Final Report Summary - BENG (Show me your colour: the neuroendocrine and molecular underpinnings of phenotypic variation in colour and aggression in cichlid fish)
Project context and objectives
The overall goal of the proposed project is to unravel the molecular and endocrine underpinnings of variation in coloration and agonistic behaviour in haplochromine cichlid fishes.
During the outgoing phase my research addressed the overall goal in the colour polymorphic species Astatotilapia burtoni, a haplochromine cichlid fish from Lake Tanganyika. Yellow and blue colour morphs are characterised by distinct differences in agonistic behaviours and hormone profiles. The melanocortin system modulates body coloration, stress physiology and behaviour, and I specifically tested as to whether the melanocortin system is differentially regulated between yellow and blue males in A. burtoni.
The project objectives from the beginning were:
1. to describe behaviour and hormone profiles of yellow and blue A. burtoni males;
2. to identify genes underlying variation in colour, aggression and physiology in colour morphs of A. burtoni;
3. to analyse the function of pleiotropic effects of hormones on colour and behaviour;
4. to apply some of the techniques that I have learned to the Lake Victoria species flock3.
Work performed and main results
1. Covariance between colour and behaviour
Using existing datasets from the Hofmann lab, I analysed for covariance between colour and behaviour. The main finding was that despite tremendous individual variation, yellow males are more aggressive than blue males. In addition, we found that yellow males showed a lower cortisol peak in response to an acute stressor than blue males. These findings are interesting and hint at a possible role for the melanocortin system since melanocortin peptides regulate both colour, behaviour and impinge directly on the stress axis (hypothalamic-pituitary-interrenal axis).
2. Gene expression in yellow and blue males
We did not find significant differences in messenger ribonucleic acid (mRNA) levels of several melanocortin-related genes in the preoptic area (POA), a major neuroendocrine integration centre in the brain. However, in the ventral skin we found that yellow males express mRNA levels of melanocortin receptors, consistent with the hypothesis that yellow and blue males differentially regulate the melanocortin system.
Before we were able to do gene-expression measurements, we had to clone the different genes that are part of the melanocortin system.
After cloning several melanocortin receptors in A. burtoni, I cloned agouti and three copies of pro-opiomelanocortin (POMC) gene in A. burtoni. Intriguingly, one copy of POMC contained a novel peptide with high expression level in the skin.
I learned how to use genomic databases, how to design primers and how to optimise quantitative polymerase chair reactions (qPCR) reactions. During my post-doctoral period, the draft genomes and transcriptomes of five cichlid species became available, which have been helpful in targeting these genes, specifically agouti and the promotor regions of all the target genes. Together with the lab technician we performed a promotor analysis of the three POMC paralogues.
Pharmacology
I tested the effect of a-MSH, a broad melanocortin agonist, and agouti, an antagonist on the MC1R, MC3R and MC4R. Using intra-peritoneal injections, we found rapid behavioural effects of both peptides, and a colour effect of a-MSH. Specifically, a-MSH increased yellowness in males and increased the rate of aggression, but only in blue males. Agouti did not affect colour but decreased the rate of aggression in both yellow and blue males. Collectively, these results suggest that a-MSH turns blue males into the yellow phenotype in terms of both behaviour and coloration. They also suggest that the melanocortin system in yellow males is up-regulated (or functionally more active) than in blue males. Our gene-expression data was consistent with that, with differences in gene expression levels of melanocortin-related genes in the skin (but not in the brain).
Heritability of yellow and blue nuptial coloration
I bred females with yellow and blue males to test for heritability of yellow and blue nuptial coloration. The results suggest that there is some limited heritability of yellow and blue coloration.
The aim during the return phase was to apply some of the techniques that I learned during the outgoing phase in understanding phenotypic variation in the Lake Victoria species flock. However, in Leiden, I became involved in a more applied-research topic on a single Lake Victoria cichlid species, namely studying the long-term consequences of human-generated noise in Haplochromis piceatus. The reason for this modification in project objective for the return phase was that the concepts and techniques that I acquired during the outgoing phase were highly suitable for studying the effects of noise on stress physiology and social behaviour in a cichlid fish. My collaborators at Leiden University included Drs Hans Slabbekoorn, Frans Witte and Niels Bouton. I trained two students, a post-doctoral researcher and a lab technician in blood collection and plasma preparation, characterisation of body coloration and quantification of social behaviour. We found that noise-treated fish courted less. We also found that group hierarchy was less defined in noise-treated fish. Collectively, these results indicate negative effects of noise on fish behaviour and group structure. The blood samples will be analysed soon to measure levels of the stress hormone cortisol. In addition, I wrote a draft for a review paper on the role of physiology in phenotypic diversification in East African cichlid fish. During the return phase I was still able to maintain collaborative ties with my former colleagues at the University of Texas at Austin. I also began writing up the results of my projects completed during the outgoing phase.
The expected final results and their potential impact and use
The societal implications of my project pertain to an overarching theme in my research, namely understanding why some groups of animals contain many species while others contain only a few. Such knowledge is of considerable applied interest because of the human impact on species diversity. My results inform us about the mechanistic basis of covariation in body coloration and aggression. Correlated traits are a big theme in speciation research; quantifying to what extent traits covary is important to predict the condition for adaptive divergence. In addition, correlated traits are important in the field of personality research and our works provides a more mechanistic understanding of individual variation in suites of traits. Furthermore, in Leiden I also became involved in understanding the effects of human-generated noise on captive cichlid fish. Underwater noise levels related to human activities are on the rise and can cause damage and disturbance to fish and interfere with their perception of relevant sounds. Previous studies focused primarily on short-term effects of loud sound bursts. Pile-driving sounds can, for example, cause auditory tissue damage and seismic airguns can drive fish away from noisy areas. However, very few studies addressed longer term consequences of more moderate human-generated noise levels.
The overall goal of the proposed project is to unravel the molecular and endocrine underpinnings of variation in coloration and agonistic behaviour in haplochromine cichlid fishes.
During the outgoing phase my research addressed the overall goal in the colour polymorphic species Astatotilapia burtoni, a haplochromine cichlid fish from Lake Tanganyika. Yellow and blue colour morphs are characterised by distinct differences in agonistic behaviours and hormone profiles. The melanocortin system modulates body coloration, stress physiology and behaviour, and I specifically tested as to whether the melanocortin system is differentially regulated between yellow and blue males in A. burtoni.
The project objectives from the beginning were:
1. to describe behaviour and hormone profiles of yellow and blue A. burtoni males;
2. to identify genes underlying variation in colour, aggression and physiology in colour morphs of A. burtoni;
3. to analyse the function of pleiotropic effects of hormones on colour and behaviour;
4. to apply some of the techniques that I have learned to the Lake Victoria species flock3.
Work performed and main results
1. Covariance between colour and behaviour
Using existing datasets from the Hofmann lab, I analysed for covariance between colour and behaviour. The main finding was that despite tremendous individual variation, yellow males are more aggressive than blue males. In addition, we found that yellow males showed a lower cortisol peak in response to an acute stressor than blue males. These findings are interesting and hint at a possible role for the melanocortin system since melanocortin peptides regulate both colour, behaviour and impinge directly on the stress axis (hypothalamic-pituitary-interrenal axis).
2. Gene expression in yellow and blue males
We did not find significant differences in messenger ribonucleic acid (mRNA) levels of several melanocortin-related genes in the preoptic area (POA), a major neuroendocrine integration centre in the brain. However, in the ventral skin we found that yellow males express mRNA levels of melanocortin receptors, consistent with the hypothesis that yellow and blue males differentially regulate the melanocortin system.
Before we were able to do gene-expression measurements, we had to clone the different genes that are part of the melanocortin system.
After cloning several melanocortin receptors in A. burtoni, I cloned agouti and three copies of pro-opiomelanocortin (POMC) gene in A. burtoni. Intriguingly, one copy of POMC contained a novel peptide with high expression level in the skin.
I learned how to use genomic databases, how to design primers and how to optimise quantitative polymerase chair reactions (qPCR) reactions. During my post-doctoral period, the draft genomes and transcriptomes of five cichlid species became available, which have been helpful in targeting these genes, specifically agouti and the promotor regions of all the target genes. Together with the lab technician we performed a promotor analysis of the three POMC paralogues.
Pharmacology
I tested the effect of a-MSH, a broad melanocortin agonist, and agouti, an antagonist on the MC1R, MC3R and MC4R. Using intra-peritoneal injections, we found rapid behavioural effects of both peptides, and a colour effect of a-MSH. Specifically, a-MSH increased yellowness in males and increased the rate of aggression, but only in blue males. Agouti did not affect colour but decreased the rate of aggression in both yellow and blue males. Collectively, these results suggest that a-MSH turns blue males into the yellow phenotype in terms of both behaviour and coloration. They also suggest that the melanocortin system in yellow males is up-regulated (or functionally more active) than in blue males. Our gene-expression data was consistent with that, with differences in gene expression levels of melanocortin-related genes in the skin (but not in the brain).
Heritability of yellow and blue nuptial coloration
I bred females with yellow and blue males to test for heritability of yellow and blue nuptial coloration. The results suggest that there is some limited heritability of yellow and blue coloration.
The aim during the return phase was to apply some of the techniques that I learned during the outgoing phase in understanding phenotypic variation in the Lake Victoria species flock. However, in Leiden, I became involved in a more applied-research topic on a single Lake Victoria cichlid species, namely studying the long-term consequences of human-generated noise in Haplochromis piceatus. The reason for this modification in project objective for the return phase was that the concepts and techniques that I acquired during the outgoing phase were highly suitable for studying the effects of noise on stress physiology and social behaviour in a cichlid fish. My collaborators at Leiden University included Drs Hans Slabbekoorn, Frans Witte and Niels Bouton. I trained two students, a post-doctoral researcher and a lab technician in blood collection and plasma preparation, characterisation of body coloration and quantification of social behaviour. We found that noise-treated fish courted less. We also found that group hierarchy was less defined in noise-treated fish. Collectively, these results indicate negative effects of noise on fish behaviour and group structure. The blood samples will be analysed soon to measure levels of the stress hormone cortisol. In addition, I wrote a draft for a review paper on the role of physiology in phenotypic diversification in East African cichlid fish. During the return phase I was still able to maintain collaborative ties with my former colleagues at the University of Texas at Austin. I also began writing up the results of my projects completed during the outgoing phase.
The expected final results and their potential impact and use
The societal implications of my project pertain to an overarching theme in my research, namely understanding why some groups of animals contain many species while others contain only a few. Such knowledge is of considerable applied interest because of the human impact on species diversity. My results inform us about the mechanistic basis of covariation in body coloration and aggression. Correlated traits are a big theme in speciation research; quantifying to what extent traits covary is important to predict the condition for adaptive divergence. In addition, correlated traits are important in the field of personality research and our works provides a more mechanistic understanding of individual variation in suites of traits. Furthermore, in Leiden I also became involved in understanding the effects of human-generated noise on captive cichlid fish. Underwater noise levels related to human activities are on the rise and can cause damage and disturbance to fish and interfere with their perception of relevant sounds. Previous studies focused primarily on short-term effects of loud sound bursts. Pile-driving sounds can, for example, cause auditory tissue damage and seismic airguns can drive fish away from noisy areas. However, very few studies addressed longer term consequences of more moderate human-generated noise levels.