In the spring of 2018, we sampled 31 uninfected and 31 infected fish from Gosling Lake and 30 uninfected and 32 infected fish from Roselle Lake using minnow traps to quantify average tapeworm size and the frequency and severity of fibrosis. Fish were sampled as part of a larger study where we sampled the first 30 uninfected fish and then continued sampling until we had found 30 infected fish for each population. Fish were categorized as uninfected if we did not find a living tapeworm. We scored fibrosis in the peritoneal cavity visually using a dissecting microscope as: 0 (no fibrosis), 1 (some fibrosis, organs do not move freely), 2 (fibrosis adhering organs together), 3 (organs adhered together and to the peritoneal wall), 4 (severe fibrosis, difficult to open peritoneal cavity) (see supplemental video). We weighed tapeworms on a digital scale; tapeworms weighing less than 0.01g were recorded as <0.01g (limit of our field scale) and were entered as 0.009g for summary statistics. If fish were infected with multiple tapeworms, we weighed all tapeworms together to get average parasite mass. We compared infection intensity between lakes using a general linear model (glm) with a Poisson distribution, and average tapeworm mass using a glm with a gamma distribution and inverse link function. We also compared the number of tapeworms above and below the threshold of our field scale per lake using a chi-square test. We compared the fibrosis scores of uninfected and infected fish between lakes using Mann-Whitney U tests. To get an additional estimate of infection prevalence for Roselle, we euthanized and preserved 169 randomly selected fish in ethanol, which were later dissected and scored as infected or uninfected (ethanol preservation is not conducive to scoring fibrosis). In June 2018, we collected fish from our three populations for breeding. Using standard in-vitro fertilization methods, we created full-sibling families from each population and transported fertilized eggs to the lab for rearing (Divino & Schultz 2014). Fish were in two rooms at the animal care facility of the University of Connecticut. Families were often, though not always, split across multiple tanks located in both rooms. All fish were ~11 months old when they were injected with different immune challenges in May 2019. Laboratory Injection Experiment We injected four different inoculants directly into the peritoneal cavity. These included 1) 20mL of 1X phosphate buffered saline (PBS, control treatment), 2) 10mL of homogenized tapeworm protein solution + 10mL PBS (tapeworm treatment), 3) 10mL of Alum (2% Alumax Phosphate, OZ Bioscience) + 10mL PBS (alum treatment), and 4) 10mL tapeworm protein + 10mL Alum (tapeworm + alum treatment). Alum is an immune adjuvant that causes the recruitment of leukocytes that initiate an immune response (Kool et al. 2012). Pilot studies demonstrated that alum injections could induce a fibrosis response in the peritoneal cavity of stickleback (Dr. Natalie Steinel, per. comm). The tapeworm protein solution was used to test if fish could recognize and respond to tapeworm antigens. By using homogenized tapeworms, our experimental design mitigates active interference by the tapeworm, which are well known to secrete a suite of immunomodulatory molecules that suppress and shift host immune responses (Coakley et al. 2016; Maizels et al. 2018; Motran et al. 2018), including in stickleback (Scharsack et al. 2013). To create this solution we used tapeworms collected from Farwell Lake (50°11'60"N, 125°35'27"W) on in 2008 that were flash frozen and stored at -80°C. We chose tapeworms from a different lake, watershed, and year to minimize any localized genetic structure of the parasite that may influence population level responses. Tapeworms were dipped in deionized water and placed in chilled 0.9x PBS. Each tapeworm was sonified on ice twice for 1 min (Branson 150 Ultrasonic Cell Disruptor, level 5). Between sonification rounds, samples were chilled on ice (5 min). The homogenized solutions were centrifuged (4°C, 4000rpm, 20min) and the supernatant was collected and pooled. We measured the protein concentration using a Red 660 kit (G-Biosciences), diluted the solution to 1mg/ml using 0.9x PBS and stored it at -20°C. Before injection, fish were lightly anesthetized using a neutral-buffered MS-222 (50-75 mg/L). We used ultra-fine syringes (BD 31G 8mm) to inject 20ul into the lower left side of the peritoneal cavity, slightly above where the end of ventral spine rests. Injections were shallow and at an angle parallel to the body to avoid injuring organs. We watched for visual distention of the peritoneal cavity to ensure solutions were being injected correctly. Solutions were prepared and syringes were loaded in a sterile culture hood. Fish were also given a small colored elastomer mark (Northwest Marine Technologies) corresponding to their treatment group injected subcutaneously just posterior to the neurocranium. While injecting elastomers can impact immune responses (Henrich et al. 2014), all fish received the same amount of elastomer in the same location across treatment groups. During the injection procedure, fish were placed on a wet sponge and had their gills covered with a wet paper towel. In total, the procedure lasted less than one minute, and fish were immediately placed in an aerated recovery tank before being returned to their home tank with negligible mortality and no noticeable adverse effects. We euthanized fish to measure fibrosis post injection at four time points: 1, 10, 42, and 90 days. These timepoints were chosen to see 1) if fish could respond within 24 hours to an immune challenge, which is when the tapeworm penetrates the gut (Hammerschmidt & Kurtz 2007), 2) how their response changed through time as the tapeworm would be growing within the peritoneal cavity (10 and 42 days). The 10 and 42 day timepoints also correspond to timepoints used in previous experiments (Scharsack et al. 2007; Weber et al. 2017a, b; Fuess et al. 2021), and 3) how this response might change within the timeframe of a breeding season, around 90 days. We used the 0-4 fibrosis scale described above. Two people (AKH & LEF) blind to treatment and population scored fibrosis (each fish was scored once). We also recorded fish mass, length, and sex. To the best of our ability, we spread treatments and time points across families within each population; sample sizes are provided in table 1. Sample sizes are small for the 90 day timepoint as this was added to take advantage of excess surviving fish. Throughout the experiment, there was a mortality rate of 11% (48 out of 418 fish), which did not appear to be driven by treatment. Mortality typically occurred days to weeks after injection.

Closely related populations often differ in resistance to a given parasite, as measured by infection success or failure. Yet, the immunological mechanisms of these evolved differences are rarely specified. Does resistance evolve via changes to the host's ability to recognize that an infection exists, actuate an effective immune response, or attenuate that response? We tested whether each of these phases of the host response contributed to threespine sticklebacks' recently evolved resistance to their tapeworm Schistocephalus solidus. While marine stickleback and some susceptible lake fish permit fast-growing tapeworms, other lake populations are resistant and suppress tapeworm growth via a fibrosis response. We subjected lab-raised fish from three populations (susceptible marine 'ancestors', a susceptible lake population, a resistant lake population), to a novel immune challenge using an injection of: 1) a saline control, 2) alum, a generalized pro-inflammatory adjuvant that causes fibrosis, 3) a tapeworm protein extract, or 4) a combination of alum and tapeworm protein). With enough time, all three populations generated a robust fibrosis response to the alum treatments. Yet, only the resistant population exhibited a fibrosis response to the tapeworm protein alone. Thus, these populations differed in their ability to respond to the tapeworm protein but shared an intact fibrosis pathway. The resistant population also initiated fibrosis faster in response to alum, and was able to attenuate fibrosis, unlike the susceptible populations' slow but longer-lasting response to alum. As fibrosis has pathological side-effects that reduce fecundity, the faster recovery by the resistant population may reflect an adaptation to mitigate the costs of immunity. Broadly, our results confirm that parasite detection and immune initiation, activation speed, and immune attenuation simultaneously contribute to the evolution of parasite resistance and adaptations to infection in natural populations.

Included in this respository is all code used for the manuscript and all datafiles. For more information about fibrosis scores in fish, please see manuscript and accompanying video. All analysis were done in R. Please contact Amanda Hund with additional questions (ahund@umn.edu). Funding provided by: James S Mcdonnell FoundationCrossref Funder Registry ID: http://dx.doi.org/10.13039/100000002Award Number: Postdoctoral Fellowship to HundFunding provided by: American Association of ImmunologistsCrossref Funder Registry ID: http://dx.doi.org/10.13039/100002570Award Number: Intersect Postdoctoral Fellowship to FuessFunding provided by: University of ConnecticutCrossref Funder Registry ID: http://dx.doi.org/10.13039/100007710Award Number: Startup funds to BolnickFunding provided by: National Institutes of HealthCrossref Funder Registry ID: http://dx.doi.org/10.13039/100000002Award Number: NIAID Grant 1R01AI123659-01A1