JOURNAL OF BACTERIOLOGY, Feb. 1973, p. 627-636Copyright 0 1973 American Society for Microbiology

Vol. 113, No. 2Printed in U.S.A.

Biosynthesis of Glycogen and Starch inCryptococcus laurentiiJOHN C. SCHULTZ' AND HELMUT ANKEL

Department of Biochemistry, The Medical College of Wisconsin, Milwaukee, Wisconsin 53233

Received for publication 30 August 1972

Cells of Cryptococcus laurentii, when grown in liquid culture on 2% glucoseclose to neutral pH, showed glycogen granules throughout the cytoplasm.Glycogen levels of C. laurentii cells reached maximal levels just before onset ofstationary phase. Concomitantly, a sharp rise in total and specific activity ofglycogen synthetase was observed. Conversely, glycogen phosphorylase reachedits highest specific activity approximately 3 hr after the glycogen peaked andremained high until most of the endogenous glycogen was utilized. Uridinediphosphoglucose pyrophosphorylase activity was always an order of magnitudehigher than glycogen synthetase during log phase, but fell off rapidly after thecells reached stationary growth. Kinetic properties of the glycogen synthetaseshowed that the enzyme is always activated by glucose-6-phosphate, althoughthe degree of activation by glucose-6-phosphate was found to be somewhatvariable. The accelerated uptake of glucose commencing with the onset ofstationary phase is explained by the rapid formation of extracellular acidicpolysaccharide, which continues as long as there is glucose in the medium. Incells grown at pH 3.4, where no detectable extracellular acidic polysaccharidewas formed, glucose uptake drastically declined when the cells reachedstationary phase. These cells also contained glycogen-like granules in thecytoplasm. The evidence presented indicates that these granules are in factglycogen, and that its structure does not resemble that of the starch excreted bycells grown at acidic pH.

Fungi of the genus Cryptococcus areencapsulated, nonfermentative, yeastlikeorganisms that multiply by budding and areclassified among the fungi imperfecti (12).These organisms excrete starch when grown atlow pH; when grown close to neutral pH, wherestarch production is not observed, they excretean acidic polysaccharide, (1, 6). The acidicpolysaccharide of C. laurentii is composed ofmannose, xylose, and glucuronic acid in a ratioof 5:2:1 (1, 9) and resembles the capsularmaterial of the organism. Our earlier studieswith C. laurentii (20), in agreement withfindings by Ruinen et al. (19), showed thepresence of glycogen granules throughout thecytoplasm of cells grown close to neutral pH.Furthermore, they revealed that most of thecell's glycogen phosphorylase and some of itsglycogen synthetase are bound to these

'Present address: Department of Biochemistry, TheUniversity of Texas Medical School at San Antonio, SanAntonio, Tex. 78229.

glycogen granules (20). This communicationprovides information on the accumulation andutilization of glycogen in C. laurentii and onactivity changes of uridine diphosphoglucose(UDP-glucose) pyrophosphorylase, glycogensynthetase, and glycogen phosphorylase duringgrowth close to neutral pH. We also presentdata that compare cells grown at low pH withthose grown at neutral pH. These data indicatethat acidic polysaccharide accumulates in themedium only when the pH is close toneutrality. They also demonstrate that theextracellular starch of cells grown at low pH isstructurally different from the glucose polymercomprising the granular material within thesecells which resembles glycogen.

MATERIALS AND METHODSOrganism. C. laurentii var. flavescens

(NRRL-Y-1401) was grown and harvested asdescribed by Ankel and Feingold (4).Media. The following liquid media were

employed. Medium I contained 2% (w/v) glucose,

627

on May 5, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

SCHULTZ AND ANKEL

0.1% urea, 0.1% KH2PO4, 0.05% MgSO4.H20, and0.2 mg of thiamine hydrochloride per liter (14). ThepH of this medium was 6.8. Unless mentionedotherwise, medium I was employed. Medium IIcontained 1.2% (w/v) glucose, 0.12% (NH4)2SO4, 0.1%KH,PO4, 0.05% MgSO4.H20, and 0.2 mg ofthiamine hydrochloride per liter. The pH of themedium was brought to 3.4 with HCI (5).

Fractionation of cell-free extracts. Alloperations were carried out at 0 to 4 C. The cells wereharvested by centrifugation at 12,000 x g for 15 min,and were washed with 10 volumes of 1% NaCl andwith 10 volumes of distilled water; the latter washingwas repeated a second time, and the cells were fi-nally washed with 10 volumes of 0.1 M tris(hydroxy-methyl)aminomethane (Tris)-hydrochloride buffer(pH 7.5) containing 0.5 g of ethylene diaminetetra-acetate (EDTA)/liter (buffer A). The packed, washedcells (5 g, dry weight) were suspended in an equalvolume of buffer A and disrupted by sonic oscillation(4). The broken cell suspension was centrifuged at28,000 x g for 15 min. The resulting supernatant fluid(S-1) was centrifuged at 100,000 x g for 30 min. Afterremoval of the supernatant fluid (S-2), the particulatefraction was suspended in 10 ml of 0.10 M Tris-hydro-chloride buffer (pH 6.5) containing 0.05 g of EDTA/liter (buffer B) and centrifuged again at 100,000 x gfor 45 min. The sediment separated into three dis-tinct layers (8). Top, middle, and bottom layers werecarefully isolated and resuspended in approximately2 volumes of buffer B.

Chemicals, enzymes, and substrates.2-(N-morpholino)-ethanesulfonic acid (MES) waspurchased from Calbiochem (Los Angeles, Calif.).Rabbit liver glycogen type III andglucose-6-phosphate were obtained from SigmaChemical Co., (St. Louis, Mo.). C. laurentiiextracellular acidic polysaccharide was a gift from D.S. Feingold, University of Pittsburgh.

UDP-14C-glucose (237 mCi/mmole), adenosinediphosphate (ADP)-l4C-glucose (228 mCi/mmole),guanosine diphosphate (GDP)- lC-glucose (28mCi/mmole), cytidine diphosphate (CDP)- l4C-glucose (8 mCi/mmole), and thymidine diphos-phate (TDP)- "IC-glucose (40 mCi/mmole) werepurchased from New England Nuclear Corp. andIntemational Chemical and Nuclear Corp. Allother chemicals not specifically listed were obtainedfrom commercial sources and were of the purestgrade available. Crystalline sweet potato ,B-amylase(a-1, 4-glucan maltohydrolase, EC 3.2.1.2), crystal-line swine pancreas a-amylase (a-1, 4-glucan 4-glu-canohydrolase, EC 3.2.1.1), Aspergillus nigerglucohydrolase, EC 3.2.1.3, and a-1, 6-glucanohy-drolase, EC 3.2.1.11), maltase (a-D-glucoside gluco-hydrolase, EC 3.2.120), UDP-glucose dehydrogen-ase (UDP-glucose: nicotinamide adenine dinucleo-tide [NAD] oxidoreductase, EC 1.1.1.22) werepurchased from Sigma Chemical Co.Chromatography. Paper chromatography was

carried out on Whatman no. 1 or 3 MM paper withthe following solvents: solvent A, 1-propanol-ethylacetate-water (7:1:2, v/v); solvent B, 95% ethanol-1M ammonium acetate, pH 7.5 (7:3, v/v) containing 1

mM EDTA; and solvent C, n-butanol-95%ethanol-water (5:1:4, v/v). Carbohydrates on paperchromatograms were detected with p-anisidinephthalate or silver nitrate-acetone followed byalcoholic sodium hydroxide spray (25).Paper chromatograms were analyzed for

radioactivity with a Nuclear-Chicago Actigraph III orby radioautography.

Analytical methods. Glucose was determinedwith glucose oxidase (Glucostat reagents ofWorthington Biochemical Corp.). Protein wasmeasured according to Lowry et al. (13) with bovineserum albumin (Sigma Chemical Co.) as a standard.The dry weight of the cells was determined after

repeated washing of the cells with cold water,followed by drying in vacuo to constant weight.

Polysaccharide analysis. Glycogen was extractedfrom the cells according to the method of Rothmanand Cabib (17). The final pellet was suspended inwater, and the glycogen content was determinedaccording to the iodine method of Krisman (11).Absorption maximum and specific absorption ofI.-glycogen complexes were determined according tothe method of Chester and Byrne (7). In the growthexperiments, the amount of starch was determinedby weighing the amount of starch isolated accordingto the procedure below and by measuring theabsorption of the iodine complex at 690 nm (11).Determination of the acidic polysaccharide.

The concentration of acidic polysaccharide in culturesupematant fluids was determined by measuring itsrelative viscosity. A standard curve was prepared byuse of known concentrations of authentic acidicpolysaccharide. The relative viscosity of a solution ofthe acidic polysaccharide is directly proportional toits concentration up to 0.02% (w/v), as described byA. Cohen (Ph.D. thesis, Univ. of Pittsburgh,Pittsburgh, Pa., 1966). Samples of the culturemedium were taken at timed intervals, the cells wereremoved by centrifugation, and the viscosity of 3 mlof supernatant fluid was measured with anOstwald-type viscometer at 37 C. All dilutions weremade with uninoculated growth medium. Toeliminate viscosity changes due to pH difference,samples from lowpH cultures were brought topH 6.4prior to the viscosity measurements. The viscosity ofthe acidic polysaccharide is essentially unchangedbetween pH 5 and 8, but decreases belowpH 4.5 (9).

Isolation and purification of starch. Theextracellular starch was isolated from 3 liters of theculture medium of C. laurentii grown in medium II.The method used to isolate the crude polysaccharideswas that of Kooiman (10). This material was dried invacuo. Its yield was 1.45 g.To a 25% solution of crude polysaccharides in

water (600 ml), 40 ml of a 2.5% (w/v) Cetavlon(cetyltrimethyl ammonium bromide) solution wasadded with stirring to precipitate all of the acidicpolysaccharide as the Cetavlon salt. The resultingprecipitate was removed by centrifugation anddiscarded. Four volumes of 95% ethanol were addedto the supernatant fluid, and the suspension wasallowed to stand at 4 C overnight. The precipitatedstarch was collected by centrifugation, washed by

628 J. BACTJOL.

on May 5, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

VOL. 113, 1973 BIOSYNTHESIS OF GLYCOGEN AND STARCH IN C. LAURENTII

resuspension in 95% ethanol, and recentrifuged. Thefinal precipitate was dried in vacuo. Its yield was0.106 g.Enzyme assays. Glycogen phosphorylase

(a-i, 4-glucan: orthophosphate glucosyltransferase,EC 2.4.1.1) activity was measured in the syntheticdirection by following the rate of incorporation of14C-glucose from "4C-glucose-i-phosphate intoglycogen as described previously (20).

Glycogen synthetase (UDP-glucose: glycogena-1,4-glucosyl transferase, EC 2.4.1.11) activity wasdetermined by following the rate of incorporation of"4C-glucose from UDP-14C-glucose into glycogen. Thereaction mixtures contained 13 mg of rabbit liverglycogen/ml,

_7.2 mm UDP-'4C-glucose (237

mCi/mmole), 7.2 mm glucose-6-phosphate (whenpresent), and enzyme in 0.21 M glycylglycine buffer,pH 7.5, in a final volume of 70 uliters. The reactionmixtures were incubated at 25 C for 10 min. Reactionmixtures containing heat-denatured enzyme servedas controls. The reaction mixtures were subjected todescending chromatography in solvent B.Appropriate areas at the chromatographic originwere excised and counted in a gas-flow counter.

UDP-glucose pyrophosphorylase (uridinetriphosphate [UTP]: a-D-glucose-i-phosphateuridyltransferase, EC 2.7.7.9) activity was measuredby spectrophotometric determination of the amountof UDP-glucose formed from UTP andglucose-i-phosphate in a two-step assay employingNAD+ and UDP-glucose dehydrogenase(UDP-glucose:NAD+ oxidoreductase, EC 1.1.1.22).The reaction mixture consisted of 0.25 MTris-hydrochloride buffer, pH 7.5, 2.5 mm MgCl2, 2.5mM UTP, 1.25 mm glucose-i-phosphate and enzyme,contained in a final volume of 0.2 ml. Afterincubation for 5 min at 30 C, the reaction wasstopped by immersing the tubes into boiling water for90 sec. Suitable samples were incubated in amicrocuvette at 30 C in the presence of 0.5 Mglycine-KOH buffer, pH 8.7, 1.2 mM NAD+, 5 mmEDTA, and 400 units of UDP-glucose dehydrogenase,contained in a final volume of 0.2 ml. The reactionwas initiated by the addition of NAD+, and theamount of UDP-glucose originally present wasdetermined from the change in absorbancy at 340 nmafter completion of the reaction.Measurement of the pH of broken C. laurentii

cells. The cells were collected by centrifugation andwashed four times with 10 volumes of cold CO2-freedistilled water. After breaking a 50% suspension ofthe cells (v/v) in cold C02-free water by sonictreatment, the pH of the broken cell suspension wasmeasured with a glass electrode (15).

RESULTSAccumulation and utilization of glycogen

by C. laurentii during growth. C. laurentiicells were grown in medium I as described inMaterials and Methods over a period of 120 hr.The levels of glycogen, glycogen synthetase,and UDP-glucose pyrophosphorylase of the

cells and the glucose content of the mediumare shown in Fig. 1A and B. Glycogenaccumulation was maximal (16.5% of the cells'dry weight) just before the onset of thestationary phase of growth. After that time, thecells began to utilize the stored glycogen. Theseresults are similar to those described forSaccharomyces cerevisiae by Rothman-Denesand Cabib (18), but the C. laurentii cells beganto accumulate glycogen considerably earlier.The appearance of appreciable amounts ofglycogen in log-phase cells was initiallyobserved by electron microscopy (20). It isstriking that endogenous glycogen is utilizedeven before exogenous glucose is significantlydepleted. This will be discussed below.Glycogen synthetase activity during

growth. Figure 1B indicates that glycogensynthetase activity, when assayed immediatelyafter disruption of the cells, was almost entirelyglucose-6-phosphate-dependent, although ameasurable increase in glucose-6-phosphate-independent activity, as well as in the totalamount of enzyme (not shown), was observedcoincident with the simultaneous rise in glyco-gen levels. The decrease in glycogen contentafter 30 hr was accompanied by a simultaneousdecrease in total and specific glucose-6-phos-phate-dependent and -independent synthe-tase activity.Glycogen phosphorylase activity during

growth. Previous studies (20) have shown thatin late log-phase cells 73% of the phosphorylaseactivity is associated with the bottom layerpreparation and that the specific activity ofthis fraction is 65 times that of the crudeextract. The activity of glycogen phosphorylasewas also measured in the crude cell extractsand in the bottom layer fraction during growthof the organism. Activities were measured inthe direction of glycogen synthesis withsaturating concentrations of both glycogen andglucose-i-phosphate present in the assay.Figure 1C illustrates the results obtained inthese studies. The specific activity of glycogenphosphorylase in the bottom layer was low inlog-phase cells, but increased 10-foldconcomitantly with glycogen breakdown andrapid utilization of the exogenous carbonsource. Similar results were obtained for thecrude extracts. The specific activity in thecrude extracts likewise increased when the cellsentered the stationary phase, although lessdramatically than observed with the bottomlayer. Total phosphorylase activity (not shown)was likewise lower in log and early stationaryphase. At 68 hr of growth, there is a 38%decrease in total phosphorylase activity as

629

on May 5, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

SCHULTZ AND ANKEL J. BACTFMOL.

-20 >0

18 ,.1614-

z-12 w

-10 0

8 J- 6

-i- 4w- 28

10 20 30 40 50 60 70 80 90 100 110 120INCUBATION TIME (HOURS)

FIG. 1. Accumulation and utilization of glycogen by Cryptococcus laurentii during growth. In A, sampleswere taken at the times indicated, and glucose and glycogen were determined as described in Materials andMethods. In B, the specific activities of glycogen synthetase in the crude extracts were measured in thepresence (7.2 mM; 0) and absence (0) of glucose-6-phosphate. The specific activity of UDP-glucosepyrophosphorylase in the crude extracts (A) was measured as described in Materials and Methods. In C, thespecific activity of glycogen phosphorylase in the bottom layer fraction (0) and in the crude extracts (0) isshown. Phosphorylase activity was measured as described (20), except that 7.3 mm glucose-l-phosphate was

employed.

compared with that measured at 33 hr, whenthe total activity was highest. The specificactivity, however, during this period remainedessentially constant.UDP-glucose pyrophosphorylase activity

during growth. UDP-glucose pyrophosphoryl-ase, as can be seen in Fig. 1B, was highest inthe early stages of growth, when glycogen ac-

cumulation was greatest, and decreased quite

rapidly after late log phase. When late station-ary-phase fraction S-2, which did not showdemonstrable UDP-glucose pyrophosphorylaseactivity, was added to active fractions from log-phase cells, no effect on the activity was ob-served. This indicates that the lack of activityin late stationary-phase preparations is not dueto inhibitors of UDP-glucose pyrophosphoryl-ase activity.

630

wcr

< wI >-c

°0 0

a. c-E

F- ,

a-E

Li -

0>- U)

>-50ij- E

(DF-

,'-ic

0 0

o Ea...

0n .C0 E

a -O-

(D Ea-0

E

Ct

zcn0

(IT

mar

0.01

.z

on May 5, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

VOL. 113, 1973 BIOSYNTHESIS OF GLYCOGEN AND STARCH IN C. LAURENTII

Exogenous glucose utilization and pro-duction of extracellular polysaccharides. Itwas surprising to observe that C. laurentii cellsreaching stationary phase started breakingdown endogenous glycogen, despite the factthat ample glucose was still present in the me-dium (82% after 30 hr, Fig. 1A), and that thisglucose was taken up in an accelerated fashionduring stationary growth. (In the next 10 hr, theglucose concentration in the medium decreasedto 15% of that originally present.) As can beseen in Fig. 2, this glucose uptake can bequantitatively related to the production ofextracellular acidic polysaccharide, whichstarts to accumulate in the medium oncestationary phase is reached.

It has been observed that cells grown at lowerpH do not excrete extracellular acidicpolysaccharide, but excrete starch instead. InFig. 2, the production of extracellularpolysaccharide is measured by determining theviscosity of the medium of cells originally at pH3.4 (medium II). These data show only a veryslight change in viscosity up to 270 hr of growthof C. Iaurentii in medium II, which is in sharpcontrast to the data obtained with cells grownin medium I. Figure 3 shows the growth curveof cells grown in medium II, the amount ofstarch found in the medium, and the levels ofglucose in the medium. These data reveal that,once the cells reach stationary phase, theglucose content in the medium does not changeappreciably and that the starch concentrationin the medium does not increase. It should bepointed out that the glucose concentration ofmedium II originally was 1.2%, whereas

0

4J

17 E

15aw

1320

a

I

7

5 >-

3 o8

,5

0 20 40 60 80 100 120 140 270

INCUBATION TIME (HOURS)FIG. 2. Acidic polysaccharide production by

Cryptococcus laurentii cultures during growth. (0)Medium I; (0) medium II.

D 120

9

-6o

uo 3J

8

-1.0

- 10.8

- 0.6

a:-- 0.4

0.2

0 25 50 75 100 125 150 175 200 225 250INCUBATION TIME (HOURS)

E2.0 01.5 t1.0 F0.8 <

0.4 zD

0.2 00

0.1

8

FIG. 3. Growth of Cryptococcus laurentii in me-dium I. Starch production (A) was measured asdescribed in Materials and Methods. Glucose in themedium (0) was measured with Glucostat.

medium I contained 2% glucose. However, C.Iaurentii cells when grown in medium I with 1%glucose behaved essentially the same as thosedescribed here, except that the amount ofextracellular acidic polysaccharide whichaccumulated in the medium wascorrespondingly less (A. Cohen, Ph.D. Thesis,Univ. of Pittsburgh, 1966).

Electron micrographs of C. laurentii cellsgrown in medium II revealed glycogen-likegranules in the cytoplasm of log-phase and ofstationary-phase cells, but must less in thelatter case. These cells also had no distinctcapsule, in contrast to cells grown in mediumI. (Electron micrographic observations weremade by J. C. Garancis, Department ofPathology, The Medical College of Wisconsin.)The granules from cells grown in mediumII were isolated as described in Materialsand Methods. The isolated material gavea brown color with iodine with an absorp-tion maximum at 493 nm and a specific ab-sorption for a 0.01% solution (w/v) of 0.26.Both values are similar to those obtained forglycogen of C. laurentii cells grown in medium I(457 nm, 0.24) and for rabbit liver glycogen(492 nm, 0.24), which were simultaneouslymeasured. They are clearly different from thevalues obtained for cryptococcal starch foundin the medium (610 to 630 nm and 0.31,calculated from the Blue Value; see 10). Theythus indicate that the granules found inside thecells grown in medium II are not starchgranules, but represent glycogen. Our datashow that C. laurentii cells, when grown inmedium I (neutral pH), continue to take upexogenous glucose in stationary phase, but usethis glucose primarily to produce extracellularacidic polysaccharide. In contrast, the samecells grown in medium II (acidic pH) cease totake up exogenous glucose once they reachstationary phase. But, under both growth

631

GROWTH -0-

a iSTARCH

- GLUCOSE -

on May 5, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

SCHULTZ AND ANKEL

conditions, endogenous glycogen is depletedduring stationary phase. These differences areundoubtedly due to the difference in pH, bothin the medium as well as inside the cells.Cultures grown for 96 hr in medium I showed apH of 7.5 in the medium, and a pH of 6.2 in thebroken cell suspension. Cultures grown for 48 or120 hr in medium II showed a pH of 2.8 in themedium and of 3.9 to 3.95 in the broken cellsuspension. It appears that the low pH causedstarch formation during log phase, butprevented its formation and that ofextracellular acidic polysaccharide duringstationary phase. In contrast, neutral pHimpaired starch formation during log andstationary phase, but caused the formation ofextracellular acidic polysaccharide duringstationary phase. In both media, however, thecells remained viable for a long time andutilized endogenous glycogen. We have noexplanation for this peculiar behavior at thepresent time.

Distribution of glycogen synthetase.Various fractions from extracts of cells grown inmedium I were assayed for glycogensynthetase. As can be seen in Table 1, themajority of the activity was associated with the100,000 x g supernatant fraction (S-2).Glycogen synthetase activity in crude extractswas found to be very unstable, and essentiallyall activity was lost within a few hours ofstorage at 0 C. In contrast, the glycogen-boundfraction was found to be considerably morestable. Increased stability could be achieved ifthe glycogen pellet was stored at -20 C as suchand suspended in buffer prior to theexperiments. Under these conditions, losses of

TABLE 1. Distribution of glycogen synthetase invarious fractions of Cryptococcus laurentii cells

Fraction ~Total Specific bustrionFraction activitya activity" butio

Sonically. disrupted.ceUs 25,330 127 100S-1 ................ 29,120 173 115S-2 ................ 21,850 190 86Top layer .............. 0 0 0Middle layer .......... 0 0 0Bottom layer .......... 300 500 1Glycogen pelletc ....... 43,700 1206

aExpressed as nanomoles of ["4C]glucose incor-porated per 10 min.

Expressed as nanomoles of ["4CJglucose incor-porated per 10 min per mg of protein.

cObtained by addition of cryptococcal glycogen(final concentration, 4%) to fraction S-2 followed bycentrifugation at 100,000 x g for 30 min.

50 to 75% activity were observed over a periodof 24 hr at - 20 C. The losses in theglucose-6-phosphate-dependent and -inde-pendent activity were nearly identical. Itwas subsequently found that the glycogensynthetase activity in the 100,000 x gsupernatant fraction (S-2) could be almostquantitatively precipitated at 100,000 x g for30 min in the presence of 40 mg of cryptococcalglycogen/ml. The total activity was actuallyfound to increase two-fold when precipitatedwith exogenous glycogen, which may indicatethat the soluble fractions contain inhibitors ofthe enzyme.

Characteristics of glycogen synthetase.Glycogen synthetase of C. laurentii has thus farnot been described. In the following, we presentsome of the characteristics of the enzyme,employing the glycogen-bound synthetaseunless otherwise mentioned. It should bepointed out that what we are describing isessentially the properties of theglucose-6-phosphate-dependent enzyme. Wehave not been able to demonstrate conclusivelythe existence of a separate glucose-6-phos-phate-independent form, nor have we beenable to separate two such forms, as in thecase of S. cerevisiae (18) or Neurospora crassa(23).Of the nucleotide sugars tested, UDP-glucose

was the best glucosyl donor. TDP-glucose wasapproximately 50% as active as UDP-glucose.Glucose-6-phosphate stimulated the incor-poration of glucose from UDP-glucose aswell as from TDP-glucose to a similar extent.We have consistently observed stimulations ofbetween 4- and 16-fold by glucose-6-phosphateof glucosyl transfer from UDP-glucose toglycogen primer in different enzymepreparations. Although we think that thesedifferences might be due to the interconversionof two forms of the enzyme, we have so far notbeen able to demonstrate such a process. Therewas no incorporation of glucose fromCDP-glucose and GDP-glucose in the presenceor in the absence of glucose-6-phosphate andonly slight incorporation from ADP-glucose,which was not stimulated by glucose-6-phos-phate.As shown in Fig. 4, the incorporation of

"C-glucose from UDP- 4C-glucose into aglycogen primer varies linearly with theenzyme concentration.

Figure 5 shows the effect of pH on enzymeactivity. The highest activity was obtainedwith glycylglycine buffer at pH 7.5. Theactivities obtained with glycylglycine bufferwere much higher than those obtained with

632 J. BACTEROL.

on May 5, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

VOL. 113, 1973 BIOSYNTHESIS OF GLYCOGEN AND STARCH IN C. LAURENTII

0

cr

r cOE

c,

8e30CD

0015

0.009

0.003

C) O 20 40 60 8014 PROTEIN (jUg)

FIG. 4. Dependence of the glycogen synthetasereaction on enzyme concentration. The incubationmixtures were as described in Materials and Methodsexcept that the glucose-6-phosphate concentrationwas 15 mM and the UDP-glucose concentration was 6MM.

Uo XC)L

zo0p0.004-

- 0002

0 o1~4.0 50 60 70 8.0 90 100 110

pHFIG. 5. Effect of pH on glycogen synthetase. The

incubation mixtures contained 13 mg of rabbit liverglycogen/ml, 17 mm glucose-6-phosphate, 6 mM UDP-"4C-glucose, and 10 Mliters of enzyme (30 Mg of pro-tein) in a final reaction volume of 60 Mliters. 7he reac-tion mixtures were incubated in the indicated buffersat 25 C for 10 min; *, 0.17 M sodium acetate buffer;0, 0.17 M MES buffer; A, 0.17 M glycylglycine buf-fer; 0, 0.17 M Tris-hydrochloride buffer; 0, 0.17Mglycine-KOH buffer.

Tris-hydrochloride buffer under the sameexperimental conditions.

Figure 6 illustrates the linear relationshipbetween time of incubation and glucoseincorporation in the absence and presence ofglucose-6-phosphate. The reaction is linear forat least 30 min in the presence and absence ofglucose-6-phosphate. In this experiment,glucose-6-phosphate stimulated the activity15-fold.Figure 7 shows that the concentration of

glucose-6-phosphate required for half maximalactivity is 2.5 mm.Addition of glycogen in excess of 13 mg/ml

(see Materials and Methods) had no effect onthe glucose-6-phosphate-dependent or -inde-

pendent activity, indicating that the glycogenconcentration is saturating. Transfer of 14C-glucose from TDP- 4C-glucose and from ADP-"4C-glucose likewise was not stimulated by in-creasing the glycogen concentration.

Figure 8 shows the dependence of the rateof glycogen synthesis on UDP-glucose concen-tration in the presence of glucose-6-phosphate(16.6 mM). From the reciprocal plot, an ap-parent Km value of 1.0 mm was calculated.

0 60

50

40o0 GIc-6-Pa:o 30z

0

o 0 10 20 30 40 50

INCUBATION TIME (MIN)FIG. 6. Effect of time of incubation and glucose-

6-phosphate on the incorporation of "4C-glucose fromUDP-'4C-glucose. The assay was as described inMaterials and Methods, with the crude extracts usedas source of enzyme. The reaction mixtures con-tained 20 Mliters of enzyme (40 ug of protein).

0

w530

a20z

S 10

_0I-

I 0 0 0 0 1.0 0.2 0.4 0.6 0.8 1.0

Y[Gic-6-P] (nM)1'

FIG. 7. Reciprocal plot of velocity versus glucose-6-phosphate concentration. The assay was as de-scribed in Materials and Methods except that theamounts of glucose-6-phosphate were varied as in-dicated and the UDP-glucose concentration was 4.uM. The reaction mixtures contained 5 uliters ofenzyme (5 Mg of protein). The insert is a plot of Sversus V where S equals the concentration ofglucose-6-phosphate (mM) and Vis the amount of "C-glucoseincorporated (nmoles/10 min).

633

on May 5, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

SCHULTZ AND ANKEL

c; 14

. 12zS 1.00

W N\ 08

3 O 06

02A>

0

0 1 2

1/[UDPG] (mM)1

FIG. 8. Reciprocal plot of velocity versus UDP-glucose concentration in the presence of glucose-6-phosphate. The assay was as described underMaterials and Methods except that the amounts ofUDP-glucose were varied as indicated. The reactionmixtures contained 5 ;liters of enzyme (5 jg ofprotein).

Reaction product. The reaction productobtained from UDP-14C-glucose behavedidentically to that obtained from"IC-glucose-i-phosphate in the glycogenphosphorylase reaction described earlier (20).Acid hydrolysis or exhaustive treatment withamyloglucosidase quantitatively releasedglucose as the only radioactive product; ,B- anda-amylases released all of the radioactivity asmaltose or as glucose and maltose,respectively. Treatment of the "4C-maltosewith 1 unit of maltase resulted in theconversion of the radioactive material to aproduct, which upon chromatography migratedidentically to gluconic acid (solvent A). Thegluconic acid arises from the action of glucoseoxidase which is present in the crude maltasepreparation. Analyses were carried outessentially as described previously for theproduct of the glycogen phosphorylase reaction(20). These data indicate that the cryptococcalglycogen synthetase employed in these studiescatalyzes transfer of glucosyl units fromUDP-glucose to non-reducing terminals of aglycogen primer, forming new a-1, 4-linkages.It should be pointed out that, when theradioactive products obtained withTDP-14C-glucose and ADP-'4C-glucose as thesubstrate were treated with a- or,-amylase,again all of the radioactivity was released asmaltose or as glucose and maltose. It thereforeappears that similar products are formed from

all three sugar nucleotides.

DISCUSSIONThe data presented here demonstrate that

glycogen metabolism in C. Iaurentii proceedsaccording to the pathway shown in Fig. 9,which is analogous to that reported formammalian cells (26), S. cerevisiae (2, 3, 17),and N. crassa (21-24). Since such a sequence ofreactions represents a "short circuit," effectivecontrol mechanisms must regulate glycogensynthesis and breakdown in this organism:when glycogen synthesis prevails, phospho-rylase activity should be turned off, whereasunder circumstances where glycogen break-down is predominant synthetase activityshould be curtailed.Measurements carried out with cell

homogenates obtained during various phases ofgrowth of the organism close to neutral pHreveal the following: glycogen starts toaccumulate in late log-phase cells and begins tobe rapidly utilized at the beginning of thestationary phase. This occurs in spite of thefact that exogenous glucose is still present andis still taken up. Simultaneously, the specificactivity of glycogen synthetase in crude cellextracts is highest when the endogenousglycogen concentration peaks and falls offconcomitantly with glycogen utilization.Conversely, glycogen phosphorylase reaches itshighest specific activity approximately 3 hrafter the glycogen peak and remains high untilmost of the endogenous glycogen is utilized. Itis striking that the specific activity ofUDP-glucose pyrophosphorylase rapidly fallsoff at the end of log phase and shows only 60%of its peak activity when the specific activity ofglycogen synthetase is maximal. This might berelated to the fact that UDP-glucose is alsoinvolved in cell wall biosynthesis, a processthat slows down when the cells cease tomultiply exponentially. Glycogen synthetase

UDP-glucosePPi

UTP

glucose-i-Pv a

Pi

lUDP

. glcogen

FIG. 9. Pathway of glycogen metabolism in Cryp-tococcus laurentii. UDP, uridine diphosphate; UTP,uridine triphosphate; PPi, inorganic pyrophosphate;P,, inorganic orthophosphate; glucose-i-P, glucose-l-phosphate.

634 J. BACTJOL.

on May 5, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

VOL. 113, 1973 BIOSYNTHESIS OF GLYCOGEN AND STARCH IN C. LAURENTII

and UDP-glucose pyrophosphorylase activitiesapproach zero when the exogenous glucosebecomes exhausted, whereas glycogenphosphorylase remains relatively constant aslong as there is glycogen in the cells.As shown in Fig. 1B, UDP-glucose

pyrophosphorylase activity in crude cellhomogenates is always more than an order ofmagnitude higher than glycogen synthetase. AsUDP-glucose is also converted to precursors ofother polysaccharides of C. laurentii(UDP-glucuronic acid, UDP-xylose [4, 8]), itssynthesis is not a step unique in glycogensynthesis, and thus control of UDP-glucosepyrophosphorylase activity at the substratelevel would be an unlikely regulatingmechanism of glycogen synthesis in C.Iaurentii.Glycogen synthetase at all levels of growth,

when measured in crude homogenates afterlimited manipulation of the cells attemperatures close to zero, shows profoundactivation by glucose-6-phosphate, indicatingthat most of the enzyme is alwaysglucose-6-phosphate dependent. We have,however, found that the degree of activation byglucose-6-phosphate in a given preparation isgenerally less (4 to 10 times) in the bottomlayer than in crude homogenates (10 to 16times). It should be pointed out that theglycogen synthetase in crude homogenates wasassayed approximately 1 hr after harvesting ofthe cells and approximately 15 min afterbreaking of the cells, whereas it took more than24 hr from the time the cells were broken untilthe bottom layer was assayed for glycogensynthetase activity. The differences in thiseffect of glucose-6-phosphate could be due to aslow transformation of the enzyme to adifferent form similar to the D - Itransformation in muscle, which has also beenshown for other fungi (23) and for yeast (18).However, in an experiment in which we haveassayed crude cell homogenates over a period oftime in the presence and in the absence ofglucose-6-phosphate, we were unable to show achange in the ratio of both activities for morethan 24 hr. The fact that the total activitymeasured after precipitation in the presence ofexcess cryptococcal glycogen is twice theactivity measured in the crude extract points tothe presence of an endogenous inhibitor(s) incrude homogenates. The presence of inhibitorymaterial in crude homogenates may provide apossible explanation for the observeddifferences in stimulation of glycogensynthetase activity. Rothman and Cabib (16)have shown that the relative stimulation of S.

cerevisiae glycogen synthetase by glucose-6-phosphate is much larger in the presenceof inhibitors. Separation of the enzyme frominhibitors by centrifugation should there-fore result in decreased activation by glu-cose-6-phosphate, which in fact has beenobserved. From the data in Fig. 1A, it appearsthat glucose uptake towards the end of logphase accelerates. This fact might lead to anaccumulation of glucose-6-phosphate in thecells and thereby increase glycogen synthesis,while at the same time glycogen phosphorylasebecomes inhibited by glucose-6-phosphate.Such inhibition has in fact been demonstrated(20).Electron microscope and chemical evidence

reveals that glycogen granules occur also in C.laurentii cells grown in acidic medium. Theacidic conditions, however, appear to impaircapsule and extracellular acidic polysaccharideformation. The relationship of theseobservations to starch production, however,remains obscure. It is clear that the findingsreported here do not provide an explanation forthe biosynthetic mechanism by which starch isformed at low pH. Failure of cryptococcal cellsto produce starch when grown close to neutralpH might be due to the fact that acidicconditions result in the induction of an entirelynew enzyme system. Another possibility mightbe that during growth at neutral pH thestarch-synthesizing system is inhibited bysome metabolite(s) that is not present ateffective concentrations during growth at lowpH. It appears certain, however, that C.Iaurentii cells grown in acidic medium are ableto synthesize glycogen. Extracellular starchproduction, therefore, cannot be theconsequence of an impairment of glycogensynthesis during growth at acidic pH.

ACKNOWLEDGMENTSThe skillful technical assistance of Johnnie Fendrix, Mary

Aumann, and Elizabeth Cavey is gratefully acknowledged.UDP-glucose pyrophosphorylase assays were kindly

performed by Utpalendu Maitra of the BiochemistryDepartment, The Medical College of Wisconsin, for whichwe express our appreciation. We also thank J. C. Garancisfor carrying out the electron microscopy and for his valuableadvice.

This work was supported by National Science Foundationgrant GB-7348 and by Public Health Service grantGM-40916 from the National Institute of General MedicalSciences. J.C.S. was a National Institutes of Healthpredoctoral fellow. H.A. is a recipient of a CareerDevelopment Award from the National Institute of GeneralMedical Sciences.

LITERATURE CITED1. Abercrombie, M. J., J. K. N. Jones, M. V. Lock, M. B.

Perry, and R. J. Stoodley. 1960. The polysaccharides

635

on May 5, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

SCHULTZ AND ANKEL

of Ciyptococcus laurentii (NRRL-Y-1401). Part 1.Can. J. Chem. 38: 1617-1623.

2. Algranati, I. D., and E. Cabib. 1960. The synthesis ofglycogen in yeast. Biochim. Biophys. Acta 43:141-142.

3. Algranati, I. D., and E. Cabib. 1962. Uridinediphosphate D-glucose-glycogen glucosyltransferasefrom yeast. J. Biol. Chem. 237:1007-1013.

4. Ankel, H., and D. S. Feingold. 1966. Biosynthesis ofuridine diphosphate D-xylose. II. Uridine diphosphateD-glucuronate carboxy-lyase of Cryptococcuslaurentii. Biochemistry 5:182-189.

5. Aschner, M., and A. Cury. 1951. Starch production inthe genus 7richosporon. J. Bacteriol. 62:350-352.

6. Aschner, M., J. Mager, and J. Leibowitz. 1945.Production of extracellular starch in cultures ofcapsulated yeasts. Nature (London) 156:295.

7. Chester, V. E., and M. J. Byrne. 1968. Carbohydratecomposition and UDP-glucose concentration innormal yeast and a mutant deficient in glycogen.Arch. Biochem. Biophys. 127:556-562.

8. Cohen, A., and D. S. Feingold. 1967. Uridinediphosphate D-xylose. Acceptor xylosyltransferase ofCryptococcus laurentii. Biochemistry 6:2933-2939.

9. Jeanes, A., J. E. Pittsley, and P. R. Watson. 1964.Extracellular polysaccharide produced from glucoseby Cryptococcus laurentii var. flavescensNRRL-Y-1401: chemical and physical characteriza-tion. J. Appl. Polymer Sci. 8:2775-2787.

10. Kooiman, P. 1963. The chemical structure of the extra-cellular "starch" produced by Cryptococcus albidusand C. laurentii var. flavescens. Antonie van Leeu-wenhoek J. Microbiol. Serol. 29:169-176.

11. Krisman, C. R. 1962. A method for the colorimetricestimation of glycogen with iodine. Anal. Biochem.4:17-23.

12. Lodder, J., and N. J. W. Kreger-Van Rij. 1952. Theyeasts, p. 371-396. Interscience Publishers, Inc., NewYork.

13. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J.Randall. 1951. Protein measurement with the Folinphenol reagent. J. Biol. Chem. 193:265-275.

14. Mager, J. 1947. Studies on the polysaccharides ofcapsulated yeasts. Biochem. J. 41:603-609.

15. Polakis, E. S., and W. Bartley. 1965. Changes in theenzyme activities of Saccharomyces cerevisiae duringaerobic growth on different carbon sources. Biochem.J. 97:284-297.

16. Rothman, L. B., and E. Cabib. 1967. Allostericproperties of yeast glycogen synthetase. II. The effectof pH on inhibition and its physiological implications.Biochemistry 6:2107-2112.

17. Rothman, L. B., and E. Cabib. 1969. Regulation ofglycogen synthesis in the intact yeast cell.Biochemistry 8:3332-3341.

18. Rothman-Denes, L. B., and E. Cabib. 1970. Two formsof yeast glycogen synthetase and their role in glycogenaccumulation. Proc. Nat. Acad. Sci. U.S.A.66:967-974.

19. Ruinen, J., M. H. Deinema, and C. Van Der Scheer.1968. Cellular and extracellular structures inCryptococcus laurentii and Rhodotorula species. Can.J. Microbiol. 14:1133-1137.

20. Schultz, J. C., and H. Ankel. 1970. Glycogen-boundphosphorylase in Cryptococcus laurentii. Biochim.Biophys. Acta 215:39-51.

21. Shepherd, D., S. Rosenthal, G. T. Lundblad, and I. H.Segel. 1969. Neurospora crassa glycogenphosphorylase: characterization and kinetics via a

new radiochemical assay for phosphorolysis. Arch.Biochem. Biophys. 135:334-340.

22. Shepherd, D., and 1. H. Segel. 1969. Glycogenphosphorylase of Neurospora crassa. Arch. Biochem.Biophys. 131:609-620.

23. Tellez-Inon, M. T., H. Terenzi, and H. N. Torres. 1969.Interconvertible forms of glycogen synthetase inNeurospora crassa. Biochim. Biophys. Acta191:765-768.

24. Traut, R. R., and F. Lipmann. 1963. Activation ofglycogen synthetase by glucose-6-phosphate. J. Biol.Chem. 238:1213-1221.

25. Trevelyan, W. E., D. P. Procter, and J. S. Harrison.1950. Detection of sugars on paper chromatograms.Nature (London) 166:444-445.

26. Villar-Palasi, C., and J. Lamer. 1960. Levels of activityof the enzymes of the glycogen cycle in rat tissues.Arch. Biochem. Biophys. 86:270-273.

636 J. BACTROL.

on May 5, 2020 by guest

http://jb.asm.org/

Dow

nloaded from