A non-canonical RNAi pathway controls virulence and genome stability in Mucorales

Abstract

Epimutations in fungal pathogens are emerging as novel phenomena that could explain the fast-developing resistance to antifungal drugs and other stresses. These epimutations are generated by RNA interference (RNAi) mechanisms that transiently silence specific genes to overcome stressful stimuli. The early-diverging fungus Mucor circinelloides exercises a fine control over two interacting RNAi pathways to produce epimutants: the canonical RNAi pathway and a new RNAi degradative pathway. The latter is considered a non-canonical RNAi pathway (NCRIP) because it relies on RNA-dependent RNA polymerases (RdRPs) and a novel ribonuclease III-like named R3B2 to degrade target transcripts. Here in this work, we uncovered the role of NCRIP in regulating virulence processes and transposon movements through key components of the pathway, RdRP1 and R3B2. Mutants in these genes are unable to launch a proper virulence response to macrophage phagocytosis, resulting in a decreased virulence potential. The transcriptomic profile of rdrp1Δ and r3b2Δ mutants revealed a pre-exposure adaptation to the stressful phagosomal environment even when the strains are not confronted by macrophages. These results suggest that NCRIP represses key targets during regular growth and releases its control when a stressful environment challenges the fungus. NCRIP interacts with the RNAi canonical core to protect genome stability by controlling the expression of centromeric retrotransposable elements. In the absence of NCRIP, these retrotransposons are robustly repressed by the canonical RNAi machinery; thus, supporting the antagonistic role of NCRIP in containing the epimutational pathway. Both interacting RNAi pathways might be essential to govern host-pathogen interactions through transient adaptations, contributing to the unique traits of the emerging infection mucormycosis.

Fig 1. NCRIP regulates a vast gene network via the cooperation of R3B2 and RdRP1.

(A) Diagram of the experimental design followed to perform macrophage phagocytosis and saprophytic assays for RNA sequencing. (B) Principal component (PC) analysis biplot of transcript abundances (measured as counts per million [CPM], mean CPM > 1.0), showing the variability across the color- and shape-coded samples. (C) Heatmap of significant differentially expressed genes (DEGs, log₂ FC ≥ |1.0|, FDR ≤ 0.05) in the depicted NCRIP mutant strains and conditions compared to the wild-type strain in the same condition as a log₂ ratio. Differential expression values and experimental samples are clustered by similarity. All DEGs responding exclusively in the saprophytic condition in both NCRIP mutants are clustered in green. DEGs in at least one NCRIP mutant responding to macrophage phagocytosis are depicted in purple. Differential expression values (in log₂ CPM) are color-coded in red for upregulation and blue for downregulation (ns stands for non-significant changes).

Fig 2. NCRIP regulates key functional categories involved in saprophytic growth.

Enrichment analysis of DEGs in each Eukaryotic Orthologous Groups (KOG) class. Significant enrichments (Fisher’s exact test, P ≤ 0.05) in a given mutant strain and condition compared to the wild-type strain are shown as uplifted rectangles. A measure of up- (red) or downregulation (blue) of each KOG class is represented as a colored scale of delta-rank values (the difference between the mean rank differential expression value of all genes in each KOG class and the mean rank differential expression value of all other genes). KOG classes and experimental conditions (mutant strains and presence/absence of macrophages) are clustered according to the similarity of their delta rank values.

Fig 3. M. circinelloides coordinates an NCRIP-dependent virulent response to phagocytosis.

Heatmap of DEGs (log₂ FC ≥ |1.5|, FDR ≤ 0.05) in the wild-type strain and the NCRIP mutant strains during phagocytosis compared to their saprophytic control conditions as a log₂ ratio. Red and blue represent up- and downregulated genes, respectively (ns stands for non-significant changes). Differential expression values (in log₂ CPM) are clustered by similarity. DEGs responding to macrophage phagocytosis in the wild-type strain but not in the NCRIP mutants are clustered together and colored in purple, constituting the NCRIP-dependent virulent response.

Fig 4. NCRIP controls the response to macrophage phagocytosis by inhibiting its targets under non-stressful conditions.

(A) The NCRIP-dependent virulent response is depicted in a heatmap. Genes are clustered by similarity to compare the response to phagocytosis in the wild-type strain and the response of the NCRIP mutants in saprophytic conditions. This comparison reveals a pre-exposure adaptation of the NCRIP mutants (magenta cluster), showing similar expression values than the wild-type response to phagocytosis. Exclusive genes of the wild-type response are clustered together (orange cluster). Differential expression values (in log₂ CPM) are color-coded to depict the degree of upregulation (red) or downregulation (blue) in each condition (ns stands for non-significant changes). (B) Bar plot of atf1, atf2, pps1 and aqp1 expression differences in r3b2Δ, and rdrp1Δ mutant strains compared with the wild-type strain in non-stressful conditions, i.e., incubation in cell-culture medium without macrophages for 5 hours. Log₂ fold-change differential expression levels were quantified by RT-qPCR and normalized using rRNA 18S as an internal control. Error bars correspond to the SD of technical triplicates and significant differences are denoted by asterisks (* for P ≤ 0.05, ** for P ≤ 0.005, and *** for P < 0.0001 in an unpaired t-test). (C) KOG class enrichment analysis of the NCRIP-dependent virulent response during macrophage interaction is grouped in pre-exposure adaptation in NCRIP mutants and wild-type specific responses. Significant enrichments (Fisher’s exact test, P ≤ 0.05) are shown as uplifted rectangles. A measure of up- (red) or downregulation (blue) of each KOG class is represented as a colored scale of delta-rank values (the difference between the mean rank differential expression value of all genes in a particular KOG class and the mean rank differential expression value of all other genes). KOG classes are clustered according to the similarity of their delta rank values.

Fig 5. NCRIP is involved in oxidative stress tolerance and mucormycosis.

(A) A bar plot showing the survival rates of the NCRIP mutants (r3b2Δ and rdrp1Δ) compared to a wild-type strain under oxidative stress. Survival assays were performed in two different minimal media (YNB and MMC) supplemented with two different H₂O₂ concentrations: 5 mM and 10 mM. Error bars correspond to the SD of technical triplicates and significant differences in survival rates were denoted by asterisks (* for P ≤ 0.05, ** for P ≤ 0.005, and *** for P < 0.0001 in a two-way ANOVA with Tukey’s multiple comparison test). (B) The virulence of r3b2Δ and rdrp1Δ mutant strains was assessed in a survival assay using immunosuppressed mice as a mucormycosis model. Groups of ten mice were infected intravenously with 1x10⁶ spores from each strain (color-coded). Survival rates were statistically analyzed for significant differences (P ≤ 0.05 in a Mantel-Cox test) compared with a virulent control strain (R7B). NRRL3631 was used as an avirulent mock control of infection.

Fig 6. NCRIP competes with the epimutational pathway to regulate transposable elements.

(A) A genomic view of centromeric chromatin (CEN4) displaying the kinetochore-binding region enrichment that marks the centromere (CEN, blue), annotation of transposable elements (colored blocks), and transcriptomic data of sRNAs (red) in M. circinelloides wild-type, canonical pathway (ago1Δ, double dcl1Δ/dcl2Δ) and NCRIP (rdrp1Δ and r3b2Δ) deletion mutant strains after 48 h of growth in rich medium. sRNA values are normalized to bins per million (BPM) mapped reads. (B) Heatmap of the differential sRNA accumulation targeting Genomic retrotransposable elements of Mucoromycotina LINE1-like (Grem-LINE1s) in the depicted RNAi mutants compared to the RNAi-proficient wild-type strain. Grem-LINE1s are numbered according to Navarro-Mendoza et al. classification [12]. Differential sRNA values (in log₂ CPM) are normalized by the trimmed mean of M values (TMM).

Fig 7. Transposable element sRNAs exhibit typical characteristics of the canonical RNAi pathway biogenesis.

(A) Representative diagram of sRNA production in the canonical and NCRI pathways. The canonical pathway is represented in two loci: Grem-LINE1, and a Serine/Threonine kinase (JGI Muccir1_3 ID: 1455000) [17]. NCRIP is exemplified by an Alkaline phosphatase (JGI Muccir1_3 ID: 1469159) [10]. Sense (positive values) and antisense (negative values) sRNAs accumulation is shown; sRNA values are normalized to bins per million (BPM) mapped reads. Grem-LINE1 open reading frames (ORF1 and ORF2 as green and red arrows, respectively) and protein domains predicted from its coding sequence are shown as colored blocks (zf-RVT, zinc-binding in reverse transcriptase [PF13966]; RVT, reverse transcriptase [PF00078]; AP, AP endonuclease [PTHR22748]; and ZF, zinc finger [PF00098 and PF16588]). (B) Sequence analysis of sRNAs mapping to the three loci shown in (A), produced in a wild-type, dcl1Δ/dcl2Δ, and r3b2Δ mutant strains. Each sequence analysis depicts a probability logo (showing the first and last five nucleotides) and a size distribution bar plot.