Original ArticleRe-examining the Manning hypothesis: androgen receptor polymorphism and the 2D:4D digit ratio
Introduction
The 2D:4D digit ratio—the ratio of the lengths of the second to fourth digits of the human hand—is a sexually dimorphic trait (Ecker, 1875). Although it varies considerably across individuals, men on average have a lower 2D:4D ratio than women. Interest in the ratio was triggered in 1998 when British researcher John Manning hypothesized that the sex difference in 2D:4D is determined by actions of testosterone in utero and that individual variation in the size of the ratio may be a proxy for individual differences in the level of fetal androgen exposure (Manning et al., 2003, Manning et al., 1998). Prenatal androgens are believed to masculinize the human central nervous system (Hines, 2011). Consequently, the prospect that 2D:4D represents a retrospective marker of the level of androgen exposure during the period when the brain is sexually differentiated has led to nearly 500 studies in the biosciences investigating links between the size of the ratio and variation in sex-related traits believed to be androgen-dependent ranging from autism, to fertility, to behavioral characteristics such as sexuality, risk-taking, or aggression (for review, see Manning, 2008).
The scientific integrity of these studies is based on the premise that individual variation in the ratio constitutes a valid marker of androgen action. The validity of 2D:4D, however, is contested. In favor of an androgen effect, the ratio is reduced (more male-typical) in patients with 21-hydroxylase deficiency who experience elevated androgens in utero (Brown et al., 2002, Ökten et al., 2002) and is female-typical in 46,XY patients with complete androgen insensitivity (Berenbaum, Bryk, Nowak, Quigley, & Moffat, 2009), in whom the androgen receptor (AR) fails to bind testosterone. These data confirm gross changes in the ratio in response to large deviations in androgen availability but do not confirm that small gradations in 2D:4D, visible among healthy persons of either sex, mirror fetal androgen exposure in a dose-dependent fashion. Variation in the ratio within each sex could potentially be an expression of multiple causative factors, not limited to or even largely reflective of androgen action (e.g., Zheng & Cohn, 2011). The view that individual variation in the ratio reflects androgen exposure in a graded, proportional manner is based largely on a single study that appeared in Evolution and Human Behavior. Manning et al. (2003) discovered that the size of the 2D:4D ratio covaries, across healthy males, with a repeat polymorphism in the gene that encodes the AR [number of polyglutamine (CAG) repeats in exon 1 of the AR gene], which is known to correlate inversely with the ability of the receptor to activate transcription of androgen-dependent target genes (Zitzmann, 2009). Manning reported that 2D:4D correlated positively with AR CAG length, supporting the view that individual differences in the ratio are a meaningful reflection of differences in the level of testosterone action.
Despite its theoretical importance, the Manning et al. (2003) result is based on only a single small sample of 50 men and has yet to be replicated. Recent work by Medland et al. (2010), using genomewide screening, and by Hurd, Vaillancourt, and Dinsdale (2011) has cast some doubt on the CAG association. In a study of aggressive behaviour, both 2D:4D and AR CAG length were measured by Hurd et al. (2011), but AR CAG length failed to show the anticipated correlation with 2D:4D (right hand: r=.006; left hand: r=−.12). Medland et al. (2010) failed to find any significant genetic association between 2D:4D and four different single nucleotide polymorphisms (SNPs) located within the AR gene, although CAG repeat expansion per se was not specifically measured.
Here, we report data in which AR CAG genotype was specifically quantified in a large sample of healthy young men in whom a dedicated measure of the 2D:4D ratio was taken. To evaluate the possibility that genetic variation in AR could lead to covariation in the production of testosterone via physiological feedback mechanisms (Crabbe et al., 2007)—which potentially could obscure any correlation—we also measured bioavailable testosterone in the same sample.
Section snippets
Participants
The current work was part of a larger investigation of associations between testosterone biomarkers, affect, and cognition. Participants were 152 physically healthy males recruited from the University of Western Ontario (M=18.72 years; S.D.=1.63; range=17–27 years). The sample was predominantly Caucasian; 10% of the sample was Asian. Participants were excluded if they reported an injury to their hands that could alter the growth of their second or fourth digits or if they had any known
Results
Consistent with previous reports (Fink et al., 2004, Hampson et al., 2008, Medland et al., 2010), the mean right hand 2D:4D for the males in our sample was M=.95, S.D.=.03, and the mean left hand 2D:4D was M=.95, S.D.=.03. Ethnic differences in 2D:4D have been reported in some subpopulations (Manning, Stewart, Bundred, & Trivers, 2004), but there was no evidence in our data that Asians and non-Asians systematically differed in either right or left hand ratios, F(1,149)=0.28, p =.595 (cf., Hurd
Discussion
The validity of the 2D:4D ratio as a gauge of testosterone action in utero has been the subject of fractious debate (Berenbaum et al., 2009, Breedlove, 2010). The idea that testosterone does underlie small individual differences in the ratio is supported primarily by Manning et al. (2003), who showed that the ratio covaried with AR CAG repeat length in a sample of 50 men, although a positive correlation was seen only for the right hand (r=.29), not the left (r=.005). Despite a much larger
Acknowledgments
We thank Christine Tenk for independent scoring of the digit ratios. This work was funded by the Natural Sciences and Engineering Research Council of Canada (E.H.). J.S. was supported by an Alexander Graham Bell Canada Graduate Scholarship. E.H. holds a chair from the Canadian Institutes of Health Research FRN: SOW97026.
References (30)
- et al.
Masculinized finger length patterns in human males and females with congenital adrenal hyperplasia
Hormones and Behavior
(2002) - et al.
Genetic variation at five trimeric and tetrameric tandem repeat loci in four human population groups
Genomics
(1992) - et al.
Second to fourth digit ratio and hand skill in Austrian children
Biological Psychiatry
(2004) - et al.
The second to fourth digit ratio and variation in the zandrogen receptor gene
Evolution and Human Behavior
(2003) - et al.
Sex and ethnic differences in 2nd to 4th digit ratio of children
Early Human Development
(2004) - et al.
A variant in LIN28B is associated with 2D:4D finger-length ratio, a putative retrospective biomarker of prenatal testosterone exposure
American Journal of Human Genetics
(2010) - et al.
Salivary testosterone levels in left- and right-handed adults
Neuropsychologia
(1996) - et al.
The ratio of second- to fourth-digit lengths and congenital adrenal hyperplasia due to 21-hydroxylase deficiency
Early Human Development
(2002) The Organizational Hypothesis: Reflections on the 50th anniversary of the publication of Phoenix, Goy, Gerall, and Young (1959)
Hormones and Behavior
(2009)- et al.
Testosterone levels and mental rotation performance in Chinese men
Hormones and Behavior
(2007)