Special Viewpoints

Malaria Control with Genetically Manipulated Insect Vectors

See allHide authors and affiliations

Science  04 Oct 2002:
Vol. 298, Issue 5591, pp. 119-121
DOI: 10.1126/science.1078278


At a recent workshop, experts discussed the benefits, risks, and research priorities associated with using genetically manipulated insects in the control of vector-borne diseases.

This is a partial report of a workshop—Genetically Engineered Arthropod Vectors of Human Infectious Diseases—jointly sponsored by the World Health Organization, the MacArthur Foundation, the National Institute of Allergy and Infectious Diseases, and London's Imperial College—originally planned for 12 September 2001 in London (but reconvened in successive sessions later in London and Atlanta). These workshops sought to encourage communication between the laboratory-oriented molecular biologists, whose work had suggested the potential of genetic control strategies, and the population geneticists, ecologists, and public health specialists, whose involvement would be crucial in moving the work beyond the laboratory. The meeting participants were charged with considering the benefits and risks of using genetically engineered arthropod vectors as public health tools and mapping out a research agenda for their development. The task of engineering different vector species and the risks associated with various methods of genetic engineering are vastly different and could not be addressed in a single report. What follows is the consensus of the working group on germline-transformed organisms developed for control of malaria transmission (authors listed above) and other participants. The reports of the working groups on paratransgenesis (transformation of obligate symbionts in insects) and on other vector-borne diseases will be presented in the near future.

In 1991, a scientific workshop in Arizona assessed the prospect for malaria control by genetic manipulation of vector populations (see the Viewpoint by Morel et al. on page 79) (1). The basic concept of genetic control of vector-borne diseases was proposed by Curtis in 1968 (2), but major advances in the molecular manipulation of Drosophila melanogaster during the 1980s encouraged reevaluation of this idea. The WHO/TDR summary document of the meeting laid out a clear list of research aims that would have to be met before a genetic control strategy could be field tested (3). These aims fell into three categories: (i) the development of genetic engineering tools that could be used with malaria vectors; (ii) the identification of effector genes that could block parasite transmission; and (iii) the development of effective methods for driving these effector genes to fixation in natural vector populations.

The first two aims have been largely achieved. Several different but effective methods of germline transformation have been developed and used in at least three species of malaria mosquito vector (4–6); two different laboratories have developed genetic constructs that significantly reduce vector competence in experimental malaria models (7, 8). A large set of molecular markers has been developed and is being used in studies of gene flow and population structure in anopheline malaria vectors (9–13). But there has been no significant progress in developing methods for driving desirable genes into wild populations and especially for ensuring the necessary unbreakable linkage between the drive system and the gene to be driven (see the Viewpoint by Scott et al. on page 117) (14).

Consideration of the potential use of genetically modified organisms (GMOs) is driven by the realization of the enormous human cost of diseases like malaria, and of the inadequacy of present control measures. Perhaps the most important theme emerging from the workshop was the recognition that control strategies involving GMOs could potentially provoke serious public mistrust and resistance to their implementation. Therefore it was strongly recommended that all work leading to the development of specific genetic control strategies targeted at malaria vectors should involve both public health specialists and scientists from disease-endemic countries and (where possible) the general public in areas where field trials could be implemented. Because field trials of genetically modified mosquitoes would have to be preceded by long-term, longitudinal studies of potential field-trial sites, the local community and its own scientists and health experts can easily be involved.

The goal of producing GMOs intended to benefit human health has been perceived more favorably by the public than that of producing GMOs for agricultural or domestic animal research. However, meeting participants strongly argued that this positive public perception could be rapidly undermined by an actual field trial of a transgenic arthropod that failed to provide a significant and tangible health benefit to the resident human community. It was therefore recommended that all preliminary research designed to lead to field trials of the efficacy of a transgenic arthropod-based disease control strategy should involve fully contained laboratory or cage environments. Release should be permitted only when all relevant parameters had been investigated in either contained environments or in open field studies that did not involve transgenic arthropods. Furthermore, field trials involving release of transgenic arthropods should take place only when all members of both scientific and local community review groups were assured that such trials had a very high probability of producing a significant and measurable public health benefit for the local community.

Many important ecological and population genetic issues must be understood before any release program can be contemplated, and such issues will be specific not only to individual vector species but also to local populations (see the Viewpoint by Scott et al. on page 117) (14). Understanding the dynamics of a natural population will require years of study, with the time frame dependent on the stability and repeatability of yearly cycles. Thus, given progress in the laboratory, it is important to start the ecological and population genetic study of potential target populations soon, as this will be the biggest scientific limitation to implementing genetic control field trials. A large number of technical problems will have to be addressed, ranging from the feasibility of producing an effective release strain to the design and assessment of release strategies with specifically predicted goals. To address such problems will require the involvement of ecologists and population geneticists. Most participants recommended that study of potential field-trial sites should be initiated immediately at multiple different locations, recognizing that the initial phase of fieldwork might show one or more of the selected sites to be unsuitable. Because the biology of vector populations at any such site would have to be studied for many years before field trials could be designed, the community cannot investigate different sites sequentially.

GMOs could be used in either of two ways for malaria control. The initial concept (expressed in the 1991 meeting) was to engineer mosquitoes with an altered phenotype that would be introduced into the population in such a way that the new trait would spread and become dominant. These strategies target the malaria parasite, rather than the mosquito itself, for reduction. There is an immediate research need for the study of drive systems in Anopheles species. These drive systems also present a potential hazard because they may generate unintended phenotypes and have unforeseen, potentially harmful ecological effects. Autonomous transposons, for example, could increase the mutation rate through multiple genomic insertions, leading to unanticipated alterations in the biology of the target species. Tight linkage of the drive system and the engineered gene is also an important issue in that its loss in the progeny of released mosquitoes could lead to loss of public health efficacy and loss of the molecular tool for future engineering efforts. Although transposon and symbiont systems have garnered the most attention to date, participants recognized the need to explore any possible drive system that could continue to propagate a released genetic construct through the target population after initial release.

An alternative use of genetic engineering for malaria control takes a more traditional approach. This involves targeting the mosquito population per se for reduction. Proposed improvements in sterile insect techniques, including release of insects carrying dominant lethals (RIDL) (15), and other mechanisms of genetic sexing may alter the prognosis for these strategies. In these situations the release of large numbers of insects presents other specific challenges: for example, the need to release only male mosquitoes so as not to increase the number or nature of mosquito bites per person per night. In the absence of an existing drive system, participants considered the use of inundative release of refractory mosquitoes as a strategy for limited field-testing of the performance of specific genetically engineered vector strains. Although considered suitable only for a small vector population with limited interpopulation gene flow (such as a real or ecological island setting), the ability to limit or quickly control unforeseen risks in the genetic manipulation of an island population will be important in early-stage trials designed to demonstrate the efficacy of particular genetic modifications of the vector population.

Although there was support for continued, intensive research in this area, a clear recommendation emerged that there should be no precipitous releases of transgenic arthropods. The malaria group was willing to recommend barring field trials of transgenic insects that were designed solely for research; others felt that initial field safety testing of the various individual elements of the engineered organism was crucial to development. The parallel processes of drug and vaccine development illustrate these two views. For either product, and indeed for engineered Anopheles mosquitoes, there is a requirement for preliminary studies of safety and efficacy in culture and in animal models before the first clinical trial is initiated. With many new drugs (other than cancer drugs), the first human trials are performed in small numbers of normal healthy volunteers, and safety is the end point examined. In these situations it would be inappropriate to endanger patients who are already sick by exposing them to a drug candidate of unknown toxicity. By contrast, when new vaccines are developed, they are most often combined with adjuvants that improve their potency or direct their effects to one or more segments of the human immune system. Under its current guidelines the U.S. Food and Drug Administration does not allow investigation of the adjuvants alone without the vaccine candidate being tested at the same time. The malaria working group requires tangible benefits at each phase of field testing. The other working groups—discussing symbionts, transducing viruses, and other mechanisms of driving traits into populations—decided to follow drug-development protocols. These differences may be appropriate given the different nature of the engineering tools and the different risks associated with each one.

Despite nearly universal recognition that enormous technical and sociological problems must be overcome before the implementation of genetic control strategies for malaria can be field tested, participants concluded that public health strategies incorporating transgenic vectors offer the potential of health benefits. Participants from disease-endemic areas, many of whom had limited prior exposure to transgenic arthropod research or policy discussions, were among the most supportive and optimistic about the public health goals such strategies hope to achieve. Participants also noted that the broad scope of biological research required for the development of genetic control strategies is likely to contribute both to the more efficient application of currently available control tools and to the development of new approaches.


View Abstract

Stay Connected to Science

Navigate This Article