Quantitative Trait Locus Analysis of Mating Behavior and Male Sex Pheromones in Nasonia Wasps

Abstract

A major focus in speciation genetics is to identify the chromosomal regions and genes that reduce hybridization and gene flow. We investigated the genetic architecture of mating behavior in the parasitoid wasp species pair Nasonia giraulti and Nasonia oneida that exhibit strong prezygotic isolation. Behavioral analysis showed that N. oneida females had consistently higher latency times, and broke off the mating sequence more often in the mounting stage when confronted with N. giraulti males compared with males of their own species. N. oneida males produce a lower quantity of the long-range male sex pheromone (4R,5S)-5-hydroxy-4-decanolide (RS-HDL). Crosses between the two species yielded hybrid males with various pheromone quantities, and these males were used in mating trials with females of either species to measure female mate discrimination rates. A quantitative trait locus (QTL) analysis involving 475 recombinant hybrid males (F2), 2148 reciprocally backcrossed females (F3), and a linkage map of 52 equally spaced neutral single nucleotide polymorphism (SNP) markers plus SNPs in 40 candidate mating behavior genes revealed four QTL for male pheromone amount, depending on partner species. Our results demonstrate that the RS-HDL pheromone plays a role in the mating system of N. giraulti and N. oneida, but also that additional communication cues are involved in mate choice. No QTL were found for female mate discrimination, which points at a polygenic architecture of female choice with strong environmental influences.

Keywords: Nasonia courtship; QTL analysis; female choice; sex pheromone; speciation.

Figure 1

Experimental design. Reciprocal intraspecific and interspecific crosses between Nasonia giraulti and N. oneida. (A) N. giraulti male × N. giraulti female (B) N. giraulti male × N. oneida female (C) N. oneida male × N. giraulti female and (D) N. oneida male × N. oneida female. The parental interspecific cross [panels (B) and (C)] generates genetically identical F₁ hybrid females that produce unique recombinant F₂ hybrid male offspring of a 50:50 genetic mixture of the two parental species. Backcrossing of F₂ hybrid males to parental strain females yielded sibships of hybrid F₃ females with a 75:25 ratio of either parental species’ genome. The haploid hybrid male genotype is indicated with the paternal species on top and the maternal species below. The diploid hybrid F₃ backcrossed female genotype consists of a recombined hybrid set, and a pure species set, resulting in 75% genome of one species, and 25% of the other species on average. The diploid hybrid female genotype is indicated with the paternal species first followed by the maternal species. The square brackets indicate species cytoplasm, which is maternally inherited. Male mating behavior and male pheromone quantity were investigated of individual F₂ hybrid males (gray shading), and female mate discrimination was investigated on clonal sibships of F₃ females (gray shading). G, N. giraulti; O, N. oneida.

Figure 2

Latency time (mean ± SE) of pure species and hybrid males crossed with either parental species females. Crosses involving N. oneida females show significantly longer latency times. The genotype labeling is as in Figure 1. Sample sizes are shown within the bars. Different lower case letters indicate significant differences in means between crosses (glm, Tukey test, P < 0.05).

Figure 3

Mating behavior progress in pure species females. Proportions reaching subsequent stage in the mating behavior process are shown for (A) intraspecific and interspecific crosses of pure species, and (B) crosses of hybrid males with either parental species females. Mounting is the most discriminatory stage in both types of crosses, but subsequent stages of mating behavior in the hybrid male crosses are also more often terminated. The genotype labeling is as in Figure 1.

Figure 4

Phenotypic and QTL mapping results of pheromone quantity. (A) Phenotypic results of pheromone quantities of pure and hybrid males. N. giraulti males have higher pheromone quantities than N. oneida males. Hybrid males show many transgressive phenotypes. There is a significant interaction effect between hybrid males and female partner species. Box plots show the median (thick horizontal line within the box), the 25 and 75 percentiles (box), and 1.5 times the interquartile range of the data (thin horizontal lines). Outliers are indicated by an open circle. The genotype labeling is as in Figure 1. Sample sizes are shown within the bars. Significant differences between males are shown with letters on top of the panel. Note the y-axis scale break. Significant differences (Mann-Whitney U-test, P < 0.05) between crosses are shown with lower case letters. (B) QTL mapping results for male pheromone quantity. The shaded region is the 95% confidence interval for the significant QTL with the vertical line indicating the QTL peak location. The dashed line shows the 5% genome-wide significance level from permutation tests out of single-QTL genome scan. Upper and lower panel show results for males mated to N. oneida and N. giraulti females, respectively. Sample sizes are shown in the left upper corner and female partner species (backcross) in the right upper corner.

Figure 5

Relationship between male pheromone quantities and copulation success of hybrid males. Male pheromone quantities significantly differ between successful and unsuccessful copulations in crosses with N. oneida females, and almost significantly in N. giraulti females (Mann-Whitney U-test, P = 0.051 and P = 0.061, respectively). Box plots show the median (thick horizontal line within the box), the 25 and 75 percentiles (box), and 1.5 times the interquartile range of the data (thin horizontal lines). Outliers are indicated by an open circle. Pheromone quantities up to 60 ng are shown. Numbers of extra outliers higher than 60 ng are listed on top of the panel. Sample sizes are shown within the bars.

Figure 6

Mate discrimination. (A) Female genomic compositions with the corresponding acceptance rate of their male partner (%), and (B) QTL mapping results for female mate discrimination. Strong mate discrimination occurs in interspecific crosses between pure N. oneida females and pure N. giraulti males, and in interspecific crosses between hybrid F₃ females whose genomic compositions were 75% of one species and 25% of the other species, and pure N. giraulti males but not pure N. giraulti females. Number of males accepted and number of pairs observed are shown in Table 2. The dashed line in (B) shows the 5% genome-wide significance level from permutation tests of the single-QTL genome scan. Upper and lower panel show results for females mated to N. oneida and N. giraulti males, respectively. Sample sizes are shown in the left upper corner, and species names of the male partner in the right upper corner.

Figure 7

Mating behavior progress in hybrid females. Proportions reaching subsequent stage in the mating behavior process are shown for different hybrid F₃ females. Failure of female arrest and male display occur more often in crosses with N. giraulti (dashed lines) than with N. oneida males (solid lines). The genotype labeling is as in Figure 1.