Coevolution and consequences of symbioses between aphids and maternally transmitted bacteria
Item Type text; Dissertation-Reproduction (electronic)
Authors Russell, Jacob Adam
Publisher The University of Arizona.
Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
Download date 25/09/2021 04:51:01
Link to Item http://hdl.handle.net/10150/280740 COEVOLUTION AND CONSEQUENCES OF SYMBIOSES BETWEEN APHIDS
AND MATERNALLY TRANSMITTED BACTERIA.
by Jacob Adam Russell
A Dissertation Submitted to the Faculty of the
DEPARTMENT OF ECOLOGY AND EVOUTIONARY BIOLOGY
In Partial Fulfillment of the Requirements for the Degree of
DOCTOR OF PHILOSOPHY
In the Graduate College
THE UNIVERSITY OF ARIZONA
2004 UMI Number: 3158146
INFORMATION TO USERS
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
UMI
UMI Microform 3158146
Copyright 2005 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 9
The University of Arizona 'w Graduate College
As members of the Final Examination Committee, we certify that we have read the dissertation prepared by Jacob Russell entitled Coevolution and consequences of symbioses between
aphids and maternally transmitted bacteria
and recommend that it be accepted as fulfilling the dissertation requirement for the
Degree of Doctor of Philosophy
Nancy Mor date
ves Carriere date Ij zoc*^ date ij oa datu
Michael Worobey date
Final approval and acceptance of this dissertation is contingent upon the candidate's submission of the final copies of the dissertation to the Graduate College. I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement.
2.00^ Dissertation Director: Nancy Moran date 3
STATEMENT BY AUTHOR
This dissertation has been submitted in partial fiilfillment of requirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.
Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.
SIGNED: 4
ACKNOWLEDGMENTS
Among the many people I'd like to thank, first and foremost, it is my honor to praise my advisor, Nancy Moran. Nancy took me on as a naive 22-year-old, straight out of my undergraduate studies at the University of Rochester. Nancy's broad knowledge of biology and her fascinating research on symbiosis attracted me to the UofA, though at times I found her credentials to be a bit intimidating. Nonetheless, Nancy showed much patience in my early years, applying pressure when necessary to keep me in line and on pace to finish my degree. She also gave me a large amount of freedom which allowed me to grow as an independent scientist. Her lab environment was always stimulating, and I never lacked what I needed to conduct successful research. I thank Nancy for all of this and for the opportunity to work along-side her and many excellent scientists in her lab. Next, I would like to thank the members of my thesis committee. Molly Hunter has been like a second advisor to me. She has always taken an active interest in my research and has gone to great lengths to give me feedback on my research and presentations, and I thank her for all of this and more. Teri Markow has been on board for several years, and I thank her for her time and for her key role in keeping me fiinded throughout my graduate career. Yves Carriere and Carlos Machado have provided valuable advice on writing, statistics, and modeling. I thank Mike Worobey for joining on short notice, and for sharp observations on my molecular analyses. Previous members of my committee include John Jaenike, Wayne Maddison, and Christina Kennedy, cind I owe them much for their generosity. In addition to my committee, I would like to thank both Howard Ochman and Michael Nachman—faculty who have devoted much time toward flirthering my education and facilitating my research. I would like to thank Kim Hammond, Helen Dunbar, Phat Tran, Michele Hofftnan, and Ben Wang for their help in the laboratory over the years. I would not have graduated without their assistance and insight. I'd like to also thank my wonderful friends and colleagues, Kerry Oliver and Steve Perlman. I've learned much from my interactions with them and am especially gratefiil for the moral support they've provided. The Moran lab has been a fantastic place to work. I can't thank all of the members here, but I'd like to acknowledge Patrick Abbot, Gordie Plague, Alex Mira, Jean Tsai, Alex Wilson, Colin Dale, Vincent Daubin, and Emmanuelle Lerat. I would not be where I am today without the love and support of my parents, Scot and Martha Russell. My father has always been very supportive, generous,s and affectionate parent. My mother has also shown much generosity toward me over the years, even going so far as to help me collect aphids. I thank them both for everything they have done. I thank my wonderful girlfi-iend, Susannah Elwyn, who has always been there for me, lending an open ear and her full support. Her love and patience were essential during my time as a graduate student. I'd like to thank the EEB, CIS, and ARL office staff and the following funding sources: the Graduate College, the Center for Insect Science, the Plant Insect Interaction group, the RTG in the Analysis of Biological Diversification, the department of Ecology and Evolution, and the NSF IGERT in genomics at the University of Arizona. 5
TABLE OF CONTENTS
ABSTRACT 7
CHAPTER L INTRODUCTION 9
1.1. Review of the literature 9
1.1.1. An introduction to symbiosis 9
1.1.2. Partner fidelity: symbiont transmission over ecological timescales 10
1.1.3. Partner fidelity: symbiont transmission over evolutionary timescales 12
1.1.4. Ecological and evolutionary impacts: phenotypic effects of symbiosis. ..13
1.1.5. Ecological and evolutionary impacts: consequences of symbiosis seen
in genomes 17
1.1.6. Symbioses between aphids and bacteria 18
1.2 Dissertation format 22
CHAPTER 2. PRESENT STUDY 24
2.1. Distributions, modes of transfer, and rates of evolution 24
2.2. AT content and rates of evolution 25
2.3. Experimental horizontal transfer of secondary symbionts 26
2.4. Fitness effects induced by symbionts vary with temperature 27
REFERENCES 29
APPENDIX A. SIDE-STEPPING SECONDARY SYMBIONTS:
WIDESPREAD HORIZONTAL TRANSFER ACROSS AND BEYOND
THE APHIDOIDEA 39 6
TABLE OF COmENTS - continued
APPENDIX B. NUCLEOTIDE CONTENT AND RATES OF EVOLUTION
OF PROTEIN CODING GENES FROM R-TYPE SECONDARY
SYMBIONTS OF THE PEA PHID 56
APPENDIX C. HORIZONTAL TRANSMISSION OF SECONDARY
BACTERIAL SYMBIONTS: TRANSMISSION AND FITNESS EFFECTS
IN A NOVEL APHID HOST, ACYRTHOSIPHONPISUM 82
APPENDIX D. CONDITIONAL FITNESS EFFECTS OF SECONDARY
SYMBIONTS: EXPOSURE TO HIGH TEMPERATURES LEADS TO
STRONG COSTS AND BENEFITS OF SYMBIONT INFECTION IN
THE PEA APHID, ACYRTHOSIPHON PISUM. 119 7
ABSTRACT
Symbiosis is a prevalent phenomenon among organisms throughout the tree of
life, including the insects which often harbor maternally transmitted bacteria. Aphids engage in symbiotic interactions with several maternally transmitted bacteria, and many
associate with microbes known as secondary symbionts. These bacteria are typically not essential from the aphid's perspective, and until recently little was known about their
roles in aphid biology or the coevolutionary histories of these symbioses.
I have addressed these mysteries in this dissertation, through use of molecular and
experimental analyses. My findings reveal that secondary symbionts are diverse and
infect members of numerous aphid and insect taxa. Though they are maternally
transmitted, their distributions can be attributed to occasional horizontal transmission
among species.
Consequences of symbiosis were observed at genomic levels, with "T-" and "U-
type" symbionts exhibiting accelerated evolution in their 16S rRNA sequences. The "R-
type" symbiont, in contrast, has not experienced accelerated substitution rates, though it
does show a recent trend toward increased AT content, as observed for other symbiotic
bacteria.
Molecular and phylogenetic evidence presented in this dissertation suggest that
secondary symbionts are generalists, capable of infecting numerous aphid hosts. Here, I
also present experimental evidence in support of this hypothesis, demonstrating that two 8 of three horizontally transferred symbionts are passed on maternally, at high efficiency, in a novel host, Acyrthosiphon pisum. However, not all efficiently transmitted symbionts would be expected to persist in populations oiA. pisum, as some reduce aphid fitness.
Finally, evidence obtained from my research and previous experimental and theoretical studies suggests that secondary symbionts should improve aphid fitness, though benefits may not accrue in all environments. Here, I examine the effects of temperature on the fitness effects induced by R-, T-, and U-type symibionts, finding that the R- and T-types confer benefits in aphids exposed to high temperatures, compared to slight and even non-existent effects on A. pisum reared under permissive temperatures.
The U-type reduced fitness of aphids reared under high temperatures, revealing a potential cost to symbiont infection that could help to explain intermediate infection frequencies. 9
CHAPTER 1.
INTRODUCTION
1.1. Review of the literature
1.1.1. An introduction to symbiosis
Many organisms engage in intimate, prolonged associations referred to as symbioses. These interactions have had profound impacts on the ecology and evolution of bacteria and eukaryotes. One of the most influential collaborations can be seen in the case of the mitochondria, energy producing organelles which reside in the cells of nearly all eukaryotes. These organelles are derived from an ancient microbe that evolved from ihs. a-Proteobacteria {Andersson et al 1998, Gray a/. 1999). Their bacterial ancestor engaged in a symbiosis with a single-celled ancestor of the eukaryotes, permitting this latter group to colonize aerobic environments.
Chloroplasts represent another example of an ancient and influential symbiotic interaction. These plant organelles arose from the Cyanobacteria, a group of photosynthetic microbes (Bremer and Bremer 1989, Martin et al. 1992). Their symbioses with the ancestor of green plants and algae evolved into a highly successfiil collaboration which conferred an autotrophic lifestyle on this lineage of eukaryotes. The benefits of photosynthesis have spread to an even wider array of taxa, through the subsequent 10 evolution of symbioses between photo synthetic eukaryotes and protist or animal hosts
(Delwiche and Palmer 1997).
Whereas mitochondria and chloroplasts represent the most widespread and ancient examples of symbiosis, organisms throughout the tree of life engage in symbioses that vary tremendously in taxonomic breadth and age of association. Through a combination of experimental investigations and studies on the genes and genomes of symbiotic participants, researchers have elucidated the evolutionary histories and consequences of these interactions. Their results reveal that symbioses have evolved on numerous occasions and display varying degrees of fidelity. They have also repeatedly shown that symbiosis plays several biological roles and has substantially impacted the ecology and evolution of most living organisms.
1.1.2. Partner fidelity: symbiont transmission over ecological timescales
In the most intimate symbioses, organisms remain in tight physical contact throughout their life-cycles. These typically involve a uni- or multicellular eukaryote and a smaller microbial symbiont, which lives inside the cells or tissues of its host. Under these circumstances, the fidelity exhibited between symbiotic partners is determined by the route of symbiont transmission. Transmission of symbionts can occur in either a vertical or horizontal fashion. Under horizontal transmission, symbionts spread among unrelated hosts through routes such as ingestion. Symbionts that undergo vertical transmission, in contrast, are passed directly from parent to offspring. The degree to 11 which symbionts rely on these modes of transfer will determine how natural selection acts to shape their effects on their hosts.
Vertical transmission tends to align the interests of hosts and symbionts, since symbiont transfer will depend on host reproduction. According to theoretical research, a reliance on this mode of transfer will favor symbionts that are beneficial toward their hosts (Ewald 1983, May and Anderson 1982). This prediction has been met in experimental studies that found bacteriophage evolve reduced virulence toward bacterial hosts when strict vertical transmission was enforced (Bull et al. 1991, Messenger et al.
1999). In contrast, horizontally transmitted symbionts do not require host reproduction for their transmission, and thus selection will not always favor reduced virulence for such microbes. Accordingly, most of the parasites that infect plants and animals spread primarily through a horizontal route. However, horizontal transmission has been documented for several mutualistic symbionts (e.g. Douglas 1995). More recent theory predicts that horizontally transmitted symbionts can indeed evolve to benefit their hosts, revealing that this route of transfer it is not a one-way ticket to parasitism (e.g. Genkai-
Kato and Yamamura 1999).
The expectation that symbionts undergoing strict vertical transmission should benefit hosts does not always hold for those infecting sexually reproducing eukaryotes.
In these hosts symbionts are typically transmitted through the maternal lineage. As a result, symbionts will spread if the number of infected daughters produced by hosts outnumbers the number of daughters produced by uninfected conspecifics (Bull 1983).
Vertically transmitted symbionts have evolved several strategies to achieve this goal. 12 selfishly manipulating host reproduction to improve their own fitness (e.g. Rousset et al.
1992, Stouthamer 1997).
1.1.3. Partner fidelity: symbiont transmission over evolutionary timescales
The modes of symbiont transmission can be followed from generation to generation through techniques such as microscopy or PGR. Yet these methods cannot address the patterns of transmission and specialization over longer, evolutionary timescales. For this purpose, researchers have utilized DNA sequences from symbiotic partners to construct phylogenies depicting their relationships. In several instances, researchers have noted that host and symbiont phylogenies are congruent (Munson et al.
1991a, Clark et al. 2000, Thao et al 2000a, Aksoy et al 1999, Bandi et al. 1995, Sauer et al 2000, Moran et al. 2003, Peek et al. 1998, Hurtado et al. 2003). Such fmdings indicate strict vertical transmission and reveal that the evolution of the symbiotic partners has been tightly and faithfully intertwined for millions of years. In contrast, some host and symbiont phylogenies are incongruent, revealing instances of horizontal transmission and a lower degree of fidelity over evolutionary timescales (O'Neill et al. 1992, Moran and Baumann 1994, Sandstrom et al. 2001, Russell et al 2003). This latter finding suggests that some symbionts are generalists, capable of infecting different host species.
Accordingly, lab studies have revealed that symbionts such as the endophytic fiingi of grasses, and the vertically transmitted Wolbachia of insects, are clearly capable of 13 infecting multiple host species (Braig et al. 1994, Rigaud et al. 2001, Sasaki and
Ishikawa 2000, Leuchtmann and Clay 1993).
1.1.4. Ecological and evolutionary impacts: phenotypic effects of symbiosis
Symbionts employ several strategies that favor their spread throughout host populations. These strategies, which range from beneficial to parasitic, have had substantial impacts on the ecology and evolution of their hosts. The themes of the strategies employed by unrelated microbes show considerable overlap, revealing that symbionts have independently evolved similar means to promote their proliferation.
Nutrition represents one component of eukaryotic biology that is often shaped by symbiosis. Several symbionts facilitate the acquisition and utilization of nutrients from the environment. For example, plants occasionally associate with soil-dwelling mycorrhizal fungi, which aid in the uptake of phosphorous (e.g. Ortas et al. 1996).
Legumes harbor rhizobial bacteria, which convert atmospheric nitrogen into ammonium, a nitrogen source that can be used by these plants (e.g. Ogara and Shanmugam 1976).
Alternatively, some bacterial symbionts synthesize essential vitamins and amino acids for their hosts. These are often found in animals feeding on nutrient poor diets, such as blood or plant sap (Baumann and Moran 1997, Moran and Telang 1998, Moran et al.
2004). Aphids, for example, harbor vertically transmitted symbionts known as Buchnera aphidicola, which synthesize a variety of vitamins and essential amino acids that are rare in the aphid diet of phloem sap (Douglas 1993, Douglas and Prosser 1992, Douglas 1988, 14
Bracho et al. 1995, Sandstrom and Mo ran 2001, Bernays and Klein 2002, Shigenobu et al. 2000, Munson and Baumann 1993, Sandstrom and Moran 1999). In exchange,
Buchnera obtain several non-essential amino acids, synthesized by their hosts (Shigenobu et al. 2000). Symbiotic bacteria of ants, tsetse flies, and weevils play similar roles through the provisioning of amino acids and vitamins (Nogge 1981, Wicker 1983,
Akman et al. 2002, Gil et al. 2003), revealing that nutritional symbioses between insects and bacteria have evolved independently on several occasions. In fact, associations with nutrient-providing symbionts appear to have played a large role in the evolution of the insects. Most notably, insects from the order Hemiptera, such as whiteflies, aphids, scale insects, psyllids, and leafhoppers, are almost ubiquitously infected with vertically transmitted symbionts. Given the established nutritional roles of several of these symbionts, their prevalence, and the tight correlation between diets of plant sap and ubiquitous symbiosis, it appears that these microbes facilitated the radiation of a large group of insects onto a previously unsuitable niche (Moran and Baumarm 1994).
In addition to nutritional roles, some symbionts confer defensive benefits, protecting their hosts from natural enemies. Two vertically transmitted bacteria of the pea aphid, Acyrthosiphon pisum, confer resistance to the wasp Aphidius ervi. A. ervi fail to develop in aphids with "R-" (Oliver et al. 2003) or "T-type" (Oliver et al. 2003, Ferrari et al. 2004) symbionts more often than in uninfected aphids. A. pisum individuals often harbor an additional symbiont ("U-type") whose presence is correlated with resistance to a fiingal parasite. Pandora (Erynia) neoaphidis (Ferrari et al. 2004). Similarly, some evidence suggests that a bacterial symbiont of Hypera postica, the alfalfa weevil, plays a 15 defensive role against parasitoids—weevils that were cured of symbiotic bacteria through antibiotic treatment were more susceptible to parasitism than their symbiont-infected counterparts (Hsiao 1996). In this instance, the identity of the defensive symbiont has not been conclusively demonstrated.
Aphids and weevils are not the only organisms that benefit from a defensive symbiosis. Staphylinid beetles of the genus Paederus harbor bacterial endosymbionts related to Pseudomonas species (Kelhier 2002). These synthesize the toxin pederin, a polyketide that protects their beetle hosts from predators (Kelhier and Dettner 1996, Piel
2002). In addition, twenty to thirty percent of grasses harbor endophytic fringi
(Leuchtmann 1992), which can protect them from herbivory (Siegel 1990, Clay 1996).
Defensive roles are not confined to microbial symbionts—indeed, several ants protect hemipteran insects or plants from natural enemies in exchange for sugar-rich secretions
(Schemske 1980, Floate and Whitham 1994, Weeks 2003). The repeated origins of defensive symbioses are a testament to the importance of natural enemies in shaping the evolution of their eukaryotic hosts.
When considering vertically transmitted symbionts of sexually reproducing eukaryotes it becomes clear that their success is not solely dependent on mutualistic strategies. As mentioned above, vertically transmitted elements will spread if the number of infected daughters bom to their hosts exceeds the number of daughters bom to uninfected members of the same species (Bull 1983). Beneficial effects of symbionts can obviously meet this requirement. But in addition, manipulative and parasitic strategies can be employed to achieve the same end. 16
Several clades of microorganisms have evolved to manipulate eukaryotic reproduction in a selfish, antagonistic attempt to gain higher representation among their hosts. Such strategies are most commonly executed within invertebrate hosts, especially the insects. For example, Wolbachia pipientis and Cardinium hertigii have independently evolved to manipulate insect reproduction through a strategy referred to as cytoplasmic incompatibility (Breeuwer and Werren 1990, O'Neill et al. 1992, Rousset et al. 1992, Hvmter et al. 2003, Zchori-Fein et al. 2004). This phenomenon typically results in lowered fecundity for uninfected females when mated to infected males. Infected females, in contrast, produce normal numbers of progeny regardless of their mate's infection status. As a result, uninfected females are at a reproductive disadvantage, allowing symbiont-infected females to spread at their expense (Caspari and Watson 1959,
Hoffmann et al. 1990). Wolbachia and C. hertigii have also been shown to induce parthenogenesis in members of the insect order Hymenoptera (Zchori-Fein et al. 2001,
Rigaud et al. 1991). In these instances, infected and unmated females produce infected daughters, favoring the spread of the inducing symbionts through production of the transmitting sex. Finally, maternally transmitted symbionts of invertebrates have evolved strategies of feminization (Rigaud et al. 1991) and male killing. Male killing is the most widespread of all the manipulative strategies, having evolved numerous times among the
Bacteroidetes, Mollicutes, and a- and y-Proteobacteria (Gherna et al. 1991, Hurst et al.
1997, Hurst et al. 1999, Williamson and Whitcomb 1979). The mechanisms behind the independent origins of these manipulative phenotypes remain undescribed. In the future. 17 it will be of interest to determine whether such homoplasy is due to convergent evolution or lateral gene transfer among symbionts co-infecting the same hosts.
1.1.5. Ecological and evolutionary impacts: consequences of symbiosis seen in genomes
Symbiosis can have profound consequences, not only on the phenotypes of the participants, but also on their genomes. Several eukaryotic and bacterial microbes that have retained a symbiotic lifestyle for millions of years have undergone a process of genome degradation. This has been characterized by rampant gene loss—symbiont genomes are much smaller than those of their free-living relatives and are among the smallest described for cellular organisms (e.g. Delwiche and Palmer 1997, Katinka et al.
2001, Mira et al. 2001, Ochman and Moran 2001, Moran et al. 2003). Symbionts of insects have also evolved at elevated rates, compared to free-living bacteria, changing more rapidly at positions subject to natural selection such as 16S rRNA sequences and non-synonymous sites of protein coding genes (Moran et al. 1993, Moran 1996, Thao et al. 2000a, Degnan et al. 2004). Finally, symbiont genomes are typically high in AT content. This reflects a substitutional bias that has arisen on several independent occasions, apparently due to the action of mutation rather than selection (Ishikawa 1987,
Baumann et al. 1995, Fukatsu and Nikoh 1998, Spaulding and von Dohlen 1998). It has been argued that the population structures of symbionts have led to genome degradation—^their small, asexual populations allow for a larger effect of genetic drift, resulting in gene loss and more rapid sequence change (Muller 1964, Moran 1996, 18
Wernegreen and Moran 1999, Funk et al. 2001). Given that much of this genome degradation appears irreversible (Moran and Wernegreen 1999), it can be concluded that the symbiotic lifestyle has often led to rapid and one-way specialization for maternally inherited bacteria.
1.1.6. Symbioses between aphids and bacteria
Aphids are phloem-feeding insects that infest a variety of plants in both natural and agricultural settings. They cause extensive damage on crops around the world, often through their roles as vectors of pathogenic plant viruses (Blackman and Eastop 2000).
Symbiosis plays an integral part in the biology of aphids, nearly all of which harbor an essential, maternally transmitted bacterium known as Buchnera aphidicola (Prosser and
Douglas 1991; Munson a/. 1991a&b). are harbored within bacteriocytes, specialized symbiotic cells that have fiinctional and structural equivalents in many other bacteria-harboring insects (Buchner 1965). They are transmitted to developing embryos inside their mothers, and have strictly relied on this route of transmission for over 100 million years as evidenced by fossil dating and congruency between aphid and Buchnera phylogenies (Munson et al. 1991a, Moran and Baumann 1994, Moran et al. 1993, von
Dohlen and Moran 2000, Clark et al. 2000). As described above, these symbionts play an important nutritional role in aphid biology, providing hosts with several essential amino acids that are rare in plant sap diets (Douglas 1993, Douglas and Prosser 1992,
Douglas 1988, Bracho et al. 1995, Sandstrom and Moran 2001, Bernays and Klein 2002, 19
Shigenobu et al. 2000, Munson and Baumann 1993, Sandstrom and Moran 1999). And like several other described symbionts and host-dependent pathogens, Buchnera genomes have undergone a process of degradation. They have lost thousands of genes, harboring genomes as small as 600 kb—several thousand kilobases smaller than those of free-living bacteria such as E. coli (Shigenobu et al. 2000, Tamas et al. 2002, van Ham et al 2003).
In addition to this gene loss, their genomes have experienced elevated rates of substitution and display strong AT bias, revealing the significance of drift in shaping their evolution (e.g. Moran 1996, Ishikawa 1987, Baumann et al. 1995).
Aphid symbioses are not limited to interactions with Buchnera—many have been shown to harbor additional bacteria known as secondary or accessory endosymbionts
(Buchner 1965, McLean and Houk 1973). These bacteria are also transmitted vertically, with no natural instances of horizontal transfer described to date (Chen and Purcell 1997,
Fukatsu et al. 2000, Sandstrom et al. 2001, Darby and Douglas 2003). The localization of these symbionts within aphid body cavities is much less precise than that of
Buchnera—secondary symbionts can reside within bacteriocytes but have also been described in the hemolymph, among other locations (Fukatsu et al 2000, Sandstrom et al 2001, Darby et al. 2001). This evidence, coupled with their sporadic distributions across hosts, led earlier researchers to hypothesize that secondary symbionts were more recently acquired guests of aphids, compared to the primary endosymbiont Buchnera
(Buchner 1965). Indeed, recent analyses on their DNA sequences have provided support for this hypothesis, as the phylogenies of three distinct secondary symbionts did not match those of their hosts from the aphid tribe Macrosiphini (Sandstrom et al. 2001). 20
At the time of the onset of my dissertation research, four groups of secondary symbionts had been described from aphids based on microscopy and 16S rDNA sequence analysis. Interestingly members from each of these clades infect the pea aphid,
Acyrthosiphonpisum, which feeds on legumes and can transmit pathogenic viruses to crops of economic importance (Blackman and Eastop 2000). The first secondary symbiont to be described has been referred to by several names including S-sym
(Unterman et al. 1989), PASS (Chen and Purcell 1997), and the R-type (Sandstrom et al.
2001). This microbe is closely related to free-living bacteria of the genus Serratia, and thus its symbiotic lifestyle evolved recently, relative to other described symbionts of insects (Sandstrom et al. 2001, Russell et al. 2003). Following this bacterium, two additional symbionts dubbed the T- and U-types (y-Proteobacteria), were described from
A. pisum and several of its relatives from the Macrosiphini (Sandstrom et al. 2001). Five additional bacterial lineages have been described as secondary symbionts of aphids (Chen et al. 1996, Fukatsu 2001a&b, Russell et al. 2003), including members of the genus
Arsenophonus which also infect other hemipterans as well as hymenopteran wasps
(Gherna et al 1991, Thao et al. 2000b, Zchori-Fein and Brown 2002, Grindle et al.
2003).
The diversity of secondary symbionts among the aphids is impressive but, until recently, their roles were undescribed. Given their reliance on vertical transmission and their high prevalence within host populations (Chen et al. 1996, Chen and Purcell 1997,
Sandstrom et al. 2001, Tsuchida et al. 2002, Simon et al. 2003) it has been predicted that these microbes should benefit their hosts, at least under some conditions. Accordingly, 21 several studies have described benefits associated with R-, T-, and U-type infection.
First, the R-type has been implicated in conferring heat tolerance in pea aphids—aphids with this microbe have higher fecundity than genetically identical, uninfected individuals when exposed to high temperatures (Chen et al 2000, Montllor et al. 2002).
Interestingly, the R-type was also shown to benefit aphids that had been cured of their essential Buchnera symbionts—symbiont-free aphids were sterile unless they harbored the R-type, which led to a partial restoration of fecundity (Koga et al. 2003). Thus, the
R-type can supplement Buchnera function, though only to a small degree. Next, as described above, both the R- and T-type bacteria confer resistance to a parasitoid wasp that is a common natural enemy of pea aphids in the field (Oliver et al. 2003, Ferrari et al. 2004). Pea aphids infected with either of these symbionts are significantly more likely to survive parasitism than their uninfected coimterparts. Similarly, the U-type symbiont has been implicated in resistance to pathogenic fungi, though the evidence to date is based only on correlative studies (Ferrari et al. 2004). The U-type also plays a role in host plant utilization by pea aphids, permitting some A. pisum to feed on clover (Tsuchida et al. 2004, but see Leonardo 2004).
These phenotypes suggest that secondary symbionts are important players in the ecology and evolution of their aphid hosts. Clearly, evidence from the field is required to bolster this claim. Moreover, additional laboratory studies are required to determine how symbionts vary in their capacities to shape host fitness and how their effects vary with environmental conditions. 22
1.2. Dissertation format
The goal of this dissertation is to dissect the symbiotic interactions between insects and their bacterial inhabitants, focusing on the specific case of aphids and their secondary symbionts. Here, I characterize the evolutionary histories of the interactions as well as the consequences of symbiosis from the perspectives of both the insects and bacteria.
In Appendix A, I describe the diversity and distributions of bacterial symbionts across the aphids and related insects, through use of diagnostic PGR and DNA sequencing. I also reconstruct the evolutionary histories of these interactions, using DNA sequence analyses and phylogenetics. Finally, I estimate the rates of evolution within symbiotic lineages, comparing these to those of free-living bacteria in order to describe the consequences of symbiosis on molecular evolution. The research in Appendix A was performed in collaboration with Amparo Latorre, Beatriz Sabater-Mufioz, and Andres
Moya at the University of Valencia. These coauthors screened European-collected aphid species for secondary symbionts, sequenced 16S rRNA genes of several R-, T-, and U- type symbionts, and performed a statistical test to compare phylogenies of U-type symbionts and their aphid hosts.
In Appendix B, I ftirther characterize the consequences of symbiosis for a recently-evolved symbiotic lineage referred to as the R-type. I sequence portions of several genes with roles in nutrient biosynthesis, describing their nucleotide content and rates of evolution. 23
In Appendix C, I analyze interactions between novel combinations of aphids and bacterial symbionts, in an attempt to determine whether secondary symbionts are generalists. I use microinjection and diagnostic PGR to create and follow novel infections in^l. pisum, and perform subsequent assays on aphid fitness to describe the outcomes of novel symbiotic interactions. Finally, in Appendix D, I again examine the consequences of symbiosis from the perspective of the aphids, focusing on the roles of secondary symbionts in conferring heat tolerance upon A. pisum. I study three symbiotic lineages, and use multiple isolates of the R-type symbiont to describe variation among bacteria in their capacities to protect hosts from the stress of heat exposure. 24
CHAPTER 2.
PRESENT STUDY
The methods, results, and conclusions for my dissertation research are displayed in the papers attached to this thesis. Here, I summarize the most important findings from each chapter.
2.1 Distributions, modes of transfer, and rates of evolution
Several aphids from the tribe Macrosiphini have previously been shown to harbor one of three secondary symbionts, dubbed the R-, T-, and U-types (Sandstrom et al.
2001). These maternally transmitted microbes were found to undergo occasional horizontal transmission, as evidenced by incongruence between insect and symbiont 16S rDNA phylogenies. To describe the distributions of these bacteria and to characterize their coevolutionary histories with a broader range of aphids, I screened 75 aphid species and 27 psyllids (phloem-feeding relatives of aphids) for R-, T-, and U-type secondary symbionts. I sequenced their 16S rRNA genes, and compared their rates of evolution to those of symbiotic and bacterial relatives. To describe their transmission modes over evolutionary timescales, I compared the genetic distances between 16S rRNA genes of symbionts in different species and used their sequences to construct phylogenies which were compared to those of their hosts. I found that R-, T-, and U-type symbionts infect several aphids outside the Macrosiphini, and in some instances even outside of the
Aphididae. Relative rates tests on their 16S rDNA sequences demonstrated that the T- 25 and U-types (but not the R-type) evolve more rapidly than free-living bacteria. This is consistent with the expectation that drift will play a larger role in the evolution of symbiont genomes due to their population structures. The sequences of symbionts in unrelated hosts were highly similar, and an analysis of genetic distances revealed no evidence for cospeciation. In addition, host and symbiont phylogenies were found incongruent, indicating horizontal transmission. Combined, these findings suggest that secondary symbionts of aphids are generalists capable of infecting numerous host species.
2.2 AT content and rates of evolution
In the research presented in Appendix A, the 16S rDNA sequence of the R-type was not found to evolve at a faster rate than that of its free-living relatives. Interestingly,
R-type symbionts and free-living Serratia were highly similar in 16S rDNA sequence
(97-98%), suggesting a recent origin for this symbiotic clade. To determine whether the
R-type shows evidence for genomic changes characteristic of more ancient symbionts, I identified and sequenced six protein coding genes from the R-type, which were primarily involved in the biosynthesis of amino acids and vitamins. One of these genes (hisG) had a premature stop codon, suggesting that it was a recently inactivated pseudogene. All had lower AT nucleotide content compared to more ancient symbiotic bacteria, but substantially higher AT content than their free-living Serratia relatives. The protein coding genes of the R-type evolved at rates comparable to those in Serratia. Combined, 26 these results reveal that the effects of the symbiotic lifestyle on R-type genome are considerably more subtle than those acting on the genomes of their ancient symbiotic counterparts but, perhaps, that the R-type has reached an early stage of genome degradation.
2.3 Experimental horizontal transfer of secondary symbionts
Results from Appendix A and previous studies have suggested that secondary symbionts of aphids are generalists, capable of colonizing novel host species in horizontal transfer events. However, little is known about the barriers to horizontal transmission of these symbionts and, thus, the forces that govern their distributions. To describe the capacity of secondary symbionts of aphids to infect novel aphid species, I transferred microbes heiween Acyrthosiphon pisum hosts using microinjection. I examined vertical transmission in these aphids, comparing efficiencies between novel and naturally occurring symbionts. I also measured the fitness effects of novel symbionts on A. pisum and compared these to the effects induced by naturally occurring symbionts.
Results revealed that secondary symbionts vary in their capacity to efficiently infect the offspring of a novel aphid host—only two out of three transferred symbionts were stably maintained in A. pisum lineages. Findings also indicated that symbionts vary in their effects on novel hosts. A novel T-type symbiont was neutral, except when reared at a temperature of 25°C, when it improved A. pisum survival. In contrast, a novel
Arsenophonus symbiont consistently depressed the fitness of aphid hosts. Combined, 27 these results reveal that inefficient vertical transmission and negative fitness effects may limit the distributions of the secondary symbionts of aphids.
2.4 Fitness effects induced by symbionts vary with temperature
The frequencies of vertically transmitted symbionts of aphids will depend on the effects the microbes have on host fitness. Previous findings suggest that R-, T-, and U- type secondary symbionts all confer benefits npon Acyrthosiphon pisum under specific environmental conditions (Chen et al. 2000, Montllor et al. 2002, Oliver et al. 2003,
Tsuchida et al. 2004). These microbes, thus, play roles in adaptation to both significant biotic and abiotic components of aphid environments. For example, the R-type has been shown to play a role in protecting A. pisum from heat, as R-infected hosts had higher fitness after exposure to a brief period of heat shock (Montllor et al. 2002). To determine the generality of the heat tolerance phenotype, I examined the effects of three R-type isolates and a T- and U-type isolate on the fitness of aphids subjected to heat shock.
Results indicate that R-type symbionts can buffer aphid hosts from the detriments of heat exposure. However, this protection varies between symbiont isolates and depends on the timing of exposure. The T-type improved the survival of heat shocked A. pisum, though it had no effects on development time or fecundity. In contrast, the U-type was harmfiil to aphids exposed to high temperatures, causing a substantial reduction in survival.
When reared constantly at permissive temperatures, T- and U-types had no effects on aphid fitness, whereas two of three R-type isolates were neutral aside from conferring a 28 slight developmental acceleration. Combining these results with previous findings, we can conclude that unrelated secondary symbionts have evolved to induce similar effects on their aphid hosts and that the outcomes of symbiosis are highly dependent on both the biotic and abiotic envrroimient. The overlapping roles of R- and T-type symbionts in both defense and heat tolerance reflect the importance of natural enemies and temperature in driving the evolution of these important crop pests. 29
REFERENCES
Akman, L., A. Yamashita, H. Watanabe, K. Oshima, T. Shiba, M. Hattori, and S. Aksoy. 2002. Genome sequence of the endocellular obhgate symbiont of tsetse flies, Wigglesworthia glossinidia. Nature Genetics 32:402-407.
Aksoy, S., X. Chen, and V. Hypsa. 1997. Phylogeny and potential transmission routes of midgut-associated endosymbionts of tsetse (Diptera: Glossinidae). Ins. Mol. Biol. 6:183-190.
Andersson, S.G., A. Zomorodipour, J.O. Andersson, T. Sicheritz-Ponten, U.C. Alsmark, R.M. Podowski, A.K. Naslund, A.S. Eriksson, H.H. Winkler, and C.G. Kurland. 1998. The genome sequence of Rickettsia prowazekii and the origin of mitochondria. Nature 396:133-140.
Bandi, C., M. Sironi, G. Damiani, L. Magrassi, C.A. Nalepa, U. Laudani, and L. Sacehi. 1995. The establishment of intracellular symbiosis in an ancestor of cockroaches and termites. Proc. R. Soc. Lond. Ser. B 259:293-299.
Baumann, P., L. Baumann, C.Y. Lai, and D. Rouhbaksh. 1995. Genetics, physiology, and evolutionary relationships of the genus Buchnera: intracellular symbionts of aphids. Aimu. Rev. Microbiol. 49:55-94.
Baumann, P., and N.A. Moran. 1997. Non-cultivable microorganisms from symbiotic associations of insects and other hosts. Anton. Leeuw. Int. J. G. 72:39-48.
Bemays, E.A., and B.A. Klein. 2002. Quantifying the symbiont contribution to essential amino acids in aphids: the importance of tryptophan for Uroleucon ambrosiae. Physiol. Entomol. 27:275-284.
Blackman, R.L., and V.F. Eastop. 2000. Aphids on the world's crops: an identification and information guide. 2"^ edition. John Wiley and Sons, West Sussex, UK.
Bracho, A.M., D. Martinez-Torres, A. Moya, and A. Latorre. 1995. Discovery and molecular-characterization of a plasmid localized in Buchnera sp. bacterial endosymbiont of the aphid Rhopalosiphum padi. J. Mol. Evol. 41:67-73.
Braig, H.R., H. Guzman, R.B. Tesh, and S.L. O'Neill. 1994. Replacement of the natural Wolbachia symbiont of Drosophila simulans with a mosquito counterpart. Nature 367:453-455. 30
Breeuwer, J.A.J., and J.H. Werren. 1990. Microorganisms associated with chromosome destruction and reproductive isolation between 2 insect species. Nature 346:558-560.
Bremer, B., and K. Bremer. 1989. Cladistic analysis of blue-green prokaryote interrelationships and chloroplast origin based on 16S ribosomal-RNA oligoneucleotide catalogs. J. Evol. Biol. 2:13-30.
Buchner, P. 1965. Endosymbiosis of animals with plant microorganisms. Interscience, New York.
Bull, J.J. 1983. Evolution of sex determining mechanisms, Benjamin Cummings Press, California.
Bull, J.J., I.J. Molineux, and W.R. Rice. 1991. Selection of benevolence in a host- parasite system. Evolution 45:875-882.
Caspari, E., and G.S. Watson. 1959. On the evolutionary importance of cytoplasmic sterility in mosquitos. Evolution 13:568-570.
Chen, D.-Q., C.B. Montllor, and A.H. PurcelL 2000. Fitness effects of two facultative endosymbiotic bacteria on the pea aphid Acyrthosiphon pisum, and the blue alfalfa aphid, v4. kondoi. Entomol. Exp. Appl. 95:315-323.
Chen, D.-Q., B.C. Campbell, and A.H. PurcelL 1996. A new rickettsia from a herbivorous insect, the pea aphid Acyrthosiphon pisum (Harris). Curr. Microbiol. 33:123-128.
Chen, D.-Q., and A.H. PurcelL 1997. Occurrence and transmission of facultative endosymbionts in aphids. Curr. Microbiol. 34:220-225.
Clark, M.A., L. Baumann, M.L. Thao, N.A. Moran, and P. Baumann. 2001. Degenerative minimalism in the genome of a psyllid endosymbiont. J. Bacteriol. 183:1853-1861.
Clark, M.A., N.A. Moran, P. Baumann, and J.J. Wemegreen. 2000. Cospeciation between bacterial endosymbionts (Buchnerd) and a recent radiation of aphids (Uroleucon) and pitfalls of testing for phylogenetic congruence. Evolution 54:517- 525.
Clay, K. 1996. Interactions among fungal endophytes, grasses and herbivores. Res. Popul. Ecol. 38:191-201. 31
Darby, A.C., L.M. Birkle, S.L. Turner, and A.E. Douglas. 2001. An aphid-bome bacterium allied to the secondary symbionts of whitefly. FEMS Microbiol. Ecol. 36:43-50.
Darby, A.C., and A.E. Douglas. 2003. Elucidation of the transmission patterns of an insect-bome bacterium. Appl. Environ. Microbiol. 69:4403-4407
Degnan, P.H., A.B. Lazarus, C.D. Brock, and J.J. Wemegreen. 2004. Host-symbiont stability and fast evolutionary rates in an ant-bacterium association: cospeciation of Camponotus species and their endosymbionts, Candidatus Blochmannia. Syst. Biol. 53:95-110.
Dehviche, C.F., and J.D. Palmer. 1997. The origin of plastids and their spread via secondary symbiosis. Plant Syst. Evol. 811:53-86.
Douglas, A.E. 1988. Sulfate utilization in an aphid symbiosis. Insect Biochem. 18:599- 605.
Douglas, A.E. 1995. The ecology of symbiotic micro-organisms. Adv. Ecol. Res. 26:69- 103.
Douglas, A.E., and W.A. Prosser. 1992. Synthesis of the essential amino acid tryptophan in the pea aphid {Acyrthosiphon pisum) symbiosis. J. Insect Physiol. 38:565-568.
Ewald, P. 1983. Transmission modes and the evolution of the parasitism-mutualism continuum. Ann. N. Y. Acad. Sci. 503:295-306.
Ferrari, J., A.C. Darby, T.J. Daniell, H.C.J. Godfray, A.E. Douglas. 2004. Linking the bacterial community in pea aphids with host-plant use and natural enemy resistance. Ecol. Entomol. 29:60-65.
Floate, K.D., and T.G. Whitham. 1994. Aphid-ant interaction reduces chrysomelid herbivory in a Cottonwood hybrid zone. Oecologia 97:215-221.
Fukatsu, T., and N. Nikoh. 1998. Two intracellular symbiotic bacteria from the mulberry psyllid Anomoneura mori (Insecta, Homoptera). Appl. Environ. Microbiol. 64:3599-3606.
Fukatsu, T., N. Nikoh, R. Kawai, and R. Koga. 2000. The secondary endosymbiotic bacterium of the pea aphid Acyrthosiphon pisum (Insect: Homoptera). Appl. Environ. Microbiol. 66:2748-2758. 32
Funk, D.J., J.J. Wernegreen, and N.A. Moran. 2001. Intraspecific variation in symbiont genomes: bottlenecks and the aphid-5Mc/?«era association. Genetics 157:477-489.
Genkai-Kato, M., and N. Yamamura. 1999. Evolution of mutualistic symbiosis without vertical transmission. Theor. Pop. Biol. 55:309-323.
Gherna, R.L., J.H. Werren, W. Weisbui^, R. Cote, C.R. Woese, L. Mandelco, and D.J. Brenner. \99\. Arsenophonus nasoniae gen. nov., sp. nov., the causative agent of the son-killer trait in the parasitic wasp Nasonia vitripennis. Int. J. Syst. Bacteriol. 41:563-565.
Gil R, F. J. Silva, E. Zientz, F. Delmotte, F. Gonzalez-Candelas, A. Latorre, C. Rausell, J. Kamerbeek, J. Gadau, B. HoIIdobler, R.C. van Ham, R. Gross, and A. Moya. 2003. The genome sequence of Blochmanniafloridanus: comparative analysis of reduced genomes. Proc. Natl. Acad. Sci. USA. 100:9388-9393.
Gray, M.W., G. Bui^er, and B.F. Lang. 1999. Mitochondrial evolution. Science 283:1476-1481.
Grindle, N., J.J. Tyner, K. Clay, C. Fuqua. 2003. Identification of Arsenophonus-type bacteria from the dog tick Dermacentor variabilis. J. Invertebr. Pathol. 83:264-266.
Hoffmann, A.A., M. Turelli, and L.G. Harshman. 1990. Factors affecting the distribution of cytoplasmic incompatibility in Drosophila simulans. Genetics 126:933-948.
Hsiao, T. H. 1996. m Ecology of Agricultural Pests, eds. Symondson, W. O. C., and Lidell, J. E. Chapman & Hall, London, pp. 51-71.
Hunter, M.S., S.J. Perlman, and S.E. Kelly. 2003. A bacterial symbiont in the Bacteroidetes induces cytoplasmic incompatibility in the parasitoid wasp Encarsia pergandiella. Proc. R. Soc. Lond. B. 270:2185-2190.
Hurst, G.D.D., J.H. Graf von der Schulenburg, T.M.O. Majerus, D. Bertrand, LA. Zakharov, J. Baungaard, W. Volkl, R. Stouthamer, and M.E.N. Majerus. 1999. Invasion of one insect species, Adalia bipunctata, by two different male- killing bacteria. Ins. Mol. Biol. 8:133-139.
Hurst, G.D.D., T.C. Hammarton, C. Band, T.M.O. Majerus, D. Bertrand, and M.E.N. Majerus. 1997. The diversity of inherited parasites of insects: the male- killing agent of the ladybird beetle Coleomegilla maculata is a member of the Flavobacteria. Genet. Res. 70:1-6. 33
Hurtado, L.A., M. Mateos, R.A. Lutz, and R.C. Vrijenhoek. 2003. Coupling of bacterial endosymbiont and host mitochondrial genomes in the hydrothermal vent clam Calyptogena magnifica. Appl. Environ. Microbiol. 69:2058-2064.
Ishikawa, H. 1987. Nucleotide composition and kinetic complexity of the genomic DNA of an intracellular symbiont of the pea Acyrthosiphon pisum. J. Mol. Evol. 24:205-211.
Katinka, M.D., S. Duprat, E. Corniliot, G. Metenier, F. Thomarat, G. Prensier, V. BArbe, E. Peyretaillade, P. Brottier, P. Wincker, F. Delbac, H. El Alaoui, P. Peyret, W. Saurin, M. Guoy, J. Weissenbach, and C.P. Vivares. 2001. Genome sequence and gene compaction of the eukaryote parasite Encephalitozoon cuniculi. Nature 414:401-2.
Kellner, R.L.L. 2002. Molecular identification of an endosymbiotic bacterium associated with pederin biosynthesis in Paederus sabaeus (Coleoptera: Staphylinidae). Ins. Biochem. Mol. Biol. 32:389-395.
Kellner, R.L.L, and K. Dettner. 1996. Differential efficacy of toxic pederin in deterring potential arthropod predators of Paederus (Coleoptera: Staphylinidae) offspring. Oecologica 107:293-300.
Koga, R-, Tsuchida, T., and T. Fukatsu. 2003. Changing partners in an obligate symbiosis: a facultative endosymbiont can compensate for loss of the essential endosymbiont in an aphid. Proc. R. Soc. Lx)nd. B. 270:2543-2550.
Leonardo, T.E. 2004. Removal of a specialization-associated symbiont does not affect aphid fitness. Ecol. Letters 7:461-468.
Leuchtmann, A. 1992. Systematics, distribution, and host specificity of grass endophytes. Nat. Toxins 1:150-162.
Leuchtmann, A., and K. Clay. 1993. Nonreciprocal compatibility between Epichloe typhina and 4 host grasses. Mycologia 85:157-163.
Martin, W., C.C. Somerville, and S. Loiseauxdegoer. 1992. Molecular phylogenies of plastid origins and algal evolution. J. Mol. Evol. 35:385-404.
May, R.M. and R.M. Anderson. 1982. Coevolution of hosts and parasites. Parasitology 85:411-426.
McLean, D.L., and E.J. Houk. 1973. Phase-contrast and electron-microscopy of mycetocytes and symbiotes of pea aphid, Acyrthosiphon pisum. J. Insect. Physiol. 19:625-629. 34
Messenger, S.L., I.J. Molineux, and J.J. Bull. 1999. Virulence evolution in a virus obeys a trade-off. Proc. R. Soc. Lond. B 266:397-404.
Mira, A., H. Ochman, and N.A. Moran. 2001. Deletional bias and the evolution of bacterial genomes. Trends Genet. 17:589-596.
Montllor, C.B., A. Maxmen, and A.H. Purcell. 2000. Facultative bacterial endosymbionts benefit pea aphids Acyrthosiphon pisum under heat stress. Ecol. Entomol. 27:189-195.
Moran, N. 1996. Accelerated evolution and Muller's ratchet in endosymbiotic bacteria. Proc. Natl. Acad. Sci. USA 93:2873-2878.
Moran, N.A., and P. Baumann. 1994. Phylogenetics of cytoplasmically inherited microorganisms of arthropods. Trends. Ecol. Evol. 9:15-20.
Moran, N.A., C. Dale, H. Dunbar, W.A. Smith, and H. Ochman. 2003. Intracellular symbionts of sharpshooters (Insecta: Hemiptera: Cicadellinae) form a distinct clade with a small genome. Environ. Microbiol. 5:116-126.
Moran, N.A., M.A. Munson, P. Baumann, and H. Ishikawa. 1993. A molecular clock in endosymbiotic bacteria is calibrated using the insect hosts. Proc. R. Soc. Lond. Ser. 8 253:167-171.
Moran, N.A., G.R. Plague, J.P. Sandtrom, and J.L. Wilcox. 2004. A genomic perspective on nutrient provisioning by bacterial symbionts of insects. Proc. Natl. Acad. Sci. USA 100:14543-14548.
Moran, N.A., and A. Telang. 1998. The evolution of bacteriocyte-associated endosymbionts in insects. Bioscience 48:295-304.
Moran, N.A., and J.J. Wemegreen. 1999. Lifestyle evolution in symbiotic bacteria: insights from genomics. Trends. Ecol. Evol. 15:321-326.
Muller, H.J. 1964. The relation of recombination to mutational advance. Mutat. Res. 1:2-9.
Munson, M.A., and P. Baumann. 1993. Molecular cloning and nucleotide sequence of a putative trpDC(F)BA operon in Buchnera aphidicola (endosymbiont of the aphids Schizaphis graminum). J. Bacteriol. 175:6426-6432. 35
Munson, M.A., P. Baumann, M.A. Clark, L. Baumann, N.A. Moran, D.J. Voegtlin, and B.C. Campbell. 1991a. Evidence for the establishment of aphid-eubacterial endosymbiosis in an ancestor of four aphid families. J. Bacteriol. 173:6321-6324.
Munson, M.A., P. Baumann, and M.G. Kinsey. 1991b. Buchnera gen. nov. and Buchnera aphidicola sp. nov., a taxon consisting of the mycetocyte-associated, primary endosymbionts of aphids. Int. J. Syst. Microbiol. 41:566-568.
Nogge, G. 1981. Significance of symbionts for the maintenance of an optional nutritional state for successful reproduction in hematophagous arthropods. Parasitology 82:101-104.
Ochman, H., and N.A. Moran. 2001. Genes lost and genes found: Evolution of bacterial pathogenesis and symbiosis. Science 292:1096-1098.
Ogara, F., and K.T. Shanmugam. 1976. Regulation of nitrogen-fixation by rhizobia export of fixed N2 as NH-4(+). Biochim. Biophys. Acta 437:313-321.
Oliver, K.M., J.A. Russell, N.A. Moran, and M.S. Hunter. 2003. Facultative bacteria in aphids confer resistance to parasitic wasps. Proc. Natl. Acad. Sci. USA 100:1803-1807.
O'Neill, S.L., R Giordano, A.M.E. Colbert, T.L. Karr, and H.M. Robertson. 1992. 16S rRNA phylogenetic analysis of the bacterial endosymbionts associated with cytoplasmic incompatibility in insects. Proc. Natl. Acad. Sci. USA 89:2699-2702.
Ortas, I., P.J. Harris, and D.L. Rowell. 1996. Enhanced uptake of phosphorous by mycorrhizal sorghum plants as influenced by forms of nitrogen. Plant Soil 184:255-264.
Peek, A.S., RA. Feldman, R.A. Lutz, and R.C. Vrijenhoek. 1998. Cospeciation of chemoautotrophic bacteria and deep sea clams. Proc. Natl. Acad. Sci. USA 95:9962-9966.
Piel, J. 2002. A polyketide synthase-peptide synthetase gene cluster from an uncultured bacterial symbiont ofPaederus beetles. Proc. Natl. Acad. Sci. USA 99:14002- 14007.
Prosser, W.A., and A.E. Douglas. 1991. An analysis of chlorotetracycline-treated pea aphid, Acyrthosiphonpisum. J. Insect Physiol. 37:713-719. 36
Rigaud, T., P.S. Pennings, and P. Juchault. 2001. Wolbachia bacteria effects after experimental interspecific transfers in terrestrial isopods. J. Invertebr. Pathol. 77:251-257.
Rigaud, T., C. Souty-Grosset, R. Raimond, J.P. Mocquard, and P. Juchault. 1991. Feminizing endocytobiosis in the terrestrial cnisXactdXiArmadillium vulgarae Latr (Isopoda)—^recent acquisitions. Endocyt. Cell. Res. 7:259-273.
Rousset, F., D. Bouchon, B. Pintureau, P. Juchault, M. Solignac. 1992. Wolbachia endosymbionts responsible for various alterations of sexuality in arthropods. Proc. R. Soc. Lond. Ser. B 250:91-98.
Russell, J.A., A. Latorre, B. Sabater-Munoz, A. Moya, and N.A. Moran. 2003. Side stepping secondary symbionts; widespread horizontal transfer across and beyond the Aphidoidea. Mol. Ecol. 12:1061-1075.
Sandstrom, J., and N.A. Moran. 1999. How nutritionally imbalanced is phloem sap for aphids? Ent. Exp. Appl. 91:203-210.
Sandstrom, J., and N.A. Moran. 2001. Amino acid budgets in three aphid species using the same host plant. Physiol. Entomol. 26:202-211.
Sandstrom, J.P., J.A. Russell, J.P. White, and N.A. Moran. 2001. Independent origins and horizontal transfer of bacterial symbionts of aphids. Mol. EcoL 10:217-228.
Sasaki, T., and H. Ishikawa. 2000. Transinfection of Wolbachia in the Mediterranean flour moth, Ephestia kuehniella, by embryonic microinjection. Heredity 85:130- 135.
Sauer, C., E. Stackebrandt, J. Gadau, B. HoUdobler, and R. Gross. 2000. Systematic relationships and cospeciation of bacterial endosymbionts and their carpenter ant host species: proposal of the new taxon Candidatus Blochmannia gen. nov. Int. J. Syst. Evol. Microbiol. 50:1877-1886.
Schemske, D.W. 1980. The evolutionary significance of extrafloral nectar production by Costus woodsonii (Zingiberaceae): an experimental analysis of ant protection. J. Ecol. 68:959-967.
Shigenobu, S., H. Watanabe, M. Hattori, Y. Sasaki, and H. Ishikawa. 2000. Genome sequence of the endocellular bacterial symbiont of aphids Buchnera sp. APS. Nature 407:81-86.
Siegel, M.R 1990. Fungal endophyte-infected grasses—alkaloid accumulation and aphid response. J. Chem. Ecol. 16:3301-3315. 37
Simon, J.C., S. Carre, M. Boutin, N. Prunier-Leterme, B. Sabater-Munoz, A. Latorre, and R Boumoville. 2003. Host-based divergence in populations of the pea aphid: insights from nuclear markers and the prevalence of facultative symbionts. Proc. R. Soc. Lond. Ser. B 270:1703-1712.
Spaulding, A.W., and C. von Dohlen. 1998. Phylogenetic characterization and molecular evolution of bacterial endosymbionts in psyllids (Hemiptera: Stemorrhyncha). Mol. Biol. Evol. 15:1506-1513.
Stouthamer, R. 1997. ^Fo/6ac/?/a-induced parthenogenesis. In: O'Neill, S., A.A. Hoffinann, and J.H. Werren (eds). Influential Passengers. Oxford University Press: New York.
Stouthamer, R., J.A.J. Breeuwer, R.F. Luck, and J.H. Werren. 1993. Molecular- identification of microorganisms associated with parthenogenesis. Nature 361:66- 68.
Tamas, I., L. Klasson, B. Canback, A.K. Naslund, A.S. Eriksson, J.J. Wemegreen, J.P. Sandstrom, N.A. Moran, and S.G. Andersson. 2002. 50 million years of genomic stasis in endosymbiotic bacteria. Science 296:2376-2379.
Thao, M.L., M.A. Clark, L. Baumann, E.B. Brennan, N.A. Moran, and P. Baumann. 2000b. Secondary endosymbionts of psyllids have been acquired muUiple times. Curr. Microbiol. 41:300-304.
Thao, M.L., N.A. Moran, P. Abbot, E.B. Brennan, D.H. Burckhardt, and P. Baumann. 2000a. Cospeciation of psyllids and their primary prokaryotic endosymbionts. Appl. Environ. Microbiol. 66:2898-2905.
Tsuchida, T., R. Koga, and T. Fukatsu. 2004. Host plant specialization governed by facultative symbiont. Science 303:1989.
Tsuchida, T., R. Koga, H. Shibao, T. Matsumoto, and T. Fukatsu. 2002. Diversity and geographic distribution of secondary endosymbiotic bacteria in natural populations of the pea aphid, Acyrthosiphon pisum. Mol. Ecol. 11:2123-2135.
Unterman, B.M., P. Baumann, and D.L. McLean. 1989. Pea aphid symbiont relationships established by analysis of 16S ribosomal-RNA's. J. Bacteriol. 171:2970-2974. van Ham, R.C., J. Kamerbeek, C. Palacios, C. Rausell, F. Abascal, U. Bastolla, J.M. Frenandez, L. Jimenez, M. Postigo, F.J. Silva, J. Tamames, E. Viguera, A. 38
Latorre, A. Valencia, F. Moran, and A. Moya. 2003. Reductive evolution in Buchnera aphidicola. Proc. Natl. Acad. Sci. USA 100:581-586. von Dohlen, C.D., and N.A. Moran. 2000. Molecular data support a rapid radiation of aphids in the Cretaceous and multiple origins of host alternation. Biol. J. Linn. Soc. 71:689-717.
Weeks, J. A. 2003. Parasitism and ant protection alter the survival of the lycaenid Hemiargus isola. Ecol. Entomol. 28:228-232.
Wemegreen, J.J., and N.A. Moran. 1999. Evidence for genetic drift in endosymbionts (Buchnera): analyses of protein-coding genes. Mol. Biol. Evol. 16:83-97.
Wicker, C. 1983. Differential vitamin and choline requirements of symbiotic and aposymbiotic S. oryzae (Coleoptera: Curculionidae). Comp. Biochem. Physiol. 76A:177-182.
Williamson, D.L. and Whitcomb, R.F. 1979. Sex ratio organisms (spiroplasmas) of Drosophila. In The Mycoplasmas, vol. 3, pp. 175-208. Whitcomb and Tully, eds. Academic Press, New York.
Zchori-Fein, E., and J.K. Brown. 2002. Diversity of prokaryotes associated with Bemisia tabaci (Gennadius) (Hemiptera:Aleyrodidae). Ann. Entomol. Soc. Am. 95:711-718.
Zchori-Fein, E., Y. Gottlieb, S.E. Kelly, J.K. Brown, J.M. Wilson, T.L. Karr, and M.S. Hunter. 2001. A newly discovered bacterium associated with parthenogenesis and a change in host selection behavior in parasitoid wasps. Proc. Natl. Acad. Sci. USA 98:12555-12560.
Zchori-Fein, E., S.J. Perlman, S.E. Kelly, N. Katzir, and M.S. Hunter. 2004. Characterization of a 'Bacteroidetes^ symbiont in Encarsia wasps (HymenopteraiAphelinidae): proposal of''Candidatus Cardinium hertigii'. Int. J. Syst. Evol. Microbiol. 54:961-968. 39
APPENDIX A.
SIDE-STEPPING SECONDARY SYMBIONTS: WIDESPREAD HORIZONTAL
TRANSFER ACROSS AND BEYOND THE APHIDOIDEA
Reprint from Molecular Ecology 12; 1061-1075. 40
MOLECULAR
ECOLOGY VOLUME 12, NUMBER 4. APRIL 2003
CC)MMe«nA8Y ^7 Holarrtic phylo;^eopraphy of the root vol^ (Micrufw implicationif tor late Ouatenwrv biogeography of high Altitudes 805 A 'feirgo" forcorai hvbridiaatton C. Bmnhoa K. E Gaibtcath, V. B. F«dot«v, j. A. Cook k M. Jaacoia O. |. MlUer & M.}. H. Vao Oppm 969 T"PSting PldsbxTtme Sworv. phyiogeographjcal aruilyi
Information on this joumai can be acoesswJ at; http: // wwv»-.blackweilptii>iishing.coin /^mals/mec
The ftnimai is covered by ACRICOLA, Ofmical AifstractH, Ciormt Mmreness Bhh^I Sckttces and Cmrent Confents.
Ti^p»rasih»vaaai^Mat0dcwellSyftrrgif.Visii — WKwill«rk>rriltyntaiJan« —rf. th, «rt> Molecular Ecology (2003) 12,1061-1075 Side-stepping secondary symbionts: widespread horizontal transfer across and beyond the Aphidoidea J. A. RUSSELL/A. LATORRE,+ B. SABATER-MUNOZ,t A. MOYAtandN. A. MORAN"^ *Umversit\/ of Arizom, Department of Ecology and Evolutionary Biology, Biological Sciences Vilest, Tucson, Arizona 85721, USA, fUniversitat de Valencia — Institut Cavanilles, GenHica Evolutiva, Edifici d'Institutos de Patema, Apartado 22085, Valhtcia, Spain Abstract To elucidate the co-evolutionary relationships between phloem-feeding insects and their secondary, or facultative, bacterial symbionts, we explore the distributions of three such microbes — provisionally named the R-type (or PASS, or S-sym), T-type (or PABS), and U- type — across a number of aphid and psyllid hosts through the use of diagnostic molecular screening techniques and DNA sequencing. Although typically maternally transmitted, phylogenetic and pairwise diveigence analyses reveal that these bacteria have been inde pendently acquired by a variety of unrelated insect hosts, indicating that horizontal transfer has helped to shape their distributions. Based on the high genetic similarity between symbionts in different hosts, we argue that transfer events have occurred recently on an evolutionary timescale. In several instances, however, closely related symbionts associate with related hosts, suggesting that horizontal transfer between distant relatives may be rarer than trans mission between close relatives. Our findings on the prevalence of these symbionts within many aphid taxa, along with published observations concerning their effects on host fit ness, imply a significant role of facultative symbiosis in aphid ecology and evolution. Keywords: aphid, PASS, psyllid, R-type, T-type, U-type Received 30 August 2002; revision received 10 December 2002; accepted 10 December 2002 host cells known as bacteriocytes (or mycetocytes). Buch Introduction nera aphidicola supplement the phloem-sap diets of their Many insects and bacteria engage in intimate associations, aphid hosts through the synthesis of essential amino acids or symbiosGS. Molecular analyses such as DNA sequencing (Douglas & Prosser 1992; Baumann et al. 1995; Sandstrom and comparative sequence analyses, which have allowed & Moran 1999). The distributions of these symbionts can for additional approaches such as in situ hybridization and be explained by a pattern of strict vertical transmission diagnostic pol5anerase chain reaction (PCR), have played (Munson etal. 1991; Rouhbakhsh etal. 19%; Clark etal. a key role in the identification of the various microbial 2000; Funk et al. 2000) since the time of a common ancestor, participants. These techniques have also been used to 84-164 million years ago (von Dohlen & Moran 2000). describe a number of intriguing patterns regarding the In addition, many aphids harbour vertically transmitted diversity of bacterial symbionts within a given host clade, bacteria that are not required for growth or reproduction. the distributions of closely related symbionts, the modes of These are often referred to as secondary endosymbionts. transmission which shape these distributions, and the ages Molecular and microscopy studies of several aphid species of associations between hosts and symbionts. have revealed that these symbionts reside in a variety of Aphids (Hemiptera: Aphidoidea) represent an ideal locations inside their hosts, are sjx)radically distributed group of organisms for studies of symbiosis. Nearly all within and across host taxa, and comprise a smaller members of the Aphidoidea harbour obligate endosym- proportion of an embryo's symbiotic flora than does the bionts known as Buchnera aphidicola (Munson etal 1991; primary symbiont, B. aphidicola (Buchner 1965; Unterman but see Fukatsu et al.1994)—vertically transmitted (i.e. from etal. 1989; Chen etal. 1996; Chen & Purcell 1997; Fukatsu parent to offepring) bacteria housed within specialized et al. 2000; Fukatsu 2001; Fukatsu et al. 2001;Sandstrom et al. 2001). Secondary symbionts of aphids belong to a number Correspondence: J. A. Russell. Fax: 520-621-9190; E-mail: of distinct lineages within the y- and a-Proteobacteria [email protected] (Unterman ef al. 1989; Chen al. 1996; Chen & Purcell 1997; © 2003 Blackwell Publishing Ltd 42 1062 J. A. RUSSELL ET AL. Fukatsu 2001; Sandstrom etal. 2001X and the Mollicutes were collected from a variety of locations in the United (Fukatsu etal 2001). Close relatives of aphid secondaries, States and Eurasia. In addition, members of four aphid with 16S rDNA sequence identities ranging from 98% to species were obtained from laboratory stocks. A list of all over 99%, have been found in distantly related arthropod aphid species involved in the survey, their taxonomic species (Chen etal. 1996; Fukatsu etal. 2001; Sandstrom affiliations, and information regarding their collection is et al. 2001). Thus, for the few hosts that have been exam presented in Table 1. ined, the distributions of several secondary symbionts Representatives of 27 psyllid species, from North and appear to be partly the result of horizontal transmission South America, along with Europe(see Thao et al. 2000b for (i.e. movement between hosts other than that from parent collection information), were also surveyed as psyllids to offspring) between host species. However, some evid have been shown to harbor a number of secondary sym ence suggests that certain associations between aphids bionts (Thao et al. 2000a), yet have not been screened for and secondary symbionts are spedalLzed and have co- the R-, T-, or U-types. The samples included were Acizzia evolved (Fukatsu 2001). uncatoides (two samples), Aphalaroida inermis, Bactericera To date, most DNA sequence-based studies of aphid sec cockerelli, Blastopsylla occidentalis, Boreioglycaspis melaleucae, ondary symbionts have focused on hosts within the tribe Cacopsylla brunneipennis, Cacopsylla myrthi, Cacopsylla pere- Macrosiphini. Moreover, although these studies provided grirw, Cacopsylla pyri, Calophya schini,Ctenarytaina eucalypti, some evidence for horizontal transfer (Chen & Purcell Ctenarytaina longicauda, Ctenarytaina spatulata, Diaphorina 1997; Darby etal. 2001; Sandstrom etal. 2001), few have citri, Glycaspis brimblecombei, Heteropsylla cubana, Heterop- employed statistical analyses to demonstrate this point, sylla texam, Neotriozella hirsuta, Psylla buxi, Trioza urticae, and little effort has been made to assess the extent to which Pachypsylla celtidismamma, Pachypsylla venusta, Panisopelma horizontal transfer has shaped secondary symbiont dis sp., Psylla sp., Russelliam intermedia, Spanioneura fonscolombii tributions. Here we use diagnostic molecular screening and Trioza eugeniae. techniques and DNA sequencing, along with pairwise To determine the phylogenetic proximity of related sym divergence and phylogenetic analyses, to explore the dis bionts found within the same host species, the 16S rRNA tributions of three facultative y-Proteobacterial secondary genes of R-, T-,or U-typesymbionts fromseveral Acyrthosi- symbionts of aphids, known as the R-type (or S-sym, or phon pisum clones, maintained in the Moran laboratory at Pea Aphid Secondary Symbiont — PASS), T-type (or Pea the University of Arizona, were sequenced directly from Aphid Bemisia-like Symbiont — PABS), and U-type (Chen PCR products as described below. These aphids were not & r^rcelJ 1997; Unterman etal. 1989; Darby etal. 2001; included in the symbiont survey, as it was known which Sandstrom et al. 2001), across a variety of aphid and psyllid symbionts they harboured prior to the start of the study. (Hemiptera: Psylloidea) species. Specifically, we compare Their 16S rDNA sequences were deposited in GenBank the phylogenies of hosts and symbionts to determine whe under the following accession numbers: AY136138 (U-type ther there is any evidence for co-spedation. Also, we com of A. pisum clone 2a from New York), AY136139 (R-type of pare 16S rDNA divergence values between primary and A. pisum clone 2BB from Wisconsin), AY136140 (R-type of secondary symbionts residing within pairs of hosts sharing A. pisum clone 9-2-1 from New York), AY136141 (T-type the same most recent common ancestor. Differences in of A. pisum clone 8-2b from New York). these values are used to determine the likelihood that cur rent secondary infections could possibly be the result of DNA extractions exclusive vertical transmission since the time of this com mon host ancestor. We also compare the rates of substitu A modified protocol of Bender etal. (1983) was used to tion in the 16S rRNA genes of R-, T- and U-type symbionts extract DNA from American and Israeli aphids. First, a to those of free-living bacteria and B. aphidicola, and esti single aphid was placed in a 1.5-mL tube, which was mate the ages of s5mibiont common ancestors through the dipped into liquid nitrogen. The aphid was then crushed use of a molecular clock. Finally, we use data on symbiont with a plastic pestle. Next, 100-200 of 'lysis buffer' prevalence within and across host taxa, along with obser [0.1 M NaCl, 0.2 M sucrose, 0.1 M Tris-HCl (pH 9.1), 0.05 M vations of symbiont-associated fitness effects, to assess the ethylenediaminetetraacetic acid (EDTA), and 0.5% sodium potential importance of these microbes in aphid biology. dodecyl sulphate] was added to each tube, and the contents were further homogenized. Samples were incubated at 65 °C for 30 min, then 8 M potassium acetate Materials and methods was added to give a final concentration of 1 M. Samples were kept on ice for a minimum of 30 min, then Samples centrifuged at 16,435 g for 15 min. The supernatant was Members of 75 species, spanning 15 subfamilies and tribes decanted into a new 1.5-mL tube and an equal volume of across the Aphidoidea (Remaudiere & Remaudiere 1997) 100% ethanol was added. After 5 min at room temperature. © 2003 BJackwelJ Publishing Ltd, Molecular Ecology, 12,1061-1075 43 DISTRIBUTIONS OF APHID-ASSOCIATED SYMBIONTS 1063 Table 1 Diagnostic molecular screening results reveal the distributions of aphid secondary symbionts Collection Accession Species Code Taxonomy location* R+ T+ Ut number Tamalia coweni Tco Tamaliinae Arizona - - - Thelaxes suberi Tsu Thelaxinae Spain - - - Stegophylla sp. Ste Phyllaphidinae Arizona - - - Chaitophorus populeti Cpe Chaitophorini Spain - - + AY136167 Chaitophorus populifoUi Cpi Chaitophorini Arizona A - - Chaitophorus leucomelas Cle Chaitophorini Spain - - - Periphyllus bulgaricus Pbl Chaitophorini Spain + + - AY136157, Cinara (Cupressobium) cupressi Ccu Cinarini Spain + - - AY136144 Cinara {Cinara) marUimae Cma Cinarini Spain + - - Cinara (Cupressobium) tujafilina Ctu Cinarini Spain + - - AY136146 Essigella califomica Eca Cinarini Arizona A A A Eulachrius pallidus Epa Cinarini Spain - - - AY136147 Lachnus sp. Lac Lachnini Arizona - - - Pterochloroides persicae Ppe Lachnini Spain + - - AY136155 Tuberolachnus salignus Tsa Lachnini Spain - - - Drepanosiphum oregonensis Dor Drepanosiphini Spain - - A Harmmelistes spinosus Hsp Hormaphidiru Washington, D.C. - - - Hormaphis cornu Hco Hormaphidini Georgia A A A Acyrthosiphon lactucae Ala Macrosiphini Arizona A - A Acyrthosiphon pisum ApiF Macrosiphini France* - - - Acyrthosiphon pisum Api E Macrosiphini England* - - - Brachycaudus cardui Bca Macrosiphini Spain - - + AY136143 Brevicoryne brassicae Bbr Macrosiphini Arizona - - - Corylobium avellanae Cav Macrosiphini Spain - - - Diuraphis noxia E>no Macrosiphini Poland - - - Hyperomyzus lactucae Hla Macrosiphini Arizona - A - Macrosiphum euphorbiae 1999 Meu 99 Macrosiphini Arizona A A + Macrosiphum euphorbiae 2001 Meu 01 Macrosiphini Arizona - + + AY136148, Macrosiphum rosae Mro A Macrosiphini Arizona - A + AY136150 Macrosiphum rosae MroS Macrosiphini Spain + - - AY136152 Megoura viciae Mvi Macrosiphini Spain - - - Myzus (Myzus) cerasi Mce Macrosiphini Spain - - - Macrosiphoniella helichrysi Mhe Macrosiphini Spain + - - AY136151 Myzus (Necfarosiphon) persicae Mpe Macrosiphini Spain - - - Roepkea marchali Rma Macrosiphini Spain - - - Sitobion fragariae Sfr Macrosiphini Spain - - - Uroleucon atripes Uat Macrosiphini Arizona - + A Uroleucon nigrotuberculatum Uni Macrosiphini Wisconsin - + - AY136162 Uroleucon pieloui Upi Macrosiphini Wisconsin - + - AY136163 Uroleucon reynoldense Ure Macrosiphijii Wisconsin - + - AY] 36164 Uroleucon rudbeckiae Uru a Macrosiphini Arizona - A + AY136165 Uroleucon rudbeckiae Urub Macrosiphini Arizona - + A AY136166 Uroleucon sonchi Uso Macrosiphini Spain - - - VJahlgreniella nervata Wne Macrosiphini Arizona - - - AY136168 Aphis craccivora AcrS Aphidini Spain + - - AY136137 Aphis craccivora Act A Aphidini Arizona A + A AY136136 Aphis fdbae Afa Aphidini Spain - - - Aphis gossypii Ago A Aphidini Arizona - - - Aphis gossypii AgoS Aphidini Spain - - - Aphis nerii AneS Aphidini Spain - - - Aphis nerii Ane A Aphidini Arizona - - - Aphis ruborum Aru Aphidini Spain - - - Aphis spiraecola Asp Aphidini Spain - - - AY136142 Hyalopterus pruni Hpr Aphidini Spain - - - Hysteroneura setariae Hse Aphidini Arizona A A A Rhopalosiphum padi Rpa Aphidini France* - - - Pemphigus betae Pbe 87 Pemphigini Utah + - - © 2003 Blackwell Publishing Ltd, Molecular Ecology, 12,1061-1075 44 1064 J. A. RUSSELL ET AL. Table 1 Continued Collection Accession Spedes Code Taxonomy location* R+ Tt U+ number Pemphigus betae Pbe 89 Pemphigini Utah - - + AY136154 Pemphigus bursarius Pbr Pemphigini Spain - - - Prociphilus (Meliarhizophagus) fraxinifolii Pfr Pemphigini Arizona - - - Pemphigus nortonii Pno Pemphigini Ohio - - - Pemphigus populi Ppi Pemphigini Spain - - - AY136158 Pemphigus populicaulis Ppc Pemphigini Arizona A A A Pemphigus populiramulorum Ppr Pemphigini Arizona - - - Pemphigus spyrothecae Psp Pemphigini Spain - + - AY136156 Thecabius populimonilis Tpo Pemphigini Arizona - - - Eriosoma lanigerum Eli Eriosomatmi Arizona - - - Eriosoma (Schizoneura) lanuginosum Elu Eriosomatini Spain - - - Eriosoma (Schizoneura) ulmi Eul Eriosomatini Spain - - - Tetraneura (Tetraneurella) akirine Tak Eriosomatini Spain - - - Tetraneura caerulescens Tea Eriosomatini Spain - - - Tetraneura ulmi Tul Eriosomatini Spain - - - AY136160 Baizongia pistaciae Bpi Fordini Spain - - - Forda formicaria Ffo Fordini Spain - - - Geoica utricularia Gut Fordini Spain - - - Geopemphigus sp. Geo Fordini Texas A + A Melaphis rhois Mrh Fordini Arizona A A A Schlechtendalia chinensis Sch Fordini Japan A A A Smynthurodes betae She Fordini Israel + N N AY136159 Chromaphis juglandicola Cju Myzocallidini Spain - - - Melanocallis caryaefoliae Mca Myzocallidini Arizona* A A A Monellia caryella Mcl Myzocallidini Arizona* A A A Myzocallis sp. Myz Myzocallidini Arizona - - - AY136153 Panaphis juglandis Pju Myzocallidini Sf>ain A - - Therioaphis trifolii f, maculata Ttr Myzocallidini New York - - - Tinocallis sp. Tin Myzocallidini Spain - - - Pterocomma populeum Ppo Pterocommatinae Spain - - - ^Samples derived from laboratory stocks. +A = ambiguous results for this symbiont; N = sample was not screened for this symbiont. samples vv^ere again centrifraged at 16,435 g for 15 min. The described (Moran etal 1999, US and Israeli aphids; van supernatant was discarded, and the pellet v^as washed Ham etal. 1997, European aphids), and all reactions in ice-cold 70% ethanol, then in ice-cold 100% ethanol. included a 'negative' control (sterile water instead of DNA) Samples were dried in a SpeedVac® and were then so as to SfHJt any possible contamination. A 'positive' suspended in 'low-TE' (1 mM Tris-HCl and 0.1 mM control (DNA from an aphid known to harbour the EDTA). relevant symbiont) was included to prevent false negatives. For European aphids, DNA was extracted from single The diagnostic primers used here (Table 2) have been individuals as previously described (Latorre et al. 1986). previously published or were newly designed based on an Psyllid DNA samples were derived from multiple indi alignment of R-, T- and U-type 168 rDNA sequences with viduals (Thao et al. 2000b). those of several closely related y-Proteobacteria. Primers spanning the intergenic spacer between the 16 and 23S rRNA genes generated PCR products that were of Diagnostic screening and DNA sequencing characteristic lengths for the R-, T- and U-type symbionts, To test for the presence of secondary symbionts, DNA which facilitated our assessment of symbiont presence/ samples were subjected to several diagnostic PCR reactions, absence. Diagnostic PCR products (obtained from R250F, each involving a diagnostic forward and 'universal' reverse T99F, or U99F with 480R) from selected samples were sub primer, which amplify a fragment of the 16-23S rRNA jected to restriction digests with the enzymes Sstl/Sad,Xbal operon. PCR reactions were carried out as previously and Clal, to further confirm infection status (Sandstrom 2003 Blackwell Publishing Ltd, Molecular Ecologx/, 12,1061-1075 45 DISTRIBUTIONS OF APHID-ASSOCIATED SYMBIONTS 1065 Table 2 Primer sequences, their positions Primer Gene and in 16S or 23S rRNA genes, and their uses in name* Primer sequence (5' —> 3') positiont Uses$ study R1279F CGAGAGCAAGCGGACCTCAC 16S; 1267-1286 DP T1279F CGAGGGAAAGCGGAACTCAG 16S; 1267-1286 DP U1279F CGAACGTAAGCGAACCTCAT 16S; 1267-1286 DP R250I^ GGTAGGTGGGGTAACGGCTC 16S; 250-269 DP; AS; S T99F§t AGTGAGCGCAGTTTACTGAG 165; 75-87 DP; AS; S U99F§ ATCGGGGAGTAGCTTGCTAC 165; 71-90 DP; AS; S 480R§ CACGGTACTGGTTCACTATCGGTC 23S; 453-476 DP; AS 1502R GTTACGACTTCACCCCAG 16S; 1484-1502 S 10F§ AGTTTGATCATGGCTCAGATTG 16S; 10-31 S; AS 559F CGTGCCAGCAGCCGCGGTAATAC 16S; 514-536 S *Names of diagnostic primers begin with an 'R', T', or U', depending on the symbiont they were designed for. tPositions of primers within the 16S and 23S rRNA genes are with respect to the sequences of Escherichia coli (GenBank accession nimibers AE000460 and AE000453, respectively). $DP = diagnostic PGR; S = sequencing; AS = amplification for products to be sequenced. §Previously described in Sandstrom et al. (2001). IPrimer spans a six base pair insertion imique to the T-type. etal. 2001). All PGR and restriction digest products were 16S rDNA sequences were at least 95% similar to a run on agarose gels, stained in ethidium bromide and visu previously described R-, T-, or U-type s3Tnbiont. Similarly, alized with UV light. we will refer to all bacteria discovered here as R-, T-, or U- PGR products, spanning portions of the 16S and 23S type symbionts only if they are at least 95% similar to all rRNA genes, were obtained for samples that amplified previously described members of these clades as revealed in diagnostic screening reactions. These products were by BLASxn searches. We use this 95% cut-off criterion as all purified, and sequenced directly (Sandstrom etal. 2001, sequences that fell into a monophyletic group with the R-, American aphids), or after cloning (Marchuk etal. 1992; T-, or U-types were less than 5% divergent from all other Martinez-Torres et al. 2001, Eurof)ean aphids). All sequen sequences within their respective clades. cing was performed at the University of Arizona GATG In contrast, samples were scored as negative if they did Sequencing Facility. Primers used in the sequencing reac not amplify, or yielded improperly sized products, in two tions are described in Table 2. relevant diagnostic PCR reactions (e.g. no amplification with R250F and 480R or with R1279F and 480R indicated a sample that was negative for the R-type). Alternatively, Assessment of diagnostic methods samples were declared negative if the partial 16S rDNA Diagnostic PGR reactions gave amplification in a number sequence of a diagnostic PGR product was less than 95% of aphid and psyllid species. However, not all amplified similar to previously described secondary symbionts. To products were identical in length to those of positive R-, T- avoid potential misdiagnosis, the few samples not meeting and U-type controls from A. pisum or to samples in which each of the criteria required to be declared negative or pos these bacteria were diagnosed through DNA sequencing itive were considered ambiguous. (Fig. 1). In particular, the primer R250F exhibited several cases of nonspecific amplification, which resulted in 16S rDNA phytogeny the discovery of 16S rDNA sequences similar to those of several free-living bacteria (these were considered likely Overlapping contigs from sequencing reactions were contaminants) along with symbionts of other insects. assembled using S£QM/4NII (DNASTAR 1997). Sequences Sequencing and BLASTTI searches revealed that the com were then compared to those in GenBank using BLASTn bination of diagnostic PGR and restriction digest reactions searches. Sequences at least 95% similar to all previously was sufficient to identify symbionts, as no samples tested described R-, T-, or U-type 16S rDNA sequences were positive with both of these techniques yet lacked a 16S assigned to the relevant symbiontgroup. The resultsof BLASTO rDNA sequence more than 95% similar to previously searches were also used to choose other symbiotic and free- described R-, T-, or U-type s5nnbionts (data not shown). living bacteria for inclusion in subsequent phylogenetic Thus, samples were scored as positive (Table 1) only if they analyses. Completed sequences were submitted to GenBank tested p)ositive with both of these methods, or if associated and their accession numbers are presented in Table 1. © 2003 Blackwell Publishing Ltd, Molecular Ecology, 12,1061-1075 46 1066 J. A. RUSSELL ET AL. (") bionts B. aphidicola (primary symbiont of the aphid Myzus rpe Ccu Mhsyllids G. brimblecombei (AF263561), H.texam (AF263562), B. cockerelli (AF263557), and B. occidentalis (AF263558); the whiteflies KXWbp (Hemiptera: Aleyrodoidea) Aleurodicus dugesii (AF286129) and Bemisia tabaci (Z11926); the aphids Uroleucott ambrosiae (AF293621, AF293622), Uroleucon rudbeckiae (AF293626), A. pisum (AF293617, AF293616, AF293618, AB033779, AB033778, psp rw A pi AB033777, AJ297720, M27040), Uroleucon helianthicola (AF293625), Uroleucon astronomus (AF293623), Alacrosiphoti- iella ludovidatuie (AF293619), Uroleucon solidaginis (AF293627), Uroleucon aenum (AF293620), and Uroleucon caligatum 3000 bp (AF293624); along v^th a symbiont of the beetle Chilomenes sua bp 1600 bp sexmaculatus (AJ272038). 16S rDNA sequences from the following free-living bacteria were also included: Pseu- iCKlDbp domonas aeruginosa (AF094720), Vibrio cholerae (X74695), Pantoea ananatis (Z96081), Escherichia coli (AB045731), Ejiter- obacter cloacae (Y17665), Serralia entomophila (AJ233427), Serratia ficaria (AJ233428), Yersinia enterocolitica (Z75316), Klebsiella pneumoniae (Y17657), and Proteus vulgaris (AJ233425). Fig. 1 Results of diagnostic molecular surveys. Brackets encompass lanes corresponding to different aphids, and abbreviations above Two analyses were performed in PAUP* 4.0b8 (Swofford brackets are as listed in Table 1. (a) Gel picture of diagnostic R- 2001) — an heuristic search using parsimony and a neighbour- type PGR and restriction digests. Lanes1 and 20 contained a 1000- joining search using distance (with the Kimura two- bp ladder (Gibco BRL) size standard. Products from R250F and parameter model of nucleotide substitution).One thousand 480R are found in lanes 2, 5, 8, 11, 14 and 17 (undigested), and bootstrap replicates were performed for each search, to lanes 3,6,9,12,15and 18 (digested with Socl/Ssrt). Products from estimate the level of support for nodes. Trees were rooted R1279F and 480R are found in lanes 4, 7, 10, 13, 16 and 19 using P. aeruginosa as the outgroup. (undigested). Note that Aspand Epa samples did not amplify with R1279F and 480R, nor were their R250F and 480R products cut with Sflcl/Ssfl to yield products of the expected sizes. This was Phylogenetic comparisons consistent with oursequencing results, which indicated the absence of the R-type. (b) Gel of diagnostic T- and U-type PGR and restric Phylogenies of several aphids in the tribe Macrosiphini tion digests. Lanes 1 and 21 contained the 1000-bp size standard, and their symbionts were compared using the RELL while lane 11 contained a lOG-bp ladder (Amersham Biosciences). (Resamphng Estimate of Log-Likelihood) method (Kishino Products obtained with U99F and 480R are found in lanes 2,5and & Hasegawa 1989), to determine the p)ossibility of sym 8 (undigested),and lanes 3,6 and 9(digested with ClaT).Undigested products obtained with the primers U1279F and 480R are found in biont persistence through strict co-speciation. This test lanes 4, 7 and 10, while those obtained with the primers T1279F gives a bootstrap probability for candidate trees without and 480R are found in lanes 14,17 and 20. Products from reactions {performing maximum likelihood estimation for each with T99F and 480R are found in lanes 12,15 and 18 (undigested) resampled data set, and can be used to compare the and in lanes 13,16 and 19 (digested with Xbal). likelihood of alternative tree topologies given the nucleotide sequence data. We specifically considered the likeli The 16S rDNA sequences obtained here, along vdth hood that the phylogeny of several U-type symbionts those of other insect symbionts (including all previously is identical, or congruent, to that of their hosts. Included published aphid R-, T- and U-type sequences) and several in these analyses were the hosts A. pisum, M. ludovicianae, free-living bacteria, were aligned in SEQUENCE NAVIGATOR U. helianthicola, U. solidaginis, U. aenum,U. rudbeckiae and 1.01 (Applied Biosystems 1994). Aligned sequences were U. astronomus, as well as their U-type associates. Parsimony imported into MACCLADE 4.03 (Maddison & Maddison searches were used to suggest a likely U-typ>e symbiont 2000) where the alignment was adjusted manually. Pub tree topology for this subset of taxa — {[(U. solidaginis, U. lished 16S rDNA sequences were included from the sym aenum) (U. rudbeckiae, A. pisum)] iU. astronomus, U. helianthicola), © 2003 Blackwell Publishing Ltd, Molecular Ecohgx/, 12,1061-1075 47 DISTRIBUTIONS OF APHID-ASSOCIATED SYMBIONTS 1067 M. ludovicianae). All nodes in this tree were supported by performed as described by Robinson etal. (1998). The bootstrap values of 80 or higher. We then compared the alignment file created for relative rates tests was also approximate bootstrap probability of this topology to used to compute pairwise divergences between the 16S those of three possible host trees based on mitochon rDNA sequences of free-living and symbiotic bacteria. drial and nuclear gene sequences (from Figs Ic in Moran Both the absolute number of differences and Kimura et al. 1999) — {[(li. solidaginis, U. aenum (U. astronomus, two-parameter corrected distances were computed using U. rudbeckiae)), U. helianthicola], M. ludovicianae, A. pisum]. PAUP* 4.0b8. All RELL analyses were executed in MOLPHY 2.3 (Adachi & Hasegawa 1996). Primary and secondary symbiont pairwise divergence comparisons Relative rates and pairwise divergences Finally, pairwise divergence comparison analyses were To determine the rates at which symbiont 16S rRNA performed in an attempt to help determine whether R-, genes are evolving, relative rates tests were performed. T-, or U-type symbionts could have persisted within The 16S rDNAsequences included all exceeded 1300 nucleo host lineages through strict vertical transmission since the tides in length and were derived from the R-type time of a common host ancestor (i.e. whether they have symbionts of A. pisum (six individuals) and U. caligatum; undergone strict co-speciation). If such a scenario has the T-type symbionts of Uroleucon nigrotuberculatum, U. occurred, genetic divergences between secondaries infecting ambrosiae (two individuals), U.rudbeckiae (two individuals), a pair of host spedes from two distinct taxa should Uroleucon pieloui, Uroleucon reynoldense, Aphis craccivora, resemble those between B. aphidicola from the same taxa. A. pisum (two individuals), Periphyllus bulgaricus, C. Specifically, because B. aphidicola undergoes strict vertical sexmaculatus and B. tabaci; the U-type symbionts of U. transmission (Munson etal. 1991; Rouhbakhsh etal. 1996; solidaginis, Brachycaudus cardui, Macrosiphum euphorbiae (two Clark et al 2000; Funk et al. 2000), and because bacterial 16S individuals), Macrosiphum rosae, M. ludovicianae, U. aenum, rRNA substitution rates range from 1 to 4% per hundred Pemphigus betae, A. pisum, U. astronomus and Chaitophorus million years (Moran etal. 1993; Ochman & Wilson 1987; populeti; B. aphidicola; and the free-living bacteria E. coli, S. Ochman et al. 1999), differences in divergences (between entomophila, S. ficaria and V. cholerae. This trimmed-down B. aphidicola and secondaries in equi-distant hosts) should alignment file (otherwise identical to that used for the 16S not exceed four-fold under strict vertical transmission of rDNA phylogeny) was opened in PAUP*b8, where missing secondary symbionts. Violations of these conditions indicate and ambiguous sites were removed. Vibrio cholerae was horizontal transfer; however, cases in which conditions defined as the outgroup, and a guide tree was constructed are not violated do not imply co-speciation, as similar (using parsimony and 100 bootstrap replicates) for divergences could simply be chance. This approach does weighting in relative rates tests (Robinson et al. 1998). The not require knowledge of phylogenetic relatedness and, bootstrap 60% majority-rule consensus tree used in relative thus, provides an alternative approach to examining the rates tests had the following topology: (V. cholerae, ((((/I. possibility of co-spedation. pisum California R, A. pisum Japan R, A. pisum Japan R, A. We constructed an alignment including several R-, pisum Japan R, A. pisum Wisconsin R, A. pisum Arizona T- and U-typ>e 16S rDNA sequences from symbionts asso- R), U. caligatum R) (S. ficaria, S. entomophila)), E. coli dated with hosts from the tribes Chaitophorini, Macro- ((((U. solidaginis U (B. cardui U, C. populeti U), U.aenum V, siphini, Pemphigini and Aphidini, and from the genera P. betae U) (M. euphorbiae U, M. euphorbiae U), M. rosae U, A. Acyrthosiphon and Uroleucon. Also included were 16S rDNA pisum U, U. astronomus U), M. ludovicianae U) (((U. sequences of B. aphidicola symbionts from spedes in the nigrotuberculatum T,U. ambrosiae Arizona T, U. reynoldense same tribes and genera: A. pisum (M27039), Diuraphis noxia T, U. pieloui T, U. ambrosiae Minnesota T, C. pyri T) (U. (M63251), M. persicae (M63249), and Uroleucon souchi (M63250) rudbeckiae T, U. rudbeckiae T, A. pisum New York T, A. pisum of the Macrosiphini; Chaitophorus vimimlis (M63252) of the Wisconsin T), P. bulgaricus T, C. sexmaculatus T), A. Chaitophorini; P. betae (M63254) and Pemphigus populi craccivora T, B. tabaci T)), Buchnera)). The ungapped and (AJ296750) of the Pemphigini; and Rhopalosiphum maidis unambiguous 1301 base pair alignment was opened (M63247), Rhopalosiphum padi (M63248), and Schizaphis in RRTREE (Robinson-Rechavi & Huchon 2000; http:// graminum (M63246) of the Aphidini. The alignment was pbil.univ-lyonl.fr/software/rrtree.html). 16S rRNA rates performed as described above. Pairwise distances were of substitution were compared between R-type, T-type, then calculated over 1296 nucleotides (no missing or U-type, B. aphidicola, E. coli and Serratia species with ambiguous sites) using the Kimura two-parameter model V. cholerae defined as the outgroup. Divergences were of nudeotide substitution, and compared between B. computed using the Kimura two-parameter model of aphidicola and secondaries infecting one member, each, of the nucleotide substitution, and relative rates tests were same two host taxa. For example, the 16S rDNA divergence © 2003 Blackwell Publishing Ltd, Molecular Ecology,12,1061-1075 1068 J. A. RUSSELL ET AL. between B. aphidicola of P. populi (Pemphigini) and M. (Table 1), adding to previous findings of the T-type in the persicae (Macrosiphini) was compared to that between related North American species Uroleucon ambrosiae, Uro the U-type symbionts of P. hetae (Pemphigini) and M. ludo- leucon astronomus and Uroleucon rudbeckiae (Sandstrom el al. viciame (Macrosiphini). When sequences were available 2001). for symbionts of more than one host species within a given taxon the smallest B. aphidicola divergence value was Phylogenetic analysis compared to the largest secondary symbiont divergence value. This minimized the differences in 16S rDNA diver The 16S rDNA phylogeny of aphid- and psyllid-associated gences, making any conclusions of horizontal transfer bacteria described here, plus their symbiotic and free- quite conservative. living relatives, is presented in Fig. 2. Several bacterial sequences obtained from aphids (and one from a psyllid) fell into one of three strongly supported clades termed Results R [parsimony (BS) = 72], T (parsimony BS = 100), and U (parsimony BS = 100). Within these clades, relationships Symbiont distributions were often unresolved because of high genetic similarity. The results of our diagnostic molecular screening sur As previously demonstrated (Sandstrom et al. 2001), the T- vey are summarized in Table 1. Each of the R-, T- and U- and U-types were sister clades (parsimony BS - 92), and type symbionts was found in one or more previously members of the R-clade clustered with species of the genus unscreened members of the Macrosiphini. The R-type was Serratia (parsimony BS = 94). found in members of six new aphid tribes — the Aphidini, Additionally, 16S rDNA sequences of aphid-associated Chaitophorini, Cinarini, Fordini, Lachnini and Pem bacteria obtained here fell into three other clades con phigini. The T-type was found in one psyllid, Cacopsylla taining previously described insect endosymbionts pyri (GenBank accession no. AY136145), and in four new (Fig. 2). First, bacteria from the distantly related aphids aphid tribes —the Aphidini, Chaitophorini, Fordini and Aphis spiraecola, a Myzocallis sp., and Wahlgreniella nervata, Pemphigini. Finally, the U-type was found in two new aphid formed a strongly supported clade (parsimony BS = 100) tribes — the Chaitophorini and Pemphigini. In total, within which included Arsenophonus triatominarum (Hj^jsa &Dale the Aphidoidea the R-type was foimd in 10/79 surveyed 1997), a symbiont of the kissing bug Triatoma infestans species, the T-type in 11/78, and the U-type in 6/76. (Hemiptera: Reduviidae),along with secondary symbionts Within the Psylloidea, 0/27 surveyed species were found of several psyllids (Thao et al. 2000a) and a whitefly to associate with the R- or U-types, while only 1 /27 psyllid (Spaulding & von Dohlen 2001). In our analysis, this clade species was positive for the T-type. Though more aphid was the sister taxon to the free-living bacterium Proteus vul and psyllid hosts tested negative than positive for each garis. We will refer to this symbiotic group as Ars — short for symbiont, it should be noted that negative results here do Arsenophonus, the genus name of the T. infestans symbiont not imply that the symbionts of interest are absent from a and of the male-killing symbiont of Nasonia vitripennis species, as only one to a few individuals were surveyed. (Ghema et al. 1991), to which our bacteria were also related These results indicate that R-, T- and U-type symbionts (data not shown). Next, a bacterium associated with are certainly not confined to members of the Macrosiphini, Eulachnus pallidus fell within a clade which included sym or even to the superfamily Aphidoidea in the case of the bionts of tsetse flies, weevils and some psylUds (parsimony T-type. Indeed, BLASTO searches using T-type 16S rDNA BS = 77). We will refer to this clade as So-So, short for sequences revealed that such molecules share a high Sitophilus oryzae Primary Endosymbiont (Heddi et al. 1999) degree of similarity (over 98%) with that of a putative and Sodalis glossinidius (Dale & Maudlin 1999) — names for male-killing bacterium described from the ladybird beetle, two symbionts within this group. Finally, bacteria from Chilomenes sexmaculatus (MEN Majerus personal commu Pemphigus populi and Tetraneura ulmi formed a mono- nication; GenBank accession no. AJ272038). Also, T-type phyletic clade (parsimony BS = 100) nested within the clade 16S rDNA sequences have previously been shown to share including theT- and U-types (parsimony BS = 92).We will refer over 98% identity with that from a secondary symbiont of to these bacteria as the V-type. While we use the term symbiont the whitefly, Bemisia tabaci (Sandstrom et al. 2001). when referring to newly described aphid associates of the Diagnostic screening revealed two cases of potentially Ars, So-So and V clades, we point out that further research stable associations between a S)mibiont tjq^e and a host on their transmission and localization is required for defin clade. First, all three surveyed members of the genus itive proof and elucidation of their symbiotic lifestyles. Cinara were found to harbour the R-type (Table 1). Also, S3mibiont 16S rDNA phylogenies conflicted with phylo- three previously unscreened Nearctic members of the sub genies of their hosts. For example, the T-type symbionts genus Uroleucon — U. pieloui, U. reynoldense and U. nigrotu- of several members of the genus Uroleucon were more berculatum — were found to associate with the T-type closely related to a T-type symbiont from the psyllid C. pyri © 2003 Blackwell Publishing Ltd, Molecular Ecology, 12,1061-1075 49 DISTRIBUTIONS OF APHID-ASSOCIATED SYMBIONTS 1069 Fig. 2 Maximum parsimony phylogeny of Myzoca/tfs sp. SS /uihts spiraecolaSS aphid-assodated bacteria and their relatives. dycMpte MmMpcomlM/SS One iiii-i H&ttnptyHattxattaSS thousand bootstrap replicates were Wah^renhUanenataSS performed for both distance and parsimony Afaaoeehonua triatomkmvm AJaurodfeuadugaaUSS searches. Bootstrap valuesfrom the parsimony Protaaswiaaris U/o/eucon ambmaiaeSS MN search are located above the relevant Lfmiauconrayno/dansaSS _ I CaeopayOa pyrtSS branches (or arrows), whUe those from the ^—• r Urolaucon r»grottri>erculatum SS distance analysis in agreement with the Urotoucon amb/osiaa SS AZ UrotaoconptalotMSS parsimony tree topology are located below Aeyrtho^pnon ptEsum SS England r ptsum SS Wt branches (or arrows). Names of free-living AeytthosiphonpisumSSH^ bacteria are in underlined italics, names of UrohuoonnidbaMaaSS - ChKamanaa maxtnaeukUm aphid symbionts (named after their hosts, Aohis cracdvoraSS AZ except for Buchnera aphidicola) are in plain -abaci SS gopftortwiagSS'Ol italics, names of symbionts of other insects (all but Sodalis glossinidius and Arsenophonus triatominarum are named after their hosts) Macmslphum FOsaaSS are presented in bold itahcs. Bacteria which . Macf08tohumeu3h0tt>laaSS'99 * Macfo^jhum at^hotblaa SS *01 are considered secondary symbionts (i.e. Acyt^io^honpiaum^W Urohucon rudbaddaa SS they reside in a host with an obligate, A^1hos^)honpi8tmSSW - Pemphigus betae SS primary symbiont) are denoted by an 'SS' Urolaucon haSanthicola SS Urolaucon astronomusSS following their listed names. Clades including LfrofeuconMacro^)honlataludovldanaaSS 5O0da0frM SS such symbionts are called R, T, U, V, Ars OialtophomspopiMaeSS and So-So, and are encompassed by bars on Brachji^uduacanMSS Ur^aucon aanum SS the right. Several aphid species names are Ya^^/yaaiOaroeomca present in multiple positions on the tree. To ^TswrBfaficana CirtaracuprasaiSS distinguish these, collection locations AcyrthoaiphonpiaumSS Japan 79 |->4<9iTtftoftb/Nvipfe<«nSSJapan (NY = New York, W1 = Wisconsin, AZ = 78^ Acyitu^fhonplaurnSS Japan Arizona, CA = California, MN = Minnesota) pfeum SSCA or dates ('01 = 2001, '99 = 1999) are placed plaumSSHY ip/bumSSAZ at the ends of host names. ApNs cracdvora SS Spain 7] — tfTDtouoon SS — ^nare b^^s/Mns SS MBcrDS49ten^/MAic/My»'SS ' xxiaabataaSS ly^buharicusSS -cMonmss pe/steaa SS Sactwfeafa eeeiMfitaf SS EtdaehnuspaWduaSS BtaatopayUaocddaniaKaSS SltopfiMua oryzaa SedaUaghaalniauaSS EnOfabactBrdoacae Eacherichiacoli Buchnara aphkfioola VibrioeholarM •'^audomonasaafualnoaa - lOchangM than to those from other Uroleucon species (Fig. 2). Also, These included the R- and U-type symbionts from Acyrthosi- the T-like symbionts of a whitefly and a ladybird beetle did phon pisum and the U-type from Macrosiphum euphorbiae. not occupy a basal position amongst the T-type symbionts, Whereas the A. pisum R-type and M. euphorbiae U-type were as would be exf)ected under co-spedation. A direct compar each monophyletic (parsimony BS = 79 and 100, respectively), ison between the phylogenies of closely related hosts and the A. pisum U-type sequences fell into a clade which also their U-type symbionts revealed several inconsistencies included the U-type of U. rudbeckiae (parsimony BS = 86). (Fig. 3) —at least four rearrangements of ttie (unrooted) symbiont tree are required to make this tree match that of the Relative rates and pairwise divergences hosts. Finally, the RELL test relative support value for the symbiont topology was a striking 100% (data not shown) — R-type 16S rRNA genes evolved at a rate consistent with indicating that this tree topology was far more likely than those of free-living bacteria such as Escherichia coli, and any of the host-tree topologies given our U-type16S rDNA Serratia, but more slowly than those of Buchnera aphidicola, dataset. the T-type, and the U-type (Table 3). T-type 16S rDNA Despite these patterns of incongruency, several sym sequences were found to evolve at a rate not significantly bionts from conspecific aphids, collected from different different from U-type or B. aphidicola 16S rDNA sequences, geographical locationsor at different times,were highly related. but significantly faster than 16SrDNA of free-living lineages. ® 2003 Blackwell Publishing Ltd, Molecular Ecology, 12,1061-1075 50 1070 J. A. RUSSELL ET AL. Aphid Host U-Type Synirfont Table 3 Relative rates tests on the 16S rRNA genes of secondary symbionts Api Api -••Uni Uru Taxon 1 Taxon 2 KI3 K,3-K23±SD Uas Uae Uae UsI R-type E. coli 0.0963 0.0985 -0.0023 ± 0.0075 0.0963 0.1233 -0.0271 ± 0.0087+ Uas R-type U-type R-type Serratia 0.0963 0.1001 -0.0039 ± 0.0048 R-type T-type 0.0963 0.1265 -0.0302 ± 0.0084t Mlu MhJ R-type B. aphidicola 0.0963 0.1487 -0.0524 + 0.0117+ E. coli U-type 0.0985 0.1233 -0.0248 ± 0.0099+ Fig. 3 Incongruency between host and symbiont trees reveals E. coli Serratia 0.0985 0.1001 -0.0016 ± 0.0070 horizontal transfer of U-type symbionts of the Macrosiphini. The E. coli T-type 0.0985 0.1265 -0.0280 ± 0.0098+ host phylogeny (left) is redrawn from Fig. 1(c) of Moran et al. 1999. E. coli B. aphidicola 0.0985 0.1487 -0.0502 ±0.0108+ The unrooted U-type symbiont phylogeny (right) was obtained Scrratia U-type 0.1001 0.1233 -0.0232 ± 0.0086+ through parsimony searches including only the U-type16S rDNA U-type T-type 0.1233 0.1265 -0.0031 ± 0.0083 sequences of the taxa shown here. Dashed arrows connect hosts U-type B. aphidicola 0.1233 0.1487 -0.0254 ± 0.0115+ to their respective symbionts (named after hosts), and suggest Serrafia T-type 0.1001 0.1265 -0.0263 ± 0.0086+ several inconsistencies. This host topology was one of just three Scrratia B. aphidicola 0.1001 0.1487 -0.0486 ±0.0112+ possible alternatives because of a trichotomy involving (Uae) T-type B. aphidicola 0.1265 0.1487 -0.0222 ±0.0116 (Usl), and (Uru, Uas); the other topologies were also incongruent with that of the symbiont tree (not shown). Name abbreviations B. aphidicola, Buchnera aphidicola. not presented in Table 1 are as follows: Mlu = Macrosiphoniella *K^3 = 16S rDNA distance between taxon x and the outgroup, ludovicianae, Uhe = Uroleucon helianthicola, Usl = Uroleucon solida- taxon 3 (i.e. Vibrio cholerae). ginis, Uae = Uroleucon aenum, Uas = Uroleucon astronomus. tDifference is significant at a = 0.05. Finally, U-tyf)e 16S rDNA sequences evolved morequickly than did those of free-living bacteria, yet more slowly than infection in A. pisum and U. caligatwn (Acyrfhosiphoti vs. B. aphidicola 16S rDNA. Uroleucon). Preliminary analyses suggested that many R-type Pairwise divergence analyses of symbiont and free-living 16S rDNA sequence divergences were inconsistent with bacteria revealed that secondary symbionts, of a given type, co-spedation. For example, divergences between sym show high levels of sequence similarity (table available bionts infecting hosts of different tribes were as low as from corresponding author). Within the R-, T- and U-clades, 0.44% when compared over approximately 1000 base all compared sequences were less than 0.5, 1.6 and 2.2% pairs. However, because these sequences were incomplete divergent, respectively. Strikingly, T-t)^ symbionts in the they were excluded from formal divergence analyses. jDsyllid C. pyri and several Nearctic aphids of the subgenus Uroleucon were identical over 1301 compared nucleotides. Discussion The U-clade was the most genetically diverse, with an average of 1.2% 16S rDNA divergence between symbionts Symbiont diversity in different host species compared to 0.4% for the T-type. Previous studies that did not include DNA sequence information suggested that a number of aphid spedes Pairwise divergence comparisons harbour vertically transmitted facultative bacterial asso- Comparisons between the16S rDNA distances of B. aphidicola dates that reside in various host tissues (e.g. Buchner and secondary symbionts infecting pairs of hosts sharing 1965; Fukatsu & Ishikawa 1998). Observations on their the same most recent common ancestor are presented in dis-tributions revealed that they had been independently Table 4. In all but one case, the divergences between B. acquired by, and possibly lost from, aphid lineages. aphidicola sequences were at least twice as large as those However, because the bacteria were not identified, no between secondary symbionts of equidistant hosts. Only conclusions could be made regarding symbiont diversity, 4/12 comparisons were consistent with the possibility the phylogenetic breadth of the hosts infected by related of strict secondary symbiont co-speciation, under the microbes, or the likelihood of occasional co-sp)edation. assumption that 16S rDNA substitution rates should not Including the six groups of bacteria described here, aphids differ by more than four-fold: the U-type infections in have been reported to assodate with secondary symbionts Pemphigus betae and Macrosiphoniella ludovicianae (Pemphigini spanning 10 distinct bacterial lineages (Unterman etal. vs. Macrosiphini), Chaitophorus populeti and M. ludovicianae 1989; Chen et al. 1996; Jeyaprakash & Hoy 2000; Darby (Chaitophorini vs. Macrosiphini), and A. pisum and U. etal. 2001; Fukatsu 2001; Fukatsu etal. 2001; Sandstrom astronomus (Acyrthosiphon vs. Uroleucon); and the R-type etal. 2001). Given our serendipitous discovery of three © 2003 Blackwell Publishing Ltd, Molecular Ecology, 12,1061-1075 51 DISTRIBUTIONS OF APHID-ASSOCIATED SYMBIONTS 1071 Table 4 16S rDNA divergences of Buchnera B. aphidicola Secondary Fold aphidicola vs. those of secondary symbionts Host taxa 165 divergence* 16S divergencet Difference for pairs of aphid hosts sharing same most recent common ancestor (see text for details) Pemphigini-Macrosiphini 0.0587 (Ppi-Mpe)t 0.0188 (PbeU-MluU)§ 3.1 0.0039 (PspT-Api"0§ 15.1 Pemphigini-Aphidini 0.0520 (Pbe-Sgr)t 0.0031 (PspT-AcrT)§ 16.8 Chaitophorini-Macrosiphini 0.0572 (Cvi-Mpe)$ 0.0188 (CpeU-MluU) 3.0 0.0039 (PblT-ApiT)g 14.7 Chaitophorini-Pemphigini 0.0692 (Cvi-Ppi)^ 0.0093 (CpeU-PbeU) 7.4 0.0031 (PblT-PspT) 22.3 Chaitophorini-Aphidini 0.0555 (Cvi-Rma)$ 0.0031 (PblT-AcrT)§ 17.9 Aphidini-Macrosiphini 0.0180 (Rma-Mpe)t 0.0039 (AcrT-ApiT)§ 4.6 Acyrthosiphon-Uroleucon 0.0200 (Api-Uso) 0.0062 (ApiR-UcaR) 3.2 0.0031 (ApiT-UamT)g 6.5 0.0101 (ApiU-UasU)§ 2.0 •Host species abbreviations not described in Table1 or Fig. 3 are as follows: Sgr = Schi2aphis graminum, Cvi = Chaitophorus vimimlis,Rma = Rhopalosiphum maidis, Uso = Uroleucon sonchi, Uam = Uroleucon ambrosiae from Arizona, Uca = Uroleucon caligatum. +The secondary symbiont types (i.e. R, T, or U) for which divergence values are presented are listed next to their host abbreviation codes. ^The smallest distance between B. aphidicola infecting members of these taxa is presented here. §The largest distance between secondary symbionts infecting members of these taxa is presented here. new aphid associates, it is likely that a substantial number symbionts having undergone an especially recent radiation of the microbial guests residing in aphids have yet to be onto their hosts. Thus, interspecies movement is probably described. an ongoing process for these bacteria. Research on other members of the Stemorrhyncha has Evidence similar to that presented here has been used to revealed similar patterns of symbiont diversity. For exam argue that a number of facultative symbionts, including ple, the psyllid Diaphorina citri, was found to associate with Wolbachia pipientis, Sodalis glossinidius, and several psyllid five distinct bacteria, which were occasionally all found to associates, have been independently acquired by different reside within the same host individuals (Subandiyah et al. host lineages (O'Neill et al. 1992; Werren et al. 1995; Aksoy 2000). In addition, at least four distinct bacterial symbionts etal. 1997; Thao et al. 2000a). Despite this evidence, the have been found in whiteflies (Qark et al. 1992;Costa et al. means by which facultative symbionts move between 1995; Jeyaprakash & Hoy 2000; Spaulding & von Dohlen arthropod species remain obscure. Some researchers have 2001), and at least five infect scale insects (Fukatsu & Nikoh proposed a role of transmission through host plants 2000; Thao et al. 2002). Thus, these related plant sapfeeders (Darby etal. 2001). Given the similarity of symbionts in are common habitats for a number of independently derived aphids and ladybird beetles (Werren et al. 1994; Chen et al. symbionts. 1996; Majerus et al. 1999; Fukatsu et al. 2001) it is tempting to propose the involvement of predator-prey interactions. Several investigations have found similar W. pipientis in Symbiont transmission hosts and their parasites, suggesting possible intersj:>ecific Within laboratory colonies, aphid secondary symbionts transmission through host-parasite interactions (Vavre undergo highly efficient vertical transmission with no etal. 1999; Cordaux etal. 2001; Noda etal. 2001). Accord observed instances of horizontal transfer (Chen & Purcell ingly, experimental studies have documented the transfer 1997; Sandstrom et al. 2001). These observations suggest of W. pipientis between a host and its parasitoid and that vertical transfer plays a large role in the persistence of between conspecific parasitoids infecting the same host symbionts within populations. Despite this importance, (Heath et al. 1999; Huigens et al. 2000). However, there is our examination of a wide range of host sp>ecies provides still no evidence of a means by which nonparasitoids can no definitive evidence for co-speciation, indicating that acquire symbionts. We note that even rare transfer events horizontal transfer is, in large part, responsible for R-, may be sufficient to explain the observed patterns. T- and U-type distributions across species. Given our In contrast to patterns suggesting interspecific hori observations of low 16S rDNA divergences between zontal transfer of secondary symbionts, several of our symbionts in different hosts, it appears that a number of observations are suggestive of symbionts persisting within horizontal transfer events occurred recently, with T-type a single host species. These cases do not necessarily imply © 2003 Blackwell Publishing Ltd, Molecular Ecolog}/, 12,1061-1075 52 1072 J. A. RUSSELL ET AL. Strict vertical transmission; they could, rather, reflect hori within their current hosts, lowering their potential for zontal transfer between only conspedfics.Overall, the phylo- interspecific horizontal transfer. To address questions genetic and genetic similarities of symbionts in a single regarding the potential host ranges of secondary sym species would not be expected if rampant interspecific bionts we propose an exiDerimental approach involving horizontal transfer was common. We therefore take these microinjection of symbionts into novel host species. observations to indicate the existence of impediments to such lateral movement. Such barriers could include limited History of associations opportunities for interspecific transfer, limited host ranges influenced by survival in, and transmission to offspring, a The overall lack of phylogenetic congruency and the new host, or frequent extinction of newly infected lineages disparity in the divergences of secondary symbionts as a result of maladaptive phenotypic effects caused by and obligate symbionts, as shown here, suggest that symbionts. Thus, despite the appareat vecency of some associations between specific hosts and their current horizontal transfer events apparently being recent, the per secondary symbionts are quite young. While high genetic sistence of symbionts within host species suggests that similarity between R-type symbionts and free-living intraspecific transfer is more common than interspecific Serratia (i.e. 2.3% divergence between the Aphis craccivora movement, enhancing opportunities for one-to-one co- R-type and Serratia entomophila over 1135 nucleotides), evolution between aphids and secondary symbionts. suggests a recent origin of symbiosis, the story for the T- and U-types is more complicated. Given that T-type 16S rRNA substitution rates are similar to those of Buchnera Symbiont distributions aphidicola (Table 3), applying the rate calibrated for B. Based on our observations of R-, T-, and U-type symbiont aphidicola (i.e. 2-4% substitution per hundred million years, distributions across multiple aphid tribes and beyond, we Moran etal. 1993) suggests that those described in this can rule out the possibility that theseclades are specialized study shared a common ancestor no more than 20-40 on hosts within the tribe Macrosiphini, or that T-type million years ago (based on the maximum divergence, of symbionts are confined to the Stemorrhyncha. However, 1.6%, between the T-types of Bemisia tabaci and Chilomenes the rarity of these bacteria within the Psylloidea (relative to sexmaculatus). However, the monophyly of the T-, U- and the Aphidoidea) suggests that the R-, T- and U-type V-types (Fig. 2) suggests that the age of symbiosis in this symbionts may be a more significant component of aphid clade of microbes is ancient — average sequence diver biology. gence between T- and U-types was around 8%, revealing Several other facultative symbionts of arthropods have a minimum age of 100 million years. Also, the elevated been found to infect even more phylogenetically diverse substitution rates of T- and U-type bacteria, vs. those of ranges of host species. For example, closely related W. pip- free-living bacteria, are consistent with a long history ietitis symbionts have been found in species spanning the of symbiosis. To determine more conclusively whether Arthropoda (Jeyaprakash & Hoy 2000; Werren & Windsor the ancestor of these bacteria was a symbiont, extensive 2000) and in nematodes (Sironi et al. 1995). Closely related screening of nonsymbiotic environments for bacteria in symbionts (over 98% 165 rDNA identity) of the genus this clade would be necessary. It would also be useful Spiroplasma have been found in hosts of the insect orders to determine whether the similar lifestyles of T-, U- and, Coleoptera (Majerus et al.1999), Lepidoptera (Jiggins et al. possibly, V-type symbionts rely on homologoios pathways 2000), Hemiptera (Fukatsu etal. 2001), as well as in ticks and whether the genes for these pathways were present in (Weisburg et al. 1989). Also, Rickettsia with over 99% 16S their common ancestor. rDNA similarity infect beetles (Werren et al. 1994), aphids (Chen et al.1996), and ticks (Philip et al.1983). These obser Importance of facultative symbiosis vations indicate that transfer among distant host taxa is a common feature of facultative symbioses between bacteria Only a few attempts have been made to estimate pheno and arthropods and suggest that some symbiont lineages typic effects associated with infection by secondary sym retain a generalized ability to live in distant host species. bionts, and each of these has focused on symbionts of the Alternatively, gene acquisition by symbionts may permit pea aphid, A. pisum. Two bacteria, the R-type and sym the colonization of novel hosts; such acquisition could bionts of the genus Spiroplasma, have been shown to be mediated by bacteriophage that are known to associ depress host fitness slightly under certain environmental ate with secondary symbionts of aphids (van der Wilk conditions (Chen et al. 2000; Fukatsu et al. 2001). Yet, the etal. 1999; Sandstrom etal. 2001). In contrast to these R-type symbiont confers benefits at high temperatures observations, high genetic similarity of symbionts within (Chen et al. 2000; Montllor et al. 2002), possibly expanding Acyrthosiphon pisum and within Macrosiphum euphorbiae the range of niches available to infected A. pisum. suggests that some symbionts are adapted to a lifestyle Furthermore, a recent study suggests that single isolates of © 2003 Blackwell Publishing Ltd, Molecular Ecology, 12,1061-1075 53 DISTRIBUTIONS OF APHID-ASSOCIATED SYMBIONTS 1073 R- and T-type sjonbionts confer resistance to parasitism phon pisum, and the blue alfalfa aphid, A. kortdoi. Entomologia by the hymenopteran parasitoid, Aphidius ervi (Oliver et al. Experimentalis et Applicata, 95,315-323. in press). Clark MA, Baumann L, Munson MA et al. (1992) The Eubacterial endosymbionts of whiteflies (Homoptera, Aleyrodoidea) con Several studies have revealed a high frequency of sec stitute a lineage distinct from the endosymbionts of aphids and ondary symbionts within aphid populations. For example, mealybugs. Current Microbiology, 25,119-123. the T-type was found in 40/40 Uroleucon ambrosiae collected Clark MA, Moran NA, Baumann P, Wernegreen JJ (2000) Co- from various regions throughout the USA (Sandstrom et al. speciation between bacterial endosymbionts (Buchnera) and a 2001), while another y-Proteobacterial symbiont, known as recent radiation of aphids (Uroleucon) and pitfalls of testing for YSMS, was found to infect 100% of surveyed individuals phylogenetic congruence. Evolution, 54,517-525. belonging to two species within the aphid genus Yamato- Cordaux R, Michel-Salzat A, Bouchon D (2001) Wolbachia infection in crustaceans: novel hosts and potential routes for horizontal callis (Fukatsu 2001). In addition, a survey of California transmission. Journal of Evolutionary Biology, 14, 237-243. A. pisum populations described the R-type in 50/57 indi Costa HS, Westcot DM, Ullman DE et al. (1995) Morphological viduals (Chen & Purcell 1997), while the U-type symbiont variation in Bemisia endosymbionts. Protoplasma, 189,194-202. was found in 219/858 individuals surveyed across Japan Dale C, Maudlin I (1999) Sodalis gen. nov. and Sodalis glossinidius (Tsuchida et al. 2002). Given the observations on their pre sp. nov.,a microaerophilic secondary endosymbiont of the tsetse fly, valence within, and incidence across, species along with Glossina morsitans morsitans. International journal of Systematic their significant effects on host fitness in a variety of eco Bacteriology, 49,267-275. Darby AC, Birkle LM, Turner SL, Douglas AE (2001) An aphid- logical and environmental contexts, we conclude that bome bacterium allied to the secondary symbionts of vi.'hitefly. facultative symbionts have probably played a significant EEMS Microbiology Ecology, 36,43-50. role in the ecology and evolution of aphids. Douglas AE, Prosser WA (1992) Synthesis of essential amino-add tryptophan in the pea aphid (Acyrthosiphon pisum) symbiosis. Journal of Insect Physiology, 38,565-568. Acknowledgements Fukatsu T(2001) Secondary intracellular symbioticbacteria in aphids of the genus Yamatocailis (Homoptera: aphididae Drepanosiphinae). We thank J. M. Michelena, for insect collecting and identification Applied and Environmental Microbiology, 67, 5315-5320. and Paul Baumann for providing psyllid DNA samples. Also, we Fukatsu T, Ishikawa H (1998) Differential immunohistochemical thank Olga Cuesta, Kim Hammond, Ben Wang and Helen Dunbar visualization of the primary and secondary intracellular for technical assistance. This research was funded by National Sci sjonbiotic bacteria of aphids. Applied Entomology and Zoology, 33, ence Foundation Grant DEB-9978518 to N. Moran, grant GVOl -177 321-326. from Ofidna de Ciencia y Tecnologia del Govern Valencia to A. Fukatsu T, Nikoh N (2000) Endosymbiotic microbiota of the Latorre, and grant BFM2000-1313 from Ministerio de Ciencia y bamboo pseudococdd Antonina crawii Onsecta, Homoptera). Teconologia (MCYT) to A. Moya. Applied and Environmental Microbiology, 66, 643-650. Fukatsu T, Aoki S, Kurosu U, Ishikawa H (1994) Phylogeny of References Cerataphidini aphids revealed by their symbiotic micro organisms and basic structure of their galls ~ implications for Adachi J, Hasegawa M (1996) MOLPHY, Programs for Molecular host-symbiont coevolution and evolution of sterile soldier castes. Phylogenetics Based on Maximum Likelihood, Version 2.3. The Zoological Science, 11, 613-623. Institute of Statistical Mathematics, Tokyo. Fukatsu T, Nikoh N, Kawai R, Koga R (2000) The secondary endo Aksoy S, Chen X, Hyf)sa V (1997) Phylogeny and potential trans symbiotic bacterium of the pea aphid Acyrthosiphon pisum mission routes of midgut-assodated endosymbionts of tsetse (Insecta: Homoptera). Applied and Environmental Microbiology, (Diptera: Glossinidae). Insect Molecular Biology, 6,183-190. 66,2748-2758. Baumann P, Baumann L, Lai CY et al. (1995) Genetics, physiology, Fukatsu T, Tsutomu T, Nikoh N, Koga R (2001) Spiroplasma and evolutionary relationships of the genus Buchnera — intra symbiont of the pea aphid Acyrthosiphon pisum (Insecta: Hom cellular symbionts of aphids. Annual Review of Microbiology, 49, optera). Applied and Environmental Microbiology, 67,1284-1291. 55-94. Funk DJ, Helbling L, Wernegreen JJ, Moran NA (2000) Intra- Bender W, Spierer P, Hogness DS (1983) Chromosomal walking spedfic phylogenetic congruence among multiple symbiont and jumping to isolate the Ace and rosy loci and the bithorax genomes. Proceedings of the Royal Society of London Series B- complex in Drosophila melanogaster. Journal of Molecular Biology, Biological Sciences, 267, 2517-2521. 168,17-33. Ghema RL, Werren JH, Weisburg W (1991) Arsenophonus tmsoniae Buchner P (1965) Endosymbionts of Animals with Plant Microorganisms. gen. nov., sp. nov., the causative agent of the son-killer trait in the Intersdence, New York. parasitic wasp Nasonia vitripennis. International Journal of System Chen DQ, Purcell AH (1997) Occurrence and transmission of atic Bacteriology, 41,563-565. facultative endosymbionts in aphids. Current Microbiology, 34, Heath BD, Butcher RDJ, Whitfield WGF, Hubbard SF (1999) 220-225. Horizontal transfer of Wolbachia between phylogenetically dis Chen DQ, Campbell BC, Purcell AH (1996) A new Rickettsia from tant insect species by a naturally occurring mechanism. Current a herbivorous insect, the pea aphid Acyrthosiphon pisum Biology, 9, 313-316. (Harris). Current Microbiology, 33,123-128. Heddi A, Grenier AM, Khatchadourian C, Charles H, Nardon P Chen DQ, Montllor CB, Purcell AH (2000) Fitness effects of two (1999) Four intracellular genomes direct weevil biology: facultative endosymbiotic bacteria on the pea aphid, Acyrthosi Nuclear, mitochondrial, prindpal endosymbiont, and Wolbachia. © 2003 Blackwell Publishing Ltd, Molecular Ecology,12,1061-1075 1074 J. A. RUSSELL ET AL. Proceedings of the National Academy of Sciences of the United States O'Neill SL, Giordano R, Colbert AME, Karr TL, Robertson HM of America, 96,6814-6819. (1992) 16S rRNA phylogenetic analysis of the bacterial endo Huigens ME, Luck RF, Klaassen RHG et al. (2000) Infectious symbionts associated with cytoplasmic incompatibility in parthenogenesis. Nature, 405,178-179. insects. Proceedings of the National Academy of Sciences of the Hypsa V, Dale C (1997) In vitro culture and phylogenetic analysis United States of America, 89, 2699-2702. of 'Candidatus Arsenophonus triatominarum', an intracellular Philip RN, Casper EA, Anacker RL et al. (1983) 'Rickettsia bellii' sp. bacterium from the triatomine bug, Triatoma infestans. Inter nov. a tick borne rickettsia, widely distributed in the United national Journal of Systematic Bacteriology, 47,1140-1144. States, that is distinct from the spotted fever and typhus Jeyaprakash A, Hoy MA (2000) Long PGR improves Wolbachia biogroups. International Journal of Systematic Bacteriology, 33, DNA amplification: wsp sequences found in 76% of sixty-three 94-106. arthropod species. Insect Molecular Biology, 9,393-405. Remaudiere G, Remaudiere M (1997) Catalogue des Aphididae du Jiggins FM, Hurst GDD, Jiggins (3D, von der Schulenburg JHG, Monde. Homoptera Aphidoidea. INRA, Paris. Majerus MEN (2000) The butterfly Danaus chrysippus is infected Robinson M, Gouy M, Gautier C, Mouchiroud D (1998) Sensitivity by a male-killing Spiroplasma bacterium. Parasitology,120,439-446. of the relative-rate test to taxonomicsampling. Molecular Biology Kishino H, Hasegawa M (1989) Evaluation of the maximum like and Evolution, 15,1091-1098. lihood estimate of the evolutionary tree topologies from DNA Robinson-Rechavi M, Huchon D (2000) RRTree: Relative-Rate sequence data, and the branching order in Hominoidea. Journal Tests between groups of sequences on a phylogenetic tree. Bio- of Molecular Evolution, 29,170-179. informatics, 16,296-297. Latorre A, Moya A, Ayala FJ (1986) Mitochondrial DNA poly Rouhbakhsh D, Lai CY, von Dohlen CD et al. (1996) The tryp morphism of Drosophila subobscura. Proceedings of the National tophan biosynthetic pathway of aphid endosymbionts (Buchn- Academy of Sciences of the United States of America, 83, 8649-8653. era): genetics and evolution of plasmid-associated anthranilate Maddison DR, Maddison WP (2000) Macclade: Analysis of Phytogeny synthase (trpEG) within the Aphididae. Journal of Molecular Evo and Character Evolution. Sinauer Associates, Sunderland, MA. lution, 42,414-421. Majerus TMO, von der Schulenburg JHG, Majerus MEN, Hurst Sandstrom J, Moran N (1999) How nutritionally imbalanced is GDD (1999) Molecular identification of a male-killing agent in phloem sap for aphids? Entomologia Experimental is et Applicata, the ladybird Harmonia axyridis (Pallas) (Coleoptera: Cocdnellidae). 91,203-210. Insect Molecular Biology, 8,551-555. Sandstrom J, Russell JA, White JP, Moran NA (2001) Independent Marchuk D, Drumm M, Saulino A, Collins FS (1992) Construction origins and horizontal transfer of bacterial symbionts of aphids. of T-vectors, a rapid general system for direct cloning of Molecular Ecology, 10,217-228. unmodified PGR products. Nucleic Acids Research, 19,1154. Sironi M, Bandi C, Sachhi L, DiSacco B, Damiani G, Genchi C Martinez-Torres D, Buades C,Latorre A, Moya A (2001) Molecular (1995) Molecular evidence for a close relative of the arthropod systematics of aphids and their primary endosymbionts. Mole endosymbiont Wolbachia in a filarial worm. Molecular and Bio cular Phylogenetics and Evolution, 20,437-449. chemical Parasitology, 74,223-227. Montllor CB, Maxmen A, Purcell AH (2002) Facultative bacterial Spaulding AW, von Dohlen CD (2001) Psyllid endosymbionts endosymbionts benefit pea aphids Acyrthosiphon pisum under exhibit patterns of cospedation with hosts and destabilizing heat stress. Ecological Entomology, 27,189-195. substitutions in ribosomal RNA. Insect Molecular Biology, 10,57- Moran NA, Kaplan ME,Gelsey MJ, Murphy TG,Scholes EA (1999) 67. Phylogenetics and evolution of theaphid genus Uroleucon based Subandiyah S, Nikoh N, Tsuyumu S, Somowiyarjo S, Fukatsu T on mitochondrial and nuclear DNA sequences. Systematic Ento (2000) Complex endosymbiotic microbiota of the citrus psyllid mology, 24,85-93. Diaphorina citri (Homoptera: Psylloidea). Zoological Science, 17, Moran NA, Munson MA, Baumann P, Ishikav^^a H (1993) A mole 983-989. cular clock in endosymbiont bacteria is calibrated using the Swofford DL (2001) PAUP*. Phylogenetic Analysis Using Parsimony insect host. Proceedings of the Royal Society of London Series B- (*And Other Methods). Sinauer Associates, Sunderland, MA. Biological Sciences, 253,167-171. Thao ML, Clark MA, Baumann L et al. (2000a) Secondary endo Munson MA, Batmrann P,Clark MA etal. (1991) Aphid-eubacterial symbionts of psyllids have been acquired multiple times. endosymbiosis: evidence for its establishment in an ancestor of Current Microbiology, 41,300-304. four aphid families. Journal of Bacteriology, 173, 6321-6324. Thao ML, Moran NA, Abbot P et al. (2000b) Cospeciation of Noda H, Miyoshi T, Zhang Q et al. (2001) Wolbachia infection psyllids and their primary prokaryotic endosymbionts. Applied shared among planthoppers (Homoptera: Delphacddae) and and Environmental Microbiology, 66,2898-2905. their endoparasite (Strepsiptera: Elenchidae): a probable case of Thao ML,Gullan PJ, Baumann P (2002) Secondary (y-Proteobacteria) interspecies transmission. Molecular Ecology, 10,2101-2106. endosymbionts infect the primary (^-Proteobacteria) endo Ochman H, Wilson AC (1987) Evolution in bacteria — evidence symbionts of mealybugs multiple times and coevolve with for a universal substitution rate in cellular genomes./ournfl/ their hosts. Applied and Environmental Microbiology, 68, 3190- of Molecular Evolution, 26, 74-86. 3197. Ochman H, Elwyn S, Moran NA (1999) Calibrating bacterial evo Tsuchida T, Koga R, Shibao H, Matsumoto T, Fukatsu T (2002) lution. Proceedings of the National Academy of Sciences of the United Diversity and geographic distribution of secondary endosymbiotic States of America, 96,12638-12643. bacteria in natural populations of the pea aphid, Acyrthosiphon Oliver KM, Russell JA, Moran NA, Hunter MS (2003) Facultative pisum. Molecular Ecology, 11, 2123-2135. bacterial symbionts in aphids confer resistance to parasitic UntermanBM, BaumannP, McLean DL (1989) Pea aphid symbiont wasps. Proceedings of the National Academy of Sciences of the United relationships established by analysis of 16S ribosomal-RNAs. States of America, 100,1803-1807. Journal of Bacteriology, 171,2970-2974. © 2003 Blackwell Publishing Ltd, Molecular Ecology, 12,1061-1075 55 DISTRIBUTIONS OF APHID-ASSOCIATED SYMBIONTS 1075 van der Wilk F, Dullemans AM, Verbeek M, van den Heuvel JF Werren JH, Hurst GDD, Zhang W, Breeuwer JAJ, Stouthamer R, (1999) Isolation and characterizationofAPSE-1, a bacteriophage Majerus MEN (1994) Rickettsial relative associated with male infecting the secondary endosymbiont of Acyrthosiphon pisum. killing in the ladybird beetle (Adalia bipunctata). Journal of Bac Virology, 262,104-113. teriology, 176,388-394. van Ham RC, Moya A, Latorre A (1997) Putative evolutionary Werren JH, Zhang W, Guo LR (1995) Evolution and phylogeny origin of plasmids carrying the genes involved in leucine of Wolbachia: reproductive parasites of arthropods. Proceedings biosjmthesis in Buchnera aphidicola (endosymbiont of aphids). of the Royal Society of London Series B-Biological Sciences, 261, Journal of Bacteriology, 179,4768-4777. 55-71. Vavre F, Fleury F, Lepetit D, Fouillet P, Bouletreau M (1999) Phylogenetic evidence for horizontal transmission of Wolbachia in host-parasitoid associations. Molecular Biology and Ewlution, Jacob Russell is a PhD candidatein the department of Ecology and 16,1711-1723. Evolution at the University of Arizona in Tucson. His research von Dohlen CD, Moran NA (2(X)0) Molecular data support a rapid focuses on the interactions between aphids and their facultative radiation of aphids in the Cretaceous and multiple origins of bacterial s5nmbionts. Nancy Moran is a Regents' Professor in the host alternation. Biological Journal of the Linnean Society, 71, 689- same department and studies the biology and evolution of 717. bacterial symbionts and their insect hosts. Professors Amparo Weisburg WG, TuUy JG, Rose DL et al. (1989) A phylogenetic ana Latorre and Andrfe Moya, and PhD candidate Beatriz Sabater- lysis of the mycoplasmas: basis for their classification. Journal of Munoz, belong to the department of Evolutionary Genetics at the Bacteriology, 171, 6455-6467. University of Valencia and study the molecular basis of symbiosis Werren JH, Windsor DM (2000) Wolbachia infection frequencies in in insects as well as the co-evolutionary interactions between insects: evidenceof a global equilibrium? Proceedings of the Royal aphids and their symbionts. Society of London Series B-Biological Sciences, 267,1277-1285. © 2003 Blackwell Publishing Ltd, Molecular Ecology, 12,1061-1075 56 APPENDIX B. NUCLEOTIDE CONTENT AND RATES OF EVOLUTION OF PROTEIN CODING GENES FROM R-TYPE SECONDARY SYMBIONTS OF THE PEA APHID 57 Abstract Symbiosis has had a profound impact on the ecology and evolution of many organisms throughout the tree of life. In several instances, these consequences can be measured through analyses of the genomes of bacterial and fungal symbionts that inhabit larger, eukaryotic hosts. The genomes of microbes that depend upon their hosts for their survival and reproduction often show evidence for degradation: they have lost thousands of genes, they display strong mutational biases towards A and T, and their sequences evolve at higher rates due to reduction in the efficacy of natural selection. Here we sequence several protein-coding genes from the R-type symbiont of aphids, a recently evolved host-dependent microbe from the y-Proteobacteria. We measure nucleotide content and the rates of non-synonymous substitutions in the R-type lineage, comparing these to rates from its close relative, Serratia marcescens. Our results suggest that this young symbiotic lineage has evolved an elevated AT content since its split from Serratia, though R-type genes are more GC biased than other free-living bacteria. In addition, the R-type has not undergone an increase in the rate of non-synonymous substitutions making it one of the most slowly-evolving host-dependent microbes described to date. Introduction Bacteria have evolved to engage in symbioses with eukaryotes on numerous occasions throughout their evolutionary histories. Many bacterial symbionts from the Chlamydia, Mollicutes, Bacteroidetes (Flavobacteria), and Proteobacteria are maternally transmitted in their eukaryotic hosts and are completely dependent upon hosts for their 58 survival and transmission (e.g. Williamson and Whitcomb 1979, O'Neill et al. 1992, Gherna etal. 1991, Unterman^/a/. 1989, Munson e/a/. 1991, Bandi e/a/. 1994, Thaoe/ al. 2003). The consequences of these symbiotic interactions have been studied in detail among the insects. Maternally inherited microbes play substantial roles in insect ecology and evolution, influencing their nutrition, reproduction, sex ratios, host plant utilization, and defense against natural enemies (e.g. Buchner 1965, Douglas and Prosser 1992, Rousset et al. 1992, Tsuchida et al. 2004, Oliver et al. 2003, Ferrari et al. 2004). Symbiosis has also had substantial impacts on the bacterial participants (Moran and Mira 2001, Shigenobu et al. 2000). This has been revealed by comparative studies on the genes and genomes of symbiotic bacteria and their free-living relatives which retain an ability to survive and reproduce in the environment. Like a number of chronic, host-dependent bacterial and fungal pathogens of vertebrates (Cole et al. 2001, Katinka et al. 2001, Fraser et al. 1995, Andersson et al. 1998, Andersson and Kurland 1998), several maternally transmitted bacteria of insects have undergone a process of genome degradation. They have lost thousands of genes (e.g. Ochman and Moran 2001, ) and their genomes are among the smallest described among cellular organisms, reaching sizes as small as 500-600 Kb (Clark et al. 2001, Shigenobu et al. 2000, Gil et al. 2003, Tamas et al. 2002, van Ham et al. 2003, Akman et al. 2002). Many display a mutational bias toward A's and T's (Ishikawa 1987, Baumann et al. 1995, Fukatsu and Nikoh 1998, Spaulding and von Dohlen 1998) and have undergone elevated substitution rates at sites subject to negative or purifying selection— 59 including the ribosomal RNA genes and non-synonymous sites of protein-coding sequences (e.g. Moran et al. 1993, Moran 1996, Thao et al. 2000). Evidence suggests that the patterns of gene loss and accelerated sequence evolution have arisen due to genetic drift, which has been promoted by several components of the symbiotic lifestyle. First, it is believed that many host-dependent bacteria can reliably obtain nutrients from their hosts (Maniloff 1996). This has relaxed selection on certain classes of genes, allowing deletional mutation biases and drift to gradually purge them from the genome (Mira et al. 2001). Second, symbiont population structures have been hypothesized to increase the role of genetic drift relative to natural selection (Funk et al. 2001). Symbionts typically undergo a bottleneck upon maternal transmission whereby small numbers of bacteria are passed onto the progeny of their female hosts (Buchner 1965, Mira and Moran 2002). The resulting reduction in effective population size leads to less effective purifying selection against slightly deleterious mutations (Ohta 1992), resulting in accelerated evolution and, occasionally, gene loss or inactivation. Finally, several maternally inherited microbes—including Buchnera of aphids, Sodalis of tsetse flies, and SOPE and SZPE of grain weevils—have lost genes involved in recombination and are believed to evolve as asexual lineages (Moran and Wernegreen 2000 for review. Dale et al. 2003). Small, asexual populations are subjected to the action of Muller's ratchet, in which mutation and drift lead to the irreversible accumulation of deleterious mutations (Muller 1964, Moran 1996). A variety of heritable bacteria engage in symbioses with phloem-feeding aphids, which cause severe damage to many economically important crops (e.g. Blackman and 60 Eastop 2000). Several have evolved from free-living ancestry within the y- Proteobacteria. The best studied among these symbionts is Buchnera aphidicola (Munson et al. 1991), an ancient symbiont that is essential for aphid growth and reproduction (Prosser and Douglas 1991). Buchnera and aphids have coevolved for over 100 million years (Moran et al. 1993; von Dohlen and Moran 2000), and since this microbe belongs to a larger clade containing other endosymbionts of insects (Lerat et al. 2003), we can conclude that its symbiotic lifestyle is considerably older. In addition to Buchnera, many aphids harbor additional bacteria, known as secondary symbionts, which are not thought necessary for host growth and reproduction (e.g. Unterman et al. 1989, Chen and Purcell 1996, Fukatsu 2000 a&b, Sandstrom et al. 2001, Darby et al. 2001). The youngest of these clades has been dubbed the R-type. The R-type is related to free- living enteric bacteria, including Serratia species (Sandstrom et al. 2001), from which their 16SrRNA genes differ by 2-3% (Russell et al 2003). This distance contrasts with those observed for most other lineages of heritable symbionts of insects, which often exceed 10% 16S rDNA divergence when compared to free-living microbes (e.g. Russell et al. 2003). Using a 2-4% rate of 16S rDNA divergence per 100 million years (Ochman and Wilson 1987, Russell et al. 2003), the maximum age of the symbiotic lifestyle in the R-type clade ranges from 58 to 115 million years—quite young, in comparison to other heritable symbionts (e.g. Moran et al. 1993, Russell et al. 2003). Here, we report sequences of several protein-coding genes from the R-type symbiont, focusing on genes involved in the biosynthesis of essential nutrients that Buchnera provide their aphid hosts. We utilize these data to ascertain whether this 61 "young" lineage displays the genomic characteristics of its ancient symbiotic counterparts and, thus, how genetic drift and substitutional biases have shaped the evolution of its genome. Materials and Methods Degenerate PGR R-type symbionts are members of the Enterobacteriaceae, which includes bacteria such as Serratia species, E. coli, and Yersinia pestis (e.g. Sandstrom et al. 2001). The genomes of the latter two microbes have been fully sequenced and annotated, as have those from several more distant members of the y-Proteobacteria such as Vibrio cholerae. Aside from its 16S rDNA sequences, the R-type genome has not been characterized. Thus, to obtain sequences of R-type protein-coding genes, we used sequences from related bacteria to design degenerate PGR primers. We downloaded amino acid sequences of the argA, aroA, cysE, dapA, hisG, ilvA, ilvC, leuA, metE, mtlD, pheA, thrA, thrB, and trpE genes of E. coli, Y. pestis, and V. cholerae from GenBank. We imported these sequences into GODEHOP ('http://bioinformatics.weizmann.ac.il/blocks/codehop.htmn where they were used to generate candidate primers for degenerate PGR. We selected primers with low degeneracy and mismatches with the homologous Buchnera sequences at the 3' end. This latter criterion ensured that we would not amplify Buchnera genes from our DNA samples, which contained a mixture of aphid and symbiont DNA. Degenerate primers designed to amplify the ribB, ribH, and thiC genes of y-Proteobacteria were kindly 62 provided by Gordon Plague. Sequences of successful degenerate primers were; 5'- ATGAATCAGACGCTACTYTC-3' (ribB Fl) and 5' GCTTCA ATKGTCACGGTAAA-3' (ribB R2) ior ribB, 5'-CTGACCGGTATCGTCTC-3' (thiC F3) and 5' GCTTTCGACATGGCGTTATC 3' (thiC R4) for thiC, 5'- GGCCCTGCTGCTGGCNGCNYTNGC-3' (aroA Fl) and 5'- TGGGGTCGTTGATGGTCACNGGNGTRTC-3' (aroAR2) for aroA, 5'- TGCGGCGGATGGACTTYGGNGGNTG-3' (hisG Fl) and 5'- CACCAGGATGGAGGAGGCNCCNARNGC-3' (hisGRl) for hisG, 5'- GCCGGCATCCAGGTGACNTTYGCNGA-3' (mtlD Fl) and 5'- GCTGCAGCTCCTGGGCYTGNGGRTC-3' (mtlD R2) for mtlD, and 5'- GGCCGAGTGCGAGCCNATGYTNGC-3' (cysE Fl) and 5'- GGTTGCTGCCGTTGAAGTGYTGRTCCAT-3' (cysE Rl) for cysE. We extracted DNA from Acyrthosiphon pisum (pea aphid) cultures which varied in their symbiont complements, as previously described (Bender et al 1983, Russell et al. 2003). DNA was amplified using touchdown PGR, in which annealing temperatures were initially high, ensuring that only strict matches would result in amplification. Annealing temperatures were lowered in later cycles to allow for a larger-scale amplification of the targeted sequences. We used PGR cycling conditions and cocktail recipes as described in Santos and Ochman (2004). For each PGR run, we included DNA from secondary symbiont-free aphids and a water sample as negative controls. To ensure that reactions and primers were working properly we used E. coli DNA as a positive control. 63 We loaded the PGR products into a 1% agarose gel, and viewed them under UV light after electrophoresis and ethidium bromide staining. Since reactions often resulted in the amplification of multiple products, we performed gel extractions on bands of the expected sizes which had amplified fi-om DNA samples of R-infected aphids (based on a 1 Kb Plus DNA ladder Qiagen), using the Qiagen gel extraction kit. Products were concentrated in a SpeedVac®, and then used for cloning. DNA sequencing and diagnostic screening All cloning reactions were conducted using the Invitrogen TopoTA® cloning kit with pCR2.1 vector according to the manufacturer's protocol. We assayed the sizes of cloned products, by PGR with the primers M13F and M13R, followed by gel electrophoresis. Inserts of the expected size were submitted for sequencing with M13F and M13R at the GATG Sequencing Facility at the University of Arizona. We assembled and visualized all sequences and chromatographs using SeqMan of the DNAStar® package. Vector sequences were removed, and remaining sequences were compared to bacterial sequences in the GenBank database using tBLASTx. Sequences selected for further analysis had a significant tBLASTx hit to the expected gene from a known R-type relative (with E-values<10"^). We designed diagnostic primers using cloned, candidate R-type sequences for the aroA, cysE, hisG, mtlD, ribB, and thiC genes. In doing so, we first these sequences with those from bacterial relatives such as E. coli, Y. pestis, and V. cholerae. Next, we designed primers that annealed to regions that were distinct and divergent in the putative R-type sequence. After confirming the utility of these primers in a small PGR assay, we 64 performed a larger survey of Acyrthosiphonpisum clones maintained in the Moran lab. Specifically, we screened DNA from aphid lines that varied in their secondary symbiont complements, including four that harbored the R-type and three (for ribB and thiC) to eight (for aroA, cysE, hisG, mtlD) lines that v^ere free of this symbiont. To frirther confirm their R-type origins, we used these same primers to obtain DNA sequences from three R-type isolates, derived from distinct pea aphid clones. Primers used for diagnostic screening and direct sequencing of R-type genes were 5'- GTTGAACGCGCGCTCGATACA-3' (RribB 2F), 5'- AGCGGTCTGGAACTGGCTGGA-3' (RribB IR), 5'- GATCGGCCTGCGCCCAGGCTCG-3' (RthiC 2F), 5'- CCCGGATGCCCCTTGGCCAGA-3' (RthiC IR), 5'- CGCTCGACAGTGAATATATC-3' (RhisG F2), 5'-CATCGCCACGTGGTTCTGCG- 3' (RhisG Rl), 5'-CTGATGTGAATCAGGCCGTG-3' (RmtlD Fl), 5'- CCGCTGCAATTCCGGTGACG-3' (RmtlD R2), 5'-TCCATGCGACATTGCTCAAA- 3' (R cysE F2), 5'-GTATCTCGCTTTCTGGTCGG-3' (R cysE Rl), 5'- TTGCGGAGGGAACGACCCGG-3' (R aroA F2), and 5'- GATGATCGTTATAAGTGCG-3' (R aroA Rl). DNA sequence analyses To describe the evolutionary patterns that characterized the R-type sequences we analyzed their rates of sequence evolution, %GC content, and codon usage. To achieve this, we first downloaded homologous sequences from related y-Proteobacteria, with the following GenBank accession numbers: U32833 {H. influenzae aroA), AY334546 (£. coli aroA), AJ414148 {Y. pestis aroA), U32743 {H. influenzae cysE), AE016768 {E. coli cysE), AJ414141 (K pestis cysE), U32729 (H. influenzae hisG), AE016762 {E. coli hisG), AJ414149 (K pestis hisG), AE004430 (V. cholerae mtlD), U00039 (E. coli mtlD), AJ414160 (Y. pestis mtlD), U32760 (H. influenzae ribB), AE016766 {E. coli ribB), AJ414144 (K pestis ribB), AE004097 (K cholerae thiC), AE016770 {E. coli thiC), AB063522 {Wigglesworthia glossinidia thiC), and AJ414158 (Y. pestis thiC). The ribB, mtlD, hisG, cysE, and aroA genes of Buchnera were all obtained from the genome sequences deposited under accession numbers APOOl 119 (mtlD) or APOOl 118 (all remaining genes). We obtained Serratia marcescens sequences from the assembled genome (These sequence data were produced by the Serratia marcescens Sequencing Group at the Sanger Institute and can be obtained from http://www.sanger.ac.Uk/Proiects/S marcescens/). This genome was not annotated, so we identified homologous sequences through tBLASTx, using R-type genes as query sequences. For each gene, we obtained a highly significant match to a single locus in the assembled S. marcescens genome. We extracted protein-coding sequences for aroA, cysE, hisG, mtlD, ribB, and thiC, finding each one to be in-tact in the regions analyzed and conserved at the amino acid level. Sequences were aligned in Sequence Navigator using the Clustal alignment option. Aligned sequences were imported into MacClade version 4.01, where the alignment was visualized and adjusted manually, using the translated amino acid sequences as a guide. At this stage, we removed sites with missing sequence data and sites to which our degenerate primers aimealed. 66 We performed relative rates tests using the program RRTree (Robinson-Rechavi & Huchon 2000; http://pbil.univ- lyo n1. fiVso ft war e/rrtree.htmn. which executes the methods of Li (1983) and Pamilo and Bianchi (1983). In these analyses, we compared the number of nonsynonymous substitutions (Kg) that have accumulated along the R-type lineage since it diverged from free-living Serratia relatives, to determine whether the R- type has undergone rate acceleration since evolving a symbiotic lifestyle. We used alignments for the six genes, which included the R-type, S. marcescens and E. coli which was used as the outgroup. We also performed relative rates tests on other free-living and symbiotic enteric bacteria, to describe variation in evolutionary rates. To determine whether R-type sequences show evidence for an AT substitutional bias, we calculated the total GC content at 3"^^ codon positions (GC3), using RRTree. We also conducted a statistical test in order to compare codon usage among organisms. Using the cusp application in EMBOSS (http://www.emboss.org') we computed codon usage tables for the five genes found to be intact in the R-type. These tables gave information on the amino acids encoded by our sequences and the numbers of times each of the 61 protein-coding codons were used. For each amino acid that contained at least one G- or C- ending codon and one A- or T- ending codon, we computed the number of codons that ended in either AT or GC. We then performed a logistic regression in JMPIN® version 3.2.6, using the individual amino acids as replicate data points and comparing codon usage among the R-type and its symbiotic and free-living relatives. This analysis specifically determined whether the R-type differed from its relatives in the usage of AT- versus GC-ending codons, with the prediction that under AT substitutional 67 bias, the R-type would utilize more of the former compared to its close relative, S. marcescens. Results Cloning, Sequencing, and PGR Screening Based on the results of tBLASTx analyses, we found that we had successfully cloned aroA, cysE, hisG, mtlD, ribB, and thiC genes (Table 1). These were considered of likely R-type origin, given that their tBLASTx matches corresponded to organisms with 16S rDNA sequences of high similiarity to the R-type (e.g. Sandstrom et al. 2001). We failed to clone the expected homologous sequences for the remainder of the genes targeted with degenerate PGR {argA, dapA, ilvA, ilvC, leuA, metE, pheA, thrA, thrB, trpE, ribH). And in addition, we were unable to successfully clone homologous sequences for any of these genes for two other symbionts of aphids, the T- and U-types (J. Russell, unpublished data). All genes were found to be intact, with the exception of hisG which had undergone a point mutation that yielded a premature stop codon at position 664-666 of the E. coli hisG gene (out of 900 total bp; accession number: AEO16762). This mutation was confirmed through direct sequencing of PGR products and appears to have arisen quite recently, as the majority of the hisG sequence remained conserved at both the nucleotide and amino acid levels. Using diagnostic primers designed from our cloned sequences, we performed diagnostic screening of our aphid stocks, to determine whether these sequences had truly originated from the R-type endosymbiont. Stocks known to be free of the R-type never 68 amplified with these primers, but in contrast we amplified products of the expected size from each of the R-infected A. pisum lines. After sequencing these products from each of three R-infected clones, we found that their sequences were identical to each other, and to the cloned sequences, except for thiC. In this case, the clone sequence differed from the direct sequences at one position due to an apparent PGR error. The GC content at 3'^'' codon positions of R-type genes ranged from 0.515 to 0.716 (Table 1). In comparison, the GG3 of the ancient symbiont Buchnera ranged from 0.139 to 0.184. Though the GC content of the R-type was higher than in the free-living E. coli (0.494 to 0.601) and Y. pestis (0.463 to 0.563), the R-type had substantially lower GC3 compared to its close relative, S. marcescens (0.736 to 0.836). Logistic regression revealed that the R-type used significantly fewer GC-ending codons (more AT-ending codons) in all genes compared to S. marcescens, though typically more compared to the free-living and symbiotic bacteria in this analysis (Table 1). Relative rates tests demonstrated that the R-type genes evolved at speeds comparable to genes of free-living bacteria, but changed more slowly than homologous sequences from ancient symbionts (Table 2). When E. coli was used as the outgroup. Kg values ranged from 0.009 to 0.166 for the R-type and from 0 to 0.160 for S. marcescens. There were no significant differences among R-type and S. marcescens rates, even under an analysis of a concatenated sequence from aroA, cysE, mtlD, ribB, and thiC, which totaled 3528 nucleotides in length. In contrast, several genes from the ancient symbionts Buchnera and Wigglesworthia evolved substantially faster than the R-type lineage (Table 2). 69 Discussion Due primarily to their asexuality and to small population sizes imposed by transmission bottlenecks, genetic drift is expected to play a large role in shaping the evolution of host-dependent symbionts (Moran 1996). Under this scenario, we should observe increased rates of substitution at positions under purifying selection, such as non- synonymous sites within protein-coding genes. Accordingly, accelerated rates of evolution at non-synonymous sites have been observed for numerous protein-coding genes in Buchnera (Brynnel et al. 1998, Moya et al. 2002), including several with roles in the biosynthesis of amino acids (Moran 1996, Wernegreen and Moran 1999, Clark et al. 1999). Moreover, the 16S rRNA genes of many heritable symbionts of insects have been observed to evolve at rates exceeding those of free-living bacteria (Thao et al 2000, Spaulding and von Dohlen 1998, Moran 1996, Degnan et al 2004, Russell et al. 2003). In a previous study, the R-type 16S rRNA gene was shown to evolve at a rate comparable to that seen for free-living bacteria (Russell et al. 2003). Our results reveal that non- synonymous sites of several R-type protein-coding genes evolve at rates comparable to those of their free-living relatives but slower than those from their symbiotic counterparts. Thus, the R-type is unique among the described maternally-transmitted bacteria of insects, representing the only symbiont that has not undergone rate acceleration. The reasons for the lack of elevated substitution rates are unclear, but the result suggests that drift has played a smaller role in protein sequence evolution within this young lineage. One possible explanation is that the effective population sizes of the R- type are larger than those for other symbionts. This would require that a larger number of R-type symbionts be passed on from mother to offspring during maternal transmission. However, R-type symbionts are considerably outnumbered by Buchnera in developing pisum embryos (Fukatsu et al. 2000) and, thus, it appears that they are actually transmitted in lower numbers in this host. Since this microbe has undergone occasional horizontal transmission between aphid species (Sandstrom et al. 2001, Russell et al. 2003) we cannot rule out that symbiont numbers are greater in other host lineages or that the symbionts that have gained common representation among the widest range of aphids are those which have preferentially escaped the perils of genetic drift. Under this latter scenario, horizontal transfer would act as a brake to genome degradation assuming that transfer occurs often enough and that lineages subjected to elevated drift are at a disadvantage. These hypotheses remain to be tested. Like many bacterial symbionts and host-dependent parasites, the R-type showed some evidence for an AT substitutional bias. AT content at codon positions ranged from 28-49% in the R-type and from 16-26% in S. marcescens. So though the R-type is not AT rich, the evidence here suggests that it has recently undergone a higher proportion of A and T substitutions compared to its free-living sister taxon, S. marcescens. We conclude that the higher AT content is a derived trait based on previous findings of high GC content in multiple Serratia species (Falkow et al. 1962) and the fact that the R-type is nested within the Serratia clade (ie. it is not an outgroup) based on phylogenetic analyses (Russell et al. 2003, Moran et al. submitted). Our analyses do not allow us to 71 decisively determine whether this nucleotide bias is due to selection on codon usage or to a true change in mutational patterns. However, we note that the usage of A versus T ending codons for single amino acids did not appear random as would be expected for a mutational bias (data not shovm). This study provides the first glimpse into the coding capacity of the R-type symbiont's genome. We obtained partial, intact sequences for five protein-coding genes from the R-type, which are known to play roles in the biosynthesis of amino acids (aroA, cysE) and vitamins (ribB, thiC), and in the metabolism of sugars (mtlD). The presence of genes involved in amino acid biosynthesis is of interest, for two reasons. First, aphids feed on phloem sap which contains low amounts of essential amino acids (Douglas 1993, Sandstrom and Moran 1999). Second, Buchnera are known to provision aphids with essential amino acids (Douglas 1988, Douglas and Prosser 1992, Munson and Baumann 1993, Bracho et al. 1995). Given that the R-type harbors genes that function in the synthesis of essential amino acids we hypothesize that this bacterium plays a nutritional role in aphid biology. To date, there is some evidence suggesting that this symbiont can supplement the functions performed by Buchnera, as Buchnera-Q:ee A. pisum have been found to reproduce only when they harbor the R-type (Koga et al. 2003). Moreover, the R-type improves performance of infected A. pisum after heat exposure (Montllor et al. 2002), a stress that is known to impair the function and survival of Buchnera (Ohtaka and Ishikawa 1991). An investigation into the role of nutrient-provisioning could help to explain these effects and to identify a source of selective pressure that has favored the 72 retention of nutrient biosynthetic genes frequently lost in host-dependent parasites (see Moran and Wemegreen 2000, for review). In conclusion, our results suggest that molecular evolution in the R-type genome is distinct from that of described bacterial symbionts of insects. The youth of the symbiotic lifestyle in this lineage suggests that the trends of genome degradation emerge in a gradual fashion and that selection may be more effective in the early stages of symbiosis. However, one out of six genes described here showed evidence for recent inactivation, which suggests that the R-type genome may be losing functional genes through mutation and drift. It will be useful to determine the patterns of gene retention and inactivation within this clade, as these will provide testable hypotheses concerning the roles of the symbionts in host biology. Here we have focused on R-type isolates from only single aphid species, Acyrthosiphon pisum. Since the R-type has been found in several tribes and families of aphids, and in hosts known to differ substantially in their ecology, comparative genomics among isolates will provide a great deal of insight into the importance of these symbionts for their aphid hosts and the forces that have shaped the evolution of this symbiosis. 73 References Cited Akman, L., A. Yamashita, H. Watanabe, K. Oshima, T. Shiba, M. Hattori, and S. Aksoy. 2002. Genome sequence of the endocellular obligate symbiont of tsetse flies, Wigglesworthia glossinidia. Nature Genetics 32:402-407. Andersson, S.G.E., and C.G. Kurland. 1998. Reductive evolution of resident genomes. Trends. Microbiol. 6:263-268. Andersson, S.G., A. Zomorodipour, J.O. Andersson, T. Sicheritz-Ponten, U.C. Alsmark, R.M. Podowski, A.K. Naslund, A.S. Eriksson, H.H. Winkler, and C.G. Kurland. 1998. The genome sequence of Rickettsia prowazekii and the origin of mitochondria. Nature 396:133-140. Bandi, C., G. DAmiani, L. Magrassi, A. Grigolo, R. Fani, and L. Sacchi. 1994. Flavobacteria as intracellular symbionts in cockroaches. Proc. R. Soc. Lond. B 257:43-48. Baumann, P., L. Baumann, C.Y. Lai, and D. Rouhbaksh. 1995. Genetics, physiology, £ind evolutionary relationships of the genus Buchnera: intracellular symbionts of aphids. Annu. Rev. Microbiol. 49:55-94. Blackman, R.L., and V.F. Eastop. 2000. Aphids on the world's crops: an identification and information guide. 2"'' edition. John Wiley and Sons, West Sussex, UK. Bracho, A.M., D. Martinez-Torres, A. Moya, and A. Latorre. 1995. Discovery and molecular-characterization of a plasmid localized in Buchnera sp. bacterial endosymbiont of the aphid Rhopalosiphum padi. J. Mol. Evol. 41:67-73. Buchner, P. 1965. Endosymbiosis of animals with plant microorganisms. Interscience, New York, NY. Chen, D.-Q., B.C. Campbell, and A.H. Purcell. 1996. A new rickettsia from a herbivorous insect, the pea aphid Acyrthosiphonpisum (Harris). Curr. Microbiol. 33:123-128. Clark, M.A., L. Baumann, M.L. Thao, N.A. Moran, and P. Baumann. 2001. Degenerative minimalism in the genome of a psyllid endosymbiont. J. Bacteriol. 183:1853-1861. 74 Cole, S.T., K. Eiglmeier, J. Parkhill, K.D. James, N.R. Thomson, P.R. Wheeler, N. Honore, T. Garnier, C. Churcher, D. Harris, K. Mungall, D. Basham, D. Brown, T. Chillingworth, R. Connor, R.M. Davies, K. Devlin, S. Duthoy, T. Feltwell, A. Fraser, N. Hamlin, S. Holroyd, T. Homsby, K. Jagels, C. Lacroix, J. Maclean, S. Moule, L. Murphy, K. Oliver, M.A. Quail, M.A. Rajandream, K.M. Rutherford, S. Rutter, K. Seeger, S. Simon, M. Simmonds, J. Skelton, R. Squares, S. Squares, K. Stevens, K. Taylor, S. Whitehead, J.R Woodward, and B.G. Barrell. 2001. Massive gene decay in the leprosy bacillus. Nature 409:1007-1011. Dale, C., B. Wang, N.A. Moran, and H. Ochman. 2003. Loss of DNA recombinational repair enzymes in the initial stages of genome degradation. Mol. Biol. Evol. 20:1188-1194. Darby, A.C., L.M. Birkle, S.L. Turner, and A.E. Douglas. 2001. An aphid-borne bacterium allied to the secondary symbionts of whitefly. FEMS Microbiol. Ecol. 36:43-50. Degnan, P.H., A.B. Lazarus, C.D. Brock, and J.J. Wemegreen. 2004. Host-symbiont stability and fast evolutionary rates in an ant-bacterium association: cospeciation of Camponotus species and their endosymbionts, Candidatus Blochmannia. Syst. Biol. 53:95-110. Douglas, A.E. 1988. Sulfate utilization in an aphid symbiosis. Insect Biochem. 18:599- 605. Douglas, A.E. 1993. The nutritional quality of phloem sap utilized by natural aphid populations. Ecol. Entomol. 18:31-38. Douglas, A.E. 1998. Nutritional interactions in insect-microbial symbioses: aphids and their symbiotic hactQrio. Buchnera. Ann. Rev. Entomol. 43:17-37. Douglas, A.E., and W.A. Prosser. 1992. Synthesis of the essential amino acid tryptophan in the pea aphid (Acyrthosiphonpisum) symbiosis. J. Insect Physiol. 38:565-568. Falkow, S., I.R. Ryman, and O. Washington. 1962. Deoxyribonucleic acid base composition of Proteus and Providence organisms. J. Bacteriol. 83:1318-1321. Ferrari, J., A.C. Darby, T.J. Daniell, H.C.J. Godfray, A.E. Douglas. 2004. Linking the bacterial community in pea aphids with host-plant use and natural enemy resistance. Ecol. Entomol. 29:60-65. 75 Fraser, C.M., J.D. Gocayne, O. White, M.D. Adams, R.A. Clayton, R.D. Fleischmann, C.J. Bult, A.R. Kerlavage, G. Sutton, J.M. Kelley. 1995. The minimal gene complement of Mycoplasma genitalium. Science 270:397-403. Fukatsu, T. 2001a. Secondary intracellular symbiotic bacteria in aphids of the genus Yamatocallis. (Homoptera:Aphididae:Drepanosiphinae). Appl. Environ. Microbiol. 67:5315-5320. Fukatsu, T. 2001b. Spiroplasma symbiont of the pea aphid, Acyrthosiphon pisum (Insecta: Homoptera). Appl. Environ. Microbiol 67:1284-1291. Fukatsu, T., and N. Nikoh. 1998. Two intracellular symbiotic bacteria from the mulberry psyllid Anomoneura mori (Insecta, Homoptera). Appl. Environ. Microbiol. 64:3599-3606. Fukatsu, T., N. Nikoh, R. Kawai, and R. Koga. 2000. The secondary endosymbiotic bacterium of the pea aphid Acyrthosiphon pisum (Insect: Homoptera). Appl. Environ. Microbiol. 66:2748-2758. Funk, D.J., J.J. Wernegreen, and N.A. Moran. 2001. Intraspecific variation in symbiont genomes; bottlenecks and the aphid-5Mc/z«era association. Genetics 157:477-489. Gheraa, R.L., J.H. Werren, W. Weisburg, R. Cote, C.R. Woese, L. Mandelco, and D.J. Brenner. 1991. Arsenophonus nasoniae gen. no v., sp. no v., the causative agent of the son-killer trait in the parasitic wasp Nasonia vitripennis. Int. J. Syst. Bacteriol. 41:563-565. Gil R, F. J. Silva, E. Zientz, F. Delmotte, F. Gonzalez-Candelas, A. Latorre, C. Rausell, J. Kamerbeek, J. Gadau, B. Holldobler, R.C. van Ham, R. Gross, and A. Moya. 2003. The genome sequence of Blochmanniafloridanus: comparative analysis of reduced genomes. Proc. Natl. Acad. Sci. USA. 100:9388-9393. Ishikawa, H. 1987. Nucleotide composition and kinetic complexity of the genomic DNA of an intracellular symbiont of the pea aphid Acyrthosiphon pisum. J. Mol. Evol. 24:205-211. Katinka, M.D., S. Duprat, E. Comillot, G. Metenier, F. Thomarat, G. Prensier, V. BArbe, E. Peyretaillade, P. Brottier, P. Wincker, F. Delbac, H. El Alaoui, P. Peyret, W. Saurin, M. Guoy, J. Weissenbach, and C.P. Vivares. 2001. Genome sequence and gene compaction of the eukaryote parasite Encephalitozoon cuniculi. Nature 414:401-2. 76 Lerat, E., V. Daubin, and N.A. Moran. 2003. From gene trees to organismal phylogeny in prokaryotes: the case of the gamma-Proteobacteria. PLoS Biol. 1:101-109. Li, W.H. 1993. Unbiased estimation of the rates of synonymous and nonsynonymous substitution. J. Mol. Evol. 36:96-99. Martin, W., L.V. Sun, J. Vamathevan, M. Riegler, R. Deboy, J.C. Brownlie, E.A. McGraw, W. Martin, C. Esser, N. Ahmadinejad, C. Wiegand, R Madupu, M.J. Beanan, L.M. Brinkac, S.C. Daugherty, S. Durkin, J.F. Kolonay, W.C. Nelson, Y. Mobamoud, P. Lee, K. Berry, M.B. Young, T. Utterback, J. Weidman, W.C. Nierman, LT. Paulsen, K.E. Nelson, H. Tettelin, S.L. O'Neill, and J.A. Eisen. 2004. Phylogenomics of the reproductive parasite Wolbachia pipientis wMel; a streamlined genome overrun by mobile genetic elements. PLoS Biol. 2:327-341. Mira, A., H. Ochman, and N.A. Moran. 2001. Deletional bias and the evolution of bacterial genomes. Trends Genet. 17:589-596. Moran, N.A., and J.J. Wemegreen. 1999. Lifestyle evolution in symbiotic bacteria: insights from genomics. Trends. Ecol. Evol. 15:321-326. Munson, M.A., and P. Baumann. 1993. Molecular cloning and nucleotide sequence of a putative trpDC(F)BA operon in Buchnera aphidicola (endosymbiont of the aphids Schizaphis graminum). J. Bacteriol. 175:6426-6432. Maniloff, J. 1996. The minimal cell genome; "on being the right size". Proc. Natl. Acad. Sci. USA 93:10004-10006. Moran, N. 1996. Accelerated evolution and Muller's ratchet in endosymbiotic bacteria. Proc. Natl. Acad. Sci. USA 93:2873-2878. Moran, N.A., M.A. Munson, P. Baumann, and H. Ishikawa. 1993. A molecular clock in endosymbiotic bacteria is calibrated using the insect hosts. Proc. R. Soc. Lond. Ser.B 253:167-171. Muller, H.J. 1964. The relation of recombination to mutational advance. Mutat. Res. 1:2-9. Munson, M.A., P. Baumann, and M.G. Kinsey. 1991. Buchnera gen. nov. and Buchnera aphidicola sp. nov., a taxon consisting of the mycetocyte-associated, primary endosymbionts of aphids. Int. J. Syst. Microbiol. 41:566-568. Ohta, T. 1992. The nearly neutral theory of molecular evolution. Annu. Rev. Ecol. Syst. 23:263-286. 77 Ohtaka, C., and H. Ishikawa. 1991. Effects of heat treatment on the symbiotic system of an aphid mycetocyte. Symbiosis 11:19-30. Oliver, K.M., J.A. Russell, N.A. Moran, and M.S. Hunter. 2003. Facultative bacteria in aphids confer resistance to parasitic wasps. Proc. Natl. Acad. Sci. USA 100:1803-1807. O'Neill, S.L., R. Giordano, A.M.E. Colbert, T.L. Karr, and H.M. Robertson. 1992. 16S rRNA phylogenetic analysis of the bacterial endosymbionts associated with cytoplasmic incompatibility in insects. Proc. Natl. Acad. Sci. USA 89:2699-2702. Robinson-Rechavi, M., and D. Huchon. 2000. RRTree: relative-rate tests between groups of sequences on a phylogenetic tree. Bio informatics 16:296-297. Ochman, H., and N.A. Moran. 2001. Genes lost and genes foimd: Evolution of bacterial pathogenesis and symbiosis. Science 292:1096-1098. Pamilo, P., and N.O. Bianchi. 1993. Evolution of the zfx and zfy genes: rates and interdependence between the genes. Mol. Biol. Evol. 10:271-281. Prosser, W.A., and A.E. Douglas. 1991. The aposymbiotic aphid—an analysis of chlortetracycline-treated pea aphid, Acyrthosiphonpisum. J. Insect Physiol. 37:713- 719. Rousset, F., D. Bouchon, B. Pintureau, P. Juchault, M. Solignac. 1992. Wolbachia endosymbionts responsible for various alterations of sexuality in arthropods. Proc. R. Soc. Lond. Ser. B 250:91-98. Russell, J.A., A. Latorre, B. Sabater-Munoz, A. Moya, and N.A. Moran. 2003. Side stepping secondary symbionts: widespread horizontal transfer across and beyond the Aphidoidea. Mol. Ecol. 12:1061-1075. Sandstrom, J., and N.A. Moran. 1999. How nutritionally imbalanced is phloem sap for aphids? Entomol. Exp. Appl. 91:203-210. Sandstrom, J., and N.A. Moran. 2001. Amino acid budgets in three aphid species using the same host plant. Physiol. Entomol. 26:202-211. Sandstrom, J.P., J.A. Russell, J.P. White, and N.A. Moran. 2001. Independent origins and horizontal transfer of bacterial symbionts of aphids. Mol. EcoL 10:217-228. 78 Santos, S.R., and H. Ochman. 2004. Identification and phylogenetic sorting of bacterial lineages using universally conserved genes and proteins. Environ. Microbiol. 6:754-759. Shigenobu, S., H. Watanabe, M. Hattori, Y. Sasaki, and H. Ishikawa. 2000. Genome sequence of the endocellular bacterial symbiont of aphids Buchnera sp. APS. Nature 407:81-86. Spaulding, A.W., and C. von Dohlen. 1998. Phylogenetic characterization and molecular evolution of bacterial endosymbionts inpsyllids (Hemiptera: Stemorrhyncha). Mol. Biol. Evol. 15:1506-1513. Tamas, I., L. Klasson, B. Canback, A.K. Naslund, A.S. Eriksson, J.J. Wernegreen, J.P. Sandstrom, N.A. Moran, and S.G. Andersson. 2002. 50 million years of genomic stasis in endosymbiotic bacteria. Science 296:2376-2379. Thao, M.L., L. Baumann, J.M. Hess, B.W. Falk, J.C.K. Ng, P.J. GuUan, and P. Baumann. 2003. Phylogenetic evidence for two new insect-associated Chlamydia of the family Simkaniaceae. Curr. Microbiol. 47:46-50. Thao, M.L., M.A. Clark, L. Baumann, E.B. Brennan, N.A. Moran, and P. Bauman. 2000b. Secondary endosymbionts of psyllids have been acquired multiple times. Curr. Microbiol. 41:300-304. Thao, M.L., N.A. Moran, P. Abbot, E.B. Brennan, D.H. Burckhardt, and P. Baumann. 2000a. Cospeciation of psyllids and their primary prokaryotic endosymbionts. Appl. Environ. Microbiol. 66:2898-2905. Tsuchida, T., R. Koga, and T. Fukatsu. 2004. Host plant specialization governed by facultative symbiont. Science 303:1989. Unterman, B.M., P. Baumann, and D.L. McLean. 1989. Pea aphid symbiont relationships established by analysis of 16S ribosomal-RNA's. J. Bacteriol. 171:2970-2974. van Ham, RC., J. Kamerbeek, C. Palacios, C. Rausell, F. Abascal, U. Bastolla, J.M. Frenandez, L. Jimenez, M. Postigo, F.J. Silva, J. Tamames, E. Viguera, A. Latorre, A. Valencia, F. Moran, and A. Moya. 2003. Reductive evolution in Buchnera aphidicola. Proc. Natl. Acad. Sci. USA 100:581-586. von Dohlen, C.D., and N.A. Moran. 2000. Molecular data support a rapid radiation of aphids in the Cretaceous and multiple origins of host alternation. Biol. J. Linn. Soc. 71:689-717. 79 Wernegreen, J. J., and N.A. Moran. 2000. Decay of mutualistic potential in aphid endosymbionts through silencing of biosynthetic loci: Buchnera of Diuraphis. Proc. R. Soc. Lond. B 267:1423-1431. Williamson, D.L. and Whitcomb, R.F. 1979. Sex ratio organisms (spiroplasmas) of Drosophila. In The Mycoplasmas, vol. 3, pp. 175-208. Whitcomb and Tully, eds. Academic Press, New York. 80 Gene (bp Function GC3® sequenced) R-type S. marcescens E. colt Y. pestis'* Buchnera'* aroA Phe, Trp, 0.6250 0.8361' 0.5056' 0.5389' 0.1417' (1089bp) and Tyr biosynthesis cysE Cys 0.5833 0.8056' 0.5478 0.4630 0.1389' (682bp) biosynthesis hisG' His 0.6629 0.8258* 0.5955 0.5281' 0.1573' (534bp) biosynthesis mtlD Carbon 0.5148 0.8164* 0.5410 0.4852 0.1475' (922bp) metabolism ribB Riboflavin 0.6552 0.7356' 0.4943' 0.5172 0.1839' (262bp) biosynthesis thiC Thiamine 0.7158 0.8306' 0.6011* 0.5628' n.a. (550bp) biosynthesis Table 1 Sequence features of protein-coding genes of the R-type and its bacterial relatives. ®GC3: Proportion of 3'^'' codon positions that contain a G or C "We compared codon usage in S. marcescens, Buchnera, E. coli, and Y. pestis to that in the R- type, using logistic regression *GC content of 3'^'' codon positions differs significantly from that in the corresponding R-type gene 'Putative pseudogene 81 Gene(s) Outgroup Taxon 1 Taxon 2 Kal-Ka2+/-sd®^' aroA E. coli R-type S. marcescens 0.006+/-0.010 cysE 0.011 +/- 0.009 mtlD -0.006+/-0.018 ribB -0.005+/- 0.013 thiC 0.009 +/- 0.005 all 0.004 +/- 0.006 aroA H. influenzae Buchnera R-type 0.055 +/- 0.025* E. coli 0.074 +/- 0.023* cysE H influenzae Buchnera R-type 0.147+/- 0.032* E. coli 0.160+/-0.032* mtlD V. cholerae Buchnera R-type 0.173 +/-0.036* E. coli 0.164+/- 0.037* ribB H influenzae Buchnera R-type 0.033 +/- 0.046 E. coli 0.073 +/- 0.041 thiC V cholerae Wiggle sworthia R-type 0.062+/- 0.018* E. coli 0.070 +/-0.018* Table 2 Relative rates tests reveal no evidence for acceleration of non-synonymous substitutions in the R-type. ®Ka "x": non-synonymous substitutions in the lineage leading to taxon x; Kal - Ka2 equals the difference between the number of non-synonymous substitutions in the lineage leading to Taxon 1 and the number occurring within the lineage leading to Taxon 2. sd = standard deviation • differences for which p<0.05 are denoted by * 82 APPENDIX C. HORIZONTAL TRANSMISSION OF SECONDARY BACTERIAL SYMBIONTS; TRANSMISSION AND FITNESS EFFECTS IN A NOVEL APHID HOST, ACYRTHOSIPHON PISUM 83 Abstract Members of several bacterial lineages are ubiquitously engaged in symbiotic relationships with insects, shaping the ecology and evolution of their hexapod hosts. Several of these microbes move between hosts through vertical transmission, a mode of transfer which promotes strong fidelity between the symbiotic partners. However, phylogenetic evidence suggests that these microbial symbionts have occasionally moved horizontally between insect species, revealing that some retain a generalized ability to infect multiple hosts. Here we examine the abilities of three vertically transmitted bacteria from the y-Proteobacteria to successfully infect a novel host, Acyrthosiphon pisum. Using microinjection, we transfer symbionts into a common host background and compare transmission efficiencies between novel and naturally occurring symbionts of A. pisum. We also compare the fitness effects of two novel symbionts to those induced by three natural^, pisum symbionts to determine whether novel symbionts should persist under natural selection. Our results reveal that symbionts vary in their capacities to utilize the pea aphid as a suitable host—one out of three novel symbionts failed to undergo efficient vertical transmission in the A. pisum background. Moreover, one out of two stably transmitted symbionts induced substantial fitness costs. These fmdings reveal that negative fitness effects and low transmission efficiency can act as barriers to horizontal transmission and, thus, play potentially significant roles in shaping the distributions of bacterial symbionts across their insect hosts. 84 Introduction Insects and bacteria have evolved a diverse array of symbiotic interactions, which play roles in nutrition (Bracho et al. 1995, Douglas 1988, Douglas and Prosser 1992, Lai et al. 1994, Moran et al. 2004, Wicker 1983), defense (Ferrari et al. 2004, Keliner and Dettner 1996, Oliver et al. 2003, Piel 2002), reproduction, and development (Caspari and Watson 1959, Ghema et al. 1991, Hurst et al. 1999, O'Neill et al. 1992, Rousset et al. 1992, Stouthamer era/. 1993,Werrener a/. 1994). Many of these bacteria are heritable— they are transferred directly from mother to offspring. This mode of transmission aligns the interests of symbiotic partners since symbiont spread depends upon host reproduction (Ewald 1983, Bull et al. 1991, May and Anderson 1982). As a result, the prevalence of a given symbiont will be determined by its net effects on host fitness and the efficiency of transmission. Both fitness benefits and highly efficient transmission will favor the spread of a heritable symbiont throughout a host population (e.g. Turelli 1994). Many clades of heritable bacteria show broad distributions across multiple families and orders of insect hosts (Bandi et al. 1995, Hypsa and Aksoy 1999, Munson et al. 1991, Sironi et al. 1995, Thao et al. 2000a&b, Werren et al. 1995, Werren and Windsor 2000). The distributions of several of these symbionts are due, in part, to horizontal transmission between species, as inferred from measures of DNA sequence divergence and comparisons of host and symbiont phylogenies (Aksoy et al. 1997, Moran and Baumann 1994, O'Neill et al. 1992, Sandstrom et al. 2001, Thao et al. 2000a). This pattern raises the questions of whether these secondary symbionts are generalists, capable 85 of successfully engaging in symbioses with multiple host species, and whether there are barriers that impede their horizontal transfer. Aphids are an ideal group of organisms for studies on symbiosis. Nearly all of these phloem-feeding insects harbor the bacterium Buchnera aphidicola (Moran and Telang 1998, Munson et al. 1991a&b), an ancient heritable symbiont that is essential for host growth and reproduction (Douglas 1992, Prosser and Douglas 1991). Many aphids harbor additional heritable bacteria known as accessory, or secondary symbionts (Buchner 1965, Chen et al. 1996, Chen and Purcell 1997, Darby et al. 2001, Fukatsu 2001a&b, Sandstrom et al. 2001, Unterman 1989). Secondary symbionts are often present at intermediate frequencies within aphid species (Chen and Purcell 1997, Haynes et al. 2003, Simon et al. 2003, Tsuchida et al. 2002) and are generally thought to be urmecessary for host growth and reproduction (but see Sandstrom et al. 2001, Fukatsu 2001a). Like Buchnera, secondary symbionts are passed from mother to offspring with high efficiency, approaching 100% in the lab (Buchner 1965, Chen and Purcell 1997, Sandstrom et al. 2001, Fukatsu et al. 2000, Darby and Douglas 2003). Recent studies have identified fitness benefits conferred by these microbes, involving resistance to natural enemies (Ferrari et al. 2004, Oliver et al. 2003), heat tolerance (Chen et al. 2000, Montllor et al. 2002), and host plant utilization (Tsuchida et al. 2004, but see Leonardo 2004). Such benefits provide adaptive explanations for symbiont prevalence within aphid populations. However, the intermediate infection frequencies of secondary symbionts suggest the existence of conditional fitness costs or occasional transmission failure in the field. 86 Secondary symbionts of aphids have evolved from phylogenetically diverse bacterial groups. Several, including Arsenophonus and the R-, T-, and U-types, belong to the gamnia-3 subdivision of the Proteobacteria and are related to free-living environmental microbes such as Escherichia coli, Proteus vulgaris. Yersinia pestis and Serratia species (Sandstrom et al. 2001, Russell et al. 2003). Whereas the R- and U- types are currently described from multiple families within the Aphidoidea, the Arsenophonus and T-type symbionts have been found in even more distantly related host taxa, including ticks, wasps, whiteflies, and psyllids (Clark reference, Darby et al. 2001, Grindle et al. 2003, Russell et al. 2003, Sandstrom et al 2001, Zchori-Fein and Brown 2002,). Small genetic distances between bacteria infecting distant host species (i.e. often less than 1% 16S rDNA divergence) and incongruent host and symbiont phylogenies reveal that Arsenophonus, R-, T-, and U-type symbionts have undergone occasional horizontal transmission between species (Russell et al. 2003, Zchori-Fein and Brown 2002). Previous experiments have demonstrated that some isolates of the R- and T-type symbionts of the pea aphid, Acyrthisiphon pisum, can stably infect novel aphid species in the laboratory, suggesting that these symbionts are generalists (Chen and Purcell 1997, Darby and Douglas 2003). However, an isolate of the R-type symbiont imposed a severe fitness cost on Acyrthosiphon kondoi after transfer from its natural host, A. pisum (Chen et al. 2000). Given the magnitude of this cost and the reliance of secondaries on aphid reproduction, the novel R-type would not likely persist within field populations of^. kondoi. 87 To examine the capacity for secondary symbionts to infect novel hosts, we transferred secondary symbionts from three aphid species into a new host, A. pisum. We describe the abilities of Arsenophonus and U-type symbionts to undergo vertical transmission in this novel background and compare their transmission efficiencies with that of a native, or naturally occurring U-type symbiont of A. pisum. In addition, we compare the effects of novQl Arsenophonus and T-type symbionts on^. pisum fitness to fitness effects exerted by native R-, T-, and U-type symbionts. Our results suggest variation in the range of potential hosts for these bacterial symbionts and reveal transmission and phenotypic barriers that likely limit their distributions. Materials and Methods Transmission Efficiency: Symbionts and Aphids A. pisum were reared on synchronously aged pre-flowering fava beans (Vicia faba) grown in a greenhouse at the University of Arizona. All cultures were reared on single plants in 4" diameter pots and enclosed by a clear SOLO® cup containing mesh tops for ventilation. These cultures were placed in Percival Scientific growth chambers which were set for 18 hours of light and a constant temperature of 20°C. Four clones of A. pisum were used in the transmission efficiency experiment- "recipient" clones 5 A and 7a and "donor" clones 2a and 8-10-1. Recipient A. pisum clones were chosen because they did not harbor the pea aphid Rickettsia symbiont (PAR), or R-, T-, or U-type symbionts, as confirmed by diagnostic PGR (Sandstrom et al. 2001); donor clones were naturally infected with heritable U-type symbionts. Each^. pisum 88 clone was started from a single female individual collected from alfalfa in New York in 2000 (7a, 2a) or 2001 (8-10-1) or from fava bean in Wisconsin in 1999 (5A). We occasionally performed diagnostic PGR screens on each of these clones to confirm their secondary symbiont complements (described in Sandstrc3m et al. 2001, Russell et al. 2003), fmding no evidence for transmission failure. A Myzocallis sp. feeding on live oak on the University of Arizona campus was found to harbor a symbiont related to Arsenophonus secondary symbionts of psyllids and whiteflies and to the male-killing Arsenophonus nasoniae of the wasp, Nasonia vitripennis (Russell et al. 2003). This symbiont appeared to be fixed within this host population, as all screened individuals tested positive in a diagnostic PGR assay. Diagnostic PGR revealed a complete absence of this symbiont from over 100 A. pisum individuals collected from NY and WI (data not shown) and a similar absence from Japanese populations of A. pisum (Tsuchida et al. 2002). Similarly, a U-type symbiont was found in all Macrosiphum euphorbiae aphids feeding on Penstemon species in Tucson, AZ. The 16S rDNA sequence of this symbiont (AYl 36149) displayed 0.8% divergence from previously sequenced pea aphid U-type symbionts (AYl36138, Russell et al. 2003). Finally, diagnostic PGR demonstrated that Aphis craccivora collected from alfalfa in Marana, AZ harbored a T-type symbiont. Its 16S rDNA sequence (AY136136) was 99.7% similar to T-type symbionts of A. pisum (AY136141, Russell et al. 2003). A. craccivora were brought back to the Moran lab and reared on alfalfa at 16 hours light, and at a constant 22°G. Individuals of each of these three species were used as donors in microinjection experiments, in an attempt to transfer their symbionts to A. pisum. 89 Transmission Efficiency: Injections Secondary symbionts reside in multiple locations within their aphid hosts, including the bacteriocytes and hemolymph (Buchner 1965, Chen et al. 1996, Fukatsu et al. 2000, Fukatsu 2001b, Sandstrom et al. 2001). Previous studies have reported the successful transfer of these bacteria between aphids by microinjection of hemo lymph from an infected aphid (Chen and Purcell 1997, Fukatsu 2001b, Oliver et al. 2003). We used a similar approach to transfer symbionts among aphids, first extracting body fluids (primarily hemo lymph) from late instar juveniles with a glass microcapillary tube pulled into a fine needle. The contents were injected into the abdominal segments of 2nd-4th instcir juveniles of secondary-free A. pisum clones 7a and 5 A. All aphids used in microinjections were immobilized on a pipet tip, which was attached to a vacuum and placed under a dissecting microscope. Injected aphids were placed onto fava bean plants and reared under the conditions described above. Field-collected Macrosiphum euphorbiae and Myzocallis individuals were used as symbiont donors; alternatively, A. pisum and A. craccivora donors were obtained from lab-reared cultures. Transmission Efficiency: Maternal Transfer To determine the capacity for symbionts to infect novel host species, we compared maternal transmission efficiencies of "novel" Arsenophonus and U-type symbionts to that of "native" U-type symbionts originating from the pea aphid, A. pisum. Multiple females, injected in March 2002 or April 2003, were reared on fava beans until 6 days post-injection. At this time, injected survivors were used to start single female lineages. Offspring bom to single females between 8-10, 12-14, and 16-20 days post- injection were collected and frozen at -80°C. Similarly, surviving injected aphids were collected at 18 or 20 days post-injection, and frozen. Single female lineages were perpetuated for several generations, using individual descendants of injected aphids to establish and maintain^, pisum lines. We froze these descendant females and several of their offspring at -80°C, up through the F5 generation. These samples were eventually subjected to symbiont screening assays to determine infection status and to aid in our selection of lines to maintain for future experiments. DNA extractions were performed on single aphids, which had been preserved at -80°C (Bender et al. 1983, Russell et al. 2003). DNA samples were assayed for symbiont infection using diagnostic PGR, in which symbiont-specific 16S rDNA primers (ArslOlSF; 5'ATCCAGCGAATCCTTTAG 3' for Arsenophonus; andU1279F: 5' CGAACGTAAGCGAACCTCAT 3' for the U-type—Russell et al. 2003) were paired with a more general, or "universal" 23S rDNA primer (35R: 5'CCTTCATCGCCTCTGACTGC 3'). In amplifying across the intergenic spacer between the 16 and 23 S rRNA genes, we obtained products of specific and repeatable lengths which further aided in our diagnostic screening. PGR reactions were conducted at 10 }xl volumes as described previously (Russell et al. 2003), Briefly, each reaction contained 5.92 p,l water, 1 ^1 of Eppendorf Taq buffer (with 15 mM Mg^^, 1 |xl of 10 mM dNTP's, 0.8 |il of each primer (at 5 (iM), and 0.08 [il of Eppendorf Taq polymerase. The recipe was the same for Arsenophonus screens, except that 5.52 [il of water were 91 added in addition to 0.4 jil of 25 mM giving a final magnesium concentration of 2.5 mM (versus 1.5 mM for the other reactions). PCR cycling conditions were: 1 cycle of 94°C for 2 minutes; 35 cycles of 94°C for 1 minute, 58°C for 1 minute, and 72°C for 2 minutes; and 1 cycle of 72°C for 6 minutes. A 3 ^1 volume of a glycerol loading buffer was added to each reaction, and 6 jll of this mixture were loaded into a 2% agarose gel. We loaded 2.5 |il of 1 Kb Plus DNA ladder (Invitrogen) into each gel to provide a standard for estimating product size. Gels were electrophoresed at 90-100V, stained with ethidium bromide, then photographed under UV Hght exposure. Samples were declared positive for a given symbiont if they yielded a PCR product of the expected size that displayed a higher intensity than the DNA ladder. The intensity criterion was applied to prevent false positives, and, thus, our estimates of transmission efficiency are conservative. Samples were declared negative if they yielded no PCR product; those yielding faint products were labeled ambiguous, and were not considered in subsequent analyses. To assess the quality of each PCR template, all samples were assayed with the primers Buch 757F (5' GAGGAATACCYKTGGCGAAA 3'—from reference 91) and 1507R(5' TACCTTGTTACGACTTCACCCCAG 3'—from reference 19), which amplify a portion of the Buchnera 16S rRNA gene. Only samples testing positive in this template quality assay were considered in transmission efficiency analyses. 92 Fitness Experiments Aphids are cyclical parthenogens, and can be reared as clonal lines in the laboratory. Using microinjections we have created several genetically identical lines that differ only with respect to the symbionts they harbor. We refer to these distinct lines here as symbiotypes. We utilized a single clone of A. pisum (5 A) for all fitness experiments. This clone was naturally free of secondary symbionts, and through microinjections we obtained a total of 6 lines that were examined in fitness experiments: 1) uninfected (N), 2) native R-type infected (R), 3) native T-type infected (TAP), 4) native U-type infected, 5) infected with the novel T-type from Aphis craccivora (Tac), 6) infected with the novel Arsenophonus (Ars). Each of the symbiont infected lines was stable and exhibited highly efficient transmission, according to diagnostic PGR. To determine the outcomes of interactions with novel and native symbionts, we measured and compared mean relative growth rate (MRGR) between genetically identical aphids harboring different secondary symbiont complements. We also measured survival, as the proportion of aphids reaching adulthood. Artificially infected aphids used in experiments were at least 10 generations removed from microinjections. In August 2002, we conducted our first set of fitness experiments, examining MRGR and survival in three separate time replicates for four natural symbiotypes of^. pisum clone 5A (uninfected, R-, T-, or U-type infected) as well as a novel symbiotype {Arsenophonus-infectQd). Adult aphids, reared under identical conditions for multiple generations, were used to produce experimental aphids which were weighed in a tin foil boat on a micro balance (Cahn 29 Automatic Electro balance®) within 90 minutes of birth. 93 These aphids were then placed on fava beans and reared at 20°C at 18L:6D. To keep track of individuals, we placed only a single aphid on each plant. We checked cultures for adults daily, beginning on day 5, and recorded the dates of adulthood for all surviving individuals. Individual aphids were weighed, for a second time, on their first day of adulthood. MRGR was then calculated as follows: (hi adult wt - hi birth wt)/tinie to adulthood. In the summer of2003, we conducted another experiment to examine the effects of the novel T-type symbiont on A. pisum hosts. MRGR was estimated for pisum 5 A symbiotypes which were uninfected with secondary symbionts or which harbored either the native (Tap) or novel (Tac) T-type secondary. This experiment was conducted at both 20 and 25°C. The design of this experiment was otherwise identical to that for the 2002 experiments, except that cultures were checked for adults at 12 hour intervals to increase resolution. Statistics—Transmission Efficiency and Fitness All statistical analyses were performed using JMPIN® version 3.2.6. Transmission efficiency results were compared between symbionts using logistic regression. Offspring of single females were scored as infected or uninfected based on PGR results. Only broods for which at least two offspring were screened were included in the analysis. Logistic regression compared the odds of transmission of the U-type symbiont of M euphorbiae (UMB) and the Arsenophonus symbiont (Ars) to that of the A. pisum U-type (UAP). In all analyses, we pooled the results of transmission in clones 7a and 5A, having found no differences between these with respect to transmission 94 efficiency. Separate analyses were conducted on offspring produced during different time intervals after injections were performed (8-10 days, 12-14 days, and 16-20 days post-injection). Logistic regression coefficients were back-transformed (inverse natural log), to determine the fold difference in transmission efficiency between the novel symbionts and the native U-type. Finally, we performed an ANOVA to compare MRGR among aphids harboring different symbionts, determining statistical differences between pairs of symbiotypes through contrast tests. Results Symbionts were detected in offspring produced 8 to 10 days post injection for 7/24 parents injected with the native A. pisum U-type (UAP) and 14/23 parents treated with the novel Arsenophonus symbiont. In contrast, novel U-type symbionts donated from M. euphorbiae (UMC) were not detected in the broods of 26 injected A. pisum at this same time-point. Nearly all aphids injected with Uap (16/17) and With Arsenophonus (7/8) gave birth to infected offspring during the 12-14 day period, while only 33% (4/12) of aphids injected with UMB yielded infected offspring during this interval. This pattern remained throughout the 16-20 day interval, when only 2/18 parents transmitted the novel UMeto offspring. Logistic regression was used to compare transmission efficiency among the three injected symbionts at the sampled time-points in this first generation (see Table 1). Transmission efficiency could not be statistically compared between UMB and UAP for the day 8-10 interval because regression coefficient estimates were unstable, apparently due 95 to the complete absence of UMB transmission at this time-point. Despite this, it was clear that UAP transmission exceeded that of UMB at this time-point—11/81 offspring of UAP injected aphids were infected compared to 0/70 offspring of UMC injected aphids. At 12- 14 days post-injection, the odds of UAP transmission was 10.2 times greater than that for Uwe (p<0.0001). Efficiency was 9.4 times more likely for UAP during the 16-20 day time interval (p<0.0001). In contrast to the novel U-type, Arsenophonus underwent relatively efficient transmission in^. pisum. At 8-10 days Tpost-inicction Arsenophonus transmission was significantly more likely than UAP inheritance by a factor of two (p=0.0036). Efficiencies were statistically indistinguishable between Arsenophonus and UAP at days 12-14 (p=0.7622), whereas UAP transmission was 2.8 times more likely than that for Arsenophonus during the 16-20 day period (p<0.0001). We present our data on transmission rates, graphically, in Figure 1, excluding the results from the day 12-14 period which were collected for aphids from only 1 out of 2 experiments. The rarity of UMB in injected aphid offspring could have arisen due to a lack of symbiont persistence within injected A. pisum or due to low transmissibility of surviving symbionts. To help discriminate between these explanations, we conducted identical logisitic regression analyses on only those offspring of single female parents testing positive for the injected symbionts. Although the magnitudes of efficiency differences between UAP and UMC were slightlylower in these analyses, the results were qualitatively similar to the previous analysis on all offspring—^the naturally occurring U-type was over 7 times more likely to undergo vertical transmission. The novel UMC still infected significantly fewer offspring than UAP at the 12-14 and 16-20 day time points (data not 96 shown), suggesting an inability to colonize developing aphid embryos. However, we cannot rule out the possibility that UMC were present at lower densities than UAP, and thus more work is needed to determine the specific cause of UMB transmission failure. To follow transmission over a longer time-course, we screened subsequent generations for U-type and Arsenophonus symbionts. Nine lineages which transmitted the UAP symbiont to offspring in the first generation were screened in the second and third generations. Transmission was nearly perfect within these lineages, with only one instance of transmission failure (Table 2). We similarly screened five descendant lineages of Arsenophonus-m]ecXQd aphids, observing incomplete transmission within 2/5 lineages, both in the passage of symbionts from the 2"^ to generation. Despite imperfect transmission at this stage, we have successfully maintained single UAP and Arsenophonus lineages within the laboratory for dozens of generations demonstrating the overall stability of these infections. In contrast, we screened approximately two dozen single individuals from the F2- F5 generations which were descended from aphids injected with UMC- None of these tested positive (data not shown). We also tested pooled samples, containing DNA from 8 aphids. Each sample contained aphids from one of four lineages observed to transmit UMC in the first generation. None of these tested positive beyond the F2 generation (data not shown) and we conclude that this UME symbiont is incapable of utilizing A. pisum as a host. Fitness 97 In our attempt to dissect the consequences of novel host-symbiont interactions, we compared the fitness effects of symbiont infection between genetically identical aphids which were either free of secondary symbionts or which harbored novel or native symbionts transferred through microinjection. In the first experiment, we reared aphids at 20°C, comparing mean relative growth rates (MRGR) between uninfected, R-type- infected, T-type-infected, U-type-infected, and Arsenophonus- 'miQcied lines of an A. pisum clone (5A). Aphids infected with Arsenophonus survived at rates comparable to those of the other symbiotypes. However, Arsenophonus infection was not without consequences, as their hosts suffered a 9-13% reduction in MRGR compared to secondary free aphids and to aphids harboring any native symbiont (Figure 2a). Our analyses indicated that this was due to both a decrease in growth from birth to adulthood and prolonged development time (data not shown). Fitness detriments were observed for Arsenophonus-'mSQoXed aphids in experiments performed up to five months later, with this novel symbiont prolonging development £ind reducing fecundity of infected A. pisum reared at 25°C (J. Russell, unpublished data). Thus, fitness costs were a consistent trait of Arsenophonus-'mfQCiQd A. pisum. In contrast, there were no significant differences in MRGR among the other symbiotypes, revealing a neutral effect of the native R-, T-, and U-type symbionts. We also performed an experiment designed to measure the consequences of harboring a novel T-type symbiont that was closely related to the native T-type of A. pisum. MRGR was measured for three symbiotypes of A. pisum clone 5 A in this experiment—uninfected, A. pisum T-type infected (TAP), etndA. craccivora T-type 98 infected (Tac)- We found no significant differences in MRGR among symbiotypes at either 20 or 25°C (Figure 2 b,c), revealing a commensal relationship between aphids and both T-types. However, those infected with the A. craccivora T-type experienced elevated survival at 25°C compared to the other symbiotypes (likelihood ratio test; PTAON = 0.0164; PTAC>TAP = 0.0132), revealing a beneficial effect of this novel symbiont. Whereas 9/53 and 10/55 TAP -infected and uninfected aphids died before adulthood, only 2/56 aphids with TAC failed to reach maturity. Discussion Many clades of heritable bacterial symbionts are widely distributed across distantly related insect hosts, revealing histories of horizontal transfer or host switching. This has led to the prediction that these symbionts retain a general ability to survive, reproduce, and undergo efficient transmission in novel hosts. Indeed, several previous studies have revealed that Wolbachia can imdergo successful horizontal transfer in the lab, exhibiting efficient vertical transmission in novel host backgrounds (Braig et al. 1994, Rigaud et al. 2001, Sasaki and Ishikawa 2000). A similar capacity has been demonstrated for two secondary symbionts of A. pisum—^the R- and T-types—^which can stably persist in the aphids A. kondoi, and A. fabae, respectively (Chen and Puree11 1997, Darby and Douglas 2003). Here, we have shown that Arsenophonus and T-type secondary symbionts are capable of persisting in a novel host, A. pisum. Transmission efficiency of Arsenophonus was comparable to that for the native U-type of A. pisum. In addition, hoXh. Arsenophonus and the novel T-type have been stably maintained in the 99 laboratory for several dozen generations, revealing the physiological suitability of A. pisum as a host. While both of these novel symbionts were capable of infecting A. pisum, our results suggest that the T-type is better suited for a lifestyle within field populations of this host. Barring an unforeseen benefit, such as resistance to natural enemies (Ferrari et al. 2004, Oliver et al. 2003), Arsenophonus-'vafQcttdi A. pisum would be outperformed by other A. pisum lineages, dooming the transferred microbe to extinction. The novel T- type, in contrast, had no negative effects on the fitness parameters we measured, and even conferred a survival advantage when reared at 25°C. Novel symbionts that undergo efficient vertical transmission after laboratory transfer have occasionally been noted to reduce insect fitness. For example, the A. pisum R-type symbiont was highly detrimental to the novel aphid host, Acyrthosiphon kondoi. A. kondoi suffered drastic reductions in fecundity and longevity, and were slower to reach maturity (Chen et al. 2000). Wolbachia have similarly been observed to exert detrimental effects on some novel isopod hosts (Rigaud et al. 2001). The mechanisms behind these fitness consequences remain unidentified, though the results clearly show that host genotypes shape the outcome of symbiosis. In contrast to our results for the novel T-type and Arsenophonus symbionts, we failed to successfully transfer a U-type from the aphid Macrosiphum euphorbiae into novel A. pisum hosts, observing only sporadic instances of vertical transmission. Transmission efficiency was low even when considering the offspring of infected parents. At this time, we cannot conclusively state whether the lack of symbiont persistence was 100 due to an immediate effect of failed transmission, or to an upstream effect of reduced symbiont survival. Nonetheless, it appears that this U-type isolate is incapable of utilizing A. pisum, revealing variation among secondary symbionts in their range of suitable hosts. Our observation is not the first case of a failed attempt at symbiont transfer. Several Wolbachia isolates undergo inefficient transmission after experimental transfer between arthropod species (Boyle et al. 1993, Clancy and Hoffmann 1997, Fuji et al. 2001, Huigens et al. 2004, Riegler et al. 2004, van Meer and Stouthamer 1999), even when moveded among close relatives. These studies have not attempted to measure the survival or persistence of symbionts within their novel insect hosts. So, again, it is unclear whether failed transmission is more typically a direct result of poor colonization of developing eggs or embryos, or an effect of reduced symbiont survival in particular host environments. Discriminating between these possibilities will help to narrow the range of plausible mechanisms that shape the host ranges of heritable microbes. For example, reduced symbiont survival would suggest a possible role of the host immune system, as insects are generally capable of eliminating bacterial infections through the action of cellular and humoral responses (Hoffmann 1995). Alternatively, an inability to colonize developing embryos or eggs would suggest that symbionts lack specific virulence factors or are unable to respond to the host-derived cues required to colonize ovaries or developing embryos. Like chronic pathogens, such as the Mycoplasmas, heritable symbionts are entirely host-dependent—^they carmot survive or proliferate in the environment, requiring 101 individual hosts for growth and transmission. This dependence is typically associated with the inactivation and loss of genes, eventually yielding highly streamlined and specialized genomes that are among the smallest of all cellular organisms (Andersson and Kurland 1998, Cole et al. 2001, Fraser et al. 1995, Mira et al. 2001, Ochman and Moran 2001, Shigenobu et al. 2000). In the face of such gene loss, we must question how heritable symbionts have retained their abilities to infect multiple hosts. For symbionts that are maternally transmitted within a particular host lineage, it is hard to envision how selection could maintain genes for wide host ranges. Pleiotropy could play an important role, if genes enabling broad host ranges play more immediate roles in the day-to-day lifestyles of symbionts. Alternatively, intermittent selection could help to account for these observations and for the patterns of highly related symbionts infecting distantly related hosts. Under this scenario, symbionts that have retained genes required for broad host ranges obtain the largest representation among the most diverse set of taxa. This process would counteract the effects of gene loss and specialization occurring for symbionts within a specific host lineage. A third possibility involves a role for lateral gene transfer, whereby mobile genetic elements or bacteriophage promote the spread of novel genes across symbionts co-infecting the same host individuals. This source for potential evolutionary novelty could partially counteract the process of gene loss, facilitating the spread of beneficial genes across an array of microbes that often infect the same host individuals. Given the paucity of data on the mechanisms governing symbiont distributions, we cannot yet say whether the responsible genes fall within the category of those known to move between bacterial species. 102 In this study, we have focused on only two of three factors that govern the patterns of symbiont distributions—^transmission effiency and fitness effects. In addition, the rates and patterns of horizontal transfer in nature will depend heavily on the opportunities presented by ecological associations. Parasitic interactions have been argued to play a role in horizontal transfer. Both Wolbachia and Arsenophonus can spread among parasitic wasps infecting the same hosts, revealing the potential importance of parasitic interactions (Huigens et al. 1999, 2004). In addition, ingestion has been demonstrated as a potential means of transmission. In two separate studies, aphids and leafhoppers acquired a heritable symbiont "infection" after feeding on contaminated diets (Darby and Douglas 2003, Purcell et al. 1994). Research into both of these avenues is needed to elucidate the processes behind widespread distributions of bacteria that play substantial roles in insect ecology and evolution. 103 References Cited Adams, D., and A.E. Douglas. 1997. How symbiotic bacteria influence plant utilization by the polyphagous aphid. Aphis fabae. Oecologia 110:528-532. Aksoy, S,, X. Chen, and V. Hypsa. 1997. Phylogeny and potential transmission routes of midgut-associated endosymbionts of tsetse (Diptera; Glossinidae). Ins. Mol. Biol. 6:183-190. Andersson, S.G.E., and C.G. Kurland. 1998. Reductive evolution of resident genomes. Trends in Microbiol. 6:263-268. Andersson, S.G.E., A. Zomorodipour, J.O. Andersson, T. Sicheritz-Ponten, U.C.M. Alsmark, R.M. Podowski, A.K. Naslund, A.S. Eriksson, H.H. Winkler, and C.G. Kurland. 1998. The genome sequence of Rickettsia prowazekii and the origin of mitochondria. Nature 396:133-140. Bandi, C., M. Sironi, G. Damiani, L. Magrassi, C.A. Nalepa, U. Laudani, and L. Sacchi. 1995. The establishment of intracellular symbiosis in an ancestor of cockroaches and termites. Proc. R. Soc. Lond. Ser. B 259:293-299. Baumann, L., P. Baumann, N.A. Moran, J. Sandstrom, and M.L. Thao. 1999. Genetic characterization of plasmids containing genes encoding enzymes of leucine biosynthesis in endosymbionts (Buchnera) of aphids. J. Mol. Evol. 48:77-85. Bender, W., P. Spierer, and D.S. Hogness. 1983. Chromosomal walking and jumping to isolate DNA from the Ace and rosy loci and the bithorax complex in Drosophila melanogaster. J. Mol. Biol. 168:17-33. Bemays, E.A., and B.A. Klein. 2002. Quantifying the symbiont contribution to essential amino acids in aphids: the importance of tryptophan for Uroleucon ambrosiae. Physiol. Entomol. 27:275-284. Bourtzis, K, M.M. Pettigrew, and S.L. O'Neill. 2000. Wolbachia neither induces nor suppresses transcripts encoding antimicrobial peptides. Insect Mol. Biol. 9:635- 639. Boyle, L., S.L. O'Neill, H.M. Robertson, and T.L. Karr. 1993. Interspecific and intraspecific horizontal transfer of Wolbachia in Drosophila. Science 260:1796- 1799. Bracho, A.M., D. Martinez-Torres, A. Moya, and A. Latorre. 1995. Discovery and molecular characterization of a plasmid localized in Buchnera sp. bacterial endosymbiont of the aphid Rhopalosiphum padi. J. Mol. Evol. 41:67-73. 104 Braig, H.R., H. Guzman, R.B. Tesh, and S.L. O'Neill. 1994. Replacement of the natural Wolbachia symbiont of Drosophila simulans with a mosquito counterpart. Nature 367:453-455. Buchner, P. 1965. Endosymbiosis of animals with plant microorganisms. Interscience, New York, NY. Bull, J.J., LJ. Molineux, and W.R. Rice. 1991. Selection of benevolence in a host- parasite system. Evolution 45:875-882. Caspari, E., and G.S. Watson. 1959. On the evolutionary importance of cytoplasmic sterility in mosquitos. Evolution 13:568-570. Chen, D.Q., B.C. Campbell, and A.H. Purcell. 1996. A new Rickettsia from a herbivorous insect, the pea aphid Acyrthosiphon pisum (Harris). Curr. MicrobioL 33:123-128. Chen, D.Q., C.B. Montllor, and A.H. Purcell. 2000. Fitness effects of two facultative endosymbiotic bacteria on the pea aphid Acyrthosiphon pisum, and the blue alfalfa aphid, kondoi. Entomol. Exp. Appl. 95:315-323. Chen, D.Q., and A.H. Purcell. 1997. Occurrence and transmission of facultative endosymbionts in aphids. Curr. Microbiol. 34:220-225. Clancy, D.J., and A.A. Hoffmann. 1997. Behavior of Wolbachia endosymbionts from Drosophila siumulans in Drosophila serrata, a novel host. Am. Nat. 149:975-988. Cole, S.T., K. Eiglmeier, J. Parkhill, K.D. James, N.R. Thomson, P.R. Wheeler, N. Honore, T. Gamier, C. Churcher, D. Harris, K. Mungall, D. Basham, D. Brown, T. Chillingworth, R. Connor, R.M. Davies, K. Devlin, S. Duthoy, T. Feltwell, A. Eraser, N. Hamlin, S. Holroyd, T. Homsby, K. Jagels, C. Lacroix, J. Maclean, S. Moule, L. Murphy, K. Oliver, M.A. Quail, M.A. Rajandream, K.M. Rutherford, S. Rutter, K. Seeger, S. Simon, M. Simmonds, J. Skelton, R. Squares, S. Squares, K. Stevens, K. Taylor, S. Whitehead, J.R. Woodward, and B.G. Barrell. 2001. Massive gene decay in the leprosy bacillus. Nature 409:1007-1011. Darby, A.C., L.M. Birkle, S.L. Turner, and A.E. Douglas. 2001. An aphid-bome bacterium allied to the secondary symbionts of whitefly. FEMS Microbiol. Ecol. 36:43-50. Darby, A.C., and A.E. Douglas. 2003. Elucidation of the transmission patterns of an insect-bome bacterium. Appl. Environ. Microbiol. 69:4403-4407. 105 Douglas, A.E. 1988. Sulfate utilization in an aphid symbiosis. Insect Biochem, 18:599- 605. Douglas, A.E. 1992. Requirement of pea aphids {Acyrthosiphon pisum) for their symbiotic bacteria. Entomol. Exp. Appl. 65:195-198. Douglas, A.E. 1998. Nutritional interactions in insect-microbial symbioses: Aphids and their symbiotic bacteria Arm. Rev. Entomol. 43:17-37. Douglas, A.E., and W.A. Prosser. 1992. Synthesis of the essential amino acid tryptophan in the pea aphid (Acyrthosiphon pisum) symbiosis. J. Insect Physiol. 38:565-568. Ewald, P. 1983. Transmission modes and the evolution of the parasitism-mutualism continuum. Ann. N. Y. Acad. Sci. 503:295-306. Ferrari, J., A.C. Darby, T.J. Daniell, H.C.J. Godfray, and A.E. Douglas. 2004. Linking the bacterial community in pea aphids with host-plant use and natural enemy resistance. Ecol. Entomol. 29:60-65. Fraser, C.M., J.D. Gocayne, O. White, M.D. Adams, R.A. Clayton, R.D. Fleischman, C.J. Bult, A.R. Kerlavage, G. Sutton, J.M. Kelley, J.L. Fritchman, J.F. Weidman, K.V. Small, M. Sandusky, J. Fuhrmann, D. Nguyen, T.R. Utterback, D.M. Saudek, C.A. Phillips, J.M. Merrick, J.F. Tomb, B.A. Dougherty, K.F. Bott, P.C. Hu, T.S. Lucier, S.N. Peterson, H.O. Smith, C.A. Hutchison, and J.C. Venter, 1995. The minimal gene complement of Mycoplasma genitalium. Science 270:397-403. Fujii, Y., D. Kageyama, S. Hoshizaki, H. Ishikawa, and T. Sasaki. 2001. Transfection of Wolbachia in Lepidoptera: the feminizer of the adzuki bean borer Ostrinia scapulalis causes male killing in the Mediterranean flour moth Ephestia kuehniella. Proc. R. Soc. Lond. Ser. B 268:855-859. Fukatsu, T. 2001a. Secondary intracellular symbiotic bacteria in aphids of the genus Yamatocallis. (Homoptera:Aphididae;Drepanosiphinae). Appl. Environ. Microbiol. 67:5315-5320. Fukatsu, T. 2001b. Spiroplasma symbiont of the pea aphid, Acyrthosiphon pisum (Insecta: Homoptera). Appl. Environ. Microbiol 67:1284-1291. Fukatsu, T., N. Nikoh, R. Kawai, and R. Koga. 2000. The secondary endosymbiotic bacterium of the pea aphid Acyrthosiphon pisum (Insecta: Homoptera). Appl. Environ. Microbiol. 66:2748-2758. 106 Gherna, R.L., J.H. Werren, W. Weisburg, R. Cote, C.R. Woese, L. Mandelco, and D.J. Brenner. 1991. Arsenophonus nasoniae gen. no v., sp. no v., the causative agent of the son-killer trait in the parasitic wasp Nasonia vitripennis. Int. J. Syst. Bacteriol. 41:563-565. Grindle, N., J.J. Tyner, K. Clay, and C. Fuqua. 2003. Identification of Arsenophonus- type bacteria from the dog tick Dermacentor variabilis. J. Invertebr. Pathol. 83:264-266. Haynes, S., A.C. Darby, T.J. Daniell, G. Webster, F.J.F. van Veen, H.C.J. Godfray, J.L. Prosser, and A.E. Douglas. 2003. Diversity of bacteria associated with natural aphid populations. Appl. Environ. Microbiol. 69:7216-7223. Heath, B.D., R.D.J. Butcher, W.G.F. Whitfield, and S.F. Hubbard. 1999. Horizontal transfer of Wolbachia between phylogenetically distant insect species by a naturally occurring mechanism. Curr. Biol. 9:313-316. Hoffmann, J.A, 1995. Innate immunity of insects. Curr. Opin. Immunol. 7:4-10. Huigens, M.E., R.P. de Almaida, P.A.H. Boons, R.F. Luck, and R. Stouthamer. 2004. Natural interspecific and intraspecific horizontal transfer of parthenogenesis- inducing Wolbachia in Trichogrammav^as^s. Proc. R. Soc. Lond. Ser. B 271:509- 515. Huigens, M.E., R.F. Luck, R.H.G. Klassen, M.F.P.M. Maas, M.J.T.N. Timmermans, and R. Stouthamer. 1999. Infectious parthenogenesis. Nature 405:178-179. Hurst, G.D.D., H. Anbutsu, M. Kutsukake, and T. Fukatsu. 2003. Hidden from the host: Spiroplasma bacteria infecting Drosophila do not cause an immune response, but are suppressed by ectopic immune activation. Insect Mol. Biol. 12:93-97. Hurst, G.D.D., F. M. Jiggins, J.H.G. von der Schulenbut^, D. Bertrand, S.A. West, I.I. Goriacheva, I.A. Zakharov, J.H. Werren, R. Stouthamer, and M,E.N. Majerus. 1999. Male-killing Wolbachia in two species of insect. Proc. Roy. Soc. Lond. Ser. B 266:735-740. Hypsa, v., and S. Aksoy. 1999. Phylogenetic characterization of two transovarially transmitted endosymbionts of the bedbug Cimex lectularius (Heteroptera: Cimicidae). Insect Mol. Biol. 6:301-304. Kellner, R.L.L, and K. Dettner. 1996. Differential efficacy of toxic pederin in deterring potential arthropod predators of Paederus (Coleoptera: Staphylinidae) offspring. Oecologica 107:293-300. 107 Lai, C.Y., L. Baumann, and P. Baumann. 1994. Amplification of trpEG; adaptation of Buchnera aphidicola to an endosymbiotic association with aphids. Proc. Natl. Acad. Sci. USA 91:3819-3823. Lefevre, C., H. Charles, A. Vallier, B. Delobel, B. Farrell, and A. Heddi. 2004. Endosymbiont phylogenies in the Dryophthoridae weevils: evidence for bacterial replacement. Mol. Biol. Evol. 21:965-973. Leonardo, T.E. 2004. Removal of a specialization-associated symbiont does not affect aphid fitness. Ecol. Letters 7:461-468. May, R.M. and R.M. Anderson. 1982. Coevolution of hosts and parasites. Parasitology 85:411-426. Mira, A., H. Ochman, and N.A. Moran. 2001. Deletional bias and the evolution of bacterial genomes. Trends in Genetics 17:589-596. Montllor, C.B., A. Maxmen, and A.H. Purcell. 2000. Facultative bacterial endosymbionts benefit pea aphids Acyrthosiphon pisum under heat stress. Ecol. Entomol. 27:189-195. Moran, N., and P. Baumann. 1994. Phylogenetics of cytoplasmically inherited microorganisms of arthropods. Trends. Ecol. Evol. 9:15-20. Moran, N.A., G.R. Plague, J.P. Sandtrom, and J.L. Wilcox. 2004. A genomic perspective on nutrient provisioning by bacterial symbionts of insects. Proc. Natl. Acad. Sci. USA 100:14543-14548. Moran, N.A., and A. Telang. 1998. The evolution of bacteriocyte-associated endosymbionts in insects. Bioscience 48:295-304. Munson, M.A., P. Baumann, M.A. Clark, L. Baumann, N.A. Moran, D.J. Voegtlin, and B.C. Campbell. 1991a. Evidence for the establishment of aphid-eubacterial endosymbiosis in an ancestor of four aphid families. J. Bacteriol. 173:6321-6324. Munson, M.A., P. Baumann, and M.G. Kinsey. 1991b. Buchnera gen. nov. and Buchnera aphidicola sp. nov., a taxon consisting of the mycetocyte-associated, primary endosymbionts of aphids. Int. J. Syst. Microbiol. 41:566-568. Oliver, K.M., J.A. Russell, N.A. Moran, and M.S. Hunter. 2003. Facultative bacteria in aphids confer resistance to parasitic wasps. Proc. Natl. Acad. Sci. USA 100:1803-1807. 108 Ochman, H., and N.A. Moran. 2001. Genes lost and genes found; evolution of bacterial pathogenesis and symbiosis. Science 292:1096-1098. O'Neill, S.L., R Giordano, A.M.E. Colbert, T.L. Karr, and H.M. Robertson. 1992. 16S rRNA phylogenetic analysis of the bacterial endos5anbionts associated with cytoplasmic incompatibility in insects. Proc. Natl. Acad. Sci. USA 89:2699-2702. Piel, J. 2002. A polyketide synthase-peptide synthetase gene cluster from an uncultured bacterial symbiont oiPaederus beetles. Proc. Natl. Acad. Sci. USA 99:14002- 14007. Prosser, W.A., and A.E. Douglas. 1991. An analysis of chlorotetracycline-treated pea aphid, Acyrthosiphon pisum. J. Insect Physiol. 37:713-719. Purcell, A.H., K.G. Suslow, and M. Klein. 1994. Transmission via plants of an insect pathogenic bacterium that does not multiply or move in plants. Microb. Ecol. 27:19-26. Riegler, M. S. Charlat, C. Stauffer, and H. Mer9ot. 2004. Wolbachia transfer from Rhagoletis cerasi to Drosophila simulans: investigating the outcomes of host- symbiont coevolution. Appl. Environ. Microbiol. 70:273-279. Rigaud, T., F.S. Pennings, and P. Juchault. 2001. Wolbachia bacteria effects after experimental interspecific transfers in terrestrial isopods. J. Invertebr. Pathol. 77:251-257. Rouhbaksh, D., N.A. Moran, L. Baumann, D.J. Voegtlin, and P. Baumann. 1994. Detection of Buchnera, the primary prokaryotic endosymbiont of aphids, using the polymerase chain reaction. Insect Mol. Biol. 3:213-217. Rousset, F., D. Bouchon, B. Pintureau, P. Juchault, and M. Solignac. 1992. Wolbachia endosymbionts responsible for various alterations of sexuality in arthropods. Proc. R. Soc. Lond. Ser. B 250:91-98. Rousset, F., D. Vautrin, and M. Solignac. 1992. Molecular identification of Wolbachia, the agent of cytoplasmic incompatibility in Drosophila simulans, and variability in relation with host mitochrondrial types. Proc. Roy. Soc. Ser. B 247:163-168. Russell, J.A., A. Latorre, B. Sabater-Munoz, A. Moya, and N.A. Moran. 2003. Side stepping secondary symbionts: widespread horizontal transfer across and beyond the Aphidoidea. Mol. Ecol. 12:1061-1075. Sandstrom, J.P., and N.A. Moran. 2001. Amino acid budgets in three aphid species using the same host plant. Physiol. Entomol. 26:202-211. 109 Sandstrom, J.P., J.A. Russell, J.P. White, and N.A. Moran. 2001. Independent origins and horizontal transfer of bacterial symbionts of aphids. Mol. EcoL 10:217-228. Sasaki, T., and H. Ishikawa. 2000. Transinfection of Wolbachia in the Mediterranean flour moth, Ephestia kuehniella, by embryonic microinjection. Heredity 85:130- 135. Shigenobu, S., H. Watanabe, M. Hattori, Y. Sasaki, and H. Ishikawa. 2000. Genome sequence of the endocellular bacterial symbiont of aphids Buchnera sp. APS. Nature 407:81-86. Simon, J.C., S. Carre, M. Boutin, N. Prunier-Leterme, B. Sabater-Munoz, A. Latorre, and R. Boumoville. 2003. Host-based divergence in populations of the pea aphid: insights from nuclear markers and the prevalence of facultative symbionts. Proc. R. Soc. Lond. Ser. B 270:1703-1712. Sironi, M., C. Bandi, L. Sacchi, B. DiSacco, G. Damiani, and C. Genchi. 1995. Molecular evidence for a close relative of the arthropod endosymbiont Wolbachia in a filarial worm. Mol. Biochem. Parasitol. 14:223-221. Stouthamer, R., J.A.J. Breeuwer, R.F. Luck, and J.H. Werren. 1993. Molecular identification of microorganisms associated with parthenogenesis. Nature 361:66- 68. Stouthamer, R., J.A. J. Breeuwer, R.F. Luck, and J.H. Werren. 1993. Molecular identification of microorganisms associated with parthenogenesis. Nature 361:66- 68. Thao, M.L., M.A. Clark, L. Baumann, E.B. Brennan, N.A. Moran, and P. Bauman. 2000a. Secondary endosymbionts of psyllids have been acquired multiple times. Curr. Microbiol. 41:300-304. Thao, M.L., N.A. Moran, P. Abbot, E.B. Brennan, D.H. Burckhardt, and P. Baumann. 2000b. Cospeciation of psyllids and their primary prokaryotic endosymbionts. Appl. Environ. Microbiol. 66:2898-2905. Tsuchida, T., R. Koga, and T. Fukatsu. 2004. Host plant specialization governed by facultative symbiont. Science 303:1989. Tsuchida, T., R. Koga, H. Shibao, T. Matsumoto, and T. Fukatsu. 2002. Diversity and geographic distribution of secondary endosymbiotic bacteria in natural populations of the pea aphid, Acyrthosiphon pisum. Mol. Ecol. 11:2123-2135. 110 Turelli, M. 1994. Evolution of incompatibility-inducing microbes and their hosts. Evolution 48:1500-1513. Unterman, B.M., P. Baumann, and D.L. McLean. 1989. Pea aphid symbiont relationships established by analysis of 16S ribosomal-RNA's. J. Bacteriol. 171:2970-2974. van Meer, M.M.M., and R. Stouthamer. 1999. Cross-order transfer of Wolbachia from Muscidifurax uniraptor (Hymenoptera: Pteromalidae) to Drosophila simulans (Diptera; Drosophillidae). Heredity 82:163-169. Werren, J.H., G.D.D. Hurst, W. Zhang, J.A.J. Breeuwer, R. Stouthamer, and M.E.N. Majerus. 1994. Rickettisal relative associated with male killing in the ladybird beetle (Adalia bipunctata). J. Bacteriol. 176:388-394. Werren, J.H., and D.M. Windsor. 2000. Wolbachia infection frequencies in insects: evidence of a global equilibirium? Proc. R. Soc. Lond. Ser. B 267:1277-1285. Werren, J.H., D. Windsor, and L.R. Guo. 1995. Distribution of Wolbachia among neotropical arthropods. Proc. R. Soc. Lond. Ser. B 262:55-63. Wicker, C. 1983. Differential vitamin and choline requirements of symbiotic and aposymbiotic S. oryzae (Coleoptera: Curculionidae). Comp. Biochem. Physiol. 76A:177-182. Zchori-Fein, E., and J.K. Brown. 2002. Diversity of prokaryotes associated with Bemisia tabaci (Gennadius) (Hemiptera:Aleyro.didae). Ann. Entomol. Soc. Am. 95:711-718. Ill Acknowledgments We thank Kim Hammond, Phat Tran, Michelle Hoffman, Jean Tsai, and Rachelle Bond for technical assistance. We also thank Susannah Elwyn, Molly Hunter, and Alex Wilson for helpfiil comments on this manuscript. This research was supported by NSF grant. J. Russell was supported by the NSFIGERT program in genomics, the Center for Insect Science, and the Plant Insect Interaction Group at the University of Arizona. 112 Days Regression equation P2' Post- injection 8-10 Y= 1.33 + 0.66Ars® p=0.0036; n.a. 95% CI: 0.22to l.II 12-14 Y = -0.289 - O.HArs"- 2.32UMe p= 0.7622; p<0.0001; 95% CI: 95% CI: -0.74 to 1.02 -1.63 to-3.01 16-20 Y = 2.21 - 1.02Ars-2.24UMe p<0.0001; p<0.0001; 95% CI: 95% CI: -0.52 to-1.52 -1.47 to-3.01 Table 1 Logistic regression analyses of symbiont transmission. ® Regression equation is Y = Po + Pi Ars. Po represents the intercept, Pi represents the In of difference in transmission failure between aphids injected with UAP VS. Ars. ^Regression equation is Y = PO + PI Ars+ P2 UMC- PO represents the intercept. PI represents the In of difference in transmission failure between aphids injected with UAP vs. Ars, and P2 represents the In of the difference in transmission failure between aphids injected with UAP VS. UMC- *Statistics reveal whether p parameters differed significantly from zero and, thus, whether the odds of transmission efficiency differed between the native symbiont and the novel symbionts. 113 Symbiont Proportion of Proportion of Infected F2 Infected F3 Uap 8/8 7/8 Uap 9/9 8/8 Uap 7/7 8/8 Uap 6/6 8/8 Uap 8/8 7/7 Uap 8/8 8/8 Uap 9/9 8/8 Ars 8/8 6/6 Ars 8/8 4/5 Ars 8/8 7/7 Ars 8/8 7/7 Ars 7/7 0/3 Table 2 Transmission efficiencies for multiple A. pisum lineages in the second (F2) and third (F3) generations post-injection. 114 FIGURE LEGENDS Figure 1 Transmission efficiencies of three injected symbionts from injected parents to their offspring. The proportions of offspring in which symbionts were detected are presented for a) 8 - 10 (hatched), or 16 - 20 (solid) days post-injection. Fractions above data bars represent the number of positive offspring/the total number of screened offspring. Values within bars represent the number of parents whose offspring were screened. Figure 2 Effects of secondary symbionts on mean relative growth rate (MRGR) of Acyrthosiphon pisum. a) Experiment performed at 20°C. N = uninfected; A = Arsenophoms-'mfeciQd; R = R- type infected; T == T-type infected; U = U-type infected. All are native A. pisum symbionts except for A. b) Experiment performed at 20°C. N = uninfected; T = native T-type infected; TAC- infected with novel T-type from Aphis craccivora. c) Experiment performed at 25°C. Abbreviations are same as in "b". Numbers within bars represent sample sizes whereas letters at the tops of bars reveal statistical differences—ie. those with different letters are significantly different (p<0.05) according to contrast tests. Mean relative growth rates were calculated as described in methods. 115 Figure 1 1 >< 0.9 4) 42/60 11/81 Ars UAp UMe Injected Symbiont Mean relative growth rate (mg/mg/day) pOOOOpOO CoCOCOCJCOifkit^A ONJ-t^0>00ON3-f^ M M o Mean relative growth rate (mg/mg/day) p P p p O O O O CO GJ CO CJ W 4^ ji. O N3 05 00 O hO O) C/) VC 3 O" O cn 3l-K •D 0 Ul Mean relative growth rate (mg/mg/day) O o p o GO Jik. Ui cn o cn o Ui *< 3 o- o" 3 •c (D 119 APPENDIX D. CONDITIONAL FITNESS EFFECTS OF SECONDARY SYMBIONTS: EXPOSURE TO HIGH TEMPERATURES LEADS TO STRONG COSTS AND BENEFITS OF SYMBIONT INFECTION IN THE PEA APHID, ACYRTHOSIPHONPISUM 120 Abstract Symbiosis is prevalent throughout the tree of life and has had a significant impact on the ecology and evolution of many bacteria and eukaryotes. The outcomes of symbiotic interactions often vary in their benevolence depending on various biotic and abiotic components of the environment. Such variation is expected to play an important role in shaping the prevalence and distributions of symbiosis throughout nature. In this study, we examine the effects of three distinct bacterial symbionts—^the R-, T-, and U- types—on the fitness of their aphid hosts under varying temperature conditions. We find that the effects of individual symbionts on Acyrthosiphon pisum vary depending on when and whether aphids are exposed to a brief period of heat shock. We also discover variation among the symbionts in their capacities to benefit to hosts after, finding that even closely related R-type differ in their exerted effects. Overall, our findings do not indicate a imiversal role for secondary symbionts in heat tolerance, but suggest the possibility for an overlapping role of R- and T-type symbionts in the tolerance of extreme temperatures. The mixture of costs and benefits conferred by the symbionts studied here suggests their potential significance in natural populations as heritable elements shaping the adaptation of aphids to their abiotic environments. 121 Introduction Bacteria commonly interact with insects in intimate associations known as symbioses. Many of these microbes behave in a fashion analogous to cytoplasmic elements, moving among hosts via maternal transmission. Due to their dependence on host reproduction, natural selection is expected to favor symbionts that increase host fitness (Ewald 1983, Bull et al. 1991, May and Anderson 1982). Accordingly, several heritable bacteria of insects have been shown to confer benefits on their hosts (e.g. Buchner 1965, Wicker andNardon 1982). For the past several decades, aphids (Insecta: Homoptera; Aphididae) have received attention from biologists interested in dissecting the mechanisms and consequences of symbiosis. These insects are specialized on diets of plant sap, which are typically lacking in essential amino acids (e.g. Douglas 1993, Sandstrom and Moran 1999). Nearly all aphids harbor an essential, maternally inherited primary symbiont known as Buchnera aphidicola (Munson et al. 1991a,b), which supplements host nutrition through synthesis and provisioning of essential amino acids (Douglas and Prosser 1992, Douglas 1988, Douglas et al. 2001, Bracho et al. 1995, Lai et al. 1994, Sandstrom and Moran 2001, Bemays and Klein 2002, Shigenobu et al. 2000, Munson and Baumann 1993). In addition, several aphids have been shown to harbor other maternally transmitted bacteria, known as secondary symbionts (Buchner 1965, Unterman et al 1989, Chen et al. 1996, Sandstrom et al 2001, Darby et al. 2001, Russell et al. 2003, Fukatsu 2001a,b). Most secondary symbionts are considered non-essential, and they are generally found at intermediate frequencies within host populations (e.g. 122 Chen and Purcell 1997, Tsuchida et al. 2002, Haynes et al. 2003; but see Fukatsu 2001a and Sandstrom et al. 2001). The significance of these symbionts has remained elusive until recently—^there is now evidence revealing that they play roles in defense against natural enemies (Oliver et al. 2003, Ferrari et al. 2004) and host-plant utilization (Tsuchida et al. 2004). In addition, "R-type" secondary symbionts can apparently confer heat tolerance m Acyrthosiphon pisum (Chen et al. 2000, Montllor et al. 2002). Temperture is an important abiotic factor shaping the biology of aphids. These insects are highly heat-sensitive, and often fail to develop or reproduce when reared at constant temperatures close to 30°C (e.g. Turak et al. 1998, Asin and Pons 2001, Wang and Tsai 2001, Wang and Tsai 2000) or when exposed to brief periods of temperatures between 30 and 40°C (e.g. Ma et al. 2003). In fact, A. pisum suffers drastic fitness reductions at constant temperatures as low as 25°C (Chen et al. 2000) and when exposed to a 37°C heat shock for periods as brief as 3 hours (Ohtaka and Ishikawa 1991). The consequences of heat sensitivity are expected to shape the ranges and distributions of natural populations of aphids, perhaps explaining the scarcity of these insects in the tropics. Findings that a secondary symbiont confers heat tolerance in A. pisum suggest that maternally transmitted microbes may play a role in adaptation under the selective forces exerted by high temperatures, thereby shaping the geographic and seasonal distributions of hosts. Here, we examine the effects of three secondary bacterial symbionts on the fitness of^. pisum, placing a large focus on their roles in aphids exposed to heat shock. In these experiments, we use genetically identical lines of aphids, artificially infected with three 123 groups of secondary symbionts from the y-Proteobacteria, known as the R-, T-, and U- types (Sandstrom et al. 2001). We compare the survival, fecundity, and generation times between infected and uninfected aphids which were either reared constantly at a permissive temperature or exposed briefly to a heat shock treatment. Using our survival data, we model the change in symbiont frequency within aphid populations, demonstrating how the fitness effects observed in our experiments could lead to rapid changes within host populations. Materials and Methods Aphids and plants In these experiments we utilized six lines of an A. pisum clone referred to as 5 A: the natural uninfected line (N) and five artificially infected lineages harboring either the R-type (R^', R^^, R'^), T-type (T), or U-type (U). We have maintained these cultures as parthenogenetic clones in the laboratory. These lines are genetically identical, differing only in their symbiont complements. This allows us to attribute any fitness differences directly to the symbionts. Artificially infected lines were estabhshed by microinjecting juveniles of the pink A. pisum clone ("5 A", from Madison, Wisconsin) with body fluids from^. pisum donors that harbored an R-, T-, or U-type symbiont (as in Oliver et al. 2003). Donors used to establish each infected line belonged to a single aphid clone collected from Cayuga or Tompkins county in New York (for R'^^, T, U), Madison, Wisconsin (for R^'), or 124 Tucson, Arizona (for R^). Before injections, each of these donor clones was screened for R-type, T-type, U-type, and PAR (Rickettsia) secondary symbionts using diagnostic PCR (see Sandstrom et al. 2001 for methods). All donors were found to harbor only a single secondary type. Injection survivors were used to start single female lineages. Using diagnostic PCR, we screened these lineages for symbionts at various intervals over the first several generations. We selected stably infected lines for continued maintenance and fiiture experiments. These lines were screened on several occasions leading up to the experiments; no instances of transmission failure were detected, in agreement with previous findings of highly efficiency maternal transfer (Chen et al. 1997, Darby et al. 2003). All artificially infected lines were reared in the laboratory for at least 10 generations before use in these experiments. Aphids were reared at 18°C 16L:8D on fava bean plants (Viciafaba) for at least two generations prior to experiments. Plants used in experiments were in the pre-flowering stage and were synchronously aged within each time replicate conducted within experiments. These plants were grown in 4 inch diameter pots, and cultures were enclosed within transparent SOLO® cups containing mesh tops. Experiments Our protocol was designed to mirror that used in a similar study by Montllor et al. (2002). Several adult aphids of comparable age were placed onto fava bean plants, and after watering, cultures were placed in a Percival Scientific® growth chamber and held at 125 18°C. We removed adults after 12-14 hours (day 1) and designated cultures for one of three treatments: "control", "day 2 heat shock", or "day 6 heat shock". Each culture contained multiple juvenile aphids, sired by several parents; several replicate cultures of each line were treated in all experiments. Nymphs subjected to the day 2 heat shock treatment were counted and placed into a separate growth chamber set at 18°C on day 2. The growth chamber temperature was gradually increased to 37.5°C over a span of two hours (+/- 0.5°C). The temperature was held constant for four hours, and then gradually decreased to 18°C. Aphids were placed back in the original growth chambers, and reared side-by-side with their control treatment counterparts at 18°C. Juveniles subjected to the day 6 heat shock treatment were handled in a fashion identical to the day 2 treatment, with the exception that nymph counting and heat shock were executed on day 6. For the control treatment, aphids were kept at 18°C for the entire experiment. To allow side-by- side comparisons of survival with the two separate heat shock treatments, these aphids were counted at either day 2 or day 6. Finding that aphids from these two classes of controls did not differ in their fitness attributes, we pooled these data for our analyses. We monitored growth chamber temperatures using Oregon Scientific® probes. Temperatures were also measured inside of cup cages using Oregon Scientific® probes or KoolTrak® data loggers. A comparison between the two measures revealed that temperatures next to fava bean plants were comparable to external temperatures. Two separate heat shock experiments were performed. The first (R-type experiment) included the N (uninfected), R^, R^', and R^^ lines of A. pisum clone 5 A, and was conducted in several blocks between September, 2003 and March, 2004. Each 126 infected line harbored a distinct R-type isolate which was donated from one of three A. pisum clones collected from New York, Wisconsin, or Arizona. In this experiment, we conducted each of the three temperature treatments. The day 2 heat shock treatment was performed over seven separate time blocks, yet in two of these nearly all of the aphids died before adulthood. We disregarded these trials, leaving us with data from five blocks. For the control and day 6 heat shock treatments, we performed experiments over five and four time blocks, respectively. The second experiment (T- and U-type experiment) was conducted in May and June of 2004, and included N, T, and U lines of clone 5 A. In these trials, only the control (three time blocks) and day 2 heat shock (two time blocks) treatments were performed. . To prevent positional effects within grovv1;h chambers, individual cultures were randomly rearranged on several occasions during each experiment. Generation time was measured as the number of days from birth to adulthood. Beginning on day 9 of the experiment, we checked cultures for adults daily until all had died or matured. We recorded the number of mature aphids in each culture observed on each day, removing adults to prevent confusion with juveniles. Using these data, and the counts of pre-treated juvenile aphids, we measured the proportion of aphids surviving to adulthood for each culture. We selected adults from each culture for fecundity assays, using those that reached adulthood earliest. These aphids were placed, individually, on fava bean plants and reared at 18°C 16L:8D. We counted offspring for each female at days 5 and 9 of adulthood (with day 1 corresponding to the day adults were first noticed and used to set 127 up single female lines). On day 5, we removed all nymphs after counting to prevent confiision w/later-born progeny. Typically, four individuals per treated culture were selected for fecundity measures. However, this sampling scheme was limited by the fact that several heat shocked cultures had fewer than four survivors. Symbiont Frequency Model Under the day 2 heat shock treatment, aphids from R-type-infected lines had unequivocally higher fitness than uninfected aphids. To model the expected change in symbiont frequency within^, pisum populations given these observed fitness differences, we computed the weighted average of survival for each line (under the day 2 heat shock treatment), thus accounting for differences in sample sizes among cultures. These values were used to calculate relative fitness for aphids in a population containing two symbiont complements—1) uninfected and 2) infected with a single R-type isolate. Since secondary symbionts are maternally transmitted and since aphids are parthenogenetic for most of their breeding season, we treated symbiont infection status in a fashion analogous to a haploid locus in population genetic models. We used the following equations to measure the change in frequency of aphid lines under viability selection: Tx+i =[(rx)(wr)(l-^))] /mean w Hx+i =[(nx)(wn)+(rx)( |i)]/mean w Where: Tx = frequency of R-type infected aphids in the generation iix= frequency of uninfected aphids in the x"' generation 128 Wr = relative fitness of the R-infected line (equal to one in all runs) w„ = relative fitness of the uninfected line (equal to the weighted average of survival for uninfected aphids divided by that for fitter R-infected aphids) H = the proportion of uninfected offspring produced by an R-infected mother mean w = (rx)(wr) + (nx)( Wn) where rx + nx = 1 In all runs, ro was set to 1/100,000. The relative fitness values, obtained from weighted averages of survival were as follows: R'^vs. N: Wr=l;Wn=0.65 R^' vs. N: Wr= 1; Wn= 0.61 R^^vs. N: wr=l;wn=0.59 Though transmission failure has rarely been observed in the lab, we included a "mutation" parameter that allowed for occasional loss of the symbiont at a rate of 2% per generation. This value equals the maximum estimate of transmission failure for the T- type secondary symbiont in the laboratory (Darby and Douglas 2003). Thus, the symbiont frequencies obtained in our analyses reflect a balance between both selection and transmission efficiency. Statistical Analyses The distributions of development times and fecundities were non-normal. We attempted several data transformations, yet none produced a normally-distributed dataset. Since these deviations violated assumptions of parametric statistical tests, we used non- 129 parametric Wilcoxon rank sum tests to analyze development time and fecundity. We conducted pairwise comparisons between all lines within each experiment, allowing us to estimate the costs and benefits of infection and to assess variation among symbionts. Logistic regression was used to analyze survival data. In these analyses, we determined the effects of rearing conditions or symbionts on the likelihood of aphid survival. All statistical analyses were performed in JMPIN® version 3.2.6. Aphids can develop into two adult forms—winged alates, or wingless morphs known as apterae. Life history parameters such as development time and fecundity are knovm to differ between these morphs, with alates exhibiting reduced offspring production and prolonged development (e.g. Dixon 1998). In our experiments, aphids typically developed into apterae, though alates were occasionally produced. Since different proportions of alates:apterae among our lines could result in substantial differences in fitness parameters, analyses on development time and fecundity were performed separately for these morphs. We also performed these analyses on pooled samples of alates and apterae. Results from pooled and separate analyses were qualitatively similar in most comparisons. Because of this, we present results from pooled analyses in this paper, citing the separate analyses only when these results produced different outcomes. Results R-type experiment 130 Exposure to heat shock at both the day 2 and day 6 stages had drastic effects on aphid performance. Compared to control aphids, heat shocked individuals suffered from drastic reductions in survival and fecundity and from prolonged development times. Many aphids were sterilized by heat shock, and some gave birth to mal-formed offspring whose limbs were stuck to the thorax and abdomen. These aphids died at very young ages and were not included in offspring counts. Though they performed poorly when exposed to heat shock on day 2, aphids infected with R-type isolates were substantially fitter than uninfected, genetically identical aphids under this treatment. R-type infection was associated with a higher odds of survival, with advantages ranging from 19 to 30% (Table 1). Similarly, R-infected aphids reached adulthood significantly faster than uninfected aphids, according to Wilcoxon rank-sum tests (Table 2). Uninfected aphids reached maturity at an average time of 13.3 days. R-infected lines matured, on average, at 11.0 (R'^), 11.6 (R'^^) and 12.4 (R^') days—demonstrating a 1-2 day acceleration compared uninfected individuals of the same clone. There were significant differences among the R-infected lines, with R'^ reaching adulthood faster than both R^"^ (p<0.0001) and R^' (p<0.0001), and R^^maturing faster than R^ (p=0.0002). Fecundity of R-infected aphids heat shocked at day 2 far exceeded that of uninfected^, pisum (Figure lb). To begin with, uninfected aphids were more frequently sterilized than their R-infected counterparts (data not shown). In addition, whereas uninfected aphids produced an average of 8.9 offspring per female, R-infected lines yielded averages of 19.6 (R^'), 20.3 (R^^), and 29.7 (R'^^) offspring. Again, non- 131 parametric Wilcoxon tests were used to compare lineages, and each R-infected line was significantly more fecund than their secondary-free counterparts (Figure lb). Analyses revealed significant differences between R-isolates, with R"^ producing more offspring than R^' and R^^ (R^ vs. R^': p=0.0034; R^ vs. R^^: p=0.0012). The presence of the R-type symbionts in aphids exposed to heat shock on day 6 was less beneficial, in comparison. Fecundities of these aphids were reduced compared to those heat shocked on day 2. Average offspring numbers were 4.6 (N), 8.6 (R^^), 2.8 (R"^^) and 11.3 (R^) (Figure Ic). The differences in fecundity between clonal lines were not significant under these conditions. Overall, sterilization was more common within R- infected lines in this treatment when compared to the day 2 heat shock treatment. But again, uninfected aphids were sterilized more often than R-infected aphids, with 57% of secondary-free aphids failing to produce any offspring. Generation times were shorter under later heat shock exposure, ranging from averages of 10.1 to 10.5 days (Table 2). Only R^' and R'^ aphids reached adulthood significantly faster than uninfected aphids (Wilcoxon rank-sum test, p<0.0001 for R'^ vs. N and p=0.0375 for R^ vs. N); development time of the R^^ line was indistinguishable from that of uninfected individuals (p = 0.8524). Again, there were significant differences among infected lines, with R'^ reaching adulthood faster than both R^' (p=0.0003) and R'^^ (p=0.0329). The developmental advantages of R'^ and R^' over uninfected aphids were each less than 0.5 days, in contrast to the 1-2 day advantages under the day 2 heat shock treatment. Though survival did not differ when comparing all aphids exposed to heat shock as later versus younger instars (data not shown), we observed a reversal in the previous benefits conferred with R-infection. Specifically, the odds ratio of survival for R-infected lines was reduced by 25-37% compared to uninfected aphids, revealing a cost to infection when heat shock was performed on older juveniles (Table 1). This reversal in symbiont effects was due to elevated survival of uninfected aphids when heat shocked at a later (vs. earlier) stage combined with a slight decrease in survival for R-infected aphids heat shocked at later (vs. earlier) times (data not shown). The highest levels of fecundity and survival, and the fastest development times were obtained for aphids reared at a constant 18°C throughout their lifespan (Figure la; Table 1; Table 2). Under these conditions, there were no significant effects of R-type symbionts on fecundity or survival. However, there was a slight, statistically significant developmental acceleration within symbiont-bearing lines (Table 2). R-infected aphids reached adulthood at 9.6-9.8 days, on average, compared to 10.0 days for uninfected aphids. The magnitudes of these differences (2-4%) were comparable to those observed for the day 6 heat shock treatment, but were substantially less than those under the day 2 treatment (7-18%). T- and U-type experiments To determine the generality of the heat tolerance effect, we performed similar experiments with additional symbionts, comparing the fitness of J. pisum infected with the T- or U-type symbionts to that of secondary-free insects. Again, heat shock on day 2 133 had drastically negative effects, causing substantial mortality, reduced fecundity, and prolonged development compared to the control treatment (Figure le). The effects of the T- and U-type symbionts on fitness were generally less pronounced than those of R-type isolates. However, both had significant effects on survival under the heat shock treatment (Table 1). Uninfected aphids were 24% more likely to survive to adulthood compared to U-type harboring aphids after heat shock exposure on day 2 (logistic regression, p=0.0439). In contrast, the T-type improved survival of heat shocked aphids by 34%, relative to uninfected aphids (p=0.0111), on par with the advantages conferred by R-type symbionts subjected to the same treatment. The significance (but not the trend) of this difference was dependent upon the inclusion of one of fourteen treated T-type cultures in which 20/20 individuals survived to adulthood. Neither symbiont caused significant changes in survival, development time, or fecundity when reared at 18°C (Figure Id, Table 1, Table 3). Under the heat shock treatment there was no difference in development time between uninfected and T-type infected A. pisum, which reached adulthood at average times of 13.2 and 13.6 days, respectively (Table 3). U-type infected aphids exposed to heat shock reached adulthood at 14.2 days, and the difference between this line and the uninfected lines bordered on statistical significance (p=0.0818 using a Wilcoxon test). Slower U-type development was due, in part, to a higher production of alates within this line compared to uninfected aphids—8/55 uninfected adults were alates compared to 18/28 for U-type lines. However, there remained a trend toward slower U-type development when only apterae were considered. 134 Interestingly, the trends in developmental times differed among alates and wingless apterae when comparing T- and U-type lines with the uninfected line. In both cases, alates of symbiont-rnfected lines tended to reach adulthood faster than uninfected alates, whereas uninfected apterae tended to reach adulthood faster than infected apterae. However, only the difference between T-infected and uninfected alates was significant (p=0.0027). Sterilization was also common among heat-shock-treated aphids in this experiment. Out of 44 single female aphids that survived for the first 8 days of adulthood, 24 produced no offspring. We observed no significant differences in the incidence of sterility when comparing across A. pisum lines. Average offspring numbers for these lines were 1.4 (U), 3.7 (N) and 6.1 (T) (Figure le). These values were statistically indistinguishable, though our analyses were limited by small sample sizes imposed by high rates of mortality. Model of symbiont frequencies To demonstrate how the observed fitness differences could lead to rapid changes in symbiont frequencies within populations, we modeled the expected change in R-type frequency, under viability selection. Survival data from the day 2 heat shock treatment were used to calculate relative fitness values for each lineage. Three separate analyses were run, each involving population containing R-Lnfected and uninfected individuals. R-infected aphids, which had higher survival under the conditions modeled, underwent a rapid increase in frequency (Figure 2). The R'^ line reached 50% frequency by generation 29, while R^' and R'^^ surpassed 50% in the 25*'' and 23'^'' generations. 135 respectively. R-type frequencies leveled off between generations 30 and 40, with values asymptotically approaching 94.29% (R^), 94.87% (R^'), and 95.12% (R'^^). A similar analysis, which assumed a 1% rate of transmission failure (vs. 2%) gave comparable results. And, of course, with 100% transmission efficiency, these symbionts would be expected to sweep to fixation. These calculations yield a conservative estimate of the rate of replacement of uninfected with the R-type infected lines because fecundity and developmental time differences not considered in the model, would amplify the fitness differences between these aphids. Discussion Our results reveal substantial costs and benefits of secondary symbiont infection that vary across temperature treatments and with the timing of heat shock exposure. All three R-type symbionts conferred substantial benefits on^. pisum hosts exposed to a heat shock treatment as young juveniles (day 2 treatment). The T-type symbiont induced a less-pronounced benefit under these conditions, improving only the survival of infected hosts. In contrast, the U-type symbiont led to a substantial reduction in survival under day 2 heat shock. Based on the results from our model, survival differences, alone, would lead to a rapid increase of symbiont frequency of R-type symbionts within host populations. T-type symbionts, though not modeled, would similarly be expected to increase in frequency under these conditions given the benefits observed here. In contrast, decreased survival of U-type infected aphids after heat shock would reduce U- type frequencies in A. pisum populations. 136 When aphids were heat-shocked as older juveniles (day 6), the outcomes of these symbioses were substantially different, as R-type-infected aphids had lower survival than those without secondary symbionts. This reduction ranged from 25-37%, comparable to the magnitude of the survival increase conferred upon R-infected aphids heat-shocked as young nymphs. Despite this survival cost, there was evidence for slight benefits of the R- type when considering development times and a trend toward higher fecundity for two of three isolates. Whereas R-, T-, and U-type infection led to strong costs and benefits in heat- shocked aphids, these microbes had minimal effects on hosts reared constantly at a permissive temperature of 18°C. Survival, fecundities, and generation times of T- and U- infected aphids were statistically indistinguishable from those of genetically identical uninfected lines. R-infected lines differed from their uninfected counterparts only when considering generation time—^two out of three such lines reached adulthood significantly faster than their iminfected, clonal counterparts. Montllor et al. (2002) foimd evidence for a benefit of the R-type under two heat shock treatments, which were performed at the same developmental stages as in our experiments. In their study, R-infected aphids had higher fecundity than genetically identical secondary-free aphids, except when they harbored an additional, rickettsial symbiont known as PAR. The authors also reported that the R-type had negligible effects on survival after heat shock, though they did not measure survival to adulthood. It is, thus, unclear whether this discrepancy is the result of differing protocols or, perhaps, genetic differences between aphids or R-type symbionts. Combined, our results suggest 137 that the R-type is not universally beneficial under heat shock. Instead, the direction and magnitude of symbiont effects on fitness depend on several factors, including the timing of heat exposure, the presence of additional symbionts, and symbiont genomes, as discussed below. Our findings on the R-type reveal that even closely related symbionts may vary in their effects on host fitness. For example, the R-type isolate derived from the Tucson- collected clone of^. pisum conferred significantly higher fecundity and faster development compared to each of the other R-isolates under the day 2 heat shock treatment. This isolate also appeared the most beneficial under the later heat shock treatment. It is worth noting that this symbiont was obtained from a region of the United States (Tucson, Arizona) that experiences temperatures far in excess of the locales from where the other two isolates originated (New York and Wisconsin). If there is, indeed, genetic variation among R-type isolates in their capacity to confer heat tolerance (as suggested here), then one might expect natural selection to yield such a pattern, whereby isolates from the hottest regions confer the largest benefits under high temperatures. Some data suggest that the R-type symbiont plays a role in heat tolerance in field populations (Montllor et al. 2002). A. pisum were sampled from three California locales in the winter or spring, and again in the summer, when temperatures were considerably higher. The R-type increased in frequency in 2/3 of these regions. However, R-type frequencies were not correlated with temperature when considered across these sites, suggesting the action of other environmental or stochastic forces. Laboratory experiments on^l. pisum have revealed a role for R-type and T-type symbionts in defense 138 against parasitoids (Oliver et al. 2003, Ferrari et al. 2004) and have suggested a role for the U-type in defense against fungi (Ferrari et al. 2004). The U-type has also been implicated in host plant utilization, according to the results of several field studies (Simon et al. 2003, Leonardo and Muiru 2003 , Tsuchida et al. 2002) and a recent laboratory analysis, which revealed a role in the utilization of clover (Tsuchida et a/.2004, but see Leonardo 2004). Therefore, future attempts to measure the natural effects of temperature on symbiont prevalence should consider these additional selective factors. The results of our study and those of Montllor et al. (2002) contribute to a growing body of literature which has revealed the importance of temperature in governing the outcomes of symbiotic interactions. And though heat tolerance is not a widely reported consequence of symbiosis, we are aware of only a few studies that have examined this possibility. In one affirmative case, the presence of a bacterial symbiont was shown to improve survival of heat shocked paramecia (Hori and Fujishima 2003). Symbionts apparently caused increased expression of host heat shock genes, revealing a possible mechanism behind the benefit. A more commonly documented consequence of high temperatures on symbiosis is one of symbiont elimination. Heat has been implicated in the global phenomenon of coral bleaching, which stems from the expulsion of dinoflagellate symbionts from their cnidarian hosts (Warner et al. 1999, Gates et al 1992). In addition, heat is known to cause symbiont mortality in several insect species (e.g. Buchner 1965, Stouthamer 1997). This phenomenon has been documented within yi. pisum, whose essential primary symbionts (Buchnera) were eliminated after a heat shock treatment resembling our own 139 (Ohtaka and Ishikawa 1991). Similarly, Montllor and colleagues (2002) noted a decrease in the number of specialized Buchnera-housmg cells (bacteriocytes) in adults that survived heat treatment, suggesting reduced Buchnera survival. They also noted that the presence of the R-type symbiont led to increased bacteriocyte persistence, thus presumably Buchnera survival, providing a possible mechanism for the R-mediated heat tolerance. High temperatures have also been noted to reduce the expression of symbiont- induced phenotypes in insects. For example, induced effects of parthenogenesis and cytoplasmic incompatibility can both be attenuated by exposing insects to heat (Stouthamer 1997, Feder et al. 1999, Hoffinarm et al. 1986, 1990, Arakaki et al. 2001). Combined, these previous findings reveal the importance of considering temperature effects when dissecting the interactions between eukaryotes and bacteria. Given the current trend of global warming, we expect that future studies on these phenomena will be of immediate relevance in understanding the consequences for symbiotic organisms in nature. 140 References cited Arakaki, N., T. Miyoshi, and H. Noda. 2001. Wolbachia-rnQdvaXedi parthenogenesis in the predatory thrips Franklinothrips vespiformis (Thysanoptera; Insecta). Proc. R. Soc. Lond. Ser. B 268:1011-1016. Asin, L., and X. Pons. 2001. Effect of high temperature on the growth and reproduction of com aphids (Homoptera: Aphididae) and implications for their population dynamics on the northeastern Iberian peninsula. Environ. Entomol. 30:1127-1134. Bull, J.J., I.J. Molineux, W.R. Rice. 1991. Selection of benevolence in a host-parasite system. Evolution 45:875-882. Bernays, E.A., and B.A. Klein. 2002. Quantifying the symbiont contribution to essential amino acids in aphids: the importance of tryptophan for Uroleucon ambrosiae. Physiol. Entomol. 27:275-284. Bracho, A.M., D. Martinez-Torres, A. Moya, and A. Latorre. 1995. Discovery and molecular-characterization of a plasmid localized in Buchnera sp. bacterial endosymbiont of the aphid Rhopalosiphum padi. J. Mol. Evol. 41:67-73. Buchner, P. 1965. Endosymbiosis of animals with plant microorganisms. Interscience, New York, NY. Chen, D.-Q., B.C. Campbell, and A.H. Purcell. 1996. A new rickettsia from a herbivorous insect, the pea aphid Acyrthosiphon pisum (Harris). Curr. Microbiol. 33:123-128. Chen, D.-Q., and A.H. Purcell. 1997. Occurrence and transmission of facultative endosymbionts in aphids. Curr. Microbiol. 34:220-225. Chen, D.-Q., C.B. MontUor, and A.H. PurcelL 2000. Fitness effects of two facultative endosymbiotic bacteria on the pea aphid Acyrthosiphon pisum, and the blue alfalfa aphid, yl. kondoi. Entomol. Exp. Appl. 95:315-323. Darby, A.C., L.M. Birkle, S.L. Turner, and A.E. Douglas. 2001. An aphid-borne bacterium allied to the secondary symbionts of whitefly. FEMS Microbiol. Ecol. 36:43-50. Darby, A.C., and A.E. Douglas. 2003. Elucidation of the transmission patterns of an insect-borne bacterium. Appl. Environ. Microbiol. 69:4403-4407. Dixon, A.F.G. 1998. Aphid ecology, 2"^ edition. London, Chapman & Hall. 141 Douglas, A,E. 1988. Sulfate utilization in an aphid symbiosis. Insect Biochem. 18:599- 605. Douglas, A.E., and W.A. Prosser. 1992. Synthesis of the essential amino acid tryptophan in the pea aphid {Acyrthosiphon pisum) symbiosis. J. Insect Physiol. 38:565-568. Douglas, A-E., L.B. Minto, and T.L. Wilkinson. 2001. Quantifying nutrient production by the microbial symbionts in an aphid. J. Exp. Biol. 204:349-358. Ewald, P. 1983. Transmission modes and the evolution of the parasitism-mutualism continuum. Ann. N. Y. Acad. Sci. 503:295-306. Feder, M.E., T.L. Karr, W. Yang, J.M. Hoekstra, and A.C. James. 1999. Interaction of Drosophila and its endosymbiont Wolbachia: Natural heat shock and the overcoming of sexual incompatibility. Am. Zool. 39:363-370. Ferrari, J., A.C. Darby, T.J. Daniell, H.C.J. Godfray, and A.E. Douglas. 2004. Linking the bacterial community in pea aphids with host-plant use and natural enemy resistance. Ecol. Entomol. 29:60-65. Fukatsu, T. 2001a. Secondary intracellular symbiotic bacteria in aphids of the genus Yamatocallis. (Homoptera:Aphididae:Drepanosiphinae). Appl. Environ. Microbiol. 67:5315-5320. Fukatsu, T. 2001b. Spiroplasma symbiont of the pea aphid, Acyrthosiphon pisum (Insecta: Homoptera). Appl. Environ. Microbiol 67:1284-1291. Gates, R.D., G. Baghdasarian, and L. Muscatine. 1992. Temperature stress causes host cell detachment in symbiotic cnidarians: Implications for coral bleaching. Biol. Bull. 182:324-332. Haynes, S., A.C. Darby, T.J. Daniell, G. Webster, F.J.F, van Veen, H.C.J. Godfray, J.L. Prosser, and A.E. Douglas. 2003. Diversity of bacteria associated with natural aphid populations. Appl. Environ. Microbiol. 69:7216-7223. Hoffmann, A.A., M. Turelli, and G.M. Simmons. 1986. Unidirectional incompatibility between populations of Drosophila simulans. Evolution 40:692-701. Hoffmann, A.A., M. Turelli, and L.G. Harshman. 1990. Factors affecting the distribution of cytoplasmic incompatibility in Drosophila simulans. Genetics 126:933-948. 142 Hon, M., and M. Fujishima. 2003. The endosymbiotic bacterium Holospora obtusa enhances heat-shock gene expression of the host Paramecium caudatum. J. Eukaiyot. Microbiol. 50:293-298. Lai, C.-Y., L. Baumann, and P. Baumann. 1994. Amplification oftrpEG; adaptation of Buchnera aphidicola to an endosymbiotic association with aphids. Proc. Natl. Acad. Sci. USA 91:3819-3823. Leonardo, T.E. 2004. Removal of a specialization-associated symbiont does not affect aphid fitness. Ecol. Letters 7:461-468. Ma, C.-S., B. Hau, and H.-M, Poehling. 2004. The effects of heat stress on the survival of the rose grain aphid, Metopolophium dirhodum (Hemiptera: Aphididae). Eur. J. Entomol. 101:327-331. May, R.M. and R.M. Anderson. 1982. Coevolution of hosts and parasites. Parasitology 85:411-426. Montllor, C.B., A. Maxmen, and A.H. FurcelL 2000. Facultative bacterial endosymbionts benefit pea aphids Acyrthosiphonpisum under heat stress. Ecol. Entomol. 27:189-195. Munson, M.A., P. Baumann, M.A. Clark, L. Baumann, N.A. Moran, D.J. Voegtlin, and B.C. Campbell. 1991a. Evidence for the establishment of aphid-eubacterial endosymbiosis in an ancestor of four aphid families. J. Bacteriol. 173:6321-6324. Munson, M.A., P. Baumann, and M.G. Kinsey. 1991b. Buchnera gen. nov. and Buchnera aphidicola sp. nov., a taxon consisting of the mycetocyte-associated, primary endosymbionts of aphids. Int. J. Syst. Microbiol. 41:566-568. Munson, M.A., and P. Baumann. 1993. Molecular cloning and nucleotide sequence of a putative trpDC(F)BA operon in Buchnera aphidicola (endosymbiont of the aphids Schizaphis graminum). J. Bacteriol. 175:6426-6432. Ohtaka, C., and H. Ishikawa. 1991. Effects of heat treatment on the symbiotic system of an aphid mycetocyte. Symbiosis 11:19-30. Oliver, K.M., J.A. Russell, N.A. Moran, and M.S. Hunter. 2003. Facultative bacteria in aphids confer resistance to parasitic wasps. Proc. Natl. Acad. Sci. USA 100:1803-1807. Russell, J.A., A. Latorre, B. Sabater-Munoz, A. Moya, and N.A. Moran. 2003. Side stepping secondary symbionts: widespread horizontal transfer across and beyond the Aphidoidea. Mol. Ecol. 12:1061-1075. 143 Sandstrom, J., and N.A. Moran. 1999. How nutritionally imbalanced is phloem sap for aphids? Entomol. Exp. Appl. 91:203-210. Sandstrom, J., and N.A. Moran. 2001. Amino acid budgets in three aphid species using the same host plant. Physiol. Entomol. 26:202-211. Sandstrom, J.P., J.A. Russell, J.P. White, and N.A. Moran. 2001. Independent origins and horizontal transfer of bacterial symbionts of aphids. Mol. EcoL 10:217-228. Shigenobu, S., H. Watanabe, M. Hattori, Y. Sasaki, and H. Ishikawa. 2000. Genome sequence of the endocellular bacterial symbiont of aphids Buchnera sp. APS. Nature 407:81-86. Simon, J.C., S. Carre, M. Boutin, N. Prunier-Leterme, B. Sabater-Munoz, A. Latorre, and R. Boumoville. 2003. Host-based divergence in populations of the pea aphid: insights from nuclear markers and the prevalence of facultative symbionts. Proc. R. Soc. Lond. Ser. B 270:1703-1712. Stouthamer, R. 1997. JFo/6ac/zza-induced parthenogenesis. In; O'Neill, S., A.A. Hof&nann, and J.H. Werren (eds). Influential Passengers. Oxford University Press: New York. Tsuchida, T., R. Koga, H. Shibao, T. Matsumoto, and T. Fukatsu. 2002. Diversity and geographic distribution of secondary endosymbiotic bacteria in natural populations of the pea aphid, Acyrthosiphon pisum. Mol. Ecol. 11:2123-2135. Tsuchida, T., R. Koga, and T. Fukatsu. 2004. Host plant specialization governed by facultative symbiont. Science 303:1989. Turak, E., R. Talent, P. Sunnucks, and D.F. Hales. 1998. Different responses to temperature in three closely-related sympatric cereal aphids. Entomol. Exp. Appl. 86:49-58. Unterman, B.M., P. Baumann, and D.L. McLean. 1989. Pea aphid symbiont relationships established by analysis of 168 ribosomal-RNA's. J. Bacteriol. 171:2970-2974. van Ham, R.C.H.J., A. Moya, and A. Latorre. 1997. Putative evolutionary origin of plasmids carrying the genes involved in leucine biosynthesis in Buchnera aphidicola (endosymbiont of aphids). J. Bacteriol. 179:4768-4777. Wang, J.-J., and J.H. Tsai. 2000. Effect of temperature on the biology oi Aphid spiraecola (Hompotera: Aphididae). Ann. Entomol. Soc. Am. 93:874-883. 144 Wang, J.-J., and J.H. Tsai. 2001. Development, survival, and reproduction of black citrus aphid, Toxoptera aurantii (Hemiptera: Aphididae), as a function of temperature. Bull. Entomol. Res. 91:477-487. Warner, M.E., W.K. Fitt, and G.W. Schmidt. 1999. Damage to photosystem II in symbiotic dinoflagellates: A determinant of coral bleaching. Proc. Natl. Acad. Sci. USA 96:8007-8012. Wicker, C., and P. Nardon. 1982. Development responses of symbiotic and aposymbiotic weevils Sitophilus oryzae L. (Coleoptera, Curculionidae) to a diet supplemented with aromatic amino acids. J. Insect Physiol. 28:1021-1024. 145 Acknowledgements We would like to thank Kim Hammond, Michelle Hoffman, Helen Dunbar and Phat Tran for technical assistance. In addition, we thank Jeff Good, Kerry Oliver, and Molly Hunter for helpful comments on the manuscript. This research was supported by NSF , and J. Russell received funding from the NSF IGERT program in genomics at the University of Arizona. T reatment Regression pr P2' P3* equation® control*^ Y = -2.60 - O.SSR"^' - p=0.0184; p=0.3199; p=0.6793; 0.25R'^^ - 0.12R'^^ 95% CI: 95% CI: 95% CI; -1.06 to-0.10 -0.74 to 0.24 -0.44 to 0.67 day 2 heat Y = -0.03 + 0.30R"" + p<0.0001; p<0.0001; p=0.0099; shock®^ 0.29R'^'^ +0.19R'^^ 95% CI: 95% CI: 95% CI: 0.15 to 0.45 0.15 to 0.43 0.04 to 0.33 day 6 heat Y = 0.90 - 0.28R'^' - p=0.0005; p<0.0001; p<0.0001; shock'^ 0.46R^^ - 0.44R'^^ 95% CI: 95% CI: 95% CI: -0.44 to-0.12 -0.61 to-0.30 -0.60 to -0.27 control"^ Y = -2.92-0.01T + p=0.9652; p=0.1544; n.a. 0.58U 95% CI: 95% CI; -0.58 to 0.56 -0.22 to 1.39 day 2 heat Y = 0.77 + 0.29T - p = 0.0111; p = 0.0439; n.a. shock^'^ 0.27U 95% CI: 95% CI: 0.07 to 0.52 -0.53 to -0.01 Table 1 Logistic regression analyses on aphid survival. ^Regression equation (for R-type experiments) is Y = po + PiR^ + P2R'^^ + PsR'^ or (for T/U-type experiments) Y = po + PiT+ P2U. ^Statistics reveal whether p parameter estimates (representing the relative difference survival compared to uninfected aphids) differed significantly from zero. 147 Treatment Statistics Aphic Line N RWI RNV RAA control number of aphids 200(17) 114(13) 216(17) 186 (20) (number of treated cultures) mean; median 10.0; 10 9.8; 10 9.7; 10 9.6; 10 days to adulthood Wilcoxon rank sum n.a. p=0.0155 p=0.0002 p<0.0001 test® day 2 heat number of aphids 116(20) 166(18) 182 (21) 177(22) shock (number of treated cultures) mean; median 13.3; 13 12.4; 12 11.6; 11 11.0; 11 days to adulthood Wilcoxon rank sum n.a. p=0.0003 p<0.0001 p<0.0001 test® day 6 heat number of aphids 201 (16) 156(14) 155(18) 143 (20) shock (number of treated cultures) mean; median 10.5; 10 10.3; 10 10.5; 10 10.1; 10 days to adulthood Wilcoxon rank sum n.a. p=0.0375 p=0.4278 p<0.0001 test® Table2 Time to adulthood for uninfected and R-type infected^, pisum. ®Pairwise Wilcoxon rank sum tests were performed, comparing development times for each R-infected line (R^', R^^, and R^) to that from the uninfected line (N) under each temperature treatment. 148 Treatment Statistics Aphid Line N T U control number of aphids 76(10) 65(11) 71 (10) (number of treated cultures) mean; median 10.1; 10 10.2; 10 10.0; 10 (days to adulthood) Wilcoxon rank sum n.a. p=0.1231 p=0.7221 test® day 2 heat number of aphids 55(11) 66 (9) 28 (8) shock (number of treated cultures) mean; median 13.2; 13 13.6; 13 14.17; 14 (days to adulthood) Wilcoxon rank sum n.a. p=0.5960 p=0.0818 test® Table 3 Time to advilthood for uninfected and T- or U-type infected A. pisum. ®Pairwise Wilcoxon rank sum tests were performed, comparing development times for each symbiont-infected line (T and U) to that from the uninfected line (N) under each temperature treatment. 149 FIGURE LEGENDS Figure 1 Distributions of fecundity, for single females varying in their symbiont infection status. Offspring production of single aphid females is represented using frequency histograms. Means, medians, and sample sizes are presented for each line under each of these treatments. Also, p-values are presented from Wilcoxon rank sum tests. Each p-value represents the comparison between fecundities for the given symbiont-infected line and the uninfected line, a) Results from control treatment for the R-type experiment, b) Results from day 2 heat shock treatment for the R-type experiment, c) Results from day 6 heat shock treatment for the R-type experiment, d) Results from the control treatment for the T- and U-type experiment, e) Results from the day 2 heat shock treatment for the T- and U-type experiment. Note, sample sizes presented represent the number of aphids measured and, in parentheses, the number of treated cultures from which these aphids were derived. Figure 2 Change in R-type frequency under viability selection. Using the weighted averages of survival under the day 2 heat shock treatment, we calculated relative fitness values. Based on these fitness values, and allowing for a 2% rate of transmission failure (see text), we plotted the change in R-type frequency (starting from 1/100,000) over time in generations (x-axis) in three separate analyses. Each analysis compared the frequency of a single R-infected line (R^*, R'^^, andR'^) to that of the uninfected line. 150 Figure la N 60- R2BB 50- n = 51 (17) >1 <> n = 40(13) median = 44 c 40- median = 40 mean = 41.1 3 mean = 38.7 p=0.2149 cr 30- £ u. 20- a 10- 0- 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 0 5 10 IS 20 25 30 35 40 45 50 55 60 65 70 # offspring # offspring 60 RTUC R' -2-1 50 n = 56 (20) n = 49(16) median = 42 c0) 40 median = 44 mean = 38.3 3 mean = 38.9 p=0.2457 p=0.9862 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 5 10 IS 20 25 30 35 40 45 50 55 60 65 70 # offepring # offspring 151 Figure lb 601 N r2BB 50 n = 63 (20) n = 66(18) >> median = 0 median = 12 c 40 30> mean = 8.9 mean = 19.6 0)CT 30 p=0.0002 20 10 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 5 10 15 20 25 30 35 40 45 50 55 60 65 70 # offspring # offspring 60- 60- RTU R9-2-I 50- 50- >» n = 71 (22) >< n = 79(21) u c 40- median = 35 c 40- median = 19 o a 3 mean = 29.7 o mean = 20.3 ff p<0.0001 w p<0.0001 £ 30- 30- IL Li. 20- 20" 10- 10- 0- 0- 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 5 10 15 20 25 30 35 40 45 50 55 60 65 70 # offspring # offspring 152 Figure Ic N 0 =68(16) n = 46 (10) median = 0 median = 1 mean = 4.6 mean = 8.6 p=0.0677 20 lOl:® 5 10 15 20 25 30 35 40 45 50 55 60 65 70 5 10 15 20 25 30 35 40 45 50 55 60 65 70 # offspring # offspring 60- RTU R9-2-1 50" 50 >v n=48(16) >. n=46(16) O c 40- median = 0 c 40" median = 1 <0 3 mean = 11.4 3 mean = 2.8 IT p=0.1655 CT p=0.3488 !» 30- O 30- u. 20- 20- 10- 10 o4 0 5 10 1S 20 25 30 35 40 45 50 55 60 65 70 5 10 15 20 25 30 35 40 45 50 55 60 65 70 # offspring # offspring 153 Figure Id 40 N n = 38 (10) >.30 n = 34(ll) median = 32 c median = 27 mean - 29.0 o mean = 26.9 §• 20 p=0.5722 10 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 5 10 15 20 25 30 35 40 45 50 55 60 65 70 # offspring # offspring U n = 40(10) median = 28 mean = 28.3 p=0.7076 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 # offspring 154 Figure le 401 N o>> 30- n = 34 (II) n = 29 (9) c median - 0 median = 0 3 mean = 3.7 mean = 6.2 g" 20 p=0.1012 10 TT-i^ 5 10 15 20 25 30 35 40 45 50 55 60 65 70 5 10 15 20 25 30 35 40 45 50 55 60 65 70 # offspring # offspring V n = 19 (8) median - 0 mean = 1.4 p=0.8893 1- - •'-f f' 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 # offspring 155 Figure 2 0.9 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 10 20 40 50 60