Infection with a harmless bacterium makes the mosquitoes that transmit dengue virus resistant to viral infection. The resistant population can rapidly replace the natural, susceptible population. See Letters p.450 & p.454
Dengue fever is the most prevalent mosquito-borne viral disease in humans. The virus is transmitted mainly by the mosquito Aedes aegypti, and so targeting this insect has been considered a viable option for controlling the incidence of the disease. In two papers in this issue (Walker et al.1 and Hoffmann et al.2), a team of researchers reports an unusual approach for making A. aegypti almost completely resistant to infection, and thereby blocking transmission of dengue virus. Moreover, when the authors released their dengue-resistant mosquitoes into the wild, the insects replaced nearly 100% of the natural, susceptible mosquito population within a matter of months.
Around 40% of the world's population is at risk of infection with dengue virus. Every year, the virus infects 50 million to 100 million people, causing classical dengue fever as well as more severe symptoms such as dengue haemorrhagic fever and dengue shock syndrome. In the absence of an effective vaccine, controlling dengue is limited to targeting the mosquitoes that transmit the virus3. Many mosquito-control strategies are based on suppressing or eliminating the insect population. By contrast, population-replacement strategies aim to replace the pathogen-susceptible mosquito population with a resistant one4.
The bacterium Wolbachia is a common endosymbiotic associate of many insects, including mosquitoes, and lives inside their cells. It is maternally inherited, and manipulates the reproduction of its invertebrate hosts in various ways to maximize the number of infected females in the next generation; this allows Wolbachia to spread rapidly through populations5. Infection with the bacterium can make insects such as mosquitoes resistant to infection by many pathogens, including viruses, malaria parasites and filarial nematodes6. If Wolbachia is artificially introduced into an uninfected vector population, it can spread and replace the wild-type population with one in which pathogen transmission is blocked5.
Researchers showed7 a few years ago that wMelPop — a virulent Wolbachia strain from a laboratory colony of the fruitfly Drosophila melanogaster — almost completely blocked dengue infection in A. aegypti. This Wolbachia strain, however, severely affected the mosquitoes' fitness, making its spread into natural A. aegypti populations difficult or even impossible8.
Going back to the source, Walker et al.1 (page 450) noted that wMel, the natural, avirulent Wolbachia strain in Drosophila, can inhibit fly infection with RNA viruses. As dengue is an RNA virus, the authors reasoned that wMel might also block infection with this virus in mosquitoes, without having the virulent effects of wMelPop. Lo and behold, when they infected A. aegypti with wMel, the mosquitoes became highly resistant to infection by dengue virus. In experiments on caged mosquitoes, Walker and colleagues further showed that the wMel infection could spread through laboratory mosquito populations at a rate consistent with predictions based on theoretical modelling.
Many studies would stop there. But the investigators took the astonishing next step of attempting a release of Wolbachia-infected mosquitoes into the wild, and, even more astonishingly, they did it in their own backyard. Queensland, Australia, has a recurring dengue problem9, and so seemed an ideal location for an initial test release.
After extensive public engagement, and development of a regulatory framework governing the release of the Wolbachia-infected insects, in January this year Hoffmann et al.2 (page 454) began releasing wMel-infected mosquitoes at two locations — Yorkeys Knob and Gordonvale, both near Cairns in Queensland (Fig. 1). The authors continued to release the insects at regular intervals over the next two and a half months, with a total of roughly 150,000 mosquitoes released at each location. The frequency of Wolbachia infection increased extensively during the releases, but more crucially, continued to climb in both areas after releases were stopped, approaching 100% in Yorkeys Knob and more than 80% in Gordonvale2.
Hoffmann and colleagues estimated that wMel exerts only a moderate (10–20%) fitness cost on the mosquitoes it infects. This suggested that a 'wave' of infection with this bacterial strain should eventually spread from local areas in which the infection was introduced. And spread it did — infected mosquitoes were detected several kilometres away from both release areas, probably owing to occasional long-distance movement of infected insects. Although in these experiments the infection is unlikely to increase in frequency outside the release area, but rather will be swamped by the local wild-type mosquitoes, the data suggest that long-range spread is possible2.
These studies mark the first time that a deliberate Wolbachia-mediated population-replacement strategy has been attempted in nature, and herald the beginning of a new era in the control of mosquito-borne diseases. The advantage of population-replacement approaches is that, once established, they are self-propagating. And because the mosquito population is simply changed rather than eliminated, effects on the ecosystem should be minimal4.
Theoretically, these strategies can also be applied to other vector-borne diseases, including malaria10. The next step will be to attempt release of Wolbachia-infected mosquitoes in an area with endemic, rather than sporadic, dengue transmission, such as southeast Asia or South America. It remains to be seen whether the introduction and spread of Wolbachia can be achieved across the highly variable worldwide range of dengue virus, and whether Wolbachia can provide consistent protection against viral strains of different genetic make-up. Nevertheless, these experiments1,2 are a groundbreaking first step towards Wolbachia-mediated replacement of A. aegypti and so elimination of the scourge of dengue fever.
Walker, T. et al. Nature 476, 450–453 (2011).
Hoffmann, A. A. et al. Nature 476, 454–457 (2011).
Morrison, A. C. et al. PLoS Negl. Trop. Dis. 4, e670 (2010).
James, A. A. Trends Parasitol. 21, 64–67 (2005).
Rasgon, J. L. Adv. Exp. Med. Biol. 627, 114–125 (2008).
Iturbe-Ormaetxe, I., Walker, T. & O'Neill, S. L. EMBO Rep. 12, 508–518 (2011).
McMeniman, C. J. et al. Science 323, 141–144 (2009).
McMeniman, C. J. & O'Neill, S. L. PLoS Negl. Trop. Dis. 4, e748 (2010).
Hanna, J. N. et al. Aust. NZ J. Public Health 30, 220–225 (2006).
Hughes, G. L., Koga, R., Xue, P., Fukatsu, T. & Rasgon, J. L. PLoS Pathog. 7, e1002043 (2011).
About this article
Developmental & Comparative Immunology (2014)
Transinfection: a method to investigateWolbachia-host interactions and control arthropod-borne disease
Insect Molecular Biology (2014)
Current Opinion in Insect Science (2014)
Analysis of Effects of Meteorological Factors on Dengue Incidence in Sri Lanka Using Time Series Data
PLoS ONE (2013)