The road to treatment 23
Interview with Tom Schwarz of Harvard
We were in touch with Thomas Schwarz, who is Professor of Neurobiology at Harvard Medical School and Professor of Neurology at the FM Kirby Neurobiology Center, Boston Children's Hospital. Thomas Schwarz and his laboratory are involved in ADOA research and are developing not one, but two potential treatments!
Can you give us some insight into how you research ADOA?
I will answer by explaining what the team at Boston Children's Hospital has done regarding ADOA and what we are trying to do by developing therapeutic directions. Our work on ADOA grew out of conversations with a family who had a very personal interest in this disease and was eager to promote and even sponsor research into ADOA. Here at Boston Children's Hospital, there is a very strong basic neuroscience community, including several scientists studying retinal ganglion cells, the cell type most prone to degeneration in ADOA. And there's also my laboratory, which has long been interested in the cell biology of neural mitochondria. This includes the mitochondria in the retinal ganglion cells, which die due to ADOA. We therefore put together a collaborative consortium that included my laboratory and that of Drs. Michael Do, Larry Benowitz and Chinfei Chen. We also have an excellent ongoing partnership with Dr. Marni Falk and her colleagues at the Children's Hospital of Philadelphia (CHoP). I have to say that most of the work was done by a fantastic postdoc in my lab, Dr. Chen Ding, PhD, with help and guidance from our colleagues and core facilities here at Boston Children's.
Our plan was to create a new mouse model of ADOA that carried the R290Q mutation found in several families. Many ADOA mutations cause what geneticists call “haploinsufficiency,” which means that the mutation simply prevents one of the two copies of the OPA1 gene from making a functional protein, and the cells have to make do with half the amount OPA1 which they should normally have. Other ADOA mutations – especially those that fall in the part of OPA1 with enzymatic activity – are worse than just making a protein that doesn't function. Instead, they make a protein that actively gets in the way and prevents the good copy of OPA1 from also functioning. These are called “dominant negative” mutations, and patients with these mutations are sometimes called ADOA+ patients. The R290Q mutation sits just at the edge of this enzymatic part, so we didn't know if it would be a simple loss-of-function mutation or a dominant negative, but we hoped it would give a strong enough phenotype in the mouse that we could use it to study the degeneration of the retinal ganglion cells.
How did you manage to change the mouse's genes, what technology did you use?
We used CRISPR technology to introduce this R290Q mutation – which turned out to be surprisingly difficult! We think this is because the CRISPR method worked too well and mutated both copies of the OPA1 gene – and cells with both copies mutated will not produce a viable mouse. But eventually we managed to create the R290Q mutant mouse line and have been studying that mouse for the past 2 years. We also obtained from CHoP a stem cell line that comes from a patient and has the same mutation – and a control group with that mutation corrected by CRISPR. This means that we can perform experiments in parallel in the mouse model and in human neurons that we can generate from those stem cells. Work with the R290Q mouse is going well. The normal functions of the OPA1 gene include allowing the inner membranes of mitochondria to fuse when two mitochondria meet. In healthy cells, including healthy neurons, many mitochondria are fused together into a network. However, in neurons from the R290Q mouse, the mitochondria are fragmented, as expected with an OPA1 mutation. The retinas of these mice are surprisingly normal and functional when the mice are young, but as they age we see many defects that parallel those seen in ADOA patients. Remember that the loss of retinal ganglion cells in ADOA patients can take decades, but mice typically live for about 2 years. So we're lucky that in our mouse model we can see that some of those retinal ganglion cells die after only 9 – 12 months.
In healthy mice, there is a normal insulating layer around the axons (the long part of the optic nerve that transmits information to the brain) of the retinal ganglion cells. The R290Q mouse also shows demyelination of nerve fibers (axons) in the optic nerve – in other words, a loss of the normal insulating layer that helps with nerve conduction in the part of the retinal ganglion cells that carries signals from the eye to the brain. We also make electrophysiological recordings (electroretinograms and visual evoked potentials) that can detect the ability of the retinal ganglion cells to fire and send signals to the brain. These recordings show profound defects as the animals age. In addition, Michael Do's laboratory has established a remarkable assay in which they can cut out the retina with the optic nerve attached to it, present precise light patterns to the retina, and record the resulting electrical activity in the optic nerve – a beautifully direct way to investigate the function of populations of retinal ganglion cells. This again suggests a change in function in the R290Q mutant mouse and begins to reveal where that change occurs.
Can you please tell us a little more about how this might lead to a therapy for ADOA?
With our model systems at hand, we test two therapeutic strategies. One of them is Gene Replacement Therapy. Simply put, this is a way to put a good copy of OPA1 back into the retinal ganglion cells. On the one hand, ADOA is a good candidate for this approach because it is relatively easy to get copies of a gene into retinal ganglion cells by injection into the vitreous humor of the eye, with a safe, non-reproducing virus that expresses the gene. . This is a form of gene therapy approved for clinical use in the retina for other conditions. Furthermore, because the degeneration is slow, there is a good time window to try to disrupt the degeneration by adding the good specimen. But there's a problem – the human OPA1 protein is made in 8 different variants (by a process called alternative RNA splicing) and it's not possible to add them all back. We're figuring out if adding just one of those variants will be enough and, if so, which variant to use.
Our second approach is to inhibit a neuronal protein called SARM1. SARM1 is an enzyme that is activated in stressed or damaged neurons and causes neurons to die – it appears to have no purpose other than to serve as a cell suicide switch. Many people in academia and in pharmaceutical and biotech companies are investing time and money in developing SARM1 inhibition as a strategy to prevent neurodegeneration. If the degeneration of retinal ganglion cells in ADOA proceeds by activating SARM1, ADOA patients may benefit from these developments.
Do you expect the treatment in development to be applicable to a broader group of ADOA patients or will it be mutation specific?
Unlike some CRISPR-based strategies that target just one specific mutation, both restoring a working copy of OPA1 and preventing SARM1 activation should work equally well for most, perhaps all, patients harboring OPA1 mutations. to have. The exception to the gene replacement approach might be in the severe dominant negative forms found in ADOA+; there would be some kind of competition between the mutated copies and the normal copies and the dominant negative mutation could still cause problems. The other caveat is that these strategies, if they work, would be relatively easy to apply in the eye, but gene therapy is more difficult for the other types of neurons affected in ADOA+.
Could the two therapeutic approaches be used in combination, or would the focus be on one or the other?
We hope that at least one of these strategies will pay off. If they both work, I see no reason why the strategies, if they work, couldn't be combined if that would increase their effectiveness. There is nothing a priori incompatible about them.
Finally, I would like to emphasize how crucial it has been to have a team of expert collaborators. This is the strength of many – none of us would have had the courage to take on this challenge without our combined expertise and none of it would have happened without the initiative of the family who fueled and supported the project.
