Posters

Cells release extracellular vesicles (EVs) of different sizes and intracellular origin. Due to their heterogenicity, the isolation of the target EV population from a mixture of supernatant-derived particles can be challenging. Anion exchange chromatography (AEX) exploits the negative charge on EV surface molecules for binding to the positively charged solid phase. CIMmultus® EV, an AEX chromatography monolith column, can separate EVs in subpopulations based on charge and offers insight into the heterogenicity of particles. Besides the availability of preparative tools for separation, combining multiple orthogonal and complementary characterization tools is crucial for defining the EV product of interest. In this work, we used a multiple-detector PATfix® system for the analysis of CIMmultus EV-fractionated samples. Samples were analyzed for the presence of EV-related tetraspanins using the fluorescence detector. PATfix MALS 3609 detector was used for the analysis of particle-containing samples and calculation of particle sizes.

The cost of mRNA production is driven by IVT reagents, particularly the co-transcriptional capping reagents. Optimization of mRNA yield is therefore crucial for lowering the cost of mRNA production. To monitor the IVT reaction over time, we implemented a rapid at-line HPLC monitoring of consumption of NTPs and production of mRNA, with a sub-3 min read-out. Use of CIMac PrimaS analytical column allowed us to determine and adjust key IVT components that influence the kinetics of mRNA production and are critical for optimization of continuous addition of reagents, i.e. fed-batch IVT.

Fed-batch reactions can also be performed by continuous feeding, requiring automated control system. We used Ambr® 250 bioreactor platform, demonstrating for the first time its potential for mRNA production. First we designed a fed-batch IVT reaction in a thermal shaker, sampled and analyzed at-line by CIMac PrimaS analytics. Based on NTP consumption kinetics, the Ambr® 250 protocol was then designed to feed a defined mixture of NTP-Mg 2+ continuously.

Quaternary amine (QA) modified monolithic columns are frequently used for purification of therapeutically relevant 
adeno-associated virus (AAV) vectors, used in gene therapy. Beside process- and sample-derived variability, 
chromatographic material variability can influence the efficiency and scalability of AAV downstream processing. This study 
presents the evaluation of highly reproducible (HR) QA-modified CIM monolithic columns, where the goal is to elute AAV 
capsids in a very narrow conductivity range, regardless of the batch and size of the the column used. 
We have developed a chromatographic test method for proving the intra- and inter-batch homogeneity of the material 
through different column sizes. The method is based on separation of AAV2/8 capsids in ascending KCl gradient on 
CIMmultus QA 1 mL columns and specimen 0.2 mL units, taken from large-scale QA monoliths up to 8000 mL in size.

Lentiviral vectors are efficient tools for transfer and stable integration of large gene inserts into the genomes of both dividing and non-dividing cells. Third generation of lentiviral vectors, which uses a fraction set of HIV genes, is replication incompetent and self-inactivating, offering a relatively safe tool for academic and industrial use, while offering a larger gene transfer capabilities compared to more commonly used AAV. As a result, several therapies using lentiviral vectors are already approved or in clinical trials, with primarily ex-vivo use. Limitations for more widespread use, as well as development of in-vivo therapies, however remain, with one of the causes being downstream purification of lentiviruses. Here we showcase a process development for CIM monolithic columns for use in downstream purification of lentiviruses. Several CIM monolithic chemistries have been tested to determine their performance for lentiviral purification, with three chemistries showing potential for further use. Of the three chemistries, two have been selected for further development. Several modifications of the two chemistries have been prepared and tested on CIMmic columns with promising initial results. The chemistries have been successfully upscaled to CIMmultus monoliths and the results from initial findings confirmed. Additional experiments are required to confirm the findings and perform optimization. The full downstream process will be developed once the most suitable chemistry is selected. During preparative, multi-angle light scatter (MALS) was used as indicator of viral presence, while analytics, used in the process development, were ddPCR and infectivity test.

Several recombinant AAV (rAAV)-mediated therapies have been approved so far, and many more are in clinical trials, yet developing an efficient rAAV process platform remains a challenge. First obstacle arises from frequent practice of upstream (USP) and downstream (DSP) process being developed separately. USP optimization activities are usually focused on achieving high viral titer, which does not always translate into maximal purity and recovery in DSP. Second significant challenge is to bridge USP and DSP into a single bioprocess, due to lack of appropriate analytical methodology capable of producing reliable and unbiased results on both sides of the bridge. Furthermore, rAAV-based gene therapy vectors require the removal of process-specific combined with product-specific impurities, as they represent serious safety threats. Here we showcase the example of AAV8 USP optimization directed towards good performance in DSP. AAV8 is produced in HEK suspension cells via plasmid transfection. Based on the screening results, the best performing set of conditions was successfully scaled up to a 5L stirred-tank bioreactor. Harvested material was then evaluated in DSP, comprised of pre-capture, and two ion-exchange chromatographic steps on CIM monolithic columns. As bridging analytics of choice, we relied on PATfix platform, (d)dPCR and Comassie Bradford protein assay measurements for thorough assessment of relevant parameters such as titer, percentage of full AAV8 capsids and total protein concentration.

In every adeno associated virus (AAV) downstream process the key steps are the enrichment of full capsids and the in-process empty/full analytics. A common approach for enrichment of full capsids is liquid chromatography using ion exchange or multimodal chemistries, which based on particles’ surface charge differences, enables separation of full (F) capsids and product related impurities. Removal of impurities is an important step as they may pose a serious health risk, as well as deteriorate the economics of the production process. A PATfix® triple detector set in combination with an CIMac™ AAVEmpty/Full analytical column by Sartorius BIA Separations was used to determine E/F ratio. The column is a strong anion exchanger which exploits capsid charge differences for separation when employing an ascending salt gradient. Usually, NaCl with the addition of 2 mM magnesium is used, as the presence of magnesium has been shown to improve the peak shape and the removal of empty capsids. According to the Hofmeister series salts have different impact on non-covalent interactions among capsid themselves and between columns and capsids. For example, antichaotrope salts were shown to improve the empty/full capsid ratio, possibly by inducing the hydrophobic interactions between them. Quaternary alkyl ammonium salts (QAAS) are known to have antichaotropic properties and they differ in the substituents. Wang et al. reported that QAASs produced a better empty/full separation than NaCl. This poster presents a screening of different eluents for empty/full analytics. Beside NaCl as a reference salt, two QAAS were chosen: tetramethylammonium chloride- TMAC and a possible nonhazardous alternative choline chloride, which is bifunctional and contains both a quaternary ammonium and hydroxyl groups. The influence of magnesium on E/F separation was also tested.

mRNA has been at the forefront of both scientific and general public interests from the start of the COVID-19 pandemic. However, there are still limited options available for rapid characterization of mRNA containing samples. For precise characterization of an mRNA sample, first the presence and concentration of mRNA molecules in the sample needs to be identified. In the second step, any contaminants in the sample coming from the IVT reaction need to be identified and quantified. All major components of the IVT reaction; nucleotides, capping reagent, enzymes and DNA template may be present in the mRNA sample. In addition, impurities such as shorter, incomplete RNA fragments, and in particular, dsRNA may also be present. Contaminants may also come from the mRNA in vitro instability, caused by spontaneous hydrolyzation of the mRNA backbone. These issues can be mitigated using appropriate analytical tools throughout the mRNA production and purification steps.

• How to increase the binding capacity of Oligo dT18?
• Can a design of experiment approach be used to optimise Oligo dT binding?
• Is the monolith available in a high throughput format for liquid handlers?
• Is it possible to use a 96-well plate Oligo dT device?

Buffer conditions (salt, additives) influence mRNA binding on Oligo dT. Three contributing factors were identified and tested: NaCl, MgCl2 and Gu-HCl, the latter leading to a capacity of >6 mg/mL.

Abstract:

Affinity-based chromatographic isolation of mRNA is robust and simple, lending itself as a useful industrial platform. mRNA constructs typically contain a 3’ polyA tail to increase stability in vivo, thereby affording the possibility of affinity purification using oligo-deoxythymidinic acid (Oligo dT) probes covalently coupled to a solid support. Poly-adenylated mRNA forms a stable hybrid with Oligo dT under high-salt conditions which is destabilized when the salt is removed, allowing mRNA to be released. Typical dynamic binding capacity (DBC) of CIMmultus Oligo dT for mRNA is 2-4 mg/mL; ever higher IVT productivity will require higher binding capacities. Screening experiments to elucidate factors affecting CIMmultus Oligo dT binding capacity for mRNA were performed in CIM® 96-well Oligo dT format. A simplified model identified NaCl, guanidine hydrochloride (Gu-HCl) and MgCl2 concentration as the key factors contributing to DBC. Buffer chemistry, buffer pH, salt type and mRNA concentration had little or no effect on DBC.

The cost of mRNA production is driven by IVT reagents, particularly the capping reagent. Optimization of mRNA yield is therefore crucial for lowering the cost of mRNA production. In order to monitor IVT reaction over time, we implemented a rapid at-line HPLC monitoring of consumption of NTPs with concomitant production of mRNA, with a sub-3 min read-out. Use of CIMac PrimaS analytical column allowed us to determine and adjust key IVT components that influence the kinetics of mRNA production and are critical for optimization of continuous addition of reagents, i.e. fed-batch IVT.

CIM® PrimaS column family combines multimodal anion exchange/hydrogen bonding properties, binding molecules with predominantly negative charge. It is used as capture method for purification of mRNA from IVT (in-vitro transcription) reaction mixture with high binding capacity. High salt wash is used to elute the plasmid and other IVT components from the column without affecting binding of ssRNA.

Microvolume spectrophotometers are commonly used as quick and easy method to measure concentration and purity of nucleic acids. DSP process for purification of mRNA includes unit operations with salt concentrations up to 2.75 M (HIC) or up to 1.25 M (Oligo dT) during load and low salt concentrations during elution.

mRNA has been at the forefront of both scientific and general public interests from the start of the COVID-19 pandemic. The demand for the mRNA product has been incredible for the last couple of years. However, there are still limited options available for a rapid mRNA quantification and characterization. In this work, mRNA analytics using a CIMac Oligo dT column is presented. mRNA is a specialized group of RNAs that carries the blueprints for building proteins from the cell’s DNA in the nucleus to the ribosomes in the cytoplasm. One of the features of mRNA molecules is a polyadenylated (poly(A)) tail on the 3’ end, that can be up to 250 nucleotides long. This feature enables mRNA to bind to the Oligo dT column. HPLC Oligo dT analytics provide a solution for fast and reproducible quantification of mRNA throughout all the process steps of mRNA production and purification. The presented method was validated using mFix4, an uncapped mRNA analog produced in-house, 3969 nt long molecule with a poly(A )tail length of 95 nucleotides.

Adenovirus is well-know gene therapy tool that gained attention as a promising vaccine delivery vehicle, specially during Covid-19 pandemics, where it was used to deliver sequence for protein S (S). Multiple serotypes have been tested in clinical trials for various applications, the most common one being human adenovirus serotype 5 (Ad5). With this in mind, we chose Ad5-S construct with GFP tag as a model vector to develop upstream process (USP) and supporting analytical tools.

Messenger RNA (mRNA) is becoming a major contributor in the fields of gene therapy and vaccines, including those developed in response to the COVID-19 pandemic. Convective Interaction Media® (CIM®) Styrene divinylbenzene (SDVB) monolithic columns are promising for high resolution purification and separation of mRNA, enabling large-scale production of this molecule. This study demonstrates the ability to prepare homogeneous SDVB monoliths with desired chromatographic properties and economical analytics over the whole size range.

Monoliths are commonly used for separation of empty and full AAV capsids on analytical and preparative scale. PrimaS approach for AAV separation employs an ascending pH gradient, which can be sensitive towards small changes in chromatographic parameters. In present work 200 µL testing units extracted from large CIMmultus PrimaS monolithic columns, were employed for the adjustment of critical chromatographic parameters.

Recombinant adeno-associated viral (rAAV) vectors are the leading gene delivery tool for treatment of a variety of diseases. While several rAAV mediated therapies have been approved so far, and many more are in clinical trials, rAAV production still faces many challenges. Key goal of rAAV upstream process development is achievieng high viral titer together with a sufficient percentage of full capsids. Moreover, analysis of complex upstream harvest samples can be challenging. Classical analytical methods such as ddPCR/ELISA offer limited information due to differences in sample preparation and basic principles for detecting empty and full capsids. Method is also time consuming and therefore less useful for following rAAV production process in real time. To overcome these limitations, we developed a PATfix Valve Switch analytical method that is based on ion exchange biochromatography and can be successfully applied for analysis of empty/full ratios in crude upstream samples.

Standard 96-well design offers a great advantage for screening many samples or conditions and supports process automatization. Our approach was multi-parallel screening of different mobile phases for rAAV capture step using CIM® SO3 0.05 mL Monolithic 96-well Plates. Buffers of different pH, sodium chloride concentrations and use of Poloxamer 188 were screened to purify AAV2/9 clarified lysate obtained from Sf9 cells. Sample was pretreated by tangential flow filtration (TFF) coupled with nuclease treatment – Kryptonase. It was shown that the optimal conditions were buffers of pH 3.5, 500 mM NaCl, with addition of Poloxamer 188. Verification of results with selected buffer resulted in high capacity (1.44E14 capsids/mL SO3), great recovery (87.7 %) and excellent protein and DNA reduction (99.98 and 99.25 %).

In each adeno associated virus (AAV) downstream process one of the key steps is enrichment of full capsids. This could be achieved by density gradient ultracentrifugation however a main drawback is its scalability. A more common approach is liquid chromatography using ion exchange chemistries, which based on particles’ charge differences, enables separation of full (F) capsids and product related impurities including non functioning AAV capsids (empty, partially filled, misfolded and wrongly packaged genome or other DNA containing subspecies).

Extracellular vesicles (EV) are lipid bound products secreted by cells. Among them, exosomes have great potential for clinical applications. Animal and human-derived components used in cell culture, such as fetal bovine serum (FBS), naturally contain exosomes that can cross-contaminate the desired product. In order to study exosomes derived from cells of interest, multiple producers have come up with exosome-depleted FBS (EV (-) FBS) generated using different approaches. In this work we evaluated commercially available EV (-) FBS supplements for residual exosome content and tested their performance in upstream exosome production process. The analysis was performed with PATfix high pressure liquid chromatography system using PATfix size exclusion (SEC) analytical method.

Removal of impurities generated in the production of adeno associated virus (AAV) is an important step as they may pose a serious health risk, as well as deteriorate the economics of the production process. The most critical subsets of these impurities include: host cell nucleic acids, host cell proteins, chromatin, capsid aggregates, capsid DNA complexes and empty capsids.
This poster introduces two new column s for performing separation of empty (E) and full (F) capsids using multimodal approach. A PATfix® system with three different detectors, i. e. absorbance, fluorescence, and light scattering in combination with three analytical columns traded as CIMac ™ AAV Empty/Full, CIMac PrimaS ™ (AAV) - Beta, and CIMac ™ PrimaT - Beta by BIA Separations Sartorius. The separation columns were used to determine and evaluate empty AAV capsids as one of the critical impurities in AAV samples. The analytical results using CIMac PrimaS™ (AAV)-Beta and CIMac™ Prima T- Beta show that other fast and reliable orthogonal HPLC methods to the CIMac™ AAV full/empty column can also be used for the separation of empty and full capsids with monolithic columns.