https://orcid.org/0000-0002-9737-4820
Figures
Abstract
Chemical communication is a widely used mode of communication for social insects and has been demonstrated to be involved in many behaviours and physiological processes such as reproduction, nutrition or the fight against parasites and pathogens. In the honey bee, Apis mellifera, the release of chemical compounds by the brood plays a role in worker behaviour, physiology, and foraging activities and colony health as a whole. Several compounds have already been described as brood pheromones, such as components of the brood ester pheromone and (E)-β-ocimene. Several other compounds originating from diseased or varroa-infested brood cells have been described as triggering the hygienic behaviour of workers. So far, studies of brood emissions have focused on specific stages of development and little is known about the emission of volatile organic compounds by the brood. In this study, we investigate the semiochemical profile of worker honey bee brood during its whole developmental cycle, from egg to emergence, with a specific focus on volatile organic compounds. We describe variation in emissions of thirty-two volatile organic compounds between brood stages. We highlight candidate compounds that are particularly abundant in specific stages and discuss their potential biological significance.
Citation: Noël A, Dumas C, Rottier E, Beslay D, Costagliola G, Ginies C, et al. (2023) Detailed chemical analysis of honey bee (Apis mellifera) worker brood volatile profile from egg to emergence. PLoS ONE 18(2): e0282120. https://doi.org/10.1371/journal.pone.0282120
Editor: Olav Rueppell, University of Alberta, CANADA
Received: July 5, 2022; Accepted: February 7, 2023; Published: February 21, 2023
Copyright: © 2023 Noël 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: All relevant data are within the manuscript and its Supporting Information files.
Funding: This study was funded by a PRIMA grant (EU H2020, #1812, “Plant-B”). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Communication is fundamental to all life forms. Chemical communication, i.e. the emission of specific chemical compounds by an individual and received by another at a specific moment, is highly present in the living world. In eusocial insects, compounds emitted and perceived by individuals of the same species constitute a pheromonal message, which is part of the chemical language repertoire. Individuals receiving this chemical message will respond by adapting their behaviour or their physiology [1–3]. This type of communication is key to the regulation of superorganism homeostasis, as it allows eusocial species to organise the colony and the division of labour [4–6].
In the honey bee, Apis mellifera, chemical communication is widely deployed across castes and individuals [3]. It is well known that the brood communicates with the adults through the use of pheromones. An iconic example of brood chemical communication is the brood ester pheromone (BEP), composed of by a mixture of 10 fatty esters (methyl and ethyl oleate, methyl and ethyl linoleate, methyl and ethyl linolenate, methyl and ethyl palmitate, methyl and ethyl stearate) [7]. Another well-known brood pheromone is (E)-β-ocimene [8]. These two pheromones have both releaser (i.e. triggers an immediate behavioural response) and primer effects (i.e. changes the physiology of the receiver). As releaser effects, BEP triggers worker brood-rearing behaviour and cell-capping behaviour (closing brood cells with wax) [7, 9], and (E)-β-ocimene increases pollen foraging and brood-feeding behaviour [10, 11]. A common primer effect of these two pheromones is the inhibition of worker ovarian growth [8, 12].
Dead brood emit oleic acid and (E)-β-ocimene, which both trigger hygienic behaviour (detection of sick brood and cleaning of the brood cell), suggesting a death-signalling role for these two compounds [13]. Hygienic behaviour is also performed on diseased brood, before it dies. This is the case for brood infected by chalkbrood (Ascosphaera apis). In order not to become a vector of contamination, adults must detect brood parasitized by the fungus before the spores reach the cuticle of the larva. The hygienic behaviour is then triggered by the emission of phenethyl acetate, 2-phenylethanol, and benzyl alcohol [14]. A variant of hygienic behaviour, VSH (Varroa Sensitive Hygiene) behaviour, targets brood cells parasitized by the varroa mite and is triggered by compounds emitted by parasitized brood cells ((Z)-6-pentadecene, (Z)-10-tritriacontene, 6 VPS (Varroa-Parasitization-Specific: tricosan-2-one, pentacosan-2-one, tetracosyl acetate, heptacosan-2-one, hexacosyl acetate and nonacosan-2-one), α-pinene and ethyl hexanoate) [15–18]. Stressed or diseased brood emit other compounds but potential behavioural responses of workers have not been assayed. Such is the case for γ-octalactone, a compound that larvae infected with the European foulbrood emit at higher levels than healthy brood (Melissococcus. plutonius) [19]. Varroa destructor infestation and Deformed Wing Virus infection increase the emission of ten chemical compounds in brood targeted by hygienic behaviour (pentacosane, hexacosane, 4-methyltetracosane, 9- and 11-methyltricosane, 11- et 13-methylpentacosane, 12- et 14-methylhexacosane, pentacosene, heptacosene, hentriacontene) [20]. With the exception of a few volatile organic compounds (VOC), the majority of chemical compounds already described as brood emissions that trigger adult behaviours are not very volatile. Little is known about highly volatile compounds of honey bee brood; studies have tended to focus on low-volatile compounds. Unlike cuticular hydrocarbons, little is known about VOC emission of brood in the other social insect [4, 21, 22]. The role of the brood on the behaviour (e.g. foraging regulation) and physiology (e.g. ovary maturation) of adults in social insects is well established. For example, in ants, brood has been shown to have a role in adult physiology and behaviour, but identification of this potential ant brood pheromone is still lacking [see review: 23]. In bumblebees, the pupal odour attracts the workers but has no impact on the physiology of the adult. The concept of a brood pheromone in the bumblebee is still discussed [24].
The emission of highly volatile chemical compounds by the brood allows information to be disseminated quickly and over a long distance within the nest. Studying emission of VOC by the brood would provide a better understanding of its chemical communication and development.
Thus far, no documentation of VOC emissions over the entire brood development is available, as studies describing VOC emissions from brood focus on specific stages of development and/or on specific compounds [25–27]. Maisonnasse et al. [27] showed the difference in (E)-β-ocimene emission between larvae and pupae (first larval stage to fifth pupal stage). Carroll & Duehl [25] carried out in situ VOC captures on unstressed brood from hatching to the pre-pupal stage, in the presence of nurses on the frame. The authors showed changes in the emission of six VOC from the brood according to age (2-heptanone, 3-carene, β-ocimene, methyl benzoate, octanoic acid and decanal) and reported the presence of 22 VOC in total (2-heptanol, 2-heptanone, 2-pentanone, 3-carene, 3-methyl-1-butanol, 6-methyl-5-hepten-2-one, α-pinene, citral, decanal, decane, ethyl benzoate, (E)-β-ocimene, geraniol, heptanal, hexanal, isobutyl acetate, isopentyl acetate, methyl benzoate, nerol, nonanal, octanal and nonane) [25]. Light et al. [26] also captured VOC in situ emitted by worker and drone brood at different developmental stages (last day of egg, first instar larvae, first, third and fifth instar pupae for drone brood, fifth instar pupae for cold-killed drone brood, fifth instar pupae for worker brood). They identified 75 VOC emitted by worker and drone brood frames that triggered an electrophysiological response in V. destructor. These different studies give an indication as to the VOC emitted by bee brood but only provide a fragmented view of VOC emissions during its development.
Similarly to "-omic" studies (genomics, transcriptomics, proteomics, metabolomics and metagenomics), capturing VOC emissions of the brood throughout its development, from oviposition to emergence, makes it possible to observe and capture the essence of brood-adult chemical communication in its entirety. Studying VOC emissions has led to a better understanding of reproduction in beetles [28] and the estimation of larval and pupal age in blowflies [29, 30]. In honey bees, VOC collection permits the study of the changes in larval food consumption and pheromonal emission over the course of their development [25, 27].
Bee brood development is characterised by a sequence of different physiological stages. The brood passes from an open brood state comprising the egg stage and five larval stages, to a capped brood state (i.e. brood cells closed with a wax cap by workers) before the emergence of the adult bee, comprising a sealed larval stage, a prepupal stage, and nine pupal stages [6]. Physiological changes in the brood can be observed through the brood haemolymph proteome [31, 32]. These important physiological changes can also be observed in brood-nurse interactions according to the developmental stage. Under the influence of brood pheromones, workers adapt the food they give to the brood according to the age of the larvae [9, 33, 34], cap brood [7], modify their foraging behaviour [10, 35], and realise hygienic behaviour [13–17].
In this study, we develop an in situ experimental approach to analyse VOC emissions of healthy worker brood during the whole developmental cycle (21 days, from egg stage to adult bee emergence). Our target is to highlight the temporal changes in brood-characterising compound presence and abundance during the entire brood development. We highlight semiochemical compounds emitted by brood that may have a role in the communication between brood and workers.
Materials and methods
Honey bee colonies and experimental setup
This study was performed with nine local honey bee colonies (A. mellifera) installed at the INRAE (Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement) research centre in Avignon (France). As the aim of this study was to capture volatile organic compounds (VOC) in healthy brood, we were very careful to use colonies that were free of parasites and disease. The health of these colonies was closely monitored throughout the experiment: colonies were qualitatively screened by our beekeepers to confirm the absence of brood illness. This study was conducted from April to July 2021, when varroa infestation is at its lowest, to maximize chances that capped brood cells were not infested by the parasite.
To ensure working with brood of a homogeneous age, queens of the nine selected colonies were caged on an empty frame for 24h to allow for egg laying. After that period, the queen was removed from the cage and no longer allowed access to the laid frame (brood frame). An empty built frame (no brood, pollen or nectar) was concomitantly used as a control in each colony (control frame). On the first day of the experiment (following queen laying), circles containing about 130 cells were drawn on both brood and control frames to select cell patches to be monitored daily, for 21 days. Each day, brood and control frames were removed from their colonies and brought to the laboratory for chemical analysis. Experimental procedures were optimised to minimise any source of stress. To do so, adult bees were gently removed from the brood frame with a soft brush, and the brood was placed directly in a warm, moist box during the short journey from the hives to the experimental room. They were kept in a room with controlled temperature and humidity (34–35°C, 65% RH). The same patch of cells was monitored each day (Fig 1), and the content of each cell was recorded on a transparent plastic sheet.
Every day for 21 days, brood and control frames were removed from their colonies. VOC of the same patch of cells (brood and control) were captured for 20 minutes with a SPME fibre, in a room with controlled temperature and humidity (34–35°C, 65% RH). After capture, frames were immediately put back in their colony while fibres were desorbed on a GC-MS. Nine colonies were used, each with a control and a brood frame (371 captures in total).
Volatile organic compound capture and analysis
VOC capture was conducted passively, using a polydimethylsiloxane divinylbenzene (PDMS DVB) Solid Phase Micro Extraction (SPME) fibre (65 μm, manual holder). The fibre was inserted inside a glass funnel (Ø 70 mm) placed upside down on the patch of cells to be monitored (Fig 1). For each capture, the funnel was pressed against the waxy surface to form a hermetic junction between the sample (cell patch) and the capture system. Captures lasted for 20min, after which frames were returned to their colonies. Each capture of a brood sample was paired with that of a control sample (seven control samples could not be captured due to bad weather conditions on day 11 and 12), thus forming a total of 371 captures (9 control + 9 brood samples per day). Qualitative and quantitative analyses of VOC were performed on a gas chromatograph coupled with mass spectrometry (GC-MS). SPME fibres were desorbed into a GC-MS (Shimadzu GCMS-QP2010SE), equipped with an electronic impact ion source (70 eV) and a ZB 5 MS column (Zebron ZB 5ms, 20 m x 0.18 mm, 0.18 μm thickness). Desorption was realised in splitless mode, with the constant column flow at 0.82 mL/min; the carrier gas was helium. The injector temperature was set at 250°C and the oven temperature program as follows: 45°C isothermal held for 5 min, followed by temperature increases at a rate of 10°C/min up to 250°C. The oven was finally held at 250°C for 4 min. The ion source was set at 200°C and m/z scanned from 40 to 350 amu. Compound identification was performed by comparison of their mass spectra and retention indices with those found in the NIST 2017 library (v2.3). This identification was further confirmed by injection of the reference standards purchased from Sigma Aldrich (except for hexenyl acetate (s24) for which the location of the unsaturation remains unknown).
Statistical analyses
Statistical analyses were performed using R (version 4.0.5). To account for non-linear trends across time, generalized additive mixed models (GAMM) were fit on total peak area for each day and each sample group (brood, control) to investigate differences between control and brood samples and their evolution over time (R package “mgcv” v1.8–39; https://cran.r-project.org/web/packages/mgcv/index.html [36]).
Three redundancy analyses (RDA) were performed on three different phases of development, based on GAMM results (“Eggs”: day 1 to day 3; “Larvae and Non-Sclerotised Capped Brood”: day 4 to day 16; “Sclerotised Pupae”: day 17 to day 21), to identify compounds that characterise the brood (R package “vegan” v2.6–2; https://cran.r-project.org/web/packages/vegan/index.html). Day and sample group (brood and control) and their interaction were included as fixed factors in the RDA to identify brood-characterising compounds. To graphically represent the RDAs, we selected the two components which explain the largest proportion of the data variance. Compounds detected in at least five out of the nine replicates in brood samples for any given day were included in the analyses. From the 3 RDAs, a list of 32 brood-characterising compounds was compiled. We then split brood development into eight characteristic periods to study the compounds or chemical families that characterised each period, as well as those that covaried together. The four days around the capping of brood cells by the workers, represent a key moment in the development of the brood, both physiologically and behaviourally [3, 6, 7, 37]. For this reason, those four days were separated into four one-day periods. The eight periods were thus defined as follows: Eggs: days 1–3; 1st to 4th instar larvae (L1-L4): days 4–7; 5th instar larvae (L5): day 8; Capping: day 9; Sealed larvae (SL): day 10; Prepupae (PP): day 11; Non-sclerotised pupae (NSP): day 12–16; and Sclerotised pupae (SP): days 17–21.
In order to classify the data, we chose to group the compounds found according to their respective chemical families. We chose this presentation not because of the potential effects on behaviour, but rather to highlight a classification based on the chemical structure of these molecules, in particular to evaluate shared patterns of variation and stage-specific emission.
As our target is to highlight the temporal changes in the presence and abundance of brood-characterising compounds during development, we normalised brood compound area values by those measured in the control (i.e. (Abrood—Acontrol)/Acontrol) for each day, compound, and colony. To avoid division by zero, a constant ten times lower than the smallest non-zero value was added to the data. We then calculated the median normalised area of each compound across the nine colonies for each day. Negative median values, indicating a higher amount of the compound in control samples than in brood samples, were set at zero. Compounds were subsequently quantified relative to the highest median normalised value observed across the 21 capture days. We represented these relative median normalized values in a heatmap with a log10 scale (adding a constant of 1). We similarly calculated normalised brood compound area by stage, analogous to that done by day. For two compounds, these stage-specific median normalised values were equal to 0 as they were observed on a single day of the four days comprising the SP stage. As before, compounds were quantified relative to the highest median value observed across the 8 developmental stages. The evolution across stages of relative median normalised compound emissions was visualized in another heatmap. We realised a clustering of the arcsine-transformed relative median normalised values per stage using a Gaussian mixture model to highlight compounds that shared stage-specific maximum emissions (R package “coseq” version 1.17.2; https://bioconductor.org/packages/release/bioc/html/coseq.html [38]). The number of clusters was identified based on the integrated completed likelihood (ICL) model selection criterion.
For comparisons with the scientific literature, we calculated the estimated emission per mg of individuals weight for (E)-β-ocimene. As the tracking of the individuals throughout the experiment did not allow us to weigh the brood, we instead used in-house brood weight data (first larval to fifth pupal stage, day 4 to day 15 in this paper, n = 9 per stage) obtained in a previous experiment on similar colonies.
Results
Total compound emission profile across development
A comparison of total peak area between control and brood samples revealed that all compounds (total peak area) are significantly more abundant in brood samples than in control samples (GAMM; Brood: estimate = 296,010, p < 2e-16; control: estimate = 200,479, p < 2e-16) (Fig 2). The quantity of compounds fluctuates significantly from 2.6e5 (day 16) to 5.4e5 (day 11) of total compound area (GAMM; Brood over time: F = 12.403, p < 2e-16; Control over time: F = 0.078, p = 0.820) (Fig 2). The first three days (egg stage) and the last five days (sclerotised pupal stages) are periods of low overall emissions, with small differences between brood and control samples (Fig 2). From day 4 (first instar larvae) to day 16 (fourth instar pupae), total compound abundance is quite variable and strikingly different between brood and control samples, with a notable peak of brood emission around days 10 and 11 (prepupae) (Fig 2). Based on these tendencies, and to improve the brood-characterising compound determination step, we separated our dataset into three groups to perform multivariate redundancy analyses (RDA): Egg (day 1 to day 3), Larvae and Non-Sclerotised Capped Brood (day 4 to day 16) and Sclerotised Pupae (day 17 to day 21).
Lines show the mean total area and the surrounding 95% standard error, for brood (blue) and control (red) samples. Dotted lines separate the three groups: “Eggs” (day 1 to 3), “Larvae and non-sclerotised capped brood” (day 4 to 16) and “Sclerotised Pupae” (day 17 to 21) considered for the multivariate analyses.
Identification of brood-characterising compounds
Based on the total emission analysis and the identification of 3 distinct phases of compound emission, we performed three redundancy analyses (RDAs).
In the largest RDA (day 4 to day 16, “Larvae and non-sclerotised capped brood” group), the first two components of the RDA explained 56.82% of the variance. Both day and sample group, as well as their interaction, were significant (Table 1). The two sample groups (brood and control) were well separated on the first component (Fig 3). Based on this analysis, twenty-seven compounds characterising the brood samples were retained (s01, s02, s03, s05, s09, s12, s13, s15, s17, s18, s24, s27, s28, s29, s30, s32, s36, s37, s38, s41, s42, s47, s52, s54, s65, s71, s74, s75, s76) (Fig 3, Table 2).
For each day: brood sample n = 9, control sample n = 9. Each dot represents a sample (blue: brood, red: control). Compounds are represented by their IDs and grey arrows. The compounds characterising the brood and retained for the rest of the analysis are shown in bold.
For each day: brood sample n = 9, control sample n = 9.
Days of detection represent periods where a given compound is found in at least 5 replicates in brood samples for a given day. *: Unsaturation remains unknown, **: Stereochemistry remains unknown.
In the “Egg” RDA analysis (days 1–3), the first two components of the RDA explained 73.95% of the variance. Both day and sample group were significant (Table 1), but the interaction was not, confirming that overall brood and control compound emissions maintain the same temporal trend. Both sample groups (brood and control) were also well separated on the first component (S1 Fig). Five compounds characterising brood samples, all included in the previous list, were highlighted by the RDA (s15, s18, s30, s37, s52) (S1 Fig, Table 2).
In the “Sclerotised pupae” RDA (day 17 to day 21), the first two components explained 64.07% of the variance. Both single factors were significant, but their interaction was not (Table 1). The two sample groups were once again well separated on the first component (S1 Fig). Twenty-eight compounds characterising the brood sample were retained, partially overlapping with the previous lists (s01, s02, s03, s05, s09, s10, s12, s13, s15, s17, s18, s27, s28, s30, s32, s35, s37, s38, s41, s45, s47, s54, s65, s71, s73, s74, s75, s76) (S1 Fig, Table 2).
Using these three RDAs, we were able to compile a final list of 32 VOC characterising the brood samples (S2 Fig, Table 2).
Evolution of brood-characterising compound emission during brood development
Each day of brood development was characterised by different emission of VOC. The 32 VOC could be classified into two groups; a group of compounds with a marked variation of emission over time and a group with a more constant emission. The group of compounds fluctuating over time included 15 compounds (in order of appearance in the Fig 4, from top to bottom: s27, s30, s42, s52, s18, s10, s12, s47, s36, s29, s73, s03, s09, s02 and s01) (Fig 4). Except for α-pinene (s15), all terpenes found in this study were in this group (s27, s30, s42, s18, s29 and s73, Fig 4). The same pattern applied to the two ketones (s12 and s36, Fig 4).
The temporal evolution of each compound area is represented in rows, with a colour gradient representing the median normalized peak area relative to the respective maximum for each compound (in log scale). The colour change represents a change in emission of the compound, with a maximum emission in yellow and a minimum in dark blue. Each column represents the age of brood development. Compounds are labelled with their IDs and chemical names. A dendrogram representing a hierarchical clustering (Euclidean distance, complete linkage) of compounds is indicated on the left.
The group of compounds with a more constant emission over time included 17 compounds (s37, s15, s74, s71, s28, s38, s65, s75, s54, s41, s05, s13, s24, s32, s35, s45 and s17) (Fig 4). The seven alkanes found in this study were in this group (s74, s71, s65, s75, s54, s05 and s13), as well as the aldehyde (s28) (Fig 4). For the alcohols and esters found in this study, their fluctuation over time depends on the compound. Four alcohols showed a marked change over time (s03, s09, s02 and s01), and six others displayed more constant emissions level (s38, s41, s32, s35, s45 and s17) (Fig 4). Two esters fluctuated strongly over time (s52, s10 and s47) and two others remained constant (s37 and s24) (Fig 4).
To study the compounds or chemical families that characterised each brood stage as well as those that covaried together, the 21 days of brood development have been divided into eight periods, as follow: Eggs: days 1–3; 1st to 4th instar larvae (L1-L4): days 4–7; 5th instar larvae (L5): day 8; Capping: day 9; Sealed larvae (SL): day 10, Prepupae (PP): day 11, Non-sclerotised pupae (NSP): days 12–16, Sclerotised pupae (SP): days 17–21. Different compounds were characteristic of the eight brood developmental stages (Fig 5). None of the compounds had its maximum at the egg stage (days 1–3), but six compounds were found in greater quantities in brood containing eggs than in the control (s15, s18, s30, s37, s52 and s71) (Fig 5 and S1 Fig). None of these six compounds were specific to the eggs; their emission increased upon hatching. In the first larval instars (L1-L4, days 4–7), four compounds reached their maximum emission (s10, s47, s52 and s12), and two other compounds neared it (s15 and s37) (Fig 5). In the last larval stage (L5, day 8), the blend began to change drastically. Compounds with their greatest quantity in the L1-L4 stage decreased (s10, s47, s52, s12 and s37), while the quantity of other compounds began to increase (s09, s35, s74, s75, s71 and s54) (Fig 5). During the capping period (Capping, day 9), 11 compounds had their maximum emission peak (s15, s74, s75, s71, s36, s01, s05, s13, s03, s02 and s41) (Fig 5). All of these compounds decreased during the sealed larval stages (SL, day 10) (Fig 5). In this stage, seven compounds had their maximum emission peak (S65, s32, s38, s54, s28, s30 and s42) (Fig 5). In the prepupal stage (PP, day 11), only three compounds reached their maximum emission peaks (s27, s29 and s24) and two compounds were slightly under maximum emission levels (s15 and s37) (Fig 5). In the non-sclerotised pupal stage (NSP, days 12–16), two compounds had their maximum emission peaks (s37 and s18) and two others were slightly under maximum emission levels (s15 and s42) (Fig 5). In the last stage of brood development (SP, days 17–21), only one compound was at its maximum (s73) (Fig 5).
The evolution of the quantity of each compound is represented in rows, with a colour gradient. The absence of compounds is represented in grey (here, both compounds are present only on a single day of the SP stage, resulting in a median of 0 for all stages). Each column represents a key stage of brood development (Eggs: day 1 to day 3; L1-L4: 1st to 4th instar larvae, day 4 to day 7; L5: 5th instar larvae, day 8; Capping: adult capping behaviour, day 9; LS: larvae spinning cocoon, day 10; PP: pre-pupae, day 11; NSP: non-sclerotised pupae, day 12 to day 16; SP: sclerotised pupae, day 17 to day 21). The width of each box is proportional to the duration of each stage. Compounds are labelled with their IDs, chemical names and chemical families.
The total amount of (E)-β-ocimene (s30) was higher during capped brood stages (SL, PP, NSP and SP) than in larvae, especially just after capping. When the amount of (E)-β-ocimene (s30) was reported per mg of individual weight, its quantity was found to be higher in larval brood stages than in capped brood stages and reached its maximum in first instar larvae (day 4) (S3 Fig).
Depending on their chemical families, the compounds highlighted in this study did not have their maximum emission peaks at the same stages of brood development. Ketones had their maximum emission peak at L1-L4 for s12, and Capping stages for s36 (clusters 1 and 3, Fig 6). Except for s37, which had its maximum peak at the NSP stage, and s24, which had its maximum peak at PP stage, the esters had their maximum emission peaks at the L1-L4 stage (Fig 6). The alcohols peaked at the late larval stages (L5, Capping and SL) (clusters 2, 4 and 6, Fig 6). Alkanes had their peaks at the capping and just after (Capping and SL) (clusters 3 and 5, Fig 6). The aldehyde had its peak at the SL stage (cluster 6, Fig 6). Finally, except for s15 which had a double peak at L1-L4 and Capping (cluster 11), all of the terpenes had their maximum emission peaks after capping of the brood cells (clusters 6, 7, 8, 9, 10 and 11, Fig 6).
Clusters group compounds with similarly timed emission peaks. Compounds are labelled with their IDs and coloured according to their chemical families. Eggs: day 1 to day 3; L1-L4: 1st to 4th instar larvae, day 4 to day 7; L5: 5th instar larvae, day 8; Capping: adult capping behaviour, day 9, LS: larvae spinning cocoon, day 10, PP: pre-pupae, day 11, NSP: non-sclerotised pupae, day 12 to day 16; SP: sclerotised pupae, day 17 to day 21.
Discussion
In this study, we analysed the VOC emitted by honey bee brood (A. mellifera) throughout its development. We identified 32 VOC characterising the brood, belonging to six different chemical families. The compounds did not have the same temporal dynamic during brood development and most of them appear to be stage-dependant. Chemical families were also specific to developmental stages.
Headspace capture is a common method for studying volatile emissions from bee brood [8, 13, 16, 27, 39, 40]. The focus of our study is to describe the landscape of chemical emissions from bee brood so that it can serve as a base for future, more functional studies. Some compounds were not fully identified (e.g. s24–2-Hexenyl acetate for which its stereochemistry is not found or s73 –Nonadecene where the location of the double bond is still unknown) and would require further identification.
To date, this study presents for the first time the emission of VOC from honey bee brood during its whole development, from egg laying to emergence. Some previous studies showed the evolution of specific compounds during the larval or early pupal stages [25, 27] or the presence of compounds in larvae or pupae at specific ages [26, 41]. A key point of our method was the monitoring of the same brood cells throughout development, with a non-invasive passive capture. No food resources (pollen, nectar, honey) were present on the brood frames, except for the larval jelly necessary for larval nutrition, which suggest that the compounds retained in this study are representative of the brood. Of the 32 compounds described in this study, 15 compounds are newly described in bees, with one compound never described either in bees or in the hive environment but has been described in floral emissions [42], and 14 of them already described as being present in the hive environment (wax, propolis, pollen, honey, ambient air) [43–48]. The other 17 compounds had already been described in bees [8, 19, 26, 40, 41, 49–55], with 13 already described has emitted by bee brood [8, 19, 26, 40, 41, 51, 52, 55].
Ocimene brood emission
(E)-β-ocimene (s30) was the compound that was captured in the greatest quantity throughout this experiment, four times higher than the other compounds. Thus, the total peak area of the full set of compounds largely reflected the dynamics of (E)-β-ocimene (s30) during brood development. (E)-β-ocimene (s30) was found every day during brood development, from the first day of the egg stage to the emergence of the adult bee. This compound has already been described in the scientific literature in the uncapped brood to early capped brood stages [25, 27]. Its presence in the emission of VOC from eggs is described here for the first time. In our study, (E)-β-ocimene (s30) is more abundant in the capped stages than in the uncapped larval stages, with a peak in the SL stage. However, when the amount of compound is calculated according to the weight of the individuals, the peak of (E)-β-ocimene (s30) appears to be on day 4, at the first instar larvae. As a consequence, the amount of (E)-β-ocimene (s30) emitted by an individual is higher in the uncapped larval stages, especially in early larval stages, than in the capped brood stages. These results are in accordance with the results obtained by Maisonnasse et al. [27], thus providing additional validation to our experimental set-up. The presence of this compound in stages other than uncapped and early-capped brood raises questions about its possible role in other stages of brood development and its involvement in chemical communication between the brood and the adults of the colony.
In addition to (E)-β-ocimene (s30), we identified three ocimene isomers: (Z)-β-ocimene (s27), α-ocimene (s29) and allo-ocimene (s42). Allo-ocimene (s42) was previously described as a putative pheromone emitted by young larvae, in a similar context to (E)-β-ocimene [55]. Here, we found the presence of allo-ocimene in more than five of our samples mainly in the 3rd and 4th instar larvae as well as from the SL stage to the end of the NSP stage. Similarly to (E)-β-ocimene, allo-ocimene triggers a rapid nervous response of the antenna [55]. In addition, the ocimene is processed by the antennal lobe [56], the first brain region processing olfactory information sent by the antennal neurons [57]. Neither of the two other ocimenes captured in this study have been described as part of brood pheromones. Nevertheless, the response of adults to (E)-β-ocimene and allo-ocimene, and the high similarity between these four ocimenes, suggest that nurses may detect (Z)-β-ocimene and α-ocimene. Like (E)-β-ocimene with the BEP, the emission of these three other ocimenes could have an additive or synergistic effect and trigger different behaviours in adults [10, 27]. Further studies on behaviour and brain processing of olfactory information of nurse bees are necessary to highlight the potential discrimination between these four ocimenes, and the potential role of the (Z)-β-ocimene, the α-ocimene and the allo-ocimene.
Compounds known to transmit pheromonal information between adult bees
In our study, we found four compounds (1-hexanol (s09), isoamyl acetate (s10), 1-octanol (s32) and 2-heptanone (s12)) already described as part of pheromone alarm mixtures emitted and received by adult bees [50, 53, 54, 58]. The alarm pheromone is composed of a cocktail of nearly twenty chemical compounds [3, 54, 59].
The compound 2-heptanone is known to be synthetized by the mandibular glands [53, 58]. During the larval stage, nurses feed the brood with royal jelly for the first stages and then with a mixture of pollen, honey and glandular extract from the mandibular and hypopharyngeal glands [60, 61]. The compound 2-heptanone has been described as a royal jelly VOC [45] and from VOC of uncapped brood frames in the presence of adult workers [25]. In our study, 2-heptanone (s12) was present in all uncapped larval stages, with a peak in 4th instar larvae (L1-L4, day 7). As the in situ capture of brood VOC did not allow us to separate brood emissions from potential emissions by larval food, our results suggest that the presence of 2-heptanone in our experiment during the early larval stages may be explained by the presence of the larval food in the brood cells. This would indicate that this compound is released from the larval food, even if we cannot rule out a possible emission by larvae.
The compound 1-hexanol (s09), isoamyl acetate (s10) and 1-octanol (s32) are synthetized in the Koschevnikov gland, close to the stinger [50, 54], and are known to being part of the alarm pheromone of adult workers [3, 54, 59]. The main functions of the alarm pheromone for colony defence are increased flight activity, recruitment of individuals and behavioural responses to moving targets. Not all active chemical compounds trigger these three functions. The common function of isoamyl acetate, 1-hexanol and 1-octanol is their ability to recruit other adults [54]. In our experiment, 1-hexanol (s09) had its emission peak at the 5th larval stage (L5, day 8) and remained abundant at the time of operculation (Capping, day 9) and at the last pupal stages (SP, days 17–18). The compound 1-octanol (s32) was captured mainly on days 8, 9 and 10 (L5, Capping, PP), with a peak at the time of operculation. Due to the synchronization between the emission of these compounds and the capping behaviour observed during the experiment, we hypothesise that 1-hexanol and 1-octanol emission by the brood may play a role in eliciting capping behaviour by workers. This hypothesis needs to be validated by behavioural assays, such as those already used to confirm the role of brood pheromone compounds as a capping signal [7].
The amount of isoamyl acetate in the Koschevnikov gland differs according to the age of the worker. Nurse bees displayed more isoamyl acetate in their gland than guards or foragers, suggesting another function in a context other than the defence of the colony [62]. This compound had already been reported as part of VOC of uncapped brood frames in the presence of adult workers [25]. In our study, we captured isoamyl acetate (s10) mainly during the larval stages (L1-L4). This suggests that broods may emit isoamyl acetate as a chemical signal for care to the nurse bees. This signal can have various meanings such as a signal that the brood is healthy, a call for feeding, for temperature or humidity control. Behavioural assays are needed to validate this hypothesis.
Compounds emitted by the brood but not described as pheromones
In this study, we found 32 compounds characterising the brood belonging to six chemical families, including ten alcohols (prenyl alcohol (s01), 2,3-butanediol (s02), 2,3-butanediol (s03), 1-hexanol (s09), 1-heptanol (s17), 1-octanol (s32), (E)-linalool furanoxide (s35), linalool (s38), phenylethyl alcohol (s41), (E)-linalool pyranoxide (s45)), one aldehyde (phenylacetaldehyde (s28)), seven alkanes (octane (s05), nonane (s13), tridecane (s54), pentadecane (s65), heptadecane (s71), nonadecane (s74), eicosane (s75)), five esters (isoamyl acetate (s10), hexenyl acetate (s24), methyl benzoate (s37), methyl salicylate (s47), γ-octalactone (s52)), two ketones (2-heptanone (s12), 2-nonanone (s36)) and seven terpenes (α-pinene (s15), β-pinene (s18), (Z)-β-ocimene (s27), α-ocimene (s29), (E)-β-ocimene (s30), allo-ocimene (s42), nonadecene (s73)). As previously mentioned, only five compounds have been described as pheromonal compounds in bees. Within the remaining 27 compounds, 13 have already been described in brood, either as being emitted by the brood (octane (s05), nonane (s13), α-pinene (s15), (E)-linalool furanoxide (s35), 2-nonanone (s36), linalool (s38), methyl benzoate (s37), allo-ocimene (s42), methyl salicylate (s47), γ-octalactone (s52), tridecane (s54), heptadecane (s71), nonadecane (s74), eicosane (s75)) [19, 25, 26, 40, 51, 52, 55], or by dead brood extracts (octane (s05), nonane (s13), eicosane (s75) tridecane (s54)) [41]. Our results are in accordance with these studies. The 14 other compounds found to characterize the brood (prenyl alcohol (s01), 2,3-butanediol (s02), 2,3-butanediol (s03), 1-heptanol (s17), β-pinene (s18), hexenyl acetate (s24), (Z)-β-ocimene (s27), phenylacetaldehyde (s28), α-ocimene (s29), phenylethyl alcohol (s41), allo-ocimene (s42), (E)-linalool pyranoxide (s45), pentadecane (s65), nonadecene (s73)) have already been described as being present in the VOC of hive products (honey, propolis, larval jelly, pollen) or floral extracts [44–47, 55, 63, 64], but we described them in honey bee brood for the first time. Nevertheless, α-pinene (s15) and β-pinene (s18) have been described as part of the odours of termite guardians [65] or as being present in Dufour’s gland of ants [66]. The presence of VOC previously measured in hive products in brood emissions is not surprising as hive products are involved in nutrition (honey, larval jelly and pollen) and protection (honey, propolis, larval jelly) against various pathogenic agents of all the individuals in the colony [33, 67–71]. These chemical compounds could be brought to the uncapped brood cells by nurses when they visit brood cells during caring behaviours. Then, compounds can be trapped in and emitted from the wax of brood cells, or released at specific times by the brood as a means of communication with adults.
Egg compounds
We found that eggs emitted a blend of five VOC: α-pinene (s15), β-pinene (s18), (E)-β-ocimene (s30), methyl benzoate (s37) and γ-octalactone (s52). Even if all of these compounds are maximally abundant at stages of brood development other than the egg stage, this study is the first to describe these 5 compounds as characteristic of eggs. The compound β-ocimene (s30) and methyl benzoate (s37) have already been found on VOC from frames containing eggs and one-day-old larvae [25]. Here, we confirm the presence of (E)-β-ocimene and methyl benzoate in egg brood cells and suggest that (E)-β-ocimene and methyl benzoate take part in the egg odour blend. Heptadecane (s71) had already been found in VOC from frames containing fourth and fifth instar larvae [26]. We found heptadecane on the first day of eggs. Here, we suggest that heptadecane takes part in the newly laid egg odour blend. α-pinene (s15) and β-pinene (s18) can be found in the VOC hive atmosphere [48]. α-pinene has also been found in VOC of larvae and pupae [26], and in VOC from brood frame with adult workers [25]. β-pinene has been found in VOC of honey [44]. In our study, frames were empty of bees and captures were done outside the hive. These two compounds were higher in brood samples than in control wax samples, thus suggesting that they could be part of the blend of VOC emitted by eggs or by cells polished by the workers for the queen to lay eggs.
Beginning our captures on the first day of the egg stage allowed us to discover that honey bee eggs and healthy larvae emit γ-octalactone (s52). γ-octalactone has been described in Apis dorsata as an alarm pheromone [72], and was recently described for the first time in A. mellifera larvae [19]. The authors found that the amount of γ-octalactone increased with the infection by European foulbrood (M. plutonius). This compound has also been described as stimulating oviposition in flies [73, 74], and as being present in the metapleural glands of ants [66], glands that allow for the marking of the nest [75]. Both of these observations corroborate our observation of γ-octalactone in eggs, and suggest that it may also be involved, along the other four compounds, in egg marking by the queen or polished cell marking by the workers. A comparison between queen and worker eggs would be interesting to solve this question.
Description of brood VOC emissions at key moments of development
Chemical compounds described in this study had their maximum emission at different times of brood development. Nevertheless, most of these compounds had their peaks during the capping and capped larval stage of the brood (day 9 to 10: Capping, SL).
Larvae have a constant need to communicate with workers in order to be fed. The modulation of the pheromone signal of larvae during their development allows an adaptation of the feeding behaviour of the worker bees [7, 9, 27]. Here, four compounds had their maximum emission at the L1-L4 stage (isoamyl acetate (s10), 2-heptanone (s12), γ-octalactone (s52) and methyl salicylate (s47)) and two at the L5 stage (1-hexanol (s09) and (E)-linalool furanoxide (s35)). The modulation of VOC emitted between the L1-L4 stage and the L5 stage could allow for modulation of worker feeding behaviour, but perhaps more importantly signal the need for capping preparation.
The capping of brood cells is a key moment in brood development. Workers close the brood cell around the 9th day after egg laying. The brood is no longer fed by workers, but the larvae feed on the jelly left on the bottom of the cell and then start to weave their cocoon and prepare for metamorphosis [6]. It has already been shown that the BEP, and particularly methyl esters present in this pheromone, trigger this capping behaviour [7]. Here, we found 11 compounds at their maximum at the time of brood capping (day 9) (prenyl alcohol (s1), 2.3-butanediol (s02 and s03), octane (s05), nonane (s13), α-pinene (s15), 2-nonanone (s36), phenylethyl alcohol (s41), heptadecane (s71), nonadecane (s74), eicosane (s75)). A complex mixture of VOC and BEP could trigger the capping of brood cells by workers. More research is necessary to determine if, and which of, these compounds alone trigger capping behaviour. The complexity of the mixture captured at this key day in brood development opens up new perspectives in understanding chemical communication between brood and adults.
After capping, the larva stretches out and starts weaving its cocoon (day 11, SL) [6]. At this time, we found seven compounds at their maximum emission (phenylacetaldehyde (s28), (E)-β-ocimene (s30), 1-octanol (s32), linalool (s38), allo-ocimene (s42), tridecane (s54) and pentadecane (s65)). Then the larva starts its metamorphosis (day 12, PP) [6]; we found three compounds at their maximum release on this day (hexenyl acetate (s24), (Z)-β-ocimene (s27), α-ocimene (s29)). Communicating at the beginning of metamorphosis could allow the workers to monitor the health of the brood. A modification of those compounds emitted by the brood just after capping, for example because of Varroa or Nosema parasitism, may trigger the cleaning behaviour by the adults [16, 19, 76].
The pupal stages were characterised by the peaking of two terpene compounds, one during NSP, and one during SP. β-pinene (s18) had its maximum emission in the early pupal stages (NSP), mainly on days 14 to 16. Nonadecene (s73) is the only compound that had its peak of emission during the last stage of brood development (SP), especially during the last 3 days before adult emergence. At the end of metamorphosis, the brood is not fed and not subject to hygienic behaviour [6]. Here, we described for the first time an emission of VOC from the brood before the emergence of the adult bee. When it could just be part of the chemical process of development and given off as a by-product, it also suggests a potential emergence signal, with a possible role in the acceptance of newly emerged workers by adults.
Interestingly, the emission of chemical families correlates with the age of the brood. Uncapped broods emitted mainly ketones, esters and alcohols, while capped broods emitted alkanes and terpenes. The emission of volatile esters (isoamyl acetate (s10), methyl benzoate (s37), methyl salicylate (s47), γ-octalactone (s52)) in larval stages is consistent with the emission of BEP from the brood at these stages [7]. These compounds could be complementary to the BEP. In addition, the brood emitted several highly volatile alkanes. Adults recognise their kin with cuticular hydrocarbon mixtures of low volatile alkanes and alkenes [77, 78]. It has been shown that long carbon chain alkanes are differentially recognised by adult honey bees [79], and it would be interesting to see if short carbon chain alkanes are also involved in nestmate recognition.
This study is the first study of VOC emissions of brood covering the entire honey bee developmental cycle, from egg laying to adult emergence. Here, we show the complexity of the volatile chemical messages emitted by the brood at different developmental stages. When Martin et al. [80] suggested low volatile compound as egg marking signal, we describe the olfactory signature made of volatile compounds of A. mellifera eggs that could also have a potential implication in the marking of queen eggs. We also show a potential emergence signal. While the compounds identified could have a potential involvement in the chemical recognition of brood and its needs by the workers, some of the compounds identified could also be the result of metabolic by-products. Our suggestions need to be assessed using appropriate biological tests. This study opens up new perspectives on the intricate relationship between brood and adult bees, and the behaviour of nurse bees towards the brood. Brood emission of molecules already described as part of chemical communication in the adult reinforces the idea that chemical communication in the bee is highly context-dependent and based on the quantity of each molecule emitted.
Supporting information
S1 Fig.
Redundancy analysis (RDA) performed on chemical compounds captured on brood and control samples (A) for “Eggs” group (day 1 to day 3), (B) “Sclerotised Pupae” group (day 17 to day 21). For each day: brood sample n = 9, control sample n = 9. Points represent samples (blue: brood, red: wax). Compounds are represented by their IDs and grey arrows. The compounds characterising the brood that are retained for the rest of the analysis are shown in bold.
https://doi.org/10.1371/journal.pone.0282120.s001
(TIF)
S2 Fig. Representative gas chromatography profiles of each brood stage.
The compounds listed are those identified as brood specific. The other peaks represent fibre peaks or peaks specific to the wax control. (a) Eggs, (b) L1-L4: 1st to 4th instar larvae, (c) L5: 5th instar larvae, (d) Capping: adult capping behaviour, (e) LS: larvae spinning cocoon, (f) PP: pre-pupae, (g) NSP: non-sclerotised pupae, (h) SP: sclerotised pupae. °: Unsaturation remains unknown, *: Stereochemistry remains unknown.
https://doi.org/10.1371/journal.pone.0282120.s002
(TIF)
S3 Fig. Quantification of (E)-β-ocimene emission for an individual from the 1st instar larval to the 5th pupal stage.
Weight data are from previous experiments. Line represents mean values and bars represent standard deviation. For each day, n = 9.
https://doi.org/10.1371/journal.pone.0282120.s003
(TIF)
S1 Table. Peak areas of the 78 compounds found for each group (brood or wax), at each day and for each colony.
https://doi.org/10.1371/journal.pone.0282120.s004
(CSV)
Acknowledgments
We thank D. Crauser and J. Senechal for giving access to the honey bee colonies, and E. Joulin for help in passing the standards on the GC-MS.
References
- 1.
Bell WJ, Cardé RT, éditeurs. Chemical Ecology of Insects. Boston, MA: Springer US; 1984.
- 2. Karlson P, Butenandt A. Pheromones (ectohormones) in insects. Annu Rev Entomol. 1959;4(1):39–58.
- 3. Slessor KN, Winston ML, Le Conte Y. Pheromone communication in the honeybee (Apis mellifera L.). J Chem Ecol. 2005;31(11):2731–45.
- 4.
Bagnères AG, Hanus R. Communication and Social Regulation in Termites. In: Aquiloni L, Tricarico E, éditeurs. Social Recognition in Invertebrates. Cham: Springer International Publishing; 2015. p. 193–248.
- 5. Hölldobler B. Multimodal signals in ant communication. Journal of Comparative Physiology A: Sensory, Neural, and Behavioral Physiology. 1999;184(2):129–41.
- 6. Winston ML. The biology of the honey bee. Cambridge, Massachussets: Harvard University Press; 1987.
- 7. Le Conte Y, Arnold G, Trouiller J, Masson C, Chappe B. Identification of a brood pheromone in honeybees. Naturwissenschaften. 1990;77(7):334–6.
- 8. Maisonnasse A, Lenoir JC, Costagliola G, Beslay D, Choteau F, Crauser D, et al. A scientific note on E-β-ocimene, a new volatile primer pheromone that inhibits worker ovary development in honey bees. Apidologie. 2009;40(5):562–4.
- 9. Sagili RR, Pankiw T. Effects of brood pheromone modulated brood rearing behaviors on honey bee (Apis mellifera L.) colony growth. J Insect Behav. 2009;22(5):339–49.
- 10. Traynor KS, Le Conte Y, Page RE. Age matters: pheromone profiles of larvae differentially influence foraging behaviour in the honeybee, Apis mellifera. Anim Behav. 2015;99:1–8.
- 11. He XJ, Zhang XC, Jiang WJ, Barron AB, Zhang JH, Zeng ZJ. Starving honey bee (Apis mellifera) larvae signal pheromonally to worker bees. Sci Rep. avr 2016;6(1):22359.
- 12. Mohammedi A, Paris A, Crauser D, Le Conte Y. Effect of aliphatic esters on ovary development of queenless bees (Apis mellifera L.). Naturwissenschaften. 1998;85(9):455–8.
- 13. McAfee A, Chapman A, Iovinella I, Gallagher-Kurtzke Y, Collins TF, Higo H, et al. A death pheromone, oleic acid, triggers hygienic behavior in honey bees (Apis mellifera L.). Sci Rep. 2018;8(1):5719.
- 14. Swanson JAI, Torto B, Kells SA, Mesce KA, Tumlinson JH, Spivak M. Odorants that induce hygienic behavior in honeybees: Identification of volatile compounds in chalkbrood-infected honeybee larvae. J Chem Ecol. 2009;35(9):1108–16. pmid:19816752
- 15. Mondet F, Blanchard S, Barthes N, Beslay D, Bordier C, Costagliola G, et al. Chemical detection triggers honey bee defense against a destructive parasitic threat. Nat Chem Biol. 2021; pmid:33495646
- 16. Nazzi F, Della Vedova G, D’Agaro M. A semiochemical from brood cells infested by Varroa destructor triggers hygienic behaviour in Apis mellifera. Apidologie. 2004;35(1):65–70.
- 17. Wagoner K, Millar JG, Schal C, Rueppell O. Cuticular pheromones stimulate hygienic behavior in the honey bee (Apis mellifera). Sci Rep. 2020;10:7132.
- 18. Liendo MC, Muntaabski I, Russo RM, Lanzavecchia SB, Segura DF, Palacio MA, et al. Temporal changes in volatile profiles of Varroa destructor ‐infested brood may trigger hygienic behavior in Apis mellifera. Entomol Exp Appl. 2021;169(6):563–74.
- 19. Kathe E, Seidelmann K, Lewkowski O, Le Conte Y, Erler S. Changes in chemical cues of Melissococcus plutonius infected honey bee larvae. Chemoecology. 2021;31(3):189–200.
- 20. Wagoner K, Spivak M, Hefetz A, Reams T, Rueppell O. Stock-specific chemical brood signals are induced by varroa and deformed wing virus, and elicit hygienic response in the honey bee. Sci Rep. 2019;9(1):8753. pmid:31217481
- 21. Le Conte Y, Hefetz A. Primer pheromones in social hymenoptera. Annu Rev Entomol. 2008;53(1):523–42. pmid:17877458
- 22. Schultner E, Oettler J, Helanterä H. The role of brood in eusocial hymenoptera. Q Rev Biol. 2017;92(1):39–78. pmid:29558609
- 23. Schultner E, Pulliainen U. Brood recognition and discrimination in ants. Insect Soc. 2020;67(1):11–34.
- 24. Starkey J, Derstine N, Amsalem E. Do bumble bees produce brood pheromones? J Chem Ecol. sept 2019;45(9):725–34. pmid:31471873
- 25. Carroll MJ, Duehl AJ. Collection of volatiles from honeybee larvae and adults enclosed on brood frames. Apidologie. 2012;43(6):715–30.
- 26. Light M, Shutler D, Cutler GC, Hillier NK. Varroa destructor mite electrophysiological responses to honey bee (Apis mellifera) colony volatiles. Exp Appl Acarol. 2020;
- 27. Maisonnasse A, Lenoir JC, Beslay D, Crauser D, Le Conte Y. E-β-Ocimene, a volatile brood pheromone involved in social regulation in the honey bee colony (Apis mellifera). PLoS ONE. 2010;5(10):e13531.
- 28. Haberer W, Steiger S, Müller JK. Dynamic changes in volatile emissions of breeding burying beetles: Volatile emissions of burying beetles. Physiol Entomol. 2014;39(2):153–64.
- 29. Frederickx C, Dekeirsschieter J, Brostaux Y, Wathelet JP, Verheggen FJ, Haubruge E. Volatile organic compounds released by blowfly larvae and pupae: New perspectives in forensic entomology. Forensic Science International. 2012;219(1‑3):215–20. pmid:22341879
- 30. Sharma A, Tomberlin JK, Delclos P, Bala M. Volatile compounds reveal age: a study of volatile organic compounds released by Chrysomya rufifacies immatures. Int J Legal Med. 2021;135(3):967–77.
- 31. do Nascimento AM, Cuvillier-Hot V, Barchuk AR, Simões ZLP, Hartfelder K. Honey bee (Apis mellifera) transferrin-gene structure and the role of ecdysteroids in the developmental regulation of its expression. Insect Biochem Mol Biol. 2004;34(5):415–24.
- 32. Woltedji D, Fang Y, Han B, Feng M, Li R, Lu X, et al. Proteome analysis of hemolymph changes during the larval to pupal development stages of honeybee workers (Apis mellifera ligustica). J Proteome Res. 2013;12(11):5189–98.
- 33. Haydak MH. Honey bee nutrition. Annu Rev Entomol. 1970;15.
- 34. Le Conte Y, Sreng L, Poitout SH. Brood pheromone can modulate the feeding behavior of Apis mellifera workers (Hymenoptera: Apidae). Journal of Economic Entomology. 1995;88(4):7.
- 35. Le Conte Y, Mohammedi A, Robinson GE. Primer effects of a brood pheromone on honeybee behavioural development. Proc R Soc Lond B. 2001;268(1463):163–8. pmid:11209886
- 36. Wood SN. Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models: Estimation of Semiparametric Generalized Linear Models. Journal of the Royal Statistical Society: Series B (Statistical Methodology). janv 2011;73(1):3–36.
- 37. Jay SC. The development of honeybees in their cells. Journal of Apicultural Research. 1963;2(2):117–34.
- 38. Rau A, Maugis-Rabusseau C. Transformation and model choice for RNA-seq co-expression analysis. Brief Bioinform. 2017;bbw128.
- 39. Nazzi F, Milani N, Della Vedova G. A semiochemical from larval food influences the entrance of Varroa destructor into brood cells. Apidologie. 2004;35(4):403–10.
- 40. Liu J, Zhang R, Tang R, Zhang Y, Guo R, Xu G, et al. The role of honey bee derived aliphatic esters in the host-finding behavior of Varroa destructor. Insects. 2022;14(1):24. pmid:36661952
- 41. Haber M, Mishyna M, Itzhak Martinez JJ, Benjamin O. Edible larvae and pupae of honey bee (Apis mellifera): Odor and nutritional characterization as a function of diet. Food Chemistry. 2019;292:197–203. pmid:31054665
- 42. Bergström LGW, Bergquist S, Stenhagen G, Gahmberg CG, Maia ACD, Nordenstam B. Floral scent chemistry within the genus Linnaea (Caprifoliaceae). Nordic Journal of Botany. 2018;36(3):njb-01732.
- 43. Cellini A, Giacomuzzi V, Donati I, Farneti B, Rodriguez-Estrada MT, Savioli S, et al. Pathogen-induced changes in floral scent may increase honeybee-mediated dispersal of Erwinia amylovora. ISME J. 2019;13(4):847–59. pmid:30504898
- 44. Duru ME, Taş M, Çayan F, Küçükaydın S, Tel-Çayan G. Characterization of volatile compounds of Turkish pine honeys from different regions and classification with chemometric studies. Eur Food Res Technol. 2021;247(10):2533–44.
- 45. Isidorov VA, Bakier S, Grzech I. Gas chromatographic–mass spectrometric investigation of volatile and extractable compounds of crude royal jelly. Journal of Chromatography B. 2012;885‑886:109–16. pmid:22245368
- 46. Jánošková N, Vyviurska O, Špánik I. Identification of volatile organic compounds in honeydew honeys using comprehensive gas chromatography. J Food Nutr Res. 2014;10.
- 47. Najar B, Ferri B, Cioni PL, Pistelli L. Volatile emission and essential oil composition of Sambucus nigra L. organs during different developmental stages. Plant Biosystems. 2021;155(4):721–9.
- 48.
Smith G, Alnasser G, Jones D, Bromenshenk J. Volatile and semi-volatile organic compounds in beehive atmospheres. In: Devillers J, Pham-Delegue M, éditeurs. Honey bees: Estimating the environmental impact of chemicals. CRC Press; 2002. p. 27–56.
- 49. Butler CG, Callow RK, Johnston NC. Extraction and purification of ‘queen substance’ from queen bees. Nature. 1959;184(4702):1871–1871.
- 50. Collins AM, Blum MS. Alarm responses caused by newly identified compounds derived from the honeybee sting. J Chem Ecol. 1983;9(1):57–65. pmid:24408619
- 51. Donzé G, Schnyder-Candrian S, Bogdanov S, Diehl PA, Guerin PM, Kilchenman V, et al. Aliphatic alcohols and aldehydes of the honey bee cocoon induce arrestment behavior in Varroa jacobsoni (Acari: Mesostigmata), an ectoparasite of Apis mellifera. Arch Insect Biochem. 1998;37:129–45.
- 52. Lee S, Lim S, Choi YS, Lee M lyeol, Kwon HW. Volatile disease markers of American foulbrood-infected larvae in Apis mellifera. Journal of Insect Physiology. 2020;122:104040.
- 53. Shearer DA, Boch R. 2-heptanone in the mandibular gland secretion of the honey-bee. Nature. 1965;206(4983):530–530.
- 54. Wager BR, Breed MD. Does honey bee sting alarm pheromone give orientation information to defensive bees? Ann Entomol Soc Am. 2000;93(6):1329–32.
- 55. Wu F, Ma C, Han B, Meng L, Hu H, Fang Y, et al. Behavioural, physiological and molecular changes in alloparental caregivers may be responsible for selection response for female reproductive investment in honey bees. Mol Ecol. 2019;28(18):4212–27. pmid:31444931
- 56. Locatelli FF, Fernandez PC, Smith BH. Learning about natural variation of odor mixtures enhances categorization in early olfactory processing. Journal of Experimental Biology. 2016;jeb.141465. pmid:27412003
- 57.
Sandoz JC. Olfaction in honey bees: From molecules to behavior. In: Galizia CG, Eisenhardt D, Giurfa M, éditeurs. Honeybee neurobiology and behavior: a tribute to Randolf Menzel. Dordrecht; New York: Springer; 2012. p. 235–52.
- 58. Butler CG. Mandibular gland pheromone of worker honeybees. Nature. 1966;212(5061):530–530.
- 59. Boch R, Shearer DA, Stone BC. Identification of iso-amyl acetate as an active component in the sting pheromone of the honey bee. Nature. 1962;195(4845):1018–20.
- 60.
Hartfelder K, Engels W. Social insect polymorphism: hormonal regulation of plasticity in development and reproduction in the honeybee. In: Current Topics in Developmental Biology. Elsevier; 1998. p. 45–77.
- 61. Peters L, Zhu-Salzman K, Pankiw T. Effect of primer pheromones and pollen diet on the food producing glands of worker honey bees (Apis mellifera L.). Journal of Insect Physiology. 2010;56(2):132–7.
- 62. Camargos AF, Cossolin JFS, Martínez LC, Gonçalves WG, dos Santos MH, Zanuncio JC, et al. Morphology and chemical composition of the Koschewnikow gland of the honey bee Apis mellifera (Hymenoptera: Apidae) workers engaged in different tasks. Journal of Apicultural Research. 2020;59(5):1037–48.
- 63. Lee FJ, Rusch DB, Stewart FJ, Mattila HR, Newton ILG. Saccharide breakdown and fermentation by the honey bee gut microbiome: Fermentation by honey bee gut microbes. Environ Microbiol. 2015;17(3):796–815.
- 64. Starowicz M, Hanus P, Lamparski G, Sawicki T. Characterizing the volatile and sensory profiles and sugar content of beeswax, beebread, bee pollen, and honey. Molecules. 2021;26(11):3410. pmid:34199969
- 65. Quintana A, Reinhard J, Faure R, Uva P. Interspecific variation in terpenoid composition of defensive secretions of european Reticulitermes termites. Journal of Chemical Ecology. 2003;29(3):14.
- 66. Morgan ED. Chemical sorcery for sociality: Exocrine secretions of ants (Hymenoptera: Formicidae). Myrmecol News. 2008;(11):79–90.
- 67. Borba RS, Spivak M. Propolis envelope in Apis mellifera colonies supports honey bees against the pathogen, Paenibacillus larvae. Scientific Reports. 2017;6.
- 68. Di Pasquale G, Salignon M, Le Conte Y, Belzunces LP, Decourtye A, Kretzschmar A, et al. Influence of pollen nutrition on honey bee health: Do pollen quality and diversity matter? Zeil J, éditeur. PLoS ONE. 2013;8(8):e72016. pmid:23940803
- 69. Klaudiny J, Bachanová K, Kohútová L, Dzúrová M, Kopernický J, Majtán J. Expression of larval jelly antimicrobial peptide defensin1 in Apis mellifera colonies. Biologia. 2012;67(1):200–11.
- 70. Kwakman PHS, Velde AA te, Boer L, Speijer D, Christina Vandenbroucke‐Grauls MJ, Zaat SAJ. How honey kills bacteria. FASEB j. 2010;24(7):2576–82.
- 71. Silici S, Kutluca S. Chemical composition and antibacterial activity of propolis collected by three different races of honeybees in the same region. Journal of Ethnopharmacology. 2005;99(1):69–73. pmid:15848022
- 72. Li J, Wang Z, Tan K, Qu Y, Nieh J. Effects of natural and synthetic alarm pheromone and individual pheromone components on foraging behavior of the giant Asian honey bee, Apis dorsata. Journal of Experimental Biology. 2014;jeb.110171.
- 73. Kempraj V, Jean Park S, Taylor PW. γ‐Octalactone, an effective oviposition stimulant of Bactrocera tryoni. J Appl Entomol. 2019;143(10):1205–9.
- 74. Pagadala Damodaram KJ, Kempraj V, Aurade RM, Venkataramanappa RK, Nandagopal B, Verghese A, et al. Oviposition site-selection by Bactrocera dorsalis is mediated through an innate recognition template tuned to γ-octalactone. Marion-Poll F, éditeur. PLoS ONE. 2014;9(1):e85764.
- 75. Cammaerts MC, Cammaerts R. Marking of nest entrances and vicinity in two related Tetramorium ant species (Hymenoptera: Formicidae). Journal of Insect Behavior. 2001;23.
- 76. Spivak M, Gilliam M. Hygienic behaviour of honey bees and its application for control of brood diseases and varroa: Part I. Hygienic behaviour and resistance to American foulbrood. Bee World. 1998;79(3):124–34.
- 77.
Blomquist GJ, Bagnères AG, éditeurs. Insect hydrocarbons: biology, biochemistry, and chemical ecology. Cambridge, UK; New York: Cambridge University Press; 2010. 492 p.
- 78. Dani FR, Jones GR, Corsi S, Beard R, Pradella D, Turillazzi S. Nestmate recognition cues in the honey bee: differential importance of cuticular alkanes and alkenes. Chemical Senses. 2005;30(6):477–89. pmid:15917370
- 79. Châline N, Sandoz JC, Martin SJ, Ratnieks FLW, Jones GR. Learning and discrimination of individual cuticular hydrocarbons by honeybees (Apis mellifera). Chemical Senses. 2005;30(4):327–35.
- 80. Martin SJ, Châline N, Oldroyd BP, Jones GR, Ratnieks FLW. Egg marking pheromones of anarchistic worker honeybees (Apis mellifera). Behavioral Ecology. 2004;15(5):839–44.