Ecological constraints and the evolution of hormone-behavior interrelationships.

Vertebrates show a diverse array of social behaviors. Equally complex are the mechanisms by which these behavioral patterns are regulated by hormones and the effects of behavioral interactions on hormone secretion. Nonetheless, comparative field and laboratory experiments indicate that general underlying themes, including mechanisms, may exist. For example, comparative studies in birds reveal that testosterone activates a type of aggression, territorial behavior, in those species that are territorial only during the breeding season. Territoriality at other times appears to be independent of sex steroid control, although qualitatively and quantitatively the behavior appears identical. Similarly, formation of pair bonds appears to be complex. In some populations such bonds are sexual, whereas in others they appear to be alliances possibly for joint defense of a territory. In cooperative groups of birds, pair bonds and alliances may exist simultaneously. Testosterone appears to be important for activation of the courtship behavior that leads to formations to sexual pair bonds. However, many investigations indicate that pair bonds in nonsexual contexts are not regulated by testosterone. Hormonal mechanisms underlying the establishment of alliances (if any) remain unknown. Clearly, these complex behavioral patterns due to seasonal changes and variation in context pose important questions for control mechanisms. One obvious question is, why this diversity in control mechanisms? It appears that there are evolutionary "costs" to high circulating levels of testosterone. They can be energetic costs or may involve increased predation risk or reduced survival after wounding. In males that express parental behavior, high circulating testosterone levels interfere with parental care, resulting in reduced reproductive success. Thus, regulation of testosterone secretion must balance the need to compete with other males as well as provide parental care. High circulating levels of testosterone for prolonged periods are also known to suppress the immune system. This latter effect may have profound implications for the development of androgen-dependent secondary sex characteristics that have evolved through sexual selection. There are several ways to avoid potential "costs" of hormone secretion at inappropriate times. A hormone may be metabolized at its target cell to another form that then binds to a different receptor (e.g., aromatization of testosterone to estradiol). Also receptors may be downregulated in tissues that would otherwise respond inappropriately in a specific life history state. On the other hand, multiple hormone mechanisms may have evolved to activate behavioral traits at the right time and in the correct context. When a behavioral trait is expressed throughout the life cycle, hormones may potentially deactivate behavior for short periods. With detailed investigations of organisms in their natural environment we can determine the potential ecological costs underlying hormone-behavior interactions that, in turn, shed light on their evolution. These data also indicate a number of problems for hormonal control mechanisms, but also indicate trends, alternatives, and hopefully in the future a more complete understanding of common mechanisms underlying behavioral endocrinology at the cell and molecular level. Only then will we be able to predict when and where specific mechanisms of hormone-behavior interactions operate and how they evolved.