This is a tuberculosis ward, and at the time this picture was taken in the late 1800s, one in seven of all people died from tuberculosis. We had no idea what was causing this disease. The hypothesis was actually it was your constitution that made you susceptible. And it was a highly romanticized disease. It was also called consumption, and it was the disorder of poets and artists and intellectuals. And some people actually thought it gave you heightened sensitivity and conferred creative genius.

By the 1950s, we instead knew that tuberculosis was caused by a highly contagious bacterial infection, which is slightly less romantic, but that had the upside of us being able to maybe develop drugs to treat it. So doctors had discovered a new drug, iproniazid, that they were optimistic might cure tuberculosis, and they gave it to patients, and patients were elated. They were more social, more energetic. One medical report actually says they were "dancing in the halls." And unfortunately, this was not necessarily because they were getting better. A lot of them were still dying. Another medical report describes them as being "inappropriately happy." And that is how the first antidepressant was discovered.

So accidental discovery is not uncommon in science, but it requires more than just a happy accident. You have to be able to recognize it for discovery to occur.

As a neuroscientist, I'm going to talk to you a little bit about my firsthand experience with whatever you want to call the opposite of dumb luck -- let's call it smart luck. But first, a bit more background.

Thankfully, since the 1950s, we've developed some other drugs and we can actually now cure tuberculosis. And at least in the United States, though not necessarily in other countries, we have closed our sanitoriums and probably most of you are not too worried about TB. But a lot of what was true in the early 1900s about infectious disease, we can say now about psychiatric disorders.

We are in the middle of an epidemic of mood disorders like depression and post-traumatic stress disorder, or PTSD. One in four of all adults in the United States suffers from mental illness, which means that if you haven't experienced it personally or someone in your family hasn't, it's still very likely that someone you know has, though they may not talk about it. Depression has actually now surpassed HIV/AIDS, malaria, diabetes and war as the leading cause of disability worldwide. And also, like tuberculosis in the 1950s, we don't know what causes it. Once it's developed, it's chronic, lasts a lifetime, and there are no known cures.

The second antidepressant we discovered, also by accident, in the 1950s, from an antihistamine that was making people manic, imipramine. And in both the case of the tuberculosis ward and the antihistamine, someone had to be able to recognize that a drug that was designed to do one thing -- treat tuberculosis or suppress allergies -- could be used to do something very different -- treat depression. And this sort of repurposing is actually quite challenging. When doctors first saw this mood-enhancing effect of iproniazid, they didn't really recognize what they saw. They were so used to thinking about it from the framework of being a tuberculosis drug that they actually just listed it as a side effect, an adverse side effect.

As you can see here, a lot of these patients in 1954 are experiencing severe euphoria. And they were worried that this might somehow interfere with their recovering from tuberculosis. So they recommended that iproniazid only be used in cases of extreme TB and in patients that were highly emotionally stable, which is of course the exact opposite of how we use it as an antidepressant. They were so used to looking at it from the perspective of this one disease, they could not see the larger implications for another disease.

And to be fair, it's not entirely their fault. Functional fixedness is a bias that affects all of us. It's a tendency to only be able to think of an object in terms of its traditional use or function. And mental set is another thing. Right? That's sort of this preconceived framework with which we approach problems. And that actually makes repurposing pretty hard for all of us, which is, I guess, why they gave a TV show to the guy who was, like, really great at repurposing.

(Laughter)

So the effects in both the case of iproniazid and imipramine, they were so strong -- there was mania, or people dancing in the halls. It's actually not that surprising they were caught. But it does make you wonder what else we've missed. So iproniazid and imipramine, they're more than just a case study in repurposing. They have two other things in common that are really important. One, they have terrible side effects. That includes liver toxicity, weight gain of over 50 pounds, suicidality. And two, they both increase levels of serotonin, which is a chemical signal in the brain, or a neurotransmitter. And those two things together, right, one or the two, may not have been that important, but the two together meant that we had to develop safer drugs, and that serotonin seemed like a pretty good place to start.

So we developed drugs to more specifically focus on serotonin, the selective serotonin reuptake inhibitors, so the SSRIs, the most famous of which is Prozac. And that was 30 years ago, and since then we have mostly just worked on optimizing those drugs. And the SSRIs, they are better than the drugs that came before them, but they still have a lot of side effects, including weight gain, insomnia, suicidality -- and they take a really long time to work, something like four to six weeks in a lot of patients. And that's in the patients where they do work. There are a lot of patients where these drugs don't work.

And that means now, in 2016, we still have no cures for any mood disorders, just drugs that suppress symptoms, which is kind of the difference between taking a painkiller for an infection versus an antibiotic. A painkiller will make you feel better, but is not going to do anything to treat that underlying disease. And it was this flexibility in our thinking that let us recognize that iproniazid and imipramine could be repurposed in this way, which led us to the serotonin hypothesis, which we then, ironically, fixated on. This is brain signaling, serotonin, from an SSRI commercial. In case you're not clear, this is a dramatization. And in science, we try and remove our bias, right, by running double-blinded experiments or being statistically agnostic as to what our results will be. But bias creeps in more insidiously in what we choose to study and how we choose to study it.

So we've focused on serotonin now for the past 30 years, often to the exclusion of other things. We still have no cures, and what if serotonin isn't all there is to depression? What if it's not even the key part of it? That means no matter how much time or money or effort we put into it, it will never lead to a cure.

In the past few years, doctors have discovered probably what is the first truly new antidepressant since the SSRIs, Calypsol, and this drug works very quickly, within a few hours or a day, and it doesn't work on serotonin. It works on glutamate, which is another neurotransmitter. And it's also repurposed. It was traditionally used as anesthesia in surgery. But unlike those other drugs, which were recognized pretty quickly, it took us 20 years to realize that Calypsol was an antidepressant, despite the fact that it's actually a better antidepressant, probably, than those other drugs. It's actually probably because of the fact that it's a better antidepressant that it was harder for us to recognize. There was no mania to signal its effects.

So in 2013, up at Columbia University, I was working with my colleague, Dr. Christine Ann Denny, and we were studying Calypsol as an antidepressant in mice. And Calypsol has, like, a really short half-life, which means it's out of your body within a few hours. And we were just piloting. So we would give an injection to mice, and then we'd wait a week, and then we'd run another experiment to save money.

And one of the experiments I was running, we would stress the mice, and we used that as a model of depression. And at first it kind of just looked like it didn't really work at all. So we could have stopped there. But I have run this model of depression for years, and the data just looked kind of weird. It didn't really look right to me. So I went back, and we reanalyzed it based on whether or not they had gotten that one injection of Calypsol a week beforehand. And it looked kind of like this. So if you look at the far left, if you put a mouse in a new space, this is the box, it's very exciting, a mouse will walk around and explore, and you can see that pink line is actually the measure of them walking. And we also give it another mouse in a pencil cup that it can decide to interact with. This is also a dramatization, in case that's not clear. And a normal mouse will explore. It will be social. Check out what's going on. If you stress a mouse in this depression model, which is the middle box, they aren't social, they don't explore. They mostly just kind of hide in that back corner, behind a cup. Yet the mice that had gotten that one injection of Calypsol, here on your right, they were exploring, they were social. They looked like they had never been stressed at all, which is impossible.

So we could have just stopped there, but Christine had also used Calypsol before as anesthesia, and a few years ago she had seen that it seemed to have some weird effects on cells and some other behavior that also seemed to last long after the drug, maybe a few weeks. So we were like, OK, maybe this is not completely impossible, but we were really skeptical.

So we did what you do in science when you're not sure, and we ran it again. And I remember being in the animal room, moving mice from box to box to test them, and Christine was actually sitting on the floor with the computer in her lap so the mice couldn't see her, and she was analyzing the data in real time. And I remember us yelling, which you're not supposed to do in an animal room where you're testing, because it had worked. It seemed like these mice were protected against stress, or they were inappropriately happy, however you want to call it. And we were really excited.

And then we were really skeptical, because it was too good to be true. So we ran it again. And then we ran it again in a PTSD model, and we ran it again in a physiological model, where all we did was give stress hormones. And we had our undergrads run it. And then we had our collaborators halfway across the world in France run it. And every time someone ran it, they confirmed the same thing. It seemed like this one injection of Calypsol was somehow protecting against stress for weeks.

And we only published this a year ago, but since then other labs have independently confirmed this effect. So we don't know what causes depression, but we do know that stress is the initial trigger in 80 percent of cases, and depression and PTSD are different diseases, but this is something they share in common. Right? It is traumatic stress like active combat or natural disasters or community violence or sexual assault that causes post-traumatic stress disorder, and not everyone that is exposed to stress develops a mood disorder. And this ability to experience stress and be resilient and bounce back and not develop depression or PTSD is known as stress resilience, and it varies between people. And we have always thought of it as just sort of this passive property. It's the absence of susceptibility factors and risk factors for these disorders. But what if it were active? Maybe we could enhance it, sort of akin to putting on armor.

We had accidentally discovered the first resilience-enhancing drug. And like I said, we only gave a tiny amount of the drug, and it lasted for weeks, and that's not like anything you see with antidepressants.

But it is actually kind of similar to what you see in immune vaccines. So in immune vaccines, you'll get your shots, and then weeks, months, years later, when you're actually exposed to bacteria, it's not the vaccine in your body that protects you. It's your own immune system that's developed resistance and resilience to this bacteria that fights it off, and you actually never get the infection, which is very different from, say, our treatments. Right? In that case, you get the infection, you're exposed to the bacteria, you're sick, and then you take, say, an antibiotic which cures it, and those drugs are actually working to kill the bacteria. Or similar to as I said before, with this palliative, you'll take something that will suppress the symptoms, but it won't treat the underlying infection, and you'll only feel better during the time in which you're taking it, which is why you have to keep taking it. And in depression and PTSD -- here we have your stress exposure -- we only have palliative care. Antidepressants only suppress symptoms, and that is why you basically have to keep taking them for the life of the disease, which is often the length of your own life.

So we're calling our resilience-enhancing drugs "paravaccines," which means vaccine-like, because it seems like they might have the potential to protect against stress and prevent mice from developing depression and post-traumatic stress disorder. Also, not all antidepressants are also paravaccines. We tried Prozac as well, and that had no effect.

So if this were to translate into humans, we might be able to protect people who are predictably at risk against stress-induced disorders like depression and PTSD. So that's first responders and firefighters, refugees, prisoners and prison guards, soldiers, you name it.

And to give you a sense of the scale of these diseases, in 2010, the global burden of disease was estimated at 2.5 trillion dollars, and since they are chronic, that cost is compounding and is therefore expected to rise up to six trillion dollars in just the next 15 years.

As I mentioned before, repurposing can be challenging because of our prior biases. Calypsol has another name, ketamine, which also goes by another name, Special K, which is a club drug and drug of abuse. It's still used across the world as an anesthetic. It's used in children. We use it on the battlefield. It's actually the drug of choice in a lot of developing nations, because it doesn't affect breathing. It is on the World Health Organization list of most essential medicines.

If we had discovered ketamine as a paravaccine first, it'd be pretty easy for us to develop it, but as is, we have to compete with our functional fixedness and mental set that kind of interfere. Fortunately, it's not the only compound we have discovered that has these prophylactic, paravaccine qualities, but all of the other drugs we've discovered, or compounds if you will, they're totally new, they have to go through the entire FDA approval process -- if they make it before they can ever be used in humans. And that will be years. So if we wanted something sooner, ketamine is already FDA-approved. It's generic, it's available. We could develop it for a fraction of the price and a fraction of the time.

But actually, beyond functional fixedness and mental set, there's a real other challenge to repurposing drugs, which is policy. There are no incentives in place once a drug is generic and off patent and no longer exclusive to encourage pharma companies to develop them, because they don't make money. And that's not true for just ketamine. That is true for all drugs. Regardless, the idea itself is completely novel in psychiatry, to use drugs to prevent mental illness as opposed to just treat it.

It is possible that 20, 50, 100 years from now, we will look back now at depression and PTSD the way we look back at tuberculosis sanitoriums as a thing of the past. This could be the beginning of the end of the mental health epidemic.

But as a great scientist once said, "Only a fool is sure of anything. A wise man keeps on guessing."

Thank you, guys.