G·A·B·B·A

Understanding the genetic basis of adaptive phenotypic variation is central to our understanding of the origins and maintenance of biological diversity. Repeated occurrence of the same phenotypes in closely or distantly related populations is a very powerful tool for testing the role of natural selection in maintenance of those phenotypes. Research into the molecular basis behind similar phenotypic change provides the best opportunity to unite long-standing ideas about the extent to which evolutionary change is constrained. Do similar phenotypes always diversify by the same genetic bases or does selection uses many alternative genomic routes to the same phenotypic ends? Do these changes mainly occur from already available variation in the genome or is adaptation dependent on the incoming mutation? In this dissertation we address these questions using different populations of Mexican blind cavefish (Astyanax mexicanus) as our model, and by taking an integrative approach using the tools of population genetics, quantitative genetics and genomics. This species is very unique, with 30 different cave populations derived from surface populations. There are numerous morphological differences between the cave adapted and closely related surface forms, including reduction in pigmentation and eye size, hypertrophy of nonoptic sensory organs, reduced metabolic rate, increased numbers of taste buds, changes in numbers of ribs as well as multiple behavioral changes. First we asked how many independent times did these morphological traits repeatedly evolved in the cave populations. We assessed genetic structure and differentiation within and among the populations using genetic data from 568 fishes from 10 cave and 11 surface localities, and 26 genetically unlinked microsatellite loci. The widespread surface localities are, with some exceptions, genetically similar to one another, whereas the cave populations are differentiated and have at least five distinct origins in the three main regions. We find lower genetic diversity in cave populations than in related surface populations due to their smaller effective population sizes, probably because of limitations in food and space. However some of the cave populations receive migrants from the surface and exchange migrants with one another, especially when geographically close. This admixture results in significant heterozygote deficiencies at numerous loci due to Wahlund effects. In cave populations receiving migrants from the surface, we identified small numbers of individuals that are both phenotypically and genotypically intermediate between the cave and surface forms, affirming gene flow from the surface. Our study confirmed that the cave populations are derived from two main surface stocks that we call “old” and “new” populations and that diverged about 6.7 Mya, based on estimates from a previous study. “New” cave populations are closer to the surface populations while “old” cave populations are more distantly related to surface and “new” populations. In addition to that, our results suggest the old stock surface populations inhabited at least three independent cave localities while there are two independent localities inhabited by “new” stock surface populations. Thus we have established evolutionary convergence that refers to changes between “old” and “new” populations and parallel evolutionary system that refers to the changes between the populations within each of these groups. This part of the study clearly established the relationship between the phenotypically similar populations and allowed us to further investigate the importance of natural selection in the parallel and convergent evolution.
In the second part of the thesis we developed and genotyped 745 SNP markers in multiple cave and surface populations and further asked: can we find loci that were repeatedly selected for in the cave environment? All together, 80 loci were identified in several independent populations and they are potentially involved in adaptation to the cave environment.
Next, we asked where these markers are positioned in the genome and whether they coincide with regions involved in the phenotypic traits. Since the physical genome of cavefish is not available we integrated our information with the data from laboratory crosses. We used an F2 cross between the cave and surface individuals and genotyped the same SNP markers in the F2 progeny. This allowed us to design a genetic map. Measures of 10 phenotypic traits that differ between cave and surface populations were available from previous studies. We used quantitative trait loci analysis (QTL) in essence correlating genotype with phenotype, to detect regions in the genome with gene loci that are responsible for each phenotype.
Some of the 80 SNPs detected as adaptive in multiple natural populations also mapped to the QTL loci for lens, amino-acid sensitivity and eye size. Those SNPs were then joined into haplotypes. Some of these haplotypes denoting putative selection were found only in “new” cave populations, but others were found both in “new” as well as “old” cave populations.
Our study supports the hypothesis that convergent adaptive phenotypic change in different populations can arise through a conserved genetic basis (shared haplotypes in new and old cave populations). Furthermore, we observed the alternative possibility that implies that natural selection can repeatedly generate similar patterns of phenotypic variation in totally novel ways (haplotypes in only new cave populations).
Finally, we asked if those selected loci represent selection/fixation of per-existing variation or new mutations. We addressed this question by comparing the ancestral allele state (surface allele) and alleles of the multiple independent populations across identified QTL regions. We observed haplotypes that were repeatedly selected in cave populations of the new lineage but were present in very low frequencies in the surface populations, or at such low frequencies as to elude detection. These suggest that adaptation from standing genetic variation plays an important role in the adaptation to the cave environment.