Scientific
Publications - Work Done by Microbiology Reader Bioscreen C
Letters in Applied
Microbiology 1998, 26,
231-235
Plantaricin D, a bacteriocin produced by Lactobacillus
plantarum BFE
905 from ready-to-eat salad
C.M.A.P. Franz, M. Du Toit, N.A. Olasupo, U.
Schillinger and W.H. Holzapfel
Federal Research Center for Nutrition, Institute of Hygiene
and Toxicology, Karlsruhe, Germany
1648/97: received 22 September 1997 and accepted 13
November 1997
C.M.A.P. FRANZ, M. DU TOIT, N.A. OLASUPO, U. SCHILLINGER AND
W.H. HOLZAPFEL. 1998.
Lactobacillus plantarum BFE 905 isolated
from `Waldorf' salad produced a bacteriocin termed
plantaricin D which was active
against Lact. sake and Listeria
monocytogenes strains.
Plantaricin D was heat stable, retaining activity after
heating at 121 °C. The
bacteriocin was inactivated by a-chymotrypsin,
trypsin, pepsin and proteinase K, but
not by papain and other non-proteolytic enzymes
tested. Plantaricin D was stable
at pH values ranging from 2.0 to 10.0. The bacteriocin
inhibited growth of L.
monocytogenes in
automated turbidity assays. Although Lact. plantarum BFE
905
harboured plasmids ranging in size from 3 to 55 kilobase pairs, loss of
bacteriocin
production could not be correlated with plasmid loss. A role for
bacteriocin-producing Lact.
plantarum of
vegetable origin in assuring the safety of vegetable foods is suggested.
INTRODUCTION
for use as biopreservatives
in vegetable products such as
prepared or ready-to-eat salads. These products are
known The
lactic acid bacteria (LAB) are involved in fermentations
to be contaminated with
food-borne pathogens, and have
of foods such as dairy and bakery products,
meat and
served as vehicles for outbreaks of food poisoning by bacteria
vegetables, and a variety of
strains are routinely used as starter
such as Listeria monocytogenes, Bacillus
cereus and Clostridium
cultures. Lactic acid bacteria
improve the shelf-life and safety
botulinum
(Beuchat 1995). Naturally-occurring bac-
of fermented foods by inhibiting
the growth of spoilage and
teriocinogenic LAB could be considered for use in the bio-
pathogenic bacteria, by
competing for nutrients and pro-
preservation of vegetable products. This paper
describes the
ducing antimicrobial compounds such as organic acids, etha-
isolation and identiflcation of
a Lact. plantarum strain from a
nol, carbon dioxide, hydrogen
peroxide and bacteriocins
ready-to-eat salad and the characterization of its
bacteriocin.
(Hammes and Tichaczek 1994; Abee et al.
1995). Bacteriocins
are proteinaceous substances with bactericidal or
bac-
teriostatic activity against sensitive bacterial species (for
MATERIALS AND METHODS
reviews see Klaenhammer 1993;
DeVuyst and Vandamme
1994; Nes et al. 1996). The use
of bacteriocins or bacteriocin-
Strain isolation and screening for bacteriocin
production
producing cultures as potential
`biopreservatives', and pos-
sibly for replacing chemical preservation (Abee et
al. 1995),
Lactic acid bacteria strains were isolated from raw and
fer- has
received much attention.
mented vegetables and ready-to-eat salads by
methods
Vegetable fermentations usually rely on the adventitious
described previously (Franz et
al. 1997), and purifled by
flora as starter bacteria and Lactobacillus
plantarum strains
streaking out on de Man, Rogosa and Sharpe (MRS) agar
often predominate in these
fermentations (Stiles and Hol-
(Merck, Darmstadt, Germany). Gram-positive, catalase-
zapfel 1997). Bacteriocinogenic Lact.
plantarum strains have
negative isolates were identifled as presumptive LAB and
been proposed to achieve more
reliable fermentations of con-
screened for antagonistic activity using the agar
spot test
sistent quality (JimeÁnez-Diaz et al.
1993; Ruiz-Barba et al.
method (Franz et
al.1997). Isolates with antagonistic activity
1994). Bacteriocin-producing LAB
may also be considered
against one of the indicator strains used (Lact.
sake DSM
20017, Leuconostoc mesenteroides DSM20343, Lact.
plantarum
Correspondence to: Dr U. Schillinger, Federal Research
Center for Nutrition,
DSM 20174 and Enterococcus faecalis DSM
20380) were kept
Institute of Hygiene andToxicology, Engesserstr. 20, 76131
Karlsruhe, Germany
(e-mail: ulrich.schillinger@bfe.uni-karlsruhe.de).
for identiflcation and
bacteriocin characterization.
© 1998 The Society for Applied Microbiology
232 C.M.A.P.
FRANZ ET AL.
Table 1 The
effect of enzymes, pH and heat treatment on
Identiflcation of bacteriocin
producing isolates
Lactobacillus plantarum BFE 905
cell-free supernatant fluid
Growth of isolates at 6, 10, 15 and 45 °C was
determined in
Ð--------------------------------------------------- --
MRS broth (Merck). Production of
gas from glucose, arginine
Treatment
Activity (AU ml-1)*
hydrolysis, fermentation of
carbohydrates, the conflguration
Ð---------------------------------------------------
-- Untreated
800
of lactic acid enantiomer
produced and the presence of dia-
Enzymes:
minopimelic acid in the cell wall was
determined by the
a-chymotrypsin (Serva, no. 17160)
0
methods of Schillinger and LuÈcke
(1987). In addition, isolates
trypsin (Sigma, no. T-8253)
0
were identifled by total soluble
cell protein proflle analysis,
pepsin (Merck, no. 7189)
0
using previously described
methods (Franz et al. 1997) and
papain (Merck, no. 7147)
800
Lact. plantarum DSM
20174 and DSM 20205 as reference
proteinase K (Sigma, no. P-0390)
0
strains.
a-amylase
(Sigma, type VIIA)
800
lysozyme (Serva, no. 28262)
800
catalase (Sigma, no. C-10)
800
Activity assay and spectrum
lipase (Sigma, no. L-1754)
800
Heat:
To determine the spectrum of
activity, a wide range of LAB
5 min at 100 °C
800
as well as L. monocytogenes, Staphylococcus
aureus and Bacillus
10 min at 100 °C
800
cereus strains
were used as indicators in agar spot tests.
20 min at 100 °C
800
Activity was assayed by using the
critical dilution assay (Franz
15 min at 121 °C
400
et al. 1997). For this, cell-free,
neutralized culture super-
pH:
natant fluid (CFNS) was prepared as follows. Supernatant
2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0
400
fluid of an 18 h culture
grown aerobically at 30 °C was
9.0, 10.0
200
obtained by centrifugation at 10 000 g and
the pH was
Ð--------------------------------------------------- --
adjusted to 6.5-7.0 using 1 N
sodium hydroxide. The super-
*AU ml-1 determined
by the critical dilution method with
natant FLuid was then heated at 100 °C for 5 min and
used in
Lactobacillus sake DSM 20017 as indicator.
critical dilution assays as
described previously (Franz et al.
1996).
lated into 190 ml
STD 1 broth without CFNS. All inocu-
lations were done in triplicate and growth of Listeria strains
Effect of heat, enzymes and pH on
bacteriocin
was monitored using the BIOSCREEN at 30 °C for 24 h.
activity
The effects of heat, enzymes and
pH on bacteriocin activity
Isolation of plasmid DNA and plasmid curing
were determined as described in
Franz et al. (1997). Briefly,
Small-scale plasmid DNA isolation
was done by the method
the effect of heat on CFNS was assessed at temperatures of
of Keppler et
al. (1991). Plasmids of known sizes were isolated
100 °C for 5, 10 and 20 min and
at 121 °C for 15 min. The
from Lactococcus lactis DSM
4645 by the same method,
effect of pH was tested on cell-free supernatant fluid
adjusted and
used as size markers. Plasmid DNA was subjected to
to a pH ranging from 2.0 to 10.0,
and the effect of enzymes
electrophoresis on 0.7% agarose gels at 100 V
(Sambrook
listed in Table 1 on supernatant fluid activity was determined
et al. 1989).
Curing of plasmids was attempted by treatment
at 1 mg/ml
1flnal
concentration (Franz et al. 1996, 1997).
of Lact.
plantarum BFE905 with novobiocin or SDS at various
concentrations. In addition,
plasmid curing was attempted by
Growth inhibition of Listeria
repeated subculturing of Lact.
plantarum BFE 905 at 43 °C.
Loss of bacteriocin phenotype was determined by
plating The
effect of the Lact. plantarum BFE 905 bacteriocin
on suitable
dilutions of treated cells onto MRS agar, and over-
growth of four L.
monocytogenes and one L. welshimeri strains
laying with the sensitive
indicator Lact. sake DSM 20017.
was assessed using the
BIOSCREEN C (Labsystems,
Bacteriocin-negative mutants were isolated and their total
Helsinki, Finland) automated
turbidometer as described by
soluble protein proflle was analysed, together with that of
the Holck et
al. (1996). Listeria strains were grown in
Standard
Lact. plantarum BFE905 parent strain, to conflrm the
identity
One Nutrient broth (STD 1) (Merck) for 18 h at 30 °C. Each
of the mutant strain as Lact.
plantarum.
culture was diluted 10-fold and 10 µl
were used to inoculate
180 ml of STD 1 broth containing
10 ml of Lact. plantarum
RESULTS
BFE 905 CFNS (1:20 dilution) in
honeycomb wells. Tenfold
diluted Lact. plantarum BFE 905
CFNS was used in a similar
An isolate obtained from a ready-to-eat `Waldorf'
salad
showed antagonistic activity against Lact. sake DSM
20017 way
(1:200 dilution). As a control, Listeria strains were
inocu- ©
1998 The Society for Applied Microbiology,Letters in
Applied Microbiology 26, 231-235
THE BACTERIOCIN PLANTARICIN D
233
and Lact.
plantarum DSM 20174 in the agar spot test. This
isolate consisted of
Gram-positive, catalase-negative rods that
did not produce gas in glucose
fermentation tests. The isolate
did not hydrolyse arginine and contained diaminopimelic acid
in the cell wall. The isolate grew
at 6 and10 °C, but only weak
growth was observed at 45 °C. The isolate produced both the
D (-)
and L (
) lactic acid enantiomers. Based on these
characteristics, as well as on the
carbohydrate fermentation
pattern (results not shown), the isolate was presumptively
identifled as Lact.
plantarum. This presumptive identiflcation
was conflrmed by total soluble
cell protein analysis, which
showed that isolate BFE 905 had a high degree of similarity
Fig. 1 Inhibitory
effect of Lactobacillus plantarum BFE
905 cell
with Lact. plantarum DSM 20205
(79.5%) and Lact. plan-
free, neutralized supernatant
fluid on Listeria monocytogenes 125.
tarum DSM
20174 (85.1%).
( ), Control; (E),
bacteriocin (1:200 dilution); (R), bacteriocin
The bacteriocin produced by Lact.
plantarum BFE 905
(1:20 dilution)
culture has a narrow activity spectrum. Among
the LAB
tested (including leuconostocs, lactobacilli and enterococci),
it inhibited only Lact.
sake DSM 20017 and Lact. plantarum
DSM 20174. The CFNS did, however,
also show antagonistic
activity towards two of three L.
monocytogenes strains tested.
Bacillus cereus and Staph.
aureus strains were not inhibited.
Activity of CFNS was at a maximum
of 800 AU/ml
1against
Lact. sake DSM
20017. This was relatively low when com-
pared to the activity of CFNS
from bacteriocin-producing
Ent. faecium and Lactococcus
lactis strains which were 6400
AU/ ml
1when
tested against the same indicator (Franz et al.
1996, 1997). When neutralized
supernatant fluid (800 AU
ml
1against Lact.
sake DSM 20017) was spotted onto a lawn
of the Lact.
plantarum BFE 905 producer organism itself, no
zone of clearing could be
detected.
The effect of heat, pH and enzymes on Lact. plantarum
BFE 905 culture supernatant fluid
is shown in Table 1. The
inhibitory compound was considered to be heat stable
as activity
remained after heating at 121 °C for 15 min. It was
inactivated by the proteases a-chymotrypsin,
trypsin, pepsin
and proteinase K, but not papain, which indicated that the
compound had a proteinaceous
nature. The inhibitory com-
pound was not sensitive to treatment with a-amylase,
lipase, Fig.
2 Plasmid proflles of Lactobacillus
plantarum BFE 905.
catalase or lysozyme (Table 1).
The antagonistic compound
Lanes 1 and 4, Lactococcus lactis DSM
4645 plasmids as
was active at pHs ranging from 2.0 to10.0, but
activity
molecular weight markers; lane 2, plasmids from bacteriocin-
decreased at pH 8.0 or above
(Table 1).
producing wild type; lane 3, plasmids from a bacteriocin-
A growth inhibitory effect of Lact.
plantarum BFE 905
negative mutant of Lact. plantarum BFE
905/M8 CFNS
was observed for all Listeria strains and as an
example,
this effect is shown for L. monocytogenes 125 in Fig.
1. Plasmid
DNA isolation revealed that Lact. plantarum
BFE905 contained plasmids ranging
in size from approxi-
mately 3 to 55 kbp (Fig. 2). Bacteriocin-negative mutants of
exhibited a
high degree (91.8%) of homology with Lact.
plantarum BFE
905 in total soluble protein proflle studies on
Lact. plantarum BFE
905 could not be generated by curing
attempts involving protoplasting,
SDS or novobiocin. A bac-
SDS-PAGE gels, thus conflrming it to be a mutant of the
parent strain. The
bacteriocin-negative trait was stable and
teriocin-negative mutant Lact.
plantarum BFE905/M8 was,
however, obtained by repeated subculturing of the
parent not
lost upon repeated subculturing at 30 °C. The mutant
Lact. plantarum BFE
905/M8 exhibited the same plasmid
strain in MRS broth at 43 °C.
This mutant did not produce
bacteriocin either in liquid or on solid medium. The strain
proflle as
the parent strain (Fig. 2).
© 1998 The Society for Applied Microbiology,Letters
in Applied Microbiology 26, 231-235
234 C.M.A.P.
FRANZ ET AL.
DISCUSSION
(Fricourt et
al. 1994). Plantaricin S is inactivated by lipolytic
and glycolytic enzymes but not by
pepsin and proteinase K
The LAB isolate 905 from `Waldorf' salad that showed antag-
(JimeÁnez Diaz et
al. 1993), and thus differs in these charac-
onistic activity was identifled
as Lact. plantarum by a com-
teristics from plantaricin D.
Plantaricin C is inactivated by
bination of classical identiflcation techniques
(biochemical
the proteases trypsin and a-chymotrypsin
but unlike plan-
and physiological tests) as well as the relatively modern
and taricin
D, it is not inactivated by pepsin and proteinase K
rapid identiflcation technique of
SDS-PAGE of whole cell
(GonzaÁlez et al. 1994).
Plantaricin D differed from plantaricin
proteins. The antagonistic
compound produced by this isolate
LC74 as the latter is sensitive to high temperature
(Rekhif had
a proteinaceous nature, as activity was lost on treatment
et al. 1994).
Based on our results, however, it could not be
with proteases. The compound was
considered heat stable,
determined whether plantaricin D is a novel bacteriocin. For
as activity remained after
heating at 121 °C for 15 min. The
such a determination the primary structure of this
bacteriocin
inhibitory compound was not inactivated by treatment with
would need to be elucidated and
compared to that of known
lipase or a-amylase, which
suggested that activity was not
bacteriocins. The complete primary structure of bacteriocins
dependent on the presence of
either a carbohydrate or lipid
from Lact. plantarum has,
however, seldom been elucidated.
moiety. The compound was inhibitory to growth of L.
mono-
One notable exception is the case of Lact.
plantarum C11,
cytogenes strains (Fig. 1) and based on
these above charac-
a multiple bacteriocin producer of which the genetics
of
teristics, it fulflls the criteria for a bacteriocin (Klaenhammer
production have been studied in
depth (Diep et al. 1996).
1993). The antilisterial activity
and heat stability suggest that
Puriflcation of the Lact. plantarum BFE
905 bacteriocin has
the bacteriocin may belong to the class II bacteriocins as
been attempted, but has so far
met with little success because
described by Klaenhammer (1993) and later, by Nes et
al.
of problems with the relatively low bacteriocin yield, as well
(1996). The bacteriocin was
tentatively named plantaricin D.
as losses in activity during puriflcation procedures
(results The
spectrum of activity of plantaricin D was considered
not shown).
to be relatively narrow
because of the LAB tested, only
Vegetable products including ready-to-eat salads are
often Lact.
sake DSM20017 and Lact. plantarum DSM
20174 were
contaminated with potentially pathogenic bacteria (Beuchat
inhibited, while among the
non-LAB, only L. monocytogenes
1995), and the use of
bacteriocin-producing, naturally-occur-
was inhibited by the bacteriocin.
The Lact. plantarum BFE
ring LAB can be envisaged as so
called `protective cultures'
905 producer culture was not inactivated by its own culture
to improve their safety. The Lact.
plantarum BFE 905 strain
supernatant fluid, which indicated the presence of an immun-
was one of a few selected LAB
(Franz et al. 1996, 1997) that
ity mechanism. It is well known
that most bacteriocin-pro-
were isolated from vegetable foods and that were considered
ducing LAB also produce an
immunity protein that protects
for application in such `biopreservation' studies. Lactobacillus
the cell from self destruction,
and that the genes for the
plantarum BFE 905 is interesting as a
potential protective
bacteriocins and immunity proteins are usually arranged in
culture because of its natural
occurrence in vegetable prod-
operons (Nes et al. 1996).
ucts, antilisterial activity and
growth at refrigeration tem-
Although a bacteriocin-negative mutant was isolated in
perature.
plasmid curing experiments,
loss of the bacteriocin-pro-
ducing trait could not be correlated with a loss of a
plasmid.
This suggested that the bacteriocin structural gene is encoded
ACKNOWLEDGEMENTS
on the chromosome. Large-scale
isolation of plasmid DNA
The authors wish to thank the European Community for
showed the same plasmid proflle
(results not shown), sug-
funds received under the Agricultural and
Industrial
gesting that there were no low copy number plasmids which
Research Program, contract
no.AGRF-CT93-6105.
could have been undetected in small-scale plasmid
isolations.
Our results were similar to those of Kato et al. (1994),
who showed
that loss of bacteriocin production following plasmid
REFERENCES
curing experiments did not result
in plasmid loss.
Abee, T., KroÈckel, L. and Hill, C. (1995) Bacteriocins:
modes of
Plantaricin D production differs from plantacin B (West
action and potentials in food
preservation and control of food
and Warner 1988) and plantaricin F (Fricourt et
al. 1994) as
poisoning. International Journal of
Food Microbiology 28, 169-
the former is produced either on
solid medium or in liquid
185.
culture, while the latter two are produced on solid medium
Beuchat, L.R. (1995) Pathogenic
microorganisms associated with
only. In addition, plantaricin D differs from
plantacin B
fresh produce. Journal of Food
Protection 59, 204-216.
(West and Warner 1988), as the
latter appears to require a
De Vuyst, L. and Vandamme, E.J. (1994) Bacteriocins
of Lactic Acid
carbohydrate moiety for activity. Plantaricin D differs from
Bacteria ed.
De Vuyst, L. and Vandamme, E.J. London: Blackie
plantaricin F because it is
active at pH values above
Academic and Professional.
Diep, D.B., HaÊvarstein, L.S.
and Nes, I.F. (1996) Characterization
4.5,whereas plantaricin F is active only at pH 4.5 or
lower ©
1998 The Society for Applied Microbiology,Letters in Applied
Microbiology 26, 231-235
THE BACTERIOCIN PLANTARICIN D
235
of the locus responsible for the
bacteriocin production in Lacto-
Kato, T., Matsuda, T., Ogawa, E. et
al. (1994) Plantaricin-149, a
bacteriocin produced by Lactobacillus
plantarum NRIC 149. Jour-
bacillus plantarum C11. Journal
of Bacteriology 178, 4472-4483.
nal of Fermentation and
Bioengineering 77, 277-282.
Franz, C.M.A.P., Du Toit, M., von
Holy, A., Schillinger, U. and
Keppler, K., Geisen, R., Schillinger, U. and Holzapfel, W.H.
(1991)
Holzapfel, W.H. (1997) Characterization of nisin-like bacteriocins
Plasmid associated bacteriocin
production by Leuconostoc carno-
produced by Lactococcus
lactis strains isolated from vegetables.
sum. Lactic, 91,
12-13 September, Caen, France.
Journal of Basic Microbiology 37, 187-196.
Klaenhammer, T.R. (1993) Genetics
of bacteriocins produced by
Franz, C.M.A.P., Schillinger, U. and Holzapfel, W.H. (1996)
Pro- lactic
acid bacteria. FEMS Microbiology Reviews 12, 39-86.
duction and characterization of
enterocin 900, a bacteriocin pro-
Nes, I.F., Diep, D.B., HaÊvarstein, L.S., Brurberg,
M.B. and Eijsink,
duced by Enterococcus faecium BFE
900 from black olives.
V. and Holo, H. (1996) Biosynthesis of bacteriocins in
lactic acid
International Journal of Food Microbiology 29, 255-270.
bacteria. Antonie
van Leeuwenhoek 70, 113-128.
Fricourt, B.V., Barefoot, S.F.,
Testin, R.F. and Hayasaka, S.S.
Rekhif, N., Atrih, A. and Lefebvre, G. (1994)
Characterization and
(1994) Detection and activity of plantaricin F an
antibacterial
partial puriflcation of plantaricin LC74, a bacteriocin
produced
substance from Lactobacillus plantarum BF001 isolated
from pro-
by Lactobacillus plantarum LC74. Biotechnology
Letters 16, 771-
cessed channel catflsh. Journal
of Food Protection 57, 698-702.
776.
GonzaÁlez, B., Arca, P., Mayo, B.
and SuaÁrez, J.E. (1994) Detection,
Ruiz-Barba, J.L., Cathcart, D.P., Warner, P.J. and
JimeÁnez-Diaz,
puriflcation, and partial characterization of plantaricin C,
a bac- R.
(1994) Use of Lactobacillus plantarum LPCO10, a
bacteriocin
teriocin produced by a Lactobacillus plantarum strain
of dairy
producer, as a starter culture in Spanish-style green olive fer-
origin. Applied
and Environmental Microbiology 60, 2158-2163.
mentations. Applied
and Environmental Microbiology 60, 2059-
Hammes, W.P. and Tichaczek, P.S.
(1994) The potential of lactic
2064.
acid bacteria for the production of safe and wholesome food.
Sambrook, J., Fritsch, E.F. and
Maniatis, T. (1989) Molecular Clon-
International Journal of Food
Research and Technology 198, 193-
ing: a Laboratory Manual 2nd
edn. Cold Spring Harbor, N.Y.:
201.
Cold Spring Harbor Laboratory.
Holck, A., Axelsson, L. and
Schillinger, U. (1996) Divergicin 750,
Schillinger, U. and Lu
Ècke, F.-K. (1987) Identiflcation
of lactobacilli
a novel bacteriocin produced by Carnobacterium
divergens 750.
from meat and meat products. Food
Microbiology 4, 199-208.
FEMS Microbiology Letters 136, 163-168.
Stiles, M.E. and Holzapfel, W.H.
(1997) Lactic acid bacteria of
JimeÁnez-Diaz, R., Rios-SaÁnchez, R.M., Desmazeaud, M.,
Ruiz- foods
and their current taxonomy. International Journal of Food
Barba, J.L. and Piard, J.-C.
(1993) Plantaricins S and T, two
Microbiology 36, 1-29.
new bacteriocins produced by Lactobacillus
plantarum LPCO10
West, C.A. and Warner, P.J. (1988) Plantacin B, a
bacteriocin pro-
isolated from a green olive fermentation. Applied
and Environ-
duced by Lactobacillus plantarum NCDO
1193. FEMS Micro-
biology Letters 49, 163-165.
mental Microbiology 59, 1416-1424.
(Full Text
online - PDF)
