Article Text


HLA and trachoma
  1. Department of Clinical Sciences
  2. London School of Hygiene and Tropical Medicine

    Statistics from

    HLA (human leucocyte antigen) molecules, which were initially studied because of their importance in organ transplant rejection, have been found to play a fundamental role in immunity, with the physiological role of presenting peptide antigens to T lymphocytes. HLA class I molecules present peptides to cytotoxic (CD8+) T lymphocytes, which are important in defence against viruses and other intracellular pathogens, whereas HLA class II molecules present peptides to CD4+ T cells, which are important in generating antibody and cellular responses mainly to antigens outside the cell. HLA molecules are thus central to immune defence against infecting micro-organisms and to the development of autoimmunity. High resolution characterisation of HLA alleles by DNA methods has recently become feasible.1

    Trachoma is the leading infectious cause of blindness. There is evidence that the clinical features both of active inflammatory disease and of the cicatrising sequelae are immunopathologically mediated. In endemic areas, familial clustering of active trachoma cases occurs, and the development of blinding sequelae is found only in a minority of subjects, raising the possibility of genetic susceptibility. In this issue of the BJO (p 431), White et al report an increased prevalence of DR16 and a decreased prevalence of DR53 class II haplotypes in Omani cases of trachomatous corneal opacity compared with a control population of students, blood and organ donors. In the Gambia, we found an increased prevalence of the class I subtype HLA A*6802 in subjects with trachomatous scarring compared with age and sex matched controls.2 Are these results compatible?

    Several issues need to be borne in mind when comparing these two studies. The distribution of HLA antigens differs between the Gambia and Oman; DR2 and its DR16 subtype are common in Oman but rare in the Gambia. The two studies did not use the same criteria for cases, or for the selection of controls. Controls should be exposed to the risk of getting the disease, and age and sex are important determinants of trachomatous sequelae. Different approaches were taken in the application of high resolution DNA typing methods to serologically defined antigens so that the subtype association found in the Gambia was not specifically examined in the Omani study. Finally the two studies used different approaches to the problem of multiple comparisons. The HLA antigens are the most variable human alleles known—for example, there are over 60 known variants of the HLA-B antigens with more being described all the time. Thus, a large number of comparisons between cases and controls is required, and there is a risk that some will appear significant as a result of chance alone. White et al elected to use the Bonferroni correction procedure, multiplying the p value by the number of tests carried out, but this increases the converse risk that important associations involving modest increases in risk may be wrongly attributed to chance. In our view, all the associations they report in Table 1 merit examination in other environments.

    What mechanisms might underlie HLA trachoma associations? Whiteet al discuss the possibilities that molecular mimicry of self antigens may play a role in pathogenesis or that HLA antigens may be receptors for chlamydia. A number of other explanations are also feasible: such associations could represent ‘holes’ in the T cell repertoire, differences in antigen recognition, signalling, or activation of suppressor T cells. The HLA genes are situated in close proximity to some 200 other genes within the MHC complex on the short arm of chromosome 6, and linkage disequilibrium between HLA haplotype and a true susceptibility allele elsewhere in the MHC might also explain putative associations. We have recently shown that polymorphism in the TNF-α promoter region, which is in the MHC between the class I and class II HLA loci is also associated with trachomatous scarring, but independently of HLA A*6802 rather than in linkage disequilibrium with it.3

    Will all this help Omani and Gambian trachoma patients? Recently, single dose azithromycin has been found to be effective treatment for active trachoma,4 and a community education programme targeted at face washing was found to reduce the prevalence of active trachoma in Kongwa, Tanzania.5 The high cost of azithromycin and the intense effort required for modest and poorly sustainable gains in the Kongwa study indicate that other interventions are likely to be needed. The identification of susceptibility or resistance genes for human ocular chlamydial infection and its sequelae may help identify those at risk, and might also identify critical mechanisms in the infection, scarring, and blinding processes either as targets for intervention or as considerations in vaccine design. But patients with trachomatous trichiasis and entropion are likely to need eyelid surgery for a while longer.


    View Abstract

    Request permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Linked Articles