Habitat loss, monocultures, and changes in plant flowering phenology are considered the major contributors to poor honey bee nutrition and subsequent honey bee declines (Kremen et al. 2002, Naug 2009, Vanbergen et al. 2013, Otto et al. 2016). Research to date has demonstrated that adequate nutrition is critical for maintaining strong immunity and survival in honey bees. Honey bee colonies that consume sufficient amounts of high-quality pollen are less susceptible to the gut parasite Nosema ceranae, have lower pathogen loads, overwinter more successfully, exhibit enhanced immunocompetence, produce drones with better semen quality, and are able to better resist stressors such as parasites, diseases, and pesticides (Alaux et al. 2010, 2011, Brodschneider and Crailsheim 2010, Di Pasquale et al. 2013, Mao et al. 2013, Schmehl et al. 2014, Niño and Jasper 2015, Jack et al. 2016, McMenamin et al. 2016, 2018, Smart et al. 2016). Poor nutrition may alter the physiology of emergent spring workers (Mattila and Otis 2006) and may subsequently influence behavior. In one study, limited nutrition during larval stages produced adults that were inefficient foragers and waggle dancers (Scofield and Mattila 2015). Hence, optimal nutrition may be a colony's first line of defense, enabling it to withstand the effects of both biotic and abiotic stressors.

Pollen and nectar are the two principal sources of macronutrients for honey bees. Carbohydrate-rich nectar supplies bees with energy, while pollen serves as the primary source of proteins, lipids, vitamins, and vital micronutrients (viz. sterols; Brodschneider and Crailsheim 2010). Micronutrients are critical and yet are understudied in honey bees (Bonoan et al. 2018). Pollen is consumed by nurse bees, whose hypopharyngeal glands biosynthesize the proteinaceous secretions that are progressively provisioned to developing larvae (Knecht and Kaatz 1990, Crailsheim et al. 1992). Hence, understanding the effects of critical micronutrients on nurse bee physiology is important in understanding its long-term implications on colony health. Research on honey bee nutrition is still a poorly developed area (Somerville 2005, Bonoan et al. 2018). Even though bee nutrition has been studied for a significant amount of time, a substantial gap in knowledge exists regarding the physiological impacts of sterols in honey bees.

Sterols play a vital role in insect physiology. They are precursors for molting hormones, act as signaling molecules that influence development, and are critical for cell membrane development and function (Behmer and Nes 2003). All insects are sterol auxotrophs in that they are unable to synthesize sterols, thereby, depending on dietary sources for these important micronutrients (Carvalho et al. 2010). In honey bees, a specific sterol called 24-methylenecholesterol (24MC) has been reported to be the most critical for colony growth and worker longevity (Herbert et al. 1980), as the researchers observed higher survival of worker bees and greater brood production in 24MC diet treatments. However, Herbert et al. (1980) do not discuss proximate mechanisms resulting in higher survival and brood rearing. Indeed, concentrations of this sterol in pupae were higher than any other sterol (Feldlaufer 1986, Svoboda et al. 1986). Elevated concentrations of this sterol have also been attributed to the formation of ecdysteroids in honey bees (Svoboda et al. 1986). Like all dietary sterols, honey bees obtain it from pollen. Colonies used for crop pollination often face nutritional stress because the quality or quantity of pollen forage available to them in such agricultural landscapes is inadequate (Naug 2009). Hence, it is crucial to understand the nutritional physiology and impacts of this important sterol in honey bees.

In this study, we examined the effects of 24MC on honey bee physiology by measuring a few important physiological parameters. The dietary concentrations of sterols vary among different insect species, but the optimal concentration of sterols appears to be around 0.1% (Clayton 1964, Jing et al. 2013) for better fitness. For species whose sterol requirements are not known, significantly higher concentrations than 0.1% were chosen in some cholesterol-related studies, and diets with higher concentrations of cholesterol, such as 1%, 3%, or even 10%, enhanced insect growth and showed no deleterious effects (Clayton 1964). In addition, plant pollens contain varying concentrations of 24MC (Standifer et al. 1968, Villette et al. 2015, Chakrabarti et al. 2019). In a previous seminal study with honey bees, Herbert et al. (1980) used 0.1% 24MC to examine the effects of this sterol on brood rearing. This study did not provide justification for selecting this specific concentration of 24MC. We speculate that these researchers chose this concentration based on information available in the literature pertaining to sterol needs of other insects. Furthermore, this study only measured effects of 24MC on brood rearing and longevity of bees and did not examine effects on any physiological parameters. Our goal in the present study was to evaluate the effects of 24MC on honey bee physiology and longevity. In our study, artificial diets with varying concentrations of 24MC (ranging from 0 to 1.0%) were formulated and fed to honey bees in a laboratory cage experiment. We chose the concentrations of 24MC to include the concentrations used in Herbert et al. (1980) and the concentrations deemed to be optimal with respect to longevity, fitness, and growth based on studies with other insects (Clayton 1964, Jing et al. 2013). For each treatment group, we measured the consumption of diet and survival of bees. In addition, we measured abdominal fat content and head protein content as abdominal fat is a crucial indicator of fitness traits in insects, including bees (Amdam et al. 2003, Toth et al. 2005, Arrese and Soulages 2010) and the head capsule houses major brood food-producing glands in bees (Kucharski and Maleszka 1998). These two physiological parameters are relatively precise indicators and are directly related to the broad and general parameters (i.e., longevity and brood rearing) that were measured by Herbert et al. (1980). Our findings provide new insights regarding sterol nutritional physiology in honey bees.

Fig. 1.

Line diagram of the experimental cage set-up depicting how the artificial diets, sugar syrup, and water were administered to the honey bees in the laboratory.

Materials and Methods

Survival

Bee mortality in each cage was recorded at 2-d intervals and total mortality was calculated at the end of each week for each replicate cage in all treatment and control groups. Kaplan–Meier survival analysis was performed based on previous studies (Klein and Moeschberger 2003, Chakrabarti et al. 2015) using MedCalc Version 18.2.1 software. Survival curves were calculated for a period up to 21 d for the study.

Table 1.

Acetone volumes added to the control and treatment groups

Diet Consumption

Diet consumption in each experimental cage was recorded each week and was calculated as the change in patty weight from the beginning of each week when it was placed in the cage to the end of the week when it was removed from the cage (seven days later). Cages that contained artificial diets, sugar syrup, and water, but no bees, were also placed in the incubator and treated identical to cages with bees. The change in diet weight within these empty cages represented the effects of evaporation. Before calculating diet consumption, the mean weekly weight loss of diet patties in these ‘evaporation control’ cages was subtracted from the patty weight loss of each experimental cage.

The average diet consumption per bee per week was calculated as follows:

where I_x = initial weight (mg) of diet placed in cage ‘y' at the beginning of week ‘x'; F_x = final weight (mg) of diet in cage ‘y' at the end of week ‘x'; x′ = mean change in diet patty weight in evaporation control cages during week ‘x'; and N_y = average number of live honey bees in cage ‘y’ during week ‘x'.

Results

Diet Consumption

Consumption data from all three replicates of a given experimental group (diet treatment) were pooled together for each week. The results for all Tukey's post hoc tests for every week are provided in  Supp Table 1 (online only). There was no significant difference in diet consumption between the two control groups C0 (week 1: 7.66 ± 0.87 mg per bee; week 2: 5.84 ± 0.46 mg per bee; week 3: 9.07 ± 0.86 mg per bee) and C1 (week 1: 8.20 ± 0.58 mg per bee; week 2: 5.67 ± 0.47 mg per bee; week 3: 9.05 ± 1.63 mg per bee) for week 1 (t = –0.519, df = 4, P = 0.32), week 2 (t = 0.274, df = 4, P = 0.60), and week 3 (t = 0.0065, df = 4, P = 0.502).

Fig. 2.

Kaplan–Meier survival curves for the experimental groups S1–S5 when compared with controls C0 and C1.The experimental duration was 3 wk (21 d). C0 and C1: control groups without and with acetone, respectively; S1, S2, S3, S4, and S5: 0.1, 0.25, 0.5, 0.75, and 1.0% dry diet weights of 24-methylenecholesterol in artificial diets.

In week 1, a significant difference in consumption was observed between treatment groups (one-way ANOVA, F_6,14 = 19.52, P < 0.001). Tukey's post hoc test revealed no significant differences between groups C0, C1, and S1 and between groups S2, S3, S4, and S5 (Fig. 3;  Supp Table 1 [online only]). In week 1, the average consumption of the sterol diets were 8.47 ± 1.01, 12.81 ± 0.44, 12.61 ± 0.25, 13.00 ± 0.44, and 14.10 ± 0.31 mg per bee for experimental groups S1, S2, S3, S4, and S5, respectively (Fig. 3). Significant differences in consumption were also observed between treatment groups in week 2 (one-way ANOVA, F_6,14 = 3.854, P < 0.05; Fig. 3;  Supp Table 1 [online only]). The average diet consumptions were 7.02 ± 1.08, 7.18 ± 0.93, 7.95 ± 0.49, 8.97 ± 0.69, and 9.63 ± 0.93 mg per bee for S1, S2, S3, S4, and S5 groups, respectively (Fig. 3). Likewise, in week 3, significant differences in diet consumption were observed (one-way ANOVA, F_6,14 = 3.757, P < 0.05; Fig. 3;  Supp Table 1 [online only]). The average diet consumptions per bee during week 3 were 9.18 ± 0.30, 10.71 ± 0.80, 11.49 ± 0.052, 11.94 ± 0.44, and 13.87 ± 1.35 mg, respectively, in groups S1, S2, S3, S4, and S5 (Fig. 3).

Discussion

To our knowledge, this is the first comprehensive study to investigate the effects of 24MC (a critical micronutrient) on honey bee physiology and survival. We found that bees consumed higher amounts of the provided artificial diets when supplemented with 24MC, suggesting that bees perceive the presence of sterol in their diet, which in turn may stimulate them to consume greater quantities of the diet. Some other phytophagous insects exhibit similar behavior. For example, silkworm larvae preferentially consumed diet blocks coated in their most physiologically important sterol over those coated in other sterols (Nagata et al. 2006). However, sterol concentrations may not be a predominant factor in diet preference when honey bees are able to choose between multiple diets varying in several nutritional components (Corby-Harris et al. 2018), wherein pollen diversity, protein to lipid ratio, or greater quantities of certain amino acids or fatty acids may serve as the dominant phagostimulant.

Fig. 3.

Mean consumption of artificial diets (mg per honey bee) in different experimental groups for 3 wk. Error bars indicate SEM. For every week, different letters indicate statistically significant differences between experimental groups. C0 and C1: control groups without and with acetone, respectively; S1, S2, S3, S4, and S5: 0.1, 0.25, 0.5, 0.75, and 1.0% dry diet weights of 24-methylenecholesterol in artificial diets.

Fig. 4.

Mean head protein content in bees that received 24-methylenecholesterol-supplemented diets or control diets C0 (without acetone) and C1 (with acetone). Diets for treatment groups S1, S2, S3, S4, and S5 contained 0.1, 0.25, 0.5, 0.75, and 1.0% 24-methylenecholesterol, respectively. Error bars indicate SEM. Different letters indicate significant differences among groups.

In our study, honey bee survival, head protein content, and abdominal lipid content were all significantly higher in bees that were fed diets with higher concentrations of sterols (even though the proportions of carbohydrates and proteins were identical among all diets). A similar effect of dietary sterols on fitness has been demonstrated in other insects. Adult female ambrosia beetles exhibited decreased longevity, locomotion, and reproduction when fed sterol-deficient diets (Norris and Moore 1980). Likewise, nymphs of the generalist grasshopper, Schistocerca americana, survived to adulthood in significantly greater proportions in less time when raised on diets containing sitosterol concentrations of 0.05% or greater than when fed diets with 0.025% sitosterol (Behmer and Elias 1999).

Fig. 5.

Mean abdominal lipid content (%) of bees receiving 24-methylenecholesterol-supplemented diets or control diets C0 (without acetone) and C1 (with acetone). Diets for treatment groups S1, S2, S3, S4, and S5 contained 0.1, 0.25, 0.5, 0.75, and 1.0% dry diet weights of 24-methylenecholesterol, respectively. Error bars indicate SEM. Different letters indicate significant differences among groups.

Although consumption of the artificial protein diets was not significantly different between S2, S3, S4, and S5 groups during any week of our experiment, at the end of 3 wk, head protein content was significantly higher in the S5 group than in the others. The protein content of the honey bee in whole body or a specific tissue is crucial to gaining insights on the physiology of bees and all correlated biological processes (Hartfelder et al. 2013). In an earlier study (Svoboda et al. 1986), 24MC was found in high quantities in the hypopharyngeal glands (located in the head capsule) of honey bees that consumed an artificial diet supplemented with 24MC. In addition, the 16-molecule architecture of the major royal jelly protein (MRJP1) oligomer was recently reported to be able to hold eight 24MC molecules (Tian et al. 2018). These findings are further supported by studies that report hypopharyngeal glands as the sites of MRJP1 production (Kucharski and Maleszka 1998). MRJP1 is an important component of the proteinaceous glandular secretions of the nurse bee hypopharyngeal and mandibular glands (both located in the honey bee head) and plays a role in age polyethism in honey bees (Buttstedt et al. 2014). The evidence of 24MC accumulation in these glands, coupled with our findings of significantly higher head protein content in bees fed greater concentrations of 24MC, suggests the potential role of this sterol (24MC) in protein synthesis in the brood food-producing glands (particularly, the hypopharyngeal glands). Thus, in our study, head proteins may be a good indicator of the state of brood food-producing glands in the experimental groups.

Abdominal fat is a crucial indicator of bee fitness and performance and regulates various aspects of insect physiology, including the synthesis and use of energy reserves (Arrese and Soulages 2010), lipid storage (Olofsson et al. 2009), detoxification (Arrese and Soulages 2010), synthesis of vitellogenin (Amdam et al. 2003, Arrese and Soulages 2010), and initiating foraging tasks in worker bees (Schulz et al. 1998, Toth et al. 2005) etc. Diet has previously been reported to influence gene expressions of the bee abdominal fat body tissues (Ament et al. 2011). In our study, lipid content in abdomens was higher in bees from sterol-rich treatment groups than in bees from the control group. Our lipid assays only quantified lipids in general, rather than distinguishing between fatty acids, triglycerides, sterols, etc. Therefore, we cannot predict what proportion each major lipid class contributed to the overall lipid content that we observed in abdomens. However, we surmise that sterols account for the difference in lipid stores between honey bees from high sterol groups and control groups.

The amount of lipids in insects may vary with life cycle stages, age, and nutritional needs (Beenakkers et al. 1985). This is especially true in honey bees, where the newly emerged adult workers (nurse bees) mature into foragers as they age (Crailsheim et al. 1992, Johnson 2010). Bees with access to greater quantities of dietary sterols were probably better able to replace the endogenous sterols they used over time and/or store excess sterols in the lipids of the abdominal fat bodies. It has been reported that nurse bees can selectively transfer significant amounts of this particular sterol from their endogenous pools to the developing brood via brood food (Svoboda et al. 1980), just as they do with amino acids and proteins when rearing brood during pollen shortages (Haydak 1970). It can be presumed that because ample fat body increases fitness in overwintering bees, sufficient consumption of sterols may potentially lead to a healthier overwintering colony in field with adequate abdominal lipid stores.

In honey bees, protein and lipid stores decline as they mature into foragers (Chan et al. 2011). We measured the effects of dietary sterols in bees that were nutritionally manipulated only in our experimental cages, as adults; the physiological parameters were examined only at the end of the experiment. Before these bees emerged from their respective cells and became part of our study, they were reared as brood in naturally foraging colonies. Thus, as larvae, they were fed by nurse bees that had access to pollen. Although we do not know the specific sterol content of this pollen, we can certainly assume that the pollen contained some phytosterols—as all plant pollens do (Villette et al. 2015). Thus, as newly emerged bees, in our experiment, they were presumably equipped with some amount of endogenous sterols. To negate the effects of endogenous sterols as a factor in our study, we thoroughly mixed all newly emerged bees of the same age group and randomly allocated them to the experimental cages. This ensured an unbiased homogenous mix of honey bees.

In our study, honey bees fed diets with a sterol concentration of 0.5% survived the longest and their head protein and abdominal lipid levels were also among the highest. Our findings provide additional insights regarding the role of this important sterol (24MC) in honey bee nutritional physiology and suggest that 0.5% concentration of 24MC may be an ideal concentration for formulating supplemental protein diets for honey bees. Research in the future should be conducted under realistic field conditions to verify the findings from our laboratory cage study. In addition, future research should also explore the feasibility of providing predetermined, specific amounts of diets to experimental bees instead of ad libitum for nuanced understanding of the effects of sterols, even though that approach appears to have significant limitations. Furthermore, the vast majority of bee species rely on pollen for their protein needs. The results of this study can be used for building an understanding of the impacts of pollen sterols on the physiology of other bee species. With multifactorial stressors contributing to pollinator decline, a fundamental knowledge of all nutritional needs of bees is crucial to improve and sustain pollinator health. Insights gleaned from this study have the potential to help formulate a more complete diet for honey bees in the future and help the beleaguered beekeeping industry.

Supplementary Data

Supplementary data are available at Annals of the Entomological Society of America online.