How should behavioural ecologists interpret measurements of immunity?

S everal important issues in behavioural ecology require an estimate of an individual’s disease resistance (e.g. Møller et al. 1999). Unfortunately, estimating disease resistance is not straightforward, as many researchers have pointed out (e.g. Lochmiller 1995; Siva-Jothy 1995; Sheldon & Verhulst 1996; Zuk 1996; Apanius 1998; Westneat & Birkhead 1998; Norris & Evans 2000; Zuk & Stoehr 2002). This challenge is not unique to behavioural ecology. Ecological toxicologists testing whether chemicals in our environment are immunosuppressive, immunologists and nutritionists studying how diet can influence the immune system, physiologists examining the effect of stress on immune systems and agriculturists and aquaculturists attempting to produce disease-resistant stock all require the ability to assess the relative strength of an individual’s immune system. The most common method of estimating immunocompetence (i.e. the magnitude and effectiveness of an animal’s immune response) in all these fields is to measure one or more components of the immune system. The implicit assumption is that these measures correlate with the ability to resist disease, or are at least an indicator of the relative ‘strength’ of the immune system. Unfortunately, recent research in these fields has shown that the connection between assays of immunity and disease resistance is complex. In this paper I will focus on the problems these complexities create for the interpretation of immune assay results. Most studies assume that a low value on an immune assay (e.g. the ability to form antibodies to a novel antigen) corresponds to lower disease resistance, or, at least an immune system that shows a less robust response to pathogens even if lower disease resistance cannot be demonstrated. Unfortunately, immunological studies demonstrate that neither assumption is necessarily true (e.g. Luster et al. 1993; Berczi & Nagy 1998; S. Wilson et al. 2001; Smith 2003). The major problems in interpreting the results of immune assays are: (1) correlations between assays of immunity and disease resistance are typically pathogen specific, (2) correlations between assays of immunity and disease resistance are sometimes weak or nonexistent, (3) research suggests that some immune components have a threshold value such that changes above that threshold value may have no biological significance, and (4) the immune system can change its response characteristics in order to optimize its defence against different kinds of intruders. Below I discuss these issues and their implications for behavioural ecologists.