What are the benefits of buffer exchange and what are the benefits of performing this work?
Buffer exchange is the process of transferring a large biomolecule from one solution to another. It can be a chore and is often a source of processing error, but it is a task that must be completed to keep the biomolecule stable or prepare it for another processing step.
This is very apparent in the case of lipid nanoparticles (LNPs). LNPs are formed where a buffer cannot remain in place for long due to the presence of an organic solvent – highlighting the need for rapid buffer exchange.
Other examples include viral vectors and AAVs produced in very dilute concentrations, frequently orders of magnitude below the levels required for dosing, which need to be exchanged and concentrated for further characterization work.
Nucleic acids, proteins and antibodies often require buffer exchange between processing and purification steps. A solution may suffer from dilution due to chromatography, the presence of too much salt, the wrong pH or the use of the wrong buffer, period. These samples should be exchanged into a buffer to ensure stability for downstream conjugation, storage or characterization. This is especially important when working with antibodies due to the amount of work that goes into the formulation, optimization, and screening. The use of lipid nanoparticles and AAVs is comparatively more recent and formulation development is less explored, but it is nevertheless important to exchange into a stable solution and at the right concentration.
What are some commonly used buffer exchange techniques?
Buffer exchange is a ubiquitous process and there are many ways to do this. Most of these methods are low tech. For example, dialysis, which can be left to run unattended over a long period of time, usually overnight.
The trade-off for being fully walk-away is that the sample is often diluted in the process, necessitating the use of an additional, separate concentration step.
Centrifugation filters pose the opposite problem – they are rapid and can be run to the required volumes, but users must attend to these throughout the whole process and monitor them. It is challenging to run multiple samples or track these easily because there is a possibility that one or two of these will run slower than or different from the rest.
Tangential Flow Filtration (TFF) remains the most high-tech of the commonly used methods, but it is not possible to walk up and use this approach. Pumps must be set up alongside an expensive cassette, and the run takes a long time. It is usually only possible to run one sample at a time, and the method is best suited to larger-scale processes of 50 mL or more.
Thankfully, Big Tuna offers several ways to offload some of this buffer exchange and sample preparation work - especially for AAVs and LNPs.
When we developed Big Tuna, we were looking for a solution that could allow users to perform buffer exchange in a robust, streamlined, walk-away manner.
Can you give our readers an overview of Big Tuna and how this helps improve the buffer exchange process?
Big Tuna is a fully automated buffer exchange and concentration platform that can handle samples in many different formats. It can be used with as little as 100 µL of sample in a 96-well format or nearly 50 mL in a 24-well format. It can also be used with a wide range of molecules and boasts a number of useful features and capabilities.
Big Tuna is a filtration plate-based exchange. It uses a consumable called the Unfilter - developed in-house at Unchained Labs. The 96-well format is an SBS format with a working range of 100 µL to 450 µL per well, while the 24-well format has a working range of 450 µL to 8 mL per well.
Both formats use regenerated cellulose membranes at varying molecular weight cutoffs - 10, 30 and 100 kilodaltons. From a cost standpoint, this is comparable to a centrifugation filter, but the plate format of Big Tuna is more amenable to handling many samples at once.
What are the key principles of Big Tuna’s operation?
Big Tuna uses a modified ultra-filtration diafiltration process. Users dispense samples into the Unfilter, and this is placed into a buffer exchange chamber. Big Tuna measures the volume using a non-contact acoustic sensor and starts a gentle pressurization cycle with orbital mixing.
After a period of pressurization, Big Tuna stops, measures the volume again, and determines how much buffer has been removed from the sample. It refills each well with either the new buffer or the sample itself and starts the process again. Each well is individually tracked because - as mentioned - sometimes samples will run differently for a variety of reasons, and it is important to ensure that faster flowing samples are not drying out while users are waiting for slow samples to complete.
This cycle is repeated until either the desired volume has been added to the sample, or it has been exchanged into the new buffer to the desired degree.
This filtration process is adaptive, meaning that it will pressurize for a conservative amount of time at the beginning, determine the flow rate for the samples inside and then set future pressurization times based on the sample flow rate. It then adjusts the pressurization time based on the fastest flow rate. This ensures that the process is completed as efficiently as possible without overconcentrating samples.
Big Tuna also features a programmable pressure level which ensures that a sample is being handled appropriately. Most samples are exchanged and concentrated at 60 psi or around 4 bar, but we found that certain sample types like AAVs and low concentrations of nucleic acids or proteins flow much faster, and can be exchanged and concentrated at as little as 15 psi or around 1 bar.
Employing orbital mixing as part of the process makes the exchange run efficiently. This can be understood by contrasting orbital mixing with dead-end filtration. Dead-end ultra-filtration will concentrate a protein at the membrane surface as the exchange is taking place. That slows down the flow and creates a concentration gradient within the sample which can result in a number of undesirable effects, such as clogging and fouling the membrane, or aggregating because the sample is at a higher concentration than it would normally be.
To avoid these problems, Big Tuna uses an orbital mixing process to ensure that the flow is consistent and uniform during the exchange process. Not only does this make the exchange faster, but it also ensures that the sample is being handled in the best possible way.
Automating this process offers a greater degree of process control than would be possible with manual methods. The operator can pre-set the percent removal per cycle: the volume of buffer removed in each pressurization cycle. Dilute or very robust samples can run at a higher percent removal, which reduces the number of required cycles to fully exchange.
For high concentrations, unfamiliar samples or where there are concerns about the buffer change being too drastic, a lower percent removal can be used to ensure a sample is gradually introduced to the new buffer without putting the sample under further stress. There are usually more cycles but of a shorter duration.
The percent exchange represents the total amount of buffer replaced. Most buffer exchange processes aim for a complete exchange - between 96% and >99%. In cases where users are looking to establish a flow property or characterize already known sample types, a lower percent exchange may be sufficient.
Big Tuna makes it possible to concentrate directly without doing a buffer exchange. This is achievable by using the full volume range of the plate. For 24-well plates, this can range from 8 mL to 450 µL.
There is also an application on the system which allows users to continue to add up to 40 mL additional sample back into that 24-well plate throughout the process, enabling over 100-fold concentration from 48 mL to 450 µL.
How do the presets available on Big Tuna benefit users?
It is impossible to use the same process for every sample. Biomolecules are very different sizes, and have very different flow properties in solution.
To help streamline process development and account for these differences, Big Tuna has presets based on the molecule type and the concentration. This allows users who have never used Big Tuna to navigate the process and establish the optimum approach to running a sample. We have done the groundwork and already programmed in the best approach for running common sample types.
Even though the pressure cycle time is adaptive, knowing that it should start at a lower pressure or at a different percent removal for a specific sample type is a helpful starting point for someone who has never used the instrument before.
Big Tuna is user-friendly and powerful, and users tend to quickly become experts in no time. At that point, users can override these presets and run it in a way that makes the most sense for their specific samples.
Can you provide our readers with some examples and case studies to illustrate Big Tuna’s features, specifically around gene therapy and vaccine preparation?
Lipid nanoparticles are a good proof-of-concept example for gene therapy vectors. LNPs are formed by the combination of two different phases - an aqueous phase which contains a genetic payload and a surfactant, and an organic phase which contains an organic solvent and the other polymers that make up the LNP.
These two phases are combined to create an emulsion, within which nanoparticles are formed to encapsulate the genetic payload.
It is important to immediately remove any ethanol (usually from the organic phase) and stabilize the molecule with the payload attached. As soon as the emulsion is formed, it must be exchanged to remove the ethanol.
For this proof-of-concept work, we took a Firefly luciferase, mRNA encapsulated LNP in 10% ethanol, buffer exchanged this directly into PBS and concentrated it threefold.
This was achieved using a 100 kDa Unfilter 24. Its larger volume and higher molecular weight cutoff ensured that flow was as fast as possible. An LNP preset on the instrument streamlined this exchange, which was at 60 psi and took around 90 minutes, including the concentration step.
The goal of this work was twofold. The first aim was to confirm that it was possible to hit the concentration target, which was the sample concentration by around threefold.
This was successfully achieved. Big Tuna can measure volume throughout the process, and at the end of this, we established that we had reduced the volume by a factor of three.
We also determined the percentage encapsulation was 98% before we started. When we were finished, we were within a couple of percent of that target, meaning that we did not damage the LNP in the process of exchanging out the ethanol.
This was all achieved in a single step without the sample having to leave the chamber or without having to stop the process and complete any additional actions.
AAVs, like many samples, are produced in low concentrations in a large volume. Our goal was to develop a way of concentrating samples easily and with minimal user intervention, so we developed a new application to enable this on Big Tuna.
Unfilter 24 can normally concentrate from 8 mL to 450 µL, but some applications start with more than 8 mL. We, therefore, established a method whereby users add 8 mL to the Unfilter 24, and then place the rest of the sample in reservoirs that sit on the deck.
This approach allows users to add an additional 40 mL of each sample for 24 samples simultaneously.
The Unfilter 24 is prepared and placed in the exchange chamber, exactly like buffer exchange. Unlike buffer exchange, after a pressurization and volume measurement step, Big Tuna adds sample instead of buffer. The instrument will repeat that cycle until the volume requested by the operator has been transferred and 8 mL of sample is in each well in the Unfilter 24.
If users want to concentrate further, they can do so in the same Unfilter by taking the 8 mL plate and concentrating this down to 450 µL. When combined, these two reduction steps result in over 100-fold volume reduction in one plate.
We took AAV9 at a volume of 18 mL, and using the steps above, added 8 mL of AAV sample to a 30 kDa Unfilter 24, and added 10 mL to a reservoir on the deck.
Once the 18 mL had been reduced to 8 mL, the concentration step was run to reduce from 8 mL to 500 µL.
Both processes use the Big Tuna AAV preset that runs at 15 psi. The concentration was around 5E11 initially, and we were able to achieve a concentration within 5% of the target. For the volume itself, we went from 18 mL to 500 µL, achieving within 5% of that target.
To confirm the capsid AAV empty/full ratio was the same before and after buffer exchange, we tested via an application on the Stunner platform called AAV Quant. We started with around two-thirds of the capsid full, and we ended with it around 75% full.
The whole process – including a 36-fold concentration - took around 90 minutes and maintained the capsid payload within the sample throughout.
Can Big Tuna also be used with the process of nucleic acids desalting?
Yes. Nucleic acids desalting is possible via many platforms, but the benefit of performing this on Big Tuna lies with the instrument’s use of orbital mixing and its uniformity of sample across the plate.
Other methods – for example, vacuum filtration – often exhibit edge effects or the smile pattern, which impacts the ability to achieve the correct concentration. This is not an issue for Big Tuna.
We performed a proof-of-concept experiment where we took double stranded DNA in TE buffer and concentrated this threefold using both available 30 kilodalton plates - the Unfilter 96 and the Unfilter 24.
In this case, we aimed to concentrate down from 2 mg/mL to around 6 mg/mL. Data from the Unfilter 96 revealed this was slightly above the target, but the volume was achieved, and there was no loss of material.
Data from the Unfilter 24 confirmed that we had that the correct amount of DNA present at the end of the process and that exchange targets were achieved.
About Unchained Labs
Unchained Labs is all about helping biologics and gene therapy researchers break free from tools that just don’t cut it. Unleashing problem-tackling products that make a huge difference in the real science they do every day. That’s their mantra, their promise and they own it.