Rapid evolution in transporters, and other membrane proteins, across the tree of life

By: Pejvak Moghimi

Twitter: @pezhvuk

I think all of us often indulge in metaphorical thinking when engaged with scientific concepts not directly visible to the naked eye, or tangible, to us. Transporters are no exception. Indeed, many of us, perhaps like to think of transporters in a similar way to the gateways that once regulated transport and commerce through the beautiful city walls of York. However, metaphors in science can quickly fall apart. And gateway metaphors are no exception. Transporters, unlike city gates of York are highly evolvable. That is not in itself ground-breaking, as all life, and the genes involved in regulating it, are subject to variety of evolutionary forces. However, Nick Lane and Victor Sojo have shown in a recent paper that membrane proteins (MPs), including transporters, undergo extraordinary rate of evolutionary changes in comparison to water-soluble proteins.

This study starts by investigating the number of orthologs of MPs and water-soluble proteins (WSPs), defined by the OMA database, among 64 species of Gram-positive and Gram-negative bacteria, Archaea, unicellular and multicellular eukaryotes. They compared the number of genes in orthologous groups (OGs) of MPs and WSPs and found that there are significantly smaller number of OGs in MPs than there are for WSPs for every species. Furthermore, they performed a logistic regression on the entire OMA ortholog data set and found that the probability of a gene coding a MP falls dramatically with increasing number of clades sharing that gene. They hypothesised that either;

  1. The sequence divergence is taking place at such a rapid rate that sequence-searching algorithms (such as the one used by BLAST) do not detect the more “hard-to-find” homologs, or
  2. It is due to disproportionate amount of gene loss in MPs.

Given the fact that OMA database uses a more rigorous version of the Smith-Waterman algorithm than BLAST, using OMA is a testament to the validity of the approach taken by the researchers to identify orthologs.

They followed the study by identifying which one of the evolutionary explanations satisfies their observations. To investigate whether rapid sequence divergence in MPs is involved in their observations, they used Nei’s sequence-diversity measure for 228,148 OMA OGs shared by any three or more species. They performed Welch’s t-test on the results and showed that MPs do indeed diverge more rapidly than WSPs. Interestingly, same analysis conducted on the same sequences, when dissected into exterior, transmembrane and interior-facing regions showed that the exterior-facing regions are under more rapid divergence rate than the transmembrane and interior-facing regions.

To determine that true gene loss, which is not to be confounded by an apparent gene loss caused by loss of homology beyond recognition, occurs at higher rate in MPs vs. WSPs they analysed prokaryotic clades (present in OMA), with 10 or greater number of closely related strains present in the database. They assumed that proteins shared by more than half of the number of clades are ancestral, and that clades, which do not share ancestral genes, represent true gene losses. The implicit assumption with this approach is that closely related strains are unlikely to have OGs that have diverged beyond recognition.

These analyses identifies that both rapid divergence and disproportionate gene loss are at play as parts of the molecular evolutionary dynamics of MPs. Authors report that previous research (citations included in the paper) demonstrates exported WSPs evolve faster than cytosolic proteins. This, together with the observations made in their work, led them to hypothesising a fundamental evolutionary principle:

“Membrane proteins evolve faster due to stronger adaptive selection in changing environments, whereas cytosolic proteins are under more stringent purifying selection in the homeostatic interior of the cell. This effect should be strongest in prokaryotes, weaker in unicellular eukaryotes (with intracellular membranes), and weakest in multicellular eukaryotes (with extracellular homeostasis).”

Two-fold effect of adaptation causes faster evolution of external sections and loss of homology in membrane proteins. Adaptation to new functions and niches causes faster evolution for outside-facing sections (top), potentially contributing to divergence beyond recognition. Other proteins may provide no advantage in the new environment, and could be lost entirely over time (centre). For simplicity, the species on the left is assumed to remain functionally identical to the common ancestor (bottom).

This principle is further supported by another experiment carried out by the authors, where they compared the rate of divergence and the degree of gene loss across the three domains of life and showed that indeed the order of the effect of these evolutionary forces is as predicted by the principle. However, the effect was still significant in multicellular eukaryotes.

This principle has great evolutionary, practical phylogenetics and medical implications. For instance, the positive selection pressure from recognition by the host results in a red-queen dynamics that drives faster evolution in MPs. Over half all known drug targets are MPs, which explains why many of the effective drugs in animal trials are unsuccessful in human trials. And last, but not least (at least to me!), if the MPs are less likely to be conserved across the tree of life, then homology searches and molecular clocks are more likely to be confounded. All of this points to the importance of the MPs in adaptability of organisms to new environment; especially microbial communities (not so gate-like after all!). All inferences, obviously, apply to transporters too, and I think, these inferences are even more exciting, in some ways, when the implications are revisited in the context of the importance of transporters in niche-adaptation. Who knows, I might write more extensively about my own views, and further investigations, regarding this matter, with a more specific emphasis on transporters.

Source: Sojo, Victor et al. “Membrane Proteins Are Dramatically Less Conserved Than Water-Soluble Proteins Across The Tree Of Life”. Molecular Biology and Evolution 33.11 (2016): 2874-2884.