www.elsevier.com/locate/vetmic

Veterinary Microbiology 106 (2005) 61–71

Production of bacteriocins by coagulase-negative

staphylococci involved in bovine mastitis

Janaına dos Santos Nascimento a, Patricia Carlin Fagundes a,Maria Aparecida Vasconcelos de Paiva Brito b, Katia Regina Netto dos Santos a,

Maria do Carmo de Freire Bastos a,*

a Instituto de Microbiologia Prof. Paulo de Goes, Departamento de Microbiologia Geral, Universidade do Brazil (UFRJ),

CCS, Bloco I, sala I-1-059, Cidade Universitaria, 21941-590 Rio de Janeiro, RJ, Brazilb EMBRAPA Gado de Leite, Juiz de Fora, MG, Brasil

Received 8 June 2004; received in revised form 20 September 2004; accepted 10 October 2004

Abstract

In the present study, 188 coagulase-negative Staphylococcus (CNS) strains were isolated from bovine mastitis cases from 56

different Brazilian dairy herds, located in the Southeast region of the country, and were tested for antimicrobial substance

production. Twelve CNS strains (6.4%) exhibited antagonistic activity against a Corynebacterium fimi indicator strain. Most

antimicrobial substances were sensitive to proteolytic enzymes suggesting that they might be bacteriocins (Bac). Amongst the

CNS producers, six were identified as S. epidermidis, two as S. simulans, two as S. saprophyticus, one as S. hominis and one as S.

arlettae. Plasmid profile analysis of these strains revealed the presence of at least one plasmid. The Bac+ strains presented either

no or few antibiotic resistance phenotypes. Three strains were shown to produce a bacteriocin either identical or similar to

aureocin A70, a bacteriocin previously isolated from an S. aureus strain isolated from food. The remaining Bac+ strains produce

antimicrobial peptides that seem to be distinct from the best characterised staphylococcal bacteriocins described so far. Some of

them were able to inhibit Listeria monocytogenes, an important food-borne pathogen, and several strains of Streptococcus

agalactiae associated with bovine mastitis, suggesting a potential use of these bacteriocins either in the prevention or in the

treatment of streptococcal mastitis.

# 2004 Elsevier B.V. All rights reserved.

Keywords: Coagulase-negative Staphylococcus spp.; Bacteriocin; Bovine mastitis; Cattle-bacteria

1. Introduction

Bovine mastitis is an inflammation of the

mammary glands usually due to a microbial infection

* Corresponding author. Fax: +55 21 2560 8344.

E-mail address: mcbastos@micro.ufrj.br (M.C.F. Bastos).

0378-1135/$ – see front matter # 2004 Elsevier B.V. All rights reserved

doi:10.1016/j.vetmic.2004.10.014

that affects milk production and quality, being one of

the most significant causes of economic loss to the

dairy industry (Soltys and Quinn, 1999; Sordelli et al.,

2000; Riffon et al., 2001; Bradley, 2002).

Although several bacterial pathogens can cause

mastitis, Staphylococcus aureus is the most important

causative agent (followed by Streptococcus spp.), and

.

J.S. Nascimento et al. / Veterinary Microbiology 106 (2005) 61–7162

once it is established in the mammary glands of the

animal, it is very difficult to eradicate (Soltys and Quinn,

1999; Lammers et al., 2001). Coagulase-negative

staphylococci (CNS) have been considered a minor

pathogen of bovine mastitis. However, many studies

have recently shown the importance of CNS infection of

the bovine mammary glands (Soltys and Quinn, 1999;

Zhang and Maddox, 2000; Younis et al., 2003).

In the efforts to avoid mastitis, some vaccines that

can reduce the severity of this illness were generated.

However, these vaccines still do not control efficiently

the development of mastitis (Leitner et al., 2003). On

the other hand, it has been verified that the

indiscriminate treatment with antibiotics, without

either a technical prescription or identification tests

of the pathogen, can contribute to an increased

resistance of these micro-organisms, making the cure

of mastitis still more difficult (Gruet et al., 2001).

Moreover, it is important to stand out that these

medicines are set free in milk for some days, after their

administration, being able to cause problems for

consumers. Thus, the identification of alternative

methods for controlling this illness is essential. One of

these methods could be the use of bacteriocins.

Bacteriocins (Bac) are antimicrobial peptides or

proteins produced by bacteria with inhibitory activity

against other bacteria. These substances, specially

those produced by Gram-positive bacteria, present a

potential use either in food industries as biopreserva-

tives or in the prevention and treatment of some

infectious diseases, having medical and veterinary

applications (Jack et al., 1995; Oliveira et al., 1998b;

Sahl and Bierbaum, 1998).

Amongst the bacteriocin-producer staphylococcal

species, S. epidermidis is the most studied. Some of its

bacteriocins are very well characterised, such as Pep5,

epidermin, epicidin 280 and epilancin K7, which are

classified as lantibiotics. Pep5 is a 34-amino acid

peptide encoded by the structural gene pepA (Kaletta

et al., 1989). Epidermin is also a well-characterised

lantibiotic produced by S. epidermidis Tu3298. Its

structural gene, epiA, codes for a 22-amino acid

peptide (Augustin et al., 1992). Epicidin 280 is also a

30-amino acid lantibiotic, produced by S. epidermidis

BN280, which has 75% of homology with Pep5

(Heidrich et al., 1998). Epilancin K7, a 31-amino acid

lantibiotic produced by S. epidermidis K7, is encoded

by the gene elkA (van de Kamp et al., 1995).

Our laboratory has been investigating bacteriocin

production by Staphylococcus strains, which is

generally associated to plasmids. Amongst the

bacteriocins produced by S. aureus and studied by

our group, aureocins A70 and A53 are the best

characterised. Aureocin A70 is a multi-peptide, non-

lantibiotic bacteriocin produced by S. aureus A70

which is encoded by an 8.0 kb-plasmid named pRJ6

(Giambiagi-deMarval et al., 1990). Aureocin A70 was

the first bacteriocin described which is composed of

four related small cationic peptides, which are

encoded by the aurABCD operon (Netz et al.,

2001). Aureocin A53 is a 51-amino acid peptide

produced by S. aureus A53. It is encoded on a 10.4 kb

plasmid, pRJ9 (Netz et al., 2002).

As well as other antimicrobial substances, bacter-

iocins produced by staphylococci (staphylococcins)

could be employed, in their purified form, on the

control of bovine mastitis. Therefore, in the present

study, in an attempt to detect new bacteriocins with

potential medical and veterinary applications, we

extended our studies on bacteriocin production to CNS

strains, which were associated with bovine mastitis in

different Brazilian dairy herds.

2. Materials and methods

2.1. Bacterial strains and culture conditions

One hundred and eighty-eight coagulase-negative

Staphylococcus strains involved in bovine mastitis

were isolated from 56 different Brazilian dairy herds

located in the Southeast region of the country, and

used in the screening for bacteriocin production.

Seventy-four Streptococcus agalactiae strains were

also isolated from bovine mastitis cases and used as

indicators. These strains were isolated according to

procedures recommended by Harmon et al. (1990).

S. aureus and S. epidermidis strains from previous

studies (Table 1) were used either as producers or as

indicators in the inhibition assays. Staphylococcus

strains were grown in either TSB (Difco) or BHI

(Difco), at 37 8C for 18 h. TSB was used to grow the

strains for DNA isolation and BHI was used in all

bacteriocin assays. The bacteria used as indicators

were grown in BHI medium, except for lactic acid

bacteria, that were cultivated in MRS medium (BBL)

J.S. Nascimento et al. / Veterinary Microbiology 106 (2005) 61–71 63

Table 1

Staphylococcus strains previously described and used in this study

Strain Relevant features Reference/source

Staphylococcus aureus

A53 Bac+ (aureocin A53), Imm+, plasmid: pRJ9 (10.4 kb) Giambiagi-deMarval et al. (1990)

A70 Bac+ (aureocin A70), Imm+, plasmid: pRJ6 (8.0 kb) Giambiagi-deMarval et al. (1990)

A70 Bac� Strain A70 cured of pRJ6 Giambiagi-deMarval et al. (1990)

Staphylococcus epidermidis

Tu 3298 Bac+ (epidermin) Augustin et al. (1992)

5 Bac+ (Pep5) Kaletta et al. (1989)

5 (cured) Bac� Pep5� Bierbaum et al. (1994)

BN280 Bac+ (epicidin 280) Heidrich et al. (1998)

K7 Bac+ (epilancin K7) van de Kamp et al. (1995)

Bac, bacteriocin; Imm, immunity.

and incubated at 28 8C for 48 h under microaerophilic

atmosphere. Bacteria were stored in their appropriate

culture medium with 40% glycerol (w/v) at �20 8Cuntil needed. When necessary, the media were

supplemented with agar at 1.5% (w/v) or 0.6% (w/v).

2.2. Assay for antimicrobial substance production

This assay was done as described previously by

Giambiagi-deMarval et al. (1990). C. fimi NCTC 7547

was used as the indicator strain for production of

antimicrobial substances.

2.3. Identification of the strains

Only the CNS strains that exhibited antimicrobial

substance production were identified, using conven-

tional biochemical tests (Kloos and Schleifer, 1986)

or, when necessary, using a commercial kit for

identification (API Staph, BioMerieux). S. agalactiae

strains were identified on the basis of Gram staining,

presence of hemolysis, negative results on tests for the

presence of catalase and esculin hydrolysis, and

positive results on tests for CAMP and hydrolysis of

sodium hypurate.

2.4. Effects of proteolytic enzymes and of 0.2N

NaOH on AMS activity

The effects of trypsin (Sigma), proteinase K

(Boehinger Mannheim), protease XXIII (Sigma)

and 0.2N NaOH on AMS activity were determined

by the method described by Giambiagi-deMarval et al.

(1990). Lack of inhibition zones when C. fimi

NCTC7547 was used as the indicator strain indicated

that the antimicrobial compound had either a

proteinaceous or an acidic nature.

2.5. Antibiotic resistance

Resistance patterns were determined by disc

diffusion on Mueller-Hinton agar (Oxoid), according

to the National Committee for Clinical Laboratory

Standards (NCCLS, 2002). The following antibiotics

(Sensifar) were used: ampicillin (10 mg), cephalothin

(30 mg), ciprofloxacin (5 mg), clindamycin (2 mg),

chloramphenicol (30 mg), erythromycin (15 mg),

gentamicin (10 mg), imipenen (10 mg), mupirocin

(5 mg), oxacillin (1 mg), penicillin (10 U), rifampin

(5 mg), tetracycline (30 mg) and vancomycin (30 mg).

The diameters of the inhibition zones were interpreted

according to the NCCLS guidelines after 24 h of

incubation at 37 8C. Strains which showed resistance

to b-lactamic drugs were also tested for the presence

of the methicillin-resistance gene (mecA) by PCR.

2.6. Isolation of plasmid DNA

Whole-cell lysates were prepared as described by

Giambiagi-deMarval et al. (1990).

2.7. DNA–DNA hybridisation assays

Southern blots and hybridisations were performed

as described by Bastos and Murphy (1988). The 5.2-kb

HindIII-A fragment of pRJ6 (encompassing all genes

J.S. Nascimento et al. / Veterinary Microbiology 106 (2005) 61–7164

required for aureocin A70 production) was used as a

probe. 32P-labelled probes were prepared by the

Random-Primers DNA Labelling System (Invitrogen)

following the manufacturer’s recommendations.

2.8. PCR amplification

Total DNA was extracted by boiling, as described

previously by Nunes et al. (1999). The primers used

for amplification of the aurABCD operon and the

genes aucA, eciA and elkA (the structural genes of

aureocin A70, aureocin A53, epicidin 280 and

epilancin K7, respectively) as well as for amplification

of the methicillin-resistance gene (mecA) and genes

orfA and orfB (presumably involved in the immunity

to aureocin A70) are listed in Table 2. Each PCR

reaction contained: 1� PCR reaction buffer (Invitro-

gen), 2.5 mM concentration of each deoxyribonucleo-

side triphosphate, 2.5 U of Taq-Polymerase (Invi-

trogen) and 50 pmol of each primer. The amplification

was done in a Programmable Thermal Controller

(PTC-100TM, MJ Research, USA) and the cycles were

performed as described by Nunes et al. (1999). A 100-

bp DNA ladder (Invitrogen) was used as the molecular

size marker.

2.9. Determination of the inhibitory spectrum

of each AMS+ strain

To determine the inhibitory spectrum of each

AMS, several strains from different species of Gram-

positive bacteria were tested as indicators. The

Table 2

Primers used in this work

Gene(s) to be amplified Relevant features Primers

aurABCD Aureocin A70 structural genes Forward: P

reverse: P

aucA Aureocin A53 structural gene Forward: A

reverse: A

orfA and orfB Immunity genes to aureocin A70 Forward: R

reverse: R

elkA Epilancin K7 structural gene Forward: K

reverse: K

eciA Epicidin 280 structural gene Forward: E

reverse: E

mecA Methicillin-resistance gene Forward: M

reverse: M

inhibitory action of the AMS produced by the

CNS strains against S. agalactiae involved in

bovine mastitis was also tested using 74 indicator

strains.

2.10. Agarose gel electrophoresis

Agarose gel electrophoreses were performed as

described by Sambrook et al. (1989). Products

amplified by PCR were analysed in 1.4% (w/v)

agarose gels and the plasmid profiles were determined

in 0.7% (w/v) agarose gels. The gels were stained in an

aqueous solution containing ethidium bromide

(0.5 mg/ml) and visualised on a UV transiluminator.

3. Results

3.1. Detection and identification of the AMS+ strains

One hundred and eighty-eight coagulase-negative

Staphylococcus (CNS) strains were isolated from

bovine mastitis cases from 56 different Brazilian dairy

herds, located in the Southeast region of the country,

and were tested for AMS production. Twelve CNS

strains (6.4%), belonging to six different herds,

exhibited antimicrobial substance production against

C. fimi (inhibition zones between 17 and 38 mm).

Among the CNS producers, six were identified as S.

epidermidis, two as S. simulans, two as S. saprophy-

ticus, one as S. hominis and one as S. arlettae

(Table 3).

Size (bp)

of the fragment

to be amplified

4B, 50-CCTTATAACTTCGAATGCT-30;5, 50-AATTATTAACAAGAGAAA-30

525

UC1, 50-GAAGTTGTGAAAACTATTA-30;UC2, 50-CATAAAACAAAGAGCCAAAGT-30

322

J61, 50-GACGAGGGTATTGCATA-30;J62, 50-CTAAGCATTCGATAAGG-30

722

71, 50-ATGAATAACTCATTATTC-30;72, 50-ATGGAA AACAAAAAAG-30

171

pin1, 50-CAGGAG GGATATATTATGG-30;pin2, 50-CAATCACTACTATTGACAATCAC-30

165

RS1, 50-TAGAAATGA CTG AACGTCCG-30;RS2, 50-TTGCGATCAATGTTACCG TAG-30

154

J.S. Nascimento et al. / Veterinary Microbiology 106 (2005) 61–71 65

Table 3

Characteristics of the bacteriocin-producer CNS strains

Strains Identification Sensitivity of the Bac to proteolytic enzymes Resistance profile Plasmid formsa

Protease XXIII Proteinase K Trypsin

A70b S. aureus S S S � 8.0 (pRJ6)

494 S. epidermidis R S S Emc >25; 15; 6.7; 5.5; 3.7; 3.2; 2.2

2166 S. epidermidis R S S � 12; 6.2

2167 S. epidermidis S R R Emc 12; 6.2

2457 S. simulans S R S Emc >27

3024 S. epidermidis S S S Ap, Ce, Gm, Ox, Pc, Tc 15; 8.1; 4.5

3154 S. saprophyticus R S S Emc 15; 8

3299 S. simulans R S S � >25; 3.5

3414 S. epidermidis R S S � >25

3419 S. hominis R R R � 4

3528 S. arlettae S S S Ap, Pc 15; 8

3576 S. saprophyticus S S S Ap, Pc 15; 8

3577 S. epidermidis S R R � >25 (2)d; 2.5; 1.5

R, resistant; S, sensitive; (�) sensitive to all the antibiotics tested; Ap, ampicillin; Ce, cephalothin; Em, erythromycin; Gm, gentamicin; Ox,

oxacillin; Pc, penicillin; Tc, tetracycline.a Size in kb.b Strain A70 was included in these experiments as a positive control.c Intermediate resistance.d The number between parentheses indicates the presence of two plasmid bands with size larger than 25 kb.

3.2. Effects of proteolytic enzymes on AMS activity

The results are also presented in Table 3. The AMS

were resistant to 0.2N NaOH, discarding the

possibility that the inhibition exhibited was due to

organic acids produced by the producer strain during

its metabolism. Except for the AMS produced by the

strain 3419, the remaining AMS were sensitive to at

least one proteolytic enzyme tested, indicating that

these AMS present a biological active proteinaceus

component in their structure, the main characteristic

of a typical bacteriocin. Therefore, from hereafter, the

AMS+ strains will be considered Bac+. Bacteriocins

produced by strains 3154, 3528 and 3576, as well as

aureocin A70, showed a high sensitivity to proteolytic

enzymes, being sensitive to all enzymes tested.

3.3. Antibiotic resistance profile

The Bac+ strains did not exhibit a great number of

resistance markers, except for strain 3024, that showed

resistance to six antibiotics (Table 3). The remaining

strains were resistant to either two (ampicillin and

penicillin) or none of the antibiotics tested. Strains

494, 2167, 2457 and 3154 presented intermediate

resistance to erythromycin. Strains that showed

resistance to one or more b-lactamic drugs were also

tested for the presence of the mecA gene, which

encodes methicillin resistance. Among the strains

tested, only strain 3024 showed amplification of the

mecA gene (data not shown).

3.4. Plasmid profiles

All strains showed at least one plasmid form

(Table 3). Strains 3154 (S. saprophyticus), 3528 (S.

arlettae) and 3576 (S. saprophyticus) exhibited an

identical plasmid profile, and, interestingly, they carry

a plasmid with a size similar to that of pRJ6 (8.0 kb), a

bacteriocinogenic plasmid previously characterised

by our group and that encodes aureocin A70. These

three strains were isolated from the same herd. Strains

2166 and 2167 (both S. epidermidis and isolated from

the same herd), also presented an identical plasmid

profile.

3.5. DNA–DNA hybridisation assays

Since strains 3154, 3528 and 3576 carry a plasmid

with a size similar to that of pRJ6, the homology

between these plasmids and pRJ6 was investigated. A

strong signal was detected to the 8.0-kb plasmids

J.S. Nascimento et al. / Veterinary Microbiology 106 (2005) 61–7166

carried by strains 3154, 3528 and 3576. No

hybridisation was observed in the remaining CNS

strains.

3.6. Amplification of the aurABCD and orfAB

operons

The presence or the absence of the aureocin A70

structural genes was also confirmed by PCR using

specific primers from sequences flanking operon

aurABCD. The host strain of pRJ6, S. aureus A70,

was used as a positive control. Strains 3154, 3528 and

3576 had a 525-bp fragment amplified. The remaining

nine strains did not present amplification of this

fragment (Fig. 1). Additionally, PCR experiments

using primers to amplify the genes orfA and orfB, that

seem to be involved in the immunity to aureocin A70,

revealed that only strains 3154, 3528 and 3576 also

had the 722 bp-fragment amplified (data not shown).

3.7. Spectrum of action of the bacteriocins produced

by the CNS strains

Strains 3154, 3528, 3576 and 3577 exhibited a

spectrum of activity wider than the other strains

(Table 4). Interestingly, the spectrum exhibited by

Fig. 1. PCR amplification of the aurABCD operon which encodes aureoc

control); (C) 494; (D) 2166; (E) 2167; (F) 2457; (G) 3024; (H) 3154; (I)

these strains was very similar to that exhibited by S.

aureus A70, inhibiting all Listeria strains tested, an

important food-borne and bovine mastitis associated

pathogen, and Micrococcus. S. agalactiae, another

common pathogen involved in bovine mastitis, was

inhibited by strains 3154, 3528, 3576 and 3577.

Strains 2166, 2167 and 3419 inhibited only C. fimi.

3.8. Cross-immunity against bacteriocins produced

by S. aureus and S. epidermidis

Protection of the bacteriocin-producer strains

against their own bacteriocins is mediated by the

so-called immunity peptides whose genetic determi-

nants are generally found in the bacteriocin gene

clusters. Immunity is generally bacteriocin specific.

Strains that produce either identical or similar

bacteriocins exhibit cross-immunity. Therefore, the

presence of cross-immunity between bacteriocin

producers is generally indicative of relatedness

between their bacteriocins.

The results of the cross-immunity tests performed

in this study are shown in Table 5. Strains 3154, 3528

and 3576 were able to inhibit the growth of strain A70

Bac�, cured of the plasmid pRJ6, but not of the wild-

type strain; strain A70 did not inhibit strains 3154,

in A70. Lanes: (A) DNA size marker; (B) S. aureus A70 (positive

3299; (J) 3414; (K) 3419; (L) 3528; (M) 3576; (N) 3577.

J.S. Nascimento et al. / Veterinary Microbiology 106 (2005) 61–71 67

Table 4

Spectrum of activity of the bacteriocins produced by coagulase-negative Staphylococcus strains

Indicator strains Producer strains

A70 494 2166 2167 2457 3024 3154 3299 3414 3419 3528 3576 3577

Bacillus megaterium F4 + � � � � � � � � � � � �Corynebacterium fimi NCTC 7547 + + + + + + + + + + + + +

Enterococcus faecium E86 NT � � � � � � � � � � � +

Lactobacillus casei t � � � � � � � � � � � �Lactobacillus casei ATCC 393 t � � � � � � � � � � � �Lactococcus lactis ATCC 11454 t � � � � � � � � � � � �Leuconostoc lactis ATCC 19256 t � � � � � � � � � � � �Listeria monocytogenes 11LM + � � � � � + � � � + + +

Listeria monocytogenes L1/2A + + � � � � + � � � + + +

Listeria monocytogenes 7848 + � � � � � + � � � + + +

Listeria innocua 397 + � � � � � + � � � + + +

Micrococcus spp. + � � � + � + + � � + + +

Paenibacillus polymyxa SCE-2 + � � � � � � � � � � � �Pediococcus pentosaceus ATCC 43200 � � � � � + � � + � � + �Streptococccus pyogenes � � � � � � � + � � � � �Streptococcus agalactie � � � � � � + � � � + + +

(+) Inhibition; (�) no inhibition; NT, not tested; t, turbid zone of inhibition. Only the indicator strains that were inhibited by at least one of the

producer strains are shown in this table. Enterococcus faecalis FA2-2, E. faecalis OG1X, Lactobacillus brevis ATCC 14869 and Leuconostoc

mesenteroides ATCC 8293 were also tested, however, they were not inhibited by any of the producer strains.

3528 and 3576 suggesting the presence of cross-

immunity among the bacteriocins produced by these

four strains. Strain 3577 was also able to inhibit strain

A70 Bac� but not A70. However, strain A70 inhibited

strain 3577, suggesting that the bacteriocins produced

Table 5

Immunity/resistance to bacteriocins among bacteriocinogenic CNS and S

Indicator strains Producer strains

494 2166 2167 2457 3024

S. aureus A70 � � � � �S. aureus A70 Bac� � � � � �S. aureus A53 � � � � �S. epidermidis Tu 3298 � � � � �S. epidermidis 5 � � � � �S. epidermidis BN280 � � � � �S. epidermidis K7 � � � � �

Producer strains Indicator strains

494 2166 2167 2457

S. aureus A70 (aureocin A70) � � + +

S. aureus A53 (aureocin A53) � � � +

S. epidermidis Tu 3298 (epidermin) + + + +

S. epidermidis 5 (Pep5) + + + +

S. epidermidis BN280 (epicidin 280) � � � �S. epidermidis K7 (epilancin K7) � � � �(+) Inhibition; (�) no inhibition.

by both strains are different. Strain 3299 inhibited only

strain A53, indicating that its bacteriocin is distinct

from aureocin A53. The producer strains of Pep5 and

epidermin were able to inhibit all the 12 CNS strains

associated with bovine mastitis, suggesting that none

. aureus strains

3154 3299 3414 3419 3528 3576 3577

� � � � � � �+ � � � + + +

+ + � � + + �� � � � � � �� � � � � � �� � � � � � �� � � � � � �

3024 3154 3299 3414 3419 3528 3576 3577

� � + � � � � +

� + � � � + + �+ + + + + + + +

+ + + + + + + +

� � � � � � � �� � � � � � � �

J.S. Nascimento et al. / Veterinary Microbiology 106 (2005) 61–7168

Fig. 2. Inhibition of Streptococcus agalactiae strains involved in bovine mastitis by Bac+ coagulase-negative Staphylococcus strains. Seventy-

four S. agalactiae strains involved in bovine mastitis were tested as indicators. (&) Number of inhibited strains and (&) percentage of inhibited

strains.

SCN bacteriocin is related to Pep5 and epidermin. The

producers of epicidin 280 and epilancin K7, however,

were neither able to inhibit nor to be inhibited by any

of the CNS strains. In such cases, the absence of

inhibition could result from either bacteriocin

immunity or resistance (i.e, inability of the bacteriocin

to act on the cells).

3.9. Amplification of the structural genes for

aureocin A53, epilancin K7 and epicidin 280

To discard the possibility that the CNS bacteriocins

are related to aureocin A53, epilancin K7 and epicidin

280, PCR experiments for the amplification of the

structural genes of these staphylococcins were

performed and none of the 12 CNS strains involved

in bovine mastitis showed amplification of the

expected fragments (data not shown).

3.10. Inhibition of Streptococcus agalactiae strains

involved in bovine mastitis

Most of the S. agalactiae strains used as indicators

(78.4%) were inhibited by strain 3299 (Fig. 2). Strains

2166, 2167, 3154, 3419 and 3577 were able to inhibit

at least 40% of the indicators strains. The percentage

of inhibition observed with the remaining strains lied

between 10 and 38%.

4. Discussion

Previous studies have shown the production of

AMS by S. aureus strains from different sources.

Giambiagi-deMarval et al. (1990) found 13 (9.5%)

AMS producers amongst 137 strains isolated from

foods and not hospitalised patients. From 1990 to

1992, 163 S. aureus strains were isolated from

different clinical specimens of human origin, in four

hospitals from Rio de Janeiro. Fifty-three strains

(32.5%) were shown to produce AMS, but only three

(1.8%) presented a high antagonistic activity against

C. fimi (Gamon et al., 1999). Oliveira et al. (1998a),

studying 46 S. aureus strains isolated from apparently

healthy cattle, identified only four (8.7%) bacteriocin-

producer strains. Recently, Nascimento et al. (2002)

studied 50 strains of S. aureus isolated from mastitic

cows and verified that 24% of these strains showed a

high inhibitory activity against C. fimi. Therefore, a

higher incidence of AMS+ S. aureus strains was found

amongst the isolates involved in bovine mastitis. In the

present work, however, amongst 188 CNS, we found

only 12 (6.4%) producer strains. AMS produced by

most strains were sensitive to proteolytic enzymes

suggesting they might be bacteriocins. The only

exception was strain 3419. However, bacteriocins

resistant to proteolytic digestion has already been

reported in the literature (Netz et al., 2002). Although

bacteriocin production is not a characteristic found in

most Staphylococcus strains, it may confer a

competitive advantage to the producer strains for

nutrition in the surroundings, helping in the occupa-

tion of determined ecological niches.

In this study, most of the Bac+ CNS strains were

identified as S. epidermidis. The remaining strains

were S. simulans, S. saprophyticus, S. hominis and S.

arlettae. Bacteriocin production has already been

described in S. epidermidis (Sahl and Bierbaum,

J.S. Nascimento et al. / Veterinary Microbiology 106 (2005) 61–71 69

1998). However, to our knowledge, this is the first

report, which describes bacteriocin production in

these four latter staphylococcal species.

The resistance profile of the 12 CNS strains to 14

different antibiotics was also evaluated. In general, the

strains did not present a great number of resistance

phenotypes. These results contrast with those

described for clinical isolates of CNS which generally

carry multiple drug-resistance determinants (Aires de

Sousa et al., 1998; Kohner et al., 1999; Santos et al.,

1999). Only strain 3024 was shown to be resistant to

six of the antibiotics tested, mainly b-lactamic drugs.

The mecA gene, detected in this strain by PCR, is

probably involved in resistance to the latter drugs.

In Gram-positive bacteria, the genetic determinants

involved in the Bac production have been found either

on plasmids or on the bacterial chromosome (Jack

et al., 1995). Amongst the 12 Bac+ CNS strains

associated with bovine mastitis studied in this work,

all strains presented at least one plasmid form. Except

for S. hominis 3419, the genetic determinants

encoding the bacteriocins could be either on the

chromosome or on plasmids carried by them. In strain

3419, the bacteriocin genetic determinant is probably

chromosomally encoded since the only plasmid found

in this strain is too small to carry all genes generally

required for bacteriocin expression (Ennahar et al.,

2000; McAuliffe et al., 2001). Three strains (3154,

3528, 3576) were shown to possess a plasmid with a

size similar to that of pRJ6 (8.0 kb), an S. aureus

plasmid which encodes aureocin A70. PCR experi-

ments were then carried out to test if the CNS strains

presented the structural genes involved in aureocin

A70 production. The amplification of the expected

fragment of 525 bp was observed only in strains 3154,

3528 and 3576. These results indicate that the DNA of

these strains possesses similar sequences to that found

in the operon aurABCD. However, despite the

amplification of the 525 bp fragment, it is not possible

to assure that the amplified sequence is identical to

that of pRJ6. This could be confirmed by DGGE

(denaturing gradient gel electrophoresis) experiments,

whose technique allows the detection of differences of

even a single base-pair in DNA fragments with the

same size (Muyzer et al., 1998). These experiments

are currently in progress.

To confirm that the 8.0 kb plasmids found in strains

3154, 3528 and 3576 are involved in bacteriocin

production, experiments of plasmid cure or transfer

are required. However, hybridisation experiments with

the HindIII-A fragment of pRJ6, the amplification of

the bacteriocin operon as well as the amplification of

the genes orfA and orfB, probably involved in the

immunity to aureocin A70, are strong evidences that

these plasmids are responsible for bacteriocin produc-

tion, being either related or identical to pRJ6. For this

reason, a more detailed characterisation of these

plasmids was not performed. Previous studies

(Giambiagi-deMarval et al., 1990; Oliveira et al.,

1998a; Gamon et al., 1999; Nascimento et al., 2002)

have shown that bacteriocinogenic plasmids of 8.0 kb

are spread among the S. aureus population. However,

this is the first report on the presence of this plasmid

also among coagulase-negative staphylococci. Such

results suggest that there is a certain level of transfer of

plasmids related to pRJ6 among different species of

this genus. Taking together these results support the

conclusion that the bacteriocins produced by the CNS

strains 3154, 3528 and 3576 are either identical or

similar to aureocin A70.

In relation to the remaining strains, the immunity/

resistance and PCR experiments showed that they

produce bacteriocins which seem to be different from

the most studied staphylococcins. However, experi-

ments aiming the characterisation of these bacter-

iocins were not performed yet.

The bacteriocins produced by the CNS strains

presented a narrow spectrum of activity against Gram-

positive bacteria when compared to the bacteriocins

produced by the S. aureus strains previously studied

by our group (Giambiagi-deMarval et al., 1990;

Oliveira et al., 1998a; Gamon et al., 1999). Only four

strains (3154, 3528, 3576 and 3577) were able to

inhibit L. monocytogenes, an important food-borne

pathogen. However, when the ability of these strains to

inhibit S. agalactiae strains involved in bovine

mastitis was investigated, five CNS strains inhibited

at least 50% of the indicator strains. These results

suggest that the bacteriocins produced by some of

these CNS strains, especially by S. simulans 3299,

may be developed into useful antimicrobial drug for

either treatment or prevention of bovine mastitis

caused by Streptococcus, the second more important

pathogen associated to this disease. However, applied

studies must be done to confirm their effectiveness ‘‘in

vivo’’. These studies must be preceded by bacteriocin

J.S. Nascimento et al. / Veterinary Microbiology 106 (2005) 61–7170

purification and analysis of its biochemical structure.

Since the bacteriocin produced by S. simulans 3299

seems to be the best candidate for prevention of

streptococcal mastitis, the experiments aiming to its

purification and characterisation are currently in

progress.

Acknowledgements

We thank Dr. Hans-Georg Sahl for sending us the S.

epidermidis strains, which produce the bacteriocins

Pep5, epidermin, epilancin K7 and epicidin 280. This

work was supported by grants from CNPq, FAPERJ

and PRONEX to M.C.F.B.

References

Aires de Sousa, M., Sanches, I.S., Ferro, M.L., Vaz, M.J., Saraiva, Z.,

Tendeiro, T., Serra, J., Lencastre, H., 1998. Intercontinental spread

of a multidrug-resistant methicillin-resistant Staphylococcus aur-

eus (MRSA) clone. J. Clin. Microbiol. 36, 2590–2596.

Augustin, J., Rosenstein, R., Wieland, B., Schneider, U., Schnell, N.,

Engelke, G., Entian, K.-D., Gotz, F., 1992. Genetic analysis of

epidermin biosynthesis genes and epidermin-negative mutants of

Staphylococccus epidermidis. Eur. J. Biochem. 204, 1149–1154.

Bastos, M.C.F., Murphy, E., 1988. Transposon Tn554 encodes three

products required for transposition. EMBO J. 7, 2935–2941.

Bierbaum, G., Reis, M., Szekat, C., Sahl, H.-G., 1994. Construction

of an expression system for mutagenesis of the lantibiotic Pep5.

Appl. Environ. Microbiol. 60, 2876–2883.

Bradley, A.J., 2002. Bovine mastitis: an evolving disease. Vet. J.

164, 116–128.

Ennahar, S., Sashihara, T., Sonomoto, K., Ishizaki, A., 2000. Class

IIa bacteriocins: biosynthesis, structure, and activity. FEMS

Microbiol. Rev. 24, 85–106.

Gamon, M.R., Moreira, E.C., Oliveira, S.S., Teixeira, L.M., Bastos,

M.C.F., 1999. Characterization of a novel bacteriocin-encoding

plasmid found in clinical isolates of Staphylococcus aureus.

Antonie van Leeuwenhoek 75, 233–243.

Giambiagi-deMarval, M., Mafra, M.A., Penido, E.G.C., Bastos,

M.C.F., 1990. Distinct groups of plasmids correlated with

bacteriocin production in Staphylococcus aureus. J. Gen. Micro-

biol. 136, 1591–1599.

Gruet, P., Maincent, P., Berthelot, X., Kaltsatos, V., 2001. Bovine

mastitis and intramammary drug delivery: review and perspec-

tives. Adv. Drug Deliv. Rev. 50, 245–259.

Harmon, R.J., Eberhart, R.J., Jasper, D.E., et al., 1990. Micro-

biological Procedures for the Diagnosis of Bovine Udder Infec-

tion. National Mastitis Council, Arlington, 34 pp.

Heidrich, C., Pag, U., Josten, M., Metzger, J., Jack, R.W., Bierbaum,

G., Jung, G., Sahl, H.-G., 1998. Isolation, characterization, and

heterologous expression of the novel lantibiotic epicidin 280 and

analysis of its biosynthetic gene cluster. Appl. Environ. Micro-

biol. 64, 3140–3146.

Jack, R.W., Tagg, J.R., Ray, B., 1995. Bacteriocins of Gram-positive

bacteria. Microbiol. Rev. 59, 171–200.

van de Kamp, M., van den Hooven, H.W., Konings, R.N., Bierbaum,

G.A., Sahl, H.-G., Kuipers, O.P., Siezen, R.J., de Vos, W.M.,

Hilbers, C.W., van de Ven, F.J., 1995. Elucidation of the primary

structure of the lantibiotic epilancin K7 from Staphylococccus

epidermidis K7 Cloning and characterisation of the epilancin-

K7-encoding gene and NMR analysis of mature epilancin K7.

Eur. J. Biochem. 230, 587–600.

Kaletta, C., Entian, K.-D., Kellner, R., Jung, G., Reis, M., Sahl, H.-

G., 1989. Pep5, a new lantibiotic: structural gene isolation and

prepeptide sequence. Arch. Microbiol. 152, 16–19.

Kloos, W.E., Schleifer, K.H., 1986. Genus Staphylococcus. In:

Sneath, P.H.A., Mair, N.S., Sharpe, M.E., Holt, J.G. (Eds.),

Bergey’s Manual of Systematic Bacteriology, vol. 2. Williams

& Wilkins, Baltimore, MD, pp. 1013–1035.

Kohner, P., Uhl, J., Kolbert, C., Persing, D., Cockerill, F., 1999.

Comparison of susceptibility testing methods with mecA gene

analysis for determining oxacillin (methicillin) resistance in

clinical isolates of Staphylococcus aureus and coagulase-nega-

tive Staphylococcus spp. J. Clin. Microbiol. 37, 2952–2961.

Lammers, A., van Vorstenbosch, C.J., Erkens, H.F., Smith, H.E.,

2001. The major bovine mastitis pathogens have different cell

tropisms in cultures of bovine mammary gland cells. Vet.

Microbiol. 80, 255–265.

Leitner, G., Lubashevsky, E., Trainin, Z., 2003. Staphylococcus

aureus vaccine against mastitis in dairy cows, composition

and evaluation of its immunogenicity in a mouse model. Vet.

Immunol. Immunopathol. 93, 159–167.

McAuliffe, O., Ross, R.P., Hill, C., 2001. Lantibiotics: structure,

biosynthesis and mode of action. FEMS Microbiol. Rev. 25,

285–308.

Muyzer, G., Brinkhoff, T., Nubel, U., Santegoeds, C., Schafer, H.,

Wawer, C., 1998. Denaturing gradient gel electrophoresis

(DGGE) in microbial ecology. Mol. Microbiol. Ecol. 3, 1–27.

Nascimento, J.S., Santos, K.R.N., Gentilini, E., Sordelli, D., Bastos,

M.C.F., 2002. Phenotypic and genetic characterisation of bac-

teriocin-producing strains of Staphylococcus aureus involved in

bovine mastitis. Vet. Microbiol. 95, 133–144.

National Committee for Clinical Laboratory Standards (NCCLS),

2002. Performance Standards for Antimicrobial Sensitivity

Testing. NCCLS, Wayne, PA.

Netz, D.J.A., Sahl, H.-G., Marcolino, R., Nascimento, J.S., Oliveira,

S.S., Soares, M.B., Bastos, M.C.F., 2001. Molecular character-

isation of aureocin A70, a multi-peptide bacteriocin from Sta-

phylococcus aureus. J. Mol. Biol. 311, 939–949.

Netz, D.J.A., Pohl, R., Beck-Sickinger, A.G., Selmer, T., Pierik,

A.J., Bastos, M.C.F., Sahl, H.-G., 2002. Biochemical character-

isation and genetic analysis of aureocin A53, a new, atypical

bacteriocin from Staphylococcus aureus. J. Mol. Biol. 319, 745–

756.

Nunes, E.L.C., Santos, K.R.N., Mondino, P.J.J., Bastos, M.C.F.,

Giambiagi-deMarval, M., 1999. Detection of ileS-2 gene encod-

ing mupirocin resistance in methicillin-resistant Staphylococcus

J.S. Nascimento et al. / Veterinary Microbiology 106 (2005) 61–71 71

aureus by multiplex PCR. Diagn. Microbiol. Infect. Dis. 34, 77–

81.

Oliveira, S.S., Povoa, D.C., Nascimento, J.S., Pereira, M.S.V.,

Siqueira Jr., J.P., Bastos, M.C.F., 1998a. Antimicrobial sub-

stances produced by Staphylococcus aureus strains isolated from

cattle in Brazil. Lett. Appl. Microbiol. 27, 229–234.

Oliveira, S.S., Abrantes, J., Cardoso, M., Sordelli, D., Bastos,

M.C.F., 1998b. Staphylococcal strains involved in bovine mas-

titis are inhibited by Staphylococcus aureus antimicrobial pep-

tides. Lett. Appl. Microbiol. 27, 287–291.

Riffon, R., Sayasith, K., Khalil, H., Dubreuil, P., Drolet, M., Lagace,

J., 2001. Development of a rapid and sensitive test for identifica-

tion of major pathogens in bovine mastitis by PCR. J. Clin.

Microbiol. 39, 2584–2589.

Sahl, H.-G., Bierbaum, G., 1998. Lantibiotics: biosynthesis and

biological activities of uniquely modified peptides from Gram-

positive bacteria. Annu. Rev. Microbiol. 52, 41–79.

Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning:

A Laboratory Manual. Cold Spring Harbor Laboratory Press,

New York.

Santos, K.R.N., Teixeira, L.M., Leal, G.S., Fonseca, L.S., Gontijo

Filho, P.P., 1999. DNA typing of methicillin-resistant Staphy-

lococcus aureus: isolates and factors associated with nosocomial

acquisition in two Brazilian university hospitals. J. Med. Micro-

biol. 48, 17–23.

Soltys, J., Quinn, M.T., 1999. Selective recruitment of T-cell subsets

to the udder during staphylococcal and streptococcal mastitis:

analysis of lymphocyte subsets and adhesion molecule expres-

sion. Infect. Immun. 67, 6293–6302.

Sordelli, D.O., Buzzola, F.R., Gomez, M.I., Steele-Moore, L., Berg,

D., Gentilini, E., Catalano, M., Reitz, A.J., Tollersrud, T., Dena-

miel, G., Jeric, P., Lee, J.C., 2000. Capsule expression by bovine

isolates of Staphylococcus aureus from Argentina: genetic and

epidemiologic analyses. J. Clin. Microbiol. 38, 846–850.

Younis, A., Krifucks, O., Heller, E.D., Samra, Z., Glickman, A.,

Saran, A., Leitner, G., 2003. Staphylococcus aureus exosecre-

tions and bovine mastitis. J. Vet. Med. 50, 1–7.

Zhang, S., Maddox, C.W., 2000. Cytotoxic activity of coagulase-

negative staphylococci in bovine mastitis. Infect. Immun. 68,

1102–1108.