Rare MTNR1B variants causing diminished MT2 signalling associate with elevated HbA levels but not with type 2 diabetes.

Aims/hypothesis: An intronic variant (rs10830963) in MTNR1B (encoding the melatonin receptor type 2 [MT2]) has been shown to strongly associate with impaired glucose regulation and elevated type 2 diabetes prevalence. However, MTNR1B missense variants have shown conflicting results on type 2 diabetes. Thus, we aimed to gain further insights into the impact of MTNR1B coding variants on type 2 diabetes prevalence and related phenotypes.

Methods: We conducted a cross-sectional study, performing MTNR1B variant burden testing of glycaemic phenotypes (N=248,454, without diabetes), other cardiometabolic phenotypes (N=330,453) and type 2 diabetes prevalence (case-control study; N=263,739) in the UK Biobank. Similar burden testing with glycaemic phenotypes was performed in Danish Inter99 participants without diabetes (N=5711), and type 2 diabetes prevalence (DD2 cohort serving as cases [N=2930] and Inter99 serving as controls [N=4243]). Finally, we evaluated the effects of MTNR1B variants on the melatonin-induced glucose regulation response in a recall-by-genotype study of individuals without diabetes.

Results: In the UK Biobank, MTNR1B variants were not associated with cardiometabolic phenotypes, including type 2 diabetes prevalence, except that carriers of missense MTNR1B variants causing impaired MT2 signalling exhibited higher HbA levels compared with non-carriers (effect size, β, 0.087 SD [95% CI 0.039, 0.135]). Similarly, no significant associations were observed with phenotypes associated with glycaemic phenotypes in the Inter99 population. However, carriers of variants impairing MT2 signalling demonstrated a nominally significant lower glucose-stimulated insulin response (β -0.47 SD [95% CI -0.82, -0.11]). A reduced insulin response was also observed in carriers of variants impairing MT2 signalling (β -476.0 [95% CI -928.6, -24.4]) or the rs10830963 variant (β -390.8 [95% CI -740.1, -41.6]) compared with non-carriers after melatonin treatment.

Conclusions/interpretation: The higher type 2 diabetes prevalence previously observed in carriers of missense MTNR1B variants causing impairment in MT2 signalling was not replicated in the UK Biobank, yet carriers had elevated HbA levels.

Data Availability: Data (Inter99 cohort and recall-by-genotype study) are available on reasonable request from the corresponding author. Requests for DD2 data are through the application form at https://dd2.dk/forskning/ansoeg-om-data . Access to UK Biobank data can be requested through the UK Biobank website ( https://www.ukbiobank.ac.uk/enable-your-research ).