Skip to main content
Loading Explore guides…
CRISPR Technology Background · Acedra
Home
Explore
Biology
CRISPR Technology Background
Sign in to save
Chat
If YouTube asks you to sign in, open the video on YouTube.
Watch on YouTube
Reload player
Transcript
130 segments
Copy
Hide video
0:03
(light futuristic music)
0:08
- 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
0:36
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.
1:07
So the CRISPR field is a very young field.
1:12
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,
1:33
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
2:08
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,
2:43
and her research uncovered the fact that many bacteria that you can isolate in various environmental niches are very abundant in these elements
2:55
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
3:05
and so I'm gonna show you a very short video that illustrates how these actually function and how the activities native
3:15
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.
3:28
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,
3:40
let's see if the audio works here. It's not, okay.
3:47
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
4:20
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
4:54
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
5:25
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.
5:38
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,
6:06
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
6:36
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.
7:07
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
7:38
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
8:08
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
8:37
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
9:10
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
9:47
to form surveillance complexes. They can search the cell looking for matching DNA or sometimes RNA sequences leading to degradation by cutting
9:59
of those matching nucleic acids.
10:03
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.
10:23
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,
10:50
it's actually already out of date but it was taken it from a review published in early 2016 that shows different examples
10:57
of CRISPR systems that are classified according to the numbers and types of cas proteins that are found in these systems
11:08
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
11:27
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
12:03
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
12:33
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
12:58
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.
13:17
And so for us
13:19
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
13:53
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
14:27
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
14:57
at a precise place in a DNA molecule or in an entire genome
15:04
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.
15:33
It's wrapped around histones, it's compacted into chromatin,
15:39
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
16:12
the Cas9 protein has to be able to unwind DNA
16:18
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?
16:49
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,
17:22
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
17:54
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
18:26
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,
18:48
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
19:19
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
19:55
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
20:20
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
20:39
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,
21:01
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,
21:29
to really understand what triggers this domain
21:34
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
21:46
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
22:09
with the target DNA sequence.
22:13
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
22:40
and certainly we can't get that information necessarily from snapshot structures like we see in crystallographic
22:49
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,
23:03
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
23:15
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
23:47
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.
24:15
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
24:46
to this slide surface,
24:50
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
25:23
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
25:56
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
26:32
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,
26:55
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
27:25
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
27:52
and can cut other DNA molecules in a catalytic fashion.
27:59
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.
28:32
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.
29:05
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
29:33
that would lead to inhibition of CRISPR systems.
29:39
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
30:01
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
30:27
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,
30:57
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
31:32
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
31:50
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
32:25
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,
32:54
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.
33:28
So it seemed very interesting. How would that work? And so to sort that out Kevin and Lucas and another student Josh Cofsky
33:38
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.
33:53
And what we found was that in every case where the HNH domain was included in the construct
34:01
we saw that there was binding to this anti-CRISPR protein. So this is just looking at direct protein protein interactions
34:12
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
34:35
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
35:07
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
35:38
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
35:49
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.
36:04
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
36:33
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
36:46
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.
37:14
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
37:46
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.
38:14
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
38:35
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
39:02
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
39:34
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
40:02
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
40:32
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
40:45
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.
41:04
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
41:31
for how we as a scientific community, global community, can move ahead to ensure
41:37
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.
42:07
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,
42:36
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.
43:02
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
43:19
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.
43:36
(student asking question)
43:44
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.
43:54
(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
44:21
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
44:34
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
45:13
engineered proteins or natural ones to control the activities of Cas9 enzymes in cells for purpose of safety
45:22
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
45:54
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
46:24
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
46:43
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.
47:11
So I think there's sort of really interesting
47:16
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)
47:41
Yes, the one that I showed you does, yeah. Yes.
47:47
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.
48:20
(light futuristic music)
Here to help you learn
Sign in to keep a private conversation with your course tutor.
Ask me anything about the material…
Sign in to chat