Development and application of an in-house reverse hybridization method for Chlamydia trachomatis genotyping


  • H. Gharsallah,

    1. Department of Microbiology and research laboratory “Microorganismes et Pathologie Humaine”, Habib Bourguiba university hospital, Medical school of Sfax, University of Sfax, Sfax, Tunisia
    Search for more papers by this author
  • O. Frikha-Gargouri,

    1. Department of Microbiology and research laboratory “Microorganismes et Pathologie Humaine”, Habib Bourguiba university hospital, Medical school of Sfax, University of Sfax, Sfax, Tunisia
    2. Biopesticides Team (LPAP), Centre of Biotechnology of Sfax, University of Sfax, Sfax, Tunisia
    Search for more papers by this author
  • F. Besbes,

    1. Department of Microbiology and research laboratory “Microorganismes et Pathologie Humaine”, Habib Bourguiba university hospital, Medical school of Sfax, University of Sfax, Sfax, Tunisia
    Search for more papers by this author
  • H. Sellami,

    1. Department of Microbiology and research laboratory “Microorganismes et Pathologie Humaine”, Habib Bourguiba university hospital, Medical school of Sfax, University of Sfax, Sfax, Tunisia
    Search for more papers by this author
  • A. Znazen,

    1. Department of Microbiology and research laboratory “Microorganismes et Pathologie Humaine”, Habib Bourguiba university hospital, Medical school of Sfax, University of Sfax, Sfax, Tunisia
    Search for more papers by this author
  • A. Hammami

    1. Department of Microbiology and research laboratory “Microorganismes et Pathologie Humaine”, Habib Bourguiba university hospital, Medical school of Sfax, University of Sfax, Sfax, Tunisia
    Search for more papers by this author


Adnene Hammami, Laboratory of microbiology, University school of medecine of Sfax, Avenue Magida Boulila, 3027, Sfax, Tunisia. E-mail:



To develop and evaluate an in-house reverse hybridization technique for Chlamydia trachomatis genotype identification.

Methods and Results

The evaluation of the developed and optimized reverse hybridization method on reference strains showed the specific detection of all genotypes. This technique showed its ability to type one inclusion-forming unit of C. trachomatis genotype E and equivalent sensitivity to the Cobas TaqMan assay. It was also able to detect mixed infections in vitro. Application of the reverse hybridization method on 38 isolated C. trachomatis strains and their respective swabs allowed the detection of six urogenital genotypes D, E, F, G, H and K and one trachoma genotype B. Genotype E was the most prevalent, detected in 73% of the swab samples. Mixed infections were detected in 26% of swab cases.


The reverse hybridization technique is simple and does not require specialized instruments. It is powerful in the diagnosis of mixed infections and is suitable for use in epidemiological studies.

Significance and Impact of the Study

This technique allowed rapid Ctrachomatis genotype identification.


Chlamydia trachomatis (C. trachomatis) is an obligate intracellular bacterium and an important cause of human diseases worldwide. The World Health Organization estimated that 92 million new cases of chlamydial infections occur each year (WHO 2001).

Chlamydia trachomatis causes various diseases including trachoma, urogenital infections and lymphogranuloma venereum (LGV). This species has been classified into 19 genotypes (A, B/Ba, C, D/Da, E, F, G/Ga, H, I/Ia, J, K, L1, L2 and L3). These different genotypes were associated with diverse clinical manifestations. Genotypes A, B/Ba and C are commonly associated with trachoma. Genotypes D-K are common in the urogenital tract and cause urogenital infections. Trachoma genotypes B and C were rarely detected in the urogenital tract (Yuan et al. 1989; Morré et al. 1998). Genotypes L1-L3 cause LGV Ctrachomatis genotypes are also subdivided into three distinct groups on the basis of their amino acid sequence homology: the B group (B, D, E, L1 and L2), the C group (A, C, H, I, J, K and L3) and the intermediate group (F and G) (Yuan et al. 1989; Zheng et al. 2007). The conventional method for the classification of C. trachomatis into genotypes was based on the serologic antigenicity of the major outer membrane protein. This protein is the immunodominant surface antigen of C. trachomatis and is encoded by the ompA gene (Yuan et al. 1989; Grayston and Wang 1975). Genotype-specific epitopes were found to be localized within four surface-exposed variable segments (VS) which are interspersed by five conserved segments (CS). VS1 and VS2 showed the greatest amount of sequence variation and genotype-specific epitopes (Baehr et al. 1988; Stephens et al. 1988).

Currently, serotyping has been replaced by genotyping as the former requires culturing of clinical isolates. Genotyping is based on the nucleotide sequence polymorphism of the commonly VS of the ompA gene that was the most common method used for C. trachomatis genotyping. Genotype identification has mainly involved PCR-based restriction fragment length polymorphism (PCR–RFLP) that is labour intensive and time consuming and omp1 gene sequencing. These two techniques do not allow the identification of multiple genotype infections, at least without the additional requirement of laborious cloning procedures (Yang et al. 1993; Jurstrand et al. 2001; Stevens et al. 2010). In the past few years, several other techniques targeting the amplification of the omp1 gene have been developed for C. trachomatis genotyping such as reverse line blot hybridization (Molano et al. 2004; Xiong et al. 2006; Quint et al. 2007) and reverse dot blot (Stothard 2001; Zheng et al. 2007). These methods are simple, fast and do not require specialized instruments. Real time PCR (Jalal et al. 2007), high resolution melting analysis (Li et al. 2010), the multiple loci variable number of tandem repeats analysis (MLVA) (Pedersen et al. 2008) and the multilocus sequence typing (MLST) (Klint et al. 2007; Bom et al. 2011) were developed and used in C. trachomatis genotyping. The MLST of Klint et al. (2007) and MLVA of Pedersen et al. (2008) techniques were found to be more discriminative than the other methods. However, they are quite laborious and expensive and thus could not be used in a routine setting.

The aim of the present study was to develop an in-house reverse hybridization method for C. trachomatis genotyping. This method was chosen from all previously developed techniques as it is powerful in the diagnosis of mixed infections. This technique is also comparable to the PCR–RFLP technique, the most used technique nowadays for C. trachomatis genotyping, in terms of discriminatory power (Lan et al. 1993). This technique was validated on reference strains and then applied on 38 isolated C. trachomatis strains and their respective swabs.


Strains and clinical specimens

Fourteen reference strains of C. trachomatis were used in this study for the validation of the reverse hybridization method: A/Har-13, B/Har-36, C/TW-3, D/UW-3/Cx, E/VR-348B, F/IC-Cal-3, G/UW-57/Cx, H/UW-43/cx, I/UW12/Ur, J/UW-36/Cx, K/UW-31/Cx, L1/440/901B, L2/434/902B and L3/404/903. These strains were kindly provided by the French National Reference Center of Chlamydia (Bordeaux, France).

Eleven micro-organisms, other than C. trachomatis, infecting the urogenital tract were selected to evaluate the specificity of our PCR and the in-house reverse hybridization method. These latter include Neisseria gonorrhoeae, Mycoplasma hominis, Mycoplasma genitilum, Ureaplasma urealyticum, Ureaplasma parvum, Staphylococcus aureus, Candida albicans, group B Streptococcus, herpes simplex virus type I and II and Human papillomavirus. Chlamydia pneumoniae was also included because of its relatedness to the C. trachomatis species.

Furthermore, 38 local C. trachomatis strains from our strain collection were used for the application of the reverse hybridization method. In addition, their original urethral and endocervical urogenital samples, previously identified as C. trachomatis positive by the Cobas Amplicor PCR method (Roche Molecular systems, Mannheim, Germany), were also tested by our in-house reverse hybridization method. Three negative Cobas Amplicor PCR clinical specimens were also included in this study. All the urogenital samples were collected from Tunisian patients in the Habib Bourguiba University Hospital of Sfax between 2003 and 2010 and kept at −80°C until used.

The 38 clinical specimens used in our study were collected from 21 men and 17 women. Among men, 19 attended for current infection, 8 of them practice clandestine prostitution. The remaining two male patients reported previous history of urogenital infection. For the 17 females, 11 of them reported a genital tract infection and six were infertile. Neisseria gonorrhoeae was detected in 10 cases from all the 38 samples. This collection can be identified as a high-risk group for C. trachomatis infection.

DNA extraction method

Total DNA was extracted from 200 μl of culture medium and direct swab specimens using the QIAmp DNA Mini kit (Qiagen GmbH, Hilden, Germany) according to the manufacturer's instructions. DNA was eluted in 50 μl elution buffer.

Amplification of the target region

A PCR targeting the region encompassing the VS1 and VS2 of the ompA gene of C. trachomatis was used in this study. The NLO and C214 primers were used. This selection was based on bioinformatic analysis showing that primers used in our study could amplify all the C. trachomatis genotypes and that the amplified region covers all the specific probes used in the reverse hybridization test. Sequences of primers used in this study are listed in Table 1.

Table 1. Primers and probes used in the reverse hybridization method and sequencing
Primer/ProbeSequence (5′–3′)GenBank accession no.Reference
  1. a

    Primers used for the amplification of the VS1-VS2 used in the reverse hybridization method.

  2. b

    Primers used for the amplification of the omp1 gene.

  3. c

    Primers used for the sequencing of the omp1 gene.

  4. d

    Probe used as positive control.

C214aBiotin-TCTTCGAYTTTAGGTTTAGATTGA Lysén et al. 2004;
Sero2Ab,cTTTCTAGAYTTCATYTTGTT Lan et al. 1993;
CGGAAGTGTGGTCTCTGCCG AF352789 Xiong et al. 2006
FCCACGAAACCTGCTGCAGAT AY464145 Zheng et al. 2007
GCTGCAACAAGTATTCCAACGT AF063199 Xiong et al. 2006;
HCCTACTACCAACGATGCAGC X16007 Xiong et al. 2006
KTAACACTGCTTTGGATCGAGC AF265239 Xiong et al. 2006;
L1GGTCAAAAAGGATGCTGTCC M36533 Xiong et al. 2006;
L3TTGAATCAAGCTGTAGTTGAGC X55700 Xiong et al. 2006;
Intermediate GroupGTCTGTGGTGGAACTGTATACAG Zheng et al. 2007

The PCR mixture, which was made up to 50 μl with sterile water, contained 1 ×  PCR buffer (50 mmol l−1 Tris-HCl pH 8·3, 10 mmol l−1 KCl, 5·0 mmol l−1 (NH4)2 SO4, and 2·0 mmol l−1 MgCl2); 0·5 μmol l−1 of each primer; 0·2 mmol l−1 each dATP, dCTP, dGTP and dTTP; 1 U of the flexi Taq DNA polymerase (Promega, Madison WI, USA) and 3 μl of DNA extract.

PCR was performed using the Gene-Amp PCR System 9700 (Perkin Elmer Cetus, Freiburg, Germany) under the following conditions: an initial cycle at 95°C for 2 min followed by 35 cycles of denaturation at 95°C for 45 s, annealing at 45°C for 45 s and elongation at 72°C for 45 s, with a final cycle at 72°C for 7 min. The amplified products were detected after electrophoresis on 1% agarose gel and staining with ethidium bromide.

In the absence of amplification, a seminested PCR was carried out, using the NLO and CT4 primers (Table 1), in a final volume of 50 μl. Two microlitres of the first PCR product was used as DNA template. The amplification conditions of the seminested PCR were the same as those of the first PCR. PCR products were visualized after electrophoresis on a 1% agarose gel by staining with ethidium bromide.

As we used in some cases seminested PCR, precautionary measures were taken to prevent DNA contamination during DNA extraction and manipulations such as dedicated pre-PCR and post-PCR rooms, irradiation of the laminar flow with UV light at 254 for 10 min to inactivate extraneous DNA and the use of aeroguard filter tips. Negative controls were included in each experiment to monitor potential contamination.

The in-house reverse hybridization method

The specific detection of Ctrachomatis genotypes was performed by the reverse hybridization test. The VS1-VS2 PCR products were hybridized to the genotype-specific probes (A, B, C, D, E, F, G, H, I, J, K, L1, L2 and L3) and to the three group probes: the B group (Gb) representing genotypes B, D, E, L1 and L2; the C group (Gc) representing genotypes A, C, H, I, J and L3; and the intermediate group (Gi) representing genotypes F and G. The sequences of the oligonucleotide probes are shown in Table 1. Probes were prepared at different concentrations (5, 10, 25, 50 and 100 μmol l−1). Then, 1 μl was spotted at a specific position on the nylon membrane. Positive (Pc) and negative controls (Nc) were included in each membrane. The positive control was the NLO primer, which will hybridize with all PCR products. Sterile distilled water was used as the negative control. The layout of the probes on the membrane used in our study is shown below:


The probes were then attached to the membrane by baking at 120°C for 30 min. To remove probes excess, the membrane was washed twice with the hybridization solution (2× sodium chloride–sodium citrate, 0·1% sodium dodecyl sulphate (SDS)) for 2 min at room temperature.

To avoid the occurrence of background in the membrane on which the probes were previously fixed, two saturation methods of the membrane were tested. The first one consists of incubating the membrane in the prehybridization solution containing 6× SSC, 5× Denhardt's solution, 0·5% SDS and 100 μl diluted salmon sperm DNA for 4 h at 42°C. Hybridization was then conducted overnight at 42°C by adding amplified DNA (40 μl, concentration ranging from 30 to 100 ng μl−1) that was previously denatured for 10 min at 100°C and directly chilled on ice for 5 min to obtain single stranded DNA.

Alternatively in the second saturation protocol, the membrane was prehybridized for 30 min at 37°C in the hybridization solution. The amplified and denatured DNA was added to the hybridization solution, and hybridization was then conducted overnight at different temperatures (25, 37 and 42°C) with shaking. The membrane was subsequently saturated for 30 min at 37°C with the TBS buffer (100 mmol l−1 Tris, 0·15 mmol l−1 NaCl) at which nonfat dry milk was added at a concentration of 0·5%. The whole set-up was previously heated at 60°C to dissolve the milk.

The saturated membranes were then incubated with agitation for 30 min at room temperature in TBST buffer (TBS, 1/100 Tween 20) containing 1 μl of the streptavidin–alkaline phosphatase enzyme (Promega). After incubation, the membranes were washed three times with the TBST buffer for 15 min. Then, the membranes were incubated for 10 min in 0·1 mol l−1 Tris buffer, pH 9·5. This solution was then removed, and the duplex biotinylated DNA-probe was revealed after incubation with the substrate of the alkaline phosphatase: the Western Blue stabilized substrate (Promega) for 30 min at room temperature and in the dark. Staining was stopped by replacing the developing solution with distilled water.

The optimization results showed that the use of nonfat dry milk as saturation method, a hybridization temperature of 37°C and a probe concentration of 25 μmol l−1 are optimal (data not shown). In fact, the use of nonfat dry milk yielded stronger signals than the Denhardt's solution. Furthermore, using 42°C as hybridization temperature showed faint signals, whereas at 25°C, nonspecific signals appeared. Finally, probe concentrations higher than 25 μmol l−1 did not increase the signal.

OmpA gene amplification, sequencing and sequence analysis

Sequencing was performed to confirm genotypes detected by the reverse hybridization method. A nested PCR targeting the ompA gene of C. trachomatis was used. The NLO and NRO primers were used in the first amplification step, while the PCTM3 and Sero2A primers were used in the nested amplification step (Table 1) (Lan et al. 1994). DNA amplified fragments were purified with the QIAquick gel extraction kit (Qiagen GmbH). Sequences were then elucidated by the deoxyribonucleic chain termination method using the PCTM3 and the Sero2A primers with the DNA sequencer ABI PRISM 3100/3100-Avant Genetic Analyser (Amersham Pharmacia Biotech, Uppsala, Sweden). The resulting sequences were aligned with those in the GenBank DNA sequence database with the basic local alignment search tool program (BLAST).


Development and optimization of the reverse hybridization method

Amplification of the target region

Using the VS1-VS2 PCR, all the reference strains, the 38 C. trachomatis strains and their 38 respective clinical samples, were successfully amplified. DNA from reference and isolated strains was amplified from the first round, whereas those of the urogenital specimens were amplified using the seminested PCR. The first PCR step amplified a 671-bp fragment, and the seminested PCR step produced a 653-bp fragment. C. trachomatis-negative Cobas Amplicor PCR clinical specimens did not show any amplification.

The specificity of the seminested PCR used for the amplification was assessed. Unspecific bands were not observed when testing Mycoplasma hominis, Mycoplasma genitilum, Ureaplasma urealyticum, Ureaplasma parvum, Staphylococcus aureus, Candida albicans, group B Streptococcus, herpes simplex virus type I and II, Human papillomavirus and C. pneumoniae.

Validation of the reverse hybridization method

Specificity of the reverse hybridization method

DNA Hybridization for reference strains A to L3 was performed using the optimal conditions (Fig. 1). Positive signals were observed for all genotypes. All the hybridizations were specific. In fact, all C. trachomatis genotypes were recognized by hybridization to one of the serogroup probes and to a genotype-specific probe. Indeed, no cross hybridization between the different genotypes and groups were detected. Although the fact that C. trachomatis-negative Cobas Amplicor PCR clinical specimens as well as negative PCR control made of water as template did not show any amplification, they were subjected to the reverse hybridization method. No hybridization patterns were observed for these negative controls.

Figure 1.

Results of the reverse hybridization method on Chlamydia trachomatis reference strains. A-L3: Results of the respective reference strains on the membrane.

Sensitivity of the reverse hybridization method

To assess the sensitivity of our PCR and the hybridization reaction, tenfold dilutions of C. trachomatis genotype E were prepared at concentrations ranging from 10−4 to 10+6 inclusion forming unit (IFU). The first round of the PCR used was able to detect up to one IFU (Fig. 2a). Hybridization results using these amplicons showed that the reverse hybridization method was able to genotype up to one IFU of C. trachomatis genotype E (Fig. 2b).

Figure 2.

Limiting dilution series of DNA from Chlamydia trachomatis serovar E. (a) Agarose gel (1%) showing amplification of serial dilution of DNA from C. trachomatis serovar E. M: Molecular weight marker, 100-bp DNA ladder and Nc: Negative control of the amplification. (b) Hybridization results of the corresponding amplicons. The layout of the probes on the membrane is shown at the left. T+ is the positive control; T- is the negative control.

Study of mixed infections in vitro

As we used an in-house reverse hybridization method, we assessed whether it could identify mixed infections in vitro. Mixture of DNA of genotypes E and G were prepared at different proportions (G/E): 1/100, 1/10, 1/1, 10/1, 100/1 and 1000/1 ng. These DNA mixtures were used for the amplification and the hybridization reactions. Differences up to one log in genovar ratios could be detected reliably (Fig. 3).

Figure 3.

Study of mixed infection of the G and E serovars in vitro. The proportion G/E (ng) used is indicated above each strip.

Typing of Chlamydia trachomatis-positive samples by the reverse hybridization method

The reverse hybridization technique was applied to 38 C. trachomatis isolates and their respective swab samples. Using our hybridization method, we were able to genotype all the strains and all the swab samples. Some sample results are shown in Fig. 4. The hybridization results of both isolates and swabs are summarized in Table 2. Generally, typing of cultured isolates showed higher intensities than those performed on direct patient swabs. Results of the genotyping conducted on isolates and swabs were concordant for the presence of specific genotypes in 92% of cases (35/38) (Table 2). Twenty-six of the 38 cases (68%) showed identical genotypes (single or mixed genotypes) between isolated strains and their swab samples. Total discrepancies were observed in only three cases (7·8%). To further confirm the genotype identity determined by the reverse hybridization method, sequencing of the ompA gene was performed for 14 strains and their respective swabs, belonging to the concordant group. The samples were chosen to cover all the genotypes detected by our reverse hybridization method. All sequenced samples matched well with the hybridization results for all the genotypes in single infection. Sequencing of three mixed infections was also performed, and a superposition of signals was observed.

Figure 4.

Genotyping results from clinical specimens and their corresponding isolates. Each column represents the genotyping results of the swab and its corresponding isolate. The letter(s) above each strip indicates the serovar(s) detected.

Table 2. Serovars distribution in the 38 Chlamydia trachomatis VS1–VS2-positive strains and urogenital samples
 Genotype strainGenotype swabNumber 38
Single infection  25 (65·8%)
Concordant results  24 (63·2%)
Discordant results  1 (2·6%)
Mixed infection  13 (34·2%)
Concordant results  2 (5·3%)
Partially concordant results  9 (23·7%)
Discordant results  2 (5·3%)

Genotyping results showed the detection of all the urogenital strains except the I and J strains. In addition, genotype B was only detected by genotyping using the clinical specimens, in 1 case. Genotypes A, C and L1 to L3 were not detected in our samples. Single infection in the swabs was detected in 28 cases (74%) and mixed infection in 10 cases (26%). For culture isolates, there were eight cases (21%) of mixed infection. Genotype E was the most prevalent, detected in 74% of cases (28/38). Genotype E is followed by genotype F which was detected in six swab; G and H in five swabs, D in four swabs, K in one swab and B in one swab.


In this study, we developed a reverse hybridization method for C. trachomatis genotyping. This technique is relatively simple, fast and requires simple technology to be performed. It comprises two steps: the first one consists of the amplification of the target region followed by the reverse hybridization step. In our study, a PCR was used for the amplification of the VS1-VS2 of the ompA gene. This target region showed the highest amount of variation between the genotypes (Yuan et al. 1989). All reference strains, local isolates and clinical samples were successfully amplified. Thus, the overall sensitivity of our PCR (the first round and the seminested) was found to be 100% when using the Cobas Amplicor test as a reference test. The sensitivity of our PCR was found to be as good as those reported in other studies, using nested PCR, ranging from 44 to 99% (Jurstrand et al. 2001; Zheng et al. 2007; Hsu et al. 2006; Gita et al. 2011). Larger sample size is, however, needed to confirm our PCR sensitivity in clinical samples. Our seminested PCR was also found to be specific as it does not allowed DNA amplification of other micro-organisms.

The amplification reaction is followed by the reverse hybridization technique. All the reverse hybridization techniques used for C. trachomatis genotyping used either a commercially available kit, the Ct-DT assay, the chemiluminescence technology (Xiong et al. 2006) or the DIG system (Zheng et al. 2007; Quint et al. 2007) for enhanced sensitivities and easier spots detection. In our study, we have developed an in-house reverse hybridization test that does not use such kits. For that, several optimization steps were performed to obtain relatively good signals. These include composition of probes, saturation method, hybridization temperature, composition of hybridization mixture, probe concentrations, enzyme volume and incubation time with the substrate (data not shown).

Using the optimized hybridization conditions, all the reference strains have been successfully genotyped. Highly specific hybridizations were seen for all of them. According to Stothard (2001), cross-hybridizations were detected between the genotype J and C probes. Such cross hybridizations were not seen in our study as we used another probe for the genotype C. Our in-house reverse hybridization method was also found to be sensitive. In fact, results of the tenfold serial dilution of titrated C. trachomatis DNA showed that our reverse hybridization method allowed the detection of amplified DNA from samples harbouring up to one IFU. Using an ompA-based DNA microarray, Ruettger et al. (2011) reported that after PCR amplification of decimal dilution series of genotype D, one IFU was still typed correctly. To our knowledge, no other data on the sensitivity of other reverse hybridization methods using titrated C. trachomatis strains are available.

It is well known that the reverse hybridization method is a powerful tool for the detection of mixed infections. As we used an in-house technique, we aimed to demonstrate the reliability of our method to detect mixed infections. These latter were detected reliably when differences in gerovar ratios were up to one log. In line with our results, Stothard (2001) studied mixed infections at different ratios and reported detection of the genotypes E and F at a ratio mixture of 1/100 ng. At 1 ng, spots were light but still visible (Stothard 2001), like we also found. In another study, using an ompA-based DNA microarray assay to genotype C. trachomatis strains, Ruettger et al. (2011) reported the identification of both genotypes D and E or E and F as long as their ratio was between 1/1 and 5/1.

To evaluate our in-house reverse hybridization method with real diagnostic samples, we tested 38 clinical culture isolates as well as their respective clinical samples. The application of the reverse hybridization showed that 68% of the genotypes were identical between isolated strains and their swab samples, whereas in 7% these genotypes were totally discrepant. These discrepancies could not be explained by the occurrence of mutations in the ompA gene. In fact, Stothard et al. (1998) reported that after 20 in vitro serial passages in Mc-Coy cells, isolates and their clinical samples showed identical ompA sequences. However, recombination events have been shown to occur in the VS2 region of the ompA gene that includes the majority of probes used in our study (Harris et al. 2012). This could explain discrepancies between strains and urogenital samples as recombination events could occur in laboratory tissue cell culture (Jeffrey et al. 2010). Furthermore, it has been suggested that culture is selective and that certain genotypes are more suited to be recovered from culture than others (Wang et al. 2011). This could allow the enrichment or the loss of some genotypes after cell culture and to variation of their initial proportion in the swabs, leading to such discrepancies. Another explanation could be the chronic inflammatory consequences of the genital reproductive tract of some of our patients. In fact, evidence of the presence of abnormal reticulate bodies in macrophages from Chlamydiae-infected patient tissue samples was suggested (Wyrick 2010). Such aberrant reticulate bodies' phenotype could not be recovered by cell culture, but the presence of chlamydial DNA in diseased tissue samples could be detected. This might results from delayed nucleic acid clearance after successful eradication of viable organisms explaining thus discrepancies. The difference of sensitivity in both the amplification and the hybridization between co-infecting genotypes could also lead to variations of the detected genotypes. In fact, Ruettger et al. (2011) reported that several fold differences between co-infecting genotypes could affect detection of the genotype at the lower level. In our study, mixed infections of G/E genotypes were detected reliably only when differences in gerovar ratios were up to one log. Discrepancies could also be related to the variation in nucleotide sequence of the primer target of co-infecting genotypes leading to different PCR performances and thus of genotyping results. Another explanation might be related to the occurrence of sampling variation especially when samples contain very low numbers of the target molecule.

As genotyping results could be biased by cell culture, only results of swabs will now be discussed. Single infections were detected in 28 cases (73%). Genotype E was the most prevalent genotype detected in 71% (27/38) of all cases. This high prevalence of genotype E is in line with the reports of Gita et al. (2011) in which genotype E was the only genotype detected in a female population. Genotype B was detected in one case. In fact, trachoma genotypes are rarely involved in urogenital infections but were previously reported (Dean et al. 2008; Bandea et al. 2001; Ikehata et al. 2000). According to the literature, genotypes D, E, F and G were reported to be the most prevalent genotypes worldwide (Jurstrand et al. 2001; Molano et al. 2004). The high prevalence of genotype E in our study has to be confirmed, however, on a larger collection of specimens.

Mixed infections were detected in our study in 26% of swab and 18% of culture isolate cases. These percentages seem to be very high as mixed infections were reported from 2 to 13% (Stevens et al. 2010; Molano et al. 2004; Xiong et al. 2006). The highest frequencies (9–13%) involved principally the use of the hybridization methods (Molano et al. 2004; Xiong et al. 2006). Thus, differences in the detection of mixed infections in the literature could be related to the genotyping method used. The high percentage of mixed infections found in our study was not likely to be related to crossreactions between the probes or potential contamination but to the patient's behaviour. In fact, eight of the nine male patients showing mixed infections practice clandestine prostitution. The remaining male patient reported previous urogenital infections.

Our in-house reverse hybridization method could reliably identify mixed infections, in contrast to the PCR–RFLP and sequencing methods. In fact, Yang et al. (1993) reported that for samples showing mixed infections by PCR–RFLP, only one genotype was detected. Reanalysis of the RFLP pattern showed minor bands that had been overlooked but that could be interpreted as representing multiple genotypes. Even when detecting single infections, Pedersen et al. (2000) reported that all strains that have been typed as ‘C’ by PCR–RFLP were found to be either F or G by sequencing. The nucleotide sequencing of the ompA gene could also result in ambiguous and uninterpretable results for mixed infections (Pannekoek et al. 2008). In fact, genotypes in the sample cannot be resolved unless the ompA PCR products were cloned and multiple clones were subsequently sequenced. Despite its advantages, the reverse hybridization technique is unable to detect new genotypes or genovariants because of the occurrence of sequence variations or mutations in the sequence of the hybridization region, resulting in aberrant results.

In conclusion, we have developed our in-house PCR-reverse hybridization method for the genotyping of C. trachomatis strains. This technique is simple, fast and does not require specialized instruments and it could thus be routinely used. Compared with sequencing or PCR–RFLP, this technique is powerful in the diagnosis of mixed infections. It is also suitable for use in epidemiological studies.


We are grateful to Dr. Radhouane Gdoura for his contribution in this work. This work is part of a doctoral thesis by Houda Gharsallah. This work received financial support from the “Ministry of Higher Education and Scientific Research, Tunisia”.