Development of a serum-free medium for dihydrofolate reductase-deficient Chinese hamster ovary cells (DG44) using a statistical design: beneficial effect of weaning of cells.

To develop serum-free (SF) medium for dihydrofolate reductase-deficient Chinese hamster ovary cells (DG44), a statistical optimization approach based on a Plackett-Burman design was adopted. DG44 cells which were normally maintained in 10 serum medium were gradually weaned to 0.5% serum medium to increase the probability of successful growth in SF medium. A basal medium was prepared by supplementing Dulbecco's modified Eagle's medium and Ham's nutrient mixture F12 with hypoxanthine (10 mg/l) and thymidine (10 mg/l). Twenty-eight different supplements were selected as variables on the basis of their growth-promoting abilities. From statistical analysis, leucine, tryptophan, lysine, proline, histidine, hydrocortisone, ethanolamine, and phosphatidylcholine were identified as important components showing positive effects on cell growth. A new SF medium (SF-DG44) was formulated by supplementing the basal medium with these components. When the weaned cells were inoculated at 1.0 x 10(5) cells/ml, a maximum viable cell concentration of 6.4 x 10(3)) cells/ ml was achieved in SF-DG44 medium. In contrast, when the unweaned cells were used, a concentration of only 4.1 x 10(5) cells/ml was reached under the same culture conditions, indicating that weaning of cells improves cell growth in SF medium. In summary, we found that development of a novel SF medium for DG44 cells was facilitated using a Plackett-Burman design technique and weaning of cells.