INHIBITION OF PHOSPHOMANNM SYNTHESIS IN HANSERULA HOLSTE‘I BY CYCLOHEXlMEDE Thesis for the Degree of M. S . MICHIGAN ST ATE WWERSIW GARY L VAN. WTSMA 1974' ‘ ““5! L L 79.x“. 1“ ‘ * 5 U University \b magma av . HDAE & snu3' ‘ anox mom mc. J LIBRARY B|NDERS ' LENNON". HIGHER! \-'\--' ¢~_. ABSTRACT INHIBITION OF PHOSPHOMANNAN SYNTHESIS IN HANSENULA HOLSTII BY CYCLOHEXIMIDE By Gary J. Van Haitsma The ascomycetous yeast Hansenula holstii Y-2448 elaborates a phosphorylated mannan which is found both in the medium and as a covalently attached capsule. Cyclo- heximide did not inhibit reproduction of whole cells, a phenomenon which could not be related to the presence of phosphomannan. However, the uptake of [1—1“C] leucine and [2,3-3H] valine into cellular protein of protoplasts was inhibited 36.3% and 50.4% respectively by cycloheximide. The incorporation of glucose into phosphomannan in the presence of cycloheximide was also inhibited in both proto- plasts (70.5%) and whole cells (49.1%). This effect was not due to a lack of necessary enzyme, since polysaccharide synthesis is not affected by cycloheximide within the time interval used. It is concluded that a step in the synthesis of phosphorylated mannan is dependent on protein synthesis. This is consistent with studies on the synthesis of cell wall mannan in other yeasts. INHIBITION OF PHOSPHOMANNAN SYNTHESIS IN HANSENULA HOLSTII BY CYCLOHEXIMIDE BY ".‘wf‘ Gary J? Van Haitsma A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Microbiology and Public Health 1974 “if: . is m «w- DEDI CATI ON This thesis is dedicated to my mother and father for their patience, understanding, and support. ii ACKNOWLEDGEMENTS I would like to express my sincere appreciation to Dr. S. H. Black for his thoughtful counsel throughout this investigation, and for his patience and critical suggestions during the preparation of this thesis. Sincere thanks are also extended to Dr. E. S. Beneke and Dr. P. K. Kindel for their guidance. The author also wishes to extend his gratutude to Dr. A. L. Rogers for his patience in listening to my problems for the duration of this effort. iii TABLE OF CONTENTS DEDICATION. . . ACKNOWLEDGEMENTS. LIST OF TABLES. . LIST OF FIGURES . INTRODUCTION. LITERATURE REVIEW . . . . . . . . . . . . . Capsular Material. . Phosphomannan Production . Phosphomannan Structure. Cell Wall Structure. . Phosphomannan and Mannan Biosynthesis. Effect of Cycloheximide on Cell Wall Biosynthesis . . Effect of 2- -deoxy- D- glucose on Cell Wall Biosynthesis . . Relationship of Peptide and Polysaccharide Synthesis. . . . . . . MATERIALS AND METHODS . . . . . . . . . . Organisms. . . Cultural Methods and Media . . . . . Phosphomannan Preparation. . . . . . . . . Cycloheximide. Preparation of Protoplasts . . . . Incorporation of Labeled Amino Acids . Incorporation of Labeled Glucose into Phosphomannan by Protoplasts . . Incorporation of Labeled Glucose into Phosphomannan by Wh01e Cells . Radioactivity. . . RESULTS . Characterization of the Capsule. . Toxicity of Cycloheximide for Whole Cells. Effect of Cycloheximide on Protein Synthesis in Protoplasts . . Effect of Cycloheximide on Phosphomannan Synthesis. ' iv Page ii iii vii 0 0 oo own-hum (N l'--l hi PHHHHFJPHH h‘ 01 -bLNOihHHFJ ta PJH \um H co NH 000 NM 0% Page DISCUSSION. . . . . . . . . . . . . . . . . . . . . . 31 LITERATURE CITED. . . . . . . . . . . . . . . . . . . 38 Table LIST OF TABLES Effect of various treatments on the cell capsule. . . . . . . . . . . Effect of cycloheximide on growth as determined by auxanographic technique. Effect of various concentrations of cyclo- heximide on growth as determined by tube dilution tests . . . . . . . . . . . . . Effect of various concentrations of cyclo- heximide on the growth of H. wingei VIA. Effect of phosphomannan on cycloheximide activity . . . . . . . . . . . . . . . . . Effect of cycloheximide on phosphomannan synthesis. . . . . . . . . . . . . . . vi Page 19 21 22 23 25 29 LIST OF FIGURES Figure Page 1 Incorporation of [1-1“C] leucine and [2,3- H] valine into protoplast protein in the presence of cycloheximide (SO 27 ug/ml) . . . . . . . . . . . . vii INTRODUCTION Phosphomannans are phosphorylated mannans character- istically produced by yeasts of the genus Hansenula; the quality of the extracellular phosphomannan has, in fact, been used in a phylogenetic classification of this genus (57). The phosphomannans have useful physical properties: in aqueous solution, they are highly viscous, thixotropic substances which may be of considerable value as thickeners and suspending agents (57). The phosphomannans also have unique biological properties: they have the ability to bind complement in vitro (26) and they induce the production of interferon in rabbits (50). Pilot plant studies have shown that phosphomannans may be economically produced (39). Therefore, information regarding the synthesis of these potentially important polysaccharides would be useful. In addition to occupying an extracellular location as in Hansenula, phosphorylated mannans are found in all yeast cell walls. These mannans, displaying a wide range in the mannose to phosphate ratio (37), are present as a protein- mannan complex (35) with the protein and mannan synthesis being interdependent (13). Moreover, by virtue of their thixotropic character, phosphomannans might, after being synthesized by the cells, form a capsular layer prior to their release. 2 The hypothesis has been suggested for Candida that extracellular and cell wall polysaccharides are synthesized by the same pathway (6). It is reasonable, therefore, to propose for Hanaenula a comparable biosynthesis of phos- phorylated mannans which are present both extracellularly and in the cell walls. It is the purpose of this study to show a requirement for protein synthesis in the production of these phosphorylated mannans produced by Hansenula. This should provide information on pathway coincidence. LITERATURE REVIEW Capsular Material Although much is known about bacterial capsules, little information is available on the subject of yeast capsules. With the exception of Cryptococcus neoformana, no direct evidence for capsule is given in the literature. Extracellu- lar polysaccharides produced by yeasts are referred to as capsular material irrespective of their relationship to the cell surface. .According to Wickerham (55), Hansenula capsulata was so named due to "the capsular material surround- ing the cells" but no further information is presented except that the colonies are extremely mucoid. Subsequently Wickerham (56) referred to capsules and capsular material in Hansenula holstii, but he presented no direct evidence for a capsule. A layer resembling a capsule is shown sur- rounding Hansenula wingei cells in electronmicrographs (6). A surrounding capsular layer may also be seen in electron- micrographs of H. holstii (Dr. S. H. Black, personal communication). Phosphomannan Production Many species of Hansenula produce extracellular phospho- mannans (15,46). Anderson et a1. (2) demonstrated that phosphomannans produced by Hansenula holstii are generated 4 extracellularly in high yields from glucose: if the initial glucose concentration in the medium is between 5 and 6 percent, greater than 50 percent of this glucose is incor- porated into the phosphomannan. Other medium constituents for optimal laboratory production of the phosphomannan have been reported (2). Phosphomannan Structure The structure of this polysaccharide is characterized by its phosphodiester linkages between the carbon-l-hemiacetal position of one mannose unit and the carbon-6 position of another (16,48). The mannose to phosphate ratio in the phosphomannan.of H. holstii is approximately 5:1 (2,16,47); however, this ratio varies for other species within the genus (47). Using periodate oxidation, Jeanes and Watson (17) investigated the glycosidic linkages which connect the a- mannosyl residues in H. holstii and reported these to be 1+2 and 1+3. The fact that these linkages are found in a 3:5 ratio led them to propose a repeating unit containing 10 a-mannosyl residues (17). A subsequent analysis of phosphomannan products produced by mild acid hydrolysis showed 3 different types were present. These products were a phosphorylated pentasaccharide (65%), a high molecular weight fragment which was resistant to further mild acid hydrolysis (9%), and other small fragments which were not characterized. It was demonstrated that the pentasaccharide had the structure P-6-Manp-a-(l+3)-Manp-a-(l+3)-Manp-a-(1+3)- Manp-a-(l+2)-Man. The high molecular weight portion is thought to be the "core" to which the other oligosaccharides 5 are linked (10). Further analysis is necessary to clarify the structure of the phosphomannan from H. holstii. Cell Wall Structure The polysaccharide content of isolated cell walls from Baker's Yeast was investigated by Northcote and Horne (35). The cell wall composition was found to be 60% polysaccharide distributed as 29% glucan and 31% mannan. Northcote and Horne also reported 13% protein and 8.5% lipid in the cell wall; however, the high percentage of lipid may be due to cell membrane contaminants. Two layers in the cell wall were observed, one of which was thought to be pure glucan and the other a protein-mannan mixture (35). Mundkur (29) also reported the occurrence of two distinct layers. In addition, he demonstrated the mannan nature of the outer layer and concluded that the inner layer must consist of glucan. Using the freeze-etch technique, Moor and Muhlethaler (27) rarely observed a distinct boundary between layers in the cell wall. The "cross fractured" cell walls showed a change in the granular structure which grows coarser toward the inner surface. This may be due to the mannan impregnating the coarser glucan structure near the cell surface. Cell wall regeneration studies of S. cerevisiae protoplasts in gelatin have shown that a glucan fibrillar network is formed on the surface of the protoplasts. This network is then masked by a "cement substance" of mannan which is produced between and over the fibrils (32). This information may explain why Moor and Muhlethaler (27) did 6 not observe a sharp change in the granularity of the wall but rather a gradual blending. The regeneration of cell walls by protoplasts of yeast cells provides a useful tool in the study of cell wall structure and biosynthesis. When protoplasts of S. cere-' visiae are cultivated in liquid media, they form only a basic fibril mesh of glucan over the cell surface; but they do not revert to normal cells (32). Necas (30,31) demon- strated that, when impregnated into gelatin, most S. cere- visiae protoplasts could regenerate their cell walls within 24 hr. The regeneration of the cell wall in protoplasts of all budding yeasts examined proceeds primarily in the same way as that for S. cerevisiae (34). This information indi- cates two distinct layers in the yeast cell wall synthesized by two different mechanisms. Phosphomannan and Mannan Biosynthesis Basic information concerning synthesis of extracellular phosphomannan and the cell wall mannans has been determined. It seems useful to present the biosynthetic pathways which produce these similar polysaccharides. A definite parallel in the early stages may be observed. Biosynthetic studies of phosphomannan produced by H. holstii showed that glucose is incorporated with the carbon chain intact (7). The glucose is epimerized to mannose as a hexose phosphate and then converted to guanosine diphosphate mannose (7). The transfer of mannose from GDP-mannose to particle bound acceptors is catalyzed by a particulate enzyme fraction in H. holstii. This same enzyme fraction also 7 transfers some of the B-phosphoryl groups from the GDP- mannose (8,20). These phosphoryl groups may be the source of the phosphates involved in the phosphodiester linkages in mannan and phosphomannan. Mayer (24) showed that GDP-mannose serves as a sub- strate for phosphomannan synthesis in Hansenula capsulata and that both the mannosyl and phosphate residues of the polysaccharide are derived from the nucleotide mannan. GDP-mannose has also been reported to be a precursor of cell wall mannan in Saccharomyces carZsbergensia (1), and, as is the case with Hansenula, the mannan is bound to particles by a particulate preparation (4). In further studies, Bretthauer and Irwin (9) observed that mannose and B-phosphoryl group from GDP-mannose are transferred to particle bound acceptors by a particulate enzyme fraction in H. holstii so that the ratio of bound mannose to phosphate is about 7:1. Bretthauer and Irwin (9) also suggested that the mannose and phosphate are present as glycoproteins. Oligosaccharides through mannotetrose containing a terminal mannose-6-phosphate have been isolated. It has been proposed that mannose and mannose-l-phosphate from GDP-mannose are transferred to certain acceptors where both mannosyl-mannose and mannose-l-P-6-mannose linkages are synthesized (9). Using S. cerevisiae, Tanner (51) presented evidence for a lipophilic mannosyl intermediate in mannan biosynthe- sis. Sentandreu and Lampen (43) also reported the presence of a lipid intermediate in the synthesis of S. cerevisiae mannan. More recently it has been reported that dolichol 8 monophosphates serve as acceptor lipids in the transfer of mannose from GDP-mannose in S. cerevisiae. This process is catalyzedfby the same particulate preparation that has been shown to synthesize mannan and phosphomannan (52). Effect of Cycloheximide on TCell WalITBiosynthesis The antibiotic cycloheximide has proved useful in studies of cell wall synthesis. Cycloheximide halts protein synthesis to a variable extent in eukaryotic cells (18,44) by preventing translocation of ribosomes on m-RNA (25). In the more resistant organisms, resistance has been shown to be a property of the 605 ribosomal subunit (38). Although the protein content of the yeast cell wall is relatively low (36), cycloheximide has been shown to inhibit the regeneration of the cell wall by sensitive protoplasts (49). Necas et a1. (33) reported that cycloheximide did not block the synthesis of the entire cell wall. Electronmicro- graphs showed that only the mannan matrix was not produced; however, the fibrillar network of glucan was readily regenerated in the presence of cycloheximide. Studies of cell wall incorporation in S. cerevisiae showed that cycloheximide inhibits the incorporation of labeled amino acid but only partially inhibits the incorpora- tion of labeled glucose (41). Subsequently it was demon- strated that the percentage of label from ll’C-glucose incorporated into the glucan portion of the cell wall was greatly increased in the presence of cycloheximide, while the percentage in the mannan was decreased radically (12). 9 Cycloheximide greatly reduces the amount of threonine incor— porated into the cell wall, as well as terminating glucose incorporation into the mannan layer after 5 minutes (12). Morris (28) reported that in Chlorella, although protein synthesis is inhibited almost immediately, cycloheximide does not affect polysaccharide synthesis for several hours. The influence of cycloheximide on the enzyme mannan synthe- tase has also been studied and no effect was detected (13). Effect of 2-deoxy-D- lucose on Cell W511 Biosynt esis A glucose analog, Z-deoxy-D-glucose, has also been used to study cell wall synthesis. Although 2-deoxy-D- glucose has little effect on protein synthesis it strongly inhibits the appearance of both mannan and protein in the cell wall (13). Relationship of Peptide and Polysaccharide Synthesis Analysis of the matrix material in the cell wall has shown it to consist mainly of mannan covalently linked to polypeptides to yield a glycopeptide (40). Considering the effects of cycloheximide and Z-deoxy- D-glucose, it appears that there is a parallel inhibition in the synthesis of protein and polysaccharide in the cell wall. Farkas et a1. (13) suggested that both components of the mannan-protein complex are necessary for the secretion of either part of that complex. Sentandreu and Lampen (42) investigated the nature of this block and found that cycloheximide caused the 10 accumulation of GDP-mannose. Cycloheximide neither blocks the epimerization of glucose to mannose nor the conversion of mannose to its nucleotide. Rather it blocks the glycosylation or synthesis of the peptide as would be expected. Sentandreu and Lampen suggested that in cell wall synthesis the glucan fibrillar network is produced independently of protein synthesis. Since the amorphous matrix of the wall is primarily a mannan-polypeptide, it is suggested that the protein moiety is synthesized first and is subsequently g1yco$y1ated during transport to its site of incorporation (42). This hypothesis would explain why both cycloheximide and Z-deoxy-D-glucose inhibit the matrix formation. MATERIALS AND METHODS Organisms A diploid strain (NRRL Y-2448) of Hanaenula holstii, which produces copious extracellular phosphomannan, and a haploid strain (VIA) of Hansenula wingei were obtained from Dr. M. E. Slodki, Northern Regional Research Laboratory, Peoria, Illinois. The strain of Baker's Yeast used was from a dried commercial source. Cultural Methods and Media Stock cultures of Baker's Yeast and VIA were main- tained on yeast maintenance medium (YM) as described by Haynes et al. (14). In order to maintain the phosphomannan producing characteristic of Y-2448, phosphomannan maintenance medium (PMM) was employed (Dr. M. E. Slodki, personal communication). For the same reason, the transfer interval (2 months) of Y-2448 was twice that used for the other organisms. PMM contained 0.1% glucose, 0.1% yeast extract, 0.1% malt extract, and 0.15% peptone. For optimum phospho- mannan production, OP medium was used. This medium contained 6% glucose, 0.1% corn steep liquor, 0.1% tryptone, 0.5% monobasic potassium phosphate, and 0.5% (v/v) Speakman's Salt Solution B (2). Winge's medium (2% glucose and 0.3% yeast extract) was used for the antibiotic testing. For 11 12 all media, agar (2%) was added when solid medium was required. Phosphomannan Preparation Phosphomannan was isolated and purified according to the method of Jeanes et a1. (16) with modifications. The cells were grown in OP medium for 120 hr at 25° C with gentle shaking on a New Brunswick rotary incubator. After this time it was assumed that the glucose had been depleted and the maximum amount of phosphomannan had been produced (2). The culture appeared very smooth and viscous at this time. The culture was diluted with 1/2 vol water and the pH was adjusted from 3.6 to about 6 with potassium hydroxide. Potassium chloride and ethanol were added with stirring to give final concentrations, in percent, of 0.5 (w/v), and 25 (w/v) respectively. The salt and alcohol thinned the culture to facilitate the removal of the cells. After standing for 75 min, the solution was centrifuged until all the cells were removed. The supernatant fluid was a brilliantly clear yellow. After increasing the concentra- tion of potassium chloride to 1% (w/v), the total concen- tration of ethanol was brought to 50% (v/v) which caused the complete precipitation of the phosphomannan. The precipitate was a soft, cohesive gumlike mass which settled rapidly. The supernatant fluid was decanted. The soluble materials that were enmeshed in the gum mass were partially removed by kneading and decantation. The phosphomannan was further purified by reprecipitation twice from an aqueous solution containing 2.5% of the gum l3 and 1% of potassium chloride with ethanol at a concentration of 50%. For later use, Phosphomannan Y-2448 was then dehydrated by adding, with vigorous stirring, a 10% aqueous solution of the gum to 15 vol absolute methanol containing 0.05% potas- sium chloride. The precipitate was filtered, washed by resuspension in methanol, and dried in vacuo under anhydrous conditions at room temperature. Cycloheximide A stock solution of cycloheximide (Calbiochem) was made up in absolute ethanol to a concentration of 10 mg/ml. Preparation of Protoplasts Protoplasts were prepared according to the method of Kozak and Bretthauer (19) with minor modifications. The cells were grown in OP medium with 2% glucose in a New Brunswick rotary incubator for 24 hrs and harvested by centrifugation (10,000xg, 10 min). They were washed in 3 vol of 1% potassium chloride to remove the adhering phospho- mannan and then brought to a concentration of 1x109 cells/ml in 0.02M Tris pH 7.6 and 0.06M 2-mercaptoethanol. After incubating at room temperature for 1.5 hr, the cells were centrifuged and washed with 4 vol of potassium phosphate buffer pH 6.5 containing 1.25M potassium chloride. Subse- quently the cells were resuspended in the same buffer at a concentration of 1x109 cells/ml and 0.04 vol snail gut juice (Industrie Biologoque Francaise) was added. This suspension was then incubated with gentle shaking at 30° C. After 7 hr, l4 practically all the cells were protoplasts as determined by plate counts and by examining the cells in hypotonic solu- tion under a phase microscope. Incorporation of Labeled Amino Acids Protoplasts were prepared as described above, removed by centrifugation (1000xg, 10 min), and washed in 1 vol of 1.25M potassium chloride in 0.04M potassium phosphate buffer pH 6.5. The cells were then suspended in a medium contain- ing 1.25M potassium chloride, 0.04M monobasic potassium phosphate, and 1% glucose at a concentration of 5x108 cells/m1 (19) and 2 m1 of this was dispensed into each of six 25 m1 Erlenmeyer flasks. The flasks were chilled on ice and 10 ul of the cycloheximide stock solution were added to three of the flasks to give a final concentration of 50 pg per m1. These flasks, along with three controls which had 10 pl of ethanol added, were then incubated at 25° C for 15 min with shaking. After being rechilled, 1 uCi of [1-1“C] leucine and 10 uCi of [2,3-3H] valine were added to each flask. After mixing well, 100 pl portions were removed and the flasks were incubated at 25° C with shaking. Portions of 100 pl were removed every half hour for the first 3 hr and a final one was taken after 4 hr. The portions were handled according to the method of Mans and Novelli (23) as adapted. Each portion was pipetted onto a 2.4 cm GF/c glass fiber disc (Whatman) and exposed to a stream of warm air for 15 sec. The discs were then placed in an ice-cold solution which contained 10% TCA (w/v), 0.1M leucine, and 0.1M valine. Approximately 3 m1 15 of the above solution was used per disc for the first wash. After standing for 60 min with occasional swirling, the TCA solution was decanted and the discs were washed in the same volume of the solution for 15 min. The liquid was then decanted and the discs were plunged into 5% TCA (w/v) at 90° C and held at this temperature for 30 min. The TCA was removed. The discs were suspended in ether-ethanol (1:1) and incubated at 37° C for 30 min. Finally the discs were suspended in ether for 15 min at room temperature. They were then air dried, placed in scintillation vials, and counted. Incorporation of Labeled Glucose into Phosphomannan by Protoplasts The procedure for incorporation of labeled glucose was the same as that for the incorporation of labeled amino acids up to the point where the amino acids were added. Here, rather than labeled amino acids, 5 uCi of [U-1“C] glucose were added to each flask and the flasks were incu- bated in a water bath at 25° C with shaking. After 3 hr, the cells were removed and chilled on ice. The protoplasts were removed by centrifugation (1000xg, 10 min) and the supernatant fluid (M) was removed and retained. Portion M was cleared of all debris by centrifugation (10,000xg, 10 min). The protoplasts were washed twice with buffer and then lysed by repeated freezing and thawing. The lysate was centrifuged (10,000xg, 10 min) and the supernatant fluid (L) removed. Fractions L and M were then centrifuged (25,000xg, 1 hr) and the supernatant fluids removed. Carrier 16 phosphomannan, 50 pg/ml, was then added to each sample. Portions of 100 pl were pipetted onto GF/c discs (2.4 cm), dried in a stream of warm air for 10 sec, and immersed in 75% ethanol containing 0.1M glucose and 0.5% potassium chloride. After standing for one hr, the ethanol was decanted and the discs were washed for another hour in the same solution. Finally the discs were washed for 30 min in ether-ethanol (1:1) and 15 min in absolute ether. After drying, the discs were placed in scintillation vials and counted. Incorporation of Labeled Glucose into Phosphomannan by Whole Cells The cells were prepared in the same manner used to make protoplasts. After being washed with 3 vol of 1% potassium chloride, the cells were centrifuged (10,000xg, 10 min) and suspended in 1.25M potassium chloride in 0.04M potassium phosphate buffer pH 6.5 to a concentration of 5x108 cells/ml. The cells were centrifuged again (5000xg, 10 min) and resus- pended to the same concentration in a medium containing 1.25M potassium chloride, 0.04M monobasic potassium phosphate, and 1% glucose. Two ml of this suspension were dispensed into each 25 ml Erlenmeyer flask used, and these were then chilled on ice. To each flask, cycloheximide was added to give a final concentration of 50 ug/ml, and the flasks were incubated in a water bath at 25° C with shaking for 15 min. They were then rechilled and 5 uCi of [U-1“C] glucose were added to each flask. Controls were prepared in the same manner except ethanol was added rather than cycloheximide stock 17 solution. The flasks were then reincubated at 25° C with shaking. After 3 hr the flasks were removed from the water bath and chilled. The cells were removed by centrifugation (10,000xg, 10 min), and the supernatant fluid was treated as stated in the preceding section and counted. Radioactivity Radioactivity on glass fiber discs was measured with a Packard Tri-carb liquid scintillation spectrometer, model 3320. The discs were immersed in 15 m1 scintillation fluid [4.0 gms 2,5-bis-2-(5-tert-Buty1benzoxazoly1)-Thiophene (BBOT) made up to 1 liter with scintillation grade toluene] and counted repeatedly to obtain mean values. The BBOT was purchased from Packard Instrument Company, Inc., and the toluene was a product of J. T. Baker Chemical Company. RESULTS Characterization of the Capsule Hansenula holstii Y—2448 cells grown for 30 hr in OP medium were harvested by centrifugation, suspended in Pelikan India ink and observed with a phase contrast micro- scope. A large clear area that represented the capsule surrounded the cell. This area had a diameter of 1.5 to 2 times that of the cell. Extensive clumping of the ink was observed in the intercellular space; however, this effect was not seen if the cells were washed prior to their sus- pension in the ink. This clumping effect was probably due to some constituent of the medium. Since washing the cells with water did not remove the capsule, further studies were done to determine the nature of the junction between the capsule and the cell wall (Table l). The results are based on the majority of the cells that initially exhibited large capsules. In the cases of treatment with water, salt, or heat, the reduction of the capsular size was proportional to the initial size of the capsule; however, when the capsules were treated with acid the rate of removal with respect to size was constant. These results indicate that the capsule is covalently bound to the cell surface. 18 19 .wocflmow xampmnm once umommw on omwo m.oH:mmmo esp momsmo omam wcm Hw>oEoH HHHSmmmo oEom momsmo Hoxfle xoppo> m :H mafia sumo mowpmpfim< «an .++++ >3 wopmofiwcfl ma Hm>oaoh pwHSmmmo Hmpoe . he ofinm>eomno o:oz LEH .u com ”Ho: 2H.o + LEA .u omN ”Ho: 25.0 ++ LEN .0 com ”Ho: zA.o ++++ Has .u com “Ho: 20.H +++ Ham .0 com ”Ho: 20.H ++ Ham .0 com ”Hum 20.H +++ LEA .u 0mm ”Hum 20.N ++++ LEN .0 com “Ho: 20.N + enamm mfimamfi mu .HNH OHQG>HOmDO 0202 flHo>OHV L SN ”Hum 0H o\ wflo< “we: mfineseomno oaoz flfio> Hy Homo 2H meofinsfiom Sfimm oHnm>Homno oaoz flHo>HV Km «xoanm>pomno ocoz mHo>HV KN oHnm>Homno ocoz mHo>HV xH Lopez omfim HmHSmmmo mo :ofluoswom mcofipflunou psoEpmoHH « madmmmo HHoo OS“ GO muGoEHmmhfl WSOfiHm> MO HUQMMMII.H mdm