Chrm3 gene and M3 muscarinic receptors contribute to salt-sensitive hypertension - but now a physiological puzzle.

In this issue, an article by Deng et al. entitled “Novel pathogenesis of hypertension and diastolic dysfunction caused by cholinergic 3 receptor signaling” reports evidence that Chrm3 and the M3 receptor (M3R) that it encodes contributes to blood pressure salt-sensitivity in both male and female Dahl salt-sensitive (DSS) rats. The study nicely demonstrates how the thoughtful application of established genetic approaches together with deep sequencing, gene editing, and information technologies can be applied to facilitate the identification and validation of candidate genes. Discovery of the role of Chrm3 in the salt-sensitive phenotype resulted from arduous chromosomal substitution mapping studies which narrowed several regions to reveal three blood pressure QTLs within Chromosomes 2 and 17 ¹. Within these narrow regions, Chrm3 was selected among three candidate genes (Igsf3, Eif3i-ps1, and Chrm3) for further analysis based on several criteria. It was the only candidate with a missense mutation and it was known to encode the muscarinic cholinergic receptor 3 (M3R), which could potentially affect cardiovascular function. Among the three candidates, Chrm3 was the only one not embryonically lethal when knocked out, thereby facilitating validation of the functional relevance and mechanisms of actions of this gene.

The study is notable since only a handful of naturally occurring gene variants have yet been identified proven to determine salt-sensitivity². The study is a reminder that although gene sequencing and editing technologies are indeed powerful tools, classic genetic breeding strategies coupled with speed congenics were required to identify the novel QTL containing Chrm3. The anonymous search for naturally occurring allelic variants linked to salt-sensitivity carried out in this study avoided focus only upon preconceived candidate genes and existing dogmas of salt-sensitivity. It thereby yielded a gene that would not have been predicted to be associated with hypertension. This was followed by the knockout of Chrm3 in the DSS rat thereby providing the important validation and proof of concept that this gene was somehow playing an important role in blood pressure salt-sensitivity. Gene editing techniques represent powerful tools for the validation of a candidate gene associated with hypertension. However, while many genes and proteins can affect changes in blood pressure an observed effect in response to gene modification does not mean the gene or a variant of that gene is important in hypertension. Enhanced M3R activity was correlated with blood pressure in DSS rats and knockout of Chrm3 lowered salt-induced hypertension in both male and female DSS rats.

How do M3 muscarinic receptors contribute to blood pressure salt-sensitivity?

The surprising discovery that M3 muscarinic receptors contribute in some way to salt-sensitivity begs important mechanistic questions. Apart from having validated the importance of Chrm3 in a commonly studied model of hypertension, the greater challenge is how to relate a gene variant to complex biological pathways at the cell and organ level of function. This is indeed the same challenge faced with the many GWAS identified SNP variants found in human populations. Although serious efforts to do this were made in the present study, the present study did not effectively identify the cellular or physiological pathways whereby Chrm3 modifies the blood pressure. Enhanced M3R activity was correlated with blood pressure in DSS rats. In vitro studies in HEK293 cells found that reduced expression of Chrm3 resulted in slower and less pronounced internalization of the muscarinic receptor M3R. Delayed endocytosis of the receptor would be expected to enhance the number of cell surface receptors and prolong functional signaling in vivo. However, studies were not pursued to determine whether and how delayed M3R internalization was functionally related to the in vivo impairment of vasodilation observed with knockout of Chrm3 in DSS rats.

Studies were carried out to determine the effects of Chrm3 knockout upon cardiac and kidney function. Both were improved in the knockout rats compared to DSS rats. However, these observations add little to the mechanistic understanding of how Chrm3 determines salt-sensitivity since reduction of arterial pressure in Chrm3 knockout rats. The reduction of arterial pressure by any means with a reduction of cardiac afterload would alone be expected to reduce hypertrophy and improve cardiac function. The same could be said of the apparent improvement of kidney function suggested by creatine clearance measurement. Neither of these observations, however, advance our mechanistic understanding of how Chrm3 and M3R altered cardiac or renal function.

Most puzzling was the paradoxical findings of impaired vasodilation of isolated cerebral and mesentery vessels in Chrm3 knockout DSS rats, despite the reduction of hypertension. Although such dissociations between vascular reactivity, total peripheral resistance and chronic levels of arterial pressures are possible in the face of large changes of arterial and venous compliances, this would be unusual and accompanied by large offsetting reductions of extracellular and blood volumes which were not determined. The contradiction between the blood pressure phenotype and vascular responses observed in Chrm3 knockout rats is not readily explained and it remains unclear how Chrm3 and alterations in M3R may be altering blood pressure salt-sensitivity.

Given the complex interacting physiological systems involved in the regulation of arterial pressure one cannot expect any single study to explain how alterations of Chrm3 could be affecting blood pressure salt-sensitivity. As carried out in part in the present study, high level screening protocols can provide useful clues for more in depth follow up studies. In the present study, it would have been expected that assessments of central and peripheral neural function would have been carried out given the known effects of muscarinic acetylcholine receptors to modify cardiac function by parasympathetic stimulation³. It is recognized that muscarinic receptors, a subfamily of G protein-coupled receptors, regulate numerous fundamental functions of the central and peripheral nervous system⁴.

From a broader perspective, the goal is to determine the functional relevance of any novel gene found to be involved in hypertension in humans or rodents. Although the importance of genomic context is unquestionable ⁵, correlated orthologous rat-human phenotype-genotype associations are common ^6–8 and although negative associations are not helpful, positive associations are likely to improve our understanding of human risk. At this point in time, the availability and application of gene editing techniques using Crisper-Cas9 in rodents provides remarkable opportunities not previously available to study the functional pathways and consequences of hypertension associated candidate genes found in rodents and human GWAS studies. However, it should be emphasized that the knockdown of a gene is only relevant to our understanding of hypertension provided that a close association of an allelic variant with the disease has first been established. Using anonymous approaches, this must be done using classic genetic approaches in rodents or by GWAS in humans.

Relevance of the non-linear structure of chromosomes upon and why hypertension cannot be explained by adding up the small effects from multiple QTLs.

Although less central to this manuscript, the author advances the theses that hypertension cannot be explained by adding up the small effects from multiple QTLs. The effects of alleles at different genetic loci of a quantitative trait can be additive, or the combined effects can be greater or lesser than the sum of individual alleles alone (e.g., epistasis). Such interactions have been clearly demonstrated in studies of congenic rats ^9–11 which have served as the basis of what Chauvet and Deng (see refs 6,8,9 in this issue by Deng et al) have previously presented as the concept of epistatic modules. Given the rapid advancement of research in this field, however, the concept of epistatic modules is increasingly less relevant to our understanding of gene-gene interactions regarding specific phenotypes. GWAS have identified several hundred single nucleotide polymorphisms (SNPs), common genetic risk variants for hypertension with small effect size ^12–14. The great majority of these variants and SNPs in linkage disequilibrium with them are in non-coding regions of the genome including intronic, intergenic or promotor regions. Of the intronic or intergenic SNPs many of these are located in potential enhancer regions. Additionally, there are thousands of long non-coding RNA (lncRNA) genes encoding many more thousands of lncRNA transcripts about which there remains little functional information ¹⁵. Elegant sequence capture-based methods (3C, 4C, 5C, HiC, ChIA-PET) are now providing increasingly detailed 3-dimensional maps of physical chromatin-fiber contact frequencies. These approaches are beginning to identify mechanisms and the relevance of changes within both the coding and non-coding regions. Interactions such as these were illustrated in a recent study by Stodola et al ¹⁶ who identified genomic regions in several chromosomes that physically interact with the renin proximal promoter. This non-linear structure of chromosomes enables transcriptional regulation by enhancers and promoters communicating with each other over large genomic distances ¹⁷. Heritable structural genomic variants and responses to perturbations caused by disrupting chromatin loops or enhancer-gene interactions can result in either graded gene expression or in all-or-nothing responses ¹⁸. It is becoming increasingly clear why hypertension cannot be explained by adding up the small effects from multiple QTLs based upon our increasing knowledge of the underlying molecular basis of epistatic modules.

Convergence of classic and modern genomic approaches.

Remarkably, the technologies needed to converge classic genetic and modern genomic approaches are now emerging that will enable one to map genes from humans to rodents and vice versa, as reflected in the Figure. SNPs associated with human disease can now be explored in ever efficient ways in animal models and in vitro cell systems obtained from rodents and humans. The functional relevance of these allelic variations can now be efficiently explored by gene editing using CRISPER-Cas9, and studies carried out to deeply explore cell signaling pathways and related higher levels of function.

Figure.

Illustration of the convergence of classic and modern genomic approaches to identify allelic variants associated with hypertension in human and animal model systems. Bidirectional approaches can now be applied to reveal the functional relevance allelic variants (SNPs) discovered human populations by GWAS or in rat or mouse models by linkage, congenics, hybrid diversity panels. Homology mapping using dense genomic sequencing data enables variants discovered in humans to be mapped to mouse and rat where function can be studied in great depth using in vitro and in vivo approaches and validated by gene editing techniques. Conversely, variants discovered in animal models can direct searches for homologous variants and studies in human subpopulations containing homologous variants.

It is more important than ever that bi-directional approaches be implemented if we are to unravel the functions of genes and genomic variants. Hundreds of organisms have been sequenced since the completion of the Human Genome Project in 2001 yielding yottabytes (10^24) of data. Advances in efforts to close the gap between genes and function are reflected by the International Mouse Knockout Consortium (IKMC) and the International Mouse Phenotyping Consortium (IMPC). Substantial progress has been made in the annotation of the 19,973 human protein coding genes based on phylogenetic profiling of 177 species (NCBI) with fewer than 6,000 remaining ¹⁹. These growing annotations represent an entry point that may suggest a pathway or organ system to be then explored in greater depth in cellular and animal model systems. The NIH TOPMed project (www.nhlbiwgs.org) aims to integrate genomic sequence and other - omic data obtained from 60 different studies with more than 120,000 subjects in efforts to reveal factors that increase or decrease the risk of disease and to identify subtypes of disease with the goal of developing more targeted and personalized treatments.

These are indeed ambitious goals with many hurdles to leap. The present study by Deng et al. revealed a novel gene involved in blood pressure salt-sensitivity, which alone required great efforts. We should anticipate that the growing application of the many newly emerging technologies and enhanced understanding of the epigenetic regulation of gene function will meaningfully advance our understanding of the relationships between sequence variants and functional pathways that lead to hypertension.