Diverse Nitrogen Sources in Seminal Fluid Act in Synergy To Induce Filamentous Growth of Candida albicans

Abstract

The pathogenic fungus Candida albicans is the leading cause of vulvovaginal candidiasis (VVC). VVC represents a major quality-of-life issue for women during their reproductive years, a stage of life where the vaginal epithelium is subject to periodic hormonally induced changes associated with menstruation and concomitant exposure to serum as well as potential intermittent contact with seminal fluid. Seminal fluid potently triggers Candida albicans to switch from yeastlike to filamentous modes of growth, a developmental response tightly linked to virulence. Conversely, vaginal fluid inhibits filamentation. Here, we used artificial formulations of seminal and vaginal fluids that faithfully mimic genuine fluids to assess the contribution of individual components within these fluids to filamentation. The high levels of albumin, amino acids, and N-acetylglucosamine in seminal fluid act synergistically as potent inducers of filamentous growth, even at atmospheric levels of CO₂ and reduced temperatures (30°C). Using a simplified in vitro model that mimics the natural introduction of seminal fluid into the vulvovaginal environment, a pulse of artificial seminal fluid (ASF) was found to exert an enduring potential to overcome the inhibitory efficacy of artificial vaginal fluid (AVF) on filamentation. These findings suggest that a transient but substantial change in the nutrient levels within the vulvovaginal environment during unprotected coitus can induce resident C. albicans cells to engage developmental programs associated with virulent growth.

INTRODUCTION

Candida albicans is an opportunistic fungal pathogen that is a benign member of the microflora of most individuals. A majority of women (≥75%) experience at least one episode of acute vulvovaginal candidiasis (VVC), and a substantial number (≥10%) suffer from recurrent chronic vaginal infections, defined as at least 4 episodes per year (1, 2). Prepubescent girls and postmenopausal women with low levels of production of estrogen rarely develop the disease. Consequently, women are susceptible to VVC when their vaginal epithelium is subject to periodic hormonally induced changes with the concomitant infusion of serum and potential intermittent exposure to seminal fluid (SF) (3, 4). Serum and seminal fluid are known to trigger yeastlike-to-hyphal morphogenic transitions in C. albicans, a developmental response linked to virulent growth (5).

Filamentous forms of C. albicans are associated with symptomatic VVC (6). True hyphae elaborated by C. albicans are elongated structures with characteristically distinct parallel walls and regularly spaced septae that confine individual nuclei. Hyphae are able to penetrate the mucosal surfaces of the host and cause damage to the epithelium. Importantly, C. albicans strains capable of forming hyphae promote symptomatic experimentally induced VVC in women, whereas mutant strains incapable of hyphal growth show diminished adherence to vaginal epithelial cells and reduced colonization and infection in a murine model of C. albicans vaginitis (7).

The vaginal environment is complex. Although the vaginal epithelium together with associated physiological secretions (vaginal fluid [VF]/cervical mucus) and resident commensal bacterial microflora contribute to maintaining an acidic environment that represses filamentation of C. albicans (8, 9), the vaginal epithelium (matrix) represents an inherently conducive environment for morphological switching of fungal cells. Particularly, the physical contact with epithelial surfaces, the constant temperature of 37°C, and elevated levels of CO₂ are established factors that favor filamentous growth (10). Additionally, in sexually mature women, the vaginal epithelium undergoes transient changes that can induce filamentation. The pH of cervical mucus is usually acidic (pH 4 to 4.5) (noninducing); however, during menses, it increases to pH 6.8 (inducing). Similarly, the introduction of seminal fluid, which is alkaline (pH 7 to 8.5), can transiently increase the pH within the vagina (11). Furthermore, serum and seminal fluid contain high concentrations of known morphogenic factors, including amino acids, proteins, and, in the case of seminal fluid, N-acetylglucosamine (GlcNAc).

In contrast to many microbial pathogens, C. albicans has a diverse metabolic repertoire, is able to colonize virtually any tissue and organ, and possesses efficient systems to extract nitrogen from diverse host niches (12). Although fungus-specific gene products involved in nitrogen uptake are likely to be essential for C. albicans virulence, only limited information is available regarding which nutrients are actually utilized during infectious growth. Within hosts, nitrogen is available primarily as free amino acids and proteins. C. albicans cells possess the means to utilize both of these forms of nitrogen and have the capacity to differentially coordinate their utilization by sensing extracellular amino acids (13). The coordinated response requires the SPS sensing pathway (13, 14). This pathway was first identified in the yeast Saccharomyces cerevisiae (reviewed in reference 15) and derives its name from the SPS sensor, a plasma membrane-localized trimeric receptor complex comprised of three core components, Ssy1, Ptr3, and Ssy5 (16). The most upstream component of the Candida SPS sensing pathway is Csh3, an endoplasmic reticulum (ER) membrane-localized chaperone that is required for the proper localization of Ssy1, the primary amino acid receptor component of the SPS sensor, and amino acid permeases to the plasma membrane (14). Thus, cells carrying csh3 null mutations have greatly reduced SPS sensor signaling and consequently exhibit a diminished capacity to take up amino acids and to filament in response to inducing amino acids (14). In systemic murine and Drosophila melanogaster infection models, C. albicans strains lacking Csh3 exhibit attenuated virulence (14, 17).

Seminal fluid is one of the few places in the human host where GlcNAc has been detected (18). This amino sugar plays important roles in a wide range of organisms from bacteria to humans (19), and the filamentation-inducing power of GlcNAc on C. albicans has been known for decades (20). In C. albicans, GlcNAc induces two signaling pathways, one leading to a switch from budding to hyphal growth and another leading to the expression of the GlcNAc transporter Ngt1 and of the genes needed for GlcNAc catabolism (21, 22).

Although the filamentation-inducing effects of serum and seminal fluid on C. albicans are well documented (23), the precise nature of the inducing signals within these body fluids remains elusive. Here, new formulations of artificial SF (ASF) and artificial VF (AVF) were developed and validated as substitutes for genuine body fluids. A systematic analysis of these artificial fluids revealed significant and complex synergistic effects between albumin, amino acids, and GlcNAc, which together potently trigger morphogenic transitions to favor filamentous growth of C. albicans. The filamentation-inducing properties of ASF exhibited long-lasting effects, even with transient exposures.

MATERIALS AND METHODS

Strains and standard media.

The C. albicans strains used in this work are listed in Table 1. Strains YJA65 (ngt1Δ) and YJA67 (csh3Δ ngt1Δ) carrying unmarked homozygous deletions of NGT1 (orf19.5392) were obtained after two successive rounds of insertion/excision using linear KpnI/SacI-restricted DNA fragments derived from plasmids pJA32 (ngt1Δ2) and pJA33 (ngt1Δ3). Insertion and excision events were monitored by the gain and loss of nourseothricin resistance, respectively (24). The two ngt1Δ alleles differ in their 5′ regions. The correct targeting of deletion constructs was confirmed by PCR, Southern analysis, and growth-based assays on selective media (data not shown). All strains were stored in 15% glycerol at −80°C and revived by streaking onto yeast extract-peptone-dextrose (YPD) for single colonies prior to the start of any experiments.

TABLE 1.

Fungal strains and plasmids used in this work

Fungal strain or plasmid	Genotype or description	Reference
Strains
Candida albicans
PMRCA18	CAI4 ura3Δ::imm434/URA3	14
PMRCA12	PMRCA18 csh3Δ3/csh3Δ3	14
YJA65	PMRCA18 ngt1Δ2::FRT/ngt1Δ3::FRT	This work
YJA67	PMRCA12 ngt1Δ2::FRT/ngt1Δ3::FRT	This work
DAY25	ura3Δ::λimm434/ura3Δ::λimm434 arg4::hisG/arg4::hisG his1::hisG/his1::hisG rim101::ARG4/rim101::URA3	59
YJA1	ura3::λimm434/ura3::λimm434 NGT1-GFP-URA3/NGT1	21
Saccharomyces cerevisiae YAF7	MATa leu2-3,112 his3-11,15 ura3-1 ade2-1 trp1-1 can1-100 rDNA::ADE2 RAD5 cir0	27
Plasmids
pSFS2	SAT1 flipper cassette (FRT-caFLP–caSAT1–FRT)	24
pRS316	URA3 CEN	26
pJA23	pRS316 containing the SAT1 flipper cassette	This work
pJA32	pJA23 ngt1Δ2 (5′ L1 and 3′ R amplicons)	This work
pJA33	pJA23 ngt1Δ3 (5′ L2 and 3′ R amplicons)	This work

Standard media, including YPD medium supplemented with 80 μg ml⁻¹ uridine, ammonia-based synthetic minimal dextrose (SD) medium, and ammonia-based synthetic complex dextrose (SC) medium, were prepared as described previously (25). Media were made solid by the addition of 2% agar.

Plasmid constructions.

Plasmids used in this work are listed in Table 1. Plasmid pJA23 was created by subcloning the KpnI/SacI fragment of pSFS2 containing the SAT1 flipper cassette (24) into KpnI/SacI-restricted pRS316 (26). Plasmids pJA32 and pJA33 were constructed by homologous recombination using gap repair in S. cerevisiae, a strategy that eliminates the need for restriction and in vitro ligation. Most S. cerevisiae strains are cir⁺; i.e., they carry endogenous 2μm plasmids that express active FLP recombinase. Consequently, recombination-based cloning strategies involving the SAT1 flipper with FLP and FRT sequences necessitate the use of cir⁰ host strains lacking 2μm plasmids. Strain YAF7 (27), generously provided by J. P. Aris, University of Florida, Gainesville, FL, was used as the host. Linear fragments of DNA comprised of 5′ and 3′ regions of the NGT1 open reading frame (ORF) flanked by short primer sequences homologous to plasmid sequences spanning the upstream KpnI (L1 and L2) and downstream SacI (R) sites of the SAT1 flipper cassette in pJA32 were amplified by using PCR with DNA from strain SC5314 as the template. Primers L1, L2, and R amplify 563-, 449-, and 489-bp fragments, respectively (sequences of primers are available upon request). L1 and R fragments, or L2 and R fragments, and KpnI/XhoI (upstream)- and SacI/NotI (downstream)-restricted pJA33 were introduced into strain YAF7. Recombination between amplified and gapped restriction fragments yielded plasmids that conferred nourseothricin resistance, selected on solid SC (lacking Ura) medium supplemented with 100 μg ml⁻¹ nourseothricin (Werner Bioagents, Jena, Germany). Plasmids pJA32 and pJA33 were rescued following passage through Escherichia coli (DH5α) cells, and the bacterial transformants were selected on LB supplemented with carbenicillin (100 μg ml⁻¹) and nourseothricin (25 μg ml⁻¹), respectively.

Artificial seminal fluid.

Artificial seminal fluid (ASF) was composed of a buffered saline solution (BSS) [61 mM Na₂HPO₄, 6.5 mM NaH₂PO₄ (H₂O), 38 mM Na citrate, 12 mM KCl, 16 mM KOH, 6.8 mM CaCl₂(H₂O)₂, 4.4 mM MgCl₂(H₂O)₆, 2.2 mM ZnCl₂, 0.27% fructose, 0.1% glucose, 6% lactic acid, 5% urea, and 1 mM uridine] supplemented with 0.7 mM GlcNAc, 1.52% albumin (essentially fatty acid-free bovine serum albumin [BSA] [catalog no. A6003; Sigma-Aldrich Sweden AB]), and the following amino acids: serine (7 mM), glutamate (3.8 mM), tyrosine (2.7 mM), leucine (2.2 mM), arginine (2.2 mM), lysine (2.1 mM), glycine (2.1 mM), valine (1.8 mM), isoleucine (1.8 mM), threonine (1.8 mM), glutamine (1.7 mM), histidine (1.4 mM), phenylalanine (0.8 mM), taurine (0.7 mM), alanine (0.6 mM), aspartate (0.6 mM), phosphoserine (0.3 mM), proline (0.2 mM), and ornithine (0.1 mM). The concentrations of amino acids were reported previously (28). The pH of ASF was 7.5 to 8.

Artificial vaginal fluid.

Artificial vaginal fluid (AVF) was composed of 52 mM NaCl, 10 mM K₂HPO₄(H₂O)₃, 10 mM KH₂PO₄, 1.5 mM CaCl₂, 2.5 mM MgSO₄(H₂O)₇, 0.2% lactic acid, 0.5% glucose, 0.04% urea, 0.03% mucin (catalog number M1778; Sigma-Aldrich Sweden AB), 0.2% BSA, and the following amino acids: glutamine (260 μM), alanine (185 μM), glycine (125 μM), proline (118 μM), valine (116 μM), lysine (93 μM), threonine (83 μM), leucine (60 μM), serine (50 μM), asparagine (44 μM), taurine (44 μM), histidine (37 μM), isoleucine (35 μM), tyrosine (30 μM), glutamate (29 μM), phenylalanine (28 μM), ornithine (27 μM), cysteine (24 μM), arginine (20 μM), citrulline (18 μM), methionine (12 μM), and phosphoserine (3 μM). The pH of AVF was adjusted to 4.2 to 4.5 with diluted HCl, and the solution was filtered through a 0.2-μm filter. AVF with a reduced BSA concentration (0.002%) was used to examine the long-term effects of ASF exposure (see Fig. 9).

FIG 9.

Mimicking of a coital event. Transient exposure to ASF has a long-term inducing effect on filamentous growth. Wild-type (PMRCA18) cells (10⁵), conditioned and incubated at 37°C as described in the text, were suspended in AVF-ASF (7:3) to reach pH 6.5 at 0 h. (Top) At 1, 1.5, 2, and 3 h, appropriate aliquots of AVF were added to obtain pH values that experimentally mimic (solid diamonds and line) the pH fluctuation in the vaginal environment during a coital event (open circles and dashed line) (11). (Bottom) Micrographs A to D depict cells at 0, 3, 6, and 22 h, respectively. The white arrows indicate hyphal branch points (C) and blastospores budding from hyphae (D). Micrographs labeled AVF and ASF are from cells incubated for 22 h in AVF and ASF, respectively. Micrographs are reproduced at the same magnification. Bar = 50 μm.

Alvarez minimal medium.

Nonstandard Alvarez minimal medium (AMM) was composed of 1.5 g liter⁻¹ yeast nitrogen base (YNB) (without amino acids and ammonium sulfate) (catalog number 233520; Difco), 5 g liter⁻¹ ammonium sulfate, 0.1% glucose, and concentrations (1.4 mM) of amino acids identical to those of AVF. Where indicated, AMM was supplemented with 0.86% BSA, 0.7 mM GlcNAc, and amino acids at concentrations (17.2 mM) similar to those of a 1:1 mix of ASF and AVF. The pH of AMM was 6.8.

Natural seminal and vaginal fluid preparations.

Seminal fluid (SF) was obtained from a healthy volunteer by masturbation after an abstinence period of 3 to 4 days. After liquefaction (20-min incubation at 20°C), the fluid was filtered through a 0.2-μm filter and stored at −20°C or used the same day. The GlcNAc content of seminal fluid was determined as follows. Samples were freeze-dried, and the sugars were hydrolyzed in a mixture of equal amounts of 4 M trifluoroacetic acid (TFA) and 4 M HCl at 100°C for 6 h. The resulting monosaccharides were chemically reduced with NaBH₄ and acetylated with anhydride acetate to obtain volatile alditol acetates, which were then analyzed by gas chromatography coupled to mass spectrometry (GC/MS): 1-μl samples were injected onto an SP-2380 capillary column (Supelco) installed in a HP 6890 series gas chromatograph coupled to an HP 5973 mass selective detector and an HP 7683 autosampler. Technical repeats were performed, and the data are presented as averages ± standard deviations. The concentration of GlcNAc in the seminal fluid samples (n = 6) was 1.35 ± 0.07 mM. Vaginal fluid (VF) (cervicovaginal mucus) was collected from the vaginal introitus of a volunteer without a history of VVC, using a sterile microcentrifuge tube. It was kept at 4°C briefly and then centrifuged at low speed (2 min at 6,200 × g). The resulting clear supernatant was used for comparison with AVF, solely with regard to pH and inhibition of filamentation. The natural samples of seminal and vaginal fluid were obtained specifically for this study from adult volunteers with their full informed oral consent, which was documented in laboratory notebooks (F.J.A.). Analyses of fluid samples were carried out anonymously, and after analysis, all samples were destroyed; no genetic material was saved. The protocols used for collection and analysis of the natural fluids have been vetted and approved by the Regional Ethical Review Board-Stockholm.

Microscopic analysis.

Growth characteristics of Candida cells were analyzed by microscopy to quantitatively assess the degree of filamentation and clumping. Cells from −80°C stocks were first grown on YPD plates and then in liquid YPD until logarithmically expanding cultures were obtained. Cells were collected by centrifugation at 6,200 × g, and the pelleted cells were washed twice with phosphate-buffered saline (PBS) and used to inoculate AVF or AMM at a starting density of between 5 × 10⁵ and 1 × 10⁶ cells ml⁻¹. Unless otherwise noted, the cultures were incubated overnight at 30°C without shaking. Cells were harvested by centrifugation (2 min at 6,200 × g), washed with PBS, and resuspended in 100 μl of fresh medium or experimental medium, as indicated. After 6 h of incubation without shaking, cells were pelleted and resuspended in a volume of 10 to 20 μl. Cell suspensions were pipetted up and down 10 times and spotted onto slides for microscopic analysis. Images were captured on several different microscopes. Regarding the results shown in Fig. 6, images of cells evenly distributed over the entire field of view were analyzed by superimposing a grid of 12 uniform squares (3 rows and 4 columns, designated A to L from top left to bottom right). Cells in squares B, C, E, H, J, and K were counted, and the percentage of filamentation was determined (hyphal cells and germ tubes were classified together against the total number of cells; pseudohyphae were rare). The data were analyzed by using analysis of variance (ANOVA) with a Bonferroni post hoc test in SPSS (IBM SPSS Statistics, version 19.0.0); a P value of <0.05 is considered significant.

FIG 6.

Albumin (BSA), amino acids (AA), and GlcNAc act differentially and in a synergistic manner to induce filamentation. Wild-type (PMRCA18), ngt1Δ (YJA65), csh3Δ (PMRCA12), and csh3Δ ngt1Δ (YJA67) cells were grown and incubated as described in the legend of Fig. 4. At 0 h, cells were pelleted and resuspended in AVF (condition C), AVF-BSS (1:1) (condition 1), or AVF-BSS (1:1) supplemented with the indicated nutrients (conditions 2 to 8). After 6 h, the growth characteristics of the cells were examined by microscopy (Fig. 7), and the extent of filamentation was quantified as described in Materials and Methods. All incubations were performed at 30°C and at ambient CO₂ levels without shaking. The values depicted are from 3 independent experimental replicates (n = 3), and percent filamentation was assessed by evaluating >1,000 cells under each condition; a total of 74,000 cells were counted. Error bars represent 1 standard deviation. The data were analyzed by using ANOVA with a Bonferroni post hoc test in SPSS (IBM SPSS Statistics, version 19.0.0); a P value of <0.05 is considered significant. The significant differences observed between conditions and strains were as follows: a P value of <0.001 for wild-type condition 2 versus condition 1, a P value of 0.002 for wild-type condition 3 versus condition 1, a P value of <0.001 for wild-type condition 4 versus condition 1, a P value of <0.001 for wild-type condition 6 versus condition 5, a P value of <0.001 for wild-type condition 7 versus condition 5, a P value of <0.001 for wild-type condition 8 versus condition 5, a P value of 0.01 for wild-type versus csh3Δ ngt1Δ cells for condition 2, a P value of 0.001 for wild-type versus csh3Δ ngt1Δ cells for condition 3, and a P value of <0.001 for wild-type versus csh3Δ ngt1Δ cells for condition 4.

RESULTS

Defined synthetic artificial seminal and vaginal fluids mimic endogenous fluids. (i) Artificial seminal fluid.

SF is composed of sperm cells within seminal plasma derived from secretions of the seminal vesicles and prostate, with minor contributions by the testes and the bulbourethral glands. SF contains citric acid, high levels of free amino acids, fructose, albumin, polyamines, GlcNAc, uridine, and zinc (18, 28–30) and typically has a pH of 7 to 8.5 (31). A synthetic formulation of SF has been reported (32). However, this formulation omitted uridine, amino acids, and GlcNAc, the latter two being known morphogens that stimulate filamentous growth of C. albicans. Consequently, to more accurately mimic genuine SF, artificial SF (ASF) was formulated to include albumin (BSA) (final concentration, 1.52%), amino acids (final concentration, 33 mM), and GlcNAc (final concentration, 1.4 mM). The amount of GlcNAc (1.35 ± 0.07 mM) corresponds to the actual concentration measured in seminal fluid (n = 6).

(ii) Artificial vaginal fluid.

Vaginal fluid (VF), or cervical mucus, contains glucose, glycogen, glycerol, lactic and acetic acids, amino acids, and urea. In addition, VF typically contains high potassium and low sodium concentrations and has an acidic pH (11, 33, 34). Although VF contains free amino acids (34), the precise composition and concentrations of amino acids are not known. Synthetic formulations of VF have been reported previously (35–37), but these formulations omitted amino acids. Since VF is mostly a plasma exudate derived from epithelial cells lining the vaginal walls, it likely contains amino acids and proteins reflecting the composition in serum. To better mimic genuine VF, an artificial VF (AVF) was formulated to reflect the amino acid composition of serum but with reduced levels (50% serum) (38), bringing the final concentration of amino acids to 1.44 mM. AVF also contains 0.03% mucin and 0.2% BSA.

(iii) Validation.

The filamentation-promoting and -inhibiting properties of the ASF and AVF formulations were tested and compared to those of the genuine fluids (Table 2). Filamentation of C. albicans in vitro is known to be stimulated at temperatures of ≥37°C, especially in combination with common inducers such as serum, GlcNAc, or Lee's medium (39). Similarly, CO₂ levels of 5 to 6% have also been shown to enhance hyphal growth in C. albicans (40). Filamentation of wild-type C. albicans cells (PMRCA18) exposed to ASF and AVF at 37°C and 5% CO₂ was examined (Fig. 1). Exponentially growing cells in YPD were washed with PBS and incubated without shaking in the presence of genuine or artificial fluids. Cells incubated in the presence of either AVF or genuine VF did not develop hyphae. In contrast, both artificial ASF and genuine SF were highly effective at inducing hyphae. The levels of filamentation induced by ASF and SF were indistinguishable. Next, the ability of ASF to induce GlcNAc-specific gene expression was examined. NGT1, which encodes the major GlcNAc transporter, is specifically induced in response to GlcNAc (21). Strain YJA1, carrying a reporter construct with green fluorescent protein (GFP) fused to the GlcNAc transporter (NGT1-GFP), was used to assess GlcNAc-dependent gene expression (Fig. 2). ASF induced Ngt1-GFP expression at levels indistinguishable from those observed with SF, and in both instances, Ngt1-GFP fluorescence was detected at the plasma membrane. Together, the results indicate that AVF and ASF appear to mimic the inhibitory and inducing properties of genuine body fluids, respectively.

TABLE 2.

Comparison of concentrations of key body fluid components

Body fluid	Hexose (%)	Amino acids (mM)	GlcNAc (mM)	Albumin (%)	pH
VF	0.6–1a	Not determined	0	0.002–0.38g	3.5–4.5i
AVF	0.5b	1.4k	0	0.002 or 0.2l	4.2–4.4
SF	0.37c	34e	1.44k	1.5c	7.2–8.5c
ASF	0.37	34	1.44	1.5	7.5–8
Plasma	0.1d	2.8–4.4f	0	5h	7.3–7.4j

^aSee reference 37.

^bSee reference 35.

^cSee reference 32.

^dSee reference 60.

^eSee reference 28.

^fSee reference 38.

^gSee reference 61.

^hSee reference 51.

ⁱSee reference 11.

^jSee reference 62.

^kThis work.

^lAs indicated in the text.

FIG 1.

Growth characteristics of C. albicans in bona fide and artificial vaginal and seminal fluids. Logarithmically growing C. albicans wild-type (PMRCA18) cells in YPD (shaking) were washed twice with PBS, suspended at 10⁶ cells ml⁻¹ in AVF, and incubated for 12 h at 37°C and with 5% CO₂ without agitation (still). The culture was washed twice with PBS and resuspended at 10⁶ cells ml⁻¹ in VF, AVF, SF, or ASF. Cells were incubated for an additional 6 h at 37°C and with 5% CO₂ without shaking and examined by microscopy. VF (top left) and AVF (bottom left) support yeastlike growth. SF (top right) and ASF (bottom right) induce filamentous growth. Bar = 50 μm.

FIG 2.

Expression of the GlcNAc transporter Ngt1 is induced in cells grown in the presence of ASF and SF. Logarithmically growing cultures of strain YJA1 (NGT1-GFP) were grown in YPD and resuspended at 10⁶ cells ml⁻¹ either in AMM supplemented with 0.86% BSA and amino acids without GlcNAc (A) or with 1.4 mM GlcNAc (B), in ASF (C), or in SF (D). Cell suspensions were incubated for 3 h at 37°C and with 5% CO₂ without shaking. Cells were examined by fluorescence (Ngt1-GFP) and phase-contrast (A and B) or differential interference contrast (C and D) microscopy. Bars = 5 μm (A and B) and 50 μm (C and D).

ASF is an effective dose-dependent inducer of filamentation at low temperature and atmospheric levels of CO₂.

The defined nature of AVF and ASF provided the basis to experimentally address the relative importance of the individual components of these fluids in inducing filamentation. Similarly, it was possible to test the contribution of environmental conditions thought to engage the morphological switches required for hyphal growth. First, the effects of temperature and CO₂ levels on ASF-stimulated filamentation were examined (Fig. 3). The preconditioned cells were washed and incubated in AVF only (control) or in a 1:1 mix of AVF and ASF for 6 h at 30°C or 37°C and with atmospheric (0.03%) or 5% CO₂. Unexpectedly, extensive hyphal growth occurred at 30°C and at the lower level of CO₂ (Fig. 3). Incubations at the higher temperature and CO₂ levels only slightly enhanced the degree of cell-cell adhesion or clumping.

FIG 3.

ASF induces filamentation at 30°C and with ambient levels of CO₂. Logarithmically growing wild-type cells (PMRC18) in YPD (shaking) were washed twice with PBS and suspended at 10⁶ cells ml⁻¹ in AVF. After 12 h of incubation, the cells were washed twice with PBS, resuspended at 10⁶ cells ml⁻¹ in AVF-ASF (1:1), and incubated for 6 h at the indicated temperatures and levels of CO₂. All incubations were performed without shaking (still). The micrographs are reproduced at the same magnification. Bar, 50 μm.

After coitus, the high nutrient content and high pH of seminal fluid are reduced by dilution with endogenous cervicovaginal fluids (11). Therefore, the possibility that different ratios of AVF to ASF may induce different levels of filamentation was explored. To examine this, cells were prepared as described above and suspended in mixes comprised of various ratios of AVF to ASF (Fig. 4). The results show that filamentation was robustly stimulated when cells were incubated in a mixture composed of a 1:1 mix of AVF and ASF and that increasing amounts of ASF in the mixture increased the degree of filamentation and clumping. The pH during the 6-h incubation period for each mixture tested was very stable, ranging from 4.44 in a 1:0 mix of AVF-ASF to 7.98 in a 0:1 mix of AVF-ASF.

FIG 4.

ASF-induced filamentation is concentration dependent. Logarithmically growing wild-type cells (PMRCA18) in YPD (shaking) were washed twice with PBS and suspended at 10⁶ cells ml⁻¹ in AVF. After 12 h of incubation, the culture was washed twice with PBS, resuspended at 10⁶ cells ml⁻¹ in the indicated mixtures of AVF-ASF, and incubated 6 h prior to microscopic analysis. The pH of the cell suspensions was measured at 0 h and 6 h; the values correspond to the measurements at 0 h. All incubations were performed at 30°C without shaking (still). The micrographs are reproduced at the same magnification. Bar = 100 μm.

These results demonstrate that ASF contains a strong inducer of filamentation that obviates the requirement for a temperature of 37°C and high levels of CO₂. Consequently, it can be concluded that a temperature of 37°C and high levels of CO₂ are environmental factors that are conducive to but not essential for filamentation. Unless otherwise noted, the induction of filamentation in all subsequent experiments was performed at 30°C and with ambient CO₂ levels using a 1:1 mix of AVF and ASF. Under these conditions, the majority of the albumin and amino acids and all the GlcNAc are provided by the ASF; the final concentrations in the mix are 0.86% albumin, 17 mM amino acids, and 0.7 mM GlcNAc. The pH of the 1:1 induction mix is close to neutral.

High levels of nitrogen sources enhance alkaline pH-induced filamentation.

C. albicans cells grow well in a yeastlike growth mode under acidic conditions, a characteristic that depends on calcineurin and its associated transcription factor Crz1 (41). The morphological switches that govern the yeast-to-filamentous growth transitions are impaired at low pH, whereas an elevated pH is known to induce filamentation. The filamentation induced at high pH requires a functional Rim101 pathway (42). Consistently, rim101Δ mutant strains are reported to exhibit greatly reduced capacities to filament at an alkaline pH of 8 (43).

The contribution of pH to the induction of filamentation by AVF and ASF was examined. As for the above-described experiments, cells were pregrown in AVF and incubated at 30°C and with ambient CO₂ levels in the presence of AVF or AVF-ASF (1:1) with the pH set to 8 or 4.5, respectively. As expected, increasing the pH of AVF from 4.5 to 8 resulted in increased filamentation and clumping (Table 3, compare conditions 1 and 2, and Fig. 5). Similarly, decreasing the pH of AVF-ASF from 8 to 4.5 reduced the level of filamentation and clumping (Table 3, compare conditions 3 and 4, and Fig. 5). Unexpectedly, the rim101Δ mutant strain exhibited obvious filamentation and clumping at high pH (Table 3, compare conditions 2 and 4, and Fig. 5) although to a lesser extent than that of the wild type. The more pronounced filamentation of the rim101Δ mutant strain at pH 8 in the nutrient-rich AVF-ASF mix (Table 3, compare conditions 4 and 2) suggested that other factors, likely the high nutrient content, may be enhancing filamentation.

TABLE 3.

Effect of pH on C. albicans filamentation

Condition	Description	Filamentationa
		WT	rim101Δ
1	AVF, pH 4.5	−	−
2	AVF, pH 8	+++	+
3	AVF-ASF (1:1), pH 4.5	++	+/−
4	AVF-ASF (1:1), pH 8	+++++	+++
5	AVF-BSS (1:1), pH 4.5	−	−
6	AVF-BSS (1:1), pH 8	+++	+/−

^aRelative levels, where − signifies no filamentation and +++++ signifies maximal filamentation.

FIG 5.

The high nutrient content of ASF induces filamentation independent of Rim101. Logarithmically growing wild-type (WT) (PMRCA18) and rim101Δ (DAY25) cells in YPD (shaking) were washed twice with PBS and suspended at 10⁶ cells ml⁻¹ in AVF. After 12 h of incubation, the culture was washed twice with PBS and resuspended at 10⁶ cells ml⁻¹ in AVF-ASF (1:1) or AVF-BSS (1:1) adjusted to pH 4.5 or 8, as indicated. After 6 h, the growth characteristics of the cells were examined. All incubations were performed at 30°C and with ambient CO₂ concentrations without shaking. The micrographs are reproduced at the same magnification. Bar = 50 μm.

To test this notion, the induction of filamentation was examined by using only the buffered saline solution (BSS) used for the preparation of ASF (see Materials and Methods); BSS lacks amino acids, albumin, and GlcNAc. Interestingly, cells grown in AVF-BSS (1:1) exhibited less filamentation than did cells grown in AVF-ASF (1:1). Reduced filamentation was noted at both acidic pH (Table 3, compare conditions 5 and 3, and Fig. 5) and alkaline pH (Table 3, compare conditions 6 and 4, and Fig. 5), indicating that the nutrients present in ASF indeed exert a stimulatory effect on hyphal growth. In summary, the high nutrient content of ASF enhances alkaline pH-induced filamentation.

Different nitrogen sources in ASF act synergistically to enhance filamentation.

To investigate nutrient-induced filamentation, cells were incubated under 8 defined conditions with various nutrient contents, including AVF (control) (condition C), AVF-BSS (1:1) (condition 1), and AVF-BSS (1:1) containing combinations of albumin, amino acids, and GlcNAc (conditions 2 to 8); condition 8 is identical to AVF-ASF (1:1). As in experiments described above, cells were grown overnight in YPD and then conditioned in AVF prior to being suspended under the appropriate conditions. The extent of filamentation was evaluated after incubation at 30°C and with ambient CO₂ levels for 6 h (Fig. 6 and 7).

FIG 7.

Growth characteristics of albumin (BSA)-, amino acid-, and GlcNAc-induced wild-type (WT) (PMRCA18) and csh3Δ ngt1Δ (YJA67) cells. Shown are micrographs of cells grown and incubated for 6 h under the conditions described in the legend of Fig. 6. Bar = 50 μm.

As expected, compared to cells maintained in AVF (control) (condition C; exclusive yeastlike growth), a shift from acidic to neutral pH induced filamentation (condition 1). The addition of GlcNAc (condition 2) or amino acids (condition 3) clearly potentiated pH-induced filamentation, and the combination of GlcNAc and amino acids induced a high degree of filamentation (condition 4). Although filamentation in the presence of albumin was described previously (44), albumin alone had only a relatively weak and statistically insignificant effect (condition 5). However, albumin clearly enhanced the filament-inducing effect of GlcNAc (compare conditions 6 and 2) and amino acids (compare conditions 7 and 3). Cells incubated with the combination of nutrients present in ASF (condition 8) exhibited the most pronounced degree of filamentation. These observations suggest that albumin exerts a significant synergistic effect on filamentation.

To address whether the filament-inducing effects of GlcNAc and amino acids required uptake, the response of mutant strains lacking NGT1, CSH3, or both CSH3 and NGT1 was examined. NGT1 encodes a high-affinity GlcNAc transporter, and homozygous ngt1Δ mutants are unable to efficiently take up GlcNAc (21). CSH3 encodes an ER membrane-localized chaperone required for efficient expression of amino acid permeases. Homozygous csh3Δ mutants have a reduced capacity to take up amino acids (14). The genotypes of the constructed strains were confirmed in growth-based assays and by Southern analysis. Unexpectedly, the addition of GlcNAc enhanced the filamentation of homozygous ngt1Δ cells (Fig. 6, green bars) to levels similar to those of wild-type cells (black bars) (compare conditions 2 and 1 and conditions 6 and 5). Likewise, csh3Δ mutants (Fig. 6, red bars) retained the ability to filament upon amino acid induction albeit at consistently reduced levels. In contrast, the csh3Δ ngt1Δ double mutant strain (Fig. 6, yellow bars) exhibited significantly reduced levels of filamentation, particularly in the absence of albumin, which underscores the synergistic contribution of albumin. Since csh3Δ and ngt1Δ cells have greatly reduced capacities to take up amino acids and GlcNAc, respectively, the observed nutrient-induced filamentation is likely independent of uptake. Intriguingly, cells incubated in the presence of albumin formed aggregates (Fig. 7).

Albumin augments GlcNAc- and amino acid-induced filamentation.

To critically test the synergistic effects of albumin on induction of filamentation, a nonstandard Alvarez minimal medium (AMM) was devised, which included the same concentrations of amino acids as in those in AVF (1.4 mM) but which lacked albumin. Table 4 summarizes the degree of filamentation and clumping of cells incubated in AMM (pH set to 6.8) without and with 0.86% albumin (conditions 1 and 5) and supplemented with 0.7 mM GlcNAc (conditions 2 and 6), 17.2 mM amino acids (conditions 3 and 7), and both GlcNAc and amino acids (conditions 4 and 8). Representative pictures of cell suspensions of the wild type and the csh3Δ ngt1Δ double mutant are shown in Fig. 8.

TABLE 4.

Clumping and filamentation in nutrient-supplemented AMM

Condition	Nutrient supplement			Relative clumping/filamentation levela
	Albumin	Amino acids	GlcNAc	WT		csh3Δ		ngt1Δ		csh3Δ ngt1Δ
				Clmp	Flmt	Clmp	Flmt	Clmp	Flmt	Clmp	Flmt
1	−	−	−	−	−	−	−	−	−	−	−
2	−	−	+	−	−	−	−	−	−	−	−
3	−	+	−	−	−	−	−	−	−	−	−
4	−	+	+	−	++	−	+/−	−	+	−	+/−
5	+	−	−	+++	++	+++	++	+++	++	+++	++
6	+	−	+	++++	+++++	++++	+++++	++++	+++++	++++	+++++
7	+	+	−	+++++	++++	++++	+++	+++++	++++	++++	+++
8	+	+	+	+++++	+++++	++++	++++	+++++	+++++	+++++	++++

^aRelative levels, where − signifies no clumping (Clmp) or filamentation (Flmt) and +++++ signifies maximal clumping or filamentation.

FIG 8.

Albumin (BSA) stimulates aggregation (clumpy growth) and enhances amino acid- and GlcNAc-induced filamentation. Logarithmically growing wild-type (WT) (PMRCA18) and csh3Δ ngt1Δ (YJA67) cells in YPD (shaking) were washed twice with PBS and suspended at 10⁶ cells ml⁻¹ in AMM (YNB with the same amino acid concentrations as those in AVF and no albumin or GlcNAc). After 12 h of incubation, the culture was washed twice with PBS and resuspended at 10⁶ cells ml⁻¹ in AMM (supplemented as indicated) (left) or AMM plus BSA (1:1) (supplemented with nutrients as indicated) (right). After 6 h, the growth characteristics of the cells were examined. All incubations were performed at 30°C and with ambient CO₂ levels without shaking. Micrographs are reproduced at the same magnification. Bar = 50 μm.

In the absence of albumin, the presence of GlcNAc or high levels of amino acids did not induce filamentation (Table 4 and Fig. 8). However, filamentation was induced when wild-type cells were challenged with both GlcNAc and amino acids (Table 4, condition 4, and Fig. 8, bottom row); individual cells were observed to filament, indicating that aggregation is not a requisite for GlcNAc- or amino acid-induced filamentation. Interestingly, csh3Δ ngt1Δ mutant cells exhibited reduced filamentation.

In contrast, when albumin was added to AMM (condition 5), clumping was evident. Furthermore, when GlcNAc, amino acids, or both were included in the medium together with albumin (conditions 6, 7, and 8, respectively), the size of the clumps and the degree of filamentation were greater than those under condition 5. Consistent with the results shown in Fig. 5 and 6, the synergistic effects of albumin, GlcNAc, and amino acids were obvious. The synergistic effect was also observed in the csh3Δ ngt1Δ mutant strains, suggesting that filamentation can be induced independently of nutrient uptake. The high degree of filamentation of cells in AVF-BSS supplemented with GlcNAc or amino acids (Fig. 5 and 6, conditions 2 to 4) is likely due to the low level of albumin (0.1%) contributed by AVF.

Transient exposure to ASF exerts a long-term inducing effect on filamentation.

The finding that ASF potently induces C. albicans cells to undergo morphological transitions raised the possibility that seminal fluid in the vaginal tract may have a similar inducing effect. To test this notion, a greatly simplified in vitro model intended to mimic the transient fluxes of nutrients and pH that occur in the vagina as a consequence of a coital event was developed. The protocol was based on measurements of vaginal pH (8, 45) and on the work of Masters and Johnson (11). Masters and Johnson determined that the pH transiently increases from a range of 3.5 to 4.0 to a range of 4.25 to 4.5 during sexual activity and that it peaks at 6.5 as a consequence of the introduction of seminal fluid. A period of 6 h is required before the pH is restored to the acidic basal pH, a process that presumably depends on dilution by vaginal fluid (Fig. 9, top, open circles and dashed line).

Wild-type (PMRCA18) cells, grown overnight in YPD at room temperature and with atmospheric (0.03%) CO₂ without shaking, were washed with PBS and incubated overnight in AVF containing 0.002% albumin (BSA). To match physiological conditions, the cell suspension was incubated at 37°C and with 5% CO₂ without shaking. These conditions were used for all subsequent incubations. Cells were pelleted and resuspended in AVF (0.002% albumin). Ninety microliters of a freshly prepared AVF-ASF (2:1) mix was added to 10⁵ preconditioned cells suspended in 10 μl of AVF. This mixture has a pH of 6.5, the peak pH measured by Masters and Johnson (11). Appropriate volumes of AVF (0.002% albumin) were subsequently added at 1 h (50 μl), 1.5 h (50 μl), 2 h (100 μl), and 3 h (100 μl) to attain pH values corresponding to the postcoital recovery process (Fig. 9, top). The growth characteristics of cells was examined at 0, 3, 6, and 22 h (Fig. 9A to D, respectively) and after 22 h for cells maintained in AVF and ASF (Fig. 9, bottom).

At 0 h, the cells were round and yeastlike (Fig. 9A). At 3 h, although the pH was <5.5, a substantial degree of pseudohyphal and hyphal growth was observed (Fig. 9B). At 6 h, when the pH was 4.7, greatly elongated and branched hyphal structures were evident (Fig. 9C, arrows). After 22 h, the pH was 4.6, the filaments seemed to swell at their ends, and extensive lateral budding of yeast cells was evident (Fig. 9D, arrows). For control incubations, no significant filamentation was observed in cell suspensions incubated in AVF (Fig. 9). Conversely, cells incubated in ASF exhibited pronounced hyphal growth, without lateral buds (Fig. 9, bottom). Similarly, cells retained in the initial AVF-ASF mix without subsequent AVF additions formed hyphae without lateral buds (data not shown). The data indicate that transient environmental changes, similar to those that occur within the vagina during coitus, can induce morphological transitions resulting in filamentous growth of C. albicans.

DISCUSSION

From the perspective of a C. albicans cell, the vulvovaginal environment represents a complex host environment. The ability of C. albicans to persist as a member of the commensal microflora requires the ability to compete for essential nutrients and to respond to changes in the host environment. One of the intriguing aspects of the vulvovaginal environment is that in addition to the monthly hormonal cycle, there is an intermittent influx of body fluids containing high concentrations of multiple nutrients. Importantly, these sources of nitrogen can induce morphological transitions favoring filamentous growth, generally considered to be the major virulence property of this polymorphic pathogen (5). In this study, improved artificial formulations of vaginal and seminal fluids were generated, with the goal of assessing how the individual components and physical properties of these fluids contribute to the induction of hyphal growth in C. albicans.

The results indicate that nitrogenous compounds present in seminal fluid, i.e., albumin, amino acids, and GlcNAc, have a direct and inducing effect on filamentation. In contrast, the relatively constant features of the vulvovaginal environment, including a physiological temperature (37°C) and CO₂ levels (5%) above ambient levels, were not required for the induction of filamentation, as ASF induced filamentation at 30°C and with atmospheric levels of CO₂. Consequently, a temperature of 37°C and high CO₂ levels appear conducive to filamentation. Furthermore, the results indicate that ASF induced filamentation in a mutant lacking Rim101 albeit to lower levels than those in the corresponding wild-type strain. This finding indicates that the Rim101 pH-sensitive pathway functions to augment nutrient-based signals. Also, the data indicate that a brief exposure of C. albicans cells to ASF had an enduring effect that promoted hyphal growth despite restoration of environmental conditions usually defined as being inhibitory for filamentation, including low pH. Lu et al. (46) reported previously that hyphal initiation and hyphal elongation are two distinct processes: an initiation event that triggers the rapid disappearance of the hyphal repressor Nrg1, followed by an elongation phase, which correlates with chromatin remodeling that impairs Nrg1 rebinding. The high nutrient content of seminal fluid apparently affects these processes and does so even at 30°C, a temperature lower than the requisite temperature of 37°C reported by Lu et al. (46). Clearly, once initiated, hyphal growth can proceed for an extended period. The events required to reset the morphological switches needed to restore yeastlike growth remain to be elucidated.

Interestingly, and not previously reported, clear synergistic effects associated with the presence of albumin were noted; together with GlcNAc and/or amino acids, albumin greatly induced filamentation and clumping of cells. Clumping leading to the formation of cell aggregates is likely the consequence of adhesins expressed on the cell surface of C. albicans. Adhesins are known to be important for adherence to epithelial and endothelial cells of the host and also to bacteria living in the same host niche, e.g., biofilms (47). Adhesion of C. albicans cells to protein-coated beads has been reported (48). The observation regarding the property of albumin to enhance cell aggregation may be relevant to understanding the expression of well-established virulence traits.

Albumin may affect filamentation due to its other properties. Previous work based on the fractionation of seminal fluid components identified two small peptides capable of hyphal induction in the presence of glucose, Mn²⁺, and an incubation temperature of 37°C (filamentation did not occur at 30°C) (49). Albumin was implicated as a possible carrier of these peptides, the sequences of which remain to be identified. More recently, bacterial cell wall fragments capable of inducing C. albicans filamentation were found naturally associated with albumin (50). In the current study, albumin did not induce filamentation when present as the sole nitrogen source. Hence, it remains unclear how the combined presence of albumin with GlcNAc and/or amino acids induces filamentation and clumping. Intriguingly, this response does not depend on the uptake of GlcNAc or amino acids; similar levels of clumping were observed for csh3Δ and ngt1Δ null mutant strains.

An important implication of this work is the need to use medium formulations that accurately reflect the conditions of the host. For example, albumin is an abundant protein present in all human tissues, including blood, at an average concentration of 5% (51). It has been shown that due to nonspecific binding, the presence of albumin in the medium increases the MIC of the antifungal echinocandins for different Candida species (52). Similarly, an acidic pH affected the inhibitory concentrations of antifungal drugs intended for the treatment of VVC in in vitro susceptibility tests compared to traditional assays performed at neutral pH (53). Based on the current results, it seems reasonable to assume that the growth and behavior of C. albicans observed in many common laboratory assays regarding VVC would be different if physiological levels of nutrients and pH were used.

Relatively few publications have reported a correlation between sexual intercourse and VVC (54–57). The dearth of information regarding this issue may be the consequence of underreporting or inaccurate reporting, and the contribution of SF to VVC may simply have been overlooked. Alternatively, it is possible that the vulvovaginal environment efficiently counteracts the virulence-promoting effects of SF. If the latter is correct, several known components missing from the greatly simplified in vitro model of the postcoital vaginal environment presented in this work may play critical roles. For example, ASF lacks sperm and bioactive molecules such as transforming growth factor β (TGF-β) cytokines that induce proinflammatory responses to defend against pathogens and facilitate pregnancy (58). Also, the model lacks host epithelial cells, the hormone estrogen, and the bacterial microflora, all of which could dampen the capacity of nitrogenous nutrients to induce filamentation. Finally, the C. albicans strains utilized in this study are common laboratory strains; fresh clinical isolates that have been minimally passaged may respond differently.

In summary, the results presented in this work have illuminated novel and important synergies among nutrients present in SF that induce hyphal growth and clumping of fungal cells. Additional studies are needed to more carefully examine the effects and the molecular mechanisms underlying the synergistic interplay of the multiple nutrient sources and to test whether the findings regarding the filament-promoting effects of ASF in vitro can be extended to C. albicans cells within the vulvovaginal microflora of women. An enhanced understanding of these processes is necessary to advance toward the ultimate goal of combating a significant quality-of-life issue for women.