– Well, good evening. My name is Eric Barker, I am Dean of the Purdue
College of Pharmacy, and it’s a pleasure to
welcome you this evening to what is now becoming sort
of a tradition around here, but we’re beginning to wrap up. This is one of the final
Ideas Festival events as we are wrapping up the 150 years of Giant Leap Celebration
here at Purdue University. Tonight we have the pleasure of welcoming Dr. Guoping Feng to campus. Dr. Feng is the Poitras
Professor of Neuroscience in the McGovern Institute
for Brain Research, at the Department of Brain
and Cognitive Sciences at the Massachusetts
Institute of Technology. He is also an institute member of the Broad Institute at MIT and Harvard, and the Director of Model
Systems in Neurobiology at the Stanley Center
for Psychiatric Research at the Broad Institute. Dr. Feng studied medicine
in Hangzhou, China, and then completed his PhD training at the State University
of New York at Buffalo, and post doctoral training at Washington University in St. Louis. Prior to joining the faculty at MIT, he was a faculty member in
the Department of Neurobiology at the Duke University School of Medicine. Dr. Feng’s research is
devoted to understanding the mechanisms regulating
the development and function of synapses in the brain,
and how synaptic dysfunction may contribute to psychiatric disorders. Using genetically
engineered animal models, Dr. Feng’s laboratory combines
cutting edge technologies and multi disciplinary approaches to unravel the neuro-biological mechanisms of neuro-developmental
and psychiatric disorders. Dr. Feng will be making
a brief presentation, and then I will join him on stage for a brief conversation and dialog, and then we will wrap up
the hour by inviting you, our audience, to provide some questions. His Ideas Festival question and theme of his presentation is, “What if breakthrough technologies
could make us smarter?” Please join me in welcoming
Dr. Guoping Feng to Purdue. (audience claps) – Thank you Dr. Barker for
the wonderful introduction, and it’s a great honor to be
part of the 150 celebration for the wonderful history
of Purdue University. I’m very glad to be here. I toured the campus and
had a wonderful time meeting with the many
extremely talented students and faculty members, so
the day so far is great, as long as I don’t
screw up this last part, it will be a wonderful visit. (audience laughs) So, I’m trying to do my best. So, today I want to, instead
of presenting my own work, I want to discuss with you
some of the new technology that has been developed, and how this new technology
may affect our daily life, for both us, and maybe for
benefiting of a lot of patients. So, you probably know, I’m focusing on how can we
change our cognitive ability, which is our intelligence, and you probably know intelligence, just like every other human trait, whether it’s behavior or actions, it’s a bell curve, right? So, most of us are in the middle, but there are a few
extremely talented ones, like you, on the one side, and many of us always wish
were smart as Einstein. Right? But there are also on the
other side of the curve, which, some of them just
have normal function, but the other side of the
curve maybe you need help that they can have a normal
life just like we have. How do we potentially
can move this bell curve all squished to this side? Right? If they say, “If it is a way, can you
improve yourself or your kids of their intelligence, will you do it?” Not there’s a possibility we can do that. So and then the questions, “Who should do it?” “Who should decide?” and “Who should pay for it?” All of these are questions
I want to discuss with you briefly and tell you the technology, why we think we can do this now. So, IQ actually is very closely
linked to success, right? There are a lot of studies. This is just a table I borrowed
from Kaufmann and show that, what kind of profession
you do is highly linked to not only your child, your adult IQ, but also linked to the
IQ when you were a child. So, this depends on what kind of IQ almost determines what kind of
successful career you can do. So the question, what determines IQ? What determines our intelligence? So how do you decide that? So probably the best evidence
comes from studying twins. You have identical twins which you have almost identical genetic material. So identical genetics, not even almost, it’s identical genetic DNA material. Then you can be raised in the same family or can be adopted into different family or even different continent. So then you can study is
their IQ same or different? So, this can determine
is the IQ determined by genetic material more or
environmental factors more. So this is kind of more of
the generalized results. So this, on the left side, Y
axis you can see similarity of our intelligence scores,
the IQ, their correlations, and these are the raising conditions, all of them are from identical twins. What you can see is
that if identical twins are reared together in the same family, you see they have very, very similar IQs. Almost 90% identical, so it’s correlation. But, if they are separated
into different families or different countries,
you can see that they are actually also very close but not as close as the identical twins but it’s still above 70%. That means that the
majority of intelligence are determined actually
by genetic material and environment does play some role, but much smaller than
the genetic material. So you can see whether
they are identical twins, or non-identical twins and
the siblings and unrelated. So you can see that there’s
a direct correlation between the genetic factors
and the IQ, the intelligence. So, how do genetics
determine the intelligence? Based on twin studies,
heritability is around 60-80% and that means the genetics
play a significant role but it’s not all of it. Because it plays a significant
role there’s a chance that if you change the genetic material you can improve impaired intelligence. So in disease conditions such
as neurodevelopmental disorder I’m gonna talk about is that
you have genetic mutations that lead to intellectual disability, but there’s also a chance
that you could change the genetic material and make
humans much more intelligent. So, how do genetics
determine intelligence? Luckily, or unluckily, in general, it’s not determined by one gene. So, how do you study that? Basically, we’re called GWAS, Genome-Wide Association Studies. You can study hundreds of thousands of people with high IQ, compared with hundreds of thousands of people with low IQ, to see what are the differences. So this is called GWAS studies. What it shows is, almost like
many other common traits, like height. Height is very heritable. If your parents are tall,
the chances you are tall is very, very high. But it is not determined by one gene. It’s determined by multiple
genes, we call them polygenic. So there are GWAS studies for IQs, and at least one study showed
there are over 500 genes, actually involved or related,
or correlated with your IQ. So, that means that your intelligence is mostly genetically determined, but not determined by a single gene. Which, is probably a good thing, right? So, then, each gene only
plays a very small role. So that makes it challenging for how we change genetic material,
to make humans smarter. On the other hand, single gene mutations can cause very severe
intellectual disability. We have so many, many of them,
neurodevelopmental disorders with very low IQ. One of them, I use the
example, average IQ is only 40. So, they can never live independently, we need to find a way to
correct these problems so that these people
can live independently and have a normal life. Because of the new technology now, we could potentially develop,
because it’s a single gene, like, we can correct this gene mutation and actually dramatically
improve or even cure this intellectual disability and really change people’s life. Because of this genetic engineering, it could be basically
part of daily life now, whether it’s treating patients,
or improving intelligence. So, what is that new technology? The new technology you probably heard, at least some of you already heard, is called CRISPR technology. Basically it’s the new type
of genome editing technology. The idea is that this
is basically a system used by bacteria to destroy viral DNA. Bacteriophage, which is
a virus to kill bacteria, and whenever they inject
the DNA into the bacteria, bacteria will pick some of the DNA, into their chromosome, so next time, if it survives, next
time you come, I know, I’m gonna destroy you. So these are the systems that
we now have been developed to use to manipulate mammalian genes. This is called CRISPR Cas9. The Cas9 is a nucleus, it can cut the DNA, so, because the virus DNA
coming and incorporated into the bacteria and the
bacteria actually remembered. So, next time, if the virus comes again, the bacteria will release a piece of RNA, use that as a guide RNA, guide the Cas9, the nucleus, to the viral DNA and cut it. So, that will destroy the virus. Now scientists, many of
them are listed here, Charpentier Lab, Feng Zhang Lab at MIT, and the Doudna Lab at Berkeley,
George Church Lab, Harvard, these actually changed the system, now we can use to manipulate
almost any cell type, in any species, almost. So the idea is you use the system, you design whatever you
want, it’s a guide RNA, as long as I match the gene,
it will bind it to the DNA, and you can cut the DNA. So, now we have all the genome sequences, so we can specifically design a guide RNA and bind to whatever gene
you want to manipulate. It will cut the DNA. Now, that a gene is cut,
your cell will try to fix it. So, that’s how we get UV irradiated. You go to the beach, your DNA’s probably damaged in some cells, but you don’t get cancer because your cells always try to repair it, and most of the time, repair perfectly. However, occasionally they make mistakes. Once they make mistakes, you have cancer, or you have other problems, or may die. So, in this case, basically,
you will have a mutation, that’s how you can generate
genetic mutation in any species. On the other hand if you say, okay, I wanna change this gene, because I know if I
change this amino acid, we’ll correct the mutation for humans, and then you can put a template, we call it precise integration, you can replace a piece of
DNA with whatever you provide. So, this made it possible to do anything. You can generate mutation,
you can repair mutation. So, if we know, does the patient have
a single gene mutation? Let’s say, Rett Syndrome. Rett Syndrome affecting
many millions of girls, and they are born normal,
then after two years later, they start deteriorating. Many of them will die at a very young age. We know these are a single
gene mutation called MECP2. We now can, potentially, go
in, correct the mutation, they will be, probably,
live a normal life. So you can do both make genetic models, or you can repair human mutations. So, this technology is unbelievably
powerful, and efficient. It can be multi-purpose. I’m now going to go into a lot of details, I imagine you can make mutations or you can repair mutations
we call knockings. But, the most important,
it’s highly efficient, you can use almost any species. You can use animals, plants, even humans. Many of them may have multiple, let’s say, intelligence is determined by 500 genes, but maybe there are five
genes that are very important and if you change them, you can slightly move the bell curve. So, this system can multiplex, we can do five genes at the same time. Now, it’s possible. These have been now, it’s
less than six years old, actually but it has been widely used. Some of the clinical
trials are being studied to try to correct some mutations. I’ll give you a couple examples. One example is, we think
now genetic engineering pigs as organ donors, right? So, there’s a lot of patients
waiting for a long time for kidney transplants. Now the idea is that you can actually modify the DNA in the pig
that will have a match and will get rid of all
the retroviral fragments in the pig DNA which is
detrimental to humans, and you correct it, then you basically, every time you need, you can
harvest a kidney from a pig and transplant to the patient. My joke is, we can do any of these things, so in the future, when you’re getting old, your organs can be replaced
one by one, by pigs, you’ll work like, everything
in your body is a pig, except your head is still human, right? (audience laughs) But they are functional. No one can tell. The other really important implication, which is even more direct,
even more impactful, is for gene therapy. There are many, many severe
neurodevelopmental disorders, many of them you probably know, and you probably encountered
or you have relatives. These are patients that have very severe neurodevelopmental disorders, they have very low IQ. So, they cannot live independently. These are very big problems,
not only for the patient, but also for the family. My main research goal
is to try to find a way to help with these patients, because they really need help
to be living independently. Now, these are many, many examples, including Rett Syndrome, I mentioned, Fragile X mental retardation,
Phelan-McDermid Syndrome, which is also very severe,
their average IQ is only 40. I’m gonna use Phelan-McDermid
Syndrome tonight to show what can we do potentially in the future to help with these patients. Their general features, all these have very severe
intellectual disability, then they have many
other problems, including Autism spectrum disorders,
seizure, sleep problems. For these severe ones, actually, genetic mutations play a key role. Actually now, we already
know, at least 25% of them are caused by mutation in a single gene. For single gene, we now can model them, and we now can potentially
go correct them, with this new technology,
or with similar technology. Because a single gene, we
call it a monogenic mutation, this is ideal for gene therapy. In the next five, 10 years,
you will see a lot of new gene therapy approaches try
to really cure this disease. I wanna mention, although
these are monogenic, which is a single gene mutation, their pathology is yet
generally not a single gene, ’cause each gene can
affect many other genes. So, it’s almost impossible to find a drug that can correct all different pathways, all different processes that it disrupted. So, gene therapy’s probably the only way to really cure them. The other approach is
circuit manipulations, or pathway interference,
these probably can correct part of the pathology but not all of them. I wanna use Phelan-McDermid
Syndrome as example. This is a mutation in
a gene called Shank3. Shank3 is critical for build
neuron-neuron interactions we call synapses. A neuron is very different,
the brain is very different from the rest of bodies. We have 80 billion neurons in the brain, we have trillions of synapses
that connect to each other. Neurons do not function autonomously, they have to interact with
other neurons to make a circuit. Everything you do, every thought you have is millions and millions
of neurons involved to form a computational
process, and have a result, and have output, so even lifting my hand is a very complex neural circuits problem. These have very low IQ and they’re really very intellectually disabled,
so you need a lot of help. We want to see, how can we study them, how can we help with them in the long run. These patients have impaired behaviors, because they are Autism Spectrum Disorder, they have social interaction problems, they have sensory problems. Sensory overload is
one of the major issues of Autism Spectrum Disorder. We made a mouse model to start with, you can see the repetitive
grooming themselves, and they also have social
interaction problems. So, if I can have this video played? So you can see, these
are, on the left side is two wildtype mice, on the
right are two mutant mice. For wildtype mice, their social is whenever there are two mice together they basically sniff other
mice, mouse’s behind, that’s their social interaction. It’s very different from humans
but it’s a social behavior. For mutants, you will see,
they even bump into each other, they will completely ignore, right? So, now they stay very far away. You see wildtype, they are
social with each other, they’re cuddling together there. I’m not sure this is really too human, but at least it’s a very
interesting biological phenomena. What caused this and why
a single gene mutation can cause mice to behave so differently? Then there’s another phenomena we think is also very interesting. So, these are two wildtype on the right, and the mutant on the left. We just in their home cage,
we put a plastic ball, in the middle, right, see what happens. This is a thing, they never saw it before, and so, you can see,
wildtype is very interested, will explore pretty soon,
they will go on top of it. Mutant, you can see, when
their head is getting close, they startle back, so that means
their whisker’s touched it. They are very sensitive,
they’re afraid of this thing, they never will play with it. So this reminds us that, this
is called sensory overload, oversensitivity, right? A lot of neurodevelopmental
disorder patients including Autism Spectrum patients, they are very sensitive to
light, noise, all these things, so that’s why one of them
will stay on the side, alone. This is a good model, probably, to study. Now, our studies show they
actually have hyperactivity in their sensory cortex. Whenever they receive information, their neurons are hyperactive, so, you see, the wildtype is on top of it, they are never even getting close to them. So, these models help us to understand how these genetic mutations
generate that kind of behavior. Then the question is, although this behavior
you can see in an adult, these are all
neurodevelopmental disorders. If we develop a gene cell, or treatment, can we actually reverse it later on? There’s a reason we call it
neurodevelopmental disorder, because it’s a development defect. So, a lot things when you develop, afterwards you cannot reverse. We call it a critical window. So, I have an accent because
I didn’t grow up here. I came 29 years old,
so I’m already way past my critical period, so my English is not as good as anyone, or my
son, who you cannot tell if you don’t see him that he was my son. So, the question is, how do you tell? How can we test what at what age we can treat a patient in the
future is still effective? So, we designed genetic
ways to test in mice first. Basically, we let the
mouse grow, like, mutant. Then in any age, we can give the drug to
restore the gene function. Just like gene therapy, similarly, and what can you reverse,
and what you cannot reverse, so, we did that. What we found is in adults, when these mice are four months old, which is a complete adult, and
then you re-express the gene, can you reverse anything? Surprisingly, in some of
our brain regions, you can. So, the top figure is
an actual physiological look at a synaptic
function, they are restored, the bottom is the structural function. Not only functionally you can restore, but structurally, you
can make new connections. That is very exciting, right? So, you can make new
connections, that means whatever defect that you
have in this brain area this is the striatum,
which is very important for motor and repetitive
behavior, all these things, and you can restore them. Now, do they restore the behavior? Yes! So you can see, there I have repetitive, compulsive grooming the skin off, but after you turn it
on, the hair grew back. The only difference is the
hair grew back as white. This has something to
do with the drug we used for the gene expression. But, they also restored
the social behavior. In the middle is the
knockout, the wildtype, you put it in the mice, let them choose whether they want to
interact with the mice or interact with an
object, and you can see wildtype like interacting
with the mice in the heatmap, and mutants like the object, but once you restore the
gene expression adult, they now go back to like
interacting with other mice. So, that means there’s some
aspect that can be restored even in adults, that gave us hope. However, that’s not the whole story, there are actually things
you cannot reverse. A lot of things you cannot reverse. For example, a motor defect,
this is what we called open field, you just
let the mouse run around to measure their activity. You can see that the
wildtype on the top line, mutant, and the restored, no difference, so there are things that you cannot. What you cannot? Anxiety you cannot, a
motor defect, you cannot, also these sensory defects that we call novel object phobias, you cannot restore. So, there are many things, you cannot restore them in adults. So, now, what if I turn
the gene on earlier? Before they become adult. This is after three weeks of birth, and you can see many things
that we cannot restore before, now you can restore them. So, this tells us that
if the mice are humans, which they are not, we
can conclude right now that at least in mice, neuronal connection and function in the adult
brain has a certain plasticity. In a certain part of the
brain, they are plastic. Even if you have developmental defects, we can still restore them. In mice. But there are also critical
development windows, which people have studied for decades. We know there are languages, there are many visual critical windows. These critical development
windows are key. Many of them, once you
pass the critical window, we cannot restore anymore. Then these we really have to
treat as early as possible if we want a full restoration of function in neural development. It’s the same thing. If we wanna improve
intelligence we may also want to do it much earlier, right? Because, what is cognition? Cognition basically is your
computation power in your brain. So, we depend on connections. If you wire something wrong, unless you correct that
wiring in your computer, your computer’s not gonna work very well. So, it’s the same kind of thing. But, your connection
made during development we’ll find, after birth, usually, that’s the critical period
you still have plasticity, after you pass that, you
don’t have that plasticity. That’s why only certain parts
of the brain is plastic. The one I showed you is
plastic, is the striatum. Striatum is for habit
formation, motor activity. Even at my age I can
still pick up a bad habit. That’s why this is very plastic. But, there are a lot
of things, people say, oh, old people are very stubborn. Yes, our cortex is fixed, so we cannot change our views anymore, or, it’s much harder to
change our views anymore. So, there are different brain regions have different plasticity. The good news, if mice are humans, there is a post-natal
window that means that we don’t have to deal with embryonically, because we cannot even diagnose them. That means after birth,
we still have a window, called a critical window,
we can still diagnose them, figure out a way to treat
them, and then correct them. This is the hope that
this might be effective. This is all on mice, right? I’ve specifically written
here if mice are like humans, and unfortunately, they are not. So, the reason is, we have
done all these things in mice, and it worked really well. However, none of them have
been translated into humans. There are many clinical trials failed, almost all of them for CNS disorder. For example, one of the most famous ones is the Fragile X, mental retardation. Three companies all
failed in clinical trials. They published wonderful
papers, in Neurons, in all different journals. Pfizer also worked really
well in Huntington’s disease, and also failed in clinical trial. The most recently, just next
to our building in Cambridge, Biogen, right? This worked really well in mice for Alzheimer’s disease, failed. That failure, overnight, cost
Biogen $18 million dollars, they’re stock dropped,
still haven’t recovered. The joke right now in the field is it’s a great time to be a mouse because we can cure anything you have. (audience laughs) Almost everything, right, so. Our goal is not to
understand a mouse, actually, we use a mouse as a model
to try to help humans. So, what do we do? That leads us to think we
need additional models. Mice is always going
to be a wonderful model in genetic manipulation,
they are mammals, so, they’re are a lot of things are conserved. But we now realize there
are a lot of things that are not conserved, either. So, one of the biggest problems, probably, by the way, it’s not the
mouse’s fault, right, so it’s always scientists,
how we use the model properly. Every model is useful. Whether it’s C. Elegans, Drosophila, they’re all useful to us for understanding the human biology, right? Cell death was discovered in C. Elegans, it’s all (mumbles) of the humans, so, every model, but for
understanding cognitive function, higher brain function, maybe
we need additional models, not to replace other models,
but first to understand them. One of the biggest problems actually is the prefrontal cortex. During evolution, the most
expanded area of the brain is the prefrontal cortex. The prefrontal cortex is the main reason we make decisions, and we
make cognitive calculations, so, if you look at this, this
is humans, macaque monkeys, and marmoset monkeys, you can
see their prefrontal cortex are much much bigger than rodents, this is a rat, actually. Many of the gray areas,
these are critical areas, are evolved in the monkeys and the humans, but they barely exist in rodents. That probably makes a very big difference, because it’s not only a fact
that there’s this brain region but since they’re all circuits. Cortical, sub-cortical circuits. So, in effect, the whole brain function. This is probably one of the reasons why so many things work so well in mice actually they do not work in humans. The idea is, then, can
we use better models? Models, we should not say better because we don’t have proof they are better, but can we use a model which is more close to evolution in humans? The CRISPR technology
now allows to do this. So, we now can put a human
mutation into monkeys to see whether they model better some of the higher
function defect in humans. Right now, all over the
world, including Japan, China, and in the US, two major
monkey models have been used. One is the marmoset,
marmoset are very small, it’s about 350 grams, almost like a rat, and they are very fluffy
so they look bigger, and the other is a macaque monkey. Marmoset is a new order monkey, so it’s a little further than the humans, and macaque is and old world monkey so it’s closer to humans, they have brain structure
much closer to human. But, each has advantages
and disadvantages, right? Macaque monkeys live for 30 to 35 years. If you wanna study late
onset diseases like Huntingdon’s disease,
or Parkinson’s disease, Alzheimer’s disease, you don’t
wanna be a graduate student to work under, you will
be a long, long time. (audience laughs) You probably will quit. So, marmoset is much shorter,
so you’ve got a chance that someone make a model
you can probably study them. With many others we try to understand whether this model can
help us to understand Autism Spectrum Disorder, like social cognition, social behavior. We’re actually working with
a large group of scientists in China, they have, so far, all the macaque monkey knockout
genetic mutation papers are published from China, and
the marmoset are from Japan. These are two leading countries in this. So we work with them together to generate a Shank3 mutation. I told you Shanks will lead to a severe intellectual disability
and so we were lucky we had homozygous monkeys
and heterozygous monkeys; in humans they’re all heterozygous. We found that in these monkeys you can give them an activity watch. Just like we carry, we say, ah, how many hours did I sleep last night? So you can do the same thing. You give them two weeks and take them off to see the data, you can
see wildtype in the blue, they have day and night
activity are very obvious. Day, have very active, night
they have very little activity. So, they sleep really well. But in the mutant you see,
heterozygous, in the middle, the day activity reduced, because the patient also
has reduced motor activity, they have a motor problem, but night activity is
dramatically increased. So, this mimics the human condition. If we have tested drugs,
so now we are testing, we identify a target
and we’re testing drugs, if we can improve the sleeping problem, it may have a better chance
to translate into humans. So, that’s a kind of use, it can help us to understand
in the development of drugs. Then the most interesting
is social activity. I showed you mice social is
sniff other mouse behind. That’s not a normal
human social, right, so, but if you see these monkeys, so, these are two monkeys. For each monkey we have a
wild type called prone-monkey. They never saw each other before. Then for the test monkey, either
it’s a wildtype or mutant. The prone monkey has a green collar on it, you can tell which, we use the same monkey to probe every mutant or control monkey. So, you put them together
with dividing in the middle. They cannot see each other. For five minutes they get used
to the smell and everything. Then a scientists will
come take the divider off, then the monkeys can see each other. You will see how they interact. Could I have the video played, please? So you can see this is a
wildtype with wildtype. So you can see this, take the divider off, now suddenly the monkeys
can see each other. What you will see is that
they immediately engaged, but they never saw each other
before so they are cautious. So, see the green color there,
that’s the probe monkey. That is our control monkey for testing. You can see, they are engaged,
they follow each other, they are very engaged and
they’re back and forth. So, they will look at each other, and so, if one goes to the other side, it will go, and it will back off, so they are very engaged
as they look at each other. That’s a normal social
behavior for monkeys. Almost like us, you know, you’re not gonna hug someone you never see,
you’ve never met that one before, you usually shake hands and you talk, and you become friends. So, that’s what they’re doing. So, next thing to do is identical. Same green monkey, the collar monkey, but the other monkey is the mutant with the heterozygous mutation Shank3 which we find in the human patient. What you will find is, could I have the video played, please? So, repetitive behavior,
repetitive flipping, that’s a very common
repetitive behavior in monkeys, in humans it’s different
repetitive behavior, but you can see a repetitive behavior. Then you will see, this person
will release the barrier, and they can see each other. You can see this is the
probe monkey coming, this monkey goes to the other side. This is the mutant monkey. The monkeys are interested in something, but never looks at the other monkey. It’s just looking outside,
looking at something, right? Even getting very close to each other, it doesn’t look, it just
goes somewhere else. So, actually, after a while
this wildtype probe monkey got frustrated, actually,
’cause it seems this monkey, this monkey’s always
interested in something, but never really interested
in the other monkey. We think of this kind of social behavior, we can understand, what is
the brain circuit defect. See, one comes, and one goes. So, they never engage each other. This monkey’s not interested. This monkey, we know engages, we’ve tested with 10 different monkeys, so. You can count that there’s a
very significant difference between the social behavior and this social behavior is much closer to what we think is the human social behavior compared to the mouse
sniff other mouse behind. If we can correct this problem, maybe there is a better chance that this can be translated into humans; this can really help us. Then, because they are
structured so similar to humans, we can actually do a functional MRI to see are there biomarkers,
activity changes in the brain, we actually found a very
significant difference between wildtype and mutant monkeys. These are non-invasive, right, for humans we do the same thing. This can be translated into humans to see whether human patients also
have this kind of defect in their neural circuits
in differing brain regions, the thalamus, striatum,
sensory cortex, visual cortex. If they do, then this can use a biomarker. Did our treatment improve the
basic function of the brain? These are macaque monkeys,
now with their help, we now have a large
colony abroad and at MIT. The marmoset we worth with are all at MIT. Now we have Shank3 marmosets now, we just mimicked a human mutation, we only have one monkey so we have to wait for the second generation
before we can study them. These could help us test gene therapy, test drug treatments, we’re
very close to Eli Lilly. I heard they are gonna
shut down the neuroscience research program, actually,
I just heard today, and the reason is, there is no market, they are not interested. They did so many years and
working so well in mice, none of them work in the clinic, so they cannot afford to keep
doing these failed things. So, we’re hoping that these
kind of things will maybe generate new interest in investment into, because we have a lot of
neurodevelopment patients, Alzheimer’s disease patients,
Parkinson’s patients, we need help, so we hope
these kind of studies will help us to improve the
development of the treatment. These monkeys we have
automated behavior tracking. These are much faster. So, computerized,
machine-learning program, so we’re working with a
lot of AI people at MIT to develop this system
to automate tracking so we don’t have to ask MIT undergraduates to watch hour and hundreds
of hours of videos to score for us, and so, computers can do this kind of work now. Now we’re mostly testing
their gene therapy approaches and drug development for these models. Now the question is if we
can use these in monkeys can we use them in humans? Yes! People already showed
that the CRISPR technology not only can you correct
a mutation in a monkey, you can actually correct
mutation in humans. This was done by Oregon Primate Center, they actually have the
University and Primate Center, they actually used human embryos. You can correct human mutations. If you can correct human mutations, then we can also change
human genes in normal people, make them better. The question is, what can you change? We actually know if someone
wants to be, ask this, you wanna be big muscled? There are genes to do that. Should we do that? So, all these have
become society questions, and ethical questions. I want everyone to think about this. It’s not science fiction, it’s real. We need everyone involved to think about what should be done,
what should not be done. There are many questions involved, there were several
meetings discussing this. The technology, we still have problems, because they have an off-target effect. But, this could be fixed
in the next few years. What if we fix everything,
it’s perfect, the technology, now what are we gonna do. So, there are targeted,
then, who needs it? For genetic disease, actually there are very, very rare cases who need it. Why? Because we have perfect
IVF technology now. You can always test and embryo, and before implantation, select
the right one to implant. There are tens of thousands
of babies born normal. Why do we need to modify the gene? They’re only very, very rare conditions. If both parents are dominant, homozygous dominant or recessive, then it’s become a problem. But these are really, really, really rare. In that case, you need to correct them. Other cases, you could
always use the IVF to select the right embryo and implant
it with no problem at all. Then the question is can we
use it to enhance human traits? There are many, many traits
we want to enhance, right? So, you want your kid to play football. Wanna be more athletic. You want the kids to go to Harvard, you wanna be more high IQ, have a little bit of competitive edge. You actually don’t have to change a lot, a significant magnitude of IQ. Small changes can give
you a competitive edge. But the question is, who
has a right to determine? Because once you change the genomic DNA it will pass along to the
next generation forever. The major problem is
we actually don’t know if you change a gene,
what else will change? It could be a lot of things become worse, one thing become better, so,
how will we determine this? Without testing in humans
you may never know. But, let’s say everything’s perfect, we can increase the
intelligence, who should decide? Who can afford to do it? If it’s really expensive,
who can afford it? Then you have equality problems. The rich is getting better,
then what do you do? You cannot compete with them anymore. So, all these are society questions we should start to think about now. So, these are not science fiction, these are totally possible to do, and it’s actually not going
to be in the far future. It will be in the near future, at least for gene
therapy to help patients, to help intellectual disability patients, I see it within 10 years. Maybe five years it will
be in clinical trials, in 10 years, maybe there are a few of them that will be developed in clinical use. One of the best examples
is Alzheimer’s disease. Now we know there are many, many factors that can contribute to
Alzheimer’s disease, but one of the biggest
risk factors is ApoE. If your ApoE4, you’re actually… (beeping) Sorry I’ll just do my… I must stop in a minute. So, ApoE4, if you’re homozygous, you have a very high risk of
developing Alzheimer’s disease. If you have ApoE2, you have
very low risk, it’s protective. Should we all go change
the human race into ApoE2? Homozygous? We can do that. Then we can very much delay happening of Alzheimer’s disease. The largest study in the Netherlands shows 1,700 people over 100
years old, very clear mind, can do anything and only
one of them has ApoE4. All of them, all the rest don’t have it. That tells you how strong is this effect. Now we can change them. There are people actually
doing that in the lab to test can we, because it’s not a gene, it’s a small fragment change
between ApoE2 and ApoE4. These are real questions
we will encounter very soon in the society, determining what should and what should not be done, and I want everyone to think about this, and you should really get
involved in the future. MIT did a survey, MIT
Technology Review did a survey a few years ago, majority of us in the US think that if you wanna change the baby just for enhanced intelligence, they’re probably going too far, but, a slight majority think
that to reduce the risk of serious disease is okay. You can have your own opinions, I think society debate
should be starting right now, and should not be
determined by a few people or even a few scientists. I’m gonna stop here, I can
acknowledge the many people involved in the lab. The monkey work actually
is a large collaboration, and I’m particularly
grateful to all the people, especially our donors
who support our research. Thank you. (audience claps) – Well, lots to talk about. I suspect for many of the
folks in the audience, the opportunity to see for the first time genetically modified
monkeys was new for some. It’s a bit scary, and so, to your point, when science fiction talks
about genetically engineering the super human race,
what we’re talking about is within a short period of time; we’ll be able to do these things. So, from your perspective, what
are some of the first things that we as a culture, as a society, really as a world society
need to think about in terms of putting guard rails around the uses of these technologies? – That’s a really good
question, it’s a tough question, probably everyone has a different opinion. My own opinion is that I think that it’s really important to
develop this technology for treating severe disorders. I meet a lot of parents,
they bring their baby, two years old, three
years old, to my office, just try to say, okay, whatever you do, please help my baby, my
baby has Rett Syndrome, it’s deteriorating, right,
they have seizures all the day, and they cry all the time. I think these are the patients
that we really can do. But, there’s also dangers to
modify our genetic material. You are not talking about drugs, you are talking about forever
changing the human DNA. On one hand, it’s kind of
directed evolution, right, you make the mutation, you make it better, but better, we think it’s better, we actually don’t know it will be better, because every gene has multiple functions. None of the genes so far we completely understand its function. So, it’s very dangerous to go on to say oh, change the gene,
we’ll enhance the ability. We’ll enhance an ability we know, but it will probably cause a lot of other problems we don’t know. So, in my view, many of
us think it should be at least temporarily banned on
any human embryo manipulation until we really have a society debate. Because, it should not be
determined by a few people. Society should have the right to decide what is the best way to move forward. Technology still can be developed, but not applying to
humans, that’s my view, and many other scientists
already voiced a similar view, and I totally agree with them. – So, one of the things that you point out in the survey in the MIT article was while making babies smarter
was certainly controversial, the use of technology for
therapeutic treatment, though, on the other hand, was
a little more accepted, and so, as we think
about neurodevelopmental or neuropsychiatric
disorders, many of them are the result of a small
battery of changes in in a battery of genes, and, it’s difficult to study those disorders, and understand the
complicated genetic changes that occur in those disorders. How do you think CRISPR and maybe some of the other emerging
technologies can help us tackle those very challenging– – Yeah, so, this is a
very, very good question. I mentioned that even in the
neurodevelopment disorders, only 25% are identifiable
genetic mutations. Many of them, majority of them so far is considered polygenic. Each variant, actually not a mutation or considered a mutation, each variant only contributed very little. Maybe 100 of them together. So, that’s why the parents
are generally normal, but the bad luck, all the mutations, we all have mutations, right? So, passed on to a single kid, and then that kid has bad luck disrupted to a certain degree then you have a
neurodevelopmental disorder. So, there are multiple ways
right now to study them, actually what people have
done is using IPSL technology, because you take the skin
cell from the patient. They have the perfect
combination for the disease, you culture them, then you
can use CRISPR technology to repair or mimic some
of the conditions to say which ones play a key role. So, I don’t believe, even when there are hundreds of mutations, I don’t believe each one
contributed similarly. So, they are probably major contributors and minor contributors, but
you just need them all together to make a very severe case. There are now technologies
that can do multiplexing. So, let’s say, with this
IPSL derived neuron, we found that 50 genes are changed. Expression changes. Can we just regulate 10 of
them, and will they improve? The answer is probably yes, because, the whole concept of polygenic is you need everything together. If you take 10% away maybe
they aren’t just much improved because they don’t have everything to make the patient
clinically significant. It is possible you don’t
have to correct everything but you have to correct
a whole bunch of them. And, CRISPR technology
is completely capable of doing multiplex, even for hundreds. People have done hundreds
of them at the same time to regulate gene expression,
or correct a mutation. That is a feasibility in the future. Of course, the more you put to change, the more off-target you will have, so that problem we have not solved. So I’m not recommending
using it in humans at all at the moment. – I’m gonna ask one more question and then we’ll open it up to
questions from the audience. So, if you have a question out there, we do have microphones on
the sides of the stage, and I’ll invite you up in just a moment. I’m gonna ask, how much more potential do you think the human brain
has left to be explored? There is probably an urban
myth out there that we use just 10% of our brain, but do you think that we are indeed
underutilizing our brains, and if so, do you envision any technology to help us unleash our brain power? – Okay, that’s a really tough question. (audience laughs) So, my view is that for most of us, we are not using all the brain power, because what I think
is, if you think about the difference between our capability when it’s athletics or brains, right? So, if you ask me play football, I could be crushed by anybody, right? So, I’m, what, you are a thousand
times worse than athletes? It’s the same thing for brain power. So, I think the best comparison is whatever on the planet
that we have found that incredible capability, but it’s
more of an individual tasks, actually not on the whole brain. So, whole brain power,
because of the combination, maybe we are, I don’t know, I’m not saying we only use 10% maybe we use more than 10%,
but for individual tasks, there are very, very smart people, there are very, very capable, extreme people have extreme memories. So, with individual tasks,
I think we can explore many, many, many forward. Because these people exist on the planet. I think the best comparison is compare the regular people like us. My memory is maybe broke. Average. But with the extreme memory, and you can tell the difference, I think the difference is huge. So, we do have a lot of power to explore. How do we explore these
kinds of abilities, use them? I don’t think neuroscientists know yet. So, there are many, many myths out there, but I really don’t think
I can provide any advice. I don’t wanna score up your exam. – Well, I do wanna open up the floor for questions from our
audience, and so if you would, please make your way to
one of the microphones, to ask Dr. Feng a question or two. – [Woman] Hello, so I do mouse behavior, I’m a grad student in my third
year, and I’m just wondering, with all the translation crisis
going on with big pharma, how is that gonna change the landscape of academic research institutions? Are we gonna start
ditching the mouse model and moving to non-human primates? Is that the only way to
be competitive nowadays? – That’s a really good question. I don’t think that’s the
only way to be competitive. I think we should never
focus on one model. I always say we actually have no proof, we have no success yet
in monkey models, right? So it’s still very explorative. It’s only because evolutionarily,
they are closer to humans, so we have a better hope. I would say we have
failed many times in mice, we have not failed yet in monkeys, so, maybe it’s also failure. Maybe the human is the only model we have, and then that would be a problem. So, there are people, there
are many different ways. IPSL is one way, but they
don’t have perfect circuits, and transplant human
cells into mouse brains, monkey brains is another way. Then people are really starting thinking, how do we develop drugs
without animal models? If animal models are 99%
failure, why do I test them? So, before you test the animal model, you want to ask if this mouse model fails, do I stop this project or do I move on? If you still move on, why
do you wanna test them? So there are a lot of debates, and I think that, only to be used, I don’t think monkey is the only way, and I’m not even sure
monkey is the best way. Human is probably the only perfect model, and even humans are not perfect, one human is not a perfect
model for another human. So, we have a lot of problems in this. But, don’t be discouraged
by big pharma’s decision. Big pharma has moved out. It’s a continual move out,
Eli Lilly announced today. They are cutting their
neuroscience research in England. That’s the only major research they have now in neuroscience. But a lot of venture
in the last few years, ventures in biotech stock. So the big pharmas decide, I’m gonna put cash in my bank account, whenever you figure out
something, I’m buying you. You said you’re not gonna sell? Not true. It’s just I have not
offered you enough money. If I offer you enough, you
will sell your thing to me. So, that’s their mentality. Actually there are many, many
ventures, starting companies, if you’re around MIT, there
are many, many ventures looking for opportunity
investing in neuroscience. So, it’s not that bad, actually,
it can get really good. So, I encourage all of you, actually, if you have a good idea, start something, talk to the venture. Actually, one of the
impressions I got today is people are very, very active in translation research here,
so it’s really fun to see. Very much like MIT culture. – Another question? – [Marshall] Hello, I’m
Marshall, I’m a second year pharmacy student, I work with Dr. Young, and my question is, if we alter genes which would alter say, the structure of transmembrane proteins or cell surface proteins,
could that lead to a potentially serious autoimmune response? ‘Cause we are born with
self-identifying antibodies, could changing those surface proteins, could that cause like
an autoimmune response similar to an organ transplant, and if so, how frequently, how serious
do you think that might be? – Yeah, so that’s also
a very good question. There are two parts of them. If we change them during
embryos, germline in humans, it’s probably not a problem. It can send it’s own protein. However if we change it later on, and it could be a problem,
since, post-natally, if the patient never sees this protein, then it will become a problem. Although it’s a human
gene, you’re introducing a new gene to this person. Luckily, most of them are heterozygous, they already have the protein,
you just increase them. So in most cases, it’s okay. The major problem is CRISPR
is a bacteria protein. You introduce, you will
generate immune reaction. How do you deal with that? That’s a problem we have not solved. So, it is possible that
if you’re long term, people have tried to use
short term delivery protein, once you fix the gene,
protein will be degraded, you are totally fine. But these have not been really
successful in the brain yet but people are working
on delivery systems. – I think we have another
question over here. – [Woman] Hello, thank you for your talk. So in your talk you
mentioned that there are certain areas that can be fixed, the critical thinking
and executive thinking, is it the part that can be fixed, or is it the part that cannot be fixed? – So, right now, in mice, that’s the part we cannot fix in adults. But, you probably can fix, if you put a gene back
much earlier, post-natally, but before they are mature. In prefrontal cortex circuits,
we believe in adults, it’s why it effects, it
doesn’t have the plasticity we want and we couldn’t fix adult mice. But, if you give them earlier the gene, restore the gene function earlier, three weeks old or two weeks
old, you can actually fix them. – [Woman] Okay, thank you,
but I was reading somewhere that adults, even older people are fully capable of learning a new language. So then how come they are
able to learn in old age? – So, really good question. That’s what I’m saying, in
some parts of the brain, so, learning language, but
they are never gonna learn as good as the young people. So, for a lot of circuits
involved learning skills, striatum, they are very plastic. Actually in mice and in humans. That’s why if older people
can pick up bad habits, or get addicted, these all involve the cortical striatal circuits, so, they have certain very significant amount of plasticity in adults. That have been done in mice
and have probably shown in humans, for example visual system. Binocular vision, right? Binocular vision in
people in poor countries, they have cataracts. If you remove them before 10 years old, they perfectly have
binocular vision because you all know if you study vision, the axon has to cross, they
have to segregate, right? But so, that’s the critical period. But if in an adult, you
take out the cataracts, they can see perfectly, but they will never have binocular vision. So, the plasticity is gone. So in the cortex, the
plasticity really are limited, but in the sub-cortical area,
they are very much active. – [Woman] So you’re saying if somebody is, what is considered an adult brain? Is 18, 19 an adult brain? – In humans, recent studies suggest that maybe after 25 you are much less plastic, so we have a long time. So, you’re all plastic. (audience laughs) – [Woman] Thank you. – Well, one of the goals
of our Ideas Festival here at Purdue for the past year has really been to provide
our campus community the opportunity to go deep into some areas around science and technology, how it interfaces with
society and culture, and I think you have certainly
done that this evening with your presentation, Dr. Feng, thank you for coming to Purdue. – Thank you. – Join me in thanking
Dr. Feng one more time. – Thank you. (audience claps) – Thank you, have a great night. – Thank you.