Roles of Candida albicans Mig1 and Mig2 in glucose repression, pathogenicity traits, and SNF1 essentiality.

Metabolic adaptation is linked to the ability of the opportunistic pathogen Candida albicans to colonize and cause infection in diverse host tissues. One way that C. albicans controls its metabolism is through the glucose repression pathway, where expression of alternative carbon source utilization genes is repressed in the presence of its preferred carbon source, glucose. Here we carry out genetic and gene expression studies that identify transcription factors Mig1 and Mig2 as mediators of glucose repression in C. albicans. The well-studied Mig1/2 orthologs ScMig1/2 mediate glucose repression in the yeast Saccharomyces cerevisiae; our data argue that C. albicans Mig1/2 function similarly as repressors of alternative carbon source utilization genes. However, Mig1/2 functions have several distinctive features in C. albicans. First, Mig1 and Mig2 have more co-equal roles in gene regulation than their S. cerevisiae orthologs. Second, Mig1 is regulated at the level of protein accumulation, more akin to ScMig2 than ScMig1. Third, Mig1 and Mig2 are together required for a unique aspect of C. albicans biology, the expression of several pathogenicity traits. Such Mig1/2-dependent traits include the abilities to form hyphae and biofilm, tolerance of cell wall inhibitors, and ability to damage macrophage-like cells and human endothelial cells. Finally, Mig1 is required for a puzzling feature of C. albicans biology that is not shared with S. cerevisiae: the essentiality of the Snf1 protein kinase, a central eukaryotic carbon metabolism regulator. Our results integrate Mig1 and Mig2 into the C. albicans glucose repression pathway and illuminate connections among carbon control, pathogenicity, and Snf1 essentiality.

Introduction

Carbon metabolism is central to the growth and survival of all organisms. It provides both energy and biosynthetic building blocks. It is tightly controlled in most organisms to enable optimal use of diverse carbon sources. The ability to adapt to changing carbon sources is especially important for commensal and pathogenic microbes because microbial competitors and host factors can cause dynamic changes in the spectrum of carbon compounds available [1, 2].

Our focus is the fungus Candida albicans. It exists primarily as a commensal resident of the GI and GU tracts of humans and other warm-blooded animals. However, upon dysbiosis of the host environment it can cause infections that include oropharyngeal candidiasis, cutaneous candidiasis, vaginal candidiasis, and systematic or intra-abdominal candidiasis stemming from colonization of the patient’s own GI tract [2, 3]. The ability of C. albicans to cause infection of diverse tissues and body sites depends upon its ability to regulate the utilization of diverse carbon sources [4].

Many of the mechanisms that govern carbon source utilization and regulation have been studied using the yeast Saccharomyces cerevisiae [5]. The extensive research from this model organism has been a useful guide for gene function analysis because genetic studies are more intractable in C. albicans. Additionally, this comparison is interesting because C. albicans is a human pathogen, so we might expect its regulation of metabolism and carbon source utilization to be different than in S. cerevisiae, a pathogen only on rare occasion. In fact, there are several unique features of C. albicans carbon regulation, such as distinctive transcriptional activators of glycolysis and alternative carbon source utilization [6, 7], loss of glucose-mediated catabolite inactivation [8], and loss of glucose-responsive post-translational modifications of the regulatory kinase, Snf1 [9]. Some differences in metabolic regulation have been directly linked to virulence [8], a connection that may inform new therapeutic strategies [4].

One form of metabolic regulation is called "glucose repression" or "carbon catabolite repression" [2, 4]. In the presence of glucose, a preferred carbon source, expression of genes for use of alternative (i.e., non-glucose) carbon sources is repressed. In S. cerevisiae, a central regulator of this pathway is the protein kinase ScSnf1, also known as the AMP-activated protein kinase. ScSnf1 is highly conserved among eukaryotes, and it functions to integrate diverse signals including metabolic and environmental changes like glucose restriction, oxidative stress, and alkaline pH [10].

ScSnf1 is activated by the upstream protein kinase ScSak1 (Snf1 Activating Kinase) and two paralogs, ScElm1 and ScTos1. ScSnf1 is required for expression of glucose-repressed genes [11], a requirement that is mediated by the transcriptional repressor ScMig1 (Multicopy Inhibitor of AL gene expression). In response to glucose, ScSnf1 is dephosphorylated, resulting in the activation of ScMig1 which represses expression of ScSnf1-dependent genes [12, 13]. The paralog ScMig2 functions mainly to augment repression of a subset of the glucose-repressed genes controlled by ScMig1 [13, 14]. Although the activation of ScMig1 is directly controlled by ScSnf1, ScMig2 has been shown to function independently of ScSnf1 [13], is transcriptionally regulated by another pathway [15], and is post-translationally modified by ubiquitination in response to carbon source [16]. In contrast, ScMig1 is regulated by carbon-responsive nuclear-cytoplasmic shuttling [17].

Elements of the Snf1 glucose repression pathway are conserved in C. albicans and in many other fungi. In C. albicans, Sak1 is required for utilization of alternative carbon sources, cell wall and membrane stress resistance, and expression of hexose transporter genes and glyoxylate cycle and gluconeogenesis genes [9]. Sak1 also influences production of long filamentous chains of cells called hyphae in a medium-dependent manner. C. albicans Snf1 clearly functions downstream of Sak1: it undergoes Sak1-dependent phosphorylation, and a strain that expresses a nonphosphorylatable Snf1-T208A mutant protein is unable to grow on alternative carbon sources [9]. Among downstream components, C. albicans, Mig1 is functionally conserved with ScMig1 as a repressor of several carbon utilization genes [18-20]. However, a mig1Δ/Δ mutation does not affect expression of most glucose-repressed genes [20]. Therefore, additional mediators of glucose repression in C. albicans have yet to be established.

One surprising feature of Snf1 function in C. albicans is that Snf1 is required for viability, unlike the situation in S. cerevisiae [9, 21–24]. Homozygous snf1 null mutations have not been recovered through transformation or selection for mitotic recombinants [21, 22], methods that readily yield homozygous mutations in nonessential genes. SNF1 essentiality has been most clearly established with a strain homozygous for a fusion of SNF1 to a conditional promoter; the strain is able to grow only when the promoter is active [24]. Viable snf1 mutations with altered function or expression [24-26] have permitted some phenotypic analysis, but it remains unclear whether the essential role of Snf1 depends upon its activity in the glucose repression pathway, or whether Snf1 has a novel second role that is essential [21].

Here we investigate the circuitry that drives the transcriptional response of C. albicans to glucose and the non-fermentable carbon source glycerol. Our findings reveal that Mig1 functions together with its paralog Mig2 to mediate glucose repression. Mig1 and Mig2 have both selective and shared functions in this glucose repression response. Each repressor also governs aspects of the snf1Δ/Δ and sak1Δ/Δ mutant phenotypes, thus tying them to the Snf1 pathway. Finally, our results argue that SNF1 is essential because of exuberant repression by Mig1, thus connecting Snf1 essentiality to its role in the glucose repression pathway.

Results

Carbon control of gene expression

To investigate C. albicans gene expression response to carbon, we compared cells growing on the non-fermentable carbon source glycerol to those growing on glucose. RNA-sequencing (RNA-seq) was performed on wild-type C. albicans in YPG (Yeast Peptone Glycerol) and YPD (Yeast Peptone Dextrose) media grown to log phase at 37°C. A total of 489 genes were significantly altered (>2-fold change in RNA levels and p<0.05) in YPG compared to YPD (S1 Table). Among those, 375 genes (77%) were up-regulated in YPG. These genes were enriched for processes that include organic acid catabolism, fatty acid catabolism, and hexose transport (ex. CTN1, ICL1, MLS1, FOX2, HGT1, GAL1; see S2 Table). The remaining 114 genes were down-regulated in YPG. These genes were enriched for processes that include carbohydrate metabolism and glycolysis (ex. ENO1, GPM1, CDC19, TDH3, PFK1, PFK2, HXK2; see S2 Table). Overall, the gene expression differences aligned well with known or expected features of cell metabolism.

We compared the differentially expressed genes in our YPG vs. YPD dataset to published gene expression datasets (Fig 1A and S3 Table). We focused on datasets in which both up- and down-regulated genes overlap with our dataset significantly (p<1E-07; Fisher’s Exact Test). We observed significant similarity to datasets intended to probe glucose repression, including growth in YEP lactate vs YEP lactate + glucose (0.01, 0.1, or 1%), incubation in YNB medium lacking a carbon source or lacking a carbon source and nitrogen, and growth in YEP maltose vs YEP glucose [20, 27]. There was also significant similarity to C. albicans gene expression changes upon co-culture with macrophages [28], in keeping with current understanding that C. albicans adapts to alternative carbon sources in the macrophage phagosome [27, 28]. Some matching datasets were unexpected; for example, we detected significant similarity to cells in M199 pH 4 vs. YPD, and in M199 pH 8 vs. YPD [29]. The similarity in gene expression changes may reflect the difference in glucose concentration between M199 (0.1%) and YPD (2%). The greatest similarity detected was to an RNA-Seq comparison of biofilm cells vs. yeast cells [30], with 260 common up-regulated genes and 38 common down-regulated genes. The growth conditions for the biofilm study were Spider medium (for biofilm cells), with the carbon source mannitol, and SD+uridine (for yeast cells), with the carbon source glucose. Therefore, it is reasonable that glucose-repressed genes were expressed at higher levels in the biofilm cells than the yeast cells. (The mig1Δ/Δ mig2Δ/Δ double mutant vs. wild type comparison, shown in the figure, is discussed below). Similarities were driven by expression of a core set of 116 genes (S1 Fig; S1G Table). These dataset comparisons underscore the correlations between our results and glucose repression responses that have been measured previously through a variety of approaches.

Fig 1

Gene expression dataset comparisons.

Comparisons were performed using our genome-wide profiling data of wild-type C. albicans grown in YPG media compared to wild-type grown in YPD media A. Fisher’s Exact Test depicting gene expression datasets most closely related to YPG gene expression. FET was used as previously described [31]. Up- or down-regulated genes with a 2-fold expression change cut-off were matched from 91 published expression datasets (S3 Table). B. Heatmap depicting gene expression of 10 metabolic genes chosen for comparison of glucose-repressed genes. Color scale corresponds to log2 fold change limits of 3.3 up (red) and 3.3 down (blue). Full gene expression datasets are available in S1 Table.

Functions of transcription factors Mig1 and Mig2 in gene expression

To define the determinants of C. albicans glucose-repressed gene expression, we investigated the transcription factors Mig1 and Mig2. Both Mig1 and Mig2 have homology to the S. cerevisiae repressors ScMig1, ScMig2, and ScMig3 (S2 Fig), particularly in the DNA binding domain. Gene expression profiling has shown that Mig1 represses some carbon utilization genes in C. albicans [19, 20]. A recent study has implicated Mig2 in control of C. albicans cell size [32], but Mig2 function has otherwise been examined only in mutant screens.

To investigate the functions of Mig1 and Mig2, we created mig1Δ/Δ, mig2Δ/Δ, and mig1Δ/Δ mig2Δ/Δ double mutant strains using a transient CRISPR-Cas9 system [33]. These strains were then complemented using the CIP10 vector [34] to introduce one copy of a wild-type allele of MIG1 or MIG2 at the RPS1 locus. For some experiments, we used a reconstituted strain in which a mig1Δ allele was replaced with a wild-type MIG1 allele at the native locus [35] because the CIP10-MIG1 plasmid yielded incomplete complementation (S4 Table). Substantial reversal of the mutant phenotypes in the complemented and reconstituted strains was verified through expression analysis for 28 genes (S4 Table).

To begin to define Mig1 and Mig2 functions, the wild-type, mig1Δ/Δ, mig2Δ/Δ, and mig1Δ/Δ mig2Δ/Δ double mutant strains were profiled using RNA-Seq in YPD medium. Additionally, the mig1Δ/Δ mig2Δ/Δ double mutant strain was profiled in YPG medium to compare its gene expression levels to the wild-type strain during growth using a non-fermentable carbon source. We found 488 genes that were significantly up-regulated (p<0.05 and >2-fold change in RNA levels) in the mig1Δ/Δ mig2Δ/Δ double mutant compared to wild-type during growth in YPD, while only 144 genes were significantly down-regulated (Fig 2A, S1 Table). These profiling data suggest that Mig1 and Mig2 function mainly as transcriptional repressors.

Fig 2

Mig1/2 regulatory effects.

A. Venn diagram of genes regulated by Mig1, Mig2, or Mig1 and Mig2. Genes included in the diagram were significantly up- or down-regulated (p<0.05) by at least 2-fold compared to the wild-type strain (CW542) in YPD medium. B. Enrichment of SYGGRG motif in Mig1/2 target gene promoters. Promoter sequence was defined as 1,000 basepairs upstream of the open reading frame of up- or down-regulated genes. C. Fold change expression of chosen Mig1 and Mig2 selective genes compared to the wild-type strain analyzed by RNA-Seq and compared between datasets. Mig1 selective genes were significantly up-regulated in the mig1Δ/Δ and mig1Δ/Δ mig2Δ/Δ, but not the mig2Δ/Δ strain profiles. Mig2 selective genes were significantly up-regulated in the mig2Δ/Δ and mig1Δ/Δ mig2Δ/Δ, but not the mig1Δ/Δ strain profiles.

To estimate direct regulatory interactions, we searched for the Mig1 binding motif SYGGRG [18] in the putative promoter regions of Mig1/2 target genes using PathoYeastract [36]. One or more Mig1 binding motifs were present in 86.4%, 93.6%, and 88.5% of genes that were up-regulated in mig1Δ/Δ, mig2Δ/Δ, and mig1Δ/Δ mig2Δ/Δ strains, respectively (Fig 2B). The overrepresentation of Mig1 binding motifs in the promoters of up-regulated genes in all three deletion mutants was statistically significant (p<0.05) compared to promoters from all genes in the genome. Mig1 binding motifs were not enriched in the genes that were down-regulated in the mutant strains (Fig 2B), so these genes may respond indirectly to mig mutations. These results support the argument that Mig1 and Mig2 function as transcriptional repressors.

Shared and selective targets of Mig1 and Mig2

The bulk of target genes are shared by both Mig1 and Mig2, but some target genes respond mainly to Mig1 or Mig2 (Fig 2A). These relationships are evident from both up- and down-regulated genes (Fig 2A). Because up-regulated genes are likely direct targets of the repressors, and because they have the most coherent functions, we focus on those genes below.

There were 67 genes that were significantly up-regulated in both the mig1Δ/Δ and mig1Δ/Δ mig2Δ/Δ mutant strains, but not in the mig2Δ/Δ mutant strain. We call these genes Mig1-selective targets. GO terms for this set of genes were enriched for transmembrane transporters of hexoses, with localization to the plasma membrane (S2 Table). Prominent among Mig1-selective targets were members of the HGT (high-affinity glucose transporters) family. The 20 HGT genes [37] encode transmembrane carbon transporters and at least one glucose sensor, Hgt4 [38]. Expression of these transporters is responsive to both the amount and type of carbon source available to the cells [37, 39]. Mig1 functions as the main repressor for six HGT genes and two predicted sugar transporters HXT5 and MAL31 in YPD (Fig 2C). This result is consistent with a previous RT-PCR experiment showing that Mig1 but not Mig2 represses HGT1, HGT2, and HGT12 [39]. These results indicate that Mig1, and not Mig2, is the major repressor of 67 Mig1-selective target genes in glucose medium.

There were 115 genes that were significantly up-regulated in both the mig2Δ/Δ and mig1Δ/Δ mig2Δ/Δ mutant strains, but not in the mig1Δ/Δ mutant strain. We call these genes Mig2-selective targets. GO term enrichment includes fatty acid catabolic processes and lipid catabolic processes (S2 Table). Mig2-selective genes include the glyoxylate cycle genes ICL1 and MLS1 [40]; the β-oxidation genes FOX2, POT1-2, and POT1; and the genes DCI1 and ECI1, which are predicted to encode enzymes involved in the β-oxidation of fatty acids (Fig 2C). We infer that Mig2, and not Mig1, is the major repressor of 115 Mig2-selective target genes in glucose medium. While Mig1 and Mig2 each repress genes involved in carbon utilization, there are functional distinctions among the Mig1- and Mig2-selective target genes.

The majority of the genes that are up-regulated in the mig1Δ/Δ mig2Δ/Δ double mutant are not up-regulated in either single mutant (258 genes; Fig 2A) indicating that Mig1 and Mig2 functions largely overlap. This profile is expected for genes where either Mig1 or Mig2 is sufficient for repression, and we refer to them as Mig1/2-shared genes. This gene set was enriched for metabolic genes with GO terms such as oxidation-reduction, transmembrane transport and lipid catabolic processes. Many of these genes encode enzymes associated with the peroxisome, similar to the Mig2-selective genes. For example, the peroxisome-related genes PEX1, PEX2, PEX4, PEX6, and PEX12 were repressed only by Mig2, but PEX3, PEX5, PEX8, PEX11, PEX13, PEX14, and PEX19 were repressed by either Mig1 or Mig2 (S1 Table). In addition, we noticed that many Mig1- and Mig2-selective genes were derepressed further in the mig1Δ/Δ mig2Δ/Δ double mutant than in either single mutant. We infer for these genes that Mig1 or Mig2 is the major repressor, but that the other Mig1/2 protein can repress weakly in the absence of the first. Therefore, while Mig1 and Mig2 each have some specific target genes, for the most part they have redundant roles in gene regulation.

The relationship between Mig1/2-regulated genes and glucose-repressed genes can be seen in two dataset comparisons. The first comparison is our assessment of gene expression profile similarity (Fig1A and S3 Table). The differentially regulated gene set from our YPG vs. YPD samples showed greatest similarity to the differentially regulated gene set from our YPD-grown mig1Δ/Δ mig2Δ/Δ double mutant vs. wild type samples. Thus there is a strong correlation between genes repressed by Mig1 or Mig2 and those repressed by glucose. Second, we compared the gene expression levels of YPD- and YPG-grown wild-type and mig1Δ/Δ mig2Δ/Δ double mutant strains. We expected that, if Mig1 and Mig2 function primarily to exert glucose repression, then under derepressing conditions (YPG medium) the double mutation should have greatly diminished gene expression impact (Fig 1B). Indeed, for many genes, this prediction is fulfilled. Genes that are repressed over 8-fold by glucose in a wild-type strain (such as ICL1, FOX2, TRY4, and JEN2) are repressed less than 2-fold by glucose in the mig1Δ/Δ mig2Δ/Δ double mutant. These analyses support the conclusion that Mig1 and Mig2 are major mediators of glucose repression.

Though Mig1 and Mig2 repression activities are modulated by carbon source, our RNA-Seq profile data showed that MIG1 and MIG2 RNA levels are not (S1 Table). These results are consistent with prior studies of C. albicans MIG1 [41] and distinct from the situation in S. cerevisiae, where carbon source affects ScMIG1 and ScMIG2 transcript levels [13, 15] as well as post-translational control [17, 42]. To determine whether Mig1 or Mig2 may be post-transcriptionally regulated, we constructed strains with HA epitope-tagged alleles. Mig1-HA was detectable on Western bots (see below); Mig2-HA was not. We examined Mig1-HA protein levels in cells grown in YPD, YPG, and Spider medium, which has the non-repressing carbon source mannitol. Mig1-HA protein levels were higher in YPD than in YPG or Spider medium, compared to a tubulin control (Fig 3A). In addition, Mig1-HA in Spider medium was detected in two electrophoretic forms (Fig 3A). Mig1-HA levels were significantly different between YPD and YPG or Spider medium (Fig 3A and 3B and S3 Fig). In contrast, MIG1 RNA levels were not significantly different between YPD and YPG or Spider medium (S1 and S5 Tables). To determine whether protein levels were responsive to glucose, we added 2% glucose to cells grown in Spider medium 10 minutes before harvesting. This addition of glucose resulted in a rapid increase in Mig1 protein levels (Fig 3A and 3B). All together, these results show that Mig1 protein levels are responsive to carbon source, while MIG1 RNA levels are not. Regulation of Mig1 at the level of protein accumulation is more similar to regulation of ScMig2 than ScMig1 [17, 42].

Fig 3

Regulation of Mig1 protein accumulation in response to carbon source.

A. Western blotting was performed on cells grown in triplicate expressing HA-tagged Mig1 or the untagged wild-type strain (CW542) grown in YPD as a control. Soluble protein was extracted from strain KL1026 grown in YPD, YPG, Spider medium, and Spider medium with 2% glucose added 10 minutes before harvesting the cells. B. Densitometric analysis of Mig1 protein was performed using FIJI. Total HA-tagged signal was compared to total tubulin signal for loading control (Dunnett test **, p<0.01).

Impact of Mig1 and Mig2 on sak1Δ/Δ mutant growth

Sak1 is the main protein kinase that phosphorylates and activates Snf1 in C. albicans [9]. Because a sak1Δ/Δ is viable, we could use growth assays to test the hypothesis that Mig1 and Mig2 function downstream of Sak1 to control carbon physiology. To test this prediction, we examined sak1Δ/Δ mig1Δ/Δ, sak1Δ/Δ mig2Δ/Δ, and sak1Δ/Δ mig1Δ/Δ mig2Δ/Δ mutant strains, as well as control wild-type and sak1Δ/Δ strains. In agreement with prior studies [9], the sak1Δ/Δ mutant grew poorly on YPG and Spider media (Fig 4). The sak1Δ/Δ mig1Δ/Δ strain grew just as poorly as the sak1Δ/Δ mutant on these media, whereas the sak1Δ/Δ mig2Δ/Δ strain had improved growth on both media (Fig 4 and S4 Fig). The sak1Δ/Δ mig1Δ/Δ mig2Δ/Δ strain had further improved growth on both media. Reconstitution of a SAK1 allele verified that the sak1Δ/Δ mutation was the cause of observed mutant growth defects (Fig 4). These growth tests support the hypothesis that Mig1 and Mig2 act downstream of Sak1 to govern carbon physiology.

Fig 4

Impact of MIG1/2 mutations on growth of sak1 or snf1 mutant and validated strains.

Tenfold serial dilutions of strains KL988(sak1Δ/Δ), KL951(sak1Δ/Δ mig1Δ/Δ), KL960(sak1Δ/Δ mig2Δ/Δ), KL955(sak1Δ/Δ mig1Δ/Δ mig2Δ/Δ), KL992(sak1Δ/Δ +SAK1), KL972(sak1Δ/Δ mig1Δ/Δ +SAK1), KL990(sak1Δ/Δ mig2Δ/Δ +SAK1), KL974(sak1Δ/Δ mig1Δ/Δ mig2Δ/Δ +SAK1), and KL953 (snf1Δ/Δ mig1Δ/Δ), KL957(snf1Δ/Δ mig1Δ/Δ mig2Δ/Δ), KL970 (snf1Δ/Δ mig1Δ/Δ +SNF1), and KL976(snf1Δ/Δ mig1Δ/Δ mig2Δ/Δ +SNF1) were spotted on YPD (glucose), YPG (glycerol), Spider media (mannitol) and Spider media with 2% glucose replacing mannitol. Growth was visualized after 48 h of incubation at 37°C.

Impact of Mig1 and Mig2 on snf1Δ/Δ mutant viability

SNF1 is essential in C. albicans [9, 21–23], a possible consequence of essential genes that are targets of the glucose repression pathway or, alternatively, a glucose repression-independent role for Snf1 that is essential [21]. In S. cerevisiae, one major function of ScSnf1 is to phosphorylate and inactivate ScMig1 [42, 43]. Therefore, we hypothesized that a C. albicans snf1Δ/Δ mutant may be inviable due to the inability to inactivate Mig1 or Mig2, leading to excessive repression of Mig1 or Mig2 target genes. Thus Snf1 essentiality would be tied to its function in the glucose repression pathway.

This hypothesis predicts that a snf1Δ/Δ mutant may be viable in a mig1Δ/Δ, mig2Δ/Δ, or mig1Δ/Δ mig2Δ/Δ mutant strain background. To test this hypothesis, we used the transient CRISPR approach to generate heterozygous and, potentially, homozygous snf1Δ mutants in each mig mutant strain. We recovered heterozygous snf1Δ/SNF1 transformants in all three strains. In addition, we recovered homozygous snf1Δ/Δ transformants in both the mig1Δ/Δ and mig1Δ/Δ mig2Δ/Δ strains, though not in the mig2Δ/Δ strain. These observations support the hypothesis that a snf1Δ/Δ mutant is inviable due to repression of key target genes by Mig1.

Growth properties of representative snf1Δ/Δ derivatives of mig1Δ/Δ and mig1Δ/Δ mig2Δ/Δ strains suggested that loss of Mig1 does not bypass the need for Snf1 entirely. For example, snf1Δ/Δ mig1Δ/Δ colonies exhibited hyperfilamentation and yellow coloration, whereas snf1Δ/Δ mig1Δ/Δ mig2Δ/Δ colonies did not (S5 Fig). In addition, the snf1Δ/Δ mig1Δ/Δ strain displayed minimal growth on YPG or Spider media, whereas the snf1Δ/Δ mig1Δ/Δ mig2Δ/Δ strain displayed substantial growth on both (Fig 4). Poor Spider medium growth of the snf1Δ/Δ mig1Δ/Δ strain was relieved when glucose replaced mannitol as carbon source (Fig 4), so the Spider medium growth defect represents a carbon utilization defect. These observations support the hypothesis that Mig1 is the major repressor of genes that are required for snf1Δ/Δ viability, and that Mig2 contributes to repression of genes that enable more flexible snf1Δ/Δ carbon physiology. Our findings indicate that Snf1 is essential because of its function in the glucose repression pathway.

Relationships among Mig1, Mig2, Sak1, and Snf1 in expression of glucose-repressed genes

To investigate how the phenotypic relationships among these regulators relate at the transcriptional level, we used 28 representative probes for Nanostring analysis from different metabolic pathways (i.e., glycolysis, gluconeogenesis, glyoxylate cycle, fatty-acid catabolism) and selected Mig1/2 target genes from our genome-wide analysis. RNA levels were assayed in a panel of single- and multi-gene mutants. SAK1 or SNF1 reconstituted strains served as controls for sak1Δ/Δ and snf1Δ/Δ mutants, and represented mig1Δ/Δ, mig2Δ/Δ, and mig1Δ/Δ mig2Δ/Δ mutant strains. RNA was isolated from cells grown in YPG medium, a condition in which we expected loss of Sak1 or Snf1 to cause reduced RNA levels from glucose-repressed genes.

We observed that 17 metabolic genes were significantly down-regulated (>2-fold and p<0.05, compared to wild type) in the sak1Δ/Δ mutant (Fig 5 and S6 Table). Of those 17 genes, only three genes (GAL10, JEN2, and PCK1) were significantly down-regulated in the sak1Δ/Δ mig1Δ/Δ mutant, and six genes (GAL10, JEN2, PCK1, HGT1, HGT2 and MAL31) were significantly down-regulated in the sak1Δ/Δ mig2Δ/Δ mutant (Fig 5 and S6 Table). In the sak1Δ/Δ mig1Δ/Δ mig2Δ/Δ triple mutant strain, only GAL10 was significantly down-regulated. All other genes showed increased expression compared to the sak1Δ/Δ mutant, reaching levels as high or higher than in the wild-type strain (Fig 5 and S6 Table). Therefore, the mig1Δ and mig2Δ mutations showed synergistic effects, reflected in the sak1Δ/Δ mig1Δ/Δ mig2Δ/Δ strain, just as they had in their effects on glucose-repressed genes in an otherwise wild-type background (Fig 2A). The fact that the reconstituted sak1Δ/Δ mig1Δ/Δ mig2Δ/Δ +SAK1 strain and the triple mutant sak1Δ/Δ mig1Δ/Δ mig2Δ/Δ strain had nearly identical gene expression profiles (Fig 5) indicates that, for this panel of genes, Mig1 and Mig2 are responsible for almost all gene expression impact attributable to loss of Sak1. These results support the conclusion that Mig1 and Mig2 mediate repression of Sak1 target genes. The synergistic effects of mig1Δ and mig2Δ mutations support the conclusion that Mig1 and Mig2 have partially overlapping roles in mediating repression.

Fig 5

Epistasis analysis among mutations in MIG1, MIG2, SAK1, and SNF1.

Heatmap of log2-fold changes in gene expression of selected carbon metabolism genes analyzed using Nanostring. RNA was extracted from strains CW542 (wild type), KL988(sak1Δ/Δ), KL951(sak1Δ/Δ mig1Δ/Δ), KL960(sak1Δ/Δ mig2Δ/Δ), KL955(sak1Δ/Δ mig1Δ/Δ mig2Δ/Δ), KL992(sak1Δ/Δ +SAK1), KL972(sak1Δ/Δ mig1Δ/Δ +SAK1), KL990(sak1Δ/Δ mig2Δ/Δ +SAK1), KL974(sak1Δ/Δ mig1Δ/Δ mig2Δ/Δ +SAK1), and KL953 (snf1Δ/Δ mig1Δ/Δ), KL957(snf1Δ/Δ mig1Δ/Δ mig2Δ/Δ), KL970 (snf1Δ/Δ mig1Δ/Δ +SNF1), and KL976(snf1Δ/Δ mig1Δ/Δ mig2Δ/Δ +SNF1) grown in triplicate in YPG medium at 37°C. Fold change values were calculated by dividing expression of each gene to the wild type. The sak1Δ/Δ mutant strain analysis is duplicated for ease of comparison to the snf1Δ/Δ mutant strains because a snf1Δ/Δ mutant strain is not viable.

The gene expression profile of a snf1Δ/Δ mutant cannot be assayed, so we used the sak1Δ/Δ strain for comparison to our snf1Δ/Δ strain panel (Fig 5 and S6 Table). Few differences were seen between the sak1Δ/Δ and snf1Δ/Δ mig1Δ/Δ strains except for the glucose transporter genes HGT1 and HGT2. Inviability of snf1Δ/Δ mig2Δ/Δ strain prevented its analysis. The reconstituted snf1Δ/Δ mig1Δ/Δ mig2Δ/Δ +SNF1 strain and the snf1Δ/Δ mig1Δ/Δ mig2Δ/Δ triple mutant strain had nearly identical gene expression profiles (Fig 5 and S6 Table). This observation supports the conclusion that Mig1 and Mig2 are responsible for almost all gene expression impact attributable to loss of Snf1. These results indicate that Mig1 and Mig2 both mediate repression of Snf1 target genes.

Mig1 and Mig2 promote tolerance of cell wall inhibitors

Carbon source influences C. albicans stress resistance and cell wall composition [44, 45], so it seemed possible Mig1 and Mig2 may be required for tolerance of cell wall inhibitors. We tested this idea with growth assays (Fig 6A and 6B) in the presence of the cell wall inhibitor caspofungin. The mig1Δ/Δ mutant showed mild caspofungin hypersensitivity, as detected with a quantitative assay (Fig 6B). The mig2Δ/Δ strain was not caspofungin hypersensitive. The mig1Δ/Δ mig2Δ/Δ double mutant showed strong caspofungin hypersensitivity. Its hypersensitivity was partially relieved by complementation with one copy of MIG2, and essentially reversed by complementation with one copy of MIG1 (Fig 6B). The double mutant was hypersensitive to another cell wall perturbing agent, calcofluor white, and its caspofungin hypersensitivity was relieved by the osmotic stabilizer sorbitol (Fig 6A). The mig1Δ/Δ mig2Δ/Δ double mutant was not hypersensitive to the ergosterol synthesis inhibitor fluconazole (Fig 6A). These results support the idea Mig1 and Mig2 are required for cell wall integrity or biogenesis, probably through impact on carbon physiology, rather than general stress tolerance.

Fig 6

Growth of Mig1/2 mutant and complemented strains during cell wall stress.

A. Plate dilution sensitivity assay. Tenfold serial dilutions of the indicated strains were spotted on YPD and YPD with 100 ng/mL Caspofungin, 100 ng/mL Caspofungin and 1 M Sorbitol, 20 μM Calcofluor White, and 100 μg/mL Fluconazole. Growth was visualized after 48 hours of incubation at 37°C. B. Quantitative caspofungin sensitivity assay in 96 well plates. Cells were incubated in liquid YPD media at 37°C containing 2-fold dilutions of Caspofungin. Absorbance was read at 600nm after 48 hours of incubation. Data was averaged from three biological replicates of duplicate measurements and normalized to cell growth without antifungal.

Mig1 and Mig2 promote filamentation and biofilm formation

Carbon source also affects one of the most prominent C. albicans virulence traits, the ability to form hyphae [46]. Therefore, we tested the mutant strains for defects in hyphal formation. Under our conditions, the mig1Δ/Δ mutant showed a modest but statistically significant decrease in hyphal length compared to wild type while the mig2Δ/Δ mutant showed no hyphal defect (Fig 7C). The mig1Δ/Δ mig2Δ/Δ double mutant had a pronounced defect in hyphal length compared to the wild type (Fig 7C). This defect was not reversed after integration of MIG1 at the RPS1 locus (strain labeled mig1Δ/Δ mig2Δ/Δ + pMIG1), though it was reversed by integration of one copy of the MIG1 at the native locus (strain labeled mig1Δ/Δ mig2Δ/Δ + MIG1). The requirement for native locus integration may reflect regulatory sequences that were absent from the plasmid used to integrate at RPS1. Hyphal formation is required to form stable multicellular biofilm communities [47], and the double mutant showed a significant reduction in biofilm depth compared to the wild-type and single mutant strains (Fig 7A and 7B). These results indicate that Mig1 and Mig2 are required for normal hyphal morphogenesis and biofilm production.

Fig 7

Mig1/2 promote filamentation, biofilm formation, and affect host cell interactions in vitro.

A. Biofilms were grown for 24 h on silicone squares in RPMI media at 37°C. Fixed biofilms were stained with Concanavalin A Alexa Fluor 594. Representative side-view projections were processed and pseudocolored using ImageJ. Scale bar corresponds to depth of the wild-type B. Biofilms were measured using confocal microscopy. Values shown are triplicate measurements from three biological replicates. ****, p<0.0001. C. Indicated strains were grown for 4 h in RPMI at 37°C. Fixed cells were imaged using DIC microscopy and a 20x objective. Hyphal lengths were measured from yeast cell to hyphal tip from >80 cells from 10 fields of view. ****, p<0.0001 D. J774.1 macrophage cytotoxicity was measured by lactate dehydrogenase release (LDH) after 5 h of coincubation with C. albicans strains as indicated. Percentage of LDH release was calculated relative to max release wells containing lysis solution. Values shown are mean with SD from duplicate measurements of three biological replicates. ****, p<0.0001. E. Human endothelial cell damage was assessed by 51Cr release following coincubation with C. albicans cells for 3 h. Percentage of Chromium release was calculated by comparison to release from wild-type (release–spontaneous/total incorporation–spontaneous)**, p<0.01.

Mig1 and Mig2 affect host cell interaction in vitro

Metabolic plasticity is required for C. albicans to escape from macrophages in vitro. Following phagocytosis, C. albicans up-regulates alternative carbon utilization gene expression [27, 48] and undergoes a yeast-to-hyphal transition that disrupts the membrane and triggers macrophage cell death [49]. Because the mig1Δ/Δ mig2Δ/Δ mutant up-regulates alternative carbon utilization gene expression, it seemed possible that the mutant would have increased capacity to damage macrophages. On the other hand, because the mig1Δ/Δ mig2Δ/Δ mutant has a defect in hyphal formation, it seemed possible that it would have diminished capacity to damage macrophages. To determine which of these phenotypes may have overriding impact, we tested the interaction of mig mutants with J774.1 murine macrophages. Lactate dehydrogenase release was used as a readout for macrophage cell damage after incubation with wild-type, mig1Δ/Δ, mig2Δ/Δ, mig1Δ/Δ mig2Δ/Δ, and the relevant complemented strains. The double mutant strain showed significantly decreased cell damage capacity compared to the wild-type or single mutant strains (Fig 7D). Thus, despite a head start that the mig1Δ/Δ mig2Δ/Δ mutant strain may have in utilizing alternative carbon sources, other aspects of its phenotype—potentially the hyphal defect—cause decreased macrophage pathogenesis.

We also assayed host cell interaction through the ability of the mutant strains to cause endothelial cell damage [50], using primary human umbilical vein endothelial cells. The double mutant strain showed a significant defect in cell damage compared to the wild-type and single mutant strains (Fig 7E). These results confirm that Mig1 and Mig2 are redundant for major biological functions and indicate that Mig1 and Mig2 are required for pathogenicity-associated phenotypes.

Discussion

Carbon metabolism and its regulation are central to the pathogenic capability of C. albicans. Carbon metabolic genes undergo dynamic expression changes in numerous colonization and infection models, and defects in carbon metabolism or its regulation disrupt the ability to colonize the GI tract, duel with macrophages and other host cells, and infect host tissues [2, 4]. Here we have extended our understanding of one major carbon regulatory pathway, the Snf1 pathway. We have identified transcription factors Mig1 and Mig2 as mediators of glucose repression in C. albicans and defined their functional activities and connections to upstream regulators Snf1 and Sak1. The Mig1/2 orthologs ScMig1/2 mediate glucose repression in S. cerevisiae, so in a broad context their functions are conserved in C. albicans. However, our studies reveal several distinctive features of Mig1/2 in C. albicans. First, Mig1 and Mig2 have relatively equivalent roles in gene regulation, whereas ScMig1 is the major mediator of glucose repression in S. cerevisiae. In addition, Mig1 is regulated at the level of protein accumulation, a difference from the regulation of ScMig1. A third unique feature is that Mig1 and Mig2 are together required for several traits associated with pathogenicity, including formation of hyphae and biofilm, cell wall inhibitor tolerance, and damage of mammalian host cells. Lastly, Mig1 is required for the essentiality of the Snf1 protein kinase in C. albicans. Snf1 is not essential in S. cerevisiae, and the riddle of Snf1 essentiality has not been addressed to our knowledge in a model system. Below we discuss the functions of Mig1 and Mig2 and their roles in C. albicans biology.

Mig1 and Mig2 function

Our results indicate that Mig1 and Mig2 largely overlap in function as repressors of glucose-repressed genes, yet each has some selective bias in target genes (summarized in Fig 8). We defined selective targets based on their significant up-regulation in one single mutant and not the other. Mig1-selective targets include carbon transporters; Mig2-selective targets include β-oxidation genes. This outcome was unexpected from work in S. cerevisiae; ScMig2 does not regulate specific glucose-repressed genes without ScMig1 [14]. In fact, we observed that Mig2 had more selective target genes than Mig1 (115 vs 67; Fig 2A). However, most selective target genes have further increased RNA levels in the mig1Δ/Δ mig2Δ/Δ double mutant compared to the mig1Δ/Δ or mig2Δ/Δ single mutants (Fig 2C). Therefore, Mig1 and Mig2 both contribute to repression of most selective target genes.

Fig 8

Summary diagram of the roles of Mig1 and Mig2 in the glucose repression pathway in C. albicans.

Sak1 and Snf1 connections drawn from published data by Ramirez-Zavala et al. [9]. Black dotted arrows indicate unclear regulatory connections. Dotted Mig1 indicates reduced protein levels in cells grown on an alternative carbon source. No Mig2 protein level data exist to our knowledge.

The majority of genes that are up-regulated in a mig1Δ/Δ mig2Δ/Δ double mutant strain show little expression change in either a mig1Δ/Δ or mig2Δ/Δ single mutant strain. These 258 genes (Fig 2A) represent targets that can be fully repressed by either Mig1 or Mig2. The large number of such genes argues that Mig1 and Mig2 most often function interchangeably.

We note that 228 genes are significantly up-regulated in a mig1Δ/Δ mig2Δ/Δ double mutant vs wild type in glucose but not in the wild type grown in glycerol vs glucose. This gene class might be taken to indicate that Mig1 and Mig2 have functions beyond a role in glucose repression. The 228 genes are enriched for functions in oxidation-reduction and peroxisome biogenesis. In yeast, peroxisomes function as sites for hydrogen peroxide reduction and fatty acid metabolism [51] and the abundance of these specialized organelles increases in response to non-glucose media such as oleic acid [52]. Therefore, some of the genes may have medium-specific glucose repression responses. Indeed, we find that 65 of the genes are up-regulated in the YEP maltose vs YEP glucose comparison [20]; another 43 of the genes are up-regulated in the YEP lactate vs 0.01, 0.1, or 1% glucose datasets [53]. The remaining genes are still enriched for oxidation-reduction and peroxisomal functions. It is possible that Mig1 and Mig2 respond to some signal in addition to presence of glucose, though the functional enrichment among these target genes suggests that the signal is related to carbon metabolism.

Some genes respond only in the mig1Δ/Δ or mig2Δ/Δ single mutants, and not in the mig1Δ/Δ mig2Δ/Δ double mutant. Most of these genes are down-regulated in the mig mutants. Because Mig1 and Mig2 seem to function primarily as repressors, we infer that down-regulated genes respond indirectly to Mig1/2 through compensatory responses to altered metabolism or other biological processes. About 10% of up-regulated genes respond only in one single mutant, and we suggest that they reflect compensatory responses as well.

Mig1 and Mig2 regulation

In S. cerevisiae, ScMIG1 and ScMIG2 RNA levels vary with carbon source [13, 15]. In addition, ScMig1 is regulated at the level of nuclear translocation [17], whereas ScMig2 is regulated through protein degradation [16]. In C. albicans we found no evidence that RNA levels vary for MIG1 or MIG2, in agreement with prior MIG1 studies [18]. However, our data show that Mig1 protein accumulates at higher levels in glucose media than in non-repressing glycerol or mannitol media. It is possible that changes in Mig1 accumulation reflect carbon-responsive differences in its degradation rate, though this mechanism is not proven.

Connection between Mig1/2 function and pathogenicity

A mig1Δ/Δ mig2Δ/Δ double mutant expresses many metabolic genes that are induced upon interaction with host cells. On that basis we anticipated that the double mutant would be “pre-adapted” to growth after uptake into macrophages and endothelial cells, and might then cause excessive damage to host cells. We observed the opposite result though; the mig1Δ/Δ mig2Δ/Δ double mutant caused reduced damage. One explanation is that metabolic flexibility is critical for effective host cell interaction, and the fact that the mig1Δ/Δ mig2Δ/Δ double mutant is locked in the glucose-derepressed state impedes interaction, perhaps at an early time when glucose is available. A second possibility is that the double mutant’s cell wall integrity defect impairs its host cell interaction. Finally, we observed that the mig1Δ/Δ mig2Δ/Δ double mutant has impaired filamentation, a phenotype closely tied to host interaction. Sak1, Snf1, and now Mig1-Mig2 have all been connected to filamentation [9, 24, 25], though the mechanisms underlying this connection is unclear. Two major activators of filamentation, BRG1 and UME6, are down-regulated in the mig1Δ/Δ mig2Δ/Δ double mutant during growth in YPG (S1 Table). A simple possibility is that levels of Brg1 and Ume6 mediate effects of Mig1/2 on filamentation.

Placement of Mig1 and Mig2 in the Snf1 pathway

Conservation of the Snf1 pathway and the prior connection of Mig1 to glucose-repressed genes has suggested that Mig1 acts downstream of Snf1 [4], and these same arguments pertain to Mig2. In addition, here we have presented epistasis tests that support the hypothesis that Mig1 and Mig2 act downstream of Snf1 and Sak1. For example, sak1Δ/Δ mutant growth defects on YPG and Spider medium were partially suppressed by a mig2Δ/Δ mutation; suppression was more effective with a mig1Δ/Δ mig2Δ/Δ double mutation. Also, snf1Δ/Δ mutant inviability was relieved by a mig1Δ/Δ mutation to yield a phenotypically abnormal strain; phenotypic abnormalities were relieved by a mig1Δ/Δ mig2Δ/Δ double mutation. Nanostring gene expression readouts also supported this hypothesis. Specifically, sak1Δ/Δ gene expression defects were partially reversed in either sak1Δ/Δ mig1Δ/Δ or sak1Δ/Δ mig2Δ/Δ strains, and a sak1Δ/Δ mutation had only minor gene expression impact in a mig1Δ/Δ mig2Δ/Δ double mutant background. Similarly, a snf1Δ/Δ mutation had only minor gene expression impact in a mig1Δ/Δ mig2Δ/Δ double mutant background. These observations are consistent with the hypothesis that Mig1 and Mig2 function downstream of Sak1 and Snf1 to mediate effects on gene expression and biological phenotype.

Mig1 and the nature of Snf1 essentiality

The requirement of Snf1 for C. albicans viability [9, 21–24] can be explained by two hypotheses, first proposed by Kwon-Chung and colleagues [21]. Snf1 may be essential because of its function in the glucose repression pathway; for example, essential genes may be under glucose repression in C. albicans. Alternatively, Snf1 may be essential because it has a novel second role in C. albicans in addition to its role in glucose repression pathway, and that second role is essential for viability. The fact that loss of Mig1 permits recover of viable snf1Δ/Δ mutants provides strong support for the first hypothesis. Moreover, the finding helps refine the hypothesis in providing candidate genes that may be the essential targets of glucose repression. Given that snf1Δ/Δ mutants could be recovered in a mig1Δ/Δ background and not in a mig2Δ/Δ background, we infer that repression of Mig1-selective targets is the cause of snf1Δ/Δ mutant inviability in an otherwise wild-type background. Based on our analysis, there are thus only 67 candidate genes (Fig 2A and S1 Table). Although many Mig1-selective target genes are known to be nonessential, we suspect that reduced expression of multiple Mig1-selective targets may have additive effects that impact viability. One simple hypothesis is that snf1Δ/Δ mutant inviability is the result of a profound glucose uptake defect, given that glucose transporter genes are major Mig1-selective targets.

Methods

Media and culture conditions

Frozen strains were maintained in 15% glycerol frozen stocks at -80°C. Streaked strains were maintained on YPD agar plates (2% dextrose, 2% Bacto peptone, 1% yeast extract, 2% Bacto Agar) and overnight cultures were grown in liquid YPD media (2% dextrose, 2% Bacto peptone, 1% yeast extract) rotating at 75 rpm at 30°C in 15 ml culture tubes. Transformants were selected on synthetic media plates (2% dextrose, 1.7% Difco yeast nitrogen base with ammonium sulfate and necessary auxotrophic supplements) or selected for nourseothricin-resistance on YPD + 400 μg/ml nourseothricin (clonNAT, Gold Biotechnology). For alternative carbon source phenotyping, strains were spotted on YPG Plates (2% glycerol, 2% Bacto peptone, 1% yeast extract, 2% Bacto Agar) and Spider Medium Plates (1% Difco Nutrient Broth, 1% D-Mannitol, 0.2% Potassium Phosphate Dibasic (K2HPO4), 2% Bacto Agar). For transcript profiling, strains were grown in filter sterilized, liquid YPD, YPG, or Spider Medium. For hyphal growth and macrophage assays, strains were grown in filter sterilized Roswell Park Memorial Institute medium 1640 media (RPMI) pH 7.4.

Strain construction

PCR products or linearized plasmids were transformed into Candida albicans cells using the lithium acetate transformation method [54]. Homozygous mutants were constructed in the SN152 or SN250 background strains [55] using the pSN69 (Candida dubliniensis ARG4), pSN52 (Candida dubliniensis HIS1), pSN40 (Candida maltosa LEU2), or pNAT [33] deletion cassettes. Homozygous mutants were created using a sgRNA targeting the gene of interest, a repair DNA template containing a selection marker, and a transient CRISPR-Cas9 system as previously described in detail [33].

For western blotting analysis, a hemagglutinin (HA) epitope-tagged Mig1 strain was constructed using strain SN152. Primers KL44 and KL45 were used to amplify the HA tag and HIS1 selection marker from plasmid pED3-HA (S2 Text) and contained homology to the 3’ end of the MIG1 coding region. A single guide expression cassette targeting the genomic region downstream of the MIG1 coding region was created using primers KL46 and KL47. The transformation mix contained the DNA amplified from pED3-HA, Cas9, and the single guide expression cassette. Transformants were selected on CSM media lacking Histidine. To construct a prototrophic strain a single guide targeting the junction between the leu2 upstream sequence and the deletion scar was created using primers KL48 and KL49 and CmLEU2 and CdHIS1 amplified from SN250 genomic DNA. The transformation mix contained the single guide expression cassette, Cas9, CmLEU2, and CdHIS1 DNA. Transformants were selected on CSM media lacking Leucine and Histidine.

For a detailed description of strain construction, see Supplemental file S1 Text.

For a complete strain and primer list, see Supplemental files S7 and S8 Tables.

Spotting plate assays

Strains were diluted in H2O to an OD600 of 3.0 measured with a spectrophotometer. Five-fold dilutions were spotted using a multichannel pipette on the indicated media. Plates were incubated for 1–3 days at 37°C or 30°C as indicated.

Hyphal microscopy and quantification

Cells from an overnight culture were diluted to an OD600 of 0.5 in glass culture tubes containing 5 ml pre-warmed RPMI media. Cultures were grown rotating at 60 rpm at 37°C for 4 h. Cells were fixed with 4% formaldehyde and stained with Calcofluor white. Cells were visualized with a Zeiss Axio Observer Z.1 fluorescence microscope and a 20x objective.

For quantification of hyphal length, 10 fields of view were acquired for each strain analyzed using a 20x DIC objective with 1.6x optical zoom. Cells were measured from the beginning yeast cell to the end hyphal tip using the segmented line tool from ImageJ (https://imagej.nih.gov/ij/). Only cells where the entire hyphal cell was clearly visible in the field of view were measured. Significance was calculated using the GraphPad PRISM one-way ANOVA test followed by Tukey post hoc analysis, p<0.05 significance.

RNA-sequencing

RNA-sequencing and data analysis were performed as previously described [56] 5 μg of extracted RNA was treated with 2 units of TurboDNAse (Invitrogen) in a 50 μl reaction. To inactivate the DNAse, NaCl, Tris-HCl pH7.5, and EDTA, were added to a final concentration of 500 mM, 200 mM, and 10 mM respectively. The RNA was PCI extracted and the supernatant containing the RNA was purified over a Zymo Research RNA clean up column and eluted with 15 μl of nuclease free water. 2 μg of total RNA was used as input for the Lexogen mRNA sense kit v2. The kit was used according to the manufacturer’s instructions for shorter amplicons. 11 PCR cycles were performed, and the libraries were run on a D1000 DNA tape (Tapestation) to assess the size and quality of the library. The concentrations of the libraries were measuring using the High sensitivity DNA assay for the Qubit (Invitrogen) The libraries were diluted to 8 nM and subjected to one lane of Illumina sequencing (Novogene), resulting in an average of 16 million reads per library.

Raw fastq reads were trimmed using cutadapt (v 1.9.1) with options “-m 42 -a AGATCGGAAGAGC” to remove Illumina 3’ adapter sequence and “-u 10 -u -6” to remove the Lexogen random priming sequences. Trimmed reads were mapped using tophat (v 2.0.8) [57] with options “–no-novel-juncs” and “-G” to align to the C. albicans SC5314 reference genome assembly 22. Primary alignments were selected using samtools (v 0.1.18) [58] with options “view -h -F 256”. Gene counts were created using “coverageBed” from bedtools (v 2.17.0) [59] with option “-S” to count stranded alignments. RNA-Seq reads mapped to the two alleles of each gene were combined for further analysis. Differential expression was assessed using DEseq2 (v 1.18.1) [60] in RStudio (v 1.1.383) using default options (alpha = 0.05).

Biofilm microscopy and quantification

Biofilms were prepared and imaged as previously described in detail [61]. Overnight cultures were diluted to an OD600 of 0.2 in 2 ml RPMI media in 6 well culture plates containing 1.5 cm x 1.5 cm sized medical-grade silicone squares. After 90 minutes, the biofilm squares were dipped in sterile PBS to wash unadhered cells and placed in a new 6 well culture plate containing 2 ml of RPMI media. Biofilm cultures were incubated at 37°C with orbital shaking at 60 rpm. After 24 h, biofilms were fixed with 4% formaldehyde/2% glutaraldehyde for 20 minutes. Fixed biofilms were washed 2x with PBS and stained with Alexafluor 594-concanavalin A for 24 h at room temperature.

Stained biofilms were indexed matched by a series of dehydration steps followed by immersion in methyl salicylate (50:50 methanol:H2O 2x, 100% methanol 2x, 50:50 methanol:methyl salicylate 2x, and then 100% methyl salicylate2x). The index matched biofilm was viewed through a coverglass and imaged using a slit-scan confocal optical unit on a Zeiss Axiovert 200 microscope with a Zeiss 40x/0.85 NA oil immersion objective.130 images were obtained at a time using a focus increment of 0.9 μm. Biofilm image stacks were processed using ImageJ (https://imagej.nih.gov/ij/) for concatenation, background subtraction, reslicing, and max projection.

For biofilm depth quantification, 3 measurements from 3 biological replicates were measured using a slit-scan confocal optical unit on a Zeiss Axiovert 200 microscope with a Zeiss 40x/0.85 NA oil immersion objective. Significance was calculated using the GraphPad PRISM one-way ANOVA test followed by Tukey post hoc analysis, p<0.05 significance.

Data analysis software

Venn diagrams analysis of gene expression data was performed using (http://bioinformatics.psb.ugent.be/webtools/Venn/). Area-proportional venn diagram images were created using the eulerr package in R. Statistical analyses were performed using GraphPad Prism version 8.00 (Graphpad Software, Inc., La Jolla). Heatmaps were created using MultiExperiment Viewer (MeV). Caspofungin minimum inhibitory concentration heatmap was created using JavaTreeView.

Macrophage cytotoxicity

The macrophage cytotoxicity protocol was performed as described previously [62]. The J774A.1 murine macrophage cell line was graciously provided by Sai Gopalakrishna Yerneni from the lab of Dr. Phil Campbell. Cells were maintained in RPMI media without phenol red, with 10% Serum and 5% penicillin/streptomycin at 37°C in 5% CO2. Cells were used from passages 8–16. 100 μl of macrophage suspension was plated at a concentration of 2.5x105 cells/ml overnight in a 96 well tissue culture treated polystyrene plate.

The following day, overnight cultures of C. albicans were subcultured in YPD media for 5 h. Subcultured cells were washed twice in PBS and diluted to a concentration of 3x106 in pre-warmed RPMI media without FBS, without phenol red, with 5% penicillin streptomycin at 37°C. Media was removed from overnight macrophage cultures and replaced with 150 μl of RPMI without FBS. 50 μl of C. albicans cells were added to each well for a MOI of 3. (3 C. albicans cells:1 macrophage). 6 wells of macrophages were not incubated with C. albicans cells for 3 spontaneous release control wells and 3 max release control wells. Cells were incubated for 5 h. Following incubation, the Pierce LDH Cytotoxicity Assay kit was used according to the manufacturer’s instructions. To achieve max lactate dehydrogenase release, 10 μl lysis solution was added to each max release control well. For positive control wells, 200 μl of a 10 mL PBS + 1% BSA freshly made stock was added to 3 blank wells and 2 μl of positive control mix from the kit was added. Supernatant from all wells was diluted 1:5 in PBS and 100 μl was pipetted into a new 96 well plate. 50 μl of substrate mix was added to each well and incubated for 30 min. at RT. Following incubation, 50 μl of stop solution was added to each well.

Absorbance was read on a Tecan at 490 nm and background absorbance was read at 680 nm. The background absorbance was subtracted from the 490 nm reading. Percent cytotoxicity was calculated according to the manufacturer’s guidelines. The assay was performed in triplicate to calculate percent cytotoxicity. Significance was calculated using the GraphPad PRISM one-way ANOVA test p<0.05 significance.

RNA extraction

For RNA-sequencing experiment and Nanostring experiment for Mig1/Mig2 strains, overnight cultures of cells were diluted to an OD600 of 0.2 in 25 ml filtered YPD media. Cells were grown for 4 h shaking at 225 rpm at 37°C.

For Nanostring experiment of Sak1/Snf1 strains, overnight cultures of cells were washed in H2O and diluted to an OD600 of 0.4 in 25 ml filtered YPG media. Cells were grown for 4 h shaking at 225 rpm at 37°C.

Cells were harvested via vacuum filtration and frozen at -80°C. RNA extraction was performed using Qiagen RNeasy Mini Kit (cat#74104) with modifications. Cells were washed from the membrane and resuspended with 1 ml of cold H2O. The cell suspension was centrifuged at top speed for 30 sec. The supernatant was removed, and the cell pellet was resuspended with 600 μl of RLT + 1% β-Mercaptoethanol. The cell suspension was transferred to a screw cap tube containing 300 μl Zirconia beads (Fisher Scientific) and 600 μl phenol:chloroform:isoamyl alcohol 25:24:1. Tubes were vortexed using a mini-beadbeater (Biospec Products) for 3 min, and centrifuged at top speed for 5 min at 4°C. 550 μl of the top aqueous phase was transferred to a new screw cap tube containing 550 μl of 70% ethanol. The mixture was transferred to a RNeasy Mini Spin Column and RNA isolation was followed according to the manufacturer’s guidelines.

Nanostring

Nanostring analysis was performed as previously described [25]. Gene expression was measured using the nCounter SPRINT Profiler. For our analysis, 28 target genes and 5 normalization genes (ARP3, FKH2, GIN4, TUP1, and CDC28) were selected for the codeset. For each Nanostring assay, 100 ng of RNA was added to the Nanostring codeset mix and incubated at 65°C overnight (16–18 h). The samples were loaded onto the cartridge according to the manufacturer’s instructions and placed in the instrument for scanning and data collection.

Heatmaps of Nanostring gene expression data represent were created using MultiExperimentViewer v4.8.1 (MeV) software. Colors represent log2-transformed ratios of gene expression comparisons. Hierarchical clustering was performed using average linkage clustering based on Manhattan distance and optimized for gene leaf order.

Endothelial cell damage assay

Endothelial cell damage by C. albicans cells was assessed as previously described [63] by measuring 51Cr release. Human endothelial cells were maintained in M199 medium as previously described [63] and loaded with 5 μCi/ml Na2 51CrO4 overnight. Cells were washed, and inoculated with C.albicans cells at a concentration of 4×104 organisms per well. Cells were incubated for 3 h and the 51Cr release was quantified using the formula: (experimental release—spontaneous release)/(total incorporation—spontaneous release). The assay was performed in triplicate using three biological replicates. Statistical significance was calculated using GraphPad Prism one-way ANOVA test p<0.05 significance.

Western blotting

Total soluble protein was extracted for Western blotting analysis from strain KL1026 and strain CW542 (for untagged negative control) in 50 ml cultures of the indicated media at 37°C, shaking at 225 rpm for 4 h. Cells were harvested by centrifugation and resuspended in FA lysis buffer (50 mM HEPES-KOH, 140 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% Sodium deoxycholate, 1 mM PMSF, and 1x proteinase inhibitor). Resuspended cells were homogenized by bead-beating with glass beads. Protein concentration was measured using a Bradford Assay(Bio-Rad). 30 μg of protein was loaded for each gel.

Protein extracts were separated on an 8% SDS polyacrylamide gel and transferred to PVDF membranes (Bio-Rad Laboratories). Ponceau S solution (Sigma) was used to assess equal loading. Membranes were blocked for 16 h at 4°C in 5% milk (Blotting-grade Nonfat dry milk, Bio-Rad Laboratories) in TBST (tris-buffered saline containing 0.05% Tween 20, pH 7.4). After washing in TBST, membranes were incubated with anti-HA monoclonal antibody (Roche; #11 583 816 001) at a 1:6,000 dilution for 1 h RT, washed with TBST, and then incubated with the secondary antibody, anti-mouse IgG-HRP(Santa Cruz Biotechnology; sc-516102), at a 1:5,000 dilution for 1 h RT. Signal was detected by enhanced chemiluminescence (Thermo Fisher Scientific, #32106) For loading control, the membranes were stripped for 10 min. using buffer containing 0.2 M glycine, 0.1%SDS, and 1% Tween 20 adjusted to pH 2.2. Membranes were washed in PBS and TBST before blocking again with 5% milk. The steps above were repeated for the stripped membrane except incubating with anti-tubulin rat monoclonal antibody(Abcam, #ab6046) at 1:6,000 dilution and Goat anti-Rat IgG HRP(Abcam #ab6734) at 1:5,000 dilution for the secondary antibody.

Blots were imaged using ChemiDoc Touch Imaging System (Bio-Rad Laboratories). Signal was quantified by densitometric analysis using FIJI (https://fiji.sc/). Mig1-HA signal was normalized to the Tubulin signal among the samples within the same blot.

Intersections of YPG vs YPD responsive genes and related datasets.

Venn Diagram includes upregulated genes from mig1Δ/Δ mig2Δ/Δ vs wild-type YPD, wild-type YPG vs wild-type YPD, wild-type engulfed by macrophages vs non-engulfed [28], pH4 vs YPD [29], and biofilm cells vs yeast cells [30]. A list of the 116 shared genes can be found in tab S1G of S1 Table. The gene set is enriched for the GO terms “monocarboxylic acid catabolic process,” “fatty acid catabolic process,” and related terms (tab S2D of S2 Table).

(PDF)

Click here for additional data file.

Multiple sequence alignment of Mig proteins from Saccharomyces cerevisiae and Candida albicans.

Protein sequences were aligned using Clustal Omega and visualized using Jalview Clustal color scheme.

(PDF)

Click here for additional data file.

Western blot of Mig1-HA tagged protein grown in media with various carbon sources.

A. Western blots are from biological triplicates used to quantify Mig1-HA tagged protein levels. Protein extracts from strains KL1026 and CW542 were made from overnight cultures washed in H2O, inoculated at OD 0.2, and grown in 50 ml cultures of the indicated medium for 4 h at 37°C. For Spider + Glucose medium, 5 ml of 20% glucose was added to a culture grown in Spider medium at 3 h and 50 min. CW542 protein extracts were used as a non-tagged negative control. Ponceau S was used as a control for proper protein transfer. Samples were visualized with an antibody against HA and membranes were stripped and reprobed with an antibody against Tubulin for loading control. B. Functional phenotyping of Mig1 protein in the Mig1-HA strain using a caspofungin sensitivity plate dilution assay from Fig 6A. The Mig1-HA strain is not more sensitive to caspofungin compared to the wild-type.

(PDF)

Click here for additional data file.

Independent isolates of sak1Δ/Δ mig1Δ/Δ and sak1Δ/Δ mig2Δ/Δ strains show similar growth phenotypes on Spider media.

Strains: Wild-type (CW542), sak1Δ/Δ (KL988), sak1Δ/Δ mig1Δ/Δ (KL951 and KL952), sak1Δ/Δ mig1Δ/Δ mig2Δ/Δ (KL955) and sak1Δ/Δ mig2Δ/Δ (KL960 and KL962) were grown overnight in YPD and tenfold serial dilutions of the indicated strains were spotted on YPD and Spider plates. Growth was visualized after 2 days of incubation at 37°C.

(PDF)

Click here for additional data file.

The snf1Δ/Δ mig1Δ/Δ strain shows abnormal filamentation and coloration compared to wild-type and the snf1Δ/Δ mig1Δ/Δ mig2Δ/Δ strain.

Strains: Wild-type (CW542), snf1Δ/Δ mig1Δ/Δ (KL953 and KL954), snf1Δ/Δ mig1Δ/Δ mig2Δ/Δ (KL957 and KL958) and snf1Δ/Δ mig1Δ/Δ mig2Δ/Δ + SNF1 (KL974) were spotted at an OD of 0.1 on YPD media and grown at 30°C for 7 days.

(PDF)

Click here for additional data file.

Strain construction.

Details of strain constructions are provided.

(DOCX)

Click here for additional data file.

Plasmid ED3-HA sequence.

The sequence of plasmid ED3-HA is provided.

(DOCX)

Click here for additional data file.

RNA-seq.

RNA-seq data are provided.

(XLSX)

Click here for additional data file.

GO terms.

GO terms associated with gene subsets are provided.

(XLSX)

Click here for additional data file.

FET YPG vs YPD.

Comparisons of datasets via Fisher’s Exact Test are provided.

(XLSX)

Click here for additional data file.

NanoString complementation.

Gene expression data for mutant and complemented strains are provided.

(XLSX)

Click here for additional data file.

NanoString spider medium.

Gene expression data for cells grown in Spider medium are provided.

(XLSX)

Click here for additional data file.

NanoString carbon codeset.

Detailed gene expression data underlying Fig 5 are provided.

(XLSX)

Click here for additional data file.

Primer list.

Sequences of primers used in this study are provided.

(XLSX)

Click here for additional data file.

Strain list.

Genotypes of strains used in this study are provided.

(XLSX)

Click here for additional data file.

10 Sep 2019

Dear Aaron,

Thank you very much for submitting your Research Article entitled 'Resolving the roles of Mig1 and Mig2 in the Sak1/Snf1 controlled glucose repression pathway in Candida albicans' to PLOS Genetics. Your manuscript was fully evaluated at the editorial level and by three independent peer reviewers all of whom work on Candida as the experimental system (fungal genetics, molecular medical mycology). The reviewers appreciated the attention to an important problem, but raised some concerns about the current manuscript. Based on the reviews, we will not be able to accept this version of the manuscript, but will be happy to review again a revised version (this constitutes major revision with respect to two reviews, and minor revision with respect to the other). As journal policy, we cannot, of course, promise publication at that time.

Your revisions should address the specific points made by each reviewer. We also ask that you include a detailed list of your responses to the review comments and a description of the changes made in the manuscript.

In revising the manuscript for further consideration at PLOS Genetics, please aim to resubmit within the next 60 days, unless it will take extra time to address the concerns of the reviewers, in which case we would appreciate an expected resubmission date by email to plosgenetics@plos.org.  The revised manuscript will be sent to one or more of the original reviewers, depending on revisions and availability of the reviewers.

If present, accompanying reviewer attachments are included with this email; please notify the journal office if any appear to be missing. They will also be available for download from the link below. You can use this link to log into the system when you are ready to submit a revised version, having first consulted our Submission Checklist.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see our guidelines.

Please be aware that our data availability policy requires that all numerical data underlying graphs or summary statistics are included with the submission, and you will need to provide this upon resubmission if not already present. In addition, we do not permit the inclusion of phrases such as "data not shown" or "unpublished results" in manuscripts. All points should be backed up by data provided with the submission.

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool.  PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org.

PLOS has incorporated Similarity Check, powered by iThenticate, into its journal-wide submission system in order to screen submitted content for originality before publication. Each PLOS journal undertakes screening on a proportion of submitted articles. You will be contacted if needed following the screening process.

To resubmit, use the link below and 'Revise Submission' in the 'Submissions Needing Revision' folder.

[LINK]

We are sorry that we cannot be more positive about your manuscript at this stage. Please do not hesitate to contact us if you have any concerns or questions.   We thank you for your support of the Public Library of Science in General, and PLOS Genetics in particular.

Yours sincerely,

Joseph Heitman, MD, PhD

Associate Editor

PLOS Genetics

Gregory P. Copenhaver

Editor-in-Chief

PLOS Genetics

Reviewer's Responses to Questions

Comments to the Authors:

Please note here if the review is uploaded as an attachment.

Reviewer #1: The manuscript by Mitchell et al describes the role of Mig1 and Mig2, two repressors that repress genes involved in carbon source utilization and catabolite repression. The authors show that Sak1 is required for growth in the absence of glucose, and that Mig1 and Mig2 are epistatic to Sak1, in a partially redundant way. The paper begins with a transcriptomics analysis of cells grown in the presence of either glucose of glycerol and includes mig1, mig2, or mig1mig2 mutants, continues to nicely phenotype double and triple mutants along with complemented derivatives to support their genetic model, then characterizes the role of Mig1 and Mig2 in several important phenotypes. The manuscript is well-written and tells a nice story, and provides a satisfying explanation for the essentiality of Snf1 in C. albicans. I have a few comments that might strengthen the manuscript.

1) The authors points on the relatedness between their data set and other datasets could be more useful with more detail on whether the same genes (a subset) drove the similarity to the differentially expressed genes in other datasets. For example, were similar subsets of genes among those that drove similarity to macrophage experiments and M199-YPD comparisons? Did the Mig1 and Mig2 and Mig1/2 regulons coincide with any of these experiments? The comparison to other datasets might be useful later on in the paper once the Mig regulons have been defined. As written, this part of the paper is rather speculative.

2) On p. 14, the following text is found: “Growth properties of representative snf1Δ/Δ derivatives of mig1Δ/Δ and mig1Δ/Δ mig2Δ/Δ strains suggested that loss of Mig1 does not bypass the need for Snf1 entirely. For example, snf1Δ/Δ mig1Δ/Δ colonies exhibited hyperfilamentation and a peculiar yellow color, whereas snf1Δ/Δ mig1Δ/Δ mig2Δ/Δ colonies did not (Fig S1).”

The authors might consider whether the yellow color is riboflavin. Several studies have shown that metabolic perturbation and alterations in cAMP signaling can change riboflavin production. Here is one paper that references riboflavin production but it is not the first (PMID:21992526). Shining a UV light on the colony may reveal characteristic fluorescence and a simple extraction can be analyzed for the reduced and oxidized spectra characteristic of this molecule.

3) Fig. 4 should be revised to have genes in all capitals in the right-hand column and the text resolution is of a lesser quality than for other figures.

4) Italics needed in at least one figure legend and references need to be joined within a single bracket in the introduction.

5) Could the authors expand on the statement that “peroxisome biogenesis [is a] functions that align well with alternative carbon source utilization”.

6) Do the transcriptomics data still need to be deposited?

Reviewer #2: Regulation of sugar usage is very important for cells, and investigating the mechanisms by which organisms control which sugars are metabolised is a significant scientific endeavor. In the manuscript the Mitchell and Filler labs investigate the roles of Mig1 and Mig2 in glucose repression in the fungal pathogen C. albicans. The basic strategy is to make single and double mutants, and investigate the phenotypes in terms of sugar usage (glucose vs alternative sources) and, because the organism is a pathogen, the consequences for pathogenicity and traits associated with this lifestyle.

In the well-investigated model yeast S. cerevisiae, glucose repression has been connected to the Snf1 kinase and its regulation of the key transcription factor Mig1 – Mig1 serves as a repressor of alternative carbon sources. High glucose, inactive Snf1, active Mig1, and repression of the use of alternative sugars; low glucose, active Snf1, inactive repressor Mig1, and alternate sugars can be used. There are lots of nuances and other players in the circuitry, but the Snf1/Mig1 interaction is central to the control process. However, S. cerevisiae is somewhat of an outlier in aspects of carbon utilization relative to other fungi, in particular its propensity to ferment glucose even in the presence of oxygen (the Crabtree/Warburg effect), so it is reasonable to ask if the glucose circuitry is also controlled by Snf1/Mig1 in a non-Crabtree yeast like C. albicans.

Basically, the results of this study say yes – the Snf/Mig regulatory circuit controls alternative carbon use in C. albicans.

First, the authors do an RNAseq analysis of glucose vs glycerol grown cells and identify gene sets that are up-regulated (~ 400) and down-regulated (~ 100) in response to the growth in glycerol. These genes are compared to the expected genes from studies in S. cerevisiae, and to data sets for other comparisons in C. albicans, and are found to represent a sensible data set. Thus C. albicans appears to have a “standard” glucose repression response.

Next, they examine whether Mig1 and Mig2, two paralogs of the Mig1/2 repressors in S. cerevisiae, are important for glucose repression. The phylogenetic relationships of Mig1/2 in S. cerevisiae and Mig1/2 in C. albicans needs to be better explained. Essentially, as C. albicans is a pre WGD yeast, and S. cerevisiae is a post WGD yeast, there are several possible ways you could get a Mig1 and a Mig2 in both species. It is worth looking to see if a prediction can be made about whether the pre WGD yeast had a Mig1 and a Mig2 that became Mig1A and Mig1B; Mig2A and Mig2B after the duplication and resolved back to Mig1A and Mig2B, or did it resolve to Mig1A, Mig1B in yeast and lose the whole Mig2 lineage from before the WGD? Lots of possible arrangements, worth seeing if something could be said, especially given that the Mig1 and Mig2 in C. albicans seem to be functionally closer to each other than the Mig1 and Mig2 of S. cerevisiae.

To investigate the functions of the two candidate repressors, they make the single and double mutants, complement the genes by dropping versions in at the standard RPS1 locus, and, in the cases where position effects prevent full complementation, they replace to WT gene at the WT locus. This provides confidence the phenotypic assessments to follow are really the result of the loss of Mig1 or Mig2 function.

The first set of phenotypes is to see which genes are de-repressed in the absence of either or both the repressors. The authors identify genes that are repressed by both (need to be mig1-mig2- to see de-repression), repressed specifically by Mig1 (de-repressed in the mig1- or the mig1-mig2- strains), and repressed specifically by Mig2. The suggest that the specifically repressed genes can be functionally sorted – Mig2 for glyoxylate and beta oxidation, Mig1 for transporters, Mig1 and Mig2 for everything else, and in fact the majority of the regulated genes. Finally, they argue the Mig1/2 regulated genes are the same genes that show glucose repression, and that glucose repression of these genes doesn’t work well in the double mutant. These data sets do not correlate perfectly of course, and the failure of glucose repressed genes to respond in the mig1/2 double mutant is not absolute (still some apparent repression of some genes), but the patterns seen strongly support the central premise here.

Next the authors address the relationship of Snf1 and the Mig1/2 elements. Loss of Snf1 is lethal to C. albicans cells, this is somewhat surprising and has greatly limited analysis of the element. Based on the hypothesis that over-suppression of the Mig gene targets could be the cause of the lethality, the authors attempted to rescue the snf1 nulls be deleting the repressors, and were able to get viable snf1 nulls in the mig1- and mig1-mig2- backgrounds. Thus, overactivity of Mig1, presumably suppression of some essential function(s), appears to be the cause of the Snf1 null’s inviability. They follow this up by assessing the impact of the mig mutants on the regulatory kinase upstream of Snf1, and show that growth defects associated with loss of this kinase are abrogated by the repressor deletions, but find that the mig2 repressor is more important, in contrast to mig1 which is the cause of the snf1 inviability.

The authors then use nanostring probes for a set of genes defined as mig1, mig2 or mig1/mig2 responsive, and probe the consequences of various perturbations. Implementing the Sak1 response needs both mig1 and mig2, but there are genes that seem to specifically require one or the other. Snf1 also seems to go through mig1/2, with some specific differences from the Sak1 data set.

Other phenotypes are probed – response to wall stress, ability to make hyphae, to interact with immune and endothelial cells. Mig1/2 function is needed, either gene alone does a pretty good job, so the proteins appear close to redundant for these functions. The data in general are convincing and sensibly interpreted (with the caveats noted below), and the results suggest logical explanations for otherwise somewhat surprising results such as the difference in snf1 and sak1 phenotypes and the essentiality of snf1.

My major concern with this work is the context in which it is embedded. At one level, the sak1, snf1, mig1 circuit controls glucose repression in C. albicans essentially as it does in S. cerevisiae, so nothing fundamental has been changed in or added to our understanding of the process. To justify the work to the audience of PLoS Genetics the authors want to note the way C. albicans has structured the circuit relative to S. cerevisiae – to emphasize the differences. But here they seem to somewhat lose their way. The fact that in C. albicans the paralogs Mig1 and Mig2 act similarly, but have a selective bias in targets is stated as surprising, when it is exactly what should happen with paralogs in this situation; they diverge in function, but modestly. The suggestion that the trajectory of Mig function “evolution” is that it has expanded in C. albicans relative to S. cerevisiae seems less likely than that the arrangement in the post WGD yeast S. cerevisiae has compacted relative to the pre WGD yeast C. albicans. If the authors are going down this route they should be providing genomic evidence from phylogenetically intermediary species, and showing the real relationships among the Mig orthologs, as previously noted.

Minor points

The authors have left PMID identifiers instead of actual references in a few places.

Reviewer #3: This work addresses the SNF1/ AMP-activated protein kinase signaling system in C. albicans. RNA seq was used to probe the relationship between the transcriptional regulators of carbon source utilization, Mig1 and Mig2, in Candida albicans, and their upstream regulators, the kinases Snf1 and Sak1. Deletion mutants in the genes encoding these regulators were created singly and in combination with each other, and the effects of the mutations on the transcriptional profile of mutants was examined. Nanostring analysis of 28 genes was used to further examine the effects of these mutations on gene expression. Growth, filamentation and biofilm phenotypes of the mutants were analyzed. Lastly, the behavior of mig1 mig2 double mutants during the interaction with model host cells, macrophages and human umbilical vein endothelial cells, was tested.

The authors first established, by RNA seq, the genes differentially expressed in wild type cells growing in the preferred carbon source glucose, or an alternative carbon source, glycerol, in complex rich medium; they found an expected assortment of genes, the majority of which were upregulated. They next obtained RNA seq datasets for deletion mutants in MIG1, MIG2 and MIG1MIG2 grown in glucose, the carbon source in which the activity of these regulators is predicted to be most important, and again found a majority of genes with increased expression among the total of genes with altered expression levels. Presence of a binding motif for Mig1 (previously established by others) was examined for promoters of genes whose expression was altered in these mutants, and was found to occur at significant frequency in genes whose expression was increased, but not those whose expression was decreased in mig1, mig2 and mig1 mig2 mutants. (No further analysis was attempted of this observation, though an important consideration is whether these regulators control the expression of other transcriptional activators that are responsible for loss of expression of this set of genes.)

The investigators then turned to the specific roles played by Mig1 and Mig2. They found a set of genes whose expression levels were increased only when MIG1 was deleted. Among them were plasma membrane hexose transporters, HGT genes, of which 8 were expressed at increased levels. Genes whose expression increased only when MIG2 was deleted included genes required for lipid catabolism; hence these regulators have distinct specificities in the processes they target.

Most genes whose expression was increased in the mig1 mig2 mutant were unchanged in either of the single mutants; the authors concluded that Mig1 and Mig2 functions largely overlap.

Constructing double mutants in SNF1 with MIG1, MIG2 or MIG1 MIG2, the authors were able to recover snf1 mutants only in a background in which MIG1 was deleted. They then examined growth phenotypes of sak1 mutants and found that their growth defects were suppressed in mutants also lacking MIG2, and further suppressed in when MIG1 was additionally, but not solely, lacking.

The authors then used a panel of 28 nanostring probes to investigate the effects of mutating these regulators singly and in combination. They observed very small numbers of differentially regulated genes among that set. The mutants were then subjected to cell wall stress using caspofungin, to which mutants lacking Mig1 were hypersensitive. Filamentation and biofilm phenotypes of mig1mig2 double mutants were shown, and the same mutants were defective in damaging macrophages and endothelial cells.

The role of the orthologous genes and their products, in particular SNF1, as an ortholog of the mammalian AMP kinase, has been studied extensively in S. cerevisiae, and in other work in C. albicans. The study is a useful compilation of datasets and shows unique results specific to C. albicans, and as such makes a significant contribution to the field of Candida albicans physiology. Superficiality in experimental design and of the analysis, to some extent limits the potential of this manuscript to move our understanding forward.

Major issues:

The analysis is somewhat superficial at several points. E.g.:

1. Is it really an “overlap of the functions” of Mig1 and Mig2 that leads to increased expression of a majority of altered genes in the mig1mig2 double mutant, but not in either single mutant? Certainly redundant activities of these transcriptional regulators could explain this observation. Alternatively, a requirement for cooperative binding at the promoters of these genes, for mutual recruitment, for nucleosome placement etc. (“combinatorial cis-regulation”) (PMID: 19029883) is possible; the authors do not consider any alternative possibilities despite the burgeoning field investigating these regulatory mechanisms, in which Mig1 is a model. Instead they simply state, for genes whose increased expression in the absence of one of the regulators, is further increased when the other is also absent, “We infer for these genes that Mig1 or Mig2 is the major repressor, but that the other Mig1/2 protein can repress weakly in the absence of the first. Therefore, while Mig 1 nd Mig2 each have some specific target genes, they also have highly redundant roles in gene regulation.”

This a priori assumption of redundant functions of these transcriptional regulators, in light of today’s large literature on this subject - (E.g.: “Understanding these rules, which are often referred to as ‘cis-regulatory grammar’, is a major challenge in modern biology,” PMID: 30570483 or “discuss how different strategies, including extensive cooperative regulation (both direct and indirect), progressive priming of regulatory elements, and the integration of activities from multiple enhancers, confer specificity and robustness to transcriptional regulation during development” PMID: 22868264) - does not seem to be a sufficient consideration of the regulatory mechanism involved.

If the “overlap of functions” idea is correct, why is the expression of 14% of genes, whose expression is higher, in mig1 or mig2 single mutants or in both than in wild type, not higher in the mig1 mig2 mutant (38/268) (Fig. 2A)? Among the genes expressed at higher levels in the mutants than the wild type, this fraction amounts to fully one third (72/216)?

Similarly, the sentences starting with “The majority of genes that are up-regulated in the mig1D/D mig2D/D double mutant are not up-regulated …” and ending with “and we refer to them as Mig1/2 shared genes” seem very unclear at best.

2. As 8 HGT genes (though I count 10 in Fig. 2C?) are expressed at elevated levels in mig1 mutants growing in glucose, Mig1 seems to act as a repressor of these genes in glucose. Is it worth considering which hexoses may the products of these genes transport as apparently it is unlikely to be glucose? Or if it is glucose, why does Mig1 normally repress them while cells are growing in glucose?

3. Using a panel of only 28 nanostring probes seems insufficient to actually move the understanding of these regulators forward. In fact, only 24 are used, because 4 probes examine the expression of the genes under study, i.e. MIG1, MIG2, SNF1, SAK1. How were these 24 genes chosen? If one were to decide that only 24 genes are sufficient, would one not choose only effectors of the processes under study, e.g. enzymes, while avoiding regulators like the transcription factors TYE7 or ACE2? Having such a small sample of genes in the set, it is inaccurate to write “Nanostring analysis indicated that there may be fairly few Mig1-selective target genes expressed in a snf1D/D mig1D/D mutant.”

4. How can a statement that “Mutations in MIG1 and MIG2 are epistatic to mutations in SAK1 and SNF1” (Fig. 4 legend) be made when no mutants in MIG1 or MIG2 alone were tested? The only mutants tested in this figure are double and triple mig1 and mig2 mutants with sak1 or snf1 mutations. While for snf1, single mutations in this gene lead to inviability, this is not the case for sak1. With this experimental design, how can the authors claim that their observed results are not due e.g. to synthetic effects between the mutations? - From S. cerevisiae, it is known that Mig1 and Mig2 act downstream of Sak1 and Snf1 so the statement about epistasis seems very logical to the reader; but if the authors aim to be showing an original result and drawing a legitimate conclusion from their own finding, the experimental design must be such that this conclusion can be drawn. To remind the authors of “epistasis,” “For any pair of genes X and Y, we evaluated the level of epistasis by comparing the fitness WXY of the double mutant with the product of the fitness values WX and WY of the corresponding single mutants,” (PMID: 15592468)

5. Issues not considered experimentally e.g. other non-glucose carbon sources, either fermentable (e.g., fructose, galactose, maltose) or nonfermentable (e.g., oleic acid, since Mig2 seemed to target lipid catabolism genes) should at least have been considered in the discussion?

6. Would it not have contributed to the analysis to use a loss-of-function strain in SNF1 other than the homozygous deletion that was known to be inviable? (e.g. the heterozygote, a strain expressing a destabilized transcript as in PMID:31118301, etc. )

6. In S. cerevisiae, participation of the heterotrimeric SNF1 holoenzyme in multiple signaling pathways is long established (PMID: 17981722); the Snf1 catalytic subunit complexes not only with a gamma subunity Snf4 but with one of 3 possible beta subunits (Gal83, Sip1 or Sip2) which regulate its subcellular localization. Its regulation involves not just an upstream activating kinase like Sak1 (or Elm3 or Tok1), but also a specific phosphatase (regulatory subunit Reg1 and the catalytic subunit Glc7) and autoinhibition modulted by the gamma subunit. Among multiple regulatory mechanims including control of transcriptional regulators beyong Mig proteins, Snf1 controls the activity of metabolic enzymes by phosphorylation, such as Pfk27, Gpd2, and Acc1 (PMID: 25512067 ). SNF1 also is activated in response to stressors like alkaline stress and sodium stress, with differential responses: accumulation in the nucleus in response to alkaline but not sodium stress (PMID:17438333). Given the complexity already known from model yeast and other eukaryotes, positing simple linear models for C. albicans is not likely to optimize our understanding of the role of this important regulator. A linear model of signaling from Sak1 -> Snf1 -> Mig1 and Mig2 as in Fig. 7 clearly cannot adequately represent the findings. If the fungal cell required only linear signaling, presumably a two component system as in bacteria would suffice to respond to the availability of glucose. A recent paper comparing SNF1 signaling of Kluyveromyces lactis with that of S. cerevisiae shows how this might be done; they devised a less crude model of their findings (PMID: 26440109).

Minor issues:

1. A reviewer’s work is made more time consuming by the absence of line numbering in the manuscript.

2. “Carbon-source dependent transcriptional regulation of ScMIG1 and ScMIG2 expression has apparently been lost for C albicans MIG1 and MIG2.” Unless the authors have evidence that the ancestral trait is carbon-source dependent transcriptional regulation of the MIG genes, while constitutive expression of these regulators is a derived trait, it is inappropriate to refer to such an evolutionary relationship. They should simply state the observation.

3. This reviewer finds the expression “up-“ or “downregulated” to be somewhat inaccurate when describing the effect of mutations of transcriptional regulators on the expression of genes. Some of the effects of these mutations may be due to regulation, i.e. compensatory mechanisms of the cell to adapt to loss of a regulator; these effects are likely to be indirect. The most direct effects of the absence of these regulators, however, are aberrantly high or low expression levels of their target genes, i.e. elevated or diminished in comparison with the wild type.

4. Setting up a hypothesis that mig1 mig2 mutants might be more fit during phagocytosis by macrophages seems a bit contrived and is not required to justify this experiment. It would be highly surprising if loss of 2 important transcriptional regulators were to increase fitness in a model host interaction. These phenotype experiments with the mutants (caspofungin sensitivity, biofilm formation, interaction with macrophages and endothelial cells) are of interest and do not need artificial justification.

**********

Have all data underlying the figures and results presented in the manuscript been provided?

Large-scale datasets should be made available via a public repository as described in the PLOS Genetics data availability policy, and numerical data that underlies graphs or summary statistics should be provided in spreadsheet form as supporting information.

Reviewer #1: No: Transcriptomics data may still need to be deposited.

Reviewer #2: Yes

Reviewer #3: Yes

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

2 Dec 2019

Submitted filename: Response to Reviews.pdf

Click here for additional data file.

18 Dec 2019

* Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out. *

Dear Aaron,

Thank you very much for submitting your revised Research Article entitled 'Roles of Candida albicans Mig1 and Mig2 in glucose repression, pathogenicity traits, and SNF1 essentiality' to PLOS Genetics. Your manuscript was fully evaluated at the editorial level and by the same three independent peer reviewers who reviewed the original submission. The reviewers appreciated the attention to an important topic.  All three reviewers recommended acceptance.  Reviewers 1 and 2 indicated that all of their comments raised had been addressed.  Reviewer 1 had one very minor comment about a supplemental table.  Reviewer 3 raised two minor remaining issues that may be considered stylistic, or may have something more to them.  In addition, this reviewer noted two minor issues that may require correction or editing.  We are therefore returning the manuscript to you as we expect that you may want to address these points raised in some fashion by judicious editing.  It is not our intent to then subject this to another round of review, as all three reviewers indicated that from their perspective the manuscript was acceptable.

We ask that you:

1) Provide your responses to the review comments and a description of any changes you have made in finalizing the manuscript.

2) Upload a Striking Image with a corresponding caption to accompany your manuscript if one is available (either a new image or an existing one from within your manuscript). If this image is judged to be suitable, it may be featured on our website. Images should ideally be high resolution, eye-catching, single panel square images. For examples, please browse our archive. If your image is from someone other than yourself, please ensure that the artist has read and agreed to the terms and conditions of the Creative Commons Attribution License. Note: we cannot publish copyrighted images.

We hope to receive your revised manuscript within the next 7 days. If you anticipate any delay in its return, we would ask you to let us know the expected re-submission date by email to plosgenetics@plos.org.

If present, accompanying reviewer attachments should be included with this email; please notify the journal office if any appear to be missing. They will also be available for download from the link below. You can use this link to log into the system when you are ready to submit a revised version, having first consulted our Submission Checklist.

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org.

Please be aware that our data availability policy requires that all numerical data underlying graphs or summary statistics are included with the submission, and you will need to provide this upon resubmission if not already present. In addition, we do not permit the inclusion of phrases such as "data not shown" or "unpublished results" in manuscripts. All points should be backed up by data provided with the submission.

PLOS has incorporated Similarity Check, powered by iThenticate, into its journal-wide submission system in order to screen submitted content for originality before publication. Each PLOS journal undertakes screening on a proportion of submitted articles. You will be contacted if needed following the screening process.

To resubmit, you will need to go to the link below and 'Revise Submission' in the 'Submissions Needing Revision' folder.

[LINK]

Please let us know if you have any questions while making these revisions.  We very much appreciate your submitting this interesting study to PLOS Genetics, and look forward to receiving the finalized manuscript which will be accepted and published.

Yours sincerely,

Joe

Joseph Heitman, MD, PhD

Associate Editor

PLOS Genetics

Gregory P. Copenhaver

Editor-in-Chief

PLOS Genetics

Reviewer's Responses to Questions

Comments to the Authors:

Please note here if the review is uploaded as an attachment.

Reviewer #1: The revised manuscript is a pleasure to read and represents a much more focused story that will be easier for readers to launch from. I have read through the paper and analyzed the new figures, and have no concerns.

The authors may consider a brief legend for supplemental tables to help with navigation within complex tables (e.g. Table S4).

Reviewer #2: Revision has dealt well with my concerns

Reviewer #3: The revised manuscript is improved and much clearer. Essentially it is a transcriptional analysis of the role of MIG1 and MIG2, which is worthwhile. It suggests that it is the inability to relieve otherwise constitutive transcriptional repression of Mig1 that makes snf1 mutants inviable.

This reviewer is still somewhat uncomfortable with the manuscript’s manner in setting up results as surprises or answers to mysteries or puzzles e.g.

- essentiality of SNF1; the original Petter et al report in 1997 discussed the possibility of Snf1 possessing “some other unknown cellular functions” unrelated to carbon source adaptation not as an explanation for their inability to isolate homozygous null mutants in this gene, but because of the heterozygotes’ phenotypes of wild type carbon source utilization patterns together with a significant decrease in growth rate as well as morphological changes.

- the finding of decreased virulence behaviors of mig1 mig2 mutants in a macrophage model, when metabolic flexibility has been described by many authors as a critical virulence determinant in C. albicans (e.g. Ramirez-Zaval et al 2017). The rationale presented by the authors in their response is not convincing to this reviewer - 3 out of the 4 important genes whose mutants they cite as showing increased virulence have not been specifically studied for this phenotype to my knowledge, and the 4th, PHO4, has decreased virulence in several in vivo models.

This is a matter of taste more than anything else and just a suggestion to the authors that their readers may appreciate a measured presentation of rationales for experiments.

Minor questions: why does the sak1 mutant appear twice in the heat map of Fig. 5?

Fig. 1B is mentioned in the text after Fig 2?

**********

Have all data underlying the figures and results presented in the manuscript been provided?

Large-scale datasets should be made available via a public repository as described in the PLOS Genetics data availability policy, and numerical data that underlies graphs or summary statistics should be provided in spreadsheet form as supporting information.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

19 Dec 2019

Submitted filename: Response to reviews PGENETICS-D-19-01346R1 revision.pdf

Click here for additional data file.

20 Dec 2019

Dear Aaron,

We are pleased to inform you that your revised manuscript entitled "Roles of Candida albicans Mig1 and Mig2 in glucose repression, pathogenicity traits, and SNF1 essentiality" has been editorially accepted for publication in PLOS Genetics. Congratulations!  We very much appreciate your detailed and clear responses to reviews, and the attention to detail in finalizing this for publication.

Before your submission can be formally accepted and sent to production you will need to complete our formatting changes, which you will receive in a follow up email. Please be aware that it may take several days for you to receive this email; during this time no action is required by you. Please note: the accept date on your published article will reflect the date of this provisional accept, but your manuscript will not be scheduled for publication until the required changes have been made.

Once your paper is formally accepted, an uncorrected proof of your manuscript will be published online ahead of the final version, unless you’ve already opted out via the online submission form. If, for any reason, you do not want an earlier version of your manuscript published online or are unsure if you have already indicated as such, please let the journal staff know immediately at plosgenetics@plos.org.

In the meantime, please log into Editorial Manager at https://www.editorialmanager.com/pgenetics/, click the "Update My Information" link at the top of the page, and update your user information to ensure an efficient production and billing process. Note that PLOS requires an ORCID iD for all corresponding authors. Therefore, please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field.  This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager.

If you have a press-related query, or would like to know about one way to make your underlying data available (as you will be aware, this is required for publication), please see the end of this email. If your institution or institutions have a press office, please notify them about your upcoming article at this point, to enable them to help maximise its impact. Inform journal staff as soon as possible if you are preparing a press release for your article and need a publication date.

Thank you again for supporting open-access publishing; we are looking forward to publishing your work in PLOS Genetics!  All best for the holiday season ahead.

Yours sincerely,

Joe

Joseph Heitman, MD, PhD

Associate Editor

PLOS Genetics

Gregory P. Copenhaver

Editor-in-Chief

PLOS Genetics

www.plosgenetics.org

Twitter: @PLOSGenetics

----------------------------------------------------

Comments from the reviewers (if applicable):

----------------------------------------------------

Data Deposition

If you have submitted a Research Article or Front Matter that has associated data that are not suitable for deposition in a subject-specific public repository (such as GenBank or ArrayExpress), one way to make that data available is to deposit it in the Dryad Digital Repository. As you may recall, we ask all authors to agree to make data available; this is one way to achieve that. A full list of recommended repositories can be found on our website.

The following link will take you to the Dryad record for your article, so you won't have to re‐enter its bibliographic information, and can upload your files directly:

http://datadryad.org/submit?journalID=pgenetics&manu=PGENETICS-D-19-01346R2

More information about depositing data in Dryad is available at http://www.datadryad.org/depositing. If you experience any difficulties in submitting your data, please contact help@datadryad.org for support.

Additionally, please be aware that our data availability policy requires that all numerical data underlying display items are included with the submission, and you will need to provide this before we can formally accept your manuscript, if not already present.

----------------------------------------------------

Press Queries

If you or your institution will be preparing press materials for this manuscript, or if you need to know your paper's publication date for media purposes, please inform the journal staff as soon as possible so that your submission can be scheduled accordingly. Your manuscript will remain under a strict press embargo until the publication date and time. This means an early version of your manuscript will not be published ahead of your final version. PLOS Genetics may also choose to issue a press release for your article. If there's anything the journal should know or you'd like more information, please get in touch via plosgenetics@plos.org.

13 Jan 2020

PGENETICS-D-19-01346R2

Roles of Candida albicans Mig1 and Mig2 in glucose repression, pathogenicity traits, and SNF1 essentiality

Dear Dr Mitchell,

We are pleased to inform you that your manuscript entitled "Roles of Candida albicans Mig1 and Mig2 in glucose repression, pathogenicity traits, and SNF1 essentiality" has been formally accepted for publication in PLOS Genetics! Your manuscript is now with our production department and you will be notified of the publication date in due course.

The corresponding author will soon be receiving a typeset proof for review, to ensure errors have not been introduced during production. Please review the PDF proof of your manuscript carefully, as this is the last chance to correct any errors. Please note that major changes, or those which affect the scientific understanding of the work, will likely cause delays to the publication date of your manuscript.

Soon after your final files are uploaded, unless you have opted out or your manuscript is a front-matter piece, the early version of your manuscript will be published online. The date of the early version will be your article's publication date. The final article will be published to the same URL, and all versions of the paper will be accessible to readers.

Thank you again for supporting PLOS Genetics and open-access publishing. We are looking forward to publishing your work!

With kind regards,

Kaitlin Butler

PLOS Genetics

On behalf of:

The PLOS Genetics Team

Carlyle House, Carlyle Road, Cambridge CB4 3DN | United Kingdom

plosgenetics@plos.org | +44 (0) 1223-442823

plosgenetics.org | Twitter: @PLOSGenetics