(light futuristic music) - So Dana's given a great introduction to gene editing and talked about some
of the nuts and bolts of how it actually works in cells and what I thought I would do
for the next 40 minutes or so is tell you about the background
to the CRISPR technology. I think there's a couple of
really interesting things to appreciate about this. First of all, the word
CRISPR has become a verb. You hear people saying to CRISPR and I wanted to sort of
tell you a little bit about where this all came from and I think it illustrates the way that new technologies often emerge. They come from unexpected directions by people that are not necessarily trying to discover new technology. They're working on fundamental questions and they come across, in this case, a mechanism that reveals itself to be something that can be harnessed for a very different purpose. So I'm gonna talk about that and then I'm gonna say a little bit about where I think this is all going. So the CRISPR field is a very young field. It's only really been
around since the mid-2000s when scientists discovered that a lot of bacteria have
repetitive DNA sequences in their genomes that came to be called CRISPRs, an acronym that stands for clusters of regularly interspaced short palindromic repeats, a big mouthful. What does that mean? Well it just means that many bacteria have a distinctive
feature in the chromosome, sometimes more than one, that include repetitive sequences shown by the black
diamonds in this cartoon that flank unique sequences
shown by the colored boxes and three research teams in the mid-2000s noticed that in many cases
these unique sequences in the CRISPR arrays come from DNA found in viruses. So this was the first indication that these might in fact be some kind of acquired immune system in bacteria and scientists also noticed that together with these arrays were CRISPR-associated or cas genes typically located nearby
in the bacterial chromosome that turned out to be encoding proteins that are part of this
adaptive immune system. And our own research at Berkeley started because of the
work of Jillian Banfield, the scientist here at
Berkeley who does research on bacteria, typically on
uncultured, unidentified bacteria, and her research uncovered the fact that many bacteria that you can isolate in various environmental niches are very abundant in these elements and so they are probably
very actively using this kind of immune system to protect themselves
from viruses in the wild. So it's some very interesting biology and so I'm gonna show
you a very short video that illustrates how
these actually function and how the activities native to these bacterial immune systems are very very nicely relevant to the work that Dana just talked about because what they do is they make targeted cuts in DNA. So in bacteria this is happening to protect cells from viruses but we can, as you heard very nicely from Dana, this can be harnessed in
various kinds of cells per gene editing. So I think my, let's see if the audio works here. It's not, okay. Well we saw, I'll narrate this. So we have viruses that
are entering a cell. You can see a virus injecting
its DNA into a bacterium and if this bacterium has a
CRISPR array in the genome it can acquire new pieces
of DNA from the virus and integrate them into the array, keeping this pattern of repeats. And the cell is then able to
make a copy of that sequence in the form of an RNA molecule that gets subsequently broken into smaller units that each
include one of the sequences derived from a virus together with the sequence
coming from the repeat. And those RNAs are then kind of marked as CRISPRs molecules by
the presence of this repeat and they combine with a
second RNA called tracer to form a structure that binds
to the protein called Cas9. So this RNA protein complex then is able to survey the cell looking for DNA sequences that have a sequence matching
the sequence of the RNA and if that match is found then this protein RNA complex
unwinds the DNA locally to allow RNA/DNA hybridization to occur. A double-stranded break
is induced in the DNA and if this happens in a bacterium that leads to degradation
of the viral DNA. But if you introduce that
system into a eucaryotic cell, a plant or animal cell
as you heard from Dana, these cells have sophisticated machinery to repair double-stranded breaks and they can do so by
introducing a small change at the site of the break or even by integrating a new piece of DNA during the process of repair and so that means that this method of introducing double-stranded
breaks in a targeted fashion into the genomes of plants and animals is a very effective way to
conduct genome engineering. So that's a very fast summary of why this system was
harnessed as a technology and why it's been so useful,
why it's really taken off over the last five years and has spread through many
different areas of biology enabling all sorts of
research and now applications that would have been very difficult if not impossible to do in the past. So I'm gonna just I really
wanna tell you three things. I want to talk a little
bit about the mechanism of this process, how this RNA-guided DNA
cutting actually works. Why do we care? Well, I'm a biochemist so I think it's just kind of cool biology but as you'll see I think once
we understand how it works we can actually harness that activity and we can think about
ways to make it better for applications that
we want to use it for. So I'm gonna tell you a
little bit about this. I want to tell you a
little bit of new biology so one of the things that's
happening in the field right now is continuing investigations not only of new CRISPR pathways but also of how these are interfacing with other proteins that
are found in bacteria and one of the very recent discoveries has been finding that many cells and the phage that infect them, the viruses that infect them, have what are called anti-CRISPR proteins, these are proteins that actually inhibit the CRISPR pathways, and these are turning out also to be potentially very
useful as technologies and I'll give you one
brief example of that. And then at the end I
wanna just say a little bit about what we think about
the societal implications of a technology that is so enabling for making targeted changes to the DNA of cells and organisms. It's really a profound
thing if you think about it 'cause it really gives
humans now the power to control evolution of
organisms in our environment and also potentially our own evolution in a very targeted fashion. So it's exciting, it's enabling, it's moving fields
forward incredibly quickly but it also brings sort
of this sense of awe and the feeling that we need to proceed with appropriate caution and respect for this powerful technology. So I'm gonna say just a few words about that towards the end. Okay but let's talk first about CRISPRs and how they operate. So this is a slide that shows the pathway of adaptive immunity in cells. So one of the things
that was very exciting in the very early days of the CRISPR field was the idea that bacteria actually have an adaptive immune system, kind of analogous to the way that we have antibodies, we have, of course, very sophisticated adaptive immune system that allows us to protect
ourselves from pathogens. Nobody thought that bacteria, very simple, single-celled organisms, would be capable of
that kind of adaptation but it turns out that they are. However, that pathway
works very differently from the way that our own
adaptive immune system works and it works by means of
these acquired sequences that are integrated into the CRISPR arrays as I showed you in that video and then these become
templates for RNA molecules that are transcribed that include copies of all
of these integrated bits of viral DNA together with the
associated repeat sequence. And note that as the
CRISPR acronym indicates these repeats are often
palindromic in character. That means that when they're made in a single-stranded form like in the RNA molecule shown here they can fold back and form structures like little hairpins that are recognized by some
of the CRISPR cas proteins and that's actually how these
RNAs are often processed is by the recognition of those sequences. So the cell very nicely
effectively labels these as CRISPR RNAs by using these unique
sequence and structure tags. And after the RNA molecules are processed into individual units that each include a
virally-derived RNA sequence they combine with one or more cas proteins to form surveillance complexes. They can search the cell
looking for matching DNA or sometimes RNA sequences leading to degradation by cutting of those matching nucleic acids. So it's really a process of adaptation, acquiring these sequences from viruses that are entering the
cell for the first time, expressing those sequences
in the form of RNA and then using the resulting
protein RNA complexes to interfere with foreign nucleic acids. And so I'm gonna say, I'm gonna
tell you a little bit more about this step here, interference, because this is really the step that through understanding how it worked allowed this system to be harnessed as a gene editing technology as you heard about from Dana. Now one of the things to
appreciate about CRISPRs is that there are many
flavors of these pathways and this is a cartoon, it's actually already out of date but it was taken it from a
review published in early 2016 that shows different examples of CRISPR systems that are classified according
to the numbers and types of cas proteins that are found in these systems and the main thing I want you to note here is that we can really
classify these systems in two broad categories
called Class 1 and Class 2. The Class 1 systems all
include multiple genes, multiple cas proteins, that have to be present in the cell and have to assemble with CRISPR RNAs to form surveillance complexes. In contrast, the Class 2
systems of CRISPR systems include a single gene, typically, that encodes one large protein that is able to combine with CRISPR RNAs and provide the cell with protection. And it was really through
studying these systems that it was possible to
harness these proteins for gene editing purposes. And this is a slide that, this is just showing, this was from a recent
conference that we attended that really just shows
examples in cartoon form of these different classes of systems. Class 1 over here including
large arrays of proteins that assemble with CRISPR molecules to form these surveillance complexes and then over here single proteins that combine with RNA to carry
out that kind of function. And so in the early days when research had focused initially
on these Class 1 systems they were very interesting, the functions were fascinating but it was hard to imagine how
you could really harness them as a technology because you would
require multiple proteins to be made in cells and to assemble and it just seemed like a many component system that might be very
complicated to get working in a heterologous cell type whereas work on these systems over here which are much simpler and were simplified further by understanding that you could combine the two natural RNAs called
CRISPR RNA and Tracer RNA into a single guide form, this converted what had
been initially a very, looked like a pretty
complicated kind of system into something that looked a lot easier to harness as a tool. And so for us one of the key questions
that we set out to address in those early days was to understand the
function of the first example of this large type of
protein, single protein, in the Class 2 CRISPR systems that could operate to protect cells in an RNA-guided fashion, protect them from viral infection, and the question was
really what is the function of this encoded protein which in the very early
days was known as CSN1 and then very quickly became called Cas9. And that line of research which was conducted with a collaborator Emmanuelle Charpentier and her student Krzysztof Chylinski revealed that Cas9 is an amazing enzyme that has the ability to
recognize double-stranded DNA at positions in the DNA sequence that match a 20 nucleotide sequence in this guiding RNA and note that I'm showing you here the guide RNA in its single-guide form where the natural CRISPR RNA which would be this part
of the sequence here and what's called the Tracer RNA which would be this part
of the sequence over here, have been combined by linking
them together covalently. So this can be made as
a single transcript, something that turned
out to be very helpful in harnessing this as a technology because it's made it easy for researchers to make RNAs that would have a desired sequence here that would track Cas9 to a
particular site in a genome and trigger its DNA cutting ability allowing it to make a
blunt double-stranded break at a precise place in a DNA molecule or in an entire genome and you've already heard quite
a lot about that from Dana. And so in our lab we've
been sort of curious about really addressing this question of how recognition really works, not just that we can see
that involves base pairing but it's really also a question of how this protein is
able to open up the DNA in a genome and you've gotta imagine that you've got the DNA in
a typical eucaryotic cell, of course it's in the nucleus, it's also highly packaged. It's wrapped around histones,
it's compacted into chromatin, different types of chromatin,
lots of proteins around are bound to it, DNA replication and repair machinery and other proteins moving through. How does this bacterial protein deal with all of that and somehow find these sequences
typically quite accurately and generate double-stranded breaks? And that's really the question
that we've been seeking to understand over the last few years. And I threw this slide in
to remind me to tell you that unlike this sort
of cartoon right here the Cas9 protein has to
be able to unwind DNA without any external source of energy. So it doesn't hydrolyze
ATP or GTP to unwind DNA. Somehow it triggers DNA unwinding
by some other mechanism. And so we'd like to
understand how that works and why do we want to know this? Well again, we're very
interested to understand what makes this enzyme
functional in eucaryotic cells, how does it deal with
chromatin, et cetera, how does it get to the
right site in the cell, what happens when it
gets to the wrong site? Can we do things to
prevent it from accessing the wrong site, make it even more accurate than it is naturally? And also can we harness its
DNA recognition activities to do other things like not just trigger covalent changes to the DNA in the genome but also to recruit other factors to positions in a genome, use it for DNA imaging, for example, and also for the kind of thing that Luke Gilbert is doing which is harnessing its
activity to regulate the expression of genes, so making changes to the
levels of proteins in cells without actually changing
the DNA sequence in cells. So one of the things that's
emerged over the last few years is that one of the ways
that Cas9 probably does this is through its ability
to change confirmation upon binding to nucleic acid and so I'm gonna show you a little video that was made by a student Ben LaFrance that morphs together a series
of crystallographic structures solved for Cas9 in
different states of assembly with nucleic acid. And so this is a movie that
starts with Cas9 protein alone, that's a crystallographic structure, and you see it morphing to
the state of the protein when its bound to the orange guide RNA and you saw a big rotation in
this gray part of the protein that reorients it to open
up a channel in the center where the guide part of the RNA is sitting and then when this protein
binds to a DNA molecule you can see additional structural changes that happen in the protein here to accommodate the
RNA/DNA hybrid that forms. Now those structures showed us that there had to be an
additional structural change. This rotation that
you're seeing right here which was modeled in at the
time that we made this video because we knew that
this part of the enzyme which is a domain in
the protein called HNH, this is actually one of the
chemical cleavers in Cas9. It's the part of the protein
that cuts the DNA strand that base pairs with the guide RNA. But in all of the crystal structures that were solved initially, which were solved using
single strands of DNA that were annealed to the guide RNA, this domain was not in the right place to actually cut the DNA. It was located quite far away. It was sort of over here
rather than in this position you can see modeled here where it would have to be positioned to actually make a cut in
the targeted DNA strand, so what was going on, and so one of the things
that emerged more recently in research that was done
by a team at Berkeley including Fugo Jiang and
David Taylor in my lab and this was a very nice collaboration with the lab of Eva
Nogales here at Berekeley who does cryoelectronmicrosopy, was at a combination of crystallography and electron microscopy allowed these students to trap a structure of the Cas9 protein bound
to its true substrate which is a double-stranded DNA molecule and so you can see that hopefully it's a little bit dark here but hopefully you can see
this double-stranded DNA. So here's the one strand is in blue and the other strand is in magenta so you can see the duplex DNA here opening up inside the protein. This targeted strand is forming a duplex with the guide RNA and then here's the non, what
we call the non-target strand is over here and remember that when Dana
Carroll talked about work that Chris Richardson is doing
at the IGI with Jacob Corn they had found experimentally that this part of the DNA is more exposed after DNA cleavage and this structure kind of
reveals the basis for that. You can see its located really
on the surface of the protein rather than buried inside
like this strand over here. And the cool thing about this structure was that we found that the, and I think this is actually an animation, I could show this to you, but it's basically, I think I'll just tell you, that this is a, if you
look at this green part of the enzyme here, in this structure this domain that cuts the target DNA strand is now actually positioned
very near where it needs to be to cut the DNA. So there's something about having the non-target strand
present in the complex that actually triggers this protein to be in the right confirmation to actually conduct chemistry. And so we're, we're
doing a lot of work now, and I'll tell you a little
bit about this shortly, to really understand
what triggers this domain to swing into place to cut the DNA because there's a lot of evidence that this is a protein that
can bind DNA quite readily even in cells and people have done experiments showing that you can detect binding of Cas9 at a lot of sort of close
but not quite perfect matches to the guide RNA and cells and yet most of those sites are never altered chemically. Why is that? Well we think one reason
is because this domain is actually very sensitive to full base pairing of the guide RNA with the target DNA sequence. And just to show you
why we think that's true I'm gonna show you one
series of experiments that are being done to test the conformational states of Cas9. So you might, if you don't do biochemistry
or think about proteins in this way you might wonder how do we know that, how can you actually figure
out structural changes in proteins, I mean, that
sounds kind of detailed and certainly we can't get
that information necessarily from snapshot structures like
we see in crystallographic structures of proteins. And so one strategy for testing conformational
changes in proteins takes advantage of the
real physical changes in distances between atoms,
or individual amino acids, in a protein if there's a
change in its structure. And so when we had this model for what we thought must be
happening in the Cas9 protein with respect to this active
site, this HNH domain that had to swing into place
to cut the target DNA strand the idea here was to introduce
pairs of chemical dyes on the surface of the enzyme that would be in very
different spatial relationships depending on the
conformational state of Cas9. So this is just showing an example where we had this HNH domain in yellow so the inactive state is shown over here, we've got a dye sitting right there. When this domain swings into
the active position shown here you can see that the position of that dye moves quite a lot and if you've got another
dye sitting over here in the protein these two start of
initially very far apart, they can't really chemically interact so you don't get much of a signal by something called
resonance energy transfer but when this domain change happens now these two dyes are very close together and you can detect a
resonance energy transfer between the two. So that's one way that you can detect these kinds of structural changes. And just to show you a little bit of data. This is actually fairly new, I don't think this
publication has appeared yet, and this is a collaboration
with Ahmet Yildiz, a biophysicist here at Berkeley, and his student Yavuz and a
student Janic Chen in my lab. What these students did was they were able to set up a system where they can tether the
Cas9 guide RNA complex to a surface and they do this by, so
here's the guide RNA, you can see it's RNA end
is being chemically linked to this slide surface, and then we can flow in
double-stranded DNA substrates for a recognition by this complex and we are using Cas9 proteins containing these pairs of
dyes that I showed you. So we can actually monitor changes in fluorescence resonance
energy transfer called FRET as a function of interactions
with these DNA substrates and by setting up an experiment like this it's very trivial to
change the DNA sequence. So we can have DNA molecules
that have a perfect match to the guide RNA and then we can also test DNA molecules that have various
mismatches to the guide RNA and see how that affects the ability of the protein to interact. And I'll just show you
a little bit of data. So what does the data for
these experiments look like? So here we've got the Cas9 RNA/DNA complex that we're monitoring formation of and over here you can see we're plotting the resonance
energy transfer signal that we're getting as a function of the complementarity between the DNA molecule
we're using in the experiment and the guide RNA. So if you just look at this last line here this is what we see if we use
just the Cas9 RNA complex, no DNA is present. The protein is in an inactive state as I showed you before and we get a FRET signal so most of the particles here have a very low FRET signal. That's because the dyes are very far apart in the confirmation, they don't interact. As we add DNA molecules that have increasing
amounts of complementarity all the way up to a perfect
match, to the guide RNA, you can see that more
and more of the molecules are populating this active state where the dyes now are very close together because this conformational
change has happened and so we're getting a very strong signal between the dyes. So we can really monitor that very nicely and the cool thing in this experiment was that we found that a lot
of the molecules initially these proteins, get stuck in
an intermediate confirmation, it's not fully off,
it's not fully inactive but it's not in the active state either. So there's sort of this intermediary state that is populated and sort
of the degree of population of this state depends,
again, on the complementarity between the DNA and the guide RNA. So it really tells us
that this is a protein that's a sensor, right,
it's really sensing the degree of a match between a target DNA and the guide RNA and that sensing is being conveyed in terms of this conformational
change of the protein. And so that really led to the idea that the protein goes
through this series of steps to get to an active state in which it starts of
in this inactive form, it goes through an intermediate and then it reaches this
sort of fully-docked state and then is able to cut the DNA and this can then dissociate and the protein gets reset and can cut other DNA molecules
in a catalytic fashion. And what we're doing right now is we're actually using
this to design versions of this protein that are even more accurate, they're even better
sensors of the target DNA than occurs in nature and we have a paper that we
just posted on the bio archive, the pre-print server, that shows design of new
mutations in the Cas9 enzyme that we think are making this protein an even better sensor, potentially more accurate at DNA cutting and thereby may be useful as a technology for gene editing as well. Okay, so let's talk a little bit about how nature fights back. And so the story of anti-CRISPRs really started kind of where
the story of CRISPRs did which was with microbiologists who were trying to understand how these systems operate in nature and of course when
there's an ongoing battle between infectious agents and their hosts they're going to be protein, there's a lot of selective pressure for both systems to
evolve ways to get around the defenses that are put up in each case. This is a cartoon that was actually made by Megan Hochstrasser who is one of our outreach coordinators here at the IGI illustrating sort of in fanciful form different ways that we could imagine that cells might come up with ways of blocking these CRISPR pathways. They could have ways of
stopping the CRISPR proteins, they could prevent binding
to the RNAs potentially and there might even be other pathways as I'll show you that would lead to
inhibition of CRISPR systems. And so Joe Bondy-Denomy, a scientist at UC San
Francisco across the bay, is one of the lead scientists whose been working on
these anti-CRISPR systems. So he's a microbiologist studying ways in which phage interact with their host bacteria and he noticed that in, there were sort of interesting examples of organisms that should have been or phage that should have been eliminated by these CRISPR pathways
that weren't somehow. And so why was that? And investigating the mechanism that led to the discovery that in many cases these organisms, these host organisms, actually encode little proteins that turn out to be inhibitory
to the CRISPR system. And so as mechanistic biochemists we've been very curious to understand again, how do these work, and how can these proteins which turn out to all be very small, they're typically under 100 amino acids, how do these actually
operate as inhibitors and can we use their mechanisms to tell us more about the
way that these proteins, in particular Cas9
enzymes, actually operate. And I'm just gonna show
you a little bit of data. Again this is unpublished, very new data. But one of the things
that was very interesting, and this work being done by students Kevin Doxzen, whose here
and how working at the IGI, and Lucas Harrington, was they noticed that one
particular anti-CRISPR protein that I'll abbreviate C1 turned out to have the
ability to block DNA cutting and the way we can test this biochemically is doing an experiment where we can put a radioactive label on one end of one strand of a double-stranded DNA target sequence and then we use this labeled DNA in an in-vitro DNA cleavage experiment and so if we do this with
a particular Cas9 enzyme you can see we get very
nice cutting of the DNA as you would expect if we don't
have any inhibitor around. As soon as we add the inhibitor though we see that DNA cleavage is blocked and so once you see that as a biochemist you say well, that could mean a couple of things. It could mean that this inhibitor just now doesn't let the
Cas9 protein even bind to the DNA substrate. Maybe it just prevents binding or it could be that it allows binding but somehow prevents DNA cutting and so to distinguish between
those two possibilities Kevin and Lucas did an experiment in which they used a
non-denaturing gel system. So this is a gel system that allows us to visualize
any trapping of proteins as they associate with nucleic acids. So if you look at the left
hand part of this gel system here we're just doing, we're taking these reactions right here, no inhibitor is present, and we just apply them to
this native gel system. So we've got our DNAs radio labeled and you can see that as we
increase the amount of Cas9 and its guide RNA that
we add to the reaction over time we get cutting of the DNA and the DNA dissociates and you can see it running faster on this, migrating faster on this gel system. So that's just what you
would expect to see. What happens when we add the inhibitor? Very interesting result. Now what we found was that we saw not only no cutting of the DNA but we saw that this radio labeled DNA becomes trapped in a much
slow migrating product and that turns out to
contain the Cas9 protein and the guide RNA. So that tells us that this inhibitor doesn't prevent binding to the DNA but somehow it prevents
the DNA from getting cut. So it seemed very interesting. How would that work? And so to sort that out Kevin and Lucas and
another student Josh Cofsky made a whole series of
variants of the Cas9 protein and I really just want
to point out one thing. So these are just cartoons that show kind of a linear cartoon of the Cas9 protein sequence. And what we found was that in every case where the HNH domain was
included in the construct we saw that there was binding
to this anti-CRISPR protein. So this is just looking at direct protein protein interactions but constructs that were
missing the HNH domain like this one right here had no binding to the inhibitor. So it really looked like
that was somehow the domain responsible for this
protein protein association and just to really bring that home they made constructs
that are shown down here that were cymeras in which the HNH domain from
this inhibited Cas9 enzyme was swapped into a different Cas9 protein and this now becomes capable
of binding the inhibitor whereas if we do the opposite we use the Cas9 protein that naturally can bind the inhibitor but we replace the HNH with
a different Cas9's domain that doesn't bind the inhibitor, now it doesn't bind anymore. So it really looks like it's specific to this catalytic domain of the enzyme. And so Lucas and Kevin were able to solve a crystallographic structure
of that interaction and this turned out to show that the actual chemical,
chemically-important residues in this HNH domain are physically blocked by this little inhibitory protein. It literally grabs on to the
sites important for chemistry in this domain and prevents them physically from interacting with DNA and the way we think this actually works, and this is now a model, is that this inhibitor grabs
on to the HNH domain in Cas9 and literally prevents it
from swinging into place and cutting the DNA as I showed you that we know
it has to be able to do. So this is a really cool example of a natural inhibitor that has the ability to trap the protein and its guide RNA on a DNA target by not physically preventing
it from associating but actually just taking advantage of the natural mechanism of cutting and preventing this conformational change. So kind of cool mechanism but again, do we care about this from a technological standpoint? And I would argue that we do and it really comes back to the work that Luke Gilbert and his
colleagues at UCSF have been doing and also the work
originally of Stanley Chi who is now at Stanford who were able to show that
you could make forms of Cas9 that were inactivated by making just targeted changes to the chemically-important
residues in the enzyme including in this HNH domain. And then linking effector
proteins to the enzyme that had the ability to repress
or activate transcription. And so this shows that, this is sort of how one can do this kind of artificially in the laboratory, but what if it's true that, what if it turns out to be the case that in phage or bacteria that have this particular
anti-CRISPR protein they can actually block
DNA cutting by Cas9 and thereby allow it to
function in a regulatory fashion kind of analogous to this. So that's something
that we're now testing. So it may be that this
really kind of allows phage or bacteria to expand the function of Cas9 naturally in cells not by making mutations in the enzyme but simply by blocking
its ability to cut DNA but retain its ability to
bind in an RNA guided fashion. Okay, so just in the
last minute or two here I just want to say a little bit about what I call responsible progress and so one of the things
that's been very exciting over the last few years is just the rapid adoption of the Cas9 technology for all sorts of gene
editing applications. And this was a cartoon that was published in Nature last year just showing that as Dana said now it's sort of over 100
organisms and cell types and many many more cell types that have been modified using this system. And just to show you some
of the very recent things that have been happening that
I think are really exciting. So this was a paper that came out. This was actually on the cover
of the journal Cell recently where a group at Coldsfer and Harver was able to use the Cas9 system to make targeted changes in tomatoes and they were doing this
to actually separate two genetic changes in these plants that had been selected by
traditional plant breeding but couldn't be separated by
traditional plant breeding because of the length
of time it would take to actually do that. They were able to knock
out one of those genes and thereby create plants that have much stronger branches. They were able to hold
heavy fruit loads like this, very practical use of gene editing and doing something that
would have been difficult if not impossible to do
with traditional methods. Why was this on the cover of Cell? Well I think it just
is sort of a harbinger of where this is all going, that in the future it's going
to be possible, we think, to make these kinds of changes in all sorts of plants, not just the major crops but also in plants that people are growing in their gardens and to do things that will create traits that are very
useful in different ways. Another recent example
was this one right here. So this was a Chinese group that published work showing that you could actually cut out integrated HIV proviruses from the genomes of mice and thereby effectively kind
of cure individual cells of viral infections. So it's kind of very reminiscent of the way that this is
actually operating in nature except here we're doing it in mice. Now will this be an
actual therapy for HIV? Probably not, right, because
it might be very hard to actually cut all the proviruses out of infected cells in these animals. On the other hand, it shows the precision of the technology and it also points to the fact that there's a lot of interest in using this system to protect organisms from viruses by harnessing these activities in ways that are analogous to the way that they operate in nature. And finally then there's germline editing so all of the editing that's
being done at the moment for clinical use is being done in somatic cells and cells that are fully differentiated but we know that this
technology works very nicely also in stem cells. It works in germline cells. This is an example from Russell
Vance's lab here at Berkeley showing a fertilized mouse
egg held by a pipette and we're seeing a needling coming in injecting Cas9 guide RNA complexes into this very early embryo and when those edits are made to the DNA they become heritable, right, they become part of the entire animal including its germ cells so they can be passed on
to future generations. And if that's done in a human being then we have a situation where we're making changes
that become embedded in human populations in principle. And so this has triggered
a lot of discussion around the world, very active conversations about the ethics of this and has led to the release very recently of this document here by the
National Academy of Sciences on human genome editing which if anyone's interested
you can download it and it really goes through all
of the implications of this, especially for human germline editing, and how sort of a roadmap for how we as a scientific
community, global community, can move ahead to ensure that there's robust research going on but that it's not, that's not impeded but that we also have
appropriate regulations around this kind of very
powerful sort of application. And you'll hear more
about that later this week as Dana said. And I'm gonna close just by mentioning that all of this research is done, of course, by
students and post-docs in various laboratories including our own but we've had many many many
wonderful collaborations and I'm just mentioning
a few of them here. These are some that I mentioned today. Karen Maxwell and Alan Jacobsen at University of Toronto are actively working on anti-CRISPRs and we've had a great time
collaborating with them as well as with Erik Sontheimer at University of Massachusetts and I mentioned these folks. Science can't be done without funding and I think we all have to be aware that right now there's a very, we're facing a very difficult situation, especially here in the United States, where there's an active push
by our federal government to cut back on research funding, especially for the kind
of fundamental research that I talked about today and so you can see that
if we really do that I think we're gonna put the
US at a severe disadvantage in terms of discovering new technologies and moving ahead in the areas of research that we know are gonna be important, not only for human health
and our environment but also for stimulating the economy, creating jobs, et cetera. So I would ask all of you to think about ways
that you can communicate the importance of the
science that you're doing and encourage your friends,
neighbors, families, who are, all of us are taxpayers, to take appropriate measures as they feel so moved to support science funding
at a fundamental level. And I'm gonna stop there. I don't know if we have time for questions but if you have any I'd be
happy to try to answer them. Yeah. (student asking question) Yeah. (student asking question) It does, it hangs onto the DNA, yep. I didn't have time to say
that but that's right. It has a very high binding
affinity to the DNA even after cutting. (student speaking) Yes, yep yep, you can do that actually using this FRET
essay we can do that. So there's a lot of nice, I think a lot of nicer
biophysical measurements like that that we can make. Now one of the caveats with all of that is that we're doing that
in vitro, of course, we're not doing it in
the context of chromatin so then we have to have other ways that we can try to
relate those measurements to what might be happening in cells and there are already
we've done some of this and there are other labs
looking at invivo imaging of Cas9 so you can really watch behaviors in the nucleus as well. So I think over time we'll build
up this very nice continuum between biochemical experiments in-vitro and then things that are
going on in-vivo in the cell. Yep. (student speaking) (laughter) You can imagine there's probably, yeah, anti anti-CRISPRS and yeah. Sure, I mean, I think it's sort of one of those things where if you can imagine it it's probably out there
somewhere in biology and if it's not out there already you can imagine ways that you could potentially
engineer such things. So I think one of the things
that's going on right now, so there's been a big push towards using engineered proteins or natural ones to control the activities
of Cas9 enzymes in cells for purpose of safety and also for the purposes of accuracy and in fact Jacob Corn, I think he just stepped out, but his group with us have recently been using
anti-CRISPRs for that purpose where you can actually
limit activity in cells to reduce the activities that might be occurring
at off-target sites. So think this is an exciting time where we're seeing very
rapid technology development around not the fundamental
core activity of Cas9 but really how we modulate
that activity in cells using some of these types of approaches. Yeah. (student speaking) Yeah yeah, that's a great question. So there's a lot of work
going on on adaptation so that means basically the acquisition of new viral sequences
into these CRISPR arrays and didn't have time to tell you about it, out lab has done a lot on this and there's other labs, of course, contributing as well but we know that there's a
particular CRISPR inner grace. It's a two protein complex, actually has six copies of those proteins that form the inner grace and that protein complex is able to find and
integrate new bits of DNA into the CRISPR array very precisely by maintaining the structure of the array and in some types of cells like in e-Coli K12 there's an enzyme that, it's actually not an enzyme,
it's a DNA binding protein that bends DNA very sharply and contributes to the
accuracy and efficiency of that integration pathway. So it's a great system. George Church's lab has actually done work using the inner grace as a technology. So they're using it as a way to do sort of cell recording where they can integrate bits of DNA into populations of cells much as happens in nature and use that as a way of
recording information. So I think there's sort
of really interesting forthcoming applications of other aspects of these systems that haven't been harnessed fully yet but take advantage of
the natural mechanism including the adaptation mechanism. You should tell me when to stop. Yeah, maybe one more and
then we'll take a break. We can continue discussing outside. (student speaking) Yes, the one that I showed you does, yeah. Yes. It is effected actually, yeah, because it turns out that those two domains are highly coupled, at least in the proteins
that we've been studying, so when you inhibit one domain even though you're blocking
just the active site that cuts one of the DNA strands it actually does affect
cutting of the other strand. (light futuristic music)
