Functional Transcriptomics for Bacterial Gene Detectives

Abstract

Developments in transcriptomic technology and the availability of whole-genome-level expression profiles for many bacterial model organisms have accelerated the assignment of gene function. However, the deluge of transcriptomic data is making the analysis of gene expression a challenging task for biologists. Online resources for global bacterial gene expression analysis are not available for the majority of published data sets, impeding access and hindering data exploration. Here, we show the value of preexisting transcriptomic data sets for hypothesis generation. We describe the use of accessible online resources, such as SalComMac and SalComRegulon, to visualize and analyze expression profiles of coding genes and small RNAs. This approach arms a new generation of "gene detectives" with powerful new tools for understanding the transcriptional networks of Salmonella, a bacterium that has become an important model organism for the study of gene regulation. To demonstrate the value of integrating different online platforms, and to show the simplicity of the approach, we used well-characterized small RNAs that respond to envelope stress, oxidative stress, osmotic stress, or iron limitation as examples. We hope to provide impetus for the development of more online resources to allow the scientific community to work intuitively with transcriptomic data.

FIGURE 1

Bacterial functional transcriptomics is facilitated by RNA-seq technology. The development of RNA-seq has expanded the range of transcriptome-based techniques that address a variety of biological questions. DROP-seq, RNA-seq of single cells compartmentalized in a droplet; scRNA, single-cell RNA-seq; dRNA-seq, differential RNA-seq; Term-seq, global mapping of 3′ ends of transcripts; ChIP-seq, chromatin immunoprecipitation followed by sequencing; RIP-seq, native RNA immunoprecipitation followed by RNA-seq; GRAD-seq, gradient profiling by RNA-seq; TraDIS, transposon-directed insertion site sequencing; Tn-seq, transposon sequencing. See reference for more details of these techniques. Image by Eliza Wolfson (https://lizawolfson.co.uk) is used under the terms of a creative commons CC-BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode).

FIGURE 2

Environmental and genetic regulation of four sRNAs that are iron responsive and/or induced by oxidative stress. Gene expression data are presented for the sRNAs OxyS, RyhB-1, RyhB-2, and STnc3080 (these data can be visualized online at https://tinyurl.com/ya7s466m and https://tinyurl.com/yb5wz7dt). Data are shown as differential expression profiles involving six discrete heat-map blocks, each block being normalized to the condition on the left-hand side. The heat maps show differential expression, a strategy that lacks accuracy when expression levels are extremely low. Absolute (A) and relative (B) expression levels of S. Typhimurium grown under 21 different conditions (SalComMac). (C) Relative expression levels of the wild-type (WT) and mutant S. Typhimurium 4/74 grown under different conditions (SalComRegulon). Before experimental validation is considered, it should be ensured that the levels of absolute expression of particular sRNAs are above the expression threshold of 10 TPM units (–37). EEP, early exponential phase; MEP, mid-exponential phase; LEP, late exponential phase; ESP, early stationary phase; LSP, late stationary phase; InSPI2, SPI2-inducing minimal media.

FIGURE 3

Environmental and genetic regulation of four sRNAs involved in the envelope stress response. Gene expression data are shown for the sRNAs RybB, RyeF, MicA, and RprA (these data can be visualized online at https://tinyurl.com/y9mskb6j and https://tinyurl.com/ybnr6jja). Panels A, B, and C are as described in the legend to Fig. 2.

FIGURE 4

Environmental and genetic regulation of six sRNAs that respond to oxygen or osmolarity. Gene expression data are shown for the sRNAs FnrS, MicA, SraL, MntS (RybA), STnc1330, and STnc4260 (these data can be visualized online at https://tinyurl.com/yat8qrql and https://tinyurl.com/y8f533gy). Panels A, B, and C are as described in the legend to Fig. 2.

FIGURE 5

Visualization of the STnc1330 sRNA transcript. RNA-seq reads are mapped to the S. Typhimurium 4/74 genome (plus strand), showing STnc1330 expression under different conditions (35, 37). (A) MEP, anaerobic shock, and NaCl shock (https://tinyurl.com/STnc1330-NaCl); (B) InSPI2 and InSPI2 low Mg2⁺ (https://tinyurl.com/STnc1330-LowMg); (C) WT InSPI2 versus ΔphoPQ (https://tinyurl.com/STnc1330-PhoPQ); (D) WT LSP versus ΔrpoS (https://tinyurl.com/STnc1330-RpoS). Height of colored tracks represents the normalized sequencing reads at that locus (scale, 0 to 100). All arrows indicate the direction of transcription; TSSs are indicated by bent arrows and predicted Rho (ρ)-independent terminators are denoted by stem-loop structures.