Introduction
Insect populations commonly face exposure to pesticides applied to target them or another species in their environment, leading to a high prevalence of pesticide resistance among insect species . At the same time, insects are frequently infected with a variety of parasites and pathogens, including bacteria, fungi, and viruses, and these agents can complement or even replace the use of chemical pesticides . The physiological effects of chemical pesticides and pathogens are not independent, however, as both can activate stress, detoxification, and immune responses . Since the evolution of pesticide resistance also often arises through these mechanisms , host responses to chemical and microbial control agents could mediate facilitation or antagonism among stressors on both proximate and evolutionary time scales . A better understanding of the molecular basis and phenotypic outcomes of these interactions would provide fundamental insight into the integration and evolution of organismal stress and immune responses while also improving the design and predictive power of pest and vector control strategies .
Pesticides target a diverse set of physiological functions in insects ranging from neurotoxic activity to the regulation of growth and development. The mechanisms of pesticide resistance show a similarly diverse set of solutions within and among pesticide classes, including target-site modifications, increased metabolic detoxification, and cuticular modifications . Organophosphates (OP), which inhibit acetylcholinesterase (AChE) to overexcite cholinergic synapses , and pyrethroids (Pyr), which disrupt voltage-gated sodium channel (vgsc) function , are two pesticide classes widely used in agricultural systems and against disease vectors and also impact important honeybee and pollinator populations (Berenbaum & Liao, 2019; Schuhmann et al. 2022). Target-site mutations and AChE gene duplications have been described for several OP-resistant insects , while Pyr resistance has been associated with mutations in the voltage-gated sodium channel . Target site mutations are not the only path to resistance, however. OP and Pyr pesticides are mainly detoxified through oxidation and hydrolysis, and resistance associated with differential expression of diverse canonical detoxification genes, including cytochrome P450s, esterases, and glutathione S-transferases (GSTs), has also been described for both pesticide types . Moreover, recent studies have implicated changes in cuticle and serine endopeptidase gene expression in increased resistance to penetration by OP and Pyr pesticides .
Recent evidence suggests that the mechanisms of pesticide resistance could also impact immune and physiological responses against parasites . For example, esterase-mediated pesticide resistance in Culex pipiens is associated with changes in immune gene expression including constitutive upregulation of antimicrobial peptide (AMP) and nitric oxide synthase (NOS) genes . Moreover, resistance-associated constitutive changes in metabolic detoxification mechanisms like cytochrome P450s or GSTs can alter concentrations of damaging reactive oxygen species (ROS) that a pathogen would encounter within the host and influence the success of pathogen colonization, growth, and transmission . For example, changes in cap‘n’collar transcription factor expression have been shown to alter detoxification gene expression and increase ROS levels, thereby conferring pesticide resistance in several species while also modifying vector competence in Aedes aegypti . Populations of OP and Pyr resistant mosquito strains in possession of target-site mutant ace-1 (AchE) and kdr (vgsc) alleles, respectively, support a higher prevalence of Plasmodium falciparum parasites, but kdr mutations were also associated with reduced midgut oocyst burden in infected individuals .
Even in the absence of evolved resistance, host exposure to pesticides may impact pathogen growth directly through contact with toxins or indirectly through the induction of insect detoxification enzymes . Exposure to pesticides can also have complex effects on components of the cellular, humoral, and oxidative stress responses of host immunity . For example, exposure to OP pesticides has been associated with increased hemocyte numbers and phenoloxidase (PO) and encapsulation activity in wax moth (Galleria mellonella ) and Colorado potato beetle (Leptinotarsa decemlineata ) larvae . However, dual exposure to OP and a pathogenic virus in silkworm larvae (Bombyx mori ) resulted in differential expression of oxidative stress and AMP genes and increased mortality . Exposure to Pyr pesticides, meanwhile, has been associated with increased melanization responses and decreased replication of Escherichia coli bacteria and decreased P. falciparum infection prevalence and intensity in A. gambiae . Pyr exposure is also hypothesized to impact the production of other immune responses including serine proteases, lytic enzymes (esterases and lysozymes), and reactive oxidative stress responses . Moreover, exposure to neurotoxic pesticides (e.g., neonicotinoids, butenolides) in bees has been associated with reduced PO activity (Czerwinski and Sadd, 2017) and increased viral loads (Harwood et al., 2022). However, the combined effects of multiple stressors on bee mortality are variable (Calhoun et al., 2021; Harwood & Dolezal, 2020).
Clearly, both pesticide resistance and exposure independently have important effects on insect immunity against pathogens. However, it remains an open question whether resistance and exposure influence host-microbe interactions in the same direction, and whether this influence arises through the same or different mechanisms. To address this gap in knowledge, we experimentally evolved resistance to two different classes of pesticides (OP and Pyr) in the red flour beetle (Tribolium castaneum ), an emerging model for studies on insect genomics, immunity, and resistance . As a stored-product pest, T. castaneum may be exposed directly or indirectly to the pesticides used to combat a variety of pests that co-inhabit stored grain facilities and that impose selection for resistance.
To investigate the interactive effects of pesticide resistance, exposure, and infection, we exposed our evolved lines to Bacillus thuringiensis (Bt), an entomopathogenic Gram-positive bacterium that has been developed into a widely used biopesticide against many insect species . While much of the research on host-Bt interactions focuses on the lethal biocontrol aspects, natural strains also occur in the environment and vary in lethality, thereby representing a selective force within natural populations . Insect resistance to Bt commonly results from changes in specific toxin-receptor interactions , providing little expectation of cross-resistance with chemical pesticides (Siegwart et al., 2015). However, Bt-resistance has been associated with increased susceptibility to bacteria-derived pesticides , and exposure to Bt has been shown to increase susceptibility to viruses and entomopathogenic nematodes . The immune responses of susceptibleT. castaneum to oral Bt infection involve significant upregulation of a suite of immune, stress, and developmental genes that potentially overlap with the response to chemical pesticides and the mechanisms of pesticide resistance. However, the mode of infection may have important implications for interactions with pesticide resistance, as T. castaneum exhibited contrasting patterns of expression of infection-related genes dependent on the route of infection,i.e., oral compared to septic infection .
To explore the complex interactions between pesticide resistance and exposure to pesticides and pathogens, we first investigated the main and interactive effects of selection regime (pesticide resistant and susceptible) and pesticide exposure on host fitness-associated phenotypes after Bt infection. Having established these phenotypes, we turned to transcriptional data to identify potential mechanisms that could explain the observed phenotypes. We first asked whether evolution regime, i.e. , evolved pesticide resistance, alone influences constitutive gene expression and the transcriptional response to infection in the absence of pesticides. We next included pesticide exposure into the regime-by-infection interaction to investigate whether pesticides facilitate or antagonize the host response to infection, and whether differential gene expression depends on the experimental evolution regime. For both investigations, we compared the results of OP and Pyr treatments to determine whether evolved resistance or exposure to pesticides with different physiological targets exert different effects on host-pathogen interactions. Our study provides a comprehensive window into the physiological and evolutionary processes that shape interactions among two important ecological stressors.