Limits...
A comprehensive analysis of adiponectin QTLs using SNP association, SNP cis-effects on peripheral blood gene expression and gene expression correlation identified novel metabolic syndrome (MetS) genes with potential role in carcinogenesis and systemic inflammation.

Zhang Y, Kent JW, Olivier M, Ali O, Cerjak D, Broeckel U, Abdou RM, Dyer TD, Comuzzie A, Curran JE, Carless MA, Rainwater DL, Göring HH, Blangero J, Kissebah AH - BMC Med Genomics (2013)

Bottom Line: QTL-specific haplotype-tagging SNPs associated with MetS phenotypes were annotated to 14 genes whose function could influence MetS biology as well as oncogenesis or inflammation.Strong evidence of cis-effects on the expression of MYO10 in PWBC was found with SNPs clustered near the gene's transcription start site.Variants of genes AKAP6, NPAS3, MARCH6 and FBXL7 have been previously reported to be associated with insulin resistance, inflammatory markers or adiposity studies using genome-wide approaches whereas associations of CDH18 and MYO10 with MetS traits have not been reported before.

View Article: PubMed Central - HTML - PubMed

Affiliation: TOPS Obesity and Metabolic Research Center, Department of Medicine, Medical College of Wisconsin, Milwaukee, Wisconsin, USA. yzhang@mcw.edu

ABSTRACT

Background: Metabolic syndrome (MetS) is an aberration associated with increased risk for cancer and inflammation. Adiponectin, an adipocyte-produced abundant protein hormone, has countering effect on the diabetogenic and atherogenic components of MetS. Plasma levels of adiponectin are negatively correlated with onset of cancer and cancer patient mortality. We previously performed microsatellite linkage analyses using adiponectin as a surrogate marker and revealed two QTLs on chr5 (5p14) and chr14 (14q13).

Methods: Using individuals from 85 extended families that contributed to the linkage and who were measured for 42 clinical and biologic MetS phenotypes, we tested QTL-based SNP associations, peripheral white blood cell (PWBC) gene expression, and the effects of cis-acting SNPs on gene expression to discover genomic elements that could affect the pathophysiology and complications of MetS.

Results: Adiponectin levels were found to be highly intercorrelated phenotypically with the majority of MetS traits. QTL-specific haplotype-tagging SNPs associated with MetS phenotypes were annotated to 14 genes whose function could influence MetS biology as well as oncogenesis or inflammation. These were mechanistically categorized into four groups: cell-cell adhesion and mobility, signal transduction, transcription and protein sorting. Four genes were highly prioritized: cadherin 18 (CDH18), myosin X (MYO10), anchor protein 6 of AMPK (AKAP6), and neuronal PAS domain protein 3 (NPAS3). PWBC expression was detectable only for the following genes with multi-organ or with multi-function properties: NPAS3, MARCH6, MYO10 and FBXL7. Strong evidence of cis-effects on the expression of MYO10 in PWBC was found with SNPs clustered near the gene's transcription start site. MYO10 expression in PWBC was marginally correlated with body composition (p = 0.065) and adipokine levels in the periphery (p = 0.064). Variants of genes AKAP6, NPAS3, MARCH6 and FBXL7 have been previously reported to be associated with insulin resistance, inflammatory markers or adiposity studies using genome-wide approaches whereas associations of CDH18 and MYO10 with MetS traits have not been reported before.

Conclusions: Adiponectin QTLs-based SNP association and mRNA expression identified genes that could mediate the association between MetS and cancer or inflammation.

Show MeSH

Related in: MedlinePlus

Manhattan plots of SNP associations with MetS phenotypes within adiponectin QTL at 14q13. Dark Blue dots depict levels of association of identifier phenotypes with all SNPs in the region determined from 1 LOD reduction from the peak [chr14: 23,131,000-36,761,868 (NCBI36/hg18)]. Vertical axis represents minus logarithm of the p-values and horizontal represents the chromosomal position (Mb). Levels of QTL-wide significance thresholds are shown by the dash lines. Red lines indicate the significant level (pα=0.05= pα=0.05=1.86×10-5) and blue lines indicate suggestive level pα=0.1= 3.72×10-5). A) Association patterns of all the SNPs of the region with RQ phenotype. Transcripts defined by UCSC genome browser (31, 32) are shown below the Manhattan plot, by blue bars. SNPs that show highest association with weight are located in gene AKAP6 (framed in red). B) SNP associations with triglycerides. SNPs that show highest associations with triglycerides are mapped to the transcription factor gene NPAS3.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3643849&req=5

Figure 2: Manhattan plots of SNP associations with MetS phenotypes within adiponectin QTL at 14q13. Dark Blue dots depict levels of association of identifier phenotypes with all SNPs in the region determined from 1 LOD reduction from the peak [chr14: 23,131,000-36,761,868 (NCBI36/hg18)]. Vertical axis represents minus logarithm of the p-values and horizontal represents the chromosomal position (Mb). Levels of QTL-wide significance thresholds are shown by the dash lines. Red lines indicate the significant level (pα=0.05= pα=0.05=1.86×10-5) and blue lines indicate suggestive level pα=0.1= 3.72×10-5). A) Association patterns of all the SNPs of the region with RQ phenotype. Transcripts defined by UCSC genome browser (31, 32) are shown below the Manhattan plot, by blue bars. SNPs that show highest association with weight are located in gene AKAP6 (framed in red). B) SNP associations with triglycerides. SNPs that show highest associations with triglycerides are mapped to the transcription factor gene NPAS3.

Mentions: This QTL region has a high density of genes with regulatory functions including signal transduction, transcription and post-transcription processing. Our fine-mapping identified SNPs associated with each of the 42 MetS phenotypes. Table 3 summarizes the SNPs most highly associated with each trait, the attributes of the variants, and the gene ontology of their annotated genes. Variants of NPAS3, a neuronal transcription factor thought to be involved in brain tumor suppression, were found to be associated with 11 of the MetS phenotypes (Fatkg, Leankg, SubQF, TAF, REE, TG, TC, LDL-c, pulse, IL1b and IL-6), suggesting a pleitropic effect (Table 3). Two of these associations (TG and TC) were borderline significant, marginally reaching the regional threshold (Table 3 and Figure 2A). SNPs near BRMS1L, a breast cancer suppressor gene, was associated with LMEDn and LDLppd. SNPs of EGLN3, a regulator of transcription factor HIF that affects apoptosis in hemangioblastoma and clear cell renal cancer, were associated with IGR and LDL-c. SNPs of NOVA1, a post-transcription processing of the GnRH paraneoplastic antigen, were associated at nominal significance with several phenotypes including weight, BMI, WC, FG, FI, HOMA, SI, AIR, and sBP.


A comprehensive analysis of adiponectin QTLs using SNP association, SNP cis-effects on peripheral blood gene expression and gene expression correlation identified novel metabolic syndrome (MetS) genes with potential role in carcinogenesis and systemic inflammation.

Zhang Y, Kent JW, Olivier M, Ali O, Cerjak D, Broeckel U, Abdou RM, Dyer TD, Comuzzie A, Curran JE, Carless MA, Rainwater DL, Göring HH, Blangero J, Kissebah AH - BMC Med Genomics (2013)

Manhattan plots of SNP associations with MetS phenotypes within adiponectin QTL at 14q13. Dark Blue dots depict levels of association of identifier phenotypes with all SNPs in the region determined from 1 LOD reduction from the peak [chr14: 23,131,000-36,761,868 (NCBI36/hg18)]. Vertical axis represents minus logarithm of the p-values and horizontal represents the chromosomal position (Mb). Levels of QTL-wide significance thresholds are shown by the dash lines. Red lines indicate the significant level (pα=0.05= pα=0.05=1.86×10-5) and blue lines indicate suggestive level pα=0.1= 3.72×10-5). A) Association patterns of all the SNPs of the region with RQ phenotype. Transcripts defined by UCSC genome browser (31, 32) are shown below the Manhattan plot, by blue bars. SNPs that show highest association with weight are located in gene AKAP6 (framed in red). B) SNP associations with triglycerides. SNPs that show highest associations with triglycerides are mapped to the transcription factor gene NPAS3.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3643849&req=5

Figure 2: Manhattan plots of SNP associations with MetS phenotypes within adiponectin QTL at 14q13. Dark Blue dots depict levels of association of identifier phenotypes with all SNPs in the region determined from 1 LOD reduction from the peak [chr14: 23,131,000-36,761,868 (NCBI36/hg18)]. Vertical axis represents minus logarithm of the p-values and horizontal represents the chromosomal position (Mb). Levels of QTL-wide significance thresholds are shown by the dash lines. Red lines indicate the significant level (pα=0.05= pα=0.05=1.86×10-5) and blue lines indicate suggestive level pα=0.1= 3.72×10-5). A) Association patterns of all the SNPs of the region with RQ phenotype. Transcripts defined by UCSC genome browser (31, 32) are shown below the Manhattan plot, by blue bars. SNPs that show highest association with weight are located in gene AKAP6 (framed in red). B) SNP associations with triglycerides. SNPs that show highest associations with triglycerides are mapped to the transcription factor gene NPAS3.
Mentions: This QTL region has a high density of genes with regulatory functions including signal transduction, transcription and post-transcription processing. Our fine-mapping identified SNPs associated with each of the 42 MetS phenotypes. Table 3 summarizes the SNPs most highly associated with each trait, the attributes of the variants, and the gene ontology of their annotated genes. Variants of NPAS3, a neuronal transcription factor thought to be involved in brain tumor suppression, were found to be associated with 11 of the MetS phenotypes (Fatkg, Leankg, SubQF, TAF, REE, TG, TC, LDL-c, pulse, IL1b and IL-6), suggesting a pleitropic effect (Table 3). Two of these associations (TG and TC) were borderline significant, marginally reaching the regional threshold (Table 3 and Figure 2A). SNPs near BRMS1L, a breast cancer suppressor gene, was associated with LMEDn and LDLppd. SNPs of EGLN3, a regulator of transcription factor HIF that affects apoptosis in hemangioblastoma and clear cell renal cancer, were associated with IGR and LDL-c. SNPs of NOVA1, a post-transcription processing of the GnRH paraneoplastic antigen, were associated at nominal significance with several phenotypes including weight, BMI, WC, FG, FI, HOMA, SI, AIR, and sBP.

Bottom Line: QTL-specific haplotype-tagging SNPs associated with MetS phenotypes were annotated to 14 genes whose function could influence MetS biology as well as oncogenesis or inflammation.Strong evidence of cis-effects on the expression of MYO10 in PWBC was found with SNPs clustered near the gene's transcription start site.Variants of genes AKAP6, NPAS3, MARCH6 and FBXL7 have been previously reported to be associated with insulin resistance, inflammatory markers or adiposity studies using genome-wide approaches whereas associations of CDH18 and MYO10 with MetS traits have not been reported before.

View Article: PubMed Central - HTML - PubMed

Affiliation: TOPS Obesity and Metabolic Research Center, Department of Medicine, Medical College of Wisconsin, Milwaukee, Wisconsin, USA. yzhang@mcw.edu

ABSTRACT

Background: Metabolic syndrome (MetS) is an aberration associated with increased risk for cancer and inflammation. Adiponectin, an adipocyte-produced abundant protein hormone, has countering effect on the diabetogenic and atherogenic components of MetS. Plasma levels of adiponectin are negatively correlated with onset of cancer and cancer patient mortality. We previously performed microsatellite linkage analyses using adiponectin as a surrogate marker and revealed two QTLs on chr5 (5p14) and chr14 (14q13).

Methods: Using individuals from 85 extended families that contributed to the linkage and who were measured for 42 clinical and biologic MetS phenotypes, we tested QTL-based SNP associations, peripheral white blood cell (PWBC) gene expression, and the effects of cis-acting SNPs on gene expression to discover genomic elements that could affect the pathophysiology and complications of MetS.

Results: Adiponectin levels were found to be highly intercorrelated phenotypically with the majority of MetS traits. QTL-specific haplotype-tagging SNPs associated with MetS phenotypes were annotated to 14 genes whose function could influence MetS biology as well as oncogenesis or inflammation. These were mechanistically categorized into four groups: cell-cell adhesion and mobility, signal transduction, transcription and protein sorting. Four genes were highly prioritized: cadherin 18 (CDH18), myosin X (MYO10), anchor protein 6 of AMPK (AKAP6), and neuronal PAS domain protein 3 (NPAS3). PWBC expression was detectable only for the following genes with multi-organ or with multi-function properties: NPAS3, MARCH6, MYO10 and FBXL7. Strong evidence of cis-effects on the expression of MYO10 in PWBC was found with SNPs clustered near the gene's transcription start site. MYO10 expression in PWBC was marginally correlated with body composition (p = 0.065) and adipokine levels in the periphery (p = 0.064). Variants of genes AKAP6, NPAS3, MARCH6 and FBXL7 have been previously reported to be associated with insulin resistance, inflammatory markers or adiposity studies using genome-wide approaches whereas associations of CDH18 and MYO10 with MetS traits have not been reported before.

Conclusions: Adiponectin QTLs-based SNP association and mRNA expression identified genes that could mediate the association between MetS and cancer or inflammation.

Show MeSH
Related in: MedlinePlus