Posted by Phillipa on December 8, 2008, at 12:49:20
Thought this might to be interest to some. Phillipa
Melatonin Receptor Gene Linked to Glucose Levels, Diabetes Risk
December 8, 2008 Variants in MTNR1B, a gene that codes for human melatonin receptor MT2, are associated with fasting glucose levels and risk for type 2 diabetes, according to 3 studies appearing online December 7 in Nature Genetics. Two studies found that rs10830963, a single nucleotide polymorphism (SNP) in MTNR1B, was significantly associated with fasting glucose levels or insulin secretion. A third study identified rs1387153, near MTNR1B, as the SNP most strongly associated with fasting plasma glucose (FPG).Melatonin receptors are present in the human brain, retina, and pancreatic islets. Secretion of melatonin, a neurohormone that regulates circadian rhythm, is lowest during the day and peaks at night, which is opposite the pattern of insulin secretion. Abnormal melatonin secretion and circadian rhythm have been previously noted in type 2 diabetes.
Nevertheless, "the idea that [circadian control] may be linked genetically to risk of impaired glucose regulation, or even diabetes, is completely new," said Joseph Bass, MD, PhD, commenting by telephone to Medscape Pathology & Lab Medicine. Dr. Bass is assistant professor of medicine and neurobiology, Northwestern University, and division head of endocrinology and metabolism, Evanston Northwestern Healthcare Research Institute, Evanston, Illinois.
Same SNP in MTNR1B Linked With Glucose Regulation
The first study consortium, the Meta-Analyses of Glucose and Insulin-Related Traits Consortium, drew on data from 10 genomewide association scans to identify loci most strongly linked to fasting glucose levels in a study population (n = 36,610) of European descent. Of 981 SNPs within a 1-Mb region containing MTNR1B, 30 showed association (P < 10−8) with fasting glucose, but the association of rs10830963 with fasting glucose was strongest. The G allele had a dose-related effect, with each copy of this allele conferring a .07 mmol/L increase in fasting glucose levels (95% confidence interval [CI], .06 .08; P = 3.2 × 10−50).
The article also noted that "strength of the association between rs10830963 and fasting glucose was unchanged after adjustment for body mass index [BMI]."
Medscape Pathology & Lab Medicine contacted senior author Gonçalo R. Abecasis, PhD, assistant professor, Center for Statistical Genetics, Department of Biostatistics, University of Michigan School of Public Health, Ann Arbor, regarding the study.
"Basically, we know that individuals with increased BMI can also show increased glucose levels as a consequence," Dr. Abecasis said in an email response. "So, we wanted to separate the chicken from the egg. Showing that adjusting for BMI did not account for the MTNR1B association with glucose suggests that these variants are influencing glucose levels 'directly' and not through BMI," he pointed out.
The second paper reported significant association between an allele (G) of the same SNP, rs10830963, and the risk of developing type 2 diabetes. Subjects for the prospective study were Swedish (n = 16,061) and Finnish (n = 2770) individuals from previous study populations; 2201 of these individuals were later diagnosed with diabetes (median follow-up, 23.5 years). The G allele was linked to a 1.11-fold increase in the risk of developing type 2 diabetes (95% CI, 1.03 1.18; P = .004).
Another facet of this study, involving 3300 nondiabetic individuals, found a dose-dependent relationship between the number of G alleles and early insulin response to glucose (corrected early insulin response to glucose (CIR): linear regression coefficient beta = −.170 ± .021; P = 5 × 10−16). Thus, as the number of G alleles increased, CIR decreased.
The authors suggest therapeutic applications of these findings: Namely, that type 2 diabetes might be treated by blocking the melatonin receptor system, thus preventing the associated decrease in insulin response. Senior author Leif Groop, MD, PhD, professor and head of unit, Department of Clinical Sciences/Diabetes & Endocrinology Research Unit, Lund University, University Hospital Malmö, Sweden, reflected on the therapeutic potential in an email to Medscape Pathology & Lab Medicine.
One of the challenges would be to ensure "that the effect would be restricted to islets. [T]here are inhibitors available, the important thing is that they do not too much interfere with other serotonin pathways in the brain," said Dr. Groop. This is a serious concern because "melatonin is a metabolite of serotonin."
Other SNPs Near MTNR1B Still Viable Candidates
The third study used genomewide association data from 2151 nondiabetic French subjects, both lean and obese, children and adults, and found 2 new SNPs significantly associated with FPG at the genomewide level. Replication analysis in nondiabetic individuals (n = 3886) confirmed only rs1387153, allele T, as a risk allele for increased FPG (beta, .08 mmol/L; 95% CI, .06 .11; P = 6.9 × 10−12).
The T allele of rs1387153 was also found to increase the risk for type 2 diabetes (odds ratio, 1.15; 95% CI, 1.08 1.22; P = 6.3 × 10−5). rs1387153 is located 28.3 kb from MTNR1B and is in linkage disequilibrium with several SNPs, including rs10830963 (the focus of the 2 preceding articles). In the present study, rs1387153 demonstrated the strongest association with FPG. However, the authors acknowledge that "all SNPs of the LD block including MTNR1B are putative candidates for...the association signal with FPG."
Senior author Philippe Froguel, MD, PhD, director of the UMR Centre National de la Recherche Scientifique, Institute of Biology and Lille University, Pasteur Institute, France, and professor and head of the Section of Genomic Medicine, Hammersmith Hospital, Imperial College London, United Kingdom, explained linkage disequilibrium to Medscape Pathology & Lab Medicine.
"Linkage disequilibrium [LD] is a phenomenon reflecting the genetic distance between DNA nucleotides, if DNA variants located on the same chromosome are transmitted together from 1 generation to another," Dr. Froguel said via email. "If the LD is 1, they are always transmitted together from a parent to his/her child.... Practically, SNPs in strong LD are difficult to distinguish, which means difficult to determine what is the 'causal' one (and the others are markers)."
He continued, "although we are sure that these SNPs support the association, it is not possible at this stage to identify the 1 that is etiological. That is the work we are currently doing." However, Dr. Froguel noted: "What is most interesting here is the contribution of the melatonin pathway, the light-sensor of the brain (regulating the circadian clock), on glucose regulation."
The Case Is Far From Closed
Taken together, the 3 studies contribute to understanding the relationship between melatonin-regulated circadian rhythm and insulin secretion. However, many questions remain. Medscape Pathology & Lab Medicine contacted Alessandro Doria, MD, PhD, MPH, director of the Genetics Core, Joslin Diabetes Center, and associate professor of medicine, Harvard School of Medicine, Boston, Massachusetts, who commented on the 3 studies by email.
"The existence of a circadian rhythm of insulin secretion and, more broadly, of glucose metabolism, has been known for many years.... It has been speculated that such rhythm may be related to the changes in melatonin levels that occur between day and night," said Dr. Doria. "If a role of melatonin as a modulator of insulin secretion were confirmed, this would provide some support for this hypothesis.
"On the other hand, some studies have shown that melatonin may also be produced by cells other than those in the pineal gland, independently from the light cycle," continued Dr. Doria. "So, a connection between melatonin and insulin secretion may not necessarily explain the circadian rhythm of glucose metabolism." In his opinion, "We need more data, both in vitro and in vivo, before we start discussing possible therapeutic applications."
poster:Phillipa
thread:867483
URL: http://www.dr-bob.org/babble/20081204/msgs/867483.html