Why Mitalipov’s ‘magic’ room for human embryo CRISPR may give us pause

Shoukhrat Mitalipov
Shoukhrat Mitalipov points to an image of a potential CRISPR’d embryo inside an incubator at the Center for Embryonic Cell and Gene Therapy in Portland, Ore. Photo credit: Rob Stein on NPR.org, re-used with permission.

The debate over whether the main conclusions of the Nature paper on human embryo CRISPR led by Shoukhrat Mitalipov’s lab at Oregon Health Sciences University (OHSU) are correct remains unresolved. Note that Nature just added an editorial alert just above the references section to their paper:

“Updated online 02 October 2017. Editorial Note: Readers are alerted that some of the conclusions of this paper are subject to critiques that are being considered by editors. Some of these critiques have been publicly deposited in preprint form. A further editorial response will follow the resolution of these issues.”

This research should be shaking us up on another level besides just the core science. Recent developments deepen societal and ethical questions about this kind of research. A unique look by NPR inside Mitalipov’s lab ups the stakes and I believe should give us pause. The surprisingly huge scale and clear clinical intent together magnify the risks.

Also, if Mitalipov’s team in particular actually might not be CRISPR’ing the human embryos the way they thought they were, could many of the human embryos being made and used in their work not be contributing to useful data production? The importance of this question is magnified by the large-scale of the embryo use. In responding to the concerns raised by the Egli preprint that questioned the central conclusions of the Ma Nature paper, Mitalipov mentioned that his lab has worked with “hundreds of human embryos”.


Were those all CRISPR’d?

How many hundreds are we talking about and where did all the human eggs come from to make those? Overall, how many human embryos are being made and/or used for genetic research around the world? It’s hard to say but CRISPR genetic modification of human embryos is likely now ongoing widely across the globe, with active work in at least four and probably more countries. Could there soon be in the range of thousands of human embryos having been CRISPR’d and if so, does that number matter?

Kathy Niakan’s group in the UK just published a knockout of the OCT4 gene in healthy human embryos that involved dozens of embryos, but only those leftover from IVF procedures, and another team, this time from China led by Zhou Songyang and Junjiu Huang, reported in a new paper that they actually cloned somatic cells to make human embryos via somatic cell nuclear transfer (SCNT), which they then modified using a somewhat different version of gene editing called “base editor” involving chemical modification of DNA.

Researchers in Sweden have explicit governmental approval permitting them to do CRISPR on healthy human embryos. Since they’ve had this green light for quite some time they probably are well into such work already and we can expect a paper from them soon. There also have been several older published papers on CRISPR genetic modification of human embryos (here, here, and here).

Taken all together this means that genetic modification of human embryos is rapidly becoming more common and that trend is likely to continue. Large-scale human embryo genetic modification isn’t going to be clearly ethically permissible in all cases. Some of the upbeat perspectives on this kind of research could be too aspirational with premature clinical objectives.

CRISPR ‘Magic’. Rob Stein was the NPR reporter invited into the Mitalipov lab. He was taken to the specific place in the lab for CRISPR of human embryos that Mitalipov reportedly referred to this way:

“This is our small room, but that’s where usually lots of big science happened,” says Mitalipov, who was born in the former Soviet Union. “We believe this room is really magic in terms of science.”

Magic? That’s a risky word for us scientists to use.

This small room is where they inject CRISPR into human embryos. The human embryos are often apparently made there earlier by IVF, including using donor eggs collected in the same building a few floors down or elsewhere.

Assembly line feel. Mitalipov and members of his team showed Stein the process by which they put CRISPR-Cas9 into the human embryos. One of the most striking aspects of Stein’s report was that the OHSU lab can attempt to CRISPR many human embryos over a short period of time. Stein reported after seeing a first CRISPR’ing of a human embryo that:

“Mitalipov and his team immediately do a second edit and then transfer the embryos to a larger incubator. The scientists will then spend the next few days monitoring live video of the two embryos, along with 17 others they had edited the weekend before, to see how they develop.”

Nineteen human embryos made and CRISPR’d in a day or two?

Even if this doesn’t happen every day, just this one lab alone could produce and genetically modify hundreds of human embryos in a short period of time. Is that scale of human embryo production and CRISPR for research ethically permissible? Who knows? It seems the justification is based in this case on hoped-for future clinical benefit. Mitalipov has indicated that he definitely wants to try this in the clinic in a heritable manner.

In addition, health risks to egg donors are often downplayed despite there being astonishingly little data on whether egg donation is safe, and anecdotal reports of cancer in young donors (see recent NYT article). The use of CRISPR in human zygotes should be understood to be transformative, both in potentially positive and negative ways. It is not just another experiment. It requires extraordinary care and sober reflection as well as transparency.

Even more transparency and openness about perspectives would be useful. It’d be great if Mitalipov penned a perspectives kind of piece to place his research here in a larger context, discussing things like preimplantation genetic diagnosis (PGD; a.k.a. “embryo screening for mutations”) and why he believes germline CRISPR is even needed (more on PGD below). I haven’t really seen him get into the meaty issues in interviews so far. For example, it’s still not clear to me why he thinks germline CRISPR would be better than PGD alone.

In the gray zone. In the last few years discussion of CRISPR human germline research has taken off. That’s great. For instance, there have been many meetings with wide-ranging discussions. These have yielded beneficial results in the form of greater awareness and exchange of diverse ideas, but we are a long way from having some clear framework for what’s permissible and what isn’t when it comes to changing the genetic code of human embryos and potentially humans. Such a framework may vary by country as well. Collectively as a field we are in a gray zone. The National Academies report released earlier this year did a nice job of outlining many challenges issues with human gene editing, but in my view it ended up not being too much in the gray zone too. They left the door open to human germline CRISPR.

Some are wanting to go through and quickly. “I don’t think I’m playing God,” Mitalipov says as quoted by NPR, “We have intelligence to understand diseases, eliminate suffering. And that’s what I think is the right thing to do.”

Don’t forget PGD. It is concerning just how often the already proven alternative of embryo screening by PGD is not brought up when people discuss heritable human genetic modification as a basis to try to prevent genetic diseases. Basically, in almost every clinical case one can imagine, there would be no point to CRISPR’ing human embryos because you can just simply screen for mutant versus non-mutant embryos without taking the big risks of genetically modifying them. In fact, you’ve got to do PGD when you do CRISPR anyway to know what is going on so why not just do PGD alone instead?

PGD isn’t perfect but from an unbiased viewpoint CRISPR has to have some very strong case for its superiority over PGD for the arguments for its eventual clinical use to make sense. If such cases cannot be made and PGD almost always is going to better, should we be CRISPR’ing massive numbers of human embryos based on the supposed justification of a reasonable expectation of safe and effective, future clinical use?

Advancing knowledge or wanting to go to the clinic? It’s important to point out that the research that Niakan and others is described as being done to advance knowledge, not to enable future germline use of CRISPR in humans, and again they used leftover IVF embryos rather than making new embryos by doing IVF themselves or cloning. Even if the Niakan team did not report a definite blockbuster scientific finding on targeting OCT4 in human embryos, I still think their work was interesting and I generally am supportive of that kind of research if done with a good scientific justification, strong transparency, proper institutional/governmental approvals, and bioethical discussions in advance. However, if many more such experiments go on to largely recapitulate what is already known from mouse studies without major unique human-specific findings then the case for use of healthy human embryos may become more difficult. See my ABCD plan for human genetic modification.

I asked Marcy Darnovsky, Executive Director of the Center for Genetics and Society about the Stein NPR article and the importance of including PGD in the dialogue about human embryo CRISPR, and here’s what she said:

“There’s no way to accurately think through the enormous safety and societal risks of human germline editing without knowing just how weak its medical justification is. You don’t need to engineer your future child’s genes to avoid passing on serious genetic conditions. The embryo screening technique PGD can accomplish that in pretty close to every situation. Embryo selection itself raises questions about what counts as serious disease and what’s just human variation. But the fact is, PGD is safe, effective, and has been available in fertility clinics for decades.”

Bottom line. As CRISPR work on human embryos is expanding around the world, societal and bioethical issues take on more urgency. Maybe we can’t or shouldn’t entirely hit “pause” on this kind of research even if it is just limited to the lab without clinical intent, but it feels like some folks have hit “fast forward” instead and there might be an almost race-like atmosphere, which boosts risks on many levels.

Broader discussion, transparency, and thoughtful contemplation are urgently needed.

7 thoughts on “Why Mitalipov’s ‘magic’ room for human embryo CRISPR may give us pause”

  1. NP Paul – I think the topic deserves some debate and I’ve spend some time on presenting some background on the issues and opinion previously.

    Yes, there is a modified issue or synthetic aspect to the editing area – not to say that anything outside of the internal process of a woman’s body isn’t man made (culture media for example in IVF). The point I believe, in terms of ethics, you are making is indeed distinct. In my mind I have no objection ethically to germline intervention if it solves a devastating inheritable disease. This is a question best weighted towards those families’ opinions that suffer from these human genetic roulette conditions. This doesn’t rule out viable alternatives but all options need to be explored imo, then evaluated and presented for their choice.

    The idea that developing and using genetic editing technology for the benefit of human reproduction seems to be something of a no go for you but that doesn’t necessarily mean that it shouldn’t be considered if and when the technology allows safe and effective intervention. I’m not a fan of draconian rules over medical innovation – freedoms, albeit with governance, to solve problems is far more effective historically.

    The idea that Mitalipov or anyone for that matter is racing ahead to produce genetically corrected babies is a sensational exaggeration. Just as the designer baby hype was over the top. No matter what it seems, even when published research work is presented that meets the scientific community’s stated interest in pursuing the research, the idea that it somehow is a fatal step towards a Gattica future is your own spin.

    My question to you is how can you support the science and the leading investigators constructively in this area of discovery when you don’t fundamentally believe it should be happening in the first place?

    The intent here is clearly to progress scientific knowledge and IF it’s possible to translate any findings into any practically acceptable applications for the benefit of patients then of course that is worth leaving the door open for. There will be no clinical translation without legal authorization – which as you know will be a long and uncertain road. Precedent in this area has already been laid.

    Fear of the unknown or uncertainty of rogue elements was the issue with cloning which hasn’t materialized as an issue after decades, so stopping discovery to contain the possibility that in all likelihood isn’t an issue is self-defeating and not practical anyway. The best way to control the science is to advance it within centers of expertise.

    I was referring to the pursuit of the science of gene editing of embryos that the NAS endorsed as well as others. The consensus was to move forward. The future it was stated would determine the technology’s clinical application but is now not anywhere near ready nor legally permitted, as was stated.

    The idea that supernumerary embryo numbers created in the IVF process could be reduced is of course conditional upon the technology working. Yes, the process would require use of donated embryos, whether by way of direct germ cell donation or via IVF fertility clinical donations.

    To put the issue of supernumerary embryo leftovers in numbers – how many embryos do you think are created in IVF fertility treatment centers annually and what number are leftovers frozen, discarded or donated? If any technology were perfected that reduced the need to create more than were implanted that would be a good things imo. Genetic editing has the potential to add a tool in the box of fertility efficiency way way in the future – but the process must be started and be supported to have that potential.

    The invalidation of the outcome due to the process doesn’t resonate given IVF has already set the precedent for embryo creation/destruction as a mainstream practice – research in this area would be minor comparatively, as was hESC line creation/research. That potential has been validated against a backdrop of similar objections.

    As far as I’m aware each IVF procedure requires eggs > millions of procedures…

    The US Federal Govt. draws a distinction on the creation of embryos for research but doesn’t outlaw the practice so yes there is a real-world distinction. However, I was pointing out that the pragmatic issue of outcome is the same for those that object to embryo experimentation which is what we were discussing. Both methods create embryos which will be destroyed so ethically that’s more than a bit of a grey area… Germline modification is not allowed for reproduction purposes.

    I’d like to read the spectrum of opinions here on the issues Paul so whatever you can due to stimulate the parties to contribute to the dialogue would be great.


    1. Hi Mike,
      If CRISPR or related genetic modification technologies can be proven to meet certain concrete criteria, then I could in theory potentially support their hypothetical future use in the germline in humans within the context of further diverse discussion about risks, social justice issues that could come up, & bioethical considerations. But it seems very unlikely. Then again, we’ve seen how fast technologies can evolve and improve.

      The criteria I have in mind include strong indications that there is some important disease-related application for which gene editing is clearly superior to PGD alone, that there is a very reasonable expectation of safety including across generations based on hard data, and that there is a reasonably strong expectation of efficacy, again based on rigorous data.

      The initial data in the handful of papers on CRISPR (and the one on base editor) use in human embryos so far seem a long, long way from meeting these hurdles. Of course, others may think my criteria are way off base (too strict, too weak, etc.) but to me they make good common sense.

      Various countries around the world will develop and continue to evolve their own approaches to oversight and in some cases laws related to this kind of research so there may not be a “one size fits all” approach to genetic modification of human embryos. We can already see this is the diversity of governmental approaches to this kind of research around the world today.


  2. Dear Admin:

    The following quote from your comments pretty well sums up the problem with how knowledgeable, expert, scientists can sometimes muddle the scientific facts to condone and promote research that is clearly unethical by current standards for research with human subjects in the U.S.

    “but we are a long way from having some clear framework for what’s permissible and what isn’t when it comes to changing the genetic code of human embryos and potentially humans.”

    Human embryos ARE humans.

    The U.S. National Academy of Science also failed humanity when they recommended conditions for which violating longstanding rules governing ethical treatment of human research subjects was recast as permissible – in the case of genomic editing performed to prevent an expected life-threatening deficiency in manipulated embryos. However, the Academy did not consider the many human embryos who would be destroyed in the research required to achieve such genetic treatments. What a terrible moral and ethical imbalance they set in motion in the U.S.! Certainly thousands of humans, and possibly millions, will die in the pursuit of genomic editing therapies before they are achieved, if in fact they can be. Moreover, collateral deaths in general experimentation justified under the stretchy umbrella of future disease-relevance will reach even higher numbers.

    When knowing scientists are willing to equivocate over the tautology that human embryos are humans, humanity is loss unless such blindness or hypocrisy is countered by other, better reasoned and less conflicted members of society.

    James @ Asymmetrex

  3. Science research on human embryos isn’t new and both Jeanne and you, as well as countless other knowledge investigators in the field, have been involved in hESC related lab work with institutional $ support. Granted less so now with iPSC but many labs still refer to the requirement for the ‘gold standard’ of hESC benchmark tests and the need for such, along with the translational programs.

    Hormonal treatment for egg retrieval to create embryos via IVF fertility treatment is a standard practice around the world today with more than 5 million healthy babies conceived. Obviously, women feel differently about the procedure given their desire to conceive and are willing to undergo the uncomfortable treatment (the published evaluation of the risks is worth a separate review posting).

    Also, eggs donations for fertility purposes to help other women is common, as are egg procurement procedures for surrogacy testing and specific donations for medical research. The ISSCR has guidelines on this: http://www.isscr.org/docs/default-source/2015-am-stockholm/eppeggprovision-haimes-0313.pdf?sfvrsn=2

    Supernumerary IVF embryos donated to science versus those created specifically for research result in the same end – destructive scientific practice for knowledge. I don’t believe it’s reasonable to purpose the moral edge to IVF leftovers, as irrespective of the intent the concept of lab work on them doesn’t merit consideration by those that object to idea of embryo experimentation or their use for the potential benefit of those in need.

    In so far as pursuing science that looks to effectively edit faults in germline cells for whatever purpose, whether it be basic research or potential translational investigation, the scientific consensus is that the work should proceed with dialogue.

    The point has been made that by correcting embryos safely it would go some way to reducing the need for multiples in the IVF process and therefore the ever-increasing number of leftover supernumerary embryos discarded

    PGD should be used universally and covered by insurance – and yes most current issues could be avoided if this were a reality but it’s not at the moment so touting its current practical value is moot. Along with that the substance of the above conversation on funding developmental science work to expand the diagnostics reality would be necessary…

    Social commentary is warranted as long as there’s an equal weight and voice to the various positions that this debate demands. Invitations to post might be a good way to frame the debate and engage.


    1. Thanks for the comment, Mike. You raise some interesting points.
      I would make a number of distinctions between human embryo use for derivation of hESC lines versus for genetic modification experiments such as using CRISPR. These aren’t equivalent and don’t have the same bioethical and societal issues. Both have potential issues, but only some overlap.

      For instance, Mitalipov has indicated if all goes well he fully wants to use CRISPR’d human embryos for reproductive production of new human beings, even though he doesn’t phrase it that way. There seem to be no if’s about that goal and others share that goal. Some talk about use of CRISPR in human embryos to improve IVF success rates….is that indirectly via knowledge or via changing embryos? CRISPR-based “fertility treatments” have been mentioned too, rather than for prevention of genetic disease.

      In contrast, the only applications I have heard of for hESC derived from embryos are in vitro research and potential cell therapy applications. Intent matters.

      When making a whole slew of CRISPR’d human embryos there is also the possibility — e.g. via publication of your methods — of enabling other labs with different ideas about bioethical or societal issues to create genetically modified people, perhaps even with the goal of changing traits.

      To my knowledge, there is no scientific consensus on pursuing CRISPR research on human embryos with future clinical intent. In fact, there’s a lot of disagreement about that. So I’d ask you to back up your statement in that regard with some sources like links, etc. Even this year’s NAS report, if memory serves, did not green light “human gene editing” research that has future clinical intent. They did leave the door open, but that’s not endorsing it.

      I don’t buy the whole “reduce the number of embryos needing to be discarded” argument because it just doesn’t make sense to me. To get to that kind of point where you have fully validated, proven safe CRISPR of human embryos in a reproductive setting with trying to make corrected mutations and hence not need to discard embryos, you’re going to have to do research probably on thousands of human embryos first, which invalidates the whole point of that kind of argument, right?

      In addition, I don’t think human egg procurement is as simple or common as your comment suggests.

      And I do think there’s an important bioethical distinction between using leftover embryos from past IVF procedures for research and de novo making new embryos via newly done IVF strictly for research use.

      I like the idea of guest posts with diverse views on this topic and will think about making some invitations. I don’t see many people voicing concerns publicly about potential human reproductive use of CRISPR and relating it to actual real life studies like the Ma paper (rather than hypothetical scenarios), but privately there is uncertainty.

  4. I can’t think of a single situation in which gene editing of human embryos for reproduction is justified. OK, I will agree that if one parent is homozygous for a dominant familial AD mutation, there would be no other way to fix it…except adoption. I think it would make more sense to put effort into making IVF and PGD more accessible. In the US, IVF and PGD aren’t covered by insurance, and in Germany, PGD is not allowed. Fixing the support system is a lot simpler than fixing the genome.

    1. I wonder how common homozygous mutant prospective parents-to-be are out there given that for some conditions a homozygous state is lethal, causes infertility, etc.

      Another possible rare scenario is if both parents-to-be carry the same mutation, but that could make CRISPR very difficult too.

      Finally, some have mentioned the scenario of a couple having so few embryos that PGD cannot reliably be done to ID an embryo without a mutation…again that’ll make CRISPR very hard or impossible as well as not all embryos into which CRISPR-Cas9 is introduced will pass QC. Probably few will.

      I agree on the health care issues. PGD really needs to be covered by insurance!

      Adoption is not mentioned often enough as well as an additional option.

Comments are closed.