January 1, 2009 (Vol. 29, No. 1)
Study Shows Process Does Not Improve Cell Culture Performance
Hydrolysates (peptones) are widely used in biopharmaceutical manufacturing to enhance cellular growth and production. Ultrafiltration is often utilized to remove large molecular weight entities including endotoxin from hydrolysates. While minimization of endotoxin is a high priority in biomanufacturing, the necessity of ultrafiltration of raw materials has not been firmly established.
Some data suggests that ultrafiltration negatively impacts hydrolysate performance in cell culture media by removing small molecular weight entities that act as growth factors in the media. Moreover, the necessity of hydrolysate ultrafiltration in cases where the material is already low in endotoxin is unclear, despite unsubstantiated claims in the industry that ultrafiltered material supports superior cell culture performance.
LucraTone™ hydrolysates are manufactured by Solabia for exclusive distribution by Millipore. Evaluation of multiple hydrolysate lots indicates that manufacture of LucraTone Soy hydrolysates yields consistently low endotoxin levels, as well as low levels of large molecular weight entities, despite the lack of ultrafiltration in the manufacturing process.
The lack of ultrafiltration of this product provided an opportunity to evaluate the impact of ultrafiltration of hydrolysates on mammalian cell culture performance. In this case study, multiple lots of both ultrafiltered and nonultrafiltered LucraTone Soy P hydrolysates were evaluated to determine their impact on mammalian cell culture performance. Results of our study suggest that there is no impact of hydrolysate ultrafiltration on cell culture performance.
Materials and Methods
Ultrafiltration of LucraTone Soy P: Five lots of LucraTone Soy P hydrolysates were prepared in 50 g/L stock solutions using MilliQ sterile water and 0.22 µm filtered. Stock solutions were separated into triplicate aliquots. Two aliquots were ultrafiltered either by normal flow or tangential flow ultrafiltration using a 5 kDa MWCO filter (Millipore). The third unfiltered aliquot served as a negative control. All three aliquots were 0.2 µm filtered, then stored at 2–8ºC prior to cell culture performance analyses. Endotoxin levels were determined before and after ultrafiltration using the LAL (Limulus amebocyte lysate) assay.
CHO Cell Culture: For this case study, a CHO cell line expressing a human IgG was grown in serum-free chemically defined medium. Exponentially growing CHO cells were harvested, counted by a Vi-Cell® counter employing the trypan blue exclusion technique, and seeded at 0.5×105 cells/mL into triplicate 125 mL shake flasks (30 mL/flask). Cells were grown in either basal CD-CHO media, CD-CHO media supplemented with 1g/L non-UF LucraTone Soy P, or CD-CHO media supplemented with 1g/L UF LucraTone Soy P.
In all instances, 8mM glutamine, 10 µg/mL puromcyin, 1X HT solution, and 0.2% (v/v) Pluronic F-68 were added to media used in the study. Protein production levels were quantified using a commercially available ELISA kit (Bethyl Laboratories) to detect human IgG. Cell densities were determined daily starting at study day four using a Vi-Cell cell counter. All experiments were performed twice, with each time point sampled in triplicate to demonstrate reproducibility. Error bars are based on one standard deviation of the mean value for the given time point. Data show representative results obtained for all five hydrolysate lots evaluated.
Non-UF LucraTone Soy P promotes cell growth and protein production of CHO cells: CHO cells were grown in the presence or absence of 1 g/L LucraTone Soy P hydrolysate over a ten-day time course. The presence of Soy P improved both cell growth (Figure 1) and protein production (Figure 2) compared to a media-only control. Similar results were observed with LucraTone Soy F, another non-UF hydrolysate (Figures 1 and 2).
Ultrafiltration further reduces endotoxin levels in soy hydrolysate: With all five soy P samples used in this study, both normal flow and tangential flow ultrafiltration reduced endotoxin to levels that were undetectable by LAL assay. Note that unfiltered LucraTone Soy P hydrolysates yield consistently low endotoxin levels even prior to ultrafiltration.
Ultrafiltered and nonultrafiltered LucraTone Soy P promote similar increases in CHO cell growth and protein production: CHO cells were grown in the presence or absence of 1 g/L LucraTone Soy P hydrolysate over ten days. The LucraTone Soy P was ultrafiltered by normal flow (NF) or tangential flow filtration (TFF) methods, and cell growth in the presence of these filtered hydrolysates were compared to a nonfiltered control. Neither NF nor TFF altered the improved cell growth (Figure 3) and protein production (data not shown) in the presence of LucraTone Soy P.
The presence of either LucraTone Soy P or Soy F hydrolysate in CHO cell culture media improved cell densities and protein production by approximately 40–50%, compared to a media-only control. These levels were observed whether or not the hydrolysates had been ultrafiltered. The differences in specific protein production among the different experimental groups was negligible (data not shown), suggesting that the observed increase in protein production in CHO cells was the result of the hydrolysate facilitating higher cell densities rather than increased protein production per cell.
The addition of either UF or non-UF soy hydrolysate to the CD-CHO media increased CHO cell proliferation independent of the method of ultrafiltration (i.e., normal flow versus tangential flow). Both the UF and non-UF conditions performed comparably among the five lots that were tested with respect to cell densities, viabilities, and protein production.
This observation supports not only that the hydrolysates had low lot-to-lot variability, but also that the method of ultrafiltration did not impact hydrolysate performance. In addition, low endotoxin levels were detected in all hydrolysate samples that were used for testing, suggesting that ultrafiltration is not necessary as an endotoxin risk-mitigating activity. It should be noted that only LucraTone hydrolysates were tested, thus these results cannot be extended to hydrolysates produced by other manufacturers without further testing.
In summary, this study suggests that ultrafiltration of LucraTone Soy P hydrolysate has no impact on CHO cell culture growth or performance, nor is it required to deliver consistently low endotoxin levels. Our results provide scientific evidence to contradict the popular assertion that hydrolysate ultrafiltration improves cell culture performance.