Journal of General Microbiology (1970), 61, 361-369
`Printed in Great Britain
`
`Nature and Properties of a Cytolytic Agent
`Produced by Bacillus subtilis
`
`By A. W. BERNHEIMER A N D LOIS S. AVIGAD
`Department of Microbiology, New York University School of Medicine,
`New York, N . Y. 10016, U.S.A.
`1970)
`(Accepted for publication 23 February
`
`SUMMARY
`The substance responsible for lysis of erythrocytes by cultures of Bacillus
`subtilis, designated subtilysin, was purified. It contained a peptide of leucine,
`aspartic acid, glutamic acid and valine and probably a lipid. Subtilysin was
`activated by Mg2+, Mn2+ and Ca2+. The rate of haemolysis was abruptly
`increased by chilling the reaction mixture. Haemolysis was inhibited by
`normal sera; the most inhibitory serum fractions contained a- and
`,&globulins. Haemolysis was inhibited by low concentrations of phosphatidyl-
`choline, phosphatidylinositol, phosphatidic acid and sphingomyelin. Subti-
`lysin possessed antibiotic properties and lysed protoplasts and spheroplasts
`derived from several bacterial species ; subtilysin was identical with surfactin,
`a peptidelipid from B. subtilis cultures that inhibits fibrin clot formation.
`Kakinuma and co-workers found surfactin to be a heptapeptide having an
`N-terminal glutamic acid in amide linkage with the carboxyl group of 3-
`hydroxy- I 3-methyl-tetradecanoic acid. Surfactin (subtilysin) possesses some
`properties in common with two other cytolytic agents of bacterial origin,
`namely, staphylococcal &-toxin and streptolysin S .
`
`I N T R O D U C T I O N
`In an earlier study a series of aerobic sporogenic bacilli was surveyed for production
`of haemolysin active against rabbit erythrocytes (Bernheimer & Grushoff, I 967).
`Among the species examined, broth cultures of Bacillus cereus, B. alvei and B. latero-
`sporus were found to contain relatively potent lysins that were activated by SH-
`compounds and that belonged to the group of oxygen-labile haemolysins of which
`streptolysin 0 is the prototype (Bernheimer, 1970). B. subtilis produced a haemolysin
`which appeared to differ from these. What can be assumed to be the same lysin of
`B. subtilis was studied by Busing (1950) and later by Williams (I957), who found the
`active moiety had the solubility properties of an organic acid. To establish more
`exactly its nature and properties, we have re-investigated this lytic agent; it is pro-
`visionally designated ' subtilysin '.
`
`METHODS
`was supplied by Dr H. S . Levinson.
`Organism. Bacillus subtilis ~ ~ ~ 1 2 2 8
`Reagents. Phosphatidylcholine and diphosphatidylglycerol were bought from
`Sylvana Co. (Milburn, N. J.) ; phosphatidylserine, phosphatidylethanolamine, sphingo-
`myelin and cerebrosides from Applied Science Laboratories, Inc. (State College, Pa.) ;
`phosphatidic acid from General Biochemicals (Chagrin Falls, Ohio) ; phosphatidyl-
`
`
`1 of 9
`
`FRESENIUS-KABI, Exh. 1023
`
`

`
`362
`A. W. BERNHEIMER A N D L. S. AVIGAD
`inositol and serum fractions from Nutritional Biochemicals Corp. (Cleveland, Ohio) ;
`cholesterol from Matheson, Coleman and Bell (East Rutherford, N. J.).
`Measurement of haemoZytic activity. In the early part of the investigation, capacity
`of subtilysin to lyse washed rabbit erythrocytes was measured as for staphylococcal
`a-toxin (Bernheimer & Schwartz, 1963) but with 0.1 % (w/v) gelatin instead of bovine
`serum albumin. This method utilized phosphate-buffered saline solution and an in-
`cubation period of 30 min. at 37"; later, the method was altered by substituting for
`phosphate buffer, tris buffer + Mg and by incubating the mixtures of lysin and cells
`for 30 min. at 37" followed by 30 min. in an ice-bath. These changes resulted in titres
`about three times as great as those obtained under the conditions first used. In the
`method finally adopted, test preparations were diluted in 0.84 % (w/v) NaCl (buffered
`at pH 7.2) with 0.01 M-triS containing 0.1 % (w/v) gelatin+o.orM-MgCI,. To I ml. of
`each of a series of dilutions increasing in about 30 % steps was added I ml. of a twice-
`washed suspension of rabbit red blood cells. The medium in which the cells were
`washed and suspended was 0.84 % (w/v) NaCl buffered at pH 7-2 with 0.01 M-tris.
`The concentration of the red cell suspension (about 0.7 %, v/v) was adjusted so that
`a sample, after lysis with saponin and addition of an equal volume of diluent, gave
`a spectrophotometric (Zeiss) extinction of 0.8 with light of 545 nm. and a light path
`of I cm. The mixtures of subtilysin dilutions and red blood cells were put in a 37"
`water-bath for 30 min., removed to an ice-bath for 30 min. and then centrifuged briefly.
`The colour of the haemoglobin in the supernatant fluids was compared visually with
`standard haemoglobin solutions, and the dilution haemolysing 50 % of the red cells
`in the suspension determined by interpolation. A unit of subtilysin is defined as that
`amount which liberates half the haemoglobin in the test red cell suspension under the
`conditions stated.
`Production of subtilysin. In preliminary experiments the quantity of subtilysin pro-
`duced in cultures utilizing a variety of media was estimated by titrating culture super-
`natant fluids for haemolytic activity; stationary and shaking cultures were used. The
`casein basal medium of Knight & Proom (1950) supplemented with a small amount
`of yeast extract gave satisfactory titres of subtilysin. 10 g. of Casamino acids (Difco
`Laboratories, Detroit, Michigan), 5 g. KH2P04, I g. NaCl, IOO mg. tryptophan and
`10 mg. cystine were dissolved in a litre of water, and the mixture adjusted to pH 7.0,
`the medium brought to a boil, filtered through paper, and distributed in 50 ml. amounts/
`250 ml. Erlenmeyer flask. Sterilization was at 123" for 20 min. To each flask was added
`0.13 ml. 20 % (w/v) solution of yeast extract (Difco Laboratories, Detroit, Michigan).
`Growth and subtilysin formation in this medium are illustrated by the data of
`Table I derived from flasks which received an inoculum of 5 ml. and incubated at
`37" in a shaking water-bath. Within limits of error of measurement the haemolytic
`activity paralleled the degree of growth. Addition to the medium of glucose, sodium
`ribonuclease or serum did not yield significantly higher titres.
`For routine production of subtilysin a seed culture was prepared by adding a loopful
`of growth from an agar plate to 50 ml. medium and incubating in a shaking water-bath
`to an extinction of about 1-0 at 650 nm. Eight flasks containing 50 ml. medium were
`each inoculated with I ml. of a IO* dilution of seed culture and incubated in a shaking
`water-bath at 37" for 17 h. Supernatant fluids of the cultures usually contained 20
`to 40 subtilysin units/ml.
`
`
`2 of 9
`
`

`
`Cytolytic agent of B. subtilis
`
`363
`
`RESULTS
`Pur$cation of subtilysin
`Unless otherwise noted purification was at room temperature. The combined super-
`natant fluids from a series of cultures (stage I, Table 2) were concentrated seven- to
`tenfold by pervaporation through Cellophane sacks. The decrease in volume was
`accompanied by formation of a precipitate (stage 2) which was collected by centrifuga-
`
`Table I . Growth of Bacillus subtilis and course of appearance of subtilysin
`Growth as
`Time of
`incubation
`extinction at
`650 nm.
`(W
`0.62
`4 7
`4
`I3 18
`2'1
`16
`3'5
`23
`24
`5'2
`* Units of haemolytic activity/ml. obtained by titrations utilizing phosphate-buffered saline and
`incubation time of 30 min.
`
`Subtilysin in culture
`supernatant fluid*
`
`Table 2 . PuriJication and recovery of subtilysin
`
`Stage I . Culture supernatant fluid
`Stage 2. Precipitate from pervaporation
`Stage 3. Acid precipitate
`Stage 4. Salt precipitate
`Stage 5. Dialysed and lyophilized
`product, 234 mg.
`
`Volume
`(ml.)
`1,055
`134
`30
`30
`-
`
`Total units
`of subtilysin
`
`Recovery of
`activity
`(%I
`I 0 0
`I 0 0
`43
`52 57
`
`tion and which contained virtually all the subtilysin activity. The supernatant fluid,
`containing less than 10 units of subtilysin/ml. was discarded. The precipitate was
`washed twice with 30 ml. 10 % (w/v) NaCl and then dissolved in 30 ml. distilled water.
`A small amount of insoluble material was removed by centrifugation, and the super-
`40. The precipitate which formed was
`natant fluid was adjusted with N-HC~ to pH
`washed twice with 30 ml. 0-1 N-acetate buffer (PH 4), and then suspended in 30 ml.
`o-og~-tris (pH 7-5). Sufficient NaOH was added to the mixture to bring to slightly
`above neutral pH and solution was left to occur overnight in the cold (stage 3).
`A small amount of insoluble material was removed by centrifugation, and solid
`NaCl was added to 10 % (w/v). After standing 60 min. the mixture was centrifuged
`for 10 min. at 10,000 rev./min. The opalescent supernatant fluid, which contained 50
`units subtilysin/ml., was discarded. The precipitate was stirred with 30 ml. 0.05 N-tris
`(PH 75) allowing several hours for solution to occur (stage 4). The solution was
`dialyzed at 4" against two changes of distilled water (1800 ml.) over about 20 hr, and
`A
`white solid (234 mg.) was obtained (stage 5). This product had
`then freeze-dried.
`a specific activity of 86 haemolytic unitslmg. The specific activities of a series of
`such products varied from 55 to 95 haemolytic unitslmg. Further purification was
`
`
`3 of 9
`
`

`
`A. W. BERNHEIMER AND L. S. A V I G A D
`
`364
`attempted by iso-electric focusing and by use of a Sephadex G-IOO column. Upon
`iso-electric focusing, subtilysin precipitated as it moved into the region pH 4 to 5;
`in Sephadex no further increase in specific activity was achieved.
`
`Physical and chemical properties
`Subtilysin was soluble in slightly alkaline water, in ethanol, acetone and chloroform;
`it was precipitated from aqueous solution by 10 % NaCl (w/v) or by acidification to
`pH 4;it did not diffuse through Cellophane. Solutions retained their haemolytic activity
`at 100' for 15 min. at pH 8.2 or 9.6, but at pH 3-0 two-thirds of the activity dis-
`appeared. The haemolytic activity of subtilysin was not affected by treatment with
`trypsin, papain or pronase. These properties suggested that the active material was of
`relatively large molecular size but was apparently not a protein.
`
`Wavelength @)
`6
`7
`
`8 9 1 0
`
`I
`
`I
`
`I
`
`I
`
`
`
`100
`
`-
`
`5
`100
`80 -
`n 8
`
`e
`
`6 0 -
`
`4 c.)
`
`.r( g 40-
`-
`
`i! c 1 '
`20
`
`"
`
`1
`
`1
`
`1
`
`1
`
`
`
`0
`1
`2000'
`
`1000
`
`10
`
`20
`
`1
`1
`1500
`Frequency (cm. -l)
`Fig. I
`Fig. I . Infrared spectrum of subtilysin.
`Fig. 2. Course of haemolysis in presence of 0.005~-MgCl, (open circles) and in absence of
`added MgP+ (solid circles).
`
`50
`
`60
`
`40
`30
`Minutes
`Fig. 2
`
`A solution containing 7 mg. subtilysin/ml. 0*05~-tris buffer (PH 7.5) appeared to be
`homogeneous in the analytical ultracentrifuge, giving a single peak having an Szow
`of about 1-2. Using a column of Sephadex G-IOO (superfine) equilibrated with 0.03 M-
`sodium borate buffer (PH 83) containing 0.1 M-KCl, and calibrated with bovine serum
`albumin, ovalbumin and cytochrome c, elution volumes plotted according to Andrews
`(1964) gave a molecular weight of about 50,000.
`Subtilysin was estimated to contain 8.8 % nitrogen (micro-Kjeldahl analysis), less
`than 0-3 % phosphorous (Lohmann & Jendrassik, 1926) and less than I % carbohydrate
`as glucose by the anthrone reaction (Colowick & Kaplan, 1957). When assayed for
`protein by the method of Lowry, Rosebrough, Farr & Randall (1951) it gave a colour
`equivalent to 4 % of that produced by an equal weight of ovalbumin. Subtilysin after
`hydrolysis for 22 hr at 110' in 6*g~-HCl yielded a ninhydrin colour equivalent to
`75 % of that given by an equal weight of ovalbumin.
`Analysis for amino acids (Blackburn, 1968) showed leucine, aspartic acid, glutamic
`acid and valine in molar ratio near 4: I : I : I. Infrared absorption (KBr) yielded a
`
`
`4 of 9
`
`

`
`Cytoi'ytic agent of B. subtilis
`365
`spectrum (Fig. I) consistent with lipid as well as of a peptide. The presence of lipid
`could also account for the solubility properties, the nitrogen content and the low
`sedimentation coefficient in relation to molecular size as determined by gel filtration.
`The formula weight of subtilysin is 1036; therefore the material formed aggregates of
`large size. Consistent with this was a tendency for subtilysin to give solutions that
`were opalescent.
`Haemolysis and inhibition of haemolysis
`Divalent cations. The following experiment shows the effect of MgC12 on haemo-
`lysis. 15 ml. tris-buffered saline (0.84 %, w/v, NaCl at pH 7-2 with 0.1 M-tris) containing
`35 haemolysin units was mixed with 15 ml. 0.7 % (vlv) washed rabbit erythrocytes in
`tris-buffered saline; both solutions were pre-warmed to 37"; a similar mixture was
`prepared with 0.005 M-MgCl, present. Samples were taken at intervals and immediately
`centrifuged, and the percentage haemolysis estimated from light absorption readings
`at 545 nm. The results (Fig. 2) show that Mg substantially increased the rate of lysis.
`A similar effect was produced when Ca or Mn were substituted for Mg.
`
`5 10 15 20 25 30 35 40 45 50 55 60
`Minutes
`Fig. 3. Effect of chilling on course of haemolysis. Open circles: reaction mixture at 37" throughout:
`solid circles: reaction mixture at 37" for 30 min. and then chilled.
`
`' Hot-cold' eflect. The experimental conditions were like those described, but the
`concentration of subtilysin was about two-thirds as great, and 0.005 M-M~CI, was in
`each of two flasks of reaction mixture: one was at 37" for the duration of the experi-
`ment: the other was at 37" for 30 min. and then chilled in an ice-bath. The results
`(Fig. 3) show that chilling increased abruptly the rate of lysis.
`Source of erythrocytes. Assays of the haemolytic activity of a single preparation of
`subtilysin with erythrocytes from four different rabbits agreed within 10 %; the sensi-
`tivity of erythrocytes derived from man and guinea pig resembled that of rabbit cells.
`About four times as much subtilysin was required to produce 50 % lysis of erythrocyte
`suspensions derived from calf, sheep and goat as for rabbit erythrocytes. The sensi-
`tivity of red cells from horse, pig, cat and chicken was less than that of rabbit cells
`but greater than that of cells from calf, sheep and goat.
`Inhibition of haemolysis by serum and serum fractions. The capacity of serum to
`inhibit the haemolytic activity of subtilysin was assayed by mixing decreasing con-
`centrations with a fixed amount (3 haemolysin units) of subtilysin, allowing the mix-
`tures of I ml. each to stand for 10 min. at 20"' and then adding I ml. 0.7 % (v/v) washed
`rabbit red cells. After 30 min. at 37" followed by 30 mi.. in an ice-bath, the mixtures
`were centrifuged briefly and the haemoglobin in the supernatant fluids estimated
`colorimetrically at 545 nm. The tests were done in tris-buffered saline containing
`
`
`5 of 9
`
`

`
`S.
`366
`A. W. BERNHEIMBR AND L.
`A V I G A D
`o*oo5~-MgC1, and 0.05 % gelatin; 50 % haemolysis was used as the endpoint of the
`titrations. About 0.005 ml. normal serum, whether from man, rabbit, calf or horse, was
`required to inhibit 2 haemolysin units, i.e. two-thirds the test amount of subtilysin.
`Assay of several Cohn fractions of human serum showed that they inhibited haemo-
`lysis in order of decreasing effectiveness as follows : a-globulin (fraction IV-I), ,8-
`lipoprotein (fraction 111-o), a-globulin (fraction IV-4) and albumin ; y-globulin did
`not inhibit up to 5 mg./ml., the highest concentration tested.
`Table 3. Efect of lipids on subtilysin haemolysis of rabbit red cells
`Concentration required
`to inhibit two-thirds of
`test amount of
`subtilysin
`(lug. Id .)
`
`Phosphatidylcholine (egg)
`Phosphatidylcholine (beef)
`Phosphatidylinositol
`Phosphatidylserine (brain)
`Phosphatidylserine (plant)
`Phosphatidylethanolaine (plant)
`Diphosphatidylglycerol (beef heart)
`Phosphatidic acid
`Sphingomyelin (beef heart)
`Cerebrosides (beef brain)
`Cholesterol
`Inhibition of haemolysis by lipids. The foregoing results suggested that lipids may
`play an important role in inhibition of lysis by serum. A variety of lipids suspended or
`dissolved in 0.85 % (w/v) NaCl at I mg./ml. were tested for inhibition of lysis by adding
`decreasing amounts to 3 haemolysin units of subtilysin. The conditions were the same as
`those used for testing inhibition by serum and serum fractions. To improve dispersion,
`most of the preparations were briefly treated by sonication before testing. The results
`(Table 3) show that low concentrations of several phosphatides, most notably phos-
`phatidylcholine, phosphatidylinositol, phosphatidic acid and sphingomyelin, in-
`hibited subtilysin haemolysis. A high concentration of cerebrosides was required to
`produce inhibition; cholesterol did not inhibit at all.
`
`Bacteriolytic and antibiotic activity
`Protoplasts and spheroplasts were prepared from a variety of bacteria according to
`methods used earlier (Bernheimer & Schwartz, 1965 ; Bernheimer, 1966). Extinctions
`of suspensions of protoplasts and spheroplasts were recorded continuously with a Cary
`spectrophotometer at 500 nm. for 30 min. at about 20°, with or without subtilysin.
`Substantial decrease in extinction was interpreted as evidence of lysis ; estimates of
`lytic activity were made from the quantity of subtilysin needed to cause half-maximum
`decrease in extinction. The results (Table 4) show that protoplasts of a variety of
`Gram-positive bacteria were lysed, whereas among Gram-negative bacteria sphero-
`plasts of Escherichia coli were lysed while those of Vibrio comma were not. The con-
`centration of subtilysin required to disrupt protoplasts of the more sensitive bacterial
`species was about the same as that needed to lyse rabbit red cells.
`Antibiotic activity was assayed by the tube dilution method. Doubling dilutions of
`
`
`6 of 9
`
`

`
`Cytolytic agent of B. subtilis
`367
`subtilysin were prepared in trypticase soy broth for all organisms with the exception
`of Streptococcus pyogenes; for the streptococcus, Todd Hewitt broth was used. Each
`assay tube containing subtilysin in 2 ml. sterile broth was inoculated with a 5 mm.
`loopful of a 20 hr culture of test bacterium, and incubated at 37" for 3 days. Growth
`was estimated visually and scored at o to 4 + . Growth of S. pyogenes was completely
`inhibited by subtilysin in a concentration of 0.6 haemolysin units/ml. (about 6 ,ug./ml.).
`Corynebacterium diphtheriae and Bacillus megaterium were completely inhibited by
`I and 2 haemolysin units/ml., respectively. Sarcina lutea and Micrococcus lysodeikticus
`were partially inhibited by I 5 haemolysin units/ml., whereas Escherichia coli was not
`affected by this concentration, the highest tested. Similar results were obtained in
`experiments with subtilysin-containing paper discs on agar plates, but S. lutea was
`more sensitive and B. megaterium less sensitive than would have been predicted from
`the data derived from the broth assays. No inhibition of growth was obtained with
`Staphylococcus aureus, Gamya tetragena, Streptococcus faecalis, Bacillus cereus, B.
`subtilis, Lactobacillus casei, E. coli or Pseudomonas aeruginosa.
`Table 4. Lysis of bacterial protoplasts and spheroplasts by subtilysin
`Approximate concentration
`producing half-maximal
`reduction in turbidity
`cug./ml.)
`5
`7
`
`BuciZZus rnegaferiurn KM protoplasts (1YSOZyme)
`Streptococcus pyugenes c203 s protoplasts
`(phageassociated enzyme)
`S. pyogenes GL 8 protoplasts (phage-associated
`enzyme)
`Escherichia coZi K 12 spheroplasts (lysozyme)
`E. coZi K 12 spheroplasts (penicillin)
`protoplasts (lysozyme)
`S. faecalis A T C C ~ ~ ~ O
`Sarcina luteu protoplasts (lysozyme)
`Vibrio comma spheroplasts (penicillin)
`* N means no lysis at 100 pg./ml.
`
`10
`
`I 0
`I 1 20
`65 N*
`
`DISCUSSION
`The agent responsible for the haemolytic activity of Bacillus subtilis was present in
`cultures in relatively large amounts, and it seems likely that it would have been seen
`before. The haemolytic moeity examined by Williams (1957) was probably subtilysin,
`and the properties of a substance termed serolysin by Aida, Koyama & Uemura (1964)
`suggest that it may be the same as the material we have studied. While the present
`work was in progress there came to our attention a paper by Arima, Kakinuma &
`Tamura (1968) describing the isolation and characterization of a B. subtilis product
`that is a potent inhibitor of blood clotting. The clotting inhibitor, named surfactin,
`is a peptidelipid whose structure has been elucidated (Kakinuma et al. 1969a, b, c) as
`
`Comparison of the properties of subtilysin and surfactin showed that they possess
`many features in common, not least being a peptide of the same amino acid composi-
`61
`24
`
`M I C
`
`
`7 of 9
`
`

`
`368
`A. W. BERNHEIMER A N D L. S. A V I G A D
`tion. A sample of surfactin supplied by Dr Kakinuma was found to have the same
`specific haemolytic activity for rabbit red cells as subtilysin. Moreover, the two
`substances showed a similar time-course of haemolysis (Fig. 4). A sample of subtilysin
`sent to Tokyo was examined for inhibition of the thrombin-fibrinogen reaction by
`Dr Kakinuma who provided the plot reproduced as Fig. 5. The clotting system con-
`sisted of 0.5 % fibrinogen (Amour Bovine Fraction 1 containing about 30 % clottable
`protein), subtilysin as indicated, thrombin 5 units/ml. and 0.01 M-tris+ 0.073 M-NaCl
`(pH 7'4). Clotting time, the interval from thrombin addition to clot formation, was
`measured at 37". The combined results left no doubt about the identity of subtilysin
`with surfactin; it would therefore seem appropriate to abandon the former designation.
`
`0
`
`80 I
`
`30
`20
`
`5 10 15 20 25 30 35 40 45 50 55 60
`150
`100
`50
`Subtilysin (pg./ml.)
`Minutes
`Fig. 5
`Fig. 4
`Fig. 4. Course of haemolysis bysubtilysin, 9 pg./ml. (open circles), and by surfactin, 10 pg./ml.
`(solid circles).
`Fig. 5. Inhibition of clotting by subtilysin.
`
`0
`
`200
`
`Surfactin possesses certain features common to at least two other agents of bacterial
`origin: staphylococcal 8-toxin and streptolysin S. As extracellular products of growth
`all are lytic, not only for mammalian erythrocytes but also for protoplasts and
`spheroplasts of certain bacteria. The lytic effects of all three are inhibited by phospho-
`lipids, and presumably phospholipids are the constituents of plasma membranes with
`which they interact to produce cell lysis. The biological activity of surfactin appears
`to depend upon the amphipathic nature of the molecule. One can speculate that the
`same may prove true of staphylococcal &toxin. It is noteworthy that surfactin and
`streptolysin S both contain peptides: neither of them is antigenic nor is &toxin.
`
`The authors acknowledge with thanks the help of Dr A. Kreger who did the assays
`of antibiotic activity and the tests of sensitivity of different red cell species. To Mr D.
`Green and Dr B. L. Van Duuren we are indebted for infrared absorption data. We
`thank Mr C. Harman for operating the amino acid analyser and Mr F. Zaboretsky
`for making the ultracentrifugal analysis. This work was supported in part by a grant
`(AI-02874) from the National Institute of Allergy and Infectious Diseases of the U.S.
`Public Health Service, by a grant from the Life Insurance Medical Research Fund,
`and by Public Health Service Research Career Program Award 5K6-AI-14, 198.
`
`
`8 of 9
`
`

`
`Cytolytic agent of B. subtilis
`
`369
`
`REFERENCES
`Y.
`
`& UEMURA, T. (1964). Serolysin produced by Bacillus subtilis 427-4.
`ADA, T., KOYAMA,
`Chemical
`3458.
`Abstracts
`P. (1964). Estimation of the molecular weights of proteins by Sephadex gel-liltration.
`ANDREWS,
`Biochemical Journal 91, 222-233.
`ARIMA, K., KAKINUMA, A. & TAMURA, G. (1968). Surfactin, a crystalline peptidelipid surfactant
`produced by Bacillus subtilis : isolation, characterization and its inhibition of fibrin clot forma-
`tion.
`Biochemical and Biophysical Research Communications 31, 488-494.
`(1966). Disruption of wall-less bacteria by streptococcal and staphylococcal
`BERNHEIMER, A. W.
`toxins. Journal of Bacteriology 91, 1677-1680.
`BERNHEIMER, A. W. (1970). In Microbial Tuxim, vol. I, ch. 5. Edited by S. J. Ajl, T. C. Montie and
`S. Kadis, New York: Academic Press. (In the Press.)
`BERNEIEIMER, A. W. & GRUSHOFF, P. (1967). Cereolysin: production, purification and partial charac-
`46, 143-150.
`terization. Journal of General Microbiology
`A. W.
`& SCHWARTZ, L. L. (1963). Isolation and composition of staphylococcal alpha
`BERNHEIMER,
`455-468.
`toxin. Journal of General Microbiology 30,
`W.
`& SCHWARTZ, L. L. (1965). Lysis of bacterial protoplasts and spheroplasts by
`BERNHEIMER, A.
`staphylococcal a-toxin and streptolysin S. Journal of Bacteriology Sg, 1387-1392.
`S. (1968).
`Amino Acid Determination. New York: Marcel Dekker, Inc.
`BLACKBURN,
`BUSING, K. H. (1950). Notizen uber das Subtilis-Haemolysin. Archiv f i r Hygiene, Berlin 133, 63-68.
`S. P.
`N. 0. (1957). Anthrone reaction. Methodr in Enzymology 3,
`84-85.
`COLOWICK,
`& -LAN,
`A., HORI, M., ISONO, M., TAMURA, G. & ARIIU,
`K. (1969a). Determination of amino acid
`KAKINUMA,
`sequence in surfactin, a crystalline peptidelipid surfactant produced by Bacillus subtilis. Agri-
`cultural and Biological Chemistry 33, 971-972.
` G. & ARIMA, K. (19696).
`KAKINUMA, A., HORI, M., SUGINO, H., YOSHIDA, I., IsONO, M., T m ,
`Determination of the location of lactone ring in surfactin. Agricultural and Biological Chemistry
`33, 1523-1524.
`G.
`& ARIMA, K. (1969~). Determination of fatty
`KAKINUMA, A., SUGINO, H., ISONO, M., TAMURA,
`acid in surfactin and elucidation of the total structure of surfactin. Agriculturd and Biological
`Chemistry 33, 973-976.
`H. (1950). A comparative survey of the nutrition and physiology of KNIGHT, B. C. J. G. & PROOM,
`
`4, 508-538.
`mesophilic species in the genus Bacillus. Journal of General Microbiology
`L~HMANN, K. & JENDRASSIK, L. (1926). Kolorimetrische Phosphorslurebestimungen in Muskel-
`178, 419-426.
`extrakt. Biochemische Zeitschrift
`LOWRY, 0. H.,
`RANDALL,
`(1951). Protein measurement with
`ROSEBROUGH, N. J., FARR, A. L. &
`R. J.
`the Folin phenol reagent. JournaZ of BiologicaZ Chemistry 193, 265-275.
`R. (1957). Haemolytic material from aerobic sporing bacilli. Journal of General
`WILLIAMS, G.
`Microbiology 16,
`6-2 I.
`I
`
`24-2
`
`
`9 of 9