Limits...
Differentiation in putative male sex pheromone components across and within populations of the African butterfly Bicyclus anynana as a potential driver of reproductive isolation

View Article: PubMed Central - PubMed

ABSTRACT

Sexual traits are often the most divergent characters among closely related species, suggesting an important role of sexual traits in speciation. However, to prove this, we need to show that sexual trait differences accumulate before or during the speciation process, rather than being a consequence of it. Here, we contrast patterns of divergence among putative male sex pheromone (pMSP) composition and the genetic structure inferred from variation in the mitochondrial cytochrome oxidase 1 and nuclear CAD loci in the African butterfly Bicyclus anynana (Butler, 1879) to determine whether the evolution of “pheromonal dialects” occurs before or after the differentiation process. We observed differences in abundance of some shared pMSP components as well as differences in the composition of the pMSP among B. anynana populations. In addition, B. anynana individuals from Kenya displayed differences in the pMSP composition within a single population that appeared not associated with genetic differences. These differences in pMSP composition both between and within B. anynana populations were as large as those found between different Bicyclus species. Our results suggest that “pheromonal dialects” evolved within and among populations of B. anynana and may therefore act as precursors of an ongoing speciation process.

No MeSH data available.


Location and putative male sex pheromone (pMSP) composition of the Bicyclus anynana populations. Sampling sites are represented by asterisks excepted for Ishasha where only genetic data were sampled. Hypothetical repartition range derived from Condamin (1973) is represented in plain red. The two CAD haplotypes corresponding to B. anynana subspecies are in green and blue dashed lines. The bar plots represent the log‐transformed average amount (ng) of each pMSP component in each population (±SD). The sample size is given for each graph (within brackets for the preliminary samples from Uganda). The number below each bar (“comp.”) codes the pMSP component: 1: (Z)‐9‐tetradecen‐1‐ol; 2: hexadecanal; 3: 6,10,14‐trimethylpentadecan‐2‐ol (the three active MSP identified from the laboratory stock of B. anynana, Nieberding et al., 2008); 4: unidentified anynana #19; 5: octadecan‐1‐ol (physiologically active in GC‐EAD tests of individuals from False Bay, Nieberding et al., unpublished data); (see Table S2 and Fig. S1 for details of their identification). The abundance of each pMSP component selected once is represented for all the populations. The counts (“pres.”) below the compounds number represent the number of individuals displaying the pMSP component in the population. The bars are light colored for the populations in which the compound was not selected as a pMSP component.
© Copyright Policy - creativeCommonsBy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5016632&req=5

ece32298-fig-0001: Location and putative male sex pheromone (pMSP) composition of the Bicyclus anynana populations. Sampling sites are represented by asterisks excepted for Ishasha where only genetic data were sampled. Hypothetical repartition range derived from Condamin (1973) is represented in plain red. The two CAD haplotypes corresponding to B. anynana subspecies are in green and blue dashed lines. The bar plots represent the log‐transformed average amount (ng) of each pMSP component in each population (±SD). The sample size is given for each graph (within brackets for the preliminary samples from Uganda). The number below each bar (“comp.”) codes the pMSP component: 1: (Z)‐9‐tetradecen‐1‐ol; 2: hexadecanal; 3: 6,10,14‐trimethylpentadecan‐2‐ol (the three active MSP identified from the laboratory stock of B. anynana, Nieberding et al., 2008); 4: unidentified anynana #19; 5: octadecan‐1‐ol (physiologically active in GC‐EAD tests of individuals from False Bay, Nieberding et al., unpublished data); (see Table S2 and Fig. S1 for details of their identification). The abundance of each pMSP component selected once is represented for all the populations. The counts (“pres.”) below the compounds number represent the number of individuals displaying the pMSP component in the population. The bars are light colored for the populations in which the compound was not selected as a pMSP component.

Mentions: We sampled four geographically distinct wild populations of B. anynana (Fig. 1 and Table 1). Notably, the populations in Uganda are of the subspecies Bicyclus anynana centralis Condamin, 1968, while others are regarded Bicyclus anynana anynana Condamin, 1968. Among the 160 sampled individuals, a subset of 35 males and 33 females was used for the chemical analyses (Table 1). Most of the males and females in each location were sampled during a single trapping session of 1 week (Tables 1 and S1). We also sampled B. anynana individuals from two laboratory stocks, the first one originating from Malawi (Nkhata Bay, from 80 females sampled in 1988, Brakefield et al. 2009) and used in most studies on B. anynana, and the second one sampled in 2006 from South Africa (False Bay, from over 70 females, sampled in 2006, de Jong et al. 2010).


Differentiation in putative male sex pheromone components across and within populations of the African butterfly Bicyclus anynana as a potential driver of reproductive isolation
Location and putative male sex pheromone (pMSP) composition of the Bicyclus anynana populations. Sampling sites are represented by asterisks excepted for Ishasha where only genetic data were sampled. Hypothetical repartition range derived from Condamin (1973) is represented in plain red. The two CAD haplotypes corresponding to B. anynana subspecies are in green and blue dashed lines. The bar plots represent the log‐transformed average amount (ng) of each pMSP component in each population (±SD). The sample size is given for each graph (within brackets for the preliminary samples from Uganda). The number below each bar (“comp.”) codes the pMSP component: 1: (Z)‐9‐tetradecen‐1‐ol; 2: hexadecanal; 3: 6,10,14‐trimethylpentadecan‐2‐ol (the three active MSP identified from the laboratory stock of B. anynana, Nieberding et al., 2008); 4: unidentified anynana #19; 5: octadecan‐1‐ol (physiologically active in GC‐EAD tests of individuals from False Bay, Nieberding et al., unpublished data); (see Table S2 and Fig. S1 for details of their identification). The abundance of each pMSP component selected once is represented for all the populations. The counts (“pres.”) below the compounds number represent the number of individuals displaying the pMSP component in the population. The bars are light colored for the populations in which the compound was not selected as a pMSP component.
© Copyright Policy - creativeCommonsBy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5016632&req=5

ece32298-fig-0001: Location and putative male sex pheromone (pMSP) composition of the Bicyclus anynana populations. Sampling sites are represented by asterisks excepted for Ishasha where only genetic data were sampled. Hypothetical repartition range derived from Condamin (1973) is represented in plain red. The two CAD haplotypes corresponding to B. anynana subspecies are in green and blue dashed lines. The bar plots represent the log‐transformed average amount (ng) of each pMSP component in each population (±SD). The sample size is given for each graph (within brackets for the preliminary samples from Uganda). The number below each bar (“comp.”) codes the pMSP component: 1: (Z)‐9‐tetradecen‐1‐ol; 2: hexadecanal; 3: 6,10,14‐trimethylpentadecan‐2‐ol (the three active MSP identified from the laboratory stock of B. anynana, Nieberding et al., 2008); 4: unidentified anynana #19; 5: octadecan‐1‐ol (physiologically active in GC‐EAD tests of individuals from False Bay, Nieberding et al., unpublished data); (see Table S2 and Fig. S1 for details of their identification). The abundance of each pMSP component selected once is represented for all the populations. The counts (“pres.”) below the compounds number represent the number of individuals displaying the pMSP component in the population. The bars are light colored for the populations in which the compound was not selected as a pMSP component.
Mentions: We sampled four geographically distinct wild populations of B. anynana (Fig. 1 and Table 1). Notably, the populations in Uganda are of the subspecies Bicyclus anynana centralis Condamin, 1968, while others are regarded Bicyclus anynana anynana Condamin, 1968. Among the 160 sampled individuals, a subset of 35 males and 33 females was used for the chemical analyses (Table 1). Most of the males and females in each location were sampled during a single trapping session of 1 week (Tables 1 and S1). We also sampled B. anynana individuals from two laboratory stocks, the first one originating from Malawi (Nkhata Bay, from 80 females sampled in 1988, Brakefield et al. 2009) and used in most studies on B. anynana, and the second one sampled in 2006 from South Africa (False Bay, from over 70 females, sampled in 2006, de Jong et al. 2010).

View Article: PubMed Central - PubMed

ABSTRACT

Sexual traits are often the most divergent characters among closely related species, suggesting an important role of sexual traits in speciation. However, to prove this, we need to show that sexual trait differences accumulate before or during the speciation process, rather than being a consequence of it. Here, we contrast patterns of divergence among putative male sex pheromone (pMSP) composition and the genetic structure inferred from variation in the mitochondrial cytochrome oxidase 1 and nuclear CAD loci in the African butterfly Bicyclus anynana (Butler, 1879) to determine whether the evolution of “pheromonal dialects” occurs before or after the differentiation process. We observed differences in abundance of some shared pMSP components as well as differences in the composition of the pMSP among B. anynana populations. In addition, B. anynana individuals from Kenya displayed differences in the pMSP composition within a single population that appeared not associated with genetic differences. These differences in pMSP composition both between and within B. anynana populations were as large as those found between different Bicyclus species. Our results suggest that “pheromonal dialects” evolved within and among populations of B. anynana and may therefore act as precursors of an ongoing speciation process.

No MeSH data available.