A case study of the diet-microbiota-parasite interplay in bumble bees.

AIMS: Diets and parasites influence the gut bacterial symbionts of bumble bees, but potential interactive effects remain overlooked. The main objective of this study was to assess the isolated and interactive effects of sunflower pollen, its phenolamides, and the widespread trypanosomatid Crithidia sp. on the gut bacterial symbionts of Bombus terrestris males. METHODS AND RESULTS: Bumble bee males emerged in microcolonies fed on either (i) willow pollen (control), (ii) sunflower pollen, or (iii) willow pollen spiked with phenolamide extracts from sunflower pollen. These microcolonies were infected by Crithidia sp. or were pathogen-free. Using 16S rRNA amplicon sequencing (V3-V4 region), we observed a significant alteration of the beta diversity but not of the alpha diversity in the gut microbial communities of males fed on sunflower pollen compared to males fed on control pollen. Similarly, infection by the gut parasite Crithidia sp. altered the beta diversity but not the alpha diversity in the gut microbial communities of males, irrespective of the diet. By contrast, we did not observe any significant alteration of the beta or alpha diversity in the gut microbial communities of males fed on phenolamide-enriched pollen compared to males fed on control pollen. Changes in the beta diversity indicate significant dissimilarities of the bacterial taxa between the treatment groups, while the lack of difference in alpha diversity demonstrates no significant changes within each treatment group. CONCLUSIONS: Bumble bees harbour consistent gut microbiota worldwide, but our results suggest that the gut bacterial communities of bumble bees are somewhat shaped by their diets and gut parasites as well as by the interaction of these two factors. This study confirms that bumble bees are suitable biological surrogates to assess the effect of diet and parasite infections on gut microbial communities.

Trace metals with heavy consequences on bees: A comprehensive review.

The pervasiveness of human imprint on Earth is alarming and most animal species, including bees (Hymenoptera: Apoidea: Anthophila), must cope with several stressors. Recently, exposure to trace metals and metalloids (TMM) has drawn attention and has been suggested as a threat for bee populations. In this review, we aimed at bringing together all the studies (n = 59), both in laboratories and in natura, that assessed the effects of TMM on bees. After a brief comment on semantics, we listed the potential routes of exposure to soluble and insoluble (i.e. nanoparticle) TMM, and the threat posed by metallophyte plants. Then, we reviewed the studies that addressed whether bees could detect and avoid TMM in their environment, as well as the ways bee detoxify these xenobiotics. Afterwards, we listed the impacts TMM have on bees at the community, individual, physiological, histological and microbial levels. We discussed around the interspecific variations among bees, as well as around the simultaneous exposure to TMM. Finally, we highlighted that bees are likely exposed to TMM in combination or with other stressors, such as pesticides and parasites. Overall, we showed that most studies focussed on the domesticated western honey bee and mainly addressed lethal effects. Because TMM are widespread in the environment and have been shown to result in detrimental consequences, evaluating their lethal and sublethal effects on bees, including non-Apis species, warrants further investigations.

Fertilizer and herbicide alter nectar and pollen quality with consequences for pollinator floral choices.

Background: Pollinating insects provide economically and ecologically valuable services, but are threatened by a variety of anthropogenic changes. The availability and quality of floral resources may be affected by anthropogenic land use. For example, flower-visiting insects in agroecosystems rely on weeds on field edges for foraging resources, but these weeds are often exposed to agrochemicals that may compromise the quality of their floral resources. Methods: We conducted complementary field and greenhouse experiments to evaluate the: (1) effect of low concentrations of agrochemical exposure on nectar and pollen quality and (2) relationship between floral resource quality and insect visitation. We applied the same agrochemcial treatments (low concentrations of fertilizer, low concentrations of herbicide, a combination of both, and a control of just water) to seven plant species in the field and greenhouse. We collected data on floral visitation by insects in the field experiment for two field seasons and collected pollen and nectar from focal plants in the greenhouse to avoid interfering with insect visitation in the field. Results: We found pollen amino acid concentrations were lower in plants exposed to low concentrations of herbicide, and pollen fatty acid concentrations were lower in plants exposed to low concentrations of fertilizer, while nectar amino acids were higher in plants exposed to low concentrations of either fertilizer or herbicide. Exposure to low fertilizer concentrations also increased the quantity of pollen and nectar produced per flower. The responses of plants exposed to the experimental treatments in the greenhouse helped explain insect visitation in the field study. The insect visitation rate correlated with nectar amino acids, pollen amino acids, and pollen fatty acids. An interaction between pollen protein and floral display suggested pollen amino acid concentrations drove insect preference among plant species when floral display sizes were large. We show that floral resource quality is sensitive to agrochemical exposure and that flower-visiting insects are sensitive to variation in floral resource quality.

Evaluation of Tandem Mass Spectrometry Experiments in the Negative Ionization Mode for Phenolamide Regioisomer Characterization.

Phenolamides are abundant specialized metabolites found in nature and consist of hydroxycinnamic acids mono- or polyconjugated with polyamines. Their participation in flower development is well-documented, and their presence in pollen raises the question of their role in pollen/pollinator interactions. The structural characterization of phenolamides is complicated by the presence of positional isomers and stereoisomers. Liquid chromatography coupled to tandem mass spectrometry in the positive ionization mode is becoming very popular in phenolamide structural characterization. However, collision-induced transamidation processes that cause the swapping of side chains have been detected, making it difficult to distinguish regioisomers with this technique. In the present report, we explore the dissociation processes undergone by the [M - H]- ions of spermidine-based phenolamides as model compounds. We describe two original competitive dissociation routes, namely, the phenolate and imidate pathways, to account for the observed fragmentation reactions undergone by collisional activated standard phenolamide anions. Whereas the phenolate pathway is regioselective at the central position for spermidine, the imidate pathway, requiring a deprotonated amide, only occurs at the extremities. Tandem mass spectrometry experiments on negatively charged phenolamide ions may then outperform their positive ionization mode counterparts for the distinction between phenolamide regioisomers and globally for the identification of phenolamides in natural extracts.

Pollen as Bee Medicine: Is Prevention Better than Cure?

To face environmental stressors such as infection, animals may display behavioural plasticity to improve their physiological status through ingestion of specific food. In bees, the significance of medicating pollen may be limited by their ability to exploit it. Until now, studies have focused on the medicinal effects of pollen and nectar after forced-feeding experiments, overlooking spontaneous intake. Here, we explored the medicinal effects of different pollen on Bombus terrestris workers infected by the gut parasite Crithidia bombi. First, we used a forced-feeding experimental design allowing for the distinction between prophylactic and therapeutic effects of pollen, considering host tolerance and resistance. Then, we assessed whether bumble bees favoured medicating resources when infected to demonstrate potential self-medicative behaviour. We found that infected bumble bees had a lower fitness but higher resistance when forced to consume sunflower or heather pollen, and that infection dynamics was more gradual in therapeutic treatments. When given the choice between resources, infected workers did not target medicating pollen, nor did they consume more medicating pollen than uninfected ones. These results emphasize that the access to medicating resources could impede parasite dynamics, but that the cost-benefit trade-off could be detrimental when fitness is highly reduced.

Bumble Bee Breeding on Artificial Pollen Substitutes.

Bumble bees are important pollinators for many temperate crops. Because of the growing demand for food from entomophilous crops, bumble bee colonies are commercially reared and placed in fields or greenhouses to guarantee sufficient pollination services. Besides, commercial colonies are increasingly used in laboratories for various bioassays under controlled conditions. For both usages, bumble bee colonies are commonly provided with sugar solution and honey bee-collected pollen pellets. However, the latter display several disadvantages since they may contain pollutants, pathogens, or toxic phytochemicals. Consequently, companies have developed pollen-free artificial diets to sustain colonies. Such diets are designed to boost worker health in the field, in complement of floral pollen collected by workers outside the colonies, but their suitability in 'closed' systems without access to floral pollen, such as in laboratory bioassays, is arguable. Here, we used microcolonies of the commercially important bumble bee Bombus terrestris L. (Hymenoptera: Apidae) to assess the suitability of five artificial pollen substitutes and three mixed diets. We also assessed the evaporation rate of the different diets as it could impact their suitability. At the end of the bioassays, microcolonies fed the artificial diets showed a reduced offspring development when compared to microcolonies fed natural pollen, which was partly offset by mixing these diets with natural pollen. By contrast, the artificial diets did not have deleterious effects on worker's health. We discuss the potential nutritional and physical causes of artificial diets unsuitability for offspring development and encourage further research to accordingly establish appropriate pollen-free diets for bumble bee breeding.

Poison or Potion: Effects of Sunflower Phenolamides on Bumble Bees and Their Gut Parasite.

Specific floral resources may help bees to face environmental challenges such as parasite infection, as recently shown for sunflower pollen. Whereas this pollen diet is known to be unsuitable for the larval development of bumble bees, it has been shown to reduce the load of a trypanosomatid parasite (Crithidia bombi) in the bumble bee gut. Recent studies suggested it could be due to phenolamides, a group of compounds commonly found in flowering plants. We, therefore, decided to assess separately the impacts of sunflower pollen and its phenolamides on a bumble bee and its gut parasite. We fed Crithidia-infected and -uninfected microcolonies of Bombus terrestris either with a diet of willow pollen (control), a diet of sunflower pollen (natural diet) or a diet of willow pollen supplemented with sunflower phenolamides (supplemented diet). We measured several parameters at both microcolony (i.e., food collection, parasite load, brood development and stress responses) and individual (i.e., fat body content and phenotypic variation) levels. As expected, the natural diet had detrimental effects on bumble bees but surprisingly, we did not observe any reduction in parasite load, probably because of bee species-specific outcomes. The supplemented diet also induced detrimental effects but by contrast to our a priori hypothesis, it led to an increase in parasite load in infected microcolonies. We hypothesised that it could be due to physiological distress or gut microbiota alteration induced by phenolamide bioactivities. We further challenged the definition of medicinal effects and questioned the way to assess them in controlled conditions, underlining the necessity to clearly define the experimental framework in this research field.

Ozone Induces Distress Behaviors in Fig Wasps with a Reduced Chance of Recovery.

Among anthropogenic environmental risks, air pollution has the potential to impact animal and plant physiology, as well as their interactions and the long-term survival of populations, which could threaten the functioning of ecosystems. What is especially alarming is that the concentration of tropospheric ozone (O3) has dramatically increased since pre-industrial times. However, the direct effects of O3 on the behavior of pollinators themselves have not been investigated so far even though insect behavior is key to their ecological interactions, which underpin the stability of ecological networks responsible for species biodiversity in ecosystems. In this study, we aim to determine the potential effects of O3 episodes at different field-realistic concentrations (0, 40, 80, 120, and 200 ppb for 60 min) on the behavior of the fig wasp Blastophaga psenes by monitoring exposed individuals hourly for 5 h after exposure. We found that ozone episodes induced major changes in insect behavior, which were already significant at 80 ppb with individuals displaying abnormal motility. The tracking over time clearly showed that exposed individuals might only have a reduced chance of recovery, with a decreasing proportion of active fig wasps despite the cessation of an O3 episode. These findings illustrate that O3 episodes can affect pollinator behavior, which may have detrimental implications for pollination systems. It is, therefore, of importance to assess the effects of O3 on insect behavior in order to predict how it could modify ecological interactions and species biodiversity in ecosystems.

Ozone Pollution Alters Olfaction and Behavior of Pollinators.

Concentration of air pollutants, particularly ozone (O3), has dramatically increased since pre-industrial times in the troposphere. Due to the strong oxidative potential of O3, negative effects on both emission and lifetime in the atmosphere of plant volatile organic compounds (VOCs) have already been highlighted. VOCs alteration by O3 may potentially affect the attraction of pollinators that rely on these chemical signals. Surprisingly, direct effects of O3 on the olfaction and the behavioral response of pollinators have not been investigated so far. We developed a comprehensive experiment under controlled conditions to assess O3 physiological and behavioral effects on two pollinator species, differing in their ecological traits. Using several realistic concentrations of O3 and various exposure times, we investigated the odor antennal detection and the attraction to VOCs present in the floral scents of their associated plants. Our results showed, in both species, a clear effect of exposure to high O3 concentrations on the ability to detect and react to the floral VOCs. These effects depend on the VOC tested and its concentration, and the O3 exposure (concentration and duration) on the pollinator species. Pollination systems may, therefore, be impaired in different ways by increased levels of O3, the effects of which will likely depend on whether the exposure is chronic or, as in this study, punctual, likely causing some pollination systems to be more vulnerable than others. While several studies have already shown the negative impact of O3 on VOCs emission and lifetime in the atmosphere, this study reveals, for the first time, that this impact alters the pollinator detection and behavior. These findings highlight the urgent need to consider air pollution when evaluating threats to pollinators.

Global warming and plant-pollinator mismatches.

The mutualism between plants and their pollinators provides globally important ecosystem services, but it is likely to be disrupted by global warming that can cause mismatches between both halves of this interaction. In this review, we summarise the available evidence on (i) spatial or (ii) phenological shifts of one or both of the actors of this mutualism. While the occurrence of future spatial mismatches is predominantly theoretical and based on predictive models, there is growing empirical evidence of phenological mismatches occurring at the present day. Mismatches may also occur when pollinators and their host plants are still found together. These mismatches can arise due to (iii) morphological modifications and (iv) disruptions to host attraction and foraging behaviours, and it is expected that these mismatches will lead to novel community assemblages. Overall plant-pollinator interactions seem to be resilient biological networks, particularly because generalist species can buffer these changes due to their plastic behaviour. However, we currently lack information on where and why spatial mismatches do occur and how they impact the fitness of plants and pollinators, in order to fully assess if adaptive evolutionary changes can keep pace with global warming predictions.

Asteraceae Paradox: Chemical and Mechanical Protection of Taraxacum Pollen.

Excessive pollen harvesting by bees can compromise the reproductive success of plants. Plants have therefore evolved different morphological structures and floral cues to narrow the spectrum of pollen feeding visitors. Among "filtering" mechanisms, the chemical and mechanical protection of pollen might shape bee-flower interactions and restrict pollen exploitation to a specific suite of visitors such as observed in Asteraceae. Asteraceae pollen is indeed only occasionally exploited by generalist bee species but plentifully foraged by specialist ones (i.e., Asteraceae paradox). During our bioassays, we observed that micro-colonies of generalist bumblebee (Bombus terrestris L.) feeding on Taraxacum pollen (Asteraceae) reduced their pollen collection and offspring production. Bees also experienced physiological effects of possible defenses in the form of digestive damage. Overall, our results suggest the existence of an effective chemical defense in Asteraceae pollen, while the hypothesis of a mechanical defense appeared more unlikely. Pre- and post-ingestive effects of such chemical defenses (i.e., nutrient deficit or presence of toxic compounds), as well as their role in the shaping of bee-flower interactions, are discussed. Our results strongly suggest that pollen chemical traits may act as drivers of plant selection by bees and partly explain why Asteraceae pollen is rare in generalist bee diets.

Pollen Protein: Lipid Macronutrient Ratios May Guide Broad Patterns of Bee Species Floral Preferences.

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.

Generalized host-plant feeding can hide sterol-specialized foraging behaviors in bee-plant interactions.

Host-plant selection is a key factor driving the ecology and evolution of insects. While the majority of phytophagous insects is highly host specific, generalist behavior is quite widespread among bees and presumably involves physiological adaptations that remain largely unexplored. However, floral visitation patterns suggest that generalist bees do not forage randomly on all available resources. While resource availability and accessibility as well as nectar composition have been widely explored, pollen chemistry could also have an impact on the range of suitable host-plants. This study focuses on particular pollen nutrients that cannot be synthesized de novo by insects but are key compounds of cell membranes and the precursor for molting process: the sterols. We compared the sterol composition of pollen from the main host-plants of three generalist bees: Anthophora plumipes, Colletes cunicularius, and Osmia cornuta, as well as one specialist bee Andrena vaga. We also analyzed the sterols of their brood cell provisions, the tissues of larvae and nonemerged females to determine which sterols are used by the different species. Our results show that sterols are not used accordingly to foraging strategy: Both the specialist species A. vaga and the generalist species C. cunicularius might metabolize a rare C27 sterol, while the two generalist species A. plumipes and O. cornuta might rather use a very common C28 sterol. Our results suggest that shared sterolic compounds among plant species could facilitate the exploitation of multiple host-plants by A. plumipes and O. cornuta whereas the generalist C. cunicularius might be more constrained due to its physiological requirements of a more uncommon dietary sterol. Our findings suggest that a bee displaying a generalist foraging behavior may sometimes hide a sterol-specialized species. This evidence challenges the hypothesis that all generalist free-living bee species are all able to develop on a wide range of different pollen types.

Comparison of the Sex Pheromone Composition of Harmonia axyridis Originating from Native and Invaded Areas.

The multicolored Asian lady beetle, Harmonia axyridis (Coleoptera: Coccinellidae), originates from South-East Asia and is now considered as an invasive species at a worldwide scale, with populations encountered in North and South America, Africa, and Europe. Several previous studies suggested that invasive populations display different behavioral and physiological traits, leading to a better fitness than native individuals. H. axyridis sex pheromone was identified recently, but only from individuals established in Europe. In this study, we compare the composition of the female sex pheromone of H. axyridis from two populations: (i) an invasive population in North America, and (ii) a native population in South-East China. We found the females originating from both populations to release in similar proportions the same five pheromonal compounds, namely ß-caryophyllene, ß-elemene, methyl-eugenol, α-humulene, and α-bulnesene. However, females from the North American strain release all five compounds in larger amount than the Chinese ones. Whether invasive individuals were selected during the process of invasion through their capacity to better call and find sexual partners remains to be confirmed.

Ensuring access to high-quality resources reduces the impacts of heat stress on bees.

Pollinators are experiencing declines globally, negatively affecting the reproduction of wild plants and crop production. Well-known drivers of these declines include climatic and nutritional stresses, such as a change of dietary resources due to the degradation of habitat quality. Understanding potential synergies between these two important drivers is needed to improve predictive models of the future effects of climate change on pollinator declines. Here, bumblebee colony bioassays were used to evaluate the interactive effects of heat stress, a reduction of dietary resource quality, and colony size. Using a total of 117 colonies, we applied a fully crossed experiment to test the effect of three dietary quality levels under three levels of heat stress with two colony sizes. Both nutritional and heat stress reduced colony development resulting in a lower investment in offspring production. Small colonies were much more sensitive to heat and nutritional stresses than large ones, possibly because a higher percentage of workers helps maintain social homeostasis. Strikingly, the effects of heat stress were far less pronounced for small colonies fed with suitable diets. Overall, our study suggests that landscape management actions that ensure access to high-quality resources could reduce the impacts of heat stress on bee decline.

Stressful conditions reveal decrease in size, modification of shape but relatively stable asymmetry in bumblebee wings.

Human activities can generate a wide variety of direct and indirect effects on animals, which can manifest as environmental and genetic stressors. Several phenotypic markers have been proposed as indicators of these stressful conditions but have displayed contrasting results, depending, among others, on the phenotypic trait measured. Knowing the worldwide decline of multiple bumblebee species, it is important to understand these stressors and link them with the drivers of decline. We assessed the impact of several stressors (i.e. natural toxin-, parasite-, thermic- and inbreeding- stress) on both wing shape and size and their variability as well as their directional and fluctuating asymmetries. The total data set includes 650 individuals of Bombus terrestris (Hymenoptera: Apidae). Overall wing size and shape were affected by all the tested stressors. Except for the sinigrin (e.g. glucosinolate) stress, each stress implies a decrease of wing size. Size variance was affected by several stressors, contrary to shape variance that was affected by none of them. Although wing size directional and fluctuating asymmetries were significantly affected by sinigrin, parasites and high temperatures, neither directional nor fluctuating shape asymmetry was significantly affected by any tested stressor. Parasites and high temperatures led to the strongest phenotype modifications. Overall size and shape were the most sensitive morphological traits, which contrasts with the common view that fluctuating asymmetry is the major phenotypic marker of stress.

Integration of non-targeted metabolomics and automated determination of elemental compositions for comprehensive alkaloid profiling in plants.

Plants produce a large array of specialized metabolites to protect themselves. Among these allelochemicals, alkaloids display highly diverse and complex structures that are directly related to their biological activities. Plant alkaloid profiling traditionally requires extensive and time-consuming sample preparation and analysis. Herein, we developed a rapid and efficient approach for the comprehensive profiling of alkaloids in plants using ultrahigh performance liquid chromatography-high resolution mass spectrometry (UHPLC-HRMS)-based metabolomics. Using automated compound extraction and elemental composition assignment, our method achieved >83% correct alkaloid identification and even >90% for medium to high intensity peaks. This represented a significant improvement in identification rate compared to generic methods used for EC determination with no a priori, such as in untargeted metabolomics studies. The developed approach was then applied to identify specific alkaloids of Aconitum lycoctonum L. and A. napellus L. (Ranunculaceae) using different parts of the plant (leaf, perianth and pollen). Significant differences in alkaloid profiles between the two species were highlighted and discussed under taxonomic and evolutionary perspectives. Taken together, the presented approach constitutes a valuable chemotaxonomic tool in the search for known and unknown alkaloids from plants.

Is non-host pollen suitable for generalist bumblebees?

Current evidence suggests that pollen is both chemically and structurally protected. Despite increasing interest in studying bee-flower networks, the constraints for bee development related to pollen nutritional content, toxicity and digestibility as well as their role in the shaping of bee-flower interactions have been poorly studied. In this study we combined bioassays of the generalist bee Bombus terrestris on pollen of Cirsium, Trifolium, Salix, and Cistus genera with an assessment of nutritional content, toxicity, and digestibility of pollen. Microcolonies showed significant differences in their development, non-host pollen of Cirsium being the most unfavorable. This pollen was characterized by the presence of quite rare δ7-sterols and a low digestibility. Cirsium consumption seemed increase syrup collection, which is probably related to a detoxification mixing behavior. These results strongly suggest that pollen traits may act as drivers of plant selection by bees and partly explain why Asteraceae pollen is rare in bee generalist diet.

Invasive plants as potential food resource for native pollinators: A case study with two invasive species and a generalist bumble bee.

It is now well established that invasive plants may induce drifts in the quantity and/or quality of floral resources. They are then often pointed out as a potential driver of bee decline. However, their impact on bee population remains quite unclear and still controversial, as bee responses are highly variable among species. Here, we compared the amino acid composition of pollen from three native and two invasive plant species included in diets of common pollinators in NW Europe. Moreover, the nutritional intake (i.e., pollen and amino acid intakes) of Bombus terrestris colonies and the pollen foraging behaviour of workers (i.e., visiting rate, number of foraging trips, weight of pollen loads) were considered. We found significant differences in pollen nutrients among the studied species according to the plant invasive behaviour. We also found significant differences in pollen foraging behaviour according to the plant species, from few to several foraging trips carrying small or large pollen loads. Such behavioural differences directly impacted the pollen intake but depended more likely on plant morphology rather than on plant invasive behaviour. These results suggest that common generalist bumble bees might not always suffer from plant invasions, depending on their behavioural plasticity and nutritional requirements.

The importance of pollen chemistry in evolutionary host shifts of bees.

Although bee-plant associations are generally maintained through speciation processes, host shifts have occurred during evolution. Understanding shifts between both phylogenetically and morphologically unrelated plants (i.e., host-saltation) is especially important since they could have been key processes in the origin and radiation of bees. Probably far from being a random process, such host-saltation might be driven by hidden constraints associated with plant traits. We selected two clades of oligolectic bees (i.e., Colletes succinctus group and Melitta leporina group) foraging on co-flowering but unrelated host-plants to test this hypothesis. We analyzed floral scent, floral color and chemical composition of pollen from host and non-host plants of these two clades. We did not find evidence for host-plant evolution in the Melitta leporina group driven by one of the assayed floral traits. On the contrary, hosts of the C. succinctus group display similar primary nutritive content of pollen (i.e., amino acids and sterols) but not similar floral scent or color, suggesting that shared pollen chemistry probably mediates saltation in this clade. Our study revealed that constraints shaping floral associations are diverse and clearly depend on species life-history traits, but evidence suggests that pollen chemistry may act as a major floral filter and guide evolutionary host-shifts.