JOURNAL OF CLINICAL MICROBIOLOGY, Sept. 1985, p. 409-415 Vol. 22, No. 30095-1137/85/090409-07$02.00/0Copyright © 1985, American Society for Microbiology

Synergistic Hemolysis Exhibited by Species of StaphylococciG. ANN HÉBERT* AND GARY A. HANCOCK

Hospital Infections Program, Center for Infectious Diseases, Centers for Disease Control, Atlanta, Georgia 30333

Received 29 April 1985/Accepted 5 June 1985

The synergistic hemolysis reactions of 61 reference strains and 189 clinical isolates representing 17 species ofstaphylococci were examined on plates of Trypticase soy blood agar (BBL Microbiology Systems, Cockeysville,Md.). Some or all of the strains of Staphylococcus aureus, S. epidermidis, S. capitis, S. cohnii, S. haemolyticus,S. hyicus, S. simulans, S. warneri, and S. xylosus produced a delta-hemolysin that gave synergistic, completehemolysis of washed human, sheep, and ox blood cells in an area of beta-lysin activity from strains of S. aureusand S. intermedius. Strains of the same nine species were positive with a commercial beta-lysin paper diskdesigned for presumptive identification of group B streptococci; most of these strains also gave synergistic,complete hemolysis with exotoxin from a strain of Corynebacterium pseudotuberculosis. None of the strains ofS. auricularis, S. carnosus, S. caseolyticus, S. hominis, S. intermedjus, S. saprophyticus, S. sciuri, or S. lentuswere positive by any of these tests for synergistic hemolysis. These results indicate that a synergistic hemolysistest could prove very useful for differentiating these species; they also suggest that one role of some of theseorganisms in human infections could be that of a synergist. Further studies of synergism may clarify the clinicalsignificance of these results.

In the past several years there has been an increasingnumber of reports of the incidence of coagulase-negativestaphylococci in clinical infections, and the pathogenic roleof these organisms is now well established (1, 7). Coagulase-negative staphylococci are frequently implicated in infec-tions of the urinary tract, heart valves, eye and ear wounds,cerebrospinal fluid shunts, and vascular and joint prosthe-ses; they are also the implied cause of subacute bacterialendocarditis, peritonitis in patients on continuous peritonealdialysis, and bacteremia in patients receiving immunosup-pressive therapy (1, 7, 17). Since 1975, Kloos and Schleifer(18-21, 28-31) have been systematically reorganizing thecoagulase-negative staphylococci into separate species withdefined phenotypic characteristics; Devriese and Hajek (6,13) have defined new species of coagulase-positive and-negative staphylococci that are most often associated withanimal infections. All of the named species are quite distinctin DNA homology studies done under restrictive reas-sociation conditions; the clinical significance of the separatespecies, however, is still being defined.

Before 1975, only three species were recognized in thegenus Staphylococcus (2); all of the coagulase-negativestaphylococci were either S. epidermidis or S. saprophyti-cus, and the coagulase-positive strains were S. aureus. Thetoxins and hemolysins of these staphylococci were thesubjects of many extensive reports. Several studies showedthat some of the less hemolytic staphylococci (4, 8), some ofthe coagulase-negative staphylococci (16, 32), and some S.aureus and S. epidermidis strains (3, 5, 23, 25) produced adelta-hemolysin that gave synergistic, complete hemolysiswith a beta-toxin-producing Staphylococcus sp. growing onagar with sheep erythrocytes. In one of those studies (32),some coagulase-negative staphylococci also gave synergistichemolysis with an exotoxin isolated from Corynebacteriumpseudotuberculosis. Since 1975, many new species of Staph-ylococcus have been defined, and the descriptions for S.epidermidis and S. saprophyticus have been amended (31);however, we are not aware of any further studies ofhemolysins or synergistic lysis in these defined species. Any

* Corresponding author.

hemolytic activity among these species in synergistic rela-tionships could possibly relate to their virulence and, there-fore, their clinical significance. In addition, consistent syn-ergistic reactions among strains within a species could helpto differentiate between the species. We have examinedreference and clinical strains of defined species of staphylo-cocci for such activities, and the results of our studies arepresented in this report.

(Part of this work was presented at the 1985 AnnualMeeting of the American Society for Microbiology [G. A.Hébert and G. A. Hancock, Abstr. Annu. Meet. Am. Soc.Microbiol. 1985, C174, p. 329].)

MATERIALS AND METHODSCultures and growth conditions. A total of 250 strains

representing 17 species of staphylococci were examinedduring this study. The 61 reference strains listed in Table 1were obtained from W. E. Kloos, Raleigh, N.C., and from theAmerican Type Culture Collection, Rockville, Md. Therandom set of 189 clinical isolates included many supplied byE. H. Gerlach, Wichita, Kans.; the other clinical strains werefrom the culture collections of laboratories at the Centers forDisease Control. All ofthe cultures were plated on Trypticasesoy agar containing 5% defibrinated sheep blood (TSA II;BBL Microbiology Systems, Cockeysville, Md.) and incu-bated aerobically for 18 to 24 h at 35°C. For prolongedstorage, 24-h growth was harvested in sterile rabbit blood andkept at -70°C. Strains removed from storage were plated onblood agar at least twice before testing; 24-h cells from thesecond or third plate were used.

Identification. All of the strains were gram-positive coccithat produced catalase. All strains were tested for coagulaseactivity by both the slide and tube tests with EDTA-rabbitplasma. Coagulase-positive strains were tested for acidproduction from glucose and mannitol anaerobically andfrom maltose and mannitol aerobically, for DNase andalkaline phosphatase activity, and for pigment production.Coagulase-negative strains were identified by the methods ofKloos (17, 19).Hemolysin definition. Several strains of each of the

staphylococcal species studied were grown on plates of

409

on February 2, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

410 HÉBERT AND HANCOCK

TABLE 1. Reference strains of Staphylococcus speciesSpecies Strain designation

S. auricularis .................. B04; ATCC 33750, 33751,33752, 33753'

S. capitis .................... ATCC 27840T, 27841, 27842,27843

S. carnosus .................... Hansen, MAS. caseolyticus ................. ATCC 13548T, 29750S. cohnii .................... DM224, SH161A; ATCC 29972,

29973, 29974'S. epidermidis ................. AW269, GH37, 1386-3; ATCC

29886, 29887S. haemolyticus ................ GH59; ATCC 29968, 29969,

29970TS. hominis.................... JL248; ATCC 27844T, 27845,

27846, 27847S. hyicus subsp. hyicus ......... 3813-D; ATCC 11249TS. hyicus subsp. chromogens .... 6-IRRb; 93-IRRaS. intermedius ................. CFDD, RK12; ATCC 29663TS. saprophyticus ............... KL20, SM295, TW111; ATCC

15305TS. sciuri .................... ATCC 29059, 29060, 29061,

29062TS. lentus .................... K-6S. simulans .................... 6L31; ATCC 27848T, 27849,

27850, 27851S. warneri .................... ATCC 27836T, 27837, 27838,

27839S. xylosus ..................... DM37; ATCC 29966, 29967,

29971T

Trypticase soy agar containing 5% defibrinated, washedblood cells of five different animal species: rabbit, sheep,horse, ox, and human. A strain of beta-lysin-producing S.intermedius was streaked perpendicular to, but not touching,the test strain on each of the five agars. All of the plates wereincubated aerobically at 35°C for 18 to 20 h and then at roomtemperature for 4 to 6 h. All synergistic hemolysis reactionswere recorded at approximately 24 h. A zone of completehemolysis within the zone of incomplete hemolysis causedby S. intermedius was a positive test (see Fig. 1). Theseplates were then reincubated overnight, and the hemolysis ofthe various species of blood cells was recorded at 48 h (9).The generally accepted terminology used by Elek (8, 9) andothers to define hemolysins was used: (i) alpha-hemolysinproduces a wide zone of complete hemolysis with blurrededges on agar containing rabbit, sheep, or ox erythrocytes,but not horse or human cells; (ài) beta-hemolysin produces awide zone of incomplete hemolysis with sharp edges on agarcontaining sheep, ox, or human erythrocytes, but not rabbitor horse cells; (iii) delta-hemolysin produces a narrow zoneof complete hemolysis with blurred edges on agars contain-ing erythrocytes of any of the animal species tested. Thezone from a delta-hemolysin is poorly developed with thecells of some animal species, and small amounts of delta-hemolysin might not show on sheep cells. A delta-hemolysinpotentiates the zone of a beta-hemolysin to complete clear-ing on sheep, ox, and human cells.

Selection of beta-lysin and exotoxin-producing strains. Sev-eral strains of coagulase-positive staphylococci and C.pseudotuberculosis were tested for beta-hemolysin andexotoxin production, respectively, to select strains to use insynergistic hemolysis tests. A technique similar to thatdescribed by Zemelman and Longeri (36) was used. A strainof S. epidermidis, known to produce a delta-lysin because ofpreliminary tests, was streaked down the center of a TSA Il

plate. Strains of S. aureus, S. intermedius, and C. pseudo-tuberculosis were then streaked perpendicular to, but nottouching, this central streak; 8 to 10 strains, 4 or 5 on eachhalf of the plate, were tested per plate. The plates wereincubated as described above for the hemolysin tests, andresults were recorded at 24 h. A strain of Staphylococcusproducing beta-lysin or a strain of C. pseudotuberculosisproducing exotoxin gave a wide zone of incompletehemolysis of sheep erythrocytes all along its line of growthand a large zone of synergistic, complete hemolysis in thearea of delta-lysin next to the central line of S. epidermidisgrowth (see Fig. 2).

Synergistic hemolysis growth tests. Strains were tested forsynergistic hemolysis by the technique first described byMunch-Petersen et al. (26) for group B streptococci. A strainof S. intermedius (AB 148) that produced beta-lysin in testsdescribed above was streaked down the center of a TSA IIplate. A strain of C. pseudotuberculosis (ATCC 19410),selected in the above tests, was streaked down the center ofanother TSA Il plate. Test strains were streaked perpendic-ular to, but not touching, the center streaks as describedabove. The plates were incubated as described for thehemolysin tests, and results were recorded after 24 h. A zoneof complete hemolysis, within the zone of incompletehemolysis caused by the beta-lysin from the S. intermediusgrowth or the exotoxin from the C. pseudotuberculosisgrowth, was a positive test (see Fig. 3).

Beta-lysin disk test. Paper disks (CAMP disk; Carr-Scarborough Microbiologicals, Inc., Stone Mountain, Ga.),impregnated with partially purified staphylococcal beta-lysinfrom a strain of S. aureus as described by Wilkinson (34),were placed along the center line of a TSA Il plate. Teststrains were streaked in a straight line to within 2 to 4 mm ofthe disks. The plates were incubated as described for thehemolysin tests, and the results were recorded at 24 h. Azone of complete hemolysis, within the zone of incompletehemolysis surrounding the disk, was a positive reaction (seeFig. 5).

RESULTSThe hemolytic reactions of one reference strain each of S.

epidermidis and S. intermedius grown for 48 h on agars witha^

FIG. 1. Hemolytic reactions (48 h) of a strain of S. intermedius(streaks of growth along the outer edge in each quadrant) and astrain of S. epidermidis (streaks of growth radiating from the centerof the plate in each quadrant) on agar with blood cells of fourseparate animal species: sheep (upper left), ox (upper right), horse(lower left), and rabbit (lower right).

J. CLIN. MICROIBIOL.

on February 2, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

SYNERGISTIC HEMOLYSIS IN STAPHYLOCOCCI 411

sheep, ox, horse, and rabbit blood cells are shown in Fig. 1;the synergistic reactions were more pronounced than thoserecorded at 24 h, but the hemolysis patterns were typical ofthose seen on the various species of cells. A very narrowzone of complete hemolysis with a blurred edge developedaround the growth of the strain of S. epidermidis on agarswith sheep and ox cells, a small clear zone with a blurrededge was seen with horse cells, and a larger clear zoneresulted with rabbit cells. A visible zone of incompletehemolysis diffused from the S. intermedius growth on theagars with sheep and ox cells, but no such activity was seenwith horse or rabbit cells. Large zones of completehemolysis developed around the S. epidermidis growth inthe area of incomplete hemolysis caused by the activity of S.intermedius on sheep and ox cells; no synergistic reactionswere seen with horse or rabbit cells. The hemolytic reactionson agar with human blood cells are not shown, but they werelike those seen with sheep and ox cells, though less intense:the zone of incomplete hemolysis was not as wide, and thezone of complete, synergistic hemolysis was smaller. Thesesame hemolytic reactions on the five species of blood cellswere seen with the other strains of S. epidermidis, S. capitis,S. cohnii, S. haemolyticus, S. simulans, S. warneri, and S.xylosus. No or only occasional, very weak hemolysis wasrecorded for strains of S. auricularis, S. carnosus, S.caseolyticus, S. hominis, S. saprophyticus, S. sciuri, and S.lentus. Strains of S. aureus had from one to three of thedefined hemolysins, and S. hyicus was not hemolytic butshowed some synergism.The patterns of hemolysis seen when coagulase-positive

strains of S. aureus and S. intermedius were tested forbeta-lysin production to determine which would be best touse in tests for synergistic hemolysis with other staphylo-cocci are shown in Fig. 2. Each of the positive test strainshad a wide visible zone of incomplete hemolysis due tobeta-lysin activity, which was then enhanced to completehemolysis in the area adjacent to the growth of a strain of S.epidermidis. The outer edge of the beta-lysin zone was from6 to 9 mm from the edge of cell growth among the strainstested. Four of 15 S. aureus and 5 of 6 S. intermedius strainshad distinct beta-lysin zones, but the other strains had

P~~~~~~~~~~~~~~~~~~~~~~~~~~u

FIG. 2. Synergistic hemolysis between a strain of S. epidermidis(vertical streak of growth) and six strains of coagulase-positivestaphylococci. The top strain on the left and ail three strains on theright are S. intermedius; the left center and bottom strains are S.aureus. Each of these six strains had a wide zoneC of incompletehemolysis that was enhanced to complete hemolysis in the zone ofdelta-lysin activity from the vertical strain.

FIG. 3. Synergistic hemolysis between a strain of S. intermedius(vertical streak of growth) and 10 strains of S. epidermidis. The leftcenter strain had a wide zone of incomplete hemolysis that wasenhanced to complete hemolysis in the zone of beta-lysin activityfrom the vertical strain. The right center strain was negative.

multiple other hemolysins and zones of complete hemolysisthat would have masked or interfered with any test forsynergistic hemolysis. Strain AB 148 of S. intermedius wasselected to test other strains for synergistic hemolysis; thisstrain produced a 7- to 8-mm zone of beta-lysin activity andhad a very narrow zone of complete hemolysis that did notconflict with tests for enhancement or synergism. The fourstrains of C. pseudotuberculosis that were tested forexotoxin production were only slightly hemolytic and gavezones of synergistic, complete hemolysis with the test strainof S. epidermidis and a reference strain (ATCC 6939) of C.equi (Rhodococcus equi). For these strains, the outer edge ofthe exotoxin zone was 6 to 10 mm from the edge of cellgrowth at 24 h and increased to 7 to 14 mm at 48 h. All fourstrains gave satisfactory results; the reference strain, ATCC19410, was selected as the test strain.Ten of the clinical isolates of S. epidermidis tested for

synergistic hemolysis with S. intermedius are shown in Fig.3. The size and shape of the clear zone of synergistic,complete hemolysis varied among strains, but a given straingave reproducible results; hazy or incomplete zones wererecorded as negative. Individual strains of S. epidermidisexhibited no or only weak hemolysis of sheep cells at 24 h,but two of the clinical isolates had moderate zones (4.5 and5.5 mm) of incomplete hemolysis on sheep blood agar.Beta-lysin activity was enhanced to complete hemolysis inthese zones, and these two strains did not give synergistichemolysis with the strains that were producing delta-lysin;therefore, the moderate-width zones were not beta-lysinreactions and were recorded as atypical delta-lysin zones.One of these strains is shown in Fig. 3.Some or all of the reference and clinical strains of S.

aureus, S. epidermidis, S. capitis, S. cohnii, S. haemolyti-cus, S. hyicus, S. simulans, S. warneri, and S. xylosus gavea distinct, clear zone of synergistic, complete hemolysiswhen tested against strain AB 148 of S. intermedius. Some ofthese reactions are shown in Fig. 3 and 4; all of the resultsare given in Tables 2 and 3. The two reference strains of S.hyicus subsp. hyicus were recorded as positive, but the areaof complete hemolysis was unique; a clear area shaped likean arc or a parenthesis developed within a zone of partialhemolysis that was clearer than the outer area of beta-lysinactivity. None of the reference or clinical strains of S.

VOL. 22, 1985

on February 2, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

412 HÉBERT AND HANCOCK

FIG. 4. Synergistic hemolysis between a strain of S. intermedius(vertical streak of growth) and strains of coagulase-negative staph-ylococci: three S. epidermidis (upper left), two S. simulans (lowerleft), one S. capitis (top right), and four S. haemolyticus (othersright).

auricularis, S. carnosus, S. caseolyticus, S. hominis, S.intermedius, S. saprophyticus, S. sciuri, or S. lentus werepositive. Many of the reference and clinical strains repre-senting most of the 17 species tested were checked againstseveral other beta-lysin-producing strains of S. intermediusand S. aureus; the results were the same as those obtainedwith AB 148.The same pattern of results was obtained when our

reference and clinical strains of staphylococci were testedwith the commercial beta-lysin disks which were designed

TABLE 2. Synergistic hemolysis exhibited by reference strains ofstaphylococci

No. of strains exhibiting a zone of syn-

Staphylococcus No. of ergistic, complete hemolysis with:

species strains S. inter- C. pseudo-tested S. aureus tuberculo-medius b tbruo(growthïa (diskY sis"

S. capitis 4 4 4 4S. cohnii 5 5 5 4S. epidermidis 5 4 4 4S. haemolyticus 4 4 4 4S. simulans 5 5 5 5dS. warneri 4 2 1 2S. xylosus 4 3 3 3S. hyicus subsp. 2 1 1 0chromogens

S. hyicus subsp. 2 (2)e 0 0hyicus

S. auricularis 5 O O OS. carnosus 2 0 0 0S. caseolyticus 2 0 0 0S. hominis 5 O O OS. intermedius 3 0 0 0S. saprophyticus 4 0 0 0S. sciuri 4 0 0 0S. lentus 1 0 0 0

a S. intermedius strain AB 148 on TSA II sheep blood agar.b Paper disks impregnated with beta-lysin from a strain of S. aureus.C C. pseudotuberculosis strain ATCC 19410 on TSA II sheep blood agar.d A very faint zone at 24 h and a definite clear zone at 48 h.e A clear, arc-shaped area developed within a zone of partial hemolysis that

was clearer than the surrounding area of beta-lysin activity (atypical reaction).

TABLE 3. Synergistic hemolysis exhibited by clinical isolates ofstaphylococci

No. (%) of strains exhibiting a zone of

Staphylococcus No. of synergistic, complete hemolysis with:

species strains C. pseudo-tested S. inter- S. aureus" tuberculo-medius' is

S. haemolyticus 23 23 (100) 23 (100) 21(91)S. simulans 17 17 (100) 16 (94) 15 (88)S. capitis 5 5 (100) 4 (80) 4 (80)S. cohnii 1 1 (100) 1 (100) 1 (100')S. epidermidis 78 63 (81) 58 (74) 54 (69)S. warneri 14 11 (79) 11 (79) 9 (64)S. xylosus 7 5 (71) 5 (71) 5 (71)'S. hominis 14 0 0 0S. saprophyticus 9 0 0 OS. sciuri 3 0 0 0S. intermedius 3 0 0 0S. aureus 15 5 (33) 5 (33) 5 (33)

a Strain AB 148.bPaper disk impregnated with beta-lysin.C Strain ATCC 19410.d Partial at 24 h but clear at 48 h.

for the presumptive identification and grouping of strepto-cocci. The results are listed in Tables 2 and 3; some of thereactions are shown in Fig. 5 and 6. All but three of thereference strains that gave synergistic hemolysis with strainAB 148 of S. intermedius also gave a positive beta-lysin disktest. The three disk-negative strains were a strain of S.warneri and both strains of S. hyicus subsp. hyicus (Table 2);however, the two S. hyicus reactions with S. intermediuswere atypical and should be considered negative, so the onlyreal discrepancy was a single strain of S. warneri. All but 7of the 189 clinical isolates gave the same results with S.intermedius and the beta-lysin disk (Table 3). The sevenstrains that were positive with S. intermedius but negativewith the disk were five strains of S. epidermidis and onestrain each of S. simulans and S. capitis. Repeated tests withthese strains gave consistent results. None of the referenceor clinical strains tested were negative with S. intermediusand positive with the paper disk containing S. aureus beta-lysin.The beta-lysin pattern of results was again seen when the

FIG. 5. Reactions of a commercial beta-lysin disk with referencestrains of coagulase-negative staphylococci: two S. haemolyticus(upper and center left), one S. hominids (lower left), and three S.epidermidis (right).

J. CLIN. MICROBIOL.

on February 2, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

SYNERGISTIC HEMOLYSIS IN STAPHYLOCOCCI 413

c

-.

FIG. 6. Reactions of a commercial beta-lysin disk with referencestrains of coagulase-negative staphylococci: two S. epidermidis(upper and center left), one S. simulans (lower left), and three S.capitis (right).

reference and clinical strains of staphylococci were testedagainst the exotoxin of C. pseudotuberculosis (Tables 2 and3). Only two of the reference strains, an S. cohnii and an S.hyicus strain, were positive with S. intermedius and thennegative with C. pseudotuberculosis (Table 2). All but 16 ofthe 189 clinical isolates gave the same results with S.intermedius and C. pseudotuberculosis. The 16 strains thatwere positive with the beta-lysin but negative with theexotoxin included nine strains of S. epidermidis, one strainof S. capitis, and two strains each of S. haemolyticus, S.

simulans, and S. warneri (Table 3). None of the strainstested were negative with S. intermedius and positive withC. pseudotuberculosis. Strains of S. aureus, S. capitis, S.

epidermidis, S. haemolyticus, and S. warneri gave strongpositive reactions at 24 h, but strains of S. cohnii, S.

simulans, and S. xylosus were very weak at 24 h and thendistinct at 48 h. These weak reactions at 24 h were recordedas negative.

DISCUSSIONThe hemolysin produced by strains from 8 of 15 species of

coagulase-negative staphylococci displayed the reactivity ofa delta-hemolysin; it was active, though variable, on theblood cells of all five animal species, and it gave synergistic,complete hemolysis with a beta-lysin on the sheep, ox, andhuman cells but not the horse or rabbit cells. We did not seeevidence of the epsilon-hemolysin discussed by Elek andothers (8, 9); it was described as a wide-zone hemolysincausing complete hemolysis of both sheep and rabbit cells,which clearly distinguished the coagulase-negative skinstrains of staphylococci. Our findings agree with otherstudies (11, 16, 22, 24), which state that the typical hemoly-sin produced by cultures of coagulase-negative staphylo-cocci is delta-hemolysin.Our coagulase-positive strains exhibited the hemolytic

activity described by Hajek (13) and Levy (as quoted inreference 8). The strains of S. intermedius produced a

beta-hemolysin, and one strain also had an alpha- or delta-hemolysin that gave a clear zone; the strains of S. aureus

displayed multiple hemolysins, including delta-hemolysin.One strain of S. hyicus subsp. hyicus gave a late-positivetube-coagulase test result; both strains of this subspecies andone strain of S. hyicus subsp. chromogens produced a

delta-like hemolysin that was only detected because of its

synergistic reaction. No hemolysins were reported in aprevious study of S. hyicus (6).

Kloos and Schleifer (18, 31) tested strains for hemolysis onbovine, human, sheep, and rabbit blood agar plates afterincubation for 24, 48, and 72 h. Their simplified scheme (19)for differentiating species of coagulase-negative staphylo-cocci includes a test for hemolysis with 5% bovine blood inP agar that is read at 72 h; hemolysis is reported as strong (+)or moderate-to-weak (±) on the basis of zone width from theculture streak. The zones are not described as clear orincomplete or as having either distinct or blurred edges. Indefining the hemolysis of a species, they sometimes used thewords partial, weak, moderate, or good but did not use theterms alpha, beta, delta, or epsilon to define the hemolysin.In their studies of nine species (18, 31), some strains of eachhad no hemolytic activity, some strains of each gave weakhemolysis, and some strains of S. haemolyticus, S. cohnii, S.xylosus, and S. warneri gave strong hemolytic reactions. Nohemolytic reactions were reported for other coagulase-negative species (6, 20, 21, 28-30).

During our studies, some or all of the strains tested fromnine different species of staphylococci gave positive syner-gistic hemolysis tests. Many times the positive synergisticreactions were completely unexpected, because the 24- andeven 48-h growth of the strains on TSA II plates had shownlittle or no visible hemolytic activity with sheep cells. Theseseemingly nonhemolytic strains were apparently producingsmall amounts of hemolysin that did not show on sheep cellsunless the cells were also attacked by a beta-lysin or anexotoxin.The results obtained with the two staphylococcal beta-

lysin test systems were comparable. The S. intermediusgrowth and disk of S. aureus beta-lysin produced positivesynergistic-hemolysis test results with strains of the samenine species of staphylococci. The negative strains of S.warneri and S. xylosus gave weak reactions at 24 h with boththe growth and disk tests, but the zones were not completelycleared and barely extended beyond the growth of thestrains. The positive strains of both of these species gavevery strong reactions at 24 h. The clear, arc-shaped areasseen with S. hyicus subsp. hyicus in the S. intermediusgrowth tests were also seen with the disk tests, but only afterprolonged incubation; at 24 h, the zones were hazy, but thearcs had not yet developed. These diminished reactions andthe strain of S. warneri that was negative only in the disk testwere the only differences seen in the reactivity of referencestrains in the two staphylococcal beta-lysin test systems. Afew more differences in reactivity were seen with clinicalisolates; seven strains from three other species were nega-tive with the disk test and positive in growth tests. It isunlikely that these differences are due to the different speciessources of the beta-lysins, since the beta-lysins from severalstrains each of S. aureus and S. intermedius gave identicalreactions in growth tests. The differences we observed wereprobably because of less reactive beta-lysin in the paperdisks. The growth test has a continuous supply of freshbeta-lysin diffusing from the growing organism, but thereactivity in the disk test is limited by the measured quantityand strength of the partially purified beta-lysin prepared bythe commercial supplier. The reactive zone around a disksoon reaches its limit, but the zone from a growing organismwill grow stronger with continued incubation. Despite thislimitation, the beta-lysin disk is a reasonable alternative to aviable Staphylococcus strain as a source of beta-lysin forsynergistic-hemolysis tests.The differences seen with the third test system, C.

VOL. 22, 1985

on February 2, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

414 HÉBERT AND HANCOCK

pseudotuberculosis exotoxin, were much greater. Moststrains gave reactions comparable to those seen with the twostaphylococcal test systems, but 48 h was required for thepositive strains of three species to produce clear zones, andseveral more strains were nonreactive with this test thanwith either of the beta-lysin tests. C. pseudotuberculosisexotoxin should not, therefore, be used to test staphylococcifor delta-lysin production, but the data we obtained demon-strated the synergistic relationships between these orga-nisms.

Synergistic hemolysis has been the subject of many pre-vious studies of a variety of microorganisms. In 1944,Christie et al. (4) described a lytic extracellular agent pro-duced by group B streptococci (Streptococcus agalactiae)that gave complete hemolysis when the streptococcus wasgrown in a zone of staphylococcal beta-lysin on sheep or oxblood agar plates. Murphy et al. (27) later called the lyticphenomenon the CAMP reaction and defined the CAMP testas a way to determine the ability of streptococci to producea lytic agent (CAMP factor) which gave a hemolytic zonewith staphylococcal beta-lysin. The acronym CAMP wasthus established, and since then a number of organisms havebeen reported as CAMP positive or said to produce theCAMP reaction with S. aureus beta-lysin; Listeria monocy-togenes, Pasteurella haemolytica, Loefflerella pseudomal-lei, Corynebacterium equi, and C. renale (10); Mobiluncusmulieris and M. curtisii (33); and Propionibacterium acnes(15). In another study (32), the purified beta-toxin of S.aureus and an isolated exotoxin of C. pseudotuberculosisgave synergistic lysis with the delta-hemolysin of S. aureusand epsilon-hemolysin of coagulase-negative staphylococci(species not determined). Under anaerobic conditions, thealpha-toxin of Clostridium perfringens gave synergistic lysison sheep (35), human, and guinea pig blood agars (12) withgroups A, B, C, and G streptococci and on sheep and humanblood agars with P. acnes (15). A reverse CAMP test wasproposed (14) with group B streptococci in a synergistichemolysis reaction for presumptive identification of C. per-fringens. Darling (5), however, argues that the acronymCAMP should only apply when testing strains of strepto-cocci against staphylococci and that the phrase "synergistichemolytic effect" or some similar one should be used todescribe the potentiation of hemolysis by any other bacteriain a zone of staphylococcal beta-lysin or similar bacterialexosubstance; we agree.We have demonstrated three of the synergistic relation-

ships that some species of coagulase-negative staphylococcihave with other organisms; on agar plates containing ox,sheep, or human erythrocytes, strains of many of the speciesthat we tested exhibited weak or no hemolysis alone but thenshowed strong, complete hemolysis in a synergistic reactionwith the beta-lysins of S. aureus and S. intermedius and theexotoxin of C. pseudotuberculosis. In light of the reportsmentioned above, it is likely that the same pattern ofreactivity would occur among these species of staphylococciif they were tested with the alpha-toxin of C. perfringens.Perhaps one role of some of these organisms in humaninfections has been that of a synergist, but when some othermore recognized pathogen was also isolated from a clinicalspecimen, the coagulase-negative staphylococcus was la-beled a contaminant. Our data suggest that (i) the coagulase-negative staphylococci, which may act as synergists, shouldnever be discounted as irrelevant when isolated from clinicalmaterial; and (ii) further studies of synergism may helpclarify their clinical significance. In addition, if the samepattern of results seen in our study of reference and clinical

strains of 17 species of staphylococci is obtained with stilllarger groups of strains, a synergistic hemolysis test mightprove helpful in differentiating the species.

ACKNOWLEDGMENTS

We thank P. B. Smith and Clyde Thornsberry for their criticalreviews of this manuscript, Robert E. Weaver and Dannie G. Hollisfor providing the strains of Corynebacterium species, and Carol G.Crowder for technical assistance with staphylococcal identifications.

LITERATURE CITED1. Aldridge, K. E. 1982. Coagulase-negative staphylococci. Infect.

Control 3:161-165.2. Baird-Parker, A. C. 1974. Gram-positive cocci. Genus Il.

Staphylococcus, p. 483-489. In R. E. Buchanan and N. E.Gibbons (ed.), Bergey's manual of determinative bacteriology,8th ed. The Williams & Wilkins Co., Baltimore.

3. Brzin, B. 1976. Camp-phenomenon in staphylococci. Zentralbl.Bakteriol. Mikrobiol. Hyg. I. Abt. Orig. A 235:373-375.

4. Christie, R., N. E. Atkins, and E. Munch-Petersen. 1944. A noteon a lytic phenomenon shown by group B streptococci. Aust. J.Exp. Biol. Med. Sci. 22:197-200.

5. Darling, C. L. 1975. Standardization and evaluation of theCAMP reaction for the prompt, presumptive identification ofStreptococcus agalactiae (Lancefield group B) in clinical mate-rial. J. Clin. Microbiol. 1:171-174.

6. Devriese, L. A., V. Hàjek, P. Oeding, S. A. Meyer, and K. H.Schleifer. 1978. Staphylococcus hyicus (Sompolinsky 1953)comb. nov. and Staphylococcus hyicus subsp. chromogenessubsp. nov. Int. J. Syst. Bacteriol. 28:482-490.

7. Doern, G. V. 1984. Coagulase-negative staphylococci-clinicalsignificance and laboratory identification. Lab. Med. 15:25-32.

8. Elek, S. D. 1959. Staphylococcus pyogenes and its relation todisease. E. & S. Livingstone Ltd., Edinburgh.

9. Elek, S. D., and E. Levy. 1950. Distribution of haemolysins inpathogenic and non-pathogenic staphylococci. J. Pathol. Bacte-riol. 62:541-554.

10. Fraser, G. 1964. The effect on animal erythrocytes of combina-tions of diffusible substances produced by bacteria. J. Pathol.Bacteriol. 88:43-53.

11. Gladstone, G. P., and W. E. van Heyningen. 1957. Staphylococ-cal leucocidins. Br. J. Exp. Pathol. 38:123-137.

12. Gubash, S. M. 1978. Synergistic hemolysis phenomenon shownby an alpha-toxin-producing Clostridium perfringens and strep-tococcal CAMP factor in presumptive streptococcal grouping.J. Clin. Microbiol. 8:480-488.

13. Hajek, V. 1976. Staphylococcus intermedius, a new speciesisolated from animals. Int. J. Syst. Bacteriol. 26:401-408.

14. Hansen, M. V., and L. P. Elliott. 1980. New presumptiveidentification test for Clostridium perfringens: reverse CAMPtest. J. Clin. Microbiol. 12:617-619.

15. Kar Choudhury, T. 1978. Synergistic lysis of erythrocytes byPropionibacterium acnes. J. Clin. Microbiol. 8:238-241.

16. Kleck, J. L., and J. A. Donahue. 1968. Production of thermo-stable hemolysin by cultures of Staphylococcus epidermidis. J.Infect. Dis. 118:317-323.

17. Kloos, W. E. 1982. Coagulase-negative staphylococci. Clin.Microbiol. Newsl. 4:75-79.

18. Kloos, W. E., and K. H. Schleifer. 1975. Isolation and charac-terization of staphylococci from human skin. Il. Descriptions offour new species: Staphylococcus warneri, Staphylococcuscapitis, Staphylococcus hominis, and Staphylococcus simulans.Int. J. Syst. Bacteriol. 25:62-79.

19. Kloos, W. E., and K. H. Schleifer. 1975. Simplified scheme forroutine identification of human Staphylococcus species. J. Clin.Microbiol. 1:82-88.

20. Kloos, W. E., and K. H. Schleifer. 1983. Staphylococcusauricularis sp. nov.: an inhabitant of the human external ear.Int. J. Syst. Bacteriol. 33:9-14.

21. Kloos, W. E., K. H. Schleifer, and R. F. Smith. 1976. Charac-terization of Staphylococcus sciuri sp. nov. and its subspecies.Int. J. Syst. Bacteriol. 26:22-37.

J. CLIN. MICROBIOL.

on February 2, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

SYNERGISTIC HEMOLYSIS IN STAPHYLOCOCCI 415

22. Kocur, M., F. Precechtel, and T. Martinec. 1966. Haemolysinsin coagulase-negative staphylococci. J. Pathol. Bacteriol.92:331-336.

23. Males, B. M., W. A. Rogers, Jr., and J. T. Parisi. 1975.Virulence factors of biotypes of Staphylococcus epidermidisfrom clinical sources. J. Clin. Microbiol. 1:256-261.

24. Marks, J. 1952. Recognition of pathogenic staphylococci: withnotes on non-specific staphylococcal haemolysin. J. Pathol.Bacteriol. 64:175-186.

25. Marks, J., and A. C. T. Vaughan. 1950. Staphylococcal delta-haemolysin. J. Pathol. Bacteriol. 62:597-615.

26. Munch-Petersen, E., R. Christie, and R. T. Simmons. 1945.Further notes on a lytic phenomenon shown by group Bstreptococci. Aust. J. Exp. Biol. Med. Sci. 23:193-195.

27. Murphy, J. M., O. M. Stuart, and F. I. Reed. 1952. Anevaluation of the CAMP test for the identification of Strepto-coccus agalactiae in routine mastitis testing. Cornell Vet.42:133-147.

28. Schleifer, K. H., and U. Fischer. 1982. Description of a newspecies of the genus Staphylococcus: Staphylococcus carnosus.Int. J. Syst. Bacteriol. 32:153-156.

29. Schleifer, K. H., U. Geyer, R. Kilpper-Bâlz, and L. A. Devriese.1983. Elevation of Staphylococcus sciuri subsp. lentus (Kloos etal.) to species status: Staphylococcus lentus (Kloos et al.)comb. nov. Syst. Apple. Microbiol. 4:382-387.

30. Schleifer, K. H., R. Kilpper-Balz, U. Fischer, A. Faller, and J.

Endi. 1982. Identification of "Micrococcus candidus" ATCC14852 as a strain of Staphylococcus epidermidis and of"Micrococcus caseolyticus" ATCC 13548 and Micrococcusvarians ATCC 29750 as members of a new species, Staphylo-coccus caseolyticus. Int. J. Syst. Bacteriol. 32:15-20.

31. Schleifer, K. H., and W. E. Kloos. 1975. Isolation and charac-terization of staphylococci from human skin. I. Amended de-scriptions of Staphylococcus epidermidis and Staphylococcussaprophyticus and descriptions of three new species: Staphylo-coccus cohnii, Staphylococcus haemolyticus, and Staphylococ-cus xylosus. Int. J. Syst. Bacteriol. 25:50-61.

32. Skalka, B., J. Smola, and J. Pillich. 1979. A simple method ofdetecting staphylococcal hemolysins. Zentralbl. Bakteriol.Mikrobiol. Hyg. I. Abt. Orig. A 245:283-286.

33. Spiegel, C. A., and M. Roberts. 1984. Mobiluncus gen. nov.,Mobiluncus curtisii subsp. curtisii sp. nov., Mobiluncus curtisiisubsp. holmesii subsp. nov., and Mobiluncus mulieris sp. nov.,curved rods from the human vagina. Int. J. Syst. Bacteriol.34:177-184.

34. Wilkinson, H. W. 1977. CAMP-disk test for presumptive iden-tification of group B streptococci. J. Clin. Microbiol. 6:42-45.

35. Williams, R. E. O., and G. J. Harper. 1947. Staphylococcalhaemolysins on sheep blood agar with evidence for a fourthhaemolysin. J. Pathol. Bacteriol. 59:69-78.

36. Zemelman, R., and L. Longeri. 1965. Characterization of staph-ylococci isolated from raw milk. Apple. Microbiol. 13:167-170.

VOL. 22, 1985

on February 2, 2020 by guest

http://jcm.asm

.org/D

ownloaded from