Evolutionary Genetics of Coloration in Cichlids

Abstract

Why are there so many species? Or, why aren’t there more species? The factors that lead to biodiversity have always been a matter of curiosity for biologists and the general public alike. Although the main patterns and the driving forces of speciation are now reasonably well-understood, there is still much to investigate. The genetic basis of speciation has a central role in testing the current theory of speciation but our knowledge is scarce and biased towards a few model organisms. How well these model organisms represent the speciation continuum is debatable. Methodological advances now allow biologists to look beyond traditional model organisms and to assess the workings of speciation in systems that, while not being genetic models are well-recognized model systems for speciation research. Cichlids are probably the most famous textbook example of speciation. The cichlid problem, i.e. existence of flocks of thousands of recently diverged species in isolated lakes, has occupied evolutionary biologists for decades. The question “why are there so many species of cichlids?” is clearly a difficult, perhaps even an unanswerable one. Nevertheless, there is consensus in the community that the extreme variation in coloration patterns, characteristic of cichlid fishes, contributes directly to these explosive speciation rates. The goals of the present thesis are to investigate the genetics of two cichlid color traits that could constitute speciation phenotypes. Aiming at providing resources to foster further forward genetic studies, a genetic map of the most investigated African cichlid, A. burtoni was constructed. The map comprises over 200 markers within 25 linkage groups. In addition to anonymous markers, genes known to influence adaptation and speciation in cichlids were also placed on the map. By including members of the Hox clusters, it was also found that the different clusters map to different linkage groups, which supports the origin of these clusters during the whole genome duplication. The present doctoral thesis also reports on two case-studies of the genetics of color-traits thought to influence speciation in cichlids. The first are eggspots. These are interesting biological traits characteristic of African cichlids. They are color patterns on the anal fins that mimic eggs and play several roles in the mating of cichlids of the haplochromine lineage. Many have suggested that eggspots are key evolutionary innovations that influence the speciation rates of haplochromines. The genetics and sexual advertisement functions of the numbers of eggspots were investigated using selective breeding and behavioral experiments, respectively. The number of eggspots was found to harbor extensive additive genetic variation and to respond asymmetrically to selection. There was also a correlated response of body size. Furthermore, no evidence was found for the previously proposed sexual advertisement functions. All of this is inconsistent with the view that such trait is under strong directional sexual selection. It seems more likely that either non-adaptive processes currently shape the distribution of eggspot numbers or that selection acting on a correlated trait does. The second case-study concerns the gold/normal polychromatism in the Midas cichlid species of the genus Amphilophus. Gold or normal colored adults exist in several different populations and species in Nicaragua. All fish start their lives with a grayish (normal) color pattern and most individuals maintain their melanophores (dark pigment cells) throughout their lives. Others undergo morphological color change due to a process of melanophore cell death and exhibit a distinctive white-to-gold coloration in adulthood. It was shown that the color morphs mate assortatively and have diverged significantly in at least two lakes. In the present thesis, the genetic architecture of the gold phenotype is shown to be a simple one. This incipient speciation phenotype is determined by a dominant allele at a single gene. It was found that the gold locus might also control mating preference. In order to identify the causative gene, a number of approaches were taken, including forward-genetics (family- and population-based mapping), candidate-gene and transcriptomic experiments. The genomic region harboring the gold locus contains only a few positional candidate-genes out of which a single one has a known pigmentation function. Significant genotype-phenotype association was found in this region in an independent, field-collected population sample. Known color-genes that determine similar phenotypes in other organisms were either unrelated to or appear to be downstream targets of the gold locus. Interestingly, the genes that showed an expression pattern consistent with a role in the process of morphological color change are also related to human pigmentation disorders (e.g. skin cancer, psoriasis). One curious and unexpected result was that heterozygous gold fish undergo color change later in life than homozygous fish. This observation has implications for the evolutionary dynamics of this allele in the wild. In summary, the present results contribute towards filling the knowledge gaps of cichlid pigmentation and speciation genetics. By questioning previous conceptions, confirming theoretical expectations and generating new and exciting hypotheses, the present thesis illustrates the importance of understanding the genetic bases of ecologically-relevant phenotypes.