Fed-batch Process Development

A generic fed-batch process has been developed for the production of monoclonal antibodies in PER.C6 cells. The process typically results in a 3- to 4-fold increase in antibody yields compared to the batch process, with yields of 1-3.5 g L after 16-18 days. The feed strategy is based on the metabolic requirements of the PER.C6 cell line. Metabolic characterization of several antibody-producing cell lines identified nutrients and medium components which 

The feed strategy involves the addition of these nutrients based on cell-specific requirements in order to supply the nutrients only as required by the culture and to limit overflow metaboHsm or the build-up of nutrients or metaboHtes that may result in reduced process performance (antibody yields) and product quality [18-25]. 

A typical feed strategy involves the addition of four to six bolus feed additions at regular intervals during a 16- to 18-day process. Similar growth and production profiles are observed for aU antibody-pro- 

Metabolic data from the fed-batch development was used to develop a supplemented batch process involving the addition of up to three of the feeds added in the fed-batch process, to the culture medium prior to inoculation of the cells. Final antibody yields are not as high as for the fed-batch process, typically an increase of 2-fold over batch yields compared to 3- to 4-fold increases for the fed-batch. However, the process offers a relatively simple way of obtaining increased antibody yields. Fig. 3.16 shows a supplemented batch culture for clone C where the final antibody concentration reached 1.3 g L .  

---

A simple mathematical model is used for quantitative description of the process and consists of a set of equations relating inputs, outputs, and key parameters of the system. The model for an alcoholic fermentation fed-batch process developed by Mayer (10) and adapted with the Ghose and Tyagi (11) linear inhibition term by the product was used as the starting point for the development of a model-based substrate sensor with product (ethanol) and biomass on-line measurements. 

Crowley et al. performed some interesting work with Pichia pastoris in a fed-batch process.19 The complex mixture was measured using a multibounce attenuated total reflectance (HATR) cell. The authors developed models for glycerol, methanol, and the product, a heterologous protein. The results are reported somewhat differently from normal chemometric results. The authors used root-mean square error (RMSE) for the product as a performance index and measured a percent difference for the methanol and glycerol. 

One of the most important developments in the history of large scale fermentations is the fed-batch process. Again, this derives from the work of Marvin Johnson at the University of Wisconsin during development of the penicillin fermentation over 50 years ago. Soltero and Johnson wrote Glucose, intermittently fed to fermentations, has given penicillin yields on synthetic medium equal to, or even better than, those obtained with lactose. Penicillin yields of twice those of lactose controls have been obtained when glucose or sucrose is continuously added to the fermentations . 

The early mammalian cell processes were typically batch. As culture media and processes have developed over the years, advances in feeding strategies for fed-batch processes have increased productivity to several grams per liter for both GS-NSO [4] and GS-CHO cell lines [5]. 

For new processes, even these companies aim to develop efficient fed-batch processes, and if MAb concentrations of >0.7 could be achieved in fed-batch systems, they would no longer use the perfusion cell-culture approach. 

Continuous processes seem to be more economical than batch and fed-batch processes. Namely, continuous processes offer some advantages, such as reduction in sterilization and re-inoculation time, and superior productivity (44). Recently a non-naturally occurring microorganism with one or more gene disruptions has been developed for the production of 1,4-butanediol (38). 

Fermentation and separation processes were developed for the pilot-scale production of PHB using an A. latus strain as well as its further processing into consumer products. Up to 1 ton/week of PHB was produced in Austria in a 15,000-L bioreactor (84). This technology is currently owned by the German company Biomer. The Biomer fermentation process is relatively simple with multistaging from the petri dish to a shake flask to a small fermentor which is then used to inoculate the production reactor (Fig. 5a). If sufficient carbon source (molasses) is added during the fed-batch process, up to 60 g/L of PHB may be obtained. The... 

A fed batch process using suspended Rhodotorula rubra cells in an aqueous medium at pH 10.6 was developed [263]. Since PAL from R. rubra is oxygen-sensitive, the biotransformation is performed under anaerobic, static condition in fed batch mode with periodic addition of concentrated ammonium dnnamate solution and pH maintenance by CO addition [263]. The fermentative process was improved by starting the process direcdy from glucose as a substrate and L-phenylalanine is now manufactured by fermentation on a scale of 8-10,000 tons per year [264]. 

Literature-cited pubhcations have focused on the development of fed-batch processes for SCO production using various feeding strategies. The application of fed-batch processes with feedback control has not been addressed extensively though. 

In an alternate process for the preparation of C-13 taxol side chain, the stereoselective enzymatic hydrolysis of racemic ci5-3-(acetyloxy)-4-phenyl-2-azetidinone 42 to the corresponding (S)-(—)-alcohol 43 has been demonstrated [53,54]. Lipase PS-30 from Pseudomonas cepacia (Amano International Enzyme Co.) and BMS lipase (extracellular lipase derived from the fermentation of Pseudomonas sp. SC 13856) catalyzed hydrolysis of the undesired enantiomer of racemic 42, producing S-(—)-alcohol 43 and the desired i -(+)-acetate 44 (Fig. 12). Reaction yields of more than 48% (theoretical maximum yield is 50%) and e.e. of more than 99.5% were obtained for the desired R-(+)-acetate. For a very efficient enzyme source (BMS lipase), a lipase fermentation using Pseudomonas sp. SG 13865 was developed. In a fed-batch process using soybean oil, the fermentation resulted in 1500 U/ml of extracellular lipase activity. Crude BMS lipase (1.7 kg containing 140,000 U/g) was recovered from the filtrate by ethanol precipitation. BMS lipase and lipase PS-30 were immobilized on Accurel polypropylene (PP). These immobilized lipases were reused (10 cycles) without loss of enzyme activity, productivity, or the e.e. of the product in the resolution process. The enzymatic process for the resolution of racemic acetate 42 was scaled up to 150 L at 10 g/L substrate concentration using inunobilized BMS lipase and lipase PS-30, respectively. From each reaction batch, 3-(R)-acetate 44 was isolated in 45 M% yield (theoretical maximum yield is 50%) and 99.5% e.e. 3-(R)-acetate 44 was chemically converted to 3-(R)-alcohol 45. The C-13 taxol side chain (41a or 45) produced... 

The most limiting factor for enzymatic PAC production is the inactivation of PDC by the toxic substrate benzaldehyde. The rate of PDC deactivation follows a first order dependency on benzaldehyde concentration and reaction time [8]. Various strategies have been developed to minimize PDC exposure to benzaldehyde including fed-batch operation, immobilization of PDC for continuous operation and more recently an enzymatic aqueous/octanol two-phase process [5,9,10] in which benzaldehyde is continuously fed from the octanol to the enzyme in the aqueous phase. The present study aims at optimal feeding of benzaldehyde in an aqueous batch system. 

The process developed on lab scale was transferred to pilot plant scale. The process is a fed-batch fermentation with a growth-phase (24h) on complex sugars and a production phase (48h). During production phase a linear feed is added containing various carbohydrates. 

Ikeda, S., Nikaido, K., Araki, F K. et al. (2004) Production of recombinant human lysosomal acid lipase in Schizosaccharomyces pombe. development of a fed-batch fermentation and purification process. Journal of... 

In addition to the enzyme s amino acid sequence, other parameters can affect the outcome of a biocatalytic process. For instance, a similar outcome in the aforementioned DERA-catalyzed statin synthesis was achieved by process improvements [21]. Using a thermostable variant of DERA (thermostability generally correlates well with tolerance to high concentrations of organic reagents or cosolvents), and fed-batch conditions, an efficient process that overcame sensitivity to high concentrations of chloroacetaldehyde was developed. 

---