Immune responses evolve to balance the benefits of microbial killing against the costs of autoimmunity and energetic resource use. Models that explore the evolution of optimal immune responses generally include a term for constitutive immunity, or the level of immunological investment prior to microbial exposure, and for inducible immunity, or investment in immune function after microbial challenge. However, studies rarely consider the functional form of inducible immune responses with respect to microbial density, despite the theoretical dependence of immune system evolution on microbe‐ versus immune‐mediated damage to the host. In this study, we analyse antimicrobial peptide (AMP) gene expression from seven wild‐caught flour beetle populations (Tribolium spp.) during acute infection with the virulent bacteria Bacillus thuringiensis (Bt) and Photorhabdus luminescens (P.lum) to demonstrate that inducible immune responses mediated by the humoral IMD pathway exhibit natural variation in both microbe density‐dependent and independent temporal dynamics. Beetle populations that exhibited greater AMP expression sensitivity to Bt density were also more likely to die from infection, while populations that exhibited higher microbe density‐independent AMP expression were more likely to survive P. luminescens infection. Reduction in pathway signalling efficiency through RNAi‐mediated knockdown of the imd gene reduced the magnitude of both microbe‐independent and dependent responses and reduced host resistance to Bt growth, but had no net effect on host survival. This study provides a framework for understanding natural variation in the flexibility of investment in inducible immune responses and should inform theory on the contribution of nonequilibrium host‐microbe dynamics to immune system evolution.

21Immune responses evolve to balance the benefits of microbial killing against the costs of autoimmunity 22 and energetic resource use. Models that explore the evolution of optimal immune responses generally 23 include a term for constitutive immunity, or the level of immunological investment prior to microbial 24 exposure, and for inducible immunity, or investment in immune function after microbial challenge. 25However, studies rarely consider the functional form of inducible immune responses with respect to 26 microbial density, despite the theoretical dependence of immune system evolution on microbe-versus 27 immune-mediated damage to the host. In this study, we analyze antimicrobial peptide (AMP) gene 28 expression from seven wild-caught flour beetle populations (Tribolium spp.) during acute infection with 29 the virulent bacteria Bacillus thuringiensis (Bt) and Photorhabdus luminescens (P.lum) to demonstrate 30 that inducible immune responses mediated by the humoral IMD pathway exhibit natural variation in both 31 microbe density-dependent and independent temporal dynamics. Beetle populations that exhibited greater 32 AMP expression sensitivity to Bt density were also more likely to die from infection, while populations 33 that exhibited higher microbe density-independent AMP expression were more likely to survive P. 34 luminescens infection. Reduction in pathway signaling efficiency through RNAi-mediated knockdown of 35 the imd gene reduced the magnitude of both microbe-independent and dependent responses and reduced 36 host resistance to Bt growth, but had no net effect on host survival. This study provides a framework for 37 understanding natural variation in the flexibility of investment in inducible immune responses and should 38 inform theory on the contribution of non-equilibrium host-microbe dynamics to immune system 39 evolution. 40 41 Keywords: ecological immunology, infection tolerance, life history trade-offs, host-parasite interactions, 42 innate immunity, Tribolium castaneum, Bacillus thuringiensis 43 44 Data Accessibility 595The RT-qPCR data and metadata will be deposited into Data Dryad upon manuscript acceptance. 596

1. In nature, hosts face shifting patterns of parasite exposure and life history trade-offs, as they develop from birth to old age. As a result, the net fitness benefit of immunological investment can change dramatically from one life stage to the next.2. Previous work has revealed a puzzling diversity of relative immune investment patterns among juvenile and adult stages, and it is not clear whether lessons learned from one particular population or species can be generalised to wild populations, after accounting for local adaptation and other variance-generating processes.3. In this study, we quantify larval and adult immune gene expression and resistance to bacterial infection in two flour beetle species (Tribolium castaneum and Tribolium confusum) from seven wild-derived populations.4. Our results provide a clear signal of higher infection-induced immunological investment and resistance in adults relative to larvae, despite variation among species in immune gene regulation.5. Better characterisation of stage-specific investment in infection resistance in natural populations can inform our understanding of life history evolution and improve predictions of disease dynamics in the wild.

In nature, hosts face shifting patterns of parasite exposure and life history trade-offs as they develop from birth to old age. As a result, the net fitness benefit of immunological investment can change dramatically from one life stage to the next. Previous work has revealed a puzzling diversity of relative immune investment patterns among juvenile and adult stages, and it is not clear whether lessons learned from one particular population or species can be generalized to wild populations, after accounting for local adaptation and other variance-generating processes. In this study, we quantify larval and adult immune gene expression and resistance to bacterial infection in two flour beetle species (Tribolium castaneum and T. confusum) from two lab-adapted and five wild-derived populations. Our results provide a clear signal of higher infection-induced immunological investment and resistance in adults relative to larvae, despite variation among species in immune gene regulation. Better characterization of stage-specific investment in infection resistance in natural populations can inform our understanding of life history evolution and improve predictions of disease dynamics in the wild.