On the evolution of myoglobin

Abstract

In previous studies, particularly of primates, a high degree of concordance was obtained between an evolutionary pattern based on comparative anatomy and another based on a reconstruction of the possible pathway of evolution of the myoglobin molecule. Accordingly, in extending our studies, we included species of uncertain evolutionary kinship, such as the tree shrew, and also increased the representation of species within the mammalian orders already studied, in the hope that it would be possible to resolve some contended issues and to establish the sequence of branching and pattern of relationship between those orders. As a first step, the phylogenetic pattern within each of the mammalian orders was constrained on zoological grounds according to generally accepted kinship based on the evidence of comparative anatomy and the fossil record. Eight phylogenetic patterns reflecting different possible kinship between the orders were chosen for detailed investigation and for each of these the most economical (parsimonious) pathway which could be obtained for the evolution of myoglobin was reconstructed. The uncertainties which are inherent in such reconstruction were increased by a high incidence of parallel substitutions. Although the eight phylogenetic patterns were widely different from one another it was found that none of the solutions had a An amino acid difference matrix is of special interest because it carries information which is more comparable with that obtained by immunological methods, where access to sequence information has not usually been available. Several different clustering procedures were applied to the amino acid difference matrix for myoglobin. As would be expected from the different assumptions on which each is based, the procedures resulted in different branching patterns which varied in their degree of zoological acceptability. Clustering of the harbour seal with the Cetacea suggested that there might be functional reasons, perhaps associated with diving, for several substitutions acquired in parallel by the myoglobins of these mammals. Repeated clustering of the horse with the sportive lemur, and a tendency for the opossum to join the primates, is associated with a high proportion of parallel substitutions. This tends to undermine confidence in phylogenetic inferences which might be drawn from the repeated clustering of the tree shrew (and often the hedgehog) among the primates. If we assume that there is an overall resemblance of the three-dimensional structure of all myoglobins, and use the crystallographic model of sperm whale myoglobin as a basis, the availability of over 30 vertebrate myoglobin sequences (and that of the mollusc Aplysia ) has provided an opportunity to consider the functional relevance of certain positions at which the nature of the amino acid residue appears to have remained unchanged, or to have changed only conservatively, during several hundred million years of evolution. Part of this conservation is attributable to the maintenance of the monomeric nature of the molecule, and it seems likely that much of the conservatism can be attributed to functional needs in initiating the folding of the molecule and in the maintenance of its tertiary structure. Concomitantly, consideration is given to the likely functional consequence of some of the substitutions which have been accepted. Atassi and his co-workers have carried out extensive immunological studies on myoglobin using antisera prepared in rabbit and goat. The availability of the amino acid sequences of rabbit and sheep (in lieu of goat) has made it possible to investigate the relation between amino acid sequence difference and the distance as measured by immunological criteria. It has not been possible to confirm Reichlin’s hypothesis that the myoglobin antigenic reactive regions of Atassi are particularly variable compared with the rest of the molecule; on the contrary it appears that differences in the immunological reactive regions occur approximately in proportion to those occurring in the molecule as a whole. Estimates of the rate of molecular evolution are of interest because its supposed constancy has been a major argument in favour of the hypothesis that a high proportion of fixed mutations are neutral, or nearly so, as far as selection is concerned. F urthermore, a near constant rate of molecular evolution holds promise for a molecular clock which might be used for dating evolutionary events. However, there is considerable variation in the rate of molecular evolution as evidenced by differences in the number of nucleotide substitutions (‘hits’) in lineages arising from the same phylogenetic branching point or by the examination of the changes at the amino acid level. In order to investigate absolute rates of evolution we sought to establish the best estimate of the date of divergence between the ancestors of the living species included in this study. These dates are based on direct fossil evidence and must be regarded as minimum dates because we have not indulged in open-ended speculations about fossils which have not yet been found. The combination of these dates with the evidence of our cladograms leads us to reaffirm our earlier finding that there are considerable differences in the amount of change in different lineages. For example, whereas one lineage (to gibbon) appears to have accepted no mutations during the past 20 Ma, another lineage (to ox) seems to have fixed seven mutations during the last 18 Ma. Goodman and others have drawn attention to the apparently low rate of molecular evolution among higher primates; a similar observation applies to the myoglobin of the Old World monkeys so far studied, but neither the myoglobins of New World monkeys nor the prosimian myoglobins share this feature. After their divergence from one another the two bird lineages included in this study appear to have fixed mutations at rates comparable with those found among mammals during the past 79 Ma. However, the number of differences between the bird and mammal ancestral stems appears to be remarkably low bearing in mind the date of divergence of their ancestors, about 293 Ma ago. Recognizing that this could be an artefact resulting from multiple changes at the same site, from undetectable back mutations and from isosemantic mutations accumulating during a long period of evolution, various procedures were adopted to transform the data in order to estimate the number of such events. None of these procedures, however, eliminated the phenomenon that during the first 214 Ma since their separation the ancestral stems leading eventually to birds and mammals seem to have fixed mutations in their myoglobin at a lower rate than the average rate prevailing during the past 79 Ma, the latter being approximately one mutation in 4 Ma. It is to be expected that sampling error will produce some fluctuations in rate, but even over the relatively long period of 79 Ma the fastest rate of fixation of mutations is about three times the slowest rate and so we are inclined to discard the molecular clock as unreliable for dating divergences, at least within this span of time. On integrating the various sections of this study we see that the changes in myoglobin have not been at random throughout the molecule. Given the constraints demanded by the functional morphology of the molecule itself and the constraints of the genetic code, it is to be expected that both will contribute to parallel change in different lineages. In the reconstructed pathways of evolution of myoglobin about 50 % of the changes were found in parallel in other lineages. There is an indication, provided by the larger number of arginine residues present in aquatic forms, that some of these parallel changes may be correlated with mode of life. The adaptive significance of the several parallel changes between the cetaceans and the pinnipeds certainly deserves physiological investigation. Regardless of the causes of parallel evolution at the molecular level this phenomenon has contributed to unexpected similarities between myoglobins and has led to difficulties in phylogenetic reconstruction of a nature already familiar to comparative anatomists.