Efficient generation of monoclonal antibodies from single human B cells by single cell RT-PCR and expression vector cloning

Abstract

We have developed an efficient strategy that combines immunoglobulin (Ig) gene repertoire analysis and Ig reactivity profiling at the single cell level. Based on surface marker expression individual cells at different stages of human B cell development are isolated by fluorescence-activated cell sorting. For each cell Ig heavy and corresponding Ig light chain gene transcripts are amplified by nested RT-PCR and cloned into eukaryotic expression vectors to produce monoclonal human antibodies of the same specificity in vitro. All reactions are performed in 96-well plates and allow cloning of large numbers of Ig genes. The recombinant antibodies are tested for reactivity with diverse self- and non-self antigens and the reactivity profile can be directly linked to the complete Ig heavy and Ig light chain gene sequence information that is obtained as part of the cloning strategy. In summary, our method to clone and express human monoclonal antibodies is unbiased, highly efficient, requires only small cell numbers and the recombinant antibodies allow direct conclusions on the frequency of specific human B cells in a diverse repertoire.

Figure 1

Strategy to clone and express human monoclonal antibodies. IgH and IgL chain genes were amplified by nested RT-PCR from single cell cDNA generated by amplification with random hexamers before ligation into eukaryotic expression vectors. 1^st PCRs were performed with forward primer mixes specific for the leader region and reverse primers specific for the respective IgH, Igκ or Igλ constant region. 2^nd PCRs were performed with forward primer mixes specific for FWR1 and respective nested reverse primers specific for the IgH, Igκ and Igλ J genes or constant regions. If 2^nd PCR primers contained restriction sites, the products were sequenced and directly cloned. Alternatively, to avoid the introduction of aa exchanges by the use of primer mixes and for the cloning of full-length Igκ genes, 2^nd PCR products were sequenced to determine the respective V and J gene. After that V gene and J gene specific primers containing restriction sites were used in another round of amplification using the 1^st PCR product as template. All PCR products were sequenced before and after expression vector cloning. For recombinant antibody production, plasmids containing inserts with 100% identity to the original PCR product were co-transfected into HEK 293 cells and antibodies were purified from supernatants after culture.

Figure 2

Reversion of somatic mutations. (A) PCR strategy to revert mutated Ig genes into their germline counterparts. Asterics indicate somatic mutations. PCR 1 amplifies a germline VH gene corresponding to the VH in the mutated clone with gene specific primers. Primers used in PCR 2 revert somatic mutations in the mutated clone. Homology of the PCR 2 forward primer to the reverse primer used in PCR 1 is indicated. The PCR 2 reverse primer is JH-specific and contains the SalI restriction site. PCR products 1 and 2 are fused via the homologous region (indicated) in a subsequent overlap PCR using the same 5′AgeI VH specific forward primer as in PCR 1 and the 3′SalI JH specific reverse primer used in PCR 2 to generate the complete germline VDJ sequence. IgL chain somatic mutations are reverted following the same principle. Overlap PCR products are cloned into the respective expression vectors. (B) Partial sequence of a representative mutated IgH gene (top) starting in FWR3 and including the CDR3 region and the conserved J_H tryptophan (W) – glycine (G) motif (FWR4). The corresponding unmutated germline configuration after successful reversion of somatic mutations as indicated on the bottom. Dots represent nucleotide identity. Dashes are non-identical nucleotides. Nucleotide exchanges and aa exchanges are indicated in bold. Germline nucleotide sequences of full-length VH3-23, D3-3 and JH4 genes are indicated in italics.

Figure 3

(A) Representative gel picture showing RT-PCR products of Igμ (450 bp), Igκ (510 bp) and Igλ (405 bp) V genes and beta-actin (302 bp) amplified from single mature naive human B cells. (B) Representative SDS gel picture of purified recombinant monoclonal human antibodies (lane #1–3) under non-reducing (left) and reducing conditions (right). For reduction, 5% β-mercaptoethanol was added to the denaturating sample buffer and all samples were incubated for 10 min at 70°C. 15 μg of proteins were subjected to a 4–12% BisTris polyacrylamide gel (BioRad). Protein bands were visualized by Coomassie blue staining. C: Control human IgG1κ antibody (Sigma). M: Molecular weight marker (BioRad).

Figure 4

Detection of self-reactive and polyreactive antibodies. (A) Typical HEp-2 cell reactivity profiles of IgG memory B cell antibodies (VB13, VB56, VB72, VB108, VB148) of healthy donor VB as determined by ELISA and immunofluorescence assay (Tiller et al., 2007). High-positive control antibody ED38 (dotted line; Meffre et al., 2004) and low ANA control serum (red line; INOVA) are shown for comparison. (B) Reactivity profiles of the high-positive control antibody ED38 (dotted line; Meffre et al., 2004), low-positive control antibody eiJB40 (red line; Wardemann et al., 2003) and negative control antibody mGO53 (green line; Wardemann et al., 2003) as tested by ELISA with dsDNA, ssDNA, LPS and insulin.