False positive non-synonymous polymorphisms of G-protein coupled receptor genes

Comprehensive Analysis of Non-Synonymous Natural Variants of G Protein-Coupled Receptors

G protein-coupled receptors (GPCRs) are the largest superfamily of transmembrane receptors and have vital signaling functions in various organs. Because of their critical roles in physiology and pathology, GPCRs are the most commonly used therapeutic target. It has been suggested that GPCRs undergo massive genetic variations such as genetic polymorphisms and DNA insertions or deletions. Among these genetic variations, non-synonymous natural variations change the amino acid sequence and could thus alter GPCR functions such as expression, localization, signaling, and ligand binding, which may be involved in disease development and altered responses to GPCR-targeting drugs. Despite the clinical importance of GPCRs, studies on the genotype-phenotype relationship of GPCR natural variants have been limited to a few GPCRs such as β-adrenergic receptors and opioid receptors. Comprehensive understanding of non-synonymous natural variations within GPCRs would help to predict the unknown genotype-phenotype relationship and yet-to-be-discovered natural variants. Here, we analyzed the non-synonymous natural variants of all non-olfactory GPCRs available from a public database, UniProt. The results suggest that non-synonymous natural variations occur extensively within the GPCR superfamily especially in the N-terminus and transmembrane domains. Within the transmembrane domains, natural variations observed more frequently in the conserved residues, which leads to disruption of the receptor function. Our analysis also suggests that only few non-synonymous natural variations have been studied in efforts to link the variations with functional consequences.

Arginine (CGC) codon targeting in the human prostacyclin receptor gene (PTGIR) and G-protein coupled receptors (GPCR)

The human prostacyclin receptor (hIP) has recently been recognized as an important seven transmembrane G-protein coupled receptor that plays critical roles in atheroprevention and cardioprotection. To date, four non-synonymous genetic variants have been identified, two of which occur at the same Arg amino acid position (R212H, R212C). This observation instigated further genetic screening for prostacyclin receptor variants on 1455 human genomic samples. A total of 31 distinct genetic variants were detected, with 6 (19%) involving Arg residues. Distinct differences in location and frequencies of genetic variants were noted between Caucasian, Asian, Hispanic and African Americans, with the most changes noted in the Asian cohort. From the sequencing results, three Arg-targeted changes at the same 212 position within the third cytoplasmic loop of the human prostacyclin (hIP) receptor were detected: 1) R212C (CGC-->TGC), 2) R212H (CGC-->CAC), and 3) R212R (CGC-->CGT). Three additional Arg codon variants (all exhibiting the same CGC to TGC change) were also detected, R77C, R215C, and R279C. Analysis (GPCR and SNP databases) of 200 other GPCRs, with recorded non-synonymous mutations, confirmed a high frequency of Arg-targeted missense mutations, particularly within the important cytoplasmic domain. Preferential nucleotide changes (at Arg codons), were observed involving cytosine (C) to thymine (T) (pyrimidine to pyrimidine), as well as guanine (G) to adenine (A) (purine to purine) (p<0.001, Pearson's goodness-of-fit test). Such targeting of Arg residues, leading to significant changes in coding amino acid size and/or charge, may have potentially-important structural and evolutionary implications on the hIP and GPCRs in general. In the case of the human prostacyclin receptor, such alterations may reduce the cardio-, vasculo-, and cytoprotective effects of prostacyclin.

Sequence variation in G-protein-coupled receptors: analysis of single nucleotide polymorphisms

We assessed the disease-causing potential of single nucleotide polymorphisms (SNPs) based on a simple set of sequence-based features. We focused on SNPs from the dbSNP database in G-protein-coupled receptors (GPCRs), a large class of important transmembrane (TM) proteins. Apart from the location of the SNP in the protein, we evaluated the predictive power of three major classes of features to differentiate between disease-causing mutations and neutral changes: (i) properties derived from amino-acid scales, such as volume and hydrophobicity; (ii) position-specific phylogenetic features reflecting evolutionary conservation, such as normalized site entropy, residue frequency and SIFT score; and (iii) substitution-matrix scores, such as those derived from the BLOSUM62, GRANTHAM and PHAT matrices. We validated our approach using a control dataset consisting of known disease-causing mutations and neutral variations. Logistic regression analyses indicated that position-specific phylogenetic features that describe the conservation of an amino acid at a specific site are the best discriminators of disease mutations versus neutral variations, and integration of all our features improves discrimination power. Overall, we identify 115 SNPs in GPCRs from dbSNP that are likely to be associated with disease and thus are good candidates for genotyping in association studies.

Identification of functional SNPs in the 5-prime flanking sequences of human genes

Background

Over 4 million single nucleotide polymorphisms (SNPs) are currently reported to exist within the human genome. Only a small fraction of these SNPs alter gene function or expression, and therefore might be associated with a cell phenotype. These functional SNPs are consequently important in understanding human health. Information related to functional SNPs in candidate disease genes is critical for cost effective genetic association studies, which attempt to understand the genetics of complex diseases like diabetes, Alzheimer's, etc. Robust methods for the identification of functional SNPs are therefore crucial. We report one such experimental approach.

Results

Sequence conserved between mouse and human genomes, within 5 kilobases of the 5-prime end of 176 GPCR genes, were screened for SNPs. Sequences flanking these SNPs were scored for transcription factor binding sites. Allelic pairs resulting in a significant score difference were predicted to influence the binding of transcription factors (TFs). Ten such SNPs were selected for mobility shift assays (EMSA), resulting in 7 of them exhibiting a reproducible shift. The full-length promoter regions with 4 of the 7 SNPs were cloned in a Luciferase based plasmid reporter system. Two out of the 4 SNPs exhibited differential promoter activity in several human cell lines.

Conclusions

We propose a method for effective selection of functional, regulatory SNPs that are located in evolutionary conserved 5-prime flanking regions (5'-FR) regions of human genes and influence the activity of the transcriptional regulatory region. Some SNPs behave differently in different cell types.

Lower rate of genomic variation identified in the trans-membrane domain of monoamine sub-class of Human G-Protein Coupled Receptors: the Human GPCR-DB Database

Background

We have surveyed, compiled and annotated nucleotide variations in 338 human 7-transmembrane receptors (G-protein coupled receptors). In a sample of 32 chromosomes from a Nordic population, we attempted to determine the allele frequencies of 80 non-synonymous SNPs, and found 20 novel polymorphic markers. GPCR receptors of physiological and clinical importance were prioritized for statistical analysis. Natural variation and rare mutation information were merged and presented online in the Human GPCR-DB database .

Results

The average number of SNPs per 1000 bases of exonic sequence was found to be twice the average number of SNPs per Kilobase of intronic regions (2.2 versus 1.0). Of the 338 genes, 111 were single exon genes, that is, were intronless. The average number of exonic-SNPs per single-exon gene was 3.5 (n = 395) while that for multi-exon genes was 0.8 (n = 1176). The average number of variations within the different protein domain (N-terminus, internal- and external-loops, trans-membrane region, C-terminus) indicates a lower rate of variation in the trans-membrane region of Monoamine GPCRs, as compared to Chemokine- and Peptide-receptor sub-classes of GPCRs.

Conclusions

Single-exon GPCRs on average have approximately three times the number of SNPs as compared to GPCRs with introns. Among various functional classes of GPCRs, Monoamine GPRCs have lower number of natural variations within the trans-membrane domain indicating evolutionary selection against non-synonymous changes within the membrane-localizing domain of this sub-class of GPCRs.

Mutant G-protein-coupled receptors as a cause of human diseases

G-protein-coupled receptors (GPCR) are involved in directly and indirectly controlling an extraordinary variety of physiological functions. Their key roles in cellular communication have made them the target for more than 60% of all currently prescribed drugs. Mutations in GPCR can cause acquired and inherited diseases such as retinitis pigmentosa (RP), hypo- and hyperthyroidism, nephrogenic diabetes insipidus, several fertility disorders, and even carcinomas. To date, over 600 inactivating and almost 100 activating mutations in GPCR have been identified which are responsible for more than 30 different human diseases. The number of human disorders is expected to increase given the fact that over 160 GPCR have been targeted in mice. Herein, we summarize the current knowledge relevant to understanding the molecular basis of GPCR function, with primary emphasis on the mechanisms underlying GPCR malfunction responsible for different human diseases.

Polymorphisms of adrenergic receptors: variations on a theme

Recent studies have revealed that most of the adrenergic receptor genes are polymorphic, leading to changes in the amino sequence of the encoded receptor. The variations occur in multiple functional regions of the receptors, and appear as haplotypes with other coding and noncoding polymorphisms in their genes. The consequences of such genetic variability have been explored in recombinant cell-based systems and in human studies. Adrenergic receptor polymorphisms have been shown to alter receptor binding, G-protein coupling, regulation, and expression compared with their allelic counterparts. Here, the genetic and molecular characterization of these polymorphisms is reviewed, as well as their potential impact on pharmacogenetics, disease risk, and disease modification.

Gene and protein domain-specific patterns of genetic variability within the G-protein coupled receptor superfamily

INTRODUCTION

Guanine nucleotide binding proteins (G-proteins) represent the targets for >50% of all therapeutics. There is substantial interindividual variation in response to agonists and antagonists directed to these receptors, which may, in part, be due to genetic polymorphisms. As a class, the sequence variability of G-protein-coupled receptor (GPCR) genes has not been characterized.

STUDY DESIGN

This variability was investigated by sequencing promoter, 5'- and 3'-UTR, coding blocks, and intron-exon boundaries, of 64 GPCR genes in an ethnically diverse group of 82 individuals.

RESULTS

Of the 675 single-nucleotide variations found, 61% occurred in > or =1% of the population sample and the nature of these 412 single nucleotide polymorphisms (SNPs) was assessed. 5'-UTR (p = 0.002) and coding (p = 0.006) SNPs were observed more often in GPCR genes, compared with 309 non-GPCR genes similarly interrogated. The prevalence of non-synonymous coding SNPs was unexpectedly high, with 65% of GPCR genes having at least one. Intron-containing genes had half as many non-synonymous coding SNPs compared with intronless genes (p = 0.0009), suggesting that when introns are not available coding regions provide sites for variation. A distinct relationship between the prevalence of non-synonymous SNPs and receptor structural domains was evident (p = 0.0006 by ANOVA), with variability being most prominent in the transmembrane spanning domains (38%) and the intracellular loops (24%). Phosphoregulatory domains, particularly the carboxy terminus, often the site for agonist-promoted phosphorylation by G-protein coupled receptor kinases, were the least polymorphic (8%).

CONCLUSIONS

There is substantial genetic variability in potentially pharmacologically relevant coding and noncoding regions of GPCRs. Such variability should be considered in the development of new agents, or optimization of existing agents, targeted to these receptors.

Pharmacology and physiology of human adrenergic receptor polymorphisms

Adrenergic receptors are expressed on virtually every cell type in the body and are the receptors for epinephrine and norepinephrine within the sympathetic nervous system. They serve critical roles in maintaining homeostasis in normal physiologic settings as well as pathologic states. These receptors are also targets for therapeutically administered agonists and antagonists. Recent studies have shown that at least seven adrenergic receptor subtypes display variation in amino acid sequence in the human population due to common genetic polymorphisms. Variations in potential regulatory domains in noncoding sequence are also present. Here, we review the consequences of these polymorphisms in terms of signaling, human physiology and disease, and response to therapy.

An Ile to Met polymorphism in the catalytic domain of adenylyl cyclase type 9 confers reduced beta2-adrenergic receptor stimulation

Adenylyl cyclase (AC) mediates signalling following activation of G(alphas)-coupled receptors such as the beta2-adrenergic receptor (beta2AR). Genetic variation in the receptor component of this pathway can alter signal transduction and the response to beta-agonists in asthma, but little is known about downstream effectors. Here, we characterize the population genomics and signalling effects of a polymorphism within the coding region of the AC9 gene that results in an Ile to Met substitution at amino acid 772 within the C1b region of the enzyme. Allele frequencies were 0.300 and 0.375 in Caucasians and Asians but were lower in African-Americans (0.163). The functional effects were studied in stably transfected HEK293 cells recombinantly expressing equivalent levels of wild-type (Ile772) and polymorphic (Met772) AC9. The polymorphic substitution results in a loss of function compared to wild-type AC9. Met772 AC9 has lower basal and beta2AR-mediated adenylyl cyclase activities compared to Ile772 AC9, as well as reduced activity following stimulation of G(alphas) by NaF. Direct stimulation of AC9 activity by Mn2+/- was also depressed in Met772 membranes, indicating decreased catalytic function, consistent with the location of residue 772. AC9 mRNA and protein were expressed in multiple human lung cell-types, including airway smooth muscle and airway epithelium. In the treatment of asthma, there is marked heterogeneity in the response to inhaled beta-agonists which is associated with polymorphisms of the beta2AR. Identification of a common AC9 variant that confers reduced enzyme activity reveals an additional polymorphism that should be considered in pharmacogenetic studies of beta-agonist therapy of asthma.

Hierarchy of polymorphic variation and desensitization permutations relative to beta 1- and beta 2-adrenergic receptor signaling

Agonist-promoted desensitization of G-protein-coupled receptors results in partial uncoupling of receptor from cognate G-protein, a process that provides for rapid adaptation to the signaling environment. This property plays important roles in physiologic and pathologic processes as well as therapeutic efficacy. However, coupling is also influenced by polymorphic variation, but the relative impact of these two mechanisms on signal transduction is not known. To determine this we utilized recombinant cells expressing the human beta(1)-adrenergic receptor (beta(1)AR) or a gain-of-function polymorphic variant (beta(1)AR-Arg(389)), and the beta(2)-adrenergic receptor (beta(2)AR) or a loss-of-function polymorphic receptor (beta(2)AR-Ile(164)). Adenylyl cyclase activities were determined with multiple permutations of the possible states of the receptor: genotype, basal, or agonist stimulated and with or without agonist pre-exposure. For the beta(1)AR, the enhanced function of the Arg(389) receptor underwent less agonist-promoted desensitization compared with its allelic counterpart. Indeed, the effect of polymorphic variation on absolute adenylyl cyclase activities was such that desensitized beta(1)AR-Arg(389) signaling was equivalent to non-desensitized wild-type beta(1)AR; that is, the genetic component had as much impact as desensitization on receptor coupling. In contrast, the enhanced signaling of wild-type beta(2)AR underwent less desensitization compared with beta(2)AR-Ile(164), thus the heterogeneity in absolute signaling was markedly broadened by this polymorphism. Inverse agonist function was not affected by polymorphisms of either subtype. A general model is proposed whereby up to 10 levels of signaling by G-protein-coupled receptors can be present based on the influences of desensitization and genetic variation on coupling.

Novel and nondetected human signaling protein polymorphisms

The frequency of single nucleotide polymorphisms (SNPs) in downstream signaling proteins was determined by combination heteroduplex HPLC and double-stranded sequencing of genomic DNA from 96-144 congestive heart failure (CHF) patients. Analysis of 56 coding exons in 9 signaling genes revealed 17 novel and 8 previously reported synonymous (no change in amino acid) SNPs, as well as one novel nonsynonymous SNP in the Rad small G protein. Because this initial analysis failed to detect numerous SNPs reported in the NCBI and Celera databases, double-strand sequencing of relevant exons from 74-91 CHF patients was used to confirm the absence of 10 previously reported nonsynonymous SNPs. Our results show that synonymous SNPs are frequent in signaling protein genes, whereas nonsynonymous SNPs are rare, suggesting a high degree of evolutionary conservation among these downstream signaling molecules. Comparisons of our results to the NCBI and Celera databases indicates that 56% of their SNP entries are not detected in our cohort. Importantly, while 31% of database SNPs were verified, 69% of SNPs detected in our cohort are not included in these databases. These findings indicate that caution may be warranted in relying exclusively on SNP databases as catalogs for polymorphic signaling protein genes.