Poor condition and infection: a vicious circle in natural populations

Abstract

Pathogens may be important for host population dynamics, as they can be a proximate cause of morbidity and mortality. Infection dynamics, in turn, may be dependent on the underlying condition of hosts. There is a clear potential for synergy between infection and condition: poor condition predisposes to host infections, which further reduce condition and so on. To provide empirical data that support this notion, we measured haematological indicators of infection (neutrophils and monocytes) and condition (red blood cells (RBCs) and lymphocytes) in field voles from three populations sampled monthly for 2 years. Mixed-effect models were developed to evaluate two hypotheses, (i) that individuals with low lymphocyte and/or RBC levels are more prone to show elevated haematological indicators of infection when re-sampled four weeks later, and (ii) that a decline in indicators of condition is likely to follow the development of monocytosis or neutrophilia. We found that individuals with low RBC and lymphocyte counts had increased probabilities of developing monocytosis and higher increments in neutrophils, and that high indices of infection (neutrophilia and monocytosis) were generally followed by a declining tendency in the indicators of condition (RBCs and lymphocytes). The vicious circle that these results describe suggests that while pathogens overall may be more important in wildlife dynamics than has previously been appreciated, specific pathogens are likely to play their part as elements of an interactive web rather than independent entities.

Figure 1

(a) Predicted development of monocytosis by sex (grey, female; black, male). Simulation for recaptured (‘R’) 22 g individuals at fixed past density (50 voles per grid) in 2005. In the simulation, ‘anaemic’ (dashed lines) are individuals with 3 million RBCs μl⁻¹, and ‘normal’ (solid lines) are voles with 8 million RBCs μl⁻¹ (eJun, early June; lJun, late June). (b) Predicted development of monocytosis for different levels of lymphocytes and population densities (solid line, 30 voles per grid; dashed line, 60; dotted line, 90). Simulation for 22 g recaptured (‘R’) males in June of 2005.

Figure 2

Predicted neutrophil change by weight (dotted line, 35 g; solid line, 25 g; dashed line, 17 g). Simulation for recaptured (‘R’) males in April at a fixed past density (50 voles per grid).

Figure 3

Simulations of the LMM describing changes of RBC counts in four weeks following (a) monocytosis (dashed, monocytosis; solid, normal) or (b) neutrophilia (dotted line, neutrophilia; solid line, normal). Three different weights (black, 17 g; dark grey, 22 g; light grey, 35 g) were simulated in (a), but in (b) the simulation corresponds to a 22 g vole. Current host density was fixed at 90 voles per grid, the average population size in summer–autumn (eJun, early June; lJun, late June).