The evolution and maintenance of intraspecific communication channels constitutes a key feature of chemical signaling and sexual communication. However, how divergent chemical communication channels evolve while maintaining their integrity for both sender and receiver is poorly understood. In the present study, we compare male and female cuticular hydrocarbon (CHC) profiles in the jewel wasp genus Nasonia, analyze their chemical divergence, and investigate their role as species-specific sexual signaling cues. Males and females of all four Nasonia species showed unique, non-overlapping CHC profiles unambiguously separating them. Surprisingly, male and female phylogenies based on the chemical distances between their CHC profiles differed dramatically, where only male CHC divergence parallels the molecular phylogeny of Nasonia. In particular, N. giraulti female CHC profiles were the most divergent from all other species and very different from its most closely related sibling species N. oneida. Furthermore, although our behavioural assays indicate that female CHC can generally be perceived as sexual cues attracting males in Nasonia, this function has apparently been lost in the highly divergent female N. giraulti CHC profiles. Curiously, N. giraulti males are still attracted to heterospecific, but not to conspecific female CHC profiles. We suggest that this striking discrepancy has been caused by an extensive evolutionary shift in female N. giraulti CHC profiles, which are no longer used as conspecific recognition cues. Our study constitutes the first report of an apparent abandonment of a sexual recognition cue that the receiver did not adapt to.
Nasonia GC-MS raw data
CSV files containing information on the GC-MS data (GC: 7890A; MS: 5975C; Agilent Technologies, Waldbronn, Germany) for all Nasonia samples, obtained with “Enhanced Chemstation”, G1701AA, Version A.03.00. Columns are identified as follows: Peak number, average retention time, peak type, peak width, peak area (integrated absolute value), start retention time, end retention time. Further description of the data generation see Materials and Methods section in the paper.
Trichomalopsis GC-MS raw data
CSV files containing information on the GC-MS data (GC: 7890A; MS: 5975C; Agilent Technologies, Waldbronn, Germany) for the Trichomalopsis outgroup samples, obtained with “Enhanced Chemstation”, G1701AA, Version A.03.00. Columns are identified as follows: Peak number, average retention time, peak type, peak width, peak area (integrated absolute value), start retention time, end retention time. Further description of the data generation see Materials and Methods section in the paper.
Normalized peak areas
CHC peak areas standardized with the normalization method of the function "decostand" of the community ecology R package “vegan” (Dixon, 2003), for a more detailed description see Materials and Methods section in the paper.
LDA scaling function values
Functions of the linear discriminant analysis (DA) calculated from the normalized CHC data on the Nasonia samples, first three functions plotted simultaneously in Figure 1 in the paper, further details on the DA see Materials and Methods and Results, respectively.
Phylogenetic distance matrix
Phylogenetic distance matrix, used in comparison with the average manhattan chemical distances between the Nasonia species based on the relative differences between their normalized CHC peak areas, performed separately for males and females (Fig. 2 and 3, respectively). The molecular distances are based on the Nasonia phylogeny, which are themselves based on the genetic divergence of a mitochondrial gene, COI sequences retrieved from NCBI, pairwise p-distances calculated with MEGA 3.0 (Werren et al., 2010) For further information, please see Materials and Methods section in the paper as well as the Tables S2 and S3, which give the average manhattan distances the phylogenetic distances were compared to.
Replicated Mantel tests
Single results of the Mantel tests, performed five times for both male CHC divergence vs. Nasonia phylogeny and female CHC divergence versus Nasonia phylogeny, all five p values and their respective averages are given for both combinations, as well as the r value. Each single test was performed with 9999 permutations.
Behavioral assays
Raw data on the behavioral assays, focussing on N. oneida and N. giraulti males (Fig. 4). Whether a copulation has occurred or not is given in the second column for each pairing in a binary coding (1 = yes, 0 = no), and the average number of occurred copulations is given for each pairing with untreated and treated female dummies. Species affiliation and treatment of female dummies indicated in the caption of each pairing. Male species affiliation can be found in the main caption on each spread sheet. Furthermore, the male arrestment time has been recorded (third column), but those data have not been used in this study. Mating results on two further N. giraulti strains are given as well.
Distribution of CHC compounds
Relative (percentages) distribution of CHC compound classes among the species and sexes of Nasonia and the outgroup Trichomalopsis, data underlying Fig. S2.