University Microfilms 300 North Zeeb Road Ann Arbor, Michigan 48106 a Xerox Education Company 72-21,030

INFORMATION TO USERS

This dissertation was produced from a microfilm copy of the original document. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the original submitted.

The following explanation of techniques is provided to help you understand markings or patterns which may appear on this reproduction.

1. The sign or "target" for pages apparently lacking from the document photographed is "Missing Page(s)". If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting thru an image and duplicating adjacent pages to insure you complete continuity.

2. When an image on the film is obliterated with a large round black mark, it is an indication that the photographer suspected that the copy may have moved during exposure and thus cause a blurred image. You will find a good image of the page in the adjacent frame.

3. When a map, drawing or chart, etc., was part of the material being photographed the photographer followed a definite method in "sectioning" the material. It is customary to begin photoing at the upper left hand corner of a large sheet and to continue photoing from left to right in equal sections with a small overlap. If necessary, sectioning is continued again - beginning below the first row and continuing on until complete.

4. The majority of users indicate that the textual content is of greatest value, however, a somewhat higher quality reproduction could be made from "photographs" if essential to the understanding of the dissertation. Silver prints of "photographs" may be ordered at additional charge by writing the Order Department, giving the catalog number, title, author and specific pages you wish reproduced.

University Microfilms 300 North Zeeb Road Ann Arbor, Michigan 48106 A Xerox Education Company 72-21,030

WHITE, Rodney Cecil, 1943- STUDIES ON CELL WALL METABOLISM OF THE GREEN ALGA CHLORELLA FYRENOIDOSA.

The Ohio State University, Ph.D., 1972 Biochemistry

University Microfilms, A XE ^OXCompany f Ann Arbor, Michigan

THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED. STUDIES ON CELL WALL METABOLISM OF THE

GREEN ALGA CHLORELLA PYRENOIDOSA

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Rodney Cecil White, B.A.

The Ohio State University 1972

Approved by

^ d v is or Department of Biochemistry PLEASE NOTE:

Some pages may have

indistinct print. Filmed as received.

University Microfilms, A Xerox Education Company ACKNOWLEDGEMENT

I wish to extend my appreciation to the entire faculty of biochem istry who always readily assisted me in the form of advice or use of equipm ent.

Very special thanks are extended to my advisor, Dr. George A.

Barber, whose interest and valuable consultations made this research p ossible.

Financial support was provided by an N .S.F. Traineeship,

Special thanks are algo given to my wife who always showed great fortitude and love throughout this work. VITA

1943 ...... Born-Houlton, Maine

1965 ...... B.A., Ricker College, Houlton, Maine

1968-1972 ...... N.S.F. Trainee, Dept, of Biochemistry, The Ohio State University, Columbus, Ohio

PUBLICATIONS

The Synthesis of 7(Adenosine 51 -Pyrophosphoryl) -D-Sedoheptulose by an Enzyme System from the Green Alga Chlorella Pyrenoidosa. R. C. White and G. A. Barber, Biochim. Biophys. Acta, In Press.

An Acidic Polysaccharide from the Cell Wall of Chlorella Pyr enoidosa. R.C, White and G. A. Barber, Biochim. Biophys, Acta, In Press. TABLE OF CONTENTS

P ag e

ACKNOWLEDGEMENT...... ii

VITA ...... iii

LIST OF TABLES ...... v

LIST OF FIGURES ...... vi

INTRODUCTION...... 1

LITERATURE REVIEW ...... 3 Synchronous Culture of Chlorella Enzyme Patterns in Synchronous Chlorella Cells Sugar Nucleotide Metabolism in Chlorella Chlorella Cell Walls

METHODS AND MATERIALS...... 17 Culture of Alga Paper Chromatography and Electrophoresis Gas-liquid Chromatography Gel Filtration and Ion Exchange Chromatography C hem icals Analytical Methods Analytical Enzymes

Experimental Procedures and R esults ...... 24 Specific Activity of Enzymes that Catalyze the formation of Sugar Nucleotides in Synchronous Chlorella Cultures ‘The Synthesis of 7(Adenosine 51-Pyrophosphoryl)-D-Sedohep- tulose by an Enzyme System from Chlorella pyr enoidosa Attempted Biosynthesis of Chlorella Cell Wall Polysaccharides Studies on Chlorella pyr enoidosa Cell Wall Polysaccharides

DISCUSSION...... 90

BIBLIOGRAPHY...... 97

IV LIST OF TABLES

Table P ag e

1. E nzym es Studied in Synchronous C u ltu re ...... 6

2. Specific Activities of Enzymes...... 33

3. Ratio of Adenosine/Phosphate/Sugar ...... 43

4. Effects of TPP and MgCl2 ...... 46

5. Results of Differentially Labeled l4C-Glucose-6- Phosphate Experiments...... 50

6. Apparent and V Values ...... 52 M m ax 7. Molar Ratios and Yields of Monosaccharides ...... 80

8. Chromatography and Electrophoresis of Aldobiuronic A c i d s ...... 88

v LIST OP PIGURES

Optical Absorbancy and Cell Numbers of the Syn chronous Culture ...... 26

Proposed Sequence of Reactions ...... 37

Structure of 7(Adenosine 51-Pyrophosphoryl)-D-Sedo- h e p tu lo s e ...... 41

Lineweaver-Burk P lo ts ...... 54

Extraction Procedure for Cell Wall Polysaccharides . 63

Elution Profile from the DEAE-Sephadex Column. . . 66

Fractionation Effected by the Bio-Gel P-300 Column. 69

Elution Pattern from the Sepharose 6B Column . . . . 71

Sedimentation Pattern of the Acidic Polysaccharide . 75

Plot of Rayleigh Interference Pattern D ata ...... 77

Results of L-Fucose Dehydrogenase Reactions . ... . 84

Elution Profile of Aldobiuronic Acids from Ag-lX8 Column ; . 87 INTRODUCTION

The plant cell wall, which functions predominantly to give struc tural rigidity to the cell, consists mainly of polysaccharides. The types of polysaccharides which are present are strain specific; how ever, most contain a high percentage of cellulose. Most of the varia tion encountered is in a group of polysaccharides often referred to as hemicelluloses or pectic substances. These mixtures of polysacchar ides are usually obtained from the wall in an impure form, i.e., small amounts of other biological material such as lipids and proteins are always present. The physical appearance and the chemical composi tion of many plant cell walls have been defined by a variety of techni ques, especially be electron microscopy (1), Although cell walls are reasonably well characterized, and are synthesized and deposited in large quantities, the means by which plants carry out this synthesis has remained essentially unsolved. This is the problem to which the research in this thesis was directed.

The unicellular green alga, Chlorella pyr enoidosa, which has a cell wall like that of many higher plants (2), was utilized in these stu dies, Chlorella cells are thought to be particularly useful for the stu dy df this problem because of a special characteristic in their growth cycle. The cell in the process of division forms 2, 4, 8, to as m any as

32 autospores which synthesize new cell walls in a relatively short period of time within the mother cell (3). By using the technique of synchronous culture, it was felt that enzyme preparations could be ob tained at the period of autospore wall formation and would contain an enriched quantity of wall-synthesizing enzymes.

Three aspects of the cell wall in synchronous Chlorella pyr enoi dosa were studied: (a) specific activities of enzymes which produce sugar nucleotides, (presumably cell wall precursors), (b) chemical composition of the cell wall, and (c) biosynthesis of cell wall poly saccharides. LITERATURE REVIEW

Synchronous Culture of Chlorella

In theory, the best m aterial to use for studying the physiological and biochemical events of a cell, especially those age-dependent events, would be a single cell. Since this is an extremely difficult task, the next best approach would be to grow cells in such a way that all the cells in a culture simultaneously pass through the same stage of the life cycle at one time. In order to achieve this type of synchronous be havior, the normal cell cycle has to be influenced in such a way that all the cells in a culture stop growing at a specific stage of their de velopment, Then, after the influencing factor has been removed, all the cells will resum e growth synchronously. Photo synthetic organisms, because of their natural diurnal behavior (4), are especially useful for achieving synchronous growth, i.e., the onset of darkness can be used as the synchronizing factor. Many techniques have been applied in order to achieve synchrony in a wide variety of cells and it has been the subject of a number of reviews (3, 4, 5).

Methods for the synchronous culture of Chlorella were first de veloped by Tamiya et_al_in 1953 (6) and later improved in 1961 (7).

Chlorella ellipsoidea cells were grown in liquid culture on a defined inorganic medium, CO-,, and lighted in such a manner as to produce maximum growth rates. Homogeneous populations of small cells were prepared from these heterogeneous cultures by differential centrifu gation. By submitting the homogeneous small cells to a light-dark se quence of 17 hrs light and 13 hrs dark, they were able to achieve a high degree of synchrony. Under those conditions the cells only divid ed in the dark and consistently yielded four daughter cells (4),

Pirson and Lorenzen in 1958 (8) developed a method using

Chlorella pyrenoidosa which involves the subjection of randomly grow ing cultures to a light-dark regime of 14 hrs light and 10 hrs dark.

After each cycle the cultures were diluted to a standard cell number with fresh medium. Almost complete synchrony was attained after two such cycles as evinced by the cells dividing only in the dark. The number of daughter cells formed per division varied between 8 and

3 2 (3 ).

Since the development of these techniques, many laboratories have applied them with minor variations. The length of the light-dark regime which is necessary to achieve synchrony depends on the strain of Chlorella being used and the growth conditions of the culture (3),

Once synchrony has been attained, the light-dark regime and dilution program must be continued or the cultures will rapidly become heter

ogeneous (5). However, Schmidt et^al (9), using a high temperature

strain of Chlorella pyrenoidosa were able to retain a high degree of synchrony over several cycles while the cultures were under continu

ous illumination.

Recently, a method has been reported which utilizes linear densi

ty gradients of Ficoll in order to obtain uniform cellular m aterial from

heterogeneous Chlorella cultures (10). These same researchers have

established that the density of Chlorella cells increases as they pro

gress through their life cycle.

Enzyme Patterns in Synchronous Chlorella Cells

The major part of the work done with synchronous Chlorella has

been on the physiological effects caused by changes in its environment.

These studies are extensive and have been covered in many reviews

(4,11,12,13).

Recently, however, synchronous cultures of Chlorella have been

increasingly utilized to study the level of individual enzymes and their

relation to specific cellular events. The enzymes that have been stud

ied are summarized in Table 1.

Schmidt et^al measured the level of the five enzymes (Table 1)

-involved in nucleic acid synthesis and observed that they all showed

characteristic sharp increases in activity just before nuclear divi

sion (9).

It was observed in the course of this research that UDP-glucose

dehydrogenase, which catalyzes the synthesis of UDP-glucuronic acid, TA B LE 1

CHARACTERISTICS OF SPECIFIC ACTIVITIES OF ENZYMES

STUDIED IN SYNCHRONOUS CULTURES OF CHLORELLA

I. Peaks during the light cycle aldolase (14) glyceraldehyde-3-phosphate dehydrogenase (14) n itrite red u c tase (15)

II. Peaks at the end of the light cycle glucose-6-phosphate dehydrogenase (14) glutam ic dehydrogenase (14) U D P-glucose dehyd rog enase (16)

III. Increases stepwise throughout the growth cycle acid phosphatase (17) alkaline phosphatase (17)

IV. Increases just before nuclear division aspartate transcarbamylase (9) glycinamide ribotide kinosynthetase (9) deoxycytidine monophosphate deaminase (9) deoxythymidine monophosphate kinase (9) dihydro-or ota se (9)

V. Remains constant throughout the growth cycle A DP -glucose pyrop h o sp h o ry lase (16) m alate dehydrogenase (14) G D P-glucose pyroph o sp h o ry lase (16) GDP-mannose pyrophosphorylase (16) U D P-glucose py ro ph o sp h o ry lase (16) U D P -L -rh am n o se sy nth etase (16)

6 shows a 19 fold increase in specific activity just before autospore for mation. Other sugar nucleotide-forming enzymes (Table 1) measured showed nearly constant specific activities throughout the growth cycle.

These observations suggest that UDP-glucuronic acid may play an im portant role in autospore cell wall formation.

By correlation of metabolite pool size at various stages of syn-

** . chronous growth to enzymatic activity, Bassham et al have shown that several enzymes of photosynthetic carbon metabolism have regulated activity (18). These methods measure the enzyme activity in vivo and probably are more reliable than the in vitro techniques. It is interes ting to note, however, that glucose-6-phosphate dehydrogenase shows maximum activity at the end of the light period by both methods (14, 18).

Transketolase, which catalyzes the transfer of a 2 carbon unit from a ketose to an acceptor aldose, was first described by Horecker et al in 1953 (19). Early work established that the donor could be a variety of ketoses and the acceptor a variety of aldoses; however, both must be phosphate esters (20). Villafranca and Axelrod, working with purified spinach transketolase, have shown that the phosphate ester specificity requirement does not hold since they obtained typical non- phosphorylated products from non-phosphorylated ketoses and aldoses

(21). Transketolase, in crude enzyme preparations of Chlorella py renoidosa, has been shown in the course of this research to transfer the two carbon unit from fructose-6-phosphate to the ribose moiety of 5(adenosine 5*-pyrophosphoryl)-D-ribose. The product of the reaction was identified as 7{adenosine 5* -pyrophosphoryl) -D-sedoheptulose (22).

This enssyme showed constant activity throughout the growth cycle.

Sugar Nucleotide Metabolism in Chlorella

Sugar nucleotides, first discovered by Leloir in 1949 (23), are ubiquitously distributed in nature and have been implicated in a large number of sugar transfer and interconversion reactions. These re actions have been extensively studied and are the subject of a number of reviews (24, 25, 26).

Sugar nucleotide metabolism was first studied in Chlorella by

Kauss and Handler in 1962 (27). They identified ADP-glucose and UDP- glucose from extracts of Chlorella pyrenoidosa and presented evidence that the ADP-glucose served as the glucosyl donor in starch synthesis, whereas UDP-glucose served as the glucosyl donor in sucrose synthe

sis, Later, Preiss et al isolated and partially purified starch synthe tase from Chlorella pyrenoidosa and showed that it was specific for

ADP-D-glucose (28).

Sanwal and Preiss (29) isolated the nucleotides from a large

quantity, (650 gms), of Chlorella pyrenoidosa grown heterotrophically

in the dark. They were able to identify the following sugar nucleotides:

UDP-D-glucose, UDP-D-galactose, UDP-xylose, UDP-arabinose,

UDP-N-acetylglucosamine, UDP-glucuronic acid, GDP-D-mannose, ADP-glucose, GDP-D-galactose, and ADP-arabinose. GDF-D-galactose and A DP-arabinose had not previously been reported to occur in plants, however, GDP-D-galactose has been reported to be present in animal tissues (30).

Barber and Chang (31) have reported the presence of an enzyme

system in Chlorella pyrenoidosa that catalyzes the reduction and epimerization of UDP-D-glucose to UDP-L-rhamnose. In bacterial sys tems this conversion proceeds via dTDP-D-glucose and the mechan

isms of this reaction has been studied by Glaser (32) and Gabriel (33).

Recently, Barber (34) has obtained evidence that Chlorella pyrenoidosa contains enzymes that catalyze the conversion of GDP-D-man-

nose to GDP-L-galactose. This transformation has previously been

shown to occur in enzyme extracts from the albumin gland of the snail,

Helix pomatia (35). The more common nucleotide derivative of galac

tose, UDP-D-galactose, has been observed to be synthesized when

UDP-glucose is incubated with enzyme extracts from Chlor ella pyre -

noidosa (31),

UDP-D-glucose dehydrogenase, UDP-D-glucuronic acid decar

boxylase, and UDP-D-xylose epimerase have been shown in this study

to be present in Chlorella pyrenoidosa. These enzymes catalyze the

following series of transformations:

1. UDP-D-glucose NAD y UDP-D-glucuronic acid C° 2 2. UDP-D-glucuronic acid ■ ■ « » UDP-D-xylose 10

3. UDP-D-xylose .... . ------> UDP-L-arabinose

This series of reactions has been shown to occur in many organisms

(36, 37) and probably provides the main source of these sugar nucleo tides for polysaccharide biosynthesis. This may not be strictly true however, since kinases and pyrophosphorylases that can catalyze the formation of these nucleotides from the free sugars have also been re ported in a few plants (38, 39). Sugar nucleotide pyrophosphorylases, that synthesize the common sugar nucleotides (UDP-D-glucose, CDP-

D-glucose, ADP-D-glucose, and GDP-D-mannose) from the appropri ate sugar-1-phosphate and nucleotide triphosphate, have also been found in Chlorella pyrenoidosa in the course of this investigation.

Thus, as indicated by the reports in the previous paragraphs,

Chlorella pyrenoidosa indeed is able to synthesize a wide variety of sugar nucleotides. The function of these compounds in Chlorella remains obscure, however, and one can only conjecture that they serve as the monosaccharide donors in the synthesis of polysaccharides.

The only synthesis described in that species that utilizes sugar nucleo tides is the synthesis of starch from ADP-D-glucose.

Chlorella Cell Walls

A. Physical structure and properties

Preston and coworkers have developed techniques by which the architecture of plant cell walls can be studied with the electron m icro- 11 scope (40), They have been able to show, by a series o£ carefully pre pared electron micrographs, that the plant cell wall consists of a series of microfibrils, with different orientations, all interwoven and embed ded in a complex m atrix (1, 40),

The isolated cell wall of Chlorella pyrenoidosa has been studied by the techniques of electron microscopy by Northcote et al (2), Their electron micrographs showed the wall to consist of two layers of m icro fibrils that are orientated at approximately 90° angles to each other; these in turn are interwoven and embedded in a continuous matrix.

These workers were able to separate the microfibrils from the matrix by alkali extraction,

Soeder (41), in an electron microscope investigation of the whole cells of Chlorella fusca [earlier referred to in his work as Chlorella pyrenoidosa (42)3, confirmed Nor the ote's work. He also presented

evidence, obtained by staining techniques, that the wall contains an

outer layer that consists predominately of lipid material.

B, Chemical properties of the cell wall

Northcote et al (2) analyzed cell walls isolated from Chlorella -

pyrenoidosa and found the dry weight to consist of 27$ protein, 9. 2$

lipid, 49.7$ carbohydrate, and 14$ inorganic material. When the walls were fractionated by alkali extraction, the insoluble m aterial was showi

to be predominately cellulose and the soluble material consisted of pro- 12 tein associated with polysaccharides referred to as hemicelluloses.

Glucosamine was present in acid hydrolyBates of the cell walls, and it was suggested that it is present in the walls in glycoprotein. By further fractionation of the alkali-soluble material, hemicellulose A, hemicellulose B, and starch were found to be present in the ratio of 9:2:1

(43). The hemicellulose A fraction was shown to be homogeneous by its electrophoretic mobility and ultracentrifugal properties and had a molecular weight of 35, 000 (43), Its monosaccharide composition was determined to be galactose, glucose, mannose, arabinose, xylose, and rhamnose in the molar ratios 7:1:2:3:3:5. The rhamnose was reported to be present in the L-configuration and the galactose in the D-config- u ratio n .

The glucosamine first reported by Northcote (2) was shown by

Mihara (44) to be present in the cell wall of Chlorella ellipsoidea as chitin. Cell walls were isolated at various stages of synchronous

growth and analyzed for glucosamine. This study showed that the in

crease in glucosamine content was proportional to total cell growth ex

cept at the stage when autospore walls were being formed within the mother cell. At this stage there was a rapid increase in the glucosa mine content due presumably to new cell wall synthesis.

In this investigation an acidic polysaccharide was isolated and

purified from the cell wall of Chlorella pyrenoidosa (45). It was shown to be homogeneous with an approximate molecular weight of 88, 000. 12 tein associated with polysaccharides referred to as hemicelluloses.

Glucosamine was present in acid hydrolysates of the cell walls, and it was suggested that it is present in the walls in glycoprotein. By further fractionation of the alkali-soluble material, hemicellulose A, hemi cellulose B, and starch were found to be present in the ratio of 9:2:1

(43). The hemicellulose A fraction was shown to be homogeneous by its electrophoretic mobility and ultracentrifugal properties and had a molecular weight of 35, 000 (43). Its monosaccharide composition was determined to be galactose, glucose, mannose, arabinose, xylose, and rhamnose in the molar ratios 7:1:2:3:3:5. The rhamnose was reported to be present in the L-configuration and the galactose in the D-config- u ratio n .

The glucosamine first reported by Northcote (2) was shown by

Mihara (44) to be present in the cell wall of Chlorella ellipsoidea as chitin. Cell walls were isolated at various stages of synchronous growth and analyzed for glucosamine. This study showed that the in crease in glucosamine content was proportional to total cell growth ex cept at the stage when autospore walls were being formed within the mother cell. At this stage there was a rapid increase in the glucosa mine content due presumably to new cell wall synthesis.

In this investigation an acidic polysaccharide was isolated and purified from the cell wall of Chlorella pyrenoidosa (45). It was shown to be homogeneous with an approximate molecular weight of 88, 000, 13

On complete acid hydrolysis, it yielded rhamnose, arabinose, xylose, mannose, galactose, and glucuronic acid in the molar ratios 11.5:2.7:

2.2:1. 0:2,9*.1.9. The rhamnose was determined to be present as the

L-enantimorph and the galactose as a 73:27 mixture of the D- and L-

enantiomorphs.

Funnett and Derrenbacker (46) isolated cell walls from three

strains of Chlorella (ellipsoidea, pyrenoidosa, and vulgaris) and

studied their amino acid composition. By washing their wall prepara

tions in 1M sodium chloride and detergent, followed by centrifugal

layering techniques, they observed a much lower protein content than

had been reported by Northcote et al (2). They proposed that these

techniques have to be applied in order to eliminate the protein and other

materials that are adsorbed to the walls during isolation. On acid hy

drolysis, each of these wall preparations yielded the same simple

pattern of eight common amino acids (aspartic acid, glutamic acid,

glycine, alanine, serine, valine, leucine, and isoleucine), It is hypo

thesized that these amino acids are hydrolyzed from small peptides in

the wall which may function like those found in the walls of gram posi

tive bacteria (47).

C. Biosynthesis of plant walls

The process by which plant cells carry out the synthesis of the

polysaccharides in their walls involves two basic problems; the first 14 is the means by which polymerization occurs from monomer units, and the second is how the large polymers are deposited outside a membrane which is impermeable to molecules as small as disaccharides (48).

Great progress has been made in understanding these problems in bacteria (47); however, very little is understood about this process in higher plants. The only fact that has been established with any cer tainty is that nucleotide diphosphate sugars serve as high energy car riers of the monomer units and that there is specificity with respect to the nucleotide as well as the sugar. There is no experimental evidence to indicate how the complex physical structure of the wall is attained.

Cellulose (j3-l, 4-glucan) which is the most abundant polysac charide in nature has been shown to be synthesized from GDP-D-glucose by particulate enzyme preparations from Phaseolus aureus (49) and cotton (50), It has also been suggested that UDP-D-glucose can also serve as the glucosyl donor for cellulose synthesis (51); however, this has been a subject of some controversy (26). The same particu late enzyme preparations from Phaseolus aureus that catalyze cellu

lose synthesis also catalyze the synthesis of a glucomannan from GDP-

D-glucose and GDP-D-mannose (52).

The synthesis of another group of plant polysaccharides that are

found in cell walls and often referred to as pectic materials have been

studied using particulate enzymes from Phaseolus aureus. UDP-D-

Galacturonic acid was shown to serve as the source of D-galacturonic 15 acid in the synthesis of pectic acid (53) and UDP-D-galactose was uti lized to synthesize a water soluble galactan (54).

Xylans and hemicellulose B (polymers of xylose and glucuronic acid) have been found in the cell walls of corn cobs (55). Particulate enzymes prepared from immature corn cobs utilize UDP-D-xylose (56) and UDP-D-glucuronic acid (57) in the synthesis of these polymers.

In algae (both marine and freshwater), which contain a wide variety of polysaccharides (58) in their walls, only one cell-free syn thesis has been achieved. Alginic acid, which is the predominant poly saccharide of the brown algae Fucus gardneri, was shown to be synthe sized from GDP-D-mannuronic acid by particulate enzymes from that organism (59). The other constituent of alginic acid, guluronic acid, was presumed to be derived from GDP-D-mannuronic acid via an epimerization to GDP-D-guluronic acid with subsequent transfer to the polymer (Z6).

All of the reported syntheses of wall polysaccharides in plants have been established by measuring the transfer of only small quanti ties of radioactive sugars, and in no case have the polymers been shown to be identical to those found in nature. The synthesized polymers have always been characterized by hydrolytic methods i.e., compari son of the hydrolytic products of the radioactive material with authentic material as to the type of monosaccharide units and type of glycosidic linkage present. Due to the restriction of available methods, the syn- 16 thetic products have not been studied at the polymer level.

Lipid (60,61,62,63), glycoprotein (62), and nucleotide-polysac charide complexes (64) have been postulated as intermediates in plant cell wall biosynthesis. Although it has been established that lipid in termediates function in bacterial systems (65,66), there has been no substantial evidence for the existence of such an intermediate in plant cell wall synthesis. This aspect of polysaccharide biosynthesis still remains essentially unsolved. METHODS AND MATERIALS

Culture of Algae

Chlorella pyrenoidosa Chick strain No. 395 was purchased from the Culture Collection of Algae, Department of Botany, University of

Indiana, Bloomington. The cells were first cultured aseptically at 25° on 1.5 x 12 cm slants of a salts medium (3) containing 1.5$ agar 6 inches from the light of two 40 watt "Grolux" bulbs (Sylvania). The cells from one such slant incubated for 14 days were aseptically intro duced into 2 1 of the sterile salts medium in a "Microferm" fermentor

(New Brunswick Scientific Co.). Illumination was provided by a sub merged, water-jacketed, quartz lamp (General Electric T-3, Xmax,

1200 mjjt) operated at 1000 watts. The medium was agitated by an im - pellor at 180 rpm and was gassed through a sparger with 1.5$ C02 in air at about 3 1/min, The fermentor vessel and salts medium were sterilized by autoclaving for 3 hrs at 120° while the lamp assembly was soaked overnight in a solution of 800 ppm benzalkonium chloride

("Roccal,11 Winthrop Laboratories). It was washed with sterile water before submersion in the fermentor.

After inoculation, the culture was incubated at 28° with constant illumination for 96 hrs. Ten liters of fresh medium were added to the

17 18 fermentor, and it was subjected to one cycle of 18 hrs light and 12 hrs dark. (Complete darkness was obtained by enclosing the fermentor vessel in a bag made of black felt enclosed in black cotton cloth.) Ten liters of the culture were then removed and replaced with fresh medi um, and the light/dark cycle was repeated. The dilution and light/

dark cycles were repeated a total of three times after which the cells appeared by microscopic examination to be nearly synchronised (3).

After the last dilution, the culture was incubated through 3 more com

plete light/dark cycles (90 hrs) without dilution. The cell density had

then reached about 3 x 107 cells/m l. Cells were harvested by centri

fugation in a Sharpies centrifuge and washed two times by suspension

in cold water and centrifugation at 7000 x g in a Sorvall refrigerated

centrifuge. The yield of cells was about 0,6 g wet weight/1 of culture.

The centrifugal pellet was frozen and stored at -60°.

When heterogeneous populations of cells were required, the 2 1

subculture was diluted with 10 1 of fresh medium at the end of four days

growth and then kept under continuous illumination until the cell density

reached about 3 x 10 cells/m l (about 3 days more). These cells were

harvested as described above. The cells were usually lyophilized and

stored at -15°. The yield was about 0. 09 g dry weight/liter.

Paper Chromatography and Electrophoresis

Partition Chromatography, unless otherwise indicated, was per- 19 formed on Schleicher and Schuell No. 589 Blue Ribbon paper in the fol lowing solvent systems: Solvent I, n-propanol/ethyl acetate/water

7:1:2; Solvent II, 95$ ethanol/lM ammonium acetate 7:3; Solvent III, isobutyric acid/1M ammonium hydroxide 10:6; Solvent IV, phenol/ water 80:20; Solvent V, ethyl acetate/pyridine/water 12:5:4; Solvent

VI, ethyl acetate/formic acid/acetic acid/water 18:3:1:4,

High voltage electrophoresis was carried out on Schleicher and

Schuell No. 589 Orange Ribbon paper on a flat plate apparatus (67) at about 25 v/cm or in a Savant Model LT-20A tank unit at 80 v/cm . The following buffer systems were used: 0. 1M ammonium formate, pH 3.7;

0.05M sodium tetraborate, pH 9. 0; 0. 5M sodium arsenite, pH 9.6.

Gas -liquid Chr omatography

Monosaccharides obtained from polysaccharide hydrolysates were reduced to the alditols and acetylated as described by Albersheim (68).

Small quantities of these were then injected into a 0.3 x 120 cm glass column containing 0. 4$ silicone, 0. 2$ ethylene glycol succinate, and

0.2$ neopentyl glycol adipate on "adiatoport S" (80/100) in a Hewlett-

Packard Model 402 gas chromatograph. At injection the flash heater was 200° and the flame detector was 250°. The oven m s preheated to

110° and held at this temperature for 6 mins after injection, then a 1° rise/m in was continued to 180°. The monosaccharides were identi fied and quantitated by comparison of their positions and the areas un- 20 der their peaks to known quantities of authentic compounds.

Gel Filtration and Ion Exchange Chromatography

Bio-Gel P-300 (CalbiochemJ’was swollen for 24 hrs in 0. 05M

Tris-HCl/1M NaCl buffer, pH 8,3. Fine particles were removed by decanting the suspension several times, A 2.5 x 50 cm column (275 ml bed volume) was packed by gravity flow, and a small amount of

Sephadex G-25 was layered on the top. It was equilibrated with 500 m l of the Tris/NaCI buffer before use.

Sepharose 6B (Calbiochem) was diluted to a thin slurry with 0. 05M

Tris-HCl/1M NaCl, pH 8,3, and packed by gravity flow into a 1,5 x

43 cm column (50 ml bed volume). The column was washed with 1 1 of the Tris/NaCl buffer. f-. | [ DEAE-Sephadex A-25-1£0 (Pharmacia) was swollen for two hrs in 0. 05M Tris-HCl, pH 8.3, and then packed by gravity flow into a 1.5 x 35 cm column (40 ml bed volume). The column was equilibrated with

500 ml of 0.05M Tris-HCl, pH 8.3.

DEAE-Cellulose was swollen in 1M NaOH, washed with water,

0. 5M HC1, and then again with water. This cycle was repeated two times, and the fines were then removed by decantation of the suspen sion. A 5 x 25 cm column was packed by gravity flow and equilibrated with 1 1 of 0. 005M sodium phosphate buffer before use. 21

C hem icals

a-D-Glucose-U i it C-l -phosphate, a-D-mannose-U 14 C-1-phosphate,

UDP-D-glucose-U14C, UDP-D-galactose-Ul4C, GDP-D-mannose-UI4C,

UDP-L-rhamnose-U 1 it C, and GDP-D-glucose-U 14 C were prepared as

described previously (31,49*69).

UDP-D-Glucose, uniformly labeled in the uridine moiety with

C14, was prepared from UTP-U14C (338 (iC//xmole) and a-D-glucose-

1-phosphate by the action of the 50-70$ (NH^SC^ fraction prepared

from Phasoleus aureus (70). It was purified by electrophoresis at

pH 3. 7 followed by chromatography in Solvent II.

D-Fructose-U^C-b-phosphate was prepared from uniformly

l4C-labeled D-fructose (100 fiC/pnole) by treatment with hexokinase

and ATP (71). It was purified by electrophoresis at pH 3,7.

D-Glucose-1 -14C-6-phosphate (6.2 jiC/pnole) and D-glucose-6-

14C-6-phosphate (7.0 fiC / jimole) were prepared by the enzyme-catal

yzed phosphorylation of the respectively labeled sugars (71). The

radioactive compounds were purified by paper electrophoresis at pH

3,7 followed by chromatography in Solvent II,

UDP-Glucuronic acid -U14C‘ was prepared from uniformly R e

labeled UDP-glucose (62 y C / jitmole) by the action of the 35-60$ (NH4)2

S04 enzyme fraction prepared from Chlorella pyrenoidosa. It was

purified by electrophoresis at pH 3.7 followed by chromatography in

Solvent II. 22

A mixture of sedoheptulosan and D-sedoheptulose was produced by treatment of sedoheptulosan (Sigma) with 0.05M HC1 for 30 min at

100° (72).

5(Adenosine 5' -pyrophosphoryl) -D-ribose was cleaved from NAD by the action of Worthington's DPNase (73). The reaction mixture con tained 2.5 /imoles NAD, 15 units DPNase, 0.75 pnoles sodium/potas - sium phosphate buffer, pH 7. 0. It was incubated for 45 min at 37°, and

5(adenosine 5' -pyrophosphoryl) -D-ribose was isolated by electrophor esis at pH 3.7,

L-Galactose and D-rhamnose was kindly supplied by Dr. Nelson

Richtmyer. L-Rhamnulose was prepared by the method of Wilson and

A jl (74).

All other reagents were purchased from commercial sources.

Analytical Methods

Radioactive compounds on paper were located by exposure to x- ray film (Kodak, no-screen). Quantitative estimations of l4C were made by immersing the sections of paper containing the I4C-activity in 0. 01 $ dimethyl POPOP, 0. 4$ PPO in toluene, and counting the samples in a Packard Liquid Scintillation Spectrometer, Model 3320,

Non-radioactive sugars were located on paper with the silver nitrate reagent, p-anisidine phosphate, or benzidine-periodate (75).

Protein concentrations were estimated by the Biuret method (76), sedoheptulose by the method of Dische (77), and total phosphate by the 23 method of Ames and Dubin (78).

Analytical Enzymes

L-Rhamnose isomerase was prepared by the method of Wilson and Ajl (74) and used as described previously (79). D-Calactose de hydrogenase (80) was purchased from Sigma. L-Fucose dehydrogenase, which also reacts with L-galactose, was prepared from fresh pork liver through the DEAE-cellulose stage (81). The dehydrogenase acti vities were monitored by following the increase in optical absorbancy at 340 nm of NADH in the reaction mixtures. EXPERIMENTAL PROCEDURES AND RESULTS

A. Specific Activity of Enzymes that Catalyze the Formation of

Sugar Nucleotides in Synchronous Ghlorella Cultures.

Description of Chlorella Cells Utilized in These Studies

Cells of Chlorella pyrenoidosa were synchronously cultured as described in the methods section. After the cell density had reached

2 x 107 cells/m l, the subsequent cycle was sampled by harvesting 2 liters of culture at 0, 9, 18, 23, and 30 hrs. The cells were collected immediately after sampling by centrifugation at 1000 x g in a refriger ated Sorvall Centrifuge, washed twice with cold water, and the centri fugal pellet was frozen and stored at -60°. The yield was approxi mately 1 gm of wet packed Chlorella cells per sample. An aliquot of each sample was used to measure the optical absorbancy and to deter mine the number of cells present. These resultB are shown in Figure

Preparation of Chlorella Enzymes

The previously frozen samples were thawed and suspended in

10 ml of cold buffer consisting of 0 .1M sodium/potassium phosphate, pH 7.0, 0. 05M j3-mercaptoethanol, and 5$ soluble polyvinylpyrroli-

24 Figure 1: Graphical representation of the number of cells and the optical absorbancy of the culture through one complete synchron ous cycle. The arrows represent the onset of darkness; O represents the cell number; and A represents the optical absorbance at 652 nm.

25 oo o> ro ro OPTICAL OPTICAL ABSORBANCY AT 6 5 2 nm CELL NUMBER PER ml OF CULTUREXIO7 Ul 0 1

HOURS OF GROWTH CYCLE 92 done or 0.05M Tris-Succinate, 0.001M EDTA, 0.001M dithiothreitol, pH 7,0, Large aggregates of cells were disrupted by briefly homo genizing the chilled suspension in a glass tube with a Teflon pestle.

The cells were broken by two passages through a cold French pressure cell (Aminco) at about 15,000 psi. All subsequent operations were con ducted in the cold. The homogenate was centrifuged at 10, 000 x g for

10 min, and the residue was discarded. Proteins were precipitated from the supernatant solution with a saturated neutral solution of am monium sulfate (Mann, specially purified). The fraction that precipi tated between 35 and 60/6 saturation was found to contain the enzymes of interest. That precipitate was dissolved in 0.3 ml of 0. 0E5M sodi ums/potassium phosphate buffer, pH 7. 0, and desalted on a 1. 5 x 20 cm column of Sephadex G-25 (fine). The effluent protein solution (3 ml) was collected, kept at 4°, and utilized immediately for enzymatic as says and protein determinations.

Identification of Sugar Nucleotides

In these studies radioactive substrates were utilized to measure the enzymatic activities by following the rate of production of the radio active product. The enzymatically produced sugar nucleotides, uni formly 14C-labeled in the sugar moiety, were identified by the follow ing procedures:

1. Paper electrophoresis in 0.1M ammonium formate, pH 3.7. 28

2. Paper chromatography in Solvent II and III.

3. Acid hydrolysis in IN HC1 for 15 min, at 100°, with subsequent

paper chromatography in Solvent I.

In each of the assays described in the next section, the 14C-label

ed sugar nucleotide was shown to be indistinguishable from the authen

tic compound when the papers were exposed to x-ray film,

UDP-Glucuronic acid-U 14 C was characterized by the following

c rite ria :

1. Incubation with Sigma's nucleotide pyrophosphate (0.2 units

Sigma nucleotide pyrophosphatase, 0.7 fjmole Tris -HC1, pH 7. 5, and

1.5 /imoles MgCl2 in a total volume of 30 /il, incubated-6 0 min at 37°),

produced a 14C-labeled compound with the identical mobility of glucur

onic acid-1-phosphate upon electrophoresis at pH 3.7.

2. Further incubation of the above l4C-sugar acid phosphate with

alkaline phosphatase (0.3 units of Sigma calf intestinal mucosa phospha tase, 2 fimoleB Tris-HCl, pH 8.2, l.Sjimoles MgCla, incubated 60 min at 37°) yielded a l4C-compound which had the identical mobility of glu

curonic acid upon electrophoresis at pH 3.7.

3. Upon heating the sugar acid in IN HC1, at 100° for 20 min, a l4C-compound was obtained which was identical with glueuronolactone upon chromatography in Solvent I.

Methods for Assaying Enzymes

1. UDP-Glucose pyrophosphorylase assay. 29

Reactions were carried out in thin walled glass capillaries and each tube contained the following reactants: 0.2 pnole -glucose-

Ul4C -1 -phosphate (0.05 flC), 0.2 pnole UTP, 0.2 pnole MgCl2 , 0.03 mg of Chlorella enzyme and 0.5 pnole of sodiums/potassium phosphate buffer, pH 7.0, in a total volume of 30 fjl. The mixtures were incubat ed at 25° for 3, 6, and 9 min at which times the reactions were stopped by placing the sealed capillaries in a 100° water bath for 2 min. The reaction mixtures were then applied directly to a sheet of.paper m ois tened with ammonium formate buffer, pH 3.7 and electrophoresed.

The areas of the electrophoretogram corresponding to UDP-glucose-U

14C (detected by UV absorption of standard UDP-glucose) and the un- reacted glucose-U 14 C-1-phosphate were cut from the paper and the 14 C- content of each was quantitated in a liquid scintillation spectrometer.

The l4C-activity recovered in the substrate and product represented

92$ of the total. The percent of the total recovered l4C-activity found as UDP-glucose was used to calculate the reaction rate in pnoles pro d u c t/m in .

2. A DP-Glucose pyrophosphorylase assay

The same procedure was used as described for UDP-glucose py rophosphorylase except that the reaction mixture contained: 0.1 ptmole

G£-D-glueose-Ul4C-1 -phosphate (0.05 fj,C), 0.2 pnole ATP, 0.15 pnole

MgCl2l 0.15 pnole 3-phosphoglyceric acid (when included), 0.03 mg of

Chlorella enzyme and 0.8 pnole Tris-succinate buffer, pH 7.0, in a 30

total volum e of 30 jil.

3. GDP-Glucose pyrophosphorylase assay

The same general procedure was used aB described for UDP-

glucose pyrophosphorylase except that the reaction mixture contained:

0.1 jimole a-D-glucose-U14C -l -phosphate (0.05 jiG), 0.15 jimole GTFf

0,15 jimole MgCl2, 0.03 mg Chlorella enzyme, 0.5 j/mole sodium/potassium phosphate buffer, pH 7. 0, in a total volume of 30jil. The iso - lation procedure required an additional step since GDP-glucose and

glucose-1-phosphate were not separated by electrophoresis. They were

eluted from the electrophoresis paper and chromatographed in Solvent

II. This chromatographic step achieves the separation of glucose-1- phosphate from the GDP-glucose.

2. GDP-Mannose pyrophosphorylase assay

The same procedure was used as described for GDP-glucose pyro phosphorylase except for the contents of the reaction mixtures. Each m ix ture contained 0.2 jimole c-D-mannose-U14C -1 -phosphate (0,04 jiC), 0.2

jimole GTP, 0.2 jimole MgClz, 0.03 mg of Chlorella enzyme, and 0.5 •J- *(* jimole Na /K phosphate buffer, pH 7.0, in a total volume of 30 jil.

5. UDP-L-Rhamnose synthetase

The procedure used for UDP-glucose pyrophosphorylase was mo

dified as follows: Each reaction mixture contained 0.005 jimole UDP-

glucose-U14C (0.05 JiC), 0. 15 jimole NADPH, 0.03 mg Chlorella enzyme,

and 1. 0 jimole Tris-HCl buffer, pH 8.2, in a total volume of 36 jil. The 31 area of the electrophoretogram corresponding to.UDP-hexose was cut from the paper, eluted with water, evaporated to dryness, and after acid hydrolysis in IN HC1, at 100° for 15 min, was chromatographed in Solvent I. The areas corresponding to glucose and rhamnose were cut out and their 14 C radioactivities . estimated « in • a liquid scintillation counter. Conversion to L-rhamnose was expressed as that percent of the total radioactivity recovered on the chromatogram as L-rhamnose.

6. UDP-Glucose dehydrogenase

Each reaction mixture contained 0. 5 ptmole NAD, 0.25jiimole

UDP-glucose, 1,4 mg Chlorella enzyme, and 75 jtfnoles sodiuny'potassium phosphate buffer, pH 7. 0, in a total volume of 0.335 ml. The reactants were mixed in 0.4 ml spectrophotometer cuvettes and equi librated at 25°. The reaction was started by the addition of UDP-D- glucose and followed continuously by charting the change in optical absorbancy at 340 nm with a Guilford Recording Spectophotometer.

When glucose, glucose-1-phosphate, or glucose-6-phosphate was sub stituted for UDP-D-glucose in the reaction mixture, no reaction was observed. Also, when NADH was incubated with the Chlorella enzyme no measurable oxidation rate was ob served. The observed increase in

optical absorbancy when UDP-D-glucose was the substrate was pre

sumed to represent the production of NADH. Since there are 2 NADH molecules produced per oxidation of each molecule of UDP-D-glucose, half of the quantity of NADH produced represents the amount of UDP- 32

D-glucuronic acid formed. To characterize the UDP-D-glucuronic

-acid formed, a small scale reaction was conducted in which UDP-D-

glucose-Ul4C was incubated with NAD. The resultant UDP-D-glucuron- 14 ic acid-U C was isolated by paper electrophoresis in 0. 1M ammonium

formate and subjected to the identification procedures described earlier

in this section.

Determination of Enzymatic Activities

Enzymes were prepared from the Chlorella cells that had been

harvested at specific stages of the growth cycle. These enzyme pre

parations were subsequently subjected to the six enzymatic assays des

cribed in the previous section and the results are shown in Table 2.

The only enzyme which showed variation in activity during the

growth cycle was UDP-glucose dehydrogenase. A 19 fold increase in

specific activity was observed at the end of the light cycle. This acti

vity was not inhibited when extracts from earlier growth stages were

added to the reaction mixture nor could the enzyme preparations from

the earlier stages be stimulated by inactive extracts from the later

stages. These results suggest that there is an age dependent synthesis

of UDP-glucose dehydrogenase which produce UDP-D-glucuronic acid

for a specific use in the later stages of development.

Metabolism of UDP-D-Glucuronic Acid

UDP-D-Glucuronic acid-U14C was prepared as described in the TABLE 2

SPECIFIC ACTIVITIES OF ENZYMES MEASURED AT DESIGNATED TIMES

THROUGHOUT THE GROWTH CYCLE

Enzyme Specific Activity (jumole product/min/mg protein)

0 hrs 9 h rs 18 h rs 23 h rs 30 h rs

UDP-glucose dehydrogenase 0.022 0.282 0.382 0.163 0.025

UDP-L-rhamnose synthetase3. 0.215 0.227 0.221 0.218 0.202

UDP-glucose pyrophosphorylase 0.285 0.304 0.317 0.300 0.290

GDP-glucose pyrophosphorylase 0.019 0.021 0.023 0.021 0.020

ADP-glucose pyrophosphorylase 0.021 0.020 0.019 0.020 0.022

A DP -glue os e pyr ophos phorylas e 0.130 0.125 0.120 0.119 0.137

GDP-mannose pyrophosphorylase 0.032 0. 030 0.029 0.032 0.030

n mole product/min/mg protein ^reaction mixture contained 3 -phosphogly eerie acid 34 methods and utilized as a substrate in reactions conducted with parti

culate, 0-30$ (NH4)2S04, . and 30-60$ (NH^SO* enzyme fractions pre pared from 18 hour Chlorella cells.

Reaction mixtures contained 5 x 10*"4 (imole UDP-D-glucuronic acid-U14C (0,04 fiC), 50 jig boiled whole cell extract, 0.15 jjmole MgCl2,

0.2 mg 0-30$ (NH4)2S04 Chlorella (or 0. 2 mg 30-60$ (NH4)2S04 Chlorella enzyme or 0.4 mg particulate Chlorella enzyme), and 0.5 pnole sodium/potassium phosphate buffer, pH 7.0, in a total volume of 30 fx1, A reaction was also included in which all the enzyme fractions were com bined and incubated with the above reactants. These reactants were in cubated at 37° for 45 min and then applied directly to a paper moistened with ammonium formate, pH 3.7, and electrophoresed. When the re sulting electrophoretograms were exposed to x-ray films, each re-

1 4 action showed the same C-activity pattern with varying degrees of in tensity. These 14C-active areas were cut from the electrophoretogram, eluted with water, evaporated to dryness in vacuo, and analyzed by paper chromatography in Solvent II and III. Monosaccharides were identified by subsequent acid hydrolysis of the nucleotides and sugar phosphates in IN HC1 for 20 min at 100° followed by paper chromato- grephy in Solvent I. UDP-D-Xylose, UDP-L-arabinose, and hydroly tic products thereof were found to be present as reaction products.

Their presence suggests that the. main metabolic function of UDP-D- glucuronic acid is the following reaction: 35 co2 UDP-D-glucuronic acid ^ ^ UDP-D-xylose

The enzyme preparations from the various stages of the growth cycle of Chlorella appeared to catalyze the reaction at the same rate. The yield of this reaction was approximately 20$ and accurate m easure ment was difficult because at least 50$ of this enzymatic activity was shown to reside in the particulate fraction.

B. The Synthesis of 7(Adenosine 5*-pyrophosphoryl)-D-Sedoheptulose by an Enzyme System from Chlorella pyrenoidosa.

Initial Observation of the Reaction System

t In the course of developing an assay system for UDP-L-rhamnose synthetase (described in part A) UDP-D-glucose-U14C and NADH (nor mally NADPH was used) were incubated with Chlorella enzyme ex tracts. Analysis of the reaction products revealed the presence of a new radioactive compound which contained approximately 30$ of the

original radioactivity. When subjected to paper electrophoresis in am

monium formate, pH 3, 7, paper chromatography in Solvent I, II and

III, and hydrolysis with acid and enzymes, it had the properties of a

sugar nucleotide. However, neither the nucleotide nor the sugar phos phate and sugar derived thereof could be immediately identified.

The following series of observations led to the identification of

the immediate precursors of this substance: 36

1. When UDP-D-glucose labeled with 14C in the uridine moiety was substituted for UDP-D-glucose labeled with 14C in the D-glucosyl moiety, the 14C-labeled nucleotide was not formed;

2. NAD stimulated the reaction more than NADH. NADP andNADPH were without effect;

3. a-D-Glucose-U14C-l-phosphate, D-glucose-U14C-6-phosphate, D- fructose-Ul4C-6-phosphate each served as a 14C-donor, but the initial rate of 14C-incorporation was greater from D-fructose-6-phosphate than from the other two sugars;

4. Incubation of only NAD with the Chlorella enzyme produced a new

UV-absorbing compound, which, when isolated and purified could be substituted for NAD as cofactor in the original reaction;

5. 5(Adenosine 5'-pyrophosphoryl)-D-ribose, obtained from NAD by the action of Worthington DPNase, also served as the cofactor.

From these observations, the sequence of reactions shown in Fig ure 2 was proposed.

Preparation of Algal Enzyme

In a typical preparation, 8.0 grams of thawed cells were sus pended in 30 ml of a buffer consisting of 0.1M sodium/potassium phos phate, pH 7.0, 0. 05M jS-mercaptoethanol, 0.001M EDTA and 5$ solu ble polyvinylpyrroline. Large aggregates of cells were disrupted by briefly homogenizing the chilled suspension in a glass tube with a UDP-D-glucose NADH

D-glucose-1 -phosphate NAD %

D-glucose-6 -phosphate 5(adenosine 5' -pyrophosphoryl) D -rib o se

D-fructose -6 -phos phate

adenosyl nucleotide sugar

Figure 2: Series of reactions that are proposed to occur in the crude Chlorella enzyme preparations and ultimately produce the im mediate precursors for the formation of the unknown nucleotide sugar,

37 38

Teflon pestle. The cells were broken by two passages through a cold

French pressure cell (Aminco) at about 15, 000 psi. All subsequent operations were conducted in the cold. The homogenate was centri fuged at 10, 000 x g for 10 mins, and the residue was discarded. Pro teins were precipitated from the supernatant solution with a saturated neutral solution of (NH4)2S04 (Mann, specially purified). The fraction obtained between 45 and 60$ saturation was found to contain the enzyme(s) of interest. That precipitate was dissolved in 0.5 ml of 0.025M sodiurn/potassium phosphate buffer, pH 7. 0, and desalted on a 1.5 x

20 cm column of Sephadex G-25 (fine). The effluent protein solution

(3 ml) was frozen and lyophilized immediately. The yield was 34 mg of powder of which 18 mg Was protein [estimated by the Biuret reaction

(76)], The material was stored at -60°.

Characterization of the Product

A quantity of the compound sufficient for characterization was pre pared by incubating the following mixture for 3 hours at 37°: 0.5 jimole

D-fructose-l4C-6-phosphate, (0.1 jiC), 0.6 pmole 5(adenosine 5' -pyro phosphoryl) -D-ribose, 3 mg Chlorella lyophilized 45-60$ (NH4)2S04 fraction, 16 jttmoles Tris-HCl, pH 8.2, in a total volume of 0,2 ml.

The mixture was applied to a sheet (18 x 60 cm) of Schleicher and

Schuell No. 589 Orange Ribbon paper moistened with ammonium for mate buffer, pH 3.7, and electrophoresed in that buffer. The UV ab sorbing compound that migrated slightly less rapidly than A DP was eluted with water, and the eluate was evaporated to about 50 |Lll in vacuo.

(Evaporation to dryness in the presence of residual ammonium for mate caused breakdown of the compound,) The eluate was applied to a

9 x 45 cm cheet of Schleicher and Schuell Orange Ribbon paper and chromatographed with Solvent III for 50 hours. In that time, the 7(adenosine 5*-pyrophosphoryl)-D-sedoheptulose was separated from 5(adenosine 5' -pyrophosphoryl) -D-ribose by about 3 cm. The residuum of isobutyric acid from the solvent was removed from the paper by irri gating it in the chromatography tank with 80$ ethanol for 8 hours. The compound was located by its radioactivity and UV absorbance and eluted with water. The yield, estimated by UV absorption at 259 m/x

[extinction coefficient of ADP-D-glucose (82)] was 0.22 junaole. In the subsequent microchemical analyses the eluate of a blank chromatogram identically treated was used to correct for. the contribution of im puri ties from the paper and solvent systems.

The structure of the compound was concluded to be 7(adenosine

5 '-pyrophosphoryl)-D-sedoheptulose (Fig. 3) on the basis of the fol lowing evidence:

1. Upon electrophoresis at pH 3.7 and chromatography in Sol vent II, the 14C labeled, UV -absorbing compound migrated like ADP-

D -glucose;

2. The ultraviolet absorption spectrum of the compound was that Figure 3: Structure of 7(adenosine 5'-pyrophosphoryl)-D-sedoheptulose.

40 HO

H — 0 — 0 — d —0

HN HO H'

0 HO 42 of an adenine nucleotide [Xmax = 259 mp; absorbancy ratios, 280 m p/

260 mp = 0. 21 and 250 m p/260 mp = 0. 77 at pH 7 (82) ];

3. The ratio adenosine/phosphate/sugar in the compound was determined to be 1:2:1 (Table 3);

4. Mild acid hydrolysis (0. 05M trifluoracetic acid, 10 min at

100°) or treatment with a nucleotide pyrophosphatase (0.2 units Sigma nucleotide pyrophosphatase, 0.7 pmole Tris-HCl, pH 7.5, and 1.5 pmoles MgCl2 in a total volume of 30 pi, incubated 60 min at 37°) pro-

14 duced a C-labeled compound that migrated upon electrophoresis at pH 3. 7 like a sugar phosphate and a UV absorbing compound indisting uishable from AMP;

5. Further acid hydrolysis of the sugar phosphate described above (1M HC1, 15 m in, at 100°) w as without effect, w h e rea s a new radioactive compound which no longer moved upon electrophoresis at pH 3.7 resulted upon its treatment with alkaline phosphatase (0,3 units of Sigma calf intestinal mucosa phosphatase, 2 pmoles Tris-HCl, pH

8.2, 1.5 pmoles MgCl2, incubated 60 min at 37°). This suggested that the sugar was linked to a phosphate of the nucleotide at a position re sistant to acid hydrolysis and hence not through a hemiacetal bond;

6. The neutral radioactive compound was not oxidized by bro mine (2$ bromine in 0.1M barium carbonate, pH 5.4, 60 min at room temp.), but it was apparently reduced by NaBH* (1 pmole NaBH*, 3 pmole Tris-HCl, pH 8. 2, in a volume of 30 pi, 60 min at room temp.) TABLE 3

CHEMICAL ANALYSIS OF ENZYMATICALLY FORMED 7(ADENO-

SINE 5* -FYROFHOSPHORYL) -D-SEDOHEPTULOSE

Analysis Total fjmole Quantity relative to adenine

Adenine 0 .22 1.0

S e dolieptulo s e 0 .2 6 1.1

Total phosphate 0 ,48 2.1

Table 3: 7(Adenosine 5' -pyrophosphoryl) -D-sedoheptulose was prepared and isolated as described in the text. The concen tration of adenine was determined by its UV absorption at 259 m/i (82), D-sedoheptulose by the cysteine-sulfuric acid method (77), and total phosphate by the method of Ames and Dubin (78).

43 44

The reduction was indicated by the change in its chromatographic mobil ity after treatment with NaBH4. These results suggested a ketose rather than aldose structure in the sugar;

7. Upon electrophoresis in sodium borate or sodium arsenite buffers, or upon chromatography in Solvent IV, the neutral radioactive compound was indistinguishable from authentic D-sedoheptulose;

8. When mixed with authentic sedoheptulosan and treated with

0.5M HC1 for 30 min at 100°, the neutral radioactive compound yielded two radioactive compounds which separated upon electrophoresis in sodium borate or sodium arsenite or upon chromatography in Solvent

IV and which were indistinguishable from authentic sedoheptulosan and

D-sedoheptulose. They were present in the ratio 5.5:1 which is in close agreement with the ratio determined by LaForge and Hudson (83);

14 9. When the intact C-labeled nucleotide was reduced with NaBH 4 and subsequently hydrolyzed with nucleotide pyrophosphatase and alka line phosphatase as described previously, the radioactive compound produced was indistinguishable by chromatography and electrophoresis from the compound synthesized by the NaBH4 reduction of authentic D- sedoheptulose. This is considered additional evidence that D-sedoheptulose in the nucleotide has an uncombined reducing group.

Determination of Enzyme Activity

Reactions were generally carried out in thin walled glass capil 45 la rie s . A typical reaction mixture contained: 0.1 jjmole 5(adenosine

5' -pyrophosphoryl) -D-ribose, 0.5 pmole (0.04 pC) D-fructose-U14C-6- phosphate, 0. 4 mg lyophilized Chlorella 45-60$ (NH4 ) 2 S0 4 fraction, 0. 02 pmole TPP, 0. 1 pmole MgCl2 and 2 pmoles Tris-HCl buffer, pH 8.2, in a total volume of 30 pi. The mixture was incubated as described in each experiment, then applied directly to a sheet of paper moistened with ammonium formate buffer, pH 3, 7, and electrophoresed. In about

90 min at 25 v/cm 7(adenosine 5 '-pyrophosphoryl) -D-sedoheptulose was separated by about 2 cm from the other principal radioactive compon ent of the mixture, D-fructose-14C-6-phosphate, The labeled com pounds were located by exposure of the paper to x-ray film. Those areas of the paper were cut out, immersed in 0.01$ dimethyl FOFOP.

0.4$ PPO in toluene, and the papers were counted in the liquid scintil lation spectrometer.

Nature of the Reaction System

The synthesis of the product was proportional to time and to en zyme concentration. Optimal activity occurred around pH 8.2, with

50$ of that activity at pH 7 and 53$ at pH 9.

The reaction was stimulated by TPP and by MgCl2 with optimal concentrations of 3.3 x 10”4M and 3,3 x 10"2M respectively (Table 4),

The nature of the substrates, products, and cofactors suggested that this is a transketolase-like reaction (84), hence several experi- TABLE 4

EFFECT OF MG++ AND TPP CONCENTRATIONS ON THE ENZYMA

TIC PRODUCTION OF 7( A DENOSINE 5 '-PYROPHOSPHORYL) -

D-SEDOHEPTULOSE

Addition C o ncen tration (M) #14C inc into product

1. none ------12.1#

2. M gCl2 3 .3 x 10“2 13.4#

3. MgCla 3.3 x 10"3 11.2#

4. M gCl2 3.3 x 10"4 11.5#

5.TPP 3.3 x 10"3 13.0#

6. TPP 3 .3 x l O " 4 13.2#

7. TPP 3 .3 x 10 _S 12.9#

8. MgCl2 and TPP 3.3 x 10"2 and 3,3 x 10**3 19.7#

9. MgCl2 and TPP 3.3 x 10"3 and 3.3 x 10”4 18.3#

10. MgCl2 and TPP 3.3 x 10 4 and 3.3x10~5 17.0#

Table 4: The reaction, mixture contained (in addition to those components listed above ); 0.1 jimole 5(adenosine 5'-pyrophosphoryl) - D-ribose, 0.05 fjpnole (0,04 fxC) D-fructose-U-l4C-6-phosphate, 0.4 mg lyophilized Chlorella 45-60# (NH4)2S04 fraction, and 2 pmoles Tris- HCl buffer, pH 8,2, in a total of 30 /il. The mixture was incubated 15 min at 3,7 C° and assayed as described in the text.

46 47 ments were performed to test that hypothesis. It was first demonstra ted that the Chlorella enzyme preparation appears to contain an ordi nary transketolase. A mixture containing the following materials was incubated for 60 min at 37°; 3 x 10 4 pmole {0,06 pC) D-fructose-14C -

6-phosphate, 0. 2 pmole D-ribose-5-phosphate, 0. 6 mg lyophilized

Chlor ella 45 -60$ (NH^SC^ precipitate, 2.5 pmoles Tris-HCl, pH 8.2, in a total volume of 30 pi. After incubation, alkaline phosphatase (0.3 units) was added to the mixture which was incubated for an additional

45 min at 37°. Radioactive D-sedoheptulose was isolated from the mix ture by electrophoresis in 0.1M sodium arsenite, pH 9.6. About 20$ of the total recovered radioactivity was in a compound indistinguishable from D-sedoheptulose.

In another experiment, yeast transketolase (Sigma) was tested for its ability to synthesize 7(adenosine 5'-pyrophosphoryl)-D-sedoheptulose. The reaction mixture contained 6 x 10~4 pmole (0. 06 pC)

D-fructose-^C-6-phosphate, 0.06 pmole 5(adenosine 5*-pyrophosphor yl) -D-ribose, 0.15 pmole MgCl2, 0.03 pmole TPP, 1.5 pmole Tris-

HCl, pH 7. 6, and 3 pi (0,02 units) Sigma yeast transketolase. The mixture was incubated at 25° for 60 min, then inactivated in a 100° bath for 2 min. A similar reaction mixture was incubated with 0. 06

pmole of D-ribose-5-phosphate in place of 5(adenosine 5* -pyrophos phoryl)-D-ribose to measure the normal activity of the enzyme. The

mixtures were treated with nucleotide pyrophosphatase and alkaline 48 phosphatase (0.2 and 0.3 units of each for 45 min at 37°), 0,2 jjmole authentic D-sedoheptulose was added to each, and they were electrophoresed on paper in 0 .15M sodium arsenite buffer, pH 9.6. The papers, after drying, were left at room temperature overnight, after which authentic D-sedoheptulose appeared as a brown spot. This area of the paper was cut out and its radioactivity determined in the liquid scintillation counter. When 5(adenosine 51-pyrophosphoryl) - D-ribose served as ketol acceptor, the D-sedoheptulose area contained only 3.9# of the total radioactivity on the paper. In the control with D-ribose 5- phosphate, 35# of the total radioactivity was found in D-sedoheptulose.

In similar experiments, Sigma yeast transaldolase was found unable to catalyze the transfer of carbons to 5(adenosine 51-pyrophosphoryl)-D- ribose from D-fructose 6-phosphate.

Evidence was obtained by the use of differentially labeled sub strates that the carbons transferred to 5(adenosine 5' -pyrophosphoryl) -

D-ribose from D-fructose 6-phosphate arise from the carbonyl end of the molecule. That would be the mechanism predicted for a transketo lase-like reaction. Differentially labeled D-fructose 6-phosphate was not available to us, so D-glucose-1-l4C -6-phosphate and D-glucose-6-

14Q-6-phosphate were used instead, (Previous observations had indi cated that radioactive D-glucose 6-phosphate rapidly gave rise to radio active D-fructose 6-phosphate when incubated with the Chlor ella enzyme * fraction.) Details of the experiment and quantitative results are shown 49 in Table 5. The small incorporation of l4C from D-glucose-6-14C-6- phosphate is presumably due to the activities of various glycolytic en zymes in the preparation.

It could not be demonstrated that the reaction catalyzed by the

Chlorella enzyme is reversible becaue of the presence of other enzymes in the preparation. Thus, when enzymatically synthesized 7(adenosine

5t-pyrophosphoryl)-D-sedoheptulose-14C and the ketol acceptor, D -ribose 5-phosphate, were incubated with the ammonium sulfate fraction and the appropriate cofactors, radioactivity was found not only in D- fructose 6-phosphate, but in a number of other unidentified compounds.

The two compounds, D-ribose 5-phosphate and 5(adenosine 5'- pyrophosphoryl)-D-ribose, were compared as substrates in the Chlorel la enzyme system by obtaining the apparent and ^ max values for each. Conditions were chosen such that each reaction was in its lin ear stage when the formation of product was measured. The mixtures contained 0.1 pmole D-fructose-l4C-6-phosphate (0.1 pC), 0.1 pmole

MgCl2, 0.02 pmole TPP, 0. 07 mg lyophilized Chlorella 45-60$(NH 4 ) 2

S04 precipitate and 3 pmoles Tris-HCl, pH 8.2, in a volume of 30 pi.

Three of the mixtures contained 0.003, 0.005 and 0,01 0 pmole respec tively of 5{adenosine 5* -pyrophosphoryl) -D-ribose and three other m ix tures contained 0.003, 0.005 and 0. 010 pmole respectively of D-ribose

5-phosphate. The mixtures were incubated 5 min at 25° and inactivated by heating in a 100° bath for 2 min7(adenosine 5*-pyrophosphoryl)-D- TABLE 5

FORMATION OF 7 (A DEN OSIN E 5*-PYROPHOSPHORYL)-D-

SEDOHEPTULOSE FROM SPECIFICALLY 14C-LABELED

SUBSTRATES BY CHLORELLA ENZYMES

Total Radioactivity recovered radio- in 7(adenosine 5' -pyro- $ Radio activity phosphoryl) -D-sedohep- activity in- Substrate (CPM) tulose (CPM) ______corporated

1. Of-D-glucose-l - 89, 364 41, 107 46$

14C -6 -phosphate

2. a- D-glucose -6- 125,000 7,500

14C -6-phosphate

Table 5: The reaction mixture contained (including the above reactants): 0.1 jLtmole 5(adenosine 5'-pyrophosphoryl)-D-ribose, 0.2 mg lyophilized Chlorella 45-60$ (NH4)2S04 fraction, 2 fjmole T ris-H C l, 0.1 //mole MgCl2, and 0.02 fjmole TPP, in a total of 30 fi1. The reac tion was incubated 1 hr at 37° C and assayed as described in the text.

50 51 sedoheptulose-I4C was isolated by electrophoresis on. paper at pH 3. 7.

It was located by the UV absorption of standard 5(adenosine 5'-pyrophos phoryl) -D-ribose which migrated at essentially the same rate as the

14C-labeled D-sedoheptulose nucleotide. The UV absorbing area of the paper was cut out and its radioactivity determined in the liquid scintillation• counter. D-Sedoheptulose - 14 C-7-phosphate, formed when D- ribose 5-phosphate, served as substrate, was isolated as the free sugar following the procedure described above for the assay of Sigma trans k eto lase.

Measurements of enzyme activity were made in triplicate for each concentration of the two substrates. Apparent K. - and V values M m ax (Table 6) were obtained from the averages of those data by the method of Lineweaver and Burk (Figure 4). They indicate that the system with

D-ribose 5-phosphate as substrate has a slightly lower apparent but a lower V than the system with 5(adenosine 5' -pyrophosphoryl)- tn3>x D-sedoheptulose as substrate.

C. Attempted Biosynthesis of Chlorella Cell Wall Polysaccharides

Introduction

Sugar nucleotides have been shown to be donors of monosaccharide units for polysaccharide biosynthesis in numerous studies (24, 25, 26).

Usually, the transfer of the sugar moiety to the polymer was catal yzed by a particulate TABLE 6

DATA FROM LINEWEAVER-BURK PLOTS

Vma?r CPM inc into S -7 -P S u b strate App K^^(M) ■______m in______ribose 5-phosphate 1.1 ±0.4 x,10"4 900 ± 60

5(adenosine 5'-pyro- 4,8 i 1.1 x 10‘4 3333 ±300 phosphoryl-D-ribose

Table 6: Components of the reaction mixtures, estimation of product formation and methods used in determining those constants are given in the text.

52 Figure 4: Lineweaver-Burk plots with D-ribose-5-phosphate (broken-line) and 5(adenosine 51-pyrophosphoryl)-D-ribose (solid-line) as substrates in the Chlorella transketolase reaction.

53 A 00 I w .\ I 0>h \ I cn \ \

—v x icr4 cpm me

w I— x o 0 1 55 enzyme fraction extracted from the organism of interest. In Chlorella this same process was assumed to occur and was studied by incubating various l4C-labeled sugar nucleotides with particulate enzymes pre pared from Chlorella cells representative of specific stages of the growth cycle (described in Part A).

Preparation of Enzymes that were Utilized in this Study

1. Particulate enzymes

Chlorella cells harvested as described in Part A were suspended in 10 ml of cold buffer consisting of 0. 1M sodium/potassium phosphate, pH 7.0, 0. 05M jS-mercaptoethanol, and 5$ soluble polyvinylpyrrolidone.

Large aggregates of cells were disrupted by briefly homogenizing the chilled suspension in a glass tube with a Teflon pestle. The cells were broken by two passages through a cold French pressure cell (Aminco) at about 15, 000 psi. The homogenate was centrifuged in the cold at

1000 x g for 15 min. The residue was discarded and the supernatant so lution was centrifuged at 20, 000 x g for 20 min. This centrifugal pel let represents the particulate enzymes. It was suspended in a mini mum of cold 0.1M sodium/'potassium phosphate, pH 7.0, buffer con taining 0.05M 8-mercaptoethanol and utilized immediately in the re actions to be described later.

2. Trypsin treated particulate enzymes

Particulate enzymes prepared as described above were treated 56 in the following manner: 0.2 ml of particulate enzyme, 0.15 jLimole

M nCl2, 20 fig of trypsin, and 20 ptmole sodium/potassium phosphate, pH 7.0, buffer in a total volume of 0.26 ml were incubated for 15 min at 30°. This mixture at the end of incubation was utilized in the parti culate enzyme reactions. In another set of experiments this material was washed two times with buffer before use in the reaction mixtures.

3. Acetone treated cells

Chlorella cells were suspended in a minimum of cold 0. 1M sodi um/potassium phosphate buffer, pH 7.0, containing 0.5M /?-mercaptoethanol and slowly poured into 40 ml of acetone at -60° which was constantly being stirred. The mixture was contained in a glass homo- genizer tube which was immersed in a dry ice, acetone bath. It was homogenized gently with a Teflon pestle and then allowed to stand for

15 min. The cells were collected by centrifugation at 10, 000 x g in a refrigerated Sorvall centrifuge previously equilibrated at -20°. The centrifugal pellet was immediately slurried and washed two times in

20 ml of cold 0 .1M Tris-HCl, pH 7. 6, 0. 05M jS-mercaptoethanol buffer.

The final centrifugal pellet was slurried in a minimum of buffer and was utilized as a source of particulate enzymes in the reaction mix tu re s .

Assay Method for the Synthesis of Polysaccharides

Sugar nucleotides uniformly ,4C -labeled in the sugar moieties were incubated in the presence of various cofactors with Chlor ella 57

particulate enzymes. After incubation the mixtures were applied di

rectly to papers moistened with ammonium formate, pH 3.7, and elec-

trophoresed. The resulting electrophoretograms, when exposed to x-

ray films, showed the 14C-activity pattern of the reaction products. jI The 14C-material that is located at the origin (exactly at the point of

application) represents polymeric material. This I4C-material was i then cut from the paper and characterized by its solubility, chromato

graphic, and hydro ytic properties.

Experiments Conducted with Particulate Enzymes

Reactions were carried out in thin walled glass capillaries, and

a typical reaction mixture contained; 5 x 10~5 jimole UDP-D-galactose-

U14C (0.04 fiC), 5 xj 10 4 Jim ole UDP-D-xylose, 5 x 10~4 jjmole GDP-D- mannose, 5 x 10"4 pmole UDP-D-glucuronic acid, approximately 1 x I 10“5 jjmole UDP-L-rhamnose, 0.15 jumole MgCl2, approximately 0.2 mg

of Chlorella particulate enzyme, and 3 jLtmole sodium/potassium phos

phate buffer, pH 7.0, in a total volume of 30 jjl. The mixture was in

cubated for 45 min at 37° and then assayed for l4C-polysaccharide m at

erial as earlier described.

A series of reactions was carried out subsequently such that each

sugar nucleotide in the above mixtures contained the 14C-label in one

reaction mixture. Enzymes prepared from cells at various stages of the growth cycle were utilized in these reactions. In none of these ex- 58 periments was there any indication of the incorporation of l4C-label into a polymer.

When these experiments were conducted using the trypsin treated particles, negative results were again obtained.

Experiments Conducted with Acetone Treated Cells

The acetone treatment produced cells which have the appearance of whole cells under a light microscope; however, the cells have be come permeable to nucleotides. This was indicated by the ability of these preparations to catalyze the synthesis of UDP-D-glucuronic acid and UDP-D-xylose from the appropriate 14C-labeled substrates (See section A for description of reactants). These preparations also were observed to contain other common enzyme activities, e.g., hexokinase, phosphohexose isomerase, phosphatase, and various epimerases.

When the acetone treated preparations were incubated with UDP-

D-glucose-U14C, CDP-D-glucose-Ul4C, or a-D-glucose-UI4C-1-phos phate in the same manner as described for the particulate enzymes, a

14C-polymeric compound was obtained at the origin of the electrophoretograms. This l4C-material was eluted from the paper with cold water, and after evaporation to dryness in vacuo, it was chromatographed on paper in Solvent V. When the chromatogram was exposed to x-ray 14* film, the C-activity pattern showed a scries of oligmers to be pre sent. These 14C-compounds were eluted and incubated with acid, snail 59 intestinal enzyme, or salivary a-amylase and the mixtures chromato graphed in Solvent I, Only 14C-glucose was observed on the chromatogram. This evidence suggests that amylase (o£-l» 4-glycosidic link ages) synthesis is occuring in the reaction mixtures. In comparative reactions, a-D-glucose-1-phosphate was a more efficient donor of the glucose unit, than was UDP-D-glucose or GDP-D-glucose. This sug gests that these sugar nucleotides are probably being hydrolyzed to a-

D-glucose-1-phosphate and that the synthesis is catalyzed from this compound by starch phosphorylase.

Since these preparations contained such an active starch synthe sizing system, it was felt that by using sugar nucleotides, which least resemble those known to be starch precursors, one would have a bet ter chance of observing the synthesis of wall polysaccharides. UDF-

D-xylose-U14C, was incubated with the acetone cells in the previously described manner. When the reaction products were analyzed by elec trophoresis at pH 3. 7, 14C-polymeric material was observed at the origin in good yield. This material was eluted from the paper and chromatographed in Solvent V. This yielded a series of oligmers which could be hydrolyzed to glucose-14C by the action of salivary a-amylase.

Thus, again the synthesis of starch was taking place. UDP-D-Glucur- onic acid-UI4C was utilized in a similar manner and yielded identical results. These results suggest that Chlorella must contain enzymes capable of converting xylose-1-phosphate to xylose-5-phosphate. This 60 would provide entry into the pentose phosphate pathway with subsequent production, of fructose-6-phosphate, which could ultimately lead to glu cose-1-phosphate. Evidence for these transformations was suggested by the presence of 14 C-fructose and I4C-glucose 1A in the reaction pro ducts as well as by the fact that unlabeled a-D-glueose-1 -phosphate com- pletely inhibited the production of starch from UDP-xylose.

D, Studies on Chlorella pyrenoidosa Cell Wall Polysaccharides

Introduction

It has been observed that enzymes extracted from Chlorella pyrenoidosa catalyze the synthesis of UDP-L-rhamnose (31), probably

GDP-L-galactose (34), and UDP-D-glucuronic acid (85) as well as several other common nucleotides (27). The presence of such enzymes suggests that there may be polysaccharides in the wall of that organism containing one or more of the monosaccharides synthesized from those nucleotide derivatives. The possible involvement of D-glucuronic acid and L-galactose seemed particularly interesting since neither has been reported to occur in Chlorella polysaccharides. These studies were directed toward the isolation and purification of a polysaccharide con taining these constituents.

Extraction of Cell Wall Polysaccharides

A suspension of 2. 0 g of lyophilized cells in 150 ml 0. 025M N a/K phosphate buffer, pH 6.8, was passed two times through a French pres 61 sure cell at 15-20, 000 psi. Whole cells were removed from the homogenate by centrifugation at 160 x g for 20 mins. Pigments and other non-polar components were removed from the material by solvent ex traction. Starch was removed by treatment with Q'-amylase and subse quent removal of the degraded glucan by dialysis. Protein was sim i larly removed by treatment with pronase. The material remaining after enzymic hydrolysis was fractionated by extraction with 4M KOH containing 0. 03M NaBH 4 under a stream of nitrogen gas at 25° for 2 hrs.

The alkali-soluble m aterial was brought to pH 8. 3 with acetic acid and dialyzed against several 1 1 changes of buffer. This material was stored in solution at 4°. The yield of alkali-soluble polysaccharide from 2 gm dry weight of cells was 46 mg (2. 3$), (determined by lyo- philization of an aliquot). Details of the extraction and fractionation procedures are given in Fig. 5.

Analysis of the Polysaccharide Fractions

Samples (5 mg) of each of the lyophilized fractions (alkali-soluble and insoluble) were suspended in 5 ml volumes of 2M trifluoroacetic acid, and hydrolysis was carried out in an autoclave for 45 min at 120°.

The hydrolysates were transferred to 50 ml round bottom flasks and evaporated to dryness on a rotary evaporator. A portion of each hydrolysate was subjected to electrophoresis in ammonium formate, pH

3.7 and paper chromatography in Solvent I. Figure 5: Details of the procedure used to extract polysacchar ide from the cell wall of Chlor ella pyrenoidosa.

62 Chlorella cells (2 g dry wt) Suspended In 150 rnl of 0. 025M N a/K Phosphate, pH 7,0. Passed through a French cell at 15, 000 psi. Centri fuged at 160 x g for 20 min.

Suncrn&tant solution Residue

Centrifuged at 20, 000 x g 20 min.

Supernatant Residue solution | Treated with 500 ml acetone 6 hr s. at room temp, j Centrifuged at 20, 000 x g for 20 min.

Supernatant Residue solution T reated with 500 ml m ethanol/chloroform 1:1, 15 h rs. at room temp. Centrifuged at 20,000 x g for 20 mins.

Supernatant Residue solution I Suspended in 4 vols. ethanol

Supernatant Insoluble Material solution (lyophilized, yield 200 mg) Suspended ir. 20 ml of 0. 1M K a/it phosphate, pH 7.0, 1 ml of human salivary amylase added, and the mix ture was incubated overnight at 3tf. Added 4 vol. of 95$ ethanol.

Supernatant Precipitate solution i Suspended in 20 m l of 0. 025M T ris.H C l, pH 7.3, containing 15 mg of Pronase, Incubated overnight, at room tump. Added 4 vol. of 95$ ethanol.

Supernatant Precipitate solution (lyophilized, yield 100 mg -54) Treated with 20 ml of -iM KOH/0. 03M NaBH* at room temp, for 2 hrs. while stirring with a stream of Nj gas. Repeated extraction once. Centrifuged at 20, 000 x g for 20 minB, I Alkali-soluble Residue (14 mg - 0,7$0) polysaccharide (46 mg - 2. 3$

Brought pH to 8. 3 with cold 50$ acetic acid. Dialyzed solution against three 1. vols. 0.05M Tris.HCl, pH 8.3, for 24 hrs. at 4*.

Final solution of polysaccharides in 0.05M Tris.HCl, pH 8.3. 64

The hydrolysate of the alkali insoluble material showed the pre

sence of glucose and glucosamine while the alkali-soluble hydrolysate

contained rhamnose, arabinose, xylose, mannose, galactose, glucose,

glucuronic acid, and two unidentified compounds.

The alkali-soluble material contained the monosaccharides of interest and was subjected to further analysis.

Purification of an Acidic Polysaccharide

The alkali-soluble polysaccharides from 2 g of lyophilized cells were applied to the DEAE-Sephadex column, and the column was washed with 6 bed volumes (240 ml) of 0, 05M Tris HC1 buffer, pH 8.3. Poly saccharides were eluted with a 400 ml linear NaCl gradient (0 to 1M) at a flow rate of 25 m l/hr. Five ml fractions were collected, and an aliquot of each was assayed for carbohydrate by the phenol-sulfuric ac id method (86). The elution profile is shown in Fig. 6. Fractions 15 to 23 were pooled and dialyzed against 1 1 of 0. 05M Tris-HCl buffer, pH 8. 3, for 24 hrs at 4°. The dialyzed solution was reduced to 10 ml in a rotary evaporator at 35° and was stored in the refrigerator.

A 2 ml sample of the material obtained from the DEAE-Sephadex column was made 2M with NaCl and applied under the layer of buffer, through a long hypodermic needle, to the top of a column of Bio-Gel

P-300. The column was eluted with 300 ml of 0.05M Tris-HCl/lM

NaCl buffer, pH 8.3. Fractions of 3 ml were collected, and an aliquot Figure 6: Elution pattern from DEAE-Sephadex of the alkali - soluble polysaccharides of Chlorella pyrenoidosa cell wall.

65 1.0

FRACTION NUMBER

O' O' 67

of each was assayed for carbohydrate by the phenol-sulfuric acid m eth

od (86). The fractionation effected is shown in Fig. 7. Material in the

first peak was collected and dialyzed against 1 1 of 0.05M Tris-HCl

buffer, pH 8.3, for 24 hrs at 4°. The volume of the dialysate was re

duced to 3 ml at 35°, and it was stored in the refrigerator.

The dialyzed Bio-Gel eluate was made 2M in NaCl and applied to

a column of Sepharose 6B. Polysaccharide was eluted from the column with 200 ml of Tris-HC1/1M NaCl buffer, pH 8.3, at a flow rate of

36 m l/hr. The elution pattern is shown in Fig. 8. The acidic poly

saccharide appeared as a single symmetrical peak. Fractions con

taining the compound were pooled and dialyzed against several 1 1 vol

umes of distilled water for 36 hrs at 4°. This solution was stored at

-15° and is hereafter referred to as the purified polysaccharide.

Analysis of the Neutral Polysaccharides

The m aterial which passed through the DEAE-Sephadex column

represents the neutral alkali-soluble polysaccharides. A 2 mg sample

of this lyophilized polysaccharide was suspended in 2 ml of 2N trifluor-

oacetic acid and hydrolyzed for 1 hr at 100°. The hydrolysate was trans

ferred to a 50 ml round bottom flask and evaporated to dryness on a ro

tary evaporator at 35°. The concentration of acid does not rise during

the evaporation since the boiling point of trifluoroacetic (69°) is lower

than that of water. This minimizes the destruction of reducing sugars Figure 7: Fractionation of the partially purified polysaccharide on a column of Bio-Gel P-300. The procedure is described in the text.

68 X OPTICAL A8S0RBANCY AT 485 nm P P P P r- ro cd w O O

O FATO NUM3E FRACTION:

cjn O Figure 8: Chromatography of the purified acidic polysaccharide on a column of Sepharose 6B. Exclusion limit for a polysaccharide on this column is 1 x 106 daltons. Blue dextran (2 m g/m l; av. mol. wt. 2 x 106) and sucrose (2 mg/ml) were added as reference compounds. Details of the procedure are given in the text.

70 .ACIDIC POLYSACCHARIDE

BLUE DEXTRAN

REACTION NUMBER 72 by the action of strong acid. i-Inositol (0.2 mg) was added as an inter nal standard, and the mixture was reduced and acetylated as described by Albersheim (68).

Upon gas-liquid chromatography, the hydrolysate yielded eight clearly separated peaks. Monosaccharides represented by the peaks were identified by adding known quantities of authentic sugars to the hydrolysis mixture and noting which peaks had increased in size.

The quantity of each monosaccharide present uas determined by comparing the area under its peak to that of the internal standard, i- inositol. An adjustment was made for the difference in yield of acetyl ated sugar alcohol from each monosaccharide (68).

The results showed that the hydrolysate contained rhamnose, arabinose, xylose, mannose, galactose, glucose, and an unidentified sugar in the molar ratios 10.3:2,1:1. 7:1. 0:3. 2:1. 0:0. 5.

The glucose present in this mixture could also be released by hydrolysis with salivary a-amylase. This indicates that it is present as a component of starch. The polysaccharides in this mixture were not subjected to further analysis since they only represented approxi mately 25 of the alkali-soluble material.

Properties of the Purified Acidic Polysaccharide

1. . Ultracentrifugal sedimentation properties

A solution containing 0.75 m g/m l of the purified polysaccharide was made 0. 2M in NaCl and centrifuged at 40, 000 RPM in a Beckman

Model E analytical ultracentrifuge. The polysaccharide sedimented as a single symmetrical peak (Fig. 9), and its sedimentation coefficient was calculated to be 2. 075 (87). The molecular weight of the com pound was estimated by the sedimentation equilibrium procedure of

Yphantis (88). This was carried out as follows: A solution containing

0.5 m g/m l of the purified polysaccharide was made 0. 1M in NaCl and

0. 03M in NaH2P04 and centrifuged at 14, 000 RPM. Equilibrium was attained after E4 hrs. The run was continued for a total of 48 hrs.

Data from a photographic plate of the Rayleigh interference pattern, plotted as In of the fringe displacements (f) against the radial distance squared (r2) in centimers (Fig. 10) gives a straight line with a slope of

1,2974, (by Chi-squares method). This slope and a value for the par tial specific volume (0.65), obtained experimentally by Northcote (43) for a similar polysaccharide, were used to calculate an approximate molecular weight of 88, 000.

2. Monosaccharide substituents

Lyophilized purified polysaccharide (2 mg) was dissolved in 2 ml of 2M trifluoroacetic acid and hydrolyzed for 1 hr at 100°. The hydrolysate was then analyzed by gas-liquid chromatography as previously described for the neutral polysaccharides.

Although a volatile derivative of glucuronic acid could be pro duced by reduction, methylation of the carboxylic group and acetylation, Figure 9: Schlieren patterns of the purified polysaccharide sedimented centrifugally as described in the text.

Figure 10: A plot of data obtained from the centrifugal sedimen tation equilibrium of the purified polysaccharide. A straight line is evidence for a homogeneous mixture of polymers (88).

76 In FRINGE DISPLACEMENT

ci Q> ro ot

2g > r s £ § srn g w t>. C P 3)> * oK otn o 3

in ro ro

-d -J the yield of the derivative was not reproducible enough to give reliable quantitative data. Therefore, the proportion of glucuronic acid in the polysaccharide was estimated by the method to follow: A 3 mg sample of the lyophilized polysaccharide was hydrolyzed in 2M trifluoroacetic acid at 121° for 2 hrs. Trifluoroacetic acid was removed in vacuo, and the residue was taken up in a small volume of water and electrophor es - ed at pH 3. 7 at about 80 v/cm for 20 mins. The bulk of the material was applied as a band on the paper, and a small portion was applied as a spot at a slight distance from the band. After electrophoresis the area of the paper that had been spotted was cut out and sprayed with the

AgN03 reagent. Material in two areas of the electrophoretogram re acted with the AgN03 reagent. One area was at the origin (neutral sugars) and the other was indistinguishable in mobility from a glucur onic acid standard. The neutral sugar and uronic acid areas of the paper were cut out and eluted with water. The quantity of glucuronic acid present in the eluate was estimated by the method of Bitter and

Muir (89). A small amount of radioactive D-galactose was added as a marker, and the mixture of neutral sugars was chromatographed in

Solvent I for 15 hrs. Galactose, which is well separated from the other monosaccharides under those conditions, was located by its radioac tivity, eluted from the paper with water, and its quantity estimated by the method of Park and Johnson (90). In each of the two methods an eluate of a blank portion of the paper was subjected to the same colori- 79 metric procedure, and the results were used to correct for reactive materials in the paper, buffer, and solvent system. The molar ratio of galactose to glucuronic acid was estimated to be 1:0.66, From this ratio and the data obtained by gas-liquid chromatography, the molar ratios and percentage yields of the monosaccharides in the polysac charide were estimated and are given in Table 7.

3. The configuration of galactose and rhamnose

Galactose and rhamnose from the polysaccharide hydrolysate were isolated by chromatography in Solvent I. The purity of each was established by gas-liquid chromatography.

L-Rhamnose isomerase (74) was used to determine the config uration of the isolated rhamnose. The reaction mixture contained 20 jLll of enzyme (ca. 0.4 mg protein), 0.1 pmole neutral glutathione, 0.1 jumole MnCl2, 0.2 j/mole of the isolated rhamnose and 0.25 jjmole T ris-

HC1 buffer, pH 7. 5, in a total volume of 30 jil. Two other reaction mix tures were prepared as controls, one with 0.2 jimole authentic D-rhamnose and the other with 0.2 fjpcnole authentic L-rhamnose. The mixtures were incubated for 30 mins at 37° then applied to a paper moistened with 0. 05M sodium borate buffer, pH 9» and electrophoresed in that buffer. Under those conditions the authentic L-rhamnose and rhamnose from the polysaccharide hydrolysate were converted almost entirely into a compound indistinguishable from authentic L-rhamnulose. D-

Rhamnose was unaffected by the enzyme. TABLE 7

ANALYSIS OF MONOSACCHARIDES IN THE

PURIFIED POLYSACCHARIDES

$ Wt or original Monosaccharide present polysaccharide* Molar ratio rhamnose 47.0 11.5 arabinose 10.3 2.7 xylos e 8.2 2.2 mannose 4.5 1.0 galactose 12.8 2.9 glucuronic ac.id 8.5 1.9

*These calculations do not include the increase in weight due to addition of water during hydrolysis. The com plete structure would have to be known before this could be accurately calculated.

80 81

The proportion of galactose in the polysaccharide present as the

D-enantimorph was determined by the use of D-galactose dehydrogen ase (80). NADH produced in the following reaction mixture was found to increase linearly with increasing concentrations of D-galactose;

3 jjl Sigma D-galactose dehydrogenase, 0.5 jjmole NAD, 4 jimoles Tris-

HC1 buffer, pH 8.6, and 0,005 to 0.015 pnole D-galactose in a total volume of 0. 3 ml. The mixtures were incubated for 30 mins at 37°.

The absorption of NADH at 340 nm was used as a measure of the oxi dation of the sugar. In a reaction mixture which contained both au thentic D-galactose and galactose isolated from the polysaccharide, the reduction of NAD was not inhibited and the results were additive. By this method the galactose sample from the polysaccharide was esti mated to contain 126 j^g^/ml of D-galactose. The quantity of total re ducing sugar in the sample was estimated as 173 jjg/'ml by the method of Park and Johnson (90), Thus 73$ of the galactose in the polysac charide is present as the D-enantiomorph. Since the sample contained only galactose according to the evidence of gas-liquid chromatography, it appeared likely that the remaining 27$ of reducing sugar represented the L-enantiomorph.

In order to obtain positive evidence for the presence of L-galac tose, the sample was incubated with L-fucose dehydrogenase partially purified from pork liver. That enzyme has been shown also to oxidize

L-galactose (81). The reaction mixture contained: 1 ml of L-fucose 82 dehydrogenase (ca, 0.1 mg protein eluted from DEAE-cellulose), 2,5 fixnoles NAD, 50 ^moles Tris-HCl buffer, pH 8. 3, and 1 fjxnole of gal actose, in a total volume of 1. 1 ml. Reaction mixtures were prepared using authentic L-galactose, authentic D-galactose and galactose iso- latedjfrom the polysaccharide hydrolysate. The reactions were car ried out at room temperature in 1 ml quartz cuvettes. The optical ab sorption of NADH at 340 nm was followed continuously with a Guilford

Model 240 recording spectrophotometer. The results shown in Fig, 11 indicate that L-galactose is present in the algal polysaccharide. Un fortunately this assay could not be made quantitative because of colored impurities in the enzyme preparation and the small amount of mono saccharide available.

4. Isolation of an aldobiuronic acid

Lyophilized polysaccharide from the DEAE-Sephadex step (Fig, 6)

(50 mg) was dissolved in 12 ml of 0.75M trifluoroacetic acid, sealed in an ampoule and heated for 4 hrs at 100°. A small amount of precipitate that formed during heating was removed by centrifugation. The super natant solution was evaporated to dryness in vacuo, and the residue was taken up in water and evaporated to dryness again. This procedure was repeated three times. The residue was then dissolved in 5 ml of water and applied to a 1,5 x 37 cm column of the anion exchange resin AG-1 -

X8 formate. The column was washed with 1 1 of water, and acidic components were then eluted with a 400 ml linear gradient of formic Figure 11: Tracing of the recorder chart obtained upon incubat ing various sugars with L-fucose dehydrogenase. At the points indi cated by the arrows substrate was added to the reaction mixtures. The distance between each point on the curves is 40 seconds.

83 OPTICAL ABSORBANCY AT 340 nm oooooooo eooo&ooooooooooo°D—GALACTOSE oooooooooooooo°°D oeooooooo&oooooooo oooooooooooooo oo 000003000000000000000000 I ^ ° 0 0 0 0 0 OOOOOOOOOOOOOOO0 OOO O o 0o° o 0 ° 0 0 0 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 3 0 I E TIM . 4 0° o° 0°° 4 O cooopooo°o°ooooooooooo oBLANK {H^O) oBLANK cooopooo°o°ooooooooooo O 0 °* o 0 ° o«,o° L-GALACTOSE o«,o° ° 0P° „°°° • © o *>0 * „ ■ 0 POLYSACCHARIDE o0 o° ° Oo oAATS FROM ooGALACT0SE CHLORELLA

84 85 acid (0 to 2M). Fractions of 3 ml were collected at a flow rate of 40 m l/hr and assayed for carbohydrate by the phenol-sulfuric acid meth od (86), Three clearly defined peaks,' labeled A, B, and C, were ob served (Fig. 12). The material in each peak was collected and sub jected to paper electrophoresis and chromatography. The results are shown in Table 8. The rates of migration of the materials in fractions

B1 and B2 suggested that they were aldobiuronic acids. Upon reduct ion of their free aldehyde groups with NaBH4, methylation of the car boxyl groups with methanolic HC1 (5$ HC1 in methanol, 16 hrs at room temperature), hydrolysis (2M trifluoroacetic acid, 2 hrs, 100°), re duction of the monosaccharides with NaBH 4 and acetylation of the re sultant sugar alcohols, the fractions were analyzed by gas-liquid chro matography. Compound B1 yielded peaks representing approximately equimolar quantities of rhamnose, glucuronic acid and mannose with a trace of xylose. Compound B2 yielded only glucuronic acid and rham nose in almost equimolar quantities. These results are interpreted as indicating that compound B2 is an aldobiuronic acid of the structure glucuronosyl-Li-rhamnose and that B1 probably consists of a mixture of two glucuronosyl disaccharides, glucuronosyl rhamnose (with a linkage different from B2) and glucuronosyl mannose. The xylose is thought to be a contaminant derived from the filter paper. In each case glucuron ic acid presumably represents the non-reducing member of the disac charide pair since the glucuronosyl-hemiacetal bonds in a polysacchar- Figure 12: Separation of the acidic components from a hydroly sate of the purified polysaccharide. The material was applied to a col umn of anion exchange resin and eluted as described in the text.

8 6 OPTICAL ABSOROANCY AT 485 nm 0.2 - 0 .4 0.6 20 040 4 30 RCIN NUMBER FRACTION 0 5 L 60 l . 70

(.W) NOIivyJ.W30N00 IOPN -4 00 TABLE 8

CHROMATOGRAPHY AND ELECTROPHORESIS OF THE URONIC

ACID-CONTAINING PRODUCTS OF PARTIAL ACID HYDROLYSIS

Electrophoresis . Chromatography F ra c tio n a t pH 3. 7_____ in Solvent I I

* A 0.05 0 .2

B1 0.80 0.25

B2 0.87 0.80

C 1 .0 1 .0

Table 8: Migration of the compounds are expressed re

lative to the migration of glucuronic acid (Rga) •

8 8 89 ide are usually the most resistant to acid hydrolysis (91).

When fraction C was similarly treated, the results of gas-liquid chromatography indicated that it was free glucuronic acid. Fraction

A, probably a large oligosaccharide, \vas present in quantities too small to permit identification of its components. DISCUSSION

Chlorella pyrenoidosa , unlike most other unicellular organisms, does not reproduce by binary fission (3). Each spherical cell increases in volume as it grows, exhibiting no other gross morphological change.

Then, over a period of a few hours, the cell forms within itself a num ber of new cells known as nautospores, " the wall of the mother cell ruptures, and the newly formed cells are released into the medium to begin the cycle again. The number of autosporeB produced is chara cteristic of the cell variety and of the growth conditions; there may be as few as two and as many as thirty-two.

The onset of autospore formation in synchronous cultures of

Chlorella has been studied by electron microscopy (44), The results of these observations on thin sections of cells suggested that the syn thesis of new cell walls takes place within the mother cell at a specific stage of the growth cycle. In the results section of this study evidence is presented that UDP-D-glucose dehydrogenase activity, which catal yzes the production of UDP-D-glucuronic acid, increases nineteen fold at a period late in the growth cycle corresponding to the period of auto spore wall formation. This increase in activity is probably due to de^

90 91 novo enzyme synthes*" rather than by the effect .of an activator. Evi dence for this was shown by the fact that this increase could not be in hibited by extracts of the earlier growth stages, nor could the enzyme preparations from the earlier stages be stimulated by inactivated ex tracts from the later stages of the growth cycle. It is well established that sugar nucleotr.des are the precursors in polysaccharide biosynthe sis. Thus, this increased enzymatic activity is interpreted as indi cating an increase in the quantity of UDP-D-glucuronic acid for utili zation in autospore cell wall formation. The UDP-D-glucuronic acid is also the starting m aterial for the production of UDP-D-xylose and

UDP-L-arabinose which probably serve as donors of the pentoses located in the cell wall.

It is possible that this increase in enzymatic activity and pre sumably the production of UDP-D-glucuronic acid indicates the onset of some other metabolic event, such as, the glucuronosylation of an un known lipid or steroid. No such compounds have been reported in

Chlorella, however, and experiments conducted in these studies did not suggest the presence of such a transfer. The only metabolic route observed for UDP-D-glucuronic acid was its decarboxylation to UDP-

D -xylose.

The nucleotide pyrophosphorylase activities measured in these studies were shown to have constant specific activities throughout the growth cycle. This, however, does not mean that the resultant sugar 92 nucleotides are produced at the same rate throughout the cycle. For in stance, their production could be stimulated by the increase of some metabolite which is produced at an enhanced rate at the stage of autospore wall formation. Chlorella ADF-D-glucose pyrophosphorylase has been purified by^Preiss et al (92) and was shown to be allosteri- cally regulated by glycolytic intermediates and inorganic phosphate.

Although no evidence has been obtained, it also is possible that the other nucleotide pyrophosphorylases are under allosteric control by metabolites unique to the autospore formation stage.

The UDP-L-rhamnose synthetase system that catalyzes the re duction and epimerization of UDP-D-glucose to UDP-L-rhamnose was shown to have a constant specific activity throughout the growth cycle.

Since L-rhamnose is a major constituent of the cell wall polysacchari des, it is felt that UDP-L-rhamnose, the proposed sugar nucleotide donor, should show a significant increase at the stage of autospore wall formation. Thus, the production of UDP-L-rhamnose must be stimu lated by some means other than de novo enzyme synthesis. The system could be allosteric ally controlled as previously described or, in this case, the availability of a cofactor could be the stimulating factor .

Bassham e£ al (18) have recently reported that 6-phosphogluconic acid production increases ten-fold at a stage which would approximately correspond to autospore formation. This in turn means that there may be a twenty-fold increase in NADPH at the stage just prior to autospore cell wall formation if the subsequent enzyme in the pentose phosphate pathway also has an enhanced activity. This increased availability of

NADPH, which is a required cofactor for the UDP-L-rhamnose synthe-^ tase system, could be the factor which controls the production of UDP-

L-rhamnose and thus would stimulate the production of quantities nece ssary for autospore cell wall synthesis.

The inability to achieve in vitro synthesis of Chlorella cell wall polysaccharides by the methods established for other polysaccharides probably does not indicate that Chlor ella utilize a different mechanism of synthesis. In most of the reported in vitro polysaccharide synthe ses only very low enzyme activities were obtained and in Chlor ella not even this amount of activity could be demonstrated. Polysaccharide synthetases are thought to be contained within the cytoplasmic m em brane and thus the condition of the membrane particles may play an important role in the synthetase activity. Chlorella has an extremely durable cell wall and rather extreme conditions must be employed in order to break the cells. It is possib le that these extraction proce dures in some manner damage the membrane particulate system, rendering it incapable of transferring the sugar moiety from the nucleo tide carrier. This same problem was encountered in the study of yeast cell wall synthesis and active synthetase preparations were not obtained until techniques of preparing spheroplasts were developed which en abled almost intact membranes to be isolated (93). 94

Cabib and Farkas (94) recently reported that chitin synthetase from several strains of yeast was present in an inactive state within the cytoplasmic membrane and could be activated by incubating the membrane preparations with the proteolytic enzyme trypsin or an acti vity factor prepared from the yeast cells. They proposed that the syn thetase remains inactive for most of the cell cycle except for a short period during which an activating factor is released resulting in rapid chitin synthesis for septum formation. It is possible that this same process occurs in the synthesis of autospore cell wall polysaccharides.

Experiments conducted with trypsin treated Chlor ella particulate enzy mes, however, did not yield any positive results.

A transketolase-like enzyme that catalyzes the formation of a com pound characterized as 7(adenosine 5*-pyrophosphoryl)-D-sedoheptulose was extracted from autotrophically cultured cells of Chlorella pyrenoidosa. The system required the substrates1 5(adenosine 5'-pyrophosphoryl) -D-ribose and D-fructose-6-phosphate and was stimulated by

TPP and MgCl2. Apparent K and V values were determined for 5-. hiex ^.denosine 51-pyrophosphoryl)-D-ribose and D-ribose 5-phosphate as ketol acceptors. Interpretation Of these values is difficult since they were obtained with a crude enzyme extract. For whatever reason; though,

5(adenosine 5 '-pyrophosphoryl)-D-ribose is converted to the D -sedoheptulose derivative more rapidly than D-ribose 5 phosphate. This is in contrast to the substrate specificity of yeast transketolase 95 which much more rapidly catalyzes the conversion of D-ribose-5-phos - phate to D-sedoheptulose-7-phosphate. Thus, if the reaction in Chlor - ella is brought about by a transketolase, the characteristics of that en zyme differ considerably from those of the yeast enzyme. Whether there is another m ore conventional transketolase in that alga cannot be determined without considerably m ore purification of the enzymes in volved. It is of interest that Villafranca and Axelrod (21) recently pur ified a transketolase from spinach leaves that uses a variety of ketol donors and acceptors including a number of non-phosphorylated sugars.

The enzymes involved in the synthesis of 7(adenosine 5 '-pyrophos phoryl)-D-sedoheptulose seem to be equally active at all stages of the life cycle. This perhaps indicates that this transketolase-like enzyme is involved in some general metabolic process which goes on at a more or less uniform rate throughout the life of the cell. No suggestion for the role of 7(adenosine 5'-pyrophosphoryl)-D-sedoheptulose in the m e tabolism of Chlor ella can be made. Indeed it may be simply an artifact of enzyme action in vitro. However, it iB indisputable that Chlor ella pyrenoidosa posseses the enzymes and substrates required to synthe size that compound, and if it does not, the cell must in some way pre vent the process from occurring.

In this work the purification of a heteropolysaccharide from the cell wall of Chlor ella pyrenoidosa is described. It appears to be a relatively homogeneous preparation since on ultracentrifugation it gave 96 a single sym m etrical peak and on equilibrium sedimentation a straight line relationship of In fringe displacement against radial distance squared was obtained (Fig. 10 ) (88).

It is felt that the most notable features of this polysaccharide are the presence of glucuronic acid and both D- and L-galactose. Neither

L-galactose nor glucuronic acid has heretofore been isolated from

Chlor ella, and, in fact, it has been especially noted that no uronic acid or L-galactose could be isolated from one strain of this species (Z, 43).

It also seems unusual to have found such a high proportion of L-rham - nose (almost 50$) in the polysaccharide. It is possible that environ mental conditions as well as constitutive factors in the strain may cause considerable variation in cell wall composition. In regard to L-galac tose, it is of interest that an enzyme system that brings about its syn thesis via the GDP-D-mannose derivative has recently been extracted from the strain of Chlor ella pyrenoidosa used in this investigation (34),

The molecular weight determined for the compound by ultracen trifugal analysis (88, 000) is considerably lower than would be expected from the rate of its movement through the various molecular sieves.

This may be attributable to the shape of the molecule. If it were highly branched, it should be more difficult for it than for a linear compound, such as a polypeptide, to pass through the pores of a filtration gel.

Hence it would be eluted from the column more quickly than the linear com pound. 97 Northcote et al (2) have concluded that the Chlor ella wall is com posed of two phases, one an organized microfibrillar structure and the other a matrix in which it is embedded. The matrix could be removed and microfibrillar material isolated by treatment of wall preparations with dilute alkali (2). Our results indicate that approximately 75$ of the alkali-soluble polysaccharide (the matrix) in this strain is com posed of acidic polysaccharides and that a large proportion of that material is made up of the purified polysaccharide (ca. 60$). It was also observed that both the non-acidic and acidic polysaccharides of the alkali-soluble fraction contain very similar proportions of neutral su gars. Thus, it is possible that the acidic polysaccharide represents a repeating unit of the cell wall m atrix that was specifically removed from a continuous network during the isolation procedure.

The cell wall m aterial that was insoluble in alkali yielded only glucose and glucosamine on acid hydrolysis. This suggests that it is unrelated to the soluble polysaccharides and is perhaps derived from a mixture of cellulose and chitin-like substance. These polysacchar ides probably make up the microfibrillar structure of this organism and account for the unusual durability of its cell wall. BIBLIOGRAPHY

Preston, R .D ., in 11 Biosynthetic Pathways in Higher Plants" (Pridham, J.B. and Swain, T., eds,), Academic Press, New Y ork, N.Y., 1965, p. 123.

Northcote, D.H., Goulding, K.J., and Horne, R.W., B iochem . J . ( 70, 391 (1958).

Kuhl, A. and Lorenzen, H. in "Methods of Cell Physiology" (Prescott, D., ed.), V. I, Academic Press, New York, 1964, p. 159.

Tamiya, H ., Ann. Rev. Plant Physiol., 17, 1 (1966).

Hoogenhout, H ., Phycologia, 2, 135 (1963),

Tamiya, H., Iwamura, T ., Shibata, K., Hase, E ., and Nehei, T., Biochim. Biophys. Acta, 1_Z, 23 (1953).

Tamiya, H ., Morimura, Y., Yokota, M ,, Kunieda, R., Plant Cell Physiol. (Tokyo), 2j 383 (1961).

Pirson, A,, and Lorenzen, H ., Z, Botan., 46, 53 (1958).

Schm idt, R.R., in "The Cell Cycle" (Padilla, G.M,, W hitson, G.L., and Cameron, I.L ., eds,), Academic Press, Inc., New Y ork, N.Y., 1969, p. 159.

Sitz, T.O., K ent, A.B., Hopkins, H, A., and Schmidt, R.R., Science, 168, 1231 (1970).

Tamiya, H. and Morimura, Y., in "Synchrony in Cell Division and Growth" (Zeuthen, E ., ed.), Inter science Publishers, New Y ork, N.Y., 1964, p. 565,

Pirson, A. and Lorenzen, H., Ann. Rev. Plant Physiol., 17, 439 (1966). 99

13. Lorenzen, H., in'Synchrony in Cell Division and Growth" (Zeuthen, E . , ed.), Interscience Publishers, New York, N.Y., 1964, p. 571.

14. Berger, C., Flora A, 157, Ell (1966).

15. Knutsen, C., Biochim. Biophys. Acta, 103, 495 (1965).

16. White, R.C., (unpublished observation).

17. Knutsen, G .f Biochim. Biophys. Acta, 161, 205 (1968).

18. Kanazawa, T., Kanazawa, K ., Kirk, M.R., and Bass ham,J.A., Plant and Cell Physiol., U , 149 (1970).

19. Horecker, B.L., Smyrniotis, P.Z., and Klenow, H ., J. Biol. Chem., 205, 661 (1963),

20. Horecker, B.L., Compr. Biochem., 15, 48 (1964).

21. Villafranca, J.J. and Axelrod, B., J. Biol. Chem., 246, 3126 (1971).

22. White, R.C, and Barber, G.A., Biochim. Biophys,, in press,

23. Caputto, R., Leloir, L. F ., Cardini, C.E., Paladini, A.C., J. Biol. Chem., 184, 333 (1950).

24. Leloir, L.F., Science, 172, 1299 (1971).

25. Hassid, W.Z., in "The Carbohydrates" (Pigman, W, and Horton D., eds.), Vol. II A, Academic Press, New York, N.Y., 1970, p. 301.

26. Hassid, W.Z., Science, 165, 137 (1969).

27. Kauss, H. and Kandler, D ., Z. Naturforsch., 176, 858 (1962).

28. Preiss, J. and Greenberg, E ., Arch. Biochem. Biophys., 118, 702 (1967).

29. Sanwal, G.G. and Preiss, J., Phytochem., 8, 707 (1969).

t 30. Kempf, E. and Mandel, P., Compt. Rend., 253, 2155 (1961).

31. Barber, G.A. and Chang, M.T.Y., Arch. Biochem. Biophys,, 118, 659 (1967). 1 0 0

32. Melo, A., Elliott, W., and Glaser, L ., J. Biol. Chem., 243, 1467 (1968).

33. Gabriel, O ., and Lindquist, L., J. Biol. Chem., 243, 1479 (1968).

34. Barber, G.A., Arch. Biochem, Biophys., 147, 619 (1971).

35. Goudsmit, E.M. and Neufeld, E.F., Biochim. Biophys. Acta, 121, 192 (1966).

36. Strominger, J.L. and Maps on, L.W., Biochem. J., 66, 567 (1957).

37. Smith, E.E., M ills, G.T., Bernheimer, H.P., Austrian, R ., B iochim . Biophys. A cta, 28, 211 (1958)..

38. Neufeld, E. F ., Feingold, D.S., and Hassid, W.Z., J . Biol. Chem., 235, 910 (1960).

39. Neufeld, E. F ., Ginsburg, V., Putnam, E.W., Fanshier, D,, and Hassid, W.Z., Arch. Biochem. Biophys., 69, 602 (1957),

40. Rogers, H. J. and Perkins, H.R., "Cell Walls and Membranes, " E. and F.N. SponLtd,, London, (1968), p. 114.

41. Soeder, C.J., Arch. Mikrobiol,, 47, 311 (1964).

42. Shihira, I., and Krauss, R.W., "Chlorella: Physiology and taxonomy of forty-one isolates," Port City Press, Baltimore, 1964.

43. Olaitan, S. A. and Northcote, D. H ., Biochem. J., 82, 509 (1962).

44. Mihara, S., Plant and Cell Physiol., 2, 25 (1961).

45. White, R.C. and Barber, G.A., Biochim. Biophys. Acta, in p r e s s .

46. Punnett, T. and Derrenbacker, E.C., J. Gen, Microbiol,, 44, 105 (1966).

47. Osborn, M.J., Ann. Rev. Biochem., 38, 501 (1969). 101

48. Rogers, H.J, and .Perkins, H.R., "Cell Walls and Membranes," E . and F.N. Spon Ltd., London, 1968, p. 161.

49. Barber, G.A. E lbein, A.D., Hassid, W. Z ., J. Biol. Chem., 239, 4056 (1964).

50. Barber, C.A. and Hassid, W.Z., Biochim. Biophys. Acta, 86, 397 (1964).

51. Brummond, D.O. and Gibbons, A.P., Biochem. Z,, 342, 308 (1965).

52. Elbein, A.D., J. Biol. Chem., 244, 1608 (1969).

53. Villemez, C.L., Sw anson, A.L., and Hassid, W.Z., A rch. Biochem. Biophys., 116, 446 (1966).

54. McNab, J.M,, Villemez, C.L., and Albersheim, P ., Biochem. J., 106, 355 (1968).

55. Whistler, R.L, and Smart, C.L., "Polysaccharide Chemistry," Academic Press, New York, N.Y., 1953, p. 135.

56. Bailey, R.W. and Hassid, W.Z., Froc. Nat. Acad. Sci., U.S., 56, 1586 (1966).

57. Kauss, H ., Biochim. Biophys. Acta, 148, 572 (1967).

58. Percival, E. and McDowell, R.H., "Chemistry and Enzymology of Algal Polysaccharides, " Academic Press, New York, N. Y ., 1967.

59. Lin, T.Y. and Hassid, W.Z., J. Biol. Chem., 241, 3283 (1966).

60. Kauss, H ., FEBS Letters, j5, 81 (1969).

61. Villemez, C.L. and Clark, A.F., Biochem. Biophys. Res. Com- m ., 36, 57 (1969).

62. Villemez, C.L., Biochem, Biophys. Res. Comm., 40, 636 (1970).

63. Colvin, J.R., Can. J. Biochem. Physiol. 39, 1921 (1961),

64. Mascarenhas, J.P., Biochem. Biophys, Res. Comm., 41, 142 (1970). 1 0 2

65. Anderson, J.S., Matsuhashi, M. t Haskin, M.A., andStromin- g e r, J.L., Proc. Natl. Acad, Sci., U.S., 53, 881 (1965).

66. Wright, A., Dankert, M., Fennesey, P., and Robbins, P.W., Proc. Natl. Acad. Sci., U.S., 57, 1798 (1967).

67. Crestfield, A.M. and Allen, F.W., Anal. Chem., 27, 769 (1955).

68. Albersheim, P., Nevins, D.J., English, P.D. and Karr, A., Carbohydrate Res., 5, 340 (1967). s 69. Barber, G.A., Methods Enzymol.,8, 307 (1966).

70. Neufeld, E. F., Biochem. Biophys. Res. Comm., 7, 461 (1962),

71. McDonald, M.R., Methods Enzymol., 1_, 269 (1955).

72. Pratt, J.W., Richtmyer, N.K., and Hudson, C.S., J. Am. Chem. Soc., 74, 2200 (1952),

73. Kaplan, N.O., Methods Enzymol., 2, 660 (1955).

74. Wilson, D.M. and Ajl, S., J. Bacteriol., 73, 410 (1957).

75. Hough, Li. and Jones, J.K.N., Methods Carbohyd. Chem., 1, 21 (1962). "

76. Layne, E ., Methods Enzymol., 3, 450 (1957).

77. Dische, Z., J. Biol. Chem., 204, 983 (1953).

78. Ames, B.N. and Dubin, D. T ., J. Biol. Chem., 235, 769 (1960).

79. Barber, G.A., Biochim, Biophys. Acta, 165, 68 (1968).

80. Wallenfels, K. and Kury, G,, Methods Enzymol., 9, 112 (1966).

81. Schachter, H,, Sarney, J ., McGuire, J ,, and Roseman, S., J . Biol. C h e m ., 244, 4785 (19&9).

82. Sober, H,A., Handbook of Biochemistry, Chem, Rubber Co., Cleveland, Ohio, 1968, p. G66.

83. LaForge, F. B. and Hudson, C.S., J. Biol. Chem,, 30, 61 (1917). 103

84. Horecker, B.L. and Smyr.miotis, P.Z., Methods Enzymol., 1, 371 (1955). “

85. White, R.C,, unpublished observation.

86. Hodge, J.E, and Hofreiter, B. T., Methods Carbohyd. Chem., 1, 388 (1962). .

87. Schachman, H.K., Methods Enzymol., 4, 55 (1957).

88. Yphantis, D.A., Biochemistry, 3_, 297 (1964).

89. Bitter, T. and Muir, H ., Anal. Biochem., 4, 330 (1962).

90. P a rk , J.T. and Johnson, M.J., J. Biol. Chem., 181, 149 (1949)

91. De, K.K. and Timell, T.E., Carbohydrate Res., 4, 177 (1967).

92. Sanwal, G.G. and Preiss, J,, Arch. Biochem. Biophys,, 119, 454 (1967).

93. Cabib, E ., Methods Enzymol. 22, 120 (1971).

94. Cabib, E. and Farkas, V., Proc. Nat. Acad. Sci., U.S., 68, 2052 (1971).