The causative agent of anthrax, Bacillus anthracis, evades the host immune response and establishes infection through the production of binary exotoxins composed of Protective Antigen (PA) and one of two subunits, lethal factor (LF) or edema factor (EF). The majority of vaccination strategies have focused upon the antibody response to the PA subunit. We have used a panel of humanised HLA class II transgenic mouse strains to define HLA-DR-restricted and HLA-DQ-restricted CD4+ T cell responses to the immunodominant epitopes of PA. This was correlated with the binding affinities of epitopes to HLA class II molecules, as well as the responses of two human cohorts: individuals vaccinated with the Anthrax Vaccine Precipitated (AVP) vaccine (which contains PA and trace amounts of LF), and patients recovering from cutaneous anthrax infections. The infected and vaccinated cohorts expressing different HLA types were found to make CD4+ T cell responses to multiple and diverse epitopes of PA. The effects of HLA polymorphism were explored using transgenic mouse lines, which demonstrated differential susceptibility, indicating that HLA-DR1 and HLA-DQ8 alleles conferred protective immunity relative to HLA-DR15, HLA-DR4 and HLA-DQ6. The HLA transgenics enabled a reductionist approach, allowing us to better define CD4+ T cell epitopes. Appreciating the effects of HLA polymorphism on the variability of responses to natural infection and vaccination will be vital in planning protective strategies against anthrax.Author SummaryThe bacterium responsible for causing the disease anthrax, Bacillus anthracis, produces a binary toxin composed of Protective Antigen (PA) and either Lethal Factor (LF) or Edema Factor (EF). Previous vaccination strategies have focused upon the antibody response to the PA subunit. However, within the field of bacterial immunity, there is a growing appreciation of the importance of the adaptive immune response, specifically led by CD4+ T cells. We identified long-term CD4+ T cell responses to PA epitopes following cutaneous human anthrax infection and vaccination, indicating that this toxin component is a principle B. anthracis antigen. To characterise the impact of polymorphism in HLA class II alleles at DR and DQ loci, we used transgenic mice to map the immunodominant epitopes from PA. This was correlated with survival in the transgenic lines following live anthrax challenge. We were able to demonstrate the differential impact of HLA class II alleles upon the CD4+ T cell immunodominant epitopes which shaped the immune hierarchy and therefore susceptibility to anthrax infection.