Fan Sozzi-Guo and Ursula Graf-Hausner look at the gentle transportation of cell suspensions with low pulsation using various pump systems.
Devices for mild/gentle transportation of cell suspensions are required in various operations in a bioprocess: transfer of inoculum, cell retention or recirculation, for preparative or analytical purposes, even transfer to downstream.
Inappropriate pumps are often used for such cell transport which results in damage to cells, be this immediately (necrosis) or indirectly (apoptosis). The main goal of this work was to find a consistent method to test various pump systems with respect to their impact on animal cells.
Hybridoma cells were grown in a spinner flask (500ml) and with a scaled up version of the system using different bioreactors (5, 13 and 42 l).
During the exponential growth phase the cells were pumped (1/2 reactor volume h-1) through a loop for variable periods of time.
A set of different analytical methods were employed in order to qualitatively and quantitatively determine inflicted cell damage directly after the stress situation: Trypan Blue staining, DAPI staining, DNA laddering, LDH-activity, lactate/glucose quotient analysis.
Four types of pump were tested during this work: peristaltic pumps, a aKobio' membrane pump, centrifugal and gear pumps.
Cell density and vitality
In an experiment using spinner flasks, nine different pumps were employed to continuously transport cell suspensions in an external loop; whereby their influence on cell growth and cell vitality was compared.
Fig. 1 shows the length of time the cells remained in the system before a recognisable cell death was observed.
Fig. 2 shows the decline in cell vitality. The influence of the different pump types varies significantly.
For instance with a gear pump there is a very rapid decline in vitality as compared to with a peristaltic pump with specially formed rollers (compare the average steepness of the values in Fig. 2).
Peristaltic pumps with specially formed rollers are the gentlest in terms of cell vitality and cell number. The MCP Process Pro280 peristaltic pump with two convex non-spring-loaded rollers could be employed to transport the cells for more than 50 hours before a recognisable cell death was observed. After 56 hours of continuous cell transport the vitality declines to 87 per cent.
The use of various analytical techniques (determination of apoptotic rate and LDH-activity) permitted a differentiation between pump induced necrotic or apoptotic cell death.
For instance the peristaltic pump with six flat rollers acted like a ball mill and induced necrotic cell death.
Fig. 3 shows the influence of a gear pump and a peristaltic pump (LDH-activity in the cell supernatant was measured).
The determination of glucose and lactose and the comparison of metabolic status based on the lactose/glucose quotient demonstrated the metabolic changes induced by pump stress. Fig. 4 shows the significant metabolic changes observed in cells which were continuously transported.
The influence of the method of cell transport on cellular behaviour in the bioreactor.
The variation in influence on the cells due to the various pump systems was verified at reactor scale. Transport of cells using peristaltic pumps with specially formed rollers proved to be the gentlest form of transport, both at small scale, in spinner flasks (see Fig. 1 and 2) and at reactor scale.
With cell transport using the peristaltic pump head Pro280 (Ismatec) with 2 convex rollers the vitality declined after 30 hours.
Reference cells
By comparison, the reference cells which were not transported had a vitality of more than 90 per cent after 90 hours of incubation (see Fig. 5).
The use of gear pumps resulted in an increase of apoptotic rate after only 10 hours; whilst a variation in vitality as compared to the reference is first observed with the peristaltic pump after 30 hours (see Fig. 6).
The difference in LDH-activity between the reference and the peristaltic pump was very small. The pump was only detrimental to cell growth.
Cell transport via gear pump showed a significant increase in LDH-activity after 28 hours. This implies that necrotic cell damage is significantly higher when gear pumps were used (see Fig. 7).
Under the established conditions, the investigation clearly shows that the rates of cell death vary strongly depending on the pumps used.
The dead cells could be differentiated into two groups: necrotic and apoptotic.
The centrifugal and membrane pumps inflicted most damage on the cells (apoptotic cell death predominating). Strong shear forces are generally attributed to gear pumps.
Instant necrosis
We found a significant amount of instant necrosis, which provides good evidence to harden the assumption. Gear pumps showed slightly better results when compared to centrifugal and membrane pumps.
The best results concerning cell growth and viability, necrosis and apoptosis were shown by peristaltic pumps with specially shaped rollers. u
Enquiry No 92
Fan Sozzi-Guo and Ursula Graf-Hausner are with Department of Chemistry, Zürich University of Applied Sciences Winterthur, Switzerland. The described study was supported by Novartis Pharma AG, Bioprocess & Technology Development.
ISMATEC SA, which provided the majority of the pumps tested, is based in Glattbrugg, Switzerland. www.ismatec.com
