Gene Drives: How to Genetically Modify an Ecosystem

334
47
What would be more unethical? Deliberately let someone be born with a Duchenne allele. Or let that same person be born with a functioning gene coding for a functioning dystrophin protein, but not knowing very well if the treatment may have some effect or something may be overlooked?

Don't we do clinical trials and tests on animals right now? I don't see how you can say we can't make inferences from animals. (ignoring how immoral it may or may not be). But these proteins aren't unique to humans. We can put disease causing allales in animals for almost all these well-known genetic diseases and it should cause very similar symptoms. This is being done right now by quite a few teams.
Also, there will be human volunteers.

If you carry this line for all medicine, we wouldn't even be allowed to use antibiotics or vaccins. In fact, there's still people arguing against vaccination.


Each year many people outright die from normal everyday pharmaceuticals. We never really know all the systems we are tuning or affecting when introducing a substance. Be it corticosteroids, statins, birth control, painkillers, etc. Let's not even get started on the central nervous system.

I suspect that when CRISPR is reliable, and it has all the promise to be so, it will be one of the treatments with the least side-effects. Because you are never tinkering with gene expression. All you are doing is fixing a mutation that completely breaks a protein. Often a point mutation resulting in a premature stop codon or a frame shift. Often you are changing only 1 basepair.
 
Last edited:
Ygggdrasil
Science Advisor
Insights Author
Gold Member
2,939
2,115
What would be more unethical? Deliberately let someone be born with a Duchenne allele. Or let that same person be born with a functioning gene coding for a functioning dystrophin protein, but not knowing very well if the treatment may have some effect or something may be overlooked?
Except in rare cases, you can do this through preimplantation genetic diagnosis (PGD) instead of gene therapy. Scientists are able to perform single cell sequencing on human oocytes or fertilized embryos in order to determine which cells carry the mutation of allele and which ones don't. They can then implant those only those lacking the particular disease alleles. In the example of Duchenne muscular dystrophy (a recessive X-linked trait), this procedure could guarantee that female carriers or males with the disease do not pass the condition down to their children (the one case where gene editing would be necessary is the case of a female carrying two copies of the disease allele, which is very rare).


As for the issue of testing, I agree that not much testing would have to be done for fixing disease alleles (other than providing evidence for the general safety of CRISPR/Cas9 gene editing, which is something that could be evaluated in human cell lines before being used in people). However, I argue above that PGD is an easier way of achieving this goal than gene editing. The area where gene editing would be required is in the introduction of new traits (e.g. knockout of PCSK9 for improved cardiovascular health). While we have limited data that such a knockout is safe (there are people lacking functional copies of PCSK9 who seem perfectly healthy), there might be the worry that such mutations are not tolerated in certain genetic backgrounds because of potential interactions of that gene with other genes in the body. For example, there are definitely cases where mutations in two different genes together can cause a disease, where either mutation alone does not cause a problem. It would be challenging to evaluate the potential for such effects in animal studies. Small scale tests in stem cell may be possible, it would be challenging to evaluate gene edits that make changes system-wide (for example, to cholesterol metabolism in the case of PCSK9).
 
120
10
While we have limited data that such a knockout is safe (there are people lacking functional copies of PCSK9 who seem perfectly healthy), there might be the worry that such mutations are not tolerated in certain genetic backgrounds because of potential interactions of that gene with other genes in the body. For example, there are definitely cases where mutations in two different genes together can cause a disease, where either mutation alone does not cause a problem. It would be challenging to evaluate the potential for such effects in animal studies. Small scale tests in stem cell may be possible, it would be challenging to evaluate gene edits that make changes system-wide (for example, to cholesterol metabolism in the case of PCSK9).
This argument has unpleasant implication. If it is unethical to try different combinations of already existing human genes then it seems like someone could argue that interracial marriage is unethical.
 
33,552
9,284
Don't we do clinical trials and tests on animals right now? I don't see how you can say we can't make inferences from animals. (ignoring how immoral it may or may not be). But these proteins aren't unique to humans. We can put disease causing allales in animals for almost all these well-known genetic diseases and it should cause very similar symptoms. This is being done right now by quite a few teams.
I don't say you cannot make inferences, but genetic changes in human can have effects different from the effects on other animals.
Also, there will be human volunteers.
For drug testing this is easy, but how can a baby volunteer before it is even born?

If it is unethical to try different combinations of already existing human genes
Oh, we try that with every child anyway. But I'm sure many will see a large difference between sampling genes at random from two humans, and biochemical modifications. Especially if you go beyond single base pair mutations that are well-known.
 
Ygggdrasil
Science Advisor
Insights Author
Gold Member
2,939
2,115
This argument has unpleasant implication. If it is unethical to try different combinations of already existing human genes then it seems like someone could argue that interracial marriage is unethical.
Since PCSK9 double deletion is very rare, we don't have a lot of data about its effects in a variety of genetic backgrounds, so it is reasonable to have these concerns (whether these concerns should halt potential trials of PCSK9 gene editing is debatable, however). Re-assortment of common variants is less concerning, especially in interracial marriages where there is presumably less chance of similar variants recombining to knock out both copies of a gene.

However, your general point that every natural conception is an experiment is important to keep in mind. Even if gene editing does run into unforeseen problems, the frequency of problems seen in conventionally conceived children is probably the appropriate point of comparison.
 
334
47
Oh, we try that with every child anyway. But I'm sure many will see a large difference between sampling genes at random from two humans, and biochemical modifications. Especially if you go beyond single base pair mutations that are well-known.
So what is the difference? Besides trying random genes is likely to have a higher risk than deliberately picking them.

That people are going to object with 'we should not play god' or 'slippery slope', I know. But I don't think those are very good reasons to condemn a child to a genetic card with an early death or significant complications/reduced quality of life.
 
33,552
9,284
So what is the difference?
Depending on what you do, there might be no difference at all.
That people are going to object with 'we should not play god' or 'slippery slope', I know.
And that is a serious issue. People don't care about ~2 mSv/year of background radiation, they care about the extra 0.0001 mSv from living close to a nuclear power plant. You can collect many signatures if you suggest to ban atoms, genes, or chemicals in general (or just dihydrogen monoxide in particular). Even PGD, where no genes are changed, is disputed.
 
Ygggdrasil
Science Advisor
Insights Author
Gold Member
2,939
2,115
An update on gene drive research:

Two groups have recently published papers demonstrating that gene drives work in mosquitoes. One study demonstrated gene drives in Anopheles stephensi, a malaria vector in the Indian subcontinent, and demonstrated that it could be used to spread malaria-resistance genes. The other group worked with Anopheles gambiae, a malaria vector in sub-Saharan Africa, and demonstrated a gene drive that affects female (but not male) fertility and thus could be used to reduce mosquito populations. Links to the studies and a news piece summarizing them are below.

Gantz et al. 2015. Highly efficient Cas9-mediated gene drive for population modification of the malaria vector mosquito Anopheles stephensi. Proc Natl Acad Sci USA 112: E6736. doi:10.1073/pnas.1521077112

Hammond et al. 2016. A CRISPR-Cas9 gene drive system targeting female reproduction in the malaria mosquito vector Anopheles gambiae. Nat. Biotech. 34: 78. doi:10.1038/nbt.3439

Summary from Nature news

It's possible that a gene drive targeting female fertility could help against the current Zika virus spread in the Americas by limiting mosquito populations. Of course, the decision to release a gene drive into the wild should not be taken lightly, and research should be done to consider any long-term unintended consequences of such action. Still, with a Zika virus vaccine potentially a decade away, gene drives seem like a solution that could be available in a few years.
 
  • Like
Reactions: mfb

Related Threads for: Gene Drives: How to Genetically Modify an Ecosystem

  • Last Post
Replies
4
Views
3K
  • Last Post
Replies
13
Views
2K
  • Poll
  • Last Post
2
Replies
33
Views
6K
Replies
26
Views
6K
Replies
3
Views
2K
Replies
10
Views
2K
  • Last Post
Replies
2
Views
2K
  • Last Post
Replies
3
Views
2K
Top