In recent years, researchers have sequenced the genomes of several Anopheles mosquito species, including those responsible for nearly all of the malaria transmission in Africa. With this information, they have begun to identify the genes underlying the insects’ ability to colonize human habitats, their reproductive biology, and their susceptibility to infection by the malaria parasite (Plasmodium spp.). If we know the genes, or variants of genes, that are responsible for key mosquito traits, such as parasite clearance or egg laying, we can theoretically introduce a genetic modification into the insects that reduces malaria transmission.

The idea has been percolating in the minds of scientists for nearly 20 years, ever since researchers developed a way to introduce genes into mosquitoes. But one challenge has hindered all such efforts to date: how to encourage the spread of a gene modification from a few lab-reared mosquitoes...

Gene drives use a variety of tricks to ensure that they, or the chromosomes that contain them, are selectively inher­ited.

In sexually reproducing organisms that have two copies of every chromosome, any single copy of a gene normally has a 50?percent chance of being passed on to an individual’s offspring. Gene drives, on the other hand, use a variety of tricks to ensure that the genetic elements, or the chromosomes that contain them, are selectively inherited. The most efficient gene drives might ensure an inheritance bias of nearly 100 percent, such that the genetic sequence doubles in frequency, relative to a similar genetic element not displaying drive, in each generation. All things being equal, after 10 generations the gene drive would have increased its frequency in the population by a relative 1,024-fold (210).

There have been several documented cases of gene drives existing in nature, with selfish genes or selfish chromosomes invading populations of several different insect species. By understanding how these selfish elements work, researchers hope to use gene drive to spread modifications into a mosquito population even if those modifications are detrimental to individual fitness. If successful, the strategy could be the beginning of the end of one of the world’s most devastating and persistent infectious diseases.

A formidable foe

In the early 20th century, Italian physician and zoologist Giovanni Battista Grassi and British army surgeon Ronald Ross determined that malaria is caused by a parasite transmitted from human to human exclusively by the bite of female mosquitoes belonging to a limited number of species from the genus Anopheles. Their discovery has motivated more than a century of research into mosquito biology, diversity, and ecology. And this knowledge of the malaria vector has directly led to successful examples of malaria control, and even elimination in some regions.

If we know the genes that are responsible for key mosquito traits, we can theoretically introduce a genetic modification into the insects that reduces malaria transmission.

Local malaria eradication has only been achieved in areas such as southern Europe and several southern US states, where a combination of factors allowed for transmission control—in particular, low numbers of infected people and a low density of efficient mosquito vectors in close proximity to humans. In some regions of sub-Saharan Africa, where 200 million people are infected with malaria every year, infected mosquitoes may bite a person almost every single day during the area’s transmission season. For these reasons, among others, Africa suffers the brunt of the malaria burden, accounting for more than 90 percent of the 438,000 malaria-caused deaths in 2015.

With current methods of malaria control—including large scale removal of mosquito breeding areas, the use of insecticide-treated bed nets, and through regional insecticide spraying—and other conventional tools such as vaccines and drugs currently proving insufficient to eliminate the disease, some researchers have turned to methods for genetically controlling the mosquito population. These approaches have the potential to be species-specific and self-sustaining without requiring repeated releases or high levels of public infrastructure.

Several examples of naturally occurring gene drive elements have been documented in insect populations in the past. For example, in Drosophila melanogaster, the transposable element P is able to copy and paste itself semi-randomly at several sites in the genome so efficiently that last century it swept through all wild populations in less than 60 years. The intracytoplasmic bacterium Wolbachia can also be considered a selfish genetic element. Found in around half of all insect species, Wolbachia is transmitted through the germline and can provide a relative fertility advantage to infected females. Theoretically, either of these genetic elements could be linked to a modification that affected mosquito fitness or the insects’ ability to fight the malaria parasite. However, the indiscriminate nature of transposable element movement in the genome and the difficulty associated with establishing Wolbachia infections in the large majority of malaria-carrying mosquito species has so far hindered the development of these approaches for malaria control. (See “Evolution Resisted,” The Scientist, October 2009.)

SPURRING SPREAD: With gene drive, it’s even possible for a deleterious allele to spread though the population, despite imposing a severe fitness cost.
See full infographic: WEB | PDF

Several years ago, our group decided to investigate whether these homing endonuclease genes could be adapted to work in the mosquito. First, we artificially inserted the natural target site of a homing endonuclease into the mosquito genome and then tested the ability of an inserted endonuclease to cut its target site and spread to homologous chromosomes. By using promoter sequences to ensure the endonuclease was specifically expressed in the germline of the mosquito, we showed that it could act as a gene drive by biasing its inheritance among the sperm or eggs that result in offspring.2

Another type of gene drive proposed to alter mosquito populations was inspired by a class of natural selfish genetic elements called homing endonucleases that exist in single-celled eukaryotes such as yeast and algae.1 Homing endonucleases are DNA-cutting enzymes that are extremely precise in their action because they recognize a very long (and therefore very rare) target sequence on a chromosome and insert the endonuclease-encoding gene within this sequence, making it resistant to further cleavage. When the endonuclease comes into contact with its chromosome’s homolog that does not contain the endonuclease gene, it cuts the intact target sequence, prompting the cell’s DNA repair machinery to use the intact chromosome, which contains the endonuclease gene, as a template to mend the break. (See illustration.)

This type of biased inheritance is dependent on the pathway the cell uses to repair the broken chromosome rather than on the nuclease itself. Indeed, we and others showed that this type of gene drive could be constructed with any type of sequence-specific DNA cutting enzyme,3 including CRISPR,4,5 by inserting these nucleases within their target site on the chromosome. CRISPR gene editing requires only a single protein (Cas9) and a small guide RNA that determines specificity by simple complementarity to the DNA target, making it relatively easy to engineer a nuclease protein to recognize sites of interest in the mosquito genome. In combination with years of research on which type of genetic modifications are most likely to reduce malaria transmission, researchers now have the tools they need to develop effective malaria control using a gene drive approach.

Implementing a gene drive against malaria

TARGETING MOSQUITOES: Strategies for using gene drive to control mosquitoes.
See full infographic: WEB | PDF

When it comes to genetically modifying mosquitoes to control the spread of disease, there are three broad strategies: spread a deleterious modification through the mosquito population to reduce overall mosquito numbers; bias the inheritance of the male sex chromosome, distorting the sex ratio toward males, which do not bite humans; or equip mosquitoes with antiparasite cargo such as an antiparasitic peptide. Each of these approaches has shown promise in early-stage lab work, and each has its own merits and considerations.

Disrupt an endogenous mosquito gene

The easiest way to imagine reducing a mosquito population would be to target a gene essential for female fertility. Individuals with only one copy of the defective gene are fertile, but the gene finds its way into all of their gametes, such that they transmit the gene to all of their offspring. As the gene drive element increases in frequency in the population, so does the probability of offspring receiving a copy of the gene drive from both parents and thus no functional copy of the female fertility gene. All such females in this class will be sterile, while males will continue to transmit the gene drive to all offspring.

Our group has recently identified three mosquito genes with roles in female fertility that may be suitable targets for such an approach, and we have shown in small laboratory populations that a gene drive specifically designed to disrupt one of these genes can spread through the population.4 An alternative but similar approach would be to target a mosquito gene essential for parasite transmission, such as a specific receptor to which the parasite binds in order to enter the cell. No such specific receptors have yet been identified for the human malaria parasite.

Distort the sex ratio

Because only female mosquitoes transmit malaria, any intervention that skews the sex ratio of a mosquito population toward males should lower disease transmission. Our group has also shown that the X chromosome in Anopheles gambiae can be selectively destroyed during sperm maturation in males by expressing nucleases that specifically target a repetitive sequence only found—in several hundred copies—on the X chromosome.6,7 The nucleases shred the X chromosome, causing nonviability in sperm that inherit it. As a result, 95 percent of all viable sperm contain a Y chromosome and produce male offspring containing the nuclease. We are currently working to link the endonuclease gene to the Y chromosome so that the nuclease can act as a gene drive to spread through the population.1 Eventually, the population would be expected to crash due to a shortage of females needed for mating.

Add a cargo

A third option is to equip mosquitoes with some sort of cargo that directly fights the malaria parasite. When the gene drive is designed to bring along such a payload, the desired effect is dependent on the nature of the cargo rather than the genomic location of the gene drive. Researchers have proposed a range of antimicrobial peptides, both natural and synthetic, for this purpose, and in 2015 researchers successfully constructed a gene drive in Anopheles stephensi that included genes encoding single-chain antibodies designed to bind and inhibit surface proteins on the malaria parasite.5 This study showed biased inheritance of the gene drive in a single generation; however, the effect on the parasite has not been demonstrated, and this approach has yet to be tested in a caged mosquito population.

L to R: toeytoey2530; piola666; MShep2. All © istock.com; L to R: toeytoey2530; ConstantinCornel; ruksil; ertyo5. All © istock.com

Gene drive considerations

Gene drives have enormous potential for the control of populations of insect vectors and pests. They are species-specific, self-sustaining, and have the potential to be long-term and cost-effective. Gene drives can complement existing approaches, such as vaccine development, antimalarial drugs, and conventional vector control approaches, including insecticides, bed nets, and other barriers.

The building of functional gene drives for malaria control is still in its infancy, however, and each of the various approaches must overcome considerable technical hurdles. Although gene drive is deliberately designed to spread despite causing a negative fitness effect on the mosquito, it will nonetheless generate a selection pressure for the evolution of resistance in the mosquito population. This selection pressure is likely to be higher for strategies that aim to reduce the reproductive output of the population than for those that alter the insect’s ability to transmit the parasite.

The most obvious way resistance might manifest is sequence variation at the target site of the endonuclease. Such variation may already exist in the population or it can accrue at a low rate if the broken target site is repaired by a DNA repair mechanism called nonhomologous end joining, rather than by copying the homologous chromosome (homology-directed repair). This risk can be mitigated in several ways, such as choosing target sites where there is very strong sequence conservation and/?or multiplexing of gene drive elements. For example, a CRISPR-based gene drive might be designed to include multiple different guide RNAs so that likelihood of resistance to cleavage by all guide RNAs is reduced, akin to combination therapy with antibiotics. Engineering tolerance into the nucleases so that the enzymes cleave target sequences despite single-base-pair variants could also help combat mosquitoes evolving resistance to the approach.

On the other hand, with a gene drive carrying antiparasitic cargo, the malaria parasite will be under selection for resistance, and the robustness of the antiparasitic effect might vary with the genetic backgrounds of the mosquito and Plasmodium strains, as well as with environmental factors such as temperature. Additionally, gene-drive elements carrying a cargo might lose it, either through mutation or recombination, but still show drive and might outcompete those with the cargo. This is not a concern for gene drives targeting an endogenous gene as there is no additional cargo to lose. Ultimately, the choice of which approach to take will depend on the specific disease-transmission scenario and the ecology of the vector species.

Any gene drive approach that shows efficacy in the laboratory must be subject to a lot of testing in strictly confined large cages to see how the intervention scales up, how stable the gene drive elements are, and whether any signs of resistance or off-target effects crop up. In the field, background studies to look at the distribution, genetic heterogeneity, and gene flow of mosquito species in a given area are essential.

In addition, as with any new technology, a careful appraisal of the approach’s ethical and ecological implications is imperative. In this light, we have welcomed recent efforts by the research community to agree on a set of principles for the safe development, implementation, and containment of gene drives.8 A recent report by the National Academy of Sciences encouraged a cautious, step-wise approach to building, testing, and implementing gene drives.9 But while overarching guidelines are helpful, deploying gene drives for control of any population should be assessed on a case-by-case basis. When it comes to targeting populations of Anopheles gambiae mosquitoes, regulators should weigh risks against the potential benefits of controlling a disease that is still causing a huge health and economic burden in the face of current malaria-control strategies.

Lastly, and most importantly, the decision to deploy a gene drive technology does not, and should not, lie with the scientists who design it. This will be a decision to be made by those countries and communities that are affected by the disease and will require an open dialogue to engage stakeholders at all levels and to ensure that consensus is reached before the technology is implemented.

Tony Nolan is a molecular biologist and senior research fellow at Imperial College London, where Andrea Crisanti is a professor of molecular parasitology.


  1. A. Burt, “Site-specific selfish genes as tools for the control and genetic engineering of natural populations,” Proc Biol Sci, 270:921-28, 2003.
  2. N. Windbichler et al., “A synthetic homing endonuclease-based gene drive system in the human malaria mosquito,” Nature, 473:212-15, 2011.
  3. A. Simoni et al., “Development of synthetic selfish elements based on modular nucleases in Drosophila melanogaster,” Nucl Acids Res, doi:10.1093/nar/gku387, 2014.
  4. A. Hammond et al., “A CRISPR-Cas9 gene drive system targeting female reproduction in the malaria mosquito vector Anopheles gambiae,” Nat Biotechnol, 34:78-83, 2016.
  5. V.M. Gantz et al., “Highly efficient Cas9-mediated gene drive for population modification of the malaria vector mosquito Anopheles stephensi,” PNAS, 112:E6736-E6743, 2015.
  6. R. Galizi et al., “A synthetic sex ratio distortion system for the control of the human malaria mosquito,” Nat Commun, 5:3977, 2014.
  7. R. Galizi et al., “A CRISPR-Cas9 sex-ratio distortion system for genetic control,” Sci Rep, 6:31139, 2016.
  8. O.S. Akbari et al., “Safeguarding gene drive experiments in the laboratory,” Science, 349:927-29, 2015.
  9. National Academies of Sciences, Engineering, and Medicine, Gene Drives on the Horizon: Advancing Science, Navigating Uncertainty, and Aligning Research with Public Values, doi:10.17226/23405, 2016.

Interested in reading more?

Magaizne Cover

Become a Member of

Receive full access to digital editions of The Scientist, as well as TS Digest, feature stories, more than 35 years of archives, and much more!
Already a member?