J Clin Pathol: first published as 10.1136/jcp.7.2.160 on 1 May 1954. Downloaded from

J. clin. Path. (1954), 7, 160.

THE USE OF A BILE-AESCULIN MEDIUM AND OF MAXTED'S TECHNIQUE OF LANCEFIELD GROUPING IN THE IDENTIFICATION OF ENTEROCOCCI (GROUP D STREPTOCOCCI) BY ALEXANDER SWAN From the Department of Pathology, King's College Hospital Medical School, University of London

(RECEIVED FOR PUBLICATION SEPTEMBER 5, 1953)

The name , originally given by 82 strains of Strep. viridans. Weatherall and Thiercelin (1899) to a Gram-positive coccus of Dible (1929), who obtained similar results, 60 of faecal origin growing in pairs and short chains, their 61 bile-tolerant, heat-resistant, mannitol- has been widely used for a group of morpho- fermenting diplococci hydrolysing aesculin, con- logically similar streptococci found in the gut of sidered, however, that the hydrolysis of aesculin man and other mammals, in dairy products, and in a bile medium was " very little more than an in soil. The limits of this group, formerly rather indicator" of bile tolerance. This conclusion was vague, were satisfactorily defined with the appli- probably responsible for the fact that the test hascopyright. cation to it of Lancefield grouping methods by been little used. The work of Williams and Hirch Sherman and his collaborators (Sherman, 1937; (1950) suggests, however, that aesculin hydrolysis Sherman, Mauer, and Stark, 1937; Sherman, has an independent value in the classification of Stark, and Mauer, 1937; Sherman, Stark, and bile-tolerant cocci, as only 83.6% of their 140 Yawger, 1937), and later by Shattock and Mattick strains of bile-tolerant streptococci hydrolysed (1943) and Shattock (1945 and 1949b). From this aesculin. http://jcp.bmj.com/ work it is apparent that the great majority, if not The present paper records an attempt to find a all, enterococci belong to Lancefield Group D, simple, rapid and accurate single procedure for and that in fact this serological property should the recognition of enterococci as a group, minor be taken as the basis of their classification. The subdivisions of which are of little importance in group will then include the following species: the routine diagnostic work of a medical labora- Strep. faecalis, with its varieties, zymogenes, and tory. Results of Lancefie!d D grouping were com- liquefaciens, Strep. durans, and Strep. bovis. pared with those of tests for bile tolerance and on September 28, 2021 by guest. Protected The degree of variation in the biological proper- aesculin hydrolysis, as well as heat resistance, on ties of Group D streptococci is well shown in a number of freshly isolated and stock strains of Table I reproduced from Shattock (1949b ; see also streptococci. Nyman, 1949). Table I covers the majority of important bio- Source of Material logical properties of the group, except bile toler- One hundred and forty-two strains of strepto- ance, or the ability to grow in the presence of cocci and eight strains of micrococci were iso- bile in concentrations up to 40% (Weissenbach, lated from human material in the laboratory of 1918), which is usually regarded as common to a general hospital. The sources of these strains all species of enterococci, and aesculin hydrolysis. are shown in Table II. Attention was drawn to the value of aesculin In addition, nine stock strains of Strep. bovis, hydrolysis in the identification of enterococci by three isolated from human intestine and' six from Rochaix (1924), and the use of aesculin in bile the rumen of the cow, and 12 strains of Strep. media was regarded as the best single differential durans were examined. Of the latter, two were test by Meyer and Schonfeld (1926), who found recovered from the human intestines, the re- that 61 of their 62 strains of enterococci hydro- mainder being mostly obtained from milk and lysed aesculin, as compared with only six among dairy products. J Clin Pathol: first published as 10.1136/jcp.7.2.160 on 1 May 1954. Downloaded from BILE-AESCULIN MEDIUM 161

TABLE I DIFFERENTIATION OF SPECIES WITHIN GROUP D REPRODUCED FROM SHATTOCK (1949a)

Strong Survival Growth Gelatin Reduction Mantl Scoe Rfioe at 600 C. Species Haemolysis Liquefaction of Litmus Mannitol Sucrose at at At Milk 30 min. 450 C. pH=9-6 Strep. faecalis .. _ _ + + ± + + + Strep. faecalis, vari- ety zymogenes + + + + + + + Strep. faecalis, vari- ety liquefaciens. . _ + + + + _ + + + Strep. durans . bovis . - | ± + + | + -

±=Varies from strain to strain. TABLE II Lancefield Grouping.-(I) The hot-HCl-hydrolysis SOURCES OF STRAINS ISOLATED IN THE HOSPITAL technique of Lancefield (1933) was employed through- LABORATORY out, using the sediment from 10 ml. of overnight glucose-broth cultures. Total Sources PeupiePresumptiveStrep. Micrococci Numbers of (2) All freshly isolated strains (150) were moreover EresmcctivrEtrcci viridans Strains grouped by a modification of the enzyme method of Faeces .. 65 1 8 74 Maxted (1948). The enzyme was prepared by freezing Urine.. .. 38 0 0 38 and thawing an agar culture of Streptomyces albus. Vagina .. 10 0 0 10 Wounds, ab- The fluid collected after thawing, which contains the scesses, etc. 5 0 0 5 enzyme, was stored without preservative at a tem- Blood cultures 2 0 0 2 Sputum .. 1 0 0 1 perature below 5° C.; under these conditions its acti- Throat .. 0 20 0 20 vity was retained for some months. Seitz filtration and pH adjustment were omitted on Mr. W. R. Total .. 121 21 8 150 Maxted's advice. The streptococci were grown in 10 copyright. ml. of glucose broth, or beef-digest broth, for 18 hours. After centrifuging, the supernatant fluid was Methods of discarded, and the deposit placed for two minutes in One hundred and thirty strains of Gram-posi- boiling water. After cooling, 0.5 ml. of the proteo- tive cocci were selected on the basis of their lytic extract was added and the mixture incubated at on McConkey medium. 50° C. overnight. As previously noted by Maxted http://jcp.bmj.com/ They appeared as small, discrete colonies, most (1948), Group D streptococci were found rather resis- of which were lactose fermenting. Microscopi- tant to the action of the enzyme, and even after this prolonged incubation the lysis was usually incom- cal examination of broth cultures showed eight plete, making recentrifugation necessary. Satisfactory of these strains to be micrococci and one to be a bacterial extracts were also obtained from enzyme- long-chained . The remaining 121 treated saline suspensions of growth on blood-agar strains, which grew in pairs or short chains, have, plates. for convenience, been referred to throughout this Streptococcal extracts prepared by both Lance- paper as "presumptive enterococci." field's and Maxted's methods were tested in the same on September 28, 2021 by guest. Protected To contrast with these, 20 strains of alpha- way by layering on to grouping serum (obtained from haemolytic streptococci were isolated from the Messrs. Burroughs Wellcome, Ltd.) in glass tubes of 1 mm. internal diameter. The results were read throats of healthy persons. These strains have during the 30 minutes. been classed with the alpha-haemolytic long- following chained streptococcus of faecal origin as "pre- Heat-resistance Test.-Approximately I ml. of an sumptive Strep. viridans." 18 to 24 hours' broth culture was placed in a 600 C. water-bath; a loopful was removed for subculture oni Methods of Teting blood-agar after 15, 30, 45, and 60 minutes. Strains surviving this temperature for 30 minutes were termed All strains were examined by the following tests: heat-resistant (Orla-Jensen, 1919 ; Dible, 1921). Lancefield grouping, heat-resistance, bile-tolerance, aesculin-hydrolysis. Bile-tolerance Test.-Heavy inocula were made on Sixty-eight strains were also tested for fermentation segments of a nutrient- containing 5% horse of mannitol and the ability to grow in nutrient broth serum, 0.1% aesculin, 0.05% ferric citrate, and suffi- containing 6.5% NaCl, on blood-agar containing cient "difco " dehydrated ox-gall powder to give a 0.06% potassium tellurite, and in milk containing final concentration corresponding to 40% bile (Wil- 0.1% methylene blue, with the decolorization of the liams and Hirch, 1950). The plates were examined dye. for growth after 24 hours' incubation. J Clin Pathol: first published as 10.1136/jcp.7.2.160 on 1 May 1954. Downloaded from 162 ALEXANDER SWAN

Hydrolysis of Aesculin.-This property was tested of them were neither heat-resistant nor bile- on the previously described bile-aesculin plate. Strains tolerant; two strains grew on bile-aesculin plates which hydrolysed aesculin produced a black colour in without hydrolysing aesculin, but neither and immediately around the colonies. Bile-tolerant was organisms which did not hydrolyse aesculin gave rise heat-resistant. to whitish colonies. The plates were incubated for None of eight strains of micrococci was heat- 24 hours; growth for a longer period was found not resistant and none hydrolysed aesculin, but six only unnecessary but also undesirable as it was liable were bile-tolerant. to obscure the results, when as many as 12 strains All Strep. bovis and Strep. durans strains were were cultured on one plate. As Weatherall and Dible bile-tolerant and hydrolysed aesculin, their extracts (1929) have pointed out, this is not truly a test of the reacted with Group D serum without concentra- capacity of an organism to hydrolyse aesculin, since tion. All strains of Strep. durans were, strains which are not bile-tolerant do not grow on this more- medium and are, therefore, not tested. over, heat-resistant, but only four out of the nine strains of Strep. bovis survived 600 C. for 30 Resuls minutes. Table IV shows the results of the subsidiary The results are summarized in Tables III and four tests, mannitol fermentation and growth in IV. media containing 6.5% NaCl, or 0.06% potassium Of the 121 " presumptive enterococci," 117 tellurite, and in milk containing 0.1 % methylene belonged to Lancefield Group D. All these blue, performed on 68 of the above 117 Group D, Group D strains were heat-resistant, bile-toler- heat-resistant, bile-tolerant, aesculin hydrolysing ant, and all hydrolysed aesculin. Of the four strains. Only 60 (88.3%) strains gave positive strains which did not react with Group D serum, results with all four tests. two were heat-resistant but intolerant to bile, and All cocci (150 strains), with the exception of two were neither heat-resistant nor bile-tolerant. stock strains of Strep. bovis and Strep. durans, These four strains were alpha-haemolytic. were grouped by both Maxted's and Lancefield'scopyright. None of the 21 strains of "presumptive Strep. methods. Both techniques gave identical precipi- viridans" reacted with Group D serum and 19 tation results with Group D sera in all cases except TABLE III one; this strain gave a positive Group D reaction RESULTS OF TESTS ON STRAINS OF ENTEROCOCCI on acid extraction but consistently negative results when extracted by Maxted's No. of method. Strains Lance- Heat Bile Aesculin http://jcp.bmj.com/ Total No. of with field Resis- Toler- Hydro- Discussion Strains Tested Common Group tance ance lysis Proper- D ties All 138 Group D strains, in this survey, and none outside this group were found to possess Presumptive entero- 117 + + + + cocci 121 2 _ + _ _ the property of aesculin hydrolysis. Bile toler-

- ance, although equally constant within the group, Micrococci 8 6 _ _ + was shared by some (six out of 33) strains out-

side it, and is therefore less specific for entero- on September 28, 2021 by guest. Protected Presumptive Strep. 2 _ + cocci. viridans 21 19 - _ Heat resistance is evidently neither a con- stant nor exclusive character of Group D strepto- Strep. bovis 9 .. 4 + + + + cocci, for five out of nine strains of Strep. bovis Strep. durans 12 .. 12 failed to survive 600 C. for 30 minutes, and on the + + + other hand two of the 25 strains of Strep. viridans (alpha-haemolytic streptococci not belonging to TABLE IV Group D) proved to be heat-resistant. RESULTS OF ADDITIONAL TESTS ON 68 STRAINS OF ENTEROCOCCI The behaviour of two strains of alpha-haemo- lytic, non-Group-D streptococci and of six strains Growth Growth Growth in in on Plate Medium of micrococci, which grew on the bile aesculin Medium Fer.men-of Containing Containing plate without hydrolysing aesculin, together with tation Contain-ing 65i Mannitol 0*06/%Pot. MethyleneO*1% the previously quoted figures from Williams and NaCl Tellurite Blue Hirch (1950), point to the value of including Numbers and Posi- 66 (97%/) 67 (98-5%.) 64 (94%) 63 (93%) aesculin in a bile medium. percentage tive of strains The bile-aesculin plate, therefore, appears to be Nega- 2 1 4 5 tive a simple reliable means for the recognition of enterococci. It is much simpler and cheaper than J Clin Pathol: first published as 10.1136/jcp.7.2.160 on 1 May 1954. Downloaded from BILE-AESCULIN MEDIUM 163

Lancefield grouping, since as many as 12 strains aesculin on a bile-containing medium it has been can be inoculated on one plate. concluded that the latter is a valid test for the The following procedure has been adopted in recognition of enterococci, having the advantage the routine work of this laboratory for the pre- of simplicity over the former. sumptive identification of enterococci. The Gram- Extracts for Lancefield grouping were prepared positive cocci are subcultured (a) in 10 ml. of by two methods: (1) the hot-HCl technique of glucose broth and (b) on a bile-aesculin plate. The Lancefield and (2) the enzymic method of Maxted. next day the morphology of the growth in glucose Both methods gave equally good results. broth is examined in a hanging drop. Strains growing in pairs and/or short chains and hydro- I am greatly indebted to Dr. A. C. Cunliffe for his lysing aesculin are classified as enterococci. The unstinted help and advice without which this work glucose broth culture is used only for the group- would have been impossible. I wish also to thank ing of those strains which fail to hydrolyse Mr. K. Tippett for general technical assistance, Mr. aesculin. W. R. Maxted for advice in regard to his method of grouping and for supplying us with his strain of It must be stressed that in this preliminary sur- Streptomyces albus, Miss M. E. Sharp for the 21 vey very small numbers of streptococci other than strains of Strep. bovis and Strep. durans, and Miss those of Group D have been examined. Thus, P. M. F. Shattock for the permission to reproduce her whereas the property of aesculin hydrolysis in a Table of variability of properties within Group D bile-containing medium has been satisfactorily streptococci (our Table I). demonstrated among enterococci, its specificity for the group remains to be proved. REFERENCES Dible, J. H. (1921). J. Path. Bact., 24, 3. Summary and Lancefield, R. C. (1933). J. exp. AMed., 57, 571. Conclusions Maxted, W. R. (1948). Lancet, 2, 255. Meyer, K., and Schonfeld, H. In an attempt to select a single reliable and (1926). Centraib. Bact., Abt. 1, 99, 402. copyright. Nyman, 0. H. (1949). Acta path. Microbiol. scand., Suppl. 83, 1. simple method for the identification of entero- Orla-Jensen, S. (1919). The Lactic Acid Bacteria. H0st, Copenhagen. cocci (Group D streptococci) the following princi- Rochaix, A. (1924). C.R. Soc. Biol., Paris, 90, 771. Shattock, P. M. F. (1945). Proc. Soc. agric. Bact., 1944, p. 69. pal tests were employed: Lancefield grouping, heat- - (1949a). Report XIIth International Dairy Congress, Stock- resistance, bile-tolerance, and aesculin hydrolysis. holm, Vol. 2, p. 598. HItggstroms, Stockholm. One hundred and seventeen Lancefield D (1949b). J. gen. AMicrobiol., 3, 80. Group -and Mattick, A. T. R. (1943). J. Hyg. Camb., 43, 173. strains were found to be heat-resistant, bile-toler- Sherman, J. M. (1937). Bact. Rev., 1, 3. http://jcp.bmj.com/ ant, and aesculin-hydrolysing. Of 25 alpha-haemo- - Mauer, J. C., and Stark, Pauline (1937). J. Bact., 33, 275. not none - Stark, Pauline, and Mauer, J. C. (1937). Ibid., 33, 483. lytic streptococci, of Group D, hydro- - and Yawger, E. S. (1937). Ibid., 33, 25. lysed aesculin, but two were heat-resistant and two Thiercelin, E. (1899). C. R. Soc. Biol., Paris, 51, 269, 551. were From the correlation Weatherall, C., and Dible, J. H. (1929). J. Path. Bact., 32, 413. bile-tolerant. very high Weissenbach, R. J. (1918). C.R. Soc. Biol., Paris, 81, 559. between Lancefield grouping and hydrolysis of Williams, R. E. O., and Hirch, A. (1950). J. Hyg., Lond., 48, 504. on September 28, 2021 by guest. Protected