Insects and Their Beneficial Microbes 9780691236230, 9780691192406

Citation preview

Insects and Their Beneficial Microbes

Insects and Their Beneficial Microbes ANGEL A E. DOUGL AS

pr i n c e ­t o n u n i v er s i t y pr es s pr i n c e ­t o n

&

ox fo r d

Copyright © 2022 by Princeton University Press Princeton University Press is committed to the protection of copyright and the intellectual property our authors entrust to us. Copyright promotes the progress and integrity of knowledge. Thank you for supporting free speech and the global exchange of ideas by purchasing an authorized edition of this book. If you wish to reproduce or distribute any part of it in any form, please obtain permission. Requests for permission to reproduce material from this work should be sent to [email protected] Published by Princeton University Press 41 William Street, Princeton, New Jersey 08540 99 Banbury Road, Oxford OX2 6JX press.princeton.edu All Rights Reserved Library of Congress Cataloging-in-Publication Data Names: Douglas, A. E. (Angela Elizabeth), 1956– author. Title: Insects and their beneficial microbes / Angela E. Douglas. Description: Princeton : Princeton University Press, [2022] | Includes bibliographical references and index. Identifiers: LCCN 2021041786 (print) | LCCN 2021041787 (ebook) | ISBN 9780691192406 (hardback) | ISBN 9780691236230 (ebook) Subjects: LCSH: Microorganisms. | Insects—Molecular aspects. | Symbiosis. | Bacteria. Classification: LCC QL493.5 .D68 2022 (print) | LCC QL493.5 (ebook) | DDC 595.7—dc23 LC record available at https://lccn.loc.gov/2021041786 LC ebook record available at https://lccn.loc.gov/2021041787 British Library Cataloging-in-Publication Data is available Editorial: Alison Kalett and Hallie Schaeffer Production Editorial: Natalie Baan Jacket Design: Karl Spurzem Production: Danielle Amatucci Publicity: Charlotte Coyne and Matthew Taylor Copyeditor: Jennifer McClain Cover image: The beewolf Philanthus rugosus (Hymenoptera: Crabronidae). Females of this insect species bear the protective symbiotic bacteria Streptomyces philanthi in their antennae. Courtesy of Martin Kaltenpoth. This book has been composed in Arno Printed on acid-free paper. ∞ Printed in the United States of America 10 9 8 7 6 5 4 3 2 1

contents

Preface  ix 1 Introduction: The Diversity of Insects and Their Microbial Symbionts 1.1 Naming the partners

1 2

1.2 Insect habitats for microorganisms

10

1.3 Insect associations with beneficial microorganisms

16

1.4 How this book is structured

19

2 The Diversity of Insect-­Microbial Associations

23

2.1 Ectosymbioses

24

2.2 Gut symbioses

37

2.3 Endosymbioses

47

2.4 The dorsal organs of lagriine beetles

65

2.5 Summary

66

3 How Insects Acquire and Control Their Microbial Symbionts

69

3.1 Environmental microorganisms acquired by insects

70

3.2 Microbes acquired from other insects

80

3.3 Routes of horizontal transmission in microorganisms with high vertical transmission rates

92

3.4 The localization and abundance of endosymbionts in insects

96

3.5 Determinants of the composition of the insect gut microbiome

100

3.6 Microbial populations through host development

110

3.7 Summary

117 v

vi  c o n t e n t s

4 Microbial Ser­vices

119

4.1 Microbial degradation of complex biopolymers in the insect diet

120

4.2 Microbial contributions to insect nitrogen nutrition

132

4.3 B vitamin provisioning

150

4.4 Sterol provisioning

157

4.5 Microbial protection against natu­ral enemies

160

4.6 Microbial detoxification of dietary toxins and insecticidal chemicals

170

4.7 Microorganisms and insect be­hav­ior

177

4.8 Microorganisms and insect tolerance of abiotic conditions

185

4.9 Summary

187

5 Harnessing Microbial Symbionts to Manage Insect Pests and Vectors of Disease

190

5.1 Native microorganisms

192

5.2 Heterologous microorganisms

205

5.3 Engineered microorganisms

215

5.4 Targeting required microbial symbionts and their interactions with the insect host

225

5.5 Outlook

228

5.6 Summary

232

6 The Insect Microbiome as a Biomedical Model

234

6.1 Insect model systems

235

6.2 The interface between the gut microbiota and the Drosophila gut

243

6.3 The gut microbiota and immune function

250

6.4 The gut microbiome and metabolic health

256

6.5 The gut microbiome, ner­vous system function, and be­hav­ior

261

6.6 Summary

270

7 Priorities for the Study of Insect-­Microbial Associations 7.1 Reinvigorating the microbiology of insect-­microbial associations

272 273

c o n t e n t s 

vii

7.2 Modes of interaction between insects and microorganisms

276

7.3 Managing microbiomes for insect health

278

7.4 Concluding comments

281

References  283 Index  317

p r e fa c e

Not all microorganisms that infect insects and other animals are pathogens. Evidence accumulated over more than a c­ entury has shown that healthy insects bear microorganisms and that many insect species require their microbial partners for sustained growth and reproduction. For much of the last c­ entury, entomologists—­being thoroughly practical ­people—­recognized two ­things: that ­these associations had ­great potential to control insect pests; and that they ­were largely intractable to study ­because most of the microbial partners could not be cultured. While the study of insect-­pathogen interactions was a mainstream discipline in entomology and microbiology, the study of beneficial microbes associated with insects was a fringe topic, pursued by relatively few researchers. Every­thing changed in the opening years of this ­century. Novel sequencing and microscopical technologies have made it pos­si­ble to identify microorganisms and investigate their functional traits without cultivation. The discipline of microbiology was transformed, and the new discipline of microbiome science was founded. A major goal of microbiome research is biomedical, to develop microbial therapeutics that ameliorate and cure chronic h­ uman diseases. Nevertheless, many of the new technologies are readily applicable to insects, reshaping our capacity to study insect interactions with their communities of resident microorganisms. This in turn has created new opportunities to harness t­ hese insect-­microbial associations for novel strategies to manage insect pests and disease vectors, and also to use insects as models for ­human microbiome research. We have an unparalleled scope for fundamental discovery and application for the public good. The rapid expansion of research on insect associations with beneficial microorganisms over the last de­cade has been exciting, perhaps even overwhelming. The primary lit­er­a­ture has ballooned, and e­ very month brings more mini reviews, perspectives, and opinion pieces on individual associations or specific ix

x P r e fa c e

aspects of insect-­microbial interactions. Making sense of this burgeoning information is demanding and made more difficult by mutual incomprehension among colleagues approaching this subject area from dif­fer­ent disciplines. Entomologists recognize that the biomedical microbiome lit­er­a­ture is impor­ tant but often strug­gle to follow it, while many microbiologists and biomedical microbiome researchers find much of the entomological lit­er­a­ture opaque. It is from many conversations with colleagues and students that I came to understand that a book about the relationships between insects and beneficial microbes could contribute to solving ­these difficulties. The specific purpose of this book is to provide a framework of the basic concepts and key studies in the field of insect interactions with beneficial microorganisms, and to explore how recent advances can be applied in insect pest management and biomedicine. My goal is to pre­sent the fundamentals of this field to assist colleagues and students in their research, teaching, and learning. I assume that readers have a university-­level education in the life sciences. The focus of this book is insects and their bacterial, archaeal, and eukaryotic microorganisms. I set myself ­these taxonomic limits as a “rule” when I started this proj­ect and, despite many temptations, I have not deviated, to ensure consistency in my subject material. ­After all, not all rules are ­there to be broken. This book does not consider noninsect arthropods, such as acarines, millipedes and centipedes, or crustaceans. Similarly, it does not address insect interactions ­either with animal endosymbionts, such as nematodes, or with beneficial viruses, such as polydnaviruses. Th ­ ese vari­ous systems illuminate our understanding of symbiotic systems in diverse and fascinating ways; but if I included them, this book would quickly have lost its focus. Even with my rule obeyed, I have not been able to include some fascinating topics and to describe the studies of many colleagues within the word limit wisely set by the publisher. Some of ­these decisions have been hard to make. I recognize fully how much of the primary lit­er­a­ture that goes unmentioned in this book has enriched my understanding of insect-­microbial associations. I have many ­people to thank. My special thanks are to colleagues who took time out of their busy schedules to read individual chapters: Arinder Arora, John Chaston, Nicole Gerardo, Cole Gilbert, Jiri Hulcr, Corrie Morreau, Hassan Salem, Jeremy Searle, Michael Turrelli, and Linda Walling; and also to Kerry Oliver and two anonymous reviewers for their most helpful advice. I am

P r e fa c e   xi

grateful to you all for your insight, thoughtful comments, and corrections of my errors. I also thank Alison Kalett, the editor at Prince­ton University Press, for her enthusiasm for this book and her good advice. Fi­nally, and as always, I thank Jeremy for his unfailing support and encouragement. June 15, 2021

Insects and Their Beneficial Microbes

1 Introduction the diversit y of insects and t heir mi cro bi a l sy mbi o n ts

The insects are a supremely successful group of animals, particularly in terrestrial habitats. By the criteria of number of individuals, number of species, ecological importance, and functional diversity, the insects dwarf all other terrestrial animals. The basis for this book is that associations with beneficial microorganisms make an impor­tant contribution to the success of the insects. For example, the role of the leafcutter ants as the dominant herbivores in many Neotropical grassland and forest habitats is founded on their cultivation of plant-­degrading fungi in their nests; the obligate blood-­feeding lifestyle of bedbugs, sucking lice, and tsetse flies is enabled by internal bacteria that produce vitamins in short supply in vertebrate blood; and many insects, including bumble bees and honey bees, harbor gut microorganisms that protect t­ hese insects from virulent pathogens. In general terms, insect associations with beneficial microorganisms are not exceptional. Most animals host beneficial microorganisms, from which they derive nutrients, protection from natu­ral enemies, or other ser­vices (McFall-­Ngai et al., 2013). Th ­ ere are, however, two ways in which microbial associations in insects are special. The first is the remarkable diversity of form and function of associations in insects, as discussed in chapters 2–4. The second way relates to the significance of insects to h­ umans. Some insects are pests and disease vectors of agricultural, medical, and veterinary importance, and ­others are valuable to ­humans, for example, as pollinators and biological control agents. As our knowledge of insect associations with microorganisms develops, it is becoming increasingly evident that this understanding can be harnessed for 1

2  c h a p t e r 1

novel strategies to control insect pests (chapter 5) and as model systems in biomedical research (chapter 6). The purpose of this chapter is to provide an overview of the insects and their microbial partners. Section 1.1 introduces the reader to the diversity of insects and microorganisms. Section 1.2 describes the essentials of insect form and function, with an emphasis on the insect structures that provide habitats for microorganisms and insect traits that facilitate and limit microbial colonization. (An extended consideration of insect structure and function is provided by Simpson and Douglas (2013).) In section 1.3, I consider the terminology used to describe interactions involving beneficial microorganisms. Fi­nally, section 1.4 outlines the contents of this book. 1.1 Naming the partners 1.1.1 The insects

­ ere is overwhelming molecular and morphological evidence that the insects Th are a monophyletic group, comprising some 28 o­ rders grouped within five subclasses (­table 1.1 and fig. 1.1A) (Misof et al., 2014). The common ancestor of insects was likely terrestrial, and the earliest fossil insects are from the early Devonian period (ca. 410 million years ago). The class Insecta, together with the classes Collembola (springtails), Protura (coneheads), and Diplura (two-­ pronged bristletails), comprise the subphylum Hexapoda, which is defined by the possession of three pairs of thoracic legs; insects differ from other hexapods in having external mouthparts. The s­ ister group of the Hexapoda is an obscure group of marine Crustacea, the class Remipedia, with fewer than 30 known species that are apparently restricted to coastal aquifers. In other words, the insects have evolved from within the Crustacea (crabs, lobsters, shrimps, ­etc.). Many aspects of the biology of insects are ­shaped by their body plan, comprising a segmented body with paired jointed appendages (fig. 1.1B), supported by a versatile exoskeleton. This body plan permits rapid and precisely controlled movement, which has facilitated the evolution of complex be­hav­ior. It is also well suited to small size, which is associated with high reproductive rates and specialization to a multitude of ecological niches unavailable to larger animals. In addition to ­these ancestral traits, two evolutionary innovations account for the success of the insects. The first innovation was the origin of flight early in the diversification of the group, with the oldest fossils of winged insects at 324 million years ago in the early Carboniferous era. All extant insects

Dragonflies and damselflies

Odonata

Cockroaches and termites

Earwigs Webspinners Rock crawlers

Mantises Heelwalkers

Blattodea

Dermaptera Embioptera Grylloblattaria

Mantodea Mantophasmatodea

Subclass Polyneoptera

Mayflies

Jumping bristletails Silverfish and firebrats

Trivial name

Ephemeroptera

Subclass Palaeoptera

Zygentoma

Archaeognatha

Subclass Apterygota2

Insect order1

­Table 1.1. Classification of insects

2.5 × 103 ≤20

2 × 103 5 × 102 32

7.5 × 103

6 × 103

3 × 103

6 × 102

5 × 102

Estimated number of species

Cockroaches are omnivores or saprophages in leaf litter and soil; termites are “social cockroaches” that feed on wood or soil Mostly nocturnal scavengers or omnivores Gregarious scavengers, mostly tropical Flightless insects restricted to high altitudes in Northern Hemi­sphere Diurnal predators Flightless predators, only known in south and east Africa

Aquatic nymphs and short-­lived adults (often less than a day) Aquatic nymphs; adults are aerial predators

Nocturnal insects in leaf litter and ­under stones

Habits

(continued)

A few cockroach species are pests of h­ uman habitations; some termites are pests of wooden buildings and other infrastructure

Examples of significance to ­humans as pests and beneficials

Grasshoppers and crickets Stick insects Stoneflies

Angel insects

Orthoptera

Phasmatodea Plecoptera

Zoraptera

Bark and true lice

Thrips

Psocodea

Thysanoptera

Coleoptera

Beetles

Subclass Endopterygota

True bugs

Hemiptera

Subclass Paraneoptera

Trivial name

Insect order1

­Table 1.1. (continued)

3.9 × 105

Terrestrial and aquatic; scavengers, predators, and herbivores

Bark lice in soil, leaf litter, bird nests ­etc.; true lice are ectoparasites of birds and mammals Feed on fungi or plants

1 × 104 6 × 103

Mostly plant-­feeding, especially on plant sap; some predators and blood-­feeders

Mostly diurnal herbivores (some species are partly or entirely carnivorous) Slow-­moving herbivores, mostly nocturnal Aquatic nymphs are predatory or herbivorous; adults are short-­lived and herbivorous or nonfeeding Gregarious inhabitants of rotting wood, leaf litter, ­etc.; scavengers, especially on fungal hyphae

Habits

1 × 105

 ≤50

3 × 103 4 × 103

2 × 104

Estimated number of species

Major crop pests; also predators used in crop pest management

Major crop pests (aphids, whiteflies, planthoppers, e­ tc.) and medical pests (bedbugs, kissing bugs) True lice are parasites of poultry, livestock and ­humans; include vectors of vari­ous disease agents Few species are major horticultural pests

Gregarious grasshoppers (locusts) are major pests of crops and vegetation

Examples of significance to ­humans as pests and beneficials

Ants, bees, and wasps

Butterflies and moths Scorpionflies

Alderflies

Lacewings

Snakeflies

Fleas

Stylops Caddisflies

Hymenoptera

Lepidoptera

Megaloptera

Neuroptera

Raphidoptera

Siphanoptera

Strepsiptera Trichoptera

Herbivorous, predatory, or parasitic, including many parasitoids; sociality evolved multiple times Mostly herbivorous larvae (caterpillars) and nectar-­feeding adults Generalist feeders in damp woodlands; few predators Aquatic carnivorous larvae; nonfeeding adults Mostly predators and crepuscular or nocturnal Mostly predators; terrestrial in temperate woodland Adults are wingless ectoparasites of mammals and occasionally birds (larvae feed on detritus in host nest or den) Endoparasites of other insects Filter-­feeding aquatic larvae; nocturnal adults feed on nectar or nonfeeding

1.5 × 105

6 × 102 1.5 × 104

2 × 103

2 × 102

5.5 × 103

4 × 102

4 × 102

1.6 × 105

Larvae are terrestrial or aquatic; adults predominantly terrestrial. Very diverse habits: scavengers, herbivores, predators, parasites, and parasitoids.

1.6 × 105

Vectors of vari­ous disease agents

Herbivorous larvae are major agricultural pests; many adults are impor­tant pollinators

Agricultural pests, e.g., tephritids, crane fly larvae; vectors of medically impor­tant disease agents, e.g., mosquitoes, sandflies; urban pests, e.g., h­ ouse fly, blow fly. Beneficial pollinators and predators of agricultural pests, e.g., syrphids Impor­tant pollinators; ­house­hold pests

2

Subclass Apterygota may be artificial, with evidence that Zygentoma is more closely related to other insects than to Archaeognatha.

The relationships among most of the subclasses are resolved, but some phyloge­ne­tic prob­lems remain. The five major ­orders (with >100,000 described species) are indicated in bold.

1

Source: The classification of Misof et al. (2014) is displayed ­here and used throughout this book.

Mecoptera

Flies

Diptera

2 3

Endopterygota

Class INSECTA

1

Remipedia (a class of Crustacea) Protura, Collembola & Diplura Apterygota* Paleoptera Polyneoptera Paraneoptera Pterygota

Within the clade Pancrustacea

A

Hexapoda

6  c h a p t e r 1

B

3 segments HEAD 5–7 fused segments; mouthparts, eyes, 1 pair of antennae

up to 11 segments

THORAX 1 pair of legs per segment, 1 pair of wings on 2nd and 3rd segment of Pterygota

ABDOMEN no legs or wings (but other appendages; e.g., sexual structures, gills, cauda in some insects)

figur e 1.1. Insects. A. Insect relationships. The class Insecta is assigned to the subphylum Hexapoda (arthropods with six thoracic legs). The ­sister group of the Hexapoda (Remipedia) is a class within the Crustacea, meaning that the Crustacea is paraphyletic with re­spect to the Hexapoda; the clade Pancrustacea has been erected to encompass both Crustacea and Hexapoda. 1: origin of insects; 2: origin of winged insects (Pterygota, comprising four subclasses, underscored in the figure); 3: origin of complete metamorphosis (morphologically dif­fer­ent larval and adult stages). The insect o­ rders are listed in t­ able 1.1. The insect body plan comprising three regions: the head, thorax, and abdomen.

other than the jumping bristletails and silverfish have wings or have evolved from winged insects; a few groups, including the parasitic lice (within the order Psocodea) and fleas (order Siphonaptera), have secondarily lost their wings. The second key innovation was complete metamorphosis, also known as holometabolism, which evolved in the common ancestor of the subclass Endopterygota. All insects of this subclass have an immature stage (larva) that is morphologically and functionally dif­fer­ent from the adult, and a quiescent, nonfeeding stage (the pupa) within which the internal tissues are remodeled to generate the adult form. Insects other than the Endopterygota are described as hemimetabolous. Th ­ ese insects have no pupal stage, and in most hemimetabolous insects the juvenile stages, usually described as nymphs, are

I n t r o du c t i o n  

7

morphologically and ecologically similar to the adult, apart from the lack of functional reproductive organs and wings. The insect o­ rders vary widely in their taxonomic diversity, from fewer than 100 described species in three ­orders to more than 100,000 species in five ­orders: Coleoptera (beetles), Lepidoptera (butterflies and moths), Diptera (true flies), Hymenoptera (ants, bees, and wasps), and Hemiptera (true bugs, including cicadas, aphids, planthoppers, and shield bugs). Collectively, t­ hese five ­orders account for 91% of all insect species, and our knowledge of most aspects of insect biology derives predominantly from research on representatives of t­ hese groups. 1.1.2 The microorganisms

Whereas insect is a phyloge­ne­tically meaningful category (section 1.1.1), microorganism is an informal term that, in common parlance, refers to organisms that cannot be seen by the h­ uman eye. B ­ ecause most very small organisms are unicellular, biologists describe taxa that are unicellular for all or most of their life cycle as microorganisms. Three instances illustrate the conventions that define the scope of the term microorganism. First, the smallest adult insects, including Nanosella (Coleoptera: Ptiliidae), a genus of beetles of length 300 µm, and vari­ous parasitic wasps (140–300 µm long), are not regarded as microorganisms, even though they are smaller than some unicellular microorganisms, such as many amoebae (Polilov, 2015). The fungi provide a second example. The fungi include both unicellular forms, e.g., yeasts, and multicellular forms comprising a network of hyphae of varying complexity. The unicellular fungi and fungi that form small or structurally s­ imple mycelia (including taxa such as Candida species, which are dimorphic, i.e., with both yeast and hyphal growth forms) are generally considered as microorganisms, even though the mycelia can often be seen with the naked eye; but t­ here is no consensus ­whether fungi that form complex macroscopic structures, such as the fruiting bodies of many fungi of the phylum Basidiomycota (mushrooms, toadstools, e­ tc.), should be treated as microorganisms or not. My final example is the viruses, which fit to the definition of microorganisms by the criterion of size (they are 0.02–0.4 µm in dimension) but are, arguably, not living organisms at all ­because they require the cellular machinery of a host cell for reproduction. Nevertheless, viruses are impor­tant ­drivers of certain interactions between insects and microorganisms, as is considered in several contexts in this book.

8  c h a p t e r 1 A

B Bacteria

Archaea

Eukaryota Bacteria

rRNA gene-based tree of life

Archaea

Eukaryota

Genome-based tree of life

figure 1.2. Relationships among living organisms. The Bacteria and Archaea are all microorganisms, and most eukaryotic phyla are also microorganisms. A. Tree based on ribosomal RNA (rRNA) sequences. B. Tree based on the sequence of multiple concatenated genes.

Most organisms are microorganisms. Phyloge­ne­tic analyses based on the small subunit ribosomal RNA gene yield three domains of organisms: Bacteria and Archaea, all of which are microorganisms, and the Eukaryota, most of which are microorganisms (fig. 1.2A). However, the validity of the three-­ domain scheme is in doubt. This is ­because analyses using the sequence of multiple concatenated genes, e.g., genes coding ribosomal proteins, yield two domains, with the Eukaryota branching from the Asgard lineage within the Archaea (fig. 1.2B) (Imachi et al., 2020). Dating t­ hese very ancient evolutionary events is extremely difficult, but ­there is some consensus that the first microbial life evolved at 3.6–3.9 billion years ago, and that microbial eukaryotes evolved at ca. 1.8 billion years ago. The eukaryotic microorganisms are often referred to as protists, the Protista or Protozoa, but t­ hese terms describe the grade of organ­ization and have no phyloge­ne­tic meaning. The several groups of large, multicellular eukaryotes (i.e., nonmicrobial eukaryotes), including the animals, terrestrial plants, and red algae, evolved from dif­fer­ent microbial lineages. The impor­tant implication for insect-­microbial interactions is that insects evolved and diversified in a world that had been dominated by microorganisms for more than 3 billion years. Which taxa in the ­great diversity of microorganisms associate with insects? Metagenomic analyses (i.e., sequencing all the DNA in a sample) assign most microbial sequences to Bacteria in most insect samples. The diversity and taxonomic identity of the microorganisms vary widely, but several phyla of Bacteria are well represented: Proteobacteria, Firmicutes, Actinobacteria, and Bacteroidetes. Relatively few Archaea are known to be associated with animals, including insects, and the principal representatives are methanogens, especially of the

I n t r o du c t i o n  

9

order Methanobacteriales within the clade Euryarchaeota—­found, for example, in the guts of termites and cockroaches (Blattodea) and scarab beetles (Coleoptera: Scarabaeidae). However, the incidence and diversity of Archaea in insects and other animals may be underestimated, linked to the difficulties in culturing many archaeal taxa and inadequate molecular resources to identify and quantify Archaea (Borrel et al., 2020). Among the eukaryotic microorganisms associated with insects, the fungi are best studied. Yeasts are very commonly detected in insect samples. The term yeast refers to unicellular fungi, and most yeasts are members of the phylum Ascomycota, while a small number of Basidiomycota also display the yeast growth form. In addition, vari­ous insects enter into specialized associations with specific Basidiomycota that form extensive mycelia and large fruiting bodies, e.g., Termitomyces (Lyophyllaceae) associated with fungus-­growing termites (subfamily Macrotermitinae) and Leucocoprinus gongylophorus (Agaricaceae) associated with leafcutter ants of the genus Atta (Hymenoptera: Formicidae). Other eukaryotic partners of insects include trypanosomatids (the order Trypanosomatida within the phylum Euglenozoa), which, although widely viewed as pathogens, have been reported in healthy insects, especially Drosophilidae (Diptera) and Hemiptera (Podipaev, 2001); and obligately anaerobic ciliates (phylum Ciliata) and flagellate protists (of the phyla Parabasalia and Preaxostyla) that inhabit the anoxic guts of some insects, notably vari­ous cockroaches and termites (Blattodea) (Brune, 2014). 1.1.3 Describing the taxonomy of insect and microbial partners

The taxonomic position of species of insects and microorganisms described in this book is summarized ­after first mention of a species in each section. For insects, I provide the order and ­family, e.g., Drosophila melanogaster (Diptera: Drosophilidae). For most microorganisms, I provide the phylum and ­family, e.g., Bacillus subtilis (Firmicutes: Bacillaceae) and Saccharomyces cerevisiae (Ascomycota: Saccharomycetaceae). Exceptionally, members of the bacterial phylum Proteobacteria are described by class (which is a Greek letter-­ Proteobacteria), e.g., Escherichia coli (γ-­Proteobacteria: Enterobacteriaceae). Specific issues of nomenclature arise for the many unculturable microoorganisms associated with insects. Formal taxonomic assignments require isolation of microorganisms into culture, and unculturable microorganisms are referred to as Candidatus taxa. For example, whiteflies bear the unculturable bacterium Candidatus Portiera aleyrodidarum. For con­ve­nience, I follow the

10  c h a p t e r 1

convention of abbreviating ­these Candidatus names to the genus name, e.g., Portiera in whiteflies. A further taxonomic issue concerns microorganisms that have been associated with insects for many millions of years. The phyloge­ne­tic status of some microbial symbionts is uncertain or disputed, with par­tic­u­lar concern that the published assignments of some ancient bacterial symbionts with very small genomes may be an artifact of long branch attraction (the erroneous interpretation that two or more distantly related taxa are closely related b­ ecause they are dif­fer­ent from other taxa in the phylogeny). For example, the intracellular bacterial symbiont of most aphids, Buchnera aphidicola, can be assigned to the order Enterobacterales (γ-­Proteobacteria) with reasonable confidence, but a more precise placement of this taxon is not assured (although some authorities assign it to the ­family Enterobacteriaceae). In this book, I indicate this uncertainty by the designation of Buchnera as “γ-­Proteobacteria: Enterobacterales incertae sedis” (incertae sedis means “uncertain placement”). Similarly, the bacterial symbiont of cicadas is Sulcia muelleri (Bacteroidetes: Flavobacteriales incertae sedis), but the bacterial symbiont of whiteflies, for which phyloge­ne­tic signal to the ­family level is assured, is referred to as Portiera aleyrodidarum (γ-­Proteobacteria: Halomonadaceae). With the application of increasingly sophisticated phylogenomic methods, many of the current phyloge­ ne­tic uncertainties and assignments of microbial taxa to incertae sedis status are likely to be resolved in the coming years. 1.2 Insect habitats for microorganisms Most insect-­associated microorganisms are localized to the cuticle, gut, hemolymph, or within cells of insects. In this section, the biology of ­these insect habitats is described. The characteristics of insect-­microbial associations in the dif­fer­ent locations are considered in detail in chapter 2. 1.2.1 The external coverings

The insect body is bounded by a light, waterproof cuticle (fig. 1.3) that protects the under­lying soft tissues against mechanical damage, desiccation, and microbial colonization. The cuticle is a composite material of chitin fibrils in a protein matrix, and its mechanical properties (strength, flexibility, ­etc.) are dictated largely by the degree of protein cross-­linking by quinones. The cuticle is largely inextensible, and sustained growth of an immature insect (nymph or

I n t r o du c t i o n  

11 Epicuticle Procuticle Microvilli Occluding junctions

epi

epi

epi

epi Basal lamina

figure 1.3. The insect integument. The insect body is bounded by a layer of epidermal cells (epi), which secrete most of the constituents of the acellular cuticle. The cuticle comprises the procuticle of chitin fibrils and cuticular proteins, many of which are cross-­linked to form a tough, inextensible structure; and the epicuticle, which confers waterproofing and is made of lipoproteins, waxes, hydrocarbons, ­etc. The epidermal cells adhere closely via lateral occluding junctions (including adherens junctions and septate junctions) to form a coherent sheet of tissue that rests basally on the noncellular basal lamina.

larva) requires the insect to undergo molting. This is a complex pro­cess that requires the formation of a new cuticle u­ nder the old cuticle, then shedding of the old cuticle (a pro­cess known as ecdysis), and fi­nally the expansion and then hardening of the new cuticle. The insect cuticle is, generally, an inhospitable habitat for microorganisms. It has low availability of w ­ ater and nutrients, and it is protected further by both microbicidal secretions from the under­lying epidermal cells and specialized glands (Yek and Mueller, 2011) and nanoscale cuticular structures that impede microbial adhesion (Watson et al., 2017). Microbial cells detected on the insect cuticle are generally transient (Zhukovskaya et al., 2013), being dislodged by insect grooming and lost with cuticle shedding at ecdysis. Despite the general inhospitality of the insect cuticle for microbial colonization, regions of the cuticle in some adult insects are specialized to facilitate colonization by specific microorganisms. For example, the cuticle of many attine ants (Hymenoptera: Formicidae of subtribe Attini) is modified to bear a dense colony of the bacterium Pseudonocardia (Actinobacteria: Pseudonocardiaceae) (see section 4.5.1). Other insects have cuticular structures that function specifically for microbial storage and transport. ­These structures are known as mycangia (singular, mycangium), following their initial description in insects harboring fungal symbionts. However, some mycangia contain both fungi and bacteria or exclusively bacteria (see sections 2.1.3 and 2.1.5). The

12  c h a p t e r 1

structural organ­ization of mycangia varies from ­simple pits in the cuticle surface to anatomically complex invaginations with a narrow opening to the exterior that is often bounded by protective setae (chitinous hairs). The diversity of mycangia has been studied most extensively in ambrosia beetles (Coleoptera: Curculionidae of subfamilies Scolytinae and Platypodinae), in which ­these structures have evolved multiple times (Hulcr and Stelinski, 2017). 1.2.2 The insect gut

As for most animals, the gut in the ­great majority of insects is a tubular structure through which ingested food is passed from mouth to anus. Although insects are exceptionally diverse in feeding habits, the fundamental plan of the gut is conserved, comprising a foregut, midgut, and hindgut (fig. 1.4). The foregut is the principal site of mechanical disruption and initial pro­cessing of food, while enzymatic digestion and nutrient assimilation is predominantly in the midgut, and the hindgut mediates selective absorption of ­water, ions, and some metabolites. The gut lumen in the foregut and hindgut is separated from the gut epithelium by cuticle that is continuous with the cuticle protecting the external surface of the insect. As for the external cuticle, the gut-­associated cuticle acts as a barrier to microbial colonization. When the insect molts, the gut cuticle is lost via the mouth and anus, and then replaced by new cuticle in the subsequent life stage of the insect. The conditions and resources in the insect gut vary widely among species, with developmental age (particularly differing very substantially between the larva and adult of holometabolous taxa) and with location in the gut (Engel and Moran, 2013). In small insects, the gut is predominantly oxic or mildly hypoxic, due to the diffusion of oxygen from the gut epithelium, which is aerobic and well supplied with tracheae, but some regions of the gut lumen of large insects can be anoxic. For many insects, the variation in pH in the gut lumen is relatively small, in the range of 6–8 pH units, but extremely alkaline conditions (pH >11) occur, for example, in the midgut of lepidopteran larvae and the hindgut of some soil-­feeding termites, and a portion of the midgut of many Diptera is highly acidic, at pH 7,000 chambers in Atta laevigata. The best-­known and most successful attine ants are the leafcutter ants. ­These species cut and deliver fresh leaves, grass, and flowers to their

28  c h a p t e r 2

fungal gardens (Li et al., 2021). The scale of this activity is so g­ reat that the leafcutter ants can play a defining role in structuring the ecol­ogy of many Neotropical forests and grasslands. The fungal partner of the leafcutter ants is a single species, Leucocoprinus gongylophorus (also known as Leucoagricus gongylophorus or Attamyces bromatificus), in the tribe Leucocoprineae (“parasol mushrooms,” Basidiomycota: Agaricaceae). L. gongylophorus is unknown outside attine ant nests. The fungal mycelium ramifies through the plant material, and many of the individual hyphae are swollen at the tip, forming nutrient-­rich structures known as gongylidia (De Fine Licht et al., 2014). The gongylidia are the sole food source for the larval ants and the principal diet of adult workers in the nest, but foraging workers also utilize food sources, such as plant sap, in the external environment. The association between leafcutter ants and L. gongylophorus is, however, just one of five modes of fungiculture in attine ants (fig. 2.1) (Mueller et al., 2018; Schultz and Brady, 2008). All other attine ants forage for dead plant material, such as fallen leaves, withered flowers, and seed husks, as well as insect frass; and only two ant genera other than leafcutters (Trachymyrmex and Sericomyrmex) cultivate L. gongylophorus with its nutrient-­rich gongylidia. Most of the remaining attine ants culture a variety of leucocoprinaceous species that, unlike L. gongylophorus, can also be found in the external environment (e.g., associated with leaf litter) where they produce fruiting structures and sexual spores. Most of ­these fungi grow in the ant nest as a mycelium but without gongylidia. Exceptionally, one leucocoprinaceous lineage adopts a unicellular “yeast” growth form, aggregating as tiny nodules (just 0.5 mm in dia­meter), in the garden, although it can grow as a mycelium and produce fruiting structures in the free-­ living condition and in culture. Yeast fungiculture is ­adopted by a single group of ants, the rimosus group of the genus Cyphomyrmex. Fi­nally, some species of the genus Apterostigma have switched to a phyloge­ne­tically distant group of the Basidiomycota: coral fungi of the ­family Pterulaceae (Chapela et al., 1994), which grow as a mycelium in the ant nest and, as for most leucocoprinaceaous symbionts, can be found in the external environment. Although the association with attine ants is not generally obligate for the fungal partner (apart from L. gongylophorus, which has, so far, not been identified in the free-­living condition), the ants do not collect the fungus from the external environment. Instead, a foundress queen initiating a new nest takes a sample of fungus from her natal nest in the infrabuccal pocket, a filtering device in the oral cavity of all ants, and she uses this to inoculate the new fungal garden. Nevertheless, the ant and fungal partners do not display strict codiversification.

I n s e c t - M i c r ob i a l A s s o c i a t ion s  29

Cyatta, Kalathommyrmex Mycetarotes, Mycetosoritis Cyphomyrmex of rimosus group Cyphomyrmex Mycetophylax, Mycetosoritis Mycetagroicus Trachymyrmex, Sericomyrmex Acromyrmex Atta

Lower attine fungiculture with various Leucocoprineae Yeast fungiculture (single clade of leucocoprinaceous fungus with single-celled morphology) Trachymyrmex fungiculture Higher attine fungiculture with Leucocoprinus Leafcutter fungiculture gongylophorus

Dead plant matter, e.g., leaf debris, seeds, and insect frass

Apterostigma Myrmicocrypta Mycocepurus

Pterulaceous fungiculture (coral fungi Pterulaceae)

Mostly live-cut vegetation

Apterostigma of pilosom group

Substrate collected by ants for fungiculture

figure 2.1. Fungiculture in attine ants. The five modes of fungiculture (bold) are mapped to the phylogeny of attine ants (left) and type of substrate collected by ants for fungiculture (right). The fungal partners in all associations have a mycelial growth form, apart from the yeast fungiculture in Cyphomyrmex of the rimosus group. The fungal species L. gongylophorus is largely restricted to higher attines, but it has occasionally been reported in the nest of a lower attine; and ants of the higher-­attine genus Trachymyrmex can occasionally host leucocoprinaceous fungi other than L. gonglyophorus. [Redrawn from fig. 2 and information in ­table 1 of Mueller et al. (2018).]

­ ere is much shuffling of partners in the associations between lower attine Th ants and vari­ous species of leucocoprinaceous fungi, and also swapping of L. gongylophorus genotypes between species and genera of higher-­attine ants (Mueller et al., 2018). Fungus farming in the termites evolved in tropical rain forests of sub-­ Saharan Africa, and fungus-­growing termites are found t­ oday in African forests and savannas, as well as parts of Asia. ­These associations can degrade lignocellulose with near 100% efficiency, and they mediate the decomposition of up to 90% of dead plant material in some arid habitats. The fungal partners belong to a single fungal genus, Termitomyces (Basidiomycota: Lyophyllaceae), which is unknown in any habitat other than nests of the Macrotermitinae. Termitomyces is cultivated on fungal combs made of plant material macerated and molded into shape by the termites and placed in semienclosed chambers, called galleries, in the termite nest. The Termitomyces adopts three growth forms: a mycelium of hyphae ramifying through the plant material, nodules

30  c h a p t e r 2

(highly modified immature mushrooms containing asexual spores), and sexual mushrooms. The fungus gardens are sustained by foraging termite workers that collect leaf litter, woody detritus, and organic debris. The plant material is then transferred to workers that tend the fungal combs in the nest. ­These latter workers consume the material, and their feces of well-­macerated and partly digested plant material is the substrate for lignocellulose degradation by the Termitomyces in the fungal comb. As described in greater detail in section 4.1.1 and fig. 4.1B, the termites feed on both fungal nodules and fungal-­digested plant material. New termite nests are inoculated with basidiospores (sexual spores) produced by Termitomyces fruiting bodies that grow out of established termite colonies. The first foraging workers of new nests collect ­these spores from the environment to initiate the fungal garden; consequently, Termitomyces can be passed between dif­fer­ent termite species. Exceptionally, departing alate reproductives of Microtermes species and Macrotermes bellicosus ingest Termitomyces basidiospores in the parental nest and transport them in their gut to the new nest, where they are released via the feces. 2.1.3 Fungal ectosymbioses in bark beetles and ambrosia beetles

The bark beetles and ambrosia beetles are weevils (Coleoptera: Curculionidae) of the subfamilies Platypodinae and Scolytinae. The adult insects raise their young in tunnels, also known as galleries, bored into woody tissue. Most species use the trunk or branches of dead or d­ ying trees, but a minority attack living trees. For example, recent outbreaks of the mountain pine bark beetle Dendroctonus ponderosae that are linked to climate change have devastated pine forests in western North Amer­ic­ a, and a virulent fungal pathogen Raffaelea lauricola (Ascomycota: Ophiostomataceae) is transmitted by the ambrosia beetle Xyleborus glabratus to live laurels including avocado, causing laurel wilt. Many bark beetles and all ambrosia beetles are associated with mycelial fungi, which they carry from their natal gallery and inoculate into their incipient gallery. The fungus grows along the walls of the gallery and extends deep into the woody substrate, from which it extracts nutrients; and the adult beetles and their larval offspring feed on the fungus. The key difference between bark beetles and ambrosia beetles is that each has a distinctive mode of feeding (fig. 2.2) (Hulcr and Stelinski, 2017; Kirkendall et al., 2015). The galleries of bark beetles are localized to the phloem tissue,

I n s e c t - M i c r ob i a l A s s o c i a t ion s  31

CURCULIONIDAE (weevils)

feeding habit Phloeophagy Scolytinae

Bark beetles AND Ambrosia beetles

tunnel and feed on phloem tissue, or

Phloeomycophagy

feed on phloem tissue and fungi in tunnel

Xylomycetophagy

feed on “farmed” fungi grown on walls of tunnels in xylem tissues; some species also feed on wood

Platypodinae Ambrosia beetles

figure 2.2. Fungal ectosymbioses in bark beetles and ambrosia beetles. The terms bark beetle and ambrosia beetle refer to the food source of the beetles, as shown. Ambrosia beetles have evolved multiple times from scolytine phloeomycophagous bark beetles, with some species displaying intermediate or variable feeding habits, and also in the Platypodinae (which includes no known species with the bark beetle feeding habit). The Scolytinae and Platypodinae are prob­ably not closely related, and the ambrosial feeding habit has evolved convergently in the two subfamilies. Note that some entomologists describe all Scolytinae as bark beetles, and the Platypodinae as pinhole borers, irrespective of the feeding habit of the beetles. [Drawn from information in Hulcr and Stelinski (2017) and Kirkendall et al. (2015).]

which lies just beneath the bark of the tree and is the most nutritious layer in wood. Some bark beetles are associated with fungi, and t­ hese species feed on a mixture of phloem tissue and fungus, a feeding habit known as phloeomycophagy. The ambrosia beetles are xylomycetophagous, meaning that they bore into the xylem tissue of wood, not the phloem tissue, and they feed mostly or entirely on the mycelium and nutritious spores of the fungal growth. Other differences between bark beetles and ambrosia beetles are correlated with their difference in feeding habits. In all ambrosia beetles, the fungi are transferred from the natal gallery to the gallery newly bored by the adult insect via mycangia (inpocketings of the adult insect cuticle), but the source of the fungi in the galleries of bark beetle hosts is variable. Some bark beetle species have mycangia for transmission of fungi from the natal gallery, some fungi are transmitted via the surface of the exoskeleton (­whether or not the insect has mycangia), and some fungi associated with the galleries are derived from free-­living fungal populations. Linked to this difference, the association with the fungi is absolutely required for ambrosia beetles but not for bark beetles. The ambrosia beetles feed on the fungus and remove frass and debris from their galleries; if the adult dies, the fungal culture becomes contaminated by other fungi and bacteria, and the beetle brood dies. In contrast, experiments

32  c h a p t e r 2

comparing the growth of bark beetles on fungus-­infested and sterile wood demonstrate that the mycangial fungi promote larval growth and survival, but they are not obligate for ­these insects (Six, 2012). The fungal partners of bark beetles and ambrosia beetles are predominantly Ascomycota, particularly Ophiostomatales (e.g., Ophiostoma in bark beetles, Raffaelea in ambrosia beetles) and Microascales (e.g., Ceratocystis in bark beetles, Ambrosiella in ambrosia beetles). ­There is unambiguous evidence for multiple fungal origins of the symbiosis, including three origins with platypodine beetles and seven origins with the Scolytinae (Biedermann and Vega, 2020). In addition, phyloge­ne­tic data demonstrate clear patterns of specificity. While fungi are rarely transferred between dif­fer­ent clades of ambrosia beetles (i.e., dif­fer­ent evolutionary origins of the symbiosis), multiple instances of partner switching are evident in comparisons between congeneric species (Skelton et al., 2019). The association in ambrosia beetles is widely accepted as an instance of fungal agriculture. The beetle is interpreted to inoculate the fungus into the woody substrate of the tunnel that it has constructed, then cultivate the fungus, and harvest it as the principal or sole food source. In this re­spect, the ambrosia beetles display behavioral sophistication comparable to the eusocial attine ants and macrotermitine termites (section 2.1.2) without eusociality (although eusociality has been reported in some ambrosia beetle species). The relationship between bark beetles and fungi does not meet the stringent criteria for agriculture (section 2.1.1), and it is broadly equivalent to other ectosymbioses involving fungi that provide a nutritional ser­vice to the insect, described in the following section. 2.1.4 Fungal ectosymbioses without farming

Several groups of Coleoptera, in addition to the ambrosia beetles and bark beetles described in section 2.1.3, are associated with ectosymbiotic fungi that provide nutrition or protection for the insect. A nutritional function has been demonstrated for the lizard beetle Doubledaya bucculenta (Coleoptera: Erotylidae), which is endemic to Japan. When the adult female deposits an egg into the internode cavity of a bamboo plant, she inoculates the cavity with the fungus Wickerhamomyces anomalus (Ascomycota: Phaffomycetaceae) carried in a reservoir by the ovipositor. The fungus proliferates and lines the cavity wall, serving as a food source for the larva. Insects that are denied access to the fungal partner die as larvae (Toki et al., 2012).

I n s e c t - M i c r ob i a l A s s o c i a t ion s  33

Another fungus, Yarrowia sp. (Ascomycota: Dipodascaceae), has been implicated in the nutrition of burying beetles (Coleoptera: Silphidae of the subfamily Nicrophorinae). Th ­ ese insects utilize carrion that has recently died. The adult beetles manipulate the carcass into a ball with a central cavity, and then the female deposits her eggs in the cavity. The larvae remain in this location, tended by their parents. Yarrowia, together with a lower biomass of several bacteria (section 2.1.5), proliferate in a biofilm-­like matrix lining the central cavity, providing a food source for the larvae. When the matrix is experimentally scraped off the cavity, larval growth is significantly depressed, confirming the nutritional significance of the association for the insect (Shukla et al., 2016). The matrix microbiota is likely derived from the adult hindgut. In the ectosymbiosis involving leaf-­rolling weevils of the subfamily Attelabinae (Coleoptera: Attelabidae), the primary role of the fungal partner is to provide protection. The adult female weevil shreds portions of a plant leaf, from which she constructs a hollow leaf-­roll chamber. She then deposits a single egg into the leaf-­roll, together with a fungal inoculum from a ventral mycangium, and fi­nally snips the leaf tissue connecting the leaf-­roll and remainder of the leaf, allowing the leaf-­roll to fall to the ground. The fungal mycelium grows rapidly and is believed to protect the leaf-­roll from microbial-­ mediated decomposition and possibly predation. The fungus associated with one attelabine species, Euops chinensis, has been identified as Penicillium herquei (Ascomycota: Aspergillaceae); one of its secondary metabolites, (+)-­scleroderolide, has been implicated as an antimicrobial agent (Wang et al., 2015). Among the Hymenoptera, vari­ous fungal ectosymbioses other than the fungus-­cultivating attine ants (section 2.1.2) have evolved, notably in other ants and in woodwasps. Ectosymbioses in ants involve ascomycote fungi maintained in plant domatia (hollow plant structures used as ant nest sites) or combined with plant fragments and honeydew to form a cardboard-­like construction material, called carton, built by ants. Confirming the early lit­er­ a­ture on myrmecophytes (“ant plants”), recent studies have shown that the domatia of many plants bear fungal patches. ­These fungi have been identified as Ascomycota of the order Chaetothyriales (informally known as black yeasts), and they are consumed by the ants as a source of nutrients (Blatrix et al., 2012; Defossez et al., 2009). Fungi of the same order are used by Lasius ants to construct nest walls (Schlick-­Steiner et al., 2008) and by vari­ous tropical arboreal ants to construct enclosed runways, also known as galleries, along tree branches (Nepel et al., 2014).

34  c h a p t e r 2

For the woodwasps (Hymenoptera: Siricidae and Xiphydriidae), the adult female inserts her eggs into dead or d­ ying wood via a long, needle-­like ovipositor, together with an inoculum of fungus derived from a mycangium associated with the ovipositor. On hatching, the larvae bore a tunnel into the xylem wood, and the fungus grows along the tunnel walls. The larvae secrete enzymes that degrade fungal material (both mycelium and asexual spores) and wood fragments, and they consume the liquid products of their external digestion (Thompson et al., 2014). Following pupation, the adult wasp emerges, the female with fungal inoculum associated with her ovipositor. The fungus in several siricid species has been identified as a basidiomycote white rot fungus of the genus Amylostereum (Basidiomycota: Stereaceae). Ectosymbiotic fungi that grow to enclose insects, protecting them against natu­ral enemies, particularly parasitoids, have been investigated in gall midges (Diptera: Cecidomyiidae) and scale insects (Hemiptera of superfamily Coccoidea). Many cecidomyiid midges induce plant galls, but members of two tribes, Asphondyliini and Lasiopterini, are exceptional in that the protective gall is not made of plant tissue but constructed by fungal mycelium (Roh­ fritsch, 2008). As a result, the insect does not interact directly with the plant but via an intervening fungal layer, which provides both mechanical protection and a source of food for midge larvae within the gall. Where investigated, the fungal partners are Botryosphaeria (Ascomycota: Botryosphaericaceae). Fungal conidia are carried in ventral pockets close to the ovipositors of females, and they are inoculated into the plant along with deposited eggs. The galls of Asphondyliini are restricted to the surface tissues of plant organs, usually flower buds and fruits, while the Lasiopterini tend to form galls in plant stems and their fungal partner often penetrates deeply into plant tissue. I conclude this section on fungal ectosymbionts with the remarkable association between armored scale insects (Hemiptera: Diaspididae) and Septobasidium (Basidiomycota: Septeobasidiaceae). As first described in the 1930s by John Couch (1931), the fungus infects a single insect, which continues to feed from the plant, delivering nutrients to the fungus via fungal haustoria inserted into the insect hemocoel. Fungal growth is predominantly external to the infected insect, and it can form a substantial fungal mat that encloses siblings of the infected insect. In this way, the fungus shields the insects against parasitoids and desiccation. The single infected insect fails to grow or develop, but it may, in princi­ple, derive high inclusive fitness from the benefit derived by its siblings. However, this intriguing scenario has not been demonstrated definitively and, in some associations, the fungal partner produces very ­limited

I n s e c t - M i c r ob i a l A s s o c i a t ion s   35

external hyphae that confer no protection, suggesting that the fungus might be a parasite. ­There are some natu­ral history reports of apparently healthy aphids (Hemiptera: Aphididae) and mealybugs (Hemiptera: Pseudococcidae) enclosed by dense fungal mycelium (Biedermann and Vega, 2020), raising the possibility of multiple evolutionary origins of relationships similar to the Septobasidium-­diaspidid scales. 2.1.5 Bacterial ectosymbioses

Two ectosymbioses with bacteria have been studied intensively, both involving Hymenoptera associated with Actinobacteria that produce antimicrobial secondary compounds. ­These are the relationships between beewolves (Crabronidae of the tribe Philanthini) and Streptomyces sp. (Streptomycetaceae), and between attine ants (subtribe Attina) and Pseudonocardia sp. (Pseudonocardiaceae). The ectosymbiont of beewolves is a single clade of Streptomyces, Candidatus S. philanthi, that evolved ca. 68 million years ago (Kaltenpoth et al., 2014). The bacteria are localized to specialized glands on the antennae of the adult female insect. The function of the bacteria is intimately related to the natu­ral history of t­ hese insects. The adult female digs an under­ground nest comprising a tunnel with multiple brood chambers. She deposits an egg into each chamber, together with one or several para­lyzed bees as food for the larval offspring, and then she strokes the ceiling of the brood chamber with her antennae, releasing an inoculum of Streptomyces. As the larva develops, it spins a silk cocoon, incorporating the Streptomyces onto the cocoon surface. Fi­nally, the Streptomyces symbiont is transferred to the antennal glands of the adult female as she emerges from the cocoon, ensuring transmission. Experimental analyses have demonstrated that the Streptomyces associated with beewolves produce a cocktail of antibiotics, comprising streptochlorin and eight piericidin derivatives, that suppresses growth of opportunistic pathogenic fungi (fig. 2.3A) residing in the soil. ­These fungi would other­wise infect and kill the larva in the humid conditions in the brood chamber (Engl, Kroiss, et al., 2018; Kaltenpoth et al., 2005). The actinobacterial symbiont Pseudonocardia sp. of attine ants is borne on the ant cuticle, and its filamentous growth form appears as “a powdery crust” to the naked eye (Currie et al., 1999). The precise location of the bacteria varies with ant species, for example, ­under the forelegs of Apterostigma and on the ventral surface of the prothorax of Acromyrmex. In many ant species,

36  c h a p t e r 2

Beewolf (Philanthini)

A

Opportunistic fungal pathogens

Streptomyces

B Leucoagaricus (fungal ectosymbiont)

Metarhizium anisopliae

Escovopsis (specialist fungal pathogen of Leucoagaricus)

Attine ant Pseudonocardia

KEY Negative interaction xxx yyy

Partners in ectosymbiosis

figure 2.3. Protective functions of ectosymbiotic bacteria of insects. A. Streptomyces philanthi protects its beewolf host from opportunistic fungal pathogens. B. Pseudonocardia sp. on the cuticle of attine ants both protects the fungus Leucoagaricus cultivated by the ants against a specialist fungal pathogen, Escovopsis, and protects the ant against the fungal pathogen Metarhizium anisopliae.

the cuticle is modified to form crypts or tubercles housing the bacteria (Li et al., 2018). Phyloge­ne­tic analyses indicate that the association evolved in the ancestor of attine ants but has been lost in several lineages, including the leafcutting ants of the genus Atta (Li et al., 2018). The Pseudonocardia symbiont of attine ants produces a range of antimicrobials, many allied to polyketides (Van Arnam et al., 2018). The targets of ­these compounds are fungi, including both insect pathogens, such as Metarhuzium anisopliae (Ascomycota: Clavicipitaceae), and a specialized fungal parasite of the fungal

I n s e c t - M i c r ob i a l A s s o c i a t ion s  37

ectosymbiont, Escovopsis (Ascomycota: Hypocreaceae) (fig. 2.3B) (Currie et al., 1999; Mattoso et al., 2012). Apart from the well-­studied associations in beewolves and attine ants, the lit­er­a­ture on bacterial ectosymbioses in insects is sparse. This may arise, at least partly, from a lack of study ­because bacterial biofilms associated with insect nests are less con­spic­uo­ us than the mycelial mats and fruiting bodies of many fungi. Molecular analyses that target the bacteria-­specific 16S ribosomal RNA (rRNA) gene have revealed that the mycangia of ambrosia beetles, the fungal gardens of attine ants and macrotermitine termites, and the feeding chamber of burying beetles contain bacteria, including taxa detected in multiple samples from the same insect species. For example, the fungal gardens of the leafcutting ant Atta texana reliably bear Acinetobacter (γ-­Proteobacteria: Moraxellaceae), Corynebacterium (Actinobacteria: Corynebacteriaceae), Propionibacterium (Actinobacteria: Propionibacteriaceae), and Pseudomonas (γ-­Proteobacteria: Pseudomonadaceae). ­These bacteria are also found in the infrabuccal pocket of foundress queens (Meirelles et al., 2016), suggesting that the bacteria are transmitted between nests along with the fungal ectosymbiont. Similarly, the matrix lining the cavity constructed by burying beetles in carrion (section 2.1.3) includes several bacteria, e.g., Dysgonomonas (Bacteroidetes: Dysgonomonadaceae), Vitreoscilla (β-­Proteobacteria: Neisseriaceae), Providencia (γ-­Proteobacteria: Morganellaceae), as well as Yarrowia fungi, and t­ hese bacterial communities are of very dif­fer­ent composition from the bacteria associated with beetle-­f ree carrion (Shukla et al., 2016). ­W hether and how t­ hese ectosymbiotic bacteria contribute to insect fitness is not well understood. Th ­ ere are indications that the bacteria associated with leafcutting ants may detoxify plant allelochemicals (Francoeur et al., 2019), and it has been suggested that the bacteria associated with burying beetles may produce antimicrobials that suppress nonspecific putrefaction of the cadaver (Shukla et al., 2016). 2.2 Gut symbioses 2.2.1 The source and fate of microorganisms in insect guts

The gut lumen of the ­great majority of insects contains microorganisms. The source of ­these microorganisms is remarkably uniform—­gut microorganisms are generally acquired orally, usually in food—­but the fate of t­ hese ingested

38  c h a p t e r 2 4 Gut epithelium Noncellular layer*

2

1

Microbe ingested into gut lumen Noncellular layer*

3

5

Gut epithelium FATE OF INGESTED MICROBIAL CELLS

1. Pass through the gut with bulk flow of food 2. Killed by gut conditions (e.g., pH, oxygen tension), digestive enzymes, or immune factors 3. Retained in gut lumen (external or internal to the noncellular layer*) and may proliferate 4. Internalized into cells of the gut epithelium or passed across the gut wall to internal tissues 5. Retained microbes or their progeny are shed and eliminated * The noncellular layer comprises an extension of the chitinous exoskeleton in the foregut and hindgut and, in many insects, the peritrophic envelope in the midgut (see section 1.2.2 for details).

figure 2.4. Alternative fates of microorganisms ingested into the insect gut lumen.

microbial cells is highly variable. Many ingested cells pass through the gut with the bulk flow of food to the feces, while other microorganisms are killed in the gut lumen or retained in a v­ iable condition (fig. 2.4). Many per­sis­tent microorganisms escape from the flow of food by adhering to the gut wall or colonizing ceca or other diverticula of the gut, and they can be retained—­ often with proliferation—­for variable periods of time, ranging from a few hours to the life span of the insect. Very commonly, individual cells from ­these per­sis­tent populations are shed into the gut lumen and eliminated in the feces. Other per­sis­tent microorganisms are planktonic in the gut lumen. ­These forms are generally motile, enabling them to adjust their position actively. In very many insects, ingested food and microorganisms are separated from the gut epithelium of the foregut and hindgut by an extension of the chitinous exoskeleton, and from the midgut epithelium by a peritrophic envelope (fig. 2.4 and section 1.2.2). Most microorganisms are restricted to the gut lumen. However, microorganisms have been reported in the ectoperitrophic space, i.e., between the peritrophic envelope and midgut epithelium (e.g., in some termites and in adult female mosquitoes). In addition, some microorganisms can also gain access to cells of the gut epithelium (section 2.2.3) or cross the epithelium to internal tissues (fig. 2.4).

I n s e c t - M i c r ob i a l A s s o c i a t ion s  39

2.2.2 The diversity and microorganisms in insect guts

Most of the information on the composition of the gut microbiota in insects has come from 16S rRNA gene amplicon sequencing studies of bacteria, with taxonomic diversity quantified as the number of OTUs (operational taxonomic units) at the 97% sequence identity. Very roughly, an OTU at the 97% identity cutoff can be treated as equivalent to a bacterial species or group of closely related species. Recent advances in sequence analy­sis have made it pos­si­ble to resolve single-­nucleotide differences in amplicon sequences, and, increasingly, taxonomic diversity is being reported as amplicon sequence variants (ASVs). Although the lit­er­a­ture on ASV diversity is, at the time of writing, l­ imited, it is already apparent that many more bacterial ASVs than OTUs are recovered from gut microbiomes of many animals, including insects. This indicates that the microbial community comprises multiple closely related strains that cannot be detected at the 97% cutoff routinely used in OTU analyses. However, the lit­er­ a­ture using ASVs is still relatively sparse and, in this chapter, I focus on OTUs to describe the broad features of gut microbiota diversity and composition. The gut of most insects bears relatively few bacterial OTUs, at least compared to mammals. For example, all of the 30 lepidopteran species in the study of Paniagua Voirol et al. (2018) and most of the 62 species of six ­orders (Blattodea, Coleoptera, Diptera, Hymenoptera, Lepidoptera, and Neuroptera) analyzed by Colman et al. (2012) comprised fewer than 100 bacterial OTUs, while mammals generally have 1,000 or more bacterial OTUs. A cross-­species metanalysis of the gut microbiota diversity in animals of body weight varying over 12 ­orders of magnitude, including 30 insect samples from 16 species, revealed that larger animals host a wider diversity of gut microorganisms (fig. 2.5) (Sherrill-­Mix et al., 2018). Larger animals have larger guts, and a large gut is predicted to promote microbial diversity in two ways. First, larger guts are likely to be more heterogeneous, increasing microbial diversity by providing a greater diversity of niches. Second, larger guts can accommodate higher absolute numbers of any microbial taxon, reducing the likelihood that random fluctuations in populations result in extinction of taxa in the gut. As Sherrill-­Mix et al. (2018) note, this pattern fits to the expectations of the species-­area relationship of island biogeography theory. Each insect (or other animal) is analogous to an island, and the number of microbial species is determined by the number of niches (dictated by island size) and the rates of colonization and extinction. A further determinant of the diversity of the microbial communities in animals is gut physiology. Microbiomes of high diversity are found in animal guts

40  c h a p t e r 2 3.5

Log10 (number of OTUs)

3.0 2.5 2.0 1.5 1.0

Animals other than insects Insects

0.5 0.0 –4

–2

0

2

4

6

8

10

Log10 (body weight) figure 2.5. Relationship between body size and number of gut bacterial OTUs in 30 insect samples and 235 samples from other animals. The insects (with number of samples per insect species in parentheses) are Blattodea (Gromphadorhina portentosa (2), Mantis religiosa (1)), Coleoptera (Tenebrio molitor (4), Megassoma actaeon (1)), Hemiptera (Boisea trivittata (2), Cimex lectularius (3)), Hymenoptera (Apis mellifera (1), Melissoda birnaculata (1), Anthidium manicuatum (1), Bombus birnaculatus (2), Vespula maculifrons (1)), Diptera (Drosophila melanogaster (4), Lucilia sericata (1)), Lepidoptera (Acronicta americana (1), Manduca quinquemaculata (2)), and Orthoptera (Acheta domesticus (3)). The regression line has a slope of 0.11 (adjusted r2 = 0.345). [Drawn from data in ­table S1 of Sherrill-­Mix et al. (2018).]

with a fermentation chamber (­Reese and Dunn, 2018). As considered in detail in chapter 4 (section 4.1), a fermentation chamber is an anoxic portion of the gut in which plant fiber is degraded anaerobically by a complex community of anaerobic microorganisms. The insects with a hindgut fermentation chamber feed on plant fiber–­rich foods, and they include cockroaches and termites (phylum Blattodea), as well as many dung beetles (Coleoptera: most dung beetles are members of the f­ amily Scarabaeidae). The bacterial diversity in ­these insects generally exceeds 100 OTUs (Colman et al., 2012; Shukla et al., 2016; Tinker and Ottesen, 2016) and, where investigated, ­these insects also have eukaryotic microbial partners. Most notably, cockroaches of the Cryptocercidae and lower termites bear obligately anaerobic protists of two phyla, Parabasalia and the Preaxostyla (order Oxymonadida) that are unknown in other habitats (Brune, 2014). The biology of t­ hese protists in the context of the life cycle of the insect hosts is considered in section 3.6.2.

I n s e c t - M i c r ob i a l A s s o c i a t ion s  41

Let us now consider the phytophagous insects, i.e., insects that feed on living plant material. Unlike the herbivorous mammals (horses, ­cattle, voles, ­etc.), which have substantial fermentation chambers, most phytophagous insects do not have a fermentation chamber, and the taxonomic diversity of their gut microbiota is not consistently greater than for omnivorous and predatory insects (Colman et al., 2012). For example, the gut microbiota of leaf-­feeding lepidopteran caterpillars has fewer than 100 OTUs (and frequently 60% prevalence, considered the “core” hindgut bacteria, are shown in bold. [Drawn from data in datafile S2 of Kwong et al. (2017).]

some individual workers, the γ-­Proteobacteria are very abundant, and the six core bacteria are reduced to very low abundance or are undetectable (Kwong and Moran, 2016). 2.2.4 Associations with single microbial taxa in insect guts

A minority of insects are consistently associated with a single microbial taxon sequestered in a well-­defined region of the gut. Entomologists in the first half of the twentieth ­century identified ­these associations in heteropteran insects of the infraorder Pentatomomorpha (Hemiptera) and in vari­ous Coleoptera (Buchner, 1965). Recent studies with advanced molecular and microscopical methods have largely confirmed the early reports, and also obtained new insights into the diversity of microbial taxa in which this lifestyle has evolved. Let us, first, consider the symbioses in the Pentatomomorpha, a species-­ rich group of phytophagous, mostly seed-­feeding, heteropteran bugs. Four of its constituent superfamilies bear highly specific symbioses with microorganisms (Sudakaran et al., 2017). The ancestral condition is a symbiosis in the

I n s e c t - M i c r ob i a l A s s o c i a t ion s  45

PENTATOMOMORPHA

Lygaeoidea: Burkholderia (Caballeronia group) in Rhyparochromidae, Berytidae, Pachygronthidae, and some Blissidae replaced by intracellular bacteriocyte symbiosis with Sodalis-like γ-Proteobacteria in Geocoridae, Artheneidae, Lygaeidae, and some Blissidae Coreoidea: Burkholderia (Caballeronia group)

Pyrrhocoroidea: Burkholderia (Paraburkholderia group) in Largidae, replaced by symbiosis with several bacteria in a more proximal region of the midgut in Pyrrhocoridae Pentatomoidea: various γ-Proteobacteria

Aradoidea KEY:

Evolutionary origin of bacterial symbiosis in ceca of distal midgut

figure 2.8. Bacterial symbioses in Hemiptera of the heteropteran infraorder Pentatomomorpha. Symbiotic bacteria are borne in ceca localized in the distal midgut for all members of the superfamilies Pentatomoidea and Coreoidea, and some members of Pyrrhocoroidea and Lygaeoidea. The symbiosis is replaced by dif­fer­ent associations in the Pyrrhocoridae and some lygaeoid families (details in figure). [Drawn from data in fig. 2 of Sudakaran et al. (2017)]

distal midgut, which includes many ceca densely packed with a single bacterial taxon (fig. 2.8). In the Coreoidea, vari­ous Lygaeoidea, and Pyrrhocoroidea of the ­family Largidae (Kaltenpoth and Florez, 2020), the bacteria are Burkholderia (β-­Proteobacteria: Burkholderiaceae); and in the Pentatomoidea, the bacteria are γ-­Proteobacteria. Within the Pentatomoidea, the Pentatomidae and Cydnidae are associated with Erwinia/Pantoea/Enterobacter-­like bacteria (γ-­Proteobacteria: Erwiniaceae), with phyloge­ne­tic evidence for multiple acquisitions, losses, and transfers between dif­fer­ent host species (Hayashi et al., 2015; Hosokawa et al., 2012). By contrast, the Acanthosomatidae, Plataspidae, and Urostylididae have each codiversified with a clade of γ-­Proteobacteria with very small genomes ( 0.001; ***= p