The Cryptococcus neoformans Rim101 transcription factor directly regulates genes required for adaptation to the host

Abstract

The Rim101 protein is a conserved pH-responsive transcription factor that mediates important interactions between several fungal pathogens and the infected host. In the human fungal pathogen Cryptococcus neoformans, the Rim101 protein retains conserved functions to allow the microorganism to respond to changes in pH and other host stresses. This coordinated cellular response enables this fungus to effectively evade the host immune response. Preliminary studies suggest that this conserved transcription factor is uniquely regulated in C. neoformans both by the canonical pH-sensing pathway and by the cyclic AMP (cAMP)/protein kinase A (PKA) pathway. Here we present comparative transcriptional data that demonstrate a strong concordance between the downstream effectors of PKA and Rim101. To define Rim101-dependent gene expression during a murine lung infection, we used nanoString profiling of lung tissue infected with a wild-type or rim101Δ mutant strain. In this setting, we demonstrated that Rim101 controls the expression of multiple cell wall-biosynthetic genes, likely explaining the enhanced immunogenicity of the rim101Δ mutant. Despite its divergent upstream regulation, the C. neoformans Rim101 protein recognizes a conserved DNA binding motif. Using these data, we identified direct targets of this transcription factor, including genes involved in cell wall regulation. Therefore, the Rim101 protein directly controls cell wall changes required for the adaptation of C. neoformans to its host environment. Moreover, we propose that integration of the cAMP/PKA and pH-sensing pathways allows C. neoformans to respond to a broad range of host-specific signals.

FIG 1

(A) Pairwise correlation analysis of the entire transcriptomes of the rim101Δ and pka1Δ mutant strains. Gene expression levels for the two mutant strains were determined in comparison with those for the wild-type strain. The transcriptomes of the pka1Δ, rim101Δ, and wild-type strains were defined using RNA-Seq analysis after incubation in tissue culture medium (DMEM) for 3 h. Shaded circles represent genes that were <2-fold differentially expressed in the wild-type and mutant strains. The adjusted r² for the entire transcriptomes was 0.515 (P, <0.001). The adjusted r² for the comparison of the significantly differentially expressed genes only was 0.923 (P, <0.001). (B) The concordance between Rim101- and Pka1-dependent gene expression supports a model in which the Rim101 transcription factor is controlled by the classical Rim/pH-responsive pathway (including Rim20 and Rim13) as well as by the cAMP/Pka1 pathway to regulate several cellular processes.

FIG 2

The rim101Δ and pka1Δ strains have both overlapping and divergent phenotypes. (A) The rim101Δ and pka1Δ mutants demonstrate similar alterations in WGA binding patterns. Cells were incubated for 24 h in tissue culture medium before staining with FITC-conjugated WGA. All micrographs were taken at the same exposure in order to distinguish the fluorescence levels of the different strains. (B) The rim101Δ and rim20Δ mutants have defects at an alkaline pH and under high-salt conditions. A total of 1 × 10⁶ cells were 10-fold serially diluted onto plates under the conditions indicated and were incubated at 30°C. (C) ENA1 shows Rim101- but not Pka1-dependent changes in expression. RNA was extracted from wild-type, rim101Δ, and pka1Δ strains after incubation for 3 h in a rich medium (YPD), under tissue culture conditions, or with high salt concentrations. ENA1 transcript levels were determined by quantitative reverse transcriptase PCR. The fold change was calculated relative to wild-type levels in YPD and was normalized to the expression of the internal control, GPD1.

FIG 3

(A to C) Rim101 binds a conserved motif, as determined by EMSAs. (A) A biotin-labeled 25-mer containing the 5′-GCCAAG-3′ motif was incubated with protein extracts from wild-type (lanes 1 and 4) or rim101Δ (lanes 2 and 3) cells. The mixtures were assessed by alterations in electrophoretic mobility by PAGE and immunoblotting using streptavidin detection. Excess unlabeled 15-mers were added to lanes 3 and 4. (B and C) A 25-mer containing the 5′-GCCAAG-3′ motif (B) or a mutated Rim101 binding motif (5′-GAGAAG-3′) (C) was incubated with protein extracts from the wild-type (lanes 1), rim101Δ (lanes 2), or rim101Δ/GFP-RIM101 (lanes 3) strain prior to electrophoresis. (D) The promoter of the ENA1 gene from C. neoformans var. grubii and its homologous sequence in C. gattii were aligned. The conserved Rim101 binding motif is highlighted. Identical nucleotides in a consensus (cons) sequence of this region are indicated by asterisks. (E) Genes with Rim101 binding sites in both C. neoformans and C. gattii were examined for Rim101-dependent transcription and were separated on the basis of induction (green) or repression (red). The positions of the binding sites were determined, and the number of genes was plotted along a representation of the distance of the Rim101 binding site from the ATG translation start site. All sequences and Rim101 site positions were obtained from FungiDB (www.fungidb.org).

FIG 4

Chromatin immunoprecipitation to detect Gfp-Rim101 DNA binding. Cells were incubated in tissue culture medium for 3 h before fixation and chromatin immunoprecipitation. PCRs were performed using primers that flanked a presumed Rim101 binding site in target gene promoters. The fold change was determined by determining enrichment in the immunoprecipitated sample relative to the no-antibody control. Actin (ACT1), which does not contain a Rim101 binding site, was used as a control.

FIG 5

Differences in gene expression and localization due to induction in tissue culture medium. (A) Normalized RNA levels of genes with differential expression in the wild-type and rim101Δ mutant strains. Strains were grown in YPD or under tissue culture (TC) conditions for 3 h before RNA extraction. RNA levels were determined by nanoString profiling. RNA levels were normalized to those of control genes as described in Materials and Methods. (B) Rim101 shows increased nuclear localization in tissue culture medium. Cells containing the Gfp-Rim101 fusion protein expressed under the control of the endogenous promoter were incubated in YPD medium (pH 6.0) or in tissue culture medium. Nuclei were stained using the Hoechst 33342 nucleic acid stain.

FIG 6

In vivo profiling of gene expression. RNA was harvested from mouse lungs (from 5 mice for each strain) infected with either the wild-type (open bars) or the rim101Δ mutant (shaded bars) strain. Expression levels were determined by nanoString profiling of candidate genes. The graph includes all genes with significantly different expression in the wild-type and rim101Δ strains (P, <0.05).