Outbreaks of fluconazole-resistant Candida parapsilosis are driven by low-biofilm-producing isolates that emerge under host selection

This is an uncorrected proof. Figures Abstract Candida parapsilosis is a major human fungal pathogen, with recent global outbreaks driven by fluconazole-resistant (FLCR-Cp) isolates that are difficult to eradicate and associated with poor clinical outcomes. However, the microbial traits enabling persistence of these outbreak lineages remain poorly defined. Here, we show that FLCR-Cp isolates responsible for prolonged, multi-country outbreaks consistently exhibit a striking low-biofilm-producing (LBP) phenotype. Contrary to the prevailing view that robust biofilm formation promotes persistence, LBP strains displayed enhanced stress tolerance, increased cell wall masking, and reduced immune recognition. These traits conferred resistance to neutrophil and macrophage killing and enhanced survival in immune cell-rich organs during systemic infection. Genome-wide transcriptomic profiling revealed extensive metabolic and regulatory rewiring in LBP strains. Whole-genome sequencing (WGS) of a global isolate collection further demonstrated that the LBP phenotype has emerged independently multiple times, supporting convergent evolution under host selection. Functional genomic analyses suggest that biofilm attenuation arises through multigenic changes, and disruption of key biofilm-associated transcriptional regulators enhanced fitness during immune interactions. Together, our findings overturn the assumption that robust biofilm formation drives outbreak persistence and instead identify biofilm attenuation as an adaptive tradeoff that promotes immune evasion and long-term survival. These results redefine our understanding of C. parapsilosis adaptation during healthcare-associated outbreaks and shift attention toward host-driven evolutionary processes than environmental persistence alone. Citation: Daneshnia F, Gunasekaran D, Bergin S, Lombardi L, Perry AM, Cai L, et al. (2026) Outbreaks of fluconazole-resistant Candida parapsilosis are driven by low-biofilm-producing isolates that emerge under host selection. PLoS Biol 24(9): e3003973. https://doi.org/10.1371/journal.pbio.3003973 Academic Editor: Jessica Brown, University of Utah, UNITED STATES OF AMERICA Received: March 20, 2026; Accepted: August 12, 2026; Published: September 1, 2026 This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication. Data Availability: All RNAseq files are available from the whole-genome sequencing database (accession number(s) GSE312575 and PRJNA1363747). Funding: This work was supported by the National Institutes of Health (NIH) National Institute of Allergy and Infectious Diseases (NIAID) R01AI132638 to M.K.M. and the NIH National Institute of General Medical Sciences (NIGMS) award R35GM156045 to C.J.N. The T.G. group acknowledges support from the Spanish Ministry of Science and Innovation (grant numbers PID2021-126067NB-I00, CPP2021-008552, PCI2022-135066-2, PLEC2023-010225, and PDC2022-133266-I00), cofounded by ERDF “A way of making Europe”, as well as support from the Catalan Research Agency (AGAUR) (grant number SGR01551); Gordon and Betty Moore Foundation (grant number GBMF9742); “La Caixa” foundation (grant number LCF/PR/HR21/00737), Fundació La Marató de TV3 (202328-31), AECC (PRYGN234923GABA), and Instituto de Salud Carlos III (CIBERINFEC CB21/13/00061- ISCIII-SGEFI/ERDF). A.C. was supported by the Fundação para a Ciência e a Tecnologia (FCT) (UID/06304/2023, LA/P/0050/2020, 2022.06674.PTDC, and PMTargets/0003/2024), the “la Caixa” Foundation and FCT under the agreement LCF/PR/HR24/52440014, the “la Caixa” Foundation under the agreement LCF/PR/HR22/52420003, and the EU Horizon 2020 research and innovation programme under grant agreement no 847507. S.M.G. was supported by FCT (2023.07620.CEECIND and 2023.12771.PEX). S.M.G. received personal support from FCT (2023.07620.CEECIND). C.M. acknowledges support from the European Union’s Horizon 2020 research and innovation program under grant agreement no. 847507 (HDM-FUN). C.J.N. acknowledges support from the Kamangar family in the form of an endowed chair. AA was supported by WueConnect Program. The funders had no role in the study design, data collection and interpretation, or the decision to submit the work for publication. Competing interests: We have reviewed the journal’s policy and declare the following competing interests: (CJN) is a co-founder and acting CEO of BioSynesis, Inc., a company developing diagnostics and therapeutics for biofilm-associated infections. (DSP) has received NIH grant support (U19AI189183) for research supporting new product development by Scynexis, and NIH grant support (R01AI109025) for antifungal drug resistance research with Prokaryotics. DSP has also received research support from Mundi Pharmaceuticals. All other authors declare no competing interests. Abbreviations: CFUs, colony-forming units; FLCR, fluconazole-resistant; FLCS, fluconazole-susceptible; GO, Gene Ontology; GSEA, gene set enrichment analysis; iNTS, invasive nontyphoidal Salmonella; LBP, low-biofilm-p; LDH, lactate dehydrogenase; MDR, multidrug-resistant; PAMP, pathogen-associated molecular pattern; PCA, principal component analysis; PRRs, pattern recognition receptors; ROS, reactive oxygen species; VCF, variant call format; WGS, whole-genome sequencing Introduction Invasive fungal infections impose a major global health and economic burden, costing billions of U.S. dollars annually [1] and causing more than 3.5 million deaths each year, with nearly 1 million deaths attributed to yeast species of the Candida genus [2]. Candida parapsilosis ranks among the top three causes of candidemia worldwide, with prevalence varying by patient population and geographic region [3,4]. The C. parapsilosis species complex comprises three genetically distinct but phenotypically similar species: C. parapsilosis sensu stricto, Candida orthopsilosis, and Candida metapsilosis [5]. Among these, C. parapsilosis sensu stricto is the most virulent and is primarily responsible for healthcare-associated outbreaks and device colonization [6]. Recent epidemiological reports from Europe, North America, the Mediterranean, and Southeast Asia have documented an alarming rise in fluconazole-resistant (FLCR) C. parapsilosis isolates, driving persistent hospital outbreaks [3,4,7,8]. Infections caused by FLCR isolates exhibit more than twice the mortality of infections with susceptible strains [9,10], and multidrug-resistant (MDR) lineages have emerged from these FLCR backgrounds [11]. Notably, FLCR isolates frequently infect antifungal-naïve patients [4] and persist in healthcare environments despite rigorous infection control measures [10]. Molecular epidemiology studies have demonstrated that isolates recovered from patient blood, hospital surfaces, and healthcare workers’ hands are often genetically indistinguishable [10], implicating both environmental and host-associated reservoirs in sustaining transmission. Yet, the microbial traits that enable these resistant lineages to persist remain poorly defined. Biofilm formation—the capacity of microbial cells to assemble structured, extracellular matrix-encased communities—has long been considered a central persistence factor in pathogenic Candida species, including Candida albicans and C. parapsilosis [3,4,12]. Fungal biofilms contribute to an estimated $43 billion in annual healthcare costs [13] due to their tolerance to antifungals, disinfectants, and immune defenses. Biofilm-derived cells can exhibit enhanced virulence [14] and facilitate colonization of new niches [12,13]. Consistent with observations in bacterial pathogens, biofilm formation in Candida species is positively associated with clinical persistence, therapeutic failure, and mortality [15,16]. Consequently, robust biofilm production has been widely assumed to underlie the success of outbreak lineages. Under this prevailing model, the persistence of FLCR C. parapsilosis—where identical genotypes circulate between patients and hospital environments [4,17]—is typically attributed to biofilm-mediated resilience. Therefore, repeated exposure to host tissues, immune pressures, and environmental stressors would be expected to select for increasingly robust biofilm formation. Indeed, both clinical [18] and experimental evolution studies [19] have demonstrated that fungal pathogens can rapidly adapt virulence-associated traits to enhance fitness within complex host niches. However, the microbiological drivers and genetic mechanisms sustaining persistent C. parapsilosis outbreaks remain unresolved. Here, we integrate phenotypic, transcriptomic, and genomic analyses to uncover an unexpected adaptive strategy in FLCR C. parapsilosis. Across outbreak isolates from multiple countries, we identify a paradoxical low-biofilm-producing (LBP) phenotype. Strikingly, biofilm attenuation correlates with enhanced host adaptation: LBP strains display increased immune evasion, resistance to macrophage and neutrophil killing, and improved survival in immune cell-rich organs during systemic infection. Transcriptomic profiling reveals extensive metabolic and regulatory rewiring in LBP strains, while whole-genome sequencing uncovers complex, multigenic alterations associated with biofilm attenuation. Disruption of key biofilm-regulatory transcription factors further enhances fitness during host immune interactions. Population genomic analyses demonstrate that the LBP phenotype has arisen independently across diverse outbreak lineages, consistent with convergent evolution under host immune selection. Collectively, our findings challenge the long-standing paradigm that biofilm hyperproduction drives fungal outbreak persistence. Instead, we propose that C. parapsilosis undergoes an evolutionary tradeoff in which biofilm attenuation enhances immune evasion and promotes long-term persistence within healthcare-associated populations. These results redefine the ecological and evolutionary forces shaping fungal outbreak dynamics and highlight host-driven adaptation as a critical determinant of epidemic success. Results Fluconazole-resistant outbreak lineages exhibit attenuated biofilm formation To define traits associated with outbreak persistence, we examined the longitudinal dynamics of bloodstream isolates of C. parapsilosis collected at Ege University Hospital (Turkey) between 2007 and 2024 [9,20,21]. During 2007–2010, all isolates were fluconazole-susceptible (FLCS). Over the subsequent decade, however, fluconazole-resistant (FLCR) isolates carrying the Erg11Y132F substitution (YCP) progressively displaced the susceptible population and dominated by 2019–2022 (Fig 1A). The rapid and sustained expansion of this lineage suggested intrinsic fitness advantages beyond drug resistance. The prevalence of C. parapsilosis isolates collected from the bloodstream of patients from 2009-2024 identifies that fluconazole-resistant isolates carrying Erg11Y132F progressively increase and outcompete the susceptible wild-type isolates (A). Crystal violet biofilm analysis (OD490) revealed that Turkish C. parapsilosis isolates carrying Erg11Y132F were LBP (upper), whereas the majority of the Brazilian (lower) C. parapsilosis isolates carrying Erg11Y132F were BPs (B). Since Brazilian YCP and WT isolates were a mixture of BP and LBPs, replicates belonging to the same isolate were color-coded to display the inter-replicate variability. LBPs from both countries showed an attenuated adherence during the early stage of biofilm formation (C). Each dot represents an independent biological replicate. Genes pertinent for adhesion in C. albicans are downregulated in C. parapsilosis LBPs. Log2 fold change of gene expression in biofilm condition was obtained for LBPs, i.e., 49B and 29T, relative to BPs from their corresponding countries of origin (35B and 1T, respectively) (D). Mitochondrial electron transport, cell redox homeostasis and aerobic respiration were exclusively upregulated in both LBP isolates at 24 h of biofilm formation. Genes belonging to the highlighted mitochondrial electron transport functions are downregulated in the biofilm condition in BPs and upregulated in LBPs. Gene names, when indicated, denote putative S. cerevisiae orthologs (E). Log2 fold change of gene expression in biofilm and planktonic conditions were obtained for BPs (combining 35B and 1T) relative to LBPs (combining 49B and 29T) at each time point and gene set enrichment analysis was performed to obtain enriched biological processes (E). Gene age was estimated and phylogenetic tree was constructed using orthologs across 15 species. “Old” genes are those conserved across Ascomycota species, “Middle” genes are conserved across CTG and non-CTG Candida species and S. cerevisiae, “Young” gene are conserved across CTG Candida species and “Unique” genes are those unique to C. parapsilosis. Genes upregulated at 24 h of biofilm formation in BPs (i.e., downregulated in LBPs) are enriched for “Young” and “Unique” genes, whereas genes upregulated at the same time point in LBPs are enriched for “Old” genes. A one-tailed hypergeometric test was conducted to statistically estimate the over-representation of each of these gene ages in differentially expressed genes and p-value was adjusted for multiple testing using Benjamini–Hochberg correction (F). BP, biofilm producers; LBP, low biofilm producers. *, **, ***, and **** represents P values ≤0.05, ≤0.01, ≤0.001, ≤0.0001, respectively. Data underlying this figure could be found in the S1 Data file. Because similar Erg11Y132F (YCP) outbreaks have been reported in Brazil, we assembled a collection of 120 isolates, comprising 95 from Turkey (historic outbreak) and 25 from Brazil (early-stage outbreak) (S1 Table). Given the long-standing assumption that biofilm formation promotes persistence in Candida species and that C. parapsilosis is shown to comparably produces biofilm relative to C. albicans [22], we quantified biofilm biomass using a validated high-throughput crystal violet assay. Of note, the crystal violet assay strongly and significantly correlated with other standard optical density measurements of biofilm quantification and with biofilm thickness measurements obtained by confocal scanning laser microscopy (S1A–S1C Fig), validating the approach. Based on this calibration, isolates with OD490 < 0.05 from the crystal violet assay were classified as low biofilm producers (LBP strains), whereas isolates with OD490 ≥ 0.05 were categorized as biofilm producers (BP strains) since those with OD490 < 0.05 remained transparent and did not show any crystal violet retention. Unexpectedly, all Turkish FLCR Erg11Y132F isolates exhibited markedly reduced biofilm formation compared to both Brazilian FLCR isolates and FLCS isolates (Fig 1B). Whereas the majority of Brazilian isolates (“collected in early-stage outbreak”) retained robust biofilm capacity, Turkish outbreak isolates uniformly displayed a low-biofilm-producing (LBP) phenotype. Additionally, while Turkish LBP isolates were all FLCR, the Brazilian counterparts encompassed a combination of FLCR and FLCS isolates (microbiological details are listed in S1 Table). Notably, impaired biofilm formation has also been reported in independent FLCR outbreaks in Brazil [23], Italy [24], and Spain [25], suggesting that biofilm attenuation is a recurrent feature of persistent lineages. Consistent with impaired biofilm formation, LBP isolates exhibited significantly reduced surface adherence (Fig 1C). Pseudohyphal development, even under conditions known to favor pseudohyphal morphology in C. parapsilosis [26], was infrequent across all isolates and did not distinguish biofilm phenotypes. Together, these findings challenge the prevailing view that enhanced biofilm formation underlies outbreak success and instead identify biofilm attenuation as a defining characteristic of dominant FLCR lineages. LBP strains exhibit metabolic rewiring and preferential activation of evolutionarily conserved transcriptional programs during biofilm development The pronounced defect in biofilm formation among LBP isolates raised a fundamental question: how is their transcriptional program reconfigured during biofilm growth? To address this, we performed comparative RNA-seq to define the regulatory and evolutionary architecture underlying biofilm attenuation in outbreak-associated strains. We selected one BP and one LBP isolate from each country: 29T and 49B (LBP, YCP lineage), and 1T and 35B (BP, FLCS lineage), where “T” and “B” denote Turkish and Brazilian origin, respectively. Because biofilm development proceeds through defined stages—early adhesion (90 min), initiation (8 h), and maturation (24 h)—transcriptomes were profiled at each time point under both biofilm and planktonic conditions. Principal component analysis (PCA) revealed tight clustering of biological replicates, confirming high reproducibility (S1D Fig). Biofilm and planktonic transcriptomes diverged progressively over time, consistent with stage-specific differentiation (Figs 1D and S1; S2A and S2B Table). Given the pronounced adhesion defect in LBP strains, we first examined expression of established adhesin genes implicated in biofilm initiation. Consistent with the phenotype, most adhesins were downregulated in LBP isolates relative to BP isolates across all timepoints (Fig 1D). To define broader transcriptional trends, we normalized biofilm transcriptomes to their corresponding planktonic states. Strikingly, mitochondrial electron transport, aerobic respiration, and redox homeostasis were the only biological processes consistently upregulated in both LBP isolates at 24 h (S1E Fig). Reciprocal normalization of BP to LBP transcriptomes identified mitochondrial-associated processes as among the most significantly downregulated pathways across nearly all stages (Fig 1E). These data indicate that LBP strains activate a coordinated mitochondrial and oxidative metabolic program during biofilm growth, a pattern reminiscent of biofilm-defective transcription factor mutants in C. parapsilosis (ace2Δ, bcr1Δ, cph2Δ, czf1Δ, gzf3Δ, efg1Δ, and ume6Δ [27]). Because biofilm regulatory networks in C. albicans are enriched for evolutionarily “young” genes unique to the CTG clade [28], we next asked whether evolutionary age distinguished transcriptional programs in BP and LBP isolates. Using comparative genomics across 15 yeast species (12 Candida and 3 non-Candida ascomycetes), we categorized genes into four age groups: “old” (conserved across all species), “middle” (shared among Candida species and S. cerevisiae), “young” (restricted to CTG Candida species), and “unique” (C. parapsilosis-specific). Genes selectively upregulated in BP strains at 24 h were significantly enriched for “young” and “unique” genes, paralleling the lineage-specific innovation described for C. albicans biofilm networks [28]. In contrast, genes upregulated at 24 h in LBP strains were predominantly “old,” evolutionarily conserved genes associated with core metabolic functions (Fig 1F and S2C Table). Thus, whereas canonical biofilm formation relies on recently evolved, clade-restricted regulatory modules, the attenuated biofilm program of LBP strains preferentially engages ancestral metabolic circuitry. Collectively, these findings indicate that outbreak-associated LBP isolates do not simply down-regulate biofilm genes; rather, they undergo a coordinated transcriptional reprogramming that shifts biofilm development away from lineage-specific adhesion networks and toward conserved metabolic processes. Whole-genome sequencing (WGS) reveals polygenic and convergent genetic architectures underlying biofilm attenuation Biofilm impairment emerged as a defining feature of Turkish YCP isolates and a subset of Brazilian isolates. Because all Turkish YCP strains displayed the LBP phenotype, we initially hypothesized that acquisition of the Erg11Y132F substitution directly impaired biofilm formation. However, CRISPR-Cas9-mediated reversion of Erg11Y132F to WT Erg11 (F132Y) in two Turkish LBPs (59T and 60T) did not increase the biofilm capacity, whereas the revertant isolates showed 16- to 8-fold reduction in fluconazole minimum inhibitory concentration (S1F Fig), indicating that fluconazole susceptibility patterns and biofilm formation are decoupled and additional genetic determinants underlie the LBP phenotype. To define these determinants and assess the evolutionary relationships among isolates, we performed high-coverage (>100×) paired-end whole-genome sequencing of 16 Turkish and Brazilian strains. Isolates were selected based on biofilm formation capacity, susceptibility profiles, and ERG11 mutations to account for various phenotypic and potentially genotypic differences. Phylogenetic reconstruction revealed that LBP isolates clustered together, forming two country-specific subclades, whereas BP isolates were more genetically diverse (Fig 2A). Unexpectedly, Brazilian LBP isolates were genetically closer to Turkish LBP isolates (2059.1 ± 57.3 mean SNP difference) than to Brazilian BP isolates (2862.8 ± 146.0 mean SNP difference), suggesting convergent or shared evolutionary trajectories associated with biofilm attenuation. Phylogenetic analysis of selected Turkish and Brazilian C. parapsilosis isolates (A). Schematic presentation of mutations selected for genetic functional analysis among the Turkish and Brazilian LBPs (B). Mutants were generated using CRISPR-Cas9 technology (Lombardi and colleagues, 2019), validated by PCR amplification and Sanger sequencing and at least 2–3 independent colonies were tested for each mutant. Biofilm (OD600) and adherence measured at 24 h and 90 min, respectively, for mutants created in the Turkish BP (1T) (C). Biofilm (OD600) and adherence measured at 24 h and 90 min, respectively, for mutants created in the Brazilian BP (35B) (D). Each dot represents an independent biological replicate for Fig A–D. Phylogenetic analysis of Italian, Turkish, and Brazilian C. parapsilosis isolates using maximum likelihood (E). Phylogenetic analysis of global collection of C. parapsilosis isolates collected from outbreaks in multiple countries (F). Canadian C. parapsilosis isolates clustered with Turkish LBPs (A01, A02, A03, B20, B21, and B22) produced a significantly lower biofilm (OD600) compared to counterparts grouped with Turkish BPs (B06, B09, D02, and K06) (G). Dashed line indicates the lowest biofilm threshold. BP, biofilm producers; LBP, low biofilm producers. *, **, ***, and **** represents P values ≤0.05, ≤ 0.01, ≤ 0.001, ≤ 0.0001, respectively. Data underlying this figure could be found in the S1 Data file. Notably, the Brazilian LBP clade included both YCP and FLCS isolates, whereas the Turkish LBP cluster consisted exclusively of YCP strains. We first explored if there exist phenotype-specific (LBP- or BP-specific) large or small chromosomal abnormalities. Our coverage analysis indicates that strain 1T has aneuploidy of chromosomes 4 and 8, having three copies of each rather than the typical two (S3A Table). Similarly, strain 47T has three copies of chromosome 6 (S3A Table). Several strains show evidence of previously discovered gene amplifications; including genes such as RTA3 (CPAR2_104610), ARR3 (CPAR2_601050), CDR1B (CPAR2_304370), and ERG11(CPAR2_303740) (S3B Table). Other genes have been deleted to one or zero copies in several strains (S3B Table). There are no genes that are amplified or deleted in all LBP strains or in all BP strains (S3B Table). Analysis of non-synonymous, stop-gain, and frameshift variants uniquely enriched in LBP isolates revealed that Turkish YCP strains shared a highly similar mutational profile, consistent with clonal expansion from a common ancestor. Gene ontology analysis identified enrichment in genes associated with ATP binding, purine binding, carbohydrate derivative binding, and ATPase activity (S2A Fig; S3C Table), although no single pathway was universally enriched across both countries. To functionally interrogate candidate determinants of biofilm attenuation, we introduced prioritized LBP-associated mutations into BP backgrounds using CRISPR-Cas9 (Turkish mutations into 1T; Brazilian mutations into 35B). Given prior evidence that biofilm defects in Candida species frequently involve transcriptional regulators [18,19], we prioritized mutations in predicted regulatory genes, including Efg1I280T and Ume6T46K, as well as additional candidates implicated by transcriptomic analysis (Fig 2B; S3D–S3J Table). Single, combinatorial (double, triple), and quadruple mutants were generated and sequence validated. Biofilm biomass and adhesion were quantified from two independent transformants per genotype relative to the wildtype (WT) (Fig 2B–2D). None of the individual Turkish candidate mutations reduced biofilm formation. However, the combination of prioritized mutations resulted in a marked reduction in adhesion (Fig 2C), indicating that biofilm attenuation in Turkish LBP isolates likely arises through cooperative, polygenic effects rather than a single dominant mutation. In contrast, analysis of a Brazilian LBP isolate (37B) identified a single mutation with strong phenotypic impact: Mnl1S557T. Introduction of this variant into a BP background substantially reduced biofilm formation and modestly decreased adhesion (Fig 2D). Introduction of Mnl1S557T into the Turkish BP strain (1T) also produced a significant, albeit smaller, reduction in biofilm formation (S2B Fig). However, reversion of S557T to wildtype (T557S) in the 35B background did not restore biofilm formation (S2C Fig), suggesting that additional genetic context contributes to the phenotype. Indeed, whole-genome sequencing analysis revealed that 35B carries 23 heterozygous mutations (Compared to CDC317) in the region 685068-897009, all of which undergone loss-of-heterozygosity in Mnl1S557T and retained in Mnl1T557S (S4 Table). Contribution of those mutations in biofilm formation warrants future studies. Expanded analysis of global C. parapsilosis genomes identified the same Mnl1S557T variant in isolates from the United States (CAS2012570) [8] and Germany (GL28) [29], and revealed that MNL1 polymorphisms are common across both BP and LBP clades (S2D Fig). Importantly, several BP isolates harbored N-terminal Mnl1 mutations, including an early stop codon (Q5*) present in Brazilian BP strains (32B, 35B, and 61B), indicating that simple loss of Mnl1 function is insufficient to confer biofilm attenuation. Moreover, ~37% of American FLCR isolates clustering with Turkish LBPs carried an alternative variant (Mnl1V563I), further supporting a complex and context-dependent role for Mnl1. Collectively, these findings demonstrate that biofilm attenuation in outbreak-associated C. parapsilosis strains is not attributable to a single resistance mutation but instead arises through distinct, and in some cases polygenic, genetic architectures. The closer phylogenetic relatedness of LBP isolates across countries, relative to sympatric BP strains, further suggests that biofilm attenuation represents a recurrent adaptive trajectory during outbreak evolution rather than a geographically restricted event. Biofilm attenuation has emerged independently multiple times across the global C. parapsilosis population To determine whether biofilm attenuation represents a geographically restricted phenomenon or a recurrent evolutionary trajectory, we expanded our comparative genomic analysis to include isolates with/without publicly available WGS datasets from additional outbreak-associated C. parapsilosis collections. A recent Italian study reported that the majority of fluconazole-resistant isolates carrying Erg11Y132F exhibited impaired biofilm formation [24]. We therefore selected 16 isolates from this cohort (15 FLCR LBP strains and one FLCS BP strain) for WGS and phylogenetic analysis. All Italian LBP isolates formed a tightly related cluster that was highly divergent from Turkish and Brazilian LBP clades (Fig 2E). In contrast, the single Italian BP isolate shared recent ancestry with Brazilian BP strains. Within-country SNP diversity was lowest among Turkish LBPs (mean 51.9 SNPs), intermediate among Italian LBPs (mean 103.8 SNPs), and substantially higher among Brazilian LBPs (936 SNPs), consistent with differing degrees of clonal expansion. Although multiple mutations were shared among Italian LBP strains, no specific biological pathways were significantly enriched (S3D–S3J Table), further suggesting that biofilm attenuation does not arise from a single conserved genetic module. We next incorporated WGS datasets from outbreak-associated isolates from the United States [8] and Canada [7]. Because biofilm phenotypes were not available for these collections, we examined their phylogenetic relationships relative to our Turkish, Brazilian, and Italian isolates. BP strains from Turkey, Brazil, and Italy segregated into multiple distinct clusters, with the largest cluster comprising Turkish BP strains intermingled with several Canadian and American isolates (Fig 2F). In contrast, LBP strains formed three major clades: (i) a large cluster of Turkish LBP strains that included 48.2% of Canadian isolates carrying Erg11K143R and 42.8% of American isolates branching from the same lineage; (ii) a smaller cluster in which 2.1% of American isolates grouped with Brazilian LBP strains; and (iii) a distinct Italian LBP clade that showed no close association with other global isolates. The close genetic relatedness between Turkish LBP strains and a subset of Canadian isolates prompted us to experimentally assess their biofilm phenotypes. Canadian isolates clustering with Turkish LBPs (A01, A02, A03, B20, B21, and B22) exhibited impaired biofilm formation, whereas isolates clustering with BP strains (B06, B09, D02, and K06) retained robust biofilm production (Fig 2G). These data demonstrate that phylogenetic clustering strongly predicts biofilm phenotype across geographically distinct collections. Our global comparative analysis supports two non-mutually exclusive routes for the emergence of LBP strains: first, dissemination of ancestral LBP lineages across countries, as exemplified by Turkish LBPs clustering with Canadian and some American isolates, reminiscent of the global clade structure described for C. auris clades [30]; and second, independent, recent clonal expansion of biofilm-attenuated genotypes, as observed in Italy. Together, these findings indicate that biofilm attenuation is not an isolated regional anomaly but rather a recurrent and convergent feature of outbreak evolution in C. parapsilosis. LBP strains exhibit enhanced stress tolerance and cell wall remodeling Loss of filamentation and biofilm capacity can, in certain contexts, confer a selective advantage. In C. albicans, gut-evolved isolates harboring loss-of-function mutations in biofilm- or filamentation-associated regulators display increased gastrointestinal fitness [19]. Similarly, C. albicans eed1Δ mutants, defective in filamentation and biofilm formation, exhibit elevated burdens during systemic infection [31] and oral colonization [32] compared to filament-competent strains. Notably, these biofilm-deficient isolates share features of metabolic reprogramming, consistent with our RNA-seq data indicating mitochondrial and oxidative metabolic remodeling in LBP strains, both of which are advantageous during interaction with innate immune cells [33–35]. Given the epidemiological pattern of persistent outbreaks—where a single genotype circulates across multiple patients—we hypothesized that biofilm attenuation may be coupled to enhanced fitness under host-relevant stress conditions. To test this, we quantified growth dynamics of all Turkish and Brazilian isolates (n = 120) across a panel of stress and nutrient conditions: rich medium (YPD), low glucose (0.2% dextrose), acidic pH (pH 4), oxidative stress (5 mM H2O2), osmotic stress (0.5 M NaCl), and elevated temperature (40 °C). Growth was measured across full curves and summarized as area under the curve (AUC), with at least two biological replicates per isolate. Across nearly all conditions—including both stress and non-stress environments—LBP strains exhibited significantly greater growth than BP strains, independent of geographic origin (Fig 3A and 3B). These data indicate that biofilm attenuation is associated with a generalized growth advantage rather than a condition-specific effect. AUC analysis under various conditions unravels that LBPs from Turkey (A) and Brazil (B) show a higher growth rate with/without stress. Cell suspension from overnight-grown isolates in YPD were adjusted at 0.2 OD600 in desired media and growth analysis was measured on an hourly basis. YFP-expressing LBPs outcompete mTorquise-expressing BPs under multiple 1:1 in-vitro competition analysis as judged by flow cytometry (C). Cell suspensions were obtained from overnight-grown isolates, 1:1 mixture of BP and LBP were prepared, and YFP and mTorquise population sizes were determined at desired timepoints (0, 3, 8, and 24 h) using flowcytometry. YFP and mTorquise-expressing isolates were generated using CRISPR-Cas9 technology (Nemeth and colleagues, 2025) and validated by both Southern blot, PCR amplification, and Sanger sequencing. Each dot represents an independent biological replicate for Fig A–C. Cell wall exposure analysis using fluorescence microscopy uncovers that LBPs have a significantly lower β-glucan, but a markedly higher mannan exposure (D). MFI of 50 fields were captured for each isolate and used for cell wall exposure analysis. AUC, area under the curve; BP, biofilm producers; LBP, low biofilm producers; YFP, yellow fluorescence protein; YPD, Yeast extract-peptone-dextrose; YPG, yeast extract-peptone-glycerol; MFI, Mean fluorescence intensity. *, **, ***, and **** represents P values ≤0.05, ≤0.01, ≤0.001, ≤0.0001, respectively. Data underlying this figure could be found in the S1 Data file. To directly assess competitive fitness, we performed pairwise in vitro competition assays. Representative LBP and BP strains from both countries (LBPs were not from same genotypes, see genomic section) were differentially labeled (YFP and mTurquoise) and co-cultured at a 1:1 ratio under multiple conditions, including fermentable and non-fermentable carbon sources (YPD, YPG), low glucose, oxidative stress, acidic pH, and combined stress (S2E Fig). Population frequencies were quantified over time by flow cytometry. Consistent with monoculture growth phenotypes, LBP strains consistently outcompeted BP strains under all tested conditions, with competitive divergence increasing over time (Fig 3C). These results demonstrate that LBP strains possess a robust and environment-independent fitness advantage in vitro. Because biofilm regulatory networks frequently intersect with cell wall remodeling pathways [36,37], and cell wall architecture critically shapes host–fungus interactions [38], we next examined cell wall architecture in a subset of isolates (n = 16 from both countries). Using fluorescence microscopy, we quantified surface exposure of chitin (WGA-Alexa 680), β-glucan (anti-dectin-1-Alexa 488), and mannan (Concanavalin A–Texas Red). LBP strains displayed significantly increased mannan exposure and reduced β-glucan exposure relative to BP strains (Fig 3D). Reduced β-glucan exposure is particularly notable given its role as a major ligand for innate immune recognition. These findings indicate that biofilm attenuation in outbreak-associated strains is coupled to enhanced stress tolerance, competitive growth, and altered cell wall architecture. Rather than representing a simple loss of virulence capacity, which is biofilm attenuation, the LBP phenotype appears to reflect a coordinated physiological reprogramming that may enhance survival under host-imposed pressures. LBP strains exhibit enhanced immunoevasion and increased tolerance to innate immune cell interactions β-glucan is a dominant fungal pathogen-associated molecular pattern (PAMP) recognized by host pattern recognition receptors (PRRs), including dectin-1, on innate immune cells. Masking of β-glucan outer mannan layers represents a well-established immune evasion strategy in fungal pathogens [38]. Given that LBP strains display reduced β-glucan exposure and increased mannan surface abundance, we hypothesized that they would be less efficiently recognized and eliminated by primary human innate immune cells. Here, we defined immunoevasion as reduced recognition and phagocytosis by innate immune cells. We therefore examined interactions of representative Turkish and Brazilian LBP and BP isolates with primary human monocyte-derived macrophages and neutrophils (n = 16 isolates, unless otherwise indicated). The LBP isolates represented distinct genotypes, as described in the genomic analyses below. For macrophage interactions, phagocytosis was quantified as intracellular colony-forming units (CFUs) at 1 h post-infection (normalized to input), and intracellular survival was assessed at subsequent timepoints relative to the 1 h internalized population. Reactive oxygen species (ROS) production was measured by dihydrorhodamine 123 fluorescence at 3 h post-infection. Macrophages from at least two independent donors were used, and extensive washing ensured that only intracellular yeast were quantified. LBP strains were significantly less efficiently phagocytosed than BP strains (Fig 4A) and elicited lower ROS production (Figs 4B and S2F). Notably the lower phagocytosis of LBPs could not be explained by their lower plastic adhesion, since (1) yeast-macrophage co-cultures were briefly centrifuged to bring the yeast cells in close proximity of macrophages, (2) the entire surface of the wells were covered by macrophages leaving limited space for attachment, and finally, (3) extensive washing steps and subsequent microscopic inspection ensured absence of extracellular yeast cells attached to the plastic. Between 1 and 3 h post-infection, LBP strains exhibited significantly greater intracellular survival relative to BP strains (Fig 4C), although survival equalized by 24 h (S2G Fig). Consistent with reduced β-glucan–dependent activation, macrophages infected with LBP strains (n = 35 isolates) produced significantly lower levels of IL-6 compared to BP strains at 24 h (Fig 4D), whereas other cytokines (TNFα and IL1β) were below detection in this assay. LBPs are immunoevasive and showed higher fitness during interaction with primary human macrophages and neutrophils. LBPs were less effectively phagocytosed by macrophages (A) and consistently macrophages infected with LBPs produced a lower ROS (B). MFI of FITC-labeled yeasts phagocytosed by macrophages and DHR123 measured by flow cytometry were used for Fig A and B. Of note, ROS gating was adjusted based on un-infected control macrophages. Furthermore, intracellular LBPs showed a significantly higher survival (CFU) compared to intracellular BPs at 3 h (normalized to average of pertinent intracellular CFU at 1 h) (C). Macrophages infected with LBPs produced a lower IL6 (D) and showed a lower damage as judged using lactate dehydrogenase assay (E). Genomic analysis of Italian, Turkish, and Brazilian C. parapsilosis isolates showed that LBPs had a higher number of genes containing premature stop-codon, frameshift, and insertion and deletions (F) and those genes were mainly implicated in gene expression, biosynthetic, and metabolic processes (G). Consistent with observations made with macrophages, neutrophils less effectively killed LBPs at all timepoints tested (CFU at each time point was normalized to pertinent initial inoculum) (H), and neutrophils infected with LBPs showed a higher degranulation (I) and shedding (J) than those infected with BPs. Finally, LBPs were immunoevasive and less phagocytosed by neutrophils (K) and produced lower ROS compared to those infected with BPs (L). Flow cytometry was used to determine the MFI of ROS, degranulation, and shedding for neutrophils infected with C. parapsilosis and gating was adjusted based on un-infected control neutrophils. Phagocytosis was determined by measuring the MFI of FITC-labeled yeasts phagocytosed by neutrophils using flow cytometry. Neutrophils and macrophages obtained from at least two healthy donors were used for analysis. Notably, phagocytosis was confirmed by extremely low association of C. parapsilosis cells with neutrophils and macrophages following cytochalasin D treatment. Each dot represents an independent biological replicate. BP, biofilm producers; LBP, low biofilm producers; ROS, reactive oxygen species; MFI, mean fluorescence intensity; CFU, Colony forming unit. *, **, ***, and **** represents P values ≤0.05, ≤ 0.01, ≤ 0.001, ≤ 0.0001, respectively. Created in BioRender. Löffler, J. (2026) https://BioRender.com/4kb4e7g. Data underlying this figure could be found in the S1 Data file. We next assessed macrophage damage at 24 h using lactate dehydrogenase (LDH) release. LBP strains caused significantly less host cell damage than BP strains (Fig 4E), and consistently their extracellular CFUs at 24 h were lower compared to BP strains (S3A Fig). Together, these findings indicate that LBP strains are less cytotoxic, and more immunoevasive and tolerant of intracellular residence during early macrophage interaction. Comparative genomic analysis further revealed that LBP strains harbored a trend towards a higher—though not statistically significant (p = 0.07)—number of genes containing premature stop codons or frameshift mutations relative to BP strains (Fig 4F and S3K Table). These mutations were enriched in genes associated with gene expression and metabolic processes (Fig 4G and S5 Table). While preliminary, this pattern is consistent with genetic streamlining observed in host-adapted bacterial pathogens [39,40] and supports the possibility that outbreak-associated LBP strains are undergoing specialization toward host-associated niches. We next examined interactions with primary human neutrophils using the same isolate set. Survival was measured at 1, 2, and 6 h post-infection relative to input. Phagocytosis, ROS production, CD62L shedding, and CD66b degranulation were quantified by flow cytometry. Across all timepoints, LBP strains exhibited significantly higher survival than BP strains (Fig 4H). Consistent with macrophage results, LBP strains were less efficiently phagocytosed (Figs 4K and S3D) and triggered significantly lower ROS production (Figs 4L and S3E). Although neutrophil degranulation was modestly increased in response to LBP strains, this difference did not reach statistical significance, whereas CD62L shedding was significantly elevated (Figs 4I–4J and S3B–S3C). Collectively, these data demonstrate that LBP strains evade innate immune recognition, reduce inflammatory activation, and exhibit enhanced survival during interactions with both macrophages and neutrophils. Rather than representing a simple loss of biofilm capacity, the LBP phenotype is coupled to coordinated cell wall remodeling and increased tolerance to immune-mediated stress. These features support a model in which biofilm attenuation is linked to enhanced host adaptation, potentially contributing to the persistence of outbreak-associated lineages. Macrophages infected with LBP strains adopt a proinflammatory transcriptional state and enhance early neutrophil fungicidal responses The combination of reduced phagocytosis, diminished IL-6 production, limited macrophage damage, and enhanced early intracellular survival prompted us to examine the transcriptional response of macrophages to LBP (29T, 49B) versus BP (1T, 35B) strains at 3 and 24 h post-infection. Using the same isolates analyzed in fungal RNA-seq experiments, macrophages were extensively washed to remove extracellular yeast, lysed, and subjected to RNA-seq. Transcriptomes were normalized to uninfected controls. Gene ontology (GO) analysis revealed distinct temporal patterns. At 3 h post-infection, macrophages infected with LBP strains were enriched for pathways associated with leukocyte cell–cell adhesion. In contrast, BP-infected macrophages showed strong enrichment at 24 h for granulocyte migration, neutrophil chemotaxis, and inflammatory signaling pathways (Figs 4B and S4A; S6 and S7 Tables). GSEA provided further resolution. At 24 h, macrophages infected with BP strains exhibited coordinated upregulation of inflammatory mediators within the “Hallmark IL-6/JAK/STAT3” pathway (consistent with higher IL6 production of macrophages infected with BP strains), including IL1B, IL6, TNF, CXCL1–3, SOCS3, and JUN (Fig 5A and S8 Table), alongside robust induction of chemokines (CXCL8, CCL2, CCL3, CCL4, CCL7, S100A8/A9) associated with granulocyte recruitment (Fig 5B and S9 Table). Enrichment plots depicting pathway enrichment (A, B: left) and heatmaps of transcript levels at the leading-edge for the “HALLMARK_IL6_JAK_STAT3_SIGNALING” pathway, and “GO:BP_GRANULOCYTE_MIGRATION” at 24 h (A, B: right). “Normalized enrichment score (NES)” and “adjusted P-value” are marked under the enrichment plots. Leading-edge analysis and heatmap showed that macrophages infected with the BPs, unlike LBPs, significantly upregulated the JAK-STAT signaling (A) and strong chemokine induction in the granulocyte neutrophil migration pathway (B). Furthermore, macrophages infected with LBPs were skewed toward an M1 profile and were more proinflammatory at 3- and 24hrs post-infection (C). Note that the P value is a nominal P-value, while NES is the normalized enrichment score (C). Supernatants collected from macrophages infected with LBPs (at 24hrs post-infection), induced a higher swarming rate (D) and fungicidal effector function (E) of neutrophils toward immobilized iRFP-expressing C. albicans yeast cells, whereas neutrophils treated with supernatants of macrophages infected with BPs showed a higher NETOSIS rate (F). Of note, supernatants were filter-sterilized using 0.2µ sterile syringe filters and MFI was used for swarming, fungicidal effector function (killing), and NETOSIS rates. Neutrophil fungicidal effector function (E) was determined by dividing the MFI iRFP-expressing C. albicans co-cultured with neutrophils over the MFI of pertinent C. albicans control at each time point. Each dot represents an independent biological replicate. LBPs showed comparable survival rate than BPs in kidney (G) and liver (H), whereas they had a markedly higher survival rate in spleen compared to those infected with BPs (I) in the context of an intravenous (tail-vein injection) systemic infection mouse model. Mice were infected with 2*108 CFU of LBPs and BPs and CFU were determined for each homogenized organ at designated timepoints by plating on YPD agar. Each dot represents an independent mouse. BP, biofilm producers; LBP, low biofilm producers; NETOSIS, neutrophil extracellular trap; CFU, Colony-forming unit; YPD, Yeast extract-peptone-dextrose; MFI, mean fluorescence intensity. *, **, ***, and **** represents P values ≤0.05, ≤0.01, ≤0.001, ≤0.0001, respectively. Created in BioRender. Löffler, J. (2026) https://BioRender.com/4kb4e7g. Data underlying this figure could be found in the S1 Data file. Previous dual species transcriptome studies demonstrated that in the initial stages of macrophage interactions with C. albicans yeast cells, the macrophages polarize toward a proinflammatory (M1-like) state, but at later time points, when C. albicans has filamented, the macrophages skew toward a less inflammatory (M2-like) state [41]. Since LBP strains exhibited hallmarks of intracellular containment—minimal macrophage damage and high survival between 1 and 3 h post-infection—we hypothesized that LBP strain-infected macrophages would exhibit an M1-like transcriptional state. Consistently, macrophages infected with LBP strains displayed a transcriptional profile enriched for genes classically associated with proinflammatory (M1-like) [41] activation at 24 h (Fig 5C and S10 Table). Notably, this occurred despite reduced cytokine release and diminished ROS induction, suggesting that LBP strains modulate the kinetics and nature of macrophage activation rather than globally suppressing inflammation. Therefore, despite reduced recognition and phagocytosis, intramacrophage LBPs tend to reprogram the immune responses toward an inflammatory state. Because tissue-resident macrophages coordinate neutrophil recruitment and activation [42], we hypothesized that macrophage-derived mediators elicited by LBP strains would differentially influence neutrophil function. To test this, we used a microfluidic assay quantifying neutrophil swarming, fungicidal activity, and NETosis [43]. Filter-sterilized supernatants from macrophages infected for 24 h were added to neutrophil-Candida co-cultures, and responses were monitored by time-lapse confocal imaging. Neutrophils exposed to supernatants from LBP-infected macrophages exhibited accelerated swarming dynamics (Fig 5D) and enhanced fungal killing (Figs 5E and S4C). In contrast, supernatants from BP-infected macrophages induced higher levels of NETosis (Fig 5F). These findings suggest that LBP strains promote a macrophage activation state that favors coordinated neutrophil recruitment and fungicidal activity rather than excessive NET formation. Overall, these data reveal a nuanced immunological phenotype: although LBP strains are less efficiently recognized and killed during direct innate immune interactions, they induce a macrophage transcriptional program that enhances neutrophil-mediated fungal clearance. This pattern is consistent with a model in which outbreak-associated LBP strains balance immune evasion with sustained host engagement, potentially contributing to prolonged persistence under immune pressure. LBP strains exhibit enhanced persistence in immune cell-rich organs during systemic infection Our in vitro and ex vivo analyses indicated that LBP strains possess a competitive growth advantage and enhanced tolerance to innate immune cell interactions. To determine whether these features translate into altered fitness during systemic infection, we evaluated representative Turkish and Brazilian LBP and BP isolates (n = 11 total strains, LBPs were not from the same genotype as shown in genomic section) in an immunocompetent murine model of hematogenously disseminated candidiasis. Mice were infected via tail-vein injection with 2 × 107 CFUs, and fungal burdens were quantified in kidneys, liver, and spleen at days 4 and 7 post-infection (3–4 mice per isolate per time point). Organs were homogenized and plated for CFU enumeration. LBP strains exhibited modestly higher fungal burdens in the kidney and liver relative to BP strains, although these differences did not reach statistical significance (Fig 5G–5H). In contrast, LBP strains displayed a significantly higher fungal burden in the spleen (Fig 5I), an organ enriched in macrophages and other immune cell populations. The preferential persistence of LBP strains in the spleen is notable given their enhanced resistance to macrophage and neutrophil-mediated killing observed ex vivo. Accordingly, LBP strains appear to display improved survival within immune cell-dense environments, consistent with their remodeled cell wall architecture and increased tolerance to innate immune stress. Together, these findings indicate that biofilm attenuation is not associated with reduced in vivo fitness. Instead, outbreak-associated LBP strains demonstrate enhanced persistence in specific host niches, particularly immune cell-rich tissues, supporting a model in which biofilm loss is coupled to host-adaptive traits that promote survival during systemic infection. Disruption of biofilm regulatory circuits enhances survival during innate immune interactions Our data thus far indicated that outbreak-associated LBP strains display enhanced survival during interactions with innate immune cells. We therefore hypothesized that attenuation of biofilm formation per se—whether arising through naturally occurring substitutions or targeted disruption of biofilm regulators—could promote fitness during host immune interactions. This hypothesis prompted us to assume that even random mutations leading to biofilm attenuation (such as Mnl1S557T) could also enhance fitness during interaction with innate immune cells. To prioritize mutants for detailed ex vivo analysis, we first screened all strains generated in both Turkish and Brazilian isolates for IL-6 induction in macrophages, as LBP clinical isolates elicited markedly reduced IL-6 production. Macrophages infected with strains harboring Mnl1S557T (Fig 6A) or Efg1I280T (Fig 6B) produced significantly less IL-6 relative to their parental WT strains, consistent with the attenuated inflammatory phenotype observed in LBP isolates. These mutants were therefore selected for subsequent functional analyses. Efg1I280T (A) and Mnl1S557T (B) induced lower IL6 production during interaction with macrophages. Neutrophils infected with mutants produced a higher ROS (C), whereas only Efg1I280T was less effectively phagocytosed by neutrophils (D) compared to pertinent wildtypes. Mnl1S557T showed a higher survival during interaction with neutrophils (E) and biofilm level (OD600) of mutants was negatively correlated to survival during interaction with neutrophils (F). Flow cytometry was used to determine the MFI of ROS for neutrophils infected with C. parapsilosis and gating was adjusted based on un-infected control neutrophils. Phagocytosis was determined by measuring the MFI of FITC-labeled yeasts phagocytosed by neutrophils using flow cytometry. Survival was determined by normalizing CFU at 2 h post-infection neutrophil against the CFU of the pertinent initial inoculum. All neutrophil effector functions obtained for each mutant were further normalized against the average of WT for each donor (to precisely capture how mutants impact neutrophil effector functions obtained from each donor and to minimize the donor-to-donor variability). Similarly, intracellular Mnl1S557T had a higher survival rate during interaction with macrophages at 3 h post-infection (normalized to pertinent intracellular at 1 h post-infection) compared to pertinent wildtype (G) and biofilm level (OD600) was negatively correlated to intracellular survival at 3hr post-infection (H). Phagocytosis (I) and survival (J) of C. parapsilosis mutants lacking key transcription factors/ kinase involved in biofilm regulation compared to parental WT strain (ATCC 22,019) during interaction with macrophages. Phagocytosis (K), ROS (L), and survival (M) of the same mutants of C. parapsilosis during interaction with neutrophils. C. parapsilosis survival during interaction with macrophages was determined by normalizing the CFU of intracellular yeast at 3 h against pertinent intracellular yeast CFUs at 1 h post-infection and data were further normalized to the average of WT for each donor (to precisely capture how mutants impact neutrophil effector functions obtained from each donor and to minimize the donor-to-donor variability). Neutrophils and macrophages obtained from at least two healthy donors were used for analysis. BP, biofilm producers; LBP, low biofilm producers; ROS, reactive oxygen species; MFI, Mean fluorescence intensity; WT, wildtype. *, **, ***, and **** represents P values ≤0.05, ≤ 0.01, ≤ 0.001, ≤ 0.0001, respectively. Created in BioRender. Löffler, J. (2026) https://BioRender.com/4kb4e7g. Data underlying this figure could be found in the S1 Data file. Under stress-free and oxidative stress conditions (YPD, pH 4 YPD ± H2O2, YPG ± H2O2), neither mutant exhibited improved growth relative to WT (S4D and S4E Fig), suggesting that the enhanced stress tolerance observed in LBP clinical isolates likely requires additional mutations. In contrast, clear differences emerged during immune cell interactions. Although neutrophils exposed to both mutants generated significantly higher reactive oxygen species (ROS) (Fig 6C), unlike clinical LBP isolates the Mnl1S557T mutant demonstrated significantly enhanced survival despite similar phagocytosis rates (Fig 6D and 6E). This observation is expected as laboratory-generated mutants do not fully represent the complex genotype of clinical LBP isolates. Survival during neutrophil interaction negatively correlated with biofilm formation capacity (Fig 6F). Likewise, intracellular survival of the Mnl1S557T mutant within macrophages was significantly increased relative to WT (Fig 6G), and macrophage survival similarly was inversely correlated with biofilm levels (Fig 6H). These findings prompted us to test whether targeted deletion of established biofilm regulators in C. parapsilosis would recapitulate this phenotype. We examined ace2Δ, bcr1Δ, cph2Δ, czf1Δ, gzf3Δ, efg1Δ, and ume6Δ mutants, previously shown to regulate biofilm formation [25]. Reintegration of the WT alleles fully rescues biofilm formation in these strains [25], confirming that observed phenotypes are attributable to the targeted deletions. Consistent with prior reports and our in vitro validation (S4F Fig), ace2Δ, bcr1Δ, czf1Δ, gzf3Δ, efg1Δ, and ume6Δ exhibited attenuated biofilm formation, whereas cph2Δ displayed biofilm levels comparable to WT. Notably, all mutants showed reduced growth rates under standard and stress conditions (S4G–S4J Fig), indicating that any fitness advantages during immune interaction are unlikely to stem from generalized growth enhancement. During macrophage interaction, phagocytosis rates were largely comparable to WT, except for ace2Δ, which showed reduced uptake (Fig 6I). Despite their growth defects, several biofilm-attenuated mutants exhibited enhanced intracellular survival: ace2Δ and bcr1Δ showed significantly increased survival, whereas cph2Δ, gzf3Δ, and ume6Δ were comparable to WT (Fig 6J). During neutrophil interaction, ace2Δ and gzf3Δ elicited lower ROS production, consistent with reduced phagocytosis (Fig 6K and 6L). Importantly, most biofilm-attenuated mutants (ace2Δ, bcr1Δ, czf1Δ, and efg1Δ) demonstrated significantly enhanced survival following neutrophil exposure, while gzf3Δ and ume6Δ exhibited survival comparable to WT (Fig 6M). Collectively, these functional analyses demonstrate that disruption of biofilm regulatory networks—whether through naturally occurring substitutions or targeted genetic deletion—does not impair, and frequently enhances, survival during innate immune interactions. These findings provide mechanistic support for an evolutionary trade-off model in which attenuation of biofilm formation is coupled to increased fitness within immune cell–rich host environments, thereby facilitating persistence during systemic infection and outbreak propagation. Discussion By integrating multi-omics analyses, functional genetics, and extensive in vitro, ex vivo, and in vivo experiments, we demonstrate that impaired biofilm formation is a recurrent and globally conserved feature of recent C. parapsilosis outbreaks, enhancing tolerance to innate immune cell effector functions. Our findings provide several novel microbiological insights. First, although biofilms serve as a robust protective shield against environmental stressors, including disinfectants, antifungals, and immune defenses, C. parapsilosis can attenuate the molecular networks underlying this ancient structure to better survive host immunity. This suggests that host pressures, rather than environmental stressors, are a major selective force shaping outbreak evolution. Supporting this notion, repeated gut passage of C. albicans favors mutants defective in filamentation and biofilm formation [19], and analogous patterns are observed in bacterial pathogens, where biofilm loss enhances host colonization or invasiveness, such as in African lineages of invasive nontyphoidal Salmonella (iNTS) [44,45]. A key question is why biofilm attenuation enhances in-host fitness. Our ex vivo macrophage studies indicate that LBP strains adopt a “yeast-locked intracellular lifestyle,” reminiscent of C. glabrata. LBP strains induce minimal macrophage damage, produce fewer extracellular yeasts at 24 h, and survive initial fungicidal activity more efficiently, correlating with higher spleen burdens in immunocompetent mice. Similar intracellular adaptation is seen in Cryptococcus gattii outbreak isolates, which exhibit enhanced macrophage fitness and virulence [46]. This yeast-locked lifestyle may facilitate bloodstream dissemination [31], conferring a selective advantage during systemic infection. Alternatively, biofilm formation may impose a metabolic cost within the host; impairing this pathway could conserve energy and enhance survival under immune pressure. Consistent with this concept, studies in Salmonella demonstrate that extracellular matrix components such as cellulose and curli fimbriae can act as PAMPs, and their loss improves infectivity [47–51]. These observations suggest that biofilm attenuation represents a conserved, trans-kingdom strategy to enhance microbial fitness against host immunity. Second, the increased fitness of LBP strains may contribute to the higher mortality observed in patients infected with YCP isolates and to their persistence in hospital settings. Turkish patients infected with LBP strains exhibit significantly higher mortality compared to BP strain infections [9,20,21], likely reflecting enhanced survival within host tissues and immune evasion. While immunopathology may also contribute, these data suggest that LBP strains can persist independently of environmental biofilm reservoirs, highlighting human carriage (e.g., healthcare workers’ hands) as a major transmission route, reminiscent of Salmonella Typhi [52–54]. These findings underscore the need for infection control strategies targeting non-environmental reservoirs. Third, our WGS analyses reveal that LBP strains can emerge from a shared ancestor or independently across distinct genotypes. Whether they originate from pre-existing biofilm-deficient isolates or evolve adaptively under host immune pressure remains unclear, but persistent immune selection clearly favors isolates capable of evading or resisting innate defenses. Fourth, despite biofilm attenuation being a recently selected phenotype, LBP strains predominantly repurpose evolutionarily conserved genes for biofilm formation. In contrast, BP strains rely on younger or species-specific genes during late-stage biofilm development, consistent with prior observations in C. albicans [28]. This suggests that evolutionarily older genes can be flexibly repurposed to maintain minimal biofilm functions while supporting immune evasion. Finally, although these findings may not directly generalize to other outbreak-causing pathogens such as C. auris, they caution against indiscriminate use of antibiofilm agents. Prolonged or repeated exposure to such compounds could inadvertently select for strains with enhanced immune evasion and systemic fitness. Investigating how antibiofilm interventions influence microbial adaptation and pathogenicity in host contexts is therefore warranted. Moreover, role of biofilm in outbreak persistence caused by other fungal pathogens, such as C. auris and C. tropicalis, requires future studies to further investigate if biofilm attenuation is C. parapsilosis-specific or generalizable to other fungal pathogens. Methods Ethics statement This study was conducted in accordance with local legislation and institutional requirements, and pertinent protocols were approved by the ethics committee of Ege University Faculty of Medicine (20-2 T/30), University Hospital of Bari (Study 1338/CE- Approved Prot. 490 – 14/9/2023, “OUTBREAKS IN TERAPIA INTENSIVA: II valore della multidisciplinarietà tra Infection Control e Antimicrobial Stewardship), and part of a FAPESP GRANT (2017/02203-7) that was reviewed and approved by local Ethical Committee at Universidade Federal de São Paulo and Antimicrobial Resistance Institute of São Paulo. Kinetic growth analysis using plate reader and flow cytometry C. parapsilosis isolates were grown in YPD overnight, followed by washing thrice with PBS, resuspending the OD600 of 0.2 in desired media, and dispensing 200 µl of cell suspensions in 96-well plates, sealed with Breathe-EasyR sealing membrane covers (Millipore Sigma). Subsequently, plates were incubated at 37 °C, unless stated otherwise, in a Tecan Microplate Reader (Infinte 2000 pro) and OD600 reading with an hour interval. To capture the entire dynamic of growth curves, we measured the AUC for each replicate in all conditions. For in-vitro competition experiments, we used YFP- and mTorquise-expressing LBP strains and BP strains, respectively. We created mixture of 1:1 of LBP strains and BP strains from different countries, inoculated pertinent media with 1:1 mixture, and monitored the proportion of each isolate in competition assays throughout the course of competition using a Celesta BD FACS Flow cytometer (BD Corporation). Biofilm assays Biofilm assays using crystal violet, optical density, and confocal scanning laser microscopy assays were carried out as previously described [11,55]. Adherence assays were performed as previously described [56] except using OD600 as a readout. Cell wall staining of mannan and exposed chitin and glucan C. parapsilosis isolates grown in YPD for 6 h (30 °C) were washed thrice with 1xPBS, followed by blocking with FACS block (0.5% BSA, 5% HI-rabbit serum, 5 mM ETDA, 2 mM NaAzide in PBS) for 30 min. Subsequently, pellets were washed thrice with FACS buffer (0.5% BSA, 5 mM EDTA, 2 mM NaAzide in PBS) at 4 °C, followed by resuspending the pellets in 5 µg/ml Fc:Dectin1 protein and incubated for 1 h at 4 °C. Next, pellets were washed thrice with PBS, stained with 50 µg/ml Wheat Germ Agglutinin-Alexa-680 to visualize exposed chitin, for 30 min at 4 °C. To visualize the exposed mannan, pellets washed thrice with PBS were stained with 50 µg/ml Concanavalin A-Texas Red for 30 min at 4 °C. DIC and fluorescence microscopy using an UltraVIEW VoX spinning disk confocal microscope (Nikon, Surrey, UK) were employed for all samples, and β1,3-glucan, chitin, and mannan exposures were measured by quantifying the MFI of each dye for at least 30 individual yeast cells of each isolate. Transformation of C. parapsilosis Transformation involving the mutations found using WGS followed the exact CRISPR-Cas9 protocol described in our previous studies [57], whereas fluorescently labeled C. parapsilosis isolates were created in accordance with a previous study [58]. All oligonucleotides used for transformation purposes are listed in S11 Table. Macrophage isolation, infection, and functional analysis Monocytes were sourced from leukopak of healthy donors attending Massachusetts General Hospital (IRB# 2014P002377). Following monolayer isolation using Ficol (ThermoFisher), monocytes were isolated using EasySep Direct Human Monocyte Isolation Kit (STEMCELL Technologies) in accordance with protocols described in our previous studies yielding high-quantity and quality monocytes [11,59]. Purity (CD45, CD16, CD14) and viability (staining with 7-AAD) were determined using flow cytometry, which were ≥93% and ≥99 in all experiments, respectively. Human macrophage-colony stimulating factor (hM-CSF, 50ng/ml) was used to differentiate monocytes into fully mature macrophages in fresh complete RPMI = cRPMI (10% heat-inactivated FBS, 1% pen-strep, 1% L-glutamine) as described previously. Overnight YPD-grown C. parapsilosis isolates washed thrice with PBS (at desired MOIs) were used to infect fully mature macrophages on day 8. To examine the intracellular survival of C. parapsilosis, macrophages were infected with an MOI = 3/1, followed by incubation in at CO2 incubator for 1 h, extensively washing macrophages to remove the non-engulfed C. parapsilosis cells at 1 h pi, adding fresh cRPMI and returning to CO2 incubator. Macrophages were lysed with ice-cold water at designated timepoints (1-, 3-, and 24-h), followed by plating on YPD agar plates and incubating plates at 37ºC for 48h and counting CFUs. Phagocytosis was determined by normalizing the CFU of intracellular C. parapsilosis at 1 h pi against the pertinent initial inoculum, whereas survival was measured by normalizing the intracellular C. parapsilosis CFUs at designated timepoints against the pertinent 1 h intracellular C. parapsilosis CFUs. To measure ROS, macrophages were stained with CellMask-Deep Red (ThermoFisher, APC700) prior to infection, followed by infection with C. parapsilosis isolates (MOI = 10/1), washing macrophages 1h pi, adding dihydroxyrhodamine 1,2,3 (DH1,2,3, ThermoFisher) and incubation for an extra 2 h. Flow cytometry was used to determine the MFI of ROS of APC700+ infected macrophages. Un-infected macrophages were used for gating the ROS. Cytokine measurement Primary human macrophages were infected with C. parapsilosis isolates at an MOI = 10. Supernatants were collected 24 h pi and subjected to cytokines were quantified using ELISA MAX Deluxe set Kits (BioLegend) in accordance with the manufacturer’s instructions [60]. Macrophage cytotoxicity measurement Macrophage cytotoxicity used a commercially available LDH kit (Sigma Aldrich) in accordance with a previously described protocol [61]. Macrophages were infected with the MOI of 8 yeasts/1 macrophages (MOI = 8/1), followed by extensively washing macrophages with PBS at 3 h pi. After an additional 21 h incubation in CO2 incubators, plates containing infected macrophages were centrifuged (1,500g, 5 min, room temperature), collecting supernatants, and subjecting them to LDH kit. Negative control included supernatant of un-infected macrophages, the value of which was subtracted from all infected counterparts, and positive control was un-infected macrophages treated with 0.25% Triton 100-X for 3 min. The subtracted value of all infected macrophages was normalized against positive control and data were presented as percentage. Neutrophil isolation and infection EasySep Human Neutrophil Isolation Kit (STEMCELL Technologies) were used to isolate fresh neutrophils from the blood of healthy donors (IRB #2014P002377), which yields highly pure and viable neutrophils (≥99%). Phagocytosis, ROS production, degranulation, shedding (all MOI = 3/1), and C. parapsilosis survival (MOI = 1/3) were determined as outlined previously [11,62]. Neutrophil swarming and analysis were carried out in accordance with previously described protocols without modification [11]. To study the impact of macrophage supernatants on neutrophil activities, macrophages infected with LBP strains and BP strains for 24 h were subjected to centrifugation (500g, 5 min, room temperature), followed by collecting supernatants, adding active 10% FBS, passing though 0.2µ filter units and storing them in CO2 incubator until use. Sixteen-well microfluidic chambers containing iRFP-expressing C. albicans were mounted on confocal microscopy (Nikon Inverted Microscope Eclipse Ti-E equipped with a CSU-X1 confocal spinning disk head (Yokogawa)), followed by adding freshly isolated neutrophils treated with pertinent supernatants and measuring swarming rates, NETOSIS, and C. albicans viability (MFI of iRFP as a proxy for survival) and capturing images every 10 min for 12 h. RNA extraction, library preparation, and sequencing Growth of C. parapsilosis biofilms was performed as described previously [22]. Biofilms were harvested by gently pipetting up and down along the bottoms of the 12-well plates and combining the biofilm slurry of the same strain from each well of one 12-well plate into a 50 mL conical tube. Planktonic cultures of a given biological replicate were combined in a 50 mL conical tube. Three biological replicates were completed for each sample and each time point. The conical tubes were centrifuged at 4,000g for 5 min and the supernatant was aspirated. Pellets were snap-frozen and stored at −80 °C until RNA extraction. Total RNA was extracted using the Ribopure RNA Purification Kit for Yeast (AM1926). mRNA was separated from total RNA using the NEBNext Poly(A) mRNA Isolation Module (E7490L). 500 ng of purified mRNA were made into cDNA libraries using the NEBNext Ultra II RNA Library Prep Kit for Illumina (E7775L). Libraries were sequenced on the AVITI 150 PE75 at the UC Davis Sequencing Core. To assess the transcriptomic responses of macrophages, they were infected with C. parapsilosis (MOI = 5/1), subjected to extensive washing with PBS at 1 h post-infection, followed by lysing macrophages using a previously described RLT buffer-based protocol. Following DNA removal using RNase-free DNase, RNA samples were subjected to purification using RNeasy kit (QIAGEN) per manufacturer’s instruction. High-quality RNA samples determined by running RNA samples on 1.5% agarose gel and NanoDrop (ThermoFisher) were used for RNA-seq using mRNA capture as described in our previous studies. Library preparation and RNAseq were performed by Azenta Life Sciences in the exact same manner as described previously [63]. RNA-seq data of macrophages infected with C. parapsilosis can be accessed with the following GEO accession number GSE344550. RNA-Seq data processing for C. parapsilosis transcriptome The paired-end reads were assessed for quality using FastQC (v0.11.9) [64]. The reference C. parapsilosis genome and its annotation were retrieved from the Candida Genome Database (CGD) [65] on 16th January, 2024. The reference genome was indexed using the subread-buildindex function in subread (v2.0.6) [66]. We then mapped the paired-end reads to the reference genome using the subread-align function in subread (v2.0.6) [66], including multi-mapped reads, mapping to the best 10 genomic loci (parameters: –multiMapping, -B 10). Transcript abundance for each gene was computed using the featureCounts function in subread (v2.0.6) [66], counting only read-pairs where both ends map to the same chomosome (parameters: -B, -C, –countReadPairs). The transcript abundance table was generated by consolidating all serial isolates, biological replicates, time point (90 min, 8 h and 24 h) and growth condition (biofilm and planktonic growth condition) using a custom Python script (v3.8.18) and were used for subsequent differential expression analyses. Differential expression and functional enrichment analysis for C. parapsilosis transcriptome Transcript abundances were used to identify differentially expressed genes in the LBP strains (29T and 49B) relative to BP strains (1T and 35B) from the corresponding country of origin across three timepoints (90 min, 8 h and 24 h) in the biofilm condition. Additionally, strain-specific gene expression in biofilm condition was obtained relative to planktonic growth for each strain independently at each time point. Differential expression analysis was performed using DESeq2 (v1.44.0) [67]. P-value estimates were adjusted for multiple hypothesis testing using independent hypothesis weighting (IHW) (v1.32.0) [68] and log2 fold change shrinkage was performed using apeglm (v1.26.1) [69]. For the heatmap comparing expression of adhesion-related genes, we obtained genes in C. albicans involved in this process [70] and identified corresponding orthologs in C. parapsilosis as annotated in CGD [65]. The heatmaps were generated using the pheatmap package (v1.0.12) in R (v4.4.3). Functional enrichment of gene categories across isolates, was performed using GSEA [71] by sorting genes in their descending order of expression. GSEA was performed using the clusterProfiler package (v4.12.6) [72] in R (v4.4.3). Gene age enrichment was performed for the set of genes differentially expressed in both LBP strains relative to the corresponding BP from the corresponding country of origin. To infer gene age, orthologs of C. parapsilosis genes were identified across 8 CTG-clade species (C. albicans, C. dubliniensis, C. tropicalis, C. guilliermondii, C. auris, C. haemulonii, C. duobushaemulonii, C. lusitaniae), 3 non-CTG species (C. glabrata, C. kefyr, C. krusei) along with Saccharomyces cerevisiae, a close relative of C. glabrata and 2 distantly related Ascomycota species (Aspergillus nidulans and Schizosaccharomyces pombe) using OrthoFinder (v2.5.4) [73]. If a C. parapsilosis gene is found only in CTG-clade species, it is considered “Young”, if it is also found in non-CTG species it is considered “Middle” age and if it is found in distant Ascomycota species, it is considered “Old”. If no ortholog of the C. parapsilosis gene is found in any other CTG, non-CTG or Ascomycota species mentioned above, it is considered “Unique” to C. parapsilosis. To estimate enrichment of different gene ages for a set of genes, a hypergeometric test was conducted, and p-value was adjusted for multiple testing using Benjamini-Hochberg correction. RNAseq data of C. parapsilosis isolates under planktonic and biofilm conditions are accessible using the following GEO accession number GSE312575. Bulk RNA-Seq data processing and enrichment analysis for macrophages’ transcriptome Raw counts from RNA-seq data and the associated sample metadata were loaded in R and harmonized by concatenating “Sample.name”, “group”, and “time point” to align column names between matrices. C. parapsilosis isolates were stratified into BP and LBP groups (BP: 1T, 35B; LBP: 29T, 49B) and timepoints (control, 3 h, 24 h). Data were analyzed using edgeR [74]; briefly, library sizes were normalized, lowly expressed genes were filtered, and a design matrix without an intercept was built on the interaction of biofilm group and time (model: ~ 0 + group:time). Differential gene expressions were tested for macrophages stimulated with “BP” versus “LBP” isolates at 3 and 24 h post-infection. Significant differentially expressed genes were defined as genes showing with |log2FC| > 1. For assessing biological pathway enrichment, we used fast geneset enrichment analysis (fgsea) on preranked gene lists (ranks = log2FC for BP-LBP) for both 3 and 24 h. Gene sets included MSigDB Hallmark (h), Canonical pathways (C2), Gene Ontology (C5), and a curated macrophage M1 versus M2 polarization geneset [71,75,76]; GSEA was run with minSize = 15 and maxSize = 500. Enrichment curves and top pathways were plotted directly from plotEnrichment. To interrogate biology of specific pathways (e.g., HALLMARK_IL6_JAK_STAT3_SIGNALING, GOBP_GRANULOCYTE_MIGRATION), leading-edge (LE) genes were extracted from the fgsea results and their group means (LBP STRAINS versus BP) at the target time point were computed from log2(CPM + 1). For visualization, heatmaps were generated with ComplexHeatmap, where rows were ordered by contribution to the observed Normalized Enrichment Score –NES (using the NES sign and the BP-LBP ranks), columns were fixed as LBP or BP (without clustering), and colors reflected row Z-scores to emphasize relative induction. Alignment and variant Calling Short sequencing reads were trimmed and filtered using Skewer v0.2.2 using parameters “-q 30 -l 35” [77]. The resulting trimmed reads were assessed for quality by FastQC (https://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Reads were mapped onto the C. parapsilosis CDC317 reference genome (GCA_000182765.2), using the Burrow–Wheeler Aligner (BWA-MEM) algorithm version 0.7.17 [78]. The aligned reads were marked for duplicates using the GATK version 4.2.0.0 MarkDuplicates tool and reindexed using GATK’s BuildBamIndex tool [79]. To produce highly accurate variants, thee separate variant calling tools were used in parallel and the results combined. Variants were called using bcftools mpileup version 1.10.2 [80], and FreeBayes version 1.3.9 [81] with standard parameters. Variants were also called with GATK using the tools HaplotypeCaller, CombineGVCFs and GenotypeGVCFs. The thee resulting files were merged using bcftools isec [79], keeping only variants that were called by at least two of the thee tools. Variants were filtered using GATK VariantFiltration [79] to remove variants with genotype quality <70 and read depth <20. Additionally, SNPs that were present in clusters of 5 or more within 20 bp were removed as artifacts, and only biallelic variants were kept. Copy number variation analysis Copy number was estimated for all C. parapsilosis genes using the short-read alignments. The mean coverage across the gene’s open reading frame was found using BEDTools coverage version 2.29.2 [82]. This was normalized by dividing by the modal genome coverage as found using BEDTools genomecov [82] and then multiplying by two to account for ploidy. The data were filtered to keep only genes that had an abnormal copy number, i.e., greater than 2.75× or less than 1.25×, in at least one strain. Chromosome ploidy was used by calculating mean coverage across the entire chromosomes and normalized as above. Functional analysis SnpEff was used to annotate the final set of variants for their predicted effect [83]. A subset of variants including only those likely to affect protein function (missense mutations, nonsense mutations, frameshifts, etc.) as input for functional GO analysis. GO analysis was performed using the Candida Genome Database website (http://www.candidagenome.org/). Orthologs of C. parapsilosis genes in Candida albicans and Saccharomyces cerevisiae were identified using CGOB [84]. Protein domains and families were predicted using InterPro [85]. Phylogenetics For phylogenetic analysis, indels were removed to keep only SNPs. A custom script was used (https://github.com/CMOTsean/HetSiteRando) to create a FASTA alignment of all sites in the genome that had a SNP in at least one sample. Heterozygous sites were randomly resolved to either the reference or alternate allele. 12,199 SNPs were used to construct the phylogeny of strains sequenced in this study, and a total of 33,534 SNPs were used in the alignment for the phylogeny including strains from other studies. RAxML version 8.2.12 [86] was used to construct a tree from this FASTA alignment using the GTRGAMMA model of nucleotide substitution with 1,000 bootstrap replicates. Trees were visualized using the iTOL website [87] and manually annotated. Whole-genome sequencing DNA was fragmented to sizes between 1 and 20 kb using a transposase that binds biotinylated adapters at the breaking point. Strand displacement was performed to “repair” the nicks left by the transposase. Fragment sizes of 3–6 kb were then selected on a 0.8% agarose gel and were then circularized. Non-circularized DNA was removed by digestion. The circular DNA was then mechanically sheared to fragments of 100 bp to 1 kb approx. and the fragments containing the biotinylated ends were pulled down using magnetic streptavidin beads and submitted to a standard library preparation. A final size selection on 2% agarose gel was done and fragments of 400–700 bp were selected for the final library. Final libraries were analyzed using Agilent High Sensitivity chip to estimate the quantity and check size distribution and were then quantified by qPCR using the KAPA Library Quantification Kit (ref. KK4835, KapaBiosystems) prior to amplification with Illumina’s cBot. Libraries were sequenced 2 × 150 bp on an Illumina’s NovaSeq instrument. Raw sequencing reads have been deposited at Short Read Archive (PRJNA1363747). Systemic infection mouse models Systemic infection mouse models were carried out according to the previously established protocol (IACUC protocol #2017N000058) [11]. Briefly, mice were infected with 2*107 CFU of each isolate using tail vein injection and sacrificed on designated timepoints, followed by harvesting and homogenizing organs and plating on YPD agar plates for CFU counting. The CFUs obtained from each organ were normalized against the CFUs of the pertinent initial inoculum used to infect mice. For experiments involving antifungal treatment, mice received desired antifungal drugs 24 h post-infection and treated every other day. CFUs of organs at desired time point were normalized against the pertinent untreated control mice and results were presented as percentages. Statistical analysis SPSS software (v.24 for Windows; SPSS, Chicago, IL, USA) was used in our statistical analysis and statistical significance was designated for P-values ≤0.05. We used Shapiro–Wilk test to determine data distribution and Wilcoxon test was used to determine statistical significance of nonparametric data. Supporting information S1 Fig. Crystal violet assay was positively correlated with standard biofilm quantification methods. (OD600) (A) and confocal scanning laser microscopy (B, C). PCA (D) and GO terminology enrichment analysis (E). Upper figure shows the biofilm level of 59T and 60T (fluconazole-resistant, carrying Erg11Y132F, and low biofilm producer) and revertant colonies (fluconazole susceptible and carrying wild-type Erg11F132Y) determined by crystal violet (OD490) and lower figure displays the fluconazole susceptibility pattern (µg/ml) of those isolates (F). The data underlying this S1D and S1E Fig can be found in https://doi.org/10.5281/zenodo.21799493 and the rest in the S1 Data File. https://doi.org/10.1371/journal.pbio.3003973.s001 (TIFF) S2 Fig. Mutations exclusive to low biofilm producers (LBP) were enriched for genes belonging to diverse biological functions. (A) Introduction of Mnl1S557T into a Turkish LBP (1T) decreased the biofilm formation (B), but its reversion of Mnl1S557T to WT did not restore the biofilm defect (C). Whole-genome sequence analysis of global collection of C. parapsilosis isolates identified various MNL1 mutations in numerous C. parapsilosis isolates from a global collection (D). Flow cytometry gating strategies using mTorquise and YFP expressing C. parapsilosis isolates (E). Flow cytometry gating strategies used to measure the reactive oxygen species (ROS) of macrophages (F) infected with C. parapsilosis. Gating strategy was drawn based on the uninfected control macrophages and mean fluorescence intensity was recorded. Macrophages infected with biofilm producers and LBPs showed similar intracellular growth rates at 24 h post-infection (G). The intracellular colony-forming units (CFUs) at each time point were normalized against the average of pertinent 1 h intracellular CFUs. The data underlying S2E and S2F Fig can be found in https://doi.org/10.5281/zenodo.21794379 and the rest in the S1 Data File. https://doi.org/10.1371/journal.pbio.3003973.s002 (TIFF) S3 Fig. Flow cytometry gating strategies used to determine the neutrophil effector functions against C. parapsilosis isolates. Low biofilm producer C. parapsilosis isolates showed a significantly lower extracellular colony-forming units (CFUs) relative to biofilm producers at 24 h post-infection of macrophages (A). Flow cytometry gating strategies used to measure the degranulation (B) and shedding (C) of primary human neutrophils during interaction with C. parapsilosis. Flow cytometry gating strategies used to measure the phagocytosis (D) and reactive oxygen species (E) of primary human neutrophils during interaction with C. parapsilosis. Gating strategy was drawn based on the uninfected control neutrophils (except phagocytosis) and mean fluorescence intensity was recorded. The data underlying figure can be found in https://doi.org/10.5281/zenodo.21794379 and the rest in the S1 Data File. https://doi.org/10.1371/journal.pbio.3003973.s003 (TIFF) S4 Fig. Multidimensional assessment of host-C. parapsilosis interface. Gene ontology enrichment analysis of differentially expressed pathways of macrophages infected with C. parapsilosis isolates. Upregulated (A) and downregulated (B) pathways of macrophages infected with C. parapsilosis isolates. Representative confocal images of neutrophils (stained with Hoescht) during interaction with iRFP-expressing C. albicans (upper) and without neutrophils (lower) (A). C. albicans survival was measured by measuring the MFI of iRFP and scale bar represents 100 µm (C). The mutants carrying Efg1I280T (D) and Mnl1S557T (E) did not show any growth advantage compared to their pertinent wildtype strains under various in vitro conditions. Biofilm analysis of C. parapsilosis mutants lacking transcription factors/kinase involved in biofilm regulation using OD600 (F). Fig G–J show growth rate of wildtype (ATCC 22,019) and the mutants derived from it under various in vitro conditions. Note that all in-vitro and biofilm experiments tested at two-independent colonies of the same mutant for rigor. Since such in vitro experiments showed consistent phenotype among independent colonies, only a single colony was used for comprehensive ex vivo analysis shown in Fig 6. The data underlying this S4A and S4B Fig can be found in https://doi.org/10.5281/zenodo.21799493 and the rest in the S1 Data File. https://doi.org/10.1371/journal.pbio.3003973.s004 (TIFF) S1 Table. Complete microbiological data of C. parapsilosis isolates collected from outbreaks in Turkey and Brazil denoted by T and B, respectively. https://doi.org/10.1371/journal.pbio.3003973.s005 (XLSX) S2 Table. Differential expression of C. parapsilosis genes in (A) biofilm, and (B) planktonic growth conditions. Differential expression analysis was performed by comparing gene expression BP strains (combining 35B and 1T) relative to LBPs (combining 49B and 29T) at each time point and log2 fold change is denoted here. (C) Gene age was estimated using orthologous gene family estimates across 15 species. “Old” genes are those conserved across Ascomycota species, “Middle” genes are conserved across CTG and non-CTG Candida species and S. cerevisiae, “Young” gene are conserved across CTG Candida species and “Unique” genes are those unique to C. parapsilosis. https://doi.org/10.1371/journal.pbio.3003973.s006 (XLSX) S3 Table. Filtered list of all variants predicted to affect protein function, such as missense variants, frameshifts, and changes to start and stop codons. Each strain is marked with a 2 if they are homozygous for the variant, a 1 if they are heterozygous for the variants, and 0 if they are homozygous for the reference. Sheet (ii) lists variants unique to BP strains, i.e., not found in LBP strains. Sheets (iii)–(vi) contain the variants which are unique to LBP strains. Sheet (vii) lists the genes that are only mutated in LBP strains, and sheet (viii) summarizes the type and number of mutations in LBP and BP strains. https://doi.org/10.1371/journal.pbio.3003973.s007 (XLSX) S4 Table. Heterozygous variants from 35B that underwent LOH in derived strains, in tabular variant call format (VCF), annotated by SIFT. Heterozygous sites are highlighted in yellow, homozygous reference in green, and homozygous alternate in red. https://doi.org/10.1371/journal.pbio.3003973.s008 (XLSX) S5 Table. Gene ontology enrichment analysis of genes containing premature stop codons or frameshift mutations in low biofilm producers compared to high biofilm producers. https://doi.org/10.1371/journal.pbio.3003973.s009 (XLSX) S6 Table. Gene ontology (GO) analysis of macrophages infected with high and low biofilm producer isolates at 3 h post-infection. https://doi.org/10.1371/journal.pbio.3003973.s010 (CSV) S7 Table. Gene ontology (GO) analysis of macrophages infected with high and low biofilm producer isolates at 24 h post-infection. https://doi.org/10.1371/journal.pbio.3003973.s011 (CSV) S8 Table. Gene set enrichment analysis (GSEA) of macrophages infected with BP strains demonstrated “Hallmark IL-6/JAK/STAT3”. https://doi.org/10.1371/journal.pbio.3003973.s012 (CSV) S9 Table. Gene set enrichment analysis (GSEA) of macrophages infected with BP strains exhibited robust induction of chemokines associated with granulocyte recruitment. https://doi.org/10.1371/journal.pbio.3003973.s013 (CSV) S10 Table. Macrophages infected with LBP strains displayed a transcriptional profile enriched for genes classically associated with proinflammatory (M1-like) activation at 24 h post-infection. https://doi.org/10.1371/journal.pbio.3003973.s014 (CSV) S11 Table. Sequence details of oligonucleotides used for manipulation of C. parapsilosis isolates using CRISPR-Cas9. https://doi.org/10.1371/journal.pbio.3003973.s015 (XLSX) S1 Data. Complete dataset used to generate figures in the manuscript. Partial FACS files could be accessed using the following link, https://doi.org/10.5281/zenodo.21794379. Data concerning Figs 1D, 1E, S1D, and S1E are deposited in public depository under the following link https://doi.org/10.5281/zenodo.21799493. https://doi.org/10.1371/journal.pbio.3003973.s016 (XLSX) Acknowledgments We thank Marina Marcet-Houben for assisting in storage, transfer and public deposition of sequencing data. Confocal microscopy data were collected in part using instrumentation acquired through NSF MRI award DMR-1625733 and additional microscopes supported by the Imaging and Microscopy Facility (IMF) at the University of California, Merced References - 1. Benedict K, Whitham HK, Jackson BR. Economic burden of fungal diseases in the United States. Open Forum Infect Dis. 2022;9(4):ofac097. pmid:35350173 - 2. Denning DW. Global incidence and mortality of severe fungal disease. Lancet Infect Dis. 2024;24(7):e428–38. pmid:38224705 - 3. Tóth R, Nosek J, Mora-Montes HM, Gabaldon T, Bliss JM, Nosanchuk JD, et al. Candida parapsilosis: from genes to the bedside. Clin Microbiol Rev. 2019;32(2):e00111-18. pmid:30814115 - 4. Daneshnia F, de Almeida Júnior JN, Ilkit M, Lombardi L, Perry AM, Gao M, et al. Worldwide emergence of fluconazole-resistant Candida parapsilosis: current framework and future research roadmap. Lancet Microbe. 2023;4(6):e470–80. pmid:37121240 - 5. Gabaldón T. Threats from the Candida parapsilosis complex: the surge of multidrug resistance and a hotbed for new emerging pathogens. Microbiol Mol Biol Rev. 2024;88(4):e0002923. pmid:39508581 - 6. Govrins M, Lass-Flörl C. Candida parapsilosis complex in the clinical setting. Nat Rev Microbiol. 2024;22(1):46–59. pmid:37674021 - 7. McTaggart LR, Eshaghi A, Hota S, Poutanen SM, Johnstone J, De Luca DG, et al. First Canadian report of transmission of fluconazole-resistant Candida parapsilosis within two hospital networks confirmed by genomic analysis. J Clin Microbiol. 2024;62(1):e0116123. pmid:38112529 - 8. Misas E, Witt LS, Farley MM, Thomas S, Jenkins EN, Gade L, et al. Molecular and epidemiological investigation of fluconazole-resistant Candida parapsilosis-Georgia, United States, 2021. Open Forum Infect Dis. 2024;11:ofae264. - 9. Arastehfar A, Hilmioğlu-Polat S, Daneshnia F, Pan W, Hafez A, Fang W, et al. Clonal candidemia outbreak by Candida parapsilosis carrying Y132F in Turkey: evolution of a persisting challenge. Front Cell Infect Microbiol. 2021;11:676177. pmid:33968809 - 10. Thomaz DY, de Almeida JN Jr, Sejas ONE, Del Negro GMB, Carvalho GOMH, Gimenes VMF, et al. Environmental clonal spread of azole-resistant Candida parapsilosis with Erg11-Y132F mutation causing a large candidemia outbreak in a Brazilian Cancer Referral Center. JoF. 2021;7(4):259. - 11. Daneshnia F, Floyd DJ, Ryan AP, Ghahfarokhy PM, Ebadati A, Jusuf S, et al. Evaluation of outbreak persistence caused by multidrug-resistant and echinocandin-resistant Candida parapsilosis using multidimensional experimental and epidemiological approaches. Emerg Microbes Infect. 2024;13(1):2322655. pmid:38380673 - 12. Lohse MB, Gulati M, Johnson AD, Nobile CJ. Development and regulation of single- and multi-species Candida albicans biofilms. Nat Rev Microbiol. 2018;16(1):19–31. pmid:29062072 - 13. Ramage G, Kean R, Rautemaa-Richardson R, Williams C, Lopez-Ribot JL. Fungal biofilms in human health and disease. Nat Rev Microbiol. 2025;23(6):355–70. pmid:39910237 - 14. Ribeiro F de C, Kemmerich KK, Gonçale JC, Junqueira JC, Mannan M, Nabeela S, et al. Candida albicans recovered from persistent candidemia exhibits enhanced virulence traits. J Infect Dis. 2025;231(4):e803–12. pmid:39693248 - 15. Rajendran R, Sherry L, Nile CJ, Sherriff A, Johnson EM, Hanson MF, et al. Biofilm formation is a risk factor for mortality in patients with Candida albicans bloodstream infection-Scotland, 2012-2013. Clin Microbiol Infect. 2016;22(1):87–93. pmid:26432192 - 16. Tumbarello M, Posteraro B, Trecarichi EM, Fiori B, Rossi M, Porta R, et al. Biofilm production by Candida species and inadequate antifungal therapy as predictors of mortality for patients with candidemia. J Clin Microbiol. 2007;45(6):1843–50. pmid:17460052 - 17. Brassington PJT, Klefisch F-R, Graf B, Pfüller R, Kurzai O, Walther G, et al. Genomic reconstruction of an azole-resistant Candida parapsilosis outbreak and the creation of a multi-locus sequence typing scheme: a retrospective observational and genomic epidemiology study. Lancet Microbe. 2025;6(1):100949. pmid:39557054 - 18. Gnaien M, Maufrais C, Rebai Y, Kallel A, Ma L, Hamouda S, et al. A gain-of-function mutation in zinc cluster transcription factor Rob1 drives Candida albicans adaptive growth in the cystic fibrosis lung environment. PLoS Pathog. 2024;20(4):e1012154. pmid:38603707 - 19. Tso GHW, Reales-Calderon JA, Tan ASM, Sem X, Le GTT, Tan TG, et al. Experimental evolution of a fungal pathogen into a gut symbiont. Science. 2018;362(6414):589–95. pmid:30385579 - 20. Ünal N, Spruijtenburg B, Arastehfar A, Gümral R, de Groot T, Meijer EFJ, et al. Multicentre study of Candida parapsilosis blood isolates in Türkiye highlights an increasing rate of fluconazole resistance and emergence of echinocandin and multidrug resistance. Mycoses. 2024;67(11):e70000. pmid:39547949 - 21. Arastehfar A, Daneshnia F, Hilmioğlu-Polat S, Fang W, Yaşar M, Polat F, et al. First report of candidemia clonal outbreak caused by emerging fluconazole-resistant Candida parapsilosis isolates harboring Y132F and/or Y132F K143R in Turkey. Antimicrob Agents Chemother. 2020. - 22. Mancera E, Nocedal I, Hammel S, Gulati M, Mitchell KF, Andes DR, et al. Evolution of the complex transcription network controlling biofilm formation in Candida species. Elife. 2021;10:e64682. pmid:33825680 - 23. Daneshnia F, de Almeida Júnior JN, Arastehfar A, Lombardi L, Shor E, Moreno L, et al. Determinants of fluconazole resistance and echinocandin tolerance in C. parapsilosis isolates causing a large clonal candidemia outbreak among COVID-19 patients in a Brazilian ICU. Emerg Microbes Infect. 2022;11(1):2264–74. pmid:36066554 - 24. Caggiano G, Fioriti S, Morroni G, Apollonio F, Triggiano F, D’Achille G, et al. Genotypic and phenotypic characteristics of Candida parapsilosis bloodstream isolates: Health Care Associated Infections in a teaching Hospital in Italy. J Infect Public Health. 2024;17(6):967–74. pmid:38631066 - 25. Torres-Cano A, de Armentia C, Roldán A, López-Peralta E, Manosalva J, Merino-Amador P, et al. Resistance to azoles in Candida parapsilosis isolates from Spain is associated with an impairment in filamentation and biofilm formation. J Fungi (Basel). 2025;11(4):299. pmid:40278120 - 26. Rupert CB, Rusche LN. The pathogenic yeast Candida parapsilosis forms pseudohyphae through different signaling pathways depending on the available carbon source. mSphere. 2022;7:e0002922. - 27. Holland LM, Schröder MS, Turner SA, Taff H, Andes D, Grózer Z, et al. Comparative phenotypic analysis of the major fungal pathogens Candida parapsilosis and Candida albicans. PLoS Pathog. 2014;10(9):e1004365. pmid:25233198 - 28. Nobile CJ, Fox EP, Nett JE, Sorrells TR, Mitrovich QM, Hernday AD, et al. A recently evolved transcriptional network controls biofilm development in Candida albicans. Cell. 2012;148(1–2):126–38. pmid:22265407 - 29. Ryan AP, Bergin S, Scully J, Zuniga-Soto E, Hession C, Barber AE, et al. Small pangenome of Candida parapsilosis reflects overall low intraspecific diversity. mBio. 2025;16(10):e0132025. pmid:40852993 - 30. Lockhart SR, Etienne KA, Vallabhaneni S, Farooqi J, Chowdhary A, Govender NP, et al. Simultaneous emergence of multidrug-resistant Candida auris on 3 continents confirmed by whole-genome sequencing and epidemiological analyses. Clin Infect Dis. 2017;64(2):134–40. pmid:27988485 - 31. Dunker C, Polke M, Schulze-Richter B, Schubert K, Rudolphi S, Gressler AE, et al. Rapid proliferation due to better metabolic adaptation results in full virulence of a filament-deficient Candida albicans strain. Nat Commun. 2021;12(1):3899. pmid:34162849 - 32. Fróis-Martins R, Lagler J, Schille TB, Elshafee O, Martinez de San Vicente K, Mertens S, et al. Dynamic expression of candidalysin facilitates oral colonization of Candida albicans in mice. Nat Microbiol. 2025. - 33. Kasper L, Seider K, Hube B. Intracellular survival of Candida glabrata in macrophages: immune evasion and persistence. FEMS Yeast Res. 2015;15(5):fov042. pmid:26066553 - 34. Case NT, Westman J, Hallett MT, Plumb J, Farheen A, Maxson ME, et al. Respiration supports intraphagosomal filamentation and escape of Candida albicans from macrophages. mBio. 2023;14(6):e0274523. pmid:38038475 - 35. Sonnberger J, Kasper L, Lange T, Brunke S, Hube B. “We’ve got to get out”-Strategies of human pathogenic fungi to escape from phagocytes. Mol Microbiol. 2024;121(3):341–58. pmid:37800630 - 36. Zavrel M, Majer O, Kuchler K, Rupp S. Transcription factor Efg1 shows a haploinsufficiency phenotype in modulating the cell wall architecture and immunogenicity of Candida albicans. Eukaryot Cell. 2012;11(2):129–40. pmid:22140230 - 37. Sohn K, Urban C, Brunner H, Rupp S. EFG1 is a major regulator of cell wall dynamics in Candida albicans as revealed by DNA microarrays. Mol Microbiol. 2003;47(1):89–102. pmid:12492856 - 38. Gow NAR. Fungal cell wall biogenesis: structural complexity, regulation and inhibition. Fungal Genet Biol. 2025;179:103991. pmid:40334812 - 39. McClelland M, Sanderson KE, Clifton SW, Latreille P, Porwollik S, Sabo A, et al. Comparison of genome degradation in Paratyphi A and Typhi, human-restricted serovars of Salmonella enterica that cause typhoid. Nat Genet. 2004;36(12):1268–74. pmid:15531882 - 40. Wang BX, Leshchiner D, Luo L, Tuncel M, Hokamp K, Hinton JCD, et al. High-throughput fitness experiments reveal specific vulnerabilities of human-adapted Salmonella during stress and infection. Nat Genet. 2024;56(6):1288–99. pmid:38831009 - 41. Muñoz JF, Delorey T, Ford CB, Li BY, Thompson DA, Rao RP, et al. Coordinated host-pathogen transcriptional dynamics revealed using sorted subpopulations and single macrophages infected with Candida albicans. Nat Commun. 2019;10(1):1607. pmid:30962448 - 42. Murray PJ. Macrophage polarization. Annu Rev Physiol. 2017;79:541–66. - 43. Hopke A, Scherer A, Kreuzburg S, Abers MS, Zerbe CS, Dinauer MC, et al. Neutrophil swarming delays the growth of clusters of pathogenic fungi. Nat Commun. 2020;11(1):2031. pmid:32341348 - 44. MacKenzie KD, Palmer MB, Köster WL, White AP. Examining the link between biofilm formation and the ability of pathogenic salmonella strains to colonize multiple host species. Front Vet Sci. 2017;4:138. pmid:29159172 - 45. MacKenzie KD, Wang Y, Musicha P, Hansen EG, Palmer MB, Herman DJ, et al. Parallel evolution leading to impaired biofilm formation in invasive Salmonella strains. PLoS Genet. 2019;15(6):e1008233. pmid:31233504 - 46. Ma H, Hagen F, Stekel DJ, Johnston SA, Sionov E, Falk R, et al. The fatal fungal outbreak on Vancouver Island is characterized by enhanced intracellular parasitism driven by mitochondrial regulation. Proc Natl Acad Sci U S A. 2009;106(31):12980–5. pmid:19651610 - 47. Rapsinski GJ, Wynosky-Dolfi MA, Oppong GO, Tursi SA, Wilson RP, Brodsky IE, et al. Toll-like receptor 2 and NLRP3 cooperate to recognize a functional bacterial amyloid, curli. Infect Immun. 2015;83(2):693–701. pmid:25422268 - 48. Tükel C, Nishimori JH, Wilson RP, Winter MG, Keestra AM, van Putten JPM, et al. Toll-like receptors 1 and 2 cooperatively mediate immune responses to curli, a common amyloid from enterobacterial biofilms. Cell Microbiol. 2010;12(10):1495–505. pmid:20497180 - 49. Tükel C, Raffatellu M, Humphries AD, Wilson RP, Andrews-Polymenis HL, Gull T, et al. CsgA is a pathogen-associated molecular pattern of Salmonella enterica serotype Typhimurium that is recognized by Toll-like receptor 2. Mol Microbiol. 2005;58(1):289–304. pmid:16164566 - 50. MacKenzie KD, Wang Y, Shivak DJ, Wong CS, Hoffman LJL, Lam S, et al. Bistable expression of CsgD in Salmonella enterica serovar Typhimurium connects virulence to persistence. Infect Immun. 2015;83(6):2312–26. pmid:25824832 - 51. Ahmad I, Rouf SF, Sun L, Cimdins A, Shafeeq S, Le Guyon S, et al. BcsZ inhibits biofilm phenotypes and promotes virulence by blocking cellulose production in Salmonella enterica serovar Typhimurium. Microb Cell Fact. 2016;15(1):177. pmid:27756305 - 52. Holt KE, Parkhill J, Mazzoni CJ, Roumagnac P, Weill F-X, Goodhead I, et al. High-throughput sequencing provides insights into genome variation and evolution in Salmonella Typhi. Nat Genet. 2008;40(8):987–93. pmid:18660809 - 53. Gunn JS, Marshall JM, Baker S, Dongol S, Charles RC, Ryan ET. Salmonella chronic carriage: epidemiology, diagnosis, and gallbladder persistence. Trends Microbiol. 2014;22(11):648–55. pmid:25065707 - 54. Crawford RW, Rosales-Reyes R, Ramírez-Aguilar M de la L, Chapa-Azuela O, Alpuche-Aranda C, Gunn JS. Gallstones play a significant role in Salmonella spp. gallbladder colonization and carriage. Proc Natl Acad Sci U S A. 2010;107(9):4353–8. pmid:20176950 - 55. Taff HT, Nett JE, Andes DR. Comparative analysis of Candida biofilm quantitation assays. Med Mycol. 2012;50(2):214–8. pmid:21539503 - 56. Gulati M, Lohse MB, Ennis CL, Gonzalez RE, Perry AM, Bapat P, et al. In vitro culturing and screening of Candida albicans biofilms. Curr Protoc Microbiol. 2018;50(1):e60. pmid:29995344 - 57. Lombardi L, Oliveira-Pacheco J, Butler G. Plasmid-based CRISPR-Cas9 gene editing in multiple Candida species. mSphere. 2019;4. - 58. Nemeth T, Zarnocki A, Ladanyi A, Papp C, Ayaydin F, Szebeni GJ, et al. PCR-based CRISPR/Cas9 system for fluorescent tagging: a tool for studying Candida parapsilosis virulence. PLoS One. 2025;20(2):e0312948. pmid:39992908 - 59. Arastehfar A, Daneshnia F, Floyd DJ, Jeffries NE, Salehi M, Perlin DS, et al. Echinocandin persistence directly impacts the evolution of resistance and survival of the pathogenic fungus Candida glabrata. mBio. 2024;15(4):e0007224. pmid:38501869 - 60. Antunes D, Gonçalves SM, Matzaraki V, Rodrigues CS, Gonçales RA, Rocha J, et al. Glutamine metabolism supports the functional activity of immune cells against Aspergillus fumigatus. Microbiol Spectr. 2023;11(1):e0225622. pmid:36475892 - 61. Pekmezovic M, Hovhannisyan H, Gresnigt MS, Iracane E, Oliveira-Pacheco J, Siscar-Lewin S, et al. Candida pathogens induce protective mitochondria-associated type I interferon signalling and a damage-driven response in vaginal epithelial cells. Nat Microbiol. 2021;6(5):643–57. pmid:33753919 - 62. Timmer KD, Floyd DJ, Scherer AK, Crossen AJ, Atallah J, Viens AL, et al. Multiparametric profiling of neutrophil function via a high-throughput flow cytometry-based assay. Cells. 2023;12(5):743. pmid:36899878 - 63. Arastehfar A, Daneshnia F, Hovhannisyan H, Fuentes D, Cabrera N, Quinteros C, et al. Overlooked Candida glabrata petites are echinocandin tolerant, induce host inflammatory responses, and display poor in vivo fitness. mBio. 2023;14(5):e0118023. pmid:37772846 - 64. Wingett SW, Andrews S. FastQ Screen: a tool for multi-genome mapping and quality control. F1000Res. 2018;7:1338. pmid:30254741 - 65. Lew-Smith J, Binkley J, Sherlock G. The Candida Genome Database: annotation and visualization updates. Genetics. 2025;229(3):iyaf001. pmid:39776186 - 66. Liao Y, Smyth GK, Shi W. The Subread aligner: fast, accurate and scalable read mapping by seed-and-vote. Nucleic Acids Res. 2013;41(10):e108. pmid:23558742 - 67. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550. pmid:25516281 - 68. Ignatiadis N, Klaus B, Zaugg JB, Huber W. Data-driven hypothesis weighting increases detection power in genome-scale multiple testing. Nat Methods. 2016;13(7):577–80. pmid:27240256 - 69. Zhu A, Ibrahim JG, Love MI. Heavy-tailed prior distributions for sequence count data: removing the noise and preserving large differences. Bioinformatics. 2019;35(12):2084–92. pmid:30395178 - 70. Rosiana S, Zhang L, Kim GH, Revtovich AV, Uthayakumar D, Sukumaran A, et al. Comprehensive genetic analysis of adhesin proteins and their role in virulence of Candida albicans. Genetics. 2021;217(2):iyab003. pmid:33724419 - 71. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A. 2005;102(43):15545–50. pmid:16199517 - 72. Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation (Camb). 2021;2(3):100141. pmid:34557778 - 73. Emms DM, Kelly S. OrthoFinder: phylogenetic orthology inference for comparative genomics. Genome Biol. 2019;20(1):238. pmid:31727128 - 74. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010;26(1):139–40. pmid:19910308 - 75. Korotkevich G, Sukhov V, Budin N, Shpak B, Artyomov MN, Sergushichev A. Fast gene set enrichment analysis. bioRxiv. 2021;60012. - 76. Xue J, Schmidt SV, Sander J, Draffehn A, Krebs W, Quester I, et al. Transcriptome-based network analysis reveals a spectrum model of human macrophage activation. Immunity. 2014;40(2):274–88. pmid:24530056 - 77. Jiang H, Lei R, Ding S-W, Zhu S. Skewer: a fast and accurate adapter trimmer for next-generation sequencing paired-end reads. BMC Bioinformatics. 2014;15:182. pmid:24925680 - 78. He L. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. arXiv V2. 2013. Available from: https://arxiv.org/abs/1303.3997 - 79. McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 2010;20(9):1297–303. pmid:20644199 - 80. Danecek P, Bonfield JK, Liddle J, Marshall J, Ohan V, Pollard MO, et al. Twelve years of SAMtools and BCFtools. Gigascience. 2021;10(2):giab008. pmid:33590861 - 81. Richter F, Morton SU, Qi H, Kitaygorodsky A, Wang J, Homsy J, et al. Whole genome de novo variant identification with freebayes and neural network approaches. bioRxiv. 2020. - 82. Quinlan AR, Hall IM. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics. 2010;26(6):841–2. pmid:20110278 - 83. Cingolani P, Platts A, Wang LL, Coon M, Nguyen T, Wang L, et al. A program for annotating and predicting the effects of single nucleotide polymorphisms, SnpEff: SNPs in the genome of Drosophila melanogaster strain w1118; iso-2; iso-3. Fly (Austin). 2012;6(2):80–92. pmid:22728672 - 84. Maguire SL, ÓhÉigeartaigh SS, Byrne KP, Schröder MS, O’Gaora P, Wolfe KH, et al. Comparative genome analysis and gene finding in Candida species using CGOB. Mol Biol Evol. 2013;30(6):1281–91. pmid:23486613 - 85. Blum M, Chang H-Y, Chuguransky S, Grego T, Kandasaamy S, Mitchell A, et al. The InterPro protein families and domains database: 20 years on. Nucleic Acids Res. 2021;49(D1):D344–54. pmid:33156333 - 86. Stamatakis A. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics. 2014;30(9):1312–3. pmid:24451623 - 87. Letunic I, Bork P. Interactive Tree Of Life (iTOL) v5: an online tool for phylogenetic tree display and annotation. Nucleic Acids Res. 2021;49(W1):W293–6. pmid:33885785