Genomic analysis of Pediococcus starter cultures used to control Listeria monocytogenes in turkey summer sausage.

Abstract

The pulsed-field technique of clamped homogeneous electric field electrophoresis was employed to characterize and size genomic DNA of three pediocin-producing (Ped+) and two non-pediocin-producing (Ped-) strains of Pediococcus acidilactici. Comparison of genomic fingerprints obtained by digestion with the low-frequency-cleavage endonuclease AscI revealed identical restriction profiles for four of the five strains analyzed. Summation of results for 10 individually sized AscI fragments estimated the genome length to be 1,861 kb for the four strains (H, PAC1.0, PO2, and JBL1350) with identical fingerprints. Genomic analysis of the pediocin-sensitive, plasmid-free strain P. acidilactici LB42 with the unique fingerprint revealed nine AscI fragments and a genome length of about 2,133 kb. Ped- (JBL1350) and Ped+ (JBL1095) starter cultures (one each) were used to separately prepare turkey summer sausage coinoculated with a four-strain Listeria monocytogenes mixture (ca. 10(5) CFU/g). The starter cultures produced equivalent amounts of acid during fermentation, but counts of L. monocytogenes were reduced to a greater extent in the presence of the Ped+ starter culture (3.4 log10 unit decrease) than in the presence of the Ped- starter culture (0.9 log10 unit decrease). Although no listeriae were recovered from sausages following the cook/shower, appreciable pediocin activity was recovered from sausages prepared with the Ped+ strain for at least 60 days during storage at 4 degrees C. The results of this study revealed genomic similarities among pediococcal starter cultures and established that pediocins produced during fermentation provide an additional measure of safety against listerial proliferation in turkey summer sausage.

Full text

Page 1

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Sept. 1992, p. 3053-3059
0099-2240/92/093053-07$02.00/0
Copyright C 1992, American Society for Microbiology
Vol. 58, No. 9
Genomic Analysis of Pediococcus Starter Cultures Used To Control
Listeria monocytogenes in Turkey Summer Sausage
JOHN B. LUCHANSKY,1,2* KATHLEEN A. GLASS,' KARTIKA D. HARSONO,' ALAN J. DEGNAN,
NANCY G. FAITH,' BEATRICE CAUVIN,1 GAIL BACCUS-TAYLOR,2 KEIZO ARIHARA 1,3
BILIG BATER,1 ARTHUR J. MAURER,2'4 AND ROBERT G. CASSENS3
Departments ofFood Microbiology and Toxicology,1Food Science,2 and Poultry Science4 and
the Muscle Biology Laboratory,3 1925 Willow Drive, University of Wisconsin, Madison, Wisconsin 53706
Received 25 March 1992/Accepted 6 July 1992
The pulsed-field technique of clamped homogeneous electric field electrophoresis was employed to charac-
terize and size genomic DNA of three pediocin-producing (Ped+) and two non-pediocin-producing (Ped-)
strains of Pediococcus acidilactici. Comparison of genomic fingerprints obtained by digestion with the
low-frequency-cleavage endonuclease AscI revealed identical restriction profiles for four of the five strains
analyzed. Summation of results for 10 individually sized AscI fragments estimated the genome length to be
1,861 kb for the four strains (H, PAC1.0, P02, and JBL1350) with identical fingerprints. Genomic analysisof
the pediocin-sensitive, plasmid-free strain P. acidilactici LB42 with the unique fingerprint revealed nine AscI
fragments and a genome length of about 2,133 kb. Ped- (JBL1350) and Ped+ (JBL1095) starter cultures (one
each) were used to separately prepare turkey summer sausage coinoculated with a four-strain Listeria
monocytogenes mixture (ca. 10V CFU/g). The starter cultures produced equivalent amounts of acid during
fermentation, but counts of L. monocytogenes were reduced to a greater extent in the presence of the Ped+
starter culture (3.4loglounit decrease) than in the presence ofthe Ped- starter culture (0.9loglounit decrease).
Although no listeriae were recovered from sausages following the cook/shower, appreciable pediocin activity
was recovered from sausages prepared with the Ped+ strain for at least 60 days during storage at 4°C. The
results of this study revealed genomic similarities among pediococcal starter cultures and established that
pediocins produced during fermentation provide an additional measure of safety against listerial proliferation
in turkey summer sausage.
In recent years, there have been several episodes of
food-borne illness involving Listeria monocytogenes (11).
Surveillance studies have shown that the pathogen is com-
mon in agricultural ecosystems (39) and prevalent on various
foods at the slaughterhouse and retail levels, most notably
on poultry (2, 13, 14, 23, 40). With the exception of sporadic
listeriosis linked to consumption of undercooked chicken
and nonreheated turkey frankfurters (1, 36), there have been
no major outbreaks of listeriosis associated with poultry.
However, conventional preservation methods, including re-
frigeration, may not preclude survival or growth of L.
monocytogenes in foods, including turkey products. For
example, the organism grew on vacuum-packaged refriger-
ated turkey loaf (24) and sliced turkey (16). Thus, the
development of innovative and complementary methods for
controlling this pathogen in poultry and other foods has
become an active area of research.
During the past decade, considerable research has focused
on the application of molecular technologies to study lactic
acid bacteria, an economically significant group oforganisms
used to prepare and preserve foods (27). Regarding this
group of organisms, relatively little information has accumu-
lated concerning genetic characterization of pediococcal
starter cultures used in meat fermentation. There have been
a few reports describing pediococci that harbor plasmids,
including plasmids encoding pediocins (9, 18, 19, 21, 33), but
thus far efforts have not been made to characterize the
chromosome of Pediococcus spp. For several other gram-
positive bacteria (for examples, see references 6-8, 22, and
*Corresponding author.
30), including other lactic acid bacteria (25, 38), the tech-
nique of pulsed-field gel electrophoresis (PFGE) has been
used to extensively characterize and size genomic DNA. In
a similar fashion, genomic analyses of pediococci by PFGE
would expand our knowledge of these genetically ill-defined
organisms and provide considerable information on genetic
relatedness among starter cultures.
In addition to acid and flavor compounds, many Pedio-
coccus spp. also produce pediocins that exhibit antilisterial
activity. The efficacy ofbiopreservatives (i.e., pediocins and
pediococci) for controlling L. monocytogenes has been
demonstrated with fresh (29), processed (3, 10a, 42), and
fermented meats (4, 12, 15, 34) as well as refrigerated
dairy-based products (32). However, there have been no
reports on the use of biopreservation systems to control L.
monocytogenes associated with turkey products. In the
present study, we compared the efficacy of pediocin-produc-
ing (Ped+) and non-pediocin-producing (Ped-) pediococcal
starter cultures for controlling L. monocytogenes in turkey
summer sausage. In addition, we utilized pulsed-field analy-
sis (i.e., genomic fingerprinting) to establish genetic relation-
ships among Pediococcus starter cultures.
MATERIALS AND METHODS
Bacteria. All strains used in this study are listed in Table 1.
Pediococcus acidilactici H (wild type of JBL1095 [5]),
PAC1.0 (also called JBL1096), P02 (also called JBL1097),
and JBL1350 were isolated from commercial starter culture
preparations or from fermented sausages preparedwith such
cultures. P. acidilactici LB42 (also called JBL1146), kindly
provided by Bibek Ray (University of Wyoming, Laramie),
3053

Page 2

3054
LUCHANSKY ET AL.
TABLE 1. Designations, characteristics, and origins of bacterial strains
Organism
designation
Relevant characteristicsa
Reference or source
P. acidilactici
JBL1095
JBL1096b
JBL1097
JBL1146
JBL1350
Pediocin AcH producer, Suc+ str-10 rf-10 (pSMB74)
Pediocin PA-1 producer, Suc+ (pSRQ10, pSRQ11)
Pedr Ped+ ccc+ Suc+
Peds Ped- ccc-
Pedr Ped- ccc+
42
18
21
33
Commercial starter
culture
PAC1.0
P02
LB42
L. monocytogenes
JBL1002
JBL1003
JBL1006
JBL1012
JBL1226
aPedr, pediocin resistant; Ped', pediocin sensitive; ccc+, contains plasmid (5); ccc, plasmid free; str-10, streptomycin resistant (1,000 pLg/ml); nf-10, rifamycin
resistant (100 pLg/ml); Suc+, ferments sucrose.
bIsolated from LACTACEL 110 (Microlife Technics, Inc., Sarasota, Fla.).
cA. M. Lammerding, Health of Animals Laboratory, Guelph, Ontario, Canada.
dJ. S. Bailey, Russell Research Center, Athens, Ga.
1ioM
103M
Meat isolate, serotype 4b
Meat isolate, serotype 1/2c
Hyperhemolytic derivative of NCTC 5105; serotype 3a
15
15
Lammerdinge
35
Baileyd
8732
AP63
Turkey frankfurter isolate, serotype 1/2a
and all L. monocytogenes strains used in this study are
pediocin sensitive (10, 10a, 33, 42). Pediococci and listeriae
were maintained as previously described (lOa, 42).
Plasmid isolation. Covalently closed circular DNA was
extracted from pediococci by the method described by
Muriana and Klaenhammer (28), but in addition to lyso-
zyme, 10RIof mutanolysin (1 mg/ml in H20; Sigma Chem-
ical Company, St. Louis, Mo.) was added to facilitate cell
lysis. Plasmid DNA was further purified by ethidium bro-
mide density gradient ultracentrifugation (26). After fraction-
ation by agarose gel electrophoresis (3 h at 70 V using
electrophoresis-grade agarose [0.7%]), gels were stained in
ethidium bromide (<1 mg/ml; Sigma) for about 15 min and
then photographed after visualization on a shortwave UV
transilluminator.
Purification, digestion, and resolution of high-molecular-
weight DNA fragments. Pediococci were passed twice in
brain heart infusion (Difco Laboratories, Detroit, Mich.)
broth at 37°C, harvested, washed, and suspended in agarose
plugs essentially as described previously (22). Intact high-
molecular-weight genomic DNA was prepared from pedio-
coccus cells embedded in agarose plugs by using the method
of Carriere et al. (8), but 20 pl of RNase (10 mg/ml in H20;
Sigma) was included during the lysozyme step, and plugs
were gently shaken while immersed in this solution. Prior to
digestion, inserts were washed once with 10 mM Tris-0.1
mM EDTA (pH 7.5) (TE buffer) plus 1 mM phenylmethyl-
sulfonyl fluoride and twice in TE buffer only, as described
previously (8). Each washing step was performed at room
temperature for 1 h.
Next, each plug was cut into about five inserts, and each
insert was placed into a sterile microcentrifuge tube for
digestion of embedded DNA with approximately 2 U of the
"rare cutting" restriction endonucleaseAscI as described by
the manufacturer (Promega Corporation, Madison, Wis.).
Following digestion for at least 16 h at 37°C with gentle
shaking, 20RIof 0.5 M EDTA (pH 8) was added to stop the
reaction. Inserts could then be stored at 4°C for 1 to 2 days
before fractionation by PFGE.
A clamped homogeneous electric field (CHEF) CHEF-DR
II (Bio-Rad Laboratories, Richmond, Calif.) pulsed-field
system and electrophoresis-grade (GIBCO-Bethesda Re-
search Laboratories, Life Technologies, Inc., Gaithersburg,
Md.) agarose were used to resolve high-molecular-weight
restriction fragments. Running buffer (0.5x Tris-borate-
EDTA) was cooled to 4°C prior to electrophoresis and
maintained at 14°C for the entire run by using an Isotemp
refrigerated circulator (Fisher Scientific, Pittsburgh, Pa.).
Electrophoresis was performed for 18 h at 200 V by using
electrophoresis-grade agarose (1%) and pulse times ramped
from 5 to 30 s. Gels were stained and photographed as
described above for plasmid DNA.
Preparation of bacteria for inoculation of sausage.
Pediococci. P.
acidilactici starter cultures JBL1095 and
JBL1350 were passed twice in MRS (Difco) broth at 30°C
and screened for pediocin production (described below)
prior to use. To prepare turkey inocula, starter cultures were
separately grown in 500 ml of MRS broth for 16 h at 30°C.
Cells were harvested by centrifugation (2,000 x g, 20 min,
4°C), suspended in 150 ml of 0.1% peptone water (ca. 109
CFU/ml), and added directly to sausage batter as described
below.
(ii). Listeriae. L. monocytogenes JBL1002, JBL1003,
JBL1012, and JBL1226 were grown individually in 10 ml of
tryptose phosphate broth (Difco) at 37°C for 16 h. The cells
from each strain were harvested by centrifugation and sus-
pended in a nominal volume of 0.1% peptone. Next, all four
cell suspensions were combined and the volume was ad-
justed to 75 ml with 0.1% peptone (ca. 5 x 107 CFU/ml)
before the four-strain mixture was added to sausage batter as
described below.
Manufacture of turkey summer sausage. Turkey summer
sausage was prepared from 90% hand-deboned turkey thigh
meat, 5% turkey hearts, and 5% turkey fat by using a
procedure modified from that of Wilson (41). Turkey com-
ponents were ground first through a 9.5-mm plate by using a
Hobart laboratory grinder (model 84142; Hobart Manufac-
turing Co., Troy, Ohio). Commercial turkey summer sau-
sage seasoning (0.5%, formulated to provide glucose [1%]
and NaCl [2.5%]; Milwaukee Seasonings, Inc., German-
town, Wis.) and NaNO2 (0.015%) were added and mixed
with 15-kg quantities of turkey meat for 3 min at 4°C in a
Hobart mixer (model 1661682). Three samples (ca. 100 g
each) were removed for analyses of uninoculated batter.
(i).
APPL. ENvIRON. MICROBIOL.

Page 3

DIFFERENTIATION OF PEDIOCOCCI
3055
Sausage batter was inoculated with 150 ml of pediococcal
starter culture (JBL1095 or JBL1350; ca. 10
batter) and mixed for 2 min at 4°C. A 7.5-kg portion of batter
was removed, ground through a 6.3-mm plate, stuffed (150 g
per chub) by using a hand stuffer (Koch Supplies, Inc.,
Kansas City, Mo.) into mahogany fibrous casings (75 by 152
mm; Vista International Packaging, Inc., Kenosha, Wis.),
and hand tied. To the remaining 7.5 kg of batter was added
75 ml of the four-strain L. monocytogenes mixture (ca. S x
105 CFU/g of batter), and the batter was mixed, reground,
and stuffed as described above. Sausages were fermented at
37°C (85% relative humidity) in a Vortron smokehouse
(model 1000; Vortron, Inc., Beloit, Wis.) to pH 5 (ca. 12 h),
cooked to an internal temperature of 66.5°C for 45 min, and
cold-showered to an internal temperature of 55°C. After
sausages were processed, individual chubs were vacuum-
packaged in gas-impermeable Curlon bags (nylon-Saran-
polyethylene; 02 transmission of 0.8 to 1.0 cm3/645 cm2/24 h
at 22.8°C, CO2 transmission of 2.5 to 3.0 cm3/645 cm2/24 h at
22.8°C, H2O transmission of 0.5 g/645 cm2/24 h at 37.8°C,
and 90% relative humidity; Curwood, Inc., New London,
Wis.) by using a Multivac AGW vacuum-packaging unit
(Sepp Haggemuller KG, Wolfertschwenden, Germany) and
then stored at 4°C (for up to 60 days) or 25°C (for up to 7
days).
Enumerating listeriae and pediococci from sausages and
determining pH and TA. Three sausage chubs from each
treatment (JBL1095 only, JBL1095 plus L. monocytogenes,
JBL1350 only, and JBL1350 plus L. monocytogenes) at each
sampling time were tested for (i) L. monocytogenes count by
direct plating and by enrichment (at the end of cook/shower
and subsequent sampling times), (ii) Pediococcus count by
direct plating, (iii) pH, and (iv) titratable acidity (TA) as
percent lactic acid. Sausage samples were taken immediately
after stuffing (time 0), during fermentation (e.g., at 4 and 8 h
after the start and at the end), at the end of the cook/shower,
and during storage at 4°C (at 1, 7, and 60 days) or 25°C (at 1,
3, 5, and 7 days). In addition, uninoculated sausage batter
samples were analyzed to determine the presence of indige-
nous L. monocytogenes and lactic acid bacteria.
To obtain sausage samples, casings were wiped with 70%
ethanol and cut with a sterile scalpel to facilitate removal of
casings. For each sample, three 25-g portions were asepti-
cally removed to separate stomacher bags for direct plating,
enrichment, and pH and TA determinations. For determin-
ing bacterium numbers, each 25-g portion was macerated
with 225 ml of 0.1% peptone water for 2 min by using a
stomacher (model 400; Tekmar Co., Cincinnati, Ohio) and
serially diluted (1:10) for direct plate counts. As described
previously (lOa, 42), pediococci were enumerated on MRS
agar (JBL1350; not genetically marked) or MRS agar plus
1,000 ,ug of streptomycin per ml and 100jigof rifamycin per
ml (JBL1095; Strr Rif) and incubated for 24 to 48 h at 30°C.
L. monocytogenes organisms were enumerated on a selec-
tive medium (modified Oxford agar) and confirmed essen-
tially as described previously (17).
The pH and TA were determined by a procedure modified
from that of Sebranek (37). In brief, a 25-g portion of sausage
was macerated in a stomacher with 225 ml of hot (ca. 70°C)
distilled, deionized water for 2 min. The homogenate was
poured into a 250-ml Erlenmeyer flask and cooled to room
temperature, and the fat layer was removed with the aid of a
pipet. The pH of the mixture was measured with a combi-
nation electrode and a Corning model 140 pH meter (Corning
Glass Works, Coming, N.Y.). The homogenate was filtered
through Whatman no. 1 filter paper, and the filtrate (100 ml)
CFU/g of
A
B
C
D
E
F
kb
16.2 -
10.1-
8.1-
4.0-
- 34.5
- chr
-9.4
FIG. 1. Comparison of plasmid profiles of Pediococcus strains.
Lanes: A, supercoiled ladder molecular weight size standard (Bio-
Rad); B, LB42; C, JBL1350; D, JBL1095; E, PAC1.0; F, P02. The
plasmid profiles shown are identical to profiles of each strain
obtained in at least nine replicate gels. chr, chromosomal DNA.
was titrated with 0.1 N NaOH to pH 8.1. The TA was
expressed as the percent lactic acid.
Monitoring pediocin AcH activity in sausages. At appropri-
ate time intervals, 10-g samples of sausage were removed
from chubs as described above, combined with 90 ml of
sterile H20, and blended to homogeneity (ca. 2 min) by using
the highest setting on a commercial blender (model 32BL97;
Waring Products Division, New Hartford, Conn.). Next, the
sausage-H20 blend was centrifuged (9,820 x g, 5 min, 4°C),
and the resulting supernatant (ca. 80 ml) was transferred to a
fresh tube and heated (100°C, 10 min) to eliminate residual
bacteria. Pediocin activity was then concentrated (80 ml
reduced to ca. 5 ml) by lyophilization. Antilisterial activity
was determined by the spot-on-lawn method and expressed
as arbitrary units per gram of sausage essentially as de-
scribed previously (lOa).
RESULTS
Preliminary characterization of pediococcal starter cul-
tures. For application to this study, experiments were con-
ducted to confirm production of pediocin AcH by JBL1095
and to determine whether JBL1350 produced or was sensi-
tive to already characterized pediocins. Strain JBL1095, a
derivative of P. acidilactici H (33), and the associated
pediocin AcH (5) have been at least partially characterized,
but essentially no information is available for JBL1350. In
this study, JBL1350 did not exhibit antimicrobial activity
related to the production of a pediocin (data not shown). On
the basis of these results, JBL1095 and JBL1350 were
employed as Ped+ and Ped- starter cultures, respectively, to
manufacture turkey summer sausage.
Plasmid analysis. Plasmid DNA was extracted from Ped+
(JBL1095) and Ped- (JBL1350) starter cultures and from
three other (PAC1.0, P02, and LB42) strains ofP. acidilac-
VOL. 58, 1992

Page 4

3056LUCHANSKY ET AL.
A B C
D
E
fF G
kb
-533
291
-242
145
-.97
48
FIG.
2. Comparison
of CHEF gel electrophoresis patterns
of
Ascl digests ofgenomicDNA from Pediococcus strains. Lanes: A,
Saccharomyces cerevisiae chromosome size standard (Bio-Rad); B,
LB42; C, JBL1350; D, JBL1095; E, PAC1L0; F, P02; G, lambda
ladder size standard (Bio-Rad). The genomic fingerprints shown are
identical to restrictionprofilesof each strain obtained in at least five
replicate gels.
tici (Fig. 1). Heretofore, the plasmid profiles of these strains
have not been directly compared (i.e.,
Plasmid DNA was not observed in the Ped-, plasmid-free
control strain, LB42 (lane B), butplasmidswere discernible
in all other strains screened (lanes C to F). In addition to at
least one large (>30 kb) plasmid, P.
PAC1L0, and P02 contained small plasmids (one each) of
similar size (ca. 9.4kb). Strain JBL1350 did not contain this
small (ca. 9.4-kb) plasmid, but at least one high-molecular-
weight band was evident (Fig. 1). These results established
that
on the
same gel).
acidilactici JBL1095,
strains JBL1095, PACLO0,
similar plasmid profiles. Further work is required to deter-
mine whether
the
resident plasmids
identical.
and P02
possess
highly
in
thesestrains
are
Genomicfingerprintingofpediococci by using CHEF. Since
Pediococcus spp. have relatively low G+C contents (ca. 37
to 44%), and on the basis of encouraging results with L.
monocytogenes
(ca.
36
to 38% G+C [22]),
low-frequency-cleavage endonuclease AscI (GGCGCGCC)
for CHEF analyses
of pediococci
genomic DNA from LB42 produced nine AscI
ranging in length from about 84 to 950 kb. CHEF analyses
revealed identical restrictionprofiles (10AscIfragments; ca.
62 to 525 kb) for P. acidilactici JBL1095, PAC1L0, P02, and
JBL1350. The sizes and numbers offragments generated by
digestion of intact pediococcal DNA with AscI are listed in
Table 2. Depending on the strain, summation of results with
AscI fragments estimated the genome sizes to be 1,861 kb
(JBL1095,
PAC1.0,
P02,
(LB42).
Control ofL. monocytogenes during turkey summer sausage
fermentation. Sausages preparedwith a Ped-
culture were compared for the ability to support the growth
of L.
monocytogenes during manufacture and storage
(Fig.3).
Levels of pediococcal
we usedthe
(Fig.
2). Digestion
of
fragments
and JBL1350)
and
2,133
kb
or Ped' starter
of
sausages
starter
cultures
APPL. ENVIRON. MICROBIOL.
TABLE 2. Restriction analysis of Pediococcus genomic DNA
Size (kb) ofAscI fragments from strainsa
Fragment
PAC1.0, P02,JBL1095,
andJBL1350
JBL1146
1
2
3
4
5
6
7
8
9
10
950.5 ± 29.5
212.5 ± 2.5
200 ± 5
178.5 ± 21.5
170 ± 20
127 ± 2
107.5 ± 2.5
102.5 ± 2.5
84±10
525 ± 35
288 ± 3
207.5 ± 2.5
180 ± 5
171.5 ± 1.5
147.5 ± 2.5
105 ± 0
91 ± 3
83±3
62.5 ± 2.5
1,861 + 5.8
Total
2,133 -
10.6
aIndividual and total fragment sizes were calculated as an average from at
least three different gels for fragments of <550 kb and one gel for fragments of
>600 kb.
decreased by the end of fermentation; pediococci were not
recovered from properly cooked sausages (data not shown).
In sausages fermented with a Ped- strain, counts of L.
monocytogenes decreased about 0.9 log10 units during the
12-h fermentation period. In contrast, counts of the pathogen
were reduced by 3.4 log10 units during fermentation of
sausages prepared with a Ped+ pediococcal starter culture.
Regardless of the choice of starter culture, no listeriae were
recovered from sausages following the cook/shower process
by either direct plating or the enrichment procedure.
Production of acid and pediocin during fermentation of
turkey summer sausage. The use of a Ped+ starter culture
rather than a Ped- strain resulted in a greater reduction in
counts of L. monocytogenes during fermentation. The pH,
TA, and pediocin activity of sausages prepared with
JBL1095 or JBL1350 were monitored to determine whether
antilisterial activity was due to acid and/or pediocin produc-
tion. The data revealed that the pH was reduced to similar
levels (reduced from an average pH of 6.3 to 5.0) at equiv-
alent rates (ca. 12 h) in sausage prepared with either starter
culture (Table 3). Likewise, the TA (expressed as percent
lactic acid) increased to the same extent in sausage prepared
with either pediococcal starter culture. More important, as
the fermentation proceeded, counts of JBL1095 decreased
by about 1.1 logl0 unit while the titer of pediocin activity
increased dramatically (from 0 to 5,000 arbitrary units per
gram of sausage in 12 h) in sausage prepared with JBL1095
(Fig. 4). Moreover, after the cook/shower, pediococci were
not detectable (<102 CFU/g) whereas pediocin activity was
maintained at high levels (ca. 5,000 arbitrary units per gram
of sausage) for 60 days at 4°C. No pediocin activity was
recovered from sausages prepared with JBL1350. These data
confirm that additional antilisterial activity (i.e., pediocin
AcH) was available in sausage prepared with a Ped+ starter
culture and that pediocin AcH was not affected by fermen-
tation, cook/shower, or storage of turkey summer sausage.
DISCUSSION
L. monocytogenes has been implicated in food-related
listeriosis outbreaks involving dairy products and coleslaw,
and the pathogen is common in meat and poultry products.
The frequent association of L. monocytogenes with fresh,
frozen, and ready-to-eat poultry (up to 60% [2, 13, 14, 23,

Page 5

DIFFERENTIATION OF PEDIOCOCCI
3057
5'
0)
:
n
4-
1-
.
3-
0
2
4
6
8
10
12
Time (h)
A
FIG. 3. Behavior of L. monocytogenes in the presence of pedi-
ococci during fermentation of turkey summer sausage (trial 2). Open
rectangles represent the four-strain L. monocytogenes mixture in
the presence of P. acidilactici JBL1350; thick crosses represent the
four-strain L. monocytogenes mixture in the presence of P. acidi-
lactici JBL1095. "A" signals the end of fermentation.
31]) and the perceived risk from poultry-related listeric
illness prompted us to evaluate the use of pediocin-produc-
ing pediococci as an additional method to ensure product
safety. Also, published information intimating similarities
among various pediocins, pediocinogenic plasmids, and pe-
diococci prompted us to investigate genomic relationships
among characterized starter cultures.
Prior to the use of JBL1095 and JBL1350 in production of
turkey summer sausage, pediocin activity, plasmid content,
and genomic fingerprints of these strains were compared
with each other and with those of characterized P. acidilac-
tici strains. Previous studies (18, 21, 33) revealed that strains
H, PAC1.0, and P02 contained a small plasmid necessary
for pediocin production and an additional plasmid of about
34.5 kb. Our data do not confirm the presence of a third
plasmid as previously reported for strains P02 (127 MDa
[21]) and H (40 MDa [33]). Aside from the initial observation
of them, there have been no reports substantiating the
presence or function of these very large plasmids in strain
P02 or H. Moreover, it is possible that the ca. 40-MDa
plasmid reported for strain H represents the dimeric form of
the 23-MDa resident plasmid. Although further work is
warranted to unravel relationships among strains and plas-
mids, Hoover et al. (20) reported that the plasmid profiles of
strains PAC1.0 and P02 were similar and that the corre-
sponding pediocins possessed similar activities and physical
properties. Our results established that the large (ca. 34.5-
kb) and small (ca. 9.4-kb) plasmids present in strains PAC1.0
and P02 are equivalent in size to plasmids in strain JBL1095
(Fig. 1). It was also of interest that the Ped- commercial
starter culture JBL1350 contained a high-molecular-weight
(>35 kb) plasmid band but was missing the small pediocin
plasmid common among the other strains. The inability to
produce pediocin (i.e., absence of pediocin plasmid) was in
large measure the reason that JBL1350 was less effective
than JBL1095 in controlling L. monocytogenes during fer-
mentation.
Thus far, the technique of PFGE-CHEF has not been
employed to size or discriminate genomic DNA from Pedi-
ococcus spp. CHEF analysis of pediococci using the enzyme
AscI revealed identical genomic fingerprints for the Ped+
strains JBL1095, PAC1.O, and P02. Moreover, the genomic
fingerprint of a Ped- pediocin-resistant commercial starter
culture (JBL1350) was identical to the AscI profiles for
strains JBL1095, PAC1.O, and P02. The plasmid-free Ped-
pediocin-sensitive strain LB42 displayed an AscI migration
profile significantly different from those of the other strains
tested. Summation of results for individually sized AscI
fragments
from
strainsJBL1095,
JBL1350 and strain LB42 estimated the genome lengths to be
1,861 and 2,133 kb, respectively. These values are in agree-
ment with pulsed-field analysis determinations of the ge-
nome sizes of other lactic acid bacteria (range of 1,700 to
2,700 kb [25, 38]). Future efforts will be directed to analyze
more strains, evaluate additional rare cutting endonucleases,
and optimize conditions for enhanced resolution of high-
molecular-weight pediococcal DNA for molecular tracking
of strains and construction of macrorestriction maps.
In the last few years, there have been several reports on
the use of pediococci and pediocins to control L. monocy-
togenes in foods (3, 4, 10a, 12, 29, 32, 34, 42). Previous
studies established that pediocin AcH (added directly or
produced in situ by JBL1095) was a potent inhibitor of L.
monocytogenes associated with all-beef wieners (lOa, 42).
Degnan et al. (lOa) implemented genetically marked Ped+
and genetically related Ped- pediococci to compare the
antilisterial activities of organic acids and pediocins in ther-
mally processed meat. These investigators demonstrated
that L. monocytogenes survived but did not grow in temper-
ature-abused vacuum-packaged wieners stored at 25°C for 8
days in the presence of a Ped- strain, whereas counts of the
pathogen were reduced by an average of 2.7 log10 units in the
presence of a Ped+ (JBL1095) strain. More recently, a
similar approach was used to confirm the antilisterial role of
pediocin production during dry fermented sausage produc-
tion (12). The results of the present work reveal the added
safeguard of bacteriocin production by pediococcal starter
cultures for reducing the potential for listerial proliferation in
poultry. More specifically, the production of acid by a Ped-
starter culture produced primarily a listeriostatic effect in
turkey summer sausage, but a listericidal response was
associated with production of pediocin AcH by a Ped+
starter culture.
In summary, the results of this study, as well as other
work cited herein, suggest that some P. acidilactici starter
cultures and pediocin-encoding plasmids are genetically re-
lated. However, the pediocin immunity genes are chromo-
somal in strain PAC1.0 (18) and presumably in P02 and are
plasmid borne in strain H (parental strain of JBL1095 [33]).
Strain JBL1350 (Ped-) was phenotypically similar to plas-
mid-cured derivatives (Ped-) of PAC1.0 and P02 because it
remained resistant to pediocin despite the absence of the
small pediocin plasmid; the location of the JBL1350 immu-
nity/resistance gene(s) is presently unknown. Our data ex-
panded upon previous reports and revealed that P. acidilac-
tici strains JBL1095, PACLO, and P02, as well as JBL1350,
have identical AscI genomic fingerprints. However, neither
the amino acid sequences of the various pediocins nor the
nucleotide sequences of the corresponding genes have been
compared. These data argue strongly for additional studies
to substantiate common parentage among strains. Our re-
sults also established that not all commercially available
starter cultures encode pediocins or are equally effective in
controlling L. monocytogenes. The fermentation of turkey
sausage with a Ped+ starter culture resulted in a dramatic
PAC1.O,
P02,
and
VOL. 58, 1992
1-b

Page 6

3058
LUCHANSKY ET AL.
TABLE 3. Changes in pH and TA during manufacture and storage of turkey summer sausage inoculated with L. monocytogenes
and/or P. acidilactici JBL1350 or JBL1095a
Result for sausage inoculated with:
JBL1350 + L. monocytogenes
JBL1350
JBL1095 + L. monocytogenes
JBL1095
Time
Trial 1
Trial 2Trial 1
Trial 2Trial 1
Trial 2Trial 1Trial 2
pH
TApH TA pHTA
pH
TApHTApH TApH TApH
TA
0b
During manu-
facture (h)
0
4
8
12
After shower
During storage
(days)
At 4°C
1
7
60
At 250C
1
3
5
7
6.14
0.46
6.44
0.436.14
0.46
6.440.43
6.31
0.48 6.410.44 6.310.48
6.41
0.44
6.14
5.72
4.94
0.47
0.57
0.81
6.48
6.22
5.75
4.94
5.07
0.43
0.48
0.58
0.82
0.81
6.11
5.82
4.93
-
4.86
0.47
0.55
0.80
-
0.80
6.49
6.21
5.70
4.99
5.08
0.43
0.46
0.58
0.79
0.81
6.39
6.17
5.53
5.14
5.17
0.42
0.48
0.60
0.68
0.75
6.41
5.99
5.41
4.98
4.97
0.41
0.51
0.60
0.78
0.76
6.39
6.19
5.48
5.07
5.13
0.41
0.51
0.59
0.73
0.75
6.35
6.00
5.44
5.02
5.02
0.44
0.52
0.60
0.77
0.77
c
4.910.80
4.94
4.97
5.05
0.78
0.94
0.88
5.07
5.13
5.20
0.85
0.88
0.94
4.97
4.96
5.02
0.77
0.93
0.90
5.13
5.14
5.21
0.84
0.89
0.96
5.18
5.17
-
0.76
0.77
-
4.96
5.04
5.12
0.79
0.90
0.95
5.13
5.11
5.25
0.78
0.79
0.82
5.02
5.20
5.17
0.77
0.78
0.91
4.99
5.10
5.02
4.94
0.81
0.79
0.85
0.92
5.070.84 4.97
4.94
4.97
4.94
0.78
0.92
0.87
0.94
5.10
-
-
5.19
0.82
-
5.20
5.18
5.11
5.12
0.80
0.77
0.88
0.91
4.98
-
0.79
-
5.14
5.14
5.06
5.01
0.80
0.76
0.90
0.92
5.02
-
-
5.20
0.78
5.13
0.95
0.94 5.070.86
0.84
aPrior to stuffing, sausage batter was inoculated with a four-strain mixture of L. monocytogenes and/or JBL1350 (Ped-) or JBL1095 (Ped+).
bValues at time zero are initial values.
c_, not determined.
reduction in counts of L. monocytogenes during fermenta-
tion. More important, the recovery of pediocin AcH activity
after 60 days of storage at 4°C established another hurdle for
L. monocytogenes in the event of insufficient processing
and/or postprocess contamination. Future efforts to con-
struct and genetically characterize bacteriocinogenic pedio-
cocci for use as starter cultures will lead to strains with
O
I
I
v
MI
ax)
n)
z
Time (h)
# t
AB
FIG. 4. Behavior of Ped+ pediococci (average of two trials) and
pediocin activity (duplicate sampling of trial 1) in turkey summer
sausage. Open rectangles represent pediocin AcH produced during
fermentation of sausage with JBL1095; thick crosses represent
JBL1095 during fermentation of sausage. "A" signals the end of
fermentation, and "B" signals the end of the cooking step.
enhanced capabilities for ensuring the safety and extending
the shelf life of foods.
ACKNOWLEDGMENTS
The technical assistance of Timothy Harried, Jodi Loeffelholz,
Ariane Nara, and Corinne Johnson is greatly appreciated. Also, we
thank Chuck Baum (UW Biotron), Lou Arrington (UW Poultry
Science), Carol Ayres (Food Research Institute), Vista International
Packaging, and Milwaukee Seasonings for providing supplies,
equipment, and/or expertise.
This project was supported in part by Hatch funds (WIS 3360 and
WIS 3444); grant 91-03814 from the National Research Initiative
Competitive Grants Program of the U.S. Department ofAgriculture;
the Beef Industry Council of the National Livestock and Meat
Board; the College of Agricultural and Life Sciences, University of
Wisconsin-Madison; and contributions to the Food Research Insti-
tute. Also, Gail Baccus-Taylor was supported by a scholarship from
the Latin American Scholarship Program of American Universities.
REFERENCES
1. Anonymous. 1989. Listeriosis associated with consumption of
turkey franks. Morbid. Mortal. Weekly Rep. 38:267-268.
2. Bailey, J. S., D. L. Fletcher, and N. A. Cox. 1989. Recovery and
serotype distribution of Listeria monocytogenes from broiler
chickens in the southeastern United States. J. Food Prot.
52:148-150.
3. Berry, E. D., R. W. Hutkins, and R. W. Mandigo. 1991. The use
of bacteriocin-producing Pediococcus acidilactici to control
postprocessing Listeria monocytogenes contamination of frank-
furters. J. Food Prot. 54:681-686.
4. Berry, E. D., M. B. Liewen, R. W. Mandigo, and R. W. Hutkins.
1990. Inhibition of Listeria monocytogenes by bacteriocin-
producing Pediococcus during the manufacture of fermented
semidry sausage. J. Food Prot. 53:194-197.
5. Bhunia, A. K., M. C. Johnson, B. Ray, and N. Kalchayanand.
APPL. ENvIRON. MICROBIOL.

Page 7

DIFFERENTIATION OF PEDIOCOCCI3059
1991. Mode of action of pediocin AcH from Pediococcus
acidilactici H on sensitive bacterial strains. J. Appl. Bacteriol.
70:25-33.
6. Brosch, R., C. Buchrieser, and J. Rocourt. 1991. Subtyping of
Listeria monocytogenes serovar 4b by use of low-frequency-
cleavage restriction endonucleases and pulsed-field gel electro-
phoresis. Res. Microbiol. 142:667-675.
7. Canard, B., and S. T. Cole. 1989. Genome organization of the
anaerobic pathogen Clostridiumperfringens. Proc. Natl. Acad.
Sci. USA 86:6676-6680.
8. Carriere, C., A. Allardet-Servent, G. Bourg, A. Audurier, and
M. Ramuz. 1991. DNA polymorphism in strains of Listeria
monocytogenes. J. Clin. Microbiol. 29:1351-1355.
9. Daeschel, M. A., and T. R. Klaenhammer. 1985. Association of
a 13.6-megadalton plasmid in Pediococcus pentosaceus with
bacteriocin activity. Appl. Environ. Microbiol. 50:1538-1541.
10. Degnan, A. J. Unpublished data.
10a.Degnan, A. J., A. E. Yousef, and J. B. Luchansky. 1992. Use of
Pediococcus acidilactici to control Listeria monocytogenes in
temperature-abused, vacuum-packaged wieners. J. Food Prot.
55:98-103.
11. Farber, J. M., and P. I. Peterkin. 1991. Listeria monocytogenes,
a food-borne pathogen. Microbiol. Rev. 55:476-511.
12. Foegeding, P. M., A. B. Thomas, D. H. Pilldngton, and T. R.
Kiaenhammer. 1992. Enhanced control of Listeria monocyto-
genes by in situ-produced pediocin during dry fermented sau-
sage production. Appl. Environ. Microbiol. 58:884-890.
13. Genigeorgis, C. A., D. Dutulescu, and J. F. Garayzabal. 1989.
Prevalence of Listeria spp. in poultry meat at the supermarket
and slaughterhouse level. J. Food Prot. 52:618-624.
14. Genigeorgis, C. A., P. Oanca, and D. Dutulescu. 1990. Preva-
lence of Listeria spp. in turkey meat at the supermarket and
slaughterhouse level. J. Food Prot. 53:282-288.
15. Glass, K. A., and M. P. Doyle. 1989. Fate and thermal inactiva-
tion of Listeria monocytogenes in beaker sausage and pepper-
oni. J. Food Prot. 52:226-231.
16. Glass, K. A., and M. P. Doyle. 1989. Fate of Listeria monocy-
togenes in processed meat products during refrigerated storage.
Appl. Environ. Microbiol. 55:1565-1569.
17. Glass, K. A., and M. P. Doyle. 1991. Fate of Salmonella and
Listeria monocytogenes in commercial, reduced-calorie mayon-
naise. J. Food Prot. 54:691-695.
18. Gonzalez, C. F., and B. S. Kunka. 1987. Plasmid-associated
bacteriocin production and sucrose fermentation in Pediococ-
cus acidilactici. Appl. Environ. Microbiol. 53:2534-2538.
19. Graham, D. C., and L. L. McKay. 1985. Plasmid DNA in strains
of Pediococcus cerevisiae and Pediococcus pentosaceus. Appl.
Environ. Microbiol. 50:532-534.
20. Hoover, D. G., K. J. Kishart, and M. A. Hermes. 1989. Antag-
onistic effect of Pediococcus spp. against Listeria monocyto-
genes. Food Biotechnol. 3:183-196.
21. Hoover, D. G., P. M. Walsh, K. M. Kolaetis, and M. M. Daly.
1988. A bacteriocin produced by Pediococcus species associ-
ated with a 5.5-megadalton plasmid. J. Food Prot. 51:29-31.
22. Howard, P. J., K. D. Harsono, and J. B. Luchansky. 1992.
Differentiation of Listeria monocytogenes, Listeria innocua,
Listeria ivanovii, and Listena seeligeri by pulsed-field gel elec-
trophoresis. Appl. Environ. Microbiol. 58:709-712.
23. Hudson, W. R., and G. C. Mead. 1989. Listeria contamination at
a poultry processing plant. Lett. Appl. Microbiol. 9:211-214.
24. Ingham, S. C., and C. L. Tautorus. 1991. Survival ofSalmonella
typhimurium, Listeria monocytogenes and indicator bacteria on
cooked uncured turkey loaf stored under vacuum at 3°C. J.
Food Saf. 11:285-292.
25. Le Bourgeois, P., M. Mata, and P. Ritzenthaler. 1989. Genome
comparison of Lactococcus strains by pulsed-field gel electro-
phoresis. FEMS Microbiol. Lett. 59:65-70.
26. Luchansky, J. B., A. K. Benson, and A. G. Atherly. 1989.
Construction, transfer and properties of a novel temperature-
sensitive integrable plasmid for genomic analysis of Staphylo-
coccus aureus. Mol. Microbiol. 3:65-78.
27. McKay, L. L., and K. A. Baldwin. 1990. Applications for
biotechnology: present and future improvements in lactic acid
bacteria. FEMS Microbiol. Rev. 87:3-14.
28. Muriana, P. M., and T. R. Klaenhammer.
phenotypic expression, and DNA sequence of the gene for
lactacin F, an antimicrobial peptide produced by Lactobacillus
spp. J. Bacteriol. 173:1779-1788.
29. Nielsen, J. W., J. S. Dickson, and J. D. Crouse. 1990. Use of a
bacteriocin produced by Pediococcus acidilactici to inhibit
Listeria monocytogenes associated with fresh meat. Appl. En-
viron. Microbiol. 56:2142-2145.
30. Pattee, P. A. 1990. Genetic and physical mapping of the chro-
mosome of Staphylococcus aureus NCTC 8325, p. 163-169. In
K. Drlica and M. Riley (ed.), The bacterial chromosome.
American Society for Microbiology, Washington, D.C.
31. Pini, P. N., and R. J. Gilbert. 1988. The occurrence in the UK
ofListeria species in raw chickens and soft cheeses. Int. J. Food
Microbiol. 6:317-326.
32. Pucci, M. J., E. R. Vedamuthu, B. S. Kunka, and P. A.
Vandenbergh. 1988. Inhibition of Listeria monocytogenes by
using bacteriocin PA-1 produced by Pediococcus acidilactici
PAC 1.0. Appl. Environ. Microbiol. 54:2349-2353.
33. Ray, S. K., W. J. Kim, M. C. Johnson, and B. Ray. 1989.
Conjugal transfer of a plasmid encoding bacteriocin production
and immunity in Pediococcus acidilactici H. J. Appl. Bacteriol.
66:393-399.
34. Sabel, D., A. E. Yousef, and E. H. Marth. 1991. Behavior of
Listeria monocytogenes during fermentation of beaker sausage
made with or without a starter culture and antioxidant food
additives. Lebensm. Wiss. Technol. 24:252-255.
35. Schillinger, U., and F.-K. Lucke. 1989. Antibacterial activity of
Lactobacillus sake isolated from meat. Appl. Environ. Micro-
biol. 55:1901-1906.
36. Schwartz, B., C. V. Broome, G. R. Brown, A. W. Hightower,
C. A. Ciesielski, S. Gaventa, B. G. Gellin, and L. Mascola. 1988.
Association of sporadic listeriosis with consumption of un-
cooked hot dogs and undercooked chicken. Lancet ii:779-782.
37. Sebranek, J. 1978. Meat science and processing, p. 139-140.
Paladin House Publishers, Geneva, Ill.
38. Tanskanen, E. I., D. L. Tulloch, A. J. Hillier, and B. E.
Davidson. 1990. Pulsed-field gel electrophoresis of SmaI digests
of lactococcal genomic DNA, a novel method of strain identifi-
cation. Appl. Environ. Microbiol. 56:3105-3111.
39. Vanrenterghem, B., F. Huysman, R. Rygole, and W. Verstraete.
1991. Detection and prevalence of Listeria monocytogenes in
the agricultural ecosystem. J. Appl. Bacteriol. 71:211-217.
40. Wenger, J. D., B. Swaminathan, P. S. Hayes, S. S. Green, M.
Pratt, R. W. Pinner, A. Schuchat, and C. V. Broome. 1990.
Listeria monocytogenes contamination of turkey franks: evalu-
ation of a production facility. J. Food Prot. 53:1015-1019.
41. Wilson, G. D. 1960. Sausage products, p. 349-372. In G. W.
Salisbury and E. W. Crampton (ed.), The science of meat and
meat products. W. H. Freeman and Co., San Francisco.
42. Yousef, A. E., J. B. Luchansky, A. J. Degnan, and M. P. Doyle.
1991. Behavior ofListeria monocytogenes in wiener exudates in
the presence of Pediococcus acidilactici H or pediocin AcH
during storage at 4 or 25'C. Appl. Environ. Microbiol. 57:1461-
1467.
1991. Cloning,
VOL. 58, 1992