Limits...
Refined analysis of the Campylobacter jejuni iron-dependent/independent Fur- and PerR-transcriptomes.

Butcher J, Handley RA, van Vliet AH, Stintzi A - BMC Genomics (2015)

Bottom Line: It was found that 202 genes were differentially expressed in at least one mutant under iron-replete conditions and 331 genes were differentially expressed in at least one mutant under iron-limited conditions.The CjFur and CjPerR transcriptomes characterized in this study were compared to those previously identified using microarray profiling and found to be more extensive than previously understood.Moreover, subsets of genes were found which are only differentially expressed when both CjFur and CjPerR are deleted and includes genes that appear to be simultaneously activated by CjFur and repressed by CjPerR.

View Article: PubMed Central - PubMed

Affiliation: Department of Biochemistry, Microbiology and Immunology, Ottawa Institute of Systems Biology, University of Ottawa, Ottawa, ON, Canada. jbutc076@uottawa.ca.

ABSTRACT

Background: The genome of Campylobacter jejuni contains two iron activated Fur-family transcriptional regulators, CjFur and CjPerR, which are primarily responsible for regulating iron homeostasis and oxidative stress respectively. Both transcriptional regulators have been previously implicated in regulating diverse functions beyond their primary roles in C. jejuni. To further characterize their regulatory networks, RNA-seq was used to define the transcriptional profiles of C. jejuni NCTC11168 wild type, Δfur, ΔperR and ΔfurΔperR isogenic deletion mutants under both iron-replete and iron-limited conditions.

Results: It was found that 202 genes were differentially expressed in at least one mutant under iron-replete conditions and 331 genes were differentially expressed in at least one mutant under iron-limited conditions. The CjFur and CjPerR transcriptomes characterized in this study were compared to those previously identified using microarray profiling and found to be more extensive than previously understood. Interestingly, our results indicate that CjFur/CjPerR appear to co-regulate the expression of flagellar biogenesis genes in an opposing and iron-independent fashion. Moreover the ΔfurΔperR isogenic deletion mutant revealed that CjFur and CjPerR can compensate for each other in certain cases, suggesting that both regulators may compete for binding to specific promoters.

Conclusions: The CjFur and CjPerR transcriptomes are larger than previously reported. In particular, deletion of perR results in the differential expression of a large group of genes in the absence of iron, suggesting that CjPerR may also regulate genes in an iron-independent manner, similar to what has already been demonstrated with CjFur. Moreover, subsets of genes were found which are only differentially expressed when both CjFur and CjPerR are deleted and includes genes that appear to be simultaneously activated by CjFur and repressed by CjPerR. In particular the iron-independent co-regulation of flagellar biogenesis by CjFur/CjPerR represents a potentially novel regulatory function for these proteins. These findings represent additional modes of co-regulation by these two transcriptional regulators in C. jejuni.

No MeSH data available.


Related in: MedlinePlus

Hierarchical clustering of genes differentially expressed under iron-limited conditions. Genes found to be differentially expressed in at least one strain under iron-limited conditions were subjected to hierarchical clustering to identify corresponding genes with similar expression profiles. Clusters I-K were split from the original clustering figure for ease of viewing. The columns each represent one strain (Δfur, ΔperR, ΔfurΔperR) and relative fold changes in expression are presented in a Log2 scale with up-regulated genes in blue and down-regulated genes in yellow. The clustering resulted in 11 main clusters (a-k). Clusters a, e and h are expanded to highlight the genes present in each cluster. See Additional file 7: Table S6 for further details and Additional file 9: Figure S3 for full histogram
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4491227&req=5

Fig3: Hierarchical clustering of genes differentially expressed under iron-limited conditions. Genes found to be differentially expressed in at least one strain under iron-limited conditions were subjected to hierarchical clustering to identify corresponding genes with similar expression profiles. Clusters I-K were split from the original clustering figure for ease of viewing. The columns each represent one strain (Δfur, ΔperR, ΔfurΔperR) and relative fold changes in expression are presented in a Log2 scale with up-regulated genes in blue and down-regulated genes in yellow. The clustering resulted in 11 main clusters (a-k). Clusters a, e and h are expanded to highlight the genes present in each cluster. See Additional file 7: Table S6 for further details and Additional file 9: Figure S3 for full histogram

Mentions: Paired end reads were aligned to the C. jejuni NCTC11168 genome (NC002163) using Bowtie V2 on the Biolinux platform [15, 16]. Approximately 97 % of reads aligned for each sequencing run (Additional file 1: Table S1). Aligned reads were visualized using Artemis with the Bamview plugin [17]. Raw read counts for each gene were tabulated (Additional file 2: Table S2) using Bedtools and also used to calculate a reads per kilobase coding sequence per million sequenced reads (RPKM) value for each gene (Additional file 3: Table S3). The calculated Log(RPKM + 1) values were used to test the homogeneity of each strain’s biological replicates by performing PCA analysis in R using prcomp [18]. Samples which appeared to cluster differently from their mates were excluded from the downstream differential expression analyses (Fig. 1, Additional file 4: Figure S1). The edgeR statistical package was used to determine gene expression fold changes using the current Genbank gene annotation information for C. jejuni NCTC11168 [19, 20]. Transcripts showing a fold change ≥ 1.5 with a FDR value ≥ 0.05 were considered to be significantly differentially expressed (Table 1, Additional file 5: Table S4, Additional file 6: Table S5, Additional file 7: Table S6) [7]. Differentially expressed genes were subsequently merged based on growth condition and clustered (using average linkage Euclidean distance) based on their expression profile using Genesis (Figs. 2 and 3, Additional file 8: Figure S2, Additional file 9: Figure S3) [21]. Differentially expressed genes for each condition were also subjected to gene set enrichment analysis (GSEA) on annotated KEGG pathways using GAGE with a FDR cutoff of <0.1 [22]. Selected significantly enriched pathways were visualized using Pathview [23]. The similarities and differences between the genes found to be differentially expressed in various strains were visualized using hive plots with differentially expressed genes as nodes and genes present in multiple strains connected with ribbons [24]. In addition, genes identified as being part of the Δfur and ΔperR transcriptomes via RNA-seq were compared to the Δfur and ΔperR transcriptomes previously characterized using microarrays [2, 4]. Genes were considered differentially expressed in the microarrays using the same parameters as in the original papers (>2 fold change, p < 0.001). Iron responsive genes were also determined by comparing the wild-type grown under iron-replete and iron-limited conditions (Additional file 10: Table S7) to the iron responsive genes previously identified via microarray profiling or RNA-seq [2, 7].Fig. 1


Refined analysis of the Campylobacter jejuni iron-dependent/independent Fur- and PerR-transcriptomes.

Butcher J, Handley RA, van Vliet AH, Stintzi A - BMC Genomics (2015)

Hierarchical clustering of genes differentially expressed under iron-limited conditions. Genes found to be differentially expressed in at least one strain under iron-limited conditions were subjected to hierarchical clustering to identify corresponding genes with similar expression profiles. Clusters I-K were split from the original clustering figure for ease of viewing. The columns each represent one strain (Δfur, ΔperR, ΔfurΔperR) and relative fold changes in expression are presented in a Log2 scale with up-regulated genes in blue and down-regulated genes in yellow. The clustering resulted in 11 main clusters (a-k). Clusters a, e and h are expanded to highlight the genes present in each cluster. See Additional file 7: Table S6 for further details and Additional file 9: Figure S3 for full histogram
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4491227&req=5

Fig3: Hierarchical clustering of genes differentially expressed under iron-limited conditions. Genes found to be differentially expressed in at least one strain under iron-limited conditions were subjected to hierarchical clustering to identify corresponding genes with similar expression profiles. Clusters I-K were split from the original clustering figure for ease of viewing. The columns each represent one strain (Δfur, ΔperR, ΔfurΔperR) and relative fold changes in expression are presented in a Log2 scale with up-regulated genes in blue and down-regulated genes in yellow. The clustering resulted in 11 main clusters (a-k). Clusters a, e and h are expanded to highlight the genes present in each cluster. See Additional file 7: Table S6 for further details and Additional file 9: Figure S3 for full histogram
Mentions: Paired end reads were aligned to the C. jejuni NCTC11168 genome (NC002163) using Bowtie V2 on the Biolinux platform [15, 16]. Approximately 97 % of reads aligned for each sequencing run (Additional file 1: Table S1). Aligned reads were visualized using Artemis with the Bamview plugin [17]. Raw read counts for each gene were tabulated (Additional file 2: Table S2) using Bedtools and also used to calculate a reads per kilobase coding sequence per million sequenced reads (RPKM) value for each gene (Additional file 3: Table S3). The calculated Log(RPKM + 1) values were used to test the homogeneity of each strain’s biological replicates by performing PCA analysis in R using prcomp [18]. Samples which appeared to cluster differently from their mates were excluded from the downstream differential expression analyses (Fig. 1, Additional file 4: Figure S1). The edgeR statistical package was used to determine gene expression fold changes using the current Genbank gene annotation information for C. jejuni NCTC11168 [19, 20]. Transcripts showing a fold change ≥ 1.5 with a FDR value ≥ 0.05 were considered to be significantly differentially expressed (Table 1, Additional file 5: Table S4, Additional file 6: Table S5, Additional file 7: Table S6) [7]. Differentially expressed genes were subsequently merged based on growth condition and clustered (using average linkage Euclidean distance) based on their expression profile using Genesis (Figs. 2 and 3, Additional file 8: Figure S2, Additional file 9: Figure S3) [21]. Differentially expressed genes for each condition were also subjected to gene set enrichment analysis (GSEA) on annotated KEGG pathways using GAGE with a FDR cutoff of <0.1 [22]. Selected significantly enriched pathways were visualized using Pathview [23]. The similarities and differences between the genes found to be differentially expressed in various strains were visualized using hive plots with differentially expressed genes as nodes and genes present in multiple strains connected with ribbons [24]. In addition, genes identified as being part of the Δfur and ΔperR transcriptomes via RNA-seq were compared to the Δfur and ΔperR transcriptomes previously characterized using microarrays [2, 4]. Genes were considered differentially expressed in the microarrays using the same parameters as in the original papers (>2 fold change, p < 0.001). Iron responsive genes were also determined by comparing the wild-type grown under iron-replete and iron-limited conditions (Additional file 10: Table S7) to the iron responsive genes previously identified via microarray profiling or RNA-seq [2, 7].Fig. 1

Bottom Line: It was found that 202 genes were differentially expressed in at least one mutant under iron-replete conditions and 331 genes were differentially expressed in at least one mutant under iron-limited conditions.The CjFur and CjPerR transcriptomes characterized in this study were compared to those previously identified using microarray profiling and found to be more extensive than previously understood.Moreover, subsets of genes were found which are only differentially expressed when both CjFur and CjPerR are deleted and includes genes that appear to be simultaneously activated by CjFur and repressed by CjPerR.

View Article: PubMed Central - PubMed

Affiliation: Department of Biochemistry, Microbiology and Immunology, Ottawa Institute of Systems Biology, University of Ottawa, Ottawa, ON, Canada. jbutc076@uottawa.ca.

ABSTRACT

Background: The genome of Campylobacter jejuni contains two iron activated Fur-family transcriptional regulators, CjFur and CjPerR, which are primarily responsible for regulating iron homeostasis and oxidative stress respectively. Both transcriptional regulators have been previously implicated in regulating diverse functions beyond their primary roles in C. jejuni. To further characterize their regulatory networks, RNA-seq was used to define the transcriptional profiles of C. jejuni NCTC11168 wild type, Δfur, ΔperR and ΔfurΔperR isogenic deletion mutants under both iron-replete and iron-limited conditions.

Results: It was found that 202 genes were differentially expressed in at least one mutant under iron-replete conditions and 331 genes were differentially expressed in at least one mutant under iron-limited conditions. The CjFur and CjPerR transcriptomes characterized in this study were compared to those previously identified using microarray profiling and found to be more extensive than previously understood. Interestingly, our results indicate that CjFur/CjPerR appear to co-regulate the expression of flagellar biogenesis genes in an opposing and iron-independent fashion. Moreover the ΔfurΔperR isogenic deletion mutant revealed that CjFur and CjPerR can compensate for each other in certain cases, suggesting that both regulators may compete for binding to specific promoters.

Conclusions: The CjFur and CjPerR transcriptomes are larger than previously reported. In particular, deletion of perR results in the differential expression of a large group of genes in the absence of iron, suggesting that CjPerR may also regulate genes in an iron-independent manner, similar to what has already been demonstrated with CjFur. Moreover, subsets of genes were found which are only differentially expressed when both CjFur and CjPerR are deleted and includes genes that appear to be simultaneously activated by CjFur and repressed by CjPerR. In particular the iron-independent co-regulation of flagellar biogenesis by CjFur/CjPerR represents a potentially novel regulatory function for these proteins. These findings represent additional modes of co-regulation by these two transcriptional regulators in C. jejuni.

No MeSH data available.


Related in: MedlinePlus