Modeling seasonal immune dynamics of honey bee (Apis mellifera L.) response to injection of heat-killed Serratia marcescens

Jana Hurychová; Jakub Dostál; Martin Kunc; Sara Šreibr; Silvie Dostálková; Marek Petřivalský; Pavel Hyršl; Dalibor Titěra; Jiří Danihlík; Pavel Dobeš

doi:10.1371/journal.pone.0311415

Abstract

The honey bee, Apis mellifera L., is one of the main pollinators worldwide. In a temperate climate, seasonality affects the life span, behavior, physiology, and immunity of honey bees. In consequence, it impacts their interaction with pathogens and parasites. In this study, we used Bayesian statistics and modeling to examine the immune response dynamics of summer and winter honey bee workers after injection with the heat-killed bacteria Serratia marcescens, an opportunistic honey bee pathogen. We investigated the humoral and cellular immune response at the transcriptional and functional levels using qPCR of selected immune genes, antimicrobial activity assay, and flow cytometric analysis of hemocyte concentration. Our data demonstrate increased antimicrobial activity at transcriptional and functional levels in summer and winter workers after injection, with a stronger immune response in winter bees. On the other hand, an increase in hemocyte concentration was observed only in the summer bee population. Our results indicate that the summer population mounts a cellular response when challenged with heat-killed S. marcescens, while winter honey bees predominantly rely on humoral immune reactions. We created a model describing the honey bee immune response dynamics to bacteria-derived components by applying Bayesian statistics to our data. This model can be employed in further research and facilitate the investigating of the honey bee immune system and its response to pathogens.

Citation: Hurychová J, Dostál J, Kunc M, Šreibr S, Dostálková S, Petřivalský M, et al. (2024) Modeling seasonal immune dynamics of honey bee (Apis mellifera L.) response to injection of heat-killed Serratia marcescens. PLoS ONE 19(10): e0311415. https://doi.org/10.1371/journal.pone.0311415

Editor: Vicente Martínez López, University of Murcia: Universidad de Murcia, SPAIN

Received: June 5, 2024; Accepted: September 18, 2024; Published: October 4, 2024

Copyright: © 2024 Hurychová 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 original data and the Python script are available as supplementary files submitted with the manuscript.

Funding: This work was supported by The Ministry of Agriculture of the Czech Republic (QK1910286). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors declare no conflict of interest.

Introduction

In a temperate climate, two distinct populations of honey bee workers occur in colonies within the year cycle, i.e., short-living ‘summer’ and long-living ‘diutinus’ bees. These two populations differ in their life span and in fundamental physiological and immune parameters [1–3]. The seasonal changes in honey bee immune defenses are triggered by the varying spectrum of pathogens present in the honey bee environment.

Individual honey bee immunity is composed of a) anatomical barriers of the body, b) cell-based immunity, which is mediated by hemocytes and their interactions with invaders, and c) humoral immunity depending mainly on phenoloxidase, lysozyme, and antimicrobial peptides (AMPs) activity [4]. Since honey bees are social insects living in colonies, social immunity such as grooming, hygienic behavior, or social fever also plays an important role in protecting the colony and the individuals against pathogens and parasites [5].

The most threatening pathogens emanate from various taxa, including parasitic mites, fungi, bacteria, and viruses, often causing co-infections [6]. In the last decades, spreading epidemics of Varroa destructor compromised the health of honey bee colonies. The mite interferes with the development of bees, negatively influences bee immunity, and serves as a vector for viral pathogens [7–9]. Furthermore, it has been suggested that bacteria can also be transmitted into the honey bee hemolymph through the feeding wound [10–12].

The bacterial pathogens of honey bees are spread worldwide and can cause septicemia in both larvae and adult bees [13, 14]. Most of the attention has been focused on Paenibacillus larvae and Melissococcus plutonius, the causative agents of the American and the European foulbrood, respectively. P. larvae and M. plutonius are highly contagious bacteria, causing decimating colony losses of managed bees. Other recognized opportunistic pathogens of honey bees, Spiroplasma apis, Spiroplasma melliferum, Hafnia alvei, Klebsiella spp., Enterobacter spp., and Serratia marcescens, are yet to be studied [15, 16]. Under normal conditions, these opportunistic bacteria coexist with their host in a commensal relationship; however, they can become pathogenic when the host’s immune system is compromised by factors such as pesticide exposure [17, 18], antibiotic treatment [19] or the presence of other pathogens that reduce immune defense [20].

In previous research, viable S. marcescens was detected in Varroa mites sampled from hives infected with this bacterium [13]. Therefore, Varroa mites were suggested as a possible vector for transferring S. marcescens into the bee hemolymph, causing sepsis and death of honey bee workers [13]. Moreover, some strains of this opportunistic pathogen, normally present in low abundance in the honey bee gut, may perturb the epithelial barrier, enter the hemocoel, and become highly virulent [20, 21]. This virulence is exacerbated by co-infection with Nosema ceranae, which disrupts gut epithelial cells [20].

Former studies discovered variable immune system reactions of larvae and adults at different life stages or castes after pathogen stimulation [9, 22–24]. Previously, differences between summer and winter (diutinus) bee populations in response to heat-killed bacteria were described [3]. The basal relative gene expression of antimicrobial peptides was higher in summer bees; however, the immune response of winter bees, including hemolymph antimicrobial activity, was more pronounced. These results were obtained by sampling bees at a single time point, providing a general view of the seasonal specificity of bee immune responses but without insights into dynamics within a prolonged period after immune activation stimuli.

In this study, we sampled bees at several time points to describe time-dependent changes in cellular and humoral immune parameters after injecting heat-killed bacteria S. marcescens. To effectively approximate the dynamics of expression or production of immune factors, a Bayesian approach was used. Bayesian statistics use available knowledge about a given parameter in a statistical model and update the model with provided observational data to determine the posterior distribution [25]. In this study, the model based on the scaled derivative of logistic functions enables the identification of important parameters in the dynamics of the measured variables, i.e. the strength (amplitude) of the analyzed immune factors and the time when the peak of the response occurs, which might be overlooked with original separated measurements. This approach provides a foundation for future investigations into the nuanced dynamics of honey bee immune responses to pathogens.

Material and methods

Experimental honey bees

The honey bee workers, Apis mellifera L., were collected from three colonies led by unrelated queens, kept at the apiary Kývalka, Czech Republic (49.1886747°N, 16.4513211°E). The colonies were maintained according to standard beekeeping practice by a professional beekeeper. Specifically, they were fed approximately 15 kg of sucrose solution (3:2 sucrose/water) for wintering, treated with flumethrin against Varroa destructor, and regularly inspected for varroosis or nosemosis. Clinical signs of varroosis and nosemosis (presence of Nosema spp. in the gut) were not observed in any of the experimental colonies. For the experiment, short-living summer bees were collected in July 2020, whereas long-living winter bees (i.e. diutinus) were collected at the end of September 2020. Honey bees taken in September from the same location and apiary were previously characterized as winter bees [2, 3]. The honey bees used in the study were age-synchronized to 10 days, according to Dostálková et al. [3]. Shortly, six frames (two from each hive) with capped brood were selected approximately one to three days before uncapping, placed in frame cages, and kept in the respective hives until the young bees emerged. The newly emerged bees were marked with color on their thoraces and returned to their colonies. After 10 days, marked bees were recollected (2,965 in summer and 1,870 in winter) and transported in a well-ventilated plywood box to the laboratory. Transport took approximately 20 minutes, and the bees were immediately processed in the laboratory as follows.

Honey bees treatment

The bees were pooled and randomly divided into four groups. Bees in the Control group were left without any treatment, whereas the CO₂ group was anesthetized with gaseous carbon dioxide. Bees in the PBS and Bacteria groups were also anesthetized with CO₂ and subsequently injected using a Hamilton syringe (Hamilton, Reno, NV, USA) in the dorsal part of the abdomen with 5 μl of sterile phosphate buffered saline (PBS; pH 7.0; Sigma-Aldrich, St. Louis, MO, USA) or 5 μl of heat-killed Serratia marcescens CCM 303 in PBS (100 bacterial cells/μl), respectively. The Serratia strain CCM 303, from the Czech Collection of Microorganisms, was cultivated on standard LB agar (Roth, Karlsruhe, Germany) at room temperature. The newly grown colonies were then resuspended in PBS to the required density and inactivated in a water bath at 80°C for 30 minutes. The bee abdomens were surface-sterilized with 5 μl of 96% ethanol prior to injection. Bees from all four groups were housed in plastic cages (100 bees/cage/0.5 dm³) prepared according to Kunc et al. [26] and kept at 34°C with sucrose solution (1:1 w/v; sucrose/water) provided ad libitum. Samples from summer bees were collected at 4, 8, 12, 16, 20, and 24 hours after the treatment. Due to the smaller number of emerging worker bees at the end of September, winter bees were sampled only at 4, 12, 20, and 24 hours post-treatment. Approximately one cage per treatment with 100 bees was sampled at each time point.

Samples collection

Hemolymph samples were collected according to Kunc et al. [2]. Briefly, the abdomen was cut off, and 2 μl of hemolymph was taken from a drop appearing from a gently pressed thorax. The hemolymph was processed according to the specific analysis. For hemocyte quantification, the hemolymph from five bees was pooled, and processed immediately. For Apidaecin 1 quantification, hemolymph from individual bees was diluted 10× with 0.1% trifluoroacetic acid. For ELISA quantification of peptides, pooled hemolymph from five bees was also diluted 10× with 0.1% trifluoroacetic acid. For antimicrobial assay, hemolymph pooled from ten bees was diluted 1.25× with phenylthiourea (1 mg/ml in PBS) to prevent coagulation and melanization, and then stored at -80°C. The remaining bee abdomens were kept at -80°C for further RNA extraction and analysis of the relative gene expression of selected AMPs.

Determination of antimicrobial activity

Antimicrobial activity in hemolymph treated with phenylthiourea was determined by radial diffusion on agar plates with Micrococcus luteus (CCM 169) as previously described [2]. Briefly, the bacterial suspension was cultured overnight in liquid LB medium (MOBIO, Carlsbad, CA, USA) with constant agitation (200 rpm; room temperature). The culture was diluted to OD₆₀₀ value of 1.5 and mixed with melted LB agar (4% agar in LB; MOBIO, Carlsbad, CA, USA) in a ratio of 1:500 (v/v). Five hemolymph samples per treatment and sampling time point were used, and 5 μl of hemolymph applied to wells prepared in agar plates supplemented with M. luteus. The resulting inhibition zones around the wells were measured after 24 hours of incubation at 30°C. The antibacterial activity of hemolymph was quantified using lysozyme standards (EC 3.2.1.17; Sigma-Aldrich, St. Louis, MO, USA).

Quantification of relative gene expression

Pools of five bee abdomens were homogenized for RNA extraction. The abdomens were smashed in plastic bags (Bioreba, Reinach, Switzerland) with a pestle in 1 ml of guanidinium isothiocyanate lysis buffer with 1% β-mercaptoethanol (Sigma-Aldrich, St. Louis, MO, USA) [27]. RNA isolation, cDNA synthesis, and qPCR were performed according to Dostálková et al. [3] with several modifications: reverse transcription was performed with an iScript^™ cDNA Synthesis Kit (BioRad, Hercules, CA, USA) in 10 μl reaction volumes according to the manufacturer’s protocol. The expected PCR product size was tested by gel electrophoresis in 2.5% (w/v) agarose gel with detection by GelRed^® Nucleic Acid Gel Stain, 10000× (Biotium, Fremont, CA, USA) and with a 50–1000 bp PCR Marker (Promega, Madison, WI, USA). To ensure the RNA samples were free of DNA contamination, we used RNA isolates as templates for PCR with the housekeeping gene EF-1 alpha. The PCR results were then visualized using agarose gel electrophoresis. For all RNA samples there was no PCR product which indicated genomic DNA contamination. The relative expression of genes of interest (apidaecin type 14, abaecin, defensin 1 and hymenoptaecin) to reference genes (RPS-5, EF-1 alpha) was calculated according to Pfaffl [28]. Briefly, the Ct values were obtained for both the gene of interest and the reference genes, as well as the efficiency of the qPCR reaction for each gene. The individual gene expression was calculated by multiplying its efficiency by the Ct value. The relative gene expression was then determined by normalizing reference gene expression to the expression of the gene of interest. The final value represents the geometric mean of the ratios for the two reference genes. This allows for accurate relative quantification of the target gene expression levels, corrected for the reference genes’ expression as internal controls. The calculated relative gene expression is independent of the control group, allowing for multiple comparisons across different experimental groups. This normalization to reference genes ensures that variations in gene expression can be compared across conditions without relying on a specific control group, enabling robust and consistent analysis. See S1 Table for primer sequences, efficiencies, and the formula used for the calculation of the relative gene expression. Nine to ten samples were analyzed per each treatment and sampling time point.

Quantification of Apidaecin 1 in the hemolymph by LC-MS analysis

Apidaecin 1 was quantified in the 2 μl hemolymph of individual bees, diluted 10× with 0.1% trifluoracetic acid. Seven to sixteen samples were prepared for each condition. Before measurement, the hemolymph samples were lyophilized and subsequently dissolved in 20 μl 5% formic acid prior to UHPLC-MS analysis. Apidaecin 1 was quantified on a UHPLC-QTOF system Dionex Ultimate 3000 UHPLC system (Thermo Fisher Scientific, Waltham, MA, USA) coupled with a Compact qTOF mass spectrometer (Bruker Daltonics, Bremen, Germany) with electrospray ionization. Apidaecin 1 was quantified using an isotopically [¹³C₆¹⁵N₄] labeled internal standard of Apidaecin 1A (purity >98%; Clonestar Peptide Services, Brno, Czech Republic) according to Danihlík et al. [29]. The LC-MS method set-up was according to Dostálková et al. [3].

Quantification of Abaecin, Defensin 1, and Hymenoptaecin by ELISA

Hemolymph pooled from five bees, diluted 10× with 0.1% trifluoracetic acid was used for the ELISA assay performed on a Corning^® 96 Well EIA/RIA Assay Microplate (Corning Inc., Somerville, MA, USA). Three samples per treatment and sampling time point were prepared. Before measurement, the hemolymph samples were lyophilized and subsequently dissolved in 600 μl of coating buffer. The volume of 100 μl of diluted samples, blanks (coating buffer), positive controls (peptide epitope supplied by the antibody manufacturer), negative controls (hemolymph from freshly emerged bee), and calibrators were pipetted per ELISA well and incubated at 4°C overnight for antigen binding. The next day, the ELISA plates were washed three times with 1× washing buffer with 0.1% (v/v) Tween 20. The unspecific binding was blocked with 200 μl of 0.5% non-fat milk (Sigma-Alrich, St. Louis, MO, USA) and incubated for 2 hours at 37°C, then again three times washed with washing buffer with 0.1% (v/v) Tween 20. The primary polyclonal rabbit antibodies (Clonestar Peptide Services, Brno, Czech Republic) were diluted in 1× washing buffer as follows: 1:500 for Defensin 1, 1:250 for Abaecin, and 1:500 for Hymenoptaecin antibody. Subsequently, the plate was incubated at 37°C for one hour, washed three times with washing buffer, and afterwards incubated at 37°C for one hour with secondary antibody (goat anti-rabbit IgG (whole molecule)-peroxidase conjugate, Sigma-Aldrich, St. Louis, MO, USA) diluted 1:3000 in 1× washing buffer with 0.1% (v/v) Tween 20. Then the plate was washed again four times with 1× washing buffer with 0.1% (v/v) Tween 20 and incubated in the dark at 37°C for one hour with 100 μl of the 2.4 mM tetramethylbenzidine substrate in phospho-citrate buffer with sodium perborate. The reaction was stopped by adding 50 μl of 0.5M H₂SO₄ to each well. The absorbance was recorded at 450 nm in a Synergy HT microplate reader (BioTek, Winooski, VT, USA). See S1 File for details on the preparation of ELISA buffers and reagents.

Hemocytes quantification

The concentration of circulating hemocytes was analyzed according to Kunc et al. and Marringa et al. [26, 30]. Shortly, 10 μl of fresh hemolymph pooled from five bees were 50× diluted in cold insect physiological saline (10 mM EDTA, 30 mM sodium citrate; pH 7.0), and stained with wheat germ agglutinin conjugated with fluorescein isothiocyanate (WGA-FITC; 1 μg/ml; Sigma-Aldrich, St. Louis, MO, USA) and propidium iodide (0.1 mg/ml; Sigma-Aldrich, St. Louis, MO, USA). The samples were incubated at room temperature for 15 min in the dark and measured using a spectral flow cytometer Northern Lights 3000 (Cytek Biosciences, Fremont, CA, USA). The results were processed in SpectroFlo^® software. WGA-FITC positive and propidium iodide negative events were gated within singlets and quantified as live hemocytes with intact membranes. Five to eight samples were prepared for each condition, i.e., each treatment and sampling time point.

Statistics

Original data gained from beforementioned analyses were employed to create probability model using Bayesian statistics. A simple Bayesian model of the response was developed in pymc python package [31]. The responses are modeled as a scaled derivative of the logistic function:

The code is in S2 File.

This enables us to have three easily interpretable parameters (Fig 1):

ampl: the highest value of the response, corresponds to the highest point on the chart;
loc: the time point of the maximal response;
scale: the speed of the response increase and decline (the model is symmetrical around the loc value).

Download:

Fig 1. A selection of functions that are fitted to data.

The figure shows the interpretation of the ‘ampl’ parameter, which corresponds to the highest point on the chart, and ‘loc’ parameter which corresponds to the time of maximal response. The ‘scale’ parameter is varied to show its effect on the speed of the response increase and decline.

https://doi.org/10.1371/journal.pone.0311415.g001

In Bayesian statistics, a prior is a probability distribution that represents the beliefs or information about a parameter before considering the current data. It expresses the uncertainty or prior knowledge about the parameter of interest. The prior distribution is updated with observed data to obtain the posterior distribution, which combines the prior information with the likelihood of the observed data to provide an updated estimate of the parameter. Our model is then specified using log-normal prior for the ampl parameter, gaussian prior for the loc parameter, and half-normal prior for the scale parameter (Fig 2). Each group and population have its own approximated parameters. The state-of-the-art No-U-Turn sampler [32] is then used to sample the posterior distribution of the parameters. The results are visualized in graphs showing ampl and loc at 95% Highest Density Interval (HDI). The values in the Result section describe the ampl or loc with 2.5–97.5% dispersion, i.e. 95% HDI, in the brackets. The original data are in S3 File.

Download:

Fig 2. Illustration of the developed hierarchical Bayesian model.

The parameters loc, ampl, scale, and sd, represents different prior distributions across the dimensions of four groups (Control, CO₂, PBS, Bacteria) and two populations (Summer, Winter). The observed data is modeled using a normal distribution function of time and the aforementioned parameters, tailored to each subgroup within the data.

https://doi.org/10.1371/journal.pone.0311415.g002

Results

Antimicrobial activity

The results obtained after the Bayesian modeling show a higher antimicrobial activity of honey bees in the winter compared to the summer population in all experimental groups (Fig 3A). The Control group shows a maximum of 4.94 (3.73–6.44) mg/ml of lysozyme in winter and 3.29 (2.62–4.04) mg/ml of lysozyme in the summer bee population (S2 Table). After the injection of heat-killed bacteria or PBS, the antimicrobial activity increased in both summer and winter bee populations compared to Control and CO₂ group. Its maximal value in the Bacteria group is higher in winter bees, with a maximum of 22.94 (19.03–28.72) mg/ml of lysozyme in comparison to summer bees, with a maximum of 7.60 (6.74–8.46) mg/ml of lysozyme. The model predicts that winter bees need more time to respond and reach the peak of their antimicrobial activity (Fig 3B). The maximal value of antimicrobial activity after bacterial stimuli was reached at 18.0 (15.1–22.9) hours in summer and 25.9 (21.8–32.0) hours in winter bees. For probability curves of the immune response, please see S1–S6 Figs.

Download:

Fig 3. Bayesian model of antimicrobial activity measured as lysozyme equivalent (mg/ml).

Comparison of four experimental groups, as well as summer (orange) and winter (blue) bee populations. (A) The highest values of the response (ampl; white dot) with 95% highest density interval (HDI; thin lines). (B) Time in hours of the maximal response (loc; white dot) with the 95% HDI (thin lines). The thick lines show 50% HDI.

https://doi.org/10.1371/journal.pone.0311415.g003

Relative gene expression and concentration of selected antimicrobial peptides

In the Control group, abaecin relative gene expression is higher in summer, with maximal value of 1.09 (0.88–1.34), compared to 0.34 (0.25–0.45) in winter (Fig 4A). Following the injection of heat-killed bacteria, the values increased sharply, reaching 3.91 (3.16–4.70) in summer and 2.30 (1.84–2.81) in winter. The highest value of Abaecin relative concentration in the honey bee hemolymph is 0.36 (0.27–0.45) and 0.24 (0.21–0.26) in the Control summer and winter honey bee, respectively (Fig 4C). After the bacterial challenge, the values reach 0.85 (0.44–1.31) in summer and 0.44 (0.35–0.54) in winter. According to the Bayesian model, the maximal value of abaecin relative gene expression after bacterial stimuli is reached at 17.3 (15.5–20.0) hours in summer and 19.4 (16.1–25.1) hours in winter bees (Fig 4B). The relative concentration of Abaecin in the Bacteria group reaches its maximum at 18.4 (11.8–26.7) hours in summer and 18.1 (14.5–23.5) hours in winter honey bees (Fig 4D).

Download:

Fig 4. Bayesian model of abaecin relative gene expression and its relative peptide concentration in the hemolymph.

Comparison of four experimental groups, as well as summer (orange) and winter (blue) bee populations. (A) The highest values of the abaecin relative gene expression (ampl; white dot) with the 95% highest density interval (HDI; thin lines). (B) Time in hours of the abaecin relative gene expression maximal response (loc; white dot) with the 95% HDI (thin lines). (C) The highest values of the Abaecin relative peptide concentration measured as absorbance at 450 nm (ampl; white dot) with the 95% HDI (thin lines). (D) Time in hours of the Abaecin relative peptide concentration maximal response (loc; white dot) with the 95% HDI (thin lines). The thick lines show 50% HDI.

https://doi.org/10.1371/journal.pone.0311415.g004

The relative expression of apidaecin 1 in the Control group reaches the highest value of 0.82 (0.69–0.97) in summer and 0.21 (0.17–0.25) in the winter (Fig 5A). Post the injection of heat-killed bacteria, the gene expression surged to 2.80 (2.46–3.16) in summer and 1.11 (0.85–1.42) in winter bees. Peptide concentration of Apidaecin 1 in the Control group is slightly lower in summer bees in comparison to winter bees, with a maximal value of 6.35 (5.05–7.68) ng/μl and 9.53 (4.68–18.92) ng/μl, respectively (Fig 5C). Its highest value in summer bees injected with bacteria is 8.21 (6.63–10.09) ng/μl, while in winter bees it reaches 41.97 (31.73–57.64) ng/μl. The highest value of the apidaecin 1 relative gene expression after bacterial stimuli is reached at 17.2 (15.3–20.2) hours in summer bees and 17.8 (15.0–22.0) hours in winter bees (Fig 5B). The highest value of Apidaecin 1 concentration in honey bee hemolymph in the Bacteria group is reached at 18.0 (14.0–24.7) hours and 25.3 (21.4–31.7) hours in summer and winter bees, respectively (Fig 5D).

Download:

Fig 5. Bayesian model of apidaecin 1 relative gene expression and its peptide concentration in the hemolymph.

Comparison of four experimental groups, as well as summer (orange) and winter (blue) bee populations. (A) The highest values of the apidaecin 1 relative gene expression (ampl; white dot) with the 95% highest density interval (HDI; thin lines). (B) Time in hours of the apidaecin 1 relative gene expression maximal response (loc; white dot) with the 95% HDI (thin lines). (C) The highest values of the Apidaecin 1 peptide concentration in ng/μl (ampl; white dot) with the 95% HDI (thin lines). (D) Time in hours of the Apidaecin 1 peptide concentration maximal response (loc; white dot) with the 95% HDI (thin lines). The thick lines show 50% HDI.

https://doi.org/10.1371/journal.pone.0311415.g005

Relative gene expression of defensin 1 in the Control group reaches the maximal value of 2.17 (1.36–3.82) in summer and 0.52 (0.35–0.73) in winter (Fig 6A). After bacterial challenge, the values rise to 13.90 (11.91–15.89) in summer and 7.13 (5.99–8.44) in winter. In the Control group, the highest value of Defensin 1 relative concentration in the honey bee hemolymph is 0.30 (0.25–0.34) in summer and 0.25 (0.23–0.28) in winter (Fig 6C). After the bacterial challenge the values reach 1.20 (0.81–1.57) in summer and 0.68 (0.50–0.87) in winter. According to the Bayesian model, the maximal value of defensin 1 relative gene expression after bacterial stimuli is reached at 14.9 (13.7–16.0) hours in summer and 24.0 (20.2–29.5) hours in winter bees (Fig 6B). The relative concentration of Defensin 1 in the Bacteria group reaches its maximum at 22.9 (18.5–29.4) hours in summer and 25.8 (19.9–32.6) hours in winter bees (Fig 6D).

Download:

Fig 6. Bayesian model of defensin 1 relative gene expression and its relative peptide concentration in the hemolymph.

Comparison of four experimental groups, as well as summer (orange) and winter (blue) bee populations. (A) The highest values of the defensin 1 relative gene expression measured as absorbance at 450 nm (ampl; white dot) with the 95% highest density interval (HDI; thin lines). (B) Time in hours of the defensin 1 relative gene expression maximal response (loc; white dot) with the 95% HDI (thin lines). (C) The highest values of the Defensin 1 relative peptide concentration (ampl; white dot) with the 95% HDI (thin lines). (D) Time in hours of the Defensin 1 relative peptide concentration maximal response (loc; white dot) with the 95% HDI (thin lines). The thick lines show 50% HDI.

https://doi.org/10.1371/journal.pone.0311415.g006

Hymenoptaecin relative gene expression in the Control group reaches the maximal value of 12.05 (7.43–27.76) in summer and 0.44 (0.25–0.68) in winter (Fig 7A). After the bacterial challenge the values rise to 24.33 (20.83–27.83) in summer and 13.84 (11.32–16.53) in winter. The highest value of Hymenoptaecin relative concentration in the honey bee hemolymph is 0.84 (0.29–1.42) and 0.23 (0.22–0.25) in the Control summer and winter honey bee, respectively (Fig 7C). After the bacterial challenge, the values reach 1.82 (1.30–2.36) in summer and 1.60 (1.21–1.97) in winter. According to the Bayesian model, the maximal value of hymenoptaecin relative gene expression after bacterial stimuli is reached at 13.5 (12.4–14.6) hours in summer and 15.9 (14.0–17.8) hours in winter bees (Fig 7B). The relative concentration of Hymenoptaecin in the hemolymph of the Bacteria group reaches its maximum at 20.1 (14.6–27.1) hours in summer and 17.0 (15.4–18.4) hours in winter honey bees (Fig 7D). For probability curves of the immune response, please see S1–S6 Figs.

Download:

Fig 7. Bayesian model of hymenoptaecin relative gene expression and its relative peptide concentration in the hemolymph.

Comparison of four experimental groups, as well as summer (orange) and winter (blue) bee populations. (A) The highest values of the hymenoptaecin relative gene expression (ampl; white dot) with the 95% highest density interval (HDI; thin lines). (B) Time in hours of the hymenoptaecin relative gene expression maximal response (loc; white dot) with the 95% HDI (thin lines). (C) The highest values of the Hymenoptaecin relative peptide concentration measured as absorbance at 450 nm (ampl; white dot) with the 95% HDI (thin lines). (D) Time in hours of the Hymenoptaecin relative peptide concentration maximal response (loc; white dot) with the 95% HDI (thin lines). The thick lines show 50% HDI.

https://doi.org/10.1371/journal.pone.0311415.g007

Concentration of circulating hemocytes

The Bayesian model shows that the concentration of hemocytes is overall higher in winter compared to summer bees (Fig 8A). In the Control group, the concentration reaches the highest value of 21730 (19380–24020) hemocytes/μl in the winter and 13300 (11900–14830) hemocytes/μl in the summer population. After injection of heat-killed bacteria we observed increased hemocyte concentration in summer bees (S6 Fig), with a maximum of 14790 (10590–19100) hemocytes/μl followed by a steady decrease. The highest value of hemocyte concentration in the summer population is reached at 5.9 (0.3–11.8) hours after injection with heat-killed bacteria (Fig 8B). For probability curves of the immune response, please see S1–S6 Figs.

Download:

Fig 8. Bayesian model of hemocyte concentration (circulating hemocytes/μl of hemolymph) measured by flow cytometry.

Comparison of four experimental groups, as well as summer (orange) and winter (blue) bee populations. (A) The highest values of the response (ampl; white dot) with the 95% highest density interval (HDI; thin lines). (B) Time in hours of the maximal response (loc; white dot) with the 95% HDI (thin lines). The thick lines show 50% HDI.

https://doi.org/10.1371/journal.pone.0311415.g008

Discussion

Applying Bayesian statistics to describe the honey bee immune response dynamics to a heat-killed bacterial pathogen offers several advantages over relying on separated measurements, which often provide limited insights into infection dynamics. In an experiment where measurements are taken at specific time intervals, Bayesian statistics allow us to model the system behavior between these points, enabling us to estimate peak values that were not directly measured. Bayesian models integrate prior knowledge with current data, allowing for a more comprehensive analysis of the immune response. Bayesian methods also account for uncertainty and variability in the data, providing a probabilistic framework that quantifies the confidence in the findings. To obtain original data, we injected 10-day-old honey bees collected from the summer and winter populations with heat-killed bacteria S. marcescens and sampled their tissues for laboratory analyses at several time points after bacteria injection. Here, we present a model based on our experimental data that provides improved approximation of the dynamics of honey bee immune response to the heat-killed bacterial pathogen, compared to a commonly used frequentist statistical data evaluation.

The honey bee immune system consists of cellular and humoral immunity. The humoral response to bacteria includes mainly melanization, secretion of AMPs and other molecules with bactericidal or bacteriostatic functions. The increase in the honey bee antimicrobial activity can be triggered not only by live pathogens, but also by the injection of pathogen-associated molecular patterns such as lipopolysaccharide [22], the outer membrane component of gram-negative bacteria, or wounding by aseptic injection of saline buffer [24]. Here, we show an increase in antimicrobial activity in both summer and winter bees after the injection of either heat-killed gram-negative bacteria S. marcescens or PBS with similar progression observed in both cases. In line with our previous studies comparing summer and winter bee populations, we have demonstrated that the winter bees are characterized by a higher constitutive level and a stronger inducible capacity of antimicrobial activity [2, 3].

Previously, Gätschenberger et al. [33] detected antimicrobial activity in one-day-old honey bee workers injected with live Escherichia coli six hours post-treatment, showing a peak at 24 hours. This level of antimicrobial activity remained stable till the end of the experiment at 72 hours post-injection. In our study on 10-day-old bees, the Bayesian model shows the expected peak of the antimicrobial activity at 18.0 hours after the injection of heat-killed bacteria in summer bees, whereas the peak in the winter bees is reached at 25.9 hours with a presumption of prolonged duration. We hypothesize that the delay in reaching the maximum in the winter population may be due to the approximately three times stronger response mounted in the winter bees.

The key element of insect humoral defense are AMPs. In the hemolymph of honey bees infected with viable E. coli, four types of AMPs were described: Apidaecins [34], Abaecin [35], Hymenoptaecin [36] and Defensin [37]. In this study, the Bayesian model confirmed our previous findings on higher gene expression of AMPs in summer bees [3]. The immune response was activated even after an aseptic injection of saline buffer, which is in agreement with previous results [24, 38, 39]. Zhao et al. [40] reported induced expression of antibacterial peptides in bees, that were in contact with Varroa mites. It is not clear if the bee immune system was activated due to the observed increase of the deformed wing virus load, possible bacterial transfer through the wound caused by the feeding mite [10, 11], or the wound itself.

Apidaecins are assumed to be the major AMPs in honey bee hemolymph challenged with bacteria, mainly due to the presence of repetitive units in the gene precursor structure, which allows rapid amplification [41]. Previously, they were reported to be the main actors against gram-negative bacteria [34]. However, in the same study, Apidaecin 1 showed no or low effect against S. marcescens. In this study, the measured concentration of Apidaecin 1 in the hemolymph of summer honey bees after injection of heat-killed S. marcescens was rather moderate, with a maximum of 8.21 ng/μl, while the winter honey bees reached a maximum of 41.97 ng/μl of Apidaecin 1. In our previous study, when measured 24 hours post-infection by heat-killed E. coli and P. larvae, the concentration of Apidaecin 1 was 23.7 ng/μl in summer and 20.3 ng/μl in the winter population [3]. It could be hypothesized, that the discrepancy between concentrations might have been caused by different bacteria used in the study. Moreover, a higher concentration of Apidaecin 1 in winter bees caused by heat-killed S. marcescens in comparison to summer bees can indicate possible higher sensitivity to that pathogen during winter [21].

Casteels-Josson et al. [37] reported that transcripts of defensin 1 occur in the honey bee hemolymph challenged with bacteria later (12 hours) than the transcripts of other AMPs and that the upregulation of defensin 1 was minimal. On the other hand, Lourenço et al. [42] detected a statistically significant upregulation of defensin 1 in bees injected with S. marcescens compared to non-injected bees already 5 hours post-treatment. In this study, a slow increase of defensin 1 gene expression in winter honey bees with a peak at 24.0 hours after bacterial injection was observed. This was not the case in summer bees, which reached the maximal value at 14.9 hours under the same conditions. Although Defensin 1 was reported to act primarily against gram-positive bacteria [37], its gene expression was stronger compared to abaecin or apidaecin 1 in both seasons. Change in defensin 1 gene expression was observed after infection with gram-negative bacteria E. coli also by Richard et al. [43], who suggested possible cross-talk among signaling pathways regulating the expression of AMPs. In the honey bees injected with heat-killed S. marcescens, we report a similar pattern in Defensin 1 concentration with a slower rise in both summer and winter populations, with peaks at 22.9 and 25.8 hours, respectively.

Abaecin and Hymenoptaecin are effective against both, gram-positive and gram-negative bacteria [35, 36]. It was also reported that they are regulated by the Imd pathway [44], which is activated mostly by membrane components of gram-negative bacteria. In our study, hymenoptaecin shows the highest gene expression of the four studied peptides, which is in agreement with our previous study describing immune reaction after 24 hours post-treatment [3]. In addition, the high relative gene expression is followed by a high relative concentration in the hemolymph, which exceeds levels of Abaecin and Defensin 1, corresponding with the results of previous studies [24, 45].

Previously, a higher hemocyte concentration in the winter honey bees compared to the summer bees was reported [2]. This study confirmed the previous observation using flow cytometry and hemolymph samples of 10-day-old honey bees. Despite the higher concentration of hemocytes, winter bees seem to rely primarily on humoral immunity. Gätschenberger et al. [33] observed melanized nodules created only in the hemolymph of the summer and not the winter honey bees injected with live E. coli. Our model also shows shifts in hemocyte concentration only in the summer bee population. Azzami et al. [45] reported that the first nodules appeared in summer honey bees at 2 hours after injection of live E. coli, and the maximal count of nodules was reached at 6 hours post-treatment. In this study, the concentration of hemocytes in summer bees was the highest at 5.9 hours after injection with heat-killed S. marcescens and decreased over time. It could be suggested that more hemocytes are released into the hemolymph within a few hours after bacterial challenge, followed by their depletion in the immune reaction, as observed in various insect species [46, 47]. We also suggest that summer bees utilize both cellular and humoral immunity, while winter honeybees rely predominantly on humoral immunity. More research is needed, as there is no information about the changes in hemocyte populations in summer bees challenged with bacteria. In addition, the functional tests analyzing phagocytosis and nodulation would shed more light on the fate of these hemocytes.

In conclusion, Bayesian statistics provide a powerful tool for integrating prior knowledge, experimental data, and uncertainty estimation, making them particularly well-suited for understanding the dynamics of the immune system and its components. This approach can update predictions continuously as new data becomes available, enhancing the accuracy and reliability of the results. In this study, humoral defense both in summer and winter bee populations was observed; however, the fundamental parameter of cellular immunity, the hemocyte concentration, responded to heat-killed bacteria only in summer bees. It could be suggested that cellular immune defense is active mainly in summer bees, while winter bees rely mostly on humoral immune mechanisms. Moreover, our results established the eligible time points for studies focused on the humoral immunity of honey bees. These findings can be applied when using heat-killed bacteria as immune stimulators in future studies focused on the induction of honey bee immunity.

Supporting information

S1 File. ELISA buffers and reagents.

https://doi.org/10.1371/journal.pone.0311415.s001

(DOCX)

S2 File. Python script.

https://doi.org/10.1371/journal.pone.0311415.s002

(ZIP)

S3 File. Original data.

https://doi.org/10.1371/journal.pone.0311415.s003

(ZIP)

S1 Table. Primer sequences and the corresponding target gene names used for PCR reactions.

https://doi.org/10.1371/journal.pone.0311415.s004

(DOCX)

S2 Table. Values of the ampl and loc with 2.5–97.5% dispersion, i.e. 95% HDI.

https://doi.org/10.1371/journal.pone.0311415.s005

(XLSX)

S1 Fig. Bayesian model of antimicrobial activity dynamics measured as lysozyme equivalent (mg/ml).

Comparison of four experimental groups: Control (CTRL; green), CO2 (CO2; yellow), PBS (PBS; red), and Bacteria (BACTERIA; purple). (A) Summer and (B) winter honey bee population.

https://doi.org/10.1371/journal.pone.0311415.s006

(TIF)

S2 Fig. Bayesian model showing dynamics of abaecin relative gene expression and its relative peptide concentration in the hemolymph.

Comparison of four experimental groups: Control (CTRL; green), CO2 (CO2; yellow), PBS (PBS; red), and Bacteria (BACTERIA; purple). Abaecin relative gene expression of (A) summer and (B) winter honey bee population. Abaecin relative peptide concentration of (C) summer and (D) winter honey bee population.

https://doi.org/10.1371/journal.pone.0311415.s007

(TIF)

S3 Fig. Bayesian model showing dynamics of apidaecin 1 relative gene expression and its peptide concentration in the hemolymph.

Comparison of four experimental groups: Control (CTRL; green), CO2 (CO2; yellow), PBS (PBS; red), and Bacteria (BACTERIA; purple). Apidaecin 1 relative gene expression of (A) summer and (B) winter honey bee population. Apidaecin 1 peptide concentration of (C) summer and (D) winter honey bee population.

https://doi.org/10.1371/journal.pone.0311415.s008

(TIF)

S4 Fig. Bayesian model showing dynamics of defensin 1 relative gene expression and its relative peptide concentration in the hemolymph.

Comparison of four experimental groups: Control (CTRL; green), CO2 (CO2; yellow), PBS (PBS; red), and Bacteria (BACTERIA; purple). Defensin 1 relative gene expression of (A) summer and (B) winter honey bee population. Defensin 1 relative peptide concentration of (C) summer and (D) winter honey bee population.

https://doi.org/10.1371/journal.pone.0311415.s009

(TIF)

S5 Fig. Bayesian model showing dynamics of hymenoptaecin relative gene expression and its relative peptide concentration in the hemolymph.

Comparison of four experimental groups: Control (CTRL; green), CO2 (CO2; yellow), PBS (PBS; red), and Bacteria (BACTERIA; purple). Hymenoptaecin relative gene expression of (A) summer and (B) winter honey bee population. Hymenoptaecin relative peptide concentration of (C) summer and (D) winter honey bee population.

https://doi.org/10.1371/journal.pone.0311415.s010

(TIF)

S6 Fig. Bayesian model of hemocyte concentration (circulating hemocytes/μl of hemolymph) dynamics measured by flow cytometry.

Comparison of four experimental groups: Control (CTRL; green), CO2 (CO2; yellow), PBS (PBS; red), and Bacteria (BACTERIA; purple). (A) Summer and (B) winter honey bee population.

https://doi.org/10.1371/journal.pone.0311415.s011

(TIF)

Acknowledgments

The authors thank Oldřich Veverka for his highly professional help with the handling and preparation of experimental colonies and Zuzana Žvátorová for technical help with measurements. While preparing this article, the author(s) used ChatGPT 3.5 and Grammarly to improve readability.

References

1. Münch D, Kreibich CD, Amdam G v. Aging and its modulation in a long-lived worker caste of the honey bee. Journal of Experimental Biology. 2013 May;216(9):1638–49.
- View Article
- Google Scholar
2. Kunc M, Dobeš P, Hurychová J, Vojtek L, Poiani SB, Danihlík J, et al. The year of the honey bee (Apis mellifera L.) with respect to its physiology and immunity: A search for biochemical markers of longevity. Insects. 2019 Aug 1;10(8). pmid:31394797
- View Article
- PubMed/NCBI
- Google Scholar
3. Dostálková S, Dobeš P, Kunc M, Hurychová J, Škrabišová M, Petřivalský M, et al. Winter honeybee (Apis mellifera) populations show greater potential to induce immune responses than summer populations after immune stimuli. Journal of Experimental Biology. 2021 Feb 1;224(3). pmid:33288532
- View Article
- PubMed/NCBI
- Google Scholar
4. Danihlík J, Aronstein K, Petřivalský M. Antimicrobial peptides: A key component of honey bee innate immunity. Vol. 54, Journal of Apicultural Research. Taylor and Francis Ltd; 2015. p. 123–36.
5. Simone-Finstrom M. Social Immunity and the Superorganism: Behavioral Defenses Protecting Honey Bee Colonies from Pathogens and Parasites. Bee World. 2017 Jan 2;94(1):21–9.
- View Article
- Google Scholar
6. Fowler AE, Irwin RE, Adler LS. Parasite defense mechanisms in bees: behavior, immunity, antimicrobials, and symbionts. Emerg Top Life Sci. 2020 Jul 2;4(1):59–76. pmid:32558901
- View Article
- PubMed/NCBI
- Google Scholar
7. Traynor KS, Mondet F, de Miranda JR, Techer M, Kowallik V, Oddie MAY, et al. Varroa destructor: A Complex Parasite, Crippling Honey Bees Worldwide. Vol. 36, Trends in Parasitology. Elsevier Ltd; 2020. p. 592–606.
8. Bruce Krejčí A, Votýpková K, Lukeš J, Votýpka J. Varroa destructor. Vol. 39, Trends in Parasitology. Elsevier Ltd; 2023. p. 487–8.
9. Kuster RD, Boncristiani HF, Rueppell O. Immunogene and viral transcript dynamics during parasitic Varroa destructor mite infection of developing honey bee (Apis mellifera) pupae. Journal of Experimental Biology. 2014 May 15;217(10):1710–8. pmid:24829325
- View Article
- PubMed/NCBI
- Google Scholar
10. Gliński Z, Jarosz J. Varroa jacobsoni as a carrier of bacterial infections to a recipient bee host. Apidologie. 1992;23(1):25–31.
- View Article
- Google Scholar
11. Kanbar G, Engels W. Ultrastructure and bacterial infection of wounds in honey bee (Apis mellifera) pupae punctured by Varroa mites. Parasitol Res. 2003 Aug 1;90(5):349–54. pmid:12684884
- View Article
- PubMed/NCBI
- Google Scholar
12. Ramsey SD, Ochoa R, Bauchan G, Gulbronson C, Mowery JD, Cohen A, et al. Varroa destructor feeds primarily on honey bee fat body tissue and not hemolymph. Proceedings of the National Academy of Sciences. 2019 Jan 29;116(5):1792–801.
- View Article
- Google Scholar
13. Burritt NL, Foss NJ, Neeno-Eckwall EC, Church JO, Hilger AM, Hildebrand JA, et al. Sepsis and hemocyte loss in honey bees (Apis mellifera) Infected with serratia marcescens strain sicaria. PLoS One. 2016 Dec 1;11(12). pmid:28002470
- View Article
- PubMed/NCBI
- Google Scholar
14. El Sanousi S.M., El Sarag M.S.A., Sumaia E. Mohamed. Properties of Serratia marcescens Isolated from Diseased Honeybee (Apis mellifera) Larvae. Vol. 133, Journal of General Microbiology. 1987.
- View Article
- Google Scholar
15. Fünfhaus A, Ebeling J, Genersch E. Bacterial pathogens of bees. Curr Opin Insect Sci. 2018 Apr;26:89–96. pmid:29764667
- View Article
- PubMed/NCBI
- Google Scholar
16. Raymann K, Moran NA. The role of the gut microbiome in health and disease of adult honey bee workers. Curr Opin Insect Sci. 2018 Apr;26:97–104. pmid:29764668
- View Article
- PubMed/NCBI
- Google Scholar
17. Al Naggar Y, Singavarapu B, Paxton RJ, Wubet T. Bees under interactive stressors: the novel insecticides flupyradifurone and sulfoxaflor along with the fungicide azoxystrobin disrupt the gut microbiota of honey bees and increase opportunistic bacterial pathogens. Science of The Total Environment. 2022 Nov;849:157941. pmid:35952893
- View Article
- PubMed/NCBI
- Google Scholar
18. Al Naggar Y, Wubet T. Chronic exposure to pesticides disrupts the bacterial and fungal co-existence and the cross-kingdom network characteristics of honey bee gut microbiome. Science of The Total Environment. 2024 Jan;906:167530. pmid:37832690
- View Article
- PubMed/NCBI
- Google Scholar
19. Raymann K, Shaffer Z, Moran NA. Antibiotic exposure perturbs the gut microbiota and elevates mortality in honeybees. PLoS Biol. 2017 Mar 14;15(3):e2001861. pmid:28291793
- View Article
- PubMed/NCBI
- Google Scholar
20. Braglia C, Alberoni D, Garrido PM, Porrini MP, Baffoni L, Scott D, et al. Vairimorpha (Nosema) ceranae can promote Serratia development in honeybee gut: an underrated threat for bees? Front Cell Infect Microbiol. 2024 May 13;14. pmid:38808063
- View Article
- PubMed/NCBI
- Google Scholar
21. Raymann K, Coon KL, Shaffer Z, Salisbury S, Moran NA. Pathogenicity of Serratia marcescens Strains in Honey Bees. McFall-Ngai MJ, editor. mBio [Internet]. 2018 Nov 7;9(5). Available from: https://journals.asm.org/doi/10.1128/mBio.01649-18
- View Article
- Google Scholar
22. Laughton AM, Boots M, Siva-Jothy MT. The ontogeny of immunity in the honey bee, Apis mellifera L. following an immune challenge. J Insect Physiol. 2011 Jul;57(7):1023–32. pmid:21570403
- View Article
- PubMed/NCBI
- Google Scholar
23. Chan QWT, Melathopoulos AP, Pernal SF, Foster LJ. The innate immune and systemic response in honey bees to a bacterial pathogen, Paenibacillus larvae. BMC Genomics. 2009 Aug 21;10. pmid:19695106
- View Article
- PubMed/NCBI
- Google Scholar
24. Randolt K, Gimple O, Geissendörfer J, Reinders J, Prusko C, Mueller MJ, et al. Immune-related proteins induced in the hemolymph after aseptic and septic injury differ in honey bee worker larvae and adults. Arch Insect Biochem Physiol. 2008 Dec;69(4):155–67. pmid:18979500
- View Article
- PubMed/NCBI
- Google Scholar
25. van de Schoot R, Depaoli S, King R, Kramer B, Märtens K, Tadesse MG, et al. Bayesian statistics and modelling. Nature Reviews Methods Primers. 2021 Jan 14;1(1):1.
- View Article
- Google Scholar
26. Kunc M, Dobeš P, Ward R, Lee S, Čegan R, Dostálková S, et al. Omics-based analysis of honey bee (Apis mellifera) response to Varroa sp. parasitisation and associated factors reveals changes impairing winter bee generation. Insect Biochem Mol Biol. 2023 Jan;152:103877. pmid:36403678
- View Article
- PubMed/NCBI
- Google Scholar
27. Evans JD, Schwarz RS, Chen YP, Budge G, Cornman RS, De la Rua P, et al. Standard methods for molecular research in Apis mellifera. J Apic Res [Internet]. 2013 Jan 2;52(4):1–54. Available from: https://www.tandfonline.com/doi/full/10.3896/IBRA.1.52.4.11
- View Article
- Google Scholar
28. Pfaffl MW. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res. 2001 May 1;29(9):45e–45. pmid:11328886
- View Article
- PubMed/NCBI
- Google Scholar
29. Danihlík J, Šebela M, Petřivalský M, Lenobel R. A sensitive quantification of the peptide apidaecin 1 isoforms in single bee tissues using a weak cation exchange pre-separation and nanocapillary liquid chromatography coupled with mass spectrometry. J Chromatogr A. 2014 Dec;1374:134–44. pmid:25435459
- View Article
- PubMed/NCBI
- Google Scholar
30. Marringa WJ, Krueger MJ, Burritt NL, Burritt JB. Honey Bee Hemocyte Profiling by Flow Cytometry. PLoS One. 2014 Oct 6;9(10):e108486. pmid:25285798
- View Article
- PubMed/NCBI
- Google Scholar
31. Salvatier J, Wiecki T v., Fonnesbeck C. Probabilistic programming in Python using PyMC3. PeerJ Comput Sci. 2016 Apr 6;2:e55.
- View Article
- Google Scholar
32. Hoffman MD, Gelman A. The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian Monte Carlo. 2011 Nov 17; http://arxiv.org/abs/1111.4246
33. Gätschenberger H, Azzami K, Tautz J, Beier H. Antibacterial Immune Competence of Honey Bees (Apis mellifera) Is Adapted to Different Life Stages and Environmental Risks. PLoS One. 2013 Jun 17;8(6). pmid:23799099
- View Article
- PubMed/NCBI
- Google Scholar
34. Casteels P, Ampe C, Jacobs F, Vaeck M, Tempst P. Apidaecins: antibacterial peptides from honeybees. Vol. 8, The EMBO Journal. 1989. pmid:2676519
- View Article
- PubMed/NCBI
- Google Scholar
35. Casteels P, Ampe C, Riviere L, van Damme J, Elicone C, Fleming M, et al. Isolation and characterization of abaecin, a major antibacterial response peptide in the honeybee (Apis mellifera). Eur J Biochem. 1990;187(2):381–6. pmid:2298215
- View Article
- PubMed/NCBI
- Google Scholar
36. Casteels P, Ampe C, Jacobs F, Tempst P. Functional and chemical characterization of hymenoptaecin, an antibacterial polypeptide that is infection-inducible in the honeybee (Apis mellifera). Journal of Biological Chemistry. 1993;268(10):7044–54. pmid:8463238
- View Article
- PubMed/NCBI
- Google Scholar
37. Casteels-Josson K, Zhang W, Capaci T, Casteels P, Tempst P. Acute transcriptional response of the honeybee peptide-antibiotics gene repertoire and required post-translational conversion of the precursor structures. Journal of Biological Chemistry. 1994 Nov 18;269(46):28569–75. pmid:7961803
- View Article
- PubMed/NCBI
- Google Scholar
38. Lourenço AP, Guidugli-Lazzarini KR, Freitas FCP, Bitondi MMG, Simões ZLP. Bacterial infection activates the immune system response and dysregulates microRNA expression in honey bees. Insect Biochem Mol Biol. 2013 May;43(5):474–82. pmid:23499934
- View Article
- PubMed/NCBI
- Google Scholar
39. Evans JD, Aronstein K, Chen YP, Hetru C, Imler JL, Jiang H, et al. Immune pathways and defence mechanisms in honey bees Apis mellifera. Insect Molecular Biology. 2006. pmid:17069638
- View Article
- PubMed/NCBI
- Google Scholar
40. Zhao Y, Heerman M, Peng W, Evans JD, Rose R, Degrandi-Hoffman G, et al. The dynamics of Deformed wing virus concentration and host defensive gene expression after varroa mite parasitism in honey bees, Apis mellifera. Insects. 2019 Jan 1;10(1). pmid:30626033
- View Article
- PubMed/NCBI
- Google Scholar
41. Casteels-Josson K, Capaci T, Casteels P, Tempst P. Apidaecin multipeptide precursor structure: a putative mechanism for amplification of the insect antibacterial response. Vol. 12, The EMBO Journal. 1993. pmid:8467807
- View Article
- PubMed/NCBI
- Google Scholar
42. Lourenço AP, Martins JR, Torres FAS, Mackert A, Aguiar LR, Hartfelder K, et al. Immunosenescence in honey bees (Apis mellifera L.) is caused by intrinsic senescence and behavioral physiology. Exp Gerontol. 2019 May 1;119:174–83. pmid:30742904
- View Article
- PubMed/NCBI
- Google Scholar
43. Richard FJ, Holt HL, Grozinger CM. Effects of immunostimulation on social behavior, chemical communication and genome-wide gene expression in honey bee workers (Apis mellifera). BMC Genomics. 2012 Oct 16;13(1). pmid:23072398
- View Article
- PubMed/NCBI
- Google Scholar
44. Schlüns H, Crozier RH. Relish regulates expression of antimicrobial peptide genes in the honeybee, Apis mellifera, shown by RNA interference. Insect Mol Biol. 2007 Dec;16(6):753–9. pmid:18093004
- View Article
- PubMed/NCBI
- Google Scholar
45. Azzami K, Ritter W, Tautz J, Beier H. Infection of honey bees with acute bee paralysis virus does not trigger humoral or cellular immune responses. Arch Virol. 2012 Apr;157(4):689–702. pmid:22258854
- View Article
- PubMed/NCBI
- Google Scholar
46. Honti V, Csordás G, Márkus R, Kurucz É, Jankovics F, Andó I. Cell lineage tracing reveals the plasticity of the hemocyte lineages and of the hematopoietic compartments in Drosophila melanogaster. Mol Immunol. 2010 Jul;47(11–12):1997–2004. pmid:20483458
- View Article
- PubMed/NCBI
- Google Scholar
47. King JG, Hillyer JF. Spatial and temporal in vivo analysis of circulating and sessile immune cells in mosquitoes: hemocyte mitosis following infection. BMC Biol [Internet]. 2013 Dec 30;11(1):55. Available from: https://bmcbiol.biomedcentral.com/articles/10.1186/1741-7007-11-55. pmid:23631603
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Münch D, Kreibich CD, Amdam G v. Aging and its modulation in a long-lived worker caste of the honey bee. Journal of Experimental Biology. 2013 May;216(9):1638–49.
View Article
Google Scholar

[2] View Article

[3] Google Scholar

[ref2] 2. Kunc M, Dobeš P, Hurychová J, Vojtek L, Poiani SB, Danihlík J, et al. The year of the honey bee (Apis mellifera L.) with respect to its physiology and immunity: A search for biochemical markers of longevity. Insects. 2019 Aug 1;10(8). pmid:31394797
View Article
PubMed/NCBI
Google Scholar

[5] View Article

[6] PubMed/NCBI

[7] Google Scholar

[ref3] 3. Dostálková S, Dobeš P, Kunc M, Hurychová J, Škrabišová M, Petřivalský M, et al. Winter honeybee (Apis mellifera) populations show greater potential to induce immune responses than summer populations after immune stimuli. Journal of Experimental Biology. 2021 Feb 1;224(3). pmid:33288532
View Article
PubMed/NCBI
Google Scholar

[9] View Article

[10] PubMed/NCBI

[11] Google Scholar

[ref4] 4. Danihlík J, Aronstein K, Petřivalský M. Antimicrobial peptides: A key component of honey bee innate immunity. Vol. 54, Journal of Apicultural Research. Taylor and Francis Ltd; 2015. p. 123–36.

[ref5] 5. Simone-Finstrom M. Social Immunity and the Superorganism: Behavioral Defenses Protecting Honey Bee Colonies from Pathogens and Parasites. Bee World. 2017 Jan 2;94(1):21–9.
View Article
Google Scholar

[14] View Article

[15] Google Scholar

[ref6] 6. Fowler AE, Irwin RE, Adler LS. Parasite defense mechanisms in bees: behavior, immunity, antimicrobials, and symbionts. Emerg Top Life Sci. 2020 Jul 2;4(1):59–76. pmid:32558901
View Article
PubMed/NCBI
Google Scholar

[17] View Article

[18] PubMed/NCBI

[19] Google Scholar

[ref7] 7. Traynor KS, Mondet F, de Miranda JR, Techer M, Kowallik V, Oddie MAY, et al. Varroa destructor: A Complex Parasite, Crippling Honey Bees Worldwide. Vol. 36, Trends in Parasitology. Elsevier Ltd; 2020. p. 592–606.

[ref8] 8. Bruce Krejčí A, Votýpková K, Lukeš J, Votýpka J. Varroa destructor. Vol. 39, Trends in Parasitology. Elsevier Ltd; 2023. p. 487–8.

[ref9] 9. Kuster RD, Boncristiani HF, Rueppell O. Immunogene and viral transcript dynamics during parasitic Varroa destructor mite infection of developing honey bee (Apis mellifera) pupae. Journal of Experimental Biology. 2014 May 15;217(10):1710–8. pmid:24829325
View Article
PubMed/NCBI
Google Scholar

[23] View Article

[24] PubMed/NCBI

[25] Google Scholar

[ref10] 10. Gliński Z, Jarosz J. Varroa jacobsoni as a carrier of bacterial infections to a recipient bee host. Apidologie. 1992;23(1):25–31.
View Article
Google Scholar

[27] View Article

[28] Google Scholar

[ref11] 11. Kanbar G, Engels W. Ultrastructure and bacterial infection of wounds in honey bee (Apis mellifera) pupae punctured by Varroa mites. Parasitol Res. 2003 Aug 1;90(5):349–54. pmid:12684884
View Article
PubMed/NCBI
Google Scholar

[30] View Article

[31] PubMed/NCBI

[32] Google Scholar

[ref12] 12. Ramsey SD, Ochoa R, Bauchan G, Gulbronson C, Mowery JD, Cohen A, et al. Varroa destructor feeds primarily on honey bee fat body tissue and not hemolymph. Proceedings of the National Academy of Sciences. 2019 Jan 29;116(5):1792–801.
View Article
Google Scholar

[34] View Article

[35] Google Scholar

[ref13] 13. Burritt NL, Foss NJ, Neeno-Eckwall EC, Church JO, Hilger AM, Hildebrand JA, et al. Sepsis and hemocyte loss in honey bees (Apis mellifera) Infected with serratia marcescens strain sicaria. PLoS One. 2016 Dec 1;11(12). pmid:28002470
View Article
PubMed/NCBI
Google Scholar

[37] View Article

[38] PubMed/NCBI

[39] Google Scholar

[ref14] 14. El Sanousi S.M., El Sarag M.S.A., Sumaia E. Mohamed. Properties of Serratia marcescens Isolated from Diseased Honeybee (Apis mellifera) Larvae. Vol. 133, Journal of General Microbiology. 1987.
View Article
Google Scholar

[41] View Article

[42] Google Scholar

[ref15] 15. Fünfhaus A, Ebeling J, Genersch E. Bacterial pathogens of bees. Curr Opin Insect Sci. 2018 Apr;26:89–96. pmid:29764667
View Article
PubMed/NCBI
Google Scholar

[44] View Article

[45] PubMed/NCBI

[46] Google Scholar

[ref16] 16. Raymann K, Moran NA. The role of the gut microbiome in health and disease of adult honey bee workers. Curr Opin Insect Sci. 2018 Apr;26:97–104. pmid:29764668
View Article
PubMed/NCBI
Google Scholar

[48] View Article

[49] PubMed/NCBI

[50] Google Scholar

[ref17] 17. Al Naggar Y, Singavarapu B, Paxton RJ, Wubet T. Bees under interactive stressors: the novel insecticides flupyradifurone and sulfoxaflor along with the fungicide azoxystrobin disrupt the gut microbiota of honey bees and increase opportunistic bacterial pathogens. Science of The Total Environment. 2022 Nov;849:157941. pmid:35952893
View Article
PubMed/NCBI
Google Scholar

[52] View Article

[53] PubMed/NCBI

[54] Google Scholar

[ref18] 18. Al Naggar Y, Wubet T. Chronic exposure to pesticides disrupts the bacterial and fungal co-existence and the cross-kingdom network characteristics of honey bee gut microbiome. Science of The Total Environment. 2024 Jan;906:167530. pmid:37832690
View Article
PubMed/NCBI
Google Scholar

[56] View Article

[57] PubMed/NCBI

[58] Google Scholar

[ref19] 19. Raymann K, Shaffer Z, Moran NA. Antibiotic exposure perturbs the gut microbiota and elevates mortality in honeybees. PLoS Biol. 2017 Mar 14;15(3):e2001861. pmid:28291793
View Article
PubMed/NCBI
Google Scholar

[60] View Article

[61] PubMed/NCBI

[62] Google Scholar

[ref20] 20. Braglia C, Alberoni D, Garrido PM, Porrini MP, Baffoni L, Scott D, et al. Vairimorpha (Nosema) ceranae can promote Serratia development in honeybee gut: an underrated threat for bees? Front Cell Infect Microbiol. 2024 May 13;14. pmid:38808063
View Article
PubMed/NCBI
Google Scholar

[64] View Article

[65] PubMed/NCBI

[66] Google Scholar

[ref21] 21. Raymann K, Coon KL, Shaffer Z, Salisbury S, Moran NA. Pathogenicity of Serratia marcescens Strains in Honey Bees. McFall-Ngai MJ, editor. mBio [Internet]. 2018 Nov 7;9(5). Available from: https://journals.asm.org/doi/10.1128/mBio.01649-18
View Article
Google Scholar

[68] View Article

[69] Google Scholar

[ref22] 22. Laughton AM, Boots M, Siva-Jothy MT. The ontogeny of immunity in the honey bee, Apis mellifera L. following an immune challenge. J Insect Physiol. 2011 Jul;57(7):1023–32. pmid:21570403
View Article
PubMed/NCBI
Google Scholar

[71] View Article

[72] PubMed/NCBI

[73] Google Scholar

[ref23] 23. Chan QWT, Melathopoulos AP, Pernal SF, Foster LJ. The innate immune and systemic response in honey bees to a bacterial pathogen, Paenibacillus larvae. BMC Genomics. 2009 Aug 21;10. pmid:19695106
View Article
PubMed/NCBI
Google Scholar

[75] View Article

[76] PubMed/NCBI

[77] Google Scholar

[ref24] 24. Randolt K, Gimple O, Geissendörfer J, Reinders J, Prusko C, Mueller MJ, et al. Immune-related proteins induced in the hemolymph after aseptic and septic injury differ in honey bee worker larvae and adults. Arch Insect Biochem Physiol. 2008 Dec;69(4):155–67. pmid:18979500
View Article
PubMed/NCBI
Google Scholar

[79] View Article

[80] PubMed/NCBI

[81] Google Scholar

[ref25] 25. van de Schoot R, Depaoli S, King R, Kramer B, Märtens K, Tadesse MG, et al. Bayesian statistics and modelling. Nature Reviews Methods Primers. 2021 Jan 14;1(1):1.
View Article
Google Scholar

[83] View Article

[84] Google Scholar

[ref26] 26. Kunc M, Dobeš P, Ward R, Lee S, Čegan R, Dostálková S, et al. Omics-based analysis of honey bee (Apis mellifera) response to Varroa sp. parasitisation and associated factors reveals changes impairing winter bee generation. Insect Biochem Mol Biol. 2023 Jan;152:103877. pmid:36403678
View Article
PubMed/NCBI
Google Scholar

[86] View Article

[87] PubMed/NCBI

[88] Google Scholar

[ref27] 27. Evans JD, Schwarz RS, Chen YP, Budge G, Cornman RS, De la Rua P, et al. Standard methods for molecular research in Apis mellifera. J Apic Res [Internet]. 2013 Jan 2;52(4):1–54. Available from: https://www.tandfonline.com/doi/full/10.3896/IBRA.1.52.4.11
View Article
Google Scholar

[90] View Article

[91] Google Scholar

[ref28] 28. Pfaffl MW. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res. 2001 May 1;29(9):45e–45. pmid:11328886
View Article
PubMed/NCBI
Google Scholar

[93] View Article

[94] PubMed/NCBI

[95] Google Scholar

[ref29] 29. Danihlík J, Šebela M, Petřivalský M, Lenobel R. A sensitive quantification of the peptide apidaecin 1 isoforms in single bee tissues using a weak cation exchange pre-separation and nanocapillary liquid chromatography coupled with mass spectrometry. J Chromatogr A. 2014 Dec;1374:134–44. pmid:25435459
View Article
PubMed/NCBI
Google Scholar

[97] View Article

[98] PubMed/NCBI

[99] Google Scholar

[ref30] 30. Marringa WJ, Krueger MJ, Burritt NL, Burritt JB. Honey Bee Hemocyte Profiling by Flow Cytometry. PLoS One. 2014 Oct 6;9(10):e108486. pmid:25285798
View Article
PubMed/NCBI
Google Scholar

[101] View Article

[102] PubMed/NCBI

[103] Google Scholar

[ref31] 31. Salvatier J, Wiecki T v., Fonnesbeck C. Probabilistic programming in Python using PyMC3. PeerJ Comput Sci. 2016 Apr 6;2:e55.
View Article
Google Scholar

[105] View Article

[106] Google Scholar

[ref32] 32. Hoffman MD, Gelman A. The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian Monte Carlo. 2011 Nov 17; http://arxiv.org/abs/1111.4246

[ref33] 33. Gätschenberger H, Azzami K, Tautz J, Beier H. Antibacterial Immune Competence of Honey Bees (Apis mellifera) Is Adapted to Different Life Stages and Environmental Risks. PLoS One. 2013 Jun 17;8(6). pmid:23799099
View Article
PubMed/NCBI
Google Scholar

[109] View Article

[110] PubMed/NCBI

[111] Google Scholar

[ref34] 34. Casteels P, Ampe C, Jacobs F, Vaeck M, Tempst P. Apidaecins: antibacterial peptides from honeybees. Vol. 8, The EMBO Journal. 1989. pmid:2676519
View Article
PubMed/NCBI
Google Scholar

[113] View Article

[114] PubMed/NCBI

[115] Google Scholar

[ref35] 35. Casteels P, Ampe C, Riviere L, van Damme J, Elicone C, Fleming M, et al. Isolation and characterization of abaecin, a major antibacterial response peptide in the honeybee (Apis mellifera). Eur J Biochem. 1990;187(2):381–6. pmid:2298215
View Article
PubMed/NCBI
Google Scholar

[117] View Article

[118] PubMed/NCBI

[119] Google Scholar

[ref36] 36. Casteels P, Ampe C, Jacobs F, Tempst P. Functional and chemical characterization of hymenoptaecin, an antibacterial polypeptide that is infection-inducible in the honeybee (Apis mellifera). Journal of Biological Chemistry. 1993;268(10):7044–54. pmid:8463238
View Article
PubMed/NCBI
Google Scholar

[121] View Article

[122] PubMed/NCBI

[123] Google Scholar

[ref37] 37. Casteels-Josson K, Zhang W, Capaci T, Casteels P, Tempst P. Acute transcriptional response of the honeybee peptide-antibiotics gene repertoire and required post-translational conversion of the precursor structures. Journal of Biological Chemistry. 1994 Nov 18;269(46):28569–75. pmid:7961803
View Article
PubMed/NCBI
Google Scholar

[125] View Article

[126] PubMed/NCBI

[127] Google Scholar

[ref38] 38. Lourenço AP, Guidugli-Lazzarini KR, Freitas FCP, Bitondi MMG, Simões ZLP. Bacterial infection activates the immune system response and dysregulates microRNA expression in honey bees. Insect Biochem Mol Biol. 2013 May;43(5):474–82. pmid:23499934
View Article
PubMed/NCBI
Google Scholar

[129] View Article

[130] PubMed/NCBI

[131] Google Scholar

[ref39] 39. Evans JD, Aronstein K, Chen YP, Hetru C, Imler JL, Jiang H, et al. Immune pathways and defence mechanisms in honey bees Apis mellifera. Insect Molecular Biology. 2006. pmid:17069638
View Article
PubMed/NCBI
Google Scholar

[133] View Article

[134] PubMed/NCBI

[135] Google Scholar

[ref40] 40. Zhao Y, Heerman M, Peng W, Evans JD, Rose R, Degrandi-Hoffman G, et al. The dynamics of Deformed wing virus concentration and host defensive gene expression after varroa mite parasitism in honey bees, Apis mellifera. Insects. 2019 Jan 1;10(1). pmid:30626033
View Article
PubMed/NCBI
Google Scholar

[137] View Article

[138] PubMed/NCBI

[139] Google Scholar

[ref41] 41. Casteels-Josson K, Capaci T, Casteels P, Tempst P. Apidaecin multipeptide precursor structure: a putative mechanism for amplification of the insect antibacterial response. Vol. 12, The EMBO Journal. 1993. pmid:8467807
View Article
PubMed/NCBI
Google Scholar

[141] View Article

[142] PubMed/NCBI

[143] Google Scholar

[ref42] 42. Lourenço AP, Martins JR, Torres FAS, Mackert A, Aguiar LR, Hartfelder K, et al. Immunosenescence in honey bees (Apis mellifera L.) is caused by intrinsic senescence and behavioral physiology. Exp Gerontol. 2019 May 1;119:174–83. pmid:30742904
View Article
PubMed/NCBI
Google Scholar

[145] View Article

[146] PubMed/NCBI

[147] Google Scholar

[ref43] 43. Richard FJ, Holt HL, Grozinger CM. Effects of immunostimulation on social behavior, chemical communication and genome-wide gene expression in honey bee workers (Apis mellifera). BMC Genomics. 2012 Oct 16;13(1). pmid:23072398
View Article
PubMed/NCBI
Google Scholar

[149] View Article

[150] PubMed/NCBI

[151] Google Scholar

[ref44] 44. Schlüns H, Crozier RH. Relish regulates expression of antimicrobial peptide genes in the honeybee, Apis mellifera, shown by RNA interference. Insect Mol Biol. 2007 Dec;16(6):753–9. pmid:18093004
View Article
PubMed/NCBI
Google Scholar

[153] View Article

[154] PubMed/NCBI

[155] Google Scholar

[ref45] 45. Azzami K, Ritter W, Tautz J, Beier H. Infection of honey bees with acute bee paralysis virus does not trigger humoral or cellular immune responses. Arch Virol. 2012 Apr;157(4):689–702. pmid:22258854
View Article
PubMed/NCBI
Google Scholar

[157] View Article

[158] PubMed/NCBI

[159] Google Scholar

[ref46] 46. Honti V, Csordás G, Márkus R, Kurucz É, Jankovics F, Andó I. Cell lineage tracing reveals the plasticity of the hemocyte lineages and of the hematopoietic compartments in Drosophila melanogaster. Mol Immunol. 2010 Jul;47(11–12):1997–2004. pmid:20483458
View Article
PubMed/NCBI
Google Scholar

[161] View Article

[162] PubMed/NCBI

[163] Google Scholar

[ref47] 47. King JG, Hillyer JF. Spatial and temporal in vivo analysis of circulating and sessile immune cells in mosquitoes: hemocyte mitosis following infection. BMC Biol [Internet]. 2013 Dec 30;11(1):55. Available from: https://bmcbiol.biomedcentral.com/articles/10.1186/1741-7007-11-55. pmid:23631603
View Article
PubMed/NCBI
Google Scholar

[165] View Article

[166] PubMed/NCBI

[167] Google Scholar

Figures

Abstract

Introduction

Material and methods

Experimental honey bees

Honey bees treatment

Samples collection

Determination of antimicrobial activity

Quantification of relative gene expression

Quantification of Apidaecin 1 in the hemolymph by LC-MS analysis

Quantification of Abaecin, Defensin 1, and Hymenoptaecin by ELISA

Hemocytes quantification

Statistics

Results

Antimicrobial activity

Relative gene expression and concentration of selected antimicrobial peptides

Concentration of circulating hemocytes

Discussion

Supporting information

S1 File. ELISA buffers and reagents.

S2 File. Python script.

S3 File. Original data.

S1 Table. Primer sequences and the corresponding target gene names used for PCR reactions.

S2 Table. Values of the ampl and loc with 2.5–97.5% dispersion, i.e. 95% HDI.

S1 Fig. Bayesian model of antimicrobial activity dynamics measured as lysozyme equivalent (mg/ml).

S2 Fig. Bayesian model showing dynamics of abaecin relative gene expression and its relative peptide concentration in the hemolymph.

S3 Fig. Bayesian model showing dynamics of apidaecin 1 relative gene expression and its peptide concentration in the hemolymph.

S4 Fig. Bayesian model showing dynamics of defensin 1 relative gene expression and its relative peptide concentration in the hemolymph.

S5 Fig. Bayesian model showing dynamics of hymenoptaecin relative gene expression and its relative peptide concentration in the hemolymph.

S6 Fig. Bayesian model of hemocyte concentration (circulating hemocytes/μl of hemolymph) dynamics measured by flow cytometry.

Acknowledgments

References