Skip to main content
NCBI home page
Royal Society Open Science logoLink to Royal Society Open Science
. 2020 Jul 22;7(7):200225. doi: 10.1098/rsos.200225

Introduced bees (Osmia cornifrons) collect pollen from both coevolved and novel host-plant species within their family-level phylogenetic preferences

Anthony D Vaudo 1,2,✉, David J Biddinger 3, Wiebke Sickel 4,5, Alexander Keller 6, Margarita M López-Uribe 2
PMCID: PMC7428236  PMID: 32874623

Abstract

Studying the pollen preferences of introduced bees allows us to investigate how species use host-plants when establishing in new environments. Osmia cornifrons is a solitary bee introduced into North America from East Asia for pollination of Rosaceae crops such as apples and cherries. We investigated whether O. cornifrons (i) more frequently collected pollen from host-plant species they coevolved with from their geographic origin, or (ii) prefer host-plant species of specific plant taxa independent of origin. To address this question, using pollen metabarcoding, we examined the identity and relative abundance of pollen in larval provisions from nests located in different landscapes with varying abundance of East-Asian and non-Asian plant species. Our results show that O. cornifrons collected more pollen from plant species from their native range. Plants in the family Rosaceae were their most preferred pollen hosts, but they differentially collected species native to East Asia, Europe, or North America depending on the landscape. Our results suggest that while O. cornifrons frequently collect pollen of East-Asian origin, the collection of pollen from novel species within their phylogenetic familial affinities is common and can facilitate pollinator establishment. This phylogenetic preference highlights the effectiveness of O. cornifrons as crop pollinators of a variety of Rosaceae crops from different geographic origins. Our results imply that globalization of non-native plant species may ease the naturalization of their coevolved pollinators outside of their native range.

Keywords: managed pollinators, pollination ecology, pollen metabarcoding, introduced species, host specialization, diet breadth

1. Introduction

The majority of wild and crop plant species require insect-mediated pollination for reproduction [1]. An ideal pollinator comprises particular traits related to its host-plant species: timing its foraging period with blooming period, exhibiting floral constancy to guarantee pollen transfer between conspecific flowers and displaying fidelity to the host-plant species over generations [2,3]. Relationships between pollinators and their host-plant species in their native geographic ranges have evolved over thousands to millions of years, leading to coevolved (morphological, behavioural, chemical and physiological) traits [3,4]. Likewise, efficient crop pollination may result from historical relationships between wild relatives of crops and their native pollinating species [5].

Most crops are cultivated globally, far outside the home range of their wild ancestors and coevolved pollinators [6]. For many pollinator-dependent crops, generalist social bees, such as western honeybees (Apis mellifera), are managed to provide crop pollination services. Yet honeybees are not efficient pollinators for many crops (e.g. due to mechanical mismatches of the bee and flower reproductive parts, or lack of sufficient movement between conspecific plants; [7–9]). A logical action to maximize pollination services in agroecosystems is to introduce coevolved crop pollinators into new geographic regions where the crop is cultivated [10–13]. However, the introduction of pollinator species into new habitats poses risks such as competition with native pollinators, pollination of invasive plant species (other than the intended crop) and importation of novel pathogens [14–18].

Several intentionally introduced bees have naturalized outside of their native geographic ranges within short periods of time [14,15,17]. Characterizing the pollen preferences of pollinators in their introduced ranges allows us to test the hypothesis that bees maintain preferences for plant species that they coevolved with in their native range (native preference hypothesis). Alternatively, bees' host-plant preferences could be exclusively driven by phylogenetic affinity independent of the geographic origin of the host-plant (phylogenetic preference hypothesis). These two hypotheses are non-mutually exclusive because bees could prefer host-plants from their native range while keeping their phylogenetic affinity for a limited number of plant lineages. However, support for the phylogenetic hypothesis (i.e. equal use of native and non-native plants from the same plant lineages) would indicate that pollinators could more easily naturalize in new habitats by the utilization of novel, but phylogenetically related, host-plant species [2,14].

In this study, we tested the native and phylogenetic hypotheses with the Japanese orchard bee, Osmia cornifrons Radoszkowski 1887 (Hymenoptera: Megachilidae). This solitary univoltine bee was introduced to the United States from East Asia (Japan, Korea, China, Eastern Russia) in 1977 to improve pollination services of fruit trees in the Rosaceae family (e.g. apples, cherries, peaches, pears, etc.; [12,19]) and has supplemented or replaced managed honeybees for fruit tree pollination in North America [20,21]. Phenologically, O. cornifrons forage in spring at the onset of blooming of commercial fruit trees in the Rosaceae family, including cherry (native to Europe) followed by peach and apple (native to East Asia). It has been demonstrated that O. cornifrons populations in Japan, Russia and the United States exhibit pollen fidelity to Rosaceae and Fabaceae species [22–24], indicating mesolectic foraging behaviour (collecting pollen from many species within one to three main plant families [2]). However, these previous studies analysed pollen via microscopy which only indicated pollen family or genus. Therefore, data are absent regarding the species-level identity of the pollen preferences of O. cornifrons, and the degree of pollen collected from native and introduced host-plant species in different landscapes [22–25].

The establishment of O. cornifrons, and abundance of crop and wild plant species from East Asia in North America [26,27], makes this an ideal system to determine how floral preferences of recently introduced pollinators may shift in new environments. Using pollen metabarcoding, we analysed pollen species identity and relative abundance in O. cornifrons larval pollen provisions. We compared O. cornifrons pollen provisions between heterogeneous landscapes of exotic and native plants [28]: orchards with a high abundance of East-Asian crops (apples and pears), orchards with a high abundance of European crops (cherries) and landscapes with abundance of host-plants from different geographic origins (suburban and wooded forest). We discuss our results in the context of how foraging plasticity to different species and abundant resources within host-plant family preferences can facilitate pollinator establishment in novel habitats.

2. Methods

2.1. Study system

We placed O. cornifrons nests (five boxes of 56-hole BinderBoard® lined with 8 mm diameter × 15 cm paper straws; www.pollinatorparadise.com) on field edges of three predominantly apple and three predominantly cherry orchards, representing typical agricultural landscapes where O. cornifrons is managed. We also placed nests in a suburban site (houses within 500 m) and in a wooded forest site (more than 3 km from orchards) representing non-orchard heterogeneous landscapes that O. cornifrons occupies (electronic supplementary material, figure S1). All sites were at least 1.8 km away from each other, beyond their reported foraging range [25,29–31]. All females were allowed to forage for their entire active period (approximately four weeks) where one to two cells are provisioned per day in ideal temperature and precipitation conditions [21]. At the end of the growing season, we collected 12 complete larval provisions from cells of different individual nests per site (N = 96) where eggs failed to hatch (and were not parasitized by mites [32]). These pollen provisions were selected because they were unused and complete and therefore constitute all pollen provisioned for that cell. No permissions were required prior to conducting field work; all sites are privately owned and work in coordination with Penn State Fruit Research and Extension Center.

2.2. Pollen metabarcoding

We extracted DNA using the Qiagen DNeasy Plant Kit. Samples were prepared for sequencing using the pollen metabarcoding protocol for the ITS2 gene region [33,34]. Negative controls were used in PCR to verify there was no plant DNA contamination. Forward and reverse primers contained barcoding sequences so each sample had a unique combination. The final pooled library was spiked with 5% PhiX control to increase sequencing quality. The library was sequenced on an Illumina MiSeq v. 2, 2 × 250 bp spiked with custom index, Read1 and Read2 sequencing primers to bind to the unique ITS2 primers [34].

We filtered forward reads for quality (maxEE = 1) and length (greater than 150 bp). We used the BCdatabaser [35] to subset the NCBI database for the marker and create a reference sequence list of all plant species present in Pennsylvania (https://plants.sc.egov.usda.gov/java/). We classified reads by selecting the match with highest identity to a reference species over the entire amplified region with a global alignment using VSEARCH [36] and a threshold of at least 97% sequence identity. Across all 96 samples, we obtained a total of 6 800 673 reads passing our quality filters, classified 4 922 865 reads and averaged 51 279.84 ± 4872.85 s.e. classified reads per sample. Using the R package phyloseq [37], we imported the final read abundance and taxonomic tables, normalized the read abundance by sample such that each sample totalled 100 (relative read abundances) and filtered taxa that represented less than 1% per sample. We verified the consistency of species identification and abundance by comparing our results to amplicon sequence variants and therefore kept our original dataset for analysis. For classification and filtering scripts, refer to our Dryad data entry (https://doi.org/10.5061/dryad.ffbg79cqn). We also verified consistency in species identification by comparing against the species characterized in the same apple orchards in Kammerer et al. [28].

We combined the final abundance and taxonomic tables in JMP Pro 14.3.0 (SAS Institute Inc., Cary, NC, USA). We classified plant species as East-Asian origin, or North American or European. Final filtering removed plant genera that were extremely rare in the dataset, those that contained less than 0.3% of the total counts per site across all sites. Because relative pollen abundances (RRA) by light microscopy and ITS2 metabarcoding have been previously strongly correlated with the primers applied here [33], we used RRA as a proxy for pollen abundance in larval provisions [38–40]. We analysed RRAs by site or landscape using the mean of RRAs from 12 samples per site and used RRAs as a population wide estimate of pollen use and not sample-wise abundance estimations [38,41,42] (see electronic supplementary material, table S1 for species list and metadata, and for raw data see https://doi.org/10.5061/dryad.ffbg79cqn).

2.3. Statistical analysis

To determine diversity of O. cornifrons pollen preferences, we estimated and compared Richness, Shannon diversity and Simpson dominance indices between landscapes and sites nested within landscape using generalized linear model (GLM) in JMP (Richness and Simpson with Poisson and Shannon with normal distribution). We used non-metric multidimensional scaling (NMDS) using PRIMER v6 (Plymouth Routines in Multivariate Ecological Research, Auckland, NZ) to determine differences in provisions between landscapes for pollen species or genera. NMDS was conducted using a Bray–Curtis similarity matrix; ANOSIM was conducted with 10 000 permutations to analyse pairwise differences between landscapes.

To test the native hypothesis, we investigated relative proportions of East-Asian versus non-Asian pollen of all plant species across all provisions between landscapes (GLM). To test the phylogenetic hypothesis, we analysed differences in pollen family and genera collected across landscapes (GLM). To test the interaction of the native and phylogenetic hypotheses, we used a 2 × 2 chi-square contingency analysis to determine differences in proportions of pollen collected from Asian versus non-Asian, and Rosaceae/Fabaceae versus non-Rosaceae/Fabaceae pollen. These tests were conducted by summing data at the genus level to avoid negative bias, where the average genus representation with high counts is reduced by rare species (with low counts) within the genus. Because O. cornifrons exhibited strong preference for Rosaceae, we further investigated differences in the abundance of pollen collected from Asian versus non-Asian Rosaceae species between landscapes with NMDS and GLM. All GLM analyses were conducted in JMP with landscape and site nested in landscape as independent factors, with average RRA as the dependent variable with Poisson distributions; Wilcoxon tests were used for post-hoc comparisons. To reduce repeating significance values, all reported results are significant at p < 0.05; χ2 values are reported for contingency analysis unless GLM is indicated.

3. Results

Osmia cornifrons pollen provisions comprised five to eight species, one to six genera, and four families on average. Simpson's dominance averaged 0.45 at the genus level indicating low diversity and that this species is indeed a mesolectic forager [2] (i.e. as their collections were dominated by plant species within few plant families [2,28]). Diversity and Simpson's dominance did not differ between landscapes, except species richness was highest in cherry orchards (electronic supplementary material, figure S2; Species Richness). The identity and composition of pollen provisions differed between landscapes, except between apple and suburban sites at the genus level, revealing that suites of pollen collected were landscape dependent (electronic supplementary material, figure S3a,b, species: R = 0.35; genus: R = 0.29).

Only 18 of 73 plant species we detected with metabarcoding were of East-Asian origin (electronic supplementary material, table S1). Across all samples, the average abundance of East-Asian pollen per sample was higher than non-Asian plants (GLM: χ2 = 840.1). But this effect was landscape dependent (GLM: χ2 = 1154): East-Asian pollen was more abundant than non-Asian pollen in apple (χ2 = 1425) and wooded landscapes (χ2 = 197), and lower in cherry (χ2 = 37) and suburban landscapes (χ2 = 93) (figure 1; electronic supplementary material, figure S4). Rosaceae pollen was by far the most represented among samples (approx. 60% per sample, GLM: χ2 = 24 700; electronic supplementary material, figure S5). This was consistent across all sites and landscapes except the suburban site where Cercis (Fabaceae) pollen was frequently collected, probably due to the high density of ornamental planting of C. canadensis in the urban areas compared to agricultural and wild landscapes (figure 1; electronic supplementary material, figure S4).

Figure 1.

Figure 1.

Open in a new tab

Osmia cornifrons larval provision host-plant genera by landscape coloured by plant family. Data in left panel are mean relative read abundances ± s.e. Boxes outlined in black are East-Asian origin. Right panel indicates proportions of pollen collected from East-Asian and/or Rosaceae/Fabaceae pollen versus not. Note that Rosaceae genera are the most represented across all landscapes and East-Asian abundances vary by landscape.

We further explored whether O. corniforns more frequently collected pollen from East-Asian Rosaceae species versus North American or European Rosaceae species. Indeed, there were differences between landscapes in composition of Rosaceae species collected (figure 2; electronic supplementary material, figure S3c,d, NMDS species: R = 0.31; genus: R = 0.25) and differences in Asian versus non-Asian Rosaceae pollen collected (χ2 = 851). East-Asian Rosaceae pollen was more abundant in provisions than non-Asian Rosaceae pollen in apple orchards (χ2 = 1425), suburban (χ2 = 280) and wooded landscapes (χ2 = 197), and less abundant in cherry orchards (χ2 = 37; figures 1 and 2).

Figure 2.

Figure 2.

Open in a new tab

Rosaceae plant species in Osmia cornifrons larval provisions by landscape. Data are mean relative read abundances ± s.e. Box colour represents target orchard crops of the particular landscape (green), non-target crops (blue), or not crop species (burgundy); boxes outlined in black are East-Asian origin.

4. Discussion

Our results reveal that O. cornifrons collect significant amounts of pollen from Rosaceae host-plants from their native East-Asian geographic origin even in naturalized habitats where alternative host-plants are abundant. However, across sites, they broadly use pollen from Rosaceae species from outside their native range indicating their capability to use host-plants species they did not coevolve with (figures 1 and 2). Our results also identify important secondary pollen hosts (e.g. North American Cercis canadensis related to the East-Asian host C. chinensis). Overall, O. cornifrons foraging is mostly driven by phylogenetic pollen preferences [2,14] showing high representation of rosaceous pollen (detecting Rosaceae species in 96% of all samples) of East-Asian, European and North American origin (electronic supplementary material, figure S4 and S5). The ability of O. cornifrons to use pollen sources from related preferred taxa, including crops, may facilitate adaptation to novel environments (figures 1 and 2).

The northeastern United States, where this study occurred, shares the same temperate forest vegetation biome as East Asia (and Northern Europe) and is now home to many East-Asian exotic plant species [26,27]. The pollen provisions of O. cornifrons exhibited a heterogeneous mix of East-Asian, European and North American species of Prunus, Rubus and Cercis (genera with native species to each geographic region; figures 1 and 2). Although only 25% of all plant species found in our dataset were introduced to North America from East Asia [26,27], O. cornifrons collected significant amounts of pollen from plants from this geographic origin (approx. 40% across landscapes, χ2 = 840.1, p < 0.01; figures 1 and 2; [15]). Interestingly, approximately 90% of the East-Asian pollen collected was in the Rosaceae family (figure 1). As expected, East-Asian pollen from apples (and pears) was dominant in provisions from apple orchards. However, the nests were placed along field edges where other flowering species and families in edges and forests were available for collection [28] suggesting their intentional choice of crop pollen. In wooded landscapes, where North American native species are more abundant, East-Asian pollen was also over-represented in pollen provisions. In landscapes dominated by European cherry trees, the bees collected pollen from not only cherries, but European, North American, and Asian Rosaceae species evenly, indicating that they foraged on more than just the most abundant host-plant surrounding the nest.

Our results indicate that O. cornifrons family-level host-plant preferences drive their foraging behaviour, corroborating that this is a mesolectic species that specializes on a few plant families: Rosaceae and Fabaceae [2]. Given O. cornifrons' limited foraging range, the variation between sites that we detected is probably driven by local availability of host-plant species. For instance, where Cercis was planted in high abundance in suburban landscapes, as opposed to orchards, we found this type of pollen in high abundance. We did not directly measure flowering plant species abundance at each site during O. cornifrons' foraging period, but replication across a variety of sites reveals similar foraging trends. Many plant species from multiple families were blooming at the same time (electronic supplementary material, table S1, figure S4; see [28]), including mass blooming trees such as Acer (Sapindaceae) or Salix (Salicaceae), which were found in low abundance or not at all in our samples. Additionally, low diversity of larval provisions (composed of multiple foraging trips) indicates O. cornifrons forages to a subset of host-plant species available (electronic supplementary material, figure S2 and S4). Our data indicate that East-Asian Rosaceae species are indeed preferred by O. cornifrons [14], but they will use non-Asian Rosaceae host-plants based on local floral availability ([25,28,43]; figures 1 and 2). The use of non-Asian host-plants may be the result of their sampling of the landscape, a by-product of nectar foraging as opposed to active pollen foraging, a dearth of Rosaceae and Fabaceae pollen post-crop bloom in orchards, or that alternative pollen resources are needed in their diet. How bees adjust their foraging behaviour in the absence of preferred resources, or the effects of local floral abundance on species preferences warrants further investigation in more controlled settings (see [22,44]).

The ability of mesolectic bee species to exclusively use host-plant species within family level preferences supports that these pollinators are ideal for crop pollination [17]. In apple orchards, the commercial apple (Malus pumila) and its East-Asian pollinizer crabapple (M. baccata) were represented evenly in pollen loads, indicating that O. cornifrons is effectively delivering pollination services [12], probably visiting both species and transferring pollen within foraging bouts throughout the day. While cherries represent non-Asian host-plant species that may not be their preferred pollen source, it was also actively collected by the bees in cherry orchard landscapes. These results suggest that O. cornifrons is an effective pollinator of cherry crops as well. However, we did not observe the same degree of preference for Prunus pollen in cherry orchards as we observed for Malus pollen in apple orchards (figure 2). In cherry orchards, the bees also visited alternative rosaceous pollen sources (including Malus, Rosa and Amelanchier) which may result from the short cherry blooming phenology or the higher preference for East-Asian pollen from Rosaceae plants. Regardless, our results suggest that the phylogenetic affinity of mesolectic bees for host-plant families that have cosmopolitan distributions can be exploited for agricultural pollination. Another example of this type of relationships can be observed in Megachile rotundata, a mesolectic bee specialized on Fabaceae, which is a managed pollinator for a variety of crops such as alfalfa [13].

While mesolectic bees may be ideal candidates for managed crop pollinators, risks associated with the introduction of exotic bee species for pollination services include pollination of invasive plant species and competition with native bee species [14–18]. Although O. cornifrons collected pollen from North American native plants (e.g. Acer, Amelanchier, Cercis and Rubus), they also collected from invasive species and common weeds of natural and agricultural habitats (e.g. Barbarea, Elaegnus, Rosa and Taraxacum) [14]. The quantitative and qualitative results of pollen use that we report in this study suggest that O. cornifrons may not be directly competing for floral resources with other native Osmia spp. [45,46]. The eastern North American native O. lignaria lignaria prefers early-blooming trees such as Quercus and Salix [45]. Western O. californica exhibit preference for Hydrophyllum, and O. l. propinqua for Hydrophyllum, Salix, and the widespread Crataegus [47,48]. These preferred pollen hosts from North American Osmia spp. are absent from our dataset even though they bloom during O. cornifrons' foraging period [28]. Potential competitive foraging between O. cornifrons and O. l. lignaria on Cercis spp. could occur as indicated by the preference for Cercis pollen in some of our samples [20,22,24,46,49]. However, specific and detailed studies would need to determine the level and effect of competition between species [15,17,46] (including competition for nesting sites in proximity to preferred hosts).

This study provides insights for appropriate management practices of O. cornifrons. For instance, managing mesolectic pollinators that prefer pollen from crop plants allows growers to place nests within the field at long distances from edges where pollination services from wild bees are generally reduced [25,29–31,43]. However, bees could be limited to short blooming periods and low nutritional diversity of the crops (such as orchard species), where they still need phenological and phylogenetic diversity of pollen species over their active period (figure 1 for apple orchards; [23,50]). By understanding bees' host-plant preferences, we can recommend complementary native edge and hedgerow host-plant species to support bee populations (e.g. Rubus spp. and C. canadaensis for O. cornifrons in orchards; [22,23,51,52]).

Over 80 bee species have been intentionally or accidentally introduced outside of their native range globally, yet how introduced bees adapt to floral communities outside of their native range is only known for pollen generalist honeybees and bumblebees [15,17]. Our data supports that pollinators show affinity to plants from their native range and will seek these pollen sources in new habitats [14]. This study reveals that despite preferences for a subset of host-plants from their native range as pollen sources, mesolectic bees may successfully establish in new communities when put in contact with novel related host-plants from other geographic origins. Mesolectic foraging, at the plant family level provides a balance between specificity to wild and targeted crop species and the ability to use phylogenetically related pollen sources. This foraging flexibility may be due to phylogenetically conserved floral reward quality [53,54], constraints in bee physiology [55–57], shared phenology between bees and host-plants [43], and bee behaviour and floral morphology [3] that allow bees to respond with plasticity to local resource availability. The globalization of cultivated crop, ornamental and invasive plant species may further facilitate the naturalization of introduced pollinators in novel environments [58].

Supplementary Material

Figure S1
rsos200225supp1.jpeg (1.5MB, jpeg)
Reviewer comments
rsos200225_review_history.pdf (1.5MB, pdf)

Supplementary Material

Figure S2
rsos200225supp2.jpg (351.3KB, jpg)

Supplementary Material

Figure S3
rsos200225supp3.jpg (566.1KB, jpg)

Supplementary Material

Figure S4
rsos200225supp4.jpg (2.2MB, jpg)

Supplementary Material

Figure S5
rsos200225supp5.jpeg (417.1KB, jpeg)

Supplementary Material

Table S1 species list.xlsx
rsos200225supp6.xlsx (12.5KB, xlsx)

Data accessibility

Available in Dryad: (i) pollen barcoded sequences, (ii) sample metadata, (iii) ITS2 reference database of plants of Pennsylvania, (iv) code for quality filtering, sequence alignment, and determining taxonomic and OTU tables, (v) resulting combined taxonomic and abundance table, (vi) code for taxa filtering and diversity indices, and (vii) final dataset used for analysis. Supporting figures are available in electronic supplementary material. https://doi.org/10.5061/dryad.ffbg79cqn [59].

Authors' contributions

A.D.V., D.J.B. and M.M.L.-U. designed the study. A.D.V. conducted molecular lab work and data analysis. D.J.B. collected pollen samples. W.S. assisted in laboratory techniques. A.K. conducted sequence quality control and classification. A.D.V. and M.M.L.-U. drafted the manuscript. All authors critically revised the manuscript and gave final approval for publication.

Competing interests

We declare we have no competing interests.

Funding

This work was supported by the USDA-SCRI grant PEN04398, USDA-SCRI Coordinated Agricultural Project grant no. MICL05063, USDA Hatch Appropriations under Project PEN04619, and the State Horticultural Association of Pennsylvania.

Permission to carry out fieldwork

No permissions were required prior to conducting field work; all sites are privately owned and work in coordination with Penn State Fruit Research and Extension Center.

References

  • 1.Ollerton J, Winfree R, Tarrant S. 2011. How many flowering plants are pollinated by animals? Oikos 120, 321–326. ( 10.1111/j.1600-0706.2010.18644.x) [DOI] [Google Scholar]
  • 2.Cane JH, Sipes S. 2006. Characterizing floral specialization by bees: analytical methods and a revised lexicon for oligolecty. In Plant-pollinator interactions: from specialization to generalization (eds Waser NM, Ollerton J), pp. 99–122. Chicago, IL: University of Chicago Press. [Google Scholar]
  • 3.Willmer P. 2011. Pollination and floral ecology. Princeton, NJ: Princeton University Press. [Google Scholar]
  • 4.Danforth BN, Minckley RL, Neff JL, Fawcett F. 2019. The solitary bees: biology, evolution, conservation. Princeton, NJ: Princeton University Press. [Google Scholar]
  • 5.López-Uribe MM, Cane JH, Minckley RL, Danforth BN. 2016. Crop domestication facilitated rapid geographical expansion of a specialist pollinator, the squash bee Peponapis pruinosa. Proc. R. Soc. B 283, 20160443 ( 10.1098/rspb.2016.0443) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Khoury CK, et al. 2016. Origins of food crops connect countries worldwide. Proc. R. Soc. B 283, 20160792 ( 10.1098/rspb.2016.0792) [DOI] [Google Scholar]
  • 7.Garibaldi LA, et al. 2013. Wild pollinators enhance fruit set of crops regardless of honey bee abundance. Science 339, 1608–1611. ( 10.1126/science.1230200) [DOI] [PubMed] [Google Scholar]
  • 8.Junqueira CN, Augusto SC. 2017. Bigger and sweeter passion fruits: effect of pollinator enhancement on fruit production and quality. Apidologie 48, 131–140. ( 10.1007/s13592-016-0458-2) [DOI] [Google Scholar]
  • 9.Bobiwash K, Uriel Y, Elle E. 2018. Pollen foraging differences among three managed pollinators in the highbush blueberry (Vaccinium corymbosum) agroecosystem. J. Econ. Entomol. 111, 26–32. ( 10.1093/jee/tox344) [DOI] [PubMed] [Google Scholar]
  • 10.Batra SWT. 1994. Anthophora pilipes villosula SM (Hymenoptera, Anthophoridae), a manageable Japanese bee that visits blueberries and apples during cool, rainy, spring weather. Proc. Entomol. Soc. Washington. 96, 98–119. [Google Scholar]
  • 11.Batra SWT. 1997. Solitary bees for Vaccinium pollination. In 6th Int. Symp. on Vaccinium Culture, Orono, ME, 12 August, pp. 71–76. [Google Scholar]
  • 12.Bosch J, Kemp WP. 2002. Developing and establishing bee species as crop pollinators: the example of Osmia spp. (Hymenoptera: Megachilidae) and fruit trees. Bull. Entomol. Res. 92, 3–16. [DOI] [PubMed] [Google Scholar]
  • 13.Pitts-Singer TL, Cane JH. 2011. The alfalfa leafcutting bee, Megachile rotundata: the world's most intensively managed solitary bee. In Annual review of entomology, vol. 56 (eds Berenbaum MR, Carde RT, Robinson GE), pp. 221–237. Palo Alto, CA: Annual Reviews. [DOI] [PubMed] [Google Scholar]
  • 14.Cane J. 2003. Exotic nonsocial bees (Hymenoptera: Apiformes) in North America: ecological implications. In For nonnative crops, whence pollinators of the future? (eds Strickler K, Cane J), pp. 113–126. Lanham, MD: Entomological Society of America. [Google Scholar]
  • 15.Goulson D. 2003. Effects of introduced bees on native ecosystems. Ann. Rev. Ecol. Evol. Syst. 34, 1–26. ( 10.1146/annurev.ecolsys.34.011802.132355) [DOI] [Google Scholar]
  • 16.Hedtke SM, Blitzer EJ, Montgomery GA, Danforth BN. 2015. Introduction of non-native pollinators can lead to trans-continental movement of bee-associated fungi. PLoS ONE 10, e0130560 ( 10.1371/journal.pone.0130560) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Russo L. 2016. Positive and negative impacts of non-native bee species around the world. Insects 7, 69 ( 10.3390/insects7040069) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Mallinger RE, Gaines-Day HR, Gratton C. 2017. Do managed bees have negative effects on wild bees? A systematic review of the literature. PLoS ONE 12, e0189268 ( 10.1371/journal.pone.0189268) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.McKinney MI, Park Y-L. 2012. Nesting activity and behavior of Osmia cornifrons (Hymenoptera: Megachilidae) elucidated using videography. Psyche 2012, 1–7. ( 10.1155/2012/814097) [DOI] [Google Scholar]
  • 20.Biddinger D, Ngugi H, Frazier J, Frazier M, Leslie T, Donovall L. 2010. Development of the mason bee, Osmia cornifrons, as an alternative pollinator to honey bees and as a targeted delivery system for biological control agents in the management of fire blight. Penn Fruit News 90, 35–44. [Google Scholar]
  • 21.Park MG, Joshi NK, Rajotte EG, Biddinger DJ, Losey JE, Danforth BN. 2018. Apple grower pollination practices and perceptions of alternative pollinators in New York and Pennsylvania. Renewable Agric. Food Syst. 129, 1–14. ( 10.1017/S1742170518000145) [DOI] [Google Scholar]
  • 22.Haider M, Dorn S, Sedivy C, Muller A. 2014. Phylogeny and floral hosts of a predominantly pollen generalist group of mason bees (Megachilidae: Osmiini). Biol. J. Linnean Soc. 111, 78–91. ( 10.1111/bij.12186) [DOI] [Google Scholar]
  • 23.Nagamitsu T, Suzuki MF, Mine S, Taki H, Shuri K, Kikuchi S, Kikuchi S, Masaki T. 2018. Effects of forest loss and fragmentation on pollen diets and provision mass of the mason bee, Osmia cornifrons, in central Japan. Ecol. Entomol. 43, 245–254. ( 10.1111/een.12494) [DOI] [Google Scholar]
  • 24.Russo L, Danforth B. 2017. Pollen preferences among the bee species visiting apple (Malus pumila) in New York. Apidologie 48, 806–820. ( 10.1007/s13592-017-0525-3) [DOI] [Google Scholar]
  • 25.Biddinger DJ, Joshi NK, Rajotte EG, Halbrendt NO, Pulig C, Naithani KJ, Vaughan M. 2013. An immunomarking method to determine the foraging patterns of Osmia cornifrons and resulting fruit set in a cherry orchard. Apidologie 44, 738–749. ( 10.1007/s13592-013-0221-x) [DOI] [Google Scholar]
  • 26.Fridley JD. 2008. Of Asian forests and European fields: Eastern U.S. plant invasions in a global floristic context. PLoS ONE 3, e3630 ( 10.1371/journal.pone.0003630) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Ricklefs RE, Guo QF, Qian H. 2008. Growth form and distribution of introduced plants in their native and non-native ranges in Eastern Asia and North America. Divers. Distrib. 14, 381–386. ( 10.1111/j.1472-4642.2007.00457.x) [DOI] [Google Scholar]
  • 28.Kammerer MA, Biddinger DJ, Rajotte EG, Mortensen DA. 2016. Local plant diversity across multiple habitats supports a diverse wild bee community in Pennsylvania apple orchards. Environ. Entomol. 45, 32–38. ( 10.1093/ee/nvv147) [DOI] [PubMed] [Google Scholar]
  • 29.Gathmann A, Tscharntke T. 2002. Foraging ranges of solitary bees. J. Anim. Ecol. 71, 757–764. ( 10.1046/j.1365-2656.2002.00641.x) [DOI] [Google Scholar]
  • 30.Guedot C, Bosch J, Kemp WP. 2009. Relationship between body size and homing ability in the genus Osmia (Hymenoptera; Megachilidae). Ecol. Entomol. 34, 158–161. ( 10.1111/j.1365-2311.2008.01054.x) [DOI] [Google Scholar]
  • 31.Zurbuchen A, Landert L, Klaiber J, Muller A, Hein S, Dorn S. 2010. Maximum foraging ranges in solitary bees: only few individuals have the capability to cover long foraging distances. Biol. Conserv. 43, 669–676. ( 10.1016/j.biocon.2009.12.003) [DOI] [Google Scholar]
  • 32.McKinney MI, Park Y-L. 2012. Distribution of Chaetodactylus krombeini (Acari: Chaetodactylidae) within Osmia cornifrons (Hymenoptera: Megachilidae) nests: implications for population management. Exp. App. Acarol. 60, 153–161. ( 10.1007/s10493-012-9629-7) [DOI] [PubMed] [Google Scholar]
  • 33.Keller A, Danner N, Grimmer G, Ankenbrand M, von der Ohe K, Rost S, Härtel S, Steffan-Dewenter I. 2015. Evaluating multiplexed next-generation sequencing as a method in palynology for mixed pollen samples. Plant Biol. 17, 558–566. ( 10.1111/plb.12251) [DOI] [PubMed] [Google Scholar]
  • 34.Sickel W, Ankenbrand MJ, Grimmer G, Holzschuh A, Härtel S, Lanzen J, Steffan-Dewenter I, Keller A. 2015. Increased efficiency in identifying mixed pollen samples by meta-barcoding with a dual-indexing approach. BMC Ecol. 15, 20 ( 10.1186/s12898-015-0051-y) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Keller A, Hohlfeld S, Kolter A, Schultz J, Gemeinholzer B, Ankenbrand MJ.2019. BDdatabaser: on-the-fly reference database creation for (meta-)barcoding. EcoEvoRxiv. ( ) [DOI] [PubMed]
  • 36.Rognes T, Flouri T, Nichols B, Quince C, Mahe F. 2016. VSEARCH: a versatile open source tool for metagenomics. PeerJ. 4, 22 ( 10.7717/peerj.2584) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.McMurdie PJ, Holmes S. 2013. phyloseq: An R package for reproducible interactive analysis and graphics of microbiome census data. PLoS ONE 8, e61217 ( 10.1371/journal.pone.0061217) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Deagle BE, Thomas AC, McInnes JC, Clarke LJ, Vesterinen EJ, Clare EL, Kartzinel TR, Eveson JP. 2019. Counting with DNA in metabarcoding studies: how should we convert sequence reads to dietary data? Mol. Ecol. 28, 391–406. ( 10.1111/mec.14734) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Lamb PD, Hunter E, Pinnegar JK, Creer S, Davies RG, Taylor MI. 2019. How quantitative is metabarcoding: a meta-analytical approach. Mol. Ecol. 28, 420–430. ( 10.1111/mec.14920) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Pornon A, Baksay S, Escaravage N, Burrus M, Andalo C. 2019. Pollinator specialization increases with a decrease in a mass-flowering plant in networks inferred from DNA metabarcoding. Ecol. Evol. 9, 13 650–13 662 ( 10.1002/ece3.5531) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Bell KL, Burgess KS, Botsch JC, Dobbs EK, Read TD, Brosi BJ. 2019. Quantitative and qualitative assessment of pollen DNA metabarcoding using constructed species mixtures. Mol. Ecol. 28, 431–455. ( 10.1111/mec.14840) [DOI] [PubMed] [Google Scholar]
  • 42.Richardson RT, Curtis HR, Matcham EG, Lin CH, Suresh S, Sponsler DB, Hearon LE, Johnson RM. 2019. Quantitative multi-locus metabarcoding and waggle dance interpretations reveal honey bee spring foraging patters in Midwest agroecosystems. Mol. Ecol. 28, 686–697. ( 10.1111/mec.14975) [DOI] [PubMed] [Google Scholar]
  • 43.Joshi NK, Otieno M, Rajotte EG, Fleischer SJ, Biddinger DJ. 2016. Proximity to woodland and landscape structure drives pollinator visitation in apple orchard ecosystem. Front. Ecol. Evol. 4, 38 ( 10.3389/fevo.2016.00038) [DOI] [Google Scholar]
  • 44.Kratschmer S, Petrovic B, Curto M, Meimberg H, Pachinger B. 2019. Pollen availability for the horned mason bee (Osmia cornuta) in regions of different land use and landscape structures. Ecol. Entomol. 45, 525–537. ( 10.1111/een.12823) [DOI] [Google Scholar]
  • 45.Raw A. 1974. Pollen preferences of three Osmia species (Hymenoptera). Oikos 25, 54–60. ( 10.2307/3543545) [DOI] [Google Scholar]
  • 46.Centrella ML. 2019. Regional and local drivers of mason bee (genus Osmia) decline across the eastern seaboard. Ann Arbor, MI: Cornell University. [Google Scholar]
  • 47.Williams NM, Tepedino VJ. 2003. Consistent mixing of near and distant resources in foraging bouts by the solitary mason bee Osmia lignaria. Behav. Ecol. 14, 141–149. ( 10.1093/beheco/14.1.141) [DOI] [Google Scholar]
  • 48.Levin MD. 1966. Biological notes on Osmia lignaria and Osmia californica (Hymenoptera: Apoidea, Megachilidae). J. Kansas Entomol. Soc. 39, 524–535. [Google Scholar]
  • 49.Kraemer ME, Favi FD. 2005. Flower phenology and pollen choice of Osmia lignaria (Hymenoptera: Megachilidae) in central Virginia. Environ. Entomol. 34, 1593–1605. ( 10.1603/0046-225X-34.6.1593) [DOI] [Google Scholar]
  • 50.Vaudo AD, Tooker JF, Grozinger CM, Patch HM. 2015. Bee nutrition and floral resource restoration. Current Opin. Insect Sci. 10, 133–141. ( 10.1016/j.cois.2015.05.008) [DOI] [PubMed] [Google Scholar]
  • 51.Park M, et al. 2012. Wild pollinators of eastern apple orchards and how to conserve them: college of agriculture and life sciences. Cornell University, Penn State University and The Xerces Society. See www.northeastipm.org/park2012. [Google Scholar]
  • 52.Vaughn M, Lee-Mäder E, Cruz JK, Goldenetz-Dollar J, Gill K, Borders B. 2015. Hedgerow planting (422) for pollinators: Pennsylvania installation guide and job sheet. Portland, OR: The Xerces Society. [Google Scholar]
  • 53.Vaudo AD, et al. 2020. Pollen protein:lipid macronutrient ratios may guide broad patterns of bee species floral preferences. Insects. 11, 132 ( 10.3390/insects11020132) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Ruedenauer FA, Spaethe J, van der Kooi CJ, Leonhardt SD.. 2019. Pollinator or pedigree: which factors determine the evolution of pollen nutrients? Oecologia 91, 349–358. ( 10.1007/s00442-019-04494-x) [DOI] [PubMed] [Google Scholar]
  • 55.Vaudo AD, Stabler D, Patch HM, Tooker JF, Grozinger CM, Wright GA. 2016. Bumble bees regulate their intake of essential protein and lipid pollen macronutrients. J. Exp. Biol. 219, 3962–3970. ( 10.1242/jeb.140772) [DOI] [PubMed] [Google Scholar]
  • 56.Vaudo AD, Patch HM, Mortensen DA, Tooker JF, Grozinger CM. 2016. Macronutrient ratios in pollen shape bumble bee (Bombus impatiens) foraging strategies and floral preferences. Proc. Natl Acad. Sci. USA 113, E4035–E4042. ( 10.1073/pnas.1606101113) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Giri S, Rule DC, Dillon ME. 2018. Fatty acid composition in native bees associations with thermal and feeding ecology. Comp. Biochem. Physiol. Part A 218, 70–79. ( 10.1016/j.cbpa.2018.01.013) [DOI] [PubMed] [Google Scholar]
  • 58.Brown J, Cunningham SA. 2019. Global-scale drivers of crop visitor diversity and the historical development of agriculture. Proc. R. Soc. B 286, 20192096 ( 10.1098/rspb.2019.2096) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Vaudo AD, Biddinger DJ, Sickel W, Keller A, López-Uribe MM. 2019. Data from: Introduced bees (Osmia cornifrons) collect pollen from both coevolved and novel host-plant species within their family-level phylogenetic preferences Dryad Digital Repository. ( 10.5061/dryad.ffbg79cqn) [DOI] [PMC free article] [PubMed]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Citations

  1. Vaudo AD, Biddinger DJ, Sickel W, Keller A, López-Uribe MM. 2019. Data from: Introduced bees (Osmia cornifrons) collect pollen from both coevolved and novel host-plant species within their family-level phylogenetic preferences Dryad Digital Repository. ( 10.5061/dryad.ffbg79cqn) [DOI] [PMC free article] [PubMed]

Supplementary Materials

Figure S1
rsos200225supp1.jpeg (1.5MB, jpeg)
Reviewer comments
rsos200225_review_history.pdf (1.5MB, pdf)
Figure S2
rsos200225supp2.jpg (351.3KB, jpg)
Figure S3
rsos200225supp3.jpg (566.1KB, jpg)
Figure S4
rsos200225supp4.jpg (2.2MB, jpg)
Figure S5
rsos200225supp5.jpeg (417.1KB, jpeg)
Table S1 species list.xlsx
rsos200225supp6.xlsx (12.5KB, xlsx)

Data Availability Statement

Available in Dryad: (i) pollen barcoded sequences, (ii) sample metadata, (iii) ITS2 reference database of plants of Pennsylvania, (iv) code for quality filtering, sequence alignment, and determining taxonomic and OTU tables, (v) resulting combined taxonomic and abundance table, (vi) code for taxa filtering and diversity indices, and (vii) final dataset used for analysis. Supporting figures are available in electronic supplementary material. https://doi.org/10.5061/dryad.ffbg79cqn [59].

Articles from Royal Society Open Science are provided here courtesy of The Royal Society

RESOURCES