The development of therapeutic recombinant antibodies involves accurate characterization of immunoglobulin variable light (VL) and heavy (VH) chains. However, it has been reported that the use of subgroup or isotype-specific primers for the amplification of monoclonal antibody (mAb) variable domains introduces heterogeneities within the variable domains, or amplifies aberrant productive Ig domains. To address these issues, we have combined the rapid amplification of cDNA ends PCR (RACE-PCR) for the full-length VL and VH amplification, with peptide mass fingerprinting of the corresponding Ig chain. Using this strategy, we amplified full-length cDNA chains of SAF34 and SAF32, two potential therapeutic mAbs against neurodegenerative diseases directed to the prion protein (PrP). We report an unambiguous correlation between hybridoma cDNA sequences and protein fingerprints of the variable domains of each mAb, indicating the discrimination between mutated, pseudo-genes and functional Ig genes. As a proof of principle for this dual strategy of full-length PCR amplification of variable domains and their characterization by MALDI-TOF, we show that the corresponding scFvs recognize the native PrP and retain full capacity to bind to human PrP, as does the parental mAb. This finding addresses the need for reliable light and heavy chain characterization, a key factor for humanization of mouse antibodies and for its use in passive immunotherapy applications.