Polyclonal Antibody Production for Membrane Proteins via Genetic Immunization.

Membrane proteins are the molecular interface between host and pathogen, yet these key proteins provide unique challenges for structural elucidation1 and for their use in developing not only therapeutics2 but also diagnostics and vaccines. Target-specific monoclonal antibodies have allowed the determination of novel membrane protein structures during electron cryomicroscopy by increasing the size of the target3, and during crystallography4 by stabilizing unique protein conformations5,6,7,8, providing crystal lattice contacts9,10,11, and allowing structure solution via molecular replacement10,11,12. Monoclonal antibodies that recognize specific conformations of membrane-embedded signal pathway proteins allow the exciting development of novel therapeutics13. Towards antibody production for membrane proteins, there is often a limitation in the availability of highly-purified or natively-folded target antigen. Therefore, we explored the genetic immunization approach14 in order to generate antibodies that target membrane proteins. Surprisingly, the efficiency of genetic immunization as applied to membrane proteins is unknown, since application of this method has been described only or for collections of soluble proteins15 or for individual membrane protein targets16,17,18. For these individual membrane proteins, the reported operational serum dilutions of ≤1:200 for the human thyrotropin and neurokinin-1 GPCRs16,17 and for human nephrin18 suggest room for improvement. The biolistic approach, using only genes as the source of antigen, has generated monoclonal antibodies that recognize native epitopes of membrane proteins17,18, including modifications such as glycosylation18. Additional groups have used genetic immunization alone to generate antibodies with therapeutic potential, albeit using proprietary methods19,20. Here we describe an efficient approach that yielded antibodies against the majority of 17 membrane proteins from Biosafety Level 3 pathogens. The SCHU S4 isolate of Francisella tularensis is one of the most pathogenic bacteria known due to its capacity for fatal infection from as few as ten cells21. F. tularensis causes the disease tularemia and is a model intracellular bacterial pathogen given its capacity to evade the immune response and to infect numerous cell types22. The arthropod-borne African swine fever virus (ASFV) causes an untreatable, highly-lethal hemorrhagic porcine disease that is an economic threat in Africa and eastern Europe23. The endemic existence of both these pathogens throughout numerous environmental sources makes eradication implausible24,25. Investigations with endogenous protein from these organisms are constrained by biosafety requirements and select agent status. To support studies of membrane proteins that are important in pathogenesis, we developed DNA-based approaches to generate and characterize antibodies against a set of membrane proteins (Supplementary Table 1) that were targeted for structural studies as part of the U.S. National Institutes of Health’s Protein Structure Initiative (PSI:Biology). Many of these targets are expected to provide novel membrane protein structures as they lack obvious sequence homologs outside of the Francisella genus and the Asfarviridae virus family.