WO2008107663A2

WO2008107663A2 - Inhibitory product removal

Info

Publication number: WO2008107663A2
Application number: PCT/GB2008/000726
Authority: WO
Inventors: Steven Doig; Michael Wright; Christopher John Knowles
Original assignee: Ekb Technology Ltd
Priority date: 2007-03-02
Filing date: 2008-03-03
Publication date: 2008-09-12
Also published as: GB0704025D0; WO2008107663A3; EP2129769A2

Abstract

A process for the removal of an ionic component from a cell culture medium includes the provision of the culture medium in a first compartment of a compartmentalised electrokinetic cell in which the electrodes are isolated from the culture medium. A second compartment is located between the first compartment and one of the electrodes and is separated from the first component by means allowing transfer of the ionic component between the compartments. A DC electric field is applied between the electrodes to cause the ionic component to pass from the first compartment into the second compartment. Apparatus for cultivation of animal cells, using the process of the invention, is also provided.

Description

INHIBITORY PRODUCT REMOVAL

Field of the invention This invention relates to the removal of inhibitory products from mammalian cell culture, resulting in higher cell densities and increased productivity of expressed proteins, including bioactive and therapeutic proteins.

Background to the invention Many types of animal cell can be grown as mono-cultures, under controlled conditions, in bioreactors. Such cells include mammalian cells, for example CHO cells, NSO cells, BHK cells, hybridomas and many others, and insect cells such as Sf9 cells and Sf 21 cells and many others. These cell types can be used as expression vectors for a wide variety of heterologous protein products, but have become the most widely used system for the manufacture of therapeutic protein products, biologies, for use in human medicine. Such biologies include monoclonal antibodies, macromolecular hormones and fusion products amongst many others. The key advantage that animal cell expression vectors have over their microbial counterparts, such as E. coli or yeast, is that complete post-translational modification of the gene-product can be achieved. Modifications such as methylation, hydroxylation, acetylation, as well as cleavage of signal sequences and proper molecular assembly via chaperones are all performed. However, perhaps the most important post- translational modification is glycosylation, since this often imparts specificity, activity and immunogenicity. Glycosylation patterns in animal cells, such as those mentioned above, are often close to those in humans, and therefore animal cell based expression of human therapeutics is often preferred to microbial expression due to the higher product quality that is achieved.

In their natural state most animal cells grow in close proximity to one another and have a relationship with their neighbouring cells and the strata upon which they are growing. Because of their natural history many animal cells will therefore only grow in vitro when provided with a suitable surface and this strategy is often termed adherent cultivation.

However, for mass production of heterologous products, suspension culture is preferred.

A major step forward in the development of scalable and highly productive bioprocesses was therefore the adaptation of cell-lines that could thrive without the need for surface adherence. Such cell-lines are now widely used and are commercially available.

Furthermore, the development of genetic / physiological and screening tools, such as techniques for improved transfection efficiency, multiple sites of genome integration, gene enhancers, robotic and high throughput selection methods and many other methodologies has resulted in the availability of cell-lines with improved specific productivities so that 20 to 100 pg of heterologous protein can be produced per cell per day. A third major advancement has concerned the development of chemically defined growth media capable of supporting the growth of high cell densities. Such media are now used generically and many formulations have been developed for specific cell types and applications and are commercially available from suppliers such as Invitrogen and Sigma-Aldrich Fine Chemicals and many others. Prior to these developments growth media usually contained several protein types and often blood serum. Along with problems due to cost and variability, this raised concerns over the transmission of animal viruses and other adventitious agents, and thus the inadvertent spread of diseases.

Further process development concerned the innovation of scalable vessels for the cultivation of animal cells. Because of their larger size and more fragile morphology, animal cells do not grow well in the same types of vessels as those used for large-scale microbial cultivation. Animal cells are more prone to physical damage caused by shear forces caused by a rotating impeller and can also be harmed by the shear forces at the interface of a rising gas bubble and the events that occur when such a bubble disengages. Nevertheless, with small modifications, such as the shape of the impeller used in a stirred tank or the geometry of the sparger in an airlift vessel, large-scale cultivation of animal cells is currently practised commercially. Reproducible cell growth and production of heterologous protein products have been shown from bench scale systems (1 to 10 litres) up to manufacturing scale process of more than 10,000 litres.

Many commercial animal cell cultivations can routinely produce 1 to 5 g/L of heterologous protein product. This improvement in productivity has been in part due to the development of novel cell-lines, improvements to cell-specific protein yield, commercial availability of defined and nutritional media and understanding of the importance of the growth vessel's geometry. However, key to this success has been the development of a variety of operating strategies, such as batch, fed-batch, continuous cultivation and perfusion cultivation. In this way cell densities of >10⁷cells/ml_ can be achieved in fed-batch and >10⁸cells/ml_ can be achieved in some perfusion systems.

In a batch process all the media components, excluding aeration gases and CO₂ for pH control, are added along with an inoculation quantity of cells at the beginning of the process and cell-growth, nutrient consumption and metabolite production occur unhindered. The shortcoming of this strategy is that only limited quantities of carbon, often in the form of glucose, and amino acids can be used due to the constraints of solubility and growth inhibition that they exert. The fed-batch process circumvents this 5 problem in that additions of nutrients are made to the growing culture, either daily or as a continuous or variable flow into the growth vessel. In this way the total amount of carbon and nitrogen (as well as other vital nutrients) can be increased without the associated problems mentioned above and thus greater cell densities can be achieved. However, two additional bottlenecks are encountered: A limit to the volume of feed that can be 10 added; and an unavoidable accumulation of toxic metabolic by-products such as lactate and ammonium. Perfusion technology has been developed to solve these problems. It is essentially composed of a standard cultivation vessel linked to a continuous cell separation device, either as an external loop or fitted internally.

15 Many types of cell separation device have been considered, for example cross-flow microfiltration membranes, gravity separators and many others. Whatever separation device is employed, the overall strategy of perfusion is to continuously remove and

^; discard spent media from the growing cell culture and to replenish it with a continuous

'• feed of fresh media, whilst at the same time retaining all of the growing cells within the

20 vessel. In this way a theoretically limitless volume of feed can be added, and the concentrations of toxic metabolites, such as lactate and ammonia, are kept low due to their constant removal. The drawbacks to perfusion technologies are (a) the inherent increase in process complexity results in a greater contamination potential and more difficult validation, (b) the product no longer results from an identifiable batch, (c) the

25 cells are, for the most part, kept at a low growth rate which is often inappropriate for maintaining a high specific productivity, and (d) large quantities of growth media are consumed which has clear cost implications.

Similar problems and solutions to those described above have also limited the

30 performance of many microbial fermentation processes. For example, lactic acid producing bacteria, such as Lactobacillus sp and Lactococcus sp, are inhibited by their major metabolic product, lactic acid. Other well known examples include propionic acid inhibition of Propionobacteria sp, but perhaps the best characterised is acetic acid production by Escherichia coli. Particular problems caused by the accumulation of acidic

35 metabolites such as these are a reduction in yield of biomass, a reduction in growth rate and the final biomass concentration, and a reduction in expression of heterologous products. Several strategies have been developed to solve or circumvent this generic problem. One solution has been to provide a highly controlled feeding strategy and/or only provide carbon sources that are preferentially oxidised rather than reduced, so that the production of acidic metabolites is minimised. An alternative approach, less widely practised, has been to selectively extract any toxic metabolites as they are produced; there are several reports where this has been achieved using either water-immiscible solvents with hydrophobic reactive extractants like quaternary amines, or using ion selective adsorbent beads or membranes.

A configuration of ion selective membranes, either cation or anion selective, within an electrical field, can be used to transfer anions or cations, such as organic acids and organic amines, from one phase to other. Termed electrodialysis, this technology has been developed with the aim of selective extraction of small organic acid metabolites either directly from microbial fermentation media, or from clarified microbial fermentation broth. In this way the productivity of several fermentation processes have been increased either through an increased lifetime per batch, or an increased specific productivity per cell. The application of electrodialysis to animal cell cultivations with the aim of in situ extraction of toxic, charged metabolites has not been reported. However, process improvements in culture longevity and productivity have been noted when simpler Donan dialysis techniques have been applied. Moreover, electrophoretic extraction of charged metabolites from a hybridoma culture using agar salt bridges also yielded positive results. The drawback of these approaches has been the mechanical complexity of the process and the additional expenditure of growth media against which the culture broth must be dialysed.

Statements of the invention

According to the present invention there is provided a process for the removal of an ionic component from a cell culture medium, the process comprising providing the culture medium in a first compartment of a compartmentalised electrokinetic cell in which the electrodes are isolated from the culture medium, a second compartment being located between said first compartment and one of said electrodes and being separated from said first compartment by means allowing transfer of said ionic component between said compartments, and applying a DC electric field between said electrodes to cause said ionic component to pass from said first compartment into said second compartment. Accordingly, the invention described in this patent sets out a process for continuously removing major inhibitors produced by growing animal cells cultures, including ions derived from organic acids, eg carboxylates such as lactate, and nitrogen containing ions such as ammonium. Furthermore, embodiments of the present invention can include the replacement of ions lost during the removal of inhibitory compounds.

The present invention also provides apparatus for the re-cultivation of animal cells, the apparatus comprising a compartmentalised electrokinetic cell in which the electrodes are isolated from the culture medium, a first component for containing the culture medium, a second compartment being located between said first compartment and one of said electrodes and being separated from said first component by means allowing transfer of said ionic component between said compartments, and means for applying a DC electric field between said electrodes to cause said ionic components to pass from said first compartment into said second compartment.

Detailed description of the invention

The present invention will now be described, by way of examples only, and with reference to the accompanying drawings which are as follows:

Figure 1 shows diagrammatically an electrokinetic cell having two distinct circuits; Figure 2 shows a two component cell for the continuous removal of lactic acid from recirculating growth cultures;

Figure 3 shows a two component cell for the continuous removal of ammonia from recirculating growth cultures;

Figure 4 shows a three component cell for the continuous removal of lactic acid and ammonia from recirculating growth cultures;

Figure 5 shows a three component cell for the continuous removal of lactic acid from recirculating growth cultures and the continuous replenishment of anions lost from the feed circuit;

Figure 6 shows a three component cell for the continuous removal of ammonia from recirculating growth cultures and the continuous replenishment of cations lost from the feed circuit.

Figure 7 shows the effect of lactate ions on CHO-S cell growth; Figure 8 shows the effect of ammonium ions on CHO-S cell growth; and Figure 9 is a plot of integral viable cells vs. coulombs (a measure of the current applied).

In the first embodiment of the invention, a simple two compartment electrokinetic cell is constructed with two circuits of recirculating fluids. In the first circuit, the feed circuit, an actively growing culture of mammalian cells which may or may not be expressing a product is circulated through a simple two compartment electrokinetic cell (Figure 1 and 2) and return to a cultivation vessel in a closed loop. A second circuit comprising of liquid of variable composition that is circulated in a closed loop, which includes a reservoir and the second compartment (Figure 1 and 2). The electrokinetic cell that forms a part of this apparatus can have a number of formats. In all instances it will comprise of a cathode, usually made from stainless steel but can be made from other materials and an anode, usually made from platinum or iridium coated titanium, but can be made from other materials. The anode and the cathode are separated from the feed and collection circuits by means of ion selective membranes. The ion selective membranes used^" to effect the separation of the electrodes from the two active circuits can be anion selective, cation selective or bipolar membranes. The choice of membrane used is dependent upon the configuration of the cells. To maintain an electric field in aqueous solution, simple electrolytes such as inorganic salt solutions or dilute acids or bases to aid conductivity.

For removal of lactic acid from actively growing mammalian cell cultures, a two compartment cell comprising an anion selective membrane on the "cathode" side of the electrokinetic cell, and is also bound by another anion selective membrane on the anodic side, this creates a catholyte chamber and the feed circuit (Figure 2). The collection circuit is created through the use of a cation selective membrane on the side next to the anode which in turn serves to separate this circuit from the anode. By application of a DC field, lactic acid is continuously extracted from the feed circuit transport across the central anion selective membrane to the collection circuit where it is prevented from reaching the anode by the cation selective membrane. The loss of cations from the feed solution containing actively growing mammalian cells is prevented by the application of an anion selective membrane adjacent to the cathode (Figure 2). This is essential for minimising the loss of nutrients from this stream. In a simple example, a DC electric field of two amps was applied to the system (Figure 2), the volume of each reservoir was 0.7L and collection circuit contained 10OmM Na₂SO₄ and the feed circuit contained 10OmM Na₂SO₄ together with 10OmM lactic acid. The pH was not controlled. Over a period of 60 minutes, lactic acid was completely removed from the feed circuit and accumulated in the collection circuit, demonstrating the potential to remove lactic acid from a feed stream.

For the removal of ammonia from actively growing mammalian cell cultures, a two compartment cell comprising an Cation selective membrane on the anode side of the electrokinetic cell in turn bound by an cation selective membrane creates a anolyte chamber and the feed circuit (Figure 3) the collection circuit is created through the use of an anion selective membrane which in turn serves to separate this circuit from the cathode. By application of a DC field, ammonia is continuously extracted from the feed circuit and is in turn transported across the central cation selective membrane to the collection circuit where it is prevented from reaching the cathode by the application of an anion selective membrane. The loss of anions from the feed solution containing actively growing mammalian cells is prevented by the application of an cation selective membrane adjacent to the anode (Figure 3). This is essential for minimising the loss of nutrients from this stream. In a simple example, a DC electric field of two amps was applied to a system identical to the one described previously (Figure 3), the volume of each reservoir was 0.7L and collection circuit contained 10OmM Na₂SO₄ and the feed circuit contained 10OmM Na₂SO₄ together with 10OmM NH₄OH. The pH was not controlled. Over a period of 60 minutes, ammonia was completely removed from the feed circuit and accumulated in the collection circuit, demonstrating the potential to remove lactic ammonia from a feed stream.

The accumulation of lactic acid and ammonia in cultures containing actively growing mammalian cells has been shown to be deleterious. A null CHO-S¹ (Invitrogen) suspension cell line was grown in CD-CHO supplemented with 40ml/l 20OmM L- glutamine and 10ml/l HT supplement. The effect on the growth of CHO S cells of adding exogenous lactic acid and ammonia (independently) was measured (Figures 7 and 8). It was demonstrated that both ions inhibited the rate of cell growth. At 3g/L lactate the cell number at 7 days was half that without exogenous addition. Above 4.5g/l there was no cell growth. Ammonium was equally inhibitory with cell numbers being halved at 15mM.

It was clear that both ionic metabolites were inhibitory and cell growth could be improved if they were removed, assuming no detriment to the CDCHO and that cells can grow in the electrical field.

¹ Chinese Hamster ovary cells adapted to serum-free conditions. Ion selective membranes exhibit limited degrees of selectivity. They are selective for electric charge and/ or valency. Very limited selectivity is seen in terms of molecular weight which has limited practical use. As a consequence of this, one of the key challenges faced in applying an electric field to a complex growth medium used to cultivate mammalian cells is the potential for loss of key nutrients and the deleterious effects that this may have on the growth of such cells. To investigate this, a closed loop consisting of an electrokinetic stack comprising three compartments (Figure 3) and a cell reservoir with CD-CHO recirculating inside the central feed circuit and driven by a peristaltic pump was assembled in a 37⁰C, 5% (v/v) CO₂ incubator. The pH of the growth medium was maintained by CO₂ in the headspace. Four experiments were performed each using 1000ml of media exposed to electrical voltage of OV, 10V, 20V or 30V. Each experiment lasted for 240 minutes and 200ml aliquots were collected after 60min, 120 min, 180min and 240min. These samples were analysed using a NOVA bioprofiler 400 (glucose, lactate, ammonia, glutamate/glutamine, sodium) and micro-pH electrode. If pH deviated from 1.7.-1 A, alkali (15OmM sodium bicarbonate+150mM sodium carbonate solution) or acid (10OmM sulphuric acid) was added to restore pH to -7.2. Further, the 200ml aliquots were filter sterilised and then seeded with CHO cells and the growth was monitored daily in a batch flask for 7 days. The results from this work are shown in Figure 9. The graph shows the IVC (the area under the curve of the growth profile) against the number of coulombs applied. Clearly the application of electrical current had a detrimental effect on the ability of CDCHO to support CHO growth. After 5000 units of charge had been applied the media was not able to support cell growth due to depletion of some vital nutrients. The composition of CDCHO is a trade secret and therefore it was not possible to determine the keys components present.

In an effort to circumvent this problem a further development of the invention was deployed comprising a three compartment cell (Figure 5). The central feed circuit is bound by two anion selective membranes which allow only anions to leave the circuit. A fresh medium circuit is created by creating a second compartment by applying an ion exhange membrane, preferably a cation exchange membrane between the cathode and the anion exchange membrane creating the feed circuit. A third chamber, the collection circuit which collects lactate waste from the feed circuit is created by interposing a cation exchange membrane between the anode and the anion exchange membrane encompassing the feed circuit. When a DC electric field is applied to this reactor stack, anions are stripped from the feed circuit, migrating across an anion selective membrane (towards the anode), into the collection circuit. Further electromigration to the anode is prevented by the cation selective membrane which provides a barrier to migration. Anions from the fresh medium circuit are, in turn transported across an anion selective membrane (towards the anode), into the feed circuit, replacing the anions lost from this circuit, ensuring no net loss of nutrients from the mammalian cell culture circuit. Using this embodiment of the invention, it is possible to remove only lactate from the feed stream and ammonium would not be extracted in this system.

However in a further development, a second stack (Figure 6) comprises a central feed circuit is bound by two cation selective membranes which allow only cations to leave the circuit. A fresh medium circuit is created by creating a second compartment through application of an ion exhange membrane, preferably a anion exchange membrane between the anode and the cation exchange membrane creating the feed circuit. A third chamber, the collection circuit which accumulates ammonium ion waste from the feed circuit is created by interposing an anion exchange membrane between the cathode and the cation exchange membrane encompassing the feed circuit. When a DC electric field is applied to this reactor stack, cations are stripped from the feed circuit, migrating across a cation selective membrane (towards the cathode), into the collection circuit. Further electromigration to the cathode is prevented by an anion selective membrane which provides a barrier to migration. Cations from the fresh medium circuit are, in turn transported across a cation selective membrane (towards the cathode), into the feed circuit, replacing the cations lost from this circuit ensuring no net loss of nutrients from the mammalian cell culture circuit. Using this embodiment of the invention, it is possible to remove ammonium ions from the feed stream.

The effectiveness of this approach was investigated. By applying DC field to the three compartments set up comprising of a fresh medium circuit, feed circuit and collection circuit, the nutritional status of the culture medium was maintained for a longer period of time than similar experiments where Na₂SO₄ was used in the fresh medium circuit in place of fresh growth medium (Figure 9).

The effect of DC electric fields on the growth of CHO S cells was determined. At low voltages of 0, 5 and 10, cell viability did not reduce during six hour experiments. However, at 20 and 30V the viability did decrease. Cell number and viability were measured using an automated device - CASY. Trypan-blue exclusion was used in conjunction with the CASY counts to assess cell viability. At 0, 5 and 10V, the cells appeared highly viable, round and refractile throughout. At 30V, it is thought the cells lysed immediately after inoculation. This resulted in white protein streaks seen in the feed reservoir. The cells did not die at low voltages in the electrokinetic membrane stack.

Claims

1. A process for the removal of an ionic component from a cell culture medium, the process comprising providing the culture medium in a first compartment of a compartmentalised electrokinetic cell in which the electrodes are isolated from the culture medium, a second compartment being located between said first compartment and one of said electrodes and being separated from said first compartment by means allowing transfer of said ionic component between said compartments, and applying a DC electric field between said electrodes to cause said ionic component to pass from said first compartment into said second compartment.

2. A process according to Claim 1 , wherein said ionic component is an ion derived from an organic acid or is a nitrogen containing ion.

3. A process according to Claim 1 or Claim 2, wherein the ionic component is a carboxylate or a nitrogen-containing cation.

4. A process according to any of the preceding claims, wherein the ionic component is lactate and/or ammonium.

5. A process according to any of the preceding claims, wherein at least one other ionic component is fed to said first component to replace any of said other component which is lost from said first compartment.

6. A process according to any of the preceding claims, wherein a third compartment is provided and the process effects removal of both negative and positive ions from the culture medium.

7. Apparatus for the re-cultivation of animal cells, the apparatus comprising a compartmentalised electrokinetic cell in which the electrodes are isolated from the culture medium, a first component for containing the culture medium, a second compartment being located between said first compartment and one of said electrodes and being separated from said first component by means allowing transfer of said ionic component between said compartments, and means for applying a DC electric field between said electrodes to cause said ionic components to pass from said first compartment into said second compartment.

8. A process according to Claim 1 and substantially as herein described.

9. A process as described herein with reference to Figure 1 in combination with either Figure 2 or Figure 3 or any of Figures 4 to 6.

10. Apparatus according to Claim 5, and substantially as described herein.

11. Apparatus substantially as described herein with reference to Figure 1 in combination with either Figure 2 or Figure 3 or any of Figures 4, 5 and 6.