Getting started with CRISPR: a review of gene knockout and homology-directed repair

this integrated DNA technologies webinar on getting started with Chris Berg genome editing my name is dr. has Becker I'll be serving as the moderator for today's presentation the presentation today will be given by Justin Barr justin is the product manager of functional genomics at IDT where he works closely with researchers to develop new tools for improving molecular biology research justin has a bachelor's in science from the University of Iowa with an emphasis in genetics and biotechnology and prior to joining IDT Justin was a part of a research group at the University where he studied Effects of vWF and Adam s Adam T s13 on endothelial cell dysfunction Justin's presentation today should last about 40 minutes and following that presentation will take as many questions as we can from the audience as attendees you have been muted but we encourage you to ask your questions at any time or make comments at any time during or after the presentation and you can do that by entering them into the questions box and we go to webinar control panel and you'll find that at the right hand side of your screen and there's a little down arrow you can click on that and make the window larger type of question into there and there's actually a thing too on the right-hand side of the bar and the newer interface where you can just pop that window out and type your question in makes it a little easier to type in that larger space at the end of the presentation then I will pass as many of those long to Justin as I can we will get to quite a few of them if we don't get to your question we will follow up with you by email so please you know make sure that you ask your questions and you will get a response from us also in case you need to leave early today or you want to revisit this webinar we are recording the presentation and we'll make that recording available on our vimeo channel at our youtube channel shown on the screen and it'll also be on our website we have a video library under our support and education tab so IDT DNA calm and we have a whole video library there's a host of CRISPR webinars already there that address other areas of CRISPR research and we also have a variety of other topics on NGS and qPCR and a whole host of others polakov about topics we also get a lot of questions about slide the slide deck will be posted it's actually already posted at the SlideShare site that is shown and that will also be sending you the link to all of this stuff in a follow-up email so with all of that being said I'm going to hand this over to Justin Elle and Justin can get started great thanks so much hands and thanks to everybody for joining the webinar today I'm excited to be talking about how to get started with CRISPR we'll be reviewing gene knockout and a little bit about homology directed repair as well my goal for this webinar is to really provide what I hope to be an easy introduction to designing your guide RNA for CRISPR and then also to give you some sense of what you might need if you want to jump straight into homology directed repair so if you don't have a CRISPR background already that's okay we'll touch on the basics as I go through and then as Anne's mentioned I'm happy to answer your questions at the end so to get started a quick overview of what I'd like to cover first a basic CRISPR workflow to give us some background to consider as we move through the rest of this steps and then I'll actually jump straight into planning for a knockout experiment so if your goal is to disrupt gene function and perhaps assess the impact of that gene we'll talk about designing a guide RNA to actually disrupt that gene we'll move into some discussion of delivery methods and comparing things like life affection and electroporation and then at the end I'll spend a few minutes talking about homology directed repair templates and things to consider and then how you would actually go about designing and ordering that type of repair template so this slide provides an overview of a current CRISPR workflow if you're working with CRISPR cassadine protein itself and a guide RNA that's the approach that we really like to use in our alt our system is based around delivering the calf line protein along with a chemically synthesized guide RNA now the goal here is to induce a double strand break at a specific site within the DNA of your cells the way this is accomplished is by first complexing together a guide RNA that tells the catheline protein specifically where to go and where to induce a double strand break once these two are complex they form a ribonucleoprotein complex that has a 20 base specific sequence as part of one of the two rnas that are that are used and that 20 main sequence provides the specificity to the target DNA such that ideally only in those locations a double strand break occurs forecast 9 nucleus which is what I'll be referring to today that's typically a 20 base proto spacer sequence and those 20 bases are immediately upstream of a motif ngg that's a commonly referred to as the pam site or proto spacer adjacent motif and that ngg sequence always has to be present in the target just downstream of the 20 bases target DNA so that the cast 9 molecule will notice stop and interrogate whether or not it has a 20 base match with its guide RNA assuming that it does have a match the cast 9 RMP will induce a double strand break and the wild-type version of the protein that cleavage event happens between the 17th and 18th base of the target DNA and then is a blunt ended cut so there's no overhang left once that double strand break has occurred then we either rely on a non-homologous end joining process and that's the more error-prone process we're typically an insertion or deletion event can occur or we rely on homology directed repair where a template piece of DNA is used to align and repair either as an error-free process or if you're introducing a specific mute nation that can sometimes be included in the middle of that repair template and we'll talk about that towards the end of the presentation so getting started with CRISPR if this is your first CRISPR experiment coming up it can be a bit overwhelming because we have so many different ways that you can actually implement CRISPR calf nine genome editing overall the goal is to be able to get cast nine protein a guide RNA to complex together into that ribonucleoprotein complex but getting to that point can be done through a number of different methods whether that's delivering those into the cast nine or the guide RNA in a plasmid form of DNA or as linear DNA such as a gbox gene fragment or perhaps delivering those components as mRNA or transcribed single guide RNA all of these methods can work very well on their advantages and disadvantages to all of them but to really narrow this down and simplify the process we feel very confident that the best way to deliver CRISPR reagents is just to work with the closest to finish products as you can so that means working with delivering the functional catheline protein along with a ready to functional and ready to use guide RNA this allows you to actually pre complex these two and form your calf line rather nuclear protein complex before to even deliver to the cells and so that allows you to have an active complex put it into cells and it can be immediately active to induce double strand breaks within those cells so this is the method that we'll be using as I explained further delivery methods and approaches and the rest of this webinar so taking the basic workflow for a crisper experiment I've kind of broken it down into five distinct steps here so it begins by designing your guide RNAs so figuring out specifically where in the genome you'd like to target and then how to identify which 20a sequence is what you'd like to use to try to target and induce that double strand break once you've designed those guide RNAs then you need to order and prepare your guide RNAs and complex them together with the gasline protein so you actually assemble that rather nucleoprotein complex from there you're ready to move into the delivery stage which typically is done either through life affection electroporation or microinjection into cells and then finally after the CRISPR delivery has occurred and some time has elapsed for that editing to take place the genomic DNA is collected from those samples and you move on to analysis and we'll talk about each of these steps in a little bit more detail so starting with the guide RNA design we'll take a look at that first for a gene disruption knockout type experiment where the goal here is to select a guide RNA that matches a coding sequence for gene of interest and induces a double strand break within say an axon of that gene and then we use the non-homologous end joining process and the error-prone nature of that process to create an insertion or deletion event ideally disrupting the function of the gene either through knocking the causing a frameshift mutation or perhaps introducing a stop codon prematurely so the first just actually identify what type of design tool you're interested in using if you type crisper design just into an online search you'll find that there are a lot of different tools out there and it's great to have a lot of choices because each of these has unique advantages or disadvantages and so I want to point out a couple important considerations to think about as you choose what type of tool you use for your your actual experimental design all these are good tools I don't really wouldn't avoid any of them but it's important to think about what type of species or cell line was used to set up that tool what if it's based off of an algorithm from experimental data how does that compare to the type of experiment you're doing or where all the experiments run on human cell lines and you're using a mouse or perhaps a zebrafish experiment in addition how was the cast 9 and the guide RNA actually delivered to those cells many of the algorithms developed have been based off of a lentiviral delivery system which can be very effective and great solution for high-throughput CRISPR analysis but when you switch from that type of delivery method to a DNA or direct RNA delivery of a guide or even arrive a nuclear protein delivery of cast 9 we actually see that there is a noise strong correlation between the guide RNA design and what's effective in one system versus another so the source of these components and the method for delivery can make a big big difference in how effective a particular guide sequence might be and finally all these tools will have some kind of design output and many of them provide both an off target score and an on target score corresponding to how likely this guide RNA sequence is to cut somewhere unwanted in the genome and then how likely this particular guide RNA sequence is to cut efficiently at the intended target site now sometimes you can see both scores and rank them based on either metric as you'd like and sometimes these are actually weighted together and it's important to understand what that weighting factor is so that you can decide which of these factors is most important to you for the particular experiment you're working on lastly some of the tools will also filter out guide RNA sequences either due to poly base runs or some other metric and typically how that filtering is done is pretty transparent but it's important to have some idea of how the tool is designed and whether or not it matches well with the experiment that you're trying to do so I'm going to go through two different examples take a look at two different free easy-to-use tools that are online for choosing guide RNAs the first one we'll take a look at is based off of a custom track on the UCSC genome browser so a very commonly used tool the nice thing about this is this custom track already has all the guide RNA sequences designed and you can simply look at the gene sequence you're interested in and determine if any of the particulars that are shown will fit well for what you're trying to do we'll also take a look at the CRISPR design tool from MIT this one has been around for quite some time and well it doesn't provide an on target CRISPR analysis or ranking it does a very nice job of providing a simple output for off target analysis and so this is another common tool and we'll take a walk through that that tool as well the latter tool is very convenient if you don't have a commonly used genome so the genome browser track only works for a number of species and sometimes you can use the MIT tool for an even wider number of particular species so to go ahead and jump right into some use of these tools I'll begin with the genome browser which is simply genome UCSC edu and I already have the homepage pulled up here but what I'll do is jump straight into the human genome tract we'll go ahead and pull up the latest assembly of the human genome and I should have the default view here so if you don't regularly use this genome browser your screen should look a lot like this when you pull it up for the first time lots of information much more than we need for right now so to simplify things I'm simply going to use the hide all button that eliminates all the custom tracks that have been added right now so I'll hide everything and I'm going to turn on just two different tracks the first one is the NCBI reference sequence track and the second one is this crisper track and that's the one that will actually show us all the guide RNA sequences that are possible and so now if i refresh I can see that I have both of those tracks present along with my location in this case on chromosome 7 so just as an example let's try to find a guide RNA sequence that would work well for disrupting the beta-actin gene and this human genome so we can type in act B here and we're looking for this first gene target and I can simply click go and now I see the entire beta-actin gene in this case based on the arrows it's being transcribed from right to left as I look at it and I can see both the exons and the introns of this gene so in choosing a location for your guide RNA if I'm trying to disrupt a gene I generally choose or start with exons that are earlier in that transcript just so that if I'm causing a disruption in that protein it happens earlier in the coding and is more likely to lead to a non-functional protein it just makes it a little bit easier to analyze the results and make sure you don't have a partially functional protein that isn't actually accomplishing what you're trying to do so in this case we'll take a look at this are the exon here and I can just drag a box over the steel bar and then it'll allow me to zoom in on just that exon so now you can see the CRISPR tract and a little bit better detail and what this track is showing is a color-coded scheme of all the guide RNAs that exist within this region of the genome so anywhere where there's an ngg sequence it tests the 20 bases upstream of that Pam sequence and for all of the 20 base sequences and in grayscale the specificity score is below 50 on a scale of 1 to 100 and generally that's a little bit more worrying so it's it has some off-target potential it might actually cut somewhere else in the genome and unless there are no other options I typically would avoid such a guide RNA for the guide RNAs that are colored we can see we have green yellow and there's one red one here in the corner those all have specificity scores above 50 and then the color represents the on target editing score so basically they're considered safer and then we move from higher scores for on target to kind of moderate and then this red one is a lower score or something below 30 for the on target cutting efficiency so the nice thing about this custom track is it allows you to very quickly see all the potential guide RNAs quickly determine which ones might be predicted to be better and then if you're using certain browsers I'm using Chrome right now you can actually hover your cursor over each guide RNA and it will tell you the specific scores for in this case three different algorithms that are being used the first is the MIT specificity score which is a 91 which is quite good since this is out of 100 the latter two scores are the andar to different on target metrics you can find out a little bit more about them outside of the scope of this webinar but if you click on the track name itself you can read about exactly what the difference is between the score from the 2016 bench paper or the merino mateus score so let's go ahead and move ahead with this particular guide RNA sequence I can tell it's within the exon of me to act in here and it looks like the numbers are all pretty high so this might have a good chance of working well if I click on that guide RNA sequence it brings me to a screen that gives me the entire guide RNA sequence the ng G or Pam sequence the specifics about the actual scoring algorithms for specificity and efficiency and then this is really nice it shows you the number of mismatches the count of off target sites with a given number of mismatches so I can see I have no direct copies of this exact 20 base sequence and I'm also safe from one or two base mismatches but as soon as we go to three mismatches I have 14 particular sites that exist off target and then we start to see what those off target results look like down below now you can kind of just by looking at this tell whether or not you think this is likely enough to prevent an off target editing event and in general having two or more mismatches have a pretty high likelihood of disrupting or at least severely limiting the potential for cast 9 to cut off off target particularly if you're using a ribonucleoprotein approach delivering cast 9 as a protein so this this looks like a pretty good guide RNA and I'll go ahead and copy that and we'll order this in just a moment here so we'll say this is the beta-actin guide RNA one and we'll save that so now let's take a look at the MIT tool and this is simply CRISPR MIT edu and this is just another nice alternative where you simply enter a search name which is good example for now and then you choose your sequence type whether it's a region not distinct to the genome or a unique genomic region for any of these given species so we've got a lot of options here and then after that you simply paste in the sequence one limitation to this tools it does only allow 250 bases to be submitted and it takes a little bit of time for it to actually do a alignment test of all of the potential CRISPR sites within that sequence so rather than waiting for those results I've actually pre-populated some here so that we can take a look at what that looks like after you submit as a job to this site so here we have a graphical display of the sequence that I've submitted and then arrows showing which direction each of the potential guide RNAs may be and this one split by which strand the particular guide RNA is on typically for the wild-type cast nine it doesn't matter which strand you target so in both cases it will induce a double strand break so any of these options would be good and then I'm actually going to take a look here at the specificity scores this is simply the same algorithm as what we saw on the genome browser and in this case the highest scores in 87 and as you move your cursor over these results you can also see the different off target sites so that you can very quickly get an idea of how likely they are to actually cut at an unwanted site in this case the highest risk guide RNA off target site has four mismatches at these four positions it also will tell you if you have a likelihood of actually hitting a coding sequence and what that specific gene would be so lots of great off target information here you can see that while this one's quite good at 87 if we take a look at this lower ranking score only two mismatches exist for this first potential high-risk spot so that hopefully gives you some sense of how you can use either of these tools to quickly choose a guide RNA let's go ahead and copy this guide RNA sequence and we can use that as our second example we'll see it test guide RNA too so now we have two different guides we've designed to actually order these it's quite straightforward you just hop over the IDT website choose the order menu and then we have our genome editing section on the right hand side so I'll choose CRISPR cast 9 we do also offer the CPF 1 CRISPR system that's outside of the scope of today's webinar but we do have a recent webinar that discusses the CPF 1 system in greater detail so clicking on the cast 9 link we can scroll down to the ordering section and what we're ordering now is the actual crispr RNA or CR RNA and that's the shorter of the two RNAs that actually directs the cast 9 nucleus on where to go so since these are new RNAs that we haven't tried before I'm going to order on the tuned animal scale I simply click order here and then this should look familiar if you've ordered DNA primers from IDT before so we'll copy the sequence name will copy the sequence and we'll we'll go ahead and order both of these so that's all there is to it so I'm putting it a twenty based DNA sequence not including that ngg Pam type this is the only part of the sequence that you need and then when I could continue and eventually add these to the shopping cart it will automatically convert them to no longer RNA format so it's pretty easy to go straight from the design tool to ordering in terms of the other components to the experiment the CR RNA forms an RNA duplex with a tracer RNA and that trace your RNA is a universal RNA so you can use one tracer for all of your crisper experiments so in this case with this new project coming up onto a shorter one trace your RNA there's five animals of this which is plenty to go with the tuned animals of each of those CRS I'm beginning with and then I'll need some cast nine nucleus as well order small cast nine nucleus and for now that's we'll go ahead and start with ordering those and I simply add them to the order and so that's we see that these now are in the correct for annotation and we have everything that we need so I would just move on but checking out and I have some other ones in my card here so that's how to go ahead and go through the ordering we'll hop back over to the presentation here and now that we have our guide RNAs we'll place that order and that will be on the way we'll look at assembling the ribonucleoprotein complex so this is really actually quite straightforward the two rnas arrive in standard tubes like you're used to if you've worked with our primers before it's dried RNA in a tube those are resuspended just by adding either te buffer or a duplexing buffer to that tube the tracer RNA actually comes with duplex buffer already so if you don't have a buffer at the lab you want to use that will actually come with the buffer that you can use to resuspend both all of the nucleotides once they've been resuspended to the same concentration you can mix them at a one-to-one molar ratio and then they simply go through a quick heating and cooling step to release any secondary structure and allow the CR RNA and the trace your RNA to form that intended duplex structure so then you have your proper guide RNA and you again mix that guide RNA duplex with cast 9 nucleus at a one-to-one ratio and these form that ribonucleoprotein complex very readily so that these steps don't take long at all and at that point you have your active RNP and you're ready to move on to delivery through ly perfection electroporation or micro injection all together these steps typically take less than an hour so it's a pretty quick experiment so now we're ready to actually deliver the RMP to the cells or to the samples that you're working with so I'll talk about the differences between these three common delivery methods so with liposuction this is probably the easiest of the delivery methods because you don't need an instrument to even get started you simply order the lipids we work and recommend we work with and recommend RNA IMAX I like the fact Amin RNA IMAX and have found CRISPR max to also work very well it does work for a number of cell types but unfortunately it's not compatible compatible with a full range of cell types so some more challenging cells we'll talk about with electroporation one nice thing about life affection is that you need very small amounts of the actual RNP so even the smallest scale products that we offer as the CR RNA and cast 9 and trace your RNA will last for many experiments sometimes even over 100 experiments from our small-scale options so it can be a very high throughput way to run and test a lot of experiments especially if you're working in a 96-well plate and you can assess these quickly it makes it a very inexpensive way to do a lot of CRISPR testing all at once as I mentioned not all cell types are compatible with my perfection and some cells start to show toxicity when the lipid level is too high when that happens generally we need to move on to electroporation and this is much more compatible with challenging cell types like primary cells or induced pluripotent stem cells we've used electroporation and collaborated with other users on electroporation to deliver the alt our RMP system to a very wide range of cell types including primary cells and iPS CS so there this can work very well the only kind of caveats or electroporation is that up front it does take some optimization effort to make sure you have all the conditions just right to work well for those cells so that you have good delivery and high viability of those cells the nice thing about electroporation systems particularly the nucleus action system from Lanza is that it does offer a high screw put option so you can actually electric rate an entire plate at once it does require higher amounts of the ribonucleoprotein complex though so you can't stretch your reagents as far with electroporation but it still works very well with these challenging cell types of course you have to have the electric grader to begin with so that initial investment exists unless you have a hopefully lab nearby that might let you borrow their instrument for the day the consumables can start to add up as well so those are some of the cons to the electroporation option lastly for micro injection this is of course much more model organism related so if you're interested in making a new Mouse strain or your micro injecting into zebrafish or C elegans this is typically what's used one great thing about micro injection is that you know exactly where the ribonucleoprotein is being delivered you're not relying upon the protein to find its way into the nucleus but rather you can actually micro inject directly into the nucleus and know that that genomic DNA is accessible to the protein because of that you can typically use a lower revenue clear protein amount so again this probably isn't as low as like perfection but you can have a lot of injections out of a small amount of ribonucleoprotein with microinjection it does typically require very good hands or experience having done many micro injections in the past and you need the equipment to actually do the micro injections so sometimes this can be in a more expensive method as well to go into further detail I think would take up a little bit too much time but we do have and I wanted to point out some very detailed guides for really walking through on a step by step basis delivering via live affection and electroporation to solve two different cell types these are available on the IDT crisper Tass nine website and as well as for the CPF one system and something that has been really great is we've actually been able to share some user submitted protocols where we haven't done any model organism work here in our Rd group but we've had collaborators that have decided to share their protocols with the wider community using these reagents to deliver into mice and zebrafish and C elegans so we have a new protocol coming soon for actual Mouse electroporation mouse zygote electroporation and if you use these tools and find a very efficient method that you'd like to share with the community please feel free to send us some that our way we'd love to share it and I think that helps everybody accomplish better research so now we've successfully delivered these two cells and we'll quickly talk about collecting genomic DNA and then moving on to analysis of those results of the CRISPR editing collecting DNA is pretty straightforward I'm guessing many of you have already done this this is detailed much more specifically in the protocol but in short you have a couple PBS washes of yourself life the cells we use the quick extract solution and we transfer them to a new container and vortex that material before heating them and doing a final evolution in spin down step so at that point you're ready to go with your genomic DNA and you can move on to analyzing those samples and looking for those CRISPR edits and the way that we do that is using an assay called the t7 under nucleus one assay so this enzyme recognizes mismatches in double-stranded DNA until you make take advantage of take advantage of this we begin with a PCR amplification step designing primers that flank the CRISPR target site that you're interested in so in the example on the here if the CRISPR target sites right in the middle of this segment of DNA amplify and cross that region if CRISPR editing has occurred you'll end up with some wild-type DNA that wasn't that it did or it was edited and repaired successfully without errors or you might end up with some insertions or some deletion events after the first initial rounds of PCR there's one final Heating and Cooling step which allows all these different strands to denature and then reform potentially introducing hetero duplexes right at that CRISPR target site so if a wild-type strand binds to an insertion strand or deletion strand you end up with predictable mismatches at that target site and those are the substrates for the t7 endo nucleus one enzyme to cleave that double-stranded DNA providing you with digestion products that can be run out on a gel capillary electrophoresis system or fragment analyzer and you should have predictable products where you can see a full-length band and then compare it to the amount of the digested products to have an idea of how much editing has occurred in those samples so this is a very simple assay it doesn't require a lot of extra steps other than PCR and the digestion with the t7 and a nucleus one and it's something that can really be done in high-throughput so we use this routinely have very fast results to confirm CRISPR editing has occurred to downside but I'll point out to this assay is that it this enzyme doesn't detect single base insertion deletion events so it does actually under represent the overall level of editing that occurs in those cells but it still gives you a very good relative idea of how much editing has occurred between different guide RNAs that you're testing and gives you enough of an idea of whether or not you can move on to the next steps you're interested in and the experiment to be absolutely sure of what those edits are typically Sanger sequencing or next-generation sequencing is still the best method to confirm exactly what the products of those edited cells will be so we'll spend the last few minutes here talking about homology directed repair and just as a reminder this is actually introducing a single or double stranded DNA product to actually create a specific change at that CRISPR cut site so that when the repair event happens through homology directed repair you end up with a new mutation at that site assuming that you have high levels of hgr occurring so from a preparation standpoint this looks very similar to what we saw for the initial Arn P preparation the only difference is that now you'd introduce your HDR template and the final step along with the delivery of the ribonucleoprotein complex so ultra all of the nucleotides these are our higher fidelity Elega synthesis products are commonly used as HDR templates and we'll talk about why that is but in terms of a workflow it doesn't change things very much you're simply adding in this extra all of the nucleotide does that last delivery step in your experiment so some considerations to keep in mind to begin with for HDR the size of the mutation you're trying to create really dictates what type of template you'd like you should use so for point mutations and smaller sized insertions or tags I highly recommend using single-stranded Allah goes and for the highest fidelity the ultra marala vis that IDT offers are very common choice standard defaulting tends to be sufficient for purification and the overall unit length for ultra Murs allows up to 200 paces so we typically recommend for these smaller insertions homology arms which we'll talk about in a minute that match either side of the cut site ranging between 30 to 60 nucleotides so that means you can easily fit any approximately hundred base insertion and sometimes even a little bit larger within the length of restrictions for an ultra olive a nucleotide or you might just be simply making a single point mutation and that's just fine too for even larger insertions though if you're moving beyond a 150 or larger basis typically double stranded donor material is used and the reason for that is not only the size limitations for ultra muroga nucleotides but we typically have to move to double-stranded DNA for those really large inserts and when we move to double-stranded DNA it seems that longer homology arms are needed to have high rates of HDR typically a homology arm greater than are up to or greater than 500 bases in length so either using a donor plasmid that's generated or something like a g blocks gene fragments as the donor template are typically used the last consideration it refers to positioning so wherever your desired mutation site will be ideally you'll choose a CRISPR guide RNA site that's as close to that mutation as possible and then in addition you know design your HDR template so that it really sits on top of that mutation and CRISPR cut site that seems to really drive the HDR rates as high as they could potentially be for that experiment and we'll look at some data too so that you can have some sense of what you can expect of course that's very cell type and experiment dependent as well so this was a survey experiment where HDR templates a centered on a CRISPR target site were tested with varying arm lengths as shown on the left so from just under 30 bases to slightly over 90 bases and these are centered on the CRISPR cleavage site and the insertion element here is actually in restriction enzyme site so an eco r1 site was put in with on either side of the insertion site sequence that matches the sequence of the wild-type DNA on either side of where that CRISPR RNP will cleave we'll take a look at very specific HDR template design in just a couple of slides so the goal here was to try to assess HDR by looking at whether or not cleavage through this insertion of a restriction enzyme site occurs this particular experiment used a guide RNA targeting emx one in x29 three cells it was done via lipe affection of the RNP into those cells the ultra marala goes were standard default and they were delivered at a three nano molar concentration either as double-stranded DNA or single-stranded DNA from either strand of that template so after 48 hours after delivering these to the cells the genomic DNA was isolated PCR amplified and then tested either through the t7 endonuclease 1 digestion I described earlier or through eco r1 digestion to determine how much editing overall had occurred through non-homologous end joining and how much of the editing occurred through homology directed repair where that eco r1 site was now present after HDR had happened so here's the data from that experiment and what we see is in blue this targeting strand to the Strand that the CR RNA there sorry that the RMP actually binds to in orange the non targeting strand and then in green the double-stranded DNA template on the top shown by the points are the level of editing based on the t7 endonuclease 1sa so we expect this to be higher and then the rates of homology directed repair are indicated by the bars so we range from mid to high 30s across the board on this experiment and in additional experiments that we've run we find that overall 30 to 60 nucleotide homology arm-length seems to provide quite good inconsistent results so that tends to be the range that we work with now for ongoing HDR studies when you move to longer insertions I think that the homology arm-length may vary at that point but as a general rule this is a good guideline to start with when you're designing homology arm tempura homology directed repair templates for simple and small insertions now one thing I'll mention about this is you'll notice that the double-stranded DNA equal r1 insertion seems to be quite high almost on par with the single-stranded DNA and it might be tempting to consider using a double-stranded DNA template but what we actually found was occurring in this case is that this is artificially high because of the type of incorporation that's happening with a double-stranded template and if we take a look at the fragment analysis results of both the single-stranded template and the double stranded HDR template what we see is with the single-stranded template we have our wild type product the full-length amplicon shown in the tall red bar here and then we have the two cleavage products after the ego r1 digest so as predicted we have two cleavage products and we're seeing this type of HDR occur where we have just insertion of the desired restriction enzyme site with the double-stranded DNA however we actually have a duplication of the digestion products we have our expected HDR event which matches the peaks for the single-stranded template but we also have these larger products that are the result of actually having a blunt ended insertion of this double stranded DNA product so we actually have the non-homologous end joining process being used to essentially force this donor double-stranded donor into the cut site so we have a replication of those homology arms so this is of course not what you want in most of cases and in the experiment so when you can single-stranded templates tend to avoid this issue entirely we haven't ever seen this type of product with a single-stranded HDR donor another interesting thing is as you test various sites and attempt to achieve HDR the actual rate of HDR can really vary quite widely across different sites even when keeping all other conditions the same so here's just an example of that shown with five sites where some of them have really terrific HDR rates where we're seeing even up to 50 percent HDR and these cells but in other sites we just see not very high levels of HDR occurring so this tends to be very specific to both the guide RNA the gene where you might be targeting and might even be specific to the specific HDR template so sometimes you do need to test more than one site and we'll take a look at designing the actual HDR guides here so taking a look on a sequence level of what an HDR repair template will look like on the top we have a wild-type sequence and in this example we're looking at changing this particular codon into a stop codon to try to disrupt this particular sequence from being translated successfully so we're making two changes shown in red here and and the way that we'd want to get started is to first find all of the Pam sites all the ng GS and the sequence so we can see we have four particular Pam sites and in this example we'll pick the closest one to where we're trying to introduce the actual mutation so that gives us this particular protospace your sequence just upstream of that Pam site and as I mentioned or the cast 9 nucleus will cut between bases 17 and 18 of that proto spacer so here's where we'll actually cut and now we just need to design the HDR template and the thing that makes this really easy is the HDR template should simply be the desired mutation in the sequence that you'd like to have at the end so in this case I can quite literally copy the sequence from why I want as my desired end product and in this case order this as a single-stranded oligonucleotide with 30 to 60 bases upstream and 30 to 60 bases downstream and my desired mutation I'm showing just one strand here but both strands could be tested and as we saw in certain from site to site sometimes the sense or the anti sense strand will or work better one thing to note though and one last consideration on designing that HDR template is it's important to think about after that HDR has occurred whether or not your guide RNA sequence as a risk of re cutting at that site that you've mutated now in this case we have two mutations that have been introduced and they're very close to the actual pam site itself so that should be quite disruptive but it's not a guarantee that this won't be cut again even this even with these mismatches here so if you're concerned about that you can always introduce other silent mutations either within the proto spacer region itself or even within the pam sequence so that's something to consider how your final HDR and template will affect re cutting events to take a look at an example with a larger insert in this case we're looking at putting in 33 bases between the two bases and red here of the wild-type sequence so our desired edit looks like this we're putting in 33 bases in the middle and again we'd find all of the Pam sequences still choose the closest one in this example you could always test a couple of them and here's where we're going to cut it's not right on top of the see once but that's still okay and then our HDR template will again look like our finished product that we want to have 30 to 60 basis to 33 nucleotide injured and then thirty to sixty basis downstream in this example we have much more protection against tree cutting events because it's very unlikely that this insert will match these final five bases of the proto spacer and then again it may not even have a ngg pam site in this example so this one wouldn't be so much of a worry but it is something I like to point out to make sure that you can avoid those that risk of recutting event in terms of synthesis options for HDR templates you could use a standard Oliver nucleotide just like you're ordering a primer and variety to you that's offered up to 100 basis and length the ultra Murali the nucleotides it's a slower synthesis process we have a higher fidelity and those allow up to 200 bases and links and one thing I'm excited to briefly mention this is an unreleased product but we are planning to actually start offering very long single-stranded DNA fragments that will actually go up to 2,000 bases in length so this is an early beta testing phase but if that's something that you're interested in and the ultra alga nucleotides are a length restriction for you I do have a sign-up URL here that you can either write down now or collect from the slides later and that final allow you to get information as we move further in towards releasing that product so in summary CRISPR guide RNA Designs has really become simple with a lot of free online tools I don't have a specific recommendation for which tool you should use but I do recommend understanding how those tools work what the rankings are based upon and whether or not they relate well to the type of experiment that you're doing of course efficient ribonucleoprotein delivery can achieved through any of the delivery methods that we discussed and there are optimized protocols for all of these and many applications on the IDT website and then if you do decide to try the mantra directed repair it's a great way to introduce simple mutations and insertions and it in your at your target CRISPR site the repair templates are pretty easy to design and they're synthesized quickly for your experiment I want to point out that our research and development team is really responsible for developing the tools for the altar system that we have today they've worked very hard on optimizing the system and making the best guide RNA system that we could it's optimized both in length and chemically modified and we really think that it's a really great way to go so hopefully you'll give it a try and I really appreciate their efforts some additional resources and support are available online so we have protocols we have the user methods that I described and some frequently asked questions and just visit our website and go to the CRISPR cast lines section and check out that support tab if you have questions that aren't addressed there we have a really great scientific application support group and these are scientists from the bench that have actually a lot of experiments experience with a lot of our tools that we offer and they can provide suggestions on how to really optimize your experiments you can also visit us at IDT DNA comm forward slash contact us for region specific phone contacts if you'd rather give us a call and with that I really appreciate everybody's time and attention my pleasure to be able to give this webinar today and I'm happy to answer any questions all right so it's not very much Justin for that presentation um if you have an re ask a question still type it into the questions box at the right hand side of your screen we've already got some of those we're not going to make it through all of them in this portion but we can respond to you individually by email as well so you will get an answer I want to direct you quickly to the chat box there's a link there to both the slides and to the the beta testing site for the longer single-stranded DNA that Justin had mentioned so we'll squeeze in a few questions here I know we're at the top of the hour if you do need to go that's fine like I said we'll be following up with you by email and you'll get links to the webinar recording and the flight deck all right so let's just get started here Justin there were a couple of questions actually about stability of the guide RNAs and the ribonucleoprotein complex itself so the two rnas come together at that a short double stranded region between the two of them knew Neil those and there were questions about how stable is that when you go to transect it or electorate that and then also is that more stable when it's incorporated into ribonucleoprotein complex you have any comments on that sure yeah that's that's a really great question so there are this correct there's a short region of homology between the CR RNA and the tracer RNA that readily forms a duplex as long as there's sufficient salt content present in that buffer so that's why we recommend rhesus pending both of those RNAs and either a te buffer or what we use is a duplexing bhaskar to keep the potassium acetate buffer and with the sufficient salt content those two are that interaction is thermodynamically favorable and then it is actually further stabilized once it has complex with the cast nine protein so that complete cast line RNP once that complex is prepared it's actually quite stable and we've actually done it number of tests with you know freeze-thaw cycles and dilution of these are NPS and found that it can actually last for quite a while so that's one convenient thing for preparing your experiments is you can actually prepare the full iron P and keep it in the refrigerator for easily up to a week at a time so that you don't have to worry about scrambling on the day of an experiment to get all these things ready if you want to store them longer-term we actually recommend once the protein is diluted to store it at minus eighty so that it freezes very quickly but it does remain active after freezing the only thing we wouldn't recommend in terms of long term stability is putting it at minus twenty without that glycerol content present just because we don't want it to freeze slowly and risk deforming or disrupting the protein itself okay you actually jumped ahead to my other question about storing the arm piece oh that was great oh the next thing that I have here for you Justin is a couple of questions about the design tools that you showed so um in the crit and the MIT design tool do you know why there are two different does zebrafish genome options there is a Dan that is one this is Dan rear so Dan danio rerio yeah then there's another one that's a GRC z10 yes so those are just different builds of the actual genome for the zebrafish assembly so the danio rerio seven is an earlier assembly and I believe the reference genome 10 is the most recent zebrafish build I'm a little rusty on that specific species but I think that that's just a difference between which reference assemblies is going to be scanned when just looking for those off target effects okay and then this is an interesting question working backwards a little bit if somebody finds a published guide RNA and they want to come back and see like what off-target tools something like this might predict is there a way to do that yeah that's a great question and actually I really like that you can do that with the MIT tool so the one thing that you will need I've been able to do this by putting in that 20 base sequence and the Pam itself so you'll need the ng G or whatever the N is so if you have your 20 base sequence so let's say we put this one back in here and then you know what the ng G is I can't remember if it was c GG here but this should be enough where you could actually submit this choose the correct target genome and it will look at the sequence identify that there's should only be yes only one it can only fit one particular target within here and it should still give you that same output so I love this as I see a pretty cool question actually did not know the answer to that yeah so I don't know if this will run fast enough but we should should be able to see that in action here in the mean times we're not just watching that takeaway for Clemens to move on to the next thing again regarding the MIT tool does the MIT score for the gr for the guide RNA only mean less off target effects are is there any prediction in there about the efficiency that somebody might expect from that guide RNA yeah that's correct so that a might this particular this tool does only show you off target rankings it doesn't provide any kind of on target analysis it it's on target of in particular the one that I think is sensitive to how that on target ranking was built and so I think that that's where more variability comes into play largely from what we have seen taking a look at the research that has come out of IDT's R&D group is that it's quite challenging to successfully predict how efficient a guide RNA sequence will be there are a lot of guide RNAs that I've seen the tools predict to be bad that end up being good and some that are predicted to be good that are bad and so I often when doing my own designs tend to more heavily weight the off-target risk but that's really a personal experimental decision I don't tell anyone how to do that too much but I wouldn't always rule out an experiment or a guide RNA because in the on target efficiency appears to be low it might still prove to be an effective site okay um all right the next question I have got a couple of delivery questions here for you so for homology directed repair you have your HDR template if you add that to the transaction or electroporation or whatever of your your ribonucleoprotein or guide RNA cation whatever form you chosen to do that in does that extra molecule impact the effectiveness of the transaction itself that's a really interesting question and and what we've seen is for electroporation delivery in particular the presence of that extra DNA molecule can actually help improve the rate of delivery so we've actually created and I didn't mention this during that presentation today but we actually offer a separate specific carrier DNA molecule called the electroporation enhancer for CRISPR cast 9 and for CP f1 and this is actually a long single-stranded DNA that was designed to avoid homology with human mouse and rat genomes and its only purpose is to actually help improve the rate at which ribonucleoprotein complex gets into the cell via electroporation so a homology directed repair template can actually act much in the same way except here it actually has a secondary purpose of also mediating homology directed repair sure well this next question is interesting I seen publications that talk about things like this but I don't know if we have any experience with this if you want to make two cuts in your genomic sequence of two different pam sites how far apart do those two sites need to be in order for there to be space for the RMP to act on both of them if that's a really good question and that's something that we started exploring a little bit as we've looked into multiplexing ribonucleoprotein it does seem like the footprint of the Cassadine molecule needs a sufficient amount of space to be able to make to proximal cuts and it also depends upon whether or not the to guide RNAs are on the same strand or on opposing strands so if we think about the worst case scenario where it's to guide RNAs on the same strand and so the orientation of the both RMPs are on the same strand I would recommend at least 20 if not moving to maybe 30 or more bases apart from one another we're still studying that and I can hopefully have that included in some future online material but it does seem like there is a it's a minimum of about a 20 base base necessity to make sure that both R and P molecules can can fit into that space and induce the break okay on the next question that I have is how long is the cast nine active inside the cells after you do the electrician's that's a good question and it's going to depend a bit upon the cell type we have done some testing and tried to kind of track the cast 9 after delivery it seems like four cells that have high turnover typically cast nine stops being detectable by Western by around between three to seven days it's kind of a long time frame of course it's very cell type dependent but it in terms of lifespan or for delivering reagent it's pretty short most of the editing happens within 24 or even 48 hours we see a plateau of the actual editing events and the cast line last comfortably that long and not much longer so it's delivering cast line as a ribonucleoprotein complex is really the perfect solution in that it's long they're long enough to do what it needs to do but not enough time to kind of cause trouble and stick around for too long that's really why we don't prefer the DNA delivery approach where you end up with Castle and sticking around longer than you'd like okay one more question for you is there a is there a good method for suppressing the mechanism for non-homologous end joining in order to increase homology directed repair efficiency there are a lot of different methods that have been discussed in the literature the most common of which is probably this cr7 small molecule which they kind of think of it might be a ligase inhibitor but the verdict from what I've seen is still out on whether or not there's real one one real runaway winner at suppressing a non-homologous end joining a lot of things have been tested and that's certainly something that we're interested in to help boost the rates of HDR but it's not something that we've have a clear answer on yet so I think I think there are certainly some options out there I think we haven't yet determined what other negative impacts or other risks might come along with some of those not on August enduring inhibitors but yes things things certainly exist and and the most common thing is looking for small molecules that inhibit nhej a okay so we still have quite a few questions to go and we are about 10 minutes past your time and if you quite a few people are leaving now so if we haven't responded your question we still will respond and I really appreciate you participating in the discussion and asking these great questions so you'll definitely be hearing from us we'll also be sending you the link to the recorded webinar like I said the links to the slides and to that beta testing site kriby for the longer DNA single-stranded DNA are in the chat box right now but we'll send your link stuff also in that email so you can have access to that later you don't have to grab everything now so with that being said this was really great Justin thank you very much thanks everyone for participating we definitely be coming back in the near future with more CRISPR webinars as well as other molecular biology topics I think certainly we'll be talking more about homology directed repair in the future especially with those longer single-stranded pieces of DNA coming in the future so look forward to those things anything else Justin nope that's it it was a pleasure and thanks everyone for joining the webinar today thank you everyone very much have a good day

Channel: Integrated DNA Technologies

Views: 7,277

Rating: 5 out of 5

Keywords: CRISPR, genome editing, gene knockout, homology-directed repair, HDR, gBlocks, RNA design

Id: OLyjOyeBMUo

Length: 70min 8sec (4208 seconds)

Published: Thu Mar 16 2017