A simple and novel method for GII norovirus genome clone with generic primers


  • L. Xue,

    1. School of Bioscience and Bioengineering, South China University of Technology, Guangzhou, Guangdong, China
    2. Guangdong Institute of Microbiology, State Key Laboratory of Applied Microbiology (Ministry-Guangdong Province Jointly Breeding Base), South China, Guangdong Provincial Key Laboratory of Microbial Culture Collection and Application, Guangdong Open Laboratory of Applied Microbiology, Guangzhou, Guangdong, China
    Search for more papers by this author
  • Q. Wu,

    Corresponding author
    1. Guangdong Institute of Microbiology, State Key Laboratory of Applied Microbiology (Ministry-Guangdong Province Jointly Breeding Base), South China, Guangdong Provincial Key Laboratory of Microbial Culture Collection and Application, Guangdong Open Laboratory of Applied Microbiology, Guangzhou, Guangdong, China
    • School of Bioscience and Bioengineering, South China University of Technology, Guangzhou, Guangdong, China
    Search for more papers by this author
  • X. Kou,

    1. Guangdong Institute of Microbiology, State Key Laboratory of Applied Microbiology (Ministry-Guangdong Province Jointly Breeding Base), South China, Guangdong Provincial Key Laboratory of Microbial Culture Collection and Application, Guangdong Open Laboratory of Applied Microbiology, Guangzhou, Guangdong, China
    Search for more papers by this author
  • J. Zhang,

    1. Guangdong Institute of Microbiology, State Key Laboratory of Applied Microbiology (Ministry-Guangdong Province Jointly Breeding Base), South China, Guangdong Provincial Key Laboratory of Microbial Culture Collection and Application, Guangdong Open Laboratory of Applied Microbiology, Guangzhou, Guangdong, China
    Search for more papers by this author
  • W. Guo

    1. Guangdong Institute of Microbiology, State Key Laboratory of Applied Microbiology (Ministry-Guangdong Province Jointly Breeding Base), South China, Guangdong Provincial Key Laboratory of Microbial Culture Collection and Application, Guangdong Open Laboratory of Applied Microbiology, Guangzhou, Guangdong, China
    Search for more papers by this author


Qingping Wu, Guangdong Institute of Microbiology, State Key Laboratory of Applied Microbiology (Ministry-Guangdong Province Jointly Breeding Base), South China Guangdong Provincial Key Laboratory of Microbial Culture Collection and Application, Guangdong Open Laboratory of Applied Microbiology, No. 100, Xianlie Zhong Road, Guangzhou 510070, China.

E-mail: wuqp203@yahoo.com.cn



This study aims to establish a novel method for cloning GII norovirus genome using generic primers rationally designed based on multiple alignments of 96 GII norovirus genome sequences.

Methods and Results

Based on conservative analysis of 96 GII norovirus genome sequences available in GenBank, three fragments encompassing the full-length genome were rationally designed. Fragments A, B and C were amplified by primers N1F/N2819R, N2689F/COG2R and COG2F/adaptor, respectively. Meanwhile, the sensitivity of the novel primers was evaluated, which could achieve 101 RTPCRU, as determined by the common detection primer pair JV12/JV13. The availability of the novel protocol was verified by sequencing two norovirus strains with different genotypes.


Primers for GII norovirus genome clone were rationally designed, and a novel GII genome clone method was established.

Significance and Impact of the Study

The three-fragment cloning method can be used as a universal tool to collect information on the genome of norovirus strains for future evolution and antivirus studies.


Noroviruses (NoVs) are the most important causative agents of acute epidemic and sporadic gastroenteritis among persons of all ages worldwide. More than 90% of nonbacterial gastroenteritis is caused by NoVs (Zheng et al. 2010). NoV infection generally includes a latent period of 24 h to 48 h, followed by active infection with typical gastroenteritis symptoms, such as stomach ache, diarrhoea and fever (Patel et al. 2009). NoVs are transmitted mainly by the faecal–oral pathway through food and water contaminated with a very low infectious dose (approx. 18 viral particles) (Teunis et al. 2008; Maunula and von Bonsdorff 2011). Thus, NoV outbreaks are frequently reported in hospitals, nursing homes, schools, restaurants and military bases.

NoVs belonging to the family Caliciviridae have a positive-sense, single-stranded RNA genome of 7.5–7.7 kb in size, including three open reading frames (ORFs) (Green et al. 2000). ORF1 encodes a polyprotein cleaved into six nonstructural proteins. ORF2 encodes a major structural protein VP1 composed of two domains, the shell domain (S) and the protruding domain (P). The latter domain is further divided into two subdomains, P1 and P2. ORF3 encodes a minor capsid protein VP2 (Glass et al. 2000), which is rich in basic amino acids and has been proposed to enhance viral stability and expression (Bertolotti-Ciarlet et al. 2003). According to the amino acid sequence of VP1, NoVs can be divided into five genogroups (GI to GV), of which GI, GII and GIV can infect humans (Zheng et al. 2006). GI and GII are further subdivided into 8 and 17 genotypes, respectively, and the numbers are still growing for NoV rapid evolution. Well, GII.4 is the predominant strain in more than 90% of disease cases globally, including China (Zeng et al. 2011, 2012).

Over the past three decades, many NoVs have been isolated from humans and animals, most of which are genetically and antigenically diverse. The error-prone RNA replication and recombination between viruses cause the great diversity of NoVs (White et al. 2007; Bull et al. 2010). Furthermore, the accumulated mutations of the P2 domain produce different GII.4 NoV variants. Genotyping based solely on a genome part, such as sequences of the major capsid or the RNA-dependent RNA polymerase, is not sufficient to characterize the recombinant NoVs (White et al. 2007). Thus, characterization of viruses based on analysis of the entire nucleotide sequence would be helpful for further accurate analyses.

Meanwhile, the lack of suitable reproduction and culture system in vitro hinder NoV studies. Complete genome studies are important in establishing phylogenetic and evolutionary relationships, which are also the core problems in antivirus research. The first full-length human NoV genome sequence obtained in 1990 greatly promoted the development of NoV molecular studies and detection methods. However, full-length genome sequences of strains in developing countries, including China, and even in industrial countries, except Japan, remain insufficient. Nowadays, some NoV genome cloning methods were developed, such as long RT-PCR of near-full-length genomes (Tellier et al. 2008) or segmentation PCR, in which NoV genome was cut into three or more fragments (Yun et al. 2010; Park et al. 2011; Shen et al. 2011). But these methods are often designed for special strains and require strict experimental conditions. Thus, the development of a universal method for collecting complete genome sequences is necessary for diagnostic, vaccine development and evolution mechanism studies.

This study introduces a three-fragment cloning method that uses three generic primer pairs for amplifying and cloning NoV GII sequences. The near-full-length genome sequences of two strains of different genotypes were determined using this novel method. The proposed method was proven as a simple accessible tool for collecting GII NoV genome sequences.

Materials and methods

Clinical samples

Two faecal specimens containing NoV strains GII.4 and GII.6 were collected from patients with acute gastroenteritis in Guangzhou, China. The specimens were converted to 10% (w/v) suspensions in phosphate-buffered saline (pH 7.3; treated with diethylpyrocarbonate) for RNA extraction or conservation at −80°C.

Reference genome sequences

For the rational design of primers for genome cloning, 96 NoV GII reference strains with full-length or near-full-length genome sequences were collected from GenBank. Their accession numbers are shown in Table 1.

Table 1. Description of the NoVs used for full-length genome-based genetic analyses
VirusGenotype aAccession no.VirusGenotypeAccession no.
  1. a

    All NoVs genome sequences were genotyped using the NoV automated genotyping tool (www.rivm.nl/mpf/norovirus/typingtool) (Kroneman et al. 2011).

Saitama U1GII.12 AB039775 Saitama U201GII.3 AB039782
Saitama U3GII.7/GII.6 AB039776 Hiroshima 1999GII.12 AB044366
Saitama U4GII.7/GII.6 AB039777 Gifu96GII.12 AB045603
Saitama U16GII.6 AB039778 YURIGII.22 AB083780
Saitama U17GII.6 AB039779 YURI 32073GII.4-1996 AB083781
Saitama U25GII.8 AB039780 GIFU99GII.7/GII.6 AB084071
Saitama U18GII.3 AB039781 Swine43GII.19/GII.11 AB126320
Chiba 2005GII.12/GII.4-2003 AB220921 SakaiGII.12/GII.4-2003 AB220922
Ehime 2005GII.12/GII.4-2003 AB220923 TCH2004GII.b/GII.3 AB365435
CamberwellGII.4-1987 AF145896 Vietnam 026GII.12/GII.10 AF504671
MD145GII.4-1987l AY032605 Snow Mountain virusGII.c/GII.2 AY134748
Mc37GII.12/GII.10 AY237415 Langen1061GII.4-2002 AY485642
CS-G12002GII.4-2002 AY502020 Farmington HillsGII.4-2002 AY502023
Oxford-B5S22GII.4-2002 AY581254 Oxfod-B4S2GII.4-2002 AY587983
Oxford-B4S5GII.4-2002 AY587984 Oxford-B4S6GII.4-2002 AY587985
Oxford-B4S4GII.4-2002 AY587986 Oxford-B4S7GII.4-2002 AY587987
Oxford-B4S1GII.4-2002 AY587988 Oxford-B4S16GII.4-2002 AY587989
Dresden174GII.4-1996 AY741811 Neustrelitz260GII.16 AY772730
HunterGII.4-2004 DQ078814 OsakaNI2004GII.d/GII.2 DQ366347
NVgz01GII.12 DQ369797 Carlow2002GII.4-2002 DQ415279
MK04GII.2 DQ456824 MD2004GII.4-2002 DQ658413
NZ327GII.4 EF187497 SK2002GII.4-2002 EF202567
SK2005GII.4/GII.4-2004 EF202568 Shanxi 2006GII.7/GII.14 EF670650
NSW696TGII.4-2006b EF684915 TCH186GII.4-2002CN EU310927
Leverkusen267GII.20 EU424333 PC15GII.4/GII.4-2004 EU921344
PC51GII.e/GII.4-2007 EU921388 PC52GII.b/GII.3 EU921389
CUK-3GII.4-2006b FJ514242 CHDC5191GII.1/GII.4 FJ537134
CHDC2094GII.4 FJ537135 CHDC3967GII.4/GII.4-1987 FJ537136
CHDC4108GII.4-1987/GII.4 FJ537137 CHDC4871GII.1/GII.4 FJ537138
NSW287RGII.4-2006b GQ845024 NSW3639GII.4-2006b GQ845366
NSW001PGII.4-2010 GQ845367 NSW505GGII.4-2001/GII.4 GQ845368
NSW390IGII.e/GII.4-2007 GQ845369 NSW199UGII.g/GII.12 GQ845370
Maizuru08GII.7/GII.14 GU017903 HS194GII.4-2006b GU325839
New Orleans 1805GII.4-2010 GU445325 Saga 2008GII.7/GII.14 GU594162
E99-13646GII.7/GII.6 GU930737 CBNU1GII.12/GII.3 GU980585
SH2GII.4-2006b GU991353 SH5GII.4-2006b GU991354
SH312GII.3 GU991355 NSW305PGII.4-2006b HM748971
NSW881ZGII.4-2010 HM748972 NSW892UGII.4-2008 HM748973
JB-15GII.4-2008 HQ009513 Ch6GII.11 HQ392821
HS210GII.g/GII.12 HQ449728 HS206GII.g/GII.12 HQ664990
New OrleansGII.4-2010 JN595867 NSW004PGII.g/GII.12 JQ613568
NSW895PGII.g/GII.12 JQ613569 NSW006DGII.4-2010 JQ613570
NSW817LGII.4-2010 JQ613571 NSW217IGII.4-2006b JQ613572
NSW295EGII.4-2010 JQ613573 CBNU2GII.4-2006b JQ622197
5MGII.4-2002 JQ798158 VNM10002GII.4-2006b JQ911595
VNM10003GII.4-2006b JQ911596 VNM10012GII.4-2006b JQ911597
VNM10037GII.4-2006b JQ911598 CHDC5191GII.1/GII4 JX023286
HarwaiiGII.1 U07611 Lordsdale virusGII.4-1987 X86557
NSW0514GII.4-2012 JX459908    

Primer design

For simplifying the amplification and cloning operation, three overlapping fragments with suitable length encompassing the full-length NoV genomic RNA were designed based on the multiple genome sequence alignments, including fragment A corresponding to nucleotides 1–2819, fragment B corresponding to nucleotides 2696–5100 and fragment C corresponding to nucleotides 5048–7547 and the 3-poly(A) tail. The primers were rationally designed and synthesized based on the result of NoV GII sequence alignment (Table 2). In addition, primers of COG2F/COG2R used as detective primers were selected for amplifying the second and third fragments (Kageyama et al. 2003).

Table 2. Primers designed for NoVs genomic RNA amplification
NamePrimer sequencePositionaPolarity
  1. a

    Location of the 5′ end of the primer in the nucleotide residue of Lordsdale virus (X86557).


RNA extraction

The viral RNA was extracted from 140 μl of 10% (w/v) faecal suspensions using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions, and the RNA pellet was resuspended in 50 μl of DNase-free and RNase-free water.

One-step RT-PCR

The first two overlapping fragments were amplified by one-step reverse transcriptase-polymerase chain reaction (RT-PCR) from the extracted viral RNA according to the manufacturer's instructions. Briefly, one-step RT-PCR mix (TaKaRa, Dalian, China) was performed in 20 μl of reaction mixture containing 0·8 μl of enzyme mix, 10 μl of 2 × RT-PCR buffer, 1.0 μl of 10·0 mmol l−1 primers, 2.0 μl of extracted RNA and 5.2 μl of RNase-free water. Primers of N1F/N2819R and N2696F/COG2R were used for amplifying fragments A and B, respectively. RT-PCR was carried out under the following reaction conditions: incubation at 50°C for 30 min; initial denaturation at 94°C for 3 min; 30 cycles of amplification at 94°C for 30 s, 54 C for 30 s and 72°C for 3 min; and a final extension at 72°C for 7 min. The detection primers of JV12 and JV13 were also used in this study. RT-PCR was conducted following previously described reaction conditions (Vinje and Koopmans 1996).

3′ rapid amplification of cDNA ends (3′ RACE)

The third fragment of the NoV genomic RNA was amplified by the 3′ RACE method (Frohman et al. 1988). In brief, oligo dT-adaptor (M13-M4) (TAKARA) was used as the reverse primer for the cDNA synthesis with AMV reverse transcriptase (TAKARA) in 10 μl of reaction mixture following the manufacturer's instructions. Subsequently, the RT reaction mixture was PCR-amplified with LA Taq (TAKARA) using the primers of M13-M4 and COG2F for 30 cycles of 30 s at 94°C, 30 s at 55°C and 3 min at 72°C, followed by a final extension at 72°C for 7 min.

Sensitivity comparison

Endpoint dilution RT-PCR was used to compare the sensitivity of JV12/JV13 and the new primers. To obtain the estimate of the endpoint detection limit, a tenfold dilution series of the RNA template was tested by RT-PCR in duplicate. The same RT-PCR mixture was made up, and the same RT-PCR thermocycle programme was performed. The detection limit of the primers was defined as the highest dilution (most diluted sample) in which a visible PCR amplicon could be observed by gel electrophoresis.

Clone and sequence

The RT-PCR products were analysed by 1% agarose gel electrophoresis in TAE buffer, followed by staining with GoldView and visualization under UV light. All amplicons (fragments A, B and C were approx. 2819, 2504 and 2500 bp, respectively) were then gel-extracted and cloned into the pMD19T simple vector (TAKARA). The recombinant plasmid was transformed into Escherichia coli DH5a competent cells (TAKARA). After confirmation by colony PCR using primers M13F and M13R, colonies were multiple randomly picked and sequenced by primer walking. Sequencing was performed on an automated sequencer (454GS FLX, Majorbio Co. Ltd., Shanghai, China).

Sequence analysis

After multiple alignment using ClustalX for Windows (version 1.81) with the default parameters for gap opening and gap extension (Thompson et al. 1997), the phylogenetic relationship between the strains in the present study and the reference strains was assessed using MEGA version 5.05 (Tamura et al. 2011). Briefly, a phylogenetic tree was generated using the neighbour-joining algorithm and Kimura 2-parameter distance model. The reliability of different phylogenetic groups was evaluated by bootstrap test (1000 bootstrap replications).


Rational design of generic primers based on the alignment of NoV GII genome sequences

Three overlapping fragments (approx. 2600–2800 bp) covering the full-length genome were designed to simplify amplification and cloning. Meanwhile, for designing generic primers that can be used for most NoV GII strains, 96 full-length or near-full-length NoV GII genome sequences were collected from GenBank as reference. Based on the sequence alignment results, four conserved regions were selected as template for designing primers.

The initial 24 bases at the 5′ end of the NoV genome were analysed based on the 90 reference strains, of which six strains were not included because of disintegration. The number of times that the four nucleotides were present in each position was listed, and the primer of N1F was designed based on the conservation of this region. The same analysis was carried out for the other two regions (location of 2696–2721 and 2797–2819) based on the 94 genome sequences, and the primers of N2819R and N2696F were correspondingly designed. Furthermore, region N/S was the most important area for designing detective primers because of its conservativeness. Thus, the primers of COG2F and COG2R were also used to amplify fragments B and C (Kageyama et al. 2003).

Nucleotide frequency in positions of primer binding sites was shown in Table 3. The primer of N1F was completely conservative, so we used it for the amplification of fragment A instead of the 5′ RACE method to simplify the cloning process. Well, some sites in the binding regions of N2696F and N2819R were variable, so degenerate primers were designed to cover all GII NoV genotypes.

Table 3. Nucleotide frequency in positions of primer binding sites in NoVs (GII)
Position1–24 (24 bp)
n 838484909090909090909090909090909090909090909090
Position2696–2721 (26 bp)
n 9494949494949494949494949494949494949494949494949494
Position2797–2819 (23 bp)
  1. a

    Sequences are listed as RNA for the RT primer binding site and as cDNA for the forward primer. The most frequent nucleotide(s) are presented.

n 9494949494949494949494949494949494949494949494

Full genome sequencing of the NoV strains

The RNA extracted from faecal samples was used to amplify three NoV genome fragments. Meanwhile, a series of tenfold dilutions of RNA was prepared to compare the sensitivity of the genome cloning primers and JV12/JV13, which showed that JV12/JV13 pair was more sensitive than the new primers (Table 4).

Table 4. Sensitivity comparison of genome cloning primers and detection primers
  1. a

    Positive RT-PCR.

  2. b

    Negative RT-PCR.


After agarose gel electrophoresis, the gel-purified amplicons were cloned into pMD19T simple vector, and the recombinant bacteria were sequenced to obtain the NoV genome. Near-full-length amplicons from two clinical samples were sequenced, and the entire nucleotide sequences were constructed using a method for connecting overlapped sections in separate sequence runs as a single sequence. Their genome sequences were 7559 and 7550 bp (GenBank accession numbers JX989074 and JX989075). The sequences include the 5′ noncoding region, complete ORF1, complete ORF2, complete ORF3 and 3′ noncoding region.

Sequence analysis

A phylogenetic tree was constructed by the alignment of the NoV genome sequences, showing that strains GZ2010-L87 and GZ2010-L96 belong to GII.4-2010 and GII.6, respectively (Fig. 1). Furthermore, genotyping results were confirmed using the Norovirus Genotyping Tool version 1.0 developed by Kroneman et al. (2011).

Figure 1.

Phylogenetic tree comparing the GZ2010-L87 and GZ2010-L96 with full-length genome sequence from representative norovirus strains. The sequences were aligned with ClustalX for Windows version 1.81 and the tree calculated with MEGA version 5.05. A neighbour-joining method was used based on the Kimura 2-parameter distance model. Bootstrap analysis was carried out using 1000 replicates, and the results were expressed as percentage at the nodes. Local NoV strains were designated by location, year and identification number and were highlighted with black dots. Guangzhou (GZ).


NoVs are the most important causative agents of acute gastroenteritis worldwide, and the emerging behaviour of NoVs has stimulated a considerable interest in molecular epidemiology and evolution mechanism. However, studies on NoVs are usually limited by the underestimated rate of recombination, error-prone RNA replication and lacking suitable culture system in vitro. Therefore, genome sequencing is an important method for tracking and monitoring the genetic diversity of NoVs. However, only a few complete NoV genome sequences are available in GenBank, especially in China.

In the present study, a three-fragment NoV genome cloning method was developed using three pairs of broadly reactive primers that are appropriate for NoV GII. The method simplified the amplification and cloning of full-length NoV genome sequences. To identify conserved regions in the NoV genome, most NoV GII full-length or near-full-length genome sequences were collected from GenBank as reference strains for alignment, except those detected since 2004 in Japan, because these sequences almost have high homology and belong to GII.4 genotype.

The NoV GII genome was 7·5–7·7 kb in size. The near-full-length genome sequences can be amplified by a pair of broad primers in some studies. However, long-fragment PCR increased the difficulty of the whole process, and the genome information obtained was not complete (Tellier et al. 2008). Other protocols for obtaining the genome sequences were reported. The NoV genome was also amplified by segmentation PCR in some studies. The virus genome was divided into several fragments (approx. 1 kb in size) and amplified with specific primes. However, the specificity of the primers increased the risk of obtaining full-length sequences (Chhabra et al. 2010; Park et al. 2011; Shen et al. 2011). The three-fragment PCR method was also described in previous studies (Katayama et al. 2002; Yun et al. 2010). However, their primers were generally designed not by the alignment of most NoV genome references, but by taking the most similar genome sequence as reference. Thus, they could not be widely used. Meanwhile, the new method we developed in this study had some superiority. First, the novel method could be used more extensively than others, because the generic primers in this study were designed rationally to coordinate with most NoV GII. Second, three overlapping fragments that encompassed the full-length genome were designed with appropriate length (2·6–2·8 kb). Hence, the divided fragments can be easily amplified and cloned by general molecular biology methods, so it decreased the operational difficulty of NoV genome clone. Third, NoV genome sequences could be saved as DNA pattern by the proposed method, which would provide materials for other NoVs studies.

The primers used in this study were designed based on the alignment results of full-length NoV GII genome sequences from GenBank. The 5′ end of the NoV genome was very conservative. Given that no variation was found, we designed N1F as the forward primer, which was very similar to the primers used previously (Katayama et al. 2002; Tellier et al. 2008), to amplify fragment A instead of the 5′ RACE method. The conservative regions between the position of 2696–2721 and 2797–2819 were also designed as the overlapping area for the first time. Fragment C was amplified according to the method used for amplifying the 3 kb fragment at 3′ end of the NoV genome (Katayama et al. 2002; Motomura et al. 2008; Nenonen et al. 2009). The popular primers of COG2F and COG2R were selected as the forward primer of fragment C and the reward primer of fragment B. Other primers, such as JV12/JV13 or G2SKF/G2SKR, could have the same result for amplifying fragment C (Vinje and Koopmans 1996; Kojima et al. 2002).

As an application, the complete sequences of two NoV strains obtained in 2010 were determined. The GZ2010-L87 strain belongs to a variant of GII.4 New Orleans (GII.4-2009) (99% homology) that shared the changes in the P2 subdomain of VP1. In addition, the GZ2010-L96 strain classified into GII.6 had 95% homology with the Saitama U16 strain, which was reported as early as 2002 in Japan. GII.4 was regarded as the predominant strain because of its clinical and epidemiologic features, but other genotypes such as GII.6, which retained their infectivity over a long time, should also be given more focus. The near-full-length genome information of the two NoV strains was obtained in only 2 days (not including the sequencing process). Thus, the new method greatly simplified the whole process.

In the recent years, the importance of NoV genome information has drawn increasing attention, not just for limitations in classifying viruses into genogroups based on partial nucleotide sequences, but also for the further research on evolution mechanism. In the present study, a three-fragment method was developed to obtain information on the complete genome sequences of NoV GII. This method was designed to amplify all the GII with three pairs of primers, irrespective of the variation within hypervariable regions of recombination events. The novel method simplifies amplification and cloning by dividing the full genome into three fragments with suitable sizes. Overall, the proposed method provides a simple tool for determination of the full-length genome sequences of the viruses detected and for investigation on the molecular epidemiology and evolution mechanisms of NoVs.


This work was supported by the National Natural Science Foundation of China (31271878) and the National Key Technology Research and Development Program (2012BAK08B07).