Diagnostic Assay for Rickettsia japonica

We developed a specific and rapid detection system for Rickettsia japonica and R. heilongjiangensis, the causative agents of spotted fever, using a TaqMan minor groove binder probe for a particular open reading frame (ORF) identified by the R. japonica genome project. The target ORF was present only in R. japonica–related strains.

formation from this genome project, we performed DNA sequencing for this 216-bp ORF to determine whether the specifi c DNA sequences are conserved in all R. japonica strains and other closely related strains, including R. heilongjiangensis (5) and Rickettsia sp. LON (6).
R. heilongjiangensis is also a causative agent of spotted fever in northeastern Asia and has been classifi ed within the R. japonica group (5). Several studies have reported that Rickettsia. sp. LON strains also have similar sequences to R. japonica. Our PCR can easily distinguish Rickettsia sp. LON strains (LON-2, LON-9, and LON-13) from R. japonica strains. This test can help in the diagnosis because Rickettsia sp. LON strains have only been isolated from ticks and may not be pathogenic in humans (6).
DNA sequencing was performed by using an ABI PRISM BigDye Terminator version 3.1 Kit (Applied Biosystems, Foster City, CA, USA) with an ABI 3130 sequence detector. DNA sequences were aligned by ClustalW software (http://clustalw.ddbj.nig.ac.jp/top-e. html) with an open gap penalty of 15, a gap extension penalty of 6.66, a gap distance of 8, and a maximum division penalty of 40. For determination of the DNA sequence for the 216-bp ORF, the primer pair of JapoSP5′ (5′-ACAACATCAATATTATAATTAGTATCC-3′) and JapoSP3′ (5′-TTCACGTATGTCTATATATGCTGCAG CG-3′) was used to amplify a 564-bp section, including this ORF, because this unique DNA sequence was located as the inserted sequence of the homolog for R. conorii RC1338 (Figure, panel A).
The nucleotide sequence of this ORF was identical among 5 of the R. japonica strains: DT-1, YH, FLA-1, HH-8, and HH-9 (100%); the sequence was highly conserved with signifi cant identity in R. heilongjiangensis (99.5%), except for the Rickettsia sp. LON strains (92.1%; Figure, panel B). This 216-bp ORF had been previously applied to BLAST searches with NCBI blast (nblast) for humans, mice, and others (http://blast.ncbi.nlm.nih.gov/ Blast.cgi). The results showed that no similar ORF had been reported to date. Therefore, we focused on this conserved region of the 216-bp ORF to develop a TaqMan minor groove binder (MGB) probe (Applied Biosystems) that could detect the pathogenic R. japonica group, including R. heilongjiangensis, with a high degree of specifi city.
Oligonucleotide primers (SpRija5′ and SPRija3′) and the TaqMan MGB probe (SpRijaMGB) were designed by using Primer Express software version 2.0 (Applied Biosystems Figure, panel B). The detection probe was labeled with the fl uorescent reporter FAM (carboxyfl uorescein labeling) at the 5′ end; the nonfl uorescent quencher and MGB were labeled at the 3′ end ( Figure,   Real-time PCR was performed by using an ABI 7500 system (Applied Biosystems). DNA polymerase (perfect real-time PCR) for the PCR was obtained from Takara Bio (Kyoto, Japan). A 20-μL sample was added to each well of a 96-well microplate (Thermo Fisher Scientifi c Inc., Waltham, MA, USA) according to the manufacturer's instructions. Thermal cycle protocol was performed as follows: fi rst incubation stage, 20 s at 95°C; second stage, 5 s at 95°C and 34 s at 60°C. The second stage was repeated 45 times. For analysis of real-time PCR, the threshold line was fi xed at 0.2 to avoid detection of nonspecifi c fl uorescence. This detection procedure can be completed within 1 h.
The reactivity of this assay was examined by using various copy numbers of synthetic R. japonica DNA fragments that were amplifi ed by the primer pairs JapoSP5′ and JapoSP3′ within the R. japonica genome. Genomic DNA of R. japonica strain YH was prepared from cultivated bacteria according to methods proposed by Furuya et al. (7). A calibration curve was generated with 5 calibrators, ranging from 10 2 to 10 9 copies/well in triplicate. We found a linear correlation (R>0.99) between the detection cycle numbers and R. japonica DNA copy numbers from 10 2 to 10 9 copies/ reaction (data not shown).
A total of 26 rickettsial strains, classifi ed into 11 species, were used in this study (Table 1). Specifi cities for this TaqMan PCR are also summarized in Table 1. Genomic DNA of R. prowazekii and R. rickettsii were prepared from antigen slides (Panbio Inc., Sinnamon Park, Queensland, Australia) by using a Gentra Puregene kit (QIAGEN, Valencia, CA, USA). Genomic DNA of other Rickettsia strains was also prepared from cultivated bacteria according to methods proposed by Furuya et al. (7).
Our results showed that this novel assay could detect all 5 R. japonica strains and 1 R. heilongjiangensis strain used in this study. However, it could not detect R. rickettsii, R. prowazekii, or other Rickettsia strains. These results indicate that the combination of probes and primers in this study had high specifi city for the pathogenic R. japonica group. Nonspecifi c reactions were not observed when genomic DNA from human or murine fi broblasts was used in any of the assays (data not shown).
The detection limits of this PCR were compared to those of conventional PCRs by using serially diluted genomic DNA. The conventional PCRs, designated as Rj5-Rj10 and R1-R2 assays, were designed to detect the 17-kDa antigen gene of Rickettsia, by using a primer set of Rj5 (5′-CGCCATTCTACGTTACTACC-3′) and Rj10 (5′-ATTCTAAAAACCATATACTG-3′) (7) and R1 (5′-TCAATTCACAACTTGCCATT-3′) and R2 (5′-TTTACAAAATTCTAAAAACC-3′) (15), respectively. Since 1996, these assays have been used for molecular diagnosis of JSF in clinical laboratories in Japan. Recent studies have suggested that a TaqMan PCR assay may be as much as 100× more sensitive than these assays (data not shown). Therefore, in our study, the TaqMan PCR was assumed to be much more sensitive than the conventional assays that are known to show false-negative results, even for a patient with acute-stage JSF.
We requested clinical samples to verify the validity of our assay. Eighteen DNA templates were extracted from blood clots collected from 18 patients in the acute stages of illness (male:female ratio 1:1; average age 64.1 years [range 27-88 years]); average number of days after onset  of fever 4.6 [range 2-7 days; Table 2]). These templates were reexamined by using our TaqMan PCR. Although the conventional assays could not detect the presence of any Rickettsia DNA, 9 of 18 samples displayed positive results with the TaqMan PCR ( Table 2). The blood clot from the patient in whom Scrub typhus disease was previously diagnosed was used as a negative control in this real-time PCR assay, resulted were not detected. Our TaqMan PCR is currently available to clinical laboratories that need to rule out false-negative results in molecular diagnoses.

Conclusions
JSF is a threat to public health in Japan. Our results suggest that an R. japonica-specifi c 216-bp ORF may have been conserved throughout the R. japonica species and closely related Rickettsia spp. The newly developed realtime PCR system, which demonstrated a high level of sensitivity and specifi city, may be a useful tool for laboratory diagnosis.