A general aim was to combine molecular typing methods with clinical background information to increase epidemiological knowledge about Chlamydia trachomatis infections. An outbreak of Lymfogranuloma venereum (LGV), caused by a more invasive variant of C. trachomatis, was reported from the Netherlands in 2003 among men who have sex with men (MSM). All Chlamydia positive specimens from a venereal disease clinic for MSM in Stockholm during one year were genotyped. No spread of LGV was found, apart from three symptomatic cases. The same ompA genotypes were found among MSM in Melbourne, but the genotype distribution was different compared to findings among the heterosexual population in Sweden. The standard method for genotyping of Chlamydia is ompA-sequencing, but it has low resolution because one genotype predominates. A multilocus sequence typing (MLST) system based on five targets was developed. In a sample of 47 specimens, 32 variants were found with MLST, but only 12 variants with ompA-sequencing. The polymorphisms in the hctB gene, one MLST target, are caused by an element of 108 bp that is present in two to four repetitions and in different variants. Although the DNA-binding function of Hc2 that is encoded by hctB has been studied, our findings of a considerable size variation show that new studies are needed. In 2006, specimens with a 377 bp deletion in the cryptic plasmid covering the target region for diagnostic test systems from Abbott and Roche were discovered in Sweden. Applying MLST to these specimens indicated that there was a single clone, denoted nvCT. The proportion of nvCT in all detected Chlamydia cases was higher (20% to 65%) in counties using Abbott/Roche compared to counties using the BectonDickinson test system (7% to 20%). The proportions of nvCT converge in counties with high or low levels when detection systems were adjusted to detect nvCT.

Chlamydia trachomatis is an obligate intracellular bacterium of major human health concern, causing urogential chlamydia infections, lymphogranuloma venereum (LGV) and trachoma. Chlamydia is one of the most common sexually transmitted infections worldwide and can cause infertility. In the first four papers described herein we used a high resolution multilocus sequence typing (MLST) system to investigate the epidemiology of C. trachomatis, and showed that MLST is superior to conventional ompA genotyping with respect to resolution. In the fifth paper we simplified the methodology by developing and validating a multilocus typing (MLT) DNA microarray based on the MLST system. In more detail, MLST analysis of consecutive specimens from 2006 in Örebro County in Sweden, and comparison to specimens from 1999-2000, showed that the new variant C. trachomatis (nvCT) is monoclonal and likely has appeared in recent years. MLST analysis of LGV specimens from men who have sex with men (MSM) showed that the increase of LGV in Europe in the last decade indeed was a clonal outbreak, contrary to the USA where LGV might have been present all along. In the third paper, clinical symptoms could not be correlated with the MLST genotypes, suggesting, together with the combined results of all previous studies, that bacterial factors, if important, need to be understood in the context of host factors. MLST analysis of specimens from a high incidence C. trachomatis area in North Norway revealed interesting epidemiological details concerning unusual genetic variants, the nvCT and MSM, but found no significant difference in genetic diversity compared to two other geographic areas in Norway. Lastly, we developed a MLT array that provides high resolution while being rapid and cost-effective, which makes it an interesting alternative for C. trachomatis genotyping. In conclusion, the MLST system and the MLT array have proven to be useful tools and should now be applied in further investigations to improve our understanding of C. trachomatis epidemiology.