Baculovirus expression cassette vectors for rapid production of complete human IgG from phage display selected antibody fragments.

For the expression of human intact IgG antibodies, we have constructed a set of baculovirus expression vectors designed to facilitate rapid insertion of heavy and light chain genes of Fab or scFv antibodies derived from phage display antibody libraries. By linking them to human constant or Fc regions, expression of complete human immunoglobulin molecules was achieved in insect cells by infection with recombinant baculovirus. The IgG expression cassette vectors are based on the backbone vector which contains two back to back polyhedron and p10 promoters. The IgG expression cassette elements, including the authentic IgG lambda or kappa and heavy chain signal sequences, as well as light chain (lambda or kappa) and heavy chain constant region genes are combined in a single vector and are controlled by the p10 and polyhedron promoter respectively. Either of VL or Fab-L and VH or Fab-Fd genes from common phage display systems can be directly inserted into one of the cassette vectors through in-frame cloning sites. This design of a single cassette vector combining heavy and light chain expression elements allowed rapid production and secretion of correctly processed and assembled intact immunoglobulins from recombinant baculovirus infected insect cells. The recombinant antibodies showed the expected molecular size of the H2L2 heterodimer in non reducing SDS-PAGE. No apparent differences were found between the expression level of heavy and light chains, and antigen binding function was preserved. For various antibodies, yields between 6 and 18 mg/l IgG were obtained.

Effects of unpaired cysteines on yield, solubility and activity of different recombinant antibody constructs expressed in E. coli.

New E. coli vectors based on the pOPE/pSTE vector system [Gene 128 (1993) 97] were constructed to express a single-chain Fv antibody fragment (scFv), a scFv-streptavidin fusion protein and two disulfide bond-stabilized Fv antibody fragments (dsFvs) utilizing different side chain positions for disulfide stabilization. All of these constructs encoded fusion proteins carrying five C-terminal histidine residues preceded by an unpaired cysteine. The influence of this cysteine, which was originally introduced to allow the chemical modification of the fusion proteins, was assessed by exchanging the two amino acids CysIle in front of the carboxy terminal His-tag to SerHis in all constructs. Yield and antigen-binding activity of the antibody constructs were compared after standard lab-scale periplasmic expression in Escherichia coli. The removal of the unpaired cysteine resulted in a significant increase in antigen-binding activity of the crude periplasmic extracts. Further, a three-five fold increase of yield and a significantly improved purity were observed after immobilized metal affinity chromatography (IMAC) with all four constructs.

Protein stabilization through phage display.

RNase S consists of two proteolytic fragments of RNase A, residues 1-20 (S20) and residues 21-124 (S pro). A 15-mer peptide (S15p) with high affinity for S pro was selected from a phage display library. Peptide residues that are buried in the structure of the wild type complex are conserved in S15p though there are several changes at other positions. Isothermal titration calorimetry studies show that the affinity of S15p is comparable to that of the wild type peptide at 25 degrees C. However, the magnitudes of DeltaH(o) and DeltaC(p) are lower for S15p, suggesting that the thermal stability of the complex is enhanced. In agreement with this prediction, at pH 6, the T(m) of the S15p complex was found to be 10 degrees C higher than that of the wild type complex. This suggests that for proteins where fragment complementation systems exist, phage display can be used to find mutations that increase protein thermal stability.

Determination of phage antibody affinities to antigen by a microbalance sensor system.

Over the past decade, phage display has maturated to be a frequently used method for the generation of monoclonal antibodies of human origin. The essential step of this method is the "biopanning" of phage carrying functional antibody fragments on their surface on an immobilized antigen. The screening of large combinatorial gene libraries with this method usually leads to a set of diverse clones specifically binding to the antigen that need to be characterized further. Beside its specificity, the key parameter to be determined is the affinity of the recombinant antibody fragment to its antigen. Here, we present a mass sensitive microsensor method that allows the estimation of antibody affinity directly from the phage supernatant. Binding of phage antibodies to the antigen immobilized on a quartz crystal microbalance (QCM) induced a mass dependent decrease in frequency. This principle was used to determine the apparent affinity of a single-chain (sc)Fv antibody against the RNA polymerase of Drosophila melanogaster presented on the surface of a filamentous phage (M13) from its association and dissociation rates. The apparent affinity obtained is in accordance with the affinity of the scFv fragment as determined by conventional equilibrium enzyme-linked immunosorbent assay (ELISA) and plasmon resonance methods.

Epitope structures recognised by antibodies against the major coat protein (g8p) of filamentous bacteriophage fd (Inoviridae).

To map the accessible surface of filamentous bacteriophage fd particles, the epitope structures of polyclonal rabbit serum and three mouse monoclonal antibodies raised against complete phage were analysed. Western blot analysis confirmed the major coat protein, gene VIII product (g8p or pVIII), to be the antigen. Overlapping peptides were synthesised by spot synthesis on cellulose membranes, covering the whole sequence of g8p. Each of the three tested monoclonal antibodies, B62-FE2, B62-GF3/G12 and B62-EA11, reacted with a core epitope covering ten amino acid residues at or near the amino terminus of g8p. The epitope recognised by B62-FE2 consists of the ten N-terminal amino acid residues of g8p. Extension of the amino terminus by various sequences did not inhibit binding, indicating that a terminal amino group is not essential for the interaction. Both B62-GF3/G12 and B62-EA11 recognise internal epitopes covering amino acid residues 3 to 12 of g8p. The epitopes of the polyclonal rabbit serum were also confined to the 12 N-terminal amino acid residues. The contribution of individual amino acid residues to the binding was analysed by a set of peptides containing individual amino acids exchanged by glycine. Accessible residues were Glu2, Asp4, Asp5, Pro6, Lys8, Phe11 and Asp12. The positions of the essential amino acid residues within the epitope are in accordance with a helical conformation of the amino-terminal region of g8p. Further, the results suggest new designs of phage display screening vectors to improve their performance in analysing non-linear epitopes.

Epitope mapping by phage display: random versus gene-fragment libraries.

We present a comparative study on epitope mapping of four monoclonal antibodies directed against four different antigens using alternative phage display techniques and peptide scanning: mAb215 reacts with the largest subunit of RNA polymerase II, mAbBp53-11 with the tumor suppressor protein p53, mAbGDO5 with the Hantaan virus glycoprotein G2 and mAbL13F3 with the Hantaan virus nucleocapsid protein. Epitopes were determined (i) by gene-fragment phage display libraries, constructed by DNaseI digested random gene fragments cloned into the 5' terminus of the pIII-gene of fd phage and (ii) by random peptide phage libraries displaying 6mer and 15mer peptides at the N-terminus of the pIII protein. Using the gene-fragment phage display libraries a single round of affinity selection resulted in the determination of the corresponding epitopes for all monoclonal antibodies tested. In contrast, biopanning of 6mer and 15mer random peptide libraries was only successful for two of the antibodies (mAbBp53-11 and mAbGDO5) after three or four rounds of selection. For the anti-p53 antibody we recovered the epitope from both the 6mer and 15mer library, for mAbGDO5 only the 6mer library displayed the epitope sequence. However, screening of the random peptide libraries with mAb215 and mAbL13F3 failed to yield immunopositive clones. Fine mapping of the epitopes by peptide scan revealed that the minimal epitopes recognized by mAbBp53-11 and mAbGDO5 consist of four and five amino acids, respectively, whereas mAb215 requires a minimal epitope of 11 amino acids for antigen recognition. In contrast, mAbL13F3 did not react with any of the synthesized 15mer peptides. The limits of the different methods of epitope mapping tested in this study are compared and the advantages of the gene-fragment phage display system are discussed.

Characterization of the epitope recognized by a monoclonal antibody directed against the largest subunit of Drosophila RNA polymerase II.

The epitope recognized by monoclonal antibody MAb215 generated previously against Drosophila melanogaster RNA polymerase II was mapped to amino acid residues 806-820 of the largest, 215 kDa, subunit located in a region conserved within the largest subunits of pro- and eukaryotic RNA polymerases. The affinities of MAb215 and of a recombinant single-chain Fv fragment (scFv215) were determined for binding to the enzyme as well as the fusion protein and synthetic peptides used for epitope mapping. In addition, amino acid residues of the epitope important for binding to MAb215 were identified using peptides carrying single amino acid substitutions. The epitope is not involved in the polymerization reaction or the DNA unwinding process since no inhibitory effects of the monoclonal antibody were observed in nonspecific in vitro transcription using denatured calf thymus DNA or double stranded oligo dC-tailed T7 DNA as template. In contrast, MAb215 inhibits accurate in vitro transcription from the Krüppel gene promoter and from the adenovirus-2 major late promoter. Preincubation of template DNA with the nuclear extract had no effects on inhibition supporting the notion that the epitope does not participate directly in the formation of preinitiation complexes. The same inhibitory effects were observed using scFv215. The results provide further evidence that recombinant antibody fragments produced in Escherichia coli possess the same specificity and similar affinity as their parental antibodies and demonstrate that scFv fragments are useful tools for analysis of transcriptional processes.