BACTERIOCIN TYPING OF STAPHYLOCOCCI

Skalka, B.: Bacteriocin Typing of Staphylococci. Acta vet. Bmo, 61,1992: 179-187. The report describes bacteriocin typing of staphylococcal strains using the deferred method. The collection of active strains comprised 11 strains, Q of which were isolated by the author, namely 3 S. intermedius strains, 2 S. hominis strains and 1 strain each of S. epidermidis, S. simulans, S. sciuri and S. haemolyticus, and two additional strains, namely S. aureus producing bacteriocin Bac Rl and S. simulans biovar staphylolyticus. The active strains varied in spectrum and intensity of their effects on both indicator stains and examined strains. Using the active strains a total of 2 083 staphylococcal strains representing a great majority of known staphylococcal species were typed. The highest sensitivity was shown by coagulase-positive species, lower sensitivity was found in coagulase-negative, novobiocin-sensitive species and the lowest sensitivity was found in coagulase-negative, novobiocin-resistant species. The bacteriocin typing described in the report may prove useful for detailed characteristics of staphylococcal strains. Bacteriocin typing, deferred method, Staphylococcus spp., active strains, indicator strains, examined strains Our previous reports on antibiotic products of staphylococci have described the characteristics of bacteriocins, also called staphylococcins, bacteriocin-like substances, together with information on their frequency within the genus Staphylococcus and their application in presumptive bacteriocin typing of staphylococci (Skalka 1985, 1986a, 1986b). Till now practical utilization of staphylococcins and staphylococcin-like staphylococcal antibiotic substances has been limited to laboratory studies in spite of some Folitary descriptions of their successful clinical use in staphylococcal infections (Lachowicz 1965, Gerber and Nowak 1976) and in the therapy of a skin disease of Dermatophilus congolensis aetiology (Zaria 1991). Laboratory use of staphylococcin producers has been described by Ivanov (1970, J etten and Vogels (1972), Pulverer and J eljaszewicz (l976) and Smarda and ObdrHlek (1981). The object of the present study is to present our own set of active strains and its use in bacteriocin typing of staphylococcal species. Materials and Methods


Bacterial Strains
Bacterial strains were divided into the following three groups: a set of active strains or producers of antibiotic substances, a set of known indicator strains and a set of strains to be examined.The identification of staphylococci and the biotyping of S. aureus were carried out on the basis of generally accepted recommendations (Devriese 1984, Kloos andSchleifer 1986).
The list of active and indicator strains gives their working designation (W), species, original laboratory designation (L) and Collection code (C).Strain 1 is known by its production of lysostaphin (Schindler and Schuhardt 1964) and strain 2, by its production of bacteriocin Bac R, (Warren et a1. 1974), The remaining strains for bacteriocin typing were isolated in our laboratory.Of these only strain 23 is not deposited in any collection.CCM strains are deposited in the Czechoslovak Collection of Microorgar>isms and • (CNCTC) and their characteristics are listed .inthe Catalogues of the two collections (Kocur et aI. 1982, Sourek 1990).

Bacteriocin Typing Technique
The technique of pre-cultivation of active strains (deferred method) with their subsequent devitalization with chloroform vapour and spraying of a suspension of the tested or known indicator strain was used in most strains of the set of producers of antibiotic substances.Only in strain 2 whose bacteriocin Bac R, is sensitive to chloroform a modification consisting in cultivation of "lawn bands" was employed.Details of the technicue were reported pr'!viol.lsly(Skalka 1986a(Skalka , 1986b)).

Results
The collection of active strains for typing was asembled on the basis of comparison of the effects of more than 250 presumptively identified producers of antibiotic substances.To 9 strains isolated in our laboratory the following two • • generally known strains were added: strain 1 producing lysostaphin and strain ,2 producing bacteriocin Bac R 1 • The set of staphylococcal indicator strains, to which two corynebacteria were added, was assembled on the basis of our previous experience.-= no effect, + = less than 2 mm, + + = 2 mm to 5 mm, + + + = more than 5 mm showed the widest spectrum and the highest intensity of activity .onall the indicator strains.Only' a slightly weaker intensity of the effects on some indicator strains was seen in strain 6. Strains 1 and 16 exerted marked antibiotic effects on both coagulase-positive and coagulase-negative staphylococcal indicators but proved inactive on both corynebacteria.Strain 23 produced marked inhibition of coagulase-positive staphylococcal indicators but had negative effects .oncoagulase-negative and corynebacterial indicators.Strains 2 and 4 showed a similar spectrum of activity, the effects of strain 2 on the two corynebacteria being more intense.Strain 5 was intensely active towards S. aureus indicatots and had a weaker effect on S. intermedius, coagulase-negative staphylococci and on the two corynebacteria.Strain 14 exerted intense effects on most coagulase-positive.indicators, weak effects on coagulase-negative indicators and no effect on corynebacterial indicators.Strain 8 and 10 were the least effective of the active strains.The spectra and intensity of the effects on the active strains are shown in Table l.
The sensitivity of the staphylococci examined varied from species to species and particularly from strain to strain.The inhibition of growth of a tested strain manifested by a zone greater than 2 mm in diameter round the spot of the active strain was scored as a positive effect; differences in its intensity were not included in the final evaluation.All coagulase-positive strains were sensitive to lysostaphin of strain 1, whereas the sensitivity of coagulase-negative strains was about 50%, being highest in S. chromo genes strains and lowest in S. sciuri strains.Sensitivity to Bac RI of strain 2 was characteristic of S. aureus while S. intermedius and all strains coagulase-negative strains were resistant and only a small number of S. hyicus strains were sensitive.S. aureus and S. intermedius strains were highly sensitive to strains 6, 16 and 22 while the sensitivity of S. hyicus and all coagulase--negative strains to exosubstances of these strains ranged from 30% to 85%.Antibiotic substances of strains 4 and 5 exerted effects on S. aureus biovar strains, which represented the group of the sensitivity, and on about 50% of S. intermedius strains and on only a small number of the remaining staphylococci under study.Strain 23 inhibited the majority of S. aureus strains, had less effect on S. chromo-• .genes, S. hyicus and S. intermedius strains, little effect on coagulase-negative, novobiocin-sensitive staphylococci and no effect on coagulase-negative, novobiocin-resistant staphylococci.Relatively narrow spectra of effects were shown byexo--substances of strains 8, 10 and 14.From total assessment of the results it appears that the highest sensitivity was shown by coagulase-positive strains, a much weaker.sensitivity was recorded for coagulase-negative, novobiocin-sensitive strains .and the l~west sensitivity was shown by coagulase-negative, novobiocin,,:,resistant -strains,particu1arly by,S.sciuri.Complete resistance was shown by less than 5% •of the strains examined.Within species there were groups~haringlhe ~ame cha-.racter of s~nsitivity.The results are illustrated by 4 examples in Fig. 1-4

Discussion
Our previous studies (Skalka 1986a(Skalka , 1986b) ) enabled us to assemble both a col-• lection for bacteriocin typing of staphylococci and a collection of indicator strains.Combined with our experiences with exfoliatin producers (Skalka et al. 1983a(Skalka et al. , 1983b)), the afore-mentioned studies made it possible to use Bac Rl producer and two corynebacterial indicators.In the light of the reports on the possibility of differentiating staphylococci and micrococci by means oflysostaphin (Poutrel .and Caffin 1981, Baker 1984, Geary and Stevens 1986, Harvey and Gilmour 1988) we extended our collection of active strains by including the producer of this enzyme' complex." -In spite of attempts at uniform terminology in the field of bacteriocins the respective suggestions along this line have not been accepted by some writers (Piard and Desmaud 1992).For the most part, however, the recommendation has been accepted that bacteria-produced antibiotic substances should be termed bacteriocins or bacteriocin-like substances depending on as to how they meet the classical criteria (Tagg et al. 1976).Except lysostaphin of strain 1 and bacteriocin Bac Rl the antibiotic substances of the remaining strains of our active set meet the criteria required for bacteriocin-like exosubstances.
The results obtained with the typing set strains on indicator strains show that the set comprised strains having a wide spectrum and a high intensity of activity along with strains producing weaker effects.However, the inclusion of the latter contributed to the achievement of greater differentiation of typed staphylococci.
From the results obtained with lysostaphin producer it appears that the concentration of this enzyme complex achieved upon the growth of the producer on nutrient medium was not sufficient to affect all staphylococci, particularly those coagulase-negative, which agrees with the observations reported by Poutrel and Caffin (1981).Investigators who used commercial disks having a high concentration of this exosubstance described 100 per cent effects (Geary andStevens 1986, Harvey andGilmour 1988), but they tested mainly S. aureus strains.Baker (1984) who studied the sensitivity of a wider spectrum of staphylococcal species to commercial lysostaphin disks found a high resistance in S. haemolyticus, S. warneri and S. epidermidis strains.
The finding that the effect of bacteriocin Bac Rl was limited only to S. aureus strains corresponds to our previous experience (Skalka 1986b).
The activity of the remaining strains of the bacteriocin set as assessed on tested strains correlated with the results of their testing on known indicators.Of particular interest are great effects of strains 4 and 5 on S. aureus biovar C and S. intermedius, considering their weak activity on the other S. aureus biovar strains on the strains of the remaining staphylococcal species, also described previously (Skalka 1968b).However, we have not succeded as yet in obtaining antibiotic strains the effect of which would be reciprocal to the effects of strains 6 and 23, i. e. the effects of which would be higher on coagulase-negative staphylococci and weak on thoSe coagulase-positive.
From the results reported here it appears that the strains most sensitive to the set of active strains used in our study were those of S~ aureus biovar C, followed by S. intermedius and S. aureus biovar A and biovar B strains.Much weaker sensitivity was shown by coagulase-negative strains, particularly by those resistant to novobiocin.In the present study the results of bacteriocin typing of the strains examined do not give the degree of intensity of the effects of active strains as used in testing on the indicators.Such assessment, possibly extended to include further degrees of intensity is recommendable for routine bacteriocin typing.
Comparison of the present results with those reported by other writers is difficult considering the use of different active strains and different numbers of strains examined and, last but not least, the changes in the taxonomy of staphylococci.Our observations on the sensitivity of staphylococci as a whole correspond to those of Ivanov (1970) and our findings of higher sensitivity of coagulase--positive strains are in keeping with the data reported by Pulverer and Jeljaszewicz (1975) and Smarda and Obdrzalek (1981).Higher sensitivity of coagulase-negative than coagulase-positive staphylococci was reported by J etten and Vogels (1972) but only for the sensitivity to staphylococcin 1580.
The bacteriocin typing of staphylococci described in the present report is feasible in terms of cost, well-suited for detailed characteristics of strains and may prove useful in epizootic and epidemic studies.An analogous method has • been recommended for a similar purpose also in other gram-positive species (Watson 1985).

Fig. 2 .Fig. 3 .
Fig. 2. Result of bacteriocin typing of S. aureus biovar C stram (A).For explanation of the figures designating active strains see Materials and Methods.

Fig. 4 .
Fig. 4. Result of bacteriocin typing of S. haemolyticus strain (H).For explanation of the figures designating active strains see Materials and Methods.