Discussion
Our study experimentally evolved pesticide resistance in flour beetles
(T. castaneum ) to investigate the combined interactive effects of
pesticide resistance and pesticide exposure on immunity and
host-pathogen interactions. Pesticide-resistant insects and insects
exposed to Btt both developed more slowly than their
counterparts, indicating sub-lethal fitness costs to resistance and
infection, respectively. We found that Pyr-resistant insects exhibited
dampened transcriptional immune responses to Btt infection
relative to their control-regime counterparts, although overall survival
and development did not differ among OP- and Pyr-resistant and
susceptible insects after Btt exposure. While the evolution of
resistance did not substantially influence host infection phenotypes,
exposure to pesticides did. Specifically, we observed that larvae were
more likely to survive OP exposure if they were also orally exposed toBtt , with gene expression patterns suggesting dampened
transcriptional stress responses in those beetles. Taken together, our
data suggest that evolved resistance, pesticide exposure, and bacterial
infection all exert contrasting effects on life history parameters that
could ultimately influence the demography of insect populations in
complex ways. The factorial nature of our study design, moreover,
provides a new window into the physiological basis and fitness
consequences of host responses to overlapping environmental stressors.
Insects, including T. castaneum , can rapidly evolve resistance to
pesticides through mechanisms that enhance metabolic detoxification and
cuticular modifications . Here, populations selected for pesticide
resistance evolved it within 6-8 generations , demonstrating increased
survival and faster development after pesticide exposure compared to
control-selected populations. Differential gene expression between
pesticide- and control- selection regimes suggested that OP resistance
is associated with constitutive changes in cuticle gene expression while
Pyr resistance is associated with oxidoreductase activity, heme binding,
and cuticular modifications, consistent with previous studies in OP and
Pyr resistant insects . Thus, while both OP and Pyr resistance involved
changes in cuticle gene expression, Pyr resistance was also associated
with changes in metabolic detoxification.
Oral Btt infection did not decrease survival in our study but was
associated with slower development in T. castaneum larvae,
indicating a direct cost of infection or an indirect cost from mounting
an immune response, as previously observed in the same system and in
other insects challenged with Bt toxins . Previously documented
variation in resistance in T. castaneum against Cry3 proteins
effective against other Tenebrionid beetles may explain why Btthad negative effects on development but did not instigate high mortality
. Here, exposure to Btt induced the upregulation of AMPs
(attacins and defensins) and one pathogenesis-related protein as well as
differential expression of detoxification genes, including P450s and a
CEH, GST, an ABC transporter, and several cuticle and development
related genes. Previous transcriptomic studies of oral Bt infection inT. castaneum have identified similar transcripts, including
attacin and defensin AMPs, a pathogenesis-related protein, a P450, and a
GST , supporting the assumption that Btt oral exposure in our
experiments presented an immune challenge despite the lack of
infection-induced mortality.
Previous work has suggested that exposure to pesticides can reduce
defenses against pathogens and stimulate an array of changes in humoral,
cellular, and oxidative stress components of immunity . Our results
reveal a strong interaction effect between Btt and OP exposure
that manifests at both phenotype and gene expression levels. For
example, dual oral exposure to Btt and OP improved larval
survival and partially rescued development time relative to OP exposure
alone, without exacerbating Btt -induced mortality. At the same
time, Btt co-exposure modified the differential expression of 175
genes relative to OP exposure alone (Fig. 3B). In contrast, Pyr exposure
did not influence survival with Btt infection, but larvae exposed
to both Pyr and Btt also partially rescued development time
compared to those exposed only to pesticides. However, in contrast to
the effects on gene expression observed with dual OP and Bttexposure, Btt co-exposure with Pyr had negligible effects on gene
expression with the overexpression of only one cuticle-related
transcript. It is possible that the positive effects on development time
seen with pesticide and Btt co-exposure compared to single
exposure to pesticides are part of an advantageous response to increased
risk (Roth & Kurtz, 2008) as has been seen in virus-infected
caterpillars reared on more toxic diets .
There is also the potential for indirect antagonistic interactions
between OP and Btt , e.g. through induction of cross-protective
host defenses. For example, while Btt exposure induced
upregulation of attacins, cytochrome P450s, and cuticle-related
transcripts, combined OP and Btt exposure dampened the expression
of these transcripts. It is possible that this dampened expression of
immune and detoxification genes may underlie reductions in
immunopathology or ROS damage corresponding to increased fitness, or the
benefits to fitness may be due to differences in resource allocation
when faced with dual pressures . Interestingly, single exposure to OP orBtt induced downregulation of an ABC transporter but this
transcript was upregulated with co-exposure to OP and Btt . ABC
transporters have been identified as receptors for Bt Cry proteins and
aid in membrane pore formation , and the additional function of ABC
transporters as xenobiotic pumps may help to mediate Bt toxicity . It is
also possible that OP and Btt interact directly, as would be the
case if Btt actively degrades OP for energy or if OP facilitatesBtt germination rates. Our in vitro data suggest thatBtt vegetative cells in liquid medium replicate faster in the
presence of OP, lending credence to a direct interaction. On the other
hand, there was no difference in germinated spore counts between
control-diet and OP-diet disks, suggesting that direct interactions, if
they matter in vivo , are not relevant until after the host
ingests the diet.
Supporting the idea of localized cross-protective mechanisms, a previous
study of oral vs. septic Btt infection in T. castaneumdemonstrated that oral infection induced greater changes in
cuticle-related gene expression compared to septic infection . In our
study, OP-exposed individuals were not protected from greater mortality
upon septic Btt infection, suggesting that infection route has
important implications for pesticide-pathogen interactions. We found
cuticle-associated gene expression changes induced by OP exposure, and
cuticular modifications of gut tissues may provide an additional layer
of protection from oral Bt infection . Future work dissecting the
relative importance of these direct and indirect interactions will
facilitate our understanding of organismal adaptations to multiple
stressors and will be crucial for refining integrative pest management
(IPM) strategies employing both chemical and biopesticides.
The impact of evolved pesticide resistance on immunity is still poorly
understood , and in the absence of pesticides we did not observe any
significant interactions between pesticide resistance and Bt exposure on
survival or development. Few genes were differentially expressed between
OP resistant and susceptible beetles after Btt exposure, butBtt exposure in Pyr-resistant populations was associated with
dampened upregulation of several genes, including two defensins, a
histidine-rich glycoprotein, and a pathogenesis-related protein, that
were more strongly upregulated in Pyr-susceptible populations afterBtt exposure. It is also possible that Pyr-resistant populations
had constitutively higher immune gene expression resulting in reduced
immune activation upon Bt infection. Moreover, an apolipophorin-III
transcript, previously shown to be important in immunity against
coleopteran-specific Bt toxins , was upregulated in Pyr-resistant
populations. In contrast, cuticle-related protein and neuropeptide
transcripts were downregulated in Pyr-regime larvae exposed toBtt . These immune gene changes associated with Pyr resistance may
have important effects on interactions with other pathogens, and future
research into Pyr resistance interactions with other bacterial or viral
pathogens may yield important insights.
While we did not observe any phenotypic differences with pesticide
selection regime and Btt infection here, we also did not observe
mortality from Btt alone and it is possible that the immune
challenge was not strong enough to elicit an interaction with resistance
mechanisms. However, the differential expression of immune and cuticle
functions in Pyr-regime larvae exposed to Btt could have
important implications for insect interactions with other pathogens or
xenobiotic pressures by modifying detoxification capabilities or
providing an additional barrier against infection or toxins .
Furthermore, changes in oxidoreductase activity observed after Pyr
selection are likely to alter the status of reactive oxygen species
(ROS) within the host and may impact immunity against different
pathogens . Finally, it is possible that adaptation to pesticides could
constrain host evolutionary responses to pathogens, or vice versa.
Future experiments could simultaneously select for dual resistance to
both pesticides and pathogens to see if evolutionary trajectories are
similar or constrained relative to selection under a single stressor.