Choosing an appropriate technique to detect prevalent and lesser-known gastrointestinal pathogens relies heavily on balancing the specificity, sensitivity, turnaround time and, of course, the costs
Monika Malecki PhD
Oliver Schildgen PhD
Department of Molecular Pathology,
Institute of Pathology,
Kliniken der Stadt Köln gGmbH,
Gastroenteric diseases remain a major health problem worldwide. Up to five billion cases of gastroneterits are reported every year – in the UK alone, 1.5 million people have gastroenteric symptoms every year.(1) In many of these cases the causative agent remains undetected. Because the source can be either bacterial, viral or parasitic, the identification of the causative agent poses an important diagnostic challenge. To apply an adequate treatment and to avoid transmission for both medical and economic reasons, healthcare institutions must implement a suitable diagnostic algorithm in stool diagnostics.
Most gastroenteric diseases are self-limited and do not require specific medication other than symptomatic treatment. Thus, further analysis of the samples is only necessary in particular cases, for instance when the illness takes a severe course and the patient suffers from bloody diarrhoea. However, recent diagnostic techniques are dramatically challenged by pathogens with high recombination potential and a high number of genetic variants. These involve enterohaemorrhagic Escherichia coli (EHEC) and human norovirus, which represent two constantly evolving, and notably precarious, pathogens that frequently cause outbreaks of gastroenteritis.
Norovirus is the most common virus causing gastroenteritis in adults in industrialised countries.(2) As far as children are concerned, norovirus is the second most common pathogen responsible for gastroenteritis. The virus is transmitted via the faecal–oral route either from person to person by smear infection, by foodborne infection or by aerosolisation. With an infectious dose of fewer than 20 virus particles, norovirus is extremely contagious and resistant in its environment, which is why it often emerges as a nosocomial infection as well as in places where people socialise.(3)
The genus norovirus is a human pathogenic member of the family of Caliciviridae.
Caliciviruses are single-stranded RNA, non-enveloped viruses with a spherical capsid and icosahedral symmetry. Norovirus comprises five genogroups (GI–GV) subdivided into at least 31 genotypes, of which GI, GII and GIV are infectious for humans. Furthermore, GI and GII have the highest diversity and are commonly associated with acute gastroenteritis.(4)
To classify norovirus nucleotide sequence analysis of the virus, reading frame 2 (ORF2) has been established. The highly conserved ORF2 encodes the major viral capsid protein VP1, which plays a regulatory role in virus replication.(5) At the same time, the protruding domain of VP1 is highly variable, inducing antigenic drift and antigenic shift within the norovirus genus. On the one hand, the genetic variability of norovirus results in epitope variations challenging serological assays; on the other hand, potential mutations at the primer-binding sites complicate molecular assays.
Norovirus diagnosis is based primarily on three diagnostic tools: direct detection of viral particles using electron microscopy, viral antigen detection with enzyme-linked immunosorbent assay (ELISA), and detection of viral RNA with reverse transcription polymerase chain reaction (RT-PCR). Which of these tests is appropriate depends on a couple of questions: Are we dealing with occasional cases of illness or are we in an acute outbreak situation? Is our facility a standard diagnostic laboratory or are we capable of performing molecular biological assays? And, as always, costs are another important aspect that should not be underestimated.
Norovirus was first identified in 1970 using electron microscopy, which was rapidly applied into routine diagnostics to detect norovirus particles in stool specimens. The specificity is its greatest advantage over other methods. Other than ELISA and PCR, electron microscopy is unaffected by mutations with regards to the typical morphology of the virus. By using electron microscopy, a number of viral pathogens with characteristic morphology can be identified. Electron microscopy is simple to perform but it requires an electron microscope and an experienced microscopist, and provides a low sensitivity (approximately 106 virus particles per gram of stool). Because of advanced expertise as a vital prerequisite, low sensitivity and high costs over time, this method has become an unsuitable diagnostic tool in day-to-day routines. Nevertheless, this method might be auxiliary in outbreak situations where an unknown agent has been tested negative for norovirus by either ELISA or RT-PCR.
To date, with its moderate limit of detection reaching from 104 virions/ml stool suspension, ELISA is widely used in the detection of norovirus, especially in laboratories with no or limited access to molecular biological technologies. Most common ELISA tests are performed with the RIDASCREEN® (R-Biopharm) and IDEIATM EIA Test (Oxoid). Both assays detect norovirus genogroups I and II and perform similarly to RT-PCR, with a moderate specificity of approximately 93–100%, and a low sensitivity of approximately 57–80%.(6–9) Interestingly, the IDEIA assay shows reactivity to a greater range of norovirus genotypes than the RIDASCREEN assay.(10)
RIDAQUICK from R-Biopharm is another immunological test based on quick immunochromatographic screening. It identifies both genotypes I and II in a turn-around time of 20 minutes and with no requirement for laboratory facilities. Compared with RT-PCR, its specificity averages 99% and shows a sensitivity of around 50–80%.(6)
To summarise, immunological tests are very simple to handle, fast and cost-effective and have no requirement for additional instrumentation. Yet they struggle with the antigenic diversity of norovirus, thereby making them an accurate first-line screening test in norovirus gastroenteritis outbreaks, which require rapid examination of a high number of samples. By contrast, immunological tests are not suitable for single sample analysis and because of their low sensitivity, it is highly recommended to add a molecular testing when immunological tests are negative.
Nowadays most virology laboratories performing molecular tests rely on quantitative RT-PCR (qRT-PCR). After nucleic acid extraction of the stool specimen, the viral RNA is transcribed by reverse transcriptase into cDNA using either random or norovirus-specific oligonucleotide primers. The cDNA is then amplified based on norovirus-specific primers complementary to the highly conserved ORF1–ORF2 region. As few as 102 copies/ml of stool suspension are sufficient to detect the clinically most relevant genotypes of norovirus with a high specificity and sensitivity (both >90%).(11)
To date there are several commercially available qRT-PCR kits for norovirus detection with either RT and PCR running separately or in a single step reaction (for example, QuantiTect Virus+ROX Vial Kit from Qiagen; Calici/Astrovirus Consensus from Argene; SmartNorovirus from Cepheid). Single-step RT-PCR is fast, reduces hands-on time and minimises chances of contamination. Two-step RT-PCR should be employed if subsequent amplification of other pathogens in addition to norovirus or research of new genotypes is required.
Noroviruses may also be detected through multiplex PCR assays, in which a broad range of gastrointestinal pathogens is identified simultaneously. Since different primers are applied in one batch, the conditions of the PCR need to be accommodated and may lead to reduced sensitivity. In this regard, the GPP assay from Luminex has emerged as a suitable multiplexing alternative, combining RT-PCR with bead hybridisation to detect 15 different gastrointestinal pathogens in one reaction mix in a turnaround time of five-to-six hours.
Compared with other methods, real-time RT-PCR for norovirus gives the highest sensitivity and specificity. It is even able to detect new genotypes and requires only a short handling time. Regrettably, real time RT-PCR depends on a real-time cycler and along with the reaction kits amounts to the most costly detection method.
Enterohaemorrhagic E. coli belongs to a variety of serovars of the bacterium E. coli, which are characterised by their ability to produce Shiga-like toxin (also called verotoxin). When infecting humans, EHEC induces a variety of inflammatory events within the intestine, such as haemorrhagic colitis. In few cases, the infection is associated with haemolytic uraemic syndrome (HUS), which can be life threatening. HUS is accompanied by renal impairment, haemolytic anaemia, thrombocytopenia, and central nervous system and myocardial damage, which in the worst of cases can be fatal. Approximately 8% of patients suffering from EHEC develop HUS, mainly children under the age of five years.(12–14)
Although EHEC does not belong to the most prevalent bacterial pathogens, as far as its severe symptomatology and its precarious treatment with antibiotics are concerned, EHEC has notable relevance in stool diagnostics. The infective dose of EHEC is only 100 colony-forming units, so person-to-person transmissions and its appearance in various foods require an overall survey and proper diagnostics.
The ‘gold standard’ for Shiga-like toxin detection in stool specimens is a cytotoxicity assay for Vero cells derived from African green monkey kidney. Vero cytotoxicity is a time-consuming and labour-intensive method requiring cell culture equipment and providing results at the earliest 48–72 after infection, making this method inappropriate for day-to-day routines.
E. coli O157:H7 is the most frequent EHEC variant worldwide.(15) E. coli O157:H7 screening is performed with sorbitol-MacConkey agar. Because the bacterium is not able to ferment sorbitol, it forms translucent colonies, which can be tested for their reaction with antisera to the O157 antigen. However, samples identified as E. coli O157:H7 should be passed to a reference laboratory for confirmation. The Centres for Disease Control and Prevention recommends testing by culture and toxin profile simultaneously.
Nevertheless, most frontline laboratories test solely for O157:H7.
However, the recent O104:H4 outbreak in Germany shows just how important it is to consider non-O157 EHEC strains. To identify O104:H4 and other rare EHEC strains, direct testing via toxin detection as well as indirect testing via serologic detection, has been established. After culture, the pathogen ELISA or PCR can be employed for direct detection. A number of ELISA tests are commercially available for Shiga-like toxin detection (for example, Premier EHEC from Meridian). Comapred with conventional culture, ELISA is faster, serogroup-independent and has high sensitivity comparable to Vero cytotoxicity.
The performance of EHEC real-time PCR is of equal quality because it is fast and accurate. In addition to a variety of available monoplex PCR tests (for example, mericon® VTEC stx 1/2 Kit from Qiagen), multiplex assays combining the detection of EHEC with other gastrointestinal pathogens have been developed (for example, ProGastro SSCS by Gen-Probe and GPP by Luminex). In an outbreak setting it might be reasonable to test for a variety of pathogenicity factors other than Stx1 and Stx2, for example, eae (Intimin) and hylA (Hemolysin), which are present in O157:H7 but not in O104:H4. Furthermore, EHEC can be detected via anti-lipopolysaccharides (LPS)–IgM antibodies against E. coli serogroups performed by ELISA, Western Blot or latex agglutination test (for example, by Deika-Seiken). That said, LPS does detect novel EHEC variants, of which the serogroups are known.
Choosing an appropriate technique to detect prevalent and lesser-known gastrointestinal pathogens relies heavily on balancing the specificity, sensitivity, turnaround time and, of course, the costs. Ideally, clinical laboratories should screen all diarrhoeal stool specimens for a broad range of potential pathogens using methods that are not restricted to serogroups.
For both norovirus and EHEC, PCR-based screening proves to be the most sensitive and specific technique available. However, if a laboratory cannot fund PCR, it is best advised to make use of an alternative method. Any of the alternatives discussed above for norovirus, namely electron microscopy, ELISA and qRT-PCR, are very useful for epidemiological examination of gastroenteritis outbreaks. Yet in terms of sporadic cases of diseases, it is highly recommended to apply two of these assays simultaneously. Likewise for the detection of EHEC, a rapid PCR assay is used ideally as a preliminary screen and then supplemented by conventional confirmation methods, such as culturing.
- Wheeler JG et al. Study of infectious intestinal disease in England: rates in the community, presenting to general practice, and reported to national surveillance. BMJ 1999;318(7190):1046–50.
- Lindesmith L et al. Human susceptibility and resistance to Norwalk virus infection. Nature Med 2003;9:548–53.
- Teunis PFM et al. Norwalk virus: How infectious is it? J Med Virol 2008;80(8):146–76.
- Zheng DP et al. Norovirus classification and proposed strain nomenclature. Virology 2006;346(2):312–23.
- Subba-Reddy CV et al. Norovirus RNA synthesis is modulated by an interaction between the viral RNA-dependent RNA polymerase and the major capsid protein, VP1. J Virol 2012;86(18):10138–49.
- de Bruin E et al. Diagnosis of Norovirus outbreaks by commercial ELISA or RT-PCR. J Virol 2006;137(2):259–64.
- Kirby A at al. An evaluation of RIDASCREEN and IDEIA enzyme immunoassays and the RIDAQUICK immunochromatographic test for the detection of norovirus in fecal specimens. J. Clin Virol 2010;49(4):254–7.
- Morillo SG et al. Norovirus 3rd Generation Kit: An improvement for rapid diagnosis of sporadic gastroenteritis cases and valuable outbreak detection. J Virol Methods 2011;173(1):13–16.
- Derrington P et al. Norovirus Ridaquick: a new test for rapid diagnosis of norvirus. Pathology 2009;41(7):687–8.
- Gray JJ et al. European multicenter evaluation of commercial enzyme immunoassays for detecting norovirus antigen in fecal samples. Clin Vaccine Immunol 2007;14(10):1349–55.
- Kageyama T et al. Broadly reactive and highly sensitive assay for Norwalk-like viruses based on real-time quantitative reverse transcription-PCR. J Clin Microbiol 2003;41(4):1548–57.
- Tarr PI et al. Shiga-toxin-producing Escherichia coli and haemolytic uraemic syndrome. Lancet 2005;365(9464):1073–86.
- Boyce TG et al. Escherichia coli O157:H7 and the haemolytic-uremic syndrome. N Engl J Med 1995;333(6):364–8.
- Bell BP et al. A multistate outbreak of Escherichia coli O157:H7-associated bloody diarrhea and haemolytic uremic syndrome from hamburgers. The Washington experience. JAMA 1994;272(17):1349–53.
- Friedrich A et al. Prevalence, virulence profiles and clinical significance of Shiga toxin-negative variants of enterohemorrhagic Escherichia coli 0157 infection in humans. Clin Infect Dis 2007;45(1):39–45.