Discussion
Our study provides evidence that P. ochraceus may show genetic variation for tolerance of or resilience to SSWS. By comparing allele frequencies between symptomatic and asymptomatic individuals during an outbreak, we detected significant differences across multiple genomic regions, consistent with a polygenic basis for this phenotype. Based on Fst analyses in individual SNP loci, three loci were detected to be under selection in at least one of the three geographic populations sampled. Multivariate analyses examining genome-wide changes revealed differences between symptomatic and asymptomatic sea stars, and 10 of these differences occurred in more than one population. Our study is the second to detect a genomic signal consistent with selection caused by a strong epizootic event of SSWS starting in 2013.
Schielbelhut et al. (2018) detected allelic shifts in healthy P. ochraceus adults and juveniles before and after the SSWS outbreak in California. They found three loci putatively under selection and reported on 100 discriminatory haplotypes between time periods. Interestingly, the 87 SNP loci we detected as most discriminatory in our samples did not occur within 10 kb of those reported by Schielbelhut et al. (2018). The nature of the sampling between studies, which differ in time period, the health status of the sea stars, and geographic location, likely contributes to differences in power to detect associated genetic variants. For instance, elevated sea water temperature can lead to additional stress responses to SSWS (Eisenlord et al., 2016; Kohl et al., 2016). Differences in thermal stress between the two regions may also cause different selective pressures. For instance, daily deviation from annual sea water temperatures in Oregon were more prevalent in 2013/2014 than in California (Miner et al., 2018). Differences in allele frequencies between Schielbelhut’s and our study could indicate the differences in synergistic selective pressures from SSWS and temperature. Additionally, we analyzed our genetic data for each subpopulation separately, detecting a signal of population structure. Schielbelhut et al. (2018) did not detect significant structure and performed a region-wide analysis. By accounting for subpopulation structure, we were able to isolate more genes that would have been confounded in a full population wide study.
While our DAPC analysis identified 85 outlier loci in total, we proceeded conservatively by further requiring that different populations shared neighboring loci (within 1 Mbp). The 10 clusters identified harbor over 100 predicted protein-coding genes. With the current dataset, we cannot identify regions more specifically, and hence many of these genes are likely linked to relevant alleles. We argue that these regions may have small but cumulative effects of adaptive genetic variation associated with SSWS tolerance or resistance, which is expected for a polygenic trait. This trait may require higher powered studies for ascertaining causative loci with more precision (Gagnaire & Gaggiotti, 2016). Nevertheless, these functional regions may serve as candidates in future studies of response to SSWS challenges, such as differential gene expression or monitoring of allele frequency changes. In fact, differential gene expression between symptomatic and asymptomatic individuals were related to innate immunity and tissue remodeling (Fuess et al., 2015; Gudenkauf & Hewson, 2015; Ruiz‐Ramos et al., 2020). Many of these genes are related to stress response, defensive apoptosis and tissue degradation. Echinoderms have complex immune system that allows for innate immunity through primitive immune memory and cytokine-like systems (Mydlarz, Jones, & Harvell, 2006). In addition, recent studies suggest that microbial activity on the diffusive boundary layer between water and animal could reduce oxygen uptake (Aquino et al., 2021), suggesting metabolic genes may be also play important roles in this response. We did not see similar genes specifically related to Ruiz-Ramos et al. (2020), suggesting other mechanisms other than immunity could be important to SSWS resistance, such as metabolic processes in response to hypoxic conditions (Aquino et al., 2021). Genes with small effects found in our study may aid in cellular processes related to SSWS resistance.
Assessing marine diseases is difficult due to large gaps in knowledge regarding how pathogens disperse and propagate in marine systems (MCCALLUM et al., 2004). The vast majority of our understanding of wildlife diseases comes from terrestrial systems (Harvell et al., 2004). In addition, there is higher taxonomic diversity of hosts and pathogens in marine systems. The complexity of marine diseases is likely a culprit as to why isolating the causation of SSWS remains to be unresolved (Eisenlord et al., 2016). Mitigation techniques for addressing outbreaks when a causative agent is unknown should run parallel with studies attempting to determine the cause (Groner et al., 2016). With the rising sea water temperatures resulting in the higher prevalence of marine diseases (Harvell et al., 2002; Tracy, Pielmeier, Yoshioka, Heron, & Harvell, 2019), we are likely to see similar scenarios of disease outbreaks impacting marine species more frequently and having little time to address management or conservation plans.
One method of addressing our lack of knowledge on SSWS and other disease outbreaks is to determine whether there is a potential for the organism of interest to adapt to the syndrome. For example, selectively rearing disease-resistant oysters in hatcheries has been a useful tool in avoiding disease outbreaks that are decimating the wild? populations (Agnew et al., 2020; Dégremont, Garcia, & Allen, 2015). Selectively breeding programs in both the eastern and Pacific oyster,Crassostrea virginica and Crassostrea gigas , has been a useful tool in producing disease tolerant lines. Of course, not all marine species are amenable for selective breeding. Assessing genomic variation in natural populations can address whether these species have the genetic makeup for adaptation to marine diseases on their own. Our dataset is hence valuable in that captured phenotypic and genetic variation during a period of high SSWS incidence.
Our study joins that of Schielbelhut et al. (2018) to show that P. ochraceus has the potential to adapt to SSWS challenges. A number of candidate genes were isolated that may aid in SSWS resistance, although none of these appear to function via changes in gene expression associated with SSWS infection. Assessing genetic variation to address susceptibility and adaptative potential should continue to run parallel with studies that try to determine the causative agent for SSWS and other marine disease. These studies should also be done both in locally as well as across geographic regions as we have seen variation in how local populations respond differently.