Abstract
Marked norovirus (NoV) diversity was detected in patient samples from a large community outbreak of gastroenteritis with waterborne epidemiology affecting approximately 2,400 people. NoV was detected in 33 of 50 patient samples examined by group-specific real-time reverse transcription-PCR. NoV genotype I (GI) strains predominated in 31 patients, with mixed GI infections occurring in 5 of these patients. Sequence analysis of RNA-dependent polymerase-N/S capsid-coding regions (∼900 nucleotides in length) confirmed the dominance of the GI strains (n = 36). Strains of NoV GI.4 (n = 21) and GI.7 (n = 9) were identified, but six strains required full capsid amino acid analyses (530 to 550 amino acids) based on control sequencing of cloned amplicons before the virus genotype could be determined. Three strains were assigned to a new NoV GI genotype, proposed as GI.9, based on capsid amino acid analyses showing 26% dissimilarity from the established genotypes GI.1 to GI.8. Three other strains grouped in a sub-branch of GI.3 with 13 to 15% amino acid dissimilarity to GI.3 GenBank reference strains. Phylogenetic analysis (2.1 kb) of 10 representative strains confirmed these genotype clusters. Strains of NoV GII.4 (n = 1), NoV GII.6 (n = 2), sapovirus GII.2 (n = 1), rotavirus (n = 3), adenovirus (n = 1), and Campylobacter spp. (n = 2) were detected as single infections or as mixtures with NoV GI. Marked NoV GI diversity detected in patients was consistent with epidemiologic evidence of waterborne NoV infections, suggesting human fecal contamination of the water supply. Recognition of NoV diversity in a cluster of patients provided a useful warning marker of waterborne contamination in the Lilla Edet outbreak.
INTRODUCTION
Norovirus (NoV), which belongs to the Caliciviridae family, is considered to be the major cause of acute nonbacterial gastroenteritis in all age groups, worldwide (38). As small (27 nm), nonenveloped, positive single-stranded RNA viruses, NoV show marked genomic diversity, low infectious dose, and considerable stability in environmental waters (15, 36, 44). These viral properties play a central role in waterborne infections of NoV etiology. A recent U.S. survey estimated waterborne infections to total 19 million cases/year (41), with NoV accounting for ca. 18% of drinking-water outbreaks associated with gastroenteritis (4). Investigation of NoV waterborne outbreaks is hampered by NoV diversity and noncultivability in standard cell cultures (11). Genomic diversity challenges the molecular methods used to detect NoV in clinical samples and in contaminated waters, the point source of waterborne infections.
These robust yet fastidious viruses cause diarrhea, nausea, and vomiting of sudden onset and short duration (24 to 72 h), characterized by high concentrations of NoV RNA in feces and vomitus (19). Therefore, NoV are readily transmitted by the fecal-oral route through contaminated foods such as raspberries, water or bivalves, aerosols, or fomites (6, 9, 25, 29). Two contrasting patterns of outbreaks are recognized (32). Costly long-running outbreaks occur in semiclosed settings such as cruise ships or health care units where NoV genogroup II (GII) strains may dominate (12, 34), while GI or GII strains may be implicated in sporadic infections or widespread nonseasonal epidemic outbreaks (23, 25). Since human NoV are acid tolerant, bile tolerant, and moderately resistant to chlorine disinfection (10), virus is excreted directly into sewage systems, and high levels are detected in incoming wastewater (8). Wastewater sequencing studies (31) reveal the diversity of human NoV GI and GII strains circulating in the population as sporadic, epidemic, or asymptomatic infections. Secondary wastewater treatment reduces but does not effectively remove NoV from wastewaters (8). Residual viruses are released to surrounding waters, a potential source of contamination for water treatment plants (WTPs), recreational bathers, and bivalves alike, as demonstrated by the high nucleotide similarity of human NoV outbreak strains and NoVs detected in bivalves and surface waters (25, 35). Such findings confirm the stability of human NoV in aquatic environments (36) and the importance of vigilance in the use of recycled river waters as drinking-water resources.
Tracing the point source of a community outbreak is of immediate importance, as is a prompt response with remedial action, especially when waterborne infection is suspected since large numbers of individuals are at risk. However, waterborne outbreak investigations are demanding. Apart from the problems of demonstrating NoV in water samples where collection time is usually delayed following “flash” contamination events, waterborne outbreaks often correlate with maintenance or emergency incidents, disinfectant failure or flooding at WTPs or wastewater treatment plants (WWTPs), heavy rain, or human error (20, 33). Significantly, the first warning of waterborne infections may come through increased reporting of gastroenteritis from the public (28).
In the present study, we focus on molecular findings from clinical samples obtained during investigation of a large outbreak of NoV gastroenteritis of waterborne epidemiology, affecting a small municipality in West Sweden (27). The outbreak was characterized by early detection of marked diversity of NoV GI strains in patient feces. These findings raise interesting questions since waterborne NoV infections frequently go undetected or are missed early in an outbreak because of over-reliance on the absence of bacterial indicators of microbial contamination in drinking water (25).
MATERIALS AND METHODS
Outbreak background.
During September 2008 an extensive outbreak of gastroenteritis affected Lilla Edet (population, 13,000) situated on the River Göta, Västra Götaland, Sweden. Outbreak investigations reported previously included assessment of questionnaires issued to households during the outbreak, and virological analysis of water samples carried out at the Swedish Institute for Communicable Disease Control, Solna, Sweden (27). Lilla Edet WTP draws recycled water from the River Göta, the recipient for 13 industrial plants, including WWTPs at Lilla Edet, upstream Trollhättan (population, 44,500), and Vänersborg (population, 22,000), approximately 87,500 people in all. River water quality is monitored with microbial, chemical, and turbidity analyses. Good communications give early warning of maintenance or alarm events allowing corrective intervention at downstream WTPs to reduce contamination risks (42).
Clinical samples.
Fecal samples were collected from patients with gastro-enteric symptoms attending Lilla Edet primary health care center from 12 to 26 September. Appeals were made via local radio on September 14 to ensure an adequate number of samples for laboratory examination.
Nucleic acid extraction.
Feces were processed according to principles for prevention of cross-contamination in sample handling and PCR processing (26). Total nucleic acids and RNA (100-μl volumes) were extracted from 10% fecal suspensions as described previously (36).
Real-time reverse transcription-PCR (RT-PCR).
Total nucleic acid extracts (10 μl) were examined in validated real-time TaqMan systems targeting NoV GI, GII, rotavirus (RoV), and astrovirus (HuAsV) (16, 36), sapovirus (SaV) G1, G2, and G4 (39), and adenovirus (HuAdV) (18). Master mixes (40 μl) with primer and probe concentrations of 300 and 200 nM, respectively, were prepared and amplified as described previously (36). Real-time RT-PCR controls included viral positive feces samples, water controls, and a plasmid construct designed to control viral amplification procedures (16). Cycle threshold values (CT) provided semiquantitative assessment of viral load where CT varies inversely with viral concentration, with the copy number estimated from standard curves prepared from 10-fold serial dilutions of plasmid containing TaqMan viral target sequence inserts (16).
RT-PCR and sequence analysis.
RNA (10 μl, 10%) was examined in RT-PCR targeting the NoV RNA-dependent RNA polymerase (RdRp)-N/S capsid-coding region (∼900 nt). NoV GI strains were detected with the primer pairs NV36/GISKR and JV12y/GISKR (24, 45, 47) in parallel RT-PCR master mixes (40 μl) prepared and amplified as described previously (35). Products (5 μl) were nested into PCR master mixes (45 μl) containing primers JV12y/GISKR or NV69/GISKR (47). Amplicons were separated, purified, and cycle sequenced with the primers JV12y, GISKR, and NV69 and a BigDye Terminator kit (v3.1; Applied Biosystems, Foster City, CA) (35). Sequences analyzed on Avanti3130XL genetic analyzer (Applied Biosystems) were aligned with GenBank reference strains using Sequencher version 4.9 (Gene Codes Corp., Ann Arbor, MI) and the BLAST tool (National Center for Biotechnology Information [NCBI] website). NoV GII strains were amplified in RT-PCR with the primer pair NV36/G2SKR (24) or JV12y/G2SKR, followed by “semi-nesting” into parallel PCR mixes containing the primer pairs JV12y/G2SKR, or NI/G2SKR (14). Thermocycling, product detection, and cycle sequencing were as described for NoV GI, but using the sequencing primers JV12y, NI, and G2SKR. SaV genotyping was based on sequence analysis of products (805 bp) amplified by RT-PCR as described for NoV, but using the primers N290 (5′-GATTACTCCAAGTGGGACTCCA-3′) modified from S290 (21) and SaV1245R (39). HuAdV-positive samples were confirmed by sequence analysis of PCR products amplified using HuAdV primer pairs (37).
Long-fragment sequence analysis and cloning studies.
NoV RdRp-major capsid-coding regions (∼2.1 kb) of 10 representative patient strains were characterized by long-fragment sequence analysis (35). Poly(T) primer 5′-T25VN-3′ (30) targeting viral poly(A)-3′ tail was used in parallel RT-PCR master mixes with primer NV36, NV69, or JV12y to amplify NoV GI strains (35). Amplified mixes (5 μl) were nested in parallel PCR mixes containing primer JV12y, NV69, or GIFFN combined with GICapA or GV7 (13, 15, 46). Purified products were cycle sequenced with GI primers (Table 1). NoV GII-positive samples were amplified in parallel RT-PCR master mixes with the primer pair NV36/T25VN or JV12y/T25VN, followed by nested or seminested PCR with the primer pair JV12y/GIICapC or N1/GIICapC (35, 46). Amplicons were cycle sequenced with GII primers (Table 1).
Table 1.
Sequencing primers used in long fragment analysis of NoV GI and GII strains detected in patient feces
| NoV genotype(s) | Primer | Polarity | Oligonucleotide sequence (5′–3′) | nt positiona | Source or reference |
|---|---|---|---|---|---|
| GI.4, GI.7, GII.6 | JV12Y | + | ATACCACTATGATGCAGAYTA | 4553–4572* | 45 |
| GI.3, GI.4, GI.7, GI.9 | NVp69 | + | GGCCTGCCATCTGGATTGCC | 4733–4752* | 47 |
| GI.3, GI.4 | GICap2 | + | GGBAATGCYTTYACDGCKGG | 5700–5719* | 15 |
| GI.3, GI.4, GI.7, GI.9 | GIFFmix | + | ATHGAACGYCAAATYTTCTGGAC | 5075–5097* | 22 |
| + | ATHGAAAGACAAATCTACTGGAC | 22 | |||
| + | ATHGARAGRCARCTNTGGTGGAC | 22 | |||
| GI.3, GI.4, GI.7, GI.9 | G1SKR | – | CCAACCCARCCATTRTACA | 5671–5653* | 24 |
| GI.3, GI.4, GI.7, GI.9 | GIFFN | + | GGAGATCGCAATCTCCTGCCC | 5327–5347* | 13 |
| GI.3, GI.4, GI.7, GI.9 | GIcapR1 | – | CGCTTGATGTAGCGTCCTTAGAC | 5391–5369* | 36 |
| GI.3, GI.4, GI.7, GI.9 | GICapA | – | GGCWGTTCCCACAGGCTT | 6914–6897* | 46 |
| GI.4 | FS1(I.4) | + | GTTGTATCACAATAATGAC | 5861–5879* | This study |
| GI.4 | FS3(I.4) | + | CATCTGGACAGAGGGTGCTCA | 6239–6259* | This study |
| GI.4 | RS6(I.4) | – | GTATAGTGCCAATATAGTC | 6491–6473* | This study |
| G1.4 | RS8(I.4) | – | GTCATTATTGTGATACAACAC | 5879–5859* | This study |
| GI.7 | F7n(I.7) | + | GACCACACCTTAATCCCTTTC | 5614–5633* | This study |
| GI.7 | F1n(I.7) | + | CAGCTAGTCAAGTGGCCCGCATAC | 6205–6228* | This study |
| GI.7 | R1n(I.7) | – | CAATCACAAGCTCCGATGTCA | 6329–6309* | This study |
| GI.3, GI.9, GII.6 | GV7 | – | IATCATCTCYTTRTCATG | 7108–7091* | 15 |
| GI.9 | NAF(I.9) | + | GAGATT GCT GCTGATGTTGTATAC | 5947–5970* | This study |
| GI.9 | NBR(I.9) | – | GAACTCTAGAGTTAGACATAAC | 6151–6130* | This study |
| GI.9 | NCR(I.9) | – | CTTGCAACACAACTAGCTGAC | 6272–6252* | This study |
| GI.9 | NEF(I.9) | + | GTCTAACTCTAGAGTTCCATC | 6135–6156* | This study |
| GI.9 | NGR(I.9) | – | CAAATTCTTGTGGAAGCAGAC | 6800–6780* | This study |
| GII.4 | N1 | + | GAA TTC CAT CGC CCA CTG GCT | 4495–4515† | 14 |
| GII.4 | G2polF | + | TGG AYT TTT AYG TGC CCA G | 4983–5001† | 36 |
| GII.4 | G2polR | – | CGA CGC CAT CTT CAT TCA C | 5099–5081† | 36 |
| GII.4 | GIICapRC | – | GRT TRA CCC ARG ANT CAA A | 6648–6630† | 15 |
| GII.4 | G2F1(II.4) | + | CCT ACA GTT GAG TCA AGA ACT A | 5736–5757† | This study |
| GII.4 | G2F3(II.4) | + | CCA ACT GTC TCC TGT CAA C | 5912–5930† | This study |
| GII.4 | FX1(II.4) | + | CAG AAT GTA CAA TGG TTA TGC A | 5363–5384† | This study |
| GII.4 | R2(II.4) | – | GAA GGT GCA GAT GTT GAC A | 5942–5924† | This study |
| GII.4 | G2SKR | – | CCR CCN GCA TRH CCR TTR TAC AT | 5389–5367† | 24 |
| GII.6 | GV4 | + | AGA IIT IAG CAC ITG GGA GGG C | 5035–5056† | 15 |
| GII.6 | GIICapC | – | CCT TYC CAK WTC CCA YGG | 6684–6667† | 46 |
Cloning studies were carried out to control sequence major capsid-coding regions of new strains and to confirm mixed NoV GI infections. A TOPO TA cloning kit was used according to manufacturer's instructions (Invitrogen, Carlsbad, CA). Ten clones were picked for control sequencing of colonies from each of the six patient samples amplified.
Phylogenetic and comparative sequence analysis.
Phylogenetic analysis included neighbor-joining methods using MacVector 7.2 software (Accelrys, Inc., San Diego, CA) with bootstrapping (1,000 replicates) to assess the reliability of branch nodes. Sequences were analyzed with Sequencher v4.9 and the BLAST tool (NCBI).
Prevalence of NoV infections in upstream municipalities.
Laboratory records at Sahlgrenska University Hospital Virus Laboratory, Gothenburg, Sweden, were examined to assess prevalence of NoV infections in Lilla Edet, Trollhättan, and Vänersborg during the 3 months prior to the outbreak.
Bacterial and protozoan pathogens.
Feces were examined for Campylobacter spp., Salmonella spp., Shigella spp., Yersinia enterocolitica, enterotoxic Escherichia coli, hemolytic enteropathogenic E. coli. Entamoeba histolytica, Giardia lamblia, and Cryptosporidium parvum (43).
Nucleotide sequence accession numbers.
The sequences described in the present study have GenBank accession numbers GU296356, JN183159 to JN183166, and JN603244 to JN603274.
RESULTS
Outbreak investigations. (i) The outbreak.
Indication of a large number of ill individuals complaining of acute gastroenteritis was reported to the Department of Communicable Disease Control and Prevention on 11 September 2008. “Boil water” orders were issued on the same day (27). Analysis of questionnaire responses indicated that approximately 2,400 individuals fell ill with acute gastroenteritis during the outbreak that had its peak on 9 and 10 September 2008 (Fig. 1). Cases were not restricted to a single region of the town or to the intake of a particular food, such as bivalves, but symptoms were associated with consumption of tap water from Lilla Edet municipal supply (27). A reduction in cases following the prompt issue of “boil water” orders strengthened the epidemiology of a waterborne outbreak.
Fig 1.
Date of onset of gastrointestinal symptoms for 379 inhabitants (dark gray bars) in Lilla Edet waterborne outbreak obtained through a questionnaire survey. NoV positive findings (black bars) in fecal samples (light gray bars) collected on the given date. (Modified with permission from Larsson and Ekvall [27].)
(ii) Water examination.
NoVs were not detected in raw or drinking water sampled during the outbreak, or in iced water prepared before outbreak onset, on examination in another laboratory (27). Somatic coliphages, indirect indicators of human or animal fecal contamination, were detected in raw and drinking-water samples collected on 1 day.
(iii) River conditions.
On 2 September, Lilla Edet WTP authorities were notified of emergency repairs 25 km upstream at the Trollhättan WWTP, with discharges to a tributary entering River Göta 8.5 km upstream of Lilla Edet. Repairs were completed on 4 September, but heavy rainfall caused sanitary overflows, which were maximal at the Trollhättan WWTP on 6 September, with overflows from Lilla Edet WWTP entering the river 500 m upstream of the Lilla Edet WTP. Marked river water turbidity was detected downstream of Lilla Edet on 6 September, with E. coli at a 1,100 most probable number (MPN)/100 ml and at a 1,200 MPN/100 ml recorded 5 and 40 km downstream, respectively, on 8 September (42).
Clinical samples.
Feces samples were obtained from 50 patients with gastroenteric symptoms attending the medical center between 12 and 26 September. The collection date is shown in the modified epicurve (Fig. 1), prepared from the date of onset of symptoms described by 379 affected individuals in the questionnaire survey (27). A peak in sample collection (Fig. 1) indicated a prompt response to radio calls made on 14 September.
TaqMan real-time hydrolysis probe detection.
Enteric viruses were detected in 36 of 50 patient samples examined in real-time TaqMan systems: NoV GI strains in 31, NoV GII in three, SaV in one, RoV in three, and HuAdV in one (Table 2). All age groups were affected, which is typical of NoV outbreaks. The predominance of NoV GI infections was reported to the Medical Officer of Health within 4 h of receipt of the first five samples. Detection of NoV according to sample date is shown in Fig. 1. Real-time systems provided direct evidence of mixtures of viral agents in three adult patients (Table 2). Patient 5 (P5) showed a dual infection with NoV GI (CT = 25) and NoV GII (CT = 33). P12 was positive for NoV GI (CT = 26) and RoV (CT = 33); P50 was positive for NoV GI (CT = 29) and reacted weakly with HuAdV (CT = 39). Viral loads presented in Table 2 were estimated from CT values and the copy number per gram of feces.
Table 2.
Real-time PCR detection of enteric virus in patient samples and sequence-based genotyping of NoV strains
| Patient | Age (yr)/gender | Real-time virus PCR | CTa (virus copy no.) | NoV genotypeb | Sequence |
|---|---|---|---|---|---|
| P1 | 27/M | NoV GI | 31.2 (8.3 × 105) | NoV GI.4 | S1 |
| P2 | 57/F | NoV GI | 27.4 (1.1 × 107) | NoV GI.9c | S2 |
| P3 | 61/M | NoV GI | 27.3 (1.2 × 107) | NoV GI.3 | S3 |
| NoV GI.4 | S3b | ||||
| P4 | 52/F | NoV GI | 30.1 (1.7 × 106) | NoV GI.7 | S4 |
| P5 | 40/F | NoV GI | 25.4 (4.4 × 107) | NoV GI.4 | S5 |
| NoV GI.7 | S5b | ||||
| NoV GII | 33.3 (3.0 × 105) | NoV GII.6 | S5c | ||
| P6 | 58/F | NoV GI | 32.6 (3.1 × 105) | NoV GI.4 | S6 |
| P7d | 66/M | NoV GI | 30.2 (1.6 × 106) | NoV GI.9c | S7 |
| NoV GI.4 | S7b | ||||
| P8 | 58/F | NoV GI | 32.5 (3.4 × 105) | NoV GI.7 | S8 |
| P9 | 53/F | NoV GI | 32.3 (3.9 × 105) | NoV GI.4 | S9 |
| P11 | 10/F | NoV GI | 36.5 (2.1 × 104) | NoV GI.4 | S11 |
| P12 | 55/M | NoV GI | 25.7 (3.7 × 107) | NoV GI.4 | S12 |
| RoV | 33.3 (2.9 × 105) | ||||
| P13 | 5/M | SaV | 25.6 (3.9 × 107) | SaV GII.2 | S13 |
| P14 | 33/F | NoV GI | 28.2 (6.4 × 106) | NoV GI.4 | S14 |
| P15 | 3/M | RoV | 26.0 (2.9 × 107) | ||
| P16 | 59/M | NoV GI | 32.5 (3.4 × 105) | NoV GI.4 | S16 |
| P17 | 12/M | NoV GI | 38.5 (5.9 × 103) | NoV GI.7 | S17 |
| P18 | 8/F | NoV GII | 30.6 (1.4 × 106) | NoV GII.6 | S18 |
| P19 | 1/F | RoV | 31.1 (9.0 × 105) | ||
| P20 | 40/F | NoV GI | 29.3 (3.2 × 106) | NoV GI.4 | S20 |
| P21 | 44/F | NoV GI | 29.5 (2.7 × 106) | NoV GI.7 | S21 |
| P23 | 21/M | NoV GI | 35.6 (4.2 × 104) | NoV GI.7 | S23 |
| P24 | 24/F | NoV GI | 29.5 (2.7 × 106) | NoV GI.7 | S24 |
| P25 | 91/F | NoV GI | 25.5 (4.2 × 107) | NoV GI.4 | S25 |
| P26 | 92/F | NoV GI | 29.0 (3.8 × 106) | NoV GI.3 | S26 |
| P29 | 91/F | NoV GI | 28.6 (4.8 × 106) | NoV GI.3 | S29 |
| P32 | 37/M | NoV GI | 28.7 (4.6 × 106) | NoV GI.4 | S32 |
| P33 | 37/F | NoV GI | 35.6 (4.2 × 104) | NoV GI.7 | S33 |
| NoV GI.4 | S33b | ||||
| P35 | 61/F | NoV GI | 34.5 (8.7 × 104) | NoV GI.4 | S35 |
| P37 | 64/M | NoV GI | 33.7 (1.5 × 105) | NoV GI.4 | S37 |
| P40 | 42/F | NoV GI | 38.6 (5.7 × 103) | NoV GI.4 | S40 |
| P43 | 53/F | NoV GII | 32.2 (4.3 × 105) | NoV GII.4 (2006b) | S43 |
| P44 | 58/F | NoV GI | 36.6 (1.9 × 104) | NoV GI.4 | S44 |
| P45 | 79/M | NoV GI | 35.5 (4.5 × 104) | NoV GI.4 | S45 |
| P47 | 8/M | NoV GI | 30.4 (1.4 × 106) | NoV GI.7 | S47 |
| P48 | 64/F | NoV GI | 28.2 (6.4 × 106) | NoV GI.9c | S48 |
| NoV GI.4 | S48b | ||||
| P50 | 37/F | NoV GI | 29.1 (3.7 × 106) | NoV GI.4 | S50 |
| HuAdV | 38.5 (7.0 × 103) | HuAdV type 2 |
CT, cycle threshold value. The virus copy number per gram of feces is given in parentheses.
Based on RNA-dependent RNA polymerase-N/S capsid sequencing.
New NoV genotype, proposed as GI.9.
Mixed infection with Campylobacter sp. and NoV GI strains.
Comparative sequence analysis and cloning studies.
NoV strains were characterized by sequence analyses of RdRp-N/S capsid-coding regions (∼900 nt) and aligned with the GenBank reference sequences for NoV GI.1 to GI.8 and NoV GII.1 to GII.17 (49). Most of the patient strains assembled with the NoV GI reference sequences. A total of 21 sequences were classified as NoV GI.4 strains, 9 were classified as GI.7 strains, and 3 (S3, S26, and S29) grouped as GI.3 strains (49). However, the nucleotide dissimilarity between S3, S26, and S29 sequences and GI.3 reference strains (U04469, AY038598, AF414403, AF414405, AF145709, and AF439267) increased from 15% to a limiting value of 20 to 23% when longer sequences (2.1 kb) covering hypervariable capsid-coding regions were compared. The remaining sequences S2, S7, and S48 assembled into a distinct group showing high dissimilarity (>22% nucleotides) from reference sequences, indicating the detection of a new genotype.
Long-fragment analysis (2.1 kb) of the unassigned group S2, S7, and S48, using the BLAST tool (NCBI), showed that these three sequences had high nucleotide similarity (96 to 97%) to patient strains detected in France (EF529736 and EF529737; coverage, 40%), Canada (EF078286; coverage, 38%), and Japan (AB112132 to AB112138). Japanese strains were named NoV GI.13 based on 558 nt of the more conserved NoV RdRp-N/S capsid-coding region covering 25% of the 2.1 kb reported here (23).
Sequences S3, S26, and S29 (2.1 kb) assigned to GI.3 were most similar (98% similarity, ca. 36% coverage) to unpublished patient strains detected in France (EF529738 to EF529740) and to patient strains from shellfish outbreaks (AB112100 and AB112127), named GI.14 by Japanese workers upon analysis of 555 nt of the RdRp-N/S capsid-coding region (23). NoV GI.7 strains detected in Lilla Edet patients differed by 8 to 11% nucleotides from all previously reported GI.7 strains found in GenBank, including patient strains AJ277609 and AB112122 and strain AB504701 from sewage in Japan.
Three strains of NoV GII were identified: NoV GII.4 (S43) and GII.6 (S5c and S18) (Table 2). S5c occurred as a mixed infection with NoV GI.4 and GI.7. SaV GII.2 was detected in P13, a 5-year-old child.
Control sequencing of clones prepared from amplicons of major capsid-coding regions confirmed the nucleotide sequences of the proposed new genotype strains (n = 3), proposed as NoV GI.9, and the mixtures of NoV strains detected in five patient samples (Table 2).
Phylogenetic analysis.
Phylogenetic trees were constructed using neighbor-joining methods based on predicted capsid amino acid sequences (530 to 550 amino acids [aa]) from 10 representative NoV strains detected in patient feces. The diverse strains clustered with reference genotypes as shown in the tree (Fig. 2), confirming the comparative sequence analyses. Lilla Edet strains were of genotypes NoV GI.3, GI.4, and GI.7 and three sequences of a new genotype cluster confirmed by cloning-based sequence studies. Definition of the new genotype, proposed as GI.9, was determined by full capsid amino acid analysis and 26% amino acid dissimilarity from other reference strains. This value (i.e., 26%) is much higher than any pairwise distances (0 to 14.1%) found between strains of the same genotype (49). Notably, Lilla Edet GI.3 strains formed a separate sub-branch showing 13 to 15% amino acid dissimilarity to GenBank GI.3 reference strains.
Fig 2.
Neighbor-joining tree based on analysis of NoV capsid amino acids (approximately 530 aa, ORF2). “S” identifies individual patient strains. Genotypes are defined on branches beside the GenBank accession numbers of the reference strains. S2 and S48 clustered as a new genotype, proposed as NoV GI.9. Relevant bootstrap values (1,000 replicates) are indicated at the branch nodes, and the genetic distance per nucleotide/site is shown by the bar.
Prevalence of NoV in Lilla Edet and upstream municipalities.
There was no epidemiological evidence of NoV infections in the region in the 3 months prior to the outbreak (27). Of 23 NoV-positive samples detected in Västra Götaland over the study period, one was from upper River Göta region, a child with NoV GII gastroenteritis in Trollhättan 80 days prior to the outbreak.
Bacterial and parasitic pathogens.
Campylobacter spp. were isolated from 2 of 50 feces samples examined. NoV GI.4 and NoV GI.9 strains were detected in one campylobacter-positive sample (P7).
DISCUSSION
This report describes the virological investigation of an extensive nonseasonal community outbreak of acute nonbacterial gastroenteritis with strong epidemiologic indices of waterborne contamination of a municipal water supply. The marked diversity of NoV GI strains detected in patient feces raises pertinent questions on the potential of molecular methods to improve clinical detection and early recognition of NoV infections of waterborne origin.
The Lilla Edet outbreak, in September 2008, was characterized by the detection of diverse NoV GI strains in patient feces. Sequence analyses of strains from the first five patient samples gave an early indication of marked genomic diversity, including NoV GI.3, GI.4, and GI.7 and a new genotype cluster confirmed by cloning studies (Table 2, Fig. 2). The NoV GI.4 and I.7 strains showed high nucleotide similarity to GenBank reference strains, but the Lilla Edet GI.3 strains formed a separate sub-branch showing 13 to 15% capsid amino acid dissimilarity with GI.3 reference strains. Three strains clustered in a new genotype, proposed here as GI.9, based on 26% capsid amino acid dissimilarity (530 to 550 aa) from established genotypes, although NoV genotype classification is unresolved (49). Predominance of diverse NoV GI strains in patients of all ages proved a distinctive feature of Lilla Edet outbreak, an extensive outbreak with epidemiologic indices of waterborne infection despite absence of direct molecular evidence of NoV contamination of tap water (27).
If, as in the World Health Organization definition of an outbreak, detection of two or more individual cases of NoV infection and further recognition of a cluster of sick individuals all using a common water source is required before water can be recognized as the point source of contamination in a NoV outbreak, then the Lilla Edet outbreak satisfies that premise (1). An untimely combination of heavy rain, emergency repairs, discharges from upstream WWTPs, and suboptimal chemical disinfection were important factors in contamination incidents associated with the Lilla Edet waterborne outbreak (17, 27). The most serious risk incidents in water treatment relate to suboptimal particle removal or disinfection treatment failure (48). Heavy rainfall with release of enteric pathogens to recipient waters is a known hazard of sewage overflows (33), and NoVs have been recorded in River Göta waters following early releases from northern WWTPs in rainy conditions (2). There was no epidemiological evidence of NoV infections in the region in the 3 months prior to the outbreak, although sporadic infections or “underreporting” of NoV infections cannot be excluded. Although the significance of asymptomatic shedding in outbreak settings has not yet been clarified (40), these are important aspects of NoV epidemiology since NoV concentration in wastewaters may reflect asymptomatic excretion, as well as disease prevalence. There is concern that recycled drinking water may cause endemic low-level gastrointestinal infections (48).
Vigilant monitoring of recycled river water quality and good communications are essential to reducing the risk of human fecal contamination at raw-water intakes, particularly since secondary treatment, the most common wastewater processing used in Sweden, can reduce viral load but does not remove NoV from wastewater (8). Although both NoV GI and GII strains are detected in effluents, GI strains appear to be more resistant to WWTP processing and show stability in aquatic environments (8, 36), which may account for the predominance of NoV GI strains detected in Lilla Edet patients. Raised proline content in NoV capsid protein P domain, water channels, and pre-stress effects induced by protruding domains may confer stability on capsid structure and contribute to the persistence of NoV in recycled river waters (3, 5, 7).
NoV diversity detected in patient feces raises important questions. Can strain diversity be used as a significant marker in the identification of a cluster of patients affected by NoV infection associated with waterborne contamination? Supporting evidence is provided by previous reports of multiple NoV GI and GII strains detected in patient outbreaks traced to the ingestion of contaminated water or sewage-contaminated bivalves (13, 25, 29). In contrast, the Lilla Edet investigations were based on real-time molecular methods targeting the highly conserved ORF1/2 junction of the genome (22). These techniques have greatly improved the speed and range of viral study in clinical and environmental settings, extending our awareness of NoV genomic diversity. “Coexistence” of multiple NoV strains in Japanese patients was traced to shellfish ingestion, implying the consumption of sewage-contaminated bivalves (23). This is relevant to our understanding of the genotype diversity detected in Lilla Edet outbreak, where infections were associated with the consumption of municipal water. Bivalves or other common foodstuffs were not implicated. Therefore, we consider that the diversity of NoV GI strains detected in Lilla Edet patients and the multiple strains noted in a few were findings consistent with NoV infections transmitted by waterborne contamination, strengthening epidemiological indices that human fecal contamination of the drinking-water supply was the outbreak point source.
Experience from the Lilla Edet outbreak emphasizes the importance of rapid diagnostic response to reports of gastroenteric infections from the public. We found the call through media for fecal samples from affected people to be an effective measure, but this step could have been initiated even earlier in the outbreak. NoV findings in the first five patient samples (Table 2) prompted rapid sequencing, followed by bioinformatics that revealed a multitude of human NoV strains suggestive of the contamination of drinking water by human feces. With outbreak preparedness, including the sampling, transport, and analysis of about 10 samples and the preliminary sequencing of informative “signature” regions of viral strains, we suggest that diagnostic results in similar norovirus outbreaks during optimal handling may be obtained within 3 days.
In conclusion, the diversity of NoV strains detected in Lilla Edet patients proved a significant indicator of waterborne infection derived from fecal contamination of a water supply. Greater awareness of the significance of NoV diversity in a cluster of patients should prompt questions of waterborne contamination, thereby improving identification and investigation of waterborne outbreaks. As RNA viruses with rapid replication and error-prone reverse transcription NoV exhibit remarkable genomic diversity that challenges molecular diagnostic methods, as shown in the present study, where real-time RT-PCR revealed strains with 26% amino acid dissimilarity to established genotypes, criteria defining a new genotype (49). The strain diversity detected in this waterborne outbreak emphasizes the importance of extended molecular surveillance of noncultivable NoVs.
ACKNOWLEDGMENTS
This study was supported by grants from the Swedish International Development Co-operation Agency, SIDA (SWE-2003-108k), the Swedish Council for Working Life and Social Research (2010-0895), and the LUA-ALF Foundation (grant 7346), Sahlgrenska University Hospital.
We thank Birgitta Bidefors, members of the Virus Detection Unit, Sahlgrenska University Hospital, Gothenburg, Sweden, and Margareta Thorhagen, Swedish Institute for Communicable Disease Control, Solna, Sweden, for skilled assistance. We also thank Olof Bergstedt, Chalmers University of Technology, Gothenburg, Sweden, and Kjell-Olof Hedlund, Swedish Institute for Communicable Disease Control, Solna, Sweden, for advice and discussion.
Footnotes
Published ahead of print 13 January 2012
REFERENCES
- 1. Andersson Y, Bohan P. 2001. Disease surveillance and waterborne outbreaks, p 115–133 In Fewtrell L, Bartram J. (ed), Water quality: guidelines, standards, and health. World Health Organization Water Series, London, United Kingdom [Google Scholar]
- 2. Åström J, Petterson S, Bergstedt O, Pettersson TJ, Stenström TA. 2007. Evaluation of the microbial risk reduction due to selective closure of the raw water intake before drinking water treatment. J. Water Health 5(Suppl 1):81–97 [DOI] [PubMed] [Google Scholar]
- 3. Baclayon M, et al. 2011. Pre-stress strengthens the shell of Norwalk virus nanoparticles. Nano Lett. 11:4865–4869 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Brunkard JM, et al. 2011. Surveillance for waterborne disease outbreaks associated with drinking water—United States, 2007–2008. MMWR Surveill. Summ. 60:38–75 [PubMed] [Google Scholar]
- 5. Cao S, et al. 2007. Structural basis for the recognition of blood group trisaccharides by norovirus. J. Virol. 81:5949–5957 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Caul EO. 1994. Small round structured viruses: airborne transmission and hospital control. Lancet 343:1240–1242 [DOI] [PubMed] [Google Scholar]
- 7. Choi J-M, Hutson AM, Estes MK, Prasad BV. 2008. Atomic resolution structural characterization of recognition of histo-blood group antigens by Norwalk virus. Proc. Natl. Acad. Sci. U. S. A. 105:9175–9180 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. da Silva AK, et al. 2007. Evaluation of removal of noroviruses during wastewater treatment, using real-time reverse transcription-PCR: different behaviors of genogroups I and II. Appl. Environ. Microbiol. 73:7891–7897 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. de Wit MA, et al. 2007. Large outbreak of norovirus: the baker who should have known better. J. Infect. 55:188–193 [DOI] [PubMed] [Google Scholar]
- 10. Duizer E, et al. 2004. Inactivation of caliciviruses. Appl. Environ. Microbiol. 70:4538–4543 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Duizer E, et al. 2004. Laboratory efforts to cultivate noroviruses. J. Gen. Virol. 85:79–87 [DOI] [PubMed] [Google Scholar]
- 12. Fankhauser RL, et al. 2002. Epidemiologic and molecular trends of “Norwalk-like viruses” associated with outbreaks of gastroenteritis in the United States. J. Infect. Dis. 186:1–7 [DOI] [PubMed] [Google Scholar]
- 13. Gallimore CI, Cheesbrough JS, Lamden K, Bingham C, Gray JJ. 2005. Multiple norovirus genotypes characterised from an oyster-associated outbreak of gastroenteritis. Int. J. Food Microbiol. 103:323–330 [DOI] [PubMed] [Google Scholar]
- 14. Green J, Gallimore CI, Norcott JP, Lewis D, Brown DW. 1995. Broadly reactive reverse transcriptase polymerase chain reaction for the diagnosis of SRSV-associated gastroenteritis. J. Med. Virol. 47:392–398 [DOI] [PubMed] [Google Scholar]
- 15. Green J, et al. 2000. Capsid protein diversity among Norwalk-like viruses. Virus Genes 20:227–236 [DOI] [PubMed] [Google Scholar]
- 16. Gustavsson L, Westin J, Andersson LM, Lindh M. 2011. Rectal swabs can be used for diagnosis of viral gastroenteritis with a multiple real-time PCR assay. J. Clin. Virol. 51:279–282 [DOI] [PubMed] [Google Scholar]
- 17. Hartlid C. 2009. Microbiological risk analysis of the municipal drinking-water supply in Lilla Edet: possible causes of a waterborne disease outbreak. Masters thesis University of Gothenburg, Gothenburg, Sweden: (In Swedish [summary in English].) [Google Scholar]
- 18. Heim A, Ebnet C, Harste G, Pring-Akerblom P. 2003. Rapid and quantitative detection of human adenovirus DNA by real-time PCR. J. Med. Virol. 70:228–239 [DOI] [PubMed] [Google Scholar]
- 19. Hohne M, Schreier E. 2004. Detection and characterization of norovirus outbreaks in Germany: application of a one-tube RT-PCR using a fluorogenic real-time detection system. J. Med. Virol. 72:312–319 [DOI] [PubMed] [Google Scholar]
- 20. Hrudey SE, Payment P, Huck PM, Gillham RW, Hrudey EJ. 2003. A fatal waterborne disease epidemic in Walkerton, Ontario: comparison with other waterborne outbreaks in the developed world. Water Sci. Technol. 47:7–14 [PubMed] [Google Scholar]
- 21. Jiang X, et al. 1999. Design and evaluation of a primer pair that detects both Norwalk- and Sapporo-like caliciviruses by RT-PCR. J. Virol. Methods 83:145–154 [DOI] [PubMed] [Google Scholar]
- 22. Kageyama T, et al. 2003. Broadly reactive and highly sensitive assay for Norwalk-like viruses based on real-time quantitative reverse transcription-PCR. J. Clin. Microbiol. 41:1548–1557 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Kageyama T, et al. 2004. Coexistence of multiple genotypes, including newly identified genotypes, in outbreaks of gastroenteritis due to Norovirus in Japan. J. Clin. Microbiol. 42:2988–2995 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Kojima S, et al. 2002. Genogroup-specific PCR primers for detection of Norwalk-like viruses. J. Virol. Methods 100:107–114 [DOI] [PubMed] [Google Scholar]
- 25. Kukkula M, Maunula L, Silvennoinen E, von Bonsdorff CH. 1999. Outbreak of viral gastroenteritis due to drinking water contaminated by Norwalk-like viruses. J. Infect. Dis. 180:1771–1776 [DOI] [PubMed] [Google Scholar]
- 26. Kwok S, Higuchi R. 1989. Avoiding false positives with PCR. Nature 339:237–238 [DOI] [PubMed] [Google Scholar]
- 27. Larsson C, Ekvall A. 2009. Report on Lilla Edet waterborne disease outbreak. Svenskt. Vatten 3:16–17 (In Swedish.) [Google Scholar]
- 28. Laursen E, Mygind O, Rasmussen B, Ronne T. 1994. Gastroenteritis: a waterborne outbreak affecting 1600 people in a small Danish town. J. Epidemiol. Community Health 48:453–458 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Le Guyader FS, et al. 2006. Detection of multiple noroviruses associated with an international gastroenteritis outbreak linked to oyster consumption. J. Clin. Microbiol. 44:3878–3882 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Liu BL, Clarke IN, Caul EO, Lambden PR. 1995. Human enteric caliciviruses have a unique genome structure and are distinct from the Norwalk-like viruses. Arch. Virol. 140:1345–1356 [DOI] [PubMed] [Google Scholar]
- 31. Lodder WJ, et al. 1999. Molecular detection of Norwalk-like caliciviruses in sewage. Appl. Environ. Microbiol. 65:5624–5627 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Lopman BA, Adak GK, Reacher MH, Brown DW. 2003. Two epidemiologic patterns of norovirus outbreaks: surveillance in England and Wales, 1992–2000. Emerg. Infect. Dis. 9:71–77 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Marsalek J, Rochfort Q. 2004. Urban wet-weather flows: sources of fecal contamination impacting on recreational waters and threatening drinking-water sources. J. Toxicol. Environ. Health A 67:1765–1777 [DOI] [PubMed] [Google Scholar]
- 34. Maunula L, Von Bonsdorff CH. 2005. Norovirus genotypes causing gastroenteritis outbreaks in Finland 1998–2002. J. Clin. Virol. 34:186–194 [DOI] [PubMed] [Google Scholar]
- 35. Nenonen NP, Hannoun C, Horal P, Hernroth B, Bergstrom T. 2008. Tracing of norovirus outbreak strains in mussels collected near sewage effluents. Appl. Environ. Microbiol. 74:2544–2549 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Nenonen NP, Hannoun C, Olsson MB, Bergstrom T. 2009. Molecular analysis of an oyster-related norovirus outbreak. J. Clin. Virol. 45:105–108 [DOI] [PubMed] [Google Scholar]
- 37. Nenonen NP, Hernroth B, Chauque AA, Hannoun C, Bergstrom T. 2006. Detection of hepatitis A virus genotype IB variants in clams from Maputo Bay, Mozambique. J. Med. Virol. 78:896–905 [DOI] [PubMed] [Google Scholar]
- 38. Noel JS, Fankhauser RL, Ando T, Monroe SS, Glass RI. 1999. Identification of a distinct common strain of “Norwalk-like viruses” having a global distribution. J. Infect. Dis. 179:1334–1344 [DOI] [PubMed] [Google Scholar]
- 39. Oka T, et al. 2006. Detection of human sapovirus by real-time reverse transcription-polymerase chain reaction. J. Med. Virol. 78:1347–1353 [DOI] [PubMed] [Google Scholar]
- 40. Phillips G, Tam CC, Rodrigues LC, Lopman B. 2010. Prevalence and characteristics of asymptomatic norovirus infection in the community in England. Epidemiol. Infect. 138:1454–1458 [DOI] [PubMed] [Google Scholar]
- 41. Reynolds KA, Mena KD, Gerba CP. 2008. Risk of waterborne illness via drinking water in the United States. Rev. Environ. Contam. Toxicol. 192:117–158 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. River Göta Watercourse Control Association 2008. Report of the River Göta Watercourse Control Association, p 29 River Göta Watercourse Control Association, Göta, Sweden: (In Swedish.) [Google Scholar]
- 43. Smittskyddsinstitutet 2002. Referensmetodik för laboratoriediagnostik vid klinisk mikrobiologiska laboratorier, 2nd ed, vol 1 Smittskyddsinstitut, Solna, Sweden [Google Scholar]
- 44. Teunis PF, et al. 2008. Norwalk virus: how infectious is it? J. Med. Virol. 80:1468–1476 [DOI] [PubMed] [Google Scholar]
- 45. Vennema H, de Bruin E, Koopmans M. 2002. Rational optimization of generic primers used for Norwalk-like virus detection by reverse transcriptase polymerase chain reaction. J. Clin. Virol. 25:233–235 [DOI] [PubMed] [Google Scholar]
- 46. Vinje J, Hamidjaja RA, Sobsey MD. 2004. Development and application of a capsid VP1 (region D) based reverse transcription PCR assay for genotyping of genogroup I and II noroviruses. J. Virol. Methods 116:109–117 [DOI] [PubMed] [Google Scholar]
- 47. Wang J, et al. 1994. Sequence diversity of small, round-structured viruses in the Norwalk virus group. J. Virol. 68:5982–5990 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Westrell T, Bergstedt O, Stenstrom TA, Ashbolt NJ. 2003. A theoretical approach to assess microbial risks due to failures in drinking water systems. Int. J. Environ. Health Res. 13:181–197 [DOI] [PubMed] [Google Scholar]
- 49. Zheng DP, et al. 2006. Norovirus classification and proposed strain nomenclature. Virology 346:312–323 [DOI] [PubMed] [Google Scholar]