Improved cell line development by a high throughput affinity capture surface display technique to select for high secretors

https://doi.org/10.1016/S0022-1759(99)00181-7Get rights and content

Abstract

A novel process is described which permits rapid and objective selection of rare cells from a heterogeneous population based on quantity of secreted target protein. The process involves construction of an immobilised affinity surface display matrix that specifically binds secreted target product which is then detected using a fluorescent labelled ligand. Cells with the highest fluorescence can then be sorted using conventional flow cytometric technology. Overall, the whole process can be completed in less than 4 h during which time in the region of five million cells can be analysed. Cells are rapidly selected for in a quantitative manner compared to traditional methods which can take several months and have a reduced probability of finding low abundance high secretors due to practical limitations imposed on the number of cells which can be screened.

Introduction

Selection of high producing cell lines is an important first step in the development of any bioprocess. For protein production from animal cells, such selected cell lines have traditionally been delivered by rounds of limiting dilution cloning followed by product analysis. However, there are several drawbacks to this route. In the first instance, it is both labour intensive and costly. Secondly, the whole process is time consuming because two rounds of cloning are required to improve the theoretical confidence of achieving clonality. Clonality is important to avoid overgrowth of high producers by low productivity variants (Frame and Hu, 1990; Ozturk and Palsson, 1990) which usually have higher growth rates than high producers (Richieri et al., 1991). As such, the whole process can take in excess of 8 months to complete and even then there is no guarantee that the cloned cell line will be stable and so useful for industrial bioprocessing. Finally, selection of the highest producers can be compromised by practical limitations on the number of cells that can be screened thereby potentially reducing the efficiency of selection of low abundance, high productivity cells. Clearly there is a need to develop an objective, rapid and simple method to select for high producing cell lines.

Flow cytometry (FC) has made it easier to monitor productivity and to isolate cells with specific characteristics (Al-Rubeai, 1999). Important advantages of FC include the ability to screen large numbers of cells rapidly, the capability to distinguish cell sub-populations and the ability to efficiently select low abundance cells demonstrating the desired characteristics. There are two possible approaches for the selection of high productivity cells using FC, both of which have been successfully applied to the selection of hybridoma cells. The first approach is based on the cell surface antibody content; hybridoma cells displaying an increased amount of cell surface antibody can be identified and recovered through the use of fluorescent labelled antibodies to the hybridoma product. Antigen specific hybridomas, isotype switch variants (Dangl et al., 1982), higher avidity variants (Martel et al., 1988) and bispecific hybridomas (Jantscheff et al., 1993) have all been identified and cloned based on characterisation of cell surface immunoglobulin. While these groups have shown a qualitative correlation between cell surface and secreted antibody, a quantitative correlation has not been broadly documented. Indeed, a kinetic analysis of hybridoma clones reported no correlation between the amount of cell surface antibody and the amount of antibody in cell culture supernatants (Meilhoc et al., 1989). However, it is possible that this discrepancy in quantitative analysis was due to the authors overlooking the effects of cell surface area. High surface antibody content may be related to the larger surface area of cells during S and G2 cell cycle phases and may give false values for cells that are actually low producers. Normalisation of cell surface antibody concentration per unit area and synchronisation of cells so that they are examined in the same cell cycle phase represent approaches for optimisation which may enable the application of this simple method of product monitoring and selection (Cherlet et al., 1995).

Recently, approaches have been developed to select cells based on secreted antibody as an alternative strategy to circumvent some of the limitations of cell surface antibody selection. Two broad approaches have been developed, namely the affinity matrix and the gel microdrop (Gray et al., 1995; Kenney et al., 1995; Manz et al., 1995). The former method is based on creation of an artificial affinity matrix, specific for the secreted product of interest. Secreted molecules bind to the affinity matrix on the surface of the secreting cell and are subsequently labelled with specific fluorescent reagents for flow cytometric analysis and cell sorting. The matrix itself is created by direct linkage of avidinated specific capture antibodies to the previously biotinylated cell surface. Secreted antibodies are retained directly on the surface of the originating cell and product cross feeding is prevented by using a medium of low permeability. Thus, it is possible to analyse and select cells on the basis of secretory activity. The potential efficiency of this technique may be dictated by a combination of the affinity constant and capacity of the cell surface matrix for the secreted product, and the product diffusion rate in the low permeability medium (Frykman and Scrienc, 1998).

Microdrop encapsulation involves, as its name implies, complete encapsulation of single cells in agarose beads (Gray et al., 1995; Kenney et al., 1995). These beads contain specific capture antibodies and so simultaneously capture secreted product and prevent cross feeding of product between cells. Again the quantity of secreted product is determined by binding of a fluorescent ligand and the cells are sorted or cloned by FC while still encapsulated within the microdrop. The ability to quantify secretion at the single cell level is a unique and compelling feature of these technologies, since it is not otherwise possible to analyse and rapidly sort individual cells quantitatively for secretory activity.

However, these techniques are not without shortcomings. Microdrop encapsulation is frequently described as being user unfriendly, needing much optimisation of conditions for each cell line used and also requires dedicated equipment and extra time for encapsulation. Additionally, to ensure single cell occupancy of the microdrops only about 5% of the drops created actually contain a cell, consequently around 95% of the particles analysed by the sorter are devoid of cells and are in effect wasted. Also, decapsulation can reduce the viability of some cell types commonly used for expression of recombinant proteins, most notably myeloma NS0 cells. The direct binding approach addresses some of these problems. For example, it is simpler and requires neither encapsulation or decapsulation of the cells, but it does require chemical modification of some of the reagents and the product saturation limit is theoretically an order of magnitude lower than for encapsulated cells (Frykman and Scrienc, 1998). Despite these shortcomings the simplicity of direct binding makes it the approach most likely to achieve the aims highlighted above. Additionally, incorporation of some form of correction for cell size effect, which is a key factor in determining secretion levels for mammalian cells (Holmes and Al-Rubeai, in press), would be desirable but has yet to be considered by either of the methods described above.

Clearly there is interest in developing a generic technology for selecting high producing cells based on secreted product. Such a technology needs to be cheap, simple enough to become routine, high throughput, very rapid, objective and adaptable to many different secreted products. The development of such a technology was the purpose of this work.

Section snippets

Cell line

NSO 6A1 (100)-3, a myeloma cell line expressing b72.3, a chimeric antibody specific to the breast tumour antigen TAG73, was given to this laboratory by Lonza Biologics. This cell line utilises the powerful GS promoter system for high level expression and was originally cloned by the limiting dilution method (Bebbington et al., 1992). These cells were routinely maintained by passaging every third day into fresh GMEM (Gibco, UK) supplemented with 5% Foetal Calf Serum (Gibco), 500 μM glutamic

Receptor localisation

The specific and essentially irreversible binding of avidin and biotin has great potential not only for immobilisation of ligands to cell surfaces but also for amplification, due to avidin having a valency for four biotin molecules. Additionally, biotinylation of cells and proteins is a relatively straightforward procedure and there are also many biotinylated antibodies readily available commercially at relatively low cost. We designed our surface display capture matrix as shown in Fig. 1 where

Acknowledgements

This work was funded by the EC framework IV programme.

References (18)

There are more references available in the full text version of this article.

Cited by (59)

  • Selecting cells expressing high levels of recombinant proteins using the GPI-anchored protein with selenocysteine system

    2021, Journal of Bioscience and Bioengineering
    Citation Excerpt :

    However, despite important advances in these methods, the rapid selection of cells expressing high levels of secreted recombinant proteins is still difficult to achieve. Several strategies have been proposed to select high-producing cell clones, such as the gel microdrop secretion assay, and the crosslinking of an affinity matrix on the cell surface within a gelatin-based low-permeability medium to capture the secreted recombinant proteins, followed by the additions of fluorescently-labeled antibodies and the subsequent processing by a cell sorter (32,33). However, the methods are often technically challenging and time consuming.

  • A Nanoparticle-Based Affinity Sensor that Identifies and Selects Highly Cytokine-Secreting Cells

    2019, iScience
    Citation Excerpt :

    We designed our cytokine capture surface as shown in Figure 1A. In our approach, cells first undergo surface biotinylation followed by the attachment of neutravidin and a biotinylated IL-6 capture antibody to form the capture surface (Holmes and Al-Rubeai, 1999). The capture surface enables the cytokine molecules secreted by cells to be immobilized on the cell surface immediately upon their release, before they become diluted in the medium.

  • Mitochondrial membrane potential identifies cells with high recombinant protein productivity

    2019, Journal of Immunological Methods
    Citation Excerpt :

    A group of FACS and magnetic bead selection techniques allow identification and selective enrichment of producing cells based on immobilization of the desired secreted proteins on or close to the cell surface. These methods include cold-capture (Brezinsky et al., 2003; Pichler et al., 2009), matrix-aided surface capture (Bohm et al., 2004; Holmes and Al-Rubeai, 1999) and gel microdrop technology (Hammill et al., 2000; Powell and Weaver, 1990). However some of the methods can display poor correlation between fluorescent signal intensity and relative cell productivity due to diffusion of secreted product and subsequent binding to nearby non-producer cells (Kumar and Borth, 2012).

  • High-efficiency secretory expression of human neutrophil gelatinase-associated lipocalin from mammalian cell lines with human serum albumin signal peptide

    2016, Protein Expression and Purification
    Citation Excerpt :

    This approach can be applied to the identification and isolation of cells secreting antibodies [34], membrane associated proteins, and smaller soluble proteins such as tPA [35]. Several selection methods, such as gel microdrop and affinity capture surface display, are used to retain the secreted protein near the producer cells [36–38]. Various strategies, such as biotinylated agarose in conjunction with avidin and biotinylated capture antibodies, are used to capture the secreted protein [39].

View all citing articles on Scopus
View full text