The Ndr/LATS kinase Cbk1 regulates a specific subset of Ace2 functions and suppresses the hyphae-to-yeast transition in Candida albicans

The Regulation of Ace2 and Morphogenesis (RAM) pathway is an important regulatory network in the human fungal pathogen Candida albicans. The RAM pathway’s two most well-studied components, the NDR/Lats kinase Cbk1 and its putative substrate, the transcription factor Ace2, have a wide range of phenotypes and functions. It is not clear, however, which of these functions are specifically due to the phosphorylation of Ace2 by Cbk1. To address this question, we first compared the transcriptional profiles of CBK1 and ACE2 deletion mutants. This analysis indicates that, of the large number of genes whose expression is affected by deletion of CBK1 and ACE2, only 5.5% of those genes are concordantly regulated. Our data also suggest that Ace2 directly or indirectly represses a large set of genes during hyphal morphogenesis. Second, we generated strains containing ACE2 alleles with alanine mutations at the Cbk1 phosphorylation sites. Phenotypic and transcriptional analysis of these ace2 mutants indicates that, as in Saccharomyces cerevisiae, Cbk1 regulation is important for daughter cell localization of Ace2 and cell separation during yeast phase growth. In contrast, Cbk1 phosphorylation of Ace2 plays a minor role in C. albicans yeast-to-hyphae transition. We have, however, discovered a new function for the Cbk1-Ace2 axis. Specifically, Cbk1 phosphorylation of Ace2 prevents the hyphae-to-yeast transition. To our knowledge, this is one of the first regulators of the C. albicans hyphae-to-yeast transition to be described. Finally, we present an integrated model for the role of Cbk1 in the regulation of hyphal morphogenesis in C. albicans. Importance Regulation of Ace2 and Morphogenesis (RAM) pathway is a key regulatory network that plays a role in many aspects of C. albicans pathobiology. In addition to characterizing the transcriptional effects of this pathway, we discovered that Cbk1 and Ace2, a key RAM pathway regulator-effector pair, mediate a specific set of the overall functions of the RAM pathway. We have also discovered a new function for the Cbk1-Ace2 axis; suppression of the hyphae-to-yeast transition. Very few regulators of this transition have been described and our data indicate that maintenance of hyphal morphogenesis requires suppression of yeast phase growth by Cbk1-regulated Ace2.


Importance 22
Regulation of Ace2 and Morphogenesis (RAM) pathway is a key regulatory network that plays a 23 role in many aspects of C. albicans pathobiology. In addition to characterizing the transcriptional 24

Introduction 19
Ace2 (we will use Ace2 to refer to the C. albicans ortholog and ScAce2 to refer to the S. 1 cerevisiae ortholog for the remainder of the text) lacks the C-terminal Cbk1 phosphorylation 2 motif but has sites in the NES and in the N-terminal region (Fig. 3B). Willger et al. performed a 3 comprehensive phosphoproteomic study of C. albicans during hyphae induction and confirmed 4 that Ace2 S136 and S151 are phosphorylated (30). There is evidence that, in the SC5312 strain 5 background (23), Ace2 has two isoforms depending on the translational start site (Fig. 3B). The 6 N-terminal Cbk1 phosphosite is between the two start sites and, thus, would only be expected to 7 be present in the Ace2L form that is proposed to associate with the plasma membrane and play 8 a role in septin dynamics. Ace2S is the isoform that localizes to the nucleus and regulates the 9 expression of cell separation genes (23). 10 To assess the role of Cbk1 phosphorylation in Ace2 function, we used a CRISPR/Cas9 11 approach to mutate the serine and threonine Cbk1 consensus phosphorylation sites of the 12 endogenous ACE2 to alanine (33). The details of this strain construction are described in the 13 materials and methods. In this way, we constructed a strain in which the only ACE2 allele 14 lacked the Cbk1 phosphorylation sites in the NES (S136A, S151A, ace2-2A) and a strain that 15 lacks all three of the Ace2, Cbk1 phosphorylation sites (S49A, S136A, S151A, ace2-3A). The 16 initially isolated phosphosite mutants were heterozygous at the ACE2 locus with a wild type 17 allele and the desired mutant allele. We deleted the wild type allele using standard homologous 18 recombination and confirmed that the only remaining allele contained the S/T-A mutations by 19 Sanger sequencing. Thus, the resulting strains are heterozygous at the ACE2 locus and, if the 20 Cbk1 phosphorylation sites are required for function, the strains would be expected to have 21 phenotypes that are similar to an ace2∆∆ homozygous deletion. In our phenotypic 22 characterization data presented below, ace2∆∆ is used as the control; as shown in 23 Supplementary Data, ace2∆/ACE2 strains show no haploinsufficiency and thus phenotypic 24 changes are due to the mutations and not to changes in gene copy number (Fig. S1). 25 26 Phospho-acceptor amino-acids at Ace2 consensus Cbk1 substrate motifs are required 1 for normal cell separation in C. albicans during yeast phase growth 2 A phenotype of ACE2 and CBK1 mutants that is conserved across all yeast species 3 studied to date is decreased cell separation due to reduced septum degradation (8,9,15). 4 Therefore, we compared the cell separation characteristics of the ace2-2A and ace2-3A mutants 5 to wild type and ace2∆∆ mutant strain under standard yeast culture conditions (30 o C, YPD 6 media). The harvested cells were fixed and examined by light microscopy to determine the 7 relative proportion of small, medium, and large cell clusters (Fig. 4A). Consistent with previous 8 literature (8,9,15), WT cultures were dominated by small clusters (1-2 cell) while the ace2∆∆ 9 mutant cultures were almost entirely comprised of large cell clusters (>5 cells, Fig. 4A). The 10 ace2-2A mutant showed an intermediate phenotype with increased numbers of medium and 11 large cell clusters that were not changed significantly by mutation of the non-NES Cbk1 site 12 (ace2-3A). As previously found for ace2∆∆ mutants (8,9,15), ultrasonication of the cell clusters 13 formed by the ace2-2A and ace2-3A mutants did not alter the size of the clusters which is 14 consistent with a failure of the cells to separate and is inconsistent with non-covalent 15 aggregation (data not shown). 16 As noted above, this phenotype is attributed to the decreased expression of septum-17 degrading enzymes such as CHT3, DSE1, and SCW11 in the ace2∆∆ mutant (Fig. 4B&C). The 18 expression of these genes is also decreased in both ace2-2A and ace2-3A mutants during both 19 yeast and hyphal cell growth (Fig. 4B&C). As with the cell separation data, mutations within the 20 NES site are responsible for the majority of the observed effects on septum-degrading gene 21 expression. This is consistent with the findings that Ace2S form of the protein is responsible for 22 transcriptional regulation of cell separation genes while the site in the Ace2L form does not play 23 a significant role in the expression of these genes (23). These data are consistent with previous 24 studies in S. cerevisiae indicating that Cbk1 phosphorylation of the NES in Ace2 is required for 25 expression of cell septum-degrading enzymes and proper cell separation (10). 12 1 Cbk1 phospho-acceptor sites in a putative nuclear export signal site are required to 2 concentrate Ace2 to daughter cell nuclei in C. albicans. 3 In S. cerevisiae daughter cells, phosphorylation of ScCbk1 consensus motifs in the NES 4 of ScAce2 prevents its export and, thereby, concentrates ScAce2 within the nuclei of new 5 daughter cells while it is relatively excluded from mother cell nuclei (10). Mutations that prevent 6 ScCbk1-mediated phosphorylation of ScAce2 result in the transcription factor localizing to both 7 daughter and mother cell nuclei (10). To determine if a similar process occurs in C. albicans, 8 we tagged the C-terminus of ACE2 in wild type and ace2-2A strains with green fluorescent 9 protein (GFP); we and others have shown previously that WT alleles retain function (15, 34) and 10 we confirmed that the phenotypes of the ace2-2A-GFP strains did not differ from the parental 11 strain (Fig. S2). 12 Consistent with previous reports, WT Ace2 localized to the daughter cell in budding 13 yeast phase (Fig. 5A). In contrast, ace2-2A-GFP signal was present in both cells of mother-14 daughter pairs (Fig. 5B). This same pattern of mislocalization is consistent with that reported by 15 the Weiss lab for strains containing ScACE2 alleles lacking Cbk1 phosphoacceptor sites in the 16 NES domain of the protein (10). This localization pattern also provides an explanation for the 17 intermediate cell separation phenotype (relative to the ace2∆∆ strain) observed for the ace2-2A 18 strains because some cells localized Ace2 to the daughter cell nuclei whereas in the null mutant 19 there is a complete absence of protein. We attempted to characterize the effect of the ace2-2A 20 allele on localization in hyphae but the relatively few wild type hyphae that show the canonical 21 Ace2 localization to the distal most nuclei of the hyphae prevented our ability to make firm 22 conclusions (Wakade and Krysan, unpublished results). These data indicate that, as in S. 23 cerevisiae, Cbk1 phosphorylation of Ace2 within the putative NES is required to concentrate 24 Ace2 in daughter cell nuclei during yeast phase growth. 25 26 Cbk1 phosphorylation of Ace2 affects susceptibility to chitin-targeted cell wall stressors 1 Ace2 function has been implicated in maintaining cell wall homeostasis through multiple 2 pathways (7). First, ace2∆∆ mutants are resistant to the chitin binding molecule, calcofluor 3 white (CFW). This is most likely due to an increase in the chitin content of the wall in the region 4 of septum due to decreased expression of chitinases. Consistent with our observation that the 5 ace2-2A/3A mutants have decreased expression of CHT3, they are also resistant to 6 concentrations of CFW that inhibit the growth of WT cells (Fig. 6A); similarly, both ace2∆∆ and 7 the ace2-2A/3A mutants are resistant to Congo Red, a molecule that interacts with both chitin 8 and glucan components of the cell wall (Fig. 6B). Ace2 has also been shown to affect the 9 mannoprotein layer of the cell wall through a separate signaling pathway involving Cek1 (21); a 10 manifestation of this function is the increased susceptibility of the ace2∆∆ mutant to the 11 glycosyl-transfer inhibitor, tunicamycin (Fig. 6C). The ace2-2A/3A mutants grow similar to wild 12 type cells at concentrations of tunicamycin that inhibits ace2∆∆ growth, indicating that this 13 function of Ace2 is independent of Cbk1. Thus, the cell wall functions regulated by Cbk1-Ace2 14 appear to be mainly limited to cell septum-related processes while other cell wall functions of 15 Ace2 are independent of Cbk1. As a result, the colony morphology of ace2∆∆ mutants on YPD plates at 30 o C takes on a 23 wrinkled or scalloped appearance while wild type strains form typical smooth colonies (Fig. 7A). 24 This well-described phenotype suggests that Ace2 may play a role in suppressing the hyphal 25 morphogenesis program in daughter yeast cells during non-inducing conditions. As shown in 26 14 We had previously reported that ace2∆∆ mutants undergo hyphal morphogenesis in 19 liquid media containing a variety of inducing agents (32), although the tempo of hyphae 20 formation is modestly slower in SM. Upon re-examination of this process, we noted that ace2∆∆ 21 strains developed lateral yeast cells at sub-apical segments of the hyphal filament before they 22 were evident on WT hyphae (Fig. 8A). In addition, we observed yeast phase budding from the 23 mother cells from which the hyphal structure emerged, indicating that, in the absence of ACE2, 24 hyphal-mother cells had re-entered the cell cycle earlier than WT cells ( Fig 8A). As hyphae 25 mature, a hyphae-to-yeast transition eventually occurs leading to the emergence of yeast buds 26 at sub-apical cell compartments within the hyphae (31, 35). Quantification of this observation 1 confirmed that at 4 hours induction, only 10% of WT hyphae displayed lateral yeast cells at sub-2 apical segments whereas 35% of ace2∆∆ mutants had formed lateral yeast cells (Fig. 8B). 3 Finkel et al. had previously reported that Snf5 regulates ACE2 expression in SM and our 4 inspection of photomicrographs of snf5∆∆ mutants suggested that there may be increased 5 lateral yeast formation in these strains as well (18). Consistent with that assessment, 80% of 6 hyphae formed by the snf5∆∆ mutant have lateral yeast at a time point when wild type cells 7 have only 10% (Fig. 8B). Our data indicate that Ace2 and Snf5 repress lateral yeast formation 8 and further suggest that Snf5 and Ace2 are required to ensure that the transition from hyphae to 9 yeast does not occur prematurely. In addition, it appears that ace2∆∆ is required to prevent 10 hyphal mother cells from re-entering the cell cycle prematurely. 11 We next asked whether Cbk1 was required for the ability of Ace2 to suppress lateral 12 yeast cell formation during the maintenance phase of hyphae formation. As shown in Fig The hyphae-to-yeast transition has not been studied to nearly the same extent as the 24 yeast-to-hyphae transition (31,35). In pioneering work, the Koehler lab found that PES1, a 25 pescadillo homolog, is required for the hyphae-to-yeast transition and is essential in yeast 26 phase cells but not during hyphal phase growth (31). Conversely, increased expression of 1 PES1 induces increased lateral yeast formation. We, therefore, examined our RNA sequencing 2 data set to determine the expression level of PES1 in ace2∆∆. Indeed, PES1 expression is 3 increased 3.8-fold in hyphal ace2∆∆ relative to wild type but is actually reduced in yeast ace2∆∆ 4 cells (Tables S3 & S5); the same trend was observed by Mulhern et al. in their transcriptional 5 profile of ace2∆∆ mutants (16). To confirm this observation, the expression of PES1 was 6 measured by qRT-PCR in WT and ace2∆∆ strains after 3 hours of induction with SM (Fig. 9A). 7 The expression of PES1 was increased in the ace2∆∆ strain relative to wild type, confirming the 8 genome-wide expression profiling data and strains containing Ace2 Cbk1-phosphosite 9 mutations also have increased expression of PES1 (Fig. 9A). Consistent with the increased 10 lateral yeast formation in the ace2-3A strain relative to ace2-2A mutants, the ace2-3A mutant 11 shows a trend toward increased PES1 expression relative to the ace2-2A mutant (Fig.8B). 12 Interestingly, the archetypal hyphal specific gene HWP1 is expressed at increased levels in the 13 ace2∆∆ deletion and the Cbk1 phosphosite mutants as well (Fig. 9C), indicating that, despite 14 increased development of lateral yeast, the hyphal transcriptional program is still strongly 15 induced. Thus, the expression of well-characterized reporter genes for both the yeast-to-16 hyphae and the hyphae-to-yeast transcriptional programs is upregulated in ace2 mutants. This 17 expression profile suggests that the Cbk1-Ace2 axis plays an important role in regulating the 18 balance between morphogenic transcriptional programs during hyphae formation in C. albicans. 19 20

Discussion 21
The RAM pathway regulates a wide range of processes in C. albicans including cell 22 cycle-associated daughter cell separation, hyphal morphogenesis, cell wall integrity and 23 biosynthesis, biofilm formation, oxidative stress resistance, and mammalian infection (14-20). 24 Although the phosphorylation motif of the key kinase in the RAM pathway, Cbk1, has been well-25 described (10), the substrates that carry out the effector functions of the pathway remain largely 26 uncharacterized in C. albicans with the exceptions of Bcr1, Ssd1, and Fkh2 (27, 28, 29). For 1 example, prior to the work described above, it was unclear what functions of the RAM pathway 2 were due to its eponymous transcription factor Ace2. 3 Through transcriptional profiling of both cbk1∆∆ and ace2∆∆ mutants and genetic 4 analysis of strains containing mutations at the Cbk1 phosphosites, we have found that, although 5 both Cbk1 and Ace2 affect the expression of a large number of genes and have a number of 6 phenotypes, the Cbk1-Ace2 axis directly regulates a specific subset of these functions. During 7 yeast phase growth, the expression of only ~10% genes are concordantly affected in cbk1∆∆ 8 and ace2∆∆ mutants. The cell septum degrading enzymes CHT3 and SCW11, well 9 characterized effectors of the RAM pathway (7), are among these and are also downregulated 10 in Cbk1-phosphosite mutants of ACE2. As expected from studies of Cbk1 and Ace2 in S. 11 cerevisiae, the ACE2 phosphosite mutants have partial cell separation defects and alterations in 12 the daughter cell specific nuclear localization localization of Ace2 (10). 13 The small overlap between the expression profiles of cbk1∆∆ and ace2∆∆ indicates that 14 Cbk1 is likely to have additional transcriptional effectors and that Ace2 is likely to have other 15 regulatory partners. With regard to the former, Bcr1 (27) and Fkh2 (28) have been shown to be 16 Cbk1-regulated transcription factors. Bcr1 is important for biofilm formation but it's effect on 17 gene expression during yeast phase growth has not, to our knowledge, been characterized (36). 18 Fkh2, interestingly, plays a role in both hyphal morphology and cell separation. During yeast 19 phase growth, the expression of SCW11 and CHT2 is reduced in fkh2 mutants (28). It is 20 therefore possible that Cbk1 regulates cell separation genes through both Fkh2 and Ace2 and 21 that other Cbk1-regulated genes are Fkh2 targets. A survey of the annotated C. albicans 22 transcription factors indicates that an additional 27 have sequences that match Cbk1 23 phosphorylation motifs. Of these, a highly likely Cbk1 substrate is Ash1. Like Ace2 it is also a 24 daughter cell-associated transcription factor that plays a role in hyphae formation under specific 25 conditions. Its expression is also dependent upon both Cbk1 and Ace2 (Table S2&3). Ash1 has 26 a HTRSRS Cbk1-consensus sequence in the N-terminus and S22 is phosphorylated during 1 both yeast and hyphal phase growth according to two large-scale phosphoproteomic studies in 2 C. albicans (30,38). 3 It is also likely that another contributing factor to the role of Cbk1 in gene expression is 4 its regulation of Ssd1, an RNA binding protein that suppresses the expression of a specific set 5 of genes (29). In S. cerevisiae, ScCbk1 phosphorylates ScSsd1 to suppress its ability to target 6 bound mRNAs to the P-body for storage and/or degradation (39, 40, 41). This set of S. 7 cerevisiae genes is enriched in cell wall and cell separation genes but contains others as well 8 (11,39). In C. albicans, the only transcript that has been shown to be affected by Cbk1-Ssd1 is 9 Nrg1 which is down-regulated at the initiation of hypha formation (29). It is unlikely that this is 10 the only gene that is targeted by Ssd1, although additional studies will be needed to identify 11 these transcripts. Overall, Cbk1 has a broad effect on C. albicans gene expression through a 12 combination of its modulation of Ace2 activity, regulation of other transcription factors, and 13 inhibition of Ssd1. 14 Ace2 also has broad effects on gene expression beyond cell separation as noted above 15 ( Fig. 1&2, Tables S1 and S3). Willger et al. found that Ace2 is phosphorylated at 16 sites in 16 addition to those that match the Cbk1 motif, suggesting that other kinases are likely to be 17 involved in the regulation of Ace2 function (30). Indeed, genetic studies by Van Wijlick et al. 18 indicate that Ace2 functions in a kinase cascade comprised of Cst20, Hst7, and Cek1 to 19 regulate protein mannosylation and susceptibility to the glycosylation inhibitor tunicamycin (21). 20 Our data are consistent with their model in that the hypersusceptibility of ace2∆∆ to tunicamycin 21 is not recapitulated by ACE2 mutants lacking Cbk1 phosphorylation sites (Fig. 6). down-regulated in ace2∆∆ were bound by Ace2-HA (17). One possibility is that the HA-tag is 8 altering the function of Ace2 in a manner that is not evident from phenotypic analysis; although 9 previous studies have shown that C-terminally tagged Ace2 proteins bind to targets such as 10 SCW11 and CHT3 (42). It seems from these data and observations that additional experiments 11 are needed before a definitive conclusion can be made regarding direct targets of Ace2 and the 12 binding motifs that determine such targets. 13 Under many conditions, Ace2 is dispensable for C. albicans hyphal morphogenesis. 14 Despite this, transcriptional profiling data reported by Mulhern et al. (16) and reported herein 15 clearly indicate that it affects the expression of many genes during hyphal morphogenesis 16 (Fig.2, Table S3). Our data indicate that a significant set of genes are upregulated in the 17 absence of Ace2, suggesting that it functions either directly or indirectly as a repressor of gene 18 expression during hyphal growth. Interestingly, HWP1, a gene that is only expressed during 19 hyphae, and PES1, a gene associated with hyphae-to-yeast transition are both upregulated 20 (Fig. 9A&B). PES1 is essential during yeast phase growth and promotes the hyphae-to-yeast or 21 lateral bud formation in hyphal cells (31); it is not essential for hyphal growth and forced 22 expression drives lateral yeast cell formation in hyphae. Our data indicate that loss of Ace2 23 function and loss of Cbk1 regulation of Ace2 lead to increased expression of PES1 and early 24 formation of lateral yeast bud or hyphae-to-yeast transition (Fig.8&9). To our knowledge, this 25 represents a novel function of Ace2 during morphogenesis and, as such, Ace2 is one of only a 26 20 small set of genes that have been demonstrated to affect the hyphae-to-yeast transition (31, 39, 1

41). 2
We propose that the Cbk1-Ace2 axis functions during the maintenance phase of hyphae 3 formation to suppress the lateral yeast growth program. During hyphal morphogenesis, Ace2 4 RNA and protein levels are initially low and then increase to peak at approximately 5 hours post 5 induction in SM (32). Thus, except under specific conditions, Cbk1-regulated Ace2 does not 6 appear to be required for initiation of hyphal formation. However, Ace2 appears to play an 7 important role in sub-apical compartments by maintaining the hyphal transcriptional program or, 8 alternatively, suppressing the yeast program (see below). 9 We and others have shown that Ace2-GFP is most clearly localized to the nuclei of the 10 leading hyphal tip (15, 34). In addition to this well-established role in daughter cells, our data 11 indicate that Cbk1-regulated Ace2 also functions in sub-apical hyphal cells as well. Although we 12 have not been able to observe Ace2-GFP signal in the nuclei or cytoplasm of sub-apical hyphal 13 cells (Wakade and Krysan, unpublished data), we suspect that this is due to the low overall 14 expression of Ace2 from its native promoter; indeed, long exposure times are needed to 15 visualize Ace2-GFP in the nuclei of daughter yeast cells or hyphal tip cells in our experience. 16 The phenotypic data for the ace2∆∆ and ace2-2A/3A strains, however, provide compelling 17 evidence that Ace2 functions in the sub-apical and mother cell compartments of the hyphae. 18 Interestingly, our genetic data also indicate that the recently described Ace2L isoform 19 may be required for this function (Fig. 3, 8 &9; reference 23). Because Calderón-Noreña DM et 20 al. found that the Ace2S isoform is likely responsible for much of the transcriptional regulation 21 attributed to Ace2 (23), it is possible that suppression of the yeast program is not a direct result 22 of Ace2 binding to DNA; indeed, Ace2L is proposed to be membrane associated. Clearly, 23 additional work will be needed to fully characterize the mechanistic details of this newly 24 described Ace2 function. 25 Regardless of the specific molecular mechanism, the function of Ace2 in the sub-apical 1 and mother cell compartments of the hyphal structure is consistent with its known cell cycle 2 functions during yeast phase growth (9, 10) and the proposed cell cycle state of sub-apical 3 compartments of C. albicans hyphae (6, 43). Specifically, Ace2 is well-established to play a 4 critical role in early G1 during yeast phase growth in both S. cerevisiae (9, 10, 44) and C. 5 albicans (45). As the cell progresses through G1 the amount of nuclear Ace2 is reduced by 6 decreased ScCbk1 phosphorylation (44) and ScAmn1-mediated ubiquitin-mediated degradation 7 Ace2. This decrease in ScAce2 is associated with the transition to START (46). In contrast to 8 yeast growth, Ace2 protein levels increase as hyphal morphogenesis progresses (32), 9 suggesting that hyphal cells maintain an early G1-like state. Interestingly, CaAMN1 expression 10 is downregulated >3-fold relative to yeast cells after 4 hours of SM induction (Table S1 and S3), 11 suggesting that decreased Amn1-mediated proteasome degradation of Ace2 could contribute to 12 the steady increase in its protein level as hyphal morphogenesis progresses (32). 13 Sub-apical and mother cells of hyphae have been shown to be arrested in G1 phase (6, 14 44). Gow and colleagues have developed a model in which the highly vacuolated nature of the 15 sub-apical compartments limits the amount of cytoplasm leading to an effectively smaller cell 16 size (43,48). Since cell size is a critical determinant of the cell transitioning from G1 to START, 17 lateral yeast cell or branching does not occur until the ratio of cytoplasm to vacuole increases. 18 Our data indicate that Cbk1-phosphorylated Ace2 plays a role in maintaining G1 and inhibiting 19 the transition to START in both mother cells and sub-apical compartments. To our knowledge, 20 this is the first genetic evidence for proteins that function to repress the transition from G1 to 21 START in C. albicans hyphae. 22 Taking these new observations together with previous work from multiple labs including 23 our own allows us to construct the following model integrating the function of the Cbk1-Ace2 24 axis during hyphal morphogenesis under standard liquid medium induction conditions (Fig. 10). 25 After a hyphae-inducing signal is sensed, Cdc28 is activated (48)  We have previously shown that Brg1 directly activates ACE2 expression during hyphal 8 morphogenesis and that expression peaks during the maintenance phase of the process (32). 9 Finkel et al. found that Snf5 is also required for ACE2 expression (18) and we show that it is 10 also required for suppression of lateral yeast formation, indicating it plays a role in maintaining 11 proper levels of ACE2 during morphogenesis. Finally, as proposed above, Cbk1-phosphorylated 12 Ace2 functions, at least in part, to suppress lateral yeast formation, likely through delaying the 13 transition from G1 to START in subapical hyphal cell compartments. 14 In summary, Cbk1 regulates a specific set of Ace2 functions during both yeast and 15 hyphae phase growth in C. albicans. While both Cbk1 and Ace2 have pleotropic effects on cell 16 physiology, the majority of those effects are independent of one another. We have also 17 proposed an integrated pathway through which Cbk1 affects hyphae formation during standard 18 laboratory induction conditions which involves its regulation of the transcriptional regulators 19 Nrg1, Brg1, and Ace2. Lastly, our data indicate that Cbk1-Ace2 is required to delay the G1-20 START transition in hyphae and, thereby, suppress the hyphae-to-yeast transition. supplementary Tables S1 and S2, respectively. All heterozygous or homozygous strains of C. 12 albicans were constructed from SN152 background using the auxotrophic marker either LEU2 13 or ARG4. The ace2Δ/ACE2 strain was generated as described previously (26). Briefly, pCR4-14 TOPO plasmid carrying ACE2::LEU2 amplicon was digested with the SbfI enzyme and 15 subsequently the linearized plasmid further inserted into the SN87 background (56). The 16 transformants were selected on SD plates lacking leucine and correct integration was confirmed 17 by PCR using ACE2.P1 and ACE2.P2 primers. 18 The transient CRISPR-Cas9 system was used to generate strains with deletion 19 mutations (33). To generate ace2ΔΔ strains, a disruption cassette with 5' and 3' flanking 20 sequences homologous to the 5' and 3' regions of ACE2 was generated by amplification of 21 CmLEU2 cassette from pSN40 plasmid (56) with primer pairs ACE2.P3 and ACE2.P4. The 22 CaCas9 expression cassette was PCR amplified from plasmid pV1093 (33) and using CAS9.P1 23 and CAS9.P2 primers whereas sgRNA expression cassette was generated using the split-joint 24 PCR method (33). Briefly, in the first step the SNR52 promoter was amplified using primer pairs 1 ACE2.P5 and ACE2.P6 whereas sgRNA scaffold were PCR amplified by using ACE2.P7 and 2 ACE2.P8 primers. In the second step, the SNR52 promoter and sgRNA scaffold were fused by 3 primer extension in which the 20 bp guide RNA sequence act as complementary primers. In the 4 third round, the fused PCR product was PCR amplified by using the nested primers ACE2.P9 5 and ACE2.P10 to harvest the sgRNA cassette. For fungal transformation, 1 μg of CaCas9 6 cassette and 1 μg sgRNA cassette was co-transformed with the 3 μg of deletion construct, 7 using the standard lithium acetate transformation method (33, 56). 8 To generate a strain with serine/threonine to alanine mutations at Cbk1 consensus 9 phosphorylation sites, a fragment of ACE2 (1-622 bp) that contained the respective mutations at 10 T49A, S136A and S151A was synthesized by GeneScript and cloned into pUC19. This 11 fragment was used as a template to generate strains containing the ace2-2A (S136A, S151A) 12 and the ace2-3A (T49A, S136A, S151A) mutations. To do this, we used a double CRISPR 13 approach. We first isolated C. albicans ARG4 gene from a SC5314 background and PCR 14 amplified using ARG4.P1 and ARG4.P2 primers. An ARG4 cassette targeted to the putatively 15 neutral dpl200 locus was amplified using ARG4.P3 and ARG4.P4 primers. The 16 phosphodeficient ACE2 allele was amplified by using ACE2.P11 and ACE2.P12 primers. To 17 achieve the transformation, 1 μg of CaCas9 cassette, 1 μg of each sgRNA-ACE2 and sgRNA-18 ARG4 cassette was co-transformed with the 3 μg of ACE2-2A/3A along with 1 μg of ARG4 19 cassette. The transformants were selected on synthetic media lacking Arg. 20 To characterize the resulting transformants, the region of ACE2 corresponding to the 21 repair fragment was PCR amplified using ACE2.P11 and ACE2.P12 primers and analyzed by 22 Sanger sequencing.
From multiple transformations, the majority of isolates were 23 heterogeneous comprising of one copy of WT-ACE2 whereas other copy contained two (S136A, 24 S151A) or three (T49A, S136A, S151A) mutations in ACE2. We then deleted the remaining WT 25 copy of ACE2 as described earlier (26) and the transformants were selected on SD plates 1 lacking leucine and arginine. The correct integration was confirmed by PCR using ACE2.P1 and 2 ACE2.P2 primers and the presence of the ACE2 mutations was confirmed by Sanger 3 sequencing. In this way, we generated strains in which the only copy of ACE2 contained 4 mutations in the Cbk1 phosphorylation sites. 5 Strains with GFP fused to the C-terminus of ACE2 were generated by homologous 6 recombination using a cassette that was derived from pMG2120 (57) and primers ACE2.P13 7 and ACE2.P14. The PCR amplified NAT1-marked, GFP cassette was purified and transformed 8 into either the SN152 reference strain or ACE2-2A strain. Two independent clones of the 9 respective strain were generated, and correct integration was confirmed by PCR. The resulting 10 strains showed no change in growth or morphogenesis phenotypes relative to the parental 11 strains. All PCR amplified or cloned fragments were confirmed by sequencing. Correct 12 insertions of two independent clones were verified by PCR and used for the further experiments. Microscopy. For colony morphology analysis, plates were incubated for the indicated time and 23 imaged using Nikon ES80 epifluorescence microscope equipped with a CoolSnap charge 24 coupled device (CCD) camera using 40x magnification. Live cell fluorescent imaging was 1 carried out with the multi-photon laser scanning microscope (SP8, Leica Microsystems). 2 Overnight cultures in YPD media were collected and back diluted either into the fresh YPD or in 3 the hyphal inducing media to achieve either yeast or hyphal growth, respectively. Cells were 4 then harvested, washed twice into 1x PBS and resuspended into 1x PBS prior to imaging. The 5 samples were treated with Hoeschst 33342 and incubated for 15 minutes in the dark at room 6 temperature followed by washing twice with PBS. For Hoechst 33342 dye excitation wavelength 7 of 350 nm and emission wavelength of 461 nm was used whereas to visualize the GFP tag, 8 excitation wavelength of 395 nm and emission wavelength of 509 nm was used, and sequential 9 images were acquired using 25x water immersive objective lens with 3.32x zoom factor. Images 10 were further analyzed by using the ImageJ software. 11 Transcriptional analysis by quantitative reverse transcription-PCR (qPCR). Strains pre-12 cultured overnight in YPD at 30 o C, back-diluted into either fresh YPD or in hyphae-inducing 13 media and collected after 4 to 5 hours of incubation either at 30 0 or 37 0 C. RNA was isolated 14 using RiboPure TM kit and reverse transcribed using an iScript cDNA synthesis kit (170-8891; 15 Bio-Rad). The qPCR reaction was performed using IQ SyberGreen supermix (170-8882; Bio-16 Rad) and primers used in this study were listed in S3 table. Briefly, each reaction contained 10 17 μl of the SYBER Green PCR master mix, 0.10 μM of the respective primers and 150 ng of 18 cDNA as a template in a total volume of 20 μl. Data analysis was performed using 2 -ΔΔCT 19 method and ACT1 was used as an internal control. Data reported here are the means from 3 20 independent biological replicates performed in a triplicate. 21 22 23 29 ACKNOWLEDGEMENTS. This work was funded by NIH grants 1R01AI098450 (DJK) and 1 1R01AI33409 (DJK); the funders had no role in study design, data collection and interpretation, 2 or the decision to submit the work for publication. We thank Manning Huang and Aaron Mitchell 3 (Pittsburgh/University of Georgia) for helpful discussions regarding construction of the Ace2 4 mutants using CRISPR/Cas9 and for providing snf5∆∆ mutants. We thank Scott Moye-Rowley 5 (Iowa) and Melanie Wellington (Iowa) for helpful discussions.  for exclusive localization to daughter cell nuclei in yeast phase cells. Exponential cells for the 20 ACE2-GFP (A) and ace2-2A-GFP were harvested and stained with Hoechst before imaging with 21 the indicated channels. The outlines of the indicated cells were generated to highlight mother-22 daughter pairs. Scale bar 5 µm. 23 Figure 6. Cbk1 phospho-acceptor site mutants of ACE2 are resistant to chitin-binding molecules 1 but not tunicamycin. A ten-fold dilution series (0.1 OD 600 initial cell density) was plated on YPD 2 as well as YPD containing calcofluor white (50 µg/mL, A), Congo red (400 µg/mL) and 3 tunicamycin (4 µg/mL, C). The plates were incubated for 3 days at 30 o C before being 4 photographed. The images are representative of two to three independent replicates. 5