C. trachomatis infection is the most common sexually transmitted disease (STD) in the United States [¹]. Mounting evidence has indicated that it not only evokes nongonococcal urethritis (NGU), cervicitis, pelvic inflammatory disease (PID), salpingitis, orchitis, and epididymitis, but also increases risk of invasive cervical cancer [²,³] and gives rise to serious reproductive disorders such as infertility [⁴,⁵], miscarriage/premature birth/missed miscarriage [^6-9], and neonatal conjunctivitis [¹⁰].

A large number of methods have been established for screening and diagnosis of C. trachomatis infection [¹¹]. Nevertheless, few of these assays meet the requirements of outpatient diagnosis, especially in terms of sensitivity, specificity, time, and simplicity of operation. Currently, non-culture assays for C. trachomatis screening typically adopt immunochromatography-based methods. Technologies based on chromogenic reactions of specific microbial enzymes have been widely applied in bacterial identification systems and chromogenic media [^12-15]. However, no chromogenic assay for detecting C. trachomatis has been made available to date.

Our previous findings [¹⁶] suggested that C. trachomatis might have very high α-mannosidase activity. The purpose of the study was to establish a novel screening method for C. trachomatis infection without culture that would be rapid and convenient for use in outpatient clinics.

In our developed assay, only C. trachomatis samples tested was positive for α-D-mannosidase activity; but OD₅₁₂ of both other organisms and cell cultures which were not inoculated with C. trachomatis used in the study was all below 0.100, which fell into the range of negative results negative with 0.150 (OD₅₁₂) as the cut-off value, even if the reactions were allowed to proceed for 1 h at 37°C. Of the 553 clinical samples, 132 samples were positive with OD₅₁₂ ranging from 0.161 to 1.955, and 421 samples were negative with OD₅₁₂ ranging from 0.013 to 0.142. C. trachomatis detection results of 553 cases with culture, LCR and enzymatic method used an expanded gold standard as the reference standard were showed in Table 4. The enzymatic method was least reliable when prostate specimens were used. The sensitivity of the enzymatic method was 91.5% (95% confidence interval [CI], 85.9% to 97.1%), and the specificity of this method was 90.0% (95% CI, 87.3% to 92.7%) (Table 5). The sensitivity and specificity of the LCR were 94.7% (95% CI, 90.2% to 99.2%) and 100% (95% CI, 100.0% to 100.0%), respectively. However, in those specimens other than prostate fluid samples, the sensitivity and specificity of the enzymatic method were 91.8% (95% CI, 86.0% to 97.6%) and 98.3% (95% CI, 97.1% to 99.5%), in the meantime the sensitivity and specificity of the LCR were 95.3% (95% CI, 90.8% to 99.8%) and 100% (95% CI, 100% to 100%) (Table 6), respectively. There were no significant differences in performance between the enzymatic method and the expanded gold standard (P > 0.05).

The result of the chlamydial suspension quantified with 617 IFU/ml was light pink, and might be considered positive (OD₅₁₂ = 0.162). Meanwhile, the result of the chlamydial suspension was colourless if quantified with 126 IFU/ml, and might be considered negative (OD₅₁₂ = 0.098). Therefore, the LOD was 617 IFU/ml for C. trachomatis. In addition, the LOD was 10^-3 u/ml (OD₅₁₂ = 0.155) for α-D-mannosidase.

There are various well-known methods for detecting C. trachomatis, including cell culture-, immunology-, molecular biology-, and biochemistry-based methods. Cell culture is complicated to perform and requires experience to produce accurate results, and it also has more stringent requirements for the sampling swabs and transporting before inoculation [²¹,²²]. Therefore, cell culture is rarely used in clinics. Among the available immunological methods, serological tests for the C. trachomatis antibody have significant limitations [²³], but methods for C. trachomatis antigen detection (mainly referring to the lipopolysaccharide, LPS), especially immunochromatography-based methods, have been widely used due to their simplicity of operation. There are two ways to extract the LPS antigen for C. trachomatis immunochromatographic assays: heat extraction and acid extraction. Neither of these methods guarantees the full extraction of LPS as an intact antigen, which influences the sensitivity of this method. The biochemical detection of glycogen in C. trachomatis inclusions [²⁴] is greatly affected by Candida spp. that often exist in these specimens and are especially common during the female menstrual cycle and pregnancy. With the increasing glycogen in vaginal epithelial cells, this method may also cause false positives. The detection of C. trachomatis in the United States and Europe has mainly focused on molecular biology methods [^25-30]. Although these methods are both high sensitivity and high specificity, it can be challenging for molecular biology methods to meet the requirements of the actual application in clinical screening.

Enzymatic studies of C. trachomatis, especially for enzymes with diagnostic significance, have not been reported in the literature. Previous studies [³¹] and our research have shown that C. trachomatis secretes extracellular enzymes with high α-D-mannosidase activity. Although some organisms used in the study such as C. albicans have genes encoding α-1,2-mannosidase [³²], α-D-mannosidase activity was invisible by naked eyes with 6-chloro-3-indolyl-α-D-mannoside as substrate. This may be because the extracellular α-D-mannosidase from these organisms is much less or the enzyme activity is relatively low. Previous records [³¹,^33-35] on substrates of α-D-mannosidase mainly involved p-nitrophenol-α-D-mannoside and 4-methylumbelliferyl-α-D-mannoside. However, 6-chloro-3-indolyl-α-D-mannoside is a novel chromogenic substrate, and colour reaction of its chromogen is much more sensitive especially in the case of the presence of an azo reagent such as fast violet B salt.

Our results showed that clinical specimens such as urethral discharge, cervical secretions, and vaginal secretions did not interfere with the chromogenic detection of α-D-mannosidase activity to screen for C. trachomatis, although prostate massage liquid produced more false positive results. Some human sperm surface proteins possess α-D-mannosidase activity [³⁶], which may be the reason that prostate specimens produce less reliable results. Serotype D was only one of the most prevalent (11.1%), and no serovar L2 was found in China [³⁷]. Although the results of this study suggested that other serotypes, such as serotypes E, F, G, K, H, J, I, and Ba, at least most of them, might have α-D-mannosidase activity, there seems to be some limitation of tests on C. trachomatis cultures based on only one strain of serotype D. Further studies and more comprehensive clinical evaluations should be conducted due to little research on α-D-mannosidase activity of C. trachomatis. In addition, our studies did not evaluate C. pneumoniae or C. psittaci; the α-D-mannosidase activities of these species should be studied as well.

The present study demonstrated that there were no significant differences between the enzymatic method and the reference method when prostate specimens were excluded. Therefore, α-D-mannosidase activity may be a useful marker for C. trachomatis in urogenital tract specimens, with many advantages, such as its speed, ease of use, convenience, and need for no special equipment. Taken together, these data show that the enzymatic method has great potential as a clinical method for C. trachomatis screening.

The authors declare that they have no competing interests. We are applying for one patent (CN patent 102286608A) relating to the content of the manuscript, and the authors do not have any objection on the patent right of authorship and ownership.

ZYW conceived of the study and designed all the experiments and drafted the manuscript. ZYW, GYF, SMW and DCQ performed the experiments. ZYW and ZQW performed the statistical analysis. ZQW and JC provided valuable insight for designing the study and revising the manuscript. All authors read and approved the final manuscript.

The pre-publication history for this paper can be accessed here:

http://www.biomedcentral.com/1471-2334/13/36/prepub