'OMICs' approach to dissect gene-gene and gene-nutrient interaction in complex diseases and associated traits
Chronic Pancreatitis using Tropical Calcific Pancreatitis as a model
We had never imagined that, what began as a simple replication study of observations made on chronic pancreatitis (CP) patients of Caucasian origin would blossom into a full-fledged career objective for us. It was in 1996 that David Whitcomb and his group identified ‘Gain of Function’ mutation in the cationic trypsinogen gene (encoded by PRSS1) to be causally associated with CP. This proved the 100-year-old hypothesis proposed by Chiari in 1896 that premature intrapancreatic activation of trypsinogen causes CP. Since distinctive contrasting features such as early age of onset, large intra-ductal calculi, accelerated course of the disease and a high susceptibility to diabetes and pancreatic cancer compared to Western CP characterize the common form of CP in Indians, termed as ‘Tropical calcific pancreatitis (TCP)’, our laboratory, in 2000 asked a simple question, “How would mutations in PRSS1 alone cause such extreme phenotypes that characterize one clinical condition, chronic pancreatitis”. The idea was to investigate the influence of nutritional factors in modulating the genetic susceptibility and thus leading to severe phenotype of TCP.
To our utter surprise, we could neither detect the reported nor any novel mutations in the PRSS1 gene in TCP or any other form of CP (Alcoholic, Hereditary and Idiopathic) through multi-centric studies in the country. However, we observed strong association of a founder mutation (p.N34S) in pancreatic secretory trypsin inhibitor (encoded by SPINK1), which established that TCP is a genetic disease and has different genetic susceptibility compared to individuals abroad. This was further substantiated by unique association of propeptide variants in cathepsin B (CTSB) with TCP. These observations led to inclusion of TCP as a genetic disease in Online Mendelian Inheritance in Man (OMIM, #608189) with SPINK1 and CTSB variants predicting susceptibility to its occurrence. We further demonstrated lack of any role for coding and copy number variants in other trypsinogen genes thus confirming no direct role for trypsinogen genes in the pathogenesis of CP in Indians. Thus, we proposed a different model for genetic susceptibility to CP in Indians and concluded that clinical genetic testing of trypsinogen genes has no predictive value but SPINK1 and CTSB variants will be more relevant to predict risk susceptibility to CP in Indians.
Subsequent to these observations, we have identified different spectrum of mutations in Chymotrypsin C (CTRC) and Carboxypeptidase A1 (CPA1), which shows that besides genetic heterogeneity, allelic heterogeneity also plays important role in the pathogenesis of CP in Indians. Overall, our studies in Indian CP patients laid foundation to the over-riding hypothesis that mutational and genetic heterogeneity may underlie the susceptibility to complex diseases in India, especially if they have a variable clinical phenotype.
Consolidating on the observations from candidate gene studies, we have performed gene-expression profiling on pancreatic tissue samples and identified differential regulation of pathways involved in tissue remodeling (apoptosis-necrosis model), cell differentiation, inflammatory responses, insulin and calcium signaling playing key role in the pathophysiology of CP. Results so far indicate that both 'trypsin-dependent' and 'trypsin-independent' pathways contribute to the development of TCP and associated phenotypes.
To further expand our understanding of the molecular pathways involved in the disease, we are undertaking a System Biology approach by exploring genetic (further candidate gene and transcriptomic studies) and epigenetic (methylation, microRNAs) mechanisms regulating the gene expression in the diseased condition.
We had never imagined that, what began as a simple replication study of observations made on chronic pancreatitis (CP) patients of Caucasian origin would blossom into a full-fledged career objective for us. It was in 1996 that David Whitcomb and his group identified ‘Gain of Function’ mutation in the cationic trypsinogen gene (encoded by PRSS1) to be causally associated with CP. This proved the 100-year-old hypothesis proposed by Chiari in 1896 that premature intrapancreatic activation of trypsinogen causes CP. Since distinctive contrasting features such as early age of onset, large intra-ductal calculi, accelerated course of the disease and a high susceptibility to diabetes and pancreatic cancer compared to Western CP characterize the common form of CP in Indians, termed as ‘Tropical calcific pancreatitis (TCP)’, our laboratory, in 2000 asked a simple question, “How would mutations in PRSS1 alone cause such extreme phenotypes that characterize one clinical condition, chronic pancreatitis”. The idea was to investigate the influence of nutritional factors in modulating the genetic susceptibility and thus leading to severe phenotype of TCP.
To our utter surprise, we could neither detect the reported nor any novel mutations in the PRSS1 gene in TCP or any other form of CP (Alcoholic, Hereditary and Idiopathic) through multi-centric studies in the country. However, we observed strong association of a founder mutation (p.N34S) in pancreatic secretory trypsin inhibitor (encoded by SPINK1), which established that TCP is a genetic disease and has different genetic susceptibility compared to individuals abroad. This was further substantiated by unique association of propeptide variants in cathepsin B (CTSB) with TCP. These observations led to inclusion of TCP as a genetic disease in Online Mendelian Inheritance in Man (OMIM, #608189) with SPINK1 and CTSB variants predicting susceptibility to its occurrence. We further demonstrated lack of any role for coding and copy number variants in other trypsinogen genes thus confirming no direct role for trypsinogen genes in the pathogenesis of CP in Indians. Thus, we proposed a different model for genetic susceptibility to CP in Indians and concluded that clinical genetic testing of trypsinogen genes has no predictive value but SPINK1 and CTSB variants will be more relevant to predict risk susceptibility to CP in Indians.
Subsequent to these observations, we have identified different spectrum of mutations in Chymotrypsin C (CTRC) and Carboxypeptidase A1 (CPA1), which shows that besides genetic heterogeneity, allelic heterogeneity also plays important role in the pathogenesis of CP in Indians. Overall, our studies in Indian CP patients laid foundation to the over-riding hypothesis that mutational and genetic heterogeneity may underlie the susceptibility to complex diseases in India, especially if they have a variable clinical phenotype.
Consolidating on the observations from candidate gene studies, we have performed gene-expression profiling on pancreatic tissue samples and identified differential regulation of pathways involved in tissue remodeling (apoptosis-necrosis model), cell differentiation, inflammatory responses, insulin and calcium signaling playing key role in the pathophysiology of CP. Results so far indicate that both 'trypsin-dependent' and 'trypsin-independent' pathways contribute to the development of TCP and associated phenotypes.
To further expand our understanding of the molecular pathways involved in the disease, we are undertaking a System Biology approach by exploring genetic (further candidate gene and transcriptomic studies) and epigenetic (methylation, microRNAs) mechanisms regulating the gene expression in the diseased condition.
Type 2 Diabetes (T2D) and its clinical course
Type 2 diabetes (T2D) is known to hit Indians at least a decade earlier and at a lower body mass index (BMI) than the Europeans. Interaction of 'thrifty phenotype', increased food consumption and decreased physical activity are attributed to the alarming rise in the prevalence of diabetes in India. Despite having the greatest number of diabetic subjects in any one country, genetic basis of T2D and related intermediate traits is not completely understood in India.
In order to understand the above, we have generated a multi-centre collaboration from different parts of the country that takes care of ethnic diversity as well as provides comprehensive coverage of Indian population. Additionally, these cohorts provide details of clinical course and thus allow us to investigate gene-environment interaction in determining the basis of various micro-vascular complications.
The boom of Genome Wide Association Studies (GWAS), about a decade ago, accelerated the discovery of genetic variants associated with T2D in Caucasians. Our attempts to replicate these results in an adequately powered cohort showed the contrast in genetic susceptibility of T2D in Indians. While we replicated the strong association of TCF7L2 and 8 other common variants with T2D in Indians, albeit with larger effect size, we found that FTO variants that predict risk of T2D through BMI in Caucasians, do not entirely mediate the risk through BMI and other anthropometric parameters such as waist-hip ratio etc. As a result, our initial study challenged the role of FTO as an exclusive obesity gene, which we further confirmed through meta-analysis of various studies across the world.
These observations led us to an Indian GWAS to identify T2D susceptibility loci specific to Indian population. In addition to replicating several European T2D loci, we have identified 3 novel loci with GWAS significance and 4 loci with suggestive significance to be associated with T2D and related intermediate traits such as obesity, fat-mass, insulin resistance and secretion in Indians. Currently, we are performing targeted Next Generation Sequencing (NGS) to identify variants that will prove causal association of these regions with T2D. We are also undertaking both, in vivo and in vitro approaches using animal and cell line models to determine the functional impact of the identified loci.
Type 2 diabetes (T2D) is known to hit Indians at least a decade earlier and at a lower body mass index (BMI) than the Europeans. Interaction of 'thrifty phenotype', increased food consumption and decreased physical activity are attributed to the alarming rise in the prevalence of diabetes in India. Despite having the greatest number of diabetic subjects in any one country, genetic basis of T2D and related intermediate traits is not completely understood in India.
In order to understand the above, we have generated a multi-centre collaboration from different parts of the country that takes care of ethnic diversity as well as provides comprehensive coverage of Indian population. Additionally, these cohorts provide details of clinical course and thus allow us to investigate gene-environment interaction in determining the basis of various micro-vascular complications.
The boom of Genome Wide Association Studies (GWAS), about a decade ago, accelerated the discovery of genetic variants associated with T2D in Caucasians. Our attempts to replicate these results in an adequately powered cohort showed the contrast in genetic susceptibility of T2D in Indians. While we replicated the strong association of TCF7L2 and 8 other common variants with T2D in Indians, albeit with larger effect size, we found that FTO variants that predict risk of T2D through BMI in Caucasians, do not entirely mediate the risk through BMI and other anthropometric parameters such as waist-hip ratio etc. As a result, our initial study challenged the role of FTO as an exclusive obesity gene, which we further confirmed through meta-analysis of various studies across the world.
These observations led us to an Indian GWAS to identify T2D susceptibility loci specific to Indian population. In addition to replicating several European T2D loci, we have identified 3 novel loci with GWAS significance and 4 loci with suggestive significance to be associated with T2D and related intermediate traits such as obesity, fat-mass, insulin resistance and secretion in Indians. Currently, we are performing targeted Next Generation Sequencing (NGS) to identify variants that will prove causal association of these regions with T2D. We are also undertaking both, in vivo and in vitro approaches using animal and cell line models to determine the functional impact of the identified loci.
1-C metabolic cycle
Neural Tube Defects
Neural tube defects (NTDs) are one of the common severe congenital malformations. Its prevalence is one of the highest and quite variable across different parts of the country (ranging from 0.6 to 11.4 per 1000) thus representing a long-term public health problem in India. Impairment in one-carbon metabolism reflected by elevated maternal serum homocysteine levels and presence of homocysteine raising variants like 677C>T and 1298A>C in methylene tetra hydrofolate reductase gene (MTHFR) have been linked to NTDs. Series of studies from different parts of the world have demonstrated that peri-conceptional folic acid supplementation prevents occurrence and recurrence of NTD in the offspring. India has also recommended the above following on results of MRC, UK trial, although there is no comprehensive study to investigate the role of folic acid and one-carbon metabolism in the causation of NTDs. This assumes importance in the background of the observations from earlier studies that large number of Indians are hyperhomocysteinemic and vitamin B12 deficiency is more prevalent (~50-60%) than folate deficiency (~1-2%) in Indian population.
In one of the first most comprehensive multi-center study comprising 370 case and 740 control trios from four parts of India, we confirmed higher maternal homocysteine levels in mothers with an NTD child and established lack of association of maternal folate deficiency and MTHFR risk variants with NTDs in their children. In contrast, we identified low holo-transcobalamin levels (representative of tissue B12 levels) in case mothers and thus demonstrated a significant role of B12 deficiency and polymorphisms in transcobalamin (TCN2) in the etiology of NTDs in Indians. Together, our study queries the rationale of peri-conceptional folate supplementation to pregnant mothers and indicates a need to revisit this policy by including B12 supplementation in the regimen.
Currently, we are investigating the role of variants in 20 other genes that influence one-carbon metabolism to comprehensively dissect gene-gene and gene-nutrient interaction in the pathogenesis of NTDs. We are also undertaking exome sequencing approach in select families with multiple affected children to identify rare variants with large effect size. This will aid in understanding causal pathways in this disease that has demonstrated classical gene-gene and gene-nutrient interaction and provides unique opportunity to prevent the genetic burden on the Indian society.
Neural tube defects (NTDs) are one of the common severe congenital malformations. Its prevalence is one of the highest and quite variable across different parts of the country (ranging from 0.6 to 11.4 per 1000) thus representing a long-term public health problem in India. Impairment in one-carbon metabolism reflected by elevated maternal serum homocysteine levels and presence of homocysteine raising variants like 677C>T and 1298A>C in methylene tetra hydrofolate reductase gene (MTHFR) have been linked to NTDs. Series of studies from different parts of the world have demonstrated that peri-conceptional folic acid supplementation prevents occurrence and recurrence of NTD in the offspring. India has also recommended the above following on results of MRC, UK trial, although there is no comprehensive study to investigate the role of folic acid and one-carbon metabolism in the causation of NTDs. This assumes importance in the background of the observations from earlier studies that large number of Indians are hyperhomocysteinemic and vitamin B12 deficiency is more prevalent (~50-60%) than folate deficiency (~1-2%) in Indian population.
In one of the first most comprehensive multi-center study comprising 370 case and 740 control trios from four parts of India, we confirmed higher maternal homocysteine levels in mothers with an NTD child and established lack of association of maternal folate deficiency and MTHFR risk variants with NTDs in their children. In contrast, we identified low holo-transcobalamin levels (representative of tissue B12 levels) in case mothers and thus demonstrated a significant role of B12 deficiency and polymorphisms in transcobalamin (TCN2) in the etiology of NTDs in Indians. Together, our study queries the rationale of peri-conceptional folate supplementation to pregnant mothers and indicates a need to revisit this policy by including B12 supplementation in the regimen.
Currently, we are investigating the role of variants in 20 other genes that influence one-carbon metabolism to comprehensively dissect gene-gene and gene-nutrient interaction in the pathogenesis of NTDs. We are also undertaking exome sequencing approach in select families with multiple affected children to identify rare variants with large effect size. This will aid in understanding causal pathways in this disease that has demonstrated classical gene-gene and gene-nutrient interaction and provides unique opportunity to prevent the genetic burden on the Indian society.
Feto-Maternal interactions in programming of metabolic syndrome and related intermediate traits
Chronic non-communicable diseases (NCDs), type 2 diabetes (T2D) and cardiovascular disease [CVD; ‘cardio-metabolic disease’) are leading causes of morbidity and premature death worldwide. T2D is a full-blown epidemic in India. According to International Diabetes Federation, India harbors an estimated 65 million diabetic adults and this is set to rise to 109 million by 2035. The prevalence of diabetes in India has increased in geometric progression over last 2-3 decades, which suggests influence of diet and life style changes in addition to genetic susceptibility to be important factors in causing this epidemic. Although, close to 70 T2D genetic loci have been identified till date, their contribution to heritability is only 10-12%. Further, it is difficult to change the genetic constitution of individuals as a way to modify the risk susceptibility. Current approaches to preventing NCDs focus mainly on treating adult obesity and risk factors like glucose intolerance, which can, at best be called secondary prevention at an ‘end-stage’ of the problem but not enough to stem the predicted future epidemics of above NCDs. Hence, in order to prevent this menace, it is crucial to identify the stage at which the susceptibility is first established, to identify the modifiable factors and the window of opportunity when the probable interventions may work. In recent times, the idea of prevention of diabetes has taken a paradigm shift with identification of intrauterine environment and feto-maternal interactions to be major determinants of diabetes and related intermediate traits including obesity, insulin resistance and secretion.
Epidemiological studies over last two decades have confirmed the long-term consequences of impaired growth and development in early life such that people of lower birthweight were at higher risk of CVD, T2D and metabolic syndrome in adult life. These studies led to the ‘Developmental Origins of Health and Disease’ (DOHaD) hypothesis, which proposes that when a fetus or young child is sub-optimally nourished, its metabolism becomes permanently altered (‘programmed’) resulting in impaired childhood development and vulnerability to adult chronic diseases. This is possible through maternal under-nutrition (Nutrient-Mediated Teratogenesis; NMT) or through over-nutrition (Fuel-Mediated Teratogenesis; FMT).
Our group is dynamic part of consortium of prospective cohorts (SNEHA) set up to explore the relationship between maternal nutrition and fetal growth and future risk of T2D. Supporting the notion of NMT, maternal hyperhomocysteinemia (especially high folate and low B12 levels) have been shown to be associated with insulin resistance and larger fat mass in child [Pune Maternal Nutrition Study (PMNS; C S Yajnik)]. On the same note, maternal diabetes was found to be associated with higher glucose and insulin concentrations in children and greater adiposity, reassuring the idea of FMT [Parthenon Study; Caroline Fall]. Using homocysteine raising genetic variants in MTHFR and other genetic variants influencing the homocysteine levels in a mendelian randomization approach, we have established causal role of maternal homocysteine (due to folate/B12 imbalance) in fetal growth.
Since homocysteine is a summative marker of one-carbon metabolism (OCM), which regulates gene expression through methylation, we are currently investigating the pathways that are influenced due to B12 deficiency. We are taking multi-pronged approach to identify the molecular basis of NMT due to B12 deficiency. These include comparison of methylome of children born to mothers with and without hyperhomocysteinemia and methylome of children before and after B12 and B12/folate intervention, followed by pathway analysis and confirmation of functional bases of differentially methylated regions. We are performing similar analysis to investigate the FMT by comparing the cord blood methylome of gestational diabetic mothers and normal mothers. Finally, in a recently funded study, we will be investigating the OMIC profile of cord blood and placenta of kids born to women with and without B12 intervention. Overall, these approaches will allow us to identify pathways (common or unique) that programme similar intermediate risk traits like obesity and insulin resistance and aid in proposing probable interventions to augment maternal (parental) health and prevent the epidemic of NCD in future generations.
Chronic non-communicable diseases (NCDs), type 2 diabetes (T2D) and cardiovascular disease [CVD; ‘cardio-metabolic disease’) are leading causes of morbidity and premature death worldwide. T2D is a full-blown epidemic in India. According to International Diabetes Federation, India harbors an estimated 65 million diabetic adults and this is set to rise to 109 million by 2035. The prevalence of diabetes in India has increased in geometric progression over last 2-3 decades, which suggests influence of diet and life style changes in addition to genetic susceptibility to be important factors in causing this epidemic. Although, close to 70 T2D genetic loci have been identified till date, their contribution to heritability is only 10-12%. Further, it is difficult to change the genetic constitution of individuals as a way to modify the risk susceptibility. Current approaches to preventing NCDs focus mainly on treating adult obesity and risk factors like glucose intolerance, which can, at best be called secondary prevention at an ‘end-stage’ of the problem but not enough to stem the predicted future epidemics of above NCDs. Hence, in order to prevent this menace, it is crucial to identify the stage at which the susceptibility is first established, to identify the modifiable factors and the window of opportunity when the probable interventions may work. In recent times, the idea of prevention of diabetes has taken a paradigm shift with identification of intrauterine environment and feto-maternal interactions to be major determinants of diabetes and related intermediate traits including obesity, insulin resistance and secretion.
Epidemiological studies over last two decades have confirmed the long-term consequences of impaired growth and development in early life such that people of lower birthweight were at higher risk of CVD, T2D and metabolic syndrome in adult life. These studies led to the ‘Developmental Origins of Health and Disease’ (DOHaD) hypothesis, which proposes that when a fetus or young child is sub-optimally nourished, its metabolism becomes permanently altered (‘programmed’) resulting in impaired childhood development and vulnerability to adult chronic diseases. This is possible through maternal under-nutrition (Nutrient-Mediated Teratogenesis; NMT) or through over-nutrition (Fuel-Mediated Teratogenesis; FMT).
Our group is dynamic part of consortium of prospective cohorts (SNEHA) set up to explore the relationship between maternal nutrition and fetal growth and future risk of T2D. Supporting the notion of NMT, maternal hyperhomocysteinemia (especially high folate and low B12 levels) have been shown to be associated with insulin resistance and larger fat mass in child [Pune Maternal Nutrition Study (PMNS; C S Yajnik)]. On the same note, maternal diabetes was found to be associated with higher glucose and insulin concentrations in children and greater adiposity, reassuring the idea of FMT [Parthenon Study; Caroline Fall]. Using homocysteine raising genetic variants in MTHFR and other genetic variants influencing the homocysteine levels in a mendelian randomization approach, we have established causal role of maternal homocysteine (due to folate/B12 imbalance) in fetal growth.
Since homocysteine is a summative marker of one-carbon metabolism (OCM), which regulates gene expression through methylation, we are currently investigating the pathways that are influenced due to B12 deficiency. We are taking multi-pronged approach to identify the molecular basis of NMT due to B12 deficiency. These include comparison of methylome of children born to mothers with and without hyperhomocysteinemia and methylome of children before and after B12 and B12/folate intervention, followed by pathway analysis and confirmation of functional bases of differentially methylated regions. We are performing similar analysis to investigate the FMT by comparing the cord blood methylome of gestational diabetic mothers and normal mothers. Finally, in a recently funded study, we will be investigating the OMIC profile of cord blood and placenta of kids born to women with and without B12 intervention. Overall, these approaches will allow us to identify pathways (common or unique) that programme similar intermediate risk traits like obesity and insulin resistance and aid in proposing probable interventions to augment maternal (parental) health and prevent the epidemic of NCD in future generations.
Web Design: Divya Sri Priyanka T