Advertisement

Baculovirus expression system scaleup by perfusion of high-density Sf-9 cell cultures

Authors

  • Antoine W. Caron,

    1. Groupe Ingénierie des Cellules Animales, Institut de Recherche en Biotechnologie, Conseil National de Recherches du Canada, 6100 ave Royalmount, Montréal, P. Québec, Canada, H4P 2R2
    Search for more papers by this author
  • Rosanne L. Tom,

    1. Groupe Ingénierie des Cellules Animales, Institut de Recherche en Biotechnologie, Conseil National de Recherches du Canada, 6100 ave Royalmount, Montréal, P. Québec, Canada, H4P 2R2
    Search for more papers by this author
  • Amine A. Kamen,

    1. Groupe Ingénierie des Cellules Animales, Institut de Recherche en Biotechnologie, Conseil National de Recherches du Canada, 6100 ave Royalmount, Montréal, P. Québec, Canada, H4P 2R2
    Search for more papers by this author
  • Bernard Massie

    Corresponding author
    1. Groupe Ingénierie des Cellules Animales, Institut de Recherche en Biotechnologie, Conseil National de Recherches du Canada, 6100 ave Royalmount, Montréal, P. Québec, Canada, H4P 2R2
    • Groupe Ingénierie des Cellules Animales, Institutde Recherche en Biotechnologie, Conseil National de Recherches du Canada, 6100 ave Royalmount, Montréal, P. Québec, Canada, H4P 2R2
    Search for more papers by this author

Abstract

A perfusion system based on a 4-L stirred tank bioreactor and a custom-designed tangential (cross-flow) filter was assembled to realize a scaleup of the Baculovirus Expression Vector System (BEVS). When perfused with 1 to 1.5 vol/day, Spodoptera frugiperda (Sf-9) insect cell cultures grew from 4 × 106 to 15 × 106 cells/mL over 3 to 4 days. The possibility of maintaining high specific production of recombinant VP6 protein (from bovine rotavirus) after baculovirus infection of the high-density cultures was then assessed. The process consisted of a growth phase in TNMFH + 10% FBS, followed by infection with Bac-BRV6L recombinant baculovirus and a shift to a low-serum (0 to 1%) medium for perfusion during the production phase. Multiple runs were executed, each including a battery of shaker flask controls at various cell densities and serum concentrations. On average, specific rVP6 production in the bioreactor amounted to 76% of that found in 20-mL shaker cultures simulatingthe bioreactor's high cell density, low serum concentration, and medium renewal rate. Mechanical stress generated by cell/medium separation in theperfusion process reduced cell growth rate but had minimal effect on rVP6production. Our results also indicated that serum concentration during the infection phase affected the rVP6 specific production in a cell density–dependent fashion. Although the feasibility of the cell density scale up was demonstrated, optimization is still needed to achieve a truly cost-effective process.

Ancillary