Community Research and Development Information Service - CORDIS

  • European Commission
  • CORDIS
  • Projects and Results
  • Final Report Summary - INTENSO (Gaining Productivity, Cost Efficiency and Sustainability in the Downstreaming Processing of Bio Products by novel Integration and Intensification strategies)
FP7

Intenso Report Summary

Project ID: 312004
Funded under: FP7-KBBE
Country: Germany

Final Report Summary - INTENSO (Gaining Productivity, Cost Efficiency and Sustainability in the Downstreaming Processing of Bio Products by novel Integration and Intensification strategies)

Executive Summary:
The European Life science and chemical industries increasingly depend on efficient, sustainable, and cost-effective bioprocessing platforms to remain competitive. A critical assessment of current bottlenecks during (bio) manufacturing clearly indicates that the recovery and purification of biologicals in large scale in responsible for many inefficiencies.
For ATPE, (Aqueous Two-Phase Extraction), a variety of cell strains, bioproducts, and upstream processes have been developed. Application of pulsed electric fields to yeast and bacterial strains producing different recombinant proteins has proven to facilitate efficient release of the bioproducts. Rational experimental design on ATPS formulation, followed by surface plotting and analysis of variance has resulted in the optimization of ATPS composition for the separation of ZERA®-fused proteins, VLPs and LYTAG-driven affinity partitioning of immunoglobulins, from culture media supernatants. The effect of chemical coatings for the EBA (Expanded Bed Adsorption) process has been investigated, providing a better process optimization by minimizing biomass-bead interactions. Experimental methods have been developed and validated to characterize chromatographic beads and are, for the first time, applicable to intact beads. The associated development of xDLVO calculations has opened the door to probe entire interaction energy surfaces to pinpoint ideal process conditions for a given system of cells and chromatographic media, which was successfully achieved with high cell density broths. The first results generate with convective flow sorption technology (CFS) of the purification of different bio-products on radial monolithic columns were obtained for different pDNA molecules, mRNA and two different VLP materials. Purification efficiency of radial monolithic columns for purification of pDNA was compared with commercially available pDNA purification kits. A prototype was designed that will be used to perform preliminary tests for simultaneous filtration and capture of the crude or partially processed feedstock.
For hybrid disposable cartridges (HDC), protocols and sufficient materials of the biological products pDNA, mRNA, recombinant proteins and monoclonal antibodies have been produced for further downstream processing. A standard grafting protocol with subsequent immobilization of ion exchange (SP/Q/DEAE) functional ligands was established for the gPore adsorbents. An understanding of the purification behavior of pDNA on anion exchange gPore adsorbents in terms of adsorption capacity was achieved. A preliminary prototype cartridge was designed, with additional equipment for optimal back pressure, chemical and physical resistance. And the prototype cartridge based gPore NW was utilized for direct target product recovery from crude or partially processed feedstock and subsequent purification in intermediate or polishing application.

Project Context and Objectives:
In an increasingly globalized market, the European life science and chemical industries have to depend more strongly on a streamlined cost-efficient bioprocessing platform. These processes are broadly differentiated into the upstream (e.g. the product manufacture, for instance cell-culture or fermentation) and downstream (primary recovery, high resolution purification and polishing/formulation of the product) processing operations. While the upstream part is well understood and significant improvements have been made in this field, the science of downstream processes (DSP) is relatively rigid, more challenging and highly interdisciplinary. Therefore, new approaches must be found in order to deal with the reliance of the DSP unit operations on one or more differences in the physicochemical characteristics of the target product compared to the contaminant entities present in the feedstock. Because these unit operations are usually a trade-off between yield and purity, efficient primary recovery linking fermentation and DSP is of the utmost importance. Measures to alleviate this problem could, for example, be the reduction of the number of unit operations required or the redefinition of unit operation sequences. The biotechnological market of today is just beginning to produce new classes of biological agents, from monoclonal antibodies to pDNA, virus-like particles or nanoplexes, for which the established downstream processes, based mostly on packed bed chromatography are insufficient.
At its core, the Intenso project is about identifying the bottlenecks of the currently used methods, and finding new solutions to circumvent them. This is highly relevant for all biotechnical products, as DSP account for more than 80% of the manufacturing costs, which makes the development of cost effective systems mandatory. The Intenso approach will not concentrate solely on the effectiveness of the new methods, but will strive to lead to a sustainable approach, factoring in risks derived from economic (cost estimation operating-costs estimations and profitability), environmental (potential burden spots, waste water treatment costs), and social developments, also including government policies and legal constraints.
In practice Intenso proposes an integration / intensification strategy resting on four technological pillars, targeting innovative new technologies, coupled with horizontal activities such as impact assessment, demonstration, training and dissemination. The four technical pillars are:
Aqueous Two Phase Separation (ATPS) will explore conventional vs. novel systems for extraction of products, with special emphasis put on an increase of product solubility and on continuous processing. Especially for the ever-growing field of monoclonal antibodies, this technology seems very promising. It is equally viable for the intensification and cost reduction of the production process for other high-value biomolecules or nanoplexes, for example pDNA, virus-like particles (VLPs), cellular bodies, and intracellular as well as secreted proteins. ATPS can be paired with electroextraction for the processing of intracellular proteins. In order to establish ATPS as a viable alternative, the physicochemical properties of the targeted products will be matched against those of the corresponding feedstock. To achieve this, different polymers and salts will be employed to maximize the solubility of the product. The relevant parameters, for example, phase, pH, and molecular weight, will be investigated in micro-reactors and on the pilot scale, combined with an exploration into possibilities for continuous separation, for example by employing hydrocyclone technology.
The suitability of second generation Expanded Bed Adsorption (EBA) to the primary recovery of products will be evaluated. This technology can be integrated very early in the bioprocessing scheme, directly after the fermentation / cell culture step. EBA has been known to suffer from undesirable interactions between biomass components and fluidized adsorbent beds, compromising hydrodynamics and dynamic binding capacity, and thereby the process performance. Direct capture of proteins of biopharmaceutical interest will lead to simultaneous clarification, concentration, and partial purification – all in a single step. Second generation EBA adsorbents, smaller in size and denser at the same time, have been developed, allowing the use of faster fluidization velocities. Consequently, this is associated with higher drag forces that could remove attached biomass particles and lead to lower buffer consumption. It will therefore be imperative to understand the thermodynamics of biomass deposition onto the adsorbent beads, and to observe critical operational parameters, such as flow rate, biomass concentration, expansion factor and adsorbent type.
In this part of the project, Convective-Flow Sorption Modules (CSFM), for example Monolith technology, will be researched based on their comparative performance and industrial applicability. Due to their special mass transfer properties, they allow for the rapid capture of macromolecular entities and supra-molecular structures, while at the same time reducing the environmental footprint of associated bioprocessing activities. Overall, the goal is to enable the efficient purification of biomolecules beyond the traditional enzymes, proteins, and monoclonal antibodies, including entities of larger size like pDNA, VLPs, or vaccine viral particles. The main reasons why conventional methods fail are due to the limited binding capacity and the irreversible attachment of products to adsorbent or processing surfaces. Due to the absence of diffusional mass transfer limitations, this technology has the potential to bind pDNA with higher capacity than beaded adsorbents and prove useful in the purification of virus like particles or mRNA. Coupled with radial columns to add a second level of intensification it can afford more rapid processing at reduced resource consumption. Systems based on convective (Monolith) media will be subjected to chromatographic experiments, determining parameters like product concentration, flow rate, mobile phase composition, mobile phase conductivity and pH. Different types of materials, e.g., ion-exchangers and affinity materials will be tested, and product quality will be monitored.
Within this technological pillar, Hybrid Disposable Cartridges (HDC), a hybrid approach created by merging existing textile technology with material / polymer chemistry will be used to generate a cost-effective and disposable system supporting eco-friendly bioprocessing. The goal here is to evaluate the performance of gPore NW materials, which can be functionalized and used to efficiently purify a range of bioproducts, including viruses. The objective is to tailor the gPore NW system, which has already shown excellent results in the recovery and purification of natural as well as recombinant proteins, to other products, like pDNA. The system, based on a delicate combination of functional and bi-component fibres integrated into a disposable and recyclable element (the cartridge), will allow for tailored flow distribution properties. These cartridges will be subjected to chromatographic experiments, monitoring parameters like product concentration, flow rate, mobile phase composition, conductivity, and pH, in order to observe their influence on the adsorption behaviour of the targeted species. Different types of ligands, e.g., ion-exchangers and affinity materials, will be tested. Dissemination and exploitation of the results were of course always considered during the project. As a research project located at the precipice of the cutting-edge, exploitation might not be immediately feasible, i.e. additional efforts will be required in the future. Nonetheless, the obtained results were disseminated in the most comprehensive and impactful way possible.
Project Results:
The main scientific and technical results of the Intenso project are described below:
Research efforts of ATPE (Aqueous Two-Phase Extraction) are centered on the development of novel strategies and methods for the selective extraction and separation of bioproducts from cells, culture media and cell lysates. This work is based on the use of mild permeabilization technologies and the application of Aqueous Two Phase Extraction (ATPE) to the separation of proteins tagged with polypeptides with predictable partitioning behavior.
One of the goals is the use of irreversible plasma membrane permeabilization, induced by pulsed electric field (PEF), for selective recovery of recombinant and native proteins from microbial cells, including different bacteria and yeast species, as an alternative to mechanical disruption, avoiding cell debris generation. PEF-assisted extraction is a relatively new, highly promising method for the recovery of bioproducts from various types of cells and is suitable for a large scale biomass processing. This technology is based on the loss of the membrane barrier function induced by high-intensity electric field pulses (electropermeabilisation) and the subsequent leakage of the soluble cytosolic content out of the cells. This research is performed by UniSofia using natural or genetically modified bacteria and yeast strains, engineered by BIOMEDAL, kindly supplied by external collaborators (ARTES Biotechnology GmBH and Gottfried Wilhelm Leibnitz University, Germany) or obtained as commercial products. The work included first the optimization of cultivation conditions for each strain (e.g., media composition, incubation temperature), in order to prevent a decrease in cell wall porosity and to increase biomass yield and bioproduct solubility. Then a suitable combination of electrical parameters ensuring irreversible plasma membrane permeabilisation was defined for each yeast strain. It was found that PEF treatment of recombinant strains provokes not only plasma membrane permeabilisation, but also an increase in the cell sensitivity to lytic enzyme as previously observed by UniSofia in different non-recombinant strains from yeast species. This finding (the first and for the moment the only data on applicability of PEF treatment for recovery of recombinant proteins from yeast) allowed developing protocols for high efficient and relatively selective release of recombinant proteins for all systems tested, based on PEF treatment and subsequent incubation of the cells with low concentration of the lytic enzyme Lyticase. This new approach allowed 70% recovery of Escherichia coli beta-galactosidase (b-gal, 545 kDa) expressed in a modified Saccharomyces cerevisiae strain, with a purification factor of 2.5, and 85% recovery of recombinant ferritin heavy chain 1 (FTH1, 480 kDa), expressed in a modified Hansenula polymorpha strain, with a purification factor of 1.8, both enzymes retaining their functional activities. Recovery efficiency in these combined, PEF/lytic enzyme treatment experiments was significantly higher than in control assays with samples obtained by standard mechanical disruption, especially in experiments performed with S. cerevisiae strains expressing bacterial laccase, where the combined treatment allowed release of this enzyme with 4 times higher volume and specific activity than in control cell lysates, indicating mechanical sensitivity or inhibition by cell debris. Similar experiments are being conducted using recombinant strains of Pichia pastoris. In order to achieve scalability, the use of a less expensive substitute of Lyticase, namely Glucanase (a commercial enzyme preparation obtained from Trichoderma harzianum, used in wine industry), has been explored. Results showed that 0.1-0.5 mg/ml Glucanase enhanced release of FTH1 in PEF-assisted extraction almost as efficiently as Lyticase, indicating that Glucanase can be used at low concentrations in the process, without a significant increase of contaminants or proteolytic degradation. Scalability and optimization of energy consumption of the PEF treatment has been explored with non-recombinant yeast available in large quantities (pressed baker's yeast, spent brewer's yeast, Kluyveromyces lactis and Pichia membranifaciens cultured in a 5l fermentor in Sofia University) and utilized as a source of total protein, native enzymes and biologically active compounds. It was demonstrated that the recovery efficiency can be preserved at high cell concentrations (up to 300 mg wet weight/ml) and at flow rates at least 60 ml/min. In these experiments new pulsing chambers with higher volume, temperature control, and changeable electrodes (Fe, Ti) were tested. Experiments on purification of the recovered enzymes (alcohol dehydrogenase, b-gal) were performed.
Dovetailed, downstream processing of PEF extracted proteins has been explored by ATPE in collaboration with UniLisbon, with preliminary results showing efficient separation of electroporated cells from extracted protein, in the case of FTH1.
ATPE procedures have been developed by UniLisbon for the separation of proteins and other bioproducts studied in the project, namely like virus-like particles (VLP), recombinant proteins fused to the Zera® peptide tag, and monoclonal antibodies. Several ATPSs were tested and optimized for the extraction and purification of a Human Immunodeficiency Virus (HIV)-GFP VLP from Chinese Hamster Ovary (CHO) cell culture. Particular focus was given to the selection of the most suitable system and extraction conditions. PEG-Sulfate revealed to be the most promising system for extraction of HIV-GFP VLP from CHO cell cultures. However, further purification and concentration is needed and after the ATPS step a back-extraction step using a new ATPS or by precipitation with PEG could be introduced in the DSP, or alternatively cation exchange chromatography using fiber cartridges. Since ATPS optimization requires the evaluation of a relative high number of parameters and volumes consumption, a lab-on-a-chip microfluidic device was developed for the optimization of the extraction conditions of an (HIV)-GFP VLP, by extraction using ATPS for expediting bioprocess design in a cost-effective manner, reducing the time and sample/reagent volumes required and lab-scale process optimization closer to the large-scale processes.
ATPE procedures have been developed also for Zera®-fused proteins. Zera® is a 93 amino acids, self-assembler proline-rich peptide domain derived from the maize seed storage protein γ-zein and developed by ERA Biotech (former INTENSO partner, acquired and substituted in the project by current partner ZIP Solutions). When fused to a protein of interest (POI) Zera® is able to induce, in vivo, the formation of endoplasmic reticulum (ER)-derived protein bodies (PBs) that facilitate the recovery and purification of fused recombinant proteins by density-based separation methods, such as centrifugation. Zera®-fused proteins can also be produced and secreted by prokaryotic hosts, like Bacillus spp cells, purified and induced to aggregate into PBs in vitro. This approach allows to further reduce production costs as well as contamination with undesired proteins. These in vitro formed PBs can be administered as vaccines and are able to induce a strong immune response already in the absence of adjuvants, since the oligomeric assembly into PBs mimic the conformational effect of adjuvants. In this project, ATPE-based separation, from the culture media, of secreted Zera®-fused proteins has been explored. It was found that the Zera® domain drives precipitation of Zera®-fused proteins at low concentrations of ammonium sulphate, and therefore a simple precipitation step with this salt, resulting in separation from the vast majority of the contaminants present in the growth medium, was applied before the ATPE process. The effect of several parameters, including PEG MW, tie-line length, pH, sample load and neutral salt addition on several Zera®-fusion proteins partitioning in PEG – phosphate ATPS were thus studied. All the optimization performed with Zera® fusion model proteins (Zera®, Zera®-Bla and Zera®-LipTla) has been then extrapolated to a product of interest, namely the fusion Zera®-PAP, a tumor associated antigen widely used in immunotherapy against prostate cancer. Zera®-PAP partitioning was predicted in ATPS of PEG/phosphate based on the partitioning studies of Zera®, Zera®-Bla and Zera®-LipTla and their properties Although these separations were just evaluated through SDS-PAGE and Western Blotting qualitative analysis, the goal was to extrapolate a good system to Zera®-PAP purification based on its MW and hydrophobicity and the results obtained with Zera®, Zera®-Bla and Zera®-LipTla separations and their characteristics. Since Zera®-PAP has 46.4 kDa and is hydrophobic, a system with low TLL, PEG MW between 6 000 and 8 000, pH between 8-9 and Loading percentage between 20 and 30%, was explored for Zera®-PAP selective separation. Thus, two ATPS conditions were tested: 10% PEG 6000 or 8000 – 10% Phosphate pH 8.5 – 25% loading, for Zera®-PAP purification. A purification scheme was proposed for the purification of Zera®-PAP which includes a first precipitation step with ammonium sulphate followed by an extraction step using PEG/phosphate ATPS. Results showed that the system with PEG 6000/phosphate provides a higher selectivity and the protein in the PEG-rich phase is mainly Zera®-PAP. After the ATPS step, the Zera®-driven in vitro aggregation into PBs has been successfully performed, allowing the isolation of PBs with a very high purity, that can be directly administered as immunotherapy. The immunogenicity of these preparations will be evaluated in vivo in tumor regression assays.
Another protein purification technology, based on the combination of the choline-binding tag LYTAG, with a self- excisable intein, has been developed by Biomedal in this project. With this method, any protein or polypeptide can be expressed in Escherichia coli as a fusion to its C terminus of an Intein-LYTAG double module, captured in a Q-anion exchanger (a choline structural mimetic), extensively washed with high ionic force and recovered (> 90%) in the mobile phase, with high purity (> 90%), by induction of intein-promoted cleavage with a reducing agent.
LYTAG has also been successfully applied to the purification of monoclonal antibodies (mAbs) by ATPE using a recombinant, dual ligand named LYTAG-1xZ. This ligand, developed and produced by Biomedal in high-density E. coli cultures, includes two separate domains: LYTAG, exhibiting affinity for PEG (another choline structural mimic) and the protein A-derived, IgG binding domain Z. This strategy exploits the ability of the LYTAG-1xZ ligand to bind to and drive partitioning of IgG molecules to the PEG-rich phase of an ATPS, and can be dovetailed with a downstream purification step based on chromatography on Q-anion exchangers. The LYTAG-1xZ ligand-based separation technology was tested with pure human serum antibodies, clear CHO cell supernatant and complex medium containing hybridoma cells and proved to be especially effective when dextran was used as the bottom phase component. Successful partitioning of mAbs to the PEG rich-phase was accomplished using 7% PEG 3350 and 6% dextran, with an increase in the partition coefficient of one order of magnitude. Integration of harvesting and ATPS affinity extraction steps were evaluated with PEG/dextran/LYTAG-Z systems proving to be an alternative strategy for integrating the clarification and the primary recovery of mAbs in a single step. An IgG recovery yield of 89% with a purity of 42% and a clarification higher than 95% was achieved with 7% PEG 3350 and 6% dextran 500,000 when using the LYTAG-1xZ dual ligand. Strong anion exchange resins containing quaternary ammonium groups (Q) were the chosen matrices to perform further purification and polishing of the mAbs. A two-step elution chromatographic method, based on the affinity of the LYTAG-1xZ to quaternary ammonium groups, was successfully developed in order to first capture the LYTAG-1xZ-mAb complexes and then sequentially elute the mAb (by pH decrease) and the LYTAG-1xZ ligand (using the LYTAG specific ligand, choline), to improve the purity obtained after ATPE and to recycle the ligand.
Three different anion exchange matrices charged with quaternary amines (a choline analogue) – CIMmultusTM QA, HiTrap Q FF and Q fiber cartridge (gPoreQ), were tested for the purpose and all were able to process the ATPS PEG-rich phase directly, without any pre-conditioning step. CIMmultusTM revealed to be the most suitable column for this polishing step, allowing a recovery yield of 99%. The overall process had an overall yield of 88% and a final purity of 94%. The polishing step allowed an increased on purity from 43% to 94%, maintaining the overall yield of the process, and the further recovery of the dual tag ligand by choline competitive elution. Cell harvesting, mAb clarification and ATPS affinity extraction steps, developed and optimized at a lab scale (2 g system), were evaluated at larger scales (up to 20 Kg system) at BOKU facilities. An extraction yield of 89%, a protein purity of 50% and a cell clarification around 100% were achieved (using ATPS optimized at a lab scale). Comparatively, and using the same conditions, in the case of the 2 kg system, values of 99.8% and 59.5% were achieved for extraction yield and protein purity, respectively and a cell clarification of about 98% was accomplished. The results obtained validated the developed platform based in ATPS as a reliable alternative strategy for the clarification and primary mAbs recovery at pilot scale. The polishing step using a Q-fiber cartridge was also scaled-up at BOKU facilities (cartridge volume of 150 ml) and similar results were obtained in terms of yield and purity compared to the lab scale (1ml volume). The overall yield and purity of the process (cell harvesting + affinity ATPS + anion exchange) was comparable to the lab scale performance.
Finally, continuous phase separation in ATPE has been explored by adapting the hydrocyclone technology. A first prototype has been developed by BHR, by combining variations in different parameters including vortex finder, cylinder and underflow diameters. Four different approaches were considered, with BHR attempting to achieve the eighty five/fifteen split of heavy to light phase objective with these low density difference systems using cyclonic techniques. This is attractive because hydrocyclones contain no moving parts and separation is driven only by the liquid flow. A small diameter commercial hydrocyclone was purchased but was found to be unable to separate the PEG / Salt / Phosphate two phase model system studied. Based on guidelines taken from published literature and consideration of commercial cyclone designs, a modular hydrocyclone was manufactured, allowing a wide range of geometric configurations to be tested. Experiments were performed using a 4˚ cone angle and a range of vortex finder, under-flow and cylinder diameters were tested. Following testing with the model ATPS system, the modular hydro-cyclone technology has been employed for the continuous separation of Lysate-fused proteins. Results were presented in terms of separation efficiency as a function of flowrate. Separation efficiency is defined as the volume fraction of a given component in an outlet, relative to the volume fraction of that component in the inlet. Both under-flow and over-flow outlets were considered. Two ATPS systems were tested as part of this project, System One, with cell lysate expressing a LYTAG-fused protein (13.0% PEG 3350, 11.5% Phosphate buffer, pH 7.0, 20.0% cell lysate, Total mass ~ 3.1Kg, Top phase: 45% (v/v), Bottom phase: 55% (v/v)), and System Two, without cell lysate (16.25% PEG 3350, 14.38% Phosphate buffer, pH 7.0, Total mass ~ 2.6kg, Top phase: 47% (v/v), Bottom phase: 53% (v/v)). Results showed that a large number of geometric factors affected the degree of separation achieved, including the cylinder diameter, vortex finder diameter, under-flow diameter and cylinder height. In most of the configurations tested, System One which was infused with proteins achieved better separation than System Two. Separation efficiencies of about 60-40% were obtained with some geometries tested using System One. In conclusion, the variation in separation efficiencies between Systems One and Two and the effect of multiple geometric variables on the performance with both Systems suggests that cyclone geometry will have to be tested and optimised with any specific system before large scale application. The effect of hydrocyclone scale could also be tested, but scale-up of an optimised geometry can reliably achieved through parallelisation.

The Expanded Bed Technology (EBA) for the direct sequestration of bioproducts from a crude feedstock was researched based on their comparative performance and industrial applicability. A classical strategy employed for the downstream processing of macromolecules has been the capture of the targeted species onto a solid phase like porous beads, modified microfiltration membranes, or reactive surfaces. The immobilization of selective ligands creates the opportunity to exploit significant adsorption behaviour differences between products and unwanted contaminants present in the feedstock. Moreover, direct adsorption methods have the capability to handle crude extracts containing suspended biological particles without previous clarification. Simultaneous concentration and selective isolation of a defined product from a raw extract or fermentation broth becomes then a viable technological pathway. This translates into increased yield and cost efficiency. For some low added value products the cost of the adsorbent is a key issue; this is many times also true for most costly (bio) products. Finite bath contacting at process scale offers an opportunity to exploit cheaper adsorbents whose lack of geometric refinement and/or lack of inherent particle density make them unsuited to fixed-bed or fluidized-bed applications. This technique can be used with advantage when recovering the target substance from a large volume of crude feedstock even in the case that suspended biological particles are present (e.g., there is no need for previous intensive clarification). A variation of this method is the sorption onto fluidized adsorbent beads, eventually with a recycle loop for enhanced capture performance. Recently, the utilization of designed adsorbents for protein purification has resulted in the introduction of the fluidized and classified adsorption system, commercially termed as ‘expanded bed adsorption’. This operation allow for simultaneous clarification, concentration, and partial purification in one single step and directly from a non-clarified feedstock. The central property of the liquid fluidised bed from the point of view of protein (and bio product) processing is the increased inter-particle distance within the bed. This allows the introduction of particle-containing feed suspensions without the risk of blocking the bed. In an ideal situation the soft-solid materials that are present in a complex feedstock will pass all the way through the column and will leave the system while the targeted compound will bind to the fluidised stationary phase. After washing, the protein of interest can be eluted and a semi-purified, clarified, and concentrated product solution will be obtained. EBA technology has been known to suffer from undesirable interactions between biomass components and fluidised adsorbent beds thereby compromising dynamic binding capacity and process performance. Effective utilisation of EBA technology requires an interaction free environment. Direct capture will be proposed for proteins of biopharmaceutical interest (either secreted or intracellular) and for mAbs. EBA can integrate the processing of such entities by simultaneous clarification, concentration, and partial purification –in a single step.
This goal of the research relates to development of downstream processing by design and scale-up of economic separation and purification processes for complex biochemical mixtures by improvement of process efficiency; e.g., low water use, water recycling and treatment. It also relates to “development of strategies for process intensification; e.g., continuous processes and novel reactor designs” and to “improvement of process development and optimisation e.g., by utilisation of sophisticated statistical tools”. A second generation of EBA adsorbents will be studied. These novel adsorbents are smaller in size that the previous generation of commercial materials (e.g. particles of smaller diameter) and also denser. The combination of the mentioned two attributes would allow for a faster fluidisation velocity due to the size / density of the fluidised particles and due to the improved mass transfer properties of the adsorbent. This, in turn, would allow for higher process productivity. Since a faster flow velocity is associated with higher drag forces that could remove attached biomass particles, lower buffer consumption would be required. Studies will be performed to understand the thermodynamics of biomass deposition onto the adsorbent beads and to observe system performance as a function of critical operational parameters e.g., flow rate, biomass concentration, expansion factor, and adsorbent type. Operational windows will be defined so as to optimize overall process performance and robustness. Technological indicator will be the ability to process at least two different targeted macromolecules directly from the crude feedstock. These products will have to belong to a different class of proteins or mAbs and will have to be produced in two different biological systems. Total process yield will be optimised to reach 70+% while preserving the biological activity of the targeted species. Contaminant material will be reduced to at least 80+%. EBA will have to economise process water (buffer) usage at least 25+% in comparison with the packed bed mode of operation (baseline).
The first section of the results deals with EBA optimization using mobile phase modification strategy. It builds upon development of methods that can be adapted to chromatographic beads in order to characterize them, namely, the capillary rise method, and the streaming potential method and further optimization of xDLVO calculations, supplemented with specially designed calculators for streamlined and comprehensive energy calculations, which eventually lead to predictions for ideal mobile phase compositions. Further application of results for optimized cycle for expanded bed adsorption processes employing high cell density broths (CHO cells) for the primary recovery of monoclonal antibodies, where ideal conditions were predicted beforehand.
Our strategy to describe interaction energies has gone far beyond single point calculations to cover an entire range of conditions, which has now resulted in predictive capabilities for opti-mal EBA process conditions in terms of process condition. This is realized by measuring sur-face energy and zeta potential variations as a function of pH, ionic strength as well as salt type. The surface energy components of these hydrated beads did not vary significant over the measure range of pH and ionic strength. This is expected since the beads do not bear pH sensitive ionic groups in the measured pH range. Furthermore, similar to the behaviour of liquids like water, the surface energies of the adsorbents were also shown to remain roughly the same when ionic strength was increased. What does change drastically over pH,ionic strength and salt type is the electrostatic behaviour of the material, which is manifested from its zeta potential. The obtained characterization data (4 pH points, each with 4 points of varying ionic strengths) for both cells (CHO) and adsorbent beads were interpolated over the entire pH and ionic strength range and were used to generate interaction energy surfaces, which highlight favourable and unfavourable conditions for biomass-adsorbent interactions.

The robustness of these calculations comes into play when incompatibilities with the product of interest arise. For example, let us suppose that the product precipitates or aggregates in the presence of Na2SO4. A new salt type-dependent integrated energy surface can be generated and one can go a step further and eliminate another salt, e.g., (NH4)2SO4, which is typically used to “salt out” proteins, and generate a new salt type-dependent integrated energy surface. These calculations lead to an ideal adsorbent selection when considering similarly functionalized matrices, e.g., mixed-mode beads. Furthermore, they can used to minimize cell aggregation, which can pose a problem in EBA processes due to the relatively large clumps they form, which are difficult to wash out. The Intenso project’s work package focused on EBA deals with the optimization and scale-up of the EBA process for monoclonal antibodies in three consecutive improvement cycles. These optimizations are focused on minimizing biomass interferences and consequently improving capture of the antibodies with good yield as well as purity. This is achieved by analysing current process conditions and observing the influences of solution chemistry, both in the equilibration and loading phase (to optimize yield), as well as the column wash phase (to optimize purity). For this cycle, changes to solution chemistry can be guided by extended DLVO calculations that predict interaction energies between adsorbent and mammalian cells in given solution conditions.
A clear multivariate interaction is observed from the experimental results with high cell density CHO cells. xDLVO-based experimental design for optimization of EBA processes was planned to study the effect of pH, salt type and ionic strength, as well as the combined effect of all parameters. xDLVO calculations were made for a normal fermentation run with CHO cell densities higher than140 million/mL, cell viability greater than 85%, and IgG titres of at least 5-10 g/L. Based on experimental results we found that mAb recovery increased by ~10% after optimisation.
The second section of the results deals with EBA optimization using surface modification of EBA adsorbent using cell repellent polymers strategy. Using microplate methods cell cepellant polymers were screened and shortlisted three best polymers. The EBA experiments were conducted at various process conditions at laboratory scale using naked beads and with polymer coated using shortlisted polymers. The tested in pilot scale at demonstration.
The goal of this optimization run is to use the best of the shortlisted polymer having cell repellent property to check its efficiency in process condition. Inception of the shortlisting was done by measurement of surface energies and zeta potential of the colloids involved (Biomass and polymeric beads and surfaces), then microplate method was developed for high throughput screening of the cell repellent polymer for both CHO and yeast cells. Plot of CTI Vs kT of each interacting body enabled us to rank the polymer. This process makes an enormous difference in testing of 3 potential polymers in comparison to all twenty-six potential candidates in process condition. Our measurements and calculations have led to 3 possible cell repellent polymers which can be investigated in the first EBA optimized run using BioToolomics Protein A adsorbent (BTPA). These are (a) coating of BTPA with 0.1% poly acrylic acid in 20 mM phosphate buffer pH 7.4. (b) Coating of BTPA with 0.1% poly methacrylic acid in 20 mM phosphate buffer pH 7.4. and (c) coating of BTPA with 0.1% poly vinyl sul-phate in 20 mM phosphate buffer pH 7.4. Best of the two polymers were tested of Protein A adsorbent from DSM biologics having a Biomedal protein Ligand (DSMPA).
The dynamic binding capacity (DBC) at 10% and 50% breakthrough for the IgG with combination of type of protein A adsorbent and polymer coating was determined by frontal studies. The UpFront FastLine 10 column was packed was packed with 5 ml of adsorbent and mounted to an ÄKTA FPLC/Explorer 100 system and equilibrated with 20 mM phosphate buffer (pH 7.4) and 150 mM NaCl. A solution of IgG (3 mg/mL) was loaded at linear flow velocities ranging from 150 cm/h for BTPA adsor-bent, 300 cm/h for DSPA-PAA adsorbent and 450 cm/h for DSMPA, DSMPMA. In expanded bed mode experiments were performed with cells and without cells. In case of without cells, the bed was loaded with 3 mg/ml of IgG in equilibration buffer and the breakthrough was estimated. In case of with cells, load consist of 10 mil-lion/ml CHO cells with more than 95% viability was spiked with 3mg/ml if IgG to mim-ic the fermentation broth. This was loaded on the column with respective flowrate and the IgG concentration was estimated using ELISA and SPR for BTPA and DSMPA adsorbents respectively. Experiment with BTPA and coated adsorbents were conducted at 3 stages. Every stage consists of one control (naked) and a polymer coated adsorbent. Experimental order was as follows, PAA, PMA and PVS. Bed was stable with Naked, PAA and PMA runs, whereas with PVS bed was collapsed at the mid of the run. This can be due to the attraction between the beads or due to the deposition of the CHO cells on PVS coated beads. Experiments with DSMPA adsorbents were conducted in single batch. Here the main observation is PAA coating resulted in increases bed expansion factor at less cross flowrate in compared to the naked and PMA coated adsorbent. This gives the advantage of more residence time for the protein to interact with the adsorbent. PAA coated adsorbent gave 2 expansion factor at 300 cm/h whereas naked DSMPA and PMA coated adsorbents were required 450 cm/h to expand to the factor of 2.
Effect of the cell repellent polymer coated adsorbent was tested pilot scale as INTENSO project demonstration. The cell repellant polymer was shortlisted based on xDLVO calculation and cell deposition experiments conducted using microplate method. The Dynamic Binding Capacity Values were increased ~ 50% in the presence of PMA-shielded EBA Beads (Prot. A). this clearly demonstrates the effectiveness of the cell repellence property of the polymer at pilot scale experiment. Also proves, colloid theory is an effective tool to guide EBA optimization and material design.

The main results, technical and scientific, of the convective flow sorption (CFS) will be outlined below.
Different viruses and virus-like-particles (VLPs) were successfully purified using CIM® (Convective Interaction Media) monolithic supports and an upscale of developed downstream processes kept the advantages of methods developed on laboratory scale.
A chromatographic process based on monoliths for purification of infective Baculovirus without prior concentration step has been established. Produced Baculovirus were harvested by centrifugation, filtered through 0.8 m filters and directly loaded onto radial 1 mL anion exchange chromatographic monoliths operated at a volumetric flow rate of one bed volume per minute. The virus was eluted with a step gradient of NaCl. Recovery of infectious virus was highly influenced by composition and age of supernatant. Substantial protein and DNA content decrease was achieved in main virus fraction. Infective virus could be 52-fold concentrated within 20.5 h of downstream process time and simultaneously an 82-fold volume reduction was possible when loading 1150 mL (2.1×108 pfu/mL) onto 1 mL scale support.

The second studied application was development of chromatographic purification of enveloped virus-like particles. These are increasingly used as vaccines and immunotherapeutics. Frequently, very time consuming density gradient centrifugation techniques are used for purification of VLPs. However, the progress towards optimized large-scale VLP production increased the demand for fast, cost efficient and scale able purification processes.
We have developed two orthogonal methods for purification of VLP directly from the clarified culture supernatant based on anion exchange and HIC/mixed mode monoliths. The methods are able to separate VLP from aberrant forms and other extra cellular particles.
The first developed chromatographic procedure for purification of HIV-1 gag VLPs produced in CHO cells is based on anion-exchange chromatography. The clarified and filtered cell culture supernatant was directly processed on an anion-exchange monolith. The majority of host cell impurities passed through the column, whereas the VLPs were eluted by a linear or step salt gradient; the major fraction of DNA was eluted prior to VLPs and particles in the range of 100–200 nm in diameter could be separated into two fractions. The earlier eluted fraction was enriched with extracellular particles associated to exosomes or microvesicles, whereas the late eluting fractions contained the majority of most pure HIV-1 gag VLPs. DNA content in the exosome-containing fraction could not be reduced by benzonase treatment which indicated that the DNA was encapsulated. Many exosome markers were identified by proteomic analysis in this fraction. We present a laboratory method that could serve as a basis for rapid downstream processing of enveloped VLPs. Up to 2000 doses, each containing 1×109 particles, could be processed with a 1 mL monolith within 47 min. The method compared to density gradient centrifugation has a 220-fold improvement in productivity.
In the second approach we have used hydroxyl-functionalized polymethacrylate monoliths, providing hydrophobic and electrostatic binding contributions, for the purification of HIV-1 gag virus-like particles. The clarified culture supernatant was conditioned with ammonium sulfate and after membrane filtration, loaded onto a 1 mL monolith. The binding capacity was 2x1012 per mL monolith and was only limited by the pressure drop. By applying either a linear or a step gradient elution, to decrease the ammonium sulfate concentration, the majority of double-stranded DNA and host cell protein impurities could be removed while the particles could be separated into two fractions. Proteomic analysis and evaluation of the p24 concentration, showed that one fraction contained majority of the HIV-1 gag and the other fraction was less contaminated with proteins originated from intracellular compartments. We were able to process up to 92 bed volumes of conditioned loading material within three hours and eluted in average 7.3x1011 particles per particle fraction, which is equivalent to 730 vaccination doses of 1x109 particles.
Both developed method have overall advantages in regard to purity, yield, productivity and environmental impact. In this WP also the analysis of bionanoparticles/VLP have been pre-validated using SEC-HPLC, SEC-HPLC-MALLS and Nanoparticle tracking analysis.
To serve the increasing amount of mRNA needed for project partners, mRNA production process was optimized for high yields of mRNA (up to 0.5 g) per batch. Thereby in vitro transcription (IVT), capping and polyadenylation processes were explored in detail and improved. mRNA production was optimized by varying different additives. Yields of mRNA up to 8 mg/mL IVT could be achieved. This was shown for at least two different mRNA sequences.
Radial chromatography employing monoliths was found to be efficient for purification of messenger RNA. BIA Separations together with Ethris have developed the method for mRNA purification on CIM C4 HLD monolithic support in descending NaCl gradient. Several optimization steps (such as thermal pre-treatment of mRNA material, fine tuning of sample preparation, in-line dilution with the adjustment buffer) were additionally implemented to increase the purity as well as productivity of the process. The capacity of C4 HLD column for mRNA was determined to be 4.5 mg mRNA/ml of C4 HLD monolith with the total recovery of 85%. Biological stability of CIM purified mRNA with the in-line dilution method was tested by Ethris with transfection of mRNA to alveolar epithelial cells (A549) using a standard transfection reagent and it was confirmed that purified mRNA using CIM C4 HLD monolith retains its biological activity. Additionally the method was successfully transferred from axial chromatography to radial chromatography and later on scaled up from 1 ml radial to 8 ml radial columns. The method has been also compared on laboratory scale to industry standard (which is at the moment precipitation), where chromatographic method shows comparable results regarding purity, yield and productivity. Due to proven ease of upscaling the purification process on chromatographic monoliths together with increase in productivity on larger monoliths we assume that on industrial scale chromatography would overcome the precipitation process.
Due to stringent requirements for the purity and efficacy of the plasmid DNA (pDNA) as a pharmaceutical product, chromatography is often used in the downstream process. Monolithic stationary phases are efficient for separation of large biomolecules such as pDNA. High ligand density butyl-modified (C4 HLD) monolithic support is the most used for pDNA purification and the goal of the study was a development of a hydrophobic methacrylate monolith with improved resolution for pDNA isoforms separation and removal of host contaminants. After a detailed search for an appropriate ligand, a pyridine-modified monolithic support was chosen and tested under descending ammonium sulfate linear gradient. The purification process was optimized according to the most efficient pDNA isoforms separation and the quantification of the main impurities during the purification steps was performed, as well as the purity and the recovery of eluted supercoiled (sc) pDNA isoform. Usage of pyridine-modified monoliths resulted in a more efficient separation between pDNA isoforms, with a similar dynamic binding capacity and recovery as C4 HLD monoliths (over 3 mg/mL and 90%, respectively). Pyridine and C4 HLD monoliths were equally efficient for the removal of the main process impurities, but pyridine exhibited higher purity in terms of sc pDNA homogeneity (98%) comparing with C4 HLD (95%), showing to be a suitable alternative to C4 HLD for this polishing step.
Sample displacement chromatography (SDC) is a chromatographic technique that utilises different relative binding affinities of components in a sample mixture and has been widely studied in the context of peptide and protein purification. SDC was implemented in a separation of pDNA isoforms under overloading conditions, where sc pDNA acts as a displacer of oc pDNA or linear isoform. Since displacement is more efficient when mass transfer between stationary and mobile chromatographic phases is not limited by diffusion, we investigated chromatographic monoliths as stationary phases for pDNA isoform separation. CIM monoliths with different hydrophobicities and thus different binding affinities for pDNA (CIM C4 HLD, CIM-histamine and CIM-pyridine) were tested under hydrophobic interaction chromatography (HIC) conditions. SD efficiency for pDNA isoform separation was shown to be dependent on column selectivity for individual isoform, column efficiency and on ammonium sulfate (AS) concentration in loading buffer (binding strength). SD and negative mode elution often operate in parallel, therefore negative mode elution additionally influences the efficiency of the overall purification process. Optimisation of chromatographic conditions achieved 98% sc pDNA homogeneity and a dynamic binding capacity of over 1 mg/mL at a relatively low concentration of AS. SDC was successfully implemented for the enrichment of sc pDNA for plasmid vectors of different sizes, and for separation of linear and and sc isoforms, independently of oc:sc isoform ratio, and flow-rate used. This study therefore identifies SDC as a promising new approach to large-scale pDNA purification, which is compatible with continuous, multicolumn chromatography systems, and could therefore be used to increase productivity of pDNA production in the future.
The industrial-scale monolithic columns – proving the homogeneity over the whole volume scale is described below.
BIA Separations is the leading developer and manufacturer of CIM® (Convective Interaction Media) monolithic chromatographic columns for production, purification, and analytics of large biomolecules. BIA Separations mission is to develop and produce CIM® monolithic columns of highest quality and provide superior research and method development services for purification and analytics of biomolecules.
The binding capacities for very large biologicals using CIM monoliths are as a rule much higher as for conventional columns and simultaneously very high flow rates can be used, therefore the productivity of the process based on monoliths (mass of purified product per hour) is superior which makes such columns very valuable for biopharmaceutical markets. Different CIM chromatographic monoliths were tested and studied within WP4 as CSFM for purification of delivered biomolecules.
Due to the intrinsic properties of polymerization of CIM monoliths one of the main issues is their scalability to industrial-scale size as well the homogeneity of the structure. Within WP4 the scalability of the CIM radial flow chromatography columns up to 8 L unit was proven using different chromatographic tests, such as pulse response test, separation of mixture of biological molecules in linear gradient, binding capacity measurements. Additionally, the homogeneity/uniformity of the largest scales (800 mL and 8 L columns) was proven with tests, where chromatographic monoliths were produced and first tested as a whole. Then they were cut to smaller disks that were further tested by different mechanical as well as chromatographic tests. It turned out that the disks from different positions in the large monolith gave almost identical results, proving that the large scale monoliths are uniform in structure.
Combining radial flow pattern together with appropriate housing design, columns containing chromatographic monoliths capable of very high throughput can be prepared, making them comparable to the membrane one. In this way, a high throughput 8 mL radial flow column (monolith and housing) was designed and developed for extremely high flow rates, up to 70 CV/min which is the range of the flow-rates applied on membrane columns. This was achieved by proper monolith dimensions with the height of 55 mm, inner diameter of 6.0 mm and thickness of just 4.5 mm. It was proven that resolution of columns was unaffected in a broad range of flow rates, making them the material of choice for various applications, especially for purification of macromolecules and nanoparticles on a pilot/industrial scale. The column packing and refilling process is very fast and relatively unsophisticated, making thin layered GMA-EDMA monoliths even more attractive for industrial application.
Proxcys is the leading producer of High Performance Radial Flow Chromatography (HP-RFC) columns and equipment. The suitability of Proxcys packed-bed HP-RFC column in the processing of crude feed CHO harvest was accessed. Data generated were to prove suitability of this application and create a reference point to the later to be conducted Hybrid-TFF/monolithic adsorber experiments. Although Expanded Bed Chromatography (EBA) is typically the method of choice for this type of crude feed harvest processing, the choice to place this study in Workpackage 4 was made before knowing the unexpected positive outcome of the experiments.
Specially-designed Radial flow mimic columns were packed with a custom agarose Cell-Tolerant Chromatography (CTC) gel functionalized with protein A for the specific binding of monoclonal antibodies (mAb) from a high density CHO-S cell culture. The Cell-Tolerant-RFC (CT-RFC) columns should allow processing of the crude feed CHO-S cells, passing through the packed bed via its void volume. This setup was expected to show limited suitability up to moderate cell densities with a short bed-height and the experiments within the INTENSO Workpackage 4 would provide indications to the limits of the application.
Radial (mimic) columns, slice columns, serve as linear scalable models for method- and process development of radial chromatography processes. An MD301 column, with a bed volume of 25ml and a bed height of 3cm was used for the first assessment. The scale-up column, another radial-mimic column, MP1201-MKII with a bed volume of 1L and a bed height of 12cm was used for the second test series.
Starting with the highest dilution the crude feeds with cell densities of 20*106, 40*106, 60*106 and 80*106 cells/ml CHO-S were subsequently loaded on a small scale radial-mimic MD301 column with a bed volume of 25ml and a bed height of 3cm. The experiment showed excellent suitability, simplicity and effortless processing with an average superficial velocity of 100cm/h up to and including a cell density of 60*106. The highest operational pressure was around 0.2 Bar at average superficial velocities of 200cm/h. The loading with a cell density of 80*106 could not be finalized because of column blocking.
For the scale-up test a 40 fold scale-up column was selected, increasing bed volume, but since the first experiment went so effortless, also increasing the bed-height by a factor of 4.
With this scale-up experiment, protein binding and other quantification were performed. The loading was repeated at identical conditions again starting with the highest dilution with cell densities of 20*106, 40*106, 60*106 and 80*106 cells/ml CHO-S. Surprisingly identical limit was reached since again the loading with a cell density of 80*106 caused the column to block. After clearance of the blocking it was flushed and eluted.
Considering the imperfections of the experiment, overloading it with in total 16.4 gram IgG and the need to clear the blockage before elution, still the mass balance prove OK with 92.5% and the recovery of 47.9% at the yield of 7.9 gram. The purity of 98.0% IgG was excellent. Additionally host-cell protein contamination was below 0.1%, 1.8% of the elution were IgG aggregates and 0.1% was unaccounted for.
Processing of high density CHO-S crude feed in a packed bed 12cm bed height Proxcys CT-RFC column, packed with agarose Cell Tolerant gel is feasible. By choosing and completing successfully the application of a CT-RFC column with 12cm bed height, the target process column with a volume of 200 liter could be achieved.
The cell density limit for the application will be 60*106 cells per ml which may be considered a high cell density cell culture. The effortless operation and excellent result justify further development of this application.
CIM monoliths from BIA Separations were commercialized before the beginning of the INTENSO project. On the other side the Proxcys Axcys development was originally aimed for radial (convection) packed bed chromatography (see previous paragraph). Within the Workpackage 4 of INTENSO project the existing chromatographic support was combined with tangential flow filtration, thus creating a completely new foreground. The new technology enables simultaneous filtration and chromatographic capture from crude feed streams with standard chromatographic monolithic cartridges. Using the new technology a completely new product was developed, namely BiAxcys, which is a hybrid TFF/monolithic adsorber cartridge. The prototype II, which has been successfully tested, is capable of purification of plasmid DNA from partially clarified lysate, bypassing majority of industrial standard clarification steps. This ability has great advantage in use of power, reagents, consumables and processing time.
All technological improvements achieved within the INTENSO project were finally used in a proposal of CSFM-based downstream process for large scale plasmid DNA (pDNA) purification (annual production of 900 g supercoiled (sc) pDNA). The standard process, where CSFM-based process was compared to, was based on three chromatographic steps utilizing particulate chromatographic media. The main goal of this study was an evaluation of their techno-economic performance and environmental impact. The results show considerable better indicators for CSFM -based DSP in all three impact categories. The technical indicators relating to process productivities are at least three times higher for CSFM-based process. The overall economic indicators, such as total capital investment and annual operating costs are in favor of CSFM-based process, mainly due to two reasons: a) better utilization of convective chromatographic media for pDNA purification, thus resulting in less chromatographic media needed to purify the same amount of final product; b) lower costs for equipment, especially due to lower expenses for chromatographic housings in CSFM-based process. Two environmental indicators in favor of CSFM-based process are Output environmental index for whole process (less waste is introduced into the environment in CSFM-based process) and specific chromatographic support consumption (10 times lower volume of chromatographic support per amount of produced pDNA are discarded in CSFM-based system) due to higher utilization of CSFM. The only two indicators highly in favor of standard process are specific water consumption and input environmental index for whole process. We see the possibility of improvements regarding water consumption in CSFM-based process with partial changes of pre-chromatographic process steps.

Scientific and technical results for HDC (Hybrid Disposable Cartridges) are focused on the on the development and functional evaluation of novel adsorbent materials for the recovery and purification of biomolecules from cells, culture media and cell lysates. gPore NW is a revolutionary system based on the utilization of proprietary functional fibres in a three dimensional arrangement integrated into a recyclable element that will allow tailored flow distribution properties. This hybrid disposable cartridge (HDC), an approach created by merging existing textile technology with material / polymer chemistry will be used to generate a cost-effective and disposable system supporting eco-friendly bioprocessing. Current industrial chromatographic process involves intensive pre-chromatographic filtration steps, in addition to constrains in mass-transfer exchange as a consequence of diffusion-based materials. The system is intended to afford in the downstream processing (DSP) of bio products by integration and intensification strategies. The goal here is not only to evaluate the performance of gPore NW materials, which can be functionalized with appropriate ligands and used to efficiently purify a range of biological products but also assessed so as to introduce a high degree of integration and intensification.
The core of this technology is the development of a composite material based on cellulose fibers with protein-adsorptive capabilities. The adsorptive cellulose fiber utilized has undergone a chemical treatment to allow a rapid swelling of the adsorptive material. The swelling of cellulose fiber facilitates the disruption of the internal hydrogen bonds between polymer chains, which increases the polymer backbone accessibility to analytes. Composite fibers are prepared by chemical reagents or gamma rays have been employed as radical initiators for the chemical or radiation-induced grafting techniques, respectively. The free epoxide groups of poly (methacrylate) could be further chemically modified to have a range of different functionalities onto the backbone of the adsorbent, for example, anion-exchange (qarternaryammonium) or cation-exchange (sulfonyl). The chemical changes involved into the backbone of the fiber adsorbents were followed by ATR-FT-IR, which allows studying opaque samples. An alternative chemical grafting modification with glycidol and allyl glycidyl ether was performed onto hydrophilic cellulose fibers, to replace methacrylate grafting. This process was discarded as a consequence of grafting degree yields < 5% under the best reaction conditions. Water uptake experiments revealed porosities in the range of 90% of the adsorptive fibers, indicating that most of the water was able to freely move inside the fiber structure, allowing solute diffusion. The degree of swelling of fibers containing different ligands was in the range of 2.95 to 3.95 g/g in distilled water at 24 °C. These swelling values were not higher enough to affect the physical integrity of the fibrous material. The SEM images shows that the modified fibers have an average diameter of 5–20 μm in the dry state and ~20 μm in the wet state. The overall morphology of the adsorptive fibers showed the same overall aspect as virgin cellulose fibers.
The confocal microscopy studies revealed the distribution and high protein adsorption as a consequence of the protein penetration inside the fiber body, transformed in a hydrogel structure. As a consequence of its high swelling ratio fibrous adsorbent has very low dry packing density (0.2 g/mL). Experimental determination of the theoretical plate height (H) was performed at different linear flow velocity. A column (5-mm diameter) filled with randomly packed qarternaryammonium (Q) fibers were compared to Q Sepharose FF adsorbents. For the fibrous column, the H is nearly independent of flow velocity for a wide range of flow rates up to 600 cm/h. This behavior contrasts conventional porous mediums, like Q Sepharose FF columns, for which the H is strongly dependent on flow velocity. The resultant H in the range of 0.5–0.8 mm and the measured peak asymmetric (As) values (≤1.6) of the fibrous column demonstrated good packing efficiency with good reproducibility and were comparable to bead packed columns. The pressure drop in a porous medium, at a low Reynolds number, was determined by Darcy’s law. The permeability coefficient was 0.95 × 10−7 cm2 and 1.51 × 10−7 cm2 for the 5-mm and 50-mm diameter columns, respectively. These experimental values are between one- and three-times greater than membranes developed for similar purposes and between one- and two-times greater than the permeability coefficient of columns packed with ion exchange beads. Hydrodynamic performance was evaluated by residence time distribution analysis. The experimental determined Péclet number for Q fibrous support (≥60) remained constant, suggesting plug flow characteristics with minimal axial mixing within the bed. However, a reduction of DBC at very high flow velocity can be explained by a small increase in axial dispersion within the system. Could be possible the presence of a fiber entanglement may enhance the convective mass transfer and significantly reduce the stagnant zones by minimizing the diffusive distance. In addition, the short pore diffusion (fiber radius) could not play a significant role in affecting the column efficiency. Another important observation about the column hydrodynamic performance is the lack of column compaction at high ionic strength after the column regeneration with 1 M sodium chloride (NaCl).
Chromatography Performance of Q fibrous adsorbents, in static and dynamic conditions, was evaluated using the model protein bovine serum albumin (BSA). From the finite bath adsorption experiments, the static capacity of fiber was found to be 140 mg/mL. The dynamic breakthrough curve analysis (BTC) was carried out as a function of superficial velocity in two columns (5 mm and 50 mm diameter). It has been observed that there is a linear scalability in the performance (~5 mg/mL deviation) of the fibrous columns at increasing flow rates with an increasing column volume. Q fibrous adsorbents compared to Q Sepharose FF showed nearly identical BSA adsorption capacity at low flow rate and two-fold higher capacity at 600 cm/h. The binding capacities reported are also higher compared to strong anion exchange monolithic adsorbents (30 mg/mL) of CIM-QA (BIA Separations, Villach, Austria) and 40 mg/mL for UNO-Q monoliths (Bio-Rad, Hercules, CA, USA). Dynamic binding capacity (DBC) of low and high MW proteins (BSA—66.5 kDa; thyroglobulin—650 kDa) was determined under standard conditions. The breakthrough at 10% saturation for the thyroglobulin was low (6 mg/mL) when compared to BSA, as it is expected for hydrogel-bead adsorbents. However, the DBC for thyroglobulin results in being almost two-times higher than DBC of Q Sepharose FF resins. Q fibers have a significant improvement in the DBC for samples having MW proteins up to 650 kDa.
The elution step of adsorbed protein mixture on sulfonyl (SP) fibers and SP Sepharose FF columns, under different flow rates, was compared. SP fibers show a higher resolution than the commercial Sepharose, since the three adsorbed proteins of the protein mixture show higher resolution than Sepharose, using a gradient of 20CVs. Resolution parameter was maintained with the fibrous adsorbent even at high flow rates up to 900 cm/h. So increasing the flow rate by a factor of 10 is possible keeping the same resolution factor. In contrast, SP Sepharose FF shows already a loss of resolution with an increase of factor 3. The increased resolution allows a use of 50% of elution concentration with a possible reduction of the buffer volume and time saving. Artificial mixture of 5 g/L lysozyme (as a model protein to purify) and 15g/L yeast cells was produced, which realistically simulates the feedstock solution to a purification of macromolecules. The whole cell extract was given after the cell disruption on an FPLC system at a linear flow velocity of 150 cm/h. The material shows almost no pressure and no clumps, indicating high biomass compatibility. The elution of the model protein lysozyme was selectively bound to SP fibers and then recovered as a very pure protein from unfiltered biomass.
Monoclonal antibodies (mAbs) are among the most promising compounds on the market and one of the pipelines of pharmaceutical biotechnology. mAbs purification by affinity chromatography has commonly been used with Staphylococcus aureus protein A (SpA) as the ligand. In order to improve the productivity of this ligand it was designed a novel synthetic protein, called AviPure, containing only two domains of the native SpA protein in addition to a linker. This AviPure, contains a Cys amino acid in the tag region, has been used for oriented ligand immobilization process onto the composite fibers. This classical immobilization chemistry is one of the preferred ones as a consequence of the high stability to alkaline media (column cleaning conditions). This chemical coupling is a cost effective reaction under mild coupling conditions (pH and room temperature) and additional coupling reagents are required. The AviPure-fibers were used to determine the DBC @10% of human IgG sample reaching 15 mg/g fiber which has low adsorption value than commercial products. The studies have been undergoing to optimize the protocol in order to enhance the binding capacity. An immobilized metal ion affinity chromatography (IMAC) ligand adsorbent has been developed by design of novel protein His-Cys tags for simplification of purification and protein immobilization. The presence of thiol groups, which show strong affinity for Pd(II) ions, could be used to improve the His-tag purification system by changing Ni(II) by Pd(II) ion, as a novel ion couple for IMAC chromatography. Therefore, two different Cys- and His-containing six amino acid tags linked to the N-terminal group of Green Fluorescent Protein (GFP) was evaluated (See Fig. 1). The GFP was expressed in E. coli as a model of recombinant system. IMAC-Pd(II) adsorption and elution steps were studied under different buffer conditions. Both Cys-containing tagged GFPs were able to bind to IMAC-Pd(II) matrices and eluted successfully using a low concentration of thiourea solution. The IMAC-Ni(II) system reaches less than 20% recovery of the GFPCys3 from a crude homogenate of recombinant E. coli, meanwhile the IMAC-Pd(II) yields a recovery of 45%, with a purification factor of 13. Considering the GFPCys1, the classical IMAC-Ni(II) system using imidazole buffer as elution system reaches similar results than GFPHis6 protein (Kikot et al, 2014). The IMAC-Ni(II) fibers were utilized to direct recovery of GFPHis6 recombinant protein from a lysate of recombinant E. coli fermentation broth. Whole process was developed in a Post-graduate course attended by graduate students from Brazil and Argentina.
Here the module fabrication and related equipment had led to a prototype version and a cartridge volume of 145 ml and 250 ml were applied for process integration applications. The SP and Q gPore NW material was uniformly homogenized and randomly packed in to the said cartridges and packing efficiency was evaluated. This indicates that uniform packing density of the resulting fiber-packed column is known to have good packed performance. Breakthrough experiments of these fibrous adsorbents for operation in the frontal mode using standard model proteins have been shown high dynamic protein adsorption capacities around 200-300 mg/g operated at linear flow velocity of 300 cm/h. The cartridge was subjected to different operational tests include backpressure tolerance, chemical/physical resistance and evaluated chromatography performance with model proteins. The production of functional adsorbents have been scaled up according to the proprietary technology under optimized scalable conditions for conditions for protein production applications. In this work, SP gPore NW cartridges were utilized in capture and intermediate purification and Q gPore NW cartridges employed in the polishing application of mAb purification.
A model HDC system, based SP gPore NW module was demonstrated for protein production cycle to support an integration effect when utilized in early recovery and purification of mAb (~150 kDa) from crude feedstock (CHO-S cell-line, DSM proprietary culture system). The study was executed by employees of ChiPro GmbH (Bremen, Germany) at DSM Biologics facility in Groningen (Netherlands) to execute the process integration experiments. The system was able to tolerate solid suspension (loading up to 30E6 cells/ml) with minimal back pressure (>0.5 bar) while direct processing of crude feedstock operated at flow rate (150 cm/h). The adsorption capacity of the system for mAb was 35 mg/g, with 85% total recovery and the average purity was ~84%. The system showed high binding capacity (55 mg/g) and improved product purity (>85%) with high recovery (~90%) when processed with partially clarified cell free harvest. The system illustrated an integration effect by replacing cell clarification steps and indicating high biomass compatibility with reasonable binding capacity at high process flow rates.
HDC base gPore NW technology was evaluated at pilot demonstration scale of downstream mAb purification train by integrating expanded bed adsorption (EBA) in capture step and gPore NW modules in intermediate purification and polishing application. The integrated processing deliverable demonstrates that high cell density feedstock can be directly processed by EBA in the early recover and purification of mAb product and subsequently the EBA capture eluate was loaded on to the mounted SP gPore NW cartridge (250 ml) on ÄKTA pilot system in bind and elution mode in order to remove the contaminants in intermediate purification step. Finally Q gPore NW module (145 ml) was employed in the polishing step of mAb in a negative chromatography mode in order to remove the process related impurities that increases the resolution of the target product in the purification process train. The trace impurities were bound to the media and the target product was collected in flow through. And the system was striped using high salt concentration to remove trace impurities and residual proteins from the adsorbent and then stored in the 20% ethanol for subsequent steps. From the experimental results, it is clearly demonstrated that the final target product is obtained with highest purity of 96% from the initial crude feedstock containing the mAb purity of 25%. The total recovery of the target mAb obtained is of 75% and the step recovery in intermediate step and polishing steps are of 92% and 99% respectively. The residual impurity levels of HCP and DNA were not available due to lack of analytical infrastructure.
In another cycle, HDC base gPore NW technology was evaluated at pilot demonstration scale of downstream mAb purification train by integrating aqueous two-phase extraction system (ATPS) in early recovery step and gPore NW system in polishing application (an affinity chromatography on anion exchange matrices). A biphasic system composed of 6% PEG 3350 Da and 7% dextran 500 kDa proved to suitable for the recovery of mAbs in the PEG-rich phase, when LYTAG-Z, a ligand with affinity for PEG and antibodies, is added to the system. The effect of ligand in the partitioning of the antibodies revealed that a mass ratio of antibody to ligand of 1:2 allows optimal extraction of antibodies to the PEG-rich phase, allowing a recovery yield of 89% ± 0.03 and a purity of 43% ± 0.03. A two elution chromatographic method, based on the affinity of the LYTAG-Z to quaternary ammonium groups, was successfully developed in order to separate the antibody from the ligand and to improve the purity obtained after ATPS extraction. A Q gPore NW column (1 ml column volume) was tested at small scale for this purpose in a bind and elute chromatography mode and able to process the ATPS PEG-rich phases directly, without any pre-conditioning step. The results at small scale test for this polishing step, promising an overall yield of 82% and a final purity of 94%. The same system was scalable at pilot demonstrate scale utilizing Q gPore NW (150 ml cartridge volume) operated at a flow velocity of 300 cm/h and obtained target mAb yield of 81% and purity of 89% .The residual impurity levels of HCP and DNA were not determined in the experimental data. Hence the process should be reevaluated for optimal process performance under optimal process conditions. The gPore NW system allows an intensification effect with a reduction of the buffer (water) consumption by operating at higher flow rates at low working pressure condition with low energy consumption, and process time savings is possible under optimized conditions for optimal process performance. Therefore the demonstrated twostep purification system utilizing intensified technologies can be targeted on process integration and intensification to increase productivity and process economics for mAb purification operations.
The work development towards objectives and for each task of work package 5 was successfully achieved and delivered. The HDC will allow the integration and intensification of downstream purification of biological products (recombinant proteins, mAbs) for scale-up economic separation and improvement of process efficiency. They can be utilized in all three chromatography steps of capture, intermediate purification and polishing application in biomolecule downstream purification train. They offer as a single use bio separation devises and facilitates high productivity with selectivity and specificity while operating bench to process scale. HDC eliminate column packing activities at on-site facility and minimize clean in place (CIP) and cleaning validation procedures. HDC can be targeted mainly for smaller and medium batches of valuable products in downstream applications. Adsorptive fibers can be used as chromatographic media that will be integrated in to a disposable element for downstream processing of protein based products, improving cost effective and eco-friendly bioprocess solutions. The straightforward modification of cellulose fibers (cotton fibers) to prepare these protein adsorptive fibers opens the possibility to introduce the textile processing technology to scale up the HDC to employ in bioprocess applications.
Conclusion
Main goal of INTENSO project was an evaluation of the current situation of the downstream processing scenario with the aim of identifying inefficiencies and concomitantly introduce a debottlenecking overarching strategy. One of the goals was to enable the efficient and economically feasible purification of biomolecules beyond the traditional enzymes, proteins, and monoclonal antibodies. These would include entities like pDNA, VLPs, or vaccine viral particles, which are much larger in size. All of the studied biological molecules are part of most industrial R&D pipelines and offer an excellent opportunity to introduce innovative bioprocessing.
During the timeframe of the project industrial and academical partners have achieved intensification as well as integration of different up- and especially downstream processing steps on the basis of a multidisciplinary approach.
To conclude, chromatographic monoliths are almost ideal for purification of very large biological molecules in terms of separation power, capacity, recovery and molecule integrity. Radial chromatography enables scale up of chromatographic monoliths to industrial scale without compromising the speed of operation or separation efficiency. Additionally, we have developed a hybrid TFF/monolithic adsorber cartridge – BiAxcys module, enabling crude feed processing with standard CIM chromatographic monoliths monolith cartridges. The BiAxcys prototype II is capable of purification from crude feed, bypassing all industry standard clarification steps. This ability has great advantage in use of; power, reagents, consumables and processing time.
Not only monoliths, chromatographic beads were tested in radial chromatography format as well. The suitability of Proxcys radial column, packed with protein A beads in the processing of crude feed CHO harvest was successfully demonstrated up to the 1 L volume scale and bed thickness, which allows upscaling up to 200 L column.
Potential Impact:
The biotechnological sector is one of the fastest growing industries with the most significant impact on human life on earth. Come to think about it, there are very few processes that are not mediated by biotechnological science, one way or the other. Simply defined, Biotechnology consists of “the use of biological processes, organisms, or systems to manufacture products intended to improve the quality of human life” (Source: http://whatis.techtarget.com/definition/biotechnology) and more specifically, biotechnologists have broken down their science into four subdisciplines called red, white, green and blue. The so-called Red Biotechnology, “involves medical processes such as getting organisms to produce new drugs or using stem cells to regenerate damaged human tissues and perhaps re-grow entire organs (ibid). Thus, the biopharmaceutical industries are the main carriers of product development and innovation in the health care sector.
The health care sector and the pharmaceutical industry in particular, is extremely important for the European and North American economies; just to give an idea: the EU pharmaceutical sector produced an output of € 220 billion and employed approximately 800,000 people in 2012. It accounts for around 1.8% of the total manufacturing workforce and is one of the industries with the highest labour productivity (European Comission (2014): Commission Staff Working Document Pharmaceutical Industry: A Strategic Sector For The European Economy”. Brussels.) Furthermore, the EU was the world’s major trader in medicinal and pharmaceutical products in 2013, with total trade amounting to € 156.9 billion (EU28) and the value of exports reaching more than € 107.4 billion (ibid). In the United States, the biopharmaceutical industry contributions to the economy are astonishing: it employed nearly 854,000 individuals in 2014, and supported more than 3.5 million additional U.S. jobs, benefiting directly and indirectly, the creation of more than 4.4 million U.S. jobs in 2014. Furthermore, the biopharmaceutical industry accounted for more than $1.2 trillion in economic output, representing 3.8 percent of total U.S. output in 2014. This total economic impact included $558 billion in revenues from biopharmaceutical businesses and $659 billion from suppliers and worker spending. (Source: http://www.phrma.org/media/economic-impact). In spite of its massive influence, the pharmaceutical industries worldwide are confronted with major societal changes that constitute both threats and opportunities for the sector, as it has been highlighted in European Comission reports. Some of the major threats and opportunities are:
Demographic Change: being a key challenge for the EU, the number of EU residents aged 65 and over will most likely increase over the next 50 years, from 92 million in 2013 to 148 million in 2060. As the elderly are more vulnerable of experiencing chronic diseases and increased treatment costs, demographic transition is considered a major challenge for sustainability of health and care systems. By 2060, public expenditure on acute and long-term care is expected to increase between 8.5 and 9.1% of EU GDP (European Comission (2014): Commission Staff Working Document Pharmaceutical Industry: A Strategic Sector For The European Economy”. Brussels). As this situation concerns member states as a threat to long-term sustainability of excellent quality medical care, it appears to be an opportunity for large pharmaceutical companies that could potentially capitalize on the specific needs of the aging population; a group which given welfare-state reductions, could be willing to pay for healthcare beyond what their social security systems provide for.
Global Epidemics: in a time and day when travelers can take diseases from one world region to the other through trasatlantic flights and spread it in days to large populations at international airports, (some examples of potentatial dangers can be found in the Zika virus, Chicungunya Fever and Ebola), there are old and new diseases which pose new public health challenges. These challenges have also been clearly identified in the recent WHO (World Health Organization) “Priority Medicines for Europe and the World” Report 17 to which the Commission has contributed. These developments remind us of the need for a reliable supplier in order to ensure the availability of medicines in times of crisis and have sparked a response at EU level with the elaboration of a “Joint Procurement Agreement of medical countermeasures” (Ibid). These fast and highly critical social events pose new pressures on biopharma companies, which are expected to provide solutions, while at the same time, the demanded actions might not be viable in financial terms. The ethical questions that that the “for-profit” orientation of much of the pharma industry raises will continue to be heavily scrutinized by social interest groups and politicians. The industry response must be a careful balancing act that addresses public opinion, while at the same time, remains sustainable.
Brexit: On June 23, 2016, the United Kingdom voted to leave the European Union in an advisory referendum referred to as the “Brexit” (British exit) vote. In this context, few realized the implications that the vote will have on healthcare, and in particular, the pharmaceutical industry, which generates over 10% of the UK’s gross domestic product (GDP) and employs 70,000 people. The cultural change and the economic impact in the British Pound are already undeniable, and yet, it is expected that the the consequences will continue to be far-reaching. For the UK´s pharmaceutical industry, every stage of drug development and commercialization will likely be affected, from the earliest stages of research to pricing and reimbursement throughout the world, and while it is yet to be seen whether the UK will have continued access to the European market, it has three main routes to EU market access: the European Economic Area (EEA), the European Free Trade Association (EFTA), and the World Trade Organization (WTO) (Source: https://www.pharmaceuticalonline.com/doc/brexit-s-impact-on-the-global-pharmaceutical-industry-market-authorization-pricing-0001 ). Each one of these options presents its advantages and disadvantages, and the risks are clear: either pharmaceutical manufacturers advocate and agree with the UK government about the next steps aimed at ensuring the industry and minimizing potential catastrophic consequences, or the British pharmaceutical industry will be entirely separated from the EU system, to the detriment of both the UK and EU. The “Brexit” trend may also carry unpredictable outcomes for EU and US pharma industries at large, particularly when voters are supporting protectionist models that increase commercial barriers and seek to deter the free flow of goods and products.
In light of these potential opportunites and threats, joint decision making between national states and the biopharmaceutical industries seems more relevant than ever. Therefore, the importance of IMI, a parthership between the EU and the European Federation of Pharmaceutical Industries and Associations (EFPIA) which seeks to move sustainable innovation forward. Through the IMI-2 programme, IMI has a budget of €3.3 billion for the period 2014-2024 (Source: http://www.imi.europa.eu/content/press-release-socioeconomic-report).
The focus of its Strategic Research Agenda (SRA) for the period 2014-2024 is on delivering ‘the right prevention and treatment for the right patient at the right time’ by keeping a strong focus on developing new medicines, as well as tools and methods to speed up patient access to new medicines.
The IMI goals are consistent with social and global change trends which will continue to demand constant medical innovation (to face the threats imposed by ever changing markets; increased aging populations and global epidemies) as discussed above, as well as the most advanced form of medicine: patient-tailored programs –also known as personalized medicine- where the future of the industry seems to be.

As it is stated on the Intenso website http://intensoproject.eu/project-outline/ , the project “works on identifying the bottlenecks of the currently used downstream processing methods and finding new solutions to circumvent them” In the context of leading the way in the process of supporting the development of new downstream processing technologies, the Intenso team came to realize that, in addition of working in the development and support of new technologies, it is absolutely necessary to gain in-depth understanding of the decision making contexts in which organizations decide upon investment in new technologies. As those who work in this field know, the process of acquisition of a new technology is one where many lenses and perspectives are present: that of scientists and researchers; business and supervisory organizational teams; national and international regulatory frameworks and last but not least, ongoing relationships and commitments between suppliers and customers.
The results obtained during the project provide an insight on various aspects related to decision-making processes in the context of biotechnological companies that either develop or adopt DSP technologies mostly in the European context. These results should be thoroughly considered when crafting a strategic agenda aimed at reaching out potential customers for new technologies, and also, for understanding of organizational buying behavior in the biotech industry.
The question is not whether quality should be sacrificed at the expense of costs, or whether companies are willing to make significant investments for the sake of quality. The trend is a “value for money” mentality where decision-makers want to maximize the financial advantage of their investment, while at the same time buying the best possible technology that meets their own special needs and the regulatory frameworks. This mentality is shared by both business and technical staff in most bio-industry organizations.
Typically, decision-makers will try to upscale their technologies and work with what they have, unless the market places a significant pressure for change. In this sense, suppliers must overcome concerns that companies might have about their capacity to provide robust technologies that meet their needs and remain in a highly competitive market. If clients perceive that they run the risk of not fulfilling their business and regulatory goals because of the adoption of a new technology, they will not go for it, no matter how attractive the technology might be. In this sense, they seek for warranties that their investment is worth it, and in this sense, the name, size and technical/business capacities of suppliers increase trust in this business interactions. This poses particular challenges for startup companies.
Biotech organizations are trying to maximize their profit and decision-makers will be guided by business principles. Any new technology that is brought to their attention has to make a business case, and the gain promised has to be significant. It is important to consider than in many organizations, decision-making processes are very hierarchical, particularly when business management is involved; therefore, providers have to speak a managerial language that is equally focused on business and science.
It is particularly tough for new small and startup companies to penetrate a highly regulated market where customers are reluctant to seek change. This is particularly true of pharma companies which have to comply with very strict regulatory frameworks. However, it is possible to enable a potential investment if the product is highly innovative and it makes market sense; also if it offers performance improvement regarding yield and purity outcomes in products; if customer companies are assured of robustness and scalability and if it has a scientific base. In this context, small companies will be smart in getting joining large scale projects like Intenso, where they will be building a case for all these requirements in supportive scientific environments.
Decision making for new DSP technologies is context based. It depends on the size of the organization; the focus areas of decision-makers; the ways of interacting when management or business people are making the final decisions, and even, how conflict is deal with, will contribute to a particular kind of business outcome.

Main dissemination activities and exploitation
At the beginning of the project, a wide variety of dissemination actions were identified as possible forms to disseminate and communicate the project findings. The partners agreed on a dissemination strategy and the consortium was very active in disseminating the project idea and later on the project results. The aim was to inform as many groups and related companies as possible dealing with the downstream processing as well as the European biotech industry and related fields. This plan was monitored and the consortium was able to conduct successfully most of the defined actions.
At the beginning of the project the project partners started with internal workshops to inform their employees about the project and gain useful feedback for the internal project team. During the project the partner’s customers were updated by internal demonstration events about the project outcome.
The project dissemination activity started with a project description which was aimed to be implemented in each partner’s intranet. A project logo was designed for a common presentation to the public or scientific audience. The project website http://intensoproject.eu/ was curated as a communication tool and kept updated with the latest news of the project to reach a broader public and to get in touch with interested companies. The website acts as an overview and repository of the information generated within the project and used for dissemination purposes. This can include, but is not limited to, posters presented at conferences, links to relevant papers and digital versions of the flyers or to interesting conferences. It’s used as a communication tool between the project and interested partners of the European biotech industry, research institutions and device manufactures. The website was kept updated all the time. For distributing project news the consortium prepared newsletters. The project newsletters are also published there and used to supply interested readers with insights into the motivation and goals of the participants and to keep the interested audience up-to-date about the project findings. The intention was to edit the newsletter in a regular way regarding the information about the project progress. During the project four newsletters were published. A project flyer was prepared with the main information about the project idea. It was distributed at several conferences, for example at the “Micro Biotec 2013” as well as at workshops forums. The cross-fertilization activity brought the project in touch with the project BioIntense, located at the Technical University of Denmark, Department Chemical and Biochemical Engineering, also a FP7 funded project, which is working on a similar topic than Intenso. During the project run-time representatives of the projects met twice at General Assembly meetings. The first contact was at an Intenso central meeting in Lisbon/Portugal in February 2014 and the second at a BioIntense meeting in Lund/Sweden in September 2015. Both times a representative was invited to the meeting to present their project. The meeting was fruitful both times for the projects but as the main research objectives were too different no deeper contact was traced. At the universities in Lisbon, Sofia, Vienna and Bremen internal lectures were conducted that were included to the curriculum promoting the new findings to the students. I.e. a lecture and demonstration of ATPS technique was given at the University of Sofia and in Vienna. One of the lectures for PhD students and postdocs was hold with the title: “Purification of viruses and virus-like particles by monoliths”. The project objective was challenging enough to motivate several students writing their thesis to gain the Master of Science or the PhD. Specific training material was developed to exploit the project findings and to educate the relevant workers on the new methods. The Monolith Summer School was conducted twice in Slovenia where several partners contributed to by presenting posters or talks. Presenting the motivation and results of the Intenso project at European and worldwide conferences and workshops, fairs and exhibitions provided the opportunity to not only disseminate the project to a wide audience, but also enabled communication with other experts of the downstream processing and biotechnology for the collection of feedback. Presentations were also a good first contact opportunity, paving the way for joint future commercial interests. Conferences of interest were for example the 9th European Symposium on Biochemical Engineering Science ESBES, Affinity 2013, PREP 2013, Biopartioning and Purification BPP 2013, 2014 BIO International Convention, 16th European Congress on Biotechnology, Plasma Product Meeting, MicroBiotec2015, or the 9th Annual Bio Innovation-Process & Manufacturing Leaders’ Summit.
Besides the publications in non-scientific magazines (BioPharm International) and newspapers (Weserkurier), the consortium published in scientific journals like “Journal of Chromatography A”, “Journal of Molecular Recognition”, “Separation and Purification Technology”, proceedings of the 2015 IEEE 4th Portuguese Meeting on Bioengineering (ENBENG), “Separation and Purification Technology” and others.
The research in the project was focussed on four different methodologies of downstream processing technology – ATPS (Aqueous Two Phase Extraction), EBA (Expanded Bed Adsorption), CFS (Convective Flow Systems) and HDC (Hybrid Disposable Cartridges). More information can be found below. As a continuation strategy how to enter the market within two years after the project’s end, the consortium decided to agree on a Memorandum of Understanding as a pre-contract as the best way for conducting this aim. The consortium is also planning an “open downstream processing group” (DSP). This group is conceptualized as a sustainable resource for DSP knowledge that could outlive the project and continue to fulfil Intenso’s primary goal, the strengthening of the European biotech industry. The result was the “BioIntenso Inaugural Conference” in Brussel that took place on January 24th. Here is the website: http://biointenso.com. This conference was open to the interested public as a first exploitation event of the project results. The topics were a brief summary of the project and its main results, a presentation of the DSP group, a speech about the trends of DSP and DSP in industry. A project video was produced and shown at the conference. It summarised the project idea, the integrated demonstrator and the ideas of “BioIntenso”. The video is available here: http://intensoproject.eu/videos.

For exploiting the project results please find a brief overview of the main project results of each pillar:
Product description: LYTAG-driven mAb ATPE and purification
The downstream process (DSP) developed for mAb production is depicted as follows: This process incorporates an initial mAb harvest / cell removal step by ATPE based on a PEG/NaPA ATPS, combined with mAb pre-capture with the bifunctional ligand LYTAG-1xZ, designed to mediate affinity adsorption to anion exchange chromatography (AEC) media such as gPore Q-nanofibers or Q-sepharose. In the next step, the LYTAG-1xZ~mAb complexes extracted in the PEG-rich phase are captured in the selected Q-medium and host cell proteins (HCP) adsorped by non-specific interactions are eliminated using a high ionic strength wash buffer (WB). Captured mAb is then eluted by disrupting its interaction with the IgG binding domain (1xZ) of the bifunctional ligand, using a low pH, mAb elution buffer (mAb EB). The LYTAG-1xZ ligand bound to the Q-medium is recovered by elution with choline and recycled. Finally, the Q-medium can then be sanitized by standard cleaning in place (CIP) procedure using NaOH, and equilibrated to be used in the next production batch.
Product description EBA
The EBA system is macroscopically not that dissimilar from a standard packed bed chromatography system. The main differences are pumps that are able to provide higher flow rates, a unique flow distribution system, and the chromatographic beads themselves. These beads are composed of statistically distributed sizes and contain a heavy core of tungsten carbide. This counteracts most of the limitations imposed by first generation EBA materials and leads to a stable fluidized bed. For optimal performance of the system, a modified flow distribution system is also included in the value offering, which allows for optimal distribution of the feedstock all throughout the column diameter. This allows the processing of greater amounts of material in the same timeframe, which reduces the very expensive residence time of a product in downstream processing facilities. Lastly, a mathematical framework has been generated that allows the calculation of the optimal process conditions for each individual product, based on xDLVO (extended Derjaguin, Landau, Verwey and Overbeek) theory. This unlocks the full potential of EBA and is considered as a service provided to potential customers, analyzing their specific product to find the most favorable process conditions. EBA has the potential to replace multiple unit operations, for example centrifugation, concentration and partial purification, thus reducing cost, energy and liquid (buffers, water) consumption, time needed and increasing the potential yield.
The operating principle of EBA is simple. Once the unclarified feedstock is applied onto the fluidized bed, the cells and cell debris travel generously around the adsorbent beads and ultimately escape through the top of the column. The target molecule interacts with the beads through specific ligands and becomes adsorbed. Subsequent elution can be accomplished either in the packed bed mode or otherwise in the expanded bed mode at reduced superficial velocity. In EBA, an appropriate stationary phase is fluidized by applying the mobile phase in an upward direction consequently, generating a bed of enlarged interparticle spaces. The expanded void portion of the fluidized bed permits the application of particle containing feedstock without the possibility of obstructing the bed. Therefore, purification of biomolecules by adsorption turns out to be promising to start with biomass containing suspensions.
Product description BiAxcys
The BiAxcys module is targeted for direct purification of biomolecules from crude feed streams. The Proxcys Axcys development was originally aimed for radial (convection) packed bed chromatography. The combination with the BIASep chromatographic monolith and housing design made by Proxcys result in a hybrid TFF/monolithic absorber cartridge, enabling crude feed processing with standard BIASep monolith cartridges.
There are two main sample distribution pathways, tangential for filtration part and transversal for capture step.
To purify directly from a crude feed, a crude feed recycle-loading path was designed from Product inlet spiralling up to the Recycling outlet. This path enables debris to recycle, and not immediately clog the Inlet filter nor Monolithic adsorber. Debris gets continuously washed off the Inlet filter and kept in suspension by high recycling feed velocity. A small fraction of feed permeates through the Outer filter and Monolithic absorber as mobile phase. The mobile phase then exits the column via the Outlet channel by the Product outlet.

The main benefits of BiAxcys module are summarized below:
• Integration of two DSP operations, namely tangential flow filtration and capture of biopharmaceutics using radial monolithic chromatography, enabling faster processing of the sample. This feature results in higher productivity in biomolecules production and less degradation of the product due to shorter exposure to the feedstock.
• 'Plug and play' and disposable bioprocess devise to provide flexibility in a multiproduct environment could generate very significant cost savings by eliminating support structure, expensive column foundations, validation costs, and plant hardware.
• Due to the monolithic stationary phase, the module is especially beneficial for the processing of large biopharmaceuticals, such as plasmid DNA, viruses and virus-like-particles.

Product description hybrid disposable cartridges
Hybrid Disposable Cartridges (HDC), a hybrid approach created by merging existing textile technology with material / polymer chemistry will be used to generate a cost-effective and disposable system supporting eco-friendly bioprocessing. gPore NW is a proprietary technology that acts as a chromatography media for unique selectivity and fast purification of target biomolecules from feed streams. The gPore NW adsorbents are generated by mixing functional fibers with inert fibers in a unique combination and integrated into disposable elements that will allow customized flow distribution properties. The system exhibits biomass compatibility and high resolution with bead like capacity and run at high flow rates with low pressure drop thereby reduces buffer cost and process time. They can be utilized in all three chromatography steps of capture, intermediate purification and polishing application in biomolecule downstream purification train. They offer as a single use bio separation devises and facilitates high productivity with selectivity and specificity while operating bench to process scale. HDC eliminate column packing activities at on-site facility and minimize clean in place (CIP) and cleaning validation procedures. They can be utilized for integration and intensification in the downstream processing for early recovery and purification of recombinant proteins and monoclonal antibodies. HDC can be targeted mainly for smaller and medium batches of valuable products in downstream applications.
The HDC Key Performance Indicators:
• Tolerance to biomass as EBA [integration]
• High flow rates as MADS [productivity]
• Increased capacity as BEADS [efficiency]
• Flexible functionality as IEX / AFFINITY [selectivity]
• Low pressure drops as easily SCALABLE
• Excellent mechanical strength and stability
• Integration & Intensification effect is POSSIBLE
• Disposability as FILTERS [Plug & Play format]

The HDC Product portfolio:
• Ion exchange gPore NW (Available)
Strong cation (SP), strong anion (Q) and weak anion (DEAE) chemistry
• Affinity based gPore NW (under development)
Protein A and IMAC (Immuno metal affinity chromatography)

Target biomolecules:
Recombinant proteins, antibody, non-antibody, enzymes, VLP’s (virus like particles) vaccines, dyes, natural bioproducts, etc.

Scale & form:
Laboratory to pilot process scale (customize) in a wide range of device formats like discs, stacks, capsules, columns and modules. Currently available HDC scale from 1 ml to 250 ml modules.

Equipment compatibility:
Products work with a wide range of standard laboratory and process equipment’s, including FPLCs, pumps, vacuum, centrifuges and syringes

Target industries:
Life sciences R&D, biopharmaceutical manufacturing, vaccine production, enzyme production, nutraceutical processing or in the food and dye sector

List of Websites:
www.intensoproject.eu

Related information

Reported by

JACOBS UNIVERSITY BREMEN GGMBH
Germany
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top