Effects of used brood comb and propolis on honey bees (Apis mellifera L.) and their

associated bacterium, Melissococcus plutonius

Thesis

Presented in Partial Fulfillment of the Requirements for the Degree Master of Science in

the Graduate School of The Ohio State University

By

Stephanie K. Murray, B.S.

Graduate Program in Entomology

The Ohio State University

2019

Thesis Committee

Dr. Reed M. Johnson, Advisor

Dr. P. Larry Phelan

Dr. Rachelle M. M. Adams

1

Copyrighted by

Stephanie K. Murray

2019

2 Abstract

Wax brood comb is the place where larval honey bees are reared, making it the physical heart of a honey bee colony. Generations of larval rearing darken the color of the wax and leave behind layers of excrement, silken cocoons and sometimes bacterial or fungal spores that are harmful to developing or adult honey bees. Additionally, the chemical nature of beeswax creates a sink for many compounds to be absorbed—these include pesticides and beekeeper-applied acaricides, as well as pheromones produced by larval honey bees. Finally, propolis—a dark brown antimicrobial substance collected by bees from plant buds—can also become concentrated around the rims of wax comb cells.

As beekeepers become more aware of the risks caused by potential pesticide and microbe build-up, a practice known as brood comb replacement is becoming more popular in the

United States. With this practice, beekeepers are systematically removing and replacing old brood combs after several years in use. However, there is little research on the effects of old brood comb on honey bee survival or the frequency with which beekeepers should replace brood comb. Additionally, some beekeepers prefer to save old brood combs, as they are preferred by honey bee swarms that are settling into a new nest space.

This thesis aims to elucidate some of the potential benefits of old darkened brood comb by recording differences in colony preference for, larval survival on, and antimicrobial activity of wax combs that have or have not been used for larval rearing.

Overall, we observed no differences in colony preference for or larval survival on comb ii treatments that were or were not previously used for larval rearing. However, there were antimicrobial effects of extracts made from both darkened brood comb and light honey comb against a honey bee pathogen, Melisococcus plutonius—the bacterial agent responsible for a larval bee disease called European Foulbrood. Extracts of propolis were also found to inhibit M. plutonius growth, suggesting that the antimicrobial effects of wax may derive from the propolis incorporated into wax. However, the antibacterial components of wax combs are still unknown and should be studied further.

iii Acknowledgments

I would first like to acknowledge the University and the CFAES Research

Enhancement Competitive Grants Program for funding the research included in Chapter 3 of this thesis. Next, I would like to thank my graduate advisor, Dr. Reed Johnson for his kindness, support and guidance throughout my graduate career at OSU. Reed has given me room to explore avenues of research not typically covered in his lab and has made great efforts to connect me with professionals who share similar interests. He has given me time to learn from my mistakes, but also pushed me in the right direction when needed. Finally, Reed is always supportive and encouraging when I want to explore new opportunities, or when circumstances feel discouraging. I would also like to thank my committee members Dr. Rachelle Adams and Dr. Larry Phelan for their guidance and support. Rachelle has not only provided lab space and equipment to make my research possible, but she has also helped guide me through experimental design and push me through the writing process. Larry has not only provided expertise on chemical ecology but has also greatly developed my critical thinking skills through one-on-one conversation and lecture activities. Both Rachelle and Larry have always had open doors when I was feeling overwhelmed through this process.

I would also like to acknowledge the many lab members and colleagues who made my research possible through equipment preparation, data collection, expert guidance, donation of space and equipment and being a listening ear. Colleagues outside iv of our lab include Dr. Thaddeus Ezeji, Dr. Christopher Okonkwo, Dr. Vanessa Hale, Dr.

Christopher Madden, Andrew Mularo and Robert Filbrun. Lab members include Jacob

Shuman, Jessica Lyons, Colin Kurkul, Tyler Eaton and Sreelakshmi Suresh.

I would like to thank many beekeepers throughout the state of Ohio that have encouraged my research through their genuine interest and engaging conversation.

Beekeepers from the Tri-County Beekeepers Association and the Greater Cleveland

Beekeepers Association have welcomed me to participate and present at meetings.

Conversation with local beekeepers has provided me with a great deal of joy, inspiration and encouragement.

Finally, I would like the thank my friends and family who have been immensely supportive throughout this journey. I would be nowhere without the unconditional love, support and encouragement that my parents and siblings have always given me. They helped remind me to smile and laugh through the mentally trying times of a earning a graduate degree. My grandparents and extended family have also been involved and encouraging throughout all my years in school. Lastly, I would like to acknowledge my dear cat, Lulu, for the companionship and joy that she has brought me during my time in

Ohio.

v Vita

June 2013 ...... Ringgold High School

May 2017 ...... B.S. Biology, California University of

Pennsylvania (Cal U of PA)

Aug 2016 – May 2017 ...... Lab Assistant, Cal U of PA

Aug 2018 – May 2019 ...... Graduate Teaching Assistant, The Ohio

State University (OSU)

Aug 2019 – Present ...... Instructor, OSU’s Agricultural Technical

Institute

February 2019 ...... OSU SEEDS Grant

March 2019 ...... Entomological Society of America, North

Central Branch 1st Place M.S. Poster

Expected December 2019 ...... M.S. Entomology, The Ohio State

University

Fields of Study

Major Field: Entomology

vi

Table of Contents

Abstract ...... ii Acknowledgments ...... iv Vita ...... vi List of Tables...... ix List of Figures ...... x Chapter 1. Literature Review ...... 1 1.1 Introduction ...... 1 1.2 Brood Comb as a Composite Material ...... 3 1.3 Pesticide Residues and Microbial Contaminants ...... 5 1.4 Benefits of Old Brood Comb: Pheromones, Propolis and Social Immunity ...... 9 1.5 Brood Comb Replacement in the United States ...... 12 1.6 Research Objectives ...... 15 1.7 References ...... 15 Chapter 2. Effects of comb used for brood rearing on honey bee (Apis mellifera L.) preference and survival ...... 20 2.1 Abstract ...... 20 2.2 Introduction ...... 21 2.3 Materials and Methods ...... 23 2.3.1 Brood Comb Treatments ...... 23 2.3.2 Choice Arena and Honey Bee Package Installation ...... 24 2.3.3 Larval Survival Estimation ...... 26 2.3.4 Statistical Analysis ...... 27 2.4 Results ...... 28 2.4.1 Choice Experiment ...... 28 2.4.2 Larval Survival...... 28 2.5 Discussion ...... 31 vii 2.6 Acknowledgements ...... 34 2.7 References ...... 34 Chapter 3. The effects of propolis and brood comb extracts on the causative agent of European Foulbrood (Melissococcus plutonius) in honey bees ...... 37 3.1 Abstract ...... 37 3.2 Introduction ...... 38 3.3 Materials and Methods ...... 41 3.3.1 Comb and Propolis Samples ...... 41 3.3.2 Comb and Propolis Extractions ...... 42 3.3.3 Bacterial Cultures ...... 42 3.3.4 Bacterial Growth Tests ...... 43 3.3.5 Statistical Analysis ...... 44 3.4 Results ...... 46 ...... 48 3.5 Discussion ...... 51 3.6 Acknowledgements ...... 54 3.7 References ...... 55 Conclusion ...... 60 Bibliography ...... 63

viii List of Tables

Table 1. Protocol for Melissococcus plutonius cultivation...... 46 Table 2. Matching GenBank sequence descriptions...... 46

ix List of Figures

Figure 1. Average rate of overwinter colony loss for responding U.S. beekeepers using brood comb of various ages (2008-2019)...... 14 Figure 2. Preparation of honey comb and brood comb treatments...... 24 Figure 3. Installing honey bee packages into choice arenas...... 25 Figure 4. Timeline of data collection for larval survival on unused or used brood comb treatment...... 27 Figure 5. Average larval survival on honey comb or brood comb estimated by visual inspection (A-B) and image analysis (C-D)...... 30 Figure 6. Total brood area on honey comb and brood comb treatments...... 31 Figure 7. Experimental colony design...... 45 Figure 8. Difference in optical density (OD) of E. coli growth relative to the solvent control for all ethanol and methanol-suspended treatments...... 48 Figure 9. Difference in optical density (OD) of S. saprophyticus growth relative to the solvent control for all ethanol and methanol-suspended treatments...... 49 Figure 10. Difference in optical density (OD) of M. plutonius growth relative to the solvent control for all ethanol and methanol-suspended treatments...... 50

x Chapter 1. Literature Review

1.1 Introduction

Today, honey bee colonies are easily managed thanks to the invention of the

Langstroth hive and moveable frames. Frames can be either wooden or plastic and hold a thin layer of wax that worker bees use as a template to draw out their wax combs

(Magnum, 2015). Worker bees utilize energy from food stores to build the hive’s wax from glands on their abdomens (Snodgrass et al, 2015) —1 kg of beeswax is energetically equivalent to 7.5 kg of honey (Seeley & Morse, 1976; Seeley, 1985). Wax combs built by honey bees include both brood comb and honey comb. These two types of comb begin the same—as lightly-colored wax scales made-up of mostly hydrocarbons

(Tulloch, 1979), manipulated and formed into hexagonal cells fit for larval, or brood, development and food storage. It is only the activities within in the hive and within the apiary that determine whether we call a frame of beeswax brood comb or honey comb.

In a feral hive, a queen-right colony will keep brood in the center of the chamber while nectar and pollen are stored just outside of this patch of brood. This creates a circular pattern in the comb, with honey storage on the outermost part and brood development on the innermost part of the circle (Schneider, 2015). This pattern is still seen in managed beehives, unless there is an addition of queen excluders and honey supers by beekeepers. Queen excluders are thin metal grids that are put between large 1 deep boxes and smaller medium boxes or supers. These excluders allow the smaller worker bees, but not the larger queen bee, to pass through to the supers. It is through these management techniques that beekeepers can produce frames of honey within a super (Delaplane, 2015; Killion, 2015). As beekeepers, we refer to any comb that holds developing larvae as brood comb, and comb that is strictly for food storage as honey comb.

The use of moveable frames in the hive also gives beekeepers the ability to store and reuse brood combs for many years. It was once more common to reuse brood combs for years until damaged (Jaycox, 1979), but beekeepers in the United States have recently started replacing brood combs on a regular basis. This practice is common among beekeepers for brood comb frames, rather than honey comb, as brood rearing causes major build-up of frass and silken cocoons in the wax (Jay, 1963; Jay, 1964; Hepburn &

Kurstjens, 1988). In addition to brood rearing build-up, wax combs also become a sink for pesticide residues (Smith & Wilcox, 1990; Johnson et al, 2010; Mullin et al, 2010;

Wu et al, 2011) and microbial pests (Pankiw et al, 1970; Koenig et al, 1986; Di Pinto et al, 2011). Every life-sustaining activity of a colony is performed on the wax comb, so researching the costs and benefits of brood comb replacement is important for bee health and survival.

Pesticide residues in brood comb have demonstrated sublethal effects on larval and adult honey bees (Wu et al, 2011), and these residues also made bees more susceptible to infection by Nosema ceranae, a fungal parasite (Wu et al, 2012).

Additionally, the rate of microbial infection may also be higher in bees reared on old brood comb, as fungal and bacterial spores are likely retained in the layers of brood 2 rearing associated build-up (Pankiw et al, 1970; Koenig et al, 1986; Di Pinto et al, 2011).

For example, bacterial cells of Melissococcus plutonius, the causative agent of brood disease European Foulbrood (EFB), were retained in brood combs after treatment by gamma irradiation, and cells were still viable to cause EFB infection when placed in a healthy hive (Pankiw et al, 1970).

It is due to this potential for deleterious waste and residue build-up that beekeepers are proactively removing old brood combs and replacing them with fresh plastic or wooden foundation frames. Though the intention is to improve colony health and survival, research on this topic is ambiguous, as some results point to the opposite effect. For example, larval survival was actually higher on old brood comb in two previous studies (Free & Winder, 1983; Berry & Delaplane, 2001), and a similar relationship has been observed in data from a national beekeeper survey. Contrary to the goals of brood comb replacement, results of an 11-year survey of beekeepers demonstrate a positive relationship between older brood comb and overwinter success (Bee Informed

Partnership, 2019). The following review will discuss the conflicting evidence on the effects of old brood comb, including brood rearing build-up, pesticide residues and microbial contaminants, as well as the beneficial components that may be responsible for increased survival.

1.2 Brood Comb as a Composite Material

Brood comb begins as a pliable, lightly-colored wax, but quickly changes from a single wax phase to a two-phased material after several cycles of brood rearing (Hepburn

& Kurstjens, 1988). After each brood cycle, the developing larvae leave a layer of silk and frass within each occupied cell (Jay, 1963; Jay, 1964; Hepburn & Kurstjens, 1988). 3 Although each larva begins her adult life by cleaning out her cell, the silken cocoon she has spun is embedded in the wax, resulting in a darker and thicker composite material with each brood cycle. In other words, after each brood cycle, the ratio of silk to wax becomes greater, eventually resulting cells with a smaller diameter (Hepburn &

Kurstjens, 1988).

Following the last larval molt, worker larvae will turn and stretch out along the length of their cells, while spinning a silken cocoon that is applied all sides of the inner cell walls. The entire spinning process takes anywhere from 24-48 hours for a worker larva to complete. It is during this time that larval frass is released within the cell. Worker larvae begin defecating into their cells 12-18 hours after cocoon spinning begins (Jay

1963, 1964). Larvae produce both a colorless, pollen-free frass followed by a darker, yellow, pollen-bearing frass (Jay, 1963; Jay, 1964; Hepburn & Kurstjens, 1988).

Defecation takes place during cocoon-spinning, and therefore, frass becomes trapped between layers of silk that are left behind in the cell walls (Jay 1963, 1964).

Without the addition of silk in the cell walls, single-phase wax makes a delicate, pliable comb in which mechanical properties vary with temperature. Mechanical properties such as strength, breaking strain, and stiffness all decrease with increasing temperature (Hepburn & Kurstjens, 1988). Although the mechanical properties of two- phase brood comb exhibit the same decreasing pattern, the addition of silk increases the strength, breaking strain, and stiffness of the comb overall (Hepburn & Kurstjens, 1988).

For example, the strength of single-phase wax at 25°C was measured at about 1.0 MPa, while that of brood comb containing 34% silk was measured at about 4.3 MPa (Hepburn

& Kurstjens, 1988). Honey bee silk is made-up of an alpha-helical protein called fibroin. 4 Unlike those of pure wax and brood comb, the mechanical properties of silk fibroin alone remain unchanged with increasing temperature (Hepburn & Kurstjens, 1988).

In addition to silk and frass incorporation, brood comb can also be altered by the addition of propolis (Strehle et al, 2002) or the build-up of pesticide residues (Smith &

Wilcox, 1990; Johnson et al, 2010; Wu et al, 2011) and microbial pathogens (Pankiw et al, 1970; Koenig et al, 1986; Di Pinto et al, 2011). Propolis—an antimicrobial resin that bees collect from plant bud exudates (Marcucci, 1995)—is frequently added to the rims of comb cells (Strehle et al, 2002) and, like caulking, used to seal any cracks in the hive body (Marcucci, 1995). In a feral colony—often within a tree cavity—the hive walls are completely covered in propolis, referred to as a propolis envelope (Seeley & Morse,

1976). This resinous substance exhibits structural properties similar to those of single- phase beeswax (Hepburn & Kurstjens, 1988). Though propolis exhibits some antimicrobial properties, the build-up of pesticides and microbes could present some challenges to honey bee health. Additionally, the thickening of comb cells over time can also present a problem for colony health, as a smaller cell produces smaller workers bees

(McMullan & Brown, 2006) with a potentially smaller foraging range and functional proboscises (Waddington & Herbst, 1987; Greenleaf et al, 2007).

1.3 Pesticide Residues and Microbial Contaminants

The number of honey bee colonies in the U.S. has been in decline for years

(vanEngelsdorp et al, 2009), and pesticides are often given the blame. Pesticides— insecticides in particular—can have a negative effect on honey bee health; however, there are multiple factors to blame for the steady decline in colonies (vanEngelsdorp et al,

2009). In addition to insecticides, honey bees are faced with microbial pests that cause 5 disease, parasitic pests like the Varroa mite, and viruses that parasites help transmit

(Sumpter & Martin, 2004; vanEngelsdorp et al, 2009; Rosenkranz et al, 2010). For example, the Varroa mite (Varroa destructor) is responsible for transmitting several honey bee viruses, such as deformed wing virus (DWV), acute paralysis virus (APV), and

Kashmir bee virus (KBV) (Sumpter & Martin, 2004). Varroa mites vector diseases as they feed on honey bee fat body tissues (Ramsey et al, 2019), presenting a new route of virus transmission in a Varroa-infested colony. DWV and APV are typically detectable in honey bees at low levels, but mites that feed on an individual bee hosting an overt viral infection can quickly spread the virus throughout an entire colony or across an apiary

(Sumpter & Martin, 2004).

Because Varroa mites are one of the most pressing threats to honey bee health, it is imperative for beekeepers to control mite populations, often by applying pesticides directly to their colonies. Acaricides often used by beekeepers in the U.S. include synthetic organics, natural products, and organic acids (Johnson et al, 2010).

Unfortunately, finding acaricides that successfully kill Varroa mites without harming honey bees is a difficult challenge. Additionally, acaricides may stick around in beeswax for years. Synthetic acaricides, such as pyrethroid tau-fluvalinate (Apistan®) and organophosphate coumaphos (Checkmite+®), are lipophilic, having an estimated half-life of five years within wax comb (Bogdanov, 2004; Johnson et al, 2010). Natural products, such as thymol (Apilife Var® and Apiguard®) and organic acids, such as formic acid

(MiteAway I® and MiteAway II®), are not as persistent in wax, but still present potential threats to honey bee health (Johnson et al, 2010).

6 In addition to beekeeper-applied acaricides, honey bees are also exposed to agrochemicals that are applied to nearby fields. Neonicotinoid and phenylpyrazole insecticides are frequently applied in U.S. agriculture (Johnson et al, 2010), and their systemic nature makes active ingredients available in nectar and pollen collected by foraging bees (Cutler & Scott Dupree, 2007; Johnson et al, 2010). Although these residues can be found in pollen, honey and adult bees, they are less frequently found in comb, due to their low solubility in wax (Johnson et al, 2010). A study by Wu et al

(2011) found 23 insecticide and miticide residues in brood comb samples from migratory beekeeping operations, but the three most frequently detected residues were those of beekeeper-applied acaricides. Acaracide residues included fluvalinate, coumaphos, and the metabolite coumaphos oxon. These residues were found at average levels of 6,712 ppb, 8,079 ppb, and 596 ppb, respectively (Wu et al, 2011). Over a period of three months, pesticide residues present in these used brood combs quickly migrated to newly drawn out control combs in this experiment (Wu et al, 2011). The residues present in treatment brood combs and reused control combs had sub-lethal effects on both developing larvae and adult bees. Larval development and adult emergence were delayed in bees exposed to pesticide residues, and longevity of adult bees from hives exposed to pesticide residues was four days shorter (Wu et al, 2011).

Wax brood comb has become a sink for miticide residues, and it may also harbor microbial pests that cause common larval honey bee disease, such as chalkbrood.

Chalkbrood is caused by the spore-forming fungus Ascosphaera apis and is characterized by larvae that first swell with a white, fur-like fungal infection, but later become dry and chalk-like after they have died in their cells (Bailey & Ball, 1991). Compared to honey 7 comb or newly drawn foundation frames, Koenig et al (1986) found higher rates of chalkbrood disease on three types of old brood comb: 1) 5-30 years old, 2) 30-45 years old, or 3) of various ages, but fumigated with ethylene oxide. Among the old comb treatments, infection rates were highest on the oldest comb and lowest on the comb that had been fumigated, but colonies on all old comb were significantly more infected than those on honey comb or newly drawn foundation frames (Koenig et al, 1986). Since fumigation would reduce microbes but not pesticide residues, these results suggest that pesticide residues are not the only deleterious factor built up in old brood comb.

Significantly higher levels of chalkbrood in brood comb greater than 30 years old indicates that A. apis spores remain viable in the comb for many years.

In addition, exposure to pesticide residues in the presence of microbial contamination can have a synergistic effect on colonies. Colonies reared on old comb with elevated levels of synthetic miticide residues showed higher susceptibility to

Nosema ceranae (Wu et al, 2012). Nosema ceranae is another spore-forming fungal parasite of honey bees; however, this fungus causes disease in adult bees by robbing the midgut of nutrients and causing stress and defecation (Bailey & Ball, 1991). In addition to fungal pathogens, honey bees are also faced with bacterial diseases, such as European and American foulbrood, caused by Melissococcus plutonius and Paenibacillus larvae, respectively (Forsgren, 2010; Hansen & Brodsgaard, 1999).

European foulbrood (EFB) and American foulbrood (AFB) are among the most economically damaging honey bee diseases (Forsgren, 2010). Both diseases are wide- spread, affecting nearly every region in which apiculture is practiced. Both bacterial agents are anaerobic, proliferating within the larval honey bee midgut, and are transferred 8 through feeding and other hive activities, such as hygiene or removal of infected larvae

(Hansen & Brodsgaard, 1999; Bailey, 1983; Forsgren, 2010). Infected larvae of both diseases become sunken and discolored (Hansen & Brodsgaard, 1999; Forsgren et al,

2013). In addition, P. larvae infection causes a rubber-like consistency of sunken larvae, which later dry and form hard scales that adhere to the cell walls (Hansen & Brodsgaard,

1999). These scales retain thousands of P. larvae spores, making infection hard to eradicate from comb and other equipment (Hansen & Brodsgaard, 1999; Di Pinto et al,

2011). Melissococcus plutonius is not a spore-forming bacterium; however, cells can also be retained in wax combs, as demonstrated by Pankiw et al (1970). Although they found

EFB infection to be inversely related to gamma irradiation dose, the disease was not eradicated even at the highest dose (Pankiw et al, 1970). AFB infection is typically treated with oxytetracycline hydrochloride and scorching or burning infected hive equipment (Hansen & Brodsgaard, 1999), while EFB infections are typically treated with oxytetracycline hydrochloride and replacement of combs (Forsgren et al, 2010).

1.4 Benefits of Old Brood Comb: Pheromones, Propolis and Social Immunity

Despite the negative potential for pesticide and microbial contamination in old brood combs, some research has demonstrated higher honey bee survival in hives with old brood comb. A survey by the Bee Informed Partnership (BIP) (2017) revealed a positive relationship between brood comb age and overwinter colony survival. Berry and

Delaplane (2001) found mixed effects of brood comb age on colony health. Although they measured significantly higher larval survival when reared on old, dark comb, area of brood, capped brood, weight of emerging bees and adult bee population were significantly higher on new comb. Although pesticide residues are often among the 9 compounds being absorbed, wax combs also absorb pheromones emitted by bees within the hive, such as brood pheromone. Brood pheromone is made up of 10 fatty aliphatic esters, three of which are important in modulating brood acceptance and feeding: methyl stearate, methyl linoleate and methyl palmitate (Le Conte et al, 1995). Because brood pheromone stimulates pollen foraging (Free & Williams, 1974; Pankiw et al, 1998) and larval feeding by nurse bees (Le Conte et al, 1995), old brood comb that has likely absorbed and retained the pheromone is hypothesized to improve brood rearing activity

(Free & Winder, 1983; Berry & Delaplane, 2001). Free & Winder (1983) found that artificial or dead larvae placed into old comb cells were accepted and reared by nurse bees significantly more frequently than those placed into new comb cells.

Beekeeping lore states that the smell of old brood comb is attractive to honey bee swarms looking for new nest cavities. To improve their efforts at capturing a honey bee swarm, beekeepers will place old brood comb in their swarm boxes or hives. Swarm preference for nest spaces holding old brood comb compared to nest spaces holding no comb has also been empirically demonstrated (Visscher et al, 1985); however, the mechanism by which honey bees were attracted was not examined. Visscher et al (1985) hypothesize that cavities holding old brood comb are preferred by swarms because the cavities are either more conspicuous due to olfactory cues (such as brood pheromone or other pheromones present in the wax), or because preferring a space with constructed comb cells is adaptive for honey bees.

The build-up of propolis is another potential benefit of used beekeeping equipment, including old brood comb. After collecting antimicrobial plant resins from plant buds and exudates (Marcucci, 1995), resins remain mostly uniform (Strehle et al, 10 2003), but are sometimes supplemented by wax and other organic matter during their integration into the hive (Marcucci, 1995). Propolis composition largely depends on the time and location of resin collection (Marcucci, 1995); however, North American propolis is mostly collected from Populus spp (Andelković et al, 2017). Johnson et al.

(1994) examined propolis composition and activity from several sites in the United

States, one of them being western Ohio. Ohio propolis was chemically diverse, possessing more than 30 UV-absorbing compounds including phenolics and flavonoids that suppressed the growth of wax moth larvae when added to their diet (Johnson et al,

1994). Activity of propolis against bacteria, fungi, viruses, cells and protozoa is largely due to flavonoids, aromatic acids, esters and caffeic acid derivatives (Marcucci, 1995).

Benefits of propolis to honey bee colonies include increased brood viability and adult longevity (Nicodemo et al, 2014), decreased viral titers of DWV (Drescher et al, 2017), and downregulation of expensive immune genes and antibacterial peptides (Simone et al,

2009; Borba et al, 2015). Although many of these conclusions come from hives experimentally supplemented with propolis and bees selected for propolis-collection, colonies have also been observed self-medicating by collecting propolis (Simone-

Finstrom & Spivak, 2012). In this experiment, colonies challenged with chalkbrood (A. apis) were observed collecting a significantly higher amount of propolis than their healthier counterparts (Simone-Finstrom & Spivak, 2012). The use of propolis to self- medicate is a form of social immunity—a phenomenon in which individuals of a superorganism perform altruistic behaviors that will benefit the colony rather than themselves (Cremer et al, 2007; Simone et al, 2009). Finally, propolis extracts derived from poplar propolis demonstrated antimicrobial activity against the bacterium and 11 fungus responsible for American foulbrood (P. larvae) and chalkbrood (A. apis), respectively (Wilson et al, 2017).

1.5 Brood Comb Replacement in the United States

Although beekeepers in the U.S. have historically used their brood combs until they become heavily contaminated or damaged (Jaycox, 1979), comb management practices of European beekeepers are increasingly being adopted. European beekeepers have long practiced brood comb replacement, as they believe that old brood combs are a reservoir for microbial contamination (Jaycox, 1979). European beekeeping books recommend adding 5-8 fresh foundation frames per hive every year and most European beekeepers replace their brood combs after 3-4 years of use in the apiary (Jaycox 1979).

Although replacing brood comb every 3-5 years is often recommended in the U.S., there has been little research to support this recommendation. Specifically, there is no empirical research suggesting the frequency with which brood comb should be replaced.

While old brood comb can negatively affect the health of a colony (Koenig, 1986;

Wu et al, 2011; Wu et al, 2012); however, freshly drawn, new brood comb presents its own economic, energetic and biological drawbacks. Frequently replacing old brood comb is not only costly for beekeepers (Hoopingarner & Sanford, 1990), it also requires energetic expenditures for honey bees forced to build new wax combs. An average hive contains nearly 1,200 g of wax, requiring bees to spend energy equivalent to 7.5 kg of honey (Seeley & Morse, 1976; Seeley, 1985). The time and energy bees spend on wax production represent trade-offs with accumulating resources early in the season. In addition to energetic cost and despite potential sublethal effects of old brood comb (Wu

12 et al 2011), Berry & Delaplane (2001) found that brood survival was significantly lower in newly drawn combs than in old brood combs. Finally, sublethal effects on larvae are likely due to synthetic miticide build-up in old brood combs (Wu et al, 2011); however,

Mullin et al (2010) found that the small layer of wax applied to fresh foundation frames is often contaminated with that same miticide residues.

Efforts from the BIP have provided years of survey results on U.S. beekeeper comb practices and their overwinter success rates. Results from these surveys demonstrate a positive relationship between older brood comb and colony overwinter success (Figure 1) (Bee Informed Partnership, 2017). Additionally, data from 2012-2015 show higher overwinter loss in apiaries that replaced more brood comb (Bee Informed

Partnership, 2012; 2013a; 2013b; 2014; 2015).

13

Figure retrieved and adapted from the Bee Informed Partnership (https://bip2.beeinformed.org/survey/). Respondents include commercial beekeepers, sideline beekeepers and hobbyists. Error bars represent standard deviation. Response n = 29,323 from 2008-2019.

Figure 1. Average rate of overwinter colony loss for responding U.S. beekeepers using brood comb of various ages (2008-2019).

14

1.6 Research Objectives

To disentangle the benefits and detriments of old brood comb on honey bee health and preference, I sought out to answer the following research question: Does comb used for brood rearing and the presence of propolis have a biological effect on honey bees or their associated microbial pests?

I formulated two main objectives to answer this question: (1) determine the effects of used brood comb on colony preference and larval survival in the field and (2) determine the antibacterial effects of used comb and propolis on M. plutonius (the bacterium responsible for European foulbrood) in vitro.

1.7 References

Andelković, B., Vujisić, L., Vučković, I., Tešević, V. & Vajs, V. (2017). Metabolomics study of Populus type propolis. Journal of Pharmaceutical and Biomedical Analysis, 135, 217-226.

Bailey, L & Ball, B. V. (1991). ‘Fungi’ in Honey Bee Pathology. Academic Press: London, UK. Pp. 53-63.

Bailey, L. (1983). Melissococcus pluton, the cause of European foulbrood of honey bees (Apis spp.). Journal of Applied Bacteriology, 55, 65-69.

Bee Informed Partnership. (2012a). The Bee Informed Partnership Management Survey Results (2011) Brood Comb Management. Retrieved from https://beeinformed.org/wp-content/uploads/2012/04/Brood-Comb- managment.pdf

Bee Informed Partnership. (2012b). The Bee Informed Partnership Management Survey Results (2012-2013) Brood Comb Management. Retrieved from https://beeinformed.org/wp-content/uploads/2014/11/Comb-Management- Summary.pdf

Bee Informed Partnership. (2013). The Bee Informed Partnership Management Survey Results (2012) Brood Comb Management. Retrieved from https://beeinformed.org/wp-content/uploads/2013/08/title-Brood-Comb- managment.pdf 15 Bee Informed Partnership. (2014). The Bee Informed Partnership Management Survey Results (2013-2014) Brood Comb Management. Retrieved from https://beeinformed.org/wp-content/uploads/2014/09/Comb-Management- Summary.pdf

Bee Informed Partnership. (2019). Average Winter Loss per Beekeeper. Retrieved from https://bip2.beeinformed.org/survey/

Berry, J. A., & Delaplane, K. S. (2001). Effects of comb age on honey bee colony growth and brood survivorship. Journal of Apicultural Research, 40(1), 3–8.

Bogdanov, S. (2004). Beeswax: Quality issues today. Bee World, 85(3), 46-50.

Borba, R. S., Klyczek, K. K., Mogen, K. L. & Spivak, M. (2015). Seasonal benefits of a natural propolis envelope to honey bee immunity and colony health. Journal of Experimental Biology, 218, 3689-3699.

Cremer, S., Armitage, S. O. A. & Schmid-Hempel, P. (2007). Social immunity. Current Biology, 17, R693-R702.

Cutler, G. C. & Scott-Dupree, C. D. (2007). Exposure to clothianidin seed-treated canola has no long-term impact on honey bees. Ecotoxicology, 100(3), 765- 772.

Delaplane, K. S. (2015). ‘Management for honey production’ in Graham, J.M. The Hive and the Honey Bee. Dadant & Sons: Hamilton, IL, USA. Pp. 505-506.

Di Pinto, A., Novello, L., Terio, V., & Tantillo, G. (2011). ERIC-PCR genotyping of Paenibacillus larvae in Southern Italian honey and brood combs. Current Microbiology, 63(5), 416–419.

Drescher, N., Klein, A., Neumann, P., Yañez, O. & Leonhardt, S. D. (2017). Inside honeybee hives: Impact of natural propolis on the ectoparasitic mite Varroa destructor and viruses. Insects, 8(15).

Forsgren, E. (2010). European foulbrood in honey bees. Journal of Invertebrate Pathology, 103, S5-S9.

Forsgren, E., Budge, G.E., Charriere, J., & Hornitzky, M.A. (2013). Standard methods for European foulbrood research. Journal of Apicultural Research, 52(1), 1-14.

Free, J.B. & Winder, M.E. (1983). Brood recognition by honeybee (Apis mellifera) workers. Animal Behavior, 31(2) 539-545.

Greenleaf, S. S., Williams, N. M., Winfree, R., & Kremen, C. (2007). Bee foraging ranges and their relationship to body size. Oecologia, 153(3), 589–596.

16 Hansen, H. & Brodsgaard, C. J. (1999). American foulbrood: a review of its biology, diagnosis and control. Bee World, 80(1), 5-23.

Hepburn, H. R., & Kurstjens, S. P. (1988). The combs of honeybees as composite materials. Apidologie, 19(1), 25–36.

Hoopingarner, R. & Sanford, M.T. (1990). The costs of beekeeping—I. Survey of commercial beekeepers. American Bee Journal, 130(6), 405-407.

Jay, S. C. (1963). The development of honeybees in their cells. Journal of Apicultural Research, 2(2), 117–134.

Jay, S. C. (1964). The cocoon of the honey bee, Apis mellifera L. The Canadian Entomologist, 96(5), 784–792.

Jaycox, E. R. (1979). Comb foundation: Are we using enough? American Bee Journal, 119(7), 515–516.

Johnson, K. S., Eischen, F. A. & Giannasi, D. E. (1994). Chemical composition of North American bee propolis and biological activity towards larvae of greater wax moth (Lepidoptera: Pyralidae). Journal of Chemical Ecology, 20(7), 1783- 1791.

Johnson, R. M., Ellis, M. D., Mullin, C. A., & Frazier, M. (2010). Pesticides and honey bee toxicity – USA. Apidologie, 41(3), 312–331.

Killion, E. (2015). ‘The production of comb and bulk comb honey’ in Graham, J.M. The Hive and the Honey Bee. Dadant & Sons: Hamilton, IL, USA. Pp. 562- 563.

Koenig, J. P., Boush, G. M., & Erickson, E. H. (1986). Effect of type of brood comb on chalk brood disease in honeybee colonies. Journal of Apicultural Research, 25(1), 58–62.

Le Conte, Y., Sreng, L. & Poitout, S. H. (1995). Brood pheromone can modulate the feeding behavior of Apis mellifera workers (Hymenoptera: Apidae). Apiculture and Social Insects, 88(4), 798-804.

Magnum, W. (2015). ‘A brief history of beekeeping in the United States’ in Graham, J.M. The Hive and the Honey Bee. Dadant & Sons: Hamilton, IL, USA. Pp. 25- 50.

Marcucci, M. C. (1995). Propolis: chemical composition, biological properties and therapeutic activity. Apidologie, 26(1), 83–99.

17 McMullan, J. B., & Brown, M. J. F. (2006). The influence of small-cell brood combs on the morphometry of honeybees (Apis mellifera). Apidologie, 37(6), 665– 672.

Mullin C. A., Frazier M., Frazier J. L., Ashcraft S., Simonds R., et al. (2010). High levels of miticides and agrochemicals in North American apiaries: Implications for honey bee health. PLoS ONE 5(3): e9754

Nicodemo, D., Malheiros, E., de Jong, D. & Couto, R. (2014). Increased brood viability and longer lifespan of honeybees selected for propolis production. Apidologie, 45(2), 269-275.

Pankiw, P., Bailey, L., Gochnauer, T. A., & Hamilton, H. A. (1970). Disinfection of honeybee combs by gamma irradiation. II. European foul brood disease. Journal of Apicultural Research, 9(3), 165–168.

Ramsey, S. D., Ochoa, R., Bauchan, G., Gulbronson, C., Mowery, J. D., et al. (2019). Varroa destructor feeds primarily on honey bee fat body tissue and not hemolymph. PNAS, 116(5), 1792-1801.

Rosenkranz, P., Aumeier, P. & Ziegelmann, B. (2010). Biology and control of Varroa destructor. Journal of Invertebrate Pathology, 103, S96-S119.

Schneider, S. S. (2015). ‘The honey bee colony: Life history’ in Graham, J.M. The Hive and the Honey Bee. Dadant & Sons: Hamilton, IL, USA. Pp. 74-78.

Seeley, T. D. & Morse, R.A. (1976). The nest of the honey bee (Apis mellifera L.). Insects Sociaux, 23, 495-512.

Seeley, T. D. (1985). Honey bee ecology. Princeton University Press: Princeton, NJ, USA. Pp 201.

Simone, M., Evans, J. D. & Spivak, M. (2009). Resin collection and social immunity in honey bees. Evolution, 63(11), 3016-3022.

Simone-Finstrom, M. D. & Spivak, M. (2012). Increased resin collection after parasite challenge: a case of self-medication in honey bees? PLoS One, 7(3), e34601.

Smith, R. & Wilcox, M. M. (1990). Chemical residues in bees, honey, and beeswax. American Bee Journal, 130(3) 188-192.

Snodgrass, R. E., Erickson, E. H. & Fahrbach, S. E. (2015). ‘The anatomy of the honey bee’ in Graham, J.M. The Hive and the Honey Bee. Dadant & Sons: Hamilton, IL, USA. Pp. 141.

18 Strehle, M. A., Jenke, F., Fröhlich, B., Tautz, J., Riederer, M., et al. (2003). Raman spectroscopic study of spatial distribution of propolis in comb of Apis mellifera carnica (Pollm.). Biopolymers - Biospectroscopy Section, 72(4), 217–224.

Sumpter, J. T. & Martin, S. J. (2004). The dynamic of virus epidemics in Varroa- infested honey bee colonies. Journal of Animal Ecology, 73(1), 51-63.

Tulloch, A. P. (1979). Beeswax – Composition and analysis. Bee World, 60(2), 47-62.

Vanengelsdorp D., Evans J. D., Saegerman C., Mullin C., Haubruge E., et al. (2009). Colony collapse disorder: a descriptive study. PLoS One, 4(8), e6481.

Visscher, P. K., Morse, R. A., & Seeley, T. D. (1985). Honey bees choosing a home prefer previously occupied cavities. Insectes Sociaux, 32(2), 217–220.

Waddington, K. D., & Herbst, L. H. (1987). Body size and the functional length of the proboscis of honey bees. The Florida Entomologist, 70(1), 124–128.

Wilson, M. B, Pawlus, A. D, Brinkman, D., Gardner, G., Hegeman, A.D., et al. (2017). 3-Acyl dihydroflavonols from poplar resins collected by honey bees are active against the bee pathogen Paenibacillus larvae and Ascophaera apis. Phytochemistry, 138, 83-92.

Wu, J. Y., Anelli, C. M., & Sheppard, W. S. (2011). Sub-lethal effects of pesticide residues in brood comb on worker honey bee (Apis mellifera) development and longevity. PLoS ONE, 6(2), e14720.

Wu, J. Y., Smart, M. D., Anelli, C. M., & Sheppard, W. S. (2012). Honey bees (Apis mellifera) reared in brood combs containing high levels of pesticide residues exhibit increased susceptibility to Nosema (Microsporidia) infection. Journal of Invertebrate Pathology, 109(3), 326-329.

19 Chapter 2. Effects of comb used for brood rearing on honey bee (Apis mellifera L.) preference and survival

2.1 Abstract

In recent years, beekeepers in the U.S. have changed the way they manage frames of old brood comb darkened through rearing many generations of worker honey bees.

Previously, beekeepers preferred to reuse brood frames until they became damaged, but they are now becoming more aware of the effects that potential microbial and pesticide residues in old combs can have on developing bees. This awareness has led to systematic replacement of old brood comb in many U.S. beekeeping operations. Although older brood comb may present risks to colony health, there are also potential benefits. For example, used brood comb contains more propolis, which has antimicrobial properties, and is preferred by swarms looking for a hive site. We designed two experiments to test honey bee preference for and survival on used brood comb. We hypothesized that bees would prefer used brood comb and larval survival would differ between treatments.

Honey bee choice within the colony was recorded three days after installation of an artificial swarm of honey bees and subsequent larval survival on comb treatments was estimated using two different methods. No difference was measured for honey bee preference (p = 0.7) or larval survival (visual estimation, p = 0.8; image analysis, p =

0.7). Results suggest that colonies have no preference for used brood comb within the

20 nesting cavity, and that neither used brood comb nor honey comb confers any immediate benefit on larval survival.

2.2 Introduction

Although it was once common for beekeepers to reuse brood comb for many seasons until it became damaged (Jaycox, 1979; Hoopingarner & Sanford, 1990), brood comb replacement is an increasingly popular practice among North American beekeepers. Driving the shortened replacement cycle is the potential for older brood combs to harbor pesticide residues (Johnson et al, 2010; Wu et al, 2011) and harmful microbes (Pankiw et al, 1970; Jaycox, 1979; Koenig et al, 1986; Di Pinto et al, 2011).

Additionally, old brood comb becomes layered with silken cocoons and larval excrement that are responsible for the darkened color and increased weight (Jay, 1963; Jay, 1964;

Hepburn & Kurstjens, 1988). As brood comb thickens, cell diameters shrink, resulting in smaller bees (McMullan & Brown, 2006). A reduction in body size could lead to less efficient honey production and pollination, as body size has been correlated with proboscis length (Waddington & Herbst, 1987) and foraging range (Greenleaf et al,

2007).

Also responsible for the darkening of brood comb is the build-up of propolis along the rims of cells (Strehle et al, 2003). In contrast to waste and contaminants, propolis build-up can benefit bee health. Honey bee health benefits include lower bacterial loads and immune gene expression, as well as higher colony survivorship

(Simone et al, 2009; Borba et al, 2015). The benefits of propolis are likely due to the antimicrobial effects of flavonoids and aromatic acids present in plant resins from which

21 propolis is derived (Marcucci, 1995). Another benefit of old brood comb is swarm attraction. Old brood comb has been used to attract swarms for years, and research shows that bees prefer nest spaces that hold previously used brood comb over nest spaces that hold no comb (Visscher et al, 1985). However, the ultimate reasoning behind this behavior is still unknown. Nest spaces holding old brood comb could be more conspicuous to bees looking for a new home, or they could be more attractive as the existing comb would save bees the energy investment required to synthesize new wax.

By removing potentially contaminated old combs, beekeepers are aiming to improve the health and survival of their colonies. However, there has been little research on frequency or effect of brood comb replacement outside of the data collected by The

Bee Informed Partnership (BIP) through an annual survey of U.S. beekeepers from 2008-

2019. Despite the perceived benefits of brood comb replacement, survey results demonstrate a positive relationship between the presence of older brood comb in apiaries and higher winter survival rates (Bee Informed Partnership, 2012a, 2012b, 2013, 2014,

2019). This relationship between comb age and colony survival could be due to genuine biological benefits conferred by older combs or may be confounded by the fact that more experienced beekeepers also tend to have older comb in their operations.

We designed an experiment to test honey bee preference for and larval survival on honey comb or used brood comb. To test honey bee preference for used brood comb, we installed artificial swarms, or packages, with a choice arena holding both honey comb and brood comb applied to plastic foundation. We hypothesized that colonies would preferentially establish on frames with brood comb treatment. We also designed an experiment to estimate larval survival on the different comb treatments. Due to the 22 ambiguity in brood comb replacement research, we hypothesized that larvae would exhibit differential survival when reared on honey comb or brood comb treatments.

Larval survival was estimated using a common visual estimation method, involving a

Liebfelder grid (Delaplane et al, 2013); however, this method is subjective and time- consuming. To test a new method of larval survival estimation, we also used digital photography (Booth et al, 2006) and image analysis software to count and compare the proportions of capped and uncapped cells within the brood area of each frame. Capped cells are those that contain pupating bees, while uncapped cells represent larvae that have died and been removed by worker bees. If successful, this method could save time and effort of researchers, and potentially provide a more precise estimate of larval survival.

2.3 Materials and Methods

2.3.1 Brood Comb Treatments

Previously drawn honey comb and darkened, used brood comb were collected to make foundation treatments. All comb came from apiaries managed at the Ohio

Agricultural Research and Development Center (OARDC) in Wooster, Ohio. The oldest frames in these apiaries are seven years old at most. Neither honey comb nor brood comb from these apiaries were treated with lipophilic miticides.

Using a clean hive tool, honey comb was scraped from plastic foundation and melted in water at up to 93°C (Crock-Pot; SKU: SCV800-B). Water was discarded, and wax was re-melted and filtered through a sieve and several layers of cheesecloth (Grade

40). The resulting product (Figure 2A) was divided and one portion was used as our honey comb treatment and the other as the base for our brood comb treatment. Due to the build-up of silken cocoons, used brood comb cannot be melted and filtered, so this 23 treatment was made by pulverizing darkened, used brood comb in a food processor

(Black and Decker, FP25000 PowerPro) with added liquid nitrogen, and then used brood comb powder was added to a portion of the melted honey comb treatment at a 1:2 (w/w) powder:wax (Figure 2C). The darkened comb-wax mixture was not heated above 77°C.

Both treatments were melted and applied to new deep plastic frames (Pierco, Riverside,

CA; SKU: 9FRBW) with a small paint roller at 4-6 grams per side (Figure 2D).

A) Melted and filtered honey comb as honey comb treatment; B) used brood comb scraped from brood frames; C) brood comb treatment made from melted honey comb and powdered used brood comb; D) plastic foundation frames treated with 8-12 g honey or brood comb.

Figure 2. Preparation of honey comb and brood comb treatments. 2.3.2 Choice Arena and Honey Bee Package Installation

Twenty-five standard 1.4 kg packages of bees headed by Italian or Carniolan queens (Rossman Apiaries, Moultrie, Georgia, USA) were installed in choice arenas among two apiaries—one at the Waterman Agriculture and Natural Resources Laboratory in Columbus, Ohio and the other at the OARDC in Wooster, Ohio. Packages were purchased and installed among three locations at four different time points in Spring

2018: ten in Columbus on April 12, three in Columbus on April 18, seven in Wooster on

April 26, and five in Columbus on May 7. Arenas were made of standard Langstroth eight-frame deep boxes (scraped clean of any propolis, wax, or debris) with a division

24 board sugar syrup feeder and three frames on each side of it (Figure 3A). On either side of the feeder, two of the frames were randomly treated with either honey comb or brood comb. The outermost frames on both sides were untreated new plastic foundation.

For package installation, the sugar syrup feeder and the two innermost frames were removed, and workers were shaken into the middle of the deep box (Figure 3B). Next, queens were released directly in the middle of the box (Figure 3C). Finally, feeders were filled with a 1:1 (w/w) sucrose:water solution to support colony establishment and wax production. Lids were replaced and colonies were allowed 72 hours before being reopened to record the side on which they had begun drawing wax in preparation for brood rearing. Seven days after package installation, frames were moved between colonies so that only frames with the chosen treatment were present in a particular hive.

For example, if the colony was established on honey comb treatment, we replaced the brood comb treated frames, and vice versa. Sugar syrup feeders were moved from the middle to the edge of the deep box and refilled with 1:1 sugar water solution.

A) choice arena made from clean eight-frame deep box, sugar syrup feeder, four treated frames and two untreated frames; B) shaking workers into middle of choice arena by removing feeder and two innermost frames; C) releasing queen directly into the middle of the choice arena.

Figure 3. Installing honey bee packages into choice arenas. 25 2.3.3 Larval Survival Estimation

Ten hives from Waterman Agriculture and Natural Resources Laboratory were removed from this experiment—eight of them due to a difference in methodology

(installed on April 12), and two of them due to queenlessness. One hive from OARDC was removed due to an error in data collection. To increase our sample size and statistical power, five new hives were added at Muck Crops Agricultural Research Station in

Willard, Ohio (installed on May 24). These hives were randomly installed on either all honey comb or all brood comb treatments.

Two methods were used to estimate larval survival: 1) visual estimation using the

Liebfelder grid (Figure 5B) (Delaplane et al., 2013) and 2) image analysis using particle counting with ImageJ software 1.52d (National Institutes of Health, Bethesda, Maryland)

(Figure 5D). Seven days after package installation, we used visual estimation (Figure 4A;

5B) to determine the number of eggs and young larvae. This cohort of brood was followed for another seven days to estimate survival on different comb treatments. Using visual estimation, we compared the area or estimated number of eggs and young larvae from day seven to the estimated number of capped brood on day fourteen.

Fourteen days after package installation digital photographs of brood frames were taken for image analysis (Figure 4B; 5D). We compared the number of uncapped cells

(presumed to have contained dead larvae that were removed by worker bees) to the number of capped cells in the circular area containing capped brood. To do this, images were loaded into ImageJ as 8-bit images with binary, black and white thresholds. Black and white scale was adjusted appropriately for each image, allowing the capped cells to appear white and the empty cells black. Next, the brood area was traced and selected, and 26 particles—or empty cells—were analyzed and counted for each image. Typically, cell size was adjusted to about 500 pixels and empty cells on the edge of the traced area were excluded from particle counting. By using the grid during photographing and counting the average number of cells within one square of the Liebfelder grid (Figure 5B), we could better estimate the proportion of larvae that did survive to capping, as well as the proportion of capped cells within the brood area. Figure 4 details the timeline of data collection for both methods.

A) Estimating the number of eggs and young larvae on day seven, and capped brood on day fourteen using the Liebfelder method. B) Photographing brood on day fourteen to estimate the number of uncapped and capped cells in the brood patch with ImageJ software.

Figure 4. Timeline of data collection for larval survival on unused or used brood comb treatment. 2.3.4 Statistical Analysis

A one-tailed binomial test was used to determine if preference for used brood comb exists. Visual estimates of larval survival were compared by a two-way mixed

ANOVA with repeated measures (eggs and young larvae on day 7; capped brood on day

14) between honey comb and brood comb treatments. Assumptions of normality, equal 27 variances and sphericity were tested with Shapiro-Wilk’s test, Bartlett’s test and

Mauchly’s test, respectively. For image analysis, the proportion of capped brood was logit transformed to achieve a normal distribution, and a two-tailed t-test was used to compare proportion of capped brood on honey comb and brood comb. Repeated measures ANOVA was not appropriate for this comparison, as the population of brood was only photographed once. Lastly, we used a two-tailed t-test to compare overall brood area on comb treatments. Brood area was calculated by adding the number of empty cells and the number of capped cells from the image analysis data set. All statistics were performed in R 3.5.0 (The R Foundation for Statistical Computing, Vienna, Austria).

2.4 Results

2.4.1 Choice Experiment

Of the 25 packages installed, thirteen colonies were established on honey comb and twelve packages were established on brood comb. Colonies showed no preference for honey comb or brood comb treated frames inside the hive (binomial test, p = 0.7).

2.4.2 Larval Survival

Survival rates of larval honey bees were estimated for 19 colonies. Using visual estimation, mean survival of eggs to capped brood was 67.2% ± 6.2% SEM (n = 9) and

70.5% ± 4.2% SEM (n = 10) on honey comb and brood comb treatments, respectively

(Figure 5A). The proportion of eggs and young larvae visually estimated to have survived to capping did not differ between hives on honey comb or brood comb treatment (two- way mixed ANOVA, F = 0.038, df = 1, p = 0.8). Using estimation through image analysis, mean proportion of capped brood within the brood area was 79.8% ± 4.6% SEM

28 (n =9) and 79.0% ± 10.1% SEM (n = 10) on honey comb and brood comb treatments, respectively (Figure 5C). The proportion of capped brood did not differ between honey comb and brood comb treatments (t-test, t = 0.4, df = 15.1, p-value = 0.7). Additionally, there was no difference in the total area of the brood patch between honey comb and brood comb (Figure 6; honey comb: mean = 3297.2 cells ± 547.7 cells SEM; brood comb: mean = 3981.3 cells ± 490.1 cells SEM; t-test, t = -0.9, df = 16.5, p = 0.4).

29

Average percentage of larvae survived to capping (honey comb = 67.2% ± 6.2% SEM; brood comb = 70.5% ± 4.2% SEM) estimated by visual grid inspection (B). C) Average percentage of capped brood (honey comb = 79.8% ± 4.6% SEM; brood comb = 79.0% ± 10.1% SEM) estimated by image analysis (D). Honey comb n = 9; brood comb n = 10.

Figure 5. Average larval survival on honey comb or brood comb estimated by visual inspection (A-B) and image analysis (C-D).

30

Honey comb: max = 6583, Q3 = 3878, median = 2951, Q1 = 2270, min = 1108, mean = 3297.2 ± 547.7 SEM. Brood comb: max = 6094, Q3 = 5221, median = 3972.5, Q1 = 2894.3, min = 1816, mean = 3981.3 ± 490.1 SEM.

Figure 6. Total brood area on honey comb and brood comb treatments.

2.5 Discussion

Previous research has shown that honey bee swarms prefer nest spaces with used brood comb over those that hold no brood comb (Visscher et al., 1985); however, we found no preference for brood comb over hoeny comb when presented inside the hive during package installation. This is in contrast to beekeeping anecdotes, as well as results of Visscher et al. (1985). Their results demonstrated a strong preference for nest spaces holding used brood comb; however, this was the preference of swarming colonies in search of a new nest space, and their preference was over hive spaces containing no comb

(Visscher et al., 1985). Therefore, our study does not refute the findings or hypotheses of previous research, since we only tested colonizing preference when bees were given a choice within a hive.

Visual estimation using a Liebfelder grid is a commonly used method for hive content assessment (Delaplane et al, 2013). Using this method, there was no observable difference between survival of eggs and young larvae to capping on honey comb or brood 31 comb treatments (Figure 5A). Our experimental method using digital photography and image analysis demonstrates the same result (Figure 5C). Standardization of a method such as this would be a useful tool in honey bee research, as it would save time, allow for data collection to be spread among several researchers, and potentially provide a more precise estimate of larval survival. Using a visual estimation is not only time consuming, but also subjective, thus requiring that the same researcher collect all data throughout an entire experiment. Additionally, to reduce the effects of handling bees and hive equipment, visual estimation is done relatively quickly by scanning and estimating the number of squares within the grid that are filled with eggs, larvae, or capped brood.

Taking a digital photo of the brood area would not only require less handling time, but it may also allow for a more precise estimate by using software to count the exact number of capped and uncapped cells in the photo. Precision of both methods needs to be tested by repeating a similar experiment and physically counting the number of capped and uncapped cells on frames to allow a comparison between physical counts and estimates made by both visual inspection and image analysis.

In addition to larval survival, we found no statistical difference between the areas of brood on honey comb and brood comb (Figure 5, t-test, p = 0.4). Through our experimental design, hives on honey comb and brood comb were forced to use these treatments as a foundation with which they would draw out their own comb—this design eliminates the factor of cell size when comparing used brood comb to fresh comb or honey comb treatments. Because old brood combs have layers of silk embedded in the cell walls, the cells inevitably shrink over time (Jay, 1936; Hepburn & Kurstjens, 1988).

By eliminating the factor of cell diameter, we can conclude from this study that 32 constituents of used brood comb have no effect on the area of brood produced by the queen. Berry & Delaplane (2001) measured a significantly larger area of brood on previously constructed fresh comb compared to previously constructed and used brood comb. Researchers hypothesized that the difference in brood area was likely due to the decrease in cell diameter (0.3 mm smaller) of used brood comb and the queen’s difficulty measuring cell width before laying a worker or drone egg (Koeniger, 1970).

Overall, the biological effect of used brood comb on larval honey bee survival is ambiguous. Our study demonstrates no evidence for benefit by used brood comb not exposed to lipophilic acaricides. However, previous research has shown that used brood combs exposed to lipophilic pesticides lead to delayed larval development and decreased adult longevity (Wu et al, 2011). Residues accumulated most frequently and in the highest amounts were those of common beekeeper-applied acaricides, tau-fluvalinate

(Apistan®), coumaphos and coumaphos oxon metabolite (Checkmite+®) (Wu et al.,

2011). This previous research, along with our results, supports the notion that beekeepers using synthetic organic acaricides, such Apistan®, Checkmite+®, ApiVar® and

Amitraz®, should continue replacing brood combs on a regular basis. Additionally, Wu et al. (2011) demonstrated that acaricide residues quickly migrated from old brood combs to newly drawn control combs used in their experiment, meaning that leaving even one treated frame in the hive can potentially contaminate any other brood or honey frames.

Alternatively, natural products and organic acid acaricides, such as Apilife Var®,

Apiguard® and MiteAway II®, accumulate much less in the wax comb (Johnson et al,

2010; Wu et al, 2011). Our results demonstrate the beekeepers applying acaricides such as these may not benefit as much from frequently replacing brood combs. Notably, we 33 did not account for microbial build-up or shrinking cells in this experiment, and these are also important factors to consider when deciding how frequently brood comb should be replaced. Previous research shows that colonies reared on old brood comb develop significantly higher rates of chalkbrood infection than colonies reared on either freshly drawn honey comb or empty foundation frames (Koenig et al., 1986). Finally, smaller cell sizes produce smaller adults bees (McMullan & Brown, 2006); however, the frequency with which brood combs need to be replaced to avoid this outcome has yet to be researched. McMullan and Brown (2006) have shown that a cell size reduction of 0.5 mm produced smaller honey bee features including head width, tracheal diameter and mass. In conclusion, the frequency with which beekeepers replace their brood combs will largely depend on apiary management and practices. Beekeepers using synthetic organic acaricides should continue regular replacement, while those using natural and organic acaricides may consider reusing brood combs for a longer period. All beekeepers should practice sterile techniques within the apiary to avoid spreading microbial diseases and replace brood combs after any such outbreak to avoid harboring any fungal or bacterial spores. Aside from pesticide residue and microbial disease build-up, brood combs should also be replaced after the cell walls have thickened by about 0.5 mm.

2.6 Acknowledgements

I would like to thank Jacob Shuman and Jessica Lyons for their assistance in preparing hive equipment and collecting data. Additionally, I would like to thank Robert Filbrun for providing field space for colonies at Muck Crops Agricultural Research Station.

2.7 References

Bee Informed Partnership. (2012a). The Bee Informed Partnership Management Survey 34 Results (2011) Brood Comb Managment. Retrieved from https://beeinformed.org/wp-content/uploads/2012/04/Brood-Comb-managment.pdf

Bee Informed Partnership. (2012b). The Bee Informed Partnership Management Survey Results (2012-2013) Brood Comb Management. Retrieved from https://beeinformed.org/wp-content/uploads/2014/11/Comb-Management- Summary.pdf

Bee Informed Partnership. (2013). The Bee Informed Partnership Management Survey Results (2012) Brood Comb Managment. Retrieved from https://beeinformed.org/wp-content/uploads/2013/08/title-Brood-Comb- managment.pdf

Bee Informed Partnership. (2014). The Bee Informed Partnership Management Survey Results (2013-2014) Brood Comb Management. Retrieved from https://beeinformed.org/wp-content/uploads/2014/09/Comb-Management- Summary.pdf

Bee Informed Partnership. (2019). Average Winter Loss per Beekeeper. Retrieved from https://bip2.beeinformed.org/survey/

Berry, J. A., & Delaplane, K. S. (2001). Effects of comb age on honey bee colony growth and brood survivorship. Journal of Apicultural Research, 40(1), 3–8.

Booth, D. T., Cox, S. E., Meikle, T. W. & Fitzgeral, C. (2006). The accuracy of ground- cover measurements. Rangeland Ecological Management, 59, 179-188.

Borba, R. S., Klyczek, K. K., Mogen, K. L., & Spivak, M. (2015). Seasonal benefits of a natural propolis envelope to honey bee immunity and colony health. Journal of Experimental Biology, 218(22), 3689–3699.

Delaplane, K. S., Van Der Steen, J., & Guzman-Novoa, E. (2013). Standard methods for Apis mellifera research. Journal of Apicultural Research, 52(1), 1–12.

Di Pinto, A., Novello, L., Terio, V., & Tantillo, G. (2011). ERIC-PCR genotyping of Paenibacillus larvae in Southern Italian honey and brood combs. Current Microbiology, 63(5), 416–419.

Greenleaf, S. S., Williams, N. M., Winfree, R., & Kremen, C. (2007). Bee foraging ranges and their relationship to body size. Oecologia, 153(3), 589–596.

Hepburn, H. R., & Kurstjens, S. P. (1988). The combs of honeybees as composite materials. Apidologie, 19(1), 25–36.

Hoopingarner, R., & Sanford, M. T. (1990). The costs of beekeeping—I. Survey of commercial beekeepers. American Bee Journal, 130(6), 405–407.

35 Jay, S. C. (1964). The cocoon of the honey bee, Apis mellifera L. The Canadian Entomologist, 96(5), 784–792.

Jay, S. C. (1963). The development of honeybees in their cells. Journal of Apicultural Research, 2(2), 117–134.

Jaycox, E. R. (1979). Comb foundation: Are we using enough? American Bee Journal, 119(7), 515–516.

Johnson, R. M., Ellis, M. D., Mullin, C. A., & Frazier, M. (2010). Pesticides and honey bee toxicity – USA. Apidologie, 41(3), 312–331.

Koenig, J. P., Boush, G. M., & Erickson, E. H. (1986). Effect of type of brood comb on chalk brood disease in honeybee colonies. Journal of Apicultural Research, 25(1), 58–62. https://doi.org/10.1080/00218839.1986.11100694

Koeniger, N. (1970). Factors determining the laying of drone and worker eggs by the queen honeybee. Bee World, 51, 166-169.

Marcucci, M. C. (1995). Propolis: chemical composition, biological properties and therapeutic activity. Apidologie, 26(1), 83–99.

McMullan, J. B., & Brown, M. J. F. (2006). The influence of small-cell brood combs on the morphometry of honeybees (Apis mellifera). Apidologie, 37(6), 665–672.

Pankiw, P., Bailey, L., Gochnauer, T. A., & Hamilton, H. A. (1970). Disinfection of Honeybee Combs by Gamma Irradiation. II. European Foul Brood Disease. Journal of Apicultural Research, 9(3), 165–168.

Simone, M., Evans, J. D., & Spivak, M. (2009). Resin collection and social immunity in honey bees. Evolution, 63(11), 3016–3022.

Strehle, M. A., Jenke, F., Fröhlich, B., Tautz, J., Riederer, M., Kiefer, W., & Popp, J. (2003). Raman spectroscopic study of spatial distribution of propolis in comb of Apis mellifera carnica (Pollm.). Biopolymers - Biospectroscopy Section, 72(4), 217– 224.

Visscher, P. K., Morse, R. A., & Seeley, T. D. (1985). Honey bees choosing a home prefer previously occupied cavities. Insectes Sociaux, 32(2), 217–220.

Waddington, K. D., & Herbst, L. H. (1987). Body size and the functional length of the proboscis of honey bees. The Florida Entomologist, 70(1), 124–128.

Wu, J. Y., Anelli, C. M., & Sheppard, W. S. (2011). Sub-lethal effects of pesticide residues in brood comb on worker honey bee (Apis mellifera) development and longevity. PLoS ONE, 6(2), e14720.

36 Chapter 3. The effects of propolis and brood comb extracts on the causative agent of European Foulbrood (Melissococcus plutonius) in honey bees

3.1 Abstract

Surveys of beekeepers in the U.S. show that colonies living on older, darkened brood comb survive winters better than those living on comb that has been replaced. One explanation for the increased survival observed on old brood comb is the build-up of propolis—a resinous, antimicrobial substance collected by bees from tree buds—in old brood comb. The presence of propolis in old brood comb has been noted in previous research, especially concentrated around the rims of wax cells. To test for beneficial properties of propolis and comb used for brood rearing on honey bee health, I exposed three bacterial species to two types of ethanol extracts of (1) propolis, (2) dark comb previously used for brood rearing and (3) light comb previously used for honey storage, but not brood rearing. Bacterial species tested include two aerobic model species

(Escherichia coli and Staphylococcus saprophyticus) and one anaerobic species

(Melissococcus plutonius) that causes European Foulbrood (EFB) disease in larval honey bees. Bacterial growth in the presence of extracts was measured by optical density for twelve hours, so that any inhibition could be recorded and used to find the minimum inhibitory concentration (MIC) of active extracts. Escherichia coli growth was not inhibited by any treatments. Propolis extracts showed some inhibition of S. saprophyticus growth and strong inhibition of M. plutonius growth. Comb used for both brood rearing

37 and food storage also showed inhibition of M. plutonius growth at high concentrations.

The active components of brood comb and honey comb are currently unknown and should be further examined. Propolis supplementation should be further examined as a potential bee-safe and food-safe method for EFB treatment.

3.2 Introduction

Melissococcus plutonius is the bacterium responsible for the larval honey bee disease known as European Foulbrood (EFB) (Bailey, 1957; 1983, Bailey & Collins,

1982). Larvae infected with M. plutonius pass on the bacterium either through defecation or dying within their wax cells. Bacterial cells are then spread to healthy larvae by adult bees through wax cell cleaning and larval care, such as feeding (Bailey, 1983). Infection causes discoloration and loss of internal body pressure, leaving brood cells filled with flaccid, yellow/brown larvae (Forsgren et al, 2013). Though not as threatening as

American Foulbrood (AFB) (Hansen & Brodsgaard, 2015), EFB is often responsible for large brood losses during mid-summer months in North America and Europe (Bailey,

1983; Forsgren et al, 2013). To help prevent the spread of EFB and other microbial diseases, beekeepers in the U.S. have recently started removing and replacing older, used brood combs on a regular basis, similar to practices of European beekeepers (Jaycox,

1979). Although the intention is to improve colony health and survival, research and surveys on the topic of brood comb replacement demonstrate that it may not be the most effective practice for every apiary. Berry and Delaplane (2001) demonstrated better brood survival on old brood comb. Additionally, national surveys of beekeeper practices and success rates performed by the Bee Informed Partnership (BIP) (2012; 2013a; 2013b;

38 2014; 2015; 2019) indicate a positive relationship between old brood comb and colony survival over the winter.

Despite the potential for microbial build-up, old brood comb also contains a build-up of propolis (Strehle et al, 2003). Though it is mostly known for its use in sealing cracks within the hive structure, propolis has antibacterial qualities as well. Propolis is a resinous substance that bees collect from plant bud exudates (Marcucci, 1995). There has been considerable research into using bee propolis to treat bacterial infections in humans

(Marcucci, 1995), and honey bee colonies supplemented with more propolis exhibit lower virus levels and lower expression of individual immune genes and antibacterial peptides (Simone et al, 2009; Borba et al, 2015; Drescher et al, 2017). Recent analysis of wax combs shows that propolis is also concentrated around the rims of wax cells (Strehle et al, 2003). Many of the compounds present are responsible for its antibacterial properties—these include flavonoids, aromatic acids, hydroxy acids, esters and terpenes

(Marcucci, 1995); however, propolis composition can differ greatly with location and plant origin (Huang et al, 2014). Propolis collected from temperate climate regions largely originates from Populus species (Andelković et al, 2017), and samples from Ohio have previously shown presence of many flavonoids and phenolics (Johnson et al, 1994).

To explain the observed benefits of old brood comb, we wanted to test the activity of propolis and used brood comb against the honey bee pathogen, M. plutionius. To test for general antimicrobial effects of propolis and comb used for brood rearing, we exposed three bacterial species to two types of ethanol extracts of lightly-colored honey comb, darkened brood comb, and propolis collected from the same source hives, and determined the minimum inhibitory concentration (MIC) of each. The MIC is the lowest 39 concentration of treatment that successfully inhibits growth (Kowalski et al, 2005;

Jorgensen & Ferraro, 2009; Vassalo et al, 2018). We hypothesize that the MIC will be lowest for propolis extracts and highest for honey comb extracts. In other words, propolis will be most effective at inhibiting bacterial growth and honey comb will be least effective.

Bacterial species used include M. plutonius, as well as two model bacteria,

Escherichia coli and Staphylococcus saprophyticus. The latter species are not associated with honey bee disease; however, they are pathogenic bacteria in mammals, are relatively easy to cultivate and are frequently used in antimicrobial experiments (Feng, Weagant &

Grant, 1998 in Rahman, Richardson & Sofian-Azirun, 2010). Model bacteria are useful because M. plutonius is a fastidious, anaerobic bacterium that requires unique growth media (Bailey & Collins, 1982; Hornitzky & Smith, 1998; Forsgren et al, 2013; Nakai et al, 2018). Additionally, M. plutonius is a gram-positive bacterium, meaning the cells are composed of one plasma membrane and a thick layer of peptidoglycans (Harrop,

Hocknull & Lilly, 1989; Forsgren, 2010; Dicks, Endo & Reenan, 2014; Forsgren et al,

2013). To account for differences between inhibition of gram-positive and gram-negative bacteria, our model bacteria represent both cell types, whereas S. saprophyticus is also gram-positive and E. coli is gram-negative (Percival & Williams, 2014). Gram-negative bacterial cells are characterized by a thin layer of peptidoglycans and an extra plasma membrane (Harrop et al, 1989). It is likely due to this extra membrane that propolis extract is not typically an effective inhibitor of E. coli growth (Tosi et al, 1996; Mirzoeva,

Grishanin & Calder, 1997; Sforcin et al, 2000; Rahman et al, 2010).

40 Previous experiments by Ristivojevic et al. (2013; 2016) demonstrated successful inhibition of E. coli using a propolis extract resuspended in methanol. However, ethanol extracts are most typically used in propolis research, as it is often aimed at human use

(Marcucci, 1995; Ahmed, 1995) and many of the antimicrobial compounds present, such as flavonoids and aromatic acids, are soluble in ethanol (Marcucci, 1995). For these reasons, we decided to test both ethanol extracts, as well as ethanol extracts resuspended in methanol.

In addition to hive health benefits, propolis is a natural component of the hive environment, is plant-derived and would likely be safe to use when bees are collecting honey intended for human consumption. This is in contrast to existing medicines for

EFB, such as Terramycin®, or oxytetracycline hydrochloride (OTC)—the current most effective treatment for EFB (Wilson, 1974; Gilliam & Argauer, 1981; Sporns et al, 1986;

Matsuka & Nakamura, 1990; Forsgren, 2010). If antimicrobial compounds present in propolis or used brood comb are effective inhibitors of M. plutonius growth, they could be potential food-safe treatments for EFB.

3.3 Materials and Methods

3.3.1 Comb and Propolis Samples

To collect comb and propolis, two experimental colonies were set up in standard

Langstroth equipment at each of three Ohio apiaries in spring of 2018. Apiaries are located at (1) the Waterman Agriculture and Natural Resources Laboratory in Columbus

(40.0104° N, 83.0400° W), (2) the Ohio Agricultural Research and Development Center in Wooster (40.7818° N, 81.9305° W), and (3) the Muck Crops Agricultural Research

Station in Willard (41.010200° N, 82.731350° W). New or thoroughly cleaned hive 41 equipment was used at each site to provide darkened brood comb, lightly-colored honey comb, and propolis, which were harvested from each site in March 2019. A metal queen excluder was used to restrict brood rearing to the bottom box so that the upper box would contain only honey comb (Figure 7). Comb and propolis samples were stored at -20°C.

3.3.2 Comb and Propolis Extractions

We performed two types of ethanol extractions on all comb and propolis samples harvested. Hive products were first pulverized using liquid nitrogen and a food processor

(Black & Decker FP2500B). A separate food processor was used for each type of hive product (honey comb, brood comb or propolis) to prevent cross contamination. Two grams of each pulverized sample were placed in 20 mL of 95% ethanol (1:10 w/v) and sonicated (Kendal HB-S-49DHT) for 60 minutes at up to 60°C. The resulting extract was filtered, evaporated and resuspended in 10 mL of either 70% ethanol or pure methanol. In total, 18 extracts were used: ethanol-suspended and methanol-suspended extracts from brood comb, honey comb and propolis derived from colonies at each of three locations.

Extracts were stored in darkness at 4°C.

3.3.3 Bacterial Cultures

Melissococcus plutonius was isolated from European Foulbrood-infected larvae at the Ohio Agricultural Research and Development Center in Wooster, Ohio in Spring

2018. This anaerobic bacterium was isolated and cultured by inoculating modified basal medium (Forsgren et al, 2015) with infected larvae that were ground and suspended in saline solution. Growth medium contained yeast extract, glucose, sucrose, L-cysteine, 1M

KH2PO4 and 2.5M KOH (Fisher Scientific: Mfr. #AC400405000, D16-500, P285-500 and Sigma Aldrich: Mfr. #P5958-250G, W326305-100G). To prevent the growth of 42 secondary bacteria, nalidixic acid was dissolved in 0.1M NaOH (Fisher Scientific: Mfr.

#AAB2509606, S318-100), filter sterilized, and added to the medium at a concentration of 3 µg/ml. Melissococcus plutonius was cultivated in an anerobic chamber at 37°C.

Cultivation protocol is detailed in Table 1. Sequencing confirmed that our culture matched to two GenBank entries for M. plutonius: CP006683 and HF569134 (Table 2).

Cultures were mixed with 50% glycerol for long-term storage at -80°C. Strains of two aerobic bacterial species, E. coli and S. saprophyticus, were cultivated in Mueller-Hinton media (Fisher Scientific: Mfr. #OXCM0405B) at 37°C and temporarily stored at 4°C to delay growth, as needed.

3.3.4 Bacterial Growth Tests

Bacteria were grown in 96-well plates to test inhibitory properties of extracts.

First, an eight-step two-fold dilution series was made for each extract. Test wells contained a final volume of 200 µL, made up of 190 µL of growth medium, 5 µL of bacterial culture standardized to 5 x 105 CFU/mL (Reller et al, 2009) and 5 µL of extract.

Positive control and solvent control wells contained growth media, culture and 5 µL of oxytetracycline (Forsgren, 2010) or solvent, respectively. Sterility control wells contained

200 µL of growth media only. Finally, wells used for standard growth contained 195 µL of growth media and 5 µL of bacterial culture standardized to 5 x 105 CFU/mL.

Following methods for standard growth inhibition assays (Reller et al, 2009), the microplate reader (Biotek 800 TSI) maintained incubation at 37°C and was set to shake and record absorbance or optical density (OD) at 600 nm every 5 minutes for 12 hours. A

12-hour assay provided sufficient time to record the mid log-phase, or mid growth-phase of all bacteria. The average time point at which the untreated wells reached mid log- 43 phase for each plate was used to determine inhibition of bacterial growth for each treatment. The lowest concentration of each treatment that inhibited growth—or was significantly different from the average OD of the plate solvent controls—was considered the MIC for each bacterial species. Propolis dilutions 1-5 were excluded from analyses due to oversaturation from extracts, causing too much noise in OD readings. For every extract and bacterium tested, there were three biological replicates (from three different apiaries) and three technical replicates of each (n = 9).

3.3.5 Statistical Analysis

The difference in OD between treatments and solvent controls at the time of mid- log phase was used to identify inhibition of bacterial growth for each species.

Specifically, we used Kruskal-Wallis and Dunn’s Multiple Comparison post-hoc test with

Bonferroni correction to identify the MICs or OD readings that were significantly different from solvent control (Kowalski et al, 1987; Vassallo et al, 2018). Kruskal-

Wallis was used due to a non-normal data set (Shapiro-Wilk’s test), and Dunn’s Multiple

Comparison was used to test for differences between the OD for each treatment and the solvent controls. All statistics were performed in R 3.5.0 (The R Foundation for

Statistical Computing, Vienna, Austria).

44

Two deep boxes (A) separated by a queen excluder (B), keeping brood comb (D) in the lower box and honey comb (C) in the upper box. Propolis (E) was collected from hive crevices.

Figure 7. Experimental colony design.

45 Table 1. Protocol for Melissococcus plutonius cultivation.

Growth media adapted from Bailey, 1957 and Forsgren et al, 2013. All steps were performed using sterile equipment and in a sterile environment, including a biosafety cabinet or an anerobic chamber.

Growth Media (makes 1 L) Cultivation in Growth Media

1. Mix the following into 800 mL DI 1. Use sterile pestle to prepare H20: suspension of 2-3 diseased brood a. 10 g yeast extract in saline and transfer suspension b. 10 g glucose to anaerobic chamber c. 10 g starch (solid medium) or 2. Inoculate growth medium: sucrose (liquid medium) a. Add ~20 µL suspension to 20 d. 0.25 g L-cysteine mL liquid growth medium OR e. 20 g agar (solid medium only) b. Pipette ~20 µL suspension 2. Add 100 mL 1M KH2PO4 onto agar plate and spread 3. Adjust pH to 6.6 using 2.5M KOH with sterile plate spreader 4. Adjust final volume to 1000 mL 3. Incubate in anaerobic conditions using distilled water at 37°C and check daily 5. Autoclave for 15 min at 115°C 6. Dissolve nalidixic acid into 0.1M NaOH (30 mg/mL) and filter sterilize (store at -20°C) 7. After cooling media, add filter sterilized nalidixic acid at a final concentration of 3 µg/mL 8. Transfer to anaerobic chamber

Table 2. Matching GenBank sequence descriptions.

Sequence matches for our bacterial sample can be located using the accession number at https://www.ncbi.nlm.nih.gov/genbank/. Published sequences (Haynes, et al. 2013) are referenced in the bibliography.

Accession Organism Sequence Authors Published? Description CP006683 M. plutionus Complete genome Djukic, M. et al. No HF569134 M. plutonius Partial galK gene Haynes, E. et al. Yes

3.4 Results

Escherichia coli was not significantly inhibited by any treatments, including the positive control (OTC) (Figure 8; OTC vs. EtOH and MeOH control p > 0.05); however, 46 OD of E. coli treated with OTC was significantly lower than the OD of the untreated standard growth well (p < 0.001). Staphylococcus saprophyticus was successfully inhibited by OTC and the highest concentration of ethanol-suspended propolis extract

(Figure 9; OTC vs. EtOH and MeOH control p = 0.003; 0.03 propolis EtOH vs. EtOH control p = 0.02). Therefore, the MIC of ethanol-suspended propolis extract against S. saprophyticus is 0.03 or 3% crude extract. Finally, M. plutonius was successfully inhibited by OTC and some dilutions of each hive product treatments (Figure 10; OTC vs. EtOH and MeOH control p < 0.001). The MIC of ethanol-suspended honey comb extract is 0.5 or 50% crude extract (Figure 10A; p < 0.001) and the MIC of methanol- suspended honey comb extract is 1.0 or 100% crude extract (Figure 10B; p < 0.001). The

MIC of ethanol-suspended brood comb extract is 50% crude extract (Figure 10C; p <

0.001) and the MIC of the methanol-suspended brood comb extract is 100% (Figure 10D; p < 0.001). The MIC of ethanol and methanol-suspended propolis extracts against M. plutonius is 0.02 or 2% (Figure 10E-F; p < 0.001). Although the range of differences in

OD relative to the solvent controls is larger for some ethanol-suspended extracts than for methanol-suspended extracts, there were no statistical differences between hive products between the two solvents (p > 0.05).

47

No effect of ethanol (A) or methanol (B) honey comb treatments (p > 0.05); no effect of ethanol (C) or methanol (D) brood comb treatments (p > 0.05); no effect of ethanol (E) or methanol (F) propolis treatments (p > 0.05); no effect of oxytetracycline (OTC) (p > 0.05).

Figure 8. Difference in optical density (OD) of E. coli growth relative to the solvent control for all ethanol and methanol-suspended treatments.

48

No effect of ethanol (A) or methanol (B) honey comb treatments (p > 0.05); no effect of ethanol (C) or methanol (D) brood comb treatments (p > 0.05); inhibition of growth by 0.03 (3%) propolis extract (E, p = 0.02); no effect of methanol propolis treatments (F, p > 0.05); inhibition of growth by oxytetracycline (OTC) (p = 0.003). Asterisk (*) indicates significant difference from zero OD.

Figure 9. Difference in optical density (OD) of S. saprophyticus growth relative to the solvent control for all ethanol and methanol-suspended treatments.

49

Inhibition of growth by 0.5 (50%) ethanol honey comb extract (A, p < 0.001) and undiluted methanol honey comb extract (B, p < 0.001); Inhibition of growth by 50% brood comb extract (C, p < 0.001) and undiluted methanol brood comb extract (D, p < 0.001); inhibition of growth by 0.02 (2%) ethanol (E) and methanol (F) propolis extracts (p < 0.001); inhibition of growth by oxytetracycline (OTC) (p < 0.001). Asterisk (*) indicates significant difference from zero OD.

Figure 10. Difference in optical density (OD) of M. plutonius growth relative to the solvent control for all ethanol and methanol-suspended treatments.

50 3.5 Discussion

The lack of effect of hive products on E. coli was not surprising, as many previous studies have shown similar results (Tosi et al, 1996; Mirzoeva et al, 1997;

Sforcin et al, 2000; Rahman et al, 2010). Although we did model our methanol- suspended extractions after a study that resulted in successful inhibition of E. coli by propolis extract (Ristivojevic et al, 2013; 2016), we saw no significant inhibition of E. coli with any of our ethanol or methanol-suspended hive product extracts. Figure 8 demonstrates some inhibition of E. coli by OTC; however, the median OD difference compared to solvent controls is around -0.09 and was not statistically different from the solvent control. The positive control did inhibit growth compared to our untreated standard growth wells, but it was not significantly different from the level of inhibition imposed by the solvent controls. While Ristivojevic et al. (2013; 2016) did see some inhibition of E. coli with methanol propolis extracts, the MICs of this species and other gram-negative bacteria were much higher than those of the gram-positive bacteria tested.

Previous research, as well as our results, support the hypothesis that antimicrobial activity of propolis is less effective against E. coli due to its cellular structure. Additionally, OTC is not typically used as a positive control for experiments on E. coli inhibition. For example, Ristivojevic et al. (2013; 2016) used three positive controls including streptomycin, ampicillin and rifampicin. OTC was used in our experiment due to its well- established activity against our bacterium of interest, M. plutonius (Forsgren, 2010); however, the effectiveness of OTC and other positive controls against E. coli may differ due to mechanisms of action or bacterial resistance (Luzzatto et al, 1968; Wehrli, 1983;

Chopra & Roberts, 2001). OTC belongs to a group of broad-spectrum tetracyclines that 51 were previously effective against both gram-positive and gram-negative bacteria (Chopra

& Roberts, 2001).

Staphylococcus saprophyticus was more susceptible to inhibition by OTC, as well as by propolis extracts. Previous research shows strong activity of propolis against gram- positive bacteria relative to that of gram-negative bacteria. Low concentrations of propolis are particularly active against S. aureus (Mirzoeva et al, 1997; Sforcin et al,

2000; Rahman et al, 2010), a bacterium belonging to the same genus. In our study, ethanol-suspended propolis extract was active against S. saprophyticus at a relatively low concentration of 32-fold dilution. We observed no inhibition of S. saprophyticus by the methanol-suspended propolis extract; however, we did exclude 5 of the higher concentrations tested due to oversaturation within test wells that made OD readings hard to interpret. It is likely that some of these higher concentrations of methanol-suspended propolis extract inhibit the growth of S. saprophyticus.

Overall, M. plutonius was the most susceptible to inhibition by hive product extracts and OTC. OTC is the current most effective treatment for M. plutonius. Our results suggest that bees are surrounding themselves with antimicrobial compounds that are not only present in propolis, but also in the honey comb and brood comb. Active compounds of both brood and honey comb are not yet confirmed, but a study by Strehle et al (2003), as well as beekeeping anecdotes, suggest that propolis is often present around the rims of wax cells. With little to no evidence, beekeepers and researchers alike often refer to propolis presence in old brood comb as one of the factors responsible for its change in color. However, there is no research comparing the presence of propolis between honey comb and brood comb cells. Researchers have hypothesized that propolis 52 is added around wax cells to repair comb or add structural stability (Darchen, 1968;

Chauvin, 1992 in Strehle et al, 2003), but it is unknown whether this stability is preferred in one comb type over the other.

Additionally, there are antimicrobial properties of other bee products, such as the honey and beeswax themselves. Like propolis, honey has been used in human medicine for millennia; however, its antimicrobial benefits stem more from the enzymes present— some of which originate from plants and others from honey bee hypopharyngeal glands

(Aurongzeb & Azim, 2011). Honey has shown inhibitory properties toward many gram- negative and gram-positive bacteria (Aurongzeb & Azim, 2011). Studies on the antimicrobial effects of beeswax are lacking compared to those of propolis and honey.

Beeswax is made up of mostly hydrocarbon chains, free fatty acids, free fatty alcohols and wax esters (Fratini et al, 2016). Additionally, compounds from plant materials, such as pollen or propolis, are also built-up in the wax combs, and the origins of antimicrobial benefits have yet to be studied. However, Fratini et al (2016) reviewed some research that has shown inhibitory effects of crude beeswax extracts, as well as beeswax in concert with other natural products like propolis or olive oil, on both gram-negative and gram- positive bacteria. A study on antimicrobial activity of crude beeswax extract took samples from wax combs that were 1-2 years old, but there was no mention of what the comb had been used for, or if propolis was present in the wax (Kacániová et al, 2012).

Further research should be done to elucidate the antibacterial effects of beeswax.

Using various extraction solvents, researchers can target specific active compounds according to their solubility; and further fractionation through GC-MS or HPLC-MS will allow compound identification (Bruschi et al, 2003). As the antibacterial properties of 53 propolis are well-established and demonstrate activity against M. plutonius, propolis should be further examined as a treatment for European Foulbrood (EFB) infection in honey bee colonies. There is an existing body of literature supporting the use of propolis in honey bee immunity, as bees selected for propolis-collection demonstrate increased survivorship and longevity (Nicodemo et al, 2014), and colonies supplemented with propolis exhibit lower infection rates and immune gene expression (Drescher et al, 2017;

Simone et al, 2009; Borba et al, 2015). Aside from breeding bees for increased propolis- collection, two of the most practical ways to supplement colonies with propolis include applying an extract or stimulating propolis collection through hive design (Borba et al,

2015; Drescher et al 2017). Finally, if medication by propolis is an effective control method for EFB infection, propolis treatments could be used in hives producing honey for human consumption. In contrast to OTC, propolis is a natural, plant-based substance that is already present within the hive. OTC residues present in honey of treated hives prevents beekeepers from producing honey for human consumption during EFB treatment (Wilson, 1974; Gilliam & Argauer, 1981; Sporns et al, 1986; Matsuka &

Nakamura, 1990). Further experiments should treat EFB-infected hives in the field or

EFB-infected larvae in-vitro with topical or oral propolis extracts to observe larval survival and infection rates.

3.6 Acknowledgements

I would like to thank all those that made my research possible through donation of space, equipment, time and expertise. First, I would like to acknowledge Dr. Christopher

Okonkwo, Dr. Christopher Madden, Colin Kurkul and Andrew Mularo for their expertise and assistance in bacterial cultivation, susceptibility testing and data collection. Tyler 54 Eaton sequenced and confirmed our M. plutonius isolate. Dr. Rachelle Adams provided lab space, equipment and aerobic bacterial strains. Dr. Vanessa Hale and Dr. Thaddeus

Ezeji provided lab space and equipment for anaerobic cultivation. Robert Filbrun provided field space for colonies at Muck Crops Agricultural Research Station. Finally,

Sreelakshmi Suresh assisted in data transcription.

3.7 References

Ahmed, F. E. (1995). Toxicological effects of ethanol on human health. Critical Reviews in Toxicology, 25(4), 347-367.

Andelković, B., Vujisić, L., Vučković, I., Tešević, V. & Vajs, V. (2017). Metabolomics study of Populus type propolis. Journal of Pharmaceutical and Biomedical Analysis, 135, 217-226.

Aurongzeb, M. & Azim, M. K. (2011). Antimicrobial properties of natural honey: a review of literature. Pak. Journal of Biochem. Mol. Biol. 2011, 44(3), 118-124.

Bailey, L. & Collins, M. D. (1982). Reclassification of ‘Streptococcus pluton’ (White) in a new genus Melissococcus, as Melissococcus pluton nom, rev; comb. nov. Journal of Applied Bacteriology, 53, 215-217.

Bailey, L. (1957). The cause of European foul brood. Bee World, 38, 95-89.

Bailey, L. (1983). Melissococcus pluton, the cause of European foulbrood of honey bees (Apis spp.). Journal of Applied Bacteriology, 55, 65-69.

Bee Informed Partnership. (2012a). The Bee Informed Partnership Management Survey Results (2011) Brood Comb Management. Retrieved from https://beeinformed.org/wp-content/uploads/2012/04/Brood-Comb- managment.pdf

Bee Informed Partnership. (2012b). The Bee Informed Partnership Management Survey Results (2012-2013) Brood Comb Management. Retrieved from https://beeinformed.org/wp-content/uploads/2014/11/Comb-Management- Summary.pdf

Bee Informed Partnership. (2013). The Bee Informed Partnership Management Survey Results (2012) Brood Comb Management. Retrieved from https://beeinformed.org/wp-content/uploads/2013/08/title-Brood-Comb- managment.pdf

55 Bee Informed Partnership. (2014). The Bee Informed Partnership Management Survey Results (2013-2014) Brood Comb Management. Retrieved from https://beeinformed.org/wp-content/uploads/2014/09/Comb-Management- Summary.pdf

Bee Informed Partnership. (2019). Average Winter Loss per Beekeeper. Retrieved from https://bip2.beeinformed.org/survey/

Berry, J. A., & Delaplane, K. S. (2001). Effects of comb age on honey bee colony growth and brood survivorship. Journal of Apicultural Research, 40(1), 3–8.

Borba, R. S., Klyczek, K. K., Mogen, K. L. & Spivak, M. (2015). Seasonal benefits of a natural propolis envelope to honey bee immunity and colony health. Journal of Experimental Biology, 218, 3689-3699.

Bruschi, M. L., Franco, S. L. & M. Gremião, P. D. (2003). Application of an HPLC method for analysis of propolis extract. Journal of Liquid Chromatography & Related Technologies, 26(14), 2399–240.

Chopra, I. & Roberts, M. (2001). Tetracycline antibiotics: Mode of action, applications, molecular biology, and epidemiology of bacterial resistance. Microbiology and Molecular Biology Reviews, 65(2), 232-260.

Dicks, L. M. T., Endo, A. & Van Reenan, C. A. (2014). ‘Minor Genera of the Enterococcaceae (Catellicoccus, Melissococcus, and Pilibacter)’ in Lactic Acid Bacteria: Biodiversity & Taxonomy. John Wiley & Sons: Hoboken, NJ, USA. Pp 239-243.

Drescher, N., Klein, A., Neumann, P., Yañez, O. & Leonhardt, S. D. (2017). Inside honeybee hives: Impact of natural propolis on the ectoparasitic mite Varroa destructor and viruses. Insects, 8(15).

Forsgren, E. (2010). European foulbrood in honey bees. Journal of Invertebrate Pathology, 103, S5-S9.

Forsgren, E., Budge, G. E., Charriere, J., & Hornitzky, M. A. (2013). Standard methods for European foulbrood research. Journal of Apicultural Research, 52(1), 1-14.

Fratini, F., Cilia, G., Turchi, B. & Felicioli, A. (2016). Beeswax: a minireview of its antimicrobial activity and its application in medicine. Asian Pacific Journal of Tropical Medicine, 9(9), 839-843.

Gilliam, M. & Argauer, R. J. (1981). Oxytetracycline residues in surplus honey, brood nest honey, and larvae after medication of colonies of honey bees, Apis mellifera, with antibiotic extender patties, sugar dusts, and syrup sprays. Environmental Entomology, 10(4), 479-482.

56 Hansen, H. & Brodsgaard, C. J. (1999). American foulbrood: a review of its biology, diagnosis and control. Bee World, 80(1), 5-23.

Harrop, A. J., Hocknull, M. D. & Lilly, M. D. (1989). Biotransformations in organic solvents: a difference between gram-positive and gram-negative bacteria. Biotechnology Letters, 11(11), 807-810.

Haynes, E., Helgason, T., Young, J.P., Thwaites, R., & Budge, G.E. (2013). A typing scheme for the honeybee pathogen Melissococcus plutonius allows detection of disease transmission events and a study of the distribution of variants. Environmental Microbiology Reports, 5(4), 525-529.

Hornitzky, M. A. & Smith, L. (1998). Procedures for the culture of Melissococcus pluton from diseased brood and bulk honey samples. Journal of Apicultural Research, 37(4), 293-294.

Huang, S., Zhang, C. P., Wang, K., Li, G. & Hu, F. L. (2014). Recent advances in the chemical composition of propolis. Molecules, 19, 610–632.

Jaycox, E. R. (1979). Comb foundation: Are we using enough? American Bee Journal, 119(7), 515–516.

Johnson, K. S., Eischen, F. A. & Giannasi, D. E. (1994). Chemical composition of North American bee propolis and biological activity towards larvae of greater wax moth (Lepidoptera: Pyralidae). Journal of Chemical Ecology, 20(7), 1783-1791.

Jorgensen, J. H. & Ferraro, M. J. (2009). Antimicrobial susceptibility testing: a review of general principles and contemporary practices. Medical Microbiology, 49, 1749- 1755.

Kacániová, M., Vukovic, N., Cheblo, R., Hascik, P., Rovna, K., et al. (2012). The antimicrobial activity of honey, bee pollen loads and beeswax from Slovakia. Archives of Biological Science, Belgrade, 64(3), 927-934.

Kowalski, R. P., Yates, K. A., Romanowski, E. G., Karenchak, L. M., Mah, F.S. et al. (2005). An ophthalmologist’s guide to understanding antibiotic susceptibility and minimum inhibitory concentration data. Ophthalmology, 112(11), 1987.e1-e6.

Luzzatto, L., Airprion, D. & Schlessinger, D. (1968). Mechanism of action of streptomycin in E. coli: Interruption of the ribosome cycle at the initiation of protein synthesis. PNAS, 60(3), 873–880.

Marcucci, M. C. (1995). Propolis: Chemical composition, biological properties and therapeutic activity. Apidologie, 26(1), 83–99.

Matsuka, M. & Nakamura, J. (1990). Oxytetracycline residues in honey and royal jelly. Journal of Apicultural Research, 29(2), 112-117. 57 Mirzoeva, O. K., Grishanin, R. N. & Calder, P. C. (1997). Antimicrobial action of propolis and some of its components: the effects on growth, membrane potential and motility of bacteria. Microbiology Research, 152, 239-246.

Nakai, Y. Ishihara, M, Arai, R. & Takamatsu, D. (2018). A scientific note on improved isolation methods for Melissococcus plutonius from diseased Apis mellifera larvae. Apidologie, 49(3), 1-3.

Nicodemo, D., Malheiros, E., de Jong, D. & Couto, R. (2014). Increased brood viability and longer lifespan of honeybees selected for propolis production. Apidologie, 45(2), 269-275.

Percival, S. L. & Williams, D. W. ‘Escherichia coli’ in Microbiology of Waterborne Diseases. Elsevier: Amsterdam, Netherlands. Pp. 89-117.

Rahman, M. M., Richardson, A. & Sofian-Azirun, M. (2010). Antibacterial activity of propolis and honey against Staphylococcus aureus and Escherichia coli. African Journal of Microbiology Research, 4(16), 1872-1878.

Reller, B. L., Weinstein, M., Jorgensen, J. H. & Ferraro, M. J. (2009). Antimicrobial susceptibility testing: a review of general principles and contemporary practices. Clinical Infectious Diseases, 49(11), 1749–1755

Ristivojević, P., Dimkić, I., Trifković, J., Berić, T., Vovkc, I., et al. (2016). Antimicrobial activity of Serbian propolis evaluated by means of MIC, HPTLC, bioautography and chemometrics. PLoS One, 11(6), e0157097.

Ristivojevića, P., Andrićb, F., Trifkovićb, J. D., Vovkc, I., Stanisavljeviće, L. Z., et al. (2013). Pattern recognition methods and multivariate image analysis in HPTLC fingerprinting of propolis extracts. Journal of Chemometrics, 28, 301-310.

Sforcin, J. M., Fernandes, A., Lopes, C. A. M., Bankova, V. & Funari, S. R. C. (2000). Seasonal effect on Brazilian propolis antibacterial activity. Journal of Ethnopharmacology, 73(1-2), 243-249.

Simone, M., Evans, J. D. & Spivak, M. (2009). Resin collection and social immunity in honey bees. Evolution, 63(11), 3016-3022.

Sporns, P., Kwan, S. & Roth, L. A. (1986). HPLC analysis of oxytetracycline residues in honey. Journal of Food Protection, 49(5), 383-388.

Strehle, M. A., Jenke, F., Fröhlich, B., Tautz, J., Riederer, M., et al. (2003). Raman spectroscopic study of spatial distribution of propolis in comb of Apis mellifera carnica (Pollm.). Biopolymers - Biospectroscopy Section, 72(4), 217–224.

58 Tosi, B., Donini, A., Romagnoli, C. & Bruni, A. (1996). Antimicrobial activity of some commercial extracts of propolis prepared with different solvents. Phytotherapy Research, 10, 335-336.

Vassalo, J., Besinis, A., Boden, R. & Handy, R. D. (2018). The minimum inhibitory concentration (MIC) assay with Escherichia coli: an early tier in the environmental hazard assessment of nanomaterials? Ecotoxicology and Environmental Safety, 162, 633-646.

Wehrli, W. (1983). Rifamipin: Mechanisms of action and resistance. Reviews of Infectious Diseases, 5(3), S407-S411.

Wilson, W. T. (1974). Residues of oxytetracycline honey stores by Apis mellifera. Environmental Entomology, 3(4), 674-676.

59 Conclusion

Overall, there is a benefit of darkened wax combs previously used for brood rearing, but the same benefit was exhibited by wax combs used for food storage as well.

These combs exhibit antibacterial activity against Melissococcus plutonius, the bacterium responsible for European Foulbrood disease. This activity was not observed against human pathogens, Escherichia coli or Staphylococcus saprophyticus. Future research should (1) document antibacterial effects of comb extracts on more honey bee pathogens, and (2) elucidate the active components of comb extracts responsible for antimicrobial activity. This research will provide a clearer understanding of what is and is not beneficial about used wax combs, allowing beekeepers to make more informed decisions on hive equipment management.

Additionally, there was no difference in larval honey bee survival on comb treatments that had or had not been used for brood rearing previously. Previous research has demonstrated ambiguous effects of old brood bomb on larval honey bee survival.

Berry and Delaplane (2001) demonstrate increased survival on old brood comb, likely due to the presence of brood pheromone stimulating larval care from adult worker bees.

On the other hand, Wu et al (2011) demonstrated delayed development of larvae reared in old comb that was contaminated with beekeeper-applied acaricides. Considering these results and the results from our study, beekeepers should continue to replace brood combs that have been exposed to agricultural pesticides or synthetic acaricides, but may not 60 benefit economically from the replacement of brood combs that are not heavily contaminated. More research should be done on the sublethal effects on old brood comb that was not exposed to synthetic acaricides.

Finally, colonies demonstrate no preference for comb treatments that has or has not been used for brood rearing when given the choice inside the hive. This is in contrast to beekeeping anecdotes and research by Visscher et al (1985) showing colony preference for previously used brood comb. However, colony preference for old brood comb has only been demonstrated on a landscape-level, in which swarms are searching for a new nest place, whereas this thesis tested for preference at the colony-level, in which small colonies were placed in a new nest space containing used and unused comb treatment on either side. Perhaps at this level, bees are no longer using conspicuous olfactory cues from old comb or searching for constructed comb that will save worker energy. The mechanisms behind swarm preference for nest spaces holding used brood comb should be studied further.

In conclusion, the results of this thesis suggest that constituents of comb used for brood rearing have no immediate effect on larval honey bee survival, but they do have antibacterial effects on M. plutonius. Additionally, constituents of comb used for food storage also exhibit antibacterial effects. To better inform beekeepers of the economic and biological costs and benefits of brood comb replacement, more research should be done on the sublethal effects of uncontaminated, old brood comb, as well as identifying the active compounds present in brood comb and honey comb. Propolis was highly active against M. plutonius, suggesting that compounds present in propolis may also be present in wax combs. Regardless of whether or not propolis compounds are responsible for 61 activity of wax combs, propolis should be further researched as a bee-safe and human- safe treatment for hives infected with European Foulbrood.

62 Bibliography

Ahmed, F. E. (1995). Toxicological effects of ethanol on human health. Critical Reviews in Toxicology, 25(4), 347-367.

Andelković, B., Vujisić, L., Vučković, I., Tešević, V. & Vajs, V. (2017). Metabolomics study of Populus type propolis. Journal of Pharmaceutical and Biomedical Analysis, 135, 217-226.

Aurongzeb, M. & Azim, M. K. (2011). Antimicrobial properties of natural honey: a review of literature. Pak. Journal of Biochem. Mol. Biol. 2011, 44(3), 118-124.

Bailey, L. & Ball, B. V. (1991). ‘Fungi’ in Honey Bee Pathology. Academic Press: London, UK. Pp. 53-63.

Bailey, L. & Collins, M. D. (1982). Reclassification of ‘Streptococcus pluton’ (White) in a new genus Melissococcus, as Melissococcus pluton nom, rev; comb. nov. Journal of Applied Bacteriology, 53, 215-217.

Bailey, L. (1957). The cause of European foul brood. Bee World, 38, 95-89.

Bailey, L. (1983). Melissococcus pluton, the cause of European foulbrood of honey bees (Apis spp.). Journal of Applied Bacteriology, 55, 65-69.

Bailey, L. (1983). Melissococcus pluton, the cause of European foulbrood of honey bees (Apis spp.). Journal of Applied Bacteriology, 55, 65-69.

Bee Informed Partnership. (2012a). The Bee Informed Partnership Management Survey Results (2011) Brood Comb Management. Retrieved from https://beeinformed.org/wp-content/uploads/2012/04/Brood-Comb- managment.pdf

Bee Informed Partnership. (2012b). The Bee Informed Partnership Management Survey Results (2012-2013) Brood Comb Management. Retrieved from https://beeinformed.org/wp-content/uploads/2014/11/Comb-Management- Summary.pdf

Bee Informed Partnership. (2013). The Bee Informed Partnership Management Survey Results (2012) Brood Comb Management. Retrieved from

63 https://beeinformed.org/wp-content/uploads/2013/08/title-Brood-Comb- managment.pdf

Bee Informed Partnership. (2014). The Bee Informed Partnership Management Survey Results (2013-2014) Brood Comb Management. Retrieved from https://beeinformed.org/wp-content/uploads/2014/09/Comb-Management- Summary.pdf

Bee Informed Partnership. (2019). Average Winter Loss per Beekeeper. Retrieved from https://bip2.beeinformed.org/survey/

Berry, J. A., & Delaplane, K. S. (2001). Effects of comb age on honey bee colony growth and brood survivorship. Journal of Apicultural Research, 40(1), 3–8.

Bogdanov, S. (2004). Beeswax: Quality issues today. Bee World, 85(3), 46-50.

Booth, D. T., Cox, S. E., Meikle, T. W. & Fitzgeral, C. (2006). The accuracy of ground- cover measurements. Rangeland Ecological Management, 59, 179-188.

Borba, R. S., Klyczek, K. K., Mogen, K. L. & Spivak, M. (2015). Seasonal benefits of a natural propolis envelope to honey bee immunity and colony health. Journal of Experimental Biology, 218, 3689-3699.

Bruschi, M. L., Franco, S. L. & M. Gremião, P. D. (2003). Application of an HPLC method for analysis of propolis extract. Journal of Liquid Chromatography & Related Technologies, 26(14), 2399–240.

Chopra, I. & Roberts, M. (2001). Tetracycline antibiotics: Mode of action, applications, molecular biology, and epidemiology of bacterial resistance. Microbiology and Molecular Biology Reviews, 65(2), 232-260.

Cremer, S., Armitage, S. O. A. & Schmid-Hempel, P. (2007). Social immunity. Current Biology, 17, R693-R702.

Cutler, G. C. & Scott-Dupree, C. D. (2007). Exposure to clothianidin seed-treated canola has no long-term impact on honey bees. Ecotoxicology, 100(3), 765-772.

Delaplane, K. S. (2015). ‘Management for honey production’ in Graham, J.M. The Hive and the Honey Bee. Dadant & Sons: Hamilton, IL, USA. Pp. 505-506.

Delaplane, K. S., Van Der Steen, J. & Guzman-Novoa, E. (2013). Standard methods for Apis mellifera research. Journal of Apicultural Research, 52(1), 1-12.

Di Pinto, A., Novello, L., Terio, V., & Tantillo, G. (2011). ERIC-PCR genotyping of Paenibacillus larvae in Southern Italian honey and brood combs. Current Microbiology, 63(5), 416–419.

64 Dicks, L. M. T., Endo, A. & Van Reenan, C. A. (2014). ‘Minor Genera of the Enterococcaceae (Catellicoccus, Melissococcus, and Pilibacter)’ in Lactic Acid Bacteria: Biodiversity & Taxonomy. John Wiley & Sons: Hoboken, NJ, USA. Pp 239-243.

Drescher, N., Klein, A., Neumann, P., Yañez, O. & Leonhardt, S. D. (2017). Inside honeybee hives: Impact of natural propolis on the ectoparasitic mite Varroa destructor and viruses. Insects, 8(15).

Forsgren, E. (2010). European foulbrood in honey bees. Journal of Invertebrate Pathology, 103, S5-S9.

Forsgren, E., Budge, G.E., Charriere, J., & Hornitzky, M.A. (2013). Standard methods for European foulbrood research. Journal of Apicultural Research, 52(1), 1-14.

Fratini, F., Cilia, G., Turchi, B. & Felicioli, A. (2016). Beeswax: a minireview of its antimicrobial activity and its application in medicine. Asian Pacific Journal of Tropical Medicine, 9(9), 839-843.

Free, J.B. & Winder, M.E. (1983). Brood recognition by honeybee (Apis mellifera) workers. Animal Behavior, 31(2) 539-545.

Gilliam, M. & Argauer, R. J. (1981). Oxytetracycline residues in surplus honey, brood nest honey, and larvae after medication of colonies of honey bees, Apis mellifera, with antibiotic extender patties, sugar dusts, and syrup sprays. Environmental Entomology, 10(4), 479-482.

Greenleaf, S. S., Williams, N. M., Winfree, R., & Kremen, C. (2007). Bee foraging ranges and their relationship to body size. Oecologia, 153(3), 589–596.

Hansen, H. & Brodsgaard, C. J. (1999). American foulbrood: a review of its biology, diagnosis and control. Bee World, 80(1), 5-23.

Harrop, A. J., Hocknull, M. D. & Lilly, M. D. (1989). Biotransformations in organic solvents: a difference between gram-positive and gram-negative bacteria. Biotechnology Letters, 11(11), 807-810.

Haynes, E., Helgason, T., Young, J.P., Thwaites, R., & Budge, G.E. (2013). A typing scheme for the honeybee pathogen Melissococcus plutonius allows detection of disease transmission events and a study of the distribution of variants. Environmental Microbiology Reports, 5(4), 525-529.

Hepburn, H. R., & Kurstjens, S. P. (1988). The combs of honeybees as composite materials. Apidologie, 19(1), 25–36.

Hoopingarner, R. & Sanford, M.T. (1990). The costs of beekeeping—I. Survey of commercial beekeepers. American Bee Journal, 130(6), 405-407. 65 Hornitzky, M. A. & Smith, L. (1998). Procedures for the culture of Melissococcus pluton from diseased brood and bulk honey samples. Journal of Apicultural Research, 37(4), 293-294.

Huang, S., Zhang, C. P., Wang, K., Li, G. & Hu, F. L. (2014). Recent advances in the chemical composition of propolis. Molecules, 19, 610–632.

Jay, S. C. (1963). The development of honeybees in their cells. Journal of Apicultural Research, 2(2), 117–134.

Jay, S. C. (1964). The cocoon of the honey bee, Apis mellifera L. The Canadian Entomologist, 96(5), 784–792.

Jaycox, E. R. (1979). Comb foundation: Are we using enough? American Bee Journal, 119(7), 515–516.

Johnson, K. S., Eischen, F. A. & Giannasi, D. E. (1994). Chemical composition of North American bee propolis and biological activity towards larvae of greater wax moth (Lepidoptera: Pyralidae). Journal of Chemical Ecology, 20(7), 1783-1791.

Johnson, R. M., Ellis, M. D., Mullin, C. A., & Frazier, M. (2010). Pesticides and honey bee toxicity – USA. Apidologie, 41(3), 312–331.

Jorgensen, J. H. & Ferraro, M. J. (2009). Antimicrobial susceptibility testing: a review of general principles and contemporary practices. Medical Microbiology, 49, 1749- 1755.

Kacániová, M., Vukovic, N., Cheblo, R., Hascik, P., Rovna, K., et al. (2012). The antimicrobial activity of honey, bee pollen loads and beeswax from Slovakia. Archives of Biological Science, Belgrade, 64(3), 927-934.

Keoniger, N. (1970). Factors determining the laying of drome and worker eggs by the queen honeybee. Bee World, 51, 166-169.

Killion, E. (2015). ‘The production of comb and bulk comb honey’ in Graham, J.M. The Hive and the Honey Bee. Dadant & Sons: Hamilton, IL, USA. Pp. 562-563.

Koenig, J. P., Boush, G. M., & Erickson, E. H. (1986). Effect of type of brood comb on chalk brood disease in honeybee colonies. Journal of Apicultural Research, 25(1), 58–62.

Kowalski, R. P., Yates, K. A., Romanowski, E. G., Karenchak, L. M., Mah, F.S. et al. (2005). An ophthalmologist’s guide to understanding antibiotic susceptibility and minimum inhibitory concentration data. Ophthalmology, 112(11), 1987.e1-e6.

66 Le Conte, Y., Sreng, L. & Poitout, S. H. (1995). Brood pheromone can modulate the feeding behavior of Apis mellifera workers (Hymenoptera: Apidae). Apiculture and Social Insects, 88(4), 798-804.

Luzzatto, L., Airprion, D. & Schlessinger, D. (1968). Mechanism of action of streptomycin in E. coli: Interruption of the ribosome cycle at the initiation of protein synthesis. PNAS, 60(3), 873–880.

Magnum, W. (2015). ‘A brief history of beekeeping in the United States’ in Graham, J.M. The Hive and the Honey Bee. Dadant & Sons: Hamilton, IL, USA. Pp. 25-50.

Marcucci, M. C. (1995). Propolis: chemical composition, biological properties and therapeutic activity. Apidologie, 26(1), 83–99.

Matsuka, M. & Nakamura, J. (1990). Oxytetracycline residues in honey and royal jelly. Journal of Apicultural Research, 29(2), 112-117.

McMullan, J. B., & Brown, M. J. F. (2006). The influence of small-cell brood combs on the morphometry of honeybees (Apis mellifera). Apidologie, 37(6), 665–672.

Mirzoeva, O. K., Grishanin, R. N. & Calder, P. C. (1997). Antimicrobial action of propolis and some of its components: the effects on growth, membrane potential and motility of bacteria. Microbiology Research, 152, 239-246.

Mullin C. A., Frazier M., Frazier J. L., Ashcraft S., Simonds R., et al. (2010). High levels of miticides and agrochemicals in North American apiaries: Implications for honey bee health. PLoS ONE 5(3): e9754

Nakai, Y. Ishihara, M, Arai, R. & Takamatsu, D. (2018). A scientific note on improved isolation methods for Melissococcus plutonius from diseased Apis mellifera larvae. Apidologie, 49(3), 1-3.

Nicodemo, D., Malheiros, E., de Jong, D. & Couto, R. (2014). Increased brood viability and longer lifespan of honeybees selected for propolis production. Apidologie, 45(2), 269-275.

Pankiw, P., Bailey, L., Gochnauer, T. A., & Hamilton, H. A. (1970). Disinfection of honeybee combs by gamma irradiation. II. European foul brood disease. Journal of Apicultural Research, 9(3), 165–168.

Percival, S. L. & Williams, D. W. ‘Escherichia coli’ in Microbiology of Waterborne Diseases. Elsevier: Amsterdam, Netherlands. Pp. 89-117.

Rahman, M. M., Richardson, A. & Sofian-Azirun, M. (2010). Antibacterial activity of propolis and honey against Staphylococcus aureus and Escherichia coli. African Journal of Microbiology Research, 4(16), 1872-1878.

67 Ramsey, S. D., Ochoa, R., Bauchan, G., Gulbronson, C., Mowery, J. D., et al. (2019). Varroa destructor feeds primarily on honey bee fat body tissue and not hemolymph. PNAS, 116(5), 1792-1801.

Reller, B. L., Weinstein, M., Jorgensen, J. H. & Ferraro, M. J. (2009). Antimicrobial susceptibility testing: a review of general principles and contemporary practices. Clinical Infectious Diseases, 49(11), 1749–1755

Ristivojević, P., Dimkić, I., Trifković, J., Berić, T., Vovkc, I., et al. (2016). Antimicrobial activity of Serbian propolis evaluated by means of MIC, HPTLC, bioautography and chemometrics. PLoS One, 11(6), e0157097.

Ristivojevića, P., Andrićb, F., Trifkovićb, J. D., Vovkc, I., Stanisavljeviće, L. Z., et al. (2013). Pattern recognition methods and multivariate image analysis in HPTLC fingerprinting of propolis extracts. Journal of Chemometrics, 28, 301-310.

Rosenkranz, P., Aumeier, P. & Ziegelmann, B. (2010). Biology and control of Varroa destructor. Journal of Invertebrate Pathology, 103, S96-S119.

Schneider, S. S. (2015). ‘The honey bee colony: Life history’ in Graham, J.M. The Hive and the Honey Bee. Dadant & Sons: Hamilton, IL, USA. Pp. 74-78.

Seeley, T. D. & Morse, R.A. (1976). The nest of the honey bee (Apis mellifera L.). Insects Sociaux, 23, 495-512.

Seeley, T. D. (1985). Honey bee ecology. Princeton University Press: Princeton, NJ, USA. Pp 201.

Sforcin, J. M., Fernandes, A., Lopes, C. A. M., Bankova, V. & Funari, S. R. C. (2000). Seasonal effect on Brazilian propolis antibacterial activity. Journal of Ethnopharmacology, 73(1-2), 243-249.

Simone, M., Evans, J. D. & Spivak, M. (2009). Resin collection and social immunity in honey bees. Evolution, 63(11), 3016-3022.

Simone-Finstrom, M. D. & Spivak, M. (2012). Increased resin collection after parasite challenge: a case of self-medication in honey bees? PLoS One, 7(3), e34601.

Smith, R. & Wilcox, M. M. (1990). Chemical residues in bees, honey, and beeswax. American Bee Journal, 130(3) 188-192.

Snodgrass, R. E., Erickson, E. H. & Fahrbach, S. E. (2015). ‘The anatomy of the honey bee’ in Graham, J.M. The Hive and the Honey Bee. Dadant & Sons: Hamilton, IL, USA. Pp. 141.

Sporns, P., Kwan, S. & Roth, L. A. (1986). HPLC analysis of oxytetracycline residues in honey. Journal of Food Protection, 49(5), 383-388. 68 Strehle, M. A., Jenke, F., Fröhlich, B., Tautz, J., Riederer, M., et al. (2003). Raman spectroscopic study of spatial distribution of propolis in comb of Apis mellifera carnica (Pollm.). Biopolymers - Biospectroscopy Section, 72(4), 217–224.

Sumpter, J. T. & Martin, S. J. (2004). The dynamic of virus epidemics in Varroa-infested honey bee colonies. Journal of Animal Ecology, 73(1), 51-63.

Tosi, B., Donini, A., Romagnoli, C. & Bruni, A. (1996). Antimicrobial activity of some commercial extracts of propolis prepared with different solvents. Phytotherapy Research, 10, 335-336.

Tulloch, A. P. (1979). Beeswax – Composition and analysis. Bee World, 60(2), 47-62.

Vanengelsdorp D., Evans J. D., Saegerman C., Mullin C., Haubruge E., et al. (2009). Colony collapse disorder: a descriptive study. PLoS One, 4(8), e6481.

Vassalo, J., Besinis, A., Boden, R. & Handy, R. D. (2018). The minimum inhibitory concentration (MIC) assay with Escherichia coli: an early tier in the environmental hazard assessment of nanomaterials? Ecotoxicology and Environmental Safety, 162, 633-646.

Visscher, P. K., Morse, R. A., & Seeley, T. D. (1985). Honey bees choosing a home prefer previously occupied cavities. Insectes Sociaux, 32(2), 217–220.

Waddington, K. D., & Herbst, L. H. (1987). Body size and the functional length of the proboscis of honey bees. The Florida Entomologist, 70(1), 124–128.

Wehrli, W. (1983). Rifamipin: Mechanisms of action and resistance. Reviews of Infectious Diseases, 5(3), S407-S411.

Wilson, M. B, Pawlus, A. D, Brinkman, D., Gardner, G., Hegeman, A.D., et al. (2017). 3- Acyl dihydroflavonols from poplar resins collected by honey bees are active against the bee pathogen Paenibacillus larvae and Ascophaera apis. Phytochemistry, 138, 83-92.

Wilson, W. T. (1974). Residues of oxytetracycline honey stores by Apis mellifera. Environmental Entomology, 3(4), 674-676.

Wu, J. Y., Anelli, C. M., & Sheppard, W. S. (2011). Sub-lethal effects of pesticide residues in brood comb on worker honey bee (Apis mellifera) development and longevity. PLoS ONE, 6(2), e14720.

Wu, J. Y., Smart, M. D., Anelli, C. M., & Sheppard, W. S. (2012). Honey bees (Apis mellifera) reared in brood combs containing high levels of pesticide residues exhibit increased susceptibility to Nosema (Microsporidia) infection. Journal of Invertebrate Pathology, 109(3), 326-329.

69