Fig 1.
(A) Schematic representation of pSplice vector construction. pSplice is derived from pOE scFv.Fc vector. pOE scFv.Fc vector can express the fusion protein only in mammalian cells while pSplice vector can express the fusion protein both, in E. coli and mammalian cells. The hatched pink portion represents peptide sequence AQPAMA, which is common to the hybrid (Hyb sig) and the modified pelB (Mod pelB) signal sequences. Expression in the bacterial system is directed by the Lac promoter (light blue) and a modified pelB signal peptide (dark green followed by hatched pink), which are placed within an intron between the 5’ and 3’ splicing sites (ss). Production of scFv-Fc in 293F cells is controlled by CMV promoter (light green) and a hybrid signal peptide (solid pink followed by hatched pink) when the intron is spliced out in 293F cells. (B) Schematic representation of the intron and the main changes introduced in the signal sequences. In brown is the hybrid signal sequence of pOEscFv.Fc. In pink is the point where the open reading frame of hybrid signal sequence is disrupted by insertion of the intron (blue line). Bold letters (nucleotide sequence) and arrows in the intron represent the splice donor and acceptor sites. In black is the pelB signal sequence. Immediately below, in green is the modified pelB signal sequence. Within this peptide sequence an A to Q mutation is indicated in red. The corresponding codon creates the splice acceptor site. In magenta is an insertion of a leucine in pelB signal sequence. The corresponding codon creates a polypyrimidine site in the intron.
Fig 2.
Upper panel: Codons and corresponding amino acids of pelB signal sequence showing splicing signal modifications in different versions of pSplice vectors. Introduction of branch point is indicated in brown. Introduction of the T-rich polypyrimidine tract (PPT) is shown in purple italic letters. Lower panel: Regions of engineering poly-G runs around -10 nucleotide of lac promoter are indicated by red letters in boxes.
Fig 3.
SDS-PAGE analysis of pSplice vectors in E.coli and 293F cells.
(A) Non reducing SDS-PAGE of scFv.Fc of clone 1 and 2 expressed from pSplice version 2, 3, 4 and 5 vectors in E.coli (lanes 1, 2, 3 and 4, respectively). (B) Non reducing SDS-PAGE of scFv.Fc of clone 1 and 2 expressed from pSplice version 2, 3, 4 and 5 vectors and the parental pOEscFv.Fc (lanes 1, 2, 3, 4 and 5, respectively) in 293F cells. Mammalian expression vector, pOEscFv.Fc is used to compare how the dual expressing pSplice vectors perform in comparison to pOEscFv.Fc vector.
Fig 4.
Expression levels of scFv.Fcs in mammalian cells.
Octet based quantitation of expression level at day 3 and 6 post-transfection of scFv.Fc clone 1 and 2 from pSplice version 2 (red), 3 (green), 4 blue) and 5 (magenta) vectors. The mean values for each group are derived from experiments done in triplicates and the s.d. is represented by error bars.
Fig 5.
Antigen binding activity of scFv.Fc of clones 1 and 2 expressed from different pSplice vectors in bacteria (A and B, respectively) and in 293F cells (C and D, respectively).
The symbols represent the following: circle-pSplice v.2; square- pSplice v.3; triangle- pSplice v.4; inverted triangle- pSplice v.5 and diamond- pOEscFv.Fc (only for mammalian expression in C and D). The scFv.Fc proteins did not bind to a negative control antigen (shown in S3 Fig). The mean values for each group are derived from experiments done in triplicates and the s.d. is represented by error bars.
Fig 6.
scFv.Fcs expressed in mammalian cells are biologically active.
Opsonophagocytosis activity of scFv.Fc fusion proteins for clone 1, clone 2 and negative control clone in supernatants from 293F cells transfected with the respective pSplicev.5 plasmid. Black and white bars represent OPK activity of scFv.Fc at 10 μg/ml and 1 μg/ml, respectively. The mean values for each group are derived from experiments done in triplicates and the s.d. is represented by error bars.
Fig 7.
Upper panel: H1 peak (properly processed N-terminus corresponding to EVQLLESGGGLVQPGGSLR) after UPLC separation of tryptic peptides of clone 2 scFv.Fc expressed in bacteria (A) and 293F cells (B). Peak corresponding to the estimated molecular weight of a correctly matured N-terminus is shown. Middle panel: (C) and (D) MS spectra of (A) and (B) in the upper panel. The estimated molecular weight for the H1 peptide with a charge of +1 is expected to be 1896.0185 da. The m/z of the major peaks in (C) and (D) correspond closely to the peptides shown in (A) and (B), respectively. Lower panel: (E) and (F) ms/ms spectra of the same sample in (A) and (B). The different peptides that the labelled peaks correspond to are as follows: Peak 1: PGGSLR; peak 2: QPGGSLR; peak 3: VQPGGSLR; peak 4: EVQLLESG; peak 5: SGGGLVQPGGSLR; peak 6: LESGGGLVQPGGSLR and peak 7: LLESGGGLVQPGGSLR. Automated and manual search for improperly processed signal peptide yielded only about 0.4% of AMA-H1 peptide in scFv.Fc purified from 293F cells (not shown).