Webinar
Real-Time Capacitance Control
for cell-cultivated products

Thank you for joining today’s joint webinar with ABER Instruments and Extracellular hosted by myself, Claire Thomas representing ABER, and Ellen Parker from Extracellular. We will just wait a couple more minutes for people to join before we begin the webinar. Just a reminder, we will be accepting questions throughout the webinar and we’ll be answering questions at the q and a session at the end. Thank you everyone for joining. We can start now. First of all, good morning Ellen. Thank you for joining us. Good morning. To begin with, can start with some introductions. So my name is Claire Thomas. I’m the Product Application and Lab Specialist for ABER Instruments. My role is mainly within the technical support team, but I also work in the labs. I’ve been with the company for coming up to five years now, and I’m very excited to be joined by Ellen. Ellen, would you like to introduce yourself? Yeah, of course. Hi everyone. My name is Ellen Parker. I’m a senior process engineer here at Extra Cellular. I’ve been with the company for around three years and I lead projects here in process development, mainly in process optimization and also in our scale up team scaling processes up to two thousand litre scale. Thanks for having me. Thank you, Ellen, and thanks for taking the time to join me. So I’ll be beginning the presentation by setting the scene to ABER’s technology. Then we can get to the fun part, is the data which Ellen will present. Firstly, just to summarise what the webinar will entail, for the initial part of the webinar I will try to answer a few basic questions such as who ABER is as a company, why is there a need for the technology like ABER’s in the present industry, and how does the technology work and its basic principles. Then most importantly I hand over to Ellen who will introduce extracellular. So Ellen, please go ahead and tell us what your presentation will be about. Yeah, so we’ll then be looking at how ABER’s technology has been integrated into our workflows here at Extracellular. So looking at developing a mammalian continuous process using ABER’s technology and also how we’ve translated that process into plant cell culture too. Thanks Ellen, looking forward to hearing about it! So to begin with, start by answering the first question, Who are ABER? ABER Instruments is a biotech company based in Aberystwyth in Mid West Wales in the UK. This is where the company was founded in nineteen eighty eight and this is where our headquarters are located. We are also employee owned which means all of us have a stake in our company which we’re very proud of. We have also set up a subsidiary in the US in twenty seventeen in Northern Virginia on the East Coast. But we also have a wide global distribution network, which seems somewhat surprising given our remote location next to the sea with beautiful views like these. When the weather is good, I’ll say that. Okay. We are the inventors of the technology, specifically in dielectric spectroscopy for biomass measurements. And we’re the industry experts with over thirty five years of experience in the field. And we’re entirely dedicated to biomass measurements in bioprocesses. We’ve recently expanded our offerings by launching the POTURA range, which utilises bio-reflectance for total cell concentration. However, for the purpose of this webinar, we will focus purely on the capacitance technology. We also have a worldwide customer base with systems in over one hundred and forty countries and our technology is used extensively by many major pharmaceutical companies to good effect. For many products, the transition from R and D to process development all the way through to manufacturing is an important one. The ABER technology has transitioned successfully through to production over the past fifteen to twenty years. It can be used in a range of bioreactors from mini to small glass bioreactors all the way up to stainless steel vessels, meaning the technology is fully scalable with your process. We also supply single use solutions for capacitance measurements via our partnership with Sartorius to produce a patch probe called the Bio PAT Viamass. But more recently, we’ve introduced the NEO through our partnership with Thermo Fisher. And you’ll see some examples of these on the right. Now to answer the question of why is there a need for the technology. Offline measurements and manual counting methods are historically the gold standard for monitoring biomass and cell growth. But offline measurement methods can present a number of challenges. Some of these might be the time and effort that goes into taking samples, especially if you’re using a manual cell counter. Insufficient sample points, so not getting a full picture of your cell data and behaviour over the course of your run. There can be a risk of contamination if you have to open the bioreactor to take samples. There might be a risk of error that can come from a variety of sources, maybe during sampling or maybe during sample preparation procedures. Some counters might have inherent variability, or if you have different people taking counts, you can get slight different results. So overall, this leads to just not having complete knowledge of your critical process parameter, which is biomass. Using ABER capacitance, which is a real time and in line technology, can have a number of benefits. So firstly, the measurement is real time, so you’re getting data on the cells and the process almost immediately. You can use the data to monitor the buyer process at the most basic level. But then you can also control parameters such as feed rate and cell bleeds to get a level of automation, which Ellen will demonstrate later with some of her data. You can use capacitance on a variety of different platforms, and it’s scalable as mentioned previously. And there’s no need to take samples. So that mitigates some of the time, cost, and risk for error of contamination that I mentioned earlier. And most importantly, you’re getting accurate and consistent measurements. So overall, you have a more complete knowledge of your critical process parameter, which is biomass. Finally, I’ll discuss my favourite part which is how the technology actually works, the technical side of things. The ABER technology measures two basic parameters, capacitance and conductivity. In simple terms, capacitance is the ability of any substance to hold charge around itself measured in picofarads per centimetre, Whereas conductivity is the ability of any substance to give away charge, and this is measured in milli siemens per centimetre. In order to understand the principle of technology, we need to return to the basic structure of the cell membrane. The The cell membrane is impermeable with respect to charge, specifically live cells, meaning that there is no free flow of ions across the membrane when under the influence of an electric field. And this is a key characteristic of the cell that allows us to measure the viable biomass. Here is a simple animation to best illustrate how the capacitance measurement works. So when an electric field is applied across a suspension, the ions move to their respective electrodes. So the positive move to negative terminal and vice versa. As soon as they reach the cell membrane, there’s a polarization effect of the live cell membrane resulting in a build of charge. This is what acts as a tiny capacitor. The greater the number of live cells, the greater the capacitance measured. As we’re measuring the build of charge around the cell, note this is not a reflection of cell density, but also a reflection of cell size. Therefore, capacitance is actually directly proportional to the membrane bound by volume of the suspension. So basically put, capacitance is everything biological in an suspension, whereas conductivity correlates to everything chemical. In contrast to this, as cells begin to die, they cannot hold charge around themselves anymore as membrane isn’t intact. So they’re leaky with respect to charge. So it doesn’t mean that they have literal holes in them. It just means they allow the free fire lines across that membrane. This means dense cells no longer have the ability to hold charge around themselves and are basically invisible to the technology. Likewise, solid media particles and gas bubbles can’t be measured by the technology as they can’t hold charge around themselves. They do not have that bilayer to have that buildable charge under the influence of an electric field, so aren’t measured. In summary, live cells with intact membranes can be polarized and therefore measured by the technology, whereas dead cells, which are leaky with respect to charge, are not measured by the technology. Solid media particles and gas bubbles are also not measured. Another take home point is that this is not a measure of just the number of cells but also the size of cells. Therefore, the resulting capacitance is directly proportional to the total membrane bound volume of those cells. This can be correlated with cell concentration if the cells are uniform or don’t have a rapidly changing volume, and most customers have success in this. In terms of cells, the technology is commonly and successfully used with mammalian cells. So for example, CHO cells, HEX cells, NSero, and you’ll see later in Ellen’s data, quail cells are also used. But we’ve also worked with a range of cell types including insect cells, yeast such as saccharomyces and pitya, both gram positive and negative bacteria, fungi, and finally plant cells, which again Ellen will have some examples of later. The main consideration in this is that the cells have an intact cell membrane and therefore can be measured by the technology. Now I’ll finish off with applications. So overall in terms of applications, the most basic use of the ABER technology is to monitor your cell concentration in real time, which gives you a detailed fingerprint of the process. Not only can the technology be used to monitor the process, it can be also used to control the process. Now this could be controlling the cell concentration at a specific value or controlling the nutrient feed based on capacitance or even harvest point detection. Ellen will have more examples of this a little later. I will now hand over to Ellen who will present some great data collected at Extracellular showcasing their novel process with ABER there to support in the optimization. I will comment here that if you have any questions, please put those in the chat section and we can address those in the Q and A session. Thank you for waiting patiently, Ellen. Please go ahead. Thank you, Claire. So I’m now going to talk about how extracellular has utilized ABER’s capacitance technology in developing a stable continuous process. So starting from a perfusion process and then moving into steady state culture, both with mammalian and plant cell culture. So first of all, who are we? So Extracellular is a contract development and manufacturing organisation dedicated to accelerating the cell based product industry. We’re based here in Bristol in the UK and we have three facilities consisting of research and development labs and then also two scale up and manufacturing spaces. We work with customers and corporates all over the world, and although currently based in the UK, we have plans to establish production across the world in the next couple of years. So what do we do? So Extracellular supports the cell based product industry, whether that is in media development, reducing the cost of your media or improving the ability of your media to allow cells to grow or doing cell development to really generate that perfect cell line for our customers. And then moving on into the engineering side of things, so we’ll do process development, process optimisation, development of continuous processes, and then moving on into the final stage of the process, into contract manufacturing and scale up. So running up So the cell based product the cell based product industry faces immense commercialization challenges. First off, yield improvements. So the industries need to transfer from traditional fed batch or perfusion processes to a continuous process to improve the final yields that you get out of a process. And these processes will also require heavy amounts of process optimisation to really get the most out of the process. Process stability is another challenge that needs overcoming. So media is going to be a major cost driver of your product And so optimising fusion rates or optimising your process to maximise media efficiency is going to be crucial, a crucial challenge for the industry to overcome. And also managing oxygen transfer rates and thermal rates as well is going to be really important as we develop the process. Moving on, scale up is a huge challenge in the industry. So currently Most people are working, you know, bench top laboratory scale volumes. But really in order to create meaningful quantities of products to be able to commercialise, we need to be scaling beyond these volumes up to, you know, two thousand, ten thousand, fifty thousand litre scale. And all of these things play into the final cost of the product. So, The main challenge at the moment is the cost, and so we need to create innovative ways in order to drive that final cost per kilo down to a place that allows us to commercialize products. So, where do we start in terms of doing that process optimization? So, at Extracellular, we did some process optimisation using a quail muscle cell line and initially did a parallel baseline with a batch, a fed batch and a perfusion process. So as you can see from the graph on the left, the perfusion process reached ten times the cell density and five times the biomass, then the batch and the fed batch process. And this was the first time when we utilized ABER’s capacitance technology. So the graph on the right shows the relationship between cell density and the capacitance. And as we pushed into higher and higher cell densities, you see the viability start to drop and that reflects nicely as the capacitance didn’t increase to the same extent as the cell density as it was only measuring the viable cells. This provided us some early insights into how we use real time cell density to have a better idea of and a better understanding of what’s going on in our bioreactors and start to introduce some ideas about improved process control and some automation as we move forward. So as I mentioned about industry challenges, we had some control limitations with the perfusion process. Despite reaching five times the biomass, high density cell cultures prove a lot more difficult to control. And in this process, we saw that oxygen transfer was the limiting factor. So you can see in the line in blue, which is the DO, you see that crash around the one hundred and eighty hour mark, even maximum gas flow rates and oxygen percentages. The cells were just consuming too much oxygen in order to maintain the DO at thirty percent. So we see that crash in DO. We also saw some temperature instability. So with the perfusion process, we were working at very high perfusion rates of around six to seven vessel volumes a day of media. And you can see that it’s difficult to control the temperature during that time. And yeah, as I mentioned, high perfusion rates means that this process is not the most efficient in terms of media usage. And so all in all, we developed a process that although was reaching very high cell densities, it’s not scalable. You So we kind of thought well what do we do here and having the capacitance probe allowed us to think more about how we can automate a process to start developing the continuous process. So, that’s what we did. So what did this look like in terms of developing the continuous process? Well, we used the perfusion data to provide us with key insights into defining these points for the continuous. Because of the unstable nature of the perfusion process, we knew that we wanted to limit the vessel volumes per day to two. This would also correspond with keeping the glucose concentration above one gram per liter because we’re in the baselining in the batch process that anything below this would then have a negative impact to cell growth. So we started to define this area where we wanted to target in terms of our starting cell density and the cell density that we would end up holding the process at. We also knew that we wanted to be able to create meaningful quantities of biomass from the continuous process. So as you can see on the graph on the left with the wet cell weight, we knew that we wanted to have the process running around thirty grams a liter so that if the cells had a twenty four hour doubling time and we removed fifty percent of the culture, we’d be generating around fifteen grams a day of biomass from the one meter system. And so all of these fed into generating a set point for our capacitance reading. And you can see from the graph on the top right that when we try and find this sweet spot, you see the KRASS set point, starting set point of around thirty. So the perfusion data gave us a really good idea of where to start in terms of using the capacitance probe to start to develop the continuous process. So what did this look like in real life? Well you can see from the graph, image on the left hand side, this is the continuous setup at extracellular. So you can see the one meter bioreactor on the right hand side connected to an ATF2 and a hollow fibre membrane from Repligen. And then you see on the left hand side the harvest bottle. So we had a pump that was continually bleeding cells out of the bioreactor and into the harvest bottle on a scale on the left hand side. We generated a few control loops in order to maintain this steady state culture. So the first control loop that we defined was the media fill pump. So that was connected to the weight of the bioreactor. So as the cells were being harvested, or as the waste media was leaving the process, the media in pump would automatically kick on and refill the bioreactor to maintain out volume. The second control loop, as I’ve mentioned, was the capacitance probe that was connected to the harvest pump. So we had a capacitance set point and as soon as the cells proliferated past that, it would automatically kick on the pump to bleed the cells and maintain the system in equilibrium. So, This made it incredibly easy to control with a really like high level of accuracy. And on that you can see the process diagram. We also made sure that the total VVD from the system was capped at two, so having flow sensors on the harvest and the waste permeate line just to ensure that we were sticking to those targets of two VVD as our perfusion rate. So what did this look like? So using automation from ABER’s technology allowed us for to produce an incredibly stable and flexible continuous process. So you can see the actual capacitance very nicely mirrored our set point. As soon as we altered those set points, the harvest pump would kick on. To allow for us to change the process depending on what was going on in the system. And you can see our capacitance set point validation, that we managed to produce a high accuracy of maintaining different set points using these control loops. And you can see from the graph on the right hand side that the harvest pump was continually on during this process. So you see this continuous cell bleed that we were aiming for. And it allowed us to be really flexible if we wanted to change a set point based on what the offline data was informing us. We could increase or decrease the capacitance set point and the pumps would act accordingly. So you can see about halfway through the steady state culture, we did have to drop the capacitance set point significantly. So what happened there? So you can see that the capacitance set point and the capacitance readings very nicely mimicked cell density. We first allowed the cells to grow exponentially up to around twelve million cells per ml before starting the harvest and maintaining steady state culture at around eight million cells per ml or a capacitance of around thirty to thirty five. But at day twenty five, we did a fifty percent dropout and halved the capacitance in the bioreactor. The reason for this was because of the metabolite profile. So looking at the glucose and lactate graph, you can see that although the cells are being maintained in steady state culture, you see the glucose start to decrease throughout the first kind of ten days of steady state culture down to a minimum of zero point three grams a litre and correspondingly the lactate increased to just under five grams a litre. And it was at this point in a conventional process where you would have to do a harvest, you would have pushed the cells to their maximum abilities and it would be time to end. However, the automation that we have during this process allowed us to recover the process. So what we did was we halved the capacitance set point, it kicked on the harvest pump and we drained around fifty percent of the bioreactor. This subsequently was refilled with fresh media and you can see that the metabolite profile was almost reset within a day during this process with both the glucose and the lactate stabilising between two and three grams a litre. And it was then that we could then start increasing the capacitance set point and bring the density in the bioreactor back up to where we started. But we see a different profile here. So as we start to increase the cell density, we don’t see the same trend. We see that the glucose and the lactate concentrations actually tend to stabilise. So having this flexibility in our process allowed us to not only extend the process for a further two weeks, but also to wean the cells into this environment with a lower concentration of glucose and a higher concentration of lactate, which almost kind of allowed them to the system to maintain an equilibrium despite increases in cell density. So even though we were changing the cell density and the metabolite profile was changing, we do see a high level of process stability. So you can see the population doubling time over the process remained extremely stable. So with an average of twenty four point five hours. We see that despite all of these changes, the doubling time maintained relatively consistent. And when we look at final outputs from the eight hundred millilitre working volume, we generated three hundred and twenty point eight grams of biomass. The process in total used sixty six point two litres of media, giving us a perfusion rate of around one point eight, one point nine. So you can see as well from the cell morphology, the differences between day zero and day thirty six, we see high levels of cell line stability, there aren’t large changes in morphology or cell viability during this process. And in terms of the harvest outputs, you can see a nice comparison between the actual harvest that we got out of the bioreactor versus the capacitance set point. So you can see that really, really high level of process control due to the ability to control the process from the capacitance probe. But again, the Continuous Continuous process isn’t perfect and your optimal process mode is gonna be dependent on a variety of factors. For example, media cost, if you have a very high media cost, you might want to opt for a batch or a repeat batch process that really maximizes your media efficiency. And you also need to choose a cell line that can cope with continually proliferating over a long period of time. So making sure that you have a cell line that can grow consistently over the course of many months is gonna be an important factor when you think about running a continuous process. But if you can find these things, then the continuous process is a lot more efficient in terms of final yields. For example, if we were to run to generate the same amount of biomass from a continuous process, we would have to run somewhere between twenty five and thirty back to back batch bioreactors, which again is going to take up your equipment utilisation and also your time. So being able to generate a process that can run for long periods of time in the same vessel is going be a big time and equipment saver in terms of producing your final yields. So I’ve spoken about how ABER’s technology can be integrated into mammalian cell culture for food ingredients like in the cultivated meat space. But we also wanted to assess the technology in other cell based industries. So whether that’s bio-actives from plant cell culture for supplements, cell derived ingredients, or even personal care ingredients such as in the skincare industry. So we wanted to see how the technology transfer worked in that respect. So we started off by using model plant cell line, the tobacco cells, BY2 cells, to understand how the capacitance was mimicked the packed cell volume and the wet weight of this process. So after doing a couple of one litre baselines, we then scaled up to a three litre stir tank and also the ten litre bubble column. And the great thing about ABER’s technology is we could easily integrate it both at different scales. So we use the Eppendorf BioBlue systems for our stir tanks and then also into different bioreactor types as well. So moving to the bubble column, we could still use ARBA’s technology here. And as you can see from both graphs, not only did we observe comparable growth data between the stir tank and the bubble column, we also saw that the capacitance was a great way of measuring plant cells. So it really validated for us that we could transfer this technology that we developed for mammalian cell culture across into helping to support customers in other industries. And both of these bioreactors produced greater than three hundred grams a litre of final tobacco product after an eight day parallel process, which was again great to see. So we went on to develop, thought about developing the continuous process for plant cell production now that we knew that we could transfer ABER’s capacitance probe into the plant cell culture. And so what we did was more of a repeat batch style process using the capacitance probe. So as you can see on the left, we allowed the cells to proliferate up until a certain point before doing an eighty to ninety percent dropout, drop the cell density back down, and then again allow the cells to proliferate. So why did we do this? So the continuous process was a challenging process to transfer from mammalian culture to plant culture due to the technology and equipment requirements of it. So as I mentioned, mammalian cell culture with the continuous process, it uses ATF two with a hollow fiber membrane. But due to the size and the morphology of the plant cells, you can see on the right hand side, the plant cells grow in these kind of Lego like chains. They’re just too large to be able to integrate into hollow membrane technology. So, we had to find a way to do a continuous process without using fusion. And so what we did instead was, again, this repeat batch, but the great thing about having the capacitance probe is it allowed us to determine the specific point at which to harvest. Instead of waiting, you know, anywhere between twelve and twenty four hours to sample, we had capacitance readings every ten seconds, which really allowed us to optimize the perfect moment to harvest so that the cells didn’t overgrow and impact the the next passage. You can also see the linear regression model that the packed cell volume and the capacitance model each other very nicely with the R squared value of zero point nine. So we were confident that the capacitance for the plant cell cultures was giving us a really good idea of what was going on in the bioreactor. The other challenge that we had with plant cell culture was the sampling. So traditionally in mammalian cell culture we would just use a syringe attached to the sampling lower lock port of our reactors. But with the plant cell culture, these cells aren’t compatible with that. We find that the plant cells tend to get tangled around the sampling port and so it impacts what your morphology of your cell looks like when you put it under the microscope and also things like viability and affecting your pack cell volume value that you get out at the end. So we instead swapped for welding on a sampling bottle and then pressurising some sample out of the bioreactor. What that meant was we were sampling much larger volumes. So in the six hundred millilitre system, we were sampling anywhere between twenty and forty millilitres a day, which is obviously a significant portion of your bio reactor daily to be able to sample. But because we had the capacitance probe, we didn’t have to sample as frequently because the capacitance was giving us a really good idea of what was going on in there. So using ABER’s technology not only reduces operator interference and reduces any risk of contamination, things like that. It also was maximizing our yields of plant cell culture due to eliminating the need for us to sample as frequently. So what does this mean in terms of scaling up? So we’re all working together towards a point in the industry where we can commercialise these cell based products. So how does this process optimisation and the use of automation and improved processes really like play out into the final product? You can see from the graph on the left that when we operate at the two thousand litre scale, having higher yields from a repeat batch or a continuous process while working in the same two thousand meter bioreactor, it’s gonna massively reduce your cost per kilogram. You’re having higher yields whilst carrying out a similar sort of work. And so you see that the cost of goods dramatically drops during these for these processes. And when we look more towards the fifty thousand litre scale, you see this impact even more. So we see that with continuous perfusion we are getting down to a cost per kilogram that is something that we could commercialise. In the cell based product industry. But it does highlight the importance of media costs here because for processes like perfusion and continuous perfusion, your media requirements are much higher and so that’s gonna add to your cost. So when we look at the final cost of goods and how we reduce those, media reducing the media cost is obviously important, but what I’ve spoken about a lot today is this increased automation and how this can impact your final cost of goods. So it can impact in one of two ways. By understanding your process a lot better and having more data points using ABER’s capacitance probe, we can really we can increase the media efficiency. Optimizing those perfusion rates means that we’re using less media and it overall drops the media costs and so your overall cost of goods reduces. It also reduces the labor if we can have more online measurements and we rely less heavily on the offline sampling. And again, reduced labor reduces your cost of goods. So I think moving towards this higher level of automation is gonna really help to accelerate this industry and get these products to market faster. And finally, I just wanted to take the opportunity to thank the amazing team that’s been generating all of this data and I’d be more than happy to answer any questions that you might have. Thank you very much. Thanks Ellen for the presentation. That was absolutely great. It’s really good to see some examples of the work you do at Extracellular. I can see that you’ve been really busy. Also, thank you everyone for your time, and we will start the q and a session shortly.