9709 - Laserfiche WebLink

Home Browse Search

programs (Frankham et al. 2002). However, the evidence described above indicates that it is a very serious problem for species subject to long-term captive breeding prior to reintroduction into the wild. Clearly, genetic adaptation to captivity should be minimized for populations likely to be used for reintroduction. In this review, I address in the context of conservation biology the issues of (i) the genetic basis of adaptive changes in captivity and its implications, (ii) factors determining the extent of genetic adaptation to captivity, and (tii) the means for minimizing the deleterious impacts of genetic adaptation to captivity on reproductive fitness of populations returned to the wild. By necessity, much of the experimentation has been done using Drosophila, as threatened species are unsuitable for experimentation. This species has proven to be a reliable model in conserva- tion and evolutionary genetics and animal breeding for species with similar breeding systems (Frankham 2000). Genetic basis of adaptation to captivity Knowledge of the genetic basis of adaptation to captivity adds to our understanding the process and its consequences, and aids in devising means to minimize the deleterious consequences upon reintroduction. In populations at equilibrium in their wild environments, genetic diversity for reproductive fitness is believed to be due largely to rare, deleterious, partially recessive alleles in mutation-selection balance (Falconer & Mackay 1996; Charlesworth & Hughes 2000). Rodriguez-Ramilo et al. (2004) concluded on the basis of combined experimentation and modelling that genetic variation for larval viability in Drosophila melanogaster is due to solely to alleles in mutation-selection balance. Further, Frankham (1990) showed consistently asymmetrical response to selection for fitness characters, as expected from the above hypothesis. Rare, partially recessive alleles result in additive genetic variation and readily contribute to selection response (Frankham & Nurthen 1981; Falconer & Mackay 1996). Balancing selection results in alleles with intermediate frequencies, but the proportion of loci involved seems small. A large body of evidence indicates that heterozygote advantage occurs at few loci (Charlesworth & Hughes 2000; Swanson-Wagner et al. 2006), evidence indicates that frequency-dependent selection is uncommon (Dolan & Robertson 1975), and few polymorphic loci seem to be maintained by selective forces that vary in time or space (Hedrick 2006). Given that genetic adaptation to captivity is overwhelmingly deleterious when captive populations are returned to the wild, alleles that were previously rare and deleterious in the wild, but are favoured in captivity are expected to form the main basis of genetic adaptation to captivity. Five lines of empirical evidence support this scenario. First, null alleles that are rare in wild populations have been found at elevated frequencies in hatchery (captive) fish populations (Leary et al. 1993). Second, Bush et al. (1976) @ 2007 The Author Journal compilation @ 2007 Blackwell Publishing Ltd GENETIC ADAPTATION TO CAPTIVITY 3 found that a rare a-glycerol phosphate allele with lowered enzyme activity in screwworm flies (Cochliomyia hominivorax) increased in frequency in captivity and reduced wild fit- ness as it resulted in compromised flying ability. Third, an ADH allele with an initial frequency of 1 % in the olive fruit fly (Dacus oleae) increased to about 40% in captivity (Zouros et al. 1982). Fourth, crossing replicate D. melanogaster popu- lations that had been maintained for 50 generations in captivity with effective sizes of 500 led to outbreeding depression in captive fitness, indicating that the replicate populations had utilized at least partially different rare alleles (Woodworth et al. 2002). Fifth, crosses of replicate Drosophila populations of Ne = 500 and of Ne = 250 at gen- eration 45 yielded substantial benefits in fitness, compared to their uncrossed parent populations when they were compared in the 'wild' environment, implicating the involvement of rare alleles that are partially recessive in the wild (Woodworth et al. 2002). If initially rare alleles are involved in genetic adaptation to captivity, they should result in selective sweeps that reduce genetic diversity at nearby neutral loci as they rise in frequency. In accord with this prediction, microsatellite genetic diversity was lost at a greater than neutral rate in Drosophila populations maintained for 50 generations at effective population sizes of 25, 50, 250 and 500, and selec- tive sweeps is the only credible explanation (Montgomery, ME et al. unpublished data). Of eight loci, seven showed faster than neutral loss of genetic diversity, indicating that initially rare alleles at many loci are involved in genetic adaptation to captivity. There is little evidence on the dominance in captivity of rare alleles that are partially recessive in the wild, but they may differ. For example, the effects of crossing captive population on fitness were in different direction when measured in captivity and in wild conditions, as described in the fourth and fifth points above. Factors determining genetic adaptation to captivity The cumulate genetic change in reproductive fitness in captivity over t generations (GAt) can be predicted from the breeder's equation (Frankham et al. 2002), as follows: ( )t-1 GAt - 5h2 I 1 - 2~e (eqn 1), where 5 is the selection differential, h2 the heritability (dependent upon genetic diversity for reproductive fitness), Ne the effective population size, and t the number of generations in captivity. 5h2 is the response to selection in the first generation and the expression after sigma reflects loss of genetic diversity due to drift in subsequent generations. Genetic adaptation is measured in the same units used for the character being considered.