Research paper
Prediction of recombinant protein production in an insect cell–baculovirus system using a flow cytometric technique

https://doi.org/10.1016/j.jim.2007.06.005Get rights and content

Abstract

The baculovirus expression vector system (BEVS) utilising the Autographa californica nucleopolyhedrovirus (AcMNPV) is widely becoming the system of choice for the production of many recombinant protein products due to the high yields obtained. However, there is a need to develop a simple reliable on-line method to monitor the production of recombinant proteins that have no intrinsic reporter properties. Here we utilise flow cytometry to measure cell size, granularity and DNA content in a single step analysis and correlate these parameters with the production of the recombinant protein β-galactosidase. Clear correlations between these parameters and productivity are made with forward and side scatter signals showing the highest correlation coefficients. Measuring these parameters does not require any processing of the cells from culture to analysis. These parameters can therefore be used successfully to predict the amount of recombinant protein product in a BEVS system on-line.

Introduction

One of the early applications of baculoviruses was as biological control agents in agriculture and forestry. Baculoviruses have since been proven to be efficient vectors for the production of a wide variety of gene products/recombinant proteins including monoclonal antibodies and tumour necrosis factor-β (TNF-β) (zuPutlitz et al., 1990, Chai et al., 1996). The particular aspect of the baculovirus replication strategy that lends itself for the efficient production of recombinant protein is its biphasic life cycle (Volkman et al., 1976). Virions are formed as singular particles, and embedded in occlusion bodies composed mainly of polyhedrin protein. Polyhedrin is expressed very late in the infection cycle so the rapid production of vast quantities of polyhedrin is due to the hyperexpression of the gene, controlled by a very strong promoter (Vlak and Rohrmann, 1985).

A baculovirus expression vector system (BEVS) requires a host cell. Most of the data on baculovirus infection in vitro are derived from AcMNPV infection of insect cells, Sf-21 and Sf-9, which are derivatives of primary cell lines from the lepidopteran, fall armyworm, Spodoptera frugiperda (Francki et al., 1991). BEVS systems with these insect cell lines are commonly used to produce recombinant protein. It is a very productive system with the strong polyhedrin and p10 promoters coupled with high cell densities achievable with insect cell cultures, > 1 × 107 cells/ml (Elias et al., 2000).

The potential of the BEVS system as an industrial and commercial process is clear, although to date no approved BEVS-made product is on the market for human vaccine or therapeutic use. However, there are a number of products in advanced clinical trials (Cox, 2004): for example, PROVENGE®, from Dendreon Corporation, is a therapeutic for prostate cancer in late stage phase III clinical trials and has received fast track review status from the FDA; FLUBLOK™, from Protein Sciences Corporation, is a therapeutic against influenza in phase III clinical trials (Cox, 2004). With recent developments in improving the host cell line and continued research into improving the baculovirus vector, the potential of the BEVS system is increasing. Therefore, there is a value-added benefit for improved monitoring methods of the BEVS process.

Models have been produced that simulate cell population dynamics, virion densities and product titres for a wide range of MOIs that concur with experimental data (Power et al., 1994, Wong et al., 1996). These models are very useful in understanding and predicting the route and outcome of processes, however, it is risky to predict the harvest strategy as a model does not account for the efficacy of the virus stock (e.g. the infectivity and replication competence of the virus). Therefore, monitoring of the particular process is ultimately essential to observe the propagation of the virus in the culture.

Flow cytometry is a very versatile monitoring tool that allows for simple, on-line measurement of morphological and many physiological parameters, primarily based on the principle of measuring light scattering and fluorescence (Al-Rubeai and Emery, 1993). Many cellular components can be conjugated with fluorochromes that will emit a specific light wavelength profile after excitation (Al-Rubeai and Emery, 1993). The use of flow cytometry has revealed many otherwise unknown properties by examining the intrinsic properties of individual cells in mammalian and insect cell cultures (Chai et al., 1996, Al-Rubeai and Emery, 1993, Al-Rubeai et al., 1992, Al-Rubeai et al., 1993, Al-Rubeai et al., 1995, Leelavatcharamas et al., 1999, Gueret et al., 2002, Kioukia et al., 1995, Farmer et al., 1989, Schorpf et al., 1990) with the first attempt to demonstrate the application of flow cytometry to process identification and control by Al-Rubeai et al. (1992). They used flow cytometric measurement of the G1 phase of the cell cycle to drive the nutrient feed rate strategy in a perfusion culture of insect cells. Later Kioukia et al. (1996) reported further exploitation of the potential of flow cytometry to closely monitor viral infection and protein expression in insect cell–baculovirus cultures. Flow cytometry has been used to characterise the sub-populations of Sf-9 cells showing that tetraploid clones are superior to diploid clones as host cell lines in BEVS for producing recombinant protein (Jarman-Smith et al., 2002, Jarman-Smith et al., 2004).

Physiological and morphological changes upon infection by viruses have been observed in many cell lines used for the production of proteins and viruses. During the late phase (6–24 h post infection) infected insect cells stop dividing and increase in diameter (Luckow, 1991). The nucleus is also enlarged due to the increase in the amount of viral DNA, a parameter that can be monitored by flow cytometry. Cell size is increased as a result of viral infection and multiplication (Palomares et al., 2001, Zeiser et al., 2000, Janakiraman et al., 2006). In the very late phase (24–36 h) cells cease to produce budded virus and begin production of recombinant protein (Luckow, 1991). The nucleus and cytoplasm will contain assembled baculovirus so should demonstrate change in granularity, and along with the morphological changes listed above, an influence on flow cytometric forward scatter and side scatter should be observed. Correlation with the time of maximum modal insect cell diameter and the time of maximum recombinant protein concentration has been demonstrated (Palomares et al., 2001).

Here we utilise a one-step, single-cell analysis to monitor various cellular parameters (DNA content, cell size and cell granularity) in an insect cell–baculovirus system and correlate them with the production of recombinant protein.

Section snippets

Cell line, cell culture conditions and viral stock

The insect cell line used here was Sf-9 (ATCC CRL 1711) a sub-clone of the primary cell line IPBL-Sf-21 (obtained from the Department of Cancer Studies, University of Birmingham). Cells were routinely maintained in 100 ml spinner flasks (DURAN bottles) in TC-100 medium (Gibco, Paisley, UK) supplemented with 5% v/v foetal calf serum (FCS; PAA Laboratories Ltd., Yeovil, UK), 50 U/ml penicillin, 50 μg/ml streptomycin and 0.05% w/v Pluronic F-68 (BASF). The cultures were incubated at 27 °C and

Results and discussion

An increase in viable cell number was observed up to 24 h post infection for MOI 1.0 and up to 50 h post infection for MOI 0.1 (Fig. 1). This is in a stark contrast to higher MOIs of 10 and 50 when a steady decline in viable cell number post infection was observed. This can be attributed to the fact that at low MOIs the proportion of initial cell infection is low, therefore only the infected proportion of cells stops dividing; the non-infected cells continue to divide, the proportion of which

Conclusions

This paper reports on the use of simple flow cytometric markers to predict productivity in an application where the recombinant protein has no simple monitoring marker. It is shown that the forward and side scatter parameters can be obtained in a single step analyses and can be used for monitoring product formation. Live or fixed culture samples are analysed in a few minutes and a simple model can be used to predict productivity. Monitoring of DNA content at an early stage after infection has

References (36)

  • J.M. Vlak et al.

    The nature of polyhedrin

  • M. Al-Rubeai et al.

    Flow cytometry in animal cell culture; using FC to optimise in vitro cell culture processes

    Bio/Technology

    (1993)
  • M. Al-Rubeai et al.

    Cell cycle and cell size dependence of susceptibility to hydrodynamic forces

    Biotechnology and Bioengineering

    (1995)
  • CoxM.M.J.

    Commercial production in insect cells: one company's perspective

  • V. Deparis et al.

    Use of flow cytometry to monitor infection and recombinant human r-1,3/4 fucosyltransferase production in baculovirus infected Sf9 cell cultures

    Biotechnology Progress

    (2003)
  • C.B. Elias et al.

    Enhanced growth of Sf-9 cells to a maximum density of 5.2 × 107 cells per mL and production of β-galactosidase at high cell density by fed batch culture

    Biotechnology and Bioengineering

    (2000)
  • R.I.B. Francki et al.

    Fifth Report of the International Committee on Taxonomy of Viruses

    (1991)
  • V. Gueret et al.

    Rapid titration of adenoviral infectivity by flow cytometry in batch culture of infected HEK293 cells

    Cytotechnology

    (2002)
  • Cited by (0)

    View full text