Macrophage phagocytosis of Coccidioides promotes its differentiation into the parasitic form Authors: Jane Symington, Mark Voorhies, Bevin C. English, Apoorva Dabholkar, Anita Sil Published: 2026 ABSTRACT ABSTRACTCoccidioides is an endemic fungus that is increasing in prevalence and causes life-threatening diseases in immunocompetent people. In the environment, the spores (arthroconidia) develop into hyphae; however, when they are inhaled by a mammalian host, they develop into a unique form called the spherule. The transition to spherule can be triggered in vitro with elevated temperatures and high CO2 levels, but how the host triggers Coccidioides spherulation is not known. We used live imaging to investigate how macrophages affect the fate of Coccidioides arthroconidia. Under tissue culture conditions, arthroconidia quickly developed into hyphae. Remarkably, the addition of macrophages promoted spherule development and delayed hyphal formation. Macrophage supernatants were not sufficient to promote spherule development, and chemical blockade of phagocytosis inhibited spherule formation. Transcriptomics analysis of Coccidioides co-cultured with macrophages revealed a signature concordant with spherules grown in vitro and allowed the identification of a core set of 143 spherule-specific transcripts. Additionally, we identified 229 Coccidioides transcripts with significantly higher abundance in the presence of macrophages compared to in vitro generated spherules, suggesting that these factors may be needed for growth in the presence of innate immune cells; 48 induced transcripts were predicted to encode secreted proteins, suggesting a function at the host-pathogen interface. Taken together, this work highlights the capacity of macrophages to promote development of the parasitic form of Coccidioides and lays a foundation for uncovering host-pathogen signaling as well as Coccidioides factors that are critical for pathogenesis.IMPORTANCEValley Fever is a disease caused by inhalation of the spores of the fungus, Coccidioides spp. It can present like the flu, pneumonia, bone infections, or meningitis. Once inhaled, the spores change into a pathogenic form that allows the fungus to spread throughout the body and cause disease. How the spores make this transition in the body is not well understood. We investigated how immune cells affected this transition. We found that engulfment of spores by innate immune cells stimulated the transition to the pathogenic form of the fungus. We determined which fungal genes are induced during interactions with innate immune cells, potentially identifying genes that may be critical for the development of the pathogenic form. This work helps us understand how this pathogen is taking advantage of our immune system to survive and cause disease. INTRODUCTION Coccidioides spp. are fungal pathogens that cause Coccidioidomycosis or Valley Fever in both immunocompetent and immunocompromised individuals. Although many experience a self-limiting disease, others develop severe disseminated forms, including meningitis, which requires lifelong therapy (1). Coccidioidomycosis cases are rising nationally, and the endemic area may be spreading, making better diagnostics and therapeutics essential. Our understanding of this pathogen is limited in part because Coccidioides has a unique parasitic morphological form called the spherule. In the environment, Coccidioides grows as hyphae, but when the spores (arthroconidia) are inhaled, they undergo a morphologic transition to the spherule, which can grow as large as 40–100 μm in diameter. The spherule is filled with hundreds of internal cells called endospores (2), which are released upon spherule rupture. Each endospore can develop into a spherule and further spread the disease. Our understanding of spherule development is largely based on in vitro experiments with specialized media, elevated temperatures (39°C), and elevated CO2 (~10%–20%) (3, 4). These studies have produced key understandings of the timing of events in the development of spherules from arthroconidia (5, 6); however, how well these in vitro studies mimic events within a mammalian host remains unknown. Many open questions about Coccidioides-host interactions remain, including how the transition to spherules is triggered in the host and how arthroconidia avoid clearance by innate immune cells. The initial infection site with Coccidioides is the lung. Respiratory pathogens like Coccidioides must overcome the host defenses of the lung, including macrophages, other immune cells, and epithelial cells, to cause disease. Many successful respiratory pathogens, including Histoplasma capsulatum and Mycobacterium tuberculosis, subvert anti-microbial processes in macrophages to survive in the host (7, 8). Early papers suggested that a variety of immune cells could kill Coccidioides arthroconidia, delay germination of hyphae, or possibly promote spherulation (9–11). Here, we take advantage of live-cell imaging in a Biosafety Level 3 facility as well as transcriptomics studies to assess how macrophages affect the developmental and molecular fate of Coccidioides arthroconidia. We determined that murine bone marrow-derived macrophages (BMDMs) stimulated a significant increase in spherule formation and a robust delay in hyphal formation in a phagocytosis-dependent manner. Using RNA-seq to determine the transcriptional response of Coccidioides to macrophage co-culture, we observed that the fungus accumulates transcripts associated with in vitro spherulation, as well as a unique set of transcripts that were more highly abundant in the presence of macrophages. Together, these studies elucidate how Coccidioides adapts to the presence of host immune cells by forming spherules and inducing the expression of potential virulence factors to cause disease. RESULTS Bone marrow-derived macrophages promote Coccidioides spherule development We used live imaging to compare the germination and subsequent development of arthroconidia in tissue culture conditions (37°C, 5% CO2, in bone marrow-derived macrophage media) alone or in the presence of BMDMs. In the absence of macrophages, arthroconidia primarily developed into hyphae, rarely forming spherules (Fig. 1A; Video S1). Strikingly, in the presence of BMDMs, many arthroconidia developed into spherules (Fig. 1B and C; Video S2). The spherules formed in the presence of macrophages were also larger in diameter than spherules formed under tissue culture conditions (Fig. 1D). In addition to promoting spherulation, the presence of macrophages significantly delayed the time it took for hyphae to first be observed in the field, which we described as “time to first hyphae” (Fig. 1E). Since time to first hyphae is a function of both the rates of arthroconidial germination and hyphal growth, we note that either or both could have been affected by BMDMs. Macrophages promote spherule formation and delay hyphae. Representative image of (A) arthroconidia alone or (B) arthroconidia with BMDMs at a multiplicity of infection of 0.1 (1 arthroconidia for every 10 BMDMs) on day 3 of the experiment; scale bar represents 30 μM. Representative spherules are labeled with blue arrowheads. (C) The number of spherules per well (each point represents data from five fields per well with three independent wells per condition). (D) Average diameter of spherules on day 3 of infection in wells with arthroconidia alone or with BMDMs represented as a box and whisker plot. (E) Time to first hyphae in frame over 3 days, with pictures taken once an hour, five fields per well. Data presented (C, D, and E) are representative of at least three independent experiments. Unpaired t-test used for panel C and Mann-Whitney test for panels D and E; error bars in panels C and E indicate standard deviation, ***P ≤ 0.001 and ****P ≤ 0.0001. We investigated whether these morphological findings were affected by the number of arthroconidia per macrophage (multiplicity of infection, or MOI). At the lower MOI of 0.01, there were still significantly more spherules in the presence of macrophages than when the same arthroconidial inoculum was grown alone (Fig. 2A). Notably, the size of the spherules at MOI 0.01 was larger than the spherules that developed at an MOI of 0.1 (average diameter of 54 µm vs. 29 µm) (Fig. 2B). The size of the spherules that formed in the presence of macrophages was more comparable to the size of spherules observed previously in vivo (40–100 μm), in contrast to the smaller spherules that are characteristic of in vitro spherulation (10–20 μm) (2, 12). At this lower inoculum, the time to observation of first hyphae in the microscopy field was longer, as sparser arthroconidia mean hyphal germination is a rarer event; however, there was still a significant increase in time to first hyphae in the wells with macrophages compared to those without macrophages (Fig. 2C). Macrophage induction of spherulation and delay of hyphae is independent of MOI and Coccidioides species. (A) The number of spherules per well (each point represents data from five fields per well with three independent wells per condition) in wells with arthroconidia alone or BMDMs at MOI 0.01. (B) Average diameter of spherules on day 3 of infection at MOI 0.1 or MOI 0.01 represented as a box and whisker plot. (C) Time to first hyphae in the frame of arthroconidia grown alone or with BMDMs at MOI 0.01. (D) Number of spherules per well and (E) time to first hyphae for Coccidioides immitis RS arthroconidia alone or with BMDMs at MOI of 0.1 or 0.01. (F) Number of spherules per well and (G) time to first hyphae of WT and ryp1∆ C. posadasii Silveira arthroconidia alone or with BMDMs at MOI 0.1 or 0.05. Pictures taken once an hour, five fields per well. Data for each figure represent five 40× fields counted per well, with three wells per condition. The data presented are representative of at least three independent experiments. Unpaired t-test used for panel A, Mann-Whitney test for panels B and C, one-way ANOVA for panels D and F, and Kruskal-Wallis test for panels E and G, *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001. These experiments were performed with the C. posadasii Silveira strain. To determine if a different Coccidioides species behaves similarly in the presence of macrophages, we performed the same experiments with C. immitis, which has more than 90% homology in predicted proteins with C. posadasii and a similar clinical manifestation (13). We found that macrophages also induced spherulation of C. immitis RS arthroconidia (Fig. 2D) and delayed the time to first hyphae (Fig. 2E). To characterize fungal genes that are required for spherulation in the presence of macrophages, we co-cultured macrophages with mutant arthroconidia lacking the transcription factor RYP1. We have previously shown that RYP1 is essential for the development of the spherule in vitro and in a murine infection model (6, 14). We found that RYP1 is also essential for spherulation in the presence of macrophages (Fig. 2F). Interestingly, we found that the time to first ryp1Δ hyphae in the field was delayed in the presence of macrophages (Fig. 2G), implying that macrophages inhibited germination and/or hyphal growth rates of the ryp1∆ mutant independent of spherule formation. Spherule development is dependent on contact with live macrophages Secreted compounds from macrophages could be responsible for promoting the spherulation of Coccidioides during co-culture. To test this hypothesis, we utilized a transwell system to image arthroconidia in the lower well that either had no exposure to macrophages, had macrophages present in the same lower well, or had macrophages seeded on the insert. Macrophages seeded on the insert were separated from Coccidioides by a 0.4-μm membrane that allowed media and secreted compounds to move between the insert and main well but prevented direct contact between the arthroconidia and macrophages. The presence of macrophages in the same wells as arthroconidia resulted in an increased number of spherules per field and a delay in time to first hyphae, as expected (Fig. 3A and B). However, when macrophages were separated from arthroconidia by the insert, they were unable to stimulate spherulation (Fig. 3A). Additionally, the time to first hyphae in these wells was indistinguishable from wells with no macrophages (Fig. 3B). These data indicate that contact between arthroconidia and macrophages was essential for spherule stimulation and delay of hyphae. Phagocytosis of arthroconidia is required for macrophages to promote spherule formation and delay hyphae. The number of spherules per field (A) and time to first hyphae per field (B) for arthroconidia cultured with or without macrophages in the well with arthroconidia or with macrophages separated by a transwell. The number of spherules per well (C) and time to first hyphae per field (D) arthroconidia with live macrophages or heat-killed or methanol-treated macrophages (at the same density or twice as dense as live macrophage well). The number of spherules per well (E) and time to first hyphae per field (F) of arthroconidia cultured with or without macrophages in the presence or absence of 10 μM cytochalasin D. For all samples, images were counted on day 3. For panels A and B, nine fields per condition were quantified, and for panels C–F, five fields per well and three wells per condition were quantified. The data presented are representative of at least three independent experiments. Kruskal-Wallis test was used for panels A–D and F, and one-way ANOVA was used for panel E. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001. Since contact with macrophages was necessary for the promotion of spherulation in our conditions, we investigated whether contact with dead macrophages was sufficient to induce spherulation and delay time to first hyphae. We exposed arthroconidia to live macrophages, heat-killed macrophages, or methanol-fixed macrophages. When arthroconidia were plated with BMDMs that were heat-killed or fixed with methanol, there were no changes in spherule number (Fig. 3C) or the time to first hyphae (Fig. 3D). Thus, contact with the surface of dead macrophages was not sufficient to induce spherule development. Macrophages do not induce spherulation when phagocytosis is blocked To test if phagocytosis was required for macrophage promotion of spherulation and delay of hyphae formation, we exposed arthroconidia to BMDMs in the presence of cytochalasin D, an actin polymerization inhibitor that inhibits phagocytosis as well as secretion. There was no increase in spherulation of Coccidioides exposed to cytochalasin D-treated macrophages compared to arthroconidia grown alone, with or without cytochalasin D (Fig. 3E). Of note, exposure of arthroconidia to 10 µM cytochalasin D during in vitro spherulation conditions did not inhibit the ability to make spherules or produce endospores in vitro (Fig. S1). The delay in hyphal formation noted in the presence of macrophages was eliminated if those macrophages were treated with cytochalasin D (Fig. 3F). Co-culture with Coccidioides arthroconidia leads to macrophage death Many pathogens that survive phagocytosis by host immune cells manipulate host cell death to survive within or to defend against those cells. We examined the effect of arthroconidia infection on macrophage survival by measuring the release of lactate dehydrogenase (LDH) from host cells as a proxy for host-cell lysis. When macrophages were infected at an MOI of 0.1, there was no significant LDH release until day 2 of infection. When macrophages were infected at a higher MOI of 1, they showed significant LDH release at days 1, 2, and 3 post-infection (Fig. S2). We concluded that co-culture with Coccidioides arthroconidia led to the lysis of BMDMs. Coccidioides induces a unique transcriptional program when challenged with macrophages To further characterize macrophage-induced spherulation, we performed RNA sequencing of Coccidioides co-cultured with macrophages. To overcome the challenge of low fungal RNA yield from co-culture, we utilized multiple MOIs to increase the range of fungal burden at early time points, subjected samples to bead beating to optimize lysis of fungal cells, and increased the sequencing depth. We examined the transcriptome of WT Coccidioides grown alone for 2 days, or co-cultured with BMDMs at an MOI of 1 (Hi) for 1 day (with greater RNA yield for the early time point) and at an MOI of 0.1 (Lo) for 1 or 2 days (when the majority of macrophages are still viable [Fig. S1]). To distinguish which transcriptome changes may be specific to the formation of spherules rather than the delay of hyphal formation, we leveraged the ryp1Δ mutant, which cannot make spherules in the presence of macrophages (Fig. 2F), although it did have a delay in time to first hyphae in the presence of macrophages (Fig. 2G). We reasoned that transcriptome changes that correlated with spherulation would not occur in the ryp1∆ mutant. Therefore, we profiled the transcriptomes of WT Coccidioides and the ryp1Δ mutant in the presence of macrophages at the same MOIs and time points, each in triplicate (Fig. 4A). Coccidioides induces transcriptome changes in response to BMDMs. (A) Schematic of RNA-seq conditions with uninfected macrophages, macrophages exposed to WT arthroconidia (blue), and ryp1∆ arthroconidia (red). “Hi” indicates MOI = 1, and “Lo” indicates MOI = 0.1. (B) PCA plot of each sample with WT in blue, ryp1∆ in red, Coccidioides alone at D2 is represented by circles, Coccidioides exposed to BMDMs at Lo MOI by triangles (open at D1 and filled at D2), and Hi MOI at D1 by squares. (C) Hierarchically clustered heatmap of differentially expressed transcripts for seven conditions, each compared to WT Coccidioides grown alone in tissue culture conditions. The seven conditions are WT or ryp1∆ Coccidioides grown in the presence of BMDMs (at the indicated MOI and time point) or ryp1∆ Coccidioides grown alone under tissue culture conditions. Heatmap colors indicate log2 differential expression ratios relative to WT alone in units of counts per million. Depth-normalized abundance of transcripts (as log2 counts per million) of two Coccidioides genes known to be expressed in spherules (PSP1 [D] and SOWgp [E]). Dots represent individual observations, and bars are averages in log space. Transcript abundance was quantified with KALLISTO relative to the most recent C. posadasii Silveira gene annotation (GCA_018416015.2). In the text below, we give systematic gene names from this annotation (starting with D8B26_) as well as the corresponding gene names from the previous C. posadasii Silveira gene annotation (starting with CPSG_) and orthologous genes from C. immitis RS (starting with CIMG_). Further analysis was restricted to 8,274 well-represented transcripts that had at least 10 counts in at least 9 samples (Table S1). Principal component analysis of the expression profiles revealed two independent components accounting for 65% of the variance. The first component accounted for 46% of the variance and separated the samples by exposure to BMDMs and time, regardless of genotype (WT or ryp1Δ), and the second component accounted for 19% of the variance and distinguished samples with spherules from those with majority hyphae (i.e., WT Coccidioides with BMDMs vs. WT alone or any of the ryp1Δ samples) (Fig. 4B). Differential expression analysis was carried out with the negative binomial model of edgeR to account for counting noise in the low-depth data (due to limiting fungal material in co-culture), combined with log-normally distributed biological and replicate variance. For the 8,274 well-sampled transcripts, we estimated contrasts for each WT and ryp1Δ infection time point, and for ryp1Δ alone, each relative to WT alone; 3,753 transcripts were at least 2-fold differential in at least one of these seven contrasts at a false discovery rate (FDR) of 5% (Fig. 4C; Table S2). Hierarchical clustering revealed eight distinct expression patterns as indicated by the colored regions on the right side of the heat map (Fig. 4C). More transcripts decreased in abundance in response to macrophages in both WT and ryp1Δ Coccidioides compared to WT arthroconidia grown in the same conditions in the absence of host cells (“D2 hyp”). Interestingly, prior reports of in vitro grown young spherules vs. mature hyphal cultures also noted that more genes were downregulated than upregulated in early spherules compared to mature hyphae (15). Approximately one-third of the differentially expressed transcripts showed increased abundance in WT Coccidioides in response to BMDMs, including genes previously identified as encoding spherule-associated factors, such as SOWgp, PSP1, MDR1, and OPS1 (Fig. 4C, D, and E) (16, 17). Other transcripts included in this subset were PDC1 (putative pyruvate decarboxylase), UAZ1 (putative urate oxidase), and CCC1 (a putative calcium transporter). BMDM-induced genes in WT Coccidioides also included transcripts encoding hypothetical proteins of interest that were previously shown to be more significantly induced in animal models vs. in vitro, including D8B26_007421 (CPSG_01366/CIMG_09001, a putative secreted protein), D8B26_000085 (CPSG_05795/CIMG_05576, an immunoreactive protein that has been found in sera from immunized or Coccidioides-infected dogs) (18), and D8B26_005342 (CIMG_00509, another putative secreted protein) (19). Thus, our data suggested that the presence of macrophages could sufficiently mimic the host environment to elicit the transcription of these host-associated genes in Coccidioides. We were interested in transcripts that were dependent on RYP1 for increased abundance since these transcripts might correlate with the ability of the fungus to generate spherules. Transcripts that were increased in WT Coccidioides exposed to BMDMs but lower in ryp1Δ Coccidioides exposed to BMDMs included D8B26_001957 (encoding a protein containing a Major Facilitator Superfamily [MFS] domain), D8B26_005576 (ortholog of iron transport multicopper oxidase FET3), and D8B26_000555 (a urease). Members of the siderophore biosynthesis cluster (genes SID3, SID4, NPS1, OXR1, and MFS1 [20, 21]) were also in this group, but interestingly, they were only significantly increased in abundance in the WT samples exposed to BMDMs at day 1 but not at day 2. This temporal regulation could be due to the local environment—at day 1, many of the WT Coccidioides remained within or closely associated with macrophages (potentially experiencing iron deprivation), whereas by day 2, the spherules were larger and no longer internalized. Macrophage-induced transcripts common to both WT and ryp1Δ Coccidioides could represent a generalized Coccidioides response to the stress of host cells independent of spherulation. One example was D8B26_008184, which exhibits homology to the transcription factors RlmA from Aspergillus niger and Smp1/Rlm1 from Saccharomyces cerevisiae and is proposed to be induced by cell wall stress in other fungi (22–24). Another gene with increased transcript abundance in all samples with BMDMs was D8B26_001686, which encodes a protein of the Nramp family of metal transport proteins (similar to S. cerevisiae Smf1-3 transporters), which suggests that phagocytosed arthroconidia experience differences in metal availability within the host cell. D8B26_002597, which was enriched in all samples with BMDMs, especially at 24 h, appears to encode a non-ribosomal peptide synthase-like protein that might generate a natural product in response to host cells. These changes in the transcriptome in response to macrophages highlight that the arthroconidia are likely experiencing cell wall stress and changes in nutrient availability and thus induce the transcription of a set of genes independent of spherulation. Since BMDMs promoted Coccidioides spherulation and inhibited hyphal growth, we expected that hyphal-associated transcripts would show decreased abundance in WT Coccidioides exposed to BMDMs. Indeed, many genes associated with hyphal growth in Coccidioides and other fungal pathogens were downregulated in WT Coccidioides during macrophage infection. One such gene was D8B26_004243, which encodes a putative Zn2C6 transcription factor orthologous to Histoplasma Nos1, which is upregulated in hyphae, as well as Aspergillus spp. RosA and NosA, a putative regulator of sexual development (25, 26). Transcript levels of D8B26_003897, a putative glucosidase similar to the A. fumigatus hyphal-associated glycosylhydrolase Crf2 (27), also showed decreased abundance in WT Coccidioides in the presence of macrophages. The Coccidioides orthologs of two transcription factors that are key players in the hyphal program in Histoplasma, Stu1 (D8B26_002234), and Fbc1 (D8B26_003963) (28, 29), both showed decreased transcript abundance in WT Coccidioides in the presence of macrophages. In the case of the ryp1 mutant, although there was also a delay in the appearance of hyphae during macrophage co-culture (Fig. 2G), the mutant is unable to form spherules and can only grow as hyphae. Correspondingly, for the majority of hyphal-associated genes, there was no obvious decrease in expression at the time points we surveyed. Coccidioides transcripts induced during macrophage co-culture include both core and host-specific spherule-associated genes We hypothesized that the presence of BMDMs induced a core set of spherule-associated genes that were also induced under in vitro spherulation conditions, as well as a set of potential virulence factors or effectors that were specific to the presence of host cells. To identify the core genes, we leveraged our laboratory’s recent work creating a transcriptomic atlas of Coccidioides arthroconidia, developing spherules, and developing hyphae (6). We compared the gene expression data from the following three sets (Fig. 5A; Table S3): (i) WT Coccidioides with BMDMs (spherule) vs. WT Coccidioides alone (hyphal); (ii) ryp1∆ Coccidioides with BMDMs (hyphal) vs. ryp1∆ Coccidioides alone (hyphal); and (iii) in vitro spherules vs. in vitro hyphae at days 1, 2, or 3 post-germination. We considered genes in a given set to be induced if their transcripts were significantly more abundant for at least two of the three conditions in that set. In vitro spherules had 947 differential transcripts with increased abundance in comparison to in vitro hyphae, the majority of which were not shared with the other contrasts, possibly due to the unique temperature and CO2 utilized to culture these cells in vitro. In the presence of BMDMs, WT and ryp1Δ Coccidioides had 455 and 395 differential transcripts with increased abundance compared to each of their corresponding Coccidioides alone controls. Although a majority of the differentially expressed transcripts in response to macrophages were unique to either WT or ryp1Δ Coccidioides, 56 were shared only among the two. Interestingly, the shared set could represent stress-response transcripts that increased in abundance during macrophage co-culture independent of spherule formation and Ryp1 activity. More significantly, this analysis indicates that the majority of genes induced in wild-type Coccidioides in the presence of macrophages were dependent on Ryp1 and/or spherule formation. BMDMs trigger the induction of core spherule-associated genes and macrophage-specific spherule-associated genes. (A) Schematic of different samples used for the analysis in panel B. (B) UpSet plot comparing the different gene sets depicted in panel A. “Mɸ WT” are transcripts with significantly increased abundance in at least two-thirds of the WT Coccidioides samples exposed to BMDMs vs. WT Coccidioides alone; “Mɸ ryp1” are transcripts with significantly increased abundance in at least two-thirds of the ryp1∆ Coccidioides samples exposed to BMDMs compared to ryp1∆ Coccidioides alone, and “in vitro S WT” are transcripts with significantly increased abundance in at least two-thirds in vitro spherule samples compared to in vitro hyphae at the same time point (6). The horizontal black bars on the left show the size of each individual set of transcripts with increased abundance as described above. The vertical black bars show the magnitude of the overlap between each of these sets (overlapping sets are indicated by connected black circles on the bottom). Each gene can only be assigned to one category per condition. The green box surrounds the subset of transcripts with significantly increased abundance in both in vitro and BMDM-associated spherules. Orange box surrounds subset of transcripts with significantly increased abundance only in WT Coccidioides exposed to BMDMs. (C) Heatmap of transcript abundance for WT Coccidioides grown in the presence of BMDMs at different time points and MOIs vs. WT Coccidioides grown alone at day 2 (D2 hyp); ryp1∆ Coccidioides grown in the presence of BMDMs at different time points and MOIs vs. ryp1∆ Coccidioides grown alone at day 2 (D2 hyp); or WT Coccidioides grown in vitro as spherules at 39°C at 10% CO2 at individual time points compared to hyphae grown at 25°C at the corresponding time points. The green and orange bars indicate the same sets of transcripts as marked in panel B. Heatmap colors indicate log2 differential expression ratios in units of counts per million. We then identified a set of 143 core spherule-enriched transcripts with increased abundance in both WT Coccidioides in the presence of macrophages and in vitro spherules (green box in Fig. 5B and green bar in Fig. 5C; Table S4). As proof of principle, we note this set included many of the previously identified spherule-associated genes such as SOWgp, PSP1, OPS1, PDC1, UAZ1, and TYR2 (16, 17). The 229 transcripts that showed higher abundance only in WT Coccidioides in the presence of macrophages (orange box in Fig. 5B and orange bar in Fig. 5C; Table S5) could represent virulence factors or other proteins required by Coccidioides during interaction with macrophages. During this analysis, we noted that not all transcripts showed a sustained increase in expression throughout the experiment. To further analyze the expression patterns, we examined the temporal and MOI-dependent expression of the core spherule-specific genes (Fig. 5C, upper/green bar) and macrophage-induced genes (Fig. 5C, lower/orange bar). The final subsets were independently hierarchically clustered and concatenated into a single heatmap (Fig. 5C). Half of the core genes were up in all three WT-with-BMDM conditions, with the other half up only in the day 1 samples. For the macrophage-induced genes, 70% were up only in the day 1 samples and not in the day 2 samples. This observation is intriguing because, in the context of macrophage co-culture, the developing spherules tended to be intracellular at day 1 and extracellular at day 2, suggesting that these fungal cells were in distinct environments. One of the most interesting transcripts with increased abundance in WT Coccidioides associated with BMDMs was D8B26_000085, which encodes a highly seroreactive protein that is found in infected human lungs (19). Antibodies to D8B26_000085 persist in the serum of naturally infected or vaccinated dogs (18), indicating that it is expressed and antigenic in animal infections. Other genes with similar patterns included D8B26_005548 (which contains an MFS domain), D8B26_000635 (a cytochrome c peroxidase), and D8B26_006970 (a presumed amidase). To focus on potential secreted effectors that were induced when Coccidioides was co-cultured with macrophages, we determined which transcripts were predicted to encode proteins with a signal peptide; 48 transcripts met this criterion, including SOWgp, TYR2, and D8B26_000085 (CPSG_05795/CIMG_05576). Next, since many fungal effectors in plant pathogens (30) and in Histoplasma (31) are small, cysteine-rich proteins, we further narrowed the list to small proteins that were cysteine-rich (<250 amino acids and ≥4 cysteines); 10 genes encoded proteins that fit these criteria, including the superoxide dismutase SOD3 (D8B26_006671), which is a virulence factor in Histoplasma (32); ELI1 (D8B26_007114), which is a protective antigen of unknown function (33); two small hypothetical proteins previously shown to be abundant in vivo (D8B26_007421 [CPSG_01366] and D8B26_005342 [CIMG_00509]); a predicted cutinase, which is expressed in the WT day 1 samples with BMDMs (D8B26_005437); and an ortholog of A. nidulans CalA that plays a role in conidial germination (D8B26_001315) (34). These transcripts may represent key molecules at the interface between the host and pathogen during Coccidioides infection. DISCUSSION As the endemic region for Coccidioides expands, putting more people at risk of disease, there is an urgent need for fundamental knowledge of how this pathogen responds to the host to cause disease. Here, we showed that phagocytosis of Coccidioides arthroconidia by macrophages induced development of the pathogenic form of Coccidioides, allowing the fungus to subvert these key innate immune cells to cause disease. Furthermore, we show that Coccidioides exposed to BMDMs expressed key genes for spherule growth as well as potential virulence factors. Finally, we have identified a core set of spherule-associated transcripts with higher abundance in spherules, independent of temperature and CO2 conditions. This set will be critical for identifying targets for diagnostics or vaccine candidates in the future. One distinguishing feature of Coccidioides in comparison to many other fungal pathogens is its ability to cause disease in immunocompetent individuals. The ability of Coccidioides arthroconidia to respond to macrophages by inducing spherule development and the subsequent difficulty of phagocytes to destroy the large spherule may contribute to the ability of this environmental fungus to defy the innate immune response. We have shown that after phagocytosis by macrophages, arthroconidia respond by forming spherules and that the average size of macrophage-associated spherules is larger than observed during in vitro spherule development. Coccidioides exposure resulted in macrophage cell lysis, suggesting that intracellular Coccidioides might cause a physical strain that provokes host cell death or that Coccidioides might actively trigger a host cell-death pathway to escape the macrophage. Alternatively, macrophage lysis might be an effort on the part of the host cell to injure the pathogen and limit infection or stimulate immune cell recruitment. Fungi and other pathogens have developed strategies to manipulate host cell death pathways to their advantage (35–38), and investigating this process in Coccidioides infections may elucidate key elements of pathogenesis. This interaction between the fungus and innate immune cells could contribute to disease progression and potentially to the dissemination of the organism to extra-pulmonary sites. Notably, the specifics of the host response, which are likely influenced by the initial interaction of the fungus with phagocytes, are likely to play a role in dissemination since it is known that polymorphisms in IFNγ, IL-12, and Dectin-1 are associated with the likelihood of disseminated coccidioidomycosis (39). This work represents the first transcriptional study of Coccidioides in the presence of macrophages. We overcame the technical challenges of a relatively low Coccidioides RNA to macrophage RNA ratio with efficient disruption of fungal cells, choice of time points, and multiplicity of infection, as well as increased depth of sequencing. One of the strengths of this study is that the conditions where WT Coccidioides grew as hyphae or as spherules had identical CO2 levels and temperature; thus, differential genes in the data set reflect factors associated with spherule development and host-pathogen environment rather than temperature or elevated CO2 alone. The identification of a core set of spherule-associated genes will be useful to our understanding of spherule biology as well as the development of Coccidioides vaccines. Additionally, we were able to observe nuance in the timing of gene expression. Coccidioides transcripts that are only expressed early in macrophage co-culture may reflect a germination program that gives rise to spherules, such as D8B26_001315, the ortholog of A. nidulans CalA. D8B26_001315 is up in early WT Coccidioides samples exposed to BMDMs as well as in early in vitro spherules. Additionally, it was striking that macrophages induced the increased abundance of transcripts encoding Coccidioides proteins that were previously observed in human lung samples, including D8B26_000085, one of the most immunogenic proteins encoded by Coccidioides. These observations suggest that macrophage-Coccidioides co-culture provides a simplified model system for analyzing important host-fungal interactions. An open question is the mechanism used by arthroconidia to sense the intracellular environment to induce the spherule form. The phagosome is a complex and dynamic environment that poses many potential challenges to enclosed pathogens, including changes in pH, exposure to reactive oxygen species, and limitation of nutrients. What role each of those processes plays in spherulation or inhibition of hyphal germination is unknown. Furthermore, spherule formation ultimately results in the release of internal cells called endospores, and the interaction of macrophages and endospores remains to be fully explored. An understanding of how these different Coccidioides cell types subvert the host immune response will provide strategies to leverage or modulate the host immune response to promote disease resolution instead of progression. MATERIALS AND METHODS Strains Most experiments were conducted with the WT C. posadasii strain Silveira (BEI # NR-48944), a kind gift from Dr. Bridget Barker, Northern Arizona University. We also utilized C. immitis strain RS (BEI # NR-48942) and ryp1Δ deletion mutant in the C. posadasii strain Silveira background (6, 14). Cell culture Bone marrow-derived macrophages (BMDMs) were derived from bone marrow obtained from the femurs and tibias of 6- to 8-week-old female C57BL/6 mice as previously described (40). Bone marrow cells were grown in bone marrow macrophage media (BMM) containing Dulbecco’s modified Eagle medium (DMEM high glucose) with 10% conditioned media from CMG 14-12 cells (source of m-CSF), 20% fetal bovine serum (FBS), 110 µg/mL penicillin and streptomycin, 2 mM L-glutamine, and 110 µg/mL sodium-pyruvate, for 7 days then frozen in 40% FBS and 10% DMSO for future use. For each experiment, an aliquot was thawed and plated to rest overnight before each experiment. Cultures were maintained at 37°C and 5% CO2. Live imaging Live imaging was performed on the Etaluma LS720 microscope with Etaluma camera with Aptina sensor using the Olympus phase contrast 40× objective with air lens (LCACHN40XIPC). Cells were maintained in a stage top incubator (Okolab) to control temperature, humidity, and CO2. For most time course experiments, brightfield images of 5–9 fields per well were taken once an hour. The number of spherules per field was quantified at the final time point. Spherule number per well was defined as the aggregate count of spherules in five fields per well. Spherule diameter was measured for all spherules in the field using Fiji (ImageJ2 version: 2.9.0). Time to first hyphae was quantified as the time when the first arthroconidia germinated into a hypha in the field of view or a hypha entered the field of view. Unless otherwise noted, experiments were performed with BMDMs in 48-well tissue culture-treated cell culture plates (Corning 3548), seeded at 0.75–1 × 105 macrophages per well and infected at an MOI of 0.1 arthroconidia per macrophage. Other MOIs utilized were 0.01 and 1. For transwell experiments, six-well transwell plates (Costar 3450) were seeded with macrophages at varying densities in the well or in the transwell. For the experiments with dead macrophages, BMDMs were heat-killed at 65°C for 5 min or methanol-fixed by resuspending in 70% methanol for 10 min, then spun for 5 min at 1,200 rpm and resuspended in fresh BMM before plating. Cells were plated at the same density as the live BMDMs (1 × 105 cells per well) or at twice the density (2 × 105 cells per well). Cytochalasin D (Sigma) was resuspended in DMSO and added at a final concentration of 10 μM to BMDM 30 min prior to the addition of arthroconidia and remained throughout the infection. Arthroconidia were grown in converse media with 10 µM cytochalasin D or DMSO in spherulation conditions (shaking culture at 39°C, 10% CO2) (5). Lactate dehydrogenase (LDH) release assay BMDMs were seeded (7.5 × 104 cells per well) in 48-well plates and infected at an MOI of 0.1 or 1 in triplicate as described above. At the indicated time points, supernatants were removed from triplicate wells and spun at 550 × g to pellet debris. The resulting supernatants were stored at 4°C until LDH quantification, as described previously (41). BMDM lysis was calculated as the percentage of total LDH from supernatant of wells of uninfected macrophages lysed in 1% Triton X-100 at the time of infection. The total LDH at later time points can be greater than the total LDH from the initial time point due to continued replication of BMDMs over the course of the experiment, resulting in an apparent lysis that is greater than 100%. RNA-seq analysis BMDMs were seeded 1 day prior to the infection in 6-well plates. On the day of infection, WT and ryp1Δ arthroconidia were added to the wells at an MOI of 0.1 or 1, and the plates were spun at 550 × g for 5 min. After 1 h, the media were removed and replaced with fresh BMM, except for the Coccidioides alone wells. At each time point, the cells were washed twice with PBS, and then QIAzol (Qiagen) was added to the wells. After 5 min, duplicate wells were then combined for each sample, with triplicate samples collected for each condition. Each sample underwent bead beating for 2 min with 0.5 mm zirconia/silica beads (BioSpec) to fully lyse fungal cells prior to RNA extraction. RNA was extracted using the Direct-zol RNA Miniprep Plus isolation kit (Zymo). Libraries were made with the NEBNext polyA mRNA magnetic isolation module and NEBNext Ultra II Directional RNA Library Prep kit with dual-indexed multiplexing barcodes (New England Biolabs) per the manufacturer’s instructions. Libraries were pooled, and sequencing was performed by the Chan-Zuckerberg BioHub-San Francisco on one lane of Novaseq S4 (Illumina). Transcript abundances were quantified based on the annotated C. posadasii Silveira genome (42), BioProject PRJNA664774. Relative abundances (reported as TPM values [43]) and estimated counts (est_counts) of each transcript in each sample were estimated by alignment-free comparison of k-mers between the reads and mRNA sequences using KALLISTO version 0.46.2 (44). Further analysis was restricted to transcripts with estimated counts ≥10 in at least nine samples. Differentially expressed genes were identified by comparing replicates for contrasts of interest using the glmQLFit, glmQLFTest, and topTags functions in edgeR version 4.0.16 (45). Genes were considered significantly differentially expressed if they were statistically significant (at 5% FDR) with an effect size of at least 2× (absolute log2 fold change ≥1) for a given contrast. We estimated contrasts for each of the WT and ryp1∆ infection conditions, Hi MOI (1) day 1 and Lo MOI (0.1) days 1 and 2, as well as ryp1∆ alone, all relative to WT alone, which grew as hyphae (D2 hyp). Transcripts were clustered on these seven contrasts using maximum linkage hierarchical clustering with uncentered Pearson distances as implemented in Bio.Cluster in BIOPYTHON 1.83 (46). To compare macrophage-induced gene expression changes to those induced by in vitro spherulation conditions, we utilized data from Homer et al., 2025 (6). For the in vitro comparisons, we contrasted arthroconidia grown in spherulation conditions (39°C, 10% CO2, in Converse media, shaking) and in vitro hyphae conditions (30°C, ambient CO2, in Converse media, shaking) on days 1, 2, and 3. For the comparisons with macrophages, we contrasted the gene expression of our WT and ryp1Δ Coccidioides in the presence of BMDMs (Hi MOI D1, Lo MOI D1, and Lo MOI D2) to the same strain grown alone at day 2 in the hyphal form (D2 hyp). Genes were considered induced in each set if they were significantly different for at least two of the three conditions in that set.