Body condition is central to animal life histories because the expression of many traits critical to survival and reproductive success is condition-dependent [1,2]. Condition-dependence is, therefore, a topic of broad interest in both natural and sexual selection. One particularly striking example of the fundamental role of condition-dependence is in the context of mate choice [3-5]. Females often choose among males on the basis of condition-dependent signals, which honestly advertise male quality as the expression of these signals may trade-off with other life-history traits [6,7]. Many explanations have been put forward to explain the ubiquity of condition-dependent life histories and signals, with one influential theory predicting that the adaptive significance of condition-dependent signals may arise from the large number of genes that may influence variation in condition, thereby offering females the opportunity to assess a substantial proportion of male genomes in determining male quality [8]. Under this hypothesis, selection favours females who based their mate choice decisions on condition-dependent signals because such behaviour increases the females' chances of obtaining good genes for their offspring. This line of reasoning can be extended to predict that one class of genes that may be of particular interest to females are those loci that contribute to variation in parasite resistance [9], a major determinant of reproductive success and survival in many species [10]. The condition-mediated immunocompetence-handicap hypothesis (CMIH) [7,11-14], proposes that females base their mate choice decisions on condition-dependent male signals in order to obtain genes that confer high immunocompetence for their offspring.
A key requirement of the CMIH hypothesis, and other related life history hypotheses [15-19], is the presence of positive genetic covariance between body condition and immune response. The CMIH hypothesis proposes that it is this genetic covariance that enables a condition-dependent signal to advertise the quality of the parasite resistance genes that a male carries [14]. Although there is abundant evidence for positive phenotypic associations between body condition and immunocompetence [12,15,16,18,20-22], phenotypic analyses are insufficient to validate the CMIH hypothesis because it remains unknown whether females selecting males with good body condition simply obtain a healthy mate, or if they actually acquire genes for their offspring that confer high immunocompetence. An additive genetic component has been established in several experimental systems for both body condition [[23], but see [24]] and immune response [[25-29]; but see [30-32]]. There is also evidence of a genetic correlation between immune function and sexual signals [33,34], between immune function and life history traits [35] and between body condition and male signal [36]. As far as we are aware, however, it has not been empirically demonstrated that variation in immune response is mediated by genetic variation in body condition, a key element of the CMIH hypothesis [14].
The overall aim of this study was to test directly for genetic covariance between indices of body condition and immunocompetence in a small passerine bird, the zebra finch Taeniopygia guttata. Zebra finches provide an ideal opportunity to determine if this critical genetic association exists for two reasons. First, this species is a model system for the study of sexual selection, in which female choice is based on a number of condition-dependent male signals that include song rate and bill colour [37-40]. Second, there is phenotypic evidence of condition-dependent expression of immunocompetence in this species [4,16]. We therefore used this system to investigate the genetic basis of covariation between an index of body condition and an index of immunocompetence using a cross-fostering experiment. Here we implement the cross-fostering experimental design of Riska et al. [42] to estimate additive genetic components of variance. An important advantage of this method was that it allowed us to partition the genetic covariance between these traits into sources attributable to direct additive genetic covariance, additive maternal genetic covariance, and the covariance between these sources.
Answers to all your Biology Questions
