A litany of methods have been put forward to stop the spread of malaria by limiting the ability of the Anopheles mosquito to act as a disease vector. Yet the Plasmodium parasite that causes the disease remains responsible for well over half a million deaths annually. Two recent reports show that gene drives, outfitted with a technological update, may be a promising way to check the disease's spread.

Gene drives in Anopheles achieve super-Mendelian inheritance, in this case of female infertility. Adapted from Hammond et al. (2015).

Gene drives were proposed decades ago as a way of forcing genetic changes to sweep through populations rapidly. A genetic variant normally has a 50% chance of being passed on to progeny, requiring generations under selective pressure to reach a high frequency in a large population. Gene drives work by initiating gene conversion, turning one variant copy in a given individual into two, and thereby ensuring close to 100% transmission in the next generation.

The CRISPR-Cas9 system has recently emerged as an effective and heritable auto-edit mechanism. The technology uses Cas9 nuclease to cut at a genomic site targeted by a short guide RNA (gRNA). DNA sequence between two cuts can be replaced with a templated sequence—in this case, the allele coding the gene drive mechanism—by homology-dependent repair (HDR).

CRISPR-based gene drives were recently demonstrated in budding yeast and the fruit fly, the latter by Ethan Bier and his group at the University of California, San Diego. To tackle Anopheles, they teamed up with Anthony James at the UC Irvine campus (Gantz et al., 2015), who had shown previously that small, single-chain antibodies (known as nanobodies) targeting two Plasmodium falciparum proteins can halt its transmission.

The researchers integrated the two nanobody effectors into a locus controlling eye color in Anopheles stephensi, a malarial vector on the Indian subcontinent. They also included a fluorescent reporter, germline-expressed Cas9 nuclease and a gRNA targeting the same locus. Outcrosses resulted in over 99% transmission, and the inducible effectors were transcribed when the mosquitoes imbibed a blood meal.

Despite this success, the researchers detected a sex-dependent transmission difference that would compromise performance in the wild. When gene drive–bearing females were outcrossed to wild-type males, new alleles were often generated by an untemplated repair process rather than by HDR. James believes that the male pronucleus in fertilized zygotes may be physically too far from the female DNA template, favoring the alternate repair pathway. Because many new alleles alter the gRNA site, they disarm the gene drive mechanism in their progeny.

An alternative approach was spearheaded in Anopheles gambiae, the main malaria vector, by Tony Nolan, Andrea Crisanti and Nikolai Windbichler from Imperial College London (Hammond et al., 2015). They opted to hobble the disease vector by coupling a gene drive to female infertility. Drastically reducing mosquito populations could be a general strategy for slowing the spread of other serious human diseases such as dengue fever and West Nile virus.

The researchers identified three genes likely to be required for fertility and targeted them with a gene-disrupting cassette bearing germline-expressed Cas9, a gRNA and a fluorescent reporter, achieving transmission rates from 91% to nearly 100%. They also detected a small number of new alleles from alternate repair or incomplete gene conversion that would lead to gene drive immunity.

Gene drives could be a potent weapon against other pathogens or their vectors; they could make at-risk species immune to infection, combat invasive species or remove weeds from farmers' fields. But they are limited to diploid organisms that reproduce quickly enough to effect meaningful change.

More important, unintended consequences of species engineering and the ethics of species eradication need broader discussion. Scientists are exercising abundant caution to ensure that potential effects are calculated carefully and safeguards are in place to avoid releasing a 'mosquito zero' that could trigger a gene drive before enough development and testing. Recent discussion has focused on the need for confinement and ecological isolation (used in the two studies mentioned here), phased testing and the ability to use additional gene drives to immunize populations against the original drive. Although gene drives are species specific, they should be designed in a way that minimizes any risk of transfer to nontarget organisms.

The scientists leading these studies stress that gene drives have great potential as complementary tools to be deployed alongside existing controls. For now, they face the challenges of avoiding acquired resistance and grappling with genetic variability in the wild.