ANTIMICROBIAL AGENTS AND CHzMOTHERAPY, June 1977, p. 1010-1016 Copyright 0 1977 American Society for Microbiology
Vol. 11, No.6' Printed in U.S.A .
Bactericidal Activity of Macromomycin, an Antitumor Antibiotic FRANCIS L. A. BUCKMIRE* AND JEFFREY L. WINKELHAKE Department of Microbiology, The Medical College of Wisconsin, Milwaukee, Wisconsin 53233 Received for publication 1 February 1977
Susceptibility testing by the broth dilution method showed that all the grampositive but only some of the gram-negative bacteria tested were susceptible to the antitumor antibiotic, macromomycin (MCR; NSC 170105). The minimal inhibitory concentration for the susceptible organisms was less than 3 ,g/ml. The action of MCR was bactericidal; however, at very high concentrations (50 ,ug/ml and above) some organisms occasionally escaped death. None of the escaped organisms was resistant to MCR. In combination with other commonly used antibiotics, MCR displayed partial synergy for Pseudomonas aeruginosa (from a minimal inhibitory concentration of >100 to 12.5 ug/ml with 100 ,ug of chloraimphenicol per ml) and for Bacillus pumilus and Staphylococcus aureus (from 1.6 to 0.4 ,ug/ml and below) with polymyxin B. As with mammalian cells, 125I-labeled MCR was irreversibly bound to both gram-positive and -negative bacteria. Treatment with trypsin of the mI-labeled MCR-exposed cells did not release the bound MCR or reverse its lethal effect. When in solution in a protective buffer at 4°C, MCR was stable for up to 45 days; at 370C, however, 25% of its bactericidal activity was lost in 72 h. Loss of activity was enhanced 16-fold in the presence of both heated and unheated pooled human sera. Urine had no effect on the activity of MCR.
Macromomycin (MCR; NSC 170105) is an unstable, antitumor (7, 9-12) polypeptide antibiotic discovered in 1968 by Chimura et al. (3). Recently, Yamashita et al. (16) have shown that its activity can be stabilized by certain substances, in particular the disaccharides, lactose and maltose. This has facilitated the purification of MCR to homogeneity (J.L. Winkelhake and F.L.A. Buckmire, Cancer Res., in press). As with other anititumor antibiotics recently isolated, the mode of action of MCR has not been studied in detail. Kunimoto et al. (6) and Winkelhake and Buckmire (Cancer Res., in press) have shown that radioiodine ('25I)-labeled MCR readily binds to mammalian cell surfaces and, presumably from that location, causes cessation of deoxyribonucleic acid (DNA), but not ribonucleic acid or protein, synthesis. Kunimoto et al. (7) have hypothesized that the interaction between MCR and the mammalian cells is analogous to the interaction between certain colicins and bacterial cells (4). Since the initial published data (3), there have been no other studies on the effect of MCR on bacteria. This communication compares the antimicrobial action of purified MCR on some
gram-positive and -negative bacteria with several commonly used antibiotics. MATERIALS AND METHODS Organisms. The bacteria used in this study were: Escherichia coli ATCC 25922 and another clinical isolate, Pseudomonas aeruginosa ATCC 27853 and five other clinical isolates, Staphylococcus aureus ATCC 25923 and two. clinical isolates, Bacillus pumilus ATCC 172, two laboratory strains of Bacillus subtilis, three strains of Haemophilus influ-
enzae type b, and four strains of Spirillum serpens. With the exception of H. influenzae and S. serpens, the cultures were cultivated on Trypticase soy agar (TSA) (Baltimore Biological Laboratories, Inc.)
plates. H. influenzae was grown on TSA supple-
mented with oxidized nicotinamide adenine dinucleotide (P-S Biochemicals, Milwaukee, Wis.), recrystallized hemin (Nutritional Biochemicals Corp., Cleveland, Ohio), and glucose (Calbiochem, La Jolla, Calif.) (1, 5), and S. serpens was cultivated on a Casamino Acids-salt medium containing agar (Difco Laboratories, Detroit, Mich.) (2). For all experiments, cultures were grown at 37°C for 8 to 14 h, harvested, and suspended in either the growth medium or HEPES (N-2-hydroxyethyl piperazine-N'-2ethanesulfonic acid) buffer. Cultures (excluding H.
influenzae) were maintained by bimonthly transfers and stored at 4°C. H. influenzae was stored at -70°C
on solid agar media, incubated overnight at 37°C,
1010
VOL. 11, 1977
BACTERICIDAL ACTIVITY OF MACROMOMYCIN
in Trypticase soy broth (TSB) containing 20% glycerol after 9 h of growth at 37°C. MCR. MCR was purified by sequential salt precipitation, ultrafiltration, ion exchange and molecular sieve chromatographies, and stabilized during lyophilization in 1% lactose-phosphate-buffered saline (Lac-PBS), pH 7.2, by a modification (Winkelhake and Buckmire, Cancer Res., in press) of the procedure of Yamashita et al. (16). During purification, fractions were monitored for activity by an agar-plate bacteriocidal technique, with B. pumilus as the test organism. Preparation homogeneity was determined by sodium dodecyl sulfate-polyacrylamide slab-gel electrophoresis, using 12.5% gel by the method of Laemli (8). A single protein band, which migrated between ribonuclease (13,700 daltons) and lysozyme (11,400 daltons) standards, was obtained. Since the electrophoresis technique employed is sensitive to about 10 ,ug of protein, our preparation was estimated to be >97% pure. Drug concentrations were determined spectrophotometrically with an 11.6 absorbancy value at 280 nm equivalent to 1%. Purified MCR was stored lyophilized in Lac-PBS at -70°C with no apparent loss of activity. MCR was radioiodinated with cellulose-immobilized lactoperoxidase (15; Winkelhake and Buckmire, Cancer Res., in press). Gamma radiation was monitored by a Beckman GM100 gamma counter with a data computer. The specific activity of 125I_ labeled MCR was 4 ± 0.5 x 106 cpm/mg. lodination did not cause loss of bactericidal activity. Other antibiotics. Actinomycin D, bacitracin, chloramphenicol, penicillin G, polymyxin B, and tetracycline were obtained from Calbiochem, La Jolla, Calif.; streptomycin sulfate came from E. R. Squibb & Sons, New York, N.Y.; gentamicin reagent was obtained from Schering Laboratories, Bloomfield, N.J.; and bleomycin sulfate (Blenoxane) was from Bristol Laboratories, Syracuse, N.Y. Stock solutions (1 to 2.5 mg/ml) of all antibiotics were prepared in sterile glass with distilled water (weight corrected for potency) and then sterilized by membrane filtration (pore size, 0.2 ,um; Gelman Instrument Co., Ann Arbor, Mich.). Appropriate dilutions were made in sterile TSB to yield 100 or 250 ,ug of each drug per ml. Broth dilution susceptibility tests. Serial twofold broth dilution tests were performed in the growth medium in a volume of 1 ml per tube. Cultures for these experiments were grown on agar plates, harvested in the growth medium without agar, adjusted to an optical density of between 0.4 and 0.55 at 600 nm, and diluted 100- to 1,000-fold to produce a final bacterial population in each tube of 106 to 107 colonyforming units (CFU) per ml. The minimal inhibitory concentration (MIC) was defined as the lowest concentration of drug that prevented macroscopic growth after incubation of 18 to 24 h at 37°C. Minimal bactericidal concentration (MBC) was determined by subculturing 1 ,ul (calibrated loop) from each clear tube onto a TSA plate. The MBC was defined as the lowest drug concentration preventing all growth on subculture. Kill curves. Side arm flasks containing MCR in
1011
TSB were inoculated with an appropriate dilution of B. pumilus suspension to produce a final bacterial population of approximately 107 CFU/ml and incubated at 30°C while shaking. Samples were removed at various time intervals, diluted in TSB, and plated in triplicate on TSA plates. Antibiotic inactivation test. HEPES buffer (0.01 M, pH 7.55) containing MCR was inoculated with approximately 108 CFU/ml of B. pumilus and incubated for 30 min at 37°C. After centrifugation at 7,500 x g for 10 min, the pelleted cells were washed twice and plated. The supernatant obtained after initial centrifugation was filter sterilized (pore size, 0.2 ,um; Gelman Instrument Co., Ann Arbor, Mich.), reinoculated with 108 CFU/ml, and, after incubation at 37°C for 30 min, plated on TSA plates. As a control, the above procedure was simultaneously performed on cells suspended in buffer without MCR. Synergy. To test for synergy, serial twofold tube dilutions of MCR were made as described above, and, to each dilution, the same concentration of a second antibiotic was added. Three different concentrations of the second antibiotic were used. Synergy was defined as a reduction of the MIC values of both antibiotics by fourfold or more. Partial synergy was defined as a fourfold or greater reduction in the MIC value of one compound but less than fourfold reduction in the MIC value of the second antibiotic. Indifference to combined antibiotics was defined as a twofold or less reduction in MIC values. Antagonism was defined as a fourfold increase in the MIC value of either antibiotic. 125I-labeled MCR binding experiments. Bacterial cell suspensions in HEPES buffer with and without 125I-labeled MCR were incubated either at 4 or 37°C for various time intervals. After incubation, the suspensions were centrifuged at 7,500 x g for 10 min, and the pelleted cells, washed twice with cold HEPES buffer to remove unbound MCR, were resuspended to the original. volume in HEPES buffer. These suspensions were again incubated at either 4 or 37°C in the presence and absence of 1% trypsin for up to 30 min and centrifuged, and the pellet was resuspended to the original volume in HEPES buffer. Both the supernatants and cell suspensions were quantitatively analyzed for radioactivity. In cases where CFU determinations were to be made, sterile materials were used throughout. Serial dilutions were made in TSB, and the CFU values were determined on TSA plates. Chemicals and buffers. All chemicals were of reagent grade. The HEPES buffer concentration throughout this study was 10 mM, pH 7.55. PBS contained 10 mM Na2HPO4, 150 mM NaCl, 1 mM CaCl2, and 0.5 mM MgCl2. 6H20, pH 7.2. Lac-PBS contained 1% lactose in PBS.
RESULTS Antibacterial effects of MCR. MCR displayed strong antibacterial effects against some gram-positive and -negative bacteria (Table 1). It was, however, much more effective against the gram-positive bacteria. The MIC was below
1012
TABLE
BUCKMIRE AND WINKELHAKE 1.
ANTIMICROB. AGZNTS CHICMOTHER.
Comparative activity of MCR and various other antibiotics against selected species of bacteriaa Bacteria
Antibiotics E. coli
P. aeruginosa
S. serpensb
B. pumilus
S. aureus
H. influenzae type bc 2.5/2.5 _d
1.6/12.5 MCR >100/>100 1.6/1.6 >100/>100 1.3/1.3 50/100 25/50 Streptomycin 12.5/12.5 100/100 1.6/1.6 100/>100 1.6/1.6 >100/>100 >100/>100 Bacitracin >100/>100 >100/>100 >100/>100 0.8/>100 0.2/>100 Tetracycline 0.8/>100 25/>100 0.8/25 6.3/>100 12.5/>100 Chloramphenicol 6.3/>100 3.1/3.1 >100/>100 1.6/>100 0.4/12.5 Actinomycin D 50/50 >100/>100 >100/>100 50/100 >100/>100 Polymxyin B 25/25 1.6/3.1 0.4/25 Bleomycin 637 a The numbers refer to MIC/MBC in micrograms per milliliter. b The medium used for S. serpens was Casamino Acids + Ca2+ + CB (see Materials and Methods). c Nicotinamide adenine dinucleotide, hemin, and glucose were added to the medium. d _, Not tested.
2 ,ug/ml for S. aureus and B. pumilus, the two gram-positive organisms tested by the tube dilution method, as well as for two clinical strains of S. aureus and two B. subtilis strains tested by a plate dilution method. With the gramnegative organisms, on the other hand, MIC values ranged from 2.5 Ag/ml, for organisms such as H. influenzae type b (three strains) and S. serpens (four strains), to greater than 100 pug/ml, for E. coli (two strains) and P. aeruginosa (six strains). When compared with some other commonly used antibiotics, the MIC of MCR was lower
than the majority against the gram-positive organisms tested (Table 1). Although its activity compared favorably with the other antibiotics tested against the gram-negative bacterium S. serpens, it was not as effective as gentamicin, streptomycin, tetracycline, chloramphenicol, polymyxin B, and bleomycin againstE. coli or gentamicin, tetracycline, and polymyxin B against P. aeruginosa. Additionally, the MBC of MCR was the same as the MIC for most organisms tested. S. aureus was the notable exception. MBC and MIC values were generally the same for all the bactericidal antibiotics.
Upon prolonged incubation at 370C, macroscopic culture growth occasionally occurred at concentrations higher than the MIC of MCR. Such growth may be explained (a) by the spontaneous mutation to resistant cells, (b) by inactivation of the antibiotic, or (c) by a prozonelike effect similar to that seen during antibodycomplement-mediated bactericidal activity (14). Cells, isolated from growth at these higher MCR concentrations (50 to 100 ,ug/ml) and subcultured once, were as susceptible as the origi-
nal culture, thus indicating they were not resistant mutants. To determine whether the presence of other antibiotics could enhance the antibacterial activity of MCR, the MIC of MCR for two gramnegative and two gram-positive organisms was determined in the presence of various concentrations of other antibiotics. Table 2 shows that, for P. aeruginosa with the chloramphenicol concentration at 100 gg/ml, the MIC of MCR was reduced from >100 to 12.5 jug/ml. No similar synergistic or antagonistic effect was observed with the other antibiotics against either P. aeruginosa or E. coli. With gram-positive organisms, partial synergy was observed between polymyxin B and MCR. With B. pumilus, the MIC of MCR was reduced to 0.1 ,Ag/ml in the presence of 20 ,ug of polymyxin B per ml, and with S. aureus, the MIC was reduced to 0.4 ,ug/ml in the presence of 100 ,ug of polymyxin B per ml. Bactericidal activity of MCR. Table 3 shows the effect of exposure time on the ability of MCR to kill bacteria suspended in a nongrowth medium (HEPES buffer). At effective concentrations (1 ,ug/ml and above), the bactericidal effect of MCR was immediate on B. pumilus, whereas on S. aureus and S. serpens the effect was time dependent. This time-dependent effect varied with the concentration of the antibiotic used. For example with S. serpens, at a concentration of 1 ug/ml (about one-half of the MIC), there was no immediate killing-, however, after 20-min of incubation in the presence of the antibiotic, approximately 98% of the cells were killed. At 10 ,ug/ml (four times the MIC), immediate killing of 80% ofthe cells, increasing to about 99% within 20 min, occurred.
VOL. 11, 1977
TABLE
BACTERICIDAL ACTIVITY OF MACROMOMYCIN
1013
2. MIC values of MCR for various bacteria in the presence of other antibioticsa MIC of MCR (gtg/ml) for:
there was a rapid recovery of cells, such that by 2 h the CFU was approximately 5% of the initial inoculum. Between 2 and 24 h, the growth rate was similar to that of the control culture. B. S. Antibioticeb (lsg/m1) p. aeruTable 4 shows that the action of MCR was E. col pumi- ausince cells plated directly from a bactericidal, a lus reus suspension containing MCR (5 ,ug/ml) gave the Gentamicin same number of CFU as cells removed from the * _d _ * (6.5) suspension and washed prior to plating. To de>100 >100 (1.3) termine the fate of MCR in the reaction mix>100 >100 (0.3) ture, the cell suspension containing MCR was Chloramphenicol centrifuged, and the sterile supernatant was 12.5 * (100.0) inoculated with fresh, untreated cells, incu* >100 (20.0) bated at 37°C, and plated. No loss in viability of 1.6 >100 (2.0) these cells occurred, thus indicating that the Penicillin G >100 * (100.0) MCR was either adsorbed or inactivated by the >100 >100 (20.0) original cells. >100 >100 (2.0) Binding of 125I-labeled MCR to bacterial Streptomycin To test the possibility that MCR was cells. * * (100.0) cells suspended in HEPES buffer adsorbed, * >100 (20.0) were exposed to 125I-labeled MCR and washed >100 >100 (2.0) free of excess antibiotic, and the amount of Polymxyin B bound MCR was determined. When B. pumilus 0.4 * (100.0) 1.6 was exposed to 125I-labeled MCR, the antibiotic 0.1 (20.0) 1.6 1.6 (2.0) was adsorbed to the cells (Table 5). The amount bound was time independent, since similar a MIC values of the various antibiotics for the amounts of 125I-labeled MCR were absorbed at cultures are given in Table 1. b Parenthetical numbers are antibiotic concentra- all time intervals tested (from 0 to 15 min). Similar results were obtained with the gramtions added to MCR-treated bacterial cultures. c *, This concentration of the second antibiotic negative organism, S. serpens. The table also was either greater than or equal to its MIC for the shows that significant amounts of bound MCR organism tested, and in those cases no growth oc- (over 20%) were released after incubation at curred. 37°C. Incubation at 4°C reduced the amount d _, Not tested. released to about 9%. Neither the release of
Tested organisms always left a residual per3. Effect of drug concentration and time of centage of surviving cells, even after extensive TABLE exposure on antimicrobial action of MCR exposure to the antibiotic. These surviving orMCR Length ganisms did not form colonies when plated at of exSurvival CFU/ml the higher antibiotic concentrations (50 ,ug/ml Organism concn posure (% (,ug/ and above), but as the MCR concentration deml) (min) creased during serial dilutions to below the B. pumilus 0 6.5 x 108 100 0 MIC for the culture growth occurred. Suscepti0.01 6.7 x 108 103.1 15 bility tests on some of the surviving cells indi0.1 6.0 x 108 92.3 15 cated that their resistance to the antibiotic had 6.7 x 107 10.3 1.0 0 not changed. Microscopic observations also in1.0 6.3 x 107 9.7 15 dicated no obvious qualitative or quantitative 5.0 15 9.0 x 106 1.4 10.0 0 1.1 x 10v 0.2 morphological differences in the treated and 10.0 1.4 x 106 0.2 30 untreated cells. 10.0 60 8.6 x 105 0.1 To determine the influence of the suspending 20.0 15 1.2 x 106 0.2 medium on bacteriostasis, growth media con- S. aureus x 108 0 0 4.6 100 taining MCR were inoculated with B. pumilus, 10 0 4.6 x 108 100 and the CFU was determined at various time 10 30 1.0 X 106 0.2 intervals. After 5 min of exposure to MCR, the 10 60 8.6 x 105 0.2 CFU had decreased to approximately 0.2% at S. serpens 0 0 100 1.6 x 108 1 0 1.7 x 108 106.3 the two concentrations (1 and 10 ug/ml) tested 1 20 2.8 x 108 1.8 (Fig. 1). At 10 jug/ml, the CFU continued to 10 0 2.9 x 107 18.1 decrease, as a function of time, to approxi10 20 1.2 x 106 0.8 mately 0.02% in 24 h. At 1 ,ug/ml, however,
1014
+n U)
ANTimicROB. AGENTS CI-INMOTHER.
BUCKMIRE AND WINKELHAKE
stability of MCR in solution. At 40C, no activity loss of MCR stabilized in Lac-PBS occurred for up to 45 days, the longest interval tested. At 370C, however, there was a 16-fold loss in activity by 96 h. The addition of urine to solubilized MCR had no effect on its activity. When serum (heated or unheated) was added, however, the inactivation rate greatly increased (Table 7). The logarithm plot of MCR residual activity yielded straight lines as a function of time for each of the suspending fluids, indicating a firstorder inactivation.
74
10 S
106
1 10,
C
0
0u
4
12 16 8 Time ( Hours)
20
24
FIG. 1. Growth of B. pumilus in the presence of MCR. Side arm flasks with TSB containing 0,1, and 10 pg/ml (A, B, and C, respectively) of MCR were inoculated with B. pumilus and incubated while shaking at 37C. At various time intervals, samples were removed and plated for viability. The numbers in parentheses are the percentages of the original inoculum.
cell-bound MCR at 370C nor its antibacterial action was affected by treating the MCR-adsorbed cells with trypsin. Influence of cell concentration on the MIC. As other studies (13) on the influence of bacterial cell concentration on MIC show, inoculum size profoundly affected the MIC of MCR (Table 6). Whereas an inoculum of 107 CFU/ml gave an MIC of 1.6 ,ug/ml, 10- and 100-fold decreases in cell concentration gave MIC values of 0.39 and 0.1 j,g/ml, respectively, and 10- and 100-fold increases gave 12.5 and 50 jg/ml, respectively. Stability of MCR in body fluids. Other workers (16) have shown that when MCR suspended in PBS was lyophilized it lost activity. This activity loss was enhanced by increasing the temperature. However, when certain divalent cations and disaccharides were added prior to lyophilization, the rate of inactivation was retarded. At 350C, for example, there was a 10% activity loss in 31 days when maltose was added, whereas almost all activity was lost in the absence of the protective agent (16). Table 7 presents the effect of certain body fluids on the
DISCUSSION MCR used in this study was purified to homogeneity, which allowed us to unequivocally determine the antibacterial action of the drug. Homogeneity was facilitated with lactose as a stabilizing agent (Winkelhake and Buckmire, Cancer Res., in. press) during purification. Stability studies on purified MCR dissolved in Lac-PBS buffer clearly indicate that the drug was unstable in solution at 37 but not at 4°C (Table 7). However, since there was no significant reduction in MCR activity within 24 h at 370C, the MIC values we obtained are considered valid. Inactivation of MCR at 370C was also enhanced, and to the satme degree (at least 16-fold in 72 h), by both heated and unheated pooled human sera, thus indicating that this inactivation was probably not due to serumrelated enzymatic cleavage of the polypeptide. In their thermal inactivation studies on purified MCR, Yamashita and co-workers (16) calculated that the lyophilized powder in maltose would lose 10% of its activity in 31 days and 17 months at 35 and 40C, respectively. Although the spectrum of the bacterial genTABLz 4. Irreversible killing of B. pumilus by MCR Samplea
MCR concn
CFU/ml
Survival
A
0
9.8 x 108
100
B
5
5.0 x 106 4.8 x 10r 2.8 x 108 2.7 x 108 3.0 x 108
C
0.5
0.5 5 S from sample A 100 D 96.4 S from sample C E 107.1 F W from sample C a Samples A and B were plated by serial dilutions directly from the reaction mixtures. For sample C, cells in the reaction mixture were pelleted, washed, suspended in fresh buffer, and then plated by serial dilutions. For samples D, E, and F, untreated cells were suspended in supernatant (5) from sample A and supernatant (S) and washings (W) from sample C, respectively, and then plated by serial dilutions. All samples were incubated for 10 min at 25°C prior to plating for viability.
BACTERICIDAL ACTIVITY OF MACROMOMYCIN
VOL. 11, 1977
era used was limited, our data confirm the preliminary studies of Chimura et al. (3) that gram-positive bacteria are susceptible to MCR. In fact, MCR was as effective on gram-positive organisms as most of the other commonly used antibiotics with which it was compared (Table 1). In addition, our data show that certain gram-negative bacteria were also susceptible. In all cases, MCR was bactericidal. Despite the lethal action of MCR, a population ofresidual cells that were not killed always remained. For these, high MCR concentrations (more than 30 times the MIC for. the culture) were bacteriostatic, since these cells were able to divide (form colonies) only when the concentrations of MCR were reduced below the MIC for the culture. Repeated exposure of these surviving cells to high concentrations of MCR, however, did not lead to resistant cells. Although the commonly encountered gramnegative bacteria such as E. coli and P. aeruginosa were resistant to MCR, partial synergy was obtained against P. aeruginosa with chloramphenicol, and this lower concentration of MCR (MIC of 12.5 ,ug/ml) was lethal. A similar additive bactericidal effect was also observed
TABLE 5. Binding of 1251-labeled MCR to B. pumilus cellsa Amount of2of 21-s la-
Time of expo-
sure to
MCR
(min)
beled
Trypsin
MCR ad-
treated
sorbed
beled
MCR
CFU (
survival)
re-
leased
(cpm)
0Ob 7.4 x 108 (100) 0 5,557 Yes 1,520 1.2 x 106 (0.2) 15 6,328 Yes 1,747 0.9 x 106 (0.1) 15 6,154 No 2,080 1.0 x 106 (0.1) a Cell suspensions in 0.01 M HEPES buffer were incubated at 37°C with MCR for the indicated time periods and washed three times with cold buffer by centrifugation at 7,500 x g for 10 min. After washing, the suspensions were incubated at 37°C for 20 min with and without 1% trypsin, centrifuged as before, and the amount of MCR released as well as viability CFU were determined. b Control culture. TABLE 6. Effect of cell concentration on the MIC of MCR for B. pumilus Cell 1.3 4.2
1.3 4.2
1.3 4.2 4.2
concn
109 x 108 x 108 x 107 x 107 x 106 105
x
MIC of MCR (yg/ml) ................
50
................
12.5
................
3.1
................
1.6
................
0.8
................
0.4
................
0.1
1015
TABLE 7. Time- and temperature-dependent inactivation of MCR in solutiona MIC (Qg/ml)
Time (h)
Temp °C
Lac-
PBS 1.6 3.1 1.6 3.1 3.1 6.3 25
Urine
Serum
Heated serum
4 0 _b 4 1,080 1 37 1.6 1.6 1.6 24 37 3.1 6.3 6.3 48 37 6.3 25 25 72 37 12.5 >200 >200 37 96 25 >200 >200 a Portions of MCR were preincubated with equal volumes of the respective reagents for the times and temperatures indicated. After incubation, the MIC of MCR was determined as described in Materials and Methods. B. pumilus was the indicator organism. b _, Not tested.
between polymyxin B and MCR against the gram-positive bacteria. No antagonism was encountered between MCR and any of the other antibiotics used. Indifference was generally the rule. As with mammalian cells (6; Winkelhake and Buckmire, Cancer Res., in press), small quantities of MCR bound irreversibly to bacterial cells within 10 to 20 min, and the adsorption was temperature independent. However, in contrast to mammalian cells, treatment of bacteria with trypsin either before or after exposure to MCR did not release the bound MCR or alter the lethal effect of the drug. Thus it seems that the mechanism governing the association between MCR and bacterial cells may be different from the surface receptor-specific association of mammalian cells postulated by Kunimoto et al. (7) and demonstrated by Nussey and Winkelhake (personal communication). The fundamental mode of action of MCR on mammalian cells (inhibition of DNA synthesis, [6; Winkelhake and Buckmire, Cancer Res., in press]) and bacterial cells (not as yet determined) may well be the same, however. The time taken for the bactericidal effect of MCR to manifest varied with the organism used. With B. pumilus the effect was immediate, whereas with S. aureus and S. serpens at least a 10-min lag took place. This time difference in responses may represent a difference in the degree of surface control of DNA synthesis or in the cell wall porosity of the organisms, rather than a difference in the mode of antibiotic action. There have been only a few studies on MCR antitumor effects (7, 9-12) and possible modes of action (6, 7). These studies suggested to Ku-
1016
BUCKMIRE AND WINKELHAKE
nimoto et al. (7) that MCR may inhibit DNA s'nthesis while attached to eucaryotic cell surfaces by a mechanism analogous to that postulated for the interactions between certain colicins and E. coli (4). The data presented so far, however, have not been unequivocal. The recent developments of a sensitive-specific assay for detection of minute quantities of the drug (Winkelhake and Buckmire, Cancer Res., in press) and the methodology to stabilize it (16) will undoubtedly stimulate more definite studies in these areas.
ANTiMICROB. AGZNTS CHZMOTHZR. 6.
7.
8.
9. 10.
ACKNOWLEDGMENTS We wish to thank B. Houghton for technical assistance and W. T. Bradner, Bristol Laboratories, for the generous gifts of MCR.
11.
LITERATURE CITED Buckmire, F. L. A. 1976. Identification and quantitation of capsular antigen in capsulated and noncapsulated strains of Haemophilus influenzae type b by crossed-immunoelectrophoresis. Infect. Immun. 13:1733-1742. Buckmire, F. L. A., and R. G. E. Murray. 1970. Studies on the cell wall of SpiriUum serpens. I. Isolation and partial purification of the outermost cell wall layer. Can. J. Microbiol. 16:1011-1022. Chimura, H., M. Ishizaka, M. Hamada, S. Hori, K. Kimura, J. Iwanga, T. Takeuchi, and H. Umezawa. 1968. A new antibiotic, macromomycin, exhibiting antituimor activity and antimicrobial activity. J. Antibiot. 21:44-49. Hardy, K. G. 1975. Colicinogeny and related phenomena. Bacteriol. Rev. 39:464-515. Herriott, R. M., E. Y. Meyer, M. Vogt, and M. Modan.
12.
1.
2.
3.
4.
5.
13. 14. 15.
16.
1970. Defined medium for growth of Haemophilus influenzae. J. Bacteriol. 101:513-516. Kunimoto, T., M. Hori, and H. Umezawa. 1971. Reversible binding of macromomycin, a macromolecular peptide antibiotic, to cell membranes. J. Antibiot. 24:203-205. Kunimoto, T., M. Hori, and H. Umezawa. 1972. An inhibitor of membrane function of tumor cells. Cancer Res. 32:1251-1256. Laemli, V. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227:680-685. Lippmsan, M. M. 1974. Mechanism of action of macromomycin (NSC 170105), an antitumor polypeptide. In Vitro 9:370. Lippman, M. M. 1974. In vitro studies of macromomycin (NSC 170105)L effects on cultured TA3Ha and L1210 cells. Cancer Chemother. Rep. 58:181-187. Lippman, M. M., and B. J. Abbot. 1973. Modification of transplantability and tumor growth following treatment of L1210 and TA3Ha cells with macromomycin (NSC 170105). Cancer Chemother. Rep. 57:501-503. Lippman, M. M., W. R. Laster, B. J. Abbot, J. Venditti, and M. Baratta. 1975. Antitumor activity of macromomycin B (NSC 170105) against murine leukemias, melanoma, and lung carcinoma. Cancer Res. 35:939-945. Sanders, C. C., and W. E. Sanders, Jr. 1975. Effects of procedural variations on the activity of aminoglycosides in vitro. Am. J. Clin. Pathol. 63:438-445. Winkelhake, J. L., and D. L. Kasper. 1972. Affinity chromatography of anti-meningococcal antiserum. J. Immunol. 109:824-833. Winkelhake, J. L., and G. L. Nicolson. 1976. Aglycosylantibody effects of exoglycosidase treatments on autochthonous antibody survival time in the circulation. J. Biol. Chem. 251:1074-1080. Yamashita, J., N. Naoi, K. Watanabe, T. Takeuchi, and H. Umezama. 1976. Further purification and characterization of macromomycin. J. Antibiot. 29:415-423.
Vol. 11, No.6' Printed in U.S.A .
Bactericidal Activity of Macromomycin, an Antitumor Antibiotic FRANCIS L. A. BUCKMIRE* AND JEFFREY L. WINKELHAKE Department of Microbiology, The Medical College of Wisconsin, Milwaukee, Wisconsin 53233 Received for publication 1 February 1977
Susceptibility testing by the broth dilution method showed that all the grampositive but only some of the gram-negative bacteria tested were susceptible to the antitumor antibiotic, macromomycin (MCR; NSC 170105). The minimal inhibitory concentration for the susceptible organisms was less than 3 ,g/ml. The action of MCR was bactericidal; however, at very high concentrations (50 ,ug/ml and above) some organisms occasionally escaped death. None of the escaped organisms was resistant to MCR. In combination with other commonly used antibiotics, MCR displayed partial synergy for Pseudomonas aeruginosa (from a minimal inhibitory concentration of >100 to 12.5 ug/ml with 100 ,ug of chloraimphenicol per ml) and for Bacillus pumilus and Staphylococcus aureus (from 1.6 to 0.4 ,ug/ml and below) with polymyxin B. As with mammalian cells, 125I-labeled MCR was irreversibly bound to both gram-positive and -negative bacteria. Treatment with trypsin of the mI-labeled MCR-exposed cells did not release the bound MCR or reverse its lethal effect. When in solution in a protective buffer at 4°C, MCR was stable for up to 45 days; at 370C, however, 25% of its bactericidal activity was lost in 72 h. Loss of activity was enhanced 16-fold in the presence of both heated and unheated pooled human sera. Urine had no effect on the activity of MCR.
Macromomycin (MCR; NSC 170105) is an unstable, antitumor (7, 9-12) polypeptide antibiotic discovered in 1968 by Chimura et al. (3). Recently, Yamashita et al. (16) have shown that its activity can be stabilized by certain substances, in particular the disaccharides, lactose and maltose. This has facilitated the purification of MCR to homogeneity (J.L. Winkelhake and F.L.A. Buckmire, Cancer Res., in press). As with other anititumor antibiotics recently isolated, the mode of action of MCR has not been studied in detail. Kunimoto et al. (6) and Winkelhake and Buckmire (Cancer Res., in press) have shown that radioiodine ('25I)-labeled MCR readily binds to mammalian cell surfaces and, presumably from that location, causes cessation of deoxyribonucleic acid (DNA), but not ribonucleic acid or protein, synthesis. Kunimoto et al. (7) have hypothesized that the interaction between MCR and the mammalian cells is analogous to the interaction between certain colicins and bacterial cells (4). Since the initial published data (3), there have been no other studies on the effect of MCR on bacteria. This communication compares the antimicrobial action of purified MCR on some
gram-positive and -negative bacteria with several commonly used antibiotics. MATERIALS AND METHODS Organisms. The bacteria used in this study were: Escherichia coli ATCC 25922 and another clinical isolate, Pseudomonas aeruginosa ATCC 27853 and five other clinical isolates, Staphylococcus aureus ATCC 25923 and two. clinical isolates, Bacillus pumilus ATCC 172, two laboratory strains of Bacillus subtilis, three strains of Haemophilus influ-
enzae type b, and four strains of Spirillum serpens. With the exception of H. influenzae and S. serpens, the cultures were cultivated on Trypticase soy agar (TSA) (Baltimore Biological Laboratories, Inc.)
plates. H. influenzae was grown on TSA supple-
mented with oxidized nicotinamide adenine dinucleotide (P-S Biochemicals, Milwaukee, Wis.), recrystallized hemin (Nutritional Biochemicals Corp., Cleveland, Ohio), and glucose (Calbiochem, La Jolla, Calif.) (1, 5), and S. serpens was cultivated on a Casamino Acids-salt medium containing agar (Difco Laboratories, Detroit, Mich.) (2). For all experiments, cultures were grown at 37°C for 8 to 14 h, harvested, and suspended in either the growth medium or HEPES (N-2-hydroxyethyl piperazine-N'-2ethanesulfonic acid) buffer. Cultures (excluding H.
influenzae) were maintained by bimonthly transfers and stored at 4°C. H. influenzae was stored at -70°C
on solid agar media, incubated overnight at 37°C,
1010
VOL. 11, 1977
BACTERICIDAL ACTIVITY OF MACROMOMYCIN
in Trypticase soy broth (TSB) containing 20% glycerol after 9 h of growth at 37°C. MCR. MCR was purified by sequential salt precipitation, ultrafiltration, ion exchange and molecular sieve chromatographies, and stabilized during lyophilization in 1% lactose-phosphate-buffered saline (Lac-PBS), pH 7.2, by a modification (Winkelhake and Buckmire, Cancer Res., in press) of the procedure of Yamashita et al. (16). During purification, fractions were monitored for activity by an agar-plate bacteriocidal technique, with B. pumilus as the test organism. Preparation homogeneity was determined by sodium dodecyl sulfate-polyacrylamide slab-gel electrophoresis, using 12.5% gel by the method of Laemli (8). A single protein band, which migrated between ribonuclease (13,700 daltons) and lysozyme (11,400 daltons) standards, was obtained. Since the electrophoresis technique employed is sensitive to about 10 ,ug of protein, our preparation was estimated to be >97% pure. Drug concentrations were determined spectrophotometrically with an 11.6 absorbancy value at 280 nm equivalent to 1%. Purified MCR was stored lyophilized in Lac-PBS at -70°C with no apparent loss of activity. MCR was radioiodinated with cellulose-immobilized lactoperoxidase (15; Winkelhake and Buckmire, Cancer Res., in press). Gamma radiation was monitored by a Beckman GM100 gamma counter with a data computer. The specific activity of 125I_ labeled MCR was 4 ± 0.5 x 106 cpm/mg. lodination did not cause loss of bactericidal activity. Other antibiotics. Actinomycin D, bacitracin, chloramphenicol, penicillin G, polymyxin B, and tetracycline were obtained from Calbiochem, La Jolla, Calif.; streptomycin sulfate came from E. R. Squibb & Sons, New York, N.Y.; gentamicin reagent was obtained from Schering Laboratories, Bloomfield, N.J.; and bleomycin sulfate (Blenoxane) was from Bristol Laboratories, Syracuse, N.Y. Stock solutions (1 to 2.5 mg/ml) of all antibiotics were prepared in sterile glass with distilled water (weight corrected for potency) and then sterilized by membrane filtration (pore size, 0.2 ,um; Gelman Instrument Co., Ann Arbor, Mich.). Appropriate dilutions were made in sterile TSB to yield 100 or 250 ,ug of each drug per ml. Broth dilution susceptibility tests. Serial twofold broth dilution tests were performed in the growth medium in a volume of 1 ml per tube. Cultures for these experiments were grown on agar plates, harvested in the growth medium without agar, adjusted to an optical density of between 0.4 and 0.55 at 600 nm, and diluted 100- to 1,000-fold to produce a final bacterial population in each tube of 106 to 107 colonyforming units (CFU) per ml. The minimal inhibitory concentration (MIC) was defined as the lowest concentration of drug that prevented macroscopic growth after incubation of 18 to 24 h at 37°C. Minimal bactericidal concentration (MBC) was determined by subculturing 1 ,ul (calibrated loop) from each clear tube onto a TSA plate. The MBC was defined as the lowest drug concentration preventing all growth on subculture. Kill curves. Side arm flasks containing MCR in
1011
TSB were inoculated with an appropriate dilution of B. pumilus suspension to produce a final bacterial population of approximately 107 CFU/ml and incubated at 30°C while shaking. Samples were removed at various time intervals, diluted in TSB, and plated in triplicate on TSA plates. Antibiotic inactivation test. HEPES buffer (0.01 M, pH 7.55) containing MCR was inoculated with approximately 108 CFU/ml of B. pumilus and incubated for 30 min at 37°C. After centrifugation at 7,500 x g for 10 min, the pelleted cells were washed twice and plated. The supernatant obtained after initial centrifugation was filter sterilized (pore size, 0.2 ,um; Gelman Instrument Co., Ann Arbor, Mich.), reinoculated with 108 CFU/ml, and, after incubation at 37°C for 30 min, plated on TSA plates. As a control, the above procedure was simultaneously performed on cells suspended in buffer without MCR. Synergy. To test for synergy, serial twofold tube dilutions of MCR were made as described above, and, to each dilution, the same concentration of a second antibiotic was added. Three different concentrations of the second antibiotic were used. Synergy was defined as a reduction of the MIC values of both antibiotics by fourfold or more. Partial synergy was defined as a fourfold or greater reduction in the MIC value of one compound but less than fourfold reduction in the MIC value of the second antibiotic. Indifference to combined antibiotics was defined as a twofold or less reduction in MIC values. Antagonism was defined as a fourfold increase in the MIC value of either antibiotic. 125I-labeled MCR binding experiments. Bacterial cell suspensions in HEPES buffer with and without 125I-labeled MCR were incubated either at 4 or 37°C for various time intervals. After incubation, the suspensions were centrifuged at 7,500 x g for 10 min, and the pelleted cells, washed twice with cold HEPES buffer to remove unbound MCR, were resuspended to the original. volume in HEPES buffer. These suspensions were again incubated at either 4 or 37°C in the presence and absence of 1% trypsin for up to 30 min and centrifuged, and the pellet was resuspended to the original volume in HEPES buffer. Both the supernatants and cell suspensions were quantitatively analyzed for radioactivity. In cases where CFU determinations were to be made, sterile materials were used throughout. Serial dilutions were made in TSB, and the CFU values were determined on TSA plates. Chemicals and buffers. All chemicals were of reagent grade. The HEPES buffer concentration throughout this study was 10 mM, pH 7.55. PBS contained 10 mM Na2HPO4, 150 mM NaCl, 1 mM CaCl2, and 0.5 mM MgCl2. 6H20, pH 7.2. Lac-PBS contained 1% lactose in PBS.
RESULTS Antibacterial effects of MCR. MCR displayed strong antibacterial effects against some gram-positive and -negative bacteria (Table 1). It was, however, much more effective against the gram-positive bacteria. The MIC was below
1012
TABLE
BUCKMIRE AND WINKELHAKE 1.
ANTIMICROB. AGZNTS CHICMOTHER.
Comparative activity of MCR and various other antibiotics against selected species of bacteriaa Bacteria
Antibiotics E. coli
P. aeruginosa
S. serpensb
B. pumilus
S. aureus
H. influenzae type bc 2.5/2.5 _d
1.6/12.5 MCR >100/>100 1.6/1.6 >100/>100 1.3/1.3 50/100 25/50 Streptomycin 12.5/12.5 100/100 1.6/1.6 100/>100 1.6/1.6 >100/>100 >100/>100 Bacitracin >100/>100 >100/>100 >100/>100 0.8/>100 0.2/>100 Tetracycline 0.8/>100 25/>100 0.8/25 6.3/>100 12.5/>100 Chloramphenicol 6.3/>100 3.1/3.1 >100/>100 1.6/>100 0.4/12.5 Actinomycin D 50/50 >100/>100 >100/>100 50/100 >100/>100 Polymxyin B 25/25 1.6/3.1 0.4/25 Bleomycin 637 a The numbers refer to MIC/MBC in micrograms per milliliter. b The medium used for S. serpens was Casamino Acids + Ca2+ + CB (see Materials and Methods). c Nicotinamide adenine dinucleotide, hemin, and glucose were added to the medium. d _, Not tested.
2 ,ug/ml for S. aureus and B. pumilus, the two gram-positive organisms tested by the tube dilution method, as well as for two clinical strains of S. aureus and two B. subtilis strains tested by a plate dilution method. With the gramnegative organisms, on the other hand, MIC values ranged from 2.5 Ag/ml, for organisms such as H. influenzae type b (three strains) and S. serpens (four strains), to greater than 100 pug/ml, for E. coli (two strains) and P. aeruginosa (six strains). When compared with some other commonly used antibiotics, the MIC of MCR was lower
than the majority against the gram-positive organisms tested (Table 1). Although its activity compared favorably with the other antibiotics tested against the gram-negative bacterium S. serpens, it was not as effective as gentamicin, streptomycin, tetracycline, chloramphenicol, polymyxin B, and bleomycin againstE. coli or gentamicin, tetracycline, and polymyxin B against P. aeruginosa. Additionally, the MBC of MCR was the same as the MIC for most organisms tested. S. aureus was the notable exception. MBC and MIC values were generally the same for all the bactericidal antibiotics.
Upon prolonged incubation at 370C, macroscopic culture growth occasionally occurred at concentrations higher than the MIC of MCR. Such growth may be explained (a) by the spontaneous mutation to resistant cells, (b) by inactivation of the antibiotic, or (c) by a prozonelike effect similar to that seen during antibodycomplement-mediated bactericidal activity (14). Cells, isolated from growth at these higher MCR concentrations (50 to 100 ,ug/ml) and subcultured once, were as susceptible as the origi-
nal culture, thus indicating they were not resistant mutants. To determine whether the presence of other antibiotics could enhance the antibacterial activity of MCR, the MIC of MCR for two gramnegative and two gram-positive organisms was determined in the presence of various concentrations of other antibiotics. Table 2 shows that, for P. aeruginosa with the chloramphenicol concentration at 100 gg/ml, the MIC of MCR was reduced from >100 to 12.5 jug/ml. No similar synergistic or antagonistic effect was observed with the other antibiotics against either P. aeruginosa or E. coli. With gram-positive organisms, partial synergy was observed between polymyxin B and MCR. With B. pumilus, the MIC of MCR was reduced to 0.1 ,Ag/ml in the presence of 20 ,ug of polymyxin B per ml, and with S. aureus, the MIC was reduced to 0.4 ,ug/ml in the presence of 100 ,ug of polymyxin B per ml. Bactericidal activity of MCR. Table 3 shows the effect of exposure time on the ability of MCR to kill bacteria suspended in a nongrowth medium (HEPES buffer). At effective concentrations (1 ,ug/ml and above), the bactericidal effect of MCR was immediate on B. pumilus, whereas on S. aureus and S. serpens the effect was time dependent. This time-dependent effect varied with the concentration of the antibiotic used. For example with S. serpens, at a concentration of 1 ug/ml (about one-half of the MIC), there was no immediate killing-, however, after 20-min of incubation in the presence of the antibiotic, approximately 98% of the cells were killed. At 10 ,ug/ml (four times the MIC), immediate killing of 80% ofthe cells, increasing to about 99% within 20 min, occurred.
VOL. 11, 1977
TABLE
BACTERICIDAL ACTIVITY OF MACROMOMYCIN
1013
2. MIC values of MCR for various bacteria in the presence of other antibioticsa MIC of MCR (gtg/ml) for:
there was a rapid recovery of cells, such that by 2 h the CFU was approximately 5% of the initial inoculum. Between 2 and 24 h, the growth rate was similar to that of the control culture. B. S. Antibioticeb (lsg/m1) p. aeruTable 4 shows that the action of MCR was E. col pumi- ausince cells plated directly from a bactericidal, a lus reus suspension containing MCR (5 ,ug/ml) gave the Gentamicin same number of CFU as cells removed from the * _d _ * (6.5) suspension and washed prior to plating. To de>100 >100 (1.3) termine the fate of MCR in the reaction mix>100 >100 (0.3) ture, the cell suspension containing MCR was Chloramphenicol centrifuged, and the sterile supernatant was 12.5 * (100.0) inoculated with fresh, untreated cells, incu* >100 (20.0) bated at 37°C, and plated. No loss in viability of 1.6 >100 (2.0) these cells occurred, thus indicating that the Penicillin G >100 * (100.0) MCR was either adsorbed or inactivated by the >100 >100 (20.0) original cells. >100 >100 (2.0) Binding of 125I-labeled MCR to bacterial Streptomycin To test the possibility that MCR was cells. * * (100.0) cells suspended in HEPES buffer adsorbed, * >100 (20.0) were exposed to 125I-labeled MCR and washed >100 >100 (2.0) free of excess antibiotic, and the amount of Polymxyin B bound MCR was determined. When B. pumilus 0.4 * (100.0) 1.6 was exposed to 125I-labeled MCR, the antibiotic 0.1 (20.0) 1.6 1.6 (2.0) was adsorbed to the cells (Table 5). The amount bound was time independent, since similar a MIC values of the various antibiotics for the amounts of 125I-labeled MCR were absorbed at cultures are given in Table 1. b Parenthetical numbers are antibiotic concentra- all time intervals tested (from 0 to 15 min). Similar results were obtained with the gramtions added to MCR-treated bacterial cultures. c *, This concentration of the second antibiotic negative organism, S. serpens. The table also was either greater than or equal to its MIC for the shows that significant amounts of bound MCR organism tested, and in those cases no growth oc- (over 20%) were released after incubation at curred. 37°C. Incubation at 4°C reduced the amount d _, Not tested. released to about 9%. Neither the release of
Tested organisms always left a residual per3. Effect of drug concentration and time of centage of surviving cells, even after extensive TABLE exposure on antimicrobial action of MCR exposure to the antibiotic. These surviving orMCR Length ganisms did not form colonies when plated at of exSurvival CFU/ml the higher antibiotic concentrations (50 ,ug/ml Organism concn posure (% (,ug/ and above), but as the MCR concentration deml) (min) creased during serial dilutions to below the B. pumilus 0 6.5 x 108 100 0 MIC for the culture growth occurred. Suscepti0.01 6.7 x 108 103.1 15 bility tests on some of the surviving cells indi0.1 6.0 x 108 92.3 15 cated that their resistance to the antibiotic had 6.7 x 107 10.3 1.0 0 not changed. Microscopic observations also in1.0 6.3 x 107 9.7 15 dicated no obvious qualitative or quantitative 5.0 15 9.0 x 106 1.4 10.0 0 1.1 x 10v 0.2 morphological differences in the treated and 10.0 1.4 x 106 0.2 30 untreated cells. 10.0 60 8.6 x 105 0.1 To determine the influence of the suspending 20.0 15 1.2 x 106 0.2 medium on bacteriostasis, growth media con- S. aureus x 108 0 0 4.6 100 taining MCR were inoculated with B. pumilus, 10 0 4.6 x 108 100 and the CFU was determined at various time 10 30 1.0 X 106 0.2 intervals. After 5 min of exposure to MCR, the 10 60 8.6 x 105 0.2 CFU had decreased to approximately 0.2% at S. serpens 0 0 100 1.6 x 108 1 0 1.7 x 108 106.3 the two concentrations (1 and 10 ug/ml) tested 1 20 2.8 x 108 1.8 (Fig. 1). At 10 jug/ml, the CFU continued to 10 0 2.9 x 107 18.1 decrease, as a function of time, to approxi10 20 1.2 x 106 0.8 mately 0.02% in 24 h. At 1 ,ug/ml, however,
1014
+n U)
ANTimicROB. AGENTS CI-INMOTHER.
BUCKMIRE AND WINKELHAKE
stability of MCR in solution. At 40C, no activity loss of MCR stabilized in Lac-PBS occurred for up to 45 days, the longest interval tested. At 370C, however, there was a 16-fold loss in activity by 96 h. The addition of urine to solubilized MCR had no effect on its activity. When serum (heated or unheated) was added, however, the inactivation rate greatly increased (Table 7). The logarithm plot of MCR residual activity yielded straight lines as a function of time for each of the suspending fluids, indicating a firstorder inactivation.
74
10 S
106
1 10,
C
0
0u
4
12 16 8 Time ( Hours)
20
24
FIG. 1. Growth of B. pumilus in the presence of MCR. Side arm flasks with TSB containing 0,1, and 10 pg/ml (A, B, and C, respectively) of MCR were inoculated with B. pumilus and incubated while shaking at 37C. At various time intervals, samples were removed and plated for viability. The numbers in parentheses are the percentages of the original inoculum.
cell-bound MCR at 370C nor its antibacterial action was affected by treating the MCR-adsorbed cells with trypsin. Influence of cell concentration on the MIC. As other studies (13) on the influence of bacterial cell concentration on MIC show, inoculum size profoundly affected the MIC of MCR (Table 6). Whereas an inoculum of 107 CFU/ml gave an MIC of 1.6 ,ug/ml, 10- and 100-fold decreases in cell concentration gave MIC values of 0.39 and 0.1 j,g/ml, respectively, and 10- and 100-fold increases gave 12.5 and 50 jg/ml, respectively. Stability of MCR in body fluids. Other workers (16) have shown that when MCR suspended in PBS was lyophilized it lost activity. This activity loss was enhanced by increasing the temperature. However, when certain divalent cations and disaccharides were added prior to lyophilization, the rate of inactivation was retarded. At 350C, for example, there was a 10% activity loss in 31 days when maltose was added, whereas almost all activity was lost in the absence of the protective agent (16). Table 7 presents the effect of certain body fluids on the
DISCUSSION MCR used in this study was purified to homogeneity, which allowed us to unequivocally determine the antibacterial action of the drug. Homogeneity was facilitated with lactose as a stabilizing agent (Winkelhake and Buckmire, Cancer Res., in. press) during purification. Stability studies on purified MCR dissolved in Lac-PBS buffer clearly indicate that the drug was unstable in solution at 37 but not at 4°C (Table 7). However, since there was no significant reduction in MCR activity within 24 h at 370C, the MIC values we obtained are considered valid. Inactivation of MCR at 370C was also enhanced, and to the satme degree (at least 16-fold in 72 h), by both heated and unheated pooled human sera, thus indicating that this inactivation was probably not due to serumrelated enzymatic cleavage of the polypeptide. In their thermal inactivation studies on purified MCR, Yamashita and co-workers (16) calculated that the lyophilized powder in maltose would lose 10% of its activity in 31 days and 17 months at 35 and 40C, respectively. Although the spectrum of the bacterial genTABLz 4. Irreversible killing of B. pumilus by MCR Samplea
MCR concn
CFU/ml
Survival
A
0
9.8 x 108
100
B
5
5.0 x 106 4.8 x 10r 2.8 x 108 2.7 x 108 3.0 x 108
C
0.5
0.5 5 S from sample A 100 D 96.4 S from sample C E 107.1 F W from sample C a Samples A and B were plated by serial dilutions directly from the reaction mixtures. For sample C, cells in the reaction mixture were pelleted, washed, suspended in fresh buffer, and then plated by serial dilutions. For samples D, E, and F, untreated cells were suspended in supernatant (5) from sample A and supernatant (S) and washings (W) from sample C, respectively, and then plated by serial dilutions. All samples were incubated for 10 min at 25°C prior to plating for viability.
BACTERICIDAL ACTIVITY OF MACROMOMYCIN
VOL. 11, 1977
era used was limited, our data confirm the preliminary studies of Chimura et al. (3) that gram-positive bacteria are susceptible to MCR. In fact, MCR was as effective on gram-positive organisms as most of the other commonly used antibiotics with which it was compared (Table 1). In addition, our data show that certain gram-negative bacteria were also susceptible. In all cases, MCR was bactericidal. Despite the lethal action of MCR, a population ofresidual cells that were not killed always remained. For these, high MCR concentrations (more than 30 times the MIC for. the culture) were bacteriostatic, since these cells were able to divide (form colonies) only when the concentrations of MCR were reduced below the MIC for the culture. Repeated exposure of these surviving cells to high concentrations of MCR, however, did not lead to resistant cells. Although the commonly encountered gramnegative bacteria such as E. coli and P. aeruginosa were resistant to MCR, partial synergy was obtained against P. aeruginosa with chloramphenicol, and this lower concentration of MCR (MIC of 12.5 ,ug/ml) was lethal. A similar additive bactericidal effect was also observed
TABLE 5. Binding of 1251-labeled MCR to B. pumilus cellsa Amount of2of 21-s la-
Time of expo-
sure to
MCR
(min)
beled
Trypsin
MCR ad-
treated
sorbed
beled
MCR
CFU (
survival)
re-
leased
(cpm)
0Ob 7.4 x 108 (100) 0 5,557 Yes 1,520 1.2 x 106 (0.2) 15 6,328 Yes 1,747 0.9 x 106 (0.1) 15 6,154 No 2,080 1.0 x 106 (0.1) a Cell suspensions in 0.01 M HEPES buffer were incubated at 37°C with MCR for the indicated time periods and washed three times with cold buffer by centrifugation at 7,500 x g for 10 min. After washing, the suspensions were incubated at 37°C for 20 min with and without 1% trypsin, centrifuged as before, and the amount of MCR released as well as viability CFU were determined. b Control culture. TABLE 6. Effect of cell concentration on the MIC of MCR for B. pumilus Cell 1.3 4.2
1.3 4.2
1.3 4.2 4.2
concn
109 x 108 x 108 x 107 x 107 x 106 105
x
MIC of MCR (yg/ml) ................
50
................
12.5
................
3.1
................
1.6
................
0.8
................
0.4
................
0.1
1015
TABLE 7. Time- and temperature-dependent inactivation of MCR in solutiona MIC (Qg/ml)
Time (h)
Temp °C
Lac-
PBS 1.6 3.1 1.6 3.1 3.1 6.3 25
Urine
Serum
Heated serum
4 0 _b 4 1,080 1 37 1.6 1.6 1.6 24 37 3.1 6.3 6.3 48 37 6.3 25 25 72 37 12.5 >200 >200 37 96 25 >200 >200 a Portions of MCR were preincubated with equal volumes of the respective reagents for the times and temperatures indicated. After incubation, the MIC of MCR was determined as described in Materials and Methods. B. pumilus was the indicator organism. b _, Not tested.
between polymyxin B and MCR against the gram-positive bacteria. No antagonism was encountered between MCR and any of the other antibiotics used. Indifference was generally the rule. As with mammalian cells (6; Winkelhake and Buckmire, Cancer Res., in press), small quantities of MCR bound irreversibly to bacterial cells within 10 to 20 min, and the adsorption was temperature independent. However, in contrast to mammalian cells, treatment of bacteria with trypsin either before or after exposure to MCR did not release the bound MCR or alter the lethal effect of the drug. Thus it seems that the mechanism governing the association between MCR and bacterial cells may be different from the surface receptor-specific association of mammalian cells postulated by Kunimoto et al. (7) and demonstrated by Nussey and Winkelhake (personal communication). The fundamental mode of action of MCR on mammalian cells (inhibition of DNA synthesis, [6; Winkelhake and Buckmire, Cancer Res., in press]) and bacterial cells (not as yet determined) may well be the same, however. The time taken for the bactericidal effect of MCR to manifest varied with the organism used. With B. pumilus the effect was immediate, whereas with S. aureus and S. serpens at least a 10-min lag took place. This time difference in responses may represent a difference in the degree of surface control of DNA synthesis or in the cell wall porosity of the organisms, rather than a difference in the mode of antibiotic action. There have been only a few studies on MCR antitumor effects (7, 9-12) and possible modes of action (6, 7). These studies suggested to Ku-
1016
BUCKMIRE AND WINKELHAKE
nimoto et al. (7) that MCR may inhibit DNA s'nthesis while attached to eucaryotic cell surfaces by a mechanism analogous to that postulated for the interactions between certain colicins and E. coli (4). The data presented so far, however, have not been unequivocal. The recent developments of a sensitive-specific assay for detection of minute quantities of the drug (Winkelhake and Buckmire, Cancer Res., in press) and the methodology to stabilize it (16) will undoubtedly stimulate more definite studies in these areas.
ANTiMICROB. AGZNTS CHZMOTHZR. 6.
7.
8.
9. 10.
ACKNOWLEDGMENTS We wish to thank B. Houghton for technical assistance and W. T. Bradner, Bristol Laboratories, for the generous gifts of MCR.
11.
LITERATURE CITED Buckmire, F. L. A. 1976. Identification and quantitation of capsular antigen in capsulated and noncapsulated strains of Haemophilus influenzae type b by crossed-immunoelectrophoresis. Infect. Immun. 13:1733-1742. Buckmire, F. L. A., and R. G. E. Murray. 1970. Studies on the cell wall of SpiriUum serpens. I. Isolation and partial purification of the outermost cell wall layer. Can. J. Microbiol. 16:1011-1022. Chimura, H., M. Ishizaka, M. Hamada, S. Hori, K. Kimura, J. Iwanga, T. Takeuchi, and H. Umezawa. 1968. A new antibiotic, macromomycin, exhibiting antituimor activity and antimicrobial activity. J. Antibiot. 21:44-49. Hardy, K. G. 1975. Colicinogeny and related phenomena. Bacteriol. Rev. 39:464-515. Herriott, R. M., E. Y. Meyer, M. Vogt, and M. Modan.
12.
1.
2.
3.
4.
5.
13. 14. 15.
16.
1970. Defined medium for growth of Haemophilus influenzae. J. Bacteriol. 101:513-516. Kunimoto, T., M. Hori, and H. Umezawa. 1971. Reversible binding of macromomycin, a macromolecular peptide antibiotic, to cell membranes. J. Antibiot. 24:203-205. Kunimoto, T., M. Hori, and H. Umezawa. 1972. An inhibitor of membrane function of tumor cells. Cancer Res. 32:1251-1256. Laemli, V. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227:680-685. Lippmsan, M. M. 1974. Mechanism of action of macromomycin (NSC 170105), an antitumor polypeptide. In Vitro 9:370. Lippman, M. M. 1974. In vitro studies of macromomycin (NSC 170105)L effects on cultured TA3Ha and L1210 cells. Cancer Chemother. Rep. 58:181-187. Lippman, M. M., and B. J. Abbot. 1973. Modification of transplantability and tumor growth following treatment of L1210 and TA3Ha cells with macromomycin (NSC 170105). Cancer Chemother. Rep. 57:501-503. Lippman, M. M., W. R. Laster, B. J. Abbot, J. Venditti, and M. Baratta. 1975. Antitumor activity of macromomycin B (NSC 170105) against murine leukemias, melanoma, and lung carcinoma. Cancer Res. 35:939-945. Sanders, C. C., and W. E. Sanders, Jr. 1975. Effects of procedural variations on the activity of aminoglycosides in vitro. Am. J. Clin. Pathol. 63:438-445. Winkelhake, J. L., and D. L. Kasper. 1972. Affinity chromatography of anti-meningococcal antiserum. J. Immunol. 109:824-833. Winkelhake, J. L., and G. L. Nicolson. 1976. Aglycosylantibody effects of exoglycosidase treatments on autochthonous antibody survival time in the circulation. J. Biol. Chem. 251:1074-1080. Yamashita, J., N. Naoi, K. Watanabe, T. Takeuchi, and H. Umezama. 1976. Further purification and characterization of macromomycin. J. Antibiot. 29:415-423.
