Dr. Karen Parker: The Causes & Treatments for Autism

Welcome to the Huberman Lab Podcast where we discuss science and science-based tools for everyday life.

I'm Andrew Huberman, and I'm a professor of neurobiology and ophthalmology at Stanford School of Medicine. My guest today is Dr. Karen Parker. Dr. Karen Parker directs the Social Neuroscience's research program at the Stanford University School of Medicine. The goal of her laboratory's research is to understand the biological basis of social functioning at every stage of the lifespan. So this includes the bonds that form between infant and parent or parents,

as well as the bonds that occur between children as they grow up, which of course form the template for social functioning when we become adults. Dr. Parker's research is heavily focused on autism and indeed on all forms of autism spectrum disorders. Today we discuss autism, we talk about the prominent theories and current understanding of the biological basis for autism, as well as what still remains mysterious and unresolved about the causes of autism.

You may have heard that the incidence, or perhaps just the diagnosis of autism, has dramatically increased in the last 10 to 15 years. And today we discuss why it is, in fact, that the incidence, not just the diagnosis, but the incidence of autism, has so dramatically increased. And perhaps most excitingly, Dr. Parker shares with us brand new research findings from her laboratory that point to a new understanding of what causes autism as well as a novel treatment for autism.

Before we begin, I'd like to emphasize that this podcast is separate from my teaching and research roles at Stanford. It is, however, part of my desire and effort to bring zero cost to consumer information about science and science-related tools to the general public. In keeping with that theme, I'd like to thank the sponsors of today's podcast. Our first sponsor is Element. Element is an electrolyte drink with everything you need and nothing you don't. That means plenty of salt, magnesium and potassium, the so-called electrolyte, and no sugar.

Salt, magnesium, and potassium are critical to the function of all the cells in your body, in particular, to the function of your nerve cells, also called neurons. In fact, in order for your neurons to function properly, all three electrolytes need to be present in the proper ratios. And we now know that even slight reductions in electrolyte concentrations or dehydration of the body can lead to deficits in cognitive and physical performance.

Element contains a science-backed electrolyte ratio of 1,000 milligrams of sodium, 200 milligrams of potassium and 60 milligrams of magnesium. I typically drink element first thing in the morning when I wake up in order to hydrate my body and make sure I have enough electrolytes.

And while I do any kind of physical training and after physical training as well, especially if I've been sweating a lot, if you'd like to try element, you can go to drink element. That's LMT dot com slash Huberman to claim a free element sample pack with your purchase. Again, that's drink element LMT dot com slash Huberman. Today's episode is also brought to us by waking up.

Waking Up is a meditation app that includes hundreds of meditation programs, mindfulness trainings, yoga nidra sessions, and NSDR non-sleep depressed protocols. I started using the Waking Up app a few years ago because even though I've been doing regular meditation since my teens and I started doing yoga nidra about a decade ago,

My dad mentioned to me that he had found an app, turned out to be the waking up app, which could teach you meditations of different durations and that had a lot of different types of meditations to place the brain and body into different states and that he liked it very much. So I gave the waking up app a try and I too found it to be extremely useful because sometimes I only have a few minutes to meditate, other times I have longer to meditate and indeed I love the fact that I can explore different types of meditation

to bring about different levels of understanding, about consciousness, but also to place my brain and body into lots of different kinds of states, depending on which meditation I do. I also love that the waking up app has lots of different types of yoga nidra sessions. For those of you who don't know, yoga nidra is a process of lying very still, but keeping an active mind. It's very different than most meditations. And there's excellent scientific data to show that yoga nidra and something similar to it, called non-sleep deep rest, or NSDR,

can greatly restore levels of cognitive and physical energy even with just a short 10-minute session. If you'd like to try the Waking Up app, you can go to wakingup.com slash Huberman and access a free 30-day trial. Again, that's wakingup.com slash Huberman to access a free 30-day trial. And now for my discussion with Dr. Karen Parker. Dr. Karen Parker, welcome. Thank you. It's great to be here.

This is going to be perhaps one of the longer conversations that we've been able to have over the years, in part because whenever I see you on campus, we're heading in our respective directions. But I'm very excited because the topic of autism is one that is on a lot of people's minds. And I think the first question that always comes up, it seems, is

whether or not the frequency of autism is indeed increasing or whether or not the field of medicine is getting better at detecting what was always there over time. Do we have any clear answers to that?

Well, I think it's a multifactorial answer. So we're getting better at detecting autism, right? So in the past, we were diagnosing kids at nine or 10 years of age, right? And now clinicians are able to reliably diagnose kids at two to three years of age, right? So there's more people.

There are pediatricians have autism screeners in house. So when you bring in your baby and over the first couple of years of life, you're filling out screeners that are looking for autism symptoms, right? So there's just a lot more awareness around autism. But the rates have increased to now one in 36 US children have a diagnosis of autism, which is over two years ago, it was one in 44. So one in 36.

Wow, I feel like it was just yesterday when it was one in 80. But is one in 36 the average across boys and girls? Does it skew differently if you look at just?

Male birth versus female birth. Yeah, that's a great question. So autism is male biased in prevalence. So you have, and again, the studies vary. I mean, it's worth noting the autism is a highly clinically heterogeneous disorder, which means that if you've met one kid with autism, you've met one kid with autism, right?

So we have to bear that in mind as we have this conversation. But different studies show that for every one girl, there's three to four boys that are impacted by autism. So there's differences in the prevalence rate, and also there's different monitoring sites. So the way in the US that these data are generated is the CDC has 11 monitoring sites.

across the country and so they follow children and then that's where the prevalence rates come from and they release new prevalence rates every few years.

So if physicians are able to detect autism early, say in a two-year-old or a three-year-old, to imagine that they're working off of tests that don't rely heavily on language. Because even though it can get some verbose two in three-year-olds, most two in three-year-olds don't have a very extensive vocabulary. And I'm guessing that they're also relying on things like visual gaze, among other things.

We've already made clear that this is not a discussion to allow people to diagnose themselves or others. But with that said, what are some of the diagnostic tools that people use? Is it language? Is it vision? Or does it present as?

abnormal auditory processing and maybe you could give us a sampling. So autism is a behavioral diagnosis, right? So unlike other areas of medicine where you might be able to take a blood test or there's other sort of tools, it's all a behavioral diagnosis by an expert. So usually a psychiatrist or a psychologist and they look for two core features. So the

So this is based on the DSM-5, and the two core features are pervasive social interaction challenges and the presence of restricted repetitive behavior. But there are a lot of people with autism who have anxiety. There are a lot of people with sensory challenges. There are a lot of people with seizure disorders, sleep disorders. So again, it's

Each person with autism has this sort of unique collection of traits, and that's how they get diagnosed.

We're going to talk a lot today about interventions, but how early are some of the behavioral interventions? And I should just say any interventions introduced nowadays. So if someone brings their child to the pediatrician and they take one of these tests and that child is deemed as having autism, will the one year old or the two year old immediately go into behavioral interventions?

Well, so usually you need to have the diagnosis of autism and then there are behavioral interventions or a variety of different ones that are used. There are some studies where because autism is highly heritable, you can have one child with autism and then if you have subsequent children, you're at an increased risk of having subsequent children with autism and these are called baby sibling studies.

So what you're doing is enriching the population of infants that you follow prospectively who are more likely to receive an autism diagnosis. And there are studies where some of those children are enrolled in behavioral studies, even when they're, quote unquote, at risk. I've heard before that.

you know, parents in which one or typically both parents are, say, of the engineering, math, physics, quote unquote, hard science type are more likely to have autistic children. Is that true? I mean, did that bear out in the data? You know, if you look at profession or

undergraduate major, does any of that correlate with the probability of having an autistic child? Yeah, well, what I can say is that there's been some studies. So what we know is that autistic traits are continuously distributed across the general population. And there was a study and there's a couple different instruments that are used

to be able to measure these autistic traits. So there's something called the social responsiveness scale, and then that's a US-based instrument, and there's an autism quotient that's a similar measure that was designed in England. And what we know from work with the AQ is that individuals that are in intense STEM fields like engineering, physics, and math have a greater burden of autistic traits, even if they don't have an autism diagnosis.

Okay, so that leads me to wonder whether or not this whole business of a spectrum is actually multiple spectra spectrums. Is it spectrums or spectra? Wait, someone will put it in the comments on YouTube. We know that for sure. Please let me know. I would like to know. What is the plural of spectrum spectrums?

You know, because when we hear the word spectrum, we think, okay, there's a spectrum of severity, right? And in fact, I have some experience with severe autism, not in my family, but where I went to undergraduate university, UC Santa Barbara, down the way from that school was the Devereaux School, which was a school, which has been there for a long time, that parents would send their kids if they were, quote unquote, severely autistic. It was actually where Dustin Hoffman went to

study for his role in Rain Man. And the kids who were really delightful, they used to come into town every once in a while to the coffee shop where I'd study. And they would also continue on from there to Kmart, which is why the Dustin Hoffman character would say, got to go to Kmart, got to go to, he would do that repetition, right? That Kmart was down the road from our, you know, our college housing and the Devereaux school. Those kids were literally in a away from home facility full time.

And I spoke to some of the parents at one point, and they were at that facility, meaning the parents had sent their children away to live there full time. Of course, they'd get visits and they'd get visits home because they were

I suppose we could say at the far end of some spectrum that made it, at least to the parents, idea impossible for them to be at home. Okay. Now, at the other end of the spectrum, if one is just simply thinking in terms of severity, I know people who have self-identified as autistic. That's how they've referred to it. So I feel comfortable saying that they've said, I am autistic.

And they seem pretty high functioning, meaning they have driver's licenses, drive cars, or healthy relationships, and manage life apparently well. They have some traits that, yes, I would agree, are a little bit different, right? So this is where we get into neurodivergence. But I guess the point is, should we think about autism as on a spectrum, or given the fact that there are these collections of different traits,

Could there be a spectrum of severity, also a spectrum of more stereotype behaviors, another spectrum that intersects with that, that has to do with obsession with a particular topic. You could imagine that there are 50 or 60 different spectra, or spectrums. I still don't know which one to say. And that when we talk about the spectrum,

We're really talking about something that's in multiple dimensions and not just one line that goes from severe to mild. Does that make sense? Yeah, I mean, I think this is where understanding the biological basis of behavior would then allow us to be able to say, here's these different dimensions, right? But not understanding the biology. You're left with, okay, are we lumpers or splitters? How do we think about this? Because autism is highly heritable, so there's about

40 to 80% of autism is genetic, right? So these vary wildly, right? But the common thinking is that the majority about 50% of autism is associated with common genetic variants. And so the way that we've always thought about this is that there is this, you know, autism is largely an inherited

polygenic condition and but what I mean by that is that you have a lot of common variants that are additive and so if you think about this collection of common genetic variants that underlie this spectrum right so if you have less of a dosing of some of these common variants you might see somebody who's a lot more who's higher functioning like you said and if you end up with one of these single gene highly penetrant

disorders, you might see severe intellectual disability and sort of lower functioning on the other end of the spectrum. But I think that there is a lot that we don't know. And what you're bringing up, I think, underlines sort of an issue with autism, which is common for many brain disorders, which is like, if you don't understand the underlying biological basis, it also gets very difficult to diagnose and treat, right? And that's where we are with a lot of different psychiatric and neurodevelopmental

disorders. To date, has there been any specific neural network that we can point to and say, ah, that's the neural network that seems to be different in people who are on the autism spectrum. I saw a study published recently that seemed to point to the idea that the genes that are altered in autism at least include a large number of genes that are altered

or the proteins that are the consequence of those genes are altered and exist at the synapse, at the connections between neurons. And I'm asking it that way because, you know, some years ago, I was at a talk on autism at Stanford and someone raised their hand and says, do we even know that autism is a brain issue?

right? Couldn't it be an issue of, you know, the immune system or the cardiovascular system, which at the time seem like, okay, gosh, of course it's, but wait, then you stop and you think, that's a really good question. How do we know it's a challenge of the brain? Right. I think that's a great question, right? And there may be people talk about autism, right? And so when you think about where the major player is, you know, we're at the infancy of thinking about this, right? And so maybe for some people,

it's more of a brain-based disorder. Maybe for some people it's, you know, the connection with the gut and the brain, right? I think what's also really tricky, right? So one thing that you have to ask is what are the barriers to progress in understanding autism, right? And so the way I think about this is that

Let's just take for a moment that this is a brain disorder. How do you study it in people, right? So, you know, it's very difficult to get access to either cerebral spinal fluid, which is a fluid that bathes the brain, brain tissue biopsies. It's very hard to get people, especially children that are really impacted into a brain scanner, right? Because they can't sit still. They may have sensory issues. They don't want to go into a scanner, right? So a lot of the tools that

neuroscientists or psychiatrists have to think about looking at the brain are limited, right? And then the other part is how do you model? So the other way we might think about getting access or thinking about model systems.

What we need to do is think about the control animals. And we need to make sure that the species that we're modeling them in has features of control humans, if you will. So we need to have complex cognitive abilities. We need to have complex social skills. We need to have an organism that has vision as its primary sensory modality, potentially sleep consolidating. So we need to think about all of those. And the tricky part, I think, until

fairly recently was that we were doing all of this work in mouse models. And the control mice just fundamentally lack many of the characteristics that are needed to model autism with fidelity. And I think that's when we look at drug development pipelines, about 50% of preclinical failures. So that would be something that's tested in an animal that works and then fails in a human clinical drug trial.

50% of those failures can be attributed to poorly selected animal models. And so I think part of where we will be getting traction is picking, developing sophisticated models as a sort of point of entry into being able to understand some of these things that are really difficult to study in people.

It's such a key point. And for those that have not heard of preclinical models, preclinical models are nonhuman models. So it could be mouse could be nonhuman primate could be flies or worms for that matter. But we're going to talk a lot about nonhuman primate preclinical models and the work that you've been doing. And of course, also the work that you've been doing in humans.

the other animal. The other primate, right, exactly. I love to remind people that we're primates, old world primates. So thank you for doing that. So you've been talking about the genetic influences on autism. And of course, genes in the environment interact, right? It's never nature or nurture. It's always an interaction in that.

isn't just about the epigenome, it's also just about the fact that nature impacts the genome and our genome impacts the way that we interact with the environment, et cetera. So what is the role of the environment in autism, both the frequency and the presentation of autism?

lots of different epidemiological studies. So advanced parental age, prematurity, severe prematurity is a risk factor for autism, maternal illness during pregnancy. So there's a bunch of different things that have been associated with an increased risk for autism.

In terms of environmental influences and how they can intersect with biology, one of the things that I was really struck by in the early 2000s, that at least by my read of the literature, hasn't really gone anywhere, was this idea that was proposed by Pashko Rakesh, who used to run the neurobiology department.

at Yale, expert in brain neuroanatomy and non-human primates and in humans, embryology, really a luminary of our field, and a series of papers exploring how the migration of neurons during early development, you know, as you and I both know, but most people out there probably don't know because we haven't covered this in the podcast. It's not typical dinner table conversation, you know, when a human embryo is developing that

the neurons are born at one location and they migrate out some distance to their final resting place where then they grow out their connections and connect with one another. And that process of neuronal migration is oh so critical for the eventual wiring of the brain. And Rakesh had this idea that perhaps, and I really want to emphasize perhaps, that the more frequent incidents of autism

might be correlated with the increase in early prenatal ultrasound. And he had these papers published in a number of really high-profile journals, including Procease and National Academy and Science, and elsewhere showing that in a mouse model, if you do ultrasound,

With each successive ultrasound, you got more migration errors. To me, it was an interesting example of the environment, frequency of ultrasound and cell migration, having some sort of interaction. But it seemed like it never went anywhere. It never got tacked to, okay, you should keep in mind the number of ultrasounds that you're getting for your child. And of course, ultrasounds are critical for pregnant women to get.

because they can stave off a number of developmental issues, and they're super important. But, you know, we've heard about ultrasound, you know, within the scientific literature, and then occasionally we'll hear other theories about, okay, it's having two parents who are both engineers, and then we'll hear, oh, you know, it's toxicity in the food environment. We've heard, you know, hypotheses about vaccines or the adjuvants that the vaccines are contained in, you know,

in that large cloud of theories, has anything really emerged from them? It's like, okay, there really seems to be at least one major risk factor, environmental risk factor, because I feel like all those theories come up, get some popular press, bunch of papers are published, sometimes those papers are retracted, like in the case of the vaccines. And then the theory kind of dies.

So is there any specific environmental influence on autism that we can say? Yes, there really seems to be something there.

Yeah, I mean, it's a really spectacularly good question. I think the tricky part about it is that every single person that comes into a trial has a different genetic background, right? And so until we can have these a priori stratified trials where you could then, you know, as a good scientist, you would only manipulate maybe one, two variables at a time, right? But when you're doing these large epidemiological studies because you can't

It's very difficult to do experimental studies, right, especially with developing children. I think that's an incredibly difficult study to do, right? So there's been an interest in this field of there's these neurogenetic syndromes that have high penetrance for autism, which basically means that you could have a disorder

or, you know, another genetic condition, let's say, it doesn't have to be a single gene, but that a lot of those kids tend to also get an autism diagnosis. And so there has been work in, like, so for instance, Fragile X is a good example, where because autism is so diverse in terms of clinical presentation,

that let's say you have a medication that could work for a handful of kids in the trial, you may not be statistically powered to see it, right? So the way I think about the autism world is there's so little we don't know. So think about being in a dark room and you have a flashlight and you only see where you shine the light, right? And so if you think about a very heterogeneous, genetically heterogeneous study,

It's going to be very difficult to tease out these pieces because an environmental risk factor might be a driver for one kid but not another, right? And so I think what we need to do is to have these genetically defined subgroups of individuals and then be able to test the gene by environment interactions or in this genetically defined group of individuals. Can we test this certain medication to see if it's beneficial for this subgroup of children?

So you mentioned fragile acts, which we know presents with autism-like symptoms in some cases. And then I think of another disease like Timothy syndrome, a mutation in an L-type calcium channel, which for those of you that don't know what these L-type calcium channels are, they're not just important for the function of neurons in the brain. They're really important for the function of neurons and other tissues, including the heart tissue.

Kids with Timothy syndrome have cardiac issues and they have autism. So, you know, I think it's important for us to kind of explore this a bit because in most people's minds, you know, kids with autism have autism and occasionally they'll have other issues, you know, gut issues or heart issues or musculoskeletal issues. But we often think that that's the consequence of the autism. But oftentimes they have multiple things going on.

And the autism actually could be secondary or independent of the other thing that's going on. So this is what leads me back to this idea of a spectrum. Is it possible that what we call autism is actually 50 different disorders or 50 different conditions, depending on what wants to call them?

What is autism really? What does it really center around? I think here, maybe it's useful to go, do we go to the diagnostic criteria? How do we decide if a child has autism if they also have a bunch of other things that are challenging them? I think that that's the $64,000 question. Again, in other areas of medicine, let's think about cancer biology.

decades ago, somebody would come in with cancer and you would hit them with radiation chemotherapy, and that was the best that we could do. But with the invention of a lot of molecular tools, you can remove a tumor and you can do molecular profiling and even have personalized medications made to attack that tumor.

what's really tricky when you have a behavioral diagnosis that's not biologically defined, you see a lot of heterogeneity. So it's incredibly difficult, I think, to answer this question because we don't know how many kinds of autism's there are, right? Like there will be people who say if you have a disorder like Fragile Axe or Prader-Willi syndrome or

Timothy syndrome or a variety of these other conditions. I've heard clinicians say, well, that's not really autism, right? That's a piece of fragile acts, right? But if it's a behavioral diagnosis and they meet behavioral criteria, it becomes this weird circular argument, right? So like until we really understand what autism is, I think that

it's going to be very tricky to start sub-defining different aspects of the condition.

I'd like to take a quick break and acknowledge one of our sponsors, Athletic Greens. Athletic Greens, now called AG1, is a vitamin mineral probiotic drink that covers all of your foundational nutritional needs. I've been taking Athletic Greens since 2012, so I'm delighted that they're sponsoring the podcast. The reason I started taking Athletic Greens and the reason I still take Athletic Greens once or usually twice a day is that it gets to be the probiotics that I need for gut health. Our gut is very important. It's populated by

gut microbiota that communicate with the brain, the immune system, and basically all the biological systems of our body to strongly impact our immediate and long-term health. And those probiotics and athletic greens are optimal and vital for microbiota health. In addition, athletic greens contains a number of adaptogens, vitamins and minerals that make sure that all of my foundational nutritional needs are met. And

It tastes great. If you'd like to try athletic greens, you can go to athleticgreens.com slash Huberman, and they'll give you five free travel packs that make it really easy to mix up athletic greens while you're on the road and the car on the plane, et cetera. And they'll give you a year's supply of vitamin D3K2. Again, that's athleticgreens.com slash Huberman to get the five free travel packs and the year's supply of vitamin D3K2.

Well, this is probably a good time for us to think about the work that you've done in terms of trying to tack the biology of social communication and behavior. Those things interact, not just language, but also behavior to

autism in humans using non-human primate models. And then of course, to also discuss some of the work that you've been doing in humans. And we can't have that discussion without first having a discussion about two neuropeptides that I think most people have heard of at least one of them. And I think there's a lot of misunderstanding about, but you're going to clarify that for us, which are oxytocin and vasopressin. So before we dive into the important work that you've been doing on

vasopressin in particular, but also oxytocin and autism. What are oxytocin and vasopressin really? Okay. So they're these small, little peptide. They're nine amino acids long, so very tiny. They only differ by two amino acids, and they're these ancient peptides that are hundreds of millions of years old.

And in almost any species studied, whether it's the current version, you might have vasotocin or other mesotocin, which are sort of precursor forms in other species, but they're highly evolutionarily conserved. And they're involved in social behavior in pretty much any, it could be egg laying, it could be, you know, but reproduction and social behavior across the phylogenetic taxes. So house cats make vasopressin and oxytocin.

Humans obviously make vasopressin and oxytocin, and pretty much every other species that has to interact with and connect with other members of its species. Especially mammals, right? So oxytocin and vasopressin are pervasive in mammalian species. Do the different species tend to make oxytocin and vasopressin in similar brain areas and tissues?

Yes, but not completely overlapping, but I think the thing that the beautiful mystery about these and the infuriating piece of them is that because they're so structurally similar, they can have similar effects and there's four receptors that they bind to. So if you think about a hormone or a neurotransmitter, so oxytocin, vasopressin, if you think about them like a key and a receptor like a wok and you have to put them together to open a door or open behavior,

they can bind to these four receptors. So it can be very difficult to disentangle which one is acting and at which receptor and where in the brain. Oh, so oxytocin vasopressin are chemically similar. Yes. Interesting. Yes. And where would you say lies their greatest

output divergence, which is just nerd speak for it. Is there an example of something that oxytocin does that vasopressin doesn't and vice versa? Yeah. Okay. So what's really fascinating is these two neurotransmitters or hormones were discovered for their peripheral effects, which basically means not in their brain, but somewhere in their body. And so oxytocin's involved

in uterine contractions and milk let down. And so was during lactation. So people sort of always thought of it as the female hormone. And then vasopressin has, at least in the peripheral system, has been involved in urinary output regulation, blood pressure.

we only knew about their physiological roles as sort of classic hormones for decades. And what was interesting is these naming conventions are fascinating medicine, right? So you could name a virus after where it was first found, right? Or it could be named after somebody who discovered the disease like Alzheimer's, for instance, is a good example.

And what was interesting, oxytocin was only named once, vasopressin was named twice. So it's either called arginine vasopressin or anti-diuretic hormone. And so it had two different names. And so as you can imagine, sometimes genes are named twice. And so somebody in cancer is studying one gene and somebody in autism is studying another. And they're not even communicating because they don't even realize that they've at least historically. Now we have all kinds of gene annotation sites, so it's less likely to happen now.

But what was fascinating is these hormones were named oxytocin as Greek for quick birth.

So for decades, people only appreciated their physiological roles, but there are neuroanatomists saying, hey, so these are both made, they're made in a lot of different places, but the action sort of happens in the hypothalamus where they're made. And there were anatomists that said, wait, these sort of project back into the brain. What are these doing in the brain? And one of my favorite historical stories was I had a mentor, a colleague like a, you know, who I didn't train with, but he was,

a real source of wisdom to me for many years. And his name is Court Peterson. And he told me this wonderful story about this Duke zoologist named Peter Klopfer. And Peter was studying ungulates, so sheep and goats. And he wrote a story of paper

in 1971 called Mother Love What Turns It On. And one thing about science is I love going back and seeing where do the pearls of wisdom come from. And so he wrote this and said, oxytocin is orchestrating all these events of motherhood. And there are sheep and goats in particular that have offspring that are precocious, meaning they're basically born ready within an hour. They can run with the herd, unlike our species, which is all-tricial, meaning we have very helpless infants.

And mom needs to bond really quickly with that baby if it's going to be running around and you only, you know, from an evolutionary perspective, you want to be investing in the baby that's yours, not somebody else's, right? And he hypothesized that it was oxytocin that was being

toe released into the brain and during milk let down, that was what turned mother love on. And that was really the beginning of this whole field of thinking. And so that opened up thinking about oxytocin in rodent maternal care and a variety of other instances.

Can I just briefly interrupt you? Because I find this so interesting, and I know it's interesting to everyone listening as well, because yes, and thank you for making it clear the oxytocin has many different roles. But this role of mother love and bonding to infant.

has me needing to ask whether or not the idea was that oxytocin is released in the mother when she interacts with her own baby. And that leads me to the question, is oxytocin also released in the baby in reaction to the mother? And how long is that effect

lasting because in order to have a pervasive bond with that baby and not just some other baby and of course we still have visual cues and you know we know our baby versus another baby most instances their rare exceptions or perhaps not so rare exceptions but leaving those aside you know

The mechanism that would allow for mother-infant bonding and infant-mother bonding by way of oxytocin presumably is something that is literally changing their brains, saying, you are the center of my life. And the baby, of course, is saying, well, you are my life.

because you are the source of life, right? And certainly for the early part of life, and nowadays it seems that that can extend well into the teens and twenties for some people. But how is oxytocin working? Is it working over the course of minutes, hours? Is there some specificity of this baby and this mom that links them in some more pervasive way? I mean, how is oxytocin doing this magic of bonding? Yeah.

I mean, it's very species specific, right? So I think that, and you need to think about like the evolutionary history of the species, right? So if you think about sheep or goats, the early studies that were done are you, the passage through the vaginal canal was what, you know, you would activate it oxytocin receptors that way.

But if you gave an oxytocin antagonist, meaning you would give into the brain something that blocked the oxytocin receptor. So if the oxytocin is being released into the brain, but you have a pharmacological agent blocking its ability to bind to its receptors, these sheep and goats wouldn't bond to their baby, for instance.

So, literally, the passage of the baby out of the vaginal canal triggers the oxytocin pathway, the release of oxytocin. Right, and elactation does too. Nature is so beautiful, because if you had to pick one event to trigger the release of oxytocin, if oxytocin's role is to create bonding with offspring, that would be the event, because that's a tough one to mistake. Right.

But what I will say, because I think you will, you know, to avoid you getting attacked on Twitter or wherever you might get attacked. I'm going to get attacked anyway. If not for this discussion, then another one. But I'm tougher than anyone. But it's really species specific, right? So if you think about our species and a lot of primates species, we live in these extended family groups. And that's how we evolved. And so unlike a goat or a sheep that might live in a herd where there's a lot of non-relatives,

We lived in a community of relatives, right? And so we, and we do all kinds of care of extended relatives. And so you wouldn't necessarily expect in a primate species where you have this long rearing history where help from the family and, and by parental care where, where sort of everybody is sort of like, it takes a village to raise the baby, we readily adopt in our, in primate societies, right? And so,

You know, like I had a cease. I mean, I'll tell you something personal. I had a C section and had I had a lot of postpartum complications. And so lactation didn't work out that well for me. One of my friends would say my massive DV T's and pulmonary emboli. And so I almost died after my son was born the first time. And so I didn't have a vaginal delivery. I couldn't DV T's a divine thrombosis.

Yeah, and it was sort of like welcome to motherhood and I was in the ICU and had to get a filter put in an inferior vena cava filter to stop me from dying because I had scattershot clots all over my lungs. And so I didn't really, you know, I didn't do a vaginal delivery. I had a C-section and I wasn't really able to lactate and man, I love that baby, right? So, you know, I can give, you know, what I will say is

It's really different in primates and we don't really understand how bonding occurs. But what I will say is that bonding between a mother, you really need to think about the evolutionary selective pressure. So I was an evolutionary biologist before I found neuroscience, right? And so I really, everything I do, I think about from an evolutionary perspective.

So, but it is many people go into the oxytocin, vasopressin field because they have a lot of questions about social interactions, right? Like I think if you think about us as being social is actually one of the, one of the

core characteristics of our species, right? So social interactions are rewarding from infancy. They keep us alive, as you mentioned, right? And so I think it's not an accident that the way we think about disorder in our species is many disorders are disorders because of lack of social connectedness, right? So it could be something like autism where, you know, there's these pervasive social interaction impairments. It could be something like drug abuse where

You know, a risk factor for drug abuse is feeling socially disconnected and alone, right? Social isolation or loss of a loved one is a very strong predictor of the onset of a stress-related depressive anxiety disorder.

when and how oxytocin is released. You mentioned mother infant bonding. I think you said yes, that the infant is also releasing oxytocin. We think. So it's bidirectional.

We think, I think most of the work has been done in mom would be, and again, this has not been really done well in primates, right? So we're extrapolating this information from species that have different evolutionary histories than us, right? So it's goat, sheeps, prairie voles, mice, rats. So what do we know about the role of oxytocin in humans? I mean, we know it's there. Yeah.

We presume based on the animal models that it's involved in mother infant bonding and presumably romantic partner bonding, at least you hear that a lot. It was unfortunately nicknamed the Love Hormone. And the reason it's unfortunate it was is that while that might cue attention to oxytocin and I'm a big fan of people paying attention to biological phenomena,

it discards the other and many roles of oxytocin. But what can we say about oxytocin in humans, if anything? Like do we know that it does? I mean, we're just, so we're assuming based on the animal models that it does something. I mean, this is very different than like dopamine where there's tons of animal model data, but we know, but they're brain imaging where we know where dopamine is expressed. And do we even know where oxytocin receptors are expressed in the human brain? Presumably that information is,

is out there. Recently, but again, there's a lot of specificity. And I think if you're thinking about disorders, you would then have to study those specific subpopulations, right? And you need, you know, a lot of this work has been done. So you have to think about how do we study it, right? So the best way to study it would be to have radio tracers where you could then, which we do have for dopamine and other compounds, where you would then go and see where after somebody's performed a task, do we see, you know, activation, right, or uptake?

There are some imaging studies. They're usually done giving intranasal oxytocin. And then you basically ask questions about, okay, we give you oxytocin intranasally, which presumably enters the brain. We could talk about reasons why we think that. And then we have you perform on some task, right? And so, you know, there's evidence if you give oxytocin, it diminishes the amygdala's response to fearful stimuli, right? So that it might have

this sort of prosocial effect. And it was actually data like that that caused people to start thinking initially about oxytocin. And those were data in humans. That's right. It reminds me that there was this brief moment where oxytocin wasn't just being discussed as the love.

Hormones, it was being discussed as the trust hormone. Also far too simple heuristic, but again, I think it's cool that the press picks up on these things and at least tells people about what's being discovered and we just always have to be careful not to

Have it lead to the assumption that that's the only role of a given of a given hormone. So it can reduce Apparently it can reduce the output of the amygdala in some way this brain area associated with threat detection And so you could imagine how that would bias the person toward being more prosocial, right?

Have there been studies exploring the role of oxytocin in making autistic children more prosocial? And behind that question, I suppose, is the assumption you can verify or not that autistic children are less prosocial than other children? Is that true? Or is it that autistic kids are just maybe more prosocial with the one friend they really, really like? I happen to know some kids with autism or

however you want to phrase it, and they have close friends, and they seem to really like those specific friends a lot. They seem very happy when they show up at the door and all the hallmarks of healthy social mind, but it is true that they are uncomfortable in groups and where there's a lot of noise, a busy birthday party is overwhelming for them, but see them playing with one or two friends.

You could see all that and assume, okay, it's just kind of an introverted kid. Actually, it kind of reminds me of me, you know? I mean, I don't have a problem with crowds, but I much prefer to be with a small group of friends or one close friend. Yeah, I hear you. I'm not way to. Right. So, you know, how do we think about this? Okay. Well, I would say,

the social features of autism are interesting, right? And so you might have, there was an attempt a long time ago, like 1979, there was a woman named Lorna Wing who tried to subtype the social features of autism, right? And so there could be people that are socially avoidant.

and really just don't want to have social interactions. There could be kids that are active but odd, which means that they have an interest in being social, but maybe they don't read social cues right and they interact in ways that other kids

don't understand or make could cause bullying, right? Something between your high school. Yeah, exactly. And that's often why some autistic kids do better with adults, right? Because adults know how to sort of channel discussions with somebody who might be a little socially awkward, right? But there's different phenotypes. I mean, people having a disinterest in social interactions could be that they're highly socially anxious, right? That making eye contact makes them anxious. You could have somebody who has

maybe is relatively, let's say, socially intact, if you will, but they have overwhelming sensory abnormalities that make it very difficult to interact with other people, right? And so let's just say,

Again, that's another caveat. There have been some studies administering oxytocin to individuals with autism. And again, these are these single dose studies. So the first studies that were done were looking at single dose oxytocin in males because some of the, and we can talk a little bit about why oxytocin versus vasopressin, which vasopressin actually would have been my choice based on the animal literature. And we can talk about that.

But this oxytocin was given to males, partly because it wouldn't, the idea would be that the off-target effects in the peripheral nervous system, i.e. milk, like down uterine contractions, are not gonna happen in males, right? And so that it was deemed that they might be safer subjects. Males are often also the go-to for research studies, as you may have talked about on your podcast before, too. Yeah, something that fortunately is changing. Yes, absolutely. Thanks to a mandate by the NIH. Correct.

I had to just kind of smile slash raise my eyebrows a little bit at the idea that, you know, the assumption that oxytocin administer to males, yes, one can see why it wouldn't cause milk like down or uterine contractions. Right.

But of course, there could be other peripheral effects of oxygen in males. But they had to pick one, so they went with males. And there is this higher incidence of autism in males. So it's not a terrible place to start. You just would hope that they would also do the experiment on females. So they're doing this by nasal spray. So intranasal. One dose. Correct. And for reasons that I don't understand, it's 24 international units. And I think maybe somebody did the first study using it. And this is how science happens. And it worked. And so then everyone uses that protocol.

And so then there's been a lot of studies looking at, you know, there's one reading the mind in the eyes. So can you look at pictures of somebody's eyes and then ask, what is the emotion that they're feeling, right? After receiving this introduction or placebo.

Where's your eye gaze going in a picture, right? So one of one of the theories is that people with autism may at least a subset of them lack social motivation. So maybe they're not looking in the places like eyes where you receive a lot of social cues that are relevant to social communication.

And so some of these early studies showed that a single dose of oxytocin in people that had high functioning autism, so they were verbal, like you said they could come in for studies, and that it looked like it had some potential effectiveness, and so there became a really strong interest in the field to think about oxytocin potentially as a therapy for autism.

And is oxytocin available over the counter? Does it require a prescription? I mean, you see sites that are selling it, but that doesn't mean anything these days. There's gray market. There's all sorts of stuff going on. But I know people that have used oxytocin, there's actually a market for, and by the way, folks, I'm not suggesting this, but someone the other day told me that they've been regularly taking

oxytocin ketamine nasal inhalations as part of their work with their licensed therapist on PTSD type stuff relating to, let's just call it relational trauma. So that's happening. But let's just think about oxytocin alone for the moment.

parents of autistic kids able to like bioxytocin nasal spray? No, so it would need to be written like the prescription would be need to be written by a by a physician. And it's not on the market, right? So there's one thing we should say is there's only two drugs that are approved by the FDA to treat autism and they're both antipsychotics, which they

They treat associated features like irritability, and they have off-target effects like weight gain, and so we don't have any medications that are currently approved in the US or anywhere else for that matter to treat the core features of autism. Interesting and unfortunate. Hopefully that will change in the not too distant future.

Do we know that children with autism, people with autism, because I'm going to just sort of assume that autism is stable over the lifespan? Like if a child is diagnosed with autism, are they going to be an adolescent and adult with autism?

So I would say that in a lot of cases, autism has lifelong impact, but there are people who outgrow their diagnosis. There are people who respond well to behavioral therapy. I mean, obviously, it's not the cure-all for everybody. There's lots of people who go through intensive behavioral therapy and probably see minimal benefit. But it's certainly something that

occurs in childhood, for the diagnosis occurs in childhood, and it, you know, for most people will then be present across the lifespan. So we could say people with autism because each study sometimes will have adults, sometimes you'll have teenagers, sometimes you'll have kids.

Is it known whether or not people with autism, assuming they meet the criteria for being autistic at that moment, have lower natural circulating or active levels of oxytocin? Because it's one thing for a nasal spray of oxytocin to improve social functioning. It's another to know that the effect is addressing an underlying biological deficit.

It's such a great question. Okay, so we should unpack that because there's been a lot of work in this area. So the first question is where are we measuring the oxytocin, right? So we mentioned oxytocin has all kinds of effects in the body as well as the brain. And it's released into the blood, but it's also released directly into the brain. And there's variable evidence about if you measure it in blood, is it a readout of the brain or not, right? Or should you be looking at something like spinal fluid that's maybe a better biochemical proxy of the brain?

Most studies, so what I will say is there's been a handful of small studies where there's been some benefit, maybe no benefit, small effects. We did a study

that was a small study at Stanford, and it was based on mouse genetic data, and I'll sort of walk you through what we did. So there's multiple mouse models of these neuro-genetic syndromes where people have social impairment, right? We can quibble about whether that's autism or not, but that they have social impairment.

And so that there are this fragile X-mouse, there's a proder willy syndrome mouse, which is the Magel 2 gene that gets manipulated, and then there's a catnap 2-mouse. And in all of those instances, when you genetically modify those mice, you see a reduction of oxytocin in the hypothalamus.

And what's interesting is that in those instances where you see this genetic modification, you do see lower blood levels in these genetically defined models. What's really cool is you can give oxytocin across development in those models, and at least in the catnap2 mouse, you can restore oxytocin neuron number two equivalent of control animals, suggesting that oxytocin is doing something

in these oxytocin deficient animals. So these are not an oxytocin gene manipulation, but these are these syndromes where you see as a consequence of manipulating genes for these syndromes that oxytocin gets knocked down. And so our thinking when we went into our clinical trial was what if its blood oxytocin levels

that there are going to be a subset of individuals that just make less oxytocin, humans, and that maybe those are the individuals who stand to benefit the most from treatment. And so we were the first group to ask

You know, across this range of individuals who showed up and we did in all the trials that we'll talk about today. These are done with my colleague Antonio Hardin at Stanford who's a child psychiatrist. And we always have double blind, meaning that the investigative team is blind and that the

or unaware, I should say. They're unaware of treatment, and then the families and the children are unaware. And then the randomized meaning there was an equal chance you could get either drug or placebo, and they're controlled, right? Okay, so we asked if we know what your pre-treatment blood oxytocin level is.

Who's gonna benefit from treatment? And we've had a couple of really interesting things. One was that the lower your baseline, so your pre-treatment, blood oxytocin level, you showed much greater benefit from the oxytocin intervention. These are- One intervention, one nasal spray. This was four weeks, sorry, I should have clarified. This is four weeks of treatment being administered oxytocin twice a day. Okay. And so we saw effectiveness there.

Sorry to interrupt so much, but just males, male and female subjects. We did. But again, you know, because the autism is male biased in prevalence, even if you make this heroic effort to over recruit, try to get more girls. And in the study, we usually try to aim for the prevalence rate because it's difficult to get girls just because there's fewer of them. Got it. Okay. But boys and girls were included. Correct. They're taking oxytocin over the period of several four weeks. And if they started off with lower based on levels of oxytocin,

you observed a benefit of the oxytocin treatment in those individuals. What about the individuals who had normal time to high levels? You didn't see much benefit, right? And so that was a cue to me to think that there may be a subset of individuals that, for whatever reason, they have lower oxytocin and they may stand to benefit more from treatment. And none of the prior studies had looked at blood oxytocin levels. And so what we had thought was that, well, maybe,

If everybody had measured baseline blood oxytocin levels, maybe there would have been more positive outcomes. But there's a lot of controversy in this field about whether oxytocin is a treatment for autism. So after we completed that trial, there was a large multi-site

What's called a phase three oxytocin treatment trial that was done at I think five sites and they gave oxytocin for an extended period of time and they showed no benefit and and Were they looking to see who started off with low levels of oxytocin that pre treatment?

So what was interesting about that study, and there were a lot of issues with it, was that oxytocin is something where you have to, if you look at it, it degrades. It's like, that's kind of what I joke about, right? So you need to take it. We take, when we go in, we have like these really intense protocols, right? So you go in and we have vacutainer tubes that are cold and we put them on ice and then the phlebotomus takes the blood from the child. So a lot of technical gymnastics.

And then we make sure we spin it in a centrifuge cold, and then we pipette it onto dry ice, so we have very minimal loss of the signal. And so if you don't adhere to those rigid protocols, which is very difficult to do across multiple sites,

It can be very difficult to get an accurate read of oxytocin. And so I think for me, it's still an open question. They didn't see the blood oxytocin predicted response in that study. The data weren't provided in the paper. It was just said that they didn't. But it's still an open question to me. Like what if there was a group of children who had low oxytocin levels and they could benefit, right? There's other people.

where they'll say, no, no, no, we don't think that chronic oxytocin's a good idea that what you really should be doing is just giving it before a behavioral therapy session, right? And so that, you know, maybe that is the way. So if you give it acutely, like in those early studies we talked about, that maybe oxytocin diminishes fear. We know that oxytocin decreases

the stress axis, the hypothalamic pituitary adrenal axis, and then it can diminish anxiety in animal models. That's well established. In a former life, I was a stress researcher, so I've spent a lot of time thinking about this. The sad thing is that once you have a negative trial, there isn't a lot of interest in funding

the work going forward, right? And so I think it's still really an open question about if there is a subset of individuals that could benefit from oxytocin replacement therapy, right? And it's, and until there's money to do that work, we may not ever know the answer.

It will be important for that work to be done eventually. Hopefully the field will return to it despite whatever trends might be happening now. I think it's important to know for the parents of autistic children whether or not there were any negative effects of oxytocin administration. In particular in the children that

did not benefit from oxytocin treatment. The rationale is the following. Well, of course, these things require a prescription if a parent has a child with autism, especially if they're young enough that the behavioral interventions could possibly stand a good chance of inducing neuroplasticity, rewiring of the neural circuits that underlie social connection. Well, then there's this time-limited window

in which those parents presumably are willing to try most anything provided it's safe. So let's assume, and I'm making up these numbers now because I haven't seen this study, but according to what you told me, that let's say a third of the autistic boys and girls that come in have low baseline levels of oxytocin. They're the ones that are going to benefit from this oxytocin intervention. The other two thirds don't.

Well, given the difficulties of measuring based on levels of oxytocin, most people don't have access to those kind of resources. If it's safe to give oxytocin, no matter what, well, then if I were that parent, I'd be knocking on my physician's door saying, hey, give me oxytocin spray because my kid might fall into that one third category. If and only if

It turns out that oxytocin is safe to give. But if there's a risk profile that doesn't justify that kind of shotgun approach, well, then I wouldn't do that. So is oxytocin spray safe? And if so, why doesn't every physician who has a patient with autism give them oxytocin nasal spray?

It's a great question, and I know that I'm a parent of three children, and I know this sense of like, you would do anything to help your child, right? And so I think the tricky part is that, so one thing I will say is that all of the studies, and there's been many of them, have shown that oxytocin is relatively safe in a pediatric population, right? The tricky part is I don't know, there's physicians that really pay attention to clinical trials, and if they don't see a benefit,

they may not be willing to write the prescription, right? So until we could identify a group of children that could benefit, you know, we need to create the opportunity for physicians to recognize that this could potentially still be a treatment, right? But that work, you know, but I think the tricky part in what I will say is, and we can maybe talk a bit about vasopressin, which, you know, my feeling is that if I was placing bets and having to choose between these two, my money would be on vasopressin.

Well, we are definitely going to talk about vasopressin in detail. I mean, the reason I mentioned that hypothetical scenario is just the sense of urgency. And in some cases, desperation that parents feel and, you know, times ticking. And if oxytocin safe, then, you know, I guess I'll put in my vote that, you know, parents should at least talk to their physician, maybe even hand them the study.

to consider. But I can also understand the perspective of a pediatrician who says, well, listen, it was a small number of kids that benefited. You're welcome to try it, but it doesn't seem like the results are that impressive. But this gets to a bunch of larger issues about medical care and randomized controlled trials and the desperation of parents and kids to treat neurodevelopmental challenges.

I just want to ask because it feels relevant in a real way. You know, if ultimately the goal of improving symptom profiles in autistic kids is about improving social cognition and social behavior. And that process involves rewiring of brain circuits, neuroplasticity. Is there any reason to think that

other approaches to inducing neuroplasticity would be beneficial, even if they're not in the biological pathways that are disrupted in autism. I think, for instance, about the now extensive use of SSRIs for the treatment of depression. In some cases, it works. In some cases, it doesn't. Side of profiles are a serious concern.

as I've discussed on this podcast before, but ultimately, we know that depression is not a serotonin deficiency. In most cases, SSRIs are atypical antidepressants like Propriorone, Wellbutrin, and things of that sort. When they work, they probably work because of their ability to induce neuro, or assist neuroplasticity. Also, the trials on psilocybin are not really about

psilocybin, they're about neuroplasticity, at least the trials for depression, right? There may be other uses of psilocybin that relate more directly to the effects of psilocybin. But ultimately, you know, what we're talking about here is the attempt to rewire the brain in a specific way, whether or not it's assisted by oxytocin or some other mechanism. So the question

is are there trials happening where people are exploring, say, psilocybin, MDMA, which, by the way, we know, increases oxytocin and serotonin dramatically, as well as things like atypical antidepressants in kids that have autism, not because we think that those autistic kids are deficient in any of the neurochemicals that these drugs would target, but that these drugs can help rewire the brain. And ultimately, that's what these kids need.

It's a really great point. And there might be subsets of kids, right? There might be kids where there would be a medication that would target other pathways, but that potently releases oxytocin, right? But there might be kids that have an oxytocin deficiency, right? But I think that that circles back to your point at the beginning where our point is that autism is a very heterogeneous condition and being able to know before you begin a trial, right?

Who am I gonna put into it? And what is my primary outcome? Like one measure that I think is gonna move the needle, right? Like it kind of requires a crystal ball. So there's a lot of guesswork that goes into this. But I would very much like to see, I will say one other thing that, I have a colleague named Adam Guistalo who's at the University of Sydney. And he published a paper a year or two ago now suggesting that oxytocin may be most effective in kids at younger ages.

And I don't quote me somewhere between two and five or three insects or something. We'll find the paper and put it in the show notes. Yeah, but you know, so it could be, to your point about neuroplasticity, that oxytocin may be maximally beneficial in younger ages, right? And if these studies are these hodgepages across ages and across sort of different social phenotypes,

Finding that signal is really important, right? And maybe age is a driver or maybe, you know, low blood oxytocin, regardless of what age you are, or maybe in Adam's case, if you recruit really young children, you're likely to see a benefit just because the brain is wiring up and it's more plastic at, you know, younger ages.

Yeah, that's also a vote, in my opinion, for early examination of kids, right? Like parents really need to get autism screening and perhaps maybe the

most important thing is to make autism screening as available and as inexpensive as possible for everyone because of the importance of early intervention even if it's purely behavioral intervention but certainly if it's behavioral and drug intervention. But the clinic wait times are really long, right? So you have to have a specialist who's capable to diagnose autism and so you could have a clinic where you know you're showing

trouble some features and a parent wants to get their kid into a clinic and you could have a 12-month or 18-month wait time, right? And so there were a lot of people that are thinking about are there laboratory-based tests that we can develop, maybe either for detection or clinical referral, right? So could we come up with a biomarker panel, for instance, where we might be able to say, wow, here's some

Here's a panel where we think this child is at reasonable risk for developing autism. Can we make sure they're prioritized for getting a diagnosis right so we can get them an early intervention, but right now we don't we don't have that right so.

having some sort of laboratory-based test, whether it could be biological, or if we could do something with eye gaze. And there's a lot of companies working on these things now to say this may not, you know, and also, obviously, again, autism is always

controversial in this field, right? There's so many different stakeholders. A lot of clinicians will say, well, I don't want a 30 second video clip replacing expert clinical opinion. There's good reasons for them to feel that way. But I think if there was a way to prioritize people that are in this line, we could get diagnoses faster.

Well, you wouldn't want false positives, but I would think that a 30 second video clip provided it's of something useful. It's going to be more valuable than nothing given the time sensitivity. Uh, what are some of the barriers to getting this behavioral testing to be not just more prominent, but pervasive? Like it seems to me that, well, I recall in school, they gave us the hearing test. We all marched on the bus. We get the beep test and, you know, um,

you know, for hearing challenges, we get vision tests, you get the Babinski reflex test, not the moment you come out of the womb, but pretty soon after. I mean, why isn't this stuff happening for autism?

for every kid. Yeah. It's not scalable, right? So you, these interviews with parents and the tests that you do can take hours, right? And any given clinician, even if they're working really long hours, there just aren't that many people that are, have the extensive training needed to make these expert diagnoses, right? And so I think that there's, you know, clinicians that are doing the absolute best they can, but they can only see a certain number of, of people a week, right? Does that have to be a physician? Sorry to interrupt. Does it, or could it, could, you know,

Could a well-trained technician do this? Yeah, well, I mean, I think technically it's a DSM diagnosis, right? So it's usually somebody with a clinical degree. So it would be a clinical psychologist. It could be a behavioral pediatrician. It could be, you know, a child psychiatrist or child neurologist. But I mean, again, that requires years and years of training. And if we look in areas where

people have fewer access to resource. I mean, particularly in impoverished areas, the mean age of an autism diagnosis is years later than in wealthy areas where there's many different medical specialists with parents that aren't working three jobs and can sit waiting around and really lobby and really advocate for their kids because

you know, if they don't show up for work that day, they're not going to get fired from their job, right? And so I think that, you know, if there's some sort of solution that allows there to be a more democratic approach to saying we need a really quick way, like you said, to be able to identify at-risk children, especially if it's a blood test or something like that, you know, it could be incredibly impactful.

Are there human trials exploring MDMA, methylene dioxide, methamphetamine also referred to as ecstasy, and or psilocybin for treatment of autism? So I was aware that maps had an MDMA trial in autism. I don't know what's happened with that.

Yeah, perhaps it's still ongoing. I'll check the maps site. I'm in communication with them from time to time. I mean, the reason for asking it, of course, you know, but maybe in case some of the listeners don't is the MDMA causes these massive increases in serotonin. That seems to be the major source of the MDMA effect, so to speak.

based on the work of our colleague, Rob Malenka, and at least one human study comparing MDMA to very high dose oxytocin treatment, ruled out the oxytocin spike that's induced by MDMA as the source or the only source, but of course, these chemicals can synergize. But based on its chemical profile, oxytocin release, massive serotonin release,

dopamine release and propensity to enhance neuroplasticity. I mean, assuming all the safety protocols were there, seems like not the perfect drug, but not a bad choice if, of course, it's inducing the kind of plasticity that someone with autism would be seeking.

Right. I mean, I think the tricky part, especially in children, is there's going to be a reluctance to potentially give them psychedelics. And so is there a way to modify the chemical compound to be something that parents might be more willing to give to their children? Right. And I totally agree with that. I guess to play devil's advocate, not against you. But I'll just state it very directly. And then I'll take the heat as necessary.

I've done two episodes about the, uh, the drugs that, you know, millions, tens of millions, if not hundreds of millions of parents are already giving their kids for ADHD, which are, um, include amphetamines, including dioxine. Methamphetamine is actually a prescription drug for a very small subset of kids with ADHD, but things like Adderall, Vivance, even methylphenidate Ritalin. I mean, these are amphetamines. They induce dopamine release and norepinephrine release. And again, I'm not suggesting people, um, give their kids MDMA.

to try and ameliorate symptoms of autism, but something chemically similar to it ought to be developed or at least explored in a human trial, in my opinion. Well, time will tell. I'll reach out to the maps group and see what's happening. Let's talk about vasopressin because there's a lot to discuss there. You told us this is a molecule that chemically is very similar to oxytocin.

Is it manufactured in the human brain and body? Yes. Okay. Do we know a subset of the sites that it's known to be produced and where some of its actions are? And you mentioned the kidney and the antideroretic hormone roles, but within the brain, like what brain areas have neurons that make vasopressin?

We'll have the receptors. Yeah, I mean, the receptors are all over the brain. And again, it varies depending on the species. And the way the receptors are measured are post-mortem tissue, which can be very difficult to get good.

samples, right? And so we need to have that caveat going in. But yeah, I mean, it's made in the hypothalamus and it's released all over the brain. And there's vasopressin receptors all over the brain, right? And what's really interesting about vasopressin, I always sort of joke that oxytocin, you know, always saw it today in the sun, if you will. And the vasopressin was sort of the stepchild that was like left, you know, sort of behind. And the

The reason why I find this fascinating is again, like I think back to my roots as a evolutionary biologist, behavioral neuroscientist. And what was interesting is that there were studies in the early to mid 1990s showing that vasopressin was critical for male social behavior.

And so there was work, you know, there was a variety of people, and I think Rob Molinka mentioned this on the podcast he did, about, you know, there was a group of people like Sue Carter, Larry Young, Tom Insol, some of these early people. And they gave vasopressin to male prairie voles, and vasopressin was what induced pair bonding with a female mate, and also paternal care.

And as I recall, those experiments were done in the context of looking at polygamy versus monogamy of these prairie voles. Prairie voles versus like a different species. So same genus, but a different species. So it might be a montane vol or, you know, highly related, but these other species. So prairie voles are monogamous, the males. Well, I mean, that was the. 50% divorce rate.

Yeah, that was not, I don't think it's that bad, but they're doing better than we are. We should look to them for pointers. And all the divorce folks are saying, wait, why'd you say better? I have some divorce friends that have said divorce is like the greatest thing. So we always say like doing better, doing worse, right? Anyway, that's a whole other podcast and certainly not the you room and lab podcast, but or maybe it is, but or will be, but yeah, my understanding is that you have certain voles that mate with

almost exclusively with one other vol for their entire lifespan. And then you have other voles located elsewhere that in those colonies, they mate with lots of different voles. So the males and females have lots of different partners, raised young with lots of different partners, mating with lots of different partners, and that if you

give vasopressin, then you can make the, I always want to call them polyamorous, but I don't know if they love each other. I'm going to answer for more files and assume they love each other. The polygamous moles, not polyamorous, but polygamous moles, then become monogamous.

Well, yeah, I would say that is probably not the take-home message. So the take-home message would be they had, let's say that there was like the good voles, right? Which are the prairie voles? And they were the ones that formed these monogamous parabonds. Dad participates in paternal care with mom. They co-raise babies together. And then dad chases off intruders, right? And then there's the more a social voles. And so these are like the montane voles.

Well, we'll see. It's a complicated story, but there's these montane voles where males and females live separately. Females like maybe live on the males territory. The male mates with a few different females absolutely doesn't provide any paternal care at all. Mom raises babies by herself, right? So that's these are really the two. There's like 1950s versus 2020. Yes. Yes.

to broadly stereotype. And if you give, okay, so for parivoles, they're sort of primed to form bonds and to be the males, to be good daddies, if you will. And all you have to do is give them a single injection of vasopressin and

you know, or you can give an antagonist and usually the way they form the bond is through mating, right? So they, you put them with a female, they mate, they cohabit for a bit. There's been all kinds of parametric studies. I can't remember how many hours it takes to form a parabond.

But then you can do these things called partner preference tests. And then you can say, here's the guy that you made it with, here's this guy you don't know, and you can do it for males, and you can do it for females, and they pick their partner, or they choose to go hang out with their partner. The montane voles, you know, either after mating with somebody may either be equal, or maybe they'll even go spend time with the new individual. So the cleanest story was that prairie voles are monogamous, montane voles are not monogamous, but in the prairie voles,

you could give vasopressin, instead of made it to habitation, and you could turn on, like, you know, a bond with somebody after only living with them for a very short period of time, right? Or you could induce paternal behavior. And I was working with the Voll species in grad school. I think the most

interesting scientific experience that I've ever had, right? And you and I both know this, right? When you're young, you're actually the person doing the work, right? As you become, you know, the head of your lab, you're mostly writing grants and giving talks, right? And then you get to hear about the super cool things that everybody in your lab is doing, right?

Eventually, the members of your laboratory kick you out of the lab. Exactly. They literally say, like, get out of here. You're leaving things in the wrong place. Whereas initially, you're telling them, hey, that's in the wrong place within a year or two. For me, I think it took about four or five years. But by about year six, I was.

demoted to my office to just write grants and write paper. I was told that one time I was back there and I tried to wait and I was like so excited what they were working on and they basically just said go write grants and bring in more money right like that was kind of their attitude like we get to be the ones who get to do the cool stuff so back when I got to actually do the science.

I remember I had this species where, and again, I told you I came at this from an evolutionary perspective. So these were called meadow voles. And I found them very interesting. So when I showed up in my thesis advisor's lab, I said, I really want to study oxytocin and vasopressin. And I really want to study voles. And I know you have a voles species. And she said, why don't have prairie voles? I have these meadow voles. And I'm studying them because they're so sensitive to light. And they changed their behavior based on light. And she said, well,

You can do what you want, but our grants basically have to have a circadian component. So she said, you got to work that in, but then we kind of struck this deal. So I was hanging out in the animal rooms, and I thought it was really fascinating. So she had animals that were either on short day lengths or long day lengths, so the mimicking summer and winter.

And I was noticing that on winter day lengths, the males were hanging out with the females. And when the female had a litter, he was like participating. And I was like, whoa, these are not supposed to be monogamous animals. And so I went into the field research and they were doing all these radio telemetry studies. And so like if you, these are probably explain what those are, putting a little transmitter under the skin, it's painless for the animal, but that allows the researcher to

monitor the behavior of the animal remotely without having to, you know, put them in cages and stuff. And so this is like under field conditions. And voles are everybody's favorite snacks. So they have like a very limited lifespan in the wild. I mean, like on the order of months. And so like if you have a short lifespan, like you should just keep reproducing, right? And so what was interesting is at the end of the summer days, as you're going into winter, territories collapse and males are found with females.

and they co-raise babies, it makes sense. If you're going to have a litter and mom needs to get up to go eat, you need somebody to sit there and warm those babies so they're going to die because they're going to freeze to death, right? So I started saying like, wow, I think these meadow voles are good dads. Like I'm noticing this. And so I told my thesis advisor, I want to study how oxytocin and vasopressin can

Maybe this is involved in tracking these evolutionary mating strategies. And so again, like the coolest experience I ever had was on these males that were housed under short day lengths. So they were like winter males. I was able to put vasopressin directly into their brains. And it was like turning on a light switch. And they ran around the cage, picked up all these babies, put them in a nest, and huddled over them. And if you put a placebo into their brain, nothing happened.

And so to me, I always filed that away in the back of my mind of like, wow, vasopressin is this really interesting hormone. And maybe someday I did a postdoc on something else, but it was always back in the back of my mind of I really want to return to this.

It's so incredible that a eight amino acid long peptide could basically turn these relatively negligent fathers into very attentive fathers. Yes. Yeah, it was fascinating.

I mean, it just speaks to the power of the peptide vasopressin. It also speaks to the power of brain circuitry. It also speaks to the idea that brain circuitry is often sitting latent in the background, ready to be activated, that it's not just about neuroplasticity and building up a new circuit that some forms of neuroplasticity are about unveiling what's

what's already there. And that peptides can act like switches, which kind of makes sense on the one hand, but I've never heard of a result as dramatic as that.

I'm presuming you're gonna tell us that then led you to go back to vasopressin and explore its ability to induce good parenting and negligent fathers. I haven't studied that yet. No, well, so I think that, you know, my mom always says, chance favors the prepared mind. And so I was doing my postdoc at Stanford and I got recruited to stay on the faculty. And I, you know, had been doing work in stress, vulnerability and stress resilience. And I really, and I love doing that work.

But I still felt this tug of, you know, I had spent all this time in a psychiatry department where I was surrounded by clinicians. And I realized that a lot of the stuff that I was doing had clinical relevance, right? And so sometimes you sort of meet the moment, right? And so right as I was transitioning to have my own lab in my department, there was a bunch of stuff going on. So there were a lot of very dedicated parents who were lobbying for funding for autism research because it was

horrifically underfunded. Really? Horrifically underfunded. Wow. I mean, at rates of 1 in 36 kids. Well, not at the time, right? So it was 1 in 150 or whatever it was back then. But there were all these parents, and I mean, again, they're heroes in my eyes, that they advocated so much for their loved ones. And so there was, you know, they started forming parent grassroots organizations that have culminated. They all started joining together, which is now Autism Speaks.

And then there was a man named Jim Simons, who runs one of the most successful hedge funds in the world, and he decided, wow, I'm gonna, you know, let's put money into autism, right? Does he have a personal link to autism? I would have to ask him. Because oftentimes, not always, but oftentimes when you hear about

wealthy donors, devoting a lot of money to one area of science, there's a familial thing there. A member of their family or a close friend has this challenge and they really want to see that challenge. Absolutely. A lot of money I've gotten for my lab from philanthropists and what I will say is the most impactful work I've ever done is through philanthropy. They're crazy ideas that no funding agency ever touches.

But yeah, so they both put a lot, you know, there was a lot of emphasis. And so because the Simon's Foundation started issuing requests for applications, there was a group at Stanford that formed and it was a clinician with a basic scientist. And my chair at the time said, well, you know, almost nothing is known about the biological basis of autism.

why don't you go, I'm going to introduce you to the head of child psychiatry, you should go talk to this group. And so as I was preparing my slides and realizing that social interaction impairments were a core feature of autism, I thought, wow, these neuropeptides may really be

you know, a part of this puzzle. And so that's actually really how I got pulled into autism research was through that. And it was, I was, you know, everybody at the time was very interested in oxytocin. And, you know, I remember thinking, so we actually did probably the most definitive blood oxytocin study because there was this idea, again, like this marketing campaign of like the oxytocin deficit hypothesis of autism.

And, you know, given how clinically heterogeneous autism was, we got money actually from the Simon's Foundation, and we did the first study with maybe 200 kids. And what we were able to show was that blood oxytocin was not a marker of autism, right? So it wasn't like there was a bimodal distribution, meaning two completely non-overlapping levels of oxytocin in people with autism, people without autism.

So, the lower your blood oxytocin levels, actually, regardless of who you were, you could be a child with autism, you could be an unaffected sibling with autism, or you could be an unrelated control child. And it was the lower your blood oxytocin levels, the greater your sort of social difficulties.

And the slopes were different. They started at different points because the behaviors were obviously different. But that's what got us thinking about our clinical trial, which is that blood oxytocin level is not going to be this great differentiator between people with and without autism, right? But we might be able to find a subgroup who could benefit from treatment. But what I like so much about your approach, the way you described it, is that

It sets aside. We don't want to say discards, but it sets aside this thing that we call autism, which is already hard to define and diagnose, and there's all these different spectrums, and he just says, okay, children with autism have challenges in social cognition, social behavior, social bonding. So do adults with autism, for that matter. Let's just focus on that.

and not worry so much about whether or not somebody is diagnosed as autistic or not, and just focus on what are some of the potential neuropeptide deficits or overexpression of neuropeptides that may in some way relate to those social challenges.

And then one can circle back to the question about autism in collecting those data. But it also points to this idea that when we go after a disease, like Alzheimer's, we can often miss the possibility that Alzheimer's, while it has deficits and cognition and memory, could also be a bunch of other things, like a metabolic disorder of the body. And so maybe you go after a particular symptomology and try to attack that. And you might actually

potentially treat or cure multiple diseases. It's a very different approach, and I hope people are catching on to the subtlety, but also the potential impact of that.

because if I heard correctly, you said there are people who are not autistic who have social functioning deficits, and they, too, have less circulating oxytocin. Right. So I would say we haven't studied people where we brought them in and characterized it, right? So these are typically developing kids.

But what we did is in the abilities that are typical of a control child, we still saw that gradient, right? And so I think it just sort of begs the question about, you know, what is oxytocin's role in human sociality, right? I mean, I think there's just so much that we don't understand about both of these molecules in terms of their

disease liability if they're low or they're healing potential if we are able to use them as modulators of other therapies.

So how did you move from oxytocin to vasopressin? You mentioned that everyone was all excited about oxytocin, still the one that we hear the most about. Yeah. Although after this podcast episode and when I start blabbing about vasopressin to everybody, maybe that'll change, but I think it's going to take a lot more than that. But maybe it's because the name is

There's something about oxytocin that like kind of sounds like the love looks like the love hormone, but like vasopressin should be renamed. Right. Well, it should be called something else like not anti-deretic hormone, not vasopressin. I mean, you're going to tell us how critically important it is, perhaps even more important than oxytocin for autism and social functioning. So I don't know, by the end of this podcast, we'll come up with a new name.

it's needed, right? Well, I'll put it out there. Okay, so how did you get to visa press? Okay, so it was interesting with oxytocin because we didn't, you know, and again, I was skeptical that we would see these big group differences, but you know, it was a little bit of like, okay, you know, what everyone's saying, this is not going to be the big solution, right? And so,

I actually came at it from the work that we did in monkeys. And so I think I mentioned previously at the beginning of the podcast that there were a lot of limitations that I saw. And then sometimes if you come into a field, you know, when you're, you're a little bit of an outsider, right? Like I'm not a clinician. I don't see autism patients, but I also, I have this really strong interest in social behavior and the biology of it. And so I was thinking about what are

What are things that we need to do to better address the challenges in autism? So one of them was, why are we looking in blood, right? Like if you look at neurological conditions, there has been a lot of progress made by doing biomarker discovery in cerebral spinal fluid, right? So like the biological substrates or clues of markers of say, various forms of dementia or

or MS were first found in spinal fluid, right? Because it's the fluid that bathes the brain in the spinal column. And so if you're looking for the biochemistry of an illness, that's the closest fluid that you can get to the brain, right? So a lot of draw just won't do it. Maybe, right? So that was part of my thinking. But then there was the issue of the animal models, right? So there was drug after drug after drug that was tested in mice and they failed in human clinical trials. And so

it made me start thinking, could we develop a primate model of naturally occurring social impairments, right? So can we, because in autism, these social impairments are, if you will, naturally occurring, right? And so, you know, this is these spontaneously occurring children. And so it made me wonder, could we identify monkeys in a large colony that have social impairments in after talking to

to clinicians who treat these children, can I spend a lot of time validating a monkey model where there will be monkeys that have features that look like they have direct relevance to core autism symptoms? And so what I did was there's a primary center, the California National Primary Research Center, and so what we did is, so I think I mentioned earlier that there's these

surveys that can be used to look at autistic traits in the general human population, right? And so we refined one of these and we did what we call back translates. So basically it's an instrument that's used for humans and then what we did is modified it to be able to

to use this rating scale in rhesus macaques, which are an old-world monkey, and I know you're familiar with them. And I was interested in looking at old-world monkeys because they're some of the closest relatives to human that are used in biomedical research. And as I mentioned previously, these autistic traits are continuously distributed across the general human population, and that this genetic

say, let's call it genetic liability, which is a fancy way of just saying that we think that there's a genetic risk that underlies this continuum of behavioral traits, right? So if we think that that's true in humans and in one of our closest relatives, and we think that

some of these genes create proteins that then are what sets up the developing brain to develop in the way that autistic brains develop. So let's just assume that that's the premise. That's what we went in with. Can we find rhesus macaques that are just living in large outdoor colonies and identify animals that might be good models for autism? And the answer is yes. We could do this all kinds of different ways. One is we could just take people and

score monkey behaviors outside their cages while they're interacting with their peers. We can use rating scales. And again, the rating scale we use, it's called the social responsiveness scale. So this is called the MCAC social responsiveness scale revised. It's a mouthful. But what it allows us to do is measure autistic-like traits in monkeys. And we can also bring monkeys in for experimental tests to see where their eyes look or how do they perform

How do they respond to videos of other monkeys, you know, if they're making affiliate of overtures, do they do like, you know, you know, McCax global, which is a positive response? Well, they do that, right? I'm going to apologize for interrupting again, but I just had to tell people this because I spent time up at the UC Davis Primate Center as a graduate student. And by the way, what we're referring to here are non-invasive observational studies, at least thus far.

So these are monkeys living in large, uh, exclosures, not enclosures, large exclosures, um, forming colonies and social relationships. And, um, you know, I think anyone that sees monkeys at the zoo and we all learn that monkeys go, and they don't eat. If you want a monkey to like you, you learn this working with macaques. Um, first of all, they don't eat the affiliate of call is a hoo.

They do this really not in the little ones. You do have very well. I spent a lot of time with these monkeys and the little ones, they do this thing where they go, I used to nurse the little ones every once in a while. They go, and they're just like, you know, it just like makes your heart melt. I think there must have been an oxytocin dump at that moment. That's probably happening right now. But if you want the monkeys to like you, you have to give an affiliative facial gesture, which is not a smile. That's actually an aggressive gesture. So as Karen, Dr. Parker just showed you, it's lip smacking, which is

Yeah. So if you see a monkey at the zoo and you want it to pay attention to you, you're going to have to lip smack. And if it doesn't, either you're not doing it right or it just doesn't like you. Exactly. Great. All right. Thanks. Now we'll go back to the study of or the establishment of this really key experiment. Right.

So then what we did is we identified these animals and we spent a lot of time. So one of the things that I do as one of my areas of expertise is validating animal models. So a lot of, like I mentioned, like a lot of reason why experiments fail is people will take an animal off the shelf and say, oh, I'm going to do this, right? But if you're studying a disorder that's characterized by

visual issues, is it the best thing to do in a nocturnal species that has old-faction as its primary sensory modality? Or is it better? And again, I will say all models have value. There's reasons you just have to

you know, you basically have to stand by what you're modeling. And so I think one of my the biggest issues I have with the sort of mouse phenotyping mafia is that, you know, there's this group of tests that they use and they use it in every single disorder, right? And then if there's a positive hit, it's like, oh, this is like, you know, this test is really for Parkinson's today, but it's for depression tomorrow, right? And so

So, my goal was to devise very specific tests that would allow us to evaluate core features of autism in this model. And the answer is we found it, right? So, if you look at monkeys that spend a lot of time alone, they have a much greater burden of autistic-like traits measuring on this rating scale.

diminished social motivation. So other monkeys will come up and interact with them, but they don't engage in social overtures them that much themselves. They do less grooming, less affiliative behaviors. They, in some of the work that we're doing, they don't lip smack back, and we can talk a little bit about that. We did a pharmacological probe, and we can talk a bit about what vasopressin does to that, which is kind of exciting.

And so we spent a lot of time validating this behavioral phenotype, right, to say that we really feel like there are core aspects of it that are allowing us to model autism, right? And I have a paper which, if you want to put it in, it's all about creating this monkey model and the power of doing it and where it took us clinically.

We'll provide a link to that in the show note captions. I also just want to throw up my vote for the fact that you did this work because, again, I don't disparage mouse model work, but we've just seen over and over again the incredibly small fraction of mouse models that lead to valid therapeutics in humans. And there's just a lot of differences between

primate brains and rodent brains. And we have a very elaborate frontal cortex, a bunch of other circuitry that mice, if they have that, they probably use it for other things. And it's just very hard to draw conclusions from those models. And they're great for probing functions that are, let's just call them more autonomic type functions and for doing some of the initial investigations. But I think while I don't

want to see every research lab switch over to primates. I think one has to be really thoughtful about the kinds of experiments one does with primates at all. This sort of behavioral assessment and the identification of a primate model for autism seems like a very good use of human resources.

Right. Well, and the other thing I will say is that there were medications that were only tested in rodents that when they were tested in people had really negative consequences. I can give you two examples. So one is thalidomide, which was a morning sickness medication that was given to women that were pregnant. And the safety testing and toxicity testing was done only in mice. I didn't know that. Yes.

And that's why it went on the market. It went on the market in Europe. And there were all these children born with profound limb abnormalities. When they went back and tested the drug in marmosets, neither Reese's monkeys or cinemologous monkeys, an old world monkey, they had the limb abnormalities. And so all they had to do, and again, I as an animal lover treat the life of a single monkey or a single mouse for that matter, an individual monkey, excuse me, or individual mouse for that matter,

as critical. I am a species. I do think there's a difference between their life and our lives when it comes to what study one does. But just the idea that

these severe developmental defects in humans could have been avoided by doing an experiment, perhaps even on one marmoset. And again, I feel for the life of discomfort of that marmoset. But the idea that that could have saved so many human lives is just striking.

Well, and there was also that straight drug MPTP that was a synthetic heroin, right, that caused like overnight Parkinsonianism, right, when like, I think the dopamine cells were just ablated, right? But when you went and looked in mice, MPTP didn't have those effects. It was only in primates and other humans and other primates, right? So, and I agree with you, I am an animal lover. I think that we have to be very careful whenever we do any animal.

experiments, right? And so you really need to have a good justification. I think for any science that's done, I will say that upfront. And, you know, we have this, you know, new generation of stem cell and organoid work, which I think is going to, you know, allow us to make all kinds of disease progress, right? So without having to study whole animal models.

or in complementary, right? But I mean, I think, again, I think we need to pick the model based on the question we're asking, right? And so if you want to have a medication that's safe and well tolerated, you know, when people were effective and you want to move the needle on complex social cognition, you want to be testing it in a species that also has complex social cognition.

Look, the Netflix show Chimp Empire. People haven't seen it. They should watch it. When you watch it, you realize they're very much like us. And dare I say, we're very much like them. Oh, yeah. It's far and away different than watching a bunch of mice.

Yes, and I'm not being disparaging of mice. I'm assuming they have that mice also have complex social cognition roles also have complex social cognition But it's of the mouse vol type and we don't know really even what to look for Right, but with primates. There's you know, affiliate of gays. There's you know, affiliate of grooming. There's ostracization of individuals in a troop I mean, there's there's a you know, banding it taking care of other babies There's all sorts of interesting dynamics that map so clearly on the human behavior and vice versa

Yeah. Yeah. So you establish this colony up at Davis at the regional and primate center where you identified some monkeys that we don't know if they have autism, but you could see that they were less socially affiliative.

Right, and I would never say they have autism. Like I will say that upfront, you know, they have features that resemble human autism and that allow us to model this, right? So we started studying those animals. And what we wanted to do was do some biomarker discovery. So what we wanted to ask was, are there any molecules that allow us to differentiate these, what we'll call them naturally low social or low social monkeys from socially competent, high social monkeys?

And so we measured a bunch of different readouts of neurotransmitter systems that were either involved in mammalian social behavior, had been implicated in idiopathic, meaning autism that doesn't have a genetic cause, or these neuro-genetic syndromes that we've been talking about, where there's pathways that are really associated with them. And so if we measured a bunch of these systems with 93% accuracy, without even knowing who the monkey was if they were lower high social,

we could just put them in the low social or high social bucket. And was this by blood draw or cerebral spinal fluid? So this was, it was everything. We did blood, we did CSF, and we put all these measures into the hopper. We did a discriminant statistical analysis, which was like a machine learning algorithm where we just said, here's all this information, help me classify if this individual is high or low social.

Cerebral spinal fluid is collected by spinal tap, correct? And my understanding, I've never had one, but that spinal tap is, of course, more invasive than a blood draw, but it still is done as an outpatient thing in humans. Like you can go in, get a needle inserted into the lower spine by an expert. They're going to draw cerebral spinal fluid. I mean, not that much more invasive and time consuming than getting a

a needle into your vein for a blood draw, right? I mean, we think of it as technically a little bit more challenging.

They're CSF draws in humans all the time. So in theory, this could map to a human study. And it did, which we'll talk about. Very cool. So we went out and we did this. I have a spectacular statistician who's, we spent a lot of time together. His name's Joe Garner. And he is a statistical genius. And so he developed this and we do all of our work together, or I would say 95% of it. We just love working together.

And he developed a statistical winnowing strategy to identify what were the key drivers and what was fascinating is in this first monkey cohort, it was the cerebral spinal fluid levels of vasopressin that were really what was driving this classification, right? So if we just knew your levels of your vasopressin in spinal fluid, but not in blood, interestingly, we could

pretty closely perfect to perfect classify you as high or low social. And so then we replicated that again in another monkey cohort because obviously as a scientist you always want to replicate your work. And then if it was really a biomarker, meaning it's a molecule in the body that gives us an indication of something. And in this case it's an indication of your social functioning.

We were able to look at monkeys and we saw that the vasopressin was consistent across measurement times. So there was a wide variety of vasopressin levels, but within an individual monkey, it was pretty much the same, right? So that's what you want to see with the biomarker. And then we showed that the vasopressin levels were closely linked to

group spent grooming, time spent in grooming. And as we mentioned, I think we mentioned earlier, grooming is in many monkey species, a critical behavior that solidifies social bonds and maintains them. And so the individuals with the lowest CSF phase or press and levels had spent the least amount of time in grooming.

grooming other monkeys. Yeah, this allopathic grooming is a very interesting behavior. And from watching chimp empire, I can tell you that new relationships are established in many ways by monkeys, these chimps, chimpanzees, sort of offering their back for grooming and if another

chimp elects to yes groom that chimp, then it establishes a some form of trust. And it all seems to have to do with proximity, like how close are you going to let me get to you vice versa. And humans, we talk about personal space and there's a whole set of things related to consent in this whole allopathic grooming thing. And then

If a chimp misbehaves on an outing, then they aren't groomed by others, and they can actually get parasitic infections, and it can be very costly. It's very interesting to just think of allopathic grooming as not a primitive of language, but a whole language into itself.

Absolutely. Yeah, and also just critical for the species. So that was really interesting to me that we were seeing these hints that vasopressin could be, you know, really important. But of course, you know, somebody will say, and I will say upfront, monkeys don't have autism, right? So then the question becomes,

does this have what's called translational value? So can I see this observation in animal model and will it provide fundamental insights into humans, right? And so I wanted to get cerebral spinal fluid from people to test this hypothesis because we had in parallel done a study looking at blood vasopressin levels and people within without autism. And we didn't see a group difference there.

unlike this really profound difference that we saw when we looked at spinal fluid in the monkeys. And again, I think I mentioned the blood vasopressin levels were indistinguishable if you were high or low social monkeys. So there was something about looking more proximate to the brain that was giving us more information than, say, the blood alone.

And so I said, I wanted to get spinal fluid. And like you said, people do this all the time. How would we? But it's not going to be a first pass, especially when we don't really have any evidence in people to go in for what we would call a research lumbar puncture. And so I had to get really creative about how do I get spinal fluid from children? And what we did was we piggybacked onto a clinical indication for a spinal fluid draw.

And we did this. So I tried to get funding for this. This is like, again, I mean, I think this is important for people to know how science is done. And so I wrote all these grant applications. Nobody would fund it. They said this is really interesting. It's too high risk. You won't be able to pull it off. And I don't usually back down from a challenge. If I think something's a good idea and I want to do it, I'm going to find a way to do it. If it's impossible, that's one thing. But if it's hard to do, it doesn't mean you shouldn't do it. You just have to figure out how to do it. And so I always try to see

bridges where other people see barriers, right? And so it's like, well, how can I access spinal fluid? And so I went around talking to all my friends who were on, and Stanford's really wonderful because it's such a small school, right? And so you're on all these different committees with all these different people. And so a lot of committees. Lots of committees. Exactly. But it's really cool because you're on them with people from all different departments.

I know people in departments that I wouldn't otherwise know. And you get to know these people well in these many committees. And where we live, it's a small community, right? Maybe we're the experiment, Karen. Maybe there's a, I always wonder whether or not there's a larger experiment, like not on monkeys, not on the, the patients or the clone job, but like we're maybe worthy experiments, right? And they're looking at how we interact on committees. Anyway, please continue.

So I started going up to people that I knew and said, hey, if you're taking spinal fluid, can I get a little bit of extra? And of course, we got IRB approval, meaning we had ethics approval and all this. Or you could get the remnant sample and obviously, again, get consent from the families. So we could either get a little bit extra when it was being drawn for research indications. So they were getting a spinal tap, no matter what. And then we were just either we're getting a little bit extra, or we were going to get the remnant that they were going to throw out.

So you usually take more than you need because you don't want to have to do another spinal tap, right? And so we were able to go around and I hustled around and got all these people involved to help me. We put hot pink stickers on the lumbar puncture trays so that in the emergency room, so if somebody was doing a spinal tap, they would call us. So we knew about it and we could get samples again under people's consent.

So we got all these people involved and we finally got samples from children with autism and children without autism. And then we also made sure that whatever they were being worked up for was negative, right? So we got the sort of healthiest people we could given that everybody was coming in for a medical reason to have a lumbar puncture.

And in this first study, we had seven children with autism, seven children without autism, and we could nearly perfectly classify 13 out of 14 individuals by just knowing their CSF vasopressin level alone, which is pretty remarkable given that there isn't a biological indicator, a robust biological indicator that we know.

So basically in this relatively small cohort, yeah, having low vasopressants. Correct. It's a biomarker of autism. Correct. And again, and what I will say is in our monkey studies and in our human studies, CSF oxytocin level became our control, right? So in our monkeys, there were no difference in CSF oxytocin by group.

And then in this first study, there were no differences in CSF oxytocin levels. A sample size of 14 is intriguing, but given autism so clinically heterogeneous, we want to replicate it. And so I knew that there was a

a professor at the NIH named Sue Sweden who was collecting cerebral spinal fluid as part of a research study because she was interested in immune parameters and folate deficiency. So she had children that were medically healthy and they were getting, you know, just like at NIH, get these huge workups, right? So they were very well characterized participants. So we were able to look at, and again, we also, this is the first time we were able to look at girls. So we had a small sample of girls

And we had boys and we basically just asked the question, can we replicate this? And I was very interested in, well, will oxytocin be what's different in the girls, right? So maybe there will be some sex specificity here and it will see low CSF vasopressin in the males and low CSF oxytocin girls. That was not the case. What we found was that if

in the individuals with autism, regardless of their biological sex, that they all had lower CSF vasopressin levels than the individuals without autism. And because they were so well characterized, we were also able to show on a gold standard research diagnostic assessment of autism. So it's an assessment that's used

in a research situation to validate an autism diagnosis by an expert clinical opinion, that the lower your vasopressin levels in spinal fluid, the greater your social symptom severity, your clinical symptom severity. And then we asked, it's like, well, vasopressin's involved in social behavior, but it's not really that involved in restricted repetitive behaviors. And that was actually the case.

It was the CSF visa press and track the social symptom severity, not the repetitive symptom severity, suggesting that there might be other biological measures that could be included as a way to, you know, have a more powerful way to differentiate people with and without autism.

And so then I was really, so that was really exciting to replicate that. And then I had a colleague named John Constantino, who is now at Emory, but he used to be at WashU. And I knew that John, I had been at a meeting in, I think it was 2010. And I found out that he had what I will call liquid gold. So he had this minus 80 sea freezer that had a bunch of neonatal infant CSF samples that he had.

from human infants. And he had collected them. And again, this was under ethical approvals. And it was basically these infants came in for something that needed to be worked up that was very rare. But if they had it, they could die, so they needed to get a medical treatment for it.

but the vast majority of these children ended up being healthy. So it was a pretty healthy sample, if you will, right? And so I knew he had all these samples and I said to him, wouldn't it be really interesting if we teamed up and we look at this CSFA's repressant finding in children before the period when behavioral symptoms first manifest, right? And so- Yeah, so sorry again to-

I think it's important because this was a question that I was thinking about earlier and I imagine many other people were too. You find these monkeys that have social interaction deficits. You find kids that have social interaction deficits and you see that there's low vasopressin in both groups. This extends to male and female children.

But then of course, the question becomes, well, maybe they have low vasopressin because of so many years or even months of social interaction deficits, right? That the direction of causality isn't clear. And so when you said liquid gold, you know, referring to the CSF from these infants, um, taken prior to any opportunity for social interaction beyond just, you know, whatever interaction they had with their, their mother up until the point the, the CSF draw was taken. Um, this really gets at the issue of causality.

Right, so it's a quasi-perspective, you know, because it was banged and then a lot of time went by, right? And so what we realized we could do was, and this was a heroic undertaking on John's part. So these were, these samples were collected back on paper medical records. So he had to trace 2,000 paper, yeah, exactly.

So he had to trace 2000, I think, paper medical records to an electronic medical record. And then what we did is we, he looked to see who went on to develop autism and who didn't, right? So then what we had was spinal fluid samples that have sort of been waiting in the freezer, if you will. And then we could ask, you know, do individuals who later receive an autism diagnosis many months or even years later already have low vasopressin levels as infants?

And the reason why this was a compelling question to ask is there's evidence to suggest that behavioral therapies are more effective the younger the child is, right? And if you think about it, if behavioral characteristics of autism emerge across development, you know, what if, and this was my, this is sort of my, we haven't, we haven't substantiated this yet, but this is like sort of my big question. What if all these autism susceptibility genes, some in interact,

and converge upon a few common pathways in the brain. And so for years, people have talked about this excitatory inhibitory balance theory of autism. But what if vasopressin is one of those pathways because it's so critically involved in social functioning? And so what I was interested in, and so let's just say for a moment, your genes are set at birth,

What if the vasopressin is already low in the brains of these infants? And so it puts them on this very different trajectory where you have this cumulative effect of there may be a little bit less socially interested and maybe they're not making the eye contact. And if there was a way to intervene really early, even potentially with a vasopressin replacement therapy, that you might be able to put them on a different developmental trajectory. So that was my big,

what if question? And what was really remarkable was, so I had been asking John, hey, can I have your spinal fluid samples? And he finally agreed after he saw a couple of those papers, understandably, he wanted to make sure that we already had shown something in people and animals that were sort of, if you will, symptomatic with social impairment.

And what we found was, yes, this was the case. So it was a small sample, it needs to be replicated, but individual is so infants that went on to have an autism diagnosis later in life already had low CSF vasopressin levels. Oxytocin levels did not differ between infants.

that received a subsequent autism diagnosis and those that didn't. Suggesting that we have a biomarker that might really be a good readout for clinical referral or risk management monitoring. Incredible. You're telling us that levels of vasopressin correlate with social cognition deficits. I think that warrants a brief discussion about cerebral spinal fluid.

I teach neuroanatomy to medical students. So forgive me for having to ask this. But I think of cerebral spinal fluid as the stuff that exists in the ventricles and down the central canal of the spinal cord and provides essential nutrients and for neurons and other cell types in the brain.

It's also a reservoir for chemicals coming from the brain, which is why the spinal tap is useful. But in the context of a cerebral spinal tap,

And you're measuring CSF and you're seeing, okay, lower levels of vasopressin in these individuals with these challenges with social deficits. Does that mean that they're making less vasopressin? Does it mean, I mean, it could have gone the other way too. Like they're dumping too much vasopressin into the CSF and it's not able to function in the brain. Like, you know, what do we know about CSF and what does it mean?

Right, well, I mean, it's a great question. So I think this is just the tip of the iceberg, right? So I think of the CSF as sort of like the kitchen sink of the brain, right? And what we need is real specificity. And so, I mean, my working hypothesis, and we'll talk a little bit about pharmacology, is that there's a deficiency in vasopressin production and individuals with autism, but there's a lot of elegant experiments that need to be done

to be able to answer this question. So we have funding currently to look in postmortem human brain tissue to look at in both blood, CSF, and hypothalamic tissue where vasopressin is made.

to look at inner relationships, right, which is very difficult to do, but also to see if there's a fewer number of vasopressin-producing cells and if vasopressin gene expression is diminished, right, because that would help us begin to answer, is this a production issue, right? So if you think back to the prairie voles,

They're sort of primed to be parental, right? Or in my case, the meadow voles, right? But you can do this in any volle species or at least the two that I'm thinking of. And you put vasopressin into the brain and then all of a sudden it unlocks this behavior, right? So is it possible that children with autism or at least a subset of them? All you have to do is replace vasopressin and that there might be a subset of these kids minimally that could benefit from vasopressin replacement, if you will.

Is there any evidence for excessive urination in kids with autism? Which, if anyone's going, what, why is he asking that? If you recall, vasopressin is also anti-diuretic hormone. I suppose the other question is, could you, has anyone looked at