A Wolbachia wMel Transinfection in Aedes albopictus Is Not Detrimental to Host Fitness and Inhibits Chikungunya Virus

Background Wolbachia inherited intracellular bacteria can manipulate the reproduction of their insect hosts through cytoplasmic incompatibility (CI), and certain strains have also been shown to inhibit the replication or dissemination of viruses. Wolbachia strains also vary in their relative fitness effects on their hosts and this is a particularly important consideration with respect to the potential of newly created transinfections for use in disease control. Methodology/Principal Findings In Aedes albopictus mosquitoes transinfected with the wMel strain from Drosophila melanogaster, which we previously reported to be unable to transmit dengue in lab challenges, no significant detrimental effects were observed on egg hatch rate, fecundity, adult longevity or male mating competitiveness. All these parameters influence the population dynamics of Wolbachia, and the data presented are favourable with respect to the aim of taking wMel to high population frequency. Challenge with the chikungunya (CHIKV) virus, for which Ae. albopictus is an important vector, was conducted and the presence of wMel abolished CHIKV dissemination to the saliva. Conclusions/significance Taken together, these data suggest that introducing wMel into natural Ae. albopictus populations using bidirectional CI could be an efficient strategy for preventing or reducing the transmission of arboviruses by this species.


Introduction
Aedes albopictus is a medically important invasive mosquito that has expanded from Asia into many new regions of the tropics and also into temperate regions of the Americas and Europe; it transmits a number of arboviruses including dengue and chikungunya [1]. All known wild populations of Ae. albopictus are naturally superinfected at very high frequency with two strains of the maternally inherited bacterium Wolbachia pipientis, called wAlbA and wAlbB [2,3]. Recently, we reported the generation of an Ae. albopictus line transinfected with the wMel strain of Wolbachia from Drosophila melanogaster, which proved unable to transmit dengue virus in lab challenges and induced complete bidirectional CI when crossed with a naturally wAlbA plus wAlbB (wAlbA/B)-infected line [4]. The wMel strain, and the related wMelPop [5] have also been shown to strongly inhibit viral replication in Aedes aegypti, as well as causing chronic immune upregulation in that naturally Wolbachia-uninfected host [6][7][8].
If wMel is to be used to reduce the dengue transmission capacity of wild Ae. albopictus, a critical factor is whether mosquitoes infected with this strain incur major fitness penalties. It has previously been reported that the natural wAlbA/B strain superinfection actually provides a fitness advantage in the form of increased longevity and fecundity compared to an uninfected line [9,10]; given that wMel is not specifically adapted to Ae. albopictus and vice versa, significant fitness costs compared to wildtypes are possible. The overreplicating wMelPop strain variant is virulent and can approximately halve the lifespan of both its D. melanogaster and Ae. aegypti hosts [11,12], and a wMelPop transinfection in Ae. albopictus produced a greatly reduced egg hatch from intra-strain matings [13]. Population dynamic mathematical models have emphasised the importance of fitness parameters in Wolbachia-based population replacement strategies [14,15]. It is therefore very important to gain a comprehensive understanding of any fitness consequences of replacing wAlbA/B with wMel in Ae. albopictus, including longevity of both male and female mosquitoes, fecundity and egg fertility rates for females and relative mating competitiveness for males.
Chikungunya virus (CHIKV) is viewed as an important potential emerging or re-emerging pathogen in many parts of the world [15,16]. There have been a number of recent epidemics where Ae. albopictus has been incriminated as a major vector or as the sole vector, for example in the Indian Ocean (La Reunion), Italy, southeastern France, central Africa, and the Kerala region of India [17][18][19][20][21][22][23][24]. It has been of particular concern that two viral adaptations, single amino acid substitutions in the envelope glycoproteins (E1-A226V which arose around 2005 and E2-L210Q identified in 2009), have significantly improved the efficiency of dissemination and transmission by Ae. albopictus [25][26][27][28]. Therefore another major aim of this study was to assess whether the wMel-transinfected line shows any impairment in its ability to transmit CHIKV, in light of the strong inhibition of dengue virus that has already been observed in this line [4], and that fact that a wMelPop Wolbachia transinfection in Ae. aegypti has been shown to produce an inhibitory effect with respect to CHIKV [7].

Mosquito strains
The Uju.wMel line [4] was generated by microinjection of wMel from D. melanogaster into a Wolbachia-uninfected (cured) Ae. albopictus strain UjuT. Uju.wt was generated by introgression of the natural wAlbA/B infection of Ae. albopictus from the Ascoli strain (Italy) into UjuT background for four generations [4] followed by a further three here; the resulting line was .99% UjuT nuclear background and contained both wAlbA and wAlbB. All Ae. albopictus colonies in Oxford were maintained at 27uC and 70% relative humidity on a 12-h light/dark cycle.

Fitness and male mating competitiveness
Females were blood fed at six days post eclosion, and aspirated into small tubes for individualized laying two days later. Eggs were matured for five days before being counted to give fecundity data and then hatched in deoxygenated water in small plastic vials. Second instar larvae were counted to give hatch rate. Uju.wMel female egg hatch was lower than for wildtype colony controls in early generations, likely as a result of variable Wolbachia density causing CI effects. Therefore, between G 6 and G 12 selection for high hatch was applied to make the Wolbachia density more uniform, by discarding approximately one quarter of broods which showed the lowest hatch rates.
Longevity was assessed in 30 cm630 cm630 cm BugDorm cages (MegaView Ltd., Taiwan) containing one hundred male and one hundred female mosquitoes (which were sexed as pupae). The mosquitoes were provided with moist cotton and supplied with sucrose ad libitum. A blood meal was provided at seven days, and subsequently every fourteen days; oviposition substrates were provided three days after blood feeding. Dead mosquitoes were counted and removed every four days. To assess developmental times, eggs were hatched and larvae reared at 100 larvae/litre and given 1 mg liver powder per larva per day; the number of adults eclosing per day was counted until all pupae eclosed.
Male mating competitiveness was assessed using three independent replicates of 50 male Uju.wMel : 50 male Uju.wt : 50 females (either Uju.wMel or Uju.wt), sexed at the pupal stage. Adults were allowed to emerge in Bugdorm cages and were left to mate for ten days. Females were then blood fed, and two days later were aspirated into small plastic vials for individualized laying. Eggs were hatched five days after laying; given the complete CI (zero egg hatch) observed in both directions in crosses between wildtype and wMel-transinfected parents [4], the male parent was scored according to whether or not embryos hatched -hatching indicating that both contained the same Wolbachia strain. Spermathecae were dissected and examined under a dissecting microscope for the presence of stored sperm for all females that produced eggs with zero hatch, and data from any females without sperm were disregarded.

Viral strain
CHIKV E1-226V was isolated from a patient on La Reunion Island in November 2005 [23]. This strain is characterized by a substitution of alanine by valine at position 226 of the E1 envelop glycoprotein in a region which is responsible for fusion of viral and cellular membranes within the endosome [29]. This anonymized strain was provided by the French National Reference Center for Arboviruses at the Institut Pasteur. Viral stocks were produced on Ae. albopictus C6/36 cells that were infected at a multiplicity of infection (MOI) of 0.1 PFU (plaque forming units)/cell for 48 h at 27uC. Stocks were constituted after two passages on C6/36 cells, titrated on Vero cells, and stored at 280uC until use.

Mosquito infections
Mosquito adults were starved for 24 hours and then allowed to feed on artificial blood-meals consisting of a virus suspension (1:3), washed rabbit erythrocytes (2:3), and 5 mM ATP. The blood-meal was provided in glass feeders covered with a pig intestine membrane and maintained at 37uC. The titer of the blood-meal was 10 7.5 PFU/mL. Females placed in plastic boxes were allowed to feed for 15 min. Engorged females were sorted on ice, transferred in cardboard containers, provided with sugar solution and maintained in BSL-3 insectaries at 28uC until day 7 postinfection (in line with previous experiments showing that viral titres gradually decreased after day 7 post infection [30])..
To evaluate transmission capacity, 35-50 females were anaesthetized on ice, their wings and legs were removed to induce stress and force salivation. The proboscis was inserted into a 20 mL tip containing 10 mL of foetal bovine serum. After 45 minutes, saliva was collected with the medium and expelled under pressure into a 0.5 mL tube containing 40 mL of DMEM. Samples were later titrated by focus fluorescent assay [31] on Ae. albopictus C6/36 cells. The transmission rate was estimated as the percentage of mosquitoes with infectious saliva among tested mosquitoes.
Saliva samples were titrated by focus fluorescent assay on C6/ 36 Aedes albopictus cell culture. Cells were grown to confluent monolayers in 96-well plates and infected with 10-fold serial dilutions of virus. After incubation at 28uC for 3 days, plates were stained using hyper-immune ascetic fluid specific to CHIKV (provided by the French National Reference Center of Arboviruses at the Institut Pasteur) as the primary antibody and a conjugate anti-IgG mouse (Bio-Rad) as the secondary antibody. The focus

Author Summary
The tiger mosquito Aedes albopictus is an invasive disease vector whose range has expanded throughout the tropics, and some temperate regions, in recent decades from its native South East Asia. It is an important vector of human viruses including dengue and chikungunya; in recent years a mutation has been detected in chikungunya virus that specifically increases transmission efficiency by Ae. albopictus, causing concern that epidemics of this disease will become more widespread and severe. Here we show that when transinfected with a strain of the symbiont Wolbachia called wMel, originating in fruitflies, the ability of the mosquito to transmit the mutated chikungunya virus was abolished in lab experiments. Furthermore, the wMel strain was shown to produce no detectable fitness costs in this new host, examining numbers of eggs produced, egg hatch, lifespan, and male mating competitiveness compared to wildtypes. This is encouraging with respect to developing the system for use in the control of dengue and chikungunya viruses. fluorescent units were counted with a fluorescent microscope and the titer was calculated as FFU/saliva.

Uju.wMel fitness parameters and male mating competitiveness
The egg hatch rate and the fecundity per single gonotrophic cycle observed in the wMel-transinfected Ae. albopictus line Uju.wMel [4] were not significantly different to the natural wAlbA/B-superinfected strain Uju.wt (hatch rate, p = 0.369, Wilcoxon rank sum test; fecundity, p = 0.738, Wilcoxon rank sum test) ( figure 1A & B). This is consistent with observations in other novel transinfections over time [32][33][34]. Additionally, the fecundity of Uju.wt was significantly higher than UjuT (p = 0.028, Wilcoxon rank sum test), supporting previous reports of increased fecundity in the presence of the natural Ae. albopictus Wolbachia superinfection compared to a cured Wolbachia-uninfected line [9,10]. As with all characterization experiments presented here, the three lines used (UjuT, Uju.wMel and Uju.wt) contained over 99% identical nuclear background, following introgression with UjuT males, in order to control for any genetic differences between the lines that may have arisen through bottlenecking. The fecundity of the Uju.wMel line contrasts with the significant fecundity reduction previously observed for the wPip transinfection of Ae. albopictus [35], and the hatch rate is much higher in the wMel transinfected line than that previously observed for a wMelPop strain transinfection in Ae. albopictus, which averaged in the 10-20% range [13].
Neither wMel nor the natural wAlbA/B infections appear to have an effect on the longevity of female Ae. albopictus in these lab cage conditions ( figure 2A). However, the longevity of Uju.wMel males was significantly increased compared to both the uninfected UjuT (p = 0, log rank test) and the naturally superinfected Uju.wt (p = 0, log rank test) (figure 2B). The average male lifespan was increased from 33.4 days (UjuT) or 33.6 days (Uju.wt) to 52.5 days (Uju.wMel). The reason for this increase in male lifespan is unknown. The lack of a significant difference in longevity in either sex between Uju.wt and UjuT contrasts with a previous report [9]. This difference between studies could have been a result of the seven generations of introgression that were used here to homogenize genetic backgrounds. No difference was found between the larval/ pupal development times of Uju.wMel and Uju.wt; the mean time taken to eclose from egg hatch was 10.8561.18 days (n = 505) and 10.8761.52 days (n = 951) for Uju.wMel and Uju.wt respectively (p = 0.931 Log rank test; error represents one standard deviation).
Male mating competitiveness of Uju.wMel males was compared to that of the naturally infected Uju.wt males using three  independent replicates of 50 male Uju.wMel : 50 male Uju.wt : 50 females (either Uju.wMel or Uju.wt). Hatching eggs from an individual female indicated that a compatible mating had occurred with a male with the same Wolbachia strain, whilst no hatch indicated an incompatible cross (data from females with empty spermathecae were disregarded). No significant difference was found between the competitiveness of the wildtype and transinfected males under these conditions (p = 0.652, Chi-squared analysis using a likelihood framework) ( figure 3).

CHIKV susceptibility
The effect of wMel on CHIKV inhibition was assessed by allowing adult females to feed on blood-meals containing a virus suspension, washed rabbit erythrocytes and 5 mM ATP. After 7 days, 35-50 females were anaesthetized on ice and salivation was induced. The saliva was titrated by focus fluorescent assay on Ae. albopictus C6/36 cells. CHIKV was present in 52% and 44% of Uju.wt and UjuT individuals' saliva, whilst no CHIKV was found in any Uju.wMel saliva (figure 4A), demonstrating complete viral inhibition caused by wMel. No significant difference was observed between the viral titers found in Uju.wt and UjuT (p = 0.4129, Wilcoxon rank sum test) (figure 4B), suggesting that the presence/ absence of wAlbA/B has no effect on CHIKV. This supports previous research showing no overall significant difference in CHIKV titers between wAlbA/B infected and uninfected Ae. albopictus [36].

Discussion
In many regions of the world the most important disease transmission risk associated with Ae. albopictus is increasingly chikungunya [16,17], and the relative importance of its transmission role compared to Ae. aegypti is probably greater for this disease than for dengue [18][19][20][21][22][23]. The data presented here provides a clear indication that the introduction to high frequency of the wMel transinfection in wild populations of Ae. albopictus should prevent it from acting as a chikungunya vector in those populations, in addition to a likely impact on the transmission of dengue. The strong inhibition of CHIKV by Ae. albopictus is noteworthy in light of the fact that the viral strain used possesses the envelope glycoprotein E1-A226V mutation, which has greatly increased the transmission potential by Ae. albopictus [25][26][27][28]. It would be useful in future studies to conduct challenges with other viral strains, particularly one that carries the E2-L210Q mutation, although this does not produce as strong a positive effect on viral fitness in Ae. albopictus as the E1-A226V variant does [27]; thus the expectation is that the degree of inhibition of other CHIKV variants produced by wMel would be at least as strong as that observed here. It was recently reported that the inhibition of CHIKV also occurs in the presence of wMel transinfection in Ae. aegypti [37], and thus the use of Wolbachia offers considerable promise for future preventative control measures against this emerging disease.
The mechanisms of viral resistance conferred by Wolbachia have yet to be fully elucidated. It has been suggested that a reactive oxygen species (ROS)-dependent activation of the Toll pathway, and subsequent increased expression of antimicrobial peptides, is responsible or plays a major role in wAlbB-transinfected Ae. aegypti [38]. However, no evidence of increased antimicrobial peptide expression has been found in Uju.wMel [4]. These experiments included qRT-PCR of both a defensin and a cecropin, which were both highlighted as important effectors in Ae. aegypti infected with wAlbB [38]. Whilst the involvement of immune genes has not been ruled out, other possible pathogen inhibition mechanisms include direct production of ROS by Wolbachia, as observed for other bacteria [39], and competition for resources such as cholesterol between the host, the bacterium and the virus [7]. Interestingly, high experimental CHIKV titers have previously been shown to produce a reduction in density of wildtype wAlbA/B Wolbachia in Ae. albopictus [36], which supports the hypothesis of resource competition between them. It will be important, with respect to the effectiveness of disease control strategies, to examine whether 100% blockage of viral transmission by wMel is maintained after many generations of carriage of this Wolbachia strain.
We previously reported that Uju.wMel is fully bidirectionally incompatible with a wAlbA/B wildtype line, which provides a mechanism for contained replacement of the naturally occurring wildtype Wolbachia strains in natural populations, since the two could not stably co-exist within a population as long as they are freely interbreeding, and the majority strain would replace the minority [4]. A mathematical model [15] provides an example of the introduction of Wolbachia into discrete seasonal populations of mosquitoes. It is clearly undesirable to release large numbers of biting females to achieve a population majority in a target population. However, by releasing a high percentage of males (the majority of adult males emerge before the females and pupae can also be separated into sexes by size) and beginning releases at the start of the rainy season, a relatively modest program of releases over a period of months should be sufficient to take a Wolbachia strain that produces complete bidirectional CI with wildtypes to fixation in a population if there are no major fitness costs of the transinfection under field conditions. At the same time there would be the benefit of suppressed overall population size of biting females due to the releases of incompatible males. Once established at 100% frequency, the transinfection would be resistant to moderate levels of immigration of wildtypes from surrounding populations, since immigrating wildtype females would be incompatible with a majority of the males they encountered.
The fitness parameters that have been measured here are all favorable with respect to the aim of introducing the wMel transinfection into natural Ae. albopictus populations for the purpose of disease prevention. The wMel-transinfected males were fully competitive with wildtype males in the simple cage mating assays employed. Extrapolations to what will occur under natural conditions must be made with caution. Semi-field contained greenhouse trials would give a better approximation to nature for longevity and mating competitiveness assays than insectary cages. Ultimately replacement dynamics can be fully assessed in open field trials; the relative effects on fitness of wMel in the greater genetic heterogeneity of wild populations is another unknown. Inn general however, small-cage lab assays have been found to be a reasonably reliable indicator of field success in the Aedes release programs conducted to date [40,41]. Detailed examinations of the relative effects of Wolbachia in Ae. albopictus on other life history parameters, particularly using different larval rearing conditions and examining effects on adult size, developmental time and larval and adult competitiveness, and also lifetime measures of fecundity, would also be useful future laboratory-based studies.
Even if significant reductions in the mating competitiveness of wMel-infected males occurred under field conditions, this could be abrogated by a sufficiently high ratio of released males to wild type males. Various scenarios can be envisaged where the long-term effectiveness of such a Wolbachia-based strategy for disease control might gradually become compromised, such as the occurrence and spread of viral mutations that escape the inhibition, but such mutations are by no means inevitable. Based on all the data collected so far, the use of wMel offers a promising new method to prevent or reduce dengue and chikungunya transmission by Ae. albopictus.