G. Russotti et al., PILOT-SCALE HARVEST OF RECOMBINANT YEAST EMPLOYING MICROFILTRATION - A CASE-STUDY, Journal of biotechnology, 42(3), 1995, pp. 235-246
In order to develop a cost-effective recovery process for an intracell
ular product, crossflow microfiltration was studied for the harvest of
a recombinant yeast under severe time constraint. It was required to
process yeast broth in a short period of time to minimize the risk for
product degradation. Preliminary microfiltration studies employing fl
at sheet membranes showed high throughout with initial fluxes on the o
rder of water fluxes (> 1000 LMH, regime I, < 2 min), followed by a ra
pid decay towards a low pseudo-steady state flux (20 LMH, regime II, >
2 min). Exploitation of these high fluxes and control of their eventu
al decline were crucial in establishing a rapid crossflow filtration p
rocess. The effect of several parameters, such as initial cell concent
ration, shear rate, transmembrane pressure, membrane pore size and med
ium composition on filtration performance were investigated to better
understand the flux decline mechanisms. We found that the major contri
butor to flux decay was reversible fouling by the cake formation on th
e membrane surface. Within the operating boundaries of our microfiltra
tion system, large-pore membrane (0.65 mu m) was much more desirable f
or harvesting our yeast (10 mu m size) without cell leakage than small
er pore ones (0.22 mu m and 0.45 mu m). Among adjustable operating par
ameters, feed flow rate (i.e., shear rate) exerted significant impact
on average flux, whereas manipulation of transmembrane pressure afford
ed little improvement. Although initial cell concentration affected ad
versely the permeation rates, growth medium components, especially soy
-peptone, was deemed pivotal in determining the characteristics of cel
l cake, thus controlling yeast microfiltration.