Adv Biochem Engin/Biotechnol (2003) 85: 43–93
`DOI 10.1007/b11045 CHAPTER 1
`
`Bioprocessing of Therapeutic Proteins
`from the Inclusion Bodies of Escherichia coli
`
`Amulya K. Panda
`Product Development Cell, National Institute of Immunology, Aruna Asaf Ali Marg,
`New Delhi – 110067, India. E-mail : amulya@nii.res.in, amulya_p@hotmail.com
`
`Escherichia coli has been most extensively used for the large-scale production of therapeutic
`proteins, which do not require complex glycosylation for bioactivity. In recent years tremen-
`dous progress has been made on the molecular biology, fermentation process development
`and protein refolding from inclusion bodies for efficient production of therapeutic proteins
`using E. coli. High cell density fermentation and high throughput purification of the recombi-
`nant protein from inclusion bodies of E. coli are the two major bottle necks for the cost effec-
`tive production of therapeutic proteins. The aim of this review is to summarize the develop-
`ments both in high cell density, high productive fermentation and inclusion body protein re-
`folding processes using E. coli as an expression system. The first section deals with the
`problems of high cell density fermentation with an aim to high volumetric productivity of
`recombinant protein. Process engineering parameters during the expression of ovine growth
`hormone as inclusion body in E. coli were analyzed. Ovine growth hormone yield was im-
`proved from 60 mg L–1 to 3.2 g L–1 using fed-batch culture. Similar high volumetric yields
`were also achieved for human growth hormone and for recombinant bonnet monkey zona
`pellucida glycoprotein expressed as inclusion bodies in E. coli. The second section deals with
`purification and refolding of recombinant proteins from the inclusion bodies of E. coli. The
`nature of inclusion body protein, its characterization and isolation from E. coli has been
`discussed in detail. Different solubilization and refolding methods, which have been used to
`recover bioactive protein from inclusion bodies of E. coli have also been discussed. A novel
`inclusion body protein solubilization method, while retaining the existing native-like sec-
`ondary structure of the protein and its subsequent refolding in to bioactive form, has been
`discussed. This inclusion body solubilization and refolding method has been applied to re-
`cover bioactive recombinant ovine growth hormone, recombinant human growth hormone
`and bonnet monkey zona pellucida glycoprotein from the inclusion bodies of E. coli.
`
`Keywords. Fed-batch fermentation, Volumetric productivity, Recombinant protein, Inclusion
`body, Refolding, Purification
`
`1
`
`2
`
`2.1
`2.2
`
`Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`
`46
`
`Parameters Influencing the Productivity of Therapeutic Protein
`from E. coli . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`Cell and Molecular Biology Considerations . . . . . . . . . . . . .
`Process Engineering Considerations . . . . . . . . . . . . . . . . .
`
`47
`48
`50
`
`© Springer-Verlag Berlin Heidelberg 2003
`
`APOTEX EX1023
`
`Page 1
`
`

`
`44
`
`3
`
`A.K. Panda
`
`High Cell Density Fermentation . . . . . . . . . . . . . . . . . . .
`
`Parameters Affecting High Cell Growth of E. coli . . . . . . . . . .
`3.1
`3.1.1 Nutrient Formulation . . . . . . . . . . . . . . . . . . . . . . . . .
`3.1.2 Acetate Inhibition . . . . . . . . . . . . . . . . . . . . . . . . . . .
`3.1.3 Nutrient Feeding Strategy for High Cell Growth . . . . . . . . . .
`3.1.4 Maximum Achievable Cell Concentration . . . . . . . . . . . . . .
`3.1.5
`Plasmid Stability . . . . . . . . . . . . . . . . . . . . . . . . . . .
`3.2
`Induction Strategy for High-Level Protein Expression . . . . . . .
`3.2.1 Effect of Inducer Concentration and Time of Induction . . . . . .
`3.2.2 Maintenance of Specific Cellular Protein Yield . . . . . . . . . . .
`3.2.3 Metabolic Burden due to Gene Expression . . . . . . . . . . . . .
`3.2.4 Effect of Oxygen on Gene Expression . . . . . . . . . . . . . . . .
`3.2.5 Amino Acid Mis-Incorporation During Protein Expression . . . .
`3.3
`Development of High Cell Density Fermentation Process for
`Ovine Growth Hormone . . . . . . . . . . . . . . . . . . . . . . .
`3.3.1 Expression of Ovine Growth Hormone . . . . . . . . . . . . . . .
`3.3.2 Effect of Acetate on Cell Growth and r-oGH Expression . . . . . .
`3.3.3 Kinetics of Inclusion Body Production During Batch Fermentation
`3.3.4 Effect of Yeast Extract During High Cell Density Fermentation . . .
`3.3.5 High Productive Fermentation Process for Ovine
`Growth Hormone . . . . . . . . . . . . . . . . . . . . . . . . . . .
`
`4
`
`High Throughput Purification . . . . . . . . . . . . . . . . . . . .
`
`Recombinant Protein as Inclusion Bodies . . . . . . . . . . . . . .
`4.1
`Inclusion Body Formation, Isolation and Characterization . . . .
`4.1.1
`Solubilization of Inclusion Body Proteins . . . . . . . . . . . . . .
`4.1.2
`4.1.3 Renaturation of Solubilized Recombinant Proteins . . . . . . . . .
`4.1.4
`Improved Methods of Protein Refolding from Inclusion Bodies . .
`
`5
`
`Novel Method of Protein Solubilization from Inclusion Body
`of E. coli . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`Solubilization and Refolding of Ovine Growth Hormone (oGH) . .
`5.1
`Isolation and Purification of r-oGH Inclusion Body from E. coli . .
`5.1.1
`Solubilization of r-oGH from Inclusion Body . . . . . . . . . . . .
`5.1.2
`5.1.3 Refolding and Characterization of Recombinant oGH . . . . . . .
`5.2
`Solubilization and Refolding of Human Growth Hormone (hGH)
`5.2.1
`Solubilization of r-hGH from Inclusion Body . . . . . . . . . . . .
`5.2.2 Effect of b-Mercaptoethanol . . . . . . . . . . . . . . . . . . . . .
`5.2.3
`Purification and Refolding of Recombinant hGH . . . . . . . . . .
`5.3
`Solubilization and Refolding of Bonnet Monkey Zona Pellucida
`Glycoprotein C (bmZPC) . . . . . . . . . . . . . . . . . . . . . . .
`Purification and Solubilization of Inclusion Body . . . . . . . . .
`5.3.1
`5.3.2 Refolding, Purification and Characterization of r-bmZPC . . . . .
`5.4
`Ideal Method for Solubilization and Refolding
`of Inclusion Body Protein . . . . . . . . . . . . . . . . . . . . . .
`
`51
`
`52
`52
`53
`55
`56
`56
`57
`58
`58
`59
`60
`61
`
`61
`62
`63
`64
`66
`
`67
`
`70
`
`71
`72
`73
`74
`76
`
`78
`78
`79
`79
`80
`81
`81
`82
`83
`
`84
`85
`86
`
`87
`
`Page 2
`
`

`
`Bioprocessing of Therapeutic Proteins from the Inclusion Bodies of Escherichia coli
`
`6
`
`7
`
`Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`
`References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`
`45
`
`88
`
`89
`
`Symbols and Abbreviations
`
`adenosine triphosphate
`ATP
`ATR-FTIR attenuated total reflectance Fourier transform infrared
`bGH
`bovine growth hormone
`BFGF
`basic fibroblast growth factor
`bmZPC
`bonnet monkey zona pellucida glycoprotein C
`bgal
`beta-galactosidase
`CD
`circular dichroism
`CDNA
`complementary deoxyribonucleic acid
`CER
`carbon dioxide evolution rate
`C/N
`carbon to nitrogen ratio
`DEAE
`diethylaminoethyl
`DO
`dissolved oxygen
`DTT
`dithiotheritol
`DMSO
`dimethyl sulfoxide
`EDTA
`ethylenediaminetetraacetic acid
`GSH
`glutathione reduced
`GSSH
`glutathione oxidized
`hGH
`human growth hormone
`IFN-g
`interferon-g
`IGF-1
`insulin like growth factor-1
`IL-1b
`interleukin-1b
`IL-2
`interleukin-2
`isopropyl thio b-D galactopyranoside
`IPTG
`kDa
`kilodalton
`LB
`luria bertani
`M
`mass
`nicotinamide adenine dinucleotide dihydrogen
`NADH2
`Ni-NTA
`nickel-nitrilotriacetic acid
`OD
`optical density at 600 nm
`oGH
`ovine growth hormone
`PAGE
`polyacrylamide gel electrophoresis
`PDI
`protein disulfide isomerase
`PEG
`polyethylene glycol
`PHB
`polyhydroxybutyrate
`PMSF
`phenylmethylsulfonyl fluoride
`oxygen partial pressure
`PO2
`PPI
`prolyl-peptidyl isomerase
`r-bmZPC recombinant bonnet monkey zona pellucida C
`rhaBAD
`rahmnose BAD
`r-hGH
`recombinant human growth hormone
`
`Page 3
`
`

`
`46
`
`A.K. Panda
`
`radioimmunoassay
`RIA
`ribonucleic acid
`RNA
`messenger ribonucleic acid
`mRNA
`ribosomal ribonucleic acid
`rRNA
`recombinant ovine growth hormone
`r-oGH
`sodium dodecyl sulfate
`SDS
`S-200 HR Sephacryl 200 high resolution
`TCA
`tricarboxylic acid
`tRNA
`transfer ribonucleic acid
`V
`volume
`ZP
`zona pellucida
`ZPA
`zona pellucida A
`ZPB
`zona pellucida B
`ZPC
`zona pellucida C
`m
`specific growth rate
`
`1I
`
`ntroduction
`
`The ultimate goal of recombinant fermentation research is the cost effective pro-
`duction of desired protein by maximizing the volumetric productivity, i.e., to ob-
`tain the highest amount of protein in a given volume in the least amount of time.
`Such bioprocessing for recombinant protein using genetically modified organ-
`isms requires a stable high-yielding recombinant culture, a high productive
`fermentation process and cost effective recovery and purification procedures.
`Escherichia coli species have been most widely used as host for the expression of
`recombinant proteins [1, 2]. Advantages of using E. coli as expression system is
`the enormous amount of data available on its cell biology, fermentation process
`development and its ability to produce large quantities of recombinant proteins
`in an inexpensive way. The successful large-scale cost-effective production of in-
`sulin by Eli Lilly (USA) and bovine growth hormone by Monsanto Corporation
`(USA) attest to the versatility and economic potential of E. coli-based therapeu-
`tic protein production. Although E. coli cannot be used to produce complex gly-
`coproteins or proteins having multiple disulfide bonds, in past 20 years recombi-
`nant DNA technologies have enabled us to produces huge quantities of thera-
`peutic proteins that might otherwise have been difficult [3, 4].
`Recombinant protein expression using E. coli as host is frequently associated
`with the formation of intracellular aggregates as an inclusion body [5]. The vol-
`umetric yield of the protein is thus is a function of both unit cell concentration
`and specific cellular protein yield. Optimization of high cell density fed-batch
`fermentation processes is thus one of the key steps for enhancing the volumet-
`ric yield of recombinant proteins [6, 7]. High level expression of protein in the
`form of an inclusion body facilitates the isolation of the protein of interest from
`the cytoplasm at the cost of its native structure. Renaturation of recombinant
`proteins from inclusion bodies into the bioactive form is cumbersome, results
`in low recovery of the final product and also accounts for the major cost in over-
`
`Page 4
`
`

`
`Bioprocessing of Therapeutic Proteins from the Inclusion Bodies of Escherichia coli
`
`47
`
`all production of recombinant proteins [5, 8]. However, in the cases where a
`simple high-yielding protein refolding process is developed for the aggregated
`recombinant protein, high-level expression of protein as inclusion body pro-
`vides a straightforward strategy for the cost-effective production of therapeutic
`protein. Thus, in spite of the problems associated with the inclusion body in
`E. coli, they have been extensively used for the commercial production of thera-
`peutic protein. High cell density fermentation and improved refolding of the in-
`clusion body proteins are thus the two major bioprocess engineering consider-
`ations for enhancing the overall yield of recombinant proteins from E. coli.
`The objective of the present review is to emphasize the importance of high
`productive fermentation as well as high throughput purification of bioactive
`therapeutic protein from the inclusion bodies of E. coli. Understanding of basic
`biological aspect of the expression system at the molecular level and translating
`this information at process level is imperative for efficient and cost-effective
`production of therapeutic compounds. Parameters that influence the high cell
`density fed-batch aerobic growth of E. coli while maintaining a stable plasmid
`of interest have been analyzed. Novel ways of fed-batch fermentation process
`considering most of these factors have been discussed in detail to maximize the
`volumetric yield of recombinant ovine growth hormone expressed as inclusion
`body in E. coli. Solubilization and refolding of inclusion body protein to the
`bioactive conformation severely limits the overall efficiency of the therapeutic
`protein production from E. coli. Solubilization of the inclusion body protein
`without disturbing the existing native-like secondary structure while using a
`low concentration of a chaotropic salt, its refolding and purification into bioac-
`tive forms have been described. Finally the novelty of high cell density fermen-
`tation processes and improved refolding of inclusion body proteins have been
`applied to a few other proteins expressed in E. coli and process development
`strategies have been discussed. Apart from reviewing the recent trends in bio-
`processing of recombinant protein from E. coli, the review discusses the fer-
`mentation and inclusion body protein refolding process developed at the Na-
`tional Institute of Immunology, New Delhi.
`
`2P
`
`arameters Influencing Productivity of Therapeutic Protein from E. coli
`
`Numerous genetic and environmental factors influence the expression of
`cloned gene product in recombinant E. coli which is most frequently used
`prokaryotic expression system for the production of heterologous proteins [9].
`At the molecular level, strength of transcriptional promoters, plasmid stability,
`copy number, mRNA stability, translational efficiency, localization, status and
`the stability of the expressed foreign protein in the host influences the expres-
`sion levels. These factors influence the metabolic state of the host during gene
`expression, which in turn, can be controlled and manipulated during fermenta-
`tion to maximize the yield of the expressed protein [10]. The expressed protein,
`depending on its localization, can be purified and recovered in the bioactive
`form. It is interesting to note that even though E. coli does not provide an oxi-
`dizing environment for disulfide bond formation leading to the aggregation
`
`Page 5
`
`

`
`48
`
`A.K. Panda
`
`during expression, such aggregated proteins have been successfully refolded in
`vitro to achieve the bioactive conformation. The successful production of
`insulin [11], bovine growth hormone [12] and tissue plasminogen activator [13]
`from inclusion bodies of E. coli indicates that complex proteins can be
`expressed and purified using an E. coli-based expression system.
`With recent advances in gene cloning, development of better cell growth
`process and improvement in refolding yield of the inclusion body protein, the
`efficiency of protein production using E. coli has increased rapidly. It has been
`realized that most of the factors which influence the efficiency of the overall
`protein production process act in a very complex and interactive way at differ-
`ent stages. Putting the gene in front of the promoter is no longer considered as
`the end of a successful recombinant expression system. In fact, it is realized of
`late that the beginning of the problem for successful production of recombinant
`protein starts only after a stable recombinant construct is ready for expression
`at the shaker flask culture level. Innumerable overlapping factors need to be
`taken into consideration for optimization of a recombinant fermentation
`process. These factors can be divided primarily into two broad groups: the first
`cell and molecular biology considerations and the second process engineering
`considerations. Cell and molecular biology considerations deal mainly with the
`level of expression, destination, location and state of the protein produced using
`E. coli as expression system. Factors that influence these things in an expression
`system are host, vector/promoter system and the origin and the nature of the
`protein of interest. By contrast, process engineering considerations deal with
`the large-scale culture of the recombinant organism and the recovery of the ex-
`pressed protein. It aims at high volumetric yield and high throughput recovery
`of the expressed protein in bioactive form. The interactions of these broad
`groups of factors also need careful analysis to optimize the level of protein pro-
`duction using recombinant E. coli.
`
`2.1
`Cell and Molecular Biology Considerations
`
`Most of the cell biology parameters that influence the productivity of the re-
`combinant protein from E. coli include the host organism: expression vector
`and promoter system, gene dosage or plasmid copy number, stability of plas-
`mid, promoter strength, induction strategy, mRNA stability, ribosomal popula-
`tion, tRNA concentration, codon bias, amino acid concentration and finally the
`localization of the expressed protein [2, 9]. Extensive reviews highlighting the
`influence of above factors have been published from time to time [9, 14, 15].
`What is more important is to consider the above cellular factors and see their
`implications during the large-scale processing of recombinant proteins, partic-
`ularly during high cell density fermentation and high throughput purification
`of the protein. In spite of extensive knowledge of the genetics and molecular
`biology of E. coli, it is sometimes difficult to express a gene efficiently in this
`organism. This may be due to unique structural features of the gene sequence,
`initiation of translation, stability and translational efficiency of mRNA, major
`differences in codon usage between the foreign gene and native E. coli, toxicity
`
`Page 6
`
`

`
`Bioprocessing of Therapeutic Proteins from the Inclusion Bodies of Escherichia coli
`
`49
`
`of the protein and degradation of expressed protein by host cell proteases.Vari-
`eties of promoters are now available for the efficient expression of a therapeutic
`protein in E. coli [16, 17]. A chemical inducer like IPTG or heat induction pro-
`vides the best available promoter for gene expression having high strength and
`regulatory capability and have been mostly used for the production of therapeu-
`tic proteins. The recently developed arabinose promoter is probably the closest
`ideal promoter in terms of strength, regulation and controlled expression [18,
`19]. The rhamnose inducible promoter (rhaBAD) has also been used for expres-
`sion of L-N carbamoylase using E. coli as the host [20]. The use of a dual pro-
`moter has also been reported to have helped in high level expression of recom-
`binant protein in E. coli [21]. The most important cellular parameter for con-
`trolling and optimizing gene expression is the translation process [22, 23].
`Stability of mRNA, secondary structure of the mRNA [24, 25] and ribosomal
`population affect the overall yield of the recombinant protein [9, 26]. It has been
`widely documented that the translation capacity of the cell remains the most
`crucial factor for the efficient expression of a recombinant protein [9]. Efficient
`design of the expression vector considering the end effects of translational ini-
`tiator, enhancers, terminator and m-RNA stabilizer need careful assessment for
`high-level expression of the foreign protein using E. coli as host.
`The major draw back of the E. coli as expression system is the inability to do
`many post-translational modifications found in eukaryotic protein, limited
`ability to facilitate disulfide bond formation due to the reducing nature of cyto-
`plasm and lack of secretion mechanism for the efficient release of the expressed
`protein into the culture medium. The decision to target the expressed protein
`into the cytoplasm, periplasm or the culture medium depends on balancing the
`advantages and disadvantages offered [15]. Real secretion of protein in to the
`extracellular medium is rare and periplasmic expression most of the time re-
`sults in low level of recombinant protein expression. Exceptions are the secre-
`tion of leptin [27] and the high-level accumulation of IGF using a dual pro-
`moter system [21]. High-level expression of protein into cytoplasm leads to ac-
`cumulation of denatured protein in the form of inclusion bodies [28, 29].
`However, with recent understanding of the structure function of the inclusion
`body protein, recoveries of bioactive protein with reasonable yield have been
`achieved for many proteins [30]. The high initial level of expression compen-
`sates loss during recovery of protein from inclusion bodies. In fact, most of the
`commercially available proteins from E. coli are expressed as inclusion bodies
`and then suitably refolded into the bioactive form. Thus for high level expres-
`sion of recombinant protein which does not require post-translational modifi-
`cation for bioactivity, expression in the form of inclusion body and its subse-
`quent purification and refolding becomes the most cost effective way of thera-
`peutic protein production.
`Another molecular biology consideration, which needs attention, is the im-
`provement of host cell metabolism for improved expression. This includes co-
`expression of chaperone [9, 14], use of tag or fusion protein for expression in
`soluble form or efficient purification [15, 31]. Using a dual promoter system and
`by delineating the cell growth phase from that of expression phase, very high
`levels of recombinant IGF1 have been successfully expressed in E. coli [21].
`
`Page 7
`
`

`
`50
`
`A.K. Panda
`
`Metabolic engineering of host cells for low acetate secretion [32, 33] and incor-
`poration of gene for improving oxygenation of the large-scale culture [34] have
`also been used for maximizing the level of foreign gene expression in E. coli.
`During large-scale processing of the recombinant E. coli the above cellular
`parameters influence the metabolic status of the cells which, in turn, necessi-
`tates special attention both during cell growth and protein recovery to optimize
`the overall protein production process. More important parameters like plas-
`mid stability, inducer concentration and time of induction, kinetics of growth
`product relations, harvest time and efficient protein recovery processes need
`special attention to scale-up the shaker flask result successfully to high cell den-
`sity fed-batch fermentation. Interrelations and effects of many such factors in-
`fluence the overall yield of the finished products and thus need to be analyzed
`in the context of the overall process performance rather than the single unit op-
`eration stages.
`
`2.2
`Process Engineering Considerations
`
`In most of the cases, E. coli expressed proteins are associated with intracellular
`accumulation as inclusion bodies. The volumetric yield of a recombinant pro-
`tein will depend on both the biomass concentration as well as specific cellular
`protein yield [6, 7]. Thus, after successful construction of the recombinant cul-
`ture for optimal expression of the protein, high productive fermentation and
`efficient protein recovery process development are the two major constraints
`for efficient cost effective production of therapeutic protein. In recent years,
`tremendous progress has been achieved both in terms of development of high
`productive fermentation and high throughput protein purification from inclu-
`sion bodies. These two factors are discussed in detail in this review with an aim
`to increase the volumetric production of therapeutic protein using E. coli.
`High volumetric yield of the protein can be achieved by increasing the cell
`concentration in the reactor volume. High cell density fermentations using fed-
`batch culture techniques are routinely used to maximize the yield of recombi-
`nant proteins in E. coli [35, 36]. Most of the high cell density fermentation
`processes have been developed using recombinant E. coli. The highest E. coli cell
`concentration around 200 g dry cell weight per liter of fermentation broth has
`been achieved for polyhydroxybutyrate expression [37]. Recombinant E. coli ex-
`pressing polyhydroxybutyrate in high cell density fermentation produces as
`high as 2.8 g L–1 d–1 of product, emphasizing the enormous capacity of the
`E. coli system to produce heterologous protein. It is expected that with proper
`bioprocessing the yield of product can be matched to the chemical synthesis
`process of polymer production [38].Assuming a 20% expression level of any re-
`combinant therapeutic protein at such a high cell concentration, it is expected
`in the near future that an expression level of 10–20 g L–1 of the recombinant
`protein can be easily produced through high cell density fed-batch fermenta-
`tion of recombinant E. coli.
`Efficient recovery of bioactive protein from the inclusion bodies is the major
`constraint for the successful production of therapeutic protein from E. coli. In
`
`Page 8
`
`

`
`Bioprocessing of Therapeutic Proteins from the Inclusion Bodies of Escherichia coli
`
`51
`
`general, the yield of bioactive recombinant protein from inclusion bodies of
`E. coli is around 10–20% of the total expressed protein. The solubilization of
`protein from the inclusion body by high concentrations of chaotropic reagents
`results in the loss of its secondary structure, leading to the random coil forma-
`tion of the protein structure [39, 40]. Loss of secondary structure during solubi-
`lization and the interaction among the denatured protein molecules leading to
`their aggregation are considered to be main reasons for the poor recovery of
`bioactive proteins from the inclusion bodies [41]. Refolding at low protein con-
`centration and use of urea for solubilization have been reported to be the most
`costly factors in the production of recombinant insulin from E. coli [8].
`Apart from the above-mentioned problems mainly encountered during the
`bioprocessing of therapeutic proteins, other factors that are inherent to the fer-
`mentation process need careful consideration. Recovery to fermentation cost
`for bioactive protein from E. coli is around 3 to 5, whereas for antibiotics like
`penicillin it is around 1, indicating the importance of downstream operation for
`cost effective production of therapeutic proteins [42]. Like many fermentation
`processes, therapeutic protein production involves lots of water for different
`processing steps. More importantly, protein folding and polishing need high
`quality water, which has to be treated specially before discharge into the envi-
`ronment. Huge water requirements during therapeutic protein production will
`adversely affect the process efficiency and power consumption and increase
`many-fold the overall cost of the product [8, 13]. Thus it is also essential to de-
`velop low water requiring, environmental friendly technology for the produc-
`tion of therapeutic proteins, particularly during refolding of inclusion body
`proteins.
`
`igh Cell Density Fermentation
`
`3H
`
`Essentially, high cell density growth involves the culturing of a recombinant
`organism to a very high cell concentration (>20 g L–1 dry cell weight) by em-
`ploying fed-batch fermentation methods [6, 36, 43, 44]. Operation of fed-batch
`fermentation helps in increasing the unit cell concentration in the reactor and
`thus improves the volumetric yield of the protein. As recombinant protein ex-
`pression using E. coli results mostly in intracellular accumulation, the volumet-
`ric yield depends on both the final cell concentration as well as the specific cel-
`lular protein yield [6]. In high cell density fermentation, maximizing cell con-
`centration helps in increasing the volumetric productivity of recombinant
`proteins. High cell density culture of E. coli, apart from improving the volumet-
`ric productivity, also provide advantages such as reduced culture volume, en-
`hanced downstream processing, reduced waste water, lower production cost
`and reduced investment on equipment [7]. Apart from this, it is essential that
`cell growth be achieved in an optimal time period to improve the overall pro-
`ductivity of the recombinant protein. Toxicity of acetate, slow growth rate, in-
`stability of plasmid, depletion of amino acid pools to sustain a high rate of pro-
`tein synthesis affect the specific cellular yield of recombinant protein at high
`cell concentrations. It is expected that an analysis of all these parameters during
`
`Page 9
`
`

`
`52
`
`A.K. Panda
`
`high cell density fed-batch growth of E. coli will lead to high volumetric pro-
`ductivity of the desired protein. Such approaches have been used extensively to
`increase the volumetric yield of recombinant products in prokaryotic systems,
`particularly those employing Escherichia coli as host [36].
`
`3.1
`Parameters Affecting High Cell Growth of E. coli
`
`Innumerable overlapping factors must be taken into account while optimizing
`high cell density E. coli growth for protein expression [35, 36]. Composition of
`medium, physical parameters during growth and operating conditions are the
`most important factors that influence the cell growth. Limitation and/or inhibi-
`tion of substrates, limited capacity of oxygen supply, formation of metabolic
`byproducts and instability of plasmid during long hours of cultivation are the
`major problems encountered during high cell density growth of E. coli. Most of
`the time these depend on host strain, vector and strength of promoter. Dense
`culture requires large amounts of O2 to support good growth and thus necessi-
`tates an unconventional aeration strategy to maintain the dissolved oxygen con-
`centration at a suitable level throughout the growth period. In most cases with
`E. coli used as a host for recombinant protein, the production phases start after
`induction with a suitable inducer. Thus, in principle, the growth phase and the
`production phase can be delineated in the same vessel for a high volumetric
`yield of the recombinant protein. However, in many cases, the operation of the
`reactor during cell growth influences the specific yield of the recombinant pro-
`teins. Thus, while developing fed-batch operation to increase unit cell growth in
`the reactor, it is equally essential to take care of the factors which affect the spe-
`cific yield of the recombinant protein.
`
`3.1.1
`Nutrient Formulation
`
`One of the essential requirements during fed-batch operation is to supply nutri-
`ents to promote cell growth [45, 46]. To limit their toxicity to the growing cells
`nutrients such as glucose, ammonia, salt are fed in approximation of their re-
`quirement. The accumulation of nutrients at high concentration inhibits
`growth and recombinant protein expression [6]. High glucose causes the Crab-
`tree effect, leading to the accumulation of acetate which is inhibitory to cell
`growth [47]. Ammonia inhibits gene expression [48] and the absence of a metal
`salt may hamper the enzymatic activities of many enzymes vital for cell metab-
`olism [45]. In general, most of the media used for high cell growth of E. coli have
`mostly glucose as carbon source, with major salts like phosphate, sodium potas-
`sium, magnesium, ammonia, sulfate, iron, minor trace elements and complex
`nitrogenous materials. High-density growth in general is initiated with a low
`concentration of the most required substrate and the nutrients are added later
`in the growth period [49]. Typically, glucose >50 g L–1, ammonia>3 g L–1,
`P>10 g L–1, Mg>8. g L–1, Fe>1.2 g L–1, Mo>0.8 g L–1 B>44 mg L–1, Cu>4 mg L–1,
`Mn>68 mg L–1, Co>0.5 mg L–1, Zn >38 mg L–1 inhibit E. coli growth [35]. Ideally,
`
`Page 10
`
`

`
`Bioprocessing of Therapeutic Proteins from the Inclusion Bodies of Escherichia coli
`
`53
`
`Table 1. Examples of high cell density growth of E. coli
`
`Host strain
`(E. coli)
`
`B
`HB101
`TG1
`KA197
`X90
`W3110
`TG1
`K12
`BL21
`N48301
`XLi-Blue
`CGSC4401
`K12
`
`Culture method/feeding
`
`Cell density
`(g L–1)
`
`Year
`
`Reference
`
`DO stat
`DO control with cell recycle
`Specific growth rate controlled
`Exponential feeding
`Exponential feeding
`Exponential feeding
`Exponential feeding
`Dialysis culture
`Exponential feeding
`Specific growth rate
`pH-stat
`pH stat
`Dialysis culture
`
`125
`145
`110
`77
`95
`45
`148
`190
`80
`50
`201
`119
`190
`
`1979
`1990
`1991
`1991
`1993
`1994
`1995
`1997
`1997
`1998
`1999
`2000
`2002
`
`[49]
`[51]
`[52]
`[43]
`[44]
`[53]
`[54]
`[55]
`[56]
`[57]
`[37]
`[58]
`[59]
`
`the components should be added to the fermenter at the same rate at which they
`are consumed so as to prevent nutrient accumulation up to toxic levels while
`still promoting good growth. Another factor which needs attention during
`medium formulation is the solubility of many components, particularly while
`making concentrated solutions for fed-batch addition. High concentrations of
`glucose, yeast extract and trace elements need careful composition to avoid pre-
`cipitation. The osmolarity of the medium can affect nutrient yield, cell growth
`and specific recombinant protein yield [6, 50]. Hence it is also essential to take
`care of medium’s osmolarity so that its detrimental effect is minimum during
`high cell density growth. Table 1 summarizes examples of high cell concentra-
`tions of E. coli at a range of 50 to 100 g L–1 achieved using fed-batch fermentation.
`Apart from medium formulation, the operating conditions such as pH, tem-
`perature and, more importantly, O2 supply are very very essential for supp