Expression of Multispecific Antibodies.

Bi- and multispecific antibody formats allow the development of new therapeutic strategies to address previously unmet medical needs. However, due to the increased complexity (e.g., the interface design and the presence of multiple binders), such molecules are generally more challenging to express and purify compared to standard monoclonal antibodies (mAbs). We describe here an optimized methodology to express and purify basic bispecific antibodies using the BEAT® interface. This interface allows to generate antibodies with very high levels of heterodimer product (reported titers exceed 10 g/L) and comes with a built-in purification strategy allowing removal of residual levels of undesired product-related impurities (e.g., homodimers and half molecules).

Expression of Bispecific Antibodies Using Cellular Display.

The identification and selection of high-producing cell lines can be a resource- and time-consuming process. The screening effort can be simplified by assessing the potential for high expression (or a desired product quality attribute) of the individual cell directly in a mix of cells. Here, we describe protocols for the use of such a cellular display technology. Using alternate splicing, two mRNA constructs are generated at tunable ratios. The first mRNA codes for the secreted product, the second mRNA attaches a transmembrane domain to the antibody and directs it to the cellular membrane. The design of the basic construct as well as efficient ways to tune the strength of the cellular display is detailed in this chapter. Further, enrichment methods are provided enabling the flow cytometric sorting of a cell population based on the quantity of cellular display or on the product quality (heterodimerization level of a bispecific antibody).

Development of a 10 g/L process for a difficult-to-express multispecific antibody format using a holistic process development approach.

Ichnos has developed a multi-specific antibody platform based on the BEAT® (Bispecific engagement by antibodies based on the T-cell receptor) interface. The increased complexity of the bi- and multi-specific formats generated with this platform makes these molecules difficult-to-express proteins compared to standard monoclonal antibodies (mAbs). This report describes how expression limitations of a bi-specific bi-paratopic BEAT antibody were improved in a holistic approach. An initial investigation allowed identification of a misbalance in the subunits composing the BEAT antibody as the potential root cause. This misbalance was then addressed by a signal peptide optimization, and the overall expression level was increased by the combination of two vector design elements on a single gene vector. Further improvements were made in the selection of cell populations and an upstream (USP) platform process was applied in combination with a cell culture temperature shift. This allowed titer levels of up to 6 g/L to be reached with these difficult-to-express proteins. Furthermore, a high-density seeding process was developed that allowed titers of around 11 g/L for the BEAT antibody, increasing the initial titer by a factor of 10. The approach was successfully applied to a tri-specific antibody with titer levels reaching 10 g/L. In summary, a platform process for difficult-to-express proteins was developed using molecular biology tools, cell line development, upstream process optimization and process intensification.

Alternative splicing for tuneable expression of protein subunits at desired ratios.

The controlled expression of two or more proteins at a defined and stable ratio remains a substantial challenge, particularly in the bi- and multispecific antibody field. Achieving an optimal ratio of protein subunits can facilitate the assembly of multimeric proteins with high efficiency and minimize the production of by-products. In this study, we propose a solution based on alternative splicing, enabling the expression of a tunable and predefined ratio of two distinct polypeptide chains from the same pre-mRNA under the control of a single promoter. The pre-mRNA used in this study contains two open reading frames situated on separate exons. The first exon is flanked by two copies of the chicken troponin intron 4 (cTNT-I4) and is susceptible to excision from the pre-mRNA by means of alternative splicing. This specific design enables the modulation of the splice ratio by adjusting the strength of the splice acceptor. To illustrate this approach, we developed constructs expressing varying ratios of GFP and dsRED and extended their application to multimeric proteins such as monoclonal antibodies, achieving industrially relevant expression levels (>1 g/L) in a 14-day fed-batch process. The stability of the splice ratio was confirmed by droplet digital PCR in a stable pool cultivated over a 28-day period, while product quality was assessed via intact mass analysis, demonstrating absence of product-related impurities resulting from undesired splice events. Furthermore, we showcased the versatility of the construct by expressing two subunits of a bispecific antibody of the BEAT® type, which contains three distinct subunits in total.

Viral inactivation for pH-sensitive antibody formats such as multi-specific antibodies.

Recently developed multi-specific antibody formats enable new therapeutic concepts. Conveniently, formats with an Fc domain allow purification in well-established mAb platform processes. However, due to the structural complexity of the formats, the assembled molecules may be sensitive to extreme pH commonly used for viral inactivation. An alternative to low pH incubation for virus inactivation is the use of a mixture of tri-n-butyl phosphate (TnBP, solvent) and Polysorbate 80 (PS80, detergent). While TnBP is toxic, this combination has a long history of use in the manufacturing of human plasma-derived products that are sensitive to low or high pH incubation. Data are provided demonstrating that the solvent/detergent (S/D) treatment using TnBP and PS80 can be successfully used for pH-sensitive, multi-specific antibody formats in the clarified cell culture fluid (CCCF). A different placement of the S/D within the purification process, namely during the capture by Protein A (PA), has been evaluated. This alternative placement allows effective viral inactivation by S/D while preserving the viral reduction and viral inactivation achieved through the PA step itself, enabling the cumulation of these effects. Furthermore, the process alternative simplifies the liquid handling by reducing the added volumes of the required S/D liquids, thus reducing the amount of toxic TnBP to a minimum. Data are shown demonstrating a complete removal of TnBP and PS80 in the process.

Benchtop Bioreactors in Mammalian Cell Culture: Overview and Guidelines.

Bioreactors are manufactured apparatuses that allow the generation of a specific environment for the highly controlled cultivation of living cells. Originally used for microbial production systems, they have found widespread applications in fields as diverse as vaccine production, plant cell cultivation, and the growth of human brain organoids and exist in equally diverse designs (Chu and Robinson, Curr Opin Biotechnol 12(2):180-187, 2001; Qian et al., Nat Protoc 13:565-580, 2018). Manufacturing of biologics is currently mostly performed using a stirred tank bioreactor and CHO host cells and represents the most "classical" bioreactor production process. In this chapter, we will therefore use the cultivation of suspension Chinese hamster ovary (CHO) cells for recombinant protein production in a stirred tank bioreactor as an example. However, general guidelines provided in this chapter are transferable to different bioreactor types and host cells (Li et al., MAbs 2(5):466-479, 2010).The preparation and operation of a bioreactor (also referred to as upstream process in a biotechnological/industrial setting) is comprised of three main steps: expansion (generation of biomass), production (batch, fed-batch, or continuous process), and harvest. The expansion of cells can last from few days to weeks depending on the number of cells at the start, the cellular doubling time, and the required biomass to inoculate the production bioreactor. The production phase lasts a few weeks and is a highly sensitive phase as the concentration of different chemicals and physical parameters need to be tightly controlled. Finally, the harvest will allow the separation of the product of interest from large particles and then the desired material (cell culture supernatant or cells) is transferred to the downstream process.The raw materials used during the upstream phase (all three steps) need to be aligned with the final purpose of the manufactured product, as the presence of residual impurities may have an impact on suitability of the final product for a desired purpose.

Generation of homogeneous cell populations with tunable levels of transgene expression.

We describe here a vector construct to establish homogeneous cell populations expressing a recombinant gene of interest (GOI) at tuneable levels, including low expression levels that are difficult to generate using standard cell line development techniques. This is achieved using a tricistronic mRNA that contains an open reading frame for the gene of interest, a first internal ribosome entry site (IRES), an open reading frame for a fluorescent reporter protein (such as green fluorescent protein, GFP), a second IRES and an open reading for an antibiotic resistance gene (such as puromycin N-acetyl-transferase, PAC). The resistance gene allows convenient selection of stable cell populations. The fluorescent reporter protein allows convenient homogeneity and expression stability assessments of the cell line. The expression level of the GOI can be adjusted by using different start codons for the open reading frame. These alternate start codons will initiate the translation of the GOI with different efficiency, leading to cell populations expressing different levels of the GOI, and similar levels of the fluorescent reporter through the first IRES and the puromycin resistance gene through the second IRES to the GOI. Such cell populations are useful tools, for instance to assess the safety of potent targeted therapeutics, as they allow the simplified generations of homogenous cell populations with different levels of target protein expression between populations.

SPLICELECT™: an adaptable cell surface display technology based on alternative splicing allowing the qualitative and quantitative prediction of secreted product at a single-cell level.

We describe a mammalian expression construct (SPLICELECT™) that allows the redirection of a proportion of a secreted protein onto the cell surface using alternative splicing: whereas the majority of the RNA is spliced into a transcript encoding a secreted protein, a weak splice donor site yields a secondary transcript encoding, in addition, a C-terminal transmembrane domain. The different sequence elements can be modified in order to modulate the level of cell surface display and of secretion in an independent manner. In this work, we demonstrated that the cell surface display of stable cell lines is correlated with the level of the secreted protein of interest, but also with the level of heterodimerization in the case of a bispecific antibody. It was also shown that this construct may be useful for rapid screening of multiple antibody candidates in binding assays following transient transfection. Thus, the correlation of product quantity and quality of the secreted and of membrane-displayed product in combination with the flexibility of the construct with regards to cell surface display/secretion levels make SPLICELECT™ a valuable tool with many potential applications, not limited to industrial cell line development or antibody engineering.

Analytical assessment of clonal derivation of eukaryotic/CHO cell populations.

Cell populations that are exclusively composed by descendants from a defined single parental progenitor are referred to as "monoclonal", "clonal" or more correctly as "clonally derived cell population". Clonal derivation of cell lines used in the manufacturing of recombinant biologics is a regulatory requirement that aims to ensure a robust process and consistent quality throughout the life cycle of a product. Clonal derivation of cell lines is usually ensured by the process (e.g. two subsequent rounds of limiting dilution). Here we present an approach to analytically assess the probability of clonal derivation of existing cell populations. Using target locus amplification (TLA) followed by next generation sequencing (NGS), unique genetic features can be identified, for example the integration site of the plasmid in the host cell genome or plasmid-plasmid junctions of the plasmids used for expression of recombinant biologics. Whereas a direct assessment of clonal derivation using TLA/NGS data is challenging due to limitations in specificity, confirmed clonally derived populations generated from the cell line population can be analyzed by qPCR for the presence of the unique genetic features identified by TLA/NGS. In the present study, a statistical analysis allowed the demonstration that two independently generated CHO cell lines were clonally derived with an upper 95% confidence interval limit of a potentially present contaminating population of 1.3%.

A novel anti-CD19 monoclonal antibody (GBR 401) with high killing activity against B cell malignancies.

BACKGROUND: CD19 is a B cell lineage specific surface receptor whose broad expression, from pro-B cells to early plasma cells, makes it an attractive target for the immunotherapy of B cell malignancies. In this study we present the generation of a novel humanized anti-CD19 monoclonal antibody (mAb), GBR 401, and investigate its therapeutic potential on human B cell malignancies. METHODS: GBR 401 was partially defucosylated in order to enhance its cytotoxic function. We analyzed the in vitro depleting effects of GBR 401 against B cell lines and primary malignant B cells from patients in the presence or in absence of purified NK cells isolated from healthy donors. In vivo, the antibody dependent cellular cytotoxicity (ADCC) efficacy of GBR 401 was assessed in a B cell depletion model consisting of SCID mice injected with healthy human donor PBMC, and a malignant B cell depletion model where SCID mice are xenografted with both primary human B-CLL tumors and heterologous human NK cells. Furthermore, the anti-tumor activity of GBR 401 was also evaluated in a xenochimeric mouse model of human Burkitt lymphoma using mice xenografted intravenously with Raji cells. Pharmacological inhibition tests were used to characterize the mechanism of the cell death induced by GBR 401. RESULTS: GBR 401 exerts a potent in vitro and in vivo cytotoxic activity against primary samples from patients representing various B-cell malignancies. GBR 401 elicits a markedly higher level of ADCC on primary malignant B cells when compared to fucosylated similar mAb and to Rituximab, the current anti-CD20 mAb standard immunotherapeutic treatment for B cell malignancies, showing killing at 500 times lower concentrations. Of interest, GBR 401 also exhibits a potent direct killing effect in different malignant B cell lines that involves homotypic aggregation mediated by actin relocalization. CONCLUSION: These results contribute to consolidate clinical interest in developing GBR 401 for treatment of hematopoietic B cell malignancies, particularly for patients refractory to anti-CD20 mAb therapies.

The kinetics of polyethylenimine-mediated transfection in suspension cultures of Chinese hamster ovary cells.

The kinetics of polyethylenimine (PEI)-mediated gene transfer at early times after transfection of Chinese hamster ovary (CHO) cell in suspension were investigated using a novel in vitro assay. Addition of an excess of competitor DNA to the culture medium at various times after the initiation of transfection inhibited further cellular uptake of PEI-DNA particles. Using this approach, a constant rate of particle uptake was observed during the first 60 min of transfection at a PEI:DNA ratio of 2:1 (w/w) and a cell density of 2 x 10(6) cells/ml under serum-free conditions. The uptake rate declined considerably during the next 2 h of transfection. Both the rate and the level of PEI-DNA uptake in serum-free minimal medium were found to be dependent on the PEI-DNA ratio, the cell density at the time of transfection, and the extent of particle aggregation. These studies of the early phase of PEI-mediated transfection are expected to lead to further opportunities for optimization of gene transfer to suspension cultures of mammalian cells for the purpose of large-scale transient recombinant protein production.

Disassembly of polyethylenimine-DNA particles in vitro: implications for polyethylenimine-mediated DNA delivery.

Here a simple in vitro assay was used to investigate the disassembly of nanoparticles of polyethylenimine (PEI) and DNA. Particles were formed with various PEIs, allowed to mature for 10 min, and then exposed to different competitors (RNA, DNA, BSA or heparin) or to different conditions of pH or osmolarity. DNA release was determined by gel electrophoresis or spectroscopy. The presence of heparin or high salt yielded complete particle disassembly for all PEIs tested. The addition of RNA to particles formed with linear PEIs or branched 2 kDa PEI resulted in rapid DNA release, but RNA induced only partial disassembly of particles formed with large branched PEIs. In the presence of competitor DNA, slow disassembly was observed with particles made with linear PEIs or branched 2 kDa PEI but not for particles made with larger branched PEIs. The presence of BSA resulted in partial disassembly of PEI-DNA particles, but acidic pH did not affect particle stability. If particles were allowed to mature longer than 10 min in NaCl, subsequent heparin-mediated DNA release decreased as the incubation time and the PEI:DNA ratio increased. However, particles that matured in culture medium were disassembled by heparin independently of maturation time or PEI:DNA ratio. It was concluded that branched PEIs have a higher affinity for DNA than linear PEIs, that the intracellular disassembly of PEI-DNA particles may involve interactions between PEI and cellular RNA, and that extended maturation of PEI-DNA particles in NaCl prior to transfection may limit the intracellular release of plasmid DNA.

Multiple glycerol shocks increase the calcium phosphate transfection of non-synchronized CHO cells.

The exposure of CHO DG44 cells to an osmotic shock, after DNA uptake, results in a cellular volume decrease of approx. 55%. Repetitive osmotic shocks targeted different sub-populations of cells as was demonstrated using two different fluorescent reporter genes. Also the exposure of a calcium phosphate-DNA coprecipitate to high osmolarity in vitro caused the release of the DNA from the precipitate. The results demonstrate the importance of the osmotic shock on the efficient delivery of plasmid DNA to the nucleus of CHO cells following calcium phosphate-mediated transfection.

Reduction of adenovirus E1A mRNA by RNAi results in enhanced recombinant protein expression in transiently transfected HEK293 cells.

Human embryonic kidney 293 (HEK293) cells, a widely used host for large-scale transient expression of recombinant proteins, are transformed with the adenovirus E1A and E1B genes. Because the E1A proteins function as transcriptional activators or repressors, they may have a positive or negative effect on transient transgene expression in this cell line. Suspension cultures of HEK293 EBNA (HEK293E) cells were co-transfected with a reporter plasmid expressing the GFP gene and a plasmid expressing a short hairpin RNA (shRNA) targeting the E1A mRNAs for degradation by RNA interference (RNAi). The presence of the shRNA in HEK293E cells reduced the steady state level of E1A mRNA up to 75% and increased transient GFP expression from either the elongation factor-1alpha (EF-1alpha) promoter or the human cytomegalovirus (HCMV) immediate early promoter up to twofold. E1A mRNA depletion also resulted in a twofold increase in transient expression of a recombinant IgG in both small- and large-scale suspension cultures when the IgG light and heavy chain genes were controlled by the EF-1alpha promoter. Finally, transient IgG expression was enhanced 2.5-fold when the anti-E1A shRNA was expressed from the same vector as the IgG light chain gene. These results demonstrated that E1A has a negative effect on transient gene expression in HEK293E cells, and they established that RNAi can be used to enhance recombinant protein expression in mammalian cells.

Serum-free large-scale transient transfection of CHO cells.

To date, methods for large-scale transient gene expression (TGE) in cultivated mammalian cells have focused on two transfection vehicles: polyethylenimine (PEI) and calcium phosphate (CaPi). Both have been shown to result in high transfection efficiencies at scales beyond 10 L. Unfortunately, both approaches yield higher levels of recombinant protein (r-protein) in the presence of serum than in its absence. Since serum is a major cost factor and an obstacle to protein purification, our goal was to develop a large-scale TGE process for Chinese hamster ovary (CHO) cells in the absence of serum. CHO-DG44 cells were cultivated and transfected in a chemically defined medium using linear 25 kDa PEI as a transfection vehicle. Parameters that were optimized included the DNA amount, the DNA-to-PEI ratio, the timing and solution conditions for complex formation, the transfection medium, and the cell density at the time of transfection. The highest levels of r-protein expression were observed when cultures at a density of 2.0 x 10(6) cells/ml were transfected with 2.5 microg/ml DNA in RPMI 1640 medium containing 25 mM HEPES at pH 7.1. The transfection complex was formed at a DNA:PEI ratio of 1:2 (w/w) in 150 mM NaCl with a 10-min incubation at room temperature prior to addition to the culture. The procedure was scaled up for a 20-L bioreactor, yielding expression levels of 10