We studied the potential of combining insect immune priming with the introduction of diverse migrants to safeguard individuals from an inbred population from disease as a technique for enhancing genetic rescue efforts.

Immune priming in insects refers to the stronger immune response insects have against pathogens following exposure. This enhanced immunity can be passed on to offspring and holds promise for insect conservation efforts against diseases.

We compared the fitness benefits of individuals from a small, inbred population with two treatments: the addition of genetically diverse migrants that had not been primed and the addition of immune-primed migrants. While both types of migrants enhanced reproduction, as in cases of genetic rescue, only primed migrants led to improved survival on exposure to a pathogen.

Better immunity led to a trade-off with reproduction in immune-primed migrants, but this was not evident upon outcrossing with the target individuals, revealing synergies between hybrid vigor and immune priming.

Given the demographic constraints and stochasticity that can exacerbate the effects of disease outbreaks in small populations, our results serve as a proof of concept for combining immune priming with assisted migration to offer a proactive strategy to mitigate disease impacts while enhancing genetic diversity.

Study system
Our populations were derived from a genetically diverse stock comprising five lineages of T. castaneum (Durkee et al. 2024). The stock population, which we used for the initial experiments and migrants, was maintained at 2000 individuals housed in replicate enclosures 4cm × 4cm × 6cm containing 30g of medium (95% wheat flour, 5% brewer’s yeast), at 31°C and 54 ± 10% relative humidity. For the proof-of-concept rescue experiment, we used a single bottlenecked population, as detailed below. We chose not to include multiple distinct bottlenecked populations, as our primary objective was to assess individual-level fitness and immune responses. 

We used females in our experimental priming treatments (“maternal priming”). Bacillus thuringiensis culture for priming (and later test infections) was prepared from a −80°C glycerol stock (strain: tenebrionis, procured from Bacillus Genetic Stock Center, Ohio), cultured overnight in nutrient broth at 30°C until reaching an optical density of 0.95 at 600nm. After centrifugation, the pellet was resuspended in 1mL PBS. Bacteria were either heat-killed for priming or used live for larval infections. Heat-killed bacteria were adjusted to 1011cells/mL of PBS for priming and 107 cells/mL of PBS for live challenge. We ensured this step by plating the heat-killed bacteria on agar plates and incubating them overnight for confirmation.

For priming, adult females were chilled on ice for ten minutes to reduce movement. Under a microscope, each female was pricked between the head and thorax using a 0.1-mm pin dipped in a heat-killed bacterial solution. Sham-primed individuals were pricked in the same location but with a pin dipped in PBS, to serve as a control for the handling and slight injury. Pricked females were allowed 48 hours to heal before use in experiments.

Initial experiments to develop foundational data:

In this set of experiments, we used individuals from genetically diverse stock populations. We confirmed successful immune priming by measuring the survival of primed individuals and their offspring against disease, we examined potential immunological mechanism of priming, and we evaluated the effects of priming on offspring performance to measure potential trade-offs. Overall, we addressed four questions (Q1-Q4).

(Q1) Does immune priming enhance survival against disease in primed individuals?

To investigate the effects of immune priming on survival following pathogen challenge, we subjected female pupae to three treatments: control (unprimed), sham-primed, and primed. We then evaluated their survival after exposure to Bacillus thuringiensis (Bt).

To rear pupae for priming, we established 30 replicate groups of 50 adult beetles from the diverse stock population. Each group was placed in a 4 × 4 × 6 cm enclosure (hereafter, a patch) with standard diet. These beetles were allowed to mate and oviposit for 24 hours, after which we sieved out the adults, retained the eggs in the diet, and allowed them to develop. When they were at the pupal stage, females were separated out (by using genital papillae as a marker) for use in the priming experiment. These female pupae were given an ID based on their patch of origin and placed individually in wells of a 96-well plate containing standard diet. The females developed into adults, and they were randomly assigned to one of the three treatments: a) unprimed control, b) sham-primed, and c) primed. Females in the primed treatment received immune priming using the protocol described above. Each treatment was applied to 10 females per patch, for a total of 300 adult females per treatment (Figure 1). Ten days after priming, adults from all three treatments were challenged with live Bacillus thuringiensis (Bt) with a dose adjusted to 107 cells/mL of PBS. Individual survival was recorded on day seven, which corresponds to Bt clearance time. 

We evaluated female survival following pathogen exposure using a generalized linear mixed-effects model (GLMM) with a binomial error distribution and logit link. Treatment was included as a fixed effect, while patch was incorporated as a random effect to account for potential variation among patches. We used ‘emmeans’ for pairwise comparisons.

(Q2). Does maternal priming enhance offspring survival against disease?

To test for benefits of maternal priming to offspring (called trans-generational immune priming) we applied the same three treatments (unprimed control, sham-primed, and primed) to a second set of 300 unmated females and placed them individually in wells of 96-well plates with diet. These females were not challenged with pathogens during their lifetime. Forty-eight hours following priming, one male was added to each well, and the pairs were allowed to mate and oviposit for a total of 48 hours. After mating, the adults were removed, and the number of larvae produced by each female was counted to measure reproductive output. Each larva was assigned unique ID based on their maternal origin. These larvae were used to address this research question (Q2) regarding trans-generational effects, as well as the following two questions (Q3 and Q4) as described below. To evaluate transgenerational effects of maternal priming (Q2), we compared survival of larvae from each maternal treatment following pathogen exposure (nunprimed = 100, nsham-primed =101, nprimed = 115). 
We validated trans-generational immune priming by analyzing larval survival following pathogen exposure using a generalized linear mixed model with a logit link function. The model included treatment applied to mothers as a fixed effect and patch ID as a random effect. Maternal identity could not be included, as only one larva per female was used to evaluate survival.

(Q3). Does maternal priming alter production of hemocytes and melanization in offspring?

We measured hemocytes and the production of melanin following previously established protocols (Ghosh et al. 2023a, b), focusing on larvae from primed and unprimed mothers (n=20 and 48 larvae, respectively, for hemocytes and n=20 for both treatments for melanization). Briefly, larvae were incised laterally, and hemolymph was collected into pre-chilled Eppendorf tubes. To count hemocytes, hemolymph was diluted 1:2 with PBS, and 5 μL of this mixture was loaded onto a Neubauer hemocytometer, then counted under a microscope (200x magnification) after 15 minutes (n= 50/treatment). To assess production of melanin, we measured phenoloxidase activity by centrifuging hemolymph (1:2 with ice-cold PBS), adding the supernatant to a microplate well with 2 mM L-Dopa (substrate), and recording melanization spectrophotometrically at 490 nm after 30 minutes. Immunological parameters were analyzed using a linear mixed model, with hemocyte count or degree of melanization as the dependent variable, treatment (applied to mothers) as a fixed factor, and patch of maternal origin as a random effect.

To explore mechanisms underlying the protective effects of maternal priming that we report below, we also evaluated if the number of hemocytes or degree of melanization in siblings could explain survival following pathogen exposure in generalized linear mixed models with survival as the dependent variable, and either sibling hemocytes or melanization as predictors, including patch of maternal origin as a random effect.

(Q4). Does maternal priming lead to trade-offs in offspring development time, weight and reproduction in the absence of disease?

To evaluate if maternal immune priming leads to trade-offs for their offspring, we examined three traits, offspring development time (egg-pupa, nunprimed = 49, nsham-primed = 30, nprimed = 41), offspring weight on eclosion to adulthood (nunprimed = 21, nsham-primed = 49, nprimed = 49), and offspring reproduction. To measure reproduction, we allowed a subset of the offspring from treated mothers to develop to the adult stage and used them to set up single mating pairs (20-25 pairs/treatment) in plastic vials containing 5g of flour. After a 48-hour oviposition period, we sieved the mating pairs out. Offspring from each pair were counted on day 35.

We analyzed adult body weight, development time, and reproduction data using a linear mixed model. In this model, we considered maternal treatment as our fixed effect and patch of maternal origin as a random effect. 
Rescue Experiment: Integrating primed migrants into genetic rescue. In this section, we addressed two major questions- 

(Q5) In the context of genetic rescue, can immune priming of migrants protect the following generation from disease? And (Q6) are there associated benefits or trade-offs with maternal priming in terms of development time or reproduction in the following generation.

Creation of the target groups and migrants for the rescue experiment
The target population was one of the populations from Durkee et al. (2024). Briefly, the target population was founded with 50 individuals, thus imposing a mild demographic bottleneck. It then was reared in a challenging environment that had deltamethrin, an insecticide added to the medium for eight generations. This event simulated a sudden environmental change and imposed a selective bottleneck. At the end of the Durkee et al. (2024) experiment, the population was maintained on standard diet (without pesticide) for seven additional generations during which population size was at least 500 individuals. This history of a demographic bottleneck and strong selection, with the environment changing to more than less challenging conditions is not unlike what populations experience in nature and provides a target population that should have reduced genetic variation. More importantly, this setup simulates real-world scenarios, such as declining populations in captivity or degraded habitats, where assisted migration may play a critical role in boosting population fitness and genetic diversity.
Our migrants were derived from the genetically diverse stock population described above. In preparation for our experiment, an independent population was initiated with 50 individuals in 4 × 4 × 6-cm enclosures partially filled with 30 mL of flour and yeast medium and allowed to grow to a size of 500 individuals over the course of several generations. This represented an isolated healthy population (monitored though stable positive growth rate). We used females from this population to create primed migrants as described above, with additional females as unprimed migrants.

In our genetic rescue experiment, we added these migrants to groups of target individuals. Because the target individuals were from a population that had passed through a bottleneck, outcrossing with migrants from a genetically diverse population is expected to increase fitness by masking deleterious mutations that lead to genetic load, as is typical in efforts at genetic rescue (Whiteley et al. 2015). Our goal was then to examine whether immune priming could provide further benefits to the following generation, through enhancing survival in the presence of a pathogen. As such, we implemented three treatments: (a) groups of target individuals without any migrants (no migrants), serving as the control, (b), groups of target individuals with migrants that had not been immune primed (unprimed migrants), and (c) groups of target individuals with primed migrants. 

Replicate control groups were initiated with 50 beetles from the target population stock. Groups receiving migrants (of either type) were initiated with 40 individuals from the target stock, to which 10 unmated migrant females were added (20% of the population) (n= 50). Maintaining group size at 50 ensures that the effects of migrants on the following generation are primarily genetic rather than demographic (Hufbauer et al. 2015). Although the ideal proportion of migrants used by managers may depend upon population size, existing genetic diversity, and the specific objectives of the initiative, typically, genetic rescue endeavors, including translocation programs, incorporate an introduction of migrants ranging from 10% to 20% of the recipient population (Hogg et al. 2006; Whiteley et al. 2015). 

These experimental groups were given 48 hours for mating and oviposition, after which adults were removed. Seventeen days post-oviposition, two larvae were randomly collected from each patch and subdivided into two groups to assess the following: 1) larval survival following exposure to the Bt pathogen, and 2) development duration (n = 50 per treatment) (Q5). The remaining larvae were allowed to develop into adults, and the average offspring number was calculated (Q6), accounting for the larvae used for Q5 assessments.

We compared larval survival following pathogen exposure between the control groups, and the groups that received primed migrants and unprimed migrants using a generalized linear mixed model with a binomial error distribution. Data on development duration and offspring production were analyzed similarly. All data were tested for normality (Shapiro-Wilk) and homogeneity of variances (Levene’s test), and all statistical analyses were performed using R version 4.2.2 (2014) and IBM SPSS (2024). Graphical abstract and visual designs were created using BioRender (License: Ghosh, E. 2025).