Introduction

The chlamydial ompA gene encodes the major outer membrane protein (MOMP) [1]. The importance of ompA genovar on the severity of trachoma phenotypes is unclear, but the MOMP may influence the host immune response, and hence the clinical signs of trachoma. In the present study, we genotyped chlamydia from a series of children with trachoma in order to assess this hypothesis.

Methods

This is a secondary analysis from a cluster-randomized clinical trial conducted in Ethiopia starting in March 2003 [1]. As part of the trial, we examined and collected swabs of the upper right tarsal conjunctival from all children aged 1–5 years in each of 40 Ethiopian communities before any intervention, and then at 2 and 6 months after a single mass azithromycin distribution [2, 3]. We assessed for the presence or absence of follicular trachoma (TF) and intense inflammatory trachoma (TI) using the World Health Organization simplified grading system (Fig. 1) and processed conjunctival swabs for chlamydial DNA using Roche AMPLICOR [4].

Fig. 1
figure 1

Superior tarsal conjunctival photographs demonstrating the World Health Organization simplified trachoma grading system. The top panel shows follicular trachomatous inflammation (TF), characterized by ≥5 follicles measuring ≥0.5 mm in diameter in the upper tarsal conjunctiva. The middle panel shows the intense trachomatous inflammation (TI), characterized by inflammatory thickening that obscures more than half of the deep underling superior tarsal conjunctival blood vessels. The bottom panel shows both TF and TI. Photographs do not correspond to the individuals in the present study; they were taken at a later study visit and are shown for illustrative purposes only.

Full size image

For the present study, we chose a random sample of chlamydia-positive swabs from 21 communities at the pre-treatment, 2-month, and 6-month visits, with swabs matched by community. From each swab, we extracted DNA and amplified a region of the ompA gene with previously published primers [5]. The ompA PCR products were sequenced in both directions, edited, aligned to reference sequences (A/HAR-13; NC_007429 for A genovar and B/Tunis-864; DQ064280 for B genovar), and then assigned a genotype. Laboratory personnel were masked to the conjunctival examination findings. We used mixed effects ordinal logistic regression to determine whether genovar was a significant predictor of TF and/or TI, with age and sex as covariates and community and child as nested random effects. We coded the response variable in the following ascending order: neither TF nor TI (0), TF (1), TI (2), TF and TI (3).

Results

We identified 359 chlamydia-positive swabs for this ancillary study. Of these, we could extract DNA and determine the ompA sequence from 351 (97.8%) swabs from 274 distinct children. Within genovars, we identified 11 different A genotypes (N = 258) and 4 different B genotypes (N = 93). Type B genovars were more likely than type A genovars to have TF and TI (OR 1.74, 95% CI 1.03–2.94, P = 0.04, Table 1).

Table 1 Distribution of trachoma grades according to ompA status
Full size table

Discussion

We found that the chlamydial B genovar was more likely than the A genovar to have more intense trachomatous inflammation. Few previous studies have commented on the relationship between ompA type and clinical trachoma, although a Gambian study with very few B genovars found differing amounts of clinical trachoma in different genotypes of the A genovar [6]. An association between ompA type and host phenotype is plausible since many believe the MOMP is an important target of immunity and ompA variants could therefore elicit differential host immune responses. Limitations of this study include its relatively small-sample size and its unclear generalizability to areas with less prevalent trachoma. Nonetheless, we provide evidence suggesting that chlamydial MOMPs may play a role in the host inflammatory response in trachoma.