Discussion
Microsatellites are a useful tool to rapidly acquire genetic information that informs conservation, breeding, and management decisions. They are especially informative for identifying population genetic structure among subpopulations and have the capacity to infer sibship relationships among individuals within a subpopulation (Koch, McCabe, Love, & Cox-Foster, 2021; Van Eeckhoven et al., 2022). While the front-end of microsatellite development can be expensive, using established microsatellites is a cost-effective method for population genetics studies relative to reduced representation genome sequencing on a per-specimen basis in the market today (Guichoux et al., 2011). Microsatellite analyses have uncovered cryptic genetic diversity among different bee species (Koch, Rodriguez, Pitts, & Strange, 2018), and have been useful markers for correlating biological phenomena such as population declines and genotype-by-environment associations (Cameron et al., 2011; Koch, Vandame, et al., 2018; Pitts-Singer, Cane, & Trostle, 2014). Furthermore, microsatellite marker development is on the rise across agriculturally important solitary bee species (Neumann & Seidelmann, 2006; Strange et al., 2017; Van Eeckhoven et al., 2022). In this study, we expand microsatellite marker availability and utility in the solitary bee genus Osmia , with special focus on O. lignaria .
Our identification and filtering strategy using the O. lignariagenome assembly uncovered a high volume of candidate microsatellite loci that can be tested using additional molecular techniques. We present those candidate loci in Supplementary Tables 1 - 3. In our study, we failed to find many perfect microsatellites that resulted in a product size greater than 241 nts under our filtering strategy. However, we were able to identify sizes exceeding 241 nts when we further examined imperfect microsatellite loci. An imperfect microsatellite is the result of a mutation at a microsatellite locus that often creates an imperfect motif (Behura & Severson, 2015). Imperfect microsatellites are suspected to be more stable than perfect microsatellites as they are less likely to incur slippage mutations (Sturzeneker, Haddad, Bevilacqua, Simpson, & Pena, 1998). Interestingly, in our study of O. lignaria , we found significantly more genotyping error in the perfect microsatellites relative to the imperfect microsatellites. Based on this result, we suggest that further microsatellite development studies should examine genotype error in the context of microsatellite type when identifying loci for population genetic studies. Markers that succumb to high genotyping error will likely result in deviations from HWE and should be removed from population genetic analyses (Hoffman & Amos, 2005; Morin et al., 2009).
Our characterization and assessment of novel O. lignariamicrosatellites underscores the importance of including genotype error rates in population genetic studies. Genotyping error can significantly impact estimates of genetic diversity, population structure, and sibship relationships (Hoffman & Amos, 2005). In turn, these errors impact the interpretation of genetic data and can lead to poor management decisions for livestock and wildlife if not controlled. Genotyping error in a microsatellite analysis is the product of diverse phenomena including poor quality of template DNA, allelic dropout, and misprinting (Hoffman & Amos, 2005). Even in cases where well established microsatellite markers are used for population genetics studies (Hoffman & Amos, 2005), poor DNA template quality has caused 50% of the genotyping error (Morin et al., 2009). This phenomenon underscores the importance of performing pilot studies and strategic re-amplification of samples when conducting genetic analyses (Hoffman & Amos, 2005).
The goal of this study was to present novel microsatellite markers forO. lignaria and characterize their utility. Application of the microsatellite markers on two Intermountain North America populations ofO. lignaria in Idaho found no differences in population genetic diversity across populations and low, but significant, population structure. Furthermore, we found no significant difference betweenHO and Nei’s HE in both populations, and thereby, provide no evidence to suggest inbreeding is occurring within either population. This inference is further supported by low FI S values and permutation tests. We estimated that ~60% of the alleles (1 - [Uniq. alleles/No. Alleles]) identified in the study are shared between both populations. Combined, these results suggest that it is likely that contemporary dispersal (i.e., gene flow) is taking place across populations. This is not surprising as the populations studied are ~30 km apart and native residents of the Bear River Mountains (Fig. 1). Furthermore, sibship analysis provides evidence for full sibling families within each population. Thus, the novel microsatellites have the capacity to identify genetic lineages/families within a population, which is important information for biological research such as the characterization of nest founding behaviors and offspring survival (Tepedino & Torchio, 1994).
Our assessment on the utility of the novel microsatellites in otherOsmia species supports future population genetic studies of the genus. To date, microsatellite markers have been developed for O. bicornis (Neumann & Seidelmann, 2006; Van Eeckhoven et al., 2022), which is in the same clade (bicornis clade) as O. lignaria . The subgenusOsmia , especially the clade bicornis , possess a diversity of Osmia species that are important pollinators of crops including O. cornuta , O. cornifrons , and O. taurus(Branstetter et al., 2021; Osterman et al., 2021). We found that the novel microsatellite loci amplified in up to 18 of the 22 in one specimen of O. cornuta (mean marker amplification = 9.9 ± 2.0 [n = 9]) (Supplementary Table 7). Osmia cornuta is native to north Africa and Europe and was introduced to the U.S. from Spain to pollinate crops in 1984 (Torchio & Asensio, 1985). While O. cornuta has not been established in the U.S. due to biological limitations (Torchio, Asensio, & Thorp, 1987; Torchio & Asensio, 1985), female O. cornuta have been found to visit between 9,500 and 23,600 almond flowers implicating high pollination efficiency (Bosch, 1994). Given the significance of O. cornuta to agriculture particularly in Europe, the novel microsatellite markers will have the capacity to characterize population genetic structure and diversity of managed populations to guide future management and breeding strategies. Finally, the invasion of the intentionally introduced O. cornifrons and accidentally introduced O. taurus to North America may benefit from population genetic study (LeCroy, Savoy-Burke, Carr, Delaney, & Roulston, 2020). Specifically, we expect that the novel microsatellites to answer questions concerning the colonization timing of these non-native bees, underlying genetic diversity and structure, and potentially their rate of expansion throughout North America.
The use of microsatellites to characterize population genetic diversity in bee pollinators has a long history. Based on Google Scholar (https://scholar.google.com/), the first peer-reviewed paper that characterizes microsatellites in bees (Hymenoptera: Anthophila) was on A. mellifera and Bombus terrestris by Estoup et al. (1993). Over the last 29 years, microsatellites continued to be developed for honey bees (Apis spp.) (Estoup, Garnery, Solignac, & Cornuet, 1995; Solignac et al., 2003), bumble bees (Bombus spp.) (Estoup, Scholl, Pouvreau, & Solignac, 1995; Reber Funk, Schmid-Hempel, & Schmid-Hempel, 2006; Stolle et al., 2009), mason bees (Osmiaspp.) (Neumann & Seidelmann, 2006; Van Eeckhoven et al., 2022), stingless bees (Melipona spp.) (Peters, Queller, Imperatriz Fonseca, & Strassmann, 1998), orchid bees (Euglossa spp.) (López-Uribe, Santiago, Bogdanowicz, & Danforth, 2013; Paxton, Zobel, Steiner, & Zillikens, 2009), and alfalfa leafcutting bees (Megachilerotundata F.; Megachilidae) (Strange et al., 2017), among others. The development of microsatellite markers in bee pollinators has been critical in supporting research studies of bee evolution and ecology and in making informed decisions on their management in agricultural and wildlife conservation settings. For example, microsatellites have proven instrumental in informing honey bee breeding decisions (Bourgeois et al., 2008; Delaney et al., 2009; Jensen et al., 2006), estimating bumble bee decline (Cameron et al., 2011; Lozier & Cameron, 2009), and determining the impacts of habitat fragmentation on gene flow in orchid bees (Soro, Quezada-Euan, Theodorou, Moritz, & Paxton, 2016; Suni & Brosi, 2011).
Osmia lignaria has been heavily adopted by producers to pollinate a diversity of orchard and berry crops that span across a broad geographic range with diverse climates (Bosch & Kemp, 1999; Boyle & Pitts-Singer, 2019; Sheffield, 2014; Torchio, 1976). Furthermore, populations endemic to the continental U.S. lends to both genetic and physiological regional differences (Branstetter et al., 2021; Pitts-Singer et al., 2014). Commercial suppliers in the western U.S. are sourcing O. lignaria from different parts of their native range including Washington, Utah, Idaho, and California, but may sell such bees directly to customers or to other distributors across the U.S. As new O. lignaria producers enter the market in response to the growing demands for integrated crop pollination (Isaacs et al., 2017), the sustainability of O. lignaria sourcing, management, and breeding would benefit from knowledge on underlying population genetic diversity and structure. For example, illegal trap nesting of O. lignariaand other solitary bees on public lands is commonplace throughout the Intermountain West. In fact, the Cub River field site is associated with a history of illegal trap nesting for Osmia species (Tepedino & Nielson, 2017). If left unchecked, trap-nesting could decimate endemic O. lignaria populations, ultimately reducing the genetic diversity available to commercial enterprises. Application of population genetic diversity data with microsatellites could uncover resilient or imperiled populations and ultimately guide sustainable trap-nesting pursuits. In conclusion, we anticipate the novel markers developed in our study to support critical investigations of O. lignaria evolution, ecology, conservation, and livestock development.