Novel determinants of cell size homeostasis in the opportunistic yeast Candida albicans

The basis for commitment to cell division in late G1 phase, called Start in yeast, is a critical but still poorly understood aspect of eukaryotic cell proliferation. Most dividing cells accumulate mass and grow to a critical cell size before traversing the cell cycle. This size threshold couples cell growth to division and thereby establishes long-term size homeostasis. At present, mechanisms involved in cell size homeostasis in fungal pathogens are not well described. Our previous survey of the size phenome in Candida albicans focused on 279 unique mutants enriched mainly in kinases and transcription factors (Sellam et al. PLoS Genet 15:e1008052, 2019). To uncover novel size regulators in C. albicans and highlight potential innovation within cell size control in pathogenic fungi, we expanded our genetic survey of cell size to include 1301 strains from the GRACE (Gene Replacement and Conditional Expression) collection. The current investigation uncovered both known and novel biological processes required for cell size homeostasis in C. albicans. We also confirmed the plasticity of the size control network as few C. albicans size genes overlapped with those of the budding yeast Saccharomyces cerevisiae. Many new size genes of C. albicans were associated with biological processes that were not previously linked to cell size control and offer an opportunity for future investigation. Additional work is needed to understand if mitochondrial activity is a critical element of the metric that dictates cell size in C. albicans and whether modulation of the onset of actomyosin ring constriction is an additional size checkpoint.


Introduction
Cell size is subject to homeostatic control such that a population of proliferating cells exhibits limited size variation and a reproducible size distribution. At the cellular level, size uniformity is maintained through an elaborate coordination between cellular growth and division (Jorgensen and Tyers 2004;Cook and Tyers 2007). Eukaryotic cells must achieve a critical size at Start, a short interval in late G1, for commitment to cell division (Jorgensen and Tyers 2004;Cook and Tyers 2007). Historically, mechanisms of size control have predominantly been studied in the model yeasts Saccharomyces cerevisiae and Schizosaccharomyces pombe (Wood and Nurse 2015;Xie et al. 2022). In S. cerevisiae, the G1 cyclins Cln1, Cln2, and Cln3 trigger Start via activation of the Cdc28 kinase subunit (Tyers 2004;Jorgensen and Tyers 2004). Former studies indicated that the main target of Cln3-Cdc28 is Whi5, an inhibitor of the SBF transcription factor complex that governs the transcriptional program of the G1-S transition. Cln3-Cdc28 activates SBF by direct 1 3 phosphorylation and inactivation of Whi5 (Costanzo et al. 2004;de Bruin et al. 2004), triggering a transcriptional cascade and ultimately enabling cells to pass Start and engage division. However, a recent investigation has upended this model by showing that Cln3-Cdc28 directly phosphorylating Rpb1, a subunit of the RNA polymerase II, at SBFregulated promoters to promote transcription (Fisher 2021;Kõivomägi et al. 2021). Dilution of Whi5 is also thought to be a mechanism by which growth is connected to the Start machinery (Schmoller et al. 2015(Schmoller et al. , 2022. As cells progress and increase their biomass through G1, Whi5 concentration is diluted which consequently leads to the release of SBF inhibition and cell cycle commitment. However, recent studies have shown that Whi5 concentration is stable throughout G1 phase which question the Whi5 dilution model (Dorsey et al. 2018;Blank et al. 2018;Litsios et al. 2019;Sommer et al. 2021).
Cell size is also modulated by nutrient availability that dictates the biosynthetic capacity in dividing cells (Björklund 2019;Xie et al. 2022). For microorganisms, this allows adaptation to the fluctuating environmental conditions of colonized niches and consequently optimizes cell fitness (Lenski and Travisano 1994). Cell size is also a critical determinant for opportunistic fungi, allowing access to specific niches inside the host or even allowing escape from phagocytic immune cells (Wang and Lin 2012). For instance, the pathogenic dimorphic fungus Histoplasma capsulatum differentiates two categories of propagules with contrasted size called macroconidia and microconidia. The smaller size of infectious microconidia allows them to enter the alveolar space and germinate easily as compared to the larger microconidia (Wang and Lin 2012). Titan cells which are a subset of yeast cells of Cryptococcus neoformans, the causative agent of cryptococcosis, are enlarged to reach up to 20 times the usual size which enables them to evade phagocytosis (Zaragoza and Nielsen 2013;Hommel et al. 2018). Interestingly, the host immune system is able to sense the cell size of the opportunistic yeast Candida albicans, tailoring the immune response to avoid damage to tissues surrounding the site of infection (Branzk et al. 2014). While these observations underscore the intimate link between cell size and fungal fitness and its impact on host-pathogen interactions, the molecular mechanisms underlying size control in fungal pathogens remain only partially explored.
We have recently undertaken a detailed genetic analysis of cell size control in C. albicans, which represented the first regulatory circuits coordinating growth and division in an opportunistic fungus (Chaillot et al. 2017;Sellam et al. 2019). A clear association between size control and fungal virulence were underscored in these works as one third of C. albicans size genes were also essential for the expression of different fungal virulence traits . These studies also uncovered the conservation of core regulators of growth and cell cycle in size homeostasis and revealed other size determinants that were specific to this pathogenic yeast. For instance, an unanticipated finding was the discovery of the role of the stress-activated protein kinase/High Osmolarity Glycerol (HOG)/p38 pathway as a potent regulator of Start specifically in C. albicans . While the HOG pathway has no role in size control in S. cerevisiae, the orthologous HOG pathway in human cells were found to be critical to maintain size uniformity through a mechanism that was equivalent to that of C. albicans (Liu et al. 2018;Sellam et al. 2019). We also uncovered that Ahr1, a zinc finger transcription factor specific to C. albicans and a key modulator of its virulence, acts as a nexus between nitrogen metabolism and cell size control (Chaillot et al. 2022). Overall, these studies highlight the fact that while some size regulators are well conserved, C. albicans has unique size control mechanisms, reconfigured through evolution, to accommodate fungal fitness in the contrasting and dynamic colonized niches of the human host.
Our previous analysis of the size phenome in C. albicans focused mainly on mutants of regulatory proteins including kinases and transcription factors . To uncover novel size regulator in C. albicans and identify additional examples of evolutive innovation of cell size control in pathogenic fungi, we expanded our size genetic survey to include 1301 strains from the GRACE (Gene Replacement and Conditional Expression) collection (Roemer et al. 2003). Our current study uncovered both conserved and novel biological processes that were essential for cell size homeostasis in C. albicans. While, both C. albicans and S. cerevisiae proliferate by budding and share core cell cycle and growth regulations, a limited overlap was observed when comparing the size phenome of both fungi. In particular, our work delineates novel biological processes that were not previously linked to cell size control and offer an opportunity for future investigation.

Cell size determination and data processing
Cell size determination was performed using a Z2-Coulter Counter (Beckman). C. albicans cells were grown overnight in YPD-doxycycline at 30 °C, diluted 1000-fold into fresh YPD-doxycycline and grown for 5 h at 30 °C to reach a final density of 5. 10 6 -10 7 cells/ml, a range in which size distributions of the WT strain do not change. A total of 100 µl of exponentially growing cells was diluted in 10 ml of Isoton II electrolyte solution, sonicated three times for 10 s and used for size quantification in the Z2-Coulter Counter. Size distribution data were normalized to cell counts in each of 256 size bins. Data analysis and size distribution clustering were performed using custom R scripts as previously described ).

Determination of critical cell size
Critical size of mutants was determined using budding index as a function of size. G1 daughter cells were obtained using the JE-5.0 centrifugal elutriation system (Beckman, Fullerton, CA) as described previously (Tyers et al. 1993). C. albicans G1-cells were released in fresh YPD-doxycycline medium and fractions were harvested at an interval of 10 min to monitor bud index.

New cell size phenome in C. albicans
We have previously screened 279 C. albicans mutants for their size distribution and uncovered novel potent regulators that modulate size at Start (Chaillot et al. , 2022Sellam et al. 2019). Here, we expanded our cell size genetic survey to include 1301 non-essential mutants from the GRACE (Gene Replacement and Conditional Expression) collection, which bear a gene deletion at one locus and an integrated tetracycline-regulated allele at the other locus (Roemer et al. 2003). A total of 18 mutants were in common with the set of mutants investigated in our former work , which leave the opportunity to analyze 1283 new mutants for cell size defect (representing ~ 30% of all predicted nonessential genes in C. albicans).
Clustering of size distributions of the GRACE collection discriminated both large and small mutants (which we refer as lge and whi phenotypes, respectively) from the rest of the conditional mutants with comparable size to the parental wild-type (WT) strain CAI4 ( Fig. 1A and Table S1). We defined a significant change in size as a 20% increase or decrease in median size, as compared to WT, as applied in our previous screen ). Accordingly, a total of 40 whi and 11 lge mutants were identified (Table S2). Size phenotypes of two repressible mutants, including the swi6 and ssk1 previously shown as large and small mutants , respectively, were confirmed by our analysis (Fig. 1B).
Gene Ontology (GO) enrichment analysis revealed that small size mutants were predominantly defective in functions related to rRNA processing and ribosome biogenesis (Fig. 1C, F). A similar finding was reported in S. cerevisiae and also in our former surveys of size phenome in C. albicans (Jorgensen et al. 2002;Soifer and Barkai 2014;Chaillot et al. 2017;Sellam et al. 2019). As ribosome biogenesis dictates the rate of protein production and thus underlies the cell's capacity to grow Jorgensen and Tyers 2004), the small size of those mutants could be explained by the fact that they passed Start with a reduced cell mass.
Inactivation of many genes with mitochondrial functions including mitochondrial genome maintenance and organization (IRC3, FIS1, MDM20, MSU1) and, protein import and assembly (MJD1, COX19) led to small size mutants (Fig. 1C, H). Genes whose repression led to whi phenotype were also enriched in cytokinesis, specifically in the formation of the contractile actomyosin ring (MYO1, IQG1, NAT3) (Fig. 1C, E). Disruption of many other genes also reduced cell size in C. albicans, including genes implicated in amino acid metabolism (SSY1, CAR2, ARO80, OPT7), carbon metabolism (ICL1, orf19.1797, RTG1), among many other processes ( Fig. 1G and Table S2). Conversely, mutations in processes related to cell cycle regulation (swi6, lte1), chromatin control (hst1, zds1), and cell wall organization (kre1, chs3) conferred an increase of cell size (Fig. 1D, I, J). Overall, much as in the case of other species, cell size in C. albicans is a complex trait that depends on diverse biological processes.

Limited overlap of the fungal cell size phenome across species
C. albicans and S. cerevisiae represent yeast genera separated by ~ 70 M years of evolution (Salichos and Rokas 2013) and share the morphological trait of budding, as well as core cell cycle and growth regulatory mechanisms (Berman 2006; Cote et al. 2009). However, an extensive degree of rewiring of both cis-transcriptional regulatory circuits 1 3 and signaling pathways was uncovered across many cellular and metabolic processes between the two yeasts (Lavoie et al. 2009;Blankenship et al. 2010;Li and Johnson 2010).
To assess the extent of conservation and plasticity of the size phenome between the two species, genes that affect cell size in C. albicans were compared to their corresponding . Surprisingly, we found minimal overlap between homologous counterparts of size mutants in both species ( Fig. 2A). Only one small size mutant was common between C. albicans and S. cerevisiae, namely zds1, a regulator of both mitotic entry and exit (Rossio and Yoshida 2011). Similarly, only one large size mutant was shared, that is that of the SBF subunit Swi6, a transcriptional regulator of the G1-S transition (Jorgensen and Tyers 2004). Unexpectedly, the C. albicans small sized mutant mdm20 exhibited a large phenotype in S. cerevisiae. To confirm these disparities in the size control network independently of the particular cut-offs used to define large and small mutants in different studies, we examined the relationship between median size of all screened C. albicans GRACE mutants and their counterparts in S. cerevisiae. Overall, no correlation was observed between the size phenomes of the two species (R 2 = 0.0146; Fig. 2B).

Modulation of start by novel and known mechanisms
Previous work has shown that mutations affecting core cellular process such as ribosomal and nucleolar proteins and cell division lead to a dramatically decreased growth rate and corresponding size defect (Tyson et al. 1979;Jorgensen et al. 2002). To identify bona fide Start regulators, rather than slow growth mutants, mutations that significantly impacted growth rate were excluded from our dataset. We calculated the doubling times of the 51 size mutants identified in this work and those that had more than 10% increase in doubling time as compared to the parental WT strains were omitted. A short list of 33 (26 whi and 7 lge mutants) strains was retained as high confidence size mutants of C. albicans (Table 1). Start onset of five potent C. albicans Start regulators Myo1, Iqg1, Chs3, Kre1, and Sun41 in addition to the conserved regulators Swi6, were characterized by assessing bud emergence in G1-elutrated cells of their mutant strains. As shown in Fig. 2C, the critical cell size was significantly reduced in myo1 and iqg1 suggesting an accelerated Start in these whi mutants. Conversely, chs3, kre1, sun41, and swi6 Lge mutants passed Start at larger size as compared to the WT. These data support the notion that the processes related to cell wall biogenesis and actin ring constriction are critical determinants of Start onset and cell size homeostasis.
The larger size of cell wall mutants could be explained by a prolonged G1 phase as a consequence of a reduced amount of newly synthesized cell wall materials required for the formation of the daughter cell and the septal wall. Constriction of the actomyosin ring formed by F-actin and the type II myosin Myo1 is essential for cytokinesis and the completion of the mitotic cell cycle (Cheffings et al. 2016). In the budding yeast, the actomyosin contractile ring is coordinated with and guides the formation of the primary septum, which deposit new cell wall materials between the dividing cells (Juanes and Piatti 2016). As many components of the C. albicans actomyosin ring act as potent size regulators, it is tempting to speculate that cell size is also monitored at the G2/M transition prior to cytokinesis. Alternatively, this suggests that in addition to the Start checkpoint of the mother cell, size could be gauged sequentially at G2/M by ensuring that the daughter cell have reached an optimal size to promote cytokinesis. A similar mechanism of size control at G2/M was proposed in S. cerevisiae whereby Swe1, a protein kinase that coordinates entry into mitosis at the G2/M transition, was shown to activate a cell size checkpoint to monitor cell size of the daughter cell (Harvey and Kellogg 2003;Turner et al. 2012).
Inactivation of genes related to ribosome biogenesis and carbon and nitrogen metabolism resulted in altered cell size (Fig. 1C, G, Table 1), consistent with our previous finding underscoring the role of these processes in the timing of Start onset and critical cell size (Chaillot et al. , 2022Sellam et al. 2019). Perturbation of mitochondrial genes led to a whi phenotype which might be linked to a decrease of energy supply or metabolic capacity of the cells as previously hypothesized in S. cerevisiae (Blank et al. 2008) (Fig. 1C). In metazoans, growth factor signaling such as the insulin/mTOR (mammalian target of rapamycin) pathway is essential for size homeostasis, a process that requires the contribution of mitochondria (Lloyd 2013;Miettinen and Björklund 2016). Future efforts describing the specific mitochondrial process required for size control in C. albicans will lend further insights into how fungal mitochondria signal to the Start machinery. Fig. 1 Cell size phenome acquisition in C. albicans. A Clustergram of size profiles of the GRACE conditional mutant collection. Size distributions of each mutant (rows, represented as a heatmap) were normalized as percentage of total counts, smoothed by averaging over a seven-bin sliding window, and hierarchically clustered using a custom R script . Two parts of the clustergram were magnified showing both whi and lge phenotypes. B Size distribution of the WT strain (CAI4) and the previously characterized whi (ssk1) and lge (swi6) mutants. C, D GO terms enrichment of the whi (C) and lge (D) size mutants identified in this study. p values were calculated using hypergeometric distribution (http:// go. princ eton. edu/ cgi-bin/ GOTer mFind er). E-J Size distributions of different C. albicans size mutants. The indicated mutant strains and the WT control (CAI4) were grown to early log phase in rich YPD-doxycycline for gene repression and sized on a Beckman Coulter Z2 Channelizer. Size distributions were shown for mutants of actomyosin ring (E), ribosome biogenesis (F), amino acid metabolism (G), mitochondrial genes (H), cell wall (I), and bud emergence (J)

Conclusion
The current investigation uncovered both known and novel biological processes that were essential for cell size homeostasis in C. albicans. This work also confirmed the plasticity of the size control network as few C. albicans size genes overlapped with those of the budding yeast S. cerevisiae. Many new size genes of C. albicans are associated with biological processes that were not previously linked to cell size control and offer an opportunity for future investigation. More work is needed to understand if mitochondrial activity is a critical element of cell size control in C. albicans and whether modulation of the onset of actomyosin ring constriction is an additional size checkpoint. Start regulators in C. albicans. Early G1-phase cells of WT (CAI4) and mutant of different potent Start regulators were isolated by centrifugal elutriation, released in a fresh YPD-doxycycline, and monitored for bud emergence