Computational tools for prioritizing candidate genes: boosting disease gene discovery

At different stages of any research project, molecular biologists need to choose - often somewhat arbitrarily, even after careful statistical data analysis - which genes or proteins to investigate further experimentally and which to leave out because of limited resources. Computational methods that integrate complex, heterogeneous data sets - such as expression data, sequence information, functional annotation and the biomedical literature - allow prioritizing genes for future study in a more informed way. Such methods can substantially increase the yield of downstream studies and are becoming invaluable to researchers.     

PCK1 and PCK2 as candidate diabetes and obesity genes

The PCK1 gene (Pck1 in rodents) encodes the cytosolic isozyme of phosphoenolpyruvate carboxykinase (PEPCK-C), which is well-known for its function as a gluconeogenic enzyme in the liver and kidney. Mouse studies involving whole body and tissue-specific Pck1 knockouts as well as tissue-specific over-expression of PEPCK-C have resulted in type 2 diabetes as well as several surprising phenotypes including obesity, lipodystrophy, fatty liver, and death. These phenotypes arise from perturbations not only in gluconeogenesis but in two additional metabolic functions of PEPCK-C: (1) cataplerosis which maintains metabolic flux through the Krebs cycle by removing excess oxaloacetate, and (2) glyceroneogenesis which produces glycerol-3-phosphate as a precursor for fatty acid esterification into triglycerides. PEPCK-C catalyzes the conversion of oxaloacetate + GTP to phosphoenolpyruvate + GDP + CO₂. It is in part the tissue-specificity of this simple reaction that results in the variety of phenotypes listed above. Briefly: (1) A 7-fold over-expression of PEPCK-C in the livers of mice causes excessive glucose production. (2) Mice with a whole-body knockout of Pck1 die within 2–3 days of birth, not from hypoglycemia, but probably because the Krebs cycle slows to approximately 10% of normal in the absence of cataplerosis. (3) Mice with a liver-specific knockout have an inability to remove oxaloacetate from the Krebs cycle, which leads to a fatty liver following a fast. (4) An adipose-specific knockout of Pck1 results in a fraction of the mice developing lipodystrophy due to lost glyceroneogenesis and a consequent decrease in fatty acid re-esterification. (5) Finally, disregulated over-expression of PEPCK-C in adipose tissue increases fatty acid re-esterification leading to obesity. These varied experimental phenotypes in mice have led us to postulate that abnormal production of PEPCK isozymes encoded by two PEPCK genes, PCK1 and PCK2, in humans could have similar consequences (Beale, E. G. et al. (2004). Trends in Endocrinology and Metabolism, 15, 129–135). The purpose of this review is to further explore these possibilities.     

Ranking candidate genes in rat models of type 2 diabetes

Background  

Rat models are frequently used to find genomic regions that contribute to complex diseases, so called quantitative trait loci (QTLs). In general, the genomic regions found to be associated with a quantitative trait are rather large, covering hundreds of genes. To help selecting appropriate candidate genes from QTLs associated with type 2 diabetes models in rat, we have developed a web tool called Candidate Gene Capture (CGC), specifically adopted for this disorder.     

Comprehensive analysis of photoreceptor gene expression and the identification of candidate retinal disease genes.

To identify the full set of genes expressed by mammalian rods, we conducted serial analysis of gene expression (SAGE) by using libraries generated from mature and developing mouse retina. We identified 264 uncharacterized genes that were specific to or highly enriched in rods. Nearly half of all cloned human retinal disease genes are selectively expressed in rod photoreceptors. In silico mapping of the human orthologs of genes identified in our screen revealed that 86 map within intervals containing uncloned retinal disease genes, representing 37 different loci. We expect these data will allow identification of many disease genes, and that this approach may be useful for cloning genes involved in classes of disease where cell type-specific expression of disease genes is observed.     

DNA polymorphism analysis of candidate genes for type 2 diabetes mellitus in a Mexican ethnic group

Type 2 diabetes mellitus is a complex metabolic disorder resulting from the action and interaction of many genetic and environmental factors. It has been reported that polymorphisms in genes involved in the metabolism of glucose are associated with the susceptibility to develop type 2 diabetes mellitus. Although the risk of developing type 2 diabetes mellitus increases with age, as well as with obesity and hypertension, its prevalence and incidence are different among geographical regions and ethnic groups. In Mexico, a higher prevalence and incidence has been described in the south of the country, and differences between urban and rural communities have been observed.We studied 73 individuals from Santiago Jamiltepec, a small indigenous community from Oaxaca State, Mexico. This population has shown a high prevalence of type 2 diabetes mellitus, and the aim of this study was to analyze the relationship between the Pst I (insulin gene), Nsi I (insulin receptor gene) and Gly972Arg (insulin receptor substrate 1 gene) polymorphisms and type 2 diabetes mellitus, obesity and hypertension in this population. Clinical evaluation consisted of BMI and blood pressure measurements, and biochemical assays consisted of determination of fasting plasma insulin and glucose levels. PCR and restriction enzyme digestion analysis were applied to genomic DNA to identify the three polymorphisms. From statistical analysis carried out here, individually, the Pst I, Nsi I and Gly972Arg polymorphisms were not associated with the type 2 diabetes, obese or hypertensive phenotypes in this population. Nevertheless, there was an association between the Nsi I and Pst I polymorphisms and increased serum insulin levels.     

复杂疾病基因鉴定的计算生物学方法

过去20多年复杂疾病易感基因鉴定的主要方法是连锁分析和关联研究。因为连锁分析确定的数量性状位点通常很宽, 加之对区域内大部分基因的功能以及基因功能和疾病之间联系的认识十分有限, 所以从数量性状位点到基因的识别是一个挑战。近年来发展了一些利用公共数据库的信息预测疾病易感基因的计算生物学方法。文章简要介绍了DGP、GeneSeeker、Prioritizer、PROSPECTR and SUSPECTS及Endeavor 5种计算生物学方法的基本原理, 以2型糖尿病/肥胖和骨质疏松症易感基因的预测为例说明它们的应用方法, 并讨论了这些方法的局限及应用前景。     

Identification of candidate genes in the type 2 diabetes modifier locus using expression QTL

To identify new genetic determinants relevant to type 2 diabetes (T2D), diabetic F2 progeny were generated by intercrossing F1 mice obtained from a cross of BKS.Cg-Lepr(db)+/+m and DBA/2, and T2D-related phenotypes were measured. In the F2 population, increased susceptibility to diabetes and obesity was observed. We also detected the major quantitative trait loci (QTL) modifying the severity of diabetes on chromosome 9, where peaks of logarithm of odds (LOD) overlapped for three traits. To identify candidate genes in the QTL intervals, we combined "expression QTL" (eQTL), taking mRNA levels as quantitative traits, and "interstrain sequence variations, including cSNPs." As a result, four genes were identified from cosegregation of clinical QTL with eQTL and 13 genes were found from interstrain cSNPs as candidates in the T2D modifier QTL. Our combined approach shows the acceleration of the discovery of candidate genes in the QTL of interest, spanning several megabases.     

MicroRNAs and type 2 diabetes/obesity

MicroRNAs belong to a newly identified class of small non-coding RNAs that have been widely implicated in the fine-tuning of many physiological processes such as the pathogenesis of type 2 diabetes(T2D) and obesity.Microarray studies have highlighted an altered profile of miRNA expression in insulin target tissues in diabetic and obese models.Emerging evidences suggest that miRNAs play significant roles in insulin production,secretion and actions,as well as in diverse aspects of glucose homeostasis and adipocyte differentiation. The identification of tissue-specific miRNAs implicated in T2D and obesity might be useful for the future development of effective strategies for early diagnosis and therapeutic intervention of obesity-related medical complications.     

Vaspin gene expression in human adipose tissue: association with obesity and type 2 diabetes

Recently, vaspin was identified as an adipokine with insulin-sensitizing effects, which is predominantly secreted from visceral adipose tissue in a rat model of type 2 diabetes. In this study, we examined whether vaspin mRNA expression is a marker of visceral obesity and correlates with anthropometric and metabolic parameters in paired samples of visceral and subcutaneous adipose tissue from 196 subjects with a wide range of obesity, body fat distribution, insulin sensitivity, and glucose tolerance. Vaspin mRNA expression was only detectable in 23% of the visceral and in 15% of the subcutaneous (SC) adipose tissue samples. Vaspin mRNA expression was not detectable in lean subjects (BMI<25) and was more frequently detected in patients with type 2 diabetes. No significant correlations were found between visceral vaspin gene expression and visceral fat area or SC vaspin expression. However, visceral vaspin expression significantly correlates with BMI, % body fat, and 2 h OGTT plasma glucose. Subcutaneous vaspin mRNA expression is significantly correlated with WHR, fasting plasma insulin concentration, and glucose infusion rate during steady state of an euglycemic-hyperinsulinemic clamp. Multivariate linear regression analysis revealed % body fat as strongest predictor of visceral vaspin and insulin sensitivity as strongest determinant of SC vaspin mRNA expression. In conclusion, our data indicate that induction of human vaspin mRNA expression in adipose tissue is regulated in a fat depot-specific manner and could be associated with parameters of obesity, insulin resistance, and glucose metabolism.     

Analysis of PBX1 as a candidate gene for type 2 diabetes mellitus in Pima Indians

The human proto-oncogene PBX1 codes for a homeodomain containing protein that modulates expression of several genes, including those contributing to regulation of insulin action and glucose metabolism. PBX1 is located on chromosome 1q22, a region linked with type 2 diabetes in Pima Indians, Caucasians, and an Old Order Amish population. We have investigated the PBX1 genomic sequence to identify polymorphisms that may contribute to diabetes susceptibility in the Pimas. PBX1 is composed of nine exons spanning approx. 117 kb and is located within 300 kb of microsatellite D1S1677, which marks the peak of linkage to diabetes susceptibility in the Pima Indians. We detected 16 single nucleotide polymorphisms in PBX1 including one causing a glycine to serine substitution at residue 21. Comparison of the frequencies of the polymorphisms between affected and unaffected Pima Indians did not detect any significant differences, indicating that mutations in PBX1 do not explain the linkage of 1q with type 2 diabetes in this population. The genomic structure of PBX1 provides a basis for similar systematic examinations of this candidate locus in other populations in relation to both type 2 diabetes and other metabolic disorders.     

Ranking candidate disease genes from gene expression and protein interaction: a Katz-centrality based approach

Many diseases have complex genetic causes, where a set of alleles can affect the propensity of getting the disease. The identification of such disease genes is important to understand the mechanistic and evolutionary aspects of pathogenesis, improve diagnosis and treatment of the disease, and aid in drug discovery. Current genetic studies typically identify chromosomal regions associated specific diseases. But picking out an unknown disease gene from hundreds of candidates located on the same genomic interval is still challenging. In this study, we propose an approach to prioritize candidate genes by integrating data of gene expression level, protein-protein interaction strength and known disease genes. Our method is based only on two, simple, biologically motivated assumptions--that a gene is a good disease-gene candidate if it is differentially expressed in cases and controls, or that it is close to other disease-gene candidates in its protein interaction network. We tested our method on 40 diseases in 58 gene expression datasets of the NCBI Gene Expression Omnibus database. On these datasets our method is able to predict unknown disease genes as well as identifying pleiotropic genes involved in the physiological cellular processes of many diseases. Our study not only provides an effective algorithm for prioritizing candidate disease genes but is also a way to discover phenotypic interdependency, cooccurrence and shared pathophysiology between different disorders.     

Capn10, a candidate gene responsible for type 2 diabetes mellitus in the OLETF rat

The rat strain Otsuka Long-Evans Tokushima Fatty (OLETF) is an animal model for type 2 diabetes mellitus. Nidd8/of has been identified as one of 14 quantitative trait loci (QTLs) involved in the diabetes by a whole genome search in 160 F2 progenies obtained by mating the OLETF and F344 rats. Comparative mapping between human and rat indicated that the Nidd8/of genomic region, near D9rat21 on rat chromosome 9, contains the calpain10 (Capn10) gene, which is putative type 2 diabetes-susceptibility gene in humans. In this study, we found no difference in Capn10 mRNA expression in the heart, liver, skeletal muscle and pancreas between OLETF and F344 rats at 5 and 10 weeks of age. However, we found a single nucleotide polymorphism (SNP) (A/A genotype in OLETF and G/G genotype in F344 and LETO rats) at the base 583 downstream from the translation start site in the rat Capn10 cDNA sequence. This SNP was deduced to substitute serine (OLETF) for glycine (F344 and LETO) at the 195 amino acid residue within the protease domain of rat Capn10. Because serine is generally not interchangeable with glycine in respect of the protein structure and function, it was deduced that the A/A genotype in OLETF is not a 'safe' mutation. This non-conservative amino acid substitution might be associated with susceptibility to type 2 diabetes in OLETF rats.     

Bioinformatics-driven identification and examination of candidate genes for non-alcoholic fatty liver disease

Objective

Candidate genes for non-alcoholic fatty liver disease (NAFLD) identified by a bioinformatics approach were examined for variant associations to quantitative traits of NAFLD-related phenotypes.

Research Design and Methods

By integrating public database text mining, trans-organism protein-protein interaction transferal, and information on liver protein expression a protein-protein interaction network was constructed and from this a smaller isolated interactome was identified. Five genes from this interactome were selected for genetic analysis. Twenty-one tag single-nucleotide polymorphisms (SNPs) which captured all common variation in these genes were genotyped in 10,196 Danes, and analyzed for association with NAFLD-related quantitative traits, type 2 diabetes (T2D), central obesity, and WHO-defined metabolic syndrome (MetS).

Results

273 genes were included in the protein-protein interaction analysis and EHHADH, ECHS1, HADHA, HADHB, and ACADL were selected for further examination. A total of 10 nominal statistical significant associations (P<0.05) to quantitative metabolic traits were identified. Also, the case-control study showed associations between variation in the five genes and T2D, central obesity, and MetS, respectively. Bonferroni adjustments for multiple testing negated all associations.

Conclusions

Using a bioinformatics approach we identified five candidate genes for NAFLD. However, we failed to provide evidence of associations with major effects between SNPs in these five genes and NAFLD-related quantitative traits, T2D, central obesity, and MetS.     

A search for candidate genes for lipodystrophy,obesity and diabetes via gene expression analysis of A-ZIP/F-1 mice

Genome scans for diabetes have identified many regions of the human genome that correlate with the disease state. To identify candidate genes for type 2 diabetes, we examined the transgenic A-ZIP/F-1 mouse. This mouse model has no white fat, resulting in abnormal levels of glucose, insulin, and leptin, making the A-ZIP/F-1 mice a good model for lipodystrophy and insulin resistance. We used cDNA-based microarrays to find differentially expressed genes in four tissues of these mice. We examined these results in the context of human linkage scans for lipodystrophy, obesity, and type 2 diabetes. We combined 199 known human orthologs of the misregulated mouse genes with 33 published human genome scans on a genome map. Integrating expression data with human linkage results permitted us to suggest and prioritize candidate genes for lipodystrophy and related disorders. These genes include a cluster of 3 S100A genes on chromosome 1 and SLPI1 on chromosome 20.