Despite several obstacles, writer Philip Ball has finally witnessed the successful growth of his neural organoid or “mini-brain”. What started as a study of dementia and its implications for the brain has developed into a wider scientific, ethical and philosophical exploration of consciousness, biological identity and personhood. Here Philip reflects on the final stage of his ‘brain in a dish’.
I had never imagined quite how beautiful my neurons would be. Not mine in particular – anyone’s web of nerve cells, when stained with fluorescent dyes that reveal their signature proteins, would doubtless look as marvellous in their intricate complexity. But we don’t, as a rule, get to see them. Our life depends on their staying safely inside the cranium to exchange the molecules of thought. I have been truly privileged to witness these neurons that bear my genome and that grew from my flesh as they spread across their little dish.
I have been lucky too. I got my “mini-brain”, but only just. Biology being what it is, nothing can be guaranteed even in the hands of the most competent experimentalists. Many of the induced pluripotent stem cells (iPS cells) that UCL researchers Selina Wray and Chris Lovejoy produced from the participants of the Brains in a Dish project failed to transform successfully into neural cultures and the organ-like structures called “organoids” – a faulty or contaminated batch of reagents seems the most likely explanation.
But at least one of the cultures made from my fibroblast cells made it all the way, developing into a mass of neurons with the rudimentary brain-like structuring characteristic of an organoid. Chris stained and sectioned this sample to obtain several gorgeous images: red, green and blue constellations with the telltale neural layering we were looking for.
Chris and Selina also grew several two-dimensional neural cultures, which more readily reveal the tangled, interconnected networks of cells. The three-dimensional organoids are in some ways the most provocative of structures, but the 2D layers are the most evocative, showing in detail the kinds of pathways along which our thoughts travel and out of which our consciousness somehow emerges.
No one knows quite how that happens, but it seems to depend on several key properties of the neural network: the sheer number of cells, their density, diversity, and degree of connectivity. It is only via these somewhat crude measures, I discovered at a meeting on the ethics of brain organoids in Oxford in June, that we can gain any sense of how distant these structures are from any suspicion of hosting genuine “thought” – and of what might be needed to approach that fuzzy boundary.
I began this project feeling clear that brain organoids, for all their wondrousness and value to biomedicine, are not in any real sense small brains. I had no fear that I would become concerned about their ‘welfare’.
Nor did I. But I am less sure now about two things. First, the ethical questions about the possibility of consciousness that seemed no more than amusing philosophical games are already now being seriously debated by knowledgeable experts. This is not because anyone is about to make a mini-brain that poses a problem of this nature (both the motivation and the technique are lacking), but because it at least no longer seems absurdly speculative to imagine that scenario, and it’s wise to stay ahead of the game.
Second, I feel more strongly the artificiality of my sense of biological identity. That’s to say, this sense is a real thing: it seems no more meaningful to call it an ‘illusion’ than it is to deny the emotions we experience. But I can no longer map that sense onto any physical description of the human body. I am more conscious than ever of our composite and contingent nature: we are communities of cells. Those rare biologists who think hard about the question of individuality seem to agree that it has no hard-and-fast meaning, despite the individual organism being indispensible to evolutionary theory.
Here’s what I mean about these two things. Among the reasons why a brain organoid is not really a mini-brain are that:
It lacks the true morphology of a brain – the shape and organization of parts. It is like a rough schematic and unfinished sketch of a brain.
It is smaller: typically a brain organoid has a few million neurons, compared to around 86 billion in an adult brain.
It has less complexity – a tiny fraction of the typically 1014 synaptic connections in an adult brain – and a neuron firing rate of just 3-4% of that in the brain.
It has no interface with the world: no sensory input or connection to other organs.
The shortcomings of morphology are largely due to the fact that, like most tissues in the body, a developing brain in an embryo and fetus needs to be guided by signals from the surrounding body in which it grows. In a petri dish these signals are absent. The limitations on size are due to the fact that a brain organoid can only grown to around the size of a small pea before cells in the interior become starved of nutrients in the culture medium, and die. The absence of sensory functionality isn’t just a question of inputs and outputs: neurons actually depend on getting sensory input during a critical period of brain growth in order to develop properly.
But some of these limitations are already being overcome. It’s conceivable, for example, that some of the developmental signals in a real embryo might be mimicked in vitro by the right chemicals to improve the morphological similarity to a genuine brain. And just a week or so after my brain organoid was grown, two groups reported success in growing vascular networks – blood vessels – in human brain organoids by grafting human brain organoids into adult mouse brains. This approach of grafting into host animals has already been demonstrated for other human organoids – for example, for liver organoids grown in vitro as “organ buds” (that is, to a point where the iPS cells are committed to a “liver” lineage) and then transplanted into mice.
There is ongoing research aimed at growing full-sized human organs such as pancreases in large animals such as pigs and cows. One of the most promising approaches so far creates a “niche” in the host animal by using either mutant strains or gene editing to produce host embryos that lack an intrinsic ability to grow those organs themselves. If iPS cells from another species are added to the embryos at a very early stage in development, they will “exploit” the “gap” by developing into the missing organ. This has been achieved already with mice-rat chimeras and preliminary results show that human iPS cells can survive in pig and cow chimeras, although the efficiency with which they grow is not high at this point.
That seems unlikely to provide a fundamental barrier, however. The brake on such studies at this stage is ethical and regulatory: at present, the US National Institutes of Health has imposed a moratorium on federal funding of research that makes human-animal chimeras, while the ethical questions are thrashed out.
As for interfacing brain organoids with the environment: quite aside from ongoing work to connect neurons with artificial electronic systems, in a recent talk at University College London, Madeline Lancaster of Cambridge University reported that she and her co-workers have observed mechanical activity in muscle tissue stimulated via connections to a brain organoid.
It’s worth noting too that consciousness is not a property of the complete human brain. We might still not know quite what consciousness is or how to define it, but one thing is for sure: we know it (in people) when it’s there, and rather amazingly, we know that the neurons that create it are localized in one particular region in the cortex. The cerebellum, meanwhile, accounts for fully 80% of all the neurons in the human brain, yet people can live and survive and experience consciousness (despite other cognitive impairments) with severe lesions to the cerebellum, or even without a cerebellum at all. So what if a way were to be found to stimulate specifically the consciousness-hosting part of the brain in a human brain organoid?
With all this in mind, talk of a pig with a human brain at the Oxford meeting, while speculative, was not absurd in principle. It’s time to start wondering, at the very least, whether brain organoids might at some stage develop something like sentience or feeling. When a neuroscientist of the stature of Christoph Koch, once a collaborator with the discoverer of DNA’s gene-encoding nature Francis Crick, says that we have to start thinking about the point at which a brain organoid might feel pain, you have to take notice – and ask what regulatory measures need to be in place.
Again, let there be no doubt: no one is trying to grow a “human brain in a dish”, nor does any researcher have any good reason to do so. But it’s up for discussion.
Meanwhile, the very existence of human-animal “chimeras” – and they have been around for decades now in research, mostly consisting of human cells in rodent hosts – creates questions about biological identity. The notion of species is in any event approximate; cross-species hybrids such as ligers and tions, the result of unions between lions and tigers, are well known, and some hybrids occur in the wild. But these are genetically uniform, their genomes a fusion of those of the parents. Chimeras, on the other hand, are trans-species genetic patchworks, and show that even rather extreme genetic disparities between cells (humans and mice diverged 90 million years ago) needn’t prevent them from getting along together in a single organism.
The fact that we are communities of diverse cells is inherent already in our dependence on our symbiotic bacteria, especially within the gut – we are not viable organisms without the collaboration of cells not “our own”. But the capacity now to take our cells and keep them alive and proliferating not just in vitro but in the bodies of other organisms brings home with special force the gap between a sense of individual identity based either on genomics or bodily integrity, and the biological reality of what cells can accomplish.
The opportunities for exploiting this fact in biomedicine are exhilarating. Organoids such as “mini-brains” are tremendous models for investigating disease and testing drugs. The culturing of tissues both in vitro and in surrogate hosts truly has the potential to transform organ transplantation and human health. The ability to edit genomes with precision takes such possibilities to another level: iPS cells from a patient could have a faulty gene “corrected” in order to grow functional tissue-compatible organs that the intended host cannot create for him- or herself.
But these advances face us too with a new perspective on the body as a collective of cells that operate with a combination of autonomy and mutual dependence. Actually this is not really a new perspective at all: it was precisely the one advocated by the nineteenth-century architects of cell theory such as the physiologist Rudolf Virchow, and by the pioneers of tissue culture such as the surgeon Alexis Carrel in the early twentieth century. “Cells congregate in societies, which are called tissues and organs,” wrote Carrel in 1935.
As the “cell view” of biology became eclipsed by the “gene view” in the later twentieth century, this picture was obscured. We were – and still are – encouraged to regard ourselves either as mere vehicles for genes, or as individuals defined by our genomes. But today, cells are coming back. They are the future of biology and medicine: not merely our “building blocks”, but entities with a life of their own, from which our own arises.
This is the third and final post from Philip Ball in a series of articles, reflecting on his experiences as part of Created Out of Mind's Brains in a dish project, which aims to stimulate curiosity about the healthy and ageing brain, what might happen when a neurodegenerative condition like a dementia occurs and the cutting-edge technologies providing new insights into dementia.