Technical Comments

Comment on Papers by Evans et al. and Mekel-Bobrov et al. on Evidence for Positive Selection of MCPH1 and ASPM

See allHide authors and affiliations

Science  24 Aug 2007:
Vol. 317, Issue 5841, pp. 1036
DOI: 10.1126/science.1141705


Evans et al. and Mekel-Bobrov et al. (Reports, 9 September 2005, p. 1717 and 1720, respectively) reported that human genetic variants of Microcephalin (MCPH1) and abnormal spindle-like microcephaly associated (ASPM) are under strong positive selection. We genotyped these variants in 9000 children and find no meaningful associations with brain size and various cognitive measures, which indicates that contrary to previous speculations, ASPM and MCPH1 have not been selected for brain-related effects.

Homozygous null mutations in the genes Microcephalin (MCPH1) and ASPM (abnormal spindle-like microcephaly associated) cause primary microcephaly in humans, a condition in which brain size is reduced to a third, with no other symptoms except mental retardation (1). Sequence variation in humans strongly suggests that for both genes, a new haplotype emerged ∼6000 (ASPM) and ∼40,000 (MCPH1) years ago and rose to frequencies in Europe and Asia of ∼40% (ASPM) and ∼80% (MCPH1) as a result of positive selection (2, 3). Given the phenotype of the recessive null mutations, it has been speculated that some brain-related feature, such as brain size or cognitive function, was the selected phenotype. However, the phenotypic consequences of the selected alleles remain unknown (4, 5).

We genotyped nearly 9000 children from the ALSPAC study (Avon Longitudinal Study of Parents and Children) for two single-nucleotide polymorphisms that are characteristic for the selected haplogroups D of MCPH1 and of ASPM (6). The selected alleles have a frequency of 42.7% (ASPM) and 81.4% (MCPH1), well in accordance with published frequencies (2, 3). Many phenotypic measurements are available for the majority of the ALSPAC children (7). We initially focused on head circumference at birth, which correlates well with brain size (1), especially during early development (8), as well as general cognition (total, verbal, and performance IQ), working memory, attention span, and motor performance (6). We found no meaningful differences across the genotypes for these measurements (Table 1).

Table 1.

Brain-related measurements and their association with ancestral and derived alleles of ASPM and MCPH1 in children of the ALSPAC cohort.

View this table:

We then extended our analysis to 30 other variables like weight, height, allergy, lung function, and occurrence of various psychological and physiological disorders (6). In none of these measurements (table S1), did we observe a robust effect of the genotype. An absence of an effect on brain size corresponds to the results of a study measuring brain volume using magnetic resonance imaging of 120 individuals (5) and with paleontological data indicating that human brain size has not changed in anatomically modern humans (9). The absence of an effect on IQ measures corresponds with the findings in a recent study with half as many individuals as analyzed here (4). Because the selected phenotype must be large enough to confer around 1 to 5% more off spring per generation on average (2, 3), and in view of our relatively large sample size, we conclude that it is unlikely that the haplogroups D of ASPM and MCPH1 conferred some selective advantage due to the wide range of brain-size– and brain-function–related phenotypes that we have investigated. Assuming that the two haplogroups D indeed rose to high frequencies because of positive selection [see (10), however], what could have been the selected phenotype? Because MCPH1 and ASPM are widely expressed in fetal and adult tissues and are involved in functions such as spindle organization and DNA damage signaling not exclusive to the brain (1), the causes for selection may involve functions frequently found to be under positive selection in mammals, such as immunity and spermatogenesis (11). Although our present study only addresses the phenotypic consequences of the two human haplogroups D, such a scenario would also be compatible with the findings that amino acid changes in ASPM and MCPH1 were repeatedly positively selected in nonhuman primate lineages for which there is no evidence of a change in brain size or cognitive function (12).

Supporting Online Material

Materials and Methods

Table S1

References and Notes

View Abstract

Navigate This Article