THE UI'IIVERS TTY OF MANITOBA

A STUDY OF THE ù1ETABOLISM OF A NOVEI, DINUCLEOSIÐE

POLYPHOSPHATE (HS3) FOUND IN ITAPIT4AL IAN CELLS:

POSS]BLE REGULATTON OF TTUC],EÏC ACTD BIOSYNTHESÏS

BY HS3 DURING STËP-DOWN GRO¡]TH CONDITIONS

BY

SI{EE HAN GOH

A THES]S

SUBI'IITTED TO THE FACULTY OF GR.ADUATE STUDIES

IN PART]AL FULLFTLIqENT OF THE REOUTREUENTS FOR

THE DEGREE OF DOCTOR OF PHTLOSOPHY

. DEPARTMENT OF I{ICROBÏOI,OGY

WTNNIPEG , 1UANTTOBA APRTL, 1979 A STUDY OF THE METABOLISI4 OF A NOVEL DINUCLIOSIDE POLYPHOSPHATË (HS3) FOUND IN MAMMALIAN CILLS : POSSIBLT REGULAT]ON OF NUCLTIC AC]D BIOSYNTHISIS

BY HS3 DURING STEP.DOI,,IN GROWTH CONDITIONS.

BY

SI.IEE HAN GOH

A dissertation slrbmitted to the Ftculty of Gradtrate Studies ol the University of Marlitoba in partial fulfillment of tlre requirerrrents ol the degree of

DOCTOR OI. PI{ILOSOPHY o! 1979

Pemrission has bcen granted to the LIBRARY OF TIIE UNIVER- SITY OF MANITOBA to lend or sell copies of thìs dissertätion, to the NATIONAL LIIIRARY OF CANADA to microfilni this dissertation and to lend or sell copies of the film' and UNIVERSITY MICROFILMS to publish an abstrâct ol'this dissertation'

'l'lìc author reserves otlrer publicatiorr rights, and neither the dissertation nor extensive extracts frorìl it rnay be printed or other- wise reproduced witlìout the autl'ìor's writtell pernrission To Siew See and my parents Acknowledgements

I would like to lhank Ðr. Herb B- LeJohn for his advice, interest and patience during the course of this research progranme. The friendship and. constructive criticisms of Linda, Renate, Dave, Glen, Rob and. other members of the Dept. of }licrobiology are gratefully acknowledged -

I would also wish to express my deepest gratitude to Dr. Jím Vtright for his generosity and inval"uable advice \¡rith man¡nalian t.issue culture teihniques and the gift of numerous cell lines used for this study.

The counsel of Drs. I. Suzuki and PeÈer Ï,oevren is much appreciated. ABS TRACT

l-v A novef dinucleoside polyphosphat.e (HS3) prevously found ín various fungi has now been detected in numerous cultured mammalian ce11 lines. Physical and chemj.caf analyses of HS3 (McNaughton et. aI' 1978) show that the molecule consists of a glutamyl,-ADP- sugar moiety which is covalently linked to a UDP-mannitol-- tetraphosphate . Physiological studi.es demonstrated that as long as de novo purine biosynthesis was bfocked (by nutrient deprivation, drugs or mutation) and no exogenous purines (anil some pyrimidines) \^rere supplied, I1S3 accumulation took place. For exampfe, the addition of de novo purine inhibitors (50 or biosynthesis Lf-azaserine ¡:g,/ml) MTx (f .Ju¡a) I resulted in HS3 synthesis as did the \,¡ith- drawal- of glutamine (2 mM). The addition of hypoxanthine, inosine, adenine, adenosine, guancsine, uridine and cytidine, but not thymidine, (al-l- at 0.3 mM) strongÌy suppressed HS3 accumulation in g lutamine- starved cHo cells. CHO purine salvage nutants (HGPRT-) failed to do likewise when supplied v/ith high concentrations of hypoxanthine, guanosine and inosine (0.3-2. 0 m.l'!) . A cHo mutant (GAT -auxotrophic for adenosine, thymidine and glycine) with blocked de novo purine biosynthesis, accumulated HS3 only when adenosine (0.1 mI'I) , but not glutamine, was withdrardn f r'om its grohrth medium. Thus HS3 accumulation may be related to a lack of either precursors for purine nucleotide biosynthesis or of purinê nucleotides themselves.

The rate of HS3 synthesis in CHO l,tT ceLls increased 5-6 fold sholtly after gluta¡nine withdrarval and hras maintained at that Level for at least 6 hr before declínÍng slowly back to the control- rate by 22 hr. The pool sizes of IIS3 increased and decreased in concert with rate changes. The replenishment of glutamine and

adenosine to starved CHO WT and CHO GAT- celfs respect- ive]y, resulted in an immediate depletion of the

accumulated IIS 3. An inverse relationship r^¡as observed betr\'een nucleic acid (DNA and RNA) synthesis and HS3 accumulation. When DNA and RNA synthesis decreased, HS3 synthesis increased and vice versa. No such correlation was apparent for protein synthesis. The inhibition of protein synthesis by either puromycin (100 ¡¡9,/n1) or cycloheximide (10 ¡rg,/m1) or that of Rl¡A synthesis by actinomycin D (1 ¡¡glml), stimulated HS3 depletion in glutanine- starved CHO M cel1s. Both fungal- and mammaLian HS3 v/ere found to be equally potent inhlbitors of in vitro and in vivo

v.l manì¡nalian RNA synthesis. AIso, studies by Lewis et. al. (1977) showed that HS3 strongly inhibited partially purified ribonucleotide reductase from CHO celLs. Thus the decrease in DNA synthesis during periods of rapid HS3 accumualtion may be the indirect result of a deficiency in deoxyribonucfeotides. The data suggest t,hat HS3 may be involved in the regulation of nucLeic acid biosynthesis in mammalian cells. TÃ3LE OF CONTENTS

PAGE Acknowledgements . . iii Abstract. v Tabre of contents. viii List of Tables' xvii List of Figures. xix Abbreviations. . . . . xxii rntroduction''''' ' xxvi

H i stori ca l- Tissue Culture. 2 Development of Tissue Culture Techniques. 2 Growth Requirements and Characteristics of cultured Marûnalian Cell-s. 5 Marnmalian Cells: RNA Biosynthesis and Amino Acid Withdrar¡a1. 9 Physiology, Biochemistry and Possible Roles of Polyphosphates and Unique Phosphory- lated Nucleosides in Procaryotic and Eucaryotic Cellutar Regulation. . 11

vr- 11 PAGE

Polyphosphates... 12 Adenosine 3' :5' Cyclic Monophosphate. . 16 Biosynthesis, Catabolism and possi- ble Regul-atory Ro1e(s).... l-6 Cyclic AMP in Bacteria. 18 Cycti-c AIUP in Èhe Fungi. 20 Cyclic AMP in Manmalian Ce11s in culture. 22 Guanosine 3':5' Cyclic Monophosphate.. 24 Bacterial MS (ppGpp and pppGpp) Nucleotides 26 The stringent Responbe . 26 fn Vivo Synthesis of ppGpp and Pppcpp. 27 In Vitro Synthesis of ppGpp and pPpcpp. 32 Effects of ppcpp and pppcpp on CeIl- ular Enzymes, Translation and

Transcripti.on In Vitro 34 Influence of Carbon and Nitrogen

Sources on ppcpp, pppcpp and rRNÄ

Metabolism. 37

HPN Compounds.... 39 Diguanosine Polyphosphates. .. 4l ix PAGE

Diadenosine Polyphosphate s . . . 42 Triadenosine Pentaphosphate. 43 Dinucl-eoside Polyphosphates (HS) ..... 44 Other Pol-yphosphorylated Nucleos j-des. 48 Mammal-ian Purine.and Pyrimidine l4etabolism. 49 Regulation of De Novo Purine Biosynth- etic Enzynes in J"lammalian cells and Tissues. 49 PRPP Synthetase and IntracelLufar PRPP Leve.ls 52

5 -P ho spho r i bo s y I - 1-Pyrophosphate Amidotrans fera se. 54

Regulation of A-tvlP and GI'IP Synthesis from IMP. 57 Induction, Repression and Derepression of De Novo Purine Biosynthetic Pathway Enzymes. 58 Regulatj-on of De Novo Pyrimidine Bio- synthesis in lrlammal-ian Cells and Tissues. 59 lluman Genetic Disease and Defective Purine and Pyrimidine Metabofism. -.. - 63 PAGE

cout . . 63

Lesch-Nyhan Syndrome. 64

Immunodeficiency Syndromes. . 65

Oroticaciduria and Xanthinuria.... 69 Platerials and Methods

Organisms. 72

Mainrnal ian Cel1 lines . 72

Fungal Cells. 72

Gror'Jth Media, Sera and Cul ture-rrare s. t5

Radioisotopes. . . ¡ 73

Miscellaneous Materials..... 73

MeÈhods. 75

crowth of Cell Lines. 75 -'n2a crowth, - Orthophosphate Labelling and Formic Acid Extraction of Cell-s for Analysis of HS PoLyphosphorylated NucLeotides During Various Growth

Conditions. 76 chromatography of "p-LabeÌled Formic

Acid Extract from Àtamrnalian ce]Ls... . '7 '7 DNA, RNÀ and Protein Analysis. 78

DNA PuÌse Labelling Experiments. . . 79

XI PAGE

Protein Pulse Labelling

Experiments 79 continuous RNA Labelling Experiments 80 Det.ermination ot -1lC-adenine rncor- poration into Nucleosides and Nucleotides 81 Determination of ATP Levels Using the Firefly Luciferin/luc ifera se lqethod.. 81 Isolation of HS3 From Ach]ya and

CHO I{T cells. 83 Isolation of DNA-Dependent RNA Pofymerases From CHo ÌlT Cells. 84 Cel1 Permeabi l-i zation and Assay

for RN.A synthesis. . 87 Assay of RNÀ synthesis By Permeabilized

CHo t^tT cells In the Presence of

Achlya HS3... BB

Re sul ts Hs3 Synthesis by cHo wT cells. 90 Effect of Varying Concentrations of L-Glutamine on HS3 and GTP Synthesis by cHo v,lT cetts. 99 xii PÀGE

Effect of L-clutanine or !-fsol-eucine Deprivation on HS3 MetaboLism

by CHO WT Cel1s. tO2 Effect of L-Glutarnine, Adencsine and

Thymidine Deprivation on HS3

Metabol"ism by CHO GAT- Cells, IO2 Effect of L-Glutamine Withdrawal and the

Replenishment with purines on HS3

Metabolism by CHO lfT and CHO purine

Salvage Mutant Cel1s. IO5 Influence of purine and pyrimidine Compounds, MethoÈrexate. and Azaserine on the AccumulaÈion and Depletion of HS3 in Three CHO CeIt Lines Incubated in the presence or Absence of

L-clutamine l_09

Determination of ATp Level-s in CHO VtT Cells Àfter Exposure to Various Grovrth Conditions and Metabolic Inhibitors. II0 Effects of Gtutamine Deprivation or Sufficiency on'"c-Adeninea^ ïncorporation into Adenine Nucleoside xii i P1\GE

and Nucleotides by CHO yHD 13 Cells.. tI3 Effects of L-Glutamine and Adenosine Deprívation and Replenishment on Rate

of AccumuLation and pool Size of HS3.. 115 a) FaÈe of --P-Labelled HS3 Accumulated by CHO WT Cell-s During cLutamine Deprivation Follor^'ing 32ní *ith- drawal From the Incubation lrledium..... !25 Time Course Studies of the Effects of s-Fluorouracil on Elevated Hs3 Levels Accumulated During L-GLutamine Starvation or Methotre*ate Treatment.. I28 Effect of L-Glutamine and Adenosine Starvation on Protein Synthesis. 132 Effect of L-clutamine and Adenosine Ðeprivation on Rate of 3H-Thymidine

Incorporation lnto DNA... 135 Effect of L-Glutamine or Adenosine t{ith-

drawal and Replenishment on RNA

Accumulation by CHO l{T and CIiO GAT

CeÌIs. 139 Effects of L-Glutamine and Adenosine Deprivation and Replenishment on the

Grovrth of CHO WT and CHO GÀT Cells.. . 140 xiv PAGE

Effects of AchLya and CHO vtT HS3 on

partially purified CHO Ì{T Cel,1 DNA_

Dependent RNA polymerases.. 146 Effect of purified Ãchlya HS3 on the 3II-uTp rncorporation of into RNA by permeabilized CHO ltT Cells. I4g

Effects of Antibiotics on HS3 Synthesis.. 155 Survey of HS3 Metabolism by Various

Cultured Manr¡nalian Cel1 Lines. 159 Discussion

HS3 Synthesis by CHO WT CeLls. 166 HS3 Accumulation: The possible Involvement of Other Nutritional Changes Besides

clutamine Deprivation ]-67 Effect of Glutamine Deprivation and

Metabolic Inhibitors on cTp and ATp Level,s. I74 HS3 Metabolism: Effect of AntibioÈics... - 176 HS3 Leve1s: Elevation by MTX Treatment or clutamine Deprivation and ÐepJ-etion by 5-FU. l7B HS3 as a possible Cellul-ar Regul_ator: Changes in Rate of Accumulation and

Pool- Size of ItS3. ]-g2 PAGE

Intracel_luLar HS3 Levels: Correlation vrith protein, DNA and RNA Synthesis

and Gror^rth. . 184

protein Synthesis. 194 DNA Synthesis.... I85

RNA Synthesis.... 1e6

cro\,¡th. 19 0 Possible Effects of HS3 and Nucleoside Triphosphate LeveLs on Nucleic Acid Synthesis. 19I clutamine Avail-ability and HS3 Metabol-ism by Various Cultured lrlamiralian Cell

Lines . Conclusion. . l9g Bib1iography.... - 199 LIST OF TABLES

TABLE PAGE 1. R-r values of nucLeotides .. 97 2. Comparison of chemical and physical

properties of AchLya and mammalian HS3 ..... 9g 3. Effect of L-gl-uÈamine, adenosine and thynidine availability on HS3 metabol-ism

by CHO cAT cells ...... , 106 4. Effect of L-glutamine and exogenous purines on intracellular HS3 levels in

CHO I{T and CHO purine salvage mutants 109 5. Influence of purine and pyrimidine

compounds, methotrexate and azaserine on

HS3 metabolism in CHO WT and purine salvage mutant (yll 2I and YIID 13) cells incubated in growth medium vrith or vrithout supplemented

T,-glutamine .... lI l- 112

6. Determination of ATP concentrations in CHO V'fT ce1ls after exposure Èo different growth conditions and metabolic inhibitors...... II4 7. (a) Effect of L-glutamine r^rithdrawal and repLenishnent on rates of 3H-rdR incorporation into DNÀ in CIIo I,JT celLs....137

xva _t 1À3LE PAGE (b) Effect of adenosine $rithdrar^¡al and rep.l-enishment on rates of 3H-tdn incorporation into DNA in CI{O GAT- cel1s. 13I 8. Effect of Achlya and CHO $7T cetl HS3 on the in vitro activities of DNA-dependent

RNA polymerases from CHO I{T cells. I5O 9. HS3 inhibition of RNA synÈhesis in

permeabilized CHO WT ce]ls. 154

10. Effect of cycloheximide and actinomycin D

on HS3 level-s on CHO I'lT cells. 158 11. Determi.nation of leveI" of 32pi labefled HS3 in various mammalian ielf lines

cultured with or vrithout 2 nM

L-glutamine 162

xvl..t.t LIST OF FIGURES

FIGURE PAGE 1. Schematic diagram of the de novo purine

biosynthetic pathl^ray. 50 2. Schematic diagram of the de novo pyrimidine biosynthetic pathvray. 5I 3. One-dimensionaf autoradiogram o¡ 32pi

LabeLled formic acid extracts from CHO vlT cells incubated in the presence or

absence of glutamine..... 93 4. Two-dímens Íonal- autoradiogram of 32pi labelled formic aci-d extract from

glutamine- starved CHO I{T cells. 95 5aab. Glutamine concentra tion- dependen t 32. effect. on --P-l-abelLed HS3 and GTP levels

in cHo v{T cel-ls. 101 6. L-Glutamine and L-isoleucine starvation effect on HS3 accumulation by CHO l4lT cells. 104 7. ïncorporatlon of I4c-udenine into adenosine and adenosine nucleoÈides by L-glutamine-deprived or supplemented

YHD 13 CHO cells. 118

xJ-x FIGURE PAGE 8. !-Glutamine availability and lates of

HS3 synthesis by CHO tr{T ceI1s. I2O 9a&b. Effect of L-glutamine or adenosine deprivation on intracellular HS3 pooL sizes in CHO cell lines. I22 f0a&b. Effect of nutrient (L-glutamine or adenosine) replenishmênt on accumulated HS3 in CHO ce]I lines. I24 ia 11. Fate of -'pi-1abe11ed IIS3, accumulated

by CHO V{T ce11s during L-gl-utamine deprivation, following 32pi "ithdtur.t from the incubation medium. I27 I2a, Effect of 5-FU on IIS3 accumulated by b&c. cHo wT cel1s and L5178Y nouse leukemia lymphoblasts during L-glutamine deprivation or methotrexate treatment. 131 l-3a&b. Effect of L-gl-utamine or adenosine deprivation on protein synthesis by CHO ce]I lines. 134 14aeb. Effect of L-glutamine or adenosine availability on RNA synthesis by CHO ceII lines. I42 FIGURE PAGE 15a&b. Effect of L-glutamine or adenosine availability on the growth of CHO cell lines. I45 16. DEAE-SephadexA-25 chromatographic

profiles of DNA-dependent RNA

polymerases isolated from CHO WT cet1s.... l-48 L7. Time-course study of the incorporation ) of 'H-UTP into RNA by permeabilized

CHO V{T celIs. 153 18. Effect of antibiotics on HS3 accumulated during L-glutamine deprivation of C¡IO lqT cells. :..... 157 19a&b. Two-dimensional autoradiograms of formic acid extracts from 32ni 1.bul-L.d hu¡nan foreskin f ibrobi"asts incubated

vrith and without !-gì.utamine . J.64

xx1 ABBREVTAT I ONS

ADA adenosine d.eaminase ADP adenosine 5'-diphosphate

O(-¡"ISM alpha-minimal essential medium AMP adenosine 5 '-monophosphate APRT adenine phosphoribosyltrans ferase deficient ATP adenosine 5'-triphosphate cAIqP cycli-c 3' ,5'-adenosine monophosphate CGMP cyclic 3r ,5 t -guanos j-ne monophosphate CHO Chinese hamster ovary cm cenÈimetre cpm(CPM) counts per minute d.ATP deoxyadenosine 5'-triphosphate dCDP deoxycytidine 5'-diphosphate DFCS dialysed fetal calf serum dGTP deoxyguanosine 5'-triphosphate FCS fetal calf serum

5-FdUMP 5- fluorodeoxyuridine 5 ' -monophosphate fig. figure 5-FU 5-fLuorouraci.l GDP guanosine 5'-diphosphate gln glutamine

xx.t r clqP guanosine 5 ' -monophosphate GTP guanosine 5 r-triphosphate HCI hydrochloric acid HGPRT hypoxanthine/guanine phosphoribosyltrans ferase defícient hr hour

IMP inosine 5 ' -monophosphate M molar min minuÈe (s) ml- mil-lil-itre mM millimolar m mo1 miLlimole MTx methotrexate Mvil molecular hteight OMP orotidine 5 | -monophosPhate PEI polyethY lene imine Pi inorganic Phosphate ppcpp guanosine 3'-diphosphate, 5'-diphosphate (l1S I) pppcpp guanosine 3 ' -triphosphate , 5r-diphosphate (MS rr) PRA phosphor ibosYl amine PRPP 5 -pho sphor ibo s yl- l-pyropho sphate TCA trichloroacetic acid TdR thymidine

xx.Ì 1l TEAB trie Èhylammoni um bicarbonate Tris tris (hydroxymethyl) amino methane uci microcurie UDP uridine s'-díphcsphate ug microgram

UMP u¡idine 5 '-monophosphate UTP uridine 5r-triphosphate UV ul-traviolet WT 'wi1d typel (+) supplemented (-) not supplemented

XXJ.V TNTRODUCTI ON It is apparent now that compl-ex biochemical regulatory functions, at the enzyme leveI, have been evolved in living organisms. Yet it is evidently desirable that the diverse cellular acti-vities should be coord.inated in some fashion so as to ensure bal-anced growth during periods of rapid nutrient changes in the organism's grohrth environment. Since the isolation and character í zation of various smal1 yet unique nucleotides from both procaryotes and eucaryotes, there has been an increasing interest in attempting to understand the functions of these molecules in the reguLation of growth. The data on thê role(s) of cyclic adenosine monophosphate and bacterial- guanosine nucleotides (ppcpp and pppGpp ) in procaryotíc Arowth regulation is someh¡hat convi.ncing. The involvement of ppcpp and pppGpp in regulating transcription during the bacteriaL rstringent response' is clear. To date, however, no definitive data have been presented to demonstrate the existence of ppGpp and pppGpp in cultured mammalian cells. cyclic adenosine monophosphate and cyclic guanosine monophosphate have been implicated in regulating eucaryotic arowth. Other

xxvJ. nucleotide-containing compound.s have also been discovered and purported to have important cellular functions. Ho\4rever, diguanosine tetraphosphate has been found to be limited to some crustaceans whil-e the functi.on of diadenosíne tetraphosphate, r^rhich seemingly is more univelsal in occurrence, has yet to be demonstrated. It is apparent that not all cellu1ar responses are mediated by these aforementioned compounds. Among these responses is the coordinated physiological changes associated with the removal of an essential amino acid from the grovrth medium of cuftured mammalian cell-s, a process commonly knovrn as the 'p]eíotypic responset. In 1975, three rather unique dinucleoside polyphosphates hrere reported to exist in various fungi. These compounds were given the pseudonyms HSl, HS2 and HS3. One of them, HS3, was partially characterj.zed recently as consisting of a glutamyl- ADP-sugar moiety covalently linked in an unknown manner to a [tDP-mannitol- tetraphosphate. In vivo and in vitro studies have demonstrated that there is an apparent correlation beth'een changes in the concentrations of these dinucleoside pol-yphosphates and the rates of RNA and ÐNA synthesis. Their possible

xxvLr role(s) in fungal sporulation has also been implicated. It is thus likely that these hitherto unknov¿n compounds may be invol-ved with cel-Iular regulation during adverse growth conditions. A study of one of these dinucl-eoside polyphos- phates, HS3, found in cuLtured mammalian celLs constitutes the basis of this thesis. The aims of this research undertaking are essentially threefold; firstly, to establish the presence of one of these HS molecules (HS3) in various cuLtured mammalian cell 1ines. Secondly, to understand the physioJ.ogy involved in the netabolism of HS3 in various 'wiId-typer and mutant cell l-ines and thirdJ-y to observe the relationship betr4reen changes in the levels of intracelfular HS3 and activities of various cellular macronolecular processes during different growth conditions. fn the course of this study, it became apparent that. HS3 metabolisn is indivisibly related to purine-pyrimidine and nucleic acid metabolism. It is hoped that this thesis vrilI, in some very small- way, contribute indirectl-y to a better understanding of purine-pyrimidine metabolism and lead to a better appreciation of some of the human genetic diseases that are related to purine and pyrimidine dysfunction.

xxvl r l H I S TORICAL (1) Tissue CuLture

(a) Ðevelopment of Tissue Culture Techniques

The deveì-opment cf marnmalian cell culture techniques has provided scientists with a powerful tool to probe the intricacies of thê structure and metabolic processes of clonal mammalian cells. The basic tools of modern day ce11 culturists consists of glass and plastic substrata, defined growth media and various types of sera. The early problems encountered in attempts to establish permanent ceIl fines in vitro $rere t$ro-fold. Firstly, a proper gror,rth- supporting environment had to be formulated and secondly the difficulty in obtaining cell clones from a si.ngle ceIl for genetic studies had to be overcome. In !9I2, Carrel (1912) reported his success in keeping a chicken muscle explant viable and proliferating for many years using a groh¡th medium composed of Ringer's salt solution (Ringer 1882), chicken plasma and muscl-e extract. Until the earLy 1950s, this medium and others of similar composition r^rere the primary types of growth medj-a used for cell culture, Despite the use of such complex media, large numbers of single clones could not 3 be generated until Sanford, Earle and f,ikely (1948) showed that the growth medium had to be first 'conditione¿lr before cells seeded at a very low density will proliferate. This was achieved by successful-ly culturing single cells in very srnall volumes of g.rovrth medium. Alternatively, Puck and Marcus (f955) reported that through the use of an i.rradiated layer of rfeeder' ceIls, most of the cells from a HeLa ceII (Gey et. aI. 1952) population could be cloned. These results suggested that the cloning procedure could be more successful and simplified if the various nutrients essential for growth could be identified. This led to the rapid development of chemically defined growth media. EagJ-e (1955a, 1955b) was the first to successful-ly formulate a defined mixture of nutrients for the propagation of mammalian cel1s in cuLture. The inedium consisted. of a mixture of thirteen essential amino acidsr vitamins, cofactors, salts, carbohydrates and small amounts of eiÈher human or dialysed horse sen¡m. One imporÈant observation from these studies was the differences in the concentration of specific nutrients required for optirnal grohrth of HeLa (Gey et. al. 1952) and mouse L cells (Sanford et. al. 1948). Today, a myríad of defined growth media, each catering to the specific nutriÈional 4

requirements of various ce1l types are used. Hor,rever these are all similar to the original synthetic media formulations (Eagle 1955a, t955b). Another major development in mammalían ceLl culture followed the introduction of the use of chenically defined gro\^rth media. Up to 1960, only límíted success was achieved in propagating diploid primary cefls with a stable genomic configuration for extended periods of time. ?he importance of the establishment of such cell cultures is t\4'o- f old. f irstl-y, a stable diploid karyotype will allow such cells to be used for precise genetic studies and secondly the objection to the use of heteroploid celf fines displaying similar phenotypic propertíes of malignant cells (Hayflick and Moorhead 196t) for vaccine production could be circumvented. Hayflick and Moorhead (1961) first reported the isolation and characteri zation of numerous strains of human fetal fibroblasts which were amenabfe for culture and vaccine production. These ceIl strains, besides having a stable diploid karyotype, also had grorl'th and morphological characteristics of normal- primary cell strains. The advent and refinement of tissue culture techniques have proven to be an invaluable tool for probing the biochenical. and molecul-ar basis for hu:nan diseases and the Prenatal diagnosis of fetuses which may have various genetic refated illnesses' And novt, with the introduction of cell- hybridization, chromosome mapping and genetic cloning techniques, tissue cul-ture will" continue to be an excíting and fruitful field for research'

(b) cror,Jth Reguirements and characteristics of crrltured Manmalian cells

With improvement in cel1 culture techniques' cel1 culturists have, in recent years, increasingly focused their attention on the control of manunalian cefl gror^'th and division by serum macromolecular factors and lovr moLecular weight nutrients- Much of thê data suggest that normal mammalian cellul"ar processes are tightly controlled by a complex interaction of such grov'Èh factors and simple nutrients (Ho11ey f975; GosPodarowicz and Moran 1976). All cetl lines studied have shown that serum is an absofute growth requirement' This l-ed to early attempts to isofate specific growth factors from serum' The first macromoLecular seru¡n fraction isolated was fetuin (Lieberman and ove 1957;Fisher et' aI' 1958) and itwasshol'ntobeessentialforpromotingceJ.lattachment and stretching on a glass substratun' Since then' 6

nunerous groh'th factors have been isolated from serum, vari-ous mammalian tissues and even from establ_ished cuLtured ma¡nmalian cells (Gospodarowicz and Moran ]-976). Recent publications (Rizzino and Sato I978; Hayashi and Sato 1976) reported the successful growth of various cell l-ines in serum free media which had been supplemented with purified fetuin, numerous hormones, transferrin and 2-mercaptoethanol. These studies by Sato and coworkers (Rizzino and Sato l-978; Hayashi and Sato 1976) and Gospodarowicz and Moran (1976) strongly suggest that one of the major roles of serum in ceII culture is for the supply of hormones. Accumulated data for Èhe effects of low molecuLar weight nutrients on the gror¡¡th of ce11 lines are extensive. However, at least for mouse L cells (Sanford et. aI. 1948) and HeLa cells (Gey et. aI . 1952) , the essential nutrilites are D-glucose, l,-arginine, L-cysteine, l,-glutamine, f-histidine, L-isoleucine, L-J-eucine, ].-lysine, L-methionine, L-phenylalanine, !-threonine, L-tryptophan, L-tyrosine and L-valine. Of all the amino acids supplied, L-glutamine is required at the highest concentrations as cell growth is rapidly arJrested on removal or depletion of it (Eagle et. aI. 1955; Ley and Tobey 1970; Pardee I974). Glutamine is an imporÈant 7 metabolite considering j.ts involvement in protein bíosynthesis, d.e novo synthesis of nucLeotides, as an energy source, synthesis of coenzymes and in several aminating reactions (Stadtman 1973¡ Zíelke et. aI . l-976) . Vitamins, cofactors and ions such as choline, foÌic acid, nicotinarnide, pantcthenic acid, pyridoxal, riboflavin, thiamine, N.+, K+, Mg++, c"++, c1- and pon3- are also essential- (Eag1e 1955c). Trace elements and other vitamins may be supplied by the addition of serum. The unique physiological roles of some of the ions in manu¡alian tissues was firsÈ recognized by Ringer in l-882. It is obvious from studies that most of the physioLogical processes require some ions as cofactors. Calcium, particuLarly its role in the cAMp response, is of paramount importance in the regulation of ce11 gror^rth (Rasmussen 1970). Of interest are tr^ro recent reports (Rubin and Sanue 1977; Barnes and Cotowick t977) which demonstrated Èhat exÈraceIlular complexes of Ca++, ppi )- and HPO4- are involved in the stimulation of ce1lu1ar proliferation, sugar uptake and thymidine incorporation in mouse 3T3 ce1ls. Transforned ceÌls in cul-ture have been shoÌ,rn to have different norphology and grohrth requirements from normal primary ce11 lines. Transformed celIs general-l-y 8

have a lower requirement for the macromolecular serum factors for growth (?emin 1966). In certain instances, different fractions of serum stj-mulate grovrth of norma] and transformed celts (paul et. aI. l97L; Rudtand et. al. I974). Also transformed cell-s continue to proliferate at suboptimal nutrient conditions unlike normal untransformed cells (Holley 1975) and will eventually die if fresh medium is not supplied. This has led to a postulation that transformed ce1ls may have lost their so called restriction point (R-point) control (pardee 1974) . this has some supportive evidence in that, in vivo, tumour cells require an ample supply of nutrients to survive (Folkman et. a1 . 7974). Some of the other characteristics expressed in cul-ture by transformed cell-s as compared to normal cells are an unlimited proliferative lifespan (HayfLick and Moorhead 1961), drasÈic karyotypic alterations (Levan and Biesele 1959, Todaro And creen 1963; Hayflick And Moorhead 1961-), ability to groh¡ in suspension culture (Macpherson and Montagnier 1964), disoriented growth patterns (stoker and Abel 1g62; Temin and Rubin 1958), poor attachment and spreading abílity on a solid substrate (Abercrombie 1970; Ambrose et. a1. I970) possibly due to changes in the size and number of actin cables (Tucker et. a1. ]-97g) . different composition of 9

the proteins and morphology of the plasma membrane (Hynes 1976; Robbins and Nicolson L975) , 1o$/ intracellular level-s of cAMP when grown to high densities (Otten et. al. L97!) , the agglutinability by wheat germ agglutinin (Aub et. al. 1965; Burger and Goldberg 1967), and the ability to produce tumours when inocuLated into appropriate animals (Ear1e and Nettleship 1943). The better understanding of such in vitro properties of transformed cel1s in culture, hopefully, may shed some light on the mechanisms involved in Þ '¡1!g tumourÍgenesis.

(2t l4ammalian Cells: RNA BiosynÈhesis and Amino Acid

W i thdrawa l-

In bâctería, various cel-lular processes including the accumulatj-on of rj-bosomal--RNA are abruptly curtaii"ed on starvati-on for an essential- amino acid (Cashe] and Gallant 1974; see tThe Stringent Response' in this 'Historical'). Unlike this well-studied rstringent response' of bacterial ceIIs expressed during amino acid deprivation (stent and Brenner 1961; Cashel and Gallant 1974) , the physiological consequences of amino acid r¡rithdravral from the growth media of cultured mammalian cells remain obscure. For example' the total RNA synthesis 10

in amino acid starved mouse L-cetls and Landschutz ascites tumour ceLls was unimpaired (Sko1d and Zetterberg \969 ¡ Shields an¿l Korner 1970) . Other reporÈs, hovrever, have shown that deprivaÈion of other mammalian cell l-ines of essential amino acids led to a decrease either in uridine incorporation into RNA (Bol-csfoldi et. al. 1971; Snulson and Thomas 1969) or the synthesis of pre-rRt{A (Vaughan 1972; Jolicoeur and Labrie 1974; Grummt et. aI. 1976) and r-RNA (Nakashima et. al. 1976). Concorünitantly, mRNA synthesis was not affected (Jolicoeur and Labrie 1974; Nakashima et. al. 1976). The biochemical mechanisms underlying this regulation of rRNA synthesis, however, are in general unknown. Gru¡nmt and GruIürt (1976) proposed that rRNA synthesis during amino acid deprivation is directly controlled by the actuaL concentration of ATP and GTP. These nucleotides have been found to be depleted upon amino acid withdrawaL. Recentl-y, a sinj-lar study using a different ceI1 Line failed to show any significant changes in ATP and GTP pool sizes before and after amino acid withdrah¡aI even though rIìl':A synthesís was severely limited (Dehlinger et. a1. I977). other proposed control mechanisms for rRNA synthesis include the requirement of a rapidly turning-over protein factor lirhich is not 1I

produced during amino acid starvation (Cereghini and Franz-f ernanô'ez L976), or the production of a regulatory molecule, anal-ogous to bacteríal ppcpp,during amino acid deprivation which may directly interact with the transcríption machinery (Goldberg and St' John 1976)' concrete information on this aspect of mammalian cell- growth regulation is therefore rudirnentary-

(3) Physiology, Biochemistiy and Possible RoIe(s) of Polyphosphates and Unique Phosphorylated Nucleosides in Procaryotic and Eucarvotic Ceflular

Regu I ation

The relationship betvreen all J-iving cells and the external milieu is essentially one of either survíva1 or death. Thus it is essential that cel]-s must have evolved certain intrinsic regulatory mechanisms with t'thich they could rapidly adjust their ceLlular metabolism according to the prevailing grovrth environment.The isolation and charactereri zation of various types of polyphosphates and unique phosphorylated nucl-eosides produced during altered growth conditions have led to the current interest in their possible role as mediators of cellular regulation of grobtth in procaryotes and eucaryotes' the 12

folLowing is a histo¡ical overview of some of these polyphosphates and unique phosphorylated nucleosides which have been found thus far in both procaryotes and eucaryotês.

(a) Polyphosphates (PoIyP)

Polyphosphates (polyP ) were first isolated and characterized from yeast by Wiame (1947r 1948) and Schmidt et. al. (1946). Since then, polyP have been isolated from a large number of microorganisms, higher plants and animals (Harold 1966). ft has been observed that there is a considerable heterogeneity in the size and quantity of endogenous polyP in different organisms (Harold 1966). On the contrary, PolyP with chain lengths of 3,4,10 and 300 have been isolated from Saccharomyces cerevisiae

(Harold 1966; Ludwig et. al. 19771 . The biosynthesis of polyP requires only one reaction that is catalysed by polyphosphate kinase (Harol"d 1966). This enzyrne which was originally found in yeast (yoshida and Yamataka 1953) and later purified from Escherichia col-i (Kornberg et. a1. 1956), catalyses the reaction .ATP + (Pi) n ADP + (pi) n+l 1L It is Mg"' dependent and is strongLy inhibiÈed by NaF 13

and ADP (Kornberg et. al. 1956). The inhibition of the reaction by ADP suggested that a low in .rirro ATPTAnP .rati-o would result in a reversal of the teaction. Unlike the biosynthesis process, the catabolism of po1yP is catalysed by numerous enzymes. PresumabJ.y, a low cellular ATPTaon ratic will resuLt in degradatlon of polyP by the reversal of the polyphosphate kinase reaction. Another degradative enzyme isolated was polyphosphate adenosinemonophosphate phosphotransferase. The purified enzyme (Dirheimer and Ebel 1,965) requires )+ Mg- and catalyses the transfer of phosphate from fairty high chain length polyP to AMP. Szymona (1962a) and Szymona et. aI. (1962b) were the first to discover pol-yphosphate glucokinase which utilizes pofyP to phosphorylate glucose to gl-ucose- 6-phosphate . This Mg2+ dependent enzyme could also use glucosamíne but not mannose or fructose as substrates. PolyP gJ-ucokinase is present in numerous mycobacteria and in Corynebacterium diphtheriae but is absent in Escherichia coIi, Aspergillus aerogenes and AsÞergillus niger (Szymona et. aI. 1962b). Polyphosphatases are apparently heterogeneous and widespread in various microorganisms and animal tissues.

In Saccharomyces cerevisiae, polyphosphata se s that are specific for short chain polyP r^¡ere found (Harold 1966) , 74

whj-Ie those purified from Corynebacterium xerosis are specific for high molecuLar weight polyp. ALso no short chain polyP intermediates r^'ere detected after hydroJ-ysis suggesting that the enzymatic attack proceeded sequentiall-y from the terminal- ends (Szymona et. al-. I962b) . In contrast, the polyphosphatase from Aspergillus niger cleaved polyP into smaller fragments vrithout Iiberation of inorganic phosphate (saÌI et. a1. 1956). Final]y a rather unique pol-yP degradative enzyme, polyphosphate fructokinase, lras isol-ated from Mycobacterium phfei (Szymona and SzumiÌo 1966). This enzyme catalyses the transfer of Pi from polyP and ATP to fructose. The biosynthesis of this enzyme is inhibited by glucose and induceil in fructose gror4rn cells. In spite of the accumuLation of data on polyP metabolism, 1ittle is known about its biologj-ca1 functions. Physiological studies, however, related to pol-yP metabolism have shed some light on its significance. A cornmon feature of polyP metabolism is the large fluctuation in intracellular pool sizes which itself is dependent on the grovrth conditions. PolyP l-evels are Low in organisms undergoing rapid grolt'th while el,evated leve.Is prevail during conditions of nutritional imbal-ance and thus resulting in growth inhibition (Harold 1966). In 15

bacteria, polyp accumulation is triggered by nitrogen (SaLl et. al. 1956), sulfur (HaroLd 1966) and zinc (Vtinder and OrHara 1962) starvation. Related to nitrogen starvation was the inverse retationship observed betvreen nucleic acid biosynthesis and polyp levels in Aerobacter aerogenes (Harold 1966). Owing to the fact that upon resumption of grovrth and the concommiÈant degradation of po1yP there is incorporation of the released pi into nucleic acids, it has been suggested (Harold 1966) that polyP may act as a phosphate storage sink during periods of non-growth. In yeast and filamentous fungi, sinilar

physioJ-ogical properties have been observed. The involvemenÈ of potyp in growth and. differentiation of Aspergillus niger and physarum polycephal-um is wel.l dccumented (Nishi 1g6I; Goodman et. al, 1969). rn Aspelgi11us niger, Iarge amounts of poLyp are found in

the spores. During the germination process polyp r^¡a s found degraded by elevated levels of polyphosphatases to Pi thus serving as a source for phospholipid and nucleic acid synthesis (Nishi 1960; Nishi 196I). Goodman et. aI. (1969) reported that in physaru¡n polycephalum the level- of ínsoluble polyp was l-ow during periods of active RNA and DNA synthesis and high during the differentiation (encystment) process. Hildeb¡andÈ and Sauer (1977) proposed I6

that insoluble polyP may be involved in the differentiation process of this organism by sþecificalJ-y inhibiting rRNA synthesis. Obviously the existence of huge amounts of po1yP in so many diverse organisms and plants points to an important role for it in celluLar regulation. Data to ascribe a definitive role (s) for polyP, however, remain elusive.

(b) Adenosine 3r:5r Cyclic Monophosphate (CAMP)

(i) Cyclic AMP: Biosynthesis, CataboLism and Possibl-e Regulatory Role (s)

Cyclic AÌ4P was first isolated and characterized by Sutherland and RaIl (1958). The synthesis of cAMP from ATP is catalysed by a plasma membrane Located enzyme, adenylate cyclase. The activity of this enzyme is dependent on a divalent cation 1t'lg2+or Mr,2+) b,-rt is strongly inhibited by ca2+ (Robison et. a]. 1971) .

The catabolism of CAMP to 5'ÀMP is effected by a cAMP phosphodiesterase which was first found in heart, brain and liver extracts (sutherland and Ra11 1958). f,ike adenytate cyclase, this phosphodiesterase is also I7

)L Mg'' dependent and its activity is strongly inhibited by the me thylxanthines , theophyttine and caffeine (Robison

et. al. 19 71) . The implication of cAMp as a possible regulatory molecule stems from observations that certain physiological responses resulting from hormonal stimulation were

interposed by transient elevation of intracel-IuIar cAI,Ip levels (see Robison et. al. 197l). The best understood of this phenomenon is the action of epinephrine and glucagon on glycogenolysis. Data shor,v that upon hormonal stimulation of specific tissues, cAMp leveIs become elevated (Sutherland and Rall 1958). This cyclic nucleotide then activates a specific c.AMp-dependent protein kj,nase which in turn acÈivates phosphorylase b kinase through direct phosphoryl-ation of the kinase (lfalsh et. a] . 1968; Robison et. al. 1971), thus ultimateJ"y leading to gl-ycogen catabolism. Numerous cAltp-dependent protein kinases have since been found in various mammaLian tissues and other eucaryotes (see Robison et. al. 1971 and Rosen et. aI. 1977 ) ft should be pointed out that the t$ro exist.ing classes of cAMP-dependent protein kinases are themselves regulated in a complex way by a heat and acid stabÌe inhibitory protein found in numerous mammalian tissues (Rosen et. al. I977). Recent studies have found other protein kinases 18

that are CAMP independent (Rosen et. a1. 1977\. Hor^rever little is known about the regulation of these proteins. It is pertinent to point out here that numerous hormonaf actions are independent of cAMP metabolism (Robison et.

a]. 1971) . CycIic AMP has now been detected in bacteria, various unicellular organisms, most animal tissues and in some plants (Rickenberg 1974; Robison et. aL. 1971). Hol^rever or,ring to the voluminous amount of data, only some of the salient aspects of cAMP metabolism and function (s) vri11 be discussed in the following sections.

(ii) cyclic AMP in Bacteria

llhen bacteria are cultured under glucose-rich conditions, the synthesis of various inducible enzymes necessary for the utilization of other sugars are either repressed or occur at greatly reduced rates even in the presence of the specific inducing sugar. The bacteria will synthesize the enzymes required for the metabolism of other energy-yielding sugars only when the gfucose supply is exhausted (Monod 1,9471. This intriguing metabolic feature of bacteria is known as catabolite repression (Magasanik 196l). I9

Iuakman and Sutherland (1965) were the first to discover that glucose could Iolrer cAMp ]evels in E. col,i. This led to a series of elegant studies which provided an insight into one of the functions of cAMp in procaryotes. Pastan and Perlman (1970) demonstrated that in the presence of both glucose and lactose, high levels of exogenousl-y supplied cAMP could overcome glucose-mediated repression of the synthesis of f-galactosidase and other inducible enzymês. The subseguent isotation of appropriate bacterial mutants proved inval-uable in these studies. For exampLe, adenylate cyclase deficient E. coli were unable to ferment various sugars even in the absence of glucose unless c.A.È4P !ìras provided. Revertants of such mutants vrere invariably found to possess normaL adenyl-ate cyclase acÈivity (Perlman and Pastan 1969). Another group of mutants failed to metabolize various sugars even in the presence of cÂMP. The lesion was subseguently found to be due to an absence of a cAMp binding protein known as either the catabol-ite gene activator protein (CAP) or the cAMP receptor protein (CRP) (Emmer et. al. 1970¡ Zubay et. aI. 1970). Using a cell-free in vitro transcription system, de Crombrugghe et. aJ-. (197I) showed that the synthesis of lac messenger RNA required both cAllP and CRP. Pastan and his collaborators (Pastan 20

and Perlman 1972¡ de Crombrugghe et. aI. 1971) have proposed a model for the mechanism of cAMp-CRp action on the lac transcriptíon system. Cyclic AMp first binds to CRp form a complex hrhich in turn binds to a site on the plomoÈer of the 1ac operon. This then facilitates the effective binding of RNA polymerase to another promoter region leading to the initiation of transcription. There is genetic evidence to suggest that these ti,ro postulated promoter sites exist (Beckwith et. a1. 1972). Thus the rate of transcription may be regulated by the relative amounts of intracellular cAMp which in turn is dependent on the externa]- source of gJ-ucose in the gro\,/th environment. This phenomenon clearly demonstrates the irnportant second 'messenge!' (Robison et. aI. 1971) role of cÀMp in bacteria.

(iii) Cyclic AMP in the Fungi

Since the demonstration of cAMp in various fungi (see Robison et. al. I97I), the possible involvement of this cyclic nucleotide in the metabolism of such organisms has been reported. The classical organism used in most of these studies is the celluLar slime rnoul-d, Dictyostelium discoideum. In the vegeÈative phase, the organism occurs as a unicelluLar amoeba. On starvation, 2I the amoeboid cells aggregate into multicetlular structures which further differentiate into fruiting bodies. The aggregation phenomenon due to starvation is unique in that initially, only a few anoeba acÈualIy release cAMP. These cells form the aggregation nucleus, which in turn excretes CAMP in pulses (Bonner 1971-) thus attracting other nearby amoebae to form larger aggregation centres. This aggregation process is also specifically stimulated by exogenously added cAMp (Konijn et. al. 1,967). The role of a CAMP phosphodiesterase in regulating the cyclic nucleotide leve1s and thus the aggregation process has been suggested by Riedel and.Gerisch (1971) . D. discoideum amoebae apparently excrete a cAMp phosphodiesterase during vegetative gro\^rth. Under unfavourable gror^¡th cond.itons, however, just prior to aggregation, an inhibitor of this enzyme is released into the medium (Riedel et. al. l_973). The lack of this CAMP phosphodiesterse inhibitor in non-aggregating mutants suggested that the development phase ôf D. discoideum is initiated, in part, by changes in the extracell-ular levels of cA¡{P (Riedel and Gerisch 1971). In addition to the aggregation ro1e, CAMP added at high concentrations to amoeboid cells caused them to 22

differentiate only into stal_k cells (Bonner 1970). Other effects of cA.tqp in fungi include the ce11 cycJ.e-dependent inhibition of a protein kinase activity in physarum pol-ycephatum (Kuehn t972) and the reversal of glucose repression on sporulation in yeast (Tsuboi et. aI. 1972). Also cAMp may be involved in glycogenolysis in Neurospora crassa (Te1lez-fnon and

Torres 1970') , the regulation of citraÈe fermentation and sucrose utilization by Aspergíllus niger (I,ùold l_974) and during the germination process of Bl-astocladiella

e¡nersonii (Gomes et. al. I97g) -

( iv)

The most interesting studies on cAMp metabolism have been those related to its rol-e in the regutation of gro$rth of mammalian cells in culture. Burk reported that gro$/th was impeded by the addj.tion of cAMp phosphodie s terase inhibitors to normal and trânsformed baby hamster kidney cells (Burk 1968). Subsequently, Ryan and Heidrick showed the direct inhibitory growth effect of cA.l"lP on cultured cel1s (Ryan and Heidrich 1968). Various ce11 lines derived from different Èi_ssues are now known to have a decreased growth rate in the presence ¿3

of added CAMP (or cAIvlP anaLogues) or various agents which raise intracellular cAMP Levels (Pastan et. 41. 1975). Further support for the role of CAMP as a physiological growth inhibitor was provided by Pastan and co-workers. Johnson and Pastan reported that prostaglandin EI sloÌ^red the grohrth of cultured fibroblasts. This agent which is knor^rn to activate adenylate cyclase and therefore raises intracellular eAMP was found to be ineffective in inhibiting growth of a mutant resistant to prostaglandin Et activation. Also prostaglandin Br, r^lhich does not

elevate intracellular CAMP r^ra s not gror^tth inhibitory (Johnson and Pastan 1971; Johnson et. aI. !972). ChoLera toxin, which has been found to inhibit DNA synthesis and therefore growth, probably acts Like Prostaglandin E, in being able to activate adenylate cyclase (Hollenberg and Cuatrecases 1973). Serum deprivation of normaf cells results in growth inhibition at the G, phase of the celI cycle and with the conconnitant el-evation of intracellular CAMP (Pardee 1974; Kram et. al. 1973). On the other hand, the effect of nutrient starvation on celluLar c-A-ÌlP renains unclear' Jolicoeur et. a1. (1974) reported that CAMP concentration in Landschutz tumour cells was unperturbed by totaf amino acid starvation while CAMP accumulated in BaIb 3T3 cells deprived of glutamine and histidine (Pastan et. 41. 1975). 24

This díscrepency may be due tÕ Physiol-ogical differences betrlreên the ce11 lines examined' Besides being a possibLe growth regulator, CAMP has also been implicated in affecting the differentiated staÈe of cells in culture (Baum et. a1. 1978; Mill-er et. al- 1978) One of the characteristics of transformed cel-l-s in culÈure, unlike normaL cells, is their ability to continue to groh¡ beyond confluency and to maintain low levels of

CAMP under such conditions (Otten et. al. 1971) . carchman et. al-. (1974) demonstrated the relationship between the transformeil and normal state of NRK cetls with regard to intracellular CAMP l-evels by ínfecting such cel1s with a temperature sensitive Kirsten sarcoma virus. Tomkins and his collaborators have suggested that the 'pleiotypic program' (Hershko et. al . 1971) in mammalian cells is regulated by CAMP, the rpleiotypic regulator'. Further, they have suggested that the transformed state may be the result of a loss in the control of CAMP metabolism or in the action of this effector (Kram et. al. 1973) '

(c) cuanosine 3' :5' -Cvclic l4onophosphate (CGMP)

Cyclic GMP was originalty discovered in 1963 in rat urine (Ashman et. aI. l-963). Subsequently, it vras tt

detected in various animal tissues. Guanylate cycl_ase, the ccMP biosynthetic enzyme, has sínce been found in all- mammalian tissues examj-ned other than sperm. The enzyme is also present in insects, bacteria, birds, fish, fungi and probably in plants. phosphodies terase s specific for cGMP degradation have also been found (coldberg and Haddox 1977). In comparison ccMp, like cAMp, is able to activate numerous cGMp specific protein kinases (Kuo and Greengard 1970; Nishiyama et. aI . 1975). The mechanism of acÈion (Linco1n et. a1. L977) and the physiological significance of the phosphorylation of specific proteins by these CGMP protein kj_nases remain unclear. In contrast, CGMP appears to have an antagonistic. regul,atory effect on those bidirectionally controlLed .functions which are also mediated by cAMP. This phenomenon is al-so known as the 'Yin - Yang' effect (GoLdberg et. al. Ig74). For example during the celt cycle of Novikoff hepatoma cells, hígh extraceLlular levels of cGMp were found during mítosis htith corre spondingJ-y low concentrations of cAMp (Zeilly and Goldberg 1977). Other 'yin - yang' reguLated biological functions include cardiac muscle contraction, J-ysosomal secretion by leucocytes, neuronal excitabil-ity, lymphocyte proliferation, response of pclymorphonuclear leucocytes to chemotactic stinuJ-i and T-j.ymphocyte mediated 26

cytotoxicity (see coldberg and Haddox Ig77). Another significant difference beth'een cGMp and cá{P metabolism is the lack of evidence to shor^r that agents which stimulate the accumuLation of intracellular CGMP also activate guanylate cyclase activity in vitro (Goldberg and Haddox 1977) unl-ike that for cAMp metabolism (Robison et. al-. 19771 . This is further substantiated by the fact that guanylate cyclase is relatívely insensitive to hormonal activation (Goldberg and Haddox l-977). These differences suggest that ccMP functions in an unique way; a role(s) which has yet to be unravelled.

(d) BacÈerial Ms (ppcpp and pppcpp) Nucleorides

(i) The Stringent Response

fn bacteria, at least, the deprivation of a gror^rth essential amino acid or the li¡nitation of transfer RNA (t-RNA) aminoacylation will result in an abrupt cessation of numerous ce1lular activities commonly knoh'n as the rstringent response' (Stent and Brenner 1961). Among.the cellular activities curtailed are the accumulation of stabLe RNÀ (CasheL and GaLlant I974) , biosynthesis of Lipids (Sokawa et. a1. 1968), nucleoÈides (Gall,ant et. al. 27

l97I), polyamines (Holffa et. al. Ig74) , uptake of purines and pyrimidines (Nierlich J.968; Edtin and Neuhard L968) , phosphate (Cashe1 and Ga1lant 1968), glucose (Sokawa and Kaziro 1969) and an increase in cellular proteoJ-ysis (Sussman and Gilvarg I959).

Bacterial mutants which failed to restrict RNA accumulation during amino acid starvation and are rreLaxedr therefore said to possess control of RNA synthesis were subsequently isolated (Borek et. a1. 1955; Stent and Brenner 1961). This anomaly vüas the resul-t of a mutation at a genetic locus caLled the RC gene (Stent and Brennet 1961) , which has sj.nce been renamed the relA gene- Thus the wiLd type bacteriai. stringent stïain bears + the reIA allele, while the rreLaj{edr strain the reIA allefe (Fiil and Friesen l96B). Fiil (1969) d.emonstrated -! that the relA' gene product is the cytoplasmíc mediator of RNA accumulation. The rol-e of the relÄ+gene product is discussed in the next section.

(ii) In Vivo Synthesis of ppcpp and pppcpp

Oh/ing to the Õbservation that many d.iverse physiological effects of amino acid starvation in bacteria (as described in preceCing section) are not the direct result of blocked RNA accumul.ation (CasheI and GaLtant I96g; caf l-ant and Harada 1969) , it was proposed by Cashe1 and Gallant (1968) that the function of the relA gene product was essential for the synthesis of a metabofic inhibitor. Subsequently t\,úo unusual nucfeotÍdes, rmagic spot' I and II (MSI and MSII) ,were isolated but only from acid extïacts of amino acid starved relA+ E. col-i and not from relA cell-s (Cashef f969r Cashef and Gaflant 1969).

Lund and Kjeldgaard (1972) have estimated that level s of ppGpp, accumulated under stringent conditions, are about the same as intracellular ATP concentrations. Using other E. cofi strains (Cashel 1975; Gaflant and Harada 7969') , it has been estimated that during amino acid deprivation the cellular ATP concentration ís

10-12 mM, while the cTP fevef drops to 2 mM from 4 mM.

During this period, ppcpp normally rises to about 4 n-t4. pppcpp is typically about one-third that of ppcpp though its been reported that the accumufation of pppcpp is higher than ppcpp in Bacillus subtilis (Sv¡anton and Edlin 1972). There are exceptions whereby relA+ bacterial strains did not accumulate pppGpp under stringent conditions (Cashel l-969;discussed laÈer) . Chemical analyses of purified ppcpp sho\,red it to be guanos ine- 5 ' - diphosphate- 3 ' -diphosphate (Cashel and Ka.lbacher 1970; Sy and Liprnann 1973) while pppGpp is 29

t probably guanosine-5 -triphosphate- 3 ' - díphosphate (Haseltine et. al. 7972). Studies of the kinetics of ppcpp metabolism suggested that this unique molecule may be involved in blocking RNA accrmuLation during the 'stringent response' . For example, q'hen. RNA synthesis was directly inhibited with antibiotics, no ppGpp was formed; thus implying that ppGpp production is not a secondary consequence of blocked RNA synthesis (Cashel and caLLant 1969; Gallant et. aI I97O¡ Lazzarini et. aL 1971). Also ppcpp accumulation conmences within seconds on amino acid deprivation while the inhibition of RNA synthesis is not apparent until a minute later (cashel 1g6g; callant et. aI. 1970; Fiit et. a1. 1972). On reversal of the stringent state, the recovery of RNA synthesis is preceded by a rapid first-order decay of the accumulated ppcpp (Cashel 1969; Stamminger and Lazzarini I974'). Another feature of ppGpp metabolism is the restoration of RNA synthesis and the absence of ppcpp ûrhen protein synthesis inhibitors were added Èo amino (CasheL acid sta¡ved ==ÀA* cells 1969; Lund and Kjeldgaard I972). This result !r'as simply explained by the trickle- charging of the t-RNÀ species specific for the missing amino acid by free a¡nino acids generated during .intracellular protein turnover when protein synthesis 30

r¡as totally inhibited (Kurland and Maaloe 1962,) . Kaplan et. aI. (1973) have reported, hovrever, that using temperature sensitive t-RNA charging mutants, both ppGpp and pppcpp accumuLation \^rere inhibited by tetracyclíne even when one species of t-RNA \,t¡as completely deaminoacy lated. These data suggest that the presence of both deacyl-ated t-RNA and an intact protein synthesizing apparatus are necessary for relA+ bacteria to exhibit the stringent state and for the generation of ppGpp anC pppcpp (Cashel- 1975). This agrees vríth the proposal that these unique guanosine nucleotides are produced by an 'idling reaction' of protein synthesis (Cashel and caLlant 1969). The inhibitory effect of rifampicin, an mRNA synthèsis inhibitor, on ppcpp synthesis during amino acid deprivation is únknown. It is possibl_e that nRNA is reguired for the 'idì.ing reactiont. and thus for ppcpp synthesis (!Íong and Nazar 1970; de Boer et. al . 1973). The data available, therefore suggest that the relA gene product is a factor associated rdith the bacterial- protein synthetic apparatus.

Recentl-y another genetic locus, designatêd spoT for rspotlessr, has been implicated in the bacterial stringent response. The gene product of the spoT locus is probably involved in the conversion of ppcpp to pppcpp as evidenced by the following j_n vivc observatj.ons. 31

Firstly certain E. coÌi spoT strains faired to accumurate any appreciable levels of pppGpp during amino acid starvation even though intracellutar concentrations of ppcpp hrere nearl-y triple those attained by spoT+ strains under identical growth conditions (Lafffer and Galfant 1974) Secondly, v¡hen spoT- muÈants were replenished with the missing amino acid, the decay of ppcpp was approxirnately ten to thirty fol-d slower than for gpT+ strains (],affler and callant 1974; Stammj.nger and Lazzarjlni lg:.4\. This possible mechanism of ppGpp and pppcpp metabolism in spoT bacterial sËrains has yet to be demonstrated in in vitro êxperiments. But recent in vivo and in vitro sÈudies ( Chaloner-Larsson and yamazaki 1976; Fiil et. aL. Ig77 wyer et. al . 1976; Heinemeyer et. a1 . 1978) have resulted in the postulation of a different model for the consequences of a mutation at the spoT locus (Sy Ig77). The model encompasses the follor¡/ing salienÈ features. Firstly, the primary result of the spoT mutation is a reduction in the rate of ppcpp caÈabolism to cDp. Secondly, pppcpp is the precursor of ppcpp and not vice versa as suggested by LaffLer and callant (I9741 . Thirdly, ppcpp negatively controLs the conversion of GTp to pppcpp and fourthly, a catabolite of ppcpp negativety conÈrol-s the conversion of pppcpp to ppcpp. An important point to note is that these stringent spoT E. coli 32

behave exactly like stringent spoT+ celfs with regard to the reciprocal rel-ationship between rates of RNA synthesis and the levels of íntracetlular ppcpp (Cashel I969, Lund and Kjeldgaard L972; Stamminger and Lazzariní Ig74).

(iii ) I¡ ylqrg Synthesis of ppcpp and pppcpp

The synthesis of ppGpp and pppGpp using an in vitro E. coli system was first reported by Haseltine et. aI. (1972). GDp and cTp r4'ere the required substrates

for ppcpp and pppGpp synthesis respectively vrith ATp being Èhe pyrophosphate donor. Other reguírements for this synthesizing system, besides buffers and sa1ts, are high salt \"ashed ribosomes and a so calLed 'stringent (Haseltine factor' et. al . 1972) present in the 0.5 M NH4CI vrash of ribosomes. When a full complement of amino acids was added to the ppcpp-pppcpp synthesizing mixture, no ppcpp or pppcpp \4,ere made with the high sal-t washes from the ribosomes of either stringent or reLaxed bacterial strains. fn the absence of added amino acids, however, only the synthesizing assay containing the 'stringent factor' from stringent strains was able to produce considerabLe amounts of ppcpp and pppcpp. This is in agreement vrith the in vivo synthesis of these unique guanosine nucleotides (Cashe] 1969; Cashel and callant 1969). In vitro mixing experiments showed that without added amino acids, ppcpp and pppcpp b'ere synthesized even when both high sal-t ribosomal- v,/ashes from stringent and relaxed sÈrains vrere present (B1ock and HaselÈine I973). This confirmed the genetic studies which showed thaÈ the relA+ allele is dominant (Fiil 1969) . ft is of interest to note that all independent relA mutanÈs have been found to produce 'some residual 'stringent factor' activity as detected by the in vitro ppcpp-pppcpp synthesizing assay. The lo$, activity from each nutant vras directly reÌated to its characteristic thermoLability (BIock and Haseltine

1973) .

Other studies, using highly purified 70S ribosomes, indicaÈed ÈhaÈ both mRNA and t-RNA were also required for the in vitro synthesis of bacterial pPcpp and pppcpp (Ilaseltine and Block I973; pederson et. a1. 1973) , thus sugEesting that the original- ribosomal preparations of Haseltine et. aI . (1972) were actually contaminated with these tr4ro RNA species. Using a highl-y purified modification of the Haseltine et. al-. (1972) procedure, pederson and coworkers (1973) 34

and Haseltine and Block (1973) were able to demonstrate that maximum rates of ppcpp and pppGpp synthesis could be achieved in the presence of a complete protein synthesis initiation complex v\¡ith an uncharged t-RNA binding to its specific ccdon situated at the acceptor site. This mechanism of guanosine tetra- and pentaphosphate synthesis is similar to that previously proposed by Cashel and callant (1969).

(iv) Effects of ppcpp and pppcpp on Cel1ular Enzymes, Translation and Transcription In Vitro

Nunerous in vivo and in vitro studies suggest that many bacÈerial cellular processes may be bl,ocked during periods of elevated intracellular ppcpp and pppcpp levels. For example acetyl-CoA carboxylase which is invol-ved in lipid biosynthesis is inhibited by ppcpp (Po1akis et. al. 1973). The de novo biosynthesis of both AMP and GMP from IMp cari be curtailed by ppcpp through the inhibit.ion of adenylosuccinate synthetase and IMP dehydrogenase respectively (Galtant et. al I97l). Using isolated bacteriaL membrane vesicles, Hochstadt-Ozer and Cashel (1972) were able to show that ppcpp potently inhibited purine uptake and the activities of various purine phosphoribosyl-transferases. ppcpp and pppcpp, at a concentration of 1 n¡,1, inhibited bacterial ADp- glucose synthetase (a raÈe limiting enzyme for gJ_ycogen synthesis) by 618 and 77? respectively (Dietzler and Leckie I9771. The effect of ppcpp and pppcpp on bacterial protein synthesis is at present rather ambiguous. In vitro experiments demonstrated that ppcpp, being quite similar in structure to cTP, inhibited protein synthesis by preventing the formation of the protein synthesis initiation complex (yoshida et. aI. I972). However it is reasonable to assume that in vivo protein synthesis would itself be blocked during amino acid deprivation. ?he physiological significance of the relationship between ppGpp and protein synthesis remains obscure. The inverse reLationship bethreen rRNA accumulation and ppcpp and pppcpp levels is weLl documented. SÈudies using partially purified DNA-dependent RNA polymerases showed that ppGpp specifically inhibited RNA synthesis (Cashel 1970). Subseguent reports by Reiness et. al. (1975) and van Ooyen et. al. (19?6) demonstrated that rRNA synthesis was preferentialty and significantly inhibited by ppcpp and pppcpp probably through the inhibition of rRNA chain initiation ( van Ooyen et. al. 1976; Travers 1973; Travers 1976). The templates used 36

in the highly purified in vitro transcrip tion-tran s lation system were either E. coli DNA or phage DNA carrying an E. coli rRNA cistron (Reiness et. aI. l-975t van ooyen et. a1. 1976). The E. col-i DNA-depèndênt in vitro synÈhesis of various ribosomal- proteins, protein elongation factors and RNA polymerase subunit ô(, vrere also inhibited by ppcpp (Chu et. aI . 1976; Lindahl et. aI. 1976') . The in v.ivo synthesis of protein elongation factors, at 1east, was previously shown to be under stringent control (Furano and llittel I976). Data support the concfusion that these physiological blocks by ppcpp occur at the leve1 of initiation of transcription (Lindahl et. al. 1976). In contrast to the nurnerous in vitro inhibitory effects of ppcpp and pppcpp, one of these, ppGpp, has a novel feature in that it was found to stirnulate the in vitro transcription of lactose, arabinose, tryptophen and histidine operons of E. coli anC salmonelLa typhimurium (Yang et. aI. L974; Stephens et. aI. 1975). These in vitro observations suggest that ppGpp and pppGpp affect bacterial metabolisrn in cornplex ways, particularly the transcription process. It has been proposed that ppcpp functions as a general- transcriptional effector (Reiness et. aI. 1975) through 37

j-ts binding to the ÐNA-dependent-RNA polyrnerase holoenzyme. Depending on hov, the ppcpp-holoenzl¡me complex interacts with specific aene/operon promotor sites, ppGpp should have a positive, negatj-ve or no effect on the transcription of the gene or operon in question (yang et. aI. I974r. Through this mechanism, it is conceivable that ppcpp and pppcpp are able to coordinate some of the cell-ular processes so as to provide optimal survival conditions during rapid changes of the nutritional- state in bacteria. The isolation and characteri zation of RNA polynerase mutants and the sequencing of specific genetic promotor sites might lead to a better understanding of this intriguing phenomenon.

(v) Influence of Carbon and Nitrogen Sources on ppcpp, pppcpp and rRNA Metabolism

In contrast to the 'stringent responser due to amino acid deprivaÈion, bacterial- cells undergo a differently program¡ned metabolic tesponse upon a shift from either a rich complex medium to a glucose minirnal- one or from a rich carbon source to a poorer one. What is unique is that both relaxed and stringent strains h'ere able to shut-off RNA accumuLation to varying degrees with the onset of such a nutritional shift 38

(Maaloe and KjeLdgaard 1966, Lazzaríni and lfins.low 1970). Concommitantly, both bacterial strains were able to accumulate ppGpp (Lazzarini et. aI . L971; Harsh¡nan and Yamazaki 1971r Winslow l97I). The intraceltular levels of ppcpp in both relA+and Ig_lA-. downshifted cel1s, however, were only about 25-509 of those àttained by amino acid starved bacteria lLazzarini et. al. I971, Harshman and Yamazaki J.971; Vtinslovr 1971). pppcpp ccncentrations remained either at basal ]evels or decreased sJ-ightly during stepdohrn conditions (Harshman and Yamazaki 1971), a response similar to that expressed by amino acid starved spoT mutants (Laff1er and Gal-Iant L974). The depressíon in rRNA aicumutation in downshifted cel-ls is the resul_t of a d.ecrease in rRNA synthesis and an increase in rRNA degradation (Erlich 1,972; Norris and Koch 1,972). ?he rol-e of ppGpp in inhibiting rRNA synthesis is clear (Reiness et. aI. 1975; van Ooyen et. aI. 1976). fn studies where rRNA levets were found to decrease specifically by degradation during downshift, no ppcpp was produced (Lazzarini et. al. 1969; Erl-ich et. a.I. I975't. Thus during downshift, ppGpp specifically affects rRNA synthesis whil-e another mechanism exists for the preferential degradation of new].y made rRNÀ 39

(Lazzarini et. al-. 1969). Gallant and Lazzarini (1976) have hypothesized that specific rRNA nucleases subject to control, possibly by yet to be discovered nucleotides, may be involved in this rRNA degradative process.

(e) HPN Compounds

The regulation of bacteriaf sporulation has been an intrÍguing probl_em. for research scientists. Using Bacillus megaterj-um, Elmerich and Aubert (1973) found Èhat sporulation by certain mutants with lesions in the purine biosynthetic pathr^ray was derepressed when grown in a glucose-ammonia containing medium. They suggesÈed that certain intêrmediates in the de novo purine biosynthetic pathvray are j-nvolved in bacterial sporulation, possibl_y by mediating the synthesis of either a repressing compound or inhibiting the production of an inducing compound. Rhaese and coworkers (1972, Ig76) r,rere the first to report the existence of four acid extractable, high]y phosphorylated compounds, HpN I, II , III and IV in the sporulating bacte,rium, Bacillus subtilis. HPN f, If, III and IV were subsequently reporÈed to be adenosine 3',5'-diphosphate (ppApp), adenosine 3'_ 40

diphosphate 5t-Èriphosphate (pppApp), uridine 3 ' - monophosphate 5 | -monophosphate- Z-diphosphate (ppzpup, r.,rheïe Z is an unknown sugar) and adenosine 3',5'-tríphosphate (pppAppp) respectively (Rhaese et. aL. 1977). HpN I and fV have al,so been reported to be synthesized ín vitro using isolated bacteriaÌ membrane vesicles with ATp as a substrate (Rhaese and croscurth 1976). These and other studies by Rhaese et. al . (1976; 19771 led the authors to suggest the possible involvement of some of these HPN compounds in bacterial sporulation. When B. subtilis attains the stationary gror,¡¡th phase, presumably due to the depLetion of essential growth nutri-ents, or rdas starved for nitrogen and phosphate, sporulation would inevitably occur. At the onset of sporulation, HPN III and IV accumuLated rapídly within the cell. A mutant srhich l4'as impaired in HpN fV synthesis was found to be asporogenous. Interestingly, a revertant of this mutant regained both the ability to sporulaÈe and synthêsize HPN IV (Rhaese et. al. 1977). But the mechanism(s) by which these iIPN compouncls regulate sporogenesis remains to be elucidated. It is nob, pertinent to point out that the physiology of B. subtilis sporulation, particularly 4I

\dith regard to the tj-me of intracellular accumulation of HPN fIl and IV and the nutritional growth conditions required for inducj-ng their synthesis exhibits a similarity to the involvement of HS compounds in the sporulation process of the eucaryote, Ach1ya (see part (i) of this section). WheÈher these tÌ4ro sets of hiqhly phosphorylated compounds are similar, both in structure and function(s), awaits more rigorous chemi.cal and biochemical studies.

(f) Diguanosine PoLyphosphates

Numerous pol-yphosphorylated nucl-eosides have been found in various eucaryotes. The earl-iest of Èhese,

e1-e4 diguanosine 5 ' - tetraphosphate (GenG) and p1-p3 ai- guanosine 5'-triphosphate (cerG), r^rere found in high concentrations in the encysÈed embryo of the brine shrímp, Artemia salina and in Daphnia gg¡1¿lg (Finamore and i,¡arner 1963t Vùarner and Fj-namore 1965; Oikawa and Smith 1966). Clegg and coworkers (1967 ) suggested that Gp4G may be usêd as a source of guanine and adenine nucleotides for nucleic acid biosynthesis during the development of A. salina from the embryonic to laval stage. This period is ¡narked by an absence of 42

de novo purine biosynthesis (CIe99 et. al. 1967 ). Recently Renart et. a1. (1976) showed that GpnG is a potent activator of GMP reductase from A. salína . This may bè physiologically significant as this enzyme coulil be a major pathr"Tay through which cMP derived from cp4c hydrolysis is converted to ÃMP (Renart et.al. 7976).

(s) Diadenosine Polyphosphates

A compound similar to Gp4c, diadenosine 5',5"'- 1, P P4-t.tr.phosphate (Ap4A) has been detected in numerous mam¡nalian cell- lines (Rapaport and Zamecnik 1976'). The obsêrvation of the inverse relationship betweên Ap4A levels and the doubJ-ing times of cefls led Rapaport and zamecnik (1976) to propose that this molecule may act as a positive 'pleiotypic mediator' as contrasted with the negative 'pleiotypic mediator' role of cAMP (Kram et. al. 1973). .A recent report by Gn¡nmt (1978) demonstrated that fairly high concentrations of Ap4A could stimulate DNA synthesis in G, arrested but not exponentially growing, permeabilized, baby hamster kidney cel1s. !ùhether this molecule acts as a storage form of ATP or as a grovrth regulator remains 43

to be elucidated (Rapaport and Zamecnik I976).

(h) Triadenosine Pentaphosphate

Since the discovery of interferon in 1957 (see review by Friedman 1977), littLe is knorvn about the mechanism(s) involved in its potent antivira.I effect. But with thê discovery of a low molecular weight trinucleotide, produced in cells synthesÍsing interferon, that can inhibit cel-I free protein synthesis, the underlying mechanj-sm of interferon action may soon be better understood (Roberts et. al. 1976') . This protein synthesis inhibitor has since been characterized as pppA2rp5'42'p5'A and given the pseudonym rtwo-five Ar (Kerr and Brown 1978). The data availabl-e suggest that one action of interferon is to induce the biosynthesis of the enzyme that is responsibì-e for synthesising this inhibitor mo1ecule. Double stranded RNA then activates this trinucleotide rsyntheÈase' to produce rtr'/o-f ive A' using ATP as the substrate. Its been postulated that tlìis rtwo-fíve A' mol-ecule in turn probabJ.y activates a nuclease that degrades cellular nRNA, thereby effectively stopping protein synthesis and thus viral production. The specificity of the activated nuclease remains an interesting question (Roberts et. al . I976; Sen et. al-. 1976; 44

Hovanessian et. al-. ]-977 ¡ Kerr and Brown I97B) - Further studies on the in vitro and ín vivo action of this unique trinucleotide may provide a more detailed insight into the compl-ex induction by interferon of the antiviral state ín mammalian celts (Friedman 1977).

(i) Dinucleoside PolyphosphaÈes (HS)

Three unique pol,yphosphorylated nucleosides have been isolated from whole-celL acid extracts of a variety of fungi (T,éJohn et. a1. l-975). These compound.s may also be isolated by osmotic shock tteatment of whole ceLl-s, thus suggestj-ng that they may be l-ocated on the cell membrane (LéJohn et. at. 1978) . One of these three compounds, HS3, was subsequently shoh'n to consist of a glutamyl-ADP-sugar moiety which is covalentì.y linked to a UDP-sugar- tetraphosphate

(McNaughton et. 41. 1978). The other thro, HS1 and HS2, which have been partiatly characterised appear tc have tr.i¡o uridines each r^rith tvrelve and. ten phosphates respectively in contrast to the eight present in HS3 (McNaughton et. al. 1978). Ðata have been presented by LéJohn et. aI. (1978) implicating these HS molecules as reguLators of sporuLation and nucleic acid metabolism. For exanpLe, it was observed that a rapid intracellular accumulation of all threê HS compounds occurred just prior to sporuJ-ation. This accelerated accumulation of HS molecules by Achl-ya could be triggered either by the depletíon of phosphate in the grov¡th medium o¡ transferring the ce11s from a rich growth medium to one devoid of nutrients. These nutritional_ 'shocks, invariably led to sporulation. Hohrever, when the étarvation medium, for example, was suppleme¡ted with glutamine or various purine and pyrimidine bases and nucleosides, HS acculnulation and sporulation vrere concomitantly aborted. It hras therefore suggested (Lé.Iohn et. al. 1978) that the intracellular accumulation of alL HS molecutes may be a prerequisite for the sporuLation process. The reLationship betrdeen HS compounds and nucleic acid metabolism during the growth cycle of Achlya is clear. An inveJrse relationship was observed !ùhen the rates of RNA and DNA synthesis were compared to HS synthesis (LéJohn et. aL. I975; LéJohn et. aI. I97B). ?his suggested that these HS molecules may be acting as negative effectors of RNA and DNA synthesis in vivo. fn vitro experiments demonstrated that HS3 and HS2, 46 at physiological concentrations, were equafly potent inhibitors of isol-ated Achtya DNA-dependent RNA polymerases (McNaughton et. al. 1975; LêJohn et. a1. 1978). Lewis et. at (1977) have also shown that HS3 and Hs2 are strong inhibitors of the isolaÈed Achl-ya ribonucleotide reductase, an enzyme involved in the conversion of ribonucleoside diphosphates to their respective deoxyribose forms (Larsson and Reichard 1966). The possible involvement of these HS compounds in regulati.ng fungal energy-linked transport systems have also been postulated (stevenson and LéJohn 1978; Goh and !éJohn 1978). It is thus likely that these polyphosphorylated dinucleosides indeed play important regulatory rol-es in the growth and development of Achlya. More recenÈly, one of the fungal HS compounds, HS3, was also detected in numerous cuLtured mammalian ceLl lines (Goh and LêJohn ]-9771 . PhysiologicaL studies clearly demonstrated that HS3 accumulated onfy rdhen glutamine h¡as removed from the normal" growth medium as isoleucine deprivation was vrithout effect (Goh and LêJohn !977 ¡ Go}j. et. al-. 197'11 . or,¡ing to the very important role of glutamine ín de novo purine biosynthesis (Uahler and Cordes I97I\ , studies vrere 47 performed using various drugs which affect de novo purine biosynthesis and with mammalian ceff mutants defective in de novo and salvage pathr,rays for purine nucl-eotide biosynthesis. The results demonstrated that HS3 accumulation may be due to a l-ack of either precursors for purine nucleotide biosynthesis or of purine nucleotides themselves (Goh and LéJohn 1977¡ Go}] et. al . 1977). For example a CHO GAT- mutant, auxotrophic for glycine, adenosine and thymidine, owing to its defective fofate metabolism (I"lcBurney and !ùhitemore I974\ accumulated HS3 only when it was deprived of adenosine. Further more, azaserine and methotrexate' which are analogues of glutamine and foLic acid respectively, and which block de novo purine biosynthesis were active in causing the accumulation of HS3 (Goh and léJohn 19771 . AIso, HS3 that accumulated during glutamine deprivation was depleÈed by all exogenously supplied purine and pyrimidj-ne bases and ribosides; the only exception being thymidine (Goh and LéJohn 19771. During glutamine starvation of cHO wT and cHo GAT- cells, it was observed that DNA and RNA synthesis r^rere inhibited in vivo (Goh and LéJohn 19771 . Studies by Lewis et. al. (1977) showed that both Achlya and mammal-ian cell HS3 r^'ere potent inhibitors of ribonucleotide reductases 48

frcm CHO cells. ft was subseguently suggested (Goh et. aI. 19771 that both DNA and RNA synthesis in mammalian cells may be regulated in a similar fashion as in Achlya (McNaughton et. al. f975; Lewis et. al . I976), by HS3 nolecules. The relevant details of these studies of HS3 metabolism in mammalian cells witl be presented in the 'Resufts' section of this thesis.

(j) Other Polyphosphory 1a ted Nuclesides

Other unusual phosphorylated nucleosides with possible physiological functions GTP-l-ike .include the 'phantom spot' (Gal-Lant 1976) which is produced by E. coli in response to an energy source do\,/nshift. The authors proposed that this molecule may mediate the accumulation of RNA. Tr,\¡o other novel nucleotides, designated DSI and DSII, r4rere recently found in E. coli (Loêwen 1976). Their physiological function (s) remains to be e lucidated. 49

(4) MaÍmalian Purine and Pyrimidine Metabolism

Since the discovery of uric acíd by Scheele in 1776 and the subseguent isolation of nucleic acids by

Miescher and Kossell (tfyngaarden and Kel1ey I972) , purine and pyri.midine metabol_ism have become fruitful areas for biochemical research. The de novo biosynthetic pathways of purines and pyrimidines are illustrated by figures 1 and 2 respectively. The regulation of purine and pyrimidine biosynthesis in mammalian ce1ls and tissues is complex. Most of the studies which have been carried out report the biochemical activities of partiafly purified enzymes. To extrapolate these findings to their functions in intact ce1J.s requires great caution. The pitfalls associated with such ín vj,tro sÈudíes have been discussed (Henderson et. al-. 19771 . Also nêw enzymes rel,ated to purine and pyrimidine metabolism are stil-l- being reported. The following discussion ís concerned with those enzymes of the purine and pyrimidine biosynthetic pathways which studies have suggested to be possible important sites of regulation.

(a) Regulation of De Novo Purine Biosynthetic Enzymes in Mammalian Cells and Tissues 50

Figure (1)

ATp + Ribose- 5-pho sphate I 5 -Pho sphoribo sy*l- 1-pyrophosphate :

5 - pho sptroii i uo sy t am íme

I Glycinamide Ribonucleot j-de

N-Forlnylglycinamide'l' Ribonucleotide

I N-Formylglycinamidine Ribonucleotide

I 5-Àminoimi da zole Ribonucleotide

I 5-Amino- 4-Imidazol-ecarboYylicacíd RibonucLeotide

I 5-A¡nino- 4- ImidazoLe-N-suclinocarboxamide Ribonucleotide I ¿ 5-Amino-4-Imidazolecarboxamide Ribonucleotide I 5-Formamido- 4- Imida zole Carboxa¡nide Ribonucleotide *I II4P

Adeny Losucc ina te Xanthosinê llonophosphate I J i AMP GI,lP q.1

Figure (2)

L-Glutamine + 2ATp + HCO3 ! v Carbamyl Phosphate

I v Carbamyl Aspartate

II Dihydroorotic Acid

úI OroÈic .Acid

I ¿ Orotidine-5 | -Monophosphate vI Uridine- 5 | -Monophosphate 52

(i) 5-Phosphoribosyl-l--pyrophosphate Synthetase (pRpp Synthetase) (EC 2.7.6.1) and Intraceflular 5-Phcsphoribosyl- 1-pyrophosphate Levels

PRPP syntheÈase, which supplies the necessary ceLlular PRPP for de novo purine and pyrimidine biosynthesis and purine and pyrimidine salvage reactions (Mahler and Cordes l-971) plays a pivotaÌ role in purine and also pyrimidine metabolism. PRPP synthetase catalyses the production of pRpp by the following reaction: Pi/I4g-)+ ribose- s-phosphate + ATp , pRpp + AIvp The activity of PRPP synthetase is reguJ.ated in a complex manner by magnesium, inorganic phosphate, ATp,

ribose- s-phosphate , 2 , 3-díphosphoglycerate and end-product nucleotides. The enzyme has an absolute reguiremenÈ for inorganic phosphate, its alLosteric activaÈor (Fox and Kel-Iey I9721 . This is consistent with the observatj-on that under physiologicaL pi concentrations, leveLs of pRpp in human erythrocytes hrere diminished in the presence of purine nucleosides or^ring to the Iowering of intracellular Pi by the nucleosides (Planet and Fox 1976') . The nhrlich ascites tumour ce11's pRpp synthetase, r^rith Mg-ÀTp being its true substrate, r^'as strongly activated by magnesi'.rm 53

(Mu¡ray and l{ong ),967) . pRpp and 2 , 3_diphosphoglycerate were found to be competitive inhibitors (Murray and Wong f967r Fox and KelJ_y l9Z2; Hershko et. a1. 1969) while the end-producÈ puïine nucfeotides $rere non_competitive inhibitors wiÈh the nucl_eoside di- and triphosphates being more potent than the nucÌeoside monophosphates (Fox and Ketly 1972). These nucleotides regulate partialty purified PRpp synthetase by a mechanism described by Fox and Kelley (1972) as 'heterogenous metabolic pool inhibition' . The degree of inhibiton being dependent on the absoLute concentration of each of the purÍne nucl-eotides in the total ceÌIular nucleotide poo1. This mechanism of inhibition has aLso.been observed in vivo (Bagnara et. a1. 197 4) . Certain pRpp synthetase mutant celIs associated with an accel-erated rate in pRpp and purine synthesis have recently been described (Becker et. al. I973¡ Zoref et. al. 1975). A similar condition exists in tissues and cell-s deficient in the enzyme hypoxanthine_guanine phosphoribosyltransferase (HGpRT) (EC 2.4.2.g) (Seegniller et. aI. L967; RosenbLoom et. aJ.. 1968). Studies using an HcPRT-deficient cuLtured rat hepatoma celì line suggest that the cause for purine overproduction appears to be the result of an elevated pRpp synthetase activity rather 54

that a PRPP 'sparing, effecÈ steming from reduced HGPRT activity (craf et. a1. 1976). Other factors are also known to affect pRpp 1eve]s.

When human lymphoblasts hrere deprived of glutamine, an essential precursor for purine biosynthesis (Mahler and Cordes J.97Il , intracellular pRpp level,s feLÌ dramatically (Skaper et. al-. 1976). The authors suggested that this phenomenon may be the result of a Lack of a substrate for PRPP synthetase. Though pyrimidines are not known to inhibit eucaryotic pRpp biosynthesis (I^Iood and Seegmiller Ig73), purine biosynÈhesis in human fibroblast was inhibited by exogenous addition of orotic acid. The data suggest that this inhibitory effect of ototic acid is most l-ikely reLated to a deptetion of intraceltuLaï pRpp

(Kelley et. aI. 197 O) .

(ii) 5-Phosphoribosyl--l-pyrophosphate Amidotransferase (Êc 2.4.2.14Ì,

The first enzymatic reaction in nearJ.y alI metabolic pathways is generaJ.J.y an irnportant regul-atory protein (Wyngaarden 19721 and pRpp amidotran sferase , the first enzyme in the de novo purine biosyntheÈic pathr,ray is no exception. This enzyme catalyses the synthesis of phosphoribosylamine (PRÀ) by the fofLowing reaction:

PRPP + clutamine "'o , phosphoribosyl amine + ug2* clutamate + ppi DDi ?Dì

--- Studies with both partially purified enzymes and intact mammalian cel1s suggest that the overaÌI rate of de novo purine synthesis is dependent on the levels of PRA (Henderson 1972) - Two other enzyme activities capable of catalyzing the synthesis of PRÃ have also been detected. The first of these, known as PRPP aminotransferase , utilizes ammonia rather than glutamine as the substrate (Reem J.972). whether this activity represents. a distinct new protein or is a subunit of PRPP amidotrans ferase remains unknown. The physiological importance of the second. enzyme activity, ri-bose- 5-phosphate aminotransfera se r,rhich has ribose-5- phosphate, ATP and ammonia as substrates for pR.A synthesis has yet to be shown in vivo (Reem 1968).

The glutamine-dependent PRPP amidotran s fe ra s e reguires Mg2+ or Mrr2+ a cofactor and is stabilized by "s high Pi leve1s (Rowe and lfyngaarden 1968). This enzyme from avian livers has a molecular weight (MW) of about 210,000 and threfve atoms of non-haem iron (Caskey et. al 1964;

Hartman 1963) . The 2l-0,000 MVJ component from pigeon liver in 56

turn consists of four identical subunits. rt is dissociable into two 100,000 MW species by dilution of the enzyme preparation and into the 50,000 MV,¡ units by thíol reagent treatment. In the presence of Alvlp and GMp, the enzyme is

converted into the inactive 1ô0,000 MW species ( Rovre and Wyngaarden 1968) while high concentratíons of pRpp wí11 Ìeverse this process (rtoh et. al . Ig76). In contrast, the PRPP arnidotrans ferase from human placenta exists in two MW forms of 133,000 and 270,000 (HoLmes et. aI. 1973b). The enzyme activity correlates directly with the amount of it present in the 133,000 MW form, suggesting that this smaller enzyme species is the acÈive form (HoLmes et. a1. 1973b). Thus purine ribonucleotides which pRpp inhibit the human amidotran s ferase converts it into the large MW form while pRpp reverses the process (Hol_mes et. al. 1973b). Though the human and avian pRpp amidotrans ferase show rather distinct strucÈural and cataJ-ytic differences, both are equally sensitive to inhibiÈion by the end-products of the purine biosynthetic pathbray i.e. IMp, AMp and GMp. This phenomenon is very specific as purine 2r and 3 '- ribonucleotides , | 5 - deoxyribonucleotides , ribonuc leos ides, free bases and pyrimidine compounds were ineffective (Vlyngaarden and Ashton Ì959,. Holnes et. aI. L973a). 57

(iii) Regulation of AMp and cMp Synthesis from IMp

IMP can be derived from the de novo purine biosynthetic pathway or through the salvage of hypoxanthine by HcpRT (Mahler and Cordes l97I). fn vitro studies have revealed that the intracellular concentratlon of IMP is very low ín mammalian tissues and maybe limiting for adenylosuccinate synthetase (EC 6.3.4.4.) and IMP dehydrogenase (EC 1.3.1.14) (van der I,ùeyden and

Kelley 1974; Holmes et. al. 1974). Using cuLtured human lymphoblasts, it l4râs observed. that the de novo fMp branch is coordinately regulated (Hershfield and Seegmiller 1,976) . For example if. guanine is added exogenously to the growth medium, the synthesis of GMp is selectively inhibited while the other branch leading tp A.t'lP synthesÍs is stimulated. Exogenously added ad.enine produced a similar but complementary effecÈ. The effects of exogenous adenine and guanine on the util-ization of de novo derived IMp for AMp and cMp synthesis can be explained by a simple selective feedback inhibition of one branch by an end product and the sinultaneous stimulation of the other branch pathway owing to the increased availabitity of Èhe ' competition-I imi ted' substrate, Ilvtp (Hershfield and Seegrniller Ig76l . 58

(iv)

purine Biosynthetíc pathway Enzymes

Data on the inductioltr rêpïession and derepression of the enzymes of the de novo purir,å biosynthetic pathü/ay in mammalian cell_s remain very sketchy. McFall and Magasanik (1960) reported that enzymes of the entire purine biosynthetic pathway may be repressed j-n mouse ¡ celi-s which have been cu1Èured for several- generations in the presence of adenine or guanosine, Appârently under such conditions all the purines of the intracellular soLub]-e pools were derived exclusively from the exogenousLy supplied purine. In whole animal. studies (Reem and Friend 1967) , pRpp anidotrans ferase activity, which is normally absent in mouse spl-een, appeared four days following infection with Friend leukemia virus. The rise in activity peaked after six to nine days followed by a gradual decLine. The mechanism of induction of the pRpp amidotransferase acÈivity by the virus is unknown. 59

(b) Regulation of De Novo pyrimidine Biosynthesis in Ma¡Enalian Cells and Tissues

The regul-ation of de novo pyrimidine biosynthesis in mammalian cell-s is both compLex, and at times ambiguous. The first reaction that is unique to the de novo pyrimidine biosythetic pathway is the synthesis of carbamyl aspartate; the reaction is catalysed by aspartate transcarbamyLase (EC 2-!.3.2). Mammalian aspartate transcarbamylase, hor,ùever, has always been copurified with two other enzymic activities, carbamyL phosphate synthetase (EC 2.7.2.9) and dihydroorotase (EC 3.5.2.3) which cataLyse the syntheses of carbamyl phosphate and dihydroortic acid respectively (Mori and Tatibana I975i Kempe et. aL. 7976; Ito and Uchino 1972). A recent study has shown that all three enzymes from SV40 transformed hamster ce1ls are covatently linked as a multifunctional high nolecular vreight protein (Coleman et. aL. Ig77). Persuasive data suggest that carbamyl phosphate synthetase may be an important regulatory enzyme because it has been shown to catalyse a rate-Limiting step for pyrimidine biosynthesis (Hager and Jones 1967a¡ Hager and Jones L967b). This enzyme is allosterically feedback 60

inhibited by UTp which ís an end-product of the pyrimidine pathvray (Levine et. al. I97L). Aspartate trânscarbamylase , on the other hand is insensitive to pyrimidine nucleotide feedback inhibition (Smith er. al . 1972) though pyrimidine nucleosides are potent inhibitors (Bresnick 1963) . Of interest is the observation that purine deoxyribonucleos ides and deoxyr5.bonucleotides strongly inhibited aspartate transcarbamylase activity (Bresnick 1962), suggesting that purine derivatives may be directl_y involved in regulating de novo pyrimidine synthesis. The last enzyme of the multifunctional enzyme complex, dihydroorotase, appears to be regulated, in vitro, Like aspartate transcarbamylase (Coleman et. al. I977) with respect to its inhibition by pyrimidine and purine de¡ivatives (Bresnick and Blatchford 1964). vùhether the activj_ties of calbamyl phosphate synthetase, aspartate transcarbamylase and dihydroorotase are coordinately regulated by end- products of de novo pyrimidine synthesis remains unclear (Bresnick and Blatchford 1964) though the biosynthesis of the three enzymes of this enzyme complex are coordinately regulated (Kempe et. a1. I976). Of the other enzymes involved in mam¡nalian de novo pyrimidine biosynthesis, orotate phospho- rÍbosyl trans ferase (EC 2.4.2.J_0) may be another key control point because firstly , this enzyme utilizes 6t

PRPP, a common intermediate required for both purine and pyrimidine biosynthesis (Mahler and Cordes 1921) and secondly it appears to be rate-liniting in Ehrlich ascites cells (Shoaf and Jones 1973). Recent sìu¿ies have shown that both adenine and adenosine are Èoxic to cul-Èured rnammalian cells (Green and Chan 1973; Ishii and Green 1973). This phenomenon may be the resul-t of a block of pyrirnidine biosynthesis at the level of orotate phosphoribosyltrans ferase since orotate \^ras found to accumulate under these conditions (Ishii and Green 1973). This may be the direct result of a reduced availability of PRPP for OMp biosynthesis owing to competiti.on fo¡ this metabolite by the adenine salvage pathway. Älternatively, the resulting excessive accumul-ation of adenine deoxynucleosides and deoxynucleotides would inhibit the aspartate transcarbamyJ-ase activity (Bresnick

7962). Hoogenraad and Lee (1974) reported that <1e novo pyrimidine synthesis j.n rat hepatoma celì-s grown in continuous culture was inhibited by the addition of 0.5 mM uridine to the growth medium. The activity of orotate phosphoribosyl Èransferase decreased while the other enzymes for de novo pyrirnidine biosynthesis were unaffected (Hoogenraad and Lee Ig74,). Because uridine, UMP and UTP did not affect the in vitro activity of 62

orotate phosphoribosyltransferase, this translation and transcription dependent inhibition of this enzlzme may be due to a depl,etion of pRpp by excessive saLvage of uracil from the breakdown of uridine. The actual- mechanism of inhibition however remains obscure

(Hoogenraad and Lee I97 4) . The last enzyme for de novo UMp biosynthesis is orotidine- 5 ' -phosphaÈe decarboxylase (EC 4.1.1.23). This enzyme which has been detected in nurnerous rnammalian tissues and cells (Smith eÈ. al. 1972) is subjected to end-product inhibition by UMp and to a lesser degree by other purine and pyrimidine nucleotides (appel 196g). UMP pl-ays a central role in pyrimidine metaboJ.ism, particularly in its conversion to the other tr"ro pyrimidines, cytosine and thymine (Mahler and Cordes 1971) Like de novo purine biosynthesis, very littte is known about the regulation of the synthesis of the enzynes of the de novo pyrinidine biosynthetic pathway in mammalain tissues and ce1ts. Besides the known coordinated regulation of the synthesis of the multifunctional enzyme complex described previously (Kempe et. al. 1976) , it is now knok¡n that reticulocytosis leads to a compl-ete loss of dihydroorotaÈe dehydrogenase in the mature erythrocytes (Snith et. al_. J-g72). 63

(s) urd Pyrimidine Metabolism

(a) Gout

Numerous human gênetic diseases are directly reLated to defects in the anabolism and caÈabotism of purines and pyrimidines. The classical of these is gout which is characterized by the deposition of sodium uïate crystaLs in various limb joints. This dísease is associated with an overproduction of purines leading to a hyperuricaemic state (Wyngaarden and Kelley lg:.2). For example, some patients have been reported to possess mutant 5-phosphoribosyl-l-pyrophosphate (pRpp) synthetases vüith either an increase in specific activity (Becker et. a1. 1973) or a lack of feed-back resistance (zoret et. al. 1975). Others lack various purine salvage enzymes leading to an accumulation of pRpp (Rosenbloom et. â1. 1968). The end result of these enzymic defects is an accel-eration of de novo purine biosynthesis and an excessive production of uric acid. 64

(b) Lesch-Nyhan Syndrome

A fairly rare human disorder associated vrith severe neurological abnormalities is the Lesch-Nyhan syndrome (Lesch and Nyhan 1964). This x-linked disease is related to a total- or nearly total deficiency of hypoxanthine-guanine phosphoribosyl-transferase (HGpRT)

activity in various tissues and is accompanied by an enhancement of de novo purine synthesis (Rosenbloom et. aI. 1968; Seegrniller et. a1. 1967). The imporÈance of HGPRT in human cel-l-ular metabol-ism was better appreciated after Adams and Harkness (1976) reported that there is a significant increase in HGPRT activities in the normal human cerebral cortex and medulla after the age of two years. This seems to correlate well with the chronological observation that in Lesch-Nyhan syndrome patients, a rapid deterioration in the behavioural and neurological abnormalities occur after this a9e (Nyhan 1973). But the biochemical relationship(s) between HGPRT deficiency and the neurological symptoms remains obscure. 65

(c) fmmun ode fi ci encv Syndromes

Two other human genetic disorders known as severe combined immunodefíciency diseases are associated with nucLeic acid metabolism. Giblett et. al. (1972) fírst reported absence of adencsine deaminase (ADA, EC 3.5.4.4), which is involved in the conversion of adenosine to inosíne and ammonia, in t¡ro young patients with T_cel1 and B-ce11 dysfunction. Subsequent studies have established at least tv¡o dozen famil_ies known to be afflicted with this disease (Hirschhorn et. al-. 1975; Meuvrissen and Pollara 1974). Becausê adenosine is toxic to cul_tured lymphoid ce11s (Green and Chan 1973) and because patients rtith AD.A syndrome have ê.l-evated levels of ad.enine, adenosine and adenosine nucleotides (Mil1s et. al . ]-976), it was thought that these purines are probably responsible for the severe lymphocytopenia observed in these patients. Carson et. aL. (1977) recentLy, hoh,ever, reported the presence of high activities of ADA and adenosine kinase that are specific for deoxyadenos ine,, deoxyinosine and deoxyguanosíne in the lymphocytes of newbcrn human tissues. Sirnmonds et. aL. (192g) then shovred that deoxyadenosine $ras far more toxic Èo mitogen_ 66

stimulated lymphocytes than adenosine. Contrary to the study by Mi1ls et. af. (1976), thro research groups have nohr reported that there is a substantial elevation of deoxyadenosine nucl-eotides and a decrëase in ATP l-eveIs in the erythrocytes of two patients wíth severe combined immunode fj-ciency associated vrith a lack of ADA activity

(Cohen et. al, 1978; Coleman et. al . 1978) . It is nord thought (Cohen et. al. 1978; Coleman et. al. 1978) that severe combined immunodeficiency disease may be due to the inhibition of DNA synthesis and hence mitogenesis of precursor T- and B-ce]1s through the inhibition of ribonucl-eotide reductase by elevated levels of deoxyadenosine triphosphate (Moore and Hurlbert 1966). Whether elevated levels of deoxyATP also persists j-n the lyrnphocytes of these immunodeficient patients remains to be elucidated. IJo\,Jever the enormous difficulty in obtaining sufficient lymphocytes for such studies from these patients is a drawback. This disease, if detected very early in infancy, may ultimateLy be amenable to treatment.

The other im¡nunode ficiency disease, vüith a concommitant absence of purine nucleoside phosphorylase (EC 2.4.2-1), was first reported in 1975 (GibLett et. al. 1975). The disease is characterized by an impairment in T-cell immune functions (Giblett et. al 1975). Purine 67

nucleoside phosphorylase, which catalyses the reversible

conversion of guanosine, deoxyguanosine, inosine and. deoxyinosine to their respective bases (parks and Agarwal, 1972), is generally believed to function in vivo in the direction of nucÌeoside breakdown. This is a reasonabfe assumption because with the exception of deoxyguanosine (Anderson J,973't, there are no known mammaLían nucleoside kinases capable of convêrting guanosine, inosine and deoxyinosine to their respective nucl,eotides (Milman et. at. 1976). The four substrates of purine nucleoside phosphorylase have been found to be elevated in the body fLuids of patients which lack this enzyme (Cohen et. aI. 1976). DeoxyGTP moreover, has recently been observed in the erythrocytes of purine nucleoside phosphorylase- deficienÈ, immunodeficient children, but not in normal controls (Cohen et. al. 1978). Studies show that of the four substrates of purine nucl-eoside phosphorylase, deoxyguanosine is the most toxic to mouse T-celL lymphoma celL lines (Chan 1978; cudas et. al-. 1978). The cytotoxic effect of deoxyguanosine was dependent on it being phosphorylated to deoxycTP by deoxycytidine kinase (Gudas et. al. l-978). Furthermore, the extractabl-e pool of deoxycTP was markedly depressed in the presence of deoxyguanosine and its cytotoxicity could be reversed by the simultaneous addition of deoxycytj-dine and hypoxanÈhine to the growth medium (Chan l-97g) . Owing to these findings, it has been suggested (Chan I97g; Gudas et. al. 1978) that the T-cel1 dysfunction in purj-ne nucLeoside phosphorylase deficient patients may be the result of an excessive accumutation of deoxycTp in the lymphoid tissues of these patients. This deoxynucLeotide is a known inhibitor of cyÈidine diphosphate reduction to dCOP (Moore and Hurlbert 1966) and de novo purine biosynthesis (Chan 1979). This conclusion of deoxycTp toxicity is supported further by the observation that J-lznphoid tissues have the highest deoxycytidine kinase activities in the body and the greatest ability to phosphorylate deoxyguanosine (Durham and fves 1969; Carson et. aI. L977). The reLevance of these findings of an in vitro animal lymphoma model must await simitar studies r,ri th human tissues. ïn contrast to the combj-ned immunodeficiency disease associated !ùith an absence of adenosine deaminase activity (Giblett et. al. Ig72) , several cases of dominantly transmitted hereditary hemolytic anemia with elevated l-eve1s of erythrocyte adênosine deamÍnase has now been reported (Valentine et. a1. ]-g77). This disease is associated r,rith decreased ATp levels in the red bl"ood 69

cells presumabty due to the result of the high adenosine deaminase activity (Valentine et. af. 1977). This hemolyÈic phenomenon may thus be directly rer.ated to the lack of ATp for maj.ntainíng cellufar metabolism in the non-nucleated erythrocyte which are incapable of de novo purine biosynthesis (Valentine et. aL. :-g77).

(d) Oroticaciduria and Xanthinuria

Two other human hereditary disorders directly related to defective purine and pyrimidine metabolism are xanthinuria and oroticaciduria, Xanthinuria Ís characterized by a deficiency of xanthine oxidase activity, The resuLting high level-s of serum xanthine and hypoxanthine lead to urinary xanthine stones formation and occasional myopathy (lVyngaarden L972). Oroticaciduria is associaÈed wlth an inability to convert orotic acid t.o UMp due to a lack of orotidine 5 ,_ phosphate decarboxyl-ase activity in alt patients vrhile most oÈhers also have a defective orotate phosphoribosyltransferase (Zellner et. aI. l_976) thus suggesting tr"ro forms of the syndrome. patients suffering from oroticaciduria are characterized by retarded grordth and development, anaemj-a and excessive orotic acid 70

excretion (Smith et. aI. lg72). Replacement therapy using eíther uridine or pyrimidine nucleotides has helped alleviate the pathological symptoms (Smith et. a1. I97 2)

The biochemical- mechanisms of some of these human genetic diseases related to aberrant purine and pyrimidine metabolism remain obscure. A better understanding of the regulation of purine and pyrimidine biosynthesis and the development of more precise methods for guanÈitating nucreobases, nucleosides and nucr-eotides will be necessary to delineate the biotogy of such human diseases. MATERIALS AIiD METHODS 72

MATERIÀLS

(1) organisms

Mammalian Cell Lines

CHO WT, CHO cÀT , mouse 3T3 fibrobl-asts and the CHO purine salvage mutants, yH 21 and yHD l-3 ( originally from Dr. L. Chasin, Columbia University) were kindly supplied by Dr. J.A. WrighÈ, DepÈ. of Microbiology, U. of Manitoba. Mouse L5l7gy lymphoblasts and SV40 transformed mouse 3T3 cell,s were kind gifts fron Dr. W. Hryniuk (Manitoba Cancer Treatment and Research Foundation, Winnipeg) and Dr. R. Sheinen (Ontarío Cancer fnstitute, Toronto)

respectively. The Lesch-Nyhan (On Ser) and normaf human (El San) skin fibroblasts rdere obtained from the American Type Culture Co1.l_ection, Rockvj-l1e, Md. , USÂ. Àl-L the

rest of the celI lines studied were purchased from FLord laboratories, Rockville, Md. , USA.

Fungal Cel l-s

The fungal strain from which HS3 rdas purified was obtained originatly from Dr, J.S. Lovett, purdue University, and designated Achlya sp. (1969) (LéJohn and Stevenson f970). 1''

(2) crolr'th Media, Sera and Culture-vrares

Afphâ-minimal essential medium (ê{_MEM) was obtained from Flov, laboratories, Rockvil-le, Md., USAr while Fisher,s medium r¿ra s prepared according tc the method of Fisher and Sartorelli (1964). Mycoplasma tested and virus screened dialysed and undialysed fetal ca.If sera were purchased from Grand Island Biological Co., Grand fsland, Nev, york. Culture plates and tubes were obtained frorn either Lux Scientific Corporation, Ca., UsA or Falcon, ca. , usA.

(3) Radíoisotopes

AII labelled chemicals were obtained from Àmersham. The specific activities are reported in the appropriate'Methods' and,Resutts' sections. 32p-ortho- phosphate (carrier free; 8 nci/m1) vras obtained in dilute HCI (pH 2-3).

( 4') MiscelLaneous Material-s

(a) Atl chemicals used h,ere of analytical grade and ¡4rere purchased from one of the following: Sigma 74

Chemical Co., Difco Labotatories, Calbiochem, J.T. Baker Chemical Co. and Fisher Scíentific. (b) Bacto-trypsin from Difco Laboratories, DeÈroj-t, Michigan.

(c) NCS Tissue Solubilizer from Ä,mersham/Searle Corp. , I1linois. (d) pEf-cellulose plates from Brinkman Instruments (canada ) t,td. (e) Gelman (25 mn) glass fibre fil,ters, type A_E from Ge1man fnsttument Co., Michigan. /5

METHODS

(1) Growth of Cell Lines

All cej_l Ìines, except L517gy mouse ]ymphoblasts, routinely cultured 'ere in Brockway bottles at 370c in a humidified 5g CO2 atmosphere incubator. Ce]Is vrere subcultured on reaching confluency, using 0.058 trypsin in phosphate buffered saline to dislodge them from the glass surface. 1517gy mouse lymphobtasts, which do not attach to sol-id surfaces, Irere grolrn as stationary cuLÈures in 100 ml serum bottr-es. subcur.turing was carried out when the cefl density approached 4-5 X l05 cel.Islmt. The incubatj-on conditions lrrete âs described above.. The split ratios fo¡ the permanent cell- lines varied while a l:4 ratio was rigorousÌy foltowed for all primary cel.l Iines. primary cel1 lines were used bet\¡Jeen passage numbers 3_35 for al1 experiments - The grohrth medium used for culturing cell-s was o(- minimal essential medium ( < _MEM) (stanners et. aI. 1971) . The only exceptions being for CHO GAT and L5l.78y cel] lines in which o( plus -¡ræu adenine/adenos ine (0.I mM) and thymidine (0.1 mM) and Fisher,s medium (pisher and 76

and Sartorelli, 1964) respectively were substituted. In all cases, the grovrth media were supplemented ¡"¡ith I0? fetal calf serum and penicillin and streptomyc.in added at 60.6 mg/Iltre and 68.6 mg/Iitre respectively.

(2) 2) Gror,eth, "'p-Orthophosphate Labell-ing and ForF.ic

Polyphosphorylated Nucleotides Durinq Various Grcr,!'th Conditions

Cells srere gror4rn Èo near monolayer stage as described above. After trypsinization, the cel-1s were recovered by low speed centrifugation, plated in 60 X 15 ¡n¡n plastic tissue cul-ture plates and. a1l-owed to gror.v for at least 24 hr before use. permanent and primary ceII lines were used at densities of bet$reên 0.5-1.0 X l-06 and O.2-0.5 x 106 cells/plate respectivety for aLL experiments. --P-l-abelIing?? For studíes, the ceLls were first washed with 5 ml of the appropriate pre-warmed medium containing 109 fetal calf serum. Three millilitres of the 10? dialysed fetal calf serum (ÐFCS) containing incubation growth medium was next added to each cuLÈure pl,ate together 32e-orthophosphate. with 50 ¡tI of. The celLs were then incubated for various times under normal growth conditions. 77

For pulse laberling studies, the incubation time was 30 min. The 1abe11in9 was terminated by aspirating Èhe incubation mêdium off, followed by a 5 m1 wash wiÈh ice_cold 0.94 NaCl-. Ce1l-s were then rapidly frozen for later use or extracted immediatety for at least 30 min at 4oC with 32p_l.be11ing 200 ¡:t of I M formic acid,/plate. Fo, studies L5178Y of mouse lymphoblasts, the ce]Is r,,rere cultured as described. Aliquots of the ceII suspension (3_4 X 105 ce11s,/m1) were dispensed into 17 X 100 rnm Fal_con plastic tubes and the cells recovered by 1ow speed centrifugation. They h¡ere washed once hrith an equal voLume of the appropriate medium and resuspended in the same medium to give a final, density of between 0.15-0.3 X 106 cel]s,/m1. The experiment was initiated with the addition of 50 ¡I 32p-orthophosphate of to 17 X I00 run Falcon ptastic tubes

containing 3 ml of the ceIl suspension. fncubation vras carried out in a 37oC shaking waÈer-baÈh. LabeIÌing vras termj-nated by low speed pelleting of the ceLfs and the supernatant aspirated. Formic acid extraction of celIs ri7as as desc¡ibed above.

32p-LabelLed (3) chromatoqraphy of Formic .Acid Extract fron Mammalian Celts

Ten microlitres of the formic acid extract was spotted on H2O-$rashed polyethytene imine (pEJ) celluLose plates and developed one-dimens ionally with f.5 u KH2PO4 ?o

at pH 3.65 by ascending chromatography until the solvent front reached 15 c¡n from Èhe origin. Two_dimensional chromatography of formic acid extracts on pEI_cellulose plates was carried out by developing with 3.3 M ammonium fomate and 4.28 boric acid, pH 7.0 in the first dimension. After a 15 min wash in methanol, foLlor¡ed by drying, chromatography in the second dimension vras vrith the phosphate soLvent as defined above. In both dimensions, the solvent fronts were all_owed to migrate to t5 cm from the origin. After chromatography, the pEI-cetlulose plates were exposed to Kodak Rp-t4 Roya1 X-Omat fitms for 24_48 hr before devel-opment. The appropriaÈe labeLl-ed areas were identified, cut out and placed in I0 m1 Bray,s (1960) scintil-lation fluid. The radioactivity was determined with a Beckman fS-230 liquid scintillation spectrometer.

(4) DNA, RNA and protein Analysis

Cells for aII these êxperiments were cultured as described when they were labelled with 3 2p_orthophosphate for isolation and analysís of HS polyphosphorylated nucleosides (see sections l_and 2 of 'Methods,). 79

(a) DNA Pulse Labellinq ExÞeriments

Before initiating radioisotope labe1ling, the cells were first washed with 5 ml of the appropriate pre_h,armed medium, Three miltilitres of fresh medium containing I0â diaì-ysed fetal- calf serum (DFCS) and 2 pcilmf 3H_taR (methyl-3H-rdR, 27 cilm mol) were then added followed by a l-5 min incubation. The pulse was terminated by rernoving the radioactive medium by aspiration. The cel-l,s were washed once hrith 5 mL of ice_coId O.9B NaCl and then 5 ml of ice-cold 108 trichloroacetic acid (TCA) was added. Af ter at ],east Ì hr at 4oC, the cell_s in TcA were removed from plate the and the entire suspension filtered through a glass fibre filter (Gelman) foll_owed by three successive washes of 5 mL of ice-cold 5? TCA and dried with 10 ml, of 70? ethanol. The dried filter was dissolved in I ml of NCS tissue soLubitiser for f hr at 45oC, and the ¡adioactivity determined following addition of 9 ml of toluene-based scintillant.

(b) protein pulse Labelling Experiments

The procedure hras exactly as described for DNA pulse J.abell-ing studies. protein was labefLed wi.th 'H(4,5)? L-leucine (t ¡:ci,zml; 1 ci,/m mol) . 80

(c) Continuous RNA Labelling Experi¡nents

The washing procedure of celLs prior to initiation of labelling and processing for radioactivity determination were exactly as described for DNA pulse 1abelling experiments. 'H(5, 6) uridine (t ¡ci,zmt, 45 ci/m mol, 5 ¡rM final concentration) was added to the cell,s at zero time. At specific intervals, the 1abelling was terminated and 3__ H-urrdLne J-ncorporation into total RNA was dete¡mined. To correct for the possible incorporation of 3H-uridine into DNA, after its conversion to thymidine, the following procedure as descrj-bed by Skold and Zetterberg (1969) was

used. DupJ.icate samples were taken for each ti¡ne point, one for anaì.ysis of totaL radioactiviÈy incorporated, the other for actual incorporation into RNA. The tatter sample was processed as fol-lows. The cel_Ls were flooded with I0B TCA, then scraped from the plates and particulate matter collected by centrifugation. The pellet v/as suspended in l-.5 ml lM NaOH for l-6-18 hr at 37oC, neutralized hrith HCL, folloh¡ed by reprecipitation with ice-cold 5? TCA. The precipitate was finatly colÌected on a glass fibre fil-ter and radioãctivity deternined as described for DNA synthesis (4a). The amount of l-abel in RNA was calculated as the difference in radioactivity between Èhe two sampLes. (5)

into Nucleosides and Nì.¡cleotides

Experimental procedures l^rere exactty as described for-32 -'p-orthophosphate labelling studies. 14"(e)_Adenine (1.1 pCi,/ml; 60 mci,/m moL) vras added folfowed by a 2 hr incubation. Thereafter the cells were extracted vüith formic acid, as described under section (2) of 'Methods, and 15 ¡11 spotÈed on Whatman #1 chromatography paper. The chromatogrâm, vrith appropriate standard nucl-eotide markers, was developed overnight r,\¡ith saturated

(NH4) 2SO4-0.1 Itf sodium ace tate- isopropanol (75219:2) by descending chromatography. After drying, the rel-evant chromatographic strips r.i'ere cut into I X 3 cm portions and radioactivity determined by counting in Bray,s scintiL.lant. The positions of the standards were observed under UV_light.

(6) Determinatj.on of ATp Levels Using the Fireffy Luci fe rin/Luci fera se Method

Cells were cuLtured and washed as described under sections (l) & (2) of 'Methods'. Three millilitres of the appropriate medium was added fo.rlowed by a 2 hr incuL¡atio¡ 82

under normal growth conditions. The incubation was terminated by aspiration of the medium, then the cells were r¡ashed wiÈh 5 ml, ice-cold 0.98 NaCI followed by immediate freezíng j.n an ethanol-dry ice-bath. The procedures for the extraction and assay of ATP \'íere essentially similar to those described by Grummt et. al. (1977). One mil-lilitre of ice-co1d 508 ethanol was added to each frozen culture. After I hr at 4oC, the ceLls were scraped off the plate vrith a rubber 'policeman'. Suspensions rrere used directly for ATP determinations. Each viat of firefly lantern powder (¡lorthington Biochem. Corp., New Jersey, USA) sras reconstituted with 5 ml of lIrO and 100 pl was used per reaction. This was made fresh each time. Other constituents of the assay in a final volume of 2 rnl were 40 pf of the ethanol--ce11 suspensicn extract, 10 mM M9SO4 and 10 mM glycylglycine buffer pH 8.0. The .ATP standard curve v¡as constructed lvith the same volume of 508 ethanol in each sample. The light emission from each reactj-on vial- v¡as determined for 60 sec by an ATP photometer, nodel 2000 (JRB Inc., California) at L5 sec following the addition of the ethanol-cel1 suspension to the complete assay mix. 83

(7) Isolation of HS3 From Achlya and CHO wT Cells

The method of groÌring Achlya for cotd osmotic- shock treatment has been described ( Cameron and LéJohn

1978). The crude osmotic-shock fluid containing HS3 r,,ra s concentrated under vacuum to beÈhreen 0.1-0.2 m1 . The entj-re sample was fineJ-y streaked onto pEI-cell_ulose plates and chromatographed one-dimensionally using J-.5 M KH2pO4 pH 3.65 sol-vent and a sarnple of 32p-l-.be11ed formic acid exÈract from AchLya included as marker. The areas on the 32p-1ubuLl"d chromatogram corresponding to the HS3 sample were then cut out following extensive washing of the plates r,rith methanoL to reduce the inorganic phosphate contaminant. After drying, unlabelled HS3 was eluted from the cut areas overnight at 40C with 50 mM triethylamine bicarbonate (TEAB) buffer. The eluate v¡as concentrated under vacuum and washed several times rarith H2O to remove residual TEAB buffer. The concentrated sample was desalted by passing iÈ through a Sephadex G-l_0 colunn and eluting it with H2O. The fractions containing .the firsr UV-absorbing peak were pooled and concentrated to dryness. The residue was dissolved in I mI of H2O and the concentration of lls3 determined by its A26O extinction, assuming a molar extinction coefficient of L4.5 X 103 anC a MW of 1800 for HS3. The sample hras then reconcentrated 84

to dryness and resuspended j-n an appropriate vol-ume of H2O for use.

For the isolation of HS3 from CHO WT cells, g-I0 Brockway bottles of CHO cells, at 3 X l-07 cel-Is,/bottle were used. Each bottle of cells was first washed with

30 m1 of glutamine-free o{ -¡fEM + 10A DFCS, followed by a 4 hr incubation under normal grolrth conditions with the above medium. The starvation was termj.nated by pouring off the medium, vrashing the ce11s with 20 ml, of ice-cold 0.99 NaCL and then frozen- The frozen ce1ls were then extracted with 1 M formic acid for 30 min at 4oC. The isolation and guantitation of HS3 from this extract $rere exactly as described above for Achlya,

(8) Isolation of DNA-Dependent RNA polymerases From CHO WT Cells

CHO VIT cells were cul-tured in suspension cul-ture hrith conÈinuous stirring at 37oC and were useC r,vhen the culture was at mid-Log phase i.e. 3-5 X 105 cel-ts,/ml . For the isolation of DliA-dependent RNA po.lymerases, approximate Ly 2-6 X 108 cell-s r"rere harvested by low speed centrifugation and washed once h¡íth 0.99 NaCL. Â11 manipulations beyond this step were carried out at 4oC. 85

The vrashed cefls r.{ere resuspended in Buffer B (50 mM Tris-HCI, pH 7.5, 5 mM MSC12,0.1 mM EDTA, 258 gJ-ycerot (v/v), I mM dithiothreitol, 1.7 mM phenylmethylsul fony 1 fluoride (PMSF) and (NH4) 20 mM 2so4 at a concentration of 1 X 10" cel-ls,/mf of buffer. The ce1l suspension was sonicated five times for 30 sec periods with 60 sec cooling inÈervals with an Insonator modeÌ I000 sonícator (Ultrasonic Systems Inc., Farmingdale, Nei,v york) using a microtip set at a power leve1 of 5. The ce1l debris was removed by centrifugation at 12,000 X g for l-0 min. The supernatant was next subjected to high speed j.n centrifugation a Beckrnan ultracentrifuge using a Beckman 40 Ti rotor. Centrifugation was for t hr at protarnine 40,000 rpm. sulfate at 10 mglm1 in Buffer A (identical to Buffer B except for the omj-ssion of I m¡f DTT, 1.7 mM PMSF and 20 mtr{ (NH4)rSOn) was added sJ.owly, r{ith continuous stirring, to the hígh-speed centrj-fugation supernatant, to a final concentration of 1 mg/nrt protamine suLfate. The preparation was stirred for an additional 15 min and the precipitate sedimented at i-2,000 x g for 15 min. The supernatant fracÈion hras discarded and the pellet resuspended in Buffer C (essentiatly Buffer B plus 120 mf'î (NH4 ) ,SOn ) . After centrifugation at 12,000 x g for 15 min, the supernatant was saved and Èhe pellet 86

re-extracted with Buffer C as just described. Both supernatant fTactions r,rere then pooled together_ The protamine suf f ate eluate r^/as dil-uted l:6 with Buffer A to give a finaf concentration of 20 rnM (NH¿) ZSO¿. The difuted fraction was edsorbed to a 1.5 X 6 cm cofumn of DEAE-Sephadex A-25 and the resin washed with 2_3 column volumes of Buffer B. The pïoteins were eluted with

a linear gradient of 40 mM to 500 m¡t (NH¿)rso* (500 m1

totaL gradient volume ) in Buf f er A containing f m.¡.î DTT and 1.7 mM PMSF, Fractions of 0.8 ml- were collected and RNA polymerase activity of each fraction was assayed by 3H-Utp measuring the incorporation of into TCA- precipitable material using a reaction mixture consisting

of 70 mM Tris-HCl-, pH 7.5, 25 mM 2 -mercaptoe thanol, 2 mM 3"-urn Mnc12, 0.5 mM cTp, GTp and ATp, 4 ¡ci of (36.8 Ci,zm mol), 5 pg of denatured calf thymus DNA and 15 pl of enzyme fraction in a final volume of 60 ¡rt. The reaction mixture was incubated at 30oC for 30 min before 40 ¡rl of the assay mixture r,ùas spotted on a piece of r¡,Ihatman #t filter paper. The dried filter paper was washed sequentiafly for 10 min periods with ice-cold 1OB TCA, 5E TCA (twice) , 95å ethanol and acetone. After complete drying, the radioactivity on the filter was determined in _f0 ml- of a toluene-based. scintilfant using a Beckman LS-230 liquid scintifLation spectrometeï. 87

(e) =a=

Cells were cultured as described previously in section (1) ,MeÈhods'. of Foï permeabil i zation of CHO WT cells, the procedure of ¡ewis et. al-. (197g) was foJ-lowed, with minor modifícations. CeIls ürere first trypsinised, pelleted by centrifugation and Èhen washed once r,rith o(-¡æ¡l warmed plus l0â fetal calf serum. The celLs v¡ere resuspended at a concentration of 2-3 X I06 cells,/ml and incubated for 30 min at 22oC in a permeabilizing buffer consisting of 1A Tween-8o, 0.25 M sucrose, 0.01 M Hepes buffer (pH 7.2) and 2 mlvt dithioth¡eito1. The permeabilized cells were centrifuged and resuspended in the permeabilizing buffer at a cell density of 3 X 106 ce]ls,/ml . 3H-UTP Each assay for the incorporation of into RNÀ by permeabilized CHO VùT cells consisted of 300 ¡:I of the cell suspension and 200 ¡:l of the RNA assay mix as described by Castellot et. al. (1978). Essentially the RNÀ assay mixture consisted of 35 m¡,I Hepes (pH 7.4), 80 mM KCl, 4 mM MgCl, , Z. 5 rntr.f potassium phosphate (pH 7 . 4) , 0.75 mM CaCIr, 50 mM sucrose, 0.5 mM MnC12, 4 mM ATp, 0.25 mM CTP and GTp and 0.0L mM UTp (2.5 ¡:Ci,/sample). The cell-assay mix was incubated at 37oC in a shaking water_ bath. The reaction was terminated by centrifuging the 88

cel-ls, removing the supernatant by aspiration and 5 rnl- of ice-cold 58 TCA added to the pelLet. After at least

L hr at 4oC, the precipitate was filtered through a glass fibre filtêr followed by four successive washes of 5 ml ice-cotd 58 TCA and th¡ee 5 ml Ìrashes using 70å ethanol before drying. The filter was processed for radioactiviÈy determination as described for section (4a) of 'Methods'.

(10) Assav of RNA Synthesis By permeabilized CHO vtT

Cells fn the presence of Achlya HS3

Preparation of perrneabilized cells and the RNA procedure assay were as d.escribed previcusly. Achlya HS3, at a final concentration of I00 ¡:g/m1 was added to the appropriate samples. The incubation time for the assay was t hr at 37oc. 89

RNS ULTS 90

(1) HS3 Synthesis by CHO lVT CelLs

Studies were carried out to determine if cu1Èured CHO I^7T cells can produce the dinucleoside polyphos_ phates , HSl, 2 and 3, whích were previously found in various fungi (LeJohn et. aI. 1975) and characterized recently by McNaughton et. al. (1978). HS compounds accumulated in significant quantities when fungal cel1s are starved of nutrients and especially so when glutamine is absent or being depJ.eted in a growing medium (LeJohn et. aI. 1978). Glutamine is present in most synthetic media at concentrations of betvreen 2_4 mM¡ which is usually 5-L0 fold greater than that of any of the other amino acids added as supplements (Stanners

et. a1. 1971; Fisher and Sartorelli 1964). It is an essential amino acíd, for growÈh of cultured ma¡nmalian cells is rapidly arrested either upon its removal- or upon its depletion (Eagle et. a.j.. 1955; Iey and Tobey 1970; Pardee !974). This is not unexpected considering the involvernênt of glutamine in numerous cetluLar

processes (Stadtman 1973) . Since stringent bacterial- strains produce

PPGpP and pppcpp in response to deprivation of an essential- amino acid (Cashel 1969; Cashel- and Galtant !969), it was decided Èo use a simil-ar approach and 91

see if glutamine r,rithdravral v/ould precipitate the accumulation of these HS compounds in marunalian ce11s. As illustrated in fig. 3, panel IA, onfy ¡IS3 was found in appreciabLe amounts in cells exposed to gJ.utamine-free grovrth medium. ln the presence of 2 mM glutami-ne, HS3 remained at lov/ l-eveLs (fig. 3, panel rB). 32"-l.b"ll"d two other very faintly spots appeared to chromatogtaph at regions where fungal HSl and HS2 would no¡ma11y be found. Hor+rever, from the labe1ling patterns, HSl and HS2 levels appeared to be unaffected by changes in the supply of glutarnine. The nature of 32p-lubuIl"d these tr"¿o entities hás not been inve s ti ga ted . Figure 4 is an autorad.iograrn of a for¡nic acid extract 32p-Iubelled fto. qlutamine-deprived CHo wT cel-Is chromatographed two-dimensionally on pEl_celIuIose pLate. As can be observed, HS3 mígrates subseguent to G?P in both dimensions. This Èwo_dimensionaL chromato_ graphic system \^ras util-ized occasional-ly to check for the purity of the presumptive IlS3 obtained by the Figure 3: Effect of L-glutamine deprivation on HS3 leve.ls in CHO I^7T celf s. The gro\,rth of cell-s was as described in 'Methods,. The cel1s were washed once with 5 m.I of prewarmed a( -¡tu¡4 supplemented lvith or without glutamine plus 10å DFCS. Three miffilitres of the 32ni appropriate medium containing SO ¡rt of r,ras then added to each pfate, followed by a 2 hr incubation. The labelling was terminated

by removal of the medium by aspiration and. washing of the celÌs with 5 mI of ice-cold 0.93 NaCl. The cell-s were then extracted for 30 min at 4oC by the addition of 0.2 mI of f M fonnic acíd to each plate. For chromatography, 10 ¡rl of the formic acid extract v,¡as appfied on PEI-cellulose Èhin layer plates and deveLoped by ascending chromatography. The procedures for chro- matography and the determination of the '\) levefs of "-P-labelled compounds have been described in the 'Methodsr. (IA) _ glutamine; (IB) + gfulaminê *þr߀,.

lllJ.l t:.,,,,, *

- :.r" 't$lxnu¡*q

Ë*Ë 94

Fì-gure 4: Autoradiogram of a ttro- dimens ional chïo-

matogram of 1M formic acj-d extract of CHO WT ceJ-ls starved for glutamine for 2 hr. â,1 "pi labelling procedure was as descrihed in the legend for fig. 3. The first

dimension ( 1) was developed in 3. 3 M am¡nonium formate-t and 4,2? boric acid at pH 7.0. After chromatogïaphy in the first dimension, the plate was washed ín methanol for 15 min and then dried. The second

dimension ( 2 ) was developed in 1. 5 M KH2PO4, pH 3.65. The solvent front was -) allowed to migrate about 1,5 cm from the origin in both dimensions. _+z .s I I È ¡ t è-'i i.,'r:1 ihi::1 96

one-dimensionaL L.5 M KH2PO4 solvent system. In a1l_ instances, 32p-labelled the Hs3 obtained one-dj.mens ionaL ly rras total-ly recoverable by the two_dimensionaL chromatographic system. This, together with physical and chemical analyses of mammalian HS3 isolated fron PEt-cetl-u1ose plates suggest that the 32n_1.b.1]"d entity with a Rf vatue of 0.36 (table L) consists essentially of HS 3.

Manmalian HS3 has been found to co_ chromatograph with fungal HS3 as shor{n by their Rf values in table l. Also, chemical and physicaL analyses of mammalian and fungar HS3 suggest thaÈ they are simira¡, if not identical molecuLes (table 2). Bacterial ppcpp and pppGpp did not co-chromatograph v¡ith mammalian and fungal HS nucleotides either one- or two-dimensionalLy (table 1). Ãt no time in these studies were ppcpp and pppcpp ever detected in mammal,ian ceLls cultured under different growth conditj-ons. 97

TABLE 1

Rf Values of Nucteotides on pEI-Cetlulose pÌates.

_ . Nucleoti de l- D.lmenslon z-^nd D.lmensr on

(AchJ-ya) HS3 0 .22 0. 35 HS3 (Mammalian) o.2I 0.36 PPGPP (MS I) 0.33 0.28 pppcpp (t4s Ir) 0.29 0.17

GTP 0.38 0. 4s

lst dimension developed by ascending chromatography with 3.3M annonium formate and 4.28 boric acid, pH 7.0. 2nd dimension devel-oped by ascending chromatography \dith 1,5!r KH2PO4' PH 3.65. TÀBLE 2

Comparison of the properties of Achlya and mammalian HS3 (rnodified from Lewis et. aI. I977).

Prope rty Ach1ya HS3 ¡lammaÌian HS3 (cHo wT celfs)

R--l r o.2za(0.:s)b 0.21(0.36) DEAE- Sepha¿lex-A2 5 elution at: (i) pH 8.0 (ii) 1. 1M 1. lM PH 3.6 0.28M 0.27M at pH 7. O ^max7\min at pH 7.0 26Onm 26Onn À 232nm 2 3lnm "250/260 pH ar 7.0 0.84 . "280/260À 0. 84 ar pH ?.0 0.45 -'290/260 0.41 at pH 7. O 0. t6 ResisÈance Èo alkaline 0. 16 phosphatase * resistanÈ Combined treatment hrith resistant phosphatase and pyrophosphata se degraded Components degrade d adenos ine + _2't uri dine + +- gL utama te 8 phosphates + + mannitol + + + N.T. sugar X + N.T. lTwo dimensional chromatography descibed in'¡rethods'. on PEf-cellulose as a .st .t- O1mens1on.-. b ¿-^nd 241"o dimension - d"t..ted in HS3 isolated fibroblasts. from Lesch-Nyhan human *fncubated at 3 7oc for t phosphatase. hr brith bacterial alkaLine +Present, N. T. - noÈ tested. 99

(2)

As previously illustrated (fig. 3), the lack of 2 mM glutamine in the culture ¡nedium resul-ted in an accumul,ation of intracel-Iular HS3 by CHO WT ceI1s. An experiment r4ra s performed Èo determine the concentration effect of glutamine availabiliÈy on HS3 and GTp te.iels in CHO WT ce1ls. The results are iltustrated in figures 5a and 5b. tn a 2 hr labelling experiment, glutamine at 0.1nlr $/as adequate to reduce cellu1ar HS3 to the r.ever.

observed when a normal quantity (2 mM) of glutamine was supplied. When GTp 1evels were moniÈored, an inverse rerationship 32p-tub"1l"d between HS3 and crp levers was observed (see figs. 5a and 5b), for an increase in HS3 red to a corresponding decrease in GTp and vice versa. It can be observed from figures 5a and 5b that 32p-lubufrea the ratios of ffi (cpm) at gluramine concentrations of 0 pM and 2 mlq were 0.37 and 0.047 respectively. However, fluctuations in the ffi ratios especially for g lutamine- s tarved CHO þIT cells have been observed. The average ffi ratio was 0.65. These variations in the ffi ratios were similar to those of other cel.l. lines that l¡rere studied _ Figure 5: Effect of varying glutamine concentrations 32pi (0-2 mM) on incorporation into (a) HS3 and (b) cTP in CHO t{T cell-s. Experímental-

conditions r,i¡ere as described in the J-egend to fig. 3. HS3- CPM (¡lO-3 I -N

9 9 l. eÊ GTP-CFU f¡ lO-1, o )l I () o f' I 3 a

o

oF 702

(3) Effect of L-clutamine or L-Isoleucine Deprivation on HS3 Metabolism by cHo wT celts

To ascertain vrhether HS3 is produced j.n response to either a general deficiency in amino acids or specifically for gJ-utamine (see fig. 3), CHO VüT cells were deprived of another essential amino acid, isoleucine. This nutrient has been impìicated in regulating the ce1I cycle of CHO ce11s (Ley and Tobey 1970). The results of fig. 6 shor'ü that isoleucine deprivation did not

Lead to an accumuLation of intracellular HS3 (see paneL lc) .

(41 Effect of L-ctutamine, Adenosine and Thymidine

Deprivation on HS3 Metabol_ism by CHO cAT Cel1s

A CHO mutant which is auxotrophic for glycine, adenosine and thymidine (GAT ) ovring to defective folate metabolism (McBurney and lvhitmore L974) was utilized to further define HS3 metabotism. Since HS3 accumulation appeared to be triggered specifically by glutamine deficiency (see fig 6) and owing to the importance of this amino acid in ce11ular meÈabolism, particularly in its role in purine metaboLism (Stadtman l-973; Mah1er 103

Figure 6: Effects of glutamine (g1n) and isoleucine (ile) starvation on HS3 accumulation by CHO VfT celLs. The autoradiogram shoi,rs the relative HS3 1eve1 in cells incubated under different growth conditions (a) -91n, +i1e; (b) +91n, -ile; (c) +gIn, +i1er (d) -g1n, -ile. The experimental procedures were as described in the legend to fig. 3. 104

GrP --,rl : lt HS3 -, |l *å 105

and Cordes ¡.97]-l , it was reasoned. that HS3 accumulation rnay be re.l-ateil to a lack of purines. An experiment

was devised in which HS3 accumulation by the CIIO GAT- mutant was monitored with respecÈ to the availability of glutamine, adenosine and thymidine. The resuLts show (tab1e 3) that only the r^rithdrawal of adenosine

f rom the complete growth medium i . e . o( -trlrlt plus l-04 DFCS, adenosine and thymidine, resulted in a signifj-cant (four foLd) accumulation of HS3 by this auxotroph.

(5) Effect of L-clutamine Withdrawat and The Replenishment with purines gn HS3 Metabolism by

The preced.i-ng results (tabIe 3) suggested Èhat

HS3 accumulation in cultured mammalian cells may be induced by a lack of an intracelluIar supply of purines. It can be hypothesized that cells with bl-ocked de novo purine biosynthesis and lacking purine salvage enzymes would exhibit elevated HS3 leveLs even &,hen supplied with an exogenous purine. 'Vùildtype' CHO and CHO purine salvage mutants rarere therefore examined for their abj_1ity to accunulate HS3 during glutamine st.arvation r when 106

TABLE 3

Effect of L-glutamine, adenosine and thymidíne availability on HS3 meÈabolism by CHO cAT- cel-Is.

HS3 (expressed as ? of control)

+ glutamine + adenosine + thymidine l-00 - glutamine + adenosine + th1'rnidine g 6 + glutamine + adenosine - thymidine . 96 + gluÈamine - adenosine + thymidine 437 + glutamine - adenosine - thymidine 4t3

32p-crthophosphate The control is the amount of incorporated into HS3 after a 2 hr incubation in complete c(_ ¡,tEM + 10C DFCS + 0.1mM adenosine + 0.1mM thynidine. The gl,utamine concentration used r4ras 2 mM. LeveLs of HS3 r,Jere detected as described in the legend to fig, 3. I07 provided with an exogenous source of purines. Table 4 summarises the results obtained. Hypoxanthine, adenine and adenosine were effective in antagonising glutamine_ less induction of HS3 accumulation in vrildtype CHO cells. Both HGPRT sal_vage muÈants, yH 2I and yHD 13 however, were unabLe to do likel^¡ise when supplied with hypoxanthine t up to 2 m.t\¡ hypoxanthine l^ras without ef f ect. By contrast, adenine and adenosine preventêd Hs3 from accurnuLating in both types of purine salvage mutants during 9luÈamine deprívation. lt r^,a s surprising that YHD 13, whích is HcpRT-/ApRT , was abLe to utj-Lize adenine and adenosine to depress HS3 level-s. Under normal grorrth conditions, the purines mentioned had negligibLe effects on the Levels of HS3 in aLl three cell lines (see table 4). 108

TABLE 4

Effect of L-glutamine and exogenous purines on intracellular tlS3 levels in CHO trvT and CHO purine salvage mutants.

Cel1 line HS3 as A of control

'wiLd type' cHo 10.9 (+ L-gln ) L-gl-n + 0.36 m¡t ?.7 (-\* hypoxanthine) 1.4 T,-gln + 0.36 m.t4 i 7.r t- L-gln + 0.18 nt'4 ) !.9 (- L-g1n + 0.3 mM adenosine) 7.9 (+ L-gIn + 0.3 mM ) L0.5 (- L-gln + 0.3 mM adenine) (+ 1f.3 r-gLn + 0.3 mM " )

YH 21 (HGPRT-) 191-9 (- L-g1n + 0.3 mM hypoxanthine) 11.5 (- L-gIn + 0.3 mM adènine) 12.0 (+ L-g1n + 0.3 mM " )

YHÐ 13 (HGPRT- !04.7 G T,-gln + 0.3 mM hypoxanrhine) ,/eenr- ¡ .3 (- L-s1n + 1.0 mM 10_1.8¡0 t i- L-91n + 2.0 nM ) 8.0 (- L-gfn + 0.3 mM adenine) 9.3 (+ L-g1n + 0.3 mI4 adenosine)

The control for each cel] type is the anount of 32pi incorporaÈed ínto HS3 after a 2 hr incubation in L- glutamine (I-gln) -deficient medium and is expressed as 1008. Hypoxanthine, adenine and adenosine brere added to the groh'th medj.um (t L-glutamine) at the start of all the experiments. yH 21 is a hypoxanthíne/guanine phosphoribosyl_ transferase CHO mutant and yHD 13 is a hypoxanthine/ guanine phosphoribosyl trans fera se-aden ine phosphoribosyl_ transferase CHO mutant. The experimental procedures rarere as described in the legend to fig. 3. 109

(6) Infl,uence of Puríne and Pyrinidine Compounds, Methotrexate and .Azaserine on the Accumulation and Depl-etion of HS3 in Three CHO Cell Lines Incubated in the Presence or Absence of L-Glutamíne

ft was of interest to determine the infl-uence of various purines and pyrimidines and antagonists of de novo purine biosynthesis on HS3 metabolism since it has become apparent that the physiological levels of this polyphosphoryLated compound aïe somehow related to changes in the supply of purines. Table 5 surnmarises the effects of various purj_ne and pyrimidine compounds, nethotrexate (MTX) and azaserine

on the accumulation and depletion of HS3 in CHO WT ce1ls and the th'o purine salvage mutants yH 2l and yHD 13. In Èhe presence of glutamine (see under r+' columns) none of the exogenousl-y added purines and pyrimidines altered significantly HS3 level-s in these ce11 Iines although hypoxanthine, adenine and adenosine slightly depressed the levels of this polyphosphorylated compound in CHO WT cells. In contrast both azaserine and IITX, which are potent inhibitors of de novo purine biosynthesis (French et. ä1. 1963r Blakl-ey J.969), r^rere active in eÌevating intraceJ-lular HS3 levels in both CHO WT and

YHD 13 cells - 110

In the absence of exogenous glutamine, a]l the purine bases and nucleosides tested erere very potent

in reducing the quantity of HS3 accumulated by CHO WT cells (see under r-' colurnns); but in the case of yH 2l and YHD 13 celL lines hypoxanthine, inosine and guanosine

ûrere rather ineffective though adenine exhibited the same

potency in these two celI lines as was for CHO WT cells. In the case of pyrimidines, th]¡¡nidine, thymine and orotic acid did not antagonise the accumulation of HS3 while 5-fLuorouracif f fts metabolic derivatives are known to inhibit thymidylate synthetase and other cellu1ar functions (HeideLberger 1lZ5)] , uridine and cytidine, in decreasing order of effectiveness, did. In comparison, allopurinol , a xanthine analogue which potently inhibits xanthine oxidase (EC 1.2.3.2) (Pomales et. al. 1963), v¡as as effective as cytidine.

(7t Determination of ATP Levels in CHO WT Cell-s After Exposure to Various cror"¡th Conditions and MetaboLic Inhibitors

Results have been presented in the preceding sections which showed a correlation betr,Jeen an el,evation F ts TABLE 5 P

rnfluence of purine and pyrimidine compounds, methdtrexate and azaserine on HS3 ¡netabolism in cHo wr and purine sarvage mutant (vu.zr and yHD 13)ce11s incubated in growth nedíum with or without supplemented L-glutamine.

Relative amounts of HS3 in cells

CHO I{T YH 2I YHD 13

Compound addedl (+ : -) glutamine2 (+ : -) glutamine (+ : -) glutamine ni1 (1.00 1.00) (1.00 : 1.00 (1.00: I.00 hypox an thine (0. 70 0.09) (0.90 : 1.10 (1.02 : 1.10 inos ine (1.03 0. r0) (1.00 : 0.90 (1.05 : 0.90 aden ine (0.7s 0.13) (1.20 : 0.12 (1.00 : 0.09 adenosine (0.67 0.08) deoxy ade nos ine (1.04 0.1s) guanosine (1.00 0.I3) (,I.20 -. I.20 cytidine (1.10 0.ss) thymidine (1.10 r.2rl (1. 10 : 1.2 0 (1.10 : 1.10 thymi ne (0.96 r.01) uri dine (o .92 0.35) (1. 00 : 0.40 5- fluorouracil (0.90 0. 11) (o.77 : 0.17 orotic acid (r.10 1.03) azaserine (1. 70 1.33) me tho trexa te (2.32 -) (1.80 : a llopurinol (- 0.s6) The experinental procedures were as described in the legend to fig. 3. I Compounds added to the medíum at the start of all experíments, were used at 0.3 mM (fina1 concentration) except for MTX (I uM), alJ-opurinol (100 ug,zml), and azaserine (50 ug,/ml). 2 * = (cpm:HS3 in cells + gtutamine + compound) (cpm:HS3 in cells + gluÈamine)

(cprn: HS3 in cel-l-s - glutamine + compound) (cpm: HS3 in cel1s - glutamine) cpm = counts/minute

H P N) 113

of intracelluLar HS3 and a btock in de novo purine biosynthesis el-icited by either giutamine deprivation (fí9. 3) or antimetabotite treatment (Èabte 5). Further, there is an inverse relationship betÌ{een HS3 and cTp levels (figs. 5a and 5b). ExperimenÈs were therefore performed to determine the fate of celluLar ATp concenÈrations in ceIls exposed to conditions which resuLted in an accumulation of HS3. Glutamine deprivation and the addition of either MTX or azaserine led to moderate decreases in ethanol_ extractabl-e ATp as determinéd by the firefJ.y luciferin,/ l-uciferase assay (table 6). MTX, at 10 pM, Losrered intracellular ATp level by about 128 whll_e gtutamine

starvation and. azaserine treatment depressed ATp concentrations by 208 and 3OE respectively.

(8) Effects of clutamine Deprivation or Sufficiency 1^ on -'C-Adenine lncorporation Into Adenine "rr..

Since gl-utamine deprivaÈion resulted in a decrease in both cellular ATp (t.ab1e 6) and GTp (fiq. 5b), a 2 hr labelling study was carried out to determine 14c-.d"r,ir," the retative íncorporation of into adenine TABLE 6

ATP concentrations, as determined by the firefly luc i ferin,/luci ferase assay method, in cHo wT cells afÈer exposure to different growth condit.ions and metabolic inhibitors.

Growth Condition ATP (B of control )

Complete growth medium 100 Compl-ete growth medium glutamine 79.9 CompLete growth medium + l- 0 t¡l\,l lvlTx 88 .2 Compl-ete growth medium + 50 ¡tg,/ml azaserine 69 .9

The control, which is expressed as L009, is the ATp level- in ceLrs incubated in complete growth medium i.e. o( -¡íEI'I + 109 DFCS. The ATp concentrations were determined 2 hr after exposure to specific grovrth conditions and metabolic inhibi.tors, 115

nucleoside and nucleotides by glutamine-starved and . unstarved ce11s. Figure 7 represents a one_dirnensional paper chromatograph shovring the leve]s of acid_ 14C-"d"rrirru extractable 1abef l-ed nucleoside and nucLeotídes frorn YHD 13 ceIls. Glutamine- starved ceIls had the

label accumulating predominantly in adenosine, AMp

and ADP with hardly any in ATp. !{hen glutamine was supplied, the l-abel- was incorporated almost entirely in adenosine and ATp which contained the bul-k of the labe1 supplied.

(9) Effects of L-clutamine and Adenosj-ne Deprivation and Replenishment on Rate of Accumulation and Pool Size of HS3

Since ít was demonstrated that gJ.utamine and adenosine withdrawal caused the accumul_ation of HS3 in CHO ÌùT and CHO cAT- cel-Ls respectively (fig 3; table 3), experiments Ì{ere performed to determine the changes in the rate of accumulation and pool size of this molecule when such cells were subjected to these same nutritionaf conditions. Figure 8 illustrates the results of an experiment in 32pi which was used to pulse l-abel cHo vtT cells during 116

glutamine starvation and the rate of labe1ling HS3 determined. The rate of HS3 accumulation .increased by 5-6 fold in 30 min over the contro'l culture and was subsequently maintained at this rate for at least 5 hr. From 6 hr onwards, the rate of HS3 accurnulation declined slowly reaching control values by 30 hr. rn a continrro.r= 32P-lubelring experiment, as sho\n'n in fig 9a' the intracellular pool of HS3 accumulated by CHO WT cells increased rapidly upon initiation of starvation. HS3 leveL increased five to six fold above the controL culture by 5-7 hr and this was followed by a slow decline to control Level. rn a similar experiment (fig. 9b), Hs3 accu¡nuLated by CHO GAT cells during adenosine starvation foltovred essentially the same pattern as that observed for CHO wT cells during glutamine deprivation.

When CIIO WT cells that had been starved for 2 mM glutamine were refed with this amino acid, the high intracellular HS3 Level decreased rapidJ-y to control levels within 15 min. This is shown in fig. l-oa. A similar phenomenon was observed (fiS. tOb) when adenos ine-s tarved cHO GAT cel1s were replenished rvith

0.1 mM of this nucleoside. LL7

Figure 7: Chromatographic profiles of one-dimensional_ chromatograms of 1M formic acid extracts obtained from YHD 13 ceLls in.cubated in growth medium with ( G-O ) and vrithout ( EHI ) 2 mu glutamine and label-l-ed with. 3.3 of *'C-adenine14 ? 1a FCi L(8--=C)adenine; 1 mCi,/m per plate 60 ' mol l'I ' for 2 hr. The experimental procedures for gïowth, label1ing and formic acid extraction of cell_s were as descrj-bed in the legend to fig. 3. The chromatography method has been described in the rMethods'. N 'g x E ocl

!o E o CL (,o .s

o, cÊ !to ()I ç

5t0 119

Figure 8: Effect of glutamine starvation on the rate of HS3 accumulation in CllO i{T cells. The experimental protocol was as described in legend to fig. 3 except that the ceLls were pulsed with 32pi for 30 min at the intervals specified. HS3 : (þ-{ ), minus 2 mI,l glutaminet ( EF_E ) , plus 2 mM glutamine. HS3 - CPM (xtO?)

_-{ 3 (D

:t co õ \27

Figure 93 Effect of nutrient (glutamine or adenosine) starvation on intracell-ular HS3 pool sizes in CHO cell lines. The experimental protocol was as described in Legend to fig 3. In these continuous label]ing experiments , 3'rí was added to the ceLls at zero time and samples taken at the

times specified. (a) CHO WT cells: HS3: ( o--€ ), minus glutamine; ( EHI ), plus glutamine. (b) CHO GAT cells: HS3: ( Ò-_---o ), minus adenosine; ( EF-+l ) , plus adenosine. (o) (b)

Oc :

TL c) jr (Í) T

Trme (hours) 123

Figure 10: Replenishment of limiting nutrient (glutamine or adenosine) to cHo cel,l- Lines and its effect on HS3 poo1s. Experimental procedure as described in J_egend to fig. 3. (a) CHO WT ce11s: ltS3: (o-__{ ), minus glutamine; ( g_----g ), plus gLutaminet ( e----O ), glutamine (2 nM) added to

glutamine- s Èarved cultures at the time

indicated by the arroh'. (b) CHO GAT- ce1ls: HS3: (H), minus adenosine; (E}---"-4 ), ptus adenosine; (G---€ ), adenosine (0.1 mM) added to adenosíne- starved cel-ls at the time indicated by the arrohr. HS3 - CPM (xlQ-r¡ oÀ) o

s -l d(¡=

J co g an 125

't.t (10) Fate of "P-Labelled HS3 Accumulated by cHo wT Ce1ls During Glutamine Deprivation Following ta "pi withdrav¡aI From the Incubation Medium

As illustrated by 32p-1"b.11ing studies (figs. I and 9a), glutamine starvation of CHO WT cells resulted in an increase in the rate of accunulation and pool size of HS3. An experiment was devised to ascertain )a the fate of "P-tabel-led HS3 during the amino acid starvation period. The results of fig lL show that r.vhen ?? exogenous "-Pi was withdrawn from ceLls incubated in glutamine-deficient medium, the prelabelled HS3 remained constant at its elevated leveÌ whife cells not starved for glutamine retained their low basal HS3 ta 1evet. Because "p-tls3 did not decrease during the next 2. hr, it. can be concluded that the molecuLe was not being converted to some other product as would be expected if unlabelted phosphate in the fresh medium was äcting as a 'chase' . Glutamine r4rithdrar,ral, therefore ' was stimul,ating HS3 synthesis. The question remains unanswered whether glutamine acts by stimul-ating !tS3 utilisation or conversion to some other metabolite (s) because of the rapid decline in its level when glutamine r^ta s added to cel1s rn'ith accumulated HS3 (fig. lOa) I'igure 11: Determinatj-on of leve1" of 32ei-pr.-

labelled HS3 in CHO WT celJ-s following the withdrawal of 32pi fro* the incubation medium. The grohrth, washing, 32pi- J.abelling of cells and the quantiÈation

of HS 3 \^rere as de scribed in legend to fig 3. 32pi-lubulling was carried out for 2 hr and aft.er r,irhich the incubation medíum was completely replaced with rcold.r medium. (È--...{ ), HS3 ( minus glutamine, p1rr" 32li at zero tine); (EH¡), Hs3 ( plus glutamine, plus 32"i zero time); (O-----O HS3 "a ), 32ni (minus gtutamine, r"*o.r.d at 2 hr) ; 32pi ( EÞ-- -tr ), us¡ (plus glutamirre, removed at 2 hr. 127

32Pi removed

ñP J o o :' \'-.-' fL (J rrI U) - o'5

----tr-tr tr

o1234

Time ( hr ) 728

(11) Time Course studies of the Effects of s-Fluoro- uracil (5-FU) on El-evated HS3 Levels Accumul-ated Durinq l-Gl-utamine Starvation or MTX Treatment

As describeil previously (see figs. 8, 9a and

table 5), glutamine deprivation or exposure to MTX resultèd in an accumulation of HS3 by CIIO WT celfs while 5-FU effectively antagonized the increase in HS3 induced by the glutamineì.ess condition (see table 5). These observations, coupLed with a recent report (Ullman et. aL. 1978) which suggested the possible antagonistic effect of ¡lTx on 5-FU cytotoxicity, l-ed to a time-course evaluation of the activities of these tuto corEnon antineoplastic drugs on HS3 metabolism in CHO WT and L5178Y mouse leukemia lymphoblasts. The resuLts are il-lustrated in figures I2a, b and c. The addition of 5-FU to gtutamine- starved

CHO v'¡T ceLl-s resulted in an immediate depletion of HS3 to control leve1s within 2 hr (see fig. 12a). In contrast, fig. 12b shows that the addition of 5-FU to MTx-treated ce1ls did not elicite an immediate depJ-etion of the HS mol-ecule though the continued accumulation of the polyphosphory lated dinucleoside was severely

restricted. A moderate decline in ce1lular HS3 129

was, hor"¡ever, observed 2 hr following 5-FU exposure. By the third hour, the level- of HS3 had decreased to 508 of non-treated cuLtures. When L5178Y mouse

leukemj-a lymphoblasts hTere treated wíth 10 PM llTX for 3 hr followed by the addition of 5-FU, as iIl-ustrated in fig. f2c, the intracellular HS3 which had been accumulated \^ras depleted. to controf level-s r+ithin 2 hr. These results are comparable with those reported in fig. l2a. The data which have just been presented suggest that while glutamine withdrawal or MTX treatment evoked an increase in HS3 synthesis, 5-FU was effective in antagonizing this IIS3 response . The mechanism(s) for this phenome.non, though intriguing, rema.ins to be el-ucidated. 130

32p-]ub"lled Figure 12: Effect of 5-FU or, HS3 accumulated during glutamine starvation or MTX treatment using conti.rrrou= 32p- 1abe1ling procedure. Experimental- protocoL as described in legend to fig. 3 and in rlqethods'. The arrovrs indicate the time when 0.5 mM 5-FU lras added. (a) cHo wT cel-ls: (E}-+l ), pLus glutaminer (6'+ ), minus glutamine; ( O----O ), minus glutamine pÌus 5-FU. (b) cHO WT cells: ( EHt ), minus MTX; ( H ), plus Mrx (1 pM); ( O---€ ), plus MTX (1 pM) plus 5-FU. (c) L5178Y cells: ( E}_-{ ), minus MTX; ( o,--.{ ) , plus MTX (10 ¡rM) ; ( O - - --O ) , plus l,tTx (10 ¡:M) plus 5-FU. lo) (b) (c) 5 lr) Þ4 j

ofL5 -Á. I gb (n)rO f t q I \ I a.

24

TIME ( hr ) I32

(I2) Effect of L-Glutamine and Adenosine Starvation on Protein Synthesis

Bacterial ppcpp l¡as been shown to be a potent inhibitor of in vitro bacterial protein synthesis (Yoshida et. a1. 1972) by specificalJ.y blocking the formation of the protein synthesis initiation complex.

It r,ta s therefore of interest to determine v'rhether or not a correlation could be observed betr,,¡een the intracellular pool sizes of HS3 and the rate of protein biosynthesis using -H-Leucine pulse label-Iing experiments. As shor^rn in fig. l3a, the rate of 3H-leucine incorporati.on by CHO WT cells was maximally inhibited within one hour following glutanine deprivation. Upon replenishment of gl-utamine to the starved culture, the optimal- rate of protein synthesis was achieved rapidly. Horvever, when adenosine was withdrahrn from the grovrth medium 3H-leucine of CHO GAT-ceIIs, the rate of incorporation into protein \¡ras unimpaired (fig. 13b) for at least 5 hr. Since results show (fig. 3 and table 3) that glutamine and adenosine vriÈhdrarlrral from the growth media of CHO VIT and CHO GAT-ceLIs respectively, invariably resulted in an accumulation of HS3, protein synthesis, therefore, is apparentì.y independent of the levels of intracellular HS3- 133

Figure L3: Effect of nutrient starvation and

replenishment (glutamine or adenosine ) on rate of 3¡i-leucine incorporatJ-on into proteins. Cells were pulsed for 15 min with 3n-leucine (3 pCi,/pl-ate) at the times indicated. (a) CHO wT celfs: (H ), minus glutamine; ( EHI ), plus gJ-utamj-ne; ( O--- -O ), glutamine (2 mM) added to glutamine- starved cultures at the tirne indicated by the arrow. (b) CHO cAT- cells: ( O€ ), m.inus adenosinei ( E]_-{I ), plus adenosine- Protein - CPM (xtO3)

=! (D3 co õ 135

(13) Effect of L-Glutanine and Adenosine Deprivation on Rate of- -H-Thymidíne3__ Incorporation Into DNA

rn vitro studies (Lewis et. al-. 1976; Lewis et. al. 1977) have shor"tn that HS3 is a potent inhibitor of both ribonucleotide reductases from Achlya and CHo ce1ls. The enzyme is involved in catalysing the reduction of ribonucleoside diphosphates to deoxyríbonucleoside diphosphates for DNA synthesis (Larsson and Reichard 1976). Also in Achlya, an inverse relationship betr,teen rates of Dl¡A synthesis and intracellular HS3 levels was observed (Lewis et. aI . 1976). It vtas therefore pertinent to determine if a sinil-ar phenomenon exists in cultured mammalian ce11s subjected to fluxes in nutrient availabilíty. I'ab1e 7a illustrates the effects of g]-utamine starvation and its replenishment on the rate of 3tt-thyrnidine incorporation into DNA by CHO I¡üT celÌs, The rate of thymidine incorporation dropped to about 252 of control within 1.5 hr following the removal of glutamine from the grobrth medium. Upon readdition of 2 mM glutamine to glutamined- starved cultures, the rate of DNA synthesis increased rapidly, reaching 75à of the controf rate wi.thin t hr.

CHO cAT cells were similarly affected by adenosine withdrawal (tab1e Zb) , âIÈhÕugh a marked 3it-thymidine drop in the rate of incorporation into DNA was not observed before 2 hr post starvation. HohTever, the readdition of adenosine to ceLls deprived of this nutrient resulted in a rapid recovery in ÐNA synthesis- Ilfithin t hr, the rate was about 90ã of control.

The data presented are consistent with Èhe results obtained fro¡n the Ach1ya studies (Lewis et. a1. 1976). It is therefore J-ikely that HS3 may regulate DNA synthesis in mamrnaLian ce1Ìs by exerting its effect on deoxyribonucleos ide diphosphate synthesis. IABLE 7A

Effect of ],-gluta:nine withdrahral and repfenishnent on ') rates of 'U-fdn incorporati-on into DNA in CHO WT cell-s.

Time (hr) after L-glutamine tH-tdR incorporated into ü/ithdravral DNA expressed as ? of control,

0.5 572 l-. 0 402 1.5 242 2.5 25.52 ?< 322 rime (hr) after addition of L-glutamine to 2.5 hr starved ceLIs

The cells were pulsed for 15 min wiÈh 3H-ra* (¡ ¡:Cilplate; 22 Cí/n mol) at the specified intervals. For replenishment of nutrient, L-glutanine (2 mM) r¡as added 2.5 hr afÈer initiation of starvation. The control, expressed as 100t, is the amount of 3H-TdR incorporated into DNA by non-starved cells in 15 min. IABLE 7b

Effect of adenosine r,rithdrawal and replenishment on 2 rates of "¡l-tdR incorporation into DNA in CHO GAT celts.

Time (hr) after adenosine 3n-rd* incorporated Ínto withdrawal DNA expressed as ? of control

0.5 89.7s 1.0 81. 5? 2.0 30.4å 3.0 78 .22

4.0 Ì5 C

Time (hr) after addition of adenosine to 3 hr starved cells

1.0

The procedure for 3H-td^ (3 pCi,/p1ate) pulse Iabelling r.iras exactly as described for Tab]e 7a. For repJ_enish- ment of adenosine (0.1 mM), the nutrient was added 3 hr after initiation of starvation- 1,3 9

(14) Effect of L-cl,utamine or Adenosine withdravral

and ReplenishmenÈ on RNA Accumulation by CHO WT and CHO cAT- Cell_s

The consequence(s) of amino acid withdrawal on RNA metabolism in cultured ma¡¡¡nal-ian ce1ls remains paradoxical (see 'Historical'). However since it was observed (figs. 8, 9a and b, 10a and b) that nutrient avaiLability is closely relaÈed to changes in intra- celLul-ar HS3 levels, experiments vrere performed to determine : (i) the effects of nutrient starvation and replenishment on RNA synthesis by CHO ¡JT and CHO GAT- cells and (ii) if changes in RNA synthesis could be correlated with fluctuations in cellular HS 3 l-evels.

The resuLts in fig 14a show that total_ RNA accumulated , as measured by 3H-uridine incorporation, was 50? of control after 5 hr of glutamine deprivation. It should be noted that uptake of uridine was not affected during this time period. Upon glutamine 3H-uridine replenishment, incorporation of into RNA increased rapidly and paralleled that for control cultures. When CHO GAT- ce1ls were deprived of adenosine, as illustrated by f.íg. I4b, the accumulation of RNA r40

was drastically reduced when compared to the controLs. Upon replenishment of adenosine (0.1_ mtl) , Rt{A accumulation resumed at control- level-s after a short 1ag. The data clearly shor^r that there is an inverse relationship between HS3 levels and RN.A, synthesis, suggesting that the HS compound may be involved in the regulation of some aspect of RNA biosynthesis. Results to be presented in following sections srilI further support this hypothesis.

(15) Effects of L-Glutamine and Adenosine Deprivation

and Replenishment on the Grorrth of CHO WT and CHO GAT Cells

The data presented in the two preceding sections (13 and 14) of 'Resultsr showed an inverse relationship between IIS3 Levels and RNA and DNA synthesis. It Ì^ra s therefore of interest to determine if changes in gror¡rth rates during nutrient deprÍvation and replenishment. could be correlated with the above observations. The effects of glutamine starvation and its subsequent replenishment on the growth rates of CHO F¡T cells are ill-ustrated in fig- 15a. A sharp reduction t47

Figure 14: Effect of nutriênt starvation and replenishment (glutamine or adenosine) ? on -H-uridine incorporatj-on into RNA. ? -H-Uridine ( 3 ¡:Ci,/p1ate ) hras added to ceLl-s at zero time and samples taken

at times indicated. (a ) CHO 'vfT cells : ( G_{ ), minus glutamine;

( Eþ_-.{t ), plus gÌutamine; (e.---O ) glutamine (2 mM) added to glutamine- starved celLs at the time indicated by the arrohr. (b) CHO GAT cells: ( o_€ ), minus adenosinei ( EHl ), pfus adenosine; ( O---€ ), adenosine (0.1 mÌll) added to adenosine- starved ce1ls at the Èime indicated by the arrohr. (o) (b)

5 rf) þ

,a¿

z Él

,o

345 12 34

TIME (hr ) 143

in ceI1 proliferatj.on can be observed upon glutamine deprivationt ceI1 numbe¡s increased by only 10g in 8-10 hr after initiation of starvation and they remained constant thereafter. Upon replenishment of the cells vrith glutamine, the rate of gïowth, after a .Lag, increased. to the rate of controL cuLtures. Figure l-5b illustrates the results of a similar study done with CHO cAT cells. when staïved of adenosine, the growth rate of GAT cells decreased rapidly and by 14-15 hr, growth had ceased completely.

When the cells were re-supplemented with O.l mM ad.enosine, the growth rate increased quickly and reachecl that of the control culture.

Besides RNA and DNA synthesis, it is apparent nor,r that there is also an inverse rel-ationship betvreen HS3 leveLs (figs. lOa and b) and grov/th rates. t44

Figure 15: Effects of nut¡ient starvation and

replenishment (glutamine or adenosine ) on the grohrth of cultured marnmalian ce1Is. CeIl numbers were determined by trypsinizing the cell_s in each cuLture pl"ate (0.058 trypsin) and counting them with a Coulter counter. The arrovrs indicate the time at vrhich

either glutamine (2 mU) or adenosine (O.f mM) ¡4ras added Èo nutrient starved cultures. The expêrinental procedures f or grorilrth, washing and incubating the cells were as described in ,Methods'. (a) CHO WT cell-s: ( Q{ ), minus gtutamine; ( E]-{ ), plus gÌutanine; ( ), minus glutamine, plus glutamine. (b) CHO GAT cel-ls: ( H ), minus adenosine; ( E}_-{ ), plus adenosine; ( Ár--¿\ ), minus adenosine, pÌus adenosine. (bt (ol

rif) "l þ ,l oo .I E J z. .l o q C) I 2t t o rot5ao5 ro 15 20

T|ME ( hr ) I46

(16) Effects of AsÞIIg and cHo !ÍT Hs3 on Partially Purified cHo wT cell- DNA-Dependent RNA Pol-ynerases

The results hrhich have been presented (figs 10a and b, l4a and b) showed an inverse relationship between RNA accumulation and cel-lular HS3 l-evel, whereby an increase in HS3 is related to a decrease in RNA synthesis and vice versa. Recently LéJohn et. a]. (1978) reported that HS3 vias a potent inhibitor of the enzymic activities of Achlya DNA-dependent RNA polymerases. Similar studies were performed to determine the effect of purified HS3 on the in vitro activities of ¡nammalÍan DNA-Cependent RNA polymerases.

DNA-dependent RNA polyrneraçes r'ite re isolated from mid-Iog phase CHO $fT celJ-s grown in suspension culture. The DEAE-Sephadex A-25 chromatographic profile of the poJ-ymerases are shown in fig. 16. Peak B is RNA polymerase II' as sho¡ùn by its sensitivity to

I pg /m3. o(-amanitin (see table B), while peak A, which is resistant to the bicyclic peptide, is RNÀ polymerase I. RNA polymerase r and II vrere eluted from the column at 0.14 11 (NH4)2So4 and 0.25 M

(NH4 ) 2SO4 respectively. ]-47

Figure 16: DEAE-Sephadex A-25 chromatography ( I.S x 6 cm col-umn) of CfiO VÍT DNA-dependent RNA polymerases using a linear gradient (0.04-0.5 M) of ammonium sulfate. Fractions of 0.8 ml- were collected and enzymic activities determined as described in 'Methods'. Peak A is RNA polymerase I and peak B is RNA polymerase If. Froction Number I49

Table 8 illustrates the inhibitory action of Achlya and CHO WT cell HS3 on mammalian DNA-dependent RNA polymerase activities. RNA poll,.merase I and If activities 1^'ere inhibited by 75? and 89å respectively in the presence of 20 yg/ml of Achlya HS3.In comparison, HS3 from CHO WT celfs, at a final concentration of

16.6 ¡tg /nI , inhíbited RNA polymerase I and. II by 63å and 84? respectively. At highêr concentrations of HS3, the activíties of the polymerases vrere inhibited even further. In the case of RNA polymerase fI, the inhibition was 95ã in the presence of 41.5 ¡:g/m1 HS3. These results support the hypothesis that HS3 rnay regulate RNA synthesj-s, possibLy at the transcription leve l- .

(17) Effect of Purified Achl-ya HS3 on the tncorporation

of -H-UTP into RIIA by Permeabilized CHO WIt Cells

Since HS3 seems to inhibit the transcription apparatus of CHO WT cells in vitro, it was of interest to determine whether the effect could be manifested under in vivo or near in vívo conditions. For this TABLE 8

Effect of Achlya and CHO WT cell HS3 on the in vitro activities of DNA- dependent RNA polymerases from CHO WT cells. RNA Polymerase I RNA Polymerase II

Achlva HS3 ( uS,/mI ) B Tnhibition

0 + o(-amanitin 96 IO 0 0 0 1 l( -6 3 35 )1 5 48 43 l0 64 64 ¿U 75 89

cfiO I,üT HS 3 (¡¡glml- )

0 n 0 o. J 35 74 16. 6 63 84 41. 5 83 95

The assay procedure is described ín 'Methods'. The amount of 3u-Utp incorporated into TCA insoluble material in the abse¡ce of HS3 is expressed as 0g inhibition. o( -amanitin ( l- ¡rgrzml ) riras used ho élistinguish between RNA polymerase I and II (Please see tResults') . 151

permeabilized cells hrere used to assay for HS3 effect on transcription. The time course incorporation of 3H-UTP into

TCA precipitable material by penneabilized CHO tlfT cells, using an in vivo assay specific for RNA synthesis is

shown in fig. Il . The incorporation of labelted UTp was linear, at Least for the first hour. This time period was chosen for the subsequent studies on the effect of Achlya HS3 on RNA synthesis by permeabilized

CHO vlT cells. It was found that about 25A of the TCA insolubl-e material labelJ-ed in t hr was resistant to

NaOH hydrolysis i.e. non-RNA material . lVhen 100 ¡rSlm1 (fina1 concenÈration) of Ach1ya HS3 was added to the 'in vivor assay mix, the â incorporation of 'H-UTP into RNA by permeabiJ.ized CHO !ùT cells was found to be inhibited by 378*as shonn in table 9 . Higher concentrations of HS3 vrere not tested.

The in vitro (tabfe 8) and in vivo (table 9 ) inhibitory effects of HS3 on transcriptj-on suggest that this polyphosphoryJ.ated dinucleoside is probably involved in the regulation of RNA synthesis in cultured mammalian ceIls.

* Actually 49.38 considering that 259 of incorporated material is NaOH-resistant Figure 17: Time course incorporation of 3H-uúp into 10å TCA precipitable material by permeabiJ-ized CHO WT cel1s using an in vivo assay for RNA synthesis; 0.9 x L0" cells were used in the assay for each time point. fo I9¿ x E ()o

E o) o o o. ()oi .c fL F- +l) ff) TABLE 9

HS3 inhibition of RNA synthesis* in permeabilized CHO cef 1s.

HS3 concentration (Fg /nI) å Inhi bition

0 0

f00 37 (Average of

two expt. )

3H-urn incorporation into TCA insoluble materj-aI by permeabilized CHO celfs r^¡as measured as described 3tt-Utp in 'Methods'. The amount of incorporated into TCÄ insofuble material- in the absence of HS3 is expressed as 0å inhibition. A correction as described for fig. 14a & b, was made for the possible incorpo- 3u-urp ration of into DNA (pLease see pg B0 of 'llethods'). 155

(18) Effect of Antibiotics on HS3 Biosynthesis

The precedÍng results suggested that nucteic acial synthesis in these ceIl lines suffered from nutrient deprivation, whí1e HS3 accumulated under such starvation condition. Therefore, antibiotic inhibitors of protein and nucleic acíd synthesís rrere used to determi-ne srhat effect they might have on HS3 synthesis. The effects of puromycin (100 pg/ml), cyclo- heximide (10 pg,zml) and actinomycin D (1 Fglml) on HS3 synthesis by glutamine-deprived CHO l{T cells are shor\rn in fig. L8. Actinomycin Ð, was the most effective in depleting accumul-ated intracel-IuIar HS3. I{ithin t hr after its addition, rls3 level decreased to control guantities. Both cycloheximide and puromycin were also able Èo Io\^rer HS3 levels, although at a much red.uced rate. In the presence of glutamine in the grolrth medium, neither 10 ¡rg,/ml- actinomycin D nor cycloheximide had any effect on the inÈraceLl-u1ar

HS 3 leveLs as shor^¡n in table 10. 156

Figure 18: Effect of antibiotics on accunulated 32_. -pi-]abelÌed- HS3 during gLutamine starvation of CHO l,lT cells. The experi_ mental protocol was exactly as described in the legend to fig, 3. Antibiotics were added to gl-utamine- starved cells at the time indicated by the arrow. ( F-{ ), minus glutamine; ( EH ), plus glutamine; (H ), minus gÌutamine + actinomycin D (I pg/nl-); ( ¿\_1\ ), minus glutamine + cycloheximide

(10 ¡:g/mL); ( I-_-! ), minus glutamine + puromycin ( 100 Fg,h1). Time (hour) TABLE 10

Effect of cycloheximide and actinomycin D on HS3 leveLs in clto wr cells. HS3 (expressed as ? of control)

+ g lutamine 9.6 g l utamine 100.0 + glut,amine + cycloheximide 7.9 glutamíne + cycloheximide 52 .8

+ glutamine 10. 7 g lutamine 100.0 + glutamine + acti.nomycin D 9.8 glutamine + actinomycin D 8.3

The control, expressed as looa, is the amount of 32p-l.burled Hs3 present in ce]-ls after gLutamine deprivation. The time of incubation was 2 hr and 3 hr for the cyclohexi¡nide and actinomycin D experiments respectively. The antibioticsrat r0 ps/mL, were added at the start of the incubation perì-od. The experiment.al procedures were as clescribed in the J-egend to fig. 3. 159

(19) Survey of HS3 Metabolism by Various çuftured Mammalian CeIl Lines

A variety of cultured mammaLian ce11 l-ines have been tested, in the presence or absence of 2 nL¡4 glutamine, for the production of HS3. Tabte l_l surnrnarizes the results obtained, though some have already been presenÈed in preceding sections of the rResultsr. All permanent cell-s lines, except CHO GAT cells, were found to accunulate HS3 o\ding to glutamine withdravral. The reason for the lack of HS3 accumulation by CHO GAT ce1ls (al-so see table 3) will be presênted in the 'Discusslon'. In contrast to the above observations, tr,Jo anomalies are apparent from the results for Èhe primary ceLl lines. First WI-38 cells, a ce1I line of fetal origin (Hayflick and Moorhead 1961) incorporated onl-y loh¡ 1evels of 32pi into IlS3 upon gLutamine deprivation, This ís j-llustrated - I{S 3 by the ]o\,r ffi (cpm) ratio (table 11). Secondly, in spite of the presence of glutamine, all the rest of the primary ce1l lines surveyed incorporatêd substantial . _32 amounts ot pi into HS3. This can be contrasted L,ith 32p-l"beIled the HS3 accumulated by glutamine- supple- mented permanent cel-1 lines (see HS3 percentages and 160

HS3 ãñÞ (cpm) ratÍos given for: norlna1 foreskin fibroblasts and cHo vlT and cHo cAr cetLs). The Lower ä# (cpm) ratio for the glutamine supplemented human foreskin fibroblasts, for example, is due to a higher amount of Labelled cTP as compared to the gtutamine starved conditíon. Owing to the unique HS3 physiological response by primary cell lines to glutamine sufficiency, it became of interest to determine if the HS3 as detected in primary cett lines by one-dimensional chromatography (table 11), is similar to the HS compound isoLated from Achlya and CHO !ÍT cells. Two-dímensionaf chromatography and some chemical analysis were performed. Figuïes l-9a and I9b are two-dimensional- autoradiograms of formic 32p-1.bu11.d acid extracts from human foreskin fibroblasts incubated either in the presence or absence !¡c ? of glutamine. The ffi (cnm) ratios of 0.72 and 1.08 for plus and minus glutamine respectively are nearÌy identical to those obtained from one-dimensional chromatography (tab]e l-1). Further, the two- dimensional chromatographic profile of HS3 from human foreskin fibroblasts (figs. l9a and b) is sir¡il"ar to that obtained for CHO VìrT cells (f ig. 4 and table 5 ) .

Chemical analysis of HS3 isol-ated from Lesch-Nyhan human 161 fibrobl-asts indicate a 1:1 adenosine to uridine ratio (Èable 2). Thus the HS3 isol-ated from primary ceJ,l lines is probably similar, if not identical_ to the Àchlya and CHO I{T HS compound. t62

TABLE 11 32p-labe11ed Determination of HS3 in various mamnalian cel1 lines cultured in the presence or absence of 2 m¡4 L-gJ.utamine. I"a 3

+ gLutamine - glutamine

Permanent ceIl lines cHo !ùT _ 12.1 (0.045¡* 100 (0.683) CHO GAT 96.0 (0.049) " (0. 090) CHO HGPRT- (YH 21) 76 .7 CHO HGPRT /APRT (YHD 13) 14. I Bal-b 3T3 (mouse) ls. I BaIb SV 3T3 (mouse ) 10.5 PHK 2I/C! T3 15. 1 HeLa 24.7 ¿517 8Y 41.0 Prima 11 lines ) wr-38 -90.s (0.183) " (0.214) Lesch-Nyhan human fibrob l as t 9 4.3 Normal human fibroblast 99.0 Normal- human foreskin f ibrob 1a s t 9s.0 (0.68) " (1.07) African green monkey kidney cel1s l-06.0

1 t" "P-labeLled HS3 Level-, after 2 hr in the absence of gLutamine is expressed ) as 100?. - An average of values after 2-5 hrs in the absence or presence of glutamine.(Wt-39 only) . * parentheses Hs3 values in are ratios of õlp- (cpm,l . The experimental procedures and the quantitation of HS3 were exactly as described in the legend for fig. 3 and 'Methods ' . 163

Figure 19: Tvro- dimensional- autoradiograms of 1 M formíc acid extracts of 32pi-rabeLLed human foreskin fibrobLasts ( a primary cell líne) cultured for 2 hr in growth mediun (a) with glutamine and (b) lvithout glutamine. The experimental procedures hrere exactLy as described in the legend to fig 3. Two-dimensional- chromatography \^¡as as described in legend to fig. 4. The ratios (cpm) are : plus "f åH glutaminê = 0.72; mì.nus glutamine = L.08. )64

ì..,..:,Í ;

------) I 155

DISCUSSION AND CONCLUSION 166

(1) HS3 Synthesis by cHO vfT Ce1ls

Only one of the fungal HS compounds, HS3, ¡,ras found to accumulate hrhen glutamine was r4¡ithdrawn from the gror^rth medium of CHO WT cells (fig. 3). The absence of HSl and HS2 cannot be compLetely ruled out. If they are present, their intracellutar 1eve1s, at best, are not affected by glutarnine availability. After a 4 hr glutamine starvaÈion period the HS3 concentration in CHO WT cells was estimated at 0.12 mM (Lewis et. aI. L977) or 216 FSlml cel1 volume assuming the ¡fr.¡ of flS3 as 1800 (McNaughton et. al. 1978) . Obviously the estimated concentration represents a minimum value as the calcul-ation did not take into consideration the per.centage recovery and the lability of the molecule during the isolation process. Physical and chemicaJ. analyses strongly suggest that the mam.malian HS3 is essentialty identical to its fungal counterpart (tabte 2). At no time during the course of this study vtas procaryotic ppcpp or pppcpp ever detected in mammal-ian cells. Their presence could have been easiÌy detected chrornatographi ca1ly in one- and two- dimensionaL analyses on thin layer pEf-cetl.ulose I67

plates (table 1). Furthermore, published data argue against eucaryotic cell production of procaryotic ppGpp and pppcpp (Smulson 1970; Mamont et. al. 1976; Thammana et. a1. 1976r Martini et. al. !977 ¡ Richter anil Isono 1977).

l2) HS3 AccumuLation: The Possible Involvement of Other Nutritional Chánges Besides Glutamine Ðeprivation

As had been reported (fig. 3) , gfutamine deprivation resulted in the accumulation of Hs3 in mammalian ce1ls. Ho\"rever, in stringent bacterial strains, ppGpp and pppcpp r^re re produced in response to a l-ack of various essentiaL amino acids (Cashel and caflant 1969; Cashel and Gallant 1974; Gallant and Lazzarini 1976). 3ut this phenomenon does not exist in cultured mammaLian cells with respect to HS3 accumulation. Vlhen CHO WT ceJ-ls were deprived of another grovith- essential amino acid, isoleucine (Ley and Tobey 1970) , an accumulation of intracellular HS3 hras not observed (fig. 6). This suggest that HS3 synthesis may not be related to a simple case of starvation for an essential amino acid. 158

ohring to the importance of glutamine as a precursor for de novo purine biosynthesis (Mahler and Cordes I97J,) , it became apparent that HS3 accumul-ation may possibly be related to a decrease in the synthesis of new cellular purines. This was substantiated through the use of tvro potent inhibitors of de novo purine biosynthesis. Azaserine, a glutamine analogue, specifically inhibits the amidotrans ferase invol-ved j_n the conversion of formylglycinamide ribonucleotide to 5-aminoimida zole ribonucleotide by forming a covalently bound en zyme- inhibi tor complex (French et. aI. 1963). IIITX is a strong reversj-bIe inhj_bitor of dihydrofolate reductase (EC l-.5.1.3) (Blak1ey 1969). The subsequent formation of 5, lo-methyl-enetetrahydrofolate , the active coenzyme requj-red for purine biosynthesis (Mahler and Cordes 1971) wouLd thus be inhibited. V'rhen these inhj-bitors were added individually to cells incubated even in the presence of glutamine, HS3 accumuLation was evoked (table 5). A CHO GAT auxotrophic mutant was subsequently utilized to further define HS3 metabolism. The de novo biosynthesis of purine nucleoÈides and thymidylates are blocked ín this mutant owing to a defect in folate r69

metabolism. IIol^rever its growth can be supported by supplementj_ng o< -MEIrf bTith adenosine, thymidine and glycine (McBurney and l4lhitmor e L974). Since the previous data supported the hypothesis that a stimulation of HS3 synthesis is reLated to a btock in purine biosynthesis and possibly a lack of nucleotides, it was reasoned that the removal of adenosine from the growth supporting medium of CHO GAT cells should lesul-t in an accumulation of HS3 as this nucleoside is the sole source of purines. Indeed regardless v/hether or not glutamine and thymidine r^re re present, there was an accumulation of HS3 in this cell- fine as long as adenosine hras withdlawn (tab1e 3). Owíng to this strong link between HS3 production and an aberrant ¿le novo purine pathway, CHO WT and CHO purine salvage pathway mutants were used to determine if an exogenous purine supply could suppress the accumulation of HS3 during nutrient deprivation. The Lack of hypoxanthine/guanine phosphoribosyl_ transferase (IlGpRT) has been associated with the Lesch_ Nyhan syndrome, a human neurological disorder (Seegmilì.er et. al 1967'l . This enzyme is of utmost importance to tissues and ceJ-Is h'hich cannot biosynthesize their own 170

purines and are dependent on an exogenous purine source

like hypoxanthine. The possibl-e impottance of HGPRT during mammalian developnìent was dj_scussed recently

by Adams and Harkness (1976). CHO mutants, yHD 13 and YH 21, lacking HGPRT (Chasin, privaÈe communi- cation) were found to be unable to utilize high concentrations of exogenously supplied hypoxanthine, inosine and guanosine to suppress the accumulation

of HS3 during glutamine deprivation white CHO WT cel-ls could (tables 4 and 5). The reason why inosine and guanosine could not be utitized to suppress HS3 accumulation is because there are no knov,rn eucaryotic nucleoside kinases which can catalyse the conversion of j-nosine anil guanosine directly to their mono- nucleotides (Friedmann et. a1. 1969). Thus these nucleosides can only be converted to mononucteotides via HGPRT after their catabolism to the respective bases by nucleoside phosphorylase (Milman et. a1. 1976). This reaction is generally believed to function in vivo in the direction of nucleoside breakdown to the base and ribose- I-phosphate (Milman eÈ. aL. 1976t

Li and Hochstadt Lg76). Therefore, onÌy CHO WT cells with functional HGPRT were able to utilize the bases and nucleosides. 17I

The observed depletion of IIS3 by adenine, adenosine and deoxyadenosine (tables 4 and 5) requires some clarification. Adenosine is readily converted to iÈs nucLeotide by adenosine kinase (Chan et. al. 1973), while deoxyadenosine can be catabol-ized to adenine by an enzyme found in manmal-ian tissues (Snyder and Henderson 1973). The purine base may then be salvaged by adenine phosphoribosyltransferase (APRT) to A¡4p. However the effectiveness of adenine in preventing HS3 accumulation by YHD 13 cel-ls is perplexj-ng since

.it is known to be deficient in both HcpRT and ÀpRT (Chasin, private communication). But yHD l-3 nust possess some residual APRT activity since it could 'tÀ incorporate -'C-adenine into various nucleotides (see fig. 7). It is pertinent to poinÈ out that all the purine compounds tested are readily utitized for adenine and guanine nucleotide synthesis by mammalj,an ce1ls (fig.7, Cook and Vibert 1965; Rapaport and Zamecnik l-976). Studies by T,éJohn et. aI. (1978) have shown that \^rith the exception of inosine and deoxynucleotide s , those purines tested on mammalian ceIls (table 5) hrere also active in inhibiting the accumulation of HS3 by nutrient- starved Achl-ya. It is possible that the ineffectiveness of inosine and 172

deoxyadenosine in the fungal_ system may be due to the absence of inosine phosphorylase (MiLman et. aI. I976) and the deoxyadenosine to adenine converting enzyme (Snyder and Henderson 1973). An altetnative to examining the relationship between HS3 accumulat.ion and the supply of exogenous purine compounds was to block in vivo purine catabolism. Allopurinol, a hypoxanthine analogue commonly used in the treatment of gout, is a potent inhibitor of xanthine oxidase (EC 1.2.3.2), an enzyme which catalyses the degradation of hypoxanthine and xanthine to uric acid (Pomales et. a]. 1963) . The anafogue was found to be marginally effect,ive, at the concentration used, in inhibiting HS3 accumul-ation in glutamine- starved CHO IfT cells (table 5). It sras surprising to find that the pyrinidines

S-fluorouracil (5-FU), uridine and cytidine r^rere effective inhibitors of HS3 accumulation while thymidine was vrithout effect (table 5r LéJohn et. aI. 1978). The possible role of 5-FU in this pheno¡nenon r^ri1l be considered in detail in a later section of this rDiscussionr. ft is not clear how uridine and cytidine affect IlS3 accumul-ation. Recently Skaper et. al. (1976) Ìeported that PRPP levels in human lymphoblasts lrere r73 decreased by glutamine withdrawaf from the growth medium. Since boÈh uridine and cytidine are potent inhibitors of aspartate transcarbamylase (Bresick 1963) , these two nucleosides may spare the consumption of PRPP at the orotate phosphoribosyltransferase step of d.e novo pyrimidine biosynthesis. Altetnatively uridine coul-d be exerting its pRpp-sparing effect by its j.nhibitory action on orotate phosphoribosyÌtrans- ferase activity (Hoogenraad and Lee 1974). This may be an explanation for the lower potency of cytidine (Èable 5) owing to the necessity of its deamination to uridine to exert its effect. Whatever the mechanism by which a PRPP-sparing effect is achieved, the greater availabiliÈy of pRpp should result in an enhancement of the rêutili.zaÈion of purine bases for nucLeotide biosynthesis. This and oÈher proposed mechanisms related to the inhibition of HS3 synthesis by various purines and pyrimidines should be considered specuìative at the monent. LéJohn (unpubì.ished) has suggested that (a) the I¡S3 molecule may act as a high-energy polyphosphate storage compound during nutritional stress; (b) the so-called inhibition of IlS3 biosynthesis by various exogenously supplied purines is in fact due to the consumption of the high energy phosphates of HS3 for nucleotide biosynthesis and (c) the constituent ADp and UDP moieties could be further used for the replenishment of all the nuc.Leotide pools. The data discussed support the hypothesis that a decrease in intracellular purine nucleotides may be responsible for evoking the HS3 response in mammalian ceffs.

(3) Effect of clutamine Deprivation and l"tetabolic Inhibitors on GTP and ATP Levels

The hypothesis described above is further supported by data obtained from measurements of cTp and ATP ce1lul-ar pools in ce1ls cultured under conditions r"Thereby IlS3 is known to accumulate. Under glutamine deprivation, an increase in HS3 resulted in a correspondlng decrease in GTp pools (figs. 5a and b). .' Hs3 ft is important to note that variations fn rne õlÞ rartos !'rere observed from experj-ment to experiment for each of the cell lines studied. Hoh¡ever, the average ratio was around 0.7 to 0.6 for a 2 hr experiment. These r75

varíations may be due to subtle differences in the nutritional state of each batch of celfs owing to the possibility of diffe¡ences in the amount of free amino acids in each batch of dialysed fetal calf serum. The levels of HS3 and GTp are indeed sensitive to slight fluctuations in the glutamine concentration (figs. 5a and b).

With respect to ATp poo1s, treatment of CHO WT cel1s with Èhe de novo purine biosynthesis inhibitors, azaserine or MTX (French et, al. 1963; Bj-akley 1969), or glutamine deprivation resulÈed in a 10-308 decrease

in ATP concentrations wj.thin a 2 hr period (table 6) whiLe HS3 accumulation was rapidly occurring during this time (fig. 3, table 5). A 308 decrease in ÀTp concentrations, within 2 hr, has also been observed for amino acid-sÈarved Ehrl-ich asciÈes tumour ce11s (live anil Kaminskas 1975; Grurnmt and GrummL 1976). 14c-adenine .Also incorporation sÈudies, using yHD 13 ce1ls, support these observations as the generation of ATP from very l-ow concentrations of exogenously 1t suppì.ied -'C-adenine \^7as completely inhibited during gJ-utamine starvation (fig. ?).

Thus the inverse relationship betr^/een HS3 176 levefs and purine nucleoside Èriphosphate concentrations is clear. The decreased leveLs of these nucleotid.es is probably due to a combination of a lack of the generation of the triphosphates and an inhibition of the synthesis of new purine nucleotide precursoïs. This may be further compounded by the possibility that some of the intermediates of the de novo purine biosynthetic pathway may be shunted off for HS !:io- synthesis (LêJohn et. a1, 1978).

(4) HS3 Metabolism: Effect of Antibiotics

Puromycin, cycloheximide and actinomycin D are al1 potênt. inhibitors of eucaryotic ce11 metabolism. The principal mode of action of ptrromycin and cycJ-oheximide is the inhibition of protein synthesis while actinomycin D inhibits RNA synthesis by forming complexes vrith DNA ( GotÈleib and Shaw 1967; Kersten and Kersten I97 4) . Recent studies have shown that ppcpp concentrations l^rere restrained to basal levefs when chloramphenicol , a procaryotic protein synthesis inhibitor, was added to amino acid starved E. coli I77

(Cashel 1969; Lund and KjeJ.dgaard I9721 . In comparison, actinomycin Ð, cyclohexímide and puromycin, in decreasing

order of potency, inhibited and even lor,rered the HS3 accumulated. during glutamine deprivation (fig. 1B). Thê depletion of the accumulated HS3 by these antibiotics may be reLated to a sparing effect cf nucleosid.e triphosphates sínêe it r4/as suggesÈed that the accumuLatíon of HS3 rnay be evoked by a lack of purine nucleotides (tables 3, 4, S and 6). This is interesting since it has been suggested by Grunìnt and Grunmt (1976) that the resumption of RNA synthesis following cycloheximide treatrnent of amino acid starved cells may be related to the expansion of ATp and GTP pools to l-evel-s comparablé to non-amino acid starved cultures. AlÈernatively, these antibiotics may lower accumulaÈed HS3 by either direcÈly stimulating its degradation, inhibiting its biosynthesis, or both. This may be accomplished through the effect of these antibiotj.cs on the synthesis of various 'enzymes' involved in HS3 metabolism. Hor4'ever, this is purely speculative as no supportive evidence ís currently available. One interesting feature related to these antibiotic studies has been the observation that 178

treatment of amino acid. starved eucaryotes with protein synthesís inhibitors has consistently resulted in a restoration of RNA synthesis (Hershko et. al . 1971, Foury and Goffeau 1973; Grummt and Grurnmt l_976; Gross and Pogo I976i Pogo and Zbtzezna l_978). This paradoxical response called'phenotypic relaxednessl ( Grummt and Grummt 1976) is observed only when cyclo- heximide is supplied to cells under conditions of metabolic stress. It is têmpting to suggest that this paradoxical tesponse may be related , in some way, to the cycloheximide effect on HS3 accumul-ation during glutamine deprivation and the possible role of HS3 in RNA synthesis (tabLes I and 9; fig. I4a) .

(s) HS3 Levelst Elevation by MîX Treatment or Glutanine Ðeprivation and Depletion by 5-FU

I*IethoÈrexate (l'lTX), a folic acid analogue and

5-fluorouracil- (5-FU), a thymine analogue are conunon drugs currently used either separately or in combination to treat a varieÈy of neoplasms in man. MTX is an inhibitor of dihydrofolate reductase (B1akley J.969) and thus rhe addition of this drug to L79

cel-ls ü/oul-d ultimatety result in an inhibition of the

biosynthesis of certain amino acids (Blakley 1969) , purines (Mah1er and Cordes 1971) and thymldylates (?fahba and Friedkin 1961) ovring to a lack of reduced folates. High concentrations of ltTX wiIl also directly inhibit thymidylate synthetase (EC 2.1-.I.45) (Borsa and Whítmore 1969), an enzyme involved in the bio- synthesis of thymidine monophosphate from deoxyuridine monophosphate. The cytotoxicity of 5-FU is generally ascribed to thê inhibition of thymídylate synthetase by iÈs derivative 5- fluorodeoxyuridine monophosphate (SFdUMP) (Heidelberger 1975). In the absence of its cofactor, 5, 10 methyl-enetetrahydrofolate (I^lahba and Friedkin 1961) , thymidylate synthetase binds poorly

to sFdUllP. In its presence, the enzyrne forms an irreversible complex with it and sFdUI{p (Santi et. aI. 7g74t. Thus it beca¡ne apparent that the cytotoxicity of the fluorinated pyrimidine analogue may be antagonized by MTX orring to the lack of reduced folates (Ullman et. a1. L978). The actual mechanism(s) for this antagonism remains Èo be resolved. It is also likety that the other nucleoside and nuc]-eotide derivatives

of 5-FU may affect other enzymic activities and RNA metabolism as weIl (Heidelberger 1975). t80

Or"ring to these intriguing observations, the effect of MTX and 5-FU on HS3 metabolism may be of some significance. It has been reported that glutamine deprivation or MTX treatment of CUO WT and L5178y cells resulted in HS3 accumulation (figs. 3, 9a, 12a,b,c and table 5) and a cor¡esponding decreâse in purine nucleoside triphosphate pools of both celL Lines (fiqs. 5 and 7; tabi-e 6; Hrynj.uk et. a1. 1973) and also dGTP levels in L5178Y cell-s (Tattersall and Harrap 1973). This aspect that a decrease in puríne nucleotide biosynthesis and/or pools is refated to HS3 accumulation has already been discussed j-n previous sections. The results shohred that 5-1U, at 0.3 In}I, inhibited the synthesis and even depleted the HS3 accumulated by CHO VIT and L5178Y ceLts during glutamine starvation or MTX treatment (figs. 12a,b and c). This phenomenon may be related to a purine sparing effect previously observed for 5-FU (TattersaLL et. a1. 1973). This may be achieved in at least three ways: firstly, by the inhibition of thymidylate synthetase (santi et. al. 1974) and thus sparing the consumption of dUMp, whose biosynthesis consumes ATp (Mah1er and Cordes I97I) i and secondly, DNA synthesis in nutríent- s tarved or I81

.!,1Tx-treated cells may be further inhibiÈed due to the lack of dTTP, and again sparing the consumption of the other deoxynucJ-eoside triphosphates. This has indeed been demonstrated for L517gy celLs by Tattersall and Harrap (1973). Third1y, 5-FU itself or one or more of its metabolic derivatives may be invotved in further blocking cellu1ar RNA synthesis (Heidelberger 19751 . Alternatively, rather than due to a nucleotide sparing effect, the effectiveness of 5-pU in depj_eting

HS3 pools may be related, in some yet unknoeln r^lay, to the unusual uracil-containing end of the HS3 molecule (table 2; Goln and LéJohn I97i.). one interesting observation from these studies is the fact thaÈ in CHO I^IT ce11s, the rate of HS3 depletion by 5-FU was far greater for glutarnine- starved cultures hThen compared with MTx-treated ones (figs. 12a and b) . The greater inhibition of purine nucleotide biosynthesis by MTX over glutamine deprivation is unlikely as l,lTX treatment, even at l0 ¡:M, depteted .ATP level-s by only 128 as compared Eo 2OZ for glutamine starvation (table 6). A plausibLe explanation may be related to the work of Santi et. al. (1974) , !'rhere they found that in the absence of reduced 782

folates, 5-FdUMP bound poorly to thymidylate synthetase. Thus the greater availabilíty of reduced folates during glutamine deprivation than MTX treatment would lead to a greater purine nucleotide sparing effect by 5-FU. The fact that the depletion of HS3 in MTx-treated L5178y cells by 5-FU \^ras very rapid when contrasted with similarly treated CHO WT cells (figures 12b and c) may be related to the tên-fold increase in the supplemented fol-ic acid in the growth medium of L5l78Y ceLls than in Èhe o{ -¡l¡¡l used. for culturing CHO cells (Fisher and Sartorelti 1964i Stanners et. al. t97r) .

Though the resulÈs discussed here may have some relevance to the vrork reported by Ullman et. aI. (1978), the physiological role (s) of 5-FU in depleting HS3 remains obscure,

(6) HS3 as a Possible Cellufar Regulator: Changes in Rate of Accrmulation and pooL Size of HS3

In bacteria, the deprivation of an essential- amino acid evokes a cellular phenonenon known as 183

the 'stringent responser (Stent and Brenner 196l) , r,íhereby numerous ceIlular processes are blocked rStringent (see Response' in historical). ppcpp and pppGpp have been implícated in regulating these diverse metabolic processes. The abiliÈy of bacteria to rapidly alter their intracellutar concentration of ppcpp in response to abrupt changes in the nutritional environment is a tremendous physio_ J-ogical asset (Cashel 1969; Gallant and Haracta 1969; callant et. al. 1970; Fiil et. al. 1972; stamminger

and l,azzarj,ni 1974; Cashel 1975). Data have been presented (figs. 8, ga and b,Loa and b) demonstrating the rapid changes in HS3 metabolism during transient nutritional imbalance¡ The parallelism of HS3 metabolisn with ppGpp metabolism suggest that this eucaryotic

dinucleoside polyphosphate may be involved in some aspect of the regulation of eucaryotic cel_j- grol^,th. The data from figure lI suggest that, at j_east in CHO WT ceLLs, HS3 degradation is very sLow during glutamine starvation. Thus it is likely that the inc¡ease in HS3 concenÈrations during glutarnine starvation may be the result of an acceleration in HS3 biosynthesis (figs. I and 9a). L84

There is no data, at present, to support a mechanism(s) responsible for the rapid depLetion of accumul-ated HS 3 f oll-or^ring the replenishment of the omitted nutrient (figs 10a and b). However, the rapid fluctuation in the intracellular levels of HS3 suggest that changes in the biosynthesis and degradation and/or utilization of the different moieties of the dinucleoside polyphosphate must be operating in the eucaryotic cel1.

(7) Intrace]Jular HS3 Levels: Correlation lvith Proteín, DNA and RNA Syntheses and crowth

(a) Protein Synthesis

Based on the däta obtained (figs. 9a and b; 10a and b; 13a and b) , it can be concluded Èhat the rate of protein synthesis is independent of the intra_ cellular leveLs of HS3 in mammalian celLs. This is in agreement r.rith the results reported for fungal cells by léJohn et. al-. (1975). The drastic inhibition of "H-leucine incorporaÈion into proteins by glutamine- deprived CHO WT cells is probably the resul-t of a deficiency in glutamine and purine nucleotides for protein synthesis rather than related to some in vivo effect 185 of HS3 on the protein synthetic process. ?his is in contrast to the observed correlation beÈr,reen an increase in intracellular ppcpp and a decrease in protein synthesis in procaryotes (Stent and Brenner f961; Cashel 1969) r and where in vitro studies (yoshida et. al. 1972) have shoh'n that the formation of the protein synthesis init.iation complex was inhibited by pPcpp.

(b) DNA Synthesis

DNA pulse label-l-ing studies (tables 7a and b) demonstrated an inverse relationship between the rate of ÐNA synthesis and HS3 concentrátions (figs. 9a and b; 10a and b). A similar correlation was observed during the grovrth cycle of the fungus, AchLya (Lewis et. al. I976). Further, it was reported that the activity of ribonucleotide reductase, an enzyme invol-ved in the conversion of ribonucleoside diphosphates to their deoxy-derivatives (Larsson and Reichard 1966), decreased as HS3 levels increased (Lev,¡is et. aI. l-976). Subsequently it was shoqrn that HS3 potently inhibited the ribonucLeotide reductases from CHO V,¡T cells and Achlya, while having 186

no significant effect on the E. coli enzyme (Lewis et. a1. 1976; lewis et. al, 1977). The inhibition by HS3 was found to be non-competitive with respect to the substrates (],ewis et. al. 1,977). The pool sizes of the various deoxynucJ-eoside triphosphates in a variety of ce11 types have been shor4rn to be very small and without replenísh- ment are sufficient.to support DNA synthesj-s for only short periods of tíme (Reichard J_972; VtaJ.ters et. a1. 1973). Therefore, Èhe rapid decline in DNA synthesis during glutamine (Ley and Tobey 1970; pardee 1974;

table 7a) or adenosine (table7b) deprivation may be rel-ated to a self-j.nposed starvation of celtular deoxyribonucleotides by the action of HS3 on the ribo- nucLeotide reductase.

(c) RNA Synthesis

' Stable RNA accumulation by bacteria during the 'stringent response' is severely restricted (see iStríngent Responser in Historicat). paraLlel_ studies have been employed to determine if a similar phenomenon occurs in mammalian cells during metabolic stress.

Hovrever t.he results reported ( see '¡.tammatian Cell-s : t87

RNA Biosynthesis and Amino Acid Vtithdrawal , in Historical,) does not allow a definite conclusion to be reached regarding. the biochemical consequences of amino acid ?rithdrar.ral in cultured ma¡nmalian ceLls. f t is 1ikety that a factor contributing to thêse conflicting resu]-ts is the use of different starvation protocoLs (eg. different amino acids, different growth medía, Length of starvation, the use of non-dialysed serurn in some cases etc. ) . Studies wiËh CHO WT and CHO GAT cel]s showed "H-uridine? that incorporation into RNA during nutrient starvatj-on was inhibited, but ¡esumed without any appreciable lag at control levels upon nutrj.ent replenishment (figs. l4a and b ). ft is pertinent to 3¡t-uridine point out that the decrease in incorporation into RNA during nutrient withdrawal was not the result of a decrease in uridine upÈake. Similarly, the uptake of uridine by Achl-ya (Lé.rohn êt. at-. l-978) and human l-iver cells (Botcsfoldi et. aI. l-971) $,as also unaffected during glutamine deprivation. Because of these in vivo observations, in vitro studies were performed to determine $¡hat effect isolated HS3 has on the activities of the various DNA_ dependent RNA poLymerases f rom CHO V,fT cell-s. 188

The isolation procedure of these DNÄ-dependent RNA polymerases from CHO WT ceJ-Is, as described in rMethods', was essentialJ.y according to the young and Whitely method (1975). The salt concentrations at which RNA polymerases I and II were eluted out (fig. 16) are in agreement vTith those isolated from rat liver

(Lindell et. al. 1970). peak B was identified as R¡¡A polymerase II, as shown by the enzyme's sensitivity to its specifíc inhibitor, o( -amanitín (Lindell et. al. 1970). RNA polymerase tfÏ, which is el-uted by 0.45 tl ammonium sulfate (Young and Whitely 1975) was found in very small quantities. RNA polymerase I is implicated in the synthesis of ribosomal RNA. while RNA polymerase II is general-Ly believed to be invol_ved in the synthesis of heterogenous nuclear RNA, part of which is probably precursor messenger RNA (Chambon 1975). RNA poLymerase III probâbly catalyses 55 and pre-45 RNA syntheses (Chambon 1975).

Interestingly, both Achlya and ¡na¡nmatian HS3 were found to be equally potent in inhibiting, in a concentratÍon dependent manner, the in vitro activities of RNA polymerase f and Ir fron CHO vtT cel-Is (tabte 8). Studies with Achlva RNA polynerases have shov,¡n a similar 189

HS3 inhibitory effect (LeJohn et. al, 1978). The mechanism of the in vi-tro inhibition of eucaryotic RNA pofymerases by HS3, however, remains to be elucidated. These in vi,tro studies \4rere extended to a t whol-e- celLr nucleotide-permeable system (Lewis et. a1. 19?8) used in conjunction hrith an assay specific for RNA synthesis (castellot et. aI. 1978). HS3 vras also found to be a fairly potent inhibitor of RNA synthesis in this intact, permeabílized maÍìrnalian ce11 system (tabIe 9). The concentration of HS3 used in this inhibition study was only approximately one-ha1f its estimated physio- logicaJ. concentration in 4 hr-glutamine- starved CHO rdT rHS3 cel,ls (see Synthesis by CHO WT Cellsr of 'Discussion').

The inhibitory action of the fungal and mamnalian HS3 on mammalian DNA-dependent RNA polymerases is another exampLe of the similarity between the HS3 molecules from both eucaryotic organisms (tab1e 2). Based on these in vitro and in vivo data, it is suggestive that HS3 may be of physiologicat importance in regulating some aspect of RNA synthesis in eucaryotes. 190

(d) Growth

Si-nce. it was observed that the biosynthesis of nucleic acids in CHO cel1s was severely restricted on nutrient starvatj,on, experimenÈs were performed to determine r"rhat effect nutrient starvation might have on gror^rth. IÈ rrra s observed that growth was severely restricted on glutamine (fig. 15a) and adenosine (fig. l5b) starvation in CHO WT and GAT celts respectively. The data are consistent r,rith those previously reported by other investigators (Eag1e et. aL. J_955; Ley and Tobey 1970; McBurney and Whitmore I974). The residuaf increase in celt numbers (figs 15a and b) , despite nutrient deprivation and the accompanied inhibition of RNÃ and DNA synthesis may be explained by the fact that those cells in the non-synchronous cell population which are betvrêen the end of G2 and M phases of the ce1J. cycle (Howard and peLc 1953) may stij-t escape the block and proceed through mitosis. this is possible since DNA synthesis, which is restricted to the S phase, and RNA synthesis are essenÈiaIly absent during late c2 and mitosis (M phase) (Mitchison 197I). It is apparent that gJ.utamine deprivation of L91

CHO WT cells is more effêctive in inhibiting grov¿th $'hen compared to adenosine starvation of CHO GAT celLs_ This rnay be related to the fact that proÈein synthesis, which continues unabated during mitosis in Chinese hamster celIs (Taylor t960), sras strongly inhibited . in glutanine- starved CHO I{T celLs while remaining unimpaired in adenosine-deprived CHO GAT-ceIls (figs.

13a and b) .

(8) Possible Effects of HS3 and Nucleoside Triphosphate Levels on NucLeic Acid Synthesis

Results have been presenÈed which showed that during specific nutrient deprivation, a high intra- cellular HS3 leve1 is associated with decreased purine nucleoside triphosphate pools, inhibition of growth and the reductj-on of RNA and DNA synthesis. However, rdhat is difficulÈ to resoLve, based on the avaj-l-able data, is hthether HS3 or ATp and GTp levels are

directly involved in regulating, in vivo, RNA and DNA synthesis through their effects on the DNA-dependent RNA polymerases and the ribonucleotide reductase during nutrient deprivation. On the one hand, both ATp and cTp pools were found to decrease during gfutamine sÈarvation; 192

a 2 hr starvation period led to a 20? decnease in ATp levels (table 6). Thus the lor¿ering of these essential nucleotides may be directly responsible for the observed inhibition of RNA and possibly DNA synthesis. However Jolicoeur and Labrj.e (1974) reported that during a 2.5 hr glutamine starvation of Landschutz ascites cel1s, only 455 pre-ribosomaf RNA synthesis r,ras pre- ferentially inhibited (about 608) while 4s, 5s and polysomal mêssenger RNA synthesis was not significantly affected. This observation sÈiLL does not exclude the possibility that separate nucleoside triphosphate pools for ribosomal- and messenger RNA synthesis exists ( Grunmt and crunmt 1976). As there is no evidence to

suppolt this hypothesis, the in vitro effect of HS3 in inhibiting RNA polymerase f activity may therefoïe be of physiologíeal importance. On the otheï hand, the significance of the ínhibition of RNA pol-ymerase II by HS3 (table 8) is unclear as in vivo sÈudies (Jolicoeur and Labrie 1974) demonstrsted that messenger RNA synthesis was not affected by glutarnine starvation. It is possible that in vivo, HS3 mey have

a quantitatíve1y different effect on R¡.JA poLymerase II activity during metabolic stress. However this is 193 purefy speculative and must await other in vitro studies, for example using native DNA and DNÀ of kno$rn coding f,unêtion as templates and the measurement of the rate of biosynthesis of specific messenger RNAS. The possible effects of decreases in ATP and GTP pools on DNA synthesis during nutrient deprivation is obscure. It has been suggested (ELford et. a1. 1970; Cory and l{hitford 1972) that ribonucleotide reduction is the rate-limiting step for DNA synthesis. The substrate specificity and the rate of reaction of ribonucleotide reductase are regul-ated in a complex, allosteric fashion by the levels of ATP and other deoxyribonucleoside Èriphosphates (Reichard J-972¡

Elford 1972). Also d.A,TP acts as a. general feedback inhibitor of the enzyme (Larsson and Reichard 1966). It \,tould be futile to attempt Èo speculate on the consequences of decreased cellular ATP and GTP contents on the ribonucleotide reductase activity since not much information is avaifabfe tûith regard to the absolute pool sizes of the various ribo- and deoxyribonuc leotide s in ce11s under condiÈions which would result in HS3 accumulation. Vùhat is clear, however, is that in vitro HS3 rirhen compared with dATP hras found to be a more potent inhibitor of the reductase (Lewis et. al. 1977\. !94

(e) Glutamine Availabilitv and HS3 Metabolis¡n by Various Cultured MaÍunal-ian Ce11 Lines

Glutamine withdrawal studies pertaíning to HS3 metabolism by various CHO ceLl lines were extended to other cultured marnmalian ce11 lines. In alI cases (tabl-e 11) only one of the fungal polyphosphoryl-ated dinucLeosides (ìfcNaughton et. al. 1978), HS3, vras detected. The data suggest that (a) the existence of HS3 ís widespread and probably exisÈs in most, if not all,cu1tured rnammalian cel1 l-ines and (b) the absence of HSL and HS2 in ¡nammalian ceIls may be reLated to an intrinsic physiological difference between them and fungal cells; namely that all three HS compounds rnay be somehow involved in the fungal sporulation process (LéJohn et. af. L978). The removal of glutamine from the growth mediumresulted in an accumulation of HS3 by aII but one of the permanent cell li-nes. The faj-lure by CHO GAT-ceLls to accumuLate Hs3 is nost 1ike1y due to the presence of adenosine in its grohrth medium. The implications of this observation have already been discussed. The primary ce11 lines with the exception 195

of wI-38, exhibited ä similar physiotogical response as pêrmanent ones in that they accumulated HS3 upon gl-utamine starvation (tabì-e 11) . 32pi The lack of label-f ed HS3 in l.tr-38 cells even after 5 hr of glutamine deprivation is perplexing because there r,ras a significant incorporation of 32pi into other intracellutar nucleotides. It i_s not clear whether this phenomenon manifested by WI-3g cells is related to Èheir IÍmiËed in vitro proliferative potentiat and embryonic origin (Hayf]ick and Moorhead 1961; Hayflick 1965), since none of the other cel,l- lines surveyed possess both of these characteri stics . Alternatively, it may be possibLe that glutamine sufficiency is attained by the rapid induction of glutamine synthetase (EC 6.3.1.2). However this appears unlikely considering the short period of starvation (2 hr) Besides, its been reported that glutamine- starved !{I-38 ce11s did not r.=rr.. growth even when a high levet of glutamine synthetase was induced (Viceps and Cristofalo 1975). A survey of other primary embryonic celI lines should suggest vrhether or not the lack of 32ni an accumulation of Iub"Iled HS3 during glutamine deprivation is a characteristic of such ce.lIs. 196

Another interesting observation h¡as the fact that during glutamine sufficiency all the primary cell lines tested, with the exception of I{I-38, were found to 32pi incorporate simílar high amounÈs of into HS3 hrhen compared with the glutamine starvation condition (table 11; figs l9a and b). The absence of the phenomenon just described in permanent celI lines may be the result of some transformation characteristic (s) of such ce11 lines (see I Growth Require¡nents and Characteri stj. cs

of Cultured Mammalian Ce11s, in Historical). For exampJ_e in their altered serum and nutrient requirements (paut et. al. 1971; Rudland et. al. !974\ or changes in the composition of the proteins and morphology of their plasma membrane (p.obbins and Nicol-son 1975; Hynes 1976). Hov¡ever mouse 3T3 and BHK 2I,/CI 13 ceIls, for example, other than being permanent cell línes (Aaronson and Todaro 1968r Macpherson and Stoker f962) exhibit very few, if any, of the characteristics of Èhe transformed state. The two phenomena discussed, with regard to HS3 metabolism in primary cel_1 lines are interesting probl,ems. Further careful analysis wiIl be required to determine the significance of these observations. For example it would be highly informative if the subcellular localization and concentration of HS3 in mammalian cells couLd be quantitated. At teast in ]-97 the fungus Achl-ya, HS3 has been found to be associated with a proteogLycan located on the cell- wa]l-membrane complex (Cameron and LëJohn 1978). The HS molecule is probably synthesized at the ce1l membrane and, depending on the gro¡1rth condition, can apparently be transl-ocated into the cel1 or excreted. into the medium (LêJohn, unpublished data). Since it Ì4ras found (tabLe 2) that HS3 contains adenine, uracil, pentose and hexose sugars, glutamate and phosphates, attempts to labe1 the HS ¡nolecule a 1^ with -t¡ and .*c precursors of de novo purine and pyrimidine pathways and the nucleosides and bases themselves have been without success. This may be due to the dilution of the supplemented radioactive precursors by the rrichr medium in which the ce1ls are cuLtured in. LéJohn (unpublished data) has success- fu11y 1abe1led Ach1ya HS3 using simÍl,ar but high specific activity precursors. The label1ing patterns of IlS3 Ltith the HS3 molecule precursors are similar to those .,. achieved with "Pi (Lê.Tohn, unpubì.ished data) . Conc Lus ion

One of the three polyphosphorylated dinucteosides (McNaughton et. al. ), HS3, has now been detected in all mamnalian ceII line tested. The results which have been presented are consistent r^'ith two hypotheses: (l) ItS3 accumul-ation, ovring to an increase in synthesis, in cultured mammalian cel-ls is in some yet unknor4rn way intimately linked with a deficient supply of purine nucleotides and/or its precursors. (2) HS3 may be physiologically important in regulating eucaryotic RNA and DNA metabolism through its effect on the activities of the DNA-dependent RNA poL)zmerases and ribonucleotide reductase. 799

B IBLÏ OGRAPHY B i bf iography

Aaronson, S.A. & c..f . Todaro ( j-968) Devel_opment of 3T3-l,ike Lines from Bal-b/c mouse embryo cuLtures J. CeII. Physiol. 72:!4I-748. Abercrombie, M. (1970) Contact inhibition in tissue culture. In Vitro, 6:I28-I42 . Ada¡ns, A. & R.A. Harkness (L976) oevefopmental changes in purine phosphoribosyl transferase s in hunan and ïat tissues. Biochem. J., 760:565-576. Ambrose, E., U. Batzdord & R. Stuart (1970) Subsurface sÈructures in normal_ and malignant cel_l-s. Nature, 227:397-399. And.erson, E.P. (1973) Nucleoside and nucl-eotide kinases. fn : The enzymes vo1 9, P.D. Boyer (ed.) Academic Press, Ner1, York. pp 49-96. Appel, S.H. (1968) Purification and kinetic properties of brain orotidine 5'-phosphâte decarboxylase. ,J. Bio1. Chem. , 243:3924-3929. Ashman, D.F., R. Lipton, M.IU. Melicov, & T.D. price (1963) Isol-ation of adenosine 3 ' , 5 ' -monophosphate and I | guanosine 3 , 5 -mono-phosphate from rat urine. Biochem. Biophys. Res. Commun., 11:330-334. Aub, J.C., B,H. Sanford & M.N. Cote (1965) Studies on the reactivity of tumour and normal cells to â \^rheat germ agglutinin. Proc. Nat. Acad. Sci. (U.S.A.), 54:396-399 . Bagnara, .A.S., A.A. Letter & J.F. Henderson (1974) Multipte mechanisms of regulation of purine biosynthesis de novo. Biochem. Biophys. Acta., 374:259-270. 201

Barnes, D.W. & S.p. Colowick (19771 Stimulation of sugar uptake and thymidine incorporaÈion in mouse 3T3 cells by calciìlm phosphate and other extracellular particles. proc. Nat. Acad. Sci. (U.S.A.), 7 4:5593-5597. Baurn, 8.J., J. Moss, S.D. Breul & R.G.Crystal- (1979) Association of elevated Levels of adenosine 3':5'_ monophosphate with a selective decrease in collagen production. .T. Bio1. Chem., 253:3391-3394. Becker, M.M., L.,f . Meyer, A.W. ülood & ,I.8. Seegmiller (1973) Purine overproduction in man associated with phosphoribosylpyrophosphate synthetase activity. Science, 17 9 zJ-123-J-126. Beckh¡ith, J., T. Grodzicker & R. Arditti (1972) Evidence for thro sj-tes in the lac promoter region. J. MoI. Biol. 69:155-160. BLake1ey, R.L. (1969) The bioche¡nisÈry of fol-ic acid and related pteridines A. Neuberger & E.L. Tatum (eds.), vo1 L3, North Holland publ-ishing Co., AmsÈeLdam, pp.267-331. Block, R. e V0..A,. Haseltinê (1973) Thermolabitity of the stringent factor in re1 mutants of Escherichia coli- J. MoI. BioL. , 77:625-629. Bonner,J.T.(1970)Inductionofsta1kce11differentiation by cyclic .AMp in the cel1ular slime mold Dictyostelium discoideum. proc. Nat. Acad. Sci. (U.S.A.), 65:1L0-1I3. Bonner, J.T. (1971) Aggregation and differentiation in ceLluLar slime molds. Ann. Rev. MicrobíoI. , 25:75-92. Borek, E., A. Ryan & J. Rockenbach (1955) Studies on a mutant of Escherichia coli vrith unbalanced ribonucleic acid synthesis. J. Bacteriol., 71:3IB-323. Borsa, J. & G.F. V'lhitmore (1969) Studies relating to the . mode of action of methotrexate III. tnhibition of thymidylate synthetase in tissue culture cel1s and in cell-free systems. UoI. Pharmacol., 5:318-332. Bray, G. (1960) A simple efficient liguid scj.ntil-lator for counting aqueous soLutions in a liquid scintillation counter. Anal. Biochem. , I2279-285. Bresnick, E. (1961) Feedback inhibition of aspartate transcarbamylase in liver and in hepatoma. Cancer Res. , 22: 1246-125I. Bresnick, E. (1963) Inhibition by pyrimidines of aspartate transcarbamylase partially purified from rat liver. Biochem. Biophys. Acta. , 672425-434. Bresnick, E. & K. Blatchford (1964) Inhibition of dihydroorotase by purines and pyrimidines. Biochem. Biophys. Acta. , 8l-:150-157. Burger, IvI.M. & A.R. Goldberg (1967') Idêntification of a tumor-specific determinant on neoplastic ceJ-1

surfaces. Proc. Nat. Acad. Sci. (U.S.A.), 57 z 359-366. Burk, R.R. (1968) Reduced adenylate cycJ-ase activity in a polyoma vírus transformed cell line. Nature 2I9: !27 2-).27 3- Cameron, L.E. & H.B. féJohn (1978) Isolation and analysis of fungal phosphorylated proteoglycan. Can. .f. Biochem. , 56t237-245. Carchman, R.4., G.S. ,Johnson, I. Pastan & E.M. Scolnick (1974) Studies on the levels of cyclic A¡,IP in cells transformêd. by wíld-type and temperature- sen sitive Kirsten sarcoma virus. Cel-1, I-58-64. CarreL, A. (1912) On the permanent life of tissues outside of the organism. J. of Expt. Med., l-5:516-528. 203

Carson, D.4., J. Kaye & J.E. Seegmil_ler (1977) lymphospecific toxicity in adenosine deaminase deficiency and purine nucleoside phosphorylase deficiency : Possible role of nucleoside kinase(s). Proc. Nat. Acad. Sci. (U.S.A.), 74¿5677-5691 . Cashel, M. (1969) The control of ribonucleic acid synthesis in Escherichia coli. IV. Relevance of unusual- phosporylated compounds from amino acid starved stringent strains. . ,J. Biol. Chem. , 244:3133-3143. Cashel, M. (1970) Inhibition of RNA potymerase by ppcpp, a nucleotide accumuLated during the stringent response to ami,no acid starvation in E. coli. Cold Spring Harbor Symp. Ouant. Bio1., Nevr york 35 ¿ 407-413 . Cashel, M. (1975) Regulation of bacterial ppcpp and pppcpp. Ann. Rev. Micro., 29:301-318. Cashel, M. e ,f. Gallant (1968) Controt of RNA synthesís in Escþerichia co1i. I. Amino acid dependence of Èhe synthesis of the substrates of RNA polymerase. J. Mo1. Bio1. , 34:31-7-330. Cashel, M. & ,I. Gallant (1969) Two compounds implicated in the function of the RC gene of Escherichia coli. Nature, 22L¿ 834-841 . Cashel, M. e .1. Gallant (1974) CelluLar regulation of guanosine tetraphosphate and guanosine pentaphosphate. In : Ribosomes M. Nomura, A Tissieres and P. LengyeL (eds.) Cold Spring Harbor Laboratory. pp. 733-745. Cashel, M. & B. Kafbacher (1970) The control of ribonucleic acid synthesis in Escherichia coIi. V. Characteri- zatlon of a nucleotidé associated with the string- ent response. J. Biol. Chem., 245:2309-2318. 204

Caskey, C.T., D.M. Ashton & ,f .8. I{yngaarden (1964) The enzymology of feedback inhibition of glutamine phosphoribosylpyrophosphate amidotransferase by purine ribonucleotides. ,I. BioL. Chem. , 239: 2570-2579. CasÈellot, J.J.,Jr., M.R. MiIler & A.B. pardee (1978) Animal cells reversibly permeable to snall mol-ecules. Proc. Nat. Acad. Sci. (U.S.A.), 752351-355. Cereghini, S. & M.T. Fran ze-Fernande z (I976) Ehrl_ich ascites cel-ls Dl,¡A-dependent RNA polymerases: Effect of amino acids and protein synthesis inhibition. FEBS Lett. , 41:161-165. Chaloner-Larsson, G. & H. yamazaki (1976) Synthesis of r- guanosine 5 triphosphaÈe- 3 '-diphosphate in a EpgT strain of Escherichj_a coli,, Can. J. Biochem. , 54-935-940. Chambon, P. (1975) Eucaryotic nuclear RNA polynerases. Ann. Rev. Biochem. , 44:613-638. Chan, T-s (1978) Deoxyguanosine toxicity on 1ymphoid ceLls as a cause for j-¡nmunosuppress ion in purine nucleoside phosphorylase deficiency. CeII, 14:523-530. Chan, T.S., K. Ishii, C. Long & H. creen (1973) purine excretion by mammaJ.ian ce1ls deficient in adenosine kinase. J. ce1I. physio1., 81:3I5-321. Chan, V.L., G.F. V{hitmore & L. Siminovitch ( j-972) Mammal-ian ce1ls with aLtered forms of RNA polymerase If. proc. Nat. Acad. Sci. (U.S.A.), 69:311-9-31-23. Chu, F., H-F. Kung, P. Caldr^reIl, H. Weissbach & N. Brot (1976) DNA dependent synthesis of protein Ll2 from Escherichia coli ribosomes, in vitro. proc. Nat. Acad. Sci. (U.S.A.\, 732 3156-3I59. 205

Cl-e99, ,J.S., A.H. lfarner e F.S. Finamore (1957) Evidence for the function of pl, p4-diguanosine 5'-tetra- phosphate in the development of Artemia salina. ,I. Biol,. Chem. , 242¿1938-1943. cohen, 4., D. DoyLe, D.vt. uartj-n,Jr., & A,J. Ammann (1g76) Abnormal purine metabolism and purine overproduction in a patient deficient in purine nucleoside phosphorylase. N. Eng. J. Med., 293: I44g-I454. Cohen, .A., L.J. Gudas, A.J. AIÌunann, c.E.,f . Staa]. & D,W. Martin, Jr. (1978) Deoxyguanosine triphosphate as a possibJ.e toxic metabolite in purine nucfeoside phosphorylase deficiency. J. CIin. fnvest., 61 : l4 05-14 09 . Cohen, .A.,R. Hirschhorn, S.D. Horowitz, A Rubinstein, S.H. Po1mar, R. Hong e D.W. Martin,Jï. (1978) Deoxyaden_ osine triphosphate as a potentiatty toxic metabol,ite in adenosine deaminase deficiency. proc. Nat. Acad. S ci. (U. S.A. | , 75:.472-476. coleman, ¡{.s., J. Donofrio, ,J..1. Hutton, L. Hahn, A. David, B. Lampkin and ,f. Dyminski (1928) rdenri_ fication and quantiÈation of adenine deoxynucleotide s in erythrocytes of a patj-ent with adenosine deaminase deficiency and severe combined immuno_ deficiency. J. Biol. Chern., 253:16J-9-1626. Coleman, P.F., P.p. Suttl-e & c.R. Stark (1927) purification from hamster cells of the multi functional- protein that initiates de novo synthesis of pyrimidine nucleotides. J. Biot. Chem., 252= 6379-6385. Cook, J.L. & M. Vibert (1966) The utilizaÈion of purines and their ribosyl derivatives fo¡ the fornation of adenosine triphosphate and guanosine triphosphate in the rabbit reticul-ocyte. ,t. BioI . Chem., 241:158-160. 206

Cory, ,t. & T. Whitford, ,.fr. (t972) Ribonucleotide reductase and DNA synthesis in Ehrl-ich ascites tumor ce1ls. Cancer Res., 32:13 01-1306. Boer, H.À., A.J.,f van Ooyen, G. Ab & M. Gruber (1973) The role of messenger RNA and peptidyl-tRNA in the synthesis of the guanine nucleotides MSI and MSII by ribosomes in vivo. FEBS Lett., 30:335-338. De Crombrugghe, 8., B. Chen, !v. Anderson, p. Nissfey, M. Gottesman, R. Perlman & I. pastan (1971) Lac DNA, RNA pol)rmerase, cyclic .A¡{p receptor protein, cyclic AMP, lac repressor and inducer are the essential elements for controlled lac trnascription. Nature, 23J.¿ I39-!42. Dehlinger, P.J., T.a. Hamj-lton & M. Litt (1977) Amino acid control of stabLe RNA synthesis in Friend leukemia cells in relation to intracellular purine nucleoside triphosphate leve1s. Eur. ,f. Biochem. , 77:495-499. Dietzler, D.N. & M.P. Leckie (1977) Regulation of ÀDp- glucose synthetase, the rate-Limiting enzyme of bacterial glycogen synthesis, by the pleiotropic nucleotides ppGpp and pppcpp. Biochem. Biophys. res. Commun. , 77 21459-7467. Dirheimer, c. e J.p. Ebel (1965) Characterj-sation d'une polyphosphate-AMP-phosphotran s fera se dans Corynebacterium xerosis. Compt. Rend., 260:3789-3790. Durham, J.P. & D.H. Ives (1969) Deoxycytidine kinase I. Distribution j_n normal and neoplastic tissues and interrelationships of dêoxycytidine and t-l - arabinofuranosyl--cytosine phosphorylation. MoI. Pharmacol. , 5:358-375. Eagl"e, H. (1955a) The specj.fic amino acid requirements of a human carcinoma cel-l (strain HeLa) in tissue culture. J. ExPt. Med., 1.02:37-48. 207

gagIe, H. (1955b) The specific amino acid requirements of a mammalian celI (strain L) in tissue culture. J. Biol. Chem. , 2!4:839-g52. Eag1e, H. (1955c) Nutrition needs of mammalian ce11s 1n

tissue culture. Science, I22 z 43- 46 - Eagle, H., V.f . Oyama, M. Levy, C.L. Horton & R. Fleisch_ man (1955) The grovrth response of mammalian cells in tissue culture to L-gLutamine and L-glutamic acid. J. Bio1. Chem., Lgt6O7-6I6. EarLe, W.R. e A. Nettleship (1943) production of malignancy in vitro. V. Results of injections of cul,tuïes into mice. J- NaÈ. Cancer fnst., 4=2!2-227. Edlin, c, sr J. Neuhard (1967) Regulation of nucleoside trj-phosphate pooJ.s in nscherichia co1i. J. Mol. Biol. , 242225-230. Elford, H- (.1972) Functj-ona1 regulatj-on of mammalian ribo- nucleotide reductase. Adv. Enzyme Regulation. 10:19-38. Elford, H.L. (1974) SubcelluLar l-oêalization of ribonucLe- otide reductase in Novikoff hepaÈoma and regenerat- ing rat Liver. Arch. Biochem. Biophys., 163:537-543. Elford, H., M. Freese, E. passamani & H. Morris û,970) Ribonucleotide reductase and celJ- proJ.iferation. I. Variations of ribonucleotide reductase activity with tumor growth rate in a series of rat hepatomas. J. Biol. Chem. , 24525228-5233. Elmerich, c, & J.p. Aubert (l-973) Involvement of the biosynthetic pathhray of purine nucleotides in the repression of bacterial sporulation. Biochem. Biophys. Res. Commun., 55:937-942. 208

Em¡ner, M., U. de Crombrugghe, l. pastan & R. perlman (1970) The. cyclic AMp receptor protein of E. coli : Its role in the synthesis of inducible enzymes. proc. Nat. Acad. S ci. (U.S.A.\, 66:4gO-487. Erlich, H. (l-972) The regulation of RNa synthesis and stability by nucleotide pools in Escherichia coli. Ph.D. thesis, University of lrlashington. Erlich, H., R.A. Lazzatini & J. call,ant (1975) Synthesis and turnover of ribosomal pNA in guanine-starved cells of E. coli. J. Biol. Chem., 250:3057-3061. Fiil, N. (1969) A functional- analysis of the rel gene in Escherichia coli. J. !1o1. Bio1., 45:195-203. Fii1, N. e J.Ð. Friesen (l-968) Isotation of relaxed mutants of Escherichia co1i. ,I. Bacteriol. , 952729-73!. Fiil, N.P., K. von Meyenburg & J.D. Friesen (1972) Accurnulation and turnover of guanosine tetraphosph- ate in Escherichj-a coli. ,f. Mol. Biol-. , jI:769-783. Fiil, N.P., B.M. Willumsen, J.D. Friesen & von Meyenburg, K. (1977) Interaction of all,eles of the rel A, rel C and EpeT genes in E. coli : Analysis of the inÈer- conversion of GTP, ppcpp and pppcpp. MoI. Gen. cenet. , 150:87-101. Finamore, F.J. & A.H. Vtarner (1963) The occurrence of pl, P'-diguanosine^ 5 | -tetrephosphate in brine shrimp eggs. J. Biol. Chem. , 2382 344-348. Fisher, F.A. & A.C.I. Sartorelli (tr964) Development, maintenance and assay of drug resistance. Methods Med. Res. , IO:247-253. Fisher, H.W., T.T. puck & G. sato (f958) Molecular growth requirements of singì-e mammalian cells: The action of fetuin in promoting cett attachment to gLass. Proc. Nat. Acad. Sci. (U.S.A.), 44:4-10. 209

Folkman, J., M. Hochberg & D. Knighton (!974) Self-regula- tion of grolarth in three dimensíons : The role of surface area limitatíon. In : Control of prolífera_ tion in animal ce]ls. B. Clarkson & p,. Baserga (eds.) Cold Spring Harbour laboratory, Ner,, york, pp 933-g42. Poury, F. & A. coffeau (1973) Sti¡nulation of yeast RNA synthesis by cycloheximide and 3',5'-cyclic AMp. Nat. New Bio1. , 245|.44-47. Fox, I.H. & W.N. Kelley (1971) phosphoribosyl-pyrophosphate in man : Biochemical and clinical significance. Ann. Intern. Med. , 742424-433. Fox, I.H. & i{.N. KeLley (1972) Human phosphoribosylpyroph_ osphate synthetase : kinetic mechanism and. end.- product inhibition. J. Blo1. Chem., 247:2I26-2I3I. French, T.C., I.B. Davj-d, R.A. Day & J.M. Buchanan (1963) Azaserine-reactive sulfhydryl group of 2-formamido- N-ribosyl-acetamide 5'-phosphate : L-glutamine amido- ligase (adenosine dÍphosphate). .I. Bio1. Chem., 2382277]--2I77. Friedman, R.M. (1977) Antiviral activity of interferons. Bacteriological Rev., 4I:.543-567. Fr j-êdman, T., ,f .E. Seegmiller e J.H. Subak-Sharpe (1969) Evidence against the existance of guanosine and inosine kinase in human fibrobfasts in tissue culture. Expt. Cell-. Res., 56:425-429. Furano, A.V. & F.P. Vùittel (1976) Synthesis of elongation factors Tu and G are under stringent control in Escherichia coli. J. Biol. Chem., 25Izg98-901. Gal-lant, J., H. Er1ich, B. Hal1 & T. Laffter (1970) Analysis of the RC function. Cold Spring Harbor Symp. Quant. Bio1., New york, 35:397-405. 210

Gallant, J. & B. Harada (1969) The control_ of RNA synthesis in Escherichia coli III. The functional reJ.ationship betr4'een purine ribonucl-eoside triphosphate pool sizes and the rate of ribonucleic acid accumulation. .I. Biol. Chem. , 24423125-3132. Gallant, J., J. Irr & M. Cashel (1971) The mechanism of amino acid control of guanylate and adenylate biosynthesis. J. Bio1. Chem., 246:5812-5816. Gall-ant, J. & R.A. Lazzarini (1976) The regulation of ribosomal RNA synthesis and degradation in bacteria. In : protein synthesis VoJ_.2, E.H. ¡{cconkey (ed.) Marcel Dekker, Inc., Nev, york and Basel , 347 pp. Gall-ant, J., L. Shell & R. Bittner (1976) A novel nucleot- ide implicated in the response of E. col-i to energy source downshíft. CeII, 7:75-84. Gey, G.O., W.D. Coffman & M.T. Kubicek (1952) Tissue cult- ure st.udies of the proliferative capaciÈy of cervical carcinoma and normal epithel-ium. Cancer Res. , 12:26 4-265. Giblett, 8., A..f . Änmann, D.V,f . Wara, R. Sandman & L.K. Diamond (1975) Nucleoside phosphorylase deficiency in a child with severely defective T-ceII immunity and normal B-cet1 immunity. Lancet, l:1010-1013.

Giblett, E., J.E. Anderson, F. Cohen, B. poll.ara & H. J. Meuwj-ssen (1972) Adenosine-deaminase deficiency in two pãtients with severeJ.y inpaired cellular immunity. Lancet, 2 z!067-1069 - Goh, S.H. & H.B. Lêjohn (1977) cenetical and biochemical evidence that a novel dinucl-eoside polyphosphate coordinates salvage and de novo nucleotide biosynthetic pathr^rays in rnammalian ce1ls. Biochem. Biophys. Res. Commun., 74:256-264. 2r7

Goh, S.H. & H.B. LéJohn (1978) clucose transport in Achlya : Characteri zation and possíble regulatory âspects. Can. J. Biochem. , 562246-256. Goh, S.H., J.A. WÌighÈ e H.B. LêJohn (1977) possible regulation of macromolecular biosynthesis in mammal-ian celJ-s by a novel dinucleoside poLyphosph_ ate (Hs3) produced during step-down gror,rth conditions. J. CeII. physío1., 93:353-362. Goldberg, A.L. e A.C. St. .tohn (1976) Intraceltular protein degradation in mammatian and bacterial ceLls : parl- 2. Ànn. Rev. Biochem. , 45:747-903. Goldberg, N.D. & M.K. Haddox (1977) Cyclic cMp metabolism and involvement in biological regul-ation. Ann. Rev. Biochem. , !52823-896. Goldberg, N.D., M,K. Haddox, E. Dunham, C. Lopez & ,l.I4l. Hadden (1974) The yin-yang hypothesis of biological control : Opposing infLuences of cyclic G-Ì,4p and cyclic A.lvIP in the règuLation of cel1 protiferation and other biological processes. In : Control- of proliferation in animal cells. B. Clarkson & R. Baserga (eds.) Cold Spring Harbor Laboratory. Ne\,r York, pp 609-625. Gomes, S.H., L. Mennucci & J.C. Da Costa Maia (I9Zg) Adenylate cyclase activity and cA¡4p metabolism during cyto-differentiation of BlastocladieIla emersonii. Biochem. Biophys. Acta 541:I90-I98. Goodman, 8.M., H.ú¡. Sauer, L. Sauer & H.p. Rusch (I969) Polyphosphate and other phosphorous compounds during grohtÈh and differentiation of physarum polycephalum, 15 : 132 5- 13 3I. Gospodarolrri cz , D. & J.S. Moran (1976) cror^'th factors in mammalian celI culture. Ann. Rev. Biochem., 45: 531-558. )1)

Gottleib, D. & p.D. Shaw (l_967) Antibiotics Vot. t. Mechanism of action. Springer_VerLag, ¡Iew york , pp. 259-277, 283-307, j14-725. Graf, L.H.Jr., J.A. McRoberts, T. lt. Harrison & D.W. Martin, .tr. (1976) Increased pRpp synthetase activity in cultured rat hepatoma cell-s containing mutations in the hypoxan thine-guanine phosphoribosyltrans fera se gene. J. CeLl. physiol. , B8:33I-342. Green, H. & T-s Chan (1973) pyrimidine starvation induced by adenosine in fibroblasts and lymphoid cel-1s : Role of adenosine deaminase. Science, .1g2:g36_937. Gross, K.J. & A.O. pogo (1976) Control of ribonucl-eic acid synthesis in eucaryotes.3. The effect of cyclohexi_ mide and edeine on RNA synthesis in yeast. Biochemistry , 15 =2082-2086. crummt, !.. (1978) Diadenosine 5',5',,-p],p4-tetraphosphate triggers initiation of in vitro DNA replication in baby hamster kidney cells. proc. Nat. Acad. sci. (U.S.A.), 75:37I-375.

Grummt, f. & F. crummt (1976) Control_ of nucleolar RNA synthesis by the intracellular pool sizes of ATp and cTP. CeII, 7:447-453. crummt, I., V.A. Smith & F. crummt (l-976) Amino acid starvation affects the initiation frequency of nucleolar RNA po.j-ymerase. Ce11, 7 :439-445. Gudas, L.J., B. Ullman, A, Cohen & D.W. Martin,Jr. (197g) Deoxyguanosine toxicity in a mouse T-lymphoma : Relationship to purine nucleoside phosphoryl-ase- associated inrnune dysfunction. CeIt, l4:53L-53g. Hager, S.E. & M.E. Jones (1967a) fnitial steps in pyrimid- ine synthesis in Ehrl.ich ascites carcinoma in vitro If. The synthesis of carbamyLphosphate by a soluble glutamine-dependent carbarnyl phosphate synthetase. J. Biol. Chen. , 242:5667-5673. 2I3

flager, S.E, e M.E. Jones (l-967b) A glutamíne-dependent enzyme for the synthesis of carbamyl phosphate for pyrimidine biosynthesis in fetal rat 1iver. J. Biol. Chem. , 242:5674-5680- Harol-d, F.M. (1966) fnorganic polyphosphates in biology : Structure, metabolism and function. BacterioJ.ogica I Rev. 30:772-7 94 Harstunan, R.B. & H. yamazaki (197I) Fo¡mation of ppcpp in the relaxed and stringent strains of Escherichia coli during díauxie lag. Biochemistry, 10:39g0-3982. Hartman, S.C. (1963) t¡e ínteraction of 6*diazo-5-oxo- I,-norleucine with pho sphoribosy lpyropho sphate amidotrans ferase. J. Biol. Chem., 23822036-204:.. Haseltine, W.A. & R. Block (l-973) Synthesis of guanosine tetra- and pentaphosphate reguires the presence of a codon-specific uncharged transfer ribo- nucleic acid in the acceptor site of ribosomes. Proc. Nat. Acad. sci. (u.s.A.), 70:1564-1568. Haseltíne, Vf .4., R. B]ock, w. cilbert & K. Weber (t972) MSI and MSII made on ribosome in idling step of protein synthesis. Nature, 238:3BI-384. Hayashi, I. & c. Sato (1976) Replacement of serum by hormones permj-ts grohrth of cells in a defined medium. Proc. Nat. Acad. Sci. (U.S.A.l, 259:132-I34. Hayflick, L. (1965) The limited in vitro lifetime of human diploid ceIl strains. Expt. CeLl Res., 37 :614-635. Hayf1ick, L. & P.S. Moorhead (1961) The serial cultivation of human diploid cell strains. Expt. Cefl Res., 25:585-62I. Heidelberger, C. (1975) Antineoplastic and inrnuno- suDDres.l nts Part fI, A.c. Sartotell,i and A.c. Johns (eds.), Springer-Verlag, Nelr york, Pp. 193-23t. 2r4

Heinemeyer, E.A., I\I. ceis & D. Richter (197g) Degradation of guanosine 3'-diphosphate 5r_diphosphate in vitro by the spoT gene product of Escherichia co1i. Eu¡. ,]. Biochem. , 89:125-131. Henderson, J-8. (1972) Requlation of purine biosynÈhesis. Am. Chem. Soc., Vtashington D.C., pp. 49_64. Henderson, J.F., ,f .K. Lowe & ,f. Barankiewicz (19771 purine and pyrimidine metabol-ism : pathways, pitfalls and pertubations. fn : purine and pvrimidine metabolism. Ciba Foundation symposium. 4g (new series) ¡ Elsevier, Excerpta Medica, North Holtand, pp. 3_14. Hershfield, M.S. e J.E. Seegmil_ler (1976) Regulation of de novo purine biosynthesis in human ]ympboblasts. Coordinate control of proximal- ( rate_determining ) steps and the i¡qsi¡1c acid branch point. ,I. Biol . Chem. , 25I:?348-7 354. Hershko, .A., p. Mamont, R. Shietds e c.M. Tomkins (l-971) Pleiotypic response. Nature New Biol . , 232=206_21I. Hershko, .A., A. Razim & J. Mager (1969) Regulation of the synthesis of 5-phosphoribosyl_l_pyrophosphaÈe in intact red blood ce1ls and in celI_free preparations. Biochem. Biophys. Acta, 194: 64_7 6. Hildebrandt, A. e H.W. Sauer (1977) Transcription of ribosomal RNA in the life cycle of physarum may be regulated by a specific nucfeola¡ initiation inhibitor. Biochem. Biophys. Res. Comlnun ., 74:466_ 472. Hirschhorn, R., N. Beratis, F.S. Rosen, R. parkman, F. Stern & S. pol_mar (I975) Adenosine_dea¡nina se defíciency in a chil-d diagnosed prenatally. Lancet, l:7 3-7 5. ¿13

Holffa, E., J. ,Janne & J. Pispa (1974) The regulation of polyamine synthesis during the stringent control in Escherichia col-i. Biochem. Biophys. Res. Conmun., 59:1104-1111. Hollenberg, M.D. & P. Cuatrecasas (1973) Epidermal qrowth factor 3 Receptors in human fibroblasts and moduLa- tion of action by choJ.era toxin. Proc. Nat. Acad. Sci. (U.S.A.) , 7022964-2968- HoLfey, R.W. (1975) Control of grohrth of mammafian cells in ceIl culture. Nature. , 258:489-490- Holmes, E.Il., J.A. McDonald, J.M. Mccord, J.B. Wyngaarden & W.N. KelIey (1973a) IIuman glutamine phosphoribos- ylpyrophosphate amidotrans ferase : kinetic and regulatory properties. J. Biol. Chem., 248:144-150. Holmes, E.!f ,, J.B. Wyngaarden & !f .N, Kelley (1973b) Human glutamine phosphoribosylpyrophosphate ami-dotransferase : trno molecufar forms interconver- tible by purine ribonucleotides and phosphoribosyl- pyrophosphate. ,f. Biol. Chém. , 2482 6035-6040- Ho1mes, E.W., Pehlke, M. & vt.N. Kelley (1974) Human IMP dehydrogenase. Kinetics and regul-atory properties. Biochem. Biophys. Acta. , 364¿209-2L7- Hoogenraad, N.J. & D.C. ],ee (1974) Effect of uridine on de novo pyrimidine biosynthesis in rat hepatoma celLs in culture. J. Biol. Chem., 249:.2763-2768. Hovanessian, 4.G., R.E. Broú7n & I.II. Kerr (1977) Synthesis of low molecular weight inhibitor of protein synthesis vrith enzyme from interferon-treated celIs. Nature, 268¿537-539. Hoh¡ard, A. & S.R. Pelc (1953) Synthesis of desoxyribonucl- eic acid in normal and irradiated cel-ls and its relation to chromosorne breakage. Heredity, London, (SuPPI .) , 6:26I-273. 2r6

Hryniuk, W.M., L,W. Brox, ,J,F. Henderson & T. Tamaoki (1975) Consequences of methotrexate inhibition of purine biosynthesis in L517By ce1ls. Cancer Res., 35 z I427-!432 . Hynes, R.O. (1976) Ce1] surface protej-ns and malignant transformation. Biochem. Biophys. Acta, 458:73-107. fshii, K. & H. Green (t973) Lethality of adenosine for cultured mammalian cells by interference with pyrimidine biosynthesis. ,J. Cel] Sci., 13:429-439. Ito, K. & H. Uchino (1972) Controt of pyrimidine biosynthesis ín human lymphocytes. J. Biol. Chem., 248:389-392 - Itoh, R., E.W. HoLmes & ,I.8. Wyngaarden (1976) pigeon 1íver amidophosphoribosyltransferase : ligand- induced alterations in molecular and kinetic properties. J. Biol. Chem., 25I:2234-2240. Johnson, c.S. e I. Pastan (197J.) ¿¡¿¡ge ín growth and morphology of fibroblasts .by prosÈaglandins. J. Nat. Cancer Inst., 47 2!357-]-364. Johnson, G.S., I. Pastan, C.V. peery, J- Otten & M.C. WilJ.ingham (1972) ProstagJ.andins . In: Cellular bio.logy and the infLamrnatory process. p.W. RamweIl and B,B. Pharriss (eds.). P1enum press, Nevr york, pp. 195-2i-9. ,folicoeur, P. & F. Labrie (1974) Control of RNA synthesis by amino acids in Landschutz ascites tumor ceLl-s. Can. J. Biochem. , 52¿867-876. .folicoeur, P., A. Lemay, F. labrie & A.L. Steiner (1974) Phosphorylation of chromosomal and ribosomal- proteins and intracellular levels of cyclic 3',5,- adenosine monophosphate and cyclic 3' ,5'-guanosine monophosphate during amino acid starvation in Landschutz tumour ce1ls. Expt. Cell- Res., 89:231-240- 2r7

Kaplan, S., A.c. Atherly & A. Barrett (1973) Synthesis of stable RNA in stringent Escherichia coli cells in the absence of charged transfer RNA. proc. Nat. Acad, Sci. (U.S.A. ) , 7Ot6g9_692. Kel1ey, !f.N., Ì.H. Fox & J.B. Wyngaarden (1970) Regulation of purine biosynthesis in cultured human cells f. Effects of orotic acid. Biochem. Biophys. Acta., 215: 512-5t-6. Kempe, T.D., E.A. Sr^tyryd, !1. Bruist & c.R. Stark (1976) Stable mutants of mammalian cells that overproduce the first three enzymes of pyrimidine nucleotíde biosynthesis. Cell, 9:541-550. Kerr, I.M. & R.E. B¡own (l-978) pppA2'p5'A2'p5'A : An inhibitor of protein synthesis synthesized with an enzyme fraction from interferon_treated cel1s. proc. Nat. Acad. sci. (U.S.A.), 75:256_260. Kersten, H. & Vù. Kersten (1974) fnni¡itors of nucleic acid synthesis. fn : Molecular biotogy, biochemistry and biophysics. Springer-Ver1ag., New york, Heidelberg, Ber1in, pp. 40-67. Konijn, T.M., J.G.C. van d.e Meene, J.T. Bonner & D.S. Barkley (1967) The acrasin activity of adenosine_ 3r-5r cyclic phosphate. proc. Nat. Acad. Sci. (U.S.A. ) , 58:1152-1154. Kornberg, 4., S.R. Kornberg & E.S. Simms (1956) Meta_ phosphate synthesis by an enzyme from Escherichia colj-. Biochem. Biophys. Acta. 26:.215_227. Kram, R., P, Mamont & G.M. Tomkins (1973) pleiotypic

control by adenosine 3r:5'-cyclic monophosphate : A nodel for gror^rth control in animat cells. proc. Nat. Àcad. Sci. (U.S.A. | , 70t1432_7436. 2r8

Kuehn, c.D. (1972) CetL cycle variation in cyclic adenosine I 3 , 5 ' -monophosphate dependent inhibition of a protein kinase from physarum pofycephalum. Biochem. Biophys. Res, Connun . , 49:4I4-4I9. Kuo, J.F. & P. creengaard (1970) Cyclic nucleotide- dependent protein kinases VI . Isolation and partial purification of a protein kinase activated by guanosine 3 5 t -monophosphate. J. Bio1. chem., 245:2493-2498." Kur1and, C.c. & O. Ìtaaloe (1962) ¡sgulation of ribosomal and transfer RÌtrA synthesis. J. Mol. Bío1, , 42193-210. Laffler, T, & J. GaLlant (1974) Spor, a new genetic locus invotved in the stringent response in E. coLi. Cell, L: 27-30. Larsson, A. & P. Rej-chard (1966) Enzymatic synthesis of deoxyribonucleotides : IX and X. J. Biol-. Chem., 24122533-2549. Lazzariní, R.4., M. Cashel & J. calLant (1971) On the regulation of guanosine tetraphosphate l_eveIs in stringent and relaxed strains of Escherichia coIi. J. Biol. Chem. , 24624381-4385. Lazzarini, R.4., K. Nakata e R.M. !ùinsl_ow (1969) Coordinate control of ribonucleic acid synthesis during uracil deprivation. ,f. Biol-. Cheml, 24423092-3100. Lazzarj-,ní, R.A. & R.M. Winslol^' (1970) The regulation of RNA synthesis during gror^7th rate transitions and amino aci,d deprivation. Co.td Spring Harbot Sl¡mp . ouant. Bio1., Neh¡ York, 35:383-390, LéJohn, H.8., L.E. Cameron, c.R. Klassen & R.U. Meuser (1978) Effects of L-glutamine and HS cornpounds on growth and sporulation metabolisn of AchÌya. Can. J. Biochen. , 56:227-236. , ,,, 1.t.,, .:r ..: :,. -.

2I9

LêJohn, H.8., L.E. Cameron, D.R. McNaughton & G.R. Kl,assen (1975) Diguanosine nucl_eotides of fungi that regulate RNA polymerases isolated and partially characterized. Biochem. Biophys. Res. Commun., 66:460-467 . lêJohn, H.B. & R.M. Stevenson (1970) Multiple regulation processes in nicotinamide adenine dinucl-eotide- specific glutanic dehydrogenases. J. Bio1. Chem., 245:3890-3900. Lesch, Iq. e !v-L. Nyhan (1964) A familial disorder of uric acid metabolism and central nervous system functj,on. Æn. J. Med. , 36:551--570. levan, A. & ,I.J. BieseLe (1958) Role of chromosomes in cancerogenesis, as studied in serial tissue culture of mammalian cells. Ann. New york Acad. Sci., 7I: 1022-1053 . Levine, R.L., N.J. Hoogenraad & N. Kretchmer (1971_) Regulation of activity of êarbamyl phosphate synthetase from mouse spleén. Biochemistry, 10:3694- 3699. Lewis, W.H., B.A, Kuzik & J.A. wright (1978) Assay of ribonucleotide reduction in nucleotide-permeable hamster celIs. J. CeIl. physiol. , 94:287-298. Lewis, W.H., D.R. McNaughton, S.H. coh, H.B. LéJohn & J.Â. Wright (1977) Inhibition of mammalian ribonucleotide reductase by a dinucleotide produced in eucaryoÈic cells. J. Cell physiol". , 93:345-352. Lewis, I4t.H., D.R. McNaughton, H.B. LéJohn & J.A. Wright (1976) Regulation of fungaJ, ribonucleotide reductase by unusual dinucleotides. Biochem. Biophys. Res. Colunun., 71:128-135. 220

Ley, K.D. & R.A. Tobey (1970) Regulation of initiation of DNÀ synthesis in Chinese hamster ce11s. J. Cell. Bio1., 47:453-459. l,i, C-C & J. HochstadÈ (1976) .t4embrane-associated enzymes involved in nucteoside processing by plasma membtane vesicles ísolated Ítom Lg2g ce1ls grown in defÍned r¡ediu¡n. J. Bio1. Chem., 25:I181-IlB7. Lieberman, I. & P. Ove (1957) Purification of a serum protein required by a mammal-ian celL in tissue culture. Biochem. Biophys. Acta, 25:449-450. l.incoln, T.I!1., I{.L. Dills,Jr., & J.D. Corbin (1977) Purification and subunit composition of guanosine 3' : 5 ' -monophosphate-dependent protein kinase from bovine lung. J. Biol. Chem. , 252:4269-4275. Lindahl , r., L. Post & M. I\¡omura (1976) DNA-dependent in vitro synthes j-s of ribosomal proteins, protein elongation factors, and RNA polymerase subunit o( : Inhibition by ppcpp. CeII,.. 9: 439-449. lindeIJ., T.J., F. v¡einberg, p.ld. Morris, R.G. Roeder & r.,I. J. Rutter (1970) Specific inhj,bition of nuclear polymerase fI by O(-amanitin. Science, I70=447-449, Live, T.R. & E. Kaminskas (1975) Changes in adenylate energy charge in Ehrlich ascites tumor ceLl"s deprived of serum, glucose or amino acids. J. Biol-. Che¡n. , 25 0 217 8 6-17 89 . Loer4ren, P.C. (1976) Novel nucleoÈides fron E. coli. isolated and partially characterized. Bioché¡n. Biophys. Res. Commun . , 7O:1210-12I8. Ludwig, ,I.R.II, S.G. Oliver & C.S. McLaughlin (1977) The effect of amino acids on gror,ùth and phosphate metabolism in a prototrophic yeast strain. Biochem. Biophys. Res. Cornmun., 79:16-23. 22r

Lund, E. & N.O. Kjeldgaard (L972) Iletabolism of guanosine tetraphosphate in Eschêrichia coli. Eur. J. Biochem. 28: 316-326 - Maaloe, O. & N,O. Kjeldgaard (1966) The regulation of RNA synthesis. fn : Control of Macromolecular Synthesis. B.R. David (ed.), W.A. Benjamin, Inc., Ner^r York , pp. 12 5- 15 3 llacpherson, I. & L Montagnier (1964) ¡g¿¡ suspension culture fo¡ the seÌective assay of cells transformed by polyoma virus. ViroLogy, 23229I-294. Macpherson, I. & M. Stoker (1962) polyoma transformation of hamster ceLl clones-an investigation of genetic factors affecting cell competence. Virol-ogy l6: 14 7- 151 . Magasanik, B. (L961) Catabolite repression. In : Cellular regulatory mechanisms. Cold Spring Harbor Symp. Quant. BioI., New york, 26.2249-256. Mahler, H.R. & E.H. Cordes (1971) BioLogical Chemistry. ^nd z- edltLon. Harper and Ror4r, New york, pp. g19_g47. Makman, R.S. e E.hf . Suthertand (1965) Adenosine 3':5' phosphate in Escherichia co1i. J. Biol. Chem., 240:1309-1314. Mamont, M., À. Hershko, R. Kram, L. Schacter, J. I,ust & c.M. Tomkins (1972) The pleiotypic response in mammalian cells: Search for an intracel_l-uLar mediator. Biochem. Biophys. Res. Connun., 4g: 1378-1384. Martini, O., J. Irr & D. Richter (1977) euestioning of reported evidence for guanosine tetraphosphate synthesis in a ribosome system from ¡nouse embryos. CeIl, 12 : L 12 7- 1131. 222

McBurney, M.W. e c.F. Whitmote (I974) fsolation and biochemical characterization of fotate deficient mutants of Chinese hamster ce1ls. Cell, 2: L73-182. McFall, E. e B. ¡4agasanik (1960) The control of purine biosynthesis in cultured mammalian cells. .1. BioÌ. Chem. , 235¿2103-2!07. McNaughton, Ð.R., c.R. Klassen & H.B. LéJohn (1975) PhosphoryLated guanosine derivatives of eukaryotes : Regulation of ÐNA-dependent RNA potymerases I, II and IlI in fungal development. Biochem. Biophys. Res. Conunun., 66:468-474. McNaughton, D.R., G.R. Klassen, p.C. Loer^ren & H.B. LéJohn (1978) Recharacterization of fungal dinucl-eoside polyphosphate (HS3). Can. J. Biochem. , 56z217-226. Meur^rissen, H.J. & B. Pollara (1974) Adenosine deaminase deficiency : The first inborn error of metabolism noted ín immunodeficiency disease. J. pediatr., 84:315-316. Miller, R.8., R. Hackenberg & H, cershman (1978) Regulation of gJ-utamine synthetase in culÈured 3T3-Ll- cells by insulin, hydrocortisone and dibutyryl cyclic AMP. proc. Nât. Acad. Sci. (U.S,A.), 752I4I9-J,422- MiJ.ls, G.C., F.C. Schmalstieg, K.B. Trimmer, A.S. coldInan & R.M. Goldblum (1976) purine metabolism in adenosine deaminase deficiency. proc. Nat. Acad. Sci. (U.S.A. \ , 73:.2867-287I. Milman, c., Ð.L. Anton & J.L. Weber (1976) Chinese hamster purine-nucteoside phosphorylase : purification, structural and cätalytic properties. Biochem., I5: 4967-497 3. llitchison, J.M. (1971) The biology of the cetl cycle. Cambridge Univ. press, T,ondon & New york, pp. tl6- 121, r30-133. 223

llonod, J. (I947) The phenomenon of enzymatic adaptation. crov/th, lI:223-289. Ivfoore, E.C. e R.B. Hurlbert (l-966) Regulation of manunalian deoxyribonucleotide biosynthesís by nucleotides as actj-vators or inhibitors. ¡I. Biol. Chem. , 24Iz 4802-4809 - Mori, M, & M. Tatibana (1975) purÍfication of homogeneous glutamine-dependent carbamyl phosphate synthetase from Àscites hepatoma cells as a complex with aspartate transcarbamylase and dihydroorotase. J. Biochem. (tokyo¡, 7B=239-242. Muhammed, 4., A. Rodgers e D.E. Hughes (1959) purification and properties of a polymetaphosphatase from

corynebacterium xerosis. ,J. cên. Microbiol. , 20: 482- 495 . Murray, A.W. & P.C.L. Wong (1967) s-phosphoribosyl pyrophosphate synthetase from Ehrlich ascites-Èumor ceIls : Effect of magnesium and ATp concentration on the enzymic activity. Biochem. Biophys. Res. Conmun. , 29:582-587 . Nakashima, T,, M. Ku$rano, S-i. Akiyama & H. Endo (1976) Effect of L-asparaginase and asparagine deprivation on RNA meÈabolism in mouse leukemia L517gy cells in suspension culture. Cancer Res., 363 3616-362I. Nier1ich, D.P. (1968) Amino acid control over RNA synthesis: A re-evaLuation. proc. Nat, Acad. Sci. (U.S.Ä.), 60:1345-1352. Nishi, A. (1960) Enzymatic studies on the phosphorus metabo.lism in germinating spores of Àspergillus niger. J. Biochem., (Tokyo) , 482758-767. Nishi, A. (1961) Rol,e of polyphosphate and phospholipid ín germinating spores of Aspergillus niger. J. Bacteriol. , 81: l0- 19 . 224

Nishiyama, K., H. Katakami, ¡I. Yamamura, y. Takai, R. Shimomura & Y. Nishizuka (1975) Functional specifi- | city of guanosine 3 : 5 ' -monophosphate-dependent and adenosine 3r :5r-monophosphate-dependent protein kinases from si.lkworm. J. Biot. Chem., 25021297-1300. Norris, T.E. & A.L. Koch (19721 Effect of groÌ"rth rate on the rel-ative rates of synthesis of nessenger, ribosomal and transfer RNA in Escherichi-a cofi. J. Mol. Biol. , 64:633-649. Nyhan, W.L. (1973) The Lesch-Nyhan Syndrome. Ann. Rev. Med. , 24:41-60. Oikawa, T.G. & M. Smith (1966) NucLeotides in the encysted embryos of Daphnia magna. Biochem. , 5:I5I7-I52I. OÈten, J. G.S. Johnson & f. Pastan (1971-) Cyclic AMp in fibroblasts : Relationship to grovrth rate and contact inhibition of grovrth, Biochem. Biophys. Res. Commun., 44 -.1192- 119 8. Pardee, A.B. (1974) A restriction point for control of nor¡nal animal cefl proliferation. PToc. Nat. Acad. Sci. (U.S.A.), 7I¿1286-1290. Parks, R.E.Jr. e R.P. Agar\^'al (1972) Purine nucleoside phosphorylase. In : The enzymes. 34ed. P.D. Boyer (ed.) Academic Press, New York, pp. 483-514. Pastan, I.H., G.S" ,fohnson and W.B. Anderson (1975) Role of cyclic nucLeotides in growth control-. Ann. Rev. Biochem. , 44 ¿ 49I-522 . Pastan, I. & R. Perlman (1970) Cyclic AMP in metaboLism. Nature Ner4r Bio1. , 229 ..5-7 . Pastan, I. & R.L. Perlman (1972) negulation of gene transcription in Escherichia coli by cycJ.ic AItlP. In : Advances in cyclic nucLeotide research, vol, I., Greengaard, P., G.A. P.obison & R. Paoletti, (eds.) Raven Press, New York, pp. 11-15. 225

Patterson, D. (1975) Bíochemical Genetics of Chinese hamster cell mutants with devj-ant purinê metabolism : Biochemical analysj-s of eight mutanÈs. Somatic Ce11 Genet. , 1: 91-1I0. Paul, D., A. Lipton & f. Klinger (1971) Serum factor requirements of norma.L and simian virus 4O-trans_ formecl 3T3 mouse fibroblast. proc. Nat. Acad. Sci. (U.s.A. ), 68:645-648. Pederson, F.S., E. I,und & N.O. Kjeldgaard (1973) Codon specific, tRNA dependent in vitro synthesis of ppGpp and pppcpp. Naturê t¡er1' Biol. , 243213-15. Perlman, R. & f. Pastan (1969) pleiotropic deficiency of carbohydrate utilization in adenyl cyclase deficient mutant of Escherichia coli. Biochem. Biophys. Res. Cortunun., 37 : 151- J_5 7. PLanet, G. & t.H. Fox (1976) Inhibition of phosphoribosyl- pyrophosphate synthesis by purine nucleosides in human erythrocytes. ,I. Bio1. Chem., 251:5839-5g44. Pogo, A.O. & V.J. Zbrzezl]a (1978) Cycloheximide stimul-ates ribosomal RNA transcription in amino acid-starved ascites tumor cells. FEBS Lett., 88:135-l_38. polakis, s.E., R.B. Guchhait & M.D. rane (1973) stringent control of fatty acid synthesis in Escherichia coli: possíb1e regulation of acetyl CoA carboxylase by ppcpp. J. BioL. Chem., 248:7957-7966. Pomales, R., S. Bieber, R. Friedman & c.H. Hitchings (1963) Augmentation of the incorporation of hypoxanthine into nucleic acids by the administration of an inhibitor of xanthine oxidase. Biochen. Biophys. Acta, 7221L9-120. 226

Prasad, K.N. & S. Kumar (1924) Cyctlc AMp and the differêntiation of neuroblastoma cells in culture. In : Control of prol-íferation in animal celIs. B. Clarkson and R. Baserga (eds.). CoLd Spring Harbor T,aboratory, New york, pp. 5gI-594. Puck, T.T. & p.f. Marcus (1955) A rapíd method for viable ce11 tit.ration and clone production vríth HeLa cef l-s in tissue culËure : The use of X-irradiated ce1ls to supply condiÈioning factors. proc. Nat, Acad. Sci. (U.S.A, ), 41:432-437. Rapaport, E. & P.C. zamecnik (1975) Incorporation of adenosine into ATp : Formation of conpartmentatized ATp. proc. Nat. Acad. Sci. (U.S.A.), 7323122_3125. Rapaport, E. & P.C. Zamecnik (1976) presence 'lÀ of diadenosine 5',5"'-p-,p'-tetraphosphate (ApnA) in mammalian celLs in Level-s varying widely with proliferative activity of the tissue : A possibl_e positive "pleiotypic acÈivator". proc. Nêt. Acad. Sci. (u.s.A. ) , 73:3984-3988. Rasmussen, H. (1970) Ce1l communication, calcium ion and cyclic adenosine monophosphate. Science, I].0:404_4I2. Reem, c.H. (1968) g¡2ymatic synthesis of 5r phospho- ribosylamine from ribose- 5-phosphate and amrnonia, an Alternatíve first step in purine biosynthesis. J. Biol. Chem. , 24325695-570)-. Reem, G.H. (1972) Oe novo purine biosynthesis by tr^,o pathvrays in Burkitt lymphoma cells and in human spJ.een. J. CLin. Invest., 5Ì:105g-1062. Reichard, P. (1972) ConÈrol of deoxyribonucleotide synthesis in vitro and in vivo. In : Advances in Enzyme Regulation. G. I{eber (ed.) perganon press, Oxford, pp. 3-16. 227

Reiness, G., H. L. Yang, G. Zubay & M. Cashel- (1975) Effects of guanosine tetraphosphate on cell-free synthesis of Escherichia coti ribosomal RNA and other gene products. Proc. Nat. Acad. Sci. (U.S.A.) 7 2:288I-2885 . Renart, I'1.F., J. Renart, M.A.c. Sille¡o c A. Sill-ero (1976) Guanosine monophosphate reductase fron Artemia salina : Inhibition by xanthosine monophosphate and activation by diguanosine tetraphosphate. Biochem., J.5 z 4962- 4966. Rhaese, H.,J., H. Dichtelmul-Ier & !'.M. Giesel (1972) Unusual phosphorylated substances associated with sporulation. In : Spores V H.O. HaLvorson, R. Hanson & L.L. Campbell (eds. ) Anerican Soc. for Microbiology, Washington D.C-, pp.I74-!79. Rhaese, H.J., R. Gradê & H. Ðichtelmul1er (1976) Correla- tion of initiation of differentiation with synthesis of highly phosphoryl-ated nucl"eotides in BaciLlus subtilis. Eur. J. Biochem. , 64.205-213. Rhaese, H.J. & R. croscurth (1976) Studiès on the control of development : Role of regulatory nucleoÈides synthesized by membranes of Baci-11us subtilis in initiation of sporulation. Proc. Nat. Acad. Sci. (U.S.A.), 73:331-335. Rhaese, H.J., J.A. Hoch & R. Groscurth (l-977) Studies on the control of development : Isolation of Bacill-us subtilïs mutants blocked early in sporulation and defective in synthesis of highly phosphorylated nucleotides. Proc. Nat. Acad. Sci. (U.S.A.\, 742 II25-II29. Richter, D. & K. Isono (1977) The mechanisn of prote j-n synthesis. In : Current topics in nicrobiology and immunology, 76¿8L-L23. 228

Rickenberg, H.V. (1974) Cyctic AMp in procaryotes. Ann. Rev. Micro. , 28:353-369. Riedel, V, & c. Gerisch (1971) Regulation of extracell-ular cyclic AMP-phosphodiesterase activity during development of DictyosteLium discoideum. Biochem. Biophys. Res. Comnun . , 42zII9-I24. Riedel, V., G. Gerisch, E. Meuller c H. Berg (1973) Defective cyclic adenosine 3 r :5' phosphate- phosphodies terase (cAMp) regulation in morphogenetic mutants of Dictyostelium discoideum. J. MoI. Biol., 742573-585. Ringer, S. (1882) ¡. Physiol. , 4:29-42. Rlzzíno, A. & c. Sato (1978) cror^¡th of embryonal carcinoma cell-s in serum-free medium. proc. Nat, Acad. Sci. (U.S.A. ) , 75:1844-1848. Robbins, J.C. & c. Nicolson (l_975) Surfaces of normal and. transformed. cell-s, In : Cancer 4 : A comprehensive treatise. F.F. Becker, (ed.), plenum press, New York and London, pp. 12-40. Roberts, W.K., A.G. Hovanessian, R.E. Brohrn, M.J. Clemens & I.M. Kerr (l-976) Interferon-mediated protein kinase and low-molecular-we ight inhibitor of protein synthesis. Naturê, 2642477-480. Robison, 4.c., R.V{. Butchet & E.W. Sutherland (1971) Cyctic AUP. Academic press, New york and London, pp. 53L. Rose¡, O.¡.f., R. Rangel-Aldoa & J. Erlichman (1977) Soluble cyclic AMP-dependent protein kinases : Review of the enzyme isolated from cardiac muscLe. fn : Current topics in celluIar regulation. B.L. Horecker & E.R. Stadtman (eils. ) Vol. 12 Academic press, New York, pp. 39-7 4. ))o

RosenbLoom, F.M., J.F. Hend.erson, I.C. Catdwell, W.N. Kel-l-èy & J.E. Seegmiller (1968) Biochemical basis of accelerated purine biosynthesis de novo in human fibroblasts lacking hypoxan thine- guan ine phosphoribosyl-transferase. J. Biot. Chem. , 243l. 116 6- 117 3 . Rora'e, P.B. & J.B. ?{yngaarden (1968) clutamine phospho- ribosylpyrophosphate amidotransferase . puri fication , substructure, amino acid composition and absorption spectra. ,J. Biol. Chem., 243=6373-6383. Rubin, H. e H. Sanue (1977) Complexes of ínorganj.c pyro- phosphaÈe, orthophosphate and calcium as stimulants of 3T3 cel1 multiplicaÈion. Proc. Nat. Acad. Sci. (U.s.A. ) , 74:5026-5030. Rudl-and, P.S., !V.8. Eckhart, D. Gospodarowj_cz & Ilf . Seifert (1974) Celt transformation mutants are not susceptible to growth activation by fibrobl_ast grohrth factof at permissive temperatures. Nature, 250:337-339. Ryan, W.L. & M.L. Heidrich (1968) Inhibition of cell | groh¡th in vitro by adenosine 3 , 5 ' -monophosphate. Science, !62: J-484-1485 . Sal-l, T., S. Mudd & J.C. Davis (1956) Factors conditioning the accumulation anil disappearance of metaphosphate in cells of Corynebacterium diphtheriae. Arch. Biochem. Biophys., 60:130-146. Sanford, K,K., W.R. Earle & c.D. Likely (1948) The gro\dth in vitro of single isol-ated tissue ceIls. J. Nat. Cancer Inst. , 9 z 229-246 . Santi, D.V., C.S. McHenry & H. Somner (1974) I{echanism of interaction of thymidylate synthetase with 5- fluorodeoxyuridylate. Biochemistry, l-3: 471-480. 230

Schmidt, G., L. Hecht e S.J. Thannhauser (1946) The enzymatic formation and thè accumulation of large amounts of a metaphosphate j-n baker's yeast under certain conditions. J. Bio1. Chem., 166:7?5-776- Seegmiller, J.8., F.M. Rosenbloom & W.N. Kelley (1967) Enzyme defect assocíated with a sex-linked human neurological disorder and excessj_ve purine synthesis. Science, 155:16 82-l-6 84. Sen, G.C., B, Lebleu, G.E. Brovrn, M. Kawakita, E. S1atteïy & P. Lengyel (1976) Interferon, double-stranded RNA and mRNA degradation. Nature, 264:3jO-372. Shields, R. & A. Korner (1970) Regulation of mammalian ribosome synthesis by amino acids. Biochem. Biophys. Acta.204:521-530. Shoaf, T. e M.E. Jones (1973) Uridylic acid synthesis in Ehrlich ascites carcinoma. Properties, subcellutar distribution and nature of enzyme complexes of the six biosynthetic enzymes. Biochemistry, 12: 4039-4051. Simmonds, H.4., c.S. Panayi & V. Corrigall (t978) A role for purine metabolism in the immune response : Adenosj.ne-deaminase activity and deoxyadenosine catabolism. lancet, 1:60-63. Skaper, S.D., R.Ð. V{illis & J.E. Seegmiller (1976) IntracelluLar 5-phosphoribosyl-J--pyrophosphate : Decreased availability during glutamine limítation. Science, 19 3:5I7-588. Sko1d, o. & A. zetterberg (1969) studies on the amino acid regulation of RNA-synthesis in rna¡runal_ian ce1ls in tissue culture. Expt, Celt. Res., 55:289-294. 23I

Smith, L.H.,Jr., C.M. Hugulêy,Jr. & J.A. Bain (1972) Hereditary orotic acj-duria. In : The metabolic basis of inherited diseases. J.B. Stanbury, J.B. wyngaarden e D.S. Frederickson (eds.) ¡I9 ¡a, Mccrarlr-Hull, Neþ, York, pp, 1003-1009. Smulson, M. (1970) Amino acid deprivation of human cells : Effects on RNA synthesis, RNA polymerase and ribonucleoside phosphorylation. Biochem. Biophys. Acta. , f99:537-540. Smulson, Iu.E. e ,J. Thomas (1969) RibonucLeic acid biosynthesis of human cells during amino acid deprivation. J. Bio1. Chem., 24425309-53!2- Snyder, F.F. & J.F. Henderson (1973) Altelnative pathways of deoxyadenosine and adenosíne metabolism. .f. Biol-. Chem., 2!925899-5904. Sokawa, Y, & Y. Kazj.ro (1969) Amino acid-dependent controL of the transport of <. -methyt glucoside in E. col"i. Biochem. Biophys. Res. ConInun., 34:99-103. Sokawa, Y., E. Nakao & Y. Kaziro (1968) On the nature of the control by RC gene in E. coli : Amino äcid- dependent controf of lipid synthesis. Biochem. Biophys. Res. Co¡rxmun., 33 : 10 8- 111. Stadtman, E.R. (1973) A note on the significance of glutamine in intermediary metabol-ism. In : Enzymes of glutamine metabolism. S. Prusi-ner & E.R. Stadtman, (eds.), Academic Press, NeI^' York, pp. 1-6. Stamrninger, G. & R.N. Lazzarínj- (1974) Altered metaboLism of the guanosine tetraphosphate, ppcpp, in mutants of E. coLi. CeII , l:85-90. Stanners, C.P., G.t. Elicein & H. Green (1971) Two types of ribosornes in mouse-hamster hybrid celLs. Nature (Lond. ) 19 71, 230252-54. 232

Stent, G.S. & S. Brenner (1961) A genetic Locus for the Tegulation of ribonucleic acid synthesís. proc. Nat. Acad. Sci. (U.S.A.l, 4722005-20l-4. Stephens, ,f .C., S.W. AÌtz & B.N. Ames (1975) Guanosíne 5r-diphosphate 3'-diphosphate (ppcpp) : positive effector for histidine operon transcripÈion and general signal for amino-acid deficiency. proc. Nat. Acad. Sci. (U.S.A. l, 7224389-4393. Stevenson, R.Iq., & H.B. L6John (1978) possible regulaticn of nucl-eoside transport and RNA synthesis in Achlya. Can. J. Biochem. , 562207-216. Stoker, M. & P. Abel (1962) Conditions affecting trans- formation by polyoma virus. CoId Spring Harbor sl¡mp. Quant. Biol., Ne14¡ vork, 27 =375-382. Sussman, A.J. e C. Gilvarg (1969) Protein turnover in amino aci-d-starved strains of Escherichia coli K-12 differing in their ribonucleic acid conÈro1 . J. Bio1. Chem. , 244:6304-6306. Sutherfand, E.l{. & T.Vil. RalI (1958) Fractionation and characteri zation of a cyclic ad.enine ribonucleotide formed by tissue particles. ,1. Bio1. Chem., 232: 10 77- 10 91 . Swanton, M. 6, c. Edlin (1972) fsoLation and characteriza- tion of an RNA relaxed mutant of B. subtilis. Biochem. Biophys. Res. Conmun., 46:.583-588. Sy, J. (1977) In vitro degradation of guanosine 5'.- diphosphate, 3'-diphosphate. Proc. Nat. Acad. Sci. (U.S.A. )' 74 :5529-5533. Sy, Y. & F. Lipmann (1973) Identification of the synthesis of guanosine tetraphosphate (MSI) as ínsertion of a pyrophosphoryl group into the 3'-positíon in guanosine 5'-diphosphate, Proc. Nat. Acad. Sci. (U.S.A.), 7O:306-309. Szymona, M, (7962a) purification and properties of the new hexokinase utilizing inorganic polyphosphate. Acta. Biochem. polon. , 9: 165-190. Szymona, Iu¡., O. Szymona & S. Kutesza (1962b) On the occurrence of inorganic polyphosphate hexokinase in some microorganisms. Acta. MicrbioL. pol-on., 11:287-300. szymona, o. & T. Szumilo (1966) Adenosine triphosphate and inorganic polyhosphate fructokinases of Mycobacterj_um phLei. Acta. Biochem. polon., 17: 129-144. TattersalL, M.H.N. e K.R. Harrap (f973) Changes in the deoxyribonucl-eoside triphosphate pools of mouse ],5178Y lymphoma celLs following exposure to methotrexate or 5- fl-uorcuracil. Cancer Res., 33: 3086-3090. Tattersall, M.H.N. & R.C. Jackson, T.A. Connors & K.R. Harrap (1973) Combination chemotherapy : The intetaction of methotrexate and 5-fLuorouracil. Eur. J. Cancer, 9 :7 33-7 39 . Taylor, ,t.H. (1960) Nucleic acid synthesis in Ìelation to the cel-l cyc1e. Ann. N.y. Acad. Sci. , 9OI4O9-42I. TeLlez-Inon, M.T. e H.N. Torrês (1970) lnterconverÈible forms of glycogen phosphorylase in Neurospora crassa. Proc. Nat. Acad. Sci. (U.S.A.), 66:459-463. Temin, H.M. (1966) Studies on carcinogenesis by avian sarcoma viruses III. The differential effect of serum and polyanions on multiplication of uninfected and convetted cel]s. ,1. Nat. Cancer Inst., 37 z167- 17s. Temin, H.M. e H. Rubin (1958) Characteristícs of an assay for Rous sarcoma virus and Rous sarcoma cetls in tissue culture. Virology, 6:669-688. 234

Tha¡nmana, P., R.R. Buerk c J. Gordon (1976) Absence of ppGpp production in synchronj-sed Balb,/c 3T3 cells on isoleucine starvation. FEBS Lêtt., 53:187-190. Todaro, G.J. & H. creen (1963) Quantítative studies of the growth of mouse embryo ce1ls in culture and their development into established lines. J. Cell. Bio1., !7¿299-313. Travers, A. (1973) The stringent response - 21 years on. In : Control of Transcription. B.B. Biswis, R.K. Mandal, A. Stevens and VÍ.E. Cohn (eds.) plenum Press, New York, pp. 67-92. TÌavers, A. (1976) Modulation of RNA polymerase specificity by ppcpp. Molec. Gen. cenet., ]-472225-232. Tsuboi, M., S. Kamisaka & N. Yanagishima (1972) Effect of cyclic 3',5' adenosine monophosphate on Èhe sporul- ation of saccharomyces cerevisiae, plant CelI. Physiol. , 13:585-588. Tucker, R.üI., K.K. Sanford & F.R. F.rankel (1978) Tubulin and actin in paired nonneoplastic and spontaneously transformed neoplastic ceLl l_ines in vitro : Ffuorescent antibody studies. Ce11, 13:629-642. Ullman, 8., M. Lee, D.I,l. Martin,Jr. & D.V. Santi (1978) Cytotoxicity of s-fluoro- 2 I -deoxyuridine : Require- ment for reduced fol-ate cofactors and antagonism by methotrexate. Proc. Ì¡at. Acad. Sci. (U.S.A.), 75:980-983. Valentine, W.N., D.E. Paglia, A.P. Tartaglia & F. cilsanz (f977) Hereditary hemolytíc anemia with increaseil red celI adenosine deaminase (4O-to 70-fol-d) and decreased adenosine triphosphate. Science, 195: 783-784. 235

Van Der Weyden, M.B. e W.N. Kelley (1974) il uman adenylo- succinate synthetase. Partial, purification, kinetic and regulatory properties of the enzyme from placenta. J. Biol . Chem., 249:.7282-7289. Van Ooyen, A.J.J., M. cruber & P. ,forgensen (1976) The ¡nechanism of action of ppcpp on rRNA synthesis in vitro. Ce11, 8:123-128. Vaughan, Iq.H.,Jr. (1972) Conparison of regulation of synthesis and utilization of 45S ribosomal precursor RNA in diploid and heÈeroploid human ceIls in response to valíne deprivation. Expt. Ce11. Res., 75223-30. viceps, D. e v.J. cristofalo (1975) Glutamine synthetase activity in WI-38 cell-s. J. Cel1. Physiol. , 86t9-I4. wahba, À.J. e M. Friedkan (1961) Direct spêctrophotometric evidence for the oxidation of tetrahydrofolate during enzymatic synthesis of thymidylate. J. Biol. Chem., 236 z I1-L2. Vlalsh, D.4., D.P. Perkins & E.G. Krebs (1968) An adenosine 3 ' , 5 '-monophosphate-dependent protein kinase fro¡n rabbiÈ skeletal muscles. ,J. BioL. Chem., 243:.3763- 3765. Wa1ters, R.A., R.A. Tobey & R.L. Ratliff (1973) Ce11-cyc1e dependent valiations of deoxyribonucleos ide tri- phosphate pools in Chinese hamster ce1ls. Biochem. Biophys. Acta. , 319:336-347. vùarner, A.H. & F.J. Finamore (1965) Isolation, purification, 3-diguanosine and characteri z ation of Pl, P 5'-tri- phosphate from brine shrirnp eggs. Biochem. Biophys. Acta. , 108:525-530. Weyer, V'1.J., de Boer, H.A., de Boer, J,G. & Gruber, M. (L976'l The sequence of pppcpp and ppGpp in the reaction scheme for magic spot synthesis- Biochem. Biophys. Acta., 442:123-I27. 236

lViame, J.Iq. (1947) The metachronatic reaction of hexamêta- phosphate. J. Am. Chem. Soc., 6923146-3147. Wiame, ,1.M. (1948) The occurrence and physiological behaviour of two metaphosphate fracÈions in yeast. J. Biol. Chem. , ]-782919-929. I{ind.er, F.c. e C. OrHara (1962) Effect of iron deficiency and zinc deficiency on composition of Mycobacterium smegmatis. Biochem. J., 82298-L08. ¡Vinsloltr, R.M. (1971) A consequence of the Ief g".re during glucose to l-actate do$¡n shíft in Escherichia coli. J. Biol. Chem. , 246:4872-4877. Wold, W.S.M. (1974) The cj.tric acid fermentation by Aspergillus niger : Regulation of growth, ceLl adhesion and citric acid. accumulation by zinc and adenosíne 3',5'-cycJ.ic monophosphate. ph.Ð. Thesis. U. of Manitoba. I{ong, T.J. & R.N. Nazar (1970) Relationship of the MS nucleotides to thê reguLation of ribonucl.eic acid synthesis in Escherichia ccjli. ,I. BioÌ. Chem., 245: 459I- 4595 . Wood, A.W. & ä.8. SeegmilJ.er (1973) properties of 5- phosphoribosyL- 1-pyrophosphate amidotran s fera se from human lymphobì.asts. J. Biol. Chem., 248:138-143. Ifyngaarden, J.B. (1972) Gluta:nine phosphoribosylpyro- phosphate amidotransferase. In : Current topics in cellular reguJ-ation, 5. B. Horecker & E. Stadtman (eds.¡. Àcademic Press, Neh' york, pp. 135-J.76. Íûyngaarden, J.B. (1972) Xanthinuria. Tn : The metaboLic basis of inherited diseases. J.B. SÈanbury, J.B. Wyngaarden & D.S. Frederickson (eds. ) ¡I9¡4., McGraw-Hi1l, Nehr york, pp. 992-1002. 237

!{yngaard.en, J.B. e D.M. Ashton (1959) The regulation of activity of phosphoribosylpyrophosphate amÍdo_ transferase by purine ribonucLeotides : a potentiaf feedback control of purine biosynthesis. J. Biol-. Chem. , 234 z 1492-1496 . Vüyngaarden, J.B. & W.N. Kelley (1972) eout. In : The metabolic basj.s of inherited disease. J.B. Stanbury, J.B. Wyngaarden & D.s. Fredrickson (eds.) :rd¡a., lilccraù¡-Hi11, Nev, yotk, pp. Bg9-968. Yang, H-L., G. Zubay, E. Urm, G. Reiness e $Í. Cashel (1974) Effects of guanosine tetraphosphate, guanosine pentaphosphate and f - Y ^tnylenyl-guanosine pentaphosphate on gene expression of Escherichia coli in vitro. proc. Nat. Acad. Sci. (U.S.A.), 7I263-67. Yoshida, A. e A. Yamataka (1953) On the metaphosphate of the yeast. J. Biochen. (Tokyo) , 40:95-94. Yoshida, M., A. Travers & B.F.C. Clark (1972) Inhibition of translation initiation iomplex formation by MS I. FEBS. Lett. , 23 z ]-63-166 . Young, H.A. e H.R. Whitely (1975) DeoxyribonucÌeic acid_ dependent ribonucleic acid polymerases in the dimorphic fungus Mucor rouxii. J. Biol. Chem., 2502479-487. Zeí!íï, C.E. & N.D. GoldberS (lg77l Ce] t- cycle-related changes of 3r:5'-cyclic cMp l_eveIs in Novikoff hepatoma celIs. proc. Nat. Acad. Sci. (U.S..A.), 74: I052-1056. p.T. Zielke, R.H., Ozand, .T.T. Til-den, Ð.A. SevdaÌian & M. Cornblath (1976) Groh'th of human diploid fibro_ blasts in the absence of glucose utilizâtion. proc. Nat. Acad. sci. (U.s.À.), 7Z:4110-4114. )10

Zollner, 2., W. crobner & A. Rauch-,.tanssen (1976) The effect of different purines and pyrirnidines on biosynthesis. In : Advances in experimental_ medicine and bíoLogy. 768. M.M. Muller, E. Kaiser & J.E. Seegmiller (eds.), ptenurn press, New york and London, pp. 269-27 3. Zoref., 8., A. de Vries & O. SperlinS (1975) Mutant feed- back-res istant phosphoribosylpyrophosphate synthetase associated with purine overproduction and gout. Phosphoribosyl pyrophosphate and purine metabolism in cultured fibroblasts. J. CLin. Invest., 56: I093-1099. Zvbay, G., D. Schwartz & J. Beckh'ith (1970) Ifechanism of activation of cataboL ite-sensitíve genes : A positive control system. Proc. Nat. Acad. Sci. (U.S.A. ) , 66:104-t10.