WOLBACHIA DISTRIBUTION AND REPRODUCTIVE EFFECTS IN POPULATIONS OF A
NEOTROPICAL BEETLE, CHELYMORPHA ALTERNANS BOH. (CHRYSOMELIDAE)
by
GWEN P. KELLER
(Under the Direction of KENNETH G. ROSS)
ABSTRACT
Wolbachia is a group of maternally-inherited endocellular bacteria that manipulate arthropod reproduction to favor infected females. While strains of these bacteria have been found in all insect orders where at least 17% of all insect species infected, few studies have investigated the distribution and effects of Wolbachia in natural populations. Here I present a review of Wolbachia research in the context of host biology (chapter 2) and I describe original research (chapter 3) that details the distribution and effects of Wolbachia in a Neotropical beetle,
Chelymorpha alternans, that is infected with two Wolbachia strains, wCalt1 and wCalt2. In chapter 3 I describe the two strains based on three Wolbachia genes, I track their distribution across space and time in host populations throughout the Panamanian isthmus, I determine strain effects on host reproduction, and I show a correlation between environmental effects and the ongoing loss of one strain. This is the first study to show the loss of a Wolbachia strain from natural populations and I discuss the possible factors responsible for the loss.
INDEX WORDS: Wolbachia, Chelymorpha, cytoplasmic incompatibility, strain loss WOLBACHIA DISTRIBUTION AND REPRODUCTIVE EFFECTS IN POPULATIONS OF A
NEOTROPICAL BEETLE, CHELYMORPHA ALTERNANS
by
GWEN P. KELLER
University of Georgia, 2005
A Dissertation Submitted to the Graduate Faculty of The University of Georgia in Partial
Fulfillment of the Requirements for the Degree
DOCTOR OF PHILOSOPHY
ATHENS, GEORGIA
2005 © 2005
GWEN P. KELLER
All Rights Reserved WOLBACHIA DISTRIBUTION AND REPRODUCTIVE EFFECTS IN POPULATIONS OF A
NEOTROPICAL BEETLE, CHELYMORPHA ALTERNANS
by
GWEN P. KELLER
Major Professor: Kenneth G. Ross
Committee: Patty A. Gowaty Joe V. McHugh Donald Champagne Wyatt W. Anderson
Electronic Version Approved:
Maureen Grasso Dean of the Graduate School The University of Georgia May 2005 DEDICATION
With great affection I dedicate this work to Lois Edgell and Joaquin Recuero, two loyal friends whose cheerfulness and humor buoyed me through the trials of pursuing this degree.
Your faith in me, and constant friendship has made all the difference in my life. Muchisimas gracias mis queridos amigos!!
iv ACKNOWLEDGEMENTS
I wish to thank my advisors, Ken Ross and Patty Gowaty for always being available, encouraging and marvelous! Cheers to Jack Werren for his willingness to share his knowledge of all things Wolbachia, and for his amazing grant writing abilities that enabled much of this research. Special thanks to Don Windsor for introducing me to the world of leaf beetles, especially Chelymorpha, and to Jenniffer Saucedo whose extraordinary beetle husbandry contributed greatly to this work. I also fondly thank the kind folks at the Smithsonian Institution in Washington, D.C. and the Smithsonian Tropical Research Institute in Panama for logistical and financial support.
v TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS ...... v
LIST OF TABLES...... vii
LIST OF FIGURES ...... viii
CHAPTER
1 INTRODUCTION TO THE DISSERTATION...... 1
2 INTRODUCTION AND LITERATURE REVIEW OF THE BIOLOGY OF
WOLBACHIA IN ARTHROPODS ...... 4
INTRODUCTION...... 5
EARLY HISTORY ...... 6
WOLBACHIA MANIPULATION OF HOST REPRODUCTION...... 8
REFERENCES ...... 22
3 REPRODUCTIVE EFFECTS AND GEOGRAPHIC DISTRIBUTIONS OF TWO
WOLBACHIA STRAINS INFECTING THE NEOTROPICAL BEETLE,
CHELYMORPHA ALTERNANS BOH. (CHRYSOMELIDAE)...... 28
INTRODUCTION...... 29
MATERIALS AND METHODS...... 31
RESULTS...... 39
vi CHAPTER
3 DISCUSSION...... 46
REFERENCES ...... 56
4 CONCLUSIONS...... 71
vii LIST OF TABLES
Page
Table 1: Collection and site information...... 60
Table 2: Egg hatch rates from crosses ...... 62
Table 3: Mitochondrial CO1 haplotype and nucleotide diversity estimates...... 63
Table 4: Parameters for invasion model ...... 64
viii LIST OF FIGURES
Page
Figure 1: Chelymorpha alternans populations sampled in Panama ...... 65
Figure 2: Wolbachia infection status and C. alternans haplotype distributions in Panama
populations ...... 66
Figure 3: Frequency of the wCalt2 strain in populations with varying length of dry season in 19
populations ...... 67
Figure 4: Egg hatch rates for females ...... 68
Figure 5: The effects of double infected males from Remedios (Rem), Gamboa (Gam) and Cana
(Can) on egg hatch rates...... 69
Figure 6: Statistical parsimony network of C. alternans haplotypes...... 70
ix CHAPTER 1 - INTRODUCTION
1 INTRODUCTION Increasing interest in reproductive parasites of arthropods over the past 10 years has led to the discovery of several species of maternally-inherited endosymbiotic bacteria that manipulate host sex ratios to favor infected female hosts. Among these bacteria, Wolbachia is the most common and shows a broad range of host effects including parthenogenesis, male feminization, male-killing, and cytoplasmic incompatibility (O’Neill et al. 1997; Werren 1997;
Stouthamer et al. 1999).
Standard techniques of bacterial identification rely on morphology and nutritional require- ments in culture. However, endocellular bacteria, such as Wolbachia, have eluded culture because of their fastidious requirements. Wolbachia survive only within host cells, thus culturing Wolbachia requires culturing insect cells, which is difficult to do. Thus, culturing techniques for identification of Wolbachia are impractical considering that Wolbachia infect at least 17% of species in the class Insecta, a group that is estimated to contain 106 - 306 species.
The recent and rapidly evolving development of molecular genetic techniques such as clon- ing, polymerase chain reaction (PCR), and automated DNA sequencing has significantly advanced the study of fastidious microbes. These techniques allow researchers to quickly isolate and identify DNA of bacterial strains and thus avoid the time-consuming tasks involved in culturing fastidious bacteria. Using eubacterial DNA primers, strings of oligonucleotides that are conserved across eubacterial species, researchers isolated Wolbachia genes from insect tissues and created Wolbachia-specific primers for those genes. These primers specifically amplify
Wolbachia products. Because of their specificity, researchers are able to quickly screen host tissues for the presence of Wolbachia strains. By cloning Wolbachia products from PCR reactions and then sequencing them, it is possible to determine the number of strains present in a single host by determining the number of variable sequences present in the sample taken from
2 that host. To verify the identity of strains, additional genes should be sequenced because cryptic strains produced by recombination could be present but not obvious if they share the same DNA sequence at one locus. Likewise, to determine whether the Wolbachia genes represent true strains rather than DNA sequences that have been transferred to the host genome, it is best to sequence several Wolbachia genes as it is unlikely that the entire Wolbachia genome would be transferred to the host.
The development of molecular techniques for identifying Wolbachia strains has created a landslide of studies investigating the distribution, variation, and evolution of Wolbachia in insects. However, there is a paucity of studies investigating the effects of Wolbachia in natural insect populations. My research aims to shed light on the effects of Wolbachia in natural populations of Neotropical insects.
In the following chapters I present a review of Wolbachia-insect research to date and I present original research conducted with insects from Panama. I have examined one species,
Chelymorpha alternans in detail and show that two strains infecting this beetle, wCalt1 and wCalt2, cause different degrees of cytoplasmic incompatibility. I also show that the infection is not recent but invaded beetle populations about 100,000 years ago as a double infection but now one of the strains is disappearing. The last invasion reduced mitochondrial diversity which is slowly recovering. This study is important because it shows that there are more factors affecting insect evolution and ecology than previously thought. It appears that microbes have been intimately influencing insect reproductive biology for eons and must be considered as an important force in insect evolution.
3 CHAPTER 2 - INTRODUCTION AND LITERATURE REVIEW OF THE BIOLOGY OF
WOLBACHIA IN ARTHROPODS
4 INTRODUCTION It is widely recognized that symbiotic microorganisms play a fundamental role in the evolution and ecology of insects. Such mutualisms are usually maintained through vertical (maternal) transmission and may be obligate or facultative, parasitic or beneficial.
Several clades of bacteria and yeast have developed mutually obligate relationships with insects in exchange for secure housing in the form of specially modified host cells (bacteriocytes) and a constant supply of nourishment, the microbes provide essential nutrients, parasitoid resistance or reproductive functions for their hosts (1, 3, 16, 18, 22, 56, 58). However, other bacterial groups have evolved as obligate parasites; i.e. they depend on their host for survival yet they are not essential for the survival of their hosts. Some of these parasitic endosymbionts affect host fitness by manipulating host reproduction to enhance their own transmission (36, 33, 36, 37, 40, 87).
The discovery of one such reproductive parasite, Wolbachia pipientis (Rickettsiales: a- proteobacteria), has stimulated renewed interest in symbioses and insect reproductive anomalies
(4, 25, 26, 30, 38, 44, 46, 60, 65, 71, 76, 81, 83, 91, 98).
Wolbachia are obligate maternally inherited endocellular bacteria that infect reproductive tissues of arthropods and filarial nematodes. These fastidious microbes have been found in 15 -
76% of insect species (42, 59, 84, 85, 91, 93, 95), 35% of isopod species (9), 75% of filarial nematode species (78), and some mite species (13). As reproductive parasites, these bacteria manipulate insect host reproduction in a variety of ways to increase the frequency of infected female hosts and thereby increase their own transmission. Phenotypes of Wolbachia include parthenogenesis induction in haplodiploid wasps (36, 61, 72, 74-77), feminization induction in isopod crustaceans (9, 64-67), death of early male embryos, also known as "male-killing" in
Nymphalid butterfiles, Coccinelid beetles and Drosphilid flies (24, 25 37-39, 44), and
5 cytoplasmic incompatibility in many insect families found in several orders (6, 7, 10, 11, 30, 34,
35, 62, 73, 81-83, 87).
Because of the ubiquity of Wolbachia and its effects on host reproduction, this bacteria plays a significant role in insect ecology and evolution. The vast number of studies generated by interest in Wolbachia over the past decade is daunting. Here I include an overview of research on Wolbachia that infect insects, with a focus on their ecology and evolution.
EARLY HISTORY OF WOLBACHIA RESEARCH
FIRST INSIGHTS Early studies of reproductive failures among strains of Culex pipiens mosquitoes led researchers to identify a Rickettsial bacteria, later dubbed Wolbachia pipientis, as the causative agent of egg mortality (2, 28, 32, 50, 51, 97). Wolbachia was found in repro- ductive tissues and egg cytoplasm of the mosquitoes. By treating the insects with tetracycline, the bacteria were eliminated, and normal reproduction was restored among most mosquito strains. The type of reproductive failure seen in the mosquito studies was termed "cytoplasmic incompatibility" or CI. CI results when fertilization fails to produce a viable embryo due to cytoplasmic egg factors that are incompatible with fertilizing sperm. In the early mosquito crossing experiments (1950s - 1970s) researchers suspected that more than one bacterial strain might be involved because the degree of reproductive failure varied among the strain crosses. At that time, conventional staining and bacterial culturing techniques could not distinguish among
Wolbachia strains. Consequently, Wolbachia research stalled until routine and affordable molecular techniques became available in the early 1990s.
MOLECULAR IDENTIFICATION It is impossible to culture Wolbachia outside of host cells. For that reason modern molecular tools, such as the polymerase chain reaction (PCR) and automated DNA sequencing, are necessary to distinguish Wolbachia strains. These techniques
6 have enabled researchers to show presence and effects of Wolbachia in a wide range of invertebrate hosts (9-11, 18, 31, 42, 48, 57, 59, 61, 67, 78, 80, 90-93, 95). Three Wolbachia genes, varying in sequence conservation, are currently used to identify strains and to create strain phylogenies. The ribosomal gene, 16S, is commonly used to identify bacterial strains. As a conserved gene it is useful for establishing deep phylogenetic relationships. The work of O'Neill et al. (52) was the first to show that Wolbachia consists of many strains that belong to a unique clade of a-Proteobacteria in the Rickettsial group.
The cell-cycle gene, ftsZ, was used to establish finer phylogenetic relationships among
Wolbachia strains and to determine the relationships of strains that induced different host repro- ductive phenotypes. With this gene, Werren et al. (90) showed that there was no pattern of coevolution of Wolbachia and its hosts. This study also confirmed results from a previous 16S rDNA study showing that little or no concordant patterns of evolution resulted between
Wolbachia and host lineages (59), suggesting that horizontal transmission of Wolbachia between insect orders was common. The ftsZ analyses resolved two major clades, subgroups A and B, infecting arthropods. Based on ftsZ DNA synonymous substitution rates these clades are estimated to have diverged 58-67 mya (90). Surveys of filarial nematodes show that they have
Wolbachia in their reproductive tissues as well but these strains belong to two novel clades, C and D, which show strong co-cladogenesis with their hosts (1, 72, 78).
The wsp gene, which codes for an outer membrane protein, has been used for even finer resolution of Wolbachia strain evolution since it appears to be evolving more rapidly then either
16S or ftsZ (84, 99). Studies of Wsp reveal a few cases of possible horizontal transfer between parasitoids and their hosts (86, 95), as predicted earlier by Werren et al. (90). Additionally,
7 many more clades, or supergroups, of Wolbachia have been described using Wsp (52). As well, clades of strains infecting related hosts are appearing as strain sequencing advances (48, 80).
Wolbachia DNA sequences are accumulating in GenBank at an astonishing rate (5). With greater numbers of sequences available, detailed phylogenetic analyses to determine the evolution of different Wolbachia phenotypes are emerging. Hinrich et al. (33) examined the origin of male-killing Wolbachia. Using novel and published ftsZ and wsp sequences, the authors were able to construct phylogenies for each of the genes, but could not find a unified
Wolbachia phylogeny using both genes; the ftsZ and wsp sequences isolated from the same strain could vary in their location in the phylogeny of each gene. Likewise, Jiggins (8) found that the rate of recombination among Wolbachia strains of arthropods is similar to that of a free- living relative, Ehrlichia ruminata. If recombination occurs between different Wolbachia strains in the same host, then phylogenies of genes will not be concordant and real phylogenies of the bacteria may be difficult to construct. It appears that tracing the origin of a reproductive phenotype is more problematic than originally thought. Rather than using molecular markers to track the trait, researchers will have to find the Wolbachia gene ultimately responsible for host reproductive manipulations.
WOLBACHIA MANIPULATION OF HOST REPRODUCTION
Since Wolbachia are maternally inherited in egg cytoplasm, male hosts are an evolutionary dead end for the bacteria. The success of the bacteria is dependent on the reproductive success of infected female hosts. By manipulating host reproduction to favor infected females
Wolbachia ensure their own survival. The results of these manipulations create a variety of host reproductive phenotypes. These include i) feminization of males whereby Wolbachia completely eliminate males from infected isopod populations and create female biased broods in one
8 lepidopteran species by converting males to functional females (9, 46, 65, 66); ii) parthenogenesis whereby arrenotokous (parthenogenetic production of haploid males) haplodiploid wasp and mite species are converted to thelotoky (parthenogenetic production of diploid females) when infected with Wolbachia (36, 61, 74-77); iii) "Male-killing" whereby males are eliminated by arresting their development in the egg stage and only females survive.
This effect in found in beetles, butterflies, and flies (24, 26, 38, 39, 44, 45, 88); and iv) cytolplasmic incompatibility whereby infected females gain a reproductive advantage over uninfected females (6, 7, 11, 14, 28, 30, 34, 35, 47, 50, 51, 68, 82, 97).
FEMINIZATION OF GENETIC MALES Feminization induction (FI) of genetic males was originally described in isopods (64-66). This phenotype occurs as a result of Wolbachia interference with the development of the androgenic gland that produces male hormones. In the absence of these hormones, males (ZZ) develop ovaries and a female phenotype (ZZ +
Wolbachia). Uninfected females are heterogametic (ZW). With a high rate of maternal transmission, the feminizing bacteria have caused the disappearance of the W chromosome in some populations. Curing with high temperatures (> 30º C) or antibiotics restores the production of males. The effects of FI Wolbachia in adult males were investigated by injecting uninfected males with an FI inducing strain. These males maintained a functional androgenic gland and secreted male hormones but were “feminized” by the Wolbachia. Bouchon et al. (9) interpret these results as competition between a bacterial factor and the male hormone (MH) for MH receptors. The Wolbachia factor may bind more tightly to the MH receptor thereby blocking the action of the male hormone.
In a study to determine whether FI Wolbachia of isopods forms a monophyletic clade,
Bouchon et al. (9) isolated and sequenced the Wolbachia wsp gene from the majority of infected
9 isopod species across Europe. They discovered that FI strains belong to the B subdivision, and most fall within one clade. To test whether the effects of FI Wolbachia strains were host specific, the researchers infected cured isopod species with Wolbachia strains harbored by other isopods. They found that the FI strains tended to induce FI in isopods closely related to the original host, whereas the effects on more distantly related hosts were unpredictable. The results ranged from feminization, to no effect, to lethal effects similar to the pathogenic “popcorn” strain found in lab cultures of mutant Drosophila melanogaster (55).
Only one insect species, the Asian corn borer moth, Ostrinia furnacalis, has been found with a strain of feminizing Wolbachia (46). A few all-female broods were discovered in field populations of this crop pest. All-female lines were raised in the lab for three generations by out- crossing with males from normal sex-ratio broods, and no males were produced. Upon treating moths from all-female broods with tetracycline, all-male progeny resulted. Since sex- determining chromosomes in lepidopterans are ZW for females and ZZ for males, the authors suspected that genetic males were feminized by a bacterial agent.
PARTHENOGENESIS INDUCING WOLBACHIA Parthenogenesis inducing (PI) Wolbachia have been found in over 30 parasitic wasp species (36, 72, 74, 76). Normal haplodiploid wasps not infected with Wolbachia have arrhenotokous parthenogenesis, by which they produce males from unfertilized haploid eggs. Female progeny are produced from fertilized eggs. When infected with PI Wolbachia, thelytokous parthenogenesis occurs, so that diploid female offspring are produced from haploid eggs due to an incomplete first mitotic division (77). Usually, few to no males are produced. Stouthamer et al. (75) found that curing Trichogramma parasitoid wasps of their Wolbachia infections resulted in the normal production of males (~25% of offspring).
10 Given sufficient time, Wolbachia-induced parthenogenesis could lead to gene pool fragmentation with the evolution of new species from different parthenogenetic lines. Such may be the case in a couple of species of parasitoid wasps. Hunter (36) discovered that partheno- genetic species of Encarsia parasitoid wasps, naturally infected with Wolbachia, have changed their oviposition behavior compared to uninfected congeneric species. As autoparasitoids, these wasps usually oviposit diploid female eggs into whitefly or scale insect larvae (primary hosts).
Haploid male eggs are placed in conspecific or congeneric female larvae (secondary host). If male eggs are mistakenly deposited into primary hosts, they usually do not develop. Using virgin females cured of Wolbachia, Hunter found that one parthenogenetic species, E. formosa, laid its unfertilized eggs almost exclusively in the primary host. These eggs developed into males and survived. Of the few unfertilized eggs laid in the secondary host, none developed.
The wasp's oviposition behavior had become fixed in favor of the production of female offspring in the primary host. Curing the other Wolbachia-infected parthenogenetic species, E. hispida, did not change its oviposition behavior. Females oviposited in both primary and secondary hosts, but males only developed in the primary hosts. Notably, males of both parthenogenetic species contained little or no mature sperm in their testes. It appears that PI Wolbachia infections have persisted long enough in these species that evolutionary changes in behavior and development have occurred as a result.
MALE-KILLING WOLBACHIA The death of male embryos is attributed to Wolbachia infections in several insects: coccinellid beetles (38), flour beetles (26), nymphalid butterflies
(25, 44), and Drosophila species (24, 39). The mechanisms used by Wolbachia to recognize developing males and inhibit their development are unknown. Because females are
11 heterogametic in butterflies and homogametic in beetles and flies, the mechanisms by which males are recognized and killed may differ for each species.
The benefits to infected female larvae include i) reduced competition for food, ii) increased survival by consuming the eggs containing unhatched brothers and iii) reduced detrimental effects of inbreeding. As yet, nothing is known about the molecular and cellular mechanisms of male-killing by Wolbachia.
The vertical transmission rates of male-killing Wolbachia in many species are quite high.
Laboratory reared broods of infected Adalia bipunctata beetles produced on average 2.8 % male progeny (44). Females sporadically produced a few male offspring. Laboratory lines of
Wolbachia-infected Acraea encedon butterflies produced no males.
CYTOPLASMIC INCOMPATIBILITY The most common phenotype expressed by
Wolbachia in arthropods is cytoplasmic incompatibility, or CI. Embryonic death occurs when uninfected females mate with infected males (unidirectional incompatibility) or when infected females mate with males infected with a different Wolbachia strain (bidirectional incompatibil- ity) (6, 7, 13, 34, 35, 60, 73, 81-83). Infected females mated to uninfected males or males harboring the same Wolbachia strain produce infected, viable offspring. The spread of
Wolbachia within and among populations is facilitated by the ability of infected females to produce fertile offspring by mating with either infected or uninfected males. Uninfected females are at a reproductive disadvantage because they can only produce viable offspring with uninfected males. Turelli and Hoffman (82) documented the rapid spread of Wolbachia through populations of Drosophila simulans in California. They found that Wolbachia dispersed at the rate of 100 km per year. Though earlier data suggested that fly dispersal averaged about 10 km per generation, commercial fruit transport may have facilitated long-range dispersal.
12 In further studies of Wolbachia in Drosophila, Turelli and Hoffman (83) described two important parameters in the establishment and spread of the bacteria: the relative hatch rate from incompatible crosses and the rate of maternal transmission. In Drosophila, they found that old males tended to loose their infections or carried reduced bacterial densities. If most females were uninfected, then mating with uninfected males would limit the spread of the bacteria.
However, once the frequency of Wolbachia in a population crossed a critical threshold, the spread to fixation should be rapid.
Studies at the cellular level have shown that when uninfected eggs are fertilized with sperm from infected males, embryogenesis fails when condensed paternal chromosomes do not line up properly on the mitotic equator during metaphase (17, 48, 63). Presgraves (62) has shown that unknown alterations in the sperm nucleus, not factors in the ejaculate fluid, are the means by which Wolbachia control male contributions to reproduction. One study using Nasonia wasp eggs showed that after fertilization, faulty breakdown of the paternal pronuclear envelope slowed the movement of paternal chromosomes to the mitotic plate (79). These eggs could not divide properly and died as a result. In haplodiploid species CI can affect fertilized eggs in two ways, either eggs die because the paternal chromosome interferes with proper mitotic development as happens with Leptopilina wasps (72), or the eggs develop as haploid males if the paternal chromosome remains dormant and is lost as in Nasonia wasps (12, 60). A consequence of CI in
Nasonia is that all-male broods consisting of normal haploid males and CI-induced haploid males occur (12, 60). Either phenotype of CI in haplodiploids greatly benefits infected females by eliminating their competition, uninfected females, and increases the pool of potential mates.
The duration of Wolbachia once it becomes established in a species is not well known. It is possible to estimate the time since the latest Wolbachia sweep, though it is not possible to know
13 whether previous sweeps have occurred. Since both Wolbachia and host mitochondria are maternally inherited, when Wolbachia sweep through a population, they are escorted by one or a few mitochondrial haplotypes that were associated with the original infection. Over time, new mtDNA mutations accumulate and genetic diversity increases. However, molecular studies show that Wolbachia-infected species have far fewer mitochondrial haplotypes than closely related uninfected species (68, 70). Shoemaker et al. (68) showed that Wolbachia-infected D. recens had much lower mtDNA haplotype diversity compared to its uninfected sister species, D. subquinaria. Yet the level of nuclear genetic diversity in each species was about the same, as would be expected after a genetic sweep that involved only the mitochondrial genome. Thus if mitochondrial haplotypes are tightly linked to Wolbachia, then measuring the rate of mtDNA haplotype evolution of infected hosts may allow an estimation of time since the last Wolbachia invasion. In this way, Shoemaker et al. estimated that D. recens experienced its last Wolbachia sweep about 50,000 years ago. An estimate of nuclear genetic divergence between sister species allowed Shoemaker and colleagues to determine whether the Wolbachia invasion occurred before or after the species split. In the case of Drosophila recens and D. subquinaria, the
Wolbachia invasion happened much more recently than the divergence of the sister species.
Additionally, the original infection of D. recens appears to have taken place before secondary contact of these species, after the retreat of the Wisconsin glacier 10,000 years ago.
The oldest estimate for a Wolbachia infection involves a double infection of a chrysomelid beetle, Chelymorpha alternans. Keller et al. (47) estimate that the last invasion of this species occurred approximately 100,000 years ago. Based on these data and the incongruent phylogenies of Wolbachia strains and host insects it appears that Wolbachia do not persist through evolutionary time with their hosts. Hosts may evolve resistance so that the bacterium is
14 eventually lost or the bacterial genome may accumulate too many deleterious mutations to allow the bacteria to survive. For instance, the wMel strain of Drosophila melanogastor has lost its ability to induce CI in its host, but when transferred to a novel host, D. simulans, CI was restored, indicating that the natural host was developing resistance (53). Although hosts may evolve resistance, Wolbachia are able to occasionally move between unrelated hosts and have thus survived for millions of years.
WOLBACHIA AND HOST FITNESS Wolbachia have been found that may increase male fertility in two diploid species and that may be necessary for female fertility in one species of wasp. Wade and Chang (80) found that infected male Tribolium beetles sired more offspring than their uninfected "conspecific" competitors. However, beetles were kept in large groups and their behavior was not monitored. The superior fertility credited to infected males could have been due to cryptic female choice where females, regardless of their infection status, preferred mating with infected males (21). In another study, Hariri et al. (27) discovered that 43% of tetracycline-cured male stalk-eyed flies were infertile, whereas 17% of infected males were infertile, indicating that the bacterium is necessary for male fertility, though tetracycline may have a negative effect on fertility. Studies of the wasp, Asobara tabida, revealed a triple infection where antibiotic curing of two of the strains had no effect on female fecundity whereas the third strain was found to be necessary for normal ovary development (20).
LETHAL WOLBACHIA An unusual strain of Wolbachia appeared in a laboratory culture of
X-chromosome deficient Drosophila melanogaster (47). Compared to the average life-span of normal flies, the life-span of the infected mutant flies was considerably reduced. Normal life- spans were restored with the elimination of the Wolbachia. This virulent strain was labeled
"popcorn" because of its widespread proliferation throughout brain, muscle, and retinal tissues.
15 Early death of the mutant flies was caused by degeneration of these tissues as bacteria proliferated and induced cell lysis.
WOLBACHIA AS AN AGENT OF EVOLUTION Apart from creating mtDNA bottlenecks as a consequence of a Wolbachia sweep, the effects Wolbachia on the evolution of host genomes are not well known. Shoemaker et al. (70) show that in addition to reducing mtDNA diversity, mtDNA of infected insects appears to have a higher mutation rate than that of uninfected relatives. Thus Wolbachia depress mtDNA diversity within species but accelerate diversity between species. This certainly has consequences for phylogeny estimation as a Wolbachia sweep may carry either ancestral or recently derived haplotypes to fixation.
The dynamics of gene flow among populations may be drastically affected when Wolbachia alter sex-ratios. If males are the sex that tend to disperse but are eliminated by parthenogenesis- inducing Wolbachia then populations may become genetically isolated. On the other hand, populations with few males as a result of male-killing or feminizing Wolbachia, act as sinks for males migrating from other populations. Thus gene flow into the population is increased as uninfected males migrate in and mate with resident females, while gene flow out is reduced due to the lack of resident males.
Conflicts between host and bacterial genomes no doubt occur. Mate choice decisions made by uninfected females have different consequences for host success depending on whether they choose uninfected or infected mates. Normally, conspecific mates produce viable offspring.
When Wolbachia induce CI, an insect's choice of mate will affect the reproductive success of both parents. Uninfected females and infected males suffer reduced success if they mate. If they could recognize and choose the proper mate to produce viable offspring, then Wolbachia might not be able to spread. However, if genes affecting mate choice behavior are located in the
16 nuclear genome and they are in linkage equilibrium with the Wolbachia, then a Wolbachia sweep may move too quickly for selection to act on nuclear genes. Since infected females produce viable offspring with infected and uninfected males, the nuclear genome of uninfected females can flow to the offspring of infected females and mating behavior remains invisible to selection.
Perhaps a model of gene flow among the differently infected individuals could predict the possibility that mates might evolve the ability to detect the Wolbachia status of their mate. If host fitness improves with free mate choice (21, 46), yet epigenetic factors like Wolbachia limit reproductive success, can reproductive behaviors evolve to avoid incompatible pairings? Tests of the effects of Wolbachia on mate choice, genome conflict, and nuclear genetic structure of infected populations remain to be worked out.
Evidence is accumulating for the role of Wolbachia as a speciation agent. Populations that become reproductively isolated may travel separate evolutionary trajectories. Breeuwer and
Werren (11) described bidirectional CI in two species of Nasonia wasps. When cured of their infections, these formerly incompatible species produced viable offspring. Shoemaker et al. (68) demonstrated the role of Wolbachia in maintaining reproductive isolation between two closely related species of Drosophila. These species are broadly sympatric and mate on the same species of fungus. They remain reproductively isolated because Wolbachia -infected D. recens females are reluctant to mate with uninfected D. subquinaria males. In the reciprocal cross CI inhibits production of hybrids. Independent, pre-mating barriers to gene flow are maintained by
Wolbachia in one direction of interspecies crosses, uninfected D. subquinaria females are essentially sterilized by infected D. recens males. In the other direction a behavioral pre-mating barrier deters gene flow across species boundaries because infected D. recens females refuse to mate with uninfected D. subquinaria males.
17 Research by Hunter (36) and Jiggins et al. (32) document changes in insect reproductive behavior as a consequence of their Wolbachia infections. In the case of Encarsia parasitoid wasps, changes in oviposition behavior were altered by Wolbachia and are now permanent (36).
Usually, females of this genus lay unfertilized (male) eggs in a whitefly host. Fertilized (female) eggs are oviposited into other Encarsia larvae. Infected females are parthenogenetic; they do not mate, yet they lay their eggs in Encarsia larvae. Females cured of their bacteria did not revert to their previous host use patterns. Acraea butterflies alter their mating behavior when infected
(32). In the wild, over 90% of females from some populations of A. encedon and a sister species,
A. encedana, are infected with male-killing Wolbachia. In these species, where males usually form leks in the forest to compete for females, the lack of males in the Wolbachia-infected populations has caused a sex-role reversal. Females form leks on ridges vying for male attention and leave the leks soon after mating. Antibiotic curing caused a reversion to normal behavior.
FUTURE DIRECTIONS OF WOLBACHIA RESEARCH Scientific interest in Wolbachia has grown from a curiosity to a discipline. With molecular techniques, the many roles of Wolbachia are being revealed. Wolbachia has the ability to cause parthenogenesis, feminization of genetic males, male-killing, and cytoplasmic incompatibility. It has become essential for reproduction in one insect species. Though the exact mechanisms employed by Wolbachia to manipulate their hosts still remain largely unknown, much of research currently underway is focusing on molecular and cellular aspects of Wolbachia functions. (100). For example, by monitoring
Wolbachia in fly ovaries, Frydman et al. (27) have found that the wMel strain of D. melanogastor accelerates egg production by transporting host proteins to oocytes by shuttling along microtubules using host motor proteins. Under competitive conditions infected flies produced 20% more eggs than did uninfected flies. One study by Hurst et al. (39) showed that a
18 male-killing Spiroplasma bacteria of Drosophila "recognizes" and kills males by activation of a male-killing toxin induced by host dosage compensation proteins in the egg. Hurst and company found that the male-killer effect was blocked if any one of the five genes was disabled when they used mutant flies that had one of five dosage compensation genes knocked out.
Several Wolbachia genomes are sequenced or nearly sequenced (96, 100). Genomic work shows that the Wolbachia genome is very plastic. Genome sizes vary: 0.8 Mb for the wUni strain from the parasitoid wasp Muscidifurax uniraptor; 1.08 Mb for the wBru strain from the filarial nematode Brugia malayi; 1.27Mb for wMel from the fruitfly D. melanogaster; and 1.6
Mb for wRi from the fruitfly D. simulans. Most Wolbachia genomes include transposable elements and bacteriophages that may be responsible for altering synteny, accelerating the rate of gene degradation and moving genes among Wolbachia genomes. Recombination among strains occurs frequently (43) and even bacteriophages of Wolbachia show recombination (8). Multiple infections are common and recombination may allow the bacterium to quickly adapt to new host environments. Werren and colleagues (94) are now in the process of sequencing 40 genes from
40+ strains in order to understand how Wolbachia have evolved to survive in so many different hosts.
Genomic research shows that Wolbachia has an intact and conserved type IV secretion system as well as proteins with ankyrin repeats (15, 29); both of which are involved in bacteria-host interactions. The type IV secretion system is responsible for secreting effector molecules, protein or DNA, into the host environment to modify it for bacterial residence. The Wolbachia genome (wMel) has the greatest number of genes with ankyrin repeats found so far among prokaryotes. Ankyrin repeats are involved in protein-protein interactions and have been associated with virulence in other bacteria. Proteins bearing these motifs might be released to
19 interact with host molecules and may be associated with host manipulation in Wolbachia (41).
The function of Wolbachia proteins with ankyrins repeats remains to be investigated.
The ability of Wolbachia to move horizontally between hosts in different orders is another puzzle. Potential vectors include parasitic mites, parasitoid wasps and flies, strepsiptera and parasitic nematodes. There is some indication that parasitoid wasps may vector Wolbachia because some studies of host/parasitoid communities have shown that the parasitoid and host share the same strains (86). Likewise, parasitic strepsipterans and several species of planthopper they parasitize carry the same Wolbachia strain (57). More laboratory manipulations of parasitoids, hyperparasitoids, and hosts need to be done in order to understand Wolbachia movement through communities.
Studies on the use of Wolbachia as an agent to drive desirable control genes into insect pests and pathogen vectors are in progress (72, 100). Introduction of new genes or inappropriately expressed genes into arthropod populations may produce refractory vectors that are no longer capable of harboring or spreading infectious disease agents. Programs that use sterile males to control disease and agriculturally important crop pests might benefit by releasing Wolbachia- infected males instead. B. Eberhard (pers. com.) has suggested that sterile-male release programs for controlling Mediterranean fruit flies may not be as effective as originally planned.
Irradiating the males to sterilize them causes behavioral changes that affect their ability to successfully court mates. Males with Wolbachia are not known to loose their mating vigor. In insects where females only mate once or are refractory to remating for some time, CI matings should be effective in managing those populations. The effects of human impact on Wolbachia dispersal have yet to be fully explored. Where habitats are altered, and arthropod communities change, Wolbachia may be able to move into new hosts on new continents. The introduction of
20 foreign arthropod species into new habitats across the planet may have changed the global balance of Wolbachia infections.
Wolbachia is an intriguing microbe capable of adapting to a large variety of arthropod hosts.
By hiding in host vacuoles it is able to survive without eliciting a host immune response. With its ability to recombine and occasionally move horizontally among hosts, Wolbachia is able to outrun Müller's Rachet and survive despite host resistance. Future studies focused on molecular research will expose the mechanisms involved in host-Wolbachia interactions and ultimately reveal how evolution has shaped this peculiar yet clever bacterial group.
21 LITERATURE CITED 1. Bandi, C., Anderson, T.J.C., Genchi, C. and Blaxter, M.L. 1998. Phylogeny of Wolbachia in filarial nematodes. Proc. R. Soc. Lond. B 265: 2407-2413.
2. Barr, A.R. 1980. Cytoplasmic incompatibility in natural populations of a mosquito, Culex pipiens L. Nature 283: 71-72.
3. Baumann, P., Lai, C.-Y., Baumann, L., Rouhbakhsh, D., Moran, N. A. and Clark, M.A. 1995. Mutualistic associations of aphids and prokaryotes: biology of the genus Buchnera. Appl. Env. Micro. 61(1): 1-7.
4. Beard, C.B., Durvasula, R.V. and Richards, F.F. 1998. Bacterial symbiosis in arthropods and the control of disease transmission. Emerging Infectious Diseases 4(4): 581-591.
5. Benson, D.A., Karsch-Mizrachi, I., Lipman, D.J., Ostell, J. and Wheeler, D.L. 2004. GenBank: update. Nucleic Acids Res. 32(Database issue): D23–D26.
6. Binnington, K.C. and Hoffman, A.A. 1989. Wolbachia-like organisms and cytoplasmic incompatibility in Drosophila simulans. J. Invert. Pathol. 54: 344-352.
7. Bordenstein, S.R. and Werren, J.H. 1998. Effects of A and B Wolbachia and host genotype on interspecies cytoplasmic incompatibility in Nasonia. Genetics 148: 1833-1844.
8. Bordenstein, S.R. and Wernegreen, J.J. 2004. Bacteriophage flux in endosymbionts (Wolbachia): infection frequency, lateral transfer, and recombination rates. Mol. Biol. Evol. 21(10): 1981–1991.
9. Bouchon, D., Rigaud, T. and Juchault, P. 1998. Evidence for widespread Wolbachia infection in isopod crustaceans: Molecular identification and host feminization. Proc. R. Soc. Lond. B 265: 1081-1090.
10. Bourtzis, K. and O’Neill, S.L. 1998. Wolbachia infections and arthropod reproduction. Bioscience 48: 287- 293.
11. Breeuwer, J.A.J., 1997. Wolbachia and cytoplasmic incompatibility in the spider mites Tetranychus urticae and T. turestani. Heredity 79(1): 41-47.
12. Breeuwer, J.A.J. and Werren, J.H. 1990. Microorganisms associated with chromosome destruction and reproductive isolation between two insect species. Nature 346: 558-560.
13. Breeuwer, J.A.J. and Jacobs, G. 1996. Wolbachia: intracellular manipulators of mite reproduction. Exp. Appl. Acarol. 20: 421-434.
14. Bressac, C. and Rousset, F. 1993. The reproductive incompatibility system in Drosophila simulans: DAPI- staining analysis of the Wolbachia symbionts in sperm cysts. J. Invert. Pathol. 61 (3): 226-230.
15. Brownlie, J. 2004. Characterization of a type-IV secretion system identified from the insect endosymbiont Wolbachia pipientis. 3rd International Wolbachia Conf., Heron Island, Australia.
16. Buchner, P. 1965. Endosymbiosis of Animals with Plant Microorganisms. New York: Interscience.
17. Callaini, G., Giovanna Riparbelli, M., Giordano, R. and Dallai, R. 1996. Mitotic defects associated with cytoplasmic incompatibility in Drosophila simulans. J. Invert. Pathol. 67(1): 55-64.
22 18. Chen, X.A. 1999. Concordant evolution of a symbiont with its host insect species: Molecular phylogeny of genus Glossina and its bacteriome-associated endosymbiont, Wigglesworthia glossinidia. J. Mol. Evol. 48(1): 49-58.
19. Clark, M.E., Veneti, Z., Bourtzis, K. and Karr T.L. 2003. Wolbachia distribution and cytoplasmic incompatibility during sperm development: the cyst as the basic cellular unit of CI expression. Mech. Dev. 120(2): 185-98.
20. Dedeine, F., Vavre, F., Fleury, F., Loppin, B., Hochberg, M.E. and Boulétreau, M. 2001. Removing symbiotic Wolbachia bacteria specifically inhibits oogenesis in a parasitic wasp. PNAS 98: 6247–6252.
21. Dobson, S.L., Bourtzis, K., Braig, H.R., Jones, B.F., Zhou, W., Rousset, F. and O’Neill, S.L. 1999. Wolbachia infections are distributed throughout insect somatic tissues and germ line cells. Insect Biochem. 29: 153-160.
22. Douglas, A.E. 1998. Nutritional interactions in insect-microbial symbioses: Aphids and their symbiotic bacteria Buchnera. Ann. Rev. Entomol. 43: 17-37.
23. Drickamer, L.C., Gowaty, P.A. and Holmes, C.M. 2000. Free female mate choice in house mice affects reproductive success and offspring viability and performance. Animal Behav. 59: 371-378.
24. Dyer K.A., Jaenike J. 2004. Evolutionarily stable infection by a male-killing endosymbiont in Drosophila innubila: molecular evidence from the host and parasite genomes. Genetics 168(3): 1443-55.
25. Dyson, E.A., Kamath, M.K. and Hurst, G.D.D. 2002. Wolbachia infection associated with all-female broods in Hypolimnas bolina (Lepidoptera: Nymphalidae): evidence for horizontal transmission of a butterfly male-killer. Heredity 88: 166-171.
26. Fialho, R.F. and Stevens, L. 2000. Male-killing Wolbachia in a flour beetle Proc. R. Soc. Lond. B 267: 1469-1474.
27. Frydman, H. 2004. Host factors involved in Wolbachia germline targeting during Drosophila melanogastor development. 3rd International Wolbachia Conf., Heron Island, Australia.
28. Ghelelovitch, S. 1952. Sur le déterminisme génétique de la sterilité dans le croisement entre differentes souches de Culex autogenicus Roubaud. C.R. Acad. Sci. Pari 24: 2386-2388.
29. Greve, P. 2004. Characterization and transcriptional analysis of two gene clusters for a type-IV machinery in feminizing intracellular symbiont Wolbachia. 3rd International Wolbachia Conf., Heron Island, Australia.
30. Guillemaud, T. Pasteur, N. and Rousset, F. 1997. Contrasting levels of variability between cytoplasmic genomes and incompatibility types in the mosquito Culex pipiens. Proc. R. Soc. Lond. B 264: 245-251.
31. Hariri, A.R., Werren, J.H. and Wilkinson, G.S. 1998. Distribution and reproductive effects of Wolbachia in stalk-eyed flies (Diptera: Diopsidae). Heredity 81 (3): 254-260.
32. Hertig, M. 1936. The Rickettsia, Wolbachia pipientis, (gen. et sp. n.) and associated inclusions of the mosquito, Culex pipiens. Parasitology 28: 453-486.
33. Hinrich, J., van der Schulenberg, G., Hurst, G.D.D., Huigens, T.M.E., van Meer, M.M.M., Jiggins, F.M. and Majerus, M.E.N. 2000. Molecular evolution and phylogenetic utility of Wolbachia ftsZ and wsp gene sequences with special reference to the origin of male-killing. Mol. Biol. Evol. 17(4): 584-600.
34. Hoffman, A.A. and Turelli, M. 1988. Unidirectional incompatibility in Drosophila simulans: inheritance, geographic variation and fitness effects. Genetics 119: 435-444.
23 35. Hoffman, A.A. and Turelli, M. 1997. Cytoplasmic Incompatibility in Insects. In Influential Passengers. O'Neill, S.L., Hoffman, A.R., and Werren, J.H. eds. Oxford University Press.
36. Hunter, M.S. 1999. The influence of parthenogenesis-inducing Wolbachia on the oviposition behavior and sex-specific developmental requirements of autoparasitoid wasps. J. Evol. Biol. 12: 735-741.
37. Hurst, G.D.D., Hurst, L.D. and Majerus, M.E.N. 1997. Cytoplasmic sex-ratio distorters. In Influential Passengers. O'Neill, S.L., Hoffman, A.R., and Werren, J.H. eds. Oxford University Press.
38. Hurst, G.D.D., Jiggins, F.M., von der Schulenburg, J.H.G., Bertrand, D., Werren, J.H., West, S.A., Goriacheva, I.I., Zakharov, I.A., Stouthamer, R., and Majerus, M.E.N. 1999. Male killing Wolbachia in two species of insects. Proc. R. Soc. Lond. B 266: 734-740.
39. Hurst, G.D.D., Bentley, J. and Veneti, Z. 2004. The interaction of male-killing with the sex determination system in Drosophila. 3rd International Wolbachia Conf., Heron Island, Australia.
40. O'Neill, S.L., Hoffman, A.R., and Werren, J.H. eds. 1997. Influential Passengers. Oxford University Press.
41. Iturbe-Ormaetxe, I., Burke, G. Riegler, M. and O'Neill, S. 2004. Distribution and function of ankyrin domain proteins in the insect endosymbiont Wolbachia pipientis. 3rd International Wolbachia Conf., Heron Island, Australia.
42. Jeyaprakash, A. and Hoy, M.A. 2000. Long PCR improves Wolbachia DNA amplification: wsp sequences found in 76% of 63 arthropod species. Insect Mol. Biol. 9: 393–405.
43. Jiggins, F.M. 2002. The Rate of Recombination in Wolbachia Bacteria. Mol. Biol. Evol. 19(9): 1640–1643.
44. Jiggins, F.M., Hurst, G.D.D. and Majerus, M.E.N. 1998. Sex ratio distortion in Acrea encedon (Lepidoptera: Nymphalidae) is caused by a male-killing bacterium. Heredity 81: 87-91.
45. Jiggins, F.M., Hurst, G.D.D. and Majerus, M.E.N. 2000. Sex-ratio-distorting Wolbachia causes sex-role reversal in its butterfly host. Proc. R. Soc. Lond. B 267 (1438): 69-73.
46. Kageyama, D., Hoshizaki, S. and Ishikawa, Y. 1998. Female-biased sex ratio in the Asian corn borer, Ostrinia furnacalis: evidence for the occurrence of feminizing bacteria in an insect. Heredity 81(93): 311- 316.
47. Keller, G.P., Windsor, D.M., Saucedo, J.M. and Werren, J.H. 2004. Reproductive effects and geographical distributions of two Wolbachia strains infecting the Neotropical beetle, Chelymorpha alternans Boh. (Chrysomelidae, Cassidinae). Mol. Ecol. 13: 2405–2420.
48. Kyei-poku, G.K., Colwell, D.D., Coghlin, P., Benkel, B. and Floate, K. D. 2005. On the ubiquity and phylogeny of Wolbachia in lice. Mol. Ecol. 14(1): 285-94.
48. Lassy C.W. and Karr T.L. 1996. Cytological analysis of fertilization and early embryonic development in incompatible crosses of Drosophila simulans. Mech. Develop. 57(1): 47-58.
50. Laven, H. 1951. Crossing experiments with Culex strains. Evolution 5: 370-375.
51. Laven, H. 1959. Speciation by cytoplasmic isolation in the Culex pipiens complex. Cold Spring Harb. Symp. Quant. Biol. 24: 166-173.
52. Lo, N., Casiraghi, M., Salati, E., Bazzocchi, C. and Bandi, C. 2002. How Many Wolbachia Supergroups Exist? Mol. Biol. Evol. 19(3): 341–346.
24 53. McGraw, E.A., Merritt D.J., Droller, J.N. and O'Neill, S.L. 2001. Wolbachia-mediated sperm modification is dependent on the host genotype in Drosophila. Proc. Roy. Soc. Lond. B Biol. Sci. 268(1485): 2565-70.
54. Milosovic-Brockett, M., Alavi, H. and Anderson, W. 1996. The relative effects of female fecundity and male mating success on fertility selection in Drosophila pseudoobscura. PNAS 93: 3080-3082.
55. Min, K.T. and Benzer, S. 1997. Wolbachia, normally a symbiont of Drosophila can be virulent, causing degeneration and early death. PNAS 94: 10792-10796.
56. Moran, N.A. and Telang, A. 1998. Bacteriocyte-associated symbionts of insects. Bioscience 48(4): 295- 304.
57. Noda, H., Miyoshi, T., Zhang, Q., Watanabe, K., Deng, K. and Hoshizaki, S. 2001. Wolbachia infection shared among planthoppers (Homoptera: Delphacidae) and their endoparasite (Strepsiptera: Elenchidae): a probable case of interspecies transmission. Mol. Ecol. 10(8): 2101-6.
58. Oliver, K.M., Russell, J.A., Moran, N.A, and Hunter, M.S. 2003. Facultative bacterial symbionts in aphids confer resistance to parasitic wasps. PNAS 100: 1803–1807.
59. O'Neill, S.L., Giordano, R., Colbert, A.M.E., Karr, T.L. and Robertson, H.M. 1992. 16S rRNA phylogenetic analysis of the bacterial endosymbionts associated with cytoplasmic incompatibility in insects. PNAS 89: 2699-2702.
60. Perrot-Minnot, M., Guo, L.R. and Werren, J.H. 1996. Single and double infections with Wolbachia in the parasitic wasp Nasonia vitripennis: Effects on compatibility. Genetics 143: 961-972.
61. Plantard, O., Rasplus, J.-Y., Mondor G., Le Clainche I. and Solignac, M. 1999. Distribution and phylogeny of Wolbachia inducing thelytoky in Rhoditini and Aylacini (Hymenoptera: Cynipidae). Ins. Mol. Biol. 8(2): 185-191.
62. Presgraves, D.C. 2000. A genetic test of the Mechanism of Wolbachia-induced cytoplasmic incompatibility in Drosophila. Genetics 154: 771-776.
63. Reed, K.M., and Werren, J.H. 1995. Induction of paternal genome loss by the paternal-sex-ratio chromosome and cytoplasmic incompatibility bacteria (Wolbachia): a comparative study of early embryonic events. Mol. Repro. Dev. 40: 408–418.
64. Rigaud, T., Souty-Grosset, C., Mocquard, J.P. and Juchault, P. 1991. Feminizing endocytobiosis in the terrestrial crustacean Armadillidium vulgare Latr. (Isopoda): Recent acquisitions. Endocyt. and Cell Res. 7: 259- 273.
65. Rigaud, T., Mocquard, J.P., and Juchault, P. 1992. The spread of parasitic sex factors in populations of Armadillidium vulgare Latr. (Crustacea, Oniscidea) Effects on sex ratio. Gen. Sel. Evol. 24: 3-18.
66. Rigaud, T. and Juchault, P. 1993. Conflict between feminizing sex ratio distorters and an autosomal masculinizing gene in the terrestrial isopod Armadillidium vulgare Latr. Genetics 133: 247-252.
67. Rousset, F., Bouchon, D., Pintureau, B., Juchault, P. and Solignac, M. 1992. Wolbachia endosymbionts responsible for various alterations of sexuality in arthropods. Proc. R. Soc. Lond. B 250: 91-98.
68. Shoemaker, D.D., Katju, V. and Jaenike, J. 1999. Wolbachia and the evolution of reproductive isolation between Drosophila recens and Drosophila subquinaria. Evolution 53: 1157-1164.
69. Shoemaker, D.D., Dyer, K.A., Ahrens, M. , McAbee K. and Jaenike, J. 2004. Decreased diversity but increased substitution rate in host mtDNA as a consequence of Wolbachia endosymbiont infection. Genetics 168: 2049–2058.
25 70. Shoemaker, D.D., Keller, G.P. and Ross, K.G. 2003 Effects of Wolbachia on mtDNA variation in two fire ant species. Mol. Ecol. 12: 1757–1771.
71. Sironi, M., Bandi, C., Sacchi, L., Disacco, B., Damiani, G. and Genchi, C.1995. Molecular evidence for a close relative of the arthropod endosymbiont Wolbachia in a filarial nematode. Mol. Biochem. Parasitol. 74: 223-227.
72. Star, P. 1999. Biology and distribution of microbe-associated thelytokous populations of aphid parasitoids (Hym., Braconidae, Aphidiinae). J. Appl. Ent. 123: 321-235.
73. Stevens, L. and Wade M. 1990. Cytoplasmically inherited reproductive incompatibility in Tribolium flour beetles: The rate of spread and effect on population size. Genetics 124: 367-372.
74. Stouthamer, R. 1997. Wolbachia-induced parthenogenesis. In Influential Passengers. O'Neill, S.L., Hoffman, A.R., and Werren, J.H. eds. Oxford University Press.
75. Stouthamer, R., Luck, R.F. and Hamilton, W.D. 1990. Antibiotics cause parthenogenetic Trichogramma to revert to sex. PNAS 87: 2424-2427.
76. Stouthamer, R., Breeuwer, J.A.J., Luck, R.F. and Werren, J.H. 1993. Molecular identification of microorganisms associated with parthenogenesis. Nature 361: 66-68.
77. Stouthamer, R. and Kazmer, D. 1994. Cytogenetics of microbe associated parthenogenesis and its consequences for gene flow in Trichogramma wasps. Heredity 73: 317-327.
78. Taylor, M.J. and Hoerauf A. 1999. Wolbachia bacteria of filarial nematodes. 1999. Parasit. Today 15(11):437-442.
79. Tram, U. and Sullivan, W. 2002. Role of delayed nuclear envelope breakdown and mitosis in Wolbachia- induced cytoplasmic incompatibility. Science. 296(5570): 1124-6.
80. Tsutsui N.D., Kauppinen, S.N., Oyafuso A.F. and Grosberg, R.K. 2003. The distribution and evolutionary history of Wolbachia infection in native and introduced populations of the invasive argentine ant (Linepithema humile). Mol Ecol. 12(11): 3057-3068.
81. Turelli, M. 1994. Evolution of incompatibility-inducing microbes and their hosts. Evolution 48: 909-911.
82. Turelli, M. and Hoffman, A.A. 1991. Rapid spread of an incompatibility factor in California Drosophila . Nature 353: 440-442.
83. Turelli, M. and Hoffman, A.A. 1995. Cytoplasmic incompatibility in Drosophila simulans: Dynamics and parameter estimates from natural populations. Genetics 140: 1319-1338.
84. van Meer, M.M.M., Witteveldt, J. and R. Stouthamer. 1999. Phylogeny of the arthropod endosymbiont Wolbachia based on the wsp gene. Insect Mol. Biol. 8(3): 399-408.
85. Vandekerckhove, T.T.M., Watteyne, S., Willems, A., Swings, J.G., Mertens, J. and Gillis, M. 1999. Phylogenetic analysis of the 16S rDNA of the cytoplasmic bacterium Wolbachia from the novel host Folsomia candida (Hexapoda, Collembola) and its implications for Wolbachial taxonomy. FEMS Micro. Let. 180: 279-286.
86. Vavre, F., Fleury, F., Lepetit, D., Fouillet, P. and Boulétreau, M. 1999. Phylogenetic evidence for horizontal transmission of Wolbachia in host-parasitoid communities. Mol. Biol. Evol. 16(12): 1711-1723.
26 87. Vavre, F., Fleury, F., Valardi, J., Fouillet, P. and Boulétreau, M. 2000. Evidence for female mortality in Wolbachia-mediated cytoplasmic incompatibility in haplodiploid insects: epidemiologic and evolutionary consequences. Evolution 54(1): 191-200.
88. Veneti Z., Toda M.J. and Hurst G.D.D. 2004. Host resistance does not explain variation in incidence of male- killing bacteria in Drosophila bifasciata. BMC Evol Biol. 4(1): 52.
89. Wade, M. J. and Chong, N.W. 1995. Increased male fertility in Tribolium confusum beetles after infection with the intracellular parasite Wolbachia. Nature 373: 72-74.
90. Werren, J.H., Zhang, W. and Guo, L.R. 1995a. Evolution and phylogeny of Wolbachia: reproductive parasites of arthropods. Proc. R. Soc. Lond. B 261: 55-71.
91. Werren, J.H., Windsor, D.M. and Guo, L. 1995b. Distribution of Wolbachia among Neotropical arthropods. Proc. R. Soc. Lond. B 262: 197-204.
92. Werren, J.H. 1997. Biology of Wolbachia. Ann. Rev. Entomol. 42: 587-609.
93. Werren, J.H. and Windsor, D.M. 2000. Wolbachia infection frequencies in a sample of North American insects: evidence of a global equilibrium? Proc. R. Soc. Lond. B 267: 1277–1285.
94. Werren, J.H. 2004. Integrative studies in Wolbachia-insect interactions. 3rd International Wolbachia Conf., Heron Island, Australia.
95. West, S.A., Cook, J.M., Werren, J.H., and Godfray, H.C. 1998. Wolbachia in two insect host-parasitoid communities. Mol. Ecol. 7: 1457-1465.
96. Martin Wu, et al. 2004. Phylogenomics of the reproductive parasite Wolbachia pipientis wMel: a streamlined genome overrun by mobile genetic elements. PLOS Biology 2 (3): 0327-0341.
97. Yen, J.H. and Barr, A.R. 1973. The aetiological agent of cytoplasmic incompatibility in Culex pipiens. J. Inv. Pathol. 22: 242-250.
98. Zchori-Fein, E. and Perlman, S.J. 2004. Distribution of the bacterial symbiont Cardinium in arthropods. Mol. Ecol. 13: 2009–2016.
99. Zhou W, Rousset F, O’Neill SL (1998) Phylogeny and PCR-based classification of Wolbachia strains using wsp gene sequences. Proc. R. Soc. Lond. B 265, 509–515.
100. 3rd international Wolbachia Conference, August 2004. Heron Island, Australia.
27 CHAPTER 3 - REPRODUCTIVE EFFECTS AND GEOGRAPHIC DISTRIBUTIONS OF
TWO WOLBACHIA STRAINS INFECTING THE NEOTROPICAL BEETLE,
CHELYMORPHA ALTERNANS BOH. (CHRYSOMELIDAE)
28 INTRODUCTION
Wolbachia are a wide-spread group of endocellular bacteria (Rickettsiae) found in 15-76% of all insect species (Werren et al. 1995b; West et al. 1998; Jeyaprakash & Hoy 2000; Werren
&Windsor 2000) that enhance their own transmission by manipulating host reproduction in various ways, including feminization of males, induction of thelytokous parthenogenesis, male- killing, and cytoplasmic incompatibility, CI (O’Neill et al. 1997; Werren 1997; Stouthamer et al.
1999). Among insects, CI is a frequent result of Wolbachia and leads to the production of inviable eggs when uninfected females mate with infected males (Hoffmann et al. 1990; Reed &
Werren 1995; Lassy & Karr 1996; Callaini et al. 1997). However, infected females are compatible with both infected and uninfected males, thus the reproductive advantage the bacteria impart to infected females enables Wolbachia to spread (see Hoffmann and Turelli 1997).
Infections with two different strains are not uncommon in nature (Jeyaprakash & Hoy 2000;
Werren & Windsor 2000) and in this case, double infected males are incompatible with single infected females bearing either strain (Rousset & Solignac 1995; Perrot-Minnot et al. 1996;
Dobson et al. 2001). Double infections are thus expected to invade populations at the cost of single infections (Perrot-Minnot et al. 1996).
Few studies of the dynamics of multiple infections in natural populations have been carried out in detail and most concern Drosophila species. These studies show that complex compatibilities exist between multiple and single infections in Drosophila, i.e. CI effects of single strains may not be additive when strains occur as multiple infections (Charlat et al. 2002;
James et al. 2002), strain segregation gives rise to single infections (Kondo et al. 1990; Solignac
29 et al. 1994; Merçot et al. 1995; Baudry et al. 2003), host factors may control strain density independent of the number of co-infecting strains (Ikeda et al. 2003; Mouton et al. 2003) and host genotype may affect the expression of CI (Rousset & Solignac 1995; Guillemaud et al.
1997; Charlat et al. 2003). Most studies of multiple infections include insects from temperate climates, though surveys for Wolbachia infections reveal that insects in the Neotropics are more likely to harbor multiple strains (34% of infected insects) than are insects in temperate zones (5 –
7% of infected insects; Werren and Windsor 2000). Here we present results of studies on the dynamics of two Wolbachia strains, wCalt1 and wCalt2, in populations of the Neotropical tortoise beetle, Chelymorpha alternans Boh., in Panama and address possible environmental effects influencing the frequency of one strain.
C. alternans is found from sea level to approximately 1000 m elevation in disturbed but unburned habitats along forest, river and stream edges on various species of Convolvulaceae
(“morning glory” family). Experimental studies indicate that C. alternans enters diapause under conditions simulating the dry season (Pullin & Knight 1992). Our observations indicate that as the dry season develops adults become increasingly scarce with immature stages largely absent.
At the onset of the wet season, in late April of most years, adults reappear. In contrast, adults in nearly non-seasonal habitats near the Caribbean coast remain active and reproductive throughout the year (Windsor & Keller unpublished observations). Since elevated temperatures (Feder et al.
1999; Hurst et al. 2000; Snook et al. 2000) and host diapause (Perrot-Minnot et al. 1996) have been shown to affect the transmission of Wolbachia, it is possible that the distribution of some
Wolbachia strains in tropical insects are affected by low heat tolerance or host adaptations to extreme environments. Because previous work showed that C. alternans was infected with at least two strains (Werren et al. 1995b) and this species is known to be distributed throughout
30 Panama we investigated the distribution and reproductive effects of Wolbachia strains based on the following questions: 1) How many Wolbachia strains infect C. alternans and how are these strains distributed among host populations, i.e. is the double infection invading? 2) Do
Wolbachia in C. alternans cause CI and what are the effects of each strain? 3) What is the maternal transmission rate of each strain and how does this affect bacterial invasion and persistence? 4) How has the last Wolbachia sweep affected host mitochondrial diversity? and 5)
Does climate affect the distribution of strains?
MATERIALS and METHODS
BEETLE COLLECTION AND MAINTENANCE Beetles were sampled at 24 sites across
Panama from December 1997 to November 2002, with some sites resampled up to four times.
Adults and larvae were maintained in the laboratory (12 hours light, 60% humidity, 26o C) on fresh leaves of Merremia umbellata (Convolvulaceae). Leaves were soaked in a 2% solution of bleach (0.352 M NaClO) for 2 minutes, and rinsed three times in fresh water to remove the bleach. This treatment reduced fungal growth on beetles and their eggs. Non-extracted remains of adult beetles are stored as vouchers at 4o C in 95 % ethyl alcohol at the Smithsonian Tropical
Research Institute (STRI) in the Republic of Panama. Beetles raised for CI studies were either first generation offspring of wild-caught females or were second or later generation antibiotic- treated (cured) stocks and were kept as virgins (2-5 weeks of age) until crosses were arranged.
Beetles collected for the study of maternal transmission were allowed to lay 2 - 4 egg masses, then stored at -80o C until tested for Wolbachia.
DNA TEMPLATE PREPARATION DNA was extracted from whole larvae or reproductive tissues of adults previously frozen at –80o C. Insect tissue was ground in extraction buffer (5%
Chelex 100 and 0.4% proteinase K in sterile deionized water) with a sterile pestle, vortexed for
31 10s, then heated at 56o C for 35 min then 95o C for 10 min. After extraction the samples were again vortexed for 10s, cooled to 4o C and then spun in an Eppendorf centrifuge at 14K rpms for
2 minutes to sediment the Chelex and cellular debris. If a sample contained a lipid layer or was not clear then 10 ul of supernatant was removed, avoiding the lipid layer, then added to a tube containing 10 ul of sterile H2O and respun at 14K rpms for 1 minute. This step usually eliminated problems of non-DNA contamination and interference in subsequent polymerase chain reactions (PCR). Extractions were used for both Wolbachia and C. alternans molecular studies and were held at –20o C while in use, or at –80o C for longer storage.
STRAIN IDENTIFICATION We cloned and sequenced portions of the Wolbachia 16S
(483bp) and wsp (560 bp) genes from one beetle from each of the Gamboa and Cana populations. The gene fragments were PCR amplified, ligated into a plasmid vector (pUC 19), transformed into E. coli, then plated onto 1.5% LB agar plates containing 0.095% ampicillin and
XGal. Clones were isolated and each was resuspended in 50 µl of sterile water. One µl of each suspension was used in a 10 µl PCR reaction: 1 µl 10X buffer (ABI), 0.8 µl MgCl2 (10mM), 1
µl nucleotide mixture (8 mM, equal parts each nucleotide), M13 forward and reverse primers (10 mM each), 5.65 µl dH2O, and 0.5 U of Taq polymerase (Amplitaq, ABI). Cycle sequence reactions were prepared using quarter reactions of the ABI Prism dRhodamine dye terminator kit and clones were sequenced on an ABI 377 automated sequencer. Sequences were aligned using
Sequencher ver. 3.1 (Gene Codes Corporation) and the sequences for each gene were compared to determine the number of strains present in each beetle.
We also directly sequenced partial fragments of Wolbachia 16S (952bp), ftsZ (1003bp), and wsp (560bp) genes (O’Neill, et al. 1992; Werren et al. 1995a; Zhou et al. 1998; MJ Research
PTC - 200 thermal cycler) from 1-2 beetles from three populations, Gamboa (Panama province),
32 Cana (Darien province) and Arenas (Veraguas province). Both forward and reverse DNA strands were sequenced as above.
To determine whether variants of each Wolbachia strain were present in C. alternans populations we sequenced a 490 bp portion of the ftsZ gene using ftsZ primers for 1-2 beetles from seven populations (n = 12 beetles, 5 beetle haplotypes) for the wCalt1 strain and four populations (n = 4 beetles, 3 beetle haplotypes) for the wCalt2 strain. We also sequenced a 560 bp portion of the wsp gene for the wCalt1 strain from 1 - 3 single infected beetles from five populations (n = 11 beetles, 6 beetle haplotypes).
POPULATION SAMPLING FOR WOLBACHIA STRAINS Strain-specific primers were created to enable us to track the distribution and temporal frequency changes of the two strains, wCalt1 and wCalt2, in populations of C. alternans. These primers amplify a 490 bp region of the ftsZ gene: wCalt1 ftsZ F – 5’ CAAGCACTAGAAAAGTCGTTA, wCalt1 ftsZ R – 5’
AAGCCCTGGCATAACCATCAGA, wCalt2 ftsZ F – 5’ CAAGCGTTAGAGAAGTCATTG, wCalt2 ftsZ R – 5’ CAGTCCTGGCATGATCATCAAA. PCR protocols followed Werren et al.
(1995a). Separate PCR cocktails for each set of primers were prepared and run simultaneously for all populations sampled. A positive control containing both Wolbachia strains and a negative control containing no DNA were run with each PCR reaction. PCR products were visualized on
1% (w/v) TBE-agarose gels stained with ethidium bromide. Samples that tested negative for both strains were retested for extraction quality with insect-specific 28S rDNA primers (Werren
1995b) and were removed from the study if again no amplification products were produced.
Because only the wCalt1 strain was found as a single infection, it is to this strain that we refer when we mention the “single infection”. A “double infection” refers to the presence of both strains in the same beetle. A “mixed infection” or “polymorphic infection” refers to populations
33 that have both double and single infected individuals.
CYTOPLASMIC INCOMPATIBILITY Three experiments were performed to test the strength of CI within and between populations. First, reciprocal crosses were made between populations fixed for the double (wCalt1 + wCalt2, Gamboa and Cana) and single (wCalt1 only,
Guarumal and Santa Fe) infections (see table 1, figure 1 for locations). Next, reciprocal crosses were made between beetles originating from Gamboa stocks consisting of naturally double infected beetles (D), and single infected (S) and uninfected (U) beetles derived from double infected beetles by antibiotic treatment. Last, we performed reciprocal crosses between single and double infected beetles from Remedios, a population polymorphic for the infection types.
We also crossed single and double infected Remedios lines with lines from Gamboa and Cana to determine whether the lack of CI among Remedios beetles was a property of males or females.
To cure Gamboa beetles of Wolbachia we injected 0.05 – 0.1 ml of sterile 0.9% rifampcin solution into 10 female and 10 male abdomens three times a week for two successive weeks.
Ultimately, two unrelated females gave rise to our uninfected lines. One single infected line of the wCalt1 strain was produced in the same manner. We did not encounter the second strain, wCalt2 as a single infection in any partially cured beetles and therefore no crosses were made with beetles singly infected with this strain. Injected beetles were mated to each other, allowed to produce several egg masses, and then were frozen (-80 oC) until tested for Wolbachia status.
Three to five offspring from each brood were tested to verify the status of the infection in the brood. Uninfected lines were mated to each other to produce the second generation of uninfected beetles. These lines were then used in crosses after curing was again confirmed. The third and subsequent generations of single infected beetles were used for crossing studies.
Eight or more replicates for each reciprocal cross were made. While it was not possible to
34 perform all crosses simultaneously, we initiated sets of crosses at the same time to investigate particular CI effects, i.e. within-site crosses for Remedios and Gamboa, and between-site crosses involving Gamboa beetles crossed with beetles from other populations. After producing at least
50 eggs, adults were frozen and later tested for the presence of the Wolbachia strains. Following larval hatching, we recorded the number of eggs in each egg mass and the number of emerged larvae. Hatch rates were calculated as the proportion of successfully hatching larvae averaged across females for each set of crosses. CI was calculated as the hatch rate from a particular cross divided by the average of all compatible crosses (95.6%). We tested the normality of both our original proportions and log and arc-sin transformed data using the Shapiro-Wilk W normality test (Sall & Lehman, 1996). Because the hatch rates in most comparisons were not normally distributed we report results as medians and 10th and 90th quantiles and used nonparametric tests for comparing medians.
Since CI may be different under field and laboratory conditions we measured the hatch rates of eggs produced in the lab by wild-caught single and double infected females collected from a mixed population, Remedios. Data for females that appeared to be sperm-limited (hatch rates less than 20%) were removed prior to the analyses.
INFECTION RATES OF OFFSPRING FROM FIELD-COLLECTED BEETLES Wolbachia infections depend on efficient maternal (vertical) transmission for maintenance in insect populations. If transmission is less than 100% then the bacteria may only persist if they also induce strong CI. Because the frequency of the double infection varied among populations we measured the rate of maternal transmission by female beetles collected from three populations that varied in their frequency of the double infection: Gamboa, fixed for the double infection,
Guarumal, fixed for a single infection of the wCalt1 strain, and Remedios, a population with both
35 single and double infected beetles.
Female beetles were collected from Gamboa in 2000, Guarumal in 2001 and Remedios in
2001-2002 and maintained in the laboratory (variable light, 26-28o C, 75% humidity) in separate containers that were examined every other day for egg production. Egg masses were transferred to sterile petri plates and placed in an incubator (13L: 11D, 26-28o C, 70-75% humidity) until larvae emerged 10-11 days later. We tested 10 offspring from each female for the presence of each strain using strain-specific primers. While PCR of Wolbachia from Guarumal single infected, 1-day old larvae gave reliable results, i.e. results similar to population estimates, we found that reliable PCR amplification of Wolbachia from offspring of Gamboa double infected females could only be obtained from three-week old adults. We have not further explored the reason for differences in PCR amplification of larvae from the different sites. Since there were double infected beetles in the Remedios population, we also tested Remedios progeny for
Wolbachia as three-week old adults. Egg hatch rates were calculated as in the CI studies.
CLIMATE AND ELEVATION EFFECTS ON WOLBACHIA DISTRIBUTION Rainfall patterns (and temperatures) are markedly different between Atlantic and Pacific sites and could affect the distribution of Wolbachia in host populations where temperatures are extreme or where the host experiences extended diapause due to the lack of rain. For that reason we compared the length of the dry season to the frequency of the less common strain, wCalt2, to determine whether the distribution of this strain was affected by climate. A dry season month was considered as a month in which the average rainfall was less than 200 mm. The rainfall data used in our comparisons were collected between 1956 – 1983 by the Panamanian Institute of
Hydrological and Electrical Resources (I.R.H.E.) with at least 10 years of collection data averaged by month for each site. Many of the I.R.H.E. weather stations were at or near most of
36 our collecting sites (n = 18) but we were unable to match five sites (2, 10, 13, 14, 17; see table 1, figure1) with weather stations.
As an additional test of temperature effects on strain distribution we examined the frequency of the wCalt2 strain with change in elevation because mean annual temperature drops 0.5o C with a rise of 100 meters in elevation. Our sites varied from 0 – 1000m, corresponding to a range of approximately 5 degrees C.
BEETLE MITOCHONDRIAL DNA We sequenced a 1277 bp portion of the mitochondrial cytochrome oxidase gene (CO1) from two to 10 beetles from each of 24 C. alternans populations
(n = 63 beetles). CO1 was amplified by PCR using two primer pairs (Simon et al. 1994) in separate reactions, C1-J-1718 and C1-N-2191 (494 bp) plus C1-J-2183 and TL2-N-3-14 (783 bp) each in a volume of 25 ml: 0.5ml DNA sample, 2 ml 10X buffer (Applied Biosystems Inc., CA,
USA), 2 ml MgCl2 (25mM), 0.5ml nucleotide mix (4 mM each), 0.5 ml each primer (20mM), 0.10
U Taq DNA polymerase (Amplitaq, ABI) plus distilled, deionized water. PCR cycling conditions were: 95o C for 1 min, 35 cycles of (95o C for 30s, 45o C for 1 min, 68o C for 2 min), then 68o C for 10 min. Sequencing was performed as previously described.
To test whether CO1 of C. alternans has evolved under neutrality we calculated Tajima’s D
(Tajima 1989) and Fay and Wu’s H (Fay and Wu 2000) statistics and ran 10,000 coalescent simulations for each statistic to create 95% confidence intervals. Tajima’s D is used to determine whether there is an excess of rare haplotypes, as expected after a selective sweep, population bottleneck or other processes such as background selection, and is based on the difference between two estimates of nucleotide diversity, q! and qW, where q! is the average of pairwise nucleotide differences and qW is the number of segregating sites within a population
(Watterson 1975). Fay and Wu’s H statistic is used to detect declines in genetic diversity due to
37 selective sweeps or demographic events while being relatively insensitive to background selection (Fay and Wu 2000). H is calculated as q! - qH, where qH is an estimate of nucleotide diversity based on the frequency of derived variants. Diversity estimates and Tajima’s D statistic were calculated using the program DnaSP (Rozas and Rozas 1995) while the Fay and Wu H statistic was estimated using a program provided by J. Fay on the website
(http://crimp.lbl.gov/htest.html). Two specimens of Chelymorpha vittata Champ., a close relative of C. alternans (Keller, Windsor and Werren unpublished data), were sequenced to provide an outgroup to determine the number of derived variants needed for the Fay and Wu H statistic. Because only two C. vittata specimens were sequenced and their sequences were identical, the number of C. alternans derived variants may be inflated due to an underestimate of polymorphisms in C. vittata.
An unrooted haplotype network was constructed with the program TCS ALPHA, v. 1.01
(Clement et al. 2000) that uses statistical parsimony to infer haplotype relationships by the method of Templeton et al. (1992). Three ambiguities in the haplotype network were resolved by assuming the haplotypes were more likely to be related to haplotypes from the same population than to haplotypes from other populations (Crandall and Templeton 1993).
To determine whether host haplotype was correlated with host infection status, as is expected during and shortly after a Wolbachia sweep, we performed a contingency test comparing haplotype by infection status. Because of small sample sizes we combined haplotypes into groups based on the inferred network: [1], [3a,b], [4 a,b,c], [6 a,b,c,d,e,f,g] and [8 a,b,c].
To establish whether the mtDNA diversity of C. alternans is reduced compared to uninfected relatives we tested seven Chelymorpha species (C. vittata, C. gressoria Boh., C. sp. nov., C. testaceomarginata Boh., C. praetextata Boh., C. cinctipennis Boh., and C. cribraria Fabr.) and
38 three closely related Stolas species (S. aenovittata Champ., S. pictilis Boh., and S. n. sp.) for
Wolbachia using Wolbachia general 16S primers but found no uninfected relatives to use for comparison.
RESULTS
STRAIN IDENTIFICATION AND STRAIN VARIANTS We recognized two strains, wCalt1 and wCalt2, based on cloned products of the Wolbachia wsp gene from beetles of two populations (Gamboa and Cana) and by direct sequencing of the wsp and ftsZ Wolbachia gene fragments from beetles of three populations (Gamboa, Cana and Arenas). Cloning produced two wsp sequences (15 clones - wCalt1 (2 clones), wCalt2 (13 clones)) and one 16S sequence (23 clones). Direct sequencing produced identical wsp and 16S sequences to the cloned sequences of each gene, respectively, and two ftsZ sequences. Total sequence divergence between the strains was 11% for wsp (560 bp), 4% for ftsZ, not including indels (1003 bp), and 0% for 16S
(948 bp). Sequences for each strain are deposited in Genbank, accession numbers AY566419 –
AY566426.
The identity of each strain was confirmed when we found beetles infected with only one or the other strain. The wCalt1 strain was commonly found as a single infection in many populations whereas the wCalt2 strain was never found as a single infection in the field.
However, we discovered one male infected with only the wCalt2 strain in our lab stocks of
Gamboa beetles. From single infected beetles we sequenced the wsp, ftsZ and 16S sequences associated with each strain.
Comparisons of sequences for each strain from beetles of many populations and diverse haplotypes revealed no genetic variation for either strain (4 wCalt2 ftsZ sequences [3 beetle
39 haplotypes], 12 wCalt1 ftsZ sequences [5 beetle haplotypes], 11 wCalt1 wsp sequences [6 beetle haplotypes]; table 1)
POPULATION INFECTION FREQUENCIES AND TEMPORAL CHANGES Nearly all
(747/753) beetles sampled from 24 populations were infected with at least one Wolbachia strain, wCalt1 (table 1, figure 2). This strain was found as a single infection in 8 populations and as a co-infection with wCalt2 in 16 populations. The wCalt2 strain was never detected as a single infection in field populations.
We noticed a distinct geographical pattern to single and double infections (figure 2).
Populations in western Pacific Panama, including the Azuero and Sona peninsulas, (populations
6 – 13) were almost entirely single infected (177 beetles; 96.6% single infected, 2.8% uninfected,
0.06% double infected). Outside of this region, seven populations (4, 5, 14, 15, 18, 21, 23) polymorphic for the infections (311 beetles; 47.9% double infected, 51.8% single infected and
0.3% uninfected) and eight populations with predominantly double infections (N=265 beetles,
98% double infected, 2% single infected) occurred throughout Panama (populations 1, 3, 16, 17,
19, 20, 22, 24).
To determine the temporal dynamics of Wolbachia we resampled 11 populations over four years. We found that significant decreases in the frequency of the double infection occurred from 1999 – 2002 in two populations, Las Lajas (population 4, df = 1, G = 4.70, P < 0.05) and
Remedios (5, df = 1, G = 3.98, P < 0.05). From July 1999 to December 1999, the frequency of double infected adults in the Remedios population fell from 70% (n=21) to 27% (n=11).
Subsequent to this decline the frequencies of the infections in the Remedios population did not change significantly between sampling periods (Dec. 1999 – Nov. 2002, G = 0.124, df = 4, n.s, table 1). Non-significant declines of the double infection occurred in Portobelo (21) (df = 1, G =
40 3.67, P > 0.05) and Curundu (18) (df = 1, G = 1.19, P > 0.10). Overall, the number of double infected individuals sampled at all sites declined significantly (i.e. a decline in the frequency of the wCalt2 strain) in collections made during 2000-2002 (G = 9.44, N = 437 beetles, df = 1, P <
0.01).
MATERNAL TRANSMISSION OF WOLBACHIA BY FIELD-COLLECTED BEETLES
Host maternal transmission is one of the key elements, besides CI, affecting the maintenance of
Wolbachia infections. Ideally, maternal transmission should be measured by mating infected females to uninfected males to avoid the loss of uninfected ova to the effects of CI, which would result in an inflation of the maternal transmission rate. A weakness in our design is that we used wild-collected females that were mated in the field, most likely to infected males, to measure vertical transmission. Transmission rates may be over estimated by a few percent, that is, females from these sites may have produced some uninfected ova that failed to develop due to
CI. However, the egg hatch rates of these females were similar to the hatch rates of infected females crossed to uninfected males in later CI studies, so our estimates of maternal transmission may not be overly exaggerated.
We investigated maternal transmission in three populations with different infection states, double (Gamboa), single (Guarumal) and mixed (Remedios). In Gamboa, transmission of both
Wolbachia strains was 100% (10 females, 100 progeny) and in Guarumal transmission of the wCalt1 strain was 99.2% with the remainder uninfected (8 females, 80 progeny). In Remedios, maternal transmission of the wCalt1 strain by both double infected (n = 12, 115 offspring) and single infected (n = 11, 109 offspring) females was 98.7%. Transmission of the wCalt2 strain by double infected females was lower, 83.1%.
EFFECTS OF CLIMATE AND ELEVATION ON STRAIN DISTRIBUTION It is possible
41 that climate influences the frequency of double and single infections, perhaps through effects on transmission. We therefore examined population infection frequencies and climate characteris- tics. The length of the dry season was negatively correlated with the distribution of the wCalt2 strain (figure 3). Populations with five or more months of dry season had significantly lower frequencies of the wCalt2 strain than those populations with shorter dry seasons (200mm: RS =
0.554, n = 19, P << 0.01). However, elevation had no effect on the distribution of the wCalt2 strain (RS = 0.331, n = 22, P > 0.05).
Because the majority of populations with single infected beetles are clustered together in central Pacific Panama, it is difficult to determine whether the loss of the wCalt2 strain is actually due to local climate or early loss with subsequent invasion of single infected beetles into the region. However, in support of the climate theory for loss, several haplotypes in three independent haplotype lineages (1, 6 [b,d,e], and 8 [a]) were found in single and double infected beetles and the double infected beetles with these haplotypes were not found in the dry region.
Further experiments are needed to determine whether climate is responsible for these patterns.
CYTOPLASMIC INCOMPATIBILITY - CROSSING STUDIES We tested the strength of
CI within and between populations with four experimental designs, i) Crosses between naturally double (D) (Gamboa and Cana) and naturally single (S) infected populations (Guarumal and
Santa Fe), ii) Crosses using GB lines that were naturally double infected, plus single and uninfected (U) lines derived from double infected lines by antibiotic treatment, iii) crosses between naturally double and single infected beetles from the Remedios population, and iv) crosses between Remedios beetles and double infected beetles from Gamboa and Cana.
We found that both infection types, single (wCalt1) and double (wCalt1 + wCalt2), caused CI
(figure 4, table 2). Crosses between populations with double and single infections and crosses
42 among Gamboa lines gave the same basic results. Compatible crosses produced 88.5 – 98.8 % hatch rates, while the hatch rates of the incompatible crosses were significantly lower. For the D x S (male x female) cross, hatch rates were 62 – 78% (35 – 18% CI, MWU, c2 = 4.36, df = 1, P =
0.037), for the S x U cross, 56 – 74% (41 – 22% CI, MWU, c2 = 11.70, df = 1, P = 0.001) and for the D x U cross, 7 – 21% (93 – 78% CI) (c2 = 32.38, df = 1, P = 0.001). Therefore, consistent with other systems, double infected males are incompatible with single and uninfected females and single infected males are incompatible with uninfected females.
Individuals from the Remedios population gave different results. The median hatch rate for the Remedios D x S cross (89.5%) was within the range of compatible hatch rates from other populations (89.5% - 97.7%, N = 14) but was significantly lower than the compatible Remedios
S x S cross of 96.6% (MWU = 118, c2 = 4.012, df = 1, P = 0.045) indicating that the double infection in males of this population induces marginal CI (7.6%).
To determine whether the low CI of the Remedios D x S cross was a property of males or females we mated beetles from this population to beetles from Gamboa, a population where
Wolbachia causes strong CI (table 2). The cross of Gamboa D males to Remedios S females resulted in a moderately reduced hatch rate, 74.2%, that was significantly lower than the hatch rate of the Remedios D x S cross, 89.5% (MWU = 116, c2 = 6.205, df = 1, P = 0.013), but not significantly different from the incompatible Gamboa D x S cross, 73.1% (MWU = 44, c2 =
0.096, df = 1, P = 0.757), indicating that the weak CI seen in Remedios was not due to sperm rescue by single infected females. We then crossed Remedios D males with Gamboa S females to compare the CI effect of Remedios males in this population. The resulting hatch rate, 89.9%
(table 2), was nearly identical to that of the Remedios D x S cross, 89.5%, and not significantly different from the compatible Gamboa S x S cross (MWU = 54, c2 = 1.244, df = 1, P = 0.107).
43 As a further comparison of Remedios strains we crossed S and D Remedios males with Gamboa
U females to determine whether the wCalt2 strain in the double infection caused any greater incompatibility than the wCalt1 strain alone. Though the hatch rate of the S x U cross, 60.8%, was somewhat higher than the D x U cross, 53.0%, and both were significantly lower than compatible crosses (S x U: MWU = 27, c2 = 10.473, df = 1, P = 0.001; D x U: MWU = 17, c2 =
21.049, df = 1, P = 0.001), the difference between them was not significant (MWU = 115, c2 =
0.517, df = 1, P = 0.472) further indicating that the wCalt2 strain in Remedios males appears to cause little or no CI.
CYTOPLASMIC INCOMPATIBILITY - POPULATION COMPARISONS The level of CI induced by double infected males varied across populations (figure 5). Uninfected females from
Gamboa stocks crossed with D males from three populations, Remedios, Gamboa and Cana, showed an increasing trend in the CI effect (i.e. decreasing egg hatch rate), respectively.
Remedios males caused the least amount of CI (53%), compared to Gamboa (20.9%) and Cana males (6.8%) (V-W test c2 = 14.93, df = 2, P < 0.0006, Tukey-Kramer HSD; table 2, figure 5a).
A similar trend was seen when the D males were mated to single infected females from either the
Gamboa (figure 5b) or Remedios (figure 5c) populations. Cana males induced the greatest amount of CI followed by Gamboa and Remedios males, respectively. Cana males had significantly reduced egg hatch compared to the other males (V-W test c2 = 23.09, df = 2, P <
0.0001, T-K HSD) when crossed to Remedios S females, but the differences among males mated to Gamboa S females were not significant (V-W test c2 = 3.44, df = 2, P = 0.1789).
CYTOPLASMIC INCOMPATIBILITY – FIELD HATCH RATES To investigate possible levels of CI in field populations of Remedios we examined the hatch rates of eggs produced by field-collected S and D females. We found that the hatch rates of these single (median 88.3%,
44 range 33.2 – 98.5%, N=31) and double (94.7%, 59.1 – 98.8%, N=16) infected females did not differ significantly (MWU = 356, c2 = 0.770, df =1, P = 0.380). This indicates that the double infection in Remedios causes weak to undetectable levels of CI in the field, however, larger samples sizes may reveal the weak CI effect uncovered in the laboratory experiments.
MITOCHONDRIAL DNA DIVERSITY AND WOLBACHIA SELECTIVE SWEEP The mitochondrial haplotypes found in C. alternans, their inferred phylogenetic relationships and
Wolbachia infection statuses are presented in figure 6. A total of 22 haplotypes with twenty- three segregating sites (18 synonymous and 5 nonsynonymous mutations, 16 parsimony informative) were found among 63 infected beetles collected from 24 locations in Panama.
Haplotypes were deposited in Genbank, accession numbers AY563955 – AY563976. Genetic diversity estimates for all samples (N = 63), and subsets of double (N = 36) and single (N = 27) infected beetles are given in table 3. Estimates of mitochondrial genetic diversity were similar for double and single infected beetles due to the large number of haplotypes shared between them. We found no correlation between host haplotype and associated infection type (G = 8.39, df = 4, P > 0.05). The inferred haplotype network (figure 6) shows that haplotypes of most single infected beetles are the same as or are derived from haplotypes of double infected beetles.
This is consistent with the production of single wCalt1 infections from double infected lineages.
To determine whether sequences were evolving in a non-neutral fashion, indicative of a recent selective sweep or demographic event such as a range expansion or population bottleneck, we performed Tajima’s D and Fay and Wu’s H tests on 1) all sequences, and 2) separately for sequences from single and double infected beetles. CO1 appears to have evolved in a neutral fashion as all D and H values were non-significant (table 3). These tests provide no evidence for a recent sweep (or bottleneck), indicating that the current infections were established long ago.
45 The power of these tests to reject the null hypothesis, that of neutral evolution, depends on large samples sizes, i.e. > 50, and a specific window of time since the demographic or selective event occurred because the addition of new mutations obscures events that took place in the remote past (Simonsen et al. 1995). Though our sample size was large (N=63), the last Wolbachia sweep of C. alternans populations may have occurred too long ago to be detected by these tests.
Assuming that the beetle DNA is evolving in a neutral fashion, as suggested by the neutrality tests, based on the pairwise mitochondrial sequence divergence rate of approximately 2.3% per million years for invertebrate mitochondria (Brower 1994) and the synonymous substitution rate of 5.7% per million years in Drosophila (Tamura 1992) we estimated that the last sweep occurred approximately 100,000 - 125,000 years ago.
OTHER WOLBACHIA EFFECTS It was not feasible in this study to measure the lifetime (1 yr) fecundity of C. alternans, so instead we measured the time it took females to produce 50 eggs
(two egg masses) once they were paired with males of the same infection status. Among
Gamboa beetles (antibiotic-treated and infected) we found a marginally significant female effect due to differences between uninfected (19.2 ± 1.3 s.e. days), and single infected (14.5 ± 1.4 s.e. days) females (K-W = 6.45, c2 = 6.452, P = 0.040, df = 2; Tukey-Kramer post-hoc comparison of means, HSD = 0.324). However, double infected females (16.3 ± 1.0 s.e days) did not differ significantly in this measure from either single infected or uninfected females (MWU = 271.4, c2 = 0.197, df = 1, P = 0.657). The interpretation of these results are complicated because the data for each group were collected at different times and the grandsires of the single infected and uninfected lines had been treated with antibiotics.
DISCUSSION
Basic theories of the invasion process of CI-Wolbachia suggest that once the frequency of
46 Wolbachia in a population passes a critical threshold, the infection will spread due to the relative reproductive advantage the bacteria gain from inducing CI. (Caspari & Watson 1959; Hoffmann et al. 1990; Turelli 1994). Factors important in the spread of Wolbachia include the strength of
CI induced, the efficiency of maternal transmission, and the relative fecundity of infected females. This applies both to single infections in uninfected populations and double infections in single infected or uninfected populations (Hoffmann & Turelli 1997). Because mitochondria and
Wolbachia are maternally transmitted and thus co-segregate, analyses of mitochondrial variation can be important in revealing patterns of Wolbachia invasions. As a Wolbachia sweep proceeds, the mitochondrial haplotype associated with the initial infection hitchhikes as the infection invades and thus becomes linked with the invading strain(s) (Turelli et al. 1992; Rousset &
Solignac 1995). Once the invasion is complete, intraspecific mtDNA diversity is reduced to the haplotype associated with the last invasion. Given the tight association between Wolbachia and host mtDNA, the reduction of mitochondrial variation serves as a genetic footprint of the movement of Wolbachia through host populations (Turelli et al. 1992; Shoemaker et al. 1999).
With time, however, the correlation between haplotype and infection type may degrade with the accumulation of mtDNA mutations and strain loss (Solignac et al. 1994; Turelli 1994; James et al. 2002) and may be complicated when more than one strain is involved.
In the present study we found that the strains infecting Chelymorpha alternans, wCalt1 and wCalt2, comprised two infection types - either a double infection of both strains or a single infection of only the wCalt1 strain; the wCalt2 strain was never found as a single infection.
There was a distinct geographic pattern to the distribution of the infection types. Populations in a large region of western Pacific Panama were exclusively single infected whereas populations outside this region were either completely double infected or were polymorphic for single and
47 double infections. We formed two general scenarios to explain the distribution of infection types. The first suggests that there is an ongoing sweep of a double infection that has replaced a preexisting single infection in all populations of the country, except western Pacific Panama.
The second suggests that the double infection has already swept across Panama and strain sorting has left some populations with mixed and single infections. The interpretation of our findings assumes that intraspecific horizontal and paternal transmission of Wolbachia (Hoffman et al.
1990; 1998) and paternal transmission of mitochondria (Kondo et al. 1990) are negligible and have not contributed to the distribution of Wolbachia or mtDNA among conspecific beetle hosts and thus associations of beetle haplotypes and Wolbachia strains are solely due to vertical
(maternal) transmission.
Our results support the second scenario, a long-standing infection of two strains with secondary loss of the wCalt2 strain in some populations. We found that i) the frequency of the double infection across populations was either stable or decreased with time, ii) the wCalt2 strain showed reduced maternal transmission in some populations, iii) the levels of nucleotide diversity were similar for both single and double infected beetles, iv) single and double infected beetles shared the same or similar haplotypes, and v) tests of neutrality for the evolution of the mitochondrial CO1 gene revealed no evidence for a recent Wolbachia sweep. Our findings thus indicate that the last sweep occurred as a double infection in the distant past.
If the double infection is not currently invading single infected populations how do we explain the occurrence of exclusively single infections in populations of the western Pacific region and frequent single/double infection polymorphisms in others? Single infected populations may have formed during the original invasion if strain sorting occurred as the double infection swept through beetle populations, or the loss of the wCalt2 strain may have occurred
48 following the invasion. Several lines of evidence suggest the latter. First, four years of population sampling revealed fluctuations in the frequency of the wCalt2 strain in several populations. Some populations, mostly those on the Pacific slope, showed declines in the frequency of the wCalt2 strain while others, on the Atlantic slope, registered some loss but then recovered. Next, studies of maternal transmission indicated that transfer of the wCalt2 strain to offspring was incomplete for some double infected females from one population (Remedios) polymorphic for the infections. This population was one that experienced a decrease and then stability in the frequency of the wCalt2 strain. Further, the distribution of infection types within the haplotype network indicates that most single infections were formed after the initial sweep of the double infection, although there are two single infected haplotype lineages (3 and 5) that could have formed by loss of the wCalt2 strain during the initial sweep. Considering only vertical transmission of the strains, a double infected haplotype could only be produced through an unbroken chain of double infected ancestral haplotypes that emerged since the initial invasion of the double infection. Haplotypes that are currently polymorphic for infection types must have originated from double infected ancestors. That extant ancestral haplotypes are found in both double and single infected beetles means that the loss of the wCalt2 strain occurred since these haplotypes diverged from their double infected ancestors. Either loss of the wCalt2 strain, and not the wCalt1 strain, occurred randomly as each new haplotype in double infected beetles emerged or else recent environmental changes have caused nearly simultaneous loss of the wCalt2 strain across several extant haplotypes.
Processes that might lead to the elimination of the wCalt2 strain include host resistance, strain competition, and environmental curing by excess heat or induction of prolonged host diapause.
We have not explored the first two possibilities but have some evidence that the wCalt2 strain
49 may be restricted in its distribution by effects of the dry season in part of the range of the host.
At sites where the dry season extends beyond 4 months, we found a significant decline in the frequency of the wCalt2 strain. The length and intensity of the dry season vary across the
Isthmus and beetle activity is ultimately constrained by rainfall. The Atlantic slope is perennially moist, experiencing fluctuations in rainfall with few months receiving less than
200mm of rain and beetles are active year-round. The Pacific slope is seasonally dry and hot with some regions receiving little or no rain for 5 or more months. During the Pacific dry season, which generally lasts from January through April or longer, host plants wither, beetles become scarce and may go into dormancy or diapause until rain and longer days return (Pullin &
Knight 1992).
It is known that both diapause (Perrot-Minnot et al. 1996) and intense heat for short periods
(Feder et al. 1999; Hurst et al. 2000; Snook et al. 2000; Stouthammer et al 1999; Werren 1997) reduce Wolbachia densities and even cure some insects of their bacterial infections in laboratory environments, so it is possible that these factors may have similar effects in natural insect populations. All populations with exclusively single Wolbachia infections occurred in the seasonal western Pacific region, an area that experiences a longer and more intense dry season than the rest of the country, 5+ months. Most of the polymorphic populations also occurred in this area and along the Pacific coast. We found that populations that experienced an average of five months of dry season varied in the frequency of the wCalt2 strain (0 – 100%). Five months may be the limit of tolerance of dry season conditions for the wCalt2 strain. Some populations that averaged five months of dry season might experience an occasional extended dry season due to an infrequent event, such as El Niño. Because we had access only to data averaged over 10+ years these infrequent anomalies were not apparent. However, such an event may be sufficient
50 to eliminate the wCalt2 strain from some or all beetle hosts in the area. The loss of the wCalt2 strain may happen if i) this strain occurs at lower densities in host tissues than the wCalt1 strain and during an extended dry season all bacterial titers are reduced which results in the loss of wCalt2 but not wCalt1, or ii) physiologically, this strain is maladapted to long periods of high temperatures or host diapause and is lost whereas the wCalt1 strain remains unaffected.
Extensive deforestation in Panama (Ibañez 2002) has contributed to the increased dryness of
Panama and may ultimately be causing the decline of the wCalt2 strain in this area. Certainly more studies are needed to determine what factors, heat, low humidity or host diapause are responsible for the loss of the wCalt2 strain.
STRAIN VARIATION AMONG POPULATIONS Since Wolbachia and mtDNA co- segregate, we sequenced genes from each strain to determine whether the strains had also accumulated mutations. No variation was found for either ftsZ or wsp sequences of either strain.
Lack of variation within Wolbachia strains is expected when an infection is invading (Turelli et al. 1992). It is more difficult to explain the lack of variation in long-standing infections where the host shows post-sweep genetic variation (Shoemaker et al. 2003). One might expect that since host mitochondria and Wolbachia share the same history that parallel variation in
Wolbachia genes might be detected. However, mitochondrial mutation rates are substantially higher than those typically found in bacteria, therefore the bacterial genes sampled may not have acquired detectable mutations in the time since the most recent sweep.
CYTOPLASMIC INCOMPATIBILITY AND FIELD HATCH RATES We found that the double infection of C. alternans induced moderate CI in laboratory crosses with single infected beetles. However, we found no evidence that the double infection is supplanting the single infection in mixed or single infected populations. We even found that the frequency of the
51 double infection was declining in two populations.
As with maternal transmission, the expression of CI varied among populations. Hatch rates from compatible crosses, both within and between populations, were reasonably consistent and ranged from 88.5 – 98.0%. However, the hatch rates of incompatible crosses, and thus CI, varied among populations. Double infected males from an eastern population, Cana, caused the strongest CI when mated to either uninfected or single infected females, whereas double infected males from a western population, Remedios, caused almost no more CI than single infected males when crossed with uninfected females, and very little CI when crossed with single infected females. Gamboa males gave intermediate CI results. The Cana and Remedios populations are approximately 610 kilometers apart and Gamboa is located almost exactly midway between them. We have only just begun to explore the reasons for these site differences of CI but have evidence that environmental factors associated with the length of the dry season may be affecting the success of the wCalt2 strain in some regions. Other possible explanations for the differences among populations include i) the occurrence of undetected strain variants among populations that might differ in their abilities to modify and/or rescue sperm, ii) differences in strain titers among and within populations which result in variable CI levels when crosses are made between populations, iii) varying host rescue mechanisms among populations which may be particularly effective against local strains but not other variants.
Remedios is a population polymorphic for the infection types and here we studied the strains in some detail. We found that the wCalt2 strain of the double infection in Remedios has reduced ability to induce CI. Although we have not measured Wolbachia density in these beetles, it is possible that the low CI of double infected males is due to a low density of the wCalt2 strain in this population. Other findings such as incomplete maternal transmission and a decrease in the
52 frequency of the wCalt2 strain in the Remedios population plus the fact that this strain is never found as a single infection are in keeping with this theory. If the density of the wCalt2 strain is lower in general this could explain why strain sorting always leads to single infections of only the wCalt1 strain. Variable Wolbachia densities can lead to strain segregation (Sinkins et al.
1995; Clancy & Hoffmann 1997) and reduced CI levels (Breeuwer & Werren 1993; Perrot-
Minnot & Werren 1999; Noda et al. 2001). However, nuclear restorer genes may also act to circumvent Wolbachia modifications (Roussett et al. 1991; Turelli 1994).
We found that uninfected females were able to partially rescue sperm from double infected males, however, the success of rescue depended on the male’s population of origin: Remedios
(53% hatch rate) > Gamboa (20.9%) >Cana (6.8%). The double infection was also rescued by the single infection in females from two different populations where again the strength of rescue depended on the male’s population of origin (Remedios>Gamboa>Cana). From these results it seems that the strength of CI is dependent on male rather than female factors, suggesting that sperm modification varies across populations. Variability of sperm modification could be due to variable Wolbachia titers (either total or strain-specific) in sperm cysts (Veneti 2003). Since we have indications that the frequency of the wCalt2 strain in some populations is negatively affected by the length of the dry season, it is possible that these conditions also affect wCalt2 titers in males of these populations. Males with decreased wCalt2 titers would have reduced CI in matings with single infected females and thus the double infection would not be able to invade where the negative effects of the dry season are extreme. Further studies of the effects of climate on local strain densities are needed to understand these population differences.
DYNAMICS OF WOLBACHIA IN A POPULATION POLYMORPHIC FOR THE
INFECTION Changes observed in the frequency of the wCalt2 strain in some populations may
53 be due to imperfect maternal transmission and possibly to environmental curing. In the
Remedios population, we measured an 11 – 14% per generation loss of the wCalt2 strain due to strain sorting whereas we were unable to measure any loss of the wCalt1 strain. Likewise, in the lab, CI induced by the double infection was low, 7.6%. Given these results we might expect the frequency of the double infection to decrease with time. Initially, over a six-month period we detected a significant decline in the frequency of the wCalt2 strain in Remedios from 64% to
27% where it then remained stable in samples taken over the next 3 years.
To determine whether our observations of double and single infection frequencies in
Remedios followed predictions we used a model developed by Hoffman and Turelli (1997, p. 65-
67), that incorporates parameters for double infections and imperfect maternal transmission.
Using the June 2001 values from field (f) and laboratory (l) studies (table 4) for the model parameters, µ - segregation rate (l), H – relative hatch rate for incompatible matings (l), p – frequency of infection (f), F – relative fecundity of infected classes (l), the model predicted that the wCalt2 strain in the Remedios population would disappear (frequency less than 1%) in 56 generations, 9 – 14 years (4 – 6 generations per year), due to incomplete maternal transmission and low CI. At the same time, the single infection would to go to 95% fixation. However, three years of population data collected after the initial decline of the wCalt2 strain do not fit these predictions. Rather than decreasing to 11 - 15%, as the model predicts, the frequency of the wCalt2 strain has remained relatively stable over the last three years (table 1). One reason for this stability may be that C. alternans has overlapping generations which violates the model’s assumption of discrete generations. Overlapping generations may slow the rate of decline of the double infection if double infected adults of each generation survive throughout the breeding season. However, inaccuracies in our point estimates of the parameter values could also explain
54 the deviation between observed and predicted, and these values almost certainly vary over time and space.
Our studies of Wolbachia in a Neotropical beetle show that C. alternans has a long-standing infection of two Wolbachia strains, wCalt1 and wCalt2. The wCalt1 strain occurs in all populations, induces weak CI, and has almost complete transmission, whereas wCalt2 occurs in two-thirds of populations only as a double infection with wCalt1, induces moderate to strong CI in conjunction with wCalt1 as a double infection, and is not completely transmitted in all populations. Environmental factors associated with dry season conditions appear to limit the distribution of the wCalt2 strain and may be responsible for its decline in some Pacific populations. Further studies are planned to evaluate the role of environment, bacterial titer and host effects on the dynamics of double infections in this system.
55 REFERENCES
Baudry E, Emerson K, Whitworth T, Werren JH (2003) Wolbachia and genetic variability in the birdnest blowfly Protocalliphora sialia. Molecular Ecology, 12, 1843-1854.
Breeuwer JAJ, Werren JH (1993) Cytoplasmic incompatibility and bacterial density in Nasonia vitripennis. Genetics, 135, 565-574.
Brower AVZ (1994) Rapid morphological radiation and convergence among races of the butterfly Heliconius erato inferred from patterns of mitochondrial DNA evolution. Proceedings of the National Academy of Science USA, 91, 6591-6495.
Callaini G, Dallai R, Riparbelli M G (1997) Wolbachia-induced delay of paternal chromatin condensation does not prevent maternal chromosomes from entering anaphase in incompatible crosses of Drosophila simulans. Journal of Cell Science, 110, 271–280.
Caspari E, Watson GS (1959) On the evolutionary importance of cytoplasmic sterility in mosquitoes. Evolution, 13, 568-570.
Charlat S, Nirgianaki A, Bourtzis K, Merçot H (2002) Evolution of Wolbachia-induced cytoplasmic incompatibility in Drosophila simulans and D. sechellia. Evolution, 56, 1735-1742.
Charlat S, Bonnavion P, Merçot H (2003) Wolbachia segregation dynamics and levels of cytoplasmic incompatibility in Drosophila sechellia. Heredity, 90, 157-61.
Clancy DJ, Hoffmann AA (1997) Behavior of Wolbachia endosymbionts from Drosophila simulans in D. serrata, a novel host. American Naturalist, 149, 975-988.
Clement M, Posada D, Crandall KA (2000) TCS: A computer program to estimate gene geneologies. Molecular Ecology, 9, 1657-1659.
Crandall KA, Tempelton AR (1993) Empirical tests of some predictions from coalescent theory with applications to intraspecific phylogeny reconstruction. Genetics, 134, 959-969.
Dobson SL, Marsland EJ, Rattanadechakul W (2001) Wolbachia-induced cytoplasmic incompatibility in single- and superinfected Aedes albopictus (Diptera: Culicidae). Journal of Medical Entomology, 38, 382-387.
Fay J, Wu CI (2000) Hitchhiking under positive Darwinian selection. Genetics, 155, 1405-1413.
Feder ME, Karr TL, Yang W, Hoekstra JM, James, A.C. (1999) Interaction of Drosophila and its endosymbiont Wolbachia: natural heat shock and the overcoming of sexual incompatibility. American Zoologist, 39, 363-373.
Guillemaud, T. Pasteur N, Rousset F. (1997) Contrasting levels of variability between cytoplasmic genomes and incompatibility types in the mosquito Culex pipiens. Proceedings of the Royal Society, London series B, 264, 245- 251.
Hoffmann AA, Turelli M (1997) Cytoplasmic incompatibility in insects. Pp. 42-80 in S.L. O’Neill, A.A. Hoffmann, and J.H. Werren, eds. Influential Passengers: inherited microorganisms and arthropod reproduction. Oxford Univ. Press, New York.
56 Hoffmann AA, Turelli M, Harshman, LG (1990) Factors affecting the distribution of cytoplasmic incompatibility in Drosophila simulans. Genetics, 126, 933-948.
Hoffmann AA, Hercus M, Dagher E (1998) Population dynamics of the Wolbachia infection causing cytoplasmic incompatibility in Drosophila melanogastor. Genetics, 148, 221-231.
Hurst, GDD, Johnson, AP, Von der Schulenberg, JHG, Fuyama, Y (2000) Male-killing Wolbachia in Drosophila: A temperature sensitive trait with a threshold bacterial density. Genetics, 56, 699-709.
Ibañez R, Condit R, Angehr G, Aguilar S, Garcia T, Martinez R, Sanjur A, Stallard R, Wright SJ, Rand AS, Heckadon S (2002) An ecosystem report on the Panama Canal: monitoring the status of the forest communities and the watershed. Environmental Monitoring and Assessment, 80, 65-95.
Ikeda T, Ishikawa H, Sasaki T (2003) Regulation of Wolbachia density in the Mediterranean flour moth, Ephestia kuehniella, and the almond moth, Cadra cautella. Zoological Science, 20, 153-7.
James AC, Dean MD, McMahon ME, Ballard JWO (2002) Dynamics of double and single Wolbachia infections in Drosophila simulans from New Caledonia. Heredity, 88, 182-189.
Jeyaprakash A, Hoy MA (2000) Long PCR improves Wolbachia DNA amplification: wsp sequences found in 76% of 63 arthropod species. Insect Molecular Biology, 9, 393-405.
Kondo R, Satta Y, Matsuura ET, Ishiwa H, Takahata N, Chigusa SI (1990) Incomplete maternal transmission of mitochondrial DNA in Drosophila. Genetics 126, 657-663.
Lassy CW, Karr TL (1996) Cytological analysis of fertilization and early embryonic development in incompatible crosses of Drosophila simulans. Mechanisms of Development, 57, 47-58.
Merçot H, Llorente B, Jacques M, Atlan A, Montchamp-Moreau C (1995) Variability within the Seychelles cytoplasmic incompatibility system in Drosophila simulans. Genetics, 141, 1015-1023.
Mouton L, Henri H, Bouletreau M, Vavre F (2003) Strain-specific regulation of intracellular Wolbachia density in multiply infected insects. Molecular Ecology, 12, 3459-65.
Noda H, Koizumi Y, Zhang Q, Deng K (2001) Infection density of Wolbachia and incompatibility level in two plant hopper species, Laodelphax striatellus and Sogatella furcifera. Insect Biochemistry and Molecular Biology, 31, 727-737.
O’Neill SL, Karr TL (1990) Bidirectional incompatibility between conspecific populations of Drosophila simulans. Nature, 348, 178-180.
O’Neill SL, Giordano R, Colbert AME, Karr TL, Robertson HM (1992) 16S rRNA phylogenetic analysis of the bacterial endosymbionts associated with cytoplasmic incompatibility in insects. Proceedings of the National Academy of Science USA, 89, 2699-2702.
O’Neill SL, Hoffman AA, Werren JH (1997) Influential Passengers: Inherited microorganisms and arthropod reproduction. Oxford University Press Inc. New York. 214 pp.
Perrot-Minnot MJ, Guo LR, Werren JH (1996) Single and double infections with Wolbachia in the parasitic wasp Nasonia vitripennis: effects on compatibility. Genetics, 143, 961-972.
Perrot-Minnot, M.J. and J.H. Werren. 1999. The dynamics of Wolbachia infection in the parasitic wasp Nasonia vitripennis: selection on incompatibility and bacterial inheritance patterns. Journal of Evolutionary Biology, 12, 272-282.
57 Pullin AS, Knight TM (1992) Induction and termination of reproductive diapause in a Neotropical beetle, Chelymorpha alternans (Coleoptera, Chrysomelidae). Journal of Zoology, 227, 509-516, Part 3.
Reed KM, Werren JH (1995) Induction of paternal genome loss by the Paternal-Sex-Ratio chromosome and cytoplasmic incompatibility bacteria (Wolbachia): A comparative study of early embryonic events. Molecular Reproduction and Development, 40, 408-418.
Rozas J, Rozas R (1995) DnaSP, DNA sequence polymorphism: an interactive program for estimating population genetics parameters from DNA sequence data. Computer Applications in the Biosciences (CABIOS), 11, 621-625.
Rousset F, Solignac M (1995) Evolution of single and double Wolbachia symbioses during speciation in the Drosophila simulans complex. Proceedings of the National Academy of Science USA, 92, 6389-6393.
Sall J, Lehman A (SAS Institute) (1996) JMP Start Statistics, Duxbury Press, Wadsworth Publishing Company, Belmont, CA.
Shoemaker DD, Katju V, Jaenike J (1999) Wolbachia and the evolution of reproductive isolation between Drosophila recens and Drosophila subquinaria. Evolution, 53, 1157-1164.
Shoemaker DD, Keller G, Ross KG (2003) Effects of Wolbachia on mtDNA variation in two fire ant species. Molecular Ecology, 12, 1757-1771.
Sinkins SP, Braig HR, O’Neill SL (1995) Wolbachia pipientis: bacterial density and unidirectional cytoplasmic incompatibility between infected populations of Aedes albopictus. Experimental Parasitology, 81, 284-291.
Simon C, Frati F, Beckenback A, Crespi B, Liu H, and Flook P (1994) Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved PCR primers. Annals of the Entomological Society of America, 87, 651-701.
Simonsen K, Churchill GA, Aquadro CF (1995) Properties of statistical tests of neutrality for DNA polymorphism data. Genetics, 141, 413-429.
Snook RR, Cleland S, Wolfner M, Karr TL (2000) Offsetting effects of Wolbachia infection and heat shock on sperm production in Drosophila simulans: Comparative analyses of fecundity, fertility, and accessory gland proteins. Genetics, 155, 167-178.
Solignac M, Vautrin D, Rousset F (1994) Widespread occurrence of the proteobacteria Wolbachia and partial cytoplasmic incompatibility in Drosophila melanogaster. Comptes Rendus de l’Academie Sciences Paris, Serie III, 317, 461-470.
Stouthammer R, Breeuwer JAJ, Hurst GDD (1999) Wolbachia pipientis: microbial manipulator of arthropod reproduction. Annual Review of Microbiology, 53, 71-102.
Tajima F (1989) Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics, 123, 585-595.
Tamura K (1992) The rate and pattern of nucleotide substitution in Drosophila mitochondrial DNA. Molecular Biology and Evolution, 9, 814-825.
Templeton AR, Crandall KA, Sing CF (1992) A cladistic analysis of phenotypic associations with haplotypes inferred from endonuclease mapping and DNA sequence data. III. Cladogram estimation. Genetics, 132, 619-633.
Turelli M (1994) Evolution of incompatibility–inducing microbes and their hosts. Evolution, 48, 1500-1513.
Turelli M, Hoffmann AA, McKechniw SW (1992) Dynamics of cytoplasmic incompatibility and mtDNA variation in natural Drosophila simulans populations. Genetics, 132, 713-723.
58 Veneti Z, Clark ME, Zabalouc S, Karr TL, Savakisa C, Bourtzis K (2003) Cytoplasmic incompatibility and sperm cyst infection in different Drosophila-Wolbachia associations. Genetics, 164, 545-552.
Watterson GA (1975) On the number of segregating sites in genetical models without recombination. Theoretical Population Biology, 7, 256-276.
Werren JH (1997) Biology of Wolbachia. Annual Revue of Entomology, 42, 587-609.
Werren JH, Windsor DM (2000) Wolbachia infection frequencies in insects: Evidence of a global equilibrium? Proceedings of the Royal Society, London series B, 267, 1277-1285.
Werren JH, Zhang W, Guo L (1995a) Evolution and phylogeny of Wolbachia: reproductive parasites of arthropods. Proceedings of the Royal Society, London series B, 261, 55-71.
Werren JH, Windsor D, Guo L (1995b) Distribution of Wolbachia among Neotropical arthropods. Proceedings of the Royal Society, London series B, 262, 197-204.
West SA, Cook JM, Werren JH, Godfray HCJ (1998) Wolbachia in two insect host-parasitoid communities. Molecular Ecology, 7, 1457-1465.
Zhou W, Rousset F, O’Neill SL (1998) Phylogeny and PCR-based classification of Wolbachia strains using wsp gene sequences. Proceedings of the Royal Society, London series B, 265, 509-515.
ACKNOWLEDGEMENTS. I would like to thank O. Sanjur and R. Collin (STRI) for helpful suggestions on molecular analyses, D. Molbo (STRI), R. Collin, K. Ross (UGA) and anonymous reviewers for helpful suggestions on previous versions of the manuscript, N.A.G. Smith (STRI) for the generous use of his field vehicle, Mark Brown (UGA) for the suggestion and loan of micro-syringes for injecting antibiotics into beetles, the Smithsonian Institution for a Scholarly
Studies Program Grant (DW, JW, project #410-3410-140253) and the Smithsonian Tropical
Research Institute for a pre-doctoral fellowship (GK) and other forms of logistical support.
59 Table 1. Collection and site information (site, province, coordinates, elevation, dry season length and date of collection), infection frequencies (status: 2 – wCalt1+ wCalt2, 1 – wCalt1, 0 – uninfected); total number sampled (N) for each collection date; beetle haplotypes (H), and number of haplotypes sampled (n). Haplotypes from double infected beetles are in bold font and haplotypes of single infected beetles are in regular font.
Elev. # Dry Infection frequencies Site# Sites Coordinates (m) Months Date 2 1 0 N H n . 1†∆ Chiriqui Grande, 8'56"N; 82' 09"W 20 3 Aug. 1998 100% 4 2a 1
Bocas del Toro Jun. 2002 75% 25% 8
Feb. 2003 100% 4
2 Legani, 8'56"N; 82' 03"W 50 na Feb. 2003 100% 1 2b 1
Bocas del Toro
3 Los Planes, Chiriqui 8'35"N; 82'15"W 800 5 Jul. 1999 87.5% 12.5% 8 1,7a,b 3
Nov. 2002 100% 9
4f Las Lajas, Chiriqui 8'10"N; 81'51"W 20 5 Jun. 1997 100% 4 6b,c,d 5
Jun. 1999 86% 14% 14
Jan. 2002 50% 50% 20
5f Remedios, Chiriqui 8'13.5"N; 81' 50"W 30 5 Jul. 1999 64% 36% 22 6b,b, 10
d,e,e
Dec. 1999 27% 73% 11
Jun. 2001 24% 72% 4% 61
Jan. 2002 28% 72% 22
Nov. 2002 24% 76% 34
6∆ Guarumal, Veraguas 7'50"N; 81'15"W 50 5 Dec. 1997 93% 7% 27 6e 3
Jul. 2000 86% 14% 7
Aug. 2001 100% 12
Jan. 2002 100% 15
7∆ Arenas, Veraguas 7'27"N; 80' 52"W 20 5 Dec. 1997 100% 3 6f 3
Oct. 1998 100% 13
8 Pedasi, Los Santos 7'33"N; 80' 2" W 30 11 Jul. 2002 100% 18 1,6g 2
9 Santa Fe, Veraguas 8'32"N; 81'06"W 500 6 May 1999 100% 14 5 2
10 Natá, Cocle 8'18"N; 80'31.5"W 50 na Jan. 2002 100% 3 6g 1
11 Anton, Coclé 8'22"N; 80'17"W 20 10 Jul. 2000 100% 10 6g 2
12 Barrigon, Coclé 8'37"N; 80'36"W 100 8 Oct. 1997 100% 6 3a,3b, 3
6a
Nov. 1998 89% 11% 9
Jul. 2000 91% 9% 11
60 13 La Pintada, Coclé 8'35"N; 80'26"W 100 na Jul. 2000 5% 95% 21 3a, 3
8a,b
May 2002 100% 6
14 Toabré, Coclé 8'38"N; 80'20"W 100 na May 2002 44% 56% 9 8a,a,c 4
15f Coclecito, Colon 8'49"N; 80' 31"W 50 4 Jan. 1998 70% 30% 10 1,8a 3
Jul. 2000 70% 30% 10
Jul. 2002 82% 18% 17
16∆ Cerro Campana, 8'41.2"N; 79' 55.5"W 450 5 Dec. 1999 100% 9 8a 1
Panama
May 2000 100% 16
17∆ Cerro Galera, 8'56"N; 79' 39.5"W 250 na Apr. 1999 100% 12 8a 3
Panama
Jun. 2002 100% 14
18 Curundu, Panama 8'59"N; 79'32"W 50 8 Jul. 1999 70% 30% 10 4b,8a 2
Jul. 2001 44% 56% 9
19†∆ Gamboa, Panama 9'12"N; 79'42"W 60 4 May 1998 100% 8 4b,8a 2
Dec. 1998 100% 10
May 1999 100% 14
Dec. 1999 100% 10
Apr. 2002 100% 17
20 Achiote, Colon 9'14"N; 80'02"W 20 3 Nov. 1998 89% 11% 9 8a 1
21†f Portobelo, Colon 9'32"N; 79' 40" 10 4 Feb. 1998 100% 8 4b,8a 2
Jan. 2002 70% 30% 10
22f El Llano-Carti, 9'18N; 78'58" W 350 5 Jul. 1998 100% 3 4a,a,b 3
Panama
Sep. 2000 100% 13
Dec. 2002 83% 17% 6
23 Santa Fe, Darien 8'28" N; 78' 9'W 50 5 Jan. 1998 50% 50% 10 4a,a 2
24†∆ Cana, Darien 7'45.4"N; 77'41.6"W 500 5 Nov. 1998 100% 15 4d 1
Apr. 2001 100% 10
Sep. 2002 100% 31
† populations sampled for ftsZ from wcalt2 ∆ populations sampled for ftsZ from wcalt1 f populations sampled for wsp from wcalt
61 Table 2. Egg hatch rates from crosses. The median, n, and 90th and 10th quantiles for egg hatch rates (percent) from crosses, within and among sites, of virgin uninfected (0), single (1), and double infected (2) C. alternans. Gamboa 0 and 1 infection types were created by antibiotic-treatment of double infections.
Male Site
------Remedios ------Guarumal Santa Fe ------Gamboa ------Cana
Female Site Infection 1 2 1 1 0 1 2 2
Remedios 1 96.9 [10] 89.5 [16] 95.6 [10] 74.2 [12] 13.1 [12] (100.0-51.4) (97.2-71.9) (98.6-74.3) (87.7-61.1) (77.4-0.00)
2 96.8 [12] 96.6 [11] 97.6 [6] 88.5 [11] (100-66.8) (100-40.4) (100-61.2) (99.7-77.1)
Guarumal 1 97.2 [24] 77.7 [20] (100-80.5) (96.3-43.4)
Santa Fe 1 97.7 [11] 61.8 [11] (100-42.4) (82.4-40.1)
Gamboa 0 60.8 [11] 53.0 [18] 89.5 [17] 73.7 [10] 20.9 [24] 6.8 [7] (96.7-4.5) (79.1-7.3) (99.5-65.8) (93.4-36.8) (51.0-5.8) (81.8-0)
1 98.0 [8] 89.9 [8] 93.2 [10] 96.6 [10] 73.1 [9] 63.9 [8] (100-94.2) (96.7-61.8) (98.3-66.4) (100-46.2) (91.1-45.8) (95.3-31.3)
2 95.8 [21] 96.2 [10] 96.1 [23] 98.8 [10] 91.1 [40] 95.8 [15] (100-78.3) (100-74.2) (100-77.5) (100-57.6) (100-50.1) (97.3-64.2)
Cana 2 94.9 [12] 95.7 [12] 89.5 [12] (98.7-85.2) (99.7-45.6) (99.6-64.0)
62 Table 3. Mitochondrial CO1 haplotype and nucleotide diversity estimates (from single and double infected beetles). N - number of samples, p - nucleotide diversity, SD - standard deviation, q - neutral expectation of p, D – Tajima’s D statistic, H – Fay and Wu’s H statistic.
Number % Number of Pairwise of Haplotype variable genetic N haplotypes diversity sites (S) p SD (p) q (S) SD(q) D H distance All Sequences 63 22 0.929 23 0.0023 0.0001 0.0036 0.0012 -1.157 0.740 0.712 Single inf. 27 15 0.937 12 0.0019 0.0001 0.0024 0.0010 -0.626 0.873 0.658 Double inf. 36 13 0.908 17 0.0025 0.0001 0.0034 0.0013 -0.900 0.820 0.715
63 Table 4. Parameters for invasion model. Field and laboratory values for parameters of the Hoffmann and Turelli (1997) model of infection frequency shifts over time in double infected populations. The values for maternal transmission, µ, were estimated from eggs hatched in the laboratory from field-collected females. Values for infection frequencies, p, were estimated from field collections. Laboratory crosses between infection types produced the hatch rates, H. Fraction of offspring that did not bear the same infection status as their mother, u. Infection states, AB – double infection, A – single infection, 0 – uninfected.
______Parameters June 2001 Jan. 2002 ______uAB 0.132 0.234 uAB,A 0.088 0.234 uAB,0 0.044 0 uA,0 0 0 pAB 0.30 0.29 pA 0.67 0.67 p0 0.03 0.04 HA,AB 0.927 0.927 H0,AB 0.530 0.530 H0,A 0.608 0.608 ______
64 Figure 1. Chelymorpha alternans populations sampled in Panama. The Panama Canal is depicted by the vertical grey line transecting the country.
65 Figure 2. Wolbachia infection status and C. alternans haplotype distributions in Panama populations. Open circles represent single infected populations, filled circles are double infected populations and half-filled circles are populations with both infection types. Haplotypes numbers correspond to those in table 1 and figure 6.
66 Figure 3. Frequency of the wCalt2 strain in populations with varying length of dry season in 19 populations. Black dots represent Pacific slope populations and grey dots are Atlantic slope populations.
67 Figure 4. Egg hatch rates for females with no Wolbachia (0), with the wCalt1 strain (1), or with wCalt1 + wCalt2 strains (2) when crossed with males infected with 0, 1, or 2 strains. Male infection status corresponds to lines inside the figure.
68 Figure 5. The effects of double infected males from Remedios (Rem), Gamboa (Gam) and Cana (Can) on egg hatch rates when mated to a) uninfected Gamboa females, b) single infected (wCalt1) Gamboa females, and c) single infected (wCalt1) Remedios females.
69 Figure 6. Statistical parsimony network of C. alternans haplotypes. The square box represents the most likely common ancestral haplotype and circles represent derived haplotypes. Circle size relates to sample size, 1 – 12 samples per haplotype, and shading corresponds to the proportion of double- (black) or single- (white) infected beetles with those haplotypes. Grey-filled circles represent unsampled haplotypes. Dashed line stands for alternate haplotype connections. Each line between nodes represents one nucleotide substitution.
70 CHAPTER 4 - CONCLUSIONS
71 CONCLUSIONS Wolbachia is a fastidious bacteria whose nature has only recently been exposed with the advent of molecular techniques over the past decade. Residing in host reproductive tissues, Wolbachia is able to manipulate host reproduction to increase its own fitness. Initial Wolbachia research focused on determining the breadth of hosts affected by this bacterial group. It is now known that Wolbachia principally infect nematodes and arthropods, especially insects. It is also known that Wolbachia has diverse effects on host reproduction yet what effect this has on host fitness and survival in natural populations is not well studied. Also, the mechanisms that facilitate Wolbachia manipulation of host reproduction are unknown.
Current studies of Wolbachia are shifting towards molecular and cellular research that explores the means of host-bacteria interactions and bacterial functions. Work is presently being done determine how Wolbachia modify sperm, the basic prerequisite for host invasion by CI
Wolbachia.
Wolbachia is an intriguing microbe capable of adapting to a large variety of arthropod hosts.
By hiding in host vacuoles it is able to survive without eliciting a host immune response. With its ability to recombine and occasionally move horizontally among hosts Wolbachia is able to outrun Müller's Rachet and survive despite host resistance. Future studies focused on molecular research will expose the mechanisms involved in host-Wolbachia interactions and ultimately reveal how evolution has shaped this peculiar yet clever bacterial group.
Because of the ubiquity of Wolbachia in insects it will be important to continue ecological and evolutionary studies in order to understand the underlying role of this microbe on host ecology and evolution. For instance, how does Wolbachia affect male-female conflict over control of reproduction? What role does Wolbachia play in host speciation? What are the consequences of host mitochondrial sweeps on host survival? How often does Wolbachia affect
72 host behavior? What are the costs and benefits to bearing Wolbachia infections? These are questions that await further examination, but with the current level of interest in Wolbachia it will not be long until they are answered.
73