Journal of Veterinary Diagnostic Investigation

Fingerprinting of poultry isolates of 24(6) 1166­–1171

© 2012 The Author(s)
Reprints and permission:
Enterococcus cecorum using three sagepub.com/journalsPermissions.nav
DOI: 10.1177/1040638712463563
http://jvdi.sagepub.com
molecular typing methods

Dona Saumya Wijetunge, Patricia Dunn, Eva Wallner-Pendleton,

Valerie Lintner, Huaguang Lu, Subhashinie Kariyawasam1

Abstract. Enterococcus cecorum is an emerging challenge to the broiler industry. The organism has been implicated in
septicemia, spondylitis, arthritis, and osteomyelitis in commercial broilers and broiler breeders, which lead to economic losses
attributed to increased mortality and culling rates, decreased average processing weights, and increased feed conversion ratios.
The current study evaluated the genetic variability of 30 clinical isolates of E. cecorum from outbreaks in Pennsylvania,
using 3 molecular typing methods, namely, pulsed-field gel electrophoresis (PFGE), randomly amplified polymorphic DNA
analysis, and enterobacterial repetitive intergenic consensus–PCR (polymerase chain reaction), in order to understand their
genetic relatedness and to identify possible pathogenic clones. The study revealed the existence of genotypic polymorphism
among E. cecorum associated with clinical disease. Of the 3 typing methods used, PFGE analysis demonstrated higher genetic
variability of E. cecorum isolates compared to PCR-based methods. Also, each molecular typing method was evaluated in
terms of typeability, discriminatory power, and reproducibility for application of these typing methods in fingerprinting of
E. cecorum in future reference. Pulsed-field gel electrophoresis provided the most reliable results with greater discriminatory
power and higher reproducibility compared to the 2 PCR-based methods.

Key words: Broiler breeders; chickens; commercial broilers; Enterococcus cecorum; fingerprinting; molecular typing.

Introduction other bacteria, there is no typing method available for

E. cecorum. Accordingly, the objective of the current study
Enterococcus cecorum, which was formerly known as was to analyze the genetic variability of E. cecorum and to
Streptococcus cecorum, is an emerging pathogen of the broiler evaluate its genetic relatedness in relation to the source of
industry. The organism was first isolated from chicken intes- infection.
tinal flora in 1983.4 Enterococcus cecorum is a facultatively Various nucleic acid–based techniques have been used to
anaerobic, catalase-negative, alpha-hemolytic, Gram-positive type and characterize the genetic variability of entero-
coccus that belongs to an atypical group of enterococci that cocci.5,6,16 Such molecular techniques include pulsed-field
do not grow in 6.5% NaCl.3 Even though E. cecorum was gel electrophoresis (PFGE), plasmid profiling, and poly-
previously known as a commensal in poultry, it has recently merase chain reaction (PCR)-based methods such as ran-
been recognized as an important pathogen of commercial domly amplified polymorphic DNA PCR (RAPD-PCR),
broilers and broiler breeders. Infection with E. cecorum is repetitive extragenic palindromic sequence PCR (Rep-PCR),
apparent in various forms where spondylitis, osteomyelitis, and enterobacterial repetitive intergenic consensus PCR
arthritis, and septicemia are the most predominant signs.12,13 (ERIC-PCR). Of these methods, PFGE is considered the
Sporadic cases of septicemia, endocarditis, and osteomyeli- “gold standard” for molecular typing of many bacterial spe-
tis caused by E. cecorum have also been reported in immune- cies.15 The PFGE SmaI microrestriction patterns of entero-
compromised human patients.1,7,14 cocci have been used extensively for genotyping and
Despite the acknowledged status of the organism as a fingerprinting of other species of enterococci.9,15,16 Similarly,
poultry pathogen, the exact mode of transmission, port of
entry, and virulence mechanisms of E. cecorum are largely
unknown. Therefore, appropriate epidemiological surveil- From the Department of Veterinary and Biomedical Science, Animal
lance is a necessity for disease investigation, and implication Diagnostic Laboratory, The Pennsylvania State University, University
Park, PA.
of preventive and control measures in an outbreak situation.
Genotyping to determine the clonal diversity or to track a Supplemental material available at jvdi.sagepub.com/supplemental.
particular clonal type plays a pivotal role in epidemiological 1
Corresponding Author: Subhashinie Kariyawasam, Department of
surveillance. Although various genomic “fingerprinting” Veterinary and Biomedical Sciences, The Pennsylvania State University,
approaches have been widely used for these purposes with University Park, PA 16802. sxk91@psu.edu
 Enterococcus cecorum genotyping 1167

RAPD-PCR, Rep-PCR, and ERIC-PCR also have been used RAPD-PCR and ERIC-PCR procedure
for molecular characterization of enterococci species present
in dairy products and sea foods, as well as in clinical set- Extraction of genomic DNA was performed using a genomic
tings.2,9,16 In the present study, 3 typing methods, namely, DNA extraction kitg according to the manufacturer’s instruc-
PFGE, RAPD-PCR, and ERIC-PCR, were used for the pur- tions. Both RAPD-PCR and ERIC-PCR were performed as
pose of genotyping and assessing the genetic relatedness of previously described.9 The RAPD-PCR assay was carried
30 isolates of E. cecorum associated with clinical disease. out in 50-μl reaction volumes containing 3 mM of MgCl2,
400 μM of deoxyribonucleotide triphosphates (dNTPs),
1 μM of M13 primer (5′-GAGGGTGGCGGTTCT-3′), 1.25
Materials and methods U of Taq DNA polymerase,j and 200 ng of template DNA in
Isolation and identification of E. cecorum strains 1× PCR buffer.j Cycling parameters used for amplification
were an initial denaturing at 94°C for 60 sec, followed by 35
Bacteria were isolated from clinical cases submitted to the cycles of 94°C for 60 sec, 40°C for 20 sec, and 72°C for 80
Animal Diagnostic Laboratory at the Pennsylvania State sec, and final extension at 72°C for 5 min. The ERIC-PCR
University (University Park, Pennsylvania) from 2008 to reactions were carried out in 25-μl reaction volumes contain-
2011. For bacterial isolation, tissue samples were inoculated ing 3 mM of MgCl2, 400 μM of dNTPs, 1 μM of ERIC1-R
onto Columbia colistin and nalidixic acid agar plates with primer (5′-ATGTAAGCTCCTGGGGATTCAC-3′), 1.25 U
5% sheep blood and incubated at 37°C for 48 hr. Bacteria of Taq DNA polymerase,j and 200 ng of template DNA in 1×
were presumptively identified as E. cecorum by Gram- PCR buffer. Amplification conditions were as follows: ini-
staining, an automated identification system,a growth char- tial denaturing at 94°C for 3 min, 35 cycles of 94°C for 30
acteristics in 6.5% NaCl broth,b and ability to ferment sec, 48°C for 60 sec, and 72°C for 5 min, and final extension
raffinose.c Presumptively identified bacteria were confirmed at 72°C for 7 min. The PCR products were analyzed follow-
as E. cecorum by 16S ribosomal RNA gene sequencing.8 ing electrophoresis on 1.5% agarosek at 120 V for 2 hr in 1×
TAE buffer (40 mM of Tris-acetate, 2.5 mM of EDTA; pH 8).
Gels were stained with ethidium bromide,d and the images
PFGE procedure
were captured using a gel documentation system.i
Preparation of DNA plugs of E. cecorum was performed
with a previously described rapid DNA preparation proce-
Data analysis
dure with minor modifications.10 Briefly, bacterial pellets of
2.5 ml of overnight cultures were resuspended in 0.5 ml of Banding patterns obtained in each typing method were ana-
2× lysis solution containing 12 mM of Tris-HCl (pH 7.4), lyzed using a commercial software.l Similarity percentages
2 M of NaCl,d 20 mM of ethylenediamine tetra-acetic acid of fingerprints were analyzed using dendrograms constructed
(EDTA; pH 7.5), 1% Brij,e 0.4% deoxycholate,e 1% sodium by Dice coefficient and the unweighted pair group method
lauroylsarcosine,f 1.0 mg/ml of lysozyme,d and 100 µg/ml of with arithmetic averages with 1.5% band position tolerance.
RNase A.g Immediately after resuspension of the bacterial Similarity index >80% was used to assign genotypes with
pellet, 0.5 ml of 1.6% pulsed field certified agarose,h cooled each method. Simpson Index was used to evaluate the dis-
down to 50°C, was added. The mixture was vortexed briefly criminatory power of each molecular typing method (http://
before being loaded into the plug molds.h Once the plugs darwin.phyloviz.net/ComparingPartitions/index.php).
were solidified, the plugs were lysed first in 3 ml of 1× lysis
solution containing 0.5 mg/ml of lysozymed and 100 µg/ml
of RNaseg at 37°C for 2 hr with gentle shaking (100 rpm/ Results
min) and then in 3 ml of solution containing 10 mM of Tris Enterococcus cecorum strains
(pH 7.4), 1 mM of EDTA, 100 mg/ml of proteinase K,f and
1% sodium dodecyl sulfated at 50°C for 1 hr. Finally, the A total of 27 E. cecorum strains isolated from commercial
plugs were washed twice with 7 ml of TE buffer (10 mM broilers and broiler breeders with osteomyelitis (n = 6),
Tris-HCl [pH 7.4] and 0.1 mM of EDTA) at 50°C for 30 min spondylitis (n = 12), septicemia (n = 2), arthritis (n = 5), cel-
and stored in 5 ml of TE buffer until used. According to the lulitis (n = 1), and yolk sac infection (n = 1) were used in the
literature, enterococci genomes have a low G+C content11; current study (Table 1). An additional 3 strains were isolated
therefore, restriction digestion was performed using SmaIg from cecal contents of a commercial broiler with spinal
for 2 hr at 25°C, and bands were separated using a commer- abscesses. Origin and source of the isolates are summarized
cial PFGE systemh on 1.5 % pulsed-field certified agaroseh in Table 1.
for 24 hr at 14°C with switching times ramped from 0.5 to
35 sec and a gradient of 6 v/cm. Enterococcus faecalis
PFGE analysis
American Type Culture Collection (ATCC) 29212 was used
as the molecular weight size standards. The restriction patterns SmaI restriction digestion of E. cecorum generated 12–16
were visualized by ultraviolet transillumination.i fragments with molecular weights ranging approximately
1168 Wijetunge et al.

Table 1. The source, origin, and genotype of Enterococcus cecorum isolates submitted to the Animal Diagnosis Laboratory, The
Pennsylvania State University (University Park, PA) during 2008 to 2011.*
Date submitted
Company Farm Bird type for necropsy Isolate ID Diagnosis Pulsotype ERIC type RAPD type
A A1 Commercial broiler 02/19/2009 1 Omphalitis P20 E8 R7
  A2 Commercial broiler 10/15/2010 26 Arthritis P14 E12 R7
B B1 Commercial broiler 02/09/2009 2 Arthritis P21 E7 R7
  B1 Commercial broiler 02/24/2009 16 Arthritis P1 E7 R5
  B2 Commercial broiler 03/31/2009 12 Osteomyelitis P13 E6 R7
  B2 Commercial broiler 04/01/2009 13 Osteomyelitis P8 E5 R7
  B3 Commercial broiler 07/06/2009 20 Arthritis P9 E5 R8
C C Broiler breeder 01/22/2009 3 Spondylitis P19 E5 R7
D D1 Commercial broiler 01/05/2009 4 Septicemia/hepatitis P17 E9 R4
  D2 Commercial broiler 07/21/2009 17 Cellulitis P6 E10 R6
E E1 Commercial broiler 11/13/2008 8 Osteomyelitis P13 E6 R7
  E2 Commercial broiler 11/13/2008 10 Arthritis P18 E8 R10
  E3 Commercial broiler 12/10/2008 5 Osteomyelitis P15 E5 R8
  E4 Commercial broiler 12/05/2008 6 Septicemia/pericarditis P15 E5 R8
  E5 Commercial broiler 11/26/2008 7 Osteomyelitis P16 E5 R8
  E6 Commercial broiler 11/02/2009 23 Spondylitis P12 E6 R7
F F1 Commercial broiler 04/11/2009 11 Spondylitis P15 E5 R8
G G1 Commercial broiler 03/02/2009 14 Spondylitis P11 E6 R7
  G2 Commercial broiler 03/02/2009 15 Spondylitis P11 E6 R7
H H1 Broiler breeder 10/13/2010 22 Spondylitis P11 E6 R7
I I1 Broiler breeder 03/09/2010 24 Spondylitis P2 E6 R9
J J1 Commercial broiler 03/04/2010 25 Spondylitis P14 E5 R7
  J1 Commercial broiler 03/10/2010 28 Spondylitis P14 E5 R7
K K Commercial broiler 09/30/2009 29 Osteomyelitis P4 E3 R9
L L1 Broiler breeder 05/16/2011 32 Spondylitis P5 E1 R5
  L2 Commercial broiler 06/13/2011 34 Spondylitis P14 E2 R7
M M1 Commercial broiler 08/02/2011 35 Spondylitis P13 E11 R7
  M2 Commercial broiler 08/02/2011 36 Cecal contents P7 E4 R1
  M3 Commercial broiler 08/02/2011 37 Cecal contents P10 E4 R2
  M4 Commercial broiler 08/02/2011 38 Cecal contents P3 E4 R3
* ERIC = enterobacterial repetitive intergenic consensus; RAPD = randomly amplified polymorphic DNA.

from 10 kb to 400 kb (Fig. 1A). Based on >80% similarity different genotypes were observed (Table 1, Fig. 2B). Of the
index, 21 pulsotypes were observed for the 30 E. cecorum 30 isolates tested, 7 isolates provided unique genotypes (E1,
isolates included in the study (Table 1, Fig. 1B). Within the E2, E3, E9, E10, E11, and E12) whereas genotype E5 con-
21 pulsotypes, 17 isolates had unique fingerprints. A total of tained the highest number of isolates (n = 9) with similarity
4 isolates showed the fingerprints of pulsotype 14, and pul- ranging from 81% to 100%. The Simpson Index of the
sotypes 11, 13 and 15 enclosed 3 isolates within each group ERIC-PCR was 0.857 (95% CI: 0.770–0.938). Based on the
with similarity ranging from 85% to 100%. The discrimina- 95% CI, statistically no difference was observed in the dis-
tory power of PFGE was high with a Simpson Index of 0.966 criminatory power between ERIC-PCR and PFGE. However,
(95% confidence interval [CI]: 0.937–0.994). Upon repeated the reproducibility that ERIC-PCR demonstrated was infe-
testing, PFGE provided exactly the same banding patterns rior to that of PFGE (see supplemental data).
ensuring the reproducibility of the test (see supplemental
data).
RAPD-PCR analysis
The M13 primer yielded amplicons ranging from 750 bp to 5
ERIC-PCR analysis
kb (Fig. 3A). Ten unique genotypes were observed for the 30
The ERIC-1R primer generated 2–7 amplicons with E. cecorum isolates tested (Table 1). The majority of isolates
molecular weights ranging from 200 to 5,000 bp (Fig. 2A). (46.66%) belonged to the RAPD genotype 7, with a similar-
According to the dendrogram with >80% similarity, 12 ity ranging from 85% to 100% (Fig. 3B). The RAPD-PCR
 Enterococcus cecorum genotyping 1169

Figure 1. Results of pulsed field gel electrophoresis (PFGE) examination of Enterococcus cecorum clinical isolates. A, PFGE of SmaI
macrorestriction profiles of E. cecorum clinical isolates. Lanes 1, 30: SmaI macrorestriction profiles of E. faecalis; lanes 2–29: fingerprints
of E. cecorum isolates. B, dendrogram based on Dice coefficient with 1.5% position tolerance. Cut-off value of 80% similarity was used to
assign the pulsotypes.

Figure 2. Results of enterobacterial repetitive intergenic consensus–polymerase chain reaction (ERIC-PCR) analysis of Enterococcus
cecorum clinical isolates. A, amplicons generated with the ERIC-1R primer. Lane 1: 100-bp DNA ladderg; lanes 2–24: fingerprints of
E. cecorum isolates. B, dendrogram based on Dice coefficient with 1.5% position tolerance. Cut-off value of 80% similarity was used to
assigning the ERIC-PCR types.
1170 Wijetunge et al.

Figure 3. Results of randomly amplified polymorphic DNA (RAPD) analysis of Enterococcus cecorum clinical isolates. A, amplicons
generated with the M13 primer. Lanes 1, 25: 1-kb DNA ladderg; lanes 2–24: fingerprints of E. cecorum isolates. B, dendrogram based
on Dice coefficient with 1.5% position tolerance. Cut-off value of 80% similarity was used to assigning the RAPD polymerase chain
reaction types.

exhibited the lowest discriminatory power, which was 0.731 Both PFGE and ERIC-PCR methods demonstrated a high
(95% CI: 0.577–0.885) and the lowest reproducibility com- genetic variability in E. cecorum isolates. Interestingly, all 3
pared to both PFGE and ERIC-PCR methods (see supple- typing methods gave indistinguishable fingerprints for iso-
mental data). lates 14 and 15 from company G with 100% similarity.
These 2 isolates originated from infected birds on 2 different
farms (farms G1 and G2) in outbreaks that occurred during
Discussion the same month. This suggests that the source of E. cecorum
Enterococcus cecorum is an emerging pathogen of concern outbreaks in farm G1 and farm G2 might be from a common
to the broiler industry. Previously, it was recognized as a source (company G). The same explanation would be applied
commensal that inhabits the intestinal tract of avian species to isolates 5 and 6 that originated from company E. Also,
as well as mammals. Since 2006, several outbreaks of spon- isolates 25 and 28 from farm J1, which were isolated from
dylitis, vertebral abscesses, progressive lameness, and septi- the same outbreak at different time points, produced identi-
cemia associated with E. cecorum in broiler breeders and cal fingerprints with all 3 typing methods. This implies that
straight-run broilers were reported from many countries the outbreak on farm J1 was associated with the same clone
including the United States.12,13,17 Increased mortality and of E. cecorum. Isolates 2 and 16 from farm B1 and isolates
high culling rates associated with the disease could be eco- 12 and 13 from farm B2 were identified as different geno-
nomically devastating to the industry. Because the route of types by all 3 typing methods, suggesting that more than
entry, the mode of transmission, and the virulence mecha- one clonal type of E. cecorum was associated with these
nisms of E. cecorum infections are unknown, identification outbreaks.
of the source of outbreaks is required for control and preven- It is evident that pulsotyping appears to be a promising
tion. The objective of the current study was to assess the method for genotyping E. cecorum because of its superior
genetic polymorphism of clinical isolates of E. cecorum of discriminatory power and reproducibility compared to
diverse origins in order to understand their relatedness and to RAPD-PCR and ERIC-PCR. However, it is a time-consuming,
identify possible pathogenic clones using 3 molecular typing labor-intensive, cumbersome technique for routine clinical
methods, namely PFGE, RAPD-PCR, and ERIC-PCR. application. The requirement of expensive equipment and
Also, the typeability, discriminatory power, reproducibility the lengthiness of the procedure are major disadvantages of
of results, and feasibility of the technique of the each typing PFGE over PCR-based methods. In the present study, a mod-
method were evaluated. ification of a previously published protocol of PFGE10 was
 Enterococcus cecorum genotyping 1171

followed, which entailed a relatively simple method requir- References

ing only 2 days to obtain fingerprint patterns. Although
PCR-based methods are very easy to perform and relatively 1. Ahmed FZ, Baig MW, Gascoyne-Binzi D, Sandoe JA: 2011,
inexpensive compared to PFGE, the results are more diffi- Enterococcus cecorum aortic valve endocarditis. Diagn Micro-
cult to analyze and interpret due to the appearance of multi- biol Infect Dis 70:525–527.
ple weak bands in the PCR profiles when repeated. 2. Bedendo J, Pignatari AC: 2000, Typing of Enterococcus fae-
In conclusion, E. cecorum isolates used in the present cium by polymerase chain reaction and pulsed field gel electro-
study revealed genetic polymorphism as detected by all 3 phoresis. Braz J Med Biol Res 33:1269–1274.
molecular typing methods. Furthermore, genotypes indistin- 3. Devriese LA, Ceyssens K, Haesebrouck F: 1991, Character-
guishable by all 3 methods revealed genetic relatedness istics of Enterococcus cecorum strains from the intestines of
between certain isolates, which could be rationalized by the different animal species. Lett Appl Microbiol 12:137–139.
source of infection and nature of outbreaks. Nevertheless, 4. Devriese LA, Dutta GN, Farrow JAE, et al.: 1983, Streptococ-
PFGE, RAPD-PCR, and ERIC-PCR techniques used in the cus cecorum, a new species isolated from chickens. Int J Syst
present study for typing of E. cecorum revealed differences Evol Microbiol 33:772–776.
in terms of discriminatory power, reproducibility, ease of 5. Drahovská H, Kocíncová D, Seman M, Turna J: 2002, PCR-
performance, and interpretation of results. Pulsed-field gel based methods for identification of Enterococcus species. Folia
electrophoresis provided the most reliable results with Microbiol (Praha) 47:649–653.
greater discriminatory power and higher reproducibility 6. Gordillo ME, Singh KV, Murray BE: 1993, Comparison of ribotyping
compared to the 2 PCR-based methods. Although the dis- and pulsed-field gel electrophoresis for subspecies differentiation of
criminatory power of ERIC-PCR was not statistically differ- strains of Enterococcus faecalis. J Clin Microbiol 31:1570–1574.
ent from that of PFGE, reproducibility was relatively low. 7. Greub G, Devriese L, Pot B, et al.: 1997, Enterococcus ceco-
The RAPD-PCR gave the lowest discriminatory power rum septicemia in a malnourished adult patient. Eur J Clin
among the 3 typing methods. Further studies have to be Microbiol Infect Dis 16:594–598.
undertaken with a larger number of isolates comprising both 8. Hall L, Doerr KA, Wohlfiel SL, Roberts GD: 2003, Evaluation of
clinical and commensal E. cecorum collected from diverse the MicroSeq system for identification of mycobacteria by 16S
sources for further evaluation of PFGE as a useful molecular ribosomal DNA sequencing and its integration into a routine clini-
typing method for epidemiological surveillance of E. cecorum cal mycobacteriology laboratory. J Clin Microbiol 41:1447–1453.
outbreaks. 9. Martín-Platero AM, Valdivia E, Maqueda M, Martínez-Bueno M:
2009, Characterization and safety evaluation of enterococci isolated
Sources and manufacturers from Spanish goats’ milk cheeses. Int J Food Microbiol 132:24–32.
a. Sensititre Gram-positive identification system, TREK Diagnostic 10. Matushek MG, Bonten MJ, Hayden MK: 1996, Rapid prepara-
Systems Inc., Cleveland, OH. tion of bacterial DNA for pulsed-field gel electrophoresis. J Clin
b. Remel Inc., Lenexa, KS. Microbiol 34:2598–2600. Erratum in J Clin Microbiol 1997 35:536.
c. Key Scientific Products Inc., Stamford, TX. 11. Palmer KL, Carniol K, Manson JM, et al.: 2010, High-quality
d. Fisher Scientific, Pittsburg, PA. draft genome sequences of 28 Enterococcus sp. isolates. J Bacteriol
e. G-Biosciences, Maryland Heights, MO. 192:2469–2470.
f. Sigma-Aldrich, St. Louis, MO. 12. Perez S: 2004, Spinal cord lesions in broiler chickens. Vet Rec 155:536.
g. Promega Corp., Madison, WI. 13. Stalker MJ, Brash ML, Weisz A, et al.: 2010, Arthritis and
h. Bio-Rad Laboratories, Hercules, CA. osteomyelitis associated with Enterococcus cecorum infection
i. Alpha Innotech Corp., Santa Clara, CA. in broiler and broiler breeder chickens in Ontario, Canada. J Vet
j. TopTaq DNA Polymerase, Qiagen Inc., Valencia, CA. Diagn Invest 22:643–645.
k. Seakem agarose, Cambrex Corp., East Rutherford, NJ. 14. Tarr PE, Sakoulas G, Ganesan A, et al.: 2004, Hematogenous
l. BioNumerics version 4.0, Applied Maths Inc., Austin, TX. enterococcal vertebral osteomyelitis: report of 2 cases and
review of the literature. J Infect 48:354–362.
Declaration of conflicting interests 15. Tenover FC, Arbeit RD, Goering RV, et al.: 1995, Interpreting
The author(s) declared no potential conflicts of interest with chromosomal DNA restriction patterns produced by pulsed-field
respect to the research, authorship, and/or publication of this gel electrophoresis: criteria for bacterial strain typing. J Clin
article. Microbiol 33:2233–2239.
16. Valenzuela AS, Benomar N, Abriouel H, et al.: 2010, Isola-
Funding tion and identification of Enterococcus faecium from seafoods:
The author(s) disclosed receipt of the following financial support antimicrobial resistance and production of bacteriocin-like sub-
for the research, authorship, and/or publication of this article: stances. Food Microbiol 27:955–961.
This study was supported by the Pennsylvania Department of 17. Wood AM, MacKenzie G, McGiliveray NC, et al.: 2002, Iso-
Agriculture and Broiler Check-off Program of the Pennsylvania lation of Enterococcus cecorum from bone lesions in broiler
University. chickens [letter]. Vet Rec 150:27.