The relationship of maternal rank, 5-HTTLPR genotype, and MAOA-LPR genotype to temperament in infant rhesus monkeys (Macaca mulatta)

Abstract

Temperament is a construct whose manifestations are quantifiable from an early age, and whose origins have been proposed as "biological." Our goal was to determine whether maternal rank and infant genotype are associated with five measures of temperament in 3- to 4-month old rhesus monkeys (Macaca mulatta), all of whom were born and reared by their mothers in large, outdoor, half-acre cages. Maternal rank was defined as the proportion of animals outranked by each female, and the two genes of interest to us were monoamine oxidase and serotonin transporter, both of which are polymorphic in their promoter regions (MAOA-LPR and 5-HTTLPR, respectively), with one allele of each gene considered a "plasticity" allele, conferring increased sensitivity to environmental events. Our large sample size (n = 2014-3140) enabled us to examine the effects of individual genotypes rather than combining genotypes as is often done. Rank was positively associated with Confident temperament, but only for animals with the 5-repeat allele for MAOA-LPR. Rank had no other effect on temperament. In contrast, genotype had many different effects, with 5-HTTLPR associated with behavioral inhibition, and MAOA-LPR associated with ratings-based measures of temperament. We also examined the joint effect of the two genotypes and found some evidence for a dose-response: animals with the plasticity alleles for both genes were more likely to be behaviorally inhibited. Our results suggest phenotypic differences between animals possessing alleles for MAOA-LPR that show functional equivalence based on in vitro tests, and our data for 5-HTTLPR revealed differences between short/short homozygotes and long/short heterozygotes, strongly suggesting that combining genotypes for statistical analysis should be avoided if possible. Our analysis also provides evidence of sex differences in temperament, and, to our knowledge, the only evidence of differences in temperament based on specific pathogen-free status. We suggest several directions for future research.

Keywords: behavior genetics; behavioral inhibition; monoamine oxidase-A promoter; plasticity alleles; serotonin transporter promoter.

Figure 1

Analysis 2. Probability of being behaviorally inhibited as a function of the 5‐HTTLPR genotype (S/S vs. L/S vs. L/L). Error bars represent 95% confidence intervals; the estimates come from the best model (M2)

Figure 2

Analysis 3. Temperament ratings as a function of the MAOA‐LPR genotype (5/5 or 5/‐ vs. 6/6 or 6/‐ vs. 7/7 or 7/‐). Error bars represent 95% confidence intervals; the estimates come from the best models (for Confident: M3; for the other ratings: M2)

Figure 3

Analysis 3. Confidence score as a function of rank and the MAOA‐LPR genotype (5‐repeat vs. 6‐repeat vs. 7‐repeat). Error bars represent 95% confident intervals; the estimates come from the best model (M3)