561 Journal of Food Protection, Vol. 81, No. 4, 2018, Pages 561–568 doi:10.4315/0362-028X.JFP-17-354 Copyright Ó, International Association for Food Protection
Research Paper
Development and Validation of a New Protocol for Detecting and Recovering Clostridium difficile from Meat Samples MAJDA BIASIZZO,* STANKA VADNJAL, URSKA HENIGMAN, MANJA KRIZMAN, ANDREJ KIRBIS, URSKA JAMNIKAR-CIGLENECKI
AND
Institute of Food Safety, Feed and Environment, Veterinary Faculty, University of Ljubljana, Gerbiˇceva 60, 1000 Ljubljana, Slovenia MS 17-354: Received 22 August 2017/Accepted 31 October 2017/Published Online 8 March 2018
ABSTRACT There is no recommended protocol for detecting and isolating Clostridium difficile present in food samples. Here, we have evaluated the recovery of C. difficile in meat samples after incubating them in various enrichment broths. The media were as follows: cycloserine-cefoxitin fructose broth supplemented with taurocholic acid, D-cycloserine, cefoxitin, and lysozyme; cycloserine-cefoxitin mannitol broth with taurocholate and lysozyme; and cycloserine-cefoxitin fructose broth supplemented with taurocholic acid, C. difficile moxalactam norfloxacin selective supplement, and lysozyme. Samples were inoculated with various strains and quantities of C. difficile and then enriched in the different broths for 1, 4, and 7 days. C. difficile was isolated on agar plates and detected with quantitative real-time PCR (qPCR). The procedure using enrichment in cycloserine-cefoxitin fructose broth supplemented with taurocholic acid, D-cycloserine, cefoxitin, and lysozyme and incubation for 4 days for qPCR detection and 7 days for isolation (plating on C. difficile agar base with added C. difficile selective supplement and 7% [v/v] defibrinated horse blood after alcoholic shock and centrifugation) was validated. Samples of different kinds of meat and meat preparation were contaminated and used for validation of the chosen protocol. The sensitivity of detection with qPCR was 100%, and the sensitivity of the isolation method was 96%. Key words: Clostridium difficile; Detection method; Enrichment; Food; Isolation method; Meat samples
Clostridium difficile is a gram-positive, spore-forming, anaerobic bacterium capable of causing disease in animals and humans that can range from asymptomatic colonization, diarrhea, and colitis to severe disease syndromes and even bowel perforation, sepsis, and death (11, 32). C. difficile infection is the most commonly diagnosed cause of antimicrobial-associated and hospital-associated diarrhea in humans (9). The incidence and severity of disease seem to have increased in the last decade, and cases have also been reported in populations that were previously thought to be at low risk, such as young individuals and pregnant women (32). About 32% of cases reported have been classified as community associated (11, 20). Asymptomatic colonization of healthy individuals has also been reported (10, 24). Recently published data that compared human, animal, and food isolates of C. difficile strongly suggested that animal reservoirs and transmission via foods are possible sources of community-associated infection (30, 31), although there is still no clear evidence that contaminated food has led to clinical infection (9, 36). C. difficile has also been isolated from animals such as calves, pigs, chickens, wild animals, and pets (9). It was isolated in different amounts from animals with clinical manifestation of diarrhea and also from healthy animals (11, * Author for correspondence. Tel: (386) 1 4779 827; Fax: (386) 1 4779 174; E-mail:
[email protected].
18, 26, 28, 33). Certain strains seemed to be highly species specific, whereas others were found in a variety of species, including humans; thus, the possibility of human-to-animal transmission or zoonotic transmission has been suggested (11, 35). If animals are a potential source of C. difficile, food could be one of the transmission routes from animals to humans (8, 12). C. difficile has been isolated from a variety of foods, mostly of animal origin, but also from ready-to-eat salads (3). The contamination rate of foods by C. difficile in Europe seems to be lower than the rates reported in the United States and Canada (9, 11, 22, 34, 35). Some protocols have been published and used for molecular detection and isolation of C. difficile from different types of samples, including clinical, environmental, and food samples (1, 4, 19, 21). Different media and sample treatments have been reported, although no standardized or worldwide recommended protocol is being used for isolation. The number of spores in food is usually low (37), so this is the main reason why C. difficile needs to be enriched before detection and isolation (31). Different enrichment protocols may have significantly different impacts on the final results, making comparisons misleading or impossible. Differences in reported prevalence may also be due to the use of different methodologies (22). Here, we have compared different enrichment broths and enrichment times and studied the possible impacts of the occurrence of different C. difficile strains on molecular
562
BIASIZZO ET AL.
detection; furthermore, we have investigated the possible impacts of different solid isolation media on the isolation of C. difficile. Because results of food contamination studies with C. difficile from Europe and North America differ significantly (7), we sought to determine whether different enrichment media have an impact on recovery rate. The aim of the study was to implement an optimal protocol for use in detecting C. difficile in food samples.
MATERIALS AND METHODS Bacterial strains and inoculum preparation. The three C. difficile strains used in this study were ATCC 9689 (hereafter strain 1 [S1]), and strains T7-MB (S2) and CD 400 (S3) (Veterinary Faculty, University of Ljubljana, Ljubljana, Slovenia). S2 is ribotype 014/020, toxinotype 0, isolated from calves. It is the most common serotype in humans in Europe (5) and also in humans and animals in Slovenia (16). S3 is ribotype 045 and was isolated from piglets. Each strain was revived on blood agar plates (blood agar base no. 2 [Oxoid, Basingstoke, UK] supplemented with 5% defibrinated ovine blood) and then transferred to cooked meat medium (Oxoid). After 2 days of incubation under anaerobic condition at 378C, the strains were prepared for inoculation. Anaerobic conditions were provided by a generator sachet (GENbag anaer, bioM´erieux, Inc., Marcy-l’Etoile, France) for anaerobic bacterial incubation placed in anaerobic polycarbonate jars (Oxoid). The inoculum for spiking the samples was prepared by diluting 100 lL of each strain in a peptone-salt solution (Natriumchlorid-NaCl, Merck, Darmstadt, Germany) and Proteose Peptone (Biolife, Milan, Italy) to give the final concentration. For prevalidation tests, each C. difficile strain was prepared separately to identify any impact of specificity of strains on detection and isolation. For further validation, a cocktail composed of all three strains was used for inoculation. A cocktail was prepared by mixing 100 lL of each strain kept in cooked meat medium into a peptone-salt solution to give a final concentration of 5 to 20 CFU per sample. In parallel with the sample inoculation, the concentration of C. difficile for each prepared inoculum was also quantified. Inoculum was diluted and transferred onto blood agar plates on which colonies were counted after incubation for 48 h on 378C under anaerobic conditions. Samples and inoculations. For prevalidation tests, minced meat (a mixture of bovine and porcine meat) samples were prepared and divided into sterile plastic bags (10 g) and stored at 708C until tested. The meat was tested for the presence of Enterobacteriaceae in accordance with ISO 21528-2:2004 (720 CFU/g) (14), sulfite-reducing clostridia in accordance with ISO 15213:2003 (,10 CFU/g) (13), aerobic mesophilic bacteria in accordance with ISO 4833-1:2013 (1.7 3 106 CFU/g) (15), and anaerobes in accordance with ISO 4833-1:2013 (25 CFU/g) (15). Plates for anaerobes were incubated in anaerobic polycarbonate jars (Oxoid), with an inserted generator for anaerobic bacterial incubation (GENbag anaer, bioM´erieux, Inc.). An adequate volume of prepared inoculum was added to 10 g of meat samples to give the target concentrations: 500 to 1,000; 50 to 100; and 10 to 50 C. difficile CFU per sample. In the prevalidation tests, 54 inoculated minced meat samples, 8 noninoculated minced meat samples, inoculated enrichment broths without matrix, and enrichment broths were tested. For validation, 44 raw meat samples of different origins were prepared. The types of samples were minced meat (n ¼ 5), spiced
J. Food Prot., Vol. 81, No. 4
minced meat (n ¼ 4), poultry meat (n ¼ 8), spiced or marinated poultry meat (n ¼ 4), and marinated pork meat (n ¼ 1); each sample was prepared in duplicate. One of the duplicate units was inoculated with a cocktail of all three C. difficile strains and one unit was not inoculated. Prevalidation study design. For prevalidation tests, samples were enriched in three different broths for 1, 4, and 7 days. After these times, detection with quantitative real-time PCR (qPCR) and isolation on different agar plates were performed. Before plating on agar plates, different sample treatments and centrifugation were also used. The results were then compared, and an optimal procedure was chosen for further validation. Enrichment. Three enrichment broths were used for the prevalidation study: (i) TCCFBþL: cycloserine-cefoxitin fructose broth (CCFB) supplemented with taurocholic acid (TCCFB), D-cycloserine, cefoxitin, and lysozyme (L). The broth base was composed of the following ingredients: 40 g/L Proteose Peptone (Biolife), 5.0 g/ L disodium hydrogen phosphate (Merck, Kenilworth, NJ), 2.0 g/L sodium chloride (Merck), 0.1 g/L magnesium sulfate (Sigma, St. Louis, MO), 6.0 g/L D-()-fructose (Fisher Scientific, Loughborough, UK), and 1.0 g/L taurocholic acid sodium salt (Merck). After autoclaving and cooling, C. difficile supplement (250.0 mg/L D-cycloserine and 8.0 mg/L cefoxitin; Oxoid) and 0.01 g/L lysozyme (Serva, Heidelberg, Germany) were added to the broth base. (ii) TCDMNþL: C. difficile moxalactam norfloxacin (CDMN) selective supplement (500.0 mg/L cysteine hydrochloride, 12.0 mg/L norfloxacin, and 32.0 mg/L moxalactam; Oxoid) and 0.01 g/L lysozyme (L; Serva). After autoclaving and cooling the broth base (same composition as for TCCFBþL), this selective supplement and lysozyme were added to the broth base. (iii) CCMB-TAL: cycloserine-cefoxitin mannitol broth (CCMB) with taurocholate and lysozyme (TAL; Anaerobe Systems, Morgan Hill, CA). Broth (5 mL) was ready to use and poured into tubes with lids, with the option to create anaerobic conditions. For enrichment in TCCFBþL or in TCDMNþL, 10 g of sample was placed into a plastic bag. Broth (90 mL) was added and homogenized in a peristaltic blender (BegMixer, Interscience, Paris, France) for 1 min. For enrichment in CCMB-TAL, 1 g of the sample was added to a tube, commercially poured with 5 mL of broth, and homogenized on a vortex mixer (Vibromix 10, Domel, Zˇelezniki, Slovenia). Each prepared sample (in a bag or a tube) was enclosed in an anaerobic jar with a generator and incubated at 378C for 1, 4, and 7 days. After incubation, molecular detection of C. difficile was carried out using qPCR and also isolation on agar plates. Molecular detection method. Total DNA was isolated using the High Pure PCR template preparation kit (Roche, Basel, Switzerland), with an additional step of bead beating on a MagNA Lyser instrument (Roche). Enrichment broth (1.5 mL) together with 106-lm-diameter glass beads (Sigma) was added to a microcentrifuge tube that was centrifuged for 5 min at 10,000 3 g. The supernatant was then discarded, and 200 lL of phosphatebuffered saline, 200 lL of tissue lysis buffer, and 15 lL of lysozyme (10 mg/mL) were added to the tube. This step was followed by bead beating on a MagNA Lyser instrument (Roche) at 6,400 rpm for 45 s and incubation at 378C for 15 min. The cycle
563
PROTOCOL FOR DETECTING AND RECOVERING C. DIFFICILE
J. Food Prot., Vol. 81, No. 4
TABLE 1. Results using TCCFBþ L for enrichmenta,b 1 Day of enrichment
7 Days of enrichment
Alcoholic shock Sample
Inoculum
M MþS1
1,000 CFU
MþS2 MþS3
Alcoholic shock
qPCR
BA
CCFA
qPCR
BA
CCFA
þ þ þ þ þ þ
þ þ þ þ þ þ
þþþ þþþ þþþ þþþ þþþ þþþ
þþþ þþþ þþþ þþþ þþþ þþþ
4 Days of enrichment 1 Day of enrichment Sample
Inoculum
TCCFBþL M MþS1
100 CFU
MþS2 MþS3 MþS1 MþS2 MþS3 a
b
10 CFU
7 Days of enrichment
Alcoholic shock
Alcoholic shock
qPCR
qPCR
BA
CCFA
qPCR
BA
CCFA
þ þ þ þ þ þ þ þ þ þ þ þ
þ þ þ þ þ þ þ þ þ þ þ þ
þþþ þþþ þþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ
þþþ þþþ þþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ
þ þ þ þ þ þ þ þ þ þ þ þ
þþþ þþþ þþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ
þþþ þþþ þþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ
BA, blood agar (plate); CCFA, C. difficile agar with supplements; M, meat sample; MþS1, MþS2, MþS3, meat samples inoculated with C. difficile strains S1, S2, and S3, respectively. , no growth; þ, ,10 CFU per plate; þþ, 10 to 100 CFU per plate; þþþ, .100 CFU per plate.
of bead beating and incubation were repeated. This step was followed by an additional bead beating step at 6,400 rpm for 45 s, after which the samples were incubated at 1008C for 5 min and allowed to cool to room temperature. The samples were then further centrifuged at 10,000 3 g for 5 min. The supernatant was transferred to another 1.5-mL microcentrifuge tube, and total DNA was isolated according to the manufacturer’s instructions. If qPCR was not performed the same day, the DNA was stored at 708C. qPCR was performed with the previously described primers and probes (25). Reactions were performed with 2 lL of DNA in a 25-lL reaction volume with FastStart Universal Probe Master (ROX) mix (Roche) as described previously (4). Positive and negative controls were included in each qPCR run. qPCR was performed on an ABI 7500 Fast instrument (Applied Biosystems, Foster City, CA). Isolation on solid media. After an enrichment step for 1, 4, and 7 days, directly from the enrichment broth or after treatment, samples were inoculated onto two types of solid agar plates: blood agar and C. difficile agar base with added C. difficile selective supplement and 7% (v/v) defibrinated horse blood (CCFA; all from Oxoid).
For treatment, 1 mL of enriched sample and 1 mL of 96% ethanol were added to a centrifuge tube and mixed by inverting (10 times); the tube was then left at room temperature for 30 min. After 10-min centrifugation at 4,000 3 g, the supernatant was removed, and a loopful (10 lL) of the sediment was streaked onto the two types of solid agar plates. Plates were then incubated for 48 h at 378C under anaerobic conditions. Colonies were identified on the basis of morphological criteria (grayish, nonhemolytic, swarming colonies) and typical odor. Validation protocol. Enrichment in TCCFBþL and anaerobic incubation for 1 day were selected for the molecular method, whereas enrichment in TCCFBþL and anaerobic incubation for 4 and 7 days followed by alcoholic shock were selected for the isolation method.
RESULTS AND DISCUSSION Minced meat was chosen as the appropriate matrix for the validation protocol because it is already known that background microbes can affect the effectiveness of examinations. The number of C. difficile spores in food is insufficient to enable detection without enrichment (7, 38);
564
BIASIZZO ET AL.
J. Food Prot., Vol. 81, No. 4
TABLE 2. Results using CCMB-TAL for enrichmenta,b 1 Day of enrichment
7 Days of enrichment
Alcoholic shock Sample
Inoculum
M MþS1
1,000 CFU
MþS2 MþS3
Alcoholic shock
qPCR
BA
CCFA
qPCR
BA
CCFA
þ þ þ þ þ þ
þ þ þ þ
þ þ þ þ
þ þ þ þ þ þ
þþþ þþþ þþþ þþþ þþþ þþþ
þþþ þþþ þþþ þþþ þþþ þþþ
4 Days of enrichment 1 Day of enrichment Sample
Inoculum
M MþS1
100 CFU
MþS2 MþS3 MþS1 MþS2 MþS3
10 CFU
7 Days of enrichment
Alcoholic shock
Alcoholic shock
qPCR
qPCR
BA
CCFA
qPCR
BA
CCFA
þ þ þ þ þ þ þ þ þ þ þ þ
þ þ þ þ þ þ þ þ þ þ þ þ
þþþ þþ þ þþ þþþ
þ þþþ þþþ þ þ þ þþþ þþþ
þ þ þ þ þ þ þ þ þ þ þ þ
þþ þþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ
þþ þþ þþþ þþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ
a
BA, blood agar (plate); CCFA, C. difficile agar with supplements; M, meat sample; MþS1, MþS2, MþS3, meat samples inoculated with C. difficile strains S1, S2, and S3, respectively. b , no growth; þ, ,10 CFU per plate; þþ, 10 to 100 CFU per plate; þþþ, .100 CFU per plate.
therefore, enrichment is necessary for successful detection. Different enrichment broths have been studied and include the following: CCFB (21), CCFB supplemented with 0.1% sodium taurocholate (1, 2), CCMB-TAL (7), CDMN (17, 29), CDMN supplemented with sodium taurocholate (35, 38), and brain heart infusion supplemented with yeast extract, cysteine, and taurocholate (34) or with cefoxitin, cycloserine, and taurocholate (6). We chose three basic enrichment broths described in the literature: TCCFB, TCDMN, and CCMB-TAL. However, the recovery of C. difficile is reported to be higher with the addition of sodium taurocholate and lysozyme to the enrichment broth (23, 39), so they both were added into CCFB and CDMN, whereas they are also commercially present in CCMB-TAL.
Prevalidation. When TCCFBþL was used as the enrichment broth, detection of C. difficile at concentrations of 1,000, 100, and 10 CFU per sample with the molecular qPCR method after 1, 4, and 7 days was positive for all spiked samples and negative for all negative controls (Table 1). However, the isolation of C. difficile from spiked samples was not successful after 1 day of incubation at a
concentration of 1,000 CFU per sample, but after 7 days, all spiked samples were positive and negative controls were negative. Thus, samples were inoculated with 100 and 10 CFU per sample and further analyzed only after 4 and 7 days of incubation (Table 1). When CCMB-TAL was used as the enrichment broth, molecular detection of C. difficile was positive in all spiked samples with a concentration of 1,000 CFU per sample after 1 and 7 days and also for spiked samples with concentrations of 100 and 10 CFU per sample after 1, 4, and 7 days (Table 2). After 1 day of enrichment, the isolation rate was better than in TCCFBþL. However, when using CCMB-TAL, C. difficile could not be isolated from two of six samples spiked with 1,000 CFU per sample. The incubation time was therefore prolonged for the samples spiked with lower concentrations (100 and 10 CFU per sample). Nevertheless, the isolation was negative for 4 of 12 spiked samples after 4 days, whereas C. difficile was isolated from all spiked samples after 7 days (Table 2). When TCDMNþL was used as the enrichment broth, molecular detection of C. difficile with qPCR was positive for all spiked samples and negative for negative controls at
PROTOCOL FOR DETECTING AND RECOVERING C. DIFFICILE
J. Food Prot., Vol. 81, No. 4
565
TABLE 3. Results using TCDMNþ L for enrichmenta,b 4 Days of enrichment 1 Day of enrichment Sample
Inoculum
M MþS1
1,000 CFU
MþS2 MþS3 MþS1
100 CFU
MþS2 MþS3 MþS1 MþS2 MþS3 TCDMNþL
10 CFU
7 Days of enrichment
Alcoholic shock
Alcoholic shock
qPCR
qPCR
BA
CCFA
qPCR
BA
CCFA
þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ
þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ
þ þþ þ þþ þ þ þþ
þ þþ þ þþþ þ þ þ þ þþ
þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ
þþ þ þ þþþ þ þ þþþ þ þþ þþ þþ þþþ
þþ þ þþ þþþ þ þ þþþ þ þþ þþ þþ þþþ
a
BA, blood agar (plate); CCFA, C. difficile agar with supplements; M, meat sample; MþS1, MþS2, MþS3, meat samples inoculated with C. difficile strains S1, S2, and S3, respectively. b , no growth; þ, ,10 CFU per plate; þþ, 10 to 100 CFU per plate; þþþ, .100 CFU per plate.
all concentrations (1,000, 100, and 10 CFU per sample) after 1, 4, and 7 days of enrichment (Table 3). Isolation was successful for 9 (50%) of the 18 spiked samples after 4 days of enrichment and for 12 (67%) of the 18 spiked samples after 7 days of enrichment (Table 3). The results obtained with the isolation method show that TCCFBþL was the most efficient enrichment broth regarding recovery and multiplication. Isolation was 100% after 4 and 7 days of enrichment, whereas when using CCMB-TAL, isolation was 67 and 100%, and when using TCDMNþL, isolation was 50 and 67%, respectively. Lister et al. (21) compared CCFB and CCMB-TAL for use in stool samples. They reported that growth of C. difficile had been detected a day later if samples had been enriched in CCMBTAL rather than in CCFB. However, evaluation of recovery after that prolonged time was better in CCMB-TAL than in CCFB. In our study with food samples, enrichment for 4 days showed better recovery when using TCCFBþL compared with CCMB-TAL, whereas after prolonged enrichment to 7 days, isolation was 100% and growth was evaluated as very good (i.e., the number of C. difficile colonies on the plates inoculated after enrichment was .100 CFU per plate) in both enrichment broths. We added taurocholate and lysozyme to CCFB, whereas commercially they are added to CCMB-TAL. Adding those components to CCFB could be one of the reasons why recovery and growth performance were better than those measures reported by Lister et al. (21). Our findings show that there is no evident
difference in the effectiveness of broths if taurocholate and lysozyme are added to CCFB. TCDMN and TCCFB (comparable to TCDMNþL and TCCFBþL in our study, but without lysozyme) were compared by RodriguezPalacios et al. (27). They found that recovery varied with the method used and reported better results for TCDMN than for TCCFB. Their study was very limited, referring only to four positive samples. We obtained the opposite results: when both broths were used with added lysozyme, the recovery and growth performance were better in TCCFBþL. Other method variables have been reported that may also have an impact on recovery rate and reported prevalence, and they include quantity of sample, enrichment time, and treatment of the samples before plating (8, 17, 28, 31, 38). Centrifugation and treatment to reduce other contaminants seem to be necessary for successful isolation of C. difficile. Results from the isolation of spiked samples without any treatment after enrichment were poor (data not shown). We also tried to treat the samples with heat shock (808C for 10 min), but without subsequent centrifugation. The results of the isolation were comparable to those of untreated samples (data not shown). To evaluate heat shock relative to alcoholic shock, the centrifugation should be performed for both procedures, with this being the reason why our results for that phase could not be compared. It seems that the multiplication of C. difficile in enrichment broth is not sufficiently intensive, so that isolation would not
566
BIASIZZO ET AL.
J. Food Prot., Vol. 81, No. 4
TABLE 4. Validation results of testing unspiked and spiked samples: qPCR after 1 day and isolation after 4 and 7 days of enrichment in TCCFBþ La,b Nonspiked sample
Spiked sample, mixture of strains (5–20 CFU) added
Alcoholic shock 1 Day No.c
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
4 Days
Alcoholic shock
7 Days
1 Day
4 Days
7 Days
Matrix
qPCR
BA
CCFA
BA
CCFA
qPCR
BA
CCFA
BA
CCFA
Spiced minced meat 1 Spiced minced meat 2 Broiler meat Spiced minced meat 3 Pork neck marinated Broiler meat 1 Chicken wok 1 Broiler meat 2 Broiler meat 3 Chicken wok 2 Mixed minced meat Spiced minced meat 4 Spiced chicken wings Broiler meat 4 Broiler file Minced meat 1 Broiler legs Minced meat 2 Minced meat 3 Chicken skewers Broiler filed legs Minced meat 4 TCCFBþL TCCFBþL TCCFBþL
NT NT NT NT NT NT NT NT NT NT NT NT NT
NT NT NT NT NT NT NT NT NT NT NT NT NT
þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ
þþ þþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ NT NT NT NT NT NT NT NT NT NT NT NT þþþ þþþ NT
þþ þþ þþþ þ þþþ þþ þþþ þþþ þþþ þþþ NT NT NT NT NT NT NT NT NT NT NT NT þþþ þþþ NT
þþ þþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þ þ þþ þ þ þþ þ þ þþ þþþ þþþ þ
þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþ þþ þþþ þ þ þþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþ
a
BA, blood agar (plate); CCFA, C. difficile agar with supplements. , no growth; þ, ,10 CFU per plate; þþ, 10 to 100 CFU per plate; þþþ, .100 CFU per plate; NT, not tested. c Numbers represent a different lot of the product. b
be successful without concentration by centrifugation. In our experiments, the isolation rate after treatment using alcoholic shock and subsequent centrifugation was highest. Alcoholic shock also eliminates most of the background microbiota in raw meat samples and facilitates effective isolation. When samples were spiked and transferred onto nonselective blood agar plates, the background microbiota still made isolation difficult. However, CCFA plates are selective and allow the growth of only a few anaerobes. Determination of atypical bacteria growing on CCFA was performed by a matrix-assisted laser desorption–ionization time of flight instrument (Bruker Daltonics, Bremen, Germany); the presence of other clostridia (Clostridium butyricum and Clostridium bifermentans) and Hafnia alvei was confirmed. We also observed differences in the recovery of different C. difficile strains (S1, S2, and S3). The best isolation performance was observed for S3: isolation from all the enrichment broths was complete after 4 and 7 days. After 4 days of enrichment in CCMB-TAL, S1 and S2 were not isolated from two of six spiked samples with a particular strain, at different concentration levels. When TCDMNþL was used, the largest differences were observed between
strains. After 7 days of enrichment, C. difficile was isolated from all spiked samples with S3, from four of six with S1, and from only two of six with S2.
Validation. For the validation, we prepared meat of various origins and meat preparations to determine whether the spices and other additives can influence the performance of detection or isolation methods (Table 4). Regarding the results of prevalidation tests, enrichment into TCCFBþL for 1 day was selected for the molecular method, whereas 4 and 7 days followed by isolation on CCFA plates after treatment with alcoholic shock and centrifugation was selected for the isolation method. Detection by qPCR was positive for all spiked samples and negative for nonspiked samples. Isolation after 4 days of enrichment gave the expected results. However, for 1 of 10 tested samples, growth on CCFA plates was poor (10 CFU per plate), so enrichment was prolonged to 7 days for all samples. The growth of C. difficile on the plates after 7 days of enrichment was very good. From the other 12 spiked samples, C. difficile was isolated from 11 samples onto CCFA and from 9 samples onto blood agar plates. In
PROTOCOL FOR DETECTING AND RECOVERING C. DIFFICILE
J. Food Prot., Vol. 81, No. 4
TABLE 5. Capability of the isolation method defined with the validation (sensitivity, specificity, and positive and negative deviation) Result
Positive Negative Together Sensitivity Specificity Positive deviation Negative deviation
Positive
Negative
Together
24 0 24
1 25 26
25 25 50
0.96 1.00 0.00 0.04
general, growth on CCFA plates was better than on blood agar plates because of the amount of background microbiota, in particular, other anaerobes, which were not eliminated by alcoholic shock. Altogether, the sensitivity and specificity of qPCR were 100%. For the isolation procedure using CCFA and blood agar as isolation plates, the sensitivity was 95% (21 of 22) on CCFA and 86% (19 of 22) on blood agar plates (Table 5). We conclude that 4 days of enrichment is the minimum enrichment period for adequate isolation of C. difficile from raw meat samples, but 7 days is even better. We established that sample enrichment in TCCFBþL at 378C for 4 days for qPCR detection and for 7 days for the isolation are the most appropriate protocols. For the isolation, treatment of samples using alcohol, centrifugation, inoculation onto CCFA plates, and incubation at 378C under anaerobic conditions for 48 h is the most efficient. It is reasonable to consider that our protocol could also be appropriate for testing raw meat samples in the presence of spices and marinades. Validation of the method was performed using inoculated samples, so additional study is required to evaluate its efficiency when testing naturally contaminated samples.
ACKNOWLEDGMENT This work was supported by the Slovenian Ministry for Higher Education, Science and Technology (research program P4-0092).
REFERENCES 1. Arroyo, L. G., J. Rousseau, B. M. Willey, D. E. Low, H. Staempfli, A. McGeer, and J. S. Weese. 2005. Use of a selective enrichment broth to recover Clostridium difficile from stool swabs stored under different conditions. J. Clin. Microbiol. 43:5341–5343. 2. Avberˇsek, J., U. Zajc, J. Mi´cunovi´c, and M. Ocepek. 2013. Improved detection of Clostridium difficile in animals by using enrichment culture followed by LightCycler real-time PCR. Vet. Microbiol. 164:93–100. 3. Bakri, M. M., D. J. Brown, J. P. Butcher, and A. D. Sutherland. 2009. Clostridium difficile in ready-to-eat salads, Scotland. Emerg. Infect. Dis. 15:817–818. 4. Bandelj, P., K. Logar, A. M. Usenik, M. Vengust, and M. Ocepek. 2013. An improved qPCR protocol for rapid detection and quantification of Clostridium difficile in cattle feces. FEMS Microbiol. Lett. 341:115–121. 5. Bauer, M. P., D. W. Notermans, B. H. van Benthem, J. S. Brazier, M. H. Wilcox, M. Rupnik, D. L. Monnet, J. T. van Dissel, E. J. Kuijper, and E. S.Group. 2011. Clostridium difficile infection in Europe: a hospital-based survey. Lancet 377:63–73.
567
6. Bouttier, S., M. C. Barc, B. Felix, S. Lambert, A. Collignon, and F. Barbut. 2010. Clostridium difficile in ground meat, France. Emerg. Infect. Dis. 16:733–735. 7. Curry, S. R., J. W. Marsh, J. L. Schlackman, and L. H. Harrison. 2012. Prevalence of Clostridium difficile in uncooked ground meat products from Pittsburgh, Pennsylvania. Appl. Environ. Microbiol. 78:4183–4186. 8. de Boer, E., A. Zwartkruis-Nahuis, A. E. Heuvelink, C. Harmanus, and E. J. Kuijper. 2011. Prevalence of Clostridium difficile in retailed meat in the Netherlands. Int. J. Food Microbiol. 144:561–564. 9. Freeman, J., M. P. Bauer, S. D. Baines, J. Corver, W. N. Fawley, B. Goorhuis, E. J. Kuijper, and M. H. Wilcox. 2010. The changing epidemiology of Clostridium difficile infections. Clin. Microbiol. Rev. 23:529–549. 10. Furuya-Kanamori, L., J. Marquess, L. Yakob, T. V. Riley, D. L. Paterson, N. F. Foster, C. A. Huber, and A. C. Clements. 2015. Asymptomatic Clostridium difficile colonization: epidemiology and clinical implications. BMC Infect. Dis. 15:516. 11. Gould, L. H., and B. Limbago. 2010. Clostridium difficile in food and domestic animals: a new foodborne pathogen? Clin. Infect. Dis. 51:577–582. 12. Hensgens, M. P., E. C. Keessen, M. M. Squire, T. V. Riley, M. G. Koene, E. de Boer, L. J. Lipman, E. J. Kuijper; European Society of Clinical Microbiology and Infectious Diseases Study Group for Clostridium difficile (ESGCD). 2012. Clostridium difficile infection in the community: a zoonotic disease? Clin. Microbiol. Infect. 18:635– 645. 13. International Organization for Standardization. 2003. Microbiology of food and animal feeding stuffs—horizontal method for the enumeration of sulfite-reducing bacteria growing under anaerobic conditions. ISO 15213:2003. International Organization for Standardization, Geneva. 14. International Organization for Standardization. 2004. Microbiology of food and animal feeding stuffs—horizontal methods for the detection and enumeration of Enterobacteriaceae—part 2: Colony-count method. ISO 21528-2:2004. International Organization for Standardization, Geneva. 15. International Organization for Standardization. 2013. Microbiology of the food chain—horizontal method for the enumeration of microorganisms—part 1: Colony count at 308C by the pour plate technique. ISO 4833-1:2013. International Organization for Standardization, Geneva. 16. Janezic, S., M. Ocepek, V. Zidaric, and M. Rupnik. 2012. Clostridium difficile genotypes other than ribotype 078 that are prevalent among human, animal and environmental isolates. BMC Microbiol. 12:48. 17. J¨obstl, M., S. Heuberger, A. Indra, R. Nepf, J. K¨ofer, and M. Wagner. 2010. Clostridium difficile in raw products of animal origin. Int. J. Food Microbiol. 138:172–175. 18. Keel, K., J. S. Brazier, K. W. Post, S. Weese, and J. G. Songer. 2007. Prevalence of PCR ribotypes among Clostridium difficile isolates from pigs, calves, and other species. J. Clin. Microbiol. 45:1963– 1964. 19. Kilic, A., M. J. Alam, N. L. Tisdel, D. N. Shah, M. Yapar, T. M. Lasco, and K. W. Garey. 2015. Multiplex real-time PCR method for simultaneous identification and toxigenic type characterization of Clostridium difficile from stool samples. Ann. Lab. Med. 35:306–313. 20. Lessa, F. C. 2013. Community-associated Clostridium difficile infection: how real is it? Anaerobe 24:121–123. 21. Lister, M., E. Stevenson, D. Heeg, N. P. Minton, and S. A. Kuehne. 2014. Comparison of culture based methods for the isolation of Clostridium difficile from stool samples in a research setting. Anaerobe 28:226–229. 22. Lund, B. M., and M. W. Peck. 2015. A possible route for foodborne transmission of Clostridium difficile? Foodborne Pathog. Dis. 12:177–182. 23. Nerandzic, M. M., and C. J. Donskey. 2009. Effective and reducedcost modified selective medium for isolation of Clostridium difficile. J. Clin. Microbiol. 47:397–400. 24. Ozaki, E., H. Kato, H. Kita, T. Karasawa, T. Maegawa, Y. Koino, K. Matsumoto, T. Takada, K. Nomoto, R. Tanaka, and S. Nakamura.
568
25.
26.
27.
28.
29.
30.
BIASIZZO ET AL.
2004. Clostridium difficile colonization in healthy adults: transient colonization and correlation with enterococcal colonization. J. Med. Microbiol. 53:167–172. Penders, J., C. Vink, C. Driessen, N. London, C. Thijs, and E. E. Stobberingh. 2005. Quantification of Bifidobacterium spp., Escherichia coli and Clostridium difficile in faecal samples of breast-fed and formula-fed infants by real-time PCR. FEMS Microbiol. Lett. 243:141–147. Pirs, T., M. Ocepek, and M. Rupnik. 2008. Isolation of Clostridium difficile from food animals in Slovenia. J. Med. Microbiol. 57:790– 792. Rodriguez-Palacios, A., R. J. Reid-Smith, H. R. Staempfli, D. Daignault, N. Janecko, B. P. Avery, H. Martin, A. D. Thomspon, L. C. McDonald, B. Limbago, and J. S. Weese. 2009. Possible seasonality of Clostridium difficile in retail meat, Canada. Emerg. Infect. Dis. 15:802–805. Rodriguez-Palacios, A., H. R. St¨ampfli, T. Duffield, A. S. Peregrine, L. A. Trotz-Williams, L. G. Arroyo, J. S. Brazier, and J. S. Weese. 2006. Clostridium difficile PCR ribotypes in calves, Canada. Emerg. Infect. Dis. 12:1730–1736. Rodriguez-Palacios, A., H. R. Staempfli, T. Duffield, and J. S. Weese. 2007. Clostridium difficile in retail ground meat, Canada. Emerg. Infect. Dis. 13:485–487. Rupnik, M. 2007. Is Clostridium difficile-associated infection a potentially zoonotic and foodborne disease? Clin. Microbiol. Infect. 13:457–459.
J. Food Prot., Vol. 81, No. 4
31. Rupnik, M., and J. G. Songer. 2010. Clostridium difficile: its potential as a source of foodborne disease. Adv. Food Nutr. Res. 60:53–66. 32. Rupnik, M., M. H. Wilcox, and D. N. Gerding. 2009. Clostridium difficile infection: new developments in epidemiology and pathogenesis. Nat. Rev. Microbiol. 7:526–536. 33. Songer, J. G., and M. A. Anderson. 2006. Clostridium difficile: an important pathogen of food animals. Anaerobe 12:1–4. 34. Songer, J. G., H. T. Trinh, G. E. Killgore, A. D. Thompson, L. C. McDonald, and B. M. Limbago. 2009. Clostridium difficile in retail meat products, USA, 2007. Emerg. Infect. Dis. 15:819–821. 35. Varshney, J. B., K. J. Very, J. L. Williams, J. P. Hegarty, D. B. Stewart, J. Lumadue, K. Venkitanarayanan, and B. M. Jayarao. 2014. Characterization of Clostridium difficile isolates from human fecal samples and retail meat from Pennsylvania. Foodborne Pathog. Dis. 11:822–829. 36. Weese, J. S. 2010. Clostridium difficile in food–innocent bystander or serious threat? Clin. Microbiol. Infect. 16:3–10. 37. Weese, J. S., B. P. Avery, J. Rousseau, and R. J. Reid-Smith. 2009. Detection and enumeration of Clostridium difficile spores in retail beef and pork. Appl. Environ. Microbiol. 75:5009–5011. 38. Weese, J. S., R. J. Reid-Smith, B. P. Avery, and J. Rousseau. 2010. Detection and characterization of Clostridium difficile in retail chicken. Lett. Appl. Microbiol. 50:362–365. 39. Wilcox, M. H., W. N. Fawley, and P. Parnell. 2000. Value of lysozyme agar incorporation and alkaline thioglycollate exposure for the environmental recovery of Clostridium difficile. J. Hosp. Infect. 44:65–69.
