Episode 5: Biomolecule purification, advances in chromatography, and automating lab workflows with robotic liquid handlers (with Dr. Patrick Kates and Dr. Todd Mullis, IMCS)

Andrew Lee: [00:00:00] And hello, I’m Andrew, co founder and chief scientific officer at IMCS. Welcome to our podcast series, Imagine More Create Solutions, where we talk about all sciences. Today, our discussion will be related to biomolecule purifications and microchromatography and the use of microchromatography on automation platforms. IMCS manufactures and provides microchromatography consumables under the brand name IMCStips and this consumable or this tool is used across different automated liquid handlers: Hamiltons, Dynamic Devices some semi automated systems like Integra, and these are used for biomolecule purifications, whether it encompasses protein purifications, nucleic acid purifications. And we’ll cover some of those topics here with our 2 scientists from IMCS, Patrick Kates, our principal scientist, and Todd Mullis, our research scientists leading the forefront of IMICS tips, will discuss the past few years worth of research using our product on various liquid [00:01:00] handlers.

Patrick Kates: Thanks, Andrew. My name is Patrick Kates, Principal Scientist at IMCS. And as Andrew alluded to, I mostly work on IMCS tips, which are microchromatography columns that are in tip form and are adaptable to automated liquid handlers. And most of what I do is on the research and development side of IMCS tips using new resins or commercially available resins for Addressing customers purification needs. , My background is, I did my doctorate work at Princeton in biophysical chemistry with a focus on inorganic molecules where I mostly interrogated mechanisms of enzymes or small molecule catalysts. So I think that kind of trended me toward. New [00:02:00] technologies and understanding how we can fix problems and new technologies or apply new technologies to fixing problems.

So that’s all for me. So, with me is Todd. Thanks, Patrick.

Todd Mullis: My name is Todd Mullis. I am a research scientist here at IMCS, and I’ve been with the company now for coming up on nine years, and my primary role is on the apps team, managing the apps team. And the way that I talk about that is that essentially when customers are interested in using the tips, I kind of handled the technical side all the way from the initial engagement where we’re getting the overview of the application and needs from the customer to the point of delivering the scripts on the various robotic systems that we use with the I mix tips. [00:03:00] My background is I got my doctorate degree at USC, mostly focused on mass spec based proteomics, specifically phosphoproteomics and developing a high throughput workflow using the Hamilton and Integra systems for global phosphoproteomic work, particularly in developing an assay for early cancer detection for colorectal cancer.

Andrew Lee: That’s great. So, you know, our focus here isn’t gonna cover all of the protein purifications. We’re just gonna take some of these highlights, and again, the highlight will be on two of the recent publications that we had. But let’s cover some generals defining protein purifications, and its importance in the technical field here [00:04:00] you know, protein purification has been around for a long time and it’s been important for any enzyme characterizations or recombinant protein characterizations, but more over a lot of the biotherapeutics. Like, antibody driven those purification strategies and its subsequent characterizations have been pretty crucial for antibody characterizations.

I mean, they rely on protein a, or these affinity purification methods. For some of the recombinant proteins, you know, they do these affinity tags with the immobilized metal affinity chromatography and you do a poly histidine tag and you do your nickel purifications or cobalt purifications. So, these are kind of common purification using the resins.

But are there any other common techniques that you guys are familiar with? Let’s go with Todd. In [00:05:00] your background with the phosphoproteins or phosphoproteomics, I mean, what were some other common techniques that you were playing around with?

Todd Mullis: Yeah, so at least on a lot of them, the mass spec side a peptide desalting step prior to mass spec injection is really common, just using a reverse phase.

Most commonly C 18 for the phospho peptide workflow. It’s important to enrich for phosphopeptides. That’s commonly done with some sort of metal base. So initially it was titanium dioxide. it was very, very common in either spin tubes or magnetic beads, and with my work with the IMCS tips we moved towards a a titanium base, but it was a custom developed resin by Tymora Analytical and Over the [00:06:00] years that the metal base that’s used there, it has shifted that binds to the fossil groups on those peptides.

But in general, you need that enrichment because of the low. Level of phosphopeptides in biological samples.

Andrew Lee: Yeah, exactly 1 of the key points that you’ve really nailed on the head is the levels of a certain target or protein of interest is at a very low level and you need various techniques to.

Enrich it and to get a cleaner version of it, or purified version of it for various downstream analysis. In Todd’s case, he was talking about mass spectrometry, but for a variety of applications, or even in our. In house applications, enzyme characterizations we can’t have contaminating extraneous enzymes making spurious results.

So affinity purification is your initial go to for the [00:07:00] protein purifications, but there must be other techniques out there, depending on the protein property, Patrick, I mean, you’ve written quite a bit around this too. So do you mind covering some of those concepts?

Patrick Kates: Sure.

And it’s not just about proteins, right? So as Todd was talking about, there’s phosphor peptides, but there’s also things like nucleic acids and for. Those you have the traditional spin column type purification, where the DNA is essentially precipitated out onto silica or onto a silica membrane. And then there’s also things like ethanol precipitation or recrystallization of small molecules that can be used to purify a molecule of interest out from contaminating molecules or cellular, but yeah, there are plenty of different tags that you can apply to proteins for affinity. Or there’s also ion exchange where you’re [00:08:00] essentially manipulating the fact that there’s a charge group on the proteins. And there are varying degrees of charge and then rinsing those off with, salt or rinsing the contaminants off with salt and then purifying off the protein with a higher level of salt.

There’s also buffer exchange or size exclusion chromatography where you select for a protein of interest based on its size or its uh, essentially globular size. Um, and and we actually use that in house. And that’s one of our commercial products in the form of the size X tips, which essentially separate out small molecules from large molecules.

In this case, small molecules being salt, allowing the end user to put their protein or biomolecule of interest into a buffer of interest.

Andrew Lee: [00:09:00] You know, I think 1 of the challenge with the purification is that there’s so many tools out there and so many different protein and so many different protein properties or biomolecule properties after covered that.

Plasma DNA, it’s absorption to based on the ionic strength and the solvents, you can take advantage of those physical properties to selectively enrich certain targets of the bottle. So.

Starting with Patrick, what sort of references did you start off with when you were doing your protein purifications? How did you obtain your background and learning? I guess the general general concepts and some of the key phrases around protein purification. I

Patrick Kates: guess my background on this topic was mostly.

Chemistry and biochemistry courses in undergrad I went to University of [00:10:00] Maryland and then I did some work in an undergraduate lab where I did protein purification on some recombinant gro-EL and gro-ES proteins. And there, it was actually, this is a long time ago. So it was kind of important to do things without an affinity tag because we were interested in crystallizing them and, So we had to be able to use the P.

I. of the protein. And do ionic ion exchange both cation and anion exchange to purify the protein in its constitutive state rather than with an affinity tag. So that kind of taught me the basics of protein chromatography. And then really the development of all of. The modern resins has has kind of made it more [00:11:00] accessible to, to the end user.

I think I don’t think you need an extensive background to be able to understand that, you know, the, the resin has a binding site that is made specifically for a target of interest and your target of interest. You can attach to essentially any protein that you want recombinantly. And here’s how you find it.

And here’s how you will elute it. And everything else that doesn’t bind to it is being washed away. The same concept of chromatography that you learned in school can be applied to all the different purification techniques that we’re going to talk about today.

Andrew Lee: You mentioned the keyword PI. That’s not principal investigator, but the ISO electric point, right? The log value of the ISO electric points. And that. Depends on the charge density of the protein, [00:12:00] given a certain pH condition, and you can take advantage of that isoelectric point of the protein and it’s charged state to bind to an opposite charge of resin so a positively charged resin will bind to negatively charged uh proteins kind of very intuitive, self explanatory. But if you understand, it sounds like if you understand the fundamentals of chromatography, that will help you segue into protein purifications and different approaches that are out there in protein purifications or any biomolecule purifications. So, Todd, what got you down this road of protein

Todd Mullis: purification?

So in undergrad, I worked in a large organic chemistry lab where we were running columns by hand to kind of purify in a low throughput way large organic molecules. And so that was my [00:13:00] first interaction with chromatography and the basics of. As you apply a biomolecule over some sort of stationary phase, you have different separation times.

And so that that was my first interaction with it. And then when I came on with I was working in our production team and we use various protein purification methods on that team. And that gave me a really good introduction or further I guess, introduction into protein purification with affinity, but also buffer exchange and my manager and I work together to do some optimization on watch buffer loading buffers and elution buffers to ensure that not only where we’re getting a, a concentrated final product, but also a a really [00:14:00] pure final product.

And so that was my kind of for a into chromatography and then then transitioning over onto the I mix upside. That’s when I really had to dive deep into various chromatography. Techniques and really TIVA, I guess GE at the time was a really helpful resource to give an overview of a lot of the different chromatography techniques the ones that we talked about.

And, and really as I went further into my research with IMCS, since a lot of those. We had to learn how some of those techniques translate into the I’m except, but some of them we had to take different approaches to ensure that we were getting not only a, again, a concentrated protein. But also ensuring that we were getting good purification [00:15:00] away from the contaminant.

Andrew Lee: It’s interesting that you brought up because it’s

probably about 170 page general content on this exact theme of fundamentals of purification and it’s. Troubleshooting techniques and the general concepts. So, I think that 1 probably has the better background than what we can cover within 30 minutes or so. 1 of the key things I want to start to segue into is the different techniques for verification.

And I think we covered. Based on the resins that you select, you can use sort of different chromatography techniques, but then there are selective approaches for particular patients. So, selective solubility you can take a minimum sulfate at different concentrations, and based on the selective solubility of the protein target, you can actually precipitate out certain proteins with the money of sulfate or salts or [00:16:00] pad.

And then you’ve got spin columns, mag beads. But we here are talking about this tip based purification. How is that any different than the other techniques that are out there? Patrick, you want to take a stab at that 1st.

Patrick Kates: Yeah, so the tips are unique in that they aren’t a packed column bed. So you can think of a resin or a, bead mixing with a solution of protein, and it can be done in a couple of ways. You can either pass liquid over a packed resin bed or a set of beads, or it can be mixed on a tabletop with a free flowing set of resin or beads.

In this case, the IMCS tips are a loosely packed resin bed. And what that means is that when you aspirate the [00:17:00] particles, the resin particles get pulled up along with the solution and distributed. Now, the benefit of this. Is that this is what we call turbulent mixing the benefit being that the surface area of the resin is fully exposed.

So there’s no issue, especially when you dispense down of. Channeling or the idea that liquid can pass through the resin bed without touching any of the resin. And that can happen in a poorly packed column. Now, the benefit of column based or packed bed chromatography is that you can essentially put more pressure and push liquid into the interior pores of the resin of interest.

So, in effect. You can get higher binding capacities in a shorter amount of time. [00:18:00] In this case, the IMCS tips more mirror the resin being mixed on a tabletop mixer. So, floating around in solution and being able to bind or what people call is equilibration type chromatography. This is typically going to be sufficient to reach the amounts of material that are needed to be purified by the end user. And without. The inconsistencies provided by something like a packed bed column that is not packed correctly.

The other benefit that I’m X tips have that makes them unique is there’s a filter. At the bottom of the tip, and so essentially you can carry the resin from step to step and so you can go and you can equilibrate your resin and then you can. Pick the tip up. And in doing so, you’ve dispensed all the liquid from the equilibration step and [00:19:00] you can go to the next.

Well, which might be your sample. You can mix it dispense all of your sample and then move over to the next. Well, wash, et cetera, et cetera to illusion. And the benefit of doing this is that you can do your entire purification in 1 plate. Or 96 samples at a time. Whereas on a typical column system like an FPLC, you can only pass through your equilibration, your sample, your wash, one at a time, over a column.

And so. You’re going to have to sequentially do your purifications, whereas here you can parallel process the purifications.

Andrew Lee: Yeah, Patrick. I mean, that was really good coverage on the technology and some of the advantages and comparison to existing technologies out there. So, these new techniques and advances and [00:20:00] biomolecule purifications previously, I remember, like, you kind of covered it. Patrick is pouring your own columns. I remember running silica columns.

for small molecule purifications and then even using some other augerous resins for protein purifications using charge dependence or charge characteristics to actually do ion exchange chromatography. But nowadays there’s a lot more resins out there that Can be leveraged to do a variety of different chromatography techniques.

But I think that raises another challenge is that you now have the techniques to express so many proteins and to screen a lot more variations of proteins, but the purification techniques that are out there isn’t really conducive for these high throughput workflows. So. Maybe based on your experience highlight some breakthroughs of different purification methods or the techniques that are out there and different modalities and [00:21:00] maybe some pros and cons for these different modalities.

Patrick.

Patrick Kates: So I think it’s. Best to start with, as you said, the packed columns we probably all had to do this at some point as an undergrad and those pack columns could be for gravity flow. So you’re essentially controlling the rate of liquid passing through your column by. Gravity and those of you both and organic and biomolecule purification know that this can be incredibly painful and long.

But then there are also pressurized column systems. So, and I think that was a pretty big step forward. And that’s the idea of. Applying a pressure onto the top of the column, pushing the liquid through under pressure, allowing the solution to be pushed into the interior pores and [00:22:00] allowing for faster particle transfer.

Now, the downside to are they are a bit finicky. If you’re not properly or well trained, you can dry columns to the point where they are no longer serviceable. And then also, there’s still only 1 sample at a time. And some of the steps, even if it’s a simple desalting step may take a very large column that really doesn’t, you know, it doesn’t move very fast.

So it might take anywhere from, 10 to 20 minutes for a PD 10 column, or it might take 3 hours if it’s a sizable size exclusion column. So there’s a little bit of a throughput issue when it comes to usage. And then I guess when you’re looking at. A bunch of small samples that you have to purify.

At the same time, there are a couple of different options. You have you [00:23:00] have I’m X tips as I talked about previously, but there are also mag be purifications and spin plates and spin plates can come in a couple of different forms. They can either be, resin based. So resin stacked into a spin plate or membrane based.

So the bottom of the spin plate can be coded in a membrane that allows fast particle transfer. And typically the downside for filter membrane spin plate would be capacity. You’re limited by the amount of layers of that membrane that they’ve been able to coat onto the bottom of a plate. In general, we tend to find that the spin plate membranes transfer the material very quickly but capacity tends to be the thing you run up against and then spin plates that are filled with resin kind of Can [00:24:00] be Used in 2 ways, they can be used as a centrifugal spin plate, or they can be used as a positive pressure manifold, essentially mimicking a, a set of pushing down positive pressure through a resin bed.

Unfortunately, for spin plates, the resin beds are not packed as well as they are in F PLCs. And because of this, there tends to be a lot of channeling and the purifications are not nearly as efficient as they are on the FPLC. And then the final, I guess, purification technique for large scale or a bunch of samples would be mag beads.

And mag beads have become quite adaptable to any sort of workflow. Any of the resins that you’re able to find, you’re usually able to find in a mag bead format. And the idea here is that you can [00:25:00] apply an external magnet to essentially keep the mag beads in place in this case, either in a plate, or on the edge of something some vessel to be used to transfer them to the next step. And one of the issues we see with mag beads is capacity issues because unlike resins, mag beads are not porous. And so all of the either affinity particles or charged particles are coating the outside of the mag bead. And so essentially what you see in terms of surface area is what you get in terms of purification.

And there is also the issue of actual handling of the mag beads is not the most desirable platform. And so yeah, I think in terms of emerging technologies and the different technologies out there for purification, you have the FPLC, you have spin plates in both. [00:26:00] Column and membrane form, and you have magnets and then that brings us to the tip based technologies, which can be used for myriad applications, which is, I guess.

What we’re talking about today,

so

Andrew Lee: the, you know, the new technologies that are out there and chromatography, I think it’s just so many that are developing in recent years, actually, not only the techniques. So, Patrick, thanks for covering the technique side of things, you know, spin columns, batch extractions, which is equilibrium driven or Mac beads and then even within the tip world.

There’s actually a couple of different. Platforms within the tips, you have your pack beds or your loose containing residents and there’s pros and cons to all of this. And, of course, we’ll link our paper back in 2020 about techniques paper that covers these concepts.

So I wanna go back to this use of tip based technology to [00:27:00] streamline the automation side. One other disadvantage might be on the tips or even on the mag beats, especially for high throughput, is that you are using larger volumes of samples. So five mil samples. Most automation systems are in a 96 array, eight by 12, and they’re capped at about one or two mils of.

Liquid sample. However, if you kind of array that in a 24 well block, you can increase that sample size to 4 mils, 5 mils and given the size of a pipette tip at 1 mil, you’re sort of limited on how you can do your purifications. Especially using a pipette, but Todd, you’ve actually done some recent work.

That was quite interesting on how you can actually overcome some of these challenges with large volume sample preparations, right? Where you have your target protein or I think it was a protein base. We’re not going to go too much into detail due to [00:28:00] some sort of confidentiality. But again, we’re going to go through the general concepts of the workflow that we’ve been working on for.

I’d say about over 5 years, how do you get large volumes of samples into a 1 mil purification concentrated and diluted? So, do you want to kind of cover that kind of workflow? And then, of course, the key aspect of that purification strategy was also coupling it to a size exclusion chromatography to get your desalting and your buffer exchanged to the target or to get to the buffer of choice.

Todd Mullis: So 1 of the things as we have gained more experience in the protein purification area is that we’re running into situations where where customers had. Like you were saying, large volumes and low concentrations of antibody samples. And so this was a challenge [00:29:00] because, as you said we could purify 24 samples at a time using, you know, multiple tips or or 1 tip but there’s challenges there in terms of throughput.

And in the case of where you’re using multiple tips in in 1 24 well plate you deal with concentration issues where you have to allude over those 2 to 4 tips. And so you have a more dilute sample, which when you’re starting with an already diluted or low concentration, there’s just challenges when you have to meet a set concentration on the back end.

So one of the approaches that we went with was we took that five or six mil sample and split it over 2 to 3 96 well plates so that one, we could get the throughput needs for customers who are wanting to run not hundreds, but even thousands of antibody samples and. Instead of going [00:30:00] based on just a kind of binding capacity, what we, we decided to use even more resin so that in those low concentration samples, we could find the sample really quickly, and so another thing that we learned is that as we moved over those plates as we were doing the binding steps, we actually needed to increase the mixing cycles for later. The 2 and 3 plate because what we found is that the binding was happening really fast in the 1st place.

But slower in the 2nd and 3rd plate. And so we made that adjustment in the, in the scripting and that led to 1, a really quick workflow. And it met the throughput challenges that we had so that we could. Purify 96 samples at a time rather than 24 at a time. And then we followed that up with a buffer exchange, which is a really nice workflow [00:31:00] that takes around 20 minutes.

That allows us to really quickly buffer exchange the samples out of the acidic condition. And so in that case, so as to limit the amount of time that those antibody samples are exposed to the. Acidic condition, we would pre equilibrate the size X tips and then immediately following the elution using the protein a tips we would immediately load that.

Into the size, except so that within a few minutes you have those buffer exchange, purified protein samples, that you can then move on to the downstream analysis

Andrew Lee: seems like a very complicated workflow. I mean, even though the concept is pretty general, right? You’re taking a large sample volume, large number of samples. Thousands of samples, [00:32:00] each at 4 or 5 mils and running these purifications. I mean, imagine if you’re a grad student and the professor requests that you do these library arrays and start purifying it.

I don’t think you could graduate in 5, 6 years. The key concept here is. high throughput biomolecule purifications using different strategies. And you can only achieve this by understanding the fundamentals of chromatography and then what sort of strategies you’re going to use for the, for the baseline purifications.

So in the primary application that Tom was talking about is probably an affinity purification coupled to a size exclusion for buffer exchange. And that’s been pretty popular across the board for high throughput automated workflows. And I think that’s a pretty amazing workflow, especially if you can actually couple those on an automated liquid hand.

So you’re churning out a lot of purified proteins and [00:33:00] in a buffer of choice in a more stable format. And that’s compatible with your downstream analytics. So I’m going to sort of pivot back to more into what other challenges do you see in these kind of high throughput protein purifications? Is it more on the automation programming side?

Or is it more on the protein knowledge? What sort of challenges? Do you see when you’re working with projects or customers and we’ll start with Todd.

Todd Mullis: Yeah, I would say one of the, I guess two most common situations is that we’re working with a scientist who really needs to to ramp up their purification throughput, but may not have any experience on a automation system. And so it’s really intimidating. If they don’t have a outside automation team to approach [00:34:00] those systems and begin to try to learn the programming and then implement a method on that system, uh, and then the alternative side of that is that you have someone who is on the automation team who has been tasked with with automating a workflow, but may not have the kind of biochemistry knowledge of what certain steps impact the purification, including the yield purity and what sort of downstream analytics are paired with the purification. Process and so that’s kind of the most common two scenarios that come up and that’s where at least my team is able to step in.

We have both the expertise on the automation side and the protein purification side to help bridge that gap between the automation and the purification.

Andrew Lee: So, Todd, you actually point out a very interesting point, and it’s cross boundary knowledge. So, a lot of the biochemists, they’re familiar with their proteins and their [00:35:00] end goals, their research theme, what sort of purity levels of protein do they need? How do they characterize it? How do they purify the protein?

However, they’re not familiar with the coding or the automation side of things, and that can be quite scary. Thank It’s a daunting task to engage into this engineering practice of taking a very expensive automated liquid handler and trying to implement your purification. Initial engagement might actually be very reluctant because again, that sort of programming takes a lot of effort, a lot of energy to do something very simple that you can, you can kind of do by hand initially.

However, if you think really into this the automation and throughput takes a lot of effort to get it consistent. I’m going to switch over to Patrick. I mean, he probably has some experience in house of trying to implement an automated workflow and to compare it against different platforms. And some of the challenges that [00:36:00] he and his team, his staff scientists have faced.

Patrick Kates: Yeah, and I, I think that this kind of will blend itself into. Todd statement, right? Because we’re also interested in, I guess, the basic tenets of what is controlling this interaction and what can we expect from the interaction? And we need to know those questions so that we can relate them to our customers.

And 1 of the things Todd talked about was in his 1st binding step Of the 3 binding steps was he observed that it bound really quickly and then it bound less quickly in a 2nd and then even less quickly in the 3rd. So he had to increase time. And the reasoning behind this is actually talked about a little in that biotechniques paper where some of the initial research I did was whether we can look at the kinetics of [00:37:00] binding or the affinity of binding the uh, and so the experiment was to basically take different concentrations of GFP and bind them to in this case, it was a GFP that was tagged with a his tag and to bind them to a nickel resin. And. Do this over different amounts of time and then look at the rate of binding and by doing so you can get what is essentially a K on rate and then by leaving it on the resin and monitoring it over time, you can get a K off rate and the cool thing about this is you can essentially get an affinity constant from the K on over K off and that’s important because you have a defined affinity constant for your residents of interest.

To the analyte of interest, and that might change slightly in the case of like an antibody [00:38:00] might not bind as well to protein A as another antibody. But in general, they’re going to be on the same order of magnitude. And by understanding whether that affinity constant is. Governed by a quick K on so a very rapid placement onto the resin or a very slow K off.

So basically, once you bind, it’s never coming off. We can figure out how long we need to allow the resin to be interacting with the molecule of interest before we move on to the next step. And we, we see that happen in, in the workflow Todd was talking about. And so that was some early work that we did that I think it seemed challenging because looking at the times of protein binding in an IMCS tip is challenging because there’s a change in volume over time because of the [00:39:00] aspiration and dispense cycles.

But what we found is that it really does closely mirror equilibrium binding. And so, in a sense, you can mimic the tip behavior via a tabletop experiment if you want to first, but we can figure out what the loose parameters are to give to clients in the hopes that they would get the best experience from the IMCS tips and the lowest amount of time, because after all, that’s one of the primary benefits of automation is we’re trying to free up a pair of hands and Make it as quick as possible to process as many samples as possible efficiently,

Andrew Lee: you know, Patrick, you covered an interesting aspect, kinetics and using the tips for measurement of kinetics for this binding interaction between 2 analytes for the resident and the analyte of interest. There are other techniques already out there, you know, the bio layer interferometry. The [00:40:00] surface plasma resonance and the 3rd technique that I’m not too familiar with, but creating coupled interferometry, but I guess we could throw in a 4th with 5 steps as another way of measuring pay on chaos.

But I think these are really designed for those measurements, right? Those, those affinity measurements, especially at low concentrations. But it does dictate a pretty important aspect of how the interactions are engaged and those interactions dictate your purification strategies. There isn’t really 1 universal resin that’s going to purify every single protein out there.

There’s no universal single protein purification technique. That’s going to purify. And that’s going to be your go to method because every protein has different. Physical properties and your end goal is. Different and leveraging either affinity based on your kinetics [00:41:00] or if you want a higher purification really for crystal structure analysis or crystallization, you’re going to need a variety of different techniques out there.

And then, of course, your throughput needs are also going to be different and your end quantity is going to be also important to note.

You know, we covered on our biotechniques paper, a couple of different key points, the kinetics, the different binding techniques for the purification techniques. We did mention the size exclusion chromatography here and there, but the. Really, the concept of it is. more covered in our SLAS technology paper published in 2022.

And in that paper, you actually have that coupled purification technique, the affinity, Plus the buffer exchange that Todd was talking about for 1 of the implementation, but also covers a plasmid DNA purification using ion exchange coupled to the buffer exchange and in that paper, we’ll link it to our podcast here, [00:42:00] but the general concepts are very similar that you do have dispersive pipette extraction for either the plasmid DNA.

Or the protein that’s affinity purified, and then you do a single directional size exclusion chromatography for your desalting.

So Todd, what’s the latest kind of interesting workflows that you’ve implemented for 1 of the customers without giving out too much

Todd Mullis: details? Yeah. So we had a customer come to us where they were using a fixed bed technology and it was taking their team eight hours for just the purification step.

And they were just wanting to increase the throughput for their workflow. And because ultimately this was fitting into a, even a larger automation workflow that they were wanting. And the eight hours was just not going to be good [00:43:00] enough in order for them to scale up the process. And so they reached out to us about using the tips for not only the affinity purification, but also the buffer exchange. And so we came outside, did a proof of concept that essentially showed that using the same type of techniques that Patrick was talking about with the variable binding and ensuring that we’re having a sufficient binding across the different, I guess, sample aliquots that we had and essentially showed that we could take that 8 hour process and turn it into a 2 hour process for the affinity purification, plus the 20 minutes for the buffer exchange, and that really is going to be a game changer for their overall workflow, because instead of having a full workday, Fully focused on the affinity purification they could run that same [00:44:00] method 2 or 3 times in a single day. That then fits within their larger automation workflow. And the customer was really happy about the experiments that we conducted that essentially showed that we could really make their process a lot more efficient.

Andrew Lee: That’s awesome. That’s I mean, improving your throughput by 2 to 3 fold. That’s pretty amazing. And then again, with the automation, you can probably try to do this 24 hours. I mean, think about the number of proteins that you’re going to purify and then now your downstream bottleneck might be your characterizations,

so, that example of working with customers, that’s very exciting on new workflows and throughput optimization by Todd. As we come to kind of wrap things up, I want to kind of leave a little teaser. Are there any new workflows or concepts that Patrick you’re working on that? Could be exciting for the audience.

Patrick Kates: Sure. There are a [00:45:00] couple. First we have the combinatorial aspect that Todd talked about with affinity and buffer exchange or really any purifications that followed by buffer exchange. And those are talked about in that SLAS tech paper that is linked where you can essentially do a purification from cells of a recombinant protein, buffer exchange that purified product into A buffer of interest and do your assay on deck.

And that can be, you know directed evolution or something like that, where, as long as you have an analytical technique on deck you can completely automate your process. And then also, I think, talked about net tech paper is the idea of a low endotoxin plasmid DNA purification, which combines an anion exchange purification of the plasmid DNA. And unfortunately, that DNA is left [00:46:00] in ethanol and high salt solution, which wouldn’t be that conducive to being used. So, while even though it’s low endotoxin, you’re not going to be able to use it until you’re able to combine it with our buffer exchange tip which puts it into buffer or water or whatever, and it does not have any of those contaminating salts or ethanol.

And so you’re left with a low endotoxin plasmid DNA. That’s suitable for transfection. And then 1 of the final workflows that we’re working on right now, and should be coming out soon is. Really, one of the most basic of workflows and that’s silica plasmid purification. And we’ve, I think, gotten to the point where we’re going to be able to do lysis and clearance of the cell lysate on deck and follow it up with a silica purification that yields a high amount of plasmid with [00:47:00] very low contamination. So I’m looking forward to that coming out soon.