basic research. However, central laboratories, which are not merely mykothekes, would probably by able to solve these problems on a broader basis and fill the gap between theory and practice. This would achieve a more rational and successful use of fungi in industrial fermentation processes.

DISCUSSION

N. K. NOTANI: Do I understand correctly that the episome-like particle, on changing from the integrated to the free phase, brings about a change in the expression but produces no lysis? IAEA-SM-134/24 91

K. ESSER: Yes, you are right. J. WEIJER: Would you agree with the statement that parasexuality as a result of heterokaryosis is genetically as effective in bringing about recombination as karyogamy? K. ESS ER: This is exactly my opinion. S. I. ALIKHANIAN: What do you mean by senescence of fungi, and what are its indications? K. ESSER: By senescence I mean the ageing of mycelia after a prolonged vegetative propagation. Its indications are decrease in growth rate, loss of metabolic properties and alterations in morphology. S.I. ALIKHANLAN: How do senescent fungi differ from the juvenile ones? K. ESSER: They are distinguished by chromosomal or extrachromosom alterations or by both.

IAEA-SM-134/5

PHYSIOLOGICAL AND GENETICAL STUDIES ON YEASTS OF THE GENUS Candida

C . GAILLARD IN, H. HESLOT Institut national agronomique, Paris, France

Abstract

PHYSIOLOGICAL AND GENETICAL STUDIES ON YEASTS OF THE GENUS Candida. The degree of ploidy has been studied in Candida tropicalis and Endomycopsis lipolytica by determining the DNA content, the shape of the survival curve after X-irradiation and the response to mutagenic agents. Sexual behaviour has also been investigated. A number of auxotrophic mutants were induced.

INTRODUCTION

A number of yeast species are able to utilize paraffins as their carbon source. From an industrial point of view this is an important characteristic because it enables proteins to be made at low cost, especially in those developing countries where petroleum is available. Yeast strains used for this purpose belong to the genus Candida, and more precisely to Candida tropicalis and Candida lipolytica. The latter has recently been shown by Wickerham et al. [14] to sporulate when complemen­ tary mating types are mixed on appropriate media. The discovery of the perfect sexual stage has shown that the species is a member of the genus Endomycopsis so it has been renamed E_. lipolytica. Genetic studies on E_. lipolytica are at present being made in our laboratory, for example, studies on the induction of auxotrophs, segregation at meiosis and at mitosis etc. Although no sexuality has been described in C_. tropicalis, we considered that it would be interesting to investigate its genetic status. Several methods were used for this purpose: shape of the gamma-ray survival curves, measurement of DNA content and p-fluorophenylalanine treatments.

M ATERIAL AND METHODS

Strain

We used a strain given us by the Institut Français du Pétrole, Candida tropicalis 29a, originally isolated from soil.

Culture media

The complete medium contained: yeast extract 0.5%, glucose 3% and agar 1.5%. The minimal medium was that described by Leupold [8 ] which contains a few growth factors (calcium pantothenate, nicotinic acid, meso- inositol and biotin). We also used the nitrogen-poor medium of Elkind and Beam [4] and the phosphate-poor medium of Jeener and Brachet [7].

93 94 GAILLARDIN and HESLOT

Haploidization was induced by culturing on complete medium supplemented with 5 g/1 p-fluorophenylalanine (p-FPA). We also used minimal medium supplemented with 100 mg/l 5-fluorouracil (5-FU).

Growth curves

Liquid cultures were made in Klett vials containing 20 ml of medium which was seeded with 5 X 105 cells/ml and constantly agitated at 30-31°C. Growth was followed with a nephelometer.

Irradiation

Irradiation was performed in a gamma cell at doses ranging between 25 and 250 kR.

Mutagenic treatments

Two alkylating agents were used, either nitroso methyl urethane (NMU) 5 X 10 '4 (v/v) or nitroso-guanidine (NG) 300 7 /ml. Treatments were for 1 hour at 2 5°C. The ultraviolet dose was 2000 ergs/cm 2, and the gamma-ray dose was 75 kR. Stable auxotrophs could only be induced by NG and by gamma rays.

DNA determ ination

The method of Burton [2] was used, with a few modifications:

2 0 m in in TCA 10% at 0°C 3 washings in TCA 5% 2 washings in ether-alcohol (3; 1 ), 3 min at 60°C 3 washings in TCA 5%.

MICROBIOLOGICAL STUDIES

On solid medium, strain 29a forms dull-white round colonies with an irregular margin, generally with folds in the centre. Individual cells show a great range of sizes from 1 to 10 ^m, Most of them are between 4 and 5 jum. At the end of the exponential phase numerous hyphae are formed, some of them reaching a length of 40-50 /um. As shown on Fig. 1, the strain exhibits a growth optimum at 30-31°C, but is still able to grow at 37-38°C. The mean generation time on complete glucose medium is 45 min. Strain 29a assimilates glucose, sucrose, maltose and galactose. It is able to ferment all these sugars, especially glucose. It also grows perfectly on ethanol or glycerol at 28°C. The minimal medium of Leupold [8 ] that we use contains two groups of growth factors namely: (a) calcium pantothenate 1 mg/l; nicotinic acid 10 mg/l; mesoinositol 10 mg/l; and (b) biotin 0.01 mg/l. We studied the effect of suppressing (a) or (b), or of adding thiamine. lAEA-SM-134/5 95

FIG. 1. Growth of strain 29a on complete medium at 25 °C (fjm = 0. 5), 30 °C (fjm= 1.3) and 35е С (fim = 1).

Figure 2 shows that strain 29a grows on a strictly minimal medium, without any growth factor. The addition of biotin or thiamine has no effect (generation time (tg) = 105 min). However, the addition of the (b) mixture distinctly im proves the growth rate (tg = 52 min at 25°C; tg = 45 m in at 30°C) as well as the final yield.

SENSITIVITY OF STRAIN 29a TO GAMMA RAYS

We tested comparatively the radioresistance of strain 29a cultivated either on complete medium or on Leupold minimal medium. Cells were irradiated either during the exponential or the stationary phase. Figures 3 and 4 show the survival curves. Cells in the exponential and stationary phases show a strikingly distinct behaviour. In the first case the curves have a shoulder and the initial straight part of the curve extrapolates at 200% which would be in favour of a diploid state. In the second case a typically one-hit curve is obtained which would point to a haploid state. These results agree well, as we shall see later, with the DNA content of exponential- and stationary-phase cells: there is always a 2 :1 ratio between these two states. 96 GA ILL ARD IN and HESLOT

F IG .2. Growth of strain 29a on different minimal media (MM) at 25 and 30°C A : 25°C (|im = 0 . 75) MM, or MM + (b) or MM + thiamine В : 30°C (мш = 0.7 5 ) MM, or M M +(b) or MM + th iam in e С : 25°C (p m = 0 .8 ) MM, or MM + (a) or MM + (a) + (b) D : 30°C ( д т = 1) MM, or MM + (a) or MM + (a) +(b).

In the stationary phase an important fraction of the population is more resistant and still survives at the highest dose applied (250 kR). We in­ vestigated the possibility to alter this behaviour of stationary-phase cells by growing them, for 8 hours, on each of the following culture media:

(1) The Elkind and Beam medium, poor in nitrogen. (2) The Jeener and Brachet medium, poor in phosphates.

Figure 5 indicates that the radioresistant fraction of the population disappears when incubated in a nitrogen-poor medium. On the other hand, transfer to a phosphate-poor medium has no effect. These results seem to imply that a depletion of the amino-acid pool is effective. As Schmidt et al. IAEA-SM-134/5 97

FIG.3. Gamma-ray survival curve of strain 29a (exponential phase). О Minimal medium ® Complete medium.

[1 2 ] have shown that a starvation of phosphates leads to a depletion of nucleo­ tides and total RNA in Saccharomyces cerevisiae , our data can be inter­ preted to show that nucleic-acid synthesis is not involved in the resistance phenomenon. In order to test this hypothesis we tried to alter or to block either the protein or the m-RNA synthesis. The first objective can be attained by incubation in minimal medium supplemented with p-FPA, or by using an amino-acid auxotroph (asp~ lys~) kept for a few hours in minimal medium. The second objective is achieved either by incubation with 5-FU or by using a mutant ura~ that has been kept for some hours in minimal medium. F ig u re s 6 and 7 lead us to conclude that interference with protein synthesis increases the sensitivity, whereas a block in the RNA synthesis has little or no effect. 98 GAILLARDIN and HESLOT

FIG.4. Gamma-ray survival curve of strain 29a (stationary phase). О Minimal medium ® Complete medium.

The fact that a population of stationary-phase cells comprises a resistant fraction must be compared with analogous observations of Moustacchi [11] on S. cerevisiae. She showed that buds were particularly radioresistant due to a proteinaceous-resistance factor. In our experiments this would explain the sensitizing effect of 5-FU, p-FPA and of amino-acid auxotrophic mutations. We found that the sum of budding cells and of big cells (> дш) was constant in all experiments and equals 18%. This is precisely the fraction of the population which is resistant (see Figs 5, 6 and 7).

TREATMENT OF STRAIN 29a BY p-FLUOROPHENYLALANINE

As the gamma-ray survival curves indicated that strain 29a was possibly diploid, attempts were made to confirm this assumption by reducing it to the haploid state. To do this, a treatment was made with p-fluorophenylalanine (p-FPA ). This compound, which is incorporated into proteins, is known to induce chromosome losses at mitosis thus favouring production of haploids. Success has been achieved by Lhoas [9] and McCully and Forbes [3] with Aspergillus niger and A_. nidulans, and by Gutz [6 ] and Adondi and H eslot [1] with Schizosaccharomyces pombe. Emeis [5] made unsuccessful attemps with S. cerev isiae. IAEA-SM-134/5 99

FIG. 5. Gamma-ray survival curve of strain 29a. A : Phosphate-poor medium (stationary phase) В : Nitrogen-poor medium • C ontrol О Incubation in poor medium.

By plating 107 cells on a Petri dish containing complete medium supple­ mented with 5 g/1 p-FPA, only a few hundred small colonies developed after 5 days. These colonies contain cells of a sub-normal size. Some were isolated by micromanipulation, transferred into normal medium and then the population was subjected to a cell-size analysis (Fig. 8 ). If a haploid is supposed to have half the volume of a diploid the linear dimensions should be in the ratio D/H 1.26. This is almost exactly the value found for the three strains Hl, H3 and H4, compared with 29a. 100 GAILLARDIN and HESLOT

100 15 0 2 0 0 2 5 0 GAMMA DOSE {kR)

FIG. 6. Gamma-ray survival curve of strain 29a (stationary phase) A : Incubation with p-FPA В : Incubation with 5-FU в C ontrol О T re a te d .

More generally, out of 50 colonies selected on the p-FPA medium:

38 gave a ratio of 1 6 gave a ratio of 1. 23 - 1. 28 6 gave a ratio of > 1.28

The latter died quickly which leads us to suppose that they had less than the haploid complement of chromosomes. IAEA- SM-134/5 101

FIG.7. Gamma-ray survival curve of strain 29a (stationary phase). A : M utant ura~ В : Mutant asp" lys" • Supplemented medium О Minimal medium.

Six presumed haploids, HI to H6 , whose linear dimensions showed a D/H ratio of 1.26, were subjected to a number of tests to make sure of their degree of ploidy.

(a) Gamma-ray irradiation

Strain 29a and HI to H6 were gamma irradiated in the dose range 25 to 250 kR. Figure 9 shows the survival curves of 29a, Hl, H2 and H5. Although there is a range in the radiosensitivity, they give one-hit curves. However, the multi-hit character of the 29a survival curve is more question­ able and these experiments, by themselves, are clearly not sufficient to draw a firm conclusion. 102 GAILLARDIN and HESLOT

7o. 7o

50 • 5 0 •

- ~l i—i i—i , 123456789 10 123456789 10 29 a Ш

FIG. 8. Cell-Size distribution of strain 29a and derived strains HI, H2, H3, H4 and H6. Limits of classes, expressed in microns, are as follows : 1 : 1.66 - 2.5 6 ! 5.8 - 6.6 2 : 2.5 - 3.3 7 : 6.6 - 7.4 3 : 3.3 - 4.1 8 : 7.4 - 8.3 4 : 4.1 - 5.0 9 : 8.3 - 9.1 5 : 5 .0 - 5.8

(b) Dry weights

With the same assumptions as above, one would expect the dry weight of haploid cells to be half that of diploids cells. Strain 29a and presumed haploids HI to H6 were cultivated in minimal medium until the stationary phase. Dry-weight determinations gave the results indicated in Table I. The agreement is good for Hl, H2 and H4 and less satisfacto ry for H3, H5 and H6 . IAEA-SM-134/5 103

FIG. 9. Gamma-ray survival curves of strain 29a and presumed haploids Hl, H2 and H5. • 29a; x HI ; + H2; О H5.

TABLE I. DRY WEIGHTS

Dry weight/108 cells H/D Strain (m g) (%)

29a 2 .2 2 5 100

HI 1 .1 2 2 50

H2 1 .1 4 5 51 .

H3 1.3 2 5 60

H4 1 .0 4 2 48

H5 1.4 4 3 65

H6 1 .2 6 5 57 104 GAILLARDIN and HESLOT

TABLE II. DNA CONTENTS OF THE STATIONARY PHASE

DNA expressed as y-deoxyadenosine/108 cells

29a 1 .7 1 5 1.0

HI 0.9 0 1 1 .9 3

H2 0 .8 5 7 2 .0 2

H3 0 .9 1 5 1 .9 2

H4 0 .8 6 0 2 .0 2

H5 0 .8 9 3 1 .9 6

H6 0 .8 7 5 2 .0 0

TABLE III. DNA CONTENTS OF THE EXPONENTIAL PHASE

DNA expressed as Strain D/H y-deoxyadenosine/108 cells

29a 2.7 1 4 1 .0 C andida . H I 1 .4 2 1.93 tro p icalis H2 1 .3 5 2.01

42 3 -3 0 .9 6 5 1 .9 4 Endomycopsis 4 2 3 -1 2 0.9 7 1 1 .9 2 lip o ly tica 423 bis 1 .8 5 1.0

423 12 J Haploid mating types (Wickerham)

423 bis : diploid ( Wickerham)

fc) DNA content

Cells were cultivated in minimal medium and the DNA content was measured in the stationary phase by the method of Burton [2]. The results are given in Table II. The D/H ratio was found to be very close to the theoretical value of 2. However, Mori and Onishi [10] found that the ratio between diploids and haploids of S. cerevisiae varied between 1.67 and 1. 72. This conflicting result prompted us to study the situation in Endomy- copsis lypolitica where both haploid mating types and diploid, from Wickerham et al. [14], are available. Results of the DNA contents (in the exponential phase) are reported in Table III. Here again the D/H ratio is n e a r 2. Therefore, there are good reasons to believe that haploidization was effectively induced in 29a by p-fluorophenylalanine treatment. IAEA-SM-134/5 105

TABLE IV. EVOLUTION OF DNA CONTENT

Number of transfers Strains 1 2 4 7

29a 2 .86 2 .8 8 2 .8 6 2 .8 5

1 1 .42 1 .4 2 1 .7 5 2 .7 5

2 1 .3 7 1 .8 7 2 .8 1 2 .8 6

3 1 .4 4 1 .81 2 .6 7 2 .6 7

4 1 .50 1 .6 7 2 .3 2 2 .8 1

5 1 .3 6 1 .4 8 2 .0 8 2 .7 6

6 1 .0 8 1 .2 5 2 .8 2 2 .8 3

7 1 .4 5 1 .4 7 2 .8 5 2 .8 6

8 1 .4 4 1 .6 7 2 .8 1 2 .8 5

9 1 .4 2 1 .96 2 .6 8 2 .8 4

10 1 .16 1 .6 8 2 .0 5 2 .7 8

11 1 .43 1 .4 4 2 .2 3 2 .8 6

12 1 .4 8 1 .5 8 2 .6 8 2 .7 9

13 1 .10 1 .4 5 2 .4 7 2 .8 7

However, further investigations showed that the haploid state was un­ stable. In order to study this phenomenon 13 new haploid derivatives of 29a were isolated and their DNA content was followed as a function of time and number of transfers into fresh medium. Table IV shows that the DNA content rises regularly until, after seven transfers, the 13 strains are indistinguishable from 29a.

COMPLEMENTATION EXPERIMENTS

A number of auxotrophic mutants were induced by gamma rays and NG in haploid H2, H3, H4 and H6, namely:

H2 (6) ura~ H3 (11) unknown req u irem en t H4 (6)asp~ H4 (9)his~ H6 (3) phe~try* and we alsohad at our disposal two mutants, already described, in strain 29a, namely: 29a (1) asp~ l.ys~ and 29a (32) ura~. Complementation experiments were performed by mixing, in minimal medium, strains with different requirements. The 21 combinations between the seven auxotrophs were tested. In all cases, except one, prototrophic colonies appeared at a frequency of at least 3-4 for 105 cells of each parent. 106 GAILLARDIN and HESLOT IAEA-SM-134/5 107

This is very superior to the background of revertants, estimated by appro­ priate controls. The com bination H4 (6) asp~ X 29a (1) asp' lys~ gave no prototrophs. This could possibly be accounted for by allelism of the asp~ mutations. The three combinations H3 (11)X H4 (6), H3 ( 11 )X H4 (9) and H4 (6)X H4 (9) gave rise exclusively to red prototrophs. This red colour disappears on minimal medium supplemented with 50 mg/l adenine. Because this was a good marking character, these red prototrophs were subjected to further a n a ly sis. A priori, several situations could give rise to prototrophs. For example: (1) h etero k ary o sis, (2) tru e diploidy o r (3) recom bination of p aren tal geno­ types. In an attempt to distinguish between these possibilities the prototrophs were treated with p-FPA and we tried to recover the parental genotypes. Table V shows the result of such experiments and proves that the red proto­ trophs contain the genetic information supplied by both parents. As explained above, haploids are unstable and when the DNA content of H3 (11), H4 (6) and H4 (9) was retested, a fter se v e ra l tra n s fe rs , it was found to be at the same level as that in the initial strain 29a. However, their auxotrophic requirements were still the same. The three combinations H3 (11) X H4 (6), H3 (11) X H4 (9) and H4 (6) X H4 (9) w ere rep eated on m inim al medium, but white prototrophs were exclusively obtained in the same propor­ tions as in the earliest crosses. Although their colour had changed, p-FPA treatment allowed us to recover, as previously, the parental types in equal proportions. When crossed, they gave also white prototrophs.

DISCUSSION AND CONCLUSIONS

The various criteria used to differentiate between strain 29a and p-FPA derivatives show that the first is presumably diploid and that the second are haploids. None of these haploids can be kept as such for a long period; there is always a shift towards the diploid state. Two hypotheses can be proposed to explain such a situation: (1) aneuploids were obtained rather than than haploids; (2) the haploid cells fuse to give diploids. Emeis [5] was unable to obtain true haploids of j3. cerevisiae by p-FPA treatment, but only aneuploids. These were relatively stable. With C. tropicalis ,29a, a similar situation possibly occurred. Then, during successive divisions, the diploid state would be restored. The other hypothesis is that haploid cells fuse spontaneously to give diploids. This would imply that incompatibility genes do not exist. We favour this second hypothesis because auxotrophs H4 (6) and H4 (9), derived from the same haploid H4, form prototrophs when mixed in minimal medium. So this strain of jC. tropicalis is presumably homothallic. As mentioned previously, mixing haploid with diploid or diploid with diploid auxotrophs on minimal medium also gives rise to prototrophs, but these prototrophs invariably become diploids if one relies on the DNA content. In a diploid X diploid combination, two assumptions can be made to explain these findings:

(1) Two diploid cells fuse but, shortly after, a mitotic reduction occurs giving rise to a diploid. However, a microscopic examination of these m ix tu r e s has not allowed us to detect big cells (4n). 108 GAILLARD IN and HESLOT

(2) Each of the diploids gives rise spontaneously to haploid cells. Some of these fuse to form diploid prototrophs.

As shown on Fig. 8, strain 29a has essentially big cells, but also small cells comparable in size to the smallest produced by the haploids. On the other hand, a class of big round cells with a thick wall (possibly chlamydospores), which are rarely present in strain 29a cultivated in -minimal medium and which have a high frequency in red prototrophs, bud by forming small cells which could be haploids. We have sent our strains to Dr. R. de Miranda, at the Centraal Bureau voor Schimmel Cultures, in Delft, for taxonomic studies. Dr. R. de Miranda (private communication) has seen chlamydospores and thinks that these strains of C_. tropicalis should be considered as belonging to a new species of Syringospora. Van der Walt [13] has recently described the life cycle of S. albicans and has shown a number of characteristics which seem to have their parallel in our strains:

(1) H om othallism (2) Spontaneous transformation, n-»2 n by nuclear fusion of a budding cell. (3) Production of haploid cells by chlam ydospores. (4) Presence of a quiescent haploid stage. (In strain 29a the very small cells do not germinate when plated. )

A still unexplained fact is that p-FPA treatment of red prototrophs gave exclusively the parental types. No auxotrophic recombinants were obtained. It may be that marker genes on H3 (11), H4 (6) and H4 (9) are located on the same chromosome. However, this seems rather unlikely. A second possibility is that these prototrophs are not true diploids, but simply heterokaryons. A firm conclusion on this point must await cytological stu d ie s.

ACKNOWLEDGEMENTS

This work has been performed with the help of the Délégation Générale a la Recherche Scientifique et Technique (Contract No. 70 02 182).

REFERENCES

[1] ADONDI, G., HESLOT, H., Mut.Res. 9 (1970) 41. [2] BURTON, K., Biochem.J. 62 (1956) 315. [3] McCULLY, K .S ., FORBES, E ., G enet.R es. 6 (1965) 352. [4] ELKIND, M .M ., BEAM, C .A ., R ad iat. Res. 3 ( 1965) 88. [5] EMEIS, C.C ., Z.Naturf. 218 (1966) 816. [6] GUTZ, H., J.Bact. 92 (1966) 1567. [7] JEENER, R ., BRÄCHET, J ., E nzym ologia И ( 1943) 222. [8] LEUPOLD, U., Schweiz. Z.allg. Path. Bakt. 18 (1955) 1141. [9] LHOAS, P., Nature 190 (1961) 744. [10] MORI, H., ONISHI, H., Appl. Microbiol. 15 (1967) 928. [11] MOUSTACCHI, E., Thesis, Paris (1964). [12] SCHMIDT, G., SERAYDARIAN, K., GREENBAUM, L .M-, LISS, M., HICKEY, M.D., THANNHAUSER, S.I.. Fed.Proc. 14 (1955) 277. [13] Van der WALT, J.P., Mycopath. Mycol.appl. 40 (1970) 231. [14] WICKERHAM, L.J., KURZMAN, C .P., HERMAN, A .I., Science 167 (1970) 1141. IAEA-SM-134/5 109

DISCUSSION

R. HUTTER: Do you have any information on the doubling times of your so-called haploids, and what is your estimate of the DNA replication time in your strain? C. GAILLARDIN: The growth of the haploid strains HI to H6 has been studied either with a complete or a minimal medium. Under both conditions they had the same doubling time as the original strain 29a. Since we didn't perform any experiment on synchronous cultures of our strains, we were unable to measure the DNA. content of the cells during the budding cycle, and we don't know anything about the DNA replication time. P. DUPUY: Were you not impeded in your experiments by the presence of pseudomycelium, which produces cell lumps and probably multinucleate cells? C. GAILLARDIN: It is true that Candida tropicalis, in the stationary phase, forms hyphae, multinucleate cells and cell lumps, all of which could account for the higher radioresistance of a part of the cell population. But our results on the radiosensitivity of cells obtained in the stationary phase remained unchanged when we filtered the irradiated population on a glass filter which let through only the single cells. P. DUPUY: The radioresistance tails can also be explained by the presence of budding cells. They presumably contain radio-protective substances, the effect of which can be eliminated by the use of a radio­ sensitizer, for example, iodoacetamide. C. GAILLARDIN: This is the hypothesis commonly used to explain the very similar results obtained in strains of Saccharomyces cerevisiae under these conditions. In fact, we observed that the population fraction which appears to be radioresistant in our experiments was quantitatively comparable with the budding cell fraction. Both these fractions remained generally constant (about 17-18% of the total population) through all the treatments studied here. Moreover, in experiments with 5-fluorouracil where the radioresistant fraction was found to decrease, we observed a correlative equivalent decrease in the number of budding cells. Neverthe­ less, we haven't tried radiosensitizing substances such as iodacetamide, nor have we attempted to work on purified budding-cell preparations. So, although it is quite likely that the radioresistance is due to the budding cells, we have no firm evidence to support this statement. HUGUETTE de ROBICHON-SZULMAJSTER: In regard to the radio­ resistance shown by some of your strains, I noted in Fig. 3 that the behaviour of the wild strain 29a differs markedly, depending on the culture medium used. It would appear that there is no radioresistant population when the culture is grown on complete, i. e. amino-acid-rich medium, whereas radioresistance does appear in the cultures grown on minimal medium. Since it is probable that in a complex medium most biosynthesis pathways are repressed or inhibited and since, moreover, you have shown that nitrogen starvation, and not phosphate or uracil starvation, is responsi­ ble for the disappearance of radioresistance in cultivated strains in a minimal medium, I wonder whether the hypothetical radiation-protection substance is not derived from an intermediate metabolite formed during the biosynthesis of one or more amino acids. Considering further that you were able to make this radioresistance disappear in a double mutant 110 GAILLARDIN and HESLOT through combined aspartate and lysine starvation, I should like to know whether you have tried the effect of simple starvation of any one of these amino acids, and possibly by using other strains containing auxotrophs in the case of other amino acids. C. GAILLARDIN: We initially performed some experiments with other mutants requiring amino acid, for instance meth~ mutant, which gave us the same results as asp~ lys~, but we didn't mention it, because we regarded the results as less significant — a single character would reverse and thus would have given us an ambiguous result. It doesn't seem, therefore, that the disappearance of radioresistance in the asp~ lys~ mutant is due to the particular type of mutations. We haven't in­ vestigated what happens in a medium supplemented with either asp or lys. B. H. NGA.: Could the inability to induce haploidization in some diploids in Candida be due to the presence of balanced lethals in these diploids? C. GAILLARDIN: The hypothesis of a balanced lethality is very difficult to prove or disprove with the few results we have. One can admit, in this hypothesis, that p-fluorophenylalanine (p-FPA) would induce haploidization for all markers except for those which are located on the same chromosomes as the lethal balance. During successive divisions the diploid state would be restored. But we don't understand why, in the case of all the red and other white recombinants we studied, this restoration process was quicker than for the wild 29a, which has the same genetic background. As we can cross two "diploid" strains, we may suppose that spontaneous haploidization can occur, since we never obtained tetraploids or cells of larger size which could account for tetraploidy. This spontaneous haploidization should also occur in our red strains. Since we haven't been able to isolate them as auxotrophic progeny, there is good evidence . to prove that they can't germinate, as we have shown in the case of the little cells given by 29a. In a lethal balance hypothesis, these true haploids shouldbelethal. On the other hand, there are good reasons for believing that a meiosis takes place in the chlamydospores produced by our red strains. We have been able to isolate the products of these meioses; they give rise to diploid cultures (by self-diploidization), which are of parental types, or recombinants in a few cases. This seems to rule out balanced lethality. However, we need more information on this point. S. I. ALIKHANIAN: Did I rightly understand that your original strain was a haploid and grew on petroleum hydrocarbons? If so, do the diploids retain the ability to grow on hydrocarbons? C. GAILLARDIN: As far as we know, our original 29a strain is a diploid and has a very strong tendency to diploidize, when it produces, either spontaneously or after p-FPA treatments, haploid derivatives. Since the strain is homothallic, we think that it is homozygous for all its genes in the diploid state. Thus the genes controlling growth on hydrocarbons should be present in the haploid isolates, as they were in the parental strain. We haven't investigated the ability of these haploids to grow on hydrocarbons. J. WEIJER: Did you ever consider the possibility that the difference between the haploid and diploid states in Candida is due to polyteny rather than to differences in the number of genomes present? IAEA-SM-134/5 111

С. GAILLARDIN: I have no information on the action of p-FPA on polyteny. We haven't performed any ultramicroscopio studies on our strain, nor measured the DNA content during the life cycle. But the polyteny hypothesis could perhaps account for the differences in DNA contents we observed between cells taken in the stationary or exponential phase.

IAEA-SM-134/1

VEGETATIVE INSTABILITY IN FUNGI The role of chromosome aberrations

J, A. ROPER The University, Sheffield, United Kingdom

Abstract

VEGETATIVE INSTABILITY IN FUNGI: THE ROLE OF CHROMOSOME ABERRATIONS. In the fungus Aspergillus nidulans, strains with a chromosome segment in excess of the haploid genome are unstable. During vegetative growth they produce two classes of variants. The more frequent class arises by deletions from either of the segments present in duplicate. These deletions, variable in size, occur by an intrachromosomal process. By deletions, a quantitatively haploid state is reached in one or more steps. The other class probably carries newly arisen tandem duplications on either duplicate segment. These variants are highly unstable and show successive genetic changes of various kinds. In some, genetic material probably part or all of the newly-generated duplication — is lost from its segment of origin; sometimes this is apparently transposed elsewhere in the genome. The resulting types accumulate, successively, mutations which affect both morphology and stability. Instability of this type has now been observed in an industrial strain of Pénicillium. Mitotic non-conformity, as this instability has been termed, will generally be an undesirable feature of an industrial strain; however, it is just possible that it could be used to generate new strains of merit.

Many cases of genetic instability involving chromosomal changes are known in plants, animals and microorganisms. Overall they appear, in both cause and effect, to be very diverse phenomena with no obvious, unifying feature. Most are difficult to analyse in formal genetic, let alone molecular, terms. However, the study of instability at mitosis is important for several reasons. The fidelity of chromosome replication is presumed to reside primarily in the accuracy of the semi-conservative replication of DNA. and in the processes of DNA repair. The many instances of mitotic instability suggest that there are other, as yet unidentified, components of this fidelity. Events occurring during normal development in certain species have parallels in some analysed cases of instability. Perhaps instability involves some mechanisms which, when controlled and programmed, have a role in normal development. In the context of practical microbial genetics, instability raises difficulties and, possibly, new approacaes to strain improvement. An understanding of instability may help in the handling of strain which, but for their lack of stability, would be of practical value; it may also suggest new ways of producing useful variants. Some species of fungi have characteristics which make them almost ideal tools for studies in this field. Colonial growth permits the detection of mitotic variants with a modified growth rate or morphology. In suitable cases use of conidial colour mutants further facilitaiés the detection of mosaic colonies and variant sectors. Isolated vegetative variants can be subjected to forms of genetic analysis which would be impossible for the equivalent somatic tissues of higher organisms. Studies of haploid and diploid strains, in species where the latter can be synthesized, are likely to provide important details of the causes and effects of instability; in

113 114 ROPER addition, the diploids may offer more valid comparison with situations found in higher organisms. This account summarizes work, in the fungus Aspergillus nidulans, on a system of instability which has parallels both in other microbial species and in higher organisms.

MITOTIC NON-CONFORMITY IN Aspergillus nidulans

The first example of this system of mitotic instability was found by Bainbridge and Roper [1], From certain crosses of morphologically- normal parents, two thirds of the progeny were normal and one third had reduced linear growth rate and a characteristic "crinkled" morphology. Ascus analysis showed that 25% of the ascospores were inviable. Examin­ ation of the parental pedigrees showed that, in each such cross, one parent was chromosomally standard while the other had an apparently non-reciprocal translocation of a segment of linkage group III to linkage group VIII [2]. The two-thirds normal progeny were of the parental chromosomal types — standard or translocated; inviable ascospores were presumed to lack the relevant segment of linkage group III; crinkled progeny seemed likely to have this linkage group III segment in duplicate (Fig. 1). In the last class, the reduced growth rate and crinkled morphology were due, presumably, to chromosome imbalance. Confirmation that this class carried a segment in duplicate was obtained through their entirely unexpected property of vegetative instability. After a few days' incubation crinkled colonies produced sectors which, in varying degree, approached wild type in growth rate and morphology. Thiosulphate-independent,

v m ______i ______

X ■ и ------, ------^

VIH ______| ______I Parental types and

Inviable deletion 1,1 ,и • re c o m b in a n t VJ)t

Duplication 111 ' %

recombinant VIII -I-

FIG. l. Results of the cross ofachromosomally-normalstrain to one with a segment of linkage group III (solid line) translocated to linkage group VIII (dotted line), s, a recessive allele determining requirement for thiosulphate. IAEA-SM-134/1 115 crinkled progeny were obtained from a cross involving the s !2 allele (thiosulphate requirement) which was known to be located on the relevant linkage group III segment (Fig. 1). Some individual colonies of this type, each grown from one uninucleate conidium, produced both thiosulphate- independent and thiosulphate-requiring sectors. This showed that these crinkled progeny were heterozygous for s !2 and it suggested also the way in which variant sectors arose. It seemed likely that they resulted from deletions, perhaps variable in size, from one or other duplicate segment. Outcrossing of the variant sectors, to strains which were chromosomally standard and to others with the III - VIII translocation, confirmed the deletion hypothesis. Sectors arose by loss of apart — variable in size and position - of either duplicate segment. When less than the whole segment was deleted, the resulting variant was still unstable and produced yet further sectors.

I - У ad+ bi +

y + ad b¡

FIG. 2. A strain carrying a duplicate segment of linkage group I (solid line) terminally attached to linkage group 11 (dotted line). Mutant alleles determined: yellow (as opposed to wild-type green) conidia, ad and bi, requirement, respectively, for adenine and biotin.

A different duplication strain, obtained by Pritchard [3], was valuable in elucidating further aspects of this instability. The particular merit of this strain lay in the duplication of a segment of linkage group I which carries the loci of two known nutritional m arkers and the locus for yellow (as opposed to wild-type green) conidia (Fig. 2). Results obtained with this strain may be summarized as follows [4]. The strain had reduced growth rate, and, like the earlier duplication strain, crinkled morphology. It too was vegetatively unstable; in fact, instability' has been found in all of the five different duplication strains tested and seems very likely to be a general property of such strains. Deletions occurred from either the translocated or untranslocated segment. Deletions which included the dominant £+ allele gave sectors with yellow conidia; others gave green sectors with improved growth rate and, when the ad+ and bi+ alleles were deleted, requirement for adenine and biotin respectively. Without out- crossing it was possible sometimes to partly specify the size and position of the deletions (Fig. 3). A quantitatively haploid or near-haploid genome was reached either in one step or in successive steps; in a minority of cases stability was reached only after deletions, inevitably non-overlapping, from both of the duplicate segments. Loss of chromosomal material occurred by an intrachromosomal process since in no instance did a variant show altered linkage arrangement of the duplication segment markers. Variants which had lost and ad, but not M, showed that deletions were at least sometimes interstitial rather than terminal. Since no terminal marker has been identified yet on this chromosome segment, the relative proportions of interstitial and terminal deletions are unknown. 116 ROPER

FIG. 3. Sectors produced by a colony of a duplication strain heterozygous for the gene determining yellow conidia. For illustration, sector frequency has been increased by using trypan blue [15].

Study of a haploid duplication strain gives no direct evidence on the role of chromosome imbalance in instability. Itmight be supposed that deletions are frequent in a normal, haploid genome and that the resulting hypohaploid nuclei are not detected. A duplication would permit expression of part of this normal instability by preserving nuclei with a deletion from either duplicate segment. Indeed, such types would be not only viable, they would also have selective advantage because of their reduced degree of chromosome imbalance. On the other hand it might be supposed that duplications provoke instability. This latter explanation was established by a study of haploid and diploid strains, with and without translocation and duplication [ 5]. The most important comparisons were of a duplication haploid (Fig. 2), a standard diploid and a diploid with an extra linkage group I segment as in the duplication haploid. Colonies of the standard diploid, heterozygous for two conidial colour markers, gave only rare sectors which could be explained fully by the well-established processes of mitotic crossing-over and haploidization [ 6]. However, under the same standardized conditions, the unbalanced diploid produced ten times as many sectors as did the duplication haploid, although sectors of the former had less advantage than those of the latter IAEA-SM-134/1 117 over their respective parents. Whereas sectors from the duplication haploid arose by deletion from either duplicate segments, those from the unbalanced diploid arose largely or exclusively from deletions involving only the translocated segment. Almost all of these sectors were still diploid; about 1 0 % were, in fact, hypodiploid as a result of deletions which extended from the translocated segment into the distal end of linkage group II [7]. A. further study, with qualitatively and quantitatively similar results, has been made using strains carrying duplications of the previously- discussed linkage group III segment [8 ]. In these two types of unbalanced diploids the primary events of instability were confined solely or largely to the appropriate linkage group I and linkage group III segments respectively. Taken together, these results established two points. First, from the relative stabilities of the various strains, it was shown that chromosome imbalance provokes instability. Second, from the variants produced by different unbalanced diploids, it could be concluded that the initial events of instability are confined largely or solely to those segments involved in im balance. These findings showed that balance of chromosome segments is a factor in the fidelity of chromosome replication. Imbalance leads to what Nga and Roper [5] have termed "mitotic non-conformity”. The term was proposed to avoid the possibly misleading concept of mitotic error and to convey that parent and daughter nuclei may not conform in genotype. Current knowledge of DNA replication neither predicts nor offers expla­ nation for mitotic non-conformity. However, there is increasing evidence that duplications are involved in many cases of instability in microorganisms and higher organisms. This suggests interference or interaction of unbalanced segments in replication. In a tentative explanation of the findings in Aspergillus, it was proposed that each chromosome segment has its own site, perhaps on the nuclear membrane, at which replication is initiated. Competition for sites, in unbalanced strains, might lead to replication errors in the competing segments. An explanation based on limiting and segment-specific initiation substances would serve equally well. The work of Rosenberger and Kessel [9] may be relevant. From a study of chromatid segregation at mitosis in A. nidulans, they proposed segregation units which carry DNA strands of the same age. Further, to explain the distribution of nuclei of the same age within a hypha, they proposed membrane sites to which segregation units are attached; the attachment sites are presumed to initiate replication and to be limited in num ber.

FIG. 4. Unequal sister chromatid exchange resulting from breaks (arrows) at different points. Crossing-over within an intrachromosomal loop yields similar results and could give, in addition, circular fragments. 118 ROPER

In formal genetic terms the origin of deletions can be explained by unequal sister chromatid exchange or by crossing-over within an intra- chromosomal loop [4]. Either process might be expected to yield new tandem duplications as well as deletions (Fig. 4). In addition to the relatively frequent sectors with improved growth rate and morphology, duplication strains have given, less frequently but regularly, types with deteriorated morphology. Most of these, at least when freshly isolated, were far more unstable than their duplication parent [4, 10]. Analysis showed that they rarely, if ever, arose by deletions; the evidence suggested a different class of genetic change with properties consistent with those of a newly-arisen tandem duplication. In a number of the most unstable variants the locus of morphological deterioration and increased instability was located, as an apparently single genetic change in each case, in one or other duplicate segment. A bewildering range of genetic changes has been observed in these variants. In some there was strong evidence that part or all of the proposed new duplication may transpose elsewhere to the non-duplicated parts of the genome. This generally reduced the degree of instability of the strain; but the chromosomal changes, whatever their nature, continued to provoke further mutations as long as both they and the original duplication existed together in. the same strain. These further mutations, arising in àll linkage groups, affected morphology and stability [10]. Some lineages, analysed to fourth-order variants, had accumulated many mutations and were still unstable. Deteriorated variants showed yet other genetic changes. In diploids formed between a standard haploid and certain deteriorated strains, recessive lethal mutations were generated with a frequency greatly exceeding the spontaneous frequency in normal diploids [11]. There was some degree of specificity in the production of lethal mutations; independent isolates of certain diploids frequently, or in some cases invariably, showed a recessive lethal, and the locations were specific for each particular diploid. Aneuploids, especially those disomic for linkage group III, were observed at high frequency in the analysis of some variants; they were found among colonies from platings of conidia of certain deteriorated strains and in the progeny from crosses of some deteriorated by normal. It is at present impossible to offer a plausible explanation for the various genetic changes which deteriorated variants undergo. This must await, at the least, further formal genetic analysis. What is of interest is that many elements of mitotic non-conformity parallel phenomena which occur in cases of instability in microorganisms and higher organisms. Further study of mitotic non-conformity may show an underlying unity, of cause and effect, in some instances of instability for which diverse explanations are now offered. It should also reveal further features of the control of mitotic fidelity.

PRACTICAL SIGNIFICANCE OF MITOTIC NON-CONFORMITY

Strains with duplicate genetic material may arise from crosses in­ volving certain chromosome aberrations. Highly developed lines of, for example, Pénicillium chrysogenum probably carry many aberrations as a result of repeated mutagenic treatments. Genetic analysis in this species strongly implicates translocations [12]. In particular, strains IAEA-SM-134/1 119 which differ from each other in a non-reciprocal translocation are likely to yield, on crossing, a proportion of unstable, duplication progeny. In a pilot survey with A. nidulans, 40% of the gamma radiation-induced translocations were judged to be non-reciprocal in that, on crossing to a standard haploid, they gave progeny of which one third were unstable [13]. In addition, duplication strains occur spontaneously and may be prefer­ entially selected when particular phenotypes are sought [3]. The extent to which mitotic non-conformity may complicate a program of strain improvement cannot yet be assessed but there are indications that the phenomenon is of more than academic interest. Already a high titre strain of P. chrysogenum has been shown to have instability of a type best interpreted as mitotic non-conformity [14]. As a mechanism for generating diversity mitotic non-conformity has no established merit. The deletions occurring in duplication strains reveal only the genetic variation which had been masked in the duplicate segments. In deteriorated variants new mutations arise spontaneously but there is no a priori reason to suppose that these might have practical value. The acute problem is likely to concern strains which, but for their instability, would be of practical use. If these do carry a chromosome segment in duplicate it might be possible to stabilize them by using, for example, balanced lethal mutations [10]. However, there could be a more positive aspect of duplication strains. Future programs of strain improvement might indicate the desirability of duplicating, in an otherwise haploid genome, segments of chromosome carrying particular structural or regulatory genes. Even in an imperfect fungal species it would be technically feasible to synthesize such strains and to ensure an adequate degree of stability. The program would not be easy but, with an improved under­ standing of the biochemical and genetic control of the relevant biosynthesis, it might be seen as one rational and valuable approach to strain improvement.

REFERENCES

[1 ] BA INBRIDGE, B .W ., ROPER, J. A ., J. gen, M icrobiol. 42 (1966) 417. [2 ] KÄFER, E ., G en etica 33 (1962) 59, [3] PRITCHARD, R.H., Symp. Soc. gen. Microbiol. 10 (1960) 155. [4 ] NGA, B. H ., ROPER, J. A ., G enetics 58 (1968) 193. [5 ] NGA, B .H ., ROPER, J. A ., G enet. Res. 14 (1969) 63. [6 ] PONTECORVO, G ., ROPER, J. A ., A dvanc. G enet. 5 (1953) 218. [7] ROPER, J.A ., NGA, B.H., Genet. Res. 14 (1969) 127. [8 ] ROPER, J .A ., WATMOUGH, W ., unpublished. [9 ] ROSENBERGER, R. F . , KESSEL, M ., J. Bact. 96 (1968) 1208. [1 0 ] AZEVEDO, J .L ., ROPER, J .A ., G enet. Res. 16 (1970) 79. [1 1 ] AZEVEDO, J .L ., ROPER, J. A ., J. gen. M icrobiol. 49 (1967) 149. [12] MACDONALD, K.D., HUTCHINSON, J.M ., GILLETT, W .A., J. gen. Microbiol. 33 (1963) 385. [1 3 ] ORTORI, S. G ., ROPER, J. A ., unpublished. [14] BALL, C ., International Symp. on Genetics of Industrial Microorganisms, Prague, 1970 (in press). [1 5 ] COOKE, P ., ROPER, J, A ., WATMOUGH, W ., N ature 226 (1970) 276.

DISCUSSION

N. K. NOTA.NI: How do you visualize the m echanism of transposition? Is it analogous to the one involved in the (chromosomal) translocations? 120 ROPER

J.A.. ROPER: No. The initial translocations were all induced by X- or gamma-irradiation. We can only speculate about the possible mechanism of the proposed transpositions. Nga and I suggested that deletions and tandem duplications could arise by unequal sister chromatid exchange, or perhaps by crossing over within an intrachromosomal loop. The latter process could give, apart from deletions and tandem duplications, circular fragments. Perhaps intrachromosomal loops are more likely in a segment with a tandem duplication. One might then think of the transposition as something similar to release, and integration elsewhere in the genome, of an episome. However, I must stress that this is highly speculative. HUGUETTE de ROBICHON-SZULMAJSTER: The model you have proposed seems to imply a single point of attachment for each chromosome at the nuclear membrane during replication. However, Halvorson et al. , studying replication timing — measured by sequential occurrence of gene products — for genes located on the same chromosome with the use of synchronous cultures of Saccharomyces cerevisiae have shown, at least in the case of chromosome V, that there does not seem to be an ordered replication beginning either at the ends or at the centromere. The results obtained by these authors indicate the existence of many initiation points on each chromosome, which do not necessarily lead to replication of each initiated segment in the same direction. To maintain your model, would it be necessary, by analogy with these results and so far as they relate to only one chromosome in S. cerevisiae, to postulate the existence, for each chromosome, of multiple points of attachment, either to the nuclear membrane or to any other kind of support? J. A. ROPER: Yes. I would certainly think of more than one initiation site per chromosome. In fact, I would suppose that even the segment carried in duplicate has more than one such site. This might help in understanding how only a small part of a duplicate segment may be lost by deletion at any one mitosis. M. CAURIE: Can the instability noticed in some of your cultures be due to infection by the extracytoplasmic particles which Dr. Esser mentioned (SM-134/24)? J. A. ROPER: No. All our findings point to a chromosomal basis only for the instability. J. WEIJER: Is there, in your opinion, a link between the instability of duplications and the breakdown (in a stepwise manner) of the diploid nucleus during haploidization? J. A.. ROPER: Yes, there is a link and I have two reasons for thinking so. Para-fluorophenylalanine inhibits diploid strains, and also strains with duplications, perhaps by the same mechanism. The other reason is that diploid strains with a duplication are far more unstable than haploids with a duplication. This was a surprising finding, since the former are less unbalanced than the latter. We can explain this tentatively by supposing that the diploid, an "unnatural" state for this fungus, is already in difficulty in reproducing its genome. S. I. ALIKHANIA.N: Are you sure that the translocation was non- reciprocal? J. A. ROPER: Although geneticists generally frown on the possibility of non-reciprocal translocations, we are quite sure that at least four of the five translocated strains studied are effectively non-reciprocal. Of IAEA-SM-134/1 121 course, we can never exclude the possibility that a small, non-essential fragment has been reciprocally translocated, since the cytology of Aspergillus is not refined enough to enable us to verify this possibility. However, the genetic evidence is consistent with non-reciprocity.

General comments by the Session Chairman

S. I. A.LIKHANIAN: Today we have heard p resentations of very interesting papers. I should like to mention, in particular, the papers of K. Es ser, J.A.. Roper and H. Heslot and C. Gaillardin. Their works are well known, but what they spoke of has not only abstract, theoretical significance but, strangely enough, great practical value. Esser spoke of the ageing of fungus strains and of problems arising out of this phenomenon — those of "keeping" the fungus strains young. This matter is of great importance in industrial microbiology, since we often encounter cases where, for lack of optimum storage conditions, the industrial properties of strains are found to be lost. Of no less importance is the very confusing case of instability in Aspergillus, which is associated with chromosomal aberration. The instability mechanism described by Roper may also be relevant for the useful (industrial) properties of fungi, and for this reason it is very helpful to know the mechanisms responsible for instability. Thus,selection of microorganisms is being increasingly reinforced with theoretical principles. A work of great interest was reported by Heslot and Gaillardin. The attempt to develop hybridization methods for Candida growing on hydrocarbons is very pertinent, since an ever-increasing number of countries are organizing production of cellular protein with the use of yeast growing on hydrocarbons. Unfortunately, all Candida strains used so far in protein production are of the wild type. Development of the genetic system of Candida will be of great value in strain improvement. I would like to mention that work in this direction is being carried out at our institute (Institute of Genetics of Industrial Microorganisms, Moscow) and at Leningrad University (Inge-Vechtomov, Simarov, Soom, et al. ).

IAEA-SM-134/2

INSTABILITY AT MITOSIS IN Aspergillus nidulans

B.H . NGA University of Singapore, Singapore

Abstract

INSTABILITY AT MITOSIS IN Aspergillus nidulans. Strain instability in Pénicillium chrysogenum has recently been reported and some methods of overcoming this have been suggested. Results from analyses and studies of mitotic instability in Aspergillus nidulans may help in the understanding of the processes of instability in Pénicillium. In Aspergillus nidulans, haploid strains with a non-reciprocal translocation can be obtained by irradiation and strains with duplication of the segment translocated can be obtained from crosses of such strains with a normal haploid strain free of translocations. Colonies of duplication strains show a distinct crinkled morphology and have a reduced linear growth rate. Crinkled colonies are unstable in vegetative culture and produce sectors of two broad classes. The first class of variants, more frequently observed, were of improved phenotype, the other had a deteriorated phenotype. Genetic analysis of these variants — meiotic and mitotic — showed that im­ proved variants arose by loss of part of all or all of one of the duplicate segments. It has been shown that some of the deteriorated variants arose from the duplication strain as a result of single gene mutations in the genome. Two possible mechanisms of this instability are discussed in the light of these analyses. It might be supposed that there are fairly frequent losses of chromosome segments throughout the genome of a normal haploid. The resulting hypohaploid nuclei would be inviable. In duplication strains, however, variant nuclei with loss from either duplicate segment would remain viable. The duplication would serve an entirely passive role. Alternatively, it might be supposed that the duplication provokes instability. These alternatives have been investigated by a detailed study of mitotic instability in: (1) normal haploids, (2) haploids with translocation, (3) haploids with duplication, (4) normal diploids, (5) diploids heterozygous for a translocation, and (6) diploids with a segment in triplicate. The results of these analyses are presented and discussed. Some methods of obtaining relatively stable strains from unstable duplication strains are also discussed.

INTRODUCTION

Recent studies have shown that strains of the filamentous fungus, Aspergillus nidulans, with a duplicate chromosome segment are unstable at mitosis [1,2]. Strains carrying a segment of chromosome III in duplicate were derived from crosses of a normal untranslocated haploid strain to one with a non-reciprocal III - VIII translocation [1]. Two other duplication strains have been examined and are both mitotically unstable. Duplication strains, which have a characteristic 'crinkled 1 morphology and a reduced linear growth rate, sectored to give variants of two broad classes. The majority of the variants showed phenotypic improvement, the others had deteriorated phenotype. Improved variants arose when the nuclei lost a variable part of one or the other of the duplicate segments [2]. Some of the improved variants gave further improved sectors at mitosis and these were approaching the standard haploids in morphology and stability. Until recently little was known about the deteriorated variants. More than twenty different deteriorated variants have now been analysed. From their results Azevedo and Roper [3] proposed that the deteriorated variants arose by tandem duplications of one of the segments carried in duplicate;

123 124 NGA this supports the suggestion that phenotypic deterioration and enhanced instability resulted from tandem duplications on one or other duplicate segment [2]. They further proposed that transposition of a variable part of the tandem duplication to non-duplicated parts of the genome gave greater stability. This paper is to outline the mitotic instability in Aspergillus nidulans and to discuss some of the ways these studies may contribute to a better understanding of strain instability in some industrial microorganisms.

DUPLICATION STRAINS

A strain with a duplication for a segment of chromosome III was syn­ thesized and described by Bainbridge and Roper [1 ]. A second duplication strain was derived from a duplication strain described by Pritchard [4]: it had a duplication for a segment of chromosome I [2 ]. This strain was shown by Nga and Roper [2] to have a non-inverted duplicate segment of a part of the right arm of chromosome I, including its right end, terminally located on chromosome II. Thus it had this segment in excess of the haploid com­ plement (see Fig. 1, strain 1). Another duplication strain had a duplication for a different segment of chromosome I. The last mentioned was one of a class of 'crinkled' segregants from a cross between a strain with a non­ reciprocal I - VIII translocation and a normal haploid strain. This class of strains was reduced in linear growth rate and had slightly crinkled morphology. As is the case with the other two duplication strains, these strains showed the characteristic inhibited growth in media supple­ mented with p-fluorophenylalanine. To test their stability in vegetative culture, conidia of each of four isolates were inoculated at a single point in the centre of twenty Petri dishes containing complete media (CM). The dishes were incubated at 37°C for seven days and nineteen faster-growing sectors were observed to grow out from the parent colonies. These stable segregants approached the standard haploid in growth rate and morphology and were not inhibited by p-fluorophenylalanine. An explanation of these results is that the sectors arose by loss from the duplication parents of a variable part of one of the duplicate segments. A more complete analysis of mitotic instability is given in the ensuing sections.

pro 1 poba 6 У + +

У+ ad 2 0 bi I

pro 1 poba 6 У+ ad 2 0 bi 1

У +

FIG. 1. Duplication strains. Chromosome I (unbroken line) and II (broken line) of strains used in this analysis. Centromeres are designated by open circles. Mutant alleles determined the phenotypes: y_, yellow conidia (as opposed to wild-type green) ; ad 20, bi 1, paba 6 and pro 1, requirement, respectively, for adenine, biotin, p-aminobenzoic acid and proline. IAEA-SM-134Æ 125

bi i

pro 1 pobo 6

Y+ od 2 0 bil

pro 1 pobo 6 У + +

У+ od 2 0 bi 1

FIG. 2. Chromosomes I(unbroken line) and II (broken line) of the haploid strains tested. MSE markers: su 1 - ad 20. suppressor of ad 20. _y and ad 20 (Chromosome I); w 3 (II) (e p ista tic to j and white conidia; gal 1 (III) and fac A (V), inability to use, respectively, galactose and acetate; pyro 4 (IV) ; S 3 (VI), nie 8 (VII) and ribo 2 (VIII) requirement, respectively, for pyridoxin, thiosulphate, nicotinic acid and riboflavine.

INSTABILITY OF DUPLICATION STRAINS

Extensive studies of instability in duplication strains 1 and 2 (Fig. 1) have been made [2, 3, 5-8]. Conidia of 1 and 2 were inoculated on the centre of Petri dishes containing complete media. After seven days' incubation at 37°C most of the colonies showed vegetative instability giving sectors most of which had improved morphology and a few deteriorated. Strains with various combinations of chromosomes, III to VIII, carrying this duplication were similarly unstable. Furthermore, the nature of the genetic markers on the duplicate segments did not affect the instability [5] .

Studies of some of the phenotypic classes arising at mitosis

Recent work has shown that improved variants arose by loss of a variable part of one or other duplicate segment [2]. The loss might be interstitial. A number of the original deteriorated variants derived from strains 1 and 2 were analysed [3]. These strains had a linear growth rate usually about that of the parent duplication strains. In order to locate the deter­ minants of deterioration, diploids were synthesized between master strain E (MSE) (see Fig. 2), which carries markers on all eight chromosomes, and each variant. Subsequent mitotic haploidisation of each diploid and the characterization of the haploid segregants enabled the location of the deter­ minants of morphological deterioration to their chromosomes. The deterio­ rated variants were of the following types [3].

1. Those with deletion from chromosome I 2. Those with mutations in chromosome I or II -1 complex 3. Those with mutations in chromosomes other than II or I 126 NGA

On the basis of the results of the studies of some deteriorated variants and their mitotic segregants Azevedo and Roper [3] proposed that deteriora­ tion resulted from tandem duplication in either duplicate segment. The transposition of a variable part of the tandem to non-duplicated regions of the genome gave greater stability.

THE ROLE OF THE IMBALANCE OF CHROMOSOMAL MATERIAL IN INSTABILITY

To examine the role of the imbalance of chromosomal material, relative instabilities of haploid and diploid strains with or without translocations and duplications were studied [5]. Six classes of strains (Fig. 2) used in this analysis were; standard haploid (A); translocated haploid (В); duplication strain (C); D, E and F were diploids obtained by combining A, В and С respectively with MSE. С and F were single representatives of a class whose members differed in residual genotype. Conidia of each strain were plated on CM at a density of about five colonies per dish. A count of sectors which were non-parental in colour and which differed from the parents in morphology was made when the colonies were observed to have approxi­ mately identical mean diameter. This provided a measure of mitotic in­ stability. No sectors were obtained for strains A and B; D and E gave the expected low frequency of mitotic segregants from mitotic crossing over and haploidisation; С gave about 45 sectors per 100 colonies and F about 501 sectors per 100 colonies. The variant sectors from F were diploid and most of them had lost part or all of the translocated duplicate segment. These results suggested that imbalance of chromosome segments provokes mitotic instability. It appeared that instability was restricted to the segments in duplicate. In a recent paper, however, strains with deletions which pre­ sumably included part of the chromosome II of the II-1 complex had also been obtained [6 ].

SOME FACTORS AFFECTING MITOTIC INSTABILITY

Temperature of incubation affect the frequency of sectoring. Results of Lieber and Roper (unpublished) showed that mitotic instability had a negative temperature coefficient. The influence of chemical agents on deletion in duplication strains has been reported [7,8]. Formaldehyde at a final concentration of 1/200 000 (w/v) seemed to increase the frequency of improved sectors. Low-density platings of conidia of duplication strains 1 and 2 on this media gave 10 0 % viability. Conidia of each of the duplication strains 1 and 2 (Fig. 1) were inoculated at the centre of Petri dishes containing CM, with or without formaldehyde, and incubated for seven days. Counts of sectors were then taken (Table I). A slight reduction in linear growth rate was observed for the treated colonies. In both the treated and the control colonies sectors began to be observed after about five days of incubation. Improved sectors that arose by loss of part or all of either of the duplicate segments were regularly obtained. These sectors showed a distinct advantage in growth rate on both media when compared with the duplication parents. There was no obvious inhibiting IAEA-SM-134/2 127

TABLE I. MEAN NUMBER OF SECTORS PRODUCED BY CONTROL AND FORMALDEHYDE-TREATED COLONIES OF STRAIN I a

No. of dishes Yellow Green

Formaldehyde treatment 35 6 .5 0 .6

C ontrol 35 0 .4 8 0 .3

a Formaldehyde-treated colonies of strain 2 also had an increased frequency of sectors; mean number of green sectors being 4.7 and yellow 0.45.

effect of media on the growth of translocated haploids arising by deletion from the untranslocated duplicate segment. It is, however, possible that variants with certain specific deletions might be inhibited when they occurred in CM dishes and so might not be manifest as sectors; these same classes of variants might establish as sectors in treated dishes. Experiments are in progress to determine the genotype of some of the variants from the treated and control colonies. The enhanced frequency of yellow sectors, in the case of 1, and green, in the case of 2 , in the treated colonies could be attributed to preferential deletion from the translocated duplicate segment as has been suggested by Cooke, Roper and Watmough[7] for the induction of deletions by trypan blue.

MITOTIC INSTABILITY AND INDUSTRIAL MICROORGANISMS

Mitotic instability has been known to occur in certain strains of several industrial microorganisms. As the productivity of the strains might be affected by instability it is of importance to understand the causes of mitotic instability and the means by which the instability may be overcome. A case of chromosomal instability in Pénicillium chrysogenum has been reported [9]. In the case of mitotic instability in a duplication strain in Aspergillus nidulans we obtained stable haploid strains by chromosome substitution [2 ]. For this, diploids synthesized between the duplication strain and a haploid master strain were haploidised and from the haploid segregante strains carrying the desired combinations of chromosomes I, III to VIII and chromo­ some II of the master strain could be selected. These segregante would not have the duplication and would be stable. This provided a method for obtaining stable haploids from unstable duplication strains in microorganisms in which mitotic haploidisation of diploid was convenient. Loss of part of one or other duplicate segment from duplication strains gave strains of which some are vegetatively stable [2 ]. If high productivity is related to the particular duplication, or a part of it, stable strains retaining the duplication would then be sought for. Relatively stable balanced lethal strains in which part of the duplication is retained can be obtained from the duplication parents [2]. Azevedo [10] has suggested the use of balanced lethal systems to prevent mitotic instability in duplication strains. 128 NGA

It is also known that changes in physical or chemical conditions of culture and in the genotype of the duplication strains may affect their instability. Recent work of Azevedo and Roper [3] has shown that some of the deteriorated variants from the duplication parent which had single new mutations were almost stable at mitosis although they still carried the duplication.

REFERENCES

[1] BAINBRIDGE, B.W ., ROPER, J.A ., Observations on the effects of a chromosome duplication in Aspergillus nidulans, J.gen.M icrobiol. 42 (1966) 417. [2] NGA, B.H., ROPER, J.A ., Quantitative intrachromosomal changes arising at mitosis in Aspergillus nidulans, Genetics_58 (1968) 193. [3] AZEVEDO, J.L., ROPER, J.A ., Mitotic non-conformity in Aspergillus: Successive and transposable genetic changes, Genet.Res. 16 (1970) 79. [4] PRITCHARD, R.H., A genetic investigation of some adenine-requiring mutants of Aspergillus nidulans, Ph.D. Thesis, University of Glasgow (1956). [5] NGA, B.H., ROPER, J.A ., A system generating spontaneous intrachromosomal changes at mitosis in Aspergillus nidulans, Genet.Res. 14 (1969) 63. [6] ROPER, J.A ., NGA, B.H., Mitotic non-conformity in Aspergillus nidulans: The production of hypodiploid and hypohaploxd nuclei, Genet.Res. 14 (1969) 127- [7] COOKE, P., ROPER, J.A ., WATMOUGH, W., Trypan-blue-induced deletions in duplication strains in Aspergillus nidulans, Nature, Lond. 226 (1970) 276. [8] PALMER, H.M ., ROPER J.A ., Induced deletions in duplication strains of Aspergillus nidulans, Aspergillus News L etter 11_( 1970) 20. [9] BALL, C ., The genetics of Pénicillium chrysogenum, Aspergillus News Letter 10 (1969) 14. [101 AZ.EVEDO, J.L., Recessive lethals induced by nitrous acid in Aspergillus nidulans, Mutat.Res. 10 (1970) 111. LA EA-SM-134/22

MICROBIAL GENETICS AND THE CONTROL OF THE PATHOGENS IN AGRICULTURAL INDUSTRIES

S .G . GEORGOPOULOS N.R.C. "Democritus”, Athens, Greece

Abstract

MICROBIAL GENETICS AND THE CONTROL OF THE PATHOGENS IN AGRICULTURAL INDUSTRIES. In the agricultural industries pathogenic microorganisms are controlled with the use of resistant genotypes of the hosts and the use of chemicals that are toxic to the pathogens. The use of resistant varieties has faced very serious problems because of great genetic variability with respect to virulence in the pathogens. In this case, the understanding of microbial genetics has greatly contributed to the solution of practical problems. In the use of toxicants, on the other hand, the appearance of new pathogen genotypes with genes overcoming toxicity has not been a matter of much practical concern so far. This is explained to be due to the nature of the toxicants rather than to reduced mutability of the pathogens. A few examples are mentioned indicating that, with the use of more selective toxicants in practice, resistance mutations will give rise to more and more problems and that knowledge of genetic mechanisms is a prerequisite to the solution of such problems.

INTRODUCTION

Pathogenic microorganisms, mostly fungi, are harmful to the agricul­ tural industries by causing diseases of plants and plant parts in the field and spoilage of the agricultural produce in storage and transport. For the control of these pathogens man has mainly employed:

a. Genes which provide the host with defense mechanisms preventing the establishment and/or development of the disease and b. Chemicals (fungicides) which are added to protect the plant or the product by killing the pathogen or inhibiting its growth.

This paper examines the genetic mechanisms in microorganisms with respect to the value and prospect of our two main methods of controlling pathogens in the agricultural industries. In comparing the two methods from this viewpoint, experience so far seems to indicate that pathogens have been very successful in overcoming defensive mechanisms of the host, through some genetic change, but not in overcoming the toxicity of the fungicides. I think we know why this has been so and we now have considerable reason todoubt whether it will remain so, as is explained in this paper.

MICROBIAL GENETICS AND PATHOGENICITY

Under the impression of the damage that is caused by pathogenic fungi we may tend to forget that in nature resistance is the rule and suscepti­ bility is the exception. In other words, most fungi do not have the right genetic constitution in order to be pathogenic to most plants and when there is pathogenicity we have a high mutual host/pathogen specificity. In the majority

129 130 GEORGOPOULOS of cases susceptibility of a plant for a fungus appears to depend on a specific biochemical inter-relationship between the two organisms. In a number of instances the "gene for gene" hypothesis, originally proposed by Flor [1], has been shown to provide the genetic basis for this inter-relationship. According to F lo r's hypothesis for each genetic locus in the host, governing resistance and susceptibility, there is a specific and related locus in the parasite that governs its virulence and avirulence. We have two types of genetic control of resistance to the pathogenic fungi. In the so-called "horizontal" resistance we have a system of many genes, each of negligible individual effect. To overcome this resistance would require many mutations in the pathogen. Horizontal resistance is, therefore, equally effective against all of the pathogen genotypes and it is persistent. The multigenic system, however, never gives the level of resistance that would be satisfactory from the practical point of view. For this reason breeders have turned to wild species from which "major" resistant genes (R-genes) are introduced into the cultivated varieties. This type of "vertical" resistance is very effective but it only takes one mutation at the corresponding locus (V-gene) in the pathogen to overcome the effect of the R-gene. For this reason, R-gene resistant varieties often have to be withdrawn after a few years because the genetic mechanisms of the pathogen have created the need for a new R-gene to be found and introduced into the host genotype. The work done for practical purposes, i.e. the breeding of resistant varieties, has greatly assisted the understanding of the genetics of the pathogens. In cases where genetic analysis in the pathogen has met with technical difficulties, putative pathogen genotypes can be constructed on the basis of genetic analysis of crosses between differentially sensitive varieties of the host. As an example I can mention the phycomycete Phytophthora infestans causing blight of the potato plant and storage rot of the tubers. Little is known on the genetics of this fungus but the pathogenic genotypes (races) are named on the basis of the corresponding R-genes for resistance in the host [2 ]. In this case and in all the cases where we have the gene- for-gene system, every time a new R-gene is introduced the number of possible pathogen genotypes is doubled and for n resistance genes in the host we can predict 2 n pathogenic races of the fungus [3]. In an analogous manner, the methodology of fungal genetics has helped to identify the genes for resistance in the host and in this way to solve practical problems with appropriate breeding work. In the case of apple scab, for example, which is caused by the ascomycete Venturia inaequalis, it is easier to do crosses and genetic analyses in the pathogen than in the host. Thus, Boone and Keit in 1957 [4] recognized seven loci for pathogeni­ city to apple in V. inaequalis. On the basis of this information, putative host genotypes were constructed and could be used for breeding new resistant v a rie tie s .

MICROBIAL GENETICS AND FUNGICIDE RESISTANCE

In the use of fungicides the genetic mechanisms of the pathogen have not created many practical problems so far. In other words, we very seldom had to discontinue the use of a fungicide because of development of fungicide- tolerant genotypes of the pathogen as we had to abandon disease-resistant IAEA-SM-134/22 131 varieties of the host because of the appearance of new pathogenic races. Thus, in the absence of the economic stimulus, studies on genetic mecha­ nisms of response to fungicides have been very limited as compared to the several decades of work on genetics of pathogenicity. The rarity of fungicide-resistance problems so far would be accounted for if fungicide-resistant strains showed reduced ability to compete in nature because of reduced pathogenicity [5]. However, the few studies conducted in the last few years indicate that it is the nature of the fungicides that we have used extensively in the past which has not permitted develop­ ment of resistant strains. In other words, fungicides whichpermit resistance mutations in the laboratory have also encountered some related difficulty in practice. The main example is what we have named [6 ] "the aromatic hydrocarbon group" to the members of which resistance is known to develop readily through mutation in the laboratory. And from the fact that strains resistant to these particular compounds have also developed in nature and have created practical problems, we can say that the genetic mechanisms in the pathogen would have caused талу more such problems if we had used more fungicides of this nature. This far, the practical problems because of resistance to members of the "aromatic hydrocarbon group" are not many, but the use of these compounds in practice is also rather limited. Diphenyl and sodium orthophenylphenate, for example, do not belong to the most widely used fungicides at all. Their use, however, against post-harvest decay of citrus resulted in widespread occurrence of resistant strains of Pénicillium spp. [7, 8 ] . In California, at least, the use of sodium orthophenylphenate had to be discontinued [9]. Other compounds of the group, resistance to which has created practical problems, are hexachlorobenzene [10] and 2, 6 -di-chloro-4-nitroaniline [11]. Laboratory studies have shown that all mutants selected on one of the aromatic hydrocarbon fungicides are cross-resistant to all of them in the group [6 ]. Five loci for resistance have been identified [12] in Nectria haematococca (Hypomyces solani), two in Neurospora crassa [13] and two in A spergillus nidulans [14] . Outside the aromatic hydrocarbon group, dodine (n-dodecyl-guanidine acetate) is the only commercial synthetic fungicide to which fairly well defined mutational resistance has been obtained in the laborotory [15, 16]. It is not insignificant that, in practice also, apparent resistance to dodine in V. inaequalis has rendered the application of this compound ineffective over large areas in New York State [17]. To most of the other agricultural fungicides that have been used exten­ sively, resistance development is very difficult even under optimal conditions. The explanation given by most experts is the non-specific nature of these compounds. In the case of toxicants which act by selective enzyme inhibition, high-level resistance can often develop, e.g. through a mutational modifica­ tion of the sensitive enzyme [18] or a shift to an alternate pathway [19]. When, however, the toxicant acts as a general enzyme poison rather than as an inhibitor of a particular system, a similar mutation would have very little effect because the compound would be able to hit at other sensitive sites. It is accepted that most of the fungicides of today are of this general inhibitor type. The diethyldithiocarbamates, for example, are known to inhibit at least 20 different enzymes [ 20] . Obviously, to effectively overcome toxicity in this case would require several mutations in the pathogen. 132 GEORGOPOULOS

Thus, from the viewpoint of genetic changes in the pathogen that are required to overcome the control measures, the use of multisite inhibitors as fungicides is analogous to the use of multigenic resistance systems of the host (horizontal resistance). Like these systems, the multisite inhibitors are also not very effective and the good results that we have often been able to obtain with them are due to the protective use of rather high concentrations of the chemical on the outside of the plant or the product [21 ]. It is generally agreed now that the chemical control of pathogens in the agricultural indus- tires cannot afford to stick to the non-selective type of toxicant and, in fact, a few very effective systemic fungicides, such as benzimidazole, oxathiin and pyrimidine derivatives and also some antifungal antibiotics, have been intro­ duced in recent years. Such compounds can be used therapeutically as well as protectively and seem to act as specific inhibitors. The oxathiins, for example, seem to act on succinic dehydrogenase [22]. If this is so, then the use of these new fungicides will be analogous to the use of R-gene resistant varieties of the host plant (vertical resistance). The control will be highly successful originally but the pathogen is likely to become able to overcome toxicity by a simple genetic change. In fact, shortly after the introduction of the one of the highly effective benzimidazole derivatives resistance was found not only in the laboratory [23] but also in the field [24]. Resistance to the oxathiins also seems to develop readily with single gene mutations [19]. It is rather early to try to predict what the situation will be if and when specific inhibitors take some of the room now occupied by the non-selective fungicides in the agricultural industries. It is not unlikely, however, that the success of the selective fungicides will not last longer than some of the successes of breeding resistant varieties of the host. Maybe, every time a specific toxicant is introduced for large-scale fungicidal applications we should start preparing for replacing it with another specific toxicant showing no positive cross-resistance relationship with the former. We may have to use the selective fungicides not singly but in combination, so that we affect more than one cell process in the pathogen, or we may have to alternate them on the same crop. Finally, we may have to go back to the old- fashioned multiside fungicides from time to time — in much the same way as we realized the value of the multigenic resistance to blight in potato, for example, following the discovery that R-gene specific pathogenic races of the fungus appear as fast as each R-gene is introduced into the host genotype [2 ] . In any event, the search for new fungicides will continue as the breeding for resistance has continued in spite of the high pathogenic variability experienced in the fungi. In the course of these studies the understanding of genetic mechanisms in the pathogen will undoubtedly derive benefits. But it is also this understanding that will permit some safe predictions as to the future of the chemicals which will be introduced and that will contribute to the solution of some of the resistance problems which should be expected to be more frequent than they have been in the past.

REFERENCES

[1] FLOR, H .H., Phytopathology 45 (1955) 680. [2] GALLEGLY, M .E., Rev. Phytopath. 6 (1968) 375. [3] PERSON. C.. Can.J.Bot. 37 (1959) 1101. [4] BOONE, D.M ., КЕГГ. G. W., Phytophathlogy 47 (1957) 403. LAEA-SM-134/22 133

[5] DAY, P.R., A.Rev.Phytophath. 4 (1966) 245. [6] GEORGOPOULOS, S.G., ZARACOVITIS, C., A.Rev. Phytopath. 5 (1967) 109. [7] DURAN, R., NORMAN. S.M., Pl.Dis.Repti. 45 (1961) 475. [8] HARDING, P.R., Jr., Pl.Dis.Reptr. 43 (1959) 649. [9] ECKERT, J.W ., in Fungicides (TORGESON, D .C., Ed.) Academic Press 1 (1967) 287. [10] KUIPER, J., Nature 206 (1965) 1219. [11] LOCKE, S.B., Phytopathology 59(1969) 13 (Abstr.). [12] GEORGOPOULOS, S. G., PANOPOULOS, N .J., Can.J.Genet.Cytol. 8 ( 1966) 347. [13] GEORGOPOULOS, S.G., KAPPAS, A., MACRIS, B., Neurospora Newsl.No. 10 ( 1966) 8. [14] THRELFALL, R.J., J.gen. Microbiol. 52 (1968) 35. [15] KAPPAS, A., GEORGOPOULOS, S .G., Experiential (1968) 181. [16] KAPPAS, A., GEORGOPOULOS, S .G., Genetics 66 (1970) 617. [17] SZCOLNIK, M ., GILPATRICK, J.D ., Pl.Dis.Reptr. 53 (1969) 861. [18] LEWIS, D., Nature 200 (1963) 151. [19] GEORGOPOULOS, S .G ., SISLER, H .D ., J.B a c te rio l. 103 (1970) 745. [20] ULFVARSON, U., in Fungicides (TORGESON, D.C., Ed.) Academic Press 2 (1969) 303. [21] GEORGOPOULOS, S. G., BioScience 19 (1969) 971. [22] GEORGOPOULOS, S.G., VOMVOYANNI, V ., Proc.2nd Int.Congress Pesticide Chemistry, 1971 (in press). [23] HASTIE, A .C., GEORGOPOULOS, S. G., J.gen.M icrobiol, (in press). [24] SCHROEDER, W .T., PROVIDENTI, R., Pl.Dis.Reptr. 53 (1969) 271.

ROLE OF PHYSICAL AND CHEMICAL MUTAGENS IN INDUSTRIAL MICROBIOLOGY RESEARCH (Session 4)

Chairman

H. HESLOT (France)

IAEA-SM-134/14

INDUCTION OF AMYLASE-PRODUCING MUTANTS IN Aspergillus oryzae BY DIFFERENT IRRADIATIONS*

J. MEYRATH, M. BAHN, H.E. HAN** Institut für angewandte Mikrobiologie, Hochschule für Bodenkultur, Vienna r and H. ALTMANN Reactor Centre, Seibersdorf, Austria

Abstract

INDUCTION OF AMYLASE-PRODUCING MUTANTS IN Aspergillus oryzae BY DIFFERENT IRRADIATIONS. Some properties of amylases in general, and of fungal a -amylase in particular, as well as their applica­ tions have been elucidated. The different methods of industrial amylase production are compared and attempts are made to explain the different behaviour of amylase-producing Aspergillus oryzae under the various methods of cultivation. Attempts have been made to devise a simple screening test on the basis of these observations. The first screening was done by taking the ratio of the diameter of the dissolved starch zone to that of the colony as a measure for specific amylase-producing ability after irradiation with u.v. rays, gamma rays and fast neutrons. It has been confirmed that high irradiation doses are not particularly suitable for the induction of a relatively large proportion of favourable mutants. While there was a relatively large proportion of promising mutants on these screening plates, the results in actual production tests did not show as high a proportion of enhanced mutants.

Enzymes are among the most useful industrial products to be derived from the cultivation of microorganisms. Applications for various enzymes are steadily increasing and, considering how extremely difficult it is to synthesize these molecules, we will have to turn to the microbes also in the future and improve their desirable properties in order to maintain a suffi­ ciently high supply of these desirable compounds.

1. Some general properties of amylases

From the practical point of view it is useful to consider here as amylases all those enzymes which are involved in the hydrolysis or transglycosylation of starch as well as of the starch dextrins and oligosaccharides. The action of a-amylase consists mainly in breaking down starch molecules to lower molecular dextrins, whereby the a-1, 4-glucosidic linkages are hydrolysed. This action results in a rapid decrease in the viscosity of starch pastes and the rapid loss of the capacity to form a blue colour with iodine. The enzyme does not derive its name from its ability to attack a-glucosidic linkages, but from the fact that the products of hydrolysis show the optical activity corresponding to the a-sugars. Amongthe microorganisms, amylases of the ce-type are produced by many lower and higher fungi, by Actinomycetes, and by true bacteria.

* The work reported here has been carried out under the partial support of the IAEA Research Contract N0.935/RB awarded to the senior author. ** IAEA Fellowship holder under official nomination from the Korean Government.

137 138 MEYRATH et al.

The other well-known enzyme involved in starch breakdown, /3-amylase, is characterized by its ability to split maltose from the non-reducing ends of the starch molecules without dextrinisation. Since (З-amylase splits only a-l, 4-glucosidic linkages from the one end of the chain and since starch contains a considerable proportion of amylopectin of which the branched chains are linked by a ~ 1, 6 -glucosidic linkages, /З-amylase does not hydro­ lyse starch completely. Instead, the breakdown stops as soon as the branching points have been reached, and a - 1,4-linkages in between the branching points remain intact. The residual dextrin which is formed still gives a blue-violet colour with iodine. The term /З-amylase is derived from the fact that the maltose produced is in the /3-form. The result of this action is a slow decrease in viscosity, a slow change in ability to form a blue colour with iodine, but comparatively rapid increases in reducing groups and fermentable sugar (i.e. maltose). There are many claims in the literature that various microorganisms, in particular moulds, form /З-amylase. In practically all the cases this is due to a misconception of the term /З-amylase, since there was no proof of any maltose (and maltose only) being formed from starch or higher dextrins. In a considerable number of strains /3-amylase has been shown to be definitely absent. The pronounced saccharogenic activity, which is often present in many mould-culture filtrates, is now known to be due to another enzyme, glu со amyl ase. The glucoamylase discovered first in moulds is able to split both a -1,4- and a -1, 6 -glucosidic linkages and, similarly to /З-amylase, it attacks the ends of the glucosidic chains to form (3-glucose. An enzyme which solely attacks a-l, 6 -glucosidic linkages has been found in mould-culture filtrates and is called a - 1 , 6 -glucosidase (and is probably identical with oligo-1, 6 -glucosidase). These enzymes, isolated from Aspergillus oryzae and A_. niger, have been shown to attack some residual dextrins which contain a high proportion of a - 1 , 6 -glucosidic linkages. Also of some practical importance are the transglucosidases found in many moulds. These enzymes are characterized by the fact that glucose molecules are transferred from some dextrins and oligosaccharides to other dextrins and oligosaccharides with the result that new types of dextrins, different in properties from the original ones, are formed. Another enzyme only indirectly involved in starch breakdown is o'-glucosidase, which hydrolyses maltose to glucose and is found in many moulds, yeasts and bacteria. Mention should also be made of the report that a so-called y-amylase has been isolated from Aspergillus awamori. Thus, the variety of amylases produced by microorganisms is quite considerable. They do not always occur at the same time in one culture, but even if only two or three are present, the problem of estimating their individual activity accurately can be of considerable magnitude.

2. Applications of amylases

Mould amylases in the form of "Koji" (mouldy bran or mouldy grain) have been used for more than 1700 years in Japan for the manufacture of alcoholic beverages from rice. The use of mouldy bran in North America for the production of industrial alcohol from starchy material was introduced IAEA-SM-134/14 139 over 50 years ago. It has since been shown that it is more economical to hydrolyse the starch for the alcoholic fermentation by mould amylases instead of malt, because the rate of secondary fermentation by yeast and the yield of alcohol are increased. In the production of alcoholic beverages in the western world, microbial amylases have not been used to any great extent, but the increased yield of alcohol as compared with that from malt does, of course, present a certain attraction, and a change in future might be expected in this respect, particu­ larly if starch-containing adjuncts are used. Direct effects on the flavour of the product seem unlikely since very pure amylase preparations showing a high activity per unit weight of material can be obtained. A different question is, of course, the desirability of the presence of residual unfermentable dextrins which provide "body" to such products as beer. Other important applications of amylases are in food processing. Most of the bread produced in the UK and North America is now supplemented with fungal amylase to improve the gassing ability of the dough and to improve the texture of the finished bread. It is important to use fungal amylases for this purpose since the heat-resistant bacterial amylases are not sufficiently inactivated during the baking process. Until a few years ago syrups were conventionally produced from starch by acid hydrolysis. Some of these syrups, i.e. those with a high content of fermentable sugars, contained also a fair proportion of bitter-tasting rever­ sion products such as oligosaccharides with linkages other than a - 1, 4-, This can be completely prevented now by the use of microbial amylases free of transglucosidase. Other industrial applications of microbial amylases are the production of precooked baby foods from cereals, the clarification of syrups for the chocolate and cream industries, the recovery of scrap candy, and the specific removal of starch from fruits and fruit juices and from pectin solu­ tions. In medicine, the use of amylase preparations, as a digestive aid in cases of a deficiency of this enzyme, is on the increase. Since techniques have now been developed to obtain various crystalline amylases with exactly known properties (some of these enzymes are available commercially) one has a comparatively easy means at hand to help elucidate the structure of unknown polysaccharides.

3. Some properties of a-am y lase s

Although the a-amylases of various origin have in common the ability to hydrolyse a- 1, 4-glucosidic linkages with the formation of lower dextrins, these (pure) enzymes are, nevertheless, not identical as can be seen from a compilation by Fischer and Montmollin (1951) in Table I. It is of some interest to mention here some finer points such as the variation of optimal pH of a-amylase with the strain of one species, i.e. A. oryzae (Nagahomo, 1939), thus indicating a possible difference in the composition and/or structure of the amylase molecule for various strains. Although such variations occur, it is noteworthy that immunologically the a-amylases of fungal origin seem to be identical but not however the a-amylases with the saccharogenic amylases (Roy, 1955). So far no sufficiently detailed studies appear to have been made to see whether the glucoamylases or oligo- 1 , 6 -glucosidases produced by the moulds Aspergillus oryzae, A. niger, A. awamori and Rhizopus japonicus are identical. 1 4 0 MEYRATH et al.

TABLE I. PROPERTIES OF a-AMYLASES OF DIFFERENT ORIGIN

H um an Hog Source M alt A. oryzae B .subtilis saliva and pancreas pancreas

Activity/mg N 2350 2400 3 600 400 0 6 200

Activity/mg protein 315 310 500 630 980

<7oN 13.4 12.9 1 5 .8 1 5 .8 1 5 .8 Optimal pH for a c tiv ity 4 .7 - 5 .4 5 .5 - 5 .9 5 .3 - 6 .8 6 .9 6 .9

Optimal pH for stab ility 4 .9 - 9-1 5 .5 - 8 .5 4 .8 - 8 .5 7 .0 - 8 .5 4 .8 - 1 1 .0

Isoelectric point 5 .6 - 4 . 2 5 .3 5 .3

P saccharogenic dextrinogenic

ac tiv ity 9 .8 9 .8 9 .8 9 .8 9 .8

Activation by Cl” - - + + +

Energy of activation (c a l/m o le ) 7 050 10 650 13350 (0 - 15°C) 13 500 13 500 9150 (15 -40eC)

4. Industrial production of ¿y-amylases

Among the amylases, a-amylase has become particularly important since it is now so extensively used in the baking industry. The organism used for the production of this enzyme is Aspergillus oryzae.

(a) Production on solid substrates.

Until a few years ago wheat bran was almost the only raw material used for a-amylase production on an industrial scale. The use of this material, popularized mainly by Takamine, implied that the amylase-producing orga­ nism was grown in the form of so-called solid cultures. The principle employed is to add just enough moisture to wheat bran to allow the mould to develop within the substrate and form eventually a uniformly overgrown mass, called mouldy bran. Temperature and moisture content are the most difficult environmental factors to keep under control in a battery of bran-loaded trays for enzyme production. The temperature in the solid mass of bran rises after some 12 h of incubation; this increase in temperature can even be noticed in small cultures such as conical flasks with a layer of |-1 in. of bran. The tempera­ ture in the incubators containing large masses of bran can rise, in fact, to such high levels that self-sterilisation of the cultures would be the result were it not for artificial cooling which is applied in some form or other. IAEA-SM-134/14 141

It goes without saying that the handling of the large number of trays requires a considerable number of man hours. Jeffreys (1948) has tried to mechanize this process to a considerable extent but, as a whole, it does not appear as though the various enzyme producers have made adequate efforts to obtain fully mechanized processes. When the maximum yield of enzyme is reached the mouldy bran can be dried at temperatures not exceeding 50°C and used as such (e.g. for saccha­ rification of starchy materials used in alcoholic fermentation), or it can be stored until the time is convenient for the extraction, which is usually done by percolation. With a heavy inoculation (e.g. about 107 conidia/g of bran) and a suitable temperature, the maximum enzyme yield can be obtained in as short a tim e as about 24 h.

(b) Subm erged culture

The rapid advances in submerged cultivation methods had a considerable impact on the processes of amylase production. These methods must seem attractive to anyone anxious to mechanize a process more fully. Further­ more, this deep-culture method would allow the use of purer raw materials with the result that: ( 1 ) enzyme extracts are less strongly coloured; (2) enzyme powders have a higher specific amylase activity; and (3)the enzyme can be much more easily crystallized, should this be desired. Normally the raw materials for deep-culture production of fungal amylase are flour or starch, supplemented with inorganic salts, and sometimes such compounds as dried tankage, stillage, corn steep liquor, or yeast extract. These latter organic compounds contribute again considerable proportions of impurities to the substrate, but they contain very valuable stimulating agents for some strains. It does not seem appropriate to group strains so as to differentiate between those suitable for stationary culture and those suitable for submerged culture without considering the substrate composition. In fact, there is information available showing that a particular change in nutrient composition has different effects in submerged and in stationary cultures. Thus, in our investigations on citric acid fermentation by Aspergillus niger it became clear that while a certain combination of trace elements was stimulating to the production of citric acid in surface cultures, these same trace elements in deep culture and in otherwise identical conditions were inhibitory to citric acid formation. The effects of inoculum size, which we have studied recently in some detail, give a similar picture. In certain substrates there is little difference in maximum yield of cell material between cultures originating from large or small inocula when grown in stationary form; however, in deep culture, these same substrates show that cultures from small inocula give a considerably lower yield of mycelium than those from large inocula (Meyrath, 1964). While in bran cultures inoculation is preferably done with conidia, in deep culture with liquid substrates, mycelium is also used as inoculum. Nevertheless, each batch is started with a well-sporulated culture of the mould that has been produced in the laboratory and grown for a relatively short period in submerged culture, resulting in the germination of conidia and the formation of some mycelium. This whole culture is then transferred into the actual fermentation vessel where most of the amylase is produced. 142 MEYRATH et al.

E 100« • Surface 3 О Shake ¿■Deep * Vermiculite □ßron

2

1

2 4 6 days

FIG.l. Production of a-am ylase by various methods of cultivation in high substrate concentrations.

Instead of two, several such stages could be used. The exact procedure is, of course, a trade secret and is bound to vary quite markedly between various enzyme producers.

(c) Synthetic substrates in solid culture Mention should be given here also of the possibility to combine certain desirable points of submerged and of bran cultures. One advantage of sub­ merged cultures lies in the use of synthetic or semisynthetic substrates. This allows considerable freedom in the choice of their composition and also results in purer enzyme preparations. The enzyme producer who is geared for the use of solid cultures (bran) could, nevertheless, take advantage of the synthetic substrates by using supporting (or "solidifying") material which does not contribute any impurities to the enzyme extract. An ideal substance in this respect is vermiculite (Meyrath, 1965). Liquid substrates can be adsorbed on the porous material and, in many respects, the properties of bran cultures are imitated. In fact, the rapid development and high yield of enzyme of the bran cultures can also be obtained with vermiculite cultures. A comparison of the various methods of cultivation with our strain is shown in Fig. 1. The most optimal conditions for the individual methods of cultivation (surface, submerged, bran, vermiculite) were selected.

5. Observations on gene expression in a-amylase formation by Aspergillus oryzae

(a) Type of amylase produced Among the various components of the amylase system the presently used strain produces virtually only a-amylase. There does not seem to be any IAEA-SM-134/14 143

TABLE II. RATIO OF DEXTRINOGENIC TO SACCHAROGENIC ACTIVITY (Ed/E s ) WITH DIFFERENT INOCULUM SIZES OF Aspergillus oryzae CONIDIA ON BRAN SUBSTRATE

106 conidia/ml 18 X 106 conidia/ml

Time of cultivation (h) ^ ^ E^

39 6 .0 0 1 .0 5 6 .1 0 0 .8 2 5

48 6 .1 2 1 .3 4 6 .0 4 1 .0 0

62 6.11 1 .2 5 6 .1 0 1 .2 6

72 6 .0 4 1.11 6 .21 1.10

135 7 .5 5 1 .1 7 7 .61 1 .2 0

TABLE III. RATIO OF DEXTRINOGENIC TO SACCHAROGENIC ACTIVITY (Ed/E s) IN Aspergillus oryzae USING VARIOUS INITIAL pH VALUES IN BRAN SUBSTRATE

Initial pH = 5 .9 5 Initial pH = 3 .9 5 Initial pH = 3 .9 5 T im e of Ed (H 2 S 04) Ed (H 3 P 04) Ed cultivation (h) pH Es pH Es pH Es

39 6 .0 0 1 .0 5 3 .9 0 _ 3 .9 6 _

48 6 .1 2 1 .3 4 3 .8 7 1 .00 3 .9 2 -

62 6.11 1 .2 5 4 .0 4 1 .28 3 .9 8 1 .2 2

72 6 .0 4 1 .11 4 .0 2 1 .1 2 3 .9 7 1 .0 9

135 7 .5 5 1 .1 7 4 .7 2 0 .8 2 4 .2 5 -

Initial pH = 5.11 Initial pH = 5 .0 4 Initial pH = 6 .7 5 T im e o f (H 2 S 04) Ed (H 3P 0 4) Ed (NaOH) Ed cultivation (h) pH Es pH ES pH Es

39 5 .1 5 1 .0 5 5 .0 6 1.07 6 .6 8 1 .1 3

48 5 .2 5 1 .2 9 5 .0 5 1.03 6 .5 7 0 .9 6

62 5 .2 7 1 .1 5 4 .9 8 1 .2 6 6 .6 2 1 .1 5

72 5 .2 8 1 .0 0 4 .7 2 1 .1 4 6-62 1 .11

135 7 .0 4 1 .1 2 6 .2 6 1 .2 5 8 .0 0 1 .1 8 144 MEYRATH et al.

• maltose. IO<ÿl H о maltose, 5 çyl x maltose.8 g/l:glucose.2g/l ® maltose, 5 q/l;glucose,5q/l a maltose, 2 g/l;gkJcose,8g/l 600

3 0 0 -

•mg MDW ЮС/ml "So 2З0 3ÔO

FIG. 2. Influence of glucose in combination with maltose on a-am ylase production. Note: The enzyme units here are on a different scale than in Fig.l.

• maltose fbche О moltos« Kodak x cenobios« yellow dextrin. Merck white dextnn. KWD 6 0 0 Noredux dextrin. 160/60 white dextrin. KWC Noredux dextrin. M 157 Noredux dextrin.M155 glucose

3 0 0

mg MDw/lOOml

FIG.3. Relation between a-am ylase formation(E) and growth(MDW = mycelium dry weight) using various carbon sources. Note: The enzyme units here are on a different scale than in Fig.l.

significant quantity of typically saccharogenic amylase. This can be concluded from the fact that separation studies on ion-exchange columns only gave one main fraction of amylase — with dextrinogenic activity (Meyrath, 1957). Furthermore, the ratio of dextrinogenic to saccharogenic activity of the amylase extract under various methods of cultivation and at various stages of culture development remains constant, as can be seen in Tables II and III. Nevertheless, there are at least two fractions of a-amylase, which can be considered as isoenzymes, that are always found in the wild type of our strain. These two fractions seem to have very closely related properties as they cannot be fractionated by chromatographic methods,neither on ion-exchange columns nor on paper chromatograms. It is very doubtful whether the two forms of a-amylase crystals which can be obtained from A. oryzae coincide with the two electrophoretically distinct fractions since we have often been able to observe that the one or the other crystal form occurred during the recrystallization procedures. The activity of a-amylase from A_. oryzae can most conveniently (i.e. most simply, accurately and quickly) be estimated by determining the slope IAEA-SM-134/14 145 of the decrease in iodine-dextrin colour in a colorimeter with a neutral density filter, since there is proportionality between intensity and time over about 50% of the reaction curve.

(b) Inducibility of a-amylase in Aspergillus or.yzae

The fact that industry uses mostly starchy raw materials for fungal o'-amylase production does not necessarily mean that strach is necessary as an inducing agent for a-amylase production. It is true that on some raw materials more a-amylase is formed than on others and that glucose or sucrose are not very suitable carbon sources for our strain (Fig. 2). Further­ more, it can be shown that maltose is as good a carbon source as various qualities of starch or dextrin is as far as amylase production in the actively growing phase of the culture is concerned; in fact, glycerol can be as good a carbon source as starch or maltose and with cellobiose there is a tendency to produce more a-amylase in the actively growing phase than with maltose, starch or dextrins (Fig. 3). There is an interesting analogy to these latter observations. Tonomura (1959) found that isomaltose can be a better carbon source than maltose or starch. Whether these effects are to be interpreted as a direct inductive phenomenon cannot be decided as there are no molecular biological investigations in this respect.

(c) Environmental effects

The environmental effects on a-amylase formation in fungi are so numerous that it is out of the question to discuss them all here, but some of them have been chosen to be able to discuss some significant points in the general problem of gene expression. A phenomenon of recurrent occurrence is the fact that caramelized carbohydrates tend to yield more a-amylase than non-caramelized ones. We thought that this phenomenon might be of interest as there exists a parallel to trace-element supply. Thus, we could show that caramelized sugars (glucose or maltose) react very markedly with Cu++. This reaction manifests itself strongly in the abolition of characteristic u.v. -absorption peaks. We interpreted these reactions as chelating reactions. Now it has been shown (Steiner, 1960) that for our strain a Cu++deficiency in a strictly defined fermentation medium is of advantage to a-amylase formation although growth is not markedly affected under these conditions (Table IV). Copper is not a component of the a-amylase molecule, but calcium is. Whether Cu+ + is able to prevent the incorporation of Ca+ + into the molecule remains to be investigated, but it does not seem to be very likely as there is no relation between a-amylase formation and Ca++ supply; in fact it is very difficult to prepare media so deficient in Ca++that a-amylase production is impeded. Allied with trace-element effects are also the very marked effects of inoculum size on the properties of the ensuing cultures. Although no transfer effect of trace elements is involved, various combinations of these elements influence markedly the characteristics of cultures originating from different inocula with regard to the rate of growth, the amylase formation and the maximal yields. Inoculum size proved to be a very suitable tool to investi­ gate a phenomenon which may be considered as physiological differentiation. This differentiation also showed a parallel to morphological differentiation, i.e. conidia formation. Thus, whether a culture in advanced phase is to 146 MEYRATH et al.

TABLE IV. INFLUENCE OF COPPER IONS ON SPECIFIC AMYLASE FORMATION (E/MDWa) IN Aspergillus oryzae (Steiner 1960)

Incubation time (days) \i% Cu+ +/1 4 5 6 7

0 6 .2 7 .3 1 4 .4 1 3.0

2 7 .6 8 .8 1 3 .4 18.0

4 5 .6 6 .3 1 0 .2 1 0 .5

10 4 .7 4 .8 6 .2 7 .3

20 3 .6 4 .7 4 .2 5 .9

40 3 .8 4 .4 4 .6 5 .9

100 3 .9 4 .6 5 .6 6 .3

200 4 .1 3 .3 5 .0 5 .9

2000 3 .3 3 .7 5 .2 5 .8

20 000 4 .2 3 .6 5 .2 6 .2

a E/MDW = Enzyme activity per unit weight dry mycelium.

mg. dry wc. mycelium/100 ml

FIG. 4. Incorporation of N O-O) and excretion of organic N(A-A.A-A) in cultures of Aspergillus oryzae using large (•- •, A-A) and small (O - O, A- A) inocula in deep culture (low substrate concentration). (From McIntosh and Meyrath, J.gen. Microbiol. 33 (1963) 57.) IAEA-SM-134/14 147

FIG.5. Specific a-am ylase formation ( E/MDW ) at various stages of growth(MDW) in deep culture (low substrate concentration) using large ( • -• ) and small (O-O) inocula.

FIG.6. Influence of inoculum size on a-am ylase production in a high-concentration substrate "solidified" with vermiculite. show rapid or slow growth is determined already at early stages of culture development. Poor growth in advanced phases (brought about by either large or small inocula depending upon trace-element supply) was allied with the formation of growth-inhibitory substances at very early stages of growth, while stimulating substances (found always in somewhat later stages of growth) obviously have been unable to counteract fully the action of inhibitory sub­ stances. No inhibitory substances could be found in the very early stages of those cultures that showed rapid growth in advanced phases (Meyrath and McIntosh, 1965). These inhibitory substances can be separated by paper 148 MEYRATH et al. chromatography (work in progress). In Aspergillus oryzae rapid growth and high maximum yield of cell matter were allied with a higher conversion co­ efficient of the carbohydrate supplied; furthermore, those cultures that were able to grow fast but gave low maximal yields of cell matter were charac­ terized by an inefficiency to incorporate inorganic N into cell matter. How­ ever, the excretion of organic nitrogenous compounds could, under these circumstances, be very strong (Fig. 4). It was also evident that inefficiency in incorporating inorganic N into cell matter and excretion of organic nitro­ genous compounds was allied with high a-amylase yields both in absolute as well as in relative (to mycelium) terms (Fig. 5). Another example of a marked inoculum-size effect in »-amylase formation can be seen in Fig. 6 where solid cultures (vermiculite) were used as a test. Considering that inoculum size can markedly influence the rate and extent of a-amylase production, the differentiating effect for growth seems to apply also for single enzymes such as a-amylase. The kind of compounds that are able to initiate this differentiation at an early stage of culture development and which are involved here has not yet been investigated, nor has their degree of specificity. Amylase production in industrialized countries is preferably carried out by submerged fermentation. Hence, it will normally be advisable to do all screening tests for high-yielding strains with this method of propagation in order to circumvene most of those problems that are tied up with trans­ ferring results obtained on surface culture to a submerged culture method. Our strain, while giving industrially attractive yields in submerged culture, performs best in surface culture under the conditions tested so far, as can be seen in Fig. 1. An interesting question to answer is what makes a culture perform better (e.g. with regard to a-amylase production) in surface culture than in submerged culture? First, the kind of submerged culture used needs some qualification. Figure 1 shows that while higher yields can be obtained in shaken culture than in deep culture, the cultivation time to reach the maximal point is longest of all the variants tested and is longer than in the surface production in liquid media. Amazingly enough, production on solid media (such as moistened bran or substrate-impregnated vermiculite) is most productive, at least as far as rate of production is concerned, and maximal yields are also among the highest if one calculates the amylase activity per unit volume of the liquid phase. Deep cultivation produces yields which are considerably lower than those under any other conditions and has a fermentation time that is about equal to that of solid cultures. Surface cultivation on liquid media takes considerably longer to reach maximal yields, these being of a similar order of magnitude as those from solid cultures. It should be pointed out that the results given here were selected insofar as the best substrate composition tested was chosen for each method of cultivation; this was considered advisable since there is an interdependence between medium composition and method of cultivation. The above phenomena would seem of interest from several points of view: (1 ) a-amylase production on solid media such as bran (a by-product in agriculture) can be advantageous to developing countries as no sophisti­ cated machinery is required. The only point to consider is that extraction from bran involves a certain dilution, but in many cases a dried fermented bran with high amylolytic activity can be used; (2 ) it would seem possible to present some kind of explanation for the phenomenon of different behaviour under different methods of cultivation, since still cultures can be made to IAEA-SM-134/14 149 produce a-amylase rapidly or slowly, depending upon whether solidifying agents are incorporated into the substrate, provided that the whole cake is always sufficiently porous to allow the development of the organism through the whole substrate. It would appear to us that the phenomenon of different behaviour under different methods of cultivation is quite akin to that produced by inoculum size, i.e. the establishment of particular microclimates at early stages of culture development which are of particular importance for the properties of the culture at advanced stages. It is easy to perceive that the excretion of metabolic compounds produces a different environment if these compounds are immediately diluted into the medium (as in shake or deep culture) instead of remaining for some time in the immediate vicinity of the mycelium, as is the case in surface cultures on liquid media or even more so in solid cultures. Other observations also led to the conclusion that a metabolic pattern imposed on a fungal culture in the early stages of development keeps on being reproduced in more advanced phases, even if now the conditions are such that a different metabolic pattern would be expected. Thus, a culture of A. oryzae endowed with a high rate of a-amylase production does not adopt the lower rate of a-amylase production of a shake culture if the method of cultivation is changed from surface to shake culture, after 48 hours of incubation with only about 1 /4 of total amylase produced at that stage. The disadvantageous converse of changing a shaken culture into a surface culture has been shown to be equally true (Volavsek, 1970).- In conclusion, the author believes that, as a working hypothesis, there are good reasons to consider both the inoculum size effects and the effects on cultivation methods as effects of physiological differentiation. Considering that we are beginning to understand why deep cultures and surface cultures behave differently, we did not think it particularly advisable within our mutation program to screen and select for high-yielding strains in submerged culture. Instead, it was more attractive to us to try to make use of some very simple screening methods on solid cultures which would indi­ cate the promising mutants at the very earliest possible stage of the screening p ro g ra m .

6 . Selection of mutants

(a) Survivor curves

A prerequisite for mutant selection is a knowledge of the survivor curves under the specific methods of testing. Figure 7 shows the survivor curves of Aspergillus oryzae conidia under the influence of the mutagenic agents u.v. irradiation, gamma rays and fast neutrons. According to several authors the selection of mutants is normally done at a survivor ratio of about 50%, lower survivor ratios are used at the start of a mutation program if a wild^type strain is mutated. This was one of the reasons why we decided to considerably increase the percentage kill. Another reason was the fact that conidia of A_. oryzae are multinucleate and it was thought best to use very high irradiating doses so that any survivors^ are likely to arise from only one nucleus in the conidium. The highest degree of survivors in our experiments was about 1 ; 10 2, the lowest about 1 : 1 06. If only 1 in every 102 conidia survives, the chance that more than one but of 150 MEYRATH et al.

Neutrorà

FIG.4 . Survivor curves of Aspergillus oryzae conidia under the influence of various kinds of irradiation (•-•» u.v.), (O -O , gamma-rays), (□ -□ , fast neutrons).

five or six nuclei in one conidia remain viable must be extremely small. It is expected that the mutants thus selected are more stable and express more obviously any change in their ability to produce a-amylase.

(b) Screening techniques

The ability of individual strains to produce and excrete amylases mani­ fests itself easily by plating out dilutions of suspensions of microorganisms on starch-containing substrates. If sufficiently high concentrations of starch (e.g. 4%) are used a halo of dissolved starch around the colonies is easily perceptible. With lower concentrations of starch it is advisable to use an iodine solution for detecting amylase activity. In order to be able to save the colony or to be able to carry out repeated tests it is possible to cut off a thin slab of the lower part of the agar plate and incubate the cultures again. A thin layer can, for example, be cut off by transferring the agar plate onto a stainless-steel dish so designed that a thin wire is led over the edge of the dish leaving the cut-off part in a small 2-mm depression. The colour reaction can be performed on this slice. In the present experiments we used 4% starch as ingredient in a suitable production medium, and a partial re-precipitation of starch was obtained by incubating the plates for at least four days at about 4°C. In this way quite an opaque medium was obtained where clearing zones could easily be m easu red . The irradiated conidia of A_. oryzae were diluted so as to obtain not more than 10 colonies per plate. Thousands of colonies were screened according to this procedure. IAEA-SM-134/14 151

0 5 10 15 20mm Diameter of colonies

FIG. 8. Distribution of colony diameters (probability plot) at three and four days of incubation after irradiation of conidia with u.v. and with gamma rays to yield 1% survivors. С = control (non-irradia ted conidia).

As expected, the size of the colonies after irradiation varied very greatly. This can be seen (Fig. 8 ) from that part of the frequency curve (sum of frequencies in a probability graph) extending up to about 50% of all colonies which is very much flatter than for the wild strain. The fact that there are few large colonies, i.e. larger than the average wild-type colonies after gamma irradiation, is shown by the sharp increase in the slope of the frequency curves at approximately, or somewhat below, the point for the mean colony diameter of the wild strain. With u.v. irradiation (of the same degree of mortality as for gamma rays) the average colony diameter is, in fact, considerably smaller than for gamma rays. Figure 8 also shows that the irradiated cultures are particularly retarded in colony growth at the beginning of their development, while a considerable proportion of them, especially those close to the mean colony size of the wild strain, do recover and show faster growth. This is indicated by the tendency of the frequency curves to form a pronounced change in slope after extended periods of growth. The wild strain showed remarkable constancy in size of colony and appearance. The deviation from the mean of the colony diameter was quite small, as indi­ cated by the comparatively steep line in a probability graph. (The fact that a straight line is obtained shows that the distribution is normal. ) A priori, it would not be disadvantageous to use poorly growing strains for industrial amylase production as long as high titres of amylase are ob­ tained. In fact, poor growth would probably prove to be of advantage as there are considerable technological problems in handling heavily grown mashes with filamentous fungi. It was thought most advisable to select those mutants with highest ratios of clearing-zone diameter to colony diameter irrespective of the colony size. A distribution of these ratios for several hundred colonies for both the wild strain and the irradiated culture for 152 MEYRATH et al.

R i ti ó

FIG. 9. Distribution of ratios of amylase activity to mycelium (diameter of halo to that of colony) on screening plates for irradiated conidia (1% survivors) after three and four days of incubation. С = control (non-irradiated conidia).

FIG. 10. Influence of various u.v. doses on distribution of ratios (amylase to mycelium) after four days of incubation. IAEA-SM-134/14 153

О

FIG. 11. Influence of various gamma ray doses on distribution of ratios (amylase to mycelium) after four days of incubation.

FIG. 12. (a) Absolute (E) and (b) specific (E/MDW) amylase production in surface culture of some selected mutants after irradiation once with various u.v. doses. 154 MEYRATH et al.

selected doses after incubation for three and four days respectively is shown in Fig. 9. Whereas these ratios again showed a normal distribution with a very low degree of variance for the wild strain, a skewed type of distribution was obtained for the mutants which showed a considerably larger degree of variance than for the wild strain. It can be seen also that after short periods of incubation irradiated conidia (particularly with u.v. ) produce colonies with a higher (specific) amylase activity, and that after further incubation (4th day) this activity is being reduced so that now some 50% of the colonies only show increased amylase formation. The results of the 4th-day readings at various doses of u.v. irradiation are in Fig. 10. It is obvious, as has often been described for all mutagen treatments, that those doses with a moderate degree of kill are more suitable for selecting mutants than those with a high degree of mortality. At 30 min of u.v. treatment (corresponding to a 99.999% kill) there are, in fact, only disadvantages for selecting higher-amylase-yielding mutants. A similar tendency in the relation between higher-yielding mutants and mutagen dose has been obtained with gamma radiation although the sequence is not so regular. Furthermore, at very high doses (150 krad, corresponding to a 99.999% kill) there is still a fair proportion of colonies showing a specifically higher activity than the wild strain (Fig. 11). At lower doses (70 to 110 krad) more than 50% of the colonies show a higher amylase activity. Some hundred mutants were picked off and their colony morphology and microscopical properties were checked. They were then transferred several times on a sporulating medium (some mutants had lost their ability to produce conidia), checked again for colony morphology and tested for amylase pro­ duction in surface culture on liquid media.

(c) Performance of the isolated mutants

Progress curves of both mycelium and amylase production were made and some of the results are shown in Figs 12(a) and (b). It can be seen that while there were considerable differences in the size of the colonies as well as in the ratios of zone diameter to colony size on the screening plates, differences in the production tests were relatively small. Many of the poorly growing mutants on the screening plates turned out to grow quite well after several transfers. There were no marked differences to the wild strain either in the time to reach the maximum enzyme production or in the time to reach the maximum mycelium production. Having shown earlier that inoculum size can exert a considerable in­ fluence on absolute and relative (to mycelium) amylase production in A. oryzae, both in submerged and in solid cultures, we should be aware of the fact that these performance tests were carried out under our standard conditions of inoculation, i.e. about 107 conidia/100 ml. The screening of the cultures, however, was carried out on individual colonies that developed from single conidia and probably even from single nuclei, thus representing extremely small inocula from which these cultures had been formed. Therefore, strictly speaking, we have selected those mutants for best performance at their very early stages of development when there was very little influence on the cultures by excreted metabolic products, since the accumulation of these is of course strongly dependent on the number of germinating conidia per unit volume. IAEA-SM-134/14 155

If the performance over the wild strain shouldprove to beconsiderably- enhanced by using small inocula then ways would have to be found in industry to make use of both the advantages of large inocula (i.e. rapid initial develop­ ment) and those of small inocula (i. e. higher specific amylase production).

BIBLIOGRAPHY

FISCHER, E.H., de MONTMOLLIN, R., Helv.chim.Acta 34 (1951) 1994.

JEFFREYS, G .A ., Food Industries 20 ( 1948) 688.

MEYRATH, J., Über die Bildung von Amylases durch Aspergillus flavus-oryzae, P.G. Keller, Winterthur, Switzerland (1957).

MEYRATH, J., Experiential (1964) 257.

MEYRATH, J., J.Sci.Fd Agrie. 16 (1965) 14.

MEYRATH, J., McINTOSH, A.F., Can.J.Microbiol. И (1965) 67.

NAGAHOMO, T ., J.agrie.Chem. Soc. Japan 15 ( 1939) 753.

ROY, D.K., Ann.Biochem. exp. Med. 15(1955) 37.

STEINER, K., Zentralblatt Bakt.Parasit. 11, 114 (1960) 47.

TONOMURA, K., NAKAMURA, N., SUZIKI, H., TANABE, O., Report Ferm.Res.Inst. Chiba (1959) 41.

VOLAVSEK, G., Thesis, Vienna (1970).

IA£A-SM-134/11

MUTATION STUDIES IN Streptomyces aureofaciens

M. BLUMAUEROVÁ, A.A. ISMAIL*, Z. HOàîÂLEK, Z. VANÊK Institute of Microbiology, Czechoslovak Academy of Sciences, Prague, Czechoslovak Socialist Republic

Abstract

MUTATION STUDIES IN Streptomyces aureofaciens. Biochemical mutants of Streptomyces aureofaciens, which were blocked in the biosynthesis of tetracycline antibiotics and marked by additional blocks in the primary metabolism, were induced by physical and chemical mutagens. U.V. light and N-methyl-N'-nitro-N-nitrosoguanidine were the most efficient mutagens in both mutation steps. The mutagen and strain specificity, some correlations between the biosynthetic activity and other mutant characteristics as well as the convenience of the methods used are discussed.

INTRODUCTION

The study of genetic regulation of biosynthetic pathways has, so far, been limited mostly to the primary metabolism. The mechanisms governing the synthesis of secondary metabolites (antibiotics, pigments, etc.) has not been studied in detail because of the complicacy of the experimental ap­ proaches. For any genetic analysis of these mechanisms, the two-stage preparation of mutants is necessary. It includes, firstly, the induction of blocks in various steps of the secondary metabolic pathway and, secondly, the genetic marking of such mutants by additional blocks of the primary metabolism. This paper summarizes the experience we have gained in the application of the two-step mutation process in various strains of Streptomyces aureo­ faciens differing mutually from each other with respect to the production of tetracycline antibiotics and their biologically inactive derivatives.

ISOLATION OF MUTANTS BLOCKED IN THE BIOSYNTHESIS OF SECONDARY METABOLITES

For the preparation of mutants, two standard strains of S. aureofaciens were used — the low-producing strain, Bg, and the producing strain, 84/25, derived from the former one. The effect of u.v. light, gamma radiation, X-rays, N-methyl-N 1 -nitro-N-nitrosoguanidine (MNNG), nitrous acid and nitrogen mustard was compared [1]. The production of secondary metabo­ lites was followed by means of paper chromatography in methanolic extracts of mycelium and in fermentation fluid of submerged cultures [2]. On the basis of altered morphological characteristics, the collection of 635 mutants

* Permanent address: National Research Centre, Microbial Genetics Research Unit, Dokki, Cairo, United Arab Republic.

157 158 TABLE I. COMPARISON OF THE BIOSYNTHETIC ACTIVITY OF PARENT STRAINS AND BLOCKED MUTANTS OF Streptomyces au reo facien s 3 UTEJJS LMURV et l. a t e BLUMAUEROVÁ Í + I I + I Í + + I + + I + + SSBp шедпад Г u « О ^ tu к вГ о о + + + + > > + I + + . z О H D S л с •S „ I 2 •о : — я ь й •о •о g = д J= л ; О 2 S •§ &-а ' ^ ,5 ’s к и и в а о 2 E ' * 'Sö exÜ í -S g О e _с 1) С М) И & В г ев I со J3 « >> « « 2 3-5 s Р о (3 о о ьо <4 W л О >> Ç Е_ ° О «О i id 5crt о <и Í O g с Í) Ч) ° . 2 с « - U « §Q 2 « о g .2 *’н 4J » ï а ©я <и 3-Ё' а S « 2 «J *2 1

ТЗ Ä E •S < > и Û U 1 1 Х> JD 3 л Н <3 42. 6 ? I <и (U ■§e ■§ < он IAEA-SM-134/11 159

TABLE II. SUMMARIZED FREQUENCIES OF SPONTANEOUS AND INDUCED MUTANTS OF Streptomyces aureofaciens BLOCKED IN THE BIOSYNTHESIS OF SECONDARY METABOLITES 2

Mutant class M utagen^ I II IIIe IV V VI-XI XII

- 98.1 0 0 1 .6 0 .3 0 0

u .v . lig h t 8 3 .9 0 .3 5 .4 6 .5 3 .6 0 .3 0 .0 0 9 d

y-radiation 6 9 .8 0 4 .7 2 0 .5 5 .1 0 0

X -rays 7 3 .7 0 .1 1 .2 1 8 .9 6 .0 0 0

N -M ustard 94.1 0 0 3 .8 2 .1 0 0

MNNG 8 1 .8 0 .7 0 1 1 .4 6 .6 0 .5 0

h n o 2 8 7 .0 0 0 7 .8 4 .7 0 .5 0

a Frequencies of individual metabolic types in per cent of the total population size of the two parents (84/25 and Bg). k The following doses of mutagens were used: u.v. light, 40 and 50 sec; y-radiation and X-rays, 100 kR; MNNG, 1 mg/ml, 60 min (in 0-02M phosphate buffer, pH 6.7); N-mustard, 0.01M, 30 min(in.0.15M phosphate buffer, pH 8.0); nitrous acid, 0.05M, 20 min (0.2M acetate buffer, pH 4.0). A detailed description of the methods has been published elsewhere [1]. c Induced in the strain 84/25 only. d Induced in the strain Bg only.

was isolated from about 60 000 colonies and evaluated. According to the results of the chromatographic analyses, it was possible to classify all the mutants tested into 12 metabolic groups (Table I). Mutants in seven of these groups produced new substances differing in their physico-chemicalproperties from the metabolites of the parent strains; in nine groups, the production of different pigments was detected. The frequencies of the occurrence of individual metabolic types in treated and untreated populations of the two parent strains are summarized in Table II. It is evident from the comparison of data in Tables I and II that the most frequent alteration, both that induced with all mutagens used or that spontaneously originated, is the loss of the ability to produce aureovocin and substance В (ob­ served in 7 from 12 metabolic types), or its remarkable decrease (in a large number of mutants in the first group). On the other hand, mutations leading to the complete suppression of tetracycline biosynthesis (groups II, III, V, XI and XII) were induced in both parents less frequently. Among mutational alterations causing production of qualitatively different metabolites, the block in méthylation at C6 of the tetracycline skeleton (detected in four mutant groups) most frequently took place. In mutants of the 6th and 10 th group this block apparently caused a complete loss of the specific methyltransferase, whilst in mutants of the 7th and 8th group (pro­ ducing both tetracyclines and demethyltetracyclines) it caused only a de­ crease of the enzyme effectiveness or its quantitative deficiency. The pro­ duction of demethyltetracyclines was usually connected with the formation of red fluorescing substances F and G and of red-violet or red-brown pigment. 160 BLUMAUEROVÁ et al.

g Я

I -S

E 5

•S 4) ft H

©■< •s *s H £

ó S complete medium containing 15 pg of acriñavine/ml.

z b Time c of irradiation Concentration 40 and 1 50 mgAnl sec. and 3 mg/tel applied for 30, 60 and 120 min in Tris-maleate buffer (pH 9.0) [11]. IAEA-SM-134/11 161

Table II also shows the relative specificity of the action of the mutagens used. Qualitative alterations in metabolism leading to the production of new substances were induced with u.v. light, MNNG and nitrous acid only. Gamma radiation and X-rays generated high morphological variability 1 in both strains. However, they induced only quantitative alterations in the production of standard metabolites or the loss of the ability to produce them. The two mutagens had also a rather specific inhibition effect on the pro­ duction of aureovocin, accompanied by a decrease in tetracycline production (group IV). This type of mutant was induced 2-5 times more frequently than after treatment with other mutagens. All mutants in the 4th and 5th group were asporogenic. A positive correlation between the decrease of the biosynthetic activity and of the sporulation was observed in mutants of the 1st group. A number of mutants (groups I, III, IV, V, XI and XII) had a considerable decrease in viability and differed from the standard strains by altered growth rate. Further study of the mutants [6 ] showed that the blocks in the secondary metabolism were often connected with a number of other biochemical changes (the respiratory activity, the utilization of carbon sources, etc. ). In mutants of groups I, III, IV, V, VII and VIII the additional changes in biosynthetic activity 2 w ere observed. However, complete reversion to the parent type never took place. These results suggest that the phenotypic expression of most mutants resulted from multisite mutations influencing not only the tetracycline biosynthesis itself, but also some other biochemical processes. Phenotypic charac­ terization of each group under consideration was usually the same for all representatives irrespective of the type of mutagen used.

ISOLATION OF NUTRITIONAL AND DRUG-RESISTANT MUTANTS

For the second mutation step, five stable well-sporulating mutants (B-96, NMG-28, NMG-2, NMG-10 and UV-0202) representing metabolic types IIA, IIB, VI, IX and X, respectively, were selected. Their mutability was compared with that of two strains of standard type (NRRL 22093, 84/25). For the induction of mutations, u.v. light and MNNG proved to be the best, whilst N-mustard was completely ineffective. Out of three methods tested, the "delayed enrichment" [8 ]4 was found to be the most convenient for the detection of auxotrophs. The highest yields of mutants were obtained when complete medium was added after 48 hours starvation. The auxotroph frequencies detected after "delayed enrichment" in various metabolic types are summarized in Table III. The yield of mutants obtained by "total isolation" was the same or lower but the low yield did not justify the laboriousness of this method. With the "filtration procedure" [12] a large number of leaky mutants was isolated. However, the yield of true auxotrophs was lower than with former techniques.

1 The alteration in the size, shape, surface and consistence of colonies and in the character of sporulation [1, 2]. 2 For example, the partial restoration of the ability to produce aureovocin or tetracyclines, or the production of other new substances that had not been previously found either in standard strains or in any type of mutants. 3 Original wild strain of S. aureofaciens [7] ; under our conditions it produced the same metabolites as the the two formerly tested standard strains (for comparison see Table I) and differed from them only by a lower level of tetracyclines (100 fig/fail). 4 Method was modified according to Vladimirov and Mindlin [9] . 162

TABLE IV. COMPARISON OF THE MUTAGENIC EFFECT OF U.V. LIGHT AND MNNG ON AUXOTROPH YIELDS IN Streptomyces aureofaciens NRRL 2209 AND NMG-2 fe - •£><-* * 5 ? -2 СМ со со СМ о о U с СО СО со Н Н

в fl 1 « В e f g I О <+— IAEA-SM-134/11 163

The results of the selection of acriflavine-resistant mutants are sum­ marized in Table III. These mutants occurred with low frequency, both in treated and untreated populations of most strains tested, and differed from each other by the resistance level to 15, 30 or 45 jug per ml of acriflavine. M utants re sista n t to streptom ycin (25 ßg/ml) were not detected either after direct plating of dense spore suspension on drug-containing medium or by additional replication of colonies obtained by total isolation from medium without streptomycin5. Most auxotrophic mutants obtained in individual experiments required arginine. This requirement is a special characteristic of the species S. aureofaciens6. As shown in Table IV, the frequencies of arginine mutants induced in the strains NRRL 2209 and NMG-2 by different doses of u.v. light and MNNG are presented as an example. With one exception7, no other arginine auxotroph responded to citrulline or ornithine, all being apparently blocked in one of the two final steps of arginine biosynthesis. Negative results of cross-feeding and recombination experiments demonstrate the identity of the site of the genetic block. It seems that the polynucleotide region controlling this biosynthetic step is — like one of the loci responsible for the synthesis of aureovocin — rather unstable and sensitive to all mutagens irrespective of the mechanism of their action. In contrast to the above discussed block in the biosynthesis of aureovocin, in this case neither the spontaneous occurrence of arginine m utants 8 nor the spontaneous restoration of the arginine production was observed. All arginine mutants isolated from standard strains belonged only to the m etabolic types I, IV and V, i.e. they had decreased or completely blocked ability to produce aureovocin and tetracyclines (Table I). Also in mutants isolated from the strains B-96 and NMG-28, the deficiency in arginine biosynthesis was usually connected with a decrease in aureovocin production. The comparison of spectra of all auxotrophic mutations detected in various metabolic types after u.v. and MNNG treatment is presented in Table V. Within the remaining types of mutants, the requirement for methionine, purine bases or uracil was the most frequent. The spectrum of mutations was always characteristic for the strain under consideration. The highest mutability was observed in strains NMG-2 and NMG-28 (both strains were obtained from the strain 84/25 after MNNG treatment) and the lowest one in strain 84/25 and in its mutant UV-0202. The growth require­ ments for amino acids other than arginine were often associated with quanti­ tative alterations in the secondary biosynthesis. Blocks in methionine and adenine biosynthesis induced with MNNG, however, led to the complete loss of the original producing activity of the prototroph parents. All mutants belonged to the metabolic types V or XI only and their activity did not change even on medium containing a high concentration of the required growth factor.

5 Recently. 13 mutants were isolated from the strain NMG-2 that appeared to be streptomycin resistant; their characterization is in progress. 6 The production of a large number of arginine mutants in S. aureofaciens is also described by Polsinelli and Beretta [13] and by Sermonti [14]. 7 The u.v. -induced mutant of the strain NRRL 2209 growing both on arginine and ornithine. 8 The e x c ep tio n a l strain is 84/25 in which all spontaneously originated mutants (Table III) required only a rg in in e . 164 S1 > и < PQ а> j о w со +JН г о .

-а SPECTRUM OF INDUCED NUTRITIONAL MUTANTS IN VARIOUS STRAINS OF Oí

оо оо оо оо oto оо оо о оо оОООО Н Nоооо ОО ОО ОО л ю о з - о > ю - ю о t- ю ю о> со с- оз со о ю л N OОО NОО и N i — IОО О Ю г н о о о о оо оо оо оо оо 0 0 5ООо ОООо NОООО g £ g 1 з Th оз 2 Cs)“ СО оо о 2 о о о о о о о о о оо о ОО О ОооОООО ОО оо оо оо оо оо оо СО Оi—< о с— о со 2 о LMURV et l. a t e BLUMAUEROVÁ ООО ООО ОО ОО 2 о I — I 03 oto oto 003 I—I 03 с а < мооdoоо оо о « — <оооо 5 03 05 2 о оо о о оо о 2 о о оо о 2 о 2 О 2 о оо

Relative distribution of mutant phenotypes in the evaluated collection of 345 mutants isolated from total number of 46 579 survivors. Proposals made by Demerec et al. [15] were followed for the genetic nomenclature. IAEA-SM-134/11 165

After determination of mutual fertility [16] of mutants representing various metabolic types and differing from each other by genetic markers, 20 of them were subjected to additional mutagenic treatment to obtain double or triple marked strains for genetic mapping. The metabolic pathway of tetracycline biogenesis from a hypothetical non-aketide precursor includes a relatively small number (1 1 ) of enzymatic reactio n s [17]. Genetic determinants controlling these reactions represent not less than 11 structural genes. Therefore, it is reasonable to assume that the position of at least 11 loci responsible for the control of individual biosynthetic steps could be mapped on the chromosome.

REFERENCES

[1] MRAÓEK, M., BLUMAUEROVA', M., PALECkOVA, F., HOäfXLEK, Z., Regulation of biosynthesis of Secondary metabolites: XI. Induction of variants in Streptomyces aureofaciens and the specificity of mutagens, Mutat.Res. 1_ (1969) 19. [2] BLUMAUEROVÀ, M., MRA¿EK, M., VONDRÁCKOVÁ, J ., P.ODOJIL, M.. HOSTÁLEK, Z., VANËK, Z., Regulation of biosynthesis of secondary metabolites: IX. The biosynthetic activity of blocked mutants of Streptomyces aureofaciens, Folia microbiol., Praha 14 (1969) 215. [3] URX, M ., VONDRÁCKOVÁ, J.. КОУАЙК, L., HORSKY, О., HEROLD, M., Papierchromatographie der Tetracyklinstoffe, J. Chromatog. 1Л (1963) 62. [4] LEVINE J., GARLOCK, E.A., FISHBACH, H., The chemical assay of aureomycin, J.Am.pharm.Ass. (Sei.Ed.) 38(1949) 473. [5] PODOJIL, M., VANËK, Z., VOKOUN, J., CUDLÍN, J., BLUMAUEROVA', M.. VONDRÄCEK, M., HASSALL, C .H ., Secondary metabolites produced by biochemical mutants of Streptomyces aureofaciens, 1st Int.Symp. on Genetics of Industrial Microorganisms, Prague, 1970, Abstract Book, p .106. [6] BLUMAUEROVA, M ., Genetic Aspects of the Regulation of the Biosynthesis of Tetracycline Antibiotics, Ph.D.Thesis, Inst.Microbiol., Czechoslovak Acad.Sei., Prague (1969). [7] DUGGAR, B.M., Aureomycin: a product of the continuing search for new antibiotics, Ann.N.Y. Acad. Sei. 51 (1948) 177. [8] LEDERBERG, J., TATUM, E.L., Detection of biochemical mutants of microorganisms, J.biol.Chem. 165(1946) 381. [9] VLADIMIROV, A .V ., MINDLIN, S.Z., A study on the inheritance of the antibiotic production capacity in the course of genetic recombination in Actinomyces rimosus, the producer of Oxytetracycline (in Russian), Genetika 2 (1967) 152. [10] LEDERBERG, J . , LEDERBERG, E .M ., R eplica p la tin g and in d irect sele ctio n of b a c te ria l m utants, J.B a c t. 63 (1952) 399. [11] DELIC, V ., HOPWOOD, D.A., FRIEND, E.J., Mutagenesis by N-methyl-N* -nitro-N-nitrosoguanidine (NTG) in Streptomyces coelicolor, Mutat.Res. 9 (1970) 167. [12] BRAENDLE, D .H., SZYBALSKI, W., Heterokaryotic compatibility, metabolic cooperation and genetic recombination in Streptomyces, Ann.N.Y.Acad.Sei. 81 (1959) 824. [13] POLSINELLI, M ., BERETTA, M ., Genetic recombination in crosses between Streptomyces aureofaciens and Streptomyces rimosus, J.Bact. £1 (1966) 63. [14] SERMONTI, G., Genetics of Antibiotic-producing Microorganisms, Wiley & Sons Ltd., London (1969). [15] DEMEREC, M., ADELBERG, E.A., CLARK, A .J., HARTMAN, P.E., A proposal for a uniform nomencla­ ture in bacterial genetics, Genetics 54 (1966) 61. [16] SERMONTI, G., CASC1ANO, S., Sexual polarity in Streptomyces coelicolor, J.gen.M icrobiol. 33 (1963) 293. [17] VANËK, Z., CUDLÍN, J., BLUMAUEROVA, M., HOSTÁLEK, Z., How many genes are required for the synthesis of chlortetracycline, Folia microbiol., Praha 15 (in press).

DISCUSSION

H. HESLOT: Is there a quick method which could be applied to single colonies for detecting induced mutants with increased tetracycline production? 166 BLUMAUEROVÁ e t a l.

Z. HOSTÁLEK: The screening techniques for the detection of variants with increased antibiotic synthesis always involve problems. This is also true of tetracyclines. H. HESLOT: I have been struck by the great instability exhibited by Streptomyces aureofaciens. Is any investigation being carried out at present to explain this phenomenon? Do you have difficulty in keeping your mutant strains in stable condition? Z. HOSTÁLEK: The instability of some blocked mutants of S. aureofaciens is an unpleasant problem for us. This phenomenon re­ duces the possibility of the choice of individual partners for recombina­ tion. The cause of instability has not been studied in detail. M. MRACEK: What is your opinion regarding the influence of genetic regulation and conditions of fermentation (e.g. aeration, etc.) on the pro­ duction of secondary metabolites in the case of your blocked mutants? Z. HOSTALEK: The level of individual metabolites is strongly in­ fluenced by cultivation conditions. This is a quite common occurrence in the case of all quantitative features, and environmental conditions play a major role here. BIOSYNTHETIC PATHWAYS AND THEIR REGULATORY MECHANISMS (Session 5)

Chairman

S.G. GEORGOPOULOS (Greece)

IAEA-SM-134/31

REGULATION OF AMINO-ACID BIOSYNTHESIS AND INDUSTRIAL PRODUCTION OF AMINO ACIDS

R. HUTTER Institute of Microbiology, Swiss Federal Institute of Technology (ETH), Zurich, Switzerland

Abstract

REGULATION OF AMINO-ACID BIOSYNTHESIS AND INDUSTRIAL PRODUCTION OF AMINO ACIDS. Advances in the industrial production of microbial products, secondary and primary metabolites, have in most cases been made by mutating a parent strain and empirically selecting for better producing mutants. But the knowledge accumulating on the pathways of amino-acid biosynthesis and on their regulation allows a directed approach to strain development and a good understanding of some critical points. To maximize the effect of their efforts industrial microbiologists must take advantage of the advances made in molecular genetics, especially in the fields of regulation, mutation techniques and recombination procedures. Some recent ideas on the molecular regulation of amino-acid biosynthetic enzymes are discussed: regulation in branched pathways, the functioning of enzyme synthesis under derepressed conditions, the question of coordinacy versus non-coordinacy of enzyme synthesis, the mode of regulation (translational or transcriptional). Then three industrial processes of amino-acid production are analysed with respect to their regulatory properties, the production of glutamic acid and lysine by Corynebacterium glutamicum and of tryptophan by Hansenula anómala. It is common to these three processes that critical regulatory blocks can be recognized and that they have to be eliminated by mutation, circumvented by tricks or bypassed by feeding precursors posterior to the blocks.

Advances in the industriell production of microbial metabolites have in most cases been made by mutating a parent strain and more or less empirically selecting for better producing mutants. This is not only true for secondary metabolites as antibiotics or alkaloids, but is also valid for primary metabolites as amino acids. Fundamental studies on biosynthetic sequences and their regulatory properties are usually only undertaken at a stage when satisfactory yields for industrial production have already been reached. But the knowledge accumulating on the pathways of amino- acid biosynthesis and on their regulation allows a precise and directed approach to strain development — or at least a good explanation for what w as found. The biosynthetic sequences for the amino acids have, with the exception of a few details, become clear. Amino acids are usually synthesized by the same routes in sill microbes analysed (exceptions are, for example, lysine [1] and arginine [2]). Diverse strains can, however, differ considerably in their regulatory patterns; our knowledge is far from complete in this respect. In spite of intensive efforts the molecular mechanisms of regulation of enzyme synthesis are not yet completely understood. Only a preliminary picture can be drawn. Some recent ideas on the molecular regulation of amino-acid biosynthetic enzymes are discussed. Then three microbial processes for amino-acid production are analysed: fermentation of glutamic acid and lysine by Corynebacterium glutamicum and of tryptophan by Hansenula anómala.

169 170 MÜTTER

multiple enzymes

multivalent (concerted) or cumulative (additive) feedback inhibition апф^ог repression

sequential feedback inhibition and/or repression _ _ _ ;-----

A — j------► В ------► C F|

reversion of feedback inhibition anchor repression

FIG. 1. Possibilities of repression and feedback inhibition in branched pathways.

P O E D Ц С В A Gene Sequence >-+"•♦ I------\ t ------H ------> / I I \ X Proteins ASase PRTase ISOase/hGPSase TSascA TSase В

Chorismic» Anthranfoc* ^ Y . т ...... Peartirtne . , PRA^ CDRP —> InGP Tryptophan KcactK)ns acid ® acri ® © © ^

FIG. 2. The tryptophan operon of Escherichia coli [21], Pj and P2 = promotors 1 and 2; О * operator; E, D, С, B, A = structural genes. For abbreviations see Table I. IAEA-SM-134/31 171

A. MOLECULAR ASPECTS OF REGULATION OF AMINO-ACID BIOSYNTHETIC ENZYMES

The regulation of amino-acid biosynthetic pathways has been investigated phenomenologically (which enzyme is regulated by which metabolite) in several systems. Some common knowledge has evolved. First, regulation is possible for enzyme synthesis (repression-derepression, frequently acting on all enzymes of a specific pathway) and for enzyme activity (feedback inhibition of the first enzyme of a specific pathway by the end product; stimulation). Second, biosynthetic sequences of amino acids are not independent of each other, but are connected by either using the same enzymes for parallel reactions (e.g. isoleucine-valine [3]) or by forming a branched system of "families", e. g. aromatic amino acids [4] or the "aspartate family" [5]. For the formation of branched biosynthetic sequences a number of regulatory patterns has been established (Fig. 1; [ 6, 7]). As is discussed for L-lysine (cf. Section C) it is easier to circumvent some regulatory types than others to get high excretion of the desired metabolite. Studies on the molecular details of enzyme regulation have been undertaken mainly with two systems, the tryptophan biosynthetic enzymes of Escherichia coli and the histidine biosynthetic system of Salmonella typhimurium. Within the scope of this short discussion we have to restrict our attention to a few points of immediate interest and we will concentrate on the tryptophan system. No attempt for completeness will be made! The structural genes of the tryptophan system of E. coli are genetically linked and form a regulatory unit (tryptophan operon; Fig. 2). In wild-type strains two regulatory mechanisms are evident, feedback inhibition of anthranilate synthetase and repression of enzyme synthesis by tryptophan. In tryptophan-requiring mutants (trp - auxotrophs, trp~) enzyme synthesis is derepressed under conditions of tryptophan limitation [8-11]. Regulatory mutants of different classes have been observed: Anthranilate synthetase may become feedback insensitive for tryptophan [9], enzyme synthesis may become constitutive by mutations in the trp-operator (oc mutants [1 2 ]) or in the unlinked regulatory gene trp R (trp R" mutants [8 , 11, 13]). Three problems of regulation behaviour have been solved, namely the functioning of enzyme synthesis under derepressed conditions, the question of coordinacy or non-coordinacy of enzyme synthesis under different conditions and the type of regulation (translational or tran­ scriptional). The three aspects are discussed separately.

Genes f? 0 E D $ С

DNA

ч DNA-dependent RNA-potymerase 05 per trp-operon) with 1 trp-mRNA ribosomes polypeptide chains (20 per mRNA) (1 per ribosome)

FIG. 3. Transcription and translation of the tiyptophan operon of Escherichia coli under derepressed conditions [18, 20], Gene symbols see Fig. 2. 172 HÜTTER

TABLE I. LEVELS OF TRYPTOPHAN BIOSYNTHETIC ENZYMES UNDER DIFFERENT REGULATORY CONDITIONS [21]

G ene Enzym e Specific activity b De repression ac tiv ity (en zy m e u n its/m g protein) ratio m easured 3 trp R trp R

E ASase 0. 0042 1. 28 305

D PRTase 0. 0038 1. 17 308

С I nG PS ase 0.0135 1. 06 79

В T Sase В 0. 19 1 0.6 56

A T Sase A 0. 23 11.9 52

Strain Tryptophan in growth Molecules of gene R atio medium (Mg/ml) product/cell A/E E A

trp R+ 50 26 157 6 .0

trp R+ 0 624 996 1 .6

trp R" 50 5675 6660 1. 2

a Abbreviations used: ASase, anthî anil ate synthetase; PRTase, anthranilate-phosphoribosyl transferase; ISOase, PRA-isomerase; InG PS ase, indolglycero-phosphate synthetase; TSase B, tryptophan synthetase 0g; TSase A, tryptophan synthetase a; PRA, N-(5'-phosphoribosyl)-anthranilic acid; CDRP, 1 -(o-carboxylphenylamino) -1 -deoxyribulose 5-phosphate; InGP, indole-3-glycerophos- phate. b Cells grown in minimal-glucose medium containing 50 u g /m l L-tryptophan. R+ and R strains carry the mutation trp E-FBR 19, rendering anthranilate synthetase insensitive to feedback inhibition by tryptophan. One unit of each enzyme is defined as that amount of enzyme which catalyses the conversion of 0.1 fimole of substrate to product in 20 min at 37eC.

TABLE II. TRYPTOPHAN MESSENGER RNA (trp-mRNA) LEVELS AND ENZYME SYNTHESIS UNDER DIFFERENT CONDITIONS [23]

Strain and R elative Rela ive R atio condition am ount o f am ount oi enzym es (l)/(2)+(3) trp-m RNA PRTase a T Sase A (3) (2) (3)

"Repressed" (wildtype in 0. 45 0. 22 0 .1 8 a p p r. 2:1 medium containing 80 M g /m l trp)

" N orm al" (wildtype in 1 .0 1. 0 1. 0 1:1 minimal medium)

" Derepressed " ' 3, 7 3. 1 4. 8 appr. 1:1 (trp R" mutant)

a For abbreviations see Table. 1. IAEA-SM-134/31 173

Two processes can be distinguished within the machinery responsible for the synthesis of enzyme constituents: starting from the promotor site Pi (see Fig. 3) multicistronic messenger RNA-molecules are synthesized by D NA-dependent R N A -polym erase enzym es [14, 15]. At 30°C the m e s se n g e r elongation rate is appr. 1200-1500 nucleotides per minute; it therefore takes about 5 minutes to synthesize a complete tryptophan message of appr. 6700 nucleotides length [16,17, 18,19]. A cluster of ribosomes (under derepressed conditions appr. 20 ) follows the polymerase immediately, and polypeptide chains are synthesized. Under the same derepressed conditions a new messenger RNA starts to be synthesized every 20 seconds [18, 2 0 ]. Enzyme synthesis is not coordinate for the whole operon under all conditions. Under conditions of enzyme repression synthesis is only coordinate within the gene group E-D and within the gene group C-B-A.; under conditions of derepression or constitutivity enzymes are synthesized nearly coordinately (Table I, [21]). Furthermore, the ratio between multi­ cistronic tryptophan messenger RNA (trp-mRNA) and tryptophan enzyme content in the cells is not constant (Table II): regulation of enzyme syn­ thesis is translational (in the range normal-repressed) as well as transcrip­ tional (in the range normal-derepressed, but partially also in the range normal-repressed) [22, 23]. A detailed picture of the molecular mechanism for repression of amino- acid biosynthetic enzymes cannot be drawn. The involvement of different components has been demonstrated or postulated: feedback centre [9,24], repressor protein [8,13], the specific tRNA synthetase [25-2 7], the specific tRNAs [28, 29]. But the interaction of the components is not clear. We will therefore leave this subject and focus our attention on the application of some of our knowledge to applied problems.

B. PRODUCTION OF L-GLUTAMIC ACID

Glutamic acid is the first amino acid which has been produced by microbes on an industrial scale [30, 31]. A variety of producer strains have been described, but those used industrially may all be closely re­ lated and belong to the family Corynebacteriaceae [32, 33]. The enzymology and regulation of glutamic-acid production has been studied mainly in the strain No. 2247 designated Brevibacterium flavum and some strains of Corynebacterium glutamicum (resp. Micrococcus glutamicus). Starting from glucose the tricarboxylic-acid cycle inter­ mediate cc-ketoglutaric acid is produced via pyruvic acid and citric acid, «-ketoglutaric acid is transformed to L-glutamic acid by glutamate dehydrogenase (Fig.4). Three steps in the process merit our special interest. First, the producer strains examined do not possess a measur­ able o-ketoglutarate dehydrogenase activity; the o-ketoglutaric acid synthesized cannot be transformed to succinic acid but is available for amination to L-glutamic acid ([30, 34, 35]; Fig. 4(1)). Second, L-glutamate dehydrogenase is inhibited in its "forward" reaction by L-glutamic acid, but only at high product concentrations; in strain No. 2247 50% activity inhibition occurs with appr. IO"1 M L-glutamic acid ([36]; Fig. 4 (2) ). Product inhibition has also been observed in strain MB-1645 [37]. Third, the excretion of large amounts of L-glutamic acid is only possible in 174 HUTTER

Glucose I I

Pyruvate

FIG. 4. Production of L-glutamic acid with Corynebacterium glutamicum. О Reactions critical for L-glutamic acid production, see text.

media with low levels of biotin; under these conditions cell membranes are highly permeable for the product ([39]; Fig. 4 (3) ). The excretion of the product eliminates the danger of further conversions to other metabolites. Partial starvation for biotin is only possible in biotin-auxotrophs (bio") [32, 38, 40]. For completeness, it may be mentioned that cells can also be made permeable for L-glutamic acid by other tricks, e. g. by the addition of penicillin [39]. To summarize, we can recognize three critical points for L-glutamic acid production; no or low cf-ketoglutarate dehydrogenase activity leaves a-ketoglutaric acid available for amination by L-glutamate dehydrogenase; this enzyme is only inhibited by high levels of L-glumatic acid; the use of bicT-strains permits partial starvation for biotin and facilitates excretion of the product.

C. PRODUCTION OF L-LYSINE

The fermentative production of lysine by mutants of Corynebacterium glutamicum is a beautiful example for yield improvement by directed search for specific mutants. L-lysine is produced by the same route in all bacteria hitherto tested. However, the regulatory properties of the biosynthetic enzymes are different in diverse species (see Ref. [ 6]). H ere we will only com pare Escherichia coli and C. glutamicum (see Ref. [ 7]). In E^_ coli the branched biosynthetic sequence of the "aspartate family" of amino acids is characterized by a great number of feedback sites and by the presence of multiple enzymes for several steps (Fig. 5; [5, 6]). To produce a maximal flow along the route aspartic acid -» aspartic-ß- semialdehyde -» lysine a number of regulatory sites would have to be eliminated, e. g. regulation of the three aspartokinases, repression of the lysine biosynthetic enzymes, feedback inhibition of dihydrodipicolinate synthetase. lAEA-SM-134/31 175

Aspartic acid

-I* ! 4 - Aspartokinase III, ll(l ♦ ▼ 1» 13-Aspar ty I phosphate i i Aspart ic - ß- semialdehyde

* "■■X »J '•« * / Homoserine ► ' •./ \ ✓ » / « /Ф * \ A Lysine Methionine Threonine Г t - 7 !

FIG. 5. Regulation of amino acids of the "aspartate family" in Escherichia coli [5, 6], ------unknown or not considered ------lysin e; ...... methionine*, ------threonine; s/ v n / ч isoleucine

I Repression by the corresponding amino acid

J* Induction by the corresponding amino acid

i * Feedback inhibition by the corresponding amino acid

Reversion of feedback inhibition by the corresponding amino acid

Loss o f en z y m e ac tiv ity by m utatio n

Loss of feedback inhibition by mutation

In C. glutamicum the scheme is different: no multiple enzymes, no feedback inhibition of dihydrodipicolinate synthetase and no repression of the lysine biosynthetic enzymes (Fig. 6 ; [41-44] ). It becomes clear that in such an organism the flow of aspartic acid to lysine is only limited at two check points, namely the multivalently regulated aspartokinase and the diversion to homoserine by the single homoserine dehydrogenase. Three ways have been found by which these two blocks can be eliminated by one single mutation:

(1) The use of homoserine auxotrophs (hsr~), in which the homoserine dehydrogenase activity has been eliminated [41, 45-47]. The level of external threonine is critical for lysine production [46]. If threonine is kept sufficiently low, no inhibition of the multivalently regulated aspartokinase occurs and aspartic acid flows uninhibited to aspartic-/3- semialdehyde. Aspartic-ß-semialdehyde cannot be diverted to homo­ serine but is available completely for lysine production. Similarly threonine auxotrophs can be used. 176 HUTTER

Aspartic acid

ß-Aspartyl phosphate i Aspartic-ß-scmialdehyde

Homoserine / / / Ljrane Methionine Threonine (>50g/l) * * + + * Isoleucine

FIG, 6- Lysine production by a homoserine-auxotrophic mutant of Corynebacterium glutamicum [41,45, 46]. О Reactions critical for lysine production, see text; for other signs see Fig. 5.

(2) The use of threonine- (or methionine-) sensitive mutants. The mutants have been isolated as prototrophic revertants from a hsr'-strain and possess only low levels of homoserine dehydrogenase [48]. They are only capable of synthesizing minimal internal levels of threonine. Due to this lack of sufficient threonine the enzyme aspartokinase will not be inhibited, even at high levels of lysine. Furthermore, aspartic-ß- semialdehyde will only be diverted to homoserine with a reduced rate but can mainly be used for lysine synthesis. (3) The use of mutants resistant to the lysine analogue S-(2-aminoethyl)- L-cysteine [49]. It is postulated that the defect lies in a feedback- resistant aspartokinase. In such a strain aspartic acid is freely converted to aspartic-ß-semialdehyde, which is only transformed to homoserine to a limited extent by a feedback-normal homoserine dehydrogenas e.

To summarize, we can learn that the elimination of multivalent regu­ lation in branched pathways and the removal of feedback inhibitions in a specific branch leads to abundant production of the end product. Conditions for a successful application are the knowledge of the regulatory set-up and the possibility to eliminate the critical blocks by mutation.

D. PRODUCTION OF L-TRYPTOPHAN

The industrial production of tryptophan by fermentation has been achieved with yeasts, especially Hansenula anómala [50, 51, 52]. IAEA-SM-134/31 177

Biosynthesis of tryptophan (Fig. 2) in yeast is mainly regulated by feedback inhibition of anthranilate synthetase by tryptophan; enzyme re ­ pression is absent or very weak [53]. The feedback block is bypassed by feeding anthranilic acid or indole as precursors. But an inhibitory effect of high levels of anthranilic acid has been observed [52]. In order to allow effective feeding by anthranilic acid it is necessary to use a mutant resistant to high levels of anthranilic acid [50]. Still, a combined feeding of anthranilic acid and indole is more bénéficiai than either precursor alone [51]. The avoidance of the single critical feedback block by feeding precur­ sors posterior to this block and the use of a mutant resistant to high levels of this precursor supplies satisfactory production conditions.

DISCUSSION AND SUMMARY

In the efforts to elucidate the molecular mechanisms of regulation of amino-acid biosynthetic enzymes the use of mutant strains has been an indispensable condition. The analysis of such mutants, in comparison with the parental wild-type strain, has widened our knowledge on the processes of enzyme synthesis and on their regulation. Similarly, the use of mutants is an indispensable instrument for the industrial microbiologist to increase the yields of his products. He has to make use of this asset by applying the findings of molecular genetics, especially in three areas: regulation studies, mutation techniques and recombination procedures. It has to be admitted that, in general, a considerable lag and gap exist between the advances in molecular and cell biology and their application in industrial microbiology. But for specific problems, that gap can be closed. Without doubt the use of radiation and radioisotopes has played, and will play, an important role besides chemical mutagenesis.

REFERENC ES

[1] VOGEL, H .J., "Lysine biosynthesis and evolution", Evolving Genes and Proteins (BRYSON, V ., VOGEL, H .J., Eds), Academic Press, New York (1965). [2] UDAKA, S., J. Bacteriol. 91 (1966) 617. [3] UMBARGER, E., DAVIS, B. D. , "Pathways of amino acid biosynthesis", Ch. IV in Vol. Ill, The Bacteria (GUNSALUS, I.C ., STANIER, R.Y., Eds), Academic Press, New York (1962). [4] GIBSON, F., PITTARD, J., Bact. Rev. 32(1968) 465. [5] STADTMAN, E. K. , Bact. Rev. 27(1963) 170. [63 COHEN, G ., "Le Métabolisme Cellulaire et sa Régulation", Hermann, Paris (1967). [7] HÜTTER, R., Path. Microbiol. 34 (1969) 195. [8] COHEN, G ., JACOB, F., Compt. rend. Acad. Sei. Paris 248 ( 1959) 3490. [9] SOMMERVILLE, R. L., YANOFSKY, C ., J. molec. Biol. 21(1965) 747, [10] MATSUSHIRO, A., KIDA, S., ITO, J ., SATO,, K., IMAMOTO, F., Biochem. biophys. Res. Comm. 9 (1962) 204. [11] ITO, J., CRAWFORD, I. P., Genetics 52 (1965) 1303. [12] HIRAGA, S., J. molec. Biol. 39 (1969) 159. [13] MORSE, D.E., YANOFSKY, C. , J. molec. Biol. 44 (1969) 185. [14] IMAMOTO, F., YANOFSKY, C ., J. molec. Biol. 28 (1967) 1. [15] IMAMOTO, F., MORIKAWA, N.. SATO, K., J. molec. Biol. _13 (1965) 169. [16] IMAMOTO, F., Proc. natn. Acad. Sei. USA 60 (1968) 305. 178 HÜTTER

[17] BAKER, R. F., YANOFSKY, C., Proc. natn. Acad. Sei. USA 60 (1968) 313. [1 8 ] MORSE, D .E ., BAKER, R. F . , YANOFSKY, C ., Proc. natn. A cad. Sei. USA 60 (1968) 1428. [19] ROSE, J.K ., MOSTELLER, R. D ., YANOFSKY, C ., J. molec. Biol. 51 (1970) 541. [20] Correspondent, "RNA at Cold Spring Harbour", Nature 226 ( 1970) 1093. [21] MORSE, D.E., YANOFSKY, C., J. molec, Biol. 38(5968) 447. [22] STUBBS, J.D ., HALL, B. D ., J. molec. Biol. 37 (1968) 289, [2 3 ] LA VALLE, R ., DEHAUWER, G ., J. m olec. Biol. 51 (1970) 435. [2 4 ] KOVACH, J . S . , BERBERICH, M. A ., VENETIANER, P ., GOLDBERGER, R. T . , J. B acteriol. 97 (1969) 1283. [25] DOOLITTLE, W. F., YANOFSKY, C ,, J, Bacteriol. 95 (1963) 1283. [26] ITO, K., HIRAGA, S., YURA, T ., Genetics 61 (3969) 521. [27] DELORENZO, F., AMES, В. N., I. Biol. Chem. 245 (1970) 1710. [28] WONG, J.T ., MUSTARD, M .H., Biochim. biophys. Acta 174 (3 969) 513. [2 9 ] ROTH, J. R ., SILBERT, D. F ., FINK, G .R ., VOLL, M .J ., ANTON, D ., HARTMAN, P .E ., AMES, B. N. , Cold Spring Harbour Symp. Quant, Biol. 31 (1966) 383. [30] HUANG, H. T ., "Microbiol production of amino acids", Progr. Industr. Micorbiol, 5 (HOCKENHULL, D. J.D ., Ed.), Heywood, London ( 3964) 55, [31] Monosodium Glutamate and Glutamic Acid, Chem. Progr. Rev. No. 25 (POWELL, R., Ed.), Noyes Dev, Corp., Park Ridge N.J. (1968). [32] VELDKAMP, H ,, VAN DEN BERG, G., ZEVENHUIZEN. L. P. T. M ., Ant. Leeuwenhoek 29 (1963) 35. [33] KINOSHITA, S., Adv. appl. Microbiol. 1(1959) 201. [34] TANAKA, K., AKITA, S., KIMURA, K., KINOSHITA, S., Amino Acids 1 ( 1959) 62. [35] SHIIO, I., OTSUKA, S., TAKAHASHI, M., J, Biochem. 50 (1961) 164. [36] SHIIO, I., OZAKI, H., J. Biochem. 68 (1970) 633. [3 7 ] NUNHEIMER, T .D . , BIRNBAUM, J. , IHNEN, E. D. , DEMAIN, A. L ., A ppl. M icrobiol. 20 (1970) 215. [38] OISHI, K., AIDA, K., Agr. Biol. Chem. 29 (1965) 83. [39] DEMAIN, A. L., BIRNBAUM, J ., Current Topics Microbiol. Immunol. 46 (1968) 1. [40] KIMURA, K., J. gen. appl. Microbiol. i)(1963) 205. [41] NAKAYAMA, K ,, KITADA, S., KINOSHITA, S., J. gen. appl. Microbiol. 7 (1961) 145. [42] NAKAYAMA, K., TANAKA, K., OGINO, H., KINOSHITA, S., Agr. Biol. Chem. 30 (1966) 611. [43] MIYAJIMA, R., SHIIO, I., Agr. Biol. Chem. 34 (1970) 1275. [44] MIYAJIMA, R., SHIIO, I., J. Biochem. 68 (1970) 311. [45] MIYAJIMA, R., OTSUKA, S.I., SHIIO, I., J. Biochem, 63 (1968) 139. [46] DAOUST, D.R., STOUDT, Т.Н ., Dev. industr. Microbiol. 7 (1966) 22. [47] SANO, K., SHIIO, I., J. gen. appl. Microbiol. 23 (1967) 349. [48] SHIIO, I., SANO, K., J. gen. appl. Microbiol. _J£(1969) 267, [49] SANO, К,, SHIIO, I,, J, gen. appl. Microbiol. 16 (1970) 373, [50] TERUI, G ,, NIIZU, H., Biotechn. Bioeng. Symp. Nr. _1 (1969) 33. [53] EBIHARA, Y., NIITSU, H., TERUI, G., J. Ferment. Technol. £7(1969) 733. [5 2 ] RUBAN, E .L ., LOBYREVA, L. B ., A ppl. Biochem . M icrobiol. USSR J .( 1965) 74. [53] Unpublished results.

DISC USSION

HUGUETTE de ROBICHON-SZULMAJSTER: You have mentioned that biotin deficiency affects the permeability barrier towards glutamic acid. Does it favourably affect the excretion of other amino acids too? R. HÜTTER: Biotin deficiency, as such, does not lead to the maximum excretion of all amino acids. The same strain which may preferentially excrete glutamic acid under conditions of extreme biotin deficiency may preferentially excrete lysine if more biotin is supplied to the growth medium. VASSILIKI VOMVOYANNI: Regarding the tryptophan operon of Escherichia coli, I should like to ask, first, whether the second prompter site (P 2 ) has been genetically identified and, second, whether IAEA-SM-134/31 179 the differences in the co-ordination of enzyme synthesis under repressed or derepressed and "normal" conditions have any physiological significance. R. HÜTTER: The presence of a second promoter was first established in Salmonella typhimurium on the basis of studies on deletion mutants. So far as I know, no mutants at the P 2 site are known. We must assume that the type of control maintained under different growth conditions does have a physiological significance for E. coli. But we can only speculate on this point; the reason may be to save the energy needed for enzyme synthesis. H. ALTMANN: Do you think that RNAase and RNAase-inhibitors are the limiting and regulatory factors in translation? R. HÜTTER: RNAase and RNAase-inhibitors may well contribute to translational control of enzyme synthesis, as is indicated by the role of RNAase in polarity and by the RNAase activity of ribosomes. But, so far as I am aware, no conclusive picture can as yet be drawn. RNAase and RNAase-inhibitors are certainly not the only regulatory factors in trans­ lation. The initiator codons, too, may have the same function. 5.1. ALIKHANIAN: You have mentioned a lysine production of 0 > 50 mg/ml. Did you yourself obtain these results? If so, what was the carbohydrate content in the growth medium in which these lysine synthesis levels were obtained? R. HÜTTER: The data presented are taken from the literature, especially from the publications of Nakayama et al. and Shiio et al. (see Refs [41, 35, 36, 48] in my paper). 5.1. ALIKHANIAN: It is a pity that you did not dwell upon tryptophan at greater length. I should like to know what tryptophan synthesis levels you obtained and in what growth medium (with or without a precursor). R. HÜTTER: To elucidate the basic mechanisms of regulation of enzyme synthesis in eukaryotes we are studying the tryptophan system in Saccharomyces cerevisiae. This organism is very suitable for basic research but not for commercial production of tryptophan. The condi­ tions used for industrial production of tryptophan by Hansenula anómala have been described by Terui and Ebihara (see Refs [50, 51] of the paper).

IAEA-SM-134/26

REGULATION OF AMINO-ACID BIOSYNTHESIS IN Saccharomyces cerevisiae

HUGUETTE de ROBICHON-SZULMAJSTER Laboratoire d'enzymologie, Centre national de la recherche scientifique, Gif-sur-Yvette, France

Abstract

REGULATION OF AMINO-ACID BIOSYNTHESIS IN Saccharomyces cerevisiae. The regulatory mechanisms known to be operative in amino-acid biosynthesis in yeast are briefly reviewed. Emphasis is given to the analysis of the regulatory pattern which controls threonine, methionine and isoleucine-valine biosyntheses in Saccharomyces cerevisiae. The implication of these findings upon selection of amino-acid over-producers in yeasts are discussed.

INTRODUCTION

The biosynthesis of amino acids in yeasts, including some of the regulatory aspects, has been recently reviewed [1]. In most cases the biosynthetic routes are essentially identical to their counterparts in bac­ teria. There are a few exceptions such as lysine biosynthesis, which is entirely distinct between bacteria and yeasts [2 ], and methionine bio­ synthesis, parts of which are the same and parts of which differ radically in the two types of microorganisms (see Ref. [1] for review). It follows that for such amino acids unique regulatory patterns have to be expected. In fact, this assumption, which is discussed later, has already been verified in methionine biosynthesis. Another more general reason for the existence of distinct regulatory patterns in yeasts, as compared with bacteria, resides in the absence of operon-like organization of structural genes coding for the successive steps in a given biosynthetic pathway (see Ref. [1] for review). It follows that a great diversity in regulatory means is observed within each pathway, the combination of which finally results in a fine ajustment, as efficient and perhaps more versatile, than that observed in bacteria, especially in the case of branched pathways. It seems then a prerequisite that a good biochemical and genetical understanding of these regulatory patterns should precede any rational approach for selecting mutants. This knowledge will form the basis for selecting those biochemical mutants in which such intricate regulation patterns are disorganized with the consequent over-production of amino acids. This paper classifies the regulatory tools that have evolved as a result of natural selection in yeast.

A. GENE ENZYME RELATIONSHIP: UNIFUNCTIONA.L, MULTI­ FUNCTIONAL AND ISOFUNCTIONA.L ENZYMES

In most cases there is no linkage between structural genes encoding for the numerous enzymes involved in amin-acid pathways [3]. For example, structural genes for histidine biosynthesis are dispersed

181 182 ROBICHON-SZULMAJSTER over at least seven chromosomes or fragments. Even when mapping is not provided for all the genes involved, it has been possible to show that they segregate independently. This is true for methionine synthesis which requires the participation of some 13 genes, all of which are unlinked. However, there are exceptions to this rule. These exceptions seem, mainly, to serve the purpose of channelling intermediates and, secondarily, to provide a means of regulating activities of two enzymes together. Although not directly concerned with amino-acid biosynthesis, one very characteristic situation of metabolic regulation worthy of mention is represented by the multi-enzyme complexes [4, 5]. In Saccharomyces cerevisae the first two steps in the pyrimidine biosynthesis are catalysed by a complex molecule which contains a carbamyl phosphate synthetase, an aspartate transcarbamylase and a regulatory unit. These three entities are synthesized as a multifunctional complex which permits: ( 1 ) a very efficient channelling at the molecular level of an intermediary product (carbamyl-phosphate) which is also synthesized by an independently coded carbamyl-phosphate synthetase involved in arginine biosynthesis; and (2) a very efficient end-product inhibition exerted by UTP on both activities at once. Another multienzyme complex comprises three enzyme catalysing steps (2, 3 and 10) involved in histidine biosynthesis [ 6]. All three proteins seem to be encoded by a single gene, his4, which contains three discrete regions A, В and C. In this case there does not seem to be any regulatory purpose for the association. The reason for this remains unclear. It seems, however, unlikely that such an association is only incidental since this phenomenon has also been observed in S. lactis [7] and Neurospora c ra s s a [ 8]. It was shown for the first time in Escherischia coli that there exist two distinct aspartokinases which catalyse the first step in the complex pathway that leads, in this organism, to threonine, methionine and lysine biosynthesis [9]. Although the two enzymes catalyse the same reaction they can be distinguished by their regulatory properties, one being feed­ back inhibited by threonine and the other feedback inhibited by lysine. A third enzyme, whose synthesis is under methionine control, has since been characterized (see Ref. [ 10] for review). However, the existence of such isofunctional enzymes should not be generalized. For example, in S. cerevisiae only one aspartokinase exists [9,11]. Thus, in this organism, the regulatory function that the isofunctional enzymes are obviously serving in E. coli has to be replaced by other regulatory means in S. cerevisiae. On the other hand, isofunctional enzymes have also been described in S. cerevisiae. There are the two carbamyl phosphate synthetases already mentioned — one being inhibited and repressed by UTP [4, 5], the other repressed by arginine [4], and the two DAHP aldolases catalysing the first step in aromatic amino-acid biosynthesis, one being feedback inhibited by phenylalanine and the other being feedback inhibited by tyrosine [12-14]. However, these examples of multifunctional or isofunctional enzymes, although constituting selective materials for the study of a special group of enzymes, are only few. Nevertheless most of the reactions, which are subject to all sorts of regulatory effects (repression, induction, feedback inhibition or activation, etc. ), are commonly catalysed by single enzymes IAEA-SM-134/26 183

-►S

A c H1 S - W H C -- H M E T l

lASPh— ► AspP+y-* ASA ^-^H Sr HSP—--»Ft h RI-— УКВ^-*АНВ—KB p^DHMV p*KIL— »I IL |

Pyr*/Т

>|v a l |

I ‘ 1 I 1 1 iso4

ÍI hom3 1 hom2 1 hom6 1 , th rl thr4 isol iso2 iso5

{ frag .2 FIG. 1. Biosynthesis of threonine, methionine, isoleucine and valine in Saccharomyces cerevisiae. ASP: aspartate; AspP: ß-aspartyl-phosphate; ASA: aspartic semialdehyde; HS: homoserine; HSP: homoserine phosphate; THR: threonine; AcHS: O-acetyl homoserine; HC: homocysteine; MET: methionine; APS: adenosine phosphosulphate; PAPS: phosphoadenosine phosphosulphate; KB: a-ketobutyrate; AHB: acetohydro- butyrate; DHMV: Dihydroxymethylvalerate; KIL: ketoisoleucine; IL: isoleucine; Pyr: pyruvate; AL: acetolactate; DHIV: dihydroxyisovalerate; KV: ketovaline; VAL: valine. Gene nomenclature Mutated alleles lead to the following requirements: hom3, 2, 6: homoserine; met2, 8 : methionine; thrl, 4 : threonine; isol : isoleucine; iso2, 3, 4, 5 : isoleucine + valine.

encoded by single and unlinked structural genes. This situation seems to be true, for example, for threonine and methionine biosynthesis and for isoleucine and valine biosynthesis in S. cerevisiae (see Fig. 1). In these two pathways, mutants for each of the loci are missing only one enzymatic activity and there is no evidence for more than one enzyme catalysing each of the steps [11, 15]. Since, in such pathways, intermediary compounds do not accumulate and end-products are not over-produced in normal conditions [16], regulatory devices other than operon organization, multifunctional or isofunctional protein must be operative. Our purpose in studying the regulation of these two pathways was to uncover the molecular nature of such regulatory devices.

B. MULTIPLICITY AND DIVERSITY IN REGULATORY DEVICES

In the two pathways considered here, each of the enzymes that has been studied so far (10 of the 13 steps in threoninine and methionine bio­ synthesis, 4 of the 5 steps involved in isoleucine and valine biosynthesis) has been found subject to some regulation of its synthesis. In addition, at least half of these enzymes have been found subject to regulation of their activity. 184 ROBICHON-SZULMAJSTER

I. Monovalent repression-, pleiotropic effects

The synthesis of at least five enzymes involved in methionine bio­ synthesis is subject to methionine-mediated repression and is unaffected by threonine. One of these, homoserine dehydrogenase, also takes part in threonine biosynthesis. The other four enzymes, homoserine-O- transacetylase, homocysteine synthetase, ATP sulphurylase and sulphite reductase are only concerned with methionine biosynthesis? On the basis of regulation studies, they constitute a regulatory group (methionine Group I enzymes) which is characterized as follows [19,20]:

1. Their rate of synthesis is subject to a very large range of variation (as much as 100 fold between the higher degree of repression and the higher degree of derepression). Such a susceptibility to end-product repression is rather unusual in yeast in which less than a 1 0 -fold effect is most common. 2. Their repressibility is affected, to the same extent, by conditions which lead to an important lack of charging of tRNAmet (experiments using a thermosensitive mutant carrying a modified methionyl-tRNA synthetase [21]). These results lead to the hypothesis that methionyl-tRNA. is some­ how involved in the regulatory process. 3. Their repressibility is also affected, to a comparable extent, by a regulatory mutation, eth2 , unlinked to the structural genes it governs. 4. Their synthesis is co-ordinate.

All these properties show that the Group I enzymes constitute, despite the absence of linkage between structural genes, a single regulatory unit "regulon", responding to a single pleiotropic repressor. In addition, the absence of response of homoserine dehydrogenase synthesis to this regulatory system leads to the postulation of the existence of at least one other methionine repressor in yeast.

II. M ultivalent re p re ssio n

Multivalent repression, which is exerted jointly by the end-products of a given pathway, was first observed in E. coli for isoleucine, valine and leucine biosyntheses [22]. An almost identical repression has also been found in yeast [23,24]. At least isoleucyl-tRNA seems to be involved in the regulatory process [25]. Since, as previously noticed, there is only one aspartokinase in yeast and this enzyme is involved in the synthesis of both threonine and m e­ thionine, it was not surprising to find that both amino acids participate in repression of its synthesis [11,19, 20]. There are also indications in favour of threonyl-tRNA. involvement [21]. Since aspartokinase synthesis does not seem to belong to the regulatory Group I described above, it

1 By studying different mutants we have shown that methionine biosynthesis seems to be unique in yeast as compared with other organisms since the biological intermediate, homocysteine, is synthesized directly from O-acetyl homoserine and H2S. Consequently, the sulphate assimilation pathway belongs to methionine biosynthesis and is only secondarily involved in cysteine biosynthesis — this amino acid being easily formed from methionine through cystathionine. These conclusions are further confirmed by the regulatory properties of enzymes involved in sulphide biosynthesis [ 17,18]. IAEA-SM-134/26 185 is unlikely that methionyl-tRNA would be involved. It follows that in multivalent repression not all of the end-products need to be acylated prior to its participation in the regulatory process. Such a scheme has recently been proposed from quite different arguments for the multivalent repressive effects exerted by isoleucine, leucine and valine in bacteria [27].

III. Antagonistic repression: end-product induction

The first term designates a phenomenon which has been observed for two enzymes in threonine and methionine biosynthesis,namely aspartic semi-aldehyde dehydrogenase and homoserine kinase, the syntheses of which are repressed by threonine and "induced" by methionine [11]. On the other hand, chorismate mutase and prephenate dehydrase, which are both subject to feedback inhibition exerted by tyrosine, are "induced" by phenylalanine [28]. The antagonistic effects, as well as the opposite effects, of isoleucine and valine upon the activity of yeast threonine de­ aminase [29-31] (properties of which have been observed in all organisms in which this last enzyme has been studied) seem to testify for the biological need to keep the production of some amino acids in a rather precise balance. Disturbances of this balance by absolute (primitive) regulatory mechanisms would create the necessity for more refined (corrective) regulatory mechanisms such as the antagonistic repression briefly mentioned here.

IV. Molecular mechanisms underlying repression

According to the Jacob-Monod model,' repression was shown to occur at the transcriptional level in bacteria. The repressor, a protein molecule [32] with a high affinity for a DNA. [33] containing the operator, can block transcription of the whole operon. However, there has been recent supporting evidence, also in bacteria, for the existence of a second repression mechanism which occurs at the translational level. Moreover, it is expected that the transcriptional repression mechanism would apply mainly to variation between normal and derepressed levels of enzymes, whereas the translational repression mechanism would apply mainly to variation between normal and repressed levels of enzymes [34, 35]. In S. cerevisiae, the normal (growth of wild type in minimal medium) levels of enzymes involved in methionine biosynthesis are more or less equally distinct from repressed and derepressed levels, leaving the possibility to measure repression as well as derepression. As far as methionine biosynthesis is concerned (and this applies to all regulatory mutations in amino acid pathways in yeast) the regulatory mutations so far studied, have no effect on spontaneous (normal) level of enzymes in the wild type, or on the rates of synthesis between normal and derepressed levels, even though they interfere with repressibility, i. e. with the rate of synthesis between normal and repressed levels. It is therefore tempting to suggest that the regulatory genes so far uncovered would be mainly concerned with the mechanism of repression at the translational level [19, 20]. This assumption supports the finding that this type of 186 ROBICHON-SZULMAJSTER

regulatory mutant does not, or only slightly, over-produces the amino acid concerned. On the contrary, mutations which would affect the tran­ scriptional mechanism would be expected to increase the rate of enzyme synthesis to the derepressed levels and should then be over-producers. If this reasoning is correct such mutants remain to be found. It would not be surprising to find that over-production would need impairment of both transcriptional and translational repression mechanisms.

Feedback inhibition or activation

Regulation of enzyme activity is very efficient in amino-acid pathways in yeast. However, perhaps as a result of the absence of operon organization, more than one step is generally subject to feedback inhibition. For example, in the threonine and methionine biosynthetic pathways the following obser­ vations have been made:

1. Common part of the pathway

The first step, catalysed by aspartokinas e, is subject to allosteric feedback inhibition exerted by threonine [11, 36]. The second step, catalysed by aspartic semi-aldehyde dehydrogenase, is activated by the bicarbonate anion [37]. The third step, catalysed by homoserine dehydrogenase, is feedback inhibited by methionine [38].

2. Threonine-specific enzymes

The first step, catalysed by homoserine kinase, is feedback inhibited by threonine [11, 15, 39].

3. Methionine-specific enzymes

The second step, catalysed by homocysteine synthetase, is subject to feedback inhibition exerted by methionine [17, 40]. The first step in the sulphate assimilation pathway, ATP sulphurylase, is feedback inhibited by the end-product of this branch, the sulphide anion [ 17, 41].

CONCLUSIONS

The regulatory pattern shown in Fig. 2 accounts for all the observations made so far on threonine and methionine biosynthesis in S. cerevisiae and can be summarized as follows:

1. Multiplicity of regulatory points. 2. Diversity in response of successive enzymes. 3. The presence of more than one repressor involving a given amino acid and acting in a given pathway. 4. Probable existence of two kinds of regulatory genes (transcriptional and translational). IAEA-SM-134/26 187

Oxaloacetate

FIG. 2. Regulatory pattern of threonine and methionine biosynthesis in Saccharomyces cerevisiae. kW '.w 1 Threonine mediated repression m i I it I Methionine mediated repression Methionine mediated "induction" ------Inhibition + -f + •+ A ctiv atio n

5. Complementarity of effects due to both end-products. 6 . Complementarity of repressive and inhibitory effects, but also 7. Corrective effects due to antagonistic repression or to inhibition and activation.

The complexity of the interactions between these regulatory devices, which are used by the same organism within one pathway that only leads to two end-products, makes it unlikely that impairment at a single point (whether by mutation or by interfering with one enzyme activity) would considerably disturb the overall regulatory pattern. Probably, in most cases, successive barriers will have to be removed in order to obtain yeast strains that are suitable for the production of amino acids. This goal will be attained through satisfactory knowledge of the regulatory patterns, through mutagenesis and selection directed towards the most critical points of these patterns and, most probably, the accumulation of different regulatory mutations concerned with a given pathway by recombination. REFERENCES

[1] de ROBICHON-SZULMAJSTER, H., SURDIN-KERJAN, Y ., The Yeasts, Vol. 2 (ROSE, A. H., HARRISON, J.S ., Eds), Academic Press (in press). [2] BROQUIST, H .P., TRUPIN, J.S., A. Rev. Biochem. 35 (1966) 231. [3] HAWTHORNE, D. C ., MORTIMER, R.K., The Yeasts, Vol. 1 (ROSE, A. H., HARRISON, J.S ., Eds), Academic Press (in press). [4] LACROUTE, F., PIERARD, A ., GRENSON, M ., WIAME, J. M ., J. gen. Microbiol. 40 (1965) 127. 188 ROBICHON-SZULMAJSTER

[5] LUE, P.F., KAPLAN, J.G ., Biochim. biophys. Acta 220 (1970) 365. [6] FINK, G.R., Genetics, Princeton 53 (1966) 445. [7] TINGLE, M ., HERMAN, A ., HALVORSON, H .P., Genetics, Princeton 58 (1968) 361. [8] AHMED, A ., Molec. Gen. Genetics 103 (1968) 185. [9] STADTMAN, E.R., COHEN, G. N., LeBRAS, G., de ROBICHON-SZULMAJSTER, H., J. biol. Chem. 236 (1961) 2033. [10] TRUFFA-BACCHI, P., COHEN, G. N., Am. Rev. Biochem. 32(1968) 662. [11] de ROBICHON-SZULMAJSTER, H ., SURDIN-KERJAN, Y ., CHEREST, H ., Genetics of Industrial Microorganisms (VANEK, Z., HOSTÁLEK, Z ., CUDLIN, J ., Eds), Czech. Acad. Sc. Publ. (in press). [1 2 ] LINGENS, F ., GOEBEL, W ., UESSELER, H ., B iochem . Z. 346 (1966) 357. [1 3 ] LINGENS. F ., SPROSSLER, B ., GOEBEL, W ., Biochem . biophys. A cta 121 (1966) 164. [ 14] MEURIS, P ., LACROUTE, F ., SLONIMSKI, P., Genetics, Princeton 56 ( 1967) 149. [15] de ROBICHON-SZULMAJSTER, H., Bull. Soc. Chim. biol. 49 (1967) 1431. [16] BOURGEOIS, C ., Bull. Soc. Chim. biol. 51 (1969) 935. [17] CHEREST, H., EICHLER, F., de ROBICHON-SZULMAJSTER. H., J. Bacteriol. 97 (1969) 336. [18] CHEREST, H., TALBOT, G ., de ROBICHON-SZULMAJSTER, H., J. Bacteriol. 102 (1970) 448. [19] CHEREST, H ., de ROBICHON-SZULMAJSTER, H ., Genetics of Industrial Microorganisms (VANEK, Z., HOSTÁLEK, Z., CUDLIN, J., Eds), Czech. Acad. Sc. Publ. (in press), [20] CHEREST, H ., SURDIN-KERJAN, Y .. de ROBICHON-SZULMAJSTER, H ., J. Bacteriol. (in press). [21] McLAUGHLIN, C. S ., HARTWELL, L. H ., Genetics, Princeton 61 (1969) 557. [2 2 ] FREUNDLICH, M ., BURNS, R. О ., UMBARGER, H .E ., Proc. natn. A cad. Sc. U .S .A . 48 (1962) 1804. [2 3 ] BUSSEY, H ., UMBARGER, H. E ., J. B acteriol. 98 (1969) 623. [24] MAGEE, P.T ., HEREFORD, L.M ., J. Bacteriol. 98 (1969) 857. [25] McLAUGHLIN, C .S., MAGEE. P. T ., HARTWELL, L.H., J. Bacteriol. 100 (1969) 579. [26] NASS, G., HASENBANK, R., Molec. Gen. Genetics 108 (1970) 28. [27] HATFIELD, G. N., BURNS, R. О ., Proc. natn. Acad. Sc. U.S.A. 66 (1970) 1027. [2 8 ] LINGENS, F ., GOEBEL, W ., UESSELER, H ., Eur. J. B iochem . ¿ (1 9 6 7 ) 363. [29] CENNAMO, C., BOLL, M ., HOLZER, H., Biochem. Z. 340 (1964) 125. [30] HOLZER, H., CENNAMO, C., BOLL, M., Biochem. biophys. Res. Comm. 14 (1964) 487. [31] de ROBICHON-SZULMAJSTER, H., MAGEE, P. T ., Europ. J. Biochem. 3 (1968) 492. [32] GILBERT, W., MULLER-HILL, B., Proc. natn. Acad. Sc. U.S.A. 56 (1966) 1891. [33] RIGGS, A.D., BOURGEOIS, S., NEWBY, F., COHN, M., J. molec. Biol. 34 (1968) 365. [34] LA VALLE, R.. DEHAUWER, G., J. molec. Biol. 51 (1970) 435. [35] McLELLAN, W .L., VOGEL, H.J., Proc. natn. Acad. Sc. U.S.A. 67 (1970) 1703. [36] de ROBICHON-SZULMAJSTER, H ., CORRIVAUX, D., Biochim. biophys. Acta 73 (1963) 248. [37] SURDIN, Y., Europ. J. Biochem. 2(1967) 341. [38] KARASSEVITCH, Y ., de ROBICHON-SZULMAJSTER, H ., Biochim. biophys. Acta 73 ( 1963) 414, [39] WORMSER, E.H ., PARDEE, A.B., Archs Biochem. Biophys. 78 (1958) 416. [40] W1EBERS, J.L ., GARNER. H.R., J. biol. Chem. 242(1967) 5644, [41] de VITO, P.C., DREYFUSS, J., J. Bacteriol. 88 (1964) 1341.

DISCUSSION

S. G. GEORGOPOULOS: I wonder whether, in the mutant you have mentioned with reduced methionine-induced repression of three different enzymes, the possibility of reduced amino-acid uptake was eliminated. HUGUETTE de ROBICHON-SZULMAJSTER: Mutants with reduced methionine uptake have also been obtained. However, the strains on which we are carrying out regulatory studies show no change at all in any amino-acid uptake system. ESTHER BALOGH: Could you please tellus what influence chloramphenicol and puromycin have on methionine uptake? Both these compounds have influence in the case of Saccharomyces carlsbergensis. HUGUETTE de ROBICHON-SZULMAJSTER: I have never investigated the influence of these compounds on amino-acid uptake in Saccharomyces cerevisiae. I was unaware of such influence. IAEA-SM-134/35

SPECULATIONS ON GENETIC LOCI CONTROLLING THE BIOSYNTHESIS OF TETRACYCLINES*

Z. HOSTALEK, M. BLUMAUEROVA, J. CUDLÍN, Z . VANEK Institute of Microbiology, Czechoslovak Academy of Sciences, Prague, Czechoslovak Socialist Republic

Abstract

SPECULATIONS ON GENETIC LOCI CONTROLLING THE BIOSYNTHESIS OF TETRACYCLINES. The paper presents a theoretical consideration on the genetic determinants of the biosynthetic pathway of chlortetracycline. After condensation of the malonate units to form the hypothetical intermediate, 11 bio­ synthetic steps are postulated which result in the final product. Each step is catalyzed by the corresponding specific enzymes. These views are confirmed by the results of genetic work where blocked mutants were used to estimate the number and sequence of genetic loci controlling the formation of tetracycline antibiotics.

Past meetings attended by geneticists investigating industrial micro­ organisms and by research workers studying the biosynthesis of secondary metabolites [1 , 2 ] have revealed the gaps that exist in our knowledge of the genetic control of the production of natural compounds. We understand fairly well the biosynthetic steps, the enzyme reactions and the genetic control of the biosynthesis of primary metabolites, e. g. amino acids. This enables a rational approach to the problem of obtaining high-production strains and of developing technologies for the production of a desired metabolite. The situation is much more complex in the biosynthesis of secondary metabolites, such as antibiotics, pigments, alkaloids, etc. This is also true for compounds of the tetracycline series, produced by Streptomyces aureofaciens and Streptom.yces rimosus. In a previous paper [3] we tried to evolve a theoretical concept of the possible pathways of the biosynthesis of chlortetracycline. We proceeded from the present knowledge of the chemistry of tetracycline antibiotics [4-6]. The considerations resulted in the realization that some 200 enzymes partici­ pate directly or indirectly in the pathway from glucose to chlortetracycline. On the simplified assumption of ' one gene one enzyme' it could be postulated that som e 200 genes participate in the control of the synthesis of tetracycline com pounds. The present paper considers only the final part of the whole biosynthetic pathway, namely the enzyme reactions taking place at the level of the formed tetracycline ring. Chlortetracycline belongs, biogenetically, to the oligoketides. It is formed through condensation of eight molecules of malonyl-CoA and one molecule of malonamyl-CoA. The mechanism of formation of malonyl-CoA

"r This work was supported by the International Atomic Energy Agency under research contract N0.845/RB.

189 190 HOSTÁLEK et al.

COS- F pyruvate -AcSCoA *с°г» MaSCoA CH Ксоон

glutamate *

2-oxoglutorate-

CO-COOH COSCoA c o s -f asparagine - СНг c h : \ c h 2- c o n h 2 c o n h , CONHo

c h 2- c o s - f cos-e

c o - c h 2- c o n h 2 V n ^ V c o n ^ CONH, OOOO OOÖÖ

FIG .l. Postulated biosynthetic origin of the tricyclic nonaketide.

M e - C 6 A - r i n g ® Cyclizotion

FIG.2. Biosynthesis of tetracycline-type compounds. Transformation of hypothetical tricyclic nonaketide.

as a biosynthetic intermediate of tetracyclines is still a matter of discussion [7]. The postulated pathway of malonamyl-CoA formation is given in Fig. 1. This is the first reaction which is characteristic for the formation of the tetracene nucleus and which does not occur in other metabolites (except the glutarimide ring of cycloheximide and related antibiotics [8 , 9]). C ondensa­ tion of malonamyl-CoA with other molecules of malonate probably takes place on a protein matrix. The hypothetical nonaketide is then cyclized and aromatized. IAEA-SM-134/35 191 © OH-C4 -2H H20-C4o, i2<1 Ci

©

NHa 2x Me O H -C 6 2H

4 -oxodehydrocMortefro- 4-oxochlortefracycline cycline

FIG.4. Biosynthesis of tetracycline-type compounds; chlortetracycline series.

The enzyme, or rather the enzyme complex, catalyzing the reaction ("anthracene synthase") represents the first step in the biosynthetic pathway, assumed to be under the control of one of the specific tetracycline loci. In contrast to the preceding steps, the reaction is relatively species-specific and characteristic for the producers of tetracyclines which is typical of the biosynthetic processes of secondary metabolism. 192 HOSTÁLEK et al.

4 -oxodehydrotetrocych'ne 4 -oxotetracycline

FIG. 5. Biosynthesis of tetracycline-type compounds; tetracycline series.

@ NH2 2 x Me 0 H -C 6

Cl Me NH? Cl Me NMeo Cl Mp пн NMeo

OH 0 OH 0 chlortetramid - green

FIG.6. Biosynthesis of tetracycline-type compounds; chlortetramid compounds.

Figure 2 shows two possibilities for the transformation of the hypotheti­ cal tricyclic derivative formed from the nonaketide by triple dehydration. In the initial stage, méthylation proceeds in position 6 (branch BI and B2) followed by cyclization of ring A (branch BI and B3) and by the removal of the oxo group at Ce­ lt follows from Fig. 3 that in the next step the biosynthetic pathway of tetracyclines divides into two further branches. The decisive step is a hydroxylation of ring A in position 4, followed by oxidation, hydration at C4a and C j2a anc^ finally chlorination in position 7. In the next step the tetra­ cycline derivatives branch into two further routes (Fig. 4). On replacing IAEA-SM-134/35 193

(И)

nh2 2 x Me OH-Ce

ОН О OH OH oureovocidin

FIG. 8. Biosynthesis of tetracycline-type compounds; aureovocidin compounds.

the oxygen of ring A with an amino group, double N-méthylation takes place and anhydrochlortetracycline is formed. The last biosynthetic steps are hydroxylation in position 6 and the final reduction which give rise to chlo rtetracycline. Figure 5 shows the metabolites formed under the conditions of blocked chlorination. The final product is tetracycline. Figure 6 shows the forma­ tion of two hypothetical intermediates, methylchlortetramid-blue and chlortetramid-green. If the metabolic blocks of the preceding diagram are joined by a block of chlorination, methyltetramid-blue and tetramid-green are formed (Fig. 7). The whole diagram is supplemented by Fig. 8 w here no oxidation of ring A took place, and by Fig. 9 where no hydroxylation at C4 occu rred . 194 HOSTÁLEK et al.

@

NH2 2xMe OH-Cß

Ç1 Me OH OH B1- ^CONH2 он о OH OH methylhydroxychlorpretetramid

Me OH ■------o ç œ c , он о OH OH methyl hydroxypretetram id © FIG. 9. Biosynthesis of tetracycline-type compounds; pretetramid compounds. OH-C4 -2H « гО-С4о .С|2о Cl

Figure 10 shows the beginning of the biosynthetic series in which méthyl­ ation in position 6 was inhibited, the series being analogous with the methyl­ ated one. If the fourth ring of the hypothetical precursor is not closed, the number of intermediates and final metabolites is substantially reduced (Fig. 11). The enzymes modifying ring A cannot play a role here. It appears, however, that the enzyme hydroxylating tetracenes in position 6 is relatively unspecific, this being evidenced by the two isolated metabolites of this series, "anthrone" and protetrone. IAEA-SM-134/35 195

OH-C 4 - 2 H C l - C 7 NHj 2xMe 0H-C 6

-COOH A ■ c o n h 2

"o n th ro n e

0 H - C 6 -2H

О

•COOH A- C 0 N H 2

pro te tro ne

FIG.11. Biosynthesis of tetracycline-type compounds; tricyclic compounds.

Figure 12 summarizes the enzyme reactions considered here, including all the intermediates and final products. The metabolites shown by full circles have already been isolated, the empty circles represent substances which are so far hypothetical. There is a total of 72 metabolites of which 27 are already known and 45 are hypothetical. All are derived from a postulated tricyclic derivative by combination of 11 enzyme reactions: méthylation at C6, cyclization of ring A, removal of the oxo-group from Cg, hydroxylation at C4, oxidation of ring A, hydration at C4a and C12a, chlorination at C7, transamination, double N-méthylation, hydroxylation at Cg and final reduction. The maximum of metabolites is formed after hydro­ xylation at Cß, which points to the relatively low specificity of the corre­ sponding enzyme. The metabolic blocks at the level of these enzymes result in the production of the corresponding metabolites which accumulate in the culture medium. On the simplified assumption that one gene is responsible for the production of any one protein, it could be speculated that in the final phase of biosynthesis (after condensation of the malonate units to the tricyclic nonaketide) there are at least 11 structural genes in action. The existing studies of the genetic determinants of the biosynthesis of tetracycline antibiotics provided only scant data on the genetic control of the formation of these compounds. The genetics of tetracycline-producing streptomycetes was focussed, at first, on the genetic mapping of loci control­ ling the biosynthesis of amino acids and other primary metabolites. Alacevic [1 0 ] presented a report on the determination of the relative distances and on the sequence of some loci controlling the synthesis of proline, histidine, arginine, tryptophan, tyrosine and nicotinic acid in S. rimosus. In an ana­ lysis of the segregating phenotypes obtained from a large number of hetero­ clones crossing different types of auxotrophic mutants, the segregation of some non-selected markers was also followed (morphological properties, 196 HOSTÁLEK et al.

pigmentation and production of the antibiotic). Detailed results of this study have not been published so the linkage between genes participating in the biosynthesis of tetracyclines with nutritional markers is not clear at present. The results of complementation and recombination experiments with S. rimosus, carried out by Mindlin et al. [11] indicate the existence of at least two separate groups of loci controlling the biosynthesis of Oxytetra­ cycline. For studying the genetic control of the biosynthesis of the antibiotic, inactive mutants of S. rimosus have been used. It was shown that mutants of different metabolic groups possess blocks in the synthesis of Oxytetra­ cycline in different genetic loci. It was even possible to establish the rela­ tive arrangement of the various loci. The results of the crosses suggest that these determinants are situated in two regions of the genetic material, one of them containing two, the other four loci. Close linkage was demonstrated in the second region. The arrangement of the genetic loci controlling the synthesis of Oxytetracycline in relation to other determinants could not be demonstrated, however. Close linkage between some loci and the locus for XA EA - SM -13 4 /3 5 197

Streptomycin resistance is the only one known. The character of the bio­ chemical reactions controlled by the individual mapped loci has not yet been identified. There are several serious reasons for the unsatisfactory state of our present knowledge. In a number of the variants obtained the biosynthetic block has only a quantitative character. Most öf these strains produce, in addition to the mutant metabolite, tetracycline antibiotics. An important role is played here by the environmental conditions affecting the phenotypic expression. It is difficult to define the basis of a genetic block. Frequently we are dealing, in this connection, with a genetic change in the primary metabolism but not in the biosynthetic pathway itself. This is exemplified by some mutants of S. aureofaciens (IVth group according to Blumauerová et al. [12]) producing chlortetracycline and tetracycline at low concentrations. These mutants, complemented metabolically [13] with low-production variants of the Xllth group, produced increased amounts of both tetracyclines and aureovocin (aureovocidin is to be found in Fig. 8 ). This example also shows the typical metabolic improvement of primary metabolism: an increased inflow of the principal building units for the tetracene skeleton of the chlor­ tetracycline molecule. The production of aureovocin reflects the precursors not processed further to the chlortetracycline molecule. We demonstrated the formation of aureovocin even in S. rimosus strains. The production of new "mutant" metabolites may obviously be accomplished even in other.strains of the standard type. By changing the medium compo­ sition or conditions of aeration, thereby affecting the activity of enzyme systems of the biosynthetic pathway, one may attain changes in the production of the standard metabolite (chlortetracycline) as well as of other by-products, both positively and negatively. For crosses one must use mutants in which only the basis of the genetic block is defined. These mutants can be identified by further nutritional and drug-resistance markers. In a secondary mutagenic treatment, in the pre­ paration of auxotrophs, changes in the original activity of the blocked mutants may frequently be observed. This reduces the choice when selecting suitable strains. A further obstacle to successful crossing may be in the mutual fertility of the partners or in a low viability of the blocked mutants charac­ terized by additional markers. In the present experiments we observed that one of the suitable types of blocked mutants for crosses was a strain which had completely lost its ability to produce chlortetracycline and tetracycline under simultaneous production of aureovocin (Und group of mutants according to Blumauerová et al. [12]). This means that the oxidation of the hydroxyl group at C 4 of the A ring of the molecule which proceeds at an early stage (see Fig.3) is a very sensitive site decisive for the biosynthesis of chlortetracycline. Recently, we have followed the biosynthetic activity of haploid re­ combinants and the possibilities of a linkage relation between reference markers and those of secondary metabolism during crosses of different types of blocked mutants. The results were compared with crosses of S. rimosus auxotrophs differing mutually by blocks in the biosynthetic pathway of Oxytetracycline, aureovocin and other related compounds. Similarly, we have done interspecific crosses between S. rimosus and jS. aureofaciens. Preliminary experiments suggest the map position of some loci controlling the secondary metabolism of the two species. Detailed results will be published elsewhere. 198 HOSTÁLEK et al.

REFERENCES

[1] VANEK, Z., HOSTALEK, Z., Biogenesis of Antibiotic Substances, Publ. House Czech. Acad. Sei., Prague, Academic Press, London (1965). [2] VANEK, Z., HOSTALEK, Z., CUDLÍN, J., Genetics of Industrial Microorganisms (Proc. Symp. Prague, 1970), Academia, Prague (in press). [3] VANEK, Z., CUDLÍN, J., BLUMAUEROVÁ, M., HOSTÁLEK, Z., How many genes are required for the synthesis of chlortetracycline?, Folia microbiol., Praha 15 ( 1971) 227. [4] McCORMICK, J.R.D ., "Biosynthesis of tetracyclines", Biogenesis of Antibiotic Substances (VANEK, Z., HOSTALEK, Z ., Eds), Publ. House Czech. Acad. Sei., Prague, Academic Press, London (1965) 73. [5] McCORMICK, J.R.D., "Point-blocked mutants and the biogenesis of tetracyclines", Genetics and Breeding of Streptomyces (SERMONTI, G., ALACEVÍC, M., Eds), Yug. Acad. Sei. and Arts, Zagreb(1969) 163. [6] PODOJIL, M., VANEK, Z., VOKOUN, J., CUDLÍN, J., BLUMAUEROVÁ, M., VONDRÁCEK, M., HASSALL, C .H ., Secondary metabolites produced by biochemical mutants of Streptomyces aureofaciens, 1st Int.Symp.on Genetics of Industrial Microorganisms, Prague, 1970, Abstract Book, p.106. [7] BËHAL, V., VANÊK, Z., Regulation of biosynthesis of secondary metabolites. XII. Acetyl-CoA carboxylase in Streptomyces aureofaciens, Folia microbiol., Praha 15 (1970) 354. [8] VANËK, Z ., PÛÏA, M., CUDLÍN, J., VONDRÁCEK, M., RICKARDS, R. W., Biogenesis of cycloheximide, Folia microbiol., Praha 14 (1969) 388. [9] CUDLÍN, J., PÖZA, M., VANËK, Z., RICKARDS, R.W., Biogenesis of streptimidone. Folia microbiol., Praha 14(1969) 406. [10] ALAÖEVld, M., Genetics of tetracycline producing streptomycetes, 1st Int.Symp. on Genetics of Industrial Microorganisms, Prague, 1970, Abstract Book, p .44. [11] MINDLIN, S.Z., ZAYTZEVA, Z.M ., SHISHKINA, T.A ., Genetic and physiological studies of inactive mutants of Actinomyces rimosus, the producer of Oxytetracycline, Genetika 3 (1968) 126. [12] BLUMAUEROVÁ, M., ISMAIL, A.A., HOÍÍÁLEK, Z., VANÊK, Z., Mutation studies in Streptomyces aureofaciens, these Proceedings. [13] BLUMAUEROVÁ, M., HO§TrÁLEK, Z., MRACe k , M., PODOJIL, M., VANEK, Z., Regulation of biosynthesis of secondary metabolites. X. Metabolie complementation of blocked mutants of Strepto­ myces aureofaciens. Folia microbiol., Praha 14 (1969) 226.

DISCUSSION

VERA JOHANIDES: I should like to congratulate you on the chemical work reported in your paper. I think the genetic part will take a lot more work in the future. I wonder what you mean by specific secondary metabolite synthesis of species. Many species have been invented by pharmaceutical firms, and therefore the question of species in this genus is an open problem. Experiments on the cosynthesis of different wild Streptomyces species and inactive mutants of tetracycline-producing Streptomycetes have confirmed that many species in the genus can synthesize a precursor for the tetracycline m o le c u le s. V V / Z. HOSTALEK: In our paper we have pointed out that most biosynthetic reactions of secondary metabolism are restricted to certain species only (e.g. formation of malonamide). I agree with you that many criteria used in the Streptomyces taxonomy are quite primitive. The relationship between individual producers of compounds of the cycline type is indeed an open p ro b le m . S. I. ALIKHANIAN: I have listened to your presentation with great pleasure, and I wish to ask you a question. Did you try to localize your mutants on chromosomes? This could be useful in the subsequent use of interspecific hybridization of Streptomyces griseus and Streptomyces aureo­ faciens and for comparing your system of genetic regulation with the system developed by us in collaboration with Mindlin and Alacevic in Yugoslavia. LAEA-SM-134/35 199

Z. HOSTÁLEK: The work on the localization of individual genetic loci controlling the biosynthesis of tetracyclines is in progress. In our labora­ tory we are trying to obtain either crosses between^, aureofaciens mutants blocked in different stages of tetracycline synthesis or interspecific crosses between S. aureofaciens and S. rimosus to estimate the map position of the individual reactions.

IAEA-SM - 1 3 4 /32

PROTEIN SYNTHESIS AND PRODUCTION OF TETRACYCLINE IN Streptomyces aureofaciens

K . MIKULÍK, J. KARNETOVÁ, A . KREMEN, J. TAX , Z . VANEK Institute of Microbiology, Czechoslovak Academy of Sciences, Prague, Czechoslovak Socialist Republic

Abstract

PROTEIN SYNTHESIS AND PRODUCTION OF TETRACYCLINE IN Streptomyces aureofaciens. The interaction of the protein-synthesizing machinery with tetracycline in tetracycline-producing and non-producing mutant strains of Streptomyces aureofaciens was studied recently in this laboratory. It was found that the sedimentation pattern of ribosomes isolated during growth remained unchanged up to 24 h, i.e. up to the time when the concentration of tetracycline in the fermentation medium did not exceed 500 /ig/ml. The ribosomal preparations of 48- and 72-h cultures contained ribosomal aggregates differing in size from normal ribosomes, and formations resembling polysomes were observed. The highest level of binding was equivalent to 320 molecules of tetracycline per ribosome. The binding of tetracycline to ribosomes of Streptomyces aureofaciens was found to be highly reversible. It was possible to detect the level of the irreversible binding when 3H-tetracycline of a high specific radioactivity was used. It was calculated, on the basis of the above experiments, that one molecule of tetracycline was bound per ribosome. The S-30 fraction containing endogenous mRNA was used to study the effect of tetracycline on protein synthesis in vitro. The results of these experiments showed that the in-vitro protein synthesis of the tetracycline-producing strain is more resistant to drug effect than a similar system isolated from sensitive bacteria. The results obtained in this laboratory suggest that the drug is bound both to the ribosomal RNA and the ribosomal proteins.

INTRODUCTION

During the past few years many data have been reported on the effect of tetracyclines on cell and biochemical pathways. Evidence from both in-vivo and in-vitro experiments has indicated that tetracyclines interact with a wide variety of reactions, including respiration [1 , 2 ] , synthesis [3-7] and enzyme activities [8-12]. Particular attention has been paid to the elucidation of the inhibition effect of the drug on protein synthesis. Previous work with bacterial ribosomes has indicated that tetracyclines have a profound effect on their physical properties. For example, Day [13] reported that 100-S ribosomal dimers are dissociated to monomeric 7 0-S ribosomes in the presence of tetracycline (TC). The effect of chlortetracycline (CTC) in vivo on the size and content of polysomes has been reported in experiments by Cundliffe [14]. At present, most of the evidence favours the view that tetracyclines prevent the binding of aminoacyl-tRNA to the aminoacyl acceptor site of ribosomes [15-18]. However, higher concentrations of tetracycline (about 5X 10‘4 M) inhibit also the release of tRNA from ribosomes [19]). In spite of extensive studies on the mode of tetracycline action on sensitive organisms, the physiological role played by the drug in the meta­ bolism of microorganisms which produce them is still uncertain. The

201 202 M1KULÍK et a l.

purpose of the experiments described in this paper was to study the interaction of tetracycline with the protein-synthesizing machinery of Streptomyces aureofaciens and to clarify the mechanism of that interaction.

MATERIAL AND METHODS

The tetracycline-producing strain (84/25) and the non-producing strain (B-96) of Streptomyces aureofaciens were prepared by u.v. -irradiation of a parent strain Bg. Strains were grown under aeration at 28°С [20]. The cells were harvested from4-to72-h old cultures, washed twice with lOmM Tris-HCl buffer (pH 7. 5) containing lOmM magnesium acetate, 60mM KC1 and 6mM 2-mercaptoethanol (TCMK).

Effect of tetracycline on in-vivo incorporation of 14C-leucine into the cell of S. aureofaciens

Radioactive U 14C-L-leucine, specific activity 83 mCi/millimole, was added at the beginning of the cultivation of_S. aureofaciens in an amount of 20 цC i/60 ml of medium. In experiments following the effect of tetracycline (TC), 50- 1000 ßg/m l TC was supplied at 4 h of cultivation. At different intervals, 1-ml aliquots were poured into 2 ml of hot 10% TCA. After 20 min at 90°C, samples were chilled, filtered and washed on membrane filters (Synpor 6 , 0.40 мт pore size). Radioactivity was counted in a Biospan Nuclear Chicago counter.

Uptake of -^H-tetracycline by whole cells and stability of TC during experiments

An am ount of 500 y.g 3H -T C /m l (10 mCí) was added to 4-h cultures of S. aureofaciens■ Samples were taken at suitable time intervals. After centrifugation at 30 000 X g for 5 min at 2°C, the supernatant was used for chromatographic analysis and spectrophotometric estimation of TC [21]. Chromatography was carried out as described by Urx et al. [22]. The radioactive spots were separated and radioactivity measured in Bray's solution. The sediment was suspended in 3 ml of 1 OmM Tris-HCl buffer at pH 7. 5 and filtered through a membrane filter. The cell suspensions on filters were washed three times with 5 ml of the buffer, then dried and the radioactivity counted.

Preparation of ribosomes and ribosomal subunits

The isolation of ribosomes from alumina-ground extracts is described elsewhere by Mikulik [23]. The ribosomal pellet was washed twice with lOmM Tris-HCl buffer (pH 7. 5) containing lOmM magnesium acetate, 60mM KC1 and 6mM 2-mercaptoethanol (ICMK), and the ribosomes were purified by centrifugation through 30% sucrose in the same buffer. Purifi­ cation through sucrose was repeated twice more. Ribosomes were dis­ sociated into subunits by dialysis against TCMK buffer containing 0. ImM magnesium acetate. The ribosomal subunits were layered over 28 ml of 10 - 30% su cro se gradient and centrifuged in a SW 25.1 ro to r for 15§ h at IA EA-SM-134/32 203

19 600 revs/min on a Spinco Model L2-65B centrifuge. Separated subunits from three tubes were combined and centrifuged for 5 h at 48 000 revs/min in a 50. 1 rotor. The separation procedure was repeated twice.

Gel filtration of ribosomal aggregates

An amount of 3.7 5 mg ribosomes was incubated in a mixture containing lOmM Tris-HCl buffer at pH 7. 5, lOmM MgCl?, 30mM NH¿C1, 6mM 2-mercaptoethanol and ImM 3H -tetracy clin e (1 iuCi). A fter 90 min of incubation at 30°C, the mixture was centrifuged at 105 000 X g for 30 min. The supernatant was discarded and ribosomal aggregates were washed three times with 10 ml of the above buffer to remove free tetracycline. The resulting sediment was suspended in 1 ml of buffer and passed through a 2 cmX 50 cm Sephadex G50 column.

Aminoacyl synthetase preparation

Aminoacyl synthetase activity was estimated in the partially purified 105 000 X g supernatant fraction [24]. The supernatant fraction was stirred and 0.2 ml of 5% streptomycin sulphate slowly added. After 15 min of stirring, the precipitated material was removed by centrifugation at lOOOOXg. The sediment was removed and the supernatant was adjusted to pH 5.2 with 0.1NHC1. The mixture was centrifuged at lOOOOXg for 10 min. The supernatant was discarded and the sediment was suspended in 0. 1W[ Tris-HCl buffer (pH 7. 6 ) containing 6mM dithiothreitol. After an 8 -h dialysis against the same buffer the solution was centrifuged at 15 000 X g for 10 min. The supernatant solution was mixed 1 : 1 with glycerol and stored at -20°C until required. This solution was used as a source of aminoacyl synthetase activity.

Cell-free amino acid incorporating system

The activity of the S-30 fraction and the effect of TC were assayed using an in-vitro 14C-valine-incorporating system [24]. The specific activity of L - 14C-valine (New England Nuclear Corp.) was 50 mCi/millimole. The reaction mixture was incubated at 37°C for 25 min and incubation was stopped by adding 3 ml of 10% TCA. The samples were heated in a water bath for 20 min at 90° С, then chilled in ice and filtered under suction through a membrane filter. The filters were dried and the radioactivity was measured.

Sedimentation velocity analysis

This was done in a Spinco Model E analytical ultra-centrifuge. The sedimentation was studied at 37 020 revs/min and 20°C. The bar angle was 60° .

Isolation of rRNA

Ribosomal RNA was isolated by the method of Click and Hackett [25]. The ratio of phenol to buffer was 1:3. The buffer solution (pH 5.0) con­ tained sodium acetate 0. 1M, NaCl 0. IM, EDTA 0.01M and 0. 5% benthonite. 204 MIKULÍK et al.

FIG. l.(A) Incorporation of 14C -leucine into proteins of tetracycline-producing strain of Streptomyces aureofaciens (o - o) and production of tetracycline (•-•). (B) Effect of tetracycline on incorporation of 14C-leucine into proteins. Course of ^ -leu cin e incorporation after addition of tetracycline to 4-h cultures: 50 ng/ml (+ -+ ); 500 fig/ml (o-o) and 1000 jig/ml (Д— Д). In the experiment with 500 pg TC/ml the level of tetracycline was followed during the experiment (• -• ). For experimental details see section on Material and Methods.

The ribosomes were mixed for 15 min with phenol-buffer mixture and the emulsified extract centrifuged for 5 min at 12 000X g to sep arate the emulsion into two phases. The aqueous phase containing rRNA was separated and residues of phenol were removed by extraction with anhydrous ether. After removing the ether with N2, two volumes of anhydrous ethanol were added to precipitate rRNA. The rRNA was reprecipitated five times from pH 5.0 buffer. IAEA-SM-134/32 205

FIG.2. Incorporation of 14C-leucine into proteins of tetracycline-nonproducing strain of Streptomyces aureofaciens. 14 С -leucine incorporation in absence (o - o) and presence of 500 |ig/ml tetracycline (• — •).

Spectral data

Optical rotatory dispersions were measured at 30°С with a JASCO model ORD/UV-5. Ultra-violet and visible spectra were recorded by means of a Perkin-Elmer Model 402 UV-VIS spectrophotometer. NMR spectra were obtained using a Tesla BS 487 A 80 MHz instrument.

RESULTS AND DISCUSSION

Our experiments were designed to study the relationship between in-vivo protein synthesis and the production of tetracycline (TC). Figure 1A shows that the incorporation of 14C-leucine into proteins increased up to 12 h of cultivation, while the concentration of tetracycline in fermentation broth was relatively low (about 50 Mg/ml). The increase of TC production was accompanied by a decrease of 14C-leucine incorporation. After finishing the cultivation at 7 2 h, the production of TC was 2050 Mg/ml. In the next experiments (Fig. IB), the differences in sensitivity to TC with TC-producing and non-producing strains were followed. The addition of 500 ßg T C /m l to a 4-h-old culture of the TC-producing strain caused 37% inhibition of protein synthesis and complete inhibition of TC production. Under the same experimental conditions with the non-TC-producing strain (Fig. 2), 14C-leucine incorporation into the proteins was totally inhibited. When the 500 Aig TC/ml was added to 16-h cultures of the TC-producing strain, synthesis of tetracycline was not influenced [26]. It was of particular interest to test whether TC, which was added to young growing cultures, was modified to an inactive form or whether the 206 MIKULÍK et al.

TABLE I. STABILITY OF TETRACYCLINE AFTER ADDITION OF Streptomyces aureofaciens TO YOUNG GROWING CELLS

Radioactivity Spectrophotometric Incubation Strains o f 3H-te tra c y c lin e a estimation of tetracyclineb (h) (counts/ min) (¿ig/m l)

84/25 1150 495 B-96 1295 480

84/25 1270 480 B-96 1200 490

84/25 1079 482 B-96 1198 475

84 /2 5 1076 483 B-96 1285 491

a Samples were collected and centrifuged at 5000 x g for 15 min. Then 50 /Д of supernatants were chromatographed in the solvent system [22]. Radioactive spot (Rf 0.47) was separated and the radioactivity was measured, b Tetracycline was estimated according to Levine et al. [21].

FIG.3. Uptake of 3H-tetracycline by tetracycline-producing (• —•) and nonproducing (o —o) strains of Streptomyces aureofaciens. The 4-h-old cells were incubated at 30°C with 500 /ig/ml of 3H-tetracycline (10 pCi). The samples were suspended in lOmM Tris-HCl buffer at pH 7.5 and filtered through a membrane filter. Radioactivity of washed cells was counted in a Bray’s solution. IAEA -S M -1 3 4 /32 207

FIG.4. Distribution of tetracycline among the subcellular fraction. Ribosomes (•-•) and S-100 fraction (o— o) were prepared as described in the section on Material and Methods.

higher resistance of the TC-producing strain against tetracycline was connected with a modification in permeability. Results presented in Table I demonstrate that no significant quantity of tetracycline was inactivated or transformed during the course of the experiments with both strains studied. Significant differences in uptake of 3H-tetracycline were encountered when whole cells of tetracycline-producing and non-producing strains were incubated with the drug. The data are presented in Fig.3. At a concen­ tration of 500 pg/m l of tetracycline in the medium, the influx of the drug was relatively rapid. After 1 5 to 20 min the curves reached a plateau. The amount of tetracycline accumulated by the non-tetracycline-producing strain was higher, by about 23%, than in experiments with the tetracycline-producing strain.

Intracellular localization of tetracycline and sedimentation pattern of the ribosom es

These experiments were performed to ascertain whether intracellular accumulation of TC during the growth of the tetracycline-producing strain is accompanied by changes in the sedimentation properties of ribosomes. The results presented in Fig.4 indicated that the amount of TC bound to subcellular fractions rapidly rose at about 24 h of cultivation, comparable with the production of TC (Fig. 1). The sedimentation profile of ribosomes also remained consistent up to 24 h, i.e. up to the time when the concen­ tration of TC in the cultivation medium did not exceed 500 ¿ig/ml (Fig. 5). 208 M1KULÍK et a l. IAEA-SM-134/32 209

FIG.6. Binding of 3H-tetracycline to ribosomes and ribosomal subunits of the tetracycline-producing strain of Streptomyces aureofaciens. Cells of 10-h cultures were treated for 2 h with 3H-tetracycline 500 iig/ml (20 /jCi). 20 Az60 units of ribosomal subunits (A) and 19 A 260 units of ribosomes (B) were layered on М-ЗО^о sucrose gradients and centrifuged for 15i h at 19 600 revs/min. One-ml fractions were collected. The sediment on the bottom of the centrifuge tubes was suspended in 1 ml of TCMK buffer at pH 7.5. (— ) 3H-tetracycline in counts/min. ( ----- ) Absorbancy at 260 nm.

These results also show that 100-S ribosomal dimers of_S. aureofaciens are more stable in the presence of tetracycline than 100-S particles isolated from E_. coli [13]. The preparation isolated from 48- and 72-h cultures contains ribosomal aggregates which sediment rapidly at the beginning of the centrifugation. Evidence for the formation of ribosomal aggregates in the presence of higher concentrations of CTC (ImM) was reported in our previous paper [27].

Binding of tetracycline to ribosome and ribosomal subunits

It has been reported that tetracycline binds to 70-S ribosomes [13, 28] and ribosomal subunits. The amount of tetracycline bound to 30-S sub­ units was considerably greater than to 50-S subunits [29]. This was con­ firmed in our experiments with ribosomes of S. aureofaciens. 500 /ug/ml 3H -tetracy clin e (20 ßCi) were added to 10-h cultures. After 2 h of cultivation at 28°С the cultures were chilled and then washed twice with TCMK buffer (pH 7.5). Ribosomes isolated from these experiments were dialysed against TCMK buffer containing either lOmM or 0. ImM magnesium acetate. Ribosomes and ribosomal subunits were separated on a sucrose density gradient. Figure 6 shows that radioactivity of 3H-tetracycline was associated with 70~S and 100-S dimers. In experi­ ments with 0. ImM Mg2* most of the radioactivity of TC was associated 210 MIKULÍK et al.

m l

FIG.7. Sephadex'G50 gel filtration of the tetracycline-ribosome complex. Washed complex was applied on a 2 cm x50 cm Sephadex G50 column and eluted with lOmM Tris-HCl buffer at pH 7.5 containing lOmM MgCl2, 30mM NH4C1 and 6mM 2-mercaptoethanol. (•— •) radioactivity of 3H-tetracycline, (x— x) absorbancy at 260 nm.

with 30-S subunits. A small, but significant, radioactivity of TC binds to 50-S particles. A considerable part of the radioactivity was connected with sedimentations of ribosomal aggregates at the bottom of the centrifuge tubes. These aggregates are stable when treated with trypsine or pancreatic ribonuclease. The highest level of binding observed was equivalent to 320 molecules of tetracycline per ribosome. This value was calculated from the absorbance and radioactivity of the 70-S peak fraction. Binding of tetracycline to ribosomes of ÍS. aureofaciens was largely reversible, like that reported in experiments with ribosomes of B. megatherium [28]. When SH-tetracycline of high specific activity was used it was possible to detect the level of irreversible binding. Figure 7 shows an elution profile of the tetracycline-ribosome complex on a Sephadex G50 column. About 1% of the total radioactivity was connected with the ribosomes. From these experiments it was concluded that one molecule of tetra­ cycline was tightly bound to a ribosome.

Activation of amino acids and in-vitro protein synthesis of tetracycline- producing strain of S. aureofaciens

Aminoacyl-tRNA synthetases fulfil two important roles in the protein-synthesizing machinery, i.e. activation of amino acids and trans­ lation of amino-acid specificity into base-pairing specificity. IAEA-SM-134/32 211

FIG. 8. Activity of aminoacyl-tRNA synthetases at different intervals of growth and production of tetracycline. Partially purified preparations of aminoacyl synthetases (see M aterial and Methods) were incubated in a reaction mixture containing 0.1M Tris-HCl buffer at pH 7.5, 0.01M KCl, 0.02M magnesium acetate, 0.005M ATP, 0.005M CTP, 8 A260 units of E. coli tRNA, 0.003M mixture of unlabelled amino acid plus 0.012 jjmoles of each of the 14C-amino acids tested. Incubation, 15m m at37°C . The reactions were stopped by addition of 2 ml of cold 10% TCA. The samples were filtered through a membrane filter, washed three times with 5 ml of ice-cold 5°jo TCA and the radioactivity was measured. Charging experiments with leucine (o—o), valine (Д—Л) and alanine (□—□ ). Production of tetracycline ( • —•).

To obtain more detailed information about interaction between the initial steps of protein synthesis and production of tetracycline, the charg­ ing of E. coli tRNA's with radioactive leucine, valine and alanine was estimated. These reactions were catalysed by partially purified amino­ acyl synthetases that were isolated at different intervals of growth of S. aureofaciens■ The concentration of proteins of partially purified enzymes was adjusted to 1 mg/ml. Figure 8 shows the results of this experiment. The maximum aminoacyl synthetase activity for all substrates studied was observed with preparations isolated from 12-h cultures. There exists a direct, if not proportional, relationship between depression of aminoacyl synthetase activity and the production of the drug. Similar conclusions were reported by Hosïâlek et al. [30] who studied the interaction of individual enzymes of the tricarboxylic acid cycle and the production of tetracycline by S. aureofaciens. In our next experiments we studied the effect of tetracycline on an in-vitro protein-synthesizing system. Table II shows the composition of the system prepared from 10-h cultures of S. aureofaciens. The incorpora­ tion of 14C-valine into protein was proportional to the amount of S-30 frac­ tion added. The non-preincubated mixture contains endogenous mRNA, 212 MIKULÍK et al.

TABLE II. CHARACTERISTICS OF IN-VITRO PROTEIN SYNTHESIS IN Streptomyces aureofaciens DIRECTED BY ENDOGENOUS mRNA

Radioactivity Incubation components (counts/min)

Complete + non-preincubated S-30 fraction (1 m g /m l) 1176

Complete + non-preincubated S-30 fraction + DNase (5 (jg/m l) 1128

Complete + non-preincubated S-30 fraction + RNase (5 jjg/m l) 4

Complete + preincubated S-30 fraction (1 m g /m l) 110

The reaction mixtures contained, in^mole/ml: 100 Tris-HCl at pH 7.8, 10 magnesium acetate, 50 KC1, 6 2-mercaptoethanol, 1 ATP, 0.03 GTP, 5 phosphoenolpyruvate, 20 fjg PEP-kinase, 0.05 mixture of 20 L-amino acids minus valine, 0.018 L-valine (125 mCi/millimole) and an S-30 fraction. Samples were incubated at 37eC for 25 min, deproteinized with lOfy mTCA . Precipitates were washed three times with b°jo TCA and counted on a Biospan Nuclear Chicago counter. Background, 2 imp/min.

FIG.9. The effect of tetracycline on in-vitro protein synthesis directed by endogenous mRNA. The compo­ sition of reaction mixtures was the same as those described in Table II. IAEA-SM-134/32 213

FIG. 10. Spectral changes of tetracycline as a function of Mgz+ concentrations. Solution of tetracycline (5X10"5 M) in lOmM Tris-HCl at pH 7.5 was incubated for 15 min at 30°C with different amounts of MgCl2. (0) control without Mg2+; (1) with 10-4 M MgCl2 ; (2) with 10-3 M MgCl2; (3) with lO'Z i£MgCl2.

sensitive to RNAse action. Protein synthesis in this system is not dependent on the presence of DNA. After preincubation of the S-30 fraction to destroy endogenous mRNA, synthesis of proteins was lowered to 10% of the maximal value observed with the non-preincubated fraction. The dialysed S-30 fraction without preincubation was used to study the effect of tetracycline on protein synthesis in the tetracycline-producing strain. Figure 9 shows the per cent inhibition of protein synthesis obtained at different levels of tetracycline. The results of these experiments suggest that the protein-synthesizing system of S. aureofaciens is more resistant to the drug effect than a similar system isolated from sensitive bacteria. Craven et al. [31] have reported that 40 Mg/ml of tetracycline causes 83% inhibition of poly U directed protein synthesis of tetracycline- sensitive cells of E. coli. The same inhibition effect was observed in experiments with !3. aureofaciens at a drug concentration of 200 Mg/ml.

Spectral studies of the ribosome-tetracycline binding interaction

The question remains open as to which part of the tetracycline mole­ cule is involved in the binding interaction with ribosomes or its components. In the present experiment the effect of ribosomes and rRNA on spectral properties of tetracycline is described. Figure 10 shows the changes in the tetracycline spectrum as a function of magnesium concentrations. The increase in Mg2+ concentration from 0 to lOmM results in a shift of u.v. absorption m axim um of Amax = 228 nm to Amax = 242 nm. The curves pass through two isobestic points at 280 nm and 345 nm. A slight shift to a higher wavelength can be also observed in the region from 376 nm to 385 nm. This shift is accompanied by consider­ able increase in maximum absorbance. 214 MIKULÍK et al.

nm

FIG.12. Optical rotatory dispersion of tetracycline (10~4 M) in lOmM Tris-HCl buffer at pH 7.5 without Mg2+ (----- ) and with ЮшМ MgCl2 (— ).

The question of which chemical group(s) of tetracycline is involved in binding with metal ions was the subject of considerable investigation. On the basis of potentiometric titrations of tetracycline [32] it was concluded that for Cu2+, Ni2+ and Zn2+ the binding group is the dimethylamino moiety of C 4 and the hydroxyl at C3 or C12a (see Fig. 11). This suggestion was corrected by means of the. d-d spectra of the transition-metal complex of tetracycline, anhydrotetracycline and dedimethylaminotetracycline [33]. The spectra of Ni2+ complexes indicate that only the oxygens are co­ ordinated to the metal. The bonding takes place through two oxygens of the 1, 2, 3-tricarbonylmethane system; the amide oxygen at C2 and the hydroxyl at Ci or C 3 . However, the enolized /З-diketone group at Сц and IAEA-SM-134/32 215

b

8 0 0 J o b c/s from HMDSO

FIG.13. Protein magnetic resonance spectra of tetracycline (0.1M) in DMSO-d6 (a) without and (b) with 0.1M MgCl2.

С i2 was not excluded from consideration. X-ray [34] and infra-red data [35] have shown a considerable amount of intramolecular H-bonding of the Сю "C n - C 12-C! position of tetracycline in solution. The data coming from CD studies [36-38] show that all of the u.v. bands of tetracycline are optically active. The BCD chromophor (Fig. 11) is responsible for the band at 360, 320, 285 and 225 nm and the ring A contributes the 260 nm band. The formation of a tetracycline chelate with Mg2+ under physiological conditions was estimated in our next experiments, using optical rotatory dispersion ORD and proton-magnetic resonance. Figure 12 shows the ORD of tetracycline in Tris-HCl buffer (pH 7. 5) with and without Mg2+. The molar ratio of tetracycline to Mg2+ was 1:100. These results indicated that the BCD chromophor is involved in complex formation. Similar results were reported by Mitscher et al. [38] studying the formation of tetracycline chelates with Al3+, Ca2+ and Mg2+, using circular dichroism measurement. These findings show that at pH 7.5 a BCD chelate is formed and the A ring complex is less evident. Complex formation of tetracycline on addition of M gCl2 (molar ratio 1:1) was also studied using the NMR techniques. Spectra were obtained in an 80 MHz instrument. The results, however, cannot be interpreted unequivocally. Firstly, the experiments could not be performed under physiological conditions. The solubility of tetracycline in water at about pH 7 is too low for obtaining a reasonable NMR spectrum in a single scan and, moreover, proton exchange of the hydroxyl groups makes it impossible to obtain resonance from these protons. For these reasons we used tetracycline solutions in DMSO-dß (from 0. 04M to 0. 1M). Secondly, it is difficult to identify the respective resonance peaks of the hydroxyl groups of interest. The spectra in Fig. 13 show a typical change of two of the tetracycline resonances on addition of MgCl2. We believe that these results support the findings obtained by other methods. In order to study the effect of ribosome concentration on the spectral properties of tetracycline, the reaction mixture containing lOmM Tris-HCl buffer (pH 7.5), lOmM magnesium acetate and tetracycline was incubated 216 MIKULÍK et al.

FIG. 14. Interaction of 70-S ribosomes with absorption spectra of tetracycline. Ribosomes at increasing concentrations (from 20 Л 260 units to 80 A260 units) were incubated in Tris-HCl buffer at pH 7.5 containing 5 x 10'5 M tetracycline and 10"2 M magnesium chloride. Incubation, 15 min at 30°C. Solid line: tetra­ cycline without ribosomes. Broken line: decrease of absorbance at 385 nm of tetracycline upon addition of increasing concentrations of ribosomes.

FIG. 15. Interaction of 50-S ribosomal subunits with absorption spectra of tetracycline. Solid line: 5 x l0 -5 tetracycline in the Tris buffer at pH 7.5 with 10“4 M magnesium chloride. Broken line: effect of increase of concentration of 50-S subunits (from 15 A260 to 60 A260 units) on absorption spectra of tetracycline. with different concentrations of ribosomes. Figure 14 shows that when ribosome concentrations increase, the absorbance of tetracycline at 385 nm considerably decreases. Similar results were obtained in experi­ ments with both 50-S and 30-S ribosomal subunits (Figs 15 and 16) with the exception that reaction mixtures contained 0. ImM magnesium acetate and that a hypochromic effect was observed at 376 nm. Ribosomes represent a very complicated system in which rRNA, as well as ribosomal proteins, may be involved in a binding interaction with tetracycline. In subsequent experiments we studied the effect of rRNA IAEA-SM-134/32 217

FIG. 16. Interaction of 30-S ribosomal subunits with absorption spectra of tetracycline. Solid line: 5 x 10"5 M tetracycline in Tris buffer at pH 7.5 with 10’4 M magnesium chloride. Broken line: effect of increasing concentrations of 30-S subunits (from 15A260 to 6OA260 units) on absorption spectra of tetracycline.

Wavelength (nm)

FIG.17. Absorption spectra of tetracycline with different amounts of rRNA in the presence of Mg2+ (— ). The blank sample contained lOmM Tris-HCl buffer at pH 7.5 with lOmM MgCl2. The different spectra were obtained by combining the tetracycline (5 x 10“5 M) in the buffer at pH 7.5 with increasing concen­ trations of rRNA (from 5A260 to 25 AZ60 units) in the sample beam and the tetracycline-buffer solution in the reference beam. Individual samples were incubated before measurement for 15 min at 30°C. The cell path was 1 cm. Tetracycline (----- ). T e tra c y c lin e plus rRNA ( — ). 218 MIKULÍK et al.

FIG.18. Absorption spectra of tetracycline with different amounts of rRNA in the absence of Mg2+. Control- tetracycline 5x 10"5 M in lOmM Tris-HCl buffer pH 7. 5 С----- ). Tetracycline in the above buffer after incubation with increasing amounts of rRNA (— ).

1.4-

1.2 -

1.0-

200 250 300 350 400 450 500 550 600 650 Wavelength (nm)

FIG. 19. The different spectra of tetracycline with DNA. These spectra were obtained similarly as described in Fig. 17. Tetracycline С— ). Tetracycline plus DNA (— ). DNA of E. coli was obtained from Calbiochem, USA.

on the spectral properties of tetracycline. Figure 17 demonstrates that rRNA in the presence of magnesium (1 OmM) causes a decrease of absorbance at 385 nm of tetracycline. On the other hand, when rRNA was incubated with tetracycline without magnesium, the hypochromic effect at 385 nm was not observed (Fig. 18). These experiments suggested that the BCD chromophor of tetracycline forms a chelate with magnesium which interacts most probably with the 2'-OH group of ribose and the PO 2 group of rR N A . IAEA-SM-134/32 219

Wavelength ( nm)

FIG.20. Absorption spectra of tetracycline in the presence of ribosomal proteins. Ribosomal proteins were isolated by the LiCl-urea method [27] and dialysed for 6 h against lOmM Tris-HCl buffer at pH 7.8. (A) After incubation in the buffer without MgCl2 and 5 x l0 -5 M tetracycline. (----- ) Tetracycline without ribosomal proteins. (— ) Tetracycline with increasing concentrations of ribosomal proteins (from 0.05 to 0.2 mg/ml). (B) After incubation in the buffer with lOmM MgCl2 and 5X10"5 M tetracycline. (----- ) Tetracycline without ribosomal protein. (— ) Tetracycline with increasing concentrations of ribosomal proteins (from 0.05 to 0.2 mg/ml).

This suggestion is supported by experiments following the effect of DNA on the spectral properties of tetracycline. Results of these experiments (Fig. 19) show that the presence of DNA in the same concentration as in experiments with rRNA has no effect on the 385 nm band of tetracycline, both with and without magnesium. However, different conformations of DNA and rRNA, and the possibility of interaction of the Mg-tetracycline complex with the phosphodiester group of rRNA, cannot be excluded from consideration since the strength of interaction of the P0 2 group in RNA is 220 M1KULÍK et al. quite different from that in DNA [39]. Changes in the absorption spectra of tetracycline molecules bound to ribosomal proteins are shown in Figs 20A and B. When-the ribosomal proteins were incubated with tetra­ cycline solution without Mg2+ (A) a hypochromic effect at the 370 nm region of tetracycline and a shift to a longer wavelength (from 370 nm to 376 nm) was observed. In the presence of lOmM magnesium chloride in the reaction mixture (B) the hypochromic effect or shift in wavelength of tetracycline Chromophore was not encountered.

CONCLUDING REMARKS

The relationship between antibiotic synthesis and other metabolic pro­ cesses of the cell has been the subject of numerous discussions. Our approach to the study of this problem was to correlate the formation of the antibiotic with the protein-synthesizing machinery of the producing microorganism. As an experimental model we used Streptomyces aureofaciens which produces a broad spectrum of metabolites, including well-known inhibitors of protein synthesis such as tetracycline and chlortetracycline. The first question that had to be answered was whether tetracycline can pass through the cell wall and interact with the protein-synthesizing system of the cell. These experiments showed that tetracycline was accumulated by whole cells, but the uptake of the drug was less in experiments with the tetracycline-producing strain than under the same conditions with a non-TC-producing strain. The addition of tetracycline at the beginning of the exponential phase to a TC-producing strain results in a partial inhibition of protein synthesis and a complete inhibition of tetracycline production. When tetracycline was added at 16 h of cultivation, synthesis of tetracycline was not influenced [26]. These results suggest that tetracycline interacts with the synthesis of specific enzymes which are needed for the bio­ synthesis of the drug rather than with their activities. When зн-tetracycline was used we were able to demonstrate that ribosomes of the tetracycline-producing strain bound the drug like ribo­ somes of TC-sensitive bacteria. However, the polysomes, as well as the ribosomes of S_. aureofaciens, are stable in the presence of low concentrations of tetiacycline (50 ^g/ml) [27]. At higher concentrations of tetracycline (500 Mg/ml) ribosomes form aggregates, different from untreated ribosomes or preparations of young growing cells in size and sedimentation properties. These results are in accordance with our experiments following the effect of the drug on in-vitro protein synthesis. This system is more resistant to the tetracycline effect than similar ones of bacteria. There is not enough information about the mechanism of resistance of protein synthesis to tetracycline. Recently, Craven et al. [31] isolated a tetracycline-resistant mutant containing ribosomes resistant in vitro. Resistance to the antibiotic can be removed or decreased by salt washing. However, it cannot be concluded whether this resistance is located in a ribosomal protein or in the initiation factors. Studies of drug-macromolecule interaction using spectroscopic tech­ niques provided much useful information. In our experiments the question as to which part of the tetracycline molecule is involved in the binding interaction with ribosomes or rRNA was studied. At the physiological pH IAEA-SM-134/32 221 of 7. 5 the BCD chromophor of tetracycline (most probably its enolized /З-diketone group at Сц and C12) forms a chelate with Mg2+ which interacts with rRNA. On the other hand, the spectral interaction of ribosomal protein with tetracycline was observed only in the absence of Mg2+. More detailed information about this type of linkage is under examination.

ACKNOWLEDGEMENTS

The authors wish to thank Dr. A. Kotyk for valuable help in the preparation of the manuscript. This research was supported by the International Atomic Energy Agency under Research Contract N. 845/RB.

REFERENCES

[1] JOHNSON, E.J., COLMER, A.R., Antibiotics Chemother. 7 (1957) 521. [2] SNELL, J.F., CHENG, L., Devs ind. Microbiol. 2 (1961) 107. [3] CREASER, E.H., J. gen. Microbiol. 12 (1955) 288. [4] KATAGIRI, H., SUZUKI, Y., GROSSOWICZ, N ., Antibiotics 14 (1961) 134. [5] MELNYKOVYCH, G ., JOHANSSON, J. Bacteriol. 77 (1959) 638. 16] ALEXANDER, B., Appl. Microbiol. 8 (1960) 69. [7] deVRIES, H., KROON, A .M ., Biochim. biophys. Acta 204 (1970) 531. [8] ROKOS, J., BURGER, M ., PROCHÁZKA, P., Nature 18lTl958) 1201. [9] ROKOS, J . , BURGER, M ., PROCHÁZKA, P ., A ntib io tik i 4 (1959) 3. [10] ROKOS, J., MÁLEK, P., BURGER, M ., PROCHÁZKA, P., KLOC, J., Antibiotics Chemother. 9 (1959) 600. [11] ARORA, K .L., KRISHNA MURTI, C.R., J. scient. Res., Djakarta 19 (1960) 103. [12] BELDING, M ., KERN, F., J. Lab. clin. Med. -61 (1963) 560. [13] DAY, L.E., J. Bacteriol. 91 (1966) 1917. [14] CUNDLIFFE, E., Mol. Pharmacol. 3 (1967) 401. [15] HIEROWSKI, M ., Proc. natn. Acad. Sei. USA 53 (1965) 594. [16] SUAREZ, G., NATHANS, D., Biochem. biophys. Res. Commun. 18 (1965) 743. [17] SARKAR, S., THACH, R.E., Proc. natn. Acad. Sei. USA 60 (1968) 1479. [18] BODLEY, J.W ., ZIEVE, F .J., Biochem. biophys. Res. Commun. 36 (1969) 463. [19] ISHITSUKA, M ., KAJI, A ., Proc. natn. Acad. Sei. USA 66 (197ÖT 168. [20] HOSfÁLEK, Z., Folia microbiol., Praha 9 (1964) 78. [21] LEVINE, J., GARLOCK, E.A., FISCHBACH, H ., J. Am. Pharm. Assoc. 38 (1949) 473. [22] URX, M ., VONDRÁCKOVÁ, J., KOVArIk , L ., HORSKY, О ., HEROLD, M ., J. C hrom atog. 11 (1963) 62. [23] MIKULÍK, K ., QU YEN, N .. BLUMAUEROVÁ, M ., VANËK, Z ., FEBS Letters 5 (1969) 131. [24] M ATTHAEI, J .H ., NIRENBERG, M .W ., Proc. n atn . A cad. S ei. USA 47 (1961) 1580. [25] CLICK, R. E., HACKETT, D .P., Biochim. biophys. Acta 129 (1966) 74. [26] BLUMAUEROVA, M ., Ph. D. Thesis, Czechoslovak Acad. Sei., Prague 1969. [27] MIKULÍK, K ., BLUMAUEROVÁ, M ., VANÊK, Z ., LUDVÍK, J ., Folia microbiol., Praha 16 (1971) 24. [28] CONNAMACHER, R.H., MANDEL, H .G ., Biochem. biophys. Res. Commun. 20 (1965) 98. [29] MAXWELL, I.H., Mol. Pharmacol. 4 (1968) 25. _ [30] HO§TALEK, Z ., TINTËROVA, M ., JECHOVÁ, V., BLUMAUEROVÁ, M ., SUCHY, J ., VANËK, Z ., Biotechnol. Bioeng. 9 (1969) 539. [31] CRAVEN, G.R., GAVIN, R., FANNING, T ., Cold Spring Harbor Symp. Quant. Biol. 34 (1969) 129. [32] DOLVISIO, J.T ., MARTIN, A.N., J. Med. Chem. 6 (1963) 16. [33] BAKER, W .A ., J r ., BROWN, P .M ., J. A m . c h e m . S oc. 88 (1966) 1314. [34] DONOHUE, I., DUNITZ, J.D., TRUEBLOOD, K.N., WEBSTER, M .S., J. Am. chem. Soc. 85 (1963) 851. [35] KALNIN'SH, K .K., BELENSKIJ, B.G., Dokl. Akad. Nauk SSSR 157 (1964) 619. 222 MIKULÍK et al.

[36] MITSCHER, L.A ., BONACCI, A .C ., SOKOLOSKI, T .D ., Tetrahedron Letters 51 (1968) 5361. [37] MITSCHER, L.A ., BONACCI, A .C ., SOKOLOSKI, T .D ., Antimicrobial Agents Chemother. (1968)p.78. [38] MITSCHER, L.A., BONACCI, A .C ., SLATER-ENG, B., HACKER, A .K., SOKOLOSKI, T.D ., Antimicrobial Agents Chemother. (1969) p .ll. [39] BRAHMS, J., MAURIZOT, J.C ., PILET, J., Biochim. biophys. Acta 186 (1969) 110.

DISC USSION

S.G. GEORGOPOULOS: You have presented data on the in-vitro sensitivity of protein synthesis to tetracycline only for the tetracycline- producing strain. Is the in-vitro sensitivity of the non-producer higher or is it only in tetracycline uptake that the two strains differ? K. MIKULIK: Recently, we also estimated the sensitivity of the protein synthesizing systems of three non-tetracycline-producing strains of Streptomyces aureofaciens. The results of these experiments show that the protein-synthesizing systems of these strains were more sensitive to antibiotic action than such systems isolated from tetracycline- producing strains. IAEA-SM-134/9

SELECTION OF MUTANTS NOT ACCUMULATING STORAGE MATERIALS

H .G. SCHLEGEL, V. OEDING Institut für Mikrobiologie der Gesellschaft für Strahlen- und Umweltforschung m .b . H. München, Göttingen, Federal Republic of Germany

Abstract

SELECTION OF MUTANTS NOT ACCUMULATING STORAGE MATERIALS. The production of single-cell protein has been recommended in order to increase the protein resources for animal and human nutrition. Microorganisms generally accumulate reserve materials (triglycerides, poly-S-hydroxybutyrate or glycogen) when growing under conditions of optimal biosynthetic efficiency for biomass production. These reserve materials are non-proteinaceous; their production should therefore be avoided. One way to suppress the production of non-proteinaceous matter is the utilization of mutants which are blocked in the synthesis and accumulation of reserve materials. Methods to select for and isolate such mutants have been developed using the hydrogen oxidizing bacterium, Hydrogenomonas eutropha strain H16. This bacterium is very well suited for the production of single-cell protein on the basis of molecular hydrogen, oxygen and carbon dioxide, and is seriously considered for application in bioregenerative systems (space trips) and for food production from electricity via water electrolysis. The methods for the isolation of lipid-free mutants involve the use of ultra-violet light as the mutagen and 32P-phosphate as a selective agent. By applying the KP-suicide method it is possible to selectively kill the wild-type cells and leave those mutants unimpaired which do not accumulate the lipid poly-S-hydroxybutyrate. Another selective method, which has been developed recently, is based on sucrose density centrifugation. Details of these methods and the physiological properties of the mutants isolated so far are discussed.

The production of single-cell protein has been recommended in order to increase the protein resources for animal and human nutrition [ 1 ]. When microorganisms grow under conditions of optimal biosynthetic efficiency for biomass production they generally accumulate reserve materials (triglycerides, poly-/3-hydroxybutyric acid or glycogen) which can amount to 70% of the cellular dry weight. These reserve materials are non-proteinaceous so their production should be avoided. One way to suppress the production of non-proteinaceous matter is to use mutants which are blocked in the synthesis and accumulation of reserve materials. Methods have been developed with Hydrogenomonas eutropha strain HI 6 to select for and isolate mutants that do not accumulate storage material (poly-/3-hydroxybutyric acid = PHB). This strain is able to grow at the expense of the oxidation of gaseous nydrogen and with carbon dioxide as the only carbon source. It also lends itself to the production of single-cell protein [2]. The methods comprise the induction of the mutation, the selection of the mutants by the 32P-suicide method and by sucrose density gradient centrifugation and then the recognition of the m utants.

Induction of mutations by u. v. -irradiation

To increase the mutation rate, u.v. -radiation was chosen as the mutagenic agent. Cell suspensions, taken from the stationary phase of growth, were washed in saline, suspended to 1 0 7 с ells/m l and exposed

223 224 SCHLEGEL and OEDING

30 60 sec 90 irradiation time

FIG. 1. Inactivation of Hydrogenomonas eutropha strain H16 by ultraviolet irradiation. Fructose-grown cells were harvested in the stationary phase of growth, washed and re-suspended in saline to 107 cells/ml. Samples were taken and plated on nutrient broth agar.

to the radiation of a u. v. -lamp (Marggraf, Berlin) at a distance of 6 cm. Photoreversion was prevented by performing all the manipulations in dim light. The inactivation curves exhibit multi-hit kinetics (Fig. 1). To induce mutations, the cells were exposed to a u.v. -dose of 50 sec which corresponds to a survival of 1 0 -3.

Recognition of PHB-free mutants on agar-plates

After growth on nitrogen-poor agar plates (0.005% NH4CI; 0.5% fructose), colonies of wild-type cells can be differentiated from colonies of PHB-free mutants by two methods. (1) By staining with Sudan black В (0.02% in 96% ethanol) and following the differentiation by ethanol. The wild-type colonies retain the dye and appear dark blue whereas the colonies of the PHB-deficient cells appear light grey. (2) By discriminating between the appearance of the colony types without staining. The wild-type colonies appear milky or opaque (due to the PHB grana) whereas the mutant colonies are translucent. The latter method is advantageous since replica plating can be dispensed with. This method has been used throughout this project.

ENRICHMENT OF MUTANTS LACKING POLY-/3-HYDROXYBUTYRIC ACID

Four mutants have been isolated from irradiated cultures without any previous enrichment procedure. From the number of colonies investigated, a mutant frequency of about 10‘ 5 can be deduced. Therefore, enrichment methods have been elaborated and employed.

(a) Selective killing of wild-type cells by antibiotics

The colistine method proved to be inapplicable. Although the storage material enables the cells to grow in the absence of an exogenous carbon and energy source, growth is rather modest and not sufficient to permit inactivation by colistine. IAEA-SM-134/9 225

exposure time(h)

FIG. 2. Decrease of viability during contact of H16 cells with various gradient solutions.

(b) Selective killing of wild-type cells by 32P-phosphate

This technique proved to be more successful [3].

(c) Density-gradient centrifugation

The separation of PHB-free cells from those having accumulated large amounts of PHB has been achieved by density-gradient centrifugation. Wild-type cells survive the contact with the various solutions usually used for density-gradient centrifugation for several hours. The following compounds and concentrations were chosen: dextran T 500:50{per cent, w/v); ficoll: 50 (per cent, w/v); sucrose: 80 (per cent, w/v); and semi-saturated CsCl solution (20°C). While dextran is tolerated without loss of viability, the other solutions are more or less bactericidal (Fig. 2). For routine work sucrose was used.

PROPERTIES OF LINEAR GRADIENTS

Linear gradients of sucrose, dextran and ficoll were prepared from equal amounts (6.5 ml) of solutions of two different concentrations using a two-cylinder mixing device. A mixture of mutant cells and PHB-rich wild-type cells was suspended at a final gradient concentration corresponding to the lowest sucrose or polymer concentration employed, and 0.2 m l of this suspension was layered on top of the gradient. A Beckman ultra­ centrifuge, Spinco Model L 65 (Rotor SW 40 Ti), was used for centrifugation. The position and width of the bands could be easily detected and measured when cellulose-nitrate tubes were used (Table I). The broad bands in the ficoll and dextran gradients are the result of the rather low gradient; both polymers are viscous and could therefore not be used at higher concen­ trations. For these reasons, sucrose proved to be the best suited for the separation of mutants and the easiest to handle. 226 SCHLEGEL and OEDING

TABLE I. LOCATION AND WIDTH OF BANDS OF PHB-DEFICIENT AND PHB-RICH CELLS IN LINEAR GRADIENTS OF VARIOUS SUGARS

Concentration in Centrifugation PHB'4 H16 Sugar per ce n t tim e (w /v) (m in) (Cm above the tube bottom)

F icoll 50 - 20 90 3 .0 - 4. 1 5. 8 - 6. 8

D extran 50 - 20 120 5. 2 - 6 .1 6. 6 - 7 .4

Sucrose 80 - 20 120 1 .9 - 2 .3 2. 8 - 3. 0

number of fractions ( 5 d rops each}

FIG. 3. The separation of PHB-containing and PHB-fee cells by sucrose density-gradient centrifugation. Cells of the PHB-free mutant PHB*4 and of the wild type of H16 were mixed in equal proportions to a final concentration of 4 CP/o sucrose and layered on top of a linear sucrose-density gradient (40-90°Jo). T he tubes were centrifuged for 120 min at 30 000 g, punctured, and fractions of 5 drops each were collected.

number of f ractions(3 drops each)

FIG, 4. Measurement of the density of PHB-containing and PHB-free whole cells (o-o) and of PHB granules (•■-•-•) by equilibrium centrifugation in a CsCl gradient ( ...... ). The data for PHB granules were obtained in a second run. The cells were suspended in a semisaturated CsCl solution; the density gradient was obtained by a 20-h run at 218 000 g. The tubes were punctured, and fractions of 3 drops each were co llected . IAEA -S M -134/9 227

SEPARATION BY MEANS OF A LINEAR SUCROSE GRADIENT

The sucrose gradient was prepared as described above and its density was determined by refractrometric measurements. After centrifuging for 120 min the PHB-deficient cells exhibited a peak at 73% sucrose and the PHB-rich wild-type cells a peak at 62% (Fig. 3). In samples taken from both bands only 0.5% wild-type cells were found among the PHB-deficient mutants and 12% mutants among the wild-type cells. As has been shown with CsCl-equilibrium centrifugation, the rate of sedimentation of both types of bacteria in the sucrose gradient is mainly due to the difference in density (Fig. 4). The position of the bands corresponds to a density of 1.350 g/ml for the PHB-deficient mutant, РНЕГ4, and of 1.244 g/ml for the PHB-containing wild type. The low density of the wild-type cells is due to the inclusions of PHB. The value of 1.244 for the PHB-containing wild type does not agree with the density (1.3277 g/ml) of PHB-grana obtained from PHB-rich cells by Na- hypochlorite treatment. Whether an artefact due to complex formation with Cs-ions was involved remains uncertain.

SEPARATION BY MEANS OF A. DISCONTINUOUS GRADIENT

On the basis of data obtained from linear gradients, a discontinuous gradient was prepared by filling the centrifuge tube with successive layers of different concentrations of sucrose solution. First with 2 ml of 80% su c ro se solution, then 5 m l of 65% solution, 4 m l of 50% and finally with 2 ml of 30% solution. While the lighter wild-type cells were retained at the top of the 65% sucrose solution, the heavier mutant cells accumulated at the top of the 80% layer. The incidence of contamination with the other cell type in each band was similar to that after separation in the linear gradient.

APPLICATION TO OTHER BACTERIA

Linear sucrose-gradient separation was used on Pseudomonas aeruginosa, Bacillus subtilis and Azotobacter vinelandii to test the separation of the log-phase cells from the cells rich in reverse materials. After centrifugation in a 90-30% linear sucrose gradient A. vinelandii exhibited the typical separation pattern shown in Fig. 5. Centrifugation of both cell types in separate centrifuge tubes confirmed these results. On the basis of our results, this centrifugation method appears to be applicable to other genera for the separation of cells of different densities.

ISOLATION AND YIELD OF MUTANTS OF STRAIN H16

After enrichment by sucrose-density centrifugation and plating on nitrogen-poor fructose agar, an average of 2% of the growing colonies were found to be PHB-deficient mutants. Forty-nine independent mutants 228 SCHLEGEL and OEDING

5 10 15 20 25 30 number of fractions

FIG. 5. Separation of PHB-rich and PHB-poor cells of Azotobacter vinelandii by sucrose density-gradient cen trifu g atio n (ЗО-ЭО^о sucrose; 90 m in at 30 000 g).

t(m in)

FIG. 6. Rates of hydrogen oxidation by washed cells of the wild type H16 and of the PHB-deficient mutant PHB"4 in the presence (solid lines) and absence (broken lines) of carbon dioxide. The vessels contained washed cells which had been harvested in the log-phase (open symbols) or which had been incubated, after washing, for 24 h under storage conditions (full symbols). They were filled with a gas mixture of either 80°]o H2 + 10°/o O2 + IÇPjo CO2 or with a СОг-free hydrogen-oxygen mixture; in the latter case the centre vessel contained 0. 2 ml of 20°}o KOH. IAEA-SM-134/9 229

t(h)

FIG. 7. Incorporation of radioactive carbon dioxide by wild-type cells and by mutant cells in a nitrogen-free mineral solution under a hydrogen-oxygen atmosphere. The vessels contained washed cells which had been harvested in the log-phase (open symbols) or which had been incubated, after washing, for 24 h under storage conditions (full symbols). were isolated; further assays showed that 32 of these mutants were "leaky" (i. e. they accumulated PHB at a diminished rate), whereas 17 mutants did not produce PHB at all. Seven of the PHB-negative mutants were tested to determine their growth rate with fructose as a substrate. There were no significant differences in the growth rates of the mutants compared with that of the wild-type strain.

Respiratory control in PHB-negative mutants

The rate of hydrogen oxidation by washed, autotrophically-grown cells of the wild-type strain HI 6 in the absence of a nitrogen source is influenced by carbon dioxide. In the presence of C0 2 the oxidation rate is almost three times as high as in its absence. This phenomenon has been tentatively explained as a control of the electron transport by limiting concentrations of ADP which, in the presence of CO 2, is regenerated by C02-fixation and PHB-synthesis. The availability of PHB-negative mutants made it possible to test this hypothesis (Fig. 6). In these experiments the stimulation factor for the wild-type cells was found to be 2.3, for the mutant РНЕГ4 it was 2.1, although the absolute rates were less than with the wild type. Absolute oxidation rates exhibited 230 SCHLEGEL and OEDING by the mutants agree with those obtained with the wild type when tested after a 24-h incubation period under autotrophic storage conditions. Three mutants were checked in this manner. Although the mutants are not able to synthesize and accumulate PHB, hydrogen oxidation was subject to a stimulation by carbon dioxide (average stimulation factor 2.1). These results do not support the hypothesis outlined above, and further studies are required. All mutants tested so far incorporate radioactive carbon dioxide in the absence of a nitrogen source. The rate of incorporation is similar to that of wild-type cells which previously had accumulated PHB to their maximum storage capacity (Fig. 7). In summary, the results obtained so far indicate that sucrose density gradient centrifugation is the method of choice for the enrichment of lipid-free mutants. The growth rate of the PHB-negative mutants with fructose does not markedly differ from that of the wild-type strain Hydrogenomonas eutropha. There is no reason to assume that lipid-free mutants cannot be used for the production of single-cell protein.

REFERENCES

[1] SCHLEGEL, H. G., CLAUS, D., LAFFERTY, R. M ., Mikroorganismen im Dienste del menschlichen Ernährung, Zbl. Bakt. I Orig. 2 1 2 (1969) 303. [2] SCHLEGEL, H. G ., "From electricity via water electrolysis to food", Fermentation Advances (PERLMAN, D .. Ed.), Academic Press, New York (1969). [3] SCHLEGEL, H. G ., LAFFERTY, R. M ., KRAUSS, I., The isolation of mutants not accumulating poly-ß-hydroxybutyric acid, Arch. Mikrobiol. 71 (1970) 283.

DISCUSSION

J. MEYRATH: I would like to know if the stoichiometric relationship is retained between hydrogen and carbon dioxide utilized and the cell material produced under the growth conditions of the organism? H. G. SCHLEGEL: The H2/C 0 2 ratio depends on the growth, rate and on the type of limiting nutrient. It varies from 4.5 to 9 and is most economical when cells grow at their highest rate. Furthermore, it is lower when oxygen limits growth and less economical when growth is limited by hydrogen or carbon dioxide. J. MEYRATH: What are the technological difficulties in supplying the organism with sufficient hydrogen? To what extent does the transfer of the rather insoluble hydrogen gas present difficulties? H. G. SCHLEGEL: The interactions of gases like hydrogen, nitrogen or oxygen with water are well known. Since no data on the ratio of the diffusion of oxygen to that of hydrogen were available, experiments were performed in order to adapt the ratio of the diffusion rates to that of the consumption rates (by the cells). A. gas mixture consisting of 25% oxygen, 10% carbon dioxide and 65% hydrogen proved to be the optimum. J. MEYRATH: Is there a possibility of changing the accumulation of nitrogen-free storage material by varying the supply of the nitrogen source? IAEA-SM-134/9 231

H. G. SCHLEGEL: Under conditions of mass culture (enzymes), suppression of storage material synthesis by increasing the concentration of the nitrogen source appears to be impossible, since for high cell clusters other factors (e. g. oxygen) limit growth and stimulate the synthesis of sto rag e m aterial. Z. HOSÎALEK: I would like to add a short comment on the significance of metabolic paths connected with the formation of reserve compounds for production of industrially important microbial metabolites. In the case of oligoketide ("derived from acetate") compounds, e. g. tetracyclines, there exists a close relationship between energy metabolism, subsequent formation of storage compounds (polyphosphates) and production of antibiotics. In the standard-type strains of Streptomyces aureofaciens we encounter high activity of the tricarboxylic acid (TCA) cycle and formation of polyphosphates. The production of tetracycline is low. In the case of high-production mutants, on the other hand, the high production of antibiotics continues and no polyphosphate accumulates at all. The low activity of the TCA cycle and of the respiratory chain and therefore the damaged formation of reserve compounds make possible the preferent utilization of acetyl CoA. for biosynthesis of tetracene nonaketide. HUGUETTE de ROBICHON-SZULMA.JSTER; Do you think that the density gradient technique you have used can also be applied to organisms which accumulate glycogen granules or other carbohydrate compounds? H. G. SCHLEGEL: The sucrose density gradient technique has been applied in order to isolate bacterial endospores lacking Ca-dipicolinate; these spores are much lighter than the wild-type spores. No experiment on the separation of glycogen-less cells is known. However, since glycogen has a higher specific weight than the cell, it should be possible to separate glycogen. Glycogen-deficient mutants of Escherichia coli have been isolated without enrichment by Preiss.> R. HÜTTER: Do mutants and the wild type have the same cell size? I ask this question because sucrose density-gradient centrifugation is used with Saccharomyces cerevisiae to separate large and small cells. H. G. SCHLEGEL: No data are available on this point. According to microscopic observations, the mutant cells do not differ from wild­ type cells without storage material. R. HÜTTER: Are the mutants revertible, i. e. are they caused by single mutations? H. G. SCHLEGEL: No revertants have been isolated so far. R. HÜTTER: Do you have additional information on the nuclear state of your organism other than the u. v. -killing curve? H. G. SCHLEGEL: No.

IAEA-SM-134/28

COMPLEXITY OF GENETIC CONTROL OF BIOCHEMICAL PROCESSES IN FUNGI AS EVIDENCED BY STUDIES ON RESISTANCE TO TOXICANTS

VASSHIKI VOMVOYANNI, S. G. GEORGOPOULOS, A. KAPPAS N. R. С . " Democritus ", Athens, Greece

Abstract

COMPLEXITY OF GENETIC CONTROL OF BIOCHEMICAL PROCESSES IN FUNGI AS EVIDENCED BY STUDIES ON RESISTANCE TO TOXICANTS. The recognition of all the genes affecting a biochemical process may present difficulties in many cases. Such difficulties are not encountered with mutations for resistance to toxic compounds, all of which can easily be selected for and recognized. Work in our laboratory with a number of fungitoxicants indicates that multigenic systems control sensitivity in fungi. These systems may involve multiple allelomorphs and modifiers in addition to the main resistance genes which may or may not show interaction when present in the same genome. Because a change in sensitivity reflects some modification of the respective cellular process, the complexity of the genetic control of the former may give a measure of the possibilities for modification of the latter.

It is generally agreed that the understanding of the co-ordinated function of the complex metabolic processes is of great importance for the practical utilization of cell activities, such as in the fermentation industries. This understanding can best be achieved through combined genetic and biochemical studies, and by this approach several of the fundamental biosynthetic pathways and regulatory mechanisms are now being elucidated. This quite successful methodology is often severely handicapped because of the difficulty in selecting suitable mutants. This is especially true for cell functions such as protein synthesis, DNA replication and transcription. In these cases, mutated phenotypes are very often lethal and cannot be isolated because it is difficult to substitute for the normal products of these processes during the selection procedure. Indirectly, the problem of isolating suitable mutants can often be solved through the selection of mutants with altered sensitivity to inhibitors which are selective for the process studied. The value of this "sensitivity to inhibi­ tor approach" has been well demonstrated in the studies on the antibiotic resistance in bacteria, on which a large amount of knowledge, especially on protein biosynthesis, has been based. In this paper we shall try to demonstrate how complex the genetic control of biochemical processes appears to be in the fungi, judging from genetic systems controlling sensitivity to fungicides. We shall refer to three polygenic systems for resistance to fungicides on which we have worked: one for resistance to what has been named "the aromatic hydrocarbon group" [ 1 ], one for resistance to n-dodecyl- guanidine acetate (dodine) and one for resistance to cycloheximide (Actidione).

233 234 VOMVOYANNI et al.

Of these three types of fungicides only cycloheximide is well established as a selective inhibitor of one process, namely protein biosynthesis in eukaryots [2]. For the aromatic hydrocarbons the issue does not appear to have been resolved [3], while dodine has been shown to act by disrupt­ ing membrane structure and inhibiting a number of membrane-bound enzymes [4, 5]. Implications of the genetic data on resistance on the understanding of the cellular processes involved are discussed with respect to cycloheximide, and speculations regarding the other two polygenic systems are attempted.

DESCRIPTION OF THE POLYGENIC SYSTEMS

The genetic control of resistance was studied in Neurospora crassa for cycloheximide, and in Nectria haematococca (Hypomyces solani) for aromatic hydrocarbons and dodine. Mutants resistant to cycloheximide and to dodine were selected after u.v. or gamma irradiation of conidia and plating on agar medium containing appropriate concentrations of the respective toxicant. With the aromatic hydrocarbons, resistant mutants are obtained as fast-growing sectors from colonies exposed to one of these compounds. Random ascospore analysis of crosses to compatible wild- type strains was used to establish the genetic nature of resistance in each case. Allelism was tested by similar analysis of at least 20 ascospores from each mutant X mutant cross. Table I shows the results of such tests for 60 mutants of N. crassa resistant to cycloheximide. As shown, resistance was found to result from mutation at one of at least four chromosomal loci, actj_4, two of which, act-1 and act-2, were known from previous studies [6 ]. It is also shown that mutation at one locus (act- 1 ) might be more frequent than at an other. Similar results were obtained with 100 mutants of N. haematococca resistant to the aromatic hydrocarbon fungicides (cnb mutants) which have been grouped [7] in five classes (loci cnbj_5). With dodine, 1 2 resistant mutants were studied and of these six resulted from mutation at the dod-1, four at the dod-2, one at the dod-3 and one at the dod-4 locus [8 ]. It is thus shown that in all three cases many genes are directly involved in the sensitivity of growth and hence of cell activities to the toxicant.

TABLE I. CLASSIFICATION OF 60 СYCLOHEXIMIDE-RESISTANT MUTANTS IN Neurospora crassa ACCORDING TO THE MUTATED LOCUS AND TO THE LEVEL OF RESISTANCE

L evel of Genetic group resistance (M g/ml) a c t-1 a c t-2 a c t-3 Others

100 17 4 2

40 15 1 1

20 8 1

5 6 1 4 IAEA-SM-134/28 235

TABLE II. RESPONSE OF SEGREGANTS FROM MUTANT X WILD-TYPE CROSSES ON DIFFERENT CYCLOHEXIMIDE CONCENTRATIONS

Mutant No. S ensitive Tolerant segregants classified O riginal G en etic crossed to segregants on the basis of tolerance level resistance group w ild type 5 jig/ml 20 fig/ml 40 jig/ml 100 \ig /m l ST4 cycloheximide

IX 12 7 6 5 a c t - 1

19 10 4 4 20 a c t-1

68 ? 6 2 20 a c t-1

73 6 3 4 20 a c t-2

125 15 10 9 20 a c t-3

As can be seen in Table I, resistance to cycloheximide can be of one of four different levels. This was determined by testing each mutant for ability to grow on a number of toxicant concentrations in agar medium. Differences in the level of resistance between mutants of the same genetic group can be explained on the basis of either allelomorphs of the resistant genes or modifying genes at other loci. By determining the resistance level of ascospore progenies from 30 mutant wild-type crosses the presence of modifiers has been established in five cycloheximide-resistant mutants as shown by the data in Table II. Twenty-five mutants, on the other hand, were shown to be monogenic and of these 11 were act-1. The latter were divided in two subclasses according to the resistance level (40 and 100 Mg of cycloheximide per ml of medium respectively) and, obviously, this difference is due to two different forms of the same resistant gene. With dodine, four different resistance levels, one for each resistance gene, were recognized. With the small number of mutants that were studied, no evidence of allelomorphs was obtained, but in two instances the presence of modifiers was established [8 ]. The situation is quite different with the aromatic hydrocarbons. In spite of the five loci for resistance that were recognized, and of the large number of mutants studied, resistance was always of practically the same level on the basis of both spore germination and hyphal growth [9]. From the viewpoint of interaction of resistant genes when more than one are brought into the same haploid nucleus in recombinants, additivity was found in the case of dodine resistance [8 ]. With cycloheximide, positive interaction has been observed by Hsu [6 ] between act-1 and act-2, but this point was not extensively studied in our work. It should be noted that the double-resistant recombinant studied by Hsu was capable of very limited growth in the absence of the antibiotic. With the aromatic hydrocarbons, the double mutants were found phenotypically indistinguishable from the single tolerant strains [1 0 ].

BIOCHEMICAL IMPLICATIONS

In the fungi, practically nothing is known on the genetic control of protein biosynthesis, which is apparently of great importance from the viewpoint of practical utilization of microbial activities. The following 236 VOMVOYANNI et al. discussion first examines how the recognition of the polygenic system for resistance to cycloheximide in N. crassa, as described in the previous chapter, might be utilized for the understanding of the genetic control of the process affected by this toxicant. In the cyclic process of the m-RNA-directed formation of successive peptide bonds on 80S ribosomes, cycloheximide selectively inhibits movement of petidyl-tRNA from the "acceptor" to the "donor" site [11]. The enzymatic fraction TF-II, normally present in the soluble phase, is required for this movement for which it is well documented that the formation of a GTP-ribosome-TF-II complex, and the hydrolysis of GTP into GDP and Pi are prerequisite steps [12]. Though it has been proposed that cycloheximide inhibits protein synthesis by action on the TF-II, the in-vitro experiments done so far have localized resistance on the ribosomal fraction [13, 14]. This is not unexpected, of course, because the ribosome must actively participate in the TF-II activities. Four single-gene mutants of N. crassa, one for each of the cyclo- heximide-resistance loci recognized in our work, were studied as to the effect of cycloheximide on protein synthesis in cell-free systems. In all cases high-level resistance in vitro was found, and by appropriate hybrid systems is was shown that the mutations affect the ribosomal fraction [15]. These mutations must affect some vital aspects of the function of the ribosome and not only the cycloheximide-ribosome inter­ action. The best evidence in support of this view is the poor growth of the double resistant homokaryon (act-1, act-2) described by Hsu [ 6 ], which apparently results from reduced efficiency of the protein synthesizing system. For protein synthesis to proceed at a normal rate it appears that all of the act genes except, perhaps, one, must be in the wild-type form. Since all four types of mutants were selected on a specific inhibitor of translocation, it is most likely that translocation is the step affected by these mutations although direct assay has not been done as yet. From what has been stated so far, it seems that the genetic control of trans­ location in N. crassa must be quite complex. Firstly, because at least four genes affecting cycloheximide sensitivity appear related to translocation, and some of the modifying genes recognized may also act at this step; secondly, because translocation may undoubtedly be affected through mutational modification of TF-II independent of the ribosome; and thirdly, because, even if we consider the ribosome only, we cannot exclude the possibility of ribosomal mutations affecting translocation independent of sensitivity to cycloheximide. It cannot yet be understood how the available genetic information on the resistance to cycloheximide can help in the elucidation of structure- function relationships on the ribosome, because basic knowledge on this aspect is still quite limited [16]. It should be emphasized, however, that the genetic complexity of ribosomal functions, as indicated from the data given in this paper and also by similar studies on resistance to protein - synthesis inhibitors in bacteria [17], must supply the protein-synthesizing system with stability and hence leave small possibility of genetic improve­ ment in this respect. With dodine and the aromatic hydrocarbons, resistance has not been demonstrated in cell-free systems. It is, therefore, entirely possible that the mutations cause reduced permeability of cytoplasmic membranes to the toxic molecule. Threlfall [18], in fact, has obtained some experi- IAEA-SM-134/28 237 mental evidence indicating that in Aspergillus nidulans mutational resistance to some aromatic hydrocarbon derivatives results from reduced uptake of the toxicant. If all five cnb genes for resistance to aromatic hydro­ carbons in N. haematococca which were mentioned in the previous chapter affect permeability then the genetic control of this process must be a complex one. The polygenic system for resistance to dodine may also indicate complex genetic control of some property of the membranes because the membrane-dodine interaction appears [4, 5] to be the basis for the fungitoxicity of this compound. It is entirely possible, however, that at least some of the dod genes may affect the dodine detoxification which has been observed in a related fungus [19].

REF ERENC ES

[1] GEORGOPOULOS, S. G ., ZARACOVITIS, C ., Rev. Phytopath. 5 (1967) 109. [2 ] SIEGEL, M .R. . SISLER, H .D ., N ature 200 ( 1963) 675. [3] GEORGOPOULOS, S.G ., ZAFIRATOS, C .. GEORGIADES, E., Physiologia PI. 20 (1967) 373. [4 ] BROWN, I. F ., SISLER, H .D ., Phytopathology 50 (1960) 830. [5] SOMERS, E., PRING, R.J., Ann. appl. Biol. 58 ( 1966) 457. [6] HSU, K.S., J. gen. Microbiol. 32 (1963) 341. [7] GEORGOPOULOS, S. G., PANOPOULOS, N .J., Can. J. Genet. Cytol. 8 (1966) 347. [8] KAPPAS, A ., GEORGOPOULOS, S. G ., Genetics 66 ( 1970) 617. [9] VOMVOYANNI, V ., GEORGOPOULOS, S. G ., Phytopathology 56 ( 1966) 1330. [10] GEORGOPOULOS, S. G ., Phytopathology 53 ( 1963) 1086. [11] McKEEHAN, W ., HARDESTY, B., Biochem. biophys. Res. Commun. 36 (1969) 625. [12] LIPMANN, F., Science 264(1969) 1024. [1 3 ] SIEGEL, M .R ., SISLER, H .D ., B iochim . biophys. Acta 103 (1965) 558. [14] COOPER, D., BANTHORPE, D. V ., WILKIE, D., J. molec. Biol. 26 (1967) 347. [15] VOMVOYANNI, V., to be published. [16] KURLAND, C.G ., Science 169 (1970) 1171. [17] SCHLESSINGER, D., APIRION, D., Rev. Microbiol. 23 (1969) 387. [18] THRELFALL, R.J., J. gen. Microbiol. 52(1968) 35. [19] BARTZ, J. A ., MITCHELL, J.E ., Phytopathology 60 ( 1970) 350.

DISC USSION

K. MIKULIK: I should like to ask you a few questions. First, did you analyse the system (or systems) which is (or are) responsible for resistance to Actidione? Second, did you find any differences in ribosomal proteins isolated from the parent strain and from your cycloheximide- resistant strains ? And, lastly, was your in-vitro protein-synthesizing system directed by endogenous mRNA or poly U? VASSILIKI VOMVOYANNI: Mutations at the act 1 -4 loci resu lt in resistance in cell-free systems, and we can conclude that they control the elements of the protein synthesis apparatus. It may be possible, on the other hand, to isolate resistant mutants which will differ from the wild type in the permeability of the cell membrane to cycloheximide or in the ability to detoxify the molecule. As regards your second question, we extracted ribosomal proteins by the LiCl-urea method and analysed these proteins on polyacrylamide gels. We were able to identify 25 bands. 238 VOMVOYANNI et al.

At this resolution no difference was observed between the wild type and the resistant ribosomes. Lastly, the resistance was studied in vitro in systems directed by either endogenous mRNA or poly U. The same results were obtained in both cases. Of course, we don't know to what extent our endogenous mRNA-directed cell-free systems can initiate new peptide chains, and therefore questions on the possibility of cycloheximide differentially affecting initiation and elongation by sensitive and resistant systems have not been answered. MUTATION AND SELECTION OF INDUSTRIALLY USEFUL MICROORGANISMS (Session 6)

Chairman

K. ESSER (Federal Republic of Germany)

IAE A-SM -1 3 4 /2 1

TECHNIQUES FOR THE DEVELOPMENT OF NOVEL MICROORGANISMS*

H. I. ADLER Oak Ridge National Laboratory, Oak Ridge, Tenn., United States of America

Abstract

TECHNIQUES FOR THE DEVELOPMENT OF NOVEL MICROORGANISMS. This paper is directed towards the non-specialist. It is an attempt to discuss some novel approaches for the production and maintenance of microbial mutants. A short discussion of the use of radiation, both ionizing and non-ionizing, as a mutagenic agent is presented. Likewise, some of the more useful chemical mutagens are briefly discussed. Attention is called to the existence of anti-mutagens. Emphasis is placed on the use of various techniques that allow the microbial geneticist to recombine properties of various mutants in one organism. This latter point is illustrated by the use of an example involving morphological mutants of Escherichia coli. References to recent monographs and reviews of chemical and radiation mutagenesis are in clu d ed .

INTRODUCTION

In this paper I will touch only lightly on areas that are well covered in recent reviews of microbial genetics or which I suspect will be discussed by others at this symposium. I will try to point up some less common ideas that may at some future time be applicable to situations encountered in the fermentation and allied industries. I will direct my remarks only partially to bacterial geneticists. Primarily I am trying to reach those who may be in a position to influence approaches to the use of microorganisms in fermentation and other industrial applications.

THE USE OF RADIATION AS A MUTAGENIC AGENT

Although numerous reports exist demonstrating that ionizing radia­ tion can produce mutations in a variety of microorganisms, little can be said about the mechanisms involved and for most organisms the approach is of limited usefulness as a practical tool for the production of mutant forms. An exception might be in a case where the organism does not allow penetration of u.v. light or chemicals to DNA. In such cases the ability of ionizing radiation to penetrate deeply into biological materials is an advantage. With regard to mechanisms I can only say that it is clear that both single and double strand breaks are produced by ionizing events in DNA. We should bear in mind that sources of ionizing radiation are still relatively unavailable in most laboratories and require elaborate shielding in order to protect the investigator. It is for this reason perhaps that not much is yet known about the mutagenic effects of ionizing radiation on micro­ organisms . There simply are not enough people with the appropriate facilities and interest.

* Research sponsored by U. S. Atomic Energy Commission under contract with the Union Carbide Corporation.

241 242 ADLER

The Situation with regard to ultra-violet, light mutagenesis is quite different. A large literature exists in this field, most of it centered around the effects of 2537 nanometer radiation obtained from a variety of low pressure germicidal lamps. For many years u.v. light was perhaps the most commonly used mutagen for studies with microorganisms. Currently there are many investigators attempting to unravel the molecular mechanisms involved. For a recent review see Drake [l]. The study of ultra-violet mutagenesis is made more complicated and also more interesting as a result of the discovery that microorganisms have systems capable of repairing some fraction of the ultra-violet damage that would otherwise lead to mutation. There are several such repair systems and several potentially mutagenic molecular lesions. Because of the existence of repair systems, the yield of mutations observed after ultra-violet treatment of a bacterial suspension depends in large measure on the post-irradiation treatment of the cells. For example, for many strains of Escherichia coli the mutation yield declines if the cells are held in a saline suspension after irradiation and before a period of growth and division. A number of mutant strains incap­ able of carrying out various steps in the repair process have been isolated. Some of these have been discussed earlier [2]. The ability or inability of a strain to carry out repair very much influences the yield of mutants obtained after treatment with ultra-violet light.

As a practical tool for the production of mutants in microorganisms, ultra-violet light has much to recommend it. Sources of radiation are relatively inexpensive and easily handled. The mutagenic agent is not very dangerous for the experimenter and requires only reasonable precau­ tions for the shielding of sensitive tissues. Furthermore, the fact that many investigators are currently determining the mechanisms involved in u.v. mutation, makes it likely that those interested in using it as a prac­ tical tool in the future will have a firm base on which to operate. One example of this suggests itself to me at the present time. I have mentioned that in Escherichia coli and a few other organisms mutants lack­ ing the ability to repair ultra-violet DNA have been isolated. In some of these mutants the ability to propagate is unusually sensitive to ultra­ violet radiation, and among these sensitive strains a certain fraction have a greatly enhanced u.v. mutability. For example, strains lacking the ability to excise pyrimidine dimers exhibit an enhanced u.v. mutability. I emphasize that only certain u.v.-sensitive strains exhibit this enhanced mutability. For example, strains that are unusually sensitive to u.v. because of a deficiency in their recombination mechanism are resistant to u.v.-induced mutagenesis, and, in fact, some of these are extremely stable to u.v. and one cannot observe mutations induced by any dose. For a recent discussion of this subject see Witkin [3]. It should be clear that an understanding of the genetic makeup of a given strain with regard to its ability to carry out repair or undergo recombination will prove useful to the experimenter desiring to induce mutations with ultra-violet light. Such genetic information is rapidly becoming available for a number of species.

CHEMICAL MUTAGENESIS

In recent years a large variety of chemicals have been used in order to induce mutations in microorganisms. Most of the interest in these chem­ icals stems from the fact that they are proving useful in unraveling the IAEA-SM “134/21 243 molecular changes that occur during the mutation process. I will not attempt a detailed review of the currently popular chemical mutagens, hut refer you to a recent book entitled, "Chemical Mutagens," by Fishbein et al [U]. I will, however, make a few comments about the major classes of currently popular chemical mutagens and call to your attention some chemical mutagens popular in an earlier period that still may be useful in situations where one is more interested in obtaining large numbers of mutants than in working out the molecular details of the mutation process.

Nitrous Acid

Nitrous acid is a powerful mutagen in viruses, some bacteria, yeast, and Neurospora. Most evidence suggests that the mutations induced are of the transition type, i.e., changes of one of the nitrogenous bases in DNA to another; for example, a transition from adenine to guanine, or cytosine to uracil. In a practical framework then, nitrous acid can be a useful mutagen in cases where one is interested in affecting the genetic material at a point in the molecule without major disruption of adjoining sites.

Base Analogues

Several compounds that can be incorporated into DNA in place of the normally occurring nitrogenous bases are mutagenic. Five-chlorouracil, 5-iodouracil, and 5-bromouracil are amóng the more common ones. Bromo- deoxyuridine, may be even a better mutagen than bromouracil, apparently because it competes more effectively for the position normally occupied by thymine than does bromouracil. As a group these compounds are useful mutagens for a variety of microorganisms.

Acridines

The acridines have recently been used in efforts to understand the process of mutagenesis in microorganisms. In at least some organisms, such as Aspergillus, acridines tend to induce unstable mutants, possibly as a result of interference with chromosome segregation during cell division. In at least some organisms, the acridines are effective in inducing frame shift lesions. In using acridines as mutagenic agents, one must be cog­ nizant of the fact that the acridines can sensitize cells to the mutagenic action of visible light. It is, therefore, necessary to avoid visible light if one is interested in studying the direct mutagenic actions of these compounds.

Alkylating Agents

The alkylating agents, such as mustard gas, have been studied in a variety of organisms. For a review see Loveless [5]. They are effective mutagens in many microorganisms. Some of the more frequently used ones are ethylmethane sulfonate, ethylethane sulfonate, and diethyl sulfate. The first chemical mutagen to be discovered was mustard gas, di-(2-chloro- ethyl) sulfide. A large literature relating the action of these mutagens to their known chemical interactions with nucleic' acids exists and will continue to grow. N-methyl-N'-n'itro-N-nitrosoguanidine is certainly an alkylating agent but seems to differ in its mechanism of action from the others. I call it to your attention primarily because it is a very potent mutagen for many microorganisms. 244 ADLER

Some Miscellaneous Mutagens

In this short section I call to your attention the existence of some mutagenic agents that are not currently under careful study, but are useful in a practical sense for the production of mutants. Some of them have the advantage of not being as dangerous for the investigator as those I have just discussed. For example, manganous chloride studied by Demerec & Hanson [6] can be a useful mutagen. Also I remind you that hydrogen peroxide either alone or in combination with various organic materials can be a useful mutagen. A discussion of some of these mutagens for which we have little understanding of mechanism can be found in Braun's "Bacterial Genetics" [7]. I do not claim to have made an exhaustive survey of the currently available chemical mutagens. The references contain much more interesting detail.

Antimutagens

A few reports have appeared indicating that some compounds can reduce the raté of both spontaneous and induced mutations. For example, Novick [8] reported that adenosine, guanosine and inosine could lower the spontaneous mutation rate in ]S. coli about threefold. More recently Puglisi [9] observed that actinomycin D and basic fuchsin are antimtitagenic in Saccharomyces cerevisiae. Curiously, some compounds that are anti- mutagenic can also act as mutagens in a different environment. At this time little can be said about the mechanism of antimutagenesis, and not many groups are attacking the question directly. Clearly, this is an important subject worthy of attention both from the point of view of those who are interested in maintaining important mutant strains and those who are interested in the details of mutagenesis. A powerful antimutagen would be an extremely valuable practical and analytical tool.

SELECTIVE TECHNIQUES

I hesitate to discuss this topic because it is very obvious to most workers that if a powerful technique for the isolation of the desired mutant is available, it should most certainly be employed. It is clear that a wise investment of research time is often made in devising highly selective techniques because once such a technique is available the ques­ tion of what mutagenic agent to use may become immaterial, and it is often possible to isolate the desired microorganism simply making use of the spontaneous background of mutations. Perhaps the best selective tech­ niques are those which allow the growth of the mutant form in some environment which completely inhibits the growth of the original parental culture. Less powerful techniques are those which allow the identifica­ tion based on colony or cell morphology, the production of a pigment, or some similar phenomenon of the mutant form in a large population of parental types.

Although it is not entirely appropriate for discussion in this section, I want to point out the possible usefulness of a special class of mutants, the conditional mutants. These mutants, which are obtained by the use of special selective techniques, are currently being studied in many laboratories [10, 11]. The most common of these are conditional temperature mutants which are isolated under selective conditions that favor, organisms in which the mutant character can only be expressed in a defined temperature range. One can see with relative ease that conditional mutants might have application in the fermentation industry. IAEA-SM-134/21 245

THE USE OF RECOMBINATION AND OTHER TECHNIQUES IN THE CONSTRUCTION OF DESIRED MUTANTS

A large amount of effort has been spent in understanding the phenomena of transformation, transduction, and conjugation. Most of this work has concentrated on a few species of microorganisms. In my opinion, a very reasonable long-term investment in research would be to explore these phenomena in a greater range of species, particularly those species that are important in the fermentation industries.

One of the most powerful approaches to the creation of organisms with specialized genotypes is the production of recombinants of pre­ existing mutants. I will restrict my remarks to some illustrations in Escherichia coli since I am most familiar with this bacterium. In any laboratory devoted to the genetic study of E. coli, and there are many, the production of strains with specified genotypes occupies a large fraction of the ingenuity and effort; and exquisite techniques have been developed. In the case of IS. coli, the applicable approaches are trans­ duction and conjugation. Transduction is a particularly useful technique if one is interested in changing the genotype of a very short region on the bacterial chromosome without introducing new genetic material at other locations. In this technique a bacteriophage is used as the vector by which a short region of desired genetic information is carried from one strain into another. A large number of bacteriophages capable of carrying out transduction is available. Some of these bacteriophages are useful for many loci of the genome whereas others can be used only in connection with the transfer of specific loci. A discussion of transduction, trans­ formation and conjugation is readily available in almost any modern text of genetics. For a detailed presentation I suggest an examination of "The Genetics of Bacteria and Their Viruses" [12]. The phenomenon of phage- mediated transduction has been reported for Escherichia coli, several Shigella species, Pseudomonas species, Staphylococcus, Proteus, and Bacillus subtilus■ Clearly it can be a powerful tool in the production of recombinants with desired characteristics, and time invested in finding the appropriate phages and conditions for other species of microorganisms might pay off handsomely.

Conjugational methods are often used for Escherichia coli■ In this technique cells of two strains, each having some part of the desired genotype, are mixed under conditions where DNA is transferred from one cell to the other, and the desired recombinant class is then isolated by the application of selective techniques. I will attempt to illustrate the use of this technique by one of our own studies. In this illustration, which will be accompanied by a 16 mm film, I will make use of morphological mutants and show how the unusual properties of two mutants can be combined into a single organism which has never been isolated directly from a natural environment nor is it likely to ever arise as a result of the application of a mutagen to a wild-type population of Escherichia coli. In passing, I would like to point out that morphological mutants may not be of the greatest interest to those who are concerned with industrial fermentations, but they should not be dismissed altogether. Some of the mutants I will discuss are of such unusual size and form that their behavior in centrifugation and fil­ tration techniques is different from that of their wild-type parent and thus might introduce new parameters in industrial processes in which these tech­ niques are employed for the removal of microorganisms.

The first mutant of interest to us in this context is E_. coli K12 AB1899- It was isolated after exposure of the wild-type K12 strain to ultra-violet light [13]. Among the survivors of this exposure was a clone 246 ADLER that had the following characteristics. It was, and. is, sensitive to ultra-violet light, X-rays, a variety of chemicals, and tends to form long multinucleate filaments after exposure to any of these agents. These fila­ ments represent the equivalent of as much as 200 to 300 cells without cross­ walls separating the individuals. We have suggested that the formation of these filaments results from the fact that the mutation in AB1899 has enhanced the sensitivity of the cross-septation mechanism to a variety of agents [lU]. The growth mechanism has not been similarly sensitized and therefore these cells grow normally after exposure to injurious agents but fail to form cross-septa. A single genetic locus is involved in this phenomenon, and it has been mapped at a position close to the gene control­ ling sensitivity to the bacteriophage Тб. Since the normal mode of growth in these ]3. coli strains is by extension along the long axis of the cylindrical cell without much change in width, an extended period of growth without division gives rise to the very long filaments which are illustrated in the time-lapse film. During the growth of such filaments, DNA synthesis, protein synthesis, RNA synthesis, and various metabolic activities occur in a nearly normal way. Apparently only one of the last steps in cell division has been affected by mutation at this locus. We were curious as to how this mutant locus might express itself in different genetic backgrounds. In particular we wanted to determine its expression in a strain of Escherichia coli that we had isolated after treatment of wild-type with the chemical mutagen Triethylenemelamine (ТЕМ). This TEM- derived mutant is unusual in that its cells are not cylindrical as is usual for Escherichia coli, nor do they grow by extension along any particular axis. They grow by a process of expansion in all directions, similar to the inflation of a balloon. We hypothesized that, if the mutant gene which weakens the cross-septation mechanism could be introduced into this second mutant, we might be able to produce giant cells of a type previously not demonstrated. In order to produce the double mutant, we did the following. A strain of Escherichia coli which acts as a donor of genetic material in conjugation experiments and carries the mutant gene for cross- septation had earlier been isolated by Markovitz [15]. This strain can ferment the sugar lactose and is streptomycin sensitive. The gene control­ ling lactose fermentation in E. coli is very close to the gene controlling cross-septation. This donor strain was mixed with a recipient strain that produced the amorphous cells which grow by expansion in all directions. The recipient strain had previously been treated with a mutagen and a lactose negative streptomycin resistant mutant of it had been isolated. A mixture of the two organisms was plated on a medium in which lactose was the only carbon source and streptomycin was present. The recipient strain cannot grow under these conditions, nor can the donor. The only cells that can produce colonies on such a medium are those combining the streptomycin resistance of the recipient and the ability to ferment lactose introduced from the donor. Clones of such recombinant strains were isolated and tested for their ability to express the deficient cross-septation gene. We expected a reasonable fraction of the lactose positive recombinants to contain the deficient cross-septation gene due to the close linkage between the two genes. In our efforts to isolate the appropriate recombinant, we were aided by the fact that the gene controlling cross-septation also has an influence on colony morphology and lactose positive recombinants with the appropriate colony morphology are almost certain to have incorporated the gene leading to deficiency in cross-septation. When the double mutants isolated were exposed to radiation, our hypothesis was borne out. The individual ceils grow as is their normal mode, that is, by expansion in all directions, but fail to divide. Since this growth goes on for a long period of time, giant amorphous cells are formed. These are approximately 500 to 1000 times the volume of a normal wild-type E. coli. These structures, illustrated in the time-lapse film and the references [1 6 , 17], are not IAEA-SM-134/21 247 seen under any naturally occurring conditions, and in fact are not readily identifiable as 13. coli cells by the casual observer. They are larger than many mammalian cell types. It should be emphasized that these large structures are not viable in the sense of being able to reproduce. They will grow for several hours, but finally fail to increase in volume, stop the various macromolecular syntheses, and frequently lyse. I have used this particular illustration to emphasize the degree of control we can exercise over a microbial system by the application of a combination of mutational and recombinational approaches.

I would like to at this time call your attention to one other kind of morphological mutant that may be of use in an industrial setting. I do this because as a rule one does not think in terms of morphological mutants as having useful industrial application. Most attention is con­ fined to those mutants with altered biochemical properties.

The mutant I refer to is one which was isolated after triethylene- melamine treatment of wild-type J¡. coli ■ It regularly produces small anucleate cells during its growth and division. These "minicells" are produced by a division-like process occurring near the ends of the cylindrical IS. coli cells. The minicells are surrounded by normal walls and membranes and contain a sample of normal cytoplasm. They do not contain appreciable quantities of DNA. They are metabolically active and persist for long periods of time, but will not reproduce. Such minicells can be isolated from the parental population by density gradient techniques previously described [18]. Preparations of minicells of great purity can be obtained. It has occurred to us that they might be very useful in any situation where one wants a metabolically active population encased in a normal wall and membrane but does not want replication of the cell to take place. For example, minicells might make very interesting antigens since they could be introduced into a host animal in great quantity without any concern that they would replicate and kill the host. Preliminary indications are that such work might be profitable. A minicell-forming mutant of Salmonella typhimurium has been isolated and it has been demonstrated that minicells do indeed make an excellent antigen.

Although I do not know of any case where transformation has been used as a means of establishing microbial recombinants for industrial application, the possibilities have great appeal. A recent report indicates the potential for transformation as a means of producing recombinants in an organism used in the production of amylase and alkaline protease [19]- In transformation experiments one isolates the DNA of a microorganism from the cells and ma­ nipulates it independent of other cellular constituents. This means that mutagens, both physical and chemical, can be applied in a relatively simple environment. The treated DMA is then mixed with the recipient strain and mutant genes are incorporated at a reasonable frequency. Transformation techniques do not work with all strains of microorganisms, but the phenom­ enon has been demonstrated in a wide variety and certainly has great potential for the future since it is currently allowing detailed analysis of the molecular mechanisms of mutation and therefore will certainly form a sound basis for future application.

In this discussion I have attempted to make the point that producing mutants is only one of the valuable techniques we have at our disposal in microbial systems. Organisms of real potential can be created by recom­ bining the properties of existing mutants. The techniques for doing this have been highly developed in a few genera, but are lacking in many others; and as I have said, I think it would be very rewarding.to explore these phenomena more in genera of interest in the fermentation industries. 248 ADLER

REFERENCES

[1] DRAKE, J. W., The Molecular Basis of Mutation, Holden-Day, San Francisco (197O). [2] ADLER, H. I., "The genetic control of radiation sensitivity in bacteria," Ch. it, Vol. 2 Advances in Radiation Biology (Augenstein, L. G., Mason R., Zelle, M. R., Eds.), Academic Press, New York (1966). [3] WITKIN, E., Ultraviolet mutagenesis in strains of ]3. coli deficient in DNA polymerase, Nature New Biology 229 (1971) 8l. [It] FISHBEIN, L., FLAMM, W. G., FALK, H. L., Chemical Mutagens, Academic Press, New York (1970). [5 ] LOVELESS, A., Genetic and Allied Effects of Alkylating Agents, Pennsylvania State University Press, University Park (1966). [6] DEMEREC, M., HANSON, J., "Mutagenic action of manganous chloride," Cold Spring Harbor Symp. Quant. Biol. l6_ (1951) 215. [7] BRAUN, W., Bacterial Genetics, W. B. Saunders, Philadelphia (19 6 1). [8] NOVICK, W., "Mutagens and antimùtagens," Brookhaven Symp. Biol. 8_ (1956) 201. [9 ] PUGLISI, P. P., Antimutagenic activity of actinomycin D and basic fuchsine in Saccharomyces cerevisiae, Molec. Gen. Genetics 103 (1968) 2U8. [10] TAI, P., KESSLER, D., INGRAHAM, J., Cold sensitive mutations which affect ribosome synthesis, J. Bacteriol 9J_ (1969 ) 1298. [11] O'DONOVAN, G., KEARNEY, C., INGRAHAM, J., Mutants of Escherichia coli with high minimal temperatures of growth, J. Bacteriol. 90 (19б5) 6ll. [12] HAYES, W., The Genetics of Bacteria and Their Viruses, Blackwell Scientific Publications, Oxford (1968). [13] HOWARD-FLANDERS, P., SIMON, E., THERIOT, L., A locus that controls filament formation and sensitivity to radiation in Escherichia coli K-12, Genetics кЦ (196U ) 237- [ll+] ADLER, H., HAKDIGREE, A., Analysis of a gene controlling cell division and sensitivity to radiation in Escherichia coli, J. Bacteriol. 87. (196U ) 720. [1 5 ] MARKOVITZ, A., Regulatory mechanisms for synthesis of capsular polysaccharide in mucoid mutants of Escherichia coli K-12, Proc. Natl. Acad. Sei. U.S. 51 (196U ) 239. [16 ] ADLER, H., TERRY, C., HARDIGREE, A., Giant cells of Escherichia coli, J. Bacteriol. 9¿ (1968) 139- [1 7 ] ADLER, H., FISHER, W., HARDIGREE, A., Cell division in Escherichia coli, Trans. N. Y. Acad. Sei. Series 11 31. (1969) 1059- [18] ADLER, H., FISHER, W., COHEN, A., HARDIGREE, A., Minature Escherichia coli cells deficient in DNA, Proc. Natl. Acad. Sei. U. S. 57 (l96fl 321. [1 9 ] COUKOULIS, H., CAMPBELL, L., Transformation in Bacillus amyloligue- faciens, J. Bacteriol. 105 (1971) 319-

DISCUSSION

S. G. GEORGOPOULOS: Your filamentous mutant reminds me of the work of Bartnicki-Garcia and Nickerson on changing the growth of Mucor rouxii from yeast-like to filamentous by modifying the environment. This work showed clear-cut differences in the composition of cell walls between filamentous and yeast-like thalli. Have you examined a similar possibility in Escherichia coli? H.I. ADLER: We did look for differences in the cell walls of filamentous and normal forms, but did not find any. IAEA-SM-134/21 249

J. WEIJER: Would you not agree with the statement that the use of ionizing radiation as a mutational tool in industrial microbiology and genetics has certain advantages with regard to stability, in terms of reversion of the mutation induced? H. I. ADLER: I agree with you that there are certain special conditions in which ionizing radiation may be the mutagen of choice. P. DUPUY: In order to get an idea of the growth rate of normal fila­ mentous and irregular cells, I should like to know whether the various sequences in the film you showed during your presentation were taken at the same speed. H.I. ADLER: All sections of the film were taken at the same framing rate.. The growth rate of filaments, as measured by increase in length, is somewhat greater than the rate observed for a comparable population of normal-length cells. HUGUETTE de ROBICHON-SZULMAJSTER: You briefly mentioned that by using extracts or specific fractions of wild-type strains it is possible, at least in the case of the mutant which elongates without dividing, to obtain a phenotypic reversion of this abnormal process. Could you please elaborate on this point? H.I. ADLER: The material isolated from normal cells which can cause division of filaments is associated with a particle sedimenting at about 100 S. The particle is heat-labile, sensitive to lipases, phospho- lip ases, detergents and so m e proteases. It does not contain RNA or DNA. We are at present attempting to define these particles more precisely because they may play a part in the normal control of cell division. K. ESSER: Could you please say something about the physiological action of the genes responsible for morphology in your bacterial mutants? Are these blocks correlated with the carbohydrate metabolism responsible for the formation of the cell wall? H.I. ADLER: We have postulated that the mutation at the "ion" locus leads to a deficiency in the synthesis of a compound essential for cross wall and cross septum formation. As regards your second question, these compounds may in fact be normal components of the cell wall or membrane. C. EMBORG: You mentioned that some microorganisms could not be mutated by u.v. light or by chemical agents owing to lack of penetration by the mutagenic agents. Could you give examples of strains in which this observation applies? H.I. ADLER: I was referring to a theoretical possibility involving an organism surrounded by a capsule or sheath which might prevent the penetration of u.v. light or a chemical mutagen. For such an organism, ionizing radiation might be the most useful mutagen.

IAEA-SM-134/2 5

ИНДУЦИРОВАННЫЙ МУТАГЕНЕЗ ПРИ СЕЛЕКЦИИ ПРОМЫШЛЕННЫХ МИКРООРГАНИЗМОВ

С. И.АЛИХАНЯН Всесоюзный научно-исследовательский институт генетики и селекции промышленных микроорганизмов, М осква, Союз Советских Социалистических Республик

Abstract — Аннотация

INDUCED MUTAGENESIS IN THE SELECTION OF MICROORGANISMS. The paper summarizes the extensive experience accumulated by scientists in the application of various mutagens for improving the yield of useful forms of microorganisms, A review is presented of the role-and importance in this work of the choice of mutagen, the determination of the effective mutagenic dose, the effectiveness of selection in the various phases of the breeding of new strains with the help of mutagens, correlative mutability, etc. The paper also reviews the position with regard to the problem of the speci­ ficity of mutagenesis and of the possibility of resolving it in the context of the creation of productive strains. Finally, a summary is presented of the special features of selection work on producers of compounds of the amino acid and nucleotide type, and stress is laid on the importance of biochemical mutations for obtaining controlled changes in the cell metabolism.

ИНДУЦИРОВАННЫЙ МУТАГЕНЕЗ ПРИ СЕЛЕКЦИИ ПРОМЫШЛЕННЫХ МИКРООРГАНИЗ­ МОВ. В докладе подводится итог большому опыту, накопленному учеными в области приме­ нения различных мутагенов в работе по повышению продуктивности полезных форм микро­ организмов. Рассматривается роль и значение в этих работах выбора мутагена, установ­ ления эффективных доз мутагена, эффективности отбора на различных уровнях процесса выведения новых штаммов под воздействием мутагенов, коррелятивной изменчивости и т.д. Рассматривается состояние проблемы специфичности мутагенеза и возможности решения этой проблемы в деле создания продуктивных штаммов. В заключительной части доклада рассматриваются особенности селекционной работы с продуцентами соединений типа амино­ кислот, нуклеотидов, подчеркивается значение биохимических мутаций для получения на­ правленных изменений метаболизма клетки.

Селекция микроорганизмов с применением разнообразных мутагенов продолжает оставаться до настоящего времени методом, наиболее интен­ сивно используемым в работах по повышению продуктивности полезных форм микроорганизмов. Богатейший опыт накоплен по применению му­ тагенов в селекции продуцентов антибиотиков. Наиболее ценные резуль­ таты получены с продуцентом пенициллина Penicillum chrysogenum (Штауф­ фером, Алиханяном и др.), с продуцентом стрептомицина Streptomyces griseus (Дюлане, Стенли, Алиханяном и др.). Эти результаты, а также данные, полученные в целом ряде других случаев, позволяют, на наш взгляд, сделать ряд существенных обобщений. Во-первых,нельзя говорить об абсолютном преимуществе какого- либо определенного мутагена или группы мутагенов в селекции микроор­ ганизмов. Положительные результаты были получены при использовании как физических факторов — ультрафиолетовых, рентгеновских и гамма- лучей, быстрых нейтронов,так и многочисленных химических факторов - мутагенов,главным образом,азотистой формы иприта,этиленимина, диэ- тилсульфата и ряда других соединений. Установлено, что перечисленные м у ­ тагены отличаются друг от друга по эффективности индуцирования мута-

251 252 АЛИХАНЯН

ций - по таким признакам, как морфология колоний, окраска спор, пиг- ментообразование, биохимическая недостаточность. Однако в отношении повышения продуктивности штаммов, т. е. хозяйственно ценных призна­ ков, так называемых количественных признаков, заметного превосход­ ства одних мутагенов над другими выявлено не было. Основываясь на целом ряде фактов, можно утверждать, что характер мутагена не играет решающей роли в селекции, т. е. что целый ряд мутагенов, таких, как пе­ речисленные выше, почти одинаково эффективны в селекции. При этом разница в эффективности разных мутагенов больше зави­ сит от особенностей штаммов. Так, например, показано, что роль мута­ гена в значительной степени определяется этапом селекции, на котором его применяют, т. е. тем, насколько часто данный продуцент уже подвер­ гался отбору с применением одного и того же мутагена. Дикий штамм, ранее не подвергавшийся селекции, дает очень высокую изменчивость при обработке мутагеном, в то время как высокоактивный штамм, получен­ ный в результате многократного отбора под действием мутагенов, т.е. продукт длительной селекции, слабее реагирует на мутаген. Трудность определения преимущества одного мутагена над другим в селекции микроорганизмов объясняется еще и тем, что количественный признак - синтез вторичного метаболита — находится под полигенным контролем, т. е. зависит от действия целой системы генов. Наконец, большую роль в определении преимущества того или иного мутагена играет система генетической регуляции синтеза, под контро­ лем которого находится интересующий селекционера метаболит. Доста­ точно привести в качестве примера продуценты двух антибиотиков: эри­ тромицина и новобиоцина. Если многолетняя селекция продуцентов пени­ циллина, стрептомицина, аурэомицина,террамицина дает на каждом этапе отбора под действием мутагена ощутимые результаты, повышая их про­ дуктивность, то селекция этих двух продуцентов с непрерывным исполь­ зованием самых разнообразных мутагенов не дает сколько-нибудь замет­ ных результатов. Таким образом, сравнение нескольких мутагенов в се­ лекции Act. erythreus или Act. spheroides вряд ли приведет к выявлению преимущества того или иного мутагена, ибо сама система генов, участ­ вующих в регуляции синтеза, препятствует дальнейшей фенотипической реакции новых мутаций в генах этой системы. Если же генетическая си­ стема продуцента эффективно реагирует на мутаген, то при сравнении нескольких мутагенов не удается обнаружить существенной разницы в ак­ тивности мутагенов. Во-вторых, известно, что продолжительная селекция с применением одного и того же мутагена часто приводит к падению его эффективности. В этом случае для повышения эффективности селекции весьма полезна смена мутагена. Так, после многократного применения ультрафиолета целесообразно использовать мутаген иной породы, - например химичес­ кий мутаген или рентгеновские лучи, — мутагенный характер действия которых отличается от механизма действия предыдущего мутагена. Та­ кой эффект нами использован, и в его основе лежат сведения о различ­ ном механизме действия разных мутагенов. Среди возможных причин, повышающих эффект индуцированной изменчивости при смене мутагена, может быть изменение проницаемости клеточной мембраны или оболочки как эффекта действия на клетку предыдущего мутагена. Третье обобщение связано с выбором эффективной дозы мутагена. Как известно, при воздействии на клетки микроорганизма ионизирующей IAEA-SM-134/2 5 253

радиации эффект находится в линейной зависимости от дозы. В такой же зависимости находится эффект от дозы при обработке клеток хими­ ческими мутагенами, - с тем отличием, что при достижении какого-то по­ рога дозы эффект уменьшается и кривая приобретает ярко выраженную тенденцию падения вниз. Ожидалось, что максимальная изменчивость по интересующему селек­ ционера признаку, т. е. по числу вариантов, превышающих среднюю актив­ ность штамма, должна быть достигнута при наиболее высокой дозе мута­ гена. Многократная проверка действия рентгеновских и ультрафиолетовых лучей, а также химических мутагенов показала, что наибольшее число ва­ риантов микроорганизмов с повышенной активностью возникает в диапа­ зоне умеренных доз, часто не совпадающих с теми дозами, которые вызы­ вают максимальное число морфологических и других альтернативных му­ таций. Более того, использование доз, индуцирующих максимальное чис­ ло альтернативных мутаций, главным образом морфологических, сопро­ вождается возникновением вариантов с нарушенной способностью к био­ синтезу антибиотика, увеличением числа минус-вариантов. В связи с этим можно вспомнить, что наиболее успешные результа­ ты в селекции продуцента пенициллина под действием ультрафиолетовых лучей получены при дозе, обеспечивающей 1% выживаемости клеток. Не лишне при этом заметить, что, вероятно, этой особенностью изменчивос­ ти по количественным признакам можно объяснить отсутствие в арсенале генетиков какого-то одного или, может быть, нескольких высоко мута­ генных факторов. Вероятно, умеренные дозы мутагена или умеренные мутагены, не вызывающие резких перестроек генотипа клетки, наиболее полезны в се­ лекции микроорганизмов. На основании большого экспериментального материала напрашивает­ ся четвертый вывод. Повышение активности штамма под действием му­ тагенов, как правило, не связано с коренным изменением его морфоло­ гии. Напротив, резкое изменение колониально-морфологических призна­ ков очень часто сопровождается полной потерей или существенной поте­ рей продуктивности штамма. Так, например, было показано, что высо­ кий процент вариантов с пониженной продуктивностью или даже с полной потерей способности синтезировать продукт встречается среди колоний, резко отклоняющихся по своим морфологическим признакам от исходного типа. Однако это не значит, что резкие сдвиги продуктивности никогда не могут сопровождаться изменениями морфологии штамма, его микро­ скопических признаков, размера и формы колонии, степени споруляции, строения воздушного и субстратного мицелия и т. п. Обычно эти морфо­ логические изменения не выходят за пределы, характерные для индивиду­ альной изменчивости данного штамма. Подчас они настолько незначитель­ ны, что измененные колонии трудно отличить от неизмененных в односпо­ ровом рассеве. Выявить происшедшие изменения порой бывает можно только при тщательном сравнении исследуемого штамма со штаммом, вновь полученным в результате дальнейшего отбора. Таким образом, на вопрос о корреляции морфологических признаков и продуктивности штаммов в селекции с использованием мутагенов мож­ но дать однозначный ответ: установить связь между морфологическими признаками и продуктивностью штаммов не представляется возможным, такая корреляция не установлена. Вместе с тем, колонии более актив- 254 АЛИХАНЯН

ных штаммов всегда отличаются от колоний менее активных штаммов (размером колоний, характерными особенностями гифов у мицелиальных микроорганизмов, характером и степенью споруляции, цветовыми оттен­ ками и т.п .). Эти отличия столь незначительны, что их трудно обнару­ жить в односпоровом или одноклеточном рассеве, — настолько трудно, что по этим незначительным отличиям колонии не поддаются отбору. Такое отсутствие связи между морфологическими признаками и про­ дуктивностью штамма осложняет отбор и делает селекцию микроорганиз­ мов очень трудоемким процессом. В этом отношении достойно внимания предложение Аутона из Англии об отборе по едва обнаруживаемым отли­ чиям сектора в гигантских колониях, выращенных из обработанной мута­ генами суспензии спор. Широкое применение мутагенов в селекции повысило продуктивность многих микроорганизмов в 100-300 раз. Такие штаммы, как правило, — результат многократных отборов с применением мутагенов, результат постепенного накапливания небольших приростов активности в течение 10-20, а иногда и больше поколений (ступеней отбора). Столь многократ­ ная обработка мутагенами одного и того же генома приводит к насыще­ нию данного генома мутациями и, в конечном счете, влечет за собой сни­ жение вероятности возникновения новых влияющих на продуктивность му­ таций в этом геноме, т. е. к падению эффективности отбора под действи­ ем мутагена. У таких штаммов резко падает коэффициент изменчивости. Явления, происходящие при этом, напоминают картину, наблюдаемую при снятии кривых "доза-эффект", когда повышение дозы мутагена до опре­ деленного уровня вызывает возрастание частоты мутаций до максимума, после достижения которого дальнейшее повышение дозы приводит к паде­ нию частоты мутаций. Уменьшение эффективности отбора у высокоактивных штаммов мо­ жет быть обусловлено накоплением в их геноме различных физиологичес­ ких мутаций, снижающих жизнеспособность и нарушающих физиологичес­ кую корреляцию. При дальнейшей обработке такого штамма мутагеном вновь возникшие мутации оказываются гибельными для клеток. Проис­ ходит своеобразная стабилизация признака в популяции. Возникает воп­ рос, можно ли экспериментально нарушить стабилизацию признака ? Мож­ но ли снять эффект нарушения физиологических корреляций? В литерату­ ре описаны случаи дестабилизации признака посредством радикальной пе­ рестройки генотипа под действием мутагена, известной под названием "большой" мутации. Описаны случаи, когда в результате возникновения "большой" мутации резко повышался коэффициент изменчивости и, как следствие, восстанавливался эффект отбора у высокоактивных штаммов под действием мутагенов. Мутагены, физические и химические, в селекции микроорганизмов применяют уже на протяжении почти 25 лет. С их помощью достигнуты значительные успехи в селекции большинства промышленно важных про­ дуцентов. Тем не менее, в области применения мутагенов в селекции до настоящего времени имеется много неясных и нерешенных вопросов. Нуж­ но признать, что метод селекции микроорганизмов с использованием му­ тагенов несет на себе отпечаток эмпиризма. В настоящее время четко наметилась тенденция развития исследова­ ний индуцированного мутагенеза в селекции микроорганизмов в направле­ нии разработки проблем специфичности мутагенеза. IAEA-SM-134/25 255

Существует мнение, что решение этой проблемы приблизит нас к по­ лучению направленных мутаций и, тем самым, —к существенному повыше­ нию эффективности селекции с использованием мутагенов. В разработ­ ке этой проблемы исследователи встречаются с двумя трудностями: пер­ вая —отсутствие ясности в этой проблеме на уровне теории мутагенеза, вторая - осложнение этой трудности в свете особенностей генетики ко­ личественных признаков. В настоящее время проблема специфичности мутагенеза еще очень далека от разрешения. По нашему мнению, пробле­ мы специфичности мутагенеза в том виде, как она сформулирована в сов­ ременной литературе, не существует. При этом следует помнить, что в общих чертах концепция специфичности мутагенеза обычно сводится в ли­ тературе к предложению о том, что определенные мутации индуцируются определенными мутагенами и что путем подбора соответствующих мута­ генов можно получить мутацию с желательным фенотипом. Эту концепцию можно правильно оценить, приняв во внимание два об­ стоятельства: молекулярную природу мутаций и сложность проявления му­ таций, связанную с многоэтапностью мутационного процесса. Как извест­ но, в основе молекулярного механизма мутагенеза лежит изменение по­ рядка чередования нуклеотидов в молекуле ДНК, в свою очередь связанное с заменой А-Т-пары на Г-Ц-пару и наоборот. Вследствие этого, воз­ можности изменения гена ограничены всего двумя типами замен (мы соз­ нательно не упоминаем здесь о трансверсии, а также о выпадении и встав­ ке пар оснований — как об очень редких событиях). Таким образом, если какой-либо мутаген способен вызывать замены в обоих направлениях, то он практически может индуцировать любую мута­ цию в любом районе любого гена. В случае же, когда используются мута­ гены, которые могут вызывать замены только в одном направлении, ин­ дуцирование любой мутации, связанной с заменой пар оснований, может быть обеспечено использованием, как максимум, всего двух мутагенов — при условии, что замены, вызываемые этими мутагенами, противополож­ но направлены. Исходя из рассуждений, следует сделать вывод, что материальные, точнее - молекулярные, основания специфичности мутагенеза и возмож­ ности подбора "специфических" мутагенов настолько ограничены, что вряд ли можно всерьез считать реальной идею такого подбора мутагенов, который обеспечивал бы получение фенотипически определенных мута­ ций. С учетом второго аспекта мутагенеза — многоэтапности процесса — проблема специфичности еще более осложняется. Например, если допус­ тить, что одним из этапов мутагенеза является проникновение мутагена через оболочку обрабатываемой клетки, то однажды подобранный мута­ ген, специфически эффективный в отношении этой клетки, может стать неэффективным, вследствие нарушения проницаемости оболочки после по­ лучения у клетки очередной мутации. Следовательно, в данном случае можно говорить об относительной специфичности, характерной для каж­ дого этапа отбора, т. е. о подборе для каждой новой ступени отбора но­ вого "специфического" мутагена. В арсенале генетиков имеется целый ряд селективных методов полу­ чения определенных мутаций у микроорганизмов. Так, метод селектив­ ных сред дает возможность отбирать определенные биохимические мута­ ции. Использование селективных штаммов, несущих амбер- и охра-мута- ции, позволяет получить мутации по определенным трипломам в молекуле ДНК. Применение этих и других приемов селективного отбора мутантов 256 АЛИХАНЯН

находится в согласии с концепцией не специфического мутагенеза, а управ ляемого мутагенеза, т.е. разработки методов, позволяющих выявлять и отбирать лишь определенные типы мутаций. Нам кажется, что исследования, направленные на повышение эффек­ тивности мутагенов в селекции микроорганизмов, должны проводиться по линии усовершенствования методов избирательного отбора мутантов, т. е. по линии повышения эффективности управляемого мутагенеза. Сравнительно недавно начало складываться новое направление в се­ лекции микроорганизмов, связанное с получением биохимических мута­ ций. Успехи работ в этом направлении базируются на широких биохимичес­ ких исследованиях по расшифровке путей биосинтеза ряда метаболитов и являются предпосылкой к решению управляемого мутагенеза. Речь идет о селекции продуцентов аминокислот, нуклеотидов, вита­ минов, т. е. веществ, объединяемых в группу первичных метаболитов. Как известно, молекулы первичных метаболитов значительно проще, чем моле­ кулы вторичных метаболитов. Поэтому, в то время как механизмы био­ синтеза большинства вторичных метаболитов изучены только в общих чер­ тах, основные пути биосинтеза аминокислот, нуклеотидов, отчасти вита­ минов изучены у микроорганизмов достаточно детально. Известно, что концентрация первичных метаболитов в клетке находится под строгим кон­ тролем, осуществляющимся с помощью хорошо изученных механизмов реп­ рессии и ретроингибирования, в результате чего в обычных условиях пер­ вичные метаболиты не накапливаются в клетках в больших количествах. В противоположность этому, вторичные метаболиты часто образуются в больших количествах; исследование механизмов регуляции их биосинтеза находится на самом начальном этапе. Наличие детальных сведений о путях биосинтеза аминокислот и нукле­ отидов, о механизмах, контролирующих скорость их синтеза, а также воз­ можность применения генетических методов блокирования различных звеньев этого процесса определяют специфичность методов селекции про­ дуцентов соответствующих первичных метаболитов, основанных на исполь­ зовании ауксотрофных мутантов с нарушенной системой регуляции. Характерной особенностью ауксотрофных мутантов является их спо­ собность накапливать промежуточные продукты биосинтеза, предшествую­ щие генетически блокированной реакции, причем промежуточные продук­ ты биосинтеза могут подвергаться дальнейшим биохимическим превраще­ ниям. Вместе с тем, пути биосинтеза различных метаболитов неразрывно связаны между собой, в силу чего блокирование одной химической реак­ ции часто приводит к изменению скорости других реакций, сопряженных с первой. Таким образом,в результате одной биохимической мутации в процессе биосинтеза данного метаболита может возникнуть большое ко­ личество всевозможных изменений. Тем не менее, получение ауксотроф­ ных мутантов микроорганизмов оказалось чрезвычайно ценным методом в селекции продуцентов некоторых метаболитов. Ценность этого метода селекции заключается прежде всего в том, что его применение очень сильно сокращает длительность и трудоемкость селекционного процесса и приближает нас к тому, что мы назвали управ­ ляемым мутагенезом. Использование в селекции микроорганизмов,у которых расшифрова­ ны пути биосинтеза первичных метаболитов, дает ощутимые преимущества благодаря тому что у них можно предвидеть результат определенной био­ химической мутации. IAEA-SM - 134/25 257

В промышленном производстве триптофана, лизина, глутаминовой кис­ лоты, инозиновой кислоты в настоящее время применяют такие биохими­ ческие мутанты, а в арсенале генетиков имеются наготове высокопродук­ тивные мутанты, обеспечивающие возможность организации крупномас­ штабного производства и других аминокислот, как только в том появится необходимость. Вышеописанный метод селекции микроорганизмов весьма прост и вы­ сокоэффективен. Однако следует помнить, что эмпирическое получение биохимических мутантов не может привести к каким-либо положительным результатам, тем более если речь идет о вторичных метаболитах. Изуче­ ние путей биосинтеза тех или иных соединений — обязательное условие успешной селекции методом блокирования различных звеньев в цепи био­ химических реакций. Таким образом, данный путь селекции, открываю­ щий неограниченные просторы для управляемого мутагенеза и весьма про­ стой, с точки зрения селекционера, оказывается достаточно сложным и трудоемким при проведении биохимических исследований. На примере создания сверхактивных штаммов продуцентов аминокис­ лот и пуриновых соединений можно убедиться в большом значении для ус­ пешного проведения направленной селекции сведений о генетическом кон­ троле синтеза метаболитов. По своей продуктивности эти штаммы, обра­ зующие до 30 мг/мл лизина и свыше 50 мг/мл глутаминовой кислоты, на­ много опередили многочисленные штаммы продуцентов антибиотиков, — активности лучших из них в результате многолетних селекционных работ достигли, в пересчете на весовое содержание, всего 15-20 мг/мл (пени­ циллин, стрептомицин, террамицин). В то же время повышение активно­ сти других продуцентов, - например олеандомицина, эритромицина, ново- биоцина, гиберрелловой кислоты и ряда ферментов, — все еще не дает эф­ фекта более чем 1-2-3 мг/мл. Большие перспективы намечаются в использовании для мутагенной селекции гибридных микроорганизмов. Было показано, что размах как спонтанной, так и индуцированной ультрафиолетовыми лучами изменчивости по пенициллинообразованию у диплоидных штаммов Penicillum chrysogenum выше, чем у исходных гап­ лоидных штаммов. В одном из описанных случаев после обработки конидий диплоида про­ дуцента пенициллина мутагеном среди небольшого числа вариантов были отобраны два, превышающие активность диплоида на 25-35% . К сожалению, несмотря на перспективность использования гибридных штаммов в мутагенной селекции, сведений в литературе о подобных рабо­ тах очень мало. Говоря об использовании физических и химических мутагенов в се­ лекции микроорганизмов, следовало бы учесть способность фагов инду­ цировать колоссальную изменчивость у микроорганизмов, иногда превы­ шающую изменчивость, возникающую под действием самых сильных мута­ генов. Эффективность этого пути селекции может быть использована у форм микроорганизмов, имеющих специфические умеренные фаги. На сегодняшний день успехи индуцированного мутагенеза позволяют нам утверждать, что наиболее эффективным путем селекции промышлен­ ных микроорганизмов является использование физических и химических мутагенов с учетом тех общих закономерностей, которым мы посвятили данное сообщение. Эти закономерности построены на анализе материа­ лов, полученных в работе с разнообразными микроорганизмами — проду­ 258 АЛИХАНЯН

центами антибиотиков, аминокислот, факторов роста, витаминов, фермен­ тов и других метаболитов, промышленное производство которых расши­ ряется из года в год.

DISCUSSION

М. MRAÖEK: What mutations do you have in mind when you speak about plus-variants? S. I. ALIKHANLAN: I mean the so-called "small" mutations, which possibly form the basis of genetic modifications. Choosing the "small" mutations step by step, we can, after several steps of selection, noticeably shift the activity of the strain. A. S. R. C. ZAGA.LLO: Could you please tell us something about the induction of polyploidy in microorganisms and the possibility of its application in industry? S. I. ALIKHA.NIAN: The production of polyploids in yeast has been reported (Imshenetsky et al. ) but the work represents laboratory in­ vestigations. No work on polyploidy in bacteria has yet been reported in the USSR. IAEA-SM-134/12

RADIATION IMPROVEMENT OF Candida No. 25 FOR PETROPROTEIN FERMENTATION

SHU-HSUN TING, CHUAN-HSIAN LI, CHING-SONG LEE Chia Yee Solvent Works, Chinese Petroleum Corp., Taiwan, Republic of China

Abstract

RADIATION IMPROVEMENT OF Candida No.25 FOR PETROPROTEIN FERMENTATION. For the purpose of studying the production of petroprotein by fermentation with petroleum as the sole carbon source, Candida No. 25 was isolated in July 1964 from the soil around the diesel-oil. storage tank in the Kaoshiung Refinery. To improve the fermentation efficiency of the strain it was first exposed to u. v. - irradiation. After the radiation was performed several times, the results showed that the new isolated mutant grew rapidly in the high concentration of hydrocarbon. Not only was there an increase in cell concentration in the broth from 8 to 35 g/1, but there was also a shortening in the necessary fermentation period from 72 to 36 hours. The improved strain showed great stability after being transferred from a 5-litre fermentor in the laboratory to a 5000-litre fermentor and even to a 150 000-litre fermentor in which more than ten running tests were made. Now, further experiments are under way with X-rays and gamma rays to achieve improved strains of microorganisms that would be suitable for petroprotein fermentation on a commercial scale.

INTRODUCTION

For the purpose of studying the production of petroprotein by fermentation with petroleum as the sole carbon source, researchers of Chia Yee Solvent Works have isolated many petroleum-propagating microorganisms from the soil around the Kaoshiung Refinery, Chinese Petroleum Corporation, Taiwan, since 1964. Ultraviolet radiation has been tried as a means of genetically improving Candida No. 25, which is one of the microorganisms isolated from the soil.

MATERIAL AND METHOD

1. Culture and culture medium

The petroleum used in this research was normal paraffin supplied by Mitsubishi Shoj Kaisha Ltd., Japan. An analysis of the normal paraffin is shown in Table I. The petroleum medium used contained 1 to 4 g of n-paraffin per hundred millilitres of medium and other inorganic nutrients.

2. Radiation and screening

The ultraviolet source was a 15-W electrical germicidal lamp hung from the ceiling of a sterile room. The primary wave length was at 2537 A. Cell suspensions of Candida No. 25 were prepared by shaking the culture in fermentation medium for 72 h and then diluting with sterile

259 260 SHU-HSUN TING et al.

TABLE I. ANALYSIS OF NORMAL PARAFFIN

Specific gravity (15eC) 0.7 7 8 0

N-paraffin purity {% 9 4 .4

N -paraffin hom ologue distribution (wtÿo)

П"С 12 0 .1 n-C . 0 .1 13

n - c i4 8 .5

n -C is 4 9 .5

n -C is 2 2 .8

n - c i, 8 .6

n - C 1B 2 .8

" 'C 19 1 .0 П-C go 0 .5

П-C 21 0 .3

n-C 22 0 .1 n -C z3 0 .1

Mean molecular weight 2 1 9 .4

Iso-p araffin contents (wt^o) 5 .4

Aromatics content (wt%) 0.1 2 7

Bromine number 0 .0 4 5

Sulphur content (ppm) 2 .3

Flash p o in t (eC) 124

Colour (APHA) 40 (Saybolt + 15)

TABLE II. RADIATION CONDITIONS

No. of R adiation Rate of No. of cells/ml No. of strains rad iatio n tim e killing that survived selected tre atm en ts (m in)

1 5 99.3 1 30 3

2 5 99. 53 18 2

3 6 99.18 42 4

4 6 99 .7 0 14 3

5 6 99. 64 16 5 IAEA-SM-134/12 261

UV EXPOSURE - MIN.

FIG .l. Ultraviolet survival curve of Candida No. 25.

TABLE III. PETROLEUM FERMENTATIVE PROPERTIES OF THE Candida STRAINS AFTER EACH RADIATION

No. of Cell concentration Percentage u. v. - at the end of 45 h protein S train . irradiations (g/D co n ten t (N x 6.2 5 )

Parent strain 0 5-7 4 9 .8 6

2 5 -U 2 1 18-24 4 9 .2 8

2 5 -U „ 2 24-30 49 .5 3

2 5 -U 214 3 28-32 49 .8 8

2 5 - U 2m. 4 30-34 50 .1 3 5 32-36 50.67 2 5 - U 2,4u water to make a suspension of about 8000 cells/ml. Then 0. 5 ml of this suspension was spread over the surface of the petroleum agar. The petroleum agar spread with cell suspension was then u.v. - irradiated at a distance of 20 cm from the ultraviolet source. The radiation time was controlled to make the cell-survival ratio less than one per cent. After irradiation, the first screening work was done by incubating the inoculated petroleum agar at 30°C for five days. The large colonies were selected, isolated, purely cultured and then used for the next screening. The second screening was done by inoculating 500-ml duplicate flasks with the selected strain from the first screening. The flasks were shaken at 33°C. The cell concentration was determined periodically during the fermentation. Strains with a high growth rate in the 500-ml fermentation flasks were selected and further tested in a 10-litre Jar fermentor from Marubishi Co., Japan. The medium for each fermentor was 3 litres. The operation temperature was kept at 33°C. Each batch was fermented for 45 h and the cell concentrations were determined every 4 h. 262 SHU-HSUN TING et al.

TABLE IV. FERMENTATIVE STABILITY OF Candida 25-U21414 IN A 10-LITHE FERMENTOR

Cell concentration at Yield of biomass No. of the end of 45-h based on n-paraffin e x p e rim en t fermentation (°!°) (g П)

1 36.7 91 .7 5

2 3 5 .8 89. 50

3 3 6.2 91 .5 0

4 3 4.3 85 .7 5

5 ■34.1 8 5 .2 5

6 3 3 .5 83 .7 5

7 3 6 .4 91.00

8 3 4.8 87 .0 0

9 3 5.1 87 .7 5

10 3 6 .0 90 .0 0

11 3 3.8 84. 50

12 3 4.6 86 .5 0

13 3 5.3 85 .7 5

14 3 3.6 8 4 .0 0

15 3 5.9 8 9 .7 5

16 3 4.1 8 5 .2 5

The average cell concentrations: 35. 01.

Standard deviation: 1. 03.

TABLE V. COMPARISON BETWEEN Candida 25 " u 2 i4 i 4 AND PARENT STRAIN

Morphology and Improved strain Parent strain growth characteristics (25-U 2l414)

Shape and size of cell Long elliptical E llip tica l 5 X 6 (jm 5-6 X 6-7 pm

Colony ap p e aran ce in Virgin : irregular R egular petroleum agar medium Centre: rough and sm ooth

Growth rate of colony after 10 days (diameter in ram) 1 .5 3 .2 IAE A-SM “134/12 263

INCUBATION AT 33 ° С - HOURS

FIG.2. Comparison of petroleum fermentation ability between Candida No. 25 and Candida 25-U:111<.

FERMENTATION AT 3 5 “ С - HOURS

FIG.3. Fermentation of Candida 25-lL.... in 150 000 litre air-lift fermentor. ------21414 Conditions: n-paraffin 2 wfÿo to mash; mash volume 95 000 litres; pH 3-4.

RESULTS

After the first radiation, the best fermentative strain was then used for the next radiation. In this way the strain was successively improved until the fifth irradiation with ultraviolet rays. The radiation conditions and survival curve are shown in Table II and Fig. 1. The ability of Candida No. 25 to ferment petroleum after each irradiation in a 10-litre Jar fermentor are shown in Table III. In order to test the stability of Candida 25-U2i414| its fermentative ability was determined every two weeks by inoculating it in a 10-litre Jar fermentor under the same operational conditions. The results of 16 experiments are listed in Table IV. The morphology and growth characteristics of Candida 25-U21414 and of the parent strain are compared in Table V and Fig. 2. 264 SHU-HSUN TING et al.

TABLE VI. COMPOSITION OF PETROPROTEIN PREPARED FROM DRIED CELLS OF Candida 25-U2i4i4

Crude protein (7o) 51 .1 5 Vitamin B6 (mg^o) 0 .2 4

C rude fa t (°Jo) 2 .4 3 Vitamin B12 (mg °Jo) 0 .1 2

Crude fibre (%) 7 .9 6 Vitamin D (mg^o) 1 0 .0

C rude ash (°Jo) 8 .1 6 Nicotinic acid (mg^o) 1 4 .6

Nitrogen-free extract (°¡o) 2 2 .6 5 Pantothenic acid (mg °jo) 1 1.1

V ita m in B1 (mg

Vitamin B2 (mg °¡o) 4 .4 8

In 1969 and 1970, the improved Candida 25-U2i4i4 was further tested in a 5000 litre air-lift fermentor and then in a 150 000-litre air-lift fermentor. The results are consistent with those obtained in the laboratory. The growth curve of Candida 25-U914.14 in a 150 000-litre air-lift fermentor are shown in Fig. 3. After fermentation in the 150 000-litre air-lift fermentor, the cells of Candida 25-U21414 were collected, washed with water and dried. The dried cells formed a pale-white powder with a pleasant odour. The analysis of the dried cells of Candida 25-U^1414 are shown in Table VI.

DISCUSSION

After nearly one. year of experimental work on the genetic improvement of Candida No. 25 with ultraviolet irradiation, very distinct improvements have been achieved, especially after the first and second irradiations. To achieve the goal of commercializing the petroprotein fermentation process in Taiwan, our present work has been accelerated with respect to X- and gamma irradiation of Candida 2 5 -U 2 14 14 and some other petroleum- propagating microorganisms. It is hoped that some new predominant strains will be obtained.

REFERENCES

[1] ANDERSON, E.H ., J. Bact. ¡И (1951) 389. [2] BRIDGES. B. A., MUNSON, R.J., J. gen. Microbiol. 46 (1967) 339. [3] DAVIES, D.R., ARLETT, C.F., J. gen. Microbiol. 46 (1967) 329. [4] IGUCHI, S., J. agrie. Chem. Soc. Japan 23 (1950) 16, 357, 25 (1951) 283. [5] IKEDA, Y., Applied Microbiol. Engineering, Vol. 1, Kyoritsu, Japan (1956) 1-164. [ 6] SHUICHI AIBA, VITALIS MORITZ et al., J. Ferment. Technol., Osaka « (1969) 203. [7] SVIHLA, G., DAINKO, J. L., SCHLENK, F., J. Bact. 85 (1963) 399. [8] SVIHLA. G., SCHLENK, F., DAINKO, J.L., Radiat. Res. 13 (1960) 879. [9] WATSON, J.D ., J. Bact. 60 (1950) 697. IAE A-S M -1 3 4 /1 2 265

DISCUSSION

H. HESLOT: Have you carried out any taxonomical study on your Candida No. 25 strain? Is it a Candida tropicalis or a Candida lipolytica?. Does it require thiamine for growth? SHU-HSUN TING: Yes, we have. It is a variety of Candida guilliermondii, according to our preliminary taxonomical study. It does not require thiamine for growth. H. HESLOT: When do you expect to start an industrial unit? SHU-HSUN TING: In the middle of this year. J. MEYRATH: Why do you use an air-lift fermentor? What oxygen transfer rate do you get in your fermentor in operations on laboratory, pilotplant and factory scales? SHU-HSUN TING: Air-lift fermentors are used because they consume less power in large-scale fermentation. We have not yet determined the oxygen transfer rate in our fermentor. J. MEYRATH: What is the highest rate of multiplication of your fastest growing strain in the exponential phase? SHU-HSUN TING: The doubling time of our Candida 25-U21414 is about 4 hours. J. MEYRATH: Why have you now switched your attention to Candida, after finding that yields and protein content in a Pseudomonas strain were considerably higher? SHU-HSUN TING: From the industrial standpoint, the Candida cells are much easier to separate and give rise to fewer toxicity problems.

IAEA-SM-134/6

TECHNIQUES UTILISABLES POUR L' AMELIORATION GENETIQUE DE LA LEVURE DANS UNE OPTIQUE INDUSTRIELLE

P. GALZY Ecole nationale supérieure agronomique de Montpellier et P. DUPUY INRA, Station de technologie des produits végétaux, Dijon, France

Abstract — Résumé

TECHNIQUES WHICH CAN BE USED IN IMPROVING YEAST GENETICS FOR INDUSTRIAL PURPOSES. The genetics of baker’s yeast (Saccharomycescerevisiae) hasbeena subject of study for several years. The techniques used in the basic research have been described in many published articles and on the whole they are effective. Unfortunately, they cannot always be adapted readily to the improvement of in­ dustrial strains. The paper discusses these various techniques, emphasizing the changes that have to be made in them in order to meet industrial requirements. In nature, the strains are either diploid or poly­ ploid. Mutageneses and selection can be carried out more easily on haploid clones. The authors discuss techniques of sporulation and dissection of asci. For basic research, micromanipulator analysis of tetrads is indispensable. On the other hand, for a genetic improvement of the industrial characteristics of yeast, it might be preferable to use the techniques of mass dissection, by which a larger number of clones of monosporous origin can be obtained very rapidly. The most critical problem is the genetic marking of these haploid clones. Generally speaking, the strains are marked by growth-factor requirements (muta­ tion for auxotrophy). Thus, crossings can be made very simply by selection of prototrophs. These markers are a means of checking on the purity of the clones at all times and of taking precautions against contamination. As a rule, however, auxotrophy reduces the vitality of a yeast. Even if prototrophic diploids are reconstituted for industrial use, there is still reason to fear that the vitality of these diploids will be diminished, since they will be heterozygous for several auxotrophies. The paper describes various possibilities of genetic marking which are better suited to industrial requirements. Up to now, research on yeast genetics has been carried on for theoretical purposes. Attempts to apply it to the solution of industrial problems have been few. There is reason to believe that conditions are now favourable for developing industrial applications of yeast genetics.

TECHNIQUES UTILISABLES POUR L'AMELIORATION GENETIQUE DE LA LEVURE DANS UNE OPTIQUE INDUSTRIELLE. La génétique de la levure de boulangerie Saccharomyces cerevisiae est très étudiée depuis plusieurs années. Les techniques utilisées dans les recherches fondamentales ont fait l'objet de nombreuses publica­ tions et sont en général efficaces. Elles ne sont malheureusement pas toujours facilement adaptables à l'am élioration des souches industrielles. Le mémoire discute ces diverses techniques en insistant sur les modifications qu’il est nécessaire de leur apporter pour répondre aux nécessités industrielles. Dans la nature les souches sont diploides ou polyploides. Il est plus facile d’effectuer la mutagénèse et la sélec­ tion sur des clones haploides. Les techniques de sporulation et de dissection des asques sont exposées. Pour des recherches fondamentales, l'analyse des tétrades au micromanipulateur s’impose. Par contre, pour une amélioration génétique des caractéristiques industrielles de la levure, il peut être préférable d’utiliser les techniques de dissection en masse qui permettent d’obtenir très rapidement un grand nombre de clones d’origine monosporale. Le problème le plus délicat est le marquage génétique de ces clones haploïdes. En général, les souches sont marquées par des besoins en facteurs de croissance (mutation pour l’auxotrophie). Il est ainsi possible de faire des croisements très simplement par sélection de proto-

267 268 G ALZ Y et DUPUY

trophes. Ces marqueurs permettent de contrôler à tout moment la pureté des clones et de se prémunir contre des contaminations. Or, une auxotrophie diminue, en règle générale, la vitalité d'une levure. Même si on reconstitue des diploides prototrophes pour une utilisation industrielle, il est à craindre que ces diplo'ides aient une vitalité amoindrie. En effet, ils seront hétérozygotes pour plusieurs auxotrophies, Diverses possibilités de marquage génétique mieux adaptées aux nécessités de l'industrie sont exposées. Jusqu'à présent, les recherches effectuées sur la génétique des levures avaient des objectifs théoriques. Les tentatives d'application à la solution de problèmes industriels ont été peu nombreuses. Il est permis de penser que les conditions sont maintenant favorables au développement des applications industrielles de la génétique des levures.

1. INTRODUCTION

Grâce à 1' existence d1 une haplophase et d 1 une diplophase nettement séparées, les levures se prêtent fort bien à l'analyse génétique. Mal­ heureusement, la plupart des travaux sont limités aus espèces Saccharo­ myces cerevisiae Hansen et Schizosaccharomyces pombe Lindner. Ils ont eu la plupart du temps des objectifs théoriques, soit l'étude des méca­ nismes génétiques, soit la connaissance des voies métaboliques. Malgré l'importance industrielle de S. cerevisiae, on constate que les applications de la génétique ont été peu nombreuses et de faible im ­ portance. En revanche, les industriels ont fait de grands efforts pour sélectionner les souches de levures. Ce travail commence en général p ar 1 ' isolement d’un très grand nombre de souches, prélevées dans la nature ou dans l'environnement industriel. Ces souches sont ensuite essayées au laboratoire. Les plus intéressantes peuvent être expéri­ mentées au niveau industriel. Pratiquement, cette méthode est longue, difficile et le plus souvent décevante. Deux raisons importantes semblent expliquer cette situation. D 1 une part, les techniques utilisées classiquement en génétique des levures [8 ] sont mal adaptées à la solution de problèmes industriels. D' autre part, les objectifs de la sélection ont été souvent insuffisamment analysés dans certaines industries. Cette revue a pour but d'examiner dans quelle mesure les moyens gé­ nétiques dont nous disposons peuvent aider à améliorer les souches de levures utilisées industriellement.

2. CYCLES BIOLOGIQUES DES LEVURES

Le cycle biologique d'une levure conditionne les possibilités de sélec­ tion et le choix des techniques. Plusieurs cas peuvent se présenter.

2.1. Saccharomyces cerevisiae Hansen

Les levures de boulangerie et de brasserie (S. cerevisiae Hansen) sont généralement diplobiontiques et hétérothalliques. La multiplication végé­ tative se fait par bourgeonnement. Les cellules sont diplo'ides. Tant que les conditions sont favorables à la croissance, la multiplication végéta­ tive continue. Sous 1' influence de certains facteurs, la méiose se dé­ clenche. Il se forme 4 noyaux dans une cellule qui se transforme alors en asque avec 4 spores (pratiquement, le nombre de spores varie de 1 à 4). IAEA-SM-134/6 269

FIG. 1. Cycle biologique de Saccharomyces cerevisiae. 1 diplophase 5 haplophases 2 méiose 6 copulation 3 asque à 4 spores 7 zygote 4 germination

Quand 1' asque éclate, les spores germent, donnant des cellules haploides capables de se multiplier par bourgeonnement. Dans l 1 asque, 2 spores sont de signe +, ou a, et 2 spores sont de signe -, ou a. Une cellule haploïde de signe + et une cellule de signe - peuvent copuler par conjugai­ son isogamique. Il y a donc hétérothallisme, puisque le croisement n' est possible qu 1 entre cellules de signe opposé. Mais aucune différence m or­ phologique n'apparaît entre les 2 gamètes. Le signe est sous contrôle monogénique. Des changements de signe peuvent être obtenus facilement par mutation. Chez S. cerevisiae, la phase haploïde est généralement très courte dans la nature. Au laboratoire, il est par contre facile de conserver une souche haploïde. Il suffit d'éviter le contact avec des souches de signe opposé. Notons que le bourgeonnement peut s'effectuer en n'importe quel point de la cellule. La figure 1 schématise le cycle biologique.

2.2. Kluyveromyces lactis (Dombrowski) Van der Walt,

L 1 espèce Kluyveromyces lactis joue un rôle important dans les in­ dustries laitières, puisqu'elle assimile le lactose. Les espèces du genre Kluyveromyces diffèrent des Saccharomyces par la fragilité de 1' asque. Des sa maturité, les spores sont éjectées à 1' extérieur [14] . De ce fait, ces souches ne permettent pas d'effectuer une analyse commode des té tra d e s. K. lactis peut être diplobiontique et hétérothallique comme S. cerevisiae, mais la plupart des souches sont homothalliques et haplobiontiques. Les cellules en multiplication végétative sont haploïdes, elles se multiplient par bourgeonnement multilatéral tant que les conditions de milieu sont favorables à la croissance. Dans certaines conditions, deux cellules sœurs peuvent copuler et donner ainsi un zygote qui peut se transformer en asque (fig.2 ). 270 G ALZ Y et DUPUY

5

FIG.2. Cycle biologique de Kluyveromyces. 1 germination 4 zygote 2 haplophase 5 asque 3 copulation 6 libération des spores

FIG.3. Cycle biologique de Candida albicans (Syringospora albicans Quinquad). 1 diplophase 4 haplophase sexuellement inactive 2 téliospores 5 karyogam ie 3 haplophase sexuellement active

2.3. Candida (Berkhout)

Pendant longtemps, on a cru que le genre Candida n 1 avait pas de cycle biologique parce qu 1 il n 1 était pas possible de mettre en évidence des ascospores. En 1967, Van der Walt [22] a découvert des souches de Candida albicans ayant une activité sexuelle. Elles sont homothalliques et la conjugaison se produit entre la cellule et son bourgeon. Cette par­ ticularité limite beaucoup les possibilités d 1 étude de cette sexualité. La génération diploïde produit des cellules dormantes, multinucléées, qui forment des conidies après réduction chromatique (fig.3). Cette découverte est importante puisque la plupart des espèces de levures cultivées pour l 1 alimentation animale appartiennent à ce genre: C. utilis, C. tro p icalis, C. k ru sei. IAEA-SM-134/6 271

3. TECHNIQUES D'ETUDE

Les techniques utilisées en génétique des levures ont été mises au point su r S. c e re v is iæ . Elles sont en général bien adaptées à l 1 étude des levures hétéro- thalliques, mais peuvent servir aussi dans le cas des levures homo- thalliques.

3.1. Sporulation

L'aération est indispensable à la sporulation des levures. C'est pourquoi la sporulation est recherchée en général en ensemençant en surface des milieux gélosés, inclinés en tube. Il est cependant possible d'obtenir la sporulation en milieu liquide à condition d'aérer fortement [2 1 ]. Les pH (7 à 9, 5) et 1' acétate sont favorables à la sporulation de S. cerevisiæ. En présence d 1 acétate, il se forme de nombreux asques a" 4 spores. La levure peut donner des asques à 1, 2, 3 et 4 spores, mais ces derniers sont les plus intéressants à étudier au point de vue génétique. La sporulation se produit quand la croissance s' arrête. Pour les Saccharomyces, les milieux les plus utilisés dérivent du milieu de Fowel. Cependant, bien d'autres milieux ont été décrits. Cer­ tains d'entre eux sont plus particulièrement utilisés pour des espèces dont la sporulation est difficile à obtenir. Les principaux milieux de sporulation sont les suivants: — Milieu de Fowel modifié, contenant de 1' acétate et 2 g/l d'extrait de levure. L'ensemencement doit être fait avec des cellules en bon état physiologique, cultivées sur un milieu riche et prélevées en phase de croissance exponentielle ou au début de la phase stationnaire. Pour les souches qui sporulent mal, il est parfois utile d'utiliser un milieu de pré- sporulation, riche en facteurs de croissance, recommandé par Lindegren. — Milieu de Gorodkowa, contenant du glucose ( 1 g/l) et de la peptone (10 g/l). La sporulation est bonne, mais tardive, — Tranches de carottes. La sporulation y est souvent abondante. — Milieu V 8 , prepare à partir d'un mélange de jus de fruits et de légumes du commerce. Ce milieu est voisin de celui préconisé par Lindegren pour la présporulation. Il est particulièrement utilisé pour les Debaryomyces.

3.2. Etude des spores

Dissection en masse

La sporulation est rarement complète et il est parfois nécessaire de se débarrasser des formes végétatives pour ne laisser subsister que les spores. Les parois des asques sont digérées en 1 ou 2 heures en incubant en présence d 1 enzyme de 1'hépatopancréas d'escargot. Après ce traite­ ment, les spores sont facilement libérées par agitation et peuvent être séparées des cellules végétatives, car leur paroi est hydrophobe et on peut les concentrer dans l'huile de paraffine. Pour cela, on émulsionne la suspension dans de l'huile de paraffine, on centrifuge pour briser 272 G ALZ Y et DUPUY

1 ' émulsion et on retrouve les spores dispersées dans la paraffine ou à la surface de séparation. La paraffine étalée sur bon milieu gélosé donne des colonies de cellules haploides. Bien que moins intéressante du point de vue génétique que 1' analyse des tétrades, la dissection en masse des asques peut suffire pour effectuer un travail de sélection portant sur un grand nombre de cellules. Cette méthode est la seule utilisable avec les souches qui sporulent mal et celles dont les spores germent mal. En partant de quantités suf­ fisantes de culture, on peut alors obtenir quelques colonies provenant de spores uniques. Certaines espèces ont des asques fragiles qui éclatent spontanément. С' est le cas, en particulier, des Kluyveromyces et aussi de certaines Hansenula. Bien que la dissection des asques soit encore possible, si l'on choisit des asques jeunes, 1 ' extraction en masse des spores consti­ tue une technique beaucoup mieux adaptée à 1 ' étude des espèces qui pré­ sentent cette particularité.

Analyse des tétrades

Les asques destinées à être disséquées sont traitées par l'enzyme de 1 ' hépatopancréas d'escargot jusqu'à ce qu' on ait atteint le stade fa­ vorable où la paroi de 1' asque est encore visible. Il est possible de con­ server cette suspension pendant quelques jours à la glacière. La dissection se fait dans la chambre de Hawthorne, suivant la tech­ nique de Mortimer [9] . On utilise un micromanipulateur équipé d'une aiguille recourbée à son extrémité. Les asques sont placées sur une gélose déposée à la face inférieure d'une lamelle. On prélève les quatre sp o res de 1 ' asque en une seule fois et on les transporte sur la partie de la gélose qui est stérile. La lame de gélose qui a servi à faire 1' isole­ ment est placée à la surface d'un milieu gélosé, et après 2 à 3 jours, chaque spore a donné une colonie.

4. ETUDE DES SOUCHES HAPLOIDES

Chaque spore donne un clone de cellules haploïdes. Il convient d'évi­ ter une multiplication exagérée, car il arrive que les souches haploïdes perdent un chromosome et deviennent aneuploides. On conserve souvent ces souches au froid à + 5°, en tubes de gélose inclinée. Il est ainsi possible de conserver longtemps une culture avant de la repiquer (un repiquage tous les 6 mois ou tous les ans). Ces souches subissent en général divers tests.

4.1. Homothallisme

Certaines espèces de levures sont considérées comme hétérothalliques (S. cerevisiæ), d'autres sont considérées comme homothalliques (S. rouxii). En fait, une mutation monogénique suffit pour passer de l'hétérothallisme à 1'homothallisme. Il est facile de vérifier ce caractère dans un clone issu d'une seule spore. Il suffit de cultiver le clone sur un milieu favorable à la sporulation. Si le clone est homothallique, il se formera des figures de copulation des zygotes et ensuite des asques. Si le clone est hétéro- thallique, aucune figure de copulation n' apparaîtra. IAEA-SM-134/6 273

4.2. Etude du signe

Dans les souches hétérothalliques, on recherche le signe d'un clone en le cultivant en mélange avec une souche test de signe connu. Il est prudent d'essayer deux souches tests: une de signe + et une de signe -. On repère ensuite les zygotes formés au microscope. La souche test et la souche à tester sont apportées en quantités équivalentes; elles doivent être en multiplication active. Après 24 h de multiplication, les levures forment un dépôt au fond du tube. Ce dépôt est riche en zygotes de grandes tailles, en figures de copulation et en cellules diplo’ides si les deux souches sont de signes contraires. Les cellules haploïdes sont petites et arrondies; les diplo'ides sont plus grandes et elliptiques. Le signe est sous contrôle monogénique. Dans 1 'asque, il doit donc y avoir deux souches de signe + et deux souches de signe -. Le milieu utilisé a la composition suivante; Yeast extract Difco 0, 5%, glucose 0, 5%.

4.3. Marquage génétique

Les souches utilisées en génétique sont en général «marquées». Elles ont des besoins en facteurs de croissance (acides aminés, bases puriques ou pyrimidiques, vitamines). Chaque besoin est contrôlé par un gène connu. Si on a croisé, par exemple, une souche déficiente pour la synthèse du tryptophane (gène trj ) avec une souche capable de le syn­ thétiser (gène TRjJ le diploïde obtenu sera hétérozygote TRX, tr1; et syn­ thétisera le tryptophane. A la mé'iose, deux spores hériteront du gène TRj, et pousseront en l'absence de tryptophane. Les deux autres spores hériteront du gène tr j, ne fabriqueront pas du tryptophane et seront incapables de pousser si on ne leur donne pas cet acide aminé dans le milieu de culture. Les clones haploïdes issus des spores sont testés systématiquement pour les facteurs de croissance qui étaient nécessaires aux deux parents. On prépare des milieux synthétiques contenant des sels minéraux, des vita­ mines, une source d'azote et une source de carbone, c' est-à-dire ce qui est nécessaire à toutes les levures. Ce milieu de base est appelé milieu minimum [5]; sa composition sera donnée en détail plus loin. Ce milieu est gélosé (3 ou 4%). Il reçoit en outre tous les facteurs de croissance nécessaires aux souches parents, sauf un. Prenons un exemple; on croise une souche tryptophane” adénine" et histidine" avec une souche uracile* et méthionine“. Les descendants du diploide obtenu peuvent présenter ces besoins. Le milieu test recevra ces cinq facteurs moins un; toutes les souches qui ne fabriquent pas le facteur manquant ne pousseront pas. Sur chaque milieu test, 2 spores pousseront et 2 ne pousseront pas parmi les 4 spores i' un asque. Le marquage génétique permet de repérer les contaminations ou de mettre en évidence 1' aneuplo'idie d'une souche. Malheureusement, les gènes utilisés pour le marquage génétique contrôlent souvent une déficience du métabolisme. Des souches ainsi déficientes ne peuvent pas en général résister à la sélection naturelle. Dans un milieu de fermentation industriel, en présence de souches sauvages, elles seraient impitoyablement éliminées. Naturellement, il est possible, par des croisements bien choisis, d'obtenir à partir de souches marquées pour des déficiences, des souches diploides 274 G ALZ Y et DUPUY prototrophes. En effet, les gènes contrôlant une déficience métabolique sont en général récessifs en présence de 1 ' allèle sauvage. Cependant, il n' est pas certain que l'hétérozygote n'ait pas aussi une vitalité diminuée. Pour les souches destinées à l'industrie, il est pré­ férable de marquer avec des gènes responsables de la résistance à des substances toxiques [17, 18]. Les marqueurs connus chez la levure S. cerevisiæ sont d'ailleurs actuellement très nombreux [l, 2]. Il serait aussi possible d'utiliser comme marqueur les gènes de la série pl qui modifient la morphologie coloniale et donnent un avantage sélectif aux souches dans certaines conditions expérimentales [7]. Les gènes res­ ponsables de la fermentation des glucides sont facilement utilisables [1 , 2 ]. Spiegelman et coll. [20] ont proposé un indicateur mélangé d'éosine et de bleu de méthylène: les souches qui fermentent le sucre présent dans le milieu donnent des colonies rouge foncé et les autres des colonies roses.

4.4. Mutations et sélection des mutants

Les cellules haploïdes et les cellules diploides de levure ont des pro­ priétés analogues. Tous les caractères pouvant présenter un intérêt in­ dustriel chez une souche diplo'ide peuvent être étudiés dans les clones haploïdes qui en dérivent. Il est donc logique de tester ces caractères sur des souches haploïdes où tous les gènes peuvent s'exprim er dans le phénotype. .La sélection peut ainsi être très efficace. C' est aussi sur les souches haploïdes que seront recherchés les gènes et les mutations cytoplasmiques destinés à marquer des souches. Un caractère recherché peut être trouvé dans un clone haploïde ou dans la descendance d'une souche prélevée dans la nature. Il est aussi possible de cribler dans un clone une mutation spontanée, si le caractère s'y prête. C'est le cas, par exemple, des gènes de résistance qui peuvent être facilement criblés. Enfin, il peut être utile d'induire des mutations par des agents chimiques ou physiques [ 1 0 ]. Pour des souches destinées à une utilisation industrielle, on peut craindre qu' un agent mutagène physique ou chimique ne provoque plusieurs mutations dans une même cellule. Certaines mutations peuvent échapper à 1' expérimentateur et diminuer la vitalité de la souche. La sélection de mutations spontanées devrait, en principe, éviter cet écueil. Cet aspect a été évoqué dans l'introduction à la mutagénèse présentée lors de ce colloque par le Pr Heslot. Les caractères à sélectionner varient avec les objectifs industriels. Cependant, quelques problèmes sont communs à toutes les industries. Le facteur le plus important est probablement la vitesse de croissance de la souche. Cette vitesse conditionne en partie le rendement industriel. De plus, une grande vitesse de croissance permet à une souche particulière de rester dominante dans le milieu soumis aux contaminations de souches sauvages. Un calcul simple montre qu' une très faible différence dans les temps de génération permet à la souche la plus rapide d'éliminer rapidement ses concurrentes [15]. Les techniques de laboratoire permettant de me­ surer la vitesse de croissance d'une souche pendant la phase exponentielle sont bien au point [6 ]. Cependant, ces techniques sont en fait insuffisantes pour juger les souches industrielles pour deux raisons. La première est que la vitesse de croissance d'une souche dans un milieu industriel, en culture continue ou discontinue, est presque toujours beaucoup plus faible IAEA-SM-134/6 275 que la vitesse observée au laboratoire. La deuxième est que les diffé- . rences de vitesse de croissance permettant l'élimination d'une souche par une autre sont trop faibles pour être mises en évidence sans doute possible par ces techniques [6 ]. Il est pratiquement nécessaire d'im agi­ ner, dans chaque cas particulier, un système expérimental permettant de placer les souches à tester dans des conditions proches de celles de l'industrie pour déterminer quelle est celle qui élimine ses concurrentes.

5. OBTENTION DE SOUCHES DIPLOIDES

Plusieurs techniques de croisement ont été décrites. Elles ne sont pas toutes utilisables pour toutes les souches. Certaines sont mal adaptées à 1 ' étude des souches industrielles.

5.1. Technique par sélection de prototrophes

Il est commode d'utiliser dans les croisements des souches auxo­ trophes présentant des besoins en facteurs de croissance différents. Les souches sont croisées par culture en mélange sur milieu liquide. Cette culture est étalée sur un milieu ne contenant pas ces facteurs, mais pou­ vant contenir éventuellement les autres facteurs dont les deux souches ont besoin. Les cellules diploides issues du croisement donnent une colo­ nie, alors que les deux souches parentes haploïdes ne poussent pas. Il est donc facile d'isoler le diploide issu du croisement. Cette technique n'est pas réalisable sous cette forme si on ne peut pas utiliser des mutations pour 1'auxotrophie. Cependant, elle, paraît adaptable. Il suffit, en effet, pour que le croisement soit facilement réalisable, de disposer d'un milieu où les deux clones haploïdes ne poussent pas, mais où le diploïde issu de leur croisement donne des colonies normales. Par exemple, la mutation «petite colonie végétative neutre» paraît utilisable. En effet, une souche haploïde marquée par cette mutation est incapable de donner une croissance sur un milieu synthétique ayant comme seule source de carbone un substrat non fermentescible (éthanol, lactate ou acétate, etc.). Après croisement entre une souche normale et une souche «petite colonie végétative neutre», on obtient un diploïde normal. La le­ vure haploïde mutante a une grave déficience respiratoire [19], mais il semble bien que toute trace ou conséquence de la mutation ait disparu chez le diploïde et chez sa descendance par voie végétative ou sexuelle. On peut donc par exemple croiser une souche haploïde sauvage et une souche haploïde «petite végétative neutre», marquée pour un gène de ré­ sistance dominant. Le diploïde obtenu peut seul former des colonies sur un milieu synthétique à base d'acide lactique contenant 1 ' agent toxique. La génétique mitochondriale a fait récemment de très gros progrès [4, 16] et d'autres mutations cytoplasmiques que les «petites végétatives neutres» sont peut-être utilisables. Il est également possible de croiser deux souches haploïdes marquées chac.uné par des gènes de résistance domi­ nants différents. 5.2. Technique de Lindegren (1943)

Les deux souches à croiser sont ensemencées dans un tube à essais contenant un milieu liquide: Yeast extract 0, 5%, glucose 0, 5%. Après 276 GALZY et DUPUY

24 h, des figures de copulation nombreuses sont visibles. Il est possible d'isoler une figure de copulation sous le microscope à l'aide d'une micro­ aiguille de verre montée sur le micromanipulateur. Cette technique, rela­ tivement délicate, est indispensable lorsque les deux souches parentes ne possèdent pas des besoins en facteurs de croissance différents ou n'ont pas de marqueurs permettant d'utiliser la méthode précédente [12,13].

5.3. Technique de Chen (1950)

Chen croise deux cellules haploïdes de signes opposés à 1' aide du micromanipulateur. Les deux souches à croiser sont cultivées en milieu liquide [3]. On dépose sous la lamelle qui forme le toit de la chambre humide une goutte de chacune des deux cultures. On dépose également quelques gouttelettes de milieu de culture stérile. A l'aide du micromani­ pulateur, on place au bord de chacune des gouttelettes deux cellules sans bourgeon, 1 ' une au contact de 1 ' autre, provenant chacune d'une des deux souches à croiser. La copulation est observée sous le microscope; elle se produit au bout de 2 à 6 heures. Le zygote formé émet des premières cellules de la diplophase. Celles-ci peuvent être immédiatement isolées, ou laissées dans la goutte de milieu. Après 24 heures, il s' est formé dans la goutte une culture qui peut être prélevée à 1 ' aide d'une pipette. Cette technique, bien que très délicate, peut être également très utile pour 1 ' étude des souches industrielles, car il n' est pas nécessaire que les souches soient marquées.

5.4. Technique de Winge et Lausten (1938)

De la même façon, Winge et Lausten croisent les ascospores deux à deux, dans la chambre humide de Winge, à 1' aide d'un micromanipulateur [23]. Ce procédé est délicat puisqu'il faut disséquer d'abord 1'asque, puis croiser les spores. Le croisement ne réussit que si les spores sont de signes opposés; dans le cas contraire, il faut recommencer les confron­ tations des spores deux à d;ux. La phase haploïde est supprimée. Cette technique est indispensable pour 1' étude d'espèces rigoureuse­ ment diplobiohtiques, où la copulation s'effectue en général avant l'éclate­ ment de 1'asque, entre deux spores voisines: Saccharomyces ludwigii H ansen. Cette technique d'hybridation a permis à Winge de produire une levure de boulangerie offrant des avantages industriels, mais Oppenoorth avoue avoir essayé, sans succès, d'obtenir des hybrides de levures basses de brasserie.-

5.5. Utilisation des souches diploides

La souche diploide peut être multipliée indéfiniment par voie végé­ tative. Il est souhaitable d'éviter la sporulation qui se produit souvent lorsque le milieu s' appauvrit dans les vieilles cultures. A chaque re ­ piquage des collections, la culture nouvelle est transférée au froid (+5°C) avant la fin de croissance. La multiplication cellulaire est presque arrêtée, mais les cellules conservent une vie ralentie qui permet une conservation de longue durée. IAEA-SM-134/6 277

C' est sous forme de souche diploïde que les gènes intéressants sont conservés de préférence. En effet, un gène peut muter. Dans une souche haploide, un gène n 1 existe qu 1 à un seul exemplaire; s'il mute dans quelques cellules, la sélection naturelle peut faire perdre le gène initial. Au con­ traire, chez un diplofde homozygote pour un gène, la mutation ne peut porter que sur un seul des deux gènes. La cellule deviendra hétérozygote, mais, dans de nombreux cas, son phénotype ne sera pas modifié. De toute façon, il sera facile de retrouver le gène intéressant dans la population. La sélection pour les divers caractères doit être effectuée de pré­ férence sur les souches haploïdes. Cependant, il ne faut pas perdre de vue que ce sont des souches diploides ou polyploïdes qui seront en défi­ nitive utilisées dans l'industrie. Tous les tests et, en particulier, l'étude de la vitesse de croissance et du pouvoir sélectif, devront être repris de la même façon avec les souches diploides.

6 . CONCLUSION

L' utilisation de la génétique pour la sélection des levures dans une optique industrielle est possible. Mais il est nécessaire de tenir compté de deux conditions préalables. Les critères de sélection doivent être bien définis. Il faut dans chaque industrie rechercher les caractères élémentaires intéressants. Par exemple, en brasserie, on attache une grande importance à la floculence et à l'atté­ nuation limite. Il convient d'étudier avec précision le contrôle génétique et les mécanismes biochimiques de ces phénomènes. Seule une connaissance parfaite de ces mécanismes permet de dégager les possibilités de criblage et de sélection des souches. Il faut donc prévoir une étude fondamentale dans chaque cas particulier, en fonction des souhaits présentés par les industriels. Pour mener à bien ces études fondamentales, on utilisera les méthodes classiques de la génétique et de la biochimie. Les techniques génétiques choisies pour une sélection industrielle doivent être simplifiées et modifiées. En effet, une fois connus les méca­ nismes contrôlant un phénomène, il faut sélectionner des souches utilisables dans l'industrie. Ces souches doivent porter le caractère recherché, mais doivent aussi garder toute leur vigueur. Le criblage et la sélection de ces souches ne pourront donc se faire qu'avec des méthodes simples, ce qui permettra d'obtenir un grand nombre de souches dont on n' utilisera que les plus vigoureuses. Le marquage génétique devra être utilisé avec pru­ dence pour ne pas diminuer la vitalité des clones marqués et de leurs descendants. En conclusion, nous pensons qu'il est difficile d'imposer dans un milieu industriel, sauf s'il est stérile, une souche nouvelle. Les méthodes génétiques doivent tendre à placer le caractère souhaité dans un contexte génétique tel que la sélection naturelle joue en faveur de la souche qui le porte. En terminant cette revue, nous émettons le voeu que ce Colloque établisse la liste des travaux, utilisant des méthodes génétiques, effectués sur les levures en vue d'améliorer leur utilisation industrielle. Nous in­ diquons quelques buts économiquement intéressants que l'on peut atteindre par la génétique et la sélection; il s'agit des points suivants: 278 GALZY et DUPUY

— marquage de souches par un caractère permettant de les reconnaître dans une population industrielle — obtention de souches résistantes à des antiseptiques — obtention de souches produisant davantage de biomasse — amélioration de la production de composés obtenus par fermentation — obtention de souches floculant facilement.

REFERENCES

[1] BÖRSTEL, R.C., von, Microb. Genet. Bull. Suppl. 19 (1963) 1-21. [2] BÖRSTEL, R.C. , von, Microb. Genet. Bull. Suppl. 31 (1969) 1-28. [3] CHEN SHIH'YI, Sur une nouvelle technique de croisements des levures, C .r. hebd. Séanc. Acad. Soi. Paris 230 (1950) 1897-99. [4} COEN, D ., DEUTSCH, J., NETTER, P., PETROCH1LO, E., SLON1MSK1, P.P., «Mitochondrial genetics, I. Methodology and phenomenology», Symp. Soc. exp. Biol., Symp. 24 (1969) 449-96. [5] GALZY, P ., SLONIMSKI, P .P ., Variations physiologiques de la levure au cours de la croissance sur l’acide lactique comme seule source de carbone, C.r. hebd. Séanc. Acad. Sei. Paris 245 (1957) 2 423-26. . [6] GALZY, P ., Observations sur Г utilisation d'un électrophotomètre pour 1' étude de la croissance des levures, Annls Technol. agrie. INRA (1958) 453. [7] GALZY, P ., Etude génétique et physiologique du métabolisme de l'acide lactique chez Saccharomyces cerevisiæ Hansen, Thèse, Annls Technol. agrie. 13 (1964) 109-259. [8] GALZY, P.,.PLAN, C ., ALBERT, J ., La génétique des levures, technique et méthodologie, Revue Ferment. Ind. aliment. 34 5 (1969) 161-67 et _24 6 (1970) 199-206. [9] JOHNSTON, J.R ., MORTIMER, R .K ., Use of snail digestive juice in isolation of yeast spore tetrads, J. Bact. 7J 2 (1959)’292. [10] HESLOT, H ., Les mécanismes moléculaires de la mutagénèse et la nature des mutations, Annls Amél. Pl. 15 2 (1965) 111-58. [11] LACROUTE, F ., Régulation de la chaîne de biosynthèse de l'uracile chez Saccharomyces cerevisiae, Thèse, Paris (1966). [12] LINDEGREN, C .C ., LINDEGREN, G ., A new method for hybridizing yeast, .Proc. natn Acad. Sei. 29 (1943) 306. [13] LINDEGREN, C .C ., The yeast cell, its genetics and cytology, Educational Publishers, St-Louis, Missouri (1949). [14] LODDER, J., The yeasts, A taxonomic study, North Holland Publishing Co., Amsterdam (1970). [15] LUMARET, R., RADIER, M ., GALZY, P., Essais de'sélection de souches de levures destinées à Г industrie (en préparation). [16] MOUNOLOU, J.C ., JAKOB, H., SLONIMSKI, P.P., «Mutations of mitochondrial DNA», Bioch. Aspects of Biogenesis of Mitochondria, Proc. Int. Symp. Bari (1968) 473. [17] PELLECUER, М ., GALZY, P., PASERO, J., Utilisations de mutations, pour le marquage des souches de levures dans l’industrie, J. Genet, appl. Microbiol, (à paraître). [18] RADIER, M ., GALZY, P.. PASERO, J., «Genetic marking of industrial yeasts», 1st Int. Symp. on Genetics of Industrial Microorganisms, Prague 23 August 1970, Book of abstracts, 216-17. [19] SLONIMSKI, P.P., La formation des enzymes respiratoires chez la levure, Thèse, Masson et Cie, Paris (1953). [20] SPIEGELMAN, S., SUSSMAN, R., PINSKA, E., Proc. natn Acad. Sei. USA_36 (1950) 591. [21] VEZINHET, F ., Contribution à l’étude de la sporulation de S. cerevisiæ Hansen, Mycopath. Mycol. appl. 40 1.(1970) 15-48. [22] WALT, J.P., Van der, The genus Syringospora, Quinquad Emend., Mycopath. Mycol. appl. _40 (1970) 231-43. [23] WINGE, ö ., LAUSTEN, О ., Artificial species hybridization in yeast, C .r. lab. Carlsberg 22 (1938) 235. IAEA-SM-134/15

EFFECT OF MICROORGANISMS ON THE STRUCTURE OF URANIUM RAW MATERIALS

F. BARBIC Institute for Technology of Nuclear Raw Materials, BRANKA KRAJINÔANIÔ Boris KidriS Institute of Nuclear Sciences, Belgrade, Yugoslavia

Abstract

EFFECT OF MICROORGANISMS ON THE STRUCTURE OF URANIUM RAW MATERIALS. The present preliminary studies describe the possibility of extracting uranium from raw material by microorganisms. The studies include the isolation and determination of the location of the microorganisms in the mineral ore. A comparison of the ecophysiological properties of the isolated types of microorganisms from different locations permits us to elucidate the conditions of uranium extraction from the ore by micro­ organisms. The results obtained show that the bacteria belonging to the genera Thiobacillus and Ferrobacillus exist in natural ore and in the water from mines. The possibility that a higher percentage of uranium can be extracted from the previously treated mineral was also indicated in the biological assay.

INTRODUCTION

The problem of separating metals from very poor mineral wastes has been the subject of interest for several years in the Microbiological Labora­ tory of our Institute. The results obtained were applied to the recovery of copper from sterile material as a halide. Our results have proved to be of great value in the study of bacterial flora from diverse mineral locations. This flora was. used for the separation of different minerals of uranium, antimony and molybdenum [ 1-5]. The isolated bacteria were capable of transforming the mineral form of uranium into a soluble form several times faster than the chemical extraction performed in a control test under identical conditions. In fact, the autotrophic bacteria in mineral beds under natural conditions play an important role in metal extraction and oxidation of sulphur, sulphide and bivalent iron [ 6 ], Therefore it is of great interest to use this property of bacteria for practical purposes in hydrometallurgy for metal isolation. The experiments of various authors [7-10] suggest a reduction in metal exploration cost when biological methods are used instead of conventional treatment. This means that the extraction uranium from mineral ore by means of microorganisms can be applied on an industrial scale.

MATERIALS AND METHOD

The mineral ore (8 kg) was first treated with a solution of sulphuric acid and then washed with water until the pH was neutral. The composition of the treated material was: U 360 g/t; Fe +2 0. 77%; F e +3 3. 14%; S 0. 53%;С03 0. 92%; S04 0. 76%.

279 280 BARBIÓ and KRAJINCANIÓ

TABLE I. CHEMICAL COMPOSITION OF MATERIAL AFTER ALKALIZATION

U Fe S CO, E xperim ent S ° 4 Cg/t) (°lo) (°Io) во) (°!o)

1. C ulture 282 3 .6 3 0 .1 0 1 .4 4 0 .2 2

2. Culture + 246 3 .6 5 0 .3 0 1 .1 2 0 .3 3 agents

3. Controls 320. 3 .7 0 0 .4 0 0 .8 0 0 .3 5

The bacteria Thiobacillus thiooxidans and Thiobacillus ferrooxidans were cultured in different combinations and concentrations. These bacteria were isolated and identified from other mines of uranium, copper and antimony by classical methods [11, 12]. During these experiments the definite determination of autochtone microbiological material was, however, not completed. The extraction was carried out in large cylindrical vessels in the alkaline solution to which a few drops of sulphuric acid were added. The rate of flow of the alkaline solution in the individual recyclations varied from 15-32 cm 3/litre. The chemical composition of the ore was analysed at the beginning and the end of the experiment. During the process of extraction the quantity of excreted uranium, Fe +2 and F e +3 were determined as well as the pH and Eh values. The quantity of excreted uranium was determined from the difference in the weight of uranium at the beginning and the end of the experiment.

EXPERIMENTAL RESULTS The thiobacilli and the ferrobacilli were isolated and determined from mine water (from the channel and from the walls), from ore that was either very rich or poor in uranium, and from sterile material. Depending on the specimen, the number of bacteria (total number) was found to vary fro m 102 to 104 /cm3. The mine water, from which Th. thiooxidans and Th. ferrooxidans were isolated, was found to contain: uranium from 4. 80 X 10"4 to 9. 50 X 10~4 g/1 in a total amount of iron ranging from 0. 270 to 0. 300 g/1. The ratio of Fe +2 to F e +3 was 1 : 10 and the pH ranged from 4 to 7.5. Extraction (alkalization) was carried out twice and lasted for 55 days with an interval of 35 days. The experiment was done in three ways. The results of the alkalization of uranium as well as other indices of oxidation are presented in Table I. When the quantity of uranium at the beginning of the experiment (360 g/t) was compared with the quantity at the end of the experiment it was found that 45 g/t (i.e. 21.66%) was separated by alkalization in the first test, 114 g/t (31.66%) in the second test and 40 g/t (11. 11%) in the control group. Proof of biological oxidation was the reduction of total sulphur (in culture, from 0. 53% to 0.10%) and an increase of total sulphates (from 0. 70% to 1. 14%) owing to the formation of sulphuric acid under the influence of bacteria. The carbonates were also found to decrease from 0.92% to 0.23%. IAEA-SM "134/15 281

It is important to point out that the consumption of sulphuric acid in the experiment was very low in accordance with our intention. The acid consumption, i. e. the difference between the quantity of acid added and the quantity remaining at the end of the experiment, was as follows:

Culture 485 g /t Culture agents 404 g /t C ontrols 520 g /t

The differences in the consumption of H2S04 in different assays indicate that maximum biological oxidation was attained in the second assay when the extraction was also at a maximum.

DISCUSSION

Our aim was to establish conditions for alkaline extraction of uranium from ore by means of bacteria. The ore was first treated with conc. sulphuric acid and 70% of the total uranium was extracted. During this procedure the structure of the ore was drastically altered and the natural environmental conditions for bacteria were changed.

CONCLUSIONS

l.O n the basis of these preliminary studies it could be concluded that the microorganisms of the genera Thiobacillus and Ferrobacillus are present naturally in the mine of Macedonia. Their high activity in terms of iron and sulphur oxidation was confirmed in this work. 2. Bacteria used in the experiment had the capacity to convert the uranium into a soluble form and to increase the utilization of the metal. In the comparison of control assays under identical conditions the alkaline extraction in biological treatments was found to take place more rapidly. 3. It was established that the process of biological oxidation of uranium alkaline extraction can be intensified by the addition of certain known agents.

REFERENCES

[1] DJERKOVIC, L., Mikrobiologija, Yugoslavia 2 No.2 (1965) 231. [2] MARJANOVIC, D., Rep. AEC, Yugoslavia (1963). [3] MARJANOVIC, D., Rep. AEC, Yugoslavia (1964). [4] MARJANOVIC, D., Mikrobiologija, Yugoslavia 2 No.2 (1965) 243. [ 5] MARJANOVIC, D ., BARBIC, F . , Z bornik rad . Rud. F a k ., Y ugoslavia (1969) 150. [6] LJALIKOVA, N .N ., Mikrobiologija, Moscow 29 No.3 (1960) 773. [7] ZIMMERLEY, D.C. et al., Engng Min. J. 159 No. 1 (1958) 89. [8] IVANOV, N.V., Microbiologija, Moscow 36 No.5 (1967) 849. [9] MILER, R. et al., Inst. Min. Metall. 72 (1963) 217. [10] HARRISON, V.F. et al., Can. Min. J. 87 5 (1966) 64. [11] CORRICK, J.D ., SUTTON, J.A ., Bureau of Mines, Dept, of Investigations (1961) 5718. [ 12] BERGEY' S Manual of Determinative Bacteriology, The Williams and Wilkins C o., Baltimore (1957).

CURRENT AND FUTURE OUTLOOK FOR APPLIED MICROBIOLOGY RESEARCH IN DEVELOPING COUNTRIES (Session 7)

Chairman

J. MEYRATH (Austria)

OPENING REMARKS ON THE STATUS OF APPLIED MICROBIOLOGY IN DEVELOPING COUNTRIES

J. MEYRATH Institut für angewandte Mikrobiologie, Hochschule für Bodenkultur, Vienna, Austria

Short contribution

What is applied microbiology? Here I am always reminded of the expression by Louis Pasteur who said: "il n1 y a pas de sciences appli­ quées, il n'y a que des applications des sciences". Nevertheless there are today a large number of departments and institutions that are termed 1 applied science departments' , ' applied microbiology departments' , etc. Although there has been a certain shift in the meaning of the word 'applied' over the years, the older and the present-day connotations can, perhaps, be reconciled if we consider that departments and institutions with the affix ' applied' are those whose main objective is to benefit mankind in the near future with the results of their research. In contrast, the research in ' pure science' laboratories is not directed to future application. The two disciplines are of equal importance and complement each other — a fact which should be borne in mind when programs are set up for developing countries. Until some twenty years ago most ideas for ' pure microbiology' were derived from industrial applications. Nowadays ' pure microbiology' and ' pure genetics' , as brought out in this Symposium, are so far developed that they can stand on their own even though there will always be some cross-fertilization of ideas between industry and 'pure science' . It is, of course, always difficult to know in advance whether academic research will lead to a practical application. In this Symposium we have heard of two practical applications of the chemoautotrophic bacteria whose properties were once considered to be of purely academic interest. The hydrogen bacteria (Knallgas bacteria) are now being envisaged for the production of single-cell protein and the sulphur bacteria (Thiobacillus) are already being used in the leaching process of low-grade ores. We have to face the fact that there is a great deal of starvation in developing countries at the present time. Thus, even though long-term projects are very important, we have to consider that short-term projects must be put into operation at the earliest opportunity. We know that agri­ cultural improvements are slow and that industrial progress is faster if the projects are properly handled. One quick way of introducing industrial technology into a developing country could be through industrial fermen­ tations. The quickest way to raise the living standard in a given area is to set up a factory, but before this can be done the feasibility of such a plan will have to be studied. One of the major problems will be the lack of available and sufficiently qualified personnel to run the factory. This is where the universities and research institutes can step in to supply

285 286 MEYRATH the trained staff. Once the first industry has been established, the trained personnel could be invaluable in helping to set up other industries. With this objective in mind, the first industry in a developing country should be convinced of the long-term wisdom of training more personnel than are required for their immediate use. Here, perhaps, the state could inter­ vene with the necessary encouragement and financial support. This philo­ sophy was emphasized by an Expert Meeting of UNIDO in December 1969, and a typical example quoted was the transfer of personnel from existing sugar refineries to the fermentation industry. The fermentation industry lends itself particularly well to introducing technology into a developing country since there are some fermentations which can be carried out with little know-how and since the minimal factory size is usually smaller than in other major industries. I have emphasized the importance of training and education; of equal importance is the selection of research topics in universities and develop­ ment centres in developing countries. We all know about the grave short­ age of protein in these countries and for microbiologists the major interest is, of course, the so-called single-cell protein (S.C.P.). We may ask once again; Is this really a short-term project? Are the circumstances such that the production costs of S.C.P. are comparable to other sources of protein? Is the questionable acceptability a grave hindrance to introducing S.C.P.? With regard to the last question there are certainly obstacles to using S.C.P. for human consumption, but there are a lot of possibilities for its use in feedstuffs for poultry, pigs and cattle. With regard to production costs it has been said in the past (see, for example, a study done under the U.N. Economic and Social Council in 1967 on the Production and Use of Protein) that S.C.P. is not yet competitive in price with other available protein sources, in particular soy bean meal. At present this statement is probably no longer universally true, as can be illustrated by a study made in the Philippines where large quantities of soy bean meal are imported from the U.S.A. for animal feeding. One way of replacing some of this imported meal would be to convert molasses, together with other additives, into yeast. The production costs can be compiled as follows:

Costs (ff) per t dry yeast

Raw materials Molasses, 4 t at ^ 90/t 360 Urea and other additives 140 Electrical energy (aeration, agitation pumping) 82 Steam (evaporation) (from fuel) 41 L abour 90 Depreciation of the plant 97 Interest on equipment 192

T o ta l 1 0 0 2 SHORT CONTRIBUTION 287

Production costs per kg dry yeast (50% crude protein): Tf 1.0 = $0.15/lb protein. Price of 1 kg soy bean meal (40% crude protein): 1.0 to ^ 1.10.

The conditions necessary for obtaining such low production costs are: specific energy requirement (electrical energy) S 0.51 kWh/kg dry yeast; specific air supply s 17 m3/kg dry yeast; productivity ä 2.1 kg yeast/m3h. One other condition which should be strongly emphasized is that no energy is required for cooling purposes. Yeast production is strongly exothermic so that large quantities of cooling water are required provided that the average water temperature is about 5 degC below the temperature for optimal yeast growth. There are hardly any high-yielding yeast strains known that have an optimal temperature above 30 - 32°C. This tempera­ ture is frequently also that of the water found in tropical areas where many developing countries are situated. So, either one has to find a yeast strain with a considerably higher optimal growth temperature or one has to resort to artificial cooling by refrigeration which requires a lot of energy. For each ton of yeast produced, 4X 106 kcal have to be removed when grown on carbohydrates and considerably more when grown on hydrocarbons. This is equivalent to about 4.5X103 kWh. Under Philippine conditions this would add an extra cost of 370 per ton of dry yeast and an increase in the production costs of m o re than 30%. This is clearly a good example of how a developing country can adapt existing technology to its own needs. In this case a suitable yeast strain has to be selected and this strain is most likely to be found in the native region. More examples could be quoted of the efforts that developing coun­ tries could make in adapting existing technology to their own needs but this single example may suffice to illustrate the point I wanted to make.

IAEA-SM-134/30

MICROBIOLOGICAL AND PHARMACEUTICAL ACTIVITY IN GHANA

A. review

M. CAURIE Food Research Institute, Accra, Ghana

Abstract

MICROBIOLOGICAL AND PHARMACEUTICAL ACTIVITY IN GHANA; A REVIEW. The review deals with the microbiological and pharmaceutical activities in Ghana from 1900 to date. It indicates that most microbiological effort has been concentrated on cocoa disease control with comparatively little work on other crops. Similarly, soil microbiology has received practically no attention. Food microbiology, which was previously neglected, is now receiving serious attention. Work in medical microbiology has indicated an increasing drug resistance of Salmonella and Staphylococcus spp. to certain antibiotics. Pharmaceutical research in Ghana has been concentrated on alkaloid-containing plants to find new substances as well as to identify plants from which existing known substances of medicinal value may be found as alternative sources of supply. Several new finds have been reported in this program of research.

INTRODUCTION

Microbiological investigations in Ghana are conducted mainly in the University of Ghana, University of Science and Technology and by various departments of the Ghana Ministry of Agriculture. Most of the research functions of the latter department, however, have since 1962 been re­ organized into a separate, but Government controlled, Council for Scientific and Industrial Research (C. S. I. R. ). The latter body has the responsibility of co-ordinating all research activities in Ghana and directly controls a number of Research Institutes including seven that conduct research in agriculture. These Institutes are the:

Cocoa Research Institute (Ghana) Crops Research Institute Soil Research Institute Animal Research Institute Forest Products Research Institute Institute of Aquatic Biology Food Research Institute

This paper reviews the microbiological activities in agriculture and medicine in Ghana from 1900 to the present as well as the pharmaceutical activities from 1958 to date.

289 290 CAURIE

COCOA RESEARCH

Ghana is an agricultural country and naturally its research, including microbiological, activities are centered around agriculture and in particular around the cocoa industry which is the main stay of the nation's economy. Like any crop, cocoa is plagued by many diseases and one of the first to be reported (Dade, 1922, 1927a) was a collar-crack root disease caused by the fungus Armillaria melea (Vahl exFr) Kummer. This fungus develops in the medullary rays and cracks the trunk longitudinally from ground level to a height of 4ft up the cocoa tree; the tree eventually collapses and falls. Several other minor root diseases, characterized by leaf fall and die-back of twigs, are listed by Wharton (1962). Wharton (1962) also lists two blight diseases. One attacks single leaves and is called white thread disease and the other attacks whole canopies and is called the horse hair blight (Dade 1927d). These diseases are caused by the fungi Marasmius scandens Masse and Marasmius equicrinis Mull respectively. Of greater economic importance, but still of less importance in Ghana compared with other countries, is the black pod disease of cocoa. This disease is caused by the fungus Phytophthora palmivora. The disease blackens the pod tissue within 4 days from initial attack and destroys the beans inside the pods within 4 weeks if the pods are left on for so long. This disease has been studied with respect to its mode and conditions for infection (Dade, 1927b; Wharton, 1962; Bimpong, 1969), spread into the stalk and flower cushions, (Dade, 1928-29) pathogenicity (Dade, 1927e) strains and their distribution (Turner 1960a; 1960b, 1961; Wharton, 1962; Ashby, 1929; Dade, 1927b) and dissemination and control measures (Bunting, 1922, 1929; Wharton, 1955, 1958; Owen, 1951; Dade, 1927b and c; Hammond, 1958; Attafuah, 1965). From these studies it has become evident that chemical control of the disease is uneconomic because of the difficulties imposed by heavy rainfall and peasant agricultural practices. Indeed, field sanitation and management practices involving the removal of dead pods and husks from trees have proved the best method of control. In 1936 a farm er from the eastern region of Ghana showed the Ministry of Agriculture a number of badly deformed cocoa branches from his farm. The disease was investigated and identified by the plant pathologist as a severe plant disease (Stephen, 1936) and was given the name "Swollen shoot and die black". This incident was instrumental in the setting up of the inter-territorial West African Cocoa Research Institute based in Ghana to study cocoa diseases to save the industry. The apparent sudden onset and the fact that surveys showed its wide occurrence on cocoa farms in Ghana made people believe that it was the result of some environmental change such as decreased humidity and soil deterioration. In 1938, however, Posnette (1940) traced the cause to a virus infection which had for its vector mealybug insects (Cotterell 1943; Posnette and Strickland, 1948) which are associated with Crematogaster ants (Strickland, 1951b; Cornwell, 1957; Hanna et al. 1956). The importance and alarm caused by the discovery of swollen shoot in Ghana may be judged by comparing the number of published papers on black-pod disease and on swollen shoot. Between 1940 and 1960 the IAEA-SM-134/30 291 number of published papers on black-pod and on swollen shoot was in the ratio of one on black-pod to 10 on swollen shoot (Ripley, 1968). In this period over 50 virus strains which possess varying degrees of virulence and induce shoot, leaf and pod symptoms (Entwistle 1958; Dale 1962; Kenten and Legg 1965) were isolated. It was found, for example, that most Ghanaian isolates produce swellings of various sizes but the Kpeve virus also isolated in Ghana only produced leaf-mottle symptoms (Thresh and Tinsley, 1959; Posnette, 1947). Indeed, some avirulent strains of the deadly New Juaben isolate of Ghana do not produce swellings at a l l . The pattern of the spread of the cocoa virus has been shown to agree with the movement of the mealybug vector which may be by walk-migration (Cornwell, 1958), passive dispersal by air currents (Strickland, 1950; Cornwell 1960, 1955b) andby other agencies (Thresh 1958b; Posnette, 1943). While prospects for controlling fungus diseases were good, studies showed that virus diseases of cocoa can only be prevented by the painful measure of grubbing or "cutting out" of diseased trees and the elimination of alternative plant hosts (Attafuah, 1965) on farms. However, this measure was started soon after Posnette (1940) had demonstrated the virus nature of the disease. Several attempts have been made at biological control (Nicol et al. 1950, Rojter et al. 1966), but though successful in the laboratory it has proved a failure in the field (Nicol 1953). The m icro­ biological studies on cocoa have concentrated on disease control with only one publication (Knapp, 1935) on the microbiology of the wet bean with fermentation. Similarly, only three reports on the storage of the dry beans (Powell and Wood, 1951; Scott and Hudson, 1937; Anonymous, 1932) are on record.

RESEARCH ON OTHER CROPS

The detailed and sustained microbiological research on cocoa, which has enabled the fundamentals of cocoa diseases to be understood and controlled, has not characterized work on other crops. Microbiological studies on these crops have been very fragmentary and unsustained. For example, over the period under review there were three publications on cocoyam root rot (Wright, 1930; Shephered, 1938; Posnette, 1945) and two papers each on tobacco leaf curl (Shephered, 1938; Stephen, 1955) and diseases of para rubber (Dade, 1927f; Robertson, 1947). There were also single publications on Cercospora disease of the bambara groundnut (Teyegagah, 1970), bacterial soft rots of cabbages and lettuce and hard rots of other vegetables caused by Sclerotium rolfsii and Rhizoctonia solani (Dade, 1934b), virus mosaic diseases of legumes (Lister and Thresh, 1956), virus rosette disease of groundnuts (Brunt and Bonney, 1964), a pepper disease caused by the fungus Leveillula taurica (Lev) Ar (Ayesu- Offei, 1966) and a fungal disease of the Carica papaya caused by the fungus Phyllactinia cor.ylea (Clerk and Ankora, 1969a). Diseases of the tomato fruit caused by the fungus Sclerotium rolfsii (Dakwa, 1965) and its control with herbicides (Clerk and Bimpong, 1969) have also been studied. In the latter studies it was observed that S. rolfsii germinated well in a 300 ppm aqueous solution of atrazine, prometryne and ramrod, but was completely suppressed in 400 ppm aqueous solution 292 CAURIE of prometryne and 500 ppm of ramrod. A. concentration of up to 1000 ppm of atrazine, however, permitted up to 48% germination. The apparent lack of sustained microbiological activity on individual crops may be attributed to the fact that there have been no serious epidemics affecting any of these crops as has been experienced with cocoa and maize which resulted in the setting up of the West African Cocoa Research Institute (now Cocoa Research Institute, Ghana) and the former West African Maize Rust Research Unit. Another reason may be that most of these crops were not economically important to the colonial administration. This policy of waiting till trouble starts may not be good, but it is expedient with the limited amount of manpower available. The maize Rust Research Unit was established in 1950 to investigate and find a solution to a maize rust epidemic caused by the rust fungus Puccinia polys ora that hit West Africa that year. During the short life span (1950-61) of this Unit several technical papers were contributed on maize rust disease by several investigators (Stanton and Cammack, 1953; Stanton 1954; Cammack, 1956a, b, c; Rhind, 1952; Blaine, 1953). Before this Unit was set up, however, Bunting (192 5, 1926, 1928) of the Gold Coast (Ghana) Ministry of Agriculture had reported on diseases of maize and other graminaceous plants. Two economically important crops, whose diseases have received some study in Ghana, are the lime tree and the coconut palm. The most important disease affecting the lime industry in the Central Region of Ghana is a virus die-back disease, tristeza, transmitted by the black aphis T ox opt era citricidus. This disease has been studied by various authors (Posnette, 1952; Hughes, 1949, 1954; Hughes and Lister, 1953; Martin, 1954a) and has been shown to be effectively controlled by using disease-resistant rootstocks, e. g. Rhangphur lime rootstock. The Cape St. Paul or coconut wilt disease poses a difficult problem to microbiologists interested in this crop in Ghana. Some years ago Dring (1958) stated that it is not certain whether it is caused by a fungus, a virus or an eelworm. This statement is still applicable. The disease is characterized by nut dropping, yellowing, bronzing and drying out of the lower leaves and at the last stage of the disease the heart leaf dies and rots with the roots. This disease has been investigated by Leather (1959) and Martin (1954b). Fungus of the Fusarium spp. are responsible for a number of plant diseases and tuber rots and this group of fungi has been taxonomically studied by Gordon (1960).

SOILS

Crops and soils are closely linked but the microbiology of the latter has been very little investigated. A. comparative account of the nitrogen cycle in Ghanaian Savannah and forest soils has been rendered by Meiklejohn (1962). The inhibitory effect of soil extract on the growth of some fungi has also been reported by Clerk (1969b). In this study it was found that aqueous extracts of humus-rich soil inhibited the germination and growth of Beauveria bassiana and Paecilomyces farinosus. The inhibitory substance was fungistatic rather than fungicidal and was inactivated by heat treatment. IAEA “SM -1 3 4 /3 0 293

WATER

By building a dam across the Volta River in Ghana, the largest man- made lake has been formed and some microbiological observations on the lake have been made by Biswas (1967, 1969). She found that seasonal changes in temperature did not affect bacterial counts in the lake. She also found that the broad composition of the microbial population resembled that found in temperate lakes. There was, however, an indication of seasonal increase of spore formers especially in the surface and bottom layers. Presumptive coliforms have usually been low on the lake. Microbiological observations on other bodies of water have been studied by Abrahams (1963) and Chelty (1966).

FOOD

Microbiological investigation into foods in Ghana has, until recently, been very limited. Apart from Fishlock's (1930a) paper on the moulding of copra which was followed by Rasper and Kuskova (1964) on the same subject and Abrahams' (1962) investigation of a local festival food suspected of causing diarrhoea to some subjects, no previous papers had been published on any microbiological studies of foods in Ghana. In the latter investigation a considerable number of Clostridium welchii in some food samples were isolated. With the setting up of the C. S. I. R. and, in particular, of the Food Research Institute in Ghana in the latter half of the sixties, microbiological studies on food are being taken seriously within the limits of our resources. Several studies have been conducted on milk. Abrahams and Laryea (1968) have made a comparative study of the bacteriology of raw milk produced on the Accra Plains. This study has established the extent to which the quality of raw milk produced on university farms exceeds that on peasant farms. It has also been established (Caurie, 1970c) that there is considerable post-processing contamination of pasteurized filled milk sold in Accra. The shelf-life of the milk through existing distribution channels was, however, found satisfactory (Caurie, 1970d). Fermented maize dough is a basic raw material in the preparation of a number of local food items. The microorganisms responsible for the later stages of the fermentation have been shown to be a mixed population of lactic-acid bacteria and yeasts (Christian, 1970). The most common bacterium was the homofermentative Pediococcus cerevisiae. The tuber of the cassava plant is one of the staple food items in Ghana and one of the methods of preparation is to dry the peeled cut up tubers under field conditions. During the drying, which may last up to 6 or 7 days, a number of fungi grow on the exposed chips. Under humid drying conditions excessive mycofloral growth renders the chips very dirty and unappetising. The mycoflora which develop on the chips have been surveyed (Caurie, 1967). Over eighty percent of the colonising fungi were identified as Cladosporium herbarum together with smaller percentages of A spergillus, Pénicillium, Sporendonema and other species. Chemical changes caused by some Aspergillus species in the root tuber have also been studied (Clerk and Caurie, 1968). 294 CAURIE

Fermented cassava tuber pulp forms the basis for the preparation of a staple food item, gari, in Ghana. Changes in the microbial population and some concomitant chemical changes during the fermentation process have been studied (Caurie, 1970b), It was found, for example, that there are no successional changes of microorganisms during the fermentation and that both yeasts and bacteria grow together in identical rythms during the fermentation. The end of the fermentation is indicated by a decline in bacterial and an increase in yeast numbers. The stability of dehydrated foods free of microbial growth or deteriorative chemical reactions resulting from products of microbial metabolism depends on the provision of optimal storage conditions. These conditions have been studied and worked out by Caurie (1970a, 1971a). The effect of storage temperatures on the microbiological quality of some foods manufactured in Ghana has also received some study (Ako-A.ddo, 1970).

FOOD PATHOGENS Of particular importance and closely related to foods are organisms belonging to the Salmonella-Shigella group which may be taken in with the food to cause enteric fevers, dysentery and, in some cases, death. Gamble and Harris (1953) in their studies have reported that the fever symptom associated with Salmonella infection may be simulated by Pseudomonas pyocynea. Some Salmonella species found in the human have also been isolated from lizard droppings (Vella, 1956), from the python (Rewall et al. 1948) and from chickens (King and Gellatly, 1955). Hughes (1954a) considers beef as a potential source of Salmonella dublin infection in Ghana. The resistance to drugs of several Salmonella species have recently been studied by Sodhi et al. (1970) and shown to maintain a fairly high resistance to antibiotics. Other organisms likely to contaminate food from human carriers are those of the Staphylococcus species. The bacteriology of wound sepsis which abounds in these species has also been studied (Sodhi et al. , 1968). The antibiotic resistance of the most common Staphylococcus species occurring in wounds, S. pyogenes, has similarly been reported by Abrahams and Laing (1964a, 1964b). In the latter studies as much as 84% of the cultures were highly resistant to aureomycin, streptomycin and as high as 96% to chloramphenicol. Sodhi et al. (1968) found as much as 93% of the cultures were resistant to penicillin which indicated an increase of 9% in the antibiotic resistance of Staphylococcus species to penicillin in Ghana within 7 years. On the distribution of contaminants on the human skin Heman-Ackah (1959b) found no uniform distribution, but reported that the skin may be effectively sterilized with either a solution of mercuric iodide (Heman- Ackah, 1959a) or by the usual scrubbing of the skin followed by treatment with a disinfectant.

MEDICAL

A. search through the medical literature from 1900 to date reveals that though many diseases of microbial origin have been reported, little or no research has been done on the microorganisms themselves. IAEA-SM-134/30 295

Salles and Sodhi (1965) have, however, studied the characteristics of tubercle bacilli isolated in Accra while the resistance of these organisms in chemotheraphy has been studied by Bell and Brown (1961, 1963). Other microorganisms isolated and studied include Norcadia spp. from heart blood (Macfie and Ingram, 1921b), a number of anonymous mycobacteria (Sodhi et al. , 1966), bacteria responsible for bacteriuria in pregnant women in Accra (Sefcovic,et al. , 1967), and for the first time in Ghana medical history the much dreaded Cholera bacilli were isolated in Ghana last year (Pobee et al. , 1970). In plague control, Burgess (1927) was able to prepare vaccines from selected strains of Bacillus pestis and later Burgess (1930) studied the bacteriological variation in Pasteurella pestis, the causative agent of the disease. Some bacteriological tests and techniques have also been developed and investigated by Hughes (1953d) and Eddington (1953c). On the veterinary front, Oppong (1969) has studied the susceptibility of pigs, sheep and goats to the type A. virus of foot and mouth disease. The exposure of these animals to the virus showed infection in the pig and subclinical or inapparent infection in the sheep and goats. In addition to the regular microbiological research activity that goes on in Ghana a considerable amount of non-research, routine, microbiological activity goes on in private establishments and in Government Public Health Reference Laboratories, and in teaching especially at the University level.

PHARMACEUTICAL RESEARCH

Pharmaceutical investigations are conducted at both the Chemistry Department of the University of Ghana and at the Pharmacy Department of the University of Science and Technology (U.S. T. ). It is, however, at the latter University that the bulk of the investigations are carried out. Here, the Pharmacy Department has, since 1958, been specialising in alkaloid-containing plants in a special program on medicinal and toxic herbs which are abundant in Ghana and which may provide useful starting material for the development of new substances of medicinal value. This program is not only out to look for new substances but also to identify alternative plant sources from which known substances of medicinal value may be extracted. In this program at the U. S. T. , several new alkaloids have already been extracted, characterized and named. From the root bark of the plant Teclea verdoorniana Excell and Mendonca (Syn. T. grandifolia) a total of eight alkaloids have been identified (Tackie, Kofi-tsepo and Hadzija, 1967). Out of this number three have been crystallized. One of these has been shown to be identical to the known alkaloid evoxanthine while the other two are new alkaloids (Tackie and Martin, 1970). One of these new alkaloids has been shown to be a furoquinolone base and the other an acridone-type base and have been named tecleadme and verdoornine respectively; tentative structures have been assigned to them. Another new alkaloid, funiferine, has been isolated from the root bark of Tiliacora funifera by Tackie and Thomas (1965) and these investi­ gators are working on its structure. 296 CAURIE

The pharmacological value of some alkaloids extracted from local herbs have been studied. The alkaloid, isorotundifoline, extracted from the leaves of certain Mitragyna plant species has been shown to be comparable to papaverine hydrochloride in its musculotropic and spasmolytic effect (Szreniawski and Tacker, 1964). Githens (1948) has stated that many Solanum species in West Africa contain the alkaloid solanine. One of the commonest Solanum species in West Africa is S. torvum and its leaf was investigated to find the level of solanine, if any, present in it. Investigations (Tackie, 1959a) showed no alkaloidal content but it appeared that the active principle in the S. torvum leaves is a glycoside which yields a mixture of sugars of which dextrose is the principal component. This active principle was shown by Gyan (1959) to have a direct depressant effect on frog heart muscle and an atropine-like effect on an isolated guinea pig ileum. Talalaj (1964-67) has extracted a number of essential oils from various Ghanaian plant species. From the dried leaf of the lemon grass C.ymbopogon citratus (D. C. ) Stapf, 2-2.5% oil has been distilled which contains 70-78% citral (Talalaj, 1964a). Citral is a starting material for the synthesis of vitamin A. and some synthetic perfumes. The leaves, flower tops and flower heads of Lippia multiflora have yielded 0.80%, 1.5% and 2.1% oil, respectively, on distillation (Talalaj, 1964b). T he oil contained 2.6 to 3.9% cam phor. The content of essential oil from the dried leaves and flowers of Ocimum viride (Willd) were found to lie between 2.2 and 2.4%. This oil contained 39-41% of thymol as the main constituent (Talalaj, 1964c). Among the medicinal plants in Ghana are some members of the Callistemon species. By steam distillation of the dried leaves of C. lanceolatus and C. rigidus 1% oil was recovered which had a cineole content of 63% and 54% respectively (Talalaj, 1965a). The nutmeg seed is a common spice in Ghana from which Talalaj (1965b) has extracted 4% oil containing 2% phenols which are probably carv acro l. Unlike the plants so far studied the leaves of the Cedrela mexicana contain very small amounts of oil. The oil content was found rather concentrated in the wood and it contained only small amounts of phenols which are also believed to be carvacrol. (Talalaj, 1965c). The vetiver oil content of Vetiveria zizanioides Stapf grass roots in Ghana (Talalaj, 1965d) contained 19-20% ketones. The essential oil content (3-4%) of ginger grown in Ghana has been found to be of very high quality (Talalaj, 1966a). Like the Callistemon species the cineole is the main constituent of the oil extracted from the Melaleuca leucadendron L (Talalaj, 1966b). The oil content of the fruits of Xylopia aethiopica (Talalaj, 1966b) also contains 6 - 8% of cineole. Some fruits of the lime tree grown in Ghana have been reported (Talalaj, 1966c) to contain much higher levels of oil extracted by distillation than is reported in the literature. Elemi oleo — resin collected from incised wounds on Canarium schweinfurthii grown in Ghana — has been shown (Talalaj, 1966d) to be a very good source of elimi oil. This oil was reported to contain phellandrene and high amounts of free alcohols and low levels of ester. There was a complete absence of aldehydes and phenols. IAEA-SM-134/30 297

The resin from the Daniella oliveri plant has been reported to contain as much as 50% oil (Talalaj, 1966f). The oil content of the dried leaves of Eucalyptus citriodora has similarly been shown to be a good source of citronellal (60.4%) (Talalaj, 1966g). Though the characteristics of the oil from the leaf of the Cinnamomum zeylamicum Nees grown in Ghana bear close identity to Ceylon cinnamon oil, the oils from the bark of species from the two localities show differences in certain essential characters (Talalaj, 1967). These differences have been attributed to soil differences. The activity of the Pharmacy Department at U. S. T. is not only limited to medicinal herbs but extends to the investigation of non-herbal pharmaceutical study as well. Thus, the efficacy of some corticoid drugs (Liniecka, 1968) and the use of certain clay preparations in traditional medicine (Tackie, 1959b; Heman-Ackah, I960) have been reported. Tackie (1960) has also identified the active principle of a popular hair dye in Ghana and advised on its proper application. The group of workers at the Chemistry Department at the University of Ghana has, in parallel investigations, also extracted a number of alkaloids from Fagara spp. (Torto et al. , 1969) and several coumarins from Afraegle paniculata (Schum and Thom) (Adjangba et al. , 1968). In the latter studies three furo-coumarins, imperatorin, heraclenin and xanthotoxin as well as two chromenocoumarins, xanthyletin and xanthoxyletin, were extracted from various parts of the A., paniculata plant.

CONCLUSION

This review has dealt with the microbiological activities in the fields of agriculture, food and medicine from 1900 to date. It has also re­ ported on some pharmaceutical research in Ghana. Some aspects of certain subjects have received more attention than others. The reason for this is economic since at our present stage of development it is not possible to have men, in sufficient numbers, to tackle problems that are not considered of prior importance.

BIBLIOGRAPHY

ABRAHAMS, Charles. (1962). "Homowo" diarrhoea, West Afr. Med. J. П. P-

ABRAHAMS, C. A. (1963). Bacterial content of Accra coastal waters, Ent. Parasit. Res. Inst. Rep. 1962/63.

ABRAHAMS, C .A ., LAING, E. (1964a). Antibiotic resistance in Staphylococcus pyogenes isolated in Accra ( 1956-1959). Gh. Med. J. 3 p. 67.

ABRAHAMS, C .A ., LAING, E. (1964b). Antibiotic resistance in Staphylococcus pyogenes isolated in Accra (1957-1960). Association between antibiotics, Gh. J. Sei. 4 p. 124.

ABRAHAMS, C. A ., LARYEA, A.M. (1968). A preliminary investigation of the bacteriology of raw milk produced by cows on the Accra plains, Gh. Med. J. 7 p. 100.

ADJANGBA, M .S., TORTO, F. G., ASOMANING, W .A., SEFCOVIC, P. (1968). Extractives of Afraegle paniculata (Schum & Thom) Ptl. Coumarins from petroleum extract, Gh. J. Sei. £ p. 87. 298 CAURIE

AKO-ADDO, A. (1970). The effect of storage temperature on the microbiological quality of locally manufactured foods, Food Res. Inst. Report.

ANONYMOUS (1962). Seawater, fresh water and cocoa beans, Gold Coast Farmer 1 (7) p. 70.

ASHBY, S.F. (1929). Strains and taxonomy of Phytophthora palmivora Butler (P. faberi Manbl). Trans. Brit. M ycol. Soc. 14 p. 18.

ATTAFUAH, A. (1965). Microbial control in the laboratory of a fungus pathogenic to cocoa, Gh. J. Sei. 5 p. 92.

ATTAFUAH, A. (1965). Occurrence of cocoa viruses in wild plant species in Ghana, Gh. J. Sei. 5 p. 97.

AYESU-OFFEI, E. N. (1966). Studies on the germination of conidia of Leveillula taurica (Lev) Ar, M. Sc. Thesis, Univ. of Ghana.

BELL, W ,]., BROWN, P.P. (1961). Bacteriological resistance in cases of pulmonary tuberculosis in Accra. West Afr. M ed. J. UJ P* 140.

BELL, W .J., BROWN, P.P. (1963). Bacteriological resistance in patients receiving chemotherapy for pulmonary tuberculosis in Ghana, West Afr. Med. J. 12 p. 181.

BIMPONG, C.E. (1969). Studies on factors influencing zoospore motility and Chemotaxis in Phytophthora palmivora (Butl) Butl, M. Sc. Thesis, Univ. of Ghana.

BISWAS, E. Ruth, I. (1967). Bacterial ecology of the Volta Lake at Ajena, Volta Basin Res. Proj. Tech. Rep. X p. 18.

BISWAS, E. Ruth, I. (1969). Some microbiological observations on the Volta Lake, Gh. J. Sei. £ p. 21.

BLAINE, M. A. (1953). Maize rust disease in the Gold Coast, West Afr. Inter-Territorial Secret. Memo 6.

BRUNT, A .A ., BONNEY, J.K. G. (1964), Aphid and mechanical transmission of virus causing "Rosette" disease of groundnuts in Ghana, Trop. Agrie. Trin. 41 p. 299.

BUNTING, R.H. (1922). Dead cocoa pods a menance to the cocoa industry, Gold Coast Agrie. Comm. Soc. J. 2 p. 6.

BUNTING, R.H. (1925). Diseases of corn, Gold Coast Dept. Agrie. Pam. 4.

BUNTING, R. H. (1926). Local cereal diseases in the records of the mycological division, Gold Coast Dept. Agrie. Bull.7 p. 25.

BUNTING, R.H. (1928). Fungi affecting graminaceous plants of the Gold Coast, Gold Coast Dept. Agrie. Bull. 10.

BUNTING, R.H. (1929). Diseases of cocoa and measures for their control, West. Afr. Agrie. Conf. 1st P ro c ., p. 86.

BURGESS, A. S. (1927). The selection of a strain of Bacillus pestis for the preparation of vaccine with special reference to the effect of animal passage on virulence, J. Hyg. (Camb) 26_ p. 152.

BURGESS, A. S. (1930). Virulence, immunity and bacteriological variation in relation to plague, J. Hyg. (Camb) 30 p. 165.

CAMMACK, W.R. (1956a). Notes on the more important fungi affecting maize in Nigeria, West Afr. M aize Res. U nit M em o 6.

CAMMACK, W.R. (1956b). Notes on the introduction of Puccinia polysora Undeiw. into Africa, West Afr. Maize Res. Unit, Memo 7.

CAMMACK, W. R. (1956c). An analysis of the measurement of uredospores of Puccinia polysora in West Africa, West Afr. Maize Res. Unit, Memo 8.

CAURIE, M. (1967). Studies on the mycoflora associated with drying chips of the tuber of Manihot esculenta Crantz, M. Sc. Thesis, Univ. of Ghana.

CAURIE, M. (1970a). A new model equation for predicting safe storage moisture levels for optimum stability of dehydrated foods, J. Fd Technol. 5 (3) p. 301.

CAURIE, M. (1970b). The fermentation of cassava tuber pulp “ biological and chemical changes, Food Res. Inst. Report. IAEA-SM-134/30 299

CAURIE, M. (1970c). The microbiological quality of pasteurised filled milk sold in Ghana — Fan White Milk, Food. Res. Inst. Report.

CAURIE, M. (1970d). The effect of temperature on the shelflife of Fan White Milk, Food Res. Inst. Report.

CAURIE, M. (1971a). Practical approach to water sorption isotherms and the basis for the determination of safe moisture levels in dehydrated foods, J. Fd Technol. j3 (1) (in press).

CHRISTIAN, W.F.K. (1970). Lactic acid bacteria in fermenting maize dough,Gh. J. Sei. 10 p. 22.

CHELTY, M .T. (1966). Chemical and bacteriological analysis of Okobeyeye State Farm Pond, Nsawam, Inst. Aquat. Biol. Memo 26.

CLERK, G. C ., CAURIE, M. (1968). Biochemical changes caused by some Aspergillus species in root tuber of cassava (Manihot esculenta Crantz).

CLERK, G .C ., ANKORA, J.K. (1969a). The conidial apparatus of Phyllactinia corylea, Gh. J. Sei. 9 p. 111.

CLERK, G .C ., BIMPONG, C.E. (1969b). Influence of three herbicides on germination of Sclerotium rolfsii, Gh. J. Sei. 9 p. 115.

CLERK, G. C. (1969b). Influence of soil extracts on the germination of conidia of the fungi Beauveria bassiana and Paecilomyces farinosus,J. Invert. Path. ¿3p. 120.

CORNWELL, P.B. (1955b). Mealybug studies, Rept. West Aft. Cocoa Res. Inst. 1954/55, p. 45.

CORNWELL, P. W. (1957). An investigation into the effect of cultural conditions on populations of the vectors of virus diseases of cocoa in Ghana with an evaluation of seasonal population trends, Bull. ent. Res. 48 p. 375.

CORNWELL, P.B. (1958). Movements of the vectors of virus diseases of cacao in Ghana. 1. Canopy movements in and between trees, Bull. ent. Res. 49 p. 613.

CORNWELL, P.B. (1960). Movements of the vectors of virus diseases of cacao in Ghana. 2. Wind movements and aerial dispersal. Bull. ent. Res. 5J[ p. 175.

COTTERELL, G.S. (1943). Entomology Section Rep., Centr. Cacao Res. Stn. Tafo 1938 42 p. 51.

DADE, H.A. (1922). ‘Collar crack*. A new disease of cocoa, Gold Coast Agrie. Com. Soc. J. 1_ p. 241.

DADE, H.A. (1927a). 'Collar crack'of cocoa. Bull. Dept. Agrie. Gold Coast 5 p. 1.

DADE, H.A. (1927b). Economic significance of cacao pod diseases and factors determining their incidence and control, Bull. Dept. Agrie. Gold Coast 6 p. 1.

DADE, H.A. (1927c). Dissemination of cacao pod diseases by invertebrates, Gold Coast Dept. Agrie. Bull. 13 p. 93.

DADE, H.A. (1927d). Thread diseases of cacao in the Gold Coast, Gold Coast Agrie. Com. Soc. J. ^ P* 9.

DADE, H. A. (1927e). A comparison of pathogenicity of various strains of Phytopthora faberi (P. palmivora) on c a c a o pods, Gold C oast D ept. A grie. Bull._13 p. 89.

DADE, H.A. (1927f). Mouldy rot of para rubber, Gold Coast Dept.. Agrie. Pam. p. 11.

DADE, H.A. (1928a). Factors determining the incidence of diseases of cacao pods, Gold Coast Dept. Agrie. Bull. 7 p. 28.

DADE, H.A. (1928b). The relation between diseased cushions and the seasonal outbreak of *black pod* disease, Bull,Dept. Agrie. Gold Coast Yearbk. ^3 P• 85.

DADE, H.A. (1929). Further notes on cushion canker of cacao, Gold Coast Dept. Agrie. BulLYrbk. 16 p. 135.

DADE, H.A. (1934b). Collar rots of vegetables and flowers, Gold Coast Farmer 3 p. 130.

DALE, W .T. (1962). Diseases and pests of cocoa. A. Virus diseases In: Agriculture and Land Use in Ghana (WILLS, J.B ., Ed.), Oxford Univ. Press, Accra.

DAKWA, D. T. (1965). Studies on the physiology and pathogenicity of Sclerotium rolfsii Sacc causing fruit rot of tomato (Lycopersicum esculentum) Mill, M. Sc. Thesis, Univ. of Ghana.

DRING, D.M. (1958). Cape St. Paul wilt disease of coconuts in the Keta District, Ghana Farmer 2 p. 50. 300 CAURIE

EDDINGTON, G. M. (1953c). A note on the Ide test, West Afr. Med. J. (n. s. ) 2 p. 209.

ENTWISTLE, P. F. (1958). Mealybugs IV - Their relationship to swollen shoot disease of cocoa in Ghana, Ghana Farmer 2 p. 129.

FISHLOCK, W .C. (1930a). Moisture and mould in copra under Western Province conditions, Gold Coast Dept. Agrie. Bull. 22 p. 233.

GAMBLE, M .E.G ., HARRIS, F. C. (1953). Pseudomonas pyocynea simulating enteric fever and developing during antibiotic treatment, W. Afr. Med. J. (n. s. ) 2 p. 79.

GORDON, W. L, (1960). Taxonomy and habits of Fusarium species from tropical and temperate regions, Canad. J. Bot. 38 p. 643.

GYAN, E. A. (1959). An investigation into the pharmacological action of an aqueous extract of Solanum torvum leaf, W. Afr. Pharm. ^p . 61.

HAMMOND, P.S. (1958). Notes on the progress of pest and disease control in Ghana, Rep. Cocoa Conf. Lond. 1957, p. 110.

HANNA, A.D^JUDENKO, E., HEATHERINGTON, W. (1956). The control of Crematogaster ants as a means of controlling the mealybugs transmitting the swollen shoot virus disease of cocoa in the Gold Coast, Bull. ent. Res 47 p. 219.

HEMAN-ACKAH, S.M. (1959a). The evaluation of antiseptic activities of solutions of contrimide, iodine, mercuric iodide and chloroxylenol at different times after application to skins of Africans, West Afr. РЬагтЛ p. 3.

HEMAN-ACKAH, S.M. (1960). A note on native clay used medicinally in Ghana, West Afr. Pharm. 1_P*61*

HEMAN-ACKAH, S. M, (1959b). A determination of bacterial distribution on African skins treated in various ways, West Afr. Pharm. 1 p. 36.

HUGHES, W. A. (1949). Lime disease in the Gold Coast, Nature, Lond. 164 p. 767.

HUGHES, W .A ., LISTER, C. A. (1953). Lime die-back in the Gold Coast: a virus disease of the lime Citrus aurantifolia, J. Hort. Sei. 28 p. 131.

HUGHES, W.A. (1954). Lime die-back in the Gold Coast, Comm. Mycol. Conf. 5th Rep,, p. 142.

HUGHES, M. H, (1953d). A useful blood culture technique, West Afr. Med. J. (n. s.) 2 p. 94.

HUGHES, M. H. (1954a). Salmonella dublin infection in the Gold Coast, West Afr. Med. J. (n. s. ) 3 p. 59.

KENTEN, R. H ., LEGG, J. T. (1965). Observations on the purification and properties of cocoa swollen short virus, Gh. J. Sei. 5 p. 221.

KING, E. D ., GELLATLY, D. (1955). A paratyphoid С outbreak in young chickens in the Gold Coast, West Afr. Med. J. (n. s.) 4 p. 96. KNAPP, A. W. (1935). Scientific aspects of cocoa fermentation. II. Succession of organisms and explanations of high temperatures, Bull. Imp. Inst. 33 p. 147.

LEATHER, R.J. (1959). Further investigations into the Cape St. Paul wilt of coconuts at Keta, Emp. J. Expt. Agríe. 27 p. 67.

LINIECKA, A. (1968). On the general properties of some corticoid drugs and their application in some mucous membrane diseases of the mouth, West Afr. Pharm. _10 p. 18.

LISTER, R. M. THRESH, J.M . (1956). A mosaic disease of leguminous plants caused by a strain of tobacco- mossaic virus, Nature, Lond., pp. 155, 175, 1047.

MACFIE, J. W. S ., INGRAM, A. (1921b). A fungus of the genus Norcadia cultivated from heart blood, Ann. trop Med. Parasit. 15 p. 283.

MARTIN, E.B. (1954a). Virus die-back disease of limes and othei citrus diseases in the Gold Coast, Comm. Mycol, Conf. 5th Rep., p. 144.

MARTIN, E. B. ( 1954b). The coconut disease in the Keta area of the Gold Coast (Cape St. Paul wilt), Dept, of Agrie. Accra.

MEQCLEJOHN, J. (1962). Microbiology of the nitrogen cycle in Ghana soils, Emp. J. Expt. Agrie. 30p.ll5. IAEA-SM-134/30 301

NICOL, J., OWEN, H., STRICKLAND, A.H. (1950). Biological control of mealybug vectors of swollen shoot virus of cocoa, Nature, Lond. 165 p. 490.

NICOL, J. (1953). The control of mealybug vectors of cocoa swollen shoot virus, Proc. W. Afr. int. Cacao Res C o n f., p. 18.

OPPONG, E. N. W. (1969). Relative susceptibility of pigs, sheep and goats to the type A virus of foot and mouth disease, Gh. J. Sei. 9 p. 54.

OWEN, H. (1951). Cacao pod diseases in West Africa, Ann appl. Biol. 38 p. 715.

POBEE, J.O. M ., GRANT, F., SALLES, C.A. (1970). Case report of Cholera, Gh. Med. J. 9 p. 306.

POSNETTE, A. F. (1940). Transmission of swollen shoot, Trop. Agrie. Trin. 17 p. 98.

POSNETTE, A. F. (1943). Control measures against swollen shoot virus disease of cacao, Trop. Agrie. Trin. 20 p. 116.

POSNETTE, A.F. (1945). Root-rot of cocoyam (Xanthosoma sagiti folium). Trop. Agrie. 22 p. 164.

POSNETTE, A.F. (1947). Virus diseases of cocoa in West Africa. I, Cocoa viruses 1A, IB, 1C and ID, Ann. appl. Biol.34 p. 388.

POSNETTE, A .F., STRICKLAND, A.H. (1948). Virus diseases of cacao in West Africa. 3, Technique of insect transmission, Ann. appl. Biol. 35 p. 53.

POSNETTE, A.F. (1952). Virus decline disease of citrus, World Crops 4 p. 64.

POWELL, B.D., WOOD, G. A.R. (1951). Storage, transport and shipment of cocoa. 1, Prevention of moulding in storage, World Crops 1_1 p. 314.

RASPER, V., KUSKOVÄ, R. (1964). The conditions of optimal storage of dessicated coconut. 1, The spoilage of coconut by microorganisms, Listy Cukrovarnicki 80.

REWALL, R.E., TAYLOR, J., DOUGLAS, S.H. (1948). A new salmonella type (Salmonella takoradi) isolated from a python, Mon. Bull. Min Hlth. pub. Hlth. Lab. Sew 7 p. 266.

RHIND, D. (1952). The rust disease of maize, West Afr. Inter-territorial Sec., Agrie & Forestry Res. Memo 1,

RIPLEY, P. O. (1968). Coordination of agricultural research work in Ghana including forestry and hydrobiology Pt III. Bibliography of published information on agricultural research done in Ghana 1891-1966, C.S.I.R. Publication.

ROBERTSON, N. F. (1947). A disease of rubber buddings in the Eastern Province caused by Phytophthora palmivora Bull., Dept, of Agrie. Gold Coast Newsletter 2 (П) p. 33.

ROJTER, S., BONNEY, J.K ., LEGG, J.T , (1966). Investigations into the use of pathogenic fungus (Cephalosporium sp. ) as a means of controlling the mealybug (Pseudococcidae) vectors of swollen shoot virus in Ghana, Gh. J. Sei (3 p. 110.

SALLES, C .A ., SODHI, H.S. (1965). Some characteristics of tubercle bacilli isolated in Ghana, West Afr. Med. J. 14 p. 198.

SCOTT, J. L., HUDSON, W.R. (1937). Effect of sea water on mould in cocoa beans, Gold Coast Dept. Agrie. Bull. 23 p. 62.

SEFCOVIC, L., SODHI, H .S., SINCLARS, C.B.G. (1967). Bacteriuria in pregnant women (A bacteriological study of ante natal cases attending Jamestown Clinic, Accra), Gh. Med, J. 6 p. 2.

SHEPHERED, E.F. S. (1938). Cocoyam root-rot in the Gold Coast, West Afr. Agrie. Conf. 3rd (1) p. 83.

SHEPHERED, E.F.S. (1938). Tobacco leaf-curl, West Afr. Conf. 3rd (1) p .87.

SODHI, H .S., SALLES, C.A ., SHARMA, V. N., CHOWDRY, G.A. (1966). Anonymous mycobacteria, A clinical and bacteriological study, Gh. Med. J. 5 p. 88.

SODHI, H .S., DJORJEVIC, L., MINGLE. J. (1968). Bacteriological study of wound sepsis, Gh. Med. J. 7 p. 199.

SODHI, H .S., AMENUVOR, Lucia and Boi Doku, F. S. (1970). Drug resistance of salmonellae isolated at Korle Bu Hospital, Gh. Med. J. £ p. 163. 302 CAURIE

STANTON, W .R., CAMMACK, R. H. (1953). Research notes on the rust disease of maize in West Africa caused by Puccinia polysora Underw., W. Afr. Maize Research Unit Memo 1.

STANTON, W .R., (1954). A note on the rust resistance and yielding ability of some maize lines from the Imperial College of Tropical Agriculture, Trinidad, W. Afr. Maize Research Unit Memo 2.

STEPHEN, W. F. (1936). A new disease of cocoa in the Gold Coast, Gold Coast Farmer 5 pp. 122 and 124 [see also Trop. Agrie. 14 (1937) p. 84].

STEPHEN, R. C. (1955). Control measures for tobacco diseases, Gold Coast Dept. Agrie. Newsletter Suppl. 9 (101) p. 1.

STRICKLAND, A. H. (1950). The dispersal of Pseudococcidae (Hemiptera - Homoptera) by air currents in the Gold Coast, Proc. R. ent. Soc. Lond. (A) 25 p. 1.

STRICKLAND, A.H. (1951b). The entomology of swollen shoot of cacao. 2. The bionomics and ecology of the species involved, Bull. ent. Res. 42 p. 65.

SZRENIAWSKI, Z., TACKER, A.N. (1964). Preliminary pharmacological investigation of the alkaloid isorotundifoline, West Afr. Pharm. £ p. 72.

TACKIE, A.N. (1959a). A chemical investigation of Solanum torvum leaf, West Afr. Pharm. 1 p. 10.

TACKIE, A.N. (1959b). Cawu, West Afr. Pharm. J.P-66-

TACKIE, A. N. (I960). The hair dye - "Yomo", West Afr, Pharm. 2 p. 11.

TACKIE, A .N ., THOMAS, A. (1965). Characterisation of a new alkaloid (funiferine) from Tiliacora funifera (Tiliacora warneikii), Gh. J. Sei. 5 p. 11.

TACKIE, A .N ., KOFI-TSEPO, W. M ., HADZIJA, B.W. (1967). Notes on the constituents Teclea verdooriana of Ghana. Part I, West Afr. Pharm. 9 (2) p. 24.

TACKIE, A .N., MARTIN, B.D. (1970). Notes on the constituents Teclea vetdootiana of Ghana. Part 2. West Afr. Pharm. _12(1) p. 2.

TALALAJ, S. (1964a). Essential oil from Cympobogon citratus grown in Ghana, West Afr. Pharm. 6 p. 75.

TALALAJ, S. (1964b). Essential oil from Lippia multiflora from Ghana, West Afr. Pharm. £ p. 97.

TALALAJ, S. (1964c). Essential oil from Ocimum viride (Willd) grown in Ghana, West Afr. Pharm. 6 p. 113.

TALALAJ, S. (1965a). Essential oil of some Callistemon species grown in Ghana, West Afr. Pharm. 7_ p. 24.

TALALAJ, S. (1965b). Essential oil from Monadora myristica grown in Ghana, West Afr. Pharm. 7 p. 64.

TALALAJ, S. (1965c). Essential oil of Cedrela mexicana M. Roem from Ghana, West Afr. Pharm. 7 p. 87.

TALALAJ, S. (1965d). Essential oil from Vetiverai zizanioides Stapf cultivated in Ghana, West Afr. Pharm. 7 p. 106.

TALALAJ, S. (1966a). Essential oil from ginger (Zingiber officinale) grown in Ghana, West Afr. Pharm. 8_ p. 2. TALALAJ, S. (1966b). Essential oil from Melaleuca leucadendron L cultivated in Ghana, West Afr. Pharm. 8 p. 24.

TALALAJ, S. (1966c). Essential oil from lime fruits from Ghana, West Afr. Pharm. 8 p. 46.

TALALAJ, S. (1966d). Elimi oil from Canarium schweinfurthii growing in Ghana, West Afr. Pharm. 8 p. 48.

TALALAJ, S. (1966e). Essential oil of Xylopia aethiopica from Ghana, West Afr. Pharm. 8 p. 72.

TALALAJ, S. (1966f). Essential oil from Daniellia oliveri and D. ogea resins collected in Ghana, West Afr. Pharm . 8_p. 90.

TALALAJ, S. (1966g). Essential oil of Eucalyptus citriodora Hooker grown in Ghana, West Afr. Pharm. 8 p. 117.

TALALAJ, S. (1967). Essential oil of Cinnamomum zeylancium Nees growing in Ghana, West Afr. Pharm. 9 p. 1

TEYEGAGAH, A. (1970). Studies on the development, germination and survival of conidia of Cercospora canescens Ellis and Martin and the development of leafspot of Voandzeia subterránea Thouars caused by Cercospora canescens, M. Sc. Thesis, Univ. of Ghana.

THRESH, J.M . (1958b). The spread of virus disease in cacao, Tech. Bull. W. Afr. Cocoa Res. Inst. 5. IAEA-SM-134/30 303

THRESH, J.M . TINSLEY, T.W . (1959). The viruses of cacao. London, Crown Agents, Tech. Bull. W. Afr. C ocoa Res. Inst. 7.

TORTO, F.G., DADSON, B.A., SEFCOVIC, P., MENSAH, I.A. (1969). Alkaloids from Bagara spp., Gh. J. Sei. £ p. 3.

TURNER, P.D. (1960a). Strains of Phytophthora palmivora Buth from Theobroma cacao. 1, Isolates from West Africa, Trans. Brit. Mycol. Soc. 43 p. 665.

TURNER, P.D. (1960b). Variation in Phytophthora palmivora on Theobroma cacao in West Africa, Nature, Lond. 186 p. 495.

TURNER, P.D. (1961). Strains of Phytophthora palmivora from Theobroma cacao II. Isolates from non- African countries, Trans. Brit. Mycol. Soc. 44 p. 409.

VELLA, E. E. (1956). Note on S alm o n ella christiansborg, W. Afr. M ed. J. (n. s) 5_ p. 170.

WHARTON, A.L. (1955). Blackpod disease, Rept. W. Afr. Cocoa Res. Inst. 1954-55, p. 49.

WHARTON, A. L. (1958), The development and application of chemical control methods against black-pod disease in West Africa, W. Afr. J. Biol Chem. I, III.

WHARTON, A.L. (1962), Blackpod and minor diseases. In: Agriculture and Land Use in Ghana, (WILLS, J.B ., Ed.), Oxford Univ. Press, Accra.

WRIGHT, J. (1930). Root-rot of cocoyams, Gold Coast Dept. Agrie. Bull. 23.

IAEA-SM-134/33

USE OF ISOTOPES AND RADIATION IN THE STUDY OF MICROBIAL ASPECTS OF RUMINANT NUTRITION *

ESTHER BALOGH, A .A . ADEGBOLA Department of Animal Science, University of Ife, Ile-Ife, Nigeria

Abstract

USE OF ISOTOPES AND RADIATION IN THE STUDY OF MICROBIAL ASPECTS OF RUMINANT NUTRITION. The rumen can be regarded as a large fermentor where the microorganisms live in a controlled environ­ ment. The authors summarize the microbiological and physiological activities of the rumen from the point of view of nutrition. Some ideas are put forward for the possible improvement of ruminant nutrition.

DISCUSSION

J. MEYRATH: Have any tests been carried out in Nigeria on the use of urea and other inorganic nitrogen compounds for feeding cattle (ruminant) in order to reduce, or even eliminate, the expensive feeding with protein m atter ? ESTHER BALOGH: We are using urea successfully in feeding experi­ ments. We have also made attempts to use various tropical grasses with high carbohydrate and protein content.

* Since the subject matter of the paper presented at the meeting bears only a marginal relevance to the main theme of the Symposium, it was decided to print here only the abstract and the discussion that followed the presentation.

IAEA-SM-134/29

UTILIZATION OF MICROORGANISMS FOR THE PRODUCTION OF FOOD AND METABOLITES IN THE. PHILIPPINES

LYDIA M. JOSON Biological Research Center, National Institute of Science and Technology, National Science Development Board, Herran, Manila, Philippines

Abstract

UTILIZATION OF MICROORGANISMS FOR THE PRODUCTION OF FOOD AND METABOLITES IN THE PHILIPPINES. The present and future research activities of the Biological Research Center, National Institute of Science and Technology are discussed. These are in line with the government's agricultural and industrial programs. The technical and economic feasibilities of producing microbial food and metabolites on a commercial scale will depend largely on the selection and strain improvement of microorganisms by the use of radiation and radioisotopes and on the adaptation to modern fermentation technology.

INTRODUCTION

Although the Philippines has vast agricultural and natural resources, she also has the highest birth rate in Asia. Coupled with this problem of population pressure is the big imbalance in the economy which is not uncommon to most developing countries. Agricultural and industrial programs are, however, being stepped up through the application of science and technology to meet the growing demands on the social and economic structures of the country (National Economic Council, 1970). In spite of the considerable achievement made in agriculture through the scientific use of high-yielding varieties, fertilizers, irrigation and pesticides, and of fast-growing stocks and feeds for livestock and fishes, the country's import of food is still large and represents a huge drain on the foreign exchange reserves. Another big drain is the import of chemicals needed as intermediates in industries, of antibiotics for combating diseases of man, animals and plants, and of vitamins for human and animal nutrition. The government, recognizing the tremendous importance of science and technology to the development and social progress of the country, created the National Science Development Board (NSDB) and invested it with power to maximize the contribution of science and technology to national progress. One of its implementing agencies in matters relating to non-nuclear research and development activities is the National Institute of Science and Technology (NIST). Cognizant of the g reat needs of the country, i. e. the increasin g d e­ mands for more food, feeds and fertilizer, chemical intermediates, antibiotics and vitamins, etc. , and of the resources available, such as

307 308 JOSON cane sugar, molasses, coconut water, copra meal, rice bran, rice and peanut hulls, corn-steep liquor, sulphite liquor, etc. , the Biological Research Center (BRC) of the NIST has directed its research activities in microbial technology to the production of food and feeds (Baens-Arcega, 1969) and of im portant m etabolites.

CURRENT RESEARCH AT NIST

Food

The Philippines, as the biggest exporter of desiccated coconut, copra and coconut oil has immense quantities of waste coconut water. This waste product, with 4% total solids containing 2% sugar and growth factors, also consitutes a waste disposal problem to the industries. Processes for the economic utilization of coconut water in combination with other abundant by- and waste products have been developed by converting these raw materials into protein-rich and palatable foods through the activities of microorganisms.

Y east

Yeasts (Rhodotorula pilimanae, R. rubra) with desirable characteristics such as high contents of protein and vitamins, attractive colour and acceptable flavour are being propagated on coconut water and molasses. The strawberry yeast, R. pilimanae Hedrick et Burke (Hipolito, Domingo and Sarte, 1965) has an attractive strawberry colour and an appetizing flavour. Its taste was found very acceptable when used as protein supple­ ment in the preparation of cookies. Conditions for maximizing the yield and the quality of amino acids are being further studied.

Mushroom mycelium

The mushroom mycelium of Volvariella volvacea (Bull, ex Fr. ) Sing (A tacador-R am os, Palo, Villadolid and C ruz, 1967), the banana mushroom with a mushroom flavour, has been propagated in submerged culture using coconut water and sucrose. Soups and sauces prepared from this mycelium have a very desirable flavour. The protein yield and vitamin content of the mycelium are being further improved. Sub­ merged culture propagation of mushroom mycelium of different edible species capable of yielding high protein and vitamin content, and with a better flavour than the banana mushroom, is being carried out.

V inegar

A quick process for vinegar production from sugared coconut water on a pilot-plant scale is being developed with the intention that it should become a cottage industry in the rural areas of the coconut region. The slow process, developed by Diokno-Palo and Vilela (1968), for making quality vinegar from the alcohol produced from coconut water enriched with 15% sucrose using Acetobacter rancens Beijerinck var turbidans IAEA-SM-134/29 309

Frateur (Maceda and Palo, 1967) is being improved. A search for better acetic-acid producing organisms suitable for trickled and submerged methods of production is being continued.

Nata

Nata (Lapuz, Gallardo and Palo, 1967) is a cellulosic polysaccharide produced by Acetobacter xylinum from sugary materials. It is a delicious dessert when cooked in thick syrup with or without added flavour. An improved commercial production of it is being developed by using fast- growing organisms capable of synthesizing data of different textures and hardness in easily available and economical containers from coconut water or coconut milk and sucrose. Nata has domestic and export markets.

Feeds and fertilizer

To keep pace with the increased agricultural program several microbial methods for the production of feeds and fertilizer have been investigated.

Algae

An ecological study of algae in Philippine waters has been undertaken to determine the quality and quantity of algae that can be economically utilized for food, feed and fertilizer, and for valuable extractives. The propagation of unicellular and multicellular algae with desirable characteristics such as high protein content, high growth rate and digestibility combined with easy harvesting is being investigated. These algae can be used as feeds for animals and fishes and as fertilizer in rice fields. Improvement in the propagation of Chlorella pyrenoidosa Chick (Palo, Rodulfo and Balita, 1965) is being continued.

Phytoplankton

The propagation of phytoplankton as food for milk fish has been undertaken. Milk fish is extensively cultivated in fish ponds for local and foreign consumption.

Mould

Attempts have been made to improve the production of a protein supplement for poultry feed from moulded copra-meal residue. Moulded copra-meal residue is a waste product in the preparation of proteolytic enzymes by Aspergillus oryzae (Baens-Arcega, Maranon and Palo, 1956).

Y east

Yeasts for food are also being propagated for feed supplementation. 310 JOSON

Chemical intermediates

The production of chemicals which can be made economically by microbial synthesis and which are in great demand in local and foreign markets are being developed.

Organic acid

Studies are being made on the production of citric acid by yeasts and Aspergillus niger from sucrose and molasses which are both major agricultural products and by-products of the country.

Polysaccharides

The production of dextran as a blood-plasma extender by species different from those reported by Baens-Arcega and Arguelles, 1960, is being developed. Methods for extracting alginic acid, agar and carrageenin from various species of algae are being devised.

Enzym es

The microbial production of enzymes has decided advantages, such as rapid rate, relatively low cost and unlimited source of supply, over the production from animal and plant sources (Arima, 1964). The multifaceted uses of enzymes make their production of economic importance.

Proteinases

The recent innovation of adding enzymes to detergents to improve their action has spurred the search for more proteolytic enzymes that are heat stable and active at a wide range of pH, especially in alkaline conditions. These enzymes are being produced for inclusion in detergents as well as for bating material and as a meat tenderizer.

C ellulases

These enzymes can convert wood, wood by-products and other cellulosic agricultural by-products to fermentable substrates as well as increase the digestibility of grain and other cellulosic industrial and agricultural by-products for use in chicken and certain animal feeds (Casida, Jr. , L. E. , 1968). The latter activity is made use of to modify coconut flour, a product of the coconut industry, to make it acceptable as food.

P ectin ases

Pectinase from moulds and bacteria is being produced for retting of industrial plant fibres. IAEA-SM-134/29 311

Antibiotics

The importation of antibiotics in 1966 for medical and non-medical uses amounted to US$ 1 200 000 (International Trade Center, 1969). Because of their importance and the increasing demand for these compounds, their production from locally available raw materials is deemed necessary. A. systematic screening of Actinomycetes (Sevilla-Santos and de Leon, 1962; Sevilla-Santos and Bernardo), bacteria and. Basidiomycetes (Sevilla- Santos, Encinas and Leus-Palo, 1964) for potential antibiotic producers has been undertaken.

Against plant pathogens

Efforts are being concentrated on the production of antibiotics against plant pathogens, specially against the blast and blight diseases of the staple crops, rice and corn.

Anti-cancer

A.ctinomycin-like compounds were found to be produced by several streptomyces. Studies have been made on their production, isolation and identification (Joson, Sevilla-Santos, Librea and Bernardo, 1968) as well as on their pharmacology and toxicity (Angeles et al. , 1970). Further studies are in progress.

Vitamins and other growth factors

Vitamins for human, and animal nutrition are also imported in large amounts and with almost the same value as antibiotics. To ease the situation vitamins are being produced that can be incorporated into feeds.

V itam in B 12

Concomitant with the screening program for potential antibiotic producers is the search for organisms capable of synthesizing vitamin B j2 and other growth factors which can be used for human and animal nutrition. Methods for their propagation and for the production of vitamin B 12 in agricultural and industrial wastes are being investigated.

BIOTECHNOLOGICAL EXTENSION PROJECT

A. recent undertaking of the BRC is the biotechnological extension work whose objective is to bring the fruits of research and development on the economic utilization of microorganisms to the people, specially those in rural areas. These are improvements in the ancient arts of fermentation such as vinegar and nata production from coconut water, fermented fish sauces such as "patis" and "bagoong" and soy sauce (Baens-A.rcega, 1969), fermented fish known as "burong isda" (Orillo and Pederson, 1968) from surplus fishes, the preparation of "angkak" (Palo and Maceda, 1960) from rice for colouring "buro" and "bagoong", 312 JOSON vegetable preservation such as pickling and sauerkraut preparation. These different processes can supply people with food in times of scarcity and can be developed into cottage industries to augment the local in­ habitats’ income.

FUTURE OUTLOOK

The eventual commercial production of these microbial foods, feeds and metabolites will be in line with the government's efforts to maximize agricultural production and to help alleviate the balance of trade problems by establishing export-oriented industries which mainly utilize indigenous raw materials. However, these efforts are dependent on many factors. The most important of these are: (1) the selection and improvement of high-yielding strains that are capable of utilizing cheap local materials at increased temperature and which can be obtained by induced mutagenesis and hybridization (Alikhanian, 1962, 1970; Calam, 1969, 1970; Bradley, 1966; Hopwood, 1970); and (2) the knowledge and control of regulatory mechanisms of biosynthetic pathways of the products to allow overproduction and excretion (Vanek and Malek, 1964; Demain, 1966; Vezina, 1969). With the use of improved strains and the adaptation to modern fermentation technology, the yield could be greatly increased thereby making the commercial production of these products technically and economically feasible. More food for the future will be the guideline of the BRC research activities: directly by propagation of microbial cells in agricultural and industrial wastes and by-products; indirectly by production of microbial feeds and fertilizers, vitamins and growth factors to stimulate and control plant and animal growth, and antibiotics to combat plant and animal diseases. In addition to the latter, the nitrogen-fixing ability of Rhizobium as an inoculant to legumes will be made use of as well as the production of microbial insecticides such as Bacillus thuringiensis. Both of these microorganisms are extensively employed by farmers in the advanced countries. Since plant and microbial proteins are deficient in some essential amino acids, the production of these essential amino acids will be necessary for supplementing the deficiency in the diet and also for their therapeutic uses. Another application of recent research in microbiology is in the leaching of minerals and the production of sulphur from gypsum. Beck (1969) reported an increase in the efficiency of leaching waste-copper ore when the conditions were adjusted to favour the growth of Thiobacillus ferrooxidans. He also studied the role of the sulphate-reducing bacteria, Desulfovibrio and Desulfotomaculum, in sulphur production. Both gypsum and copper ores are abundant in the Philippines.

BIBLIOGRAPHY

ALIKHANIAN, S. I. (1962). Induced mutagenesis in the selection of microorganisms, Adv. appl. M icrobiol. 4 p. 1.

ALIKHANIAN, S. I. (1970). Applied advances in genetics of microorganisms, Genetika 6 (4) p. 96. Through Microbiology Abstracts 6 ( 1), 1970. IAEA-SM-134/29 313

ANGELES et al. (1970). Paper presented at the 10th Int. Cancer Congress, Houston, Texas, May 22-29,

ARIMA, К (1964). "Microbial enzyme production", Globial Impacts of Applied Microbiology (STARR, M. P., Ed.), John Wiley and Sons, New York p. 277.

ATACADOR-RAMOS, M ., PALO, M .A., VIlLADOLID, D.V., CRUZ, D.S. (1967). A study on submerged culture production of banana mushroom (Volvariella volvacea (Bull, ex Fr. ) Sing.) mycelium as a source of protein, B-vitamins, and food flavour, Phil. J. Sei. 96 (3).

BAENS-ARCEGA, L., ARGUELLES, A. LI. (1960). The dextran-forming Leuconostoc from Philippine cabbage: Some factors influencing dextran formation and specific identity of the organism, Phil, J. Sei. 89 p. 23.

BAENS-ARCEGA, L. (1969). Philippine contribution to the utilization of microorganisms for the production of foods, Biotechn. and Bioengng. Symposium, No. 1 p. 53.

BAENS-ARCEGA, L. (1970). Sauce manufacture, Process Biochem. 5 (10) p. 50.

BECK, J. V. (1Ô69). Metal recovery from low-grade ores, sulphur recovery from gypsum, Paper presented at the Expert Working Group Meeting on the Manufacture of Chemicals by Fermentation, UNIDO, Vienna, Austria, December 1-5.

BRADLEY, S.G. (1966). Genetics in applied microbiology, Adv. appl. Microbiol. 9 p. 29.

CALAM, C.T. (1969). Problems of culture improvement in industrial microbiology. Methods of industrial strain improvement, Paper presented at the Expert Working Group Meeting on the Manufacture of Chemicals by Fermentation, UNIDO, Vienna, Austria, December 3-5.

CALAM, C.T. (1970). "Improvement of microorganisms by mutation, hybridization and selection", Methods in Microbiology (NORRIS, J.R ., RIBBONS, D.W ., Eds), Academic Press Inc., London p .435.

CASIDA, Jr., L.E. (1968), Industrial Microbiology, John Wiley and Sons, Inc., New York p. 401.

DEMAIN, A. L. (1966). Industrial fermentations and their relation to regulatory mechanisms, Adv. appl. M icro b io l. 8 p. 1.

DIOKNO-PALO, N., VILELA, L. C. (1968). Possibilities of manufacturing high-grade vinegar from sweetened coconut water, Science Bulletin.

HIPOLITO, D. G ., DOMINGO, J.G ., SARTA, N.M. (1965). High-protein yeast, Rhodotorula pilim ana& t Hedrick et Burke, from Philippine strawberry fruit, Phil. J. Sei. 94 p. 179. Patent No. 1895, April 1, 1965 (Invention).

HOPWOOD, D. A. (1970). "The isolation of mutants”, Methods in Microbiology (NORRIS, J.R., RIBBONS, D.W ., Eds), Academic Press Inc., London p. 363.

International Trade Center. (1969). The Philippines as a market for selected manufactured products from developing countries, UNCTAD/GATT, Geneva.

JOSON, L. M ., SEVILLA-SANTOS, P ., LIBREA, J . f BERNARDO, Z . (1968). A ntibiotics I. Isolation and identification of an actinomycin complex produced by local streptomyces, Proc. First Post-War Philippine Science Convention, Manila, July 15-21.

LAPUZ, M .M ., GALLARDO, E., PALO, M. A. (1967). The nata organism: cultural requirements, characteristics and identity, Phil. J. Sei. £6 p. 91.

MACEDA, L. M ., PALO, M.A. (1967)* A study on an acetic-acid^forming bacterial isolate and factors influenceing its growth and production of acetic acid or vinegar from alcoholic medium, Phil. J. Sei. 96 p. 111.

National Economic Council (1970). Four-year development plan FY 1971-74.

ORILLO, C .A ., PEDERSON, C .S, (1968). Lactic acid bacterial fermentation of burong dalag, Appl. Microbiol. 16 p. 1669.

PALO, M .A,, RODULFO, B. R., BALITA, C.B. (1965). A study on the propagation and growth of local high-protein strain of Chlorella, Phil. J. Sei. 94 p. 189.

PALO, M.A. VIDAL-ADEVA, L., MACEDA, L.M. (1960). A study on Angkak (Chinese red rice) and its production, Phil. J. Sei. 89 p. 1. 314 JOSON

SEVILLA-SANTOS, P., BERNARDO, Z. Further screening of Philippine soils for antitubercular and antitumor producing streptomyces, Phil. J. Sei. (in press).

SEVILLA-SANTOS, P., ENCINAS, C .J., LEUS-PALO, S. (1964), The antibacterial activities of aqueous extracts from Philippine basidiomycetes, Phil. J. Sei. 9 p. 479.

SEVILLA-SANTOS, P., de LEON, W. (1962). A survey of Philippine soils for antibiotic producing streptomyces, Phil. J. Sei. 91 p. 241.

VANEK, Z ., MALEK, 1.(1964). "The pathway of biogenesis of natural products from acetic acid", Global Impacts of Applied Microbiology (STARR, M. P ., Ed.), John Wiley and Sons, New York p. 382.

VEZINA, C. (1969), Microbial production of therapeutic agents. Paper presented at the Expert Working Group Meeting on the Manufacture of Chemicals by Fermentation, UNIDO, Vienna, Austria, December 1-5. IAEA-SM-134/19

ACTIVITIES OF UNIDO IN THE FERMENTATION INDUSTRIES

C.S. CHIANG UNIDO, Vienna, Austria

Abstract

ACTIVITIES OF UNIDO IN THE FERMENTATION INDUSTRIES. The activities of UNIDO in the fermentation industries are described in this statement. As operational activities, i.e . technical assistance in the field, three projects were completed in the past few years and five current projects are under implementation. These activities covered technical assistance in the pro­ duction of butanol-acetone and yeast, and in research and development of the fermentation industry. As supporting activities, an expert working-group meeting on the manufacture of chemicals by fermentation was held in Vienna on 1 - 5 December 1969. Eleven papers were presented in the meeting and conclusions and recommendations were drawn up by the working group.

UNIDO's activities in the fermentation industries, as in other industries, may be divided into two broad classes: technical assistance in the field, and supporting activities. The technical assistance activities may take a number of forms such as feasibility studies, selection of processes and machinery, in­ vitation for and evaluation of tenders, supervision of construction and commissioning of industrial plants, etc. Where it seems desirable, assistance is also given in establishing experimental plants which will either demonstrate the viability of a modern manufacturing industry in a developing country or will help to develop new uses for local raw materials. Help is also given in improving existing plants which are not functioning properly, because of poor design or inadequate supervision or maintenance. Supporting activities include the organization of expert working groups, workshops and seminars. The purpose of expert working groups is to lead to the formulation of important findings which will be of value to the de­ veloping countries. The workshops and seminars are organized in a dif­ ferent way as they contain an element of transfer of knowledge to partici­ pants from developing countries and exchange of information among the different countries. The field projects which UNIDO has undertaken in industrial fermenta­ tion in recent years include the following:

Completed projects:

Project number and Description of project R em arks title

UAR-082-A (SIS) An expert was assigned to work Completed in for two months in a plant January 1969, Organic chemical specializing in the production of follow-up action in dustries organic chemicals (butanol- recommended. acetone) by the fermentation of

315 316 CHIANG

sugar-cane molasses. This plant was encountering dif­ ficulties in obtaining a satisfactory yield of the end-products.

PHILIPPINES 79-57 Assistance to the Biological Completed in Research Centre, National July 1967, Industrial Institute of Science and follow-up action fermentation Technology. recommended.

PHI-043-B (SIS) To assist the Biological Completed in Research Centre, National September 1970, Industrial Institute of Science and follow-up action fermentation Technology in the develop­ recommended. ment of industrial fermenta­ tion research. (Duration: 3 months.)

Current projects:

Project number and Description of project R em arks title

UAR-041-A (SIS) Two experts to assist in the To be implemented UAR-041-B (SIS) production of butanol-acetone in 1971. by fermentation including the following: Butanol-acetone by (1) Fermentation Technologist fermentation or Applied Technologist: to demonstrate to the plant staff microbiological and chemical techniques for process control of the butanol-acetone fermentation. (Duration: 3 months.) (2) Microbiologist or Fermentation Technologist: to demonstrate to the plant staff techniques for isola­ tion of suitable bacterial strains, and to advise on development work to improve butanol-acetone fermentation. (Duration: 3 months.)

CUB-043-B/ID An expert to assist the Expert already Research Department of the in field. Industrial Cuban Institute of Research fermentation expert on Sugarcane Derivatives (basic research) (ICIDCA) in a basic research of industrial fermentation. (Duration: 6 months.) IAEA-SM-134/19 317

CUB-043-C (RP/ID) An expert to assist the Research To be implemented Department of the Cuban in 1971. Industrial Institute of Research on fermentation expert Sugarcane Derivatives (ICIDCA) (general in general development of the development) fermentation industry. (Duration: 3 months.)

Projects under consideration:

Project number and Description of project R em arks title

G uatem ala Assistance to the Industrial Fermentation Laboratory of ÏCAITI (Central American Research Institute for Industry.)

Philippines Assistance to the yeast factories.

Within the supporting activities of UNIDO, an expert working group meeting on the manufacture of chemicals by fermentation was held in Vienna on 1 - 5 December 1969. Eleven papers were presented in the meeting as listed below:

1. "Metal Recovery from Low Grade Ores, Sulphur Recovery from Gypsum" - Jay V. Beck, USA. 2. "Problems of Culture Improvement in Industrial Microbiology" - C.T. C alam , UK. 3. "F erm en tatio n P lants and Equipm ent" - E lm e r L. Gaden, J r ., USA. 4. "Use of Water-insoluble Enzyme'Derivatives in Synthesis and Separation" - Leon Goldstein, Israel. 5. "Microorganisms and their Role in Fermentation" - C.W. Hesseltine and W.C. H aynes, USA. 6 . "Fermentation Processes employed in the Pharmaceutical Industries and their Economic Aspects" - István Horvath, Hungary. 7. "Nutritional Supplements, Vitamins, Amino Acids, and Flavouring A gents" - H .T. Huang, USA. 8 . "Energetic and Kinetic Aspects of Industrial Fermentation" - J. Meyrath, Austria. 9. "Microbial Production of Therapeutic Agents" - Claude Vézina, Canada. 10. "Industrial Chemicals: Organic Solvents, Organic Acids, Miscel­ laneous Products, Microbial Insecticides" - F. Wagner, Germany. 11. "Fermentation and Wastes Disposal" - P.A. Stevens, WHO.

Some general principles (recommendations) on the potential role of fermentation technology in developing countries were drawn up by the group in the meeting. It is hoped that UNIDO will expand its activities in this field to a greater extent in the coming years.

IAEA-SM-134/17

MICROBIOLOGICAL PROGRAM ACTIVITIES OF UNESCO

A .C .J . BURGERS Department of Science Policy and Promotion of the Basic Sciences, UNESCO, France

Abstract

MICROBIOLOGICAL PROGRAM ACTIVITIES OF UNESCO. Unesco's program in microbiology comprises the following activities: the organization of training courses and conferences, a fellowship program and the co-ordination of international activities in micro­ biology with other U.N. agencies and non-governmental organizations. The program is planned in close co-operation with the panel on microbiology of the International Cell Research Organization (ICRO).

Unesco's program, Research in Microbiology, is a part of the Life Science's program which consists of the following three activities: Interdisciplinary Brain Research, Cell and Molecular Biology and Research in Microbiology. In comparison with the programs in microbiology of other U.N. agencies, like WHO, FAO and IAEA, the program of Unesco is rather modest and concentrates mainly on the promotion of fundamental research in various fields of non-medical microbiology especially in developing countries. Unesco's program activities are:

a. The organization of training courses for postgraduate students, b. A fellowship program, c. The organization of conferences, and d. The co-ordination for international microbiological activities with other U.N. agencies and non-governmental organizations (NGOs).

Most of the activities are carried out in close co-operation with the International Cell Research Organization (ICRO), which is composed of about 200 scientists selected on the basis of their scientific competence and has its offices in Unesco. ICRO was founded with the aid of Unesco in 1962. ICRO is composed of eight different panels: 1. Control Processes and Molecular Biology, 2. Organized Metabolism, 3. Morphology and Mechanics of the Cell, 4. Interaction between Cell and Non-Cellular Environment, 5. Genetic Change and Virus-Cell Interaction, 6. Micro­ biology, 7. Fundamental Biology of Reproduction and Fertility, 8. Cell Differentiation. It is in consultation with Panel 6 that Unesco's program

319 320 BURGERS in microbiology is planned and executed. For the policy making and guidance of this program by Unesco, three levels of information and contacts exist:

1. U.N. level

Protein Advisory Group (PAG)

This competent and technical body studies various ways to increase the production of protein in the world (proteins from conventional and unconventional sources, economic aspects, marketing) and gives advice to the U.N. agencies and the Economic and Social Council of the U.N. The sponsoring agencies of the group are FAO, WHO and Unicef. Unesco is a member with full participating rights. One of the ad hoc Working Groups of the PAG deals with the production of Single Cell Proteins (SCP). Several members of the Unesco/ICRO panel on micro­ biology are members of this group. These contacts enable us to bring our program directly in line with the recommendations made by this group.

United Nations Development Project (UNDP)

In the framework of the Technical Assistance and Special Fund pro­ jects of UNDP, Unesco selects and recruits microbiologists to develop teaching and research in microbiology (TA), to assist in the establish­ ment of microbiological research laboratories (as part of the Science Faculties) and to strengthen the microbiology in existing institutes (SF) in developing countries.

2. U.N. agencies level

Microbiology, with its branches in medicine, agriculture, industry and cellular-molecular biological research including the applications of radiations, is a discipline par excellence for Inter-Agency Co-operation. To avoid duplication and the waste of funds, a close contact between the agencies is necessary. Therefore, meetings of representatives of agencies together with those of the appropriate NGOs, in which plans and projects are tabled, discussed and co-ordinated, are of great importance. The first of these meetings took place in 1970 at WHO in Geneva, in which IAEA, WHO, Unesco and representatives of IAMS, IOBB, and ICRO participated (see item 3 for explanation of the abbreviations). Good examples of inter-agency co-operation are the International Conferences on the Global Impacts of Applied Microbiology (GIAMs), which are organized by the Unesco/ICRO panel in Microbiology and sponsored by WHO and Unesco in consultation with FAO, IAMS and IBP. Training courses are another type of activity through which joint actions are being realized. Several Unesco/WHO training courses in the fields of Cell and Molecular Biology and in Brain Research exist already. It is hoped that this program can be extended and joint training courses in microbiology can be added (for example with IAEA) in the future. IAEA-SM-134/17 321

3. N G O le v e l

Unesco, through its ICRO/Unesco panel in microbiology, stays in close contact with the International Association of Microbiological Societies (IAMS), its three sections, and with the International Organiza­ tion of Biotechnology and Bioengineering (IOBB). Moreover, Unesco assisted in the creation of the World Federation for Culture Collections (WFCC) in Mexico City, 1970. In co-operation with the International Biological Programme (IBP), joint IBP/Unesco symposia have been organized at the occasion of the GIAM conferences. Several other microbiological activities organized by IBP are jointly sponsored by Unesco and IBP.

Program details

Unesco/ICRO training courses

These courses are international or regional with a duration of 2-3 weeks and the number of participants between 18 and 30. They are organized mainly in developing countries. A considerable part of the time (more than 50%) of the course is devoted to practical work at the laboratory bench.

Topics of such training courses:

Continuous Cultivation of Microorganisms (Prague, 1965), Various Aspects of Research in Microbiology (Addis Ababa, 1967), Genetics and Physiology of Bacterial Viruses (Bangalore, 1968), Biotechnology (fermentation) (Sao Paulo, 1969), Microbial Physiology and Genetics of Molecular Biology (Bombay, 1969), Bacterial Membrane and Cell Wall (Paris, 1970), Fundamental Research in Soil Microbiology (Mexico City, 1971).

Conferences a. International Conferences on the Global Impacts of Applied Micro­ biology (GIAM) These conferences are aimed at confronting high governmental officials, administrators, research workers and students with the progress and latest discoveries made in the various fields of micro­ biology and emphasize the impact which microbiology can have on the economic development of a country. (Sweden, 1963; Ethiopia, 1967; India, 1969.) It is hoped that IAEA will join us in the organization of the 4th GIAM planned for 1973 in Latin America. b. International Conference on Culture Collections Convinced that one of the pillars for progress in microbiology is taxonomy and the study of the physiological properties of microorganisms, Unesco sponsored and assisted in the organization of the International Conference on Culture Collections (Tokyo, 1968) which became the spring­ board for the creation of the World Federation for Culture Collections (WFCC) mentioned above. 322 BURGERS

Fellowships

Each biennium Unesco provides six full fellowships (9-12 months) to students from developing countries who wish to study in a specialized field of microbiology which is of direct interest to the development of their home country.

Surveys

A few years ago Unseco lauched a world survey on microbiological activities. The results of this survey are being analysed by IAMS. The first results (Africa) have been published. When the need was felt for a survey on microorganisms which are frequently used in teaching and research, the Unesco/ICRO panel on "Genetic change and virus cell interaction", compiled an international registry of microbial genetic stock strains (bacteria, bacteriophages and fungi), indicating the laboratories where they are kept and from which they can be obtained. Unesco promoted and sponsored also a world survey on bacterial culture collections; this survey is in a computerized form and has entered the second phase which is the listing of all the strains kept in collections that were assembled.

Budget

The budget of Unesco's program Research in Microbiology is in the order of $45 000 per year.

DISCUSSION

S. I. ALIKHANIAN: It is necessary, in my opinion, to establish a body to co-ordinate activities on preparing programs of assistance to developing countries in organizing microbiological industries. This body should concentrate on the genetics and selection of microorganisms. The existing UNESCO bodies concerned with microbiology deal with many problems associated with the use of microbiological synthesis products but not with the organization of such industries. In the application of microbiological synthesis the decisive role is played by productive strains. Only with success in producing improved strains can a micro­ biological industry be established. For this reason, I think that an international body concerned with the genetics and selection of m icro­ organisms would make an important contribution to success in this field, especially as regards training personnel in the developing countries. The genetics of microorganisms is the basis on which molecular biology would develop. A.C.J. BURGERS: The United Nations agencies cannot impose a policy or action on any government. The geneticists should unite and form the appropriate organization which, through its contacts with the scientific unions under the International Council of Scientific Unions, could IAEA-SM-134/17 323 approach scientific councils and governments and in this way make their voice heard at the general conferences and assemblies of the United Nations agencies. In the meantime, the agencies will certainly consider any activity which lies within their competence and within the framework of their program. N. K. NOTANI: Are there any major areas of overlap in the biological objectives of support by UNESCO, ICRO and IAEA? A. C. J. BURGERS: Yes, there are several areas where the scientific programs of the United Nations agencies overlap. However, this gives us an opportunity to organize joint interagency activities, which bring together scientists of various disciplines and have a strong and beneficial impact on the exchange of information. We are planning to convene a meeting at the working level between the representatives of the various United Nations agencies and of several non-governmental organizations to co-ordinate selected international program activities in microbiology and to adopt a program in this field for the future. J. MEYRATH: Does the UNESCO Panel on Microbiology deal with industrial microbiology, since IOBB is concerned mainly with the bio­ engineering aspects? A.. C. J. BURGERS: The activities of the Panel depend mainly on the interest of the members composing it. Initially, the microbiological activities were oriented to pure research. More recently the interest has shifted more to research in applied microbiology, which is reflected in the scientific specialization of the members. J. MEYRATH: Are the training courses organized through UNESCO and ICRO held on a regular basis or only when the need arises? A. C. J. BURGERS: In addition to short-term training courses on "hot" topics, UNESCO also sponsors regular training courses in micro­ biology lasting between nine months and one year. The short-term courses are organized in close co-operation with ICRO. Applications, accompanied by a detailed description, are submitted by scientists to ICRO for consideration. UNESCO pays only the travel costs of teachers and students. Board and lodging are normally provided by the host country. The topics to be covered in such courses are selected by UNESCO/ICRO on the basis of the needs of developing countries and the scientific level of the teaching staff and students of post-graduate level. J. MEYRATH: Has the PAG recently revised the production costs of single-cell proteins? If so, what is their present estimate of the production cost per pound of protein? А.. С. J. BURGERS: In view of the large investments made by petroleum industries for the production of single-cell proteins, it can be expected that the price of this product will be competitive with that of soya meal in the future.

IAEA "SM “134/18

MICROBIOLOGICAL PROGRAM ACTIVITIES OF THE IAEA

R. MUKHERJEE IAEA, Vienna, Austria

Abstract

MICROBIOLOGICAL PROGRAM ACTIVITIES OF THE IAEA. The paper presents a brief discussion on the rationale behind the Agency's program on the applications of nuclear techniques and technology in the microbiological disciplines providing for human health and welfare. Major emphasis is given to the development of microorganisms that are useful for the fermentation industries to meet the economic and welfare needs of the developing countries.

The experts from a number of the Member States of the Agency have discussed in an excellent way their research findings on the genetics and biochemical pathways in microorganisms. Of particular interest has been the emphasis on the metabolic processes of microorganisms which have either an established or a potential relevance to fermentation practices. We have also heard of the interest and program activities in this field of several Member States and of other international agencies. The International Atomic Energy Agency has recognized the need for more emphasis on research leading to the use and control of micro­ organisms for the health and welfare of mankind and thus, in its program, one can find several efforts directed to the study of radiation effects on microorganisms. The rationale for this increasing interest by the Agency in radiation microbiology is based on the fact that microorganisms in the fermentation industry give promise of being an economic source for food, nutritional supplements, pharmaceutical products and organic acids that are greatly needed but are lacking in adequate amounts for a large sector of the world's population. Great strides have been made in conventional agricultural practices that have led to increased productivity of crops, livestock, and the control of pests as well as in the storage and preservation of foods particularly in the temperate climates. However, the.cost of transporting these products is high even if there were an adequate amount to supply the increasing world needs. The evidence is now clear, and has been reinforced through this symposium, that strains of microorganisms, when properly fitted into the technology of the fermentation industries, can make a very significant addition to the food, medical, and industrial needs of the world. Further advantages of the use of microorganisms are that they can be transported at a very low cost and that they increase at a phenomenal rate when grown on proper media. Also of great significance is the fact that they can utilize waste products while producing substances important to the health and welfare of a country. Probably of even greater interest is that it is in the tropical and subtropical regions of the world where food supplements are in greatest need that there exist excess waste products

325 326 MUKHERJEE such as molasses, cornsteep liquors, and carbohydrate-rich root crops that could be converted to the needs of these countries through the use of m icro o rg a n ism s. A striking fact is that one worker in a yeast-producing plant can, on the average, produce about 60 tons of dry protein per year, whereas his counterpart in conventional agriculture can produce only 6 tons of dry protein per year. It would be expected that this difference in productivity may be even increased further as the fermentation technology is improved and new strains of microorganisms are developed. It is recognized that problems do exist with respect to microbial productivity of many sub­ stances particularly with respect to quality; however, these problems can be solved through research and the rate of solution will directly depend on the amount of research effort expended and the rate at which new information in this field becomes available to scientists throughout the w orld. The Agency's radiation microbiology effort is directed to encouraging the use of ionizing radiation for the improvement of microorganisms that produce beneficial products. Its interest, however, does not end there. It is concerned with the total effort leading to the economic production of antibiotics, amino acids, organic acids, vitamins, hormones, etc. and in attempting to increase the rate of information exchange in this field of microbiology. To achieve this goal the Agency's radiation biology section has started a coordinated research program on radiation microbiology. This co-ordination of research is designed to impart much needed information, guidance and encouragement for the participants from institutes of the developing countries. Opportunities for such co-ordination through the medium of information exchange is provided by research co-ordination meetings with the participants in the program as the needs are recognized and as meetings are approved by the Agency. Within our co-ordinated research program nine scientists from eight countries (including five developing and two developed countries) are, at present, participating. Their research activities include studies on mutation induction and selection of improved strains, development of genetic systems by various recombination processes, biochemical and genetic analysis of the regulatory processes for biosynthesis, and the development of suitable strains for the utilization of locally available carbon-rich substrates. The Agency has. recently awarded five fellowships to scientists for training in applied microbiology. The Agency also held a training’course on radiation microbiology in Bombay, India in 1969. This course provided for the training of 20 participants from 11 developing countries in the areas of applied microbiology. Another such training course is to be proposed for 1972. I would like to conclude by pointing out that as the development of the Agency's program in radiation microbiology was underway we found that it was extremely difficult to keep the scope of the program within narrow limits. We found that studies on the effects of irradiation are important to a number of programs within the Agency. For example: radiation steri­ lization of biomedical products (an ongoing research program) depends heavily on knowledge of the killing and mutagenic effects of radiation on microorganisms. This is also true of the Agency's FAO/IAEA program in food preservation. IAEA-SM-134/18 327

In a program that is being expanded on radiation attenuation of toxins and parasites for the production of vaccines, the effects of ionizing radiation on viruses, bacteria and protozoa can result in inhibition of the virulence of pathogenic organisms while having only a small effect on their antigenic properties. We feel confident that an increased effort in radiation microbiology will not only help in solving the nutritional, pharmaceutical and organic- acid needs of man but will also lead to information necessary to the solu­ tion of radiation preservation of food and biomedical products and the production of new vaccines. I would finally like to say that the Agency will continue to be sensitive to the needs of the developing countries in promoting radiation microbiology in their countries and that the scientists in those countries can be of particular help to the Agency's task by providing quantitative information on their pharmaceutical, organic acid, and nutrient supplement needs that can be solved through microbiological research.

CHAIRMEN OF SESSIONS

Session 1 J.A. ROPER United Kingdom Session 2 H.I. ADLER United States of America Session 3 S.I. ALIKHANIAN Union of Soviet Socialist Republics

Ses SI'on 4 H. HESLOT F ran ce

Ses Slion 5 S.G. GEORGOPOULOS G reece Sess ion 6 K. ESSER Federal Republic of G erm any Session 7 J. MEYRATH A ustria

SECRETARIAT

Scientific Secretary R. MUKHERJEE Division of Life Sciences, IAEA Administrative S ecretary Caroline DE MOL Division of Scientific and VAN OTTERLOO Technical Information, IAEA E d ito r Anne ERICSON Division of Publications, IAEA Records Officer S.K. DATTA Division of Languages, IAEA

329 LIST OF PARTICIPANTS

AUSTRIA

A ltm an n , H. österreichische Studienges. für Atomenergie mbH, Lenaugasse 10, 1080 Vienna

Fonatsch, Christa Institut fur medizinische Biologie, Hans-Sachsgasse 3, 8010 Graz

Gonzalez, Alicia Biochemie Kundl GmbH, 6250 Kundl

Heinzei, Brigitte Institut füi medizinische Biologie, Hans-Sachsgasse 3, 8010 Graz

Hitschmann, Agnes Institut für Angewandte Microbiologie, Michaelerstrasse 25, 1180 Vienna

H ofer, H. Institut für Biologie, Reaktorzentrum Seibersdorf, 2444 Seibersdorf

Katinger, H. Institute of Microbiology, University of Agriculture, Gregor Mendelstrasse 33, 1180 Vienna

Matsché, N. Department for Microbiology, Institute of Technology, Getreidemarkt 9, 1060 Vienna

Meyrath, J. Institut für angewandte Mikrobiologie, Hochschule für Bodenkultur, Michaelerstrasse 25, 1180 Vienna

Oberzill, W. Institut für Biochemische Technologie und Mikrobiologie, Technische Hochschule, Getreidemarkt 9, 1060 Vienna

Oswald, J.-D. Biochemie Kundl GmbH, 6250 Kundl

Partsch, G. Institut für Biologie, Reaktorzentrum Seibersdorf, 2444 Seibersdorf

Wutzel, H.J.G. Vereinigte Hefefabriken Mautner Markhof und Wolfrum, Mautner Markhofgasse 40, 1110 Vienna

BULGARIA

Kibarska, Totka Institute of Technical Microbiology, Geo Milev 3, Sofia

330 LIST OF PARTICIPANTS 331

CANADA

Weijer, J. Department of Genetics, University of Alberta, Edmonton, Alberta

CHINA

Li, Chuan-Hsian Chai Yee Solvent Works, Chinese Petroleum Corp., 239 Min Sheng South Road, Chia-Yi, Taiwan

Ting, Shu-Hsun Chai Yee Solvent Works, Chinese Petroleum Corp., 239 Min Sheng South Road, Chia-Yi, Taiwan

CZECHOSLOVAK SOCIALIST REPUBLIC

Н оЗШ ек, Z . Institute of Microbiology, Czechoslovak Academy of Sciences, Budejovická 1083, Prague 4 - KRC

Mikulik, K. Institute of Microbiology, Czechoslovak Academy of Sciences, Budejovická 1083, Prague 4 - KRC

DENMARK

C hristensen, E. A. Control Department, Statens Seruminstitut and Danish Atomic Energy Commission, Research Establishment. RisÖ, 4000 Roskilde

Em borg, C. Bacteriological Laboratory, Accelerator Department, Danish Atomic Energy Commission, Research Establishment, Riso, 4000 Roskilde

Johansson, A. A/S NUNC, 8 Algade, 4000 Roskilde

Laursen, P.K. A/S Grindstedvaerket, Edwin Rahrsvej 38, 8220 Brabrand

Vestberg, Karin A/S Grindstedvaerket, Edwin Rahrsvej 38, 8220 Brabrand ■

Villadsen, K .J.S. A/S Grindstedvaerket, Edwin Rahrsvej 38, 8220 Brabrand

FINLAND

Markkanen, P. Biotechnical Laboratory, The State Institute for Technical Research, Nuijatie 4 В 15, Vapaala 332 LIST OF PARTIQPANTS

FRANCE

Accolas, J. -P. Station recherches laitières, Institut national de la recherche agronomique, 78 Jouy-en-Josas

Biisson, R. Société Rhône Poulenc, 9 Quai Jules Guesde, 94 Vitry

D upuy, P. Station de technologie de produits végétaux, Institut national de la recherche agronomique, 7 rue Sully, 21 Dijon

Gaillardin, C. Service de génétique, Institut national agronomique, 16 rue Claude Bernard, Paris 5e

G alz y , P. Institut national de la recherche agronomique, (149 rue de Grenelle, Paris) Centre de recherche agronomique du Midi, 34 Montpellier

G odet, P. Société Melle-Bezons, 47 rue de Villiers, 92 Neuilly

H eslot, H. Service de génétique, Institut national agronomique, 16 rue Claude Bernard, Paris 5e

Kalabokias, G. Société Rapidase, 15 rue des Comtesses, 59 Seclin

M aldonado, P. F. Institut national agronomique, 16 rue Claude Bernard, 75 Paris 5e

Pasero, J, Société internationale de recherche BP, Rue des 4 Filles, 28 Epernon

Pelluet, J.G, ORSAN, 13 rue de Calais, Paris 9e

Peyre, Marcelle Centre de recherches, Roussel-Uclaf, 111 Route de Noisy, Romainville 93

Robichon-Szulmajster, Huguette de Laboratoire d* enzymologie, Centre national de la recherche scientifique, 91 Gif-sur-Yvette

GERMANY, FEDERAL REPUBLIC OF

Esser, K. Lehrstuhl fur Allgemeine Botanik, Ruhr-Universität Bochum, Postfach 2148, 4630 Bochum-Querenburg

H oc, S. Selecta Verlag, Karlstrasse 29, 8033 München-Planegg LIST OF PARTICIPANTS 333

M racek , M . Farbwerke Hoechst AG, 6230 F rankfurt/M . 80

O eding, V. Institut für Mikrobiologie der Gesellschaft fur Strahlen- und Umweltforschung, Grisebachstrasse 8, 3400 Gottingen

Sahm , H. Gesellschaft für molekularbiologische Forschung, Mascheroder Weg 1, 3301 Stockheim

Sahm, Ursel Gesellschaft für molekularbiologische Forschung, Mascheroder Weg 1, 3301 Stockheim

Schlegel, H. G. Institut für Mikrobiologie der Gesellschaft für Strahlen- und Umweltforschung, Grisebachstrasse 8, 3400 Göttingen

S chuster, E. Rohm G m bH , Kirschenallee, 6100 Darmstadt

T a n , T .L . Gesellschaft für molekularbiologische Forschung, Mascheroder Weg 1, 3301 Stockheim

GHANA

C au rie, M . Food Research Institute, P. O. Box 20, A ccra

GREECE

G eorgopoulos, S. G N.R. C. "Democritos", Aghia Paraskevi Attikis, Athens

Vomvoyanni, Vassiliki N.R.C. "Democritos", Aghia Paraskevi Attikis, Athens

HUNGARY

Ig a li, S. FJC National Institute for Radiobiology and Radiohygiene, Penka K. u. 5, Budapest XXII

Udvardy, Eva N. G. Richter Chemical Works, Gyomrôi út 19-21, Budapest X

INDIA

Notani, N.K. Biology Division, Bhabha Atomic Research Centre, Trombay, Bombay 85 334 LIST OF PARTICIPANTS

ISRAEL

Juven, B. The Volcani Institute of Agricultural Research, P .O . Box 15, Rehovot

ITALY

Ami ci, Alba Maria Farmitalia, Istituto Rlcerche, Via dei Gracchi 35, Milano

Bianchi, A. Instituto Sper. Orticoltura, V.B. Croce 47, 63100 Ascoli Piceno

Magaudda, G. Laboratorio Applicazioni Agricoltura (CNEN), C.S.N. Casaccia, S. Maria di Galería, Rome

KOREA

H an, H .E . Crown Brewery Co. Ltd., Seoul Present address: Institute of Applied Microbiology, Michaelerstr. 25, 1180 Vienna , Austria

NIGERIA

Adebona, A.C. University of Ife, Ile-Ife

Balogh, Esther University of Ife, Ile-Ife

NORWAY

Thomassen, S. Apothekemes Laboratorium for Specialpraeparater, SkjSyen

PHILIPPINES

Joson, Lydia M. Biological Research Center, National Institute of Science and Technology, National Science Development Board, Herran, Manila

POLAND

Makowski, J. Institute of Hydraulic Research of the Polish Academy of Science, ul. Cystersov 11, Gdansk-Oliwa LIST OF PARTICIPANTS 335

PORTUGAL

Zagallo, A.S.R.C. Department of Microbiology, Estaçâo Agronómica Nacional, O eiras

SINGAPORE

N ga, B.H . Department of Botany, University of Singapore, Bukit Timah Road, Singapore 10

SWITZERLAND

H utter, R. Mikrobiologisches Institut, Eidgenössische Technische Hochschule, Universitâtsstr. 2, 8006 Zurich

K obel, H. Sandoz AG, 4002 Basel

Sanglier, J.-J. Sandoz AG, 4002 Basel

Shepherd, D. Nestlé*s Technical Assistance Company, Linor, 1350 Orbe

TURKEY

K eskin, S. Commission à 1*énergie atomique, Centre d*études de l’énergie nucléaire, Ankara

UNION OF SOVIET SOCIALIST REPUBLICS

Alikhanian, S.I. Institute of Genetics of Industrial Microorganisms, P.O. Box 379, Moscow D-182

Sukhodolets, V. V. Institute of Genetics of Industrial Microorganisms, P.O. Box 379, Moscow D-182

Z hdanov, V. G. Institute of Genetics of Industrial Microorganisms, P .O . Box 379, M oscow D -182 336 LIST OF PARTICIPANTS

UNITED KINGDOM

MacDonald, K. D. Microbiological Research Establishment, Porton Down, Salisbury, Wilts.

Roper, J.A. Department of Genetics, The University, Sheffield S10 2TN

UNITED STATES OF AMERICA

Adler, H.I. Biology Division, Oak Ridge National Laboratory, Oak Ridge, Tenn. 37830

C h am ey , W. Schering Corp., 1011 Morris A ve., Union, New Jersey 07083

Cornwall, R. US Army Material Command, Scientific and Technical Information Team, IG Hochhaus, Room 750, 6000 Frankfurt/M., Fed. Rep. of Germany

H o ttle , G. Office of Naval Research Branch, Keysign House, 429 Oxford Street, London W .l, United Kingdom

L ein, J. Panlabs Inc., P .O . Box 81, F a y e tte v ille , N. Y. 13066

YUGOSLAVIA

Alacevic, Marija Technological Faculty, Pierottieva 6, Zagreb

Busljeta, M. "Krka" Pharmaceutical and Chemical Products, Novo M esto

Johanides, Vera Technological Faculty, University of Zagreb, Pierottieva 6, Zagreb

Krajinïanic, Branka "Boris Kidric" Institute of Nuclear Sciences, P. O. Box 522, Vin£a

Odar, A. "Krka" Pharmaceutical and Chemical Products, Novo M esto

Petravic, J. "Krka” Pharmaceutical and Chemical Products, Novo M esto LIST OF PARTICIPANTS 337

ORGANIZATIONS

CCE van Hoeck, F. Commission des Communautés Européennes, Services de biologie, 200 rue de la Loi, Brussels, Belgium

FORAT OM

Alcmann, H. Österreichische Studienges. fur Atomenergie mbH, Lenaugasse 10, 1080 Vienna (see also under Austria)

UNESCO

Burgers, A .C.J. Department of Science Policy and Promotion of the Basic Sciences, Pl. de Fontenoy, Paris 7e, France

UNIDO

Chiang, C.S. Felderhaus, Room A -111, Rathausplatz 2, 1010 Vienna

WHO

Meilland, G. WHO Liaison Officer, IAEA, Kärntnerring 11, 1010 Vienna AUTHOR INDEX

(including participants in discussions)

Numerals underlined refer to the first page of a paper by the author concerned. Further numerals denote comments and questions in the discussions.

Adegbola, A. A. : 305 Joson, Lydia, M. : 307 Adler, H.I. : 51, 80, 241, 248, Kappas, A. : 233 249 K arnetová, J. : 201 Alikhanian, S. I. : ¡3, 9, 42, 91, K rajincanié, B ranka: 279 110, 120, 121, 179, 198, K rem en, A. : 201 251, 258, 322 L ee, Ching-Song: 259 Altmann, H.: 41, 51, 53, 137, Li, Chuan-H sian: 259 179 M ehta, R. D. : 63 Bahn, M. : 137 Meyrath, J. : 9, 41, 137, 230, Balogh, Esther: 188, 305 265, 285, 305, 323 B arbié, F. : 279 Mikulik, K. : 201, 222 B lum auerová, M. : 157, 189 M ishra, K. P . : 73 B u rg ers, A. C. J. : 319, 322, 323 M racek, M. : 166, 258 C aurie, M .: 120, 289 M ukherjee, R. : 325 Chiang, C. S. : 315 Nga, B. H. : 110, 123_ Cudlm, J. : 189 Notani, N. K. : 40, 43, 51, 80, Dupuy, P.: 109, 249, 267 90, 119, 323 E m borg, C .: 71, 249 Oeding, V. : 223 Esser, K.: 83_, 91, 249 P a rtsc h , G. : 53, 61, 62 Gaillardin, C.: 93, 109, 110, 111 Peyre, Marcelle: 41 Galzy, P . : 267 Robichon-Szulmajster, Huguette de: Georgopoulos, S. G. : 61, 129, 109, 120, 178, 181, 188, 188, 222, 233, 248 231, 249 Gopal-Ayengar, A. R. : 43, 73 R oper, J. A. : 41, 42, 71, 113, 120 Han, H. E. : 137 Schlegel, H. G. : 223, 230, 231 Heslot, H .: 9, 13, 40, 41, 42, 62, Singh, B. B. : 713 71, 93, 165, 166, 265 Srinivasan, V. T. : 73 H ofer, H. : 79 Sukhodolets, V. V. : 41, 61 HoStálek, Z.: 157, 166, 189, Tax, J. : 201 198, 199, 231 Ting, Shu-Hsun: 259, 265 Hutter, R.: 109, 169, 178, Vanëk, Z. : 157, 189, 201 179, 231 Vomvoyanni, Vassiliki: 178, 233 Ism ail, A. A. : 157 Weijer, J. : 63, 71, 91, 110, Johanides, Vera: 198 120, 249 Joshi, V. R. : 43 Z agallo, A. S. R. C .: 258

TRANSLITERATION INDEX

Алиханян, C.И. Alikhanian, S.I. . w *■ f ß Exclusive sales agents for IAEA publications, to whom all orders and ■ inquiries should be addressed, have been appointed in the following countries:

UNITED KINGDOM Her Majesty’s Stationery Office, P.O. Box 569, London S.E.l UNITED STATES OF AMERICA UNIPUB, Inc., P.O. Box 433, New York, N.Y. 10016

In the following countries IAEA publications may be purchased from the sales agents or booksellers listed or through your major local booksellers. Payment can be made in local currency or with UNESCO coupons.

ARGENTINA Comisión Nacional de Energía Atómica, Avenida del Libertador 8250, Buenos Aires AUSTRALIA Hunter Publications, 2 3 McKillop Street, Melbourne, C.l BELGIUM Office International de Librairie, 30, avenue Marnix, Brussels 5 CANADA Information Canada, Ottawa, Canada C.S.S.R. S.N.T.L., Spálená 51, Prague 1 Alfa, Publishers, Hurbanovo námestie 6, Bratislava FRANCE Office International de Documentation et Librairie, 48, rue Gay-Lussac, F-75 Paris 5e HUNGARY Kultura, Hungarian Trading Company for Books and Newspapers, P.O.Box 149, Budapest 62 INDIA Oxford Book and Stationery Comp., 17, Park Street, Calcutta 16 ISRAEL Heiliger and Co., 3, Nathan Strauss Str., Jerusalem ITALY Agenzia Editoriale Commissionaria, A.E.I.O.U., Via Meravigli 16, 1-20123 Milan JAPAN Maruzen Company, Ltd., P.O.Box 5050, 100-31 Tokyo International NETHERLANDS Martinus Nijhoff N.V., Lange Voorhout 9—11, P.O. Box 269, The Hague PAKISTAN Mirza Book Agency, 65, The Mall, P.O.Box 729, Lahore-3 POLAND Ars Polona, Céntrala Handlu Zagranicznego, Krakowskie Przedmiescie 7, Warsaw ROMANIA Cartimex, 3-5 13 Decembrie Street, P.O.Box 134-135, Bucarest SPAIN Librería Bosch, Ronda Universidad 11, Barcelona-7 SWEDEN C.E. Fritzes Kungl. Hovbokhandel, Fredsgatan 2, Stockholm 16 SWITZERLAND Librairie Payot, rue Grenus 6, CH-1211 Geneva 11 U.S.S.R. Mezhdunarodnaya Kniga, Smolenskaya-Sennaya 32-34, Moscow G-200 YUGOSLAVIA Jugoslovenska Knjiga, Terazije 27, Belgrade

Orders from countries where sales agents have not yet been appointed and requests for information should be addressed directly to: Publishing Section, J International Atomic Energy Agency, Kärntner Ring 11, P.O.Box 590, A-1011 Vienna, Austria

INTERNATIONAL ATOMIC ENERGY AGENCY VIENNA, 1971 PRICE: US $9.00 SUBJECT GROUP: I Austrian Schillings 224,- Life Sciences/ (£3.75; F.Fr. 49,70,- DM 31,90) Isotopes and Radiation in Biology