Development and application of 14 microsatellite markers in the burying beetle Nicrophorus vespilloides reveals population genetic differentiation at local spatial scales

Introduction

Burying beetles (genus Nicrophorus) are relatively unusual among insects in providing elaborate parental care for their developing larvae. Reproduction centres on the fresh carcass of a small vertebrate (like a songbird or mouse), which adults locate by flight. If more than one adult of the same sex finds the carcass, they will commonly fight to determine ownership (e.g., Otronen, 1988; Scott, 1994). Defeated subordinates may stay nearby: males may sneak matings with the dominant female, while females might become intraspecific brood parasites (e.g., Müller, Eggert & Dressel, 1990; Hopwood et al., 2015). The dominant individuals of each sex then pair up and together prepare the carcass for reproduction (although mated females can singlehandedly raise young: Müller et al., 2007). They remove the fur or feathers, roll the flesh into a ball and bury it below ground in a shallow grave. During this time, parents defend their carcass breeding resource from attack by conspecifics, congenerics and other carrion-feeding insects (Robertson, 1993; Trumbo, 1990). Eggs are laid in the soil near the carcass. Newly hatched larvae crawl into a crater on the brood ball and there they solicit attention from their parents, who also stay to protect them from attack (Trumbo, 2007). About a week after the larvae hatch, the carcass is entirely consumed. The beetle parents then fly off to search for new mating opportunities or fresh carrion and the larvae disperse into the soil to pupate (Scott, 1998).

Burying beetles thus play a key ecological role as nutrient recyclers (e.g., Royle, Hopwood & Head, 2013). Furthermore, their relatively unusual natural history, and the ease with which they can be bred in the lab, means that several have become popular study species in experimental analyses of social evolution, the evolution of parental care and mating systems (e.g., Royle, Hopwood & Head, 2013; Scott, 1998). Nevertheless, despite their widespread use in the lab, relatively little work has focused on the burying beetles’ ecology in nature. This is because it is difficult to track marked burying beetles through their life course to understand patterns of dispersal (e.g., Attisano & Kilner, 2015), the likely extent of competition for limited carrion resources (e.g., Kilner et al., 2015), the degree of connectivity between populations and therefore the potential for inbreeding (e.g., Pilakouta et al., 2015). Yet this knowledge is key for interpreting the results of experiments carried out in the laboratory (e.g., Kilner et al., 2015; Attisano & Kilner, 2015; Pilakouta et al., 2015).

Furthermore, habitat fragmentation has recently been suggested to influence population structure in beetles (Keller & Largiader, 2003; Suzuki & Yao, 2014). Habitat fragmentation by deforestation has specifically been hypothesised to influence dispersal of N. americanus in the USA (Creighton et al., 2009) and N. quadripunctatus populations in Japan (Suzuki & Yao, 2014). Yet behavioural work on flight performance shows that burying beetles are capable of sustained flight over tens of kilometres (Attisano & Kilner, 2015). Therefore, a key aim of this study was to determine the extent to which burying beetles disperse in nature over relatively short distances, using genetic techniques (cf Houston et al., 2015).

Recently, molecular resources have been developed for N. vespilloides, including a genome, epigenome (Cunningham et al., 2015) and transcriptomes (e.g., Parker et al., 2015; Palmer et al., 2016). We took advantage of these newly available molecular tools to develop microsatellite markers. We used the markers to determine the extent of genetic differentiation between natural populations of N. vespilloides, deliberately choosing sites on a local scale, no more than 21 km apart (Fig. 1) to test the power of the markers we developed. These four populations were all in south Cambridgeshire and at: Waresley Woods (Latitude: 52.17487°; Longitude: −0.17354°), Gamlingay Woods (Latitude: 52.15555°; Longitude: −0.19286°), Byron’s Pool (Latitude: 52.17305°; Longitude: 0.10196°) and Madingley Woods (Latitude: 52.22658°; Longitude: 0.04303°). These four populations inhabit patches of woodland that are islands in a landscape dominated by arable farmland and urban development. Their close proximity enabled us to determine whether habitat fragmentation or beetle flight performance could better explain population structure on a local scale.

Material and Methods

Results

Discussion

We developed microsatellite markers to infer details of the burying beetle’s ecology that cannot be deduced through simple observation, but which are becoming increasingly important for the interpretation of experiments on this species in the laboratory. The process of microsatellite development was greatly facilitated by the existing N. vespilloides genome (Cunningham et al., 2015) and transcriptome (Parker et al., 2015). The available genomic tools, however, are still poorly annotated and so further detailed characterisation of the sequences containing the markers is still limited (see Supplementary Information). Nevertheless, our analyses suggest that the markers are predominantly unlinked. We rapidly developed a set of 14 (out of 20) reliable polymorphic markers for the species: i.e., 70% were successful. This proportion of successful markers is similar to that obtained for N. quadripunctatus (Suzuki & Yao, 2014), although these authors used enriched genomic libraries for marker development.

Our genetic analyses revealed significant deviations from Hardy-Weinberg equilibrium at five loci. This is most likely due to an excess of homozygotes at these loci but could also be due to the presence of null alleles in three of these markers. The only other marker with a putatively high null allele frequency (>0.05) was Nvesp_E. However, the high number of homozygotes for this marker is unlikely to be due to null alleles because it is in HWE (Kalinowski, Taper & Marshall, 2007). We are confident that our results are not biased by including markers that deviated from HWE, or potential null alleles, because we obtained similar results when we excluded these markers from our analyses. In general, the microsatellite markers showed lower observed heterozygosity than expected and the global F_IS value (0.085) also suggested some heterozygosity deficiency. We infer from these findings that there is limited gene flow between our study populations, and potentially some inbreeding as well. The consequences of inbreeding in N. vespilloides have recently been analysed experimentally in the laboratory (e.g., Mattey, Strutt & Smiseth, 2013; Pilakouta et al., 2015; Pilakouta & Smiseth, 2016) but to our knowledge, our study provides the first indication that N. vespilloides might breed with relatives in nature and it matches previous results obtained for N. quadripunctatus (Suzuki & Yao, 2014).

Consistent with our interpretation of limited gene flow between populations, we found significant pairwise population differentiation between all but one pair of populations (Waresley-Gamlingay; approximately 2.5 km apart) despite the low geographical separation between the sampling sites (maximum 20.3 km). This genetic differentiation may be the result of neutral population differentiation, or the effects of selection acting on functional genes correlated with the neutral markers (e.g., Rousset & Raymond, 1995). We think the former possibility is more likely because we obtained similar results when we ran the analyses with and without markers in HWE and because a BLASTx search showed that the markers did not generally correspond to (albeit limited) known coding genes in N. vespilloides.

Previous work on flight performance of N. vespilloides showed that these beetles have a wide distribution of flight distances in the laboratory ranging from 68 m to 26 km. Furthermore, flight durations ranged from 61 s to 6.5 h under laboratory conditions (Attisano & Kilner, 2015). Beetles tethered in flight mills may be able to fly for greater distances than naturally flying beetles because they bear less of their weight in flight. But even if we assume that beetles in natural flight cover only a third of the distance that they achieved in a flight mill, our data suggest that population differentiation within the sampling area is not solely attributable to the flight range of the burying beetle.

We suggest instead that habitat fragmentation has driven the fine-scale population structure we report here by imposing barriers that limit dispersal (cf Pascoal et al., 2009; Kanno, Vokoun & Letcher, 2011; Valtonen et al., 2014). Our analyses suggest that each ‘island’ population of burying beetles is increasingly reproductively isolated with increasing geographic distance between woodlands. Perhaps N. vespilloides is unwilling, or unable, to undertake flights across open fields and through housing. Studies of congeneric species (Nicrophorus marginatus, N. tomentosus, N. orbicollis and N. defodiens) found that the size of these woodland fragments affects the abundance and reproductive success of the resident burying beetle population (Trumbo & Bloch, 2000; Gibbs & Stanton, 2001). Our analyses suggest that the extent of their connectivity might also be an important factor for promoting gene flow and preventing populations from becoming smaller and more inbred. If our interpretation is correct, then it has important conservation implications because it suggests that there is a threshold size of woodland required to sustain an outbreeding burying beetle population. It also might help explain why populations of American burying beetles (N. americanus) have collapsed so spectacularly in recent years following substantial deforestation, causing it to become endangered (Anderson, 1982; Creighton et al., 2009).

Although Single Nucleotide Polymorphism (SNPs) are starting to be the tool of choice for population genetics/genomics studies, microsatellites still provide a cheaper alternative. Nevertheless, we anticipate that the microsatellites we have developed will prove most useful in future work for assigning parentage (S Pascoal, 2016, unpublished data) because the large brood size typically seen in N. vespilloides still makes other techniques prohibitively expensive. N. vespilloides mates rampantly and promiscuously in the laboratory (e.g., House et al., 2009) and in natural populations (e.g., Müller et al., 2007). Although several previous studies have analysed strategies used by males for securing paternity (e.g., Eggert, 1992; House et al., 2009), parentage has never before been assigned using microsatellites. The microsatellites we have developed here thus pave the way for more detailed analyses of the evolutionary causes and consequences of promiscuity in this species.

Supplemental Information