Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Gene Drives: How to Genetically Modify an Ecosystem

  1. Jul 19, 2014 #1


    User Avatar
    Science Advisor

    Genes normally have a 50-50 chance of being passed from parent to offspring, but scientists may have figured out a way to create gene drives that show up in offspring with a much higher frequency:

    This idea had been discussed for a while (it was first proposed by Austin Burt in 2003), but new gene editing methods developed in the past few years seem to make this idea much closer to reality.

    What is most exciting – and concerning – about gene drive technology is that when introduced into wild populations, organisms containing gene drives would breed with the population and could spreading the modified genes throughout the population even if the modifications decrease the reproductive fitness of the organism. The researchers imagine this technology could have a number of applications, for example, modifying mosquito populations to prevent the spread of malaria, modifying agricultural pests and weeds to deal with pesticide and herbicide resistance, and modifying invasive species to limit their ecological damage. A recent paper in the journal eLife discusses how such gene drives could be engineered and their potential applications.

    However, with such far reaching consequences, society should approach this technology with caution, and the same authors of the eLife paper have also published a policy forum paper in Science opening the conversation about how this technology should be regulated.

    In addition to the Scientific American blog post linked above, PBS also has a good, popular press summary of the papers: http://www.pbs.org/wgbh/nova/next/evolution/crispr-gene-drives/
  2. jcsd
  3. Jul 22, 2014 #2
    Very nice! Thanks for sharing Ygggdrasil!
  4. Mar 19, 2015 #3


    User Avatar
    Science Advisor

    Two studies published today provide experimental demonstration of gene drive technologies in fruit flies and yeast (two commonly used experimental organisms in biology). Both these studies make use of CRISPR/Cas9 genome editing system. Here's an exerpt from a news article in Science:


    Here are the studies:
    Gantz and Bier. The mutagenic chain reaction: A method for converting heterozygous to homozygous mutations. Science. Published online March 19, 2015. doI:10.1126/science.aaa5945
    DiCarlo et al. RNA-guided gene drives can efficiently and reversibly bias inheritance in wild yeast. bioRxiv, posted March 19, 2015. doi:10.1101/013896 (this was posted in the non-peer reviewed bioRxiv pre-print server, but it comes from a reputable group with expertise in the area)

    These studies have stirred much debate among scientists because they are capable of making large scale changes to our environment. Both groups employed safeguards in their research to prevent accidental release of their technologies into their environment, but the barrier to creating gene drives is not very high, and many research groups around the world likely have the capabilities to build similar systems. What sorts of restrictions should we place on this type of research? Should systems like these be used in the wild to, for example, combat malaria, and what criteria should we use to make that determination?
  5. Mar 19, 2015 #4


    User Avatar
    2016 Award

    Staff: Mentor

    A very interesting report.

    Sounds like something that would spread so incredibly fast that I wonder why it has not occured naturally at some point, or at least why it is not so common we see it today.

    If I understand the method correctly, you need a specific CRISPR complex for a specific gene you want to spread? And then you also need one gene for this specific CRISPR complex - and that has to spread faster as well?
    Last edited: Mar 20, 2015
  6. Mar 19, 2015 #5


    User Avatar
    Science Advisor

    Here's a diagram from the Science paper describing how it works:

    CRISPR/Cas9 is a complex between a protein called Cas9 and an RNA containing specific structural features that allow it to bind to Cas9 (called the guide RNA or gRNA). A specific part of the gRNA sequence tells the Cas9 protein which sequences to cut, and by changing that sequence in the gRNA, researchers can "program" CRISPR/Cas9 to a specific sequence of DNA. When CRISPR/Cas9 cuts the DNA on one chromosome, it activates the DNA repair pathways in the cell. These DNA repair pathways grab the homologous chromosome (remember, we have two copies of each chromosome), finds a region with a similar sequence to the damaged region, then copies the sequence from the homologous chromosome onto the damaged chromosome. So, everything between the two homology arms of the gene drive (HA1 & HA2 in the figure), which would include the Cas9 gene, the gRNA, and whatever else you include, gets copied onto the other chromosome. Thus, the Cas9, gRNA, and other additional sequences all spread together.

    You would need a specific gRNA for each specific gene you would want to spread, but the same Cas9 works for any gRNA. The study done in yeast by George Church's lab made a gene drive containing only the homology arms, the gRNA, and the target gene they wanted to spread. They supplied the Cas9 in a separate plasmid so that they could better control the process.

    That's a good question. The CRISPR/Cas9 system comes from bacteria and these bacteria control the system such that it does not target the bacterium's own DNA (it acts as a defense against viral DNA), so potentially these things have never popped up in eukaryotes (which have the homologous repair system that helps to copy gene drives).
  7. Mar 25, 2015 #6

    Filip Larsen

    User Avatar
    Gold Member

  8. Mar 25, 2015 #7
    Runaway diversity scares me. 12 new viruses this year 144 next year perhaps 20,000 the year after? To introduce a new ecosystem which could devastate the one that exists just sounds risky right off the bat. Especially one based on mass mutation of species.
  9. Mar 25, 2015 #8
    I think how fast it would spread depends at least in part on the gene(s) in question. If they gene harms the organism to much it wont be able to pass it on, and there is also http://evolution.berkeley.edu/evosite/evo101/IIIE3Sexualselection.shtml [Broken] to think about.
    Last edited by a moderator: May 7, 2017
  10. Mar 26, 2015 #9


    User Avatar
    Gold Member

    But it still seems it could spread even if it confers a sgnificant disadvantage (within limits). Prsumably that s what makes it attractive as a weapon against some mosquitoes and such, but also it sounds potentially disastrous especially if it finds a way to cross species some time after having been released in the wild. Scary stuff,
  11. Mar 27, 2015 #10
    This could also be used in eugenics in humans ? In this case it is really a thing to avoid
  12. Mar 27, 2015 #11


    User Avatar
    Science Advisor

    The Scientific American piece addresses this question:
    While gene drives aren't likely going to affect humans, the CRISPR/Cas9 technology is currently being used by various research groups to engineer human embryos and human germ cells. Science is even reporting that a paper demonstrating gene editing in human embryos has already been submitted for publication (although genetic engineering of human embryos is banned in many countries, it is legal in some major countries like the US and China). So, technologies to create heritable changes in the human genome currently exist and are accessible to many groups around the world. The danger, of course, that is we still don't understand enough about the biology of the human genome to know which genes to edit in order to alter many of the traits we care about (like intelligence).
  13. Mar 27, 2015 #12


    User Avatar
    Gold Member

    The direct risk from human genetical engineering doesn't seem the most concerning. But spreading it into the wild...
    And it's not very reassuring to hear "don't worry, it would take centuries"
  14. Mar 27, 2015 #13

    Doug Huffman

    User Avatar
    Gold Member

    Seems a good time to apply the progressive's favorite, The Precautionary Principle.
  15. Mar 28, 2015 #14
    Maybe the prophecy of the brave new world is not far away. Already with those techniques ips and cloning males are not needed anymore for reproduction, technically ? But psychologically maybe couples are still needed, like a duality principle.
    Last edited: Mar 28, 2015
  16. Mar 28, 2015 #15
    Why? Ever known someone with Huntington’s or Duchenne's?

    Fact is, every animal that mates is deeply concerned with the genetic card they will deal their offspring. Humans are also extremely obsessed by this. It is all eugetics, albeit a terrible crude form.

    We are humans because of our genes. We should treasure the quality of the human gene pool highly. CRISPR will help us improve the human gene pool. So no more need for children to die horrible deaths or people to be rejected because they are too ugly to mate with.

    It is unethical to put people on a plane that isn't tested, fixed, and found to be safe. Same will soon be true for bearing children with unedited genomes.

    In fact, looking at my personal decision, I think it would already be immoral for me to have children without paying the 500 euro to have my genome checked. And if it comes out bad, I may be forced by ethics to not have children. Even if it means I have to be lonely and without family the last 10 years of my life.

    Every child has the right to a set of genes that make a normal life possible. Just as every child has the right to get vaccinated. If the parents disagree, the child should be protected from the parents.
    I am glad I am not in the position where I have to decide on mandatory sterilization, decide on whose indivisible right is to be violated in favour of upholding the other. Let's hope CRISP works really well asap.

    We have a moral obligation here to future generations.
    Last edited: Mar 28, 2015
  17. Mar 28, 2015 #16
    It is not genetics but a more global aspect : The thing is that those methods remain expensive and hence like wealth health go to the richer there is no democratization process going on, it is a kind of peculiar.

    For example i have schizophrenia and at the beginning my parents felt guilty but with the time they disculpabilized and there remain the nature that human won't be able to control. Now for physical disease it could be a good thing but there most of the time remains randomness with genetics.
    Last edited: Mar 28, 2015
  18. Mar 28, 2015 #17


    User Avatar
    Gold Member

    @Almeisan, the position you express is very strong and not uncontroversial.
  19. Mar 29, 2015 #18


    User Avatar
    Science Advisor

    This is quite a dangerous way of looking at things. For example, how do you define "bad"?

    A recent study published in http://dx.doi.org/10.1038/ng.3243[/URL] sequenced the genomes of 2000 individuals, and found that 7.7% harbored at least one gene in which both copies carried loss-of-function mutations. This study revealed [url=http://news.sciencemag.org/biology/2015/03/one-thousand-genes-you-could-live-without]~1,000 genes seemed like they could be knocked out completely without any major adverse health effects[/url] (though, research is continuing on those individuals to see if they can find any effects, especially since some of those genes were, in fact, thought to be important). There are also documented cases where knocking out particular genes seems to be beneficial (see the case of [url=http://www.nature.com/news/genetics-a-gene-of-rare-effect-1.12773]PCSK9[/url]).

    The point here is that, if one were to sequence your genome, it is very likely that one would find many "errors" in the genome. In the vast majority of cases, however, we wouldn't know if these "errors" we identify would have any major functional consequences. Certainly, screening for known genetic disorders is important, but there's a lot of grey area in defining a genetic disorder (for example, would an allele that slightly increases one's risk of type II diabetes be considered a genetic disorder?).

    Furthermore, as we perform more studies that try to identify the genes associated with particular traits, we're quickly finding that, even for traits are very heritable like height, no one gene has a major effect on the trait. Rather, the trait is controlled by hundreds of genes that all have very small effects. This throws into question whether it would even be possible to alter traits like height or intelligence without whole-scale re-writing of the genome (something not possible without major breakthroughs in gene editing technology).
    Last edited by a moderator: May 7, 2017
  20. Mar 29, 2015 #19
    All you are saying is that we might not know right now what is good and what is bad? That's all besides the point.

    I don't know what you mean with 'errors'. For many diseases, we know very well what the ΔF508 mutation is, how the defective protein turns out, and that this is the cause of the disease. Are you really saying that undoing the 3 base pair mutation for, say cystic fibrosis, giving the patient suffering from the symptoms the allele that 29 in 30 of the people have, who seem to have perfectly fine functioning CFTR, would cause even worse symptoms?
    All you have to do is give a patient the healthy allele of one of the parents, one she or he would have gotten anyway if not for the bad luck.

    Are you really saying ordinary 'healthy alleles' cause more disease than 'diseased alleles'? You can't see the difference between starting with fixing the most heinous and obvious genetic diseases? So we can't start out to cure the most cruel genetic diseases without also tweaking everything else that we shouldn't be touching yet?

    I don't know why you introduce height or introduce traits that aren't coded for by alleles. Many of these diseases are the result of a defective protein. Genes code for proteins. It can't be more obvious. In some cases we in fact already provide the patient with medicine containing the correct protein.
    There isn't a protein that is the sole determining factor for your height. Also, the natural variation in height, as opposed to a growth disorder, isn't something that requires medical treatment.

    It's not like you are giving an example of a gene that has a 5% odds to cause cancer before age 45, but is also correlated with brain function(and might make people smarter/nicer/more social, though we have no way to know). It's not like I am saying we should cut everything suspicious out of our genome and risk actually making it worse, not better. The whole debate is the argument that we should let nature run it's course, as intended.

    All pharmaceuticals have side-effects. Often very severe ones. We still use them. Are you in opposition against this as well? Fixing the gene itself, when correctly identified and correctly replaced, has zero side-effects. CRISPR so far seems extremely selective. Of course CRISPR should meet all the standards we put on medical treatments right now. But I doubt you think I say CRISPR should get a special treatment.

    You quote all these articles to look cool, but you seem not to know what there relevance might be, if any.
    I don't see how apparently defective proteins that don't seem to cause any disease have any relevance. I am also not saying that by tomorrow we should insert genes for endocellulase or daptomycin or something.

    And yes, the idea that some genes in our genome can potentially code for proteins that only do us harm, may be true. Only more reason to look for them, not less.
    Of course many genes are going to have tradeoffs. But even there CRISPR can give a solution. You can knock out cancer-causing heart-disease inhibiting genes in certain people, do the opposite in others, until a long-term solution is found.
    Last edited: Mar 29, 2015
  21. Mar 30, 2015 #20


    User Avatar
    Science Advisor

    Sure, I would agree with this, though the line one draws between the most cruel genetic diseases and ones that should not be touched yet can be somewhat fuzzy. Diseases like cystic fibrosis, Huntington's disease, and sickle cell anemia are good examples. What about gene variants that confer disease risk? BRCA1 and BRCA2 mutations, TP53 mutations, and other mutations that confer high risk of developing cancer would probably be a good idea. But, scientists have identified numerous mutations that seem to be associated with cancers of some type and many other diseases, but often the size of the risk and the penetrance of the phenotype is unclear. This is probably an example of letting the perfect be the enemy of the good, and a slippery slope fallacy, but these are important questions that should be addressed.

    Furthermore, for most of the high risk genetic disorder that we'd like to fix are Mendelian, so these could be avoided through preimplantation genetic diagnosis of embryos rather than gene editing which currently carries some risks of introducing unintended genetic changes elsewhere in the genome. In this case, one could make the argument that this fact makes germline gene editing unethical because there are safer alternatives available that could achieve the similar goals.

    Overall, I agree with your points more than I disagree and am mainly bringing up these points as more of a devil's advocate.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: Gene Drives: How to Genetically Modify an Ecosystem