The carbohydrate surface of Mycobacterium tuberculosis: Antigenicity and antibody immunity

Abstract

MTB arabinomannan (AM) and glucan (GC) can be found on the surface of Mycobacterium tuberculosis (MTB), and are found in vivo during experimental murine tuberculosis. I extensively studied AM and GC antigens in this thesis. An AM antigen was found to be dynamically expressed on the surface of in vitro grown MTB Erdman, but not on the CDC-1551 MTB strain. The GC antigen was found on a variety of mycobacteria, and was localized within a capsular-type of material on the outside of MTB, using microscopy. The production of the AM antigen was quantitated during different stages of disease, and the amount of AM antigen detected was not always proportional to the number of MTB CFU in the infected organ. Histological examination indicated that the AM antigen diffused from bacterial clusters grown in vivo. Serological responses to MTB revealed antibodies to AM and GC, indicating that these polysaccharides are made during infection.;Unique safety and experimental protocols were developed for the chemical fixation of MTB and for nose-only aerosol delivery of MTB to mice. Other findings relevant to the genetic characterization and expression of MTB carbohydrate antigens, in addition to an antibody selection technique for isolating mutant MTB bacteria, are included as an appendix to this thesis.;In additional studies, experimental parameters that could influence antibody immune efficacy in the setting of tuberculosis were varied to determine if and when a mAb recognizing MTB AM could influence murine survival. The 9d8(2a) antibody could enhance infection in an antibody and infectious dosage dependent manner. The 9d8 (A) (IgA isotype) antibody was administered intranasally, and had only minimal effects on MTB CFU during infection. Immunization of mice with AM or GC conjugate vaccines did not prolong survival in mice challenged with MTB. However, mice immunized with an AM conjugate vaccine had slightly reduced organ CFU when challenged with BCG or MTB seven days after infection. These results are in contrast to published results that used a 9d8 (G3) antibody to show enhanced murine survival, indicating that antibody immunity to tuberculosis is not always a protective response and is dependent upon experimental parameters.