Previous Article | Next Article ![]()
Journal of Clinical Microbiology, September 2006, p. 3268-3273, Vol. 44, No. 9
0095-1137/06/$08.00+0 doi:10.1128/JCM.00803-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Kitasato Institute for Life Sciences, Laboratory of Viral Infection I, Minato-ku, Tokyo 108-8641, Japan,1 Ashikaga Red Cross Hospital, Department of Pediatrics, Ashikaga, Tochigi 326-0808, Japan,2 Eiken Chemical Co. Ltd., Kita-ku, Tokyo 114-0002, Japan,3 Kansai Medical University, Department of Pediatrics, Moriguchi, Osaka 570-8506, Japan4
Received 21 December 2005/ Returned for modification 11 June 2006/ Accepted 9 July 2006
|
|
|---|
|
|
|---|
Rubella virus infection is one of the communicable diseases in infants and children. Most patients with rubella develop mild symptoms and recover without any complications or sequelae. However, infection among young women who have no immunity against rubella virus during early pregnancy, especially within the first trimester, may cause fetal death or congenital rubella virus infection/syndrome (CRS). CRS is characterized by multiple malformations: deafness, eye abnormalities, congenital cardiac disease, neurological abnormalities, and so on (2, 31). Although the burden of CRS is not well characterized in all countries, WHO estimated that more than 100,000 cases of CRS occur every year in developing countries and 43% (91/214) of countries did not have a national immunization program for rubella in 2002 (2, 31). However, in the United States and several European countries, the indigenous circulation of rubella has been interrupted and CRS has been eliminated. This achievement of the elimination of CRS largely depends on vaccination programs with effective surveillance (2).
In Japan, nationwide outbreaks of rubella have not occurred since 1992-1993. The number of reported cases of clinical rubella decreased from 2,795 in 2003 and 4,248 in 2004 to 895 in 2005 from approximately 3,000 pediatric sentinel sites without requiring any laboratory confirmation (national surveillance data from the National Institute of Infectious Disease, Japan). Although only one case of CRS was reported annually during 1999 to 2003, 10 patients with CRS were reported to the national surveillance program in 2004 and two were reported in 2005. The issue of rubella and CRS still remains important in Japan for the elimination of CRS with the accelerated control of rubella. A single dose of monovalent rubella vaccine targeted to children aged between 12 and 90 months was introduced in a national immunization program in Japan in 1994. From April 2006, Japan has launched a two-dose immunization program with combined measles and rubella vaccine given to children aged 1 year and aged 5 or 6 years, before primary school entry.
There are some problems with diagnosing rubella and CRS from clinical information alone, because some infected individuals are asymptomatic or have mild symptoms. Different viruses such as measles, human parvovirus B19, and enterovirus cause symptoms similar to those of rubella (2, 25, 32). Also, misclassification of cases of CRS may occur, since some patients with CRS may have a single symptom or defect or may have a positive result for immunoglobulin M (IgM) enzyme immunosorbent assays (EIAs) without clinical symptoms (4, 5, 33). The development of efficient laboratory tests will enable the accurate diagnosis of rubella and CRS.
Virus isolation is traditionally performed for the laboratory confirmation of rubella virus infection, but the sensitivity of virus isolation is low because of poor conditions of sample transport and inadequate timing of the sample collection. It takes 3 to 4 weeks to obtain the results of virus isolation after three blind passages in RK13 or Vero cells (5, 22). Serological tests such as the hemagglutination inhibition method are performed principally with paired sera from the acute and convalescent phases (5). Although IgM EIA is employed for diagnosis using a single serum sample, false-positive results of serum rubella IgM tests may occur in people with human parvovirus B19 infection, with a positive heterophile test for infectious mononucleosis, or with positive rheumatoid factor (1, 24, 25, 28, 29).
Recently, several molecular biological laboratory tests for the detection of rubella virus have been developed, such as reverse transcription-PCR (RT-PCR) (3, 6, 8, 13, 15, 22). Loop-mediated isothermal amplification (LAMP) was originally developed for DNA amplification, as reported by Notomi et al. (19) in 2000. This method was also applied for the detection of RNA genomes. LAMP amplifies targeted nucleotides with six primers within 1 hour under isothermal conditions without any temperature shifts such as those used for denaturation, annealing, and extension in the PCR cycling reactions (19). In this study, we developed an RT-LAMP method for the detection of the rubella virus genome and compared the sensitivity of RT-LAMP with those of RT-PCR and virus isolation using clinical samples.
(We presented this study at the plenary session of the 46th Annual Meeting of the Japanese Society of Clinical Virology, Fukuoka, Japan, June 2005.)
|
|
|---|
The Takahashi vaccine strain developed at the Kitasato Institute, Japan (RV/Takahashi: AB03338), was used as the reference strain in this study. Also, we used four isolates as wild circulating strains: two strains (RVi/Tokyo.JPN/87CRS-w: AB238916 and RVi/Tokyo.JPN/87CRS-o: AB238917) isolated in 1987 in Tokyo, Japan; one strain (RVi/Tokyo.JPN/95CRS-n: AB238918) isolated in 1995 in Tokyo, Japan; and one strain (RVi/Kanagawa.JPN/04-s: AB238915) isolated from a patient with acute rubella virus infection in 2004 in Kanagawa, Japan. The E1 sequence data were registered with the DNA Data Bank of Japan under the accession numbers in parentheses.
RT-LAMP. The genomic RNA was extracted from 200 µl of virus culture fluids of rubella virus or clinical samples using a magnetic bead RNA purification kit (TOYOBO Co. Ltd., Osaka, Japan). RNA was resuspended in 30 µl of distilled water, and 5 µl was subjected to RT-LAMP and RT-PCR. RT-LAMP was carried out using autocycling strand displacement DNA synthesis performed using Bst DNA polymerase (New England Biolabs). The RT-LAMP primers were designed using the software program for LAMP primer design (Eiken Chemical Co. Ltd., Japan). The RT-LAMP primers were designed in the E1 region between genome positions 8476 and 8680. A diagram of the RT-LAMP primers is shown in Fig. 1. We synthesized six primers: outer primers (F3 and B3), inner primers (FIP and BIP), and two additional loop primers (F loop and B loop). FIP contained a sequence complementary to F1 linked with the F2 sequence, and BIP contained a B1 sequence linked with a sequence complementary to B2. These four primers amplified the targeted DNA. We synthesized the F loop primer between F1 and F2, and the B loop primer between B1 and B2, to enhance the sensitivity and reactivity (17, 18, 19).
![]() View larger version (37K): [in a new window] |
FIG. 1. Diagram of RT-LAMP primers for the detection of the rubella virus genome. The eight primer sites are shown in the upper panel. Sequence alignments of six LAMP primers are shown in the lower panel. Positive-sense F3 and complementary B3 were used as outer primers. FIP contains the sequence complementary to F1 linked with the F2 sequence. BIP contains the B1 sequence linked with the sequence complementary to B2. Two additional loop primers (F and B) are synthesized between F1 and F2 and between B1 and B2, respectively. The arrows show the direction of DNA synthesis.
|
As the RT-LAMP reaction progresses, the reaction by-products, pyrophosphate ions, bind to magnesium ions and form a white precipitate of magnesium pyrophosphate. Thus, the value of the turbidity is closely related to the amplification of targeted DNA (16).
RT-PCR. The purified RNA genome was converted to cDNA with avian myeloblastosis virus reverse transcriptase (Life Science Inc.) at 50°C for 1 h, using primer 9407-9426 () (5'-AGGGCGGCGGTGACGAACTT-3'). Sets of primers in the E1 region for the detection of rubella virus genes were designed as follows: 8110-8129 (+) (5'-GTCTCTTGATCACGCCCTCG-3') and 8797-8816 () (5'-GTGTTGCAGAACGGGTGTTCA-3') for the first PCR and 8476-8494 (+) (same as F3 RT-LAMP primer) and 8663-8680 () (same as B3 RT-LAMP primer) for the nested PCR. The RT-PCR was performed with 1.25 U of Taq DNA polymerase (TaKaRa BioMedicals, Tokyo, Japan) in a thermal cycler (TaKaRa BioMedicals, Tokyo, Japan) with 35 rounds of the following thermal cycling conditions; denaturation at 95°C for 30 seconds, annealing at 58°C for 30 seconds, and extension at 72°C for 1 min 40 seconds. The PCR products were confirmed by electrophoresis through a 1.5% agarose gel and stained with ethidium bromide (10, 20, 27).
To conduct the sequence analysis using the PCR products from virus isolates, we made different sets of PCR primers designed as follows: 7981-8000 (+) (5'-CGAAGACGGCTGGACTTGCC-3') and 9743-9762 () (5'-CTATGCAGCAACAGGTGCGG-3') for the first PCR and 8110-8129 (+) and 9743-9762 () for the nested PCR. The nested RT-PCR products including the target site between genome positions 8731 and 9469 were used for routine genotyping analysis (30).
|
|
|---|
![]() View larger version (40K): [in a new window] |
FIG. 2. Detection limit of RT-LAMP and RT-PCR. The Takahashi vaccine strain containing 105.5 PFU/0.1 ml was used. RNA was serially diluted by 1:10, and each dilution was subjected to RT-LAMP and RT-PCR. A result with a turbidity of 0.1 was considered to be LAMP positive. Numbers at left of the right panel are sizes in base pairs of the DNA marker.
|
Detection of rubella virus genome by RT-LAMP. Nine patients who were clinically diagnosed with acute rubella virus infection were examined. They were 0 to 8 years old, and six (66.7%) were male. The results of virus isolation, RT-PCR, and RT-LAMP are summarized in Table 1. Rubella viruses were isolated from three patients (33.3%) and designated RVi/Tochigi.JPN/04-s, RVi/Tochigi.JPN/04-h, and RVi/Tochigi.JPN/04-i, their E1 region genome sequences having the accession numbers AB238919, AB238920, and AB238921, respectively. They were classified into genotype 1D. Rubella virus genome was detected in six patients (66.7%) by RT-PCR and seven (77.8%) by RT-LAMP. The rubella virus genome was detected by RT-PCR and RT-LAMP in all three patients positive for virus isolation. Among six patients negative for virus isolation, the virus genome was detected in three by RT-PCR and in four by RT-LAMP. The backgrounds of the patients are shown in Table 2, together with the results of virological examinations. Eight patients did not have a history of immunization against rubella, and the rubella immunization history of case 1 was unknown. The rubella IgM EIA was performed in four (cases 5, 6, 7, and 9) out of nine patients and was found to be positive in all. Of two patients negative for all three tests, one patient (case 5) was positive for IgM EIA antibodies and another patient (case 3) had family contact with a laboratory-confirmed patient (case 6). Among four patients (cases 5, 6, 7, and 9) who were serologically confirmed to have rubella virus infection by IgM EIA, the rubella virus genome was detected in three by RT-LAMP and RT-PCR. Among five patients who were clinically diagnosed without serological confirmation, RT-PCR or RT-LAMP detected rubella virus genome in four patients.
|
View this table: [in a new window] |
TABLE 1. Results of virological examinations: virus isolation, RT-PCR, and RT-LAMP, using nine clinical samples obtained from nine patients suspected of rubella virus infection
|
|
View this table: [in a new window] |
TABLE 2. Immunization history and the results of serology, virus isolation, RT-PCR, and RT-LAMP for nine patients
|
|
|
|---|
The RT-LAMP primers were designed based on the Takahashi vaccine strain, since the wild strains of rubella (clades 1 and 2) were found to differ by 8 to 10% at the nucleotide level (5). The genetic differences in the E1 region between the wild strains and Takahashi vaccine strain were 2.8 to 5.8% in our study. Although RT-LAMP was not performed for clade 2, we suppose that RT-LAMP would detect other genotypes because the RT-LAMP primers were designed in the conserved region. RT-LAMP amplifies 200 to 250 bp of nucleotides with high sensitivity and specificity. The products of RT-LAMP are not appropriate for sequence analysis because most of the region amplified by RT-LAMP is occupied by the RT-LAMP primers as shown in Fig. 1. WHO recommended the E1 coding region for molecular epidemiology or a window of 739 nucleotides, 8731 to 9469, for routine molecular epidemiological analysis (30, 32, 33). As for genotyping, E1 gene should be amplified by conventional RT-PCR.
RT-LAMP for rubella would be expected to be a reliable and rapid diagnostic method in the clinical setting, because RT-LAMP showed an equivalent or higher detection rate compared with RT-PCR and virus isolation for nine samples obtained from clinically diagnosed patients. The RT-LAMP procedure has clinical advantages of simplicity and rapidity, in comparison with virus isolation and RT-PCR. Virus isolation requires complex procedures for cell culture, is not always successful, and is not appropriate for a clinical laboratory diagnostic tool. The genome amplification method always has the possibility of false positives due to cross-contamination. Since RT-LAMP is performed as a simple procedure in a single tube with sensitivity similar to that of nested PCR, cross-reaction would less likely occur in RT-LAMP than in RT-PCR.
The detection limit of RT-PCR primers used in our study (30 PFU/ml) was similar to that in previous studies, and the RT-PCR primers were designed to have the same target as the RT-LAMP primers in order to compare the sensitivities of RT-LAMP and RT-PCR. In previous reports, RT-PCR for rubella targeting the E1 region detected up to 8 infectious units of WHO international standard/ml in amniotic fluid (8, 15) and a 106 dilution of culture fluid containing 106.8 TCID50/ml of rubella virus, reported by Bosma et al. (3). Cooray et al. (6) also reported that the rubella genome RNA was detected in oral fluid containing 100 PFU/ml.
In several laboratories, real-time PCR has been developed for the detection of measles virus (21, 23), mumps virus (14), and RSV (9, 12), but real-time PCR for rubella virus has not been reported. This method is based on PCR temperature shifts and shows higher sensitivity than RT-PCR with a reduction of the risk of cross-contamination. As it takes 2 to 3 h to obtain the results by real-time PCR, RT-LAMP has the distinct benefit of a faster reaction, within 60 min. Also, Tzeng et al. (26) improved the methodology of virus isolation developed using a replicon-based reporter gene assay for the detection of rubella virus. LAMP is a simpler, more sensitive, and more rapid method than virus isolation.
Misclassification of the clinical diagnosis of the nine patients from whom we collected samples in our study might have occurred, since we did not require a serological confirmation for the definition of the cases. Among the nine patients in this study, IgM EIA was performed in four patients. The remaining five did not receive an IgM test, but they were strongly suspected to have rubella virus infection. These five samples from patients with suspected rubella were appropriate for use in our study for the following two reasons. The outbreak of rubella in this area in April 2004 was confirmed by isolating rubella virus from patients. Also, the positive predictive value of case definition is considered to be high during outbreaks (11).
This study had some limitations, including the different criteria for the clinical diagnosis of rubella, the differences in timing and procedure for sample collection, and the limited number of clinical samples. However, at least 100 samples should be tested for the assessment of the procedure according to the review on the validation of nucleic acid-based tests published by Dimech et al. (7). The purpose of this study was to establish a rapid, sensitive, and simple method of RT-LAMP for the detection of rubella virus. The detection of the rubella virus genome by RT-LAMP can be confirmed using clinical samples from patients with rubella or CRS. We should further evaluate RT-LAMP for rubella diagnosis in the field.
During the progress of rubella control toward CRS elimination, the clinical application of RT-LAMP would be helpful for rapid diagnosis when managing patients or determining the response to an outbreak. A rapid response should be made to cope with outbreaks of rubella or CRS. RT-LAMP would contribute to the rapid diagnosis of index cases for the control of outbreaks. Methods for laboratory-confirmed diagnosis such as RT-LAMP will improve the surveillance of rubella and CRS by enabling the early detection of outbreaks and decreasing the risk of misclassification of cases that may occur with clinical diagnoses.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»