Chemical analysis of honey bees (Apis mellifera) and hive products show that most managed bee colonies in North America and Europe are repositories of a suite of chemical contaminants, including an assortment of insecticides, acaricides, herbicides and fungicides -. While a number of the residues detected were insecticides, many of which are highly toxic to bees, compounds of low acute toxicity were detected most frequently and at the highest concentrations in both bees and hive products. Some of the most ubiquitous contaminants of bees and bee products, coumaphos and tau-fluvalinate , are abundant in the hive environment because both are deliberately introduced as therapeutic acaricides to control the ectoparasitic mite, Varroa destructor. Varroa is the most serious pest of managed honey bee colonies in Europe and North America and clearly plays a role in the recent colony losses associated with colony collapse disorder -. Varroa weakens colonies in two ways: directly, by consuming the hemolymph of adult and pupal bees and, indirectly, by vectoring honey bee viruses and causing immunosuppression in parasitized bees .
In the face of the serious challenges presented by Varroa, beekeeping has become dependent on management techniques to control mite infestations, with apicultural acaricides playing a major role . However, finding chemical control agents that selectively kill an arthropod pest of an arthropod host poses a unique pharmacological challenge. Synthetic pesticides that have been used as acaricides include the pyrethroid tau-fluvalinate (Apistan and Mavrik), the organophosphate coumaphos (CheckMite+, Perizin and Asuntol 50), the formamidine amitraz (Apivar and Taktik) and the pyrazole fenpyroximate (Hivastan and FujiMite). Natural products are also used for Varroa control, including the monoterpenoid thymol (ApilifeVar and ApiGuard) and the organic acids, oxalic acid (Oxivar) and formic acid (MiteAway Quick Strips). There is little doubt that bees can benefit from reduced Varroa populations through the effective use of acaricides in combination with other management techniques , , . In a chemical survey of honey bee colonies suffering from colony collapse disorder, the healthiest colonies were found to have higher concentrations of one acaricide, coumaphos .
The effectiveness of tau-fluvalinate  and coumaphos  has waned as Varroa populations have developed resistance to these acaricides. However, tau-fluvalinate and coumaphos remain common contaminants in the hive environment, partially as a result of continued application by beekeepers, and partially due to their lipophilic properties which lead to accumulation and persistence in beeswax , . Both coumaphos and tau-fluvalinate survive the wax recycling process and are present in newly manufactured wax foundation , . While amitraz itself does not accumulate in bee colonies , the amitraz metabolite 2,4-dimethyl formamide (DPMF) has been detected in both bees and wax . Oxalic acid is a natural product that can be found in honey and as an allelochemical in plants , though not at concentrations used for Varroa control. While thymol and other monoterpenoids may be naturally present in floral sources at low concentration, the high concentrations needed for acaricidal activity may noticeably contaminate honey and wax , . With the wide range of acaricides currently in use and the continued presence of lipophilic acaricides in beeswax, it is quite likely that bees will be exposed to multiple acaricides simultaneously.
In addition to the acaricides, beekeepers may also apply antimicrobial drugs to control bacterial and microsporidial pathogens. Fumagillin (Fumadil-B) is fed in sucrose syrup to control infection by the microsporidian gut pathogens Nosema apis and Nosema ceranae. Oxytetracycline (Terramycin) and tylosin (Tylan) are applied in powdered sugar or syrup to control American foulbrood (Paenibacillus larvae) and other bacterial infections. To protect harvested honey from contamination many antimicrobial drugs and acaricides are subject to a withholding period during which these therapeutics cannot be applied. As such, beekeepers are left with a relatively narrow window during which antimicrobial or acaricide applications are possible, potentially leading to a situation where multiple treatments are applied simultaneously.
In addition to compounds applied by the beekeeper, bees may also be exposed to plant protection products applied to flowers and flowering crops. Fungicides are the most abundant and common of the plant protection products found in bees and bee products because fungicides can be applied during bloom when bees are present , , . While fungicides generally appear safe for adult bees , these compounds may, in certain situations, produce harmful effects . For instance chlorothalonil (Bravo), the most commonly detected fungicide in bees and bee products , was found in “entombed pollen” in colonies suffering from colony collapse disorder . Larval and pupal mortality has been reported in bees exposed to the fungicides pyraclostrobin and boscalid, which together constitute Pristine . There are also documented interactions between the sterol biosynthesis inhibiting (SBI) fungicides and pyrethroid insecticides in honey bees -. For example, prochloraz is a SBI fungicide that functions through the inhibition of fungal cytochrome P450-monooxygenase (P450) mediated synthesis of ergosterols. Prochloraz has been shown to inhibit detoxicative P450 activity in honey bees as well, particularly in relation to detoxication of pyrethroid pesticides , , , . Pristine and a variety of SBI fungicides are used during bloom on almond orchards, when bees are present, and have been detected in pollen samples , .
With the potential for managed honey bees to experience simultaneous exposure to acaricides, antimicrobials and fungicides it is important to consider the potential for harmful interactions between these compounds. Any interactions observed could provide an insight into honey bee physiology and will shed light on bees’ mechanisms of tolerance for both natural and synthetic xenobiotics. The present study aims to test for interactions between these most abundant contaminants of the hive environment using pair-wise lethal dose bioassays in which a sublethal pre-treatment with one acaricide, fungicide or antimicrobial is followed by a series of lethal doses of an acaricide. Mortality counts were then used to fit log-probit regression lines and determine lethal dose values (LD50s). Acaricides were also combined with model enzyme inhibitors to characterize the classes of detoxicative enzymes that may be the basis for interactive effects.
Bliss , recognized three principal types of interactive effects that pesticide or drug combinations may elicit . If no interaction occurs, and a combination is found to be only as deadly as its most toxic constituent, then the components of a mixture are understood to act independently. Such “independent joint action” is the null hypothesis for the experiments presented here and is characterized by acaricide toxicity that is unchanged by prior exposure to any other compound.
Interactive effects between compounds are observed when the toxicity of a drug or pesticide combination is either more or less toxic than expected based on the toxicity of the most toxic constituent. In the context of these experiments an interactive effect is defined as a change in the toxicity of an acaricide following sublethal pre-treatment with a fungicide, antimicrobial drug or another acaricide. An agonistic interaction is defined by the elevated toxicity of a drug or pesticide combination, while an antagonistic interaction is characterized by decreased toxicity. Interpreting the biological basis of interactive effects between compounds requires that the mode of action of the drugs or pesticides are known. Additive agonistic interactions are most likely to occur when different compounds work through the same mode of action. Synergistic agonistic interactions probably occur when the compounds work through different modes of action. The mode of action for each compound is listed in Figure 1 and 2 for each acaricide, fungicide and antimicrobial compound -.
Confidence intervals (95%) are indicated below the LD50 values. Significant differences compared to the control treatment are indicated with a superscript letter: a = significant pre-treatment effect, b = significant pre-treatment*acaricide dose effect (Table S1). LD50 values taken from previous work: † = , ‡ = . Names for classical enzyme inhibitors are abbreviated as follows DEM = diethyl maleate, DEF = S,S,S-tributylphosphorotrithioate, PBO = piperonyl butoxide. A dash “−” indicates an LD50 that could not be calculated because of insufficient data.
Confidence intervals (95%) are indicated below the LD50 values. Significant differences compared to the respective treatment are indicated with a superscript letter a = significant pre-treatment effect, b = significant pre-treatment*acaricide dose effect (Table S2).