Abstract
One of the mainstays in the prevention of Chlamydia trachomatis and Neisseria gonorrhoeae infections is the availability of laboratory diagnostics with high sensitivity and specificity. Assays for diagnosis of C. trachomatis include cell culture and nucleic acid amplification tests (NAATs). The major target sequences for C. trachomatis diagnosis by NAATs are located at the cryptic plasmid and the major target used for characterisation is the omp1 gene. The gold standard for diagnosis of N. gonorrhoeae is culture. However, numerous NAATs for identification of N. gonorrhoeae and a number of molecular genetic methods for characterisation of N. gonorrhoeae have been developed. Probably no routine laboratory can attain as high sensitivity by culturing C. trachomatis or N. gonorrhoeae as by using NAATs. For that reason NAATs can be recommended for diagnosing C. trachomatis, but not as the only diagnostic assay for N. gonorrhoeae, due to lack of antibiotic susceptibility testing and specificity problems, most pronounced for pharyngeal and rectal samples. Genotyping of C. trachomatis or N. gonorrhoeae provides additional information for contact tracing. It is recommended for N. gonorrhoeae, at least in low prevalence geographic areas, but cannot today be recommended for C. trachomatis. This is due to the low genetic variability and hence the limited benefits for partner notification. However, genotyping of C. trachomatis may play an important role under special circumstances.
Chlamydia trachomatis and Neisseria gonorrhoeae infections belong to the most frequently encountered sexually transmitted infections worldwide (1). Even though the bacteriological diagnosis was not established until the 19th and 20th centuries, clinical symptoms probably associated with these bacteria are described in the ancient literature (2).
The clinical spectrum of genital diseases caused by these two bacteria is quite similar and mainly comprises urethritis in men and urethritis/cervicitis in women. However, no symptoms or only very slight symptoms such as weak discharge are common, especially in women. Oropharyngeal infections may occur in both sexes, and eye infections, acquired from infected mothers, are described in newborns. Conjunctivitis in adults is due to self-inoculation from a genital infection (3, 4).
Most complications due to C. trachomatis and/or N. gonorrhoeae infections affect women. Ascending genital-tract infections involving the endometrium, fallopian tubes and/or adjacent pelvic structures, i.e. pelvic inflammatory disease (PID), may develop with a risk of infertility, ectopic pregnancy and chronic pelvic pain (5, 6). In addition to female morbidity, the costs of these diseases are considerable (7). When the incidences of C. trachomatis and N. gonorrhoeae infections declined in Sweden during the 1980s and 1990s, this was followed by declining incidences of PID and ectopic pregnancy (5, 6). Accordingly, strategies to detect, treat and prevent genital C. trachomatis and N. gonorrhoeae infections are essential to prevent PID and its later complications. In the absence of effective vaccines, the mainstay in the prevention of C. trachomatis and N. gonorrhoeae infections is the availability of specific laboratory diagnostics with high sensitivity, effective antibiotic treatment, contact tracing and screening of high incidence groups, as well as education of the public with respect to safe sexual behaviour (5, 6, 8, 9).
This review will discuss:
- 1)
different molecular genetic methods for diagnosis and characterisation of C. trachomatis and N. gonorrhoeae,
- 2)
the strengths and weaknesses of these methods,
- 3)
the possible advantages/disadvantages of using genetic characterisation of these bacteria for understanding the epidemiology of C. trachomatis and N. gonorrhoeae infections and its usefulness in contact tracing.
BIOLOGY OF CHLAMYDIA TRACHOMATIS
The Chlamydia (C) species are distinguished from all other microorganisms by a unique growth cycle, and are placed in their own family (Chlamydiaceae). They are obligate intracellular parasites that replicate within the cytoplasm of host cells and the elementary body (EB) is adapted for extracellular survival and for initiation of infection. After adhesion to the eukaryotic cell, the EB enters the cell by endocytosis and stays in an intracellular vacuole, which develops to a so-called inclusion. Subsequently, the EB changes to a metabolically active and dividing form called the reticulate body (RB), which is adapted for intracellular multiplication. There are three Chlamydia species pathogenic to humans: C. pneumoniae, C. psittaci and C. trachomatis. All of these express a genus-specific lipopolysaccharide (LPS) antigen on the surface, which causes cross-reactions between the three species in diagnostic tests based on Chlamydia LPS antigen (10). Serological variants (serovars) A–C of C. trachomatis preferably colonise the eye and cause trachoma, serovars D–K preferably colonise the genital tract and cause genital infections, and L1–L3 cause lymphogranuloma venereum (LGV). C. pneumoniae and C. psittaci are etiological agents mostly of respiratory infections.
Some recent genome sequencing studies have provided us with a new understanding of C. trachomatis (11, 12). The first sequenced C. trachomatis, serovar D, genome consisted of a 1,042,519 base-pair chromosome (GenBank accession no. AE001273) and a 7,493 base-pair plasmid (12). Analysis of this genome showed that C. trachomatis comprises 894 likely protein-coding genes. Counterparts of enzymes characterised in other bacteria were identified in C. trachomatis to account for the minimal requirements for DNA replication, repair, transcription and translation. The inclusion membrane (Inc) proteins and the polymorphic membrane proteins (Pmp) found occupy 12% to 19% of the Chlamydia-specific genomic sequences and are, according to our present knowledge, unique to Chlamydia. The Chlamydia organism expresses a major outer membrane protein (MOMP) that is surface-exposed and forms the basis for classification of different serovars of C. trachomatis. The omp1 gene, which codes for the MOMP, is present in all three human pathogenic Chlamydia species and is used for genotype determinations of C. trachomatis. The gene contains five conserved domains and four variable domains (VDI to VDIV) that vary considerably between the species (13, 14).
LABORATORY DIAGNOSIS OF CHLAMYDIA TRACHOMATIS
Conventional diagnostic methods
Assays that are used nowadays for diagnosis of C. trachomatis include cell culture, which is the only method that is based on detection of live C. trachomatis and has in the past been considered the reference standard for assessing the validity of new diagnostic tests due to its high specificity and under optimised conditions reasonably high sensitivity. In short, a specimen is collected in an appropriate transport medium and is inoculated onto a confluent monolayer of McCoy cells in the laboratory (15). After 48–72 h of growth, intracytoplasmic inclusions have been developed in infected cells. The inclusions are detected by staining with fluorescein-conjugated monoclonal antibodies (mAbs) that are specific for the MOMP of C. trachomatis or other Chlamydia-specific structures. Cell culture is time consuming and laborious, and was in many routine laboratories replaced by antigen- or DNA/RNA detection during the 1990s.
Some laboratories use antigen detection methods such as enzyme immunoassays (EIA) and/or direct immunofluorescence assays (DFA). C. trachomatis EIA tests detect chlamydial LPS with a monoclonal or polyclonal antibody. Manufacturers (MicroTrac, Chlamydiazyme, etc.) have developed blocking antibodies that verify C. trachomatis-positive EIA results because of the potential for false-positive results. In DFA, specimen material is placed directly on a slide. Fluorescein-conjugated antibodies react with the Chlamydia surface and are visualised by fluorescence microscopy. Depending on the commercial product used, the antigen detected by the fluorescein-conjugated antibodies in the C. trachomatis DFA is either the LPS or the MOMP component. This method requires a skilled microscopist who can distinguish between fluorescing chlamydial particles and non-specific fluorescence. DFA is mostly used as a confirmatory test in some routine laboratories. As in EIA tests the anti-LPS antibodies can cross-react with nonchlamydial bacterial species, as well as with LPS of C. psittaci and C. pneumoniae. DFA with a C. trachomatis-specific anti-MOMP monoclonal antibody is considered highly specific (16).
Nucleic acid amplification tests (NAATs)
Several commercial NAATs are available nowadays, and like in-house polymerase chain reaction (PCR) assays use different DNA or RNA regions as target molecules for amplifying C. trachomatis DNA/RNA in clinical samples. The major target sequences are located at the cryptic plasmid, which is present in approximately 10 copies in each C. trachomatis organism (17, 18). Using the plasmid as target DNA could theoretically increase the sensitivity compared to using a single chromosomal gene such as the omp1 gene (19). However, some studies suggest that plasmid-free variants of C. trachomatis may on rare occasions be present in clinical samples (20), and these will not be detected if the plasmid is used as target DNA. NAATs available for detection of the cryptic plasmid in C. trachomatis in clinical specimens are the COBAS Amplicor®Chlamydia trachomatis Test (Roche Diagnostic Systems), which is a PCR, and the Becton Dickinson BDProbeTecTM ET (Becton Dickinson) test, which uses strand displacement amplification (SDA). The ligase chain reaction (LCR) for detection of the cryptic plasmid (LCx C.trachomatis® assay; Abbott Laboratories) is not available anymore. A comparison and evaluation, assessing the performance of these three NAATs was done by Van Dyck et al. (21), and shows that the sensitivity and specificity were quite similar. The Gen-Probe APTIMATM assay (Gen-Probe) for C. trachomatis uses transcription-mediated amplification (TMA) to detect a specific 23S rRNA target, which is present in hundreds of copies in each Chlamydia organism. The Gen-Probe TMA assay was included in a comparison with LCR and COBAS Amplicor assays for detection of C. trachomatis in first-void urine (22); the results were evaluated versus cell culture, and no significant differences were demonstrated between the three commercial techniques. But when using a new “gold standard”, i.e. a sample was considered to be true positive if two or more techniques yielded positive results, COBAS Amplicor showed a higher sensitivity compared to Gen-Probe TMA and LCR. The same gold standard was used in a five-city study, in a head-to-head multicenter comparison of the DNA probe test (PACE 2 DNA probe test from Gen-Probe) and NAATs (Amplicor microwell plate PCR test from Roche and the LCx LCR test from Abbott) for C. trachomatis infection in women in the United States (23). It was concluded that LCR and PCR tests performed on endocervical swabs and urine were superior to the PACE 2 test for screening C. trachomatis infection in women. Furthermore, an assay based on the nucleic acid sequence-based amplification (NASBA) technique for detection of C. trachomatis 16S rRNA in clinical samples (NucliSens Basic kit, Organon Teknika; 24) and a real-time PCR for detection of the omp1 gene (RealArt C. trachomatis TM PCR, Abbott Laboratories; 25) have also been developed. In conclusion, the validity of a diagnostic test will be a reflection of what has been chosen as comparator and reviews of screening tests for C. trachomatis conclude that commercial NAATs are more sensitive than non-NAATs (21, 26–32).
CHARACTERISATION OF CHLAMYDIA TRACHOMATIS
Conventional phenotypic characterisation
The MOMP is the major structural protein exposed on the surface of the infectious EB and RB. Serovariant-specific epitopes are associated with the MOMP of C. trachomatis and conform to serovars determined by microimmunofluorescence (33). Prototypic serovars designated A to K and L1-L3 as well as additional immunovariants (Ba, Da, Ia, etc.) have been identified. Serovars A, B, and C are associated with trachoma, D to K with urogenital infections, and L1-L3 with LGV. Since cell culture is necessary for this method the viability of the cells and the number of typeable organisms are of great importance. This method is only used in a minority of laboratories.
Genotypic characterisation
By use of MOMP directed (omp1 gene) primers in a PCR, which covers the variable domains I – IV (VDI to VDIV), it is possible not only to detect but also to distinguish between the various types of C. trachomatis in noncultured clinical samples (14). Amplification of the omp1 gene by PCR has made it possible for further characterisation either by restriction fragment length polymorphism (RFLP) or by DNA sequence analyses. There is a good agreement between RFLP and serotyping by monoclonal antibodies as was shown by Morré et al. (34), and it has also been shown that DNA sequencing is a more reliable epidemiological tool compared to RFLP (35). Typing of clinical isolates has shown that D, E and F are the most common genotypes in different parts of the world; the distributions are shown in Table 1. The genotypes D, E and F account for 55–84% in the present review, and as high as 94% in a study from New Delhi among females with urogenital infections (46). In a population of pregnant women from Thailand, genotype E was found only in 9% of the specimens (38), compared to 24–60% in other studies. For example, the prevalence of genotype E in the age group 13–23 years was found to be about 60% in a study from a youth health center in Sweden (40), and a study from Iceland reported the prevalence of genotype E to be 60–70% in the age group 17–22 years (41). The high prevalence of genotype E is also found in a study from five different areas in United States by Millman et al. (44). In that study a slightly different pattern was seen. Genotype I and J had a prevalence of 26%, compared to the other studies reviewed where prevalences of 0–12% were seen (for details see Table 1). Millman et al. also provided a quantitative assessment of the genetic diversity within each genotype, and found that E, Ba, H, Ia and F were most conserved, while J, G, D, and K were the most divergent. This is in agreement with the findings in two studies from Sweden (39, 43) which reported that only 4–6% of genotype E clinical isolates and 2–4% of genotype F diverged from the respective reference sequences (39, 43). On the contrary, almost all clinical isolates defined as genotype H in these two studies from Sweden were found to harbour substitutions distinguishing them from the reference sequence.
| No. of sequences of different genotypes (%) | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Authors (reference) | Area | Yeara | Method | B | D | E | F | G | H | I | J | K | Total |
| Ikehata et al.(36) | Japan | 2000 | RFLPb | 12(5) | 48(22) | 53(24) | 24(11) | 39(18) | 15(7) | 15(7) | 5(2) | 9(4) | 218 |
| Morré et al.(37) | Netherlands | 2000 | RFLP | NFc | 51(12) | 177(40) | 93(21) | 36(8) | 16(4) | 17(4) | 19(4) | 17(4) | 438d |
| Pedersen et al.(35) | Denmark | 2000 | DNA seq | NF | 3(10) | 15(48) | 8(26) | 1(3) | NF | NF | 1(3) | 3(10) | 31e |
| Bandea et al.(38) | Thailand | 2001 | DNA seq | 3(7) | 10(23) | 4(9) | 11(25) | 1(2) | 5(11) | 3(7) | 2(5) | 5(11) | 44 |
| Jurstrand et al.(39) | Sweden | 2001 | DNA seq | 1(<1) | 32f(14) | 112(47) | 41(17) | 6(3) | 6(3) | 7(3) | 10(4) | 21(9) | 237 |
| Sylvan et al.(40) | Sweden | 2002 | RFLP | NF | 3(10) | 18(60) | 4(13) | NF | NF | NF | NF | 4(13) | 30g |
| Jonsdottir et al.(41) | Iceland | 2003 | DNA seq | NF | 35(11) | 190(58) | 32(10) | 9(3) | 8(2) | 4(1) | 35(11) | 15(4) | 330h |
| Oehme et al.(42) | Germany | 2003 | RFLP | 1(1) | 8(9) | 42(46) | 19(21) | 6(7) | 3(3) | NF | 5(5) | 7(8) | 91 |
| Lysén et al.(43) | Sweden | 2004 | DNA seq | 9(1) | 60(9) | 264(39) | 139(21) | 77(11) | 16(2) | 7(1) | 48(7) | 58(9) | 678 |
| Millman et al.(44) | USA | 2004 | DNA seq | 5(1) | 72i(14) | 150(30) | 98(19) | 21(4) | 3(1) | 73(14) | 59j(12) | 26(5) | 507 |
| Ngandjio et al.(45) | Cameroon | 2004 | RLFP | NF | 6(18) | 14(41) | 7(21) | 5(14) | NF | NF | 2(6) | NF | 34 |
- a Year of publication.
- b Six unclassified specimens (RFLP) were further DNA sequenced and defined as B variants and five unclassified specimens were defined as D variants.
- c NF=not found.
- d 13 specimens (3%) were defined as unrecognizable PCR-based RFLP patterns.
- e RFLP was compared to DNA sequencing; the genotyping results from DNA sequencing are presented.
- f 14 specimens were defined as D/B-120 and 6 as D/B-185.
- g One specimen (3%) was impossible to type.
- h Two mixed infections were found.
- i One was defined as genotype Ba/D.
- j 24 was genotype Ja.
BIOLOGY OF NEISSERIA GONORRHOEAE
The obligate human pathogen N. gonorrhoeae belongs to the Neisseria genus of the bacterial family Neisseriaceae. The two primary pathogenic Neisseria species, N. gonorrhoeae and N. meningitidis, are genetically and morphologically similar. However, molecular, cellular and biochemical differences exist, which probably reflect the different diseases that the two species are causing (47, 48). Thus, N. gonorrhoeae mostly colonises the urogenital tract and causes gonorrhoea, while N. meningitidis preferably colonises the throat/upper respiratory tract and in some cases is the cause of meningitis and/or septicaemia.
The chromosome of the N. gonorrhoeae strain FA1090 was recently determined to consist of 2,153,944 bp (GenBank accession no. AE004969, http:www.genome.ou.edugono.html). One cryptic plasmid, several β-lactamase-encoding plasmids and different conjugative plasmids may also be carried by N. gonorrhoeae strains. Horizontal genetic exchange, through transformation and conjugation, frequently occurs between N. gonorrhoeae strains and also with other similar species, mostly commensal Neisseria or N. meningitidis (49–52). A random recombination between chromosomal genetic loci, most probably mediated by transformation, is presumed to be frequent in nature, and consequently the population structure of N. gonorrhoeae is fully sexual, i.e. panmictic or non-clonal (49). The frequent horizontal genetic exchange in combination with mutations causes a high level of genotypic and phenotypic variability, which is important for evasion or adaptation to the immune response of the host and for development or spread of antibiotic resistance mechanisms. These properties make the bacteria effective in persisting without severely damaging the host, i.e. in producing mildly symptomatic or asymptomatic infection (53), and this also stresses the importance of using adequate methods for diagnosis and characterisation of this highly variable pathogen.
LABORATORY DIAGNOSIS OF NEISSERIA GONORRHOEAE
Conventional diagnostic methods
For a diagnosis of intracellular diplococci, direct microscopy of Gram- or methylene blue-stained smears of urethral specimens from men or women is still being used in STD clinics. However, endocervical samples of women have a much lower sensitivity. The method is rapid, inexpensive and quite specific. Disadvantages are the subjectivity when interpreting the results and the unsuitability of using rectal and pharyngeal specimens. Furthermore, asymptomatic infection of N. gonorrhoeae may be less likely to be detected (53–55), and it is not possible to perform antibiotic susceptibility testing.
Culture of fastidious N. gonorrhoeae is still the gold standard for diagnosing gonorrhoea. However, the use of adequate swabs, a suitable transportation media for the specimens, and a short transportation time are of the utmost importance. In addition, optimised inoculation and incubation as well as the use of selective culture media (56) optimally combined with a nonselective media of high quality are also crucial (54, 57, 58). The presumptive diagnosis based on identification of suspected colonies with typical morphology, positive oxidase reaction, and detection of Gram-negative diplococci by microscopy has also to be confirmed by, for instance, carbohydrate oxidation assays, biochemical and rapid enzyme substrate tests, chemiluminescent DNA probes that hybridise to specific N. gonorrhoeae rRNA sequences, coagglutination tests, or monoclonal fluorescent-antibody tests (54, 55, 57–59). Under optimised conditions, culture of N. gonorrhoeae has a high specificity and sensitivity, is inexpensive, and suitable for most types of specimens. In addition, culture allows antibiotic susceptibility testing and characterisation of N. gonorrhoeae strains. Meanwhile, culture for diagnostic purposes is quite time consuming.
Different EIAs for antigen detection have also been developed for diagnosis of N. gonorrhoeae. These are rapid, easy to perform, do not require viable organisms, and may be sensitive and specific for detection of bacteria in male urethral specimens and first-void urine (60). However, they are less sensitive for endocervical specimens (57, 61, 62) and in asymptomatic infection. Rectal and pharyngeal specimens are not suitable (55, 57, 59), and it is not possible to perform antibiotic susceptibility testing.
Nucleic acid amplification tests (NAATs)
Numerous DNA/RNA-based noncultural methods, mainly NAATs, for the identification of N. gonorrhoeae have been developed. The PACE 2 assay (Gen-Probe) has been used for several years to detect 16S rRNA of N. gonorrhoeae (63) by DNA probe hybridisation. Digene Hybrid Capture 2 (HC2) CT/GC including HC2 GC-ID for N. gonorrhoeae (Digene Corporation) is a newer RNA-probe based assay that identifies specific DNA sequences of the chromosome as well as the cryptic plasmid of N. gonorrhoeae (64). In general, the hybridisation methods have a high specificity but a somewhat lower sensitivity than the NAATs. Over the years, different PCR assays have also been developed. These identify N. gonorrhoeae-specific chromosomal genes encoding outer membrane proteins such as the reduction-modifiable protein (Rmp) (65), the chromosomal putative cytosine DNA methyltransferase gene (Cobas Amplicor, Roche Diagnostics; 66, 67), the 16S rRNA gene (66), or the cppB gene located in regions of the cryptic plasmid that may also be integrated in the chromosome (68). LCRs for identification of N. gonorrhoeae chromosomal multicopy pilin genes and opa genes, respectively, have also been developed (LCx, Abbott Laboratories (69, 70)) but are not available anymore. Furthermore, a NASBA technique (NucliSens Basic kit, Organon Teknika; 24), and a TMA (APTIMA Combo 2, Gen-Probe; 71) for amplification of 16S rRNA, as well as a SDA assay for a chromosomal DNA sequence homologous to a site-specific recombinase gene, i.e. the multicopy pilin gene-inverting protein homologue (pivNg gene) (BDProbeTecET, Becton Dickinson 72), has been developed.
In general, NAATs have a higher sensitivity than culture especially for samples from the rectum and pharynx (21, 73–76). However, NAATs based on detection of genes on the cryptic plasmid, e.g. the cppB gene, result in a lower sensitivity due to the lack of this plasmid or a cppB gene integrated in the chromosome in several strains (77). False-positive samples, mostly due to commensal Neisseria species, have been identified by several of the NAATs (66, 77, 78), which emphasizes the importance of adequate interpretation of the results and resolving equivocal results by retesting the original samples (78). In addition, due to the high degree of interspecies genetic recombinations, the use of a sensitive and specific confirmative assay that targets other genes may in the future be necessary for all the NAATs, especially for extragenital specimens such as those from the pharynx (66, 77, 79).
CHARACTERISATION OF NEISSERA GONORRHOEAE
Conventional phenotypic characterisation
Auxotyping of N. gonorrhoeae isolates was described already in 1973 (80). Different auxotypes differ according to their nutritional requirements for amino acids, purines, pyrimidines, and vitamins. However, auxotyping is time consuming, labour intensive, quite expensive, and possesses limited discriminatory ability.
The internationally established serogroup and serovar determination, e.g. by using the co-agglutination technique (81), with mAbs is based on antigenic diversities of the outer membrane protein PorB. The Genetic systems (GS) panel (82) and the Pharmacia (Ph) panel (83) are the most widely used panels of mAbs. The serovar determination is fast, easy to perform, cost-effective, does not require sophisticated equipment, and gives information concerning the antigenicity of PorB. Disadvantages are suboptimal discriminatory ability, problems concerning reproducibility, limited access to GS mAbs of adequate quality, and that interpretation of the results may occasionally be subjective (84–87). In addition, the prevalence of non-serotypeable strains as well as new serovars (84, 86–88) will probably increase over time due to the ongoing evolution of the porB gene.
Antibiograms for determination of epidemiological relationships between N. gonorrhoeae isolates have low discriminatory ability and are not stable over longer time periods. Consequently, antibiograms should not be used as a reliable epidemiological characterisation method though they generate valuable clinical information and are essential for monitoring the changing patterns of antibiotic susceptibility and for adequate treatment of the patients.
Genotypic characterisation
Molecular methods for characterisation of N. gonorrhoeae that are not subject to the limitations of the phenotypic characterisation methods have been developed, and some of these are summarised in Table 2. These methods analyse the presence of different enzymes or plasmids, single genetic locus, multiple loci, or they potentially index the entire genome of N. gonorrhoeae strains.
| Method | Reference(s) |
|---|---|
| Multilocus enzyme electrophoresis(MEE), which indexes allelic variation in multiple chromosomal housekeeping enzymes | 89 |
| Plasmid analysis | 90, 91 |
| Conventional Sanger sequencing of entire or more polymorphic segments of the porB gene | 85, 87 |
| Genetic variant(genovar) determination by pyrosequencing technology on highly polymorphic segments of the porB gene | 92 |
| Hybridisation of biotinylated probes to variable regions of the porB gene | 93 |
| Ribotyping, i.e. restriction fragment length polymorphism (RFLP), of rRNA genes with identification of fragments by hybridisation of specific rRNA probes | 91 |
| Neisseria gonorrhoeae multi-antigen sequence typing [NG-MAST]), which analyses parts of the porB gene and the transferrin-binding protein B(tbpB) gene | 94 |
| opa-typing by PCR amplification followed by restriction endonuclease digestion of the opa genes | 50 |
| Multilocus sequence typing(MLST) that identifies allelic polymorphism in several chromosomal housekeeping genes | 95 |
| Subtyping by analysis of the number and sequences of repeats in the genes encoding an outer membrane lipoprotein (Lip) | 96 |
| Amplified ribosomal DNA restriction analysis(ARDRA) | 97 |
| Amplified fragment length polymorphism(AFLP) | 98 |
| Arbitrarily primed PCR(AP-PCR) or randomly amplified polymorphic DNA(RAPD) typing | 85, 97 |
| Digestion of the entire genome by high-frequency cutting restriction endonucleases followed by separation of DNA fragments by polyacrylamide gel electrophoresis(PAGE) | 99 |
| Digestion of the entire genome by rarely cutting restriction endonucleases followed by separation of DNA fragments using agarose pulsed-field gel electrophoresis(PFGE) | 100 |
The discriminatory power of some of the techniques, such as multilocus enzyme electrophoresis (MEE) and plasmid analysis, may often be insufficient, and plasmids can be lost. In addition, techniques based on interpretation of band patterns on gels or hybridisation patterns are often labour intensive and more subjective, which may cause reproducibility problems. These methods subsequently require pronounced standardisation in a national and global context, especially for interlaboratory comparisons of achieved patterns. According to evaluations of phenotypic methods and different molecular typing methods, porB gene sequencing (entire or partial), Neisseria gonorrhoeae-multiantigen sequence typing (NG-MAST), multilocus sequence typing (MLST) (by using 13 genes), pulsed-field gel electrophoresis (PFGE) with suitable restriction enzyme(s) (for instance SpeI and/or BglII) and especially opa-typing have a high discriminatory ability between N. gonorrhoeae strains (85–87, 94, 95, 97, 100, 101). Subsequently, these seem to be suitable molecular typing techniques, at least for short-term epidemiology. However, different genetic targets have a different evolutionary history, which results in a lack of complete congruence between different methods and, consequently, we need to acquire a more thorough knowledge of the molecular mechanisms and time scale of the evolutionary changes seen in different parts of the N. gonorrhoeae genome.
DISCUSSION
Advantages of the diagnostic DNA/RNA-based methods are the high sensitivity and in most cases high specificity, speed, the possibility of using non-invasive samples such as urine, no requirement for viable organisms, the potential for using samples collected by the patients themselves (102), that cost-effective pooling of specimens may be possible, and the opportunity for simultaneous detection of several agents, e.g. N. gonorrhoeae and C. trachomatis (53, 57, 58). Probably no routine laboratory for clinical microbiology can attain as high sensitivity by culturing C. trachomatis or N. gonorrhoeae as by using NAATs. By using two different NAATs with different target sequences a positive result of one assay can be confirmed by the other assay, which in special situations may be of benefit, for example in legal questions.
Disadvantages of the diagnostic DNA/RNA-based methods are the specificity, i.e. for N. gonorrhoeae, and, in some cases, decreased sensitivity due to inhibitors for both C. trachomatis and N. gonorrhoeae, expensive equipment, risk of contamination by previously amplified nucleic acid, lack of approved protocol for some specimens (pharyngeal and rectal samples, for example), and that no viable bacteria are isolated for strain characterisation or antibiograms for N. gonorrhoeae (53, 57, 58, 103). However, some strain characterisation is possible without viable bacteria and C. trachomatis is, so far, sensitive for antibiotics used for treatment, e.g. tetracycline, so antibiograms for C. trachomatis are still of minor interest. The existence of inhibitors also stresses the importance of using adequate protocols for preparation of the DNA template as well as using inhibition controls, especially in some methods. However, their influence may be decreased by dilution of the clinical samples or by storage in a refrigerator or freezer (103).
In the context of adequate transportation of specimens, reasonably high quality at routine laboratories with respect to culturing N. gonorrhoeae and low prevalence of gonorrhoea, it is at present not recommended that NAATs be used as the sole method for diagnosis due to the lack of antibiotic susceptibility patterns and the lack of possibilities for characterising the strains. However, for screening core groups of high-frequency transmitters or for use in countries with suboptimal transportation of specimens or culture, NAATs may be more suitable.
It is often not rational for a routine laboratory to have two different assays for detection of C. trachomatis. However, if both cell culture and NAATs are in use it is possible to perform NAAT if a patient sample causes cytotoxic effects on the cells and vice versa to perform cell culture when a urethral or cervix sample shows inhibition in the NAAT. Another argument for still having the opportunity to use the cell culture technique is the possibility of future antibiotic resistance of C. trachomatis. If that time were to come, it would be appropriate for the cell culture technique to be available.
Due to low sensitivity, EIA techniques can only be recommended in a few situations (53, 57, 58, 104, 105), such as in laboratories where more complicated assays are not available, for example in some developing countries.
Characterisation of C. trachomatis and N. gonorrhoeae strains can provide valuable information regarding the variants circulating in the community, and with increased knowledge of the epidemiology of Chlamydia infections and gonorrhoea, efforts to prevent spread can probably be more effective. An advantage of genotyping methods for C. trachomatis is that they can be used for non-invasive samples such as urine (38–41).
Genotyping of C. trachomatis mostly provides a higher discrimination between strains and, consequently, additional information when applied to contact tracing. Genotyping was shown to be a valuable tool in determining whether a new infection had occurred in those patients still C. trachomatis-positive at follow-up and where a new genetic variant had been introduced (43, 106). Findings of discrepant strains within sexual networks would improve contact tracing by suggesting that there were unmentioned contacts, unknown to the contact tracer (43, 106). The study by Falk et al. (106) confirmed the general experience of STD-clinics that partner notification in Sweden is probably – despite legislation – suboptimal. Two major reasons are the small catchment area of each clinic and the inadequacy of information given by the index patient to the contact tracer. Centralization of partner notification and limiting the number of clinics where people can attend would probably improve partner notification, decrease further transmission, and decrease the incidence of C. trachomatis and N. gonorrhoeae infections. However, there are to our knowledge, no published studies where partner notification efficacy of centralized and de-centralized partner notification schemes has been adequately compared.
In most reviewed studies, C. trachomatis genotype E was shown to be the most common genotype (see Table 1). This fact and the overall genetic stability of the different isolates (39, 40, 43) limited its usefulness in a partner notification context (106). On the contrary, if a protective vaccine could be developed, it would be advisable to cover genotype E isolates. Today, in unselected populations, the low genetic variability and hence the benefits for partner notification are too limited for the use of sequence-based genotyping of C. trachomatis to be recommended in routine work. In addition, technical developments and a reduction in costs may be needed to render it more cost effective (43). However, genotyping of C. trachomatis could play an important role in special circumstances, such as sexual abuse.
Genotypic characterisation of N. gonorrhoeae is significantly more discriminating, reproducible, objective and reliable as an epidemiological tool for N. gonorrhoeae than routinely performed serovar determination. These molecular techniques can be used to increase our knowledge of strain populations in the entire society or in core groups, temporal and geographic changes, emergence and transmission of individual strains, to disprove treatment failures and to confirm presumed epidemiological connections or distinguish isolates of suspected clusters (85–87, 94, 100, 107–110). This information together with epidemiological questionnaires is crucial for accomplishing accurate contact tracing, and for identifying core groups, risk behaviours, the use of effective antibiotic treatment and different preventive measures and interventions. However, in some situations serovar determination may still be useful as a primary epidemiological marker because it reveals information concerning the exposure and antigenicity of different PorB epitopes, which can be important in the context of the immune response, immune protection and, ultimately, for the development of a vaccine. The questions asked in relation to each issue should guide the use of the most appropriate method for epidemiological characterisation.
The best approach for fast, objective, portable, highly discriminative, reproducible, highly typeable, and high-throughput characterisation of N. gonorrhoeae is probably based on sequencing of one or two highly variable genes, e.g. the porB and tbpB gene. Furthermore, especially for global or long-term epidemiology and reliable phylogenetic studies, sequencing of more conserved and evolutionary neutral chromosomal housekeeping genes may be needed.
CONCLUSIONS
-
• NAATs can be recommended as diagnostic assays for C. trachomatis in routine diagnostic laboratories.
-
• NAATs cannot yet be recommended as the only diagnostic assay for N. gonorrhoeae in routine diagnostic laboratories. This is due to the lack of antibiotic susceptibility testing, specificity problems, especially for some specimens such as pharyngeal and rectal samples, and that characterisation of the isolate is not possible.
-
• Genotypic characterisation of C. trachomatis and N. gonorrhoeae provides additional information for contact tracing.
-
• Genotypic characterisation cannot yet be recommended for C. trachomatis due to limited cost-benefits in routine diagnostics, but can be recommended in special situations, such as in the case of sexual abuse.
-
• Genotypic characterisation can be recommended for N. gonorrhoeae, at least in low prevalence geographic areas, but also in specific situations, for example during investigations of outbreaks of gonorrhoea and in the case of sexual abuse.
