Evaluation of Potassium-ClavulanateSupplemented Modified Charcoal-CefoperazoneDeoxycholate Agar for Enumeration of

Campylobacter in Chicken Carcass Rinse
Jung-Whan Chon,∗ Hong-Seok Kim,∗ Hyunsook Kim, Deog-Hwan Oh, and Kun-Ho Seo

Potassium-clavulanate-supplemented modified charcoal-cefoperazone-deoxycholate agar (C-mCCDA) that
was described in our previous study was compared with original mCCDA for the enumeration of Campylobacter in pure
culture and chicken carcass rinse. The quantitative detection of viable Campylobacter cells from a pure culture, plated on
C-mCCDA, is statistically similar (P > 0.05) to mCCDA. In total, 120 chickens were rinsed using 400 mL buffered
peptone water. The rinses were inoculated onto C-mCCDA and mCCDA followed by incubation at 42 °C for 48 h.
There was no statistical difference between C-mCCDA (45 of 120 plates; mean count, 145.5 CFU/mL) and normal
mCCDA (46 of 120 plates; mean count, 160.8 CFU/mL) in the isolation rate and recovery of Campylobacter (P > 0.05)
from chicken carcass rinse. The Pearson correlation coefficient value for the number of Campylobacter cells recovered in
the 2 media was 0.942. However, the selectivity was much better on C-mCCDA than on mCCDA plates (P < 0.05).
Significantly fewer C-mCCDA plates (33 out of 120 plates; mean count, 1.9 CFU/mL) were contaminated with nonCampylobacter cells than the normal mCCDA plates (67 out of 120 plates; mean count, 27.1 CFU/mL). The C-mCCDA
may provide improved results for enumeration of Campylobacter in chicken meat alternative to mCCDA with its increased
selectivity the modified agar possess.

Keywords: Campylobacter, chicken carcass rinse, C-mCCDA, contaminants, enumeration

Introduction
Campylobacter spp. have been regarded as a major source of
gastrointestinal infection worldwide (Park and Sanders 1992;
Nachamkin and others 1998; Josefsen and others 2003). Campylobacter are a natural part of the intestinal flora in poultry (Josefsen
and others 2003; Belanger and Shryock 2007), and the number of
these bacteria in a carcass rinse can vary and, as a result, can affect
the number of these bacteria in a carcass rinse (Line 2001a; Line
and Berrang 2005; Oyarzabal and others 2005). Enumeration of
Campylobacter on chicken carcasses by using sensitive and quantitative detection methods has been a major goal for many food
quality control authorities (Josefsen and others 2003).
Although there are various quantitative methods to detect
Campylobacter that do not involve culturing of the bacteria, direct
plating onto selective agars is still regarded as an effective technique for isolating and enumerating Campylobacter from a carcass
rinse, as shown in previous studies (Line 2001; Line and Berrang
2005; Oyarzabal and others 2005). However, the differentiation of
suspected Campylobacter colonies from other contaminating species
by using selective agars are often confounded by an overgrowth
of non-Campylobacter colonies (Line 2001; Line and others 2001;
Line and Berrang 2005; Line 2006; Jasson and others 2009; Chon
and others 2012a).
MS 20131378 Submitted 9/27/2013, Accepted 12/23/2013. Authors Chon, HS Kim, H Kim, and Seo are with KU Center for Food Safety, College of Veterinary
Medicine, Konkuk Univ., Seoul, The Republic of Korea. Author Oh is with Dept.
of Food Science and Biotechnology, Inst. of Bioscience and Biotechnology, Kangwon
Natl. Univ., Chuncheon, Kangwon 200-701, Republic of Korea. Direct inquiries to
author Seo (E-mail: bracstu3@konkuk.ac.kr).
∗ Both

authors contributed equally to this work.

R 

C 2014 Institute of Food Technologists

doi: 10.1111/1750-3841.12388
Further reproduction without permission is prohibited

Most selective agars commonly used in food quality control are supplemented by a high concentration of cefoperazone, a 3rd-generation cephalosporin (Corry and others 1995).
However, resistance to cefoperazone has recently become more
widespread in food-related bacteria, causing a failure of isolation of
Campylobacter spp. in raw poultry meat (Jasson and others 2009;
Moran and others 2011; Chon and others 2012a). In particular, extended spectrum beta-lactamase (ESBL)-producing Escherichia coli may grow on selective agar because of its resistance
to cephalosporins. The ESBL-producing bacteria have frequently
been isolated in raw chicken in many countries, and have become
a common factor affecting isolation and precise enumeration of
Campylobacter in chicken meat (Warren and others 2008; Costa and
others 2010; Leverstein-van Hall and others 2011; Moran and others 2011). Therefore, eliminating the growth of ESBL-producing
bacteria on selective media increases both the sensitivity and selectivity of Campylobacter-selective agar.
Modified charcoal-cefoperazone-deoxycholate agar (mCCDA)
has frequently been used by many microbiologists for detection
and enumeration of Campylobacter in clinical and food samples
(Corry and others 1995; Engberg and others 2000; Oyarzabal
and others 2005; Moran and others 2011). Rosenquist and others
(2007) reported that the direct plating on mCCDA is an acceptable
method for quantitative detection of Campylobacter in chicken meat
(Habib and others 2011). mCCDA is formally used in Intl. Organization for Standardization (ISO) for enumeration of Campylobacter
in foods (ISO 2006; Ahmed and others 2012). However, mCCDA
is prone to contamination by ESBL-producing bacteria, thereby
rendering differentiation of suspected Campylobacter colonies difficult (Moran and others 2011; Chon and others 2012a, 2012b). In
a previous study, we described an improved mCCDA through the

Vol. 79, Nr. 5, 2014 r Journal of Food Science M923

M: Food Microbiology
& Safety

Abstract:

The plates were compared using ature (4 °C) and used within 5 d.8 79. Populations of Campylobacter recovered from a milliliter of the chicken carcass washes are shown in Figure 1. The Pearson correlation coefficient was also used to compare potassium clavulanate (Sigma.05) isolation rate and selectivity than did the original mCCDA (Chon and others 2013). St. The number of C-mCCDA and mCCDA plates positive for Campylobacter was similar. Campylobacter strains kept frozen at −70 °C were inoculated on blood agar (Oxoid. M: Food Microbiology & Safety Biochemical confirmation of contaminants on plates To identify the competing flora frequently appeared on the 2 selective agars. version 19.5%]. 6 strains of Campylobacter (3 Campylobacter jejuni strains: NCTC 11168. In the 6th trial. A single colony each of the 3 C.4A 68. U. Suspected Campylobacter colonies (up to 5 colonies from each plate) were subcultured onto blood agar and finally confirmed using colony PCR with reference to Denis and others (1999). Plates were incubated at 42 °C for 48 h under microaerobic condition.3 ± 19. enumeration of Campylobacter by using C-mCCDA was not evaluated in that study. 46 of 120 [38.7 ± 19. We cooled mCCDA.compared the number of plates positive for Campylobacter and non-Campylobacter contaminants. 10% CO2 . and DD_2 [chicken isolate]) were used in the current study. jejuni C. .7 ± 12. This result indicates that the quantitative detection of viable Campylobacter cells from a pure culture. coli strains described above were inoculated into 20 mL of Bolton broth (Bolton broth) and incubated at 37 °C for 48 h under microaerobic condition.A. Louis.5 CFU/mL) than mCCDA (160.non-Campylobacter contaminants from pure culture and chicken turer’s instructions. coli strains: SD_4 [chicken isolate]. There was no statistically significant difference (P > 0. The ESBL enzyme production was also tested using the Vitek 2 AST n224 kit (bioM´erieux) according to manufacturer’s recommendations. 79.4 115. However. Materials and Methods Table 1–The mean number of cells from the 6 Campylobacter strains. an ESBL inhibitor. Hampshire. with 20 chicken carcasses. coli Mean ± SEMa Inoculum (CFU/0. As determined by direct plating. France). U.0 59. U. Chicken carcasses were rinsed with 400 mL of buffered peptone water (Difco. Sparks. starting from pure cultures. Md. Table 3). In this study. 14 subcultured colonies (4 contaminant colonies from 4 C-mCCDA plates and 10 contaminant colonies from 10 mCCDA plates) were biochemically identified using the Vitek 2 GN kit (bioM´erieux. although C-mCCDA exhibited slightly lower mean counts (145. After shaking.05). and qualitatively evaluated the detection of Campylobacter from chicken carcass rinse on the modified agar (potassium-clavulanatesupplemented mCCDA. of recovered cells (CFU/0.mCCDA Numbers of recovered cells of Campylobacter on the 2 selective agars from pure culture are compared in Table 1.1 mL) was inoculated onto the 2 selective agars in duplicates.3%].2A a Statistically significant differences among values in rows are indicated by differing uppercase letters (P < 0.Enumeration of Campylobacter with C-mCCDA .0 ± 25.2 64.1 mL of the wash was 10-fold serially diluted in saline water. Nr. In total. each of the 3 C.1 mL of each of these dilutions of pure culture with a cell count ranging from 30 to 300 were inoculated onto mCCDA and C-mCCDA. With reference to our previous study (Chon carcass rinse on the 2 selective agars were compared using a tand others 2013). The most prominent competing microflora on both mCCDA and CmCCDA from chicken carcasses were also identified. The incubated Bolton broth was 10-fold serially diluted in saline water. All prepared plates were stored at a low temper.S.) to 1 L of the accuracy of enumeration by using the 2 selective agars. . and each dilution (0. All Campylobacter strains were from our own collection. Isolation of Campylobacter from chicken carcass rinse In total. and 3 C. South Korea. no statistical difference was found between the 2 agars (P > 0. Mo. The Pearson correlation coefficient value for the number of recovered Campylobacter cells between the 2 media was 0. Differences in the number of recovered Campylobacter cells and Preparation of mCCDA and C-mCCDA The mCCDA (Oxoid) was prepared according to the manufac.5 mg of test. recovered on the 2 selective agars. we subcultured a single contaminant that appeared on each selective agar.. and 0. Data analysis Statistical analyses were conducted using SPSS. with no statistical difference (C-mCCDA..0 ± 32.K.6 69.1 mL) on plate Species C.) by gentle shaking up to 1 min.7 61. 5.0 ± 13. plated on C-mCCDA.05.5 mg/L of potassium clavulanate. we biochemically confirmed contaminants from each of mCCDA and C-mCCDA in the 6th trial. we quantitatively evaluated the detection of Campylobacter in C-mCCDA and compared it with detection in the original mCCDA. and 85% N2 ) at 37 °C for 48 h for 2 passages. The modification of mCCDA resulted in greater (P < 0.). Marcy l’Etoile.8 CFU/mL. Fisher’s exact test in pairs.1 114. addition of 0. 45 of 120 [37.1 mL) C-mCCDA mCCDA 113. jejuni and the 3 C.00.05) to mCCDA. coli strains. CJ_Yim2 [human isolate]. Comparison of performance of the 2 selective agars Data on Campylobacter and non-Campylobacter cells recovered from the 2 selective media from chicken carcass rinses are presented in Table 2 and 3.2 ± 22. C-mCCDA was made by adding 0.05). C-mCCDA) (Chon and others 2013). and GD_4 [chicken isolate]. followed by incubation microaerobically (5% O2 . 2014 Results and Discussion Recoverability of Campylobacter from pure culture using mCCDA and C. Table 2).3 ± 44. Nr. jejuni and the 3 C. 0. Campylobacter from pure culture recovered using mCCDA and C-mCCDA We compared the number of colonies recovered from each of the 2 agars. The Campylobacter used in this study were obtained from pure culture and from chicken carcass rinses. All plates in duplicates were incubated at 37 °C for 48 h microaerobically and colonies are enumerated. CCHS12-2 [chicken isolate]. Table 1) in the number of Campylobacter cells between the 2 selective agars. . mCCDA. M924 Journal of Food Science r Vol. Bacterial strains In total.3 ± 11. is statistically similar (P > 0. 120 chickens were purchased between May 2012 and May 2013 from 5 different retail stores located in Seoul.S.A.942.

(4 of 4. Rosenquist and others 2007). which has frequently been used for the enumeration of Campylobacter (Line and others 2001. . the dense for isolation of the pathogen (Jasson and others 2009. It appears that ESBL-producing E. Many researchers reported that ESBL-producing E. Moran and others (2011) added potassium clavulanate to the Bolton broth. Ahmed and others 2012. Nr. . However. Chon and others 2012a. we identified 14 contaminants from the 2 selective agars.05) on C-mCCDA than on mCCDA as judged by the fewer number of competing microflora. rendering differentiation and enumeration difficult (Jasson and others 2009. Table 3–Comparison of the number of Campylobacter and nonCampylobacter cells recovered on C-mCCDA and mCCDA from 120 chicken carcass rinses. Vol.05). All E. and they reported improved Campylobacter isolation rates.05). Chon and others 2012a. A mean of 1. In the 6th trial (n = 20). coli were effectively eliminated on C-mCCDA owing to the presence of potassium clavulanate. Moran and others 2011. 79.9 ± 0. compared to a mean of 27. Figure 1–Campylobacter recovered from carcass rinse samples using the 2 selective agars C-mCCDA and mCCDA represented as a correlation coefficient. 2014 r Journal of Food Science M925 M: Food Microbiology & Safety Table 2–Comparison of the number of plates of C-mCCDA and growth of non-Campylobacter on the cefoperazone-containing mCCDA positive for Campylobacter and non-Campylobacter strains agars has been one of the most common confounding factors from 120 chicken carcass rinses. the selectivity was much better (P < 0. coli (8 of 10. We have also reported that potassiumclavulanate-containing mCCDA is more suitable than normal mCCDA for isolation of Campylobacter from chicken carcass rinse (Chon and others 2013). Significantly fewer (P < 0.3) 67/120B (55.1 CFU/mL on mCCDA (P < 0. coli strains were positive for the ESBL enzyme. coli may grow profusely on agars. 5. Moran and others 2011. However.05). Such contamination can mask suspicious Campylobacter and lead to false positives or negatives.1B 0 to 670 a Statistically significant differences among values in rows are indicated by differing uppercase letters (P < 0. All tested colonies isolated from CmCCDA (4 colonies) were identified as Acinetobacter spp. to our knowledge. we evaluated the quantitative detection of Campylobacter from chicken carcass rinse in C-mCCDA and normal mCCDA. . Previous studies have demonstrated the efficacy of addition of an ESBL inhibitor for selective isolation of Campylobacter from poultry meat.8 ± 38. Hayashi and others 2013).9 CFU/mL non-Campylobacter was found on C-mCCDA.0A mCCDA 160.1 ± 8. Hayashi and others 2013). indicating that the quantitative detection of these bacteria was similar in the 2 selective agar media (Figure 1). Nr. 80%). and the remainder were E.Enumeration of Campylobacter with C-mCCDA . In this study. 100%). Strains Campylobacter Non-Campylobacter Value (CFU/mL) Mean ± Range Mean ± SEMa Range SEMa C-mCCDA 145. (2 of 10. of positives/ total number of samples (%) Strains Campylobactera Non-Campylobactera C-mCCDA mCCDA 45/120A 46/120A (38.5 ± 0 to 2125 1. only 2 were Acinetobacter spp. 20%).3A 0 to 20 36. no study has evaluated the effect of potassium clavulanate on the quantitative detection of Campylobacter from chicken carcass rinse without an enrichment step.5) 33/120A (27. Oyarzabal and others 2005. Although mCCDA agar is the most popular selective agar media for the isolation of Campylobacter in poultry samples. Of the 10 contaminant colonies subcultured from mCCDA.8) (37.05) C-mCCDA (33 of 120) plates were contaminated with nonCampylobacter strains than normal mCCDA plates were (67 of 120).9A 0 to 2750 27.5) a Statistically significant differences among values in rows are indicated by differing uppercase letters (P < 0.

Prevalence of Campylobacter. 2013. Platteel T. van den Munckhof MP. Uyttendaele M. Kubota-Hayashi S. 1995. 2010. Sanders GW. Mizuno T. Harrington CS. J Microbiol Methods 88:304–10. Poultry Sci 59:2382–5. coli in chicken carcass rinses. in chicken carcass rinse. J Clin Microbiol 38:286–91. Radhouani H. Macrolide-resistant Campylobacter: the meat of the matter. and Sutterella spp. Leverstein-van Hall MA. Comparison of two types of broths and three selective agars for the detection of Campylobacter spp. Miyata M. Food Microbiol 28:1117–23. Kim HS. . Belanger AE. Kelly C. J Rapid Methods Autom Microbiol 14:110–7. Can J Microbiol 38:313–6. Evaluation of ISO 10272:2006 standard versus alternative enrichment and plating combinations for enumeration and detection of Campylobacter in chicken meat. Development of a selective differential agar for isolation and enumeration of Campylobacter spp. Yim JH. in whole chicken carcass-rinse samples. 2008. 2007. van de Sande-Bruinsma N. Nye K. Switzerland. Comparison of methods for recovery and enumeration of Campylobacter from freshly processed broilers. Blivet D. ISO 10272-1:2006. and Fisheries. Mevius DJ. Research Foundation of Korea (NRF) funded by the Ministry of Education. 2006. References M: Food Microbiology & Safety Ahmed R. Denayer S. J Food Prot 64:1711–5. Lattuada CP. JungWhan Chon and Hong-Seok Kim were also partially supported by the Brain Korean 21 (BK21) Project from the Ministry of Education and Human Resources Development. although C-mCCDA provided greater (P < 0. International Organization for Standardization (ISO). 2007). Chon JW. Geneva 20. Aalbaek B. Rajkovic A. Intl J Food Microbiol 135:248–53. Our results indicate that the increased selectivity of C-mCCDA may be a useful tool in the enumeration of Campylobacter cells in carcass rinse samples. Josefsen MH. Harvey M. Colin P. 2009. Part 1: Detection Method. 2005. Costa L. Taylor J. 2012b. 2011. Benson ST. Poeta P. quantitative detection ability of C-mCCDA to recover Campylobacter cells in pure culture and chicken carcass rinse is statistically similar (P > 0. Moran L. Allos M. Hyeon JY. Hansen TB. 79. Macklin KS. De Zutter L. Rosenquist H. Uyttendaele M. Dierikx CM. M926 Journal of Food Science r Vol. in poultry retail samples. from raw chicken.. Seo KH. Comparison of two types of plating media for detection and enumeration of Campylobacter from poultry samples. Woodford N. High prevalence of extendedspectrum b-lactamases Escherichia coli and vancomycin-resistant enterococci isolates from chicken products. Occurrence of thermotolerant campylobacters in fresh vegetables sold at farmers’ outdoor markets and supermarkets. Lett Appl Microbiol 52:614–8. L´opez-G´alvez F. J Antimicrob Chemother 60:715–23. Scharinga J. from poultry carcass rinses. Improvement of modified charcoal-cefoperazonedeoxycholate agar by addition of potassium clavulanate for detecting Campylobacter spp. Yoshida S. Conclusion According to our findings. Restoring the selectivity of Bolton broth during enrichment for Campylobacter spp. Agriculture. Chon JW. Kim H. Seo KH. O’Neill P. Dutch patients. Colin P. Madden RH. Habib I. Nachamkin IB. Corry JE. Case Postale 56. plasmids and strains. Line JE. 2000. Park JH. 2005. Comparison of enrichment vessels for recovery of Campylobacter spp. Le´on-Velarde CG. 3. Intl J Food Microbiol 163:41–6. Hyeon JY. Debevere J. Laisney MJ. Kim JH. and Technology. McGettrick S. Post DE. Natori T. Voets GM. retail chicken meat and poultry share the same ESBL genes. Kawamoto K.05) selectivity than normal mCCDA. Butler V. Cormican M. 2007. 2011. Ohkusu K. Ermel G. from broiler rinse samples. Intl J Poult Sci 4:81–4. Microbiology of food and animal feeding stuffs-horizontal method for the detection and enumeration of Campylobacter spp. Line JE. Line JE. J Food Safety 30:141–53. Bonten MJ. 2012a. Development of a m-PCR assay for simultaneous identification of Campylobacter jejuni and C. Zhang J. Barbaree JM. Appl Environ Microbiol 71:3351–4. Ho T. Odumeru JA. Forestry. Stern NJ. . Use of blood-free enrichment broth in the development of a rapid protocol to detect Campylobacter in twenty-five grams of chicken meat. Clin Microbiol Infect 17:873– 80. National ESBL surveillance group. Bengtsson A. and by the Export Promotion Technology Development Program of iPET (nr. Helicobacter. Salvat G. Stuart JC. Miller RS. Arcobacter. Improvement of modified charcoal-cefoperazone-deoxycholate agar by supplementation with a high concentration of polymyxin B for detection of Campylobacter jejuni and C. 2013. van Essen-Zandbergen A. ISO Central Secretariat. A problem of public health. Characterization of Escherichia coli from raw poultry in Belgium and impact on the detection of Campylobacter jejuni using Bolton broth.Enumeration of Campylobacter with C-mCCDA . Acknowledgments This research was supported by Mid-career Researcher Program (2012R1A2A2A-01015344) through the Natl. Oyarzabal OA. Hoorfar J. A collaborative study on a Nordic standard protocol for detection and enumeration of thermotolerant Campylobacter in food (NMKL 119. Preston and Park-Sanders protocols adapted for semi-quantitative isolation of thermotolerant Campylobacter from chicken rinse. Park CE. . Botteldoorn N. 1998. Igrejas G. Ed. Hawkey PM. Jasson V. Gerner-Smidt P. 2001. 2014 Engberg J. Hayashi M. On SL. 1 rue de Varemb´e. Science. Warren RE. Intl J Food Microbiol 26:43–76. Clin Microbiol Rev 11:555–67. Sampers I. 5. Livermore DM. Makino S. L¨ubeck PS. Habib I. Nr. Campylobacter species and Guillain-Barre´ syndrome. Part 2: Colony Count Technique. Berrang ME. Evaluation of agar plates for direct enumeration of Campylobacter spp. 2006. in human fecal samples as estimated by a reevaluation of isolation methods for Campylobacters. Seo KH. Chon JW. Intl J Food Microbiol 118:201–13. ISO/TS 10272-2:2006. 1992. Ezaki T. Ensor VM. Soumet C.05) to that of mCCDA. Line JE. Shryock TR. Evaluation of novel agars for the enumeration of Campylobacter spp. 2011. Song KY. Imported chicken meat as a potential source of quinoloneresistant Escherichia coli producing extended-spectrum b-lactamases in the UK. Kim JH. 2003. 2001. De Zutter L. Rivoal K. Song KY. Baert L. J Food Prot 64:982–6. Denis M. Fluit AC. Culture media for the isolation of Campylobacters. Appl Environ Microbiol 78:1624–6. Lett Appl Microbiol 29:406–10. Intl J Food Microbiol 165:7–10. 1999. Intl J Food Microbiol 80:177–83. 313010-3) funded by the Ministry for Food. CH-1211. Gomes C. coli. 2012. J Antimicrob Chemother 61:504–8.