(C) PLOS One
This story was originally published by PLOS One and is unaltered.
 . . . . . . . . . .



Transcriptomic atlas throughout Coccidioides development reveals key phase-enriched transcripts of this important fungal pathogen [1]

['Christina M. Homer', 'Division Of Infectious Diseases', 'University Of California San Francisco', 'San Francisco', 'California', 'United States Of America', 'Mark Voorhies', 'Department Of Microbiology', 'Immunology', 'Keith Walcott']

Date: 2025-04 

Coccidioides spp. are highly understudied but significant dimorphic fungal pathogens that can infect both immunocompetent and immunocompromised people. In the environment, they grow as multicellular filaments (hyphae) that produce vegetative spores called arthroconidia. Upon inhalation by mammals, arthroconidia undergo a process called spherulation. They enlarge and undergo numerous nuclear divisions to form a spherical structure, and then internally segment until the spherule is filled with multiple cells called endospores. Mature spherules rupture and release endospores, each of which can form another spherule, in a process thought to facilitate dissemination. Spherulation is unique to Coccidioides, and its molecular determinants remain largely unknown. Here, we report the first high-density transcriptomic analyses of Coccidioides development, defining morphology-dependent transcripts and those whose expression is regulated by RYP1, a major regulator required for spherulation and virulence. Of approximately 9,000 predicted transcripts, we discovered 273 transcripts with consistent spherule-associated expression, 82 of which are RYP1-dependent, a set likely to be critical for Coccidioides virulence. ChIP-Seq revealed two distinct regulons of RYP1: one shared between hyphae and spherules and the other unique to spherules. Spherulation regulation was elaborate, with the majority of 227 predicted transcription factors in Coccidioides displaying spherule-enriched expression. We identified provocative targets, including 20 transcripts whose expression is endospore-enriched and 14 putative secreted effectors whose expression is spherule-enriched, of which six are secreted proteases. To highlight the utility of these data, we selected a cluster of RYP1-dependent, arthroconidia-associated transcripts and found that they play a role in arthroconidia cell wall biology, demonstrating the power of this resource in illuminating Coccidioides biology and virulence.

Funding: This research was supported by the HHMI Hanna Gray Fellowship (to CH), the Program for Breakthrough Biomedical Research, which is partially funded by the Sandler Foundation (to CH), NIH R21AI172185 (to AS), NIH 5R01AI146584 (to AS), and NIH U19AI166798 (to AS) for funding. AS is a Chan Zuckerberg Biohub – San Francisco Investigator. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Here, we performed the first high-depth, high-density transcriptomic time courses of Coccidioides arthroconidia germinating into either hyphae or spherules that went on to release endospores. We leveraged the ryp1∆ mutant to define genes whose transcription is regulated by RYP1 throughout these developmental trajectories, defined morphology-specific binding targets of RYP1, and highlighted a particular role for RYP1 in direct regulation of genes expressed in the spherule morphology. Additionally, we annotated TFs, identified candidate-secreted effectors, and defined candidate endospore-associated genes. From these data, we selected a cluster of spore-associated genes that were RYP1-dependent and found that they play a role in arthroconidia cell wall biology, demonstrating the power of this transcriptomic atlas to uncover new biology. Together, these findings serve as a foundational resource for the study of this important fungal pathogen.

Despite limited molecular insight into spherulation, the RYP1 transcription factor (TF) is known to be a major spherulation regulator [ 12 ]. RYP1 is a WOPR-domain containing TF whose orthologs (such as Wor1) in other fungi are regulators of morphology transitions and development [ 13 – 16 ]. Additionally, WOPR family proteins often regulate virulence factors [ 15 – 17 ] and are required for virulence in fungal pathogens [ 12 , 18 – 21 ]. In Coccidioides, the ryp1∆ mutant is unable to form spherules and has an aberrant transcriptome in both spherule and hyphal conditions [ 12 ].

Coccidioides spp. are dimorphic fungal pathogens found in the soil in the Southwest United States and other desert regions in Central and South America [ 1 ]. In the soil, they grow as hyphae that generate vegetative spores known as arthroconidia. Upon inhalation by a mammalian host, arthroconidia germinate and form a unique host-associated morphology known as the spherule [ 2 ]. Mature spherules rupture, releasing hundreds of internal cells known as endospores which can each go on to form another spherule in a cycle called spherulation. Notably, Coccidioides can cause infection in immunocompetent and immunocompromised individuals [ 3 ]. There is currently no cure for serious disseminated infections [ 4 , 5 ]. Efforts to develop new treatments and prevention strategies have been hindered by a lack of molecular knowledge of the host form of Coccidioides, the spherule, including sparse sampling of the transcriptome during Coccidioides development. Prior studies have relied on microarray or low replicate number RNA-Seq at one or two time points during spherule formation, in different conditions varying by laboratory, using different media to induce spherules versus hyphae, and only two published datasets profiled endospores after they have been released from spherules [ 6 – 12 ]. The spherule transcriptome remains under-characterized, and the endospore transcriptome is essentially unknown.

Results

Generating high-density transcriptomics of wild-type and mutant Coccidioides under spherule- and hyphal-inducing conditions Over the course of this analysis, we used two strategies to characterize the spherule transcriptome and to identify key spherule-associated transcripts. We determined which transcripts were regulated by the critical transcriptional regulator RYP1-dependent genes, and we compared the spherulation transcriptome to the hyphal transcriptome to identify transcripts that were associated with each morphology (morphology-dependent genes). The TF RYP1 is required for spherulation in Coccidioides [12]. We reasoned that understanding the portion of the spherule transcriptome, that is, dependent on RYP1 would identify transcripts whose expression is associated with spherule formation rather than the conditions used to generate spherules. First, we germinated both wild-type and ryp1∆ arthroconidia under spherulation conditions, observed morphology by light microscopy, and performed RNA-Seq at the same time points as the previous experiment, now sampling from the same culture over time to increase consistency between subsequent time points of development (Figs 2A and S2A). Of note, the particulate matter in the ryp1∆ cultures was present in arthroconidia stocks and likely represents cellular debris carried over into spherulation cultures. This has been observed in prior literature and likely reflects the previously reported low viability of ryp1∆ arthroconidia [12]. To compare the wild-type spherules generated in our first and second experiments, we determined the time of endospore release, the quantity of endospore release, and spherule size. The wild-type strain did release endospores starting on day 3, but there was quantitatively less endospore release in this experiment (Fig 2B and S7 Table). Wild-type spherules achieved a similar diameter by day 6, as seen previously (S2B Fig and S8 Table). Again, the arthroconidia transcriptome was the most distinct within a genotype, with sets of transcripts being induced/repressed as spherules formed and transcripts showed substantial dependence on RYP1 (Fig 2C and S9 Table). We also observed a similar pattern of decreased differential transcripts between subsequent time points as the experiment progressed (S2C Fig). Therefore, we conclude that these separate spherule development trajectories are comparable except for the endospore release stage. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 2. Spherule and hyphal transcriptomes are dependent on RYP1. (A) Micrographs of fixed samples from each flask at the time of RNA harvest for one replicate of spherule growth in Converse, 39°C, 10% CO 2 . Subsequent samples were taken from the same flask over time. Endospore release was first observed on day 3 in wild type. As expected, the ryp1∆ mutant did not form spherules under these conditions. Surprisingly, it did form smaller rounded structures of unclear significance (open arrowheads) in addition to hyphae and some chains of oblong cells (black arrows) which have not been reported previously in Coccidioides to our knowledge. (B) Quantification of the proportion of each morphology in cultures on days 4–6, n ≥ 20 fields of view counted for each sample. Underlying data can be found in S7 Table. (C) Heatmap of transcript abundance over time in spherulation conditions. Transcripts that had at least 10 reads detected in at least three samples were included as mean-centered rows in this heatmap. Rows are clustered based on correlation across all columns. Log 2 (counts per million) indicated by yellow and blue shading. (D) Micrographs of fixed samples from each flask at the time of RNA harvest for one replicate of hyphal growth in Converse, 25°C. Subsequent samples were taken from the same flask over time. “Pellet” and “Hyphae” are the same biological samples prepared in different ways as described in E. Open arrows indicate hyphae forming initial arthroconidia. Black arrowheads indicate branching hyphae. Black arrows indicate chains of oblong cells similar to those observed for ryp1∆ in spherulation conditions. (E) Schematic of preparation of hyphal samples for microscopy. “Pellet” samples were placed in a 96-well plate with glass bottom and pelleted at 584 × g for 2 min prior to visualization. For “hyphal” samples, 5 µL of fixed samples containing small clumps of hyphae were placed on a slide with a coverslip prior to visualization. (F) Heatmap of transcript abundance over time in hyphal conditions, displayed in the same manner and with the same criteria for inclusion as in C. https://doi.org/10.1371/journal.pbio.3003066.g002 To simultaneously query the transcriptome of the hyphal morphology, we germinated the same arthroconidia stocks of wild type and ryp1∆ in hyphal-inducing conditions. It has been common to compare spherules grown in Converse medium to hyphae grown in a different rich medium (GYE) [7,8], but, to eliminate media-specific expression effects, we generated hyphae in Converse medium (at ambient temperature without additional CO 2 ). At each time point, we observed hyphae formation by light microscopy and the transcriptome by RNA-Seq (Figs 2D and S2D). To best capture the heterogeneity of hyphal cultures, we performed light microscopy using two different modalities (2E). Pelleted samples provided higher sensitivity for short filaments, whereas slides were used to examine longer hyphae and mature hyphal mats that did not pellet. We observed that wild-type samples formed germ tubes by day 1, with extension and early branching on day 2, followed by robust hyphal mats on day 3. We expected older hyphae to undergo arthroconidia generation and observed early evidence of arthroconidia formation on day 6 (Fig 2D, open arrows). The ryp1∆ mutant also demonstrated rare germ tubes on day 1 but appeared to have delayed hyphal branching as we did not observe branching structures until day 3 (Figs 2D and S2D, black arrow heads). On day 6, instead of early arthroconidia development, ryp1∆ demonstrated aberrant morphology with chains of rounded and oblong structures (Fig 2D, black arrows), similar to the morphology at late time points in spherulation conditions. ryp1∆ arthroconidia (same biological samples as seen in Fig 2C) demonstrated significantly different expression compared to wild-type arthroconidia, but wild-type and mutant hyphal transcriptomes started to resemble each other more closely over time (Fig 2F and S9 Table), suggesting that RYP1 is largely dispensable for the hyphal transcriptome. As observed with spherulation, we found a similar pattern of decreased differential transcripts between subsequent time points as the experiment progressed (S2E Fig).

Identifying RYP1-dependent and morphology-dependent transcripts during spherule and hyphal formation To further refine our understanding of the Coccidioides transcriptome and to elucidate the molecular role of RYP1 in Coccidioides development, we examined which transcripts are significantly differential in wild type compared to the ryp1∆ mutant at each time point of spherulation or hyphal growth, termed “RYP1-dependent” (Fig 3A). Surprisingly, the highest number of RYP1-dependent transcripts was in arthroconidia, where a role for RYP1 in gene regulation has not been interrogated previously. This effect was observed regardless of arthroconidia storage conditions prior to use (S3A and S3B Fig) and indicates a previously unknown and significant role for RYP1 in the transcriptome of arthroconidia, the infectious particle of this fungus, that bears further study. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 3. Defining RYP1-dependent and morphology-dependent transcripts. (A) Number of significantly differential transcripts between wild type and the ryp1∆ mutant at each time point specified. Transcripts that are induced by RYP1 (higher in wild type than ryp1∆) are in purple, and transcripts that are repressed by RYP1 (higher in ryp1∆ than wild type) are in green. (B) As in A but only with timepoints where paired spherule and hyphal wild-type datasets are available to highlight morphology-dependent genes. Dark purple and dark green correspond to the number of RYP1-dependent transcripts that are also morphology-dependent (significantly differential between wild-type spherules and wild-type hyphae) at that time point. (C) Scatterplot demonstrating the expression of all detected transcripts at day 3. On the x-axis, values are the ratio of wild-type spherule over wild-type hyphal transcript abundance transformed to log 2 (counts per million). On the y-axis, values are the ratio of transcript abundance of wild-type spherule over ryp1∆ in spherulation conditions, transformed to log 2 (counts per million). (D) Overlap between transcripts that are significantly differential between wild type and the ryp1∆ mutant at all timepoints of spherulation (days 1–6, excluding the 8 h timepoints since spherules and hyphae had not appeared by then), transcripts that are differentially expressed between wild type and the ryp1∆ mutant over all timepoints of hyphal formation (days 1, 2, 3, and 6), and transcripts that are morphology-dependent in wild type at all comparable timepoints (days 1, 2, 3, and 6). (E) Number of significantly differential transcripts between wild-type spherules and hyphae at each time point specified. Transcripts with higher abundance in spherules than hyphae are yellow and transcripts with higher abundance in hyphae than spherules are blue. (F) As in E, now highlighting dark yellow and dark blue transcripts corresponding to the number of morphology-dependent transcripts that are also regulated by RYP1 at the corresponding spherule time point. (G) Same graph as F except the dark yellow and dark blue transcripts now correspond to the number of morphology-dependent transcripts that are regulated by RYP1 at the corresponding hyphal time point. (H) Overlap between transcripts that are significantly differential between wild-type spherules and hyphae at all comparable timepoints (days 1, 2, 3, and 6) and transcripts that are differentially expressed between the ryp1∆ mutant in spherulation and hyphal-inducing conditions at the same timepoints. https://doi.org/10.1371/journal.pbio.3003066.g003 During germination into spherules, there were increasing numbers of RYP1-dependent transcripts until a peak at day 4, with more transcripts induced by RYP1 (purple) versus relatively constant numbers of transcripts repressed by RYP1 (green) (Fig 3A). In contrast, cells in hyphal-inducing conditions trended toward fewer RYP1-dependent transcripts over time, indicating that the wild type and ryp1∆ hyphal transcriptomes converge as both genotypes differentiated into hyphae (Fig 3A). At each time point sampled in both spherule and hyphal conditions, we highlighted RYP1-dependent transcripts that were also “morphology-dependent” (significantly differential between spherules and hyphae in wild-type culture at the same time point) or “morphology-independent” (Fig 3B, dark and light regions, respectively). In contrast to stable numbers of RYP1-dependent morphology-independent genes, there was an increase in the number of RYP1-dependent morphology-dependent transcripts in spherules from 8 h to day 3. This increase was largely driven by two groups of transcripts: (1) RYP1-activated, spherule-activated or (2) RYP1-repressed, hyphal-activated. This trend is more easily observed in Fig 3C, where the day 3 data are plotted, and in the global analysis in S3C and S3D Fig. In hyphal-promoting conditions, this trend was not observed, and both RYP1-dependent, morphology-dependent and RYP1-dependent morphology-independent transcripts decreased over time with no clear correlation between the hyphal-RYP1-dependent transcriptome and the morphology transcriptome (Figs 3B, S3C and S3D). Thus, during spherule development, RYP1 has an impact on the morphology regulon that increases with time and peaks on day 3, as well as a morphology-independent impact with constant magnitude over time. On the contray, in hyphal development, RYP1 has a largely morphology-independent impact on the transcriptome that decreases over time, indicating that wild-type and ryp1∆ hyphae converge on similar transcriptomes. Next, to further understand the role of RYP1 in Coccidioides biology, we examined the stringent set of transcripts that were RYP1-dependent across all spherule or hyphal time points (Fig 3D). There were 452 transcripts consistently RYP1-dependent across all six time points in spherulation conditions (termed “S-RYP1-dependent”) and 262 transcripts across all four time points in hyphal conditions (termed “H-RYP1-dependent”). Of these 452 S-RYP1-dependent and 262 H-RYP1-dependent genes, 79 were common to both sets (p = 1.94e-38 by Fisher exact test). While significant, this relatively low magnitude of overlap adds additional evidence for RYP1’s distinct regulatory roles in spherule and hyphae. We also found 551 consistently morphology-dependent transcripts by comparing wild-type spherules and hyphae. About 152 (135 + 17) of these strictly morphology-dependent transcripts were also consistently S-RYP1-dependent (p = 4.97e-71 by Fisher exact test), and 61 (44 + 17) strictly morphology-dependent genes were consistently H-RYP1-dependent (p = 1.69e-18 by Fisher exact test). The 17 transcripts that are S-RYP1-dependent, H-RYP1-dependent, and morphology-dependent include the gene D8B26_005342/CIMG_00509, which is already known to be spherule-induced, RYP1-dependent [12], and, interestingly, in an area of genomic introgression between the two known Coccidioides species, Coccidioides posadasii and Coccidioides immitis [23]. This central overlap of 17 is surprisingly low, although still significantly higher than expected by chance (χ2 = 1019.27, p < 0.0001), and again implies that RYP1 has two distinct regulatory roles in these two morphologies, more significant in spherules compared to hyphae. The high number of morphology-dependent genes that are RYP1-independent suggests roles for additional regulators of spherulation.

Morphology change triggers differential expression of a core set of transcripts across all developmental time points Next, we examined morphology-dependent transcripts at each shared time point of spherule and hyphal development in wild type (Fig 3E). As expected, the number of morphology-dependent transcripts increased over time as spherules and hyphae emerged. We highlighted the morphology-dependent transcripts that were also S-RYP1-dependent (Fig 3F) at the same time points and observed an increase in the magnitude of this subset of transcripts with the exception of day 6, while morphology-dependent S-RYP1-independent genes remain relatively constant (with the same exception of day 6). On the contrary, when we highlighted the number of morphology-dependent genes that are H-RYP1-dependent at the same time points, that number is relatively small (Fig 3G) and does not have a clear trend. Thus, the role of RYP1 in regulating morphology is related to its regulon in spherulation conditions, where it induces spherule-associated transcripts and suppresses hyphal-associated transcripts. In hyphae, RYP1 regulates a small subset of transcripts, but most of these seem to be morphology-independent. Finally, we defined a stringent set of transcripts that were consistently morphology-dependent across all shared spherule and hyphal time points (Fig 3H). As discussed above, 551 transcripts were consistently morphology-dependent in wild type. A total of 318 transcripts were consistently differential in the ryp1∆ mutant growing in spherulation conditions compared to hyphal conditions, even though the mutant forms hyphae under both these conditions. Given the uniform morphology, these 318 transcripts are likely responding to the difference in spherulation- and hyphal-inducing conditions (namely, temperature and CO 2 ). Surprisingly, the overlap between the 551 morphology-dependent genes in wild type and the 318 condition-dependent transcripts in ryp1∆ is low in magnitude (71 transcripts total, p = 2.56e-20 by Fisher exact test), meaning that the majority of the 551 morphology-dependent transcripts are linked to the morphology itself. Focusing on the 551 transcripts with morphology-dependent expression in wild type, 273 are consistently spherule enriched (of those, 82 are also consistently S-RYP1-dependent), and 239 are hyphal enriched (of those, 32 are also consistently H-RYP1-dependent). We examined these subsets further at the gene level to better understand the molecules involved in the Coccidioides morphologic transition. Within the spherule-enriched set, as expected, we found the transcript encoding the best-characterized virulence factor in Coccidioides, SOWgp [24] (D8B26_003939), and the previously reported spherule-associated gene PSP1 [7,8,12,25] (D8B26_002733). We also found D8B26_003869, the ortholog of BOI2 in Saccharomyces cerevisiae, a gene involved in polar growth and inhibition of cytokinesis during budding [26], which may imply a role for directed vesicle fusion with the plasma membrane or a delay in cytokinesis during spherule development. Additionally, there are two TFs (D8B26_005038 and D8B26_006698) in this group that are good candidates for regulators of spherulation in addition to RYP1. Of note, OPS1 [12,25] (D8B26_004398) and ALD1 [25,27] (D8B26_007314), genes that were previously published to be spherule-biased, were found to be spherule enriched in some early time points but not consistently at later time points of morphological development, demonstrating the power of this high-density developmental time course. Finally, despite the critical role RYP1 plays in inducing spherulation in Coccidioides, the RYP1 transcript itself does not demonstrate morphology-specific expression (S3E Fig). In the consistently hyphal-enriched transcripts, we found STU1 (D8B26_002234), the Coccidioides ortholog of Aspergillus APSES family TF STUA which regulates conidiation [28]. This finding matches the ortholog of STUA in Histoplasma, STU1/EFG1, which is extremely hyphal-biased in its expression [29]. Consistent with previous findings, the major component of the woronin body structure that plugs damaged areas of hyphal walls, HEX1 (D8B26_006047), was up-regulated in hyphal conditions compared to spherules [7]. As expected, these hyphal-associated genes were also up-regulated in ryp1∆ cells in both spherule- and hyphal-inducing conditions. Somewhat unexpectedly, the cytosolic catalase (D8B26_007217) was found to be consistently higher in hyphal conditions and the ryp1∆ mutant. This gene has been previously found to have higher expression in spherules than hyphae [8,12] in studies in which the spherules and hyphae were grown in different media. Given the discordant findings between our data and previous publications, we believe nutritional cues play a key role in regulating this particular transcript. Thus, our rich dataset identifies 551 consistently morphology-dependent transcripts that are prime effector and regulatory candidates for control of the Coccidioides developmental program, deconvolutes the effects of change in growth conditions from change in morphology, and identifies 273 spherule-enriched genes that are likely to be involved in virulence.

RYP1 binds to two distinct subsets of promoters We next sought to determine which RYP1-dependent genes displayed association with RYP1 using ChIP-Seq with an antibody generated against a peptide epitope of RYP1. While we attempted to perform ChIP on arthroconidia and multiple time points of spherule or hyphal growth (8 h, D1, D2, D4, micrographs in S4A Fig) and one time point for each morphology of the ryp1∆ mutant (micrographs in S4B Fig), consistent RYP1 binding was only detectable for spherules on days 1, 2, and 4 and hyphae on days 2 and 4 (S10 Table). This lack of binding in wild type may be due to less initial biomass and does not necessarily reflect a lack of RYP1 binding at those early time points. In the ryp1∆ mutant, we expected very little binding of the RYP1 antibody, and while we did identify sporadic peaks in individual replicates, they were not reproducible and likely represented low-level off-target binding of the antibody. We focused our subsequent analyses on those later time points with >400 detected peaks in at least two of three replicates. As expected, we observed RYP1 binding in spherules at the SOWgp promoter (Fig 4A), consistent with observations in this and prior studies [11,12,30] that have found SOWgp expression to be spherule-associated and RYP1-dependent. Additionally, we examined the RYP1 locus itself and found that RYP1 bound both upstream and downstream of the gene, suggesting a possible autoregulatory mechanism for RYP1, a known characteristic for RYP1 orthologs in other fungi [15,16,31,32] (Fig 4B). Some promoters in Coccidioides are very large and demonstrate RYP1 binding peaks far from the predicted ATG (S4C Fig), including the intergenic regions between D8B26_007678, encoding a hyphal-enriched hypothetical protein, and D8B26_007679, encoding a spherule-enriched predicted NAD kinase (S4D Fig). Finally, we found examples of RYP1 binding in hyphae and spherules (Fig 4C), including D8B26_005360 and D8B26_005361, both RYP1-repressed transcripts that encode hypothetical proteins. Upon manually reviewing the 32 genes designated as bound in hyphal samples only, we found evidence of binding in spherule samples as well and believe these are instances in which MACS did not correctly identify peaks in the paired spherule time point. Therefore, we do not think there are any examples of RYP1 binding promoters in hyphal samples alone. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 4. RYP1 regulates two distinct subsets of targets. (A) Traces demonstrating chromosomal location of fold enrichment of ChIP signal/input in spherules (yellow) and hyphae (blue) at the designated time points relative to the annotated SOWgp gene. (B) As in A but demonstrating ChIP signal/input relative to the annotated RYP1 gene. (C) As in A but demonstrating ChIP signal/input relative to D8B26_005359, D8B26_005360, and D8B26_005361 genes (genes indicated by gray arrows). (D) Barplot demonstrating the proportion of S-RYP1-dependent genes or H-RYP1-dependent genes (as defined in 3D) whose promoters have RYP1 binding by ChIP-Seq. (E) UpSet plot demonstrating the size of each individual set of genes whose promoters are bound at each designated spherule/hyphal time point (bottom left) and size of overlap between each of these sets (magnitude on top, overlapping sets demonstrated by connected black circles on bottom). Each gene can only be assigned to one unique category. (F) Overlap of spherule-specific peaks (genes whose promoter is bound in spherule timepoints only, without RYP1 binding in the corresponding hyphal time point or ryp1∆ mutant subjected to the same conditions as wild type) on days 1, 2, and 4. (G) Motif enriched in DNA sequences of RYP1 ChIP-Seq peaks found consistently in both days 2 and 4 spherule and hyphal datasets (498 sites, p = 4.7e-064 on day 4 and 193 sites, p = 3.7e-046 on day 2). (H) Percent of genes in each subset (from 3A, 3D) whose promoter regions have at least one hit for the RYP1 binding motif in 4G. Promoters are defined as the sequence upstream of the coding sequence (CDS) start until the next upstream CDS is encountered, or 10 kb maximum. Fully RYP1-dependent genes are those that are significantly differential between wild type and ryp1∆ in all spherule timepoints (days 1–6) and all hyphal time points (days 1, 2, 3, and 6). * : p < 0.05, **: p < 0.005, by Fisher exact test. (I) Motif enriched in DNA sequences of RYP1 ChIP-Seq peaks found uniquely in days 2 and 4 spherule datasets (>1,000 sites, p = 1.3e-044 on day 2, 215 sites, p = 4.5e-046 on day 4). (J) As in H but now for RYP1 binding motif in I. https://doi.org/10.1371/journal.pbio.3003066.g004 We determined which RYP1-dependent genes observed by RNA-Seq were also direct targets of RYP1 by ChIP-Seq. We found that 60% of the 452 S-RYP1-dependent transcripts also demonstrated RYP1 promoter binding in at least one spherule time point of our ChIP-Seq experiment (Fig 4D). In contrast, only 27% of the 262 H-RYP1-dependent genes had RYP1 promoter binding in hyphae. We additionally examined the percentage of the 551 morphology-dependent genes we had previously defined (and the 786 morphology-dependent genes we found from the paired RNA-Seq from this ChIP-Seq experiment, S11 Table) that had RYP1 binding in their promoters. We found ~55% of both subsets of morphology-dependent genes were RYP1 targets (S6E Fig). Thus, RYP1 plays an important role as a direct regulator of morphology in Coccidioides, where it seems to act specifically by binding promoters in spherule development. Although the loss of RYP1 influences the hyphal transcriptome, the ChIP-Seq data suggest that effect is more indirect. Next, we created an UpSet plot (Fig 4E) to group genes whose promoters had RYP1 binding detected. This analysis revealed that most genes fell into three categories: (1) genes whose promoters are bound at the spherule day 2 time point only, (2) genes whose promoters are bound in both later spherule timepoints, and (3) genes whose promoters are bound in both later spherule timepoints in addition to the latest hyphal time point. Taken together, this likely indicates two distinct regulons for RYP1: an exclusive spherule regulon and a shared spherule/hyphal regulon. Of note, there were minimal numbers of genes whose promoters demonstrate RYP1 binding in hyphae only, indicating that there does not seem to be a unique RYP1 regulon in hyphae. Given the large number of spherule RYP1 peaks detected, we further delved into those genes that had spherule-specific promoter binding (without any observed RYP1 peaks in hyphal samples or ryp1∆ samples). In examining the genes containing spherule-specific peaks across the spherule timepoints (Fig 4F), we found that the RYP1 spherule regulon appears to be dynamic, with 274 genes bound by RYP1 only in day 1 spherules, 2049 unique genes bound by RYP1 in day 2 spherules, and 882 genes bound by RYP1 in both of the later spherule timepoints. Thus, RYP1 in Coccidioides appears to have a complex and dynamic role in regulating multiple stages of the spherule morphology, suggesting it plays a role in a complex regulatory network like some of its orthologs in other fungi [15,33,34].

Two distinct RYP1 motifs in Coccidioides are enriched in promoters of RYP1-bound genes To better understand how RYP1 can regulate two distinct subsets of genes, we performed motif searches on multiple subsets of RYP1 peaks combined as illustrated in S4F Fig: (1) peaks found in promoters of genes in both spherules and hyphae and (2) peaks found in promoters of genes only in spherules. In the first group, we discovered a significantly enriched motif (Fig 4G), that is, extremely similar to the previously published RYP1 motif in Histoplasma (S4G Fig). We used MAST to search for this RYP1 spherule/hyphal motif in all Coccidioides promoters in the genome with the threshold E-value of 2.08e-04, a cutoff which proved useful for this analysis in Histoplasma [15]. Since the motif has low information content, we found that 30% of all promoters had a hit to the RYP1 motif (Fig 4H). The percent of promoters containing RYP1 motif hits was highest for genes whose transcripts are RYP1-dependent across all time points studied and genes whose transcripts are RYP1-dependent across all spherule time points (S-RYP1-dependent). Since this motif was derived from peaks found in both spherule and hyphal morphologies, unsurprisingly, the enrichment of the RYP1 binding motif in the promoters of H-RYP1-dependent genes and morphology-dependent genes was also quite high (~40%). Finally, we found no enrichment of the motif in the 3,599 RYP1-dependent transcripts in arthroconidia, suggesting that RYP1 may control this regulon indirectly through a second major regulator. Together, this motif analysis indicates that the presence of the RYP1 motif in promoters alone does not explain the varying impact of RYP1 on spherules, hyphae, and arthroconidia that we observed by RNA-Seq. Similar to peak distribution in promoters, the motif was found distributed over a wide range of distances, with a strong bias toward proximity to the gene (S4H Fig). Next, we examined the number of RYP1 motif hits per promoter for each of these subsets of RYP1-motif-containing promoters (S4I Fig). Interestingly, about 40% of S-RYP1-dependent genes with motif hits had more than one motif hit, including some promoters with up to five total RYP1 motif hits. This trend toward more motif hits was unique to S-RYP1-depenent genes and may provide a clue toward the mechanism by which RYP1 has more impact on the transcriptome in spherules despite a shared DNA binding sequence in both spherules and hyphae. Second, we performed motif searches on peaks that were found only in spherule conditions and discovered a novel motif that has not been reported before for RYP1 association in any organism (Fig 4I). Given the extremely different sequence from the canonical RYP1 motif described above, we hypothesize that this motif reflects recruitment of RYP1 to these promoters via interaction with a second (unknown) regulator that binds this motif directly. Using a more stringent E-value of 1e-06 given the higher information content in this motif, we found that it was present in 20% of promoters across the genome. This motif was significantly enriched in S-RYP1-dependent and morphology-dependent gene promoters (Fig 4J). There was a similar distribution of the position of this motif relative to the gene compared to the canonical RYP1 motif except for mildly decreased numbers of motif hits more than 8 kb from the gene ATG (S4J Fig). Unlike the canonical RYP1 motif discussed above, there was not a similar trend toward increased numbers of RYP1 motif hits per promoter in any gene subsets (S4K Fig). Interestingly, S-RYP1-dependent and morphology-dependent genes have significantly longer promoters than other gene subsets, which may accommodate both motifs we identified and potentially more numbers of the canonical RYP1 motif in S-RYP1-dependent genes (S4L Fig). Thus, we find that distinct RYP1-associated motifs, number of motifs per promoter, and potentially combinatorial motifs could contribute to the ability of RYP1 to possess two distinct regulons.

Defining endospore-associated transcripts Since the endospore form is even less characterized than spherules, we used the RNA-Seq data corresponding to cultures for which we observed the most endospore release to identify potential endospore-enriched transcripts. Specifically, we interrogated day 3 and later time points from Fig 1, and also performed additional RNA-Seq from spherules formed in DMEM + 20% FBS and harvested at day 3, when released endospores were abundant (S5A Fig). These spherules were generated from the same arthroconidia as described in Fig 2. We defined endospore-enriched transcripts as those that were all consistently differential on days 3–6 compared to days 1 and 2 in the experiment from Fig 1, all significantly differential in DMEM conditions compared to RPMI + 10% FBS conditions (S5B Fig, also generated from the same arthroconidia as described in Fig 2) which did not exhibit endospore release, and all significantly differential in DMEM conditions compared to day 1 and 2 spherules, day 1 and 2 hyphae, and arthroconidia from the same experiment. For all these differential comparisons, we enforced criteria that the direction of differential expression had to be consistent for the transcript across the comparisons made. Of the transcripts meeting the above criteria, there were 18 transcripts with an increase in expression in samples containing endospores and two transcripts that demonstrated consistent decrease in expression in samples containing endospores (Fig 5B and S13 Table). The transcripts with increased abundance included MEP1, a metalloprotease which is known to play a role in masking endospore recognition by the immune system [40], and CTS1 (misannotated as two separate transcripts D8B26_000666/7 in the current genome), an endochitinase that has been previously characterized to have maximal expression when endospores are present in culture [41]. Interestingly, the endospore-enriched transcripts include two other secreted serine proteases, D8B26_003356 and D8B26_007338. Both transcripts that are consistently downregulated in endospore-containing cultures have no available annotation data. This list of genes represents the first and strongest candidates for factors intimately involved in endospore biology.

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3003066

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/