Identification of lymphogranuloma venereum-associated Chlamydia trachomatis serovars by fluorescence in situ hybridisation – a proof-of-principle analysis


  • Hagen Frickmann,

    Corresponding author
    1. Department of Tropical Medicine at the Bernhard Nocht Institute, German Armed Forces Hospital of Hamburg, Hamburg, Germany
    2. Institute for Microbiology, Virology and Hygiene, University Hospital Rostock, Rostock, Germany
    • Corresponding Author Hagen Frickmann, Department of Tropical Medicine at the Bernhard Nocht Institute, German Armed Forces Hospital of Hamburg, Bernhard Nocht street 74, D-20359 Hamburg, Germany. Tel.: 0049 40 6947 28743; Fax: 0049 40 6947 28709; E-mail:

    Search for more papers by this author
  • Andreas Essig,

    1. Institute for Medical Microbiology and Hygiene, University Hospital Ulm, Ulm, Germany
    Search for more papers by this author
  • Sven Poppert

    1. Institute for Medical Microbiology, Justus Liebig University Giessen, Giessen, Germany
    2. Bernhard Nocht Institute for Tropical Medicine, Hamburg, Germany
    Search for more papers by this author


We describe a proof-of-principle evaluation of a fluorescence in situ hybridisation (FISH) procedure to identify Chlamydia trachomatis serovars L1–L3, the causative agents of lymphogranuloma venereum, in cell cultures based on newly designed DNA probes. Rapid and easy-to-perform FISH could facilitate the diagnosis of lymphogranuloma venereum without nucleic acid amplification or serotyping, but requires broader evaluation studies, for example, in tropical high-endemicity regions.


Nous décrivons une évaluation d'une preuve de principe d'une procédure d'hybridation fluorescente in situ (HFIS) pour l'identification des sérotypes L1 à L3 de Chlamydia trachomatis, agents étiologiques de la lymphogranulomatose vénérienne, dans des cultures cellulaires, en utilisant des sondes d’ADN nouvellement conçues. La procédure HFIS est rapide et facile à effectuer et pourrait faciliter le diagnostic de la lymphogranulomatose vénérienne sans sérotypage ou amplification d'acide nucléique, mais elle nécessite des études d’évaluation plus larges, par exemple, dans les régions tropicales à forte endémicité.


Describimos una evaluación de la prueba de principio del procedimiento de hibridación in situ con fluorescencia (FISH) para identificar los serovares L1–L3 de Chlamydia trachomatis, los agentes causantes del linfogranuloma venéreo, en cultivos celulares basados en un nuevas sondas de ADN. Una prueba rápida y fácil de hacer como el FISH podría facilitar el diagnóstico de linfogranuloma venéreo sin amplificación de ácidos nucléicos o serotipado, pero requiere de estudios de evaluación más amplios, por ejemplo en regiones tropicales altamente endémicas.


Lymphogranuloma venereum (LGV), caused by Chlamydia trachomatis serovars L1, L2; L2a, and L3 and characterised by ulcers and lymphadenopathy as well as invasive spread with fistulas, fibrosis and strictures consecutively in later stages, is locally endemic in Africa, Asia, South America and the Caribbean but is comparatively rarely observed in central Europe (Rampf et al. 2004). Although LGV outbreaks with the newly identified variant L2b occur in European promiscuous men having sex with men (Bremer et al. 2006), the tropics remain the major region of endemicity (Rampf et al. 2004).

Nucleic acid amplification testing (NAT) techniques are the method of choice for the diagnosis of chlamydial infections. The former diagnostic gold standard of cell culture is nowadays only infrequently used in microbiological diagnostic routine due to its poor sensitivity compared with NAT (Moller et al. 2008), although NAT may fail if new genetic variants emerge as described (Ripa & Nilsson 2007).

However, if chlamydiae are grown in cell culture, further identification is needed as soon as characteristic inclusion bodies are detected microscopically. A fluorescence in situ hybridisation (FISH) protocol using a hierarchical probe set for the rapid discrimination of Chlamydia spp. in cell culture has been described (Poppert et al. 2002). FISH is based on the specific binding of fluorescence-labelled oligonucleotide probes usually to ribosomal RNA of the target organisms. The procedure is faster and easier to perform than NAT (Moter & Göbel 2000), thus allowing for a convenient discrimination of chlamydiae in cell monolayers. However, a FISH probe with specificity for LGV-associated C. trachomatis serovars L1–L3 has not been described so far and was not considered in the previous publication (Poppert et al. 2002).

When we succeeded in isolating a C. trachomatis L2b strain whose identity was confirmed by sequencing of a 1100-bp fragment of the ompA gene in Hela229 cells from a penile ulcer of a 19-year-old man having sex with men (Rampf et al. 2004), the C. trachomatis serovar L1–L3-specific DNA probe LGV 16S 457 (5′-TCC-AGC-GGG-TAT-TAA-CCG-CC-3′) was designed using the software ARB (Ludwig et al. 2004; Kumar et al. 2006) and evaluated in silico using the software probeCheck (Loy et al. 2008). In silico evaluation suggested the combined use of the sulfoindocyanine Cy3-labelled LGV 16S 457 probe with the non-labelled competitor probes LGV 16S 457 comp1 (5′-TCC-GGC-GGG-TAT-TAA-CCG-CC-3′) and LGV 16S 457 comp2 (5′-TCC-AGC-GGG-TAT-TAA-CCG-TC-3′) to block non-specific binding to closely related chlamydiae whose sequence showed only a one-base-pair mismatch with the target sequence of the LGV strains (Table 1).

Table 1. In silico evaluation using the software probeCheck (December 2013). Probe binding to be expected within the zero-mismatch range and one-mismatch range. Chlamydia spp. strains were not detected in the two-mismatch range and three-mismatch range (data not shown). Bold and underlined type indicates mismatches in the target sequence
Mis-match countRibosomal RNA sequence corresponding to the probeMatching with known sequences
Probe LGV 16S 457: 5′-TCC-AGC-GGG-TAT-TAA-CCG-CC-3′
0GGC-GGU-UAA-UAC-CCG-CUG-GAChlamydia trachomatis serovar L2 (n = 3; accession numbers ACUI01000001; AM884177; CP002024), Chlamydia trachomatis 434/Bu animal isolate (n = 1), Chlamydia trachomatis not further characterised (n = 15), unidentified strains (n = 3)
1GGC-GGU-UAA-UAC-CCG-CCG-GA (covered by competitor probe LGV 16S 457 comp1)Chlamydia suis (n = 5), Chlamydia muridarum (n = 1)
1GAC-GGU-UAA-UAC-CCG-CCG-GA (covered by competitor probe LGV 16S 457 comp2)Chlamydia trachomatis strains (human and animal isolates) 6276s (n = 1), 70 (n = 1), 6276 (n = 1), D(s)2923 (n = 1), 70s (n = 1), D/UW-3/CK (n = 1), A/HAR-13 (n = 1), G/11074 (n = 1), E/11023 (n = 1), G/9301 (n = 1), G/9768 (n = 1), G/11222 (n = 1), D-LC (n = 1), D-EC (n = 1), L2c (n = 1), E/150 (n = 1), A2497 (n = 1), Sweden2 (n = 1), Chlamydia trachomatis not further characterised (n = 24), unidentified strains (n = 2)
1GGC-GGU-UAA-UAC-CCG-CUU-GA (not covered by a competitor probe)Rubrobacterineae bacterium BR7-21 (n = 1)

The LGV-chlamydiae-specific probe/competitor probes combination was applied in combination with a sulfoindocyanine Cy5-labelled CHLALES 16S 523 probe (5′-CCT-CCG-TAT-TAC-CGC-AGC-3′) targeting all Chlamydiales and a fluorescein-isothiocyanate (FITC)-labelled probe CHLCAE 16S 574 (5′-CTT-TCC-GCC-TAC-ACG-CCC-3′) targeting all Chlamydiaceae as described (Poppert et al. 2002) to confirm the presence of Chlamydiaceae in cellular vacuoles (Figure 1). Optimal hybridisation conditions for the new probe/competitor probes combination were assessed by increasing the formamide concentration in the hybridisation buffer from 10% to 50% in stages at 46 °C as described (Moter & Göbel 2000; Poppert et al. 2002) (data not shown), and the optimum was confirmed at 30% formamide as intended during the probe design. Further increase in the formamide concentration led to reduced fluorescence intensity due to reduction in the strength of binding of the probes to the target organisms in infected cells. The estimation of the required number of bases for the probe and the competitor probes was based upon the authors' experience, although software programs such as mathFISH (Yilmaz et al. 2011) are nowadays available for calculating the binding strength of a probe.

Figure 1.

Correct identification of the patient's C. trachomatis serovar L2b strain as an LGV-causing strain by the Cy3-labelled LGV 16S 457 FISH probe (red fluorescence) in combination with two non-labelled competitor probes (a). In contrast, no LGV 16S 457 probe-associated red fluorescence was observed in cells infected with a C. trachomatis serovar K strain (b). Presence of Chlamydiaceae in the respective vacuoles of the infected cell monolayers was confirmed by the Cy5-labelled CHLALES 16S 423 probe (infrared fluorescence, visualised as dark blue) staining all Chlamydiales and the FITC-labelled CHLCAE 16 573 probe (green fluorescence) staining all Chlamydiaceae. Merged images demonstrate overlays of the fluorescence channels, thus proving specific probe binding at corresponding positions. Microscopy was performed with a Leica TCS SP confocal laser scanning microscope.

Fluorescence in situ hybridisation (FISH) was performed as described (Poppert et al. 2002) with minor modifications. Fixation of the infected cells on microscopic slides and inactivation of the bacteria were performed with 2% paraformaldehyde, and afterwards, the slides were air-dried. Hybridisation was performed in hybridisation buffer (30% formamide, 0.9 m sodium chloride, 0.02 m Tris-HCl, 0.01% sodium dodecyl sulphate) in the moist chamber at 46 °C for 90 min. The non-intercalating DNA-stain 4,6-diamidino-2-phenylindole (DAPI) was added to the washing buffer (0.1 m sodium chloride, 0.005 m EDTA, 0.02 m Tris-HCl, 0.01% sodium dodecyl sulphate) as detailed (Poppert et al. 2002) to prove specific probe binding to sites where bacterial DNA is present. Slides were washed for 15 min at 46 °C. Afterwards, the slides were mounted using the mounting medium CitiFluor AF1 (Citifluor Ltd., London, UK) and stored cold and dark in the fridge until they were assessed under a Leica TCS SP confocal laser scanning microscope (Leica MICROSYSTEMS, Wetzlar, Germany).

In vitro evaluation with a previously described Chlamydia spp. strain panel including C. pneumoniae (TW 183, Washington Research Foundation, Seattle, WA, USA), C. trachomatis serovar K (UW31/CX, ATCC VR 887) and a clinical C. psittaci isolate (Essig et al. 1995; Poppert et al. 2002) supplemented with LGV-associated C. trachomatis L1–L3 strains (ATTC VR 901B, VR 902B, VR 903) showed specific binding to the LGV-associated strains C. trachomatis L1–L3 only. The non-target organisms were negative by LGV-specific FISH. Subsequent hybridisation of the clinical isolate C. trachomatis serovar L2b with the new LGV-specific probe/competitor probes combination led to bright specific fluorescence (Figure 1).

If it is intended to pursue cell culture-based diagnosis of chlamydial infections despite its limitations regarding sensitivity (Moller et al. 2008), FISH may allow for a rapid and easily performed identification of LGV-associated C. trachomatis strains in cell cultures. However, the rare occurrence of LGV in Germany (Rampf et al. 2004) and the associated persisting lack of clinical isolates of LGV-associated C. trachomatis serovars in our laboratory did not allow for a broad in vitro evaluation of our new probe/competitor probes combination with any considerable number of clinical isolates, which constitutes a limitation of this proof-of-principle analysis. The very limited number of strains used in the panel for empirical testing makes it important to confirm the probe specificity against a larger panel of closely related strains. We recommend such an evaluation study in high-prevalence areas to further validate the diagnostic value of our LGV-FISH procedure in routine diagnosis.

Further, FISH-based detection of C. trachomatis directly in human cells from swabs has been demonstrated (Kapur et al. 2006). Evaluation studies of the LGV-FISH protocol described here in high-prevalence settings should include testing with such primary materials to assess the technique's potential use as a rapid diagnostic procedure without prior cell culture. Such FISH approaches require no more than the standard equipment of a microbiological diagnostic laboratory such as an incubator and a fluorescence microscope. The price of the reagents of one FISH reaction is as low as 1 USD if published patent-free probes are used, allowing the use of the technique even in resource-limited areas.

In well-equipped laboratories in wealthy countries, PCR will remain the gold standard for the diagnosis of chlamydial infections due to its superior sensitivity in case of chronic infections, with 10³ chlamydiae/μl sample being sufficient for a positive result (Moller et al. 2008). However, if resources are scarce, as they are in many tropical countries, inexpensive and easily performed FISH allows for the detection and differentiation of chlamydiae at least if high bacterial density is guaranteed, for example in the case of early infection.

After thorough validation studies, LGV-FISH might further be used for basic scientific purposes, for example, to correlate the spatial distribution of chlamydiae in infected tissues with corresponding histopathological findings as demonstrated for other pathogens (Moter & Göbel 2000) or to analyse the intracellular developmental cycle of C. trachomatis serovars L1–L3 as demonstrated for C. pneumoniae (Poppert et al. 2002).

In summary, the probe/competitor probes combination described here completes the previously described probe panel for the FISH-based discrimination of chlamydiae (Poppert et al. 2002), increasing its value for both diagnostic and scientific purposes.


We are grateful to Ulrike Simnacher and Adelinde Steingruber for excellent technical assistance. A poster abstract containing data from this note has been published as follows: ‘Essig A, Rampf J, Sunderkotter C, et al. Molecular diagnosis of Lymphogranuloma venereum using 16S rRNA-based fluorescence in situ hybridisation (FISH). Int J Med Microbiol. 2004; 294:91’.