A Chlamydia trachomatis Strain With a 377-bp Deletion in the Cryptic Plasmid Causing False-Negative Nucleic Acid Amplification Tests

AFTER AN ALMOST STEADY increase for 10 years in reported Chlamydia trachomatis (Ct) genital infections in Sweden, there was a leveling out in 2005 to 2006. In some areas of the country, like the county of Halland (290,000 inhabitants), we noted a decrease in diagnosed infections of 25% in the beginning of 2006. In Halland we have used nucleic acid amplification tests (NAAT) to diagnose Ct infections since 1995, beginning with Abbott LCx (Abbott Laboratories, Abbott Park, IL), October 2002 to March 2006, Roche Amplicor PCR (Roche Molecular Systems, Branchburg, NJ), and then Abbott m200 real-time PCR, all with target areas on the cryptic plasmid. The unexplained decrease raised a suspicion of impairment of kit quality or a change in the target area in the microbe. By using ARTUS C. trachomatis (Hamburg, Germany) kit targeting the genomic ompA gene and not the cryptic plasmid, we found positive samples that had tested negative in Abbott's test. DNA sequencing of the area in the cryptic plasmid containing the target region for the Abbott m 2000 assay of 12 of these discrepant samples revealed a 377-bp deletion (GenBank accession no. EF 121757).1 The deletion involves the entire target area for the Abbott m2000 assay as well as most of the target area for the assays supplied by Roche. Data on the exact target regions for the Abbott and Roche assays were kindly supplied by the companies. The DNA sequencing of part of the cryptic plasmid was done in collaboration with Abbott Diagnostics, USA. A similar deletion in ORF1 has been described in an equine Chlamydophila pneumoniae N16, in 1990 by Wills et al.2 Pickett et al.3 suggested that the deletion in N16 was responsible for a decreased copy number of plasmids in the microbe. If this is the case in the described mutated Ct clone is not known. From mid-September 2006 through February 2007, we have used both Abbott's m2000 rtPCR and ARTUS C. trachomatis MOMP PCR on 5785 consecutive clinical samples sent to our laboratory. Six hundred twenty-one samples turned out positive for Ct, of which 19.5% (121 of 621) was detected only with the ARTUS kit. In Figure 1 the number of detected wild-type and variant-type Ct in Halland are shown by monthly intervals from January 2004 through February 2007. Since testing for both wild types and the mutant clone was performed the 5 months, October 2006 to February 2007, the corresponding number of diagnosed infections for those months during 2004 to 2005 and 2005 to 2006 are shown in Figure 2 for comparison. The decrease in wild-type infections and replacement and increase in infections by the mutant strain is clearly shown.Fig. 1: Number of diagnosed Chlamydia trachomatis infections by PCR in Halland County monthly from January 2004 through February 2007. The 5-months period, October to February, is highlighted, giving the numbers of wild-type infections and the mutant-strain infections.Fig. 2: Representative real-time PCR amplification curves from 4 samples positive for Chlamydia trachomatis with a cryptic plasmid carrying the 377-bp deletion (black diamonds) and from 2 samples without the deletion (grey triangles).To verify the presence of the 377-bp deletion in ARTUS MOMP-positive/Abbott m2000-negative samples, a mutation-specific, real-time PCR was designed. In the LightCycler-based method, FRET probes were designed to bind to sequence flanking the deletion to produce fluorescence signal only when binding to mutant sequence (primers: CTP200, TCC GGA TAG TGA ATT ATA GAG ACT; and CTP201, CCT AAA AGT GTT TTT TCT GGC; probes: CTP-LC with LCred-640 label, CGG ATC CCT TGT ACA ATC AAT TTA CCG; and CTP-FL with fluorescein label CTA GAG GAC TTA CCT CTT CCC CAG AAC A). Sample DNA was extracted using Abbott sp2000 as recommended by the manufacturer, and 5 μL of DNA extract was added to a 20-μL PCR reaction mixture. Samples were amplified and detected on the LightCycler 1.0 using LightCycler FastStart DNA master HybProbe reagents (Roche Diagnostics). The following temperature profile was used: 1 cycle of 95°C for 10 minutes for enzyme activation followed by 50 cycles at 95°C for 15 seconds, 50°C for 10 seconds, and 72°C for 15 seconds. Real-time data were collected at the end of the annealing segment (50°C); all temperature transition rates were set to maximum 20°C/s. Specific detection of mutant sequence over wild-type sequence was validated by testing 12 mutant samples verified with DNA sequencing and 20 samples positive for Ct in the ARTUS kit and in Abbott m2000 assay. Figure 2 demonstrates the clear-cut distinction between the wild types and the mutant strains in the assay. For clarity only a smaller number of representative samples has been included. The variant strain has been reported from all over Sweden in proportions from 10% to 65% of the total number of infected patients. Serotyped isolates have all been serovar E, the most common Ct serovar in Sweden (Unemo M et al. to be published). The only reported presence outside Sweden is in Norway where 2 cases, 1 of them a Swedish visitor, was detected recently.4 In Denmark, another neighboring country, no cases have been found in spite of testing 2000 patient samples (Skov-Jensen, Statens Serum Institut, Copenhagen, Denmark, personal communication). Maybe the strain is still expanding in other countries but has not yet reached detectable levels. Possibly there are factors in Sweden, not present in the neighboring countries, favoring the expansion of the clone. Some of these factors could be a high diagnostic coverage with more than 400,000 samples analyzed each year in a population of 9,000,000 in combination with the fact that Roche tests dominate the Swedish market. Only a few laboratories use the Becton Dickinson ProbeTec (Sparks, MD) assay targeting a different area of the plasmid and therefore detect the mutated clone. The Gen-Probe APTIMA CT (San Diego, CA) assay targeting rRNA should detect the clone but is not used in Sweden. Thus there is a possibility that the found mutated clone has been selected for in Sweden. Only symptomatic patients have been treated; no contact tracing have been performed, and screened patients asymptomatically infected with the variant have not been diagnosed. Our experience describes an expected pitfall in NAAT analysis: case Ct PCR. Dual target tests have an advantage and are now developed by both Abbott and Roche.