Initially posted 10 December 2009 by evomed
Are humans still evolving? In Chapter 6, we reviewed the current debate on the pace and direction of human evolution. It might seem that modern medical technologies and welfare provisions have removed the impact of natural selection on the human population: those who in the past would have died young, today survive into the reproductive period. Yet one problem of this argument is that natural selection works through differential reproductive success rather than differential survival alone. Reproductive technologies and social changes, however, are likely to have an effect. The promise of ‘engineered’ genes in embryos conceived using assisted reproduction technologies is an exciting medical development, yet unlikely to produce evolutionary change, as these directed mutations would probably be swamped by countless mutations emerging naturally in the babies born every year. In contrast, large-scale social developments—such as postponement of pregnancies towards the late reproductive period, artificial restriction of fertility through use of contraception, and sexual selection—are likely to produce an effect. Furthermore ‘traditional’ selective pressures, such as dietary and ecological changes as well as infectious diseases, continue to operate. For instance, a recent study of kuru showed that even a disease that appeared not more than a century ago and persisted in a limited geographic area can produce a sufficient selective pressure to effect an observable genetic change.
Contrary to the suggestions that selection no longer operates, scientists working with large sets of genetic data (3.9-million HapMap SNP dataset) have proposed that the rate of evolution in humans is actually accelerating. This acceleration was proposed to be caused by the larger size and lesser reproductive isolation of the current human population, allowing for a higher probability that a potentially advantageous mutation could occur and persist. But estimating the direction of natural selection in modern societies has been difficult, as scientists lacked sufficiently detailed human data. A recent paper by Byars et al in PNAS capitalized on the rich resource provided by the Framingham Heart Study, an epidemiological study begun in 1948 with the aim of identifying factors contributing to cardiovascular disease, and named after its location, Framingham in Massachusetts, U.S. The aim of the PNAS study was to show that natural selection was operating on contemporary humans and to predict evolutionary changes for some traits with medical significance. While the Framingham Heart Study now follows the third generation, Byars and colleagues took into account only the original and the offspring cohort, as they had completed their reproduction. Furthermore, in contrast to the original study, this one focused on women only. Cultural/behavioural variations (education, smoking, medication) were taken into account as covariates, and secular demographic change was dealt with by dividing women into 6 groups depending on their date of birth, and then measuring their relative reproductive success in comparison to the mean of their cohort. Finally, the variation of a measurement (such as total cholesterol) across longer time span was taken into account by constructing a response surface of each trait for age and time, measuring each individual deviation from the average value, and then calculating a single representative value (the average of the residuals) for each individual.
The traits found to be favoured by natural selection include lower stature, low total cholesterol (yet higher body weight), lower blood pressure, later menopause and earlier beginning of childbearing. The estimated rates of projected evolution are, however, relatively low, and indicate a pace of evolution that is, compared to other animals and accounting for generation time, slow to moderate.
The conclusions of this study go against the previously mentioned predictions of fast evolution, yet it must be remembered that albeit the study was done on a large sample, it was still a limited population, from a relatively restricted geographic area and it comprised no more than two generations. An even larger and longer study would probably produce a more reliable prediction. Furthermore, as the Framingham Heart Study focused on cardiovascular disease, many parameters of potential interest were not measured. For instance, the observed favoured lengthening of the reproductive period is probably based on associated changes in sex hormone levels, so for instance measuring progesterone and estradiol might have resulted in larger observable effects. But all in all, this study provides a valuable piece of information and it is hoped that other large epidemiological and clinical studies in the future would include evolutionary biologists and consequently bring more data into the field.