Integration of Transcriptomic and Proteomic Approaches Reveals the Temperature-Dependent Virulence of Pseudomonas plecoglossicida

Abstract

Pseudomonas plecoglossicida is a facultative pathogen that is associated with diseases of multiple fish, mainly at 15-20°C. Although fish disease caused by P. plecoglossicida has led to significant economic losses, the mechanisms of the temperature-dependent virulence are unclear. Here, we identify potential pathogenicity mechanisms and demonstrate the direct regulation of several virulence factors by temperature with transcriptomic and proteomic analyses, quantitative real-time PCR (qRT-PCR), RNAi, pyoverdine (PVD) quantification, the chrome azurol S (CAS) assay, growth curve measurements, a biofilm assay, and artificial infection. The principal component analysis, the heat map generation and hierarchical clustering, together with the functional annotations of the differentially expressed genes (DEGs) demonstrated that, under different growth temperatures, the animation and focus of P. plecoglossicida are quite different, which may be the key to pathogenicity. Genes involved in PVD synthesis and in the type VI secretion system (T6SS) are specifically upregulated at the virulent temperature of 18°C. Silencing of the PVD-synthesis-related genes reduces the iron acquisition, growth, biofilm formation, distribution in host organs and virulence of the bacteria. Silencing of the T6SS genes also leads to the reduction of biofilm formation, distribution in host organs and virulence. These findings reveal that temperature regulates multiple virulence mechanisms in P. plecoglossicida, especially through iron acquisition and T6SS secretion. Meanwhile, integration of transcriptomic and proteomic data provide us with a new perspective into the pathogenesis of P. plecoglossicida, which would not have been easy to catch at either the protein or mRNA differential analyses alone, thus illustrating the power of multi-omics analyses in microbiology.

Keywords: Pseudomonas plecoglossicida; proteomics; temperature; transcriptome; virulence; white nodules disease.

Figure 1

PCA analysis (A), hierarchical clustering of all genes (B) and the DEGs (C), and numbers of DEGs (D). For each temperature treatment, there are three replicates (12-1, 12-2, 12-3; 18-1, 18-2, 18-3; 28-1, 28-2, 28-3). The plot demonstrates clear separation of the 3 sample groups (A). For hierarchical clustering, green and red indicate decreased and increased expression, respectively. Transcripts were clustered by hierarchical clustering using the complete linkage algorithm and Pearson correlation metric in R.

Figure 2

Functional annotation of DEGs based on GO and KEGG database. Histogram presentation of clusters of GO classification of DEGs in 12°C (A) and 28°C group (B). Distributions of the KEGG pathways in 12°C (C), and 28°C group (D).

Figure 3

Hierarchical clustering of commonly up-regulated (A) and down-regulated DEGs (B), numbers of DEGs in enriched functions (C), and the validation of RNA-seq by qRT-PCR (D). For hierarchical clustering (A), blue and red indicate decreased and increased expression, respectively. Transcripts were clustered by hierarchical clustering using the complete linkage algorithm and Pearson correlation metric in R. For QPCR analysis of the expression of randomly selected novel genes (B), data are presented as mean ± S.D. (n = 3). Means of treatments not sharing a common letter are significantly different at P < 0.05.

Figure 4

Relationship between protein changes and transcript changes in response to different temperatures. For summarization of proteins that have mRNA measurements, proteins (top bar) are grouped with regard to their responses to temperature. Each protein group was then subgrouped with regard to corresponding mRNA changes (bottom pies). The comparison between intracellular protein changes and transcript changes were listed in (A,B), while the comparison between extracellular protein changes and transcript changes were listed in (C,D).

Figure 5

Temperature-dependent regulation of PVD synthesis involved in virulence of Pseudomonas plecoglossicida. (A) Presence of PVD in the supernatants of cultures under different temperatures. Data are presented as mean ± S.D. (n = 3). Means of treatments not sharing a common letter are significantly different at P < 0.05. (B) The distance of the advancing color-change front in the CAS-blue agar under different temperatures. Data are presented as mean ± S.D. (n = 3). Means of treatments not sharing a common letter are significantly different at P < 0.05. (C) qRT-PCR analysis of the expression of pvds1, pvds2, pvds3, pvds4, and pvds5 after stable gene silencing compared with the control. Data are presented as mean ± S.D. (n = 3). Means of treatments not sharing a common letter are significantly different at P < 0.05. (D) Presence of PVD in the supernatants of stable silenced strains compared with the control. Data are presented as mean ± S.D. (n = 3). Means of treatments not sharing a common letter are significantly different at P < 0.05. (E) The distance of the advancing color-change front in the CAS-blue agar of stable silenced strains compared with the control. Data are presented as mean ± S.D. (n = 3). Means of treatments not sharing a common letter are significantly different at P < 0.05. (F) Growth of wild-type and stable silenced strains in the absence of 2, 2′-Dipyridyl (n = 3). (G) Growth of wild-type and stable silenced strains in the presence of 2, 2′-Dipyridyl (n = 3). (H) Biofilm formation of wild-type and stable silenced strains in the presence or absence of 2, 2′-Dipyridyl. Data are presented as mean ± S.D. (n = 6). Means of treatments not sharing a common letter are significantly different at P < 0.05. (I) The cumulative survival of E. coioides injected with wild-type and pvds1-, pvds2-, pvds3-, pvds4-, and pvds5-RNAi strains during 10 days post-challenge.

Figure 6

The dynamic distribution of RNAi strains in host. E. coioides were intrapleurally injected with WT P. plecoglossicida strain, PVDS2-RNAi strain and icmF-RNAi strain, respectively. The numbers of RNAi strains bacteria in spleen, liver, head kidney, trunk kidney and blood after injection of PVDS2-RNAi strain (A) and icmF-RNAi strain (B) were compared with the WT strain using qPCR at 1, 6, 12, 24, 48, 72, and 96 h. Data are shown as means ± SD from three independent biological replicates. * P < 0.05, **P < 0.01.

Figure 7

Temperature-dependent regulation of T6SS involved in virulence of Pseudomonas plecoglossicida. (A) qRT-PCR analysis of the expression of hcp, dotU, and icmF after stable gene silencing compared with the control. Data are presented as mean ± S.D. (n = 3). Means of treatments not sharing a common letter are significantly different at P < 0.05. (B) Biofilm formation of wild-type and stable silenced strains. Data are presented as mean ± S.D. (n = 6). Means of treatments not sharing a common letter are significantly different at P < 0.05. (C) The cumulative survival of E. coioides injected with wild-type and hcp-, dotU-, and icmF-RNAi strains during 10 days post-challenge.