International Journal of Food Microbwlogy, 13 (1991) 119-130 © 1991 Elsevier Science Publishers B.V. 0168-1605/91/$03.50
119
FOOD 00406
Antagonistic effect of coryneform bacteria from red smear cheese against Listeria species Natalia Valdes-Stauber, Harald G/Stz and Martin Busse Department of Bacteriology, South German Experimental and Research Station in Dairying, Technical University of Munich at Weihenstephan, Freisin~ F.R.G. (Received 28 September 1990; accepted 18 January 1991)
A to~l of 187 coryneform bacteria were isolated from red smear cheese and screened for inhibitory effects against 16 strains of Listeria species. Culture filtrates from Brevibacterium linens (16 strains), Arthrobacter nicotianae (4 strains) and Arthrobacter nucleogenes (3 strains) showed clear zones of inhibition. The antagonistic effect was seen against 26 to 87% of 91 l.a'steria strains tested. A. nicotianae and A. nucleogenes were more effective against Listeria innocua and Listeria ivanovii than against Listeria monocytogenes. No species specifically was observed for B. linens, but there was a difference regarding the inhibitory activity of individual culture filtrates. When culture filtrates of the test strains were added to 1.2steria broth cultures, the maximum growth level was not attained. Inhibition in broth cultures was dependent on the concentration of culture filtrates. Culture filtrates from the late stationary phase had a stronger inhibitory effect on the growth of L, mono~ytogenes. The nature of the inhibitory effects remained unclear. Attempts to characterize the nature of the antagonism showed that the culture filtrates lost their inhibitory activity upon heating, and the molecular size of the inhibitory substances were greater than 12-14 kDa. Key words: l.,isteria; Antagonism; Cheese smear coryneform bacteria
Introduction
Cheeses are known to be sources of foodborne listeriosis (Terplan, 1989). Due to production processes and ripening procedures, semi-soft and soft cheeses are to be considered as problematic cheese types. Whereas Listeria spp. were isolated from 2-3% of all cheese samples, the proportion of positive samples in soft cheese with moulded surface and red smear came close to 10% (Terplan et al., 1986; Breer, 1986). These cheese types offer favourable ecological conditions for listerial growth (Terplan et al., 1986; Ryser and Marth, 1987, 1989). Recent reports have demonstrated inhibition of Listeria by different strains of bacteria used for food or cheese manufacture (Harris et al., 1989; Racchach et al., 1989). Lactic acid bacteria can inhibit Listeria by producing organic acids, hydroCorrespondence address: H. Gt~tz, Bakterioiogisches lnstitut, SVFAM, Weihenstephan. Technische Universitltt Mfinchen, Vt~ttingerstr. 45, D-8050 Freising- F.R.G.
120 gen peroxide and other inhibitory, substances (Carminati et al., 1989: Schaak and Marth, 1988a,b). Some bacteria, present in ripened soft cheese, have the potential to inhibit Listeria (Asperger et al., 1989). Hug-Michel et al. (1989) isolated Arthrobacter s p p and Serratia liquefaciens from cheese smear which could inhibit Listeria on solid media. These inhibitory effects were seen only in model experiments. Attempts to apply these observations to red smear cheeses were not successful (Asperger et al., 1989). Coryneform bacteria are important in the ripening of red smear cheeses (Mulder et al., 1966; E1-Erian, 1969; Seiler, 1986). Thus, the aim of this study was to evaluate the inhibitory effect of the red smear surface flora on various Listeria species.
Materials and Methods
Maintaining of cultures All cultures used in this study were grown on agar slants and stored at 4 ° C. They were then subcultured in broth media for the individual experiments. Test strains Coryneform bacteria were isolated from various red smear cheese types by surface plating, and were purified repeatedly for further studies. For isolation of coryneform bacteria we used plate count agar (PCA; Merck) to which 3% ( w / v ) NaC1 (modified PCA) was added (SeLler, 1986). The isolates were characterized morphologically and further differentiated using 53 physiological tests according to Seiler (1983). The test strains were isolated from 12 samples of red smear cheese comprising nine different types (i.e. seven from F.R.G. and one each from France and Switzerland). Indicator strains For preliminary screening tests, 16 strains of Listeria were used as indicator organisms (Table I). For further studies, 91 Listeria strains belonging to five confirmed species from the culture collection of our institute were used. These consisted of Listeria monocytogenes (42 strains, all serotypes), Listeria innocua (34 strains, all serotypes), Listeria ivanovii (11), Listeria seeligeri (3) and Listeria welshimeri (ATCC 35897). Furthermore, a strain of Listeria grayi (ATCC 19120), Listeria murrayi (ATCC 25401) and Jonesia denitrificans (ATCC 14870) were included in the study. Preparation of cultures The test strains were grown in modified PC broth (Peptone from casein 5.0 g / l , yeast extract 2.5 g / l , o( + )glucose 1.0 g/l, NaCI 30 g/l, pH 7.0) at 30 ° C for 72 h under continuous shaking to ensure aeration of the culture. The indicator strains were grown in tryptose broth (TB, Merck) at 30 ° C for 24 h. A volume of 0.1 ml of
121 these cultures was then further subcultured in 5 ml of tryptose broth and incubated at 30 ° C for 24 h.
Culture filtrates A 72-h broth culture of a test strain was centrifuged at 3000 rpm for 20 rain, and the supernatant was used for screening purposes. Before use the supernatant was neutralized with 0.1 N N a O H or 0.1 N HC1, and finally filter-sterilized (Millipore, 0.02 ~ m pore size). The sterile culture filtrate was heated at 60 ° C in a water bath for 30 rain to test for heat stability of the inhibitory substances. The sterile culture filtrates were dialysed against PBS buffer (NaCI 7.56 g / l , N a 2 H P O 4 0.724 g / l , K 2 H P O 4 0.210 g / l , H 2 0 dest. 1 1, p H 7.2) at 4 ° C for 24 h (Visking dialysis tubing, 12000 MW cut-off; Serva). Nisin (Sigma) in varying concentrations was tested against Listeria strains in solid media to compare with the culture filtrates. Antagonism in solid media The method described by Arthur and Barry (1980) was used to study the inhibitory effects of 187 coryneform bacteria strains against the selected set of 16 indicator strains of Listeria spp., using the agar diffusion test (well assay technique). The method was further applied to study the inhibition spectrum of the test strains against 91 Listeria strains, and to characterize the inhibitory principle. For preparation of plates, a Listeria strain (0.4 ml of a 24-h culture grown at 30 o C) was inoculated into 15 ml of soft tryptose agar (TB, Merck with 8 g agar/1), well mixed and poured into a petridish (~ 10 cm). Following solidification of the agar, up to 9 wells of 6 m m diameter were aseptically punched in the agar plates. Subsequently, volumes of 40/~1 each the sterile culture filtrates were pipetted into the wells. As a control, the same amount of modified PC broth was pipetted into one well. The plates were then incubated at 30 ° C. Growth and inhibition were detected using Henry's illumination (Henry, 1933). Plates were checked for inhibitory zones after 24, 48 and 72 h of incubation. The distance between the edge of the well and the inhibition zone was measured using a calliper. Antagonism in liquid media To test the inhibitory effect of neutralized culture filtrates in broth cultures (Hoover et al., 1989), three Listeria strains (SLCC 2372, N C T C 7973 and SLCC 1694), three positive test Strains of Brevibacterium linens (M18, M21 and M24), one positive Arthrobacter nicotianae strain (G05) and one negative test strain of B. linens (R01) were chosen. Choice of strains and combinations was based upon the observations made in solid media. Coryneform bacteria, whose culture filtrates showed strong inhibitory effects, were tested against sensitive as well as relatively resistant Listeria strains. A 24-h culture of Listeria strains grown in TB at 30 ° C was diluted in TB to a level of 106 cells/ml. 1 ml of this suspension was mixed with 1 ml of a sterile culture filtrate of a test strain. To serve as control, 1 ml of sterile modified PC broth was
122
a d d e d to the Listeria suspension. T o investigate the effect of the relative c o n c e n t r a tion of the culture filtrates, 1 ml of the Listeria s u s p e n s i o n was m i x e d with 2 ml, 1.5 ml, 1 ml a n d 0.1 ml of the culture filtrates respectively, to o b t a i n ratios of 1 : 2, 1 : 1.5, 1 : 1 and 1 : 0.1. The v o l u m e was a d j u s t e d to 3 ml with m o d i f i e d PC b r o t h . As a control, 1 ml of the Listeria s u s p e n s i o n was m i x e d with 2 ml of sterile m o d i f i e d PC broth. T h e b l a n k consisted of a m i x t u r e o f 1 ml TB a n d 2 ml m o d i f i e d PC b r o t h . T h e e x p e r i m e n t was carried out in m i c r o t i t r a t i o n plates. A v o l u m e o f 2 0 0 / d was filled in each of three wells. T h e plates were sealed a n d i n c u b a t e d at 3 0 ° C for 4 days. G r o w t h of the Listeria strains was m e a s u r e d at 620 n m ever)" 2 or 4 h, using an E L I S A r e a d e r ( E A R 400, S L T L a b i n s t r u m e n t s , A u s t r i a ) . Before r e a d i n g s the plates were shaken for 10 rain. R e a d i n g s of the three wells were averaged. T h e b l a n k was substracted, to yield the final results.
Results I n h i b i t i o n of Listeria spp. in the a g a r diffusion tests s h o w e d up as lysis of confluent Listeria growth. Listeriae a p p e a r e d to g r o w n o r m a l l y in the first 24 h. A f t e r 48 h, however, clear and, in c e r t a i n cases, u n c l e a r i n h i b i t i o n zones were o b s e r v e d a r o u n d the wells (Fig. 1). T h e size o f these zones was s o m e t i m e s i n c r e a s e d
Fig. 1. Inhibition of L. monocytogenes (1031) by culture filtrates of coryneform bacteria in solid media (agar diffusion assay).
123 TABLE I Listeria strains used for screening of coryneform bacteria
Species
Serovar
Code no.
Strain designation
L monocytogenes
1/2a 1/2a 1/2a 1/2c 3c 3b 4b 4c 4d 4e "7" 6a 6b n.d. d 5 1/2b
1006 1040 1440 1001 1032 1031 1003 1019 1033 1018 1034 2011 2023 2406 3009 4007
NCTC a7973 ATCC 15313 food isolate SLCC b 2372. ATCC c 19112 SLCC 2479 SLCC 1694 SLCC 2375. ATCC 19115 SLCC 2376, ATCC 19116 SLCC 2377, ATCC 19117 SLCC 2378, ATCC 19118 SLCC 2482 sLCC 337% ATCC 33090 SLCC 5640 food isolate SLCC 4769 SLCC 3954
L innocua
L. ivanovii L seeligeri
a b c d
National Collection of Type Cultures. Special Listeria Culture Collection. American Type Culture Collection. not determined.
TABLE II Preliminary screening of coryneform bacteria for inhibitory activity against Listeria species (agar diffusion assay) Test strains Species Brevibacterium linens Brevibacterium imperiale Brevibacterium f e r m e n t a n s Brevibacterium spp. Arthrobacter a m m o n i a g e n e s Arthrobacter citreu$ A rthrobacter nicotianae A rthrobacter nucleogenes Arthrobacter spp. Corynebacterium fascians Corynebacterium insidiosum C o ~ n e b a c t e r i u m spp. M y c o b a c t e r i u m spp.
Total
Inhibitory. effect a Number of isolates 51 5
Clear zone 16 _ b
Unclear zone 9 _
3
-
1
1 39
-
-
4
-
-
15 3
4 3
-
22
-
-
15 1
-
-
13 1
-
-
187
23
10
a Number of isolates being inhibitory to one or more Listeria strains included in Table 1. b - ; no inhibition observed.
124 after 72 h (data n o t included). A l t h o u g h i n h i b i t i o n zones were also present e m p l o y ing Nisin, n o initial growth of L i s t e r i a was observed a r o u n d the wells (data not shown). Heat-treated culture filtrates of B. linens strains M15, M16, M19 a n d M24 lost their i n h i b i t o r y activity (data not included). C u l t u r e filtrates of five B. l i n e n s strains (M15, M18, M21, M22 a n d M24) were dialysed a n d the i n h i b i t o r3' substances were f o u n d to be n o n - d i a t y s a b l e (data not included). The L i s t e r i a strains used to screen the 187 c o r y n e f o r m strains i n c l u d e d in this study are shown in T a b l e I. T h e results o b t a i n e d are presented in T a b l e II. F o r further e x a m i n a t i o n s , only those test strains were considered, whose culture filtrates showed a clear zone of i n h i b i t i o n against at least one L i s t e r i a strain. O u t of the 33 test strains, which were considered positive b y m e a n s of the p r e l i m i n a r y screenings, the culture filtrates of only 23 strains showed a clear i n h i b i t i o n zone. E x a m i n a t i o n of the 23 c o r y n e f o r m test strains just m e n t i o n e d against 91 L i s t e r i a strains showed that no single test filtrate had an a n t a g o n i s t i c effect against all L i s t e r i a strains. I n h i b i t o r y effects against L. m o n o c v t o g e n e s , L. i n n o c u a a n d L. i v a n o v i i are presented in T a b l e III. TABLE III Percentage of Listeria strains inhibited by broth culture filtrates of coryneform bacteria (agar diffusion assay) Coryneforms species Breoibacterium hnens
Arthrobacter mcotianae
Arthrobacter nucleogenes
Listeria spp. (no. of strains included)
code no. I01 M10 Mll M12 MI4 M15 M16 M17 M18 M19 M20 M21 M22 M23 M24 M25 106 G05 G12 M04 H02 H06 H07
(42)
(34)
(11 )
Total a (91)
57.1% 76.2% 69.0% 61.9% 66.7% 78.6% 57.1% 52.4% 90.5% 64.3% 57.1% 59.5% 54.8% 71.4% 66.7% 45.2% 38.1% 61.9% 52.4% 54.8% 50.0% 47.6% 57.1%
5.9% 88.2% 61.8% 29.4% 23.5% 85.3% 32.4% 41.2% 91.2% 35.3% 29.4% 41.2% 8.8% 52.9% 41.2% 2.9% 47.1% 79.4% 73.5% 76.5% 79.4% 79.4% 82.4%
9.1 ~ 81.8% 54.5% 63.6% 9.1% 54.5% 63.6% 63.6% 63.6% 54.5% 45.5% 27.3% 36.4% 45.5% 27.3% 27.3'~ 72.7% 90.9% 72.7% 63.6% 81.8% 81.8~ 81.8%
31.9% 81.3% 64.8% 50.6% 42.9% 76.9% 49.5% 48.4% 86.8% 52.8% 45.1% 49.5% 35.2% 61.5% 51.7% 26.4% 47.3% 72.5% 63.7% 63.7% 64.8% 63.7% 69.2%
L monocytogenes
L innocua
L. ic'anotrii
The total number of strains also includes 3 strains of L seeligeri and 1 strain of L welshimeri.
IV
For
origin
-
effect
+
H07
-
no inhibition
b
+
1 and
zone;
see Table
+
+
M04
H02
++
+
GI2
H06
+ +
-
106
++
M25
GO5
+ +
++
MI9
M24
++
MI8
+ +
++
M17
+ +
++
MI6
M23
++
M15
M22
++
M]4
+ +
++
M12
+ +
++
MII
M21
++
M20
+ +
MI0
+
.
= clear
Materials
+
+
.
-
++
++
+ +
++
+ +
+ +
+
+ +
+ +
++
++
++
++
++
++
++
++
+
+
1006
and
-
-
+
.
.
.
-
.
+ +
+
+
+
.
++
++
.
+
++
+
+
+
+
+ +
1031
.
.
_
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
zone
Methods.
.
.
.
.
.
.
.
.
.
.
.
-
-
-
-
-
-
-
-
+
-
.
(+)
( < 3 mm);
.
-
-
-
-
+
1033
+
+
+
-
++
-
+
+
+
+
+
+
+
-
-
+
-
_
i
clear
Listeria
1040
+ +
.
selected
(+)
.
against
1032
bacteria
inhibition
"
of coryneform
monocytogenes
1001
L
filtrates
101
of culture
of strains
nucleogenes
b Symbols:
•
A.
A. nicotianae
B. linens
Antagonistic
TABLE
.
.
.
.
++
++
+
+
-
-
.
-
.
.
.
.
.
zone
+ +
++
++
++
++
-
++
++
+
++
+
"
(agar
.
.
.
.
.
.
.
.
.
.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
+
_
-
+
--
-
-
-
-
-
-
-
-
-
-
(+)
-
-
-
_
inhibition
ivanovii "
3009
L
assay)
( + ) = unclear
2406
diffusion
( > 3 ram);
-
-
-
-
+
+
+
+ +
+
+
+
+
+
+
+
+
+
_
2023
innocua
strains
2011
L
and
inhibition
+ +
++
+
+
++
++
-
+ +
+ +
+ +
+ +
+ +
++
++
++
++
++
++
++
++
++
+ +
1440
species
L
seeligeri "
+
+
+
zone.
+
+
+
+ +
++
++
+
++
+ +
+ +
+ +
+ +
+ +
++
++
++
++
++
++
++
++
++
+ +
4007
126 O. 35 [ L~ m._onocytog_,ene.._sss(I031 ) ~
control; no addition of culture f i l t r a t e
O" 30 ] -~9- B tinens (R01)
[]
A nicottanae ( G O 5 ) ~
~0.20
~
~~_~~_~.~ .
÷
o.~o/
015
O. 05
oI 6
8
10
12
14
16
18.
20
22
24
26
28
30
32
34
36
T i m e (h) at 30°C
Fig. 2. Inhibitory effect of sterile culture filtrates of various bacteria on the growth of L. raonocytogenes
(1031) in liquid media. Culture filtrates of A. nicotianae and A. nucleogenes were considerably more inhibitory of L. innocua and L. ioanooii. Only 38-62% of L. monocytogenes were inhibited. Within the B. linens strains, it appeared that a higher proportion of L. monocytogenes was inhibited by B. linens than L. innocua. The sterile filtrates had varying levels of activity. Of particular interest are the culture filtrates of B. linens M10, M15 and M18, which are active against a high proportion of strains of L. rnonocytogenes, L. innocua and L. ioanovii investigated. None of the test strains included in Table III were inhibitory to L. welshimeri, L. grayi, L. rnurravi or Jonesia denitrificans (data not shown). Two of the three strains of L. seeligeri tested were, however, inhibited by all 23 coryneforms (data not shown). The Listeria strains investigated showed varying sensitivity to the sterile culture filtrates of coryneform bacteria (Table IV). Furthermore, 11 L. rnonocytogenes, one L. ivanovii and one L. seeligeri strains were extremely sensitive to the culture filtrates, in being inhibited by 96% of the 23 tested filtrates (results not shown). In contrast, three L. monocytogenes and two each of L. innocua and L. ivanovii strains showed almost total resistance to 22 out of the 23 culture filtrates (results not shown). The inhibitory effect of the various sterile culture filtrates on growth of L. monocytogenes (1031) is depicted in Fig. 2. Similar inhibitory patterns were also observed with respect to two other strains of L. monocytogenes (data not given). Further incubation up to 96 h did not result in any increase in listerial growth. N o inhibition, however, was observed with B. linens (R01). The effect of dilution on the inhibitory properties of a 72-h culture filtrate of B. linens (M21) against L. monocytogenes (1031) is presented in Fig. 3. Additional tests
127 0.30
L. monocyto~enes (1031) control; no oddition of c u l t u r e t i l t r o t e
0.25
~
Culture f i l t r o t e (1:2)
[]
--t-- C u l t u r e f l i t rote (1 : 1.5) c o
0.20
+
C u l t u r e f i l t r o t e (1:1,
~
Culture f i l t r o t e (1:0.1)
/ /
~u ~ " l = [ " ~ - D
p /
I'1 I/_/
~
~u 0.~5 £ o
f
~
#k-,---__._~
I
I
28
30
o
119
FOOD 00406
Antagonistic effect of coryneform bacteria from red smear cheese against Listeria species Natalia Valdes-Stauber, Harald G/Stz and Martin Busse Department of Bacteriology, South German Experimental and Research Station in Dairying, Technical University of Munich at Weihenstephan, Freisin~ F.R.G. (Received 28 September 1990; accepted 18 January 1991)
A to~l of 187 coryneform bacteria were isolated from red smear cheese and screened for inhibitory effects against 16 strains of Listeria species. Culture filtrates from Brevibacterium linens (16 strains), Arthrobacter nicotianae (4 strains) and Arthrobacter nucleogenes (3 strains) showed clear zones of inhibition. The antagonistic effect was seen against 26 to 87% of 91 l.a'steria strains tested. A. nicotianae and A. nucleogenes were more effective against Listeria innocua and Listeria ivanovii than against Listeria monocytogenes. No species specifically was observed for B. linens, but there was a difference regarding the inhibitory activity of individual culture filtrates. When culture filtrates of the test strains were added to 1.2steria broth cultures, the maximum growth level was not attained. Inhibition in broth cultures was dependent on the concentration of culture filtrates. Culture filtrates from the late stationary phase had a stronger inhibitory effect on the growth of L, mono~ytogenes. The nature of the inhibitory effects remained unclear. Attempts to characterize the nature of the antagonism showed that the culture filtrates lost their inhibitory activity upon heating, and the molecular size of the inhibitory substances were greater than 12-14 kDa. Key words: l.,isteria; Antagonism; Cheese smear coryneform bacteria
Introduction
Cheeses are known to be sources of foodborne listeriosis (Terplan, 1989). Due to production processes and ripening procedures, semi-soft and soft cheeses are to be considered as problematic cheese types. Whereas Listeria spp. were isolated from 2-3% of all cheese samples, the proportion of positive samples in soft cheese with moulded surface and red smear came close to 10% (Terplan et al., 1986; Breer, 1986). These cheese types offer favourable ecological conditions for listerial growth (Terplan et al., 1986; Ryser and Marth, 1987, 1989). Recent reports have demonstrated inhibition of Listeria by different strains of bacteria used for food or cheese manufacture (Harris et al., 1989; Racchach et al., 1989). Lactic acid bacteria can inhibit Listeria by producing organic acids, hydroCorrespondence address: H. Gt~tz, Bakterioiogisches lnstitut, SVFAM, Weihenstephan. Technische Universitltt Mfinchen, Vt~ttingerstr. 45, D-8050 Freising- F.R.G.
120 gen peroxide and other inhibitory, substances (Carminati et al., 1989: Schaak and Marth, 1988a,b). Some bacteria, present in ripened soft cheese, have the potential to inhibit Listeria (Asperger et al., 1989). Hug-Michel et al. (1989) isolated Arthrobacter s p p and Serratia liquefaciens from cheese smear which could inhibit Listeria on solid media. These inhibitory effects were seen only in model experiments. Attempts to apply these observations to red smear cheeses were not successful (Asperger et al., 1989). Coryneform bacteria are important in the ripening of red smear cheeses (Mulder et al., 1966; E1-Erian, 1969; Seiler, 1986). Thus, the aim of this study was to evaluate the inhibitory effect of the red smear surface flora on various Listeria species.
Materials and Methods
Maintaining of cultures All cultures used in this study were grown on agar slants and stored at 4 ° C. They were then subcultured in broth media for the individual experiments. Test strains Coryneform bacteria were isolated from various red smear cheese types by surface plating, and were purified repeatedly for further studies. For isolation of coryneform bacteria we used plate count agar (PCA; Merck) to which 3% ( w / v ) NaC1 (modified PCA) was added (SeLler, 1986). The isolates were characterized morphologically and further differentiated using 53 physiological tests according to Seiler (1983). The test strains were isolated from 12 samples of red smear cheese comprising nine different types (i.e. seven from F.R.G. and one each from France and Switzerland). Indicator strains For preliminary screening tests, 16 strains of Listeria were used as indicator organisms (Table I). For further studies, 91 Listeria strains belonging to five confirmed species from the culture collection of our institute were used. These consisted of Listeria monocytogenes (42 strains, all serotypes), Listeria innocua (34 strains, all serotypes), Listeria ivanovii (11), Listeria seeligeri (3) and Listeria welshimeri (ATCC 35897). Furthermore, a strain of Listeria grayi (ATCC 19120), Listeria murrayi (ATCC 25401) and Jonesia denitrificans (ATCC 14870) were included in the study. Preparation of cultures The test strains were grown in modified PC broth (Peptone from casein 5.0 g / l , yeast extract 2.5 g / l , o( + )glucose 1.0 g/l, NaCI 30 g/l, pH 7.0) at 30 ° C for 72 h under continuous shaking to ensure aeration of the culture. The indicator strains were grown in tryptose broth (TB, Merck) at 30 ° C for 24 h. A volume of 0.1 ml of
121 these cultures was then further subcultured in 5 ml of tryptose broth and incubated at 30 ° C for 24 h.
Culture filtrates A 72-h broth culture of a test strain was centrifuged at 3000 rpm for 20 rain, and the supernatant was used for screening purposes. Before use the supernatant was neutralized with 0.1 N N a O H or 0.1 N HC1, and finally filter-sterilized (Millipore, 0.02 ~ m pore size). The sterile culture filtrate was heated at 60 ° C in a water bath for 30 rain to test for heat stability of the inhibitory substances. The sterile culture filtrates were dialysed against PBS buffer (NaCI 7.56 g / l , N a 2 H P O 4 0.724 g / l , K 2 H P O 4 0.210 g / l , H 2 0 dest. 1 1, p H 7.2) at 4 ° C for 24 h (Visking dialysis tubing, 12000 MW cut-off; Serva). Nisin (Sigma) in varying concentrations was tested against Listeria strains in solid media to compare with the culture filtrates. Antagonism in solid media The method described by Arthur and Barry (1980) was used to study the inhibitory effects of 187 coryneform bacteria strains against the selected set of 16 indicator strains of Listeria spp., using the agar diffusion test (well assay technique). The method was further applied to study the inhibition spectrum of the test strains against 91 Listeria strains, and to characterize the inhibitory principle. For preparation of plates, a Listeria strain (0.4 ml of a 24-h culture grown at 30 o C) was inoculated into 15 ml of soft tryptose agar (TB, Merck with 8 g agar/1), well mixed and poured into a petridish (~ 10 cm). Following solidification of the agar, up to 9 wells of 6 m m diameter were aseptically punched in the agar plates. Subsequently, volumes of 40/~1 each the sterile culture filtrates were pipetted into the wells. As a control, the same amount of modified PC broth was pipetted into one well. The plates were then incubated at 30 ° C. Growth and inhibition were detected using Henry's illumination (Henry, 1933). Plates were checked for inhibitory zones after 24, 48 and 72 h of incubation. The distance between the edge of the well and the inhibition zone was measured using a calliper. Antagonism in liquid media To test the inhibitory effect of neutralized culture filtrates in broth cultures (Hoover et al., 1989), three Listeria strains (SLCC 2372, N C T C 7973 and SLCC 1694), three positive test Strains of Brevibacterium linens (M18, M21 and M24), one positive Arthrobacter nicotianae strain (G05) and one negative test strain of B. linens (R01) were chosen. Choice of strains and combinations was based upon the observations made in solid media. Coryneform bacteria, whose culture filtrates showed strong inhibitory effects, were tested against sensitive as well as relatively resistant Listeria strains. A 24-h culture of Listeria strains grown in TB at 30 ° C was diluted in TB to a level of 106 cells/ml. 1 ml of this suspension was mixed with 1 ml of a sterile culture filtrate of a test strain. To serve as control, 1 ml of sterile modified PC broth was
122
a d d e d to the Listeria suspension. T o investigate the effect of the relative c o n c e n t r a tion of the culture filtrates, 1 ml of the Listeria s u s p e n s i o n was m i x e d with 2 ml, 1.5 ml, 1 ml a n d 0.1 ml of the culture filtrates respectively, to o b t a i n ratios of 1 : 2, 1 : 1.5, 1 : 1 and 1 : 0.1. The v o l u m e was a d j u s t e d to 3 ml with m o d i f i e d PC b r o t h . As a control, 1 ml of the Listeria s u s p e n s i o n was m i x e d with 2 ml of sterile m o d i f i e d PC broth. T h e b l a n k consisted of a m i x t u r e o f 1 ml TB a n d 2 ml m o d i f i e d PC b r o t h . T h e e x p e r i m e n t was carried out in m i c r o t i t r a t i o n plates. A v o l u m e o f 2 0 0 / d was filled in each of three wells. T h e plates were sealed a n d i n c u b a t e d at 3 0 ° C for 4 days. G r o w t h of the Listeria strains was m e a s u r e d at 620 n m ever)" 2 or 4 h, using an E L I S A r e a d e r ( E A R 400, S L T L a b i n s t r u m e n t s , A u s t r i a ) . Before r e a d i n g s the plates were shaken for 10 rain. R e a d i n g s of the three wells were averaged. T h e b l a n k was substracted, to yield the final results.
Results I n h i b i t i o n of Listeria spp. in the a g a r diffusion tests s h o w e d up as lysis of confluent Listeria growth. Listeriae a p p e a r e d to g r o w n o r m a l l y in the first 24 h. A f t e r 48 h, however, clear and, in c e r t a i n cases, u n c l e a r i n h i b i t i o n zones were o b s e r v e d a r o u n d the wells (Fig. 1). T h e size o f these zones was s o m e t i m e s i n c r e a s e d
Fig. 1. Inhibition of L. monocytogenes (1031) by culture filtrates of coryneform bacteria in solid media (agar diffusion assay).
123 TABLE I Listeria strains used for screening of coryneform bacteria
Species
Serovar
Code no.
Strain designation
L monocytogenes
1/2a 1/2a 1/2a 1/2c 3c 3b 4b 4c 4d 4e "7" 6a 6b n.d. d 5 1/2b
1006 1040 1440 1001 1032 1031 1003 1019 1033 1018 1034 2011 2023 2406 3009 4007
NCTC a7973 ATCC 15313 food isolate SLCC b 2372. ATCC c 19112 SLCC 2479 SLCC 1694 SLCC 2375. ATCC 19115 SLCC 2376, ATCC 19116 SLCC 2377, ATCC 19117 SLCC 2378, ATCC 19118 SLCC 2482 sLCC 337% ATCC 33090 SLCC 5640 food isolate SLCC 4769 SLCC 3954
L innocua
L. ivanovii L seeligeri
a b c d
National Collection of Type Cultures. Special Listeria Culture Collection. American Type Culture Collection. not determined.
TABLE II Preliminary screening of coryneform bacteria for inhibitory activity against Listeria species (agar diffusion assay) Test strains Species Brevibacterium linens Brevibacterium imperiale Brevibacterium f e r m e n t a n s Brevibacterium spp. Arthrobacter a m m o n i a g e n e s Arthrobacter citreu$ A rthrobacter nicotianae A rthrobacter nucleogenes Arthrobacter spp. Corynebacterium fascians Corynebacterium insidiosum C o ~ n e b a c t e r i u m spp. M y c o b a c t e r i u m spp.
Total
Inhibitory. effect a Number of isolates 51 5
Clear zone 16 _ b
Unclear zone 9 _
3
-
1
1 39
-
-
4
-
-
15 3
4 3
-
22
-
-
15 1
-
-
13 1
-
-
187
23
10
a Number of isolates being inhibitory to one or more Listeria strains included in Table 1. b - ; no inhibition observed.
124 after 72 h (data n o t included). A l t h o u g h i n h i b i t i o n zones were also present e m p l o y ing Nisin, n o initial growth of L i s t e r i a was observed a r o u n d the wells (data not shown). Heat-treated culture filtrates of B. linens strains M15, M16, M19 a n d M24 lost their i n h i b i t o r y activity (data not included). C u l t u r e filtrates of five B. l i n e n s strains (M15, M18, M21, M22 a n d M24) were dialysed a n d the i n h i b i t o r3' substances were f o u n d to be n o n - d i a t y s a b l e (data not included). The L i s t e r i a strains used to screen the 187 c o r y n e f o r m strains i n c l u d e d in this study are shown in T a b l e I. T h e results o b t a i n e d are presented in T a b l e II. F o r further e x a m i n a t i o n s , only those test strains were considered, whose culture filtrates showed a clear zone of i n h i b i t i o n against at least one L i s t e r i a strain. O u t of the 33 test strains, which were considered positive b y m e a n s of the p r e l i m i n a r y screenings, the culture filtrates of only 23 strains showed a clear i n h i b i t i o n zone. E x a m i n a t i o n of the 23 c o r y n e f o r m test strains just m e n t i o n e d against 91 L i s t e r i a strains showed that no single test filtrate had an a n t a g o n i s t i c effect against all L i s t e r i a strains. I n h i b i t o r y effects against L. m o n o c v t o g e n e s , L. i n n o c u a a n d L. i v a n o v i i are presented in T a b l e III. TABLE III Percentage of Listeria strains inhibited by broth culture filtrates of coryneform bacteria (agar diffusion assay) Coryneforms species Breoibacterium hnens
Arthrobacter mcotianae
Arthrobacter nucleogenes
Listeria spp. (no. of strains included)
code no. I01 M10 Mll M12 MI4 M15 M16 M17 M18 M19 M20 M21 M22 M23 M24 M25 106 G05 G12 M04 H02 H06 H07
(42)
(34)
(11 )
Total a (91)
57.1% 76.2% 69.0% 61.9% 66.7% 78.6% 57.1% 52.4% 90.5% 64.3% 57.1% 59.5% 54.8% 71.4% 66.7% 45.2% 38.1% 61.9% 52.4% 54.8% 50.0% 47.6% 57.1%
5.9% 88.2% 61.8% 29.4% 23.5% 85.3% 32.4% 41.2% 91.2% 35.3% 29.4% 41.2% 8.8% 52.9% 41.2% 2.9% 47.1% 79.4% 73.5% 76.5% 79.4% 79.4% 82.4%
9.1 ~ 81.8% 54.5% 63.6% 9.1% 54.5% 63.6% 63.6% 63.6% 54.5% 45.5% 27.3% 36.4% 45.5% 27.3% 27.3'~ 72.7% 90.9% 72.7% 63.6% 81.8% 81.8~ 81.8%
31.9% 81.3% 64.8% 50.6% 42.9% 76.9% 49.5% 48.4% 86.8% 52.8% 45.1% 49.5% 35.2% 61.5% 51.7% 26.4% 47.3% 72.5% 63.7% 63.7% 64.8% 63.7% 69.2%
L monocytogenes
L innocua
L. ic'anotrii
The total number of strains also includes 3 strains of L seeligeri and 1 strain of L welshimeri.
IV
For
origin
-
effect
+
H07
-
no inhibition
b
+
1 and
zone;
see Table
+
+
M04
H02
++
+
GI2
H06
+ +
-
106
++
M25
GO5
+ +
++
MI9
M24
++
MI8
+ +
++
M17
+ +
++
MI6
M23
++
M15
M22
++
M]4
+ +
++
M12
+ +
++
MII
M21
++
M20
+ +
MI0
+
.
= clear
Materials
+
+
.
-
++
++
+ +
++
+ +
+ +
+
+ +
+ +
++
++
++
++
++
++
++
++
+
+
1006
and
-
-
+
.
.
.
-
.
+ +
+
+
+
.
++
++
.
+
++
+
+
+
+
+ +
1031
.
.
_
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
zone
Methods.
.
.
.
.
.
.
.
.
.
.
.
-
-
-
-
-
-
-
-
+
-
.
(+)
( < 3 mm);
.
-
-
-
-
+
1033
+
+
+
-
++
-
+
+
+
+
+
+
+
-
-
+
-
_
i
clear
Listeria
1040
+ +
.
selected
(+)
.
against
1032
bacteria
inhibition
"
of coryneform
monocytogenes
1001
L
filtrates
101
of culture
of strains
nucleogenes
b Symbols:
•
A.
A. nicotianae
B. linens
Antagonistic
TABLE
.
.
.
.
++
++
+
+
-
-
.
-
.
.
.
.
.
zone
+ +
++
++
++
++
-
++
++
+
++
+
"
(agar
.
.
.
.
.
.
.
.
.
.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
+
_
-
+
--
-
-
-
-
-
-
-
-
-
-
(+)
-
-
-
_
inhibition
ivanovii "
3009
L
assay)
( + ) = unclear
2406
diffusion
( > 3 ram);
-
-
-
-
+
+
+
+ +
+
+
+
+
+
+
+
+
+
_
2023
innocua
strains
2011
L
and
inhibition
+ +
++
+
+
++
++
-
+ +
+ +
+ +
+ +
+ +
++
++
++
++
++
++
++
++
++
+ +
1440
species
L
seeligeri "
+
+
+
zone.
+
+
+
+ +
++
++
+
++
+ +
+ +
+ +
+ +
+ +
++
++
++
++
++
++
++
++
++
+ +
4007
126 O. 35 [ L~ m._onocytog_,ene.._sss(I031 ) ~
control; no addition of culture f i l t r a t e
O" 30 ] -~9- B tinens (R01)
[]
A nicottanae ( G O 5 ) ~
~0.20
~
~~_~~_~.~ .
÷
o.~o/
015
O. 05
oI 6
8
10
12
14
16
18.
20
22
24
26
28
30
32
34
36
T i m e (h) at 30°C
Fig. 2. Inhibitory effect of sterile culture filtrates of various bacteria on the growth of L. raonocytogenes
(1031) in liquid media. Culture filtrates of A. nicotianae and A. nucleogenes were considerably more inhibitory of L. innocua and L. ioanooii. Only 38-62% of L. monocytogenes were inhibited. Within the B. linens strains, it appeared that a higher proportion of L. monocytogenes was inhibited by B. linens than L. innocua. The sterile filtrates had varying levels of activity. Of particular interest are the culture filtrates of B. linens M10, M15 and M18, which are active against a high proportion of strains of L. rnonocytogenes, L. innocua and L. ioanovii investigated. None of the test strains included in Table III were inhibitory to L. welshimeri, L. grayi, L. rnurravi or Jonesia denitrificans (data not shown). Two of the three strains of L. seeligeri tested were, however, inhibited by all 23 coryneforms (data not shown). The Listeria strains investigated showed varying sensitivity to the sterile culture filtrates of coryneform bacteria (Table IV). Furthermore, 11 L. rnonocytogenes, one L. ivanovii and one L. seeligeri strains were extremely sensitive to the culture filtrates, in being inhibited by 96% of the 23 tested filtrates (results not shown). In contrast, three L. monocytogenes and two each of L. innocua and L. ivanovii strains showed almost total resistance to 22 out of the 23 culture filtrates (results not shown). The inhibitory effect of the various sterile culture filtrates on growth of L. monocytogenes (1031) is depicted in Fig. 2. Similar inhibitory patterns were also observed with respect to two other strains of L. monocytogenes (data not given). Further incubation up to 96 h did not result in any increase in listerial growth. N o inhibition, however, was observed with B. linens (R01). The effect of dilution on the inhibitory properties of a 72-h culture filtrate of B. linens (M21) against L. monocytogenes (1031) is presented in Fig. 3. Additional tests
127 0.30
L. monocyto~enes (1031) control; no oddition of c u l t u r e t i l t r o t e
0.25
~
Culture f i l t r o t e (1:2)
[]
--t-- C u l t u r e f l i t rote (1 : 1.5) c o
0.20
+
C u l t u r e f i l t r o t e (1:1,
~
Culture f i l t r o t e (1:0.1)
/ /
~u ~ " l = [ " ~ - D
p /
I'1 I/_/
~
~u 0.~5 £ o
f
~
#k-,---__._~
I
I
28
30
o
