INTEGRATION THROUGH BEHAVIOR and MORPHOLOGY in a SOCIAL INSECT by Pedr
Bacterial Symbionts at the Colony and Individual Levels: Integration through Behavior and Morphology in a Social Insect
Item Type text; Electronic Dissertation
Authors Rodrigues, Pedro A D P.
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 09/10/2021 12:27:38
Link to Item http://hdl.handle.net/10150/621295
BACTERIAL SYMBIONTS AT THE COLONY AND INDIVIDUAL LEVELS: INTEGRATION THROUGH BEHAVIOR AND MORPHOLOGY IN A SOCIAL INSECT
by
Pedro Augusto Da Pos Rodrigues
______
A Dissertation Submitted to the Faculty of the
GRADUATE INTERDISCIPLINARY PROGRAM IN ENTOMOLOGY AND INSECT SCIENCE
In Partial Fulfillment of the Requirements
For the Degree of
DOCTOR OF PHILOSOPHY
In the Graduate College
THE UNIVERSITY OF ARIZONA
2016
2
THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE
As members of the Dissertation Committee, we certify that we have read the dissertation prepared by Pedro Augusto Da Pos Rodrigues, titled BACTERIAL SYMBIONTS AT THE COLONY AND INDIVIDUAL LEVELS: INTEGRATION THROUGH BEHAVIOR AND MORPHOLOGY IN A SOCIAL INSECT and recommend that it be accepted as fulfilling the dissertation requirement for the Degree of Doctor of Philosophy.
______Date: August, 12, 2016 Diana E. Wheeler
______Date: August, 12, 2016 Judith L. Bronstein
______Date: August, 12, 2016 Goggy Davidowitz
______Date: August, 12, 2016 Kirk E. Anderson
______Date: August, 12, 2016 Scott Powell
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: August, 12, 2016 Dissertation Director: Diana E. Wheeler
3
STATEMENT BY AUTHOR
This dissertation has been submitted in partial fulfillment of the 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 an accurate acknowledgement of the 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: Pedro Augusto Da Pos Rodrigues
4
ACKNOWLEDGMENTS
I am very grateful for all the people that were involved in the building of this dissertation and in my formation. Diana Wheeler, my advisor, was very supportive and has encouraged me to adventure into new areas of expertise, which have shaped my career. It is because of Diana’s guidance that I am a more independent thinker today and more encouraged to learn and try new approaches in my research. Diana is an excellent advisor and I am honored to be her last student before her retirement. I am also grateful to Diana’s mother, Mrs. Eula Wheeler, who has greatly supported my research, including once when she offered her own car so that I could do field work at the Saguaro National Park. I am very grateful for the help I received from all my committee members: Kirk Anderson (USDA), Judie Bronstein, Goggy Davidowitz, Molly Hunter, and Scott Powell (George Washington University). They were all excellent mentors that contributed significantly to the improvement of my work. For Kirk, I am thankful for his guidance, support, and all the training I received while working part-time in his laboratory; together, our collaborations have helped me mature as a scientist and greatly advanced my understanding on the microbial ecology of social insects. Judie’s guidance made me a better ecologist, with a broader understanding of fundamental concepts, and a more sharp skeptic view of my ideas as well as others’ ideas; because of Judie, I am more interested in framing my work within big questions, and my writing has also substantially improved. Goggy gave me support and enthusiasm to the possibilities within and beyond my project, which was always motivating. Similarly to Judie, Goggy has also helped me improve my writing and frame my work for a broader audience and broader concepts. Molly’s mentorship helped me focus in more doable ideas and projects during my PhD. Molly was also very supportive and always available for whenever I needed help, from advice on academic issues to kindly letting me use her lab space and supplies for some of my projects. Finally, I am grateful for Scott’s help, which involved mentorship and direct collaborative work. Scott was the first to introduce me to the skills of baiting, collecting, and rearing Cephalotes ants. With Scott I learned a lot about the biology and ecology of these amazing ants. Scott and his wife, Bia, were also my first friends in Tucson, even before I left Brazil, and they greatly helped me adapting in my first few years in the US. All of my committee members were very helpful and generous with their time, and they provided me with numerous recommendation letters over these years, which were essential for succeeding in my applications for travel and research grants. In addition, there were tough times during my PhD, including hospitalizations, and my committee members were always very supportive, as friends would, for which I am very grateful. During my PhD several collaborators helped my work move forward. Michele C. Lanan (Deep Springs College) is responsible for the speed by which we were able to get all the work done for my first manuscript of this dissertation (Appendix A), including getting it published before my defense. Michele is extremely intelligent, hard-working and creative, and I have learned with her how to move forward motivated and enthusiastic. Michele also gave me confidence on the quality of my own work. Jacob Russell (Drexel University) and Piotr àukasik (University of Montana) introduced me to bioinformatics and were very supportive in different steps of my dissertation. With them, I also developed my second manuscript (Appendix B) of this dissertation. 5
My wife Corinne Stouthamer was always by my side, providing me with support and keeping me sane when my projects didn’t work out, when my part-time jobs where consuming most of my dissertation time, and when I took an overwhelming number of responsibilities. Beyond being always there for me, Corinne also contributed directly to the development of this dissertation, particularly Appendix C, in which she is a co-author. My family, on both the Stouthamer and on the Rodrigues sides, was always supportive and gave me confidence on my work. My family in Brazil had an enormous amount of faith in me and my work, by trusting that doing a PhD abroad was a good idea, and believing that studying microbial ecology of ants is important, even if they could not entirely understand how. The Stouthamers were always very supportive and very caring, helping me and Corinne throughout my PhD, from moving-in and moving-out of houses, to organizing our beautiful wedding. Bodil, Vishwas, Javier, Patricia Navarro, Rousel Orozco, Ming Huang, Kelly, Catherine, Norm, Daniel Silva, João Paulo, Lívia, Simeão, Mariana, Lewis, Tim, Liz, Avery, Tuan, Chan, and so many other friends, not necessarily in this order, were also very important in my formation. I am very thankful and lucky for the friends I have, who have always supported me. Finally, I want to thank NSF, which supported me in my first year via a grant to my advisor. During my PhD I was also supported by travel and research grants from the Center for Insect Sciences (CIS, University of Arizona), The Graduate and Professional Student Council (GPSC, University of Arizona), and the International Union for the Study of Social Insects (IUSSI).
6
DEDICATION
I dedicate this work to the scientists that, directly or indirectly, inspired me to get here: Paulo Oliveira (UNICAMP, Brazil), Ronaldo Zucchi (USP, Brazil), João M. F. Camargo (USP, Brazil), Edward O. Wilson (Harvard, USA), Bert Hölldobler (ASU, USA), and Diana Wheeler (UofA, USA).
7
TABLE OF CONTENTS
ABSTRACT...... 8
INTRODUCTION...... 10
FORMAT AND PRESENT STUDY...... 16
REFERENCES...... 19
APPENDIX A...... 23
APPENDIX B...... 69
APPENDIX C...... 141
APPENDIX D: PERMISSIONS ...... 188
8
Abstract
The determination of a symbiotic association as beneficial requires good assessment of the costs
and benefits involved in the maintenance and transmission of these microbes across generations.
In social insects, symbiotic associations are complex as they may involve a network of
interactions between individual and colony that result in stable associations over evolutionary
time. My goal was to investigate the roles of behavior and morphology as integrators that have
enabled the benefits of harboring gut microbes to reach both adult and growing brood in a
colony. To achieve this goal, I used turtle ants (Cephalotes), a group that has co-evolved with their gut microbes since the Eocene (Sanders et al. 2014) and that shows a variety of morphological and behavioral specializations likely connected to this symbiotic association. In my dissertation I present evidence that the specialized behavior and morphology of Cephalotes are indeed strongly associated with mechanisms that ensure stability of ant-gut microbe interactions over evolutionary time. In Appendix A, I show that a valve between the crop and midgut (proventriculus) of C. rohweri works as a filtration organ, capable of excluding possible pathogens from the mostly liquid diet consumed by turtle ants. In addition, the proventricular filter is also associated with the structuring of the gut microbiota, dividing it in at least two great groups: one upstream and another downstream of the proventriculus. Through behavioral observation and microscopy, we also suggest that the formation of the proventricular filter is only complete after young and sterile workers (callows) are inoculated with the core group of symbiotic bacteria. In Appendix B, I present results confirming that the compartmentalization of gut microbiota is also present in the congener C. varians. I compare these results with previously published data, defining the meta-communities of the gut microbiota, and demonstrate that the 9
previously recognized core microbiota is composed of compartment-specific microbial
communities and lineages. This compartmentalization of the gut microbiota is similar to the one
found in highly specialized herbivores, both vertebrates and invertebrates. In addition, I also
sampled the infrabuccal pocket, a characteristic oral cavity found in ants and that has largely
been ignored in studies of gut symbiosis. Based on my results, I provide compelling evidence
that hindgut microbes are inoculated into food particles trapped in the infrabuccal pocket, aiding
in digestion of this substrate. Moreover, I suggest that trophallaxis olays a central role in
inoculation of food and individuals, and might be responsible for the transmission of nutrients
that are predicted to result from the gut bacteria metabolism. Finally, in Appendix C I
characterize abdominal trophallaxis in C. rohweri to gain insight on its role in the context of symbiotic associations with gut microbes. I show that the hindgut contents, including bacteria, can be transmitted via abdominal trophallaxis. This interaction is found to occur between all combinations of major and minor workers, in addition to callows. The rate of solicitation of abdominal trophallaxis is higher when individuals are protein starved, indicating that hindgut content may also be nutritive. Using shotgun metagenomic data, we show that the microbiota present in the infrabuccal pocket (mostly hindgut bacteria) are indeed capable of re-utilizing nitrogen and synthesizing essential amino acids, in addition to breaking down plant material. We also report that oral trophallaxis is a possible route for transmission of crop-specific bacteria for callows, as this group has performed oral trophallaxis at a relatively higher rate than older workers. Put together, these results highlight the importance of nestmate interactions and gut morphology in the establishment and maintenance of symbiotic microbes in a social insect, 10
introducing a new model for explaining the evolution and functioning of ant-gut microbe symbiosis.
Introduction
1. Explanation of the problem and its context
Symbiosis, from pathogenic to beneficial associations, has shaped life, profoundly influencing the evolution of most extant species. In social animals, such as ants, a diversity of microbes has been found inhabiting the digestive tract, indicating a possible dependence upon microbial metabolism for acquisition of nutrients. In the sections below I briefly introduce the concept of symbiosis, focusing specifically on one of the most ubiquitous associations: animal gut associated microbes. In particular, I explore the complexity of harboring microbes in a social animal in light of the central question of this dissertation: how does behavior and morphology contribute to maintenance and acquisition of microbes, at both the individual and colony levels?
At the end, I introduce my focal study organism, species of Cephalotes and summarize the three studies contained in this dissertation.
Gut Symbiosis
Symbiosis, the living together of different species and in persistent contact, has shaped the evolution of many single and multi-cellular organisms and is one of major driving forces of evolutionary novelty and radiation (Sapp 1994; Kawecki 1998; Bronstein 2015; Douglas 2015;
Schneider et al. 2016). Outcomes of symbiotic associations vary from beneficial to detrimental 11
for one or both parts of the association. Historically, pathogenic interactions have been the best
studied type of symbiosis, leading to important discoveries such as the key role of parasites in
the evolution of sexual reproduction and generation of biodiversity (Hamilton et al. 1990;
Kawecki 1998; Morran et al. 2011). More recently, growing evidence shows that beneficial symbiosis might also represent a fundamental piece in the evolution of life, with the most iconic example being the ancient association between two prokaryotic organisms that led to the origin of all eukaryotes (Andersson & Kurland 1999).
The most common symbiotic association in animals is likely found between microbes and the digestive tract (Dillon & Dillon 2004; Ley et al. 2008). In recent years many unrecognized host- gut microbe associations have been discovered, including in humans. The rising interest in gut symbiosis is mostly due to the recent development and popularization of molecular biology tools, such as next-generation sequencing. By providing increased access to the identity of unculturable microbes and their metabolic properties, these tools have allowed the questioning of an older view regarding host-gut bacteria interactions: that most gut microbes are commensals (Shapira
2016). Evidence shows that gut symbionts can aid with digestion, recycle nitrogen waste, increase immunity and even affect host behavior (reviewed in Engel and Moran (2013)) . Despite the discovery of many beneficial microbes, the understanding of the nature of these associations is still far from complete: symbiotic microbes are often described only in terms of identity and metabolism, but disconnected from hypotheses for the net benefits of the association, the biology of the host and the symbiont, the environmental context, costs involved in the association, and the mechanisms that explain persistence over both host life cycle and evolutionary time.
12
Gut symbiosis in ants
Ants are among the most abundant and diverse of all extant animals (Hölldobler & Wilson
1990). Present in nearly all terrestrial habitats, ants strongly influence the functioning and maintenance of ecosystems (Folgarait 1998). Symbiosis between ants and gut microorganisms has been suggested as a key variable for understanding ant adaptations for exploring new niches
(Zientz et al. 2005). For instance, many arboreal ants are considered to consume a mostly herbivorous diet and are often found to associate with specific clades of gut microbes, in particular Rhizobiales (Russell et al. 2009). Arboreal ants are considered to exhibit a specific type of herbivory (sometimes referred as “cryptic herbivory”, Hunt (2003), or “functional herbivory” (Anderson et al. 2012) ) in which they do not consume plant tissues directly, but rather secretions such as extra-floral nectar (EFN) and honeydew from sap-feeding hemipterans
(Cook & Davidson 2006). Because this diet is deficient in important nutrients such as amino acids (Blüthgen et al. 2004; Wäckers & Wäckers 2005), arboreal ants are believed to depend on their gut microbes to complement their diet. Carpenter ants (genus Camponotus) for instance harbors an intracellular bacterium (Blochmmania) in their midgut tissues that has been shown to be able to recycle urea, break down uric acid (from bird droppings that the host may occasionally consume), and synthesize essential amino acids (Feldhaar et al. 2007). Associations with intracellular symbionts are found across the Camponotini tribe, but is otherwise rare among ants, having been only reported in one other genus, Cardiocondyla (Klein et al. 2016). On the other hand, communities of extracellular bacteria that inhabit the lumen of the digestive tract of ants are widely found among arboreal ants (Russell et al. 2009; Brown & Wernegreen 2016).
Whether or not these microbial communities are beneficial to ants has not been established yet, 13
but they are assumed to play digestive roles similar to the ones found in Blochmannia, in addition to possibly fixing nitrogen (Russell et al. 2009).
On establishing extracellular symbionts as beneficial symbionts
Carrying beneficial extracellular bacteria in the digestive tract may require specific mechanisms that ensure successful maintenance and transmission of microbes across generations. In insects it is hypothesized that communities of extracellular microbes must be able to survive out of the host body during transmission to other individuals, i.e. on the surface of eggs, in fecal material
(coprophagy), or in “symbiont-enclosing capsules” deposited among eggs (revised in Salem et al. (2015)). Social behavior adds another layer of complexity as nestmates may need to be re- infected at different life stages (Nalepa et al. 2001), and additional mechanisms for infection and maintenance of symbionts must ensure that beneficial microbes are not lost when individual hosts leave their mother colony to found new colonies. Transmission via coprophagy has been demonstrated in Apidae (Koch & Schmid-Hempel 2011; Martinson et al. 2012), and Isoptera
(Nalepa et al. 2001). Nonetheless, it has been recognized that the exchange of liquids among individuals via trophallaxis might represent a more specialized route of transmission (Powell et al. 2014; Nalepa 2015). In particular, proctodeal trophallaxis, i.e., the direct consumption of rectal fluids via oral-proctodeum contact, is considered a derived state of coprophagy (Nalepa
2015), and may significantly diminish the environmental stress that gut microbes experience during transfer between hosts. In addition to mechanisms for transmission across generations, maintenance of beneficial microbes may involve protection against ingestion of pathogenic or more competitive microorganisms that can cause dysbiosis (Nelson et al. 2012; Jones et al. 2013; 14
Kautz et al. 2013; Cariveau et al. 2014). Morphological and physiological specializations may also improve survival and resilience of symbiotic microbes in the gut environment (Engel &
Moran 2013; Kwong & Moran 2015; Donaldson et al. 2016). Finally, the maintenance of
beneficial symbiosis of gut microbes in a social animal involves a fundamental difference in
relation to solitary animals: the benefits of living in symbiosis must reach beyond the individual,
by improving colony health, growth and reproduction (Anderson et al. 2011; Nalepa 2015).
The features described above for animal gut-microbe symbiosis, particularly in social insects
such as ants, illustrate the complexity of these interactions and highlights the importance of
understanding the host biology in establishing the nature of these associations. In particular,
understanding the network of interactions intermediated by morphology and behavior seems to
be fundamental for the generation of hypotheses regarding the function of gut symbionts as well
as the establishment of rules that can be generalized across biological systems.
Cephalotes and gut symbionts: integration through morphology and behavior
The tribe Cephalotini is a diverse group of ants found in the Neotropics. It includes
approximately 185 species, present in deserts, rainforests and savannah (De Andrade & Urbani
1999). Represented by two genera, Cephalotes and Procryptocerus, cephalotines are known to harbor a community of bacteria and fungus in their digestive tract (Wheeler 1984; Caetano & da
Cruz-Landim 1985; Roche & Wheeler 1997; Bution & Caetano 2008). The roles of these symbionts have not yet been determined, but it has been assumed that they function as nutritional mutualists (Russell et al. 2009; Anderson et al. 2012). Cephalotini share with other arboreal ants the signature of functional herbivory, as revealed by isotopic analyses (Davidson et al. 2003; 15
Cook & Davidson 2006; Russell et al. 2009). In addition, similarly to other arboreal ants such as
Pseudomymicinae and Dolichoderinae, Cephalotini ants are known to have a specialized pouch-
like ileum, where a mass of microbes is found (Cook & Davidson 2006; Stoll et al. 2007).
Cephalotini ants are also known to perform abdominal trophallaxis, a behavior similar to
proctodeal trophallaxis, believed to be the route for vertical transmission of their gut symbionts
(Wilson 1976; Wheeler 1984). Abdominal trophallaxis is also found in other exudate-feeding
ants, such as Camponotus (Santos et al. 2005) and Dolichoderus (Cook & Davidson 2006), but as in Cephalotini, the determination of its role requires further investigation. Besides specializations in the ileum, Cephalotini also present an unusually shaped proventriculus, a valve between the crop and the midgut (Eisner 1957). The proventriculus is present in all ants, varying in size and shapes (Eisner 1957), but whether the proventriculus interacts with the diet of ants and with the gut associated microbes is not clear in most ants. Finally, ants, including Cephalotes are considered to be specialized liquid feeders (Hölldobler & Wilson 1990), but it has long been known that solid food is frequently trapped in a specialized oral cavity in the head of ants, the infrabuccal pocket (Wheeler & Bailey 1920; Eisner & Happ 1962; Urbani & de Andrade 1997).
Fungus and pollen are among the content found in the pellets formed in this cavity (Wheeler &
Bailey 1920; Urbani & de Andrade 1997; Mankowski & Morrell 2004), which may serve as food to larvae (Wilson 1976; Cole 1980; Jouvenaz et al. 1984; Blatrix et al. 2012). Despite its possible role in nutrition, and its putative contact with fluids exchanged via oral and abdominal trophallaxis, the infrabuccal pocket has been largely ignored in studies of ant-gut microbe symbiosis. 16
In Cephalotes, the most diverse genus within Cephalotini, a core community of bacteria is consistently found across different species that has co-diversified with their hosts, possibly since the Eocene (Sanders et al. 2014). In this dissertation I use species of Cephalotes as model organisms to investigate how social insects have adapted to a symbiotic lifestyle with gut microbes. Specifically, I investigate how specialized morphology of ants (gut compartmentalization, proventriculus, and infrabuccal pocket), in association with behavior
(grooming, trophallaxis) contribute to the maintenance and transmission of microbes among nestmates and within colonies.
2. Format and Present Study
This dissertation contains three manuscripts included as appendices. Through these manuscripts I present my findings from a comprehensive investigation on the roles of morphology and behavior of ants in integrating their gut microbiome at both the individual and colony level.
Below I give a brief description of the main questions and results from each of these appendices.
Appendix A: “A bacterial filter protects and structures the gut microbiome of an insect”
Cephalotes evolved in close association with a core community of extracellular microbes found throughout their digestive tract. Whereas intracellular symbionts might be protected inside specialized host cells, the core microbiota of Cephalotes is constantly exposed to the influx of food and environmental microbes that may result in dysbiosis and loss of beneficial microbes.
Here we explored the role of the proventriculus as a morphological filter that can protect the gut microbiota from food-borne pathogens. By using a combination of DNA sequencing, microscopy, and behavioral assays, we found that in Cephalotes rohweri the proventriculus is 17
capable of filtrating particles as small as 0.2uM, the size of some of the smallest bacteria ever
described. We also found that the gut microbiota in healthy individuals is compartmentalized,
with a significant division in the structure and composition of communities found upstream of
the proventriculus and downstream of the proventriculus, which is also consistent with this
organ’s filtration capability. Finally, we investigated how new, sterile adults are inoculated with
core bacteria, since the filtering ability of the proventriculus could prevent microbes from
reaching compartments such as the midgut and hindgut. We suggest that the filtering capacity of
the proventriculus might be subject to the formation of a sealing layer on top of this organ, which
was observed to happen after individuals engage in abdominal trophallaxis.
Appendix B: “Microbiome partitioning in turtle ants”
One of the main findings from Appendix A was the fact that the core microbiota associated with
Cephalotes is not uniformly distributed, or at least not in C. rohweri. The partitioning of microbial communities in different gut compartments is frequently found in highly specialized herbivores, such as ruminants, hemipterans and termites. To explore further the significance and generality of partitioning in Cephalotes in relation to their diet and social lifestyle, we sequenced the microbial communities of a congener, C. varians. In our sampling we also included the infrabuccal pocket, to test whether the food particles in this compartment also show a specific community of microbes. In addition to the taxonomic profile of bacterial communities partitioned along the gut, we were also interested in testing whether the functional profile of gut communities also varied depending on the gut compartment identity. For this goal, we used
PICRUST, a bioinformatics tool for inferring function of bacteria based on phylogenetic proximity to bacterial groups where full genomic information is available. To verify the 18
generality of our findings, we included previous published data on C. rohweri (Appendix A) and a smaller study in C. varians (Kautz et al. 2013). Since these studies sampled different regions of the 16S rRNA gene, direct comparison among studies is difficult. To allow for more accurate comparisons, we also developed and implemented a curated reference database that significantly improves the bacterial taxonomy, across different studies involving the Cephalotes microbiota.
Using these tools, our results confirm the existence of a consistent pattern of compartmentalization of the gut microbiota, separating at least three meta-communities: a crop- specific community, a midgut-specific community, and a hindgut-specific community. We also found that the infrabuccal pocket of workers harbors a mixture of crop and hindgut bacteria, indicating a novel role of trophallaxis in inoculating food particles with gut bacteria. Using functional profiles for each gut compartment, we introduce a new model for the network of interactions between microbes, host, and colony. In particular, we suggest that both oral and abdominal trophallaxis might be fundamental to linking gut microbes to nutrients that are transferred to growing larvae in the colony.
Appendix C: “Transmission of beneficial microbes mediated by behavior in a social insect”
Our results in both Appendix A and B point to a central role of trophallaxis, in particular abdominal trophallaxis, in both maintaining and transmitting gut microbes in Cephalotes. In this study we characterize abdominal trophallaxis and test its role in the nutrition of the colony.
Through behavioral observations, we found that occurrences of abdominal trophallaxis are less frequent than oral trophallaxis, but may be performed by both minor and major workers, in addition to young workers that have just recently emerged from pupation (callows). We found that abdominal trophallaxis is particularly long in duration when performed by callows within 19
their first few hours post-eclosion, but this behavior is more frequent and shorter in duration
when performed by older adults. Our results also indicate that oral trophallaxis is performed at a
higher rate by callows, which suggests that crop-specific microbes might be transmitted via a
different route compared to midgut and hindgut microbes acquired via abdominal trophallaxis.
We present compelling evidence that the source of liquid transferred via abdominal trophallaxis
is the hindgut and that the frequency of this behavior is increased when the colony is protein-
starved. Finally, through a shotgun metagenome analysis we found that the microbiota found in
the infrabuccal pellets of C. rohweri is similar to the hindgut bacteria and is able to digest plant material, recycle nitrogen and synthesize essential amino acids. These results support the scenario hypothesized in Appendix B, i.e., that the gut microbiota might benefit the colony by providing nutrients that can be transferred to the growing brood via oral trophallaxis and regurgitation of infrabuccal pellets. Together, these results in C. rohweri and previous results in
C. varians (Appendix B) strengthen the hypothesis on the central role of behavior and morphology in enabling gut symbionts to benefit the individual and the colony.
3. References
Anderson K.E., Russell J.A., Moreau C.S., Kautz S., Sullam K.E., Hu Y., Basinger U., Mott B.M., Buck N. & Wheeler D.E. (2012). Highly similar microbial communities are shared among related and trophically similar ant species. Mol. Ecol., 21, 2282-96. Anderson K.E., Sheehan T.H., Eckholm B.J., Mott B.M. & DeGrandi-Hoffman G. (2011). An emerging paradigm of colony health: microbial balance of the honey bee and hive (Apis mellifera). Insectes Soc., 58, 431-444. Andersson S.G.E. & Kurland C.G. (1999). Origins of mitochondria and hydrogenosomes. Curr. Opin. Microbiol., 2, 535-541. Blatrix R., Djiéto-Lordon C., Mondolot L., La Fisca P., Voglmayr H. & McKey D. (2012). Plant- ants use symbiotic fungi as a food source: new insight into the nutritional ecology of ant– plant interactions. Proc. R. Soc. Lond., Ser. B: Biol. Sci., 279, 3940-3947. 20
Blüthgen N., Gottsberger G. & Fiedler K. (2004). Sugar and amino acid composition of ant- attended nectar and honeydew sources from an Australian rainforest. Austral Ecol., 29, 418-429. Bronstein J.L. (2015). The study of mutualism. In: Mutualism. Oxford University Press Oxford. Brown B.P. & Wernegreen J.J. (2016). Deep divergence and rapid evolutionary rates in gut- associated Acetobacteraceae of ants. Bmc Microbiol, 16, 1-17. Bution M.L. & Caetano F.H. (2008). Ileum of the Cephalotes ants: A specialized structure to harbor symbionts microorganisms. Micron, 39, 897-909. Caetano F. & da Cruz-Landim C. (1985). Presence of microorganisms in the alimentary canal of ants of the tribe Cephalotini (Myrmicinae: Location and relationship with intestinal structures). Naturalia, 10, 37-47. Cariveau D.P., Elijah Powell J., Koch H., Winfree R. & Moran N.A. (2014). Variation in gut microbial communities and its association with pathogen infection in wild bumble bees (Bombus). Isme J, 8, 2369-2379. Cole B.J. (1980). Repertoire convergence in two Mangrove ants, Zacryptocerus varians and Camponotus (Colobopsis) sp. Insectes Soc., 27, 265-275. Cook S.C. & Davidson D.W. (2006). Nutritional and functional biology of exudate-feeding ants. Entomol. Exp. Appl., 118, 1-10. Davidson D.W., Cook S.C., Snelling R.R. & Chua T.H. (2003). Explaining the abundance of ants in lowland tropical rainforest canopies. Science, 300, 969-972. De Andrade M.L. & Urbani C.B. (1999). Diversity and adaptation in the ant genus Cephalotes, past and present (Hymenoptera, Formicidae). Stuttgarter Beiträge Naturkunde Ser. B (Geol. Paläontol.). Dillon R.J. & Dillon V.M. (2004). The gut bacteria of insects: nonpathogenic interactions. Annu Rev Entomol, 49, 71-92. Donaldson G.P., Lee S.M. & Mazmanian S.K. (2016). Gut biogeography of the bacterial microbiota. Nature reviews. Microbiology, 14, 20-32. Douglas A.E. (2015). The special case of symbioses: mutualisms with persistent contact. In: Mutualism. Oxford University Press Oxford. Eisner T. (1957). A comparative morphological study of the proventriculus of ants (Hymenoptera: Formicidae). Bull. Museum Comp. Zool., 116. Eisner T. & Happ G.M. (1962). The infrabuccal pocket of a formicine ant: a social filtration device. Psyche, 69, 107-116. Engel P. & Moran N.A. (2013). The gut microbiota of insects - diversity in structure and function. FEMS microbiology reviews, 37, 699-735. Feldhaar H., Straka J., Krischke M., Berthold K., Stoll S., Mueller M.J. & Gross R. (2007). Nutritional upgrading for omnivorous carpenter ants by the endosymbiont Blochmannia. BMC biology, 5, 48. Folgarait P.J. (1998). Ant biodiversity and its relationship to ecosystem functioning: a review. Biodiversity & Conservation, 7, 1221-1244. Hamilton W.D., Axelrod R. & Tanese R. (1990). Sexual reproduction as an adaptation to resist parasites (a review). Proc. Natl. Acad. Sci. USA, 87, 3566-3573. Hölldobler B. & Wilson E.O. (1990). The ants. Harvard University Press, Cambridge, Mass. Hunt J.H. (2003). Cryptic herbivores of the rainforest canopy. Science, 300, 916-917. 21
Jones R.T., Vetter S.M., Montenieiri J., Holmes J., Bernhardt S.A. & Gage K.L. (2013). Yersinia pestis infection and laboratory conditions alter flea-associated bacterial communities. Isme J, 7, 224-228. Jouvenaz D.P., Ellis E.A. & Lofgren C.S. (1984). Histopathology of the tropical fire ant, Solenopsis geminata, infected with Burenella dimorpha (Microspora: Microsporida). J Invertebr Pathol, 43, 324-332. Kautz S., Rubin B.E., Russell J.A. & Moreau C.S. (2013). Surveying the microbiome of ants: comparing 454 pyrosequencing with traditional methods to uncover bacterial diversity. Appl Environ Microbiol, 79, 525-34. Kawecki T.J. (1998). Red queen meets Santa Rosalia: arms races and the evolution of host specialization in organisms with parasitic lifestyles. The Am. Naturalist, 152, 635-651. Klein A., Schrader L., Gil R., Manzano-Marin A., Florez L., Wheeler D., Werren J.H., Latorre A., Heinze J., Kaltenpoth M., Moya A. & Oettler J. (2016). A novel intracellular mutualistic bacterium in the invasive ant Cardiocondyla obscurior. Isme J, 10, 376-88. Koch H. & Schmid-Hempel P. (2011). Socially transmitted gut microbiota protect bumble bees against an intestinal parasite. Proc. Natl. Acad. Sci. USA, 108, 19288-19292. Kwong W.K. & Moran N.A. (2015). Evolution of host specialization in gut microbes: the bee gut as a model. Gut microbes, 6, 214-20. Ley R.E., Lozupone C.A., Hamady M., Knight R. & Gordon J.I. (2008). Worlds within worlds: evolution of the vertebrate gut microbiota. Nat Rev Micro, 6, 776-788. Mankowski M.E. & Morrell J.J. (2004). Yeasts associated with the infrabuccal pocket and colonies of the carpenter ant Camponotus vicinus. Mycologia, 96, 226-231. Martinson V.G., Moy J. & Moran N.A. (2012). Establishment of characteristic gut bacteria during development of the honeybee worker. Appl Environ Microbiol, 78. Morran L.T., Schmidt O.G., Gelarden I.A., Parrish R.C. & Lively C.M. (2011). Running with the Red Queen: host-parasite coevolution selects for biparental sex. Science, 333, 216-218. Nalepa C.A. (2015). Origin of termite eusociality: trophallaxis integrates the social, nutritional, and microbial environments. Ecol. Entomol., 40, 323-335. Nalepa C.A., Bignell D.E. & Bandi C. (2001). Detritivory, coprophagy, and the evolution of digestive mutualisms in Dictyoptera. Insectes Soc., 48, 194-201. Nelson A.M., Walk S.T., Taube S., Taniuchi M., Houpt E.R., Wobus C.E. & Young V.B. (2012). Disruption of the human gut microbiota following Norovirus infection. Plos One, 7, e48224. Powell J.E., Martinson V.G., Urban-Mead K. & Moran N.A. (2014). Routes of acquisition of the gut microbiota of the honey bee Apis mellifera. Appl Environ Microbiol, 80, 7378-87. Roche R.K. & Wheeler D.E. (1997). Morphological specializations of the digestive tract of Zacryptocerus rohweri (Hymenoptera: Formicidae). J. of Morphol., 234, 253-262. Russell J.A., Moreau C.S., Goldman-Huertas B., Fujiwara M., Lohman D.J. & Pierce N.E. (2009). Bacterial gut symbionts are tightly linked with the evolution of herbivory in ants. Proc Natl Acad Sci U S A, 106, 21236-41. Salem H., Florez L., Gerardo N. & Kaltenpoth M. (2015). An out-of-body experience: the extracellular dimension for the transmission of mutualistic bacteria in insects. Proc. R. Soc. Lond., Ser. B: Biol. Sci., 282. 22
Sanders J.G., Powell S., Kronauer D.J., Vasconcelos H.L., Frederickson M.E. & Pierce N.E. (2014). Stability and phylogenetic correlation in gut microbiota: lessons from ants and apes. Mol. Ecol., 23, 1268-83. Santos J.C., Yamamoto M., Oliveira F.R. & Del-Claro K. (2005). Behavioral repertory of the weaver ant Camponotus (Myrmobrachys) senex (Hymenoptera: Formicidae). Sociobiology, 45, 1-11. Sapp J. (1994). Evolution by Association. Oxford University Press, USA. Schneider K., Resl P. & Spribille T. (2016). Escape from the cryptic species trap: lichen evolution on both sides of a cyanobacterial acquisition event. Mol. Ecol., 25, 3453-3468. Shapira M. (2016). Gut microbiotas and host evolution: scaling up symbiosis. Trends Ecol. Evol., 31, 539-549. Stoll S., Gadau J., Gross R.O.Y. & Feldhaar H. (2007). Bacterial microbiota associated with ants of the genus Tetraponera. Biol. J. Linn. Soc., 90, 399-412. Urbani B.C. & de Andrade L.M. (1997). Pollen eating, storing, and spitting by ants. Naturwissenschaften, 84, 256-258. Wäckers F.L. & Wäckers F.L. (2005). Suitability of (extra-)floral nectar, pollen, and honeydew as insect food sources. In: Plant-provided food for carnivorous insects: a protective mutualism and its applications (eds. Wäckers FL, van Rijn PCJ & Bruin J). Cambridge University Press Cambridge, UK, pp. 17-74. Wheeler D.E. (1984). Behavior of the ant, Procryptocerus scabriusculus (Hymenoptera: Formicidae), with comparisons to other cephalotines. Psyche, 91, 171-192. Wheeler W.M. & Bailey I.W. (1920). The feeding habits of Pseudomyrmine and other ants. T. Am. Philos. Soc., 22, 235-279. Wilson E.O. (1976). A social ethogram of the neotropical arboreal ant Zacryptocerus varians (Fr. Smith). Anim. Behav., 24, 354-363. Zientz E., Feldhaar H., Stoll S. & Gross R. (2005). Insights into the microbial world associated with ants. Arch. Microbiol., 184, 199-206.
23
APPENDIX A: A BACTERIAL FILTER PROTECTS AND STRUCTURES THE GUT MICROBIOME OF AN INSECT
Published: ISME Journal (2016) 10: 1866-1876
24
A bacterial filter protects and structures the gut microbiome of an insect
Michele Caroline Lanan 1*, Pedro Augusto Pos Rodrigues2*, Al Agellon3 , Patricia Jansma4 ,
Diana Esther Wheeler5
* These authors contributed equally to this work, sharing first co-authorship
1Chair of the Natural Sciences, Deep Springs College, Big Pine, CA 93513, USA.
2Graduate Interdisciplinary Program in Entomology and Insect Science, University of Arizona,
Tucson, AZ 85721, USA.
3School of Animal and Comparative Biomedical Sciences, University of Arizona, Tucson AZ
85721, USA.
4Department of Neuroscience, University of Arizona, Tucson, AZ 85721, USA. 5Department of
Entomology, University of Arizona, Tucson, AZ 85721, USA.
Keywords: microbiome, coevolution, eusociality, morphology, gut, alimentary canal, partner fidelity, symbiosis, Cephalotes, ants
ABSTRACT
Associations with symbionts within the gut lumen of hosts are particularly prone to disruption due to the constant influx of ingested food and non-symbiotic microbes, yet we know little about how partner fidelity is maintained. Here we describe for the first time the existence of a gut morphological filter capable of protecting an animal gut microbiome from disruption. The proventriculus, a valve located between the crop and midgut of insects, functions as a micro-pore 25
filter in the Sonoran Desert turtle ant (Cephalotes rohweri), blocking the entry of bacteria and particles >0.2µm into the midgut and hindgut while allowing passage of dissolved nutrients.
Initial establishment of symbiotic gut bacteria occurs within the first few hours after pupation via oral-rectal trophallaxis, before the proventricular filter develops. Cephalotes ants are remarkable for having maintained a consistent core gut microbiome over evolutionary time (Sanders et al.,
2014) and this partner fidelity is likely enabled by the proventricular filtering mechanism. In addition, the structure and function of the cephalotine proventriculus offers a new perspective on organismal resistance to pathogenic microbes, structuring of gut microbial communities, and development and maintenance of host-microbe fidelity both during the animal life cycle and over evolutionary time.
INTRODUCTION
Nutritional mutualisms between animals and microbes are widespread (Backhed et al., 2005,
Hongoh, 2011), often taking place in the alimentary canal where microbes can play an important role in food digestion (McFall-Ngai et al., 2013). Maintenance of fidelity between mutualistic partners seems straightforward for obligate endosymbionts that are cosseted inside special cells or organs associated with the gut (Moran et al., 2008). In contrast, microbes in the lumen of the one-way bilaterian gut generally face a downstream flow of ingested content that may flush away resident bacteria (e.g. Nyholm and MgFall-Ngai, 2004; Blum et al., 2013) or introduce non- symbiotic and pathogenic microorganisms that can be harmful for both the host and the resident bacterial community (e.g. Nelson et al., 2012; Jones et al., 2013; Cariveau et al. 2014). With 26
this consistent downstream flow of ingested food and microorganisms, how does the host maintain partner fidelity with its beneficial microbes?
In general, the composition of gut microbiomes is known to be structured through diet (Muegge et al., 2011), gut physiology (Kwong and Moran, 2015) and compartmentalization (Engel and
Moran, 2013), avoidance of parasites through hygienic behavior (Cremer and Sixt, 2009), physical barriers (e.g. peritrophic matrix, (Hegedus et al., 2009)), and innate immune systems
(Nyholm and Graf, 2012). Here we report on a novel means of host manipulation of gut microbiota: an anatomical filter capable of protecting the host and their microbiota from disturbance by non-symbiotic microbes and likely involved in promoting high specificity between host and symbiotic microbiota over evolutionary time.
Turtle ants in the genus Cephalotes (118 species) consume a mostly herbivorous diet (Russell et al., 2009) supplemented by pollen, bird feces, and vertebrate urine (Powell, 2008, Baroni Urbani and de Andrade, 1997). Cephalotes hosts 16 to 20 core bacterial strains (Sanders et al., 2014, Hu et al., 2014) that are present in large numbers in the midgut and hindgut lumen (Roche and
Wheeler, 1997) and likely play an important role in host nutrition (Russell et al., 2009, Anderson et al., 2012). Microbial communities in Cephalotes are highly similar among nestmates, within and between species (Hu et al., 2014), and have codiversified with their hosts indicating a history of vertical transmission at the colony level (Sanders et al., 2014). Such vertical transmission of microbes is usually associated with obligate intracellular symbionts (Moran et al., 2008, Engel and Moran, 2013) rather than inhabitants of the digestive tract, where horizontal acquisition of microbes with ingested food is frequently observed (Pernice et al., 2014). This level of core gut 27
microbiota stability over evolutionary time is quite unusual, and the mechanism enabling this
stability is unclear (Sanders et al., 2014, Hu et al., 2014).
We investigated the function of the proventriculus, a valve in the gut of Cephalotes rohweri, to determine whether it could serve as the mechanism for maintaining gut microbiota partner fidelity. The proventriculus separates the crop and midgut in insects and exhibits a remarkably divergent shape in Cephalotes ants, the function of which has been speculated upon (Baroni
Urbani and de Andrade, 1997, Roche and Wheeler, 1997) but remains unknown. A symbiont sorting mechanism has been recently described in the bean bug Riptortus pedestris – a gut constriction that is only permeable to the passage of their gut symbiont, an environmentally acquired bacterium in the genus Burkholderia (Ohbayashi et al., 2015). Unlike R. pedestris,
Cephalotes gut symbionts are maternally inherited and consistently found across different species of Cephalotes. A second distinction is that while Burkholderia is hosted in a special, isolated portion of the gut in R. pedestris, the symbiont community of turtle ants is found in the lumen throughout the alimentary canal and in constant exposure to ingested food. In host-gut symbiont systems, ingested, non-symbiotic microorganisms may cause detrimental changes in the community structure of the resident bacteria, resulting in lower immune response and development of diseases in the host (Sansonetti, 2004; Stecher et al., 2013). The Cephalotes proventriculus has previously been demonstrated to filter larger solid particles (>12µm, (Roche and Wheeler, 1997)) and we hypothesize that it may also prevent particles as small as bacteria from transiting the gut. To test this hypothesis, we investigated the morphology of the proventriculus, its porosity, and its association with the microbiome structure within the alimentary canal of Cephalotes. 28
MATERIALS AND METHODS
Colonies used for experiments
Cephalotes rohweri colonies were collected from Sonoran Desert scrub habitat at Tucson
Mountain Park, Tucson AZ, USA (permit from Pima County Natural Resources, Parks and
Recreation) in the spring and summer of 2012 and 2013. Colonies live in cavities bored by beetles into the branches of the tree Cercidium microphylla. During warm months, workers forage on the branches and leaves and rarely visit the ground. The average distance between collected colonies was 0.5 miles and trees were spaced widely in the habitat, ensuring that we used separate colonies. Colonies were maintained in the laboratory in nests consisting of a wood cutout between two sheets of plexiglass inside a Fluon-painted acrylic box, reared at approximately 25°C, provided with water in cotton-stoppered test tubes, and fed a diet of freeze- killed, bisected cockroaches and 20% honey water.
Imaging
All specimens used for scanning electron microscopy were dissected from freshly killed ants and dehydrated in an ethanol series followed by critical point drying. We then microdissected the dry specimens and mounted them on stubs with conductive carbon adhesive tape, followed by sputter coating with platinum. All samples were imaged with a Hitachi S-4800 Type II Field Emission
SEM (Hitachi High Technologies America, Inc., Pleasanton, CA, USA). The specimen in Fig 1C was treated with 10% KOH for 24 h, and then washed in Milli-Q water for 24 h prior to dehydration in order to remove all tissue except the cuticular layer comprising the upper surface of the proventriculus and wall of the crop. 29
Gut specimens for TEM imaging were dissected from ants and fixed in 4% formaldehyde, 0.5% glutaraldehyde in 0.1M phosphate buffer (pH 7.0) for 8 hours at 4°C, rinsed in buffer, and fixed in 2% osmium tetroxide for 30 minutes. Specimens were again rinsed and then dehydrated in an ethanol series followed by acetone, infiltrated in a series of Spurr’s resin/acetone mixtures before being embedded in Spurr’s. Ultrathin sections were cut with a diamond blade and ultramicrotome, placed on formvar film grids, stained with uranyl acetate and lead citrate, and imaged with a Philips CM-12 TEM (FEI, Hillsboro, OR, USA). Light microscopy specimens were embedded as above for TEM, but with chlorazol black and 1% methylene blue as stains. Images in Fig 1B-1E and Fig 4A-4D were colorized in Adobe
Photoshop CS6 to indicate the different parts of the structure. The illustration in Fig 1A was drawn by MCL based on SEM, TEM, confocal, and light microscopy data. Original, uncolored micrographs are provided for all data in Fig S1, Fig S2, and Fig S3.
Sequencing and analysis of gut microbiome
We sequenced ants from field-collected colonies, but kept in laboratory conditions for two months (colonies 1a, 6a, and 7a) in addition to colonies freshly collected from the field (colonies
1b, 2b, 3b, and 4b). DNA was extracted from each of the following compartments of the alimentary canal (“gut compartments” from hereafter): the crop, proventriculus, midgut, ileum, and rectum. DNA was also extracted from a rinsed leg (control), surface-sterilized larva, and scrapings from the nest interior surface. Minor workers were dissected under sterile conditions: each ant was chilled in a sterile petri dish at 0°C for 5 minutes, and repeatedly rinsed vigorously in sterile Milli-Q water. The gaster was opened by insertion of sterile ultra-fine forceps (Dumont
#5SF) between tergites 1 and 2, such that tergite 1 was lifted up and away from the body in a 30
way that prevented the exterior surface from contacting the interior of the gaster at any time. The whole intact gut was then lifted from the interior of the gaster and placed in sterile phosphate buffer with a separate pair of freshly sterilized forceps. The crop, midgut, ileum, and rectum were then separated and placed in separate tubes, again with freshly sterilized forceps for each separation. Specimens were discarded when any portion of the gut ruptured, touched another portion, contacted the outside surface of the exoskeleton, or was otherwise thought likely to have been contaminated. All tools were flame sterilized with 100% ethanol between every change in position during the dissections, and tools were cleaned via sterilization and sonication between dissections. In order to acquire sufficient DNA for sequencing we pooled each gut compartment from five workers for each colony. Sequences are available under accession number XXX.
We ground each sample with sterile pestles in enzymatic lysis buffer and extracted DNA using the Qiagen DNAeasy Blood and Tissue (Qiagen Inc., Valencia CA, USA) extraction kit following the pretreatment protocol for gram-positive bacteria with an increase of the pretreatment lysozyme incubation period to 24 h. Samples were screened for bacterial DNA using a universal 16S rRNA primer (Tet199F/1513R) and DNA in each sample was quantified with a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE,
USA). Extraction yielded enough bacterial DNA to attempt pyrosequencing (>3ng/ul) for most, but not all samples (Table S2).
DNA was sent to Research and Testing Laboratories, Lubbock, TX, USA for Roche 454 FLX-
Titanium pyrosequencing (Roche Applied Science, Indianapolis, IN, USA), using the 28F-519R bacterial assay for the V1-V3 variable regions of the16S rDNA gene (Dowd et al., 2008).
Sequence data was processed using Mothur v.1.33.0 (Schloss et al., 2009) following the standard 31
operating procedure for 454 generated data (http://www.mothur.org/wiki/454_SOP). We discarded sequences with fewer than 200 bp or more than two mismatches to the primer, aligned them to the SILVA database (Schloss, 2009, Pruesse et al., 2007), and removed chimeras using
UCHIME (Edgar et al., 2011). The remaining sequences were matched against the Mothur RDP reference database, and mitochondrial, chloroplast, archaea, eukaryote, and unknown sequences were removed.
Sequences were grouped into operational taxonomic units (OTUs) with at least 97% similarity, and representative sequences were classified both by matching against the Mothur RDP reference database and by nucleotide BLAST against the NCBI database. Statistical comparisons were conducted on data subsampled to 1685 sequences (the lowest number of sequences yielded by a gut sample). To investigate the similarity of bacterial communities along the alimentary canal, we compared samples using principal coordinates analyzes (PCoA) and hierarchical analyses with Jaccard distances as a measure of diversity within a gut compartment and Ward minimal variance criterion for sample similarity clustering. In order to investigate what variables best explained variation in the bacterial communities, we conducted PermanovaG (Chen et al.,
2012). All analyzes were done in R (Team RDC, 2008) and Mega v. 6.0 (Tamura et al., 2013).
The following R packages were used: Vegan (Oksanen et al. 2015), Ape (Paradis et al. 2004),
GUniFrac (Chen et al., 2012) and RColorBrewer (Neuwirth 2014).
Fluorescent bead experiments
To determine the size of particles capable of passing the proventriculus, adult ants were removed from lab-reared colonies and placed in closed petri dishes, where they were provided with a 100 ul droplet of 20% sucrose in Milli-Q water, 0.1% methylene blue dye, and 0.2 % yellow-green 32
fluorescent latex microsphere beads (Fluoresbrite®, Polysciences Inc., USA). We separately tested particle sizes of 6 µm, 2 µm, 0.5 µm, and 0.2 µm. Twenty-four hours after feeding, each ant was immobilized by cooling at 0°C for 5 minutes, rinsed in Milli-Q water, and dissected under sterile conditions similar to those described above. Because contamination by beads from the exoskeleton or ruptured gut could lead to false positives, great care was taken to flame- sterilize the dissecting tools and surfaces during every repositioning, and dissection tools were frequently examined under a fluorescence microscope for bead contamination. Ants were discarded if any portion of the gut ruptured, if contact between the gut and outside of the ant occurred, if the blue dye had not advanced to the rectum after 24 h (indicating that the food had passed all the way through the gut), or if they had numerous beads visibly adhered to their exoskeleton due to contact with the food droplet (posing a high risk of contamination by beads during dissection). Each portion of the freshly dissected gut (infrabuccal pocket of the mouth, crop, midgut, ileum of hindgut, and rectum) was placed under a separate cover slip in phosphate buffer and immediately examined for the presence of beads. To ensure that contact with the gut contents did not diminish the brightness of the YG fluorescent dye, we incubated 0.5um beads with midgut and hindgut contents at 27°C for 24 h, confirming that brightness of the microspheres did not decrease (n=3). To ensure that the blue dye and sugar solution did not diminish the brightness of the YG fluorescent spheres, we also incubated a sample of the mixture at 27°C for 24 h and confirmed that the brightness did not differ from a fresh sample of the spheres.
Gut compartments of ants fed with 6 µm, 2 µm, and 0.5 µm beads were examined and photographed using a Nikon Eclipse E600 fluorescence microscope (Nikon Instruments Inc., 33
Melville, NY, USA) using 20x and 40x dry lenses with 10x eyepieces (200x and 400x magnification) under FITC lighting. The number and location of beads in each gut compartment were noted and photographs of each gut compartment were taken using a Diagnostic Instruments
RT Color Spot Microscope Camera 2.2.1 (Sterling Heights, MI, USA). The smallest beads we tested, 0.2 µm, were not bright enough to view under the fluorescence microscope using dry optics. We therefore examined the gut compartments of 20 ants fed 0.2 µm fluorescent beads using a Zeiss 510 Meta Laser scanning confocal microscope (LSCM) on an Axioimager Z1 with a 40x oil plan fluor NA 1.3 lens (Carl Zeiss Microscopy LLC, Thornwood, NY, USA), under which the 0.2µm beads appeared clearly. Each gut compartment was first searched for the presence of beads visually using the multi-track line scan for the FITC and Rhodamine filters at
400x, under which beads appeared bright bluish green and autofluorescent structures such as chitin and spherocrystals (Bution and Caetano, 2010) appeared yellow green, making them easy to distinguish. Each slide was separately examined by ML or PR, who dissected the ants, and PJ, who had no prior expectation of the location of beads or of the gut compartment being viewed.
ML/PR and PJ separately noted the number and location of beads present in each slide. We then acquired confocal images of a representative region of each gut compartment using the 488nm line on the Ar laser and the 543nm green HeNe laser. In the cases where beads were found on midgut, ileum, or rectum slides, we used image z-stacks to determine whether the beads were on the outside surface of the organ (likely due to contamination during dissection), inside the lumen of the organ (likely due to passage through the proventriculus) or in the buffer outside the organ
(likely due to contamination during dissection and mounting).
Video analysis of oral-rectal trophallaxis, formation of filter layer 34
Two colony subsamples consisting of 8-10 mature workers, 3-5 larvae, and 3-5 pupae were video recorded from the time a pupa began to move its legs. A new adult (callow) was considered to have emerged from the pupa when it stood up and walked for the first time, and the time from emergence to the first instance of oral-rectal trophallaxis was calculated from the video. We chose to use the terminology oral-rectal trophallaxis because it clearly describes the contact between the ants. Previous authors have used a number of terms including abdominal trophallaxis (Wilson, 1976) and anal trophallaxis (Sanders et al., 2014).
RESULTS AND DISCUSSION
Structure of the proventriculus
The hymenopteran proventriculus is derived from the foregut crop and is similarly lined with cuticle (Eisner, 1957). Although the proventriculus of most Myrmicine ants is comprised of a simple tube and sphincter (Eisner, 1957), the genus Cephalotes has evolved a novel proventricular structure consisting of a large, flattened bulb covered in small cuticular spikes
(Fig 1). Our electron and confocal microscopy data revealed that the proventricular surface facing the interior of the crop is coated with a thick, non-cellular mucilaginous layer that is held in place by the spiky surface, leaving no obvious opening through the valve to the midgut (Fig
1C, 1D, 1E). This acellular layer is similar in some ways to the peritrophic matrix – an envelope of chitin fibers and glycoproteins produced in the midgut of many invertebrates (Hegedus et al.,
2009) – but it is produced in the foregut and remains adhered to the proventriculus rather than passing to the midgut. Liquid moving from the crop to the midgut must pass through this layer before entering the channels within the proventriculus that converge to a single tube entering the 35
midgut (Fig 1B, 1D). TEM data revealed that the tube to the midgut is ringed with muscle, suggesting that liquid is pulled through the layer by pumping action from below. The sclerotized upper surface of the valve, however, is rigid and lacks associated musculature, with no mechanism by which it could move or open.
Our micrographic evidence further indicated that although bacteria are typically associated with the crop wall and the surface of the proventricular layer facing the lumen of the crop (Fig 1F,
S1), they are absent within the proventricular layer and connecting channels to the midgut. This data may suggest that bacteria are unable to pass through the proventricular layer.
Gut microbe distribution in relation to the proventriculus
To determine whether the proventriculus plays a role in structuring the gut microbiome, we investigated 1) whether bacterial communities differ upstream (crop) and downstream (midgut) of the proventriculus, and 2) whether the proventricular surface hosts a distinct bacterial community from the crop. Using 454 amplicon pyrosequencing, we sequenced the crop, proventricular surface, midgut, hindgut ileum, and rectum of field-collected workers from seven spatially segregated colonies. We found no effect of colony source (four sequenced from freshly collected colonies vs. three kept in the lab for two months) on variation within colonies
(PermanovaG, F=1.457, p=0.234), allowing us to analyze all samples together.
We found striking partitioning of microbial communities between gut compartments, with highly similar communities occurring in all seven colonies (Fig 2). Examining the effect of location within the gut on microbial composition (blocked by colony), we found that gut compartments harbored significantly different communities of bacteria, with more variation explained by gut location than colony identity (PermanovaG, colony F=1.779, p=0.065, gutpart F=12.729, 36
p=0.001). Midgut samples yielded high numbers of reads for a single bacterial strain in the
Opitutales clade (Verrucomicrobia) (Fig 2), suggesting that this section primarily hosts an
Opitutales. The same strain was present at lower numbers throughout the gut.
While the midgut, ileum, and rectum communities were nearly identical across all colonies, the
crop and proventricular surface showed more variation. We found that the surface of the
proventriculus did not host a distinct bacterial community from that of the crop wall and crop
contents (Fig 3), although the crop community as a whole was distinct from the rest of the gut.
The crop tended to be dominated by several strains of Rhizobiales, a group of bacteria that has
been hypothesized to fix nitrogen (Russell et al., 2009) and to be involved in the digestion of pollen (Hu et al., 2014). The exception was Colony 4b, in which the crop and proventricular
samples contained predominantly Lactobacillus sp. Workers from Colony 4b were collected
when gathering extrafloral nectar from a cactus (Cylindropuntia acanthocarpa) near the host
tree, while workers from all other colonies were foraging exclusively on the host tree. Ingestion
of bacteria in the nectar is one possible explanation for the differing crop microbiota of this
colony.
Supporting our filter hypothesis, bacteria associated with the nest environment and foregut were
absent or present at only very low numbers in the midgut, hindgut, and rectum of all colonies
examined (Fig 2, Fig 4). For example, the six most numerous OTUs of the crop and
proventriculus (excluding OTU1 Opitutales from calculations) accounted for more than 90% of
sequences obtained from these organs, yet accounted for only 0.6% of sequences from the
midgut and hindgut samples. Similarly, only 4 of the 40 most numerous larval and nest
environment OTUs (90% of larval and nest reads) were present in the midgut and hindgut, 37
comprising only 0.008%. Most impressively, although 14317 Lactobacillus sp. sequences were found in the crop and proventriculus samples for colony 4b, not a single read was recovered from the midgut, ileum, and rectum samples of this colony (Fig 4).
It is worth noting that the similarity between communities of each compartment are associated with their location relative to the proventriculus, i.e. foregut communities tend to be more similar to each other than to the midgut and hindgut communities (Fig 3B). The significant partitioning of gut communities, along with the fact that most bacterial phylotypes from the foregut are not found in the midgut nor in the hindgut supports the idea that the proventriculus is involved in structuring and protecting the gut microbiome.
Porosity of the proventriculus
To determine the filtering capability and porosity of the proventriculus, we fed ants yellow fluorescent microspheres of one of the following sizes: 6µm, 2µm, 0.5µm, and 0.2µm, given in a solution of water, sugar and blue dye. After 24 h we dissected the ants and examined the gut compartments for the presence of microspheres. In no case did we find that microspheres of any of size passed through the proventriculus, and even the smallest microspheres tested, 0.2µm, did not pass beyond the crop of 20 ants (14 mature workers, 5 callow workers, 1 male, Fig S4, Table
S3). Nonetheless, the dye successfully reached the midgut and hindgut, indicating that dissolved molecules of food pass through the proventricular filter. In a more detailed examination of the position of particles, we found that 0.2µm beads were present on the surface of the filtering layer facing the crop interior but never within the proventricular channels (n=5, Fig. S5, Table S4).
Previous work has suggested that particles as large as 12µm can pass the proventriculus of C. rohweri (Roche and Wheeler, 1997). However, under light microscopy it is difficult to 38
distinguish such particles from the many spherocrystals (FigS4) produced in the midgut (Bution
and Caetano, 2010).
This remarkable filtering capacity is comparable to commonly used water purifying systems that
typically use ceramic filters with 0.2 um pore size to remove bacteria, eg. MilliQ (Millipore
Corporation). Because all but the very smallest bacteria cannot pass a 0.2µm filter (Razin and
Hayflick, 2010), we conclude that the proventriculus of Cephalotes rohweri has the capability of
excluding most ingested bacteria while allowing passage of dissolved molecules.
To determine whether the filtering mechanism is unique to the proventriculus of Cephalotine
ants, we tested the filtering ability of the proventriculus of two other ant genera. The genus
Pogonomyrmex belongs to the same subfamily of ants as Cephalotes, the Myrmicinae, but has
the simple funnel-shaped organ that is typical of the subfamily (Eisner, 1957). In four
Pogonomyrmex rugosus individuals we found that 2µm beads easily passed through the
proventriculus to the midgut (Table S6). The genus Camponotus belongs to the subfamily
Formicinae, which is characterized by a complex proventriculus with four hair-lined, sclerotized channels leading to a muscular pumping bulb that forces liquid through to the midgut (Eisner and
Wilson, 1952). In Camponotus fragilis, we found that although 6µm and 2µm beads were unable to pass the proventriculus, many thousands of 0.2µm beads passed through and were found in the lumen of the midgut (Fig S7, Table S6). This result suggests that the fine filtering capability of the Cephalotes proventriculus is a unique highly derived trait.
Although the proventricular filter prevents adult Cephalotes from digesting particles, colonies are still able to utilize solid foods by harnessing the digestive capabilities of larvae. When we fed colonies the 0.2µm bead mixture for 7 days (n=5), we found that mature workers packed the 39
beads into infrabuccal pellets and fed them to larvae, the guts of which were subsequently filled with beads (Fig S7, Table S5). Ant larvae lack the specialized gut morphology of adults but produce digestive enzymes adults lack (Hölldobler and Wilson, 1990), and are thus capable of digesting solid food. Larvae can then reciprocally provide digested liquids to feed workers
(Cassill et al., 2005). The colony as a whole therefore has two distinct digestive systems working in tandem.
Inoculation of the gut and formation of the filter layer
Previous work has suggested that workers emerge from the pupa with a sterile gut (Russell et al.,
2009), and our imaging studies also failed to detect gut bacteria in late-stage pupae and newly eclosed workers (Fig S8). If the proventriculus prevents the transit of bacteria through the gut and ants emerge from the pupal stage with a sterile gut, how do adult ants acquire the specialized bacteria found in the midgut and hindgut?
To find out, we video recorded the behavior of young, incompletely sclerotized (callow) workers from two colonies in the first 8 hours after eclosing. In line with previous observations (Wheeler,
1984, Wilson, 1976), we found that new workers solicit and consume rectal fluid (oral-rectal trophallaxis) from their nestmates soon after eclosion. In colony 8b, we observed that a new worker drank rectal fluid in repeated short bouts for 23 minutes 6 hours after eclosing (Fig 5F).
Similarly, in colony 5b a callow first drank for 11 minutes 3.25 hours after eclosing (Fig 5E).
This behavior is a potential route for microbial inoculation, provided that bacteria are able to pass through the proventriculus of new workers.
To find out whether the filtering layer is in place during this behavior, we dissected the worker from colony 5b immediately after this first bout of oral-rectal trophallaxis and prepared the 40
proventriculus for SEM imaging, along with ones from a pupa, a several day old callow, and a
mature forager from the same colony (Fig 5 A-D). Additionally, workers of different ages from
colonies 1b, 7a, 8b, and 9b were stained and examined via light microscopy to confirm that the
filtering layer developed similarly across colonies. In all cases the layer was absent in late-stage
pupae, first appeared in several day old callows, and was thickest in older workers (Fig 5). The
worker from colony 5b did indeed lack the layer (Fig 5B) at the time it consumed rectal
secretions from a nestmate. Thus the filtering layer on the proventriculus is not formed or
thicken until after the young worker adult engages in oral-rectal trophallaxis (Fig 5). New
workers therefore have only a brief window of time to inoculate their gut by drinking the rectal
excretions of nestmates before the proventriculus is sealed against further passage of bacteria.
Hypothetical role in immunity
The microbiota of Cephalotes may change when infected by pathogens such as Spiroplasma
(Kautz et al., 2013), a bacterium that is typically acquired through the alimentary canal (Bové,
1997). It is worth noting that Spiroplasma is among the smallest known bacteria (Razin et al.,
1998; Razin and Hayflick, 2010), smaller than the particle sizes tested here and potentially capable of infiltrating the proventriculus of Cephalotes. However, other microorganisms that may invade the alimentary canal of Cephalotes are typically larger than 0.2 um, e.g. nematodes
(Yanoviak et al., 2008) and microsporidians (P. Stock, personal communication). In these cases, pathogens may use alternative strategies to overcome the proventriculus filtration system, such as infecting larvae or young workers in which the proventriculus is porous or not completely sealed. This hypothesis was raised by Yanoviak et al. (2008) who noticed that workers infected by a parasitic tetradonematid nematode were generally smaller than healthy workers and 41
presented lighter cuticular pigmentation. Similarly, workers of C. rohweri infected with microsporidia also have lighter pigmentation (P. Stock, personal communication). Therefore, although the filtration capacity of the proventriculus may make the transit of parasites and non- symbiotic microorganisms difficult, the relative importance of its role in host immunity requires further testing.
CONCLUSIONS
We conclude that the unusual proventriculus and associated layer of C. rohweri is capable of excluding most if not all bacteria from entering the midgut of these ants, while still allowing dissolved nutrients to pass. To our knowledge, this is the first report of an animal organ capable of preventing ingested bacteria from transiting the gut while allowing food to pass. The recent finding of a similar mechanism in hemipterans, in which ingested microbes are sorted through a gut constriction (Ohbayashi et al., 2015), suggests that microbial-filtering organs may be also found in other insects that associate with extracellular gut microbes. A microbial filtering capability may be associated with fluid feeding, as both insect species have a liquid-based diet as adults. In the bean bug Riptortus pedestris, a microbial filter sequesters symbionts acquired from the environment, whereas in Cephalotes microbial filtering is possibly responsible for the persistent association with gut symbionts over time (Hu et al., 2014) as well as the pattern of vertical transmission and coevolution between host and microbes (Sanders et al., 2014). Other mechanisms of gut microbiome manipulation such as specificity through physiology and diet
(Kwong and Moran, 2015) may work in concert with this novel proventricular barrier to prevent the passage and establishment of foreign microbes. 42
Older workers host a pool of symbionts to inoculate new adults within colonies, while newly
mated queens are the vertical transmission route to new colonies. Sharing of the colony-level
microbiome may be important to the functioning of many eusocial animals, and a similar pattern
of oral-rectal feeding (or coprophagy) occurs in other societies including termites (Kohler et al.,
2012), bumblebees (Koch and Schmid-Hempel, 2011), and naked mole rats (Jarvis, 1981). For
Cephalotes, the proventriculus may function as part of the colony-level immune system, protecting the core gut microbiome of adults from introduction of pathogens. Processing of solids is delegated to larvae, which host an entirely different gut microbial community and act as a separate digestive system within the colony. Understanding the mechanism of microbe partitioning in this system opens up new avenues for research into host-microbe interactions and the nutritional role of the gut microbiome.
Acknowledgments: We wish to thank Wulfila Gronenberg, Mike Riehle, Tony Day, Gina
Zhang, and Joe Cicero for assistance in learning techniques. We also thank Nick Waser, Peter
Waser, Andrew Waser, Mary Price, Judith Bronstein, Amity Wilczek, Gordon Snelling, Corinne
Stouthamer, and Terry McGlynn for comments on previous drafts of the manuscript.
Author contributions: Research was conceived by ML, PR, and DW; experiments were designed by ML, PR, and DW with technical advice from PJ and AA; ML, PR, PJ, and AA conducted the experiments. Data were analyzed by ML and PR. Manuscript was written by ML with assistance from PR, all authors edited the final version of the manuscript. Figures were created by ML. 43
Supplementary information (SI) is available at ISMEJ’s website and at the end of this document.
REFERENCES
Anderson, K. E., et al. (2012). Highly similar microbial communities are shared among related and trophically similar ant species. Mol Ecol, 21:2282-2296.
Backhed, F., Ley, R. E., Sonnenburg, J. L., Peterson, D. A. & Gordon, J. I. (2005). Host-bacterial mutualism in the human intestine. Science, 307:1915-1920.
Baroni Urbani, C. & De Andrade, M. L. (1997). Pollen eating, storing, and spitting by ants.
Naturwissenschaften, 84:256-258.
Blum J. E., Fischer C. N., Miles J., Handelsman J. (2013). Frequent replenishment sustains the beneficial microbiome of Drosophila melanogaster. mBio 4:e00860-13.
Bové, J. M. (1997). Spiroplasmas: infectious agents of plants, arthropods and vertebrates. Wien
klin Wochenschr, 109:604-612.
Bution, M. L. & Caetano, F. H. (2010). The midgut of Cephalotes ants (Formicidae:
Myrmicinae): ultrastructure of the epithelium and symbiotic bacteria. Micron, 41:448-454.
Cariveau, D. P., Powell, J. E., Koch, H., Winfree, R., & Moran, N. A. (2014). Variation in gut microbial communities and its association with pathogen infection in wild bumble bees
(Bombus). ISMEJ, 8:2369-2379.
Cassill, D. L., Butler, J., Vinson, S. B. & Wheeler, D. E. (2005). Cooperation during prey digestion between workers and larvae in the ant, Pheidole spadonia. Insect Soc, 52:339-343. 44
Chen, J., et al. (2012). Associating microbiome composition with environmental covariates using generalized UniFrac distances. Bioinformatics, 28:2106-2113.
Cremer, S. & Sixt, M. (2009). Analogies in the evolution of individual and social immunity.
Philos. Trans. R. Soc. B Biol Sci, 364:129-142.
Dowd, S. E., et al. (2008). Evaluation of the bacterial diversity in the feces of cattle using 16S rDNA bacterial tag-encoded FLX amplicon pyrosequencing (bTEFAP). BMC Microbiol, 8:1-8.
Edgar, R. C., Haas, B. J., Clemente, J. C., Quince, C. & Knight, R. (2011). UCHIME improves sensitivity and speed of chimera detection. Bioinformatics, 27:2194-2200.
Eisner, T. A. (1957). A comparative morphological study of the proventriculus of ants
(Hymenoptera: Formicidae). Bull Mus Comp Zool, 116:439-490.
Eisner, T. A. & Wilson, E. O. (1952). The morphology of the proventriculus of a formicine ant.
Psyche, 59, 47-60.
Engel, P. & Moran, N. A. (2013). The gut microbiota of insects- diversity in structure and function. FEMS Microbiology Rev, 37:699-735.
Hegedus, D., Erlandson, M., Gillott, C. & Toprak, U. (2009). New insights into peritrophic matrix synthesis, architecture, and function. Annu Rev Entomol, 54:285-302.
Hölldobler, B. & Wilson, E. O. (1990). The Ants. Belknap Press of Harvard University Press,
Cambridge, Massachusetts.
Hongoh, Y. (2011). Toward the functional analysis of uncultivable, symbiotic microorganisms in the termite gut. Cell Mol Life Sci, 68:1311-1325. 45
Hu, Y., Lukasik, P., Moreau, C. S. & Russell, J. A. (2014). Correlates of gut community
composition across an ant species (Cephalotes varians) elucidate causes and consequences of
symbiotic variability. Mol Ecol, 23:1284-1300.
Jarvis, J. U. (1981). Eusociality in a mammal: cooperative breeding in naked mole-rat colonies.
Science, 212:571-573.
Jones, R. T., Vetter, S. M., Montenieiri, J., Holmes, J., Bernhardt, S. A., & Gage, K. L. (2013).
Yersinia pestis infection and laboratory conditions alter flea-associated bacterial communities.
ISME J, 7:224-228.
Kimura, M. (1980). A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. J Mol Evol, 16:111-120.
Koch, H. & Schmid-Hempel, P. (2011). Socially transmitted gut microbiota protect bumble bees against an intestinal parasite. Proc Natl Acad Sci U. S. A., 108:19288-19292.
Kohler, T., Dietrich, C., Scheffrahn, R. H. & Brune, A. (2012). High-resolution analysis of gut environment and bacterial microbiota reveals functional compartmentalization of the gut in wood-feeding higher termites (Nasutitermes spp.). Appl Environ Microbiol, 78:4691-4701.
Kwong, W. K. & Moran, N. A. (2015). Evolution of host specialization in gut microbes: the bee gut as a model. Gut Microbes, 6:214-220.
Mcfall-Ngai, M., et al. (2013). Animals in a bacterial world, a new imperative for the life sciences. Proc Natl Acad Sci U. S. A., 110:3229-3236.
Moran, N. A., Mccutcheon, J. P. & Nakabachi, A. (2008). Genomics and evolution of heritable bacterial symbionts. Ann Rev Genet, 42:165-190. 46
Muegge, B. D., et al. (2011). Diet drives convergence in gut microbiome functions across
mammalian phylogeny and within humans. Science, 332:970-974.
Nelson A. M., Walk S. T., Taube S., Taniuchi M., Houpt E. R., Wobus C. E., et al. (2012).
Disruption of the human gut microbiota following Norovirus infection. PLoS ONE 7:e48224.
Neuwirth, E. (2014). RColorBrewer: ColorBrewer Palettes. R package version 1.1-2. Available: http://CRAN.R-project.org/package=RColorBrewer.
Nyholm, S. V., & McFall-Ngai, M. (2004). The winnowing: establishing the squid–Vibrio symbiosis. Nature Rev Microbiol, 2:632-642.
Nyholm, S. V. & Graf, J. (2012). Knowing your friends: invertebrate innate immunity fosters beneficial bacterial symbioses. Nat Rev Microbiol, 10:815-827.
Ohbayashi, T., et al. (2015). Insect's intestinal organ for symbiont sorting. Proc Natl Acad Sci U.
S. A., 112:E5179-E5188.
Oksanen J, et al. (2015). Vegan: Community Ecology Package. R package version 2.3-0.
Available: http://CRAN.R-project.org/package=vegan.
Paradis E, Claude J, Strimmer K. (2004). APE: analyses of phylogenetics and evolution in R language. Bioinformatics, 20:289-290.
Pernice, M., Simpson, S. J. & Ponton, F. (2014). Towards an integrated understanding of gut microbiota using insects as model systems. J Insect Physiol., 69:12-18.
Powell, S. (2008). Ecological specialization and evolution of a specialized caste in Cephalotes ants. Funct Ecol, 22:902-911. 47
Pruesse, E., Quast, C., Knittel, K., Fuchs, B. M., Ludwig, W., Peplies, J., et al. (2007). SILVA: a
comprehensive online resource for quality checked and aligned ribosomal RNA sequence data
compatible with ARB. Nucleic Acids Res, 35:7188-7196.
Razin, S. & Hayflick, L. (2010). Highlights of mycoplasma research- an historal perspective.
Biologicals, 38:183-190.
Razin, S., Yogev, D., & Naot, Y. (1998). Molecular biology and pathogenicity of mycoplasmas.
Microbiol and Mol Biol Rev, 62:1094-1156.
Roche, R. K. & Wheeler, D. E. (1997). Morphological specializations of the digestive tract of
Zacryptocerus rohweri (Hymenoptera: Formicidae). J Morphol, 234:253-262.
Russell, J., Moreau, C. S., Goldman-Huertas, B., Fujiwara, M., Lohman, D. J., & Pierce, N. E.
(2009). Bacterial gut symbionts are tightly linked with the evolution of herbivory in ants. Proc
Natl Acad Sci U. S. A., 106:21236-21241.
Sanders, J. G., Powell, S., Kronauer, D. J., Vasconcelos, H. L., Frederickson, M. E., & Pierce, N.
E. (2014). Stability and phylogenetic correlation in gut microbiota: lessons from ants and apes.
Mol Ecol, 23, 1268-1283.
Sansonetti, P. J. (2004). War and peace at mucosal surfaces. Nat Rev Immunol, 4:953-964.
Schloss, P. D. (2009). A high-throughput DNA sequence aligner for microbial ecology studies.
PLoS ONE, 14:e8230.
Schloss, P. D., Westcott, S. L., Ryabin, T., Hall, J. R., Hartmann, M., Hollister, E. B., et al.
(2009). Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl Environ Microbiol, 75:7537-7541. 48
Stecher, B., Maier, L., & Hardt, W. D. (2013). 'Blooming' in the gut: how dysbiosis might
contribute to pathogen evolution. Nat Rev Microbiol, 11:277-284.
Tamura, K., Stecher, G., Peterson, D., Filipski, A. & Kumar, S. (2013). MEGA6: Molecular
Evolutionary Genetics Analysis Version 6.0. Mol Biol and Evol, 30:2725-2729.
Team R. D. C. (2008). R: A language and environment for statistical computing. In:
COMPUTING, R. F. F. S. (ed.). Vienna, Austria: http://www.R-project.org.
Wheeler, D. E. (1984). Behavior of the ant, Procryptocerus scabriusculus (Hymenoptera:
Formicidae), with comparisons to other cephalotines. Psyche, 91:171-192.
Wilson, E. O. (1976). A social ethogram of the neotropical arboreal ant Zacryptocerus varians
(Fr. Smith). Anim Behav, 24:354-363.
Yanoviak, S. P., Kaspari, M., Dudley, R. & Poinar, G. (2008). Parasite-induced fruit mimicry in
a tropical canopy ant. Am Nat, 171:536-544.
49
FIGURES
Fig 1. Morphology of the C. rohweri proventriculus. Panels show (A) the location of the proventriculus relative to other parts of the gut, (B) cutaway diagram showing the channels within the proventriculus, (C) TEM cross-section through the proventriculus and filtering layer, (D) SEM top view of the proventriculus with the filtering layer removed from the cuticular spines, (E) SEM top view of the proventriculus with the filtering layer in place, and (F) SEM of a cross-section of the proventriculus and associated layer showing cuticular spines, filtering layer, and bacteria on surface. In panels B-F the cuticular and underlying cellular parts of the proventriculus are colored yellow, the filtering layer is red, bacteria and detritus on the surface of the filtering layer are purple, and the crop lumen is blue. Illustration and SEM colorization by MCL. 50
Fig 2. The microbial communities found in different gut compartments (crop, proventriculus, midgut, ileum, rectum), the larva, and the nest interior. The percentage of sequences in each sample is shown for 97% OTUs found in the crop contents, proventriculus and associated layer, and midgut for seven field-caught colonies, as well as control (rinsed leg), larval, and nest interior sequences for samples that could be sequenced (Table S2). Illustration by MCL.
51
Fig 3. Principal coordinate analysis plot showing clustering among sample types (A) and (B) Hierachical clustering of gut samples (Created with R packages Vegan (Oksanen et al. 2015), Ape (Paradis et al. 2004) and RColorBrewer (Neuwirth 2014).)
52
Fig 4. The average number of total reads per sample in environmental (nest and larval), foregut, midgut, and hindgut samples for the 50 most numerous OTUs.
53
Fig 5. SEM images showing the development of the filtering layer before and after oral-rectal trophallaxis. In all panels the cuticular parts of the proventriculus are colored yellow and the filtering layer is red. Panels show cross-sections of the proventriculus and filtering layer from (A) a late-stage pupa, (B) a newly-emerged worker that has just completed oral-rectal trophallaxis, (C) a several-day-old callow, and (D) a mature worker. Panels (E) and (F) show still images from video in which callow workers (light color) participate in oral-rectal trophallaxis. The callow in (E) was dissected to produce the image in (B).
54
SUPPORTING INFORMATION
Collection information for all colonies used is given in Table S1. Original, uncolored micrographs for Figs. 1 and 2 are provided in Figs. S1 and S2. Fig. S3 shows the large numbers of bacteria present in different gut regions of mature workers. Electron microscopy indicated that bacteria were absent in these gut regions in several late-stage pupae and newly eclosed workers. Extraction from 5 workers yielded enough bacterial DNA to attempt pyrosequencing (>3ng/ul) for most, but not all samples. The nest interior sample for Colony 4b and rectum of 1b could not be sequenced due to low quality of DNA. A summary of the use of all samples for sequencing is provided in Table S2. Table S3 presents the detailed results of the bead experiment, in which no beads could conclusively be shown to have passed through the proventriculus. Detailed notes are given in the few cases where beads were present elsewhere on the slide (likely due to contamination during dissection). Fig. S4 shows representative confocal images for each gut section of the 14 mature workers, 1 male alate, and 5 callow workers (total of 20 ants) examined for the presence of 0.2µm beads. Spherocrystals and crystals can be seen in many of the images, a possible source of confusion in previous studies. The same channel gain levels were used for all images. Some images appear brighter due to proximity to larger areas of cuticle such as tracheae or the cuticular crop wall. In one of the 20 ants we examined for 0.2µm beads (Fig S4, Colony 2b ant 1 ileum), 8 beads were found on the hindgut ileum. We used image z-stacks to determine that 7 of these beads were definitely on the outside surface of the organ, probably due to contamination with the forceps. It was unclear whether the eighth bead was in a fold on the outside surface or possibly inside the organ, because it was at the same level as the wall of the organ in our image stack. The Z-stack side view for this sample is shown in Figure S6. Figure S6a shows an example of a larval gut packed with beads after being reared with workers fed the 0.2µm bead mixture. Workers pressed the beads into infrabuccal pellets and fed them to the larvae (Table S5). In four Pogonomyrmex rugosus individuals we found that 2µm beads easily passed through the proventriculus to the midgut, and in Camponotus fragilis we found that many thousands of 0.2µm beads passed through the proventriculus (Table S6, Figure S7b).
55
SUPPORTING INFORMATION: FIGURES
Fig. S1. Original, uncolored micrographs used in Fig. 1, showing (A) SEM top view of the proventriculus with the filtering layer removed from the cuticular spines with KOH, (B) SEM top view of the proventriculus with the filtering layer in place, (C) TEM section through the proventriculus and filtering layer, and (D) SEM of a cross-section of the proventriculus showing cuticular spines, filtering layer, and bacteria on surface. A linear adjustment of contrast was applied to these images to appear more clearly in print.
Fig. S2. Original, uncolored micrographs used in Fig. 3, showing SEM images of cross-sections of the proventriculus and filtering layer from (A) a late-stage pupa, (B) a newly-emerged worker that has just completed oral-rectal trophallaxis, (C) a several-day-old callow, and (D) a mature worker. A linear adjustment of contrast was applied to these images to appear more clearly in print.
56
Fig. S3. Original, uncolored micrographs showing the numerous bacteria present in (A) the hindgut ileum lumen and (B) rectal lumen of adult workers. Numerous bacteria typically adhere to the interior crop wall of adult workers, as shown in representative (C) SEM and (D) TEM (lu= lumen, b= bacteria, w= cuticular crop wall) images. A linear adjustment of contrast was applied to these images to appear more clearly in print.
57
58
Fig. S4. Confocal images for a representative area of the crop, midgut, and lumen of hindgut for each of 14 mature worker ants, one male, and 5 callow workers used to test for the passage of 0.2µm microspheres. The identity of objects in each of these images was confirmed visually under FITC lighting by both ML and PJ prior to imaging. Features are labeled as beads (b), proventriculus (pv), tracheae (tr), and contents of the lumen (c) such as crystals, spherocrystals, and other particles. Most contents of the midgut and ileum lumens did not fluoresce and appear black. Z stack analysis of beads on the ileum of Colony 2b ant 1 is shown in Fig. S6. Notes on the possible bead-like object on the midgut of Colony 6b callow 2 are in Table S3.
59
Fig. S5. Confocal images showing the location of 0.2µm beads relative to proventricular channels and surface. (A) shows a composite image of only portions of slices determined to show the interior of the proventriculus, while (B) shows a composite of the entire slices (interior and exterior). Four example slices are shown in (C), (D), (E), and (F), moving upward from the base of the proventricular bulb to the top. The approximate boundary between exterior and interior is shown as a dotted line, beads are labeled b, and tracheae are labeled tr. Note that beads did not just occur on the upper outside surface of the proventriculus, but on the lower outside surface as well (e.g. beads can be seen on the spikes wrapping around the bottom in (C).
60
Fig. S6. Confocal image and side-view image z-stacking showing the surface location of the 0.2µm beads found on the ileum of ant 1 from colony 2b. The sections of (A) surrounded by boxes were z-stacked, then rotated to produce the images in (B). Beads 1, 3, 4, 5, 6, 7, and 8 are clearly on the outside surface of the wall of the ileum (red), not within the lumen (lu). Bead 2 is located at the same level as the wall of the ileum, and might be located within a fold of the surface. Objects in the figure are labeled b1-b8 for beads, Lu=lumen of the ileum, and W=wall of the ileum.
61
Fig. S7. Confocal images showing (A) the C. rohweri larval gut lumen (lu) completely full of 0.2µm beads (b) from ingested infrabuccal pellets, and (B) the gut of Camponotus fragilis containing thousands of 0.2µm beads that have passed the proventriculus. Tracheae are labeled tr and gut content other than beads are labeled as c. Note that ant larvae have a blind-end digestive tract and wastes are sequestered until pupation and then discarded in the meconium. Larvae experimentally fed fluorescent beads expelled bright yellow meconium at pupation.
Fig. S8. Original, uncolored micrographs showing no bacteria in representative areas of (A) the crop lumen, (B) the proventricular surface, (C) the lumen of the hindgut ileum, and (D) of a pupa nearing eclosion (tan cuticular pigmentation developed). A linear adjustment of contrast was applied to these images to appear more clearly in print. The midgut is not shown, as the organ is more delicate in pupae and was damaged by the electron beam during attempted imaging.
62
Table S1. The collection location, date, and experimental use of all colonies.
Colony Collection information Use in experiments DNA extracted from gut compartments immediately March 2013, Tucson after field collection. Colony Cephalotes 1b Mountain Park then reared in lab and used for bead experiment in spring 2014. DNA extracted from gut compartments immediately after field collection. Workers March 2013, Tucson prepared immediately after Cephalotes 2b Mountain Park field collection for SEM imaging, Fig. 1. Colony then reared in lab and used for bead experiment in spring 2014. DNA extracted from gut compartments immediately after field collection. Workers March 2013, Tucson prepared for TEM imaging Cephalotes 3b Mountain Park two weeks after collection, Fig. 1. Colony then reared in lab and used for bead experiment in spring 2014. DNA extracted from gut compartments, larvae, and nest April 2013, Tucson Mountain interior immediately after field Cephalotes 4b Park collection. Colony then reared in lab and used for bead experiment in spring 2014. Colony reared in lab and used for bead experiment in spring 2014. Colony used for video May 2013, Tucson Mountain Cephalotes 5b analysis of oral-rectal Park trophallaxis. Workers prepared for SEM imaging of layer formation, Fig. 3. Colony reared in lab and used May 2013, Tucson Mountain Cephalotes 6b for bead experiment in spring Park 2014. Colony reared in lab and used May 2013, Tucson Mountain Cephalotes 8b for bead experiment in spring Park 2014. Colony used for video 63
analysis of oral-rectal trophallaxis. Worker prepared for SEM image showing proventricular layer (Fig. 1). Colony reared in lab and used May 2013, Tucson Mountain Cephalotes 9b for bead experiment in spring Park 2014. DNA extracted from gut compartments after 2 months August, 2012, Tucson Cephalotes 1a in the lab. Worker prepared Mountain Park with KOH for SEM imaging, Fig. 1. DNA extracted from gut August 2012, Tucson Cephalotes 6a compartments after 2 months Mountain Park in the lab. DNA extracted from gut August 2012, Tucson Cephalotes 7a compartments after 2 months Mountain Park in the lab. Colony reared in lab from Camponotus 1 August 2012, Tucson AZ queen and used for bead experiment in spring 2014. Workers collected from colony in field and used for Pogonomyrmex 1 June 2014, Tucson AZ bead experiment in spring 2014.
Table S2. The use of each sample for the sequencing study. This table details the reasons that certain samples were not included in the final results. Crop Colo Rinsed content Proventric Nest ny leg s ulus Midgut Ileum Rectum Larva interior 1a Did not Not Sequenced Sequen Sequen Not Not Not yield sampled ced ced sampled sampled sampled enough bacterial DNA for sequenci ng 64
6a Did not Not Did not Sequen Sequen Not Not Not yield sampled yield ced ced sampled sampled sampled enough enough bacterial bacterial DNA DNA for for sequencing sequenci ng 7a Did not Not Sequenced Sequen Sequen Not Not Not yield sampled ced ced sampled sampled sampled enough bacterial DNA for sequenci ng 1b Did not Sequen Sequenced Sequen Sequen Sequenc Sequenc Sequenc yield ced ced ced ing ed ed enough failed bacterial (low DNA quality) for sequenci ng 2b Sequenc Sequen Sequenced Sequen Sequen Sequenc Did not Sequenc ed ced ced ced ed yield ed enough bacterial DNA for sequenci ng 3b Did not Sequen Sequenced Sequen Sequen Sequenc Did not Did not yield ced ced ced ed yield yield enough enough enough bacterial bacterial bacterial DNA DNA DNA for for for sequenci sequenci sequenci ng ng ng 4b Did not Sequen Sequenced Sequen Sequen Sequenc Did not Sequenc yield ced ced ced ed yield ing enough enough failed bacterial bacterial (low 65
DNA DNA quality) for for sequenci sequenci ng ng
Table S3. Results of the microsphere experiment. Inside Insid Insi Bea infrabuc Insi e de Inside Colo d cal de midg ileu rectu ny size pocket crop ut m m Beads found elsewhere on slide 3b 6 >1000 0 0 0 0 µm 2b 6 >1000 1 0 0 - 4 in buffer floating outside midgut µm 2b 6 >1000 1 0 0 0 1 in buffer floating outside ileum µm 2b 6 >1000 0 0 0 0 µm 2b 6 >1000 6 0 0 0 µm 9b 2 >1000 >500 0 0 0 µm 9b 2 >1000 >100 0 0 0 µm 9b 2 >1000 0 0 0 0 µm 2b 2 >1000 >500 0 0 0 µm 2b 2 >1000 >500 0 0 0 µm 9b 0.5 >1000 0 0 0 0 µm 9b 0.5 >1000 0 0 0 0 µm 9b 0.5 >1000 >100 0 0 0 µm 9b 0.5 >1000 0 0 0 - µm 8b 0.5 >1000 0 0 0 0 µm 8b 0.5 >1000 16 0 0 0 µm 66
8b 0.5 >1000 0 0 0 0 µm 8b 0.5 >1000 >100 0 0 - µm 8b 0.5 >1000 0 0 0 0 µm 3b 0.5 >1000 0 0 0 0 µm 2b 0.5 >1000 0 0 0 0 µm 2b 0.5 >1000 0 0 0 0 µm 2b 0.5 >1000 0 0 0 0 µm 2b 0.5 >1000 10 0 0 0 µm 2b 0.5 >1000 0 0 0 0 µm 9b 0.2 >1000 >100 0 0 - 1 floating loose in buffer under µm coverslip with midgut 9b 0.2 >1000 >100 0 0 - 2 floating loose in buffer under µm coverslip near midgut 8b 0.2 >1000 >100 0 0 - µm 8b 0.2 >1000 >100 0 0 - 1 under same coverslip as midgut, µm but far away from organ in buffer 6b 0.2 >1000 >100 0 0 - 5 on the outside surface of the µm midgut on trachae, none inside as confirmed with image stack. 2 more floating in buffer outside of midgut, near spilled midgut contents. None inside midgut lumen. 5b 0.2 >1000 >100 0 0 - µm 5b 0.2 >1000 >100 0 0 - µm 4b 0.2 >1000 >100 0 0 - µm 4b 0.2 >1000 >100 0 0 - 1 floating in buffer outside midgut µm 2b 0.2 >1000 >100 0 0 - µm 2b 0.2 >1000 25 0 0 - µm 67
2b 0.2 >1000 >100 0 0 or - 8 beads on outside surface of ileum, µm 1 likely from forceps because they are all in a band across the middle. One bead is ambiguous as to whether it is in a fold on the outside surface or possibly inside the organ (see Fig. S4). The rest are clearly on the outside. 1b 0.2 >1000 >100 0 0 - µm 1b 0.2 >1000 >100 0 0 - µm 1b 0.2 - >100 0 - - µm 1b 0.2 - >100 0 0 - Callow µm 5b 0.2 - >10 0 - - Callow, ileum destroyed in µm dissection 6b 0.2 - >10 0 0 - Callow, possibly one bead on upper µm surface of midgut, probably not (wrong color- yellow green similar to malpighinan tubual cells that fluoresce, not blue-green of beads) 6b 0.2 - >10 0 0 - Callow µm 8b 0.2 - 15 0 0 - Callow, one bead floating in buffer µm outside ileum, three in buffer outside midgut, none inside lumen of either
Table S4. Number of beads observed in the crop lumen and inside the proventricular channels of Cephalotes rohweri workers.
In crop outside inside proventricular Colony worker is from proventricular layer channels 1b >100 0 1b >1000 0 1b >100 0 2b >100 0 6b >100 0
68
Table S5. Results of experiment rearing larvae with workers fed 0.2 µm beads. Number of beads observed inside the Colony larval gut 8b >1000 8b >1000 5b >1000 5b >1000
Table S6. Results of bead experiment with other ant species. inside infrabuccal inside elsewhere on Colony Bead size pocket inside crop midgut slide Camponotus 1 6 µm >1000 >1000 0 Camponotus 1 6 µm >1000 >1000 0 Camponotus 1 6 µm >1000 39 0 1 in buffer outside crop, 2 in buffer outside midgut, none inside midgut Camponotus 1 6 µm >1000 0 0 Camponotus 1 2 µm - >1000 0 Camponotus 1 2 µm - >100 0 Camponotus 1 0.2 µm >1000 >1000 >1000 Pogonomyrme 2 µm >10 >100 x 1 Pogonomyrme 2 µm - >100 >100 midgut x 1 ruptured, many more beads have clearly spilled out Pogonomyrme 2 µm - >100 >100 x 1 Pogonomyrme 2 µm - >100 >100 x 1
69
APPENDIX B: Microbiome partitioning in turtle ants
Microbiome partitioning in turtle ants
Pedro A P Rodrigues1*, Piotr àukasik2,3, Yi Hu2, Jacob Russell2, Michele Lanan4, Jeffrey
Oliver5, Diana Wheeler6
1 Graduate Interdisciplinary Program in Entomology and Insect Science, University of Arizona,
Tucson, AZ 85721, USA
2 Department of Biology, Drexel University, Philadelphia, PA 19104, USA
3 Division of Biological Sciences, University of Montana, Missoula, MT 59801, USA
4 Chair of the Natural Sciences, Deep Springs College, Big Pine, CA 93513, USA
5 Health Sciences Library, University of Arizona, Tucson, AZ 85721, USA
6 Department of Entomology, University of Arizona, Tucson, AZ 85721, USA
* Corresponding author
Email: [email protected]
Keywords: symbiosis, mutualisms, herbivory, Cephalotes, nutritional ecology, arboreal ants, social insects
ABSTRACT
Investigations of the gut microbiome have revealed that microorganisms previously thought to function as commensals may actually play beneficial roles that have influenced their host ecology and evolution. In insects, highly specialized herbivory has been found to be 71
accompanied by intimate associations with symbiotic bacteria. Less specialized herbivores such
as exudate-feeding ants have also been hypothesized to benefit from their association with gut
microbes, although this relationship is not well-understood. We investigated whether the gut
communities of exudate-feeding ants are structured similarly to the functional partitioning found
in the gut microbiota of highly specialized herbivores. We sampled gut bacteria of the ant
Cephalotes varians and compared our results to published data in the same species and a
congener, C. rohweri. We found that bacterial communities tend to be structured in three partitions: (1) an upper gut partition (crop and proventriculus); (2) a middle partition (midgut); and (3) a lower gut partition (ileum and rectum). The diversity of microbes increases from the upper gut toward the lower gut, and we found evidence that some bacterial phylotypes show specificity to the organs in which they are located. In addition, our results suggest that bacteria in the infrabuccal pocket, a pouch-like organ in the head of ants, play a key role in digesting solid food that accumulates in this organ. Using these patterns, we generate predictions regarding the functional potential of these microbes, and offer new models and mechanisms to explain the interplay among microbiota, nutrition and colony growth in Cephalotes.
INTRODUCTION
Beneficial host-microbe associations have been implicated in niche diversification and speciation of animals (Joy 2013). The evolution of herbivory, in particular, is often accompanied by associations with biochemically diverse microorganisms that are able to synthesize essential amino acids, vitamins and enzymes that their hosts would not be able to acquire otherwise
(Brune & Ohkuma 2011; Engel & Moran 2013). Sap-feeding insects, as well as wood-eating 72
termites, represent some of the best studied cases of long co-evolutionary history, associated
with high rates of co-diversification of host and symbiont (Douglas 2009; Bennett & Moran
2015; Bignell 2016). These highly specialized insects are unable to survive without their
microbial mutualists, and this partnership is maintained through morphological and physiological
adaptations. For instance, in order for bacterial biochemical processes to take place, specific
physical-chemical conditions must be met, which correspond to specific locations within and
among gut compartments (Bignell 2011; Donaldson et al. 2016).
Growing evidence shows that ants have evolved tight associations with gut microbes. It has beenhypothesized that, similar to phytophagous and wood-eating insects, they depend on the metabolically diverse abilities of their symbionts to acquire nutrients that are deficient in their diet (Russell et al. 2009; Anderson et al. 2012; Hu et al. 2014). Arboreal ants, in particular, have been characterized as functional herbivores, i.e., they consume mostly extrafloral nectar and honeydew from sap-feeding hemipterans (Blüthgen et al. 2000; Davidson et al. 2003). These exudate-feeding ants are a diverse and dominant group of arthropods in tropical forests
(Davidson & Patrell-Kim 1996) that often carry microbes in specialized sections of their digestive tract (Cook & Davidson 2006). One apparent paradox in this association is the fact that although gut microbes are likely to supplement the diet of adult ants, the colony demand for nutrients is highest for the sessile growing brood in the nest (Hölldobler & Wilson 1990).
Therefore, how is a carbohydrate-rich diet translated into colony growth? In this study, we characterize gut communities of an arboreal ant in terms of microbial taxonomy and function, taking into consideration their location within the digestive tract to explain how these traits may connect to colony growth. 73
Types of ant-gut bacteria associations
At least two types of symbiotic bacteria have been found to associate with the digestive tract of
ants: (1) intracellular bacteria (endosymbionts) found in specific host cells called bacteriocytes;
and (2) extracellular bacteria inhabiting the lumen of the digestive tube. Blochmannia, an endosymbiont of ants in the tribe Camponotini, is a well-studied bacterium that can recycle nitrogen and synthesize essential amino acids (Feldhaar et al. 2007). This association is considered to have driven high diversification in this tribe (Feldhaar et al. 2007), best illustrated by the hyper-diverse genus Camponotus, in which over 1000 species have been described
(Bolton 2013). Despite this success, endosymbiotic associations of nutritional significance seem to be otherwise rare among exudate-feeding ants, having been found to date only in the tribe
Camponotini and in Cardiocondyla (Klein et al. 2016). Associations with extracellular gut symbionts appear to be considerably more common (Cook & Davidson 2006; Stoll et al. 2007;
Russell et al. 2009). Ants in the genus Cephalotes, for instance, have undergone remarkable coevolution with gut symbionts (Russell et al. 2009; Anderson et al. 2012; Sanders et al. 2014).
Also known as turtle ants, Cephalotes is a diverse group of arboreal ants that hosts a core community of bacteria found consistently across species in this genus and related genera in the same tribe, the Cephalotini (Anderson et al. 2012). It has been hypothesized that these microbes might function as nutritional mutualists, providing nutrients that may be lacking in the mostly herbivorous diet of these ants (Caetano & da Cruz-Landim 1985a; De Andrade & Urbani 1999;
Cook & Davidson 2006), although this hypothesis has not been rigorously tested yet. The finding that Cephalotes harbors bacteria in the clade Rhizobiales, known to include nitrogen-fixing bacteria, has triggered great interest in determining how gut microbes contribute to the nutrition 74
of these ants (Russell et al. 2009; Hu et al. 2014). In addition, it has long been discussed whether the habit of pollen collection, frequently observed in Cephalotes but unusual among ants as a whole (Urbani & de Andrade 1997), is associated with a presumably specialized gut microbiota that might aid in digesting the recalcitrant pollen wall (Russell et al. 2009; Hu et al. 2014).
Specializations of Cephalotes and hypothetical roles of gut symbionts
The digestive tube of Cephalotes shows morphological specializations that have been associated with the evolution and maintenance of their specific microbiota (Roche & Wheeler 1997; Bution
& Caetano 2008; Lanan et al. 2016). The ileum, for instance, is enlarged compared to other ants, and filled with a mass of bacteria (Roche & Wheeler 1997; Bution & Caetano 2010). Similar enlarged gut compartments containing beneficial bacteria are typical of animals that evolved herbivory in partnership with gut microbes, both vertebrate and invertebrates (McBee 1971;
Wolcott & O'Connor 1992; Stevens & Hume 1998; Engel & Moran 2013). In termites, physicochemical gradients are involved in the determination of the location and structure of beneifical microbial communities along the gut (Schmitt-Wagner & Brune 1999; Kohler et al.
2012). Another similarity to termites is the fact that Cephalotes performs a behavior equivalent
to proctodeal trophallaxis, where a newly emerged worker (callow) will solicit and consume
rectal fluids from an older worker (Wilson 1976), a behavior thought to be associated with the
acquisition of their typical microbiota (Wheeler 1984; Lanan et al. 2016). The acquisition of
microbiota involves another morphologically specialized organ, the proventriculus, a valve
between the crop and the midgut that, in Cephalotes, has evolved a complex morphology (Eisner
1957). In these ants, the proventriculus is unique in its ability to filter particles as small as 0.2um,
supporting the hypothesis that these ants are strictly liquid feeders (Lanan et al. 2016). 75
Observations post-eclosion showed the formation of a sealing layer on top of the proventriculus
that becomes thicker as callows transition to fully pigmented adult workers (Lanan et al. 2016).
The progressive accumulation of the sealing layer suggests that the formation of the
proventriculus might be incomplete in the first hours post-eclosion, allowing for the rectal fluid
inoculum to reach and colonize the midgut and hindgut (Lanan et al. 2016). Studying Cephalotes
rohweri, Lanan et al. (2016) noted that the bacterial communities are organized according to
their localization in relation to the proventriculus. Communities of the foregut, midgut and
hindgut were consistently different, across field-collected and lab-reared colonies, indicating that
the compartmentalization is resilient to environmental conditions such as nesting and diet.
Functional herbivory in Cephalotes may be possible because of the supporting roles of microbes
in digesting pollen and fixing and recycling nitrogen in Cephalotes, but little evidence supports
these claims. The hypothesis of pollen digestion by gut bacteria of Cephalotes is mostly based on
the observation that these ants accumulate pollen grains in their infrabuccal pocket (Urbani & de
Andrade 1997), a pouch-like cavity between the mouth and pharynx. In addition, there are
potential roles of microbes in the nutritional ecology of these ants that have not yet been
investigated. For instance, Bactrocera olive flies and Camponotus carpenter ants share with
Cephalotes a mostly herbivorous diet and consumption of bird droppings, and both have been found to acquire essential amino acids from their symbionts, which also have the ability to recycle urea and break down uric acid (Feldhaar et al. 2007; Ben-Yosef et al. 2014). In addition to pollen, fungus may be found within the infrabuccal pockets of ants (Wheeler & Bailey 1920;
Mankowski & Morrell 2004), and even within the digestive tube of Cephalotes (Caetano & da 76
Cruz-Landim 1985b; Caetano 1989), but the potential for gut bacteria to play a role in fungus
digestion has not been considered.
In this study we combine information on the identity and location of symbiotic bacteria
throughout the Cephalotes digestive tube to gain insight into their nutritional roles at the individual and colony levels. Our specific goals were to (1) determine the number of bacterial meta-communities that inhabit the digestive tract of Cephalotes, (2) to test the hypotheses regarding the nutritional roles of the Cephalotes gut microbiota, and (3) to develop a new hypothesis model on ant-gut symbiosis based on the information about function and location of the gut microbiota. Gut compartments have been sampled separately before in Cephalotes varians, but because of the small sample size (n = 1 healthy colony), the authors advised results within gut compartments to be considered preliminary (Kautz et al. 2013). In the present study we add more information about the compartmentalization of gut microbes in C. varians, with a larger sample size (n=6 colonies). In addition, we include the infrabuccal pocket in our sampling, a gut compartment that has not previously been surveyed for bacteria, but that has been directly associated with pollen collection and regurgitation of food to larvae (Urbani & de Andrade
1997). In order to verify the generality of our results, we also compared our results to previously published data on C. rohweri (Lanan et al. 2016) and on C. varians (Kautz et al. 2013). In this study we also introduce a curated database for classification of Cephalotes-specific microbes, which uses a combination of data published from previous culturing and 16S rRNA cloning experiments and phylogenetic information for improving the classification of next-generation sequence data.
77
METHODS
Sample collection, dissections, DNA extraction and sequencing
Six colonies of Cephalotes varians were collected in the Florida Keys (FL, USA) and preserved in 95-100% ethanol (EtOH), in a -80o C freezer. From each colony we sampled five minor worker and carefully dissected out each of the following sections of the gut: infrabuccal pocket, crop, midgut, ileum, and rectum. All samples were surface sterilized prior to dissections, by gently holding an individual ant with a forceps and repeatedly dipping it into a bleach solution
(6%) for 1 minute, followed by two washes in distilled water (1 minute each) and a final wash in
EtOH for another minute. To avoid further contamination and cross contamination, each ant was dissected in a droplet of distilled water and as each organ was removed, the remaining organs were transferred to a new droplet. If at any step an organ ruptured, the whole set of organs was discarded and a new individual was dissected. Dissecting tools were sterilized in a bleach solution (6%) at every step of the dissection and the solution was replaced for every 5 individuals that were dissected. Each organ was placed in an individual 1.5mL tube that was kept in the -80
C freezer. In addition, we also dissected the second pair of legs of each individual to serve as a control for microbes not removed by our surface sterilization method. These leg samples and additional blank samples containing only distilled water were used as control samples, to decrease contaminants that could not be otherwise accounted for.
We extracted the DNA of all samples using a DNeasy Blood and Tissue Kit (Qiagen), following their protocol for Gram-positive bacteria. Samples were submitted to the Argonne National
Laboratory, prepared and sequenced using the 16S protocol for surveying bacterial communities as established by the Earth Microbiome Project. The amplification primers used were 515F 78
(GTGYCAGCMGCCGCGGTAA) and 806R (GGACTACNVGGGTWTCTAAT), and the sequencing was done on a 2x150bp Illumina MiSeq lane.
Sequence quality control
Sequences were assembled and processed using Mothur and following its standard procedure protocol for sequences generated by the MiSeq platform (Kozich et al. 2013). In summary, sequences were discarded if they were not within 251-255bp, or if they showed homopolymers equal or larger than 10bp, or if they presented ambiguous bases. Sequences were aligned using the Mothur-formatted version of the Silva Database (v123), and screened and filtered to contain only the 16S V4 region, targeted by the primers 515F-806R. Next, we removed sequences with homopolymers larger than 8 bases and chimeric sequences identified by the implementation of
UCHIME in Mothur. Sequences were clustered, and classified by Mothur’s implementation of the Bayesian RDP classifier, using a curated RDP training set (see below). We used a bootstrap cutoff of 80% and all sequences that could not be classified or were found to belong to lineages other than Bacteria were excluded. To decrease the presence of contaminants, we compared sequences present in gut tissues against sequences that were present in the control samples.
Sequences were removed if they met the following conditions, in order: (1) a sequence in the control samples was found in an experimental library at a relative abundance of 10% or more; (2) a sequence represented less than 0.1% of the total number of sequences in an experimental library; (3) a library lost more than 90% of its sequences during these first two filtration steps.
Filtration steps were automated by a python script developed by Piotr Lukasik and Jon Sanders.
Re-analyses of sequences published in other studies 79
In addition to the sequencing data generated in this study, we also downloaded previously
published C. varians microbiota data published by Kautz et al. (2013), and C. rohweri microbiota data published by Lanan et al. (2016). Although these papers were not focused on investigating microbiota partitioning along the gut compartments of Cephalotes, both studies surveyed the gut microbiota in gut compartments individually. We quality-filtered the data from these studies according to the parameters specified below, which allowed only long (more than
200bp) and high-quality sequences to be used in downstream tests and comparisons.
Kautz et al. (2013) sampled different gut compartments for three C. varians colonies: CSM
1280, CSM1235 and CSM 1323. Unfortunately, the latter colony was discarded because reads in these samples mostly matched Spiroplasma, which is hypothesized to be a pathogen of this ant
(Kautz et al. 2013; Lanan et al. 2016). Therefore we only used data available for colony
CSM1280, which included head (2 samples), crop (2 samples), midgut (2 samples), and hindgut
(2 samples), and data from colony CMS1235, which included only one head sample. These samples were amplified using a primer targeting the V1-V3 region of the 16S rRNA gene, 28F
(GAGTTTGATCNTGGCTCAG) and 519R (GTNTTACNGCGGCKGCTG), and sequenced with 454 pyrosequencing technology. Accession numbers were: SRS372496, SRS372498,
SRS372499, SRS372500, SRS372501, SRS372502, SRS372503, SRS372504, SRS372505.
Using Mothur, we quality-filtered these sequences following the “flowgrams-route” of the 454 standard operational procedure, that keeps higher quality and longer sequences compared to the alternative method using quality scores. Similar to the pipeline used above for the data we generated, the data from Kautz et al. (2013) were also trimmed, aligned, chimera-filtered, 80
classified and cleaned of sequences other than Bacteria. Specific parameters used in the 454 and
MiSeq Mothur pipelines are available in the Supplementary Material.
Lanan et al. (2016) sampled C. rohweri gut compartments from 7 colonies as follows: 4 crop samples, 6 proventriculus samples, 7 midgut samples, 7 ileum samples and 3 rectum samples.
Similarly to the Kautz et al. (2013) dataset, samples were amplified with the primers 28F-519R and sequenced with 454 pyrosequencing technology. An identical procedure as described above was adopted for quality-filtering sequences of this study.
Gut partitioning analyses
Core gut microbiota and compartment-specific bacteria
To describe the partitioning in the bacterial communities within the guts of C. varians and
C.rohweri, we first clustered their sequences at 97% similarity, for each dataset separately, using the average-neighbor algorithm as implemented in Mothur. These sequences were then classified using the Mothur implementation of the Bayesian RDP classifier and the CS-RDP training set with a bootstrap cut-off value of 80%. CS-RDP is our implementation of a customized RDP database containing Cephalotes-specific training set (for further details, see supporting methods in the Supplemental Material). To define the core gut microbiota of Cephalotes we used the criteria of Hu et al. (2014): an OTU was considered part of the gut microbiota if (1) it was found in more than two-thirds of the colonies, and (2) if its relative abundance across all libraries was at least 4%. Differently from Hu et al. (2014), we applied these filters to each gut compartment separately. Another additional criterion is that before applying these filters, we rarefied each gut compartment, to account for differences in sequencing depth among samples within each gut compartment. Sequences that did not meet criterion (2) were grouped and reported as “Other”. 81
Partitioning description and testing
To determine similarity among gut compartments in each study, we first generated a principal
coordinates analyzes plot using generalized UniFrac distances (alpha=0.5) (Chen et al. 2012)
from sequences subsampled to the smallest sample. Similarly to Lanan et al. (2016), we grouped
field-collected and lab-reared colonies of C. rohweri, as this variable is not explanatory of the variation observed in microbiota composition. Next, we performed a PermanovaG test, using variance-adjusted UniFrac distances (d=0, d=0.5, and d=1), to determine how much of the variance observed could be explained by colony identity or gut compartment identity. In addition to, we run a metacommunity analysis to determine the number of individual communities that can be identified along the gut. For this approach we used the Dirichlet Multinomial Mixture model (DMM) (Holmes et al. 2012), which determines metacommunities or enterotypes in the absence of metadata information regarding sample origin. Our method is mostly based on the strategy adopted by Ding and Schloss (2014). Using Mothur, we run DMM using the get.communitytype command on subsampled datasets. Because of the small number of samples, these datasets were pre-processed to combine OTU with the same taxonomy into genus-level phylotypes. This approach decreases the number of possibly spurious OTUs, highlighting differences due to abundant taxa, but at the cost of excluding the contribution of rare species to community composition. To determine the number of community types, we ran the analysis 5 times for each dataset and selected the replicate with the smallest Laplace value.
In order to generate further hypotheses about the compartment-specificity of microbes along the gut of Cephalotes, we built heatmaps for the core microbiota of each Cephalotes species.
Functional analyses 82
Functional profiles of the Cephalotes gut compartments were predicted with PICRUST v.1.0.0
(Langille et al. 2013). PICRUST uses the relationship between phylogenetic placement and
conserved metabolic function to predict the functional profile of bacterial communities based on
their identity as inferred by the 16S metabarcoding. Although Langille et al. (2013) have found
that PICRUST in some cases can be superior to shotgun metagenomes in inferring a microbiome
functional profile, we exercised caution with regard to the conclusions that we drew from the
results. PICRUST is a powerful tool, but relies on published bacterial genomes, which are mostly
composed by medically relevant and environmental bacteria, which may not accurately represent
the metabolic potential of bacteria that evolved in association with Cephalotes. PICRUST
provides a metric, weighted Nearest Sequenced Taxon Index (NSTI), that quantifies how closely a given sample is related to the available reference genomes used by PICRUST. In general, an
NSTI score of 0.03 or less indicates that in average reference genomes are 97% similar or more to the bacterial community found in a sample. For PICRUST, relatively accurate predictions for humans, mammals and even environmental samples were obtained for NSTI values near or higher than 0.05, ranging in accuracy of approximately 70% to 85% (Langille et al. 2013). We restrict our data analyses and discussion to the inference of general hypothetical patterns, with focus in gut compartments that show a NSTI score of 0.05 or less.
We first tested if functional profiles were different along different gut compartments. For each dataset, we built biom files in Mothur, using our Silva-aligned sequences and Greengenes taxonomy (gg_13_8), as required by PICRUST. The biom-file generated in Mothur was then subsampled in Qiime (Caporaso et al. 2010) before further analysis. Next, we followed the general procedure of metagenome prediction in PICRUST, which starts with normalizing the 83
number of 16S copies, followed by metagenome prediction, and then collapsing the number of functional predictions in broader categories. The output file of PICRUST presents the abundance of gene ortholog copies associated with each category of metabolic pathway as defined in the
Kyoto Encyclopedia of Genes and Genomes (KEGG) database (Kanehisa et al. 2016). PICRUST outputs a table with over six thousand KEGG gene orthologs that can be collapsed to more general pathways. When collapsed to level 3, the number of KEGG categories is decreased to
328, but contains gene orthologues of difficult interpretation, such as “cancer” or
“photosynthesis”, especially in the context of insect gut microbes. We filtered the KEGG output table to contain only metabolic pathways relevant to the generation and testing of hypothesis related to nutrition and or host-microbe interactions. To test if functional profiles differ along the digestive tube, we calculated a dissimilarity matrix using the Bray-Curtis distance. For this test, we selected 62 categories that were related to nutrition (e.g. “nitrogen metabolism”) or potential microbe-host interaction (e.g. “protein export”). The list of KEGG categories is shown in
TableS2 (Supplementary Material). Next, we visualized the differences among gut compartments by using a hierarchical clustering analyzes, taking the dissimilarity matrix and calculating its clusters with the agglomeration method “ward.D2”. To test if differences were significant, we calculated a permanova test, using gut compartments as the explanatory variable.
In order to explore specific association between function and compartments, we also selected pathways that have been implicated to be present in Cephalotes-gut microbe interactions, as well as in well-studied insect-symbiont nutritional mutualisms. Pathways associated with the ability to synthesize essential amino acids, as well as the ability to synthesize carotenoids, were selected from level 3 of collapsed KEGG categories. The ability to fix nitrogen, the ability to recycle 84
nitrogen (synthesis of urease), the ability to consume pollen through pectin digestion (synthesis
of pectinase), and the ability to consume fungus (synthesis of chitinase) were all manually
selected based on pathway information available in the KEGG database. In addition, we selected
pathways to the synthesis of vitamins such as B1, B2, B6 and B12, which together with vitamin
A (a carotenoid) are among nutrients believed to be required for successful rearing of ants
(Bhatkar & Whitcomb 1970). Because of the exploratory nature of this analysis, we only visually
contrasted the distribution of genes associated to these pathways in both C. varians and C. rohweri. A complete list of categories and their associated KEGG orthologs is available in the
Table S3 (Supplementary Material).
Bioinformatics and statistical packages
Quality control, alignment, clustering of sequences, and classification, as well as the meta- community analysis DMM, were all done in Mothur (v1.37.0) (Schloss et al. 2009). Statistics and graphs were done in R with the following packages: ade4, ape, GUniFrac, Hclust,
RColorBrewer, and vegan. MEGA 6 was used to generate maximum-likelihood phylogenetic trees. PICRUST was used in the Huttenhower’s instance of the Galaxy platform.
Data accessibility
Upon publication, sequencing data will be deposited in the SRA archive. All tests conducted in
R, as well as the codes used for generating graphs will be available in www.github.com/PedroDaPos/new-repo. Upon publication, datasets will be deposited in
Figshare.
RESULTS 85
The gut microbiota in Cephalotes varians and C. rohweri
A total of 858,187 raw reads were generated in our sampling of the Cephalotes varians gut microbiota. All blank samples contained reads, indicating that the reagents or distilled water were contaminated. After running our decontamination script, all blank samples were removed and experimental libraries had lost an average of 45 ± 12.15% from the total of each library. The total number of sequences left was 286,583. After processing for high-quality sequences,
283,991 reads were kept, with a median length of 253bp. We will refer to this dataset as “Cv16”.
These sequences were grouped into 118 OTUs (97%), among which only 21 presented abundances equal or higher to 4% within a gut compartment (Figure 1A). Fourteen core gut bacteria were identified, belonging to the orders Burkholderiales, Campylobateriales, Opitutales,
Pseudonomadales, Rhizobiales, and Sphingobacteriales (Figure 1A). Flavobacteriales, also considered to belong to the C. varians core microbiota, were classified as “other”, as their relative abundance was smaller than our cut-off of 4%. All sequences classified as other are provided in Figure S1 (Supplementary Material). In our re-analyses of C. varians gut compartment samples from Kautz et al. (2013) we retained 13,737 sequences after quality- filtering an initial batch of 14,250 sequences. We will refer to this dataset as “CvK”. The median length of sequences was 242 bp and similarly to the Cv16 dataset all core groups were also retained (Figure 1C). For C. rohweri, our re-analyses of data from Lanan et al. (2016) yielded
154,137 sequences, after quality filtering 157,745 raw reads, with a median length of 242 bp, and containing representatives of all groups considered to belong to their core microbiota (Figure
1E). Our rarefaction curves show good sampling effort (Figure S2, Supplementary Material).
Gut microbiota compartmentalization 86
The variation observed in the diversity of gut communities was best explained by their location
in the gut (Cv16, F4,29 = 11.906, P = 0.001 for organ location, and P = 0.154 for colony ; C. rohweri, F4,26 = 9.504, P = 0.001 for organ location, and P = 0.067, for colony), except for the
CvK dataset, in which both location and colony were significant explanatory variables
(F3,8=1.76, P = 0.001 for organ location, and P = 0.007 for colony).
When compared, samples appeared to cluster in three distinct groups (Figure 1B-D-F). The results from our metacommunity analysis are consistent with this pattern, as the best fit model for each dataset also resulted in three partitions (Table 1). In general, the first partition groups mostly communities belonging to the foregut, i.e., the crop and proventriculus. The second partition is dominated by midgut samples. Finally, the third partition contains ileum and rectum samples (Table 1). A few exceptions were found: all partitions of the CvK dataset were heterogeneous, containing a mix of two or more organs. In Cv16, one midgut sample
(D_YH091_midgut) was grouped with hindgut samples, and one rectum sample
(D_YH081_rectum) was grouped with crop samples. Infrabuccal pocket samples were found in both partitions 2 and 3 of the Cv16 dataset. A fourth compartment was initially found in the C. rohweri dataset, containing only the proventriculus and crop samples belonging to colony “4b”
(data not shown). These samples were excluded from our final analysis, because at closer inspection we noticed that these samples were dominated by a Bacilliales strain (Figure 1E), hypothesized to have been acquired with the nectar collected by Cephalotes (Lanan et al., 2016).
When we re-analyzed the Cv16 data without these two samples, only three groups were found, consistent with the results found in C. varians. 87
Figures 1 and 2 show that the midgut is one of the least diverse gut compartments, whereas the
hindgut compartments tend to contain most core microbes. In both C. varians and C. rohweri, members of Cephaloticoccus (Verrumicrobia) are the most abundant bacteria in midgut samples, followed in C. varians by a Campylobacteriacea. One midgut sample from the CvK dataset, however, is an exception to this pattern, as it contained almost as many phylotypes as hindgut samples (sample CMS1280_3; Figure 1C). Focusing on the distribution of Rhizobiales, we found evidence that suggests that groups in this order specialize either in the foregut or in hindgut compartments. In C. varians, RhizobiA1_2 (Cv16) and an unclassified Rhizobiales (OTU 001)
(CvK) are primarily found in the crop, and absent in most midgut and hindgut samples (Figure
2A-B). On the other hand, in hindgut samples other Rhizobiales strains are dominant, such as an unclassified Rhizobiales (OUT 006) (Cv16), RhizobiA2_1 (CvK) and RhizobiA2_3 (CvK). A similar pattern is also true for C. rohweri: two unclassified Rhizobiales (OTU 003 and OTU 004) are only found in the crop and proventriulus, whereas a RhizobiaA2_1 is found in hindgut compartments (Figure 2C). It is noteworthy that RhizobiA2_1 was found only in CvK and C. rohweri, in the same region of the gut. These two studies were also amplified and sequenced using the same set of primers and sequencing technology, which suggests that similar lineages of
Rhizobiales may specialize in the same region of the guts of different Cephalotes species.
Members of the Xanthomonadales, Burkholderiales, Flavobacteriales, Pseudomonadales and
Sphingobacteriales are almost exclusively found in hindgut samples (Figure 2A-C). For C. varians (Cv16) we also observed that the second most rich compartment in number of core microbe species are samples belonging to the infrabuccal pocket, which mostly resemble the composition of hindgut samples, but also contains the crop specialist RhizobiaA1_2. In the CvK 88
dataset, this pattern was similar, when excluding the midgut sample CMS1280_3, an unusual
sample when compared to other midgut samples, in the same study and in our study. Head
samples, which contain the infrabuccal pocket, contained both hindgut and foregut specialists in
CvK.
Functional compartmentalization
Using PICRUST, we made predictions regarding the potential functional profile(s) of the
different gut communities found in Cephalotes. In general, NSTI values were smaller than 0.05
(Table S5, Supplementary Material), which indicates that the predictions made by PICRUST are
based on genomes that share, on average, 95% similarity or higher to the bacteria identified in
Cv16, CvK and the C. rohweri datasets. A surprisingly consistent pattern was found in these
different datasets: upper gut compartments, such as crop and proventriculus, showed the smallest
NSTI values, ranging from a crop sample (CvK) with NSTI = 0.012 to a maximum NSTI value
of 0.034 found in a proventriculus sample from the C. rohweri dataset. Samples with the highest
NSTI values belonged to the hindgut: the smallest value found was an ileum sample (C. rohweri) with NSTI = 0.039 and the highest was another ileum sample with NSTI = 0.086 (Cv16). Based on these values, we are confident that the predictions made by PICRUST can be trusted for spotting patterns that are sufficient for generating hypotheses. However, we refrain from making definitive statements about the functional nature of the gut microbiome, as PICRUST is not a substitute for more accurate methods such as shotgun metagenome analysis or transcriptome analysis.
Focusing on a selected number of pathways relevant to nutritional mutualism and microbe-host interactions, we calculated a dissimilarity matrix using Bray-Curtis distances and used 89
hierarchical clustering to identify similarity the different gut sections (Figure 3). Upper gut
compartments, such as crop and proventriculus, tended to show a functional profile more similar
to each other than to lower gut samples, such as midgut and hindgut. Head and infrabuccal
pocket samples tended to cluster together with upper gut samples, although some infrabuccal
pocket samples were also observed in the lower gut grouping in Cv16 (Figure 3A). As noted
earlier, two samples of C. rohweri were contaminated by a Lactobacillales bacterium. These samples (marked with an asterisk in Figure 3C) formed a distinct grouping, distant from all other gut compartments. To evaluate whether these groupings are masking functional redundancy or similarity among the different gut compartments, we used the Bray-Curtis distances matrix to test how much of the variation observed is explained by gut location. Using permanova, we found that gut location is a significant explanatory variable in the C. rohweri dataset (F4,26, P =
0.039) and in Cv16 (F4,29, P < 0.001), but does not explain the variation observed in CvK (F3,8,
P = 0.218).
Next, we investigated specific hypothesis regarding functional abilities of microbes associated with Cephalotes. Because of the exploratory nature of PICRUST predictions and the inconsistencies observed in the CvK dataset, we chose to focus this analysis only on well- sampled datasets, such as Cv16 and the C. rohweri dataset. Nitrogen recycling, nitrogen fixation, and digestion of uric acid (bird droppings) are some of the main functions that have been attributed to be among the roles played by the Cephalotes microbiota. In general, KEGG orthologs (KO) associated with these functions are the least abundant in hindgut samples, for both Cv16 and C. rohweri. In fact, these pathways are relatively more enriched in midgut and crop samples, although these differences are more prominent in C. varians than in C. rohweri 90
(Figure 4A). Another consistent pattern observed in both datasets is the high abundance of KO
associated with breakdown of uric acid within the infrabuccal pocket and crop samples in Cv16,
and within crop and proventriculus samples in the C. rohweri dataset.
Whether microbes are able to digest pollen was evaluated through the presence and abundance of
KO associated with the biosynthesis of pectinases. Since yeast hyphae have been observed before in the guts of Cephalotes and are also reported to be present in infrabuccal pellets of ants, we included in our analysis the potential for fungus digestion, illustrated by the ability to synthesize chitinase. There was a markedly higher abundance of KO associated with these enzymes in the midgut and hindgut samples than in other compartments, for both Cephalotes species (Figure 3B). In the uppergut compartments, presence was very low and spotty among crop and proventriculus samples. On the other hand, infrabuccal pockets showed relatively low abundance of pectinase and chitinase related genes, but were more consistently found across samples.
Finally, our results support the hypothesis that gut microbes of Cephalotes are able to synthesize essential amino acids and vitamins. All gut compartments of C. varians showed similar abundance of pathway genes associated with essential amino acid biosynthesis, whereas C. rohweri midgut and hindgut samples showed a greater abundance of these genes compared to foregut samples (Figure 4D). For vitamins a similar pattern was found, except that the microbiota of both C. varians and C. rohweri showed a higher abundance of KO associated with vitamin
B12 synthesis in the uppergut samples compared to midgut, ileum and rectum (Figure 4D). For
C. varians, B6 and carotenoids synthesis also followed a similar tendency, whereas in C. rohweri 91
the abundance of KO associated with these compounds was more comparable across all compartments.
DISCUSSION
The diversity and abundance of exudate-feeding ants have been hypothesized to be associated with the evolution of a specialized microbiota capable of synthesizing nutrients and enzymes that their hosts are unable to produce (Cook & Davidson 2006). One approach to understanding patterns of colonization, function and resilience of gut microbes is through investigating their location and identity within the host (Donaldson et al. 2016). We found a consistent pattern of partitioning of the gut microbiota in two species of Cephalotes (Figures 1 and 2), comparable to that observed in highly specialized herbivores. We also demonstrated that the digestive tube of
Cephalotes can be divided into three enterotypes or metacommunities (Table 2): (1) a foregut community, composed of crop and proventriculus, and dominated by foregut-specific lineage(s) of Rhizobiales; (2) a midgut community, dominated by Cephaloticoccus; and (3) a hindgut community, the most diverse community, containing nearly all members of the core microbiota of Cephalotes. The compartmentalization of the microbiota does not necessarily imply that different communities have different roles, as functional redundancy cannot be ruled out without further investigation of their metabolic potential. Nonetheless, by using PICRUST and KEGG orthologs, we found suggestive evidence that the compartmentalization of the microbiota does reflect distinct functional profiles (Figure 3). Understanding the location and identity of bacteria has also allowed us to evaluate the plausibility of their potential nutritional role. Below, we summarize and discuss our main findings. PICRUST predictions are to be considered educated 92
guesses, as its reliability is limited to available, published reference metagenomes which do not
include Cephalotes-specific studies. Microbes in the gut of Cephalotes, nevertheless, showed high similarities with the reference genomes, ranging from 92% to 99% overlap in the 16S rRNA identity. These results can therefore be used to generate hypotheses that warrant further investigation.
Crop- and hindgut- specialized Rhizobiales
Despite the use of different sequencing technologies and primer sets, we found that members of
Rhizobiales tend to be consistently found in specific compartments, in both C. varians and C. rohweri. In Cephalotes varians, the crop is dominated by at least one strain of Rhizobiales:
RhizobiaA1_2 in Cv16 and an unclassified Rhizobiales (OTU 001) in CvK (Figure 2A-B). Two strains are found in the crop and proventriculus of Cephalotes rohweri (Figure 2C). The crop is considered the social stomach of ants and other social insects, where liquid food is stored and shared among nestmates, including larvae. Possibly because of regurgitation (oral trophallaxis), bacteria found in the crop were also detected in the infrabuccal pocket and head of C. varians
(Figure 2A-B). We hypothesize that bacteria found in the foregut of Cephalotes is exchanged among nestmates, along with liquid food. Therefore, it is possible that foregut bacteria are involved in processes at the level of the colony, such as larval nutrition. The microbial community of the crop and proventriculus showed similarity of over 95% with reference genomes in PICRUST, indicating high reliability (Langille et al. 2013). In particular, the crop showed the highest number of KEGG orthologs associated with the breakdown of uric acid
(Figure 4A). Cephalotes are known to collect bird droppings, and this finding suggests that the conversion of uric acid to urea and or ammonia may take place in this organ. In colonies of ants, 93
developing larva and the queen are the individuals with the highest demand for protein
(Dussutour & Simpson 2009). To date, larvae of Cephalotes have been shown to house unspecific groups of bacteria, which are similar to environmental bacteria found in the nest
(Lanan et al. 2016). We hypothesize that crop-specific Rhizobiales might be involved in the conversion of uric acid to more tractable forms that can be digested and absorbed by larva.
Hindgut-specific Rhizobiales are represented by an unclassified Rhizobiales (OTU 006) in Cv16 and RhizobiA2_1 and RhizobiA2_3 in CvK. In C. rohweri, RhizobiA2_3 was found primarily in the ileum and rectum samples. Due to its posterior location, a particularly relevant function for hindgut microbiota regards the ability to recycle nitrogen. Malpighian tubules, the functionally equivalent to kidneys in insects, collect urea from the hemolymph that is next transferred to the ileum and rectum. Bacteria present in these compartments are therefore the most likely candidates to participate in the recycling of nitrogen in Cephalotes. In addition to the hindgut- specific Rhizobiales lineages, members of Burkholderiales, Xanthomonadales, Pseudomonadales and Flavobacteriales were also almost exclusively found in the hindgut of Cephalotes (Figure
2A-C). We found that genes associated with urea digestion are indeed predicted to be present in this compartment (Figure 4A).
Rhizobiales are hypothesized to fix nitrogen, but in our study this ability was found to be possible throughout the gut (Figure 4A). In fact, this pathway was particularly enriched in the midgut, where members of the clade Verrumicrobiales are abundant. Similarly to members of the
Rhizobiales clade, Verrumicrobiales have also been demonstrated to be able to fix nitrogen
(Khadem et al. 2010), but it is unclear where the levels of N2 would be highest in the crop, midgut or in the hindgut. In our dissections and as described in the literature (Roche & Wheeler 94
1997; Bution & Caetano 2008; Bution & Caetano 2010) we observed that the ileum is
particularly well connected to tracheae. Most likely, this organ must experience a high metabolic
rate, requiring higher oxygenation to sustain its massive microbial load.
The infrabuccal pocket microbiota reflects the composition of hindgut microbiota
The role of the infrabuccal pocket in the nutrition of ants has been largely overlooked. Previously
considered to function primarily as a filter to prevent solid particles from reaching the crop
(Eisner & Happ 1962), for Cephalotes and related genera it has also been assumed to serve as a
storage pouch where solid particles such as pollen are collected and compacted in a pellet that is
fed to larvae (Wheeler 1984; Urbani & de Andrade 1997). Although pollen has never been
detected in the guts of C. rohweri and C. varians, it has been hypothesized that the specialized gut microbiota of Cephalotes might be involved in digesting the recalcitrant pollen walls
(Russell et al. 2009; Hu et al. 2014). A central challenge with this hypothesis is the determination of a mechanism by which gut microbes would enter into contact with the pollen.
Cephalotes is likely unable to ingest pollen past the crop due to the very fine filtering structure of its proventriculus. Nevertheless, we found that hindgut bacteria are able to enter the infrabuccal pocket. In fact, this organ showed a mixture of hindgut and crop-specific bacteria (Figures 1 and
2), which might derive from both abdominal and oral trophallaxis. Based on these results, we predict that if the Cephalotes microbiota do participate in the digestion of pollen, then most likely this digestion takes place in the infrabuccal pocket. PICRUST predicts that pathways associated with the breakdown of pectin and chitin are both abundantly present in the midgut and hindgut, but largely absent in crop samples. These genes were also consistently found across samples of infrabuccal pocket, although in very low levels. There is therefore a potential role of 95
midgut and hindgut bacteria in the digestion of pollen and fungal cells, but crop bacteria may not contribute directly to this process.
The presence of hindgut microbes in the infrabuccal pocket indicates a novel, nutritional role for abdominal trophallaxis. This behavior, previously thought to only contribute to the inoculation of newly emerged workers, might also involve the inoculation of infrabuccal pellets. The low levels of bacteria in the infrabuccal pocket and the fact that infrabuccal pellets are frequently regurgitated (Wilson 1976; Wheeler 1984; De Andrade & Urbani 1999) may suggest that the microbiota in this organ are frequently depleted. In addition, exposure to environmental microorganisms and possibly pathogens is likely detrimental to the maintenance of the infrabuccal pocket microbiota in the long term. We therefore propose that this organ is likely frequently seeded with hindgut bacteria that are necessary for digestion of solid particles present in the infrabuccal pellet. Wilson (1976) has observed that abdominal trophallaxis is performed not only among callows and workers, but also among older workers. Older workers are already colonized by core microbiota and further gut colonization is unlikely, as the proventriculus is known to filter out particles the size of most bacteria (Lanan et al. 2016). Therefore, it is plausible that abdominal trophallaxis among older workers explains the presence of hindgut bacteria in the infrabuccal pocket. Further investigation of the frequency of this behavior as well as a metagenomics survey of genes present in the infrabuccal pellet are necessary for testing this proposed mechanism.
Synthesis of essential amino acids, B vitamins and their importance to larval rearing
Adult ants in Cephalotes are able to survive for months under laboratory conditions without intake of protein. In queenright colonies, workers show undeveloped or underdeveloped ovaries 96
and, as other holometabolous insects, they do not grow in body size after pupation. Therefore,
workers have the lowest requirement for protein intake when compared to the queen or growing
larvae. Cephalotes, like most eukaryotes, are unable to synthesize essential amino acids, which are necessary for their growth. Deficiencies in vitamins in the B group have also been associated with lower success in growth and survival of insects (Fraenkel & Blewett 1943; Hosokawa et al.
2010). We hypothesize that the Cephalotes microbiota may depend on the synthesis of these
nutrients by their associated, co-evolved core microbiota. Insects are believed to require ten
essential amino acids (Zientz et al. 2004). We found that all compartments sampled show genes
associated with their synthesis. In order to infer the role of microbes in synthesizing these
nutrients we also must consider their source(s) of nitrogen. As mentioned previously, Cephalotes
are known to collect bird droppings, rich in uric acid. We earlier hypothesized that the synthesis
of uricases might be particularly important in the crop, as oral trophallaxis would be the most
direct route for sourcing nitrogen to growing larva. Following this scenario, we hypothesize that
by-products of uricases such as urea and ammonia might reach the midgut and hindgut, where
the synthesis of different amino acids could take place. As suggested earlier, hindgut bacteria
may also recycle nitrogen from the host waste, which would represent an additional source of
nitrogen. Moreover, the hindgut is the most diverse organ in Cephalotes in terms of its
microbiota and might also contain the most diverse set of pathways that would be required for
the synthesis of all ten essential amino acids. Similarly, KEGG orthologs associated with the
biosynthesis of a selected number of vitamins were found to be present in similar abundance
across all organs. The only category that shows a pattern of compartmentalization in both
Cephalotes species is the synthesis of vitamin B12, more abundantly represented in the foregut 97
and infrabuccal pocket, compared to the rest of the gut. Although vitamin B12 is present in
artificial diets for rearing ants, it is not believed to be a dietary requirement for insects (Dadd
1973). It is also not known if Cephalotes is affected by the deficiency of this vitamin. On the
other hand, the microbiota itself might be dependent on the availability of this vitamin for
survival. The microbiota of phytophagous insects and even the human microbiota, for instance,
are negatively affected by deficiencies in vitamin B12 (revised in Degnan et al., 2014).
Despite the limitations listed above, we found evidence that suggest that the gut microbiota of
Cephalotes is likely capable of synthesizing essential amino acids and vitamins, which are particularly important for the growth of larva and the colony itself. We emphasize that regardless of the site of synthesis, these nutrients should be present in the liquid food exchanged with larva via oral trophallaxis. Because of the higher diversity of the hindgut bacteria, we predict that the transfer of bacteria and or nutrients from the hindgut to the crop and infrabuccal pocket must be responsible for upgrading the liquid food that is given to growing larva. Nutrient transfer via behaviors similar to coprophagy is not uncommon. Termites, for instance, have been observed to acquire protein via proctodeal trophallaxis, where protein deprived workers performed this behavior more frequently than well-nourished control workers (Machida et al. 2001).
Coprophagy is also observed among vertebrates with a herbivorous diet, where nutrient-rich feces are reingested for absorption of nutrients, including vitamins and amino acids generated by gut microbes (Hirakawa 2001).
CONCLUSIONS AND FUTURE DIRECTIONS 98
Cephalotes shares specializations in the host and microbiota that are comparable to highly
specialized herbivory found in other animals. Based on our results on the localization of the
microbiota, we were able to generate hypotheses about the possible nutritional roles of symbionts
at both the individual and colony-levels. These hypotheses are complemented by PICRUST
predictions on the potential metabolic routes that the microbiota should show based on traits
shared by closely related taxa of bacteria where metagenomics information is available. In
addition to uncovering new hypotheses for the roles of microbes to the nutrition of ants, the
information on location of some microbes helped us develop new hypotheses for known
biological traits of Cephalotes. For instance, we propose a novel interaction between infrabuccal pocket and abdominal trophallaxis that may result in the generation and transfer of nutrients important for larval growth. A summary of our findings, as well as new proposed mechanisms involved in the nutritional ecology of Cephalotes, is presented in Figure 5. In this study we also developed CS-RDP, a curated database that organizes and provides reference sequences and names to bacterial symbionts commonly found in two species of Cephalotes.
Further investigations using metagenome and metatranscriptome of the different partitions of the gut may shed light on the mechanisms proposed here on the interaction of gut microbes and nutrition of Cephalotes.
REFERENCES
Anderson K.E., Russell J.A., Moreau C.S., Kautz S., Sullam K.E., Hu Y., Basinger U., Mott B.M., Buck N. & Wheeler D.E. (2012). Highly similar microbial communities are shared among related and trophically similar ant species. Mol. Ecol., 21, 2282-96. Ben-Yosef M., Pasternak Z., Jurkevitch E. & Yuval B. (2014). Symbiotic bacteria enable olive flies (Bactrocera oleae) to exploit intractable sources of nitrogen. J. Evol. Biol., 27, 2695-705. 99
Bennett G.M. & Moran N.A. (2015). Heritable symbiosis: The advantages and perils of an evolutionary rabbit hole. Proc Natl Acad Sci U S A, 112, 10169-76. Bhatkar A. & Whitcomb W. (1970). Artificial diet for rearing various species of ants. Florida Entomologist, 229-232. Bignell D.E. (2011). Morphology, Physiology, biochemistry and functional design of the termite gut: an evolutionary wonderland. In: Biology of Termites: a Modern Synthesis (eds. Bignell ED, Roisin Y & Lo N). Springer Netherlands Dordrecht, pp. 375-412. Bignell D.E. (2016). The role of symbionts in the evolution of termites and their rise to ecological dominance in the tropics. In: The Mechanistic Benefits of Microbial Symbionts (ed. Hurst JC). Springer International Publishing Cham, pp. 121-172. Blüthgen N., Verhaagh M., Goitía W., Jaffé K., Morawetz W. & Barthlott W. (2000). How plants shape the ant community in the Amazonian rainforest canopy: the key role of extrafloral nectaries and homopteran honeydew. Oecologia, 125, 229-240. Bolton B. (2013). An Online Catalog of the Ants of the World. Brune A. & Ohkuma M. (2011). Role of the termite gut microbiota in symbiotic digestion. In: Biology of Termites: a Modern Synthesis (eds. Bignell ED, Roisin Y & Lo N). Springer Netherlands Dordrecht, pp. 439-475. Bution M.L. & Caetano F.H. (2008). Ileum of the Cephalotes ants: A specialized structure to harbor symbionts microorganisms. Micron, 39, 897-909. Bution M.L. & Caetano F.H. (2010). Symbiotic bacteria and the structural specializations in the ileum of Cephalotes ants. Micron, 41, 373-81. Caetano F. & da Cruz-Landim C. (1985a). Presence of microorganisms in the alimentary canal of ants of the tribe Cephalotini (Myrmicinae: Location and relationship with intestinal structures). Naturalia, 10, 37-47. Caetano F.H. (1989). Endosymbiosis of ants with intestinal and salivary gland bacteria. In: Insect Endocytobiosis: morphology, physiology, genetics, evolution. (eds. Schwemmler W & Gassner G). CRC Press Boca Raton, p. 272. Caetano F.H. & da Cruz-Landim C. (1985b). Presence of microorganisms in the alimentary canal of ants of the tribe Cephalotini (Myrmicinae): location and relationship with intestinal structures. Naturalia, 10. Caporaso J.G., Kuczynski J., Stombaugh J., Bittinger K., Bushman F.D., Costello E.K., Fierer N., Pena A.G., Goodrich J.K., Gordon J.I., Huttley G.A., Kelley S.T., Knights D., Koenig J.E., Ley R.E., Lozupone C.A., McDonald D., Muegge B.D., Pirrung M., Reeder J., Sevinsky J.R., Turnbaugh P.J., Walters W.A., Widmann J., Yatsunenko T., Zaneveld J. & Knight R. (2010). QIIME allows analysis of high-throughput community sequencing data. Nature methods, 7, 335-6. Chen J., Bittinger K., Charlson E.S., Hoffmann C., Lewis J., Wu G.D., Collman R.G., Bushman F.D. & Li H. (2012). Associating microbiome composition with environmental covariates using generalized UniFrac distances. Bioinformatics, 28, 2106-13. Cook S.C. & Davidson D.W. (2006). Nutritional and functional biology of exudate-feeding ants. Entomol. Exp. Appl., 118, 1-10. Dadd R.H. (1973). Insect nutrition: current developments and metabolic implications. Annu Rev Entomol, 18, 381-420. 100
Davidson D. & Patrell-Kim L. (1996). Tropical arboreal ants: why so abundant? In: Neotropical biodiversity and conservation (ed. Gibson A). Mildred E. Mathias Botanical Garden, University of California Los Angeles, CA, USA, pp. 127-140. Davidson D.W., Cook S.C., Snelling R.R. & Chua T.H. (2003). Explaining the abundance of ants in lowland tropical rainforest canopies. Science, 300, 969-972. De Andrade M.L. & Urbani C.B. (1999). Diversity and adaptation in the ant genus Cephalotes, past and present (Hymenoptera, Formicidae). . Stuttgarter Beiträge Naturkunde Ser. B (Geol. Paläontol.). Ding T. & Schloss P.D. (2014). Dynamics and associations of microbial community types across the human body. Nature, 509, 357-60. Donaldson G.P., Lee S.M. & Mazmanian S.K. (2016). Gut biogeography of the bacterial microbiota. Nat. Rev. Microbiol., 14, 20-32. Douglas A.E. (2009). The microbial dimension in insect nutritional ecology. Funct Ecol, 23, 38- 47. Dussutour A. & Simpson S.J. (2009). Communal nutrition in ants. Curr. Biol., 19, 740-4. Eisner T. (1957). A comparative morphological study of the proventriculus of ants (Hymenoptera: Formicidae). Bull. Museum Comp. Zool., 116. Eisner T. & Happ G.M. (1962). The infrabuccal pocket of a formicine ant: a social filtration device. Psyche, 69, 107-116. Engel P. & Moran N.A. (2013). The gut microbiota of insects - diversity in structure and function. FEMS microbiology reviews, 37, 699-735. Feldhaar H., Straka J., Krischke M., Berthold K., Stoll S., Mueller M.J. & Gross R. (2007). Nutritional upgrading for omnivorous carpenter ants by the endosymbiont Blochmannia. BMC biology, 5, 48. Fraenkel G. & Blewett M. (1943). The vitamin B-complex requirements of several insects. Biochem. J., 37, 686-692. Hirakawa H. (2001). Coprophagy in leporids and other mammalian herbivores. Mammal Rev., 31, 61-80. Hölldobler B. & Wilson E.O. (1990). The ants. Harvard University Press, Cambridge, Mass. Holmes I., Harris K. & Quince C. (2012). Dirichlet multinomial mixtures: generative models for microbial metagenomics. Plos One, 7, e30126. Hosokawa T., Koga R., Kikuchi Y., Meng X.-Y. & Fukatsu T. (2010). Wolbachia as a bacteriocyte-associated nutritional mutualist. Proc. Natl. Acad. Sci. USA, 107, 769-774. Hu Y., Lukasik P., Moreau C.S. & Russell J.A. (2014). Correlates of gut community composition across an ant species (Cephalotes varians) elucidate causes and consequences of symbiotic variability. Mol. Ecol., 23, 1284-300. Joy J.B. (2013). Symbiosis catalyses niche expansion and diversification. Proc Biol Sci, 280, 20122820. Kanehisa M., Sato Y., Kawashima M., Furumichi M. & Tanabe M. (2016). KEGG as a reference resource for gene and protein annotation. Nucleic Acids Res., 44, D457-62. Kautz S., Rubin B.E., Russell J.A. & Moreau C.S. (2013). Surveying the microbiome of ants: comparing 454 pyrosequencing with traditional methods to uncover bacterial diversity. Appl Environ Microbiol, 79, 525-34. 101
Khadem A.F., Pol A., Jetten M.S. & Op den Camp H.J. (2010). Nitrogen fixation by the verrucomicrobial methanotroph 'Methylacidiphilum fumariolicum'. Microbiology, 156, 1052-9. Klein A., Schrader L., Gil R., Manzano-Marin A., Florez L., Wheeler D., Werren J.H., Latorre A., Heinze J., Kaltenpoth M., Moya A. & Oettler J. (2016). A novel intracellular mutualistic bacterium in the invasive ant Cardiocondyla obscurior. Isme J, 10, 376-88. Kohler T., Dietrich C., Scheffrahn R.H. & Brune A. (2012). High-resolution analysis of gut environment and bacterial microbiota reveals functional compartmentalization of the gut in wood-feeding higher termites (Nasutitermes spp.). Appl Environ Microbiol, 78, 4691- 701. Kozich J.J., Westcott S.L., Baxter N.T., Highlander S.K. & Schloss P.D. (2013). Development of a dual-index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the MiSeq Illumina sequencing platform. Appl Environ Microbiol, 79, 5112-20. Lanan M.C., Rodrigues P.A., Agellon A., Jansma P. & Wheeler D.E. (2016). A bacterial filter protects and structures the gut microbiome of an insect. Isme J, 10, 1866-1876. Langille M.G., Zaneveld J., Caporaso J.G., McDonald D., Knights D., Reyes J.A., Clemente J.C., Burkepile D.E., Vega Thurber R.L., Knight R., Beiko R.G. & Huttenhower C. (2013). Predictive functional profiling of microbial communities using 16S rRNA marker gene sequences. Nat. Biotech., 31, 814-21. Lin J.Y., Hobson W.J. & Wertz J.T. (2016a). Ventosimonas gracilis gen. nov., sp. nov., a member of the Gammaproteobacteria isolated from Cephalotes varians ant guts representing a new family, Ventosimonadaceae fam. nov., within the order 'Pseudomonadales'. Int.J. Syst. Evolut. Microb. Lin J.Y., Russell J.A., Sanders J.G. & Wertz J.T. (2016b). Cephaloticoccus gen. nov., a new genus of 'Verrucomicrobia' containing two novel species isolated from Cephalotes ant guts. Int.J. Syst. Evolut. Microb.. Machida M., Kitade O., Miura T. & Matsumoto T. (2001). Nitrogen recycling through proctodeal trophallaxis in the Japanese damp-wood termite Hodotermopsis japonica (Isoptera, Termopsidae). Insectes Soc., 48, 52-56. Mankowski M.E. & Morrell J.J. (2004). Yeasts associated with the infrabuccal pocket and colonies of the carpenter ant Camponotus vicinus. Mycologia, 96, 226-231. McBee R.H. (1971). Significance of Intestinal Microflora in Herbivory. Ann. Rev. Ecol. System., 2, 165-176. Roche R.K. & Wheeler D.E. (1997). Morphological specializations of the digestive tract of Zacryptocerus rohweri (Hymenoptera: Formicidae). Journal of Morphology, 234, 253- 262. Russell J.A., Moreau C.S., Goldman-Huertas B., Fujiwara M., Lohman D.J. & Pierce N.E. (2009). Bacterial gut symbionts are tightly linked with the evolution of herbivory in ants. Proc Natl Acad Sci U S A, 106, 21236-41. Sanders J.G., Powell S., Kronauer D.J., Vasconcelos H.L., Frederickson M.E. & Pierce N.E. (2014). Stability and phylogenetic correlation in gut microbiota: lessons from ants and apes. Mol. Ecol., 23, 1268-83. Schloss P.D., Westcott S.L., Ryabin T., Hall J.R., Hartmann M., Hollister E.B., Lesniewski R.A., Oakley B.B., Parks D.H., Robinson C.J., Sahl J.W., Stres B., Thallinger G.G., Van Horn 102
D.J. & Weber C.F. (2009). Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl Environ Microbiol, 75, 7537-41. Schmitt-Wagner D. & Brune A. (1999). Hydrogen profiles and localization of methanogenic activities in the highly compartmentalized hindgut of soil-feeding higher termites (Cubitermes spp.). Appl Environ Microbiol, 65, 4490-6. Stevens C.E. & Hume I.D. (1998). Contributions of microbes in vertebrate gastrointestinal tract to production and conservation of nutrients. Phys. Rev., 78, 393-427. Stoll S., Gadau J., Gross R.O.Y. & Feldhaar H. (2007). Bacterial microbiota associated with ants of the genus Tetraponera. Biol.J.Linn.Soc., 90, 399-412. Tamura K., Stecher G., Peterson D., Filipski A. & Kumar S. (2013). MEGA6: Molecular Evolutionary Genetics Analysis version 6.0. Mol. Biol. Evol., 30, 2725-9. Urbani B.C. & de Andrade L.M. (1997). Pollen eating, storing, and spitting by ants. Naturwissenschaften, 84, 256-258. Wheeler D.E. (1984). Behavior of the ant, Procryptocerus scabriusculus (Hymenoptera: Formicidae), with comparisons to other cephalotines. Psyche, 91, 171-192. Wheeler W.M. & Bailey I.W. (1920). The feeding habits of Pseudomyrmine and other ants. T. Am. Philos. Soc., 22, 235-279. Wilson E.O. (1976). A social ethogram of the neotropical arboreal ant Zacryptocerus varians (Fr. Smith). Anim. Behav., 24, 354-363. Wolcott D.L. & O'Connor N.J. (1992). Herbivory in crabs: adaptations and ecological considerations. American Zoologist, 32, 370-381. Zientz E., Dandekar T. & Gross R. (2004). Metabolic interdependence of obligate intracellular bacteria and their insect hosts. Microb. Mol. Biol. Rev., 68, 745-70.
103
FIGURES
(a) Cephalotes varians (Cv16) (b) pocket crop midgut ileum rectum 1.00
0.75 n o i t r
o 0.50 p o r P
0.25
0.00 YH081 YH089 YH091 YH092 YH097 YH113 YH081 YH089 YH091 YH092 YH097 YH113 YH081 YH089 YH091 YH092 YH097 YH113 YH081 YH089 YH091 YH092 YH097 YH113 YH081 YH089 YH091 YH092 YH097 YH113
Burkholderiales BurkhoC_1 Enterobacteriales Cephaloticoccus capnophilus Rhizobiales Unclassified (Otu006) XanthoA2_1 Alcaligenaceae Unclassified (Otu008) Campylobacteriales Salmonella Pseudomonadales Sphingobacteriales Unclassified Bacteroidetes BurkhoA1_1 Campylobacteraceae Unclassified (Otu005) Methylophilales Ventosimonadaceae Unclassified (Otu009) SphingoA_1 Bacteroidetes Unclassified (Otu021) BurkhoA2_2 Clostridiales Methylobacillus Rhizobiales Xanthomonadales BurkhoB_4 Anaerococcus Opitutales RhizobiaA1_2 XanthoA1_1 Other
(c) Cephalotes varians (CvK) (d) head crop midgut hindgut 1.00
0.75 n o i t r
o 0.50 p o r P
0.25
0.00 CMS1235 CMS1280_2 CMS1280_3 CMS1280_2 CMS1280_3 CMS1280_2 CMS1280_3 CMS1280_2 CMS1280_3
Burkholderiales Arcobacter varians Cephaloticoccus capnophilus Rhizobiales Sphingobacteriales Unclassified Proteobacteria BurkhoB_3 Flavobacteriales Pseudomonadales RhizobiA2_1 SphingoA_1 Proteobacteria Unclassified (Otu006) BurkhoC_1 FlavoA_1 Pseudomonas RhizobiA2_3 Xanthomonadales Campylobacteriales Opitutales Ventosimonas gracilis Rhizobiales incertae sedis Unclassified (Otu001) XanthoA1_1 Other
(e) Cephalotes rohweri (f)
crop proventriculus midgut ileum rectum 1.00
0.75 n o i t r
o 0.50 p o r P
0.25
0.00 1b 2b 3b 4b 1a 1b 2b 3b 4b 7a 1a 1b 2b 3b 4b 6a 7a 1a 1b 2b 3b 4b 6a 7a 2b 3b 4b
Burkholderiales BurkhoA1_2 Rhizobiales Xanthomonadales Other Alcaligenaceae Unclassified (Otu0009) BurkhoD_1 RhizobiA2_1 Dyella Xanthomonadaceae Unclassified (Otu0006) Alcaligenaceae Unclassified (Otu0011) Ralstonia Rhizobiales incertae sedis Unclassified (Otu0003) Alcaligenaceae Unclassified (Otu0013) Rhizobiales incertae sedis Unclassified (Otu0004) Unclassified Bacilli Opitutales Bacilli Unclassified (Otu0002) Burkho other 1 Cephaloticoccus primus
104
Figure 1. Bacterial diversity along the gut of Cephalotes. Relative abundance of bacteria is shown for Cephalotes varians (A and C) and Cephalotes rohweri (E). Gut location for each sample is provided at the top portion of the barplot graph, whereas colony identity is provided in the x-axis. Proportions are calculated based on the total number of reads recovered for each sample after quality-filtering. Only groups of bacteria with relative proportion greater than or equal to 4% are represented. Groups with relative proportion smaller than 4% are grouped as “other”, represented in gray. Groups of bacteria belonging to the core microbiota are highlighted in bold and represent bacteria that were present in at least 2/3 of the samples, per gut compartment. Principal Coordinate plots depict the grouping of samples based on UniFrac distances (alpha=0.5), for C. varians (B and D) and C. rohweri (F). Ellipses are drawn around the centroid of the multidimensional scaling distances calculated for samples belonging to the same gut compartment. When sample size is small (n=2 samples), the ellipses are stretched into a line connecting samples. “d” values refer to the length of each square in the grid, generated by multidimensional scaling based on the dissimilarity among samples.
105
(a) (b)
C. varians (Cv16) C. varians (CvK) t t t t u m u e u p d m p k u g g g a o t u o c d r d d r i c e e i o c l n c i e i h p m r m h
Alcaligenaceae Unclassified (Otu008) BurkhoB_3
BurkhoC_1 BurkhoA1_1
Arcobacter varians BurkhoA2_2
FlavoA_1
BurkhoB_4
Cephaloticoccus capnophilus
Campylobacteraceae Unclassified (Otu005)
Proteobacteria Unclassified (Otu006)
Cephaloticoccus capnophilus
Ventosimonas gracilis
Ventosimonadaceae Unclassified (Otu009) RhizobiA2_1
RhizobiaA1_2 RhizobiA2_3
Rhizobiales Unclassified (Otu006) Rhizobiales incertae sedis Unclassified (Otu001)
SphingoA_1 XanthoA1_1
XanthoA1_1
XanthoA2_1 2 3 2 3 2 3 2 3 5 ______3 0 0 0 0 0 0 0 0 2 8 8 8 8 8 8 8 8 1 2 2 2 2 2 2 2 2 1 9 1 2 7 3 1 9 1 2 7 3 1 9 1 2 7 3 1 9 1 2 7 3 1 9 1 2 7 3 1 1 1 1 1 1 1 1 S 8 8 9 9 9 1 8 8 9 9 9 1 8 8 9 9 9 1 8 8 9 9 9 1 8 8 9 9 9 1 S S S S S S S S 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 M M M M M M M M M H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H C Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y C C C C C C C C Colony Colony
(c)
C. rohweri s u l u t c u m i p m r u g t o t u d r n c i e c l e i e m r v o r p
Relative Abundance Alcaligenaceae Unclassified (Otu0009) 1.00
Alcaligenaceae Unclassified (Otu0011) 0.75
0.50 Alcaligenaceae Unclassified (Otu0013)
0.25 Burkho other 1
0.00 BurkhoA1_2
Cephaloticoccus primus
RhizobiA2_1
Rhizobiales incertae sedis Unclassified (Otu0003)
Rhizobiales incertae sedis Unclassified (Otu0004)
Xanthomonadaceae Unclassified (Otu0006) b b b b a b b b b a a b b b b a a a b b b b a a b b b 1 2 3 4 1 1 2 3 4 7 1 1 2 3 4 6 7 1 1 2 3 4 6 7 2 3 4 Colony
Figure 2. Hetmaps of the relative abundance and location of the core microbiota. (A) and (B) represent compartments and bacteria sampled for C. varians, in two different studies (Cv16 and CvK); (C) represent samples from C. rohweri. Black and dark shades of gray indicate middle to 106
high relative abundance, whereas light shades of gray represent low abundance, and the lightest shade indicated absence.
107
(a) Cephalotes varians (Cv16) rectum ileum rectum ileum ileum ileum rectum ileum ileum rectum pocket midgut midgut rectum rectum midgut midgut midgut midgut pocket pocket pocket pocket crop crop pocket crop crop crop crop
(b) Cephalotes varians (CvK) midgut
hindgut
hindgut
head
midgut
head
crop
head
crop
(C) Cephalotes rohweri ileum ileum ileum midgut midgut midgut midgut midgut midgut midgut rectum ileum rectum rectum ileum ileum ileum proventriculus proventriculus proventriculus proventriculus proventriculus crop crop crop proventriculus* crop*
Figure 3. Similarity of the functional profile of bacterial communities according to their location in the gut. Cladograms are plot based on hierarchical clustering analysis, using Bray-distances to distinguish samples based on the abundance of a selected subset of collapsed KEGG categories (level 3). The agglomeration method used is “ward.D2”.
108
Nitogen Metabolism (a)
16000 16000 infrabuccal pocket crop midgut ileum rectum crop proventriculus midgut ileum rectum 14000 14000 12000 12000 t t n n u u o 10000 o 10000 c c e e n n e 8000 e 8000 g g
O O
K 6000 K 6000 4000 4000 2000 2000 0 0 1b 2b 3b 4b 1a 7a 1b 2b 3b 4b 1b 2b 3b 4b 1a 6a 7a 1a 6a 7a 1b 2b 3b 4b 2b 3b 4b 1 9 1 2 7 3 1 9 1 2 7 3 1 9 1 2 7 3 1 9 1 2 7 3 1 9 1 2 7 3 8 8 9 9 9 1 8 8 9 9 9 1 8 8 9 9 9 1 8 8 9 9 9 1 8 8 9 9 9 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y
______Sample D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D Sample
Cephalotes varians Cephalotes rohweri Uricase Urea recycling Nitrogen fixation
Pollen and Fungus digestion (b)
5000 infrabuccal pocket crop midgut ileum rectum 5000 crop proventriculus midgut ileum rectum 4500 4500 4000 4000 3500 3500 t t n n u u o o 3000 3000 c c
e e n n 2500 2500 e e g g
2000 2000 O O K K 1500 1500 1000 1000 500 500 0 0 1 1 1 1 1 1 1 1 1 1 9 2 7 3 9 2 7 3 9 2 7 3 9 2 7 3 9 2 7 3 1b 2b 3b 4b 1a 7a 1b 2b 3b 4b 1b 2b 3b 4b 1a 6a 7a 1a 6a 7a 1b 2b 3b 4b 2b 3b 4b 8 9 8 9 8 9 8 9 8 9 8 9 9 1 8 9 9 1 8 9 9 1 8 9 9 1 8 9 9 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y ______
______Sample D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D Sample
Cephalotes varians Cephalotes rohweri Pectinase Chitinases
Essential amino acid biosynthesis (c)
250000 250000 infrabuccal pocket crop midgut ileum rectum crop proventriculus midgut ileum rectum
200000 200000 t t n n u u o 150000 o 150000 c c e e n n e e g g 100000 100000 O O K K
50000 50000
0 0
1 9 1 2 7 3 1 9 1 2 7 3 1 9 1 2 7 3 1 9 1 2 7 3 1 9 1 2 7 3 1b 2b 3b 4b 1a 7a 1b 2b 3b 4b 1b 2b 3b 4b 1a 6a 7a 1a 6a 7a 1b 2b 3b 4b 2b 3b 4b 8 8 9 9 9 1 8 8 9 9 9 1 8 8 9 9 9 1 8 8 9 9 9 1 8 8 9 9 9 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y
______Sample D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D Sample
Cephalotes varians Cephalotes rohweri Arginine biosynthesis Methionine biosynthesis Phenylalanine biosynthesis Valine, leucine and isoleucine Histidine biosynthesis Threonine biosynthesis Tryptophan biosynthesis biosynthesis
Vitamin biosynthesis (d)
50000 infrabuccal pocket crop midgut ileum rectum 50000 crop proventriculus midgut ileum rectum 45000 45000 40000 40000 t t
n 35000 n 35000 u u o o c 30000 c 30000 e e n n e 25000 e 25000 g g
O 20000 O 20000 K K 15000 15000 10000 10000 5000 5000 0 0 1 1 1 1 1 1 1 1 1 1 9 2 7 3 9 2 7 3 9 2 7 3 9 2 7 3 9 2 7 3 8 9 8 9 8 9 8 9 8 9 8 9 9 1 8 9 9 1 8 9 9 1 8 9 9 1 8 9 9 1 1b 2b 3b 4b 1a 7a 1b 2b 3b 4b 1b 2b 3b 4b 1a 6a 7a 1a 6a 7a 1b 2b 3b 4b 2b 3b 4b 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y ______Sample D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D Sample
Cephalotes varians Cephalotes rohweri Vitamin B1 Vitamin B6 Vitamin A and Vitamin B2 Vitamin B12 other carotenoids
109
Figure 4. Distribution and abundance of metabolic pathways that have been hypothesized to be present in Cephalotes. Barplots depict the abundance of selected KEGG orthologs as predicted by PICRUST. Cephalotes varians refer to predictions based on the Cv16 dataset. C. rohweri refer to predictions based on the Lanan et al. (2016) dataset. “Sample” information refers to the colony source.
TABLE
Table 1. Meta-community partition as modeled by Dirichlet Multinomial Mixture Model analysis. The number of partitions is the same across the three different datasets used in this study (Cv16, CvK, and C. rohweri (Lanan et al., 2016). Identity of the partition is not the same across samples, because datasets were analyzed separated.
C. varians (Cv16) C. varians (CvK) C. rohweri (Lanan et al.) Partition Organ Sample Organ Sample Organ Sample 1 Midgut D_YH081_midgut Head CMS1235_head Midgut mg_1 Midgut D_YH089_midgut Crop CMS1280_crop_2 Midgut mg_2 Midgut D_YH092_midgut Midgut CMS1280_midgut_2 Midgut mg_3 Midgut D_YH097_midgut Midgut mg_4 Midgut D_YH113_midgut Midgut midgut_1 Midgut midgut_6 Midgut midgut_7 2 Crop D_YH081_crop Crop CMS1280_crop_3 Ileum hg_1 Crop D_YH089_crop Head CMS1280_head_2 Ileum hg_6 Crop D_YH091_crop Head CMS1280_head_3 Ileum hg_7 Crop D_YH092_crop Ileum il_1 Crop D_YH097_crop Ileum il_2 Crop D_YH113_crop Ileum il_3 Pocket D_YH097_pocket Ileum il_4 Pocket D_YH113_pocket Rectum rectum_2 Rectum D_YH081_rectum Rectum rectum_3 Rectum rectum_4 3 Ileum D_YH081_ileum Hindgut CMS1280_hindgut_2 Crop crop_1 Ileum D_YH089_ileum Hindgut CMS1280_hindgut_3 Crop crop_2 Ileum D_YH091_ileum Midgut CMS1280_midgut_3 Crop crop_3 Ileum D_YH092_ileum Proventriculus prov_1 Ileum D_YH097_ileum Proventriculus prov_7 Ileum D_YH113_ileum Proventriculus pv_1 Rectum D_YH089_rectum Proventriculus pv_2 110
Rectum D_YH091_rectum Proventriculus pv_3 Rectum D_YH092_rectum Rectum D_YH097_rectum Rectum D_YH113_rectum Pocket D_YH081_pocket Pocket D_YH089_pocket Pocket D_YH091_pocket Pocket D_YH092_pocket Midgut D_YH091_midgut
111
Individual Level Colony Level
Oral Trophallaxis
By-products of solid food digestion (pollen and fungus) Nutrients acquired
s
e
d d l By-products of uric acid digestion i i
l d c via foraging (directly) i
o u
t o
r
r
e q S
f o
i o g
a
f from food pellet
o
u L p
d r
g
p Lipids (?) i
t
e a
l t
l
i e
o
Carbohydrates (?) t
n
from nectar and trophobiont exudates Carbohydrates O r
Minerals a l
Microbiota Ip Vitamins (?) T
r Growing larva Metabolism Amino acids o p
via gut microbes (indirectly) h Uricase and Pectinase a l l Digestion of foo d a Urease x i pellet (pollen and s B12 biosynthesis fungal spor es) Carbohydrates p N2 Fixation (?) Lipids Urea recycling and Urease Abdominal digestion of uric acid B vitamins biosynthesis Trophallaxis Essential amino acids Carotenoid biosynthesis Other nutrients Essential amino acids Vitamins (B complex) biosynthesis Carotenoids (Vitamin A) Transfer of midgut and hi ndgut symbionts Vitamins, caroteno ids and essent ial amino acids
Figure 5. Hypothetical scenario of the interaction between microbes and the nutrition of adults and brood in colonies of Cephalotes. On the left, a worker is represented next to the routes of liquid and solid particles, which indicate ingestion (arrows pointing down) and regurgitation (arrows pointing upwards). Gut compartments are also represented in the same individual: infrabuccal pocket (Ip), pharynx (Phr), crop (C), midgut (M), Malpighian tubules (Mt), ileum (Il) and rectum (R). The proventriculus location is indicated by a dashed line, between the crop and the midgut, highlighting its role in separating solid particles and bacteria between the crop and midgut, allowing only the passage of liquid food. Metabolic processes are also represented in relation to the position of the proventriculus: above the dashed line are processes believed to take place in the crop and infrabuccal pocket; below the dashed line are functions associated with the midgut and hindgut microbiota. In this model it is uncertain if nitrogen fixation is present in Cephalotes and, even if present, it is uncertain whether it would be represented in compartments above or below the proventriculus. For this reason, nitrogen fixation is represented with a question mark and on top of the dashed line. On the right side, colony-level processes are represented. A worker head represents nestmates in a colony (minor and major workers), receiving liquid food, nutrients and bacteria via oral and abdominal trophallaxis, represented by arrows connecting both sides of the figure. The infrabuccal pocket is also represented (Ip), where particle food may be seeded with hindgut bacteria via abdominal trophallaxis. A breakdown of nutrients acquired by workers is listed between the worker head and a representation of growing larvae of the colony. The transfer of nutrients from older workers to larvae is represented by arrows. Illustration made by PAPR; worker figures inspired by photographs of C. varians from www.antweb.org (specimens FMNHINS0000062892 and CASENT0103759).
112
SUPPLEMENTAL MATERIAL
SUPPORTING METHODS
Custom RDP database
For comparing studies that use different primers and or sequencing technology, we developed a
custom RDP training set that can help establish a common identity for the short sequences
generated by NGS technology. Culturing and cloning experiments have produced many partial or
nearly-full length sequences (~700-1400 bp) of the 16S rRNA gene that were found in gut
tissues of both C. varians and C. rohweri (e.g. Anderson et al., 2012). Next-generation sequencing technology, on the other hand, yields shorter sequences, at around 250bp long in the studies investigated here. Therefore, the correct classification of these short sequences depends of the availability of a matching longer sequence in a reference database. For C. rohweri and C. varians these longer sequences often lack genus-level or species level taxonomical information associated with their records at the NCBI Gen Bank so researchers have preferred to use higher taxonomical levels such as Order. While this approach has allowed researchers to identify a core microbiota associated with Cephalotes, specific inferences of their function, as well as cross- study comparisons, are difficult because often there is more than one phylotype of bacteria in the
Orders that have been described to be associated with Cephalotes.
Anderson et al. (2012) published a phylogeny of the gut microbiota found in Cephalotini ants and proposed names for its main clade groups. In their study, they used a large dataset of sequences from cloning and culturing experiments. Using their data information, we downloaded all sequences that were listed to be associated with the C. rowheri microbiota (90 sequences) and all sequences belonging to the C. varians microbiota (596 sequences). In addition, we also 113
downloaded sequences from another cloning experiment of gut microbes in C. varians (25 sequences), reported in Hu et al. (2014). More recently, two studies have described two gut bacteria species that are associated with C. varians (Ventosimonas gracilis and Cephaloticoccus capnophilus) and one bacterium associated with the gut of C. rohweri (Cephaloticoccus primus)
(Lin et al. 2016a; Lin et al. 2016b). Their 16S sequences were also downloaded and included in the reference database. A complete list of accession numbers for these sequences is available in
Table S1.
Sequences from C. rohweri and C. varians were processed separately. Sequences varied in size, from around 200bp to 1500bp, so the first goal was to select the longest sequences available, while avoiding as many losses as possible. Using Mothur, we aligned the sequences to the Silva database (v123), and trimmed and selected sequences that would cover at least the V1-V3 and
V4 regions of the 16S gene, which are the regions commonly targeted in NGS surveys of
Cephalotes gut microbiome. Next, we used pre.cluster and dist.seqs to reduce our dataset and calculate the distances among our sequences. We clustered sequences that were 99% similar to each other using the average-neighbor clustering algorithm in Mothur, thus defining our operational taxonomical units (OTU). In order to assign proper taxonomy to these sequences, we would require biochemical and morphological description from culture isolates of the Cephalotes gut microbiota, which are mostly unavailable in the published literature, except for three species.
Therefore, we adopted an approach similar to what has been done in investigations involving the gut microbiota of other insects, such as Apis mellifera (e.g. Newton & Roeselers 2012): based on phylogenetic placement and or closely-related and taxonomically described Bacteria, we gave each OTU a unique nickname. This solution is not ideal, as it is based on a relatively small 114
number of sequences that does not represent the entirety of Cephalotes-specific core community
of bacteria, and it may also mask different strains and even different species of bacteria.
Nonetheless, the adoption of a nickname is a step towards being able to refer to voucher
sequences when identifying gut-specific bacteria, allowing more consistent use of phylotype
names when describing the gut microbiota of Cephalotes and a more direct way for comparing findings across different studies.
To define the sequence nicknames, we first built a phylogenetic tree to infer clade names as defined by Anderson et al. (2012). We aligned and quality-filtered untrimmed reference sequences and, for each host species, built a separate maximum-likelihood phylogenetic tree of their gut microbiota, using the model GT+R in MEGA 6 (Tamura et al. 2013). Next, for each
99% OTU cluster, we retrieved untrimmed sequences and selected the longest one to be its OTU representative (voucher). To determine its similarity to previously described bacteria, we queried the NCBI nucleotide database using BLAST and the RDP trainset (v14) using the RDP classifier.
If a sequence matched a previously described bacteria in a published study with similarity over
95% in the NCBI nucleotide database, and also the same genus was identified by the RDP database with a bootstrap value over 95%, then we named that sequence with a nickname containing the consensus bacterial genus name and the epithet of the host Cephalotes species name, separated by an underscore. If this condition was not met, but there was a match of similarity of 95% or a higher to a sequence previously described as specific to the Cephalotes host species in a published study, where the bacteria was cultured, then the name assigned to this entry in the NCBI database was adopted. If none of these conditions was met, then the sequence was named after its clade grouping, according to our phylogenetic tree and the assignment 115
proposed by Anderson et al. (2012), followed by a sequential number that distinguished that
sequence from others that also failed to match a consensus genus name.
The selected representative sequences and taxonomy from both C. varians and C.rohweri microbiotas were added to the respective sequences (fasta format) and taxonomy files of the
Mothur-formatted RDP trainset (v14), which will be referred as the Cephalotes-Specific RDP training set (CS-RDP). To compare the efficiency between RDP and CS-RDP, we quantified the number of sequences that were unclassified among datasets and the number of OTUs that received a new, specific name that matched the curated CS-RDP database. The results of this comparison show a decrease in the number of unclassified sequences when using CS-RDP
(Table S6).
Classification with CS-RDP
Our reference phylogenetic trees mostly reflected the taxonomic assignment as proposed byAnderson et al. (2012), with few exceptions (Supplemental Figures). For C. varians we identified one sample that was possibly mislabeled in the supplemental material from Anderson et al. (2012): FJ477564 is identified as “XanthoA1” in our study, but as “Opitutales clade A1” in
Anderson et al. (2012), and as Xanthomonadales on its own record at the NCBI nucleotide database. For C.rohweri, samples identified as “Bukholderiales clade other” were found to be polyphyletic, being assigned to multiple clades, including PseudoA1, BurkhoB, BurkhoA1, and
BurkhoD. Another group that was not monophyletic in our phylogenetic tree was BurkhoA1.
After quality-filtering sequences, we retained 34 OTU in the C. varians reference database, with representatives from the clades BurkhoA1, BurkhoA2, BurkhoB, BurkhoC, BurkhoD, Burkho A- other, CampyloA, FlavoA, OpituA1, PseudoA2, RhizobiA1, RhizobiA2, SphingoA1, 116
XanthoA1, and XanthoA2. The C. rohweri dataset was considerably smaller and resulted in 10
different OTUs, corresponding to the clades Burkho other, BurkhoB, OpituA3 and PseudoA1. A
representative from each OTU cluster was selected, classified and nicknamed. A table
summarizing the selected sequences as well as their consensus taxonomy and nicknames is
available in Table S4 (Supplementary Material). Representative OTU sequences, as well as their
taxonomy, were then used to build the CS-RDP training set.
The improvement in classification was evaluated by comparing the number of unclassified
sequences at the genus and family levels produced by CS-RDP in contrast to the standard RDP
training set. In all cases, the number of sequences left unclassified was smaller when using CS-
RDP (Table S6). Despite the greater number of OTUs with finer classification, the method still
showed some deficiencies. In C. rohweri, for instance, the number of high quality, near-full
length 16S sequences was so small that many of the core microbiota OTU remained classified
only to the family or order level, such as three different Alcaligenaceae, and two OTU classified
as Rhizobiales (Figure 1E). In C. varians, the greater number of available high-quality sequences resulted in a much higher taxonomical resolution for most of the core groups. We also observed that the different regions of the 16S gene targeted by Cv16 and CvK may have resulted in slightly different taxonomical assignments for the core groups. In contrast, in Cv16, OTU 006 was only classified to the family level, Campylobacteriaceae, and its counterpart in CvK successfully matched the genus level of a voucher sequence, with the temporary nickname
“Arcobacter varians”. The same was also observed for Ventosimonadaceae (OTU 008 in Cv16) and Ventosimonas gracilis in CvK. Finally, when comparing Cv16 and CvK, it is also noticeable
that our classification did not always result in the same genus-level names. For instance, whereas 117
RhizobiA2_1 and RhizobiA2_3 matches were found in CvK, the only Rhizobiales with a genus match to our reference database was RhizobiaA1_2 in Cv16. It is likely that these differences may reflect biased amplification of gut bacteria by the primers used in each of these studies, but we cannot rule out sequencing error or limited sequencing depth, and, in particular, the fact that our reference database does not comprehensively cover all phylotypes found in either C. varians or C. rohweri.
118
SUPPLEMENTAL FIGURES
(A) C. varians Rodrigues 'Other' pocket crop midgut ileum rectum
0.20
0.15 n o i t r o p o r
P 0.10
0.05
0.00 YH081 YH089 YH091 YH092 YH097 YH113 YH081 YH089 YH091 YH092 YH097 YH113 YH081 YH089 YH091 YH092 YH097 YH113 YH081 YH089 YH091 YH092 YH097 YH113 YH081 YH089 YH091 YH092 YH097 YH113 Colony Bacterial Species Acidobacteria Gp1 Unclassified (Otu072) Bacteria Unclassified (Otu088) BurkhoB_3 Chryseobacterium Bradyrhizobium Acidobacteria Gp1 Unclassified (Otu093) Bacteria Unclassified (Otu089) BurkhoB_4 FlavoA_2 Methylobacterium Actinomycetales Unclassified (Otu020) Bacteria Unclassified (Otu101) BurkhoC_1 Flavobacterium RhizobiaA1_2 Actinomycetales Unclassified (Otu052) Bacteria Unclassified (Otu106) BurkhoD_1 Fusobacterium Rhizobiales Unclassified (Otu006) Actinomycetales Unclassified (Otu057) Bacteria Unclassified (Otu109) Burkholderia Gammaproteobacteria Unclassified (Otu028) Gemmobacter Actinomycetales Unclassified (Otu074) Bacteria Unclassified (Otu113) Comamonadaceae Unclassified (Otu025) Gammaproteobacteria Unclassified (Otu068) Paracoccus Actinomycetospora Bacteria Unclassified (Otu117) Comamonadaceae Unclassified (Otu034) Gammaproteobacteria Unclassified (Otu116) Rhodobacteraceae Unclassified (Otu031) Arthrobacter Bacteroides Herbaspirillum Gp3_Unclassified_(Otu061) Rhodobacteraceae Unclassified (Otu058) Brachybacterium Paludibacter Massilia Carnobacterium Rhodobacteraceae Unclassified (Otu112) Corynebacterium Porphyromonadaceae Unclassified (Otu038) Pusillimonas Lactobacillus Selenomonas Kocuria Prevotella Campylobacteraceae Unclassified (Otu005) Streptococcus Veillonella Micrococcus Bdellovibrio Caulobacter Methylobacillus Arcticibacter Rothia Betaproteobacteria Unclassified (Otu039) Phenylobacterium Chondromyces Chitinophaga Alphaproteobacteria Unclassified (Otu063) Betaproteobacteria Unclassified (Otu085) Anaerococcus Myxococcales Unclassified (Otu049) Cyclobacteriaceae Unclassified (Otu043) Alphaproteobacteria Unclassified (Otu108) Aeriscardovia Dorea Marinomonas SphingoA_1 Planococcaceae Unclassified (Otu083) Bifidobacterium Roseburia Marinospirillum Sphingobacteriales Unclassified (Otu047) Staphylococcus Gardnerella Ruminococcaceae Unclassified (Otu111) Cephaloticoccus capnophilus Sphingomonas Bacteria Unclassified (Otu026) Acidovorax Cytophagaceae Unclassified (Otu050) Parcubacteria Unclassified (Otu075) Rhodanobacter Bacteria Unclassified (Otu037) Alcaligenaceae Unclassified (Otu008) Deltaproteobacteria Unclassified (Otu094) Blastopirellula Stenotrophomonas Bacteria Unclassified (Otu070) BurkhoA1_1 Geobacter Acinetobacter XanthoA1_1 Bacteria Unclassified (Otu081) BurkhoA1_2 Enterobacteriaceae Unclassified (Otu030) Pseudomonas XanthoA2_1 Bacteria Unclassified (Otu082) BurkhoA2_2 Salmonella Ventosimonadaceae Unclassified (Otu009)
(B) C. varians Kautz 'Other' head crop midgut hindgut
0.25
0.20
0.15 n o i t r o p o r P
0.10
0.05
0.00 CMS1235 CMS1280_2 CMS1280_3 CMS1280_2 CMS1280_3 CMS1280_2 CMS1280_3 CMS1280_2 CMS1280_3 Colony
Bacterial Species Actinomycetales Unclassified (Otu025) Alcaligenaceae Unclassified (Otu018) BurkhoA2_1 FlavoA_2 RhizobiaA1_2 Brevibacterium Alcaligenaceae Unclassified (Otu024) BurkhoB_3 Flavobacteriaceae Unclassified (Otu067) Rhizobiales incertae sedis Unclassified (Otu001) Corynebacterium Alcaligenaceae Unclassified (Otu030) BurkhoC_1 Lactococcus Rhizobiales incertae sedis Unclassified (Otu012) Propionibacteriaceae Unclassified (Otu027) Alcaligenaceae Unclassified (Otu033) BurkhoD_1 Streptococcus Rhizobiales incertae sedis Unclassified (Otu021) Propionibacterium Alcaligenaceae Unclassified (Otu041) Massilia Legionella Rhizobiales incertae sedis Unclassified (Otu065) Anaerobacillus Bordetella cf HU1 Arcobacter varians Proteobacteria Unclassified (Otu006) SphingoA_1 Bacteria Unclassified (Otu052) BurkhoA1_1 Rheinheimera RhizobiA2_1 XanthoA1_1 Alcaligenaceae Unclassified (Otu016) BurkhoA1_2 FlavoA_1 RhizobiA2_3 XanthoA2_1 119
(C) C. rohweri 'Other' crop proventriculus midgut ileum rectum
0.15 n o i t r
o 0.10 p o r P
0.05
0.00 1b 2b 3b 4b 1a 1b 2b 3b 7a 1a 1b 2b 3b 4b 6a 7a 1a 1b 2b 3b 4b 6a 7a 2b 3b 4b Colony Bacterial Species Actinomycetales Unclassified (Otu0095) Acidovorax Massilia Acinetobacter Rhizobiales incertae sedis Unclassified (Otu0122) Blastococcus Alcaligenaceae Unclassified (Otu0009) Oxalobacteraceae Unclassified (Otu0037) Alkanindiges Rhizobiales Unclassified (Otu0043) Dietzia Alcaligenaceae Unclassified (Otu0011) Oxalobacteraceae Unclassified (Otu0085) PseudoA1_3 Rubellimicrobium Nocardioides Alcaligenaceae Unclassified (Otu0013) Pigmentiphaga Pseudomonas Acidocella Propionibacterium Alcaligenaceae Unclassified (Otu0016) Ralstonia Psychrobacter Flavisolibacter Saccharomonospora Alcaligenaceae Unclassified (Otu0017) Brevundimonas Bradyrhizobium Altererythrobacter Bacillus Alcaligenaceae Unclassified (Otu0045) Clostridiales Unclassified (Otu0089) Hyphomicrobiaceae Unclassified (Otu0077) Blastomonas Staphylococcus Alcaligenaceae Unclassified (Otu0047) Deinococcus Microvirga Leptospira Bacteria Unclassified (Otu0022) Alcaligenaceae Unclassified (Otu0050) Enterobacteriaceae Unclassified (Otu0054) RhizobiA1_1 Dyella Bacteria Unclassified (Otu0048) Alcaligenaceae Unclassified (Otu0057) Flavobacteriaceae Unclassified (Otu0018) RhizobiA2_1 Xanthomonadaceae Unclassified (Otu0006) Bacteria Unclassified (Otu0079) Alcaligenaceae Unclassified (Otu0078) Flavobacterium Rhizobiales incertae sedis Unclassified (Otu0003) Xanthomonadaceae Unclassified (Otu0020) Bacteria Unclassified (Otu0128) Alcaligenaceae Unclassified (Otu0151) Lactobacillales Unclassified (Otu0101) Rhizobiales incertae sedis Unclassified (Otu0004) Xanthomonadaceae Unclassified (Otu0024) Bacteria Unclassified (Otu0136) Alcaligenaceae Unclassified (Otu0157) Streptococcus Rhizobiales incertae sedis Unclassified (Otu0014) Xanthomonadaceae Unclassified (Otu0025) Bacteria Unclassified (Otu0177) Burkho other 1 Cephaloticoccus primus Rhizobiales incertae sedis Unclassified (Otu0026) Xanthomonadaceae Unclassified (Otu0070) Bacteroides BurkhoA1_2 Proteobacteria Unclassified (Otu0069) Rhizobiales incertae sedis Unclassified (Otu0033) Xanthomonadaceae Unclassified (Otu0076) Bacteroidetes Unclassified (Otu0021) BurkhoB_6 Proteobacteria Unclassified (Otu0131) Rhizobiales incertae sedis Unclassified (Otu0034) Xanthomonadaceae Unclassified (Otu0099) Bacteroidetes Unclassified (Otu0118) BurkhoD_1 Proteobacteria Unclassified (Otu0161) Rhizobiales incertae sedis Unclassified (Otu0041) Xanthomonadaceae Unclassified (Otu0126) Bacteroidetes Unclassified (Otu0119) Burkholderia Proteobacteria Unclassified (Otu0163) Rhizobiales incertae sedis Unclassified (Otu0083) Figure S1. Composition of bacterial taxa with relative abundance smaller than 4%
120
C. varians Rodrigues C. varians Kautz 0
3 crop head
ileum 5 crop midgut 3 hindgut pocket midgut
5 rectum 2 0 3 5 0 2 2 s s U U T T O O
0 f f 2 o o
5 r r 1 e e b b m m u u 5 N N 1 0 1 0 1 5 5 0 0
0 5000 10000 15000 20000 0 500 1000 1500 2000 2500 3000
Sequences per Sample Sequences per Sample
C. rohweri 0 7 crop ileum midgut 0
6 proventriculus rectum 0 5 s U 0 T 4 O
f o
r e b 0 m 3 u N 0 2 0 1 0
0 5000 10000 15000 20000 Sequences per Sample
Figure S2. Rarefaction Curves
121
(A)
J Q J J 2 Q 1 J Q 5 J Q 2 A J Q 2 4 J 5 o Q 2 5 5 Q 2 4 h J 5 4 7 2 5 5 t J Q 2 4 5 0 5 4 2 n J Q 5 5 O 2 6 a Q 4 1 7 1 J 2 5 4 3 5 5 8 p Q 5 O X J 2 5 4 5 O A 2 1 i Q 5 4 5 5 t J 2 O p u o 6 p 2 J Q 4 5 1 0 O i 2 5 t A h Q O p i 7 J 2 5 4 5 2 9 O p t u t i u 1 3 J Q 2 4 4 t A n 5 4 8 O p p i u t A 4 5 6 a J Q 4 5 3 O i u 1 2 p i t A 1 Q 4 3 t u 5 2 5 1 O A X 5 u 1 J 5 O p i A 2 J 5 1 0 t 1 Q 2 4 4 O p i u A Q J 4 7 p t 1 7 5 5 i u A 1 Q Q 2 5 8 O p i t 2 4 1 O t u A J 9 J 5 5 0 3 p i u 1 Q 5 2 p t A 1 3 J 4 5 4 0 5 O i u A Q O i t 4 2 7 4 1 3 p t u A 1 1 8 O u 5 J J 2 5 9 1 p i A 1 Q 4 t A 2 Q 5 4 8 9 O p i t u 1 J 2 4 5 O 1 i u A 1 2 t Q Q 5 3 1 O p u A J 5 1 p 1 J Q 2 4 4 3 i p i t A 1 J 5 3 9 O t u u Q 2 2 O i t A 1 J 5 4 0 5 p A u Q 2 1 0 p i A 1 5 4 4 O ut 1 J 2 1 9 O i t 1 J J Q 5 4 8 5 p u A Q Q 1 O p A 1 2 4 9 5 i i t 2 2 5 1 7 O p t u u 1 J 5 5 4 8 i A Q 4 4 5 9 O p t A J 2 i u 1 1 Q 5 5 3 O p t u A 5 3 6 0 i t 1 J 2 4 2 6 O p u A J Q 5 5 O O i t A 1 4 2 p p p u 1 Q 2 3 8 i i i A 1 1 5 t t t 1 1 J 2 7 O u A 1 1 u 1 1 u 1 1 4 A Q 5 1 8 1 1 A A o 1 p A A A A 5 A 4 O o J A 1 2 i 1 A A 1 1 1 A h 0 o o t o J 7 A A Q h o A 5 p u t o 8 o Q t o 6 o h h A i o h o A h 4 A 2 O o t n J t t 5 h t n h h 2 u t 5 5 h 1 t o Q O h t t h o h p t a n 5 A n n 1 a t t 4 n t 1 2 i n h 1 4 p n n h 1 1 J 8 t 1 n 6 a a t X n a n u a n t 1 5 5 O i X Q a 4 a a A 1 1 t 1 1 a 4 A n u a a n A A 1 X X a A 8 X 1 2 p X o J X A 1 5 0 K 2 A 1 X X a O 1 X a o o A A o Q 5 i X X A A 0 O X t h o M J 1 3 4 u 8 4 A 1 p 0 2 1 9 3 t h o o h 1 Q 2 X X h A o o p 9 t 0 5 A 1 3 t t h 1 i 4 6 O 9 9 5 6 5 2 o J t 7 n h h 6 2 6 i 0 8 t o h h u 1 9 n A Q 4 t 2 4 n n t t A 1 3 4 t t 1 p u 5 5 A 5 4 2 h 6 8 1 6 4 a A 6 6 n 6 4 t h A i 5 a a o o 1 0 2 4 A 6 6 a n n n n O t 6 5 o
5 t J 4 4 6 1 7 4 1 4 2 4 a A 0 5 5 u 2 X n o A Q 1 4 h h 1 4 4 X a a X a a 7 p 4 2 h
2 n 3 4 A 5 5 6 7 X t t 2 4 4 o A 5 4 2 X a t h o 1 1 J 2 i 5 O 5 5 A 5 2 t X X t 1 5 a X X C 2 2 4 n n 5 4 h o Q 5 u 2 5 4 6 Q 5 n h 1 Q X 2 5 t O p 2 2 5 4 n t o e 4 A 2 1 J a a h 2 5 X A 6 J 2 2 5 3 A
h J i 2 3 a 1 p 5 Q Q n t h p t Q 4 9 5 u 1 3 3 9 7 2 a n O Q 2 t o o c Q Q F X X 4 2 6 2 4 i Q J J 0 X a n J Q J A 1 t Q 5 a a A Q 6 3 1 6 3 9 0 1 0 J X h Q 1 O p u J J n h A 1 4 J p J 6 2 a t 1 J Q J 4 t o i A J Q 4 5 6 X X Q 2 n 8 p t 4 6 6 1 1 a o A A 1 5 4 9 n 7 u J 6 7 X n h A 2 5 1 J 5 5 o o O i 5 4 4 4 4 1 J t A 6 9 8 0 t h o A 5 4 p u 2 5 4 2 0 X a t a o Q p 2 5 5 9 h 4 5 h A 1 2 5 5 4 5 3 6 9 n h t h o A i 2 5 6 X n X t 1 1 J 2 t Q 2 2 3 0 1 a t 5 8 i l u 1 2 2 5 Q 4 6 4 2 n h o 1 Q 5 3 3 u Q Q 2 4 a n 1 J 6 0 t A 4 O A Q 2 J n A 2 8 s J 5 4 5 6 X a h O J Q Q Q Q 5 1 X a n t o F 5 1 J 4 6 6 2 a A o 5 p Q J J 2 5 2 4 X A 1 J 4 3 p i J J Q 2 6 X a n h t J 5 4 0 9 X o o h 4 5 6 i u 2 5 4 t t O t J Q Q 2 X a h A1 J 7 3 u A Q 2 2 5 6 7 5 6 n t h n Q 7 7 p A J Q J 5 2 6 3 t o O 1 J 2 4 2 4 2 X a 6 i t 1 Q 2 9 0 n n a h A 1 J 2 1 p u J Q 6 5 9 t o Q 5 9 A J Q 5 6 6 5 1 0 X a a X A 4 i t J Q 4 2 5 n h 2 u 1 J 2 4 4 4 6 4 X o J 3 O J 5 4 4 X a t 1 Q 5 7 p A 5 Q 5 5 4 6 8 n h 1 J 2 4 5 1 0.05 Q 2 5 2 5 X t 1 2 0 Q 1 O i t J 2 J 2 5 4 4 a A A 5 9 u Q 2 6 0 6 n o 1 2 4 4 p A1 Q Q 2 5 0 X a o A 5 1 J Q Q 4 6 4 6 h J 4 9 O i t J J 2 6 t h o A 1 Q 1 u J J Q 5 4 5 1 t 1 2 p A J Q 4 4 X n n th o J 2 5 8 O i t 1 2 5 5 2 6 a A Q 4 0 p u J 2 4 1 3 a n th o A 1 2 5 O i A Q 2 Q 6 X a 1 J 5 p t u 1 J 5 2 X n h o A Q 4 4 i A Q Q J 2 4 6 X a t h 1 2 6 5 O t u J J 5 1 n t o A J 5 5 p A 1 Q 7 2 X a n h o A Q 4 9 O i t 1 2 7 1 1 t h o 1 1 2 8 p u J 4 7 X a n t J 5 4 i A Q 4 2 1 3 X a n th A A 1 1 Q 4 4 O t u 1 J J 4 4 2 o A 2 8 p A F 5 1 5 X a n h o A 1 J 5 4 0 i 1 2 5 4 2 X a t h o 1 Q 4 O t u 2 5 4 8 9 4 n t h o A 2 8 p A Q 2 3 9 X n t h o A 1 J 5 7 i t 1 Q 5 4 3 1 3 a a n t o Q 4 4O u J J Q 2 4 9 6 X a n th A 2 8 p A 5 4 3 0 X a n th o 1 J 5 6 i t 1 J Q 2 5 4 X n h 1 Q 4 7O u J 2 5 4 4 5 0 X a a t A A 1 2 6 4 p A 1 Q 2 5 4 9 0 X n o o J 5 5 i t J Q 2 5 3 1 9 2 X a th h A 1 Q 4 7 O u A J Q 2 4 2 8 0 n t o 1 J 2 8 p 1 J Q 5 4 3 3 3 X n h A Q 5 4 6O i t u J Q 5 4 4 1 1 a a t o A 2 5 6 4 p A J 2 2 5 5 4 4 8 X n th o 1 J 4 9 i t u 1 Q 2 2 4 4 6 X a n h Q 8 4 O A J Q Q 5 5 7 3 1 X a t A 1 J 2 5 9 p i 1 J J Q 2 2 4 6 1 X n o A 1 Q 4 O t u J Q 5 7 4 a th o A 1 2 5 8 6 p i A J Q 2 4 2 8 X n h o A 1 J 4 1 t u 1 J Q 5 4 3 6 a t h o Q 8 3 O A 3 J 2 5 8 7 X n nt h A 1 2 5 5 p i 2 4 3 8 a a t o A J Q 4 O t u Q Q 5 4 3 4 X n th o 1 2 7 1 p i A 1 J J 2 5 4 1 3 X a n h 1 5 6 O t u Q 2 5 2 0 2 X a t A J Q 4 5 p i A J Q 2 4 4 9 X n o A 1 2 7 5 O t u A1 J 5 4 3 8 a th o A 1 J 5 4 p 1 Q 2 5 4 9 6 X n th o 1 Q 5 1 i t u J Q 2 5 3 9 2 a n th oA 1 J 2 5 4 O A J Q 2 54 3 0 X a n h A A Q 4 5 p i t 1 J Q 2 54 4 X a nt o o1 2 5 7 2 O u A J Q 2 4 96 7 X a th th J Q 4 5 p 1 J Q 5 6 9 X n n oA 2 4 3 O i t u J 2 4 6 8 Xa a th 1 J 5 4 p A 1 Q 5 4 21 6 n A 1 Q 2 0 i t u J 2 25 4 4 Xa ho A 2 5 6 O A 1 JQ 5 47 15 21X nt ho 1 J 4 3 p i t JQ 2 5 2 6 a t oA 1 Q 2 3 3 O u A 1 Q 2 4 7 8 X anth A 1 J 5 4 p i J Q 25 7 7 9 X n ho A Q 2 5 3 t u A J Q J4 47 8 8 a nt tho 1 J 5 4 4 O p 1 J F 5 43 0 X a n oA1 Q 2 2 0 i t u 2 5 44 0 X a th A 1 5 4 1 O p A 1 JQ 2 5 40 1 X n o J 1 9 i t u JQ 2 4 40 6Xa th oA 1 Q25 8 O A 1 JQ25 4 84 an th J 45 p i t u Q 25 4 4X n uA Q25 31 O A 1 J Q 25 40 Xa1 hpoit J 42 p i J Q 54 83 nOt Q2 02 Op t u A J 2 43 94oA 64a 541 ituA 1 JQ25 3nth 5 JQ 88 Op 1 Q 54a 77 u X 254 ituA J Q2X J4 H JQ 525O 1 J 9 F 01 254 pitu 76 03 JQ2 347O A1 54 73 A2 548 pitu Q2 KFtho 2 JQ2 80O p A1 J an tho A2 A 2 545 ituA X an tho tho A JQ2 81 Op 1 37 X an an tho 2 542 ituA 46 u X X an oA2 JQ2 59 Op 1 25 8 H 7 92 X nth A 542 ituA JQ 28 49 46 38 Xa tho A2 o J 65O p 1 30 54 25 47 an ho nth Q25 ituA F7 Q2JQ 25 758X ant a 2 4261O 1 K J JQ 54 75 X XoA 2 JQ pitu Q2 46 4 Hunth hoA 2542 A1 J 25 66 2X a ant 51O p JQ 254 029 X ntho JQ25 ituA1 JQ 73 47 1 a 2 4552Opitu KF 548 67 u XhoA2 JQ254 A1 Q2 254 0 Hant oA 2 574O pituA J JQ 0293 X anthhoA JQ254 1 F73466 X ant oA2 205O pitu K 25 770 Xanth 2 JQ254 A1 JQ 2544646 X thoAA2 561O pituA JQ 25 11Xan ntho 2 JQ254 1 JQ2547 3 Xa hoA2 791O pituA JQ 547372 Xant oA JQ254 1 JQ2 548 anth 2 506O pituA JQ2 4875X nthoA JQ2545 1 JQ254871Xa hoA2 04O pituA1 Q25 55 Xant 2 JQ25 J 2546 4654 ntho2A 4271O pituA JQ JQ25 9X a thoA 2 JQ2545 1 5472Xan thoA 44O pituA1 JQ24684 Xan A2 JQ25426 JQ25 4617a ntho o 6O pituA1 JQ25 21X anth 2 JQ254 Opitu Q2547 Hu nXthoA2 523 A1 J 730295 X a thoA JQ O KF 254687Xanantho 254752 pituA1 JQ 54878 2 JQ25 Op JQ2 303 HXua nXthoA 4573 ituA1 KF730 59 antho oA2 JQ2 Op JQ2547 6 Hu X 98P seud 54549 ituA1 KF73029 JQ2544 seudoA2 JQ254 OpituA 254499P 2 277 1 JQ P seudoA J O JQ254371 seudo Q254559 pituA1 KF730305 Hu Pseudo 0302 Hu P2 JQ254258Op ituA1 KF73 PseudoA2 JQ25 Opit JQ254374Ps eudoA 4560 uA1 JQ254680 PseudoA2 JQ254542 OpituA1 JQ254741 JQ254330 OpituA1 JQ254767 PseudoA2 OpituA1 KT258896 Ventosimonas gracilis JQ25426 9 JQ254501PseudoA2 OpituA1 JQ Ps 2 JQ254580 254688 eudoA OpituA1 JQ254842P seudoA2 JQ254577 JQ254834P seudoA2 OpituA1 JQ25 Pseu 2 JQ254252 4373 doA ituA1 KF730304 Hu seudo 54376O p JQ25470 Pseu Pdo 2 JQ2 ituA1 JQ 4P A 54256O p 254714 seudoA2 JQ2 ituA1 JQ254747PseudoA2 54263O p JQ25 JQ2 uA1 4862P seudo 182O pit J Q A2 JQ254 tuA1 2 5 4 5 0 0 P s 178O pi e u d o A2 JQ254 1 8O pituA JQ25426 OpituA1 83 JQ2541 OpituA1 556 JQ254 OpituA1 4551 A1 J Q 2 5 4 6 JQ25 Opitu 9 9 B u r k h o C 54553 ituA1 J Q 2 5 4 7 JQ2 O p 0 6 B u r k h o C 5 4 2 7 0 1 J Q J Q 2 p i t u A 2 5 4 4 6 9 B u 6 0 9 O i t u A 1 J Q 2 5 r k h o C F J 4 7 7 O p 4 4 7 2 B u r k 5 4 7 7 5 i t u A 1 J Q 2 5 h o C J Q 2 8 O p 4 4 7 3 B u r k 2 5 4 5 5 i t u A 1 J Q 2 5 h o C J Q 7 O p 4 4 7 1 B u r k 5 4 2 5 A 1 J Q 2 h o C J Q 2 O p i t u A 1 5 4 8 5 B u r 4 5 4 7 O p i t u J Q 2 5 8 k h o C J Q 2 5 8 i t u A 1 4 6 6 6 B u r k 5 4 2 7 O p 1 J Q h o C J Q 2 7 3 p i t u A 2 5 4 4 2 5 4 2 O 1 J Q 7 7 B u J Q 1 9 6 p i t u A 2 5 4 7 r k h o C 2 5 4 O J Q 2 5 B u r J Q 1 9 0 A 1 J Q 2 2 5 4 k h o C 2 5 4 p i t u A 1 J Q 2 K 5 4 4 J Q 2 8 5 2 B J Q O p i t u J Q 5 4 F 7 6 8 B u 5 4 4 7 u r k h o 2 7 5 O 1 2 5 4 7 2 4 3 0 3 r k h o J Q 5 B u C 2 5 4 9 9 i t u A 8 3 B u J Q 2 0 0 D 2 5 4 r k h o J Q 5 4 1 O p J Q 8 B u r k h J 5 4 H u J Q 7 5 6 B C J Q 2 2 0 A 1 2 J Q r k h o D Q 2 6 6 9 B u C o ma m 2 5 4 u r k h 5 4 5 p i t u 1 J Q 5 4 7 2 5 o B 5 4 6 r k h o n a 8 5 1 B o C Q 2 O t u A J 2 5 4 4 B 4 8 J Q 2 6 1 B u o D d a c u r k h J 2 4 9 p i Q 2 4 6 u r 5 0 B 5 4 r k h e a e o C 2 5 4 2 O A 1 F 5 4 8 6 B k h u r k 8 5 5 o D J Q 2 7 p i t u 1 J J 4 7 8 5 u r k o B h o B u 2 5 4 O Q 7 6 3 B u h o B r k h J Q 2 5 5 u A J Q 2 5 1 2 r k h B o D 5 4 p i t A 1 J Q 2 5 4 3 B o B Q 2 8 O i t u 2 5 4 8 0 3 B u r k J 7 6 0 O p A 1 K F 4 3 3 6 B u r k h o J 4 7 8 p i t u A 1 7 3 2 3 B u r k h o B F 4 5 7 O p i t u 0 2 u r k h o B 2 5 5 0 O A 1 9 4 h o B J Q 4 2 9 i t u J Q H B 2 5 4 1 7 O p K 2 u P J Q 2 5 1 A 1 F 5 4 u s J Q 5 7 p i t u A 1 7 3 6 6 i l l i 5 4 O i t u 1 0 3 0 B m o n Q 2 3 O p u A 1 0 u a s J 2 5 6 p i t 1 J H r k h 2 5 4 5 7 O A Q u o B Q 5 4 2 p i t u 1 J Q 2 5 uP J 2 2 6 O A 2 J Q 4 3 s i l l J Q 5 4 0 i t u 1 1 5 4 2 4 i m o Q 2 2 6 O p 1 A u A 6 5 4 2 B n a J 5 4 7 t u A i t u i t 8 1 J Q J Q 3 u r s 2 2 6 p i p O p 1 B u J Q 2 2 2 6 k h J Q 5 4 O O r k 2 5 4 5 4 B u o B 2 6 4 3 4 A J Q h J Q 5 3 3 3 r k J Q 2 3 2 5 t u 2 o B 2 4 3 3 7 6 h o 5 4 4 5 4 p i J 5 5 2 B u B u B Q 2 2 5 2 5 O Q 4 4 6 9 B r r k J Q J Q 0 J Q 2 4 7 5 u r k h h J 6 2 J 2 5 4 8 7 B k h o o B 7 7 Q 5 8 B u r o B J 4 J 2 4 6 4 u r k B F Q 5 7 6 B k h o 2 5 4 7 0 u h o B 4 2 B r k B J Q A 7 7 u r h 2 o 3 1 B k o B 5 v B u r h o 4 F l a A u k h B 8 7 v o r k o 7 B 3 l a A h o B u r 4 9 F o B k h 5 4 2 v o 2 4 9 F l a B J Q 5 4 A 2 8 9 o J Q 4 A l a v A 5 4 o F o 2 v a v J J Q F l a 9 1 F l A J Q 4 4 v o Q J Q J 2 5 6 5 9 5 l a 2 J Q Q 4 4 9 2 4 F A J J Q 5 2 5 2 2 4 4 J Q 4 oA o Q 4 6 4 5 5 7 5 2 5 1 5 v a v A J 2 J 2 5 6 7 4 4 8 6 B Q 2 Q 7 l a l o Q 5 Q 4 7 4 4 5 u J J 4 F F v A 2 4 2 4 J B 9B 7 4 7 B r k 2 5 0 l a m J 5 7 5 7 Q u u B h Q 9 7 F v o A u A Q 4 8 4 9B 2 r k r u u r o J 4 7 a i g o A J K 7 2 7 u 5 h k r k A1 4 4 1 l o r n o J F 2 5 B 4 r 4 o h k h h 2 5 5 6 8 F a v e i g A J Q Q 7 5 7 u 0B k 8 A1 o o o 2 3 l A t h i n o Q 2 3 4 r u h 5 A A A 1 J Q Q 4 7 F o c p h g J 5 J J Q 2 0 8 B k r o 4B 1 1 J 5 6 v b a S p i n e 1 J Q 2 4 K Q J 5 2 6 u h k h A1 u 2 4 3 7 l a A o 7 S h A a i A Q 2 5 8 F 2 Q 2 4 9 5 r k o r Q 0 o 0 e b 1 2 4 5 J 5 6 B h A o k J 2 5 3 F v g 6 p o c A J 5 6 7 J Q 2 4 8 1 u o 1 A1 h 4 l a i n 7 8 S g A a o i 1 J Q 5 4 8 0 3 4 Q 5 5 H r o Q 5 9 F h 0 i n o A A r i i z b 1 1 Q 4 8 6 B 0 4 2 4 7 B k A A J 2 6 A p 7 2 7 g i A A 1 J 2 7 8 u 7 J 2 5 7 6 u h 1 1 Q 3 1 s 4 4 0 h n o o A t e h o i i A J 2 5 7 B 3 0 Q 5 4 4 u A o J 4 1 o i J 5 2 p i g g i z b b i A 1 J Q Q 5 4 5 9 u r k 0 B 6 B r l 5 3 v d F 2 4 S h n o c R o o b Q 4 1 9 2 4 8 6 u k c A1 2 l a u p i n i g A a h z o b A J 2 2 7 B r h u 5 7 4 B h a Q 5 4 F e Q 5 1 h h n b 5 R i i z z o i 1 Q 2 5 7 0 B u k o H r 4 5 8 r k o l s J 2 S i o h i z b 1 K 5 4 3 h A k B u A i g J 2 Q 6 p p h g l o 2 h i i A A J 2 5 4 0 u r o 2 u h 8 4 r h 1 i Q 6 J 7 5 S S p n l 3 4 R h o i 1 J F K 4 7 0 B r k 7 B u k o n J 0 P 4 4 i A A y 4 R h z b Q 5 8 4 k h A u P o 3 u r h A a 3 u 5 7 S h h 4 R i o b i A 1 J Q 7 F 4 1 5 B u h o 2 s A k o 1 c 4 2 4 7 8 p o o 5 3 1 5 R h z o 1 J Q 2 3 7 2 u r B r k h e 5 H 6 4 g g A 2 4 6 i z b i A J Q 2 5 7 3 9 k o A i l o u o A1 2 6 Q 5 3 S n P 3 3 3 R i o i A 1 2 5 0 3 6 5 B B r A 2 l i t r h ea J 2 4 7 6 i i n o A Q 5 3 0 h z b J Q 2 4 3 B u k h 2 m h k o A1 Q 0 4 4 h g u J 2 4 4 5 h i b i A 1 Q 5 4 3 0 5 u h o o e A J 3 Q 5 5 8 0 p h n o A 5 3 2 R o 2 5 4 8 1 2 B u r o r h 1 0 J 4 8 p i H Q 5 3 3 R h z o b A 1 2 6 1 r k A n o 3 2 2 5 S h g o 2 4 6 i z o i 1 5 4 8 4 9 u r k k h A 2 a 7 3 S n 8 J 2 5 4 0 R h i b A 5 4 7 6 H h 2 A Q Q 2 4 p i g Q 5 3 4 z i A 7 1 B 3 r h o s 1 F J J 9 h n 0 J Q 2 4 6 R h i o b i 4 8 1 0 B u k o A K Q 5 2 S p i 3 J 2 3 1 h b 1 1 B u H h o A J 2 8 6 h Q 5 3 3 R i z o 1 8 4 2 u A 2 2 5 S 0 J Q 2 4 4 z o A A 8 5 B u r P u o 2 Q 6 8 p 3 4 5 R h i i i A 1 r k u A A J 4 4 8 J 5 3 2 i z b i 1 2 B u r k h 2 5 0 S 7 Q 2 5 4 4 R h 1 b B k h s l 5 5 J 6 9 h o b i A A 1 1 1 u r o c 2 2 4 9 F 2 5 3 R o o i u k h o i 1 Q 4 3 5 A z z b A A A r A ll a Q 5 5 K Q 2 6 R i i i z b i i i 1 r k h o A 2 i m l J Q 4 2 J 5 4 3 b i o h o i J 2 J 5 7 h h o b b b A 1 k A g 5 3 Q 2 5 4 2 o h z i 1 h A 2 2 o Q 4 3 2 i z o o o 1 i o i 2 J 2 5 z R R i b A A 2 n n J 5 Q 4 3 2 i R h z z z i i o A Q Q 2 7 h i i i o A b A 2 a a 2 J 5 4 h 2 7 b b i 1 c J J Q 4 8 R h h h z 1 o 2 s Q 2 5 5 5 R i o o b A 1 e J 5 R i A z J 2 3 4 R R R z z o i i a Q 3 7 h i i A i e 2 3 3 z b i h J Q 4 4 4 1 i b 4 0 7 R h h o b b J Q 6 5 5 7 4 o R J 5 7 0 3 R h z o o 3 2 2 4 9 R i z 2 4 7 9 i z z u 4 5 9 R h i i i Q Q 5 4 6 6 h 5 J Q 2 7 0 h H b J 2 5 4 4 4 R h 2 J 6 1 8 R o Q 2 5 5 3 R R 2 2 2 Q 4 z 2 2 Q J 2 7 7 7 i Q 2 6 1 A A i J J 5 4 6 8 u i i A 4 3 h A i J Q Q 2 7 3 i b 2 5 4 2 H b b 2 J 5 b 2 i J 6 5 0 R b o 2 5 7 o o A 2 Q 2 i 2 4 9 3 o o z A A b 6 z z i 2 J 2 4 u i i A i i b A 2 Q Q 8 7 z z o i 5 4 i b i 5 i h b A J Q o i z J 2 H h h b A 2 5 F o i b 2 h h o 2 i J R z b 2 0 o 2 2 K R R i z h Q 8 i z A o b 2 Q R R i 2 A 2 3 i o z J u h z i A 2 2 Q 9 i o R i i J h z 7 h b A A 0 A 2 i b A J 2 z i i 3 H R h b A i 8 h i i F 4 R o u i 0 2 R h o b b 5 5 o b z R h b b 9 R K 3 3 3 i z H o o 3 7 2 R i z o o 9 4 i o 4 4 1 R 7 h z z 4 3 z 4 h z 7 i z i 2 2 7 5 h i 5 5 9 i i F 5 3 8 5 R 8 3 9 8 h h 0 9 R h 2 2 6 h 4 3 R h K 2 2 5 5 4 2 7 3 4 R R Q Q 5 R 5 3 R R 4 4 Q Q 5 0 4 7 5 3 0
J J
2 6 4 J 5 8 J A 2 3 5 6 2
F 2 4 5 4
5 1 8
6 2
5 o Q 2 8 6 7 5 5 6 l K Q 2
J Q 4 2 5 4 7 A
F 2 5 y
Q 4 3
J Q 3
J 5 3 A
o
5 p
Q J 4 4
K J 5 4 l 4 o
2 Q 4 A l
J 2 A
5 y
5 5
m 2 o
J 5 5
y l A
o
Q p 2 l
2 2
Q 2 a
p y o
Q 2 A
J l y
m J A
Q Q p
C
J Q o
y A
m Q p
Q l
o
J a
J J A
J l p o a y
J m
6 l m
C y o
A
p
y l
7 a
C
m
a
p
A
o
y
7
p l
2 A
C a m
o
C
6 p
y l
4
A
m
1
o
a m
l
A
1
C y
5 p A
o
8
6
a
m
y
l
A
4
a o
8 C p
2
o
l
4 5
y l
C
m p A
a
o
4 7
y
C
l A
5 6
Q 8
m p
y
o
6
a
5
C
p y l A 2
o
J 4 6
6
m
p
a
l
A
2 4
4
y
C m
p o
8
5
0
a
A
Q
4
y l
o
A
5
m
C
8
m p l
Q
2 4 5 a
8
o J
9
y A
p
C m
o
2
a
l
7
y
J
l
5
a A
4
0
0
7 C
p Q
o
m
y
A
a
4
p
l
Q
C m y A
2 3
o 5 8
9
J
4
C
A
p l
9
a
o
y
J A 5
m p o
2 0
C
l A
4 6
a
5
Q
l
o
0 m
y
1
2
p l o
A
A
8
A
C
5 8
a
y
o
4
m
J
l 2
y
A
2
Q
2
C m
p
A a l
A
A 7 y o o
A
4
2 o
A 8
A 6 A
p
y Q
5
J
a
m p
o l l
A
A 7
l 8
A C
o y
Q
4
a
o l
p
o C
o 5
A
A l 1
J
o 4
o 8
o
l
2
m p
y
l A A y
y l
7
J Q
o 4 o l l
A l 4 a
p
o C
5
y
m
l l
m
o 2 y 8 o l C y
4
y 4
p p
o o y
a
J 4
4
Q
p
y 5 l y
l o y 8 m
l l 2
p y y
C
m
y l
a
4
p 8
a
5
4
p
p m
J p Q
y 5 y 7
p 2 p
p 4
y y a
8
C
p
y p
m
1
p 7 a
m Q 2 5 4
m J
C
a
2 7
C
m
p 4
p
p p
m 5
m
4
m p Q
J
8
2 C
a
m
2 5
a
a
m
m m
C
4
m m Q 5
m a 0
6
C
J a Q 1
a
4 8
m
m a m m 2
a
0
a
a a
1
J
5
C
Q
2
C C
a
m a 9
J
4
a 8 0 C
5 4
a a
a C a 5 5 C
3
C J
C
Q 2
8
C
C C 7
a 5
Q 4 8
C C
1 2
C 5
7 8 0
6
J
C C 8
C C
2
2
4 0
J
Q 5
7 7
C 2
0
7 1
0
6
0
4 7
6
9
4
2 Q
8
5
3
4 6 J
8 2
7 4
8
1 Q
1 2 8
5
4
8 7
1
9 J 2 5
8
7 9 Q
2
4 0 9
5 8
0
5
7 J
4
8 7 0 4 2
5
4 8
8 2
J Q 8 2
4
1 7
5
4
8 1 7
8
4 2
Q
7 7 4
4 J
5
2 5
4
4
4 8
5
4 4
2
Q
4 4 Q
8 5
4
5 J
4 4 5
5 2
2
5
5 4 4
Q
2
5
5
J
J Q
2 5 4 5
5
2
5 5 Q
2 2 J
2
2
5 5
2
2
J Q
Q 2
2
5 Q
2 J
2 2
Q
2 2
Q
Q J
Q J
J Q Q
2
Q Q
J
Q Q
Q J J Q Q
J
J J
Q J
Q J
J J
J Q
J J J J J
122
(B)
J r Q e r
J h
2 e Q t
5 h o J 2 t Q 4 o 5 o
4 2 4 h
J 2 o 5 1 1 Q 3 k h 4 r 4 A
2 O k 1 u 6 o J 5 r Q 5 d 1 O p u B 4 2 1 1
2 1 i u A O p t B A A J 5 u 5 1 4 e o o Q 4 i o
9 t A 0 5 p s A u d d 2 1 d O i 3 4 4 o 4 t A P u u 5 u u
1 d 4 4 8 p 3 e e 1 A e u 4 O i 4 4 5 s s A 3 t 3 s e
u 5 1 5 2 o p P P 8 A P s 2 A 4 d J i O t 3 Q P Q u 1 5 u o 4 1 p Q J 9 d 2 K A 5 3 e 5 3 A 5 M i 3 J 4 s u t 4 1 o 4 u 2 4 4 e 1 0 A 4 P d J 4 4 4 s Q 5 6 Q 5 u 1 3 5 4 2 3 6 5 1 P e J 2 A 5 O 0 0.05 2 5 4 s 4 0 2 9 o 4 p Q 2 1 P d 1 4 Q 3 J 4 i Q 4 A Q t J Q u J 4 4 u J 9 e o 2 A J 5 4 d 5 O 3 2 5 s 1 41 p 5 4 P u A 5 itu Q 2 4 se o JQ 4 A J 5 1 d 2 O 3 Q 2 4 P u 1 5 p J 4 0 e A 44 itu Q 4 6 s o 6 A J 5 4 P d J 3 O 3 2 4 2 eu 1 Q2 p Q 5 4 s A 54 itu J 2 4 P do 44 A3 Q 54 6 eu 7 O J 2 44 s A1 pi Q 4 P do tuA J 25 64 eu 3 Q 4 Ps 1 J 54 2 oA JQ Q2 45 eud 254 J 54 Ps 163 2 0 A1 J Rh JQ 444 udo Q25 izob 25 Pse 4158 iale JQ 142 oA1 Rhiz sA2 254 seud JQ obial JQ 50 P 25416 esA2 2544 oA1 0 Rhizo JQ Pseud bialesA 254445 JQ25416 2 JQ 2 Rhizobiale PseudoA1 sA2 JQ254462 JQ254165 Rhizobiale sA2 JQ254138 PseudoA1 RhizobialesA2 JQ254161 JQ254461 Pseudo sA2 A1 Rhizobiale JQ25 254159 2 4143 Xan JQ ialesA thoA2 hizob JQ25 156 R sA2 4465 JQ254 biale Xanth hizo sA2 JQ oA2 5 R iale 2544 5415 zob A2 66 X JQ2 Rhi les anth 57 bia A2 JQ oA2 541 izo es JQ 254 Q2 Rh ial 2 25 42 J 6 ob sA J 41 2 X 16 iz le A2 Q 45 ant 54 Rh bia s 25 X hoA Q2 7 o le JQ 41 an 2 J 6 iz ia o 2 4 th 41 h b er J 5 7 oA 25 R zo t o Q 4 Xa 2 Q 4 hi n r 2 13 n J 16 R E te o J 5 4 th 4 4 n r Q 41 B oA 25 8 7 E te 2 3 u 2 Q 16 1 n 5 7 rk J 4 4 0 E 4 B h 5 5 7 J J 4 u oB 2 2 1 9 o Q J Q 5 rk Q Q 4 6 r Q 2 3 h J J 5 e o 2 5 B o 2 1 t r 5 2 4 u B 4 n e 5 Q t J J 4 4 rk J 5 E J 4 2 n Q Q 1 4 5 h
Q 3 7 o 2 J J 2 5
Q E 2 3 B
J 7 Q 2 Q 5 6 B
1 1 J 5 B B u 2 5 4 Q
J 2
4 7 4 r
5 1 u u k J 4
Q 5
5 1 2 4 r h
J 4 3 r
Q J 1 4 k
2 J 5 J k J o 4 2 5 J
4 6
Q Q 3 4 h h B
Q Q Q
4 5 2 5 8
Q Q
3 9 4 B o o
2 J 2 4 5 4 2
6
2 2 2 B
2 8 u o
2
5 B o
4 4 5 Q
B 5 5 5 r t
5 u h 8 t
4 B
3 J 4
4 u k h
4 4
u 4 r e
4
4 0 h 3 u e
B r
4 J k r r 4
4 4 k
4 o r
3 k r
3 h
u 3
B Q
2 3 k
3 h
h 3 A
7
o
r
B 1 h
u 9
5 2 o o k 2
4 1
A o
B
u r
h
B 5
B B
k
o B o 1 A B
r
o
u 4
h
t u
k
u
u t h u h 1
r u 4
o o
h r
r r k
e r e
r k
k
k 2
t
o r k
h
h k r
o h h
h 6
h o
e h
t
o
o o
h o
o r
B o
t
o
e
h
o o
o u
o r t
e o
h t t
t r h
r h t h
t e h k
h e
e e h r e
e r r r o r
r D
Figure S3. Maximum-likelihood phylogenetic trees of Cephalotes-specific sequences added to the CS-RDP database. (A) Symbiotic bacteria associated with C. varians; (B) Symbiotic bacteria associated with C. rohweri.
123
SUPPLEMENTARY TABLES
Table S1: Accession numbers for RDP database, followed by taxonomical placement as in Anderson et al. (2012). When unavailable, taxonomical information refers to the NCBI sequence entry.
Cephalotes varians FJ477607_SphingoA, JQ254207_SphingoA, JQ254208_SphingoA, JQ254209_SphingoA, JQ254321_SphingoA, JQ254380_SphingoA, JQ254381_SphingoA, JQ254467_SphingoA, JQ254588_SphingoA, JQ254589_SphingoA, JQ254590_SphingoA, JQ254662_SphingoA, JQ254668_SphingoA, JQ254670_SphingoA, JQ254745_SphingoA, JQ254748_SphingoA, JQ254761_SphingoA, JQ254863_SphingoA, JQ254320_RhizobiA1, JQ254325_RhizobiA1, JQ254327_RhizobiA1, JQ254334_RhizobiA1, JQ254335_RhizobiA1, JQ254341_RhizobiA1, JQ254343_RhizobiA1, JQ254344_RhizobiA1, JQ254345_RhizobiA1, JQ254348_RhizobiA1, JQ254352_RhizobiA1, JQ254353_RhizobiA1, JQ254356_RhizobiA1, JQ254357_RhizobiA1, JQ254359_RhizobiA1, JQ254360_RhizobiA1, JQ254361_RhizobiA1, JQ254362_RhizobiA1, JQ254363_RhizobiA1, JQ254364_RhizobiA1, JQ254653_RhizobiA1, JQ254676_RhizobiA1, JQ254678_RhizobiA1, JQ254679_RhizobiA1, JQ254693_RhizobiA1, JQ254694_RhizobiA1, JQ254703_RhizobiA1, JQ254705_RhizobiA1, JQ254707_RhizobiA1, JQ254710_RhizobiA1, JQ254717_RhizobiA1, JQ254722_RhizobiA1, JQ254726_RhizobiA1, JQ254728_RhizobiA1, JQ254734_RhizobiA1, JQ254739_RhizobiA1, JQ254322_RhizobiA2, JQ254324_RhizobiA2, JQ254328_RhizobiA2, JQ254332_RhizobiA2, JQ254338_RhizobiA2, JQ254340_RhizobiA2, JQ254351_RhizobiA2, JQ254354_RhizobiA2, JQ254358_RhizobiA2, JQ254379_RhizobiA2, JQ254584_RhizobiA2, JQ254585_RhizobiA2, JQ254586_RhizobiA2, JQ254587_RhizobiA2, JQ254650_RhizobiA2, JQ254651_RhizobiA2, JQ254652_RhizobiA2, JQ254665_RhizobiA2, JQ254691_RhizobiA2, JQ254713_RhizobiA2, JQ254753_RhizobiA2, JQ254762_RhizobiA2, JQ254839_RhizobiA2, JQ254177_FlavoA, JQ254306_FlavoA, JQ254307_FlavoA, JQ254311_FlavoA, JQ254367_FlavoA, JQ254368_FlavoA, JQ254369_FlavoA, JQ254370_FlavoA, JQ254489_FlavoA, JQ254490_FlavoA, JQ254491_FlavoA, JQ254492_FlavoA, JQ254493_FlavoA, JQ254494_FlavoA, JQ254495_FlavoA, JQ254496_FlavoA, JQ254715_FlavoA, JQ254176_CampyloA, JQ254480_CampyloA, JQ254481_CampyloA, JQ254482_CampyloA, JQ254483_CampyloA, JQ254484_CampyloA, JQ254485_CampyloA, JQ254486_CampyloA, JQ254487_CampyloA, JQ254488_CampyloA, JQ254658_CampyloA, JQ254674_CampyloA, JQ254690_CampyloA, JQ254702_CampyloA, JQ254709_CampyloA, JQ254719_CampyloA, JQ254750_CampyloA, JQ254771_CampyloA, JQ254772_CampyloA, JQ254773_CampyloA, JQ254774_CampyloA, JQ254776_CampyloA, JQ254777_CampyloA, JQ254779_CampyloA, JQ254780_CampyloA, JQ254783_CampyloA, JQ254784_CampyloA, JQ254785_CampyloA, JQ254787_CampyloA, JQ254788_CampyloA, JQ254789_CampyloA, JQ254790_CampyloA, JQ254792_CampyloA, JQ254793_CampyloA, JQ254794_CampyloA, JQ254795_CampyloA, JQ254796_CampyloA, JQ254797_CampyloA, JQ254799_CampyloA, JQ254800_CampyloA, JQ254801_CampyloA, JQ254802_CampyloA, JQ254803_CampyloA, JQ254804_CampyloA, JQ254805_CampyloA, JQ254806_CampyloA, JQ254807_CampyloA, JQ254808_CampyloA, JQ254809_CampyloA, JQ254810_CampyloA, JQ254811_CampyloA, JQ254812_CampyloA, JQ254814_CampyloA, JQ254815_CampyloA, 124
JQ254816_CampyloA, JQ254817_CampyloA, JQ254818_CampyloA, JQ254819_CampyloA, JQ254820_CampyloA, JQ254821_CampyloA, JQ254822_CampyloA, JQ254823_CampyloA, JQ254824_CampyloA, JQ254825_CampyloA, JQ254826_CampyloA, JQ254827_CampyloA, JQ254828_CampyloA, JQ254829_CampyloA, JQ254830_CampyloA, JQ254831_CampyloA, JQ254832_CampyloA, JQ254856_CampyloA, JQ254866_CampyloA, JQ254876_CampyloA, JQ254474_BurkhoA1, JQ254476_BurkhoA1, JQ254479_BurkhoA1, JQ254667_BurkhoA1, JQ254685_BurkhoA1, JQ254689_BurkhoA1, JQ254723_BurkhoA1, JQ254732_BurkhoA1, JQ254740_BurkhoA1, JQ254749_BurkhoA1, JQ254754_BurkhoA1, JQ254757_BurkhoA1, JQ254764_BurkhoA1, JQ254766_BurkhoA1, JQ254782_BurkhoA1, JQ254848_BurkhoA1, JQ254854_BurkhoA1, JQ254857_BurkhoA1, JQ254865_BurkhoA1, JQ254873_BurkhoA1, JQ254175_BurkhoA2, JQ254365_BurkhoA2, JQ254366_BurkhoA2, JQ254700_BurkhoA2, JQ254708_BurkhoA2, JQ254712_BurkhoA2, JQ254730_BurkhoA2, JQ254742_BurkhoA2, JQ254751_BurkhoA2, JQ254841_BurkhoA2, JQ254845_BurkhoA2, JQ254859_BurkhoA2, JQ254860_BurkhoA2, JQ254868_BurkhoA2, JQ254869_BurkhoA2, JQ254870_BurkhoA2, JQ254879_BurkhoA2, JQ254882_BurkhoA2, JQ254470_BurkhoA_other, FJ477612_BurkhoB, JQ254303_BurkhoB, JQ254323_BurkhoB, JQ254326_BurkhoB, JQ254329_BurkhoB, JQ254336_BurkhoB, JQ254337_BurkhoB, JQ254342_BurkhoB, JQ254478_BurkhoB, JQ254657_BurkhoB, JQ254660_BurkhoB, JQ254681_BurkhoB, JQ254682_BurkhoB, JQ254686_BurkhoB, JQ254727_BurkhoB, JQ254731_BurkhoB, JQ254744_BurkhoB, JQ254760_BurkhoB, JQ254836_BurkhoB, JQ254838_BurkhoB, JQ254850_BurkhoB, JQ254853_BurkhoB, JQ254864_BurkhoB, JQ254877_BurkhoB, JQ254469_BurkhoC, JQ254471_BurkhoC, JQ254472_BurkhoC, JQ254473_BurkhoC, JQ254475_BurkhoC, JQ254477_BurkhoC, JQ254666_BurkhoC, JQ254699_BurkhoC, JQ254706_BurkhoC, JQ254725_BurkhoC, JQ254737_BurkhoC, JQ254756_BurkhoC, JQ254851_BurkhoC, JQ254852_BurkhoC, JQ254858_BurkhoC, JQ254468_BurkhoD, JQ254661_BurkhoD, JQ254669_BurkhoD, JQ254724_BurkhoD, JQ254855_BurkhoD, FJ477564_OpituA1, FJ477608_OpituA1, FJ477609_OpituA1, FJ477619_OpituA1, FJ477620_OpituA1, JQ254178_OpituA1, JQ254179_OpituA1, JQ254180_OpituA1, JQ254181_OpituA1, JQ254182_OpituA1, JQ254183_OpituA1, JQ254184_OpituA1, JQ254185_OpituA1, JQ254186_OpituA1, JQ254187_OpituA1, JQ254188_OpituA1, JQ254189_OpituA1, JQ254190_OpituA1, JQ254191_OpituA1, JQ254192_OpituA1, JQ254193_OpituA1, JQ254194_OpituA1, JQ254195_OpituA1, JQ254196_OpituA1, JQ254197_OpituA1, JQ254198_OpituA1, JQ254199_OpituA1, JQ254200_OpituA1, JQ254201_OpituA1, JQ254202_OpituA1, JQ254203_OpituA1, JQ254204_OpituA1, JQ254205_OpituA1, JQ254206_OpituA1, JQ254249_OpituA1, JQ254250_OpituA1, JQ254251_OpituA1, JQ254252_OpituA1, JQ254253_OpituA1, JQ254254_OpituA1, JQ254255_OpituA1, JQ254256_OpituA1, JQ254257_OpituA1, JQ254258_OpituA1, JQ254259_OpituA1, JQ254260_OpituA1, JQ254261_OpituA1, JQ254262_OpituA1, JQ254263_OpituA1, JQ254264_OpituA1, JQ254265_OpituA1, JQ254266_OpituA1, JQ254267_OpituA1, JQ254268_OpituA1, JQ254269_OpituA1, JQ254270_OpituA1, JQ254271_OpituA1, JQ254272_OpituA1, JQ254273_OpituA1, JQ254274_OpituA1, JQ254275_OpituA1, JQ254276_OpituA1, JQ254277_OpituA1, JQ254278_OpituA1, JQ254330_OpituA1, JQ254333_OpituA1, JQ254339_OpituA1, JQ254347_OpituA1, JQ254350_OpituA1, JQ254355_OpituA1, JQ254375_OpituA1, JQ254376_OpituA1, JQ254377_OpituA1, 125
JQ254378_OpituA1, JQ254502_OpituA1, JQ254503_OpituA1, JQ254504_OpituA1, JQ254505_OpituA1, JQ254506_OpituA1, JQ254507_OpituA1, JQ254508_OpituA1, JQ254509_OpituA1, JQ254510_OpituA1, JQ254511_OpituA1, JQ254512_OpituA1, JQ254513_OpituA1, JQ254514_OpituA1, JQ254515_OpituA1, JQ254516_OpituA1, JQ254517_OpituA1, JQ254518_OpituA1, JQ254519_OpituA1, JQ254520_OpituA1, JQ254521_OpituA1, JQ254522_OpituA1, JQ254523_OpituA1, JQ254524_OpituA1, JQ254525_OpituA1, JQ254526_OpituA1, JQ254527_OpituA1, JQ254528_OpituA1, JQ254529_OpituA1, JQ254530_OpituA1, JQ254531_OpituA1, JQ254532_OpituA1, JQ254533_OpituA1, JQ254534_OpituA1, JQ254535_OpituA1, JQ254536_OpituA1, JQ254537_OpituA1, JQ254538_OpituA1, JQ254539_OpituA1, JQ254540_OpituA1, JQ254541_OpituA1, JQ254542_OpituA1, JQ254543_OpituA1, JQ254544_OpituA1, JQ254545_OpituA1, JQ254546_OpituA1, JQ254547_OpituA1, JQ254548_OpituA1, JQ254549_OpituA1, JQ254550_OpituA1, JQ254551_OpituA1, JQ254552_OpituA1, JQ254553_OpituA1, JQ254554_OpituA1, JQ254555_OpituA1, JQ254556_OpituA1, JQ254557_OpituA1, JQ254558_OpituA1, JQ254559_OpituA1, JQ254560_OpituA1, JQ254561_OpituA1, JQ254562_OpituA1, JQ254563_OpituA1, JQ254564_OpituA1, JQ254565_OpituA1, JQ254566_OpituA1, JQ254567_OpituA1, JQ254568_OpituA1, JQ254569_OpituA1, JQ254570_OpituA1, JQ254571_OpituA1, JQ254572_OpituA1, JQ254573_OpituA1, JQ254574_OpituA1, JQ254575_OpituA1, JQ254576_OpituA1, JQ254577_OpituA1, JQ254578_OpituA1, JQ254579_OpituA1, JQ254580_OpituA1, JQ254581_OpituA1, JQ254582_OpituA1, JQ254583_OpituA1, JQ254645_OpituA1, JQ254647_OpituA1, JQ254648_OpituA1, JQ254649_OpituA1, JQ254659_OpituA1, JQ254716_OpituA1, JQ254752_OpituA1, JQ254765_OpituA1, JQ254775_OpituA1, JQ254781_OpituA1, JQ254786_OpituA1, JQ254791_OpituA1, JQ254798_OpituA1, JQ254813_OpituA1, JQ254833_OpituA1, JQ254835_OpituA1, JQ254840_OpituA1, JQ254844_OpituA1, JQ254861_OpituA1, JQ254867_OpituA1, JQ254874_OpituA1, JQ254880_OpituA1, JQ254849_OpituA3, JQ254371_PseudoA2, JQ254373_PseudoA2, JQ254374_PseudoA2, JQ254498_PseudoA2, JQ254499_PseudoA2, JQ254500_PseudoA2, JQ254501_PseudoA2, JQ254680_PseudoA2, JQ254688_PseudoA2, JQ254704_PseudoA2, JQ254714_PseudoA2, JQ254718_PseudoA2, JQ254720_PseudoA2, JQ254735_PseudoA2, JQ254736_PseudoA2, JQ254741_PseudoA2, JQ254743_PseudoA2, JQ254747_PseudoA2, JQ254767_PseudoA2, JQ254834_PseudoA2, JQ254842_PseudoA2, JQ254862_PseudoA2, FJ477621_XanthoA1, FJ477622_XanthoA1, FJ477623_XanthoA1, JQ254210_XanthoA1, JQ254211_XanthoA1, JQ254212_XanthoA1, JQ254213_XanthoA1, JQ254214_XanthoA1, JQ254215_XanthoA1, JQ254216_XanthoA1, JQ254217_XanthoA1, JQ254218_XanthoA1, JQ254246_XanthoA1, JQ254247_XanthoA1, JQ254248_XanthoA1, JQ254372_XanthoA1, JQ254382_XanthoA1, JQ254383_XanthoA1, JQ254384_XanthoA1, JQ254385_XanthoA1, JQ254386_XanthoA1, JQ254387_XanthoA1, JQ254388_XanthoA1, JQ254389_XanthoA1, JQ254390_XanthoA1, JQ254391_XanthoA1, JQ254392_XanthoA1, JQ254393_XanthoA1, JQ254394_XanthoA1, JQ254395_XanthoA1, JQ254396_XanthoA1, JQ254397_XanthoA1, JQ254398_XanthoA1, JQ254399_XanthoA1, JQ254400_XanthoA1, JQ254401_XanthoA1, JQ254402_XanthoA1, JQ254403_XanthoA1, JQ254404_XanthoA1, JQ254405_XanthoA1, JQ254406_XanthoA1, JQ254407_XanthoA1, JQ254408_XanthoA1, JQ254409_XanthoA1, JQ254410_XanthoA1, JQ254411_XanthoA1, JQ254412_XanthoA1, JQ254413_XanthoA1, 126
JQ254414_XanthoA1, JQ254591_XanthoA1, JQ254592_XanthoA1, JQ254593_XanthoA1, JQ254594_XanthoA1, JQ254595_XanthoA1, JQ254596_XanthoA1, JQ254597_XanthoA1, JQ254598_XanthoA1, JQ254599_XanthoA1, JQ254600_XanthoA1, JQ254601_XanthoA1, JQ254602_XanthoA1, JQ254603_XanthoA1, JQ254604_XanthoA1, JQ254605_XanthoA1, JQ254606_XanthoA1, JQ254607_XanthoA1, JQ254608_XanthoA1, JQ254609_XanthoA1, JQ254610_XanthoA1, JQ254611_XanthoA1, JQ254612_XanthoA1, JQ254613_XanthoA1, JQ254614_XanthoA1, JQ254615_XanthoA1, JQ254616_XanthoA1, JQ254619_XanthoA1, JQ254620_XanthoA1, JQ254621_XanthoA1, JQ254622_XanthoA1, JQ254623_XanthoA1, JQ254624_XanthoA1, JQ254625_XanthoA1, JQ254626_XanthoA1, JQ254627_XanthoA1, JQ254628_XanthoA1, JQ254629_XanthoA1, JQ254630_XanthoA1, JQ254631_XanthoA1, JQ254632_XanthoA1, JQ254633_XanthoA1, JQ254634_XanthoA1, JQ254635_XanthoA1, JQ254636_XanthoA1, JQ254638_XanthoA1, JQ254639_XanthoA1, JQ254640_XanthoA1, JQ254641_XanthoA1, JQ254642_XanthoA1, JQ254643_XanthoA1, JQ254644_XanthoA1, JQ254696_XanthoA1, JQ254697_XanthoA1, JQ254701_XanthoA1, JQ254746_XanthoA1, JQ254755_XanthoA1, JQ254763_XanthoA1, JQ254768_XanthoA1, JQ254769_XanthoA1, JQ254778_XanthoA1, JQ254846_XanthoA1, JQ254497_XanthoA2, JQ254617_XanthoA2, JQ254618_XanthoA2, JQ254637_XanthoA2, JQ254646_XanthoA2, JQ254654_XanthoA2, JQ254655_XanthoA2, JQ254656_XanthoA2, JQ254663_XanthoA2, JQ254664_XanthoA2, JQ254671_XanthoA2, JQ254672_XanthoA2, JQ254673_XanthoA2, JQ254675_XanthoA2, JQ254677_XanthoA2, JQ254683_XanthoA2, JQ254684_XanthoA2, JQ254687_XanthoA2, JQ254692_XanthoA2, JQ254695_XanthoA2, JQ254698_XanthoA2, JQ254711_XanthoA2, JQ254721_XanthoA2, JQ254729_XanthoA2, JQ254733_XanthoA2, JQ254738_XanthoA2, JQ254758_XanthoA2, JQ254759_XanthoA2, JQ254770_XanthoA2, JQ254847_XanthoA2, JQ254871_XanthoA2, JQ254872_XanthoA2, JQ254875_XanthoA2, JQ254878_XanthoA2, KF730288_Hu_Xantho, KF730289_Hu_Rhizobi, KF730290_Hu_Xantho, KF730291_Hu_Alcaliginaceae, KF730292_Hu_Xantho, KF730293_Hu_Alcaliginaceae, KF730294_Hu_Pusillimonas, KF730295_Hu_Xantho, KF730296_Hu_Xantho, KF730297_Hu_Rhizobi, KF730298_Hu_Rhizobi, KF730299_Hu_Rhizobi, KF730300_Hu_Comamonadaceae, KF730301_Hu_Xantho, KF730302_Hu_Pseudo, KF730303_Hu_Xantho, KF730304_Hu_Pseudo, KF730305_Hu_Pseudo, KF730306_Hu_Pseudisphingobacterium, KF730307_Hu_Pseudo, KF730308_Hu_Phyllobacteriaceae, KF730309_Hu_Pusillimonas, KF730310_Hu_Pusillimonas, KF730311_Hu_Pusillimonas, KF730312_Hu_Rhizobi, KM066003_Ceph_capnophilus, KT258896_Ventosimonas_gracilis. Cephalotes rohweri JQ254136_BurkhoA1, JQ254421_BurkhoA1, JQ254448_BurkhoA1, JQ254458_BurkhoA1, JQ254134_BurkhoB, JQ254137_BurkhoB, JQ254420_BurkhoB, JQ254453_BurkhoB, JQ254457_BurkhoB, JQ254426_BurkhoD, JQ254169_Entero, JQ254170_Entero, JQ254171_Entero, JQ254172_Entero, JQ254174_Entero, JQ254146_OpituA3, JQ254148_OpituA3, JQ254149_OpituA3, JQ254150_OpituA3, JQ254151_OpituA3, JQ254152_OpituA3, JQ254153_OpituA3, JQ254154_OpituA3, JQ254173_OpituA3, JQ254423_OpituA3, JQ254424_OpituA3, JQ254425_OpituA3, JQ254438_OpituA3, JQ254444_OpituA3, JQ254447_OpituA3, JQ254463_OpituA3, JQ254133_Burkho_other, JQ254139_Burkho_other, JQ254140_Burkho_other, JQ254427_Burkho_other, 127
JQ254428_Burkho_other, JQ254429_Burkho_other, JQ254430_Burkho_other, JQ254431_Burkho_other, JQ254432_Burkho_other, JQ254433_Burkho_other, JQ254434_Burkho_other, JQ254435_Burkho_other, JQ254436_Burkho_other, JQ254437_Burkho_other, JQ254455_Burkho_other, JQ254456_Burkho_other, JQ254135_PseudoA1, JQ254138_PseudoA1, JQ254141_PseudoA1, JQ254142_PseudoA1, JQ254439_PseudoA1, JQ254440_PseudoA1, JQ254441_PseudoA1, JQ254442_PseudoA1, JQ254443_PseudoA1, JQ254445_PseudoA1, JQ254446_PseudoA1, JQ254449_PseudoA1, JQ254450_PseudoA1, JQ254451_PseudoA1, JQ254452_PseudoA1, JQ254454_PseudoA1, JQ254459_PseudoA1, JQ254460_PseudoA1, JQ254461_PseudoA1, JQ254462_PseudoA1, JQ254464_PseudoA1, JQ254155_RhizobialesA2, JQ254156_RhizobialesA2, JQ254157_RhizobialesA2, JQ254158_RhizobialesA2, JQ254159_RhizobialesA2, JQ254160_RhizobialesA2, JQ254161_RhizobialesA2, JQ254162_RhizobialesA2, JQ254163_RhizobialesA2, JQ254164_RhizobialesA2, JQ254165_RhizobialesA2, JQ254166_RhizobialesA2, JQ254167_RhizobialesA2, JQ254168_RhizobialesA2, JQ254143_XanthoA2, JQ254144_XanthoA2, JQ254145_XanthoA2, JQ254147_XanthoA2, JQ254419_XanthoA2, JQ254422_XanthoA2, JQ254465_XanthoA2, JQ254466_XanthoA2, KM066004_CephPrimus.
Table S2. Selected collapsed KEGG categories (level 3) for metabolism involved in nutritional functions and host-bacteria interactions Amino acid Glycerolipid Mineral absorption Flavonoid metabolism metabolism biosynthesis Amino acid related Glycerophospholipid Nitrogen metabolism Sulfur metabolism enzymes metabolism Amino sugar and Glycine, serine and Other glycan Thiamine nucleotide sugar threonine degradation metabolism metabolism metabolism Biosynthesis and Glycolysis / Other ion-coupled Metabolism of biodegradation of Gluconeogenesis transporters cofactors and secondary vitamins metabolites Biotin metabolism Inorganic ion Other transporters Methane metabolism transport and metabolism Carbohydrate Inositol phosphate Peptidases Glutathione digestion and metabolism metabolism absorption Carbohydrate Ion channels Phenylalanine Tryptophan metabolism metabolism metabolism Carotenoid Isoflavonoid Phenylalanine, Transporters biosynthesis biosynthesis tyrosine and tryptophan 128
biosynthesis Cysteine and Linoleic acid Pores ion channels Tyrosine metabolism methionine metabolism metabolism D-Alanine Lipid biosynthesis Protein digestion and Valine, leucine and metabolism proteins absorption isoleucine biosynthesis D-Arginine and D- Lipid metabolism Protein export Valine, leucine and ornithine metabolism isoleucine degradation D-Glutamine and D- Lipoic acid Secretion system Vitamin B6 glutamate metabolism metabolism metabolism Energy metabolism Lipopolysaccharide Starch and sucrose alpha-Linolenic acid biosynthesis metabolism metabolism Fat digestion and Lipopolysaccharide Folate biosynthesis beta-Alanine absorption biosynthesis proteins metabolism Fatty acid Lysine biosynthesis Fructose and biosynthesis mannose metabolism Fatty acid Lysine degradation Galactose metabolism metabolism
Table S3. Selected KEGG pathways for hypothesis testing
Nitrogen Fixation K00531 anfG; nitrogenase delta subunit [EC:1.18.6.1] K02585 nifB; nitrogen fixation protein nifD; nitrogenase molybdenum-iron protein alpha chain K02586 [EC:1.18.6.1] K02588 nifH; nitrogenase iron protein NifH [EC:1.18.6.1] nifK; nitrogenase molybdenum-iron protein beta chain K02591 [EC:1.18.6.1] K02593 nifT; nitrogen fixation protein K02596 nifX; nitrogen fixation protein K02597 nifZ; nitrogen fixation protein K04488 nifU; nitrogen fixation protein
Nitrogen Recycling K01428 ureC; urease subunit alpha [EC:3.5.1.5] K01429 ureB; urease subunit beta [EC:3.5.1.5] K03187 ureE; urease accessory protein 129
K03188 ureF; urease accessory protein K03189 ureG; urease accessory protein K03190 ureD; ureH; urease accessory protein K03192 ureJ; urease accessory protein K14048 ureAB; urease subunit gamma/beta [EC:3.5.1.5]
Uric acid digestion K00365 uaZ; urate oxidase [EC:1.7.3.3] K01466 allB; allantoinase [EC:3.5.2.5] K01477 alc, ALLC; allantoicase [EC:3.5.3.4] Cobalamin biosynthesis (Vitamin B12) K00798 MMAB, pduO; cob(I)alamin adenosyltransferase [EC:2.5.1.17] K19221 cobA, btuR; cob(I)alamin adenosyltransferase [EC:2.5.1.17] K02232 cobQ, cbiP; adenosylcobyric acid synthase [EC:6.3.5.10] K02225 cobC1, cobC; cobalamin biosynthetic protein CobC K02227 cbiB, cobD; adenosylcobinamide-phosphate synthase [EC:6.3.1.10] cobP, cobU; adenosylcobinamide kinase / adenosylcobinamide-phosphate K02231 guanylyltransferase [EC:2.7.1.156 2.7.7.62] E2.4.2.21, cobU, cobT; nicotinate-nucleotide--dimethylbenzimidazole K00768 phosphoribosyltransferase [EC:2.4.2.21] K02226 cobC, phpB; alpha-ribazole phosphatase [EC:3.1.3.73] K02233 E2.7.8.26, cobS, cobV; adenosylcobinamide-GDP ribazoletransferase [EC:2.7.8.26]
Tocopherol/tocotorienol biosynthesis (Vitamin E) HPT, HGGT, ubiA; homogentisate phytyltransferase / homogentisate K09833 geranylgeranyltransferase [EC:2.5.1.115 2.5.1.116] K12502 VTE3, APG1; MPBQ/MSBQ methyltransferase [EC:2.1.1.295] K18534 K18534; MPBQ/MSBQ methyltransferase [EC:2.1.1.295] K09834 VTE1, SXD1; tocopherol cyclase [EC:5.5.1.24] K05928 E2.1.1.95; tocopherol O-methyltransferase [EC:2.1.1.95]
M0012 7 Thiamine biosynthesis, AIR => thiamine-P/thiamine-2P (Vitamin B1) K03147 thiC; phosphomethylpyrimidine synthase [EC:4.1.99.17] THI20; hydroxymethylpyrimidine/phosphomethylpyrimidine kinase K00877 [EC:2.7.1.49 2.7.4.7] thiD; hydroxymethylpyrimidine/phosphomethylpyrimidine kinase K00941 [EC:2.7.1.49 2.7.4.7] thiDE; hydroxymethylpyrimidine kinase / phosphomethylpyrimidine kinase / K14153 thiamine-phosphate diphosphorylase [EC:2.7.1.49 2.7.4.7 2.5.1.3] K00878 thiM; hydroxyethylthiazole kinase [EC:2.7.1.50] 130
THI6; thiamine-phosphate diphosphorylase / hydroxyethylthiazole kinase K14154 [EC:2.5.1.3 2.7.1.50] K00788 thiE; thiamine-phosphate pyrophosphorylase [EC:2.5.1.3] thiDE; hydroxymethylpyrimidine kinase / phosphomethylpyrimidine kinase / K14153 thiamine-phosphate diphosphorylase [EC:2.7.1.49 2.7.4.7 2.5.1.3] THI6; thiamine-phosphate diphosphorylase / hydroxyethylthiazole kinase K14154 [EC:2.5.1.3 2.7.1.50] K00946 thiL; thiamine-monophosphate kinase [EC:2.7.4.16]
M00125 Riboflavin biosynthesis, GTP => riboflavin/FMN/FAD (Vitamin B2) K01497 ribA, RIB1; GTP cyclohydrolase II [EC:3.5.4.25] ribBA; 3,4-dihydroxy 2-butanone 4-phosphate synthase / GTP cyclohydrolase II K14652 [EC:4.1.99.123.5.4.25] K01498 ribD1; diaminohydroxyphosphoribosylaminopyrimidine deaminase [EC:3.5.4.26] ribD; diaminohydroxyphosphoribosylaminopyrimidine deaminase / 5-amino-6-(5- K11752 phosphoribosylamino)uracil reductase [EC:3.5.4.26 1.1.1.193] K00082 ribD2; 5-amino-6-(5-phosphoribosylamino)uracil reductase [EC:1.1.1.193] ribD; diaminohydroxyphosphoribosylaminopyrimidine deaminase / 5-amino-6-(5- K11752 phosphoribosylamino)uracil reductase [EC:3.5.4.26 1.1.1.193] K02858 ribB, RIB3; 3,4-dihydroxy 2-butanone 4-phosphate synthase [EC:4.1.99.12] ribBA; 3,4-dihydroxy 2-butanone 4-phosphate synthase / GTP cyclohydrolase II K14652 [EC:4.1.99.123.5.4.25] K00794 ribH, RIB4; 6,7-dimethyl-8-ribityllumazine synthase [EC:2.5.1.78] K00793 ribE, RIB5; riboflavin synthase [EC:2.5.1.9] K00861 RFK, FMN1; riboflavin kinase [EC:2.7.1.26] K11753 ribF; riboflavin kinase / FMN adenylyltransferase [EC:2.7.1.26 2.7.7.2] K00953 FLAD1; FAD synthetase [EC:2.7.7.2] K11753 ribF; riboflavin kinase / FMN adenylyltransferase [EC:2.7.1.26 2.7.7.2]
M0012 4 Pyridoxal biosynthesis, erythrose-4P => pyridoxal-5P (Vitamin B6) K03472 epd; D-erythrose 4-phosphate dehydrogenase [EC:1.2.1.72] K03473 pdxB; erythronate-4-phosphate dehydrogenase [EC:1.1.1.290] K00831 serC, PSAT1; phosphoserine aminotransferase [EC:2.6.1.52] K00097 pdxA; 4-hydroxythreonine-4-phosphate dehydrogenase [EC:1.1.1.262] K03474 pdxJ; pyridoxine 5-phosphate synthase [EC:2.6.99.2] K00275 pdxH, PNPO; pyridoxamine 5'-phosphate oxidase [EC:1.4.3.5] From KEGG Level 3 categories Lysine biosynthesis Phenylalanine, tyrosine and tryptophan biosynthesis 131
Valine, leucine and isoleucine biosynthesis Carotenoid biosynthesis M0008 1 Pectin degradation K01051 E3.1.1.11; pectinesterase [EC:3.1.1.11] K01184 E3.2.1.15; polygalacturonase [EC:3.2.1.15] E3.2.1.67; galacturan 1,4-alpha-galacturonidase K01213 [EC:3.2.1.67] K01732 pectin lyase [EC:4.2.2.10]
Chitinase chiA; bifunctional chitinase/lysozyme [EC:3.2.1.14 K13381 3.2.1.17] K01183 chitinase [EC:3.2.1.14]
Table S4. CS-RDP taxon nomenclature
C. varians
Representati Anderson Longest Taxonomical ID or Taxonomical name ve Sequence et al. sequence nickname supported by ID within OTU JQ254184 OpituA1 KM06600 Cephaloticoccus May 6. doi: (854bp) 3 capnophilus 10.1099/ijsem.0.001141. (1,552bp) [Epub ahead of print] JQ254597 XanthoA1 FJ477621 XanthoA1.1 RDP: Xylella [46%] (997bp) (1,467bp) Blast: Stenotrophomonas [100%/92%; GU945535] 1 KT258896 PseudoA2 KT25889 Ventosimonas gracilis Apr 7. doi: (1489bp) 6 10.1099/ijsem.0.001068. (1489bp) [Epub ahead of print] JQ254481 Campylo JQ254488 Arcobacter_varians RDP: Arcobacter [100%] (986bp) A Blast: Arcobacter trophiarum [98%/97%; FN650332] JQ254492 FlavoA JQ254495 FlavoA.1 RDP: Myroides [84%] (936bp) (1,002bp) Blast: Flavobacterium columnare [97%/91%; CP003222.2]3 JQ254584 RhizobiA KF73029 RhizobiA2.1 RDP: Brucella[39%] (968bp) 2 7 Blast: Bartonella vinsonii 132
(1,425bp) [100%/95%; EU295657]4 KF730295 XanthoA2 KF73029 XanthoA2.1 RDP: Silanimonas[28%] (1,475bp) 5 Bast: Lysobacteri mobilis (1,475bp) [100%/91%]5 JQ254380 SphingoA JQ254380 SphingoA.1 RDP: (985bp) (985bp) Pseudosphingobacterium[7 5%] Blast: Parapedobacter composti [100%, 86%]6 KF730311 BurkhoA2 KF73031 BurkhoA2.1 RDP: Pusillimonas[74%] (1,455bp) 1 Blast: Bordetella petrii (1,455bp) [100%/96%]7 JQ254207 SphingoA KF73030 SphingoA.2 RDP: (985bp) 6 Pseudosphingobacterium[7 (1,453bp) 9%] Blast: Parapedobacter composti [94%, 85%]6 JQ254367 FlavoA JQ254368 FlavoA.2 RDP: Myroides[71%] (998bp) (1,000bp) Blast: Flavobacterium columnare [97%/91%]3 JQ254472 BukhoC JQ254473 BukhoC.1 RDP: Diaphorobacter[77%] (886bp) (1,004bp) Blast: Alicycliphilus denitrificans [98%/95%]8 JQ254626 XanthoA1 JQ254626 XanthoA1.2 RDP: (1,002bp) (1,002bp) Stenotrophomonas[50%] Blast: Stenotrophomonas acidaminiphila[100%/93%] 9 JQ254476 BurkhoA1 JQ254476 BurkhoA1.1 RDP: Candidimonas[39%] (1,014bp) (1,014bp) Blast: Bordetella bronchiseptica [100%/92%]10 KF730300 BurkhoD KF73030 BurkhoD.1 RDP: Brachymonas[33%] (1,453bp) 0 Blast: Comamonas (1,453bp) odontotermitis[99%/93%]11 KF730299 RhizobiA N/A RhizobiA2.2 RDP: (1,421bp) 2 Pseudochrobactrum[30%] Blast: Bartonella apis [100%/95%]12 KF730298 RhizobiA N/A RhizobiA2.3 RDP: Daeguia[56%] (1,382bp) 2 Blast: Bartonella vinsonii [100%/95%; EU295657]4 KF730301 Near N/A XanthoA2.x RDP: Silanimonas[61%] 133
(1,465bp) XanthoA2 Blast: Thermomonas hydrothermalis [98%/91%] KF730296 XanthoA2 N/A XanthoA2.2 RDP: Silanimonas[31%] (1,475bp) Bast: Lysobacteri mobilis [100%/91%]5 KF730294 BurkhoB N/A BurkhoB.1 RDP: Pusillimonas[82%] (1,457bp) Blast: Pusillimonas sp. [100%/95%]13 KF730293 BurkhoA2 N/A BurkhoA2.2 RDP: Pusillimonas[68%] (1,455bp) Blast: Bordetella sp. [100%/96%]14 KF730309 BurkhoA N/A Bordetella_cf_HU1 RDP: Pusillimonas[61%] (1,455bp) other Blast: Bordetella sp. HU-1 [93%/99%]15 Bordetella petrii [100%/97%]7 KF730310 BurkhoB N/A BurkhoB.2 RDP: Pusillimonas[82%] (1,457bp) Blast: Pusillimonas sp. [100%/95%]13 KF730291 BurkhoA1 N/A BurkhoA1.2 RDP: Pusillimonas[69%] (1,455bp) Blast: Bordetella bronchiseptica [100%/92%]10 KF730289 RhizobiA N/A RhizobiA1.1 RDP: Mycoplana[14%] (1,411bp) 1 Blast: Phyllobacterium myrsinacearum [100%/93%]16 JQ254637 XanthoA2 N/A XanthoA2.3 RDP: Aquimonas[30%] (1,019bp) Blast: Lysobacteri mobilis [98%/91%]5 JQ254608 XanthoA1 N/A XanthoA1.3 RDP: Xylella[46%] (1,003bp) Blast: Stenotrophomonas acidaminiphila[99%/92%]9 JQ254592 XanthoA1 N/A XanthoA1.4 RDP: Xylella[55%] (957bp) Blast: Stenotrophomonas acidaminiphila[100%/92%] 9 JQ254500 PseudoA2 N/A Ventosimonas cf. RDP: Serpens[26%] (1,015bp) gracilis Blast: Ventosimonas gracilis [100%/98%] JQ254478 BurkhoB N/A BurkhoB.3 RDP: Pusillimonas[55%] (1,010bp) Blast: Pusillimonas sp. [100%/95%]20 134
FJ477623 XanthoA1 N/A XanthoA1.5 RDP: Xylella[33%] (1,421bp) Blast: Stenotrophomonas acidaminiphila[99%/93%]9 FJ477612 BurkhoB N/A BurkhoB.4 RDP: Pusillimonas[58%] (1,410bp) Blast: Pusillimonas sp. [100%/95%]13 FJ477564 OpituA1* N/A Stenotrophomonas_vari RDP: (1,467bp) (XanthoA ans Stenotrophomonas[100%] 1) Blast: Stenotrophomonas sp. [99%/99%]21 KF730312 Near N/A RhizobiaA1.2 RDP: (1,411bp) RhizobiA Phyllobacterium[36%] 1 Blast: Phyllobacterium myrsinacearum [100%/93%]16
C. rohweri Representative Anderson et Longest Taxonomical ID Taxonomical name Sequence ID al. sequence or nickname supported by within OTU JQ254437 Burkho other JQ254428 Aquabacterium_sp1 RDP: (1,463bp) (1,476bp) Aquabacterium[100%]
JQ254441 PseudoA1 JQ254441 PseudoA1.1 RDP: Serpens[45%] (1,481bp) (1,481bp) Blast: Ventosimonas gracilis [98%/95%] KM066004 OpituA3 KM066004 Cephaloticoccus (1,552bp) (1,552bp) primus JQ254134 BurkhoB N/A BurkhoB.5 RDP: (1,4767bp) Candidimonas[60%] Blast: “Alcaliginaceae bacterium HU-3” [92%/97%] Bordetella sp. [100%/95%]14 JQ254140 Burkho other N/A Burkho_other_1 RDP: (1,480bp) Parapusillimonas[52%] Blast: Ventosimonas gracilis [99%/91%]
JQ254443 PseudoA1 N/A PseudoA1.2 RDP: Serpens[69%] (1,472bp) Ventosimonas gracilis [100%/94%] 135
JQ254445 PseudoA1 N/A PseudoA1.3 RDP: Serpens[67%] (1,471bp) Ventosimonas gracilis [100%/94%] JQ254453 BurkhoB N/A BurkhoB.6 RDP: (1,465bp) Pusillimonas[59%] Blast: : “Alcaliginaceae bacterium HU-3” [91%/96%] Bordetella sp. [100%/94%]14 JQ254455 Burkho other N/A Burkho_other_2 RDP: (1,471bp) Parapusillimonas[12%] Blast: Ventosimonas gracilis [100%/92%] Tree: PseudoA1 JQ254462 PseudoA1 N/A PseudoA1.5 RDP: Serpens[29%] (1,454bp) Ventosimonas gracilis [100%/94%] 1 doi: 10.1111/j.1574-6968.2011.02452.x. Epub 2011 Nov 30 2 doi: 10.1099/ijs.0.022665-0. Epub 2010 Mar 19 3 doi: 10.1128/JB.00281-12 4 doi: 10.1128/JCM.02456-07. Epub 2008 Mar 26 5 doi: 10.1099/ijs.0.000026. Epub 2014 Dec 12 6 doi: 10.1099/ijs.0.013318-0. Epub 2009 Sep 18 7 doi: 10.1099/jmm.0.46976-0 8 doi: 10.1128/JB.00365-11. Epub 2011 Jul 8 9 doi: 10.1099/00207713-52-2-559 10 Horiguchi Y, Sugimoto N, Matsuda M. Stimulation of DNA synthesis in osteoblast-like MC3T3-E1 cells by Bordetella bronchiseptica dermonecrotic toxin. Infection and Immunity. 1993;61(9):3611-3615. 11 doi: 10.1099/ijs.0.64551-0 12 doi: 10.1099/ijsem.0.000736. Epub 2015 Nov 3. 13 doi: 10.1007/s10532-013-9636-3. Epub 2013 Mar 30 14 doi:10.1016/j.chemosphere.2006.11.019 15 doi: 10.1073/pnas.0907926106. Epub 2009 Nov 30. 16 doi: 10.1007/s11274-009-0148-6 19 doi: 10.1099/ijsem.0.001068. [Epub ahead of print] 20 doi: 10.1016/j.biortech.2009.07.086. Epub 2009 Aug 31 21 doi: 10.1128/AEM.05188-11. Epub 2011 May 20
136
Table S5. NSTI values (Cv1 C. varians (CvK) C.varians 6) Sample Organ Metric Value Orga Weight Sample n Metric Value CMS1280_crop ed Weight _2 crop NSTI 0.012094 D_YH081_cr ed 0.0123 Weight op crop NSTI 98 CMS1280_crop ed Weight _3 crop NSTI 0.026996 D_YH089_cr ed 0.0121 Weight op crop NSTI 64 ed Weight CMS1235_head head NSTI 0.02977 D_YH091_cr ed 0.0173 Weight op crop NSTI 07 CMS1280_head ed Weight _2 head NSTI 0.023536 D_YH092_cr ed 0.0124 Weight op crop NSTI 97 CMS1280_head ed Weight _3 head NSTI 0.015624 D_YH097_cr ed 0.0130 Weight op crop NSTI 94 CMS1280_hind hindgu ed Weight gut_2 t NSTI 0.062722 D_YH113_cr ed 0.0129 Weight op crop NSTI 03 CMS1280_hind hindgu ed Weight gut_3 t NSTI 0.051269 D_YH081_ile ed 0.0761 Weight um ileum NSTI 52 CMS1280_midg ed Weight ut_2 midgut NSTI 0.034327 D_YH089_ile ed 0.0713 Weight um ileum NSTI 48 CMS1280_midg ed Weight ut_3 midgut NSTI 0.044281 D_YH091_ile ed 0.0700 um ileum NSTI 36 Summary Weight Organ Min Max Average D_YH092_ile ed 0.0863 0.0156 0.0297 um ileum NSTI 9 Head 24 7 0.022976 Weight 0.0120 0.0269 D_YH097_ile ed 0.0807 Crop 94 96 0.019545 um ileum NSTI 34 0.0343 0.0442 Weight Midgut 27 81 0.039304 D_YH113_ile ed 0.0823 0.0512 0.0627 um ileum NSTI 01 Hindgut 69 22 0.056996 Weight D_YH081_mi midg ed 0.0422 dgut ut NSTI 26 137
Weight D_YH089_mi midg ed 0.0391 dgut ut NSTI 19 Weight D_YH091_mi midg ed 0.0430 dgut ut NSTI 81 Weight D_YH092_mi midg ed 0.0490 dgut ut NSTI 62 Weight D_YH097_mi midg ed 0.0409 dgut ut NSTI 88 Weight D_YH113_mi midg ed 0.0357 dgut ut NSTI 81 Weight D_YH081_po pocke ed 0.0660 cket t NSTI 4 Weight D_YH089_po pocke ed 0.0491 cket t NSTI 06 Weight D_YH091_po pocke ed 0.0326 cket t NSTI 79 Weight D_YH092_po pocke ed 0.0518 cket t NSTI 41 Weight D_YH097_po pocke ed 0.0349 cket t NSTI 86 Weight D_YH113_po pocke ed 0.0411 cket t NSTI 22 Weight D_YH081_re rectu ed 0.0687 ctum m NSTI 07 Weight D_YH089_re rectu ed 0.0635 ctum m NSTI 29 Weight D_YH091_re rectu ed 0.0544 ctum m NSTI 7 D_YH092_re rectu Weight 0.0784 ctum m ed 01 138
NSTI Weight D_YH097_re rectu ed 0.0767 ctum m NSTI 17 Weight D_YH113_re rectu ed 0.0773 ctum m NSTI 27
Summary Avera Organ Min Max ge Infrabuccal 0.032 0.0660 0.0459 pocket 7 4 62 0.012 0.0173 0.0133 Crop 2 07 94 0.035 0.0490 0.0417 Midgut 8 62 1 0.0863 0.0778 Ileum 0.07 9 27 0.054 0.0784 0.0698 Rectum 5 01 59
C.rohweri Sample Organ Metric Value Weighted crop_1 Crop NSTI 0.024238 Weighted crop_2 Crop NSTI 0.013291 Weighted crop_3 Crop NSTI 0.022665 Weighted crop_4 Crop NSTI 0.019688 Weighted hg_1 Ileum NSTI 0.041818 Weighted hg_6 Ileum NSTI 0.039862 Weighted hg_7 Ileum NSTI 0.041118 Weighted il_1 Ileum NSTI 0.047642 il_2 Ileum Weighted 0.050699 139
NSTI Weighted il_3 Ileum NSTI 0.058664 Weighted il_4 Ileum NSTI 0.060987 Weighted mg_1 Midgut NSTI 0.039605 Weighted mg_2 Midgut NSTI 0.040298 Weighted mg_3 Midgut NSTI 0.039809 Weighted mg_4 Midgut NSTI 0.039314 Weighted midgut_1 Midgut NSTI 0.039415 Weighted midgut_6 Midgut NSTI 0.039353 Weighted midgut_7 Midgut NSTI 0.039391 Weighted prov_1 Proventriculus NSTI 0.028626 Weighted prov_7 Proventriculus NSTI 0.015496 Weighted pv_1 Proventriculus NSTI 0.034179 Weighted pv_2 Proventriculus NSTI 0.015336 Weighted pv_3 Proventriculus NSTI 0.024185 Weighted pv_4 Proventriculus NSTI 0.01999 Weighted rectum_2 Rectum NSTI 0.040482 Weighted rectum_3 Rectum NSTI 0.050503 Weighted rectum_4 Rectum NSTI 0.063502
Summary Organ Min Max Average Crop 0.013290734 0.024238 0.01997 Proventriculus 0.015335767 0.034179 0.022969 Midgut 0.039314072 0.040298 0.039598 140
Ileum 0.039862183 0.060987 0.048684 Rectum 0.040482088 0.063502 0.051496
Table S6. Number of unclassified sequences obtained when using the RDP database (v. 11) and our curated, Cephalotes-specific RDP database (CS-RDP). Improvement is shown by a smaller number of unclassified Bacterial sequences obtained in the family and genus levels.
Family Genus CS- CS- RDP RDP RDP RDP Cv16 43 32 57 46 CvK 51 10 76 40 Crohweri 170 50 249 191
141
APPENDIX C: TRANSMISSION OF BENEFICIAL MICROBES MEDIATED BY
BEHAVIOR IN A SOCIAL INSECT
142
Transmission of beneficial microbes mediated by behavior in a social insect
Pedro A P Rodrigues1*, Corinne M Stouthamer1, Jacqueline Lafratta2, Diana Wheeler2
1 Graduate Interdisciplinary Program in Entomology and Insect Science, University of Arizona,
Tucson, AZ 85721, USA
2 Department of Entomology, University of Arizona, Tucson, AZ 85721, USA
* Corresponding author
Email: [email protected]
Keywords: symbiosis, mutualisms, cryptic herbivory, Cephalotes, nutritional ecology, arboreal ants, social insects
ABSTRACT
The evolution of beneficial host-bacteria symbiosis is often accompanied by morphological, physiological and behavioral adaptations that result in stable interactions over the host lifetime and across generations. In social insects these adaptations extend beyond the individual enabling nestmates and developing brood to benefit from the symbiotic bacteria. Here we explore the integrative aspect of trophallactic behavior, in particular abdominal trophallaxis, by investigating its importance in colony nutrition mediated by symbiotic gut microbes. Studying Cephalotes rohweri, we demonstrate that abdominal trophallaxis involves the transfer of hindgut content to both minor and major workers. This transfer includes symbiotic bacteria that enter in contact 143
with solid food accumulated in the mouth cavity of adult workers (infrabuccal pellets). Our
metagenome analysis of infrabuccal pellets collected by Cephalotes rohweri suggests that hindgut bacteria may be capable of aiding digestion of recalcitrant matter such as plant and fungi tissues, in addition to recycling nitrogen and synthesizing 8 out of the 10 essential amino acids.
We provide compelling evidence showing that ill-nourished individuals tend to solicit and engage in abdominal trophallaxis at a higher rate than individuals fed a complete diet. In general, our results add a new perspective on how ants interact with symbiotic bacteria and the importance of trophallaxis in integrating social insect - gut bacteria interactions.
INTRODUCTION
By harnessing the metabolic diversity of prokaryotes, in particular Bacteria (Ornston & Yeh
1979), animals became capable of surviving and reproducing under conditions in which they would be otherwise unfit (Moya et al. 2008). The persistence of these associations is fundamentally dependent on how animals interact with their symbionts and on the mechanisms that ensure re-infection in every generation (McFall-Ngai et al. 2013). In animal societies, evolution of behavioral complexity and social behavior itself have been hypothesized to derive from mechanisms that could ensure contact with microbial symbionts through generations
(Troyer 1984; Lombardo 2008; Archie & Theis 2011). Ants are some of the most abundant and diverse social animals, and many of them have evolved symbiotic associations with microorganisms (Cook & Davidson 2006; Russell et al. 2009). Here we investigate the contribution of ant behavior in enabling colony members to benefit from the metabolic properties 144
of symbiotic bacteria, as well as how behavior mediates transmission of symbionts through
generations.
Beneficial ant-bacteria associations have been hypothesized to be key to explaining their
dominance in environments such as the canopy of tropical forests (Cook & Davidson 2006). The
majority of arboreal ants are exudate-feeding species, i.e., they consume a diet composed mostly
of extra-floral nectar and honeydew from sap-feeding insects (Blüthgen et al. 2000; Davidson et al. 2003). As functional herbivores, arboreal ants may depend on the ability of microbes to obtain nutrients that are insufficient in their diet, such as nitrogen (Wäckers & Wäckers 2005).
Some of the best studied microbial communities among exudate-feeding ants are found in the genus Cephalotes. Also known as turtle ants, Cephalotes species harbor relatively diverse bacterial communities that are found across species in the tribe Cephalotini, an association hypothesized to be present since the Eocene (Anderson et al. 2012; Sanders et al. 2014). Some of the mechanisms proposed to explain the long history of this association include adaptations such as a specialized proventriculus, a valve between the foregut and midgut that likely protects hosts from disbyosis (Lanan et al. 2016). For inheritance of microbes, Cephalotes seem to depend on abdominal trophallaxis for inoculation of newly emerged individuals (callows) (Wheeler 1984;
Roche & Wheeler 1997; Lanan et al. 2016), although this hypothesis has never never been completely tested.
Abdominal trophallaxis was first observed in turtle ants by Wilson (1976) when building an ethogram for C. varians. He described the behavior as the contact between the mouthparts of an individual and the tip of the abdomen of another ant (donor), while frequently stroking the donor abdomen with their antennae and occasionally with their foretarsi. Since then, abdominal 145
trophallaxis has been observed to occur in other Cephalotes (Roche 1996), the Cephalotini genus
Procrytocerus (Wheeler 1984), carpenter ants in the genus Camponotus (Santos et al. 2005), and in slave-making ants Protomognathus americanus (Stuart 1981), and Polyergus rufescens
(D'Ettorre et al. 2002), although in these cases this behavior may play a role other than transmission of symbionts (Wheeler 1984). In fact, the nature of the substance passed via abdominal trophallaxis has not been established. Wilson (1976) proposed that the origin of the liquid is likely the ovaries, as he observed that workers of C. varians show great interest in secretions that surround worker-laid trophic eggs. Even so, Wilson (1976) did not rule out other possible sources, such as the hindgut.
In cases where conspecifics interact, abdominal trophallaxis was initially proposed by Wheeler
(1984) as means to transmit microbes when studying the Cephalotini, Procryptocerus scabriusculus. Roche and Wheeler (1997) provided compelling evidence in support of this hypothesis by showing that workers that emerge in isolation harbored no bacteria in their guts compared to workers that were allowed social contact. Nonetheless, a clear connection to abdominal trophallaxis was still missing. It is likely that microbial symbionts are acquired by callows via rectal fluids that may be passed on via abdominal trophallaxis, but it is also important to note that this behavior also takes place in the absence of callows (Wilson 1976). In fact, in at least one species (C. varians) workers typically contain hindgut bacteria within their mouth cavity (infrabuccal pocket), likely acquired via abdominal trophallaxis (Appendix B). The infrabuccal pocket traps solid particles, such as pollen and fungus (Wheeler & Bailey 1920;
Urbani & de Andrade 1997), forming pellets of food that have been reported to be given to developing brood in different species of ants (Wilson 1976; Cole 1980; Jouvenaz et al. 1984; 146
Blatrix et al. 2012). Rodrigues et al. (unpublished) hypothesize that infrabuccal pellets are inoculated via abdominal trophallaxis with beneficial microbes capable of digesting recalcitrant matter such as plant wall components (pectin and cellulose) and the fungal wall (chitin). In addition, it is likely that abdominal trophallaxis may also transfer nutrients to workers that subsequently pass them to larvae via oral trophallaxis. This proposed process is similar to behavior observed in termites, where workers deprived of a protein diet show increased solicitation of hindgut content (proctodeal trophallaxis) from donors fed a complete diet
(Machida et al. 2001).
Here we investigated whether abdominal trophallaxis can be a link between social lifestyle and the ecology of host-microbial symbiotic interactions. Specifically we characterized the nature of abdominal trophallaxis by recording the frequency of this behavior, castes involved, and duration. We designed experiments to test (1) whether liquid transmitted by abdominal trophallaxis is from hindgut origin; (2) the role of this behavior in transmitting microbes across generations; (3) whether abdominal trophallaxis is dependent on the nutritional state of the colony; and (4) if metagenome content of the infrabuccal pellet is consistent with a digestive and nutritional role of hindgut bacteria.
METHODS
1. Experimental colonies
Colonies of Cephalotes rohweri were collected from dying branches of Palo Verde trees
(Cercidinium microphylla), in Tucson Mountain Park, with permission from Pima County 147
Natural Resources, Parks and Recreation (AZ, USA). All colonies were queenright and were
housed in artificial wood cutout nests placed inside boxes treated with Fluon to prevent ants from
escaping. Their diet was offered ad libitum and consisted of 20% honey water fed twice a week, along with fragments of freshly killed Nauphoeta cinerea cockroaches. Water was provided via cotton-stoppered test tubes and temperature was kept constant at approximately 25°C. In these conditions, we have successfully maintained C. rohweri colonies for over 4 years.
2. Behavioral observations
We video-recorded three colonies of C. rohweri to characterize the frequency, duration and castes involved in abdominal trophallaxis. As a reference, we also recorded the same information when oral trophallaxis was performed. From each colony we created queenless colony fragments, named after their source colony as C513, C813 and C913. Each colony fragment consisted of 5 pupae, 10 minor workers (“workers” hereafter), 10 major workers (“soldiers” hereafter), 10 larvae, and eggs. Therefore, a total of 60 adult ants were present at the beginning of the experiment, but this number changed as ants died and new workers and soldiers emerged from pupae. Although not our focal subjects, brood was included to generate more workers, as newly emerged workers (callows) usually perform abdominal trophallaxis within the first few hours after pupation. Callows are distinguished by the light coloration of their cuticle, which darkens as sclerotization is completed. For the purposes of our behavioral observations, callows are considered to be a separate behavioral caste, although it is not clear whether they perform any task in the colony while sclerotizing. Since eggs and larvae of social insects usually carry the fertility signal from the queen, which inhibits workers and soldiers from developing their ovaries
(Endler et al. 2004; Endler et al. 2006), we can assume that the social environment in these 148
queenless colony fragments is close to that of a queenright colony. Each colony fragment was housed in an artificial nest made of an Ethylene-vinyl acetate (EVA) cutout chamber sandwiched between two microscope slides, allowing us to record activities inside the nest. Each microscope slide was covered by red cellophane, to minimize agitation or other stress responses related to the light used during the recording of the videos.
Colony fragments were given one week to habituate to their new nesting conditions before recordings took place. Each colony was video-recorded on five different days (morning and afternoon) between October 6th and October 17th, 2014. From the total 90 hours of video, a subset of 10 hours per colony was selected across all days of recording. In each video we noted:
(1) nature of the trophallactic interaction: oral or abdominal; (2) the castes that were interacting
(in the case of abdominal trophallaxis, we noted which caste was donor and which caste was the receiver); and (3) duration of the interaction. Occasionally, we also noted if an individual performing oral trophallaxis had just received abdominal trophallaxis within 2-3 minutes prior to the current interaction, which may indicate an alternative route of transmission of gut content.
We used stringent criteria for recording interactions: (1) for abdominal trophallaxis (see Figure
1), an individual had to clearly contact its mouthparts with the tip of the abdomen of a donor individual; typically, it tilted its head in an angle around 30-45° in relation to the floor of the nest and spent at least 1 second in that position, during which time it repeatedly touched the abdomen of the donor with its antennae; (2) for oral trophallaxis, we only recorded workers that clearly touched mouthparts for at least one second, with their heads typically tilted towards each other
(Figure 1). 149
All colonies were kept on the same diet, temperature and water availability as their source colonies. To evaluate whether colonies were healthy when subjected to these nesting conditions, and to be able to relate observations to nest composition, we censused colony demography two to three times a week, for over two months, including the period of the observations.
To verify if the frequency of trophallactic acts was a function of the number of workers and callows present in each colony, we estimated a ratio of interactions per individual, using data from our weekly surveys of colony demography. For abdominal trophallaxis, we first summed the number of instances each caste was seen acting as a receiver, per interacting pair and per day of observation. Next, we divided this sum by the number of individuals of the same caste that were present in the same day. Finally, we calculated the mean ratio of interactions for all days, for each pair performing abdominal trophallaxis, along with its associated standard error. For oral trophallaxis, a similar calculation was performed. Since we could not determine donor or receiver in oral trophallaxis, we chose to sum, for each date, the number of instances in which each caste was seen performing oral trophallaxis, but we ignored the identity of the other interacting caste. This sum was then divided by the number of individuals recorded for the same caste in the same day. With this information, the estimated mean ratio of interactions and standard errors were calculated. These values are to be interpreted cautiously because we only recorded interactions inside the nest, but not all individuals were inside the nest during observations. In addition, for a third of the videos we had to utilize colony demography information from data recorded a day after the observations.
3. Liquid transmission and hypothetical nutritional role of abdominal trophallaxis 150
To determine (1) whether the liquid received during abdominal trophallaxis comes from the hindgut, and (2) whether deficiency in nutrition of an individual is associated with an increase in the number of times that individual solicits abdominal trophallaxis, we designed an experiment in which we could track the fate of rectal fluids of a set of workers (“donors”) that were only allowed to interact with other workers (“receivers”) via abdominal trophallaxis. At the same time, receivers were divided into groups that were fed a diet either rich or poor in protein, whereas the donor group was always fed a diet rich in protein. This setup allowed us to measure whether the frequency with which receivers solicited abdominal trophallaxis was associated with their nutritional status.
We used workers from six colonies (C1, C6, C613, C813, C814, and C913) and, for each colony, workers were removed and assigned to one of the following three groups: “Blue Group (BG)”,
“Honey and beads (HB)”, and “Honey, beads, cockroach, and SPAM ® (HBCS)”. BG contained workers (medium and large sized, including soldiers) collected from inside their source nest (20 workers) and from outside their nests (3 workers; foragers). A honey water blue solution (30% honey (v/v) and 0.1% Methylene Blue) was fed to BG workers, twice a week, for two and half weeks, together with freeze-killed, cockroach fragments and SPAM ®. The other two groups
(HB and HBCS) were composed each by 10 minor workers and 5 medium to large sized larvae.
The diet of these two groups consisted of non-dyed 30% honey water with either a protein source
(HBCS) or deficient in protein (HB). PMMA microbeads of assorted sizes (6um to 45um;
Polysciences) were added to the diet, because in pilot experiments we observed that when a diet lacks particulate food, such as the honey water only treatment, there was no formation of an infrabuccal pellet in workers. The infrabuccal pellet is important for detection of small volumes 151
of blue solution that may not be sufficient to dye the rest of the digestive tube. The concentration of beads was 0.4g/100mL of honey water. Similarly to BG, groups HB and HBCS were fed their respective diets twice a week, for two and a half weeks. Each group was housed in an artificial nest that consisted of a small petri dish (5cm in diameter), covered with red cellophane to simulate a dark nest chamber, as ants are assumed to be unable to see in the red wavelength. The nest was kept inside a larger Petri dish (14cm in diameter), which contained feeding dishes for honey water and protein food (cockroach and SPAM ®). A small hole of approximately 0.5 cm at the side of the nesting dish allowed ants to access their foraging arena.
After the two and a half weeks of habituation and diet treatment, we updated the composition records for every colony, as some workers died during the experiment. Workers of
Cephalotes rohweri often drowns or get stuck in the honey water solution, which is their primary cause of death. Next, BG workers were added to new nests containing either HB or HBCS treated workers, for the same replicate (colony). To ensure that BG workers could only interact via abdominal trophallaxis with receiver workers, we placed BG workers in an oral-trophallaxis excluder we developed. BG workers were briefly exposed to a block of ice to induce a state of inactivity, and a small droplet of water-soluble, non-toxic white glue (Elmer’s ®) was added to the top of their prothorax, which was then glued against the inner surface of a 10ul pipette tip cut in half (Figure 2). Next, glue was added to the borders and extremities of the pipette tip half, and glued against the nest surface (Figure 2). We allowed 30 minutes for the glue to dry and to ensure that ants were well secured and unable to escape. The same process was repeated for every other donor ant (BG), until two nests containing each approximately 10 glued donor ants were prepared for each colony. Each of the new nests received the respective matching colony 152
HB or HBCS group, forming the following experimental groups: HB-BG and HBCS-BG. In pilot experiments, we observed that soldiers survive for over a week while glued to the oral- trophallaxis excluder, and the excluder does not prevent receivers from soliciting abdominal trophallaxis.
We allowed experimental groups one hour of habituation to the new nesting conditions. Next, we video-recorded the HB-BG and HBCS-BG groups belonging to colonies C1 and C6, for three hours each. The frequency of abdominal trophallaxis was quantified from these videos by a blind observer who was unaware of the treatments and expected outcomes of this experiment. The same stringent criterion as described earlier was used to quantity abdominal trophallaxis. After three days, all workers from all colonies were freeze-killed and later dissected to identify the presence of blue liquid in their infrabuccal pockets.
4. Infrabuccal pocket metagenome: potential nutritional functions
We characterized the microbiota present in infrabuccal pellets of field-collected Cephalotes rowheri and described their potential functions based on a shotgun-metagenome analysis. In addition, we used metagenomic data to characterize the composition of infrabuccal pellets, i.e. plants, fungus and other organisms that might be present. Fifteen to twenty workers (major and minor) were collection from one colony (colony PAR 0115) located in Tucson Mountain Park
(with permission from Pima County, Arizona, USA). Workers were immediately flash frozen in dry ice, brought to the laboratory and transferred to a -80 °C freezer. We started by dissecting only minor workers and aimed to collect a minimum of ten infrabuccal pellets. Pellets were pooled and subjected to DNA extraction. We also recorded the relative size and coloration of pellets found in our sampling. 153
Prior to dissections, all tools and work space were sterilized with a chlorine bleach solution (8%) and tools were flame sterilized throughout the dissections. Each ant was washed for 1min in 75% sterile ethanol, followed by a 30 seconds wash in molecular grade water. Wash solutions were replaced for every 5 ants washed. To dissect infrabuccal pellets, we removed and carefully pinned the head of a worker through the foramen orifice on a sterile surface of freshly solidified wax, with the head dorsal side upwards and tilted upwards at approximately 45 degrees, such that mouthparts were visible. Next, we removed mandibles and made a cut across their openings and between the clypeus-labrum and the labium. In the opposite direction, we cut the head capsule from the mandible orifices towards the occipital foramen, thus dividing the head capsule into dorsal and ventral halves. The dorsal half was carefully pushed up along the pin holding the lower half in place, which in most cases resulted in destruction of the upper surface of the infrabuccal pocket membrane, exposing an intact pellet. In some cases we used iris scissors to further remove the infrabuccal pocket tissue in order to further expose the pellet. With a fine tip syringe, we carefully dislodged the pellet and transferred it into a lysis buffer solution in a 1.5ml microcentrifuge tube, kept in ice. The same process was repeated for all ants, with surfaces sterilized between dissections and tools flame-sterilized throughout dissections and particularly when external surfaces were touched.
Next, infrabuccal pellets were treated with lysozyme at a concentration of 20mg/ml, and incubated for 3h at 37 °C. DNA was extracted with a QIAamp DNA Micro Kit (Qiagen) following the manufacturer’s protocol, with a modified proteinase K digestion step at 56 °C overnight. Two 50ul elutions were done and the one with the highest concentration and quality was submitted to a sequencing facility (MR DNA Next Generation Sequencing Service Provider, 154
Texas, USA). In order to ensure that enough material would be sequenced, whole genome amplification was performed prior to sequencing, using REPLI-g Mini kit (Qiagen). The DNA sample was then subjected to fragmentation and a paired-end 2x300bp Ilumina MiSeq run (600 cycles). The protocol used by the sequencing facility is available at the Supplementary Material
(S1).
Data analysis
To compare the presence or absence of blue dye in the infrabuccal pellets of Cephalotes, we used a generalized linear model to perform a logistic regression, using presence/absence as our response variable and treatment, blocked by colony, as our explanatory variable. We use chi- square to compare the frequencies of abdominal trophallaxis acts observed in the group of receivers treated with a protein diet (HBCS) versus receivers that were fed a protein deficient diet (HB). All statistical tests, as well as descriptive statistical reports for our behavioral observations, were performed in R v. 3.2.5 using R Studio v. 0.99.879 (RStudio Team 2015).
The following R packages were used for data wrangling and graphing: “ggplot2” (Wickham
2009), “reshape2”(Wickham 2007), and “plyr”(Wickham 2011).
Metagenome data was analyzed using the MG-RAST pipeline and platform (Meyer et al. 2008;
Wilke et al. 2016). In summary, paired-end data was joined in MG-RAST, quality-filtered to keep only sequences with a qc-score higher than 15, de-replicated, and filtered to exclude any sequence that could be associated with human DNA. MG-RAST pipeline has an automated system for screening data for taxonomic classification as well as functional profiling, sourcing a large range of databases. For classification, we used a cutoff of 60% identity, and an e-value of
1x10^-10 which indicates the probability of a different match for a given sequence. For 155
functional profiling , we used a higher e-value, 1x10^-5, in order to keep a larger number of sequences that represent genes variants that might be poorly represented in the database sources, but that still share homology to well-known characterized organisms. Using the “KEGG- mapper” tool, we looked for enzymes and pathways that might be involved in (1) the digestion of pollen substrate (cellulose and pectin), urea and uric acid, (2) nitrogen fixation, and (3) synthesis of essential amino acids.
RESULTS
1. Behavioral observations
A total of 203 acts of abdominal trophallaxis and 1514 acts of oral trophallaxis were recorded.
Abdominal trophallaxis tended to last over 25 seconds (32 ± 6.63 s 95% CI), ranging from single contacts that could last 1 second to over 6 minutes (386s), for a median duration of 16 seconds
(Figure 3A). All pair combinations among worker, soldier and callow were observed to engage in abdominal trophallaxis, with the highest frequency recorded for major workers performing this behavior as donors and minor workers as receivers, followed by interactions between pairs of minor workers (Figure 3A). In contrast, callows performing abdominal trophallaxis as receivers showed some of the lowest frequencies of interactions, despite the fact that callows were present during all of the observations. The highest frequency for callows as receivers was observed within one single hour of interactions recorded for colony 913, involving a callow that had just emerged in the same day and a soldier that performed the task as donor. On the other hand, callows receiving abdominal trophallaxis were in average the longest interactions we observed: 166 ± 66s (SE). Typically, callows engaged in multiple sequential contacts during 156
abdominal trophallaxis. On one occasion, for instance, a single callow from C913 engaged in four contacts spaced 4-90 seconds intervals apart, and with durations ranging from 66 to 353 seconds, with a single major worker as donor. For comparison, the frequency of oral trophallaxis was highest among minor workers, followed by interactions between minor workers and major workers (Figure 3B). Contacts tended to last about 20 seconds (19.66 ± 1.84s 95% CI), ranging from 1 second-long contacts to 803 seconds (worker-worker), with a median of 9 seconds.
The frequency of interactions per caste might be a function of the number of individuals representing each caste. When we account for differences in the size of each caste population in a colony, we observe that the behavior of soliciting and receiving abdominal trophallaxis interactions was more common in callow interactions, particularly for colony C913 (Figure 4A).
For colonies C513 and C813, the distribution of interactions was more uniform among castes.
When the same adjustment was made for the frequencies observed for oral trophallaxis, a striking different pattern emerges, showing that callows engage in oral trophallaxis at a much higher frequency (per individual) than the other castes (Figure 4B).
Finally, secondary interactions were observed occasionally. In total, in 55 records of oral trophallaxis at least one individual had just performed abdominal trophallaxis within the previous few minutes. These interactions were not initially planned as tracking individuals through the colony would require specific color marking to distinguish nestmates. However, given the fact that we only recorded 203 abdominal trophallaxis acts in this study, 55 records of oral trophallaxis immediately following abdominal trophallaxis may represent a common occurrence.
This is evidence of a potential secondary route for sharing of hindgut content among nestmates. 157
The complete dataset with observations and metadata is available in the supplementary material
(S2).
2. Liquid transmission and hypothetical nutritional role of abdominal trophallaxis
From a total 100 individuals that were dissected, regardless of treatment, signs of rectal liquid transfer were detected in the infrabuccal pellets of only 27 individuals. However, we noted that one donor escaped from the oral trophallaxis excluder in colony 913. Therefore, we excluded colony C913 from further analyses, which included five pellets from the sum above. Typically, pellets were dyed in green (S3, Supplementary Material) or had only small specks of blue dye.
We dissected the guts of donors and all showed strong concentration of blue dye in their crops and infrabuccal pockets, although their hindguts were often turquoise resembling the color of receiver pellets (S3, Supplementary Material). This indicates that the blue dye changes color as it reaches the rectum. It may also be possible that the liquids that are transferred via abdominal trophallaxis are in volumes so small that the dye was not concentrated enough to be detectable in all receivers dissected. Despite the relatively low frequency of dyed pellets, we found that groups fed a diet deficient in protein (HB-BG) had a significantly higher frequency of dyed pellets compared to groups fed a complete diet (P<0.0001; Table 1). We also found that colony identity influenced the outcome of these interactions (P<0.01, Table 1). For instance, whereas in colony
C1 we found dyed pellets only in two workers of group HB-BG, in contrast to one worker from group HBCS-BG, in colony C613 we found 9 in HB-BG versus 4 from HBCS. The complete dataset for this experiment is available as Supplemental Material (S5).
The video-analysis supported the results found in our dissections, by showing that for both C1 and C6, the frequency of abdominal trophallaxis in HB-BG groups was higher than the observed 158
for the HBCS-BG group, although only significant in C1 (Figure 5). The duration of interactions
was on average 25.21 ± 17.19s (95% CI) for C1 HB-BG (n=19) and only 4 seconds long for the
only interaction recorded for C1HBCS-BG. Similarly, abdominal trophallaxis from receivers of
C6 HB-BG interacted on average for 4.71 ± 1.27s (95% CI) (n=7), whereas C6 HBCS-BG
interacted in average for 1 second with donors (n=2). It is noteworthy also that we frequently
noticed that some individuals solicited abdominal trophallaxis from several donors one after the
other. These individuals would solicit but not engage in abdominal trophallaxis, moving from
donor to donor. This is a curious behavior that was also noted by Wilson (1976) in C. varians, which described workers showing a “hunger” for the material received via abdominal trophallaxis. This is the first record since Wilson (1976) of this behavior and the first for C. rohweri.
Overall, our results are consistent with the hypothesis of liquid transfer via abdominal trophallaxis, and show an association between frequency of this behavior and nutritional state.
Nonetheless, our results should be interpreted cautiously as it is possible that the blue dye degraded in the guts of donors, which may have biased our results. In addition, only 28 acts of abdominal trophallaxis were recorded in a total of 12 hours of observation (3 hours x 2 experimental group x 2 colonies). Larger sample size and longer hours of observation are imperative for further support of these patterns.
3. Infrabuccal pocket metagenome: potential nutritional functions
The metagenome sequencing of infrabuccal pellets (colony PAR0115) yielded 2,376,611 sequences, with a mean length of 301 ± 78bp. After quality-filtering, 53.5% of the data was excluded from further analyses, resulting in 1,188,499 high quality sequences, with a mean 159
length of 251± 105bp. With data aligned, 13,121 rRNA sequences were identified, 381,671 hits
were classified as proteins and 179,446 functional categories were annotated.
Pellets were composed mostly by Bacteria (83%), followed by Virus (1%), Eukaryotes (0.4%),
and Archaea (0.05%) (Figure 6A). The bacterial community contained in the pellets included
members of the core community found in the gut of C. rohweri, such as Rhizobiales (46.8% of bacterial sequences), Burkholderiales (16.3%), Xanthomonadales (5.5%), and Verrumicrobiales
(0.2%). The low abundance of reads other than Bacteria indicates that particles that composed the pellet had their DNA either digested before extraction or unavailable inside cell walls that could not be lysed by our methods. Within eukaryotes, the most abundant groups included
Arthropoda (33%), Ascomycota (26%), Chordata (16%) and Streptophyta (14%). Plants included within Stretophyta were mostly represented by Euphorbiaceae (40%) and Salicaceae (30%), followed by Poaceae (12%) and Fabaceae (3%) (Figure 6B). At a closer look, we found that 99% of the Euphorbiaceae sequences matched the genus Ricinus, known to be a wind-pollinated plant.
The same is true for Salicaceae, where 100% of the sequences matched the wind-pollinated cottonwood genus Populus. Members of Poacaeae, the grass family, are also primarily wind- pollinated. These results strongly suggest that the majority of pollen found in the infrabuccal pellet comes from wind-pollinated plants. Among Ascomycota, most abundant groups included the Aspergillus containing class Eurotiomycetes (36%), Sordariomycetes (28%), that contain endophytes and plant pathogens such as blight causing Cryphonectria parasitica, and
Dothideomycetes (26%), known to contain endophytes and to associate with lichens. We suggest that the relative ratio of fungus to plants recorded in our study show that fungal spores may represent an overlooked but important representative of the diet of Cephalotes. However, we 160
recognize that our DNA extractions were not designed to break down walls of fungus or plants,
which may have biased the amount of genetic material that could be recovered from these
organisms.
Our functional analyses revealed that pectin, as well as chitin digestion may not be mediated by
microorganisms in the infrabuccal pellets (Figure 7A-B). Nonetheless, chitinase may still be
synthesized by the host. Fungivorous ants are known to be able to synthesize chitinase (Febvay et al. 1984; Nygaard et al. 2011). Other ant genera with available genomes show the ability to synthesize this enzyme (Bonasio et al. 2010; Wurm et al. 2011). The pathway for cellulose digestion was completely represented in our metagenome analysis (Figure 7B), which may indicate that at least partial breakdown of pollen tissues is possible. For the roles of the microbiome in increasing nitrogen availability, we found a nearly-complete pathway for uric acid digestion, with only one enzyme missing, urate oxidase. Genes that synthesize this enzyme have been reported to be present in ants of different genera (Bonasio et al. 2010; Nygaard et al.
2011; Wurm et al. 2011), which may indicate that a complementary metabolism between gut bacteria and host may exist in Cephalotes. A complete pathway for nitrogen recycling via urea degradation was also found, which may make nitrogen available for synthesis of essential and non-essential amino acids. Complete pathways for the synthesis of 8 out of the10 essential amino acids (for insects) were also found: Arginine, Isoleucine, Leucine, Methionine, Phenylalanine,
Tryptophan, Threonine, and Valine. Histidine and Lysine pathways were nearly complete with only few steps missing. We did not find evidence of nitrogen fixation.
DISCUSSION 161
In order to gain insight into the role of behavior in mediating ant-bacteria interactions, we performed a comprehensive study on trophallaxis in Cephalotes. We demonstrate that abdominal trophallaxis is a multi-faceted behavior, mediating the transfer of microbes to food substrate as well as inoculation of new adult member of the colony. In addition, the rate at which abdominal trophallaxis is solicited is affected by nutritional state of the colony, consistent with the hypothesis that rectal fluids are rich in nutrients. Moreover, our metagenomic analysis reveals that the microbiome of infrabuccal pellets, mostly composed of transferred bacteria from the hindgut, is capable of upgrading hosts’ diet with nutrients such as essential amino acids. Finally, our study is the first to investigate, at the molecular level, the composition of infrabuccal pellets, showing that besides the dominance of hindgut bacteria, pollen from mostly wind-pollinated plants and fungal groups are also represented. These results change the current perspective of how ant-bacteria interactions work, as described below. In the next sections we discuss the generality of these results in relation to other social insects and in insect-bacteria symbioses in general.
Multi-level interactions between host and microbiota in social insects: trophallaxis as an integrator
Whereas intracellular bacteria can be transmitted via ovarian tissues to new generations, transmission of extracellular bacteria depends on other mechanisms (Salem et al. 2015). In social animals, behavior may mediate maintenance of symbionts at both the individual and colony levels of the host species. In the honeybee Apis mellifera, most of the typical gut bacterial community can be acquired through brood cell cleaning and trimming (Anderson et al. 2016), a task primarily observed in 1-3 day old adult individuals (Seeley 1982). In Attini leaf-cutter ants, 162
a complex system of behaviors and castes ensures that their symbiotic fungus is maintained and
transmitted across generations (Hölldobler & Wilson 2011). More generally, however,
transmission of symbiotic bacteria via trophallactic behavior seems to be a shared trait among
social insects.
To our knowledge, oral or stomodeal trophallaxis has been reported to be involved in the
transmission of beneficial symbionts only in the honey bee (Powell et al. 2014). Nonetheless, we
observed that newly emerged workers of C.rowheri participate in oral trophallaxis at a much higher rate than other castes, which may represent an additional route for sharing of microbes.
This finding is also consistent with the fact that bacterial communities are compartmentalized along the digestive tube of Cephalotes, with some of phylotypes of the core microbiota only found in the crop (Appendix B). These crop-specific groups are, therefore, more likely to be acquired via oral trophallaxis than abdominal trophallaxis. We also observed that workers that have engaged in abdominal trophallaxis may subsequently engage in oral trophallaxis, which in turn may transmit gut microbes among nestmates. It is likely that oral trophallaxis might represent at least a secondary route of transmission of beneficial microbes in social insects, whereas abdominal and proctodeal trophallaxis might represent more specialized behaviors, likely derived from coprophagy (Nalepa 1994).
Sharing of symbionts and nutrients via oral and abdominal trophallaxis expand symbiosis from the individual to the colony as a whole, including the growing brood that may represent the end destination of nutrients acquired via bacterial metabolism. Given the benefits of these associations and their putative association with trophallaxis, one prediction is that social insects that lack trophallaxis may also lack vertically inherited gut symbionts. At least in ants, groups 163
that completely lack trophallaxis, including most ponerine ants which typically consume a
carnivorous diet (Hölldobler 1985), are relatively uncommon. Bacteria in the guts of ponerine
ants have been reported, but their role as beneficial microbes remains controversial (Oliveira et
al. 2016). Further studies are required to clarify the evolution of trophallaxis, social behavior and
sharing of symbiotic microorganisms.
Parallels between food pellets and fecal pellets
Infrabuccal pockets have been traditionally thought to work exclusively as a filtering organ,
preventing solid particles from reaching the crop, where they would become trapped (Eisner &
Happ 1962). At the same time, infrabuccal pellets have been hypothesized to be a form of food
for larvae. Indeed, in the infrabuccal pockets of larvae a variety of food items have been found,
including arthropod parts, pollen, and fungal spores (Wheeler & Bailey 1920). The surfaces of
the bodies of ants collect a variety of fungal spores and pollen (Cembrowski et al. 2015) that are
in turn collected during self- and allo- grooming and likely accumulate in the infrabuccal pocket.
The presence of hindgut bacteria in the infrabuccal pocket, along with particles of pollen and
fungus, in the infrabuccal pocket, as well as observations that pellets are given to larvae, all
suggest an alternative route of nutrient acquisition by ants that is worth further investigation.
Both social cockroaches and termites produce fecal pellets that contain partially digested detritus
and plant fibers that come into contact with gut bacteria (Nalepa et al. 2001). These pellets are
re-ingested and increase the efficiency by which nutrients are extracted from their diet. In a
parallel with fecal pellets, infrabuccal pellets of Cephalotes may also represent food that has been inoculated with beneficial gut bacteria that is subsequently regurgitated and transferred to larvae for further digestion and nutrient extraction. Our metagenomic data support the idea that 164
hindgut bacteria have the potential to at least partially digest components of the infrabuccal
pellet. In addition, mandibular and salivary glands, which open in the mouth cavity of ants
(Hölldobler & Wilson 1990), may also contribute to the digestion of food pellets prior to larval
feeding.
A caste specialized for storing microbes?
In social insects, major workers are called soldiers when they play roles related to defense of the
nest, but tasks associated with major workers may also include specialization for food processing
and a role as a live storage of nutrients (Hölldobler & Wilson 1990; Tsuji 1990; Grüter et al.
2012). Our observations suggest that an additional role that major workers may play is serving as
donors of rectal fluids during abdominal trophallaxis. While this is not an exclusive role, major
workers were observed to perform this task at a higher rate than other castes. In Cephalotes
rohweri, major workers assist in nest maintenance and brood care (personal observation), but
they are particularly specialized for defense by blocking the nest entrance with their heads
(Powell 2008). Based on our observations, we suggest that major workers may serve to store
nutrients and microbes that can be shared via abdominal trophallaxis. This function overlaps in
part with the function of repletes, but with the additional role of inoculating newly emerged
individuals and food pellets with hindgut bacteria. The production of major workers requires a
big investment in energy by the colony (Hölldobler & Wilson 2009), which make these
individuals particularly valuable and hard to replace. Perhaps, for this reason, major workers are
rarely seen foraging, including in Cephalotes rohweri colonies (personal observation), due to the higher risk of mortality outside of the nest (Dornhaus & Powell 2010). In the protected nest 165
environment, with relatively low-energy demanding tasks, soldiers might be analogous to
culturing medium that keeps a reliable source of inoculum, particularly to new adult workers.
Conclusions
Trophallaxis is likely a universal behavior in social insects that mediates symbiotic relationships
with beneficial gut microorganisms. Via trophallaxis, individual and colony are able to harness
the metabolic properties of their bacteria, in addition to serving as a medium for vertical
transmission of these beneficial symbionts. Cephalotes transmit hindgut content, including
bacteria and possible nutrients, via abdominal trophallaxis. Oral trophallaxis may also play a role
in mediating transmission of hindgut bacteria, but further investigation is necessary to test this
hypothesis. With the aid of metagenomic data, we infer that gut bacteria are able to digest the
content of infrabuccal pellets, in addition to synthesizing essential amino acids. Collectively,
these findings change many of our current views on the nutritional ecology of ants as well as
their interactions with symbionts. Future research on the roles of infrabuccal pellets on nutrition
of the colony, as well as surveys of microbes and trophallaxis behavior in other ant genera may
shed light on the integrative roles of trophallaxis in symbiont-individual-colony interactions.
REFERENCES
Anderson K.E., Rodrigues P.A.P., Mott B.M., Maes P. & Corby-Harris V. (2016). Ecological succession in the honey bee gut: shift in Lactobacillus strain dominance during early adult development. Microb. Ecol., 71, 1008-1019. Anderson K.E., Russell J.A., Moreau C.S., Kautz S., Sullam K.E., Hu Y., Basinger U., Mott B.M., Buck N. & Wheeler D.E. (2012). Highly similar microbial communities are shared among related and trophically similar ant species. Mol. Ecol., 21, 2282-96. Archie E.A. & Theis K.R. (2011). Animal behaviour meets microbial ecology. Anim. Behav., 82, 425-436. 166
Blatrix R., Djiéto-Lordon C., Mondolot L., La Fisca P., Voglmayr H. & McKey D. (2012). Plant- ants use symbiotic fungi as a food source: new insight into the nutritional ecology of ant– plant interactions. Proc. R. Soc. Lond., Ser. B: Biol. Sci., 279, 3940-3947. Blüthgen N., Verhaagh M., Goitía W., Jaffé K., Morawetz W. & Barthlott W. (2000). How plants shape the ant community in the Amazonian rainforest canopy: the key role of extrafloral nectaries and homopteran honeydew. Oecologia, 125, 229-240. Bonasio R., Zhang G., Ye C., Mutti N.S., Fang X., Qin N., Donahue G., Yang P., Li Q., Li C., Zhang P., Huang Z., Berger S.L., Reinberg D., Wang J. & Liebig J. (2010). Genomic comparison of the ants Camponotus floridanus and Harpegnathos saltator. Science, 329, 1068-1071. Cembrowski A.R., Reurink G., Arcila Hernandez L.M., Sanders J.G., Youngerman E. & Frederickson M.E. (2015). Sporadic pollen consumption among tropical ants. Insectes Soc., 62, 379-382. Cole B.J. (1980). Repertoire convergence in two Mangrove ants, Zacryptocerus varians and Camponotus (Colobopsis) sp. Insectes Soc., 27, 265-275. Cook S.C. & Davidson D.W. (2006). Nutritional and functional biology of exudate-feeding ants. Entomol. Exp. Appl., 118, 1-10. D'Ettorre P., Mondy N., Lenoir A. & Errard C. (2002). Blending in with the crowd: social parasites integrate into their host colonies using a flexible chemical signature. Proc. R. Soc. Lond., Ser. B: Biol. Sci., 269, 1911-1918. Davidson D.W., Cook S.C., Snelling R.R. & Chua T.H. (2003). Explaining the abundance of ants in lowland tropical rainforest canopies. Science, 300, 969-972. Dornhaus A. & Powell S. (2010). Foraging and defence strategies. In: Ant Ecology (eds. Lach L, Parr CL & Abbott KL). Oxford University Press Oxford, UK, pp. 210-230. Eisner T. & Happ G.M. (1962). The infrabuccal pocket of a formicine ant: a social filtration device. Psyche, 69, 107-116. Endler A., Liebig J. & Hölldobler B. (2006). Queen fertility, egg marking and colony size in the ant Camponotus floridanus. Behav. Ecol. Sociobiol., 59, 490-499. Endler A., Liebig J., Schmitt T., Parker J.E., Jones G.R., Schreier P. & Hölldobler B. (2004). Surface hydrocarbons of queen eggs regulate worker reproduction in a social insect. Proc. Natl. Acad. Sci. USA, 101, 2945-2950. Febvay G., Decharme M. & Kermarrec A. (1984). Digestion of chitin by the labial glands of Acromyrmex octospinosus Reich (Hymenoptera: Formicidae). Can. J. Zool., 62, 229-234. Grüter C., Menezes C., Imperatriz-Fonseca V.L. & Ratnieks F.L.W. (2012). A morphologically specialized soldier caste improves colony defense in a neotropical eusocial bee. Proc. Natl. Acad. Sci. USA, 109, 1182-1186. Hölldobler B. (1985). Liquid food transmission and antennation signals in ponerine ants. Isr. J. Entomol., 19, 89-99. Hölldobler B. & Wilson E.O. (1990). The ants. Harvard University Press, Cambridge, Mass. Hölldobler B. & Wilson E.O. (2009). The superorganism : the beauty, elegance, and strangeness of insect societies. 1st edn. W.W. Norton, New York. Hölldobler B. & Wilson E.O. (2011). The leafcutter ants : civilization by instinct. 1st edn. W. W. Norton & Company, New York. 167
Jouvenaz D.P., Ellis E.A. & Lofgren C.S. (1984). Histopathology of the tropical fire ant, Solenopsis geminata, infected with Burenella dimorpha (Microspora: Microsporida). J Invertebr Pathol, 43, 324-332. Lanan M.C., Rodrigues P.A., Agellon A., Jansma P. & Wheeler D.E. (2016). A bacterial filter protects and structures the gut microbiome of an insect. Isme J, 10, 1866-1876. Lombardo M.P. (2008). Access to mutualistic endosymbiotic microbes: an underappreciated benefit of group living. Behav. Ecol. Sociobiol., 62, 479-497. Machida M., Kitade O., Miura T. & Matsumoto T. (2001). Nitrogen recycling through proctodeal trophallaxis in the Japanese damp-wood termite Hodotermopsis japonica (Isoptera, Termopsidae). Insectes Soc., 48, 52-56. McFall-Ngai M., Hadfield M.G., Bosch T.C.G., Carey H.V., Domazet-Lošo T., Douglas A.E., Dubilier N., Eberl G., Fukami T., Gilbert S.F., Hentschel U., King N., Kjelleberg S., Knoll A.H., Kremer N., Mazmanian S.K., Metcalf J.L., Nealson K., Pierce N.E., Rawls J.F., Reid A., Ruby E.G., Rumpho M., Sanders J.G., Tautz D. & Wernegreen J.J. (2013). Animals in a bacterial world, a new imperative for the life sciences. Proc. Natl. Acad. Sci. USA, 110, 3229-3236. Meyer F., Paarmann D., D'Souza M., Olson R., Glass E.M., Kubal M., Paczian T., Rodriguez A., Stevens R., Wilke A., Wilkening J. & Edwards R.A. (2008). The metagenomics RAST server - a public resource for the automatic phylogenetic and functional analysis of metagenomes. BMC Bioinformatics, 9, 386. Moya A., Pereto J., Gil R. & Latorre A. (2008). Learning how to live together: genomic insights into prokaryote-animal symbioses. Nat. Rev. Genetics, 9, 218-29. Nalepa C.A. (1994). Nourishment and the evolution of termite eusociality. In: Nourishment and Evolution in Insect Societies (eds. Hunt JH & Nalepa CA). Westview Press Boulder, pp. 57-104. Nalepa C.A., Bignell D.E. & Bandi C. (2001). Detritivory, coprophagy, and the evolution of digestive mutualisms in Dictyoptera. Insectes Soc., 48, 194-201. Nygaard S., Zhang G., Schiott M., Li C., Wurm Y., Hu H., Zhou J., Ji L., Qiu F., Rasmussen M., Pan H., Hauser F., Krogh A., Grimmelikhuijzen C.J., Wang J. & Boomsma J.J. (2011). The genome of the leaf-cutting ant Acromyrmex echinatior suggests key adaptations to advanced social life and fungus farming. Genome Res., 21, 1339-48. Oliveira T.B., Ferro M., Bacci M., Souza D.J., Fontana R., Delabie J.H.C. & Silva A. (2016). Bacterial communities in the midgut of ponerine ants (Hymenoptera: Formicidae: Ponerinae). Sociobiology, 63, 637-644. Ornston L.N. & Yeh W.K. (1979). Origins of metabolic diversity: evolutionary divergence by sequence repetition. Proc. Natl. Acad. Sci. USA, 76, 3996-4000. Powell J.E., Martinson V.G., Urban-Mead K. & Moran N.A. (2014). Routes of acquisition of the gut microbiota of the honey bee Apis mellifera. Appl Environ Microbiol, 80, 7378-87. Powell S. (2008). Ecological specialization and the evolution of a specialized caste in Cephalotes ants. Funct Ecol, 22, 902-911. Roche R.K. (1996). Nutritional biology of the turtle ant, Zacryptocerus rohweri, morphological specializations of the digestive tract and associated behaviors. In. The University of Arizona. Tucson, AZ. 168
Roche R.K. & Wheeler D.E. (1997). Morphological specializations of the digestive tract of Zacryptocerus rohweri (Hymenoptera: Formicidae). J. Morphol., 234, 253-262. RStudio Team (2015). RStudio: integrated development for R. . RStudio, Inc., Boston, MA, USA. Russell J.A., Moreau C.S., Goldman-Huertas B., Fujiwara M., Lohman D.J. & Pierce N.E. (2009). Bacterial gut symbionts are tightly linked with the evolution of herbivory in ants. Proc. Natl. Acad. Sci. USA, 106, 21236-41. Salem H., Florez L., Gerardo N. & Kaltenpoth M. (2015). An out-of-body experience: the extracellular dimension for the transmission of mutualistic bacteria in insects. Proc. R. Soc. Lond., Ser. B: Biol. Sci., 282. Sanders J.G., Powell S., Kronauer D.J., Vasconcelos H.L., Frederickson M.E. & Pierce N.E. (2014). Stability and phylogenetic correlation in gut microbiota: lessons from ants and apes. Mol. Ecol., 23, 1268-83. Santos J.C., Yamamoto M., Oliveira F.R. & Del-Claro K. (2005). Behavioral repertory of the weaver ant Camponotus (Myrmobrachys) senex (Hymenoptera: Formicidae). Sociobiology, 45, 1-11. Seeley T.D. (1982). Adaptive significance of the age polyethism schedule in honeybee colonies. Behav. Ecol. Sociobiol., 11, 287-293. Stuart R.J. (1981). Abdominal trophallaxis in the slave-making ant, Harpagoxenus americanus (Hymenoptera: Formicidae). Psyche, 88, 331-334. Troyer K. (1984). Microbes, herbivory and the evolution of social behavior. J.Theor. Biol., 106, 157-169. Tsuji K. (1990). Nutrient storage in the major workers of Pheidole ryukyuensis (Hymenoptera : Formicidae). Appl. Entomol. Zool., 25, 283-287. Urbani B.C. & de Andrade L.M. (1997). Pollen eating, storing, and spitting by ants. Naturwissenschaften, 84, 256-258. Wäckers F.L. & Wäckers F.L. (2005). Suitability of (extra-)floral nectar, pollen, and honeydew as insect food sources. In: Plant-provided food for carnivorous insects: a protective mutualism and its applications (eds. Wäckers FL, van Rijn PCJ & Bruin J). Cambridge University Press Cambridge, UK, pp. 17-74. Wheeler D.E. (1984). Behavior of the ant, Procryptocerus scabriusculus (Hymenoptera: Formicidae), with comparisons to other cephalotines. Psyche, 91, 171-192. Wheeler W.M. & Bailey I.W. (1920). The feeding habits of Pseudomyrmicinae and other ants. T. Am. Philos. Soc., 22, 235-279. Wickham H. (2007). Reshaping data with the reshape package. J. Stat. Soft., 21, 1-20. Wickham H. (2009). ggplot2: elegrant graphics for data analysis. Springer-Verlag, New York. Wickham H. (2011). The split-apply-combine strategy for data analysis. J. Stat. Soft., 40, 1-29. Wilke A., Bischof J., Gerlach W., Glass E., Harrison T., Keegan K.P., Paczian T., Trimble W.L., Bagchi S., Grama A., Chaterji S. & Meyer F. (2016). The MG-RAST metagenomics database and portal in 2015. Nucleic Acids Res., 44, D590-4. Wilson E.O. (1976). A social ethogram of the neotropical arboreal ant Zacryptocerus varians (Fr. Smith). Anim. Behav., 24, 354-363. Wurm Y., Wang J., Riba-Grognuz O., Corona M., Nygaard S., Hunt B.G., Ingram K.K., Falquet L., Nipitwattanaphon M., Gotzek D., Dijkstra M.B., Oettler J., Comtesse F., Shih C.-J., 169
Wu W.-J., Yang C.-C., Thomas J., Beaudoing E., Pradervand S., Flegel V., Cook E.D., Fabbretti R., Stockinger H., Long L., Farmerie W.G., Oakey J., Boomsma J.J., Pamilo P., Yi S.V., Heinze J., Goodisman M.A.D., Farinelli L., Harshman K., Hulo N., Cerutti L., Xenarios I., Shoemaker D. & Keller L. (2011). The genome of the fire ant Solenopsis invicta. Proc. Natl. Acad. Sci. USA, 108, 5679-5684.
170
FIGURES
Figure 1. Different trophallactic acts in Cephalotes rowheri. Left and right panels: oral trophallaxis. Central panel: abdominal trophallaxis. Images extracted from video-recording of laboratory colonies. Color, brightness and contrast were modified to show only gray tones and highlight interactions.
171
Figure 2. Oral trophallaxis excluder. On top: side-view of an abdominal trophallaxis donor (left) with a single drop of glue on the thorax (in gray), glued to a halve of a pipette tip that is also glued to the surface of the nest; Bottom: photograph illustrating experimental design with donors inside oral trophallaxis excluders while receivers are freely moving. Color, brightness and contrast were modified to show only gray tones and highlight interactions.
172
(A)
(B)
Figure 3. Frequency and duration of trophallaxis in three colonies of C. rohweri. (A) Abdominal trophallaxis; (B) Oral Trophallaxis. Different castes are represented by letters and their order represents their task in the interaction: donor (left) and receiver (right); thus “CS” is an interaction where callow is the donor and the major worker (soldier) is the receiver. Castes represented: callow (C), major workers (S), and minor worker (W).
173
(A)
Abdominal Trophallaxis
2.0
1.5 C o l _
1.0 5 1 3
0.5
l 0.0 a u d i 2.0 v i d n i r e p 1.5 s n C o i t o c l _ a 8
r 1.0 1 e 3 t n i f o
o 0.5 i t a r e g a r 0.0 e v a 2.0
1.5 C o l _
1.0 9 1 3
0.5
0.0 CC CS CW SC SS SW WC WS WW Interaction
(B)
Oral Trophallaxis
Figure 4. Average frequency of trophallaxis by caste. (A) Abdominal Trophallaxis; (B) Oral Trophallaxis. Average of frequency of interactions based on the number of individuals in each caste at each day of observation. For abdominal trophallaxis, receivers are the referential for calculating the average number of interactions per pair of interaction. For oral trophallaxis, for each day of observation interactions involving each caste were grouped and their sum (per caste) was used to calculate the average number of interactions. Castes represented: callow (C), major workers (S), and minor worker (W).
174
C1 C6 20 ( 2 = 16.2, df = 1, p-value = 5.699e-05) ( 2 = 2.7778, df = 1, p-value = 0.09558)
18
16 s i x a l
l 14 a h
p 12 o r t
d 10 b A
f
o 8
q e r 6 F 4
2
0 HBCS HB HBCS HB
Figure 5. Frequency of abdominal trophallaxis in groups fed different diets. Two colonies were observed, colony “C1” and colony “C6”. Treatments are represented by “HBCS”, where receivers were fed previously a complete diet containing honey water, roaches and SPAM; “HB” refers to receivers that were fed only honey water.
175
(A)
A
X u
r a a Bacteria n n
t e t h i B m a o Archaea r e b o a c e
a n d c a a Eukaryota a y t r e e r d h e c r t a i a a O z c c c r Virus o e e t e P b h a t a a c h i e b a e y e r Other c r o a l ol e e d b
b 0 h a 0 o o t a . t Unclassified 9 . e h c 9 e O t % e e % R c a
ri a 2 A e
c % c e a a % n
e 5 % e g 1 i 7 l % a lc A
Rho er % dobact ales 9 Rho dosp irlilale e s ea ac B ad B u on a rk m rton h a ella B o om ce e l C ae t d a e % 1 p 5 0% rai ro le ia te s cter o oba b rote a ceae P c lderia et rkho r Bu ia 3%
Bacteria Other ia teobacter Firmicutes 0.6% Alphapro a unclassified (derived from Bacter ia) 0.4% ri te Cyanobacteria 0.2% c 6% a Xa Chloroflexi 0.2% b s nht om e on % s) o l adace Deinococci 0.1% 11 las te a ae (c d ceae teria o Acidobacteria 0.09% ia ac ia r a ob ob ob p hiz ctin icr n Planctomycetia 0.09% R A om a o ruc m m er m o Chlorobiaceae 0.07% V a d es u Spirochaetales 0.04% Rhizobial G e 5% ae s Synergistetes 0.03% itut P P Op se ud Thermotogaceae 0.03% 2 o % m Nitrospiraceae 0.02% 1 on a % d E ac Deferribacteraceae 0.02% n e ae 1 te Fusobacteriales 0.02% B % ro a 2 b a unclassified (derived from Poribacteria) 0.0 2% c % c t e A t e unclassified (derived from Lentisphaerae) 0 .01% r c ria o O % t c i i e Chlamydiales 0.01% 5 d n p O a 1 e o i e t m u t Gemmatimonadaceae 0 .01% t h e t e e y a s r Chrysiogenaceae 0.009% a c c e e e c t a Aquificales 0.006% a a e l l l e Fibrobacteraceae 0.003% e s c u Dictyoglomacea e 0.003% r B Mollicutes 0.002% Elusimicrobia 0.0009%
(B)
e a e c a c li a S
% 3 3
) yta s ph le pto ia tre h m S ig fro lp ed a eriv M (d ed sfii las unc
Streptophyta s
e l
a
c i s 8% s Brassicac a eae r
B s le % a 40 b e a s ea F e biac l hor a Eup s n ale la 3 Po o % S F ab 1 ace % ae
So la na ce ae 1 O 2 th % er
P o a c e a e
Selaginellales 0.4% Marchantiales 0.1% 176
Figure 6. Relative abundance of groups of organisms found in infrabuccal pellets of C. rohweri. (A) Left: Bacteria; Top-right corner: all organisms found in the pellet; (B) Streptophyta.
177
(A)
(B)
Figure 7. Carbohydrate-associated metabolism present in the community of infrabuccal pellets. Blue coloration indicates genes present; white boxes are absent enzymes. (A) Amino sugar and nucleotide sugar metabolism; arrow points to chitin; (B) startch and sucrose metabolism; arrows 178
point to pectin and cellulose. Adapted from KEGG metabolic pathways designed by Kanehisa et al. (2016).
179
Figure 8.Uric acid and urea breakdown by microbiota found in the infrabuccal pellets of C. rohweri. Arrows point to urate (uric acid) and urea. Adapted from KEGG metabolic pathways designed by Kanehisa et al. (2016).
180
TABLE
Table 1. Logistic Regression comparing the number of individuals found with blue dye in their infrabuccal pellets, using diet treatment nested in colony as explanatory variables
Df Deviance Df Resid. Deviance Pr(>Chi) NULL 92 101.757 Treatment 1 12.122 91 89.635 0.0004983 * Treatment:Colony 10 22.533 81 67.102 0.0126086 *
* Statistically significant results
181
SUPPLEMENTAL MATERIAL
Protocol used by the MR DNA facility for Whole Genome Amplification (WGA) and MiSeq sequencing
The library was prepared using Nextera DNA Sample preparation kit (Illumina) following the
manufacturer's user guide. The initial concentration of DNA was evaluated using the Qubit® dsDNA HS Assay Kit (Life Technologies). Because of low DNA concentration for all the samples, whole genome amplification was carried out by using REPLI-g Mini kit (Qiagen) followed by Nextera DNA Sample preparation. The linear amplified DNA concentration was
again evaluated (Table 1) using the Qubit® dsDNA HS Assay Kit (Life Technologies). Samples were then diluted accordingly to achieve the recommended DNA input of 50ng at a concentration of 2.5ng/uL. Subsequently, the samples underwent the simultaneous fragmentation and addition of adapter sequences. These adapters are utilized during a limited-cycle (5 cycles) PCR in which unique index was added to the sample. Following the library preparation, the final concentration of the library (Table 1) was measured using the Qubit® dsDNA HS Assay Kit (Life Technologies), and the average library size was determined using the Agilent 2100 Bioanalyzer (Agilent Technologies). The libraries were then pooled in equimolar ratios of 2nM, and 12pM of the library pool was sequenced paired end for 600 cycles using the MiSeq system (Illumina). Table 1. DNA and final library concentration and average library size.
Sample DNA DNA library Average library concentration concentration size (bp) (ng/uL) (ng/uL)
Sample.2W 39.00* 22.4 874
Sample.3W 15.50* 20.0 627
Sample.4WS 23.25* 23.4 800
Sample.5W 20.75* 25.4 824 182
Sample.6S 10.75* 22.8 825 *- Whole genome amplified DNA
SUPPLEMENTAL FIGURES AND TABLES
The following files will be made available via Figshare upon publication - S2: S2_Behavioral_Observations_and_Metada.xlsx – all information on frequency, castes, time and duration of interactions that were used to characterize abdominal and oral trophallaxis; survey data on colony demographics is also included
- S4 – Data and metadata information regarding colonies used in testing if rectal fluids are transferred via abdominal trophallaxis; also included data on whether nutritional state of receivers is associated with the frequency that they solicit abdominal trophallaxis
183
(A) (B)
Mg
IL
S3 – Coloration of digestive tube and infrabuccal pellet. (A) Midgut (Mg) and Ileum(IL), showing the typical coloration of the hindgut after being fed a blue-dyed honey water solution for over two weeks; (B) coloration of the infrabuccal pellet; visible also are the beads fed to the receiver group, along with non-dyed honey water.
184
185
186
187
S5 – Pathways for all ten amino acids considered essential for insects. Blue boxes represent genes for enzymes that were found in the pellets of Cephalotes rohweri. Adapted from KEGG metabolic pathways designed by Kanehisa et al. (2016). 188
APPENDIX D: PERMISSIONS
Permission to reproduce content
Wed, Jul 27, 2016 at [email protected]
a) To reproduce the Contribution in whole or in part in any printed volume (book or thesis) of which they are the author(s). b) They and any academic institution where they work at the time may reproduce the Contribution for the purpose of course teaching. c) To post a copy of the Contribution as accepted for publication after peer review (in Word or Tex format) on the Authors’ own web site or institutional repository, or the Authors’ funding body’s designated archive, six months after publication of the printed or online edition of the Journal, provided that they also give a hyperlink from the Contribution to the Journals web site. d) To reuse figures or tables created by them and contained in the Contribution in other works created by them. The above use of the term 'Contribution' refers to the author's own version, not the final version as published in the Journal. For open access content published under a Creative Commons license, authors can replace the submitted version with the final published version at publication. Open Access content can be reused, without permission for all non-commercial purposes. Kind regards Melissa Rose Permissions Assistant SpringerNature The Macmillan Building 4 Crinan Street London N1 9XW