A temperature-responsive network links cell shape and virulence traits in a primary fungal pathogen

Abstract

Survival at host temperature is a critical trait for pathogenic microbes of humans. Thermally dimorphic fungal pathogens, including Histoplasma capsulatum, are soil fungi that undergo dramatic changes in cell shape and virulence gene expression in response to host temperature. How these organisms link changes in temperature to both morphologic development and expression of virulence traits is unknown. Here we elucidate a temperature-responsive transcriptional network in H. capsulatum, which switches from a filamentous form in the environment to a pathogenic yeast form at body temperature. The circuit is driven by three highly conserved factors, Ryp1, Ryp2, and Ryp3, that are required for yeast-phase growth at 37°C. Ryp factors belong to distinct families of proteins that control developmental transitions in fungi: Ryp1 is a member of the WOPR family of transcription factors, and Ryp2 and Ryp3 are both members of the Velvet family of proteins whose molecular function is unknown. Here we provide the first evidence that these WOPR and Velvet proteins interact, and that Velvet proteins associate with DNA to drive gene expression. Using genome-wide chromatin immunoprecipitation studies, we determine that Ryp1, Ryp2, and Ryp3 associate with a large common set of genomic loci that includes known virulence genes, indicating that the Ryp factors directly control genes required for pathogenicity in addition to their role in regulating cell morphology. We further dissect the Ryp regulatory circuit by determining that a fourth transcription factor, which we name Ryp4, is required for yeast-phase growth and gene expression, associates with DNA, and displays interdependent regulation with Ryp1, Ryp2, and Ryp3. Finally, we define cis-acting motifs that recruit the Ryp factors to their interwoven network of temperature-responsive target genes. Taken together, our results reveal a positive feedback circuit that directs a broad transcriptional switch between environmental and pathogenic states in response to temperature.

Figure 1. Ryp proteins are required for expression of yeast-phase specific genes.

(A) Transcriptional profile comparisons of wild-type cells grown at 37°C (WT-37), wild-type cells grown at room temperature (WT-RT), and ryp1, ryp2, and ryp3 mutants grown at 37°C represented as a heatmap. For a complete list of YPS and FPS genes see Table S1 (see Materials and Methods for bioinformatics analysis). (B) Select set of YPS genes involved in virulence is shown as a heatmap. Log2-based color scale is shown. (C) Phase-specific genes listed in Table S1 were subjected to analysis for GO term enrichment. Top five GO terms and enrichment p values are shown for both YPS and FPS genes. (D) Correlation between transcriptional profiles of YPS and FPS genes in ryp mutants grown at 37°C and wild-type cells grown at RT are presented as scatter plots with Pearson correlation coefficients (r).

Figure 2. Ryp factors associate upstream of similar sets of genes.

A network view of ChIP targets was generated using Cytoscape. Each large circle is composed of individual closed circles, each of which represents an individual target gene. YPS genes are colored in yellow, and FPS genes are colored in blue. The distribution of YPS and FPS genes in each category of ChIP targets (e.g., shared targets of Ryp1, 2, and 3) was compared to the whole genome (Figure S5) using the Wilcoxon test. The p values obtained in this analysis are shown. Additionally, targets shared by Ryp1 and Ryp3, as well as targets shared by all three Ryp factors, were enriched for YPS genes (p value <0.001) as determined by hypergeometric tests performed in R.

Figure 3. Ryp factors associate upstream of genes regulating morphology and virulence.

ChIP-chip enrichment ratio versus genome coordinate for Ryp1, Ryp2, Ryp3, and Ryp4 ChIP events were plotted using Mochiview. Genome coordinates were adjusted to the start of the ORF of interest. Motif A (red triangles) and Motif B (blue triangles) locations shown in these loci were identified using MAST. Genomic regions of RYP genes (A) and virulence genes (B) are shown. Asterisks denote Motif A and Motif B probes used in the mobility shift assays in Figure 6C and 6D.

Figure 4. Ryp4 is required for yeast-phase growth.

(A) ChIP-chip enrichment ratio versus genome coordinate for Ryp1, Ryp2, Ryp3, and Ryp4 ChIP events are shown at the RYP4 locus. Genome coordinates were adjusted to the start of the ORF. Motif A (red triangles) and Motif B (blue triangles) locations shown in these loci were identified using MAST. (B, C) qRT-PCR was used to quantify relative levels of RYP4 transcript in (B) ryp1, ryp2, ryp3 insertion mutants and wild-type and (C) ryp4 knockdown mutants and wild-type controls grown at 37°C and RT. GAPDH was used as a normalizer gene. qRT-PCRs were performed with at least two biological replicates for each strain. Triplicate measurements of representative replicates are graphed as mean ± standard deviation. (D) Phase-contrast microscopy images of ryp1, ryp2, ryp3, and ryp4 mutants and wild-type controls grown at 37°C and RT are shown. Black bar equals 20 µm. (E) Transcriptional profile comparisons of wild-type grown at 37°C (WT_37), wild-type grown at room temperature (WT_RT), and ryp1, ryp2, ryp3, and ryp4 mutants grown at 37°C represented as a heatmap. Phase-specific (YPS and FPS) genes are listed in Table S3 (see Materials and Methods for bioinformatics analysis). (F) Different categories of ChIP targets are shown as circles. Ryp1, Ryp2, and Ryp3 shared targets were split into two categories depending on the presence or absence of a Ryp4 ChIP event. The distribution of YPS and FPS genes in each category was compared to each other using the Wilcoxon test. The p value obtained in this analysis is shown. In this comparison, Ryp1, Ryp2, Ryp3, and Ryp4 shared targets are further enriched for YPS genes (p value <0.001) as determined by hypergeometric tests performed in R.

Figure 5. Ryp1, Ryp2, and Ryp3 physically interact.

(A) Co-IP experiments were performed in wild-type cells grown at 37°C using Ryp2 and Ryp3 antibodies. Immunoblot shows detection of Ryp1, Ryp2, and Ryp3 in the input and elution fractions of Ryp2 and Ryp3 IPs. (B, C) Yeast-two-hybrid matings were performed with bait and prey strains listed. Diploid cells were spotted onto minimal media plates: -UHT, -Uracil/-Histidine/-Tryptophan; –UHTL, -Uracil/-Histidine/-Tryptophan/-Leucine. Growth on -UHTL plates and blue color on -UHT+X-gal plates indicate an interaction between proteins expressed in bait and prey strains. β-galactosidase activities were measured for three independent isolates of each strain. Quadruplicate measurements of representative isolates are graphed as mean ± standard deviation.

Figure 6. Ryp factors bind to specific cis-acting regulatory elements.

(A) Motif A and Motif B logos are shown. (B) Motif specificity of Motif A, Motif B, and their shuffled versions were analyzed by ROC plots. True positive rate was defined as motifs found in all ChIP events. False positive rate was defined as motifs found in unbound intergenic regions. (C, D) His-tagged Ryp proteins were generated by a coupled in vitro transcription-translation reaction in wheat germ extracts, and purified using Ni-NTA beads. Extracts with no template DNA were processed similarly and included as controls. 60-bp fragments from the CBP1 promoter were used as Motif A and Motif B probes (Figure 3B). 1 nM labeled probe and 2 µg of purified protein were used for each binding reaction. Binding to labeled Motif A or Motif B probes was competed by adding unlabeled probe in 10-fold, 50-fold, and 100-fold molar excess of the labeled probe.

Figure 7. Ryp factors can drive gene expression using specific cis-acting regulatory elements.

(A) Motif A (B, C) Motif B, and their mutated versions were cloned into UAS-less CYC1 promoter fused to lacZ gene. Point mutations made in the motifs are shown in red. RYP genes were cloned into two different vectors with two different markers. Each of these vectors were transformed into motif-containing yeast strains as labeled. β-galactosidase activities were measured for three independent isolates of each strain. Quadruplicate measurements of representative isolates are graphed as mean ± standard deviation.

Figure 8. Ryp proteins are involved in a temperature-responsive complex transcriptional network.

A model for the temperature-responsive Ryp network is shown. ChIP-chip interactions are depicted by solid arrows and indirect regulation is depicted by a dashed arrow. All four Ryp proteins are expressed at increased levels at 37°C, require each of the others for their expression, and associate with the promoter regions of RYP1, RYP2, and RYP4. Even though none of the Ryp proteins associate with RYP3 promoter, RYP3 expression levels depend on the other Ryp proteins. De novo motif finding approaches and in vivo transcriptional activation assays reveal that Ryp1 is necessary and sufficient to drive gene expression using Motif A, and Ryp2-Ryp3 co-expression is necessary and sufficient to drive gene expression using Motif B. We hypothesize that YPS gene expression in response to growth at host temperature can be achieved through many possible combinations of motifs and Ryp factor association. Three such scenarios are shown.