Association of Polymorphisms in GPR10, the Gene Encoding the Prolactin-Releasing Peptide Receptor With Blood Pressure, but not Obesity, in a U.K. Caucasian Population

The neuropeptide prolactin-releasing peptide (PrRP) and its cognate G-protein-coupled receptor, GPR10, were initially identified as possible regulators of prolactin release from the anterior pituitary lactotroph. Subsequently, however, some doubt has been cast on the assertion that the main physiological role of PrRP and its receptor is in the control of prolactin release. Both neuroanatomical studies of the ligand and its receptor, as well as the physiological responses to central administration of PrRP, suggest roles in other processes, such as feeding behavior and the central control of blood pressure (1,2). The anorectic effect of PrRP has been demonstrated by the dose-dependent reduction in nocturnal food intake that follows a single intracerebroventricular (ICV) injection of PrRP into free-moving male rats. In 24-h-fasted rats, ICV PrRP can also inhibit food intake when compared with ICV saline (1). A role for PrRP in the central control of blood pressure is suggested by the finding of a highly significant elevation in mean arterial blood pressure observed in conscious, unrestrained male rats following a single ICV injection of PrRP in comparison with their saline-treated controls (2). To date, there are no reported human genetic studies of PrRP or GPR10 in relation to obesity or blood pressure. We therefore undertook mutation detection studies of these two genes in 94 subjects with severe early-onset obesity.

Eight polymorphic sites were found in GPR10 and four in PrRP. Two of the GPR10 polymorphisms (V283I and P305L) resulted in amino acid changes; all of the variants in the PrRP gene were silent (Table 1). V283I and P305L were found at similar allele frequencies (1 and 7%, respectively) in normal-weight Caucasians. Thus, major loss-of-function mutations in either PrRP or its receptor do not appear to be a common cause of severe human obesity.

To examine whether genetic variation in GPR10 might contribute to the inherited component of variation in obesity, blood pressure, and other intermediate metabolic traits, we genotyped 1,084 unrelated U.K. Caucasian subjects from the prospective MRC Ely study (3,4) for the two common polymorphisms (G-62A and C914T [P305L]) that we identified in our initial screen. We recorded 1,084 valid genotypes for C914T and 1,026 for G-62A. Allele frequencies of the mutant alleles at the G-62A and C914T polymorphic loci in the Ely cohort were 0.25 and 0.04, respectively. Neither polymorphism deviated significantly from Hardy-Weinberg predictions, and the two variants were in linkage disequilibrium (P < 0.00001; n = 1,026). This linkage disequilibrium is explained by an increased frequency of the −62A→C914 and G-62→914T over the G-62→C914 and −62A→914T haplotypes; hence, the mutant allele at each locus appeared to be in partial linkage disequilibrium with the wild-type allele at the other. GPR10 haplotypes were constructed; however, on account of the rarity of the 914T allele, haplotypes including this allele were insufficiently common in the population to permit analysis. Therefore, haplotype analysis did not provide any additional information over and above the results obtained from the analysis of the two SNPs separately.

Neither polymorphism showed any association with BMI (Table 2), fasting glucose, insulin, triglycerides, or cholesterol (data not shown). However, significant associations between C914T and diastolic blood pressure (P < 0.01) and between G-62A and C914T and systolic blood pressure (P < 0.05) were found (Table 2). Blood pressure was higher in carriers of the −62A allele than in noncarriers. Similarly, those who were homozygous for the C914 allele had higher blood pressure readings than carriers of the 914T allele. These findings are consistent with the observed pattern of linkage disequilibrium between these two SNPs.

The ability of the variant receptor P305L to generate intracellular calcium in response to a range of PrRP concentrations (5) was examined using fluorometric imaging plate reader (FLIPR) technology applied to stably transfected HEK293 cells. The naturally occurring mutant receptor was compared with both its wild-type counterpart and a negative control (an artificially mutated form of the receptor lacking its COOH-terminal portion [H306STOP]) (Fig. 1). The P305L mutation resulted in a significant (P < 0.001) shift to the right in ligand-induced calcium flux (half-maximal effective dose [95% CI]: 2.2 × 10⁻⁹ mol/l PrRP [1.7 × 10⁻⁹ to 2.8 × 10⁻⁹] for the wild type [P305] vs. 1 × 10⁻⁸ mol/l PrRP [9.2 × 10⁻⁹ to 1.1 × 10⁻⁸] for the mutant receptor [305L]). Responses at maximally effective ligand concentrations were not significantly different. Comparable results were obtained when the shorter form of the ligand, PrRP-20 (Phoenix Pharmaceuticals, Belmont, CA) was used (data not shown). Preliminary experiments suggest that there are no major differences in ligand binding between the naturally occurring mutant and wild-type forms of the receptors (C.S., P.C., J.B., personal observations).

The failure to detect pathogenic mutations within GPR10 and PrRP in subjects with severe obesity or associations between polymorphisms in GPR10 and BMI does not exclude a role for these molecules in appetite regulation. It is unlikely, however, that genetic variation in these molecules is responsible for a substantial proportion of human obesity.

The associations found between GPR10 polymorphisms and blood pressure are intriguing and deserve further study in independent populations. How might PrRP/GPR10 influence blood pressure? PrRP elevates blood pressure when administered ICV (2). PrRP is highly expressed in areas of the brain known to be important in the control of cardiovascular function, most notably the A1 and A2 areas of the caudal medulla, where PrRP immunoreactive neurones strongly costain with tyrosine hydroxylase (6). GPR10 is encoded on human chromosome 10q25.3–q26.1 (7). While this is not a region that has been implicated in major genome-wide scans in human hypertension, it is notable that there is a major quantitative trait locus for blood pressure in rats on chromosome 1, in the region syntenic to human 10q25–26.

We conclude that genetic variation in the G-protein-coupled receptor GPR10 may have an influence on human blood pressure. As is always the case with observations of this nature, replication studies in other large populations is a priority to determine the consistency of the genetic association.