Detection of Kudoa septempunctata 18S Ribosomal DNA in Patient Fecal Samples from Novel Food-Borne Outbreaks Caused by Consumption of Raw Olive Flounder (Paralichthys olivaceus)

Abstract

Kudoa septempunctata is a newly identified myxosporean parasite of olive flounder (Paralichthys olivaceus) and a suspected causative agent of several food-borne gastroenteritis outbreaks in Japan. Here, we report the detection of K. septempunctata 18S ribosomal DNA in fecal samples of outbreak patients using an efficient method based on real-time PCR. We first performed a spiking experiment to assess whether our previously developed real-time PCR assay was applicable to detect K. septempunctata in feces. Simultaneously, we compared the relative extraction efficacy of K. septempunctata DNA using three commercial kits. Finally, our detection method was validated by testing 45 clinical samples obtained from 13 food-borne outbreaks associated with the consumption of raw flounder and 41 fecal samples from diarrhea patients epidemiologically unrelated to the ingestion of raw fish. We found that the FastDNA Spin Kit for Soil (MP Biomedicals) was the most efficient method for extracting K. septempunctata DNA from fecal samples. Using this kit, the detection limit of our real-time PCR assay was 1.6 × 10¹ spores per g of feces, and positive results were obtained for 21 fecal and 2 vomitus samples obtained from the food-borne outbreaks. To our knowledge, this is the first report to describe the detection of K. septempunctata DNA in patient fecal samples. We anticipate that our detection method will be useful for confirming food-borne diseases caused by K. septempunctata in laboratory investigations.

INTRODUCTION

In Japan, the number of outbreaks of an unidentified food-borne disease associated with raw fish consumption has increased since 2003 and has averaged more than 100 per year, reaching 158 in 2010 (7). In most outbreaks, victims ingested raw olive flounder (Paralichthys olivaceus), but no food-borne bacteria, viruses, bacterial toxins, or chemicals were detected in the implicated foods. Symptoms of the illness occur within 2 to 20 h after consuming raw olive flounder and include transient, but strong, diarrhea and vomiting, with patients typically recovering within 24 h of the onset of symptoms (7). In our recent study examining the etiological agents of this food-borne disease, we conducted epidemiological and metagenomic sequence analyses and animal experiments, which revealed that the novel myxosporean parasite Kudoa septempunctata was the causative agent of the outbreaks (5, 7). Since olive flounder is an important commercial fish and widely distributed in Japan, food-borne outbreaks caused by K. septempunctata have become a notable public health concern. However, since K. septempunctata was first described in the trunk muscles of an aquacultured olive flounder in 2010 (10), only a limited number of investigations of this parasite have been conducted in various disciplines related to the field of food hygienics. Therefore, no effective measures for preventing K. septempunctata food-borne outbreaks caused by the consumption of raw flounder have developed, other than eliminating olive flounder infected with this parasite. Accordingly, to prevent the spread of K. septempunctata food-borne outbreaks, it is important to investigate each outbreak case in detail to develop suitable public health strategies.

Outbreaks of this novel food-borne disease are typically associated with the consumption of traditional Japanese foods, particularly “sushi” (cooked rice with sliced raw fish) and “sashimi” (sliced raw fish), which are prepared using raw flounder muscle tissue containing more than 10,000 spores of K. septempunctata per g. However, since certain outbreaks had low attack rates or odds ratios for raw flounder (7), the diagnosis of this food-borne disease can be difficult using only epidemiological analyses. Therefore, the detection of K. septempunctata in flounder consumed by patients or in patient samples is important for the confirmation of K. septempunctata-associated food-borne disease.

We previously established a real-time PCR method targeting the small subunit ribosomal DNA (18S rDNA) of K. septempunctata as a rapid, accurate, sensitive, and specific diagnostic molecular technique to evaluate the infection levels of this parasite in olive flounders (6). Although this method was useful for detecting the causative foods in K. septempunctata food-borne outbreaks (6), it is often difficult to evaluate food samples that are implicated in such outbreaks, as the raw flounder items are frequently entirely consumed. Thus, a suitable, highly sensitive method to analyze patient clinical samples, such as feces and vomitus, is required for investigating outbreaks potentially associated with this novel food-borne pathogen.

The detection of pathogens in feces using light microscopy is a classical laboratory method for diagnosing parasitic cryptosporidiosis and giardiasis (2, 4, 12). However, we were previously unable to detect K. septempunctata spores in patient fecal specimens using a formalin-ether sedimentation technique (4, 12), suggesting that the number of intact spores in each fecal sample was below the detection limit for the microscopic assay. Here, we attempted to apply our highly specific and sensitive real-time PCR method to the detection of K. septempunctata in spiked clinical fecal samples. In addition, we compared the relative DNA extraction efficacy of three commercial DNA extraction kits using artificially contaminated fecal samples. Finally, we tested 45 patient samples obtained from 13 food-borne outbreaks associated with the consumption of raw flounder and 41 fecal samples from diarrhea patients epidemiologically unrelated to flounder to confirm whether our detection method can be reliably applied to the investigation of food-borne outbreaks associated with K. septempunctata.

MATERIALS AND METHODS

Real-time PCR assay.

Real-time PCR assays were performed using a previously reported method with minor modifications of the reaction cycling conditions (6). The following cycling conditions were used: 95°C for 30 s, followed by 45 cycles of amplification at 95°C for 5 s and 60°C for 31 s. PCR amplifications were performed using the primers Forward (5′-CGGTCATATCAGCCATGGATAAC-3′), and Reverse (5′-CTATCGACAAATTAATGTTCGATATGC-3′) and the PrimeTime qPCR probe (6-FAM)-TCACCATGTAAATGGTGGGAGCATTT-(Iowa BlackFQ) (Medical and Biological Laboratories, Nagoya, Japan) (6).

Comparison of three DNA extraction methods using fecal samples artificially contaminated with K. septempunctata.

Spiking experiments with K. septempunctata spores were performed to assess whether our real-time PCR assay was applicable to the detection of K. septempunctata DNA in fecal samples and to compare the relative DNA extraction efficiencies among three commercial kits. Three fecal samples (samples A, B, and C) epidemiologically unlinked to the consumption of flounder were spiked with 10-fold dilutions of a K. septempunctata spore suspension to obtain concentrations of 1.6 × 10⁴ and 1.6 × 10⁶ spores per 1 g of feces for each sample. The spore suspension was prepared from flounders infected with K. septempunctata using Percoll (GE Healthcare, Little Chalfont, United Kingdom) density gradient centrifugation as previously described (6) and was stored at 4°C until use.

Three commercial DNA extraction kits—the QIAamp DNA stool minikit (Qiagen, Hilden, Germany), the FastDNA Spin Kit for Soil (MP Biomedicals, Solon, OH), and the UltraClean fecal DNA kit (MO BIO Laboratories, Inc., Carlsbad, CA)—were used to extract DNA from the artificially contaminated fecal samples. To compare the amount of K. septempunctata DNA extracted using the three kits, 200 mg of each sample and 200 μl of DNA elution buffer were used during the extraction procedure for each kit. Extracted DNA was stored at −20°C until use. Each DNA extraction was performed in triplicate and real-time PCR was performed in duplicate for each extracted DNA sample. Mean cycle threshold (C_T) values were then compared using the Tukey-Kramer test with the significance level set at P < 0.05 to determine the most efficient DNA extraction method.

Analytical sensitivity of the real-time PCR assay using DNA extracted with the FastDNA Spin Kit for Soil.

To evaluate the analytical sensitivity of our real-time PCR assay for K. septempunctata DNA, two fecal samples (samples D and E) epidemiologically unrelated to the ingestion of flounder were spiked with a K. septempunctata spore suspension at final concentrations of 1.6 × 10⁻¹ to 1.6 × 10³ spores per 1 g of feces. Total DNA was then extracted from the spiked samples using the FastDNA Spin Kit for Soil in duplicate. Although the kit protocol recommends the use of 500 mg of starting material, we used 300 mg of each spiked sample to avoid sample loss and tube failure caused by overfilling of the matrix tube. Total DNA was eluted with 100 μl of DNase/pyrogen-free water. Real-time PCR assays were performed in triplicate using total DNA extracted from both spiked samples. Mean C_T values obtained from the duplicate measurements of samples D and E spiked with equal numbers of K. septempunctata spores were compared using the two-tailed Student t test with a significance level of P < 0.05.

Real-time PCR assay evaluation using outbreak patient samples.

A total of 45 samples consisting of 43 fecal and 2 vomitus samples obtained from patients involved in 13 food-borne outbreaks occurring between May and November 2011 in western Japan, which were epidemiologically linked to the consumption of raw flounder, were used to evaluate our K. septempunctata detection method. Health agencies in each jurisdiction interviewed the patients using standard questionnaires to obtain demographic and clinical information and asked about food exposures in the days prior to disease onset. In particular, the investigators asked about the consumption of foods containing raw olive flounder in the 24 h prior to onset, revealing that olive flounder “sashimi” was consumed in all cases, except for one, in which olive flounder “sushi” was consumed. In most patients associated with these 13 outbreaks, symptoms characterized by abdominal pain, diarrhea, nausea, and/or vomiting occurred within 1 to 14 h. All clinical samples were negative for food-borne bacteria and viruses (enterotoxigenic Staphylococcus aureus, Bacillus cereus, Clostridium perfringens, Salmonella spp., Shigella spp., diarrheagenic Escherichia coli, Campylobacter coli, C. jejuni, Vibrio cholerae, V. parahaemolyticus, and norovirus genogroups I and II). In 6 of the 13 outbreaks, 1.3 × 10⁴ to 2.2 × 10⁷ K. septempunctata spores per g were detected in the flounders ingested by patients or in the same lot of fish using microscopic morphological examination and our real-time PCR method (6). However, the causative food was not available in the remaining seven outbreaks. In addition, 41 fecal samples from patients with gastrointestinal disease in several food-borne outbreaks epidemiologically unrelated to K. septempunctata (17, 7, and 17 patients in Salmonella enterica serovar Enteritidis, C. coli and C. jejuni, and norovirus outbreaks, respectively) were also tested. Portions (300 mg) of each sample were subjected to DNA extraction using the FastDNA Spin Kit for Soil.

To confirm that no false-positive results were detected, the real-time PCR products were electrophoresed for 30 min in a 2.0% agarose gel at 100 V and visualized by staining with ethidium bromide and viewing under UV light to confirm the presence of amplicons of the expected size. In addition, a nested PCR assay was conducted using the following forward and reverse flanking primers: (forward-114F) 5′-ATGGATAACTGTGGTAAATCTAGAGCTAATAC-3′ and (reverse-200R) 5′-CCAGTTGGTCGAGTCTAATAAATGC-3′. The nested PCR mixture was prepared using PuReTaq Ready-To-Go PCR Beads (GE Healthcare, Piscataway, NJ) containing 5 pmol of each primer and 2.5 μl of a 1/100 dilution of the real-time PCR products in a total volume of 25 μl. Amplification conditions involved an initial denaturation at 94°C for 5 min, 30 cycles of denaturation at 94°C for 30 s, annealing at 55°C for 30 s, and extension at 72°C for 30 s, followed by a final elongation step at 72°C for 5 min. The amplified products were electrophoresed and visualized as described above.

To exclude the possibility that inhibition of the real-time PCR led to false-negative results, separate samples were spiked with a 1/100 volume of a K. septempunctata 18S rDNA plasmid solution (1.1 × 10³ copies/μl) and run in parallel with unspiked samples. The plasmid DNA solution was prepared as described previously (6).

Fisher exact test with a significance level of P < 0.05 was used to assess the association between the real-time PCR detection rate and the total length of time from causative food consumption to specimen collection.

RESULTS

Comparison of three commercial DNA extraction kits for use in real-time PCR.

Table 1 shows the mean C_T values of real-time PCRs performed with DNA extracted from three spiked fecal samples (samples A, B, and C) using three commercial DNA extraction kits. In assays using samples spiked at the lowest level (1.6 × 10⁴ spores/g), the lowest C_T values were obtained from DNA samples extracted by the FastDNA Spin Kit for Soil for all three fecal samples. Moreover, the lowest C_T values were also obtained for assays performed using DNA extracted with the FastDNA Spin Kit for Soil in two of the three fecal samples spiked with 1.6 × 10⁶ spores/g. These data suggested that the FastDNA Spin Kit for Soil was the most efficient of the three commercial kits for extracting K. septempunctata DNA from the spiked fecal samples. Therefore, this kit was selected for use in all subsequent experiments.

Table 1.

Mean C_T values determined using three commercial DNA extraction kits

Sample	Mean CT ± SDa at low and high concentrations of spiked K. septempunctata spores in fecal samples
	Low (1.6 × 104 spores/g)			High (1.6 × 106 spores/g)
	QIAamp	FastDNA	UltraClean	QIAamp	FastDNA	UltraClean
A	37.65 ± 0.747C	27.55 ± 0.286A	35.54 ± 0.751B	30.70 ± 0.125C	24.25 ± 0.547A	29.48 ± 0.596B
B	37.15 ± 0.435C	29.40 ± 2.264A	32.94 ± 0.330B	31.83 ± 0.366C	27.03 ± 0.323B	25.31 ± 0.212A
C	36.96 ± 0.422C	27.75 ± 0.108A	32.85 ± 0.193B	29.52 ± 0.166C	20.64 ± 0.215A	25.65 ± 0.853B

^aSuperscript letters A to C indicate the DNA extraction efficiency in each experiment, from high (A) to low (C), as determined using the Tukey-Kramer test with a significance level (P) of 0.05. Kits: QIAamp, QIAamp DNA stool minikit (Qiagen, Hilden, Germany); FastDNA, FastDNA Spin Kit for Soil (MP Biomedicals, Solon, OH); UltraClean, UltraClean fecal DNA kit (MO BIO Laboratories, Inc., Carlsbad, CA).

Analytical sensitivity of the real-time PCR assay with spiked fecal samples.

To evaluate the analytical sensitivity of the real-time PCR assay developed for K. septempunctata detection, two fecal samples (samples D and E) spiked with 1.6 × 10⁻¹ to 1.6 × 10³ spores of K. septempunctata per g were analyzed (Table 2). The mean C_T values for the real-time PCRs were determined using a threshold line of 0.35. Although amplification products were not obtained for samples spiked with 1.6 × 10⁻¹ or 1.6 × 10⁰ spores/g, the mean C_T values of samples D and E, which were spiked with 1.6 × 10¹ spores/g, were 36.83 ± 0.36 and 39.75 ± 1.24, respectively, suggesting that the detection limit of the real-time PCR assay was 1.6 × 10¹ spores per g of feces. Based on these findings, we concluded that the patient fecal samples were positive for K. septempunctata if the C_T value was ≤41 cycles, when the threshold line of the real-time PCR assay was set at 0.35 for fluorescent intensity. In addition, significant differences in the C_T values were detected between the two fecal samples spiked with 1.6 × 10¹, 1.6 × 10², and 1.6 × 10³ spores/g (two-tailed Student t test [P < 0.05]).

Table 2.

Mean C_T values in spiking experiments used to evaluate the analytical sensitivity of the real-time PCR assay for K. septempunctata in fecal samples

Sample	Mean CT ± SDa at an inoculum dose (spores/g) of:
	1.6 × 10−1	1.6 × 100	1.6 × 101	1.6 × 102	1.6 × 103
D	ND	ND	36.83 ± 0.36A	33.91 ± 0.64B	30.16 ± 0.35C
E	ND	ND	39.75 ± 1.24A	36.61 ± 0.55B	32.19 ± 0.29C

^aSuperscript letters A to C denote pairs where the difference in mean C_T values was statistically significant (P < 0.05), as determined by two-tailed Student t tests. ND, not determined.

Detection of K. septempunctata in clinical samples from patients associated with outbreaks due to the consumption of raw flounder.

To evaluate the sensitivity of our real-time PCR assay for the detection K. septempunctata in clinical samples, 43 fecal and 2 vomitus samples obtained from 13 food-borne outbreaks associated with the consumption of flounder were tested. Of 43 fecal samples, positive real-time PCR results for K. septempunctata were detected in 21 samples. In addition, both vomitus samples gave positive results in the assay. Single PCR products of the expected size (159 bp) were amplified from all 23 positive samples and the C_T values of the positive clinical samples ranged from 30.78 to 40.69. Moreover, nested PCR products were also obtained for all real-time PCR-positive samples. Eventually, K. septempunctata 18SrDNA was detected in at least one clinical sample from each of the 13 outbreaks. Importantly, no false-negative results were detected, and no PCR products were amplified from the 41 fecal samples obtained from patients involved in the food-borne outbreaks epidemiologically unrelated to K. septempunctata.

Of 40 fecal samples whose sampling dates were known, the detection rate (63.0%) of K. septempunctata in the 27 fecal samples collected within 2.5 days of the consumption of flounder was significantly higher than that (23.1%) in the 13 fecal samples collected later than 2.5 days after the consumption (Fisher exact test, P < 0.05).

DISCUSSION

In the present study, we demonstrated that our previously established real-time PCR assay was applicable for the detection of K. septempunctata DNA in clinical fecal samples. Moreover, it was determined that the FastDNA Spin Kit for Soil more efficient extracted K. septempunctata DNA from artificially contaminated fecal samples than the QIAamp DNA stool minikit and UltraClean fecal DNA kit. Notably, our results showed that increasing the duration between the intake of infected olive flounder and sample collection affected the ability of our real-time PCR assay to detect K. septempunctata in feces. Despite this limiting factor, the presently described method would be useful for laboratory investigations of food-borne outbreaks caused by this novel pathogen.

To develop a detection method with higher sensitivity for K. septempunctata DNA, we first compared the relative DNA extraction efficiencies from fecal samples artificially contaminated with K. septempunctata spores using three commercial kits. We found that the FastDNA Spin Kit for Soil, which has been used to directly isolate PCR-ready genomic DNA of microorganisms not only from soil and sediments, but also from feces (1, 9), displayed the highest efficiency in almost all assays. Therefore, this kit was used with our real-time PCR assay for detecting K. septempunctata in clinical samples of outbreak patients.

The detection limit of our real-time PCR assay used in conjunction with the FastDNA Spin kit for Soil in the spiking experiments was 1.6 × 10¹ spores per g of feces. As 300 mg fecal sample spiked with 1.6 × 10¹ K. septempunctata spores per gram theoretically contains approximately five spores, and we previously showed that a single spore of K. septempunctata contains at least approximately 3.0 × 10² of 18S rDNA copies (6), the real-time PCR detectable limit in the spiking experiment was approximately 3.0 × 10¹ of 18S rDNA copies per reaction. This detection limit was nearly equal to that of a real-time PCR assay performed using plasmid DNA as a control (1.1 × 10¹ copies of K. septempunctata 18S rDNA per reaction) in our previous study (6). These data suggest that presently described real-time PCR assay for the detection of K. septempunctata in fecal samples has a similar sensitivity to that used for plasmid-encoded K. septempunctata 18S rDNA. However, the mean C_T values for two samples that samples that were spiked with equal numbers of K. septempunctata spores (1.6 × 10¹, 1.6 × 10², and 1.6 × 10³ spores/g) significantly differed, suggesting that the consistency (e.g., solid, soft, or watery) and composition of feces might affect the efficiency of DNA extraction or amplification efficiency of the real-time PCR assay.

Of 45 clinical samples collected for the investigation of food-borne outbreaks associated with consumption of raw flounder, 2 vomitus and 21 fecal samples were positive for K. septempunctata DNA by our developed real-time PCR method. Although the C_T values of all 23 positive samples exceeded 30 cycles, single PCR products of the expected size were amplified from all positive samples and a nested PCR assay confirmed that all amplicons were specific for K. septempunctata 18S rDNA. These data indicate that that no false-positive results were obtained in the detection assays and that only a small amount of detectable K. septempunctata DNA was present in feces and vomitus from outbreak patients. In addition, since no PCR products were amplified from 41 fecal samples obtained from food-borne outbreaks epidemiologically unrelated to K. septempunctata, our real-time PCR assay using fecal matter appears to maintain the high specificity previously demonstrated for flounder tissue.

Our case file analyses of 13 food-borne outbreaks investigated in this study revealed that the median incubation time in all cases was 7 h or less. In addition, transient diarrhea and/or vomiting were the characteristic symptoms of patients. Together, the epidemiological data indicated that K. septempunctata was likely the etiological agent in these outbreaks (7). As predicted, in 6 of the 13 outbreaks, K. septempunctata spores were detected in the implicated olive flounder remnants or the same lot samples. However, because the food remnants were not obtained in the other seven cases, the causative food was not conclusively identified and other supporting evidence for the link between K. septempunctata and these outbreaks was required. As a result of the analyses performed here, in which K. septempunctata DNA was detected in at least one clinical sample from each of the seven cases, we have provided strong evidence that this pathogen was the causative disease agent in these seven outbreak cases. The present findings also indicate that our real-time PCR method using clinical samples is an effective tool for investigating K. septempunctata food-borne outbreaks.

The significant decrease in the detection rate of K. septempunctata in patient fecal specimens collected later than 2.5 days after the intake of causative food indicates that the length of the period from the time of consumption to specimen collection is an important factor for the successful detection of K. septempunctata using this assay. Myxosporean parasites, such as the well-studied Myxobolus cerebralis (8, 13), typically have a two-stage life cycle alternating between fish and annelids, although little is known about the life cycle of K. septempunctata within or outside of olive flounders. K. septempunctata outbreak patients generally recover within 24 h of the onset of symptoms and fluid accumulation induced by K. septempunctata spores does not persist longer than 4 h in the suckling mouse test (7). Based on these facts, we hypothesize that K. septempunctata likely has a similar life cycle to other myxosporean parasites and that mammals, such as humans and suckling mice, may have no association with the life cycle of this parasite. Therefore, K. septempunctata may not be able to colonize and proliferate in the mammalian gastrointestinal tract and be rapidly excreted in the feces. Our finding that K. septempunctata is predominantly detected in feces during only a short period in comparison with other diarrheagenic parasites, which infect and proliferate in the gastrointestinal tract, may support this hypothesis (2, 3, 11). However, further studies, such as investigating the mechanisms underlying the diarrheal activity induced by K. septempunctata in humans or suckling mice, are required to confirm this hypothesis and shed light on the life cycle of this emerging pathogen.

Since the consumption of raw fish is a traditional Japanese food custom, the confirmation that olive flounder infected with K. septempunctata may cause food-borne illness and outbreaks represents a major public health concern in Japan. Moreover, since Japanese foods prepared with raw fish, such as sushi and sashimi, are also popular in many developed countries, that threat of food-borne disease caused by K. septempunctata also exists outside of Japan. Thus, we anticipate that our presently described method for the sensitive detection of K. septempunctata in clinical samples may be an effective tool for the investigation of both domestic and international outbreaks caused by this novel pathogen.