https://cpw.cvlcollections.org/files/original/cbc41f815c968dc86a89be7c09e93664.pdf f10996908c5f30d2440df41244edc127 PDF Text Text | Revised: 28 July 2020 | Accepted: 25 August 2020 DOI: 10.1111/eva.13127 ORIGINAL ARTICLE Sarcoptic mange severity is associated with reduced genomic variation and evidence of selection in Yellowstone National Park wolves (Canis lupus) Alexandra L. DeCandia1 | Edward C. Schrom1 Daniel R. Stahler3 | Bridgett M. vonHoldt1 | Ellen E. Brandell2 | 1 Ecology & Evolutionary Biology, Princeton University, Princeton, NJ, USA Abstract 2 Population genetic theory posits that molecular variation buffers against disease risk. Biology, Pennsylvania State University, State College, PA, USA 3 Yellowstone Center for Resources, Yellowstone National Park, WY, USA Correspondence Alexandra L. DeCandia, Ecology & Evolutionary Biology, Princeton University, Princeton, NJ 08544, USA. Email: alexandradecandia@gmail.com Funding information National Science Foundation, Grant/Award Number: DEB-0613730, DEB-1245373 and DGE1656466; Yellowstone Forever; Princeton University Department of Ecology and Evolutionary Biology; Princeton University Center for Health and Wellbeing Although this “monoculture effect” is well supported in agricultural settings, its applicability to wildlife populations remains in question. In the present study, we examined the genomics underlying individual-level disease severity and population-level consequences of sarcoptic mange infection in a wild population of canids. Using gray wolves (Canis lupus) reintroduced to Yellowstone National Park (YNP) as our focal system, we leveraged 25 years of observational data and biobanked blood and tissue to genotype 76,859 loci in over 400 wolves. At the individual level, we reported an inverse relationship between host genomic variation and infection severity. We additionally identified 410 loci significantly associated with mange severity, with annotations related to inflammation, immunity, and skin barrier integrity and disorders. We contextualized results within environmental, demographic, and behavioral variables, and confirmed that genetic variation was predictive of infection severity. At the population level, we reported decreased genome-wide variation since the initial gray wolf reintroduction event and identified evidence of selection acting against alleles associated with mange infection severity. We concluded that genomic variation plays an important role in disease severity in YNP wolves. This role scales from individual to population levels, and includes patterns of genome-wide variation in support of the monoculture effect and specific loci associated with the complex mange phenotype. Results yielded system-specific insights, while also highlighting the relevance of genomic analyses to wildlife disease ecology, evolution, and conservation. KEYWORDS ectoparasite, genetics, infection severity, mite infestations, natural selection, RADsequencing, sarcoptic mange, wildlife disease This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2020 The Authors. Evolutionary Applications published by John Wiley & Sons Ltd Evolutionary Applications. 2021;14:429–445. � wileyonlinelibrary.com/journal/eva | 429 17524571, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eva.13127 by Colorado Parks & Wildlife, Wiley Online Library on [13/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Received: 3 June 2020 �| A classic paradigm in population genetics states that molecular diversity buffers against disease risk (Spielman, Brook, Briscoe, & Frankham, 2004). Host variation is thought to confer multiple defense strategies, thus limiting a pathogen's ability to exploit common weaknesses at the individual and population levels (Bergstrom & Antia, 2006; Hedrick, 1999). Conversely, the absence of host varia- Annual Wolf Count (Dec.) 1 | I NTRO D U C TI O N DECANDIA et al. Mange Invasion 200 Max. Mange Burden CDV Outbreak 150 100 50 0 1995 tion is expected to increase a population's vulnerability to infection, leading to disease outbreaks. This phenomenon is well supported within agricultural settings (Reiss & Drinkwater, 2018) and has been termed the “monoculture effect” (Elton, 1958). However, the universality of this trend beyond the agricultural realm remains uncertain. Elucidating the relationship between host genomic variation and 2000 2005 2010 2015 2020 Year F I G U R E 1 Annual wolf counts recorded in December 1995 through 2019 with years of CDV outbreaks, mange invasion, and maximum mange burden indicated (figure adapted from Almberg et al., 2012). Large circles represent large-scale CDV outbreaks, with smaller circles indicative of smaller outbreaks wildlife disease remains a prominent goal of molecular and disease ecologies (Blanchong, Robinson, Samuel, & Foster, 2016; DeCandia, light burden gave way to high exposure of canine adenovirus type-1, Dobson, & vonHoldt, 2018). This is particularly important for small, canine parvovirus, canine herpesvirus, and the protozoan Neospora fragmented, or reintroduced populations, where genetic diversity caninum (Almberg, Mech, Smith, Sheldon, & Crabtree, 2009). These loss may reduce evolutionary potential and threaten long-term vi- diseases were considered enzootic in the park's canids, and none ap- ability (Frankham, 2005; Spielman, Brook, & Frankham, 2004). peared to negatively impact individual fitness or population viability. Regarding disease, the inability to cope with novel or enduring Conversely, canine distemper virus (CDV) and sarcoptic mange parasites can lead to increased morbidity among individuals, and have been associated with morbidity, mortality, and reduced popu- ultimately precipitate population declines or local extirpation. This lation size in YNP (Figure 1; Almberg et al., 2012). Large-scale out- phenomenon remains understudied in wild populations, with little breaks of CDV in 1999, 2005, and 2008 (with smaller outbreaks in consensus between disparate host–parasite systems. 2002 and 2017) infected multiple carnivore hosts in the Greater A recent meta-analysis found strong support for the monocul- Yellowstone Ecosystem, leading to high levels of wolf-pup mortality ture effect in wildlife by examining the effect of population-level (Almberg et al., 2009, 2011; Almberg, Cross, & Smith, 2010; Stahler, heterozygosity on parasite success (Ekroth, Rafaluk-Mohr, & Macnulty, Wayne, vonHoldt, & Smith, 2013). While the effects of King, 2019). However, this study primarily focused on invertebrate most outbreaks were short-lived, the 2008 outbreak coincided hosts and included both laboratory- and field-based studies. Its with the invasion of sarcoptic mange in YNP wolves. Combined focus on population-level heterozygosity further excluded consider- with density-dependent mortality caused by interpack aggression ation of individual-level effects. Although relatively few in number, (Cubaynes et al., 2014), the 2008 CDV outbreak and 2007 mange in- within-population studies in wildlife have reported an inverse rela- vasion appeared to have regulated population size, which has stabi- tionship between genetic diversity and disease using neutral micro- lized between 80 and 108 wolves since (Almberg et al., 2012; Smith satellites (Coltman, Pilkington, Smith, & Pemperton, 1999; Townsend et al., 2020). et al., 2018), immunogenetic markers (Brambilla, Keller, Bassano, & Sarcoptic mange is caused by the ectoparasitic mite Sarcoptes Grossen, 2018), and genome-wide datasets (Banks et al., 2020). In scabiei and has been observed in YNP wolves every year since addition, morbidity has been associated with specific loci in multi- its invasion in January 2007 (Almberg et al., 2012, 2015; Pence ple host species (Batley et al., 2019; Donaldson et al., 2017; Elbers, & Ueckermann, 2002). Symptoms include pruritus, alopecia, eo- Brown, & Taylor, 2018; Ellison et al., 2014; Margres et al., 2018). sinophilia, hyperkeratosis, hyperpigmentation, and dermal in- Considered together, these studies highlight the importance of char- flammation (Almberg et al., 2012; Bornstein, Morner, & Samuel, acterizing genetic variation within the context of wildlife disease, 2001; Nimmervoll et al., 2013; Oleaga, Casais, Prieto, Gortázar, particularly for conservation-relevant species. & Balseiro, 2012). These symptoms are consistent with type IV We contributed to these efforts by examining host genomic (or delayed) hypersensitivity, which suggests that an ineffective variation and infection severity in a wild population of canids: gray immune response harms the host through chronic inflammation wolves (Canis lupus) inhabiting Yellowstone National Park (YNP). (Abbas, Lichtman, & Pillai, 2016). Yet, the severity of these symp- YNP wolves have been closely monitored for disease since their toms varies widely among wolves. Some individuals develop minor initial reintroduction in 1995 and 1996 (Phillips & Smith, 1997). To symptoms, rapidly clear mites, and fully recover within months. minimize risk, founders were screened for good health, vaccinated Others quickly develop severe symptoms that worsen until death against numerous canine diseases, and treated with a broad-spec- from mange or its associated dehydration, emaciation, secondary trum acaricide and anthelmintic (Almberg, Cross, Dobson, Smith, bacterial infection, or increased vulnerability to other causes such & Hudson, 2012). As a result, founders and their offspring initially as intraspecific killings (Almberg et al., 2012; DeCandia, Leverett, bore low disease loads. Within a few generations, however, their & vonHoldt, 2019). As the source of this variability remains 17524571, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eva.13127 by Colorado Parks & Wildlife, Wiley Online Library on [13/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 430 �431 unknown, mange is considered a ubiquitous yet neglected dis- characterized by lower elevations (1,500–2,200 m), serves as prime ease (Hengge, Currie, Jäger, Lupi, & Schwartz, 2006; Walton, Holt, wintering habitat for the park's ungulates, and supports a higher Currie, & Kemp, 2011). density of wolves than the interior. In contrast, the interior (7,991 We hypothesized that host genomic variation contributes to km2) is higher in elevation (>2,500 m), receives higher annual snow- differences in mange infection severity in YNP wolves. More spe- fall, and generally supports lower densities of wolves and ungulates. cifically, we predicted that infection severity would inversely correlate with genome-wide diversity. Through implementation of a family-based association study, we anticipated identification of 2.2 | Sample collection and mange classification associated loci with putative gene functions related to immunity, inflammation, and skin barrier integrity, highlighting the relevance We used archived tissue and blood samples collected by the National of specific variants to the mange phenotype. As numerous factors Park Service (NPS) during field necropsies and annual helicopter are known to contribute to disease state in wildlife, we predicted capture and handling of YNP wolves conducted in accordance with that genetic variation would be one of several variables predictive NPS Institutional Animal Care and Use Committee (IACUC permit of mange severity at the individual level, with environmental and IMR_YELL_Smith_wolves_2012). Sample collection procedures were pack-level variables also relevant. We further considered changes in also reviewed and approved by the Princeton University Institutional genomic variation through time at the population level. Here, we an- Animal Care and Use Committee (Princeton IACUC #2009A-17). ticipated reductions of genome-wide variation since the initial rein- Annually, both static and dynamic life history data were collected troduction events in 1995–1996, as YNP typically serves as a source on YNP wolves. Static metadata included sex, coat color (gray or population for surrounding areas rather than a sink for dispersers black), date of birth, natal pack, and date of death. Dynamic meta- (vonHoldt et al., 2010). We additionally predicted that mange-asso- data included annual records of pack membership, age group, and ciated alleles would reduce in frequency following the 2007 invasion social status. In cases where sex was undetermined (i.e., decom- of mange, as we hypothesized that severe infection exerts selective posed carcasses), we used a simple molecular assay of sex chromo- pressure on YNP wolves. somes to infer sex (DeCandia, Gaughran, Caragiulo, & Amato, 2016). To test these hypotheses, we generated a genome-wide dataset Pack-level variables included location in the park (northern range or of single nucleotide polymorphisms (SNPs) in a subset of YNP wolves interior), pack size, and breeding status (i.e., whether the pack con- exposed to sarcoptic mange for individual-level analyses. We then tained a breeding pair that year). genotyped these same SNPs in all wolves with biobanked blood and Frequent observations of YNP wolves also resulted in the docu- tissue for population-level analyses. The availability of samples and mentation of individual mange scores, which reflected the percent- detailed phenotypic data for YNP wolves uniquely enabled us to dis- age of body area presenting symptoms, such as hair loss or lesions. entangle genetic and environmental factors underlying this complex On a 3-point scale, a score of 0 indicated no evidence of mange, disease phenotype. As mange infects over one hundred mammal 1 indicated that ≤ 5% of the body was impacted by mange-related species worldwide including humans (Pence & Ueckermann, 2002), symptoms, 2 referred to 6%–50% of the body being symptomatic, YNP wolves serve as a case study with important implications for and 3 referred to the most severe score where > 50% of the body mammals globally. This study advances our understanding of the was presenting symptoms (Almberg et al., 2012; Pence, Windberg, & genomics underlying mange in a free-ranging carnivore, while also Sprowls, 1983). Any field-based observation or annual capturing of providing insights and predictions applicable to diverse host–para- animals by NPS resulted in a mange score assigned to the correspond- site systems. ing individual. The frequent monitoring by NPS officials, consistent method of mange score assignment, and repeated observation of 2 | M E TH O DS 2.1 | Study area the same wolves maximized confidence in disease phenotypes. In downstream statistical analyses, we used the highest mange score documented per wolf for genetic analyses and mange score at the time of observation for mixed-effects modeling. Severity classes were coded “mild” for highest score 1, “moderate” for highest score YNP encompasses 8,991 km2 of protected land in north-western 2, and “severe” for highest score 3. Wyoming and adjacent parts of Montana and Idaho in the western To estimate pack-level exposure, we used the dates of first and United States. YNP is mountainous (elevation range: 1,500–3,800 last mange observation for each pack. We then flanked these dates m), and its steep gradients in elevation, soil, and climate contribute by one month to account for asymptomatic periods that can both to varied land cover, including riparian vegetation, shrubland, grass- precede and follow infection (Arlian, 1989; Samuel, 1981). This es- land, alpine meadows, and mixed coniferous forests. We make refer- tablished the mange exposure window for each member in the pack. ence to two regions of the park, the northern range and the interior, We restricted our mange dataset to only include wolves that con- based on ecological and physiographical differences and variation tained three or more observations and were putatively exposed to in disease dynamics (see Almberg et al., 2009, 2012 for details). mange, even if the animal was assigned a mange score of 0 (following Importantly, the 1,000 km2 area of the northern range within YNP is vonHoldt et al., 2020). 17524571, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eva.13127 by Colorado Parks & Wildlife, Wiley Online Library on [13/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | DECANDIA et al. �| DECANDIA et al. 2.3 | DNA extraction and restriction site-associated DNA (RAD) sequencing excluded wolves with fewer than three observations and no history of mange exposure (based on pack membership and infection history), as it was impossible to assess mange severity class in individu- We extracted genomic DNA following the Qiagen DNeasy Blood als never challenged by mites. To complete the dataset with this final & Tissue Kit standard protocol, quantified samples with the set of samples, we implemented the STACKS v2.2 populations module Quant-iT™ PicoGreen® dsDNA Assay Kit or Qubit TM fluoromet- a second time with an additional filtering parameter (−r 0.9) to re- ric quantitation, and standardized concentrations to 5 ng/μL. We tain loci genotyped in more than 90% of wolves. We used VCFtools additionally visualized DNA extractions on 1% agarose gels to v0.1.12b (Danecek et al., 2011) to remove singletons, doubletons, identify and retain samples with high molecular weight for library and sites found on the X chromosome, due to difficulties posed by preparation. chromosomal sex determination and X-inactivation to mixed-sex We used a modified restriction site-associated DNA sequenc- study designs (Clayton, 2009). This produced our final dataset of ing (RADseq) protocol by Ali et al. (2016) to generate genome-wide high-confidence autosomal SNPs for downstream analysis. For pop- SNP data. To summarize, we digested genomic DNA with the sbfI ulation-level analyses, we genotyped these same SNPs in all wolves restriction enzyme prior to ligation of uniquely barcoded, bioti- (regardless of mange exposure history) with available biomaterial. nylated adaptors. We then pooled barcoded samples (48 samples per pool) and randomly sheared DNA to 400 bp on a Covaris LE220. Sheared libraries were enriched for fragments containing the li- 2.4 | Genetic diversity statistics gated adaptor using a streptavidin bead-binding assay (Dynabeads M-280, Invitrogen), with subsequent library preparation following We hypothesized that genetic diversity would inversely correlate the standard manufacturer protocol for the NEBNext Ultra II DNA with mange infection severity, as coded into classes mild, moderate, Library Preparation Kit (New England Biolabs). We purified and se- and severe. We used STACKS v2.2 to examine patterns of genetic di- lected libraries for fragments 300–400 bp in size using Agencourt versity between: (a) infected and uninfected wolves and (b) infected AMPure XP magnetic beads. Two libraries were pooled for each final wolves with different mange severities. Diversity metrics included sequencing library to contain 96 barcoded samples. Final libraries the percentage of polymorphic sites (%Poly), number of private al- were then standardized to 10 nM before paired-end sequencing leles (PAS), minor allele frequency (MAF), observed heterozygosity (2X150 nt) on an Illumina HiSeq 2500 or NovaSeq 6000. (HO), expected heterozygosity (HE ), and nucleotide diversity (π). We We used a custom perl script (sbfI_flip_trim_150821.pl, see assessed the statistical significance of between-group differences Appendix S1) to align all forward and reverse reads with the re- in R. For binary mange presence, we used two-tailed Welch's t tests, striction enzyme cut site into one file. We then used STACKS v1.42 as we assumed unequal variance between the two infection groups. (Catchen, Hohenlohe, Bassham, Amores, & Cresko, 2013) for For infection severity, we used analysis of variance (ANOVA), as this the initial stages of data processing, in order to manually remove allowed for inclusion of all four mange severity groups in the same poor-quality samples from the dataset before paired-end mapping. analysis. We used process_radtags to demultiplex and filter reads for > 2bp We next used ADZE v1.0 to estimate rarefied metrics of allelic barcode mismatches or quality scores below 90% using a sliding win- diversity (Szpiech, Jakobsson, & Rosenberg, 2008). As allelic rich- dow (15% of the read), and removed PCR duplicates using default ness (AR), private allelic richness (PAR), and shared PAR are heavily parameters in clone_filter. influenced by sample size, adoption of a rarefaction approach en- We completed paired-end alignments to the reference dog ables cross-group comparisons when sample sizes differ. Using this CanFam3.1 genome (Lindblad-Toh et al., 2005) using STAMPY v1.0.21 approach, AR, PAR, and shared PAR are estimated by averaging subsa- (Lunter and Goodson 2011) for samples with > 500,000 reads to mples of each group at standardized sample sizes. We set the miss- maximize sequence coverage. SAM files were sorted and filtered for ing data tolerance to 25% and calculated mean AR, PAR, and shared quality scores (MAPQ) ≥ 96 using Samtools v0.1.18 (Li et al., 2009), PAR between infection severity groups. with a final conversion to BAM format. We subsequently used STACKS v2.2 to identify, genotype, and filter SNPs with gstacks and populations for paired-end data using the Marukilow model (Maruki 2.5 | Mixed-effects modeling & Lynch, 2017). As this model assesses the statistical likelihood of each genotype call, it reduces the need for subsequent coverage fil- We used mixed-effects modeling to contextualize genetic diversity ters when paired with clone-filtering. within the broad range of factors that may influence infection sever- We implemented the populations module using all available ity in YNP wolves. Input data were derived from annual observa- samples and the flag–write_single_snp to retain only a single poly- tions conducted in YNP between 2007 and 2019, and included both morphic site per read. When samples were replicated in the li- static and dynamic life history variables for infected wolves (mange brary preparation and sequencing process, we used PLINK (Purcell status of 1, 2, or 3). Mange status at the time of observation served et al., 2007) to compare each of the replicates for proportion of as the response variable, and both random and fixed effects were missing loci and retained the sample with lower missingness. We considered during model selection. Individual wolves appeared in 17524571, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eva.13127 by Colorado Parks & Wildlife, Wiley Online Library on [13/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 432 �433 the dataset multiple times, with their static life history variables un- the reduced model one at a time and implementing lrts to assess sig- changed and their dynamic life history variables sometimes differing. nificance. We performed a similar procedure for pairwise interaction To control for repeated measures, random-effects variables terms between fixed effects contained in the reduced CLMM to see included individual identifier, pack membership at the time of ob- whether their inclusion significantly improved model fitting. We cal- servation, and year observed. Individual identifier controlled for culated Akaike information criterion adjusted for small sample (AICc) nonindependence between repeated measures of the same wolf, using AICcmodavg v2.2-2 (Mazerolle, 2019) and weighted AICc using whereas pack membership and year observed controlled for shared MuMIn v1.43.15 (Bartoń, 2019). environmental effects present within each pack (Almberg et al., 2015; Brzeski et al., 2015) and observation year (Stahler et al., 2013). As numerous wolves changed pack membership across years, we fitted 2.6 | Identifying outlier loci these variables as partially crossed random intercepts. Fixed effects included environmental (season and location in the We implemented a univariate linear mixed model in GEMMA to park), pack-level (breeding status and size of the pack), and individ- identify outlier loci associated with infection severity (Zhou & ual-level (sex, coat color, age group, social status, and standardized Stephens, 2012, 2014). We included sex and coat color as covari- observed heterozygosity or HO) variables. To determine season, we ates in the model to account for static life history variables, and assigned observations obtained during October through March as used a pairwise relatedness matrix to account for familial structure “winter,” and those obtained during April through September as within the dataset. As our dataset included wolves with unknown “summer.” For each observation, we used pack membership to de- pedigree relationships, we calculated a centered pairwise related- termine location in the park (northern range or interior), estimated ness matrix using the -gk 1 flag in GEMMA. We excluded natal pack pack size (range 1–18), and pack-level breeding status (yes or no) for as a covariate, as the pairwise relatedness matrix accounted for all that observation year. Sex (male or female), coat color (gray or black), possible relatives, rather than relying on inferred relations sharing age group (yearling or adult), and social status (subordinate or alpha) a natal pack. We adjusted the lrt p-values obtained using a modified were assigned at each observation, with missing data interpolated false discovery rate (FDR) procedure (Benjamini & Yekutieli, 2001), using adjacent observations and YNP Annual Reports (2007–2019). and used an in-house python script (vonHoldt, Heppenheimer, To account for genetic diversity, we standardized HO calculated for Petrenko, Croonquist, & Rutledge, 2017) to annotate significant out- each wolf by subtracting the mean and dividing by standard devia- liers as intronic, exonic, intergenic, or within 2Kb of a promoter in tion. We chose this measure to represent genetic diversity as stan- the reference dog genome (Lindblad-Toh et al., 2005). We predicted dardized HO provided the most inclusive estimate of genome-wide functional relevance using the Ensembl Variant Effect Predictor variation without overparameterizing candidate models. We then (VEP) web interface (McLaren et al., 2016) and queried genic sites formulated a priori hypotheses about how each variable may af- in the Ensembl, Online Mendelian Inheritance in Man (OMIM, 2018), fect mange infection severity based on existing literature (Table S1; and GeneCards (www.genec​ards.org) databases. Finally, we used Almberg et al., 2012, 2015; Candille et al., 2007; Cross et al., 2016; G:GOST in G:PROFILER to conduct gene ontology analyses (Raudvere DeCandia et al., 2018; Fazal, Cheema, Maqbool, & Manzoor, 2014; et al., 2019). We searched annotated genes for all available anno- Feather, Gough, Flynn, & Elsheikha, 2010; Mech & Boitani, 2003; tations (including molecular functions, cellular components, and Oleaga et al., 2011; Pence et al., 1983; Spielman, Brook, Briscoe, biological processes) and assessed statistical significance using the et al., 2004; Stahler et al., 2013). Benjamini–Hochberg FDR of 0.05 (Benjamini & Hochberg, 1995; We used cumulative link mixed models (CLMM) implemented vonHoldt et al., 2020). Although this analysis may be underpowered in the R package ordinal v2019.12-10 to test these hypotheses due to sample size constraints, we adjusted significance thresh- (Christensen, 2019). A type of generalized linear mixed-effects olds to account for multiple testing and decrease the likelihood of model, CLMMs are optimized for ordinal response variables and false positives (following DeCandia, Brzeski, et al., 2019; vonHoldt employ a maximum-likelihood framework for parameter estimation et al., 2020). using the Laplace approximation. We initiated model selection by constructing a null model (no fixed effects) and a global model (all nine fixed effects). We used the saturated model to check for collin- 2.7 | Population-level analyses earity between fixed effects using the check_collinearity function in the R package performance v0.4.5 (Lüdecke, Makowski, Waggoner, We next considered changes in genetic variation through time in all & Patil, 2020). We then implemented a stepwise model reduction YNP wolves with available biomaterial, regardless of mange expo- procedure, where we sequentially removed the nonsignificant term sure history. Using static metadata, dynamic observations, and YNP with the highest p-value in each CLMM (Stahler et al., 2013). We annual reports, we determined which wolves were alive in each ob- compared sequential models using the likelihood-ratio test (lrt) for servation year between 1995 and 2019. We then used ADZE v1.0 cumulative link models, and halted the reduction process when to estimate annual mean allelic richness, while controlling for sam- removal of the next fixed effect variable led to significantly worse ple size differences between years. For these analyses, we used the model fitting. We reconsidered dropped terms by adding them to missing data tolerance of 100% to ensure that the same loci were 17524571, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eva.13127 by Colorado Parks & Wildlife, Wiley Online Library on [13/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | DECANDIA et al. �| DECANDIA et al. analyzed in each year's calculation. To account for the breeding (n = 13, highest mange score 3) symptoms (Figure S1). For popula- structure of YNP wolves, we performed these analyses a second tion-level analyses, we genotyped these same 76,859 SNPs in 408 time using only known breeders. Here, we considered breeding sta- unique individuals, regardless of mange exposure history. tus to be a static life history variable, in order to increase annual sample sizes. As such, each year's calculation included all living breeders regardless of their reproduction status in that particular year. 3.2 | Genetic diversity Following examination of genome-wide allelic richness, we explored the change in per-locus allele frequencies through time. For Regarding susceptibility (i.e., binary mange presence), infected this analysis, we binned loci into three categories based on the asso- wolves (inf) exhibited higher levels of genetic diversity than wolves ciation of the focal allele (typically the minor allele) with mange se- with no detected infection (uninf) across several diversity metrics verity in GEMMA. Categories included (a) no association, (b) positive (Table S2). This relationship was statistically significant for observed association (where increased allele frequency was associated with (two-sided t test; HO, inf = 0.1986 ± 0.0006, uninf = 0.1942 ± 0.0006, more severe mange), and (c) negative association (where increased t153650 = −5.110, p < .001) and expected (two-sided t test; HE, allele frequency was associated with milder mange). We constructed inf = 0.1874 ± 0.0005, uninf = 0.1856 ± 0.0005, t153720 = −2.367, a mixed-effects model with a Gaussian likelihood and weakly reg- p = .018) heterozygosity, but not for minor allele frequency (two- ularizing skeptical priors in the R package brsm (Bürkner, 2017) to sided t test; MAF, inf = 0.1249 ± 0.0005, uninf = 0.1239 ± 0.0005, assess whether positive and negative association with mange sever- t153710 = −1.521, p = .128) or nucleotide diversity (two-sided t test; ity influenced changes in per-locus allele frequency through time. π, inf = 0.1888 ± 0.0005, uninf = 0.1876 ± 0.0006, t153710 = −1.623, In this model, standardized allele frequency was regressed on the p = .105). Although the number of private alleles was higher in the in- fixed effects of year, association with mange, and the interaction fected group, sample size differences likely contributed to this result. between them, while controlling for the locus ID of each allele as a We therefore used rarefaction to estimate mean allelic richness (AR) random effect. To improve MCMC convergence times, we included and mean private allelic richness (PAR) across standardized sample all positively and negatively associated alleles but only a randomly sizes in infected and uninfected wolves. We found that uninfected selected subsample of 500 nonassociated alleles. For each subset of wolves exhibited higher levels of allelic variation across both diver- alleles, the model was run twice: once for the years 1995–2006, and sity metrics (AR, inf = 1.9276 ± 0.0007, uninf = 1.9431 ± 0.0007; PAR, once for the years 2006–2019, to assess changes in allele frequency inf = 0.0498 ± 0.0006, uninf = 0.0653 ± 0.0007; Table S3). before versus after the 2007 mange invasion of YNP. The approach When grouped by infection severity (uninfected, mild, moderate, was repeated four times to confirm that results were consistent and severe), we consistently observed significant differences be- across independent subsamples of nonassociated alleles and across tween severity groups across diversity metrics, including observed independent Markov chains. (ANOVA; HO, F1,307434 = 7.970, p = .005) and expected (ANOVA; HE, F1,307434 = 558.900, p < .001) heterozygosity, minor allele frequency 3 | R E S U LT S 3.1 | RAD-sequencing (ANOVA; MAF, F1,307434 = 73.280, p < .001), and nucleotide diversity (ANOVA; π, F1,307434 = 283.700, p < .001). Notably, within the infected groups (mild, moderate, and severe), we observed decreasing genetic diversity with increasing infection severity across all metrics (Figure 2a–f). Uninfected wolves had the largest percentage of Our first implementation of populations catalogued 214,762 variant polymorphic loci and number of private alleles, but were otherwise sites in 510 samples. After removing duplicates, wolves with fewer intermediate when compared to other severity classes (Table S2; than three observations, and putatively unexposed individuals, we Figure S2). implemented populations a second time to create a mange-relevant Similar patterns emerged when controlling for sample size. dataset with an additional filtering parameter that removed SNPs Mean allelic richness was highest in mildly infected wolves genotyped in < 90% of individuals. This resulted in 106,936 SNP loci (AR = 1.7360 ± 0.0011), intermediate in moderately infected wolves genotyped in 117 wolves. We then filtered SNPs to remove single- (AR = 1.7059 ± 0.0012), and lowest in severely infected wolves tons, doubletons, X chromosome sites, and loci missing allelic depth (AR = 1.6347 ± 0.0016; Figure 2g, Table S3), with nonoverlapping information. The final dataset retained 76,859 SNPs genotyped in standard errors between all estimates. Regarding mean private al- 117 wolves, with 49 uninfected and 68 infected individuals. Within lelic richness, mildly infected wolves harbored the most unique the infected group, wolves exhibited mild (n = 29, highest mange alleles (PAR = 0.0425 ± 0.0004), with moderately infected wolves score 1), moderate (n = 26, highest mange score 2), and severe intermediate (PAR = 0.0323 ± 0.0003) and severely infected wolves F I G U R E 2 Genetic diversity statistics for mange-infected wolves grouped by mild (highest mange score 1), moderate (highest mange score 2), and severe (highest mange score 3) infection severity. Metrics include the following: (a) percentage polymorphic loci, (b) number of private alleles, (c) observed heterozygosity, (d) expected heterozygosity, (e) minor allele frequency, (f) nucleotide diversity, (g) rarefied mean allelic richness, and (h) rarefied mean private allelic richness 17524571, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eva.13127 by Colorado Parks & Wildlife, Wiley Online Library on [13/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 434 �(b) (c) (d) (e) (f) (g) (h) 17524571, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eva.13127 by Colorado Parks & Wildlife, Wiley Online Library on [13/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License (a) 435 | DECANDIA et al. �| DECANDIA et al. possessing the fewest (PAR = 0.0195 ± 0.0003; Figure 2h, Table model reduction procedure. The most parsimonious model contained S3). Uninfected wolves exhibited the second highest allelic rich- environmental, pack-level, and individual-level variables (Figure 4; ness (AR = 1.7190 ± 0.0011) and possessed the most unique alleles Table S6). More specifically, season (β = 0.9485, Z = 3.166, p = .002), (PAR = 0.0469 ± 0.0004; Table S3; Figure S2). breeding status of the pack (β = −2.0258, Z = −2.584, p = .010), and We additionally examined private allelic richness shared be- age group (β = 1.1769, Z = 2.016, p = .044) exhibited significant ef- tween pairwise combinations of infection groups (Figure 3, Table fects, with standardized HO approaching significance (β = −0.8434, Z S4). In general, similar groups (where highest mange score was off- = −1.911, p = .056; Table 1). These four variables appeared in all six set by one) shared more unique alleles than disparate groups (where models with ΔAICc < 2, with all other variables appearing in only one highest mange score was offset by two or three), with the exception of six top models (Table S6). Parameter estimates for these variables of the uninfected–moderate pair. As such, uninfected and mildly in- suggested that wolves experienced more severe mange in winter fected wolves shared the most alleles (0.0438 ± 0.004), followed by (environment: season) and in nonbreeding packs (pack level: breed- mildly and moderately infected wolves (0.0312 ± 0.003). In contrast, ing status). Regarding individual-level variables, adult wolves (age uninfected and severely infected wolves shared the fewest alleles group) and individuals with reduced genetic variation (standardized (0.0155 ± 0.0002), with alleles shared by mildly and severely in- HO) also experienced more severe mange. Removal of standardized HO fected wolves similarly low (0.0192 ± 0.002). Within the pairs offset from the reduced model resulted in significantly worse model fitting by one mange score, we observed an inverse relationship between (p = .041), and subsequent addition of omitted and pairwise interac- infection severity and shared private allele richness. For example, tion terms did not significantly improve AICc (p > .05). We therefore the moderate–severe pair (0.0221 ± 0.0003) shared fewer alleles retained the reduced model as our most parsimonious CLMM. This than both the uninfected–mild (0.0438 ± 0.0004) and mild–moder- model excluded location, pack size, coat color, sex, and social status as ate (0.0312 ± 0.0003) pairs. This trend also occurred for pairs offset significant predictors of mange severity at the individual level. by two mange scores. 3.3 | Mixed-effects modeling 3.4 | Identifying outlier loci We identified 410 autosomal sites significantly associated with After constructing our null and global CLMMs, we calculated the vari- mange severity after applying BY-modified FDR correction ance inflation factor (VIF) for each fixed effect variable included in (p < .004). Frequency of the mange-associated allele was positively the saturated model. We observed low collinearity between predictor associated with mange severity at 224 sites and negatively asso- variables (VIF range 1.18–3.13; Table S5) and initiated our stepwise ciated with mange severity at 186 sites. Across all 410 sites, the mange-associated allele was typically found in the heterozygous state (Table S7). Site annotations included 12 exonic, 171 intronic, 17 near promoters, and 257 intergenic sites, with VEP annotations including 20 low, four moderate, and 847 modifier effects (N.B., many sites had multiple annotations). We identified 42 gene ontological categories that passed the FDR threshold set in G:PROFILER (Table S8). Categories included four molecular functions, 16 cellular components, and 22 biological processes (Figure S3a). The majority of categories involved cell barrier function and flexibility (n = 11), cell–cell and cell–substrate junctions (n = 6), and cell differentiation and development (n = 19). Genic sites queried in the Ensembl, OMIM, and GeneCards databases returned putative functions related to innate and adaptive immunity, autoimmunity and inflammation, cell barriers and adhesion, and skin development and disorders. For example, hematopoietic prostaglandin D synthase (HPGDS) has been implicated in the resolution of delayed-type hypersensitivity responses (Trivedi et al., 2006). Similarly, protein tyrosine phosphatase, nonreceptor type 6 (PTPN6) has been linked to heightened inflammation characterized by edema, sustained inflammatory infiltrate, and the delayed wound healing (Lukens et al., 2013). Both loci exhibited decreasing minor allele frequency with increasing infection severity (Figure S3b). F I G U R E 3 Rarefied private allelic richness shared between mange severity classes Additional genes were associated with chronic skin disorders, such as psoriasis and peeling skin disease (corneodesmosin, CDSN; 17524571, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eva.13127 by Colorado Parks & Wildlife, Wiley Online Library on [13/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 436 �F I G U R E 4 Fixed effects included environmental, pack-level, and individuallevel variables. Asterisks indicate variables included in the final model. Figure created with BioRender Response Environmental Pack-Level Individual-Level *Season Location *Breeding Status Pack Size *Age Group *Standardized HO Sex Coat Color Social Status ~ + Mange Score TA B L E 1 Parameter estimates (β), standard error, Z-score, p-value, and 95% confidence intervals for variables contained in the most parsimonious model predicting mange severity at the individual level Explanatory variable SE β Z-score p-value + CI (2.5%) CI (97.5%) Season (winter) 0.9485 0.2996 3.166 .002 0.3612 1.5358 Breeding status (yes) −2.0258 0.7839 −2.584 .010 −3.5621 −0.4894 Age group (adult) 1.1769 0.5839 2.016 .044 0.0326 2.3213 Standardized HO −0.8434 0.4413 −1.911 .056 −1.7083 0.0216 F I G U R E 5 Mean allelic richness rarefied to 20 individuals for all wolves (black solid line) and all known breeders (brown dashed line) alive in each year 437 Mean Allelic Richness 1.76 1.74 1.72 All Wolves 1.70 Only Breeders Mange Invasion Max. Mange Burden CDV Outbreak 1995 2000 2005 2010 2015 2020 Year Matsumoto et al., 2008; Oji et al., 2010), inflamed skin lesions called wolves in 2019 to 122 wolves in 2003 (median = 73). Rarefied hidradenitis suppurativa (nicastrin, NCSTN; Pink et al., 2011), inelas- mean allelic richness decreased through time, with the high- tic skin termed cutis laxa (Elastin, ELN; Hadj-Rabia et al., 2013), est values calculated for 1995 (AR = 1.7456 ± 0.0011) and 1996 ichthyosis (scaly skin) associated with Refsum disease (peroxiso- (A R = 1.7476 ± 0.0011) and the lowest values calculated for 2017 mal biogenesis factor 7, PEX7; van den Brink et al., 2003; Schmuth (A R = 1.6926 ± 0.0012), 2018 (AR = 1.7017 ± 0.0013), and 2019 et al., 2013), palmoplantar keratoderma (thickening of the skin (AR = 1.6891 ± 0.0014; Figure 5). Analysis of breeding individuals around hands and feet) and alopecia (SAM and SH3 domain con- was restricted to 1995–2016 due to small sample sizes in 2017– taining 1, SASH1; Courcet et al., 2015), and epithelial cell growth 2019. These datasets ranged from 17 wolves in 2016 to 67 wolves and thickening, termed hyperplasia and hyperkeratosis (SMAD in 2003 (median = 41). Although the overall trend was similar, family member 7, SMAD7; He et al., 2002). These loci exhibited breeding individuals exhibited higher mean allelic richness than negative (ELN, SASH1, and SMAD7) and positive (CDSN, NCSTN, and the census population in all years except 2001–2003. The highest PEX7) relationships between minor allele frequency and infection values were calculated in 1995 (AR = 1.7579 ± 0.0011) and 1996 severity, with no overarching pattern evident in control loci lacking (A R = 1.7570 ± 0.0011), with the lowest values occurring in 2001 association with mange severity (Figures S3 and S4; Table S9). (A R = 1.7149 ± 0.0011), 2002 (A R = 1.7104 ± 0.0011), and 2003 (AR = 1.7076 ± 0.0011; Figure 5). 3.5 | Population-level analyses Allele frequency analyses included three bins of loci: (a) no association, (b) positive association, and (c) negative association between the focal allele (typically the minor allele) frequency and We analyzed 76,859 SNPs in 408 unique individuals observed in mange severity. The majority of mange-associated loci (n = 399/410) YNP between 1995 and 2019. Annual datasets ranged from 22 had both alleles present in YNP since 1995–1996, with the minor 17524571, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eva.13127 by Colorado Parks & Wildlife, Wiley Online Library on [13/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | DECANDIA et al. �| DECANDIA et al. allele emerging for the remaining 11 loci between 1997 and 2003 were restricted to the years after mange invaded YNP. From 1995 to (Table S10). All mange-associated alleles were therefore pres- 2006, none of these three bins of loci exhibited significant changes ent in the population as standing variation in January 2007, when in frequency on average (nonassociated, β = −0.004, 95% credible mange invaded the park. We tested for selection by constructing interval [−0.007, 0.0001]; positively associated, β = 0.000, 95% cred- a mixed-effects model to estimate whether the average change in ible interval [−0.006, 0.005]; negatively associated, β = 0.000, 95% allele frequency through the years 2006–2019 depended on mange credible interval [−0.007, 0.006]; Figure 6b). These results were con- association. As expected, randomly subsampled nonassociated al- sistent across four independent subsamples of nonassociated alleles leles did not significantly change in frequency on average during (Figure S5). this time frame (β = 0.003, 95% credible interval [−0.009,0.002]; Figure 6a). Conversely, alleles positively associated with mange severity significantly decreased in frequency on average (β = −0.041, 4 | D I S CU S S I O N 95% credible interval [−0.049, −0.034]; Figure 6a), while alleles negatively associated with mange severity significantly increased in fre- In the present study, we characterized the relationship between quency on average (β = 0.010, 95% credible interval [0.002,0.018]; host genomic variation and disease severity in a wild population Figure 6a). Critically, these significant changes in allele frequency of reintroduced canids. Through use of biobanked samples and (a) (b) F I G U R E 6 Posterior predictions of the average changes in frequency through time for alleles not associated, positively associated, and negatively associated with mange severity, with 95% credible intervals surrounding the mean. Nonassociated alleles comprise a randomly selected subset of 500 loci. The same analysis was repeated to assess changes in allele frequency (a) after mange invasion of YNP and (b) before mange invasion of YNP 17524571, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eva.13127 by Colorado Parks & Wildlife, Wiley Online Library on [13/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 438 �439 detailed phenotypic records, we calculated summary statistics of role of genetic diversity in shaping the landscape of mange infection genome-wide variation and performed a family-based association severity at the individual level. The most parsimonious model high- study to identify genomic variants linked with mange infection se- lighted the multifactorial nature of disease state in wildlife through verity. We contextualized genomics within the broad range of fac- inclusion of genetic (standardized HO), environmental (season), and tors influencing disease state in YNP, and considered changes in demographic (breeding status of the pack and individual age) vari- genomic variation through time at the population level. Through ables. Critically, we found model support for an inverse relationship these analyses, we found evidence of selection acting on mange- between genetic diversity and mange severity. This result confirmed associated loci following the 2007 invasion of S. scabiei mites in YNP. the relevance of genome-wide variation in predicting mange sever- Although numerous studies have catalogued immunogenetics in ca- ity in YNP wolves alongside environmental, behavioral, and demo- nids (Aguilar et al., 2004; Arbanasić et al., 2013; Galaverni, Caniglia, graphic factors. Fabbri, Lapalombella, & Randi, 2013; Hedrick, Lee, & Garrigan, 2002; To build upon this result and capture the complex nature of dis- Hedrick, Lee, & Parker, 2000; Kennedy et al., 2011; Marshall, ease, we explored other model relationships associated with mange Langille, Hann, & Whitney, 2016), this study was among the first severity. Regarding season, we found evidence that wolves presented to explore genome-wide variation within the context of disease se- more severe mange in winter, when cold ambient temperatures ren- verity in a wild canid population. This allowed us to test whether der thermoregulation more difficult. This result supports previous host genomic variation was predictive of disease severity, as sug- studies examining mange prevalence (Almberg et al., 2012), mortality gested by the monoculture effect observed in agricultural settings risk (Almberg et al., 2015), and energetics (Cross et al., 2016) in YNP (Ekroth et al., 2019). Results yielded system-specific insights, while wolves and other species impacted by mange (Martin et al., 2018). also contributing to the larger-scale effort of applying genomic tech- Season may also contribute to our finding that nonbreeding packs niques to wildlife disease ecology (Blanchong et al., 2016; DeCandia were more likely to present severe mange. Breeding season for YNP et al., 2018). wolves (mid-February mating; mid-April birth) directly follows the We hypothesized that host genomic variation would predict mean severity window for infected packs (September 2–February mange infection severity rather than susceptibility, given the mode 2; Almberg et al., 2015). Mangy individuals may exhibit poor body of transmission and pathology of sarcoptic mange. Transmission of S. condition during breeding season, reducing breeding likelihood and scabiei mites occurs upon contact with an infected individual or fo- efficacy (Stahler et al., 2013), as seen in other host–parasite systems mite, such as a mange-infected den (Montecino-Latorre et al., 2019; (Holand et al., 2015; Marzal, De Lope, Navarro, & Møller, 2005; Pence & Ueckermann, 2002). Presumably, exposed wolves have an Møller, 2002; Sarasa et al., 2011). Poor body condition may also equal probability of infection regardless of their genomic diversity. be influenced by age, as adult wolves exhibited worse mange than Inter-individual differences subsequently emerge due to the immune yearlings. This finding is consistent with reports for age/mange re- response mounted by the host (Nimmervoll et al., 2013; Oleaga lationships in Iberian wolves and coyotes (Oleaga et al., 2011; Pence et al., 2012), which is likely to be under genetic control (Steinke, et al., 1983), and divergent from reports in red foxes and dogs (Fazal Borish, & Rosenwasser, 2003). In the present study, we observed et al., 2014; Feather et al., 2010; Newman, Baker, & Harris, 2002). an inverse relationship between host genomic variation and mange These differences in the literature, and our study in particular, may infection severity in YNP wolves. This supports the paradigm that result from the demographics of the dataset. For example, we ex- genetic variation plays an important role in wildlife disease (King & cluded any wolf that had fewer than three observations, which may Lively, 2012; Lively, 2010; Luong, Heath, & Polak, 2007; Spielman, have systematically excluded fatal mange infections in pups, as seen Brook, Briscoe, et al., 2004). Additional evidence includes heterozy- in the Silver pack in 2010 (Smith et al., 2011). We therefore recom- gous house finches (Carpodacus mexicanus) that exhibited reduced mend further study of age-specific outcomes with mange infections disease severity and mounted stronger immune responses than ho- in both wolves and canids, more broadly. mozygous finches after experimental inoculation with Mycoplasma Following our findings that genome-wide variation significantly gallisepticum (Hawley, Sydenstricker, Kollias, & Dhondt, 2005). predicts mange severity, we discovered specific loci associated with Similarly, outbred guppies (Poecilia reticulata) exhibited lower the highest mange score recorded per wolf. These loci were found Gyrodactylus turnbulli parasite intensities and shorter infection dura- in genes related to innate and adaptive immunity, autoimmunity and tions when compared to inbred individuals (Smallbone, Oosterhout, inflammation, cell barriers and adhesion, and skin development and & Cable, 2016). Here, YNP wolves exhibiting mild mange symptoms disorders. For example, reduced minor allele frequency was asso- possessed higher levels of genome-wide variation than wolves ex- ciated with severe mange in genes HPGDS and PTPN6, which have hibiting more severe symptoms. been previously linked to immunopathology, inflammation, and de- Patterns of genome-wide variation suggested that host diver- layed wound healing in mice (Lukens et al., 2013; Trivedi et al., 2006). sity was an important predictor of mange severity in YNP wolves. Additional loci were discovered in genes related to chronic skin However, wildlife disease dynamics are known to be impacted by conditions. Associated genes included CDSN (psoriasis and peeling host environment, demography, and behavior, as well (Ezenwa skin disease; Matsumoto et al., 2008; Oji et al., 2010), NCSTN (hi- et al., 2016; Parratt, Numminen, & Laine, 2016; Silk et al., 2019). We dradenitis suppurativa; Pink et al., 2011), ELN (cutis laxa; Hadj-Rabia therefore used mixed-effects modeling to quantitatively assess the et al., 2013), PEX7 (ichthyosis; van den Brink et al., 2003; Schmuth 17524571, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eva.13127 by Colorado Parks & Wildlife, Wiley Online Library on [13/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | DECANDIA et al. �| DECANDIA et al. F I G U R E 7 Sarcoptic mange was implicated in the dissolution of the Druid Peak pack in late 2009 and early 2010, when numerous pack members became infected. This pedigree contains a subset of Druid wolves shaded to indicate mange infection severity. Genotype at the mange-associated locus contained in gene PTPN6 is indicated, when available, to illustrate how family-based association links genotypes with phenotypes while controlling for relatedness. Similar analyses were conducted for all loci analyzed by GEMMA. Dashed lines connect the same individual to parentage events occurring in different parts of the pedigree et al., 2013), SASH1 (palmoplantar keratoderma and alopecia; Although some fluctuations coincide with CDV outbreaks, the in- Courcet et al., 2015), and SMAD7 (hyperplasia and hyperkeratosis; consistent pattern suggests that disease was not the primary driver He et al., 2002). These annotations and skin conditions matched of long-term decreases in genome-wide variation. Instead, this pat- symptoms of severe mange across host taxa (Almberg et al., 2012; tern may result from YNP’s status as a source population for the Little et al., 1998; Niedringhaus, Brown, Sweeley, & Yabsley, 2019; surrounding Greater Yellowstone Ecosystem, as few wolves suc- Oleaga et al., 2012; Pence & Ueckermann, 2002), and were consis- cessfully disperse into the park (vonHoldt et al., 2010; vonHoldt tent with inflammation-induced immunopathology and hypersensi- et al., 2008). Notably, the mating structure of wolves precludes all tivity presented in allergic and autoimmune disorders (Barker, 2001; individuals from reproducing (Mech & Boitani, 2003), thereby reduc- Barnes, 2010; Bin & Leung, 2016; Esaki et al., 2015; Liang, Chang, & ing the effective population size (vonHoldt et al., 2008). While the Lu, 2016; Nattkemper et al., 2018; Quraishi et al., 2015; Rodríguez overall pattern of diversity was consistent between all wolves and et al., 2014; Sonkoly et al., 2010). While it is possible that mange-as- known breeders, reproductive individuals exhibited higher levels of sociated alleles may also influence individual-level risk during CDV genome-wide variation in all years except 2001–2003. This upward outbreaks, annotations more closely matched the pathology of shift may slow the pace of genomic diversity loss through time, al- mange. Overall, these results mirror other wildlife studies that un- though further study is needed on mate choice and its long-term covered disease-relevant genes in diverse host–parasite systems effects on variation in YNP. (Batley et al., 2019; Donaldson et al., 2017; Elbers et al., 2018; Ellison et al., 2014; Margres et al., 2018). Decline in genome-wide variation at the population level may increase the prevalence of severe mange infections in the future, The relevance of host genomics to disease risk is not restricted given the inverse relationship observed at the genomic scale. While to the individual level. Many of these processes scale to inform pop- higher levels of diversity maintained by breeders may ameliorate ulation-level dynamics through time. In YNP wolves, temporal anal- this risk, mange can still negatively affect YNP wolves. For example, yses confirmed our hypothesis that genomic variation has decreased mange has been implicated in the dissolution of previously stable since the initial reintroduction event, despite ephemeral fluctuations. packs, such as the Druid Peak pack (Figure 7; Almberg et al., 2012). 17524571, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eva.13127 by Colorado Parks & Wildlife, Wiley Online Library on [13/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 440 �441 Although two litters were born in April 2009, Druid wolves began should remain a priority during founder selection, reintroduction, to exhibit symptoms of mange infection soon after. By the end of and subsequent population management of at-risk populations. For October, the pack had lost alpha female 569F to intraspecific con- species harboring inbred genomes, we further recommend explora- flict, numerous wolves to dispersal or death, and all pups to mange tion of additional molecular mechanisms that may influence disease or its associated symptoms (Smith et al., 2009). Surviving members risk in the absence of genomic variation (such as gene regulation or fragmented into smaller groups in early 2010, and the exceptionally the host-associated microbiome). The integration of molecular and long tenure of the Druid Peak pack was over by year's end (Smith disease ecologies presents a powerful opportunity to elucidate the et al., 2011). Similar stories emerged from the Leopold, Everts, and factors underlying disease risk, as well as the evolutionary effects of Silver packs (Almberg et al., 2012Yellowstone National Park Wolf disease on wildlife. These insights can then inform best practices for Project Annual; Smith et al., 2011), emphasizing the scaling effects disease management and wildlife conservation. of mange infection on individuals, packs, and the greater YNP wolf population. AC K N OW L E D G E M E N T S As exemplified by the Druid Peak case study, mange-mediated We would like to thank Emily Almberg, Andrew Dobson, Kristin mortality and pack dissolution can impose strong selective pressures Brzeski, and Sarah Budischak for inspiring conversations at the out- on YNP wolves. Consistent with our expectations, analyses of allele set of this study, and Matthew Metz for providing data for annual frequency through time revealed signatures of selection acting on analyses. We thank the numerous Yellowstone Wolf Project techni- mange-associated loci. Similar effects have been seen in response cians who diligently collected observational and mange data on YNP to Mycoplasma galliseptum (Bonneaud et al., 2011), devil facial wolves. We would also like to thank Princeton University undergrad- tumor disease (Epstein et al., 2016), and chytridiomycosis (Savage & uate students Samantha Wu and Quin Pompi for assisting in the early Zamudio, 2016). In the present study, we observed significant reduc- stages of this work. Funding was provided by Princeton University's tions in the average frequency of alleles positively associated with Department of Ecology and Evolutionary Biology and Center for mange severity between 2006 and 2019 (after mange invaded the Health and Wellbeing. Research in Yellowstone National Park was park), with no change evident between 1995 and 2006. We addi- supported by funding from the National Science Foundation DEB- tionally observed evidence of selection increasing alleles negatively 0613730 and DEB-1245373. Funding was also received from the associated with mange severity between 2006 and 2019; however, Yellowstone Park Foundation (now Yellowstone Forever) and key credible intervals overlapped with the nonassociated group, which donors, especially Annie and Bob Graham, Valerie Gates, and Frank exhibited no change between 1995 and 2006 and between 2006 and Kay Yeager. This material is based upon work supported by the and 2019. Considered together, these results suggested that there National Science Foundation Graduate Research Fellowship under are stronger selective pressures acting to remove alleles associated Grant No. DGE1656466. with severe mange (i.e., positively associated alleles) than to increase the frequency of alleles associated with mild mange (i.e., negatively C O N FL I C T O F I N T E R E S T associated alleles). None declared. The observed relationship between host genomic variation and disease severity in YNP wolves at the individual and population lev- DATA AVA I L A B I L I T Y S TAT E M E N T els highlights the relevance of molecular variation to wildlife popula- Mapped bam files for the samples included in this study are avail- tions (Frankham, 2003). This is particularly important for host species able on NCBI’s public Sequence Read Archive under BioProjects threatened by disease, whether through epizootic outbreaks or the PRJNA577957 and PRJNA660734. Please see supplementary infor- slow invasion of enzootics (Daszak, Cunningham, & Hyatt, 2000). mation for metadata and BioSample Accession Numbers. These results support the paradigm that host genomic variation can buffer against disease risk, as seen in agriculture's monocul- ORCID ture effect. They further emphasize the importance of considering Alexandra L. DeCandia https://orcid.org/0000-0001-8485-5556 https://orcid.org/0000-0002-1793-6433 genome-wide variation and disease-relevant loci when studying Edward C. Schrom host–parasite dynamics, particularly in longer-evolved systems. Ellen E. Brandell https://orcid.org/0000-0002-2698-7013 Using YNP wolves as an example, declines in genome-wide variation Daniel R. Stahler https://orcid.org/0000-0002-8740-6075 through time may increase the likelihood of severe mange infections. Bridgett M. vonHoldt https://orcid.org/0000-0001-6908-1687 However, removal of harmful mange-associated alleles may counteract that risk. Monitoring summary metrics of diversity alongside REFERENCES disease-associated loci will enable more accurate risk assessment Abbas, A., Lichtman, A. H., & Pillai, S. (2016). Basic immunology: Functions and disorders of the immune system. New York, NY: Elsevier Inc. Aguilar, A., Roemer, G., Debenham, S., Binns, M., Garcelon, D., & Wayne, R. K. (2004). High MHC diversity maintained by balancing selection in an otherwise genetically monomorphic mammal. Proceedings of the National Academy of Sciences, 101(10), 3490–3494. https://doi. org/10.1073/pnas.03065​82101 and outbreak predictions, although YNP wolves are not currently treated or vaccinated against disease. In more heavily managed wildlife systems, similar analyses can directly inform conservation action. While further studies are needed to assess the universality of these trends, we posit that the maintenance of genetic variation 17524571, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eva.13127 by Colorado Parks & Wildlife, Wiley Online Library on [13/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | DECANDIA et al. �| Ali, O. A., O’Rourke, S. M., Amish, S. J., Meek, M. H., Luikart, G., Jeffres, C., & Miller, M. R. (2016). Rad capture (Rapture): Flexible and efficient sequence-based genotyping. Genetics, 202(2), 389–400. https://doi. org/10.1534/genet​ics.115.183665 Almberg, E. S., Cross, P. C., Dobson, A. P., Smith, D. W., & Hudson, P. J. (2012). Parasite invasion following host reintroduction: A case study of Yellowstone’s wolves. Philosophical Transactions of the Royal Society B, 367(1604), 2840–2851. https://doi.org/10.1098/rstb.2011.0369 Almberg, E. S., Cross, P. C., Dobson, A. P., Smith, D. W., Metz, M. C., Stahler, D. R., & Hudson, P. J. (2015). Social living mitigates the costs of a chronic illness in a cooperative carnivore. Ecology Letters, 18(7), 660–667. https://doi.org/10.1111/ele.12444 Almberg, E. S., Cross, P. C., Mech, L. D., Smith, D. W., Sheldon, J. W., & Crabtree, R. L. (2011). Infectious diseases in Yellowstone’s canid community. Yellowstone Science, 19(2), 16–24. Almberg, E. S., Cross, P. C., & Smith, D. W. (2010). Persistence of canine distemper virus in the Greater Yellowstone Ecosystem’s carnivore community. Ecological Applications, 20(7), 2058–2074. https://doi. org/10.1890/09-1225.1 Almberg, E. S., Mech, L. D., Smith, D. W., Sheldon, J. W., & Crabtree, R. L. (2009). A serological survey of infectious disease in Yellowstone National Park’s canid community. PLoS One, 4(9), e7042. https://doi. org/10.1371/journ​al.pone.0007042 Arbanasić, H., Huber, D., Kusak, J., Gomerčić, T., Hrenović, J., & Galov, A. (2013). Extensive polymorphism and evidence of selection pressure on major histocompatibility complex DLA-DRB1, DQA1 and DQB1 class II genes in Croatian grey wolves. Tissue Antigens, 81(1), 19–27. https://doi.org/10.1111/tan.12029 Arlian, L. G. (1989). Biology, host relations, and epidemiology of Sarcoptes scabiei. Annual Review of Entomology, 34, 139–161. https://doi. org/10.1146/annur​ev.en.34.010189.001035 Banks, S. C., Scheele, B. C., Macris, A., Hunter, D., Jack, C., & Fraser, C. I. (2020). Chytrid fungus infection in alpine tree frogs is associated with individual heterozygosity and population isolation but not population-genetic diversity. Frontiers of Biogeography, 12(1), e43875. https://doi.org/10.21425/​F5FBG​43875 Barker, J. N. W. N. (2001). Genetic aspects of psoriasis. Clinical Dermatology, 26, 321–325. https://doi.org/10.1159/00015​0964 Barnes, K. C. (2010). An update on the genetics of atopic dermatitis: Scratching the surface in 2009. Journal of Allergy and Clinical Immunology, 125(1), 16–29. https://doi.org/10.1016/j. jaci.2009.11.008 Bartoń, K. (2019). MuMIn v1.43.14. https://cran.r-proje​c t.org/web/ packa​ges/MuMIn/​index.html Batley, K. C., Sandoval-Castillo, J., Kemper, C. M., Attard, C. R. M., Zanardo, N., Tomo, I., … Möller, L. M. (2019). Genome-wide association study of an unusual dolphin mortality event reveals candidate genes for susceptibility and resistance to cetacean morbillivirus. Evolutionary Applications, 12(4), 718–732. https://doi.org/10.1111/ eva.12747 Benjamini, Y., & Hochberg, Y. (1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society, 57(1), 289–300. Benjamini, Y., & Yekutieli, D. (2001). The control of the false discovery rate in multiple testing under dependency. Annals of Statistics, 29(4), 1165–1188. https://doi.org/10.1214/aos/10136​99998 Bergstrom, C. T., & Antia, R. (2006). How do adaptive immune systems control pathogens while avoiding autoimmunity? Trends in Ecology & Evolution, 21(1), 22–28. https://doi.org/10.1016/j. tree.2005.11.008 Bin, L., & Leung, D. Y. M. (2016). Genetic and epigenetic studies of atopic dermatitis. Allergy, Asthma, and Clinical Immunology, 12, 52. https:// doi.org/10.1186/s1322​3-016-0158-5 Blanchong, J. A., Robinson, S. J., Samuel, M. D., & Foster, J. T. (2016). Application of genetics and genomics to wildlife epidemiology. DECANDIA et al. Journal of Wildlife Management, 80(4), 593–608. https://doi. org/10.1002/jwmg.1064 Bonneaud, C., Balenger, S. L., Russell, A. F., Zhang, J., Hill, G. E., & Edwards, S. V. (2011). Rapid evolution of disease resistance is accompanied by functional changes in gene expression in a wild bird. Proceedings of the National Academy of Sciences of the United States of America, 108(19), 7866–7871. https://doi.org/10.1073/pnas.10185​80108 Bornstein, S., Morner, T., & Samuel, W. (2001). Sarcoptes scabiei and Sarcoptic Mange. In W. M. Samuel, M. J. Pybus, & A. A. Kocan (Eds.), Parasitic diseases of wild mammals (pp. 107–119). Ames, IA: Iowa State University Press. Brambilla, A., Keller, L., Bassano, B., & Grossen, C. (2018). Heterozygosity– fitness correlation at the major histocompatibility complex despite low variation in Alpine ibex (Capra ibex). Evolutionary Applications, 11(5), 631–644. https://doi.org/10.1111/eva.12575 Brzeski, K. E., Harrison, R. B., Waddell, W. T., Wolf, K. N., Rabon, D. R., & Taylor, S. S. (2015). Infectious disease and red wolf conservation: Assessment of disease occurrence and associated risks. Journal of Mammalogy, 96(4), 751–761. https://doi.org/10.1093/jmamm​al/ gyv080 Bürkner, P. C. (2017). brms: An R package for Bayesian multilevel models using stan. Journal of Statistical Software, 80(1), 1–28. Candille, S. I., Kaelin, C. B., Cattanach, B. M., Yu, B., Thompson, D. A., Nix, M. A., … Barsh, G. S. (2007). A B-defensin mutation causes black coat color in domestic dogs. Science, 318, 1418–1424. Catchen, J., Hohenlohe, P. A., Bassham, S., Amores, A., & Cresko, W. A. (2013). Stacks: An analysis tool set for population genomics. Molecular Ecology, 22(11), 3124–3140. https://doi.org/10.1111/ mec.12354 Christensen, R. H. B. (2019). ordinal - Regression models for ordinal data. R package version 2019.12-10. Clayton, D. G. (2009). Sex chromosomes and genetic association studies. Genome Medicine, 1, 110. https://doi.org/10.1186/gm110 Coltman, D. W., Pilkington, J. G., Smith, J. A., & Pemperton, J. M. (1999). Parasite-mediated selection against inbred Soay sheep in a free-living, island population. Evolution, 53(4), 1259–1267. https://doi. org/10.2307/2640828 Courcet, J.-B., Elalaoui, S. C., Duplomb, L., Tajir, M., Rivière, J.-B., Thevenon, J., … Faivre, L. (2015). Autosomal-recessive SASH1 variants associated with a new genodermatosis with pigmentation defects, palmoplantar keratoderma and skin carcinoma. European Journal of Human Genetics, 23, 957–962. https://doi.org/10.1038/ ejhg.2014.213 Cross, P. C., Almberg, E. S., Haase, C. G., Hudson, P. J., Maloney, S. K., Metz, M. C., … Smith, D. W. (2016). Energetic costs of mange in wolves estimated from infrared thermography. Ecology, 97(8), 1938– 1948. https://doi.org/10.1890/15-1346.1 Cubaynes, S., MacNulty, D. R., Stahler, D. R., Quimby, K. A., Smith, D. W., & Coulson, T. (2014). Density-dependent intraspecific aggression regulates survival in northern Yellowstone wolves (Canis lupus). Journal of Animal Ecology, 83, 1344–1356. https://doi. org/10.1111/1365-2656.12238 Danecek, P., Auton, A., Abecasis, G., Albers, C. A., Banks, E., DePristo, M. A., … Durbin, R. (2011). The variant call format and VCFtools. Bioinformatics, 27(15), 2156–2158. https://doi.org/10.1093/bioin​ forma​tics/btr330 Daszak, P., Cunningham, A. A., & Hyatt, A. D. (2000). Emerging infectious diseases of wildlife - threats to biodiversity and human health. Science, 287, 443–449. https://doi.org/10.1126/scien​ ce.287.5452.443 DeCandia, A. L., Brzeski, K. E., Heppenheimer, E., Caro, C. V., Camenisch, G., Driscoll, C., & vonHoldt, B. M. (2019). The city-fox phenomenon revisited: Exploring urban colonization through multiple genetic lenses. Ecology and Evolution, 9(4), 2046–2060. https://doi. org/10.1002/ece3.4898 17524571, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eva.13127 by Colorado Parks & Wildlife, Wiley Online Library on [13/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 442 �DeCandia, A. L., Dobson, A. P., & vonHoldt, B. M. (2018). Toward an integrative molecular approach to wildlife disease. Conservation Biology, 32(4), 798–807. https://doi.org/10.1111/cobi.13083 DeCandia, A. L., Gaughran, S., Caragiulo, A., & Amato, G. (2016). A novel molecular method for noninvasive sex identification of order Carnivora. Conservation Genetics Resources, 8(2), 119–121. https:// doi.org/10.1007/s1268​6-016-0525-z DeCandia, A. L., Leverett, K. N., & vonHoldt, B. M. (2019). Of microbes and mange: Consistent changes in the skin microbiome of three canid species infected with Sarcoptes scabiei mites. Parasites and Vectors, 12(488), 488. https://doi.org/10.1186/s1307​ 1-019-3724-0 Donaldson, M. E., Davy, C. M., Willis, C. K. R., McBurney, S., Park, A., & Kyle, C. J. (2017). Profiling the immunome of little brown myotis provides a yardstick for measuring the genetic response to white-nose syndrome. Evolutionary Applications, 10(10), 1076–1090. https://doi. org/10.1111/eva.12514 Ekroth, A. K. E., Rafaluk-Mohr, C., & King, K. C. (2019). Host genetic diversity limits parasite success beyond agricultural systems: A meta-analysis. Proceedings of the Royal Society B, 286, 20191811. https:// doi.org/10.1098/rspb.2019.1811 Elbers, J. P., Brown, M. B., & Taylor, S. S. (2018). Identifying genome-wide immune gene variation underlying infectious disease in wildlife populations - a next generation sequencing approach in the gopher tortoise. BMC Genomics, 19, 64. https://doi.org/10.1186/s1286​ 4-018-4452-0 Ellison, A. R., Tunstall, T., DiRenzo, G. V., Hughey, M. C., Rebollar, E. A., Belden, L. K., … Zamudio, K. R. (2014). More than skin deep: Functional genomic basis for resistance to amphibian chytridiomycosis. Genome Biology and Evolution, 7(1), 286–298. https://doi. org/10.1093/gbe/evu285 Elton, C. S. (1958). The Ecology of Invasions by Animals and Plants. London: Methuen & Co., Ltd. Epstein, B., Jones, M., Hamede, R., Hendricks, S., McCallum, H., Murchison, E. P., … Storfer, A. (2016). Rapid evolutionary response to a transmissible cancer in Tasmanian devils. Nature Communications, 7, 12684. https://doi.org/10.1038/ncomm​s12684 Esaki, H., Ewald, D. A., Ungar, B., Rozenblit, M., Zheng, X., Xu, H., … Guttman-Yassky, E. (2015). Identification of novel immune and barrier genes in atopic dermatitis by means of laser capture microdissection. Journal of Allergy and Clinical Immunology, 135(1), 153–163. https://doi.org/10.1016/j.jaci.2014.10.037 Ezenwa, V. O., Archie, E. A., Craft, M. E., Hawley, D. M., Martin, L. B., Moore, J., … White, L. (2016). Host behaviour-parasite feedback: An essential link between animal behaviour and disease ecology. Proceedings of the Royal Society B, 283, 20153078. https://doi. org/10.1098/rspb.2015.3078 Fazal, S., Cheema, K. J., Maqbool, A., & Manzoor, F. (2014). Assessment of biological and physical environmental risk factors of sarcoptic mange in pet dogs. Pakistan Journal of Zoology, 46(2), 587–590. Feather, L., Gough, K., Flynn, R. J., & Elsheikha, H. M. (2010). A retrospective investigation into risk factors of sarcoptic mange in dogs. Parasitology Research, 107(2), 279–283. https://doi.org/10.1007/ s0043​6-010-1860-8 Frankham, R. (2003). Genetics and conservation biology. Comptes Rendus Biologies, 326, S22–S29. https://doi.org/10.1016/s1631​ -0691(03)00023​- 4 Frankham, R. (2005). Genetics and extinction. Biological Conservation, 126(2), 131–140. https://doi.org/10.1016/j.biocon.2005.05.002 Galaverni, M., Caniglia, R., Fabbri, E., Lapalombella, S., & Randi, E. (2013). MHC variability in an isolated wolf population in Italy. Journal of Heredity, 104(5), 601–612. https://doi.org/10.1093/jhere​d/est045 Hadj-Rabia, S., Callewaert, B. L., Bourrat, E., Kempers, M., Plomp, A. S., Layet, V., … Bodemer, C. (2013). Twenty patients including 7 probands with autosomal dominant cutis laxa confirm clinical and | 443 molecular homogeneity. Orphanet Journal of Rare Diseases, 8, 36. https://doi.org/10.1186/1750-1172-8-36 Hawley, D. M., Sydenstricker, K. V., Kollias, G. V., & Dhondt, A. A. (2005). Genetic diversity predicts pathogen resistance and immunocompetence in house finches. Biology Letters, 1, 326–329. https://doi. org/10.1098/rsbl.2005.0303 He, W., Li, A. G., Wang, D., Han, S., Zheng, B., Goumans, M.-J., … Wang, X.J. (2002). Overexpression of Smad7 results in severe pathological alterations in multiple epithelial tissues. The European Molecular Biology Organization Journal, 21(11), 2580–2590. https://doi.org/10.1093/ emboj/​21.11.2580 Hedrick, P. (1999). Balancing selection and MHC. Genetica, 104(3), 207–214. Hedrick, P. W., Lee, R. N., & Garrigan, D. (2002). Major histocompatibility complex variation in red wolves: Evidence for common ancestry with coyotes and balancing selection. Molecular Ecology, 11(10), 1905– 1913. https://doi.org/10.1046/j.1365-294X.2002.01579.x Hedrick, P., Lee, R., & Parker, K. (2000). Major histocompatibility complex (MHC) variation in the endangered Mexican wolf and related canids. Heredity, 85, 617–624. https://doi. org/10.1046/j.1365-2540.2000.00805.x Hengge, U. R., Currie, B. J., Jäger, G., Lupi, O., & Schwartz, R. A. (2006). Scabies: A ubiquitous skin disease. The Lancet Infectious Diseases, 6(12), 769–779. Holand, H., Jensen, H., Tufto, J., Pärn, H., Sæther, B. E., & Ringsby, T. H. (2015). Endoparasite infection has both short- and long-term negative effects on reproductive success of female house sparrows, as revealed by faecal parasitic egg counts. PLoS One, 10(5), e0125773. https://doi.org/10.1371/journ​al.pone.0125773 Kennedy, L. J., Randall, D. A., Knobel, D., Brown, J. J., Fooks, A. R., Argaw, K., … Laurenson, M. K. (2011). Major histocompatibility complex diversity in the endangered Ethiopian wolf (Canis simensis). Tissue Antigens, 77, 118–125. https://doi. org/10.1111/j.1399-0039.2010.01591.x King, K. C., & Lively, C. M. (2012). Does genetic diversity limit disease spread in natural host populations? Heredity, 109, 199–203. https:// doi.org/10.1038/hdy.2012.33 Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N., … Durbin, R. (2009). The sequence alignment/map format and SAMtools. Bioinformatics, 25(16), 2078–2079. https://doi.org/10.1093/bioin​ forma​tics/btp352 Liang, Y., Chang, C., & Lu, Q. (2016). The genetics and epigenetics of atopic dermatitis filaggrin and other polymorphisms. Clinical Reviews in Allergy and Immunology, 51(3), 315–328. https://doi.org/10.1007/ s1201​6-015-8508-5 Lindblad-Toh, K., Wade, C. M., Mikkelsen, T. S., Karlsson, E. K., Jaffe, D. B., Kamal, M., … Lander, E. S. (2005). Genome sequence, comparative analysis and haplotype structure of the domestic dog. Nature, 438(7069), 803–819. https://doi.org/10.1038/natur​e 04338 Little, S. E., Davidson, W. R., Rakich, P. M., Nixon, T. L., Bounous, D. I., & Nettles, V. F. (1998). Responses of red foxes to first and second infection with Sarcoptes scabiei. Journal of Wildlife Diseases, 34(3), 600–611. https://doi.org/10.7589/0090-3558-34.3.600 Lively, C. M. (2010). The effect of host genetic diversity on disease spread. The American Naturalist, 175(6), 98–101. https://doi. org/10.1086/652430 Lüdecke, D., Makowski, D., Waggoner, P., & Patil, I. (2020). performance: Assessment of regression models performance v0.4.5. https://cran.rproje​c t.org/web/packa​ges/perfo​rmanc Lukens, J. R., Vogel, P., Johnson, G. R., Kelliher, M. A., Iwakura, Y., Lamkanfi, M., & Kanneganti, T. D. (2013). RIP1-driven autoinflammation targets IL-1α independently of inflammasomes and RIP3. Nature, 498(7453), 224–227. https://doi.org/10.1038/natur​e12174 Luong, L. T., Heath, B. D., & Polak, M. (2007). Host inbreeding increases susceptibility to ectoparasitism. European 17524571, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eva.13127 by Colorado Parks & Wildlife, Wiley Online Library on [13/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License DECANDIA et al. �| Society for Evolutionary Biology, 20, 79–86. https://doi. org/10.1111/j.1420-9101.2006.01226.x Margres, M. J., Jones, M. E., Epstein, B., Kerlin, D. H., Hendricks, S. A., Comte, S., … Storfer, A. (2018). Large-effect loci affect survival in Tasmanian devils (Sarcophilus harrisii) infected with a transmissible cancer. Molecular Ecology, 27, 4189–4199. https://doi.org/10.1111/ mec.14853 Marshall, H. D., Langille, B. L., Hann, C. A., & Whitney, H. G. (2016). Patterns of MHC-DRB1 polymorphism in a post-glacial island canid, the Newfoundland red fox (Vulpes vulpes deletrix), suggest balancing selection at species and population timescales. Immunogenetics, 68(5), 381–389. https://doi.org/10.1007/s0025​ 1-016-0907-0 Martin, A. M., Fraser, T. A., Lesku, J. A., Simpson, K., Roberts, G. L., Garvey, J., … Carver, S. (2018). The cascading pathogenic consequences of Sarcoptes scabiei infection that manifest in host disease. Royal Society Open Science, 5(4), 180018. https://doi.org/10.1098/ rsos.180018 Maruki, T., & Lynch, M. (2017). Genotype calling from population-genomic sequencing data. G3 Genes, Genomes, Genetics, 7, 1393–1404. https://doi.org/10.1534/g3.117.039008 Marzal, A., De Lope, F., Navarro, C., & Møller, A. P. (2005). Malarial parasites decrease reproductive success: An experimental study in a passerine bird. Oecologia, 142(4), 541–545. https://doi.org/10.1007/ s0044​2-004-1757-2 Matsumoto, M., Zhou, Y., Matsuo, S., Nakanishi, H., Hirose, K., Oura, H., … Chaplin, D. (2008). Targeted deletion of the murine corneodesmosin gene delineates its essential role in skin and hair physiology. Proceedings of the National Academy of Sciences, 105(18), 6720–6724. https://doi.org/10.1073/pnas.07093​45105 Mazerolle, M. J. (2019). AICcmodavg: Model selection and multimodel inference based on (Q)AIC(c). R package version 2.2-2. McLaren, W., Gil, L., Hunt, S. E., Riat, H. S., Ritchie, G. R. S., Thormann, A., … Cunningham, F. (2016). The ensembl variant effect predictor. Genome Biology, 17(1), 1–14. https://doi.org/10.1186/s1305​ 9-016-0974-4 Mech, L. D., & Boitani, L. (2003). Wolf social ecology. In L. D. Mech, & L. Boitani (Eds.), Wolves: behavior, ecology, and conservation. Chicago, IL: University of Chicago Press. Møller, A. P. (2002). Temporal change in mite abundance and its effect on barn swallow reproduction and sexual selection. Journal of Evolutionary Biology, 15(3), 495–504. https://doi. org/10.1046/j.1420-9101.2002.00386.x Montecino-Latorre, D., Cypher, B. L., Rudd, J. L., Clifford, D. L., Mazet, J. A. K., & Foley, J. E. (2019). Assessing the role of dens in the spread, establishment and persistence of sarcoptic mange in an endangered canid. Epidemics, 27, 28–40. https://doi.org/10.1016/j. epidem.2019.01.001 Nattkemper, L. A., Tey, H. L., Valdes-Rodriguez, R., Lee, H., Mollanazar, N. K., Albornoz, C., … Yosipovitch, G. (2018). The genetics of chronic itch: Gene expression in the skin of patients with atopic dermatitis and psoriasis with severe itch. Journal of Investigative Dermatology, 138(6), 1311–1317. https://doi.org/10.1016/j.jid.2017.12.029 Newman, T. J., Baker, P. J., & Harris, S. (2002). Nutritional condition and survival of red foxes with sarcoptic mange. Canadian Journal of Zoology, 80, 154–161. https://doi.org/10.1139/z01-216 Niedringhaus, K. D., Brown, J. D., Sweeley, K. M., & Yabsley, M. J. (2019). A review of sarcoptic mange in North American wildlife. International Journal for Parasitology: Parasites and Wildlife, 9, 285–297. https://doi. org/10.1016/j.ijppaw.2019.06.003 Nimmervoll, H., Hoby, S., Robert, N., Lommano, E., Welle, M., & RyserDegiorgis, M. P. (2013). Pathology of sarcoptic mange in red foxes (Vulpes vulpes): Macroscopic and histologic characterization of three disease stages. Journal of Wildlife Diseases, 49(1), 91–102. https://doi. org/10.7589/2010-11-316 DECANDIA et al. Oji, V., Eckl, K.-M., Aufenvenne, K., Natebus, M., Tarinski, T., Ackermann, K., … Hennies, H. C. (2010). Loss of corneodesmosin leads to severe skin barrier defect, pruritus, and atopy: Unraveling the peeling skin disease. The American Journal of Human Genetics, 87, 274–281. https://doi.org/10.1016/j.ajhg.2010.07.005 Oleaga, A., Casais, R., Balseiro, A., Espi, A., Llaneza, L., Hartasanchez, A., & Gortazar, C. (2011). New techniques for an old disease: Sarcoptic mange in the Iberian wolf. Veterinary Parasitology, 181(2–4), 255–266. https://doi.org/10.1016/j.vetpar.2011.04.036 Oleaga, A., Casais, R., Prieto, J. M., Gortázar, C., & Balseiro, A. (2012). Comparative pathological and immunohistochemical features of sarcoptic mange in five sympatric wildlife species in Northern Spain. European Journal of Wildlife Research, 58(6), 997–1000. https://doi. org/10.1007/s1034​4-012-0662-y Online Mendelian Inheritance in Man, OMIM®. Baltimore, MD: McKusickNathans Institute of Genetic Medicine, Johns Hopkins University. https://omim.org/ Parratt, S. R., Numminen, E., & Laine, A. (2016). Infectious disease dynamics in heterogenous landscapes. Annual Review of Ecology, Evolution, and Systematics, 47(1), 283–306. https://doi.org/10.1146/ annur​ev-ecols​ys-12141​5-032321 Pence, D. B., & Ueckermann, E. (2002). Sarcoptic mange in wildlife. Revue Scientifique Et Technique (International Office of Epizootics), 21, 385– 398. https://doi.org/10.20506/​rst.21.2.1335 Pence, D. B., Windberg, L. A., & Sprowls, R. (1983). The epizootiology and pathology of sarcoptic mange in coyotes, Canis latrans, from south Texas. The Journal of Parasitology, 69(6), 1100–1115. https:// doi.org/10.2307/3280873 Phillips, M. K., & Smith, D. W. (1997). Yellowstone wolf project: biennial report 1995 and 1996. National Park Service, Yellowstone Center for Resources, Yellowstone National Park, Wyoming, YCR-NR-97-4. Pink, A. E., Simpson, M. A., Brice, G. W., Smith, C. H., Desai, N., Mortimer, P. S., … Trembath, R. C. (2011). PSENEN and NCSTN mutations in familial hidradenitis suppurativa (acne inversa). Journal of Investigative Dermatology, 131(7), 1568–1570. https://doi.org/10.1038/ jid.2011.42 Purcell, S., Neale, B., Todd-Brown, K., Thomas, L., Ferreira, M. A. R., Bender, D., … Sham, P. C. (2007). PLINK: A tool set for whole-genome association and population-based linkage analyses. American Journal of Human Genetics, 81(3), 559–575. https://doi.org/10.1086/519795 Quraishi, B. M., Zhang, H., Everson, T. M., Ray, M., Lockett, G. A., Holloway, J. W., … Karmaus, W. (2015). Identifying CpG sites associated with eczema via random forest screening of epigenome-scale DNA methylation. Clinical Epigenetics, 7(1), 68. https://doi.org/10.1186/s1314​ 8-015-0108-y Raudvere, U., Kolberg, L., Kuzmin, I., Arak, T., Adler, P., Peterson, H., & Vilo, J. (2019). g:Profiler: A web server for functional enrichment analysis and conversions of gene lists (2019 update). Nucleic Acids Research, 47(W1), W191–W198. https://doi.org/10.1093/nar/gkz369 Reiss, E. R., & Drinkwater, L. E. (2018). Cultivar mixtures: A meta-analysis of the effect of intraspecific diversity on crop yield. Ecological Applications, 28(1), 62–77. https://doi.org/10.1002/eap.1629 Rodríguez, E., Baurecht, H., Wahn, A. F., Kretschmer, A., Hotze, M., Zeilinger, S., … Weidinger, S. (2014). An integrated epigenetic and transcriptomic analysis reveals distinct tissue-specific patterns of DNA methylation associated with atopic dermatitis. Journal of Investigative Dermatology, 134(7), 1873–1883. https://doi. org/10.1038/jid.2014.87 Samuel, W. M. (1981). Attempted experimental transfer of sarcoptic mange (Sarcoptes scabiei, Acarina: Sarcoptidae) among red fox, coyote, wolf and dog. Journal of Wildlife Diseases, 17(3), 343–347. https:// doi.org/10.7589/0090-3558-17.3.343 Sarasa, M., Serrano, E., Soriguer, R. C., Granados, J.-E., Fandos, P., Gonzalez, G., … Pérez, J. M. (2011). Negative effect of the arthropod parasite, Sarcoptes scabiei, on testes mass in Iberian ibex, Capra 17524571, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eva.13127 by Colorado Parks & Wildlife, Wiley Online Library on [13/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 444 �pyrenaica. Veterinary Parasitology, 175(3–4), 306–312. https://doi. org/10.1016/j.vetpar.2010.10.024 Savage, A. E., & Zamudio, K. R. (2016). Adaptive tolerance to a pathogenic fungus drives major histocompatibility complex evolution in natural amphibian populations. Proceedings of the Royal Society B, 283(1827), 20153115. https://doi.org/10.1098/rspb.2015.3115 Schmuth, M., Martinz, V., Janecke, A. R., Fauth, C., Schossig, A., Zschocke, J., & Gruber, R. (2013). Inherited ichthyoses/generalized Mendelian disorders of cornification. European Journal of Human Genetics, 21, 123–133. https://doi.org/10.1038/ejhg.2012.121 Silk, M. J., Hodgson, D. J., Rozins, C., Croft, D. P., Delahay, R. J., Boots, M., & McDonald, R. A. (2019). Integrating social behaviour, demography and disease dynamics in network models: Applications to disease management in declining wildlife populations. Philosophical Transactions of the Royal Society B, 374(1781), 20180211. https://doi. org/10.1098/rstb.2018.0211 Smallbone, W., Oosterhout, C. V., & Cable, J. (2016). The effects of inbreeding on disease susceptibility: Gyrodactylus turnbulli infection of guppies, Poecilia reticulata. Experimental Parasitology, 167, 32–37. https://doi.org/10.1016/j.exppa​ra.2016.04.018 Smith, D. W., Stahler, D. R., Albers, E., McIntyre, R., Metz, M., Cassidy, K., … Bowersock, N. (2009). Yellowstone Wolf Project: Annual Report, 2009. National Park Service, Yellowstone Center for Resources, Yellowstone National Park, Wyoming, YCR-2010-06. Smith, D., Stahler, D., Albers, E., McIntyre, R., Metz, M., Irving, J., … Bowersock, N. (2011). Yellowstone Wolf Project: Annual Report, 2010. YCR-2011-06. National Park Service, Yellowstone National Park, Yellowstone Center for Resources, Yellowstone National Park, Wyoming. Smith, D., Stahler, D., Cassidy, K., Stahler, E., Metz, M., Meyer, C., … Loggers, E. (2020). Yellowstone National Park Wolf Project Annual Report 2019. National Park Service, Yellowstone Center for Resources, Yellowstone National Park, WY, USA, YCR-2020-01. Sonkoly, E., Janson, P., Majuri, M. L., Savinko, T., Fyhrquist, N., Eidsmo, L., … Pivarcsi, A. (2010). MiR-155 is overexpressed in patients with atopic dermatitis and modulates T-cell proliferative responses by targeting cytotoxic T lymphocyte-associated antigen 4. Journal of Allergy and Clinical Immunology, 126(3), 581–589. https://doi.org/10.1016/j. jaci.2010.05.045 Spielman, D., Brook, B. W., Briscoe, D. A., & Frankham, R. (2004). Does inbreeding and loss of genetic diversity decrease disease resistance? Conservation Genetics, 5(4), 439–448. https://doi.org/10.1023/ B:COGE.00000​41030.76598.cd Spielman, D., Brook, B. W., & Frankham, R. (2004). Most species are not driven to extinction before genetic factors impact them. Proceedings of the National Academy of Sciences, 101(42), 15261–15264. https:// doi.org/10.1073/pnas.04038​09101 Stahler, D. R., Macnulty, D. R., Wayne, R. K., vonHoldt, B. M., & Smith, D. W. (2013). The adaptive value of morphological, behavioural and life-history traits in reproductive female wolves. Journal of Animal Ecology, 82(1), 222–234. https://doi. org/10.1111/j.1365-2656.2012.02039.x Steinke, J. W., Borish, L., & Rosenwasser, L. (2003). Genetics of hypersensitivity. Journal of Allergy and Clinical Immunology, 3(2), S495–S501. Szpiech, Z. A., Jakobsson, M., & Rosenberg, N. A. (2008). ADZE: A rarefaction approach for counting alleles private to combinations of populations. Bioinformatics, 24(21), 2498–2504. https://doi.org/10.1093/ bioin​forma​tics/btn478 445 Townsend, A. K., Taff, C. C., Wheeler, S. S., Weis, A. M., Hinton, M. G., Jones, M. L., … Boyce, W. M. (2018). Low heterozygosity is associated with vector-borne disease in crows. Ecosphere, 9(10), e02407. https://doi.org/10.1002/ecs2.2407 Trivedi, S. G., Newson, J., Rajakariar, R., Jacques, T. S., Hannon, R., Kanaoka, Y., … Gilroy, D. W. (2006). Essential role for hematopoietic prostaglandin D2 synthase in the control of delayed type hypersensitivity. Proceedings of the National Academy of Sciences, 103(13), 5179–5184. https://doi.org/10.1073/pnas.05071​75103 van den Brink, D. M., Brites, P., Haasjes, J., Wierzbicki, A. S., Mitchell, J., Lambert-Hamill, M., … Wanders, R. J. A. (2003). Identification of PEX7 as the second gene involved in Refsum disease. American Journal of Human Genetics, 72, 471–477. https://doi. org/10.1086/346093 vonHoldt, B. M., DeCandia, A. L., Heppenheimer, E., Janowitz-Koch, I., Shi, R., Zhou, H., … Sinsheimer, J. S. (2020). Heritability of inter-pack aggression in a wild pedigreed population of North American gray wolves. Molecular Ecology, 29(10), 1764–1775. https://doi. org/10.1111/mec.15349 vonHoldt, B. M., Heppenheimer, E., Petrenko, V., Croonquist, P., & Rutledge, L. Y. (2017). Ancestry-specific methylation patterns in admixed offspring from an experimental coyote and gray wolf cross. Journal of Heredity, 108(4), 341–348. https://doi.org/10.1093/jhere​ d/esx004 vonHoldt, B. M., Stahler, D., Bangs, E., Smith, D., Jimenez, M., Mack, C., … Wayne, R. (2010). A novel assessment of population structure and gene flow in grey wolf populations of the Northern Rocky Mountains of the United States. Molecular Ecology, 19(20), 4412–4427. https:// doi.org/10.1111/j.1365-294X.2010.04769.x vonHoldt, B. M., Stahler, D. R., Smith, D. W., Earl, D. A., Pollinger, J. P., & Wayne, R. K. (2008). The genealogy and genetic viability of reintroduced Yellowstone grey wolves. Molecular Ecology, 17(1), 252–274. https://doi.org/10.1111/j.1365-294X.2007.03468.x Walton, S. F., Holt, D. C., Currie, B. J., & Kemp, D. J. (2011). Scabies: New future for a neglected disease. Advances in Parasitology, 57, 309–377. https://doi.org/10.1016/B978-0-12-38769​2-8.00016​-3 Zhou, X., & Stephens, M. (2012). Genome-wide efficient mixed-model analysis for association studies. Nature Genetics, 44(7), 821–824. https://doi.org/10.1038/ng.2310 Zhou, X., & Stephens, M. (2014). Efficient multivariate linear mixed model algorithms for genome-wide association studies. Nature Methods, 11(4), 407–409. https://doi.org/10.1038/nmeth.2848 S U P P O R T I N G I N FO R M AT I O N Additional supporting information may be found online in the Supporting Information section. How to cite this article: DeCandia AL, Schrom EC, Brandell EE, Stahler DR, vonHoldt BM. Sarcoptic mange severity is associated with reduced genomic variation and evidence of selection in Yellowstone National Park wolves (Canis lupus). Evol Appl. 2021;14:429–445. https://doi.org/10.1111/eva.13127 17524571, 2021, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eva.13127 by Colorado Parks & Wildlife, Wiley Online Library on [13/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | DECANDIA et al. � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Journal Articles Description An account of the resource CPW peer-reviewed journal publications Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Sarcoptic mange severity is associated with reduced genomic variation and evidence of selection in Yellowstone National Park wolves (Canis lupus) Description An account of the resource Population genetic theory posits that molecular variation buffers against disease risk. Although this “monoculture effect” is well supported in agricultural settings, its applicability to wildlife populations remains in question. In the present study, we examined the genomics underlying individual-level disease severity and population-level consequences of sarcoptic mange infection in a wild population of canids. Using gray wolves (Canis lupus) reintroduced to Yellowstone National Park (YNP) as our focal system, we leveraged 25 years of observational data and biobanked blood and tissue to genotype 76,859 loci in over 400 wolves. At the individual level, we reported an inverse relationship between host genomic variation and infection severity. We additionally identified 410 loci significantly associated with mange severity, with annotations related to inflammation, immunity, and skin barrier integrity and disorders. We contextualized results within environmental, demographic, and behavioral variables, and confirmed that genetic variation was predictive of infection severity. At the population level, we reported decreased genome-wide variation since the initial gray wolf reintroduction event and identified evidence of selection acting against alleles associated with mange infection severity. We concluded that genomic variation plays an important role in disease severity in YNP wolves. This role scales from individual to population levels, and includes patterns of genome-wide variation in support of the monoculture effect and specific loci associated with the complex mange phenotype. Results yielded system-specific insights, while also highlighting the relevance of genomic analyses to wildlife disease ecology, evolution, and conservation. Bibliographic Citation A bibliographic reference for the resource. Recommended practice is to include sufficient bibliographic detail to identify the resource as unambiguously as possible. DeCandia, A. L., E. C. Schrom, E. E. Brandell, D. R. Stahler, and B. M. vonHoldt. 2021. Sarcoptic mange severity is associated with reduced genomic variation and evidence of selection in Yellowstone National Park wolves (Canis lupus). Evolutionary applications, 14(2):429-445, https://doi.org/10.1111/eva.13127 Creator An entity primarily responsible for making the resource DeCandia, Alexandra L. Schrom, Edward C. Brandell, Ellen E. Stahler, Daniel R. vonHoldt, Bridgett M. Subject The topic of the resource Ectoparasitism Infection severity Mite infestations Natural selection RAD-sequencing Sarcoptic mange Extent The size or duration of the resource. 17 pages Rights Information about rights held in and over the resource <a href="http://rightsstatements.org/vocab/InC-NC/1.0/">IN COPYRIGHT - NON-COMMERCIAL USE PERMITTED</a> Format The file format, physical medium, or dimensions of the resource application/pdf Language A language of the resource English Is Part Of A related resource in which the described resource is physically or logically included. Evolutionary Applications Date Accepted Date of acceptance of the resource. Examples of resources to which a Date Accepted may be relevant are a thesis (accepted by a university department) or an article (accepted by a journal). 2020/08/25 Date Issued Date of formal issuance (e.g., publication) of the resource. 2020/09/10 Date Modified Date on which the resource was changed. 2020/07/28 Date Submitted Date of submission of the resource. Examples of resources to which a Date Submitted may be relevant are a thesis (submitted to a university department) or an article (submitted to a journal). 2020/06/03 Type The nature or genre of the resource Article