Floral hosts of leaf-cutter bees (Megachilidae) in a biodiversity hotspot revealed by pollen DNA metabarcoding of historic specimens

RESEARCH ARTICLE Floral hosts of leaf-cutter bees (Megachilidae) in a biodiversity hotspot revealed by pollen DNA metabarcoding of historic specimens Annemarie Gous1,2, Connal D. Eardley2, Steven D. Johnson2, Dirk Z. H. Swanevelder1, Sandi Willows-MunroID2* 1 Biotechnology Platform, Agricultural Research Council, Pretoria, South Africa, 2 School of Life Sciences, University of KwaZulu-Natal, Pietermaritzburg, South Africa * willows-munro@ukzn.ac.za Abstract South Africa is a megadiverse country with three globally recognised biodiversity hotspots within its borders. Bees in particular show high diversity and endemism in the western part of the country. Not much is currently known about the floral host preferences of indigenous bees in South Africa, with data only available from observational studies. Pollen metabarcod- ing provides provenance information by utilising DNA analyses instead of floral visitation and traditional microscopic identification to identify pollinator food plants, which can be time con- suming and imprecise. In this study, we sampled pollen from leaf-cutter bees (Megachilidae) specimens maintained in a historic insect collection (National Collection of Insects, South Africa) that were originally collected from two florally important areas in South Africa (Succu- lent Karoo and Savanna) and used metabarcoding to determine pollen provenance. We also sampled pollen from leafcutter bee species with wider distributions, that extend across many different biomes, to determine if these ‘generalist’ species show relaxed floral host specificity

in some biomes. Metabarcoding involved sequencing of the nuclear internal transcribed spacer 2 (ITS2) region. Amplicons were compared to a sequence reference database to assign taxonomic classifications to family level. Sequence reads were also clustered to OTUs based on 97% sequence similarity to estimate numbers of plant species visited. We found no significant difference in the mean number of plant taxa visited in the Succulent Karoo and Savanna regions, but the widespread group visited significantly more floral hosts. Bees from the widespread group were also characterised by a significantly different compo- sition in pollen assemblage. The time since specimens were collected did not have an effect on the mean number of taxa visited by any of the bee species studied. This study highlights national history collections as valuable sources of temporal and spatial biodiversity data. Introduction The use of next-generation sequencing (NGS) technology and DNA barcoding in high- throughput identification of plant origins from pollen samples has been on the rise in recent years [1–6]. There are many advantages to this method over traditional microscopic palynol- ogy, including increased time-efficiency, multiplexing samples to reduce costs, and the relative PLOS ONE PLOS ONE | https://doi.org/10.1371/journal.pone.0244973 January 21, 2021 1 / 16 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Citation: Gous A, Eardley CD, Johnson SD, Swanevelder DZH, Willows-Munro S (2021) Floral hosts of leaf-cutter bees (Megachilidae) in a

biodiversity hotspot revealed by pollen DNA metabarcoding of historic specimens. PLoS ONE 16(1): e0244973. https://doi.org/10.1371/journal. pone.0244973 Editor: Guy Smagghe, Ghent University, BELGIUM Received: June 19, 2020 Accepted: December 21, 2020 Published: January 21, 2021 Peer Review History: PLOS recognizes the benefits of transparency in the peer review process; therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. The editorial history of this article is available here: https://doi.org/10.1371/journal.pone.0244973 Copyright: © 2021 Gous et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All the data generated in the study has been uploaded to FigShare and has been made public. The DOI are as follows: : M. feline (https://doi.org/10.6084/m9.figshare.