Thread by @ToriHerridge

Okay, I am picking up my *cough* mammoth thread from yesterday.

If you are interested in the story so far, I talked a bit about the current main justification given by Colossal for their plans for "mammoth de-extinction" here:

https://twitter.com/ToriHerridge/status/1437434059194523656?s=20

🧵/1

other justifications include the collateral benefits of:

🧬 biotech development
🧠 increased knowledge of mammoth & elephant biology
🐘 improved elephant husbandry & conservation
🦏 conservation of species on the brink of extinction
👩‍🔬inspiring the next generation of scientists

These are all quite hard to quantify (and each also raises a number of different ethical considerations), which makes the problem of ⚖️ harm and good tricky.

So I am cheating for now, and leaving these for later (if I still have the will to live).

🧵/3 [forgot to number /2]

Ethical considerations need to address the balance of harm/benefit in relation to the individual subjects involved.

And so I want to talk about elephants (and any newly created GM Arctic-adapted elephants).
🧵/4

The Asian 🐘, Elephas maximus, is the extant species most likely to be used.

Asian elephants are endangered, & their numbers are decreasing. Once widespread across Asia, their habitat is increasingly fragmented & encroached upon by humans

iucnredlist.org/species/7140/4…
🧵/5

Asian elephants will be(/have already been) used as follows:

🧬 their genomes sequenced
🧪 their stem cells harvested
🧬 their genomes edited using CRISPR to insert mammoth DNA at key gene target sites

They may also be used:

🥚 for egg collection
🤰 as surrogate mums

🧵/6

If we were talking about humans, each step would require consent (although humanity has a long history of not bothering with that, even for humans).

Elephants can't consent. Thus we have a moral duty to assess the cost-benefit ratio of each step to them as individuals
🧵/7

genome sequencing, stem cell collection & CRISPR-editing pose minimal *additional* harm to the subject.

[I say additional, as captivity imposes its own problems, but so long as elephants weren't brought/bred into captivity for this explicit purpose it's an upstream issue]

🧵/8

egg collection (AFAIK not yet done successfully in elephants, but please elephant vets chip in if I am wrong) would be more intrusive/experimental, requiring hormonal stimulation of ovulation, sedation and probable surgery.

🧵/9

here is a diagram of the Asian elephant reproductive tract to give you some idea of the practicle challenges involved in reaching the ovaries.

From: nottingham.ac.uk/biosciences/do…
🧵/10

[on a personal note, I had my own eggs harvested for IVF, and it was a pretty shitty experience. Obviously I decided it was worth it. But then I got to keep my baby]

Recognising these issues, Church has posited creating elephant eggs from stem cells.

This has been done for mice, but has yet to be done for elephants AFAIK:
science.org/doi/abs/10.112…

🧵/11

Once we have an elephant egg in the lab, following advances in either of those methods [⏰], the edited nucleus will be inserted, & initial cell division would --I presume -- commence in vitro, as with human IVF.

This would allow for embryo screening before implantation
🧵/12

It's when we get to the implantation stage that the ⚖️ tips heavily to the cost side of things, and keeps on tipping

(full disclosure, so you are primed for bias, this is my big sticking point)

🧵/13

Ut-Oh

CORRECTION -- should have done my fact checks! Colossal says it will use **African** elephants at the surrogacy stage.

colossal.com/mammoth/

🎵school run interlude music🎵

right before we get to surrogacy, we are going to back track a little bit to gene editing with CRISPR. This is a VERY cool technique pioneered by Jennifer Doudna And Emmanuelle Charpentier, who rightly won a Nobel Prize for their work.

🧵/ummm?15?

CRISPR has revolutionised gene editing, and I am not going to go in to the details (here is a link for those wo want an explainer: wired.co.uk/article/what-i…).

The critical thing is it means multiple genome edits can be made quickly and relatively inexpensively.

🧵/16

The editing bit is straightforward(ish).

The trick is knowing what genes to edit.

If you want to synthesise an elephant that can survive in the Arctic, you might want to look at genes associated with:

hair, fat, temperature regulation, haemoglobin, circadian rhythmn

🧵/15

And probably also:
tooth development (grazing adaptation), digestion (ditto), tusk growth (not just for the aesthetics, but for the role they may play in habitat modification and social interactions/signalling), ear size, tail length (though you could just remove these!).

🧵/16

Here's a fun article by @PatriciaChrzan et al, with a nice summary picture:

sciencedirect.com/science/articl…

And AFIAK (?@DevoEvoMed) we are still a long way from moving from potential candidate genes (as identified by Lynch et al here: sciencedirect.com/science/articl…) to actually KNOWING what genes lead to what physical or metabolical feature.

🧵/17

Thus the gene-editing stage is also experimental, and likely play a significant role in shaping our understanding of mammoth & elephant biology and genetics.

Which I can't help but be excited about if I am honest.

🧵/18

And just like Christmas, @DevoEvoMed has done a brilliant thread on current state of knowledge on elephant/mammoth gene function.
👇
🧵/19

https://twitter.com/devoevomed/status/1437813973198442506

So, assuming:

🧬 functional genes are identified
✂️the CRISPR editing with these goes well (@DevoEvoMed is sceptical)

a critical question for me is: at what point from here do we actually know that those genes are doing what it is hoped they will do?

🧵/20

gene expression can be checked in vitro using various methods (This is outside my wheelhouse; see this explainer for more information: news-medical.net/life-sciences/…).

But stem cell gene expression ≠ certainty of mammoth-like form or function... It's doing something, but what?

🧵/21

Which is why I wanted to backtrack to the potential issues & unknowns associated with the gene editing stage.

While this stage poses minimal harm when confined to the lab, once it moves to implantation, the viability of the embryo will impact the risk level for surrogates
🧵/22

When assessing the harm-benefit ⚖️ to a surrogate African elephant female, we need to consider:

1. risks associated with cross-species pregnancy (this is a modified Asian elephant egg)

2. the likelihood of failed pregnancy (burden will increase through gestation)

🧵/23

plus all the usual risk factors associated with assisted pregnancy in captive elephants, bearing in mind:

* Labour issues & stillbirths are relatively common in captive elephants.

sciencedirect.com/science/articl…

[CW for images, they are quite upsetting]
🧵/24

* procedures are also invasive, causing pain & distress, and increased risk of infection

I am guessing the implantation procedure will be an extension of that of artifical insemination (see below; switch semen for egg), passing through the cervix

rhinoresourcecenter.com/pdf_files/124/…
🧵/25

for extant, captive elephants, these risks can be justified on the basis of:

* the elephants own reproductive needs and rights, leading to improved animal welfare

* the benefit to threatened or endangered species

🧵/26

Do these same benefits accrue to the surrogate elephant if the foetus/calf is:

⚖️ Unrelated (even at a species level)?

⚖️ Destined to be removed from the mother-calf relationship by translocation to Siberia?

In my opinion the balance tips to harm.
🧵/27

As an aside, one possible reason the African savannah elephant may have been chosen as a surrogate is the lower incidence of labour problems -- but this likely reflects the lower success rates for captive breeding in African elephants.

So there's another potential issue
🧵/28

Artificial Womb Technology has been proposed by George Church as an alternative to surrogacy.

Current artifical womb technologies include:

The biobag nature.com/articles/ncomm…

and EVE therapy ajog.org/article/S0002-…

🧵/29

The biobag has a better success rate than EVE therapy, & technological development is ongoing.

Both have only been used in late gestation at a stage equivalent to viability limits in humans (~22weeks).

Artificial womb technology is only a partial solution to surrogacy.

🧵/30

based on the appearance of the elephant foetuses here, I suspect you'd want to go beyond 24 weeks at least:

🧵/31
ncbi.nlm.nih.gov/pmc/articles/P…

Experts like Hank Greely, and @ECRomanis (who was kind enough to chat with me about AWTs, and the ethical and legal issues involved) are convinced Artifical Wombs are >10 years away

https://twitter.com/HankGreelyLSJU/status/1438016162667765760?s=20

🧵/32

Chloe Romanis' (@ECRomanis) work also introduced me to another aspect of the ethics of artificial wombs:

The implications for reproductive rights. It's really interesting. Read her work.

jme.bmj.com/content/44/11/…
🧵/33

@ECRomanis So, the use of artifical womb technology requires:

1. significant lead-in time
2. likely use of surrogate for >24weeks
3. experimental deployment as part of the de-extinction process, with unknown morbidity/mortality risks, not least...

🧵/34

4. how to transfer of fetus to the artifical womb.

Remember, C-sections are 100% fatal in elephants.

[from sciencedirect.com/science/articl…]

🧵/35

IMO, within the timeframe proposed, it would be impossible to avoid the use of multiple elephant surrogates in highly experimental procedures, with:

🏥 likely clinical risks, balanced by few --if any-- benefits, to the surrogate elephant

☠️ significant calf mortality

🧵/36

Before we get to the 🐘🧬🦣 calf, & its own rights/needs, a quick dip into research ethics to end.

I don't think any of the parties involved take animal welfare lightly, & so the more likely scenario is the work will never move beyond the lab bench
🧵/37

The potential collateral benefits to biotech development, knowledge, & conservation that might come from the lab bench stuff *are* exciting it their own right.

I think it is important & fair to acknowledge that, in reality, a balance of good will come from the lab stage.

🧵/38

Of course, we should ask ourselves why it needs to be framed as "mammoth de-extinction"

But, most likely, you are only reading this -- and I am only writing it -- because it was framed as such.

And therein, maybe, lies the answer...?

🧵/39

but anyway, research ethics.

And I am going to go back to the proposed research plan (which can be seen here colossal.com/technology/ -- scroll down!) of gene editing 🧬-->implantation 🥚-->gestation 🤰 (surrogacy and/or AW)-->birth 🦣-->introduction to Siberia🏞️

🧵/40

the legal framework & ethical framework for the use of animals in research varies between countries, and the UK -- where I am from -- is particularly stringent. This will inform my perspective.

🧵/41

In the UK, we often talk about "the 3 Rs", which are:

REDUCE
REPLACE
REFINE

nc3rs.org.uk/the-3rs

And you have to justify your proposed animal-based research, with cost/benefit analysis, in relation to these.

🧵/42

this exhaustive report from @Nuffbioethics on the ethics of research involving animals is superb.

A few screen grabs of pertinent stuff below

nuffieldbioethics.org/publications/a…
🧵/43

Tag yourself...

#AnythingGoes #OnBalance #MoralDilemma #Abolitionist

🧵/44

I think I'm an #OnBalance, with a sprinkling of #MoralDilemma. And with Moral Dilemma, an interesting aspect, often missed, is the moral dilemma over failing to act...

Anyway. let's not get too abstract.

🧵/45

We can assess Colossal's plans using this framework:

(i) Goal: to create, & birth, a gene-edited elephant
(ii) Probability of success: very low
(iii) Animals: elephants
(iv) Effects: pain, distress, metabolic burden, risk of disease and death
(v) alternatives? Not yet.

🧵/46

point (ii-iii) set the bar very high on the cost-benefit balance ⚖️.

point (v) relates to artificial wombs; delaying until these are available ticks the 3Rs, though the proposed benefits to humanity may have reduced by that time given rate of Arctic warming

🧵/47

So the critical Q is: do the justifications/potential benefits exceed the high-bar set by points (ii), (iii), (iv)?

I would say no.

[see my previous thread:

https://twitter.com/ToriHerridge/status/1437434059194523656?s=20

]

🧵/48

Other opinions are available, though, and the internal ethics review board (or the Institutional Animal Care and Use Committee, as I think it is called in the US , see: ncbi.nlm.nih.gov/books/NBK24650/) at the involved institution (?Colossal's own?) might decide otherwise

🧵/49

This comes back to diversity of views and beliefs on animal research, & the role for public opinion in shaping regulation.

I'll come back to the importance of a public voice in decision making later, but fagging here as it is key to ⚖️ assessments.

Your opinions matter.

🧵/50

Have I bored you all with the mundanity of red tape? I hope so. I like that, no matter how out-there research sounds, there are frameworks to work with.

Celebrity science shouldn't get a VIP pass.

An excellent paper by Carrie Friese & Claire Marris journals.plos.org/plosbiology/ar…
🧵/51

So my summary of this entire, ridiculously long, thread is:

I think it is highly unlikely that research into "mammoth de-extinction" will move beyond the lab bench in the next 5 years, based on practical concerns & standard research ethics approval procedures.

UNLESS...

🧵/52

UNLESS a much stronger case is for the benefits (to elephants and humanity) of creating an Arctic adapted elephant.

So the ball is in Colossal's court there.

Convince me.

Or more importantly, convince the world.

🧵/fin