I think that I have a shallow but accurate understanding of CRISPR (Cas9). What I thought were misunderstandings others had (around somatic vs. germ cells) are now coming from the scientific community. I hope and assume that they don’t misunderstand it, but perhaps they are riding a wave of the public’s misunderstanding toward an anticipated cusp of popularity, assuming that the knots in perception will get worked out later.
Before getting into the CRISPR misunderstanding, here’s a similar misunderstanding. When people first heard of human cloning, some thought (perhaps fancifully) that it meant a fully mature adult could be replicated as a fully mature adult. Science fiction movies were made and so on (e.g., digital cloning). But cloning is a simple concept and doesn’t mean that at all. It doesn’t mean a third identical adult Winklevoss brother could be created in an artificial womb, but it does mean another zygote / blastocyst / embryo / fetus / baby with the same DNA as the twins could probably be made with normal human gestation time and birth.
My understanding of CRISPR (specifically CRISPR/Cas9) is simple: it’s a gene-editing tool that changes the DNA of individual cells, like genomic search-and-replace, thereby modifying the cells before they perform their function(s), such as replication (cell division in non-neurons) and protein creation (gene expression). Most of the discussion out there I think is around somatic cells (i.e., cells throughout the body, in which case offspring are not affected), though I think the same general principles apply to germ cells (i.e., cells used in reproduction).
2017-10-25 note: CRISPR/Cas13 operates on RNA instead of DNA, which takes it out of a gene-editing role and puts it into a more traditional gene-silencing role.
Scientists are coming forward and implying that CRISPR can change the billions of copies of one’s genome in the mature human’s brain. This is 100% pure science fiction to me, with the same level of absurdity as cloning adults into adult replicas.
Whether intended or not, scientists make it sound like a mature organism’s genome is in a single, tweakable control center: make one tiny change in one tiny place and change a mature person’s genetic destiny (and that of their offspring) forever. Easy. Done. They gloss over that a mature person comprises trillions of copies of their genome, throughout the cells of their body. Only at the fertilization stage is there a genomic control center where changes propagate, and one’s genome cannot be reliably changed even into the zygote stage. They gloss over entirely how genetic changes propagate to offspring.
Here’s an analogy. If you are a farmer with acres of corn plants, you don’t look out over your cornfield and think about genetically modifying the countless multitudes of corn cells in your fields. If you’ve already planted seeds, then it’s way too late to be thinking about modifying their genomes—unless you are planning for future crop generations. The place and time to modify your corn genomes is in the seeds, before planting them. It can only be the same for all cell-based life on earth. It’s just common sense.
Coming to terms
Gene drive: the phenomenon in which the inheritance of a particular gene or set of genes is favorably biased.
Germ cell vs. somatic cell: a cell that gives rise to gametes vs. a differentiated cell in the body.
Germline: a series of germ cells each descended from earlier cells in the series, regarded as continuing through successive generations.
Meiosis: a type of cell division that reduces the chromosome number by half. In the male, this results in sperm. In the female, this results in ova.
Mosaicism: the presence of two or more populations of cells with different genotypes in one individual, who has developed from a single fertilized egg.
Why I’m scratching my head
I’m scratching my SCA3 head because of this article from 2017-06-19, titled “CRISPR/Cas9-mediated gene editing ameliorates neurotoxicity in mouse model of Huntington’s disease.”
The abstract concludes with this:
Our studies suggest that non–allele-specific CRISPR/Cas9-mediated gene editing could be used to efficiently and permanently eliminate polyglutamine expansion–mediated neuronal toxicity in the adult brain.
A quick reading of this sounds like they are talking about eliminating a genetic defect from a mature brain, but that is outright impossible. Since they are talking about eliminating toxicity only, it gets a little more nebulous:
In the SCA realm, I think there are various assumptions others make, which I am not ready to make, though I also make assumptions of my own. This is what I discern:
Most cells don’t matter?
Others assume that with HD, SCA, and perhaps other genetic brain diseases, only a small fraction of all cells needs to be fixed to fix the disease. In other words, most cells don’t matter.
Intuitively, I’m not ready to make that assumption. My assumption is that most cells do (or will) matter. A typical cerebellum comprises 69 billion nerve cells and 19 billion non-nerve cells. My assumption is that if you fix, say, a hundred million of the 88 billion cells in the cerebellum (114 per 100,000), very little net positive change could possibly take place. But that’s just an assumption on my part.
Protein creation activity is small?
Perhaps there’s a deeper assumption that the toxic polyQ protein creation in the cerebellum is at a rate that can be matched by the cell fix rate, et voilà. Thinking of the cerebellum as a machine that churns out toxic protein at a slow rate, perhaps that machine’s churn rate can be matched by CRISPR or other gene-silencing technique.
My assumptions are that such a small fix rate will leave billions of ticking time bombs that are susceptible to going off later. A lot of cells might not be participating in the toxic churn now, but then they will later.
This goes along with my guess that the best that gene silencing has to offer is disease amelioration but not eradication in an individual. And by how small a percentage? How temporarily? How long until the time bombs left behind go off?
Side note: the HD article above does use ameliorates in the title (and nowhere in the body—that is telling!), which is a word I have used in a past article.
You say potahto?
Here’s another analogy. Let’s say you are watching a potato rot over a few weeks on a kitchen counter. If all you can see is the outside, you might assume the inside is fine. Contrary to that viewpoint, I would assume that if the outside is rotten, the inside is probably also rotten.
That is how I think of cerebellar degeneration. If there are cells near the brain’s surface that aren’t doing so well, I don’t assume the inside is fine because I can’t see it (I think this is the scientific community’s stance). I simply don’t know who is more correct. I don’t know if anyone knows. How many cerebellums of people with SCA are medically dissected after death? Perhaps zero; again, I don’t know.
Inject first, ask questions later?
Transgenic animals (knockout mice in particular) are used for legitimate genetic disease tests. Transgenic means the animals are born with a (human) disease. It’s impossible to give a mature organism a genetic disease with an injection; to modify its genome, it must grow from conception and be born with the disease (i.e., using some form of genetic engineering).
Similarly, intuitively, it is impossible to rid a mature organism of a genetic brain disease via a CRISPR brain injection. Yes, of course, doing so will hit a few neurons here and there and repair them, but the minimality of the fraction of neurons is staggering, as exemplified above.
Just test it?
There seems to be a trend among some scientists: just test it and see what happens. If you get the results you’re looking for, then the nay-sayers were wrong. It doesn’t matter if no one understands why; results trump having an understanding, even if there’s no viable explanation.
My thoughts: this is very dangerous territory. This might be where we are in 2017, but it’s very unsettling. Certainly, this is the impetus behind many thousands of drug repurposing investigations, including trigriluzole.
It remains my understanding that polyQ genetic defects and their diseases, such as HD and SCA, aren’t eradicable from an individual if born (conceived) with one. I came to that realization a while ago. Still, there might be some future disease-modifying treatments that reduce symptoms by attacking the disease rather than the symptoms, but with no expectation of eradicating the disease from an individual, just ameliorating it—fractionally and temporarily. That’s the middle ground that’s opening up. And since life itself is temporary, temporary might be good enough.
Addendum—CRISPR is good for cancer, not SCA
If you’re born with SCA(3), there is no way to rid your brain of it. On the other hand, there’s little question that a CRISPR “brain bath” could fix some fraction of the troublesome brain cells. In that case, the fewer and younger the cells, the better. That is, only the fertilization stage offers a fix opportunity, but the newborn stage is the next best time.
The assumption I’m apparently supposed to make is that eventually delivery and efficacy (or is it effectiveness?) of drugs will improve to the point that SCA3 in an adult is stretched out to beyond one’s lifetime: the disease is not cured or eradicated from an individual, but perhaps the individual could avoid having to deal with the disease, except when having children, since germ cells can still pass on the defect.
That is so many decades past my lifetime, if ever, it barely registers with me. I’m not trying to say CRISPR isn’t a reason to get excited, just not for adult-onset brain diseases such as SCA. What is CRISPR potentially good for, other than germ cell or zygote manipulation?
CRISPR’s first trial applications (in 2017) are towards some cancers. What is cancer, in basic terms? Cancer is abnormal cell growth, often involving tumors, though leukemia is different (i.e., abnormally-formed white blood cells). Since cancer involves cells and the mutations of genes contained therein, it’s considered genetic, though not in the same sense as SCA, which involves a single hereditary defect in a single DNA location, replicated trillions of times; tumorous cancer is more local, random, and can be triggered externally—for example, by radiation, by cigarette smoke, etc.
Being explored now is using CRISPR to destroy cancerous cells as an alternative to chemotherapy or surgery, where (relatively few) damaged cells are essentially piggybacking on (relatively many) healthy cells, which is very different from the inherently impossible idea of modifying billions of neurons in the SCA brain.
Reports of CRISPR being curative for (e.g.) type 1 diabetes and muscular dystrophy I think are recklessly exaggerated, like they are for SCA and HD. I am not focused on understanding diabetes and so on, as I am most focused on SCA3. My understanding of diabetes and some other diseases is that when the brain is not the attack target of the disease, there is more potential to treat the symptoms. This goes for some kidney diseases and other non-brain organ diseases. Without jumping immediately to science fiction, it’s possible to conceive of non-brain organ damage that is repaired or even replaced. The underlying genetic problem is not fixed and might perpetuate a recurring damage cycle, but at least replacement is part of a conceivable scenario, whereas replacing the brain is not.
Even with muscular dystrophy, the disease attack target is the muscles, not the brain, which opens some therapeutic possibilities involving both CRISPR and stem cells, neither of which is applicable to mature-stage SCA. I don’t mean to sound like I’m predicting panaceas for everything I don’t understand, as others are willing to do for SCA and HD. Quite simply, ameliorative therapy applied to damaged muscles simply sounds more beneficial than ameliorative therapy applied to a damaged brain.
Neuronal destruction in (most of) the brain cannot be fixed—or fully prevented post-zygote stage—and parts of the brain are not replaceable. For the foreseeable decades, I think the absolute best case for genetic brain diseases such as SCA and Huntington’s disease will be super-expensive ameliorative treatments that prolong a miserable life. As such, they will be largely undesirable, and undesirability will hinder the unspoken implication—involving some science fiction—of ameliorative treatments magically evolving into cures, despite pharmaceutical companies being on a profit-driven, 20-year patent cycle.
More fundamentally, I don’t see why pharmaceutical companies would accommodate those unwilling to use IVF+PGD (USD 20,000 or so) or worst case, unwilling to use genetic testing of any kind and bank on an option for multi-USD-1,000,000 risky (i.e., spinal cord injections) and imperfect therapy on an asymptomatic newborn, or child.