Limits...
Mapping of genotype-phenotype diversity among clinical isolates of mycobacterium tuberculosis by sequence-based transcriptional profiling.

Rose G, Cortes T, Comas I, Coscolla M, Gagneux S, Young DB - Genome Biol Evol (2013)

Bottom Line: First, lineage-specific changes in amino acid sequence of transcriptional regulators were associated with alterations in their ability to control gene expression.Second, changes in nucleotide sequence were associated with alteration of promoter activity and generation of novel transcriptional start sites in intergenic regions and within coding sequences.Taken together, these findings advance our understanding of mycobacterial evolution, contribute to a systems level understanding of this important human pathogen, and more broadly demonstrate the application of state-of-the-art techniques to provide novel insight into mechanisms by which intergenic and silent mutations contribute to diversity.

View Article: PubMed Central - PubMed

Affiliation: MRC National Institute for Medical Research, Mill Hill, London, United Kingdom.

ABSTRACT
Genome sequencing has identified an extensive repertoire of single nucleotide polymorphisms among clinical isolates of Mycobacterium tuberculosis, but the extent to which these differences influence phenotypic properties of the bacteria remains to be elucidated. To determine whether these polymorphisms give rise to phenotypic diversity, we have integrated genome data sets with RNA sequencing to assess their impact on the comparative transcriptome profiles of strains belonging to M. tuberculosis Lineages 1 and 2. We observed clear correlations between genotype and transcriptional phenotype. These arose by three mechanisms. First, lineage-specific changes in amino acid sequence of transcriptional regulators were associated with alterations in their ability to control gene expression. Second, changes in nucleotide sequence were associated with alteration of promoter activity and generation of novel transcriptional start sites in intergenic regions and within coding sequences. We show that in some cases this mechanism is expected to generate functionally active truncated proteins involved in innate immune recognition. Finally, genes showing lineage-specific patterns of differential expression not linked directly to primary mutations were characterized by a striking overrepresentation of toxin-antitoxin pairs. Taken together, these findings advance our understanding of mycobacterial evolution, contribute to a systems level understanding of this important human pathogen, and more broadly demonstrate the application of state-of-the-art techniques to provide novel insight into mechanisms by which intergenic and silent mutations contribute to diversity.

Show MeSH

Related in: MedlinePlus

Overrepresentation of differentially expressed toxin–antitoxins (TAs). (A) Ratio of significant differential gene expression grouped by functional category, compared with the genome-wide representation of the category. Values on the x axis represents the difference as fold change, positive fold change indicates overrepresentation of a particular function category and negative values underrepresentation. There were 2.9-fold more toxin–antitoxins than expected (χ2, P = 0.03). (B) Validation of selected RNA-seq differentially expressed toxin–antitoxins (solid bars) by qRT-PCR (striped bars). Fold change relative to Lineage 1 expression on y axis (log10 scale) and bars colored by lineage with higher expression. Error bars for qRT-PCR indicate the standard deviation of three biological replicates.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3814196&req=5

evt138-F4: Overrepresentation of differentially expressed toxin–antitoxins (TAs). (A) Ratio of significant differential gene expression grouped by functional category, compared with the genome-wide representation of the category. Values on the x axis represents the difference as fold change, positive fold change indicates overrepresentation of a particular function category and negative values underrepresentation. There were 2.9-fold more toxin–antitoxins than expected (χ2, P = 0.03). (B) Validation of selected RNA-seq differentially expressed toxin–antitoxins (solid bars) by qRT-PCR (striped bars). Fold change relative to Lineage 1 expression on y axis (log10 scale) and bars colored by lineage with higher expression. Error bars for qRT-PCR indicate the standard deviation of three biological replicates.

Mentions: We were unable to identify direct SNP associations for the remainder of the genes showing lineage-specific patterns of differential expression (135 out of 168 differentially expressed genes and antisense). We anticipate that their differential expression reflects downstream consequences of primary mutations. Analysis of the panel of differentially expressed genes according to functional category identified a 2-fold overrepresentation of proteins involved in virulence, detoxification, and adaptation. This was driven by ten TA genes ( supplementary table S7D, Supplementary Material online), and separate classification of all TA as an independent category revealed 2.9-fold overrepresentation in the differentially expressed set compared with the genome representation (χ2, P = 0.03) (fig. 4A and supplementary table S8, Supplementary Material online). For selected TA, the pattern of differential gene expression seen by RNA-seq was confirmed by quantitative RT-PCR (fig. 4B and supplementary tables S9 and S10, Supplementary Material online).Fig. 4.—


Mapping of genotype-phenotype diversity among clinical isolates of mycobacterium tuberculosis by sequence-based transcriptional profiling.

Rose G, Cortes T, Comas I, Coscolla M, Gagneux S, Young DB - Genome Biol Evol (2013)

Overrepresentation of differentially expressed toxin–antitoxins (TAs). (A) Ratio of significant differential gene expression grouped by functional category, compared with the genome-wide representation of the category. Values on the x axis represents the difference as fold change, positive fold change indicates overrepresentation of a particular function category and negative values underrepresentation. There were 2.9-fold more toxin–antitoxins than expected (χ2, P = 0.03). (B) Validation of selected RNA-seq differentially expressed toxin–antitoxins (solid bars) by qRT-PCR (striped bars). Fold change relative to Lineage 1 expression on y axis (log10 scale) and bars colored by lineage with higher expression. Error bars for qRT-PCR indicate the standard deviation of three biological replicates.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3814196&req=5

evt138-F4: Overrepresentation of differentially expressed toxin–antitoxins (TAs). (A) Ratio of significant differential gene expression grouped by functional category, compared with the genome-wide representation of the category. Values on the x axis represents the difference as fold change, positive fold change indicates overrepresentation of a particular function category and negative values underrepresentation. There were 2.9-fold more toxin–antitoxins than expected (χ2, P = 0.03). (B) Validation of selected RNA-seq differentially expressed toxin–antitoxins (solid bars) by qRT-PCR (striped bars). Fold change relative to Lineage 1 expression on y axis (log10 scale) and bars colored by lineage with higher expression. Error bars for qRT-PCR indicate the standard deviation of three biological replicates.
Mentions: We were unable to identify direct SNP associations for the remainder of the genes showing lineage-specific patterns of differential expression (135 out of 168 differentially expressed genes and antisense). We anticipate that their differential expression reflects downstream consequences of primary mutations. Analysis of the panel of differentially expressed genes according to functional category identified a 2-fold overrepresentation of proteins involved in virulence, detoxification, and adaptation. This was driven by ten TA genes ( supplementary table S7D, Supplementary Material online), and separate classification of all TA as an independent category revealed 2.9-fold overrepresentation in the differentially expressed set compared with the genome representation (χ2, P = 0.03) (fig. 4A and supplementary table S8, Supplementary Material online). For selected TA, the pattern of differential gene expression seen by RNA-seq was confirmed by quantitative RT-PCR (fig. 4B and supplementary tables S9 and S10, Supplementary Material online).Fig. 4.—

Bottom Line: First, lineage-specific changes in amino acid sequence of transcriptional regulators were associated with alterations in their ability to control gene expression.Second, changes in nucleotide sequence were associated with alteration of promoter activity and generation of novel transcriptional start sites in intergenic regions and within coding sequences.Taken together, these findings advance our understanding of mycobacterial evolution, contribute to a systems level understanding of this important human pathogen, and more broadly demonstrate the application of state-of-the-art techniques to provide novel insight into mechanisms by which intergenic and silent mutations contribute to diversity.

View Article: PubMed Central - PubMed

Affiliation: MRC National Institute for Medical Research, Mill Hill, London, United Kingdom.

ABSTRACT
Genome sequencing has identified an extensive repertoire of single nucleotide polymorphisms among clinical isolates of Mycobacterium tuberculosis, but the extent to which these differences influence phenotypic properties of the bacteria remains to be elucidated. To determine whether these polymorphisms give rise to phenotypic diversity, we have integrated genome data sets with RNA sequencing to assess their impact on the comparative transcriptome profiles of strains belonging to M. tuberculosis Lineages 1 and 2. We observed clear correlations between genotype and transcriptional phenotype. These arose by three mechanisms. First, lineage-specific changes in amino acid sequence of transcriptional regulators were associated with alterations in their ability to control gene expression. Second, changes in nucleotide sequence were associated with alteration of promoter activity and generation of novel transcriptional start sites in intergenic regions and within coding sequences. We show that in some cases this mechanism is expected to generate functionally active truncated proteins involved in innate immune recognition. Finally, genes showing lineage-specific patterns of differential expression not linked directly to primary mutations were characterized by a striking overrepresentation of toxin-antitoxin pairs. Taken together, these findings advance our understanding of mycobacterial evolution, contribute to a systems level understanding of this important human pathogen, and more broadly demonstrate the application of state-of-the-art techniques to provide novel insight into mechanisms by which intergenic and silent mutations contribute to diversity.

Show MeSH
Related in: MedlinePlus