Sex differences in immune responses: Hormonal effects, antagonistic selection, and evolutionary consequences

Highlights

• Immune responses differ between the sexes due to suppressive and enhancing effects of steroid sex hormones

• We hypothesize that sex hormone-induced differences in immune responses may lead to sexually antagonistic selection

• Sexually antagonistic selection is likely to be stronger for type 2 than for type 1 associated immune genes

• Sexually antagonistic selection on immune-genes should increase with the degree of sexual selection

• Mating behavior and social mating system may affect selection on immunity genes, particularly on type 2 associated genes

Abstract

Males and females differ in both parasite load and the strength of immune responses and these effects have been verified in humans and other vertebrates. Sex hormones act as important modulators of immune responses; the male sex hormone testosterone is generally immunosuppressive while the female sex hormone estrogen tends to be immunoenhancing. Different sets of T-helper cells (Th) have important roles in adaptive immunity, e.g. Th1 cells trigger type 1 responses which are primarily cell-mediated, and Th2 cells trigger type 2 responses which are primarily humoral responses. In our review of the literature, we find that estrogen and progesterone enhance type 2 and suppress type 1 responses in females, whereas testosterone suppresses type 2 responses and shows an inconsistent pattern for type 1 responses in males. When we combine these patterns of generally immunosuppressive and immunoenhancing effects of the sex hormones, our results imply that the sex differences in immune responses should be particularly strong in immune functions associated with type 2 responses, and less pronounced with type 1 responses. In general the hormone-mediated sex differences in immune responses may lead to genetic sexual conflicts on immunity. Thus, we propose the novel hypothesis that sexually antagonistic selection may act on immune genes shared by the sexes, and that the strength of this sexually antagonistic selection should be stronger for type 2- as compared with type 1-associated immune genes. Finally, we put the consequences of sex hormone-induced effects on immune responses into behavioral and ecological contexts, considering social mating system, sexual selection, geographical distribution of hosts, and parasite abundance.

Introduction

In vertebrates, the highly developed immune system provides protection from pathogens and parasites, but also from pathologies that are caused by the organism itself, e.g. tissue damage induced by oxidative stress or uncontrolled growth of tumors (Abbas et al., 2014, Murphy and Weaver, 2016). However, immune responses are costly, e.g., in terms of energy, nutrients, and immunopathology, and investments in immunity must therefore be traded off against investments in other important traits and activities (Hasselquist and Nilsson, 2012, Roff, 1992, Sheldon and Verhulst, 1996, Stearns, 1992, Zuk and Stoehr, 2002). Immune costs play out differently depending on reproductive and life history strategies across species, and such differences exist also within species or populations, for example between sex and age groups (Hasselquist, 2007, Klein, 2000, Sheldon and Verhulst, 1996, Zuk and Stoehr, 2002). In terms of sex, many empirical studies have found that males and females differ in parasite burden (Klein, 2004, Schmid-Hempel, 2011, Zuk and McKean, 1996) and that there can be sex differences in immune responsiveness (Klein and Roberts, 2010, Schuurs and Verheul, 1990). Moreover, it has been known for over 30 years that sex hormones can affect immune responses (e.g., Ansar Ahmed et al., 1985, Grossman, 1985).

Folstad and Karter (1992) connected the findings on parasite burden, immune responsiveness, and sex hormones, and argued that higher parasite burdens and lower immune responses in males compared to females was a result of immunosuppressive effects of the androgen hormone testosterone in males. They named their idea the ‘immunocompetence handicap hypothesis’ (ICHH) and proposed that this mechanism could explain why secondary sexual traits in males could be honest signals of male quality (Folstad and Karter, 1992). Specifically, Folstad and Karter suggested that androgen-dependent secondary sexual signals in males are adjusted according to a trade-off between the potential cost of infection and the benefits of higher reproductive success achieved by increased mating success (Folstad and Karter, 1992). Males with stronger secondary sexual signals (for example larger secondary sexual ornaments) suffer a reduction in immunocompetence (due to suppressed immune responses) as an effect of increased testosterone levels, and these males therefore run an increased risk of infection which ensures that only high quality males with ‘good genes for disease resistance’ can afford the most elaborate secondary sexual characters (Folstad and Karter, 1992).

Previous studies have suggested that sex hormones and mating behavior may interact to affect selection patterns on immune function (Hasselquist, 2007, Rolff, 2002, Zuk, 1990, Zuk and McKean, 1996). The selective advantage of high testosterone levels in males should be determined by the strength of sexual selection, and this is often affected by the social mating system of the species/population (Hasselquist, 2007, Zuk, 1990). For example, if males may increase their mating success considerably by having high expression of testosterone-dependent traits, it will impose a selection on males for increased testosterone levels, despite the risk of having strongly suppressed immune responses. Under such conditions the difference in the strength of immune responses between the sexes may be particularly large. Variation in mating behaviors (or social mating systems) have been predicted to affect immune responses according to the following general trends (Hasselquist, 2007, Zuk, 1990):

• Larger sex differences in immune responses in species with socially polygynous mating systems.

• Larger sex differences in immune responses in species where males are more involved in fighting and thus generally have higher testosterone levels, e.g., species where intra-sexual selection is strong.

• Larger sex differences in immune responses in species with a higher degree of inter-sexual selection (but smaller sex difference than in intra-sexual selection species).

• No or only small differences in immune responses between the sexes in species with socially monogamous mating system and low degree of sexual selection.

Social mating systems and sexual selection may also emphasize specific aspects of immunity, for example related to wound healing in species with strong intra-sexual selection (Zuk and Johnsen, 1998), or to stress-induced risk of immunopathology (e.g. autoimmunity, Raberg et al., 1998). The latter could differ between the sexes if there are differences in the amount of parental care that is provided by the male and the female parent (cf. Raberg et al., 2000). Furthermore, in species with large size dimorphism between the sexes (as is often characteristic of species with strong intra-sexual selection), the larger sex may suffer increased risk of contracting infections with food-transmitted parasites and pathogens, because of the increased food intake needed to maintain a larger body (Folstad and Karter, 1992).

Many studies have investigated the links between testosterone and immune function, for example when testing the immunocompetence handicap hypothesis. In a first meta-analysis to summarize this type of data by Roberts et al. (2004), a link between elevated testosterone and immunosuppression was found, although the effect was not significant when controlled for multiple studies conducted on the same species. In this study, there was no effect of testosterone on direct measures of immune function (white blood cell counts, antibodies, or response to phytohemagglutinin), but testosterone was shown to increase ectoparasite burdens, especially in reptiles (Roberts et al., 2004). Recently, Foo et al. (2016) conducted a meta-analysis on an extended data set that included both females and males. They investigated the effects of male (testosterone) and female (estrogen) sex hormones on immune function, because previous studies have found that also female sex hormones may affect immune function (Klein, 2004, Schuurs and Verheul, 1990, Zuk and McKean, 1996). Moreover, they also included the effects of sex hormones on more detailed measures of immune responses, such as cell-mediated and humoral immunity (explained below in “The vertebrate immune system”). Foo et al. (2016) found that, in studies where sex hormones were experimentally added, the general pattern was that testosterone had a medium-sized immunosuppressive effect, while estrogen showed no clear overall pattern. The effects of estrogen depended on the immune measure in question; experimentally elevated estrogen increased parasite resistance (i.e., reduced parasite load), suppressed cell-mediated immunity, and tended to enhance humoral immunity (p = 0.07). When considering correlational studies of natural levels of circulating sex hormones, the levels of testosterone in males showed no relationship either with overall immune function or humoral immunity, but there was a significant positive relationship with cell-mediated immunity (p = 0.04). In females, natural levels of estrogen showed a positive relationship with overall immune function (p = 0.01), although it became non-significant when phylogeny was accounted for (p = 0.16). Moreover, there was no significant relationship between circulating levels of estrogen and cell-mediated immunity, though there was a tendency for a positive relationship with humoral immunity (p = 0.08).

The meta-analysis by Foo et al. (2016) (summarized in Table 1) provides evidence that testosterone has a general suppressive effect on immune function, whereas the effects of estrogen depend on the immune measure in question. However, there is a substantial heterogeneity in the results of the meta-analysis, particularly concerning the effects on cell-mediated and humoral immunity, and this encouraged us to examine the effects of sex hormones on immunity in more detail. In the present review, we summarize the knowledge about the effects of sex hormones on different aspects of the vertebrate immune system, discuss possible evolutionary and ecological consequences of such sex differences, and propose a novel hypothesis for how sex hormone-induced differences in immune responses can affect the evolution of immune genes.