An immunological marker of tolerance to infection in wild rodents

Abstract

Hosts are likely to respond to parasitic infections by a combination of resistance (expulsion of pathogens) and tolerance (active mitigation of pathology). Of these strategies, the basis of tolerance in animal hosts is relatively poorly understood, with especially little known about how tolerance is manifested in natural populations. We monitored a natural population of field voles using longitudinal and cross-sectional sampling modes and taking measurements on body condition, infection, immune gene expression, and survival. Using analyses stratified by life history stage, we demonstrate a pattern of tolerance to macroparasites in mature compared to immature males. In comparison to immature males, mature males resisted infection less and instead increased investment in body condition in response to accumulating burdens, but at the expense of reduced reproductive effort. We identified expression of the transcription factor Gata3 (a mediator of Th2 immunity) as an immunological biomarker of this tolerance response. Time series data for individual animals suggested that macroparasite infections gave rise to increased expression of Gata3, which gave rise to improved body condition and enhanced survival as hosts aged. These findings provide a clear and unexpected insight into tolerance responses (and their life history sequelae) in a natural vertebrate population. The demonstration that such responses (potentially promoting parasite transmission) can move from resistance to tolerance through the course of an individual's lifetime emphasises the need to incorporate them into our understanding of the dynamics and risk of infection in the natural environment. Moreover, the identification of Gata3 as a marker of tolerance to macroparasites raises important new questions regarding the role of Th2 immunity and the mechanistic nature of the tolerance response itself. A more manipulative, experimental approach is likely to be valuable in elaborating this further.

Figure 1. Patterns of tolerance and resistance in stratified analyses, with Gata3 as a biomarker.

Cross-sectional study (cross-sect.). (A–D) Differing associations in mature (ad) and immature male voles (j) between principal component scores representing grouped macroparasite infection (main influential species, PC^{M main}) and SVL (A), body condition (residual body weight, Weight^adj) and PC^{M main} (B), PC^{M main} and log-transformed Gata3 expression in nonspecifically stimulated splenocytes (Gata3^mit-stim) (C), and body condition and log-transformed Gata3^mit-stim (D). (E) Association between testis condition (residual testis weight, Testis^adj) and log-transformed Gata3^mit-stim in mature male voles. (F) Contour map representing log-transformed Gata3^mit-stim against variation in body condition and PC^{M main} in mature males. (G) Association between body condition and log-transformed Gata3^mit-stim in voles with overt TB or with no sign of TB (all stages). (A–E, G) Scatter of points represent actual observations (A) or (for graphical purposes) partial residuals (B–E, G) from LMMs; solid lines are model predictions averaged across all other terms in model; p values relate to main effects in LMM analyses conducted separately for mature or immature males (A, E) or Stage×Predictor interactions in LMMs containing different slopes for mature and immature males (B–D); 1 s.e. given above and below model predictions (dashed lines).

Figure 2. The dynamics of Gata3 expression in tolerance responses.

(A) Longitudinal study (Longit.). A history of exposure to ticks or fleas at previous sampling points was associated with higher levels of peripheral blood Gata3 (Gata3^blood) expression; LMM prediction with 1 s.e. above and below. (B) Longitudinal study. Positive association of Gata3 expression (Gata3^blood) in peripheral blood with weight increase (adjusted for starting weight) in the subsequent month. Scatter of points represents final weight residuals on starting weight; solid line is an LMM prediction averaged across all other terms in model; 1 s.e. given above and below model prediction (dashed lines). (C) A causal hypothesis refined using SEM of data for mature males in the cross-sectional study. Where the flow of causality (solid arrow directions) depicted in the set of relationships was supported by temporal analyses in the longitudinal study, this is indicated with an “L”. Details of analysis in Methods S1. The p values for respective parameters in the SEM are given alongside arrows (whose widths are proportional to the z score). Blue arrows indicate a positive effect and red arrows a negative effect. The arrow with a dashed margin indicates that age-specific effects on survival were additionally found for Gata3 expression (in blood) in separate analyses of the longitudinal data (not part of the SEM).

Figure 3. Association of Gata3 expression with survivorship.

Longitudinal study (Longit.). In both the full sample and in mature males, analyses of return rates indicated that higher Gata3^blood expression was associated with lower survival in smaller animals and higher survival in the largest animals. Predictions from LMMs (for details see Methods S1); 1 s.e. given above and below model predictions.