Glazier Et Al 2023 The Candida Albicans Reference Strain sc5314 Contains A Rare Dominant Allele of The Transcription
Glazier Et Al 2023 The Candida Albicans Reference Strain sc5314 Contains A Rare Dominant Allele of The Transcription
Glazier Et Al 2023 The Candida Albicans Reference Strain sc5314 Contains A Rare Dominant Allele of The Transcription
IMPORTANCE Candida albicans is a commensal fungus that colonizes the human oral
Editor James W. Kronstad, The University of British
cavity and gastrointestinal tract but also causes mucosal as well as invasive disease. The Columbia, Vancouver, British Columbia, Canada
expression of virulence traits in C. albicans clinical isolates is heterogeneous and the
Address correspondence to Damian J. Krysan,
genetic basis of this heterogeneity is of high interest. The C. albicans reference strain
[email protected].
SC5314 is highly invasive and expresses robust filamentation and biofilm formation
Virginia E. Glazier, Juraj Kramara and Tomye Ollinger
relative to many other clinical isolates. Here, we show that SC5314 derivatives are
contributed equally to this article. The order of the
heterozygous for the transcription factor Rob1 and contain an allele with a rare gain- first three authors was determined alphabetically by
of-function SNP that drives filamentation, biofilm formation, and virulence in a model last name.
of oropharyngeal candidiasis. These findings explain, in part, the outlier phenotype of The authors declare no conflict of interest.
the reference strain and highlight the role heterozygosity plays in the strain-to-strain
See the funding table on p. 20.
variation of diploid fungal pathogens.
Received 16 June 2023
KEYWORDS Candida albicans, biofilms, filamentation, virulence Accepted 27 July 2023
Published 22 September 2023
C andida albicans is a commensal fungus that is found in the human oral cavity as well Copyright © 2023 Glazier et al. This is an open-access
article distributed under the terms of the Creative
as the gastrointestinal and genitourinary tracts (1). In general, C. albicans causes two
Commons Attribution 4.0 International license.
types of disease in humans. First, mucosal candidiasis of the oral cavity and genitouri
nary tract are common among both immunocompetent and immunocompromised
individuals (2). For example, oral mucocutaneous infections occur in normal newborns
while the same disease occurs later in life in patients with reduced T-cell function such
as those living with HIV/AIDS. In addition, most women will have at least one episode
of vulvovaginal candidiasis in their lifetime. The second type of disease caused by C.
albicans is an invasive disease involving infection of the bloodstream, abdominal organs
such as liver or spleen, and the central nervous system (3). The risk of these life-threat
ening infections is increased in patients with reduced neutrophil number or function,
premature infants, and in patients who have undergone extensive abdominal surgery as
well as other conditions.
The ability of C. albicans to undergo the morphogenetic transition from yeast to
hyphae is important for the pathogenesis of both mucosal and invasive infections (4).
The formation of hyphae has been correlated with the severity of mucosal disease (5) and
with damage to deep organs after dissemination (6). In addition, hyphal formation plays
a key role in the establishment of biofilms (7). C. albicans biofilms contribute directly to
the pathogenesis of mucosal disease and biofilms that form on medical devices such as
intravascular catheters contribute indirectly to the pathogenesis of invasive disease (8).
Accordingly, understanding the transcriptional networks that regulate hyphal morpho
genesis and biofilm formation has been the focus of much research (9, 10).
Nobile et al. initially identified the zinc finger transcription factor Rob1 in a landmark
screen for transcription factors required for in vitro biofilm formation and named the
gene, regulator of biofilms 1 (9). Rob1 is also required for biofilm formation in vivo in
models of intravascular catheter infection and denture infection (9). Recently, we found
that loss of ROB1 function reduces virulence in a model of oropharyngeal candidiasis
and decreases filamentation during infection of mammalian tissue (11, 12). As part of our
interest in C. albicans haploinsufficiency (13), we observed that heterozygous deletion
mutants of ROB1 had distinct filamentation and biofilm phenotypes. This prompted a
more detailed analysis of the function of ROB1 and its two phenotypically distinct alleles.
As described below, we discovered that the reference strain SC5314 and its deriv
atives are heterozygous at the ROB1 locus with one allele showing gain-of-function
properties relative to the other under some but not all conditions. Curiously, we have
RESULTS
Rob1 affects hyphal morphogenesis and biofilm formation in an inducing-
condition- and strain-dependent manner
The effect of rob1∆∆ mutants on filamentation has been reported in the SC5314-derived
SN background in two previous screens of the TF deletion library performed by the
Johnson laboratory (9, 14). Homann et al. found that the rob1∆∆ had reduced filamenta
tion on solid agar YPD, YPD + 10% bovine calf serum (BCS), and Spider medium plates
at 37°C (9). We observed the same reduced filamentation phenotype for the library
rob1∆∆ strain on Spider medium at 37°C; at 30°C, the mutant showed an altered central
wrinkling pattern but peripheral invasion was present (Fig. 1A). Wrinkling in the central
portion of the colony is indicative of pseudohyphal growth (15) while the peripheral
invasion reflects hyphal growth. At 37°C, the colonies were smooth with no peripheral
invasion. On RPMI 1640 medium (referred to as RPMI for the remainder of the paper) and
RPMI + 10% BCS agar plates, the rob1∆∆ mutant showed strong peripheral filamentation
at both temperatures. Within a standard laboratory background, Rob1, therefore, has
temperature- and medium-dependent effects on filamentation.
FIG 1 The effect of Rob1 on C. albicans filamentation is dependent upon strain background and inducing medium. (A) Homozygous ROB1 deletion mutants
in the indicated strain backgrounds were spotted on agar plates prepared with Spider medium, RPMI medium, and RPMI supplemented with 10% bovine calf
serum (BCS). The strains were incubated at 30°C or 37°C for 3–5 days prior to photographing. Representative images from three independent experiments are
shown. (B) Overnight cultures of the indicated strains were diluted (1:50) into liquid RPMI + 10% BCS at 37°C. After 4 hours, the cultures were fixed and the
distribution of yeast, pseudohyphae, and hyphae was determined by light microscopy. An asterisk indicates that the parental and rob1∆∆ mutant had statistically
significant differences in distribution by Student’s t test (P < 0.05).
Next, we asked if the effect of Rob1 on filamentation varied with strain background
(Fig. 1A). To do so, we generated deletion mutants of ROB1 in four C. albicans well-
characterized clinical isolates (P75010, P87, P57055, and P76067) for which the effects of
other filamentationrelated TFs have been studied (16). These four isolates have different
filamentation phenotypes: P75010 and P57055 show almost no filamentation while P87
and P76067 filament on RPMI and RPMI + 10% BCS (Fig. 1A). For the low filamenting
strains, deletion of ROB1 has no observable effect except in the case of P75010 on Spider
medium at 30°C where the deletion mutant colony shows peripheral invasion. On Spider
medium, neither P87 nor P76067 undergoes significant peripheral invasion but P76067
shows central wrinkling at 30°C in both WT and the rob1∆∆ mutant but not at 37°C.
P76067 has minimal peripheral invasion on Spider medium at 37°C and this invasion is
absent in the rob1∆∆ mutant. Deletion of ROB1 in P87 reduces filamentation at 37°C on
both RPMI and RPMI + 10% BCS but not at 30°C. Curiously, P7067 does not filament well
on RPMI + 10% BCS but does so on RPMI. The deletion of ROB1 reduces filamentation of
P76067 on RPMI at both 30°C and 37°C. These data indicate that the role of Rob1 during
filamentation on agar varies with both the specific induction conditions and the C.
albicans strain background.
Huang et al. characterized the effect of other key TFs involved in hyphal morpho
genesis in these same strains using liquid RPMI + 10% BCS conditions (16). Therefore,
we examined the effect of the rob1∆∆ mutant under the same conditions. Consistent
with the results reported by Huang et al. (16), the four clinical isolates as well as the
SC5314-derived SN strain showed distinct patterns of filamentation (Fig. 1B). Under these
conditions, SN250 and P76067 were the only two strains to form more than 20% true
hyphae. P87 and P57055 predominately formed pseudohyphae while >80% of P75010
cells remained as yeast in RPMI + 10% BCS at 37°C for 4 hours (Fig. 1B). Deletion of ROB1
in SN250 essentially abrogated hyphae formation with pseudohyphae (70%) being the
dominant morphotype. In striking contrast, deletion of ROB1 in P76067 had no effect on
hyphae or pseudohyphae formation. Similarly, the rob1∆∆ mutation did not significantly
change the morphological distribution of the pseudohyphae-predominate strains P87 or
P57055 nor the yeast predominate strain P75010. Huang et al. found that the TFs Efg1,
Brg1, and Ume6 had consistent effects across the strains while the role of Bcr1 during
filamentation varied between strains (16). Our results indicate that the effect of Rob1 on
The set of genes regulated by Rob1 during hypha formation varies with
inducing conditions
The effect of Rob1 on gene expression was characterized by Nobile et al. during biofilm
formation in Spider medium using microarrays (9) and by Nanostring during kidney and
ear infection by Xu et al. (18) and Wakade et al. (12), respectively. However, the role of
Rob1 in the regulation of gene expression during in vitro hyphal induction conditions has
not been reported. Using an RNA sequencing approach, we compared gene expression
FIG 2 Rob1 is required for biofilm formation in multiple C. albicans strain backgrounds. The biofilm
formation of the indicated strains was determined using the microtiter plate density assay as described in
Materials and Methods using RPMI medium at 37°C. The bars indicate the OD600 of the biofilm at 90 min
(adhesion), 24 hours, and 48 hours. The bars indicate mean of biological triplicates performed in technical
triplicate. The asterisks indicate that the rob1∆∆ mutant is significantly different from the parental by
Student’s t test.
of rob1∆∆ mutants to WT after 4 hours of hyphal induction using RPMI and Spider
medium at 37°C. A total of 211 genes were differentially expressed (±log2 1; adjusted P
value < 0.05) in the rob1∆∆ mutant in Spider medium and 263 genes were differentially
expressed in RPMI medium relative to SN250 (Fig. 3A and B). Strikingly, only 10 genes
were down regulated in the rob1∆∆ mutant in both media and there were no genes
upregulated in both media (Fig. 3C and D). The set of Rob1-dependent genes
SC5314-derived strains are heterozygous at the ROB1 locus and the two
alleles have distinct filamentation phenotypes
We previously constructed a library of TF heterozygotes for use in genetic interaction
studies (13). As part of this work, we observed that ROB1 heterozygous deletion mutants
showed distinct filamentation phenotypes on Spider medium at 37°C. The phenotypes of
six independent rob1∆/ROB1 heterozygotes constructed in the SN/SC5314 background
are shown in Fig. 4A along with the parental strain and the rob1∆∆ homozygote. Three
heterozygotes show wild type or even increased levels of filamentation after 7 days
on Spider medium at 37°C. The other three heterozygotes show smooth colonies with
little or no peripheral invasion and are closer to the null mutant. As indicated in the
Candida Genome Database, ROB1 in SC5314 is heterozygous with a non-synonymous
C/T polymorphism at position 2902 resulting in the A allele encoding a serine and the
B allele encoding a proline (Fig. 4B). The results of Sanger sequencing of this region of
ROB1 for the six isolates is shown in Fig. 4A and the genotypes are as indicated in Fig. 4B.
The filamentous isolates (rob1-2, rob1-3, rob1-5, rob1-6) contained an S at 946 (ROB1946S/
rob1946P∆) while the isolates with reduced filamentation (rob1-1, rob1-4) contained P
at 946 (rob1946S∆/ROB1946P). For the remainder of our studies, we used strains rob1-1
to characterize the rob1946S∆/ROB1946P genotype and rob1-5 for the ROB1946S/rob1946P∆
genotype.
To determine if the filamentation phenotype of the heterozygous ROB1 mutants
varied with inducing conditions, the isolates rob1-1 and rob1-5 were plated on RPMI and
FIG 4 ROB1 in the SC5314 background is heterozygous and the two alleles have different filamentation
phenotypes. (A) Six independent isolates of heterozygous ROB1 strains derived in the SN background
were plated on solid Spider medium and incubated for 4 days at 37°C. (B) Diagrams of the Rob1 ORF
indicating the position of the SNP and the Sanger sequences of the six heterozygous strains shown in (A).
(C) The distribution of cell morphologies after 4 hours incubation in RPMI + 1% BCS. The asterisk indicates
that the indicated strain differs from SN250 in a statistically significant manner by Student’s t test (P
<0.05). (D) Micrographs of mouse ears inoculated with a 1:1 mixture of SN250 (GFP labeled) and the
indicated rob1 mutant (RFP) and imaged by confocal microscopy 24 hours post infection. Quantitation of
stimuli in mammalian tissue are sufficient to trigger filamentation in the rob1 heterozy
gotes regardless of the allele that is present.
The ROB1946P/rob1∆ strain has reduced biofilm formation in vivo and in vitro
Next, we examined the effect of the two ROB1 alleles on biofilm formation under
three conditions, Spider medium, RPMI + 1% BCS, and RPMI + 10% BCS at 37°C. The
rob1946S∆/ROB1946P strain showed reduced biofilm relative to both wild type and the
ROB1946S/rob1946P∆ strain under all three conditions (Fig. 5A) during adhesion and at
24 hours and 48 hours. The rob1946S∆/ROB1946P strain phenocopied the rob1∆∆ under all
three conditions, indicating it is significantly attenuated in biofilm formation in vitro. We
also examined the structure of the biofilms formed in Spider medium and found that
the biofilm formed by the rob1946S∆/ROB1946P strain was less dense than either WT or
the ROB1946S/rob1946P∆ strain but that there was no difference in the presence of hyphae
within the biofilm structure (Fig. 5B).
Nobile et al. showed previously that the rob1∆∆ mutant has reduced biofilm
formation in a rat intravascular catheter infection model (9). To determine if the
phenotypes observed in vitro were also present in vivo, catheters implanted in jugular
veins of rats were infected with SN250, ROB1946S/rob1946P∆, rob1946S∆/ROB1946P, and rob1∆∆
strains. The catheters were removed 24 hours post infection and the fungal burden
was determined as previously described. Consistent with previous results, catheters
infected with the rob1∆∆ mutant had reduced fungal burden (~1.5 log10 CFU/catheter)
relative to SN250 (Fig. 5C). The fungal burden of the catheters infected with the ROB1946P/
rob1∆ strain was nearly identical to the rob1∆∆ mutant, while catheters infected with
the ROB1946S/rob1∆ strain were comparable to SN250. The infected catheters were also
characterized using scanning electron microscopy. The ROB1946S/rob1∆ strain formed
hyphal structures with a structure similar to the SN250 reference strain (Fig. 5D). The
rob1946S∆/ROB1946P strain, on the other hand, forms biofilm structures that consist mainly
of yeast with some cells that appear to be pseudohyphae. It is interesting to note
that in Spider medium, the rob1946S∆/ROB1946P strain forms less dense biofilms that are
structurally similar to WT (Fig. 5B), while in vivo (Fig. 5D), the biofilm shows a dramatic
reduction in filamentous forms.
The ROB1946S allele appears to be rare and isolated to the SC5314 strain
among 224 sequenced isolates
Next, we were interested to determine the prevalence of the ROB1946P allele. To identify
other strains with this SNP, we compiled genomes from 216 C. albicans strains from the
literature (21–24) and an additional 8 strains collected from premature infants that our
group had sequenced (25). SNP calling was performed as described in Materials and
Methods. Although we do not have uniform sampling from all clades, we have been
unable to identify another C. albicans strain that contains the ROB1946S allele as either a
heterozygote or a homozygote (Table 1). In other words, all sequenced strains that we
have analyzed to date are homozygous for the ROB1946P allele which is the phenotypically
less active allele of ROB1. Our largest set of sequenced genomes comes from Clade 1,
which includes SC5314. Thus, the ROB1946S allele is rare among relatively closely related
strains. In previously reported systematic studies of filamentation and biofilm formation
across a large set of clinical isolates (22, 26), SC5314 is one of the most robust in terms of
these two phenotypes under many conditions. It is therefore tempting to speculate that
the ROB1946S allele may contribute to this feature of the strain.
The position of the non-synonymous SNP is in the C-terminal portion of the protein.
Rob1 is a zinc finger transcription factor and its likely DNA binding domain is predicted
to be in the C-terminal region of the protein (27). Gain-of-function mutations in zinc
finger transcription factors frequently are located in the C-terminus of the protein. For
example, fluconazole resistance is associated with such mutations in the zinc finger
transcription factors Tac1 and Mrr1 (28); these mutations lead to increased expression of
multi-drug efflux pumps such as CDR1 which mediate fluconazole resistance. Our initial
phenotypic data suggest that ROB1946S may represent a gain-of-function allele relative to
ROB1946P which is the predominant allele in sequenced C. albicans isolates.
Although there are a variety of mechanisms by which an allele can display pheno
types of a gain-of-function allele, the simplest and best characterized for a transcription
factor is that the changes in amino acids alter the activity of the factor. As discussed
above, the Rob1 946 position is in the C-terminus which is frequently the activation
domain of zinc finger transcription factors. If that were the case, then we would expect
that one allele would activate the expression of genes regulated by Rob1 more than
the other. We focused our analysis on three canonical hypha-associated genes (ALS3,
ECE1, and HWP1) differentially expressed in the rob1∆∆ mutant. We also examined the
expression of ROB1 to see if the alleles may auto-regulate the gene differently. We
examined the expression of these four genes during hyphal induction by RPMI with 1%
BCS because the mutants have distinct filamentation phenotypes under these conditions
(Fig. 6).
We first compared the heterozygous mutants to WT and the rob1∆∆ mutant under
these conditions. ROB1 expression was reduced by ~2-fold in both heterozygous
mutants relative to WT (Fig. 6). The similar expression of ROB1 in the two heterozygous
strains indicates that phenotypic differences are not due to differential autoregulation
but more likely due to differences in the function of the resulting protein. For ALS3,
ECE1, and HWP1, a consistent pattern of relative gene expression emerged. The three
genes were expressed in the ROB1946S/rob1946P∆ mutant at a higher level compared to the
rob1946S∆/ROB1946P mutant. The expression of the genes in the WT strain was intermediate
between the two heterozygous mutants (HWP1 and ALS3) or was comparable to the
ROB1946S/rob1946P∆ mutant (ECE1). Although the changes in gene expression between the
two alleles are not dramatic (2- to 3-fold), they correlate with the distinct filamentation
phenotypes shown by the strains and support the conclusion that the rare ROB1946S
allele found in SC5314 represents a gain-of-function allele relative to the more prevalent
ROB1946P allele.
FIG 6 The ROB1 alleles have distinct effects on the expression of canonical hyphae-associated genes
during filamentation. The indicated strains were incubated in RPMI + 1% BCS for 4 hours and RNA
isolated as described in the Materials and Methods. The expressions of ALS3, ECE1, HWP1, and ROB1 were
determined by quantitative RT-PCR using the ∆∆CT method. The bars indicate mean of two independent
experiments preformed in triplicate. Asterisks indicate statistically significant differences between the
indicated strain and SN250 by ANOVA followed by Dunnett’s correction for multiple comparisons (P <
0.05).
FIG 7 The ROB1946S allele appears to be a gain-of-function allele during in vitro filamentation and biofilm formation. (A) A
poorly filamenting clinical isolate of C. albicans that is homozygous for the ROB1946P allele was converted to an SC5314-like
ROB1 heterozygote by knock-in of the ROB1946S allele. The resulting strain shows increased filamentation relative to a strain
with knock-in of the exogenous allele on solid Spider medium and RPMI (A) and after induction in liquid RPMI + 1% BCS for
4 hours at 37°C (B). The asterisk indicates that the indicated strain differs from SN250 in a statistically significant manner by
Student’s t test (P < 0.05). (C) SN250 was transformed with knock-in constructs to generate either ROB1 homozygotes or a
heterozygote containing NAT markers in the 3′ untranslated region. The filamentation of these strains were compared to the
unmarked SN250 heterozygote on Spider medium at 37°C (C) and in liquid RPMI + 10% BCS (D). The asterisk indicates that the
indicated strain differs from SN250 in a statistically significant manner by Student’s t test corrected for multiple comparisons (P
< 0.05). (E) The biofilm properties of the SN250 knock in strains were compared in RPMI + 1 % BCS at 37°C at 24 and 48 hours.
Asterisks indicate statistically significant differences between the indicated strain and SN250 by ANOVA followed by Dunnett’s
correction for multiple comparisons (P < 0.05).
with a NAT1 marker in the 5′ region was also constructed. As shown in Fig. 7C, the
homozygous ROB1946S/ROB1946S strain shows a striking increase in filamentation on Spider
medium at a time point when the heterozygous strain shows only the beginning of
central wrinkling and the homozygous ROB1946P/ROB1946P strain shows a smooth colony.
Once again, this increase in filamentation is condition dependent because there is
no difference in the colonies of the different ROB1 strains on solid RPMI medium. In
RPMI + 1 % BCS at 37°C (Fig. 7D), the ROB1946S/ROB1946S strain forms slightly more hyphae
after 4 hours compared to either the ROB1946P/ROB1946P or ROB1946P/ROB1946S strains. The
homozygous ROB1946S/ROB1946S strain in Spider medium (Fig. 7E) shows increased biofilm
density in Spider medium with the ROB1946P/ROB1946S heterozygote intermediate between
the two homozygous strains.
FIG 8 The ROB1946P allele promotes oral colonization while the ROB1946S allele promotes invasive infection. (A and B) Survival curves for CD-1 mice (n = 10/strain)
infected with the indicated strains by tail-vein infection and monitored to moribundity. The rob1∆∆ mutant (A) was the only strain for which a statistically
significant survival time was observed (Kaplan Meier, Mantel log-rank, P <0.05). (C) The oral fungal burden of tongues harvested from cortisone-treated
CD-1 mice (5/strain) infected with the indicated strains 5 days post-infection. The bars are mean with standard deviation indicated by the error bars. The
asterisks indicate statistically significant differences between strains denoted by the horizontal lines as determined by ANOVA with Dunnett’s test of multiple
comparisons. (D) Histological analysis of tongues from infections described for panel C. The fields are representative of multiple fields evaluated.
significantly less able to establish infection or colonization, Rob1 is critical for OPC but
whether that leads to a commensal or invasive infection depends, in part, on the specific
ROB1 allele.
DISCUSSION
Here, we characterized the role of Rob1 in two virulence traits, filamentation and biofilm
formation, and in three in vivo models of infection: disseminated disease, OPC, and
vascular catheter. In vitro, the zinc finger transcription factor Rob1 indicates in C. albicans
pathobiology that is highly dependent upon the culture media in vitro. An excellent
example of this distinction is that the rob1∆∆ mutant forms filamentous colonies on RPMI
medium agar plates that show clear invasion. However, when incubated in the same
medium under liquid conditions, almost no hyphae form, and pseudohyphae predom
inate. Consistent with this environmentally contingent phenotype is our observation
that Rob1 regulates distinct sets of genes during in vitro hyphal morphogenesis in two
commonly used induction media: Spider medium and the tissue culture medium RPMI
(Fig. 3C and D). The overlap in regulated genes is a small fraction of the total number of
genes differentially expressed under the two conditions.
Furthermore, Nobile et al. found that 2,150 genes are differentially expressed in the
rob1∆∆ mutant during biofilm formation which is 10-fold higher than the number of
genes we found to be differentially expressed under hyphae induction (9). Based on
CHiP-ChiP analysis, only 2% of the differentially expressed genes in the rob1∆∆ mutant
are bound by Rob1. Thus, Rob1 appears to have indirect effects on gene transcription of
a large set of genes during biofilm formation and the indirect nature of its function is one
possible explanation for the differences in regulated genes between the two inducing
conditions. Until the transcriptional profiles of more TFs involved in filamentation are
directly compared under different conditions, it is not possible to know if such a striking
change in differentially expressed genes is general phenomenon or specific to Rob1. We
have recently compared the transcriptional profiles of EFG1, ROB1, and BRG1 mutants in
RPMI + 10% BCS and during infection of ear tissue using Nanostring probe set of 185
environmentally responsive genes (12). Although each gene regulated distinct genes
sets in the two conditions, the overlap was much more extensive than observed for the
are heterozygous at the ROB1 locus with a single SNP leading to alleles with distinct
functions. Surprisingly, all other C. albicans strains that we have examined are homozy
gous for the allele (ROB1946P) that has reduced filamentation and biofilm formation in
vitro. These data suggest that the ROB1946S allele is a gain-of-function allele relative to
the predominant ROB1946S allele. Further supporting that conclusion, the introduction
of ROB1946S into strains that poorly filament and are homozygous for the ROB1946S allele
enhances their filamentation. Additionally, converting an SC5314 strain to a homozygote
of ROB1946S also further increases its ability to filament, even in relatively weak inducing
conditions.
The presence of this ROB1 gain of function allele is likely to contribute to the
robust filamentation and biofilm phenotypes observed for SC5314, particularly in vitro.
However, it is important not to overestimate the generality of these effects. First, there
are other clinical isolates of C. albicans that have filamentation and biofilm phenotypes
that are similar to SC5314 in vitro but are homozygous for the apparently less active
ROB1946P allele (22). Second, the conversion of a poor filamenting clinical strain to the
SC5314 heterozygous genotype at the ROB1 locus improved its filamentation but it
remained far less robust than SC5314. This observation is consistent with a similar
experiment reported by Hirakawa et al. (22) in which an EFG1 loss of function allele
was replaced by a functional allele in a poorly filamenting clinical isolate. In that case,
filamentation improved but the levels remained well below that of SC5314. Conse
quently, it is likely that the phenotypic heterogeneity of these clinical isolates of C.
albicans is due to genetic/genomic heterogeneity at multiple loci. Third, both alleles of
ROB1 support wild-type levels of filamentation in mouse tissue and are indistinguishable
in terms of virulence in this model. This indicates that the advantages or distinctions of
the ROB1946S are dependent on environmental niche.
The distinctions between the functions of the two alleles in vivo are most apparent
in settings that have features of the biofilm state. A consideration of these distinctions
provides a possible model for the apparent low prevalence of the gain of function
ROB1946S allele. The increased filamentation potential of the ROB1946S and the correspond
ing increased expression of inanimate surface biofilm promoting genes such as HWP1
and ALS3 are required for SC5314 to form a robust biofilm in a vascular catheter.
However, the advantages of this allele disappear once C. albicans enters the blood stream
Strain construction
Rob1 deletion mutants. Both alleles of ROB1 were deleted in clinical isolates using
the transient CRISPR-Cas9 system (1). Briefly, cells were transformed with Cas9 and
single ROB1 guide RNA cassettes and the rob1Δ::NAT1 repair template. The NAT1 repair
template was amplified from the pCJN542 plasmid (generous gift from Dr. Aaron Mitchell
[2]) using primers with ROB1-gene-derived flanks (OL279, OL280, Table S4). The ROB1
guide RNA was amplified by a split-join PCR using the OL1, OL2, 18.005, 18.008, 18.009,
and 18.010 oligos. Transformants were selected on YPD media supplemented with
200 mg/mL nourseothricin. The correct integration of rob1Δ::NAT1 repair template and
the absence of the ROB1 ORF was checked by PCR using oligos derived from the NAT1
gene, ROB1 ORF, and ROB1 5′ and 3′ UTR (18.026, OL422–OL425, Table S4).
ROB1 allele swap. NAT1 knock-in cassette carrying the respective SNP was made
by three round PCR approach with first round primers (OL759, OL760, supplemental
material) derived from 3′ end of the ROB1 gene proximal to the 2,902C/T SNP (corre
sponding to the 946S/P change in Rob1p) and ROB1 5′ UTR region. The first round
product was then stitched to a NAT1 cassette flanked with overlapping sequence in a
second round PCR. The NAT1 cassette with ROB1 flanking sequence was amplified from
pCJN542 using the OL757 and OL758 (Table S4). The stitched product was then amplified
in a third round PCR using the flanking primers and used for transformation. Correct
knock-ins were confirmed by PCR using NAT1- and ROB1-derived primers.
(14–16) was used to perform data normalization and differential expression analysis with
an adjusted P value threshold of 0.05 on each set of raw expression measures.
were homogenized and plated for quantitative fungal burden (n = 5 per strain) or
processed for histology. The log10-transformed fungal burden data for each experiment
was analyzed by ANOVA with corrections for multiple comparisons to identify statistically
significant differences between individual strains (adjusted P < 0.05).
ACKNOWLEDGMENTS
The authors thank Aaron Mitchell (Georgia) for assistance with biofilm imaging and
review of drafts of the manuscript.
This work was supported by the following grants from the NIH: F32AI26634 (V.E.G.);
R01AI073289 (D.R.A.); R01AI141893 and R01AI081704 (R.J.B.); R01DE026600 (S.G.F.);
R01AI133409 (D.J.K.).
AUTHOR AFFILIATIONS
1
Department of Biology, Niagara University, Niagara Falls, New York, USA
2
Department of Pediatrics, Carver College of Medicine, University of Iowa, Iowa City, Iowa,
USA
3
Division of Infectious Diseases, Los Angeles Biomedical Research Institute and Harbor-
UCLA Medical Center, Torrance, California, USA
4
Department of Medicine, David Geffen School of Medicine at UCLA, Los Angeles,
California, USA
5
Department of Medicine, Section of Infectious Disease, University of Wisconsin,
Madison, Wisconsin, USA
6
Department of Medical Microbiology and Immunology, University of Wisconsin,
Madison, Wisconsin, USA
7
Department of Microbiology, University of Georgia, Athens, Georgia, USA
8
Department of Molecular Microbiology and Immunology, Brown University, Providence,
Rhode Island, USA
9
Department of Molecular Physiology and Biophysics, Carver College of Medicine,
University of Iowa, Iowa City, Iowa, USA
10
Department of Microbiology and Immunology, Carver College of Medicine, University
of Iowa, Iowa City, Iowa, USA
FUNDING
AUTHOR CONTRIBUTIONS
DATA AVAILABILITY
All sequencing data have been deposited at the NCBI GEO site with the accession
number GSE238051.
ADDITIONAL FILES
Supplemental Material
REFERENCES
1. Lopes JP, Lionakis MS. 2022. Pathogenesis and virulence of Candida 7. Blankenship JR, Mitchell AP. 2006. How to build a biofilm: a fungal
Albicans. Virulence 13:89–121. https://doi.org/10.1080/21505594.2021. perspective. Curr Opin Microbiol 9:588–594. https://doi.org/10.1016/j.
2019950 mib.2006.10.003
2. Vila T, Sultan AS, Montelongo-Jauregui D, Jabra-Rizk MA. 2020. Oral 8. Ramage G, Borghi E, Rodrigues CF, Kean R, Williams C, Lopez-Ribot J.
candidiasis: a disease of opportunity. J Fungi (Basel) 6:15. https://doi.org/ 2023. Our current clinical understanding of candida biofilms: where are
10.3390/jof6010015 we two decades on? APMIS. https://doi.org/10.1111/apm.13310
3. Pappas PG, Lionakis MS, Arendrup MC, Ostrosky-Zeichner L, Kullberg BJ. 9. Nobile CJ, Fox EP, Nett JE, Sorrells TR, Mitrovich QM, Hernday AD, Tuch
2018. Invasive candidiasis. Nat Rev Dis Primers 4:18026. https://doi.org/ BB, Andes DR, Johnson AD. 2012. A recently evolved transcriptional
10.1038/nrdp.2018.26 network controls biofilm development in Candida albicans. Cell 148:126–
4. Sudbery PE. 2011. Growth of Candida albicans hyphae. Nat Rev Microbiol 138. https://doi.org/10.1016/j.cell.2011.10.048
9:737–748. https://doi.org/10.1038/nrmicro2636 10. Do E, Cravener MV, Huang MY, May G, McManus CJ, Mitchell AP, Bahn Y-
5. Lemberg C, Martinez de San Vicente K, Fróis-Martins R, Altmeier S, Tran S. 2022. Collaboration between antagonistic cell type regulators governs
VDT, Mertens S, Amorim-Vaz S, Rai LS, d’Enfert C, Pagni M, Sanglard D, variation in the Candida albicans biofilm and hyphal gene expression
LeibundGut-Landmann S. 2022. Candida albicans commensalism in the network. mBio 13:e0193722. https://doi.org/10.1128/mbio.01937-22
oral mucosa is favoured by limited virulence and metabolic adaptation. 11. Solis NV, Wakade RS, Glazier VE, Ollinger TL, Wellington M, Mitchell AP,
PLoS Pathog 18:e1010012. https://doi.org/10.1371/journal.ppat.1010012 Filler SG, Krysan DJ. 2022. Systematic genetic interaction analysis
6. Saville SP, Lazzell AL, Monteagudo C, Lopez-Ribot JL. 2003. Engineered identifies a transcription factor circuit required for oropharyngeal
control of cell morphology in vivo reveals distinct roles for yeast and candidiasis. mBio 13:e0344721. https://doi.org/10.1128/mbio.03447-21
filamentous forms of Candida albicans during infection. Eukaryot Cell 12. Wakade RS, Ristow LC, Wellington M, Krysan DJ. 2023. Intravital imaging-
2:1053–1060. https://doi.org/10.1128/EC.2.5.1053-1060.2003 based genetic screen reveals the transcriptional network governing
Candida albicans filamentation during mammalian infection. Elife
12:e85114. https://doi.org/10.7554/eLife.85114
13. Glazier VE, Murante T, Koselny K, Murante D, Esqueda M, Wall GA, RC, Poulain J, Battail C, Wincker P, Borman AM, Chowdhary A, Fan S, Kim
Wellington M, Hung C-Y, Kumar A, Krysan DJ. 2018. Systematic complex SH, Le Pape P, Romeo O, Shin JH, Gabaldon T, Sherlock G, Bougnoux M-E,
haploinsufficiencybased genetic analysis of Candida albicans d’Enfert C. 2018. Gene flow contributes to diversification of the major
transcription factors: tools and applications to virulence-associated fungal pathogen Candida albicans. Nat Commun 9:2253. https://doi.org/
phenotypes. G3 (Bethesda) 8:1299–1314. https://doi.org/10.1534/g3. 10.1038/s41467-018-04787-4
117.300515 25. Bliss JM, Basavegowda KP, Watson WJ, Sheikh AU, Ryan RM. 2008. Vertical
14. Homann OR, Dea J, Noble SM, Johnson AD. 2009. A phenotypic profile of and horizontal transmission of Candida albicans in very low birth weight
the Candida albicans regulatory network. PLoS Genet 5:e1000783. https: infants using DNA fingerprinting techniques. Pediatr Infect Dis J 27:231–
//doi.org/10.1371/journal.pgen.1000783 235. https://doi.org/10.1097/INF.0b013e31815bb69d
15. Noble SM, French S, Kohn LA, Chen V, Johnson AD. 2010. Systematic 26. Anderson FM, Visser ND, Amses KR, Hodgins-Davis A, Weber AM,
screens of a Candida albicans homozygous deletion library decouple Metzner KM, McFadden MJ, Mills RE, O’Meara MJ, James TY, O’Meara TR.
morphogenetic switching and pathogenicity. Nat Genet 42:590–598. 2023. Candida albicans selection for human commensalism results in
https://doi.org/10.1038/ng.605 substantial within-host diversity without decreasing fitness for invasive
16. Huang MY, Woolford CA, May G, McManus CJ, Mitchell AP. 2019. Circuit disease. PLoS Biol 21:e3001822. https://doi.org/10.1371/journal.pbio.
diversification in a biofilm regulatory network. PLoS Pathog 3001822
15:e1007787. https://doi.org/10.1371/journal.ppat.1007787 27. MacPherson S, Larochelle M, Turcotte B. 2006. A fungal family of
17. Glazier VE, Murante T, Murante D, Koselny K, Liu Y, Kim D, Koo H, Krysan transcriptional regulators: the zinc cluster proteins. Microbiol Mol Biol
DJ. 2017. Genetic analysis of the Candida albicans biofilm transcription Rev 70:583–604. https://doi.org/10.1128/MMBR.00015-06
factor network using simple and complex haploinsufficiency. PLoS 28. Sasse C, Dunkel N, Schäfer T, Schneider S, Dierolf F, Ohlsen K, Morsch
Genet 13:e1006948. https://doi.org/10.1371/journal.pgen.1006948 häuser J. 2012. The stepwise acquisition of fluconazole resistance
18. Xu W, Solis NV, Ehrlich RL, Woolford CA, Filler SG, Mitchell AP. 2015. mutations causes a gradual loss of fitness in Candida albicans. Mol
Activation and alliance of regulatory pathways in C. albicans during Microbiol 86:539–556. https://doi.org/10.1111/j.1365-2958.2012.08210.x
mammalian infection. PLoS Biol 13:e1002076. https://doi.org/10.1371/ 29. Witchley JN, Penumetcha P, Abon NV, Woolford CA, Mitchell AP, Noble
journal.pbio.1002076 SM. 2019. Candida albicans morphogenesis programs control the
19. Cravener MV, Do E, May G, Zarnowski R, Andes DR, McManus CJ, Mitchell balance between gut commensalism and invasive infection. Cell Host
AP. 2023. Reinforcement amid genetic diversity in the Candida albicans Microbe 25:432–443. https://doi.org/10.1016/j.chom.2019.02.008
biofilm regulatory network. PLoS Pathog 19:e1011109. https://doi.org/ 30. Gillum AM, Tsay EY, Kirsch DR. 1984. Isolation of the Candida albicans
10.1371/journal.ppat.1011109 gene for orotidine-5’-phosphate decarboxylase by complementation of
20. Wakade RS, Krysan DJ, Wellington M. 2022. Use of in vivo imaging to S. cerevisiae ura3 and E. coli pyrF mutations. Mol Gen Genet 198:179–182.
screen for morphogenesis phenotypes in Candida albicans mutant https://doi.org/10.1007/BF00328721
strains during active infection in a mammalian host. J Vis Exp, no. 188. 31. Ene IV, Bennett RJ, Anderson MZ. 2019. Mechanisms of genome
https://doi.org/10.3791/64258 evolution in Candida albicans. Curr Opin Microbiol 52:47–54. https://doi.
21. Ford CB, Funt JM, Abbey D, Issi L, Guiducci C, Martinez DA, Delorey T, Li org/10.1016/j.mib.2019.05.001
BY, White TC, Cuomo C, Rao RP, Berman J, Thompson DA, Regev A. 2015. 32. Liang S-H, Bennett RJ. 2019. The impact of gene dosage and hetero
The evolution of drug resistance in clinical isolates of Candida albicans. zygosity on the diploid pathobiont Candida albicans. J Fungi (Basel) 6:10.
Elife 4:e00662. https://doi.org/10.7554/eLife.00662 https://doi.org/10.3390/jof6010010
22. Hirakawa MP, Martinez DA, Sakthikumar S, Anderson MZ, Berlin A, Gujja 33. Sharma A, Solis NV, Huang MY, Lanni F, Filler SG, Mitchell AP, Bahn Y-S.
S, Zeng Q, Zisson E, Wang JM, Greenberg JM, Berman J, Bennett RJ, 2023. Hgc1 independence of biofilm hyphae in Candida albicans. mBio
Cuomo CA. 2015. Genetic and phenotypic intra-species variation in 14:e0349822. https://doi.org/10.1128/mbio.03498-22
Candida albicans. Genome Res 25:413–425. https://doi.org/10.1101/gr. 34. McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A,
174623.114 Garimella K, Altshuler D, Gabriel S, Daly M, DePristo MA. 2010. The