Tsetse-Wolbachia symbiosis: Comes of age and has great potential for pest and disease control

Abstract

Tsetse flies (Diptera: Glossinidae) are the sole vectors of African trypanosomes, the causative agent of sleeping sickness in human and nagana in animals. Like most eukaryotic organisms, Glossina species have established symbiotic associations with bacteria. Three main symbiotic bacteria have been found in tsetse flies: Wigglesworthia glossinidia, an obligate symbiotic bacterium, the secondary endosymbiont Sodalis glossinidius and the reproductive symbiont Wolbachia pipientis. In the present review, we discuss recent studies on the detection and characterization of Wolbachia infections in Glossina species, the horizontal transfer of Wolbachia genes to tsetse chromosomes, the ability of this symbiont to induce cytoplasmic incompatibility in Glossina morsitans morsitans and also how new environment-friendly tools for disease control could be developed by harnessing Wolbachia symbiosis.

1. Introduction

Tsetse flies (Glossina spp.) are found throughout tropical sub- Saharan Africa and are the sole vectors of Trypanosoma spp., which cause African animal trypanosomosis (AAT) or nagana in livestock and human African trypanosomosis (HAT) or sleeping sickness in humans (Leak, 1998; Van den Bossche et al., 2010; Welburn et al., 2001). Current estimates of the World Health Organization (WHO) suggest that epidemics where tens of thousands of people were infected have declined recently, although about 60 million people in Africa continue to be at risk to contract sleeping sickness (Aksoy, 2011; Barrett, 2006; Simarro et al., 2008). In many parts of sub-Saharan Africa, AAT and the presence of tsetse is considered the major obstacle to the development of more efficient and sustainable livestock production systems and one of the most important causes of hunger and poverty (Cattand, 1995; Dyck et al., 2005; Feldmann et al., 2005; Kioy et al., 2004).

Controlling the vector, the tsetse fly, remains theoretically the most efficient and sustainable way of managing AAT (Jordan, 1986; Leak, 1998). There are currently several accepted environment- friendly methods of controlling tsetse: (i) the SAT (sequential aerosol technique) an aerial application of ultralow volume, non-residual insecticides (Jordan, 1974), (ii) stationary bait techniques, i.e. use of insecticide-impregnated targets and traps that can be odor baited (Green, 1994), (iii) the live bait technique, i.e., application of residual insecticides on livestock (Thompson et al., 1991), and (iv) the release of sterile male insects, called the sterile insect technique (SIT) (Oladunmade et al., 1990; Politzar and Cuisance, 1984; Vreysen et al., 2000). Recently, Wolbachia-based control strategies have been suggested as a tool to suppress agricultural pests and disease vectors (Apostolaki et al., 2011; Bourtzis, 2008; Bourtzis and Robinson, 2006; Brelsfoard and Dobson, 2009, 2011; Xi et al., 2005; Zabalou et al., 2009, 2004).

The aim of this review is to summarize the currently available knowledge about the presence of Wolbachia infections in tsetse flies and to describe how this symbiont could be exploited for the control of the tsetse vector species and the associated diseases.

2. Tsetse symbiotic partners

Microbial symbiotic associations, beneficial and pathogenic in nature, are ubiquitous in the Biosphere, including the insect fauna (Bourtzis and Miller, 2009, 2006, 2003; Zchori-Fein and Bourtzis, 2011). In tsetse flies, three main symbiotic bacteria have been described (Aksoy, 2000). The first symbiont Wigglesworthia glossinidia (called Primary endosymbiont) is an obligate symbiotic bacterium harbored by all tsetse flies. Absence of Wigglesworthia results in female sterility (Nogge, 1976; Pais et al., 2008). Two distinct populations of Wigglesworthia have been reported in female tsetse. The first population are intracellularly restricted in specialized epithelial- like cells called the bacteriocytes that make up the bacteriome organ in the anterior midgut. The bacteriome population is thought to provision the host with essential nutrients such as vitamins that are absent from tsetse’s single vertebrate blood-based diet (Akman et al., 2002; Aksoy, 1995; Aksoy et al., 1995; Nogge, 1981; Wang et al., 2009). Extracellular forms of Wigglesworthia have been detected in female milk secretion, through which the symbionts are presumably transmitted to the offspring (Attardo et al., 2008; Ma and Denlinger, 1974; Pais et al., 2008). In addition to its nutritional role, Wigglesworthia was recently reported to influence host immune maturation processes during the larval juvenile stages (Pais et al., 2008; Weiss et al., 2006).

The second symbiont, genus Sodalis glossinidius (called secondary endosymbiont) is present in all individuals in laboratory maintained tsetse lines. It is distributed in many tissues including haemolymph, salivary glands, milk gland, and also in the midgut where it lives in close proximity to where the trypanosomes develop (Aksoy et al., 1997; Cheng and Aksoy, 1999; Dale and Maudlin, 1999). It is also transmitted into tsetse’s intrauterine larva in mother’s milk secretions (Attardo et al., 2008; Pais et al., 2008). Unlike Wigglesworthia, the presence of Sodalis in natural populations is heterogenous (Geiger et al., 2005b). Although the genome size of Sodalis is similar to that of close free-living relatives (e.g. Yersinia and E. coli), a significant portion of the Sodalis’s genome (over 25% of its coding sequences, CDSs) is composed of pseudogenes that have been inactivated, presumably as a consequence of relaxed selection as they no longer play a vital role in the symbiosis (Belda et al., 2010; Toh et al., 2006). The role of this recently acquired symbiont in tsetse flies has not yet been determined, but it has been suggested to influence host longevity and susceptibility to trypanosome infection (Dale and Maudlin, 1999; Dale and Welburn, 2001; Geiger et al., 2007; Geiger et al., 2005a). Sodalis can be cultured in vitro (Welburn et al., 1987) and genetically transformed (Beard et al., 1993a; Dale et al., 1995). Both intron mutagenesis based and lambda red-mediated genetic modification methods have been applied to Sodalis for functional studies (Pontes and Dale, 2011; Runyen-Janecky et al., 2010). A paratransgenic methodology has been developed in order to express trypanocidal products, which can be expressed in tsetse midguts when reconstituted with recombinant Sodalis (Beard and Aksoy, 1998).

The third symbiont, Wolbachia pipientis (Wolbachia for the purposes of this review), belongs to a group of obligatory intracellular and maternally transmitted bacteria, which have been reported in filarial nematode species and are widespread in arthropods, infecting all major orders of insects, mites, spiders, scorpions, isopods and springtails (Hilgenboecker et al., 2008; Saridaki and Bourtzis, 2010; Werren et al., 2008). In insects, the presence of these bacteria in the reproductive tissues is often associated with the induction of parthenogenesis, feminization, male-killing or cytoplasmic incompatibility (abbreviated CI) (Saridaki and Bourtzis, 2010; Werren et al., 2008). CI is the most widespread reproductive effect Wolbachia has on its hosts. It can be unidirectional or bidirectional. In diplo-diploid insects, unidirectional CI is expressed as embryonic lethality, when an infected male is crossed with an uninfected female or a female infected with a different Wolbachia strain. The reciprocal cross, as well as crosses between individuals with the same infection status, is fully compatible. Bidirectional CI is expressed in crosses between individuals infected with Wolbachia strains with different CI properties (Saridaki and Bourtzis, 2010; Werren et al., 2008). Wolbachia-induced CI is currently under consideration as a novel environment-friendly tool for the population control of insect pests and disease vectors (Apostolaki et al., 2011; Bourtzis, 2008; Bourtzis and Robinson, 2006; Xi et al., 2005; Zabalou et al., 2009; Zabalou et al., 2004).

3. Prevalence of Wolbachia infections in populations of tsetse flies

Wolbachia infections seem to be widespread in tsetse flies and concern both laboratory and natural populations of several Glossina species (Table 1). The presence of Wolbachia in tsetse flies was first reported in 1993 (O’Neill et al., 1993). Using a Wolbachia specific 16S rRNA-based PCR assay, Wolbachia was detected in female reproductive tissues of Glossina morsitans morsitans (G. m. morsitans) and G. m. centralis, while it was not found in G. m. submorsitans, G. palpalis palpalis and G. p. gambiensis. Cheng et al. (2000) also detected Wolbachia in laboratory and natural populations of G. austeni and G. brevipalpis.

Table 1.

Wolbachia prevalence in adult flies of natural populations and laboratory lines of twelve Glossina species.

Glossina species	Wolbachia prevalence			References
	Natural	Laboratory	Total
G. austeni	(355/432) 82.2%	Not analyzed	(355/432) 82.2%	Cheng et al. (2000), Doudoumis et al. (2012) and O’Neill et al. (1993)a
G. brevipalpis	(17/218) 7.8%	(24/44) 54.5%	(41/262) 15.6%	Cheng et al. (2000), Doudoumis et al. (2012) and O’Neill et al. (1993)a
G. m. centralis	Not analyzed	(13/13) 100.0%	(13/13) 100.0%	Cheng et al. (2000), Doudoumis et al. (2012) and O’Neill et al. (1993)a
G. m. morsitans	(519/601) 86.4%	(119/119) 100.0%	(638/720) 88.6%	Cheng et al. (2000), Doudoumis et al. (2012) and O’Neill et al. (1993)a
G. f. fuscipes	(0/116) 0.0%	(0/46) 0.0%	(0/162) 0.0%	Cheng et al. (2000), Doudoumis et al. (2012) and O’Neill et al. (1993)a
G. p. gambiensis	(2/555) 0.4%	(0/99) 0.0%	(2/654) 0.3%	Cheng et al. (2000), Doudoumis et al. (2012) and O’Neill et al. (1993)a
G. p. palpalis	(0/48) 0.0%	(0/46) 0.0%	(0/94) 0.0%	Cheng et al. (2000), Doudoumis et al. (2012) and O’Neill et al. (1993)a
G. pallidipes	(19/1800) 1.1%	(13/247) 5.3%	(32/2047) 1.6%	Cheng et al. (2000) and Doudoumis et al. (2012)
G. tachinoides	(0/93) 0.0%	(0/17) 0.0%	(0/110) 0.0%	Cheng et al. (2000), Doudoumis et al. (2012) and O’Neill et al. (1993)a
G. longipinnis	(0/47) 0.0%	(10/10) 100.0%	(10/57) 17.5%	Cheng et al. (2000)
G. swynnertoni	(3/27) 11.1%	(10/10) 100.0%	(13/37) 35.1%	Cheng et al. (2000)
G. m. submorsitans			0.0%	O’Neill et al. (1993)a
Total	(915/3937) 23.2%	(189/651) 29%	(1104/4588)2 24.1%

^aIn the study by O’Neill et al. (1993), G. austeni, G. brevipalpis, G. m. morsitans and G. m. centralis are reported as Wolbachia-infected while G. f. fuscipes, G. p. gambiensis, G. p. palpalis, G. tachnoides and G. m. submorsitans as Wolbachia-uninfected; however, the precise number of individuals tested was not mentioned.

A recent extensive Wolbachia-specific 16S rRNA PCR-based survey of 3750 individuals from both laboratory and natural populations of nine different Glossina species originating from 10 African countries confirmed that: (a) Wolbachia is widespread in species of the morsitans complex: G. m. morsitans, G. m. centralis and G. austeni populations. The report described for the first time Wolbachia in G. pallidipes, albeit at low prevalence, (b) Wolbachia is present in the fusca complex in G. brevipalpis and (c) Wolbachia could not be detected in the species of the palpalis complex: G. p. palpalis, G. fuscipes and G. tachinoides, while it was detected in a few individuals of G. p. gambiensis (Doudoumis et al., 2012).

Wolbachia infections were heterogenous in the field, ranging from 0 to 100% in natural populations of G. austeni and G. brevipalpis and from 9.5 to 100% in natural populations of G. m. morsitans (Doudoumis et al., 2012). This observation suggests that geography may affect Wolbachia prevalence, as reported previously for field populations of the spider Hylyphantes graminicola (Yun et al., 2010), and probably indicates recent sweeps through tsetse populations. Alternatively, this phenomenon could be due to the fact that Glossina populations exhibit extensive genetic structuring, which may influence Wolbachia infection dynamics (Krafsur, 2009; Ouma et al., 2007; Ouma et al., 2011). Of the laboratory strains, the populations of G. m. morsitans, G. m. centralis, G. longipinnis and G. swynnertoni were 100% infected. Wolbachia infection was not fixed in G. brevipalpis and G. pallidipes, and the symbiont was absent from G. p. palpalis, G. p. gambiensis, G. f. fuscipes and G. tachinoides.

4. Genotyping and phylogeny of tsetse Wolbachia strains

Wolbachia is a highly diverse group of bacteria, which are assigned to a single species, Wolbachia pipientis (Lo et al., 2007). The Wolbachia strains are currently classified into 13 supergroups, A to N, while the supergroup G is considered as the artificial result of recombination events (Augustinos et al., 2011; Bordenstein and Rosengaus, 2005; Casiraghi et al., 2005; Cheng et al., 2000; Gorham et al., 2003; Lo et al., 2007; Lo et al., 2002; Ros et al., 2009; Rowley et al., 2004; Zhou et al., 1998). Single gene (mainly using the wsp gene, which encodes for the major Wolbachia surface protein) and/or Multi locus sequence typing (MLST) approaches are the main tools for genotyping Wolbachia strains (Baldo et al., 2006; Paraskevopoulos et al., 2006). Using these approaches, nine allelic profiles or Sequence Types (ST) were found in Wolbachia strains infecting G. m. morsitans, G. m. centralis, G. austeni, G. pallidipes, G. brevipalpis and G. p. gambiensis (Doudoumis et al., 2012). The wsp-based genotyping revealed the presence of nine alleles (Doudoumis et al., 2012).

MLST- and wsp-based phylogenetic analysis showed that Wolbachia strains infecting G. m. morsitans, G. m. centralis, G. austeni, G. pallidipes and G. brevipalpis belong to supergroup A, while the Wolbachia strain infecting G. p. gambiensis is a member of the supergroup B (Cheng et al., 2000; Doudoumis et al., 2012; Zhou et al., 1998). The tsetse Wolbachia strains of supergroup A are members of three separate and distantly related groups, which supports the hypothesis that horizontal transmission of Wolbachia from unrelated taxa may have occurred, as has been described for other arthropod taxa (Baldo et al., 2008; Raychoudhury et al., 2009; Ros et al., 2009; Salunke et al., 2010). It is worth noting, however, that the sibling species G. m. morsitans and G. m. centralis carry closely related Wolbachia strains, which may be the result of a hostsymbiont co-divergence process.

Schneider and colleagues recently showed that Variable- Number-Tandem-Repeat (VNTR) marker loci can be used from the characterization of Wolbachia diversity present in Glossina species. Based on this approach, and or the combined PCRblot technique developed by the same research group, it was confirmed: (a) that tsetse fly species harbor multiple cytoplasmic strains, (b) the presence of Wolbachia gene insertions onto tsetse host chromosomes and (c) the presence of low-titer infections (Schneider et al., 2013).

5. Wolbachia genomics

Since the first fully annotated Wolbachia genome of the Drosophila melanogaster strain wMel (Wu et al., 2004), three additional genomes are currently available. Two of these, wRi and wPip, are Wolbachia strains from insects, Drosophila simulans and Culex pipiens and the third is from the filarial nematode Brugia malayi (Foster et al., 2005; Klasson et al., 2009b, 2008). At least ten genomes are currently being sequenced from a diverse set of hosts, which represents just a small part of the extremely broad phenotypic diversity of Wolbachia (Genomes Online Database – GOLD and personal communication; Table 2). The majority of these sequencing efforts are in progress with so far incomplete genomes, except for the wVitB strain from Nasonia vitripennis and the wPip strain from Culex quinquefasciatus, which are permanent draft genomes (Kent et al., 2011; Salzberg et al., 2005).

Table 2.

Wolbachia genome projects.

Host	Strain	Supergroup	Genome size (Kb)	ORFs	Contigs	Status	Reference	NCBI record
Drosophila melanogaster	wMel	A	1267	1312	1	CG	Wu et al. (2004)	AE017196
Drosophila simulans	wRi	A	1445	1187	1	CG	Klasson et al. (2009b)	CP001391
Culex quinquefasciatus Pel	wPip	B	1482	1312	1	CG	Klasson et al. (2008)	AM999887
Brugia malayi	wBm	D	1080	940	1	CG	Foster et al. (2005)	AE017321
Glossina morsitans morsitans	wGmm	A	1021	932	1	CG	Aksoy S.	NK
Culex quinquefasciatus JHB	wPip	B	1542	1415	21	PDG	Salzberg et al. (2009)	ABZA00000000
Nasonia vitripennis	wVitB	A	NK	NK	523	PDG	Kent et al. (2011)	AERW00000000
Muscidifurax uniraptor	wUni	A	867	855	256	DG	Klasson et al. (2009a)	ACFP00000000
Culex pipiens molestus	wPipMol	B	NK	NK	888	DG	Parkhill J	CACK00000000
Drosophila simulans	wSim	A	1063	781	629	DG	Salzberg et al. (2005)	AAGC00000000
Drosophila willistoni	wWil	A	862	992	260	DG	Salzberg et al. (2005)	AAQP00000000
Drosophila ananassae	wAna	A	1440	1834	464	DG	Salzberg et al. (2005)	AAGB00000000
Dirofilaria immitis	wDim	C	NK	NK	NK	IP	Bandi C	NK
Onchocerca volvulus	wOv	C	NK	31100	NK	IP	Parkhill J and Birren B	NK
Diaphorina citri	ND	B	NK	1063	760	IP	Weinstock G	NK
Wuchereria bancrofti	ND	D	NK	NK	NK	IP	Birren B	NK

CG, complete genome; PDG, permanent draft genome; DG, draft genome; IP, in progress; NK, not known; ND, not designated.

Wolbachia strains have small genomes (1.02–1.7 Mb), with the smallest identified so far being that of the wGmm strain from G. m. morsitans (unpublished data). Although Wolbachia genome sizes are within the range of the Rickettsiales (0.8–2.1 Mb), unlike most Rickettsiales they contain a high number of mobile and repetitive elements (Cerveau et al., 2011; Cordaux, 2008; Leclercq et al., 2011; Wu et al., 2004). Insertion sequences, transposons and the Wolbachia bacteriophage family (known as WO) can constitute up to 21% of the Wolbachia genome (Kent and Bordenstein, 2010; Klasson et al., 2009b, 2008; Wu et al., 2004). This makes Wolbachia a likely target for large scale horizontal gene transfer (HGT) of mobile elements, but so far only scant evidence of HGT events exists. Indirect data indicate that gene transfer of minor capsid gene sequences of bacteriophage WO between Wolbachia co-infections in the same host is common (Bordenstein and Wernegreen, 2004; Chafee et al., 2010), but evidence of phage sequence transfer beyond this single gene has been demonstrated so far only a 52.2 kb phage and the flanking genes between the wVitA and wVitB coinfections of N. vitripennis (Kent et al., 2011).

6. Horizontal gene transfer events

HGT, also known as Lateral Gene Transfer (LGT), permits the movement of genetic information between distantly related species. In prokaryotes, HGT events are almost universal, occur frequently and are regarded as a driving force of prokaryotic evolution (Andam and Gogarten, 2011; Kurland, 2000). In several Wolbachia-host interactions, horizontal gene transfer events from Wolbachia to a variety of insects and nematodes have been reported (Table 3). The first reported HGT event involves the adzuki bean beetle Callosobruchus chinensis, in which an estimated 380 kb genomic fragment of Wolbachia, presumably transcriptionally inactive, was identified in the X chromosome (Kondo et al., 2002; Nikoh et al., 2008). A nearly complete Wolbachia genome was found in the second chromosome of the fruit fly Drosophila ananassae (Dunning Hotopp et al., 2007); almost 2% of the transferred genes seem to be transcribed. Whether these genes have any biological significance is unclear, as the level of transcription is low (Dunning Hotopp et al., 2007; Nikoh and Nakabachi, 2009; Nikoh et al., 2008). Also in the genomes of parasitoid wasps of the genus Nasonia and filarial nematodes of the genera Onchocerca, Brugia and Dirofilaria, a small number of Wolbachia genes have been detected (Dunning Hotopp et al., 2007; Fenn et al., 2006). In the longicorn beetle Monochanus alternatus, at least 14% of a Wolbachia genome, corresponding to more than 170 genes, is located on a chromosome (Aikawa et al., 2009).

Table 3.

Wolbachia acquired genes in insects and filarial nematodes.

Host	Estimated size of genes transferred	Cytoplasmic Wolbachia	Level of expression	References
Callosobruchus chinensis	380 kb	+	−	Kondo et al. (2002) and Nikoh et al. (2008)
Drosophila ananassae	>1 Mb	+	+c	Dunning Hotopp et al. (2007)
Onchocerca volvulus	2 Kb	+	NK	Fenn et al. (2006)
Nasonia vitripentis	491 bp	+	NK	Dunning Hotopp et al. (2007)
Nasonia longicornis	142 bp	+	NK	Dunning Hotopp et al. (2007)
Nasonia giraulti	446 bp	+	NK	Dunning Hotopp et al. (2007)
Brugia malayi	>8 kb	+	NK	Dunning Hotopp et al. (2007)
Dirofilaria immitis	978 bp	+	NK	Dunning Hotopp et al. (2007)
Monochanus alternatus	>180 kb	+	NK	Aikawa et al. (2009)
Aedes aegypta	1278 bp	−	+d	Woolfit et al. (2009) and Klasson et al. (2009a)
Aedes mascarensisa	1278 bp	−	+d	Woolfit et al. (2009) and Klasson et al. (2009a)
Acyrthosiphon pisumb	1312 bp	−	+d	Nikoh and Nakabachi (2009)
Glossina morsitans morsitans	>500 kb	+	NK	Bourtzis K and Serap A. personal communication
Acanthocheilonema viteae	>30 kb	−	+c	McNulty et al. (2010)
Onchocerca flexuosa	>30 kb	−	+c	McNulty et al. (2010)

NK, not known.

^aDirection of transfer is not clear.

^bTransfer from a Wolbachia-like bacterium.

^cExpression was detected just above background level; biological significance is not clear.

^dExpressed above background level.

Gene transfer events have also been detected in the mosquitoes Aedes aegypti (Ae. aegypti) and Ae. mascarensis, although the direction of transfer is unclear (Klasson et al., 2009a; Woolfit et al., 2009). Interestingly, the transferred genes have been classified as transcriptionally active in the salivary glands of Ae. aegypti, which is the most important mosquito vector of human dengue virus and other arboviruses, and naturally not associated with Wolbachia. This suggests that important biological functions of Wolbachia could have been incorporated into host genomes through HGT even in host species that currently do not harbor cytoplasmic Wolbachia infections but apparently did in the past. It is worth noting that only in the absence of Wolbachia, the horizontally transmitted genes present expression levels above background level, although this conclusion can not be generalized. (Table 3).

The extent of HGT events has been further strengthened with the detection of Wolbachia gene transfers to the G. m. morsitans nuclear genome. Initial characterization indicated that at least three genes (16S rRNA, fbpA and wsp) of the naturally infected Wolbachia strain of G. m. morsitans (wGmm) have been transferred to the host genome (Doudoumis et al., 2012). These transfer events were identified in natural and laboratory populations, which highlight the long co-evolution of the host-Wolbachia association. The Whole Genome Sequence of G. m. morsitans (http://www.sanger.ac.uk) led to the identification of at least two horizontal transfer events with 52% and 47% of the Wolbachia genome being transferred to the host (unpublished data). This massive gene transfer from Wolbachia to G. m. morsitans can provide further insight into the evolution and fate of laterally transferred endosymbiont genes in multicellular host organisms. The retention of large fragments of Wolbachia in the G. m. morsitans genome may also suggest that this event may be relatively recent and may not have had time to undergo extensive erosions. Interestingly, in the subspecies G. m. centralis, no (similar) chromosomal insertions have been identified as yet, which suggests that the HGT event may have taken place after the divergence of these two subspecies again supporting its relatively recent origin. However, it is also possible that the Wolbachia chromosomal inserted genes were lost in the G. m. centralis lineage (Doudoumis et al., 2012).

Taken together, these findings suggest that symbiont–host gene transfers might be occurring more frequently than previously thought, similar to the gene transfers from organelles to the eukaryotic nucleus. In the HGT events presented, a genomic Wolbachia sequence, which can be as large as the whole genome, or as small as a few genes are transferred by an as yet unknown mechanism to the host genome. Once integrated, these genes are pseudogenized through the acquisition of point mutations, insertions and/or deletions. So far in a very few cases, some of the transferred genes survive and acquire novel expression patterns, and are thus incorporated into the gene complement of the host organism. It is likely the localization of Wolbachia infections in germ-line tissues promotes the HGT events since no similar gene transfers from symbiont to host genomes have been noted with obligate endosymbionts, including Buchnera in aphids or Wigglesworthia and Sodalis in tsetse (Kurland, 2000).

7. Wolbachia-induced CI and tsetse flies

Although Wolbachia infections have been reported in somatic tissues of several arthropod hosts, the symbiont is mainly present in the reproductive tissues of tsetse flies (Aksoy, 2000; Cheng et al., 2000; Doudoumis et al., 2012; O’Neill et al., 1993). The occurrence of Wolbachia in ovaries and testes has been linked to the induction of reproductive phenotypes, the most common being CI. However, the functional role of the symbiont in tsetse flies had for long time remained elusive, because the use of antibiotics, such as tetracycline, for the establishment of Wolbachia-free lines necessary for appropriate genetic crosses, also eliminated the mutualistic Wigglesworthia and led to the production of sterile flies. Alam and colleagues (2011) recently reported the successful establishment of Wolbachia-free and still fertile G. m. morsitans lines by dietary provisioning of tetracycline supplemented blood meals with yeast extract. This allowed the performance of proper mating experiments, which for the first time provided evidence that the presence of Wolbachia is responsible for the induction of strong CI in G. m. morsitans (Alam et al., 2011). It is worth noting that this was also the first time that Wolbachia-induced CI was investigated in a viviparous insect. The cured lines were further used in comparative studies to investigate potential effects of Wolbachia infections on host fitness under laboratory conditions. The results were clear: (a) there was no major effect of Wolbachia on host eclosion rates and (b) removal of Wolbachia did not result in a decrease of the mating capacity of males (Alam et al., 2011).

Wolbachia-induced CI has been proposed as an environmentfriendly mechanism for population control of agricultural pests and disease vectors (Beard et al., 1998, 1993b; Bourtzis, 2008; Bourtzis and O’Neill, 1998; Dobson, 2003; Sinkins and Godfray, 2004). Expression of strong CI in Wolbachia-infected G. m. morsitans opens new routes for population control of the insect vector and trypanosomosis. This can be done with three potential approaches: First, Wolbachia-induced CI can be used as a population suppression mechanism in a way analogous to the SIT. This Wolbachia- based approach, known as Incompatible Insect Technique (IIT), has been developed and successfully used for population control of agriculturally important insects and disease vectors under laboratory and field conditions (Apostolaki et al., 2011; Laven, 1967; Zabalou et al., 2009, 2004). IIT could potentially be advantageous over SIT, since it does not rely on the use of irradiation to sterilize males, a process that results in the release of males with reduced fitness. In IIT, sterility is due to the presence of a Wolbachia strain generically capable of inducing CI (Bourtzis, 2008; Bourtzis and Robinson, 2006; Vreysen et al., 2011). Second, Wolbachia-induced CI could also be used as a spreading/replacement mechanism for desired phenotypes, because infected females have a reproductive advantage over uninfected females, since they can mate with both uninfected and infected males thus invading populations in nature (Bourtzis, 2008; Bourtzis and Robinson, 2006; Bourtzis and O’Neill, 1998; Brelsfoard and Dobson, 2009; Dobson et al., 2002; Hoffmann et al., 1998; Rasgon, 2008, 2007; Sinkins and Gould, 2006). Wolbachia-based suppression and/or replacement approaches have been documented with insect vector species (the mosquito species: Culex pipiens species complex, Aedes aegypti, Aedes albopictus) under laboratory, semi-field or field conditions (Atyame et al., 2011; Chambers et al., 2011; Hoffmann et al., 2011; Laven, 1967; Ruang-Areerate and Kittayapong, 2006; Walker et al., 2011; Xi et al., 2005). Most of the above-mentioned examples were based on host transinfection via embryonic cytoplasmic injections. However, such microinjections cannot easily be applied to tsetse flies, because they are viviparous. In this case, infection should be attempted through other methods, for example via maternal intrathoracic injections, as has been done in the case of Aedes aegypti (Ruang-Areerate and Kittayapong, 2006). Considering that Wolbachia can find their way to the germ line by infecting stem cells, intrathoracic injections may represent an efficient alternative method for the transinfection of viviparous species such as Glossina spp. (Fast et al., 2011; Frydman et al., 2006; Sacchi et al., 2010). Finally, Wolbachia mediated CI can be harnessed to drive trypanosome resistant paratransgenic tsetse into natural populations to replace their parasite susceptible counterparts (Alam et al., 2011). Based on the fact that all symbionts are maternally transmitted to tsetse’s progeny, resistance genes expressed by Sodalis can be propagated by CI caused by Wolbachia if perfect inheritance of Sodalis and Wolbachia can be ensured (Aksoy et al., 2008; Aksoy and Weiss, 2007; Rio et al., 2004). Given that natural tsetse populations carry heterogenous Wolbachia infections, naturally infected and uninfected strains can be used to develop novel strains that may be incompatible with indigenous populations. Recolonization of Wolbachia infected lines with modified Sodalis strains conferring parasite resistance can give rise to tsetse lines that may be used in the field to spread desirable phenotypes.

8. Harnessing and controlling pathogens for pest and disease control

The available tools for the prevention, diagnosis and therapy of HAT and AAT are rather limited or inadequate (Holmes and Torr, 1988; Simarro et al., 2008). It is generally accepted that the reduction of tsetse populations or reduction of the flies’ ability to transmit trypanosomes remain the most effective approaches to control disease.

The use of sterile insects as part of an area-wide integrated pest management (AW-IPM) approach is a proven approach in controlling dipteran pests, such as fruit flies and screwworms, as well as in the sustainable eradication of tsetse flies from the Island of Unguja, Zanzibar (Enkerlin, 2005; Klassen and Curtis, 2005; Vreysen et al., 2000; Wyss, 2000). The sustainable removal of the Glossina austeni population from the Unguja Island led to the Pan African Tsetse and Trypanosomosis Eradication Campaign (PATTEC) initiative and to a project aiming to create a zone free of Glossina pallidipes in the Southern Rift Valley of Ethiopia (Alemu et al., 2007; Feldmann et al., 2005). This project was based on the establishment of a laboratory colony of the target species at the Insect Pest Control Laboratory (former Entomology Unit) of the Joint FAO/IAEA Programme of Nuclear Techniques in Food and Agriculture, Seibersdorf, Austria, using pupae obtained from the target field population in Ethiopia. Despite its initial successful establishment in 1996, this colony experienced a steady decline over 2 years and collapsed.

Investigations revealed that up to 85% of both male and female flies had salivary gland hypertrophy (SGH), a syndrome first described in wild populations of G. pallidipes (Burtt, 1945; Whitnall, 1934), but later detected in many tsetse species from different African countries (Ellis and Maudlin, 1987; Gouteux, 1987; Jura et al., 1989; Jura et al., 1988; Minter-Goedbloed and Minter, 1989; Odindo et al., 1982; Otieno et al., 1980; Shaw and Moloo, 1993). Jaenson (1978) was the first to identify a nuclear rod-shaped enveloped DNA virus averaging 70 × 640 nm in size as the causative agent and named as the Salivary Gland Hypertrophic Virus (SGHV). This virus was also associated with testicular degeneration and ovarian abnormalities (Jura et al., 1988; Kokwaro et al., 1990; Sang et al., 1999; Sang et al., 1998) and affected the development, survival, fertility and fecundity of naturally (Feldmann et al., 1992) or experimentally (Jura et al., 1993; Sang et al., 1997) infected flies. In tsetse populations in nature, mother-to-offspring transmission, either trans-ovum or through infected milk glands, is thought to be the most likely mode of transmission of the SGHV (Jura et al., 1989; Sang et al., 1998, 1996). In laboratory maintained flies, horizontal transmission during membrane feeding (Feldmann, 1994) is the most significant route of virus infection, as each tray of blood may be used to feed up to ten successive sets of fly cages.

Due to the negative impact of SGHV on fly health, it caused the collapse of two colonies of G. pallidipes and constrains the expansion of the colony in Ethiopia (Abd-Alla et al., 2010), thus a virus management strategy is required. Management of the virus can be achieved by one of the following approaches, or a combination of: (i) change of the feeding system currently used in colony rearing; (ii) neutralization of the virus released during blood feeding by adding virus-specific antibodies to the blood; (iii) inhibition of viral replication with commercially available antiviral drugs against the DNA polymerase of similar viruses, and (iv) development and application of RNA interference strategies to silence essential virus specific gene(s) (Abd-Alla et al., 2011).

Could Wolbachia-based approaches offer an alternative route for the control of SGHV in tsetse flies? Wolbachia has recently been shown to protect Drosophila melanogaster against infections with RNA viruses (Hedges et al., 2008; Osborne et al., 2009; Teixeira et al., 2008). Although SGHV is a DNA virus, we are currently investigating Wolbachia-SGHV interactions in Glossina species. Interestingly enough, a series of recent studies showed that Wolbachia infections present in insect vector species limit the infection and/ or increase host resistance against several human pathogens. These effects were observed against pathogenic viruses such as Dengue, Chikungunya and West Nile Virus, filarial nematodes and a major eukaryotic microbial pathogen, Plasmodium (Glaser and Meola, 2010; Hughes et al., 2011; Kambris et al., 2009; Moreira et al., 2009; Mousson et al., 2010). Laboratory experiments with flies where the obligate symbiont Wigglesworthia has been eliminated while retaining Sodalis and Wolbachia indicate much greater susceptibility to trypanosome infections (Pais et al., 2008). This finding may suggest that unlike the other insect systems, where newly introduced Wolbachia infections can limit pathogen transmission through host immune induction, tsetse symbiosis is dominated by mutually beneficial interactions dominated by Wigglesworthia. In the absence of Wigglesworthia, various aspects of host physiology, including fecundity and immunity are compromised. Whether Wolbachia limit trypanosome infection in natural Glossina populations is currently under investigation.

A virulent strain of Wolbachia, popcorn, was first described in Drosophila melanogaster as over-replicating bacteria in somatic tissues, which reduce the life span of their hosts (Min and Benzer, 1997). This property was recently harnessed in order to transinfect and reduce the life span of insect vector species, such as Aedes aegypti (McMeniman et al., 2009). It is also worth noting that Wolbachia researchers combined several genetic properties of Wolbachia, such as induction of CI and limiting pathogen infection, in transinfected Aedes aegypti (infected with the wMel Wolbachia strain). It was shown that the wMel infection was able to reduce dengue virus proliferation in the host and when these Wolbachia-infected Aedes aegypti populations were released into nature, in Australia, successfully blocked dengue transmission (Hoffmann et al., 2011; Walker et al., 2011). In principle, a similar approach, perhaps combined with the virulent properties of some Wolbachia strains, could be developed for the control of tsetse flies and trypanosomosis (Alam et al., 2011; Van den Abbeele et al., 2013).

It is also worth noting that recent findings suggest that the Wolbachia strains infecting the Glossina morsitans could become virulent over-replicating bacterial strains, like the wMelPopcorn Wolbachia strain (Min and Benzer, 1997), significantly affecting fitness and fecundity of Glossina hybrid species. Schneider and colleagues suggested that this pathogenic effect may be used as a new and or complementary tool for the control of tsetse flies and trypanosomosis (Schneider et al., 2013).

9. Conclusions

Tsetse flies are vectors of the trypanosomes, which cause African sleeping sickness in humans (HAT) and nagana in animals (AAT). Given that there are no effective vaccines or drugs against this pathogen, control of trypanosomosis currently relies on vector control methods. There is an urgent need for pesticide-free and environment-friendly methods for vector and disease control. The SIT, which has successfully been used in the past to control G. austeni populations, is currently being considered as the method of choice also for the control of other Glossina species in sub-Saharan Africa. Recent developments in the field of Insect-Wolbachia Symbiosis suggest new tools to combat the vector and the diseases. The detection and characterization of Wolbachia infections in both laboratory and natural populations of several Glossina species, as well as the ability of these symbionts to induce CI, to reduce the life span of insect vector species and to interfere with major pathogens strongly suggest that Wolbachia-based sustainable approaches, in combination with SIT or as stand-alone methods, can be developed for the control of Glossina species and trypanosomosis.