Bee-population declines are linked to nutritional shortages caused by land-use intensification, which reduces diversity and abundance of host-plant species. Bees require nectar and pollen floral resources that provide necessary carbohydrates, proteins, lipids, and micronutrients for survival, reproduction, and resilience to stress. However, nectar and pollen nutritional quality varies widely among host-plant species, which in turn influences how bees forage to obtain their nutritionally appropriate diets. Unfortunately, we know little about the nutritional requirements of different bee species. Research must be conducted on bee species nutritional needs and host-plant species resource quality to develop diverse and nutritionally balanced plant communities. Restoring appropriate suites of plant species to landscapes can support diverse bee species populations and their associated pollination ecosystem services.

To fuel their activities and rear their offspring, foraging bees must obtain a sufficient quality and quantity of nutritional resources from a diverse plant community. Pollen is the primary source of proteins and lipids for bees, and the concentrations of these nutrients in pollen can vary widely among host-plant species. Therefore we hypothesized that foraging decisions of bumble bees are driven by both the protein and lipid content of pollen. By successively reducing environmental and floral cues, we analyzed pollen-foraging preferences of Bombus impatiens in (i) host-plant species, (ii) pollen isolated from these hostplant species, and (iii) nutritionally modified single-source pollen diets encompassing a range of protein and lipid concentrations. In our semifield experiments, B. impatiens foragers exponentially increased their foraging rates of pollen from plant species with high protein: lipid (P:L) ratios; the most preferred plant species had the highest ratio (∼4.6:1). These preferences were confirmed in cage studies where, in pairwise comparisons in the absence of other floral cues, B. impatiens workers still preferred pollen with higher P:L ratios. Finally, when presented with nutritionally modified pollen, workers were most attracted to pollen with P:L ratios of 5:1 and 10:1, but increasing the protein or lipid concentration (while leaving ratios intact) reduced attraction. Thus, macronutritional ratios appear to be a primary factor driving bee pollen-foraging behavior and may explain observed patterns of host-plant visitation across the landscape. The nutritional quality of pollen resources should be taken into consideration when designing conservation habitats supporting bee populations.foraging behavior | nutritional ecology | pollen | pollinator | preferences F oraging animals must obtain appropriate nutrients for growth, development, and reproduction from their environments. Bees forage in a very complex and changing environment, where floral nutritional resources (nectar and pollen) vary widely in quality and quantity among plant species (1). These resources are accompanied by myriad floral cues, including floral odors, color, morphology, and display area, and can vary dramatically in spatiotemporal availability; all these factors may influence and reinforce foraging decisions (2, 3). Worldwide declines in populations of bees and other pollinators have been linked to reduced diversity and abundance of host-plant species, likely placing bees under nutritional stress (4, 5). To develop strategic conservation protocols that preserve or restore foraging habitat that supports healthy pollinator populations, we must understand how bees forage in their environments to meet their nutritional needs. It is well established that solitary and social insects can forage selectively and regulate their intake of synthetic diets spanning a range of macronutrient nutritional qualities to reach their optimal, species-specific nutritional intake (6-8). Here we examine whether the generalist bumble bee species Bombus...

Pollinator nutritional ecology provides insights into plant–pollinator interactions, coevolution, and the restoration of declining pollinator populations. Bees obtain their protein and lipid nutrient intake from pollen, which is essential for larval growth and development as well as adult health and reproduction. Our previous research revealed that pollen protein to lipid ratios (P:L) shape bumble bee foraging preferences among pollen host-plant species, and these preferred ratios link to bumble bee colony health and fitness. Yet, we are still in the early stages of integrating data on P:L ratios across plant and bee species. Here, using a standard laboratory protocol, we present over 80 plant species’ protein and lipid concentrations and P:L values, and we evaluate the P:L ratios of pollen collected by three bee species. We discuss the general phylogenetic, phenotypic, behavioral, and ecological trends observed in these P:L ratios that may drive plant–pollinator interactions; we also present future research questions to further strengthen the field of pollination nutritional ecology. This dataset provides a foundation for researchers studying the nutritional drivers of plant–pollinator interactions as well as for stakeholders developing planting schemes to best support pollinators.

SummaryA variety of methods are used in honey bee research and differ depending on the level at which the research is conducted. On an individual level, the handling of individual honey bees, including the queen, larvae and pupae are required. There are different methods for the immobilising, killing and storing as well as determining individual weight of bees. The precise timing of developmental stages is also an important aspect of sampling individuals for experiments. In order to investigate and manipulate functional processes in honey bees, e.g. memory formation and retrieval and gene expression, microinjection is often used. A method that is used by both researchers and beekeepers is the marking of queens that serves not only to help to locate her during her life, but also enables the dating of queens. Creating multiple queen colonies allows the beekeeper to maintain spare queens, increase brood production or ask questions related to reproduction. On colony level, very useful techniques are the measurement of intra hive mortality using dead bee traps, weighing of full hives, collecting pollen and nectar, and digital monitoring of brood development via location recognition. At the population level, estimation of population density is essential to evaluate the health status and using beelines help to locate wild colonies. These methods, described in this paper, are especially valuable when investigating the effects of pesticide applications, environmental pollution and diseases on colony survival. Métodos estándar diversos para la investigación en Apis mellifera ResumenEn la investigación de la abeja de la miel, se han usado una variedad de métodos que se diferencian en función del nivel en el que se realiza la investigación. Al nivel individual, el manejo de las abejas individuales es necesario, incluyendo a la reina, las larvas y las pupas. Existen diferentes métodos para la inmovilización, mortandad y almacenamiento, así como para la determinación del peso individual de las abejas. La precisión en la sincronización de las etapas de desarrollo es también un aspecto importante de los experimentos con muestreos individuales.La microinyección se utiliza a menudo con el fin de investigar y manipular los procesos funcionales de las abejas melíferas, como por ejemplo, la formación y recuperación de la memoria y la expresión génica. Un método utilizado tanto por investigadores como apicultores es el marcado de las reinas, que sirve no sólo para ayudar a localizarlas durante su vida, sino que también permite su datación. La creación de varias colmenas a partir de reinas permite al apicultor mantener reinas de repuesto, aumentar la producción de cría o hacer preguntas relacionadas con la reproducción. Al nivel de colmena, la medición de la mortalidad intra colmena utilizando trampas de abejas muertas, el pesaje de las colmenas completas, la recolección de polen y néctar, y el seguimiento digital del desarrollo de la cría a través del reconocimiento de su ubicación, son algunas de las técnicas más útiles. Al nivel p...

Bee population declines are linked to reduction of nutritional resources due to land-use intensification, yet we know little about the specific nutritional needs of many bee species. Pollen provides bees their primary source of protein and lipids, but nutritional quality varies widely among host-plant species. Therefore, bees may be adapted to assess resource quality and adjust their foraging behavior to balance nutrition from multiple food sources. We tested the ability of two bumble bee species, Bombus terrestris and B. impatiens, to regulate protein and lipid intake. We restricted B. terrestris adults to single synthetic diets varying in protein:lipid ratios (P:L). The bees overate protein on low fat diets and overate lipid on high fat diets to reach their targets of lipid and protein respectively. The bees survived best on a 10:1 P:L diet; the risk of dying increased as a function of dietary lipid when bees ate diets with lipid contents greater than 5:1 P:L. Hypothesizing that P:L intake target of adult worker bumble bees was between 25:1-5:1, we presented workers from both species unbalanced but complementary paired diets to determine if they self-select their diet to reach a specific intake target. Bees consumed similar amounts of proteins and lipids in each treatment and averaged a 14:1 P:L for B. terrestris and 12:1 P:L for B. impatiens. These results demonstrate that adult worker bumble bees likely select foods that provide them with a specific ratio of P:L. These P:L intake targets could affect pollen foraging in the field and help explain patterns of host-plant species choice by bumble bees.

Nearly 80% of all pesticides applied to row crops are herbicides, and these applications pose potentially significant ecotoxicological risks to nontarget plants and associated pollinators. In response to the widespread occurrence of weed species resistant to glyphosate, biotechnology companies have developed crops resistant to the synthetic-auxin herbicides dicamba and 2,4-dichlorophenoxyacetic acid (2,4-D); and once commercialized, adoption of these crops is likely to change herbicide-use patterns. Despite current limited use, dicamba and 2,4-D are often responsible for injury to nontarget plants; but effects of these herbicides on insect communities are poorly understood. To understand the influence of dicamba on pollinators, the authors applied several sublethal, drift-level rates of dicamba to alfalfa (Medicago sativa L.) and Eupatorium perfoliatum L. and evaluated plant flowering and floral visitation by pollinators. The authors found that dicamba doses simulating particle drift (≈1% of the field application rate) delayed onset of flowering and reduced the number of flowers of each plant species; however, plants that did flower produced similar-quality pollen in terms of protein concentrations. Further, plants affected by particle drift rates were visited less often by pollinators. Because plants exposed to sublethal levels of dicamba may produce fewer floral resources and be less frequently visited by pollinators, use of dicamba or other synthetic-auxin herbicides with widespread planting of herbicide-resistant crops will need to be carefully stewarded to prevent potential disturbances of plant and beneficial insect communities in agricultural landscapes.

Bee viral ecology is a fascinating emerging area of research: viruses exert a range of effects on their hosts, exacerbate impacts of other environmental stressors, and, importantly, are readily shared across multiple bee species in a community. However, our understanding of bee viral communities is limited, as it is primarily derived from studies of North American and European Apis mellifera populations. Here, we examined viruses in populations of A. mellifera and 11 other bee species from 9 countries, across 4 continents and Oceania. We developed a novel pipeline to rapidly and inexpensively screen for bee viruses. This pipeline includes purification of encapsulated RNA/DNA viruses, sequence-independent amplification, high throughput sequencing, integrated assembly of contigs, and filtering to identify contigs specifically corresponding to viral sequences. We identified sequences for (+)ssRNA, (−)ssRNA, dsRNA, and ssDNA viruses. Overall, we found 127 contigs corresponding to novel viruses (i.e. previously not observed in bees), with 27 represented by >0.1% of the reads in a given sample, and 7 contained an RdRp or replicase sequence which could be used for robust phylogenetic analysis. This study provides a sequence-independent pipeline for viral metagenomics analysis, and greatly expands our understanding of the diversity of viruses found in bee communities.

Foraging behavior is a critical adaptation by insects to obtain appropriate nutrients from the environment for development and fitness. Bumble bees (Bombus spp.) form annual colonies which must rapidly increase their worker populations to support rearing reproductive individuals before the end of the season. Therefore, colony growth and reproduction should be dependent on the quality and quantity of pollen resources in the surrounding landscape. Our previous research found that B. impatiens foraging preferences to different plant species were shaped by pollen protein:lipid nutritional ratios (P:L), with foragers preferring pollen species with a ~5:1 P:L ratio. In this study, we placed B. impatiens colonies in three different habitats (forest, forest edge, and valley) to determine whether pollen nutritional quality collected by the colonies differed between areas that may differ in resource abundance and diversity. We found that habitat did not influence the collected pollen nutritional quality, with colonies in all three habitats collecting pollen averaging a 4:1 P:L ratio. Furthermore, there was no difference in the nutritional quality of the pollen collected by colonies that successfully reared reproductives and those that did not. We found however, that “nutritional intake,” calculated as the colony‐level intake rate of nutrient quantities (protein, lipid, and sugar), was strongly related to colony growth and reproductive output. Therefore, we conclude that B. impatiens colony performance is a function of the abundance of nutritionally appropriate floral resources in the surrounding landscape. Because we did not comprehensively evaluate the nutrition provided by the plant communities in each habitat, it remains to be determined how B. impatiens polylectic foraging strategies helps them select among the available pollen nutritional landscape in a variety of plant communities to obtain a balance of key macronutrients.