SUMMARY
Sapoviruses cause acute gastroenteritis in humans and animals. They belong to the genus Sapovirus within the family Caliciviridae. They infect and cause disease in humans of all ages, in both sporadic cases and outbreaks. The clinical symptoms of sapovirus gastroenteritis are indistinguishable from those caused by noroviruses, so laboratory diagnosis is essential to identify the pathogen. Sapoviruses are highly diverse genetically and antigenically. Currently, reverse transcription-PCR (RT-PCR) assays are widely used for sapovirus detection from clinical specimens due to their high sensitivity and broad reactivity as well as the lack of sensitive assays for antigen detection or cell culture systems for the detection of infectious viruses. Sapoviruses were first discovered in 1976 by electron microscopy in diarrheic samples of humans. To date, sapoviruses have also been detected from several animals: pigs, mink, dogs, sea lions, and bats. In this review, we focus on genomic and antigenic features, molecular typing/classification, detection methods, and clinical and epidemiological profiles of human sapoviruses.
INTRODUCTION
Sapoviruses cause acute gastroenteritis in humans and animals. They belong to the genus Sapovirus within the family Caliciviridae. Sapovirus infections are a public health problem because they cause acute gastroenteritis in people of all ages in both outbreaks and sporadic cases worldwide. Outbreaks often occur in semiclosed settings. Outbreaks caused by foodborne transmission of sapovirus have also been reported. In this comprehensive review, we focus mainly on human sapoviruses.
HISTORY
Sapovirus particles are small (about 30 to 38 nm in diameter) and icosahedral and have cup-shaped depressions on the surface, which is a typical calicivirus morphology (Fig. 1) (1). Sapovirus particles were first detected in human diarrheic stool samples in 1976 in the United Kingdom using electron microscopy (EM) (2), and the virus was soon recognized as a new gastroenteritis pathogen (3 – 11). However, the prototype strain of the Sapovirus genus was from another outbreak in Sapporo, Japan, in 1982 (strain Hu/SaV/Sapporo/1982/JPN), because it has been studied extensively for sapovirus virological and genetic characteristics (10, 12 – 14).
Transmission electron micrographs of sapovirus and norovirus particles from clinical samples. Scale bars indicate 100 nm. (Courtesy of Yasutaka Yamashita, Ehime Prefectural Institute of Public Health and Environmental Science, Japan.)
TAXONOMY
Sapoviruses were previously called “typical human caliciviruses” or “Sapporo-like viruses.” In 2002, The International Committee on the Taxonomy of Viruses assigned these viruses to the species Sapporo virus, genus Sapovirus, in the family Caliciviridae (15). Currently, the family Caliciviridae consists of five established genera, Sapovirus, Norovirus, Lagovirus, Vesivirus, and Nebovirus (http://www.ictvonline.org/virusTaxonomy.asp), whereas five new genera (Bavovirus, Nacovirus, Recovirus, Valovirus, and Secalivirus) have been proposed (16 – 20).
CELL CULTURE AND ANIMAL INFECTION TRIALS
Attempts to grow human sapoviruses in cell cultures (2, 4, 5, 9, 21 – 24), have been reported, and two studies describe the propagation of sapoviruses in green monkey kidney cells (23) or primary human embryo kidney cells in the presence of trypsin and actinomycin D (Table 1) (24); however, no confirmed reproduction of these data is available. Currently, only a few porcine sapovirus strains have been grown successfully in primary porcine kidney cells or a porcine kidney cell line (i.e., LLC-PK1) in the presence of porcine intestinal contents or bile acids (Table 1) (25 – 28). Bile acids likely support porcine sapovirus replication via escape from endosomes during the virus entry step (29). Also, a cellular cyclic AMP (cAMP) signaling pathway induced by intestinal contents or bile acids likely causes downregulation of innate immunity (25, 30). An infection trial of human sapoviruses in mice did not succeed (5). Currently, only the specific porcine sapovirus (Cowden strain) has been studied for its pathogenesis in gnotobiotic pigs (31 – 33).
Cell culture trials for human and animal sapoviruses
PHYSICAL CHARACTERISTICS AND STABILITY
Sapovirus is a nonenveloped virus, and the virus has a buoyant density of 1.36 to 1.41 g/cm3 (5, 22, 34, 35). The stability of porcine sapovirus under physicochemical treatment is as follows: (i) stable at pH 3.0 to 8.0 at room temperature for 1 h, (ii) sensitive to ethanol treatment (60% and 70%) at room temperature for 30 s, (iii) inactivated by 200 mg/liter (or ppm) sodium hypochlorite at room temperature for 30 min, and (iv) inactivated by heating at 56°C for 2 h (36).
GENOMIC ORGANIZATION
The sapovirus genome has a positive-sense, single-stranded RNA genome, which is approximately 7.1 to 7.7 kb in size and has a 3′-end poly(A) tail. The sapovirus genome contains two open reading frames (ORFs) (Fig. 2). ORF1 encodes a large polyprotein containing the nonstructural proteins followed by the major capsid protein, VP1 (Fig. 2). ORF2 is predicted to encode the minor structural protein VP2 (Fig. 2) (28, 37). A similar genomic organization (i.e., two ORFs, with the first ORF encoding the nonstructural proteins and VP1) is found in other calicivirus genera, such as Lagovirus, Nebovirus, and the newly proposed genera Valovirus, Bavovirus, and Nacovirus (17, 19, 20, 38, 39). The genomic organization of Norovirus, Vesivirus, and Recovirus differs from that of Sapovirus: ORF1 encodes nonstructural proteins, and ORF2 and ORF3 encode structural proteins VP1 and VP2, respectively (18, 37, 40). A third ORF (ORF3) has been predicted in several human (12, 41 – 47) and bat (48) sapovirus strains; however, its function is unknown.
Diagram of sapovirus genomic organization and the RT-PCR target regions for human sapoviruses based on the GI.1 Manchester strain (GenBank accession no. X86560). A schematic diagram of the sapovirus genomic organization, including the putative subgenomic transcript, two common open reading frames (ORF1 and ORF2), the predicted viral nonstructural proteins (NS1, NS2, NS3 [NTPase], NS4, NS5 [VPg], and NS6-NS7 [protease–RNA-dependent RNA polymerase {RdRp}]), and structural proteins VP1 and VP2 is shown. The putative cleavage sites in the ORF polyprotein and their predicted sizes are indicated, according to previous reports (37, 41, 42, 49 – 51, 55, 272). Typical amino acid motifs for NS3 (GAPGIGKT), NS5 (KGKTK and DDEYDE), NS6 (GDCG), NS7 (WKGL, KDEL, DYSKWDST, GLPSG, and YGDD) and VP1 (PPG and GWS) are also shown. An overview of the RT-PCR target regions (RdRp region, RdRp-VP1 junction region, and VP1 region) is shown, and the detailed primer information is summarized in Tables 2 to 4. The putative first amino acids of VP1 from subgenomic transcript “MEG” and the cleavage site of this motif (ME/G; the slash indicates the cleavage site) from the ORF1 polyprotein are also shown with their nucleotide positions. The putative NS7 (RdRp) region and the cleavage site between RdRp and VP1 are based on previous reports (50, 52, 55, 58). The proposed subdomains in the sapovirus VP1 from the subgenomic transcript (N-terminal variable region [NVR] [1 to 43]), N-terminal region [N] [44 to 285]), central variable region [CVR] [286 to 441]), and C-terminal region [C] [442 to 561]) (47) are also indicated.
The ORF1-encoded polyprotein is expressed and processed into at least six nonstructural (NS) proteins (NS1, NS2, NS3, NS4, NS5, and NS6-NS7) and a structural protein (VP1) by virus-encoded protease (Fig. 2) (49 – 54). In vitro studies failed to show cleavage of the NS6-NS7 protein by the viral protease (28, 49, 50, 52, 55, 56), although both the NS6 and NS7 proteins can carry out their respective functions (proteolytic and polymerase) when expressed individually in vitro (56 – 58). The NS6-NS7 protein was also detected in porcine sapovirus-infected cells (28). Similar to the case for sapoviruses, vesivirus also produces the NS6-NS7 protein (fused protease-polymerase) (52, 53, 59 – 61), whereas noroviruses and lagoviruses produce an individual protease and polymerase, NS6 and NS7, respectively (51, 53, 62 – 67). The biological functions of the other sapovirus NS proteins have not been experimentally determined; however, NS3 and NS5 have a typical calicivirus NTPase motif (GAPGIGKT) and VPg motifs (KGKTK and DDEYDE), respectively (Fig. 2) (37, 49, 68, 69). VPg is linked to the 5′ end of the viral RNA and is critical for calicivirus genome replication, transcription, and translation (37, 70).
VP1, an approximately 60-kDa protein, is a major component of the complete virus (34, 35). Two mechanisms can be considered in the production of sapovirus VP1. One is that VP1 is cleaved from the ORF1-encoded polyprotein, and the other is that VP1 is translated from a subgenomic RNA (from the 3′-coterminal RNA corresponding to VP1 to the genome end region) (Fig. 2) (71, 72). A subgenomic RNA was confirmed for the sapovirus Cowden strain during replication (25). The VP2 protein has not yet been identified in sapovirus virions; however, the expression of this protein was detected in the in vitro translation products of a porcine sapovirus full-length genomic cDNA construct and from porcine sapovirus-infected cells (28). VP2 is predicted to be a strong basic protein and is identified as an interior component of the norovirus particles (73).
The expression of VP1 in insect or mammalian cells resulted in spontaneously assembled virus-like particles (VLPs) (12, 71, 72, 74 – 81). The sapovirus VLPs are morphologically and antigenically indistinguishable from those of the native sapovirus virions found in clinical specimens (12, 74). Digitized electron cryomicrographs of the human sapovirus VLPs revealed that the icosahedral capsid is formed from 180 molecules of VP1, the same as in norovirus (76). Sapovirus VP1 could be separated into several domains: the N-terminal variable region (NVR), N-terminal region (N), central variable region (CVR), and C-terminal region (C) (Fig. 2) (47). The conserved amino acid motif “GWS” was found in the predicted N and CVR junction (Fig. 2). The “G” in this motif is strictly conserved among caliciviruses (76). Norovirus VP1 has also been separated into several domains, the N-terminal domain, shell domain, and protruding (P) domain, which is further divided into P1 and P2 subdomains (76, 82, 83). The sapovirus VP1 CVR region likely corresponds to the highly variable P2 domain of norovirus VP1 (47, 76).
GENOMIC SEQUENCE AND ANTIGENICITY
The first complete genome of a sapovirus was determined for the Manchester strain detected in the United Kingdom in 1993 (Hu/Manchester/93/UK; GenBank accession no. X86560) (41, 42), which is closely related genetically to the prototype Sapporo strain (14). Thus far, 26 (21 from humans and five from animals [porcine and bat]) complete sapovirus genomes are available in GenBank (as of 1 September 2013). The VP1-encoding region is the most diverse region in the genome (84 – 86), and sapoviruses are divided into multiple genogroups based on complete VP1 sequences. Five genogroups (GI to GV) are recognized (46, 87), and nine additional genogroups (GVI to GXIV) were recently proposed (88). To date, human sapoviruses have been classified into four genogroups (GI, GII, GIV, and GV).
Distinct antigenicity among sapovirus strains has been demonstrated by using clinical specimens (9, 43, 89 – 91), recombinant VP1 proteins (77, 92), or virus-like particles (VLPs) (74, 77, 80, 81, 93). Antigenicity differs among GI, GII, GIV, and GV strains (93, 94) and is also distinct among different genotypes within GI and GII (80, 81, 94). These experimental results also support that VP1 determines sapovirus antigenicity. The antigenic differences between human and animal sapoviruses have not yet been determined.
MOLECULAR CHARACTERIZATION
Genogroups and GenotypesThe partial RNA-dependent RNA polymerase (RdRp) or partial VP1 region (Fig. 2) or both of these regions can be used to partially characterize detected sapoviruses, as well as to investigate the similarity of the detected sapovirus for epidemiological surveys. In contrast, the RdRp-VP1 junction region (Fig. 2) is too short for such sequence analysis.
For genetic classification of sapoviruses, VP1 sequences are widely used, because this region is more diverse than the RdRp region (45, 46) and the VP1 sequence correlates with virus phenotype (i.e., antigenicity) (43, 74, 77, 80, 81, 92 – 95). The International Calicivirus Conference Committee proposed that at least the entire VP1 region sequence is necessary to designate new genogroups or genotypes. We recently established a human sapovirus classification scheme based on the complete VP1 nucleotide sequences (87). In this review, we include newly available sapovirus strains and updated sapovirus genotype numbering along with our previous analytical methods and criteria (87). The frequency histogram with pairwise distance values of 59 representative complete capsid nucleotide sequences of GI, GII, GIII, GIV, and GV sapoviruses resulted in three clearly distinct and nonoverlapping peaks (0 to 0.159, 0.198 to 0.471, and 0.522 to 0.807) (Fig. 3A). These three peaks can be considered to represent the strain, genotype, and genogroup, respectively, as previously described (87). The mean values ± 3 standard deviations (SD) for the pairwise distance peaks were 0 to 0.151, 0.170 to 0.416, and 0.489 to 0.801, respectively (Fig. 3A), and the cutoff values for the genotype and genogroup clusters were designated ≤0.169 and ≤0.488, respectively. Based on the criteria, human sapovirus GI and GII were each subdivided into seven genotypes (GI.1 to GI.7 and GII.1 to GII.7). GIV was placed into a single genotype (GIV.1), and GV was subdivided into two genotypes (GV.1 and GV.2) (Fig. 4). GV also includes sapoviruses detected from pigs (GV.3) and sea lions (GV.4). As summarized in Fig. 4, genotype numbering based on the entire VP1 sequences is inconsistent among research groups for GI.5, GI.6, and GII.2 to GII.5 (46, 47, 87, 96). The phylogenetic tree pattern based on the 59 complete VP1 amino acid sequences is similar to that based on the nucleotide sequences (data not shown); however, the pairwise distance histogram showed only two major peaks (0 to 0.480 and 0.652 to 1.115), and we cannot define the genotype range statistically (Fig. 3B). This differs from the case for noroviruses, because genotypes could be defined statistically by both VP1 nucleotide and amino acid sequences (97).
Pairwise distance distribution histograms of 59 representative sapovirus complete VP1 sequences. The vertical dashed lines indicate the cutoff limits for genogrouping and/or genotyping. (A) Limits based on nucleotide sequences, with three peaks (0 to 0.159, 0.198 to 0.471, and 0.522 to 0.807) corresponding to the strain, genotype, and genogroup range, respectively. The mean values ± 3 SD for the pairwise distance peaks were 0 to 0.151, 0.170 to 0.416, and 0.489 to 0.801, and the cutoff values for the genotype and genogroup clusters were designated ≤0.169 and ≤0.488, respectively. (B) Limits based on amino acid sequences, with two major peaks (0 to 0.480 and 0.652 to 1.115) corresponding to the strain and genogroup range, respectively.
Genogroup and genotypes of GI, GII, GIII, GIV, and GV sapovirus strains based on complete VP1 nucleotide sequences. The phylogenetic tree is based on the complete VP1 nucleotide sequences (approximately 1,690 nt) of a total of 59 sapovirus strains (representing 58 sapovirus strains corresponding to all genotypes within GI, GII, GIV, and GV and one porcine strain representing GIII). The phylogenetic tree was constructed by the neighbor-joining method with 1,000 bootstrap replications using NJPlot software (http://pbil.univ-lyon1.fr/software/njplot.html) (273). The numbers on each branch indicate the bootstrap values of ≥950. The scale represents the nucleotide substitutions per site. Each sapovirus strain is indicated as genogroup/genotype-GenBank accession number-strain name (i.e., GI.1-U65427-Sapporo). Genotyping numbers are updated based on a recent classification scheme (87). Genotyping numbers from three other reports (46, 47, 96) are summarized for comparison.
Evolution and Emergence of Predominant Sapovirus StrainsGenogroup and genotype analysis is important to characterize the currently circulating sapoviruses in the population. Emergence of genetically similar sapoviruses in multiple countries in Europe (98) and dynamic changes of genogroups and genotypes in different years among gastroenteritis patients in the same geographical area in Japan have been reported (99 – 101). Interestingly, GIV.1 strains were detected predominantly in Japan, Canada, the United States, and Europe around 2007 (98, 101 – 104). The dynamic change of the detected sapovirus genogroups in 2007 was also identified by national surveillance through regional diagnostic labs network in Japan (http://www.nih.go.jp/niid/en/iasr-table/2784-iasrtve.html; see “IASR Tables Virus” “By Season” “Gastrointestinal Pathogens” “PDF”). This is a distinct trend compared to noroviruses, in which a specific genogroup and genotype (i.e., genogroup II and genotype 4 [GII.4]) have been predominant in the past decade in Japan (http://www.nih.go.jp/niid/en/iasr-table/2784-iasrtve.html) and in multiple other countries (105 – 108). In the case of norovirus GII.4, time-ordered genetic and antigenic change of VP1 was identified (109, 110). Recent studies from Europe reported similar time-ordered genetic change in the VP1 region of the sapovirus GI.2 strains (98), as reported for norovirus GII.4 strains (111 – 115).
Due to the inconsistent genotype numbering systems used by different research groups for GI.5, GI.6, and GII.2 to GII.5 (46, 47, 87, 96) (Fig. 4), it is important to indicate which numbering system was used for genotyping, and a harmonized genotype numbering system will facilitate comparison and exchange of information from sapovirus surveillance at national and international levels.
Recombinant StrainsSapoviruses with inconsistent grouping between the nonstructural protein-encoding region (including the RdRp region) and the VP1 encoding region have been designated “recombinant” or “chimeric” strains. Both intra- and intergenogroup recombinant strains have been reported (Fig. 5). All reported intergenogroup recombinant strains were GIV (based on VP1 sequence), whereas they were clustered together with GII strains in the RdRp region (46, 85, 102, 116, 117). Intragenogroup recombinant strains within GI (118 – 120), GII (84, 121, 122), and GIII (123) have been identified.
Comparative phylogenetic analysis of sapoviruses based on complete RdRp and VP1 nucleotide sequences. Phylogenetic trees of 27 sapovirus strains (10 GI, 5 GII, 5 GIV, and 6 GV strains and 1 representative GIII strain) whose sequences covering the putative complete RdRp to the end of VP1 had been available at GenBank (www.ncbi.nlm.nih.gov/) by 1 September 2013 are shown. The trees on the left and right are based on the nucleotide sequences of the putative complete RdRp region (approximately 1,570 nt) and the entire VP1 region (approximately 1,690 nt), respectively. The phylogenetic trees were constructed by the neighbor-joining method with 1,000 bootstrap replications using NJPlot software. The numbers on each branch indicate the bootstrap values of ≥950. The scale represents the nucleotide substitutions per site. The sapovirus strains showing inconsistent clustering patterns on the two trees are indicated within dashed boxes.
Recently, a norovirus classification scheme has been reported (97). The authors used nucleotide sequences of nearly complete RdRp (1,300 nucleotides [nt]) and both amino acid and nucleotide sequences of VP1 for classification (97). Although sapovirus RdRp (NS7) is fused with protease (NS6) (Fig. 2), we defined the putative complete RdRp-encoding region (1,566 nt) for sapoviruses (Fig. 2) based on our previous in vitro studies (52, 58). Among GI, GII, GIV, and GV sapoviruses, the nucleotide sequences spanning the putative complete RdRp- and VP1-encoding regions (Fig. 2) of 26 strains (10 GI, 5 GII, 5 GIV, and 6 GV based on VP1) were available in GenBank as of 1 September 2013. All of these strains and a representative GIII Cowden strain of pig origin were used for phylogenetic analysis based on the RdRp and VP1 regions. We found conserved amino acid motif “WKGL” (Fig. 2) at amino acid positions 12 to 15 in the putative compete RdRp (NS7) region among these sapovirus strains. As shown in Fig. 5, several strains clustered differently on the phylogenetic trees based on RdRp- and VP1-encoding regions. For example, based on the RdRp region, GII and GIV strains are not well separated, as discussed previously (85). The GII.2 Mc10 and GII.3 C12 strains also cluster together in the RdRp region and were previously reported as intragenogroup recombinant strains (84). The GV.4 strain clustered together with other GV strains (GV.1, GV.2, and GV.3) in the VP1 region, but it is separated from other GV strains in the RdRp region. However, the RdRp sequence-based classification is less reliable due to the fewer available sequences compared to the complete VP1 sequences. Further accumulation of sufficient sequence data spanning the complete or sufficient length of the RdRp- to the VP1-encoding regions for all the genogroups and genotypes are critical to provide a better understanding of “recombinant” or “chimeric” strains and to establish a reliable classification scheme for the sapovirus RdRp region in the future, because the putative complete RdRp sequence data for GI.3, GI.4, GI.6, GI.7, GII.4, GII.5, and GII.7 sapoviruses are not yet available (Fig. 5). In addition, it is also critical to amplify a single PCR fragment covering the partial RdRp- and VP1-encoding region for recombination analysis to avoid the possibility of coinfection of different genogroups and/or genotypes of sapovirus strains, as discussed previously (124).
LABORATORY DIAGNOSIS
Virus Particle DetectionSapoviruses are morphologically distinguishable from other gastroenteritis pathogens (e.g., norovirus, rotavirus, astrovirus, or adenovirus) by their typical “Star of David” surface morphology under the electron microscope (1, 33, 90, 125) (Fig. 1). However, this has low sensitivity compared to nucleic acid detection methods (116, 126 – 128, 130, 131).
Antigen Detection MethodsEnzyme-linked immunosorbent assays (ELISAs) have been developed for the detection of human sapovirus antigens (91, 93, 132, 190) and have been used for the detection of sapoviruses from clinical samples (43, 91, 132 – 135, 190). However, these assays are not widely used for diagnosis due to the difficulty in detection of antigenically diverse sapovirus strains, low sensitivity compared to nucleic acid detection methods (43, 91, 93), and current lack of commercial availability. The development of a broadly reactive ELISA or immunochromatography system for the detection of sapovirus antigens depends on the combination of a panel of genogroup/genotype-specific antisera and/or using broadly reactive monoclonal antibodies. These approaches may be feasible, because a common epitope(s) likely exists among GI, GII, GIV, and GV sapovirus strains (94). Broadly reactive norovirus-specific monoclonal antibodies that recognize VP1s of different genogroups of noroviruses were also reported (136 – 140).
Nucleic Acid Detection MethodsReverse transcription-PCR (RT-PCR), especially real-time RT-PCR, has become a major and routine method for sapovirus detection from clinical specimens (i.e., feces), because of its specificity, sensitivity, and broad reactivity. Numerous primers have been designed for the detection of human sapoviruses (Tables 2 to 4). These primers are designed to amplify the partial RdRp (44, 46, 141 – 144, 146 – 153), RdRp-VP1 junction (86, 154 – 162), or partial VP1 (126, 152, 163 – 166) region (Fig. 2). Due to the high genetic diversity of sapoviruses, most of the assays include multiple or degenerate primers (Tables 2 to 4). The primers targeting the conserved motifs of the RdRp region (e.g., p290 and p289 [Table 2]) also amplify other human gastroenteritis viruses (norovirus, rotavirus, and astrovirus) (144, 167, 168). Numerous primers with distinct names are quite similar, especially for RdRp-VP1 junction-targeting primer sets (Table 3).
Primer combinations targeting the RdRp region for human sapovirus detection
Primer and probe combinations targeting the RdRp-VP1 junction region for human sapovirus detection
Primer and probe combinations targeting the VP1 region for human sapovirus detection
Multiplex RT-PCR or PCR assays, whose products were differentiated by agarose gel electrophoresis (101, 159, 163, 169), real-time RT-PCR or PCR (156, 158, 162, 170), and a microsphere-based fluorescent PCR product detection assay (e.g., Luminex technology) (160, 166), have been reported for the detection of human sapoviruses together with other gastroenteritis viruses. Although these assays aimed for simultaneous detection of multiple viruses, it is unclear whether these assays can detect all genogroups of human sapoviruses.
Sapoviruses were also detected by specific primer-independent techniques (i.e., the metagenomic sequence approach) from untreated sewage (16), sewage sludge (171), and feces from California sea lions (172), dogs (173), and humans. These approaches are not widely used for diagnosis but may be applicable for routine clinical diagnosis in the future, when the cost of such assays and data analysis is comparable to that of traditional assays.
Selection of Detection MethodsThe nucleic acid detection method is more sensitive than EM (116, 126 – 128, 130, 131) or ELISA (91). Different detection rates among different PCR assays using the same panel of specimens (clinical specimens, environmental water, and shellfish) were reported (99 – 101, 174 – 177). Assays targeting the RdRp-VP1 junction region have the highest detection rate and can be used as the first choice for sapovirus screening from clinical specimens (101, 174). The VP1-targeting RT-PCR is preferred because the products can be sequenced for reliable genotyping (99 – 101). Similar results were reported for environmental water samples (i.e., river water) (175). RdRp-VP1 junction-targeting real-time RT-PCR was also used for the detection of sapoviruses from shellfish (178, 179): however, the nested RT-PCR targeting the partial VP1 region is superior to the real-time RT-PCR and single-round RT-PCR because of the low level of viral RNA in shellfish compared to clinical specimens (177, 178). Currently, limited primer sets (47, 86, 100, 165, 175, 180) have demonstrated the ability to detect all genogroups of human sapoviruses.
Full-Genome Sequencing ApproachesFull genomic sequence analysis is still not practical for routine diagnosis. A long single-round or nested RT-PCR to amplify a 2- to 2.5-kb PCR fragment to determine the complete VP1 sequences of various sapovirus strains from clinical specimens is feasible by using forward primers targeting the RdRp and/or RdRp-VP1 junction region (Tables 2 to 4) and a reverse primer hybridized to the 3′-end poly(A) tail (Fig. 2) (46, 47, 87, 130, 174, 181, 182). In contrast, the amplification of the 5′-end 5- to 5.5-kb fragment corresponding to the beginning of the genome to the VP1 upstream region is variable because of the lack of universal primers. As a new technology, the specific primer-independent metagenomic sequencing approach (i.e., next-generation sequencing techniques) can be used to determine the nearly complete genome sequences (lacking the 5′ end or both the 5′ and 3′ ends) from fecal specimens (172, 173). 5′ rapid amplification of cDNA ends (RACE) techniques (14, 41, 42) are still necessary to determine 5′ ends to obtain the complete sapovirus genomic sequences.
CLINICAL AND EPIDEMIOLOGICAL OBSERVATIONS
Symptoms and Severity of DiseaseBased on the epidemiological data from patients with sapovirus gastroenteritis, the incubation period ranges from less than 1 day to 4 days (5, 8, 44, 130, 135, 178, 183, 184). Major clinical symptoms include diarrhea and vomiting; however, additional constitutional symptoms (i.e., nausea, stomach/abdominal cramps, chills, headache, myalgia, or malaise) are also frequently reported. Similar to the case for norovirus illness, fever is a rare clinical symptom. Diarrhea usually resolves within 1 week (4, 5, 7 – 9, 44, 104, 117, 127, 135, 183, 185 – 189); however, individuals showing symptoms for a longer time (i.e., from over a week to up to 20 days) were also reported (9, 21, 127, 186, 188, 267). In general, the severity of sapovirus gastroenteritis is milder than that for rotavirus and norovirus (Table 5) (185, 186, 191). Gastroenteritis symptoms are usually self-limiting, and patients usually recover within a couple of days; however, the symptoms, severity, and duration of disease are dependent on the individual, and sapovirus infection sometimes leads to hospitalization (22, 152, 167, 193 – 209). Mortality is rare, but it was reported from outbreaks that occurred in a long-term-care facility for the elderly (104). Human noroviruses are associated with more serious clinical complications in susceptible groups (i.e., premature neonates and immunocompromised patients) (210 – 212). No such information is available for human sapoviruses, and this requires investigation in the future.
Reported clinical severity scores for sapovirus-, norovirus-, and rotavirus-associated gastroenteritis
Subclinical (asymptomatic) sapovirus infection was also detected (4, 6, 21, 134, 135, 213 – 216). Quantitative PCR analysis revealed that asymptomatic individuals also shed sapovirus in the feces at levels comparable to those shed by individuals with gastroenteritis (182, 183).
Shedding Levels and Patterns in FecesSapovirus shedding in feces may continue after symptoms disappear (1 to 4 weeks after onset of illness) (6, 22, 174, 191). Sapovirus shedding levels in clinical stool specimens range from 1.32 × 105 to 1.05 × 1011 genomic copies/gram of stool (80, 99, 101, 116, 117, 127 – 130, 174, 178, 181 – 183). Sapovirus RNA shedding levels in feces gradually decreased after onset of illness (174). During the prolonged excretion period (i.e., 25 days and 28 days after onset of illness) in some individuals in an outbreak, both synonymous and nonsynonymous nucleotide substitutions in the VP1-encoding region have been identified (174), and this is a possible mechanism for the generation of new variants of sapovirus in vivo. Similar to the case for noroviruses, sapoviruses were also detected from an immunocompromised patient who showed prolonged diarrhea (147 days) (217), although further studies with quantitative analysis are necessary.
Sporadic CasesSapoviruses are detected worldwide (i.e., in more than 35 countries), and more than 100 papers have described sapovirus detection from clinical specimens. Among them, 13 studies detected more than 30 sapovirus strains from patients with sporadic gastroenteritis (Table 6) (89, 99, 101, 102, 126, 131, 132, 165, 185, 191, 215, 216, 218 – 226). Although different methods (electron microscopy, ELISA, and PCR assays with different primer sets) were used in these studies, the sapovirus positive rates ranged from 2.2% to 12.7%. Eight studies also detected other gastroenteritis pathogens, and sapoviruses ranked second to fourth as the major viral pathogens among patients with sporadic gastroenteritis (Table 6). Similar to the case for noroviruses (101, 131, 185, 198), sapoviruses were detected mainly in the cold season among patients with sporadic gastroenteritis (89, 99 – 101, 131, 198, 225 – 227), although different seasonal peaks among years have also been reported (132, 185). Sapovirus illnesses occur more frequently in younger children than in older children and adults (131, 191, 216).
Sapovirus positive rates in gastroenteritis patients from 13 studies that detected more than 30 strains during the study period
OutbreaksAlthough the reported outbreak numbers are less for sapoviruses than for noroviruses (145, 228 – 230), sapovirus gastroenteritis outbreaks occur throughout the year in all ages of people in various settings, such as child day care centers, kindergartens, schools, colleges, hospitals, nursing homes, restaurants, hotels, wedding halls, and ships (3 – 7, 9, 80, 98, 103, 104, 116, 117, 127, 129, 130, 145, 174, 181, 182, 186, 188, 189, 228, 231 – 237). Suspected foodborne sapovirus outbreaks have also been reported (44, 145, 178, 230, 232, 238). The largest foodborne sapovirus outbreak (n = 665) has been reported in Japan in 2010 (183). An epidemiological investigation pointed to contaminated box lunches which were prepared by food handlers who were shedding sapovirus.
Data from four studies suggest that sapovirus caused 1.3 to 8.0% of the gastroenteritis outbreaks (Table 7) (98, 145, 228, 230), and data from the other three studies reported that sapovirus was detected in 5.9 to 22.6% of outbreak samples that tested negative for norovirus or both norovirus and pathogenic bacteria (103, 104, 234) (Table 7).
Sapovirus positive rates in gastroenteritis outbreaks
Coinfections of sapoviruses and multiple enteric viruses (e.g., noroviruses, rotaviruses, astroviruses, adenoviruses, enteroviruses, kobuviruses, etc.) have also been reported among acute gastroenteritis outbreaks (10, 178, 230, 231, 235 – 237, 239). Coinfections with different sapovirus strains (i.e., different genogroups/genotypes) were also identified from oyster/clam-associated gastroenteritis outbreaks (178, 236).
Sapoviruses in Seafood, Environmental Water, and AnimalsSapoviruses genetically indistinguishable (i.e., similar or identical based on partial virus genome sequences) from those detected in human clinical specimens have also been detected from shellfish (oysters and clams) (177, 178, 240, 241) and environmental water samples (river water and wastewater) (16, 171, 175, 176, 180, 241 – 246). These sapoviruses were likely viruses of human fecal origin that were discharged into environmental waters and accumulated in shellfish (i.e., oysters or clams). As evidence, sapoviruses were detected more frequently with higher viral RNA levels from environmental water samples (i.e., sewage and river water) in the cold season (175, 176, 180, 243, 247, 248), when the number of patients with sapovirus-associated sporadic gastroenteritis increased (89, 99 – 101, 131, 198, 225 – 227). In addition, similar sapovirus strains were detected from gastroenteritis patients, wastewater, and oysters, which were collected from geographically related areas in the same season (241). In contrast, sapoviruses genetically indistinguishable from those detected in human clinical specimens have not been discovered in other animals (i.e., swine, mink, bats, dogs, and sea lions) (48, 88, 123, 172, 173, 249 – 253). Based on complete VP1 sequences, GV.3 porcine sapoviruses are closest to human strains; however, they can be clearly separated into a different genotype (Fig. 4). These results suggest the existence of interspecies barriers among human and animal sapoviruses, although further epidemiological studies for other animals and experimental infection studies using human sapoviruses in various animals are necessary. Sapovirus contamination levels were ∼1.6 × 104 copies/g of digestive tissue in various types of shellfish (oyster, cockle, and smooth clam) (179), up to 1.3 × 105 copies/liter in wastewater treatment plant influent (248), and ∼1.3 × 109 copies/liter in untreated wastewater (247).
Transmission Route and Host SusceptibilityTransmission of sapovirus is through the fecal-oral route. Sapoviruses can be transmitted from person to person via contact with sapovirus-positive feces, vomitus, or sapovirus-contaminated materials/surfaces or via contaminated food and drinking water (44, 104, 129, 130, 145, 178, 182, 183, 231, 232, 236 – 239). These transmission routes are similar to those for norovirus (254), and sapovirus may also have a low infectious dose similar to that of norovirus (i.e., 1,015 to 2,800 genomic copies) (255, 256); however, similar volunteer studies are necessary to confirm this speculation for sapoviruses. No host genetic factors for susceptibility or resistance to human sapovirus infection and disease have been identified. Susceptibility to human sapoviruses is not associated with histo-blood group antigen (HBGA) phenotypes (214). In vitro data also support no binding of sapovirus to HBGAs (257, 258). This differs from the case for the prototype norovirus (Norwalk virus): certain HBGA phenotypes (e.g., nonsecretor) of an individual are clearly related to resistance to virus infection (259, 260). Other different genogroups/genotypes of norovirus strains can also bind to HBGAs (113, 261 – 263) but lack a clear relatedness between the HBGA phenotypes and resistance to infection (264 – 266). Sialic acids have recently been reported as binding factors for porcine sapovirus (258).
ImmunityThe serological responses to sapovirus infection were demonstrated by immune electron microscopy, ELISA, or radioimmunoassay using paired sera (i.e., acute- and convalescent-phase sera) with purified virus from clinical specimens (3, 4, 7, 22, 90, 190, 267, 268). The seroprevalence studies of human sapoviruses using purified virus or recombinant capsid proteins demonstrated a gradually increasing seroprevalence rate with age, and it reached a high level (>90%) in school-age children, and remained high (80 to 100%) in sera or pooled immunoglobulin collected from adults (92, 132, 190, 269 – 271). These results suggest that sapovirus infection is common during early childhood.
Protective immunity/resistance mechanisms to sapovirus infection at the putative primary infection site (e.g., intestinal lumen) remain to be clarified, but the presence of preexisting serum antibodies to sapoviruses was associated with reduced frequencies of sapovirus infection and illness, at least for antibodies to antigenically homologous sapoviruses (267). A similar phenomenon was also observed in gastroenteritis outbreaks that occurred in mother and baby units (7). Adults who had serum antibodies to antigenically indistinguishable human sapoviruses did not show any clinical symptoms on reinfection (7). Symptomatic reinfections with a different genogroup/genotype of sapovirus were recently reported in a study from Japan (99).
CONCLUSIONS AND FUTURE DIRECTIONS
Recent epidemiological studies with improved diagnostic assays have highlighted the impact of sapovirus-associated gastroenteritis. Genetically highly diverse sapovirus strains were identified through epidemiological surveillance studies. Continuous surveillance with a broadly reactive detection system(s) and molecular characterization will permit the identification of changes in major strains as well as the emergence of new strains and an understanding of the evolution of sapoviruses among humans and animals. However, in contrast to the significant improvement in sapovirus detection methods, the basic understanding of infection/replication sites, pathological changes in infected persons, immunological responses and protective immunity to sapovirus infections in humans, infectious dose, and stability in the environment remain unknown. To date, no vaccines or antiviral drugs are available for the control and prevention of human sapovirus infections. The mechanisms of virus binding and entry into target cells and viral RNA replication and translation are undefined, partially due to the lack of a cell culture system. Extensive studies of human sapoviruses in clinical cases, the use of the cell culture-adapted porcine sapovirus strain as a model, and establishment of a human sapovirus cell culture system will improve our knowledge of sapoviruses and may lead to more targeted control measures for prevention of sapovirus gastroenteritis in the future.
ACKNOWLEDGMENTS
We thank Yasutaka Yamashita, who kindly provided transmission EM pictures for human sapovirus and norovirus, and Kelly A. Scheuer and Susan Sommer-Wagner for their editing of the manuscript.
- Copyright © 2015, American Society for Microbiology. All Rights Reserved.
REFERENCES
- 1.↵
- 2.↵
- 3.↵
- 4.↵
- 5.↵
- 6.↵
- 7.↵
- 8.↵
- 9.↵
- 10.↵
- 11.↵
- 12.↵
- 13.↵
- 14.↵
- 15.↵
- 16.↵
- 17.↵
- 18.↵
- 19.↵
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.↵
- 25.↵
- 26.↵
- 27.↵
- 28.↵
- 29.↵
- 30.↵
- 31.↵
- 32.↵
- 33.↵
- 34.↵
- 35.↵
- 36.↵
- 37.↵
- 38.↵
- 39.↵
- 40.↵
- 41.↵
- 42.↵
- 43.↵
- 44.↵
- 45.↵
- 46.↵
- 47.↵
- 48.↵
- 49.↵
- 50.↵
- 51.↵
- 52.↵
- 53.↵
- 54.↵
- 55.↵
- 56.↵
- 57.↵
- 58.↵
- 59.↵
- 60.↵
- 61.↵
- 62.↵
- 63.↵
- 64.↵
- 65.↵
- 66.↵
- 67.↵
- 68.↵
- 69.↵
- 70.↵
- 71.↵
- 72.↵
- 73.↵
- 74.↵
- 75.↵
- 76.↵
- 77.↵
- 78.↵
- 79.↵
- 80.↵
- 81.↵
- 82.↵
- 83.↵
- 84.↵
- 85.↵
- 86.↵
- 87.↵
- 88.↵
- 89.↵
- 90.↵
- 91.↵
- 92.↵
- 93.↵
- 94.↵
- 95.↵
- 96.↵
- 97.↵
- 98.↵
- 99.↵
- 100.↵
- 101.↵
- 102.↵
- 103.↵
- 104.↵
- 105.↵
- 106.↵
- 107.↵
- 108.↵
- 109.↵
- 110.↵
- 111.↵
- 112.↵
- 113.↵
- 114.↵
- 115.↵
- 116.↵
- 117.↵
- 118.↵
- 119.↵
- 120.↵
- 121.↵
- 122.↵
- 123.↵
- 124.↵
- 125.↵
- 126.↵
- 127.↵
- 128.↵
- 129.↵
- 130.↵
- 131.↵
- 132.↵
- 133.↵
- 134.↵
- 135.↵
- 136.↵
- 137.↵
- 138.↵
- 139.↵
- 140.↵
- 141.↵
- 142.↵
- 143.↵
- 144.↵
- 145.↵
- 146.↵
- 147.↵
- 148.↵
- 149.↵
- 150.↵
- 151.↵
- 152.↵
- 153.↵
- 154.↵
- 155.↵
- 156.↵
- 157.↵
- 158.↵
- 159.↵
- 160.↵
- 161.↵
- 162.↵
- 163.↵
- 164.↵
- 165.↵
- 166.↵
- 167.↵
- 168.↵
- 169.↵
- 170.↵
- 171.↵
- 172.↵
- 173.↵
- 174.↵
- 175.↵
- 176.↵
- 177.↵
- 178.↵
- 179.↵
- 180.↵
- 181.↵
- 182.↵
- 183.↵
- 184.↵
- 185.↵
- 186.↵
- 187.↵
- 188.↵
- 189.↵
- 190.↵
- 191.↵
- 192.
- 193.↵
- 194.↵
- 195.↵
- 196.↵
- 197.↵
- 198.↵
- 199.↵
- 200.↵
- 201.↵
- 202.↵
- 203.↵
- 204.↵
- 205.↵
- 206.↵
- 207.↵
- 208.↵
- 209.↵
- 210.↵
- 211.↵
- 212.↵
- 213.↵
- 214.↵
- 215.↵
- 216.↵
- 217.↵
- 218.↵
- 219.↵
- 220.↵
- 221.↵
- 222.↵
- 223.↵
- 224.↵
- 225.↵
- 226.↵
- 227.↵
- 228.↵
- 229.↵
- 230.↵