Back ground:

This year's dengue fever epidemic in Taiwan is severe, which is the second most serious since 2015. We learned from the news that the Kaohsiung City Government, the National Health Research Institutes, and Kaohsiung Medical University are collaborating on researching the use of Wolbachia to prevent dengue fever. According to the news, they have been conducting research since 2021 and plan to engage in international fieldwork and communication efforts in 2024.  Other news sources indicate that the project was initiated in 2018 and involved cooperation with WMP.


Expert Reactions: 

Prof Luke Alphey, Professor of Genetics, Department of Biology, University of York, said:

Q1. What research evidence and trials are currently known to show that the use of Wolbachia is effective and feasible in preventing dengue fever?

Confusingly, Wolbachia has two different uses in this context. One is to generate sterile males, an approach currently used in Singapore, for example. WMP use the other approach, permanent establishment of Wolbachia in the population. I’ll call these the “sterile-male” and “gene drive” approaches. Since you mentioned WMP, I won’t say much more about the sterile-male version here, except to be aware that if you look for other reports of lab or field use of Wolbachia it might be that other approach.

WMP (and others taking a similar approach but WMP are the leaders) aim to introduce Wolbachia into target populations. Wolbachia is a bacterium, but it acts here as a heritable genetic element. After initial introduction into a target population, it will bias inheritance in its favour, i.e. act as a gene drive (though WMP won’t use that term), to further spread itself in the population until the large majority of individuals are infected. Wolbachia-infected mosquitoes have significantly reduced ability to transmit dengue, so once the Wolbachia is present in most mosquitoes in an area, that area should have less transmission.

Note that although we talk about Wolbachia-infected mosquitoes, Wolbachia is not infectious in the normal sense of being transferred from one individual to a nearby one, instead it is passed from mother to her offspring, being present in the eggs she lays. Though the evolutionary record shows that Wolbachia can be transmitted between species, this is clearly a very rare event relative to the mother-to-eggs transmission.

Effective and feasible:

Numerous trials have shown that Wolbachia can be established in a population through mass release of Wolbachia-infected mosquitoes over a period of time, first in two small areas in Australia, and subsequently in other locations. Not every release has succeeded – in Rio de Janeiro it was speculated that this was because the released mosquitoes were more susceptible to insecticides than the wild population, so an insecticide-resistant version was produced; in other locations the Wolbachia have been found to be sensitive to environmental effects such as temperature fluctuations. Nonetheless, it is my impression that most introductions have been made to work, even if not necessarily first time. So ‘feasible’ is, in my view, fairly well covered.

‘Effective’ has two components:

  • do the Wolbachia establish and persist as expected (in this gene drive version – they should do the opposite in the sterile-male application!)? As noted above, this seems work pretty well - it is not necessarily the case that all failures/difficulties are reported as prominently as the successes, but there have certainly been a number of successes.
  • Does this reduce dengue transmission? For population-suppression measures such as sterile males, it is reasonable (for dengue) to assume that strong suppression, i.e. strong reduction of the target population will lead to significant reduction in transmission. For population-modification approaches such as the Wolbachia/gene-drive approach, it is not so obviously the case that presence of the modification, even in most mosquitoes, will have such an effect. However, in the case of Wolbachia it should, and mathematical models based on lab studies suggest it will. For example, a prominent paper ( concluded that “A mathematical model fitted to the data collected on the wMel strain suggested that wMel could reduce the transmissibility of dengue by 66 to 75%, enough to eliminate dengue in low or moderate transmission settings”. Most encouragingly, a single – but quite large – epidemiological study seemed to confirm this result. This was in Indonesia and concluded ( “The protective efficacy of the intervention was 77.1% (95% CI, 65.3 to 84.9)”. Various anecdotal remarks have been made about dengue reduction in other locales, but this is the only solid study that I am aware of. On the basis of that single – but quite large – study, plus lab studies and theory, I would say yes, establishment of Wolbachia is feasible and can reduce dengue transmission.

Q2. What are the limitations of using Wolbachia to stop dengue fever? What is still unclear about science and needs research to understand? For example, will this biological control method have an impact on insects other than mosquitoes?

There are quite a few unknowns around the use of Wolbachia (gene drive version).

We don’t know the molecular/biochemical basis of either ‘gene drive’ mechanism, i.e. the method (“cytoplasmic incompatibility”, CI) whereby the Wolbachia spreads through a target population, nor the reason that Wolbachia-infected mosquitoes are less able to transmit dengue. We do know that many strains of Wolbachia show CI, and also that wMel does not show strong CI in the insect it was taken from Drosophila melanogaster, even though it does when transplanted to Aedes aegypti. This implies that the host insect can evolve means to overcome CI, i.e. the spreading ability. Without attempting exhaustive detail, it is hard to predict the evolutionary trajectory of this complex mosquito-bacterium-virus-human system [Wolbachia is a bacterium]. We know, for example, that the Wolbachia-based resistance can be overcome by more virus, so that might select for virus strains that replicate to higher titres (concentrations) in humans – but on the other hand there might be downsides to the virus associated with that that prevent such strains spreading. Hard to predict. I don’t think anything like that has been seen so far, but that doesn’t mean much, such evolutionary changes take time, and widespread use of the method, even for fast-evolving things like viruses.

Other viruses – lab studies suggest that Wolbachia also reduces the ability of Aedes aegypti to transmit several other relevant viruses, such as Zika (e.g. However, it seems to increase infection by some related insect-specific viruses (, so it is hard to predict what the effect might be on some future epidemic caused by a currently-unknown virus. That’s important, because there have been two major epidemics this century of viruses transmitted by Aedes aegypti – Zika and chikungunya viruses, and one might reasonably anticipate that there might be further such epidemics in the future. One might argue that an advantage of population-suppression methods – including the sterile-male version of Wolbachia is that by removing the mosquito, all possibility of transmission by that mosquito is removed, even for entirely unknown viruses.

Regarding impact on other insects – or other elements of the ecosystem – than the intended targets (Aedes aegypti, Wolbachia, dengue virus), I think here we can be more confident. It is very hard to see any route to harm. Wolbachia could ‘jump’ from Aedes aegypti to another species, but this is highly unlikely, and probably not very harmful if it did. The relevant strain, wMel, is already present in D. melanogaster, a widespread, cosmopolitan species, so there is some environmental exposure already. A population-suppression approach would remove this invasive species, which might be considered bio-remediation, whereas the Wolbachia population-modification approach keeps it in place – but gives better protection against subsequent (re-)invasion by more Aedes aegypti. Both approaches have great potential, though they are different it’s hard to say that one is unequivocally better than the other.

Q3. What infectious disease and public health regulations are needed to prevent dengue fever with Wolbachia in local areas?

Different countries/communities will do things differently. On the whole, I think genetic-based control, for example Wolbachia-based methods, can reasonably easily incorporated into existing programmes without major changes to existing practice.

Communication/community-engagement may be more important – the idea of combating mosquito-borne disease by releasing more mosquitoes may not be a natural one to many people!

Q4. At what stage should such research be communicated to the public, local areas or the media?

That’s a much broader question – and SMC may have its own views on that!

Declare conflicts of interest: 

Prof Alphey co-founded Oxitec in 2002 and was a director until Oxitec was acquired by Intrexon in 2015. He no longer has any financial interest in Oxitec. Prof Alphey is an adviser to Biocentis Ltd and Synvect Inc, with financial and/or equity interest. Each of these companies is developing genetics-based methods for control of pest insects.

Prof Austin Burt, Professor of Evolutionary Genetics, Faculty of Natural Sciences, Department of Life Sciences (Silwood Park), Imperial College London, said:

There are 2 different ways of using Wolbachia to control dengue.

(1) Release males carrying the Wolbachia bacteria into a population that does not contain the bacteria. Matings between the released males and the natural females will produce no offspring, and so the released males are effectively sterile. This has been used as a trial in California (, and is being used in Singapore (

(2) Release females (and potentially males) carrying the Wolbachia bacteria in such a way that the bacteria becomes established in the target population and all (or the vast majority) of females have it. For reasons that are a little unclear (at least to me), females with the baceteria are less able to transmit dengue viruses. So, this approach is not reducing the numbers of mosquitoes, but changing them to no longer be able to transmit the disease. There has been good success in Indonesia ( and in Malaysia ( with 2 different Wolbachia strains.

I’m not sure which approach is being investigated for Taiwan.

The approach needs to be tried more to see the circumstances in which it will work and when it won’t. I’m not aware of any significant risks associated with the technology (other than it may not work). I’m also not aware of how much it costs.

Declare conflicts of interest: 

Prof Austin Burt is Principal Investigator for Target Malaria, a not-for-profit research consortium focussed on developing and testing gene drive mosquitoes for malaria control in sub-Saharan Africa (