Transcriptome assembly and candidate genes involved in nutritional programming in the swordtail fish Xiphophorus multilineatus

BACKGROUND: Nutritional programming takes place in early development. Variation in the quality and/or quantity of nutrients in early development can influence long-term health and viability. However, little is known about the mechanisms of nutritional programming. The live-bearing fish Xiphophorus multilineatus has the potential to be a new model for understanding these mechanisms, given prior evidence of nutritional programming influencing behavior and juvenile growth rate. We tested the hypotheses that nutritional programming would influence behaviors involved in energy homeostasis as well gene expression in X. multilineatus. METHODS: We first examined the influence of both juvenile environment (varied in nutrition and density) and adult environment (varied in nutrition) on behaviors involved in energy acquisition and energy expenditure in adult male X. multilineatus. We also compared the behavioral responses across the genetically influenced size classes of males. Males stop growing at sexual maturity, and the size classes of can be identified based on phenotypes (adult size and pigment patterns). To study the molecular signatures of nutritional programming, we assembled a de novo transcriptome for X. multilineatus using RNA from brain, liver, skin, testis and gonad tissues, and used RNA-Seq to profile gene expression in the brains of males reared in low quality (reduced food, increased density) and high quality (increased food, decreased density) juvenile environments. RESULTS: We found that both the juvenile and adult environments influenced the energy intake behavior, while only the adult environment influenced energy expenditure. In addition, there were significant interactions between the genetically influenced size classes and the environments that influenced energy intake and energy expenditure, with males from one of the four size classes (Y-II) responding in the opposite direction as compared to the other males examined. When we compared the brains of males of the Y-II size class reared in a low quality juvenile environment to males from the same size class reared in high quality juvenile environment, 131 genes were differentially expressed, including metabolism and appetite master regulator agrp gene. DISCUSSION: Our study provides evidence for nutritional programming in X. multilineatus, with variation across size classes of males in how juvenile environment and adult diet influences behaviors involved in energy homeostasis. In addition, we provide the first transcriptome of X. multilineatus, and identify a group of candidate genes involved in nutritional programming.