Transcriptional Landscape and Regulatory Roles of Small Noncoding RNAs in the Oxidative Stress Response of the Haloarchaeon Haloferax volcanii

Abstract

Haloarchaea in their natural environment are exposed to hypersalinity, intense solar radiation, and desiccation, all of which generate high levels of oxidative stress. Previous work has shown that haloarchaea are an order of magnitude more resistant to oxidative stress than most mesophilic organisms. Despite this resistance, the pathways haloarchaea use to respond to oxidative stress damage are similar to those of nonresistant organisms, suggesting that regulatory processes might be key to their robustness. Recently, small regulatory noncoding RNAs (sRNAs) were discovered in Archaea under a variety of environmental conditions. We report here the transcriptional landscape and functional roles of sRNAs in the regulation of the oxidative stress response of the model haloarchaeon Haloferax volcanii Thousands of sRNAs, both intergenic and antisense, were discovered using strand-specific sRNA sequencing (sRNA-seq), comprising 25 to 30% of the total transcriptome under no-challenge and oxidative stress conditions, respectively. We identified hundreds of differentially expressed sRNAs in response to hydrogen peroxide-induced oxidative stress in H. volcanii The targets of a group of antisense sRNAs decreased in expression when these sRNAs were upregulated, suggesting that sRNAs are potentially playing a negative regulatory role on mRNA targets at the transcript level. Target enrichment of these antisense sRNAs included mRNAs involved in transposon mobility, chemotaxis signaling, peptidase activity, and transcription factors.IMPORTANCE While a substantial body of experimental work has been done to uncover the functions of small regulatory noncoding RNAs (sRNAs) in gene regulation in Bacteria and Eukarya, the functional roles of sRNAs in Archaea are still poorly understood. This study is the first to establish the regulatory effects of sRNAs on mRNAs during the oxidative stress response in the haloarchaeon Haloferax volcanii Our work demonstrates that common principles for the response to a major cellular stress exist across the 3 domains of life while uncovering pathways that might be specific to the Archaea This work also underscores the relevance of sRNAs in adaptation to extreme environmental conditions.

Keywords: archaea; extreme environments; noncoding RNA; oxidative stress; small RNA; transcriptional regulation.

FIG 1

Genome viewer of antisense sRNAs (cis acting) (A) and intergenic sRNAs (trans acting) (B). Paired-end reads (100 bases) were mapped to the H. volcanii NCBI reference genome. Reference genes are marked as black lines with arrowheads indicating their location on the plus strand (>) or minus strand (<). Reads marked in red are transcribed from the minus strand, while blue reads are transcribed from the plus strand. Untranslated regions were predicted using Rockhopper2 (pink lines). Green lines mark discovered sRNAs. Coverage plots are in gray.

FIG 2

Number of sRNAs (total, antisense, and intergenic) discovered under no-challenge and H₂O₂ challenge conditions.

FIG 3

Transcript per million (TPM) expression levels between sRNAs and their putative cis-mRNA targets during oxidative stress. Each point represents the TPM ratio between an sRNA and its putative cis-mRNA target. TPM values are the averages from sRNA and mRNA replicates with error bars representing standard deviations among replicates. A pairwise t test was conducted between sRNA and putative cis-mRNA target replicates to infer significant difference in TPM expression between the cis pairs. Orange points indicate a P value <0.05, and gray points indicate a P value >0.05. The red line represents no change in the ratio of sRNA–cis-mRNA expression (slope = 1).

FIG 4

(A) Heat map of log₂-transformed fold change of differentially expressed antisense sRNAs (asRNAs). (B) Differential expression fold changes of upregulated asRNAs and their putative cis-mRNA targets (averages and standard deviation error bars from replicates). (C) Differential expression fold changes of downregulated asRNAs and their putative cis-mRNA targets (averages and standard deviation error bars from replicates). (D) Differential expression fold changes of all transposase-targeting asRNAs and their putative cis-transposase mRNA targets (averages and standard deviation error bars from replicates). *, both the asRNA and the cis-target transposase mRNAs have significant differential expression based on an FDR of <0.05.

FIG 5

Distribution of binding regions for antisense sRNAs. UTR, untranslated region; CDS, coding sequence.

FIG 6

Validation of differentially expressed sRNAs by Northern blots. (A) Representative Northern blot confirming sizes and differential expression patterns of an intergenic sRNA during oxidative stress. (B) Quantification of Northern blots confirming the expression of intergenic sRNAs (random primed labeling) and strand specificity of sRNAs (oligonucleotide labeling). All classes of sRNAs were confirmed: antisense (5′ UTR, 3′ UTR, CDS) and intergenic sRNAs.

FIG 7

Gene ontology enrichment analysis identifying the functional classification of gene targets of sRNAs during oxidative stress. (A) Enriched target gene functions for upregulated sRNAs. (B) Enriched target gene functions for downregulated sRNAs.

FIG 8

Distribution of differentially expressed genes during oxidative stress in H. volcanii. (A) MA plot of differentially expressed genes; each point represent a gene. Significant (FDR < 5%) differentially expressed sRNAs are color coded as upregulated (blue) or downregulated (red) and known oxidative stress response genes (yellow). (B) Gene function for the most up- and downregulated sRNAs.