Ecological and Evolutionary Relationships between Bees and their Bacterial Gut Microbiota

Item Type text; Electronic Dissertation

Authors Martinson, Vincent G.

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 23/09/2021 12:57:33

Link to Item http://hdl.handle.net/10150/223362 ECOLOGICAL AND EVOLUTIONARY RELATIONSHIPS BETWEEN BEES AND THEIR BACTERIAL GUT MICROBIOTA

by

Vincent G. Martinson

______

A Dissertation Submitted to the Faculty of the

GRADUATE INTERDISCIPLINARY PROGRAM IN INSECT SCIENCE

In Partial Fulfillment of the Requirements For the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA

2012

2

THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE

As members of the Dissertation Committee, we certify that we have read the dissertation prepared by Vincent G. Martinson entitled ECOLOGICAL AND EVOLUTIONARY RELATIONSHIPS BETWEEN BEES AND THEIR BACTERIAL GUT MICROBIOTA and recommend that it be accepted as fulfilling the dissertation requirement for the

Degree of Doctor of Philosophy

______Date: 03/08/12 Nancy A. Moran

______Date: 03/08/12 Howard Ochman

______Date: 03/08/12 Michael J. Sanderson

______Date: 03/08/12 Diana E. Wheeler

______Date: 03/08/12 Noah K. Whiteman

Final approval and acceptance of this dissertation is contingent upon the candidate’s submission of the final copies of the dissertation to the Graduate College.

I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement.

______Date: 03/08/12 Dissertation Director: Nancy A. Moran

3

STATEMENT BY AUTHOR

This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

SIGNED: ______Vincent G. Martinson

4

ACKNOWLEDGEMENTS

Nancy Moran is an excellent scientist and advisor that has an unparalleled enthusiasm for science. I thank Nancy for giving me the opportunity to help develop a new research program within her lab. Not only did she provide her expert advise on the direction of my research and presenting my work, but generously provided resources that allowed this work to happen. Thank you to Mike Sanderson, Diana Wheeler, and Noah Whiteman for contributing their time to be on my dissertation committee, and providing helpful feedback. I have learned much from Howard Ochman through classes, lab meetings, and as a committee member, and thank him for his honest discussions. I thank Molly Hunter for filling in at my oral examination and for her interest in my research. The Moran and Ochman lab members (Nicole Gerardo, Mark van Passel, John Stavrinides, Pradeep Reddi Marri, Hema Narra, Renyi Liu, Chi-Horng Kuo, Baoyu Tian, Kerry Oliver, Eva Nováková, John McCutcheon, Allison Hansen, Rahul Raghavan, Zakee Sabree, Philipp Engel, Walden Kwong, Patrick Degnan, Kevin Vogel, and Gaelen Burke) have been greatly influential in my scientific development. Thank you to the many lab technicians that have given me support over the years (Helen Dunbar, Heather McLaughlin, Tyler Jarvik, Eli Powell, and Kim Hammond). Especially Kim and Eli for wrangling irritable bee colonies and keeping the rest of the colonies happy. Thank you to the members of the Carl Hayden Bee Research Center (Gloria DeGrandi-Hoffman, Thomas Deeby, Mark Carroll, Bruce Eckholm) for access to hives and training me how to tend bees, without which I could not have performed my work. James Nieh and Jay Evans contributed critical specimens for my research. My co-authors Bryan Danforth, Robert Minckley, Olav Rueppell, Salim Tingek, Tanja Magoc, Steven Salzberg gave valuable feedback on manuscripts that became part of this dissertation. Undergraduate Jamie Moy helped to collect data. Thank you to Becky Nankivell, Sharon Richards, Teresa Kudrna, Ryan Nystrom, Ming Beckwith, Pennie Liebig, and the EEB office staff for their administrative support (and encouragement). Thank you to Andrea Grantham, Tony Day, Carl Boswell for teaching me microscopy techniques that aided in my research. Thank you to the NSF IGERT Program in Comparative Genomics and the director Michael Nachman, which not only funded much of my graduate research, but also taught me invaluable techniques, critical thinking, and about surviving in academia. Thank you to the Center for Insect Science and the NSF (through funding awards to Nancy Moran) for funding my research and travel to scientific conferences. Thank you to the Department of Ecology and Evolutionary Biology for the platform to share my data, feedback from faculty, and free food. Thank you to all my family and friends (you know who you are) for being there through thick and thin.

5

DEDICATION

To Ellen for your love, and to my family, Mom (Joanne) and Dad (Wynn) and my brothers Philip and Jonathan for all their support, encouragement, and positive thinking.

6

TABLE OF CONTENTS

ABSTRACT...... 7

INTRODUCTION...... 8

PRESENT STUDY...... 19

REFERENCES...... 23

APPENDIX A: A SIMPLE AND DISTINCTIVE MICROBIOTA ASSOCIATED WITH HONEY BEES AND BUMBLE BEES...... 29

APPENDIX B: ESTABLISHMENT OF CHARACTERISTIC GUT BACTERIA DURING DEVELOPMENT OF THE HONEY BEE WORKER...... 73

APPENDIX C: HAPPENSTANCE SEQUENCING OF THE NEAR-COMPLETE GENOME OF A BUMBLE BEE GUT SYMBIONT...... 123

7

ABSTRACT

Gut microbial communities exist in the vast majority of animals, and often form complex symbioses with their hosts that affect their host’s biology in numerous ways. To date, the majority of studies of these complex interactions have focused on the nutritional benefits provided by the microbiota; however, the natural microbiota can also influence development, immunity, and the metabolism of its host. Apis mellifera, the honey bee, harbors a distinctive bacterial community that is present in individuals from distant locations around the world; however, the basis of the bee-microbiota association is unknown. This dissertation explores properties of the bacterial microbiota within bees, including its persistence of this association, mechanisms of transmission, localization through host ontogeny, and basic metabolic capabilities that define and maintain the symbiotic relationship. Apis and Bombus species (honey and bumble bees) share a distinct bacterial microbiota that is not present in other bees and wasps. Close analysis of the A. mellifera microbiota revealed consistent communities in adult worker gut organs and a general lack of bacteria in larvae. Contact between workers and with hive materials were identified as major routes of transmission for bacterial communities, showing the importance of social behavior in this association. Genomic analysis of a gut bacterium co-sequenced with the Bombus impatiens genome revealed it as a divergent lineage of

Gammaproteobacteria, and deletions of conserved metabolic pathways, reduction in genome size, and its low GC content all suggest that the bacterial species has had a long association with its host. 8

INTRODUCTION

Explanation of the problem and its context

The evolution and physiology of animals are greatly influenced by the microbial communities residing within them (Lee & Mazmanian 2010; Ley et al. 2008; Ley et al.

2006; Wostmann 1996). Many host-microbe interactions occur within the gut, which is not only the chief area of food digestion and nutrient absorption, but also a main location for immune system activity (Chapman 1998; Hooper et al. 2002; Lee & Mazmanian

2010; Tellam 1996; Vallet-Gely et al. 2008). Gut microbial communities exist in the vast majority of animals and have formed complex symbioses with their hosts, affecting their host’s biology in abundant ways (Dehority 1997; Dillon & Dillon 2004; Mackie 1997).

Advances in molecular techniques have enabled the characterization of resident microbial communities within animal guts; however, relationships that may be critical for animal health are often overlooked.

The biology of the honey bee, Apis mellifera, has been thoroughly studied because of its importance as a pollinator in agricultural and natural ecosystems and its social behavior (Aizen et al. 2008; Klein et al. 2007; Robinson et al. 2005). Early culture-based studies of A. mellifera gut bacteria identified thousands of strains without specific colonization patterns (Gilliam 1997). Contradictory to these findings, culture- independent surveys of A. mellifera bacterial associates recovered very few species (<10)

(Jeyaprakash et al. 2003). The relatively simple composition of the microbiota is consistent among different A. mellifera subspecies and populations from distant locations 9 around the world, suggesting that a mutualistic relationship may tie A. mellifera health to its gut bacterial community (Babendreier et al. 2007; Cox-Foster et al. 2007; Mohr &

Tebbe 2006, 2007; Olofsson & Vasquez 2008).

In 2006, US beekeepers suffered colony losses of 50 to 90% from an enigmatic disease dubbed Colony Collapse Disorder (CCD) (vanEngelsdorp et al. 2007). Since then, colony losses have persisted and researchers have uncovered several potential causative agents, but unfortunately no causative agent for this disease has been identified

(Bromenshenk et al. 2010; Cox-Foster et al. 2007; Higes et al. 2009; vanEngelsdorp et al. 2009). Further, numerous species of native bumble bees have drastically decreased in geographic range in North America and Europe (Biesmeijer et al. 2006; Cameron et al.

2011). The decline of bee populations makes it necessary to understand aspects of bee biology that affect its health and defend against pathogens. A well-characterized microbiota of A. mellifera will provide insights for the future management and conservation of honey bee and bumble bee colonies.

Many insects that have difficult-to-digest diets or diets that lack critical nutrients

(e. g., wood, phloem sap) have evolved microbial gut communities to aid in nutrient acquisition (Dillon & Dillon 2004). Insect lineages often harbor vertically transmitted microbes that have been associated with a transition to a novel diet (Hongoh et al. 2005;

Hosokawa et al. 2006; Pinto-Tomas et al. 2009). Bees constitute a monophyletic clade of herbivores that feed primarily on pollen and honey and evolved from carnivorous wasps

(Danforth et al. 2006). Pollen cytoplasm is the only source of protein and fat for bees, yet its outer coating is difficult to digest (Roulston & Cane 2000). Gut bacteria may help to 10 break down the pollen coat to release nutrients for the bee. The distribution of the characteristic microbiota of A. mellifera throughout the bee and wasp phylogeny could reveal whether this set of bacteria is associated with the ancient transition from carnivorous wasp to herbivorous bee.

Phylotypes within the A. mellifera gut have only been recovered in association with this host and close relatives, despite the abundance of barcode-sequencing projects cataloging environmental bacteria. Consistent and specific association with A. mellifera suggests that the microbiota plays a role in host biology, yet most studies have focused on adult bees (Babendreier et al. 2007; Jeyaprakash et al. 2003). The community dynamics of the microbiota through host development and among the gut organs have been overlooked (Brune & Friedrich 2000; Vasanthakumar et al. 2008). Adult workers perform a succession of tasks as they age, which potentially exposes them to different microorganisms: young adults nurse larvae within the hive, whereas older adults forage pollen and nectar from flowers outside the hive (Gilliam 1997; Haydak 1970; Seeley

1985). Physiological changes during development and the worker bee’s progression through roles within the hive could provide the microbiota species separate spacio- temporal niches (Chapman 1998; Haydak 1957). Additionally, social behavior could facilitate transmission of the microbiota between generations (Hughes et al. 2008;

Lombardo 2008).

The bumble bee (Bombus sp.) microbiota contains bacteria closely related to several of the characteristic A. mellifera phylotypes (Koch & Schmid-Hempel 2011a;

Martinson et al. 2011; Olofsson & Vasquez 2009). The natural microbiota of B. 11 terrestris, which is transferred vertically within the colony, has been shown to defend against infection by a trypanosomatid parasite (Koch & Schmid-Hempel 2011b). The social transmission and wide distribution of this microbial consortium among bumble bees suggests a long association and that coevolution has occurred, potentially resulting in specialized symbionts (Koch & Schmid-Hempel 2011a; Martinson et al. 2011).

Sequencing of the B. impatiens genome resulted in the co-sequencing of a

Gammaproteobacterial genome that allowed a first view into the evolution of a member of the bumble bee microbiota and the mechanism by which the microbiota confers protection from trypanosomatid infection.

Review of the literature

Importance of microbiota research

Culture-based and microscopic surveys of animal gut contents have revealed enormous populations of microorganisms, including bacteria, archaea, fungi, and protozoa (Dehority 1997). These observations prompted Louis Pasteur to postulate in

1885 that these microbes influence digestion and nutrient absorption, and are necessary for normal host life (Wostmann 1996). The establishment of germ-free model organisms further highlighted the mutualistic interactions between the host and microbiota by showing major physiological discrepancies relative to normal microbiota animals (Falk et al. 1998). Pasteur’s original hypothesis of nutrition was confirmed; additionally, striking differences in host gut morphology, development, and a reduced ability to avoid infection were observed (Dubos & Schaedler 1960; Gordon & Pesti 1971; Savage 1977). Further 12 research has produced examples of the extent to which the biology of the host and its microbiota are intertwined. For example, normal gut bacteria can diminish Plasmodium infection in Anopheles (Cirimotich et al. 2011), the lack of microbial colonization in

Drosophila leads to severe homeostatic problems caused by mis-regulation of the metabolically central insulin-like signaling pathways (Shin et al. 2011), and the natural microbiota modulates brain development and adult behavior in mice (Heijtza et al. 2011).

Evolution of the microbiota - Certain lineages of animals have evolved specialized gut structures to harbor microbial communities that enable the host to extract carbon, nitrogen, vitamin, and energy resources from low-nutrient or recalcitrant diets (e. g., ruminants, termites) (Brune & Friedrich 2000; Mackie 1997). In turn, lineages of microorganisms have evolved to inhabit the stable and nutrient-rich gut environment, which has lead to both specialist pathogens and mutualists (Ohkuma et al. 2007). The

Gammaproteobacterial orders Enterobacteriales and Pasteurellales consist mainly of host- associated mutualists and pathogens that inhabit the mucosal surfaces of vertebrates

(Bisgaard 1993; Lan & Reeves 2006). Microbial community composition of the gut can be influenced by several factors, including diet, geography, and disease status, yet members of the same species usually have more similar microbiotae than those of sympatric species (Chandler et al. 2011; Ley et al. 2006; Savage 1977). Phylogenetic relationships of the gut microbiotae between host species can recapitulate the host phylogeny, suggesting long-term associations resulting in coevolution between the host and its microbial consortium (Brucker & Bordenstein 2012; Ley et al. 2008; Ochman et al. 2010). 13

Maintenance of microbiota - Transmission of the microbiota between host generations is essential for the maintenance of beneficial bacteria. Coprophagy and other fecal-oral routes are important for host-host transmission of microbiotae in many animals

(e.g., lagomorphs, cockroaches) (Nalepa et al. 2001; Wostmann 1996). Animals with social behavior have an advantage of increased access to beneficial microorganisms by cohabitating with multiple generations (Ley et al. 2008; Lombardo 2008). Social insects also benefit from a shared space where the microbiota can be transmitted in other ways such as contact within the physical living quarters (Hughes et al. 2008).

To persist within the animal gut, bacteria must have mechanisms to evade or cope with host defenses, as well as, a means to compete or cohabit with other microbes

(Hongoh et al. 2008; Walter et al. 2011). Molecular signaling between the host and the natural microbiota is critical for correct colonization and community homeostasis (Ashida et al. 2011; Frese et al. 2011; Hooper 2009). For example, the normally asymptomatic relationship between the mouse and Helicobacter hepaticus can be disrupted by the elimination of the type six secretion system, resulting in over-colonization of H. hepaticus and an inflammatory host immune response (Chow et al. 2010).

Apis mellifera biology and microbiota

In many ecosystems, bees play a crucial role in pollination (Aizen et al. 2008;

Klein et al. 2007). Whereas the majority of bee species are solitary, A. mellifera lives in highly social colonies that consist of several thousand individuals workers (Michener

1974, 2007; Wilson 1971). The queen bee lays one egg in each cell of the comb; where 14 the egg hatches and is attended to by nurse workers (Seeley 1985). Larvae are fed a protein-rich glandular secretion with small amounts of pollen and nectar. During larval development, the foregut is not connected to the hindgut until pupation when it defecates for the first time, whereas the adult gut is differentiated into four organs with specialized digestion functions. After eclosion the newly emerged worker remains in the hive as a nurse bee for approximately the first ten days of its life, consuming large amounts of bee bread (stored pollen) that provides the protein to produce the previously mentioned glandular secretion. The oldest workers carry out foraging flights to collect pollen and nectar to feed the colony. An individual worker lives on average 45 days; however, a colony can potentially survive indefinitely. When a colony grows too large, new queens are reared and the colony splits and a portion of the workers leave with a queen to form a new colony.

Apis mellifera microbial communities - Whereas much focus has been placed on identification and combat of A. mellifera pathogens, commensal microorganisms have received considerably less attention (Schmid-Hempel 1998; Shimanuki & Knox 2000).

Nevertheless, studies of the natural microbial communities of the honey bee go back to the early 20th century (White 1906; White 1921). Numerous culture-based surveys of bacteria within the gut of the honeybee between the 1920s and the late 1990s observed a complex and inconsistent microbiota of over 6,000 bacterial strains (Gilliam 1997).

However, the advent of DNA-based screens of the bacterial 16S rRNA gene exposed an inherent bias in culture-based approaches, revealing that only a small fraction of the microbial community was cultivable using standard techniques (Hugenholtz 2002; Staley 15

& Konopka 1985). Surprisingly, non-culture based surveys subsequently showed that the honey bee microbiota is much simpler than suggested by previous culture-based approaches and that there is a consistent set of only eight bacterial phylotypes in A. mellifera individuals from the United States, Australia, South Africa, Germany, Sweden, and Switzerland (Babendreier et al. 2007; Cox-Foster et al. 2007; Jeyaprakash et al.

2003; Mohr & Tebbe 2006; Olofsson & Vasquez 2008).

The eight dominant phylotypes within A. mellifera’s microbiota stem from five diverse bacterial classes: Alphaproteobacteria (2), Betaproteobacteria (1),

Gammaproteobacteria (2), Firmicutes (2), Actinobacteria (1). Previous cultivation work isolated the Lactobacillus and Bifidobacterium species from honey bee feces that correspond to the phylotypes in the Firmicutes and Actinobacteria (i. e., Firm-4, Firm-5,

Bifido phylotypes) (Gilliam 1997; Scardovi & Troatelli 1969). These genera are often associated with animals as members of the gut microbiota. The Alpha-1 phylotype is related to bacteria in the genus Bartonella, a group of intracellular pathogens (Minnick &

Anderson 2006). The Alpha-2 phylotype is a member of the Acetobacteriaecae. The Beta symbiont forms a distinct clade close to non-pathogenic commensals in the

Simonsiella/Alysiella cluster within Neisseriaceae, which have been isolated from the oral cavities of several mammals (Hedlund & Kuhn 2006; Kuhn et al. 1978). Previous phylogenetic reconstructions have placed the Gamma phylotypes both as members of

Enterobacteriaceae (Jeyaprakash et al. 2003) and as members of Pasteurellaceae

(Olofsson & Vasquez 2008); their conflicting placement has made it difficult to infer closely related genera. 16

Role of bacteria in colony collapse disorder - Major colony losses among beekeeping operations across the United States suggested the emergence of a novel microbial pathogen. In a metagenomic survey to associate microbial species with disease, viral, bacterial, fungal, and trypanosomatid candidates were identified (Cox-Foster et al.

2007). Healthy colonies and colonies that had suffered from colony collapse disorder

(CCD) harbored all eight bacterial phylotypes associated with the A. mellifera microbiota. Although the bacterial microbiota did not have a direct influence on disease status, the community structure did change within CCD colonies, possibly reflecting physiological changes within the gut or a defensive response from the microbiota.

Bombus microbiota

The genus Bombus includes primitively social bee species that live in colonies from tens of workers up to several hundred individuals (Michener 1974). As in A. mellifera, sociality is an important trait for acquiring and maintaining mutualistic relationships with microorganisms (Koch & Schmid-Hempel 2011b). Bombus terrestris retains a simple microbiota with only four main phylotypes, all of which were previously found in A. mellifera (Koch & Schmid-Hempel 2011a; Martinson et al. 2011; Olofsson &

Vasquez 2009). The majority of the microbiota is made up of Gamma- and

Betaproteobacteria closely related to the Gamma-1 and Beta A. mellifera phylotypes.

Less frequently, phylotypes resembling the Firm-5 and Bifido phylotypes are also found.

Work by Koch and Schmid-Hempel (2011b) has demonstrated that the microbiota of

Bombus terresteris is able to defend its host against a trypanosomatid parasite. Given the 17 close relationship of the A. mellifera and Bombus microbiotae, this finding suggests the

A. mellifera microbiota has a protective function.

Explanation of the dissertation format

This dissertation examines several aspects of the relationship between bees and their bacterial gut microbiota, including the evolutionary origins of the microbiota, bacterial community dynamics through host ontogeny and among gut organs, basic mechanisms maintaining the microbiota in the Apis mellifera colony, and the metabolic capabilities of an individual member of the bumble bee microbiota.

Appendix A describes research that determines the evolutionary association between bees and the characteristic microbiota of A. mellifera. In this study, I sequenced the bacterial 16S rRNA gene from the guts of diverse bee and wasp species and found that most had microbial communities dominated by common environmental bacteria and widespread insect parasites. In contrast, species from the closely related genera of Apis and Bombus (honey and bumble bees) harbored distinct bacterial taxa, corresponding to those in A. mellifera. I further confirmed that the distinct microbiota was restricted to

Apis and Bombus using specific primers for the dominant phylotypes to screen multiple individuals from several bee species. Finally, I constructed phylogenies for the phylotypes that verified each as a discrete clade.

The research described in Appendix B focused on the colonization dynamics of the A. mellifera gut by three dominant phylotypes throughout the host lifecycle. I enumerated the gut bacteria with quantitative PCR in different aged workers (i. e., newly 18 emerged, nurses, foragers), and found different gut organs harbored distinct bacterial communities that varied in composition and total number of bacteria. I conducted pyrotag sequencing of gut sections to catalog the entire community within each gut section and found >98% were the characteristic phylotypes. Additionally, I localized the three bacterial phylotypes with fluorescence in situ hybridization microscopy, and found that they formed a stratified community within the ileum. I used experimental rearing to determine the transmission route of the bacteria, which showed that exchange from older workers and exchange from hive surfaces played a role. Finally, I proposed the candidate names of Gilliamella apicola and Snodgrassiella alvi, for the Gamma-1 and Beta phylotypes, respectively, to provide nomenclature continuity in future research on the A. mellifera microbiota.

In Appendix C, I analyzed the genome of a Gammaproteobacterium sequenced in a separate project aimed at recovering the sequence of the Bombus impatiens genome.

Using single-copy orthologous genes I assayed the completeness of the symbiont genome sequencing and found it was nearly complete (~96%) with a relatively small size of

1.9Mb. I conducted a phylogenetic reconstruction with the 16S rRNA gene and found this genome was placed among Ca. Gilliamella apicola sequences. Further, a multiprotein phylogeny I constructed positioned the genome as a singleton clade sister to the

Enterobacteriales. I annotated the genome to ascertain the metabolism of the bacteria and found rearrangements/deletions present in key pathways for oxidative phosphorylation, losses expected to cause this organism to be anaerobic or microaerophilic. Finally, I found it has several gene categories that suggest a defensive function. 19

PRESENT STUDY

The methods, results and conclusions of this study are presented in the papers appended to this dissertation. The following is a summary of the most important findings in this document.

Previous non-culture-based 16S rRNA and metagenomic surveys revealed that the guts of honey bees (Apis mellifera), collected from geographically distant locations, harbor representatives of the same eight bacterial phylotypes (Babendreier et al. 2007;

Cox-Foster et al. 2007; Jeyaprakash et al. 2003; Mohr & Tebbe 2006; Olofsson &

Vasquez 2008). Bees are a monophyletic group of Hymenoptera that transitioned from the carnivorous diet of their wasp ancestors to a completely herbivorous diet (Danforth et al. 2006). Appendix A explores the hypothesis that acquisition of these bacteria was associated with the shift to herbivory in bees. Apis mellifera specimens were consistently colonized by the previously identified bacterial phylotypes. However, other bee species that were surveyed lacked the common A. mellifera phylotypes; instead, solitary bee species’ microbiotae were dominated by a Burkholderia species and by the widespread insect associate, Wolbachia. This result rejected the hypothesis that the evolution of bees’ dietary transition was symbiont-dependent (in at least it was not dependent on these particular symbionts). In contrast to the majority of other bee species, two Asian honey bees (Apis andreniformis and Apis dorsata) and several bumble bee (Bombus) species shared several members of the A. mellifera microbiota. The identification of distinctive bacterial phylotypes among the social lineages of Apis and Bombus suggested that 20 transmission of the microbiota is facilitated by contact between nestmates, and that coevolution may have occurred between these social bee lineages and their gut bacteria.

The microbiota is consistent within A. mellifera adults, but the dynamics of the community were unknown for other stages of this insect’s complex social development.

Appendix B contains the first culture-independent characterization of the microbiota in A. mellifera larvae, nurse worker adults, as well as that from individual gut sections of forager adults. The location and abundance of the three most numerous phylotypes (Beta,

Firm-5, Gamma-1, BFG) were described using fluorescence in situ hybridization microscopy and quantitative PCR. These tests revealed that the different gut organs contain unique community structures. Nurse and forager adults harbored similar microbiotae among their gut organs, with approximately 90% in the rectum, 8% in the ileum, 2% in the midgut, and <1% in the crop. Unbiased pyrotag sequencing supported this trend and reiterated that the three dominant phylotypes made up 73% of the total bacterial microbiota. The ileum had a stratified community with the Beta phylotype abutting the host tissue, the Gamma-1 phylotype distributed in a thick mat adjacent to the

Beta and the ileum wall, and the Firm-5 phylotype present in small patches. Strikingly, larvae seem to have very small amounts of bacteria, and often do not contain the distinctive adult microbiota phylotypes, suggesting that they secrete antimicrobial compounds or that the characteristic bacteria cannot survive in the larval gut environment. Further, newly emerged workers (NEWs) were mostly bacteria-free.

Feeding experiments showed that the common bacterial phylotypes can be transferred to

NEWs from older workers, or through natural emergence and subsequent contact with the 21 comb. Bee bread, the major source of protein and fat in the A. mellifera diet, also did not contain many of the microbiota phylotypes, suggesting the microbiota cannot survive outside the bee gut or the bee bread is defended from colonization. Candidate names

Snodgrassiella alvi (Beta) and Gilliamella apicola (Gamma-1) were proposed for two of the bee microbiota phylotypes.

Within the Bombus impatiens genome project an associated gammaproteobacterial genome was inadvertently sequenced, providing the unique opportunity to reconstruct the metabolism of a member of the bumble bee microbiota.

This microbial community has been shown to be protective against trypanosomatid parasites; however, the mechanism of defense is unknown (Koch & Schmid-Hempel

2011b). In Appendix C, the near-complete genome sequence of Ca. Gilliamella apicola strain B. impatiens.01 is reported. Multiprotein phylogenetic analysis of this genome positioned it as a singleton clade with a sister position to Enterobacteriales, splitting the commonly reconstructed unification of Enterobacteriales and Pasteurellales. The genome is relatively small (~1.9Mb) and GC poor (36.6%) compared to most other

Gammaproteobacteria, but similar to known members of the Pasteurellales, which are associates of mucosal surfaces of vertebrates (Christensen & Bisgaard 2008). The genome has few regions of synteny with other genomes, even in areas conserved across related bacteria, and has lost genes underlying large portions of the citric acid cycle and

NADH I, which are critical for oxygenic phosphorylation. Reduction and loss of pathways to survive in oxygen-rich environments suggests a long-term association with the bee gut and potential specialization to compete in the low-oxygen or anaerobic 22 environments where it is found. The genome contains several attachment genes and secretion systems that are likely used to colonize the cuticle lining of the hindgut or to attack invading microbes.

23

REFERENCES

Aizen MA, Garibaldi LA, Cunningham SA, Klein AM (2008) Long-term global trends in crop yield and production reveal no current pollination shortage but increasing pollinator dependency. Curr. Biol. 18:1572-1575.

Ashida H, Ogawa M, Kim M, Mimuro H, Sasakawa C (2011) Bacteria and host interactions in the gut epithelial barrier. Nat. Chem. Biol. 8:36-45.

Babendreier D, Joller D, Romeis J, Bigler F, Widmer F (2007) Bacterial community structures in honeybee intestines and their response to two insecticidal proteins. FEMS Microbiol. Ecol. 59:600-610.

Biesmeijer JC, Roberts SPM, Reemer M, et al. (2006) Parallel declines in pollinators and insect-pollinated plants in Britain and the Netherlands. Science 313:351-354.

Bisgaard M (1993) Ecology and significance of Pasteurellaceae in animals. Zbl. Bakt.- Int. J. Med. M 279:7-26.

Bromenshenk JJ, Henderson CB, Wick CH, et al. (2010) Iridovirus and microsporidian linked to honey bee colony decline. PLoS ONE 5:e13181.

Brucker RM, Bordenstein SR (2012) The roles of host evolutionary relationships (Genus: Nasonia) and development in structuring microbial communities. Evolution 66:349-362.

Brune A, Friedrich M (2000) Microecology of the termite gut: structure and function on a microscale. Curr. Opin. Microbiol. 3:263-269.

Cameron SA, Lozier JD, Strange JP, et al. (2011) Patterns of widespread decline in North American bumble bees. Proc. Natl. Acad. Sci. USA 108:662-667.

Chandler JA, Lang JM, Bhatnagar S, Eisen JA, Kopp A (2011) Bacterial communities of diverse Drosophila species: ecological context of a host-microbe model system. PLoS Genet. 7:e1002272.

Chapman RF (1998) The insects: structure and function. Cambridge University Press, Cambridge, UK.

Chow J, Lee SM, Shen Y, Khosravi A, Mazmanian SK (2010) Host-bacterial symbiosis in health and disease. In: Mucosal immunity, pp. 243-274. Elsevier Academic Press Inc, San Diego. 24

Christensen H, Bisgaard M (2008) Taxonomy and biodiversity of members of Pasteurellaceae. In: Pasteurellaceae: Biology, genomics and molecular aspects (eds. Kuhnert P, Christensen H), pp. 1-26. Caister Academic Press, Norfolk, UK.

Cirimotich CM, Dong YM, Clayton AM, et al. (2011) Natural microbe-mediated refractoriness to Plasmodium Infection in Anopheles gambiae. Science 332:855-858.

Cox-Foster DL, Conlan S, Holmes EC, et al. (2007) A metagenomic survey of microbes in honey bee colony collapse disorder. Science 318:283-287.

Danforth BN, Sipes S, Fang J, Brady SG (2006) The history of early bee diversification based on five genes plus morphology. Proc. Natl. Acad. Sci. USA 103:15118-15123.

Dehority BA (1997) Foregut fermentation. In: Gastrointestinal microbiology (eds. Mackie RI, White BA, Isaacson RE), pp. 39-83. Chapman & Hall, New York, NY.

Dillon RJ, Dillon VM (2004) The gut bacteria of insects: nonpathogenic interactions. Annu. Rev. Entomol. 49:71-92.

Dubos RJ, Schaedler RW (1960) The effect of the intestinal flora on the growth rate of mice, and on their susceptibility to experimental infections. J. Exp. Med. 111:407-417.

Falk PG, Hooper LV, Midtvedt T, Gordon JI (1998) Creating and maintaining the gastrointestinal ecosystem: what we know and need to know from gnotobiology. Microbiol. Mol. Biol. Rev. 62:1157-1170.

Frese SA, Benson AK, Tannock GW, et al. (2011) The evolution of host specialization in the vertebrate gut symbiont Lactobacillus reuteri. PLoS Genet. 7:e1001314.

Gilliam M (1997) Identification and roles of non-pathogenic microflora associated with honey bees. FEMS Microbiol. Lett. 155:1-10.

Gordon HA, Pesti L (1971) Gnotobiotic animal as a tool in study of host microbial relationships. Bacteriol. Rev. 35:390-429.

Haydak MH (1957) Changes with age in the appearance of some internal organs of the honeybee. Bee World 38:197-207.

Haydak MH (1970) Honey bee nutrition. Annu. Rev. Entomol. 15:143-156.

Hedlund BP, Kuhn DA (2006) The genera Simonsiella and Alysiella. In: The prokaryotes, pp. 828 -839. Springer-Verlag, New York, NY.

Heijtza RD, Wang SG, Anuar F, et al. (2011) Normal gut microbiota modulates brain development and behavior. Proc. Natl. Acad. Sci. USA 108:3047-3052. 25

Higes M, Martin-Hernandez R, Garrido-Bailon E, et al. (2009) Honeybee colony collapse due to Nosema ceranae in professional apiaries. Environ. Microbiol. Rep. 1:110-113.

Hongoh Y, Deevong P, Inoue T, et al. (2005) Intra- and interspecific comparisons of bacterial diversity and community structure support coevolution of gut microbiota and termite host. Appl. Environ. Microbiol. 71:6590-6599.

Hongoh Y, Sharma VK, Prakash T, et al. (2008) Complete genome of the uncultured Termite Group 1 bacteria in a single host protist cell. Proc. Natl. Acad. Sci. USA 105:5555-5560.

Hooper LV (2009) Do symbiotic bacteria subvert host immunity? Nat. Rev. Microbiol. 7:367-374.

Hooper LV, Midtvedt T, Gordon JI (2002) How host-microbial interactions shape the nutrient environment of the mammalian intestine. Annu. Rev. Nutr. 22:283-307.

Hosokawa T, Kikuchi Y, Nikoh N, Shimada M, Fukatsu T (2006) Strict host-symbiont cospeciation and reductive genome evolution in insect gut bacteria. PLoS Biol. 4:1841- 1851.

Hugenholtz P (2002) Exploring prokaryotic diversity in the genomic era. Genome Biol. 3:reviews0003.0001-0003.0008.

Hughes DP, Pierce NE, Boomsma JJ (2008) Social insect symbionts: evolution in homeostatic fortresses. Trends Ecol. Evol. 23:672-677.

Jeyaprakash A, Hoy MA, Allsopp MH (2003) Bacterial diversity in worker adults of Apis mellifera capensis and Apis mellifera scutellata (Insecta:Hymenoptera) assessed using 16S rRNA sequences. J. Invertebr. Pathol. 84:96-103.

Klein AM, Vaissiere BE, Cane JH, et al. (2007) Importance of pollinators in changing landscapes for world crops. Proc. R. Soc. B 274:303-313.

Koch H, Schmid-Hempel P (2011a) Bacterial communities in central European bumblebees: low diversity and high specificity. Microb. Ecol. 62:121-133.

Koch H, Schmid-Hempel P (2011b) Socially transmitted gut microbiota protect bumble bees against an intestinal parasite. Proc. Natl. Acad. Sci. USA. 108:19288-19292.

Kuhn DA, Gregory DA, Buchanan GE, Nyby MD, Daly KR (1978) Isolation, characterization, and numerical taxonomy of Simonsiella strains from oral cavities of cats, dogs, sheep, and humans. Arch. Microbiol. 118:235-241.

Lan R, Reeves PR (2006) Evolution of enteric pathogens. In: Evolution of microbial pathogens (eds. Seifert HS, DiRita VJ), pp. 273-296. ASM Press, Washington, D.C. 26

Lee YK, Mazmanian SK (2010) Has the microbiota played a critical role in the evolution of the adaptive immune system? Science 330:1768-1773.

Ley RE, Hamady M, Lozupone C, et al. (2008) Evolution of mammals and their gut microbes. Science 320:1647-1651.

Ley RE, Peterson DA, Gordon JI (2006) Ecological and evolutionary forces shaping microbial diversity in the human intestine. Cell 124:837-848.

Lombardo MP (2008) Access to mutualistic endosymbiotic microbes: an underappreciated benefit of group living. Behav. Ecol. Sociobiol. 62:479-497.

Mackie RI (1997) Gut environment and evolution of mutualistic fermentative digestion. In: Gastrointestinal microbiology (eds. Mackie RI, White BA, Isaacson RE), pp. 13-38. Chapman & Hall, New York, NY.

Martinson VG, Danforth BN, Minckley RL, et al. (2011) A simple and distinctive microbiota associated with honey bees and bumble bees. Mol. Ecol. 20:619-628.

Michener CD (1974) The social behavior of the bees: a comparative study. The Belknap Press of Harvard University Press, Cambridge, MA.

Michener CD (2007) The bees of the world. Johns Hopkins University Press, Baltimore, MD.

Minnick MF, Anderson BE (2006) The genus Bartonella. In: The prokaryotes, pp. 467- 492. Springer-Verlag, New York, NY.

Mohr KI, Tebbe CC (2006) Diversity and phylotype consistency of bacteria in the guts of three bee species (Apoidea) at an oilseed rape field. Environ. Microbiol. 8:258-272.

Mohr KI, Tebbe CC (2007) Field study results on the probability and risk of a horizontal gene transfer from transgenic herbicide-resistant oilseed rape pollen to gut bacteria of bees. Appl. Microbiol. Biot. 75:573-582.

Nalepa CA, Bignell DE, Bandi C (2001) Detritivory, coprophagy and the evolution of digestive mutualisms in Dictyoptera. Insect. Soc. 48:194-201.

Ochman H, Worobey M, Kuo CH, et al. (2010) Evolutionary relationships of wild hominids recapitulated by gut microbial communities. PLoS Biol. 8:e1000546.

Ohkuma M, Sato T, Noda S, et al. (2007) The candidate phylum 'Termite Group 1' of bacteria: phylogenetic diversity, distribution, and endosymbiont members of various gut flagellated protists. FEMS Microbiol. Ecol. 60:467-476. 27

Olofsson TC, Vasquez A (2008) Detection and identification of a novel lactic acid bacterial flora within the honey stomach of the honeybee Apis mellifera. Curr. Microbiol. 57:356-363.

Olofsson TC, Vasquez A (2009) Phylogenetic comparison of bacteria isolated from the honey stomachs of honey bees Apis mellifera and bumble bees Bombus spp. J. Apicult. Res. 48:233-237.

Pinto-Tomas AA, Anderson MA, Suen G, et al. (2009) Symbiotic nitrogen fixation in the fungus gardens of leaf-cutter ants. Science 326:1120-1123.

Robinson GE, Grozinger CM, Whitfield CW (2005) Sociogenomics: Social life in molecular terms. Nat. Rev. Genet. 6:257-U216.

Roulston TH, Cane JH (2000) Pollen nutritional content and digestibility for animals. Plant Syst. Evol. 222:187-209.

Savage DC (1977) Microbial ecology of gastrointestinal tract. Annu. Rev. Microbiol. 31:107-133.

Scardovi V, Troatelli LD (1969) New species of bifid bacteria from Apis mellifera L. and Apis indica F. a contribution to the taxonomy and biochemistry of the genus Bifidobacterium. Zentralbl. Bakteriol. Parasitenkd. Infektionskr. Hyg. 123:64-88.

Schmid-Hempel P (1998) Parasites in social insects. Princeton University Press, Princeton, NJ.

Seeley TD (1985) Honeybee ecology. Princeton University Press, Princeton, NJ.

Shimanuki H, Knox DA (2000) Diagnosis of honey bee diseases. Agriculture handbook no. AH-690. USDA, Washington, DC.

Shin SC, Kim SH, You H, et al. (2011) Drosophila microbiome modulates host developmental and metabolic homeostasis via insulin signaling. Science 334:670-674.

Staley JT, Konopka A (1985) Measurement of in situ activities of non-photosynthetic microorganisms in aquatic and terrestrial habitats. Annu. Rev. Microbiol. 39:321-346.

Tellam RL (1996) The peritrophic matrix. In: Biology of the insect midgut (eds. Lehane MJ, Billingsley PF), pp. 86-114. Chapman & Hall, London.

Vallet-Gely I, Lemaitre B, Boccard F (2008) Bacterial strategies to overcome insect defences. Nat. Rev. Microbiol. 6:302-313. vanEngelsdorp D, Evans JD, Saegerman C, et al. (2009) Colony collapse disorder: a descriptive study. PLoS ONE 4:e6481. 28 vanEngelsdorp D, Underwood R, Caron D, Hayes J, Jr. (2007) An estimate of managed colony losses in the winter of 2006-2007: a report commissioned by the apiary inspectors of America. Am. Bee J. 147:599-603.

Vasanthakumar A, Handelsman J, Schloss PD, Bauer LS, Raffa KF (2008) Gut microbiota of an invasive subcortical beetle, Agrilus planipennis Fairmaire, across various life stages. Environ. Entomol. 37:1344-1353.

Walter J, Britton RA, Roos S (2011) Host-microbial symbiosis in the vertebrate gastrointestinal tract and the Lactobacillus reuteri paradigm. Proc. Natl. Acad. Sci. USA 108:4645-4652.

White GF (1906) The bacteria of the apiary: with special reference to bee diseases. G. P. O., Washington, D. C.

White PB (1921) The normal bacterial flora of the bee. J. Pathol. Bacteriol. 24:64-U69.

Wilson EO (1971) The insect societies. The Belknap Press of Harvard University Press, Cambridge, MA.

Wostmann BS (1996) Germfree and gnotobiotic animal models: background and applications. CRC Press, Boca Raton, FL.

29

APPENDIX A: A SIMPLE AND DISTINCTIVE MICROBIOTA ASSOCIATED WITH

HONEY BEES AND BUMBLE BEES

Published: Molecular Ecology (2011) 20:619-628.

30

JOHN WILEY AND SONS LICENSE TERMS AND CONDITIONS Mar 28, 2012

This is a License Agreement between Vince Martinson ("You") and John Wiley and Sons ("John Wiley and Sons") provided by Copyright Clearance Center ("CCC"). The license consists of your order details, the terms and conditions provided by John Wiley and Sons, and the payment terms and conditions.

All payments must be made in full to CCC. For payment instructions, please see information listed at the bottom of this form.

License Number 2877881140250

License date Mar 28, 2012

Licensed content publisher John Wiley and Sons

Licensed content publication Molecular Ecology

Licensed content title A simple and distinctive microbiota associated with honey bees and bumble bees

Licensed content author VINCENT G. MARTINSON,BRYAN N. DANFORTH,ROBERT L. MINCKLEY,OLAV RUEPPELL,SALIM TINGEK,NANCY A. MORAN

Licensed content date Feb 1, 2011

Start page 619

End page 628

Type of use Dissertation/Thesis

Requestor type Author of this Wiley article

Format Electronic

Portion Full article

Will you be translating? No

31

ABSTRACT

Specialized relationships with bacteria often allow animals to exploit a new diet by providing a novel set of metabolic capabilities. Bees are a monophyletic group of

Hymenoptera that transitioned to a completely herbivorous diet from the carnivorous diet of their wasp ancestors. Recent culture-independent studies suggest that a set of distinctive bacterial species inhabits the gut of the honey bee, Apis mellifera. Here we survey the gut microbiotae of diverse bee and wasp species to test whether acquisition of these bacteria was associated with the transition to herbivory in bees generally. We found that most bee species lack phylotypes that are the same or similar to those typical of A. mellifera, rejecting the hypothesis that this dietary transition was symbiont- dependent. The most common bacteria in solitary bee species are a widespread phylotype of Burkholderia and the pervasive insect associate, Wolbachia. In contrast, several social representatives of corbiculate bees do possess distinctive bacterial phylotypes. Samples of A. mellifera harbored the same microbiota as in previous surveys, and closely related bacterial phylotypes were identified in two Asian honey bees (Apis andreniformis and Apis dorsata) and several bumble bee (Bombus) species. Potentially, the sociality of Apis and Bombus species facilitates symbiont transmission and thus is key to the maintenance of a more consistent gut microbiota. Phylogenetic analyses provide a more refined taxonomic placement of the A. mellifera symbionts.

32

INTRODUCTION

Recent non-culture-based 16S rRNA and metagenomic surveys have revealed that the guts of honey bees (Apis mellifera) from the United States, Australia, South Africa,

Germany, Sweden, and Switzerland harbor representatives of the same eight bacterial phylotypes (Fig. 1) (Babendreier et al. 2007; Cox-Foster et al. 2007; Jeyaprakash et al.

2003; Mohr & Tebbe 2006; Olofsson & Vasquez 2008). This result is contrary to results of previous culture-based surveys, which showed a complex and inconsistent microbiota of over 6,000 bacterial strains in A. mellifera guts (Gilliam 1997). The eight characteristic phylotypes constitute ~95% of bacterial 16S rRNA sequences cloned from

A. mellifera abdomens, and represent five bacterial classes (Fig. S1) (Cox-Foster et al.

2007). The association between these bacteria and A. mellifera is highly conserved despite environmental, geographic, and subspecies differences of hosts (Babendreier et al. 2007; Jeyaprakash et al. 2003; Mohr & Tebbe 2006; Olofsson & Vasquez 2008).

These observations suggest that A. mellifera has a coevolved symbiotic relationship with some or all of these bacteria, and that the associations are maintained across generations of the host.

Many insects (i.e. termites, leaf-cutter ants, aphids, etc.) have evolved non- pathogenic, persistent associations with microorganisms that provide benefits to both partners (Hongoh et al. 2008; Moran et al. 2008; Pinto-Tomas et al. 2009). Nutritional symbioses represent the majority of known insect-microbe mutualisms, and are found in multiple insect lineages that subsist on unusual or low-nutrient diets (e.g. sap, blood, 33 detritus, wood) (Moran et al. 2008). For example, acquisition of a nutrient-provisioning bacterial symbiont in a sharpshooter (Hemiptera: Cicadellidae: Cicadellinae) ancestor coincided with the transition from a diet of phloem sap to the comparatively nutrient-poor diet of xylem sap (Moran 2007), illustrating the potential impact microbial symbiosis can have on the ecology of the host insect.

Bees arose from predatory apoid wasps in the early to mid-Cretaceous, transitioning from the ancestral carnivorous life-style to herbivory, depending upon plant pollen as their sole protein source (Danforth et al. 2006). Pollen cytoplasm is nutrient- rich but is protected by a carbohydrate exine that is refractory to most digestive systems; nevertheless, diverse insects and vertebrates are able to use pollen as food (Roulston &

Cane 2000). One hypothesis for the origin of the distinct dietary habits and ecology observed in bees is that these were facilitated by the acquisition of a distinctive microbiota capable of supplementing the host with necessary nutrients, or assisting in the digestion of pollen. If so, bacteria closely related to the lineages found in A. mellifera might be ubiquitous across all bee species, which comprise a monophyletic clade within the superfamily Apoidea. Outside of A. mellifera, few bee species (and specimens) have been screened, so the distribution of these bacterial types in bees has not been clear

(Mohr & Tebbe 2006). In the current study, we further characterize the A. mellifera microbiota and determine the distribution of the same or related lineages in gut communities from members of all but one bee family, the wasp clade sister to all bees, and a distantly related wasp that has independently shifted to an exclusively pollen and nectar diet. 34

MATERIALS AND METHODS

Specimen collection and DNA extraction

Bee and wasp specimens were collected into 95% ethanol and stored at -80°C until DNA was extracted (Table S1). Specimens were collected randomly as they were foraging at flowers. The two A. mellifera samples were prepared from adults collected within a single healthy hive in January 2008 at the Carl Hayden Bee Research Center

(Tucson, AZ). For the A. mellifera hive sample a total of eighty adult digestive tracts

(crop, midgut, ileum and rectum) were dissected in 10mM MgSO4, homogenized with iris dissection scissors followed by a 1.5mL tube mini-pestle, and passed through an 8µm filter to remove most host cells.

Whole abdomens were removed and macerated with sterile dissection scissors for all specimens except larger bees (i.e. Xylocopa sp. and Bombus sp.). For large specimens, the abdominal exoskeleton was burdensome to fit into a 1.5mL tube and prevented complete homogenization of the digestive tract so it was not used in the DNA extraction.

For specimens stored in ethanol, DNA was extracted using the Gentra PureGene Kit

(Qiagen Inc., Valencia, CA, USA) preceded by a 30-minute lysozyme incubation

(outlined in the DNeasy kit, Qiagen Inc., Valencia, CA, USA) to lyse Gram-positive bacterial cells. For specimens stored in lysis buffer (.1M Tris-HCl, .1M EDTA, .01M

NaCl, .005% SDS, 500µg/mL Proteinase K), DNA was extracted with a standard phenol/chloroform reaction, followed by ethanol precipitation. Resulting DNA was analyzed for molecular weight and quantity on a 1.0% agarose gel (80 V, 2 hrs). 35

16S rRNA PCR and cloning

DNA samples for two A. mellifera samples, 11 other bee species, and three wasp species were sent to the Joint Genome Institute (JGI, Walnut Creek, CA). At JGI, partial

16S rRNA gene sequences were amplified from each specimen using universal bacterial primers 27F (5’-AGAGTTTGATCCTGGCTCAG-3’) and 1391R (5’-

GACGGGCRGTGWGTRCA-3’) with reaction conditions listed in the JGI SOP 16S18S rRNA PCR Library Creation protocol as in Warnecke et al. (2007). PCR products were screened on a 1% agarose gel (100 V, 1 hr) for the expected size along with a size ladder and positive (known bacterial DNA) and negative (no template DNA) controls.

Resulting PCR products were cloned using the TOPO TA Cloning Kit (Invitrogen,

Carlsbad, CA, USA), and 384 clones were chosen for sequencing (forward and reverse).

Sequences were automatically edited and assembled in the JGI sequencing pipeline.

Corbiculate 16S rRNA PCR and cloning

The corbiculate bees lack scopa (a dense mass of long, branched setae for pollen collection) on their hind legs, instead their tibia is modified into a corbicula, which is also known as the “pollen basket”. Corbiculates form a clade within the bee family Apidae that contains several of the most economically important bee species (i. e. A. mellifera the honey bee) and the most complex eusocial societies among the bees. There are four tribes within the corbiculates; the solitary/communal orchid bees (Euglossini), the primitively eusocial bumble bees (Bombini), and the highly eusocial stingless (Meliponini) and honey bees (Apini) (Kawakita et al. 2008). We obtained additional sequences from the 36 microbiota of several corbiculate bee species. A universal 16S rRNA PCR was performed at 20-µl reaction volume containing template DNA (~75ng), 0.8 U Taq DNA Polymerase

(New England BioLabs, Ipswich, MA, USA), 0.25 mM of each dNTP, 1x PCR buffer,

0.1 µM 27F-short primer (5’- GAGTTTGATCCTGGCTCA-3’) and 0.1 µM 1507R primer (5’-TACCTTGTTACGACTTCACCCCAG -3’). Cycling conditions were as follows: 94°C for 4 min; 35 cycles of 94°C for 30 s, 58°C for 30 s, 72°C for 90 s; and a final extension at 72°C for 10 min. PCR products were cloned into E. coli JM109 competent cells using the pGEM-T Easy Vector System (Promega Corp., Madison, WI,

USA) per manufacturer’s instructions, and grown on LB agar containing 100 µg/ml ampicillin, and appended with X-gal and IPTG for blue/white screening. Colony PCR was performed on randomly selected transformed colonies with vector primers M13F (5’-

CAGGAAACAGCTATGAC-3’) and M13R (5’- GTAAAACGACGGCCAG-3’) using the previously mentioned cycling conditions except with an annealing temperature of

55°C. PCR products were screened for the expected size on a 1% agarose gel (95 V, 50 min), and cleaned with a 15-minute incubation at 37°C with 0.2µl of Exo I and 0.2µl CIP

(New England BioLabs, Ipswich, MA, USA) followed by 15 minutes at 80°C. For each specimen, 94 clones were sequenced at the Arizona Research Labs, Genetics Core

(University of Arizona) using both forward and reverse M13 primers.

Chimeric sequence removal, classification, and heatmap construction

The 16S rRNA library from each specimen was aligned in the Green Genes online platform using the NAST alignment program (DeSantis et al. 2006), and potential 37 chimeric sequences were identified with Bellerophon (Huber et al. 2004). Chimeric sequences were manually verified with the online RDP Chimera Detection program (Cole et al. 2003) and removed from the dataset. The Ribosomal Database Project’s

Pyrosequencing Pipeline Infernal aligner (Nawrocki et al. 2009) (Bacteria alignment model) was used to align the resulting non-chimeric sequences generated in the current study, as well as those collected from previous publications exploring the A. mellifera microbiota.

Each clone library for the 16 JGI samples yielded between 182 and 360 near full- length 16S rRNA sequences, providing a total of 4799 sequences after the initial quality control. This dataset was expanded with an additional 395 sequences from five species of corbiculate bees; these sequences were obtained at the DNA sequencing facility at the

University of Arizona. In addition, we included 628 published 16S rRNA sequences, mostly shorter in length and mostly sampled from A. mellifera.

Resulting alignments were clustered into operational taxonomic units (OTUs) with RDP’s Complete Linkage Clustering method (Cole et al. 2009), using a 0.97 sequence similarity cutoff for designating phylotypes. OTUs were classified with the

RDP classifier, and resulting sequences with less than 70% bootstrap support at their

Class assignment were removed as well as any chloroplast sequences. OTUs were compared to GenBank with BLASTn to identify their top hit. The OTU frequency within bee and wasp specimens was visualized in a heatmap, created with the “heatmap.2” program within the gplots package for R.

38

Diagnostic PCR screening

DNA was extracted from bee and wasp samples (Table S1) using the methods described above. The following PCR primers (with annealing temperatures and extension times) were used to selectively amplify seven of the bacterial types: Alpha-1

(5’-CAAGTCGAACGCACTYTTCG-3’) and 1507R (5’-

TACCTTGTTACGACTTCACCCCAG -3’), 58°C, 1.5 min; Alpha-2.1-120F (5’-

GTAGGGATCTGTCCATAAGAG-3’) and Alpha-2.1-806R (5’-

GCTCCGACACTAAACAACTAGG-3’), 54°C, .5 min; Alpha-2.2-180F (5’-

GCCTGAGGGCCAAAGGAG-3’) and Alpha-2.2-702R (5’-

GCGTCAGTTCCGAGCCAGG-3’), 61°C, .5 min; Beta (5’-

CTTAGAGATAGGAGAGTG-3’) and 1507R, 50°C, .5 min; Firm-4 (5’-

AGTCGAGCGCGGGAAGTCA-3’) and 1507R, 58°C, 1.5 min; Firm-5qtF (5’-

GGAATACTTCGGTAGGAA-3’) and Firm-5qtR (5’-CTTATTTGGTATTAGCACC-

3’), 52°C, .5 min; Gamma1 (5’-GTATCTAATAGGTGCATCAATT-3’) and 1507R,

54°C, 1.5 min. The lack of sufficiently specific primer sites for both the Gamma-2 and

Bifido phylotypes precluded us from screening for these phylotypes in diagnostic PCR screens. PCRs were performed in 20 µl reactions with cycling conditions as above with positive (DNA from an A. mellifera gut) and negative (distilled water) control reactions.

PCR products were screened for the expected size on a 1% agarose gel (95 V, 50 min).

39

Phylogenetic analysis of the A. mellifera phylotypes

Sequences were aligned in the RDP Pyrosequencing Pipeline Infernal aligner

(Nawrocki et al. 2009) and manually refined using Mesquite (Maddison & Maddison

2009). Phylogenetic trees (GTR+I+γ model and 100 bootstrap replicates) were computed using a maximum likelihood framework in RAxML (Stamatakis 2006) with the CIPRES

Portal (Miller et al. 2009).

Nucleotide accession numbers

The 16S rRNA gene sequences determined in this study are in GenBank under the numbers HM108361 through HM113359, and the Cox-Foster et al. (2007) sequences are

HM107875 through HM108360.

RESULTS

Bacterial microbiota profiles of diverse bee and wasp species

After culling chimeric, chloroplast, and unassigned sequences, we retained a final dataset of 5604 16S rRNA sequences, representing 4999 near-full length sequences acquired in this study, plus 605 published sequences, mostly from A. mellifera and mostly shorter in length. The 4999 sequences were collected from a total of 20 bee species representing the phylogenetic diversity of bees, with more intensive sampling of the corbiculates, the clade containing A. mellifera. In addition, three wasp species were included as outgroups; one of these (Paragia vespiformis, Masaridae) represents an 40 independent origin of pollen-feeding, allowing us to further examine whether particular bacterial types are associated with this feeding habit. A total of 146 Operational

Taxonomic Units (OTUs) with a 3% distance cutoff were obtained after clustering. Of those, 62 OTUs (42%) were singletons, and 91 OTUs (62%) were found in only one specimen.

Major constituents of the communities

Almost half of the sequences (2416/5604) fell within a single OTU with >99% identity to database sequences of Burkholderia cepacia. The B. cepacia OTU was recovered from all specimens surveyed in this study except for the A. mellifera hive sample (20/21). Wolbachia was also abundant in the apoid wasp, Philanthus gibbosus

(Crabronidae), and in several bees: Rediviva saetigera (Melittidae), Agapostemon virescens (Halictidae), and Colletes inaequalis (Colletidae).

Apis mellifera

Almost all (>98%, 267/271 & 263/267 sequences) bacterial 16S rRNAs detected in both A. mellifera samples corresponded to one of the eight major groups of the previously described A. mellifera microbiota (Cox-Foster et al. 2007). The Alpha-2.2 and the Gamma-2 phylotypes were not detected in this screen. Sequences with top

BLASTn hits in GenBank to the Beta, Bifido, Firm-4, Firm-5, and the Gamma-1 phylotypes were the most abundant. Both A. mellifera communities were dominated by the Firm-5 phylotype, in contrast to previous surveys of the A. mellifera microbiota, 41 which were dominated by the Gamma-1 (Cox-Foster et al. 2007). This difference may reflect our use of lysozyme in DNA extractions, resulting in more efficient extraction of

Gram-positive bacteria, or it may reflect differences in developmental stages sampled.

We note that even though sampling methods were very different for the two A. mellifera samples, the profiles of symbiont abundances were similar.

The genera Apis and Bombus (corbiculate clade)

Some of the typical A. mellifera OTUs were recovered from other Apis and from

Bombus species (both in the corbiculate clade), though none had the profile characteristic of A. mellifera itself. The A. dorsata microbiota included three major OTUs, a sequence with a top BLASTn hit in GenBank to the A. mellifera Alpha-1 phylotype, plus two phylotypes of Burkholderia. The majority of sequences from A. andreniformis (63%,

49/78 sequences) corresponded to a single OTU nearly identical (>96% sequence identity) to the A. mellifera Alpha-1 phylotype. The Bombus sonorus microbiota was dominated (96%, 75/78 sequences) by a single OTU closely related to the A. mellifera

Beta phylotype (94% sequence identity). Bombus sp. from Montana and Bombus impatiens from California were both dominated (87%, 70/80 sequences and 76%, 54/71 sequences) by the B. cepacia OTU.

42

Survey of the A. mellifera gut microbiota in phylogenetically diverse bee

species

In order to further examine the host range of phylotypes corresponding to typical

A. mellifera associates, we used diagnostic PCR on a panel of bee species to survey the presence or absence of seven of the common associates. Positive amplifications for any of these sequences were obtained only within the genera Apis and Bombus (Fig. S2). A. mellifera specimens had the highest percent of positive reactions of any species (82%,

46/56 reactions). Most negative results for A. mellifera were from the Alpha-2.2, which was identified in only 2/8 specimens. A. dorsata specimens were universally infected with all symbionts except the Alpha-2.1, Firm-4, and Firm-5, while the A. andreniformis specimens were universally infected with Alpha-1 and had high infection rates of the

Beta (5/6), Firm-4 (5/6), and Gamma-1 (4/6). Ten of the twelve Bombus specimens were positive for at least one of these bacterial phylotypes, but the set of typical A. mellifera bacteria was less ubiquitous in Bombus than in Apis specimens. Specifically, among

Bombus samples screened, only the Gamma-1 (8/12), Beta (4/12), and Firm-5 (4/12) were detected. Because these diagnostic screens depended upon specific primers to amplify sequences closely related to the A. mellifera associates, mismatches due to sequence divergence within those primer regions could yield negative results.

Nevertheless, results of the PCR surveys are consistent with the near-universal absence of related sequences from non-corbiculate bee species in the 16S rRNA libraries, and both provide evidence that members of the characteristic A. mellifera microbiota are absent from most bee species. 43

Phylogenetic relationships of the A. mellifera gut microbiota

We analyzed the phylogenetic relationships of each of the A. mellifera bacterial types (Fig. S3). Six of the eight phylotypes correspond to single distinct clades, exclusively sampled from A. mellifera or close relatives; the Alpha-2 and Bifido types each correspond to two closely related lineages.

Alpha-1. The Alpha-1 sequences form a highly supported clade related to the genus Bartonella within the Rhizobiales. This clade also includes sequences from several herbivorous ant species (Russell et al. 2009).

Alpha-2. Two distinct clades were identified in the Alpha-2 phylogeny. Clade 1 clustered outside the Acetobacteraceae and included two published sequences associated with Drosophila, as well as sequences from the Xylocopa specimen. Clade 2 grouped within the genus Gluconobacter and contained only sequences from European A. mellifera. This clade contained sequences from each of the A. dorsata, Xylocopa,

Caupolicana, and Calliopsis specimens as well as Saccharibacter floricola, a bacterium isolated from pollen.

Beta. A well-supported clade corresponding to the Beta symbiont fell within

Neisseriaceae and grouped with the genera Simonsiella and Alysiella.

Bifido: The Bifido bacterial sequences from A. mellifera fell into two sister clusters within the clade representing the genus Bifidobacterium. Distributed among the A. mellifera sequences were several sequences from cultured isolates, originally obtained from A. mellifera guts. The cultured B. asteroides sequences form a clade with one set of

A. mellifera sequences, while B. coryneforme and B. indicum cluster with the sister clade. 44

Firm-4 and Firm-5. Firm-4 and Firm-5 both fell within the genus Lactobacillus with high support. Firm-4 forms a clade sister to the acidophilus clade, whereas the

Firm-5 fell within the acidophilus clade. The Firm-5 cluster contains several sequences

(GenBank accession numbers AY667698, AY667699, and AY667701) that are described as cultured from the guts of Italian A. mellifera.

Gamma-1 and Gamma-2. The Gamma-1 and Gamma-2 taxa together form a highly supported clade branching between the Enterobacteriaceae and Pasteurellaceae.

Within this separate clade are bacterial sequences cloned from an aphid, a fly, a ground beetle, and a wild boar.

DISCUSSION

Our results confirm that A. mellifera is consistently colonized by a distinctive set of bacterial species. Both A. mellifera samples in the 16S rRNA survey were dominated by these bacteria (Fig. 2), as in previous surveys based on different methodologies and on samples from different localities. Furthermore, our surveys using diagnostic PCR primers revealed consistent presence of these bacteria in a panel of A. mellifera specimens (Fig.

S2) and of many of these bacterial types in other corbiculate bees sampled, including members of the genera Apis and Bombus. In contrast, most other bee and wasp species completely lacked these phylotypes; a few bee species harbored one or two phylotypes associated with A. mellifera, but these phylotypes represented less than ten percent of their microbiota profile (Fig. 2, Fig. S2). Thus, we can dismiss the hypothesis that the A. 45 mellifera microbiota represents an ancestral set of bacterial associates that aided in the transition from predaceous wasp to pollenivorous (herbivorous) bee. Within the corbiculate clade, however, the phylogenetically restricted host range of these bacteria suggests an interdependent, coevolved association, as observed in other associations between insects and mutualistic microbes (Hosokawa et al. 2006; Moran et al. 2008;

Ohkuma et al. 2009; Warnecke et al. 2007).

The other Apis species and the Bombus species screened in this study frequently contained only a subset of the phylotypes common in the A. mellifera microbiota (Fig.

S2). These associations were detected even though specimens were collected from distant geographic locations, suggesting that the bacterial microbiota is transferred between generations and not acquired from the environment. While the majority of bee species are solitary, those within the genus Apis are highly eusocial and those within the genus

Bombus are primitively eusocial, living in colonies from tens of workers up to several thousand individuals (Wilson 1971). A potential mechanism for younger generations to acquire the characteristic microbiota could be the oral transfer of food between nestmates

(trophallaxis), found in eusocial Apinae (Michener 1974). Apis species have more frequent trophallaxis in comparison to Bombus (Wilson 1971). Another variable that may affect maintenance of the microbiota is colony establishment. Apis colonies split approximately in half to form two new colonies by colony fission (Michener 1974). In contrast, most Bombus colonies are established by individual females; these colonies grow and completely disband on an annual cycle (Michener 1974). This life cycle may impose a greater chance of microbiota loss possibly resulting in higher variation of 46 infection frequency among Bombus colonies than among Apis colonies (Fig. S2).

Overall, eusocial behavior may be one of the most important variables affecting the inheritance and maintenance of the corbiculate-specific microbiota.

While Apis and Bombus species may maintain a consistent microbiota, the other specimens in our study have strikingly different profiles of bacterial taxa in their guts

(Fig. 2) with no evident connection to phylogenetic position, collection location, or natural history. Most bee species are solitary and burrow into soil or plant materials to make nests, which are provisioned with pollen and nectar to provide the necessary nutrition for the offspring to mature (Michener 2007). Bees, thus, come into contact with many plant- and soil-associated microorganisms. In our 16S rRNA survey, a single phylotype of Burkholderia cepacia was present in nearly all bee specimens (20/21) and was prominent (>10% of sequences) in most samples (16/21) surveyed in depth. B. cepacia is commonly found in the soil rhizosphere and on plant surfaces (Compant et al.

2008); hence, this bacterium may be acquired from the environment. The other major group identified in bee and wasp specimens was Wolbachia, a common intracellular bacterium of insects, estimated to infect between 20 and 66% of all species (Duron et al.

2008; Hilgenboecker et al. 2008; Werren & Windsor 2000) and present in five of the bee and wasp species we surveyed.

Apis mellifera microbiota phylogenies

Each member of the A. mellifera characteristic gut microbiota forms one or two distinct clades, and sequence identities are high (>96%) within each such clade. Thus, 47 these lineages correspond to coherent bacterial species with wide geographic distribution

(Fig. 1, Fig. S3). Some of these clades (such as the Bifido) may consist of two or three distinct sister species. Phylogenetic classification may aid in future cultivation attempts, as the genera identified as close relatives have defined morphologies, biochemical profiles, and habitats. The Alpha-1 phylotype forms a sister clade to the genus

Bartonella, a group of intracellular pathogens that can infect many insects (Minnick &

Battisti 2009), raising the possibility that these bacteria are at least partially intracellular within A. mellifera. Closely related to the A. mellifera Alpha-1 were sequences associated with the evolution of herbivory in phylogenetically diverse ant lineages

(Russell et al. 2009). The Alpha-2 phylotype sequences are divided into two distinct clades within Acetobacteraceae. The first clade groups with several sequences found in the Xylocopa specimen and with two sequences previously found in a survey of

Drosophila-associated bacteria (Roh et al. 2008), forming a distinct clade separate from other named genera. The second clade clusters within the genus Gluconobacter and contains only sequences originating from European A. mellifera (Babendreier et al.

2007), suggesting a geographically heterogeneous distribution of these

Alphaproteobacteria. Saccharibacter floricola, a bacterium isolated from pollen, was nested within this second clade in addition to sequences from the A. dorsata, Xylocopa,

Caupolicana, and Calliopsis specimens, suggesting that this phylotype is associated with flowers (Jojima et al. 2004). The Beta symbiont forms a distinct clade close to the

Simonsiella/Alysiella cluster within Neisseriaceae. Simonsiella/Alysiella species have been isolated from the oral cavities of several mammals in which they are thought to be 48 non-pathogenic commensals (Hedlund & Kuhn 2006; Kuhn et al. 1978). The Bifido phylotype forms two sister clades within the genus Bifidobacterium and includes three

Bifidobacterium species previously cultured from the A. mellifera gut (Scardovi &

Troatelli 1969). Thus, the Bifido phylotype may correspond to two or three known species that can be cultured in the lab. The Firm-4 and Firm-5 phylotypes comprise two clades within and sister to the acidophilus group of the genus Lactobacillus. Several unpublished database sequences corresponding to the Firm-5 clade are described as being cultured from the A. mellifera gut (NCBI accessions AY667698, AY667699, and

AY667701). The Gamma-1 and Gamma-2 phylotypes are sister groups forming a clade that is sister to the family Pasteurellaceae and separate from Enterobacteriaceae. This new clade encompasses several database sequences associated with animals (Drosophila,

DQ980728; aphid, EU348326; beetle, EF608532; and wild boar, FJ612598). Potentially, members of this clade are widespread associates within insect guts (boars may consume insects).

Coevolution of gut bacteria and corbiculate bees

Sequences amplified from microbiota of Apis and Bombus formed clades together

(Fig. S3), raising the possibility of a coevolutionary association between these bee genera and a specific assemblage of bacteria. Screening several representative species from other members of this clade (the solitary/communal Euglossini and the highly eusocial

Meliponini) would provide further insights into the effects of social behavior upon this bacterial community. 49

Two Firm-5 sequences were amplified from the Colletes inaequalis specimen

(Fig. 2, Fig. S3), implying that this bacterial type is found in more disparate hosts.

However, these sequences were minor constituents in the C. inaequalis sample, raising the possibility that these bacteria were transients acquired at flowers, a shared habitat of diverse bee species.

Possible functions: Mutualist v Pathogen

The consistency of the association between A. mellifera and its characteristic microbiota suggests mutualistic relationships, at least with some members, though pathogenic effects have not been excluded.

Relatives of several members of the A. mellifera microbiota (Acetobacteraceae,

Bifidobacterium, Lactobacillus, and Simonsiella) produce short chain fatty acids such as lactic or acetic acid as waste products during the metabolism of carbohydrates (Biavati &

Mattarelli 2006; Hammes & Hertel 2006; Hedlund & Kuhn 2006; Kersters et al. 2006).

Assimilation of these compounds could supplement bee nutrition, just as short chain fatty acids produced by rumen microbes supply nearly all the energy requirements of ruminant mammals (Dehority 1997). Short chain fatty acids can be absorbed through the rectal wall in insects (Bradley 2008), and we have observed that the majority of the pollen and bacterial biomass within an adult A. mellifera is contained inside the rectum.

Overwintering Apis may obtain additional nutrition from these rectal bacteria, as consumed food is stored for longer periods of time within the rectum during winter months (Lindstrom et al. 2008; Spivak & Gilliam 1998). 50

Osmotic pressure change within the gut has been suggested as a main digestive mechanism capable of breaking pollen cells and releasing their contents (Roulston &

Cane 2000). Unlike many animals, A. mellifera mix pollen with a nectar/glandular solution before their larvae begin feeding (Michener 1974), which might facilitate digestion by introducing enzymes and/or microorganisms that can degrade resistant carbohydrates. Further, in A. mellifera pollen is stored for long periods of time, which changes its texture and nutritive qualities (Human & Nicolson 2006).

A. mellifera colonies offer a favorable environment for viral, bacterial, fungal, and protist pathogens because of the high density of individuals and the exchange of food among nestmates. Due to the risk of infection spreading through an entire hive, the A. mellifera immune system is expected to be highly developed. In contrast with this expectation, a comparison of the A. mellifera genome to Anopheles gambiae and

Drosophila melanogaster revealed a substantial decrease in recognizable immune pathway genes (71 in A. mellifera, 209 in A. gambiae, 196 in D. melanogaster) (Evans et al. 2006). This decrease in gene number was not observed in other A. mellifera gene families, suggesting that A. mellifera has a reduced immune flexibility (Evans et al. 2006;

Weinstock et al. 2006). However, social behaviors (i.e., grooming and removal of diseased brood) provide a major defensive barrier against pathogens (Evans & Spivak

2010; Spivak & Reuter 2001). Additionally, the A. mellifera gut microbiota could provide further protection against invading pathogens by producing inhibitory compounds or by monopolizing nutrients within the gut (Corr et al. 2007; Round & Mazmanian 2009).

51

CONCLUSION

In this survey of bees from a broad sample of families and subfamilies, members of the characteristic bacterial microbiota of A. mellifera were found to be absent from most species outside of the corbiculate clade, represented in our study by Apis and

Bombus. An earlier survey of short 16S rRNA segments from A. mellifera, Bombus terrestris and Osmia bicornis raised the possibility of a broader distribution in bees

(Mohr & Tebbe 2006). However, our more extensive survey suggests that most of the bacterial species identified from A. mellifera have a narrow association with Apis and

Bombus. Coevolution between the corbiculate bees and a characteristic microbiota may explain this association, but further screening of the other corbiculate tribes is needed to accept or reject this hypothesis.

This survey offers a closer view of the bacteria associated with bees; however, more sampling within and between bee species could provide a more complete understanding of the relationship between bees and their bacterial microbiota. We note that a similar set of bacterial phylotypes has been recovered from all non-culture-based studies of the A. mellifera microbiota, despite extremely different methods of sampling, amplification and sequencing (Babendreier et al. 2007; Cox-Foster et al. 2007;

Jeyaprakash et al. 2003; Mohr & Tebbe 2006; Olofsson & Vasquez 2008; this study).

Thus, although our sampling of other bee species is limited, and could be influenced by preservation method or PCR primers, our failure to retrieve the same or related phylotypes from most species sampled is striking. 52

Our results support the existence of a relatively simple microbiota in A. mellifera, an eminent model organism (Weinstock et al. 2006) and an ecologically influential and economically important insect (Morse & Calderone 2000). Learning more about this little known, but potentially essential, microbiota may have wide-reaching implications for our understanding of the basic biology of honey bees and for current practices in apiculture and agriculture.

ACKNOWLEDGMENTS

We thank Chih-Horng Kuo for assistance with the heatmap.2 program, Howard

Ochman for helpful insights, James Nieh for bee specimens, and Gloria DeGrandi-

Hoffman for access to hives at the Carl Hayden Bee Research Center. V. Martinson was supported by the National Science Foundation IGERT training grant in Comparative

Genomics to the University of Arizona, and additional research support came from NSF

0626716 to N. Moran. Some work was conducted by the U.S. Department of Energy

Joint Genome Institute, supported by the Office of Science of the U.S. Department of

Energy under Contract No. DE-AC02-05CH11231. O. Rueppell was supported by a grant from USDA-NIFA (#2010-65104-20533) and a RGB grant from the North Carolina

Biotechnology Center.

53

REFERENCES

Babendreier D, Joller D, Romeis J, Bigler F, Widmer F (2007) Bacterial community structures in honeybee intestines and their response to two insecticidal proteins. FEMS Microbiol. Ecol. 59:600-610.

Biavati B, Mattarelli P (2006) The family Bifidobacteriaceae. In: The prokaryotes, pp. 322 - 382. Springer-Verlag, New York, NY.

Bradley TJ (2008) Active transport in insect recta. J. Exp. Biol. 211:835-836.

Brochier C, Bapteste E, Moreira D, Philippe H (2002) Eubacterial phylogeny based on translational apparatus proteins. Trends Genet. 18:1-5.

Cole JR, Chai B, Marsh TL, et al. (2003) The ribosomal database project (RDP-II): previewing a new autoaligner that allows regular updates and the new prokaryotic taxonomy. Nucleic Acids Res. 31:442-443.

Cole JR, Wang Q, Cardenas E, et al. (2009) The ribosomal database project: improved alignments and new tools for rRNA analysis. Nucleic Acids Res. 37:D141-D145.

Compant S, Nowak J, Coenye T, Clement C, Barka EA (2008) Diversity and occurrence of Burkholderia spp. in the natural environment. FEMS Microbiol. Rev. 32:607-626.

Corr SC, Li Y, Riedel CU, et al. (2007) Bacteriocin production as a mechanism for the antfinfective activity of Lactobacillus salivarius UCC118. Proc. Natl. Acad. Sci. USA 104:7617-7621.

Cox-Foster DL, Conlan S, Holmes EC, et al. (2007) A metagenomic survey of microbes in honey bee colony collapse disorder. Science 318:283-287.

Danforth BN, Sipes S, Fang J, Brady SG (2006) The history of early bee diversification based on five genes plus morphology. Proc. Natl. Acad. Sci. USA 103:15118-15123.

Dehority BA (1997) Foregut fermentation. In: Gastrointestinal microbiology (eds. Mackie RI, White BA, Isaacson RE), pp. 39-83. Chapman & Hall, New York, NY.

DeSantis TZ, Hugenholtz P, Keller K, et al. (2006) NAST: a multiple sequence alignment server for comparative analysis of 16S rRNA genes. Nucleic Acids Res. 34:W394-W399.

54

Duron O, Bouchon D, Boutin S, et al. (2008) The diversity of reproductive parasites among arthropods: Wolbachia do not walk alone. BMC Biol. 6:12.

Evans JD, Aronstein K, Chen YP, et al. (2006) Immune pathways and defence mechanisms in honey bees Apis mellifera. Insect Mol. Biol. 15:645-656.

Evans JD, Spivak M (2010) Socialized medicine: individual and communal disease barriers in honey bees. J. Invertebr. Pathol. 103:S62-S72.

Gilliam M (1997) Identification and roles of non-pathogenic microflora associated with honey bees. FEMS Microbiol. Lett. 155:1-10.

Hammes WP, Hertel C (2006) The genera Lactobacillus and Carnobacterium. In: The prokaryotes, pp. 320 - 403. Springer-Verlag, New York, NY.

Hedlund BP, Kuhn DA (2006) The genera Simonsiella and Alysiella. In: The prokaryotes, pp. 828 -839. Springer-Verlag, New York, NY.

Hilgenboecker K, Hammerstein P, Schlattmann P, Telschow A, Werren JH (2008) How many species are infected with Wolbachia? - a statistical analysis of current data. FEMS Microbiol. Lett. 281:215-220.

Hongoh Y, Sharma VK, Prakash T, et al. (2008) Genome of an endosymbiont coupling N2 fixation to cellulolysis within protist cells in termite gut. Science 322:1108-1109.

Hosokawa T, Kikuchi Y, Nikoh N, Shimada M, Fukatsu T (2006) Strict host-symbiont cospeciation and reductive genome evolution in insect gut bacteria. PLoS Biol. 4:1841- 1851.

Huber T, Faulkner G, Hugenholtz P (2004) Bellerophon: a program to detect chimeric sequences in multiple sequence alignments. Bioinformatics 20:2317-2319.

Human H, Nicolson SW (2006) Nutritional content of fresh, bee-collected and stored pollen of Aloe greatheadii var. davyana (Asphodelaceae). Phytochemistry 67:1486-1492.

Jeyaprakash A, Hoy MA, Allsopp MH (2003) Bacterial diversity in worker adults of Apis mellifera capensis and Apis mellifera scutellata (Insecta:Hymenoptera) assessed using 16S rRNA sequences. J. Invertebr. Pathol. 84:96-103.

Jojima Y, Mihara Y, Suzuki S, et al. (2004) Saccharibacter floricola gen. nov., sp nov., a novel osmophilic acetic acid bacterium isolated from pollen. Int. J. Syst. Evol. Microbiol. 54:2263-2267.

55

Kawakita A, Ascher JS, Sota T, Kato M, Roubik DW (2008) Phylogenetic analysis of the corbiculate bee tribes based on 12 nuclear protein-coding genes (Hymenoptera:Apoidea:Apidae). Apidologie 39:163-175.

Kersters K, Lisdiyanti P, Kmagata K, Swings J (2006) The family Acetobacteraceae: the genera Acetobacter, Acidomonas, Asaia, Gluconacetobacter, Gluconobacter, and Kozakia. In: The prokaryotes, pp. 163-200. Springer-Verlag, New York, NY.

Kuhn DA, Gregory DA, Buchanan GE, Nyby MD, Daly KR (1978) Isolation, characterization, and numerical taxonomy of Simonsiella strains from oral cavities of cats, dogs, sheep, and humans. Arch. Microbiol. 118:235-241.

Lindstrom A, Korpela S, Fries I (2008) The distribution of Paenibacillus larvae spores in adult bees and honey and larval mortality, following the addition of American foulbrood diseased brood or spore-contaminated honey in honey bee (Apis mellifera) colonies. J. Invertebr. Pathol. 99:82-86.

Maddison WP, Maddison DR (2009) Mesquite: a modular system for evolutionary analysis.

Michener CD (1974) The social behavior of the bees: a comparative study. The Belknap Press of Harvard University Press, Cambridge, MA.

Michener CD (2007) The bees of the world, 2 edn. Johns Hopkins University Press, Baltimore, MD.

Miller MA, Holder MT, Vos, R, Midford PE, Liebowitz T, Chan L, Hoover P, Warnow T (2009) The CIPRES Portals. CIPRES. URL: http://www.phylo.org/sub_sections/portal. Accessed: February 2010.

Minnick MF, Battisti JM (2009) Pestilence, persistence and pathogenicity: infection strategies of Bartonella. Future Microbiol. 4:743-758.

Mohr KI, Tebbe CC (2006) Diversity and phylotype consistency of bacteria in the guts of three bee species (Apoidea) at an oilseed rape field. Environ. Microbiol. 8:258-272.

Moran N, McCutcheon JP, Nakabachi A (2008) Genomics and evolution of heritable bacterial symbionts. Annu. Rev. Genet. 42:165-190.

Moran NA (2007) Symbiosis as an adaptive process and source of phenotypic complexity. Proc. Natl. Acad. Sci. USA 104:8627-8633.

Morse RA, Calderone NW (2000) The value of honey bees as pollinators of U.S. crops in 2000. Bee Cult. 128:1-15. 56

Nawrocki EP, Kolbe DL, Eddy SR (2009) Infernal 1.0: inference of RNA alignments. Bioinformatics 25:1335-1337.

Ohkuma M, Noda S, Hongoh Y, Nalepa CA, Inoue T (2009) Inheritance and diversification of symbiotic trichonymphid flagellates from a common ancestor of termites and the cockroach Cryptocercus. Proc. R. Soc. B 276:239-245.

Olofsson TC, Vasquez A (2008) Detection and identification of a novel lactic acid bacterial flora within the honey stomach of the honeybee Apis mellifera. Curr. Microbiol. 57:356-363.

Pinto-Tomas AA, Anderson MA, Suen G, et al. (2009) Symbiotic nitrogen fixation in the fungus gardens of leaf-cutter ants. Science 326:1120-1123.

Roh SW, Nam YD, Chang HW, et al. (2008) Phylogenetic characterization of two novel commensal bacteria involved with innate immune homeostasis in Drosophila melanogaster. Appl. Environ. Microbiol. 74:6171-6177.

Roulston TH, Cane JH (2000) Pollen nutritional content and digestibility for animals. Plant Syst. Evol. 222:187-209.

Round JL, Mazmanian SK (2009) The gut microbiota shapes intestinal immune responses during health and disease. Nat. Rev. Immunol. 9:313-323.

Russell JA, Moreau CS, Goldman-Huertas B, et al. (2009) Bacterial gut symbionts are tightly linked with the evolution of herbivory in ants. Proc. Natl. Acad. Sci. USA 106:21236-21241.

Scardovi V, Troatelli LD (1969) New species of bifid bacteria from Apis mellifera L. and Apis indica F. a contribution to the taxonomy and biochemistry of the genus Bifidobacterium. Zentralbl. Bakteriol. Parasitenkd. Infektionskr. Hyg. 123:64-88.

Spivak M, Gilliam M (1998) Hygienic behaviour of honey bees and its application for control of brood diseases and varroa mites. Part I. Hygienic behaviour and resistance to American foulbrood. Bee World 79:124-134.

Spivak M, Reuter GS (2001) Resistance to American foulbrood disease by honey bee colonies Apis mellifera bred for hygienic behavior. Apidologie 32:555-565.

Stamatakis A (2006) RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics 22:2688-2690.

57

Warnecke F, Luginbuhl P, Ivanova N, et al. (2007) Metagenomic and functional analysis of hindgut microbiota of a wood-feeding higher termite. Nature 450:560-565.

Weinstock GM, Robinson GE, Gibbs RA, et al. (2006) Insights into social insects from the genome of the honeybee Apis mellifera. Nature 443:931-949.

Werren JH, Windsor DM (2000) Wolbachia infection frequencies in insects: evidence of a global equilibrium? Proc. R. Soc. Lond. B 267:1277-1285.

Wilson E (1971) The insect societies. The Belknap Press of Harvard University Press, Cambridge, MA.

58

Phylogeny of Bacterial Phyla

a Gammaproteobacteria “Gamma-1” Mycoplasmas ` Chlamydiales “Gamma-2” Betaproteobacteria _ Simonsiella/Alysiella “Beta”

¡ Alphaproteobacteria Spirochetes Bartonella Proteobacteria “Alpha-1” Acetobacteraceae Green sulfur Cyanobacteria “Alpha-2” Deinococcales Actinobacteria Bi!dobacterium Bacteroidetes “Bi!do” Firmicutes High G+C Lactobacillus Gram positives Low G+C “Firm-4” Gram positives Thermotogales “Firm-5” Aqui!cales

Figure 1 - Phylogenetic positions of A. mellifera bacterial gut microbiota associates. Phylogeny based on Brochier et al. (2002).

59

=*5$'

M"JJ"

S'8*$8#"P*$'(" !$*"

H+25"

!"P(++( F('J(P<*$4

A5#8/,0:(#%"&4(*"4%"

B,02"4/%"&)*C

HP*(%8 43 HP*(%8#"P*$'("

)))H+25"EI))))))H+25"E-,I))))H+25"E-,-)))))))!$*")))))))))))!(>(&8)))))))))F('JEK)))))))))F('JEL))))))M"JJ"EI)))M"JJ"E-

B

2

/I B. /B /B /-

I6

-/

I/

K. IK

-K/

-.K

6;.

6KG

666

-B-

6.L -/6

6.I

-L;

66B IB- 6K/ 6.L

-/I -;/

KB; !

"#$%&

',1./*&,

A,-25)&)* ;*%)&:,#)"6"

;*%)&-(00%+(#"

?%":")%"&,*536%"

A,-25)&),3,#5)

!"#"$%"&'()*%+,#-%)

;*%)&-(00%+(#"

7"0%465)&*"6(00"65)

9(:%'%'"&)"(6%$(#"

A,-25)&%-*"6%(3)

7,*0%6%)&2%)456(00"(

;*%)&"3:#(3%+,#-%)

.,00(6()&%3"(<5"0%)

78'"%)$*)"+,)<%2<#,

785')9):$##$)-..;

@10,4,*"&4"0%+,#3%4" !"#$%&&'(%)*

!/%0"36/5)&$%22,)5)

."5*,0%4"3"&1"##,=%

."00%,*)%)&)52"0*%35)

7()*(#"*%)&4,48(#(00%

!"#$%&'($')$*)"+,)-../

C8DEF84*$')$*)"+,)-../

./"012%,3&4"0%+,#3%45-

0$1"2'"3"45)$*)"+,)-..6

>($"4/%0(&,:,36,)6,-"

;$"*,)6(-,3&'%#()4(3) ?"4@<$A)9)=+8>448%)-..G =+8>448%)9)?"4@<$A)-..B -.*/$01,'

C8'#(P<+"*$)!$$4 !"#$%&'''''''''''''''''''''''''()*+$*, H%&'$%(&"$ H2(&"$ 7"4"'(&"$ R25$P(&"$ C'"#'8%(&"$ N"4128&"(&"$ 7$+(**(&"$)4,4, Q"+(P*(&"$ C8++$*(&"$ 7$O"P5(+(&"$

!(

!' "3

!$$4 !.*51*2+&'$2'("#)%* H28(&$" 60

Figure 2 - Frequency of bacterial types in the microbiotae of 21 specimens (three wasps and 16 bees) represented as a heatmap. (a) OTUs correspond to one of the distinctive A. mellifera bacterial phylotypes. Column ‘n’ denotes number of bacterial 16S rRNA sequences for each specimen. Lower section displays the presence or absence of each bacterial type in previously published studies. (b) OTUs recovered that did not have a top BLASTn hit to the A. mellifera bacterial types.

61

SUPPLEMENTARY RESULTS

Arizona hive Arizona individual

n = 267 n = 271 Not characteristic bacteria

Figure S1 - Proportional representation of bacterial 16S rRNA sequences from the individual sample (271 sequences) and hive sample (267 sequences) of adult A. mellifera. In each case, >98% of recovered sequences belong to one of the eight characteristic associates. Color key is the same as in Fig. 1.

62

Genus Species Family Alpha-1 Alpha-2.1 Alpha-2.2 Beta Firm-4 Firm-5 Gamma-1

Andrena species 1 Calliopsis puellae Calliopsis puellae Calliopsis subalpinus Calliopsis subalpinus Perdita ashmeadi Perdita ashmeadi Perdita callicerata Perdita exclamans Perdita exclamans Perdita larreae Perdita larreae Perdita larreae Perdita larreae Perdita larreae Perdita munita Perdita munita Perdita semicaerulea Perdita semicaerulea Perdita semicaerulea Andrenidae Perdita semicaerulea Perdita semicaerulea Protoandrena mexicanorum Protoxea gloriosa Anthophora californica Anthophora curta Apis andreniformis Apis andreniformis Apis andreniformis Apis andreniformis Apis andreniformis Apis andreniformis Apis dorsata Apis dorsata Apis dorsata Apis mellifera Apis mellifera Apis mellifera Apis mellifera Apis mellifera Apis mellifera Apis mellifera Apis mellifera Bombus impatiens Bombus impatiens Bombus impatiens Bombus impatiens Bombus sonorus

Bombus sonorus Corbiculates Bombus sonorus Bombus sonorus Bombus sp. Bombus sp. Bombus sp. Bombus sp. Apidae Ceratina apacheorum Ceratina apacheorum Ceratina apacheorum Ceratina apacheorum Diadasia opuntiae Diadasia opuntiae Diadasia opuntiae Diadasia opuntiae Diadasia opuntiae Diadasia opuntiae Diadasia opuntiae Diadasia rinconis Melissodes paroselae Melissodes paroselae Neolarra californica Nomada gutierrezia? Nomada species 1 Nomada species 1 Triepeolus verbesinae Triepeolus verbesinae Zacosmia maculata Agapostemon angelicus Agapostemon angelicus Conanthalictus new species Dufourea species? Halictus ligatus

Sphecodes mandibularis? Halictidae Anthidium cockerelli Anthidium jocosum Anthidium maculifrons Anthidium undescribed? Ashmeadiella bigeloviae Ashmeadiella rhodognatha Hoplitis biscutellae Hoplitis biscutellae Hoplitis biscutellae Lithurgus apicalis Megachile prosopidis Stelis species 9 Trachusa larreae Megachilidae Prionyx sp. Sphecidae 63

Figure S2 - Diagnostic screen based on PCR amplification with diagnostic primers for seven of the bacteria commonly associated with A. mellifera. Positives are only observed within corbiculate bees.

64

A. Alpha-1-DQ837622 Apis mellifera Alpha-1-EU055544 Collection Location Alpha-1-AY370186 North America Alpha-1-AY370185 Australia 87 Alpha-1-DQ837624 Europe 99 Alpha-1-DQ837623 Africa 86 HM108384 Apis andreniformis HM108428 Asia HM108447 95 Apis dorsata HM108439 Tetraponera-attenuata-DQ113411 100 100 Cardiocondyla-sp.-FJ477552 86 Dolichoderus-sp.-FJ477639 100 Ant associates Pheidole-sp.-FJ477645 98 100 Procryptocerus-batesi-FJ477653 Procryptocerus-batesi-FJ477652 Bartonella-bacilliformis-KC583 Bartonella-grahamii 100 Rochalimaea-elizabethae-L01260 Bartonella-tribocorum-CIP-105476 Bartionella-doshiae-Z31351 65 Bartonella-taylorii-Z31350 Bartonella Bartonella-alsatica-AJ002139 Bartonella-clarridgeiae-X97822 Bartonella-quintana-str.-Toulouse Bartonella-koehlerae-AF076237 Bartonella-henselae-str.-Houston-1 88 Pseudochrobactrum-asaccharolyticum-AM18 Pseudochrobactrum-glaciei-AB369864 Ochrobactrum-oryzae-AM490630 Phyllobacterium-catacumbae-AY636000 100 Phyllobacterium-myrsinacearum-AY512821 Phyllobacterium-bourgognense-AY785320 Mesorhizobium-amorphae-AF041442 100 Agrobacterium-tumefaciens-EF555460 Agrobacterium-tumefaciens-D14500 Mycoplana-dimorpha-EU022307 Daeguia-caeni-EF532794 Brucella-suis-ATCC-23445 Brucella-canis-23365 78 Brucella-melitensis-Melitensis-AY594215 Brucella-abortus-X13695 Ochrobactrum-anthropi-U70978 81 Ochrobactrum-tritici-AJ242584 96 99 Ochrobactrum-grignonense-AJ242581 Ochrobactrum-sp.-LMG-20564-AF452128 91 Rhizobium-etli-U28916 Bradyrhizobium-liaoningense-AB029402 100 100 Methylobacterium-extorquens-AF531770 99 Methylobacterium-sp.-MgMjW-37-AB234521 100 Caulobacter-crescentus-AJ227757 Caulobacter-sp.-FWC20-AJ227766 Wolbachia-Drosophila-AE017196 Pseudomonas-aeruginosa-DQ777865

0.2

65

B. Drosophila 100 Gluconobacter-EU409601 Collection Location Acetobacteraceae-EU0962 3 1 gut North America Alpha-2.1 98 AY370188 Apis mellifera Australia 98 HM108334 HM111901 Europe 100 72 80 HM111875 Africa HM112118 Asia 100 HM112342 Xylocopa californica HM112377 99 Acetobacter-oboediens-AJ001631 100 A.-xylinum-X75619 Gluconacetobacter-hansenii-X75620 Acetobacter-azotocaptans-AF192761 G.-diazotrophicus-X75618 Gluconacetobacter-sacchari-AF127407 81 A.-liquefaciens-X75617 Acidomonas-methanolica-AB110702 Kozakia-baliensis-AB056321 DQ837626 A. mellifera DQ837625 98 81 AJ971849 100 AJ971850 72 73 100 AJ971857 Alpha-2.2 HM108484 Apis dorsata HM109557 100 HM109591 Caupolicana yarrowi 97 Saccharibacter-floricola-NR_024819 80 100 76 HM112219 X. californica 90 HM112141 HM109496 C. yarrowi HM109419 73 98 99 HM109131 Calliopsis subalpinus HM109285 96 Gluconobacter-cerinus-X80775 G-frateurii-X82290 97 99 100 Brown-Thai-AB459531 99 Brown-Thai-AB459530 Flower associated 84 Red-wine-spoilage-EU131163 Gluconobacter-oxydans-X73820 bacterium 100 Gluconobacter-EU096234 82 Acetobacteraceae-EU409602 Drosophila gut 90 Acetobacter-lovaniensis-AB032351 Acetobacter-syzygii-AB052712 Acetobacter-peroxydans-AB032352 86 100 Acetobacter-pomorum-AJ001632 79 A.-pasteurianus-X71863 Acetobacter-estunensis-AB032349 Acetobacter-tropicalis-AB032354 70 80 Acetobacter-cibinongensis-AB052710 84 Acetobacter-orientalis-AB052706 100 Acetobacter-orleanensis-AB032350 Acetobacter-cerevisiae-AJ419843 A-aceti-X74066 98 Asaia-krungthepensis-AB102953 Asaia-siamensis-AB035416 Acidiphilium-acidophilus-D86511 Acidocella-facilis-D30774

0.09

66

C. HM108318 Apis mellifera Collection Location 95 North America HM108312 Australia 100 Africa¬AY370189

Europe DQ837616 Africa 99 Asia DQ837617 HM112094 99 HM111977 81 HM108492 Apis dorsata 91 HM108691 100 Bombus sonorus HM108714

HM108516 Bombus impatiens

Simonsiella-crassa-AF328143 100 100 Sheep Alysiella-filiformis-AB087263

Kingella-denitrificans-L06166 95 Eikenella-corrodens-AF320620 Human Simonsiella-muelleri-AF328145

Kingella-kingae-M22517 96 77 Simonsiella-sp.-AF328149 100 Cat Simonsiella-steedae-AF328154 Dog Kingella-oralis-L06164 Human

Neisseria-weaveri-L10738

Neisseria-flavescens-L06168

100 Neisseria-perflava-AJ239295 100 Neisseria-mucosa-AJ239279

Neisseria-pharyngis-AJ239281 82 Neisseria-macaca-L06169 Neisseria 100 Neisseria-sicca-AJ239292

Neisseria-lactamica-AJ239296

Neisseria-cinerea-AJ239299

97 Neisseria-meningitidis-AJ239310

98 Neisseria-gonorrhoeae-X07714

Neisseria-polysaccharea-L06167

Laribacter-hongkongensis-AF389085 89 Chromobacterium-subtsugae-AY344056

Ralstonia-solanacearum-AF207896 99 Burkholderia-cepacia-X80287

Bordetella-pertussis-AF142326

0.09

67

D. HM108346 Apis mellifera Collection Location North America EF187232 HM108337 Australia 98 85 Europe HM108324 Africa EF187236 Asia AY370184

HM113300

EF187235

B.¬asteroides-AB437355 cultured 97 B.¬asteroides-M58730

B.¬asteroides-FJ861331

HM113259

HM112025

EF187237 96 78 75 B.¬coryneforme-EF187238

B.¬coryneforme-AB437358 cultured 94 84 B.¬coryneforme-M58733

B¬indicum-D86188 cultured

Bifidobacterium-minimum-M58741

Gardnerella-vaginalis-M58744

Bifidobacterium-longum-M58739

Bifidobacterium-gallinarum-D86191 96 Bifidobacterium-pullorum-D86196

Parascardovia-denticolens-D89331 79 Bifidobacterium-magnum-D86193 96 Bifidobacterium-pseudolongum-D86195 98 Bifidobacterium-lactis-X89513 100 Bifidobacterium-thermophilum-U10151 100 Bifidobacterium-boum-D86190

Bifidobacterium-bifidum-S83624

Bifidobacterium-catenulatum-M58732 100 Bifidobacterium-ruminantium-D86197 90 Bifidobacterium-merycicum-D86192

Bifidobacterium-subtile-D89379

N-asteroides-Z36934

Frankia-sp.-L11307

0.2

68

DQ837634 Apis mellifera E. 94 HM112033 Collection Location HM111883 North America DQ837636 Australia Firm-5 HM111880 Europe L.¬apis¬AY667701 cultured Africa 86 L-SD2-FJ861330 Asia 100 EF187242 74 L.¬insectis¬AY667699 cultured AY370183 L-SDR1-FJ861329 99 HM112104

78 L.¬alvei¬AY667698 cultured 94 EF187240 100 90 HM111870 85 HM111887 91 HM112866 Colletes inaequalis HM112858 L-acetotolerans-M58801 L-delbrueckii-X52654 78 L-intestinalis-AJ306299 91 85 L-acidophilus-X61138 79 L-helveticus-X61141 100 L-sobrius-AY700063 L-hamsteri-AJ306298 100 L-jensenii-AF243176 L-johnsonii-AJ002515 L-amylophilus-M58806 spacecraft-GQ129909 HM111923 95 Apis mellifera DQ837632 86 Firm-4 DQ837633 HM108335 99 100 HM108332

100 EF187244 HM111924 100 HM112038 HM112042 L-sakei-M58829 100 L-larvae-AY667700 L-casei-D16551 70 L-plantarum-X52653 L-kunkeei-Y11374 L-buchneri-M58811 Pediococcus-acidilactici-AJ305320 100 100 L-fermentum-X61142 L-reuteri-X76328 Paralactobacillus-selangorensis-AF04974 L-mali-M58824 L-salivarius-AF089108 Leuconostoc-mesenteroides-M23035 Weissella-viridescens-M23040

0.2

69

F. HM108497 Apis dorsata HM108449 Collection Location 88 100 HM112050 Apis mellifera North America HM112130 78 Australia HM108310 Europe HM108315 Africa AY370191 95 HM111973 Asia HM112068 Gamma-1 AY370192 99 DQ837605 DQ837604 77 100 HM108563 Bombus impatiens HM108542 SS-Stomaphis-cupressi-EU348326 Gamma-2 70 74 SS-Stomaphis-longirostris-FJ655515 100 Aphids SS-Cinara-cupressi-EU348322 SS-Stomaphis-quercus-FJ655516 100 HM107876 Apis mellifera DQ837611 EF187250 100 HM108330 HM108316 ground-beetle-EF608532 63 95 wild-boar-FJ612598 Beetle, fly, and boar hillsb-J21-DQ980728 100 Bisgaard-t-14-L06086 99 Bisgaard-t-14-AY362901 Pasteurellaceae Bisgaard-t-32-AY172729 100 100 Pasteurella-testudinis-AY362926 Pasteurella-testudinis-L06090 85 Bisgaard-t-40-AY172732 Gallibacterium-anatis-AF228001 Pasteurella-multocida-AF294410 Mannheimia-haemolytica-AF060699 97 Actinobacillus-delphinicola-AY362889 83 100 Actinobacillus-delphinicola-X89377 96 Actinobacillus-scotiae-Y09653 92 Haemophilus-influenzae-X87978 Phoconebacter-uteri-X89379 Haemophilus-ducreyi-NC_002940 70 Actinobacillus-arthritidis-AF247712 Thorsellia-anophelis-AY837748 100 Phlomobacter-fragariae-AB246669 94 Arsenophonus-nasoniae-M90801 Proteus-vulgaris-AJ233425 Morganella-morganii-AB089246 Yersinia-pestis-AJ232232 100 Enterobacteriaceae 97 Serratia-entomophila-AJ233427 77 Ser-sym-Cinara-cupressi-EU348322 67 Serratia-marcescens-AY395011 Pantoea-stewartii-AJ311838 88 99 Escherichia-coli-X80725 Salmonella-enterica-AF008580 Photobacterium-leiognathi-X74686 Vibrio-cholerae-X74695

0.2

70

Figure S3 - Phylogenetic relationships of each of the bacterial species characteristic of A. mellifera (RAxML with 100 bootstrap replicates). Numbers on branches represent bootstrap support. Trees represent A. Alpha-1, B. Alpha-2, C. Beta, D.

Bifido, E. Firm, and F. Gamma symbionts.

71 enom enom enom enom enom enom enom enom enom Rancho San Bernardino Rancho San Bernardino Rancho San Bernardino Rancho San Bernardino Rancho San Bernardino Huntley Huntley Rancho San Bernardino Mt. Drabb Rancho San Bernardino Mt. Drabb Rancho San Bernardino Santa Catalina Mts. Santa Catalina Mts. Rancho San Bernardino Santa Catalina Mts. Rancho San Bernardino Santa Catalina Mts. Rancho San Bernardino Captivity Rancho San Bernardino Captivity Rancho San Bernardino Rancho San Bernardino Captivity Rancho San Bernardino Captivity Rancho San Bernardino Rancho San Bernardino T Rancho San Bernardino T Rancho San Bernardino Rancho San Bernardino T Rancho San Bernardino T Rancho San Bernardino Rancho San Bernardino T T Rancho San Bernardino Rancho San Bernardino Rancho San Bernardino T T Rancho San Bernardino Rancho San Bernardino Rancho San Bernardino T Rancho San Bernardino Rancho San Bernardino Rancho San Bernardino Rancho San Bernardino Rancho San Bernardino Santa Catalina Mts. Rancho San Bernardino Rancho San Bernardino Huntley Rancho Puerta Blanca Huntley Rancho San Bernardino Rancho Puerta Blanca Rancho San Bernardino Rancho San Bernardino Rancho Puerta Blanca Rancho San Bernardino Region ov r Sonora Sonora Sonora Sonora Sonora Montana Montana Sonora California Sonora California Sonora Arizona Arizona Sonora Arizona Sonora Arizona Sonora California Sonora California Sonora Sonora California Sonora California Sonora Sonora Sabah Sonora Sabah Sonora Sonora Sabah Sonora Sabah Sonora Sonora Sabah Sabah Sonora Sonora Sonora Sabah Sabah Sonora Sonora Sonora Sabah Sonora Sonora California Sonora Sonora Sonora Arizona Sonora Sonora Montana Sonora Montana Sonora Sonora Sonora State/P Sonora Sonora Sonora Mexico Mexico Mexico Mexico Mexico United States United States Mexico United States Mexico United States Mexico United States United States Mexico United States Mexico United States Mexico United States Mexico United States Mexico Mexico United States Mexico United States Mexico Mexico Malaysia Mexico Malaysia Mexico Mexico Malaysia Mexico Malaysia Mexico Mexico Malaysia Malaysia Mexico Mexico Mexico Malaysia Malaysia Mexico Mexico Mexico United States Malaysia Mexico Mexico United States United States Mexico United States Mexico Mexico United States Mexico Mexico United States Mexico United States Mexico Mexico Mexico Country Mexico Mexico Mexico Perditella Perditella Perdita Perditella Perdita Perdita Perdita Perditella Zadontomerus Perdita Perditella Perdita Perdita Perdita Melissodes Perdita Melissodes Hypomacrotera Nomadopsis Hypomacrotera Nomadopsis Subgenus Heliophila ribe Protandrenini Perditini Perditini Perditini Bombini Perditini Bombini Bombini Perditini Bombini Perditini Bombini Perditini Bombini Bombini Perditini Bombini Ceratini Bombini Perditini Nomadini Bombini Nomadini Bombini Perditini Nomadini Bombini Neolarrini Apini Perditini Epeolini Apini Epeolini Apini Perditini Melectini Apini Apini Perditini Eucerini Apini Perditini Apini Perditini Perditini Eucerini Apini Perditini Emphorini Apini Apini Calliopsini Emphorini Apini Apini Calliopsini Emphorini Apini Emphorini Apini Calliopsini Emphorini Apini Calliopsini Emphorini Apini Apini Emphorini Anthophorini T Emphorini Anthophorini ginae ginae ginae ginae ginae ginae ginae ginae ginae ginae ginae ginae ginae ginae ginae ginae ginae ginae ginae ginae ginae ginae r r r r r r r r r r r r r r r r r r r r r r Panu Panu Panu Panu Apinae Panu Apinae Apinae Panu Apinae Panu Apinae Panu Apinae Apinae Panu Apinae Xylocopinae Apinae Panu Nomadinae Apinae Nomadinae Apinae Panu Nomadinae Apinae Nomadinae Apinae Panu Nomadinae Apinae Nomadinae Apinae Panu Apinae Apinae Apinae Panu Apinae Apinae Panu Apinae Panu Panu Apinae Apinae Panu Apinae Apinae Apinae Panu Apinae Apinae Apinae Panu Apinae Apinae Apinae Apinae Panu Apinae Apinae Panu Apinae Apinae Apinae Oxaeinae Apinae Apinae Andreninae Subfamily Apinae Apinae Andrenidae Andrenidae Andrenidae Andrenidae Apidae Andrenidae Apidae Apidae Andrenidae Apidae Andrenidae Apidae Andrenidae Apidae Apidae Andrenidae Apidae Apidae Apidae Andrenidae Apidae Apidae Apidae Apidae Andrenidae Apidae Apidae Apidae Apidae Andrenidae Apidae Apidae Apidae Apidae Andrenidae Apidae Apidae Apidae Andrenidae Apidae Apidae Andrenidae Apidae Andrenidae Andrenidae Apidae Apidae Andrenidae Apidae Apidae Apidae Andrenidae Apidae Apidae Apidae Andrenidae Apidae Apidae Apidae Apidae Andrenidae Apidae Apidae Andrenidae Apidae Apidae Apidae Andrenidae Apidae Apidae Andrenidae Family Apidae Apidae Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Superfamily Apiformes Apiformes -08 -08 -08 r r r 2/18/07 2/18/07 2/18/07 2/27/07 2/27/07 2/27/07 2/27/07 2/27/07 2/27/07 5-Jun-08 5-Jun-08 3-Jun-08 3-Jun-08 3-Jun-08 1-Jun-09 1-Jun-09 1-Aug-08 1-Aug-08 1-Aug-08 1-Aug-08 1-Aug-08 1-Aug-08 7-May-08 7-May-08 9-May-12 1 1 12-Jun-09 12-Jun-09 21-Jun-08 21-Jun-08 21-Jun-08 21-Jun-08 12-Jun-09 12-Jun-09 21-Ma 21-Ma 21-Ma 25-May-09 13-May-08 13-May-08 17-May-09 16-May-12 16-May-12 16-May-12 17-May-09 21-May-12 21-May-12 17-May-09 21-May-12 17-May-09 17-May-09 16-May-12 16-May-12 16-May-12 16-May-12 16-May-12 17-May-09 16-May-12 17-May-09 16-May-12 16-May-12 16-May-12 25-May-09 25-May-12 16-May-12 25-May-12 17-May-12 25-May-12 21-May-12 Date collected Adult male Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult male Adult female Adult female Adult female Adult male Adult female Adult female Adult female Adult female Adult female Adult female Adult male Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult male Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult male Adult female Adult female Adult female Adult male Adult female Adult male Adult female Adult male Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult male Adult male Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult female Stage/Sex Adult female Adult female Hive Location vierecki vierecki subalpinus subalpinus Subspecies curta mexicanorum larreae larreae semicaerulea sp. larreae sp. sp. semicaerulea sp. exclamans sonorus semicaerulea sonorus sonorus larreae sonorus apacheorum impatiens semicaerulea species 1 impatiens gutierrezia? impatiens larreae species 1 impatiens californica dorsata munita verbesinae dorsata verbesinae dorsata munita maculata andreniformis andreniformis ashmeadi semicaerulea paroselae andreniformis andreniformis callicerata ashmeadi exclamans paroselae andreniformis subalpinus rinconis andreniformis mellifera opuntiae mellifera mellifera puellae opuntiae mellifera subalpinus opuntiae mellifera opuntiae mellifera puellae opuntiae mellifera mellifera gloriosa opuntiae californica species 1 Species name opuntiae riepeolus riepeolus Anthophora Protoandrena Perdita Perdita Perdita Bombus Perdita Bombus Bombus Perdita Bombus Perdita Bombus Perdita Bombus Bombus Perdita Bombus Ceratina Bombus Perdita Nomada Bombus Nomada Bombus Perdita Nomada Bombus Neolarra Apis Perdita T Apis T Apis Perdita Zacosmia Apis Apis Perdita Perdita Melissodes Apis Apis Perdita Perdita Perdita Melissodes Apis Calliopsis Diadasia Apis Apis Diadasia Apis Apis Calliopsis Diadasia Apis Calliopsis Diadasia Apis Diadasia Apis Calliopsis Diadasia Apis Apis Protoxea Diadasia Anthophora Andrena Genus Diadasia 9 1 3 2 1 6 5 4 3 2 1 5 1 75 72 1 70 67 63 60 57 55 52 80 54 44 79 36 78 49 32 77 34 45 30 29 43 31 41 42 24 27 37 40 17 39 59 21 38 20 35 19 18 25 13 16 12 14 46 41 MT MT BB3 BB2 BB1 A C Curtis 5 BBCA2 BBCA1 DVE 12 BBMT2 BBMT1 Mt. Lem Mt. Lem BRL-GK Identifier 1 samples used in diagnostic PCR. 1 A Acu Pme Pla6 Pla4 Pse4 Dop1 Bom12 Pla3 Bom Pse3 Bom10 Pex1 Bom9 Bom8 Pse2 Bom7 DN Bom6 Pla2 Bom5 Cap1 Bom4 Pse1 Nsp2 Bom3 Ngu Bom2 Pla1 Nsp1 Bom1 Nca Ador3 Pmu2 Tve2 Ador2 Tve1 Ador1 Pmu1 Zma Aand6 Aand5 Pas2 Pse5 Mpa2 Aand4 Aand3 Pca Pas1 Pex2 Mpa1 Aand2 Csu2 Dri1 Aand1 Amel7 Dop7 Amel6 Cpu2 Dop6 Amel5 Amel4 Csu1 Dop5 Amel3 Dop4 Amel2 Cpu1 Dop3 Amel1 Ame1 Pgl Dop2 Aca Ans2 tube id 72 (Bryan Danfofrth coll.) ildlife Refuge, 26 km E Douglas ildlife Refuge, 26 km E Douglas ildlife Refuge, 26 km E Douglas ildlife Refuge, 26 km E Douglas ildlife Refuge, 26 km E Douglas W W W W W

Ascher coll.)

Minckley) Melaleuca L Ascher coll.)

Arizona campus

ompkins Co. Ithaca vicinity (Bryan Danforth, coll) ompkins Co. Ithaca (John Region Cochise Co. Rancho San Bernardino Rancho San Bernardino Rancho San Bernardino Rancho San Bernardino Rancho Puerta Blanca Rancho San Bernardino Rancho San Bernardino Rancho San Bernardino Rancho San Bernardino Rancho San Bernardino Rancho Puerta Blanca Rancho San Bernardino Carl Hayden Bee Research Lab Carl Hayden Bee Research Lab Rancho San Bernardino University of Rancho San Bernardino Rancho San Bernardino Cochise Co., San Bernardino National Rancho San Bernardino Cochise Co., San Bernardino National Rancho San Bernardino Cochise Co., San Bernardino National Rancho San Bernardino Cochise Co., San Bernardino National Rancho San Bernardino T FRANCE: Uchaux (near Orange) (Bryan Danforth, coll.) Monroe Co., near Rochester (R Rancho San Bernardino Cochise Co., San Bernardino National Rancho San Bernardino 7 km S. Graskop (Bryan Danforth, coll.) Hidalgo Co., Rodeo vicinity (Bryan Danforth, coll.) Cortland Co., McLean Bog (John T 20 km NNE Eurardy Station on Australia

ork ork ork ork Y Y Y Y

estern aucluse State Arizona Sonora Sonora Sonora Sonora Sonora Sonora Sonora Sonora Sonora Sonora Sonora Sonora Arizona Arizona Sonora Arizona Sonora Sonora Arizona Sonora Arizona Sonora Arizona Sonora New Arizona Sonora New V Sonora Arizona Sonora New New Mexico Mpumalanga New W Africa

Country Australia United States Mexico Mexico Mexico Mexico Mexico Mexico Mexico Mexico Mexico Mexico Mexico Mexico United States United States Mexico United States Mexico Mexico United States Mexico United States Mexico United States Mexico United States United States Mexico United States France Mexico South United States Mexico United States United States United States gopsis r Subgenus Ashmeadiella Dasyosmia Dasyosmia Dasyosmia Ashmeadiella Megachile Heteranthidium Lithu Odontalictus Agapostemon Agapostemon Zadontomerus Zadontomerus Zadontomerus gini r ribe T Masarini Sphecini Osmiini Osmiini Osmiini Osmiini Osmiini Megachilini Anthidiini Anthidiini Anthidiini Anthidiini Anthidiini Anthidiini Apini Lithu Apini Xylcopini Emphorini Halictini Halictini Osmiini Halictini Megachilini Halictini Colletini Caupolicanini Caenohalictini Ceratini Halictini Ceratini Melittini Calliopsini Philanthini Hesperapini Sceliphrini Ceratini ginae gini r r Subfamily Masarinae Sphecinae Megachilinae Megachilinae Megachilinae Megachilinae Megachilinae Megachilinae Megachilinae Megachilinae Megachilinae Megachilinae Megachilinae Megachilinae Apinae Apinae Lithu Rhophitinae Xylocopinae Rhophitinae Apinae Halictinae Halictinae Megachilinae Halictinae Megachilinae Halictinae Diphaglossinae Colletinae Halictinae Xylocopinae Halictinae Xylocopinae Panu Melittinae Sphecinae Philanthinae Dasypodaini Xylocopinae Family Masaridae Sphecidae Megachilidae Megachilidae Megachilidae Megachilidae Megachilidae Megachilidae Megachilidae Megachilidae Megachilidae Megachilidae Megachilidae Megachilidae Apidae Apidae Megachilidae Halictidae Apidae Halictidae Apidae Halictidae Halictidae Megachilidae Halictidae Megachilidae Halictidae Colletidae Colletidae Halictidae Apidae Halictidae Apidae Andrenidae Melittidae Crabronidae Melittidae Sphecidae Apidae espoidea V Superfamily Spheciformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Apiformes Spheciformes Spheciformes Apiformes -08 -08 -08 -06 -02 -08 -02 r r r r r r r 4-Oct-05 22-Jul-00 22-Jul-00 16-Jul-00 1-Aug-08 1-Aug-08 1-Aug-08 1-Nov-07 1-Aug-08 6-Aug-06 6-Aug-06 6-Aug-06 6-Aug-06 1-Aug-08 6-Aug-06 1-Aug-08 9-May-12 9-May-12 9-May-12 18-Ap 16-Ap 18-Dec-07 18-Dec-07 21-Ma 21-Ma 21-Ma 21-Ma 23-Ma 31-Aug-04 13-May-08 21-May-12 17-May-09 17-May-12 25-May-12 16-May-12 21-May-12 17-May-09 25-May-09 10-May-08 Date collected Adult female Sex/Stage Adult female Adult female Adult female Adult male Adult male Adult male Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult male Adult male Adult female Adult female Adult female Adult female Adult female Adult male Adult female Adult male Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult female Adult male ithin the hive ithin the hive Hive Location W W 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 80 pooled # of ind. libraries A vespiformis Species sp. bigeloviae biscutellae biscutellae biscutellae rhodognatha prosopidis species 9 larreae undescribed species? jocosum maculifrons cockerelli mellifera mellifera apicalis species? californica new species opuntiae ligatus mandibularis? biscutellae angelicus odontostoma angelicus yarrowi inaequalis virescens apacheorum patellatus apacheorum subalpinus saetigera cockerelli gibbosus californicum apacheorum gus r eate the 16S rRN rachusa r Paragia Genus Prionyx Ashmeadiella Hoplitis Hoplitis Hoplitis Ashmeadiella Megachile Stelis T Anthidium Anthidium Anthidium Anthidium Apis Apis Lithu Dufourea Xylocopa Conanthalictus Diadasia Halictus Sphecodes Hoplitis Agapostemon Megachile Agapostemon Caupolicana Colletes Agapostemon Ceratina Halictus Ceratina Calliopsis Rediviva Hesperapis Philanthus Chalybion Ceratina 6 8 7 4 2 3 71 64 51 48 26 47 22 15 10 66 58 33 28 23 76 69 61 JSA JSA BND (02-7) (05-28) 125702 124642 125740 124630 125716 125750 (02-41) 125749 Identifier samples used to c A Pri Abi Hbi3 Hbi2 Hbi1 Arh Mpr Ste1 Tla Aun Ajo Ama Aco DN Lap Duf Cns Hli Sma Aan2 Aan1 Cap6 Cap5 Cap3 73

APPENDIX B: ESTABLISHMENT OF CHARACTERISTIC GUT BACTERIA

DURING DEVELOPMENT OF THE HONEYBEE WORKER

Published: Applied and Environmental Microbiology (2012) 78:2830-2840. 74

AMERICAN SOCIETY FOR MICROBIOLOGY LICENSE TERMS AND CONDITIONS Mar 28, 2012

This is a License Agreement between Vince Martinson ("You") and American Society for Microbiology ("American Society for Microbiology") provided by Copyright Clearance Center ("CCC"). The license consists of your order details, the terms and conditions provided by American Society for Microbiology, and the payment terms and conditions.

All payments must be made in full to CCC. For payment instructions, please see information listed at the bottom of this form.

License Number 2877880084563

License date Mar 28, 2012

Licensed content publisher American Society for Microbiology

Licensed content publication Applied and Environmental Microbiology

Licensed content title Establishment of Characteristic Gut Bacteria during Development of the Honeybee Worker

Licensed content author Vincent G. Martinson, Jamie Moy, Nancy A. Moran

Licensed content date Apr 1, 2012

Volume 78

Issue

Start page 2830

End page 2840

Type of Use Dissertation/Thesis

Format Electronic

Portion Full article

Order reference number

Title of your thesis / ECOLOGICAL AND EVOLUTIONARY RELATIONSHIPS BETWEEN BEES dissertation AND THEIR BACTERIAL GUT MICROBIOTA

Expected completion date Apr 2012

Estimated size(pages) 180

75

ABSTRACT

Previous surveys have shown that adult honey bee (Apis mellifera) workers harbor a characteristic gut microbiota that may play a significant role in bee health. For three major phylotypes within this microbiota, we have characterized distributions and abundances across the life cycle and among gut organs. These distinctive phylotypes, called Beta, Firm-5, and Gamma-1 (BFG), were assayed using quantitative PCR, FISH microscopy, and experimental manipulation of inoculation routes within developing bees.

Adult workers (9-30 days post-eclosion) contained a large BFG microbiota with a characteristic distribution among gut organs. The crop and midgut were nearly devoid of these phylotypes, while the ileum and rectum together contained over 95% of the total

BFG microbiota. The ileum contained a stratified community in which the Beta and

Gamma-1 phylotypes dominated, filling the longitudinal folds of this organ. Deep sequencing of 16S rRNA genes showed clear differences among communities in midgut, ileum, and rectum. In contrast with older workers, larvae and newly emerged workers contain few or no bacteria, and their major food source, bee bread, lacks most characteristic phylotypes. In experiments aimed at determining route of inoculation,

NEWs sometimes acquired the typical phylotypes through contact with older workers, contact with the hive, and emergence from the brood cell; however, transmission was patchy in these assays. Our results outline a colonization pattern for the characteristic phylotypes through A. mellifera ontogeny. We propose the names 76

Candidatus Snodgrassella alvi and Candidatus Gilliamella apicola respectively for the

Beta and Gamma-1 phylotypes.

INTRODUCTION

Animal health is greatly influenced by the microbial community within the digestive tract (Dillon & Dillon 2004; Lee & Mazmanian 2010). Whereas gut pathogens can negatively influence health, commensal microorganisms can prime immune response, confer resistance to invading pathogens, and augment nutrition (Round & Mazmanian

2009). Both vertebrates and invertebrates have consistent non-pathogenic associations with a microbial gut community (Ley et al. 2008; Ochman et al. 2010), including defined mutualistic relationships (e.g., termites (Hongoh et al. 2005), broad-headed bugs

(Kikuchi et al. 2007), and plataspid stinkbugs (Hosokawa et al. 2006)). Gut communities change with host environment, diet, and age and among gut compartments (Brune &

Friedrich 2000; Hongoh et al. 2006; Robinson et al. 2010). Most studies of gut microbiota reflect a single sample, and thus yield no information on temporal and spatial dynamics.

The social insect Apis mellifera (honey bee) is the most important pollinator globally, and the health of A. mellifera colonies has been a major concern following colony losses in the last decade (Cox-Foster et al. 2007; Klein et al. 2007). In A. mellifera, adult workers harbor a characteristic gut microbiota consisting of nine distinct bacterial phylotypes, which account for >95% of their total bacterial microbiota and 77 which have been observed repeatedly using several non-culture based methods on samples representing different environments, continents, and host genotypes

(Babendreier et al. 2007; Cox-Foster et al. 2007; Jeyaprakash et al. 2003; Martinson et al. 2011; Mohr & Tebbe 2006; Olofsson & Vasquez 2008). The repeated observation of the same distinctive bacterial phylotypes in non-culture based studies implies that most members of this gut microbiota are maintained by transmission between individuals within a hive and not by selective acquisition from the extra-hive environment

(Jeyaprakash et al. 2003; Martinson et al. 2011). Additionally, these observations suggest a symbiotic relationship potentially critical to bee health (Martinson et al. 2011; Olofsson

& Vasquez 2008). To date, no studies have addressed which organs of the gut these bacteria colonize, or how the gut microbiota changes during A. mellifera ontogeny.

The A. mellifera gut bacteria encounter a physically and nutritionally variable environment, due to the complex development and social behavior of this insect.

Furthermore, the adult gut is divided into four major organs (crop, midgut, ileum, rectum), providing different functions in the catabolism and absorption of food and different environments for bacterial symbionts (Chapman 1998; Snodgrass 1910). Adult workers perform a succession of tasks as they age, which potentially exposes them to different microorganisms: young bees nurse larvae within the hive, whereas older bees forage pollen and nectar from flowers outside the hive (Ament et al. 2008; Gilliam 1997;

Seeley 1985). In contrast to adults, larval A. mellifera have a discontinuous gut in which the foregut (crop and midgut) is not connected to the hindgut (ileum and rectum) until just before pupation, when they excrete dietary waste for the first time (Snodgrass 1910). 78

Larvae reside within a single brood cell where nurse workers feed them a highly nutritional glandular secretion with small amounts of pollen and honey (Winston 1985).

In this study, we use culture-independent methods to enumerate and visualize the microbiota of different gut organs and of bees of different ages. We focus on three abundant phylotypes within the A. mellifera gut microbiota.

MATERIALS AND METHODS

Quantitative PCR (qPCR) to estimate bacterial abundance

Workers were collected from two colonies at the USDA Carl Hayden Bee

Research Center (USDA-CHBRC) in Tucson, AZ in November 2009. To obtain known- age adults, a frame that contained capped brood was cleared of adult bees, caged, and held at 34°C for 24 hours in a dark, humidified incubator (mimicking hive conditions).

For each colony, five newly emerged workers (NEWs) were collected (day-1 bees), and

50 NEWs were marked with Testors enamel paint (Testor Corp., Rockford, IL, USA), returned to their colony, and allowed to mature naturally. Subsequently, five marked bees were collected at days 9 and 19 and three at day 30. Samples were stored in 75% ethanol at -20°C. For DNA extraction, gut organs (i.e., crop, midgut, ileum, rectum) were dissected and separated with sterile forceps and dissection scissors. DNA was extracted and quantified as in Martinson et. al. (2011) and diluted to ~75 ng/µl to standardize PCR and qPCR reactions. To verify that DNA was of sufficient quality for PCR, a control 79

PCR reaction was performed for a ~600bp fragment of the A. mellifera elongation factor

1-alpha gene (ef1α) with primers efs599 and efa923 (Vane-Wright et al. 1999).

Primer pairs that amplified 100-250bp products from the 16S rRNA gene were designed for each of the three most consistently observed and most abundant bacterial phylotypes found in the A. mellifera microbiota (Table 1). These specific phylotypes (or species) are referred to as “Beta”, “Firm-5”, and “Gamma-1” as in previous studies (Cox-

Foster et al. 2007; Martinson et al. 2011), and we use “BFG phylotypes” for the set of these three phylotypes. Standard curves were created and reactions were run as in Oliver et. al. (Oliver et al. 2006) on a LightCycler (Roche Applied Science, Indianapolis, IN,

USA).

The qPCR results were expressed as total number of bacterial 16S rRNA gene copies per sample, by multiplying by the total DNA amount in each sample, and values were normalized with log transformation. To preserve a normal distribution, samples that were below the detection level of the qPCR curve were given the value of 1000. Least squares-means analysis was performed using JMP version 8 (SAS Institute Inc., Cary,

NC, USA) to identify effects that correlated with estimates of the BFG phylotypes.

Variables (i.e., A. mellifera age, gut organ, bacterial phylotype) and all interaction terms were added into the model at the start. After each run, the factor with the highest p value was removed until only factors with a p value <0.05 remained. Significant differences within and between samples were determined by a Tukey’s HSD test. Data was back transformed for presentation. The Morisita index (a quantitative measure of beta 80 diversity) was calculated for the community profiles of each gut organ and compared statistically using ANOSIM in PAST v2.06 (Hammer & Harper 2001).

Fluorescence in situ hybridization (FISH) microscopy

Two individuals were collected from each of four age groups within a hive at the

USDA-CHBRC (Tucson, AZ, August 2010): uncapped fifth instar larvae, NEWs, nurse bees, and foragers. Adult digestive tracts were dissected and placed onto a strip of filter paper to facilitate positioning of the specimen. Tissue was fixed in 4% paraformaldehyde for 2-3 hours (adult guts) or overnight (larvae) at room temperature, rinsed in 1x PBS buffer for 20 seconds, and held in 75% ethanol at 4°C until embedding. Samples were embedded into paraffin with vacuum infiltration using a Tissue-Tek VIP Tissue

Processor (Sakura Finetek USA Inc., Torrance, CA, USA), and 5µm sections were cut with disposable blades and mounted onto Epic Plus Slides (Epic Scientific, Tualitin, OR,

USA).

A specific and exclusively binding FISH probe was designed for each 16S rRNA gene of the Beta, Firm-5, and Gamma-1 phylotypes (Table 1). Sections were cleared of paraffin prior to probe hybridization by melting the wax in a 60°C incubation for 30 min, followed by two 5 min washes in xylenes, a 7 min wash in 95% ethanol, a 7 min wash in

75% ethanol, and a 5 min rinse in ddH2O. Hybridization of probes was performed simultaneously following the protocol presented in Daims et. al. (2005). Spectral imaging was used to view sections on a Zeiss 510 meta confocal microscope. Autofluorescence was assayed for each tissue type by imaging sections as above, without FISH probes. 81

Observations were made on each gut organ for adult specimens and at two locations along the larval midgut.

Diagnostic PCR screen for bacteria in larvae

To determine the presence and identity of bacterial microbiota in larvae, DNA samples were screened with universal and specific primer sets. Specimens (3rd instar, 5th instar, and NEW) were collected from Beltsville, MD at the USDA Bee Research

Laboratory (USDA-BRL) from two healthy hives and one hive infected with European foulbrood (Melissococcus plutonius). Collections were made at the USDA-CHBRC in

Arizona (5th instar and nurse adult) and at West Campus (Yale University, West Haven,

CT) from healthy hives. DNA was subsequently extracted from surface-sterilized larvae and screened with eukaryotic ef1α primers to verify DNA quality using previously mentioned methods. Universal bacterial 16S rRNA gene primers 27f-short and 1507r

(Martinson et al. 2011) were used to screen for bacterial presence/absence. Specimens that produced an amplicon with the universal bacterial primers were further screened with seven primer pairs that each amplify specific phylotypes (Alpha-1, Alpha-2.1, Alpha-2.2,

Beta, Firm-4, Firm-5, and Gamma-1) found in the A. mellifera microbiota (primer sequences and reaction conditions are listed in Martinson et al. (2011). All PCR screens were run with a positive (adult A. mellifera gut DNA) and a negative (ddH2O) control on a 1% agarose gel (100 V, 50 min).

82

Assessing the source of the characteristic microbiota in NEWs

To determine the source of the microbiota in NEWs two cage formats were used: frame cages, cages around hive frames containing brood and bee bread, and cup cages, caged worker bees with hand-collected bee bread. For cup cage analyses, NEWs that had not been exposed to the hive environment or older workers were collected by transferring late-stage pupae from sealed brood cells to a cage, incubated at 34°C with high humidity

(to mimic hive conditions), and pupae were allowed to eclose naturally. Resulting NEWs were marked with paint. Six NEWs were placed into each of three cup cages

(construction described in Evans et al. (Evans et al. 2009)). All cup cages were provided sterile 1:1 sucrose:H2O solution (0.25µm filter) and raw bee bread ad libitum. In both cage types, bees were allowed to survive until day 9 and were then collected into 95% ethanol. Individuals that died before day 9 were collected into 95% ethanol during daily inspections. DNA was extracted and screened with diagnostic primers following described protocols (Martinson et al. 2011; U'Ren et al. 2010).

Bee bread was experimentally assessed as a source of the characteristic microbiota in a cup cage of lab-reared NEWs with no nurse workers and a diet of raw bee bread (Table S1, cups 1.1 & 1.2). Additionally, bee bread was directly PCR-screened for the characteristic phylotypes, using DNA from two samples of bee bread combined from seven comb cells. Older workers were assessed as a source of the characteristic microbiota in a cup cage of lab-reared NEWs with three nurse workers (in-hive workers collected from the same hive as NEWs) and a diet of raw bee bread (Table S1, cup 2). 83

We assessed whether nurse gut homogenate, added to bee bread, could introduce the characteristic microbiota in a cup cage with lab-reared NEWs (Table S1, cup 3).

To assess the ability of hive materials (i.e., comb, brood cell cap) to introduce the microbiota, pupae were allowed to emerge in the lab in frame cages lacking adult workers. Resulting NEWs were marked with paint and used in the frame cage tests

(Table S1). To assess the transfer of the characteristic microbiota from hive materials to

NEWs, 20 marked NEWs were placed back into a frame cage (Table S1, frame 1). To assess transfer from hive materials in combination with nurse workers, twenty NEWs and twenty older nurse workers (collected from within the same hive and marked a different color) were placed into a frame cage (Table S1, frame 2).

454 pyrotag analysis of A. mellifera gut bacterial community

DNA from distinct gut organs of three day-9 workers and one day-30 worker, dissected and extracted for the above qPCR analysis, were selected for 454 pyrotag sequencing. A ~450-bp portion of the bacterial 16S rRNA gene was amplified from each sample using the universal primers 926f and 1492r. Each sample was given a barcode sequence (Table S2) and reactions were run as in Ochman et. al. (2010) with 454 FLX

Titanium sequencing (Roche Applied Science, Indianapolis, IN, USA). Finished sequence was analyzed with the default QIIME parameters (except alterations listed)

(Caporaso et al. 2010). Raw 454 output was split into samples, barcode and primer sequence was removed, and resulting sequences were filtered for quality and length

(minimum quality score of 25, retained sequences between 460-600bp). Sequences were 84 denoised, 97% sequence similarity OTUs were picked and aligned, non-bacterial OTUs unable to properly align were removed, and chimeric OTUs were identified. Manual inspection of the alignment identified more chimeras that were removed with the QIIME- identified chimeras. The resulting dataset was processed in QIIME for OTU abundance and jackknifed beta diversity (with even taxon sampling). Alpha diversity was obtained in mothur (Schloss et al. 2009). OTUs were classified as one of the characteristic phylotypes or ‘other’ with a blastn search to GenBank. The ‘heatmap.2’ program within the gplots package for R was used to display OTUs with a frequency >0.05% of one sample or a top blastn hit to a characteristic A. mellifera phylotype in the GenBank nt database.

Phylogenetic analysis for proposed Candidatus names

Representative sequences of the Beta and Gamma-1 phylotypes and closely related genera were aligned with Infernal in the RDP (Cole et al. 2009). A maximum likelihood phylogeny was constructed for each with RAxML using the GTRGAMMA parameter and 100 bootstrap replicates (Stamatakis 2006).

RESULTS

qPCR results

Colony of origin had no significant effect on numbers of bacteria (Table 2), so bees from different colonies were pooled for subsequent analyses. Worker age had a large 85 effect on BFG abundance, mainly because NEWs collected on day 1 were nearly devoid of the BFG phylotypes, with fewer than 104 copies (Figure 1a). Day-1 individuals had at least three orders of magnitude fewer 16S rRNA gene copies than older workers, which contained more than 106 copies (Figure 1a; Table 2, p = 2.4 x 10-106, Tukey’s HSD).

Because day-1 individuals possess very few gut bacteria, and therefore are not representative of mature honey bee workers, subsequent analyses were performed both with and without day-1 individuals. After day 1, total BFG 16S rRNA gene copies continued to gradually increase with age, more than doubling between day 9 and day 30.

Averaged across all ages, there were more 16S rRNA gene copies from the Firm-

5 phylotype than from the Beta or Gamma-1 phylotypes (Figure 1b; Table 2, p = 0.0011,

Tukey’s HSD). The Beta and Gamma-1 phylotypes represented 23% and 25% of the total BFG 16S rRNA gene copies, respectively, while the Firm-5 represented 52%. When individuals from day 1 were excluded, the average total copies of the 16S rRNA gene per bee increased 5-6 fold relative to the analysis including day-1 individuals but relative proportions of the phylotypes were similar (Table 2, p = 0.2391).

Numbers of total 16S rRNA gene copies varied greatly among the organs (Figure

1c; Table 2, p = 6.6 x 10-84, p = 9.8 x 10-80, Tukey’s HSD). On average, of total BFG 16S rRNA gene copies within the entire digestive tract, the crop harbored 0.007 - 0.062%, the midgut harbored 1 - 4%, the ileum harbored 4 - 10%, and the rectum harbored 87 - 94%.

Numbers of each phylotype were characteristic for a particular gut organ, regardless of host age or bacterial type (Figure 2a & c). 86

Only day-1 samples failed to show a significant age x organ interaction (Figure

2a; Table 2, p = 1.6 x 10-33, Tukey’s HSD). All day-1 samples had very low abundance

(Figure 2a). Gut organs differed in relative abundances of the three phylotypes, and the organ x bacterial phylotype interaction was significant whether or not individuals from day 1 were excluded (Table 2, p = 3.7 x 10-5, p = 2.4 x 10-5, Tukey’s HSD). The total

BFG amount varied by several orders of magnitude between organs (Figure 2). In addition, the crop/rectum, midgut, and ileum had distinct community profiles (Figure 2b;

ANOSIM, p = <0.0001, R = 0.3575). The crop and rectum BFG communities were not significantly different (ANOSIM score 0.3504) and consisted mostly of the Firm-5 phylotype (69%, 81%), the midgut was dominated by the Gamma-1 phylotype (47%), and the ileum was dominated by the Beta phylotype (42%) (Figure 2b & c).

FISH microscopy

Autofluorescence was observed in the gut tissues, but did not obscure the FISH probe imaging of bacteria within the guts of A. mellifera workers. Both fifth instar larvae and NEW gut samples lacked signal from any of the FISH probes (and therefore are not included in Figure 3), while nurse and forager guts produced a signal for each probe

(Figure 3). These results indicate that the guts of larvae and very young workers contain very few or no bacteria. In FISH surveys, NEWs, nurses, and foragers are expected to correspond approximately to day 1, day 9 and day 30 individuals, based on known behaviors of workers bees of varying ages (e.g., Ament et al. 2008). 87

In contrast to larvae and young NEWs, FISH images revealed that nurses & foragers had substantial numbers of bacteria within their guts. A similar pattern of bacterial colonization was observed in both nurses and foragers (Figure 3). The nurse/forager crop was nearly devoid of bacteria. The nurse/forager midgut had a small number of bacteria including the Beta phylotype distributed along its entire length, but the bacterial load increased toward the posterior end of the midgut where the Gamma-1 phylotype became dominant.

The nurse/forager ileum has a bacterial profile strikingly different from that of the midgut. The large invaginations along the ileum’s length are filled with bacterial cells

(Figure 3c & d). This mass of bacteria is mainly composed of the BFG phylotypes.

Additionally, some cells only stained with the universal bacterial probe (eub339). The

Gamma-1 and Beta phylotypes are the most numerous cells in the ileum. The Beta phylotype is often directly associated with the ileum’s intima, while the Gamma-1 phylotype is found throughout the ileum’s invaginations (Figure 3f & g). The Firm-5 phylotype is not dominant, but clusters of Firm-5 can be seen.

The nurse/forager rectum can contain pollen cells that autofluoresce to some degree under most wavelengths used to excite the FISH probes, making it difficult to observe the small bacterial cells. Nonetheless, the large bacterial population within the rectum greatly outnumbers the pollen and can be easily identified in images (Figure 3e).

Firm-5 is the most abundant phylotype in the rectum, while small amounts of Beta and

Gamma-1 phylotypes are visible. Additionally, the rectum has many bacteria that were labeled only with the universal probe (eub339) and not with a specific probe, suggesting 88 that much of the rectal community does not correspond to the Beta, Firm-5, or Gamma-1 phylotypes. We note that the Firm-5 probe is specific and does not hybridize with the

Firm-4, another phylotype of Lactobacillus that may be abundant in the rectum.

Diagnostic PCR screen for bacteria in larvae

PCR screens for bacteria were performed on DNA samples from whole larvae, from healthy colonies from three locations and from a colony exhibiting symptoms of infection with European foulbrood. All samples had positive amplification of the A. mellifera ef1α gene, indicating successful extraction of PCR-quality DNA. Only six of the 35 healthy individuals (3rd instar larva, 5th instar larva, or NEW) yielded a product with the universal bacterial 16S rRNA gene primers (Table 3).

In contrast, most (10/14) of the bees from a colony previously determined to be infected with European foulbrood yielded a band with the universal 16S rRNA gene primers. Of these, 3rd instar larvae mostly lacked bacteria (1/5 positive reactions), 5th instar larvae all contained bacteria (5/5), and NEW individuals mostly contained bacteria

(4/5). Nearly all phylotype-specific PCR screens were negative; however, some individuals were positive for the Alpha-2.2 in both healthy colonies and in the European foulbrood colony (Table 3). Limited sequencing indicated that the positive universal 16S rRNA gene products in the European foulbrood colony represented multiple bacterial species, including the common secondary invader Enterococcus faecalis (Forsgren 2010).

89

Assessing the sources of the characteristic microbiota in NEWs

All DNA samples had positive amplification of the A. mellifera ef1α gene, indicating that DNA was good quality. For bees in cup cages provided only bee bread as a potential source of microbiota, several samples (3/12) yielded fungal amplicons, while bacteria were completely absent (0/12) (cups 1.1 & 1.2, Table S1). Bee bread from the hive contained both bacteria and fungi (2/2 samples); however, the characteristic microbiota phylotypes were nearly absent (Table S1). The only characteristic phylotype present was the Alpha-2.2 (Table S1).

All NEWs reared in cup cages with nurse workers were positive for bacteria and negative for fungi (Table S1). Nurses were positive for nearly all phylotypes except the

Alpha-1. NEWs exposed to these nurses contained several of the characteristic phylotypes, but were not as consistently colonized as the nurses (Table S1). NEWs exposed to nurse gut homogenate were positive for many of the characteristic A. mellifera phylotypes (Table S1). NEWs reared in a frame cage acquired several phylotypes but colonization was patchy among individuals, whether or not they were exposed to nurses

(Table S1).

Community composition of gut organs based on deep sequencing of 16S

rRNA genes

Each sample had a minimum of 2291 good sequences for data analysis (Table

S3). In addition to bacterial 16S rRNA genes, the 926f and 1492r primer set amplified the A. mellifera 18S rRNA gene, which represented a large proportion of the ileum and 90 midgut samples before they were removed for analyses (Table S4). Blastn identified multiple 97% sequence similarity OTUs for each phylotype, but manual chimera checking reduced that to 1-2 OTUs (Table S4). Each organ sample was dominated by a subset of 4 abundant OTUs, which correspond to the Beta, Gamma-1, Gamma-2, and

Firm-5 phylotypes (Figure 4). Alpha diversity metrics for each sample are listed in Table

S3. Using the weighted-UniFrac metric, PCA and UPGMA analyses separated the organ’s communities into distinct clusters. Day-30 samples were more loosely clustered with the day-9 samples, falling outside the day-9 clades on the UPGMA tree and were more distant from day-9 samples on the PCA plots (Figure 4). Pyrotag datasets are available through the NCBI Sequence Read Archive (accession number SRP008053).

Candidatus Snodgrassella alvi and Candidatus Gilliamella apicola

We propose the following candidate names for two of the organisms that were the focus of our study.

“Candidatus Snodgrassella alvi,” new lineage. Phylogenetic analysis of 16S rRNA gene sequences indicates that the Beta phylotype represents a unique clade of

Neisseriaceae, related to the genera Simonsiella and Alysiella (Martinson et al. 2011).

The corresponding 16S rRNA gene sequences have been found only within the guts of several Apis and Bombus species, and has >5% sequence divergence from other Alysiella,

Simonsiella, and other Neisseriaceae genera (Koch & Schmid-Hempel 2011a; Martinson et al. 2011). Distinguishing attributes include existence within the alimentary canal of corbiculate bees and the unique 16S rRNA gene sequence 91

TTAACCGTCTGCGCTCGCTT (positions 572-592, Escherichia coli scheme); accession AY370189 is a representative sequence. The lineage is named in reference to the entomologist Robert E. Snodgrass (1875-1962) who made important contributions to insect anatomy, morphology, evolution, and A. mellifera biology with his 1910 publication of The Anatomy of the Honey Bee (Snodgrass 1910). The epithet, alvi, is derived from the Latin word alvus, meaning “beehive or digestive organs”, and refers to

C. Snodgrassella’s presence in the bee gut.

“Candidatus Gilliamella apicola,” new lineage. Phylogenetic analysis of 16S rRNA gene sequences indicates that the Gamma-1 phylotype resides in a clade closely related to Pasteurellaceae and Enterobacteriaceae, but distinct from either (Figure 5b,

(Martinson et al. 2011). Sequences representing the Gamma-1 phylotype have been exclusively identified from within the guts of several Apis and Bombus species, and have

~6% sequence divergence from Orbus hercynius and >10% sequence divergence from members of Pasteurellaceae or Enterobacteriaceae (Koch & Schmid-Hempel 2011a;

Martinson et al. 2011). Distinguishing attributes include existence within the alimentary canal of corbiculate bees and the unique 16S rRNA gene sequence

CGAGGTCGCCTCCCTTTGTA (positions 1246-1266, E. coli scheme); accession

AY370191 is a representative sequence. The lineage is named in reference to the entomologist Martha Gilliam who pioneered the study of the microbial associates of A. mellifera during three decades of research. The epithet, apicola, is derived from the Latin word apis, meaning “bee” and the Latin suffix -cola, meaning “inhabitant of”, and refers to C. Gilliamella’s presence in the bee gut. 92

DISCUSSION

From our results, we conclude that A. mellifera has a consistent pattern of colonization by the Beta, Firm-5, and Gamma-1 phylotypes that is influenced by host age and gut morphology. The abundance and community structure of the microbiota changes through the A. mellifera life cycle, and varies among the organs of the adult gut. Bacteria were absent and potentially actively excluded from larvae and newly emerged workers.

The microbiota is established in the adult worker gut after brood cell emergence through contact with the hive and trophallaxis between nestmates; the characteristic phylotypes are maintained throughout the worker’s life, spanning diverse tasks and dietary regimens.

As observed in previous surveys (Babendreier et al. 2007; Cox-Foster et al. 2007;

Jeyaprakash et al. 2003; Martinson et al. 2011; Mohr & Tebbe 2006, 2007; Olofsson &

Vasquez 2008; Vasquez & Olofsson 2009), all naturally reared adult A. mellifera workers harbored the characteristic bacterial phylotypes within their guts (Figures 1 - 4; Table

S1). The deep sampling provided by the pyrotag data reaffirmed that the BFG phylotypes are consistent and dominant members of the microbiota (Cox-Foster et al. 2007;

Martinson et al. 2011), accounting for ~73% of the total 16S rRNA gene sequences recovered (Table S4). The Gamma-2 phylotype was also a major constituent, representing ~23% of the sequences. The rarer members of the gut microbiota, such as the Bifidobacterium phylotype, are consistently present (Figure 4) and may play important functional roles. The proportions of individual phylotypes are somewhat different from those observed in some other studies (Cox-Foster et al. 2007; Martinson et 93 al. 2011), potentially reflecting differences in DNA extraction protocols or biological differences among colonies.

Adult workers

As adult workers age, their progression of tasks requires more time outside the hive and more exposure to different foods and sources of microorganisms (Haydak

1970). Overall, the adult microbiota is fairly constant as the individual worker transitions from feeding brood within the hive to foraging in the extra-hive environment. Our results suggest that the microbiota of older workers (day 30) is larger (Figure 1a), and may shift towards higher proportions of the Gamma-1 and Gamma-2 phylotypes (day-30 samples cluster outside the day-9 samples, see Figure 4).

The crop is a muscle-lined organ of the gut that is capable of distending to accommodate nectar collected by foraging workers (Sammataro & Cicero 2010;

Snodgrass 1910). Even though the crop often contains nutrient-rich nectar that could be used as an energy source for microbes, it contains very few bacteria (Figures 1c, 2, 3a).

The frequent filling and emptying of the crop as nectar is collected and transferred to the hive for honey production could perturb the microbial community and prevent bacterial colonization. Alternatively, the enzymes added during nectar processing in the crop

(Maurizio 1975; Winston 1985) could actively deter bacteria; these enzymes have been hypothesized to be responsible for antimicrobial properties of honey (Weston 2000;

White et al. 1963). 94

Within the adult midgut, the principal site of digestion, epithelial cells secrete enzymes that digest food so that it can be absorbed through specialized midgut cells and the hindgut epithelium (Jimenez & Gilliam 1996; Terra et al. 1996). Unlike the rest of the gut, the midgut wall lacks a thin layer of cuticle called the intima (Snodgrass 1910).

Instead, the midgut epithelium produces the peritrophic membrane, a loose film that aids in digestion, protects the epithelial cells from abrasive food particles (i.e., pollen exine), and acts as a barrier to pathogens (Tellam 1996; Yue et al. 2008). This membrane is continually produced by the midgut epithelium and then shed as the meal passes, which inhibits microbial attachment (Tellam 1996). The presence of digestive enzymes and the peritrophic membrane could explain the relatively depauperate midgut (1 - 4% of the total BFG microbiota, Figures 1 & 2), even though the midgut is the largest organ in the

A. mellifera alimentary canal (Cruz-Landim & Rodrigues 1967). Further, FISH microscopy showed most midgut bacteria were located posterior, near the pylorus, which projects into the midgut (Figure 3b & c). This suggests that the midgut microbiota could be carryover from the ileum that was dissected with the midgut.

The A. mellifera ileum is a relatively small organ between the midgut and the rectum, which has deep infoldings that provide surface area for absorption of nutrients not collected in the midgut (Santos & Serrao 2006; Terra et al. 1996). Despite the midgut being much larger than the ileum, the BFG population was nearly twice as large in the ileum (5 - 10% of total BFG, Figures 1c, 2, 3c - f). In comparison to the midgut, the ileum has abundant attachment sites on its intima infoldings and access to partially digested, unabsorbed nutrients. FISH images provide further evidence that attachment is 95 important for bacterial colonization, particularly for the Beta and Gamma-1 phylotypes

(Figure 3f & g).

The ileum community appears as a stratified biofilm relative to the gut wall

(Figure 3d.1 - d.4, 3f, 3g). The biofilm is structured with the Beta phylotype abutting the host tissue, the Gamma-1 phylotype distributed in a thick mat adjacent to the Beta and the ileum wall, and the Firm-5 phylotype present in small pockets along the ileum wall.

Potentially, attachment of the Beta may enable colonization and attachment by later phylotypes, such as the Gamma-1. The resulting bacterial mat could create microgradients (e.g., nutrient, oxygen, pH), which could provide separate niches for the utilization of a variety of substrates, similar to the termite paunch community (Brune &

Friedrich 2000).

The rectum, like the crop, distends to fit more contents; this occurs continually as workers retain digested waste until they take a defecation flight to dispose of it outside the hive (Seeley 1985). In this relatively static environment, akin to the termite paunch, the contents of the rectum (mainly empty pollen exines) could serve as a nutrient source for bacteria, since the carbohydrates found in the exine are recalcitrant to direct digestion by A. mellifera (Roulston & Cane 2000; Warnecke et al. 2007). Consistent with the stable, nutrient-rich environment, the rectum harbored the majority of the microbiota, accounting for 87 - 94% of total BFG 16S rRNA genes per bee (Figures 1c & d, 2). The

Firm-5 phylotype dominated the rectal community and was ubiquitous throughout its lumen, interspersed with the digested pollen husks. Overall, the rectum contained the majority of the 16S rRNA gene copies for the BFG phylotypes and also contained 96 additional bacterial cells that did not hybridize with specific BFG probes, but only with the universal eubacterial probe. These non-BFG bacterial cells most likely represent the remaining phylotypes from the characteristic microbiota (i.e., Alpha-1, Alpha-2.1, Alpha-

2.2, Bifido, Firm-4, Gamma-2). FISH microscopy surveys of these other phylotypes would illuminate their distributions in the bee gut.

Larvae

The presence of a gut microbiota is nearly universal among animals (Ley et al.

2008), but our non-culture based methods revealed that healthy A. mellifera larvae from colonies at three geographic locations had few or no bacteria in their guts (Table 3). The scarcity or absence of gut bacteria in A. mellifera larvae seems especially odd in light of the well-characterized gut community of nurse workers that orally feed larvae

(Babendreier et al. 2007; Cox-Foster et al. 2007; Gilliam 1997; Jeyaprakash et al. 2003;

Martinson et al. 2011). A. mellifera larvae have a blind gut that prevents digested substrates from being voided until just before pupation (Chapman 1998; Snodgrass

1910). Absence or scarcity of bacteria in A. mellifera larvae has been noted on the basis of culture-based methods (Gilliam 1971; Gilliam & Prest 1987). However, other culture- based studies have suggested that larvae naturally have large amounts of the characteristic phylotypes within their guts (Forsgren et al. 2011; Mohr & Tebbe 2006;

Olofsson & Vasquez 2008). In these studies, larvae were not surface-sterilized, and contamination from the brood cell could have transferred small amounts of these phylotypes to the surface of the larvae. Alternatively, most larvae screened were 97 concurrently infected with the pathogen Paenibacillus larvae (Olofsson & Vasquez

2008), potentially altering the natural dynamics of the larval microbiota and allowing microbial colonization, as we observed for colonies known to be infected with European foulbrood. Using culture-independent methods, Mohr & Tebbe (2006) determined 16S rRNA gene profiles but not total numbers of bacteria in adult and larval A. mellifera.

They found that gut community profiles of adults were consistent over three years of sampling (and consisted of the phylotypes we report here), but that larval profiles differed from adult profiles, varied within and between sampling years, and often lacked the characteristic microbiota phylotypes (Mohr & Tebbe 2006). The irregularity of bacterial presence suggests that the larval microbiota represents the bacterial community present in their food, bee bread. In the current survey, larvae that were positive for bacteria were nearly devoid of the characteristic phylotypes except the Alpha-2.2 (Table 3). Alpha-2.2 was found in raw bee bread (Table 3) and was previously found to occur in diverse bee and wasp species (Martinson et al. 2011). Potentially, variation in the bacterial community of bee bread could explain why the presence of bacteria is variable between larvae.

Attempts at determining the protective function of probiotic bacteria have focused on introducing Lactobacillus sp. (including the Firm-5 phylotype) to larvae and subsequently infecting these larvae with a pathogen (Forsgren et al. 2011). However, these experiments relied on strains of the Firm-5 phylotype originating from adult workers and thus do not address the natural presence of Firm-5 in larvae. 98

The scarcity of bacteria in larvae suggests a growth-suppressing antimicrobial agent or strong immune response produced by the larvae and/or delivered by nurse workers during trophallaxis (Evans & Lopez 2004). In many social insects, larvae are only fed adult-processed foods, which could be altered to inhibit microbial growth or enriched for a certain subset of non-pathogenic/probiotic microbes, thus insulating the young from opportunistic pathogens (Cremer et al. 2007; Cremer & Sixt 2009). This may be especially important for A. mellifera since many of its most destructive diseases attack brood (fungal, bacterial, arachnid, and protozoan, (Schmid-Hempel 1998; Shimanuki &

Knox 2000)). These pathogens often infect host tissues by passing through the midgut wall, after being consumed (Shimanuki & Knox 2000; Yue et al. 2008). Potentially, anti- microbial compounds in larval guts prevent proliferation of the phylotypes characteristic of adult guts but fail to prevent colonization of resistant brood pathogens.

Newly emerged workers

NEWs contain few or no bacteria, as anticipated since A. mellifera goes through a complete metamorphosis in which the gut intima is shed (Kikuchi et al. 2007). In contrast to NEWs, day-9 adults have a fully developed microbiota that is not significantly different from that of older bees (Figures 1a, 2a). Because day-9 individuals are fully colonized, before their first foraging flights, the microbiota is acquired within the home colony. This system blurs the line between vertical and horizontal transmission; individuals probably do not obtain their microbiota from their mother, but through direct or indirect contact with their sisters or the hive contents. Therefore, transfer occurs within 99 the colony or “superorganism”, as in termites and leafcutter ants (Hongoh et al. 2005;

Wilson 2000). Thus, NEWs may obtain the characteristic microbiota through consumption of comb-stored honey and bee bread, through trophallaxis of nectar from older workers (Seeley 1982; Seeley 1985; Winston 1985), or through contact with the comb. Phylotypes may differ in their usual routes of transfer. In our current survey, bee bread lacked nearly all the characteristic bacterial phylotypes and did not transfer the characteristic phylotypes to NEWs (Table S1). Further, our assays showed that colonization by certain phylotypes can occur though contact with the comb or natural emergence from the brood cell and also through exposure to (and possibly trophallaxis by) older nestmates (Table S1). The crop is often referred to as the ‘social stomach of the colony’ because it distributes and receives food shared among nestmates. Although the crop contains few bacteria (Figures 1a, 2, 3a), a few transferred cells could seed a young bee’s gut and replicate into a full microbiota (Kaltenpoth et al. 2010). Normal trophallaxis could homogenize the microbiota profiles of individuals within the colony.

Coprophagy, eating of feces directly or indirectly through contact with hive surfaces, is another potential transmission route.

Potential roles of the bee gut microbiota

Studies in many animals are revealing that gut bacteria routinely perform a number of specific beneficial functions in their hosts. Bumble bees, which are close relatives of honey bees, also contain both phylotypes corresponding to the Beta and the

Gamma-1 phylotypes within their guts (Martinson et al. 2011; Mohr & Tebbe 2006), and 100 a recent study provided evidence that these gut bacteria provide adult bumble bees with protection against parasitic protozoans (Koch & Schmid-Hempel 2011b). This raises the possibility that the Beta or Gamma-1 phylotype of A. mellifera also may function in protection against disease organisms. These potential beneficial functions are critical, as

A. mellifera is the most important pollinator in agricultural systems, and thus a significant link in the human food supply.

ACKNOWLEDGEMENTS

We thank Mark Carroll, Thomas Deeby, Gloria DeGrandi-Hoffman, and Bruce

Eckholm for access to hives at the Carl Hayden Bee Research Center; Jay D. Evans for providing larvae and NEWs from the USDA Bee Research Laboratory (Beltsville, MD);

Bernhard Schink for comments on the Candidatus names; and Ellen O. Suurmeyer for assistance marking bees and helpful discussion. V. Martinson was supported by the

National Science Foundation IGERT training grant in Comparative Genomics to the

University of Arizona, and additional research support came from the Center for Insect

Science (University of Arizona) and a National Science Foundation award to N. Moran

(NSF 1046153).

101

REFERENCES

Ament SA, Corona M, Pollock HS, Robinson GE (2008) Insulin signaling is involved in the regulation of worker division of labor in honey bee colonies. Proc. Natl. Acad. Sci. USA 105:4226-4231.

Babendreier D, Joller D, Romeis J, Bigler F, Widmer F (2007) Bacterial community structures in honeybee intestines and their response to two insecticidal proteins. FEMS Microbiol. Ecol. 59:600-610.

Brune A, Friedrich M (2000) Microecology of the termite gut: structure and function on a microscale. Curr. Opin. Microbiol. 3:263-269.

Caporaso JG, Kuczynski J, Stombaugh J, et al. (2010) QIIME allows analysis of high- throughput community sequencing data. Nat. Methods 7:335-336.

Chapman RF (1998) The insects: structure and function. Cambridge University Press, Cambridge, UK.

Cole JR, Wang Q, Cardenas E, et al. (2009) The ribosomal database project: improved alignments and new tools for rRNA analysis. Nucleic Acids Res. 37:D141-D145.

Cox-Foster DL, Conlan S, Holmes EC, et al. (2007) A metagenomic survey of microbes in honey bee colony collapse disorder. Science 318:283-287.

Cremer S, Armitage SAO, Schmid-Hempel P (2007) Social immunity. Curr. Biol. 17:R693-R702.

Cremer S, Sixt M (2009) Analogies in the evolution of individual and social immunity. Philos. T. R. Soc. B 364:129-142.

Cruz-Landim C, Rodrigues L (1967) Comparative anatomy and histology of the alimentary canal of adult Apinae. J. Apicult. Res. 6:17-28.

Daims H, Stoecker K, Wagner M (2005) Fluorescent in situ hybridization for the detection of prokaryotes. In: Molecular microbial ecology (eds. Osborn AM, Smith CJ). Taylor & Francis Group, New York, NY.

Dillon RJ, Dillon VM (2004) The gut bacteria of insects: nonpathogenic interactions. Annu. Rev. Entomol. 49:71-92.

Evans JD, Chen YP, Di Prisco G, Pettis J, Williams V (2009) Bee cups: single-use cages for honey bee experiments. J. Apicult. Res. 48:300-302. 102

Evans JD, Lopez DL (2004) Bacterial probiotics induce an immune response in the honey bee (Hymenoptera : Apidae). J. Econ. Entomol. 97:752-756.

Forsgren E (2010) European foulbrood in honey bees. J. Invertebr. Pathol. 103:S5-S9.

Forsgren E, Olofsson TC, Vasquez A, Fries I (2011) Novel lactic acid bacteria inhibiting Paenibacillus larvae in honey bee larvae. Apidologie 41:99-108.

Gilliam M (1971) Microbial sterility of the intestinal content of the immature honey bee, Apis mellifera. Ann. Entomol. Soc. Am. 64:315-316.

Gilliam M (1997) Identification and roles of non-pathogenic microflora associated with honey bees. FEMS Microbiol. Lett. 155:1-10.

Gilliam M, Prest DB (1987) Microbiology of feces of the larval honey bee, Apis mellifera. J. Invertebr. Pathol. 49:70-75.

Hammer O, Harper DAT (2001) PAST: Paleontological statistics software package for education and data analysis. Palaeontol. Electron. 4:1-9.

Haydak MH (1970) Honey bee nutrition. Annu. Rev. Entomol. 15:143-156.

Hongoh Y, Deevong P, Inoue T, et al. (2005) Intra- and interspecific comparisons of bacterial diversity and community structure support coevolution of gut microbiota and termite host. Appl. Environ. Microbiol. 71:6590-6599.

Hongoh Y, Ekpornprasit L, Inoue T, et al. (2006) Intracolony variation of bacterial gut microbiota among castes and ages in the fungus-growing termite Macrotermes gilvus. Mol. Ecol. 15:505-516.

Hosokawa T, Kikuchi Y, Nikoh N, Shimada M, Fukatsu T (2006) Strict host-symbiont cospeciation and reductive genome evolution in insect gut bacteria. PLoS Biol. 4:1841- 1851.

Jeyaprakash A, Hoy MA, Allsopp MH (2003) Bacterial diversity in worker adults of Apis mellifera capensis and Apis mellifera scutellata (Insecta:Hymenoptera) assessed using 16S rRNA sequences. J. Invertebr. Pathol. 84:96-103.

Jimenez DR, Gilliam M (1996) Peroxisomal enzymes in the honey bee midgut. Arch. Insect Biochem. Physiol. 31:87-103.

Kaltenpoth M, Goettler W, Koehler S, Strohm E (2010) Life cycle and population dynamics of a protective insect symbiont reveal severe bottlenecks during vertical transmission. Evol. Ecol. 24:463-477. 103

Kikuchi Y, Hosokawa T, Fukatsu T (2007) Insect-microbe mutualism without vertical transmission: a stinkbug acquires a beneficial gut symbiont from the environment every generation. Appl. Environ. Microbiol. 73:4308-4316.

Klein AM, Vaissiere BE, Cane JH, et al. (2007) Importance of pollinators in changing landscapes for world crops. Proc. R. Soc. B 274:303-313.

Koch H, Schmid-Hempel P (2011a) Bacterial communities in central European bumblebees: low diversity and high specificity. Microb. Ecol. 62:121-133.

Koch H, Schmid-Hempel P (2011b) Socially transmitted gut microbiota protect bumble bees against an intestinal parasite. Proc. Natl. Acad. Sci. USA. 108:19288-19292.

Lee YK, Mazmanian SK (2010) Has the microbiota played a critical role in the evolution of the adaptive immune system? Science 330:1768-1773.

Ley RE, Hamady M, Lozupone C, et al. (2008) Evolution of mammals and their gut microbes. Science 320:1647-1651.

Martinson VG, Danforth BN, Minckley RL, et al. (2011) A simple and distinctive microbiota associated with honey bees and bumble bees. Mol. Ecol. 20:619-628.

Maurizio A (1975) How bees make honey. In: Honey, a comprehensive study (ed. Crane E), pp. 197-220. Heinemann, London.

Mohr KI, Tebbe CC (2006) Diversity and phylotype consistency of bacteria in the guts of three bee species (Apoidea) at an oilseed rape field. Environ. Microbiol. 8:258-272.

Mohr KI, Tebbe CC (2007) Field study results on the probability and risk of a horizontal gene transfer from transgenic herbicide-resistant oilseed rape pollen to gut bacteria of bees. Appl. Microbiol. Biot. 75:573-582.

Ochman H, Worobey M, Kuo CH, et al. (2010) Evolutionary relationships of wild hominids recapitulated by gut microbial communities. PLoS Biol. 8:e1000546.

Oliver KM, Moran NA, Hunter MS (2006) Costs and benefits of a superinfection of facultative symbionts in aphids. Proc. R. Soc. B 273:1273-1280.

Olofsson TC, Vasquez A (2008) Detection and identification of a novel lactic acid bacterial flora within the honey stomach of the honeybee Apis mellifera. Curr. Microbiol. 57:356-363.

Robinson CJ, Schloss P, Ramos Y, Raffa K, Handelsman J (2010) Robustness of the bacterial community in the cabbage white butterfly larval midgut. Microb. Ecol. 59:199- 211. 104

Roulston TH, Cane JH (2000) Pollen nutritional content and digestibility for animals. Plant Syst. Evol. 222:187-209.

Round JL, Mazmanian SK (2009) The gut microbiota shapes intestinal immune responses during health and disease. Nat. Rev. Immunol. 9:313-323.

Sammataro D, Cicero JM (2010) Functional morphology of the honey stomach wall of the European honey bee (Hymenoptera: Apidae). Ann. Entomol. Soc. Am. 103:979-987.

Santos CG, Serrao JE (2006) Histology of the ileum in bees (Hymenoptera, Apoidea). Braz. J. Morph. Sci. 23:405-413.

Schloss PD, Westcott SL, Ryabin T, et al. (2009) Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl. Environ. Microbiol. 75:7537-7541.

Schmid-Hempel P (1998) Parasites in social insects. Princeton University Press, Princeton, NJ.

Seeley TD (1982) Adaptive significance of the age polythism schedule in honeybee colonies. Behav. Ecol. Sociobiol. 11:287-293.

Seeley TD (1985) Honeybee ecology. Princeton University Press, Princeton, NJ.

Shimanuki H, Knox DA (2000) Diagnosis of honey bee diseases. Agriculture handbook no. AH-690. USDA, Washington, DC.

Snodgrass RE (1910) The anatomy of the honey bee. Government Printing Office, Washington, DC.

Stamatakis A (2006) RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics 22:2688-2690.

Tellam RL (1996) The peritrophic matrix. In: Biology of the insect midgut (eds. Lehane MJ, Billingsley PF), pp. 86-114. Chapman & Hall, London.

Terra WR, Ferreira C, Baker JE (1996) Compartmentalization of digestion. In: Biology of the insect midgut (eds. Lehane MJ, Billingsley PF), pp. 206-235. Chapman & Hall, London.

U'Ren JM, Lutzoni F, Miadlikowska J, Arnold AE (2010) Community analysis reveals close affinities between endophytic and endolichenic fungi in mosses and lichens. Microb. Ecol. 60:340-353. 105

Vane-Wright RI, Raheem DC, Cieslak A, Vogler AP (1999) Evolution of the mimetic African swallowtail butterfly Papilio dardanus: molecular data confirm relationships with P. phorcas and P. constantinus. Biol. J. Linnean Soc. 66:215-229.

Vasquez A, Olofsson TC (2009) The lactic acid bacteria involved in the production of bee pollen and bee bread. J. Apicult. Res. 48:189-195.

Warnecke F, Luginbuhl P, Ivanova N, et al. (2007) Metagenomic and functional analysis of hindgut microbiota of a wood-feeding higher termite. Nature 450:560-565.

Weston RJ (2000) The contribution of catalase and other natural products to the antibacterial activity of honey: a review. Food Chem. 71:235-239.

White JW, Schepartz AI, Subers MH (1963) The identification of inhibine, the antibacterial factor in honey, as hydrogen peroxide and its origin in a honey glucose- oxidase system. Biochim. Biophys. Acta 73:57-70.

Wilson EO (2000) Sociobiology: the new synthesis. Harvard University Press, Cambridge, MA.

Winston ML (1985) The biology of the honey bee. Harvard University Press, Cambridge, MA.

Yue D, Nordhoff M, Wieler LH, Genersch E (2008) Fluorescence in situ hybridization (FISH) analysis of the interactions between honeybee larvae and Paenibacillus larvae, the causative agent of American foulbrood of honeybees (Apis mellifera). Environ. Microbiol. 10:1612-1620.

106

a. b. 108 6 1.2 x 10 A A B AB 8 x 105 106 B B 4 x 105 4 10 C

102 Total copies of 16S rRNA copies Total Total copies of 16S rRNA copies Total 0 1 x 105

Day 1 Day 9 Beta Day 19 Day 30 Firm-5 c. Gamma-1 108 A B 106 C

D 104

102 Total copies of 16S rRNA copies Total 0

Crop Ileum Midgut Rectum

Figure 1 - Abundances of the Beta, Firm-5, and Gamma-1 phylotypes (together =

BFG) per adult worker, for different ages and gut organs, measured as copies of 16S rRNA gene. (a.) Total BFG abundance in workers at 1, 9, 19, and 30 days post emergence. (b.) Average phylotype abundances in adult A. mellifera workers (days 1 -

30). (c.) Average BFG abundances of the gut organs in adult A. mellifera workers (days 1

- 30). Letters above confidence intervals (one standard deviation) represent significance levels (Tukey’s HSD). 107

a. !"!" Day 1 Rectum Day 9 Ileum A Day 19 Midgut A A $ Day 30 !" B BC BC BC C C !"# Crop Day 1 D DEDE % !" E E DE DE

&

Total copies of 16S rRNA copies Total !"

"

Crop Ileum Day 9 Day 9 Day 9 Day 9 Midgut Rectum Day Day19 30 Day Day19 30 Day Day19 30 Day Day19 30 b. Rectum

Ileum Crop

Midgut Beta Firm-5 Gamma-1 c. Rectum !"10 Ileum A Midgut AB B 8 !" BC C CD C C D !"6 Crop E E E !"4

!"2 Total copies of 16S rRNA copies Total

"

Beta Beta Beta Beta Firm-5 Firm-5 Firm-5 Firm-5 Gamma-1 Gamma-1 Gamma-1 Gamma-1 108

Figure 2 - Comparison of the BFG community parsed by age, gut organ, and bacterial phylotype. (a.) Numbers of 16S rRNA gene copies corresponding to BFG phylotypes in A. mellifera gut organs for worker adults of different ages. (b.) Mean phylotype abundances relative to total BFG abundance for gut organs of adult workers, excluding day-1 workers. Circle area is proportional to the organ’s total BFG abundance, and the crop chart is expanded. (c.) Abundances of phylotypes in each gut organ for adult workers, excluding day-1 workers. Letters above confidence intervals (one standard deviation) represent significance levels (Tukey’s HSD).

109

Beta, Firm-5, Gamma-1 Beta Firm-5 Gamma-1 Universal Blue/White/Green Blue White Green Red a.1 a.2 a.3 a.4 a.5 L Crop

100 μm 100 μm 100 μm 100 μm 100 μm

b.1 b.2 b.3 b.4 b.5 W Midgut L 100 μm 100 μm 100 μm 100 μm 100 μm

c.1 c.2 c.3 c.4 c.5

Ileum I L

100 μm 100 μm 100 μm 100 μm 100 μm

d.1 d.2 d.3 d.4 d.5

Ileum I

50 μm L 50 μm 50 μm 50 μm 50 μm

e.1 e.2 e.3 e.4 e.5

Rectum

50 μm 50 μm 50 μm 50 μm 50 μm

f. g.

M I Ileum P L I L

100 μm 100 μm

110

Figure 3 - Localization of the Beta, Firm-5, and Gamma-1 phylotypes within the crop, midgut, ileum, and rectum of mature adult workers. Confocal microscopic images of phylotype-specific and universal bacterial FISH probes are shown, with false coloration of specific BFG and universal bacterial probes as listed. Column 1 and frames f & g show composite images for the BFG FISH probes; columns 2-4 show hybridization for the individual BFG probes; column 5 shows hybridization for the universal bacterial probe. Rows represent different gut organs; the boxed area in c.1 is enlarged in row d to show the deep infoldings of the ileum filled with bacterial cells. Abbreviations: L, gut lumen; I, cuticular intima; W, midgut wall; P, partially digested pollen; M, Malpighian tubules.

111

0.4 0.3 30 Day Day 30 Day 0.2

0.1

0.0 -0.1 PC 1 (78.35%) 30 Day -0.2 Midgut Ileum Rectum

-0.3 -0.4

0.25 0.20 0.15 0.10 0.05 0.00 -0.05 -0.10 -0.15 -0.20 PC 2 (14.66%) 2 PC

b.

30 9

9 Rectum 9

9

9

Ileum 9

30

30

9

9 Midgut 1 0.0

9 0.8

do Day ! Beta Bi Firm-4 Firm-5 0.2 Frequency

Alpha-2.2 Gamma-1 Gamma-2 a. 112

Figure 4 - Bacterial community profiles in midgut, ileum, and rectum samples from four individual A. mellifera workers (three day-9 workers and one day-30 worker), characterized using abundances of 16S rRNA gene sequences in 454 pyrotag data. (a.)

Phylotype frequencies. Dendrogram shows UPGMA clustering of bacterial communities based on weighted-UniFrac metric; all nodes have >90% bootstrap support. Each phylotype was assigned sequence clusters having top blastn hit to members of that phylotype in GenBank. No clusters comprising >0.05% of any sample had a top blastn hit other than the characteristic A. mellifera phylotypes (Alpha-2.2, Beta, Bifido, Firm-4,

Firm-5, Gamma-1, and Gamma-2), and these low frequency clusters are not shown. (b.)

Principal components analysis of the A. mellifera gut organ bacterial communities (day-

30 worker samples labeled). Pyrotag data analyzed in QIIME (Caporaso et al. 2010) using the weighted-UniFrac metric.

113

3 ) 9 ) 7 Y0965 y, boar y, X8937 AY36288 1 ! 9 sp.) 8 9 22800 Aphid 0 2 X8797 AF AF06069 s i X8937 Pasteurellaceae t Actinobacillus scotiae Apis mellifera 0 Beetle, Beetle, Candidatus 5 ana illus delphinicola Bombus _00294 uenzae c Apis mellifera 6 " m 6 8 nucleotide substitutions per site AF24771 NC Enterobacteriaceae iu

i r e 0.0 8 Gilliamella apicola inoba AF29441 t ct “Gamma-2” phylotype “Gamma-2” FJ65551 2 Actinobacillus delphinicola Ac FJ65551 8 2 EU34832 (“Gamma-1” phylotype) (“Gamma-1” 98072 6 alliba 9 us ducrey 4 Q 9 G il Phoconebacter uteri 6 ) D Haemophilus in 0 Mannheimia haemolytica 9 ) EF60853 Honey bee ( Honey 36292 aemoph 1 H Honey bee ( Honey L0609 AY 3 Bumble bee ( s 6 9 Actinobacillus arthritidis Stomaphis quercus Pasteurella multocida Stomaphis cupressi Stomaphis longirostris 9 3 , wild boar FJ61259 9 SS 8 udini SS SS 0 3 0 st 5 e Bisgaard taxa 40 AY17273 t 4 0 0 5 0 4 9 Poecilus chalcites 10 5 4 4 3 ella 9 HM10844 8 1 r 4 2 AY83774 Drosophila melanogaster 1 HM11205 eu HM11213 2 2 7 st y ( 8 DQ837611 HM215025 EF187250 HM108316 HM107876 a ! 0 Orbus hercynius HM10831 HM10831 Pasteurella testudinis P Bisgaard taxa 32 AY17272 DQ83760 10 8 DQ83760 0 AY37019 HM11197 7 AB24666 AY37019 HM11206 HM108542 M9080 HM108563 0 Bisgaard taxa 14 AY36290 Bisgaard taxa 14 L0608 HM215028 HM215029 10 9 6 HM215031 6 5 0 10 HM215032 HM215034 1 5 7 2 2 ground beetle ( 0 4 4 5 1 5 3 9 6 5 2 9 10 9 9 0 9 0 7468 4 7 9 0 4 5 0 5 3 X 6 4 9 2 7 4 0 4 4 3 hi 5 0 Thorsellia anophelis t 6 10 1 10 4 1 8 5 AF00858 0 23342 10 AB08924 AJ23342 leiogna AJ s i m 5 X8072 r 2 Phlomobacter fragariae iu Arsenophonus nasoniae AJ31183 r AY39501 e ulga ct 0 omophila X7469 s v t 10 oba eu AJ23223 t t o ia en t ho Salmonella enterica Pr 0 P 8 Escherichia coli Serra 9 Morganella morganii 9 5 Pantoea stewartii 8 Serratia marcescens 0 Vibrio cholerae Yersinia pestis 10 8 6 9 7 7 8 9 8 8 6

0 5

0 b. 10 Neisseriaceae 8 7 Neisseria ) 4 nucleotide Vitreoscilla 0.0 substitutions per site AF32814 6 AF32814 0 4 1 1 5 0 6 AJ23929 sp.) Conchiformibium, Conchiformibium, 6 9 2 Eikenella, Kingella, Eikenella, X0771 7 Snodgrassella alvi Snodgrassella 5 AF32062 AJ23928 1 Alysiella, Bergeriella, AF32815 M2251 3 sp. L0616 L0616 AJ23931 L0616 9 Simonsiella muelleri AJ23929 4 Simonsiella muelleri AF32814 cans Apis mellifera 08726 Simonsiella, Uruburuella cans ! 8 L0617 !

AB08726 5 (“Beta” phylotype) (“Beta” s AB 9 0 Bombus Eikenella 5 32815 an AJ23929 10 Neisseria lactamica c cans 8 ! L0616 ! i AF eedae 23927 3 2 tr Eikenella corrodens 38908 Neisseria sicca AJ23929 Neisseria pharyngis Neisseria gonorrhoeae 6 3 AJ m st a 7 L1073 AF deni Candidatus eedae s 4

ava s o Kingella denitri i avescens " 9 c Kingella denitri s ibiu " Neisseria meningitidis 9 u m st iella r 2 AB08726 AB08726 Neisseria polysaccharea rm 0 4 m Conchiformibium kuhniae 0 AF32814 o M2251 f Neisseria cinerea ge AY34405 ia ibiu 2 10 r 3 r 4 10 ongen 9 Honey bee ( Honey 6 hi e 7 e 4 AB08726 k 9 c rm Bergeriella denitri B Conchiformibium kuhniae ss o Neisseria 0 liformis liformis L0616 f Q25969 9 ! ! 6 Neisseria per Nei Con 4 hi H 9 58661 5

c 8 Bumble bee ( Neisseria weaveri s L0617 6 0 i 4 5 er hong AJ 7 4 s 5 ct Con 2 1 1 4 a su ui 4 Alysiella crassa 2 8 ll s 6 8 Alysiella Alysiella Kingella kingae Alysiella crassa 9 8 9 0 9 Lariba Kingella oralis 5 0 HM11317 9 uella 9 8 4 r 10 9 sp. GQ24935 8 EU05555 ruburue HM11211 9 HM11209 U ubu 8 8 HM11322 HM11309 r 0 8 AY37018 EU05554 0 HM215022 U HM108731 HM108727 0 10 10 7 HM108703 6 Chromobacterium subtsugae 0 7 Vitreoscilla stercoraria Vitreoscilla 10 0 2 0 2 7 8 10 2 3 0 8 10 5 8 3 7 0 10 a. 114

Figure 5 - Phylogenetic placement of (a.) Candidatus Snodgrassella alvi (“Beta” phylotype) within the Neisseriaceae and (b.) Candidatus Gilliamella apicola (“Gamma-1” phylotype) within the Gammaproteobacteria. Numbers on branches represent bootstrap support (RAxML with 100 bootstrap replicates).

115

e r cles y , and c X opho s

- r t) ed R n y3 y5 luor 488 or 15 F C C f e ! uo xas e b xa r e emaining o e r T l r tu A ra ttachme

a FISH p (5' empe t e b o eached 55°C, with r r t / p til it c n iable annealing ) r du

a p v o , r cle u s y c

or 5 f a length (b qPCR p 20 15 20 18 128 114 210 es p y

t y 1°C each b ylo h T ed cles of 95°C AA r T y T A C e A

c T T T A w e G A T C G G C c G T o C C T G A G T T G A A A C A T A TT C GGAA T G GG A C C C G G T A C C A amma-1 p T G T A A T T C G T C G T CC T A G C T as then l A T G GG GG A e sequen G T A C GG w C C T A C A T b or 10 min; 40 C A T T f CCC G C A o T G T A T T r T C A GG GG C C G C C G T T A A T A C A C T A T T T G CC G G C T C G A GG T irm-5, and T T T A A A cles and A F T T C

y A A C A T c T T rimer/p G T CCC G C A P C T GGA C C G T T T T tions: 95°C c ee eta, r B ea

r or all or the f f qtF qtR es - - e name b or the ! rst th f b o qtF qtR qtR r o - - - qtF as used r - w ol as 68°C c o w m5-81 m5-183 t r r amma1-459 amma1-648 rimer/p eta-1009 eta-1115 i i e o r r P B B F F G gam1-1246 eub339 G beta-572 ! r5-575 tu a r

tion p c ea empe r t

wn get o ia r r e PCR primers and FISH p e v t ti c

a ouchd t tit e and ta he annealing n T b

. es ua o ersal ba s r wing b v

Q m-5 m-5 o o r r amma-1 amma-1 eta eta r i i or 5 B F G Uni G F B oll f f he rimer/p able 1. T T P qPCR primers FISH p a 72°C 116

Table 2: Statistical analysis of qPCR resultsa

Results including individuals from days: 1, 9, 19, 30 9, 19, 30 Factor F P value F P value

Colony F1,432 0.3724 F1,312 0.5438

-106 Age (A) F3,432 2.4 x 10 * F3,312 0.0361*

-84 -80 Gut Section (GS) F3,432 6.6 x 10 * F3,312 9.8 x 10 *

Bacterial Phylotype (BP) F2,432 0.0011* F2,312 0.2391

-33 A – GS F9,432 1.6 x 10 * F9,312 0.2821 A – BP F 0.074 F 0.2386 6,432 6,312

A – GS – BP F18,432 0.8135 F18,312 0.992

-5 -5 GS – BP F6,432 3.7 x 10 * F6,312 2.4 x 10 *

aAll factors were analyzed with the standard least squares test.

117

e a v a amma-1 5/5 0/5 0/1 — — — — — 0/5 0/4 0/1 G imer g r ial p r e t c m-5 r i 5/5 0/5 0/1 — — — — — 0/5 0/4 0/1 F ersal ba v

m-4 r i 0/5 0/1 0/5 — — — — — 0/5 0/4 0/1 F

med because the uni r o f

r eta 5/5 0/5 0/1 — — — — — 0/5 0/4 0/1 B e not pe r e w

tions c lpha- 2.2 5/5 1/1 2/5 5/5 3/4 1/1 — — — — — A ea r es t a

eened) r lpha-2.1 . sc o 5/5 0/5 0/1 — — — — — 0/5 0/4 0/1 A e/n . A dash indic v d a oo r ers

k r o oulb f w lpha-1 1/5 2/5 0/1 — — — — — 0/5 0/4 0/1 A . of bees positi o opean r u ia r E , e t B c esult (n r . EF s

, and adult nurse s tion ersal Ba W c v eening E r c ea r 5/5 1/5 5/5 5/5 4/5 1/5 0/5 0/5 0/5 0/5 0/5 S Uni , N e

e v a v r

e positi t a y) y) y) a a a a a a a a een of la h e stage h h v v v v v v v v r f r r r r r r r r y) h . T (healt t erial sc s s C t c tus) and li W W d-instar la d-instar la d-instar la a esul a E E r r r en, r , MD 1 (EFB) , MD 2 (healt , MD 3 (healt B 5th-instar la Nurse adults 5th-instar la 5th-instar la 3 N 5th-instar la N 3 5th-instar la 3 e numbers indic v e e e e c a v y (st ti n

a oldfa est H olo eltsvill eltsvill eltsvill able 3: ucson, AZ (healt B T neg B B B W T C a 118

SUPPLEMENTARY RESULTS

- - 0/2 1/3 4/5 2/4 1/6 3/3 5/5 amma-1

G - - 0/2 3/3 5/5 4/4 5/6 3/3 5/5 irm-5 F

- - 0/2 2/3 5/5 4/4 2/6 3/3 3/5 irm-4 F - - eta 0/2 0/3 1/5 3/4 5/6 3/3 5/5 B - - 2/2 0/4 1/3 3/5 6/6 2/3 4/5 lpha-2.2 A

- - 0/2 0/6 1/3 1/5 2/4 3/3 5/5

lpha-2.1 A

- - 0/2 0/3 0/5 0/4 0/6 0/3 0/5 lpha-1 A ees

b eria ersal t 2/2 v 0/6 0/6 c 3/6 5/6 4/6 6/6 3/3 5/6 a age reared B c Uni

ersal 2/2 v 0/6 0/6 0/6 0/6 0/3 0/6 3/6 0/6 ungi F Uni ead

r NEWs NEWs NEWs NEWs NEWs NEWs Nurses Nurses ee b B Screened

eria in bee bread and cup/frame t c

Nurses Nurses Bee bread Bee bread Hive contact Hive contact & NEW’s microbiota Assessed source of Nurse gut homogenate

age up 2 up 3 ame 1 ame 2 up 1.1 up 1.2 C r r C C C C F F able S1. Survey of fungi and ba T Bold numbers indicate positive reactions. “-” indicates reactions were not performed because the universal bacterial primer gave a negative result. 119

ed t c y 9 y 9 y 9 y 30 y 30 y 30 a a a olle a a a c D D D D D D y a D

gan r tum c idgut leum e I M ut o R

G

T T

C A G C A T

T G T T C C erse primer v A T e Y R GG C

A T

GG C A G T

T A d primer r a GA K w r

o f AAA Y er & CT k A AA Lin T C tion a orm f

e c T T G A G G G G C T G C T G C C C C G T A G C T G GA GA A A G

T A T A C A T T A A C C T T T T G A C A T C T G A T T C A C A C T A G C T C G T G G G C T T G G de sequen C C A C C G A de and primer in A A T G A o T T T A C o G G A G C G A

c c C C T T T C C r C r a B

otag ba r y p I30.1 I09.3 I09.1 I09.2 R30.1 R09.3 R09.1 R09.2 M09.3 M30.1 M09.2 M09.1

ample ID S able S2. 454 T 120

e ap t r a 8.1 9.5 7.3 6.7 9.5 7.3 44.3 13.6 13.5 13.8 11.4 13.6 14.7 13.5 13.4 19.6 10.9 13.5 11.1 11.4 13.6 14.7 13.5 13.4 otst o richness estim B age d's r o e 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 o v G o c x e 0.4 0.3 0.5 0.8 0.3 0.3 0.4 0.5 0.9 0.9 0.8 0.5 0.4 0.3 0.5 0.8 0.3 0.3 0.4 0.5 0.9 0.9 0.8 0.5 ind Simpson x e 1.2 1.3 0.9 0.3 1.3 1.4 1.2 0.8 0.4 0.3 0.6 1.1 1.2 1.3 0.9 0.3 1.3 1.4 1.2 0.8 0.4 0.3 0.6 1.1 ind hannon S 7.0 7.0 7.0 88.5 15.0 10.0 14.0 13.0 14.0 11.0 13.0 15.0 13.0 13.0 18.8 10.5 14.0 13.0 10.3 11.0 13.0 15.0 13.0 13.0 hao 1 C s U 7 8 7 6 8 7 T 36 12 13 12 11 13 13 13 13 18 10 13 10 11 13 13 13 13 O es c otal 2790 2291 4430 5297 7445 8039 6035 2788 2290 4430 5295 7445 8039 6035 T 13889 11104 10678 13351 11156 13871 11104 10678 13351 11156 equen S I09.1 I09.2 I09.3 I30.1 I09.1 I09.2 I09.3 I30.1 ample R09.1 R09.2 R09.3 R30.1 R09.1 R09.2 R09.3 R30.1 M09.1 M09.2 M09.3 M30.1 M09.1 M09.2 M09.3 M30.1 S y a 9 9 9 9 9 9 30 30 30 30 30 30 D tum tum otag samples gan r c c r idgut idgut leum leum y I I O p M M Re Re y of 454 s t U T s ersi O U v T on O t on t lpha di A ith single ithout single able S3. W W T 121 ium r e t c a obium er r r t ium quaba c r A cus e t c er cus c t ia o oba c r r c c ium i o r r c ium e r a t r e xiba yloba ellula c ma t ellulosimic a r onia e er h holde i c olirub A P t C t tae_sedis; k P eae S r c et r on c e als C c M ia oba ia eae; R eae; eae eae; r r y eae;Singulisphae eae; eae yba eae; c c c c r e c eae c c eae x c a a a a t C c c r r r r a o eae; eae; c r eae;Bu eae; a a e e e c c r r t t t eta eta eta c c e a en t e e c c c c c c ia r t eae;Bi ! doba eae iales_in ia t ia t eae eae; r c y y y oba e r r r c c cus S c c c c cus e r t c t c m m m eae;Kocu c e ia ia ia ia eae c o t r r c r r o o o o oba c c t t t e e c n e e r oba oba eae eae; eae t t c c c t t o t t ca ia o c c c t c c holde holde holde c c r r xiba c e e hodoba hodoba hodoba omonospo k k k ia ia ia e r e c c o R R R eae;Iamia r r r lan lan lan r r r t yloba obacillus ep c A A c t c r P P P h nae oba o oba on t c olirub iales;E r r r r S A C S et ales; ales; ales; e omic ic ic ic r r r t oba r M c e e v e M P M M illales; illales; t t t eae; a iales;Bu iales;Bu iales;Bu etales; etales; etales; r r ales; ales ales; l c c c r r r c eae;La r r r c c c F c e e e y y y oba edis XI; t t t ca r iales;Bi ! doba iales;Bi ! doba iales;Neisse iales;Neisse iales;Neisse S c c c c m m m e r r r r r etales; etales; etales; etales etales; t o obiales obiales; o o o e e c c c c c t t t iales; t t obiales;Iamia c z z n holde holde holde r r y y y y y c c c c c tae k k k o oba oba oba hodoba hodospi hi hodoba hodoba hi hodospi e r r r r t r r r t m m m m m R R R R R R R obacilla e ia;E ia ia ia c t lan lan lan r r r r c ep c P P P e e e e ia; ia; ia; ia; ia; r ia; ia; n t t t t r r r r r t r r I tino tino tino tino tino ia;Bu ia;Bu ia;Neisse ia;Neisse ia;Bu ia;Neisse c c c c S e e e e e e e r r r r r r oba c cidimic c c c c olirub olirub olirub t t t t t t t v e e e e e e amily I;GpI c c c c c c c S A A A S A S A A t t t t t t F a l c c c c c c ia; ia; ia; ia; ia; ia; ia; ia; ia; ia;Bi ! doba ia;Bi ! doba F etacia; etacia; etacia; ia; eoba eoba eoba eoba r r r r r r r r r r r c c c r idiales; idiales idiales t t t t ia_Gp3;Gp3 e e e e e e e e e e e ia; r r r y y y r tae_sedis e t t t t t eoba t t eoba t t t eoba eoba eoba t eoba eoba o o o o r r t t t t t t t t e r r r r c c c c c c c c c c c eoba eoba eoba eoba eoba eoba e m m m t c e o o o o o o o t t t t t t t c c o o o r r r r r r r c o o o o o o t t t obacillales obacillales; obacillales;La obacillales r r r r r r c c c t t t t c c c c a_in oba tinoba tinoba tinoba tinoba tinoba tinoba tinoba tinoba tinoba tinoba tinoba lan r lan lan anoba v ammap ammap ammap ammap lphap lphap lphap lphap lphap lphap lphap idia;Clost idia;Clost idia;Clost c c c c c c c c c c c etap etap etap etap etap etap P P P y r r r a cidoba l G G G B B G B A B B A A A B A A A A A A A A A A A A A A C A F es; es; es; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; t t t ia; r r r r r r r r r r r r r r r r r r r r r r r r r r r r ia; r e e e es; r e e e e e e e e e e e e e e e e e e e e e e e e e e e e t e c c c t t t t t t t t t t t t t t t t t t t t t t t t t t t t e t y y y t c c c c c c c c c c c c xi c c c c c xi c c c c c c c c c c c c c es;Bacilli;La es;Bacilli;La es;Clost es;Bacilli;La es;Clost es;Clost es;Bacilli;La e e m m m t t t t t t t oide o o o r t t t o " o " r r e c c c eoba eoba eoba eoba eoba eoba eoba eoba eoba eoba eoba eoba eoba eoba eoba eoba eoba t t t t t t t t t t t t t t t t t t tinoba tinoba tinoba tinoba tinoba tinoba tinoba tinoba tinoba tinoba tinoba micu micu micu micu anoba micu micu micu c o o o o o o o o o o o o o o o o o r r r r r r r c c c c c c cidoba c c c c c y lan lan lan r r r r r r r r r r r r r r r r r i i i i i i i P P F A P F P A A P P P A A P P F F A A P P P A P P P P A P A F A F P C A P F P ia; ia; ia; ia; ia; ia ia; ia; ia; ia; ia; ia; ia; ia; ia; ia;OP10;OP10_gene ia; ia; ia; ia ia; ia; ia; ia;Chlo ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia; ia;Chlo ia;Ba ia; ia; ia r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c onsensus Lineage C Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba 1 7 6 1 1 1 2 6 1 1 2 1 4 1 1 1 3 1 1 1 8 2 1 1 1 1 1 1 1 1 1 1 98 24 77 47 35 41 10 24 12 336 272 734 201 9752 23120 96505 19254 96482 42476 otal T 0 5 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 13 54 59 13 13 29 57 319 105 1708 6035 6035 3678 y 30 a R30.1 D 0 3 1 0 0 0 0 0 8 0 0 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6 0 0 0 0 0 19 13 13 69 52 298 173 675 189 9660 11156 11156 R09.3 0 2 1 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 5 0 1 1 0 0 0 0 0 1 0 0 0 0 0 84 39 13 13 28 190 105 tum samples y 9 8039 8039 7580 c a D R09.2 Re 8 5 0 0 0 0 0 0 3 0 0 0 0 0 0 3 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 86 13 13 31 62 163 358 208 13351 13351 12421 R09.1 0 2 0 0 0 0 0 0 0 2 7 0 0 0 0 0 0 0 0 0 0 5 0 0 7 0 0 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 263 4330 7445 2839 7445 y 30 a I30.1 D 7 8 1 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 11 24 11 80 15 887 1439 6380 1833 10678 10678 I09.3 0 0 1 2 0 0 0 0 0 0 0 0 3 0 0 2 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 12 10 43 968 819 y 9 5297 2020 5295 1436 a leum samples I I09.2 D 5 2 0 0 0 0 0 0 0 0 0 0 2 0 0 4 0 0 0 7 0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 13 13 57 41 1014 1726 3160 5081 11104 11104 I09.1 1 0 6 0 1 0 0 0 1 0 8 0 0 0 0 0 0 0 0 0 0 0 0 0 0 8 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 38 347 4035 4430 4430 y 30 a M30.1 D 0 0 0 0 0 0 1 7 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 19 403 376 1490 2291 2290 M09.3 0 0 0 0 0 0 0 1 0 0 0 0 0 2 0 0 0 0 0 1 0 0 0 0 4 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 12 16 10 11 917 432 293 y 9 1111 2790 2788 idgut samples a D M M09.2 otag samples r y 0 0 3 1 1 1 0 5 1 1 2 1 2 1 1 0 1 6 2 1 1 1 1 3 4 1 1 0 1 1 0 0 1 1 0 0 1 1 0 0 p 36 21 18 11 314 174 6727 4436 2158 13889 13871 M09.1 s otal otal U T T s es within 454 T c U O T s y istic istic istic istic istic istic istic istic istic istic istic O t r r r r r r r r r r r U r s in sample s in sample e e e e e e e e e e e T t t t t t t t t t t t U U on c c c c c c c c c c c O t T T a a a a a a a a a a a r r r r r r r r r r r O O on on on on on on on on on on on on on on on on on on on on on on on on t t t t t t t t t t t t t t t t t t t t t t t t cha cha cha cha cha cha cha cha cha cha cha e simila TU abundan ------c otal otal O m-5 m-4 m-5 T T TU description r r r amma-2 amma-2 amma-2 amma-1 lpha-2.2 eta eta eta i i i w O G G F non single single G G non single single non single A non non single single B B single non non single single B single non non non single single Bi ! do single non single single Bi ! do single single single F single single single F single a 2 3 6 7 R 21 38 25 36 56 76 64 48 40 60 79 33 32 52 42 75 68 62 29 11 44 78 15 10 39 19 49 22 24 12 47 50 34 73 14 53 35 66 20 55 57 45 80 97% sequen TU ID able S4. otals with single otals without single O T T T 122 eae c ia r e t c eae oba c r e ia t r e n eae t c c ia r yta iales;E r h e t op c t ep r oba t r iales;Bi ! doba iales;Neisse S e r r t e t n c ia ia ia ia;E r r r r oplast; r e e e e t t t t ia;Neisse c c c c r e t c ia;Bi ! doba ia;Chlo eoba eoba eoba eoba r r t t t t e e t o o o o t r r r r c eoba c t o obacillales r t c tinoba anoba ammap ammap ammap ammap c etap y G G G B G A C ia; ia ia; ia ia; ia ia ia; ia; ia; ia; r r r r r r r r r r r e e e e e e e e e e e t t t t t t t t t t t c c c c c c c c c c c es;Bacilli;La es;Bacilli t t eoba eoba eoba eoba eoba eoba eoba eoba eoba a 18S rRNA gene t t t t t t t t t r tinoba anoba micu micu o o o o o o o o o e r r c y r r r r r r r r r f i i P P P P P C P A F P F P P ia; ia; ia; ia; ia; ia; ia; ia; ia ia; ia; ia ia; ia; ia; r r r r r r r r r r r r r r r e e e e e e e e e e e e e e e t t t t t t t t t t t t t t t c c c c c c c c c c c c c c c pis melli onsensus Lineage C Ba Ba Ba A Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba 1 9 8 19 15 33 37 70 60 18 12 42 16 22 231 53564 otal T 0 0 0 0 1 2 2 0 0 6 0 0 4 0 24 164 R30.1 5 0 0 3 3 3 0 7 1 0 6 1 61 16 10 18 R09.3 0 1 0 0 0 3 0 0 1 0 0 1 1 0 5 46 R09.2 1 0 0 0 2 9 0 0 1 4 1 3 7 15 13 15 R09.1 2 0 3 0 5 5 0 0 0 0 0 0 0 0 0 2810 I30.1 3 9 6 0 0 0 0 0 1 6 0 2 0 26 36 1077 I09.3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5635 I09.2 9 2 2 1 1 0 0 2 0 10 84 12 10 13 15 2534 I09.1 0 0 0 5 0 0 0 0 0 0 0 0 0 0 0 11266 M30.1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 27 12003 M09.3 0 0 0 3 0 0 0 0 1 0 0 0 0 0 0 9359 M09.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 112 8594 s U M09.1 T O ed v o em r 18S a r es oplast c r ti ! ed and m-4 m-4 m-5 m-5 TU description r r r r amma-2 amma-1 amma-1 amma-2 amma-1 amma-1 amma-1 eta n i i i i O G G G A. mellife Chlo G B Bi ! do Bi ! do F F G G G F F 0 70 71 67 51 23 63 28 59 41 27 30 54 65 58 17 anually ide TU ID himeric sequen M C O 123

APPENDIX C: HAPPENSTANCE SEQUENCING OF THE NEAR-COMPLETE

GENOME OF A BUMBLE BEE GUT SYMBIONT

To be published in an environmental microbiology journal.

124

ABSTRACT

The natural gut microbiota assists animals with the critical enterprises of development, nutrition, and defense. Genome sequencing has provided insights into the nature of these relationships. Recently the bumble bee microbiota was shown to protect against trypanosomatid parasites; however, the mechanism for this is unknown. Here we report the nearly complete genome of Ca. Gilliamella str. Bimp1 (BiG) that was inadvertently sequenced alongside the genome of the Common Eastern Bumble bee

(Bombus impatiens). BiG represents the first draft-genome from a divergent bacterial lineage related to Enterobacteriales and Pasteurellales that has been associated with guts of several diverse insect species, and that recently received the candidate name

“Gilliamella apicola” for strains present in honey bees. Metabolic reconstruction of BiG shows that genes are absent for a functioning TCA cycle, but that it has retained the cytochrome bd complex that is utilized in low oxygen environments. Anaerobic pathways for nitrate respiration, mixed acid fermentation, and citrate fermentation are present and may be important for survival in the bee hindgut. Unlike the mucus layer of the mammalian gut, the insect hindgut is lined with cuticle; therefore, the intact Flp pilus that

BiG harbors may be critical for adhesion and colonization. BiG contains a type 6 secretion system, as well as, many antibiotic/multidrug transporters, which may be critical for interactions with its host and other gut commensals or pathogens. This genome has signatures of reduction (2.0 megabase pairs) and rearrangement, similar to genomes of other host-associated bacteria. This sequence provides a glimpse into one of 125 the constituents of the bumble bee microbiota, which may contribute to the defense of its host against pathogens.

INTRODUCTION

Bacteria that reside within the animal gut have mechanisms to moderate interactions with their host, and to cohabit with other microbes (Hongoh et al. 2008;

Walter et al. 2011). These interactions can fall on a spectrum from pathogenic to mutualistic. Whereas enteric pathogens access nutrition by infecting host tissues, indigenous microbiota species often utilize substrates from the host diet and may provide the host with nutritional benefits (Ashida et al. 2011; Flint et al. 2008; Warnecke et al.

2007). For example, germ-free mice require 30% more calories than do mice with an intact microbiota (Wostmann et al. 1983). Utilization of energy sources is varied among currently sequenced gut bacteria (Walter & Ley 2011). For instance, the human associate

Bacteroides thetaiotaomicron has a wide array of polysaccharide metabolism genes, whereas the vertically transmitted Candidatus Ishikawaella capsulata (symbiont of a plataspid stinkbug) has a highly reduced genome, limited to a few energy capabilities

(Nikoh et al. 2011; Xu et al. 2003). Gut bacteria can also endow their host with defense against invading pathogens and parasites, with both active and passive methods (Stecher

& Hardt 2011). Microbes that provide benefits to their host are often transmitted between generations and form well-established interactions with their host (Bright & Bulgheresi

2010). 126

Honey bees (Apis sp.) and bumble bees (Bombus sp.) harbor distinct bacterial microbiotae (Babendreier et al. 2007; Cox-Foster et al. 2007; Jeyaprakash et al. 2003;

Koch & Schmid-Hempel 2011a; Martinson et al. 2011; Martinson et al. 2012). Whereas

Apis mellifera’s microbiota consists of eight phylotypes from four diverse bacterial classes (Martinson et al. 2011), Bombus species are commonly inhabited by only two phylotypes (Koch & Schmid-Hempel 2011a, b; Martinson et al. 2011). Here I use

"phylotype" to refer to a set of closely related strains, similar to the usual boundaries for species names in bacteria. The Bombus phylotypes are closely related to the Apis- associated lineages of Ca. Gilliamella apicola (Gammaproteobacteria) and Ca.

Snodgrassella alvi (Betaproteobacteria) (Martinson et al. 2012). Experimental exposure assays demonstrated that the natural Bombus terrestris microbiota provides an “extended immune phenotype” against the trypanosomatid gut parasite Crithidia bombi (Koch &

Schmid-Hempel 2011b). Evidence of social transmission within a colony and of vertical inheritance between generations suggests that these microbes may have reduced genome sizes, as also observed in endosymbionts and some other insect gut symbionts (Hosokawa et al. 2006; Koch & Schmid-Hempel 2011b; Martinson et al. 2012; Moran et al. 2008;

Nikoh et al. 2011).

Previous phylogenetic reconstructions have placed Ca. Gilliamella apicola both as a member of Enterobacteriaceae (Jeyaprakash et al. 2003) and as a member of

Pasteurellaceae (Olofsson & Vasquez 2008). Insect bacterial symbionts often have elevated rates of evolution causing their 16S rRNA genes to position on long branches that could misrepresent their true phylogenetic relationships. The Bombus impatiens 127 genome-sequencing project, by chance, recovered the genome sequence of a gammaproteobacterium related to the Ca. Gilliamella apicola clade. This genome sequence provides insights into the phylogenetic relationships and lifestyle of Ca.

Gilliamella apicola, as well as clues about its role in B. impatiens biology.

MATERIALS AND METHODS

Sample DNA, sequencing, identification and assembly of bacterial reads

A single male Bombus impatiens that was aged greater than 24 hours was collected from colony purchased from Biobest Biological Systems (Leamington, Ontario,

Canada). DNA was extracted from the entire B. impatiens specimen using a standard phenol/chloroform preparation.

The genome of Bombus impatiens was sequenced using Illumina sequencing technology, resulting in approximately 65-fold coverage of the genome. Three paired-end libraries were constructed, with fragment sizes of 400, 4000, and 8000 bp. The 250 megabase (Mb) (estimated size) genome was assembled using the CABOG assembler

(Miller et al. 2008) modified to handle short Illumina reads. Assembly resulted in 69,944 contigs with a length greater than 100 bp. In addition, the assembly contained 1,086,650 degenerate contigs, primarily small repetitive sequences or contigs with very low quality.

60,355,858 reads remained as unassembled singletons. The assembly is deposited in

GenBank under the accession AEQM00000000. 128

To scan these contigs for possible bacterial symbionts, we initially aligned them to the genomes of four strains of Wolbachia, which are ubiquitous endosymbionts of invertebrates. We aligned the genomic contigs against each of these four strains using the promer program from the MUMmer package (Kurtz et al. 2004), which translates both the reference and the query sequences to amino acids in all six reading frames. This allowed for more sensitive alignment than a DNA-based approach. We identified several contigs that had strong homology to at least one of the Wolbachia species. We isolated these contigs and used the Blastx program to map them against the complete non- redundant protein database at NCBI, in order to find bacteria that were closer to the symbiont in B. impatiens. The best hits from this mapping were to three

Gammaproteobacteria species: Photorhabdus asymbiotica, Yersinia enterocolitica, and

Proteus mirabilis.

We then aligned all assembled contigs to the complete genomes of each of these three bacteria. We ran DNA alignments and translated protein alignments using the nucmer and promer programs from the MUMmer package, and mapped all contigs against both the DNA and the protein sequences of the three bacterial genomes. The more sensitive protein-based alignments were compared to the annotated coordinates of the proteins in each of the bacterial genomes. Each contig that contained at least one complete protein was denoted as possibly bacterial. In addition, contigs not longer than

500 bp that contained at least a partial match at greater than 60% identity to any bacterial protein were also denoted as bacterial. This analysis identified 367 regular contigs, 1,129 degenerate contigs, and 255,589 singleton reads as possibly bacterial in origin. Next, we 129 used Blastx to search each of these contigs and reads against the entire NCBI protein database, and we eliminated any contig with a better match to a eukaryotic species than to a bacterial genome. This left 343 regular, 941 degenerate, and 255,589 singleton contigs as likely bacterial sequences.

We used the CABOG assembler’s raw output files to locate all reads used to build these bacterial contigs and extracted these reads from the original sequence files with their paired-end mates. This resulted in 615,185 mate pairs from the 400 bp insert size library, 20,716 mate pairs from the 4 Kb insert size library, 8,164 mate pairs from the 9

Kb insert size library, and 121,568 unpaired reads. These reads were assembled de novo with the CABOG assembler. The final bacterial assembly contained 1,999,325 bp in just

84 contigs. The largest contig contained 110,984 bp, and the assembly had an N50 size of 39,885 bp. The 84 contigs were combined into 33 scaffolds spanning 2,004,741 bp, with a scaffold N50 size of 98,624 bp and a maximum scaffold of 204,248 bp. The approximate coverage of the genome is 37X.

Annotation of the bacterial genes

Gene annotation was completed in the automated Integrated Microbial Genomes

Expert Review (IMG/ER) pipeline (Markowitz et al. 2012). Protein-coding sequences and RNA-coding genes were predicted within its framework using Prodigal and tRNAS- can-1.23 (Markowitz et al. 2012). Functional annotations were assigned to genes based on protein domain characterization with: COG clustering, Pfam, TIGRfam, InterPro domains, Gene Ontology (GO) terms, and KEGG Orthology (KO) terms with metabolic 130 pathway maps. Additional manual assessement with KEGG (Kanehisa et al. 2008),

EcoCyc (Keseler et al. 2009), and the MetaCyc Pathologic program (Caspi et al. 2010) was performed to check pathway completion. A metabolic map was manually created for the B. impatiens gammaproteobacterium.

Core gene phylogeny

Single-copy orthologous (SiCO) genes were selected from an original set of 203 consistently present, non-horizontally transferred core genes (Lerat et al. 2003) to reconstruct the phylogenetic placement of the B. impatiens gammaproteobacterium. SiCO genes were selected from 28 gammaproteobacterial genomes using SiCO gene lists in

MaGE (Vallenet et al. 2006) or using a cutoff ratio of >0.30 bitscore ratio cutoff in a blastp search with the 203 SiCO genes from E. coli (Lerat et al. 2003). A final collection of 89 genes were selected, individually aligned in MUSCLE (Edgar 2004), and concatenated together. Gblocks (Talavera & Castresana 2007) was used to remove poorly aligned regions and the resulting alignment consisted of 27452 amino acid sites.

Maximum likelihood reconstruction was performed with RAxML (Stamatakis 2006) on

100 bootstrap replicates using the PROTGAMMA algorithm and the WAG substitution matrix, which were selected with ProtTest v2.4 (Abascal et al. 2005). Individual SiCO gene trees were built with the same methods as the multiprotein dataset and subsequently sorted with PhyloSort (Moustafa & Bhattacharya 2008).

131

16S rRNA gene phylogeny

The 16S rRNA gene was used to reconstruct and refine the phylogenetic relationships of the B. impatiens gammaproteobacterium among a representative set of

Ca. Gilliamella apicola, A. mellifera Gamma-2 phylotypes (collected from the genera

Apis and Bombus), and closely related sequences from the NCBI database. One of the four 16S rRNA gene copies in the B. impatiens gammaproteobacterium genome was selected and aligned with the 16S rRNA dataset using Infernal (Nawrocki et al. 2009). A maximum likelihood phylogeny was constructed with RAxML using the GTRGAMMA parameter and 100 bootstrap replicates (Stamatakis 2006).

Putative horizontal gene transfer

IMG annotation pipeline identified putative horizontally transferred genes.

Further analysis of potential horizontal gene transfer was assayed with a phylogenetic pipeline modified from Moustafa et. al. (2008). Briefly, the pipeline identifies closely related genes in the NCBI database, aligns the amino acid sequences, and constructs phylogenies for each. PhyloSort (Moustafa & Bhattacharya 2008) was used to find trees which indicated horizontal transfer from Firmicutes to the B. impatiens gammaproteobacterium.

132

RESULTS

Sequencing of the B. impatiens genome resulted in the byproduct sequencing of a gammaproteobacterial isolate. From the resulting ~250 Mb of assembled data, 2 Mb were partitioned into a gammaproteobacterial genome (see Methods). Resulting contigs that represent the gammaproteobacterial genome harbor at least one ORF per contig and have an N50 length of 39.9 Kb. A blastp search for 203 SiCO genes identified 195, each with exactly one copy (96% of SiCOs). Calculation of GC content of the contigs produced a unimodal distribution with a mean of 36.6%.

The B. impatiens gammaproteobacterial genome is at least 1.99 Mb in size (Table

1). A total of 50 tRNA genes and 23 tRNA synthetase genes were identified, encompassing all 20 amino acids. Altogether, 1694 protein-coding genes were identified from the assembly, 14% (236) of them with unknown function (Table 1). Roughly 72% had functions predicted with cluster of orthologous genes, the largest categories were

Translation (9.7%), General function only (9.3%), Amino acid transport and metabolism

(8.6%), Cell wall biogenesis (7.6%), and Replication (6.8%), Coenzyme transport and metabolism (5.8%), Carbohydrate transport and metabolism (5.3%). Pseudogenes were not identified among the contigs. The primary metabolism of the B. impatiens gammaproteobacterium, as inferred from the genome sequence, is illustrated in Figure 1.

133

Phylogenetic analyses

The phylogenetic relationships between the bacterial orders of

Gammaproteobacteria included in the analysis (Enterobacteriales, Pasteurellales,

Vibronales, Alteromonadales, and Pseudomonadales) were correctly reconstructed with high bootstrap support (Figure 2a). The B. impatiens gammaproteobacterium was placed as sister to Enterobacteriales (Figure 2a). Trees created with individual SiCO genes varied in their support, mainly due to their differing sequence complexity (46 - 1407 amino acids in length; Figure 2c). The majority (61/89) of individual SiCO genes resulted in a tree topology uniting the B. impatiens gammaproteobacterium, Enterobacteriales, and

Pasteurellales. Nearly 3/4 of those trees (44/61) placed the B. impatiens gammaproteobacterium sister to, or within the Enterobacteriales with an average bootstrap support >80% (Figure 2c). The 16S rRNA gene tree contained monophyletic clades for the Gamma-1 and Pasteurellales but not for Enterobacteriales, and the relationships between the bacterial families were not well supported (Figure 2b, Figure

S1).

Putative horizontal gene transfer

A total of 54 genes were identified by IMG as putatively horizontally transferred from Firmicutes. Of those genes, PhyloSort supported horizontal transfer from

Firmicutes for only 39 (Table 2).

134

DISCUSSION

Our analyses show that the genome of a gammaproteobacterium sequenced concurrently with B. impatiens is nearly complete, as indicated by the unimodal G+C content, the presence of a complete tRNA synthetase complement, and the presence of

96% of a defined set of single copy orthologous (SiCO) genes. Previous genome sequencing projects have been shown to produce large datasets corresponding to symbiotic microorganisms (Salzberg et al. 2005); however, most of these symbionts do not assemble as well as the B. impatiens gammaproteobacterial genome retrieved in our study (84 contigs), suggesting a near-clonal population in the source DNA. The final assembly produced a genome at least 1.99 Mb in size with 1770 putative genes and a GC content of 36.6%, which makes it relatively small and AT-rich compared to genomes of other Gammaproteobacteria. Its small size and GC bias is comparable to previously sequenced host-dependent commensals and pathogens (e. g., Ca. Regiella insecticola

(Hansen et al. 2012) or Haemophilus somni (Siddaramappa et al. 2011)). It is particularly similar to members of the Pasteurellales, which associate with the mucosal surfaces of birds and mammals (Bisgaard 1993).

Our multiprotein phylogeny reconstructed a basal branching pattern joining five gammaproteobacterial orders that agrees with previously published genome-based phylogenies (Gao et al. 2009; Lerat et al. 2003; Williams et al. 2010). Protein and 16S rRNA gene analyses placed the B. impatiens gammaproteobacterium as a singleton clade that splits the generally unified relationship of Enterobacteriales and Pasteurellales 135

(Figure 2). Strong support for a sister relationship between the B. impatiens gammaproteobacterium and the Enterobacteriales was recovered in the multiprotein dataset (Figure 2). Additionally, of the individual SiCO gene trees that retrieved

Enterobacteriales and Pasteurellales as a unified group (not considering the B. impatiens gammaproteobacterium), nearly 75% supported placement as sister to, or within

Enterobacteriales (Figure 2c). In contrast, phylogenetic analysis with 16S rRNA gene sequences placed the B. impatiens gammaproteobacterium as a member of a clade sister to Pasteurellales (Figure 2b, Figure S1). However, other relationships within the

Enterobacteriales were not constructed correctly and the sister relationship had low support, suggesting that the 16S rRNA gene is not appropriate to resolve relationships at the family level in these taxa.

The 16S rRNA gene phylogeny positioned the B. impatiens gammaproteobacterium within the newly described lineage “Candidatus Gilliamella apicola”, thus it represents a strain found in B. impatiens that will be referred to as Ca.

Gilliamella str. Bimp1 (BiG). The clade that contains BiG also includes Ca. Gilliamella apicola sequences recovered from several Apis and Bombus species, as well as, sequences mostly from other animal-associated samples (fly, beetle, aphids, and wild boar).

Potentially all of these sequences represent organisms associated with insect guts. Within this clade, BiG clusters among sequences collected from native European Bombus species and from lab-reared B. impatiens colonies in North America. The BiG, obtained from individuals of a lab colony in North America, is similar (98.6% sequence similarity) to a sequence that was collected from a native B. terrestris individual in Northern Germany 136

(HM215025, Figure 2b). This suggests that the BiG phylotype is a common associate of geographically widespread Bombus species. Members of this putative insect-gut clade have >10% sequence divergence (16S rRNA gene) to taxa outside the clade. A recent survey of wild and lab-reared Drosophila species recovered many bacterial phylotypes that clustered within this clade (Chandler et al. 2011), further suggesting a relationship of clade members with insect guts. Lipid analysis of a member of this clade, Orbus hercynius (>5% divergence to bee sequences), revealed a fatty acid profile dissimilar from those in Enterobacteriales and Pasteurellales (Volkmann et al. 2010). The current genome phylogeny, multiple 16S rRNA surveys from diverse insects, and lipid profiles of

Orbus hercynius suggest this clade is a novel gammaproteobacterial family or order that is commonly associated with the insect gut.

Metabolic reconstruction of Ca. Gilliamella str. Bimp1 (BiG)

The BiG genome contains the majority of genes in predicted metabolic pathways for glycolysis, gluconeogenesis, the full pentose phosphate pathway, nucleotide metabolism, lipopolysaccharide production, peptidoglycan fabrication, heme and siroheme, and ubiquinone assembly. Biosynthetic capabilities remain intact for the majority of the 20 amino acids and many vitamins and cofactors. Critical genes are missing for several pathways, including the biosynthesis of Ile, Val, Leu, Gly, and Pro

(Figure 1). However, deeper sequencing may identify these genes because they are part of otherwise intact pathways. 137

Genomic capabilities for the TCA cycle, NADH dehydrogenase I, cytochrome c oxidase and reductase, and fumarate reductase are not complete in the BiG genome.

Large portions of these pathways are missing, which are all necessary for oxidative phosphorylation. This suggests that the microbe cannot produce energy with oxygen as the final electron acceptor. The animal hindgut is often nearly anaerobic, which might make BiG’s lost pathways unnecessary for successful colonization. Large sections of pathways seem to be deleted or rearranged on otherwise intact contigs (example in Figure

S2). Overall, the genome lacks large areas of synteny with other genomes, even in areas conserved among many species of Enterobacteriales and Pasteurellales. The loss of central metabolism attributes (e. g., TCA cycle) and the lack of synteny with other bacterial genomes suggest that this genome underwent rearrangement and reduction similar to that observed in other symbiotic genomes (Burke & Moran 2011; Nikoh et al.

2011).

Further, the genome exhibits a scarcity of mobile genetic elements, as observed in obligate host associates (Ochman & Davalos 2006). Gut bacteria with strict host associations (e. g., Helicobacter sp., Lactobacillus reuteri, Pasteurellales species) often have small genomes, suggesting that BiG may have a restricted host distribution (Walter et al. 2011; Walter & Ley 2011). Future genome sequencing of members within the insect-gut clade would provide comparisons of strains in different hosts, potentially revealing genes that are specific to each host niche (Frese et al. 2011; Hansen et al.

2012). 138

BiG contains multiple mechanisms facilitating an anaerobic lifestyle. It has retained the low-oxygen cytochrome bd complex that is critical for E. coli colonization of the mouse gut by enabling utilization of the small amounts of oxygen present (Jones et al.

2007; Jones et al. 2011). In E. coli his mechanism maximizes net energy benefit and reinforces the anoxic environment by scavenging the remaining oxygen. Additionally, the genome’s respiratory nitrate reductase A and formate dehydrogenase N are intact; these together can produce a proton motive force through anaerobic nitrate reduction, resulting in cytoplasmic nitrite (Jones 1980; Jormakka et al. 2002). The reduction of nitrite to ammonia by the NADH nitrite reductase can serve as an “electron sink” during anaerobic growth, shifting the fermentation toward the production of ATP and acetate (Cole &

Brown 1980). Additionally, this pathway may be used to shunt nitrite to ammonia for glutamate production (Stewart 1988). The bee gut is full of nitrogenous waste, though mostly in the form of uric acid and not ammonia (McNally et al. 1965). Genes for aerobic or anaerobic degradation of uric acid or urea to ammonia (uricase, xanthine dehydrogenase, urease) are not encoded in the genome of BiG. Therefore, assimilation of nitrite may be advantageous to the gut colonizers.

Recent in situ analysis of the A. mellifera microbiota showed that the majority of the bacteria were resident within the hindgut (including Ca. Gilliamella apicola)

(Martinson et al. 2012). Oxygen levels could govern colonization of the gut organs (i.e., anaerobic hindgut, aerobic foregut), restricting BiG clade bacteria to low oxygen or anaerobic environments. Fluorescent in situ microscopy of the Bombus microbiota and 139

microsensor surveys (i. e., O2) of the Apis and Bombus gut are needed to determine the breadth of this pattern (Schramm 2006).

Full pathways are present for mixed acid and citrate fermentation, which yield end products of ethanol, formate, and the short chain fatty acids acetate and lactate. Short chain fatty acids are the bulk of carbon and energy sources of ruminant animals (Dehority

1997). This raises the possibility that the bacterium is providing its host a nutritional benefit through biosynthesis of needed compounds (Breznak & Kane 1990). However, biosynthetic contributions (e. g., amino acids or vitamins) are probably not very significant because the bee diet is composed of both easily accessible mono- and disaccharides and protein-rich pollen (Kane 1997). Recent in situ experiments using gnotobiotic Drosophila show that the developmental benefits provided by a commensal gut bacterium are linked with acetate production (Shin et al. 2011). The bacterium stimulates Drosophila insulin-signaling pathways, which modulates development rate, body size, and energy metabolism. Metabolic production of acetate by BiG may similarly affect the Bombus host.

Potential interactions with the host and other gut microorganisms

Several putative host-interaction factors were present, including Sel1 repeat proteins, bacterial Ig-like domains, and bacterial α2-macroglobulins; these could be critical for recognition of this bacterial strain by the host epithelium (Budd et al. 2004). A full Flp pilus gene set was present; this apparatus is known to be critical for adhesion and biofilm formation (Tomich et al. 2007). Adhesion may play a critical role for insect gut 140 associates because, unlike mammals, insects lack the ability to secrete a mucus layer that facilitates microbial residence (Bignell et al. 1980). Therefore, microorganisms within the insect hindgut must attach to the cuticular lining to avoid elimination with feces.

Several BiG genes suggest the method by which nutrition is obtained from the bumble bee gut. Aiding in the digestion of refractory polysaccharides found in the pollen does not appear to be within the capabilities of this microorganism; however, it has transporters (bglF) and beta-glucosidases (bglA, glucoside hydrolase family 4) that likely confer the ability to degrade the products of cellulase (e. g., cellobiose and cellotriose)

(Grabnitz et al. 1991). The bee genome contains an intact gene encoding a cellulase belonging to the glucoside hydrolase 9 family, similar to endocellulases, which break cellulose randomly (Watanabe & Tokuda 2010). These can be taken up by the phosphotransferase system and degraded with its two beta-glucosidases (EC3.2.1.86).

This potentially allows the gut bacterium to metabolize the breakdown products of the cellulose coat of pollen that the bee must degrade to obtain the protoplasm (Kunieda et al.

2006; Willis et al.). Metabolism of these compounds could provide energy to this organism, since glucose and fructose (honey constituents) are mostly absorbed in the midgut (Crailsheim 1988a, b).

Ca. Gilliamella apicola str. Bimp1 contained putative horizontally transferred genes from members of the Firmicutes; notably, several sugar uptake and degradation genes were identified, including the previously mentioned bglA (Table 2). These genes may be critical for survival in the bee gut, which is exposed to numerous sugars found in nectar. An intact operon for the uptake of mannose (phosphotransferase system, PTS) 141 seems to have been transferred from a species related to Bacillus. The mannose PTS has been shown to have an extensive history of horizontal transfer in bacteria, and is mostly found in bacteria associated with animal guts (Zúñiga et al. 2005). Mannose is toxic to honeybees and potentially to bumblebees, therefore microbial mannose PTS systems could protect the host from low amounts of mannose often present in nectar (Barker &

Lehner 1974).

The genome contains four putative collagenases, which is a high number for its size. Readily accessible collagen is not present within the insect gut; however, the basal lamina that separates epidermal cells from hemolymph contains collagen fibrils (Lane et al. 1996). Collagenases have been commonly classified as virulence factors because they allow bacteria to move deeper into host tissues, suggesting that this bacterium may infect the bumble bee gut lining (Harrington 1996). Complex interactions with the host epithelium are additionally supported by the presence of multiple secretion systems. A putative hemolysin may target host cells, but alternatively could target pathogens within the bumble bee gut (e. g., Nosema bombi, Crithidia bombi), thereby providing a defensive benefit to its host.

The recently described T6SS evolved from viral tail fibers to produce a syringe- like secretion system that is present in many bacteria, and is becoming understood as pivotal to interactions among bacteria, and between bacteria and eukaryotes (Leiman et al. 2009; Mougous et al. 2006; Pukatzki et al. 2006; Schwarz et al.). The cell-puncturing device (VgrG gene) of the T6SS, is critical for attachment and can discriminate between targets (Zheng et al. 2011). The BiG genome contains three intact VgrGs and 13 142 additional VgrG fragments (Table S1), which may correspond to between 12 and 16 full

VgrGs, a relatively large number for a genome of this size (Pukatzki et al. 2009). These

VgrG elements do not display target specificity or function, but some have genes adjacent to the C-terminus that may be T6SS effectors that control interactions with the host and other microbes present in the gut (Jani & Cotter 2010; Murdoch et al. 2011) (Table S1).

For example, asymptomatic colonization of the mouse gut by Helicobacter hepaticus was disrupted when its T6SS was knocked out, leading to over-colonization by H. hepaticus and an inflammatory reaction by the host (Chow et al. 2010). This example suggests that the T6SS facilitates signaling between the bacterium and the host or that the bacterium inoculates the gut epithelium with anti-inflammatory compounds. T6SS effectors also have lytic effects (Russell et al. 2011) that can be delivered to specific eukaryotic or bacterial targets, depending upon the VgrG utilized (Hood et al. 2010; Zheng et al. 2011).

Thus, further investigation of the BiG T6SS and its biological effects on the host bumble bee, gut pathogens, and members of the indigenous microbiota may demonstrate context- dependent interactions.

The marked abundance of antibiotic/multidrug resistance transporters including several ABC, DMT, MOP, MFP, Eam/Emr, and arabinose efflux pumps also suggests that it is resistant to attack from invading microorganisms. The Bombus hive is a resource-rich environment that attracts a variety of pathogens, many of which are specific to the gut environment (Schmid-Hempel 1998). Colonization by BiG could prevent invasion of the host epithelium; alternatively, these pumps could be used to export unidentified antimicrobial compounds to actively kill invading microbes. However, 143 prediction of these short (10-50 amino acid) peptides is often missed by software programs (Fjell et al. 2007). These compounds have been shown to differentiate between eukaryotic an bacterial microbes within the gut, and could provide directed defense against specific invaders (Azambuja et al. 2004).

CONCLUSION

Recent surveys of Bombus species show a reduction in range distributions among many European and North American species, which have been correlated with infection by Nosema bombi (Biesmeijer et al. 2006; Cameron et al.). Protection from pathogens is in part mitigated by the microbiota in Bombus terrestris, and insights into the mechanism underlying this protective phenotype may lead to novel treatment options (Koch &

Schmid-Hempel 2011b). The Ca. Gilliamella str. Bimp1 (BiG) genome represents a divergent lineage of Gammaproteobacteria that is specific to honey bees and bumble bees and that belongs to a larger clade repeatedly collected from insect guts (Chandler et al.

2011; Koch & Schmid-Hempel 2011a; Martinson et al. 2011). This genome has lost critical pathways in oxygenic respiration, but has retained pathways for an anaerobic or microaerophilic lifestyle, which may restrict this organism to the insect hindgut environment. Further, this genome provides candidate mechanisms for protection against antagonistic microorganisms invading the gut.

144

ACKNOWLEDGEMENTS

We thank Ellen O. Suurmeyer for her helpful discussion and comments on the manuscript and Jen Wisecaver for bioinformatics assistance. V. Martinson was supported by the Center for Insect Science (University of Arizona) and a National Science

Foundation award to N. Moran (NSF 1046153). Sequencing and analysis were supported by NIH Director's Pioneer Award 1DP1OD006416-01 (G.E. Robinson), and in part by

NIH grant R01-HG006677 (S.L. Salzberg).

145

REFERENCES

Abascal F, Zardoya R, Posada D (2005) ProtTest: selection of best-fit models of protein evolution. Bioinformatics 21:2104-2105.

Ashida H, Ogawa M, Kim M, Mimuro H, Sasakawa C (2011) Bacteria and host interactions in the gut epithelial barrier. Nat. Chem. Biol. 8:36-45.

Azambuja P, Feder D, Garcia ES (2004) Isolation of Serratia marcescens in the midgut of Rhodnius prolixus: impact on the establishment of the parasite Trypanosoma cruzi in the vector. Exp. Parasitol. 107:89-96.

Babendreier D, Joller D, Romeis J, Bigler F, Widmer F (2007) Bacterial community structures in honeybee intestines and their response to two insecticidal proteins. FEMS Microbiol. Ecol. 59:600-610.

Barker RJ , Lehner Y (1974) Acceptance and sustenance value of naturally occurring sugars fed to newly emerged adult workers of honey bees (Apis mellifera L). J. Exp. Zool. 187:277-285.

Biesmeijer JC, Roberts SPM, Reemer M, et al. (2006) Parallel declines in pollinators and insect-pollinated plants in Britain and the Netherlands. Science 313:351-354.

Bignell DE, Oskarsson H, Anderson JM (1980) Specialization of the hindgut wall for the attachment of symbiotic microorganisms in a termite Procubitermes aburiensis (Isoptera, Termitidae, Termitinae). Zoomorphology 96:103-112.

Bisgaard M (1993) Ecology and significance of Pasteurellaceae in animals. Zbl. Bakt.- Int. J. Med. M 279:7-26.

Breznak JA, Kane MD (1990) Microbial H2/CO2 acetogenesis in animal guts - nature and nutritional significance. FEMS Microbiol. Rev. 87:309-313.

Bright M, Bulgheresi S (2010) A complex journey: transmission of microbial symbionts. Nat. Rev. Microbiol. 8:218-230.

Budd A, Blandin S, Levashina EA, Gibson TJ (2004) Bacterial alpha2-macroglobulins: colonization factors acquired by horizontal gene transfer from the metazoan genome? Genome Biol. 5:R38.

Burke GR, Moran NA (2011) Massive genomic decay in Serratia symbiotica, a recently evolved symbiont of aphids. Genome Biol. Evol. 3:195-208. 146

Cameron SA, Lozier JD, Strange JP, et al. (2011) Patterns of widespread decline in North American bumble bees. Proc. Natl. Acad. Sci. USA 108:662-667.

Caspi R, Altman T, Dale JM, et al. (2010) The MetaCyc database of metabolic pathways and enzymes and the BioCyc collection of pathway/genome databases. Nucleic Acids Res. 38:D473-D479.

Chandler JA, Lang JM, Bhatnagar S, Eisen JA, Kopp A (2011) Bacterial communities of diverse Drosophila species: ecological context of a host-microbe model system. PLoS Genet. 7:e1002272.

Chow J, Lee SM, Shen Y, Khosravi A, Mazmanian SK (2010) Host-bacterial symbiosis in health and disease. Adv. Immunol. 107:243-274.

Cole JA, Brown CM (1980) Nitrite reduction to ammonia by fermentative bacteria: short circuit in the biological nitrogen cycle. FEMS Microbiol. Lett. 7:65-72.

Cox-Foster DL, Conlan S, Holmes EC, et al. (2007) A metagenomic survey of microbes in honey bee colony collapse disorder. Science 318:283-287.

Crailsheim K (1988a) Intestinal transport of sugars in the honeybee (Apis mellifera). J. Insect Physiol. 34:839-845.

Crailsheim K (1988b) Regulation of food passage in the intestine of the honeybee (Apis mellifera). J. Insect Physiol. 34:85-90.

Dehority BA (1997) Foregut fermentation. In: Gastrointestinal microbiology (eds. Mackie RI, White BA, Isaacson RE), pp. 39-83. Chapman & Hall, New York, NY.

Edgar RC (2004) MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 32:1792-1797.

Fjell CD, Hancock REW, Cherkasov A (2007) AMPer: a database and an automated discovery tool for antimicrobial peptides. Bioinformatics 23:1148-1155.

Flint HJ, Bayer EA, Rincon MT, Lamed R, White BA (2008) Polysaccharide utilization by gut bacteria: potential for new insights from genomic analysis. Nat. Rev. Microbiol. 6:121-131.

Frese SA, Benson AK, Tannock GW, et al. (2011) The evolution of host specialization in the vertebrate gut symbiont Lactobacillus reuteri. PLoS Genet. 7:e1001314.

Gao B, Mohan R, Gupta RS (2009) Phylogenomics and protein signatures elucidating the evolutionary relationships among the Gammaproteobacteria. Int. J. Syst. Evol. Microbiol. 59:234-247. 147

Grabnitz F, Seiss M, Rucknagel KP, Staudenbauer WL (1991) Structure of the beta- glucosidase gene bglA of Clostridium thermocellum. Eur. J. Biochem. 200:301-309.

Hansen AK, Vorburger C, Moran NA (2012) Genomic basis of endosymbiont-conferred protection against an insect parasitoid. Genome Res. 22:106-114.

Harrington DJ (1996) Bacterial collagenases and collagen-degrading enzymes and their potential role in human disease. Infect. Immun. 64:1885-1891.

Hongoh Y, Sharma VK, Prakash T, et al. (2008) Complete genome of the uncultured Termite Group 1 bacteria in a single host protist cell. Proc. Natl. Acad. Sci. USA 105:5555-5560.

Hood RD, Singh P, Hsu FS, et al. (2010) A type VI secretion system of Pseudomonas aeruginosa targets, a toxin to bacteria. Cell Host Microbe 7:25-37.

Hosokawa T, Kikuchi Y, Nikoh N, Shimada M, Fukatsu T (2006) Strict host-symbiont cospeciation and reductive genome evolution in insect gut bacteria. PLoS Biol. 4:1841- 1851.

Jani AJ, Cotter PA (2010) Type VI secretion: not just for pathogenesis anymore. Cell Host Microbe 8:2-6.

Jeyaprakash A, Hoy MA, Allsopp MH (2003) Bacterial diversity in worker adults of Apis mellifera capensis and Apis mellifera scutellata (Insecta:Hymenoptera) assessed using 16S rRNA sequences. J. Invertebr. Pathol. 84:96-103.

Jones RW (1980) Proton translocation by the membrane-bound fomate dehydrogenase of Esherichia coli. FEMS Microbiol. Lett. 8:167-171.

Jones SA, Chowdhury FZ, Fabich AJ, et al. (2007) Respiration of Escherichia coli in the mouse intestine. Infect. Immun. 75:4891-4899.

Jones SA, Gibson T, Maltby RC, et al. (2011) Anaerobic respiration of Escherichia coli in the mouse intestine. Infect. Immun. 79:4218-4226.

Jormakka M, Tornroth S, Byrne B, Iwata S (2002) Molecular basis of proton motive force generation: structure of formate dehydrogenase-N. Science 295:1863-1868.

Kane MD (1997) Microbial fermentation in insect guts. In: Gastrointestinal microbiology (eds. Mackie RI, White BA, Isaacson RE), pp. 231-268. Chapman & Hall, New York, NY.

Kanehisa M, Araki M, Goto S, et al. (2008) KEGG for linking genomes to life and the environment. Nucleic Acids Res. 36:D480-D484. 148

Keseler IM, Bonavides-Martinez C, Collado-Vides J, et al. (2009) EcoCyc: A comprehensive view of Escherichia coli biology. Nucleic Acids Res. 37:D464-D470.

Koch H, Schmid-Hempel P (2011a) Bacterial communities in central European bumblebees: low diversity and high specificity. Microb. Ecol. 62:121-133.

Koch H, Schmid-Hempel P (2011b) Socially transmitted gut microbiota protect bumble bees against an intestinal parasite. Proc. Natl. Acad. Sci. USA. 108:19288-19292.

Kunieda T, Fujiyuki T, Kucharski R, et al. (2006) Carbohydrate metabolism genes and pathways in insects: insights from the honey bee genome. Insect Mol. Biol. 15:563-576.

Kurtz S, Phillippy A, Delcher AL, et al. (2004) Versatile and open software for comparing large genomes. Genome Biol. 5:R12.

Lane NJ, Dallai R, Ashurst DE (1996) Structural macromolecules of th cell membranes and the extracellular matrices of the insect midgut. In: Biology of the insect midgut (eds. Lehane MJ, Billingsley PF), pp. 115-152. Chapman & Hall, London.

Leiman PG, Basler M, Ramagopal UA, et al. (2009) Type VI secretion apparatus and phage tail-associated protein complexes share a common evolutionary origin. Proc. Natl. Acad. Sci. USA 106:4154-4159.

Lerat E, Daubin V, Moran NA (2003) From gene trees to organismal phylogeny in prokaryotes: The case of the gamma-proteobacteria. PLoS Biol. 1:101-109.

Markowitz VM, Chen IMA, Palaniappan K, et al. (2012) IMG: the integrated microbial genomes database and comparative analysis system. Nucleic Acids Res. 40:D115-D122.

Martinson VG, Danforth BN, Minckley RL, et al. (2011) A simple and distinctive microbiota associated with honey bees and bumble bees. Mol. Ecol. 20:619-628.

Martinson VG, Moy J, Moran NA (2012) Establishment of characteristic gut bacteria during development of the honey bee worker. Appl. Environ. Microbiol. 78:2830-2840.

McNally JB, McCaughe WF, Standife LN, Todd FE (1965) Partition of excreted nitrogen from honey bees fed various proteins. J. Nutr. 85:113-116.

Miller JR, Delcher AL, Koren S, et al. (2008) Aggressive assembly of pyrosequencing reads with mates. Bioinformatics 24:2818-2824.

Moran N, McCutcheon JP, Nakabachi A (2008) Genomics and evolution of heritable bacterial symbionts. Annu. Rev. Genet. 42:165-190.

Mougous JD, Cuff ME, Raunser S, et al. (2006) A virulence locus of Pseudomonas aeruginosa encodes a protein secretion apparatus. Science 312:1526-1530. 149

Moustafa A, Bhattacharya D (2008) PhyloSort: a user-friendly phylogenetic sorting tool and its application to estimating the cyanobacterial contribution to the nuclear genome of Chlamydomonas. BMC Evol. Biol. 8:6.

Moustafa A, Chan CX, Danforth M, et al. (2008) A phylogenetic approach for studying plastid endosymbiosis. Genome Inform. 21:165-176.

Murdoch SL, Trunk K, English G, et al. (2011) The opportunistic pathogen Serratia marcescens utilizes type VI secretion to target bacterial competitors. J. Bacteriol. 193:6057-6069.

Nawrocki EP, Kolbe DL, Eddy SR (2009) Infernal 1.0: inference of RNA alignments. Bioinformatics 25:1335-1337.

Nikoh N, Hosokawa T, Oshima K, Hattori M, Fukatsu T (2011) Reductive evolution of bacterial genome in insect gut environment. Genome Biol. Evol. 3:702-714.

Ochman H, Davalos LM (2006) The nature and dynamics of bacterial genomes. Science 311:1730-1733.

Olofsson TC, Vasquez A (2008) Detection and identification of a novel lactic acid bacterial flora within the honey stomach of the honeybee Apis mellifera. Curr. Microbiol. 57:356-363.

Pukatzki S, Ma AT, Sturtevant D, et al. (2006) Identification of a conserved bacterial protein secretion system in Vibrio cholerae using the Dictyostelium host model system. Proc. Natl. Acad. Sci. USA 103:1528-1533.

Russell AB, Hood RD, Bui NK, et al. (2011) Type VI secretion delivers bacteriolytic effectors to target cells. Nature 475:343-U392.

Salzberg SL, Hotopp JCD, Delcher AL, et al. (2005) Serendipitous discovery of Wolbachia genomes in multiple Drosophila species. Genome Biol. 6:402.

Schmid-Hempel P (1998) Parasites in social insects. Princeton University Press, Princeton, NJ.

Schramm A (2006) Microsensors for the study of microenvironments and processes in the intestine of invertebrates In: Intestinal Microorganisms of Soil Intertebrates (eds. Konig H, Varma A), pp. 463-473. Springer-Verlag, Berlin.

Schwarz S, Hood RD, Mougous JD (2010) What is type VI secretion doing in all those bugs? Trends Microbiol. 18:531-537.

Shin SC, Kim SH, You H, et al. (2011) Drosophila microbiome modulates host developmental and metabolic homeostasis via insulin signaling. Science 334:670-674. 150

Siddaramappa S, Challacombe JF, Duncan AJ, et al. (2011) Horizontal gene transfer in Histophilus somni and its role in the evolution of pathogenic strain 2336, as determined by comparative genomic analyses. BMC Genomics 12:570.

Stamatakis A (2006) RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics 22:2688-2690.

Stecher B, Hardt WD (2011) Mechanisms controlling pathogen colonization of the gut. Curr. Opin. Microbiol. 14:82-91.

Stewart V (1988) Nitrate respiration in relation to facultative metabolism in enterobacteria. Microbiol. Rev. 52:190-232.

Talavera G, Castresana J (2007) Improvement of phylogenies after removing divergent and ambiguously aligned blocks from protein sequence alignments. Syst. Biol. 56:564- 577.

Tomich M, Planet PJ, Figurski DH (2007) The tad locus: postcards from the widespread colonization island. Nat. Rev. Microbiol. 5:363-375.

Vallenet D, Labarre L, Rouy Z, et al. (2006) MaGe: a microbial genome annotation system supported by synteny results. Nucleic Acids Res. 34:53-65.

Volkmann M, Skiebe E, Kerrinnes T, et al. (2010) Orbus hercynius gen. nov., sp. nov., isolated from faeces of wild boar, is most closely related to members of the orders 'Enterobacteriales' and Pasteurellales. Int. J. Syst. Evol. Microbiol. 60:2601-2605.

Walter J, Britton RA, Roos S (2011) Host-microbial symbiosis in the vertebrate gastrointestinal tract and the Lactobacillus reuteri paradigm. Proc. Natl. Acad. Sci. USA 108:4645-4652.

Walter J, Ley RE (2011) The human gut microbiome: ecology and recent evolutionary changes. Annu. Rev. Microbiol. 65:411-429.

Warnecke F, Luginbuhl P, Ivanova N, et al. (2007) Metagenomic and functional analysis of hindgut microbiota of a wood-feeding higher termite. Nature 450:560-U517.

Watanabe H, Tokuda G (2010) Cellulolytic systems in insects. Annu. Rev. Entomol. 55:609-632.

Williams KP, Gillespie JJ, Sobral BWS, et al. (2010) Phylogeny of Gammaproteobacteria. J. Bacteriol. 192:2305-2314.

Willis JD, Oppert B, Oppert C, Klingeman WE, Jurat-Fuentes JL (2011) Identification, cloning, and expression of a GHF9 cellulase from Tribolium castaneum (Coleoptera: Tenebrionidae). J. Insect Physiol. 57:300-306. 151

Wostmann BS, Larkin C, Moriarty A, Brucknerkardoss E (1983) Dietary intake, energy metabolism, and excretory losses of adult male germfree wistar rats. Lab. Anim. Sci. 33:46-50.

Xu J, Bjursell MK, Himrod J, et al. (2003) A genomic view of the human-Bacteroides thetaiotaomicron symbiosis. Science 299:2074-2076.

Zheng J, Ho B, Mekalanos JJ (2011) Genetic analysis of anti-amoebae and anti-bacterial activities of the type VI secretion system in Vibrio cholerae. PLoS ONE 6:e23876.

Zúñiga M, Comas I, Linaje R, et al. (2005) Horizontal gene transfer in the molecular evolution of mannose PTS transporters. Mol. Biol. Evol. 22:1673-1685.

152

Table 1. General features of the Ca. Gilliamella str. Bimp1 (BiG) genome.

Chromosome (bp) >1 999 325 bp

Extrachromosomal elements Unknown

GC Content 36.60%

Total predicted CDS 1770

Coding density (%) 82

Average CDS length (bp) 954 bp/318 aa

Pseudogenes 0

rRNA operons

5S rRNA 7 16S rRNA 4 23S rRNA 5 tRNAs 50

153

TCC 51507 A

TCC 35646 A

TM300

. 168 . 168 r r

YK9

TCC 25745 W23 T2-87

A

TTU M1-001

WCH70

Listeriaceae bacterium Geobacillus sp. Enterococcus gallinarum EG2 Syntrophomonas wolfei Goettingen, DSM 2245B Paenibacillus curdlanolyticus Bacillus subtilis spizizenii Clostridium bartlettii DSM 16795 Enterococcus sp. 7L76 Bacillus thuringiensis sv israelensis Enterococcus faecium E980 Listeria monocytogenes 08-5923 Lactobacillus rhamnosus Lc 705 Listeria grayi DSM 20601 Bacillus coagulans 2-6 Bacillus subtilis subsp. Subtilis st Staphylococcus carnosus From Genome Desulfitobacterium dehalogenans JW/IU-DC1, Bacillus cereus cytotoxis NVH 391-98 Leptotrichia goodfellowii F024 Listeria welshimeri sv 6b, SLCC5334 Listeria welshimeri sv 6b, SLCC5334 Bacillus coagulans 36D1 Anaerofustis stercorihominis DSM 17244 Desulfitobacterium hafniense DCB-2 Staphylococcus aureus ST398 Bacillus coagulans 36D1 Enterococcus italicus DSM 15952 Listeria welshimeri sv 6b, SLCC5334 Bacillus pumilus SAFR-032 Eubacterium cylindroides Clostridium cf. saccharolyticum K10 Bacillus atrophaeus 1942 Clostridium cf. saccharolyticum K10 Enterococcus italicus DSM 15952 Listeria welshimeri sv 6b, SLCC5334 Bacillus thuringiensis IBL200 Pediococcus pentosaceus Listeria seeligeri sv 1/2b, SLCC3954 Bacillus subtilis subsp. Subtilis st 1 1 167 1 COG1335 COG3684 COG0406 COG0583 COG0620 COG1263 COG2723 COG3715 COG4687 COG3716 COG COG0225 COG0214 COG2814 COG2820 COG COG0456 COG0531 COG2323 COG1296 COG4392 COG0286 COG1605 COG0732 COG4975 COG0610 COG4905 COG0213 COG1972 COG2189 COG3587 COG0561 COG3444 COG1636 COG1418 COG03 COG0561

0.34 .. 0.37 .. 0.37 .. 0.37 ...... 0.35 0.35 0.35 .. .. 0.39 Contig GC % .. 0.35 0.36 .. 0.37 0.37 .. 0.36 ...... 0.37 0.36 ...... 0.37 0.38 .. .. 0.37 0.39 ..

fold Length (bp) f 64926 .. 47248 .. 27350 .. 65738 ...... 102764 14474 38071 .. .. 38219 Sca .. 45947 31379 .. 37254 77226 .. 20279 ...... 39885 34555 ...... 10510 59508 .. .. 51284 6516 .. Accession

fold IMG f GBi_ctg820001289446 .. GBi_ctg820001289445 .. GBi_ctg820001289437 .. GBi_ctg820001289444 .. GBi_ctg820001289410 GBi_ctg820001289416 ...... GBi_ctg820001289403 .. GBi_ctg820001289386 Sca .. GBi_ctg820001289448 GBi_ctg820001289453 .. GBi_ctg820001289435 GBi_ctg820001289436 .. GBi_ctg820001289430 .. .. GBi_ctg820001289428 .. GBi_ctg820001289427 .. .. GBi_ctg820001289423 .. GBi_ctg820001289418 .. .. GBi_ctg820001289456 GBi_ctg820001289460 .. 12 182 345 253 761 631 478 307 267 302 177 1 124 295 467 405 251 Length (aa) 150 221 485 107 88 77 44 234 417 859 1026 173 291 433 398 1035 257 628 323 244 222 197 279

1) 1

1, EC:1.8.4.12, EC:1.8.4. 1 family (TC 2.A.3.8.12) (IMGterm) T A component, Glc family (TC 4.A.1.2.5) om Firmicutes) A , L r r . r 1 genes (f , GntR family (IMGterm) r methylase Mod (EC:2.1.1.72) Bimp A . r flux permease f ranscriptional regulator ranscriptional regulato ransposase agatose-1,6-bisphosphate aldolase (EC:4.1.2.40) ype I restriction-modification system methyltransferase subunit (EC:2.1.1.72) ype I site-specific restriction-modification system, R (restriction) subunit and related helicases (EC:3.1.21.3) Amidases related to nicotinamidase T Fructose-2,6-bisphosphatase (EC:5.4.2.1) Methionine synthase (B12-independent) (EC 2.1.1.14) (IMGterm) Beta-glucosidase/6-phospho-beta-glucosidase/beta-galactosidase (EC:3.2.1.86) T Phosphotransferase system, mannose/fructose/N-acetylgalactosamine-specific component IIC Phosphotransferase system, mannose/fructose/N-acetylgalactosamine-specific component IID Peptide methionine sulfoxide reductase (EC:1.8.4. Nitrous oxide-stimulated promote Uncharacterized protein conserved in bacteria Pyridoxine biosynthesis enzyme T Arabinose e Uridine phosphorylase (EC:2.4.2.3) Gene Product Name Acetyltransferases (EC:2.3.1.-) Predicted membrane protein Serine/threonine exchange transporte PTS system beta-glucoside-specific II Hypothetical protein T Predicted branched-chain amino acid permease (azaleucine resistance) Predicted membrane protein Chorismate mutase (EC 5.4.99.5) (IMGterm) Restriction endonuclease S subunits (EC:3.1.21.3) T T Predicted membrane protein Putative glucose uptake permease Pyrimidine-nucleoside phosphorylase (EC 2.4.2.2) (IMGterm) Nucleoside permease Restriction endonuclease (EC:3.1.21.5) Predicted hydrolases of the HAD superfamily Adenine specific DN Uncharacterized protein conserved in bacteria Phosphotransferase system, mannose/fructose/N-acetylgalactosamine-specific component IIB (EC:2.7.1.69) Predicted HD superfamily hydrolase Predicted glutamine amidotransferase involved in pyridoxine biosynthesis Predicted hydrolases of the HAD superfamily . Gilliamella apicola st Ca 10 1 ed r ag

T

GBi_0070.00000200 GBi_0069.00000240 GBi_0068.00000060 GBi_0068.00000050 GBi_0068.00000040 GBi_0061.00000 GBi_0041.00000150 GBi_0041.00000140 GBi_0036.00000050 GBi_0041.00000130 GBi_0082.00000050 GBi_0029.00000900 GBi_0012.00000210 GBi_0012.00000200 IMG Locus GBi_0076.00000170 GBi_0076.00000160 GBi_0072.00000210 GBi_0061.00000100 GBi_0060.00000300 GBi_0054.00000030 GBi_0059.00000040 GBi_0060.00000290 GBi_0054.00000020 GBi_0052.00000220 GBi_0052.00000210 GBi_0052.00000200 GBi_0052.00000060 GBi_0051.00000020 GBi_0048.00000240 GBi_0048.00000220 GBi_0048.00000130 GBi_0048.00000120 GBi_0043.00000060 GBi_0041.00000170 GBi_0041.00000160 GBi_0082.00000030 GBi_0078.00000400 GBi_0082.00000040 GBi_0076.00000330

able 2: Putative horizontally transfer 2505925526 2505925484 2505925414 2505925413 2505925412 2505925263 2505924832 T 2505924831 2505924734 2505924830 2505925872 2505924690 2505924217 2505924216 IMG Gene Object ID 2505925735 2505925734 2505925638 2505925262 2505925222 2505925077 2505925164 2505925221 2505925076 2505925061 2505925060 2505925059 2505925044 2505925004 2505924940 2505924938 2505924929 2505924928 2505924868 2505924834 2505924833 2505925870 2505925796 2505925871 2505925751 154

Ile Ser Leu Ala Ser Glu Glu x2 Val Gly Lys Thr Pro Ala Arg Asp Met Ser Asp Ser PTS SSS APC APC APC ABC ABC ABC AGCS DAACS LIVCS trehalose-6-P mannose DAACS

SSS acetoacetyl-acp ascorbate ß-glucosides GDP-mannose lipid A T1SS precursor folates LPS T6SS molybdopterin ribulose-5-P glucose-6-P ABC Lipoproteins THF pyridoxal-5-P N-acetyl- ribose-5-P fructose-6-P D-glucosamine-6-P ABC Cell division TAT ABC Lipopolysaccharides HCO - T5SS purines IMP PRPP 3 ubiquinone peptidoglycan SEC UMP FAD, FMN, ribo!avin His pyrimidines glyceraldehyde-3-P isoprenoids Ile Thr Citrate 2-HCT Gly Thaimine ABC L-aspartate- glycerate-3-P Ser Trp semialdehyde Lys Putrescine ABC MFP Multidrug x4 Met Asn Asp Sulfate x2 MFS Arg PEP chorismate MFP Colicin V Ala Nitrate / nitrite MFS Cys ABC Antibiotics x3 Pro Tyr Phe ornithine Glu oxaloacetate Phosphate PiT Leu Val MOP Multidrug x3 Gln uroporyphorinogen III Cys lactate pyruvate 2-oxoisovalerate DMT Multidrug x3 Gly acetyl-CoA protoheme siroheme CoA Pantothenate SSS acetate glutathione formate Mo ABC ethanol Mixed Acid Peptide / Ni ABC Fermentation

Iron complex x2 ABC

Zn ABC ammonia

Mg2+, Co2+ x2 MIT

AmmA nitrate nitrite NDH I ATPase NDH II NRA FDN cyt bd FRD

Figure 1 - Metabolic reconstruction of Ca. Gilliamella str. Bimp1 (BiG). Circles next to each connecting line represent genes involved with that pathway; filled represent present genes and open represent missing genes. Amino acids are in solid boxes.

Vitamins and cofactors are in dashed boxes. 155

Ca. Gilliamella str. Bimp1 (BiG) B. impatiens SS Stomaphis longirostris FJ655515 a. b. * SS Stomaphis quercus FJ655516 Salmonella typhimurium SS Stomaphis cupressi EU348326 76 HM112050 * Aphids * Escherichia coli 86 HM112130 58 57 HM108449 Citrobacter koseri * HM108310 HM108315 * Klebsiella pneumoniae AY370191 91 * Yersinia pseudotuberculosis HM112068 * 74 HM111973 DQ837604 * Yersinia pestis 99 Ca. Gilliamella apicola DQ837605 65 * Yersinia enterocolitica AY370192 A. mellifera 90 HM108542 B. impatiens 67 * Serratia proteamaculans * HM108563 B. impatiens HM215029 B. terrestris 70 Providencia stuartii HM215031 B. lapidarius * HM215028 B. pascuorum 63 63 * Arsenophonus nasoniae HM215032 B. terrestris HM215034 B. terrestris * * Proteus mirabilis Ca. Gilliamella str. BiG B. impatiens * HM215025 B. terrestris Bombus sp. Photorhabdus luminescens Enterobacteriales DQ837611 50 Haemophilus in"uenzae EF187250 Gamma-2 * * HM108316 A. mellifera Haemophilus parain"uenzae * 51 HM107876 98 Orbus hercynius, wild boar FJ612598 94 Pasteurella multocida !y (Drosophila melanogaster) DQ980728 * ground beetle (Poecilus chalcites) EF608532 Beetle, boar, !y * Pasteurella dagmatis Pasteurellales Enterobacteriales * Actinobacillus succinogenes * 0.2 * Mannheimia succiniciproducens c. Actinobacillus pleuropneumoniae Individual SiCO gene trees * Phylogenetic Avg. bootstrap Avg. amino acid Haemophilus ducreyi Total SiCOs * pattern at “A” (%) length *

Mannheimia haemolytica Pasteurellales P “A” E 25 80 448 Vibrio cholerae BiG * P “A” 19 82 340 Vibrio !scheri Vibrionales E & BiG E Shewanella piezotolerans “A” * BiG 17 58 320 P Shewanella violacea Alteromonadales E “A” 0 – – Pseudomonas aeruginosa P & BiG

Pseudomonas entomophila Pseudomonadales other 28 – 194 0.2 amino acid substitutions /site E, Enterobacteriales; P, Pasteurellales; BiG, Ca. Gilliamella str. Bimp1

Figure 2 - (a) Phylogenetic placement of Ca. Gilliamella str. Bimp1 (BiG) as a singleton clade among five orders of Gammaproteobacteria. Maximum likelihood reconstruction inferred from 89 concatenated SiCO genes (27,452 aligned amino acid sites). (b) Location of BiG among members of the insect gut-associated clade that clusters separately from Enterobacteriales and Pasteurellales. Tree based on maximum likelihood with the 16S rRNA gene. (c) Proportions of the 89 individual SiCO gene trees that returned node “A” (on multiprotein tree) that unites Enterobacteriales, Pasteurellales, and

BiG; and sister relationships of those with node “A” supported with a >50% bootstrap value. Asterisks represent bootstrap support values of 100. 156

Table S1: Location of VgrG genes in Ca. Gilliamella apicola str. Bimp1

Gene Object ID Length Locus Tag VgrG Adjacent gene lengths (aa) Adjacent region function

2505925163 931 GBi_0059.00000030 intact 858 annexin repeat protein/insecticidal toxin 2505924397 786 GBi_0018.00000020 intact 322, 1413, 219, 219 unknown 2505924904 857 GBi_0047.00000020 intact contig break 2505924238 249 GBi_0014.00000010 partial 301, 1342, 274 unknown 2505924237 330 GBi_0013.00000150 partial contig break 2505925888 228 GBi_0084.00000080 partial 168, 111, 254, 239, 240, 241 unknown 2505925852 241 GBi_0081.00000010 partial 536, 246 unknown 2505925851 126 GBi_0080.00000090 partial contig break 2505925719 163 GBi_0076.00000010 partial contig break 2505925640 182 GBi_0073.00000010 partial contig break 2505925639 330 GBi_0072.00000220 partial contig break 2505925753 447 GBi_0077.00000010 partial contig break 2505925752 183 GBi_0076.00000340 partial contig break 2505925001 185 GBi_0049.00000270 partial none 2505924423 249 GBi_0018.00000280 partial none 2505925862 269 GBi_0081.00000110 partial 229, 280, 212, 1222 unknown 157

3 SS-Stomaphis-longirostris-FJ655515 100 SS-Stomaphis-quercus-FJ655516 Aphids SS-Stomaphis-cupressi-EU348326 20 100 HM112050 86 HM112130 57 HM108449 30 HM108310 58 HM108315 11 Ca. Gilliamella apicola AY370191 91 HM112068 A. mellifera 99 74 HM111973 48 DQ837604 65 DQ837605 90 AY370192 HM108542-Bombus 100 67 HM108563-Bombus HM215029-Koch 63 70 HM215031-Koch Ca. Gilliamella str. 63 HM215028-Koch HM215032-Koch Bombus sp. 21 HM215034-Koch 100 BiG B. impatiens HM215025-Koch 47 DQ837611-G2 50 26 EF187250-G2 Gamma-2 100 HM108316-G2 A. mellifera 100 51 HM107876-G2 94 wild-boar-FJ612598 hillsb-J21-DQ980728 Beetle, boar, !y ground-beetle-EF608532 58 Actinobacillus-delphinicola-X89377 100 Actinobacillus-delphinicola-AY362889 94 Actinobacillus-scotiae-Y09653 93 Haemophilus-in!uenzae-X87978 63 46 Mannheimia-haemolytica-AF060699 45 52 Phoconebacter-uteri-X89379 Haemophilus-ducreyi-NC_002940 71 Actinobacillus-arthritidis-AF247712 53 Pasteurella-multocida-AF294410 77 Gallibacterium-anatis-AF228001 37 34 Bisgaard-t-40-AY172732 100 Pasteurella-testudinis-L06090 100 Pasteurella-testudinis-AY362926 Bisgaard-t-14-L06086 89 100 98 Bisgaard-t-14-AY362901 Pasteurellales Bisgaard-t-32-AY172729 61 Thorsellia-anophelis-AY837748 100 Arsenophonus-nasoniae-M90801 95 Phlomobacter-fragariae-AB246669 Proteus-vulgaris-AJ233425 Morganella-morganii-AB089246 54 98 Salmonella-enterica-AF008580 Enterobacteriales 90 Escherichia-coli-X80725 100 71 Pantoea-stewartii-AJ311838 83 Serratia-marcescens-AY395011 Serratia-entomophila-AJ233427 Yersinia-pestis-AJ232232 100 Photobacterium-leiognathi-X74686 Vibrio-cholerae-X74695

0.2

Figure S1 - Full 16S rRNA gene tree with sequences from Pasteurellales,

Enterobacteriales, and the clade that includes Ca. Gilliamella str. Bimp1 (BiG).

158

Ca. Gilliamella str. Bimp1 (BiG) Enterobacteriales Pasteurellales

sucA sucB sucC sucD

Figure S2 - Example of the BiG genome’s lack of synteny with related genomes.

Shown is the sucABCD operon, which is part of a conserved cluster of TCA cycle genes.