Evaluating the efficiency of CHEF and CMV promoter with IRES and Furin/2A linker sequences for monoclonal antibody expression in CHO cells

Abstract

In recent years, monoclonal antibodies (mAbs) have been developed as powerful therapeutic and diagnostic agents and Chinese hamster ovary (CHO) cells have emerged as the dominant host for the recombinant expression of these proteins. A critical step in recombinant expression is the utilization of strong promoters, such as the Chinese Hamster Elongation Factor-1α (CHEF-1) promoter. To compare the strengths of CHEF with cytomegalovirus (CMV) promoter for mAb expression in CHO cells, four bicistronic vectors bearing either internal ribosome entry site (IRES) or Furin/2A (F2A) sequences were designed. The efficiency of these promoters was evaluated by measuring level of expressed antibody in stable cell pools. Our results indicated that CHEF promoter-based expression of mAbs was 2.5 fold higher than CMV-based expression in F2A-mediated vectors. However, this difference was less significant in IRES-mediated mAb expressing cells. Studying the stability of the F2A expression system in the course of 18 weeks, we observed that the cells having CHEF promoter maintained their antibody expression at higher level than those transfected with CMV promoter. Further analyses showed that both IRES-mediated vectors, expressed mAbs with correct size, whereas in antibodies expressed via F2A system heterogeneity of light chains were detected due to incomplete furin cleavage. Our findings indicated that the CHEF promoter is a viable alternative to CMV promoter-based expression in F2A-mediated vectors by providing both higher expression and level of stability.

Fig 1. Schematic representation of the vectors containing different promoters and furin/2A and IRES sequence for mAb expression.

(A) CHEF-LC-F2A-HC (CHEF-F2A) vector, (B) CMV-LC-F2A-HC vector, (C) CHEF-LC-IRES-HC vector, (D) CMV-LC-IRES-HC vector. CHEF; CHEF promoter, CMV;CMV promoter, SP; signal peptide, LC; Light chain, Furin; Furin recognition site sequence, 2A; 2A self-cleavage sequence derived foot-and-mouth disease virus (FMDV), IRES; Internal ribosome entry site, HC; Heavy sequence, pA; poly A.

Fig 2. Schematic illustration showing insertion of furin/2A sequences between light and heavy chain by overlap extension PCR.

Fig 3. Analysis of the four bicistronic vectors for mAb and mRNA expression in established stable cell lines.

Stably transfected pools were generated by transfection of CHO-S cells with various bicistronic vectors containing either IRES or F2A sequences and different promoters (CHEF or CMV). Levels of mAb and mRNA expression were measured by ELISA and qRT-PCR respectively. Black bars represent the mAb titer and gray bars represent mRNA fold induction. The error bars represent the standard deviation of three independent measurements.

Fig 4. SDS-PAGE and Western blot analysis of purified mAb in stable cell pools transfected with four bicistronic vectors.

Supernatants of three different cell pools were purified using protein-A. Purified samples were analyzed with SDS-PAGE and western blot under reducing and non-reducing condition. Commercial IgG1 was used as a positive control. (A) SDS-PAGE profile of purified samples in reduced state, lane 1; Protein molecular ladder, 2; CHEF-F2A, 3; CMV-F2A, 4; CHEF-IRES, 5; CMV-IRES, 6; positive control. (B) SDS-PAGE profile of purified samples in non- reduced state, lane 1; CMV-IRES, 2; CHEF-IRES, 3; CMV-F2A, 4; CHEF-F2A, 5; protein molecular ladder. (C) Western blot analysis of samples under reducing condition, lane 1; protein molecular ladder, 2; CHEF-F2A, 3; CMV-F2A, 4; CHEF-IRES, 5; CMV-IRES, 6; positive control. (D) Western blot analysis of purified samples in non- reduced state, lane 1; CMV-IRES, 2; CHEF-IRES, 3; CMV-F2A, 4; CHEF-F2A, 5; protein molecular ladder.

Fig 5. Western blot analyses of purified mAb in triplicate stable cell pools transfected with vectors bearing F2A-mediated expression system.

Light chains with various molecular weights (25, 28 & 30 kDa) were detected. Triplicates of each stable pool were shown with 1, 2 and 3. (A) Lane 1; CHEF-F2A (2), lane 2; CHEF-F2A (3), lane 3; protein molecular ladder. (B) Lane 1; protein molecular ladder, lane 2; CMV-F2A (1), lane 3; CMV-F2A (3).

Fig 6. Analysis of the stability of antibody expression over time by stable cell pools transfected with two vectors containing F2A sequences.

Both stable cells were cultivated for 18 weeks upon removal of puromycin as a selection marker. Every 2 weeks, antibody expression was monitored and measured by ELISA. The other pools exhibited the same expression pattern; the data from one of them was represented.

Fig 7. Screening of clonal cell lines established from each stable cell pools transfected with vectors containing F2A sequences.

Single cell were generated from CHEF-F2A and CMV-F2A cell pools by limiting dilution technique and expanded for additional analysis. The level of antibody expression was determined by ELISA. (A) Measuring the level of antibody expressed by clonal cell lines derived from CHEF-F2A cell pool. (B) Measuring the level of antibody expressed by clonal cell lines derived from CMV-F2A cell pool. (C) Clonal cell lines recovered from both cell pools were categorized according to level of antibody expression. The percentage of cell lines in each group was indicated.

Fig 8. Analysis of integrated transgene copy number in stable cell pools transfected by four bicistronic vectors.

Antibody copy number based on light chain copy number was calculated by qRT-PCR. The error bars represent the standard deviation of three independent measurements.