Bioreactor design for successive culture of anchorage-dependent cells operated in an automated manner.

A novel bioreactor system was designed to perform a series of batchwise cultures of anchorage-dependent cells by means of automated operations of medium change and passage for cell transfer. The experimental data on contamination frequency ensured the biological cleanliness in the bioreactor system, which facilitated the operations in a closed environment, as compared with that in flask culture system with manual handlings. In addition, the tools for growth prediction (based on growth kinetics) and real-time growth monitoring by measurement of medium components (based on small-volume analyzing machinery) were installed into the bioreactor system to schedule the operations of medium change and passage and to confirm that culture proceeds as scheduled, respectively. The successive culture of anchorage-dependent cells was conducted with the bioreactor running in an automated way. The automated bioreactor gave a successful culture performance with fair accordance to preset scheduling based on the information in the latest subculture, realizing 79- fold cell expansion for 169 h. In addition, the correlation factor between experimental data and scheduled values through the bioreactor performance was 0.998. It was concluded that the proposed bioreactor with the integration of the prediction and monitoring tools could offer a feasible system for the manufacturing process of cultured tissue products.

Correlation of cellular life span with growth parameters observed in successive cultures of human keratinocytes.

In serial cultures of keratinocytes isolated from human neonatal foreskin, the potential for cell attachment and cell division decreased with an increase in the number of culture passages. The specific growth rate of the cells (mu) reduced gradually with elapsed culture time and an appreciable drop in the rate was observed below mu=0.02 h(-1) owing to cellular senescence. Then, a critical point in the serial cultures was determined as the last data point keeping the mu value at more than 0.02 h(-1). In serial cultures of keratinocytes isolated from the breast skin of 22 and 70-year-old individuals, the time profiles of the mu value displayed critical points, as observed in the neonatal foreskin cultures. Plots of the mu value and average cell area against the remaining numbers of population doublings (N(dF)/t-N(d)) were found to give a relationship which overlapped irrespective of the three kinds of keratinocytes, and the propensity for a depression of growth potential [estimated as a ratio of differential value of mu to that of (N(dF)-N(d))] was much higher after the critical point than before. In addition, a data map in terms of the specific growth rate and average cell area, which could be classified into proliferative, senescent and transitional phases, was proposed for the discrimination of cellular senescence in serial cultures for the production of epithelial sheets.

Characterization of cellular motions through direct observation of individual cells at early stage in anchorage-dependent culture.

In the monolayer culture of murine fibroblasts, the characterization of cellular motions was performed by morphological observation using a tool consisting of an optical assembly connected to a computer-aided image analysis system. This tool enabled the estimation of the projected area of the cells with sufficient accuracy, being able to follow the serial behavior of the cells on the culture surface. Trypsinization was chosen as an external factor for altering the state of the cells, and the rate of cell spreading (r(S)) and time of first cell division (t1) were evaluated by subjecting the cells to trypsin treatment for t(T)=3 and 15 min. During culture of the cells treated for t(T)=3 min, the r(S) values of individual cells exhibited a broad variation, ranging from 41 to 321 microm2/h, and the average r(S) was 165+/-78 microm2/h , which was 1.5 times larger than that of the cells treated for t(T)=15 min. On the contrary, the average t1 of the cells treated for t(T)=3 min was 9.5 h which was 60% reduced as compared with that of the cells treated for t (T)=15 min. The prolonged time of trypsin treatment was considered to induce the decrease in r(S) and delay of the first cell division due to the requirement for the recovery from cell surface damage caused by trypsinization. The logarithmic plots of r(S) and t1 were found to have an inverse relation regardless of the state of the individual cells.