Identification of Immune Regulatory Genes in Apis mellifera through Caffeine Treatment

Simple Summary The preference of honeybees to consume nectar with caffeine has been recorded. To investigate the effects of caffeine to this important pollinator, we first focus on the influences on immunity which is seldom explored in insects regarding caffeine. In our results, we discovered the suppressive effects on viral pathogens and the boosting effects on immunity after caffeine treatment. At least six different latent-infecting viruses in Taiwan were suppressed by caffeine. Nevertheless, the enhancement on immunity may not be effective if the bees have not been exposed to the environment or potential natural secondary metabolites like caffeine. These findings provide a basic but valuable insight into how caffeine can aid honeybees in fighting against viral invasion. Abstract Plants and pollinators are mutually beneficial: plants provide nectar as a food source and in return their pollen is disseminated by pollinators such as honeybees. Some plants secrete chemicals to deter herbivores as a protective measure, among which is caffeine, a naturally occurring, bitter tasting, and pharmacologically active secondary compound. It can be found in low concentrations in the nectars of some plants and as such, when pollinators consume nectar, they also take in small amounts of caffeine. Whilst caffeine has been indicated as an antioxidant in both mammals and insects, the effect on insect immunity is unclear. In the present study, honeybees were treated with caffeine and the expression profiles of genes involved in immune responses were measured to evaluate the influence of caffeine on immunity. In addition, honeybees were infected with deformed wing virus (DWV) to study how caffeine affects their response against pathogens. Our results showed that caffeine can increase the expression of genes involved in immunity and reduce virus copy numbers, indicating that it has the potential to help honeybees fight against viral infection. The present study provides a valuable insight into the mechanism by which honeybees react to biotic stress and how caffeine can serve as a positive contributor, thus having a potential application in beekeeping.


Introduction
The western honeybee, Apis mellifera, is one of the most important pollinators in the world [1,2]. They contribute to the production of many different fruits, vegetables, crops, and aromatic seeds. However, honeybees are weak at resisting their stressors, which can cause direct/indirect injuries and Western honeybees (Apis mellifera) were purchased in Hsinchu county, Taiwan, and kept at the National Taiwan University. About 50~100 bees for the experiment were collected directly from one hive as a mixed group. Then, the bees were caged in different Bugdorms (18 × 18 × 18 cm, MegaView Science, Taichung, Taiwan) at 37 • C. Prior to the experiments, bees were fed with 0.7 M sucrose water with/without caffeine (0.1 mM, Sigma-Aldrich, St. Louis, MO, USA) for one week [29,35]. One frame of the hive was taken back to the lab and incubated in 30 • C, waiting for newly emerged bees to come out. The newly emerged bees were first fed with 0.7 M sucrose water and pollen as protein source for one week to stabilize their physiological condition, then caffeine (0.1 mM) was added to the diet for Insects 2020, 11, 516 3 of 12 another week. All food for the bees were changed every three days, and the dead bees were removed. After this feeding stage, all bees were collected for gene analysis or bioassays.

Deformed Wing Virus (DWV) Purification
About 80~100 bees were directly collected from their hive, and then put into cages, incubated in 30 • C. DWVs were mixed with sugar solution (0.7 M) as a regular treatment for one week. After treatment, bees were frozen in −80 • C, and then homogenized with 10 mL phosphate-buffered saline (PBS). The liquid was collected through a nylon filter to keep out residue, then we centrifuged it (16,000× g). Supernatant was moved to another tube and centrifuge again, then use Minisart ® Syringe Filters (0.45 µm) to filtrate the supernatant.

DWV Infection and Caffeine Treatment
Four treatment groups were prepared: control (0.7 M sucrose water only), caffeine only (0.1 mM) [29], DWV only (10 6 virus copies), and both (caffeine/DWV). DWV was diluted with phosphate-buffered saline (PBS) to prepare a working virus solution with 5 × 10 6 copies/µL. The sucrose water was replaced with water only one day prior to experiment and the following day the bees were forced to take 4 µL of solution mixed with 2 µL of virus and 2 µL of sucrose. After 48 h, the bees were frozen at −80 • C for total RNA extraction and gene expression analysis. After one week stabilizing the physiological condition, the bees were separated into two groups, and caffeine was then added to the diet of one of the groups for another week as the caffeine-only treatment group. Then, bees were randomly chosen from each group and forced to take 4 µL of solution mixed with 2 µL of virus and 2 µL of sucrose as virus only and both treatment group. After this feeding stage, all bees were collected for gene analysis or bioassays.

Total RNA Extraction
After the treatments, total RNA was extracted using a TRIzol™ reagent kit (Thermo Fisher Scientific). Whole bodies of two bees were pooled together for homogenization. RNA quantification was determined using a NanoDrop 2000 (Thermo Fisher Scientific).

cDNA Synthesis
cDNA synthesis of each treatment group was performed using a High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems). For each sample, 2 µg of total RNA was used. The reaction was incubated in a FlexCycler 2 PCR thermal cycler (Analytic Jena AG) for 10 min at 25 • C, 120 min at 37 • C, 5 min at 85 • C, and then stopped at 4 • C. The final products were used for quantitative real-time polymerase chain reaction (RT-PCR) analysis or stored at −20 • C for later analysis.

Real-Time Polymerase Chain Reaction (PCR) and Data Analysis
Real-time PCR was performed using a StepOnePlus™ Real-Time PCR System (Applied Biosystems) using SensiFAST™ SYBR Hi-ROX Kit (BIOLINE, London, UK). The result (fold change) was calculated following the relative quantification theory [36]. For quantitative PCR, honeybee-specific primers for including immune, viral and carbohydrate metabolism genes (Table 1) were used as described in previous studies [35,37,38]. All samples were amplified simultaneously, and three independent experiments were performed. PCR-array images were analyzed with the software R (Version XX). Fold changes were calculated using the relative quantification method (2 − Ct ) [39]. Each group of tested genes was normalized to a reference gene (18s rRNA), and fold changes in the control group were used as a calibrator. Table 1. List of real-time polymerase chain reaction (RT-PCR) primers and target genes.

DWV Titer Calculation
A plasmid containing a DNA fragment of DWV was transformed to Escherichia coli (DH5α) and then grown in LB broth (ARROWTEC) containing ampicillin as a selection marker at 37 • C for 16 h. Plasmid extraction was conducted using a Presto™ Mini Plasmid Kit (Geneaid) and the concentration was determined using a NanoDrop 2000 (Thermo Fisher Scientific). A DNA Copy Number and Dilution Calculator (Thermo Fisher scientific website) were used to determine the amount of DNA sample equivalent to 10 10 plasmid copies. Serial dilution was performed to prepare 10 10 to 10 1 plasmid copies. Real-time PCR was conducted using serially diluted plasmid samples. A standard curve was plotted using data obtained from the real-time PCR results on serially-diluted plasmid samples; regression analysis was used to calculate the virus copy numbers in the virus-treated bees.

Other Latent Infecting Viruses in Taiwan
Based on a previous study in Taiwan [35], we choose the also latent-infecting virus species to see if caffeine can aid the bees to resist them. The virus we use including Vorroa destructor virus-1 (VDV), Kashmir bee virus (KBV), Kakugo virus (KV), Israeli acute bee paralysis virus (IAPV), Sacbrood Virus Insects 2020, 11, 516 5 of 12 (SBV), black queen cell virus (BQCV), and Chronic bee paralysis virus (CBPV) to see whether caffeine can also suppress their replication or not. After treated with caffeine, honeybees were freeze-killed and the mRNA extraction, complimentary DNA (cDNA) reverse transcription, and RT-PCR was performed to compare the amount of viral infection of honeybees treated with/without caffeine.

Statistical Analysis
Gene expression level results were analyzed using one-way analysis of variance (ANOVA) followed by Tukey's post hoc test for significance using Statistica software (Version 8). The virus number was analyzed using one-way Student's t-test. A p < 0.05 indicated a statistically significant result.

Effects of Caffeine on Immunity Gene Signaling Factors and Anti-Microbial Peptides
The expression of immune response-related genes after DWV infection (Table 2), especially those involved in the Toll and Imd pathways, as well as anti-microbial peptide (AMP), were analyzed 2 days after infection. It was found that both caffeine alone (caffeine group) and DWV infection alone (DWV group) could stimulate gene expression compared with the control group, with DWV being a stronger stimulant than caffeine ( Figure 1A-C). However, the highest stimulation was observed in DWV-infected bees receiving the caffeine diet prior to infection (caffeine/DWV group), which exhibited up to 5-fold stimulation in gene expression ( Figure 1A-C and Table 2). Genes that exhibited more than a 1-fold increase in the DWV and caffeine/DWV groups include parseph and Myd88 in the Toll signaling pathway, PGRPLS, Tak1, Kenny, and Tab in the Imd pathway, and prophenoloxidase-activating enzyme (PPOact), Lysozyme-1 (Lys-1), and AMP defensin-2 ( Figure 1D). Stimulation of these factors may provide a strong immune response against pathogen infection.

Other Latent Infecting Viruses in Taiwan
Based on a previous study in Taiwan [35], we choose the also latent-infecting virus species to see if caffeine can aid the bees to resist them. The virus we use including Vorroa destructor virus-1 (VDV), Kashmir bee virus (KBV), Kakugo virus (KV), Israeli acute bee paralysis virus (IAPV), Sacbrood Virus (SBV), black queen cell virus (BQCV), and Chronic bee paralysis virus (CBPV) to see whether caffeine can also suppress their replication or not. After treated with caffeine, honeybees were freeze-killed and the mRNA extraction, complimentary DNA (cDNA) reverse transcription, and RT-PCR was performed to compare the amount of viral infection of honeybees treated with/without caffeine.

Statistical Analysis
Gene expression level results were analyzed using one-way analysis of variance (ANOVA) followed by Tukey's post hoc test for significance using Statistica software (Version 8). The virus number was analyzed using one-way Student's t-test. A p < 0.05 indicated a statistically significant result.

Effects of Caffeine on Immunity Gene Signaling Factors and Anti-Microbial Peptides
The expression of immune response-related genes after DWV infection (Table 2), especially those involved in the Toll and Imd pathways, as well as anti-microbial peptide (AMP), were analyzed 2 days after infection. It was found that both caffeine alone (caffeine group) and DWV infection alone (DWV group) could stimulate gene expression compared with the control group, with DWV being a stronger stimulant than caffeine ( Figure 1A-C). However, the highest stimulation was observed in DWV-infected bees receiving the caffeine diet prior to infection (caffeine/DWV group), which exhibited up to 5-fold stimulation in gene expression ( Figure 1A-C and Table 2). Genes that exhibited more than a 1-fold increase in the DWV and caffeine/DWV groups include parseph and Myd88 in the Toll signaling pathway, PGRPLS, Tak1, Kenny, and Tab in the Imd pathway, and prophenoloxidaseactivating enzyme (PPOact), Lysozyme-1 (Lys-1), and AMP defensin-2 ( Figure 1D). Stimulation of these factors may provide a strong immune response against pathogen infection.

Caffeine Inhibits DWV Replication by Enhancing Carbohydrate Metabolism
To further investigate how caffeine stimulates the expression of immune response-related genes, the replication of DWV in infected bees with or without the caffeine diet was analyzed. It was found that the DWV genome copy number was significantly reduced in bees receiving the caffeine diet (Figure 2A), suggesting that caffeine induced the expression of immune response related genes, Insects 2020, 11, 516 7 of 12 which in turn suppressed DWV genome replication. It is known that an optimal immune response requires a significant amount of energy [40]; therefore, we hypothesized that caffeine stimulates the expression of immune response-related genes by imposing positive effects on pathways relating to carbohydrate metabolism. To test this hypothesis, we analyzed the expression of genes involved in glycolysis and the tricarboxylic acid cycle (TCA) cycle in bees with or without the caffeine diet by real-time PCR (Table 3). Although the result had no significant difference, but still we can see the result showed the expression of genes involved in glycolysis ( Figure 2B) and the TCA cycle ( Figure 2C) has an up-regulated trend.

Caffeine also Inhibits the Infection of Other Prevalent Viruses in Taiwan
More than 8 persistent infectious viruses are prevalent among bees in Taiwan. These infected honeybees were given caffeine in their diet to evaluate its effect on viral activity in the host. Quantification of viral gene expression by real-time PCR showed a significant decrease in 6 of the viral genes in the caffeine-treated bees except for Black queen cell virus (BQCV) and Varroa destructor virus (VDV) (Figure 3). Thus, BQCV/VDV infection might not induce a noticeable immune response in honeybees.

Caffeine also Inhibits the Infection of Other Prevalent Viruses in Taiwan
More than 8 persistent infectious viruses are prevalent among bees in Taiwan. These infected honeybees were given caffeine in their diet to evaluate its effect on viral activity in the host. Quantification of viral gene expression by real-time PCR showed a significant decrease in 6 of the viral genes in the caffeine-treated bees except for Black queen cell virus (BQCV) and Varroa destructor virus (VDV) (Figure 3). Thus, BQCV/VDV infection might not induce a noticeable immune response in honeybees. CBPV: chronic bee paralysis virus. The data are presented as mean ± SD, n = 6. Values from the untreated groups were set to 1 and the values from caffeine-treated group were adjusted accordingly. Statistical analysis was performed using the Mann-Whitney U-test, * p < 0.05 relative to data collected from the control group.

Caffeine Does not Affect the Expression of Immunity Related Genes in 16-Days Old Honeybee
The results presented thus far were obtained from a mixture of bees from different tasks and ages. However, we also explored whether caffeine has a similar effect on 16-days old bees. For this experiment, newly-emerged bees received caffeine in their diet for 7 days and were subsequently fed with DWV. The expression of genes involved in the Toll and Imd pathways and AMPs were analyzed 2 days post-infection. It was found that the expression of genes involved in the Toll and Imd pathways were down-regulated in the DWV, caffeine, and caffeine/DWV groups ( Figure 4A,B). The expression of many AMPs was also down-regulated in the DWV infection and caffeine only groups ( Figure 4C). However, for several genes whose expressions were suppressed by DWV infection (including pgrps2 and cactus1 in the Toll signaling pathway, kenny in the Imd pathway, and the AMPs abaecin and defensin-2, and prophenoloxidase-activating enzyme (PPOact), and Lysozyme-1 (Lys-1)), the inhibitory effect could still be prevented significantly by a caffeine diet prior to infection ( Figure  4D). These results indicate that for 16-day-old bees, caffeine (to which they have had no previous exposure) may have negative regulatory effects on immune response-related gene expression. This, in turn, affects the overall physiological condition of the insects in a similar manner to if the insects were exposed to an adverse environment.

Discussion
In this study, the influence of caffeine on the immune system of honeybees was evaluated. Gene expression levels were measured to demonstrate the interaction between caffeine and DWV. Although the effect of caffeine on mammals is well known [41], particularly its effects on antioxidation and neural activation, there are only a limited number of similar studies on insects. The results of the present study provide a basic yet valuable insight into the effect of caffeine on gene regulation in insects. The data are presented as mean ± SD, n = 6. Values from the untreated groups were set to 1 and the values from caffeine-treated group were adjusted accordingly. Statistical analysis was performed using the Mann-Whitney U-test, * p < 0.05 relative to data collected from the control group.

Caffeine Does not Affect the Expression of Immunity Related Genes in 16-Days Old Honeybee
The results presented thus far were obtained from a mixture of bees from different tasks and ages. However, we also explored whether caffeine has a similar effect on 16-days old bees. For this experiment, newly-emerged bees received caffeine in their diet for 7 days and were subsequently fed with DWV. The expression of genes involved in the Toll and Imd pathways and AMPs were analyzed 2 days post-infection. It was found that the expression of genes involved in the Toll and Imd pathways were down-regulated in the DWV, caffeine, and caffeine/DWV groups ( Figure 4A,B). The expression of many AMPs was also down-regulated in the DWV infection and caffeine only groups ( Figure 4C). However, for several genes whose expressions were suppressed by DWV infection (including pgrps2 and cactus1 in the Toll signaling pathway, kenny in the Imd pathway, and the AMPs abaecin and defensin-2, and prophenoloxidase-activating enzyme (PPOact), and Lysozyme-1 (Lys-1)), the inhibitory effect could still be prevented significantly by a caffeine diet prior to infection ( Figure 4D). These results indicate that for 16-day-old bees, caffeine (to which they have had no previous exposure) may have negative regulatory effects on immune response-related gene expression. This, in turn, affects the overall physiological condition of the insects in a similar manner to if the insects were exposed to an adverse environment.
conditions from mixed nursing bees/forager bees and 16-day-old bees. Nevertheless, the expression of PPOact, Lys-1 and Kenny were also significantly enhanced in DWV-infected bees, similar to in the caffeine/DWV group (Figure 4), suggesting that caffeine has a marginal boosting effect on the immune system of 16-day-old bees. The differential influences of caffeine on the immune system during pathogen infection in 16-day-old bees and older nursing bees/forager bees thus require further study.

Discussion
In this study, the influence of caffeine on the immune system of honeybees was evaluated. Gene expression levels were measured to demonstrate the interaction between caffeine and DWV. Although the effect of caffeine on mammals is well known [41], particularly its effects on anti-oxidation and neural activation, there are only a limited number of similar studies on insects. The results of the present study provide a basic yet valuable insight into the effect of caffeine on gene regulation in insects.
The effect of caffeine on the immune system of honeybees is almost unknown. In humans, a high concentration of caffeine can mitigate the damage caused by inflammation [42]. This is due to the inhibition on phosphodiesterase (PDE) activity, leading to an increased level of intracellular cAMP and the activation of the PKA pathway [43]. This may explain the changes in gene expression found in the present study, since the PKA pathway is also involved in regulating the immune system [44]. Analysis of the expression level of immune genes showed that they were significantly up-regulated after caffeine treatment in DWV-infected bees ( Figure 1A-C), namely for parseph and myd88 in the Toll pathway, PGRPLC, tak1, kenny and tab in the Imd pathway, and lysozyme-1 (Lys-1), prophenoloxidase-activating enzyme (PPOact), and the AMP gene amPPO ( Figure 1D). Both the Toll and Imd pathways have been previously demonstrated to be involved in fighting against viral invasion in insects [45,46]. Our results also showed that caffeine treatment prior to DWV infection could sufficiently inhibit DWV infection ( Figure 2). This indicates that caffeine can boost the immune system by up-regulating the expression of genes involved in pathways known to influence immune responses (such as Toll and Imd) and protect honeybees from the external stress caused by viral infections. In the results, after DWV infection, the gene expression level of the immune system is up-regulated. Although it is indicated in previous study that DWV can suppress the host's immune system, this is the consequence of long-term and latent infection of Vorroa destructor and DWV [16]. In our experiment, the virus was fed to the healthy bees, so this situation is different from the previous study. This acute infection induced the spike in the immune response of honeybees against the viral invasion [47].
Until now, the effect of caffeine on gene expression in 16-day-old bees has remained unexplored. Caffeine is a natural compound present in the nectar of certain plants and, therefore, honeybees can easily consume caffeine from the environment [25,29]. Bees that are 16 days old, however, were not exposed to caffeine (and other compounds) as the foragers because they were kept in lab. The effect of a stressor and caffeine treatment was, therefore, worth investigating for these honeybees with "clean" background. Interestingly, the results are contrary to those obtained in the mixed group: caffeine does not stimulate the immune system of 16-day-old bees as it does to the mixed group in the presence of DWV infection and the expression of most genes involved in the Toll and Imd pathways and AMPs are down-regulated ( Figure 4A-C).
One previous study indicated that older honeybees, such as foragers, have a high basal gene expression level related to detoxification and immune pathways for dealing with more environmental stressors [48]. This may explain different outcomes under the same experimental conditions from mixed nursing bees/forager bees and 16-day-old bees. Nevertheless, the expression of PPOact, Lys-1 and Kenny were also significantly enhanced in DWV-infected bees, similar to in the caffeine/DWV group (Figure 4), suggesting that caffeine has a marginal boosting effect on the immune system of 16-day-old bees. The differential influences of caffeine on the immune system during pathogen infection in 16-day-old bees and older nursing bees/forager bees thus require further study.

Conclusions
In summary, caffeine has the potential to improve the resistance of honeybees to viral infection. For 16-day-old bees, caffeine has a similar effect but it is not as extensive and positive as within the mixed group. This study provides a basic knowledge of the influence of caffeine on the expression profiles of immune-related genes of honeybees. Even though the mechanism by which caffeine regulates the immune system in honeybees is not yet clear, future experiments on its effects on responses to other pathogens such as bacteria and fungi should be undertaken. This would help define a more definitive role of caffeine in promoting the immune responses of honeybees against pathogens.