Engineering anti-infective antibodies

In the past 15-20 years, advances in antibody engineering have facilitated the generation and isolation of monoclonal antibodies (mAbs) to a wide array of antigens. Consequently, mAbs have become essential therapeutic tools and currently dominate the global protein therapeutics market. The engineering of anti-infective antibodies, however, has proven quite a challenge, despite the fact that antibodies were naturally evolved to fight infections. The identification of suitable antigens, the mode of administration and the high cost associated with the production of antibody therapeutics are some of the major hurdles for the progress of anti-infective antibodies. This dissertation addresses issues concerning the development of anti-infective antibodies against two different pathogens: SARS coronavirus (CoV) and two pathogenic species of Burkholderia bacteria. To investigate the role of affinity in viral neutralization and evolution of escape mutants, we first sought to isolate an antibody with high affinity towards the receptor binding domain (RBD) of SARS-CoV. Following high-throughput screening of a library of random mutants via the APEx display system, we isolated antibodies with affinities in the range of 0.8 nM - 0.1 nM. The affinity was further improved by additional mutagenesis and DNA shuffling, and a high affinity variant (45pM) with ~300-fold improvement over the parental antibody was isolated. Evaluation of these antibodies in an in vitro assay demonstrated that neutralization of wild-type Urbani strain of SARS-CoV correlates well with the affinity of the antibody, with higher affinity leading to greater neutralization. Moreover, the antibody exhibiting the highest affinity could neutralize SARS-CoV escape mutants that evaded neutralization by both parental and lower affinity antibodies. Another important aspect for the development of anti-infective antibodies concerns the identification of suitable antigen targets to be used in the isolation of antibodies. In an effort to develop a high-throughput screening method for the isolation of antibodies to a wide array of antigens, we used a synthetic antibody (Fab) library constructed by a minimalist approach and displayed on the surface of filamentous bacteriophage. The library was screened against antigens from Burkholderia pseudomallei and Burkholderia mallei. After only three rounds of selection and enrichment against five different antigens, we obtained Fabs specific to four of the antigens as confirmed by ELISA. These results not only demonstrate the use of a synthetic antibody library for the isolation of antibodies against infectious pathogens, but also its feasibility, and potential applicability as a high-throughput screen for a variety of antigens.