I am a stem cell biologist who has been studying schizophrenia, a disease that is far too common, severe, and poorly treated. The estimates of heritability for schizophrenia range from 50 to 85 percent, with the latest twin studies suggesting it is closer to 80 percent. This means that the disease is highly genetic and largely predisposed by the DNA that you're born with.
I believe that we can do a better job predicting this disease before symptom onset, treating it before symptom onset, and understanding what the genetic risk factors for this disease are and how they are interacting. I am challenging us to think about what if we switched from a diagnosis-based to a genotype-first treatment approach. It doesn't matter whether a patient has schizophrenia or bipolar disorder; what matters is that they have a specific mutation in a specific gene.
When I started my postdoc, I asked the question: could we make human neurons from living schizophrenia patients in the lab? I wanted to take skin samples from patients and controls and turn them into induced pluripotent stem cells, which can make every cell type in the body, including all of the neurons and astrocytes in the brain. These cells are identical to the cells in the brain of each and every one of us. Then, we do a two-stage differentiation to turn these into neurons.
I am encouraged that the neurons from the patient's look a lot like the neurons from the controls. This is what you would expect, as patients walk and talk and breathe like the rest of us. If it was so simple that I just couldn't make neurons from patients, oh they died overnight, there would be something wrong with my model.
I think it's really encouraging that the cells look so similar. Now, we can explore what sort of things we would be looking for in a laboratory dish. A lot of our early work was to repeat stuff that we already knew from 50 years of clinical research. We know the neurons tend to be a little bit smaller in the post-mortem brains of patients versus controls. That's mostly because neurons are smaller; they have fewer dendrites, fewer connections, and fewer synapses between the neurons. This is exactly what we saw in our first study in the lab.
We saw fewer connections between the neurons, and a couple of years later, a second group on a completely independent cohort found the exact same phenotype by the exact same assay. This is the first continental replication of a stem cell phenotype, and we think it really justifies using stem cells to explore schizophrenia.
We are now using this model to explore how schizophrenia risk factors impact neurons and to find new drugs to improve those neurons and make them go back to looking like the controls. There is a growing number of loci in the genome that are linked to schizophrenia, and the question is for each SNP in the genome linked to schizophrenia, which genes are genes impacting and how does that result in disease. This will allow us to better diagnose disease from blood draws and hopefully better predict treatment response.
We are now incorporating a new technology called CRISPR editing, where we change the DNA of any given person and then ask the experiment. We can take a schizophrenia risk factor and put it in this control or this control or even this patient and ask how changing one genetic risk factor impacts neuronal function across different people. This will allow us to capture effects of genetic background and penetrance.
We are also trying to understand how to better do drug studies. Drug studies are really hard and really expensive, and when you're looking for completely new mechanisms of disease, you have to screen hundreds of thousands of drugs. We are working with some of the best pharmaceutical companies around to screen hundreds of thousands of drugs in silico in the computer. This will allow us to complete our first drug screen in our schizophrenia patient cells and find new drugs to target this disease.