Is there a relationship between morphological variation and genetic variation of enzyme and blood group loci in human populations?

The relationship between variability in 27 morphological traits and genetic variation at three enzyme loci (EsD, PGM-1 and AP) and three blood group loci (MN, ABO and Rhesus) was studied in two Caucasian populations. No difference was noted in the mean values of the homozygous and heterozygous classes at the loci studied in both populations. Neither was there any difference in the percentage of 'Modal' individuals among the homozygous and heterozygous classes in the two populations. However, at the EsD, PGM-1, AP and MN loci, a negative correlation was observed between morphological variance and heterozygosity in both populations. These results are in agreement with Lerner's prediction of 'Genetic Homeostasis'. Lerner's understanding of the correlation was that heterozygosity at the individual or population level was advantageous as it provided the organism with a greater buffering ability to withstand environmental change and, therefore, a greater chance of survival and reproduction. In the English population no significant departures from Hardy Weinberg expectations were observed for the three enzyme loci AP, EsD and PGM-1 and the blood group locus MN, hence it can be assumed that a random mating population was sampled. However, in the Gujarati Indian population significant departures from Hardy Weinberg expectations were noted at the EsD and AP enzyme loci, due to an excess of homozygotes and a deficiency of heterozygotes. A number of possible reasons can be given to explain this, although it is most likely due to inbreeding. On comparing the gene frequencies of the biochemical systems in this study with those recorded by other researchers, a close agreement was noted. This indicates that the population sample studied is representative of the parent population. On comparing the results of the anthropometric data in the English population with that in the Gujarati population, differences are noted in human morphology. The Gujaratis tend to be shorter in stature, have a lower body mass and less subcutaneous fat tissue. This is partially due to hereditable differences between the two populations but the possibility of some environmental effect still exists. Several theories explaining the negative correlation between morphological variation and heterozygosity have been put forward. They basically fall into two categories, those that attribute the correlation to the scored loci themselves (mostly enzyme loci), and those that consider the scored loci as markers for genetic conditions that are not detectable by the assay method involved but are, nevertheless, the causative agents of the morphological variation in individuals. The theory of additive allelic effects was directly addressed in this study. This theory implies that the observation within a population. that the least phenotypic variation exists among the most heterozygous individuals can be explained by genotype pooling. In this study the homozygote genotypes were both pooled and then separated to look for any differences in variance of the traits in the two situations. No reduction in morphological variation was noted when the homozygous genotypes were treated separately instead of pooled, hence no evidence to support the theory of additive allelic effects. At present there is no consensus about the genetic mechanism underlying the morphological variance/heterozygosity correlation or any other correlation of heterozygosity with a phenotypic trait. This is not because one cannot choose among competing hypotheses that can explain the observations but rather because it is difficult to arrive at a hypothesis that will be consistent with all (or even the bulk) of the results to date. The observed negative correlation between morphological variation and heterozygosity was more pronounced in the English population than the Gujarati population suggesting that hypotheses that place a stronger emphasis on the structure of the population than on the scored loci e.g. associative overdominance, are more likely to explain those findings.