Rapid birth-death evolution and positive selection in detoxification-type glutathione S-transferases in mammals

Abstract

Glutathione S-Transferases (GSTs) are phase II detoxification enzymes that may have evolved in response to changes of environmental substrates. GST genes formed a multigene family and in mammals, there are six classes known as Alpha, Mu, Omega, Pi, Theta, and Zeta. Recent studies in phase I detoxification system specifically the cytochrome P450s provided a general explanation on why genes from a common origin such as those in a multigene family have both phylogenetically stable and unstable genes. Genes that participate in core functions of organisms such as development and physiology are stable whereas genes that play a role in detoxification are unstable and evolve in a process known as birth-death evolution, which is characterised by frequent gene gains and losses. The generality of the birth-death model at explaining the evolution of detoxification enzymes beyond the phase I enzyme has not been comprehensively explored. This work utilized 383 Gst genes and 300 pseudogenes across 22 mammalian species to study gene gains and losses. GSTs vary greatly in their phylogenetic stability despite their overall sequence similarity. Stable Gst genes from Omega and Zeta classes do not show fluctuation in gene numbers from human to opossum. These genes play a role in biosynthesis related functions. Unstable genes that include Alpha, Mu, Pi and Theta undergo frequent gene gain and loss in a process known as birth-death evolution. Gene members of these four classes are well known for their roles in detoxification. Our positive selection screen identified five positively selected sites in mouse GSTA3. Previous studies showed two of these sites (108H and 208E) were biochemically tested as important residues that conferred catalytic activity against the toxic aflatoxin B1-8,9-epoxide. The functional significance against aflatoxin of the remaining three positively selected sites warrant further investigation.

Characterization of Dog Glutathione Transferase P1-1, an Enzyme Relevant to Veterinary Medicine

“…The substrates include epoxides, alkenals, quinones, and hydroperoxides, suggesting that GSTs have evolved as a guardrail against such toxic products that are formed ubiquitously and abundantly as byproducts of oxygen metabolism [ 4 ]. Both membrane-associated (MAPEG; [ 5 ]) and soluble [ 6 ] GSTs exist, and the main play-actors in mammalian detoxication appear to be members of the soluble GSTs of the Alpha, Mu, and Pi classes [ 7 , 8 ]. When tissues other than liver were investigated, the occurrence of the different GSTs was found to vary among organs and cell types, resulting in divergent capabilities for protection against toxic agents [ 9 , 10 ].…”

“…Dogs are treasured domesticated animals as well as experimental models in pharmacological research. GSTs representing the classes responsible for detoxication of xenobiotics [ 7 ] were identified in both canine liver [ 16 ] and kidney [ 17 ]. In particular, the Pi class enzyme GST P1-1 was demonstrated in both of these organs, as well as in the lens [ 18 ].…”

Emergence and influence of sequence bias in evolutionarily malleable, mammalian tandem arrays

“…Many types of outward-facing genes have been reported to exhibit extreme allelic diversity or rapid divergence across species, and polymorphisms in these genes underlie human phenotypic variation in drug metabolism, sensory perception, and immune response (Charkoftaki et al, 2019; Nei et al, 2008; Niimura and Nei, 2007; Schwartz et al, 2017; Semple and Dorin, 2012; Shelton et al, 2022; Sun et al, 2017; Tan and Low, 2018; Thomas, 2007). Colloquially, outward-looking genes are so diverse in copy number and sequence that a first step in GWAS is often to “throw out the ORs.” To systematically examine the degree of coding sequence variation in human genes grouped by AT/GC content, we used a dataset of rare single nucleotide variants ascertained from whole exome sequencing of >100,000 unrelated people (gnomAD v2.1.1, Figure 4A, B) (Karczewski et al, 2020).…”

Nucleotide substitution rates of diatom plastid encoded protein genes are positively correlated with genome architecture

“…The reason for the differences between diatoms and plants with respect to substitution rates may be attributed to fundamental differences in their genome content. Diatom plastomes are gene dense, with very little space dedicated to non-coding sequences and most are devoid of large repeat sequences 61 . However, the average diatom plastomes size is close to those of seed plants because they encode for more genes 7 , 9 , 10 , 62 .…”

“…CODEML model 0 was also used to estimate dN / dS values at the level of individual genes and functional groups. For genes with dN / dS > 1 in model 0, these genes were tested further with CODEML model 7 (neutral) and model 8 (positive selection) to uncover potential positively selected sites using similar methodology as described previously 61 .…”