Longevity and Plasticity of CFTR Provide an Argument for Noncanonical SNP Organization in Hominid DNA

Abstract

Like many other ancient genes, the cystic fibrosis transmembrane conductance regulator (CFTR) has survived for hundreds of millions of years. In this report, we consider whether such prodigious longevity of an individual gene – as opposed to an entire genome or species – should be considered surprising in the face of eons of relentless DNA replication errors, mutagenesis, and other causes of sequence polymorphism. The conventions that modern human SNP patterns result either from purifying selection or random (neutral) drift were not well supported, since extant models account rather poorly for the known plasticity and function (or the established SNP distributions) found in a multitude of genes such as CFTR. Instead, our analysis can be taken as a polemic indicating that SNPs in CFTR and many other mammalian genes may have been generated—and continue to accrue—in a fundamentally more organized manner than would otherwise have been expected. The resulting viewpoint contradicts earlier claims of ‘directional’ or ‘intelligent design-type’ SNP formation, and has important implications regarding the pace of DNA adaptation, the genesis of conserved non-coding DNA, and the extent to which eukaryotic SNP formation should be viewed as adaptive.

SNP Formation Bias in the Murine Genome Provides Evidence for Parallel Evolution

“…We and others have suggested that random mutation accrual over billions of years could otherwise degrade the integrity of core metabolic genes, and that regulatory mechanisms may therefore exist to influence where (and possibly when) SNPs are most likely to occur in genomic DNA ( Charlesworth B and Charlesworth D 1997 ; Loewe and Lamatsch 2008 ; Hill et al 2014 ). One such mechanism is a transition bias that favors both synonymous and conservative exonic SNPs ( Collins and Jukes 1994 ; Wakeley 1994 ; Freeland and Hurst 1998 ; Hill et al 2014 ). To further investigate relevance of the present findings to exonic patterns of evolutionary SNP accumulation, we located the four greatest and four least permissive quartets for A⇔G polymorphism within cDNAs of CFTR and dystrophin, two genes of ancient vertebrate origin ( fig.…”

“…We and others have characterized longevity of core metabolic genes in the face of a mutational “ratchet” (that over hundreds of millions of years would be capable of decimating eukaryotic exons), and suggested existence of adaptive mechanisms that regulate DNA mutation and serve to promote long-term genomic survival ( Gabriel et al 1993 ; Lynch 2010 ; Koonin 2012 ; Hill et al 2014 ). Findings from the present study furnish new evidence in support of this hypothesis.…”

“…We also note that highly specialized sequence motifs favoring SNPs within promoters and other crucial regions of noncoding DNA (e.g., CpG islands) could serve to preferentially facilitate polymorphism and diversity in a manner that directs single base mutations to the regulatory compartment ( ENCODE Project Consortium 2012 ; Gerstein et al 2012 ), while shielding against uncontrolled mutation accrual within essential protein-coding elements ( table 1 and supplementary table S1 and fig. S3 , Supplementary Material online) ( Hill et al 2014 ). In either case, future studies of mouse genomic evolution should consider the role of contextual preference and parallel SNP formation shown here for highly inbred murine strains.…”

“…SNP data were queried and downloaded from 1000 Genomes ( http://www.1000genomes.org , last accessed August, 2015) for 19 human genes of interest taken from a list extensively characterized by our laboratory for features, such as transition bias, genetic founder alleles, well-defined minor allelic frequency, intronic versus exonic SNP prevalence, haplotype block formation, synonymous versus nonsynonymous polymorphism, and conservation among multiple species (including horse, frog, zebrafish, opossum, shark, and chicken) ( Fortini et al 2000 ; Hill et al 2014 ). Intronic DNA was studied to minimize evolutionary selection bias, and data were collated according to SNP type (A⇔G, C⇔T, A⇔C, etc.).…”

“…Mutations and specific constraints that govern sites of DNA polymorphism are not well understood. Although eukaryotic single nucleotide polymorphism (SNP) formation is viewed as stochastic, preferences in the nature and location of certain single base replacements (including C⇔T and A⇔G [transition] mutations) are described among metazoans ( Lederberg JA and Lederberg EM 1952 ; Collins and Jukes 1994 ; Wakeley 1994 ; Drake et al 1998 ; Freeland and Hurst 1998 ; Sung et al 2012 ; Hill et al 2014 ). Isolated “hot spots” for single nucleotide variants exist, but there is little or no information regarding the overall contribution of such regions to eukaryotic SNP formation.…”