By Grant Jacobs 24/01/2019 1

What if you could change a gene in a population by introducing laboratory-prepared animals then just letting nature take it’s course? A gene drive is the informal name given to a process where a genetic variation is set up so that it will be inherited more often in the offspring than it would by chance.[1] In a gene drive each generation has a better than 50% chance of inheriting the new variant, so over time the chosen variant becomes the dominant variant of that gene in the population.

Not a new idea and common in nature

It’s not a new idea. It’s also something that is actually common in nature. The idea has been around since at least the 1960s, and that ‘selfish’ genetic elements can spread through populations has been known for over a hundred years. What is new in recent years is that new ways of creating genetic variants—in particular, CRISPR—have made it something that can now be tested rather than talked about.

At the moment it’s a laboratory thing that researchers are exploring. (One team’s efforts have led to approval to, eventually, do a limited field trial of non-gene-drive animals in Burkina Faso. There’s an excellent (very) long read on this and gene drives in general at Vox by Dylan Matthews.)

Potentially gene drives could be used to eliminate an unwanted species by making offspring infertile or, say, make only male offspring viable. They could also be used to reduce the transmission of a disease by tweaking the genetics of carrier species.

Moving past insects

At the moment gene drives only work in insects.[2]

Today a team from the University of California at San Diego headed by Kimberly Cooper report the first demonstration of a gene drive in mice. They were able to take laboratory mice and ‘drive’ a gene variant (allele) affecting coat colour so that after a few generations most were albino. This worked for the female mice, but not the males.

It’s short of what you’d want in a proper gene drive, and at this point is more highlighting the challenges to overcome (which could be tough), but it’s certainly moving in that direction.

Earlier work has focused on insects, partly because their short lives and large numbers of offspring mean that theoretically gene drives are more likely to be effective in insects than other types of life. In NZ one species we might consider for a gene drive for conservation efforts are wasps. Wasps can have a large impact on forests, and they are a now well-established problem in many parts of NZ. Perhaps in the future well could add mice to the list?

Of tyrosinase, mice, melanin and albinos

Laboratory mice are special breeds. You don’t just wander down an alley and collect a few! They’re careful maintained inbred lines from a small number of institutes dedicated to breeding them. The basic idea is that the genetics of the mice of any one strain will be essentially identical. This way studies on Tyrc-ch mice in Sweden can be compared to studies on the same strain of mice in California. Tyrc-ch mice are the chinchilla variant of tyrosinase used in the study. It’s one of a long list of mice with different Tyr alleles.

We and mice—and other mammals—have a tyrosinase gene. It codes for an enzyme that contains two copper atoms that help our cells make quinones from phenols and polyphenols. One type of quinone is melanin. Tyrosinase helps to make melanin from the phenolic amino acid, tyrosine. Tyrosinase is one of the reasons we need a little copper in our diets.

Melanin is what gives the colour of skin. We (mice and us!) have two copies of each gene – two alleles. If both alleles for tyrosinase don’t work, we’ll be albino or somewhere-in-between if we have gene variants that only slightly reduce tyrosinase activity. This page shows pictures of mice with different alleles of the Tyr gene (locus).

Tyrosinase is in melanosomes, organelles that create and store melanin. Melanin gives darker skin and protects us from the sun.

Coat colour is an easy thing to record (and you don’t have to sacrifice the mice). The mix of a well-defined genetic change and an easy visual way to see the outcome (coat colour) means tyrosinase alleles are popular for some genetic studies. The featured image for this post is a of a ‘chinchilla’ variant tyrosinase mouse.[3]

Copying the gene over, driving the change

DNA repair systems are critical for our survival. Every day our DNA is smashed with stray high energy light and chemical reactions. All animals have a DNA repair system to cope with this.

When DNA is repaired, the replacement for the repair has to be copied from somewhere. Usually it’s the matching copy of the same gene.

One thing you can do with CRISPR is to set things up so that the cell’s DNA repair system will (most often) use a DNA sequence you want it to to make the repair. You cut the allele (gene variant) you want altered, triggering repair of that allele. The cell’s own repair system will repair the cut, reading in the replacement sequence you intended.

In this case the researchers wanted to end up with both alleles replaced with a variant they’ve supplied.[4] If both copies of the tyrosinase gene are not working the mice will have white coats; if one copy is working the mice will have grey coats.

They inserted a copy of DNA with a ‘guide RNA’ into the tyrosinase gene. This insertion disrupts the reading of the tyrosinase gene so it’s non-functional. The guide RNA that is inserted directs the Cas9 cutting enzyme of CRISPR to specifically cut the other, normal copy of the tyrosinase gene.  But these mice only have the guide RNA, no Cas9 enzyme. Without the Cas9 enzyme no cutting can take place.

Left like this the mice have grey coats: one defective tyrosinase gene, and one normal. Gene editing requires both the guide RNA and the Cas9 enzyme to work.

If you breed these mice with mice that do have a copy of the Cas9 enyzyme, then gene editing might occur in the embryo. (By splitting the two bits across different parent mice the researchers can test what’s happening.)

By varying how they encode the Cas9 enzyme gene, researchers can restrict where the Cas9 enzyme gene is expressed: in males or females, or going all the time in all cells.

Having Cas9 running all the time isn’t a good idea

When the Cas9 cutting enzyme gene is set up to be running all the time, a DNA repair pathway that can damage the DNA, including guide RNA region, frequently occurred. That’d be no use for a gene drive ‘in the wild’. The guide RNA is what directs the cutting to the right place. If you lose that, the gene drive will stop working. (It’d also mean unreliable repair of the normal gene that was targeted for editing.)

Not much luck with males either

When they tried tweaking the Cas9 cutting enzyme gene so it was only going in males, long story short, the final outcome was similar. Why this is, they don’t know. This may be because the different type of cell replication in sperm cells compared to eggs affects how the DNA repair is working. It could also be the timing of the cutting compared to cell replication (more specifically before or during one type of cell replication, meiosis).

More work is needed learn specifically what is happening, but at this stage they’re not having much luck with males.

Better luck with females

By contrast the highest rate of inheritance of the modified tyrosinase gene they saw for females was 72% – well about the 50% you’d expect from random Mendelian inheritance. We’re getting somewhere!

This may be because the timing of Cas9 gene activity wasn’t such an issue with females. They may have ‘got lucky’ with the particular set up they used, so that luckily the cutting may have been delayed so that it fitted better to the cell replication cycle, and hence DNA repair.

Future prospects and concerns

The purpose of the research is ‘proof of principle’: can we do this thing?

This work isn’t ready to be used ‘in the wild’, and even if it were it wouldn’t be approved for use.

Their work shows that to get an effective gene drive for practical use in mice they’d have to work past that the wrong DNA repair mechanism is happening in males. They’d also want a higher frequency of gene conversion. This may be a challenging, especially if the correct type of DNA repair only occurs in specific times in embryo development.

Many laboratory experiments don’t need such high efficiency gene conversion (even if that might be better). This research shows mice could be used to explore genetic diseases or other laboratory experiments using a CRISPR-based gene drive approach.

Gene drives could be used to reduce (or eliminate) genetic variants that are essential to spreading a disease. Gene drives might also be a good conservation tool that could be used to dramatic reduce populations of an unwanted species.

This research is another reminder that we need to think about what uses we put these technologies to, or not, as the case might be. They’re not ready yet but they’re coming down the road.

Other articles in Code for life

Three kinds of knowledge about science journalism

Note to science communicators–alleles not “disease genes”

Sea stars and mosaics

Deleting a gene can turn an ovary into a testis in adult mammals

One example of why all those genomes from different species are useful to biologists

Strongest opponents of GM think they know best but actually know the least


1. This type of inheritance has been dubbed ‘Super Mendelian’. Gregor Mendel is the Austrian monk famous for growing and breeding peas, noting how traits like wrinkled peas were inherited. In Mendelian genetics genes are dominant or recessive. Dominant genes ‘rule over’ recessive ones, suppressing their effect. In the classic, idealised model, the offspring are a random mix of the parents alleles. In a gene drive, the offspring have a better than 50% chance of inheriting the ‘driven’ gene. Thus over time, the fraction of animals with the gene in the population will rise.

Recessive alleles are marked with a lowercase letter; dominant with an uppercase letter. Source:

(In practice these ideal ratios are rarely seen in life because of other aspects of how a gene contributes to the trait it contributes to. Mendel’s work still gives a conceptual framework of the basics of how genes pass on to the next generation that you can fit the complexity of the reality into. Also: Mendel, or one of his assistants, may have tweaked his numbers! – statistically they’re a little too good to be true.)

2. One thing that I find turning up in a lot of expressions of concern about new genetics is that people seem to think everything in genetics is universal. It’s understandable that people think this, but it’s not true. For example, simplified descriptions of genetics have for decades been talking about the ‘universal genetic code’ used to read genes when in practice it slightly differs in different types of life. Another thing that differs, much more dramatically, is how genes are packaged, organised and used in different types of life or different types of cells.

3. I’ve tried to find the origin of the name ‘chinchilla’ for the allele, but a quick look didn’t turn up anything useful. You’d suspect it’s been named after the epinomous South American rodent, one way or other, perhaps from a similar coat colour in some variety of chinchilla. (The linked page also suggests they make tricky pets to look after.)

4. I’d give you how this is done, but I’m writing this in the wee hours of morning and I’d like to get to sleep! More seriously, it’s better left for an article dedicated to that. The full story is quite long and detailed.


Key reference

Hannah Grunwald, Valentino M. Gantz, Gunnar Poplawski, Xiang-ru S. Xu, Ethan Bier & Kimberly L. Cooper.

Super-Mendelian inheritance mediated by CRISPR–Cas9 in the female mouse germline

Published in Nature,

(The DOI may take a day or two to become active.)

Further references

Committee on Gene Drive Research in Non-Human Organisms: Recommendations for Responsible Conduct; Board on Life Sciences; Division on Earth and Life Studies; National Academies of Sciences, Engineering, and Medicine. Gene Drives on the Horizon: Advancing Science, Navigating Uncertainty, and Aligning Research with Public Values. Washington (DC): National Academies Press (US); 2016 Jul 28. 1, Introduction. Available from:

About the featured image

The featured image is from the European Society for Pigment Cell Research (ESPCR) website. If you’re interested you can follow them on Twitter: ‎@EuSPCR. This is one of a very long list of mouse colour genes they have web pages for. Picture credit: Lluis Montoliu (Source)

One Response to “A gene drive in mice – but only for females”