Association of HLA-E Polymorphism with the Incidence of Severe Bacterial Infections in Sickle Cell Anemia.

While it is commonly admitted that the susceptibility to severe infections (septicemia, meningitis, or osteomyelitis) in Sickle Cell Anemia (SCA) patients is a major cause of mortality and morbidity, these complications remains poorly explained. However, several lines of evidence highlight the involvement of an immunogenetic modulation, including both innate and adaptive immune responses. HLA-E belongs to the HLA-class I non-classical family and is ubiquitously expressed. HLA-E molecules bind either to self-peptides from the leader sequences of various HLA-class I molecules and thus modulate NK (Natural killer) cell activation/inhibition, or to the microbial derived-peptides from human viruses or bacteria to induce T cell responses. Only two functional alleles, HLA-E*0101 and HLA-E*0103, differing by a single amino acid substitution in the α2 heavy chain domain and by their cell surface expression level, have been reported so far. In order to evaluate whether the HLA-E polymorphism could influence the incidence of severe bacterial infections in SCA, we analyzed a cohort of 99 SCA patients living in Paris, but for the majority, originating from black Africa. All patients were older than 5 years of age at inclusion. Among the cohort, 51 presented at least one severe bacterial infection and 48 did not disclose any major infectious episode in their clinical history. Statistical analysis showed that the incidence of severe bacterial infections are high when the patient’s genotype was HLA-E*0101/E*0101 (47% in the first group versus 21% in the second group; χ2 = 7.54 p = 0.006, pc = 0.01, OR: 3.28, [95%CI = 1.28–9.07]), supporting a negative effect of this genotype. The genetic association herein found between HLA-E polymorphism and severe bacterial infections is of relevance, given the emerging evidence for the involvement of HLA-E molecules in pathogen-derived peptide presentation to human CD8+ T cells and hence in host response to pathogens. Furthermore, these data are in agreement with our previous findings in bone marrow transplantation settings, were the homozygous state for HLA-E*0101 allele is also a risk factor for early severe bacterial infections. Both situations by constitutively lowering the overall threshold of resistance to infection, might unmask anti infectious genetic factors normally silent.