EP1144647A2

EP1144647A2 - Method of improving the primary energy metabolism of mammalian cell lines

Info

Publication number: EP1144647A2
Application number: EP00910644A
Authority: EP
Inventors: Roland Wagner; Noushin Irani; Manfred Wirth; Joop Van Den Heuvel
Original assignee: Helmholtz Zentrum fuer Infektionsforschung HZI GmbH
Current assignee: Helmholtz Zentrum fuer Infektionsforschung HZI GmbH
Priority date: 1999-02-05
Filing date: 2000-02-07
Publication date: 2001-10-17
Also published as: CA2360753A1; JP2002535991A; DE19904794A1; WO2000046378A3; WO2000046378A2; US6706524B2; US6291238B1; US20020052046A1

Abstract

The invention relates to a method of improving the primary energy metabolism of mammalian cell lines. According to said method, a structural gene for a cytoplasmic yeast pyruvate carboxylase is introduced into the mammalian cells by genetic engineering techniques and is expressed in said cells.

Description

Process for improving the primary energy metabolism of

Mammalian cell lines.

The invention relates to a method for improving the primary energy metabolism of mammalian cell lines, expression vectors for use in the method and recombinant mammalian cells with improved primary energy metabolism obtainable by the method.

Only a very small proportion of glucose, one of the main sources of energy, can be completely converted to CO. be oxidized. The majority is released as lactate and alanine. Since the energy gain in aerobic glycolysis is only small, the energy requirement is covered by glutaminolysis. This creates ammonium as a toxic by-product.

Various studies have shown that the key enzymes that convert the end product of glycolysis, pyruvate, into the citrate cycle (TCA) are only weakly active in cell lines (Fitzpatrick et al., 1993; Petch and Butler, 1994; Neermann and Wagner, 1996) (a list of the cited references can be found at the end of this description). A link between glycolysis and the citrate cycle should increase the amount of glucose used in energy metabolism and reduce the need for glutamine.

Overview of energy metabolism in mammalian cell lines In the cultivation of long-term mammalian lines, glucose and glutamine have a special position among the numerous essential components of the relatively complex nutrient medium. Because unlike bacteria or yeasts, both substrates are necessary as energy suppliers, with glutamine as the main serves the source of cellular energy (ATP) .. The degree of cellular energy supply from glutamine depends on the individual cell line as well as on the presence and concentration of glucose.

The primary energy metabolism of mammalian cells thus consists of glucose and glutamine oxidation and encompasses the metabolic pathways of glycolysis, glutaminolysis and the citrate cycle. A simplified overview of these primary metabolic pathways with branches and crucial enzyme functions is given in FIG. 1.

Glucose metabolism

Glyeolysis converts between 80% and 97% of the glucose in mammalian cell lines. However, in contrast to insect or primary cells, almost all of the glucose converted in the glycolysis is processed to lactate in transformed mammalian cell lines, and only a very small part (about 0.2 to 5%) of the glucose carbon or the glycolytic intermediates is obtained the energy-supplying citrate cycle.

The causes of the almost constant conversion of glucose to lactate or the blockage of the transition from glycolytic intermediates to the citrate cycle in almost all mammalian cell lines is largely unclear.

First of all, it is assumed that the activity of the mediating enzyme pyruvate dehydrogenase is very low due to low expression rates or permanent inhibition by irreversible phosphorylation. As a result of this incorrect regulation, large quantities of lactate are released into the nutrient medium and there lead to an uncontrolled acidification the culture. These circumstances mean a low efficiency in the utilization of the nutrients, a high glucose consumption and a low energy yield.

Glutamine metabolism

Glutaminolysis does not represent a single, continuous metabolic pathway like glycolysis, but rather forms a network of up to eight, partly interlinked, alternative metabolic routes, on which glutamine can be oxidized to different degrees and which thus lead to different energy yields and product combinations .

A large part of the glutamine is deaminated and introduced into the citrate cycle via the intermediate a-ketoglutarate, where it can be completely oxidized to C0 ₂ . In addition, glutamine can also be partially converted to the amino acids aspartate, alanine or lactate. These can either be secreted into the culture medium or, in the case of aspartate, can be introduced into the citrate cycle via oxaloacetate.

The choice of the routes for the complete or only partial oxidation of the glutamine determines the contribution of the glutamine to the energy balance of the cell. Studies on various mammalian cell lines showed that glutamine in the presence of glucose covers 30% to 65% of the cell's energy requirements, possibly even 98%. In general, the lower the glucose concentration in the medium, the higher the contribution of glutamine to the energy balance of the mammalian cell.

A key by-product of glutamine degradation is ammonium. Depending on the glutaminolytic metabolic branch can release one or two moles of ammonium per mole of glutamine, which has a growth-inhibiting to toxic effect on the cells. On the one hand there is a decrease in the intracellular pH value, on the other hand high amounts of ammonium lead to changes in the nucleotide and

Sugar nucleotide pools (Ryll et al., 1994; Valley et al., 1999), which has a significant influence on the expression of the N-glycosidically bound carbohydrate side chains of glycoproteins and thus changes the product quality of the therapeutic agent (Gawlitzek et al. 1998 ; Gram atikos et al. 1998).

Metabolism engineering in mammalian cell lines Attempts to significantly improve the growth and productivity of cellular systems probably require the radical change in certain substrate flows of energy metabolism. This requires very precise knowledge of key metabolic points.

However, since an irreversible intervention in the metabolism is associated with greater difficulties, especially in sensitive mammalian cells, very few attempts have been made to rationally design the metabolism in mammalian cells compared to bacteria and yeasts.

In fact, as far as we know today, there have only been six successful attempts at metabolic engineering in the so-called primary metabolism of mammalian cells.

Pendse and Bailey (1994) published the first approach in mammalian cell lines. The gene for the bacterial vitreoscilla hemo- was presumed that an increased intracellular ATP or energy level increased productivity. globin (Vhb) cloned into a tPA-producing CHO_ cell line. The resulting Vhb-expressing cell line showed reduced growth compared to the non-transfected control cell line, but a specific tPA production increased by 40% to 100%. Obviously, both growth and productivity can be influenced by increased cellular energy states.

Renner et al. (1995) presented an analysis using growth factors such as bFGF (basic fibroblast growth factor)

Relationship between cell cycle phases, growth and cyclin E expression. With high cyclin-E expression, certain CHO cells have a cell cycle with a relatively long G, phase and short S phase, and cell growth is relatively high. The authors then cloned a cyclin E expression vector in CHO cells. The transfected cell line showed higher growth rates than the non-transfected control cells, especially in protein-free basal media due to the now increased amount of Cyclin-E. In addition, the cell morphology and the cell cycle phase distribution were similar to the CHO cells stimulated with bFGF. This metabolic design of the cell cycle can influence cell growth and morphology.

Bell et al. (1995) cloned a vector with a glutamine synthetase gene into a hybridoma cell line. In contrast to the non-transfected control cell line, the transfected hybrid cell line was able to grow in glutamine-free nutrient medium due to its glutamine synthetase activity.

This transfection was a decisive intervention in the glutamine metabolism and thus in the primary and energy Metabolism of the mammalian cell. At the same time, the problem of poisoning from ammonium formed from glutamine was greatly reduced by the possibility of culturing without glutamine.

Rees and Hay (1995) cloned the complete bacterial threonine pathway into a mammalian cell line so that this cell line could grow in threonine-free nutrient medium. The cell line was transfected with two vectors, of which the genes for aspartokinase, aspartate semialdehyde dehydrogenase, homoserine dehydrogenase, homoserine kinase and threonine synthase were constitutively expressed. Thus, the complete bacterial pathway for threonine from aspartate was cloned into a mammalian cell line, for the cultivation of which the relatively expensive amino acid threonine is no longer necessary. At high aspartate concentrations in threonine-free medium, the transfected cell line showed the same growth behavior as the non-transfected control cell line in threonine-containing medium.

This metabolic engineering approach could make nutrient media much simpler and cheaper without reducing growth and productivity.

Another metabolic engineering approach did not aim at direct primary metabolism, but at glycosylation and thus the functionality and quality of glycoproteins such as EPO.

Schlenke et al. (1997) cotransfected a BHK cell line with vectors that shared the information for 2, 6-sialyltransferase and EPO. Compared to one only with the EPO The cDNA-transfected BHK cell line showed that the EPO produced by this cotransfected cell culture had a greatly increased N-acetylneuraminic acid content at the end of the carbohydrate side chains, which is due to the cocloned 2,6-sialyltransferase.

In contrast to the 2,3-sialyltransferase naturally occurring in the BHK cell line, the 2,6-sialyltransferase can bind N-acetylneuraminic acid to N-acetylgalactosamine residues of the carbohydrate side chains. Only with this change does the EPO possess human-identical properties that can be important for therapeutic use.

Analyzes carried out by the inventors on the glucose flow in mammalian cell lines revealed a bottleneck in the initial reaction of the glycolysis, which is catalyzed by the enzyme hexokinase.

Using an approach to metabolic engineering, the inventors have cloned a recombinant rat brain hexokinase gene into BHK cells in order to expand the rate-determining reaction of glycolysis with additional hexokinase activity and to increase the flow of substances and the amount of cell energy in the form of ATP.

In the recombinant cells, the hexokinase activity was up to 3 times higher than usual, the glucose consumption increased, the material flow in the glycolysis increased and the intracellular ATP concentration also increased. Glutamine metabolism was also activated. However, the cell growth rate decreased in the cell lines with strong hexokinase activity and high ATP rates. The production rates of the cells for recombinant Protein products, however, rose. A high ATP content does not seem to promote the growth rates of the cells, but possibly their productivity.

The results showed a very clear connection between the glycolytic material flow and cell growth: An increased relative or absolute glycolytic flow results in reduced cell growth or a reduced cell concentration, since the flow of the pentose phosphate pathway is likely to be reduced and the formation of growth-reducing metabolites how to increase UDP-N-acetylhexosamines.

In addition, a high ATP content seems to be important for increasing productivity. Because different steps of protein biosynthesis can be directly related to the concentration of ATP or GTP. For example, Jackson (1991) related the translation rate for certain recombinant proteins directly proportional to the availability of ATP.

This also explains the higher product formation rate found by the inventors in cells with increased ATP content.

Overall, there is a direct correlation between the cellular ATP content or the amount of ATP produced and the production rate or amount of recombinant protein. The productivity of a cell could also depend directly on the ATP content and thus on the energy metabolism of the cell. The object of the invention is therefore to provide a method for improving the primary metabolism of mammalian cell lines.

According to the invention, this object is achieved with a method for improving the primary energy metabolism of mammalian cell lines according to claim 1.

The invention further relates to expression vectors for use in the method and recombinant mammalian cells obtainable by the method with improved primary energy metabolism compared to the wild type.

The recombinant cell lines obtained by the process according to the invention show increased complete glucose oxidation, reduced lactate production, reduced glutamine consumption, higher ATP content and higher oxygen consumption.

Furthermore, the recombinant cell lines in batch cultures grow significantly longer than wild-type cell lines and manage longer with the available nutrient resources.

The recombinant cell lines can produce much longer with the existing nutrients, so that the total yield of the product increases and the process costs are reduced in comparison to conventional cell lines.

Further advantages that can be achieved with the method according to the invention will become clear in the course of the subsequent detailed description. Advantageous and preferred embodiments of the invention are the subject of the dependent claims.

Figure description

Figure 1 is an overview of primary metabolism in mammalian cells.

Figure 2a is a schematic representation of common glucose oxidation in mammalian cell lines.

FIG. 2b illustrates the strategy according to the invention for metabolism engineering by introducing a cyto-solically expressed pyruvate carboxylase gene (PYC2) from yeast into the cells according to a preferred embodiment of the invention.

Figure 3 shows the liberation of the 5 'and 3 "region of the PYC2 gene from flanking yeast-specific sequences.

FIG. 4 is a schematic drawing of the expression vector pCMVSHE-PYC2 and the selection vector pAG60.

The invention is explained in more detail below with reference to the figures.

In order to avoid the negative effect of a high glycolytic material flow, but to improve the utilization of glucose with regard to ATP production, a different path than in the prior art had to be taken. This started with the fact that the activities of all enzymes that link glycolysis to the citrate cycle, such as pyruvate dehydrogenase, phosphoenolpyruvate carboxylase and pyruvate carboxylase (PC), are little or nonexistent in cell lines compared to primary cells (Fig. 2a).

In particular, the lack of pyruvate carboxylase is already well known as an inherited disorder, with patients suffering from severe acidosis leading to premature death. In order to ensure that glucose and pyruvate carbon is ejected into the citrate cycle in cell lines, an additional metabolic side path was introduced according to the invention by introducing a cytoplasmic-expressed pyruvate carboxylase (PC).

The natural PC is expressed in the mitochondria and is strictly dependent on the presence of sufficient amounts of acetyl-CoA. However, since the amount of acetyl-CoA in cell lines was to be classified as critical due to the lack of supply of this metabolite via pyruvate, the route of increasing the gene dose, similar to that of hexokinase, could not be taken here.

However, pyruvate carboxylase is active in the yeast Saccharomyces cerevisiae, even without the presence of acetyl-CoA (Colowick and Kaplan).

After the gene for PYC2 had been purified from yeast-specific sequences, which could lead to problems with expression in mammalian cell lines, an expression plasmid in BHK-

Cells transfected and expressed in the cytoplasm. With the expression of a cytoplasmic. Pyruvate carboxylase (PYC2) from Saccharomyces cerevisiae has successfully expressed a yeast gene in mammalian cell lines for the first time.

This new step now enables the cells to convert pyruvate from glycolysis in the cytoplasm first into oxaloacetate and then with the cytoplasmic malate dehydrogenase already present to malate, which can then be introduced into the mitochondria for further oxidation via a malate-aspartate shuttle ( Fig. 2b).

Figure 2a shows a schematic representation of the usual glucose oxidation in mammalian cell lines. In the mitochondria of primary cells, pyruvate is normally deearboxylated to acetyl-CoA by the pyruvate dehydrogenase complex (PDHC) and introduced into the citrate cycle. Little or no PDHC activity is found in mammalian cell lines. Pyruvate carboxylase (PC) and phosphoenolpyruvate carboxykinase (PEPCK) activities are also very low.

FIG. 2b shows the strategy for metabolic engineering by introducing a cytosolic expressed pyruvate carboxylase gene (PYC2) from yeast into the cells. Pyruvate is now converted to oxaloacetate and introduced into the mitochondria as malate via the malate-aspartate shuttle. PYC2 competes with lactate dehydrogenase (LDH) for the cytoplasmic pyruvate pool, which means that less substrate is available for lactate formation. In addition, NAD ^{+ is} reduced by the cytoplasmic malate dehydrogenase (MDH _cyt ). In addition, the reverse reaction via the malate enzyme (ME) offers another useful cycle for the formation of the important NADPH. 1 = MDH _cyt , 2 = ME With this step, 3 major advantages have now been achieved in the new recombinant cells:

1. More glucose carbon can be completely oxidized in the citrate cycle.

2. The recombinant cytoplasmic pyruvate carboxylase competes with the lactate dehydrogenase for the substrate pyruvate and extracts this substrate amount for the formation of lactate. 3. The malate dehydrogenase reaction leads to the regression of NAD ⁺ from NADH, which is again available for glycolysis at the stage of glyceraldehyde-3-phosphate dehydrogenase. This step reduces the importance of the lactate dehydrogenase reaction for the regression of NAD ⁺ . Depending on the malate infiltration rate via the shuttle mechanism into the mitochondria, an excess of malate can be deearboxylated with the formation of NADPH by the presence of the malate enzyme with the regression of pyruvate. NADPH is used for many constructive processes and syntheses.

In the recombinant BHK cells, the glueose consumption could be reduced by a factor of 4 and the glutamine consumption by 2 times compared to the conventional BHK cells. In addition, a 1.4 times higher ATP content with comparable good growth could be demonstrated (Tab. 1).

The confirmation that glucose-carbon really does get into the citrate cycle was carried out with the help of radioactively labeled glucose and labeled pyruvate through the release of radioactive ¹⁴ C0 ₂ and a higher glucose oxidation underpinned. The cell-specific oxygen consumption, which is an indication of the oxidative metabolism, increased by a factor of 3 in the recombinant, PYC2-expressing cell line compared to the original cell. The lower amounts of substrate used result in a higher amount of product per substrate used, so that these recombinant cell lines can now be used to produce pharmaceutical proteins over a longer period of time using the existing culture medium, which ultimately contributes to reducing the overall production costs.

Table 1:

Substance change data for wild-type cells and PYC2-transfected cell clone cultures.

For this purpose, P-lasmids were produced in a manner known per se which carry the PYC2 gene from yeast and accordingly transfect BHK cells with them. It is clear to the person skilled in the art that other cells can also be used instead of the BHK cells. Success was verified using metabolism-specific data. The flux of glucose in the TCA is slightly increased. The oxygen consumption rate, an indication of the degree of oxidative phosphorylation, is also increased.

It also shows that the metabolically modified PYC-bearing recombinant cells have a lower glucose and glutamine consumption. At the same time, the lactate production rate drops. As a result, the growth cycle of the PYC cells in batch culture is extended compared to the unchanged cells. For batch fermentation, this means improved substrate utilization and thus an extended production phase.

In continuous cultures, the inventors were able to show that due to the reduced substrate consumption of the recombinant cells, the improved substrate utilization led to a higher

Cell concentration in the steady state of the chemostat leads.

In the future, EPO will be used as a model protein to investigate whether cell-specific productivity or the space-time yield in the fermenter can be increased.

Production of the gene construct

The cDNA of pyruvate carboxylase (PYC2, isoenzyme 2) from Saccharomyces cerevisiae was obtained from R. Stucka (Stucka et al., 1991). To see possible interactions of the coding

To avoid the area adjacent to the yeast vector fragment, the cDNA was removed by endonuclease restriction and ligation after Methods well known in the prior art with small PCR fragments trimmed down to the essential coding sequences (FIG. 3).

The modified cDNA was placed under the transertion control of a CMV promoter. The resulting pCMVPYC2 plasmid also contains an ampicillin resistance gene. It is clear that other promoters and resistance genes or selection markers in general can also be used.

Figure 3 shows the liberation of the 5 'and 3' region of the PYC2 gene from flanking yeast-specific sequences.

First, a 678 bp fragment was cut out from the 5 'end with Pvull, the missing 381 bp was ligated with PCR (polymerase chain reaction) via specific primers and a HindIII site was added before the coding sequence.

The same procedure was followed for the 3 'end via the Clal cleavage site by removing 469 bp and ligation with a 176 bp fragment. The fragment was provided with a HindIII and Smal site at the 3 'end.

Production of recombinant BHK cells The plasmid pCMVSHE-PYC2 was mixed with a neomycin resistance plasmid (pAG60) using the calcium phosphate coprecipitation method (Maitland and McDougall, 1977) in BHK-21A cells according to a modified protocol by Kriegler ( Kriegler, 1990) transfected (Fig. 4). After removing the precipitates, no osmotic shock was applied by adding glycine. The latter measures can, however, be carried out if the practical circumstances so require. There are also their transfection methods known in the prior art can be used.

Figure 4 is a schematic drawing of the expression vector pCMVSHE-PYC2 based on the standard expression vector pCMVSHE. It means: CMV: Cytomegalovirus immediate early enhancer promoter; SVpA: late polydenylation signal, splice sites, SV40 early late splice sites.

bibliography

Bell, S.L., Bebbington, C, Scott, M.F., Wardell, J.N., Spier,

R.E., Bushell, M.E., Sanders, P.G. 1995. Genetic Engineering of Hybridoma Glutamine Metabolism. Enzymes Mierob. Technol. 17: 98-106

Fitzpatrick, L., Jenkins, H.A., Butler, M. 1993. Glucose and glutamine metabolism of a murine B-lymphocyte hybridoma grown in batch eulture. Appl. Biochem. Biotech. 43: 93-116.

Gawlitzek, M., Valley, U., Wagner, R. 1998. Ammonium ion / glucosamine dependent increase of oligosaccharide complexity in recombinant glycoproteins secreted fro eultivated BHK-21 cells. Biotechnol. Bioeng. 57: 518-528

Grammatikos, S.I., Valley, U. Nimtz, M. Conrad, H.S., Wagner, R. 1998. Intracellular UDP-N-acetyl-hexosamine pool affects N-glycan complexity: A mechanism of ammonium action on protein glycosylation. Biotechnol. Progr. 14: 410-419

Jackson, R.J. 1991. The ATP requirement for initiation of eukaryotic translation varies aecording to the mRNA species. Eur. J. Biochem. 200: 285-294

Kriegler, M. 1990. Gene transfer and expression - a laboratory manual. Stockton Press London

Maitland, NJ, McDougall, JK. 1977. Biochemical transformation of mouse cells by fragments of herpes simplex virus DNA. Cell 11: 233-241 Neermann, J., Wagner, R. 1996. Comparative Analysis of Glucose and Glutamine Metabolism in Transformed Mammalian Cell Lines, Insect and Primary Liver Cells. J. Cell. Physiol. 166: 152-169

Pendse, G.J., Bailey, J.E. 1994. Effect of Vitreoscilla hemoglobin expression on growth and specific tissue plasminogen activator productivity in recombinant Chinese hamster ovary cells. Biotechnol. Bioeng. 44: 1367-1370

Petch, D., Butler, M.A. 1994. Profile of energy metabolism in a murine hybridoma: glucose and glutamine utilization. J. Cell. Physiol. 161: 71-76

Rees, W.D., Hay, S.M. 1995 The biosynthesis of threonine by mammalian cells: expression of a complete bacterial biosynthetic pathway in an animal cell. Biochem. J. 309: 999-1007

Renner, W.A., Lee, K.H., Hatzimanikatis, V., Bailey, J.E., Eppenberger, H.M. 1995. Recombinant Cyclin E Expression Activates Proliferation and Obviates Surfaee Attachment of Chinese Hamster Ovary (CHO) Cells in Protein-Free Medium. Biotechnol. Bioeng. 47: 476-482

Ryll, T., Valley, U., Wagner, R. 1994. Biochemistry of Growth Inhibition by Ammonium Ions in Mammalian Cells. Biotechnol. Bioeng. 44: 184-193

Schlenke, P., Grabenhorst, E., Wagner, R., Ni tz, M., Conradt, HS 1997. Expression of human 2, 6-sialyltransferase in BHK-21A cells increases the sialylation of coexpressed human erythropoietin. In: Animal Cell Technology: From Vaccines to Genetic Medicine (Carrondo, MJT, Griffiths, JB, Moreira, JLP, ed.), Kluwer Academic Publishers ^ Dordrecht, pp. 475-480

Stucka, R., Dequin, S., Salmon, J.M., Gancedo, C. 1991. DNA sequences in chromosomes II and VII code for pyruvate carboxylase isoenzymes in Saccharomyces cerevisiae: analysis of pyruvate carboxylase-deficient strains. Mol. Gen. Genet. 229: 307-315

Valley, U., Nimtz, M., Conradt, H.S., Wagner. R. 1999. Incorporation of Ammonium into Intracellular UDP-Activated N-Acetylhexosamines and into Carbohydrate Structures in Glycoproteins. Biotechnol. Bioeng., In press

Colowick, S.P., Kaplan, N. 0., 1969. Pyruvate carboxylase from Saccharomyces cerevisiae. In: Citric acid cycle of: Methods in Enzymology (Lowenstein, J.M., ed.), Vol XIII.

Claims

claims

1. A method for improving the primary energy metabolism of mammalian cell lines, in which a) suitable mammalian cells with one or more expression vectors, the i) a structural gene for a cytoplasmic yeast pyruvate carboxylase, ii) a suitable promoter for transcriptional control and iii) one or contains / contain several selection markers, are transformed or transfected, b) the cells are grown under conditions suitable for expression and selection, and c) surviving cell clones are isolated and / or fermented / cultivated.

2. The method according to claim 1, wherein the mammalian cells are BHK cells.

3. The method of claim 1 or 2, wherein the yeast is Saccharomyces cerevisiae.

4. The method according to any one of the preceding claims, wherein the promoter is the CMV promoter.

5. The method according to any one of the preceding claims, wherein the selection marker is an ampicillin resistance gene.

6. The method according to any one of the preceding claims, wherein the selection marker is a geneticin resistance gene.

7. Expression vector for use in a method according to any one of claims 1 to 6, which contains a structural gene for a cytoplasmic yeast pyruvate carboxylase.

8. Expression vector according to claim 7, wherein the yeast is Saccharomyces cerevisiae.

9. Recombinant mammalian cells with improved primary energy metabolism, obtainable by a method according to any one of claims 1 to 6.