Parasitism, Host Immune Function, and Sexual Selection Author(s): A. P. Moller, P. Christe and E. Lux Source: The Quarterly Review of Biology, Vol. 74, No. 1 (Mar., 1999), pp. 3-20 Published by: The University of Chicago Press Stable URL: http:/www.jstor.org/stable/2664375 . Accessed: 20/11/2013 12:27Your use of the JSTOR archive indicates your acceptance of the Terms Bor- gia 1986; Borgia and Collis 1989). The effi- cient parent hypothesis posits that females prefer extravagantly ornamented males be- cause such males are more efficient parents, simply because they can allocate more re- sources to paternal care than parasite-infested males (Hamilton 1990; M0ller 1990c). While the Hamilton-Zuk hypothesis provides a mecha- nism for choosy females to accrue indirect fit- ness benefits, the two alternatives provide mechanisms entirely based on direct fitness benefits. A number of different assumptions and pre- dictions have been raised for the three hypoth- eses. For example, the Hamilton-Zuk hypothe- sis predicts 1) a genetic correlation between parasite resistance and secondary sexual char- acters, and 2) a microevolutionary increase in parasite resistance as a consequence of female choice (Kirkpatrick and Ryan 1991). Their ob- vious intractability has apparently discour- aged tests of these predictions. The exact mechanism that generates a rela- tionship between the expression of secondary sexual characters and parasite load has been the subject of some debate. The original model proposed by Hamilton and Zuk (1982) assumed that secondaiy sexual characters acted as reliable indicators of male quality, based on a revealing handicap mechanism (see Andersson 1994). Alternatively, second- ary sexual characters influenced by parasites may act as true condition-dependent handi- caps, with the secondary sex trait reliably re- flecting the condition of individual males. Subsequent analyses have demonstrated that females may obtain resistance genes by choos- ing the most extravagantly ornamented males, even if their secondary sexual characters do not act as a handicap (Wedekind 1994a). Folstad and Karter (1992) proposed that the mechanism generating reliability in the sexual signaling system depended on the im- munosuppressive effects of testosterone, which is essential for full expression of male second- ary sexual characters in many vertebrates. This so-called immunocompetence handicap could act as a result of the constraints imposed by androgens on the signaling system. Alterna- tively, males may be able to optimize their level of sexual display without compromising their immune system (M0ller and Saino 1994; Wede- kind and Folstad 1994). A second type of mechanism that could directly enforce relia- bility of signaling by an indicator mechanism is based on the assumption that testosterone affects spermatogenesis, but simultaneously influences immune function negatively (Fol- stad and Skarstein 1997; Hillgarth et al. 1997). Hence females will obtain fertile and immu- nocompetent mates with low parasite loads by choosing the most extravagantly ornamented males (Folstad and Skarstein 1997; Hillgarth et al. 1997). Immunosuppression caused by androgens or corticosteroids can only apply to vertebrates, although similar tradeoff mecha- nisms based on other substances may apply to the reproductive physiology of invertebrates. The exact signals females use to determine male parasite status could be those most se- verely affected by parasites and, therefore, the most reliable signals. Since many organisms use multiple signals in sexual contexts, differ- ent signals may reveal infections with different kinds of parasites, as suggested by Wedekind (1992). The exact assessment mechanisms of male parasite infection status, and thus poten- tial parasite resistance, may be important since costly signals are particularly likely to play a role in condition-dependent sexual se- lection. Resistance genes could be revealed di- This content downloaded from 150.212.2.226 on Wed, 20 Nov 2013 12:27:52 PM All use subject to JSTOR Terms and ConditionsMARCH 1999 PARASITE-MEDIATED SEXUAL SELECTION 5 rectly from sexual signals such as odor in mam- mals, as determined by MHC-haplotype (Potts at al. 1991; Potts and Wakeland 1993; Wede- kind 1994b). For example, some studies of ro- dents and humans have now suggested that mate preferences are directly associated with MHC-dependent odors (Potts et al. 1991; Wedekind et al. 1995), and similar properties may be revealed by visual signals in other or- ganisms, such as birds (von Schantz et al. 1996). Finally, other olfactory properties of parasitized individuals, such as opionates, may reliably reveal the parasite status of potential partners (Hudson et al. 1992; Kavaliers and Colwell 1993). Male ornamentation is associated with heri- table parasite resistance, and parasite resis- tance is coupled with t