Periplaneta Americana

Host-Microbial Symbiosis Within the Digestive Tract of Periplaneta americana.

Dissertation

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy

in the Graduate School of The Ohio State University

Benjamin C. Jahnes

Graduate Program in Microbiology

The Ohio State University

2020

Dissertation Committee

Professor Zakee Sabree, Advisor

Professor Kelly Wrighton

Professor Virginia Rich

Professor Rachelle Adams

Copyrighted by

Benjamin C. Jahnes

2020

Abstract

Cockroaches as a model system allow us to explore a self-contained microbial community of incredible complexity, with the gut alone housing numerous members of the Eukarya, Bacteria, and Archaea, often in close symbiotic association with one another. This diversity provides us with the opportunity to examine a plethora of microbe/microbe and microbe/host interactions, with recent work in this area reviewed here. The ease of rearing the cockroach and its great resilience allows us to look at the cockroach as a blank slate, in the absence of gut microbiota, and set a baseline for growth through the establishment of germ-free insects. Assembling an aseptic isolation habitat of low cost and complexity is shown to aid in the maintenance of germ-free insects and allows for the selective reintroduction of gut bacterial isolates to the cockroach. The germ-free cockroach provides the potential to systematically examine interactions between the cockroach and the diversity of life that can be isolated from within the gut of the insect. In this work the germ-free cockroach is inoculated with gut microbiota through coprophagy and compared to wild-type cockroaches to examine the degree to which this behavior serves to provision the gut with symbiotic taxa, and examine the extent of bacterial stimulation of growth and development in wild-type compared to germ-free insects. A brief inoculation by coprophagy appears to endow cockroaches with a subset of the wild-type microbial community that is sufficient to induce growth phenotypes of

the hindgut that are significantly different from, but intermediate to germ-free and wild- type cockroaches. Histological sectioning along the gut further refines the site of microbial stimulation of host gut development to the posterior midgut and anterior hindgut. Finally, several closely related and strongly host-associated cockroach gut

Bacteroides are examined in relation to common mammalian gut Bacteroides, to examine to what extent insect Bacteroides have retained traits that may be required for host association and gut colonization. These native Bacteroides are then introduced to germ- free cockroaches in a defined community to examine genus-specific dynamics of gut colonization and host-growth promotion. The four reintroduced Bacteroides prove to have variable abilities to colonize a naïve gut environment and don’t positively contribute to cockroach development, in a monogeneric community. This demonstrates that more complex microbial community level interactions among gut microbiota and successional dynamics may contribute to preparing the gut environment for colonization by native gut microbiota, and host adaptation alone doesn’t guarantee microbe/host mutualism.

iii

Dedication

I dedicate this work to my mom and dad who fostered a sense of wonder and encouraged my curiosity in the natural world and my brother and sister for being co-conspirators in exploration and general supporters in life.

Acknowledgments

I wish to acknowledge the numerous undergraduates that served as mentees who kept me company in the lab and aided variously in this work, including Sema Osman, Mady

Herrmann, Jon Foltz, John Thundathil, Sophia Nicholas, and Keshap Poudel. Thank you to George Keeney for maintenance of the OSU Insectary Greenhouse Periplaneta americana colony. Thanks to Marymegan Daly and John Freudenstein for use of histology preparation equipment and lab space. Thanks to Leslie Jackson for use of a microscope and histology prep equipment.

Vita

1999……………………………………………………….West Muskingum High School

2003…………………………………………B.S. Biological Sciences, Cornell University

2004-2006…………………..……………………...U.S. Peace Corps, Niger, West Africa

2006-2013………………………………………………Flint Ridge Vineyard and Winery

2014-Present………...Graduate Teaching/Research Associate, The Ohio State University

Publications

Jahnes, B.C., Sabree, Z.L., 2020. Nutritional symbiosis and ecology of host-gut microbe systems in the Blattodea. Curr. Opin. Insect Sci. 39, 35–41. https://doi.org/10.1016/j.cois.2020.01.001

Vera-Ponce de León, A., Jahnes, B.C., Duan, J., Camuy-Vélez, L.A., Sabree, Z.L., 2020. Cultivable, Host-Specific Bacteroidetes Symbionts Exhibit Diverse Polysaccharolytic Strategies. Appl. Environ. Microbiol. https://doi.org/10.1128/AEM.00091-20

Jahnes, B.C., Herrmann, M., Sabree, Z.L., 2019. Conspecific coprophagy stimulates normal development in a germ-free model invertebrate. PeerJ 7, e6914. https://doi.org/10.7717/peerj.6914

Garrick, R.C., Sabree, Z.L., Jahnes, B.C., Oliver, J.C., 2017. Strong spatial-genetic congruence between a wood-feeding cockroach and its bacterial endosymbiont, across a topographically complex landscape. J. Biogeogr. 44, 1500–1511. https://doi.org/10.1111/jbi.12992

Thirunavukkarasu, N., Jahnes, B., Broadstock, A., Rajulu, M.B.G., Murali, T.S., Gopalan, V., Suryanarayanan, T.S., 2015. Screening marine-derived endophytic fungi for xylan-degrading enzymes. Current Science (Vol. 109). vi

Fields of Study

Major Field: Microbiology

vii

Table of Contents

Abstract ...... ii Dedication ...... iv Acknowledgments...... v Vita ...... vi List of Tables ...... x List of Figures ...... xi General Introduction ...... 1 Chapter 1. Nutritional symbiosis and ecology of host-microbe systems in the Blattodea. 4 ABSTRACT ...... 4 Gut Microbial Diversity ...... 4 Host Insect/Microbe Interactions ...... 7 Microbe/Microbe Interactions ...... 10 Microbial Community Assembly ...... 12 Conclusions and Future Perspectives...... 14 Chapter 2. A Low Cost and Modular System for Rearing Germ-Free and Gnotobiotic American Cockroaches (Periplaneta americana)...... 15 ABSTRACT ...... 15 INTRODUCTION ...... 16 MATERIALS AND METHODS ...... 30 RESULTS ...... 38 DISCUSSION ...... 49 Chapter 3. Conspecific Coprophagy Stimulates Normal Development in a Germ-Free Model Invertebrate...... 55 ABSTRACT ...... 55 INTRODUCTION ...... 56

viii

MATERIALS AND METHODS ...... 59 RESULTS ...... 64 DISCUSSION AND CONCLUSIONS ...... 74 Chapter 4. Microbial Colonization Promotes Epithelial Expansion in Gut Subcompartments of Periplaneta americana...... 79 ABSTRACT ...... 79 INTRODUCTION ...... 80 MATERIALS AND METHODS ...... 82 RESULTS ...... 89 DISCUSSION ...... 95 Chapter 5. Recruitment, persistence, and mutualism in the gut microbial community of Periplaneta americana...... 100 ABSTRACT ...... 100 INTRODUCTION ...... 100 MATERIALS AND METHODS ...... 102 RESULTS ...... 108 DISCUSSION ...... 128 Conclusions and Future Perspectives...... 133 Bibliography ...... 135 Appendix A. Supplementary Information Chapter 2 ...... 155 Appendix B. Supplementary Information Chapter 3 ...... 158 Appendix C. Supplementary information Chapter 5 ...... 165

List of Tables

Table 1 Microaerobic vs. Anaerobic Growth...... 50 Table 2 p-value Table Histological Measurements ...... 95 Table 3 Oxidative Stress Response Genes ...... 119 Table 4 Isolate Specific Primers ...... 155 Table 5 p-value Table Morphological Measures ...... 164

List of Figures

Figure 1 Oothecum Collection...... 30 Figure 2 Ootheca incubation...... 32 Figure 3 Nymph Housing...... 34 Figure 4 Gnotobiotic Inoculation...... 36 Figure 5 Germ-free Insect Validation – PCR...... 40 Figure 6 Germ-free Insect Validation – Microscopy...... 41 Figure 7 E. coli Gut Colonization Detection and Enumeration...... 43 Figure 8 Detection of Isolate PAD521 Colonization...... 44 Figure 9 Detection of Isolate PAL227 and PAB224 Colonization...... 46 Figure 10 Detection of Colonization by Isolate PAB51...... 47 Figure 11 Detection of Colonization by Polyculture of Isolates PAD25 and PAF510 48 Figure 12 Insect Maturation Rate ...... 65 Figure 13 Qualitative Comparisons of Hindgut Morphology in 5th instar Periplaneta americana. 67 Figure 14 Hindgut Microbial Biomass ...... 68 Figure 15 Gut Morphology...... 70 Figure 16 Principal Component Analysis (PCA) of Morphological Characteristics at 5th instar. 73 Figure 17 Blocking ...... 87 Figure 18 Histological Measurements...... 93 Figure 19 Conceptual Model of Cockroach Gut Dynamics ...... 98 Figure 20 PAB Isolate Phylogenetics ...... 108 Figure 21 Capsular Polysaccharide Loci ...... 110 Figure 22 Conserved Protein Glycosylation Locus ...... 114 Figure 23 New Putative Protein Glycosylation Locus ...... 117 Figure 24 Gnotobiotic Colonization by Monogeneric Inoculation ...... 124 Figure 25 Cockroach growth parameters...... 125 Figure 26 Growth of Bacteroides isolates in anaerobic conditions vs microaerophilic conditions. 126 Figure 27 Peroxide susceptibility...... 127 Figure 28 Cage Collected Ootheca Rinse Plating ...... 156 Figure 29 Cockroach Collected Ootheca Rinse Plating ...... 157 Figure 30 Wild-type Time to 5th Instar ...... 158 Figure 31 Body Mass Boxplot...... 159 Figure 32 Germ Free Cockroach Validation ...... 160

Figure 33 Foregut Length Boxplot ...... 161 Figure 34 Gut Mass Boxplot ...... 162 Figure 35 Partition of Variation ...... 163 Figure 36 PCA All instars ...... 163 Figure 37 Bacteroides fragilis Capsular Polysaccharide Loci ...... 166 Figure 38 Bacteroides In-vitro Coculture ...... 167 Figure 39 Bacteroides Post-Inoculation In-vitro Coculture ...... 168

xii

General Introduction

In microbiology, model animals represent a unique environment and exclusive habitat for certain microbes, or may serve as a foundation for the study of microbe/host interactions.

Model animals vary in ease of rearing, their suitability as hosts for specific microbiota, and the translational quality of findings within the system to other systems. The pig represents a valuable animal model whose physiology and microbiota is very similar to humans thus translational quality of research in this organism is high, but rearing pigs as lab animals is relatively difficult and space intensive (Bendixen, Danielsen, Larsen, &

Bendixen, 2010). Mice offer benefits over other mammals, in that they are relatively small yet still host a diverse gut microbiota, and genetic conservation with other animals is high, though significant resources are required to rear them aseptically (Krych, Hansen,

Hansen, van den Berg, & Nielsen, 2013; Yi & Li, 2012). As an invertebrate, Drosophila has served as an easily reared and prolific genetic model organism with substantial genetic conservation with vertebrates, but it hosts few gut microbial taxa (Douglas, 2018;

Erkosar & Leulier, 2014; Pais, Valente, Sporniak, & Teixeira, 2018). Each model organism presents unique advantages and drawbacks.

This dissertation will examine the cockroach as an invertebrate model organism with a complex microbial community that serves as an interesting microbial habitat of incredible diversity, examines the practice of rearing the cockroach under germ-free and gnotobiotic

conditions, looks at phenotypic differences between cockroaches that have varying levels of microbial colonization of the gut, and examines qualities of cockroach derived gut microbial taxa that may enable them to efficiently colonize the insect gut.

Chapter 1 introduces the gut microbial ecology within Blattodea (cockroaches and termites), reviewing recent literature on digestive microbial symbioses of this insect

Order, highlighting how they are an intriguingly diverse microbial habitat unto themselves.

Chapter 2 details a procedure for deriving germ-free and gnotobiotic Periplaneta americana, demonstrating that the organism can be reared aseptically using relatively low-cost materials with high rates of success with the potential to reintroduce native and non-native microbial taxa to the gut.

Chapter 3 outlines a study utilizing germ-free cockroaches to examine parameters of growth in the absence of bacteria compared to cockroaches colonized through coprophagy and wild-type cockroaches.

Chapter 4 follows up on chapter 3 by examining fine scale features of gut morphology across the same treatments through histological analysis.

Chapter 5 takes a look at a subset of bacteria derived from the cockroach gut and examines them in relation to a human gut-derived congeneric in terms of their genomic capacity for host colonization and then examines their colonization efficiency and interactions with the host when introduced as a defined community.

Together these studies set the stage for use of the cockroach P. americana in the systematic study of host/microbial interactions, microbial colonization dynamics, and gut microbial community ecology.

Chapter 1. Nutritional symbiosis and ecology of host-microbe systems in the Blattodea.1

ABSTRACT

Cockroaches and termites (Order: Blattodea) have been the subject of substantial research attention for over a century due, in part, to a subset of them having a strong propensity to cohabitate with humans and their structures. Recent research has led to numerous insights into their behavior, physiology, and ecology, as well as their ability to harbor taxonomically diverse microbial communities within their digestive systems, which include taxa that contribute to host growth and development. Further, recent investigations into the physiological and behavioral adaptations that enable recalcitrant polysaccharide digestion and the maintenance of microbial symbionts in cockroaches and termites suggests that symbionts contribute significantly to nutrient provisioning and processing.

Gut Microbial Diversity

Cockroaches and termites exhibit a wide range of dietary strategies that include omnivory, xylophagy, detritovory, humivory and coprophagy, which are influenced by the diverse, largely uncharacterized microbes inhabiting their digestive tracts. Recent experimental work suggest that host taxonomy (Dietrich, Köhler, & Brune, 2014; Sabree & Moran, 2014;

LiJuan Su et al., 2016), gut compartmentalization (Aram Mikaelyan, Meuser, & Brune,

2017; Rossmassler et al., 2015) and diet (Aram Mikaelyan, Dietrich, et al., 2015; Lijuan

Su et al., 2017) significantly impact gut bacterial community composition. Blattodea

1 This chapter was reproduced verbatim from Jahnes and Sabree 2020 Curr. Opin. Insect Sci. 4

species host stable core microbiota whose relative abundances respond to different diets and they are consistently dominated by members of the Firmicutes, Bacteroidetes and

Proteobacteria, with Spirochetes being distinctively abundant in some termites

(Benjamino, Lincoln, Srivastava, & Graf, 2018; Schauer, Thompson, & Brune, 2014;

Lijuan Su et al., 2017; Tinker & Ottesen, 2016). Micro-eukaryotes (i.e. protists) are more diverse and abundant in ‘lower termites’ than in cockroaches (Gijzen & Barugahare, 1992;

Martínez-Girón, Martínez-Torre, & van Woerden, 2017), yet they are prevalent in wood- feeding cockroaches (Vďačný et al., 2018). Wood-feeding ‘lower termites’ (e.g.

Mastotermitidae, Termopsidae, and Rhinotermitidae) maintain microbial communities with increased protozoan diversity relative to cockroaches and ‘higher termites’, and they are enriched with Spirochetes and Bacteroidetes that live within or on the surfaces of endemic protists (Berchtold et al., 1999; Dietrich et al., 2014; Noda et al., 2005; LiJuan Su et al., 2016). Bacterial diversity increases in ‘higher termites’ (e.g. Macrotermitinae,

Termitinae, Apicotermitinae and Nasutitermitinae) that feed on diverse diets and are characterized by a notable abundance of Spirochetes and other poorly defined taxa

(Dietrich et al., 2014; Aram Mikaelyan, Dietrich, et al., 2015; LiJuan Su et al., 2016). Diet strongly correlates with gut microbiota composition within the ‘higher’ termites, where fungus-cultivating Macrotermitinae resemble that of omnivorous cockroaches, dominated by Firmicutes, Bacteroides and Proteobacteria, and relatively few Spirochetes. The inverse is observed in wood-feeding Nausitermitinae and Termitinae gut microbiota (Dietrich et al., 2014; Aram Mikaelyan, Dietrich, et al., 2015). Interestingly, gut microbiota composition in litter-, soil- and humus-feeding Nausitermitinae and Termitinae are more

similar to omnivorous cockroaches, except for the increased abundance of Spirochetes, which further illustrates the relationship between diet and gut microbial diversity (Dietrich et al., 2014; Aram Mikaelyan, Dietrich, et al., 2015).

Comprehensive surveys of viruses, Amoebozoa, fungi, protozoans, and helminths within Blattodea gut microbial communities using nucleic acid sequencing approaches are sparse. Metagenomic profiling of the Coptotermes formosanus gut bacteriophage community identified ~566 phage phylotypes/colony with 960 unique phylotypes across three colonies (Tikhe & Husseneder, 2018), which hints at considerable bacteriophage diversity within C. formosanus and perhaps the Blattodea in general. A survey of

Entamoeba from Periplaneta americana, Blaptica dubia, and Gromphadorhina oblongonota cockroaches identified 134 unique rDNA sequences from nine new

Entamoeba clades, which greatly expands the known diversity of cockroach-dwelling

Entamoeba (Kawano et al., 2017), yet little is known about their biology and ecological roles. 18S rDNA metagenomic profiling of Blattella germanica hindguts detected a diverse array of fungi, nematodes, Candidia, flagellates and ciliates across both lab-reared and wild-caught individuals (Kakumanu, Maritz, Carlton, & Schal, 2018). While fungal and nematode sequences were ubiquitous, flagellates and ciliates were restricted to insects from a few locations and were low to absent in lab-reared insects (Kakumanu et al., 2018).

Oxymonad flagellates are abundant in the gut microbiota of wood-feeding cockroaches and termites (Hampl, 2017; Nalepa C.A., Bignell D.E., 2001) and several new species of

Monocercomonoides and Blattamonas were cultivated from several omnivorous cockroaches (Treitli et al., 2018). Thelastomatid nematodes (Nematoda: Oxyurida:

Thelastomatidae) are common inhabitants of the cockroach hindgut where they graze on gut microbiota, and their feeding impacts both host growth and development and hindgut bacterial community diversity (Mccallister, 1988; Vicente, Ozawa, & Hasegawa, 2016). In contrast, few gut-dwelling nematodes have been identified in termites, with isolated reports of Thelastomatid nematodes (Kanzaki et al., 2012) and the species description of

Stomachorhabditis fastidiosa (Massey, 1971).

Host Insect/Microbe Interactions

The expanded hindgut paunch of Blattodea (Engel & Moran, 2013; Matthew D. Kane,

1997) and specific subcompartmentalization of termite hindguts (A Brune, Emerson, &

Breznak, 1995) reflect dramatic adaptations to their abundant gut microbiota. Distinct bacterial communities have been demonstrated within subcompartments of the termite hindgut, with greater similarity between homologous subcompartments of different taxa of soil and humus feeding higher termites than among different subcompartments of an individual taxon (Aram Mikaelyan et al., 2017). The recent availability of several Blattodea genomes and transcriptomes reveal additional possible adaptations to harboring diverse gut microbiota. In particular, gene families associated with the innate immune system, including PGRP and GNBP pattern recognition molecules within IMD and Toll pathways

(Harrison et al., 2018; Terrapon et al., 2014) and antimicrobial peptide genes, have expanded through multiple duplication events (I.-W. Kim et al., 2016; Li et al., 2018).

These data suggest that they may employ a high-resolution taxon recognition system to manage their diverse gut microbiota.

Early studies eliminating or suppressing gut bacteria using germ-free methods and antibiotics demonstrated positive effects of bacterial colonization on cockroach and termite growth and development (Bracke, Cruden, & Markovetz, 1978b; Eutick, Veivers, O’Brien,

& Slaytor, 1978; Jahnes, Herrmann, & Sabree, 2019; Rosengaus, Zecher, Schultheis,

Brucker, & Bordenstein, 2011). Furthermore, drastic reduction of fat body endosymbionts

(Blattabacterium spp.) of cockroaches using antibiotics led to high mortality, slow growth, and severely reduced reproductivity (Brooks & Richards, 1955). Depletion of gut bacteria and Blattabacterium spp. using antibiotics decreased exogenous glucose-U-14C or

35 Na2 SO4 incorporation into insect-extracted amino acids, (Block & Henry, 1961; Henry,

1962) and increased urate accumulation in fat bodies (Malke & Schwartz, 1966), suggesting a microbial participation in amino acid provisioning and nitrogen cycling.

Recent 13C stable isotope analysis in Periplaneta americana and Reticulitermes flavipes supported this by demonstrating that insect-assimilated essential amino acids were likely of bacterial origin (P. A. Ayayee, Larsen, & Sabree, 2016; P. Ayayee, Jones, & Sabree,

2015). Additionally, antibiotic suppression of gut microbiota decreased the standard metabolic rate in P. americana, suggesting that gut microbiota contribute significantly to energy expenditure of the cockroach (P. A. Ayayee, Ondrejech, Keeney, & Muñoz-Garcia,

2018).

Omnivorous cockroach gut bacteria can degrade dietary polysaccharides like cellulose

(Cruden & Markovetz, 1979; Gijzen, van der Drift, Barugahare, & Op den Camp, 1994), and cellulolytic symbionts are important for termite and wood-feeding cockroach diet assimilation. While wood-feeding cockroaches and lower termites rely upon cellulolytic

protists (and their bacterial mutualists) for lignocellulose feeding (Tokuda et al., 2014), polysaccharide-degrading bacteria are important cellulolytic components of the higher termite gut microbiota (Liu et al., 2019). Beyond contributing to dietary digestion, gut microbiota can also degrade and detoxify xenobiotics. Antibiotic depletion of gut microbiota in insecticide-resistant Blattella germanica cockroaches increased indoxacarb insecticide sensitivity and partial resistance was conferred upon indoxacarb-susceptible strains following coprophagy of feces, and the microbes within, from insecticide-resistant

B. germanica (Pietri, Tiffany, & Liang, 2018). A complementary study reported that several endosulfan-degrading bacteria were isolated from Blatta orientalis digestive tracts, yet host protection against this insecticide was not examined (Ozdal & Ozdal, 2016).

Finally, antibiotic-treated B. germanica that were infected with the entomopathogenic fungus Metarhizium anisopliae exhibited 50% greater mortality than untreated, infected insects, which suggests that gut bacteria reduced entomopathogen susceptibility (F. Zhang et al., 2018).

In addition to providing host protection, gut bacteria and their products can impact host behavior. Volatile carboxylic acids (VCA) present in frass (feces) are among fecal aggregation agents used by B. germanica to attract nestmates, facilitating several beneficial outcomes (i.e. predator avoidance, mate location, access to nutrient-rich feces). Choice- based assays revealed that frass from wild-type B. germanica was more attractive than frass from antibiotic-treated insects to 1st-instar nymphs (Wada-Katsumata et al., 2015). Several

VCAs were absent in frass from antibiotic-treated insects and inoculation of frass with

aerobic bacteria cultivated from untreated insects rescued the aggregation response, strongly suggesting bacterial origins of aggregation agents.

Microbe/Microbe Interactions

Diverse interspecies symbioses exist in Blattodea digestive tracts that include protist- bacterial endo- and ecto-symbiosis, nematode-bacterial ectosymbioses, and nematode and protist predatory behaviors that too can shape the gut microbial community.

Protist-Endosymbiont Interactions. Methanogenic archaeabacteria, delta-protobacterial sulfate-reducers and spirochete acetogens are common endosymbionts of cellulolytic ciliates and flagellates native to cockroach and termite guts (Kuwahara, Yuki, Izawa,

Ohkuma, & Hongoh, 2017; Lind et al., 2018; Ohkuma et al., 2015). Recent genomic analyses have confirmed the potential for these organisms to act as H2 sinks within their hosts and hint at the importance of syntrophy in promoting endosymbiont acquisition

(Kuwahara et al., 2017; Lind et al., 2018; Ohkuma et al., 2015). Trichonympha spp. protists are abundant in several termite lineages and they harbor bacterial endosymbionts that belong to several distinct bacterial phyla. Members of the Elusimicrobia phylum, namely the ‘Endomicrobia’ spp., reside within Trichonympha spp. and have been detected in the gut lumens of termites, cockroaches and other animals (Andreas Brune, 2012). Genome annotations of five Candidatus ‘Endomicrobium trichonymphae’ phylotype Rs-D17 genomovars revealed highly similar genome content and structure, including amino acid and vitamin provisioning capabilities and intact CRISPR/Cas systems, suggesting a persistent defense against foreign DNA (Hongoh, Sharma, Prakash, Noda, Taylor, et al.,

2008; Izawa et al., 2016). Given the nitrogen demands of amino acid provisioning, it is

notable that these endosymbionts lacked diazotrophic genes observed in the cultivated

Endomicrobium proavitum, which also inhabits Reticulitermes hindguts, but do not always physically associate with Trichonympha protists(H. Zheng, Dietrich, & Brune, 2017).

Extensive genome sequence sampling of the Elusimicrobia is underway, with >130 projects (as of October 2019) registered with GenBank, which can reveal how members of this taxon have evolved across diverse hosts. The trichonymphid endosymbiont

Candidatus ‘Ancillula trichonymphae’ (Actinobacteria:Bifidobacteriales) lacks nitrogen- fixing genes yet can potentially provision amino acids and vitamins using ammonia assimilation and completely fermentative energy metabolic pathways (Strassert,

Mikaelyan, Woyke, & Brune, 2016). The Bacteroidetes trichonymphid endosymbiont

Candidatus ‘Azobacteroides pseudotrichonymphae’ resides in Rhinotermitidae and has nitrogen-fixation genes that likely support the mutualism through metabolite provisioning

(Hongoh, Sharma, Prakash, Noda, Toh, et al., 2008). The detection of a circularly permuted bacteriophage genome encoding a glutamine-tRNA absent in the bacterial chromosome in the related Can. “A. pseudotrichonymphae” phylotype ProJPt-1 adds an intriguing additional layer to symbioses in these organisms (Pramono et al., 2017).

Ectosymbionts of Gut Microbes and Microbial Grazing. Protist ectosymbionts are abundant in termite guts, with Spirochetes and ‘Synergistes’ species associated with and propelling Mixotricha paradoxa (Cleveland & Grimstone, 1964) and Caduceia versatilis

(Hongoh et al., 2007), respectively, through coordinated movements. TG2

‘Margulisbacteria’ attach to Treponema spirochetes who are themselves attached to termite flagellates, exhibiting a multilevel symbiosis (Utami et al., 2019). Hydrogenesis via

cellulosic fermentation by ‘Margulisbacteria’ is thought to benefit their acetogenic spirochete partner. The Bacteroidales ectosymbiont of Barbulanympha protists within the cockroach C. punctulatus can fix 15N, which the protist also assimilates (Tai et al., 2016), while, in contrast, the Candidatus ‘Symbiothrix dinenymphae’ (order: Bacteroidales) ectosymbiont of Dinenympha spp. protists from the lower termite Reticulitermes speratus lacks nitrogen fixation genes, yet it encodes lignocellulose degradation enzymes (Yuki et al., 2015). Bacterial ectosymbionts of nematodes belonging to the Bacteroidales termite cluster V, Rikennellaceae and Ruminococcaceae groups have been observed in two thelastomatid species residing in Panesthia angustipennis woodroach digestive tracts

(Murakami, Onouchi, Igai, Ohkuma, & Hongoh, 2019). Microbe grazing by Leidynema appendiculatum nematodes within P. americana and P. fuliginosa hindguts altered gut community composition, increasing overall taxonomic diversity and enriching the community with Proteobacteria (Vicente et al., 2016). Finally, sympatric bacterivorous ciliates in Panesthia cockroaches exhibited distinct grazing preferences amongst available bacterial species, suggesting long-term ciliate-bacterial food web associations that likely impact gut community composition (Vďačný et al., 2018).

Microbial Community Assembly

Coprophagy, trophallaxis, exuvium consumption, cannibalism, and necrophagy are common behaviors observed across Blattodea and hypothesized to aid in the transfer and retention of gut symbionts (Arthur G Appel, Sims, & Eva, 2008; Kopanic, Holbrook,

Sevala, & Schal, 2001; Mira, 2000; Nalepa C.A., Bignell D.E., 2001), including protozoa

(Hoyte, 1961b), nematodes (Ozawa et al., 2014) and methanogens (Gijzen & Barugahare,

1992). Recent 16S rRNA gene sequencing studies have demonstrated how gut bacterial communities are acquired (Rosas et al., 2018), change over time (Purificación Carrasco et al., 2014) and across different gut environments (Aram Mikaelyan et al., 2017;

Rossmassler et al., 2015; Schauer, Thompson, & Brune, 2012). Gut and frass microbial diversity in B. germanica significantly overlap, but 13-20% of gut taxa were absent from frass, suggesting that coprophagy is not the only means for gut microbial community acquisition (Kakumanu et al., 2018). Rifampicin treatment of B. germanica gut communities over two generations depleted gut community diversity to nearly 10% of untreated cockroaches, and coprophagy of frass from untreated insects resulted in the gut community rebounding to ~50% of the untreated species diversity (Rosas et al., 2018).

Single exposure conventionalization of germ-free P. americana cockroaches resulted in intermediate growth and morphological phenotypes between germ-free and wild-type cockroaches, suggesting that multiple exposures to frass and its associated microbiota are necessary for normal growth and development (Jahnes et al., 2019). Inoculations of germ- free Shelfordella lateralis cockroaches with gut microbiota from termite and mouse guts resulted in colonization of bacteria congeneric to those of a S. lateralis-native gut bacterial community, suggesting that the cockroach gut could select for specific taxa (A Mikaelyan,

Thompson, Hofer, & Brune, 2016a). Work with germ-free cockroaches has also demonstrated how culturable bacteria that are native to the gut can impact gut oxygen levels, which in turn can affect gut community membership that includes taxa that are oxygen-sensitive (i.e. Bacteroidetes) (Tegtmeier, Thompson, Schauer, & Brune, 2016).

Conclusions and Future Perspectives

Our expanding recognition of the diversity of life found within Blattodea digestive tracts will provide the basis for extended study of more specific interactions within this environment. Gut microbes participate in polysaccharide digestion, xenobiotic degradation, and amino acid and vitamin provisioning, and further detailing of the functional contributions of individual taxa to symbiotic digestion warrants attention.

Likewise, newly discovered protist-bacterial symbioses provide opportunities to dissect interspecies interactions. Finally, the Blattodea, especially cockroaches, are amenable to cultivation under both conventional and germ-free conditions, which highlights them as an excellent platform for future efforts to identify the mechanisms underlying host-microbe and microbe-microbe interactions.

Chapter 2. A Low Cost and Modular System for Rearing Germ-Free and Gnotobiotic American Cockroaches (Periplaneta americana).

ABSTRACT

Detailing the role microbes play in the development health and evolution of their hosts, and vice versa, relies upon the examination of hosts that are amenable to the partial or complete removal of their microbial symbionts. The American cockroach (Periplaneta americana) typically harbors a complex gut microbial community and it can easily be reared under germ-free conditions, making it ideal for studying complex gut microbial communities and host-microbe interactions. Effective, newly developed procedures and rearing chambers constructed from readily-available components facilitate the derivation and maintenance of germ-free, gnotobiotic and conventionalized P. americana.

Validation of germ-free insects and strategies to quantify and characterize membership in defined gut microbial communities are described. Both non-native aerobes and native anaerobic cockroach gut isolates can be consistently introduced as monocultures and a subset of these could colonize the gut of early-instar cockroaches. Gnotobiotic monocultures monitored over time demonstrate that some native cockroach gut isolates effectively and persistently colonize the cockroach gut in monoculture after a brief inoculation, while other native gut isolates fail to persist in the cockroach gut in isolation.

Assembly of a simple bipartite microbial community suggests that bacterial mutualisms or commensalism may be required for some isolates to persistently colonize the cockroach gut. This work demonstrates that the P. americana germ-free and gnotobiotic

model system can be used to examine colonization dynamics of gut microbiota, symbiotic interactions, and gut niche, with numerous other possible applications. The relatively low-cost associated with the P. americana model system broadens its accessibility to researchers from diverse disciplines.

INTRODUCTION

Diverse model systems have been used to study bacterial interactions at the gut interface. Germ-free (digestive tract free of microorganisms) and gnotobiotic (digestive tract populated with a defined community of microbial taxa) systems provide particular value in teasing out microbial contributions to health and development, and improve our understanding of how specific bacterial taxa interact with the host (Fiebiger, Bereswill, &

Heimesaat, 2016; Gordon & Pesti, 1971). Vertebrate model systems include pigs, mice, and fish, among others, and such models provide close evolutionary proximity, thus high conservation of genetic pathways with the human system, with respective divergence times of 96 million years ago (MYA), 90 MYA, and 435 MYA (Kamra, 2005; Kostic,

Howitt, & Garrett, 2013; Kumar, Stecher, Suleski, & Hedges, 2017). However, vertebrate systems often demand specialized equipment, trained technicians and sufficient accommodations to house all the above therein (Yi & Li, 2012). Furthermore, research with vertebrates often requires adherence to national mandated policies (e.g. US National

Institutes of Health Public Health Service (PHS) Policy on Humane Care and Use of

Laboratory Animals) that are enforced by institutional animal care and use committees.

Acquiring the permits to proceed with such work can take several months and significant

upfront investments to demonstrate adequate conditions for working with vertebrates, all of which can limit investigator access to these animals for experimental research.

Invertebrates can be invaluable alternatives to vertebrates (Dionne & Schneider,

2008; Feany & Bender, 2000; Gilbert, 2008; Glavis-Bloom, Muhammed, & Mylonakis,

2012; Mackay & Anholt, 2006; O’Kane, 2003; Ugur, Chen, & Bellen, 2016; Wilson-

Sanders, 2011; Wolf et al., 2006) or subjects of biological research in their own right, and some of them have become popular for the study of the host-gut microbiome interactions

(Apidianakis & Rahme, 2011; Bergman, Seyedoleslami Esfahani, & Engström, 2017;

Erkosar & Leulier, 2014; S.-H. Kim & Lee, 2014; A. C. N. Wong, Vanhove, & Watnick,

2016). Some advantages of invertebrates is their relatively rapid and prolific reproductivity, small specimen size, minimal space requirements, simplified housing and absence of policies regulating research with invertebrates (Bier & Mcginnis, 2004;

Giacomotto & Ségalat, 2010; Glavis-Bloom et al., 2012). Taken together, these advantages can dramatically lower barrier-to-entry costs for the scientific community.

Although invertebrates are more evolutionarily distant to humans than vertebrate models, diverging 797 MYA, substantial genetic homology exists (O’Kane, 2003; Reiter, Potocki,

Chien, Gribskov, & Bier, 2001) allowing for generalization of findings from model systems as far removed as Drosophila melanogaster and Caenorhabditis elegans.

Additionally, invertebrate systems are valuable when more general questions of gut microbial ecology are being considered.

The Drosophila model system has provided substantial value in understanding host-microbe interactions within the insect midgut. Among insects, Drosophila is

relatively derived, the genus having diverged from closely related flies approximately 65

MYA (Kumar et al., 2017). Drosophila is one of the most intensively studied insects and a wide array of genetic tools are available that allow easy manipulation and generation of mutants and gene reporter strains (del Valle Rodríguez, Didiano, & Desplan, 2012). The genetic tractability of the model has allowed researchers to dissect molecular interactions between microbes and the insect. Germ-free lines of Drosophila can be generated and reared under sterile conditions (Koyle et al., 2016), and these animals have provided a platform for highlighting how commensals and pathogens differentially elicit distinct

IMD and DUOX immune pathway responses (Guo, Karpac, Tran, & Jasper, 2014; Ha et al., 2009; J. H. Ryu et al., 2008) and impact epithelial cell homeostasis through EGFR and JAK/STAT cell proliferative pathways (Broderick, Nichole A., Buchon, Nicolas,

Lemaitre, 2014; Buchon, Broderick, Chakrabarti, & Lemaitre, 2009). Furthermore, examining the influence of bacteria and yeasts on nutritional provisioning under the influence of low quality diets has allowed researchers to demonstrate the importance of microbiota in nutrient provisioning (Yamada, Deshpande, Bruce, Mak, & Ja, 2015), elicitation of the TOR pathway and regulation of insulin signaling (Shin et al., 2011) and cellular metabolism (Storelli et al., 2011). Experimental evolution in Drosophila has demonstrated both host-dependent and host-independent factors associated with microbial dependent host-growth promotion. Drosophila reared on nutrient poor medium over 170 generations showed decreasing responses to bacteria when experimentally evolved Drosophila were subsequently reared under germ-free and gnotobiotic conditions (Erkosar, Kolly, Meer, & Kawecki, 2017). In contrast, experimental evolution

in gnotobiotic Drosophila demonstrated that host-growth promoting mutations within

Lactobacillus plantarum were due to bacterial adaptation to diet, not adaptation to host

(Martino et al., 2018).

The simplicity of the Drosophila gut microbiota and digestive system both helps and hinders research on host microbial dynamics. Fewer total interactions between individual microbiota and host, and among microbes within the gut, limit the overall complexity of the system and the number of variables that must be parsed in order to understand the system. Wild D. melanogaster have been shown to host around 30 microbial taxa (Corby-Harris et al., 2007; Cox & Gilmore, 2007), while lab strains more typically harbor five taxa (Broderick & Lemaitre, 2012; C. N. A. Wong, Ng, & Douglas,

2011). In lab strains of Drosophila, Firmicutes and Proteobacteria constitute the dominant phyla in the fruit fly gut, most commonly occupied by three genera,

Lactobacillus, Enterobacteria and Acetobacter, while members from groups typically found in both vertebrates and invertebrates, such as Bacteroidetes and Archaea, are largely absent (Broderick & Lemaitre, 2012; Cox & Gilmore, 2007; C. N. A. Wong et al.,

2011). An additional factor limiting the Drosophila system is that its gut microbiota appear to be largely transient and dominated by those associated with host diet (Blum,

Fischer, Miles, & Handelsman, 2013; Broderick & Lemaitre, 2012; Erkosar & Leulier,

2014). Further, Drosophila gut bacteria rarely colonize the gut in a persistent fashion, as do microbes in many vertebrate systems, demonstrated when wild-type adult fruit flies were transferred repeatedly to aseptic food, and gut bacteria dropped to undetectable levels within several days (Blum et al., 2013). Recent work has shown that a minority of

bacterial taxa (2 of 25) recovered from more diverse microbial communities of wild

Drosophila display persistent colonization of the gut (Ma & Leulier, 2018; Pais et al.,

2018), however more work is required to determine whether persistence is an exception or a trend among Drosophila gut microbiota. The Drosophila diet likely contributes to the taxonomic simplicity of its gut microbiota as simple sugars and polysaccharides comprise the bulk of the nutritional profile of its frugivorous diet (Atkinson & Shorrocks,

1977; Marlett & Vollendorf, 1994; Simmonds & Preedy, 2016), and supports bacteria of relatively narrow metabolism. As the dietary nutrients are easily assimilable, it is possible that their relative simplicity may impact the evolution of the digestive tract towards an architecture dominated by the midgut, where host digestive enzyme production and nutrient uptake is focused, and as in the majority of insects, with a minimal hindgut that serves a role in water and ion recovery (Terra, 1990). Reflecting this trend, insects that feed on similarly uncomplex substrates such as plant sap and blood have elongated midguts and relatively short hindguts with bacterial colonization often localized to the midgut, where few taxonomically restricted and highly specialized bacterial taxa can be found (Engel & Moran, 2013). In contrast, several insects from divergent orders including crickets, scarab beetles, crane flies and cockroaches, feeding on diets of greater complexity and recalcitrance, have elaborated hindguts that harbor a diverse array of bacteria with broad metabolic capabilities and provide anaerobic niches for degradation and fermentation of complex polysaccharides (Breznak, 1982; Engel & Moran, 2013;

Matthew D. Kane, 1997; Klug & Kotarski, 1980; Terra, 1990; W. Zheng, Zhao, & Zhang,

2012). Such invertebrate models with increased dietary, gastrointestinal and microbial complexity may better approximate the digestive environments of higher animals.

The American cockroach, Periplaneta americana is an omnivorous peridomestic cockroach of West African origin, that has attained a global distribution through dissemination by the shipping industry (Bell & Adiyodi, 1982; von Beeren, Stoeckle,

Xia, Burke, & Kronauer, 2015). Fossil evidence suggests that the insect order

Dictyoptera, within which cockroaches fall, emerged among early insects around 375

MYA, with cockroach-like fossils appearing around 315 MYA (Legendre et al., 2015), while molecular phylogenetic evidence suggests that Periplaneta diverged from other genera of modern cockroaches approximately 122 MYA (Kumar et al., 2017). P. americana is a well-characterized model organism that has been used in neurophysiology, neuroendocrinology and locomotion research (Aguilar et al., 2003;

Ayali et al., 2015; Huber, Masler, & Rao, 1990; Stankiewicz, Dąbrowski, & de Lima,

2012). Combined with its low-cost rearing and availability of its genome (Li et al., 2018), it will be able to substantially contribute to understanding genetics and evolutionary history. The cockroach has drawn increasing attention as a model for the study of complex gut microbial communities (Bertino-Grimaldi et al., 2013; A Mikaelyan,

Thompson, Hofer, & Brune, 2016b; Aram Mikaelyan, Köhler, et al., 2015; Sabree &

Moran, 2014; Schauer et al., 2014, 2012; Tinker & Ottesen, 2016; J. Zhang, Zhang, Li,

Liu, & Liu, 2016). Evidence of a diverse microbial community within the cockroach gut emerged from cultivation based studies, biochemical studies and microscopy, which identified bacteria of diverse morphology and metabolism (Bignell, 1977b; Bracke,

Cruden, & Markovetz, 1979; Cruden, D.L. Markovetz, 1987; Cruden & Markovetz,

1984; Hoyte, 1961a). Indications that the gut microbial community is important for cockroach development followed the suppression of the gut microbiota using antibiotics, which increased the developmental time of cockroach nymphs (Ludek Zurek & Keddie,

1996) and decreased the standard metabolic rate of late instar nymphs (P. A. Ayayee et al., 2018), but did not affect adult cockroaches (Bracke, Cruden, & Markovetz, 1978a).

Taxonomic characterization of the community membership lagged until the advent of high throughput sequencing methods. Recently, DNA barcoding has allowed a more thorough understanding of the cockroach microbial community composition, the complexity of the cockroach gut microbial community and parallels with the more complex communities of higher vertebrates (Bertino-Grimaldi et al., 2013; Sabree &

Moran, 2014; Tinker & Ottesen, 2016; J. Zhang et al., 2016).

The cockroach diet and evolutionary history appears to drive a gut architecture and microbial community composition reminiscent of omnivorous vertebrates.

Cockroach gut architecture is characterized by host/microbial digestive partitioning, where host enzymatic digestion and nutrient uptake are localized in the crop and midgut and bacterial digestion and water uptake is concentrated in the hindgut (Engel & Moran,

2013; Matthew D. Kane, 1997; Nalepa C.A., Bignell D.E., 2001), comparable to the partitioning of host and microbial digestion between the human small intestine and colon

(Borgstrom, Dahlqvist, Lundh, & SJOVALL, 1957; Guerra et al., 2012; Hayashi,

Takahashi, Nishi, Sakamoto, & Benno, 2005; Hume, 1997; Johnson, 2019; Savage,

1977). The omnivorous diet of the cockroach ensures a diversity of substrates entering

the gut with relatively labile and host digestible components such as simple sugars, starch, lipids, and proteins along with refractory components like cellulose, hemicellulose, pectin, and chitin. Host-produced digestive enzymes act upon ingested diet primarily in the crop and midgut, where nutrients are absorbed (Bell & Adiyodi,

1982; Oyebanji, Soyelu, Bamigbade, & Okonji, 2014; Swingle, 1925; Tamaki et al.,

2014). Undigested material exits the midgut into the hindgut where microbial populations are elevated compared to other compartments (Bignell, 1977b; Bracke et al., 1979;

Schauer et al., 2014, 2012). The cockroach hindgut is elaborated in a fashion similar to that of termites and other insects that feed on recalcitrant vegetable matter, presumably to allow for extended bacterial decomposition of structural polysaccharides (Engel &

Moran, 2013; Matthew D. Kane, 1997), with parallels to the colon of vertebrates (Hume,

1997). Likewise, digestate retention time is prolonged in the hindgut relative to other compartments, providing extended time for bacterial degradation of recalcitrant biomass

(Day & Powning, 1949; Snipes & Tauber, 1937). The importance of hindgut fermentation is evidenced by the detection of host assimilable short chain fatty acids

(SCFA) derived from recalcitrant polysaccharides within this compartment and the incorporation of this carbon into cockroach tissues (Bignell, 1977a; Bracke & Markovetz,

1980; Hogan, Slaytor, & O’Brien, 1985). Correspondingly, antibiotic suppression of anaerobic gut microflora significantly reduced production of SCFA and negatively impacted cockroach development (Bracke et al., 1978a; Ludek Zurek & Keddie, 1996).

With a hindgut adapted to foster microbial digestion, study of the cockroach and its gut microbiota may reveal new insight into host microbial digestive mutualisms.

The cockroach gut microbial community is relatively complex considering bacterial diversity alone, but more so considering general biotic diversity, and has been shown to include viruses, archaea, fungi, amoebozoa, protozoa, and helminths (Bell &

Adiyodi, 1982; Cruden, Gorrell, & Markovetz, 1979; Hoyte, 1961a; Kawano et al., 2017;

S. L. Vicente, Ozawa, & Hasegawa, 2016). Only the gut bacterial community will receive further attention in this work, but the fact that other diverse lifeforms reside in the cockroach gut lends the model to examination of complex interactions across several domains of life. Cultivation independent studies have identified between 200 and 2500 bacterial phylotypes at the 97% nucleotide identity threshold within the cockroach gut, with membership dominated by Bacteroidetes, Firmicutes, Proteobacteria and

Actinobacteria, which is similar to the species diversity observed in omnivorous vertebrates (Bertino-Grimaldi et al., 2013; A Mikaelyan et al., 2016b; Sabree & Moran,

2014; Schauer et al., 2014, 2012; Tinker & Ottesen, 2016; J. Zhang et al., 2016). These complex gut communities are stable after a period of microbial recruitment throughout early instars and are persistent into adulthood (Purificación Carrasco et al., 2014).

Aerobic and anaerobic cultivation methods have yielded numerous microbial isolates from the cockroach gut including important human pathogens and novel indigenous taxa, though it is unclear what proportion of the total community is cultivable (Burgess,

McDermott, & Whiting, 1973; Dugas, Zurek, Paster, Keddie, & Leadbetter, 2001;

Tegtmeier et al., 2016; Vera-Ponce de León, Jahnes, Duan, Camuy-Vélez, & Sabree,

2020). Cultivation dependent enumeration identified one to two orders of magnitude fewer CFU than direct microscopic counts (Cazemier, Hackstein, Op Den Camp,

Rosenberg, & Van Der Drift, 1997; Cruden & Markovetz, 1984), while isolation efforts have yielded as many as 160 anaerobic isolates from a single cockroach gut (Asao and

Sabree, unpublished data). The wide phylogenetic diversity of microorganisms in the cockroach gut underlies broad metabolic diversity. Lactic acid bacteria dominate the relative high oxygen conditions of the foregut compartment, fermenting labile sugars to lactate (M. D. Kane & Breznak, 1991; Schauer et al., 2012). The oxygen depleted midgut and anterior hindgut provide favorable conditions for bacteria capable of polysaccharide hydrolysis and fermentation which leads to the production of SCFA and gaseous byproducts (Bignell, 1977a; Cruden & Markovetz, 1979; Ludek Zurek & Keddie, 1996).

Correspondingly, fermentative polysaccharolytic bacteria, such as Bacteroides,

Dysgonomonas, Ruminococcaceae, and Fibrobacter, are enriched in the hindgut compartment (Bertino-Grimaldi et al., 2013; Sabree & Moran, 2014; Schauer et al., 2014;

Tinker & Ottesen, 2016). The same low oxygen conditions that promote fermentation allow dissimilatory sulfate reducing bacteria (Schauer et al., 2012), homoacetogens

(Breznak & Switzer, 1986; Ottesen & Leadbetter, 2010), and methanogens (W. W.

Sprenger, Van Belzen, Rosenberg, Hackstein, & Keltjens, 2000) to consume abundantly evolved metabolic H2 gas coupled to the reduction of sulfate, CO2, methanol and acetate in the cockroach hindgut (Lemke, Van Alen, Hackstein, & Brune, 2001; L. Zurek &

Keddie, 1998). These diverse metabolisms are important in numerous animal gastrointestinal systems from mice (Nguyen, Vieira-Silva, Liston, & Raes, 2015;

Swanson et al., 2011; C. Zhang et al., 2010) to ruminants (Stewart, Flint, & Bryant, 1997) and humans (Gibson, Macfarlane, & Cummings, 1993) where they relieve constraints of

product inhibition and help increase metabolic throughput of polysaccharolytic microorganisms (Biesterveld, Zehnder, & Stams, 1994; Williams, Withers, & Joblin,

1994). In addition to taxonomic and metabolic diversity, gut compartment heterogeneity and temporal conditions leave the potential for interesting niche dependent and successional effects on microbial colonization. Oxygen gradients within the cockroach gut vary with gut region and developmental stage (Tegtmeier, 2016; Tegtmeier et al.,

2016), leading to a successional pattern of gut colonization as early colonizers consist of aerobes and facultative anaerobes, followed by oxygen depletion and secondary colonization by anaerobes (P Carrasco, 2014; Tegtmeier et al., 2016). Oxygen and hydrogen partial pressure and redox conditions also vary radially within the hindgut and also structures the microbial community on a sub-compartmental level (Wander W.

Sprenger, Hackstein, & Keltjens, 2007). The cockroach may thus serve as an accessible model system for the study of gastrointestinal microbial communities of high taxonomic and metabolic diversity as well as a model to study microbial succession, community assembly, and niche partitioning.

The cultivability of cockroach gut taxa facilitates their characterization and understanding potential microbial community and microbe/host interactions, enhancing the value of the study system. A fumarate-respiring Chryseobacterium was isolated from the gut of P. americana and was shown to be present in the midgut and hindgut at poplulations of 1.25x105 and 6.8x106 (Dugas et al., 2001). A novel methanogen,

Methanomicrococcus blatticola, was isolated from the hindgut of P. americana and was demonstrated to utilize methanol, methylamines, and hydrogen to produce methane (W.

W. Sprenger et al., 2000). Among over 160 gut isolates cultured from the hindgut of P. americana were eleven new species of Bacteroidetes whose genomes were sequenced and their ability to degrade dietary and recalcitrant polysaccharides was assessed (Vera-

Ponce de León et al., 2020). Given the known taxonomic diversity and range of metabolisms known to be present in the cockroach gut, many more taxa and functional interactions remain to be characterized.

Cockroaches are gregarious insects, assembling in multigenerational cohorts where they are in close physical contact and participate in coprophagy, necrophagy and cannibalism, and in the process acquire and disseminate microbiota (A. G. Appel, Sims,

& Eva, 2008; Jahnes et al., 2019; Kopanic et al., 2001; Nalepa C.A., Bignell D.E., 2001).

Indeed, specific bacterial metabolites have been identified within frass that are responsible for the aggregation behavior of cockroaches, and may be an adaptive response encouraging cockroach uptake of beneficial microbial taxa (Wada-Katsumata et al., 2015). This behavior may help contribute to the retention of core bacterial taxa that can be identified across numerous individual cockroaches (Schauer et al., 2014; Tinker &

Ottesen, 2016). Individual lineages of bacterial taxa form clades unique to cockroaches and termites (Dietrich et al., 2014) suggesting that intraspecific transmission of microbiota has been frequent and consistent enough for strong host-symbiont relationships to form over evolutionary time. The strong host-symbiont relationship is further evidenced by the heavy colonization of the cockroach gut up to four weeks after a brief inoculation via frass consumption (Jahnes et al., 2019). Long-term persistent association with gut microbiota is likely to drive the evolution of host mechanisms to

cope with and regulate microbial colonization, and the cockroach model may allow us to better understand these processes. Expansions of Toll and IMD immune pathway gene families identified within the genome of P. americana may reflect an adaptation to or a requirement for coexisting with a complex gut microbiota (Li et al., 2018).

The life history of the cockroach makes it amenable to germ-free culture, allowing it to serve as a tool to understand the assembly of complex gut communities and the host/microbial interactions among a much broader suite of gut commensals. The cockroach is both fecund and easily reared, depositing eggs in heavily sclerotized and stress resistant capsules (ootheca) that are readily surface sterilized (Gier, 1947b). The cockroach may serve as a model organism of similar complexity, unlike Drosophila, to the dominate vertebrate model of gut microbial ecology, the mouse (Mus musculus), while presenting the researcher with reduced costs and rearing complexity and thus lowering the barrier of entry into the study of gut microbial dynamics.

Germ-free systems for rearing cockroaches were devised in the early 1900’s and used to investigate nutritional requirements of the cockroach (House, H. L., Patton, 1949) as well as to prepare candidate model organisms for spaceflight (Benschoter & Wrenn,

1972). The prevention of bacterial colonization via germ-free methods can be more advantageous than antibiotic depletion of microbiota, as not all bacterial taxa are removed with antibiotics and toxic effects of antibiotics on the host can be avoided. Most germ-free cockroach experiments have involved Blattella germanica and reported slower accumulation of body mass and delayed maturation to adulthood (Benschoter & Wrenn,

1972; Clayton, 1959; House, 1949). Gier mentions a single effort at rearing germ-free P.

americana, but notes that only one insect reached 4th instar and no detailed data about growth rates or morphology were reported (Gier, 1947a). More recent efforts have expanded germ-free culture to the cockroach Shelfordella lateralis, examining oxygen dynamics upon the reintroduction of native gut bacterial isolates and the assembly of complex xenobiotic communities in the cockroach gut, with cockroaches maintained as long as 4 weeks (A Mikaelyan et al., 2016b; Tegtmeier et al., 2016). Germ-free growth of

P. americana to 5th instar, over a period of 5 weeks, was attained, with demonstration of differential developmental phenotypes between germ-free, wild-type and frass- conventionalized insects and persistent gut colonization of conventionalized insects by a single frass inoculation at first instar (Jahnes et al., 2019). Given the complexity of the cockroach gut microbiota and host adaptive response to this complex community, much remains to be learned from this model organism.

Early apparatuses for germ-free rearing included bulky aseptic gloveboxes, specialized glass containers with hard-to-find components, and simple but inadequate

Falcon tubes (Benschoter & Wrenn, 1972; House, 1949; Tegtmeier et al., 2016). None of these rearing systems share characteristics of high portability and ease of handling, construction from readily available components, and robust isolation from environmental microorganisms. This work describes novel apparatuses for rearing germ-free P. americana through the first four larval instars, or roughly half of its juvenile phase. In addition, methods for sanitizing ootheca and for inoculating germ-free insects with oxygen-sensitive anaerobes are detailed. This work demonstrates that this model organism, methodology, and apparatus may be utilized to learn about colonization

dynamics of native gut isolates and opens up the possibility for numerous lines of questioning regarding gut microbial ecology and host microbe interactions in complex gut microbial communities.

MATERIALS AND METHODS

Oothecum cleaning. Ootheca were collected directly from a breeding colony of cockroaches numbering in the hundreds of insects to maintain a steady oothecum supply.

Ootheca adhering to females were selected (Figure 1), based on degree of sclerotization, with priority given to those ootheca with a darker coloration (more fully sclerotized).

Ootheca adhering to females were much cleaner than those deposited on habitat surfaces and were synchronized to hatch ~30 days after collection. Supplies for sanitizing ootheca were autoclaved in advance and the lab bench was cleaned with 70%

Figure 1 Oothecum Collection. Ootheca that were near full sclerotization (brown) were selected for cleaning.

EtOH. A maximum of 20 ootheca were placed in an aseptic 50 mL Falcon tube and agitated in ~40 ml of autoclaved dH2O for 30 seconds. Any floating ootheca were removed, as they likely suffered from damage. Ootheca and rinse water was dumped into an aseptic Buchner funnel to quickly drain away water. The Buchner funnel was stored in

70% EtOH / 2% bleach between rinse steps to remove any deposited contaminants. This rinse was repeated two more times, with continuous agitation, and a 1 mL aliquot of the last rinse water was reserved in an aseptic microcentrifuge tube for quality control (QC) plating. Two large sponges were soaked in aseptic 1% Alconox detergent and ootheca were rubbed between the sponges to remove any adhering debris. Ootheca were checked for any surface damage, cracks, or dimples, as these indicated compromised and contaminated ootheca that could leak contaminants throughout the cleaning procedure.

Ootheca were placed into a Buchner funnel and aseptic dH2O was poured over ootheca to remove detergent. Ootheca were then placed in an aseptic 50 mL falcon tube and agitated in ~40 ml of autoclaved dH2O for 30 seconds, and water was disposed. Distilled water rinses are expected to induce osmotic stress between bacterial disruption by detergent and antiseptic washes. This rinse was repeated two more times, with continuous agitation, and a 1 mL aliquot of the last rinse water was reserved in an aseptic microcentrifuge tube for

QC plating. Ootheca were placed into a 50ml falcon tube with 20ml of 1% bleach and agitated for 3 minutes. Ootheca were placed into a Buchner funnel and aseptic dH2O was poured over ootheca to remove residual bleach. Ootheca were then placed in an aseptic

50 mL falcon tube and agitated in ~40 ml of autoclaved dH2O for 30 seconds, and water was disposed. This rinse was repeated two more times, with continuous agitation, and 1

mL aliquot of the last rinse water was reserved in an aseptic microcentrifuge tube for QC plating. The last rinse was decanted and ootheca were moved to a biosafety cabinet or laminar flow hood. Lysis buffer stock was prepared with 20 mL of 1M Tris-Cl, 2 mL of

0.5M EDTA, and 12 mL Triton X-100 to 1 L mqH2O. Ten mg lysozyme was added to each 1 mL of buffer stock just prior to use and was subsequently filter sterilized. Ootheca were placed in a 15 mL falcon tube with just enough lysis buffer to submerge ootheca (~

4 mL per 20 ootheca) and agitated for 3 minutes. Buffer was decanted and ootheca were rinsed in 40 mL dH2O in an aseptic 50 mL falcon tube. This rinse was repeated two more times, with continuous agitation, and 1 mL aliquot of the last rinse water was reserved in an aseptic microcentrifuge tube for QC plating. Ootheca were deposited on a sterile filter paper within a sterile petri dish and allowed to dry. Lids were removed from microcentrifuge tubes prior to autoclaving, and ootheca were placed into individual aseptic tubes. Sterile cotton dental wick was used to plug the top of each tube and tubes were placed in an aseptic 1000 μL pipette tip box equipped with a soaked sponge to maintain humidity (Figure 2). Ootheca were Figure 2 Ootheca incubation. incubated for ~30 days until hatching. The Ootheca were incubated in cotton capped four aliquots of rinse water were plated in microcentrifuge tubes within a pipette tip triplicate on Luria-Bertani (LB) agar plates box. to monitor bacterial load after each antiseptic wash and ensure that ootheca were free of bacteria at the final wash.

Installation to habitat. Within a biosafety cabinet or laminar flow hood, microcentrifuge tubes containing day-old nymphs were placed in an aseptic Falcon tube and submerged in an ice bath or refrigerator to immobilize nymphs. Vinyl tubes (Nalgene 180, 5/8” OD x

½” ID), 7 inches in length, were prepared in advance and autoclaved to seal and sterilize them. Plastic plugs containing 2 ml 1% agar were prepared and autoclaved to serve as a water source. Sterile gamma irradiated rodent diet was cut with an aseptic scalpel into half-centimeter chunks so that it would fit into the tube and approximately 0.75 g was added to each vinyl tube. Immobilized nymphs were shaken from the microcentrifuge tube into the vinyl tube so that three or four nymphs were deposited into the habitat

(Figure 3) and the empty oothecum was

Figure 3 Nymph Housing.

Germ-free nymphs were housed in a vinyl tube with irradiated rat food and an agar plug as a water source. Air was filtered passively through a syringe filter.

retained for QC plating. Vinyl tubes were clamped with a pinchcock to trap the nymphs at one end of the tube while the other end was stoppered with the agar containing plug. A 13mm 0.22 μm sterile syringe filter was inserted into a 16 ga hypodermic needle and pierced through the vinyl tube at a shallow angle so as to travel ~1 cm through the vinyl wall of the tube to create a firm seal around the needle (Figure 3). Habitat tubes were placed inside of a surface sterilized

Rubbermaid container and incubated at 30˚ C. Food and agar was sufficient to support 4 nymphs through 5th instar, or approximately 60 days without

intervention. The empty oothecum was macerated in 0.50 mL 1x PBS and plated in triplicate on LB agar plates to detect contamination.

Bacterial Isolates. Bacterial isolates were obtained from dilutions of adult P. americana gut homogenates, plated and maintained on modified tryptone yeast glucose (MTYG) agar and liquid medium (Vera-Ponce de León et al., 2020). All isolates were repeatedly re-streaked from isolated colonies and Sanger sequenced to confirm purity and identity.

Several isolates representing the diversity of organisms obtained by these cultivation - efforts were selected for evaluating colonization efficiency. Isolates included two

Bacteroides (PAB51 and PAB224), two Dysgonomonas (PAD521 and PAD25), one

Lachnospiraceae (PAL227) and one Fusobacterium (PAF510). GFP labeled Escherichia coli was maintained on LB medium.

Inoculation of Gnotobiotic insects Freshly installed into aseptic habitats, germ-free nymphs were deprived of water overnight by pinching the habitat at its midpoint with a pinchcock to isolate them away from the agar plug. Pure liquid cultures of cockroach gut isolates were grown overnight to approximately OD 1.0 in anaerobic Balch tubes and normalized to the same OD. Cultures were transferred to an anaerobic chamber and 1 mL of culture was collected with a sterile syringe and transferred to a microcentrifuge tube.

Culture was centrifuged and 0.66 mL of supernatant was removed to concentrate cells 3x.

The cell pellet was resuspended and ~20 μL of culture was drawn into a 1 mm OD x 0.58 mm ID capillary tube. Suction from a 10 μL pipette was used to assist capillary loading.

When polycultures were introduced, concentrated culture was combined in 1:1 ratio. A small amount of vegetable oil (2-4 μL) was applied to the top end of the capillary tube to 35

seal the tube against atmospheric intrusion. Capillary tubes were moved into the biosafety cabinet where germ-free habitat tubes were surface sterilized with 70% EtOH and pierced with a 16 ga hypodermic needle at the midpoint, perpendicular to the surface of the vinyl tube. The culture-filled capillary tube was lowered through the bore of the 16 ga needle and into the habitat (Figure 4). The hypodermic needle was then removed, leaving the culture-filled capillary held firmly within the vinyl tube. The fluid level was marked on the capillary tube with indelible marker to monitor consumption was adjusted to be within reach of nymphs. The capillary tube was left in place for 24 hours before being renewed with a freshly filled capillary tube and repeated for a total of 96 hours, or 4

Figure 4 Gnotobiotic Inoculation. capillary changes. After removal of the final

Bacterial culture was provided through capillary, cyanoacrylate glue was applied to a glass capillary tube. the hole and the pinchcock was removed to allow nymphs to access the agar plug. Habitat

tubes were placed inside of a surface sterilized

Rubbermaid container and incubated at 30˚ C.

Isolate Detection and Quantification.

Gnotobiotic insects were grown to designated

instar and dissected to remove the gut, from

esophagus to anal cavity. Guts were placed

into 200 μL aseptic 1x PBS and moved to an

anaerobic chamber to be homogenized. Gut

homogenate was serially diluted and plated in

quadruplicate on MTYG medium. DNA was extracted from remaining homogenate with a Qiagen Blood and Tissue kit. Diagnostic PCR was conducted using isolate specific primers (Table 4) and the following thermocycler conditions: 3 min 95˚ C followed by 35 cycles: 1 min 95˚ C, 15s 58˚ C, 1 min 68˚ C, followed by 3 min extension at 68˚ C.

Isolate specific primers were searched against the Arb Silva 16S database and against an in house 16S database of cockroach gut isolates to ensure specificity. Isolates specific primers were further tested against closely related species within an in-house isolate collection to ensure amplification specificity. Absolute qPCR quantification was conducted on an Eppendorf Mastercycler Realplex II thermocycler using a plasmid reference with a 192 bp 16S amplicon insert from isolate Fusobacterium PAF510 with plasmid copies at 4 dilutions, between 109-103 copies per μL, to generate a standard curve against which unknowns were compared. Unknowns were amplified with corresponding isolate specific primers designed to generate ~200 bp amplicon and three technical replicates were conducted for each biological replicate. Thermocycler conditions and isolate specific primers were the same as those used for diagnostic PCR except that the final extension step was eliminated and a melt curve analysis was performed to differentiate target amplification from primer dimer. All isolates maintain one 16S gene.

Imaging of DAPI stained insect tissues and green fluorescent protein (GFP) labeled E. coli was conducted on an inverted Nikon epifluorescence microscope.

Microaerobic vs Anaerobic growth. Two identical 96 well microtiter plates were prepared in an anaerobic chamber. Wells were filled with 200 μL anaerobic MTYG medium and inoculated with 10 μL of over-night isolate culture, in quadruplicate. Control

wells contained reduced resazurin oxygen indicator (hydroresorufin). Plates were covered with polypropylene film and one plate was incubated in an anaerobic chamber, while the other was incubated in a spectrophotometer in ambient atmosphere. Endpoint OD600 readings were acquired after 44 h and T=0 OD600 readings were subtracted to determine overall growth (ΔOD600). Oxidation controls incubated in ambient atmosphere became red (hydroresorufin -> resorufin), indicating oxygen infiltration through plates or film and into the medium while those incubated in the anerobic chamber remained clear and unoxidized.

RESULTS

Oothecum cleaning. Ootheca were collected under two different scenarios, those that were naturally deposited on colony surfaces or those still attached to females (Figure 1).

Ootheca are extruded from the female ovipositor and take several days to fully emerge, sclerotize and be deposited. Deposited ootheca were often pasted to into crevices by females using a layer of chewed up fiber and were frequently smeared with frass. For this reason, deposited ootheca can be coated with a significant amount of debris. Deposited ootheca were separated into two batches based on whether they floated or sank in the initial distilled water rinse. Floating ootheca are often more mature and closer to hatching, but also may suffer from damage and dehydration. Ootheca that remain attached to the female cockroach, nearing full sclerotization, are viable and very clean.

Both deposited and undeposited ootheca were subjected to the oothecum cleaning procedure, and bacterial load was monitored after each wash step through plating of an aliquot of rinse water. Ootheca deposited on cage surfaces frequently shed bacteria after

both the distilled water rinse and Alconox wash, and were only consistently free of bacteria after the bleach wash (Figure 28A). Ootheca collected directly from females required less vigorous scrubbing with Alconox detergent and shed fewer bacteria at earlier washes, occasionally yielding bacteria free ootheca after the Alconox wash

(Figure 29). Additionally, ootheca that sank during the initial distilled water rinse shed fewer bacteria throughout the rinse process (Figure 28A vs Figure 28B). Ootheca collected directly from females were also synchronized, and consistently hatched approximately 30 days after collection. Hatch rate was 50.6% of 530 treated ootheca over a 2 year period.

Germ-free cockroach validation. Germ-free status of insects was verified using cultivation-independent techniques including PCR and microscopy (Figure 5 and Figure

6). Cockroaches maintain a vertically transmitted obligate bacterial endosymbiont of the fat body (Blattabacterium sp.) and lose reproductive viability when it is eliminated. PCR of frass DNA extracts using universal bacterial primers was used to detect gut microbiota in frass and avoid cross-reactivity from host endosymbiont DNA. Frass extracts from germ-free treated insects rarely yielded PCR amplicon, supporting the germ-free status of the overwhelming majority of treated insects (Figure 5). DAPI stained thin sections of germ-free hindgut revealed only food- associated autofluorescence and no fluorescence

Figure 5 Germ-free Insect Validation – PCR.

Gel electrophoresis of frass DNA extracts amplified by PCR with universal 16S primers. Contaminated insects yield amplicon near 1400 bp while germ free insects yield no amplicon.

Figure 6 Germ-free Insect Validation – Microscopy.

Thin sections of wild-type cockroach hindguts display dense accumulation of DAPI stained bacterial biomass at the lumen/epithelial interface, while germ-free tissues display an absence of bacterial biomass. Distinct bacterial morphologies are visible at 1000x magnification within the lumen of wild-type thin sections, while only autofluorescence is visible in the germ-free lumen.

consistent with bacterial morphology, further supporting the germ-free status of sampled insects (Figure 6).

Detection of monoculture gnotobiotic colonizers.

Insects were monoinoculated with P. americana anaerobic gut isolate Dysgonomonas

PAD521 or GFP labeled E. coli. At sampling, insects were surface sterilized with 70%

EtOH, homogenized, dilutions plated on MTYG medium, incubated anaerobically and

CFU were enumerated. Isolate Dysgonomonas PAD521 was not recovered by dilution plating however E. coli was recovered and averaged 2.5x105 CFU mg-1 body mass

(Figure 7). E. coli was further detected in-situ within the hindgut by fluorescence microscopy (Figure 7). Because no viable isolate was recovered by dilution plating

Figure 7 E. coli Gut Colonization Detection and Enumeration.

A. Enumerating gut colonization by E. coli via dilution plating B. In situ localization

of GFP labeled E. coli in the cockroach hindgut. from Dysgonomonas PAD521 inoculated insects, DNA was extracted from gut homogenates and subjected to PCR with primers designed to specifically amplify

Dysgonomonas PAD521 16S rDNA. Positive amplification within gut extracts demonstrates effective colonization of the gut by isolate Dysgonomonas PAD521 (Figure

8). Additional, anaerobic isolates introduced in monoculture proved to be difficult

Figure 8 Detection of Isolate PAD521 Colonization.

Detection of gut colonization by diagnostic PCR amplification of isolate specific

DNA.

to recover by dilution plating, including

isolates Bacteroides PAB224,

Bacteroides PAB51, and

Lachnospiraceae PAL227 leaving PCR

as the favored method for detection of

colonizers.

Bacterial succession. Insects were

monoinoculated with P. americana gut isolates Bacteroides PAB224, Bacteroides PAB51, and Lachnospiraceae PAL227.

Inoculated insects were sampled at 2nd instar to detect colonization after a single molt and at 5th instar to determine whether colonization persists through several molts with only a single 4 day inoculation during early 1st instar. Because isolates Bacteroides PAB224,

Bacteroides PAB51, and Lachnospiraceae PAL227 were not recovered by dilution plating, dPCR was used to detect isolate DNA in extracts using isolate specific primers.

All isolates were detected at second instar (Figure 9A-B, Figure 10A), however only

Bacteroides PAB51 was detected at 5th instar (Figure 10A). Bacteroides PAB51 16S copy number was quantified with qPCR conducted on gut DNA extracts at 2nd and 5th instar indicating an average of 3.2x104 16S copies mg-1 body mass at 2nd instar and an average of 3.3x105 16S copies mg-1 body mass at 5th instar (Figure 10B).

Figure 9 Detection of Isolate PAL227 and PAB224 Colonization.

Isolate colonization detected at 2nd instar by diagnostic PCR of gut DNA extracts using isolate specific primers, but not at 5th instar.

Detection of polyculture gnotobiotic

colonizers. Insects were polyinoculated

with P. americana anaerobic gut isolate

Dysgonomonas PAD25 and

Fusobacterium PAF510. At sampling,

insects were surface sterilized,

homogenized, dilutions plated on

MTYG medium, plates were incubated

anaerobically and CFU were

enumerated. Both isolates were

recovered by dilution plating and

averaged 5.2x106 CFU mg-1 body mass

for isolate Dysgonomonas PAD25 and

8.3x105 CFU mg-1 body mass for

isolate Fusobacterium PAF510 (Figure

9A). Because all isolates are not easily differentiated by colony morphology or recovered in cultivation by dilution plating qPCR was evaluated as a means of quantifying gut colonization by polycultures, relying on

amplification by isolate specific primers to differentiate each colonizing isolate.

Quantification by qPCR indicated average 16S counts of 9.4x106 copies mg-1 body mass for isolate Dysgonomonas PAD25

Figure 10 Detection of Colonization by Isolate PAB51

Isolate colonization detected at 2nd and 5th instar by diagnostic PCR of gut DNA extracts using isolate specific primers. Enumeration of bacterial populations at 2nd and 5th instars using qPCR.

and 1.3x105 copies mg-1 body mass for isolate Fusobacterium PAF510 (

B). Interestingly, in monoculture Fusobacterium PAF510 failed to persist in

Figure 11 Detection of Colonization by Polyculture of Isolates PAD25 and PAF510

A. Polyculture colonization enumerated by dilution plating. B. Polyculture colonization enumerated at 5th instar by qPCR of gut DNA extracts using isolate specific primers. C. Detection of bacterial monoculture using diagnostic PCR indicates a lack of monoculture colonization by isolate PAF510.

colonization to 5th instar while isolate

Dysgonomonas PAD25 maintained

colonization to 5th instar (Figure 11C).

Microaerobic vs. Anaerobic Growth.

Fusobacterium PAF510 reached 1.36 ΔOD600

after 44h under strict anaerobic conditions

(Table 1). Under the low-level oxygen

diffusion through polyethylene film covering

the microtiter plates, Fusobacterium

ΔOD600. Dysgonomonas PAD25 reached 1.15

ΔOD600 after 44h under strict anaerobic

conditions, while under conditions of diffusive

oxygen exposure, Dysgonomonas PAD25

reached 1.31 ΔOD600 over the same time

period.

DISCUSSION

This work demonstrated that, like other cockroaches, P. americana may be reared through early instars in a germfree state with little difficulty and may be recolonized with select microbiota. Constructed from widely available, “off the shelf” components, assembly of the rearing system can easily and inexpensively be accomplished by other researchers. The apparatus described for isolating germfree nymphs from environmental microbial 49

contaminants was demonstrated to consistently maintain the insects in a germ-free state for over six weeks, as reflected by the results of dPCR (Figure 5) and microscopy (Figure 6).

Table 1 Microaerobic vs. Anaerobic Growth.

Comparison of growth under microaerobic conditions and fully anaerobic conditions indicates a growth advantage by isolate PAD25 in the presence of oxygen and a growth defect in isolate PAF510 in the presence of oxygen.

Oothecum source (whether colony deposited or harvested directly from the female) and condition (floating vs sinking) appear to play an important role in sterilization success, as reflected in growth on quality control plates from oothecum rinses (Figure 28). In contrast, Tegtmeier et al. 2016 used a strategy of selecting floating ootheca because these were more mature and closer to hatching. In our experience floating ootheca were more frequently cracked and compromised by contamination than sinking ootheca, and sometimes flotation was due to partial desiccation of oothecum.

Those sinking ootheca collected directly from females were more consistently clean and hatch was synchronized at approximately 30 days post collection date, allowing better scheduling of research activities.

The capillary method for introduction of bacterial isolates to germ-free cockroaches combined with the soft, penetrable vinyl of the habitat enclosure allowed

inoculation of the insects without removal from enclosures, avoiding potential exposure to contaminating microorganisms that may occur with excessive handling. Additionally, the capillary method may serve as an improvement over the Tegtmeier et al. (2016) method of releasing nymphs onto culture-soaked filter paper. The glass capillaries should comparatively limit exposure of the bacterial culture to oxygen and evaporation, with only a small surface area exposed at the capillary opening. This methodology may allow the introduction of increasingly oxygen sensitive microorganisms, that would otherwise have difficulty maintaining viability with full exposure to ambient conditions. E. coli was introduced to germ-free insects and was recovered by dilution plating of gut homogenates and was observed in-situ by epifluorescence microscopy, enabled by GFP expression.

While none of the cockroach gut isolates utilized in this work have been engineered to express fluorescent proteins, this proof of concept demonstrates that GFP labeled bacteria can be viewed in whole-mounted guts, and visualized through the host lumen (Figure 7), enabling the observation of gut niche and potentially allowing microbial associations to be visualized within the gut without complicated and time-consuming histological preparation. Monoculture gut colonization by native cockroach anaerobic isolates was successfully demonstrated by detection with isolate specific primers but not dilution plating. The process of tissue collection and preparation appears to negatively affect the survivability of a number of the isolates. Interestingly, an isolate dependent pattern of gut colonization persistence was observed. As cockroaches develop through larval instars, they undergo periodic molting of their exoskeleton along with portions of their foregut and hindgut lining. This represents a period of disturbance during which gut colonization

could potentially be lost. Detection of gut isolates after one or multiple molting events is viewed as a test of the isolates ability to persist despite perturbations experienced throughout cockroach metamorphosis. Isolates Bacteroides PAB224, Bacteroides

PAB51, and Lachnospiraceae PAL227 were all detected in hindgut DNA extracts after one molting event, approximately one week after inoculation. This is sufficient time for the original inoculant to pass through the digestive tract, as gut transit in P. americana has been measured at approximately 20 hrs (Snipes & Tauber, 1937). This time period allowed multiple purges of the gut contents between the initial inoculation and sampling, leaving a low probability that detected isolate DNA was carry-over from inoculation.

This indicates that all three isolates were able to multiply and persist within the gut through one molting event. In contrast, sampling after 4 molting events, or 5th instar, revealed colonization in only insects inoculated with Bacteroides PAB51. This indicates a differential ability of native gut isolates to persist in the cockroach gut in monoculture, and suggests that some isolates are more effective colonizers than others or may be more readily retained through cycles of coprophagy and exuvium consumption. qPCR quantification of 16S rDNA at 2nd instar compared to 5th instar in Bacteroides PAB51 colonized insects demonstrates an increase in 16S count per mg of insect mass over time, indicating greater colonization with time. How this corresponds with a simple increase in gut biomass, volume, and epithelial surface area is unclear.

Two isolates, Dysgonomonas PAD25 and Fusobacterium PAF510 demonstrated differential colonization after monoculture inoculation, with Dysgonomonas PAD25 persisting to 5th instar while Fusobacterium PAF510 did not persistently colonize and

was undetected at 5th instar. Interestingly, with polyculture inoculation of the same two isolates, Fusobacterium PAF510 was able to colonize and persist in the gut to 5th instar.

This suggests that colonization efficiency may be dependent on commensal or mutualistic interactions between cohabitating microorganisms. Dysgonomonas PAD25 demonstrated a subtle increase in growth under low oxygen conditions compared to anaerobic conditions (Table 1) and its genome encodes a microaerobic cytochrome bd oxidase

(CydAB) and NADH dehydrogenase (NuoA-N) (B.C.Jahnes, unpublished data), suggesting that it may be able to metabolize small quantities of oxygen which may reduce oxygen partial pressure in the gut environment. High affinity cytochrome bd oxidase has been identified in other facultative and obligate anaerobes and is implicated in the oxidative stress response and respiratory growth in low oxygen conditions (Baughn &

Malamy, 2004; Das, Silaghi-Dumitrescu, Ljungdahl, & Kurtz, 2005; Giuffrè, Borisov,

Mastronicola, Sarti, & Forte, 2012). In contrast isolate Fusobacterium PAF510 demonstrated a marked decrease in overall growth under low oxygen conditions compared to anaerobic conditions (Table 1), suggesting that it is inhibited by oxygen exposure. Correspondingly, Fusobacterium PAF510 maintains no cytochrome oxidase or

NADH dehydrogenase in its genome (B.C. Jahnes, unpublished data). Thus, a workable explanation underlying differential colonization by Fusobacterium PAF510 is that this isolate cannot cope with oxygen concentrations within an uncolonized gut, however co- colonization with Dysgonomonas PAD25 transforms low oxygen conditions within the gut to approach an anaerobic state in which Fusobacterium PAF510 can more effectively colonize and persist. This corresponds with observations by Carrasco et al. (2014) that

Fusobacterium is absent or at low titers in early instar cockroaches and found in increasing abundance in later instar cockroaches, perhaps corresponding with increased gut anaerobiosis.

Future work will leverage the utility of this experimental system, and take advantage of the numerous bacterial isolates obtained from P. americana, to examine colonization dynamics within defined gut microbial communities of increasing complexity and effects on host development.

Chapter 3. Conspecific Coprophagy Stimulates Normal Development in a Germ-Free Model Invertebrate.2

ABSTRACT

Microbial assemblages residing within and on animal gastric tissues contribute to various host beneficial processes that include diet accessibility and nutrient provisioning, and we sought to examine the degree to which intergenerational and community-acquired gut bacteria impact development in a tractable germ-free invertebrate model system.

Coprophagy is a common behavior in cockroaches that provides access to nutrients and serves as the primary means by which juveniles are inoculated with mutualistic gut bacteria. This hypothesis was tested in the American cockroach (Periplaneta americana) by preventing coprophagy, resulting in germ-free insects that exhibited prolonged growth rates and gut tissue dysmorphias when compared to wild-type P. americana. Germ-free

P. americana that were conventionalized by allowing coprophagy from conspecifics and siblings reared under nonsterile conditions resulted in colonization of P. americana gut tissues by a dense microbial community and significant (p<0.05) recovery of wild-type level growth and hindgut tissue development phenotypes. These data suggest that although P. americana can be reared in a germ-free manner, coprophagy likely

2 This chapter was reproduced verbatim from Jahnes et al. 2019 PeerJ. 55

introduces mutualistic gut bacteria to P. americana that contribute to normal gut tissue and organismal development.

INTRODUCTION

Host-associated bacteria are increasingly recognized as being inextricably involved in the health, development and evolution of their animal hosts, and the digestive tract is a major intersection for the host and its environment, and its gut bacterial consortia. In vertebrates, microbial ecosystems located within animal digestive tracts, hereafter referred to as ‘gut microbiota’, (Yasuda et al., 2015; Costea et al., 2018) can impact host health, behavior, dietary nutrient accessibility, infectious disease susceptibility and disease morbidity (Sonnenburg & Bäckhed, 2016; Koppel, Maini Rekdal & Balskus,

2017; Parker et al., 2017; Foster, Rinaman & Cryan, 2017; Zhao & Elson, 2018).

Invertebrates also harbor sparse-to-abundant multispecies microbial communities within gastric tissues that can be as specialized as those observed in vertebrates (Dillon &

Dillon, 2004; Engel & Moran, 2013; Douglas, 2015), and the global distribution, trophic diversity, ecological contributions, and agricultural, environmental and human health impacts of insects have stimulated interest in insect-gut microbiota interactions. Although host diet accessibility and nutrient metabolism has been a major focus of insect host-gut microbiota work (Six, 2013; Engel & Moran, 2013; Brune & Dietrich, 2015; Peterson &

Scharf, 2016; Kwong & Moran, 2016; Bonilla-Rosso & Engel, 2018), these microbial assemblages have also been shown to participate in zoonotic infections (Aksoy, 2018), mediate plant-insect interactions (Hansen & Moran, 2014; Casteel & Hansen, 2014;

Shikano et al., 2017), shape host behavior (Keesey et al., 2017), and stimulate immune system functions (Ryu et al., 2008; Buchon, Broderick & Lemaitre, 2013) and host growth and development (Coon et al., 2014; Zheng et al., 2017). When the gut microbiota is comprised of, or dominated by, a single or few taxa, it is possible to link individual taxa to host benefits, as observed in Riptortus pedestris stink bugs (Burkholderia that confer pesticide resistance, (Kikuchi, Hosokawa & Fukatsu, 2007; Itoh et al., 2018)) and in honey bees (Gilliamella that degrades pollen constituents and Bifidobacterium asteroids that produce growth-promoting hormones, (Engel, Martinson & Moran, 2012;

Kešnerová et al., 2017). Additionally, amenability of these insect models, as well as

Drosophila fruit flies (Apidianakis & Rahme, 2011), to axenic/germ-free rearing has greatly facilitated efforts to assign discrete functions to specific gut taxa.

Among invertebrate model systems, cockroaches are also proving to be useful for exploring host-gut microbiota interactions. The gut microbiota of omnivorous cockroaches had been characterized primarily by cultivation and microscopy approaches

(Bracke, Cruden & Markovetz, 1979; Cruden & Markovetz, 1980, 1987; Umunnabuike

& Irokanulo, 1986; Dillon & Dillon, 2004; Thompson et al., 2012) until cultivation- independent community profiling (i.e. 16S rRNA gene amplicon sequencing) approaches became widely available. Tremendous phylogenetic diversity, spanning over 20 bacterial phyla (Mikaelyan et al., 2015), has been detected in cockroach guts, with members of the

Bacteroidetes, Firmicutes, Fusobacteria, and Proteobacteria comprising the majority of the gut microbiota, and diet appears to exert a significant impact on presence and relative abundance of the gut microbiota (Schauer, Thompson & Brune, 2012; Krych et al., 2013; 57

Sabree & Moran, 2014; Schauer et al., 2014; Dietrich, Köhler & Brune, 2014; Perez-

Cobas et al., 2015; Mikaelyan et al., 2015; Tinker & Ottesen, 2016; Zheng et al., 2017).

Additionally, cultivation under oxygen-limited conditions has coaxed new species of cockroach gut bacteria into cultivation (Tegtmeier et al., 2016, 2017).

Cockroaches are amenable to axenic rearing and such studies have helped to shed light on some possible roles of gut bacteria in cockroach development and behavior. The majority of these studies have focused on the cockroach Blatella germanica, reporting successful but delayed development to adulthood, with normal fertility and viability of young

(House, 1949; Clayton, 1959; Benschoter & Wrenn, 1972). Host aggregation, a common cockroach behavior, was found to be stimulated by gut bacterial products in Blattella germanica that were absent when B. germanica was treated with antibiotics (Wada-

Katsumata et al., 2015). Germ-free Shelfordella lateralis cockroaches have facilitated the determination of how oxygen impacts gut tissue colonization and metabolic activity of two bacterial strains isolated from S. lateralis (Tegtmeier et al., 2015). Additionally, digestive tracts in germ-free S. lateralis exposed to environmental and animal-derived inocula were capable of enriching for bacteria, primarily members of the Bacteroidetes,

Firmicutes and Proteobacteria, that were closely related to gut residents typically found in

S. lateralis gut tissues (Mikaelyan et al., 2016). Furthermore, a clear phylogenetic signal was detected between inoculum source (i.e. cockroach > termite > mice > soil) along with recapitulation of a gut microbiota composition typically observed in S. lateralis

(Mikaelyan et al., 2016). Although these efforts have made great strides in illustrating how cockroaches and their gut microbiota collaborate, less is known about host 58

physiological responses to gut microbiota colonization. Recent work has shown that consumption of feces (coprophagy) from conspecifics and nestmates is common in cockroaches (Nalepa, Bignell & Bandi, 2001; Kopanic et al., 2001), and is a means for the transfer of hindgut bacteria (Rosas et al., 2018). Additionally, coprophagy provides consumers with amino acids, lipids, carbohydrates and micronutrients that remain in the diet post-digestion and from microbes that have colonized the fecal pellet during digestion or since its deposition in the environment (Nalepa et al. 2001). Cockroach nymphs exhibit the strongest responses to aggregation pheromones present in feces

(Dambach et al. 1995), of which at least some of these chemical signals are produced by bacteria therein (Wada-Katsumata et al., 2015), suggesting that inoculation of early-stage nymphs with gut bacteria via coprophagy is important for normal development. This study seeks to detail the impact of coprophagy on P. americana physiology and development, and it is expected that physiological systems within the cockroach that are most heavily influenced by exposure to microbiota-enriched frass will highlight sites of host-microbiota interactions for further study.

MATERIALS AND METHODS

Insects. P. americana nymphs, adults and ootheca were obtained from a live collection maintained for 60+ years in the Insectary at the Ohio State University Biological

Sciences Greenhouse (Columbus, Ohio).

Ootheca Treatment. Ootheca were manually detached from gravid P. americana females and surface sterilized by three rounds of cleaning using a detergent scrub (1% 59

Alconox detergent), dilute bleach ( 0.08% sodium hypochlorite), and then an enzymatic lysis buffer (lysozyme 10mg/ml, EDTA, Tris-HCl and Triton X-100), each step followed by three rinses with sterile MilliQ water (MQW) to remove antiseptic solution and induce intermittent hypo-osmotic shock on surface-clinging bacteria (Schwinghamer, 1980;

Salema et al., 1982). An aliquot of each third MQW rinse was reserved to monitor effectiveness of previous cleanings to reduce bacterial load. Aseptic ootheca were incubated individually in sterile microcentrifuge tubes capped with sterile cotton at 31°C for approximately 30 days until hatching, which yielded up to 16 germ-free nymphs per oothecum. No decreased oothecum viability due to washing treatments was observed.

Rearing germ-free P. americana. Cohorts of three to four germ-free (GF) 1st instar nymphs were aseptically-transferred to sterile rearing chambers of approximately 16 cubic centimeters, stocked with gamma-irradiated (aseptic) rat chow and aseptic water, which was renewed weekly. Rearing chambers were passively ventilated with air filtered through a 0.22 µm membrane.

Rearing conventionalized P. americana. To introduce bacteria native to P. americana, cohorts of three to four GF 1st instars were exposed to frass taken from a lab-maintained colony of nonsterile or ‘wild-type’ (WT) P. americana in lieu of food for 3 days.

Subsequently, conventionalized (Conv) 1st instar nymphs were housed under the same conditions as GF insects except for that sterility was not maintained. As each generation of P. americana typically acquires their gut microbial community through conspecific and nestmate coprophagy (Nalepa et al. 2001; Bell et al. 2008), the conventionalization approach used reflects the normal route of gut microbe acquisition. 60

Rearing wild-type P. americana. One-day old 1st instar insects from 10 ootheca were deposited in an aquarium containing 10 adult male cockroaches from a nonsterile mixed generation colony maintained in the lab and provided with gamma-irradiated rat chow and access to MQW ad libitum; nymphs from this colony were designated ‘wild-type’

(WT). WT hatchlings were free to interact with adult cockroaches and their frass to facilitate coprophagy and subsequent acquisition of normal gut microbiota. As cannibalism of deceased nestmates is also a putative mechanism for gut microbiota acquisition, late-stage nymphs from the nonsterile colony were sacrificed and deposited in the WT colony. WT insects did not experience spatial constraints or undergo the oothecum sterilization procedure as compared to GF and Conv insects and were exposed to unfiltered air.

Quality Control. Quality control measures were employed throughout experiments to ensure maintenance of germ-free status and to confirm colonization of Conv individuals.

Oothecum rinse water, collected between each antiseptic treatment, was plated in triplicate on Luria-Bertani agar (LBA) plates (incubated at 31˚C for up to 1 week) to detect cultivable contaminants flushed from ootheca, and to provide confidence that ootheca are aseptic as they enter incubation. Ootheca shedding no contaminants at final rinses were considered to be aseptic and used for subsequent experiments. At the time of hatching and installation into habitats, one 1st instar nymph from each oothecum, or the oothecum itself, was sacrificed and homogenized in sterile 1% PBS and plated on LBA in triplicate to confirm aseptic status. At the termination of insect growth, and just prior to insect dissection, frass was collected from rearing chambers, suspended in 1% PBS, and

plated to confirm that habitats remained germ-free throughout the duration of the experiment. Bi-monthly sampling of GF-reared individuals for contamination was performed by homogenizing individuals at various instars and plating homogenate on

LBA. During sterile treatment method development, diagnostic PCR with universal 16S primers 27F (5’-AGAGTTTGATCMTGGCTCAG-3’) and 1391R (5’-

GACGGGCGGTGTGTRCA-3’) was performed on DNA extracted from homogenates prepared from germ-free insects and their frass to detect non-cultivable contaminants of the gut or habitat. Additionally, microscopic examination of DAPI-stained frass, gut contents, and gut thin-sections were performed to further evaluate the effectiveness of the germ-free, and conventionalization protocols.

Instar Duration Measurement. GF and Conv nymphs raised in cohorts of 3-4 insects were monitored individually for molting activity, and dates of instar transitions of individuals within each cohort were recorded. Aquarium rearing of WT insects in cohorts of dozens of individuals prevented tracking of individual insects and corresponding molt date and developmental times within the WT treatment. Insects dissected at subsequent instars for gut morphological measurements resulted in decreasing sample sizes at later instars, and non-parametric statistical methods were used when making comparisons across instars. Targeting early instar insects constrains the experiment to a relatively short timeframe, as rearing to adulthood is prohibitively long (6-12 months for WT insects), with GF P. americana progression to adulthood uncertain.

Morphological Measurements. Nymphs were collected as they molted to 3rd, 4th, and

5th instars, and duration (days) of these instars were recorded. Based on unpublished

preliminary data, the cumulative effects of bacterial colonization, or lack thereof, were apparent in 4th and 5th instars and thus the experiment was terminated at fifth instar.

Nymphs at designated life stages were dissected approximately 5 days after molting, at peak of feeding within instar (Valles et al. 1995), to minimize variability in gut morphology associated with instar transitions. Eight morphological metrics were collected in millimeters (mm), unless otherwise noted: body length, body width, body mass (grams, g), whole gut length, foregut length, midgut length, hindgut length, and gut mass (g). FIJI image analysis package (ImageJ) was used to perform measurements of the full length of dissected guts (esophagus to anus) and their corresponding carcasses from digital images of these tissues taken immediately following dissection. Results of statistical tests performed on comparisons of morphological measurements are reported in

(Table 5). Additionally, gut compartments and bodies were traced and lengths measured, with measurements calibrated to a scale in each photo. Mass measurements were also collected for whole GF, Conv and WT individuals prior to dissection and of their dissected digestive tracts using a microbalance. Additionally, qualitative observations of gut texture, color, opacity, and segmentation were collected from dissected 5th instar GF,

Conv and WT individuals.

Phenotype Analysis. All statistics were performed in R using vegan and FSA packages

(Ogle; R Core Team, 2013; Oksanen et al., 2015). Morphological measurements at instar rather than calendar age yielded differences in sample size, as insects aged out of instar classes at different rates; non-parametric statistics were utilized to accommodate sample size variation. Instar duration, and morphological measurements were analyzed for

statistical differences using a Mann-Whitney Test for pairwise-comparisons. Kruskal-

Wallis and Dunn Tests were performed for multiple comparisons, with Bonferroni correction-adjusted p-values. Principal Component Analysis (PCA) was performed to simultaneously examine multiple morphological variables across treatments. The function ‘Varpart’ within the vegan R package was used to partition variance among morphological variables and the PCA was constrained to the 3 variables representing

93% of the variation. Subsequently, multi-response permutation procedure (MRPP) was used to assess the significance of the observed differences between treatment centroids.

RESULTS

Germ-free insects exhibited reduced width and prolonged instar duration. GF

(n=37) P. americana individuals remained in each instar an average of two days longer than Conv (n=34) individual (Figure 12A), which resulted in longer times between molts in GF insects and prolonged developmental periods (Figure 12B) being observed. While

Conv insects molted to 5th instar after an average of 32.3 days, GF insects required an average of 38.9 days to reach the same life stage. While stadium duration of WT insects was not obtained for every instar because of the complexity of tracking individual insects in a large cohort, average age at 5th instar was 30.2 days (Figure 30). GF insects also exhibited the lowest average body width (3.84 mm) when compared to WT (4.18 mm) and Conv (4.08 mm) insects (Figure 12C). Body lengths of GF and Conv individuals were not significantly different at 3rd, 4th or 5th instars (Figure 12D), and body mass did not differ between treatments (Figure 31).

Figure 12 Insect Maturation Rate Limiting access to gut bacteria via coprophagy prolongs development in Periplanata americana. A) Duration (stadium) of individuals within each life stage; B) Cumulative development time; C) Body width measurements; D) Body length measurements. Asterisks indicate significant (p<0.05) differences given a Dunn test. n= number of individuals measured per sample.

Conventionalization of germ-free P. americana recovers normal hindgut tissue morphology and development. Distinct visual contrasts between 5th instar WT (n=23)

(Figure 13A) and GF (n=32) (Figure 13B) hindgut tissues were consistently evident.

WT P. americana hindguts are characterized by numerous lateral folds along the length of the hindgut that are visible as an undulating gut margin with lateral creases (Figure

13D). All inspected WT hindguts were opaque, turgid in texture, and generally filled with mustard-yellow hindgut contents (Figure 13A). Conv hindguts (Figure 13C) were similar to those from WT insects in that they were turgid in texture and more opaque than 65

GF hindguts (Figure 13B), and Conv hindguts exhibited frequent lateral folds and coloration characteristic of WT insects. Further, dissection of Conv revealed mustard- yellow contents, similar to those observed in WT digestive tracts, and this appears to be a mix of digestate and bacterial biomass (note the fluorescence of material in gut lumen in

Figure 14) lodged in hindgut folds. DAPI stained hindgut tissue thin sections reveal a dense mat of bacterial biomass adjacent to the host gut epithelial lining and planktonic bacterial growth deep within the lumen of WT and Conv insects (Figure 14). In contrast,

GF hindguts were uniformly flaccid, translucent, and displayed less frequent and less pronounced lateral folding (Figure 13B). Additionally, gut contents were comprised of partially digested diet and no evidence of bacteria were observed (Figure 14). No amplicon was observed after diagnostic PCR conducted on GF frass DNA extract

(Figure 32) using bacteria-specific primers. Despite thorough cleaning of fat bodies from guts at the time of dissection, the fat body specific endosymbiont Blattabacterium was detected in diagnostic PCR reactions of gut tissue DNA extracts from GF insects, negating the utility of bacteria-specific primers for contamination detection in gut homogenates by diagnostic PCR.

Figure 13 Qualitative Comparisons of Hindgut Morphology in 5th instar Periplaneta americana. Exemplars of one wild-type (A), germ-free (B), and conventionalized (Conv.; C) P. americana individual is presented. A magnified hindgut from a wild-type P. americana (D) is provided to highlight normal morphological features, including undulating gut margin highlighted by a thick white line and gut segmentation highlighted with thin dashed lines. Scale represents 1mm.

Figure 14 Hindgut Microbial Biomass Wild-type and conventionalized Periplanata americana hindguts are enriched with microbial biomass that aggregates adjacent to the epithelium, while germ-free hindguts lack microbial biomass. 4′ 6-diamidino-2-phenylindole (DAPI) stained thin-sections of hindguts from wild-type (WT) (n=3), germ-free (GF) (n=3) and conventionalized (Conv) (n=3) P. americana were viewed under epifluorescence to visualize DNA associated with epithelium and luminal contents. h – Hemocoel; l – lumen; n - gut epithelial cell nuclei; m - microbial biomass; arrows – fluorescence consistent with bacterial morphology.

Average length of the complete fore-to-hindgut tissues was shorter in GF insects

(17.13 mm) than in both Conv (18.96 mm) and WT (21.24 mm) insects (Figure 15A;

Dunn Test: GF-Conv p=0.0244, Conv-WT p=0.0028, GF-WT p=0.0054). When fore-

(Figure 33) and mid-gut (Figure 15B) sections were examined separately, average lengths did not differ significantly between any treatments at any instar, except for between GF (3.30 mm) and either WT (4.83 mm) or Conv (3.48 mm) 3rd instar midguts

(Dunn Test: WT-GF p=0.0058, Conv-GF p=0.0455). Conversely, hindgut length was significantly reduced in GF insects at 3rd, 4th and 5th instars relative to either Conv or WT insects (Figure 15C). Finally, significant differences in gut mass were only detected at

4th instar between GF and WT insects (Figure 34), with respective masses of 5.2 and 6.7 mg (Dunn Test: WT-GF p<0.0070), but variability was high within treatments.

Figure 15 Gut Morphology Conventionalization of germ-free Periplaneta americana partially recovers wild-type gut length morphology as individuals age. Whole gut (A) midgut (B) and hindgut (C) lengths were measured in germ-free (GF), conventionalized (Conv) and wild- Type (WT) individuals at third, fourth and fifth instars, and, at the sixth instar for WT and Conv individuals. Asterisks indicate significant (p<0.05) differences given a Dunn test. n= number of individuals measured per sample.

Wild-type, conventionalized, and germ-free insects bear morphological signatures that define a gradient across treatments. Of eight morphological variables examined in this study whole gut length, midgut length, and hindgut length explained 91.4% of variance across treatments (Figure 35). Principal Component Analysis (PCA) was performed to examine the relationship among treatments at each instar in reference to these three response variables. Ordination was conducted on the first two principal components, representing 84.5% and 6.9% of variance explained, respectively (Figure

16). Treatment centroids were significantly different (MRPP: p<0.001), with the Conv centroid being intermediate to that of GF and WT centroid. All vector loadings associated with morphological variables were significant, with strong linearity (p<0.001, whole gut length r²=.99, midgut length r²=.91, hindgut length r²=.89). Loading vectors for hindgut length and whole gut length demonstrate a gradient of low values in proximity to the GF centroid to high values towards the WT centroid. The vector for midgut length appears to follow a gradient associated with within-instar variation, as midgut length varies more within instar than across treatments. A second PCA was performed on data from all instars and treatments, simultaneously, with 93.1% and 1.9% of variance explained by the first two axes, yielding greater linearity of loading vectors but reduced separation of centroids (Figure 36). With multiple instars represented within each treatment, vector loadings previously associated with within-instar variation (i.e. midgut length) realign to describe an across-treatment gradient, implying that variation in midgut length is better explained by treatment than instar or within-instar effects.

Figure 16 Principal Component Analysis (PCA) of Morphological

Characteristics at 5th instar.

PCA and ordination was performed, constrained by the three variables explaining the greatest variation, specifically, whole gut length (wg), hindgut length (hg), and midgut length (mg). The first component explained 84.5% of variation, with 6.9% and 2.2% explained by the second and third components. Treatment clusters displayed the least overlap at instar 5 and are situated with Conventionalized

(Conv) treatment sandwiched between germ-free (GF) treatments and Wild-Type

(WT). Distance between treatment centroids was 0.22 between the WT and GF centroids, 0.15 between the Conv and GF centroids, and 0.12 between the Conv and

WT centroids. All vector loadings are significant (p<.000999) with r2 values of 0.99

(wg), 0.91 (hg), and 0.89 (mg). Vector length is proportional to fit.

DISCUSSION AND CONCLUSIONS

We demonstrated that germ-free (GF) P. americana exposed to frass from lab-reared, wild-type (WT) conspecifics partially restored instar duration periods and digestive tissue development, particularly in hindgut tissues, to levels observed in WT insects, and these findings suggest a relationship between exposure to gut microbiota in cockroach frass and host digestive tissue and developmental outcomes. P. americana and other cockroaches typically live in dense multigenerational aggregates that afford them easy access to waste, discarded exoskeletons, and dead bodies, and the microbes therein, of conspecifics (Mira,

2000; Nalepa, Bignell & Bandi, 2001; Kopanic et al., 2001; Rychtár et al., 2014). These social and dietary behaviors are common in cockroaches and reflect a reliable means for 74

the transmission of gut microbes that contribute to host fitness (Nalepa, Bignell & Bandi,

2001; Sabree & Moran, 2014). We hypothesized that P. americana relies upon community-acquired gut microbes for normal development, and preventing their exposure to these microbes (as in GF insects) would result in growth and developmental defects. GF insect growth stalled in 4th and 5th instars, as no insects progressed to 6th instar within the timeframe of the experiment and some insects remained in 5th instar indefinitely. Future work may determine whether germ free P. americana progress to adulthood. GF insects conventionalized by a single exposure to conspecific frass early in life (i.e. Conv insects) exhibited a partial, but significant, remediation of instar duration

(Figure 12A-C) and gut tissue development (Figure 13A-D, Figure 14 and Figure 15A-

C). Interestingly, Conv insects exhibited an initial delay in growth and maturation at first instar, compared to both WT and GF insects (Figure 12A-D). This was likely due to the

3 day exposure to frass, without food, to enforce frass consumption. Yet, the cumulative benefit of microbial colonization allowed the Conv insects to surpass the GF and approach growth and maturation levels observed in WT insects by 5th instar. Given these data, it was surmised that the observed positive growth and developmental responses to frass exposure and subsequent gut colonization were cumulative due to persistent colonization of the cockroach gut rather than a burst of growth at time of exposure. It is notable that Blattabacterium was not removed as part of making GF insects and was present in GF, Conv and WT insects, yet its contributions alone did not support normal host development in the GF insects.

The partial recovery of WT growth and development observed may be due to Conv insects having received only a single exposure to conspecific frass, instead of multiple exposures, during development. Specifically, Conv insects were restricted to a single exposure to WT frass during their first instar and were housed in small cohorts (n<4) for the duration of the experiment (i.e. through fifth instar) and each experienced at least four moltings. Molting represents a potential bottleneck event during which portions of the gut lining, along with substantial gut contents and associated bacteria, are shed (Engel &

Moran, 2013). If the exuvium and gut lining are not rapidly consumed by insects there is potential for loss of gut microbes, especially oxygen-sensitive taxa (e.g. some members of the Bacteroidetes and Firmicutes), that may not easily be horizontally reacquired through coprophagy within a relatively tiny cohort of insects of the same life-stage experiencing the same experimental conditions. Under normal conditions, developing insects living in multigenerational, population dense communities would have several opportunities to reacquire gut bacteria lost as a result of molting by coprophagy. As these experimental conditions were meant to severely limit exposure to microbes present in frass, further work is planned to determine minimum frass exposure frequency for nymphs to achieve WT growth and development.

An additional explanation for the observed results is that the absence of gut microbes in

GF insects may have limited the host’s access to dietary nutrients that are typically liberated by members of the gut microbiota, and thus prolonged their development due to inadequate access to nutrients. Cockroaches and other insects that consume a diet comprised primarily of plant biomass that is rich in recalcitrant polysaccharides like

pectin, cellulose, and hemicellulose have a characteristically enlarged hindgut colonized with diverse microbes that serves as an anaerobic digester for these biopolymers (Bignell,

1977, 2016; Breznak, 1982; Kane & Breznak, 1991; Kane, 1997; Bell, Roth & Nalepa,

2007; Watanabe & Tokuda, 2010). Exposure to conspecific frass partially recovered WT levels of gut tissue development and instar duration, which may indicate the colonization of some endemic taxa that facilitate dietary nutrient accessibility. It was notable that despite little size and weight difference at equivalent instar, the gastrointestinal tract developed differentially in GF and Conv insects, especially in the hindgut where the highest numbers of bacteria have been found. The hindgut is a major site of microbial activity along the cockroach alimentary tract, with the greatest numbers of bacteria consistently identified in this compartment through methods as diverse as microscopy, cultivation, and 16S sequencing (Bignell, 1977; Bracke, Cruden & Markovetz, 1979;

Schauer, Thompson & Brune, 2012; Schauer et al., 2014). As such, it is unsurprising to find significant divergence in growth phenotype between GF and either WT or Conv insects in this compartment in the absence of microbial influence. Additionally, the microbial and biochemical composition of the frass is likely more similar to that of the hindgut than the midgut, which may explain why the remediative value of the frass was more pronounced in these tissues.

Gut microbes have been shown to mediate nutritional and immune factors and influence physiological systems (Engel & Moran, 2013), and numerous avenues to microbial promotion of host growth have been identified in other model organisms, and are likely functioning in P. americana. Efforts to link gut microbiota to host health and

development have been spurred by the tremendous progress being made to detail the membership and composition of these communities. Model systems, especially germ-free animals, provide invaluable platforms for linking microbes to specific host outcomes. P. americana is a valuable addition to available invertebrate model systems because it harbors species of many bacterial taxonomic groups typically found in some vertebrates, including humans, is trophically omnivorous, and can be reared germ-free without antibiotics and, given this study, responds positively when exposed to conspecific frass.

The relatively low rearing costs and high fecundity add to the amenability of the GF P. americana model system, which further ensures that many of the questions raised by results obtained in this study can easily be experimentally pursued. Further efforts to scrutinize how conspecific frass and its biochemical and microbial components are linked to host growth and development are underway and may reveal details of host-microbe interactions that may be generalizable to other animals.

Chapter 4. Microbial Colonization Promotes Epithelial Expansion in Gut Subcompartments of Periplaneta americana.

ABSTRACT

The omnivorous cockroach Periplaneta americana relies upon a complex gut microbiome comprised of hundreds of bacterial taxa for normal gut tissue development.

The gut microbiota is partially transferred among nestmates by coprophagy and transfer by this route affects hindgut length and overall gut morphology. Specific sites of symbiotic action within the gut have not been examined and little is known about which of the cockroach gut microbiota promote normal development. This study reports an examination the histological morphology of the cockroach midgut and hindgut epithelial and visceral muscle layers to examine subsites within the gut whose development is most stimulated by microbial action. The posterior midgut and anterior hindgut were most affected by microbial colonization, with significantly increased cross-sectional area, and increased gut perimeter compared to non-colonized insects. The anterior midgut was not significantly affected by microbial colonization and the posterior hindgut only demonstrated increased cross-sectional area but no significant difference in perimeter or muscle thickness measures. This work homes in on gut sites where mutualistic bacteria are likely to reside and will facilitate the identification of microbial taxa with important contributions to the host, as the gut microbial communities at these sites are more closely examined in future work.

INTRODUCTION

The interface of the gut epithelial lining and gut lumen is where an organism and its microbiota are in close proximity and intimately engaged in the exchange of metabolites that include nutrients (Colombani et al., 2003; Macfarlane, Woodmansey, George, &

Macfarlane, 2005; Martens, Chiang, & Gordon, 2008; Storelli et al., 2011; Tailford,

Crost, Kavanaugh, & Juge, 2015; H. Zheng, Powell, Steele, Dietrich, & Moran, 2017), signaling molecules (Capo, Charroux, & Royet, 2016; Lee et al., 2015), and virulence factors (Ashida, Ogawa, Kim, Mimuro, & Sasakawa, 2012; Juge, 2012). Microbial stimulation of gut tissue development has been documented in numerous animal systems and attributed to nutrient provisioning and stress response (Bracke et al., 1978a; Donohoe et al., 2011; Jahnes et al., 2019; Jones et al., 2013; Patel, Maldera, & Edgar, 2013). In the

American cockroach, Periplaneta americana, microbial colonization of the gut following coprophagy induced increased growth in hindgut tissues in fifth-instar insects compared to germ-free insects of the same instar (Jahnes et al., 2019). Mutualistic gut microbiota are hypothesized to promote gut development through several nutritional mechanisms that include break-down of recalcitrant (non-host available) substrates (Cruden &

Markovetz, 1979), essential vitamin and amino acid provisioning (A. C. Wong, Dobson,

& Douglas, 2014), providing host-assimilable metabolic waste products, like short-chain fatty acids, (Donohoe et al., 2011; Ludek Zurek & Keddie, 1996), and by being directly digested by the host (Yamada et al., 2015). For example, growth rate of Drosophila raised under nutrient-limited conditions could be rescued by dietary supplementation with the fungus Issatchenkia orientalis, which promoted the uptake of radiolabeled amino

acids from the diet (Yamada et al., 2015), and Lactobacillus plantarum, which was shown to promote growth by activating amino acid-sensing cascades (Storelli et al.,

2011). Likewise, production of acetic acid by Acetobacter pomorum was essential for host growth promotion (Shin et al., 2011).

Compared to a condition in which bacteria are present in the gut, a germ-free condition may represent a state of undernutrition if the net host-nutrient-harvest is greater in the presence of a mutualistic gut microbiota that may be present under wild-type or conventionalized conditions. Starvation and undernutrition can decrease intestinal stem cell division, DNA replication (Park & Takeda, 2008), and increase apoptosis within midgut nidi (sites of rapid stem cell division and differentiation) and lead to reduced midgut length in P. americana (Park, Park, & Takeda, 2009; Park & Takeda, 2008).

Furthermore, reduced turgidity and lateral folds, likely a consequence of reduced muscularity and epithelial development, were observed in P. americana hindguts that lack bacterial colonization (Jahnes et al., 2019). A mechanistic explanation for reduced visceral muscularity was uncovered in Drosophila where suppressed energy sensing was observed to lead to reduced visceral muscle thickness and activity (Bland, Lee,

Magallanes, Foskett, & Birnbaum, 2010).

In contrast, certain gut microbiota could lead to a net decrease in host nutrient harvest because of competitive uptake of host assimilable nutrients by the gut microbial community, in which case we would expect the germ-free state to be one of comparatively increased growth and development. Conversely, from previously

mentioned work (Bracke et al., 1978a; Jahnes et al., 2019; Jones et al., 2013; Shin et al.,

2011; Storelli et al., 2011; Yamada et al., 2015) we see an increase in growth and development in microbially colonized individuals, so we suspect that the host bacterial mutualism occurs more often than competition.

We expect that histological changes in gut morphology will reflect the developmental trends in gross morphology seen in previous work, in that epithelial proliferation may lead to greater gut biomass in regions of microbial impact. This study seeks to scrutinize the cockroach digestive tract through histological examination of the gut according to subcompartment to identify those sites where decreased proliferative activity may lead to the shortened midgut and hindgut phenotypes observed in previous work under conditions of undernutrition and in the absence of microbial colonization.

MATERIALS AND METHODS

Insects. P. americana nymphs, adults and ootheca were obtained from a live collection maintained in the Insectary in the Ohio State University Biological Sciences Greenhouse

(Columbus, Ohio).

Ootheca Treatment. Ootheca were manually detached from gravid P. americana females. Ootheca that were directly attached to females, in a fully or near fully sclerotized state were preferentially selected. As such, they are relatively clean of debris and damage, and will hatch within approximately 30 days. Ootheca were surface sterilized via mechanical and chemical means including a detergent scrub (1% Alconox detergent), antiseptic solution (0.08% sodium hypochlorite), and enzymatic lysis buffer 82

(lysozyme 10mg/ml, EDTA, Tris-HCl and Triton X-100), each wash punctuated by three rinses with sterile MilliQ water (MQW) to remove antiseptic solution and induce intermittent hypo-osmotic shock on surface-clinging bacteria (Schwinghamer, 1980;

Salema et al., 1982). An aliquot of each third MQW rinse was reserved to monitor effectiveness of previous cleanings to reduce bacterial load (see Quality Control). Aseptic ootheca were incubated individually in sterile microcentrifuge tubes, capped with sterile cotton, at 30°C for approximately 30 days until hatching, which yielded up to 16 germ- free nymphs per oothecum. 1st instar nymphs from each oothecum were divided into cohorts of 3 or 4 insects divided across treatments as follows.

Germ-free P. americana. Germ-free (GF) 1st instar nymphs were aseptically-transferred to sterile rearing chambers stocked with gamma-irradiated rat chow and aseptic 1% agar, sufficient for 60+ days of feeding, and chambers were ventilated with air filtered through a 0.22 µm membrane.

Conventionalized P. americana. To introduce bacteria native to P. americana, GF 1st instars were exposed to frass taken from a lab-maintained colony of nonsterile or ‘wild- type’ (WT) P. americana mixed in a 1:10 ratio with irradiated rat food, exposed for 7 days. Subsequently, these conventionalized (Con) 1st instar nymphs were housed in sterilized rearing chambers (but sterility was not maintained), and fresh gamma-irradiated rat chow was provided on a weekly basis as required to limit growth of spoilage organisms. As each generation of P. americana typically acquires their gut microbial community through conspecific coprophagy (Nalepa et al. 2001; Bell et al. 2008), the

conventionalization approach reflects this route of gut microbe acquisition, while maintaining insects in habitats equivalent to those of GF insects.

Wild-type P. americana. One day old 1st instar insects from 10 surface sterilized ootheca were deposited in an aquarium containing 10 adult male cockroaches from a nonsterile mixed generation colony maintained in the lab and provided with gamma-irradiated rat chow and access to MQW ad libitum; nymphs from this colony were designated ‘wild- type’ (WT). WT hatchlings were free to interact with adult cockroaches and their frass, which encourages normal coprophagic behavior, that leads to the acquisition of normal gut microbiota. Late-stage nymphs from the nonsterile colony were sacrificed and deposited in the WT colony, as cannibalism and necrophagy of deceased nestmates is also a putative mechanism for gut microbiota acquisition.

Quality Control. Quality control measures were employed throughout experiments to ensure maintenance of GF status and to confirm robust colonization of Con individuals.

Oothecum rinse water, collected between each antiseptic wash, was plated in triplicate on

Luria-Bertani agar (LBA) plates (incubated at 31C for up to 1 week) to detect cultivable contaminants flushed from ootheca, and to provide confidence that ootheca are aseptic as they enter incubation. Ootheca shedding no contaminants at final rinses were considered to be aseptic and used for subsequent experiments. At the time of hatching and installation into habitats, the empty oothecum was homogenized in sterile 1% PBS and supernatant was plated on LBA in triplicate to confirm aseptic status. At the termination of insect growth, and just prior to insect dissection, frass was collected from rearing chambers, suspended in 1% PBS, and plated to confirm that habitats remained germ-free

throughout the duration of the experiment. During methodological development, diagnostic PCR with universal 16S primers 27F (5’-AGAGTTTGATCMTGGCTCAG-

3’) and 1391R (5’-GACGGGCGGTGTGTRCA-3’) was performed on DNA extracted from frass homogenates to detect non-cultivable contaminants of the gut or habitat.

Additionally, microscopic examination of DAPI-stained frass, gut contents, and gut thin- sections were performed to further evaluate the effectiveness of the germ-free, and conventionalization protocols.

Dissection. Insects were dissected in sterile PBS to extract full length gastrointestinal tracts (guts). Full length guts were linearized by teasing away tracheoles connecting adjacent folds of the gut and were allowed to relax in a droplet of PBS for photography.

Images documenting the longitudinal extent of guts adjacent to a millimeter scale were acquired.

Agar Embedding. 3% agarose was prepared, autoclaved and reserved for embedding guts. After photos had been acquired of the gut, molten agarose was pipetted into a wax mold after which dissected guts were laid across the molten agar surface and allowed to sink into the agar, while maintained in a linear orientation. More molten agar was pipetted over the surface of the gut to encase the gut near the center of the agar block.

Fixation. Carnoy solution was prepared as follows: ethanol, chloroform, and acetic acid were mixed 6:3:1 and placed into 10 ml vials. Agar-embedded guts were inserted into

vials and placed on a rocker for 24 hours after which time fixative was replaced with 70%

EtOH and samples were stored under refrigeration.

Dehydration and Embedding. Agar embedded guts were drained of 70% ethanol and placed in plastic cages. Agar was trimmed to allow freedom of movement in plastic cages. Cages were placed in 70% ethanol bath within an automatic tissue processor.

Solvents and soak times included 1hr in 70% EtOH, 1hr in 80% EtOH, 1hr in 90% EtOH,

1hr in 95% EtOH, 1hr in 100% EtOH 1, 1hr in 100% EtOH 2, 1 hr in 50:50

EtOH/Xylene, 1hr in 100% Xylene, 4 hrs in 100% Xylene, 20 min in 30% periplast dissolved in xylene, 20 min in 100% periplast @ 60 ˚C, 20 min in 100% periplast @ 60

˚C.

Blocking. Because of the long and skinny structure of the embedded guts, and the need to section them transversely, conventional blocking molds were unsuitable. Custom molds were fashioned from 10ml syringes and mounted onto wooden blocks suitable for clamping to a microtome. Agar embedded guts were pierced through the agar with a fine insect mounting needle at an approximately 45 degree angle and lowered into the block/mold assemblies, so that the gut was centered in the cylinder of the syringe bore

(Figure 17). The mold was topped with molten

Figure 17 Blocking

A 10 ml syringe was trimmed and affixed to a wood block using beeswax to serve as a mold.

wax with a heated syringe and large bore needle (16 ga). After

the wax has hardened, the mold was loosened and the block

was trimmed.

Sectioning. Ribbons were cut at 10 micron thickness, with the gut sectioned from anal cavity to proventriculus. Sections were then floated on a 50 ˚C water bath, and affixed to slides. One slide per ribbon was made, with about 20-30 slides per gut. Slides were placed on a hot plate for approximately 10 min and then moved to a histological slide dryer for 1 hour.

Staining. Slides were stored for several weeks and then dewaxed and rehydrated through a xylene, ethanol, and water series. The series was as follows: 100% xylene, 100% xylene

50:50 xylene/ethanol, 100% ethanol, 100% ethanol, 90% ethanol, 80% ethanol, 70% ethanol, 50% ethanol, distilled H2O. Soak times were 10 min, 10 min, 5 min, 1 min for all ethanol solutions, and 2 min for the water soak. Rehydrated slides were stained according to Masson’s trichrome method, with a 7.5 min soak in 50 ˚C azocarmine g in

1% acetic acid, water rinse, 30 sec in 0.1% aniline in EtOH, 2 min in acidified EtOH 87

(0.1% HCL), 2 hrs in 5% phosphotungstic acid, and 3 hrs in 2% Orange G/0.5% Analine

Blue in 7.5% aqueous acetic acid. Slides were rinsed in distilled H2O for approximately

10 seconds, until blue dye dissipated from the agar portion of the section and then slides were dehydrated. The dehydration series consisted of 10 sec in 100% ethanol, 10 sec in

100% ethanol, 10 sec in 50:50 xylene/ethanol, 10 sec in 100% xylene, 10 sec in 100% xylene. Slides were mounted and coverslipped with Cytoseal 60.

Imaging. Images were acquired on an Olympus bright field microscope with an Omax camera at 100x total magnification for cross-sectional area and circumference measurements and at 400x total magnification for muscle measurements. Five cross- sections were imaged per gut region per individual for a total of 20 measurements per treatment per gut region for cross-sectional area and circumference measurements. For muscle measurements, the same five cross sections per individual were imaged at higher magnification, capturing 4 micrographs per section, one from each quadrant, for a total of

80 images per treatment per gut region.

Measurements. Measurements were conducted in Image J using a custom macro to threshold images, create masks, and record measurements in a semi-automated fashion.

Pixel measures were converted to um and um2 units. Photos containing digestate and bacterial biomass directly adjacent to the lumen required manual demarcation of the luminal boundary to ensure accurate thresholding. Thresholded images were converted to masks of total gut cross-sectional area and luminal area, the difference of which

constituted the epithelial cross-sectional area. Perimeter measurements were obtained from the luminal mask and the total cross-sectional area mask. For muscle measurements, grid overlays were projected over micrographs to direct measurements in a random fashion to avoid measurement bias, as muscle thickness within a sample was often heterogeneous, and two measurements were taken per image.

Statistics. Because multiple measures per individual are not independent, two approaches were utilized to avoid pseudoreplication. The conservative approach used the average of measures from each individual within a treatment and gut region and evaluated the averaged data using a conventional ANOVA. Additionally, a linear mixed model was employed, including a random factor for individual to explain variation associated with multiple measures within each insect. P-values presented in the body of this work are derived from the former, more conservative approach and p-values from both analyses are presented in the supplementary section.

RESULTS

Epithelial Cross-Sectional Area. The cross-sectional area of the cockroach gut represents the quantity of epithelial and visceral-muscle biomass. The cross-sectional area of thin-sections from gut epithelial tissue was compared across three bacterial treatments within 4 different cockroach gut subcompartments (Figure 18, A-D), including the anterior midgut (AMG), posterior midgut (PMG), anterior hindgut (AHG) and posterior hindgut (PHG). Cross-sectional area within the AMG was not significantly different across treatments, while the cross-sectional area was significantly greater in the 89

WT (958 µm2) compared to GF (518 µm2) PMG (ANOVA: WT-GF p=0.010 ) and AHG

(ANOVA: WT-GF p=0.018 ) with cross-sectional area of Con (740 µm2) insects intermediate to WT and GF in the same subcompartments, but not significantly different from either. The WT PHG cross-sectional area (1128 µm2) was significantly greater than that of both the Con (777 µm2) and the GF (536 µm2) treatments (ANOVA: WT-Con p=0. 038, WT-GF p=0.002).

Gut Perimeter. The exterior perimeter of the cockroach gut reflects the degree of lateral expansion of the gut and is proportional to the overall gut diameter. The exterior perimeter of thin sections from gut epithelial tissue was compared across the same bacterial treatments and gut subcompartments as previously examined (Figure 18, E-H).

Significant differences in the length of the exterior gut perimeter were not observed in the

AMG or PHG, but were seen in the PMG and AHG. Specifically, the exterior perimeter of the PMG was greater in WT (238 µm) insects compared to GF (190 µm) insects

(ANOVA: WT-GF p=0.004). The exterior perimeter of the AHG was greater in WT (342

µm) insects compared to GF (206 µm) insects and greater in Con (312 µm) insects compared to GF (206 µm) insects (ANOVA: WT-GF p=0.002, Con-GF p=0.007).

Luminal Perimeter. The interior perimeter of thin sections from gut epithelial tissue was compared across the same bacterial treatments and gut subcompartments (Figure 18, I-

L). Significant differences in the length of the luminal perimeter were not observed in the

AMG or PHG, but were seen in the PMG and AHG. The luminal perimeter of the PMG

was greater in WT (355 µm) insects compared to GF (208 µm) insects (ANOVA: WT-

GF p=0.002). The luminal perimeter of the AHG was greater in Con (465 µm) insects compared to GF (337 µm) insects (ANOVA: Con-GF p=0.040).

Ratio of Gut Perimeter to Luminal Perimeter. The ratio of gut perimeter to luminal perimeter represents the degree to which the lumen is invaginated in the hindgut and reflects the number and depth of crypts in the midgut. This ratio was calculated and compared across treatments and gut compartments (Figure 18, M-P). Significant differences in the ratio of gut perimeter to luminal perimeter were not observed in the

AMG or PHG, but were seen in the PMG and AHG. The ratio of the PMG was lower in

WT (0.70) insects compared to GF (0.92) insects (ANOVA: WT-GF p=0.013). In contrast, the ratio of the AHG was higher in WT (0.79) insects compared to GF (0.62) insects (ANOVA: WT-GF p=0.023).

Visceral Muscle Thickness. The thickness of visceral muscle was measured at eight random points around the periphery of each gut section and compared across treatments and gut subcompartments (Figure 18, Q-T). Significant differences between treatments were only seen in the AHG where the muscle thickness in WT (7.9 µm) insects was significantly greater than in the GF (6.4 µm) insects (ANOVA: WT-GF p=0.020) and muscle thickness in Con (7.0 µm) insects was intermediate to WT and GF, but not significantly different. Muscle thickness in the PHG followed the same trend of

increasing thickness from GF (5.8 µm) to Con (6.7 µm) and WT (7.1 µm), but was not significantly different

Figure 18 Histological Measurements.

Cockroach guts were sectioned at 4 positions along the gut, anterior midgut

(AMG), posterior midgut (PMG), anterior hindgut (AHG), and posterior hindgut (PHG). Morphological metrics were quantified at each gut region across 3 treatments, germ-free (GF), conventionalized (Con), and wild-type

(WT). Morphological metrics included cross sectional area, gut perimeter, lumen perimeter, the ratio of gut to luminal perimeter and visceral muscle thickness. Statistical comparisons reflect the results of an ANOVA and post hoc

Tukey HSD test. Significant differences are concentrated in the PMG and

AHG, with some significant differences in the PHG.

Averaged measures vs linear mixed model. Multiple measures from the same insect and gut compartment were either averaged before statistical analysis or a linear mixed model was applied to compensate for multiple measures. Though p-values were lower using the linear mixed model, only two changes in significance were seen between the two methods (Table 2). In the PMG gut perimeter analysis the WT – Con comparison becomes significantly different (ANOVA: WT-Con p=0.043), as well as the PMG lumen perimeter WT-Con comparison (ANOVA: WT-Con p=0.021).

Table 2 p-value Table Histological Measurements

Comparisons of significant differences for pair-wise comparisons of treatments within each gut compartment and for each gut morphological parameter.

DISCUSSION

The PMG and AHG subcompartments are those most influenced by microbial colonization of the gut, with significant differences across all examined histological features consistently observed within these two sites (Table 2). Meanwhile, the PHG shows greater cross-sectional area under the influence of WT microbial colonization but no significant changes in perimeter or muscle thickness. The three treatments, GF, Con, 95

and WT likely reflect states of increasing bacterial diversity. Only obligate fat body endosymbionts are found in the GF insects, a community of intermediate diversity/abundance in insects conventionalized through coprophagy, and maximum diversity found in WT insects. Corresponding with this gradient of microbiome diversity, a near linear trend was frequently observed in relation to histological phenotypes with the

Con treatment measurements intermediate in magnitude to GF and WT treatments

(Figure 18). This same trend was identified in measures of gross morphology of the P. americana body and gut (Jahnes et al., 2019). A study of gut microbial community composition across the 3 treatments would help support the basis for this phenotypic gradient.

Studies of cockroach gut microbial abundance and diversity identify communities of lowest microbial abundance and Shannon diversity in the proximal gut (crop: 2.2x106 mg-1, H=0.51; midgut: 2.0x106 mg-1, H=0.48), with abundance and diversity peaking in the hindgut (2.2x107 mg-1, H=1.55) and dropping in the rectum (1.5x107 mg-1, H=1.13)

(Schauer et al., 2012). Known influences on microbial community composition within the insect gut include physiochemical gradients (Tegtmeier et al., 2016), substrate composition (Schauer et al., 2012), and host immune controls (S.-H. Kim & Lee, 2014; J.

Ryu et al., 2008; Xiao et al., 2017). Studies of digestion in the cockroach identify major sites of digestive enzyme secretion (Sakai, Satake, & Takeda, 2006), hydrolysis

(Oyebanji et al., 2014; Tamaki et al., 2014), and nutrient absorption (Treherne, 1957,

1958, 1967), and highlight the gastric cecae and anterior midgut as major sites of enzyme

secretion and nutrient uptake. In Drosophila, substantial overlap occurs between sites of innate immune activity and nutrient absorption, in the region of the cardia, gastric cecae and anterior midgut (Buchon et al., 2013), leading to the possibility that the host actively suppresses microbial populations at locations within the gut where microbial activity may compete with host assimilation of nutrients. Corresponding with this hypothesis, antimicrobial peptides and DUOX expression are elevated within the anterior midgut

(Buchon et al., 2013) while negative regulators of innate immunity are expressed in posterior regions of the Drosophila midgut (J. Ryu et al., 2008) where enzymatic and absorptive activity is decreased (Oyebanji et al., 2014; Tamaki et al., 2014; Treherne,

1967), lifting constraints on the proliferation of gut microbiota precisely as the digestate is depleted of host assimilable substrates. Correspondingly, a cultivation-based study revealed greater microbial abundance in the posterior midgut (108 ml-1) compared to the anterior midgut (107 ml-1) and three to eight-fold greater abundance within the anterior hindgut (108 ml-1) than the posterior hindgut (Bignell, 1977b). The pattern of microbial abundance in the midgut may reflect the activity of the host innate immune response in this compartment, while in the hindgut, diminishing abundance of bacteria in the posterior hindgut may reflect depletion of the digestate, a suggestion which is further supported by the high abundance of sporulating bacteria in the posterior hindgut (Cruden,

D.L. Markovetz, 1987), evidence of increasingly adverse conditions within this subcompartment.

Considering these patterns of microbial abundance and host digestion, a model explaining differential tissue morphology seen in cockroach gut histology across compartments and treatments may be proposed (Figure 19). With active suppression of gut microbiota by

Figure 19 Conceptual Model of Cockroach Gut Dynamics

Regions of host nutrient assimilation and host innate immunity overlap at the anterior of the gut while bacterial abundance and stimulation of host epithelial proliferation overlap in posterior regions of the gut.

the host innate immune system in the anterior midgut and host dominated digestion, neither bacterial competition with the host nor bacterial enhancement of digestion are taking place, and no significant effect on host anterior midgut development is seen. With host innate immune suppression in the posterior midgut and reduced host mediated digestion, bacterial proliferation and bacterial mediated digestion is enabled in this subcompartment, yielding significantly increased tissue development in WT individuals. 98

Likewise, in the anterior hindgut further bacterial proliferation and fermentative activity may yield host assimilable short chain fatty acids and drive epithelial tissue proliferation in WT individuals. In the posterior hindgut, with the likely exhaustion of microbially assimilable substrate, reduced microbial metabolism and bacterial decline yields fewer host assimilable substrates, and only results in significant differences in insect tissue proliferation across contrasting bacterial treatments in the case of cross-sectional area.

Meanwhile, the bacterial community in conventionalized individuals frequently leads to elevated tissue development that is intermediate to that of WT individuals and GF individuals, but not always significantly different from GF or WT. This suggests that a stimulatory microbial community is present in the Con treatment, but is incomplete compared to the WT community, with some microbial taxa or microbial abundance patterns skewed in a way that prohibits WT levels of development.

Examining microbial taxa common to the Con and WT treatments may uncover candidates responsible for stimulating host epithelial development. Likewise, identifying and examining microbiota at specific sites of enhanced epithelial expansion, such as in the PMG and AHG, and taxa common to both sites may help uncover host/microbial mutualisms.

Chapter 5. Recruitment, persistence, and mutualism in the gut microbial community of Periplaneta americana.

ABSTRACT

Members of the Bacteroides are common mammalian gut commensals and symbionts.

Here we show that four host adapted Bacteroides from the cockroach Periplaneta americana share numerous factors that have been identified as being necessary for host colonization and competition within the mammalian gut, suggesting that factors for host colonization are conserved across mammals and some insects. Introduction of four host- adapted Bacteroides to germ-free cockroaches revealed that not all four isolates could efficiently colonize a naïve gut environment. Genomic evidence suggests that effectively colonizing isolates code an increased number of genes for coping with oxidative stress and experimental evidence suggests that a consistently colonizing isolate demonstrates higher resistance to oxidative stress. Introduction of this group of Bacteroides to germ-free P. americana did not affect maturation rate but did reduce hindgut length of the cockroach.

INTRODUCTION

As an environment, the gut poses unique challenges and benefits to microbes dwelling within. While the gut environment is constantly moderated by the host in terms of pH, temperature, and osmotic potential, microbes have to contend with fluctuating redox potential, outflow of digestate, and host innate immune defenses. Studies within mammalian systems have begun to unravel the genetic mechanisms for host gut

100

colonization, but it is unclear whether parallel mechanisms are required for colonization of the insect gut.

The gut of the omnivorous cockroach, Periplaneta americana, provides a moderated environment with diverse substrates which foster the growth of hundreds of commensal and symbiotic gut microorganisms (Bertino-Grimaldi et al., 2013; Sabree & Moran, 2014;

Tinker & Ottesen, 2016; J. Zhang et al., 2016). While the microbial community within the cockroach gut has begun to be characterized taxonomically, little is known about the role of individual taxa in dietary digestion, development of the host, and gut microbial community ecology. Cultivation efforts have yielded isolates of diverse taxonomy from the P. americana gut and metabolomic measurements from insects and enrichment cultures indicate diverse metabolisms such as polysaccharide hydrolysis (Bignell, 1977a), acetogenesis (M. D. Kane & Breznak, 1991), methanogenesis (Gijzen & Barugahare,

1992), and sulfate reduction (Cruden et al., 1979), but few isolates have been characterized exhaustively. Recent work detailed the isolation and genome sequencing of 11

Bacteroidetes from the gut of P. americana, revealing a broad ability to degrade plant structural polysaccharides (Vera-Ponce de León et al., 2020).

Among the 11 Bacteroidetes described by Vera-Ponce de León et al. (2020) was a group of 4 Bacteroides that formed a monophyletic clade among other cockroach derived

Bacteroides indicating a probable long-term association with cockroaches. This monophyly and probable long-term association indicate that these isolates are likely strongly host-adapted. The combination of cultivable isolates with sequenced genomes and a tractable germ-free and gnotobiotic system allows us to ask questions about details of

101

bacterial genetics that support host colonization and the roles of members of the cockroach gut microbial community in cockroach health and development. Examining the genomes of these isolates in comparison to a closely related mammalian symbiont genome may provide clues as to common or diverging genetic mechanisms for host gut colonization.

Additionally, introduction of these Bacteroides to the germ-free cockroach, to form an artificial gnotobiotic gut community, may indicate the nature of their interaction with the host; whether it is commensal, mutualistic, or parasitic.

MATERIALS AND METHODS

PABacteroides Phylogenetics. A locus of high genomic conservation was identified across all 4 PAB isolates and one non-cockroach outgroup, Bacteroides fragilis, which consisted of 28 ribosomal protein genes and 6 other conserved protein genes in perfect synteny. Sequences were aligned using ClustalW, positions containing gaps were stripped, and a maximum likelihood tree was constructed using PhyML.

Genomic Features Correlating with Colonization Success. Sequenced genomes are available for all PAB isolates through Genbank (PAB224: QVMK00000000, PAB51:

QVMI00000000, PAB519: QVMJ00000000, and PAB214: QVML00000000). Genes from pathways and protein complexes associated with known host colonization factors were parsed from the genome assemblies and completeness of pathways were compared across PAB isolate genomes. Genes of interest included Rhodobacter nitrogen fixation

(rnf), sodium dependent NADH dehydrogenase (nqr), capsular polysaccharide biosynthesis (CPS), locus of Fragilis glycosylation (lfg), and vitamin B12 transporter genes (btuB). Other genes not shown to be essential for host colonization, but possibly 102

beneficial to colonization success were parsed, including reactive oxygen/nitrogen species (ROS) tolerance genes, oxygen tolerance genes, and respiratory genes.

Insects. P. americana nymphs, adults and ootheca were obtained from a live collection maintained in the Insectary in the Ohio State University Biological Sciences Greenhouse

(Columbus, Ohio).

Ootheca Treatment. Ootheca were manually detached from gravid P. americana females. It is prudent to collect ootheca that are directly attached to females, in a fully or nearly fully sclerotized state. As such, they are relatively clean of debris and damage, and will hatch within approximately 30 days. Ootheca were surface sterilized via mechanical and chemical means including a detergent scrub (1% Alconox detergent), antiseptic solution (0.08% sodium hypochlorite), and enzymatic lysis buffer ( lysozyme 10mg/ml,

EDTA, Tris-HCl and Triton X-100), each wash punctuated by three rinses with sterile

MilliQ water (MQW) to remove antiseptic solution and induce intermittent hypo-osmotic shock on surface-clinging bacteria (Schwinghamer, 1980; Salema et al., 1982). An aliquot of each third MQW rinse was reserved to monitor effectiveness of previous cleanings to reduce bacterial load (see Quality Control). Aseptic ootheca were incubated individually in sterile microcentrifuge tubes, capped with sterile cotton, at 31°C for approximately 30 days until hatching, which yielded up to 16 germ-free nymphs per oothecum. 1st instar nymphs from each oothecum were divided into cohorts of 3 or 4 insects and treated as follows.

103

Germ-free and Gnotobiotic P. americana. Germ-free (GF) 1st instar nymphs were aseptically-transferred to sterile rearing chambers stocked with gamma-irradiated rat chow and aseptic 1% agar, sufficient for 60+ days of feeding, and chambers were ventilated with air filtered through a 0.22 µm membrane. Bacterial isolates recovered from P. americana gut homogenates were reintroduced to GF 1st instars to render gnotobiotic insects with defined gut microbial communities. Isolates were cultivated under anaerobic conditions in modified tryptone yeast glucose (MTYG) liquid medium.

Culture identity and purity were confirmed prior to inoculation by PCR with isolate specific primers and dilution plating to confirm consistent colony morphology. Overnight cultures, approaching late log phase, were normalized to equivalent optical density (OD).

Aliquots were collected in an anaerobic chamber (Coy Laboratory Products, Inc.) and centrifuged in a benchtop centrifuge to pellet cells. Two thirds of supernatant was removed and cells were resuspended. Cultures were combined in equal ratio, and ~ 20 µl was drawn into 50 µl capillary tubes, and topped with mineral oil to prevent air intrusion from the upper end. GF insects, deprived of water, were exposed to culture mixes by inserting capillary tubes through habitat wall. Insects were denied access to water in order to enforce consumption of culture and visibly fed on culture within minutes of provisioning. Culture was renewed daily for 4 days, with late log phase culture on days 1 and 3, and stationary phase culture on days 2 and 4. After capillary removal, habitat penetrations were sealed with cyanoacrylate and nymphs were provided access to 1% agar for hydration. Nymphs were dissected at mid 5th instar.

104

Wild-type P. americana. One day old 1st instar insects from 10 surface sterilized ootheca were deposited in an aquarium containing 10 adult male cockroaches from a nonsterile mixed generation colony maintained in the lab and provided with gamma-irradiated rat chow and access to MQW ad libitum; nymphs from this colony were designated ‘wild- type’ (WT). WT hatchlings were free to interact with adult cockroaches and their frass, which encourages normal coprophagic behavior, that leads to the acquisition of normal gut microbiota. As cannibalism of deceased nestmates is also a putative mechanism for gut microbiota acquisition, late-stage nymphs from the nonsterile colony were sacrificed and deposited in the WT colony.

Quality Control. Quality control measures were employed throughout experiments to ensure maintenance of GF status and to confirm robust colonization of Conv individuals.

Oothecum rinse water, collected between each antiseptic wash, was plated in triplicate on

Luria-Bertani agar (LBA) plates (incubated at 31 ˚C for up to 1 week) to detect cultivable contaminants flushed from ootheca, and to provide confidence that ootheca are aseptic as they enter incubation. Ootheca shedding no contaminants at final rinses were considered to be aseptic and used for subsequent experiments. At the time of hatching and installation into habitats, the empty oothecum was homogenized in sterile 1% PBS and supernatant was plated on LBA in triplicate to confirm aseptic status. At the termination of insect growth, and just prior to insect dissection, frass was collected from rearing chambers, suspended in 1% PBS, and plated to confirm that habitats remained germ-free throughout the duration of the experiment. During methodological development, diagnostic PCR with universal 16S primers 27F (5’-AGAGTTTGATCMTGGCTCAG-

105

3’) and 1391R (5’-GACGGGCGGTGTGTRCA-3’) was performed on DNA extracted from frass homogenates to detect non-cultivable contaminants of the gut or habitat.

Dissection and DNA Extraction. Insects were dissected aseptically in sterile PBS to recover the full digestive tract. Full length guts reserved for DNA extraction were placed in sterile PBS and DNA was extracted with a Qiagen Blood and Tissue kit following the manufacturers protocol.

Monogeneric in-vitro Coculture. One hundred microliters of normalized inoculum of the four PAB isolates, combined in equal proportions, was inoculated into anaerobic

Balch tubes containing 5 ml of vitamin supplemented MTYG medium and aliquots were collected at 24 and 48h. Quantitative PCR was conducted as previously outlined to enumerate isolate populations in coculture and confirm viability of inoculum. Output culture from cockroach inoculations, which was exposed to ambient conditions within a glass 1mm capillary for 24h, was also inoculated into Balch tubes under the same conditions, with aliquots collected at the same frequency.

Isolate Quantification. Absolute qPCR or RTqPCR was conducted using isolate specific primers (Table 4). Isolate specific primers were searched against the Arb Silva 16S database and against an in house 16S database of cockroach gut isolates to ensure specificity. Isolates specific primers were further tested against closely related species within an in-house isolate collection to ensure amplification specificity. All isolates

106

maintain one 16S gene. Absolute qPCR quantification was conducted on an Ependorff

Mastercycler Realplex II thermocycler using a plasmid reference with a 192 bp 16S amplicon insert from isolate Fusobacterium PAF510 with plasmid copies at 4 dilutions, between 109-103 copies per μL, to generate a standard curve against which unknowns were compared. Unknowns were amplified with corresponding isolate specific primers designed to generate ~200 bp amplicon and two technical replicates were conducted for each biological replicate. Thermocycler conditions were as follows: 3 min 95˚ C followed by 35 cycles: 1 min 95˚ C, 15s 58˚ C, 40s 68˚ C, 10s signal acquisition at 78˚ C and a final melt curve analysis was performed to differentiate target amplification from primer dimer. Absolute RTqPCR utilized a one-step mastermix (Thermo Scientific Verso) and thermocycler conditions differed with only the addition of an initial reverse transcriptase incubation at 50˚ C for 15 min followed by a denaturing step at 95˚ C for 15 min before initiation of the previously mentioned cycling steps and melt curve analysis. 16S count data was normalized by insect body mass (mg).

Microaerobic vs Anaerobic growth. Ninety-six well microtiter plates were prepared in an anaerobic chamber. Wells were filled with 200 μL anaerobic MTYG medium and inoculated with 10 μL of over-night culture, in quadruplicate. Control wells contained reduced resazurin oxygen indicator (hydroresorufin). Plates were covered with polypropylene film and were incubated in a Fluostar Omega spectrophotometer in

-1 ambient atmosphere or with nitrogen purge gas flowing at 1 L min . OD600 readings were acquired every hour for 72h. Oxidation controls incubated in ambient atmosphere became

107

red over the course of numerous hours (hydroresorufin → resorufin), indicating oxygen infiltration through plates or film.

Peroxide Susceptibility. Fifty μL of overnight culture was spread on MTYG plates

(without cysteine) in triplicate. A 6-mm filter disk was saturated with 10 μL of 3% H2O2, placed in the center of the plate and immediately incubated anaerobically at 30°C. Zones of clearing were evaluated after 48 and 96 h.

RESULTS

PABacteroides Phylogenetics. A locus of high genomic conservation was identified across all 4 PAB isolates and one non-cockroach outgroup, Bacteroides fragilis, which consisted of 28 ribosomal protein genes and 6 other conserved protein genes in perfect synteny. With B. fragilis rooting the tree, the branching pattern followed the order

PAB51, PAB224, PAB519, and PAB214 (Figure 20).

Figure 20 PAB Isolate Phylogenetics

Maximum likelihood tree of four P. americana gut Bacteroides isolates with Bacteroides fragilis as an outgroup. Alignment created from a single contiguous locus encompassing 28 ribosomal proteins and 6 other conserved proteins. The scale represents the number of substitutions per site.

108

Potential Bacterial Genomic Factors Facilitating Host Colonization and

Competition Within the Gut.

Capsular polysaccharide biosynthesis. Capsular polysaccharide biosynthesis genes are present in all isolates and localized in distinct operons in most PABacteroides isolates (Figure 21). PAB519 codes the most capsular polysaccharide (CPS) operons, including four clusters of genes enriched with glycosyl transferases which are all preceded by a diagnostic transcription antitermination protein (UpdY), that is present in vertebrate gut

109

Figure 21 Capsular Polysaccharide Loci

Capsular Polysaccharide loci from the four P. americana gut Bacteroides isolates are represented, with diagnostic UpdY genes followed by concentrations of glycosyl hydrolase genes. UpdY genes are orange, flippases are purple, glycosyl transferases are yellow-green, dehydratases are dark-blue, dehydrogenases are light-blue, miscellaneous CPS genes are yellow, and hypothetical genes are gray.

110

111

Bacteroides (Chatzidaki-Livanis, Coyne, & Comstock, 2009). PAB51 codes a single contiguous CPS operon with some scattered loci coding CPS-associated genes, likely related to protein glycosylation. PAB224 codes a single contiguous CPS operon while

PAB214 codes a single fragmentary CPS operon without an UpdY antitermination gene at the start of the operon. Though capsular polysaccharide operons are common within

PAB isolates, the gene identity and synteny is not conserved with those CPS operons of B. fragilis or other closely related Bacteroides (Figure 37). CPS biosynthesis was demonstrated to be a factor required for host colonization (Michael J Coyne,

Chatzidaki-Livanis, Paoletti, & Comstock, 2008; Goodman et al., 2010) and a single CPS locus was shown to be sufficient for competitive colonization of the gut (Michael J

Coyne et al., 2008). One possible mechanism of colonization and competition competence conferred by capsular polysaccharide may be increased bacterial resistance towards host antimicrobial peptides (Campos et al., 2004).

In B. fragilis, regulation of 7 of 8 CPS operons is under control of invertible promoters, mediated by a site-specific recombinase (M. J. Coyne, Weinacht, Krinos, & Comstock,

2003; Krinos et al., 2001). Isolates PAB51, PAB224, and PAB519 contain the conserved promoter consensus sequence upstream of UpdY, within the CPS operon. The promoter sequence is evident in only 3 of 4 CPS operons in PAB519, while inverted repeats identical to those found in B. fragilis (M. J. Coyne et al., 2003) were present, flanking promoters, in isolates PAB51 and PAB519. Two of the three CPS promoters within

PAB519 were flanked by inverted repeats while the promoter in PAB224 was not flanked by inverted repeats. This suggests that in PAB51 and PAB519, control of CPS expression

112

follows the same mechanism as vertebrate dwelling Bacteroides, but the lack of inverted repeats in isolates PAB224 and PAB214 and the lack of a promoter sequence in PAB214 suggests that variation has developed in the regulation of the CPS locus in this cockroach dwelling Bacteroides clade.

All PABacteroides isolates code an operon conserved within the Bacteroides for protein glycosylation (lfg), which has been demonstrated to confer a competitive advantage for growth within the mouse gut and is essential for synthesis of fucosylated glycoproteins

(Michael J. Coyne, Reinap, Lee, & Comstock, 2005; Fletcher, Coyne, Villa, Chatzidaki-

Livanis, & Comstock, 2009). The synteny of the lfg locus is conserved across previously examined Bacteroidetes (Michael J. Coyne et al., 2013) and three of the four

PABacteroides isolates (PAB51, PAB224, and PAB214), with methionine tRNA- synthetase (MetG) and flippase (WzxC) followed by glycosyl transferases, however in

PAB519 the flippase and glycosyl transferase coding locus is separated and distant from

MetG, but still adjacent to genes with conserved synteny downstream of the locus including the tRNA-Met gene (Figure 22). This is an interesting departure by isolate

PAB519 considering that the synteny of this locus is conserved across numerous other taxa within the Bacteroidetes phylum (Michael J. Coyne et al., 2013), and this locus is co-transcribed with the metG gene, hinting at its importance for survival (Fletcher et al.,

2009). A cluster of inverted repeats adjacent to the glycosylation locus in PAB519 may indicate that a phase variable promoter controls the transcription of the operon, like in

CPS gene regulation. Separately on the

113

Figure 22 Conserved Protein Glycosylation Locus

A locus for protein glycosylation, conserved across Bacteroides, is represented with synteny conserved adjacent to the methionine synthase gene for all but PAB519. Flippases are purple, glycosyl transferases are green, isomerases are dark-blue, miscellaneous glycosylation genes are yellow, hypothetical genes are gray, and unrelated genes are black.

114

115

chromosome, PABacteroides isolates maintain GDP-mannose dehydratase (Gmd) and

GDP-L-fucose synthetase genes that are required for bacterial synthesis of GDP-L-

Fucose, while PAB519 maintains fucose kinase, which allows the incorporation of exogenous fucose into glycans (Michael J. Coyne et al., 2005). Because fucose is common in glycans of host epithelial surfaces it is suggested that incorporation of fucose into bacterial glycans may serve as a form of molecular mimicry that may prevent elicitation of host defenses by bacterial glycans (Michael J Coyne, Reinap, Lee, &

Comstock, 2005). Conservation of pathways for glycan fucosylation in the PAB

Bacteroides suggests that this strategy may be active in the insect gut, like in the mammalian gut.

An additional, previously undocumented, putative protein glycosylation operon is located within the isolates PAB224, PAB51, and PAB214 with a conserved location within the genome, adjacent to NqrA-F, with conserved synteny and identity of the first 5 genes and last 4 genes across all loci (Figure 23) The same locus is present in

B. fragilis, with conservation of synteny and identity within the first four genes

(BT_1163-BT_1166), but is absent in isolate PAB519. The operon in PAB224 and

PAB214 both contain flippase homologs as well as polysaccharide polymerase homologs required to constitute a functional protein glycosylation apparatus. This

116

operon is entirely lacking in PAB519 and the operon in PAB51 is reduced in length and lacking a flippase (Figure 23).

Figure 23 New Putative Protein Glycosylation Locus

A new putative locus for protein glycosylation, conserved across Bacteroides, is represented with synteny conserved adjacent to the NADH dehydrogenase Nqr locus for all but PAB519. Nqr genes are red, flippases are purple, glycosyl transferases are green, polysaccharide polymerase is pink, miscellaneous glycosylation genes are yellow, hypothetical genes are gray, and unrelated genes are black.

117

118

Oxidative Stress. Colon dwelling Bacteroides have evolved within the low oxygen environment of the gut and are more susceptible to oxidative stress than aerobes. PAB51 maintains the greatest number of genes associated with oxidative stress response, followed by PAB224, PAB519, and then PAB214 (Table 3). Isolates PAB51 and

Table 3 Oxidative Stress Response Genes

PABacteroides isolates code varying numbers of gene homologs for oxidative stress response. White – no homolog, light grey – one homolog, dark grey – two homologs, black – three homologs.

PAB224 code alkyl hydroperoxide reductase (AhpCF) while isolates PAB519 and

PAB214 do not. Isolate PAB51 has two duplications of ahpC (for a total of 3 copies), while in both isolates ahpF is truncated to include only the N-terminal domain, and lacks the pyridine nucleotide-disulfide oxidoreductase domain. However, both of these isolates contain independent thiol peroxidases (thx) which are homologous to the C-terminal deletions and may be able to work in concert with the truncated AhpF to activate AhpC, as has been demonstrated in biochemically dissected AhpF (Poole, Godzik, Nayeem, &

Schmitt, 2000).

119

PAB51 and PAB224 code an additional peroxiredoxin, related to bacterioferritin co- migratory protein (bcp), which has shown activity towards organic peroxides (Atack,

Harvey, Jones, & Kelly, 2008).

Isolates PAB51 and PAB519 both contain organic hydroperoxide resistance protein and regulator (OhrBR) while PAB224 and PAB214 do not. All isolates code rubredoxin, rubrerythrin, superoxide dismutase, catalase, and a thiol peroxidase, while isolate PAB51 codes an extra copy of rubredoxin and PAB224 codes a second superoxide dismutase homolog. All isolates contain the Bacteroides aerotolerance operon (batI) which has been shown to be important in B. fragilis host infectivity in rodent models (Tang, Dallas, &

Malamy, 1999). In contrast the PABacteroides isolate genomes don’t contain a close homolog of the oxygen protective gene Dps, found in B. fragilis (Betteken, Rocha, &

Smith, 2015). Isolates PAB51 and PAB224 code ferritin (FtnA) while isolates PAB519 and PAB214 code bacterioferritin (Bfr). These iron sequestering proteins have been implicated in a protective response to oxidative conditions (Rocha & Smith, 2004, 2013).

Isolate PAB51 is unique in having 3 homologs of ferritin compared to the single copies retained by the other isolates. All PAB isolates code catalase genes and at least one homolog of thioredoxin, thioredoxin-1, or thioredoxin C-1. PAB51 and PAB519 code peptide methionine sulfoxide reductase (MsrA), which reduces oxidized methionine

(Cabiscol, Tamarit, & Ros, 2000). PAB51 is unique in encoding glutathione synthetase.

Together isolate PAB51 codes 25 genes associated with ROS/RNS and oxidative stress response, PAB224 codes 18 genes from the same categories, while PAB519 codes 15

120

genes and PAB214 codes 11, not including genes associated with respiratory oxygen consumption.

Antimicrobial Peptide Resistance. Isolate PAB519 codes an elevated number of phosphatidylglycerophosphatase enzymes that are homologous to Bacteroides thetaiotaiomicron LpxF, which is responsible for resistance towards the cationic antimicrobial peptide polymyxin B and competitiveness in the mouse gut through cleavage of lipopolysaccharide phosphate groups and reduction of negative membrane charge (Cullen et al., 2015). While isolate PAB519 codes 4 homologs of LpxF, isolate

PAB224 and PAB214 code two homologs and isolates PAB51 codes a single homolog.

Vitamin B12. All Isolates code a single homolog of the Ton-B dependent vitamin B12 outer membrane transporter (BtuB) regulated by a B12 riboswitch. This gene and its neighbor (BtuG) are homologous and syntenic with the Bacteroides thetaiotaomicron

BtuB 1 locus (BT_1489, BT_1490), which was demonstrated to provide the least competitive advantage of the three B. thetaiotaomicron btuBG loci (Degnan, Barry, Mok,

Taga, & Goodman, 2014a). Isolate PAB51 codes the vitamin B12 ABC transporter

(BtuFCD) under regulation by a vitamin B12 riboswitch, which is adjacent to its cobalamin biosynthesis operon that can complete synthesis of cobalamin from the precursor hydroxymethylbilane, under control of an independent B12 riboswitch. Isolates

PAB519, PAB214, and PAB224 lack cobalamin biosynthesis operons, and maintain limited salvage pathways. This suggests that PAB51 may be the most competitive isolate in a vitamin B12 limiting environment, as it can likely assimilate many more B12 precursors than the other isolates, and may be a more effective scavenger of vitamin B12

121

pathway intermediates, but may not compete well with other Bacteroides with more efficient BtuBG uptake machinery.

Energy Metabolism. All isolates harbor components of the electron transport chain that enable microaerophilic respiration. Of the PABacteroides isolates, complete NADH dehydrogenase complex one machinery (Nuo A-N) genes were only identified in PAB51 and PAB224. Isolate PAB519 is missing genes for subunits Nuo E-G, while isolate

PAB214 maintains genes for only subunits Nuo E-G (though conserved adjacent genes are at the end of a contig, indicating that assemblies may be incomplete). All isolates maintain sodium translocating NADH dehydrogenase (Nqr A-F). Nqr was shown to be an essential factor for host colonization in B. thetaiotaomicron (Goodman 2009), and may function in sodium ion homeostasis in addition to energy generation. Nqr A-F is common among pathogenic microorganisms and relatively rare in environmental microorganisms, suggesting host-association may be a driver for retention of this operon.

All PABacteroides isolates contain cytochrome bd oxidase and succinate dehydrogenase enabling microaerobic and anaerobic respiration.

All PABacteroides isolates maintan NrfAH nitrite reductase which can function as a mechanism for anaerobic respiration or detoxifying host ROS. At the same time, only isolates PAB51, PAB224, and PAB519 code a cytoplasmic nitric oxide flavorubredoxin reductase, which functions exclusively in the nitrosative stress response (Costa, Teixeira,

& Saraiva, 2003; Gardner, Helmick, & Gardner, 2002), while PAB51 has duplicate copies.

122

Isolate PAB214 is unique in coding an operon for lysine degradation to butyrate via an energy conserving ferredoxin oxidoreductase.

Monogeneric in-vitro Coculture. Compared to individual PABacteroides isolate abundance at inoculation (~106 16S gene copies per isolate), increases were seen in the abundance of all isolates at 24 and 48h (107-109 16S gene copies per isolate), with isolate

PAB51 and PAB214 constituting the greatest proportion of the population in combined liquid culture (Figure 38). After exposure to ambient conditions within a glass capillary tube for 24h, during cockroach inoculation, bacteria enumerated in the inoculum (Figure

39) decreased slightly (104-105 16S gene copies per isolate), and after 48 hours of incubation in fresh anaerobic medium, the abundance of all isolates increased with isolate

PAB51 dominating to constitute the greatest proportion of the population in combined liquid culture(109 16S gene copies), while isolate PAB214 lagged behind all other isolates (107 16S gene copies).

Monogeneric Gnotobiotic Polyculture. The 4 Bacteroides isolates were introduced in equal proportions to germ free cockroaches, and whole guts were sampled at mid fifth instar (~40 days). Isolate PAB51 was consistently detected in all sampled guts, averaging

1.1x106 16S copies mg-1 body mass (Figure 24). Isolate PAB224 was detected in 4 of 13

123

guts averaging 1.4x107 16S copies mg-1 body mass (Figure 24). Neither PAB214 nor

PAB519 were detected.

Figure 24 Gnotobiotic Colonization by Monogeneric Inoculation

Quantification of Bacteroides isolate abundance by qPCR within gnotobiotic cockroach gut samples. Isolate PAB51 is present in all sampled cockroaches, while isolate PAB224 is present in 4 of 13 cockroach guts.

Cockroach Growth Monogeneric Polyculture. Cockroaches were inoculated with 4

PAB isolates and growth parameters were measured at 5th instar. Duration until 5th instar was not significantly different between germ-free cockroaches and the monogeneric gnotobiotic roaches (Figure 25). Meanwhile, the hindgut was significantly reduced in length (Wilcox p=0.0041) in the monogeneric gnotobiotic cockroaches (mean 4.07mm) compared to germ free cockroaches (mean 4.86mm) (Figure 25). No other significant

124

differences were observed between morphological parameters among treatments (Figure

25).

Figure 25 Cockroach growth parameters.

Age at molt to 5th instar was not significantly different for germ-free cockroaches from those inoculated with the four Bacteroides isolates. Hindgut length was significantly shorter in cockroaches inoculated with the four Bacteroides isolates compared to germ-free insects.

Anaerobic and Microaerobic Growth Kinetics. Given that strongly colonizing isolate

PAB51 possesses more oxygen tolerance genes, growth was compared under anaerobic and microaerobic conditions. Under microaerobic conditions optical density of isolates

PAB224, PAB214, and PAB519 peaks at 10 hours and then diminishes abruptly (Figure

26). This corresponds with the time it takes for resazurin control wells to become oxidized. Isolate PAB214 peaks at 0.12 OD600, PAB224 peaks at 0.10 OD600, and

PAB519 peaks at 0.06 OD600 (Figure 26). Optical density of isolate PAB51 steadily increases until 72h, peaking at 0.14 OD600 (Figure 26). Under a nitrogen purged atmosphere isolates PAB214 and PAB51 grow continuously, with increasing OD600 to 72 125

hours, reaching 0.26 and 0.34 OD600 respectively (Figure 26). Isolate PAB224 peaks at

63 h at 0.35 OD600, while isolate PAB519 peaks after 53 h at 0.12 OD600 (Figure 26).

Figure 26 Growth of Bacteroides isolates in anaerobic conditions vs microaerophilic conditions.

Compared to fully anaerobic growth, isolates PAB224, PAB214, and PAB519 decline in absorbance after 10 hours when medium begins to become oxidized whereas isolate PAB510 continues to increase in absorbance until the end of the incubation.

Peroxide Sensitivity. Peroxide is a host innate immune factor that has been demonstrated to control the growth of gut commensal microorganisms in Drosophila (S.-H. Kim &

126

Lee, 2014; Xiao et al., 2017). Peroxide tolerance followed the trend PAB51 > PAB224 =

PAB214 = PAB519, with average zones of inhibition of 25.8 mm, 40.3 mm, 43.0 mm, and 40.7 mm, with significant differences (ANOVA, p=1.9x10-11) between PAB51 and all other PAB isolates (Figure 27).

Figure 27 Peroxide susceptibility.

Susceptibility to peroxide is significantly greater in PAB224, PAB214, and PAB519 than in PAB51, evidenced by greater zones of clearing induced in bacterial lawns in a disc diffusion assay.

127

DISCUSSION

A number of studies have examined genetic factors within Bacteroides necessary for host colonization and competitiveness in the mammalian gut microbiome. These factors include Rhodobacter nitrogen fixation (Rnf), sodium dependent NADH dehydrogenase

(Nqr), capsular polysaccharide synthesis (CPS), vitamin B12 transporter (BtuB) and

Locus for Fragilis Glycosylation (lfg) (Comstock, 2009; Michael J Coyne et al., 2008,

2005; Goodman et al., 2010). Capsular polysaccharide functions in a protective manner by masking bacteria from the host when host-mimicking sugars such as sialic acid and fucose are incorporated at terminal positions in the capsular polysaccharide chain

(Comstock, 2009; Michael J Coyne et al., 2005). Additionally, capsular polysaccharide composition can alter the charge of the bacterial envelope repelling cationic antimicrobial peptides (AMPs) or simply impeding the interaction of AMPs with the membrane

(Breazeale, Ribeiro, McClerren, & Raetz, 2005; Campos et al., 2004; Cullen et al., 2015;

Zhou et al., 2001). While Nqr is essential for host colonization in Bacteroides thetaiotaiomicron, only slight defects are reported in growth in vitro (Goodman et al.,

2010). The prevalence of Nqr in pathogens and marine organisms, and paucity of the complex in other environmental isolates suggests that it may be necessary for sodium ion homeostasis in the sodium rich marine environment or host tissue (Barquera, 2014).

Vitamin B12 is essential for the numerous microbiota with over 86% of 11,000 studied organisms coding vitamin B12 dependent enzymes (Shelton et al., 2019). Furthermore,

75% of 311 gut taxa studied encoded Vitamin B12 dependent enzymes (Degnan et al.,

2014a). At the same time, only 37% of studied taxa had predictive marker genes for B12

128

biosynthesis (Shelton et al., 2019). This suggests a wide reliance on vitamin B12 salvage and transport to fulfill the requirement for this vitamin. At least one BtuB vitamin B12 transporter was required for competitive growth with a wildtype B. thetaiotaiomicron in vitro and BtuB transposon mutants were outcompeted by other Bacteroidetes in the mouse gut (Degnan et al., 2014a; Goodman et al., 2009). No in-vitro growth defect was seen in Rnf mutants, and it is unclear why the gene is essential for host colonization

(Goodman et al., 2010). As identified here, the PABacteroides isolates code many of the genes that have been identified as essential for host colonization within mammalian systems, suggesting that the requirements for colonizing the insect alimentary canal are similar.

All PAB isolates code genes in loci that are consistent with CPS production, though isolates including PAB224 and PAB214 only code single loci. A single CPS locus was sufficient for competitive colonization of the mouse gut by B. fragilis (Michael J Coyne et al., 2008). Loci responsible for glycosylation of proteins are present in all isolates and may be responsible for molecular mimicry of host fucosylated glycans as in B. fragilis

(Michael J. Coyne et al., 2005). In our work, a previously undescribed putative glycosylation locus was found adjacent to the Nqr NADH dehydrogenase in B. fragilis and all PAB genomes except PAB519. Future work may reveal whether this locus provides a competitive advantage in the gut like the other conserved glycosylation loci.

Interestingly, PAB519 codes several CPS loci that may be under the control of invertible promoters, providing the potential for capsular phase variability and the change of surface architecture depending on environment, like in B. fragilis. The concomitant

129

proliferation of CPS loci in PAB519 and lack of conservation in other protein glycosylation loci indicates that PAB519 may have evolved an alternate surface glycosylation strategy compared to the other PAB isolates and Bacteroidetes.

Few gut Bacteroides code complete vitamin B12 biosynthesis pathways, yet they often rely on B12 dependent methionine synthase (MetH). The importance of this requirement is demonstrated by the complete rescue of B12 transport mutant growth defects by supplementation with methionine (Degnan, Barry, Mok, Taga, & Goodman, 2014b;

Wexler et al., 2018). Strong competition within the gut environment for vitamin B12 reinforces the requirement for BtuB to compete and persist in this niche (Degnan et al.,

2014b). While all of the PABacteroides code a BtuB transporter, the more complete B12 biosynthesis pathway of isolate PAB51 may allow it to better compete in a B12 limited environment such as the cockroach gut.

As a host innate immune factor, peroxide production is controlled by DUOX and the downregulation of this gene leads to the over-proliferation of gut microbiota in

Drosophila (S.-H. Kim & Lee, 2014; Xiao et al., 2017). Gut commensals and symbionts may require defenses against host peroxide to persist in the gut niche. Isolate PAB51 displays increased peroxide resistance which may be due to duplications of ahpC and higher net basal expression of peroxidase enzyme that may result from several transcribed loci. PAB51s resistance to oxidative/nitrosative stress may allow it to colonize more closely to the host epithelium where innate defense factors are present and prior to other gut colonizers.

130

Early in the colonization of the gut, oxygen may be more abundant, and over time oxygen concentrations may decrease with increasing bacterial colonization. Higher oxygen concentrations are seen in germ free cockroaches compared to normally colonized individuals and monocolonized individuals, and insects monocolonized with a single aerobe had reduced gut oxygen concentrations compared to a normally colonized gut

(Tegtmeier et al., 2016). Pioneering aerobic or aerotolerant bacteria may be the first colonizers of the cockroach gut, drawing oxygen tensions down enough to allow increasingly oxygen sensitive members of the gut community to gain a foothold in the gut. Increased resistance to oxygen exposure in isolate PAB51 may be due to the duplications of rubredoxin and ferritin identified in the genome and increased basal expression of these proteins. Possession of at least one copy of each of these genes may increase oxygen tolerance of both PAB51 and PAB224 compared to PAB214 and

PAB519 which lack both genes, but possess single copies of bacterioferritin. The presence of oxidative stress and nitrosative stress response genes may reflect where the isolates live within the gut habitat and/or temporal ordering of colonization. Oxygen partial pressure is elevated at the luminal/epithelial interface, and PAB51 would better resist this stress compared to the other isolates.

When germ-free cockroaches were inoculated with four closely related and cockroach adapted PABacteroides only one isolate (PAB51) consistently colonized, while a second

(PAB224) occasionally colonized. In vitro coculture of PABacteroides indicates that there is not a strong competitive exclusion of any one isolate through competition.

However, coculture of PABacteroides in vitro after oxygen exposure demonstrated that

131

PAB51 proliferated to greater abundance after this stress, indicating that it may be more resistant to ambient oxygen and that environmental conditions may play a role in the colonization efficiency of the PAB isolates. Genomic evidence of elevated gene dosage related to the oxidative stress response provides some indication that PAB51 may be better equipped to cope with atmospheric oxygen and host-derived oxidants. Anaerobic vs aerobic cultivation studies demonstrate that while PAB224, PAB214, and PAB519 respond to oxygen exposure by reductions in optical density, after medium oxidation at

10hrs, PAB51 continues to increase in optical density despite medium oxidation. In addition, PAB51 was significantly more resistant to peroxide than the other PAB isolates.

These two factors are evidence of true phenotypic advantages of PAB51 in respect to oxidative stress, and may underly its success in colonizing an unpopulated gut. While in this case comparative genomics provides us some clues about what may allow PAB51 to be an effective gut pioneer, whether oxygen dynamics or ROS/RNS scavenging are truly factors remains to be determined and will require more work. Techniques such as a transposon insertion sequencing or site directed mutagenesis may help determine whether these genes are essential or beneficial factors for colonization of the cockroach gut.

Jahnes et al. 2019 reported longer hindguts in conventionalized and wild-type P. americana than germ-free P. americana. Though the hindguts were shorter under

PABacteroides colonization, it does not appear to have been detrimental to the host, as maturation rate was not significantly different than germ-free cockroaches. It appears that these isolates could be considered commensal, given their neutral effects on host maturation rate.

132

Conclusions and Future Perspectives

The literature review at the opening of this dissertation reveals that the cockroach gut microbiome is underexplored, with opportunities to uncover new eukaryotic and viral life forms in the cockroach gut and numerous microbe/microbe symbioses remain to be discovered.

The cockroach P. americana has been demonstrated here to be amenable to germ-free and gnotobiotic work, opening up the possibility of examining host microbe interactions between this cockroach and any of the diverse life forms that can be isolated from the cockroach gut microbiome.

Coprophagy is demonstrated within this work to be an important means of gut colonization that leads to organismal development that is intermediate to germ-free and wild-type insects. It remains to be seen how the gut microbial community of frass conventionalized insects compares in microbial abundance and diversity in relation to the wild-type community. A better understanding of the gut microbial taxa that are transferred by coprophagy may be achieved through 16S amplicon sequencing of frass conventionalized insects.

Histological identification of sites of microbiota-induced growth promotion highlights the posterior midgut and anterior hindgut as gut subregions that are subject to increased epithelial proliferation under microbial colonization. Efforts to detect and characterize host/microbial symbioses may be directed to these sites in order to identify gut microbial taxa that are of benefit to the host and detect transcriptional and metabolic responses to beneficial taxa. 133

While a group of four host-adapted Bacteroides were identified and genomic characterization revealed many genetic determinants for host colonization that are common to mammalian Bacteroides, two isolates fail to colonize a naïve gut environment and colonizing isolates fail to significantly improve organismal growth, suggesting that more complex dynamics are in play in regards to gut microbial community succession and bacterial symbiosis. Future work should aim to assemble a gut microbial community that better resembles the functional complexity found in that of a complex microbial community, as the feedback and synergy among numerous organisms may be required to constitute a beneficial gut community that colonizes the numerous niches of the gut effectively and leads to improved host development.

134

Bibliography

Aguilar, R., Maestro, J. L., Vilaplana, L., Pascual, N., Piulachs, M. D., & Bellés, X. (2003). Allatostatin gene expression in brain and midgut, and activity of synthetic allatostatins on feeding-related processes in the cockroach Blattella germanica. Regulatory Peptides, 115(3), 171–177. https://doi.org/10.1016/S0167- 0115(03)00165-4 Apidianakis, Y., & Rahme, L. (2011). Drosophila melanogaster as a model for human intestinal infection and pathology. Disease Models and Mechanisms, 59(4), 21–30. https://doi.org/10.1002/glia.21148 Appel, A. G., Sims, S. R., & Eva, M. J. (2008). Factors affecting coprophagy and necrophagy by the German cockroach (Dictyoptera: Blattellidae). 6th International Conference on Urban Pests, Budapest, Hungary, 13-16 July 2008, 139–142. Retrieved from https://www.cabdirect.org/cabdirect/abstract/20133307490 Appel, Arthur G, Sims, S. R., & Eva, M. J. (2008). Factors affecting coprophagy and necrophagy by the german cockroach (Dictyoptera: Blattellidae). Proceedings of the Sixth International Conference on Urban Pests, 12(Table 1), 139–142. Ashida, H., Ogawa, M., Kim, M., Mimuro, H., & Sasakawa, C. (2012). Bacteria and host interactions in the gut epithelial barrier. Nature Chemical Biology, 8(1), 36–45. https://doi.org/10.1038/nchembio.741 Atack, J. M., Harvey, P., Jones, M. A., & Kelly, D. J. (2008). The Campylobacter jejuni thiol peroxidases Tpx and Bcp both contribute to aerotolerance and peroxide- mediated stress resistance but have distinct substrate specificities. Journal of Bacteriology, 190(15), 5279–5290. https://doi.org/10.1128/JB.00100-08 Atkinson, W., & Shorrocks, B. (1977). Breeding site specificity in the domestic species of Drosophila. Oecologia, 29(3), 223–232. https://doi.org/10.1007/BF00345697 Ayali, A., Couzin-Fuchs, E., David, I., Gal, O., Holmes, P., & Knebel, D. (2015). Sensory feedback in cockroach locomotion: current knowledge and open questions. Journal of Comparative Physiology A, 201(9), 841–850. https://doi.org/10.1007/s00359-014-0968-1 Ayayee, P. A., Larsen, T., & Sabree, Z. (2016). Symbiotic essential amino acids provisioning in the American cockroach, Periplaneta americana (Linnaeus) under various dietary conditions. PeerJ, 4(April), e2046. https://doi.org/10.7717/peerj.2046 Ayayee, P. A., Ondrejech, A., Keeney, G., & Muñoz-Garcia, A. (2018). The role of gut microbiota in the regulation of standard metabolic rate in female Periplaneta americana . PeerJ, 6, e4717. https://doi.org/10.7717/peerj.4717 Ayayee, P., Jones, S., & Sabree, Z. (2015). Can 13C stable isotope analysis uncover 135

essential amino acid provisioning by termite-associated gut microbes? PeerJ, 3, e1218–e1218. https://doi.org/10.7717/PEERJ.1218 Barquera, B. (2014). The sodium pumping NADH:quinone oxidoreductase (Na+-NQR), a unique redox-driven ion pump. Journal of Bioenergetics and Biomembranes, 46(4), 289–298. https://doi.org/10.1007/s10863-014-9565-9 Baughn, A. D., & Malamy, M. H. (2004). The strict anaerobe Bacteroides fragilis grows in and benefits from nanomolar concentrations of oxygen. Nature, 427(6973), 441– 444. https://doi.org/10.1038/nature02285 Bell, W. J., & Adiyodi, K. G. (1982). The American cockroach. Physiological Entomology, 7(Accessed from http://nla.gov.au/nla.cat-vn1667079), 117–118. https://doi.org/10.1111/j.1365-3032.1982.tb00674.x Bendixen, E., Danielsen, M., Larsen, K., & Bendixen, C. (2010). Advances in porcine genomics and proteomicsc-a toolbox for developing the pig as a model organism for molecular biomedical research. Briefings in Functional Genomics and Proteomics, 9(3), 208–219. https://doi.org/10.1093/bfgp/elq004 Benjamino, J., Lincoln, S., Srivastava, R., & Graf, J. (2018). Low-abundant bacteria drive compositional changes in the gut microbiota after dietary alteration. Microbiome, 6(1), 86. https://doi.org/10.1186/s40168-018-0469-5 Benschoter, C. A., & Wrenn, R. T. (1972). Germfree techniques for establishment and maintenance of a colony of aseptic German cockroaches. Annals of the Entomological Society of America, 65(3), 641–644. Berchtold, M., Chatzinotas, A., Schönhuber, W., Brune, A., Amann, R., Hahn, D., & König, H. (1999). Differential enumeration and in situ localization of microorganisms in the hindgut of the lower termite Mastotermes darwiniensis by hybridization with rRNA-targeted probes. Archives of Microbiology, 172(6), 407– 416. https://doi.org/10.1007/s002030050778 Bergman, P., Seyedoleslami Esfahani, S., & Engström, Y. (2017). Drosophila as a Model for Human Diseases—Focus on Innate Immunity in Barrier Epithelia. Current Topics in Developmental Biology, 121, 29–81. https://doi.org/10.1016/BS.CTDB.2016.07.002 Bertino-Grimaldi, D., Medeiros, M. N., Vieira, R. P., Cardoso, A. M., Turque, A. S., Silveira, C. B., … Machado, E. A. (2013). Bacterial community composition shifts in the gut of Periplaneta americana fed on different lignocellulosic materials. SpringerPlus, 2(1), 609. https://doi.org/10.1186/2193-1801-2-609 Betteken, M. I., Rocha, E. R., & Smith, C. J. (2015). Dps and DpsL Mediate Survival In Vitro and In Vivo during the Prolonged Oxidative Stress Response in Bacteroides fragilis. Journal of Bacteriology, 197(20), 3329–3338. https://doi.org/10.1128/JB.00342-15 Bier, E., & Mcginnis, W. (2004). Model Organisms in the Study of Development and Disease. In R. P. Epstein,C. J., Erikson & and Wynsahw-Boris A. (Eds.), Molecular Basis of Inborn Errors of Development. Oxford University Press. Retrieved from http://www.expasy.ch/cgi-bin/lists?celegans.txt Biesterveld, S., Zehnder, A. J. B., & Stams, A. J. M. (1994). Regulation of product formation in Bacteroides xylanolyticus X5-1 by interspecies electron transfer. 136

Applied and Environmental Microbiology, 60(4), 1347–1352. Bignell, D. E. (1977a). An experimental study of cellulose and hemicellulose degradation in the alimentary canal of the American cockroach. Canadian Journal of Zoology, 55(Norkrans 1967), 579–589. https://doi.org/10.1139/z77-073 Bignell, D. E. (1977b). Some Observations on the Distribution of Gut Flora in the American Cockroach, Periplaneta americana. Journal of Invertebrate Pathology, 29(3), 338–343. https://doi.org/10.1016/S0022-2011(77)80040-2 Bland, M. L., Lee, R. J., Magallanes, J. M., Foskett, J. K., & Birnbaum, M. J. (2010). AMPK supports growth in Drosophila by regulating muscle activity and nutrient uptake in the gut. Developmental Biology, 344(1), 293–303. https://doi.org/10.1016/J.YDBIO.2010.05.010 Block, R. J., & Henry, S. M. (1961). Metabolism of the sulphur amino-acids and of sulphate in Blattella germanica. Nature, 191(4786), 392–393. https://doi.org/10.1038/191392a0 Blum, J. E., Fischer, C. N., Miles, J., & Handelsman, J. (2013). Frequent replenishment sustains the beneficial microbiome of Drosophila melanogaster. MBio, 4(6), 1–8. https://doi.org/10.1128/mBio.00860-13 Borgstrom, B., Dahlqvist, A., Lundh, G., & SJOVALL, J. (1957). Studies of intestinal digestion and absorption in the human. The Journal of Clinical Investigation, 36(10), 1521–1536. https://doi.org/10.1172/JCI103549 Bracke, J. W., Cruden, D. L., & Markovetz, a. J. (1979). Intestinal microbial flora of the of the american cockroach, Periplaneta americana L. Applied and Environmental Microbiology, 38(5), 945–955. Retrieved from http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=243614&tool=pmcentre z&rendertype=abstract Bracke, J. W., Cruden, D. L., & Markovetz, a J. (1978a). Effect of metronidazole on the intestinal microflora of the american cockroach, Periplaneta americana l. Antimicrobial Agents and Chemotherapy, 13(1), 115–120. Retrieved from http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=352193&tool=pmcentre z&rendertype=abstract Bracke, J. W., Cruden, D. L., & Markovetz, A. J. (1978b). Effect of metronidazole on the intestinal microflora of the american cockroach, Periplaneta americana l. Antimicrobial Agents and Chemotherapy, 13(1), 115–120. Retrieved from http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=352193&tool=pmcentre z&rendertype=abstract Bracke, J. W., & Markovetz, a. J. (1980). Transport of bacterial end products from the colon of Periplaneta americana. Journal of Insect Physiology, 26(2), 85–89. https://doi.org/10.1016/0022-1910(80)90047-5 Breazeale, S. D., Ribeiro, A. A., McClerren, A. L., & Raetz, C. R. H. (2005). A formyltransferase required for polymyxin resistance in Escherichia coli and the modification of lipid A with 4-Amino-4-deoxy-L-arabinose. Identification and function oF UDP-4-deoxy-4-formamido-L-arabinose. The Journal of Biological Chemistry, 280(14), 14154–14167. https://doi.org/10.1074/jbc.M414265200 Breznak, J. A. (1982). Intestinal Microbiota of Termites and other Xylophagous Insects. 137

Annual Review of Microbiology, 36(1), 323–323. https://doi.org/10.1146/annurev.mi.36.100182.001543 Breznak, J. A., & Switzer, J. M. (1986). Acetate synthesis from H2 plus CO2 by termite gut microbes. Applied and Environmental Microbiology, 52(4), 623–630. Broderick, Nichole A., Buchon, Nicolas, Lemaitre, B. (2014). Microbiota-Induced Changes in Drosophila melanogaster Host Gene Expression and Gut Morphology. https://doi.org/10.1128/mBio.01117-14 Broderick, N. A., & Lemaitre, B. (2012). Gut-associated microbes of Drosophila melanogaster. Gut Microbes, 3(4), 307–321. https://doi.org/10.4161/gmic.19896 Brooks, M. A., & Richards, A. G. (1955). Intracellular Symbiosis in Cockroaches. I. Production of Aposymbiotic Cockroaches. The Biological Bulletin, 109(1), 22–39. https://doi.org/10.2307/1538656 Brune, A, Emerson, D., & Breznak, J. A. (1995). The Termite Gut Microflora as an Oxygen Sink: Microelectrode Determination of Oxygen and pH Gradients in Guts of Lower and Higher Termites. Applied and Environmental Microbiology, 61(7), 2681–2687. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/16535076 Brune, Andreas. (2012). Endomicrobia: intracellular symbionts of termite gut flagellates. Journal of Endocytobiosis and Cell Research, 23, 11–15. Buchon, N., Broderick, N. A., Chakrabarti, S., & Lemaitre, B. (2009). Invasive and indigenous microbiota impact intestinal stem cell activity through multiple pathways in Drosophila. Genes and Development, 23(19), 2333–2344. https://doi.org/10.1101/gad.1827009 Buchon, N., Osman, D., David, F. P. A., Fang, H. Y., Boquete, J., & Deplancke, B. (2013). Resource Morphological and Molecular Characterization of Adult Midgut Compartmentalization in Drosophila. CellReports, 3(5), 1725–1738. https://doi.org/10.1016/j.celrep.2013.04.001 Burgess, N. R. H., McDermott, S. N., & Whiting, J. (1973). Aerobic bacteria occurring in the hind-gut of the cockroach, Blatta orientalis. Journal of Hygiene, 71(1), 1–8. https://doi.org/10.1017/S0022172400046155 Cabiscol, E., Tamarit, J., & Ros, J. (2000). Oxidative stress in bacteria and protein damage by reactive oxygen species. International Microbiology, 3(1), 3–8. https://doi.org/10.2436/im.v3i1.9235 Campos, M. A., Vargas, M. A., Regueiro, V., Llompart, C. M., Albertí, S., & José, A. (2004). Capsule Polysaccharide Mediates Bacterial Resistance to Antimicrobial Peptides Capsule Polysaccharide Mediates Bacterialjavascript:submitMSDSSearch(“US”,’’); Resistance to Antimicrobial Peptides, 72(12), 7107–7114. https://doi.org/10.1128/IAI.72.12.7107 Capo, F., Charroux, B., & Royet, J. (2016). Bacteria sensing mechanisms in Drosophila gut: Local and systemic consequences. Developmental and Comparative Immunology, 64, 11–21. https://doi.org/10.1016/j.dci.2016.01.001 Carrasco, P. (2014). Succession of the gut microbiota in the cockroach Blattella germanica. International Microbiology, 17, 99–109. https://doi.org/10.2436/20.1501.01.212 Carrasco, Purificación, Pérez-Cobas, A. E., van de Pol, C., Baixeras, J., Moya, A., & 138

Latorre, A. (2014). Succession of the gut microbiota in the cockroach Blattella germanica. International Microbiology : The Official Journal of the Spanish Society for Microbiology, 17(2), 99–109. https://doi.org/10.2436/20.1501.01.212 Cazemier, A. E., Hackstein, J. H. P., Op Den Camp, H. J. M., Rosenberg, J., & Van Der Drift, C. (1997). Bacteria in the intestinal tract of different species of arthropods. Microbial Ecology, 33(3), 189–197. https://doi.org/10.1007/s002489900021 Chatzidaki-Livanis, M., Coyne, M. J., & Comstock, L. E. (2009). A family of transcriptional antitermination factors necessary for synthesis of the capsular polysaccharides of Bacteroides fragilis. Journal of Bacteriology, 191(23), 7288– 7295. https://doi.org/10.1128/JB.00500-09 Clayton, R. B. (1959). A Simplified Method for the Culture of Blatella germanica under Aseptic Conditions. Nature, 184(4693), 1166–1167. https://doi.org/10.1038/1841166a0 Cleveland, L. R., & Grimstone, A. V. (1964). The fine structure of the flagellate Mixotricha paradoxa and its associated micro-organisms . Proceedings of the Royal Society of London. Series B. Biological Sciences, 159(977), 668–686. https://doi.org/10.1098/rspb.1964.0025 Colombani, J., Raisin, S., Pantalacci, S., Radimerski, T., Montagne, J., & Léopold, P. (2003). A nutrient sensor mechanism controls Drosophila growth. Cell, 114(6), 739– 749. https://doi.org/10.1016/S0092-8674(03)00713-X Comstock, L. E. (2009). Importance of Glycans to the Host-Bacteroides Mutualism in the Mammalian Intestine. Cell Host and Microbe, 5(6), 522–526. https://doi.org/10.1016/j.chom.2009.05.010 Corby-Harris, V., Pontaroli, A. C., Shimkets, L. J., Bennetzen, J. L., Habel, K. E., & Promislow, D. E. L. (2007). Geographical distribution and diversity of bacteria associated with natural populations of Drosophila melanogaster. Applied and Environmental Microbiology, 73(11), 3470–3479. https://doi.org/10.1128/AEM.02120-06 Costa, P. N., Teixeira, M., & Saraiva, L. M. (2003). Regulation of the flavorubredoxin nitric oxide reductase gene in Escherichia coli : nitrate repression, nitrite induction, and possible post-transcription control. FEMS Microbiology Letters, 218(2), 385– 393. https://doi.org/10.1016/S0378-1097(02)01186-2 Cox, C. R., & Gilmore, M. S. (2007). Native microbial colonization of Drosophila melanogaster and its use as a model of Enterococcus faecalis pathogenesis. Infection and Immunity, 75(4), 1565–1576. https://doi.org/10.1128/IAI.01496-06 Coyne, M. J., Weinacht, K. G., Krinos, C. M., & Comstock, L. E. (2003). Mpi recombinase globally modulates the surface architecture of a human commensal bacterium. Proceedings of the National Academy of Sciences, 100(18), 10446– 10451. https://doi.org/10.1073/pnas.1832655100 Coyne, Michael J., Fletcher, C. M., Chatzidaki-Livanis, M., Posch, G., Schaffer, C., & Comstock, L. E. (2013). Phylum-wide general protein O-glycosylation system of the Bacteroidetes. Molecular Microbiology, 88(4), 772–783. https://doi.org/10.1111/mmi.12220 Coyne, Michael J., Reinap, B., Lee, M. M., & Comstock, L. E. (2005). Human symbionts 139

use a host-like pathway for surface fucosylation. Science, 307(5716), 1778–1781. https://doi.org/10.1126/science.1106469 Coyne, Michael J, Chatzidaki-Livanis, M., Paoletti, L. C., & Comstock, L. E. (2008). Role of glycan synthesis in colonization of the mammalian gut by the bacterial symbiont Bacteroides fragilis. Proceedings of the National Academy of Sciences of the United States of America, 105(35), 13099–13104. https://doi.org/10.1073/pnas.0804220105 Coyne, Michael J, Reinap, B., Lee, M. M., & Comstock, L. E. (2005). Human symbionts use a host-like pathway for surface fucosylation. Science (New York, N.Y.), 307(5716), 1778–1781. https://doi.org/10.1126/science.1106469 Cruden, D.L. Markovetz, A. J. (1987). Microbial ecology of the cockcroach gut. Annual Review of Microbiology, 41(617–643). Cruden, D. L., Gorrell, T. E., & Markovetz, a J. (1979). Novel microbial and chemical components of a specific black-band region in the cockroach hindgut. Journal of Bacteriology, 140(2), 687–698. Retrieved from http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=216698&tool=pmcentre z&rendertype=abstract Cruden, D. L., & Markovetz, a J. (1984). Microbial aspects of the cockroach hindgut. Arch Microbiol, 138(2), 131–139. Retrieved from http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt= Citation&list_uids=6477029 Cruden, D. L., & Markovetz, A. J. (1979). Carboxymethyl cellulose decomposition by intestinal bacteria of cockroaches. Applied and Environmental Microbiology, 38(3), 369–372. Cullen, T. W., Schofield, W. B., Barry, N. A., Putnam, E. E., Rundell, E. A., Trent, M. S., … Goodman, A. L. (2015). Antimicrobial peptide resistance mediates resilience of prominent gut commensals during inflammation. Science, 347(6218), 170 LP – 175. https://doi.org/10.1126/science.1260580 Das, A., Silaghi-Dumitrescu, R., Ljungdahl, L. G., & Kurtz, D. M. (2005). Cytochrome bd oxidase, oxidative stress, and dioxygen tolerance of the strictly anaerobic bacterium Moorella thermoacetica. Journal of Bacteriology, 187(6), 2020–2029. https://doi.org/10.1128/JB.187.6.2020-2029.2005 Day, M. F., & Powning, R. F. (1949). A study of the processes of digestion in certain insects. Australian Journal of Biological Sciences, 2(2), 175–215. https://doi.org/10.1071/BI9490175 Degnan, P. H., Barry, N. A., Mok, K. C., Taga, M. E., & Goodman, A. L. (2014a). Human gut microbes use multiple transporters to distinguish vitamin B 12 analogs and compete in the gut. Cell Host and Microbe, 15(1), 47–57. https://doi.org/10.1016/j.chom.2013.12.007 Degnan, P. H., Barry, N. A., Mok, K. C., Taga, M. E., & Goodman, A. L. (2014b). Human gut microbes use multiple transporters to distinguish vitamin B12analogs and compete in the gut. Cell Host and Microbe, 15(1), 47–57. https://doi.org/10.1016/j.chom.2013.12.007 del Valle Rodríguez, A., Didiano, D., & Desplan, C. (2012). Power tools for gene 140

expression and clonal analysis in Drosophila. Nature Methods, 9(1), 47–55. https://doi.org/10.1038/nmeth.1800 Dietrich, C., Köhler, T., & Brune, A. (2014). The cockroach origin of the termite gut microbiota: patterns in bacterial community structure reflect major evolutionary events. Applied and Environmental Microbiology, 80(7), 2261–2269. https://doi.org/10.1128/AEM.04206-13 Dionne, M. S., & Schneider, D. S. (2008). Models of infectious diseases in the fruit fly Drosophila melanogaster. Disease Models & Mechanisms, 1(1), 43–49. https://doi.org/10.1242/dmm.000307 Donohoe, D. R., Garge, N., Zhang, X., Sun, W., O ’connell, T. M., Bunger, M. K., & Bultman, S. J. (2011). The Microbiome and Butyrate Regulate Energy Metabolism and Autophagy in the Mammalian Colon. Cell Metabolism, 13, 517. https://doi.org/10.1016/j.cmet.2011.02.018 Douglas, A. E. (2018). The Drosophila model for microbiome research. Lab Animal, 47(6), 157–164. https://doi.org/10.1038/s41684-018-0065-0 Dugas, J. E., Zurek, L., Paster, B. J., Keddie, B. A., & Leadbetter, E. R. (2001). Isolation and characterization of a Chryseobacterium strain from the gut of the American cockroach, Periplaneta americana. Archives of Microbiology, 175(4), 259–262. https://doi.org/10.1007/s002030000243 Engel, P., & Moran, N. A. (2013). The gut microbiota of insects - diversity in structure and function. FEMS Microbiology Reviews, 37(5), 699–735. https://doi.org/10.1111/1574-6976.12025 Erkosar, B., Kolly, S., Meer, J. R. Van Der, & Kawecki, T. J. (2017). Adaptation to chronic malnutrition leads to reduced dependence on microbiota in. Erkosar, B., & Leulier, F. (2014). Transient adult microbiota, gut homeostasis and longevity: Novel insights from the Drosophila model. FEBS Letters, 588(22), 4250– 4257. https://doi.org/10.1016/j.febslet.2014.06.041 Eutick, M. L., Veivers, P., O’Brien, R. W., & Slaytor, M. (1978). Dependence of the higher termite, Nasutitermes exitiosus and the lower termite, Coptotermes lacteus on their gut flora. Journal of Insect Physiology, 24(5), 363–368. https://doi.org/10.1016/0022-1910(78)90076-8 Feany, M. B., & Bender, W. W. (2000). A Drosophila model of Parkinson’s disease. Nature, 404(6776), 394–398. https://doi.org/10.1038/35006074 Fiebiger, U., Bereswill, S., & Heimesaat, M. M. (2016). Dissecting the interplay between intestinal microbiota and host immunity in health and disease: Lessons learned from germfree and gnotobiotic animal models. European Journal of Microbiology and Immunology, 6(4), 253–271. https://doi.org/10.1556/1886.2016.00036 Fletcher, C. M., Coyne, M. J., Villa, O. F., Chatzidaki-Livanis, M., & Comstock, L. E. (2009). A General O-Glycosylation System Important to the Physiology of a Major Human Intestinal Symbiont. Cell, 137(2), 321–331. https://doi.org/10.1016/j.cell.2009.02.041 Gardner, A. M., Helmick, R. A., & Gardner, P. R. (2002). Flavorubredoxin, an Inducible Catalyst for Nitric Oxide Reduction and Detoxification in Escherichia coli. Journal of Biological Chemistry, 277(10), 8172–8177. 141

https://doi.org/10.1074/JBC.M110471200 Giacomotto, J., & Ségalat, L. (2010). High-throughput screening and small animal models, where are we? British Journal of Pharmacology, 160(2), 204–216. https://doi.org/10.1111/j.1476-5381.2010.00725.x Gibson, G. R., Macfarlane, G. T., & Cummings, J. H. (1993). Sulphate reducing bacteria and hydrogen metabolism in the human large intestine. Gut, 34(4), 437–439. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/8491386 Gier, H. T. (1947a). Growth Rate in the Cockroach Periplaneta Americana (Linn). Annals of the Entomological Society of America, 40(2). Gier, H. T. (1947b). Intracellular Bacteroids in the Cockroach (Periplaneta americana Linn.). Journal of Bacteriology, 53(2), 173–189. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/16561260 Gijzen, H. J., & Barugahare, M. (1992). Contribution of anaerobic protozoa and methanogens to hindgut metabolic-activities of the American cockroach, Periplaneta americana. Applied and Environmental Microbiology, 58(8), 2565– 2570. Gijzen, H. J., van der Drift, C., Barugahare, M., & Op den Camp, H. J. (1994). Effect of Host Diet and Hindgut Microbial Composition on Cellulolytic Activity in the Hindgut of the American Cockroach, Periplaneta americana. Applied and Environmental Microbiology, 60(6), 1822–1826. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/16349275 Gilbert, L. I. (2008). Drosophila is an inclusive model for human diseases, growth and development. Molecular and Cellular Endocrinology, 293(1–2), 25–31. https://doi.org/10.1016/J.MCE.2008.02.009 Giuffrè, A., Borisov, V. B., Mastronicola, D., Sarti, P., & Forte, E. (2012). Cytochrome bd oxidase and nitric oxide: From reaction mechanisms to bacterial physiology. FEBS Letters, 586(5), 622–629. https://doi.org/10.1016/j.febslet.2011.07.035 Glavis-Bloom, J., Muhammed, M., & Mylonakis, E. (2012). Of Model Hosts and Man: Using Caenorhabditis elegans, Drosophila melanogaster and Galleria mellonella as Model Hosts for Infectious Disease Research (pp. 11–17). Springer, New York, NY. https://doi.org/10.1007/978-1-4419-5638-5_2 Goodman, A. L., Mcnulty, N. P., Zhao, Y., Leip, D., Mitra, R. D., Lozupone, C. A., … Gordon, J. I. (2010). Identifying genetic determinants neede to establish a human gut symbiont in its habitat. Cell, 6(3), 279–289. https://doi.org/10.1016/j.chom.2009.08.003.Identifying Goodman, A. L., McNulty, N. P., Zhao, Y., Leip, D., Mitra, R. D., Lozupone, C. A., … Gordon, J. I. (2009). Identifying genetic determinants needed to establish a human gut symbiont in its habitat. Cell Host & Microbe, 6(3), 279–289. https://doi.org/10.1016/j.chom.2009.08.003 Gordon, H. A., & Pesti, L. (1971). The gnotobiotic animal as a tool in the study of host microbial relationships. Bacteriological Reviews, 35(4), 390–429. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/4945725 Guerra, A., Etienne-Mesmin, L., Livrelli, V., Denis, S., Blanquet-Diot, S., & Alric, M. (2012). Relevance and challenges in modeling human gastric and small intestinal 142

digestion. Trends in Biotechnology, 30(11), 591–600. https://doi.org/10.1016/j.tibtech.2012.08.001 Guo, L., Karpac, J., Tran, S. L., & Jasper, H. (2014). PGRP-SC2 Promotes Gut Immune Homeostasis to Limit Commensal Dysbiosis and Extend Lifespan. Cell, 156(1–2), 109–122. https://doi.org/10.1016/J.CELL.2013.12.018 Ha, E. M., Lee, K. A., Seo, Y. Y., Kim, S. H., Lim, J. H., Oh, B. H., … Lee, W. J. (2009). Coordination of multiple dual oxidase-regulatory pathways in responses to commensal and infectious microbes in drosophila gut. Nature Immunology, 10(9), 949–957. https://doi.org/10.1038/ni.1765 Hampl, V. (2017). Preaxostyla. In Handbook of the Protists: Second Edition (pp. 1139– 1174). Springer International Publishing. https://doi.org/10.1007/978-3-319-28149- 0_8 Harrison, M. C., Jongepier, E., Robertson, H. M., Arning, N., Bitard-Feildel, T., Chao, H., … Bornberg-Bauer, E. (2018). Hemimetabolous genomes reveal molecular basis of termite eusociality. Nature Ecology and Evolution, 2(3), 557–566. https://doi.org/10.1038/s41559-017-0459-1 Hayashi, H., Takahashi, R., Nishi, T., Sakamoto, M., & Benno, Y. (2005). Molecular analysis of jejunal, ileal, caecal and recto-sigmoidal human colonic microbiota using 16S rRNA gene libraries and terminal restriction fragment length polymorphism. Journal of Medical Microbiology, 54(11), 1093–1101. https://doi.org/10.1099/jmm.0.45935-0 Henry, S. M. (1962). THE SIGNIFICANCE OF MICROORGANISMS IN THE NUTRITION OF INSECTS*. Transactions of the New York Academy of Sciences, 24(6 Series II), 676–683. https://doi.org/10.1111/j.2164-0947.1962.tb01905.x Hogan, M. E., Slaytor, M., & O’Brien, R. W. (1985). Transport of volatile fatty acids across the hindgut of the cockroach Panesthia cribrata Saussure and the termite, Mastotermes darwiniensis Froggatt. Journal of Insect Physiology, 31(7), 587–591. https://doi.org/10.1016/0022-1910(85)90116-7 Hongoh, Y., Sato, T., Dolan, M. F., Noda, S., Ui, S., Kudo, T., & Ohkuma, M. (2007). The motility symbiont of the termite gut flagellate Caduceia versatilis is a member of the “Synergistes” group. Applied and Environmental Microbiology, 73(19), 6270–6276. https://doi.org/10.1128/AEM.00750-07 Hongoh, Y., Sharma, V. K., Prakash, T., Noda, S., Taylor, T. D., Kudo, T., … Ohkuma, M. (2008). Complete genome of the uncultured Termite Group 1 bacteria in a single host protist cell. Proceedings of the National Academy of Sciences of the United States of America, 105(14), 5555–5560. https://doi.org/10.1073/pnas.0801389105 Hongoh, Y., Sharma, V. K., Prakash, T., Noda, S., Toh, H., Taylor, T. D., … Ohkuma, M. (2008). Genome of an endosymbiont coupling N2 fixation to cellulolysis within protist cells in termite gut. Science, 322(5904), 1108–1109. https://doi.org/10.1126/science.1165578 House, H. L., Patton, R. L. (1949). Nutritional Studies with Blatella germanica (L.) Reared Under Aseptic Conditions I. Equipment and Technique, (April 1949), 94– 100. House, H. L. (1949). Nutritional Studies with Blatella germanica (L.) Reared Under 143

Aseptic Conditions II. A Chemically Defined Diet. The Canadian Entomologist, 81(5), 105–112. Hoyte, H. M. D. (1961a). The protozoa occurring in the hind-gut of cockroaches. I. Responses to changes in environment. Parasitology, 51(3–4), 415–436. https://doi.org/10.1017/S0031182000070700 Hoyte, H. M. D. (1961b). The protozoa occurring in the hind-gut of cockroaches. III. Factors affecting the dispersion of Nyctotherus ovalis. Par, 465–495. Huber, I., Masler, E. P., & Rao, B. R. (1990). Cockroaches as Models for Neurobiology Applications in Biomedical Research, Volume 1. CRC Press. Retrieved from https://books.google.com/books?id=ms6r0okINBcC&pgis=1 Hume, I. D. (1997). Fermentation in the Hindgut of Mammals. In R. I. Mackie & B. A. White (Eds.), Gastrointestinal Microbiology (pp. 84–115). Boston, MA: Springer US. https://doi.org/10.1007/978-1-4615-4111-0_4 Izawa, K., Kuwahara, H., Kihara, K., Yuki, M., Lo, N., Itoh, T., … Hongoh, Y. (2016). Comparison of intracellular “ca. endomicrobium trichonymphae” genomovars illuminates the requirement and decay of defense systems against foreign DNA. Genome Biology and Evolution, 8(10), 3099–3107. https://doi.org/10.1093/gbe/evw227 Jahnes, B. C., Herrmann, M., & Sabree, Z. L. (2019). Conspecific coprophagy stimulates normal development in a germ-free model invertebrate. PeerJ, 7, e6914. https://doi.org/10.7717/peerj.6914 Johnson, L. R. (2019). Gastrointestinal physiology. Philadelphia, PA: Elsevier. Retrieved from https://library.ohio-state.edu/record=b8521888 Jones, R. M., Luo, L., Ardita, C. S., Richardson, A. N., Kwon, Y. M., Mercante, J. W., … Neish, A. S. (2013). Symbiotic lactobacilli stimulate gut epithelial proliferation via Nox-mediated generation of reactive oxygen species. The EMBO Journal, 32(23), 3017–3028. https://doi.org/10.1038/emboj.2013.224 Juge, N. (2012). Microbial adhesins to gastrointestinal mucus. Trends in Microbiology, 20(1), 30–39. https://doi.org/10.1016/j.tim.2011.10.001 Kakumanu, M. L., Maritz, J. M., Carlton, J. M., & Schal, C. (2018). Overlapping Community Compositions of Gut and Fecal Microbiomes in Lab-Reared and Field- Collected German Cockroaches. Applied and Environmental Microbiology, 84(17), 1–17. Kamra, D. N. (2005). Microbial Diversity. Rumen microbial ecosystem. Current Science, 89(1)(1), 124–135. https://doi.org/10.1146/annurev.es.06.110175.000351 Kane, M. D., & Breznak, J. A. (1991). Effect of host diet on production of organic acids and methane by cockroach gut bacteria. Applied and Environmental Microbiology, 57(9), 2628–2634. Kane, Matthew D. (1997). Microbial Fermentation in Insect Guts. In Gastrointestinal Microbiology (pp. 231–265). Boston, MA: Springer US. https://doi.org/10.1007/978-1-4615-4111-0_8 Kanzaki, N., Giblin-Davis, R. M., Scheffrahn, R. H., Taki, H., Esquivel, A., Davies, K. A., & Herre, E. A. (2012). Reverse Taxonomy for Elucidating Diversity of Insect- Associated Nematodes: A Case Study with Termites. PLoS ONE, 7(8). 144

https://doi.org/10.1371/journal.pone.0043865 Kawano, T., Imada, M., Chamavit, P., Kobayashi, S., Hashimoto, T., & Nozaki, T. (2017). Genetic diversity of Entamoeba: Novel ribosomal lineages from cockroaches. PLOS ONE, 12(9), e0185233. https://doi.org/10.1371/journal.pone.0185233 Kim, I.-W., Lee, J. H., Subramaniyam, S., Yun, E.-Y., Kim, I., Park, J., & Hwang, J. S. (2016). De Novo Transcriptome Analysis and Detection of Antimicrobial Peptides of the American Cockroach Periplaneta americana (Linnaeus). Plos One, 11(5), e0155304. https://doi.org/10.1371/journal.pone.0155304 Kim, S.-H., & Lee, W.-J. (2014). Role of DUOX in gut inflammation: lessons from Drosophila model of gut-microbiota interactions. Frontiers in Cellular and Infection Microbiology, 3(January), 116. https://doi.org/10.3389/fcimb.2013.00116 Klug, M. J., & Kotarski, S. (1980). Bacteria Associated with the Gut Tract of Larval Stages of the Aquatic Cranefly Tipula abdominalis (Diptera; Tipulidae). Applied and Environmental Microbiology, 40(2), 408–416. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/16345618%0Ahttp://www.pubmedcentral.nih. gov/articlerender.fcgi?artid=PMC291589 Kopanic, R. J., Holbrook, G. L., Sevala, V., & Schal, C. (2001). An adaptive benefit of facultative coprophagy in the German cockroach Blattella germanica. Ecological Entomology, 26(2), 154–162. https://doi.org/10.1046/j.1365-2311.2001.00316.x Kostic, A. D., Howitt, M. R., & Garrett, W. S. (2013). Exploring host-microbiota interactions in animal models and humans. Genes and Development, 27(7), 701– 718. https://doi.org/10.1101/gad.212522.112 Koyle, M. L., Veloz, M., Judd, A. M., Wong, A. C., Newell, P. D., Douglas, A. E., & Chaston, J. M. (2016). Rearing the Fruit Fly Drosophila melanogaster Under Axenic and Gnotobiotic Conditions, (July), 1–8. https://doi.org/10.3791/54219 Krinos, C. M., Krinos, C. M., Coyne, M. J., Coyne, M. J., Weinacht, K. G., Weinacht, K. G., … Comstock, L. E. (2001). Extensive surface diversity of a commensal microorganism by multiple DNA inversions. Nature, 414(6863), 555–558. https://doi.org/10.1038/35107092 Krych, L., Hansen, C. H. F., Hansen, A. K., van den Berg, F. W. J., & Nielsen, D. S. (2013). Quantitatively Different, yet Qualitatively Alike: A Meta-Analysis of the Mouse Core Gut Microbiome with a View towards the Human Gut Microbiome. PLoS ONE, 8(5). https://doi.org/10.1371/journal.pone.0062578 Kumar, S., Stecher, G., Suleski, M., & Hedges, S. B. (2017). TimeTree: A Resource for Timelines, Timetrees, and Divergence Times. Molecular Biology and Evolution, 34(7), 1812–1819. https://doi.org/10.1093/molbev/msx116 Kuwahara, H., Yuki, M., Izawa, K., Ohkuma, M., & Hongoh, Y. (2017). Genome of “Ca. Desulfovibrio trichonymphae”, an H 2 -oxidizing bacterium in a tripartite symbiotic system within a protist cell in the termite gut. ISME Journal, 11(3), 766–776. https://doi.org/10.1038/ismej.2016.143 Lee, K.-A., Kim, B., Bhin, J., Kim, D. H., You, H., Kim, E.-K., … Lee, W.-J. (2015). Bacterial Uracil Modulates Drosophila DUOX-Dependent Gut Immunity via Hedgehog-Induced Signaling Endosomes. Cell Host & Microbe, 17(2), 191–204. 145

https://doi.org/10.1016/J.CHOM.2014.12.012 Legendre, F., Nel, A., Svenson, G. J., Robillard, T., Pellens, R., & Grandcolas, P. (2015). Phylogeny of dictyoptera: Dating the origin of cockroaches, praying mantises and termites with molecular data and controlled fossil evidence. PLoS ONE, 10(7), 1–27. https://doi.org/10.1371/journal.pone.0130127 Lemke, T., Van Alen, T., Hackstein, J. H. P., & Brune, A. (2001). Cross-Epithelial Hydrogen Transfer from the Midgut Compartment Drives Methanogenesis in the Hindgut of Cockroaches. Applied and Environmental Microbiology, 67(10), 4657– 4661. https://doi.org/10.1128/AEM.67.10.4657-4661.2001 Li, S., Zhu, S., Jia, Q., Yuan, D., Ren, C., Li, K., … Zhan, S. (2018). The genomic and functional landscapes of developmental plasticity in the American cockroach. Nature Communications, 9(1). https://doi.org/10.1038/s41467-018-03281-1 Lind, A. E., Lewis, W. H., Spang, A., Guy, L., Embley, T. M., & Ettema, T. J. G. (2018). Genomes of two archaeal endosymbionts show convergent adaptations to an intracellular lifestyle. ISME Journal, 12(11), 2655–2667. https://doi.org/10.1038/s41396-018-0207-9 Liu, N., Li, H., Chevrette, M. G., Zhang, L., Cao, L., Zhou, H., … Wang, Q. (2019). Functional metagenomics reveals abundant polysaccharide-degrading gene clusters and cellobiose utilization pathways within gut microbiota of a wood-feeding higher termite. The ISME Journal, 13(1), 104–117. https://doi.org/10.1038/s41396-018- 0255-1 Ma, D., & Leulier, F. (2018). The importance of being persistent: The first true resident gut symbiont in Drosophila. PLoS Biology, 16(8), 1–8. https://doi.org/10.1371/journal.pbio.2006945 Macfarlane, S., Woodmansey, E. J., George, T., & Macfarlane, G. T. (2005). Colonization of Mucin by Human Intestinal Bacteria and Establishment of Biofilm Communities in a Two-Stage Continuous Culture System Colonization of Mucin by Human Intestinal Bacteria and Establishment of Biofilm Communities in a Two- Stage Continuous Cult. Appli, 71(11), 7483–7492. https://doi.org/10.1128/AEM.71.11.7483 Mackay, T. F. C., & Anholt, R. R. H. (2006). Of Flies and Man: Drosophila as a Model for Human Complex Traits. Http://Dx.Doi.Org/10.1146/Annurev.Genom.7.080505.115758. https://doi.org/10.1146/ANNUREV.GENOM.7.080505.115758 Malke, H., & Schwartz, W. (1966). Untersuchungen über die Symbiose von Tieren mit Pilzen und Bakterien. XII. Die Bedeutung der Blattiden-Symbiose. Zeitschrift Für Allgemeine Mikrobiologie, 6(1), 34–68. https://doi.org/10.1002/jobm.19660060106 Marlett, J. A., & Vollendorf, N. W. (1994). Dietary fiber content and composition of different forms of fruits. Food Chemistry, 51(1), 39–44. https://doi.org/10.1016/0308-8146(94)90045-0 Martens, E. C., Chiang, H. C., & Gordon, J. I. (2008). Mucosal Glycan Foraging Enhances Fitness and Transmission of a Saccharolytic Human Gut Bacterial Symbiont. Cell Host and Microbe, 4(5), 447–457. https://doi.org/10.1016/j.chom.2008.09.007 146

Martínez-Girón, R., Martínez-Torre, C., & van Woerden, H. C. (2017). The prevalence of protozoa in the gut of German cockroaches (Blattella germanica) with special reference to Lophomonas blattarum. Parasitology Research, 116(11), 3205–3210. https://doi.org/10.1007/s00436-017-5640-6 Martino, M. E., Joncour, P., Leenay, R., Gervais, H., Shah, M., Hughes, S., … Leulier, F. (2018). Bacterial Adaptation to the Host’s Diet Is a Key Evolutionary Force Shaping Drosophila-Lactobacillus Symbiosis. Cell Host and Microbe, 24(1), 109-119.e6. https://doi.org/10.1016/j.chom.2018.06.001 Massey, C. L. (1971). Two new genera of nematodes parasitic in the eastern subterranean termite, Reticulitermes flavipes. Journal of Invertebrate Pathology, 17(2), 238–242. https://doi.org/10.1016/0022-2011(71)90097-8 Mccallister, G. L. (1988). The Effect of Thelastoma bulhoesi and Hammerschmidtiella diesingi (Nematoda: Oxyurata) on Host Size and Physiology in Periplaneta americana (Arthropoda: Blattidae). Proc. Helminthol. Soc. Wash (Vol. 55). Mikaelyan, A, Thompson, C. L., Hofer, M. J., & Brune, A. (2016a). The deterministic assembly of complex bacterial communities in germ-free cockroach guts. Applied and Environmental Microbiology, (December), AEM.03700-15. https://doi.org/10.1128/AEM.03700-15 Mikaelyan, A, Thompson, C. L., Hofer, M. J., & Brune, A. (2016b). The deterministic assembly of complex bacterial communities in guts of germ-free cockroach. Applied and Environmental Microbiology, 82(4), 1256–1263. https://doi.org/10.1128/AEM.03700-15 Mikaelyan, Aram, Dietrich, C., Köhler, T., Poulsen, M., Sillam-Dussès, D., & Brune, A. (2015). Diet is the primary determinant of bacterial community structure in the guts of higher termites. Molecular Ecology, 24(20), 5284–5295. https://doi.org/10.1111/mec.13376 Mikaelyan, Aram, Köhler, T., Lampert, N., Rohland, J., Boga, H., Meuser, K., & Brune, A. (2015). Classifying the bacterial gut microbiota of termites and cockroaches: A curated phylogenetic reference database (DictDb). Systematic and Applied Microbiology, 38(7), 472–482. https://doi.org/10.1016/j.syapm.2015.07.004 Mikaelyan, Aram, Meuser, K., & Brune, A. (2017). Microenvironmental heterogeneity of gut compartments drives bacterial community structure in wood- and humus-feeding higher termites. FEMS Microbiology Ecology, 93(1), fiw210. https://doi.org/10.1093/femsec/fiw210 Mira, A. (2000). Exuviae eating: A nitrogen meal? Journal of Insect Physiology, 46(4), 605–610. https://doi.org/10.1016/S0022-1910(99)00146-8 Murakami, T., Onouchi, S., Igai, K., Ohkuma, M., & Hongoh, Y. (2019). Ectosymbiotic bacterial microbiota densely colonize the surface of thelastomatid nematodes in the gut of the wood-feeding cockroach Panesthia angustipennis. FEMS Microbiology Ecology, 95(2). https://doi.org/10.1093/femsec/fiy238 Nalepa C.A., Bignell D.E., B. C. (2001). Detritivory, coprophagy, and the evolution of digestive mutualisms in Dictyoptera. Insectes Soc., 48, 194–201. https://doi.org/10.1007/PL00001767 Nguyen, T. L. a., Vieira-Silva, S., Liston, a., & Raes, J. (2015). How informative is the 147

mouse for human gut microbiota research? Disease Models & Mechanisms, 8(1), 1– 16. https://doi.org/10.1242/dmm.017400 Noda, S., Iida, T., Kitade, O., Nakajima, H., Kudo, T., & Ohkuma, M. (2005). Endosymbiotic Bacteroidales bacteria of the flagellated protist Pseudotrichonympha grassii in the gut of the termite Coptotermes formosanus. Applied and Environmental Microbiology, 71(12), 8811–8817. https://doi.org/10.1128/AEM.71.12.8811-8817.2005 O’Kane, C. J. (2003). Modelling human diseases in Drosophila and Caenorhabditis. Seminars in Cell & Developmental Biology, 14(1), 3–10. https://doi.org/10.1016/S1084-9521(02)00162-3 Ohkuma, M., Noda, S., Hattori, S., Iida, T., Yuki, M., Starns, D., … Hongoh, Y. (2015). Acetogenesis from H2 plus CO2and nitrogen fixation by an endosymbiotic spirochete of a termite-gut cellulolytic protist. Proceedings of the National Academy of Sciences of the United States of America, 112(33), 10224– 10230. https://doi.org/10.1073/pnas.1423979112 Ottesen, E. a, & Leadbetter, J. R. (2010). Diversity of formyltetrahydrofolate synthetases in the guts of the wood-feeding cockroach Cryptocercus punctulatus and the omnivorous cockroach Periplaneta americana. Applied and Environmental Microbiology, 76(14), 4909–4913. https://doi.org/10.1128/AEM.00299-10 Oyebanji, O., Soyelu, O., Bamigbade, A., & Okonji, R. (2014). Distribution of digestive enzymes in the gut of American cockroach, Periplaneta americana (L.). International Journal of Scientific and Research Publications, 4(1). Retrieved from www.ijsrp.org Ozawa, S., Vicente, C. S. L., Sato, K., Yoshiga, T., Kanzaki, N., & Hasegawa, K. (2014). First report of the nematode Leidynema appendiculata from Periplaneta fuliginosa. Acta Parasitologica, 59(2), 219–228. https://doi.org/10.2478/s11686-014-0230-6 Ozdal, M., & Ozdal, O. G. (2016). Isolation and Characterization of α-Endosulfan Degrading Bacteria from the Microflora of Cockroaches. Polish Journal of Microbiology, 65(1), 63–68. https://doi.org/10.5604/17331331.1197325 Pais, I. S., Valente, R. S., Sporniak, M., & Teixeira, L. (2018). Drosophila melanogaster establishes a species-specific mutualistic interaction with stable gut-colonizing bacteria. PLoS biology (Vol. 16). https://doi.org/10.1371/journal.pbio.2005710 Park, M. S., Park, P., & Takeda, M. (2009). Starvation induces apoptosis in the midgut nidi of Periplaneta americana : a histochemical and ultrastructural study, 631–638. https://doi.org/10.1007/s00441-008-0737-y Park, M. S., & Takeda, M. (2008). Starvation suppresses cell proliferation that rebounds after refeeding in the midgut of the American cockroach, Periplaneta americana. Journal of Insect Physiology, 54(2), 386–392. https://doi.org/10.1016/j.jinsphys.2007.10.011 Patel, P. H., Maldera, J. a, & Edgar, B. a. (2013). Stimulating cROSstalk between commensal bacteria and intestinal stem cells. The EMBO Journal, 32(23), 3009– 3010. https://doi.org/10.1038/emboj.2013.244 Pietri, J. E., Tiffany, C., & Liang, D. (2018). Disruption of the microbiota affects physiological and evolutionary aspects of insecticide resistance in the German 148

cockroach, an important urban pest. PLOS ONE, 13(12), e0207985. https://doi.org/10.1371/journal.pone.0207985 Poole, L. B., Godzik, A., Nayeem, A., & Schmitt, J. D. (2000). AhpF can be dissected into two functional units: Tandem repeats of two Thioredoxin-like folds in the N- terminus mediate electron transfer from the thioredoxin reductase-like C-terminus to AhpC. Biochemistry, 39(22), 6602–6615. https://doi.org/10.1021/bi000405w Pramono, A. K., Kuwahara, H., Itoh, T., Toyoda, A., Yamada, A., & Hongoh, Y. (2017). Discovery and complete genome sequence of a bacteriophage from an obligate intracellular symbiont of a cellulolytic protist in the termite gut. Microbes and Environments, 32(2), 112–117. https://doi.org/10.1264/jsme2.ME16175 Reiter, L. T., Potocki, L., Chien, S., Gribskov, M., & Bier, E. (2001). A systematic analysis of human disease-associated gene sequences in Drosophila melanogaster. Genome Research, 11(6), 1114–1125. https://doi.org/10.1101/gr.169101 Rocha, E. R., & Smith, C. J. (2004). Transcriptional regulation of the Bacteroides fragilis ferritin gene (ftnA) by redox stress. Microbiology, 150(7), 2125–2134. https://doi.org/10.1099/mic.0.26948-0 Rocha, E. R., & Smith, C. J. (2013). Ferritin-like family proteins in the anaerobe Bacteroides fragilis: when an oxygen storm is coming, take your iron to the shelter. BioMetals, 26(4), 577–591. https://doi.org/10.1007/s10534-013-9650-2 Rosas, T., García-Ferris, C., Domínguez-Santos, R., Llop, P., Latorre, A., & Moya, A. (2018). Rifampicin treatment of Blattella germanica evidences a fecal transmission route of their gut microbiota. FEMS Microbiology Ecology, 94(2). https://doi.org/10.1093/femsec/fiy002 Rosengaus, R. B., Zecher, C. N., Schultheis, K. F., Brucker, R. M., & Bordenstein, S. R. (2011). Disruption of the termite gut microbiota and its prolonged consequences for fitness. Applied and Environmental Microbiology, 77(13), 4303–4312. https://doi.org/10.1128/AEM.01886-10 Rossmassler, K., Dietrich, C., Thompson, C., Mikaelyan, A., Nonoh, J. O., Scheffrahn, R. H., … Brune, A. (2015). Metagenomic analysis of the microbiota in the highly compartmented hindguts of six wood- or soil-feeding higher termites. Microbiome, 3(1), 56. https://doi.org/10.1186/s40168-015-0118-1 Ryu, J. H., Kim, S. H., Lee, H. Y., Bai, J. Y., Nam, Y. D., Bae, J. W., … Lee, W. J. (2008). Innate immune homeostasis by the homeobox gene caudal and commensal- gut mutualism in Drosophila. Science, 319(5864), 777–782. https://doi.org/10.1126/science.1149357 Ryu, J., Kim, S., Lee, H., Bai, J. Y., Nam, Y., Bae, J., … Lee, W. (2008). Innate Immune Homeostasis by the Homeobox Gene Caudal. Science, 319(February), 777–783. https://doi.org/10.1126/science.1149357 S. L. Vicente, C., Ozawa, S., & Hasegawa, K. (2016). Composition of the Cockroach Gut Microbiome in the Presence of Parasitic Nematodes. Microbes and Environments, 31(3), 314–320. https://doi.org/10.1264/jsme2.ME16088 Sabree, Z. L., & Moran, N. a. (2014). Host-specific assemblages typify gut microbial communities of related insect species. SpringerPlus, 3(1), 138. https://doi.org/10.1186/2193-1801-3-138 149

Sakai, T., Satake, H., & Takeda, M. (2006). Nutrient-induced-amylase and protease activity is regulated by crustacean cardioactive peptide (CCAP) in the cockroach midgut. Peptides, 27(9), 2157–2164. https://doi.org/10.1016/j.peptides.2006.04.009 Savage, D. C. (1977). Microbial Ecology of the Gastrointestinal Tract. Annual Review of Microbiology, 31(1), 107–133. https://doi.org/10.1146/annurev.mi.31.100177.000543 Schauer, C., Thompson, C., & Brune, A. (2014). Pyrotag sequencing of the gut microbiota of the cockroach Shelfordella lateralis reveals a highly dynamic core but only limited effects of diet on community structure. PloS One, 9(1), e85861. https://doi.org/10.1371/journal.pone.0085861 Schauer, C., Thompson, C. L., & Brune, A. (2012). The bacterial community in the gut of the cockroach Shelfordella lateralis reflects the close evolutionary relatedness of cockroaches and termites. Applied and Environmental Microbiology, 78(8), 2758– 2767. https://doi.org/10.1128/AEM.07788-11 Shelton, A. N., Seth, E. C., Mok, K. C., Han, A. W., Jackson, S. N., Haft, D. R., & Taga, M. E. (2019). Uneven distribution of cobamide biosynthesis and dependence in bacteria predicted by comparative genomics. ISME Journal, 13(3), 789–804. https://doi.org/10.1038/s41396-018-0304-9 Shin, S. C., Kim, S., You, H., Kim, B., Kim, A. C., Lee, K., … Lee, W. (2011). Drosophila Microbiome Modulates Host Developmental and Metabolic Homeostasis via Insulin Signaling, (November). Simmonds, M. S. J., & Preedy, V. R. (2016). Nutritional composition of fruit cultivars. Elsevier. Retrieved from https://www.sciencedirect.com/book/9780124081178/nutritional-composition-of- fruit-cultivars Snipes, B. T., & Tauber, O. E. (1937). Time Required for Food Passage through the Alimentary Tract of the Cockroach, Periplaneta Americana Linn. Annals of the Entomological Society of America, 30(2). Sprenger, W. W., Van Belzen, M. C., Rosenberg, J., Hackstein, J. H. P., & Keltjens, J. T. (2000). Methanomicrococcus blatticola gen. nov., sp. nov., a methanol- and methylamine-reducing methanogen from the hindgut of the cockroach Periplaneta americana. International Journal of Systematic and Evolutionary Microbiology, 50(6), 1989–1999. https://doi.org/10.1099/00207713-50-6-1989 Sprenger, Wander W., Hackstein, J. H. P., & Keltjens, J. T. (2007). The competitive success of Methanomicrococcus blatticola, a dominant methylotrophic methanogen in the cockroach hindgut, is supported by high substrate affinities and favorable thermodynamics. FEMS Microbiology Ecology, 60(2), 266–275. https://doi.org/10.1111/j.1574-6941.2007.00287.x Stankiewicz, M., Dąbrowski, M., & de Lima, M. E. (2012). Nervous System of Periplaneta americana Cockroach as a Model in Toxinological Studies: A Short Historical and Actual View. Journal of Toxicology, 2012, 1–11. https://doi.org/10.1155/2012/143740 Stewart, C. S., Flint, H. J., & Bryant, M. P. (1997). The rumen bacteria. In The Rumen Microbial Ecosystem (pp. 10–72). Dordrecht: Springer Netherlands. 150

https://doi.org/10.1007/978-94-009-1453-7_2 Storelli, G., Defaye, A., Erkosar, B., Hols, P., Royet, J., & Leulier, F. (2011). Lactobacillus plantarum Promotes Drosophila Systemic Growth by Modulating Hormonal Signals through TOR-Dependent Nutrient Sensing. Cell Metabolism, 14(3), 403–414. https://doi.org/10.1016/J.CMET.2011.07.012 Strassert, J. F. H., Mikaelyan, A., Woyke, T., & Brune, A. (2016). Genome analysis of ‘Candidatus ancillula trichonymphae’, first representative of a deep-branching clade of bifidobacteriales, strengthens evidence for convergent evolution in flagellate endosymbionts. Environmental Microbiology Reports, 8(5), 865–873. https://doi.org/10.1111/1758-2229.12451 Su, Lijuan, Yang, L., Huang, S., Li, Y., Su, X., Wang, F., … Song, A. (2017). Variation in the Gut Microbiota of Termites (Tsaitermes ampliceps) Against Different Diets. Applied Biochemistry and Biotechnology, 181(1), 32–47. https://doi.org/10.1007/s12010-016-2197-2 Su, LiJuan, Yang, L., Huang, S., Su, X., Li, Y., Wang, F., … Song, A. (2016). Comparative Gut Microbiomes of Four Species Representing the Higher and the Lower Termites. Journal of Insect Science, 16(1), 97. https://doi.org/10.1093/jisesa/iew081 Swanson, K. S., Dowd, S. E., Suchodolski, J. S., Middelbos, I. S., Vester, B. M., Barry, K. A., … Fahey, G. C. (2011). Phylogenetic and gene-centric metagenomics of the canine intestinal microbiome reveals similarities with humans and mice. ISME Journal, 5(4), 639–649. https://doi.org/10.1038/ismej.2010.162 Swingle, H. S. (1925). Digestive Enzymes of an Insect. The Ohio Journal of Science, 25(5), 209–218. Retrieved from http://hdl.handle.net/1811/2281 Tai, V., Carpenter, K. J., Weber, P. K., Nalepa, C. A., Perlman, S. J., & Keeling, P. J. (2016). Genome Evolution and Nitrogen Fixation in Bacterial Ectosymbionts of a Protist Inhabiting Wood-Feeding Cockroaches. Applied and Environmental Microbiology, 82(15), 4682–4695. https://doi.org/10.1128/AEM.00611-16 Tailford, L. E., Crost, E. H., Kavanaugh, D., & Juge, N. (2015). Mucin glycan foraging in the human gut microbiome. Frontiers in Genetics, 5(FEB). https://doi.org/10.3389/fgene.2015.00081 Tamaki, F. K., Pimentel, A. C., Dias, A. B., Cardoso, C., Ribeiro, A. F., Ferreira, C., & Terra, W. R. (2014). Physiology of digestion and the molecular characterization of the major digestive enzymes from Periplaneta americana. Journal of Insect Physiology, 70(November), 22–35. https://doi.org/10.1016/j.jinsphys.2014.08.007 Tang, Y. P., Dallas, M. M., & Malamy, M. H. (1999). Characterization of the BatI (Bacteroides aerotolerance) operon in Bacteroides fragilis: isolation of a B. fragilis mutant with reduced aerotolerance and impaired growth in in vivo model systems. Molecular Microbiology, 32(1), 139–149. https://doi.org/10.1046/j.1365- 2958.1999.01337.x Tegtmeier, D. (2016). The gut microbiota of the cockroach Shelfordella lateralis: primary colonization, succession, and metabolic response to microenvironmental conditions. University of Marburg, Marburg, Germany. Tegtmeier, D., Thompson, C. L., Schauer, C., & Brune, A. (2016). Oxygen affects gut 151

bacterial colonization and metabolic activities in a gnotobiotic cockroach model. Applied and Environmental Microbiology, 82(4), 1080–1089. https://doi.org/10.1128/AEM.03130-15 Terra, W. (1990). Evolution Of Digestive Systems Of Insects. Annual Review of Entomology, 35(1), 181–200. https://doi.org/10.1146/annurev.ento.35.1.181 Terrapon, N., Li, C., Robertson, H. M., Ji, L., Meng, X., Booth, W., … Liebig, J. (2014). Molecular traces of alternative social organization in a termite genome. Nature Communications, 5. https://doi.org/10.1038/ncomms4636 Tikhe, C. V., & Husseneder, C. (2018). Metavirome sequencing of the termite gut reveals the presence of an unexplored bacteriophage community. Frontiers in Microbiology, 8(JAN). https://doi.org/10.3389/fmicb.2017.02548 Tinker, K. A., & Ottesen, E. A. (2016). The core gut microbiome of the American cockroach, Periplaneta americana , is stable and resilient to dietary shifts. Applied and Environmental Microbiology, (September), AEM.01837-16. https://doi.org/10.1128/AEM.01837-16 Tokuda, G., Tsuboi, Y., Kihara, K., Saitou, S., Moriya, S., Lo, N., & Kikuchi, J. (2014). Metabolomic profiling of 13 C-labelled cellulose digestion in a lower termite: Insights into gut symbiont function. Proceedings of the Royal Society B: Biological Sciences, 281(1789). https://doi.org/10.1098/rspb.2014.0990 Treherne, J. E. (1957). Glucose absorption in the cockroach. The Journal of Experimental Biology, 34(4), 478–485. Retrieved from http://jeb.biologists.org/content/34/4/478.short Treherne, J. E. (1958). THE DIGESTION AND ABSORPTION OF TRIPALMITIN IN THE COCKROACH, PERIPLANETA AMERICANA L. Journal of Experimental Biology, 55(2), 862–870. Treherne, J. E. (1967). Gut absorption. Annual Review of Entomology, 12, 43–58. Treitli, S. C., Kotyk, M., Yubuki, N., Jirounková, E., Vlasáková, J., Smejkalová, P., … Hampl, V. (2018). Molecular and Morphological Diversity of the Oxymonad Genera Monocercomonoides and Blattamonas gen. nov. Protist, 169(5), 744–783. https://doi.org/10.1016/j.protis.2018.06.005 Ugur, B., Chen, K., & Bellen, H. J. (2016). Drosophila tools and assays for the study of human diseases. Disease Models & Mechanisms, 9(3), 235–244. https://doi.org/10.1242/dmm.023762 Utami, Y. D., Kuwahara, H., Igai, K., Murakami, T., Sugaya, K., Morikawa, T., … Hongoh, Y. (2019). Genome analyses of uncultured TG2/ZB3 bacteria in ‘Margulisbacteria’ specifically attached to ectosymbiotic spirochetes of protists in the termite gut. ISME Journal, 13(2), 455–467. https://doi.org/10.1038/s41396-018- 0297-4 Vďačný, P., Érseková, E., Šoltys, K., Budiš, J., Pecina, L., & Rurik, I. (2018). Co- existence of multiple bacterivorous clevelandellid ciliate species in hindgut of wood- feeding cockroaches in light of their prokaryotic consortium. Scientific Reports, 8(1), 17749. https://doi.org/10.1038/s41598-018-36245-y Vera-Ponce de León, A., Jahnes, B. C., Duan, J., Camuy-Vélez, L. A., & Sabree, Z. L. (2020). Cultivable, Host-Specific Bacteroidetes Symbionts Exhibit Diverse 152

Polysaccharolytic Strategies. Applied and Environmental Microbiology. https://doi.org/10.1128/AEM.00091-20 Vicente, C. S. L., Ozawa, S., & Hasegawa, K. (2016). Composition of the cockroach gut microbiome in the presence of parasitic nematodes. Microbes and Environments, 31(3), 314–320. https://doi.org/10.1264/jsme2.ME16088 von Beeren, C., Stoeckle, M. Y., Xia, J., Burke, G., & Kronauer, D. J. C. (2015). Interbreeding among deeply divergent mitochondrial lineages in the American cockroach (Periplaneta americana). Scientific Reports, 5, 8297. https://doi.org/10.1038/srep08297 Wada-Katsumata, A., Zurek, L., Nalyanya, G., Roelofs, W. L., Zhang, A., & Schal, C. (2015). Gut bacteria mediate aggregation in the German cockroach. Proceedings of the National Academy of Sciences, 112(51), 15678–15683. https://doi.org/10.1073/pnas.1504031112 Wexler, A. G., Schofield, W. B., Degnan, P. H., Folta-Stogniew, E., Barry, N. A., & Goodman, A. L. (2018). Human gut Bacteroides capture vitamin B12 via cell surface-exposed lipoproteins. ELife, 7. https://doi.org/10.7554/eLife.37138 Williams, A. G., Withers, S. E., & Joblin, K. N. (1994). The effect of cocultivation with hydrogen-consuming bacteria on xylanolysis byRuminococcus flavefaciens. Current Microbiology, 29(3), 133–138. https://doi.org/10.1007/BF01570753 Wilson-Sanders, S. E. (2011). Invertebrate Models for Biomedical Research, Testing, and Education. ILAR Journal, 52(2), 126–152. https://doi.org/10.1093/ilar.52.2.126 Wolf, M. J., Amrein, H., Izatt, J. A., Choma, M. A., Reedy, M. C., & Rockman, H. A. (2006). Drosophila as a model for the identification of genes causing adult human heart disease. Retrieved from www.pnas.orgcgidoi10.1073pnas.0507359103 Wong, A. C., Dobson, A. J., & Douglas, A. E. (2014). Gut microbiota dictates the metabolic response of Drosophila to diet. J Exp Biol, 217(Pt 11), 1894–1901. https://doi.org/10.1242/jeb.101725 Wong, A. C. N., Vanhove, A. S., & Watnick, P. I. (2016). The interplay between intestinal bacteria and host metabolism in health and disease: lessons from Drosophila melanogaster. Disease Models & Mechanisms, 9(3), 271–281. https://doi.org/10.1242/dmm.023408 Wong, C. N. A., Ng, P., & Douglas, A. E. (2011). Low-diversity bacterial community in the gut of the fruitfly Drosophila melanogaster. Environmental Microbiology, 13(7), 1889–1900. https://doi.org/10.1111/j.1462-2920.2011.02511.x Xiao, X., Yang, L., Pang, X., Zhang, R., Zhu, Y., Wang, P., … Cheng, G. (2017). A Mesh–Duox pathway regulates homeostasis in the insect gut. Nature Microbiology, 2(March), 17020. https://doi.org/10.1038/nmicrobiol.2017.20 Yamada, R., Deshpande, S. A., Bruce, K. D., Mak, E. M., & Ja, W. W. (2015). Microbes Promote Amino Acid Harvest to Rescue Undernutrition in Drosophila. Cell Reports, 10(6), 865–872. https://doi.org/10.1016/j.celrep.2015.01.018 Yi, P., & Li, L. (2012). The germfree murine animal: An important animal model for research on the relationship between gut microbiota and the host. Veterinary Microbiology, 157(1–2), 1–7. https://doi.org/10.1016/j.vetmic.2011.10.024 Yuki, M., Kuwahara, H., Shintani, M., Izawa, K., Sato, T., Starns, D., … Ohkuma, M. 153

(2015). Dominant ectosymbiotic bacteria of cellulolytic protists in the termite gut also have the potential to digest lignocellulose. Environmental Microbiology, 17(12), 4942–4953. https://doi.org/10.1111/1462-2920.12945 Zhang, C., Zhang, M., Wang, S., Han, R., Cao, Y., Hua, W., … Zhao, L. (2010). Interactions between gut microbiota, host genetics and diet relevant to development of metabolic syndromes in mice. The ISME Journal, 4(2), 232–241. https://doi.org/10.1038/ismej.2009.112 Zhang, F., Sun, X. X., Zhang, X. C., Zhang, S., Lu, J., Xia, Y. M., … Wang, X. J. (2018). The interactions between gut microbiota and entomopathogenic fungi: a potential approach for biological control of Blattella germanica (L.). Pest Management Science, 74(2), 438–447. https://doi.org/10.1002/ps.4726 Zhang, J., Zhang, Y., Li, J., Liu, M., & Liu, Z. (2016). Midgut Transcriptome of the Cockroach Periplaneta americana and Its Microbiota: Digestion, Detoxification and Oxidative Stress Response. PloS One, 11(5), e0155254. https://doi.org/10.1371/journal.pone.0155254 Zheng, H., Dietrich, C., & Brune, A. (2017). Genome analysis of Endomicrobium proavitum suggests loss and gain of relevant functions during the evolution of intracellular symbionts. Applied and Environmental Microbiology, 83(17). https://doi.org/10.1128/AEM.00656-17 Zheng, H., Powell, J. E., Steele, M. I., Dietrich, C., & Moran, N. A. (2017). Honeybee gut microbiota promotes host weight gain via bacterial metabolism and hormonal signaling. Proceedings of the National Academy of Sciences, 114(18), 4775–4780. https://doi.org/10.1073/pnas.1701819114 Zheng, W., Zhao, Y., & Zhang, H. (2012). Morphology and ultrastructure of the hindgut fermentation chamber of a melolonthine beetle Holotrichia parallela (Coleoptera: Scarabaeidae) during larval development. Micron, 43(5), 638–642. https://doi.org/10.1016/j.micron.2011.11.009 Zhou, Z., Ribeiro, A. A., Lin, S., Cotter, R. J., Miller, S. I., & Raetz, C. R. H. (2001). Lipid A modifications in polymyxin-resistant Salmonella typhimurium: PmrA- dependent 4-amino-4-deoxy-L-arabinose, and phosphoethanolamine incorporation. Journal of Biological Chemistry, 276(46), 43111–43121. https://doi.org/10.1074/jbc.M106960200 Zurek, L., & Keddie, B. A. (1998). Significance of methanogenic symbionts for development of the American cockroach, Periplaneta americana. Journal of Insect Physiology, 44(7), 645–651. Retrieved from http://www.sciencedirect.com/science/article/pii/S0022191098000249 Zurek, Ludek, & Keddie, B. A. (1996). Contribution of the colon and colonic bacterial flora to metabolism and development of the american cockroach Periplaneta americana L. Journal of Insect Physiology, 42(8), 743–748. https://doi.org/10.1016/0022-1910(96)00028-5

154

Appendix A. Supplementary Information Chapter 2

Table 4 Isolate Specific Primers

155

Figure 28 Cage Collected Ootheca Rinse Plating

156

Figure 29 Cockroach Collected Ootheca Rinse Plating

157

Appendix B. Supplementary Information Chapter 3

Figure 30 Wild-type Time to 5th Instar

Age at 5th instar of wild-type insects for reference to germ-free and conventionalized insects.

158

Figure 31 Body Mass Boxplot

Body mass comparisons between wild-type, germ-free, and conventionalized P. americana at four life stages. Asterisks indicate significant (p<0.05) differences given a Mann-Whitney test. n= number of individuals measured per sample.

159

Figure 32 Germ Free Cockroach Validation dPCR of DNA extracts from GF frass showing no amplification using universal primers among GF samples and amplification in one contaminated sample, in addition to the positive control.

160

Figure 33 Foregut Length Boxplot

161

Figure 34 Gut Mass Boxplot

162

Figure 35 Partition of Variation

Figure 36 PCA All instars

163

Table 5 p-value Table Morphological Measures

164

Appendix C. Supplementary information Chapter 5

165

Figure 37 Bacteroides fragilis Capsular Polysaccharide Loci

166

Figure 38 Bacteroides In-vitro Coculture

167

Figure 39 Bacteroides Post-Inoculation In-vitro Coculture

168