Machine Learning-Guided Prediction of Antigen-Reactive In Silico Clonotypes Based on Changes in Clonal Abundance through Bio-Panning.

c-Met is a promising target in cancer therapy for its intrinsic oncogenic properties. However, there are currently no c-Met-specific inhibitors available in the clinic. Antibodies blocking the interaction with its only known ligand, hepatocyte growth factor, and/or inducing receptor internalization have been clinically tested. To explore other therapeutic antibody mechanisms like Fc-mediated effector function, bispecific T cell engagement, and chimeric antigen T cell receptors, a diverse panel of antibodies is essential. We prepared a chicken immune scFv library, performed four rounds of bio-panning, obtained 641 clones using a high-throughput clonal retrieval system (TrueRepertoireTM, TR), and found 149 antigen-reactive scFv clones. We also prepared phagemid DNA before the start of bio-panning (round 0) and, after each round of bio-panning (round 1-4), performed next-generation sequencing of these five sets of phagemid DNA, and identified 860,207 HCDR3 clonotypes and 443,292 LCDR3 clonotypes along with their clonal abundance data. We then established a TR data set consisting of antigen reactivity for scFv clones found in TR analysis and the clonal abundance of their HCDR3 and LCDR3 clonotypes in five sets of phagemid DNA. Using the TR data set, a random forest machine learning algorithm was trained to predict the binding properties of in silico HCDR3 and LCDR3 clonotypes. Subsequently, we synthesized 40 HCDR3 and 40 LCDR3 clonotypes predicted to be antigen reactive (AR) and constructed a phage-displayed scFv library called the AR library. In parallel, we also prepared an antigen non-reactive (NR) library using 10 HCDR3 and 10 LCDR3 clonotypes predicted to be NR. After a single round of bio-panning, we screened 96 randomly-selected phage clones from the AR library and found out 14 AR scFv clones consisting of 5 HCDR3 and 11 LCDR3 AR clonotypes. We also screened 96 randomly-selected phage clones from the NR library, but did not identify any AR clones. In summary, machine learning algorithms can provide a method for identifying AR antibodies, which allows for the characterization of diverse antibody libraries inaccessible by traditional methods.

A point mutation in the heavy chain complementarity-determining region 3 (HCDR3) significantly enhances the specificity of an anti-ROS1 antibody.

The proto-oncogene tyrosine kinase ROS1 plays a key role in carcinogenesis through gene rearrangement to form a fusion protein with other genes, in which the C-terminal intracellular region of ROS1 participates. The possibility of wild type ROS1 overexpression through epigenetic regulation has been proposed. Here, we generated an antibody, 3B20, reactive to the N-terminal region of ROS1 to use it for the detection of wild type ROS1 in cancerous tissues. Using immunoblot and immunoprecipitation analyses, we found that 3B20 also reacted with heat shock proteins (Hsp)70s. Using homology searching, ROS1 and Hsp70s were found to share an identical amino acid sequence: DLGT. Using alanine mutagenesis of ROS1, the epitope was found to harbor this sequence. To modify the idiotope with the aim of selecting more specific antibodies, we introduced random mutations into the heavy chain complementarity-determining region 3 and successfully generated an antibody clone, 3B20-G1K, with a point mutation that only reacted with ROS1 in enzyme-linked immunosorbent assays, and in immunoblot and immunoprecipitation analysis. In immunohistochemical analysis using 3B20-G1K, ROS1 was found to be absent in normal lung tissues and was overexpressed in a case of lung adenocarcinoma.

Screening of Pro-Asp Sequences Exposed on Bacteriophage M13 as an Ideal Anchor for Gold Nanocubes.

Bacteriophages are thought to be ideal vehicles for linking antibodies to nanoparticles. Here, we define the sequence of peptides exposed as a fusion protein on M13 bacteriophages to yield optimal binding of gold nanocubes and efficient bacteriophage amplification. We generated five helper bacteriophage libraries using AE(X)2DP, AE(X)3DP, AE(X)4DP, AE(X)5DP, and AE(X)6DP as the exposed portion of pVIII, in which X was a randomized amino acid residue encoded by the nucleotide sequence NNK. Efficient phage amplification was achievable only in the AE(X)2DP, AE(X)3DP, and AE(X)4DP libraries. Through biopanning with gold nanocubes, we enriched the phage clones and selected the clone with the highest fold change after enrichment. This clone displayed Pro-Asp on the surface of the bacteriophage and had amplification yields similar to those of the wild-type helper bacteriophage (VCSM13). The clone displayed even binding of gold nanocubes along its length and minimal aggregation after binding. We conclude that, for efficient amplification, the exposed pVIII amino acid length should be limited to six residues and Ala-Glu-Pro-Asp-Asp-Pro (AEPDDP) is the ideal fusion protein sequence for guaranteeing the optimal formation of a complex with gold nanocubes.

Virus templated gold nanocube chain for SERS nanoprobe.

A M13 virus based SERS nanoprobe is presented. Gold nanocubes closely aligned into chains along the length of the virus intensify Raman signals of various reporter molecules serving as specific labels. An antibody is expressed at one end to detect the analyte. This new SERS nanoprobe holds promise for infinitesimal and multiplexed detection of any antigen.

Bispecific Her2 × cotinine antibody in combination with cotinine-(histidine)2-iodine for the pre-targeting of Her2-positive breast cancer xenografts.

PURPOSE: Cotinine has optimal characteristics as a hapten for pre-targeted radioimmunotherapy (PRIT). This study was performed to evaluate the applicability of cotinine/anti-cotinine antibody to PRIT. METHODS: We developed and prepared a tandem, single-chain, variable fragment Fc fusion protein [tandem single-chain variable fragment (scFv) Fc fusion protein] that is reactive to both human epidermal growth factor receptor 2 (Her2) and cotinine. Its simultaneous reactivity to Her2 and cotinine was tested in an enzyme-linked immunosorbent assay (ELISA) and two radioimmunoassays (RIA) employing Her2-coated RIA tubes and a Her2-overexpressing cell line. For in vivo imaging, mice bearing Her2-positive tumors were injected with a mixture of tandem scFv Fc fusion and (125)I-cotinine-conjugated histidine dipeptide ((125)I-cotinine peptide). After a delay, (125)I-cotinine peptide was injected again. RESULTS: ELISA and RIA results showed that tandem scFv Fc fusion protein successfully bound to both Her2 and cotinine. In single-photon emission computed tomography (SPECT), the complex of tandem scFv Fc fusion protein and (125)I-cotinine peptide was localized to Her2-positive tumor xenografts in mice 4 h after the first injection. Enhanced radioactivity at the site of the Her2-positive tumor lesion was monitored 1 h after the second injection. CONCLUSIONS: With these findings, we conclude that the tandem scFv Fc fusion protein and cotinine hapten system have the potential to be applied in PRIT.

Application of bispecific antibody against antigen and hapten for immunodetection and immunopurification.

We present a bispecific antibody that recognizes an antigen and a hapten and can be applied to various biological assays, including immunoblotting and immunoprecipitation. In immunoblot analysis of serum, an anti-C5 × anti-cotinine bispecific tandem single-chain variable fragment (scFv)-Fc fusion protein and cotinine-conjugated horseradish peroxidase (HRP) generated a clean signal without the high background that was observed in a parallel experiment using HRP-conjugated goat anti-rabbit immunoglobulin G (Fc-specific) antibody. In immunoprecipitation analysis of serum, use of the bispecific tandem scFv-Fc fusion protein and cotinine-crosslinked magnetic beads significantly reduced the amount of protein contaminants compared with a parallel experiment done with protein A agarose beads. In subsequent immunoblot analysis, use of cotinine-HRP as the secondary probe instead of HRP-conjugated goat anti-rabbit IgG (Fc-specific) antibody successfully eliminated the band corresponding to the bispecific tandem scFv-Fc fusion protein.