A Mouse With A Human Brain ...

Via bioethics.net comes news of a National Geographic News article called "Animal-Human Hybrids Spark Controversy". The article is, in my judgment, really confused: confused in a way that makes it much more sensationalistic than it should be, and obscures the really interesting questions that human/non-human chimeras raise.

For instance, it says that "at Stanford University in California an experiment might be done later this year to create mice with human brains." Now, I don't know what, exactly, they are planning to do at Stanford, though I can guess. But it's really unlikely that they are going to try to make mice with human brains. Why? Well, for starters, think of the size problems. You have a little mouse body a few inches long, weighing maybe an ounce, and then attached to it a mouse head big enough to house an entire human brain, weighing a couple of pounds. Leaving aside such questions as, how would it walk? How huge would its neck muscles have to be? and so forth, just ask yourself how, exactly, a human brain is supposed to fit inside a mouse cranium, even assuming that crania have some capacity to expand early in development.

You might at this point be thinking: silly hilzoy! Obviously, what the article means is that they will create a mouse with a mouse-sized human brain, just as, if someone said they were going to build a Matchbox car with a working automobile engine, they would mean a Matchbox-sized engine, not a regular one. But how would this work? A mouse-sized brain made of human neurons would not be (what we normally think of as) a human brain, any more than something small enough to fit into a Matchbox car chassis, but made of (a small number of) normal-sized engine parts, would count as a normal working engine. The obvious solution would be to make the engine, or the brain, out of tiny replica parts. But we don't have tiny little replicas of human neurons. Nor is there any reason to think that it's even possible to create a tiny version of the human neuron that works the way a human neuron works, so that if we arranged those tiny neurons the way normal neurons are arranged in the human brain, they would work (a) at all, or (b) the way a human brain does. So the idea of a mouse with a human brain, in anything like the normal sense of that phrase, is just a non-starter.

Likewise, the article raises this (im)possibility (quoting a bioethicist who should know better):

"an experiment that would raise concerns, he said, is genetically engineering mice to produce human sperm and eggs, then doing in vitro fertilization to produce a child whose parents are a pair of mice."

Just try to imagine how the logistics of this might work. The two mice mate, and conceive a child. It begins to develop. Then what? Presumably, one of two things happen: the fetus dies, or the female mouse bursts. What could not possibly happen is that a female mouse could actually, literally, carry a human infant to term and then give birth to it. (Through a mouse pelvis? After spending the better part of the entire mouse lifespan pregnant with a child that would, at birth, weigh on the order of a hundred times as much as she does? Please.)

As I said, though, all this just serves to obscure some interesting questions, to which I will now turn. (Warning: it's going to be one of my wonky posts. But it will be interesting to me to see whether anyone makes it through, and if so whether they think I'm right.)

I am going to talk about human/non-human chimeric brains. Before we get started, though, there's one crucial distinction I need to get clear on: genetic vs. cellular chimeras. A cellular chimera is an organism that has two genetically different types of cells. Each of the cells is normal, given its source; the only unusual thing is that the chimeric being has two (or more) different kinds side by side. A genetic chimera, by contrast, is a being whose genes are made of the genes of two different species. (Thus, each cell has genes from different species.) I am not knowledgeable enough to talk about human/non-human genetic chimeras, so in what follows I will stick to cellular chimeras.

There is one particular kind of moral issue that cellular chimeras don't raise. Since genetic chimeras' genes are made up of the genes of two different species, if they reproduce they can pass their chimerism on to their offspring. Cellular chimerism, by contrast, is not heritable. If you have, for instance, a mouse cell in your pancreas, and you have a child, you do not need to wonder whether your child would be born with a mouse cell in its pancreas. It's your sperm or eggs that determine what your child is like, not your pancreas, and therefore if your sperm or eggs are human, your children will be too. Since a cellular chimera will not have chimeric offspring, the moral issues (if any) that are raised by the creation of a cellular chimera concern the question: should we bring this being into existence? and not, should we loose a race of chimeras on the world?

Cellular chimeras can occur naturally: women who have carried a child, for instance, often end up with some cells from that child in their bodies, and are thus chimeras. Likewise, anyone who has had a transplant or a blood transfusion is a cellular chimera: s/he has her own cells and cells from the blood or tissue donor in her body. When you graft a tree onto a different rootstock, you are creating a chimera. This is all perfectly normal, and no one has ever raised huge objections to it (other than Jehovah's Witnesses, who object only to chimerism involving human blood.)

Moreover, cross-species chimerism has existed for a while without seeming to raise huge ethical issues. Jesse Helms, for instance, was a human-nonhuman chimera after he got a pig valve transplanted into his heart, and yet no one seemed to worry that this made him, in any morally relevant sense, less than fully human. (He joked about it, though: he is supposed to have said: "Every time I pass a plate of barbecue, I cry. It might be one of my relatives." He still ate the barbecue, though.)

Now: why aren't most people troubled by the idea of people getting pig valves transplanted into them*, except maybe on a visceral level, which tends to dissolve once you consider that the alternative might be death? I think it's because, for most of us, heart valves don't really affect who we are, or what matters about us, except in that if they don't work, we die. What matters is having something that works in there, not having a specifically human something. And this is true of a lot of other body parts as well: if (let's say) I needed a pig liver or a chimpanzee toe to function, it might be sort of gross, but if it worked I wouldn't feel that it had altered who I was.

The one part of the body that really does seem to make us who we are is the brain. Now: no one is now proposing injecting animal neurons or neural stem cells into human brains, and if they did, I can't see how it would pass normal ethics review. But scientists are transplanting human neural stem cells (which go on to form human neurons) into animals. Is this wrong?

To try to answer this question, it's important to understand that how our brains function depends on their architecture. What does this mean? Consider the difference between a team of oxen and a car. A team of oxen can pull a lot of weight. This isn't because of the structure in which you arrange them; it's because each of the oxen can pull a certain amount of weight by itself, and if you put them together, they can pull a lot. The sole function of the way you arrange them is to keep them from getting in each other's way. By contrast, the bits of a car, separately, do not have any ability to pull anything. If you attach even a tiny weight to, say, a spark plug, the spark plug will not pull it. (Not even if you crack a whip over it and say "Eee-yah! Git!", the way they do in Westerns.) With cars, you aren't just adding up the pulling power that the parts contain independently. If you don't put those parts together in the right way, they can't pull at all. (Likewise with computers: it's the arrangement of the parts, the architecture, that matters; the individual bits of silicon and solder do not, by themselves, have the power to carry out even very simple calculations.)

The brain is like a car or a computer, not a team of oxen. Individual neurons do not have the capacity to think simple little thoughts; it's a large number of those neurons taken together and arranged into a brain with a given architecture that can think. Now: imagine, say, a mouse, with an existing mouse brain, into which you transplant one human neural stem cell. What happens to it? One possibility: the neural stem cell goes off by itself and doesn't integrate with the brain. (This seems to be what often happens when the brain isn't injured. No one really knows why.) In this case, there's no effect on the brain at all. Second, it gives rise to neurons that integrate with the mouse brain, and work well. (This often happens when the brain is injured: stem cells find their way to the site of injury. No one knows how. Stem cells are cool.) Since the mouse brain architecture remains the same, and the human neuron just replaces a mouse neuron, the mouse's cognitive powers, such as they are, remain the same. Third, it gives rise to neurons, they try to integrate, but they don't play well with the mouse neurons (say they don't do the same things in response to the same signals, or they use different neurotransmitters.) Then the mouse brain sort of breaks at the affected point; how much of a problem this causes depends on the point in question.

It's sort of like replacing a nand gate in a computer with a nand gate from a much better computer. If the two are functionally interchangeable, what you get is basically the old computer, capable of doing the old computer's operations, with, perhaps, a gate that won't fail as quickly. If not -- if the materials are somehow incompatible, for instance -- then that part of the chip stops working. What absolutely does not happen is that the old computer becomes able to do some of the things that the much better computer is capable of. It's the same with brains.

There is one wrinkle in this, however. How the neurons integrate depends on the chemicals around them and so forth: the environment in which they find themselves, and what that prompts them to do. If you introduce a few neural stem cells, this environment will of course be a mouse environment. But presumably if you introduce enough neural stem cells, and they integrate into the mouse brain, they will start to create their own environment. It's hard to see how this would happen, in an adult mouse, unless the mouse brain were already seriously injured, and if it's hard to see how a normal mouse brain could develop anything like a normal human architecture, it's really, really hard to see how a massively damaged mouse brain could. Also, if you introduce human neural stem cells into a mouse brain at a sufficiently early stage in its development, it can have greater effects than I've just indicated, both because there are fewer mouse neural cells for it to interact with, and because these are at a much more malleable stage. In these cases, and (as far as I can tell) only in these cases, could you get a mouse brain that had anything like a partial human architecture.

Of course, it couldn't be very much like a human architecture, since, as I noted earlier, mouse brains are just too small. The cases in which you'd have to worry about creating something human-like would be cases in which you introduce either massive numbers of neural stem cells into an injured brain, or human neural stem cells into a developing (fetal) brain, and the brain in question belongs to an animal whose normal brain is about the same size as a human brain. A chimpanzee, for instance, or possibly a pig. Again, a few neural stem cells introduced into an adult pig or chimp brain would not produce anything human-like, but introducing them into a developing brain, or introducing a lot of them into a massively injured one, might.

But that's still only a possibility. As before, in the case of an adult pig or chimp with massive brain injury, it's really, really unlikely that any number of human cells will bring it anywhere near human functioning. It's the case of introducing human neural stem cells into an early-stage developing brain of a large-brained animal that might possibly cause concern; and this is, I think, the only case in which we might create a partially human-like brain. But even in this case, it's much more likely that the animal would end up with a brain that was all screwed up: there are many more ways for this bizarre form of neural development to go wrong than for it to go right.

The other serious moral issue that this sort of thing raises is: what would it be like for the animal? Would it suffer more than it would have otherwise? This is a long subject in itself, and I don't really have the endurance to tackle it. But one point does follow from what I've said above: while it might suffer more because its brain didn't work right, it would not suffer because, for instance, it was a mouse with the cognitive powers of a human, and was therefore accepted neither by mice nor by humans, and felt alone, in a sort of permanent and unalterable exile. Nothing like that is in the cards, as far as I can see.

* Footnote: Here I mean: troubled by the thought that it will make them somehow less human. There are other reasons to be worried by transplanting pig stuff into us. The most obvious is that they might contain pig viruses; and if you think about how you might go about making it easy for a pig virus to infect a human, it's hard to imagine what would beat knocking out that human's immune system, cutting him or her open, and sticking the virus into a nice warm comfy place inside. But that's another issue entirely.

Comments

[votermom]

It's like these scientists never watch movies. First robot soldiers for Iraq. Didn't they watch Terminator? Now human-brained rats. Ben, anyone? :)

I think the safety issue of the experiment,as you mentioned, the problem of diseases jumping species is a serious one, (exhibit A: Mad Cow Disease).

[kenB]

The whole article you quote is silly, but just to pick a nit, re this bit:

then doing in vitro fertilization to produce a child whose parents are a pair of mice.

I think they're talking about implanting the fertilized egg into a human host, so the mice would be the genetic parents without doing the hard labor.

Regarding cellular chimeras, I wish I had something serious to say, but all I could think of was this old dirty joke.

[fdl]

Prof H.:

I think you're dancing around the idea of consciousness. What makes us human is our human notion of self-consciousness. Are other animals self-aware? hard to tell. can they describe to us what their notion of self awareness is? no. at least, not yet.

If human consciousness is inextricably linked to the physical architecture of the human brain, then at some point we need to be concerned that implanting enough human brain cells into an orangutan, for example, will lead to something that is actually part human . . . i.e., have quasi human consciousness.

that, i think, is the repellant part.

Francis

[M. Scott Eiland]

Uh oh--you just *know* where this will lead. . .

They're Pinky and The Brain
Yes Pinky and The Brain
One is a genius, the other's insane
They're laboratory mice, their genes have been spliced
They're dinky, they're Pinky and The Brain Brain Brain Brain Brain

Before each night is done, their plan will be unfurled
By the dawning of the sun they'll take over the world
They're Pinky and The Brain
Yes Pinky and The Brain
The Twilight Campaign is easy to explain
They'll prove they're mousy worth, and overthrow the earth
They're dinky, they're Pinky and The Brain Brain Brain
Brain Brain Brain Brain Brain Narf!

[hilzoy]

bioethics.net cited Mrs. Frisby and the Rats of NIMH, which is what I've always thought of in this context.

Francis: I think what I meant to be arguing is: human consciousness, insofar as we understand it at all, seems to depend on the architecture of the brain. If I'm right to say that introducing human neural cells will not produce human brain architecture in most cases, then I think it won't produce a sort of hybrid consciousness. Except possibly in the very specific case I mentioned at the end.

[hilzoy]

kenB: you're right, but now what I don't get is: in what sense could mice be "the genetic parents" of a human child? I go back to the article and see this: "genetically engineering mice to produce human sperm and eggs". Well, last time I checked, human sperm and eggs get their DNA from the organism whose sperm and eggs they are; they don't just up and create a whole new genome from scratch. Or rather: they can mix and match bits from each strand of DNA in the original organism, but they can't create a human genome when the existing organism is a mouse. So what this 'genetic engineering mice to produce human sperm and eggs' could possibly be, I don't know.

What is possible is to implant human eggs/sperm, or else human germ cells (the precursor to eggs and sperm) into a mouse, so that the mouse will grow human eggs or sperm. But then the mice would not be the genetic parents; the donors of the transplanted eggs/sperm/germ cells would be. And as I noted in the post, a mouse couldn't possibly be the gestational parent of a human child. So in what sense is the mouse the child's parents at all? All it seems to have done is incubated the eggs or sperm.

And why anyone would ever do such an experiment is a complete mystery. We can do IVF by much less roundabout routes.

[sidereal]

"I think you're dancing around the idea of consciousness. What makes us human is our human notion of self-consciousness. Are other animals self-aware? hard to tell. can they describe to us what their notion of self awareness is? no. at least, not yet."

What is the observable evidence you can draw on to determine whether an entity is self-conscious (besides wearing baggy clothing. Ha! I think we mean 'self-aware'). Personally, I think our notion of human consciousness is 2 parts anthropocentric hubris, 1 part abstract thinking ability. I'm quite sure that orangutans, for example, are conscious and self-aware in every way you can articulate to some degree or another and are demonstrably more self-aware than, for example, an infant or the severely handicapped.

[R J Keefe]

My grasp of brain physiology is weak at best, but doesn't the power of the human brain depend on the cortical folding that puts neurons located at some distance on the cortical plane in much closer contact? If the folds are the architecture, then the implanted human neurons are just disconnected little spark plugs.

Beyond that, there's the likelihood that consciousness is not some sort of internal brain phenomenon but rather a sensation arising from the brain's harnessing of the senses, one of which (if I only knew how to classify it) must be speech. (Let's end, in other words, the idea of the senses as passive inputs. Knocking somebody upside the head is a sensory output.)

Sorry to wander afield, but the post is provocative.

[hilzoy]

sidereal: I didn't address that part of francis' comment, and I agree with you completely. Chimps and bonobos seem clearly to be self-conscious -- at least they do things like pass the mirror tests. Note for those who don't know what these are: put some odorless undetectable dye on an animal's forehead, where it can't see it. (You can do this while the animal is anaesthetized, so that it doesn't know you've done it.) Put the animal in front of a mirror. Does the animal, seeing that the mirror-animal has a big red spot on its forehead, touch its own forehead to investigate? Or does it assume that the mirror-animal is just another animal with a funny spot? If the former, so the argument goes, it can form the thought: that's me! I have a spot on my forehead! -- or something equivalent to that. If the animal fails, it doesn't seem to me to follow that it has no such thought. It might, for instance, just not be very interested in spots.

Anyways, I don't know offhand what research has been done on orangutan self-consciousness, but I find it hard to imagine that they don't have it. The question would be, is it somehow 'humanized'? What would this even mean?

What I think might be unique to humans is the ability to consider, explicitly, their reasons for action. (I mean: thinking something like: yes, I want do do X, but is that a good reason for doing X? I'm also assuming that we can act on the result, even if we don't always, so that it's not a purely idle question.) This is quite important, and puts all sorts of power over your own conduct within your grasp, power you wouldn't otherwise have. E.g., the ability to construct an ideal to try to live by, or a general principle to act on. I think these are extraordinarily powerful cognitive tools. (Powerful in a sort of mathematical sense ('a powerful result'), not a large guy with muscles sense.)

[Stan LS]

Great, just great! They are going to grow human brains in mice and outsource all the white collar jobs to the labs :(

[CaseyL]

Further on in the article hilzoy links to, the researcher working on 'human brains in mice' says he's already created mice whose brains were 'about 1%' human. But the article doesn't say in what way those brains were human: whether they simply contained 1% human brain cells, or whether the brain structure was 1% human (in terms of cortical folds, say). And it doesn't say anything about how that 1%, however it was measured, affected the mouse's thought patterns or behavior.

The same researcher does say he wants to try creating a 100% human brain in a mouse precisely to see if the mouse exhibits 'human cognitive behavior.' If he can pull it off, and the mouse does exhibit human cognitive behavior, then we've got a huge moral dilemma on our hands:

If something can think like a human, does that make it human? How do you define what 'thinking like a human' is? By cognitive potential? By genome? By the so-far unquantifiable criterion of 'self-awareness'? By capacity for moral thought?

Part of me is appalled by the idea of creating human-like consciousness in creatures that have no rights. Creating a human brain in a mouse for the explicit purpose of seeing if that mouse can achieve human cognition... and then experimenting on it and dissecting it... strikes me as profoundly criminal.

But another part of me is fascinated. Human biological processes are only possible because we ('we' meaning just about all of the macrofauna) are walking colonies, really, of various lifeforms. The micro-flora and -fauna in our digestive tract; the mitochondria in our cells: they're not 'us,' they're autonomous things that live in us. I vaguely remember reading somewhere that even our brain function depends on a colony of autonomous critters that lives in our synapses or axons or something - but I don't remember specifics; maybe someone else here does.

And viruses, my god, viruses! They're Nature's way of getting the genetic material from one species into another. Viruses might be one mechanism for how we acquired the colonies that make us possible. Are viruses essential to evolution? Is there a limit to evolutionary viral transference, past which an organism can't function? What might 'humankind' look like in, say, 5 million years, or 10 million, if we continue to evolve?

So that part of me - the one that wonders how much evolutionary change, and how much radical change, is possible - is fascinated by these experiments.

[Slartibartfast]

Relax. This is just more of mice running experiments on us.

[nagoya ryan]

I, for one, welcome our new mouse/human masters.

[Anarch]

Slarti, that slays me. Absolutely slays me.

Seriously: how long have you been waiting to use that line?

[Slartibartfast]

Given the abundance of opportunities for humor presented by the monumental blockbuster that is HGTTG, I may still have a few left to use at the Restaurant.

Glad I was able to tickle you, Anarch. That'll be $12.50, plus whatever sales tax applies where you live.

[sidereal]

Speaking of. . .

Is there anyone not infinitely stoked/worried about this?

[Slartibartfast]

I've done repeated simulations, and determined that that movie could not have been made in the time allotted.

Still, I'm at 11.

[kenB]

I'm not getting my hopes up. I'm rather afraid it'll turn out to be almost, but not quite, entirely unlike the original. (Radio series, that is).

[Phil]

I'm cautiously optimistic, but I'm still curious, as I posted in my own blog, as to why Zaphod appears to have the wrong number of nearly everything in this photo. Still, Stephen Fry as the voice of the Guide, and Alan Rickman as Marvin are wonderful choices.

[Dianne]

hilzoy: "no one is now proposing injecting animal neurons or neural stem cells into human brains, and if they did, I can't see how it would pass normal ethics review"

At risk of shocking everyone, I don't see this as necessarily unethical. For example, suppose that one wanted to try to cure Huntington's chorea, Parkinson's disease, or any other disease where a small part of the brain goes wrong but the cortex (generally considered the concious, thinking part of the brain) was intact. One method that has been proposed and, to a certain extent, experimentally used is to transplant neural tissue or stem cells into the affected area of the brain to try to regrow that area.

Now, suppose that it might be easier or safer to use mouse neural stem cells to regrow the substantia nigra or whatever part you want to fix. (For various technical reasons, I suspect that human stem cells would be the better choice, but for the purposes of arguing the ethical point, please assume for the moment that mouse cells work better for some reason.) Would it be wrong to inject stem cells into someone's brain, regrow the affected area, and allow them to live a relatively normal life, just because the stem cells used were non-human?

[Dianne]

BTW: Porcine valves are treated with some fairly nasty chemicals (formaldehyde, I think, although I'm not sure) to make them non-immunogenic and non-infectious. So the risk of acquiring random pig viruses is probably lower than it would at first appear.

[hilzoy]

Dianne: I wasn't clear; when I said that injecting nonhuman neural stem cells into human beings probably wouldn't make it past normal ethics review, I was thinking about the combination of (a) safety issues and (b) lack of obvious clinical rationale, leaving aside any special moral issues that injecting these cells into humans might raise.

Is it only the valves that are so treated? Can they do that with e.g. livers?

[Anarch]

Given the abundance of opportunities for humor presented by the monumental blockbuster that is HGTTG, I may still have a few left to use at the Restaurant.

Yes, but how many others are there (beyond the obvious fjord jokes) that make essential use of your handle?

[Dianne]

"Is it only the valves that are so treated? Can they do that with e.g. livers?"

Unfortunately, no. The problem is that while the valve can be made of dead and fixed tissue (it only has to act as a mechanical barrier which allows flow only at the right time in the right direction) but a transplanted organ has to be alive and capable of doing all the complex things cells do. So the valves can be fixed in formalin and still work, but transplanted organs have to be fresh and healthy (which, unfortunately, also means immunogenic and potentially infectious.) The point of a pig valve as opposed to a mechanical one is that the body's coagulation system sees the pig valve's surface as something that is supposed to be in the circulatory system and therefore doesn't make clots around it like it would a mechanical (ie metal or plastic) valve. Does that make sense or have I thoroughly confused the issue now?

[hilzoy]

No, that makes perfect sense. If your answer had been 'yes, they can do that with any tissue'. then I would have been thoroughly confused, since I have always been given to understand that there are serious problems with non-human viruses and xenotransplantation. As it is, I'm not. Thanks.

[Koneko]

I was going to say the same thing about the child with mice parents. Fertilisation would occurr in the mouse and before the embryo gets anywhere near the stage of being embedded in the mouse uterine wall, it would be transferred to a human host.

This is done a lot when creating transgenic animals, especially in the livestock industry. I've not actually heard that it has been tried, or is going to be tried in humans

[Gary Farber]

"Seriously: how long have you been waiting to use that line?"

Since he chose his name, ten million years ago, I figure.

[g]

"I've done repeated simulations, and determined that that movie could not have been made in the time allotted."

What movie?

"I was going to say the same thing about the child with mice parents."

The same thing as what?

[Gary Farber]

Oh, oops, 3 year old thread. Never mind.

[Annie]

By transplanting human neurons into a mouse with almost with no mouse neurons would create a human consciencness in the mouse.
As far as conducting experiments on a mouse with a LIVING brain (with human consciencness!) within an ANIMAL lab is simply unethical.

[hi]

hi
hi