Good afternoon. My name is Cindy Morsehead. I'm a professor at the University of Toronto in Canada. Today I'm going to share some of our work that has impacted the way that we think about promoting brain repair. I'm going to focus on neural stem cells. So these really are the building blocks for the central nervous system. They reside in the adult, primarily in a region called the subappendema. And this region lines the lateral ventricles and is really the remnant of the germinal zone during development. There's also another location in the brain as seen in the sagittal section, which is the dentate gyrus of the hippocampus. And I'm going to focus on the subappendema, which is the largest pool. These subappendemal neural stem cells, despite being most prominent in the subappendema, these are still a very rare population of cells that can exist in two forms, either quiescent or activated. And by activated, I refer to proliferative. These neural stem cells divide and give rise to progenitor cells that will migrate along a very well-defined pathway called the rostral migratory stream. And these progenitors then go to the olfactory bulb, where they turn into olfactory bulb interneurons that integrate into the circuitry. 60% of the cells that are in this lineage will undergo cell death, and 25% of them will migrate. So we can see we have a lot of different features going on in the normal adult brain. Under baseline conditions, we have cell death, cell migration, and cell differentiation. As well, the cells that are born in the adult brain are also able to give rise to oligodendrocytes, primarily ones that will go and reside in the corpus callosum. We can assay neural stem cells using a simple and robust assay called the neurosphere assay. So this in vitro assay was originally described by Reynolds and Weiss in 1992. Much to their surprise, they were able to isolate a population of stem cells, which when isolated through a dissection of the subappendema, single cells would give rise to clonally derived colonies of cells that elicited the two cardinal stem cell properties. One, self-renewal. You could take a single neurosphere, dissociate it, and get new neurospheres. And the second property is multipotentiality. You could take one of these spheres, stick it down on a sticky substrate, and the cells would give rise to neurons, astrocytes, and oligodendrocytes. Stem cells exist throughout the lifetime of the animal. They exist in something called the stem cell niche. And here I'm demonstrating what we know about the subappendemal stem cell niche. And this region, neurogenic niche, changes a lot through neonatal development and through aging. Changes occur in terms of the cells that comprise the stem cell niche, the distribution of the cells, the numbers of the different types of cells, including microglia, astrocytes. The blood supply to this region changes over time and becomes more tortuous with age. The stem cells themselves will change their morphology and their activity. So this niche, I'm going to tell you as we go through the story today, really has an impact on the activation of stem cells and the process by which stem cells can contribute to neural repair. Since their original isolation, neural stem cells from the adult forebrain have been isolated and transplanted in a number of different models to promote neural repair. Isolated cells would be expanded in vitro. They would either be transplanted into the brain or spinal cord as undifferentiated cells, or they could be differentiated into the desired cell type, whatever cell they were trying to replace, and then transplanted into the brain. It turns out that this is a promising strategy. It's being used now and continues to be studied. But there are a lot of things that need to be considered that are still being worked out in the process of designing stem cell transplantation therapies. What types of cells do we use for the neural repair? When do we transplant the cells? Where do we transplant the cells? Right into the lesion? Or do we transplant them into an environment that is perhaps less hostile? What are our measures of success? What conjunctive therapies might we also need in order to promote their survival and differentiation and integration into the system? For instance, bioscaffolds and biomaterials that might promote their survival. All of these questions are still being asked. There are over 200 stem cell-based therapies that are targeting the CNS and different disease models. What I'd really like to focus on today, however, is the use of biologics and novel therapeutics to stimulate those resident neural precursor cells that we know where they are. We know what they do under baseline conditions. Can we activate them and get them to migrate to sites of injury, expand in number because they are rare cells, and turn into the appropriate cell types to replace those that are lost due to injury or disease? We're asking a lot of the cells, but at the same time, we think it will work because this is what they normally do. We know that injury alone is able to activate neural stem cells. Stem cells, if you were to isolate them from the subappendum of a control brain shown in the blue bar, you'd get 100%. If you had an injury, and here I'm showing you the response of the stem cells after a stroke injury in an adult at seven days post-injury, you can see that there's a significant expansion. The stem cells themselves are quite responsive to injury. They're equally responsive to some drug therapies. Not only can you get them to expand in response to a stimulus, an injury stimulus, but in the presence of drugs and in combination with an injury, you can get them to expand even further. Of course, if we want to use these cells for repair, we really do need more of them because they are such a rare phenotype. What I'm going to talk to you today about is the work we've done recently with the drug metformin. We've been able to demonstrate that this drug is able to aid cells migrating to the sites of injury, expand them in number, and promote neurogenesis and oligogenesis. But why did we choose metformin? We know that metformin, the commonly used drug to treat type 2 diabetes, used by millions, well-tolerated, with a strong safety profile. It was shown by Wang et al., part of Frieda Miller's group, back in 2012, that metformin is also able to promote neurogenesis and oligogenesis from embryonic and human cells that are in the presence of metformin. This can happen both in vitro and, more recently, has been demonstrated in vivo. With that in mind, with the ability to say that metformin is able to promote oligogenesis and neurogenesis in the brain, we asked, could we use metformin to repair a brain injury? Neural stem cells, through aging, actually deplete in number. This is a graph showing you that during gestation and just around the time of birth, this is when we have the most neural stem cells that we can isolate from the brain. Over time, and with aging, there's a decrease, so that by the time you're in the aged brain, in the old age brain, there are very few stem cells left that would form these neurospheres in a dish. We don't know the exact mechanism for the decrease through aging. We don't know whether it's because of increased quiescence of the cells, they're just less active, have they become senescent, or are they actually undergoing cell death. But the bottom line of this graph is just demonstrating that there are fewer stem cells through aging that we can activate. As a result, we wanted to ask if we could activate the stem cells in a neonatal model of injury. The idea was that we would focus our attention in a time when we had the most stem cells available, as a proof of principle, and be able to activate them. The first thing that we noticed was that, in fact, the neonatal stem cells were responsive to metformin treatment. If we were to place neural stem cells from the neonatal brain in a dish in the presence of metformin, we could expand them in number, as shown in the graph. We saw a similar phenotype if we were to inject the metformin directly into the pups and then isolate the stem cells. Metformin alone was able to increase the size of the neural stem cell pool. With this in hand, and knowing we have a lot of them around, we actually did our studies in a model of neonatal stroke. Our neonatal stroke was a hypoxic ischemic insult on day 8. On day 8, we gave the animals a stroke. We started giving them metformin treatment on day 9 and for one week. We tested the animals one week after the completion of the metformin treatment on day 23 on a number of sensory motor tasks. I'm showing you here the cylinder task. As you can see from this figure on the right, you can see there's quite significant damage to the brain. Animals that receive a hypoxic ischemic insult have deficits in a number of motor tasks, as well as cognitive tasks, which I will come to later on in the talk. What we were able to show was quite exciting. We were able to demonstrate in this cylinder task that animals that received one week of metformin treatment could recover to control levels of their behavior in this task. Let me describe the task for a moment. Animals that are placed inside a cylinder, when you put the mouse inside this clear cylinder, it will stand up on its hind paws and explore the inside of the cylinder using both of its paws equally. If there's an injury, it will use the unimpaired paw more frequently. This is what we see in the graph. If you put control animals into the cylinder, they show no preference for either paw. The same thing with metformin treatment. Animals that receive stroke prefer to use their uninjured paw significantly more. Animals that received the ischemic insult, as well as metformin, either through the mother's milk or through injection directly into the pups, were back down to control levels, and they were not significantly different from controls. We then asked what was going on at the cellular level. To do this, we used a transgenic mouse that allowed us to pre-label the neural stem cells in the mice prior to the injury. We used a Nestin Cree ERT2 mouse, so a conditional reporter that had been crossed to a TD tomato. In this animal, once we give them tamoxifen, the neural precursor cells in the subapendema are labeled and expressing TD tomato. In these animals, we would then give the insult in the presence or absence of our metformin, and we asked what do these cells do in vivo. It turns out that, as I've shown here, in control animals, don't receive any injury, but still receive the tamoxifen. In the parenchyma, I'm showing you here the motor cortex, no cells are really being contributed from the stem cell pool at this time at day 23. At the time when we saw the functional improvement. The odd cell, we saw a few cells in injured animals. This animal received an injury, but did not receive metformin. There were a few cells in the parenchyma, but out in the parenchyma of animals that received both the injury and the metformin, we saw an abundance of cells all throughout the ipsilateral hemisphere, so the injured hemisphere. These cells were derived from the subventricular zone, from the neural precursor pool. When we did a phenotype analysis, it turns out that these cells are turning into oligodendrocytes and neurons in the parenchyma. Neural stem cells from the subventricular zone are migrating to the parenchyma and giving rise to new neurons and new oligodendrocytes. This was concomitant with the functional improvement. Not only could we show this functional improvement, which we were very excited about in the presence of just one week of metformin treatment, but more recently we've actually demonstrated that you can even delay that metformin treatment for a week. So delayed metformin to start a week after the injury is also sufficient to improve functional outcomes, which of course has very important clinical implications to be able to delay that metformin treatment and still see the functional recovery. In the next series of experiments, we were wondering if we could promote cognitive recovery. We know that this stroke injury is sufficient to cause cognitive impairments as well as motor impairments. And so although they don't show up until a lot later, until weeks after the injury, we would be able to do behavioral assessment of cognitive function in these same mice that receive stroke. We decided to sort of hit it with a hammer and give metformin for a much longer period of time. And one of the first tasks we used to assess their cognition was called the puzzle box task, which I will quickly go through here. So animals would be placed inside the puzzle box and there's an underpass to the goal box. And this goal box is where the animals like to be. So animals that you put into these white want to get directly through the underpass and into this goal box. We give them three trials and they're very good at this on the very first day. On the second day, we let them run into the goal box. And then on the second trial of that day, we fill this underpass with some stuffing or some bedding that requires them to dig it out to learn that they need, in order to get into their goal box, they have to acquire this new task of digging. On the third day, we make it a little bit more challenging and we put a plug, a paper plug in the underpass and the animals have to learn to pull out the plug in order to get into the goal box. As can be shown on the next slide, I'm going to just focus your attention on this graph here, the female graph. What you can see is just what I was saying that if we take a look at the red line, which is the naive line, you can see that animals rapidly get into the goal box. This is the time in seconds that it takes them to accomplish the task. When we make it a little bit more difficult, it takes them a little bit longer with the stuffing and then we make it even more difficult, we put the plug in on the last day. Males and females that have HI do much poorer on this task in terms of the length of time it takes them to learn to do the task. So you can see here that females and males both take a longer time after HI to learn the task of removing the plug. What was most interesting here is that females, you can see them in green, the green line, they can completely recover and behave exactly the same as naive in the presence of metformin. But the males did not recover, they still behaved the same way as injured mice, even though they received the metformin treatment. So in this task, what Rebecca was able to demonstrate was a clear difference in the recovery between females and males in this cognitive task in the presence of metformin. While they both showed deficits, only the females were able to recover. This was really interesting, but we wondered if it was specific to HI. We wondered if metformin was able to promote functional recovery in a cognitive task in a different injury model. So what Rebecca and Daniel did is they started with cranial irradiation, and they gave cranial irradiation on day 17, and then they gave metformin for a few weeks, and then they did behavioral testing again a few weeks later for when they could see the deficit. This cranial irradiation also causes, has been well established to cause functional impairments in cognition. So what I'm showing you in this slide are two tasks that we did with the mice, and we had very similar results where females but not males were able to, had their cognitive deficits reversed in the presence of metformin. This is a spontaneous alternating task where animals are placed in the middle of a Y, and they have to go to each of the arms sequentially in order to receive a score of one. Males that perform this task normally will do it appropriately 50% of the time. If you give them irradiation, they do poorly on this task. If you give them metformin to females, they will actually recover in this task. It turns out that males do not show any deficit in this task, so we weren't able to test males on the effect of metformin. So we did a different task for males, and this was object, novel object placement. And so what we did here is we put two objects with a male mouse into a box, and the animals have time to explore, and the animals will usually explore both objects equally. After a delay of time, you move one of the objects to a new place, and an animal should more time with the object in the new location. So control mice, males, perform poorly after irradiation and do not spend enough time relative to controls looking at the new object, pardon me, the new location of the object. If you give them metformin, they do not recover. So similar to what we saw with HI, females will recover in a cognitive task, and males will not in the presence of metformin. This told us that, quite prominently, that there was sex-dependent repair in the presence of metformin. A lot of work was done by Rebecca and Daniel to really show that there were age and sex-dependent neural precursor cell activation, and that this activation was dependent on the niche. So just to briefly highlight what I'm showing you in this graph is that at the time of birth, both males and females are responsive to metformin treatment. And what I mean by responsive is that their neural precursor pools will expand in number after birth. Males and females in the juvenile age are not responsive to metformin. In adulthood, females acquire the responsiveness to metformin again, but males never do. So males are only responsive to metformin treatment in the early times postnatally, and they are not responsive to metformin throughout the rest of life, the neural precursor pools, whereas females lose their responsiveness and then gain it back in the adult. So the observation that females but not males were responsive to metformin in the adult made us ask whether hormones might be a likely candidate to underlie these effects. So we decided to perform a loss-of-function experiment. So here again, in controls, I'm showing you that if you give metformin to a mouse, in an adult female mouse, you can see an expansion of the number of neurospheres that you generate in a dish. In an over-ectomized mouse, if you give metformin, you do not see this effect. So you lose the effect in the absence of estrogen. This is the same sort of thing was shown in a gain-of-function experiment where we have control animals at postnatal day 17 in the prepubescent phase that are not responsive, but if we give estrogen to these mice, we can make them responsive to the metformin. And again, a gain-of-function and a loss-of-function experiment indicated that estrogen was playing a role in the activation of the stem cells. It turns out that through a number of other experiments, estrogen itself is not acting directly on the stem cells, but instead it was acting through the niche. And we knew this because if we took stem cells out of their niche and put metformin on them, they were non-responsive no matter what age the stem cells came from. So metformin was only able to activate and expand the size of the neural stem cell pool if the cells were plated in the presence of their niche. And importantly, they had to be plated in the presence of the female niche. Turns out that some cell within the female niche, in response to estrogen, is making the neural stem cells responsive to metformin. So the female niche is able to induce metformin responsiveness and expansion of the neural stem cell pool, and this is not true of the male niche. Whenever I present this data, one of the questions that immediately comes up is, is there any hope for males given this need for the estrogen in activating the cells? I always like to show this slide because it says, yes, there's hope. Because remembering that metformin also increases neurogenesis and oligogenesis. And if you look at these two differentiation outcomes from neural precursor cells derived from neural stem cells, both males and females are able to give rise to increased numbers of oligodendrocytes and neurons in the presence of metformin. So both males and females can give rise to more differentiated progeny in the presence of metformin. It turns out that based on this number of other studies that we've done, it turns out metformin not only promotes oligodendrogenesis, promotes neurogenesis, and can expand the size of the neural precursor pool, which I've shown to you during this talk. More recent work that we've done, Daniel Durkacz did it, it was more recently reported. It also modulates the neuroinflammatory response. So identifying which exact mechanism is promoting the functional recovery is really a question that we're very interested in pursuing. But it turns out that metformin has all of these effects on these different cell types in the brain. Based on these studies that I've shown you, we've now reported a pilot clinical trial result that was reported in Nature Medicine that really implicated metformin as being safe, feasible for children that had acquired brain injury. And based on that, there are a number of clinical trials, full-on clinical trials that are now underway for children and adolescents with both acquired brain injury, multiple sclerosis, and cerebral palsy. These are being led by clinicians and scientists at the Hospital for Sick Children in Toronto. So much like getting the question, is there any hope for males? One of the questions I get is, what about in the aging brain? So I've already shown you that the numbers of stem cells is depleted through time in the aging brain. I've shown you that the niche is responsible for a lot of the changes that occur in terms of the activation of the stem cell pool and even the progeny, things that I didn't get time to go into today. Certainly we know there's fewer neurons born in the aged brain, and in general, the degree of plasticity is reduced in aging. And one of the examples that I can give of this, just to end the talk and end it in a bit of hope, is that neural stem cells, I've shown you this data, neural stem cells expand in response to injury, and I showed you this earlier. So this is in a young brain, you can see a two to three-fold increase in the size of the neural stem cell pool just after injury alone. In an old age brain, this does not happen. So in old age brain, you do not get the expansion in the neural stem cell pool, they're no longer responsive to injury alone. But what is promising is that the stem cells are responsive to injury plus drugs. Showing you here, cyclosporin A, we've also demonstrated it with other growth factors that the adult brain is still able to be activated in terms of the neural stem cell pool in response to injury plus drugs. So what I hope I've done today is convinced you that stimulating resident neural precursor cells is a good idea for being able to promote self-repair. And with that, I would like to thank all of the people, I've mentioned some of the people that are currently in the lab, in bold are the names of those that contributed to the story that I told today, and of course my collaborators and funding. Thank you very much.

brain repair

neural stem cells

central nervous system

subventricular zone

neurosphere assay

metformin