sign in sign up
<<
Home , Cyclospora, Cyclospora cayetanensis, Toxoplasma gondii

1348

Journal of Food Protection, Vol. 77, No. 8, 2014, Pages 1348–1354 doi:10.4315/0362-028X.JFP-13-381 Copyright G, International Association for Food Protection

Efficacy of Wash Solutions in Recovering cayetanensis, parvum, and from Basil

VENESSA CHANDRA, MARIA TORRES, AND YNE´ S R. ORTEGA* Downloaded from http://meridian.allenpress.com/jfp/article-pdf/77/8/1348/1686703/0362-028x_jfp-13-381.pdf by guest on 30 September 2021 Center for , University of , 1109 Experiment Street, Griffin, Georgia 30223-1797, USA

MS 13-381: Received 13 September 2013/Accepted 23 March 2014

ABSTRACT Parasitic diseases can be acquired by ingestion of contaminated raw or minimally processed fresh produce (herbs and fruits). The sensitivity of methods used to detect parasites on fresh produce depends in part on the efficacy of wash solutions in removing them from suspect samples. In this study, six wash solutions (sterile E-Pure water, 3% levulinic acid–3% sodium dodecyl sulfate, 1 M glycine, 0.1 M phosphate-buffered saline, 0.1% Alconox, and 1% HCl–pepsin) were evaluated for their effectiveness in removing , , and Toxoplasma gondii from basil. One hundred or 1,000 oocysts of these parasites were inoculated onto the adaxial surfaces of 25 g of basil leaves, placed in stomacher bags, and stored for 1 h at 21uC or 24 h at 4uC. Leaves were hand washed in each wash solution for 1 min. DNA was extracted from the wash solutions and amplified using PCR for the detection of all parasites. Oocysts inoculated at a concentration of 1,000 oocysts per 25 g of basil were detected in all wash solutions. At an inoculum concentration of 100 oocysts per 25 g, oocysts were detected in 18.5 to 92.6% of the wash solutions. The lowest variability in recovering oocysts from basil inoculated with 100 oocysts was observed in 1% HCl–pepsin wash solution. Oocyst recovery rates were higher at 1 h than at 24 h postinoculation. Unlike most bacteria, parasites cannot be enriched; therefore, an optimal recovery process for oocysts from suspected foods is critical. The observations in this study provide guidance concerning the selection of wash solutions giving the highest retrieval of parasite oocysts.

Foodborne diseases are a major cause of illness and cayetanensis require time to sporulate and are not infectious death in the . Over the decades, there has been when excreted, Cryptosporidium oocysts undergo sporula- an increase in the number of reported cases of foodborne tion inside the and are infectious when shed into the illness linked to fresh produce. An epidemiologic investi- environment. Toxoplasma gondii oocysts in feces gation revealed that there were 190 produce-associated sporulate 24 to 48 h after being excreted. outbreaks from 1973 through 1997 and listed Cyclospora is acquired by ingestion of contaminated drinking water, and Escherichia coli O157:H7 as previously unrecognized fresh produce containing T. gondii oocysts, or meat causes of (34). Outbreaks of parasitic containing Toxoplasma tissue cysts. foodborne diseases are on the increase in parallel with the Numerous outbreaks of parasitic associated globalization of the food supply, increased consumption of with the consumption of contaminated fruits (primarily fresh produce, and increased international travel. Scallan raspberries) (14, 15, 19), fresh produce (basil, mesclun, et al. (32) estimated 9.4 million episodes of foodborne green onions, and lettuce) (2, 3, 13, 21, 24), and water (1, illnesses annually in the United States, in which 0.2 million 12, 22, 26) have been reported since 1990, with recurring illnesses are caused by parasites. Toxoplasma is one of outbreaks from contaminated Thai basil in the leading causes of foodborne illness requiring hospital- 2001 (16) and snow peas in 2005 (4) and ization (8%) and one of the leading causes of death (24%) outbreaks in 2003 and 2009 implicating not only unpas- (32). teurized milk and apple cider but also salads from salad Cyclospora cayetanensis and Cryptosporidium parvum bars, salad mixtures, and green onions (5, 6). A recent are emerging human that have been associated outbreak of cyclosporiasis involving more than 600 cases with numerous outbreaks of diarrheal illness. Both are has been linked to consumption of a salad mix (7). coccidian parasites that cause intracellular infections, According to a 2011 U.S. Food and Drug Administration primarily in the epithelial cells of the intestinal tract when (FDA) import alert, fresh produce, including basil, from infectious oocysts are ingested. While the oocysts of C. Mexico has been refused for admission to the United States as it appeared to contain C. cayetanensis (36). Although T. * Author for correspondence. Tel: 770-233-5587; Fax: 770-233-5587; gondii has not been implicated in foodborne outbreaks E-mail: [email protected]. linked to fresh produce, oocysts in feces of felines could J. Food Prot., Vol. 77, No. 8 RECOVERY OF PARASITES FROM BASIL 1349 contaminate water or soil, thereby possibly leading to HCl (Fischer, Fairlawn, NJ), pH 2.0, plus 6.4% pepsin (Fischer, contamination of produce. Nazareth, PA). The PBS solution contained 0.14 M sodium Unlike bacteria, parasites do not multiply in the chloride, 0.002 M potassium phosphate monobasic, 0.003 M environment. Because contaminated foods may contain potassium chloride, and 0.01 M sodium phosphate dibasic in very low numbers of oocysts, very sensitive isolation and E-Pure water. detection methods are essential. Detection of C. cayetanen- Recovery of oocysts. Bags containing inoculated samples and sis, Cryptosporidium spp., and T. gondii oocysts in fresh wash solutions (200 ml of E-Pure water, 3% levulinic acid–3% produce can be done using techniques or PCR. SDS, 1 M glycine, 0.1 M PBS, or 0.1% Alconox or 100 ml of 1% However, these methods are highly dependent on the HCl–pepsin) were vigorously shaken by hand for 15 s, hand efficient recovery of oocysts from produce tissues. Washing rubbed for 30 s, and shaken vigorously for another 15 s. Wash of fresh produce with elution buffers has been used to solutions were decanted into 50-ml conical tubes. Stomacher bags remove oocysts. Typically, the fresh produce is washed with containing basil leaves were squeezed by hand to remove as much a solution of choice by manually agitating the sample for a of the wash solution as possible. Oocysts in the wash solution were Downloaded from http://meridian.allenpress.com/jfp/article-pdf/77/8/1348/1686703/0362-028x_jfp-13-381.pdf by guest on 30 September 2021 period of time. The resulting wash solutions are concen- concentrated by centrifugation (Allegra 6 Centrifuge, Beckman trated by centrifugation and further analyzed by microscopy Coulter, Palo Alto, CA) for 15 min at 2,060 | g. Most of the or PCR. supernatant wash solution was decanted, leaving 3 to 4 ml in each The purpose of this study was to determine the tube. The resultant pellets from each tube were pooled and transferred to a sterile 15-ml conical tube, followed by centrifu- effectiveness of several wash solutions in recovering C. gation as described above. The supernatant wash solution was cayetanensis, C. parvum, and T. gondii from basil. aspirated, and the pellet was brought to 1 ml with sterile E-Pure Optimization of elution methods has implications for water in a 1.5-ml microcentrifuge tube and stored at 4uC. For each outbreak investigations, as well as for surveillance studies. wash solution, one uninoculated control sample was treated as Basil was selected because it has been implicated as a described above. vehicle in numerous outbreaks of cyclosporiasis. DNA extraction. DNA was extracted from wash solution MATERIALS AND METHODS samples using a FastDNA Spin for Soil Kit (MP Biomedicals, Irvine, CA) with slight modifications to the manufacturer’s Parasites. C. cayetanensis oocysts were obtained from a protocol as follows: 200 ml of each sample was added to a Lysing naturally infected individual from Peru courtesy of Dr. Robert Matrix E Tube (MP Biomedicals). Samples were processed in a Gilman, A.B. Prisma, and C. parvum oocysts were obtained from infected calves and kindly provided by the Centers for Disease FastPrep FP120 (Thermo Savant, Holbrook, NY) at a speed setting Control and Prevention (CDC). The oocysts of both parasites were of 5.5 for 30 s. All samples were centrifuged (Eppendorf stored in 2.5% potassium dichromate. T. gondii oocysts were Centrifuge 5415 D, Eppendorf AG, Hamburg, Germany) at | | obtained courtesy of Dr. J. P. Dubey, U.S. Department of 15,700 g instead of 18,200 g as stated in the manufacturer’s Agriculture, and were stored in 2% sulfuric acid. The oocysts of instructions. The final sample volume for the DNA extract (80 ml) each parasite were counted (10 repetitions each) using a Bright- was stored at 220uC. Positive controls were extracted simulta- Line hemacytometer (Hausser Scientific, Horsham, PA), further neously. diluted to give 100 and 1,000 oocysts per 100 ml, and used to inoculate the basil leaves. PCR. A nested PCR assay was used to amplify fragments of the 18S rRNA gene of Cyclospora (20). For the primary PCR, 1 ml Inoculation of basil. Sweet basil stems (Tom Pontano & of DNA extract from each sample was used. One microliter of this Sons Farms, Vineland, NJ, Pontano Farms, Inc., Boynton Beach, primary reaction mixture was used for the secondary nested PCR. FL, and Rock Garden, Miami, FL) were purchased from a farmers’ The amplified product was about 500 bp. A nested PCR protocol market in Atlanta, GA, and stored at 4uC for no longer than 3 days for Cryptosporidium was used to amplify fragments of the 18S prior to inoculation. Leaves (25-g samples) were spot inoculated on rRNA gene (38). The template for the primary PCR was 1 mlof the adaxial surface with 100 ml of E-Pure water containing either DNA extract of each sample, and 2 ml of the primary reaction 100 or 1,000 oocysts of C. cayetanensis or C. parvum and placed mixture was used as the DNA template for the secondary nested in sterile stomacher bags (Labplas, Inc., Ste-Julie, Quebec, PCR. The internal primers yielded a product of about 826 bp. A Canada). T. gondii was inoculated with 100 ml of E-Pure water PCR for T. gondii was used to amplify a product of 529 bp at a concentration of 100 oocysts per 25 g of leaves. Inoculation targeting a repetitive DNA fragment of T. gondii as described by was done with caution to ensure that the droplets were Homan et al. (17). PCRs were done in a Mastercycler gradient homogeneously distributed on the surface of the leaves. Inoculated thermal cycler (Eppendorf 5331, Eppendorf, Westbury, NY). The leaves were kept at room temperature (21 ¡ 2uC) for 1 h or at 4uC PCR products, along with positive and negative controls and a 100- for 24 h prior to testing the efficacies of wash solutions; leaves bp DNA ladder (Quick-Load, New England Biolabs, Ipswich, inoculated with T. gondii were kept at 4uC for 24 h before testing MA), were electrophoresed in 1.5% agarose gels (Bio-Rad) wash solutions. Three inoculated samples were washed with each containing ethidium bromide (Sigma) and visualized using a test solution, and the experiments were repeated two or three times. White/UV TMW-20 Transilluminator.

Wash solutions. Six wash solutions were evaluated in this DNA purification and sequencing. Amplification products study, as follows: sterile E-Pure water (ThermoScientific, Marietta, were purified using a QIAquick PCR Purification Kit (Qiagen, OH); 3% levulinic acid (Sigma, St. Louis, MO) plus 3% sodium Valencia, CA) according to the manufacturer’s protocol. The dodecyl sulfate (SDS; Bio-Rad, Hercules, CA); 1 M glycine purified samples were sent to Macrogen USA, Rockville, MD, for (Sigma), pH 5.5; 0.1 M phosphate-buffered saline (PBS; Sigma), sequencing to confirm the identity of each parasite. DNA pH 7.0; 0.1% Alconox (Alconox, Inc., White Plains, NY); and 1% sequencing was performed in both directions using the same 1350 CHANDRA ET AL. J. Food Prot., Vol. 77, No. 8

TABLE 1. Identification rates of Cyclospora cayetanensis by PCR after washing of oocyst-inoculated basil leaves using six wash solutions

Mean % (¡SD) of samples PCR positive at indicated postinoculation time

1,000 oocysts/25 g of leavesa 100 oocysts/25 g of leavesb

Wash solution 1hc 24 hd 1 h 24 h

E-Pure water 88.9 (¡15.7) 88.9 (¡15.7) 37.0 (¡17.0) 37.0 (¡17.0) 3% levulinic acid–3% SDS 83.3 (¡23.6) 72.2 (¡7.9) 22.2 (¡31.4) 27.8 (¡7.9) 1 M glycine, pH 5.5 88.9 (¡15.7) 94.4 (¡7.9) 40.7 (¡17.0) 44.4 (¡22.2) 0.1 M PBS, pH 7 94.4 (¡7.9) 66.7 (¡47.1) 22.2 (¡15.7) 50.0 (¡23.6) 0.1% Alconox 94.4 (¡7.9) 77.8 (¡31.4) 18.5 (¡17.0) 37.0 (¡12.8) 1% HCl–pepsin 100.0 (¡0.0) 100.0 (¡0.0) 40.7 (¡17.0) 51.9 (¡6.4) a (Number of PCR-positive samples/total number of samples analyzed in two replicate tests) | 100. Downloaded from http://meridian.allenpress.com/jfp/article-pdf/77/8/1348/1686703/0362-028x_jfp-13-381.pdf by guest on 30 September 2021 b (Number of PCR-positive samples/total number of samples analyzed in three replicate tests) | 100. c 1 h, inoculated basil leaves were incubated at 21uC for 1 h prior to analysis. d 24 h, inoculated basil leaves were incubated at 4uC for 24 h prior to analysis.

PCR primers as for the original amplifications. The DNA particles and the potential for the presence of PCR sequences were then compared with the collection of sequences inhibitors. The manual shaking method was selected for in GenBank using BLASTN. use in subsequent experiments. The identification rates of C. cayetanensis in the six Microscopy. Several PCR-positive samples were selected for solutions used to wash 25 g of basil leaves inoculated with microscopy analysis for confirmation of the presence of the parasites in the wash samples. A white/UV excitation filter of 330 1,000 or 100 oocysts per 25 g are shown in Table 1. The to 385 nm was used to visualize Cyclospora and Toxoplasma lowest recoveries from leaves inoculated with 1,000 oocysts oocysts. Oocysts were also observed by differential interference per 25 g were achieved using 3% levulinic acid–3% SDS contrast microscopy at |400 magnification. Cryptosporidium and 0.1 M PBS, pH 7.0. These solutions were not further oocysts were detected using the Merifluo Cryptosporidium/Giardia evaluated for removing C. parvum from leaves. Only four Kit (Meridian Bioscience, Inc., Cincinnati, OH) following the wash solutions (E-Pure water, 1 M glycine, 0.1% Alconox, manufacturer’s instructions. and 1% HCl–pepsin) were evaluated in subsequent trials. The recovery of C. cayetanensis and C. parvum oocysts was Statistical analysis. Each test consisted of four or six wash determined at 1 and 24 h postinoculation. Based on the solutions from three inoculated samples. Each sample was results for C. cayetanensis, the studies of T. gondii were analyzed in triplicate during PCR amplification. All experiments restricted to 24 h postinoculation with the concentration of were repeated two or three times. The PCR results for each wash solution were compiled in Excel (Microsoft, Redmond, WA), and 100 oocysts per 25 g. The recovery of T. gondii oocysts was the percentage of recovery was determined. The mean percentage determined using all six wash solutions. of recovery from two or three replicate tests was determined for The identification rates 1 h after inoculating leaves with each wash solution. The normal distribution of the results was 1,000 C. cayetanensis oocysts per 25 g were consistently assessed by normal probability plots and the Shapiro-Wilk test. higher than those from leaves at 24 h postinoculation The significance of the effects of time elapsed between inoculation (Table 1). This may have been due to the shorter time for and application of wash solutions and of efficacies among wash oocysts to adhere to the leaf surfaces, thus resulting in more solutions was analyzed using R (version 2.15.2) at a 95% oocysts being removed during the washing process 1 h confidence interval. postinoculation, or to stronger attachment of oocysts at a refrigerated temperature (4uC) during the 24-h period RESULTS AND DISCUSSION following inoculation. When the leaves were inoculated with A preliminary evaluation of sample processing methods 1,000 oocysts per 25 g and washed with 1% HCl–pepsin 1 or was done in an attempt to improve oocyst recovery. One 24 h postinoculation, the mean recovery rate was 100%.The technique involved manually shaking the inoculated sam- overall recovery rates from leaves inoculated with 100 C. ples for 15 s, hand rubbing for 30 s, and shaking again for cayetanensis oocysts were higher in samples analyzed at 1 h another 15 s. The other method involved the use of a rocker than at 24 h postinoculation, with the highest recovery rate platform, in which the inoculated leaves were gently (51.9%) achieved using 1% HCl–pepsin. There were no agitated for 7.5 min, followed by inverting the bag and significant differences (P . 0.05) among the wash solutions. agitating for an additional 7.5 min. At a low concentration C. cayetanensis oocysts were not detected in the negative- of 100 C. cayetanensis oocysts per 25 g, the two methods control samples. resulted in recovery rates of 28 to 52% and 0 to 44%, The efficacy of wash solutions in removing C. cayeta- respectively (data not shown). The manual shaking method nensis oocysts from basil leaves inoculated at 100 oocysts per provided more abrasion of basil leaf surfaces, which 25 g was further evaluated. PCR-positive wash solutions from possibly could have facilitated the removal of oocysts. leaves 1 h postinoculation are shown in Figure 1A. In the first The elution of parasites comes with an increase of debris replicate, only two of nine inoculated samples were PCR J. Food Prot., Vol. 77, No. 8 RECOVERY OF PARASITES FROM BASIL 1351 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/77/8/1348/1686703/0362-028x_jfp-13-381.pdf by guest on 30 September 2021

FIGURE 1. Number of wash solutions that were PCR positive after washing basil leaves inoculated at a concentration of 100 C. cayetanensis or C. parvum oocysts per 25 g and kept at 21uC for 1 h. A sample was considered positive when at least one PCR of three in each replicate was positive. positive when leaves were washed with E-Pure water, glycine, The identification rate (%) of wash solutions positive for and 1% HCl–pepsin. Three of nine samples were positive oocysts was equal to the sum of the number of PCR-positive when washed in PBS, and none was PCR positive using 3% samples divided by the total number of PCR samples in three levulinic acid–3% SDS or 0.1% Alconox. The large amount replicates of each experiment, which totaled 27, since each of foam produced by 3% levulinic acid–3% SDS and 0.1% PCR amplification was performed in triplicate. At an inoculum Alconox during washing may have interfered with the concentration of 100 oocysts per 25 g, the mean recovery rates recovery of the small number of oocysts. In the second ranged from 18.5 to 51.9% for C. cayetanensis (Table 1) and replicate, more than four of nine 3% levulinic acid–3% SDS, 33.3 to 70.4% for C. parvum (Table 2), with 1% HCl–pepsin glycine, and 1% HCl–pepsin wash solutions were PCR and 0.1% Alconox, respectively, showing the highest recovery positive. In the third trial, water and glycine had five PCR- rates. The mean recovery rates of T. gondii were from 74.1 to positive samples, whereas 1% HCl–pepsin had four (Fig. 1A). 92.6%, with the highest recovery achieved using 1% HCl– In a follow-up study using leaves 24 h postinoculation with pepsin (Table 3). 100 oocysts per 25 g, 1% HCl–pepsin showed an increase in When a sample was considered to be positive if at least PCR-positive samples and had consistently more PCR- one of the nine PCR readings was positive (the total number positive samples (Fig. 2A). Overall, oocyst recoveries using of PCR readings in three repetitions is nine), the mean 3% levulinic acid–3% SDS and PBS were highly inconsistent. recovery rates ranged from 50 to 100% for C. cayetanensis Table 2 shows the identification rates from basil leaves and 56 to 100% for C. parvum in wash solutions from basil inoculated with 1,000 C. parvum oocysts per 25 g and kept inoculated with 100 oocysts per 25 g; the 1% HCl–pepsin for 1 and 24 h before washing with E-Pure water, 1 M and 0.1% Alconox wash solutions showed the highest glycine, 0.1% Alconox, or 1% HCl–pepsin. With the recovery rates for C. cayetanensis and C. parvum, exception of glycine wash solutions used on samples 1 h respectively. T. gondii recoveries were 89 to 100% using postinoculation, all wash solutions from leaves analyzed at 1% HCl–pepsin. Even though the mean recovery rates were both postinoculation times contained oocysts (100% higher using this analytical approach, the results are in recovery). Since all wash solutions were positive for agreement with those leading to the conclusion that washing oocysts, only two replicate experiments were performed. basil leaves with 1% HCl–pepsin or 0.1% Alconox gives Table 2 also shows the recovery efficiencies when 100 C. higher recovery rates. parvum oocysts per 25 g were applied to leaves. The mean Cyclospora and Cryptosporidium oocysts were identi- percentages of wash solutions positive for oocysts were fied in a selected number of PCR-positive samples that were slightly higher in samples 1 h postinoculation than in examined by microscopy. In the group of 1,000 oocysts per samples 24 h postinoculation, with the highest being 70.4% 25 g of leaves, 100% contained Cyclospora oocysts and recovery from leaves washed with 0.1% Alconox at 1 h. 80% Cryptosporidium oocysts. In the group with 100 However, no significant differences were observed between oocysts per 25 g leaves, 40, 20, and 30% contained postinoculation times or among the wash solutions (P . Cyclospora, Cryptosporidium, and Toxoplasma oocysts, 0.05). The numbers of wash samples PCR positive for C. respectively. parvum (Fig. 1B and 2B) and C. cayetanensis (Fig. 1A and According to studies done by Robertson and Gjerde 2A) were similar within each parasite and postinoculation (30), between one and six Cryptosporidium oocysts per sampling time. The recovery rates for T. gondii (Fig. 1C) at 100 g of produce can be detected. In our study, the nested 24 h postinoculation were similar to those for C. PCR assays detected lower numbers of oocysts, ranging cayetanensis and C. parvum. from two to less than one oocyst. Primers developed by 1352 CHANDRA ET AL. J. Food Prot., Vol. 77, No. 8

Several studies have examined the ability of various wash solutions and techniques to recover Cyclospora and Cryptosporidium from foodstuffs (8–10, 28, 29, 31).Ina recent study, Shields et al. (33) determined the recovery rates of both parasites using wash solutions containing three laboratory glassware detergents. Foods (fruits, herbs, and leafy greens) were inoculated with amounts of oocyst ranging from 3 log oocysts per 30 mg of basil leaves to 6 log oocysts per 25 g of leafy greens. The highest recovery rates (70 to 80%) were achieved with 0.1% Alconox. In our study, recovery rates of 78 to 100% were achieved when 0.1% Alconox was used to wash basil leaves inoculated at a concentration of 1,000 oocysts per 25 g. Downloaded from http://meridian.allenpress.com/jfp/article-pdf/77/8/1348/1686703/0362-028x_jfp-13-381.pdf by guest on 30 September 2021 As Shields et al. (33) noted, Alconox contains sodium dodecylbenzenesulfonate (C12H25C6H4SO3Na), a surfactant found in laundry detergents, and tetrasodium pyrophosphate (Na4P2O7), a food additive with emulsifying and dispersing properties. The mixture of these compounds appears to promote the removal of oocysts from the basil leaf surface. The efficiency of the recovery of oocysts with 0.1% Alconox, however, was highly variable. Although it worked well, the foam produced during the washing could have contributed to a loss of oocysts during the recovery process. Data from this study indicate that the recovery efficiencies for C. parvum and T. gondii oocysts per 25 g of basil were higher than those for C. cayetanensis.Itis presumed that Cyclospora oocysts exhibit some stickiness, thus enabling better adherence to the leaf surface. Other characteristic differences between oocyst morphologies may have also contributed to the differences in recovery rates. Several studies have examined the time elapsed between inoculating and washing foods for the efficacy of recovery of oocysts, but there is no standard time universally used. Cook et al. (8–10) and Robertson et al. (28, 29, 31) mentioned that the times between inoculation and washing were 1 to 3 h. In another study (24), lettuce leaves were washed with distilled water 24 h postinocula- tion. Our study evaluated postinoculation times of 1 and 24 h, and although no significant difference was observed between postinoculation times, further research is needed to determine the appropriate period of time between contam- ination and evaluation of wash solutions that would mimic practices on a commercial level. The FDA suggests a time of 72 h postinoculation for food commodities held under FIGURE 2. Number of wash solutions (of nine tested) that were refrigerated conditions (37). Increasing the time between PCR positive after washing basil leaves inoculated at a concentra- inoculating and washing produce, however, could result in tion of 100 C. cayetanensis, C. parvum, or T. gondii oocysts per 25 g better attachment, thereby affecting efficacies in removal. and kept at 4uC for 24 h. A sample was considered positive when at Studies done with human volunteers revealed that least one PCR of three in each replicate was positive. Cryptosporidium has a low infectious dose of 10 to 30 oocysts (11, 23). Although the infectious dose of Cyclos- Relman et al. (27) were initially used for detecting C. pora has not been determined, it is assumed to be low as cayetanensis in our study. However, these primers did not well (35). Therefore, it is important to be able to identify result in strong amplification. Thus, primers used by Li et al. very low numbers of oocysts during the investigation of (20) were selected as an alternative, as they were more outbreaks. Moreover, it is assumed that oocyst recoveries sensitive. The secondary primers successfully amplified the can vary, depending on factors such as the composition of targeted gene and yielded a product of about 500 bp. wash solutions and differences in food matrices and surface Selected amplified PCR products were then sequenced to morphology that could influence the oocyst adherence. The confirm that the amplified product was in fact from C. surface of raspberries, for example, which have been cayetanensis. associated with several outbreaks of cyclosporiasis, aids in J. Food Prot., Vol. 77, No. 8 RECOVERY OF PARASITES FROM BASIL 1353

TABLE 2. Identification rates of Cryptosporidium parvum by PCR after washing of oocyst-inoculated basil leaves using four wash solutions

Mean % (¡SD) of samples PCR positive at indicated postinoculation time

1,000 oocysts/25 g of leavesa 100 oocysts/25 g of leavesb

Wash solution 1hc 24 hd 1 h 24 h

E-Pure water 100.0 (¡0.0) 100.0 (¡0.0) 55.6 (¡29.4) 59.3 (¡12.8) 1 M glycine, pH 5.5 88.9 (¡15.7) 100.0 (¡0.0) 51.9 (¡6.4) 33.3 (¡11.1) 0.1% Alconox 100.0 (¡0.0) 100.0 (¡0.0) 70.4 (¡28.0) 51.9 (¡35.7) 1% HCl–pepsin 100.0 (¡0.0) 100.0 (¡0.0) 59.3 (¡23.1) 55.6 (¡19.2) a (Number of PCR-positive samples/total number of samples analyzed in two replicate tests) | 100. b (Number of PCR-positive samples/total number of samples analyzed in three replicate tests) | 100. Downloaded from http://meridian.allenpress.com/jfp/article-pdf/77/8/1348/1686703/0362-028x_jfp-13-381.pdf by guest on 30 September 2021 c 1 h, inoculated basil leaves were stored at 21uC for 1 h prior to analysis. d 24 h, inoculated basil leaves were stored at 4uC for 24 h prior to analysis. the adherence of oocysts to the fruit. Raspberries have using other foods, such as raspberries, and the effect of crevices and fine hair-like projections (trichomes) that may extended postinoculation times on rates of recovery. enhance oocyst entrapment (18, 25). Hence, it is important to determine the performance of wash solutions in different ACKNOWLEDGMENTS food matrices. This project was supported by a seed grant from Center for Food Although data from this study indicate that no Safety, the University of Georgia, and by a Hatch USDA-NIFA grant. significant differences were observed among the wash solutions, one wash solution might perform better than REFERENCES others if used under certain circumstances. In the case of 1. Anonymous. 1991. Outbreaks of diarrheal illness associated with oocyst recovery from foodstuffs which contain fats or lipids, (blue-green algae)-like bodies—Chicago and Nepal, 1989 and 1990. Morb. Mortal. Wkly. Rep. 40:325–327. wash solutions that have detergent properties (e.g., Alco- 2. Anonymous. 1997. Outbreak of cyclosporiasis—Northern Virginia- nox) might be more effective in removing oocysts. These Washington, D.C.–Baltimore, Maryland, metropolitan area, 1997. variables deserve further study. Morb. Mortal. Wkly. Rep. 46:689–691. With the emergence of parasitic foodborne illnesses 3. Anonymous. 1998. Foodborne outbreak of cryptosporidiosis— associated with the consumption of fresh produce, an Spokane, Washington, 1997. Morb. Mortal. Wkly. Rep. 47:565–567. 4. Anonymous. 2004. Outbreak of cyclosporiasis associated with snow efficient elution method to retrieve oocysts from the peas—Pennsylvania, 2004. Morb. Mortal. Wkly. Rep. 53:876–878. suspected produce is crucial. Since C. cayetanensis, C. 5. Anonymous. 2011. Cryptosporidiosis outbreak at a summer camp— parvum, and T. gondii have low infectious doses and North Carolina, 2009. Morb. Mortal. Wkly. Rep. 60:918–922. enrichment procedures for these parasites are not available, 6. Blackburn, B. G., J. M. Mazurek, M. Hlavsa, J. Park, M. Tillapaw, it is important to develop methods to maximize their M. Parrish, E. Salehi, W. Franks, E. Koch, F. Smith, L. Xiao, M. Arrowood, V. Hill, A. da Silva, S. Johnston, and J. L. Jones. 2006. detection. Our study shows that 1% HCl–pepsin had a lower Cryptosporidiosis associated with ozonated apple cider. Emerg. variability in recovering oocysts at 24 h postinoculation Infect. Dis. 12:684–686. from basil leaves to which 100 oocysts per 25 g had been 7. Centers for Disease Control and Prevention. 2013. Investigation of an applied, and the nested PCR assays detected low numbers of outbreak of cyclosporiasis in the United States. Available at: http:// www.cdc.gov/parasites/cyclosporiasis/outbreaks/investigation-2013.html. oocysts, ranging from 2 to less than 1 oocyst. These findings Accessed 20 August 2013. are important when developing methods to improve the 8. Cook, N., R. A. Nichols, N. Wilkinson, C. A. Paton, K. Barker, and sensitivity of foodborne parasite detection. More studies are H. V. Smith. 2007. Development of a method for detection of Giardia needed to determine the recovery efficiencies of parasites duodenalis cysts on lettuce and for simultaneous analysis of salad products for the presence of Giardia cysts and Cryptosporidium oocysts. Appl. Environ. Microbiol. 73:7388–7391. TABLE 3. Identification rates of Toxoplasma gondii by PCR 9. Cook, N., C. A. Paton, N. Wilkinson, R. A. Nichols, K. Barker, and H. V. after washing of oocyst-inoculated basil leaves using six Smith. 2006. Towards standard methods for the detection of Cryptospo- wash solutions ridium parvum on lettuce and raspberries. Part 1. Development and optimization of methods. Int. J. Food Microbiol. 109:215–221. Mean % (¡SD) of 10. Cook, N., C. A. Paton, N. Wilkinson, R. A. Nichols, K. Barker, and Wash solution samples PCR positivea H. V. Smith. 2006. Towards standard methods for the detection of ¡ Cryptosporidium parvum on lettuce and raspberries. Part 2. E-Pure water 85.2 ( 25.7) Validation. Int. J. Food Microbiol. 109:222–228. ¡ 3% levulinic acid–3%SDS 81.5 ( 23.1) 11. DuPont, H. L., C. L. Chappell, C. R. Sterling, P. C. Okhuysen, J. B. 1 M glycine, pH 5.5 85.2 (¡6.4) Rose, and W. Jakubowski. 1995. The infectivity of Cryptosporidium 0.1 M PBS, pH 7 74.1 (¡25.7) parvum in healthy volunteers. N. Engl. J. Med. 332:855–859. 0.1% Alconox 77.8 (¡11.1) 12. Hayes, E. B., T. D. Matte, T. R. O’Brien, T. W. McKinley, G. S. 1% HCl–pepsin 92.6 (¡12.8) Logsdon, J. B. Rose, B. L. Ungar, D. M. Word, M. A. Wilson, E. G. Long, E. S. Hurwitz, and D. D. Juranek. 1989. Large community a (Number of PCR-positive samples/total number of samples outbreak of cryptosporidiosis due to contamination of a filtered public analyzed in three replicates) | 100. water supply. N. Engl. J. Med. 320:1372–1376. 1354 CHANDRA ET AL. J. Food Prot., Vol. 77, No. 8

13. Herwaldt, B. L. 2000. Cyclospora cayetanensis: a review, focusing 25. Ortega, Y. R., and R. Sanchez. 2010. Update on Cyclospora on the outbreaks of cyclosporiasis in the 1990s. Clin. Infect. Dis. 31: cayetanensis, a food-borne and waterborne parasite. Clin. Microbiol. 1040–1057. Rev. 23:218–234. 14. Herwaldt, B. L., M. L. Ackers, and the Cyclospora Working Group. 26. Rabold, J. G., C. W. Hoge, D. R. Shlim, C. Kefford, R. Rajah, and P. 1997. An outbreak in 1996 of cyclosporiasis associated with imported Echeverria. 1994. Cyclospora outbreak associated with chlorinated raspberries. N. Engl. J. Med. 336:1548–1556. drinking water. Lancet 344:1360–1361. 15. Herwaldt, B. L., M. J. Beach, the Cyclospora Working Group. 1999. 27. Relman, D. A., T. M. Schmidt, A. Gajadhar, M. Sogin, J. Cross, K. The return of Cyclospora in 1997: another outbreak of cyclosporiasis Yoder, O. Sethabutr, and P. Echeverria. 1996. Molecular phyloge- in North America associated with imported raspberries. Ann. Intern. netic analysis of Cyclospora, the human intestinal , suggests Med. 130:210–220. that it is closely related to species. J. Infect. Dis. 173:440– 16. Hoang, L. M., M. Fyfe, C. Ong, J. Harb, S. Champagne, B. Dixon, 445. and J. Isaac-Renton. 2005. Outbreak of cyclosporiasis in British 28. Robertson, L. J., and B. Gjerde. 2000. Isolation and enumeration of Columbia associated with imported Thai basil. Epidemiol. Infect. 133: Giardia cysts, Cryptosporidium oocysts, and Ascaris eggs from fruits 23–27. and vegetables. J. Food Prot. 63:775–778. 17. Homan, W. L., M. Vercammen, J. De Braekeleer, and H. 29. Robertson, L. J., and B. Gjerde. 2001. Factors affecting recovery Verschueren. 2000. Identification of a 200- to 300-fold repetitive efficiency in isolation of Cryptosporidium oocysts and Giardia cysts Downloaded from http://meridian.allenpress.com/jfp/article-pdf/77/8/1348/1686703/0362-028x_jfp-13-381.pdf by guest on 30 September 2021 529 bp DNA fragment in Toxoplasma gondii, and its use for from vegetables for standard method development. J. Food Prot. 64: diagnostic and quantitative PCR. Int. J. Parasitol. 30:69–75. 1799–1805. 18. Kniel, K. E., D. S. Lindsay, S. S. Sumner, C. R. Hackney, M. D. 30. Robertson, L. J., and B. Gjerde. 2001. Occurrence of parasites on fruits and vegetables in Norway. J. Food Prot. 64:1793–1798. Pierson, and J. P. Dubey. 2002. Examination of attachment and 31. Robertson, L. J., B. Gjerde, and A. T. Campbell. 2000. Isolation of survival of Toxoplasma gondii oocysts on raspberries and blueberries. Cyclospora oocysts from fruits and vegetables using lectin-coated J. Parasitol. 88:790–793. paramagnetic beads. J. Food Prot. 63:1410–1414. 19. Koumans, E. H., D. J. Katz, J. M. Malecki, S. Kumar, S. P. 32. Scallan, E., R. M. Hoekstra, F. J. Angulo, R. V. Tauxe, M. A. Wahlquist, M. J. Arrowood, A. W. Hightower, and B. L. Herwaldt. Widdowson, S. L. Roy, J. L. Jones, and P. M. Griffin. 2011. 1998. An outbreak of cyclosporiasis in in 1995: a harbinger of Foodborne illness acquired in the United States—major pathogens. multistate outbreaks in 1996 and 1997. Am. J. Trop. Med. Hyg. 59: Emerg. Infect. Dis. 17:7–15. 235–242. 33. Shields, J. M., M. M. Lee, and H. R. Murphy. 2012. Use of a 20. Li, G., S. Xiao, R. Zhou, W. Li, and H. Wadeh. 2007. Molecular common laboratory glassware detergent improves recovery of characterization of -like organism from dairy cattle. Cyclospora Cryptosporidium parvum and Cyclospora cayetanensis from lettuce, Parasitol. Res. 100:955–961. herbs and raspberries. Int. J. Food Microbiol. 153:123–128. 21. Lopez, A. S., D. R. Dodson, M. J. Arrowood, P. A. Orlandi, Jr., A. J. 34. Sivapalasingam, S., C. R. Friedman, L. Cohen, and R. V. Tauxe. da Silva, J. W. Bier, S. D. Hanauer, R. L. Kuster, S. Oltman, M. S. 2004. Fresh produce: a growing cause of outbreaks of foodborne Baldwin, K. Y. Won, E. M. Nace, M. L. Eberhard, and B. L. illness in the United States, 1973 through 1997. J. Food Prot. 67: Herwaldt. 2001. Outbreak of cyclosporiasis associated with basil in 2342–2353. Missouri in 1999. Clin. Infect. Dis. 32:1010–1017. 35. Sterling, C. R., and Y. R. Ortega. 1999. Cyclospora: an enigma worth 22. MacKenzie, W. R., W. L. Schell, K. A. Blair, D. G. Addiss, D. E. unraveling. Emerg. Infect. Dis. 5:48–53. Peterson, N. J. Hoxie, J. J. Kazmierczak, and J. P. Davis. 1995. 36. U.S. Food and Drug Administration. 26 March 2012. Import Alert Massive outbreak of waterborne Cryptosporidium in 28-11. Available at: http://www.accessdata.fda.gov/cms_ia/import Milwaukee, : recurrence of illness and risk of secondary alert_95.html. Accessed 27 February 2013. transmission. Clin. Infect. Dis. 21:57–62. 37. U.S. Food and Drug Administration, Office of Foods. 8 September 23. Okhuysen, P. C., C. L. Chappell, J. H. Crabb, C. R. Sterling, and 2011. Guidelines for the verification of analytical methods for the H. L. DuPont. 1999. Virulence of three distinct Cryptospori- detection of microbial pathogens in foods. Available at: http://www. dium parvum isolates for healthy adults. J. Infect. Dis. 180: fda.gov/downloads/ScienceResearch/FieldScience/UCM273418.pdf. 1275–1281. Accessed 13 August 2013. 24. Ortega, Y. R., C. R. Roxas, R. H. Gilman, N. J. Miller, L. Cabrera, C. 38. Xiao, L., U. M. Morgan, J. Limor, A. Escalante, M. Arrowood, W. Taquiri, and C. R. Sterling. 1997. Isolation of Cryptosporidium parvum Shulaw, R. C. Thompson, R. Fayer, and A. A. Lal. 1999. Genetic and Cyclospora cayetanensis from vegetables collected in markets of diversity within Cryptosporidium parvum and related Cryptosporid- an endemic region in Peru. Am. J. Trop. Med. Hyg. 57:683–686. ium species. Appl. Environ. Microbiol. 65:3386–3391.