GNP2 Encodes a High-Specificity Proline Permease in Candida albicans

ABSTRACT The tight association of Candida albicans with the human host has driven the evolution of mechanisms that permit metabolic flexibility. Amino acids, present in a free or peptide-bound form, are abundant carbon and nitrogen sources in many host niches. In C. albicans, the capacity to utilize certain amino acids, like proline, is directly connected to fungal morphogenesis and virulence. Yet the precise nature of proline sensing and uptake in this pathogenic fungus has not been investigated. Since C. albicans encodes 10 putative orthologs of the four Saccharomyces cerevisiae proline transporters, we tested deletion strains of the respective genes and identified Gnp2 (CR_09920W) as the main C. albicans proline permease. In addition, we found that this specialization of Gnp2 was reflected in its transcriptional regulation and further assigned distinct substrate specificities for the other orthologs, indicating functional differences of the C. albicans amino acid permeases compared to the model yeast. The physiological relevance of proline uptake is exemplified by the findings that strains lacking GNP2 were unable to filament in response to extracellular proline and had a reduced capacity to damage macrophages and impaired survival following phagocytosis. Furthermore, GNP2 deletion rendered the cells more sensitive to oxidative stress, illustrating new connections between amino acid uptake and stress adaptation in C. albicans.

ScGap1, and ScPut4. C. albicans encodes 10 orthologs, which cluster into three subgroups (Gnp1 to -3, Gap1 to -6, and Put4) ( Fig. 1; see also Table S1A in the supplemental material). In general, BLASTp analysis showed that both S. cerevisiae and C. albicans AAPs are highly similar on the protein level within the same species, making precise assignments difficult (Table S1B). Nevertheless, we either generated or utilized available single-deletion strains or strains carrying multiple AAP knockouts from the respective subgroups. The resulting set was then used to comprehensively characterize the substrate specificity of the deleted permeases.
Characterization of C. albicans amino acid permease substrate specificities. To investigate amino acid specificity, we applied a screening approach using sterically similar toxic amino acid analogs. An AAP deletion strain displaying resistance to a toxic amino acid analog would indicate that the respective permease is required for the uptake of the sterically similar amino acid.
At first, we tested amino acid analogs shown to impair microbial growth, as well as other commercially available analogs, for their capacity to inhibit the growth of the SC5314 wildtype strain when added to amino-acid-free SD minimal medium. Using this approach, we identified toxic analogs of 11 amino acids: Pro, Asn, Gln, Glu, Arg, Lys, Ala, Phe, Tyr, Met, and Trp ( Fig. 2A and B; Table S2A). Next, the single AAP deletion strains and the stp2D strain, impaired in the uptake of multiple amino acids, were tested with the established screening method in order to identify AAP substrate specificities and the requirement of Stp2 for their utilization (Fig. 2C). For better comparability, the growth of each deletion strain in the presence of the particular analogs was expressed as a percentage of the respective inhibition of the wild type (see Materials and Methods; Table S2B). The retained growth of the stp2D strain on toxic analogs of Stp2-dependent amino acids confirmed the functionality of this approach.
The most striking phenotype was observed for the gnp2D strain, which was completely resistant to the proline analog 3,4-dehydro-L-proline but otherwise sensitive to all other tested analogs (Fig. 2C). Resistance to the proline analog was observed in all strains deleted for GNP2, namely, gnp1D/gnp2D, gnp2D/gnp3D, and gnp1D/gnp2D/gnp3D. Since no other AAP deletion resulted in resistance to 3,4-dehydro-L-proline, we inferred that Gnp2 is critical for C. albicans proline uptake. Consequently, the reintegration of GNP2 into the gnp2D strain resulted in restored sensitivity to 3,4-dehydro-L-proline (Fig. S1). Of note, the stp2D strain displayed moderate resistance to the proline analog, suggesting that the deletion of this transcription factor influences proline utilization.
Aside from the prominent role of Gnp2 in proline uptake, we were able to assign putative functional roles to other AAPs in C. albicans. For instance, strains carrying GNP1 deletions displayed high-level resistance to the glutamine and asparagine analogs and intermediate resistance to the tryptophan analog (Fig. 2C). The retained capacity of the gnp1D strain to grow in the presence of the asparagine analog was unique among all tested strains, suggestive of specialized uptake by Gnp1. Additional permeases are likely involved in glutamine and tryptophan uptake: the gap1D, gap3D, and gap1D/gap2D/gap6D strains were slightly resistant to the glutamine analog, while gap1D, gap4D, and, to a lesser extent, gap3D, gap5D, and gap1D/gap2D/gap6D strains were resistant to the tryptophan analog 5fluor-DL-tryptophan. Furthermore, strains carrying deletions of PUT4, GAP1, and GNP1/2/3 exhibited slight to moderate resistance to the lysine analog L-4-thialysine (Fig. 2C).
To expand the results from the toxic amino acid analog screen, we tested the GNP deletion strains for growth on SD medium supplemented with single amino acids as the sole nitrogen source. We also included the stp2D strain as its growth characteristics under the tested conditions are well defined (33,39). Overall, the amino acid capacity to promote fungal growth varied greatly; e.g., glutamine and asparagine promoted robust growth, whereas tyrosine or lysine was the least capable of supporting C. albicans growth as a nitrogen source (Fig. S2A). The stp2D strain was impaired in growth on multiple nitrogen sources but not on proline (Fig. S2B), which is well in line with previous findings (33). Strains in which GNP2 was deleted consistently had impaired growth FIG 2 Screen of C. albicans AAP mutants for resistance to toxic amino acid analogs reveals novel phenotypes. (A) Relative growth of the SC5314 wild type in SD minimal medium supplemented with toxic amino acid analogs compared to growth in SD medium alone. The analogous proteinogenic amino acid is indicated in parentheses. (B) Representative growth curves for SC5314 in SD medium alone and supplemented with 3,4-dehydro-L-proline, L-canavanine, or L-4-thialysine. (C) Growth inhibition of AAP deletion strains and the stp2D mutant relative to the wild type on each toxic amino acid analog. Red indicates growth inhibition of the AAP deletion strain equivalent to that of the wild type, while green indicates unaffected growth of the AAP deletion strain equal to the growth in SD medium alone.
Garbe et al. ® on proline as the sole nitrogen source (Fig. S2B). Furthermore, the gnp1D/gnp2D strain showed moderately impaired growth on methionine, while the gnp1D strain grew less well on threonine, and no impairment was visible on asparagine or glutamine (Fig. S2B). Oddly, strains lacking GNP1 or GNP3 had markedly increased (up to 260%) growth on lysine as a nitrogen source compared to SC5314 (Table S2C).
In order to validate that exclusively Gnp2 is important for proline uptake in C. albicans, we next tested the growth of GNP deletion strains on proline as the sole carbon source. Additionally, we included the put4D and gap1D/gap2D/gap6D strains since ScPut4 and ScGap1 mediate proline uptake in S. cerevisiae. As expected, all strains lacking GNP2 were strongly impaired in growth on solid (Fig. 3A) or liquid (Fig. S2C) medium with proline as the sole carbon source, whereas the other AAP deletion strains and the GNP2 complemented strain possessed wild-type-like growth. To exclude the presence of additional proline transporters, potentially ammonium repressed, we tested growth on proline as the sole carbon and nitrogen source. Again, only the gnp2D strain displayed absent growth (Fig. S2C). Since in S. cerevisiae, Put4 is also involved in the uptake of g-aminobutyric acid (GABA), we likewise examined all strains for growth on GABA as the sole carbon source (40). Interestingly, only strains lacking GNP2 displayed impaired growth, more pronounced in liquid medium, whereas the put4D strain was not affected (Fig. S3).
Finally, to verify that Gnp2 is indeed the main proline permease, we measured proline uptake in SC5314, the gnp2D strain, and the GNP2 complemented strain in liquid medium containing proline as the sole carbon source. After 6 h of incubation, the extracellular proline concentration was significantly reduced in the wild type and the GNP2 complemented strain compared to the medium control but not in the gnp2D strain, proving the absence of proline uptake in this strain (Fig. 3B). Altogether, our results highlight the essential role of Gnp2 for proline uptake in C. albicans and support the notion of divergent AAP functions between this pathogenic fungus and S. cerevisiae.
The functional difference of GNP1 and GNP2 is reflected in their transcriptional regulation. Since our findings clearly show that Gnp1 and Gnp2 have distinct substrate specificities, we next examined the expression patterns of the respective genes and GNP3 in order to detail their regulation on the transcriptional level. Given that the S. cerevisiae orthologs are regulated in either an SPS (ScAGP1 and ScGNP1)-or an NCR (ScGAP1 and ScPUT4)-dependent manner, we used different SPS-inducing (either glutamine or a mixture of all proteinogenic amino acids [synthetic complete [SC] medium] as the sole C source) and noninducing (either glucose [SD medium] or proline as the sole carbon source) growth conditions. To test for NCRmediated regulation, we also measured expression in the absence of ammonium sulfate The proline concentration was normalized to an internal standard (10 mM 13C5 L-proline). Measurement of the medium control was performed in technical triplicates, and measurement of the spent medium was performed in five biological replicates. Significance was tested via one-way ANOVA with the medium control as the reference, indicated by asterisks (***, P , 0.001; ****, P , 0.0001).
Gnp2-Mediated Proline Uptake in Candida albicans ® (proline or SC medium as the sole source of both nitrogen and carbon) and with proline as the sole source of nitrogen.
Under all tested conditions, the expression of GNP3 was constant and relatively low (Fig. 4). In contrast, GNP1 had a clear SPS-dependent expression pattern: in the wild type, the gene was induced in response to SC medium and glutamine (SPS inducing) compared to the SD control and remained unchanged in the presence of proline (SPS noninducing) or in the stp2D mutant under all conditions tested (Fig. 4). The strict SPSdependent regulation of GNP1 was additionally verified in a strain carrying an N-terminally truncated and, thus, constitutively active Stp2 variant (STP2tOE), where we noted constant induction compared to the wild type (Fig. S4A). Furthermore, GNP1 was expressed at similar levels in SC medium without ammonium sulfate (Fig. S4B).
The regulation of GNP2 appeared more complex. First, the availability of ammonium sulfate did not influence the GNP2 expression pattern in response to proline or SC medium, demonstrating that the expression of this gene falls outside NCR control ( Fig. 4; Fig. S4B). Next, in SC5314 cells exposed to a mixture of amino acids (SC medium), GNP2 was strongly induced compared to the amino acid-free SD medium, which was also observed in stp2D cells although to a lesser degree (Fig. 4). Therefore, we concluded SPS-independent induction of GNP2, which is further supported by the fact that glutamine alone failed to induce GNP2 expression (Fig. 4). Surprisingly, the presence of proline as the sole source of carbon had no influence on GNP2 expression levels in the wild type, while strong upregulation was visible in cells deleted of STP2 (Fig. 4). In the presence of proline as the sole nitrogen source, we observed induction of GNP2 in the wild type, which was further increased in the stp2D strain (Fig. S4B). Based on these observations, we hypothesized a repressing activity of Stp2 on the expression of GNP2 under proline-rich conditions. Indeed, the expression of GNP2 was similar to that of the wild type in the strain carrying the artificially activated Stp2 (STP2tOE), confirming that Stp2 negatively regulates GNP2 expression in C. albicans in response to proline (Fig. S4A).
Recently, the transcription factor Put3 was described as a positive regulator of GNP2 and the proline catabolism genes PUT1 and PUT2 (38). Indeed, we observed a significant increase of GNP2 expression in a strain carrying an artificially activated variant of Put3 (PUT3-GAD) (Fig. S4A) (38,41). However, no changes compared to the wild type were observed in SD medium, indicating the necessity of proline availability for this induction (Fig. S4A). Finally, we examined the expression profiles of the proline catabolism genes for potentially shared features in the regulation of proline uptake and catabolism. Interestingly, the expression patterns of both genes in response to proline resembled that of GNP2, displaying FIG 4 Expression of GNP1 and GNP2 is nutrient source and genotype dependent. YPD log-phase cells of the SC5314 wild type and the stp2D mutant were transferred to the indicated media (YNB and ammonium sulfate supplemented with either glucose, proline, or glutamine or SC medium as the sole carbon source) and incubated for 1 h at 37°C. Expression levels were measured in technical and biological triplicates, and MED15 expression was used as the internal reference. Relative expression was calculated according to the DC T method. Significance was tested via two-way ANOVA between the wild type and the stp2D mutant, indicated by asterisks (****, P , 0.0001), and between test media and the SD medium reference for every strain, indicated by pound signs (#, P , 0.0001). strong induction in the stp2D and PUT3-GAD strains (Fig. S4C). Consequently, no induction was observed for both genes in the wild-type or STP2tOE strain in proline compared to SD medium, confirming the repressive role for Stp2 in proline utilization (Fig. S4C).
GNP2 is essential for proline-induced hyphal morphogenesis in C. albicans. A recent report showed that mitochondrial catabolism of proline acts as an inducer of C. albicans morphogenesis and is directly linked to fungal escape from macrophages (28). Thus, we examined if Gnp2-dependent uptake is also required for proline-induced filamentation in vitro.
Therefore, proline-induced filamentation was examined under previously reported conditions (28). As expected, the gnp2D strain remained in the yeast form, while all other tested strains displayed prominent filamentation (Fig. 5). Indeed, all strains rapidly formed dense hyphal aggregates except the gnp2D strain, which, even after prolonged incubation, displayed only sparse flocculation (Fig. S5A). Quantification of germ tube formation after a short incubation under this condition showed an absent filamentation response of the gnp2D strain to proline (Fig. S6). When proline was replaced with arginine, an amino acid that feeds into the same pathway, the gnp2D strain displayed robust filamentation indistinguishable from those of the other tested strains (Fig. S5B). Thus, we conclude that Gnp2-mediated proline uptake is required for filamentation.
Gnp-mediated amino acid uptake is required for cytotoxicity and fungal survival during macrophage confrontation. To test the role of the Gnp1/2/3 permeases in the interaction with macrophages, we confronted strains lacking these permeases with J774 macrophages and evaluated fungal survival and the capacity to damage macrophages. Surprisingly, all three single-deletion strains were impaired in their ability to damage these immune cells, with the gnp1D strain being most affected (Fig. 6A). No additive effects were observed with strains carrying multiple deletions since the gnp1D/gnp2D, gnp1D/gnp3D, and gnp1D/gnp2D/gnp3D strains displayed damage potential similar to that of the gnp1D single mutant. Similarly, the gnp2D/gnp3D strain caused macrophage damage equally as well as the respective single-deletion strains. Of note, all tested strains displayed robust growth and filamentation in the corresponding medium (Fig. S7).
The reduced ability of the permease mutants to damage J774 macrophages was also reflected in the impaired fungal survival following confrontation with these immune cells. However, some notable differences were observed (Fig. 6B). For both the gnp1D and gnp2D strains, survival was about 60% of that of the wild-type control. In the gnp3D strain, survival was also reduced albeit not significantly. However, the gnp1D/gnp2D strain showed an even further decrease in survival after macrophage confrontation (;35%), whereas the gnp1D/ gnp3D and gnp2D/gnp3D strains had survival rates comparable to those of strains with deletions of either GNP1 or GNP2, respectively. Oddly, the triple-deletion strain displayed higher survival rates than the gnp1D/gnp2D strain (Fig. 6B). Altogether, these results demonstrate an important role of Gnp1-and Gnp2-mediated amino acid uptake for C. albicans-macrophage interactions. Gnp2-Mediated Proline Uptake in Candida albicans ® Gnp2-mediated proline uptake is beneficial for C. albicans ROS resistance. The reduced capacity to damage macrophages and to survive the immune cell confrontation of GNP deletion strains was not linked to filamentation defects under these conditions since all strains were equally capable of hypha formation upon phagocytosis (data not shown). Macrophages use a variety of mechanisms to ensure restricted growth and the eradication of the engulfed microbes, including the oxidative burst (42). Thus, we examined if Gnp1/2/3 are required for fungal adaptation to reactive oxygen species (ROS).
The ROS inducer tert-butyl hydroperoxide (t-BOOH) was added to two different media: SC medium containing all proteinogenic amino acids and SD minimal medium without any amino acids as the control. As expected, when no amino acids were present, none of the AAP deletion strains exhibited increased sensitivity to ROS stress compared to the wild type (Fig. 7). However, in SC medium, the gnp2D strain displayed markedly increased sensitivity to t-BOOH, a phenotype similar to that of the stp2D strain, indicating a beneficial effect of amino acid uptake for the fungus when exposed to ROS stress. Interestingly, the gnp1D/ gnp2D and gnp1D/gnp2D/gnp3D strains showed further increases in sensitivity, whereas the gnp1D and gnp3D strains remained unaffected. Thus, amino acid transport via Gnp2 and, to a lesser degree, Gnp1 appears important for the adaptation of C. albicans to oxidative stress.
Gnp-mediated amino acid uptake is not critical for full virulence in a murine model of disseminated candidiasis. To assess whether Gnp1/2/3-mediated amino acid uptake release from macrophages was measured. The damage capacity of different C. albicans strains was calculated relative to the high control (complete macrophage lysis through treatment with 5% Triton for 10 min). Significance was tested via one-way ANOVA against the wild type (*, P , 0.05; **, P , 0.01; ****, P , 0.0001). The experiment was performed in technical triplicates and seven biological replicates. (B) C. albicans colonies were counted after 24 h of coincubation with macrophages as the readout for fungal survival. Survival was calculated as a ratio relative to the wild type. Significance was tested via one-way ANOVA. The experiment was performed in four biological replicates. is required for full fungal virulence in vivo, we tested the gnp1D, gnp2D, and gnp3D strains; the corresponding revertant strains; and the gnp1D/gnp2D/gnp3D strain in a murine model of disseminated candidiasis. Surprisingly, despite the prominent in vitro phenotypes, no significant reduction in virulence was seen (Fig. 8). However, a slight trend toward increased survival was visible for mice infected with the gnp1D, gnp2D, and gnp3D single mutants.

DISCUSSION
Metabolic flexibility and the ability to quickly react to environmental changes are two closely intertwined characteristics of paramount importance for C. albicans and its successful lifestyle as an opportunistic pathogen (8,9). A key role in this context is the utilization of amino acids, which affects fungal virulence in various ways (21). In this work, we focused on AAP-mediated uptake of amino acids and especially proline, a particularly good source of energy, as the oxidative phosphorylation of 1 mol proline yields approximately 30 mol ATP, nearly equivalent to glucose (43). While proline promotes C. albicans growth not only as the sole nitrogen source but also as a source of carbon, in S. cerevisiae proline utilization is limited to nitrogen utilization only (38,44). This substantial difference suggests that the mechanisms of uptake and cellular fate of this amino acid differ between the two microorganisms. While the AAPs involved in S. cerevisiae proline uptake are well defined, the sensing and transport of extracellular proline in C. albicans have not been studied.
Here, we demonstrate that C. albicans possesses 10 orthologs of the four prolinetransporting AAPs in S. cerevisiae (ScAgp1, ScGnp1, ScGap1, and ScPut4). These orthologs are involved in the uptake of various amino acids, but only Gnp2 mediates proline uptake in C. albicans. Indeed, the C. albicans genome encodes six ScGap1 orthologs, but the deletion of none of them affected the sensitivity to the proline analog (30). The strain with the triple deletion of GAP1, GAP2, and GAP6 was also not impaired in growth on proline, although Gap2 and Gap6 were previously reported to complement proline utilization in an uptake-deficient S. cerevisiae strain (30). Furthermore, Gnp1 appears to play an important role in the uptake of asparagine, glutamine, tryptophan, and, to a lesser extent, threonine and methionine. The physiological relevance of Put4 remains unclear since a C. albicans strain lacking PUT4 was not impaired in proline but likely involved in lysine uptake. Moreover, Put4 appeared dispensable for the uptake of GABA, unlike ScPut4 (40). Instead, this function is mediated by Gnp2, which is required for growth on GABA as the sole carbon source, while it was expendable during growth on GABA as the sole nitrogen source (data not shown). Taken together, our results show that Gnp1 and Gnp2 are functionally distinct in C. albicans: Gnp2 is a specialized proline and GABA transporter, whereas Gnp1 has broader substrate specificity, comparable to those of ScAgp1 and ScGnp1 (45)(46)(47)(48)(49).
This specialization is further exemplified by the diverged transcriptional regulation of GNP1 and GNP2. While ScAGP1 and ScGNP1 are both regulated in an SPS-dependent manner, in C. albicans, only GNP1 is positively regulated by Stp2 (37). This correlates well with the observed substrate specificity for Gnp1, which, aside from tryptophan, coordinates the uptake of only SPS-stimulating amino acids (28). In contrast, the regulation of GNP2 is more complex. Since it was strongly induced in response to a mixture of amino acids in the wild-type and stp2D strains, although slightly lower, we concluded mainly SPS-independent induction. However, GNP2 was strongly induced in stp2D cells in the presence of proline as the sole carbon source. This observation was quite unexpected given that under this condition, the transcription factor Stp2 is considered to be retained in the cytosol (28,33). Furthermore, GNP2 was not induced in a strain with a constitutively active variant of Stp2. Interestingly, the same observations were made for the proline catabolism genes PUT1 and PUT2, which shared the GNP2 expression pattern under all tested conditions, suggesting that proline uptake and catabolism are similarly regulated. These results strongly point toward Stp2-mediated repression of proline utilization, the underlying mechanisms of which remain unclear. One possible connection is Dal81, a transcription factor involved in GABA utilization in C. albicans, shown to interact with full-length and processed Stp2 in S. cerevisiae to enhance its transcriptional activity (50,51). Another factor is Put3, previously described as a positive regulator of proline utilization in C. albicans (38). Indeed, we observed that a strain carrying an artificially activated variant of Put3 had significantly elevated transcript levels of PUT1, PUT2, and GNP2 in the presence of proline. While the dependence of proline availability for Put3 activity is a well-established mechanism for S. cerevisiae, the potential cross talk of Stp2-and Put3-mediated gene regulation in C. albicans requires further investigation (52).
The outstanding nature of proline uptake in C. albicans inevitably prompts the question of why such specialization has evolved. Since proline is a valuable source of carbon and nitrogen for this pathogenic fungus, a simple explanation would be to allow rapid adaptation to the proline-rich niches of the human host. This amino acid is especially enriched in the extracellular matrix in the structural protein collagen, composed of up to 25% proline/hydroxyproline (53). There, it might serve as an easily accessible energy source during invasive growth. Additionally, proline catabolism is directly linked to pathogenicity, as it triggers hyphal growth and is required for the capacity to damage macrophages (28). Accordingly, we found that proline-induced filamentation in vitro is dependent on Gnp2. However, gnp2D cells were fully capable of forming filaments within macrophage phagosomes. It is feasible that the uptake of arginine or ornithine compensates for the impaired ability of the gnp2D strain to import proline during phagocytosis, resulting in the activation of the same filamentation pathway (18,27,28).
Despite the retained ability to filament within the macrophage phagosome, the strains deleted of GNP2 were impaired in damaging these immune cells and had impaired survival following macrophage confrontation. Interestingly, previous studies also reported that GNP2 is induced following confrontation with neutrophils or macrophages (22,23). Since these immune cells confront C. albicans with a combination of factors, this could indicate the physiological relevance of proline uptake in adaptation to host-imposed stresses. This notion is supported by the significantly increased sensitivity to oxidative stress of strains lacking GNP2. Indeed, proline plays a prominent role in the stress adaptation of many organisms ranging from humans to plants and microbes (43). Due to its special structure, this amino acid fulfills a variety of functions: it acts not only as a direct ROS scavenger, detoxifying hydroxyl radicals or singlet oxygen, but also as a chemical chaperone stabilizing redox enzymes. It further contributes to the cellular glutathione pool, which is essential in detoxifying ROS (43). While the role of prolinemediated stress adaptation has not been investigated in C. albicans, functional proline catabolism was previously described as being essential for ROS homeostasis and virulence in Cryptococcus neoformans (54). Furthermore, the proline-mediated stress response is thoroughly studied in the model yeast S. cerevisiae (55,56). Although the underlying mechanisms are not fully understood so far, proline homeostasis and forced accumulation are beneficial in resistance to environmental stresses like oxidation, desiccation, and freezing, partially due to the positive influence on membrane and protein stability.
In this context, the importance of Gnp2 in GABA utilization is especially interesting. The GABA shunt is a highly conserved pathway in most organisms and closely intertwined with proline metabolism and the TCA cycle. In S. cerevisiae this pathway was shown to be involved in thermotolerance by reducing ROS production and in the regulation of the replicative life span (57,58). GABA accumulation also accompanies and influences stress responses in several plants (59). The physiological importance of GABA in C. albicans remains poorly investigated, with reports linking Gpt1, a putative GABA/polyamine transporter, and the transcription factor Uga3 to the utilization of this host-relevant nutrient (51,60). However, as dual-specificity GABA/proline transporters have been previously reported for other microorganisms, for instance, Bacillus subtilis, it is feasible that similar transport systems could exist in other pathogenic organisms like C. albicans (61).
Despite the impairments in growth, filamentation, and macrophage interactions, we observed no significant reduction in the virulence of GNP1/2/3 deletion strains in a murine model of disseminated candidiasis. This suggests that either certain compensatory mechanisms or residual unspecific proline uptake is sufficient to support the growth and pathogenicity of the strains in this infection model. Another possibility is that proline uptake is not critical for systemic infection but rather more relevant during the commensal growth of C. albicans. In its commensal niches, like the gastrointestinal tract, the fungus is subjected to mutualistic and competitive interactions with the associated microbiota. Recently, the metabolic interactions of C. albicans and gut bacteria were modeled using genome-scale metabolic models, revealing that proline is one of the few metabolites readily consumed by bacteria, which negatively influences fungal growth (62). Since GNP2 was found to be induced during C. albicans colonization of the gastrointestinal tract, the presence of a high-specificity proline permease might provide a competitive edge in this environment (63).
In conclusion, our results clearly demonstrate the presence of specialized proline uptake in C. albicans and its relevance for fungal physiology. The precise reason for this specialization, its advantages over redundant uptake, and the underlying mechanisms for sensing and regulation require further investigations. These results, together with the role of GABA-mediated cellular processes, should lead to a more comprehensive understanding of C. albicans pathogenesis.

MATERIALS AND METHODS
C. albicans cultivation. All C. albicans strains used in this study are listed in Table S3A in the supplemental material. Unless indicated otherwise, C. albicans was routinely cultured in yeast extract-peptonedextrose (YPD) medium (1% yeast extract, 2% peptone, and 2% glucose, solidified with 2% agar for plates) at 30°C. Precultures were grown overnight with shaking at 180 rpm and washed twice with sterile distilled water (dH 2 O) prior to subsequent experiments.
Strain construction. The C. albicans wild-type strain SC5314 was used for all gene deletions. For the generation of AAP mutants, a modified CRISPR-Cas9 approach was used to simultaneously delete both alleles of the gene of interest (64). Briefly, Cas9 was PCR amplified from pV1093. Specific guide RNAs were chosen and PCR amplified to be expressed under the control of the SNR52 promoter. Repair constructs containing the SAT1-FLP1 cassette were PCR amplified from pSFS2 (65), using long primers with 50-bp homologous flanking fragments specific to the target gene. All PCR products were column purified, and 500 ng of each fragment (Cas9, the gene-specific single guide [sgRNA], and the gene-specific repair construct) was used to electroporate C. albicans (66). Transformants were selected on YPD-nourceothricin (YPD-Nat) medium, and the correct gene deletion was confirmed by PCR. The correct clones were rendered Nat sensitive by growing them overnight on yeast extract-peptone-maltose (YPM) (2% maltose) to excise the SAT1-FLP1 cassette. For complementation, the gene of interest was amplified from genomic DNA and fused by overlap extension PCR to the SAT1 product from pSFS2, including 50 bp of homology upstream and downstream of the target gene. A guide RNA Gnp2-Mediated Proline Uptake in Candida albicans ® targeting the FLP recombination target (FRT) scar left behind after SAT1-FLP1 excision was PCR amplified. The mutant strain to be complemented was electroporated with 500 ng of each PCR fragment (Cas9, FRT sgRNA, and the target gene fused to SAT1). Transformants were selected on YPD-Nat, and the correct gene complementation was confirmed by PCR. The generation of a constitutively active Stp2, through truncation of amino acid residues 2 to 100, was performed analogously to the previously described generation of the STP2 overexpression strain (67). The truncated variant of STP2 was synthesized by Genewiz.
Growth assays. (i) Solid medium. C. albicans cultures grown overnight were set to an optical density at 600 nm (OD 600 ) of 0.5 in sterile dH 2 O, and 5 mL was spotted per strain. All plates contained 0.17% yeast nitrogen base (YNB) without ammonium sulfate and amino acids and 2% purified Bacto agar (BD, NJ, USA). For growth on defined carbon sources (10 mM Pro or GABA), the plates were supplemented with 0.5% ammonium sulfate. The plates were incubated at 37°C for 2 to 3 days, and images were taken with a ProtoCol2 colony counter (Synbiosis, UK).
(ii) Liquid medium. Growth assays in liquid medium were performed in 96-well microplates using an Infinite M Nano microplate reader (Tecan, Switzerland). If not indicated otherwise, C. albicans cultures grown overnight were set to an OD 600 of 2 in sterile dH 2 O, and 2 mL was added to 198 mL test medium per well for a final OD of 0.02. The OD 600 was subsequently measured every 20 min for 24 h or 72 h at 37°C with 10 s of shaking prior to each measurement. All measurements were performed in technical triplicates and biological duplicates. Growth was measured as the area under the curve (AUC) (GraphPad Prism 8.4.3, trapezoidal method; baseline at 0.02).
To assess growth on single amino acids as the sole nitrogen source, base medium composed of 0.17% YNB and 2% glucose was supplemented with 5 mM the indicated nitrogen source, except for tyrosine and ammonium sulfate, with 1.5 mM and 0.5%, respectively. The initial pH was adjusted to 5. The growth of SC5314 on single amino acids was calculated as a ratio of the growth on ammonium sulfate. The growth of AAP deletion strains was calculated as the ratio of SC5314 growth under the respective conditions.
Resistance to toxic amino acid analogs (Sigma-Aldrich, MO, USA) (Table S2A) was monitored through their addition to SD medium (0.17% YNB, 0.5% ammonium sulfate, 2% glucose [pH 5]). Growth in the presence of toxic analogs was calculated relative to growth in SD medium alone. Due to the various degrees of SC5314 growth inhibition through the individual analogs, the growth of the particular AAP deletion strains was calculated as a percentage of wild-type inhibition where the effect for each analog on the wild type is set to 100%. As a consequence, 100% wild-type inhibition for an AAP deletion strain indicates growth inhibition similar to that of SC5314, while 0% wild-type inhibition indicates the unaffected growth of an AAP deletion strain in the presence of an analog. The following formula was used for calculation: Quantification of proline uptake. C. albicans proline uptake was quantified as follows. C. albicans cultures grown overnight were used to inoculate 10 mL test medium (0.17% YNB, 0.5% ammonium sulfate, 10 mM proline [pH 4.5]), which was subsequently incubated at 37°C for 6 h. One milliliter of culture per sample was sampled, and after centrifugation, 100 mL of the supernatant was supplemented with 13C5 L-proline (Eurisotop, MA, USA) for a final concentration of 10 mM as the internal reference.
Ultrahigh-performance liquid chromatography (UHPLC) coupled with high-resolution mass spectrometry (MS) was carried out using a Thermo (Bremen, Germany) UltiMate HPG-3400 RS binary pump and a WPS-3000 autosampler, which was set to 10°C and was equipped with a 25-mL injection syringe and a 100-mL sample loop. The column was kept at 25°C within the column compartment of the TCC-3200 system. A Thermo Accucore C 18 reverse-phase (RP) chromatography column (100 by 2.1 mm, 2.6 mm) was used, using the gradient listed in Table S3C, at a constant flow rate of 0.4 mL/min. Eluent A was water with 2% acetonitrile and 0.1% formic acid. Eluent B was pure acetonitrile.
Mass spectra were recorded with a Thermo QExactive plus orbitrap mass spectrometer coupled to a heated electrospray ionization (HESI) source. The column flow was switched at 0.5 min from waste to the mass spectrum and at 11.5 min again back to the waste, to prevent source contamination. For monitoring, two full-scan modes were selected with the following parameters: positive polarity, a scan range of m/z 80 to 1,200, a resolution of 70,000, an automatic gain control (AGC) target of 3 Â 10 6 , and a maximum ion time of 200 ms. General settings were as follows: a sheath gas flow rate of 60, an auxiliary gas flow rate of 20, a sweep gas flow rate of 5, a spray voltage of 3.0 kV, a capillary temperature of 360°C, an S-lens radio frequency level of 50, an auxiliary gas heater temperature of 400°C, and an acquisition time frame of 0.0 to 11.5 min. For negative mode, all values were kept instead of the spray voltage, which was set to 3.3 kV.
Peak detection and integration were carried out using Thermo Scientific Xcalibur 3.0.63 Quan Browser software with the following settings: a mass tolerance of 5 ppm, a mass precision of 4 decimals, and the compounds proline (calculated for C 5 H 10 NO 2 as [M 1 H] 1 of 116.0706) and labeled proline (calculated for 13 C 5 H 10 NO 2 as [M 1 H] 1 of 121.0874). A retention time of 0.66 min was set, including the window, which was set to 30 s; the signal was the extracted ion chromatogram (XIC) from the positive full-scan experiment; the peak detection algorithm was ICIS (smoothing = 1; baseline window = 40; area noise factor = 5; peak noise factor = 10); and the peak detection method was the highest peak, with a minimum peak height (signal-to-noise [S/N] ratio) of 3. Area ratios of each analyte relative to the internal standard (IS) were determined, and the analyte concentration of each sample was calculated relative to the average of the medium control. For statistical analysis, Microsoft Excel 365, version 2108, was used. Error bars were calculated using the confidence interval with 95% probability. Significance was tested via one-way analysis of variance (ANOVA) using GraphPad Prism 8.4.3 with the medium control as the reference.
Filamentation experiments. Proline-and arginine-or Dulbecco's modified Eagle's medium (DMEM)induced filamentation in liquid medium was analyzed as follows: C. albicans cultures grown overnight were inoculated into 30 mL of test medium (0.17% YNB, 0.2% glucose, and 10 mM proline or arginine [pH 6] or DMEM [catalog number 41965; Thermo Fisher Scientific]) at a final OD of 0.3 and incubated at 37°C with shaking at 180 rpm (28). Samples were fixed at the indicated time points with 4% Histofix (Carl Roth, Germany), and morphology was assessed via differential interference contrast (DIC) imaging with a Zeiss AxioObserver Z.1 microscope. The experiments were performed in biological triplicates, and exemplary pictures are shown. For the quantification of proline-induced germ tube formation after 60 min, a minimum of 185 cells per strain from one representative experiment were counted and classified as either germ tube or yeast, and the relative proportions of both groups were calculated.
RNA extraction and qRT-PCR. Gene expression levels were measured via quantitative reverse transcription PCR (qRT-PCR) under the following growth conditions: SD medium with proline or glutamine as the sole carbon source (0.17% YNB, 0.5% ammonium sulfate, 10 mM proline or glutamine [pH 5]), proline as the sole carbon source without ammonium sulfate, or proline as the sole nitrogen source (0.17% YNB, 5 mM proline, 2% glucose [pH 5]) and SC medium with or without ammonium sulfate (0.17% YNB with or without 0.5% ammonium sulfate and 1% complete SC medium without ammonium sulfate [Biomol, Germany] [pH 5]). C. albicans cultures grown overnight were diluted to an OD of 0.3 in 50 mL YPD and incubated at 37°C with shaking at 180 rpm until early log phase was reached (OD 600 = 1). Cells were collected, washed twice with dH 2 O, and used to inoculate 10 mL test medium at an OD of 0.3. The cultures were incubated at 37°C for 60 min, and cells were harvested, snap-frozen in liquid nitrogen, and stored at 280°C until further processing. For total RNA isolation, cells were resuspended in 440 mL acetate EDTA buffer (50 mM sodium acetate, 10 mM EDTA) with 10% SDS, and 440 mL of acid-phenol-chloroform with isoamyl alcohol (Ambion, Germany) was added. The suspension was thoroughly mixed, incubated for 5 min at 65°C, frozen at 220°C for 30 min, and subsequently thawed at 65°C for 5 min. The phases were separated via 4 min of centrifugation at 16,000 Â g, the upper aqueous phase was transferred, and a 1/10 volume of sodium acetate (pH 5.3) and 1 volume of 2-propanol were added to precipitate the RNA for at least 30 min at 220°C. The supernatant was discarded after 10 min of centrifugation at 12,000 Â g, and the RNA pellet was washed twice with 80% ethanol and finally solved in 300 mL nuclease-free H 2 O. The quality and quantity of the RNA were assessed using 2100 bioanalyzer (Agilent Technologies, CA, USA) and Nanodrop (Thermo Fisher Scientific, MA, USA) measurements. qRT-PCR was performed using Brilliant III Ultra-Fast SYBR green qRT-PCR master mix (Agilent Technologies) on a Stratagene Mx3005P system (Agilent Technologies) with 100 ng/mL RNA as the template. The expression levels were calculated according to the DC T method using MED15, a subunit from the RNA polymerase II mediator complex, as the internal reference (68). The oligonucleotides used are listed in Table S3B.
Macrophage infection assays. Measurements of fungal macrophage cytotoxicity and survival were performed as previously described (69). Solely, the survival rates of AAP deletion strains were calculated relative to the wild type as follows: Mouse model of disseminated candidiasis. C. albicans strains were grown in YPD to log phase, washed, and resuspended in phosphate-buffered saline (PBS). A total of 2 Â 10 5 cells were injected into the lateral tail vein of a total of 10 female 7-to 9-week-old ICR mice per strain. All mice were monitored 3 times per day and euthanized when moribund. All experiments were approved by the Animal Welfare Committee of the University of Texas Health Science Center at Houston. Survival curves were tested for significant differences via a log rank (Mantel-Cox) test.

SUPPLEMENTAL MATERIAL
Supplemental material is available online only.