Schizophrenia, in a dish

by Kristin Sheppard 

             A few weeks ago my friend Ellen posted a great blog about stem cells. She did a great job of explaining the basics of stem cells and describing some of the possible therapies and maybe even cures for various diseases that would accompany transplantation of these cells under different conditions (and if you missed it, you can read her blog post here).

Cal Poly CIRM students class of 2014

             Over the last few weeks both Ellen and I (and everyone else in our little cohort of first-year CIRM students) have been tirelessly researching different labs, and the work that these labs are doing in the field of stem cell technology, to try and find the perfect fit for our upcoming internships. During my research I stumbled upon a whole new aspect of stem cell research that until recently has been unexplored. A post-doc from the Gage lab in the Laboratory of Genetics at the Salk Institute for Biological Sciences presented it to me as “schizophrenia in a dish.” As crazy as it sounds, this lab has been able to make a model of this disease by taking a very small skin sample from patients with schizophrenia, turning those cells back into stem cells (technically they would be called induced pluripotent stem cells or iPSCs), and then differentiating the iPSCs into neurons. By culturing these cells in a plate instead of an animal they have created a cellular model of schizophrenia, in a dish.  

              So let’s take a step back for a minute, and discuss what schizophrenia actually is. Schizophrenia affects approximately 1% of the world population. It is a mental disorder characterized by the experience of psychosis, a state where the individual loses contact with reality. To be clinically diagnosed with schizophrenia, the individual needs to satisfy the following requirements:

1)    Have at least of the two following symptoms for a significant period of time during a one-month period:

・      Delusions

・      Hallucinations (either visual or auditory)

・      Disorganized speech

・      Excessively disorganized or catatonic behavior

・      Negative symptoms (which can include a reduction in speech or speech content, flat affect, increased apathy, or social withdrawal)

2)    Functioning far below their functional level prior to the onset of symptoms.

3)    Continuous signs of the disturbance for at least six months, and at least one of those months include symptoms in their full and active form.

So if schizophrenia is a mental disorder diagnosed based on a person’s behavior and perceptions, how can a bunch of cells in a dish be helpful? Although the exact cause is unknown, schizophrenia tends to run in families, suggesting that there is a genetic component. There have been several genes found to be associated with schizophrenia, but there is still no clear gene or set of genes that cause the disease. That is where the cellular model comes in. This model allows scientists to test molecular differences and differences in electrophysiology between neurons from individuals with schizophrenia and individuals without schizophrenia. Even better than that, this model circumvents complications that come with current methods of studying this disease in humans (which generally means a post-mortem analysis of an affected individual’s brain). Common complications that accompany these post-mortem studies include an unknown history of medical treatment for the disease (or lack there of), a history of alcohol or drug abuse, and variations in the cause of death. The cellular model is a living model with identical neurons from living individuals who can likely identify their treatment history. This is important because it allows for the direct comparison with neurons from unaffected individuals to find molecular and physiological differences. It also allows for the treatment of the affected neurons with different medications to compare how these drugs affect genetically identical cells to help determine what the best treatment is.

So lets get back to what these researchers actually found. The initial findings, although important, were not that exciting. They found that when they turned the skin cells into stem cells, then differentiated them into the neural progenitor cells (the precursors for neurons), and finally neurons there was absolutely no difference between cells from individuals with schizophrenia and cells from individual without schizophrenia. More specifically, there was no difference in the differentiation process or pluripotency markers between cells from schizophrenic individuals and cells from non-schizophrenic individuals.

How to make schizophrenia in a dish

After these cells were allowed to become mature neurons, they tested the synaptic connections between these neurons. The way that they did this was particularly clever; they used a rabies virus. The rabies virus is known to spread between neurons in an animal by moving through the synapses (or connections) between neurons. However, in order to be able to use this to test the connectivity of the neurons, the researchers have to be able to distinguish neurons that were infected directly from the surrounding fluid and neurons that were infected through the synapse with another neuron. In order to do this they modified the rabies virus so that it is fluorescent red when viewed with a fluorescence microscope, and has to use a specific receptor to be able to enter the neuron from the surrounding fluid. Before putting the modified rabies virus into the surrounding fluid, they modified the neurons so that only some of them had the necessary receptor for the rabies virus. The neurons with this receptor would appear fluorescent green when viewed under the same microscope. This is important because you can tell which cells were infected from the surrounding fluid (because they will be both red and green) and which ones were infected through its connection to another neuron (because those will be only red). They then compare the number of just red cells to the number of red and green cells to determine the connectivity of the neurons, a higher proportion of just red cells means that there is better connectivity between those neurons. So these researchers found that neuron cultures from individuals with schizophrenia had fewer just red cells than neuron cultures from individuals without schizophrenia. This means that neurons with schizophrenia have lower connectivity than neurons without schizophrenia. This is a physiological difference that would not have been found without the use of the cellular model.

Neurons infected with rabies virus. The image on the left is of neurons from non-schizophrenic individuals, image on the right is of neurons from a schizophrenic individual (Brennand, 2011).

They also found that neurons from individuals with schizophrenia had fewer projections out of the cell body than neurons from individual without schizophrenia. This finding was consistent with previous observations made from post-mortem studies on the brains of schizophrenic individuals. It is however important that this was confirmed, because it is another reason why neurons from schizophrenic individuals have lower connectivity. These projections are responsible for making the connections known as synapses with other neurons, so fewer projections means fewer connections.

When they measured the electrical signals being sent between these neurons they found that there were no differences between neurons from schizophrenic individuals and non-schizophrenic individuals in the characteristics of the electrical signals or the ability of the neurons to send the signals. This would indicate that these neurons are still able to communicate with each other; it’s mainly a difference in the number of connections that allow communication between neurons and which signals are being sent that distinguishes schizophrenic neurons from non-schizophrenic neurons.

After discovering the decreased connectivity, these researches tested the effects of five common antipsychotic drugs and their ability to increase the connectivity between these neurons. The drugs that they tested were Clozapine, Loxapine, Olanzapine, Risperidone, and Thioridazine. They found that only Loxapine significantly improved the neural connectivity while they were in the dishes. This would imply that Loxapine is going to be the most effective medication for those individuals, however they also admit that it is possible that they need to adjust the doses or the timing of when they administer the other medications to be able to see an effect. Although they may need to work out the dosages, this method can be an effective way for screening for new drugs that will help to treat or even cure the disease.

It's difficult to know all the side effects of a drug when they're not tested in a cellular model first 

Finally, using this cellular model, researchers found that there were 596 unique genes with a 1.3 fold change in how many times each gene was read by the cells between the schizophrenia neurons and the non-schizophrenic neurons. Only 13% of these genes had been previously identified in publications to be associated with schizophrenia. A whole 16% had been previously linked to schizophrenia through post-mortem studies. And a whopping 25% had been previously implicated in schizophrenia in one manner or another. This is very important because it implies that there isn’t just one gene responsible, or even just a couple genes responsible for causing this disease. In fact, the researchers think that as this model is used to study more individuals with schizophrenia, it will be found that there are countless different changes in the way an individual’s cells will read their DNA. That sounds terrifying right? But the good news is that they also predict that although there may be a ton of small changes, they think there will be a pattern in what causes these changes, meaning that they think there will be a handful of common pathways that are disrupted that leads to all the countless small changes.

              So let’s do a recap. This research team from the Gage lab at the Salk Institute for Biological took skin cells from individuals with schizophrenia and used stem cell technology to turn them into neurons. They then compared these neurons to neurons that had been created in the same way, but from skin cells of individuals without schizophrenia. They found that neurons from schizophrenic individuals have a lower connectivity and fewer projections from the cell body, which would contribute to the lower connectivity. They also found that there are an amazingly large number of molecular differences in how neurons from schizophrenic individuals read their DNA, but they expect to find a common pathway that is causing all these small changes as more individuals are studied in this manner. And finally they found that this model can be used as a way to screen new medications for their effectiveness in treating the physical aspects of this disease.

              Now all of this is really cool and all, but why should anyone except for scientists and those who are affected in one way or another by schizophrenia care?  If this cellular model can work for such a complex disease as schizophrenia, it should also be able to work to study, and maybe even cure, equally complex diseases that are more common such as autism, multiple sclerosis, or even Parkinson’s disease. In this unique way stem cells now present the potential to help understand, and maybe even cure, diseases with currently unknown causes.

References

Brennand, K.J., A. Simone, J. Jou, C. Gelboin-Burkhart, N. Tran, S. Sangar, Y. Li, Y. Mu, G. Chen, D. Yu, S. McCarthy, J. Sebat, and F.H. Gage. 2011. Modelling schizophrenia using human induced pluripotent stem cells. Nature 473: 221-225.

Saha, S., D. Chant, J. Welham, and J. McGrath. 2005. A systematic review of the prevalence of schizophrenia. PLoS Med 2:e141

Andreasen, N.C., W. T. Carpenter Jr. 1993. Diagnosis and classification of schizophrenia. Schizophrenia Bulletin 19: 199-214.