Interview with Gootenberg and Abudayyeh: Using gene editing tech. to detect RNA targets

Lydia Morrison: Hello and welcome to the Lessons
From Lab and Life Podcast. I’m your host, Lydia Morrison, and I hope
that our podcast offers you some new perspective. Today I am joined by Jonathan Gootenberg and
Omar Adudayyeh who are the first McGovern fellows of the McGovern Institute for Brain
Research at MIT and they’re here with us to share their work, applying CRISPR gene editing
technology to diagnostic reporter assays. Jonathan and Omar, thanks so much for joining
me today. Omar: Yeah, thanks for having us. Excited to be here. Jonathan: It’s great to be here. Thanks so much. **H2 header: Cas13**
Lydia Morrison: Could you tell us about Cas13 and how you’re using it as a tool to detect
disease states? Omar : Yes, so Cas13 is an enzyme that is
associated with CRISPR systems and when the genome editing craze with CRISPR was taking
off, most people were really excited about this one protein, Cas9, and how we could cut
DNA and be used for genome editing. At the time we were really interested in whether
there were other enzymes within CRISPR that could also be useful for biotechnology purposes. At the time we were more interested in DNA
targeting, so we built a whole sort of computational pipeline of search, you know, you know, tens
of thousands of bacteria for a new CRISPR enzymes. And out of that search we did indeed find
more DNA targeting enzymes like Cas12 but there was this other enzyme that looked very
different than them and actually targeted RNA that we ended up calling Cas13. Omar: From that search we ended up characterizing
this enzyme showing that they could actually cut RNA, degrade RNA and in fact, even detect
RNA. So a lot of applications with Cas13 have to
do with around really sensitive nucleic acid detection for RNA targets, whether that’s
RNA viruses or, taking DNA targets, converting them into RNA and also detecting them. So we’ve been able to show Cas13 for detecting
bacteria from blood, detecting cancer DNA, even for agricultural applications, like being
able to genotype plants that may have been modified. So you can either, figure out if you know
there are soy beans that have a specific trait you’re interested in or if you want to be
able to make the engineering of plants easier and rapidly detect traits you may have introduced
into your population, it’s a, I think it’s a really powerful enzyme and has applications
from both healthcare, agriculture and even beyond. So yeah. Lydia Morrison: So you mentioned really sensitive
detection limits. What sort of sensitivity are you talking about? Jonathan: We can essentially get down to the
single molecule. If you have a single molecule in a solution
and that, so the way you measure things is molar, right? And we need a prefix in front of that. So atto is 10 to the negative 18 so it goes
nano, pico, femto, atto. Lydia Morrison: I don’t hear about atto a
lot. Jonathan : So attomolar levels. If you have a two attomolar concentration
in one microliter of that is literally a single molecule and we can detect that. So you can’t really get much more sensitive
than that. And it allows us to do a lot of things like,
you know, Omar mentioned there’s all these different applications, infectious disease,
in oncology, in agriculture where we can take these potentially very minute amounts of material
or a very minor population and a certain type of sequence and detect it. So we’re very interested in now is there’s
a lot of oncology applications where if you have a disease, if you have cancer and you’re
shedding this sequence into your blood, we can detect that at a very low frequency. So, I think it has many different applications
that we’re very excited about. **H2 header: SHERLOCK Technology**
Lydia Morrison: Wow, that sounds incredibly powerful. Could you tell us about how you took the discovery
of the Cas13 enzyme and applied that to create the SHERLOCK technology? Omar: Yeah, so when we were first characterizing
Cas13, it was really exciting to actually see it could programmably target RNA. But what we also saw was that when it was
activated by target, it would end up cleaving any other molecules in solution. At first we thought it was contamination problem. And we have our in cases and solution, like
we need to purify this again and again and try to get as clean as possible and you spent
months on repeating this. And we kept finding the same result. So we realized that Cas13 operated very differently
than other CRISPR enzymes that were known at the time. And that basically when it’s bound and recognized
in RNA target, it could end up just cleaving anything around it. We realized that while this might not make
it the best cellular tool, if you want to specifically target mRNAs or specific genes,
it could actually report on the presence of a target. Omar: The idea was if we could bring in a
reporter molecule that when cleaved would fluoresce, we could spike that in. And when Cas13 found, Zika target for example,
we cleave this reporter and release fluorescence. And so it was very like an unexpected route. We had never thought we’d go into the diagnostics
field or you know, we were not even thinking about that. We were trying to make gene therapy, gene
editing. And so we really followed this unexpected
results. And what we started doing is just seeing,
you know, can we apply Cas13, routines to all sorts of applications. You know, can we detect, you know, pseudomonas,
can we detect Zika, can we detect melanoma mutations in the blood. We just kept applying it and just kept working. We even found it could work from saliva. It worked from blood, urine, all sorts of
samples without even purification. Omar: Like a lot of diagnostics. You have to take a human sample, you have
to purify out the nucleic acid and then you put it through an instrument that’s rather
complex and it’s a lot of many steps. It takes time. It takes a skilled personnel. And we were finding that this simple enzyme
could make just a single step reaction where you could literally spit in a tube, maybe
heat that tube up for a few minutes and then just spike it into this reaction. And you could have detection on a paper strip
even. So it really made the whole idea of diagnostics
becoming outside of a complex lab setting and more maybe the home into the field. And I think we have really big aspirations. You know, you can imagine being able to detect
flu at home or being able to know if you actually have a viral disease or bacterial cold and
whether you need to go into the hospital or not. So I think, we’re really hopeful to see where
we can take it. Lydia Morrison: Yeah, that sounds like a really
beneficial application of it. What are the advantages of SHERLOCK versus
other molecular diagnostics? Jonathan : That’s a great question. There is a lot of different ways to actually
detect nucleic acids and do molecular diagnostics. One of the really nice things about SHERLOCK
is that we have basically this enzyme, Cas13, that does this detection and it’s very specific
so we can tune it so it can distinguish even a single base. And that’s very nice because in these applications,
like in cancer or certain viruses, like with drug resistance, we want to be able to actually
detect a specific mutation. And so that makes it very specific and that’s
coupled with the fact that once the protein is activated by this detection, it’s a kind
of a very catalytic mechanism where it’s going to cleave many copies of these reporter molecules. Jonathan : So it really means that we have
this very specific interaction that unlocks this very promiscuous and high turnover detection. So that means that we can dope in this reporter
molecule, which we can make very cheaply, and get very good detection even when there’s
low levels or something that’s very close in sequence to something else. I think that specificity and low cost and
sensitivity combined makes it a really robust method. This is also coupled with the other things
that Omar already discussed. Where we can take it to different outputs
and we can multiplex it so we can read out many different things at once. But I think the true power is from this Cas13
specific detection and then activation that makes it really amazing. Lydia Morrison: You mentioned that the activity
can be kind of promiscuous. Does that mean that you observe a lot of off
target events? Omar: The actual activity is promiscuous,
but the way the enzyme is activated is very specific. So the activity only activates when the Cas13
recognizes a target through its guide RNA. And the guide RNA is programmed to be complimentary
to your nucleic acid target. So only if it finds the precise sequence match
will the enzyme change in a way such that this enzyme can now start cutting anything
in solution. And in that case it’ll cut anything indiscriminately. But what we found, as Jonathan was talking
about was that, even a single nucleotide mismatch can sort of inactivate the enzyme and not
allow it to activate, giving us that ability to detect single base pair mutations really,
really well. And that’s really unique from a lot of other
assays where detecting a single mismatch can be quite difficult and you can always have
background signal from those assays. And so that’s a really big advantage of this
system and CRISPR enzymes in general. **H2 Header: RNA Editing RESCUE Technology**
Lydia Morrison: Earlier today you gave a seminar at New England Biolabs which I attended, and
you mentioned the RESCUE technology that you’ve recently developed. Can you tell us a little bit more about that? Jonathan : Yeah, so the RESCUE technology
is one of our RNA editing technologies and it’s complimentary to a repair technology
that we published back in 2017. So RESCUE is a much more recent development
and these are both technologies that rely on Cas13 being catalytically activated. So we make it so it doesn’t cut RNA anymore,
but it goes to RNA and then we can drag something with it. And what we drag with it is an RNA editing
enzyme. So the repair technology dragged with it,
an enzyme that goes from A to I and that allows us to correct certain mutations, but we wanted
to do is expand the number of edits we could make. One big motivation for this is that if you
want to make certain edits that can change your protein in a way that it functions differently,
that’s could be very interesting for a therapeutic application where we affect a pathway. Jonathan : And as we mentioned before with
RNA editing, it’s temporary. So you could actually temporarily change a
protein in a certain way. So what we did with RESCUE is that we took
this protein that we’re recruiting with Cas13, it’s called ADAR and we actually did a lot
of directed evolution and rational mutagenesis to modify it. So actually did C to U changes so it unlocks
this entire new potential base transition. That was actually a long undertaking that
we did and we found that we could actually get it to work effectively. Then we demonstrated that we could target
certain pathways like the wnt signaling pathway, we could target beta-catenin which is a member
of that, and we could actually activate that signaling and cause cells to grow just by
delivering this targeted RNA editing approach. So that was very exciting and we’re enthusiastic
to see how people use this tool, this evolved tool for different applications, both in basic
biology and in therapeutics. Lydia Morrison: That’s really interesting. So you mentioned that it’s great at temporal
regulation. If you wanted a more sustained regulation,
is there a mechanism by which you could keep up more steady state level and maintain the
repression or activation of a certain pathway? Omar: Yeah. So if you really wanted a long-term modulation
of these nucleotides, you have multiple options. So one is controlling how you deliver the
system. So for example, if you deliver the system
actually as a protein form, the protein will get turned over right after maybe a day or
two. And so your modulation would be quite transient
in that case. If you want something longer, you could then
start thinking about maybe viral delivery. So for example, if you go for AAV’s, those
viruses can actually stick around for years. And so if you deliver this tool using that
system, it would actually reach a steady state of editing within a cell and it would stick
around for quite a long time. And then of course, if you really want permanent
modulation, you could of course do DNA editing and use something like Cas9 or Cas12 where
you can install the mutation instead of n RNA and DNA, in which case it would be permanent. And last for forever. Lydia Morrison: So you did lots of screening
to identify Cas13 and Cas12. Can you tell us how machine learning or artificial
intelligence played into that data mining? Jonathan : Yeah, so when we actually discovered
these proteins, we went through a process of basically looking for certain anchors in
the genomes of these different bacteria. We essentially compiled all of the sequences
from thousands, tens of thousands actually, of bacteria and then looked for certain landmarks. Then near those landmarks we could find if
there were proteins and if there proteins that we knew what they were, we could obviously
say that. But if they were unknown proteins, we could
start to cluster them together and say, well there’s all these different proteins that
kind of co-occur with these landmarks like CRISPR arrays, what are they? So that kind of clustering and finding similarity
in the lining there was a little bit of, I’d say weak artificial intelligence process where
we could cluster things and look for what was similar. Then what we did is we eventually found these
clusters and then kind of look for them again and they turned out to be all the same protein
and that was the Cas12 and Cas13 but I think one thing that we’re very interested in moving
forward is using much more of the additional genomes. Jonathan : Of course now more genomes are
sequenced every year and there are hundreds of thousands of genomes available now. Using those genomes along with more sophisticated
ways of training, machine learning on what exactly does a protein look like in terms
of just sequence or certain features of secondary structure and use that to actually go into
these expanded data and look for proteins of interest, whether they be CRISPR proteins
or other potential proteins that could be used for genome editing or other applications. So I think we’re very excited about using
both a ton of new data as well as new approaches in annotating and predicting similarity of
proteins to delve into these data sets. Lydia Morrison: And how are you planning to
apply the knowledge that you’ve gained from Cas12 and Cas13 research and SHERLOCK and
RESCUE technologies? How are you planning to apply that in your
new lab at the McGovern Institute? Omar: Yeah, so I think, a lot of what we’ve
learned is how to explore natural diversity and ways evolution has already created useful
proteins and enzymes and how they exploit them for biotechnology. I think a lot of what we’re doing now is we’re
trying to maybe find more systems beyond CRISPR that could be useful for, gene editing and
gene therapy. There’s still a lot of limitations with how
CRISPR is used now in terms of efficiency of editing and even being able to insert,
large chunks of, you know, genetic material, like large genes for example, to get permanent
replacement. And so, you know, a lot of what we’re doing
is maybe, you know, there are systems beyond CRISPR that can be used for phage defense,
that could also be useful in this way. Omar: So we’re trying to mind for additional
signatures, we’re trying to characterize those enzymes in high throughput. So ways to screen them, whether it’s in vitro
or in bacteria. Then once you have them and can show that
there’s activity, how do we continue to engineer them using our engineering tool set? So whether it’s directed evolution or just
screening through mutagenesis and to make these enzymes even better than how they’ve
already evolved. And so we’re doing a lot of that. We’re also applying a lot of these tool sets
to other things beyond just proteins from bacteria. I think we kind of hinted at this, but what
you really need is ways to deliver these proteins to the right cells when you’re doing gene
therapy. And our tool set for getting cell are for
getting these tools to the right cell type or right tissue type is quite limited. Omar: And so we are applying a lot of these
mining and engineering approaches towards the viruses to try to either find new viruses
that have the properties that you want and then to engineer them to either hit the right
brain cell type or to be able to get into muscle better. Or even bone marrow and hitting T cells, OR
hematopoietic STEM cells, or whatever cell type you might want. So I think, there’s a lot of problems that
need solving and there’s hundreds of thousands of bacteria and other organisms we can try
to pull a solutions from. Lydia Morrison: I was wondering if you could
share with us your perspective on the use of CRISPR and gene editing technology in current
clinical trials and therapeutic approaches such as agriculture? Omar: Yeah, so I think, 2019 has been a big
year for CRISPR’s as two clinical trials, have begun. One from CRISPR Therapeutics another from
Editas Medicine. So I think that time will tell whether the
results are promising and if it’ll actually work. But I think, in the next few years we’ll start
to see CRISPR technologies actually start to become approved and enter the clinic. And I think for agriculture we’ll probably
see results even sooner. People are already trying to make different
types of crops either, taste better or improve yields, make them healthier. Like a lot of genome editing for like soy
beans for example, to get rid of unhealthy oils and put more healthier oils in them,
have already begun. And so I think, it’s going to be- the next
few years is going to be really exciting. Lydia Morrison: And do you have any thoughts
on the role or significance of RNA editors in addition to traditional DNA CRISPR technologies? Jonathan : Yeah, so having RNA editing as
a complimentary technology for DNA editing I think allows for a lot of additional things. I think they approach two different areas
in a sense because in DNA editing, there’s a lot of things that you can do where you
can target things and you dose it once, but there’s a lot of cases where it may be difficult
to deliver or it may be something that you don’t want permanently. So with RNA editing you have the capability
to, in some fields use endogenous proteins inside of cells like the natural ADAR and
just co-opt that for targeting. Jonathan : In our case we introduced this
protein, but I think that there’s an exciting area, especially for things that you only
may want to induce temporarily to have this capability where maybe you don’t want to run
forever. And of course in the safety aspect, with DNA
that’s great that you can do something permanently. But on the other side if you do something
permanently, you don’t want it there. That’s a little bit of an issue. With RNA, if you have off targets, they’re
not as much of an issue. So I think that will play into a little bit
of how these things are regulated. But I think the space is so large. If you think about just the medicines that
we can do that both technologies will do quite well. Lydia Morrison: Thank you both so much for
joining me today and I want to offer my congratulations on the amazing success that you’ve seen so
early in your career and I am super excited to see where your research lends itself to
diagnostics and clinical therapeutics in the future. So thank you. Omar: Great. Yeah, thanks for having us. Jonathan : Yeah, thanks. Lydia Morrison: And I want to make sure that
our listeners know where to go to learn more about your research. So could you tell them where to find your
website? Jonathan: Yeah, so our website is
Lydia Morrison: Awesome. Thanks so much guys. Thanks for listening to this episode of our
podcast. I hope you learned something about the amazing
possibilities CRISPR gene editing technology is enabling in healthcare and agriculture. As always, check out the transcript of this
podcast for lots of informative links, including the link to Omar and Jonathan’s new website. And don’t forget to tune in next time when
we’ll be joined by Rupali Dabas of University College London’s iGEM team will be turning
over the interview reins as she sits down with her colleagues and professors to discuss
public perceptions of engineering biology.

Add a Comment

Your email address will not be published. Required fields are marked *