Vol. 11, No. 5 Printed in U.S.A.
INFECTION AND IMMUNITY, May 1975, p. 1014-1023 Copyright 0 1975 American Society for Microbiology
Factors Affecting Filamentation in Candida albicans: Relationship of the Uptake and Distribution of Proline to Morphogenesis G. A. LAND, W. C. McDONALD,2 R. L. STJERNHOLM, AND L. FRIEDMAN* Departments of Microbiology and Immunology,* Biology, and Biochemistry, Tulane University School of Medicine, New Orleans, Louisana 70112 Received for publication 12 December 1974
When glucose was present in high concentrations, Candida albicans formed filaments in a phosphate-buffered medium, regardless of the nitrogen source. In lower concentrations of glucose, filamentation occurred only when various members of the glutamate, succinyl, or acetoacetyl-coenzyme A families of amino acids were used as sole nitrogen sources. Yeast morphology could be maintained either by replacing the amino acids in the medium with ammonium chloride or by making the medium high in phosphate or biotin. Studies using [U- 4C]proline indicated that proline was catabolized in a manner consistent with the generation of increased cellular reducing potential and that the proline label entered into the Kreb's cycle. A reduction in Kreb's cycle activity was evidenced by an initial increase and then a rapid drop of the total organic acid content of the cells as well as in specific Kreb's cycle intermediates. Filamentation under conditions of low phosphate, high glucose, and increased cellular reduction potential, accompanied by a decrease in Kreb's cycle activity, suggests that morphogenesis in C. albicans is correlated with a Crabtree-like effect, i.e., repression of mitochondrial activity. In the presence of certain body fluids in vitro, or in tissue in vivo, the yeast form of Candida albicans will convert to a filamentous form. This was first demonstrated in blood and serum acquired from debilitated individuals (26). Filamentation in combination with certain host factors may be an important factor in pathogenesis (2, 14, 32), consequently a better understanding of the mechanism controlling the yeast-mycelial (Y-M) balance might lead to a means of controlling systematic disease. Wickerham and Rettger, in 1939, published a monograph on filamentation in Candida (34), and since then there have been many attempts to determine what controls the Y-M balance in Candida. In the majority of in vitro experiments, summarized in recent reviews (3-5, 23), either serum or various complex artificial media were used to induce filamentation. Nickerson was one of the first to successfully provide a chemically defined environment conducive to filamentation (22). He postulated that the enzymatic transfer of reducing power gener-
ated within the mitochondria to the yeast cell wall via a disulfhydryl reductase controlled morphogenesis. He suggested a medium high in cysteine (and thus reduced) for the maintenance of yeast morphology, thereby providing control of the amount of reduced sulfhydryl groups within the cell wall. Widra (35) expanded Nickerson's work and proposed that phosphate and magnesium played a role in determining yeast morphology of Candida in culture, presumably by fulfilling requirements of the reductase enzyme for activity. Mardon et al. (20) found that the Y-M balance in their Candida strains was governed by the relationship of the external CO2-O2 ratio to internal pools of S-adenosylmethionine of yeasts when grown in a minimal salts-glucose medium. In 1971, Dabrowa (Proc. Int. Soc. Human and Animal Mycol., p. 51-52) reported the induction of filaments in a simple phosphate-buffered medium containing proline and an unspecified amount of glucose and biotin. The results of the preceding investigators suggest that it should be possible to define the 'Present address: Department of Microbiology, Wadley cultural conditions necessary to maintain each Institutes of Molecular Medicine, Dallas, Tex. 75235. 2Present address: Department of Biology, University of form of C. albicans in a pure state. The purpose of the present research was to attempt to define Texas, Arlington, Tex. 1014
VOL. 11, 1975
FILAMENTATION IN C. ALBICANS
stable cultural conditions for maintaining Candida in either the yeast or filamentous form and to determine the factors responsible for changes in Candida morphology. MATERIALS AND METHODS Cultural conditions. Twenty-eight' isolates of C. albicans, all of which conformed to all biochemical and morphological criteria for this species, were screened for their ability to form filaments from blastospores (yeasts) in a proline basal medium described below. All isolates formed filaments to some extent, but isolate 5865, from a patient with systemic candidiasis, was selected for this study from among those which profusely formed filaments. Transition from yeast to filamentous growth was apparent in the first generation (2 h) and was maximal by 6 h, whereupon a return to yeast-like growth began. We felt, however, that the stimulus for filamentation occurred upon placing the yeasts into the proline medium. Thus, metabolic events which control the onset of filamentation would occur within the first hour of growth, but the morphological expression of those events would not appear until after the cells attempted to divide, namely, after the 2-h incubation of the proline medium. Accordingly, intervals of incubation varied with experimental design. The basic medium in which various nitrogen and carbon sources were tested consisted of 0.008 M phosphate-buffered saline (PBS) (12) and 250 ug of biotin per liter. The latter has been shown to be a growth requirement for C. albicans (15). On the basis of these experiments, summarized in Table 1 and Fig. 2, the following basal media, prepared in PBS, were employed for most of the studies, unless otherwise specified: a proline medium containing 250 Ag of biotin per liter, 10-2 M glucose, and 10-2 M proline (for filamentous growth); or an NH4Cl medium (for yeast growth) prepared in the same manner but with 10-2 M NH4Cl instead of proline. These concentrations of glucose and biotin were not limiting inasmuch as increasing the concentration of either nutrient did not result in increased growth. Synchronized inocula for incubation in the experimental media were prepared in the following manner. Cells were first starved for 6 h at 37 C in PBS while being shaken at 120 gyrations per min (New Brunswick Scientific Co.). Synchrony was then attained by a modification of the method of Dabrowa (9). This involved allowing the suspension to settle for 30 min, followed by centrifugation of the supernatant through 15% mannitol for 5 min at 100 x g. The synchronized inoculum (Fig. 1) was washed three times in warm (37 C) PBS and placed, at appropriate cell densities (see below), into the experimental media. Cell density and percentage of filamentation were determined by triplicate counting of each culture in a hemacytometer; for each experiment there were three cultures, and each experiment was repeated three times. For example, one curve in Fig. 2 represents the averaged data of nine cultures, cultivated three at a time, and one observation point on that curve represents the average of 27 counts in a hemacytometer.
For a single count, the fungi (filament or yeast) in 4 mm2 were tallied. Usually a total of 100 to 400 fungal units were counted as follows: (i) a single yeast or a yeast and a small bud were counted as a single unit; (ii) two large yeasts together were counted as two separate units; (iii) pseudohyphae, a yeast cell with a germ tube, or true hyphae were considered filaments; (iv) as much as was possible each cell within a filament was counted as a single cell; (v) when filamentous growth overran the culture (usually after 10 h), each filamentous ball was counted as a single unit. Dry weights were determined by one of two methods: (i) washed yeasts were dried and weighed after filtration onto preweighed 0.45-Mm membrane filters (Millipore Corp.); or (ii) by a modification of the replicate culture method of Hurley (17). The modification was that of using preweighed, siliconized, soft plastic-topped scintillation vials (New England Nuclear, Boston, Mass.) rather than 4-oz (0.12liter) prescription bottles. Cultures from several experiments were accumulated and stored (at 4 C to stop growth). The containers were coded and randomized so that the individual assaying the growth was not prejudiced by a knowledge of the batch number or cultural conditions at the time of assay. Uptake and distribution of proline. L- [U- 14C Jproline (specific activity, 198 uCi/mmol), was placed into siliconized scintillation vials containing 10 ml of nitrogen-free PBS-biotin-glucose medium, with a final concentration of 10-4 M proline (3.0 uCi/ml). Starved yeasts were then suspended in the labeled proline medium at a density of 106 cells/ml and incubated at 37 C and 120 gyrations per min for 5 min. Although the concentration of proline in the labeled medium was less than that usually used, the amount was still adequate to stimulate significant filamentation (Fig. 2). Furthermore, as the cells were in the low concentration of proline for 5 min only and then immediately suspended in medium containing 10-2 M proline, the labeled cells formed filaments at a maximal rate. After 5 min, the labeled cells were washed three times in warm PBS containing 1,000 U of Pen-Strep (Grand Island Biological Co., Grand Island, N.Y.) per ml and resuspended in unlabeled proline medium. At 0, 15, 30, 45, and 60 min of incubation at 37 C, cells were harvested by centrifugation at 11,000 x g for 5 min at 4 C (8). The pH of the supernatant was determined, and the supernatant was then discarded. Each lot of pelleted cells was washed three times in cold (4 C) distilled water and resuspended in 10 ml of PBS; 5 ml of each suspension was filtered onto a preweighed 0.45-Mm membrane filter (Millipore Corp.). The filters were dried in vacuo at 80 C ovemight, weighed, placed in 15 ml of Aquasol (New England Nuclear), and counted in a Beckman LS230 liquid scintillation system. The remaining 5 ml of each of the above suspensions was disrupted in a French pressure cell at 26,000 lb/in2 (Aminco, Silver Spring, Md.), and the lysates were centrifuged at 18,000 x g for 60 min. Disruption of both yeasts and filaments was 99.1% i 0.3, as determined by direct counts of intact cells in a hemacytometer. The supernatants were decanted and
LAND ET AL.
each was adjusted to 25 ml before placement on a Dowex 50 (Sigma Chemical Co., St. Louis, Mo.; 100 to 200 mesh) column (1 by 20 cm). Each time, the column was charged previously with 2 N HCl and brought to neutrality by repeated washings with distilled water. The organic acids contained in the supernatants were eluted from the column by five 10-ml amounts of distilled water and pooled. After removal of the organic acids, the amino acids were collected by eluting with five 10-ml volumes of 3 N NH4OH. Both pools (organic acids and amino acids) were evaporated in vacuo at 80 C, weighed in plastic petri dishes (60 by 15 mm), and counted as above. The pellets remaining after the centrifugation at 18,000 x g were extracted for total ribonucleic acid (RNA), deoxyribonucleic acid (DNA), and protein by the method of Dabrowa (8). RNA was determined by the orcinol method using torula yeast RNA (Sigma Chemical Co., grade IV) as a standard, total DNA was determined by the diphenylamine method with salmon sperm DNA (Calbiochem, La Jolla, Calif.) as a standard, and total protein was determined by the method of Lowry et al. against a bovine serum albumin standard (29). Recovery of the 14C label by these extraction procedures varied between 80 to 86% throughout the repetitions of these experiments. Isolation of mitochondria. Mitochondria were isolated as described by Yamaguchi et al. (36) from several replicate cultures of C. albicans grown under the proline pulse-chase conditions previously mentioned. Fifteen to 30 mg of protein from sand-ground cells was layered onto a discontinuous sucrose gradient (27). The gradients were centrifuged in a Spinco SW26.1 rotor at 25,000 rpm for 20 h, and 0.5-ml fractions were collected. Parallel fractions from several gradients were pooled and assayed for RNA, DNA, and protein content, as well as for respiratory activity. Succinoxidase and nicotinamide adenine dinucleotide, reduced form (NADH) oxidase served as respiratory markers and were measured polarigraphically by a Clark oxygen electrode on 2 mg of protein from each fraction pool (36). Identification of pool components. After the amino and organic acid pools were separated by column chromatography, dried, and weighed, they were rehydrated to a known concentration in distilled water. One-hundred micrograms of either pool was spotted onto thin-layer chromatography sheets (6064 cellulose; Eastman Chromagram), and the sheets were developed in a butanol-acetate-water solvent (120:30:50). The amino acids were detected by the chromagen method of Moffat and Lytle (21). The organic acids, being nonreactive with ninhydrin, were detected with 0.4% ethanolic bromocresol purple at pH 5 (25). Additionally, amino acids and organic acids were chromatographed side by side with labeled standards (U-14C) and followed by 14C scanning (Packard radiochromatogram scanner). Spots of radioactivity located by the radioscan were cut out, placed in 15 ml of Aquasol, and counted. Carbohydrates were separated by a modification of the thin-layer method of Kraeger and Hamilton (19), employing 0.1% KH2PO4-saturated silica gel sheets (Eastman Chromagram, 6061 cellulose) and were detected by their reaction with 1% aniline-1% di-
phenylamine in acetone and phosphoric acid (30) (International Chemical and Nuclear Corp., Cleveland, Ohio). Individual compounds were located by radio-scanning, and the spots were cut out and counted as described above. Additionally, carbohydrate pools were chromatographed side by side with carbohydrate U- 4C standards.
RESULTS Relationship of nitrogen source to filamentation. The influence of the nitrogen source in the glucose-biotin-PBS medium on Candida morphology was examined by substituting each of the 23 amino acids described as sole nitrogen sources by Dabrowa (Table 1). With few exceptions, the results may be summarized as follows. (i) Amino acids entering metabolism by conversion to glutamate promoted more filamentation, on the average, than the other groups of amino acids tested; (ii) amino acids entering the Kreb's cycle via succinate, succinyl-coenzyme A, or oxaloacetate supported filamentation to a lesser, but significant, extent; (iii) amino acids catabolized to fumarate, acetate, or pyruvate did not significantly influence the formation of filaments; (iv) ammonium chloride in the glucose-biotin buffer supported only yeast growth; and (v) blocking of the active groups on proline (the ring-bound nitrogen or the free carboxyl group) by either ring substitution, N-acyl derivatives, or carboxy ethers did not promote filament production. Even though a number of amino acids stimulated filament production equally as well as the glutamate family at 6 h, the above generalizations between amino acid groups were more evident at 24 h, when the glutamate family of amino acids sustained more filamentation than the other groups of amino acids. The pH of the cultures grown in various amino acids did not differ significantly during the 6 h of incubation, by which time the yeasts should have been committed to filament production if such were going to occur. In fact, it was our experience that filamentation was always easily detectable within 2 h. The ratio of dry weight of culture to dry weight of inoculum suggested that at least two generations of cells occurred during that 6 h of incubation and that the cultures were synchronous through two generations of cells (Fig. 1). High concentrations of phosphate in the proline medium depressed filamentation (Fig. 2), whereas low concentrations ( 10- 1 M) of glucose supported filamentation in the NH4C1 medium. Uptake and distribution of [U-14C]proline.
VOL. 11, 1975
FILAMENTATION IN C. ALBICANS
TABLE 1. Growth characteristics of C. albicans incubated with uarious amino acids for 6 h in a glucose-biotin-phosphate buffer (pH 7.2) Nitrogen source
Proline Arginine Alanine Histidine
Thioproline Glutamate Ornithine Asparagine Aspartate Isoleucine Methionine Valine Lysine Tryptophan Phenylalanine Leucine Threonine Tyrosine Serine Glycine Cysteine Cystine p-NO2-phenyl proline Benzyl-proline N-acetyl proline L-Azetidine-2 carboxylic acid (NH4)2SO4 NH4C1
Route of entry into metabolism
Increase in growthh
pH of culture medium at 6 h
Glutamate a-ketoglutarate Glutamate a-ketoglutarate Glutamate a-ketoglutarate Glutamate a-ketoglutarate Glutamate a-ketoglutarate Glutamate a-ketoglutarate Glutamate a-ketoglutarate Oxaloacetate Oxaloacetate Succinyl CoAf Succinyl CoA( Succinyl CoA( Acetoacetyl-CoA Acetoacetyl-CoA Acetoacetyl-CoA Acetoacetyl-CoA Acetoacetyl-CoA Acetoacetyl-CoA Pyruvate-acetate Pyruvate-acetate Pyruvate-acetate Pyruvate-acetate Transamination Transamination Transamination Transamination
93 ± 5 90 ± 4 89 ± 6 68 ± 3.6 51 ± 4 38 ± 0.33 14 + 0.67 44 ± 2.1 40 ± 1 70 1 16 ± 0.33 20 + 0 70 + 3 42 + 2 40 + 1 36 + 0.33 31 ± 0.33 0 20 ± 0 8±1 3± 1 0 2 ± 0.67 0 0 0
6.3 ± 0.06 4.6 ± 0.05 4.1 ± 0.065 4.5 ± 0.07 6.3 ± 0.04 6.0 + 0.051 6.1 + 0.08 6.1 + 0.056 5.5 + 0.06 4.3 ± 0.075 5.2 + 0.06 5.5 ± 0.05 4.0 + 0.03 5.9 + 0.045 6.2 + 0.066 4.6 + 0.08 5.6 ± 0.02
5.1 ± 0.056 4.0 ± 0.03
7.0 8.0 8.0 7.4 7.0 7.2 6.8 7.6 7.6 8.0 8.0 8.0 8.0 7.6 7.8 8.0 7.8 8.0 8.0 8.0 7.6 7.8 7.2 7.2 7.2 7.0
28 ± 0.67 0
6.7 ± 0.02 6.8 ± 0.04
6.1 ± 0.04 5.5 ± 0.05
5.7 ± 0.03 5.5 ± 0.07 5.0 ± 0.04 4.2 ± 0.06 4.9 ± 0.08
Standard error of the mean of three experiments. h Ratio of dry weight after 6 h of incubation to dry weight of inoculum ± standard error of the mean of three experiments. ' CoA, Coenzyme A. a
The distribution of label in the various classes of compounds varied with time (Fig. 3). Immediately after the pulse, the bulk of the radioactivity was incorporated into free amino acids. A peak of radioactivity in organic acids was reached at 15 min and then declined to a low level. As the level of radioactivity declined in the amino acid pools, a steady increase was noted in the KOH-precipitable fraction (RNA) and the NaOH-precipitable fraction (protein). After 60 min of incubation, the RNA fraction of the pellet of lysed cells contained twice as much label as the protein. The trichloroacetic acidprecipitable fraction (DNA) did not contain significant radioactivity until 60 min, approximately one-half the first generation time of Candida in this medium. Only seven labeled amino acids were separated by thin-layer chromatography from the extract of lysed filaments, four of which are presented in Fig. 4. Proline decreased in activity after the first 15 min of incubation. Glutamate,
histidine, and aspartate showed the greatest incorporation of the seven amino acids isolated, all reaching a peak within 30 min. Glutamate and histidine were depressed at 45 min to approximately one-sixth of their peak activity. Aspartate, on the other hand, declined less rapidly and one-third of the "4C activity still remained after 60 min. Significant amounts of 14C were also found in the amino acids leucine, isoleucine, and lysine. Lysates from unlabeled yeast controls, grown in the NH4Cl medium, produced only four chromogenically separable amino acids: cysteine, methionine, leucine, and alanine. Several organic acids separated by thin-layer chromatography from lysates obtained from filaments also had acquired the 14C label (Fig. 5). Alpha-ketoglutarate reached peak activity within 15 min after the pulse and declined by approximately 60% within 30 min. Succinate activity reached a maximum at 30 min and declined to 30% of maximum activity by 60 min
LAND ET AL.
bulk of 14C radioactivity corresponded with a fraction high in respiratory activity, presumably mitochondria (Fig. 6). Analysis of the macromolecular components within pooled fractions of protein peaks developed on the gradient showed, as in extracted cells, a preferential labeling of RNA over protein and DNA (Table 2). The peak containing high respiratory activity (Fig. 6, peak 2) also contained significantly more "4C label than peaks 1 and 3. Glucosamine, indicative of cell wall material, was present in significant amounts in peak 1, with a trace in peak 3 and none in peak 2.
FIG. 1. The synchronous growth of C. albicans under conditions of filament (a) or yeast (0) formation, as determined by dry weights.
past pulse. Labeled citrate and malate were present in significant amounts during the first 30 min after the pulse, but could no longer be detected by 45 min. Six Kreb's cycle intermediates (citrate, a-ketoglutarate, succinate, malate, fumarate, and oxaloacetate) were separated by thin-layer chromatography from yeast controls grown in NH4Cl medium. Spots formed by these intermediates did not appear to change in intensity or weight during the 60 min of incubation. Large concentrations of acetate and pyruvate were found in the yeast controls but not in the proline-supplemented filamentous cultures. No "4C was detected in the carbohydrate pools. There was, however, a major change in the carbohydrate pools of yeasts undergoing germination to filaments. Carboyhydrate pools of yeasts consisted mainly of glucose, mannose, and galactose. The carbohydrate pool of the filaments showed only trace amounts of galactose and mannose, but contained high concentrations of glucose and glucosamine. Cellular location of proline uptake and distribution. Separation of disrupted 60-min cells by centrifuging them through a discontinuous sucrose gradient demonstrated that the
DISCUSSION Dabrowa suggested that proline is the only amino acid capable of stimulating significant filamentation in C. albicans (Proc. Int. Soc. Human and Animal Mycol., p. 51-52, 1971). The precise cultural conditions of her experiments and the physiological state of the fungus were not described. This necessitated development of our own culture system, using her ideas as a guideline. Since large precursor pools have been reported to occur in other yeasts (13), we decided to starve the yeasts prior to metabolic studies to avoid the influence of such pools on our results. It was also our experience that filamentation was depressed when amino acids capable of stimulating filamentation were used in conjunction with ammonium chloride. The depressing effect of the ammonium ion on
/ MOLARITY OF
FIG. 2. Effects of variouts nutrients in stimulating filamentation in normal PBS. Symbols: (X3C), decreasing molarity of proline with 10-2 M glucose and 250 gg of biotin per liter in 0.008 M phosphate buffer; ( i--), decreasing molarity of glucose with 10-2 M NH4C1 and 250 ,gg of biotin per liter in 0.008 M phosphate buffer; (D.:.t, decreasing molarity of inorganic phosphate with 10-2 M proline, 10-2 M glucose, and 250 Aig of biotin per liter.
VOL. 11, 1975
FILAMENTATION IN C. ALBICANS
tion of "C into the Kreb's cycle intermediates) was related to the initiation of filamentation. Normal proline catabolism seems to be a re4 quirement for filamentation, since chemical 4.0 D JP alteration of the active groups of proline, which presumably alters the manner in which the w molecule is catabolized, prevented filament formation. Synder et al. have also suggested that substrate-generated NADH may be related to filament formation in Candida (J. W. Snyam *:. r der, L. J. Bir, and R. C. LaChapelle, Abstr. e~~ Annu. Meet. Am. Soc. Microbiol. 1973, Mm 36, Jx ~ ~ *:: f p. 136). Our results are also seemingly in contrast with other investigators employing a glucosesalts-biotin medium for growth (20, 24). We feel, however, that it would be inaccurate to compare our data with these groups, for the following reasons. (i) Nickerson and Chung (24) TIME used a filamentous variant, whereas Mardon et tMINUTES AFTER 5 MIN PULSE] al. (20) used a biochemical variant which ferFIG. 3. Uptake and distribution of L-[U-'4C]pro- ments sucrose and is unable to form either line by C. albicans grown in PBS containing 10-2 M chlamydospores on corn meal agar or germ glucose and 250 zg of biotin per liter. Each curve tubes in serum. The strain used in this study, as represents the average of three separate experiments. stated above, is a systemic isolate which conDifferences between the values at 0 and 30 min for protein, RNA, whole cells, and amino acids were highly significant, P < 0.01; differences between the values at 0 and 15 min for organic acids and 0 and 60 min for DNA were equally significant. Symbols: (#.... .), counts per minute per milligram of dried yeast; (IL t), counts per minute per milligram of dried amino acids; (X:.:::::X), counts per minute per milligram of dried organic acids; (O-s) counts per minute per milligram of RNA; (mhoini), counts per minute per milligram of dried protein; (_), counts per minute per milligram of dried DNA.
amino acid transport has been reported to occur in other fungi (1, 11, 28). Derepression of this effect occurs by either starving the fungus for 3 h or by replacing the ammonium ions in the medium with proline. We have found, in addition to proline, that several other amino acids, chiefly of the glutamate family, were capable of stimulating filamentation. Those amino acids, catabolized through succinyl-coenzyme A or acetoacetyl-coenzyme A, also supported significant filamentation, although to a lesser degree than that seen after stimulation by the glutamate family of amino acids (Table 1). The only common feature of the amino acids supporting the best filamentation was that all ultimately caused generation of an increased amount of reducing potential within the cell, as a result of their entry into the Kreb's cycle. For example, perhaps the increased reducing power generated by the catabolism of proline through the Kreb's cycle (as evidenced by the incorpora-
2 20~~~~~~O *.
* *0 °-~~~~~*
TIME AFTER 5 MIN
FIG. 4. Incorporation of a labeled carbon skeleton from [U-14Cjproline into specific amino acids by C. albicans. Data was obtained from the average of three thin-layer plates from each of three experiments. The increase in label between 0 and 30 min for all amino acids was highly significant, P < 0.01. Symbols:
(X---- X), glutamic acid; *3-m 4), aspartic acid; (ilAA), histid ine; tD=m, proline.
LAND ET AL.
forms to all biochemical and morphological criteria used by the clinical laboratory as diagnostic of C. albicans. (ii) Whereas it is true that the basic experimental design of both groups employed a glucose-salts-vitamin medium, D there are subtle differences which we believe our studies show to be important in Candida mor: 15 phology. In the study by Nickerson and Chung (24) 0.1% yeast extract provided vitamins as well as phosphate and other nutrients for growth. The yeast extract in their medium : / \would also provide carbon and nitrogen, making IZ 100 the total carbon or energy source at least five times that used in our study. Mardon et al. (20) used 10 times more glucose in their medium 0 59 >_______t than is present in ours. In addition, they altered the C02-02 ratio of the gaseous environment of E0 their fungal cultures, which has been shown to 0 ui L have an important effect on the morphology of many fungi (16). 4W5 _10 Nickerson postulated that Y-M conversion TIME (MINUTES AFTER 5 MIN PULSE) resulted electronphosphate, nicotinamide transfer to reduced form adenine when dinucleotide FIG. 5. Incorporation of a labeled carbon skeleton (NADPH)-dependent flavoprotein (disulfhyfrom [U-_4C]proline into specific Kreb's cycle inter- dryl reductase was blocked at a site in "which a
mediates by C. albicans. Data were obtained from the average of three thin-layer plates from each of these separate experiments. The increase in label from 0 to 30 min for succinate, and the decrease from 0 to 60 min for citrate and malate, was highly significant, P
< 0.01. Symbols: (D-.-0), a-ketoglutaric acid; () ), succinic acid; (::::::A), citric acid; (E)--s), malic acid.
cellular reaction chain is coupled yielding carbohydrate
amino acid metabolism are involved in Candida morphogenesis. The marked increase and abrupt decrease in Kreb's cycle activity of
(22, 23). Mardon et al. (20) also agree that energy transfer and, more importantly, sulfur
1.5 0) o C)0.3 ~~~~~~~~~~~~~1.5 o~~~~~ 00.3 ~~~~~~~u 2 3
6gnd. u 0.1
for 60 min with cold proline. Symbols: (w), absorbance at 280 nm; (O), incorporation of [U_'4C]proline into cellular fractions (counts per minute per milligram of protein:(Av), micromoles of succinate consumed per minute per milligram of protein:(-), micromoles of NADH consumed per minute per milligram of protein.
FILAMENTATION IN C. ALBICANS
VOL. 11, 1975
TABLE 2. Distribution of 14C label in various fractions of disrupted C. albicans after 1 h of incubationa Counts/min | Fractions
Whole cells RNA DNA Protein Glucosamine5 Band' (fractions 2-5) RNA DNA Ppotein Glucosamine Band2 (fractions 17-26) RNA DNA Protein Glucosamine Band3 (fractions 47-51) RNA DNA Protein Glucosamine
per mg (dry wt)
6,240± 105 2,965± 53 376± 8 2,297+ 20
Presence Rof glu-
Rcvrdcosamine 100 45 6 37 4+
1,135± 21 113± 11 32± 11 606+ 9
100 10 3
3,455± 1,865± 155± 1,131 ±
76 36 15
1,522± 21 161±+ 12 55± 7 535+ 15
a Isolation, determination, and counting of macromolecules were done according to protocols stated in Material and Methods under uptake and distribution of proline. IUsed as a marker to denote the cell wall rich fractions. Glucosamine was determined by the thinlayer procedure described in Materials and Methods.
Candida during proline catabolism (germination) also suggests that a block in electron flow affects cell morphology. Since lowered Kreb's cycle activity is concomitant with a reduction in electron transfer, there would be fewer electrons available for maintaining the NADPH-disulfhydryl reductase system. Nickerson's data suggests that the disulfhydryl reductase enzyme is mitochondrial in nature and has high activity during yeast-phase growth and "repressed" activity during filamentous growth. Our experiments with unifoimly labeled proline suggest that the mitochondria indeed are involved in Candida morphogenesis, for even after 1 h of growth 55% of the label remains in a cell fraction containing high respiratory activity (Fig. 6). The fact that the label appeared in the RNA fraction to a greater extent than other fractions after 15 min in these experiments suggested that proline has entered the Kreb's cycle and has been catabolized to one or more of the intermediates of purine or pyrimidine biosynthesis by that time.
In studies by Dabrowa et al. (8) on Y-M conversion of synchronized Candida yeasts, RNA was synthesized at a rate twice that of protein and thrice that of DNA. In our results, after 1 h the high respiratory fraction also contained a greater amount of labeled RNA than DNA or protein. Increased rates of RNA and protein synthesis would be expected if there was major changes in carbohydrate metabolism. Changes in carbohydrate metabolism would be expected during mitochondrial repression and have been reported to occur during Y-M conversion in C. albicans (7). The fact that growth in NH,Cl favors yeast morphology may also be related to glycolysis, generation of NADPH, and changes in mitochondrial activity. The ammonium ion has been found to be a potent activator of phosphofructokinase, an enzyme important in control of glycolysis (31). Chattaway et al. (6) reported that the activity of phosphofructokinase and the hexose monophosphate shunt (HMP) was consistently higher in yeast cells when compared to filaments. Since the HMP pathway would generate NADPH for cellular division, the maintenance of yeast morphology by disulfhydryl reductase (22) would be possible. The ammonium ion also has been reported to affect activities of one of the two glutamic dehydrogenases present in fungi (1, 11, 28, 33). Under conditions of NH4+ assimilation, an NADP+dependent glutamic dehydrogenase is activated which is also correlated with an increased HMP activity (5, 33). The maintenance of yeast morphology in C. albicans in a medium containing NH,Cl, therefore, may be a result of the effect on NH4+ upon an NADP+-dependent glutamate dehydrogenase and increased HMP activity. A relationship such as this would reconcile our data and that of Chattaway et al. (6) with Nickerson's model (22), since NADPH would be generated by glutamic dehydrogenase as well as by the HMP pathway. It is also interesting to note that the second glutamic dyhydrogenase is an NAD+-linked enzyme, synthesized under conditions favoring the breakdown of glutamate. Any amino acid, such as proline, which uses glutamate as a primary intermediate in metabolism would favor an induction of the NAD+-linked enzyme. Generation of NADH during catabolism of proline might be at the expense of formation of the necessary NADPH needed to maintain cellular division and yeast morphology according to the Nickerson model; thus, filaments would form. We believe, however, that control of Y-M conversion does not lie solely in the amount of NADPH generated by the fungus, but rather in
LAND ET AL.
a complex system of controls effected by NADH, NADPH, and internal glucose-to-phosphate ratios. In the present study, morphology of Candida was influenced by high and extremely low concentrations of glucose, as well as by proline catabolism. When glucose was limiting (i.e., the glucose-to-phosphate ratio was low), NADH oxidase activity would be favored over NADPH production to meet the immediate energy requirements of the starved cells (5). This would require further depletion of the already low glucose, perhaps at the expense of glucose available for HMP activity. Cellular division, under the above conditions, could not be maintained until the HMP was active enough to provide the critical levels of NADPH; thus filaments would form. If the glucose-tophosphate ratio was high, filamentation might be brought about by the "Crabtree effect" in the starved cells. Energy requirements for the starved cells would again be immediate, favoring the NADH-oxidase system, and at the same time there would be an immediate requirement of the cell to phosphorylate glucose for glycolysis (18). Both energy (i.e., adeosine 5'-triphosphate) production and phosphorylation of glycolytic intermediates would require phosphate or phosphorylated adenylate moieties (adenosine 3'-monophosphate, adenosine 5'-diphosphate, and!adenosine 5'-triphosphate), thereby,.according to one current theory, forcing mitochondrial repression (18). As a result of the cell maintaining a balance between oxidative phosphorylation and glycolysis, the NADPH levels might be depleted. Filaments rather than yeasts would be formed, since there would be an uncoupling of energy flow (mitochondrial repression) from cellular division as predicted by the Nickerson model (22). Filamentation of starved yeasts during proline metabolism may involve a similar mechanism. If sufficient NADH were generated as proline was being utilized, NADH oxidase and oxidative phosphorylation would be favored over NADPH maintenance to accommodate the increased amount of NADH within the cell. In addition, glucose would have to be phosphorylated for glycolysis. The result would be a perturbation of the balance between glycolysis and oxidative phosphorylation, as is postulated to occur (18) when the glucose-to-phosphate ratio is high; thus repression of mitochondria would occur. The fact that the total organic acids and various Kreb cycle intermediates of Candida showed an increase (favoring energy production) and then abrupt decrease (indicative of mitochondrial repression) during proline metabolism supports that hypothesis.
In summary, C. albicans became filamentous in a minimal medium containing proline or certain other amino acids as a sole nitrogen source. The proline appeared to be catabolized in a manner which would be consistent with the generation of increased reducing potential with the Candida mitochondria, namely, by entry into the Kreb cycle via a-ketoglutarate. A decrease in Kreb's cycle activity within filaments was indicated by a marked increase and then an abrupt decrease in the concentrations of organic acids and specific Kreb cycle intermediates. High glucose has been reported to inhibit respiration and Kreb's cycle activity in Ehrlich ascites tumor cells (18). and in Saccharomyces (10), by the Crabtree effect. The Crabtree effect is postulated by Koobs (18) to occur by depletion of phosphate reserves during mobilization of excess glucose at the expense of oxidative phosphorylation. It is also interesting to note that Candida converts to filaments under similar conditions of high glucose-to-phosphate ratios. As energy production competes with glycolysis for phosphorylated adenylates, a Crabtree-like inhibition of respiration would eventually occur. The suppression of mitochondrial function in turn would cause changes in carbohydrate biosynthesis and metabolism. Changes in the types of carbohydrates available for cell wall synthesis would influence cell wall composition (7) and, ultimately, morphology. Thus, as Nickerson's model dictates, the interaction of cellular division with energy and carbohydrate metabolism would control morphogenesis in C. albicans. ACKNOWLEDGMENTS G. A. L., was a trainee in medical mycology. This investigation was supported by Public Health Service training grant 5-T01-AI00003 from the National Institute of Allergy and Infectious Diseases. LITERATURE CITED 1. Arnst, H. N., Jr., and D. W. MacDonald. 1973. A mutant of Aspergillus nidulans lacking NADP-linked glutamate dehydrogenase. Mol. Gen. Genet. 122:261-265. 2. Ateman, K. 1968. Pathology of Candida infection of the umbilical cord. Am. J. Clin. Pathol. 49:798-804. 3. Bartnicki-Garcia, S., and I. McMurrough. 1971. Biochemistry of morphogenesis in yeasts, p. 441-491. In A. H. Rose, and J. S. Harrison (ed.), The yeasts, vol. II. Academic Press Inc., New York. 4. Brown-Thomsen, J. 1968. Variability in Candida albicans (Robin) Berkhout. I. Studies on morphology and bio-
chemical activity. Hereditas 60:355-398. 5. Chandra, A. K., and A. B. Bauerjee. 1973. Oxidation of
specifically labeled glucose by Trichophyton rubrum. Acta Microbiol. Pol. 4:197-200. 6. Chattaway, F. W., R. Bishop, M. M. Holmes, F. C. Odds, and A. J. E. Barlow. 1973. Enzyme activities associated with carbohydrate synthesis and breakdown in yeast and mycelial forms of Candida albicans. J. Gen. Microbiol. 75:97-109.
VOL. 11, 1975
FILAMENTATION IN C. ALBICANS
7. Chattaway, F. W., M. R. Holmes, and A. J. E. Barlow. 1968. Cell wall composition of the mycelial and blastospore forms in Candida albicans. J. Gen. Microbiol.
12. 13. 14.
15. 16. 17.
18. 19. 20. 21.
51:367-376. Dabrowa, N., D. A. Howard, J. W. Landau, and Y. Shechter. 1970. Synthesis of nucleic acids and proteins in the dimorphic forms of Candida albicans. Sabouraudia 8:163-169. Dabrowa, N., J. W. Landau, and V. D. Newcomer. 1967. Generation time of Candida albicans in synchronized and nonsynchronized cultures. Sabouraudia 6:51-56. Dedeken, R. H. 1966. The Crabtree effect and its relation to the petite mutation. J. Gen. Microbiol. 44:157-165. Dubois, E., M. Grenson, and J. M. Wiame. 1973. Release of the "ammonia effect" on three catabolic enzymes by NADP-specific glutamate dehydrogenase less mutations in Saccharomyces cerevisiae. Biochem. Biophys. Res. Commun. 50:967-972. Dulbecco, R., and M. Vogt. 1954. Plaque formation and isolation of pure lines with poliomyelitis virus. J. Exp. Med. 99:167-182. Gancedo, J., and C. Gancedo. 1973. Concentrations of intermdiary metabolites in yeasts. Biochimie 55:205-211. Gentles, J. C., and C. J. LaTouche. 1969. Yeast as human and animal pathogens, p. 108-182. In A. H. Rose and J. A. Harrison (ed.), The yeasts, vol. I. Academic Press Inc., New York. Gilardi, G. L. 1965. Nutrition of systemic and subcutaneous pathogenic fungi. Bacteriol. Rev. 29:406-424. Graafmous, W. D. J. 1973. The influence of carbon dioxide on morphogenesis in Penecillium isariiforme. Arch. Mikrobiol. 91:67-76. Hurley, R., and V. C. Stanley. 1969. Cytopathic effects of pathogenic species of Candida on cultured mouse epithelial cells: relation to the growth rate and morphology of the fungi. J. Med. Microbiol. 2:63-74. Koobs, D. H. 1972. Phosphate mediation of the Crabtree and Pasteur effects. Science 178:127-133. Kraeger, S. J., and J. G. Hamilton. 1969. Quantitative glass-paper chromatography of fungal cell wall acid hydrolysates. J. Chromatogr. 41:113-115. Mardon, D., E. Balish, and A. W. Phillips. 1969. Control of dimorphism in a biochemical variant of Candida albicans. J. Bacteriol. 100:701-707. Moffat, E. D., and R. I. Lytle. 1959. Polychromatic technique for the identification of amino acids on paper chromatograms. Anal. Chem. 31:926-928.
22. Nickerson, W. J. 1954. Experimental control of morphogenesis in microorganisms. Ann. N.Y. Acad. Sci.
60:50-57. 23. Nickerson, W. J. 1963. Symposium on the biochemical basis of morphogenesis in fungi. IV. Molecular basis of form in yeast. Bacteriol. Rev. 27:305-324. 24. Nickerson, W. J., and C. W. Chung. 1953. Cellular division of a mutant yeast. Am. J. Bot. 41:114-120. 25. Nordmann, J., and R. Nordman. 1960. Organic acids, p. 272-290. In I. Smith (ed.), Chromatographic and electrophoretic techniques, vol. I. Pitman Press, New York. 26. Reynolds, R., and A. Braude. 1956. The filament inducing property of blood for Candida albicans. Its nature and significance. Clin. Res. Proc. 4:40-47. 27. Schnaitman, C. A. 1970. Protein composition of the cell wall and cytoplasmic membrane of Escherichia coli. J. Bacteriol. 104:890-901. 28. Schwencke, J., and N. Magana-Schwencke. 1969. Derepression of a proline transport system in Saccharomyces chevalieri by nitrogen starvation. Biochim. Biophys. Acta 173:302-312. 29. Shatkin, A. J. 1969. Colorimetric reactions for DNA, RNA, and protein determinations, p. 231-237. In K. Halsel and N. Salzman (ed.), Fundamental techniques in virology. Academic Press Inc., New York. 30. Smith, I. 1960. Sugars, p. 246-260. In I. Smith (ed.), Chromatographic and electrophoretic techniques, vol. I. Pitman Press, New York. 31. Sols, A., C. Gancedo, and G. Delafuente. 1971. Energy yielding metabolism in yeasts, p. 273-307. In A. H. Rose and J. S. Harrison (ed.), The yeasts, vol. II. Academic Press Inc., New York. 32. Taschdjian, C. L. 1970. Opportunistic yeast infections, with a special reference to candidiasis. Ann. N.Y. Acad. Sci. 174:606-622. 33. Van De Poll, K. W. 1973. Activity of the hexose monophosphate shunt in a mutant of Saccharomyces carlsbergensis lacking NADP dependent glutamate dehydrogenase activity. FEBS Lett. 32:33-34. 34. Wickerham, L. J., and L. F. Rettger. 1939. A taxonomic study of Monilia albicans, with special emphasis on morphology and morphological variation. J. Trop. Med. Hyg. 42:174-213. 35. Widra, A. 1964. Phosphate directed yeast mycelial variation in Candida albicans. Mycopathologia 23:197-202. 36. Yamaguchi, H., Y. Kanda, and K. Iwata. 1971. Biochemical properties of mitochondria from Candida albicans. Sabouraudia 9:221-230.