ANTnuczoBIAL AGzNTs AND CHEMOTHERPY, JUly 1977, p. 67-72 Copyright © 1977 American Society for Microbiology
Vol. 12, No. 1 Printed in U.S.A.
Delineation of the Relative Antibacterial Activity of Cefamandole and Cefamandole Nafate J. R. TURNER,* DAVID A. PRESTON, AND J. S. WOLD Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, Indiana 46206
Received for publication 22 December 1976 By conventional laboratory evaluation procedures, the in vitro antibacterial activities of cefamandole and its O-formyl ester, cefamandole nafate, appear virtually identical. When the activities of these two compounds were exaniined for their ability to lyse log-phase cultures of susceptible bacteria, however, cefamandole was found to be about 10 times more active than cefamandole nafate. Cefamandole nafate was shown to be rapidly converted to cefamandole in bacteriological media, with a half-life of less than 1 h at a pH of 7.0 or above. At pH 6.0, in log-phase inhibition experiments, however, cefamandole nafate is more stable, allowing delineation of the activity between cefamandole and cefamandole nafate. The efficacy of cefamandole was identical to that of cefamandole nafate in treating experimental animal infections, indicating that rapid conversion of cefamandole nafate to cefamandole occurs in vivo.
The antibacterial activity of cefamandole, 7n-mandelamido- 3 f [(1- methyl-1H-tetrazol-5-yl)thio]methyl}3-cephem-4-carboxylic acid, was first described by Wick and Preston (7), who showed that this antibiotic has very good activity against gram-negative bacteria, including ,8-lactamase-producing strains of Proteus and Enterobacter. Subsequently, this antibiotic was introduced into clinical trial as the sodium salt of the O-formyl ester of cefamandole and designated as cefamandole nafate. This derivative, which is stable in the crystalline form, is rapidly converted to cefamandole by hydrolysis of the formyl ester in vitro after dissolution. Because of rapid in vivo conversion of cefamandole nafate to cefamandole (4), the latter is the predominant circulating antibiotic after administration of cefamandole nafate to laboratory animals or humans (J. S. Wold, R. R. Joost, H. R. Black, and K. E. Briscoe, Ninth International Congr. Chemother., M225, 1975). The in vitro antibiotic activity of cefamandole nafate was shown (R. B. Kammer, D. A. Preston, and J. R. Turner, Ninth International Congr. Chemother., M226, 1975) to be equivalent to cefamandole lithium by conventional agar dilution and disk plate susceptibility tests against a number of bacterial pathogens. It was also reported that a differential in activity between these two antibiotics was detectable against Bacillus subtilis based on a decreased growth rate in log-phase cultures, although only at concentrations below 0.1 ,ug/ml. We have shown that these apparently contradictory findings are the result of two basic differ-
ences in experimental conditions between the two tests: the duration of the experiments and the pH at which these experiments were performed. MATERIALS AND METHODS Antibiotics. Cefamandole lithium and cefamandole nafate (the O-formyl ester of cefamandole sodium) were obtained from the Lilly Research Laboratories, Eli Lilly and Co. Bacterial cultures. B. subtilis (ATCC 6633) was obtained from the American Type Culture Collection (ATCC). All other bacterial strains used were originally clinical isolates collected from various sources over several years and maintained in this laboratory. Disk susceptibility tests: effect of pH. Antibiotic activity at pH 6.0 or 8.0 was determined by the disk susceptibility test method using Lilly Biochem agar no. 6 (beef extract, 0.15%; yeast extract, 0.6%; peptone, 0.6%; and agar, 1.75%) or antibiotic medium 5
(Difco).
MICs. Minimal inhibitory concentrations (MICs) were determined by the agar dilution procedure described by Ericsson and Sherris (1). Inocula of approximately 104 bacteria were applied to each spot on the surface of antibiotic-containing Mueller-Hinton agar plates by use of a multipronged inoculating device. Growth medium and conditions for log-phase cultures. Minimal medium at pH 6.0 consisted of: (NH4)2S04, 0.2%; K2HPO4, 1.4%; KH2PO4, 0.6%; MgSO4-7H2O, 0.02%; sodium citrate, 0.1%; vitaminfree Casamino Acids (Difco), 0.02%; and glucose, 0.5%. Log-phase cultures were prepared by inoculating 50 ml of medium with a sample of a 16-h culture sufficient to give a reading of 5 Klett units. The cultures were incubated on a rotary shaker at 37°C 67
68
TURNER, PRESTON, AND WOLD
ANTmiCROB. AGENTs CHEMOTHER. TABLz 1. Agar dilution MIC of two forms of cefamandole for a variety of gram-positive and gramnegative bacteria
at 180 rpm. Antibiotic additions were made after log-phase growth haft been established. The antibiotic activity was quantitated in terms of the time required for lysis of the culture to occur. The measurement used for comparative purposes was termed "time to lysis" and was the point in time after addition of the antibiotic when the Klett reading decreased or remained constant. Assay of cefamandole and cefamandole nafate in culture medium. Uninoculated medium (agar or liquid culture medium) to which antibiotic had been added was incubated under the same conditions as the experimental cultures. Samples of these media were removed at appropriate time intervals, acidified with glacial acetic acid (final concentration, 25%), and frozen until assayed. The assay was accomplished by high-pressure liquid chromatogrphy as described by Wold et al. (Ninth International Congr. Chemother., M225, 1975). Experimental infections. Streptococcus pyogenes
Geometric mean
No. Bacteria Bested
Enterobacter aerogenes Enterobacter cloacae Escherichia coli Klebsiella pneumoniae Proteus mirabilis Proteus morganii Proteus rettgeri Proteus vulgaris Salmonella typhi Serratia marcescens Shigella flexneri Staphylococcus aureus
B. SUBTIUS z
ANTIBI
25-
C
B. SUBTIUS
MEDIUM
5
AT PH I
C
8 7 6 10 7 2 3 2 1 1 1 3
Nafate
1.2 2.7 1.3 1.0 1.3 0.7 1.6 2 0.25 64 0.13 0.63
1.1 2.7 1.3 0.9 1.0 0.5 1.0 1.4 0.13 64 0.25 0.51
/
ANTIBIOTIC MEDIUM 5 AT pH 8
CEFAMANOOLE NAFATE CEFAMANDOLE LITHIUM a0
MIC (Qg/ml) Lithium
10I~~~~~~~
O-)
F aKftl I.3
,2.5
..
I
I
10
I
I
I
15 20 25 30 ZONE DIAMETER {mm}
I
35
I
I
10
I
I
15 20 25 30 ZONE DIAMETER (mm}
I
35
50 U c
25
S. AUREUS
S. AUREUS
AGARNO. 6 ATPH
AAR NO. ATPH8
F
1.3 z
10
1.W
=2
2.515 20 25 30 ZONE DIAMETER (mml
10 e
35
10
15 20 25 30 ZONE DIAMETER Imm)
35
50-
-
c2
E. COU 2525
z Cl 9
II
/
E. COU
AGAR NO. 6 AT
AGAR NO. 6 AT PH 8
.
10
/ 5~~~~~~
2.5~~~~~ I 10
I I I I 20 25 30 15 ZONE DIAMETER (mm)
-
I
I
35
10
I
I
I
IA
15 20 25 30 ZONE DIAMETER (m)
35
FIG. 1. Assay of the antibiotic activity of two forms of cefamandole in agar media at pH 6.0 and 8.0.
69
CEFAMANDOLE, CEFAMANDOLE NAFATE ACTIVITY
VoL. 12, 1977
TABLE 2. Hydrolysis of cefamandole nafate to cefamandole in several bacteriological agar media Unchanged cefamandole nafate (%) remaining at. Agar medium
Mueller-Hinton agar (BBL) Antibiotic medium 5 (Difco), adjusted to low pH Antibiotic medium 5 (Difco), adjusted to high pH Lilly biochem agar no. 6, adjusted to low pH Lilly biochem agar no. 6, adjusted to high pH a Half-life of cefamandole nafate.
Initial
pH
5min
lh
2h
3h
4h
7.4 6.5
99.9 100+
48.9 91.9
11.2 73.9
3.5 57.7
0 49.4
8.1
67.0
0
0
0
0
6.0
100+
100+
91.5
85.7
78.4
8.15
72.9
0
0
0
0
and Streptococcus pneumoniae were grown overnight in brain heart infusion broth (BBL) supplemented with 5% defibrinated rabbit blood. Decimal dilutions of the test culture were made in brain heart infusion broth without blood to yield the previously determined infective concentration of organisms. McAllister strain Swiss white mice (13 to 15 g, random sex) were infected by intraperitoneal injection of 0.5-ml volumes of the desired bacterial suspension.
Staphylococcus aureus, Escherichia coli, and Klebsiella pneumoniae were grown for 18 h at 36°C on Trypticase soy agar (BBL) slants. The surface growth was suspended in saline to a density of 109 colony-forming units/ml and diluted to predetermined infective concentrations. The final dilution of these latter cultures was made in sterile 5% hog gastric mucin at pH 7.4. As with the streptococci, these bacterial suspensions were administered by intraperitoneal injection in 0.5-ml volumes. Groups of eight infected mice were treated subcutaneously with antibiotic (0.25-ml volumes) at 1 and 5 h after infection. Survivors and deaths were recorded daily for 7 days. The amount of antibiotic required to protect 50% of the infected mice (ED50) was estimated by the method of Reed and Muench (6). RESULTS
Comparison of the in vitro activity of cefamandole nafate with that of cefamandole as determined by agar dilution MIC, led to the conclusion that there was little, if any, activity difference between the two forms of cefamandole. Table 1 shows the geometric mean MICs for various genera. The MIC of both compounds was the same for 36 of the 51 bacterial isolates tested. There was one dilution difference in favor of cefamandole nafate with 13 isolates, and a one-dilution difference in favor of cefamandole lithium with 2 isolates. Similar results were obtained with the Food and Drug Administration standarized disk susceptibility test (2). Zone sizes, with 30-jg disks of the two compounds, against 18 isolates representing
T112 (h) 0.9 3.8
Vol. 12, No. 1 Printed in U.S.A.
Delineation of the Relative Antibacterial Activity of Cefamandole and Cefamandole Nafate J. R. TURNER,* DAVID A. PRESTON, AND J. S. WOLD Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, Indiana 46206
Received for publication 22 December 1976 By conventional laboratory evaluation procedures, the in vitro antibacterial activities of cefamandole and its O-formyl ester, cefamandole nafate, appear virtually identical. When the activities of these two compounds were exaniined for their ability to lyse log-phase cultures of susceptible bacteria, however, cefamandole was found to be about 10 times more active than cefamandole nafate. Cefamandole nafate was shown to be rapidly converted to cefamandole in bacteriological media, with a half-life of less than 1 h at a pH of 7.0 or above. At pH 6.0, in log-phase inhibition experiments, however, cefamandole nafate is more stable, allowing delineation of the activity between cefamandole and cefamandole nafate. The efficacy of cefamandole was identical to that of cefamandole nafate in treating experimental animal infections, indicating that rapid conversion of cefamandole nafate to cefamandole occurs in vivo.
The antibacterial activity of cefamandole, 7n-mandelamido- 3 f [(1- methyl-1H-tetrazol-5-yl)thio]methyl}3-cephem-4-carboxylic acid, was first described by Wick and Preston (7), who showed that this antibiotic has very good activity against gram-negative bacteria, including ,8-lactamase-producing strains of Proteus and Enterobacter. Subsequently, this antibiotic was introduced into clinical trial as the sodium salt of the O-formyl ester of cefamandole and designated as cefamandole nafate. This derivative, which is stable in the crystalline form, is rapidly converted to cefamandole by hydrolysis of the formyl ester in vitro after dissolution. Because of rapid in vivo conversion of cefamandole nafate to cefamandole (4), the latter is the predominant circulating antibiotic after administration of cefamandole nafate to laboratory animals or humans (J. S. Wold, R. R. Joost, H. R. Black, and K. E. Briscoe, Ninth International Congr. Chemother., M225, 1975). The in vitro antibiotic activity of cefamandole nafate was shown (R. B. Kammer, D. A. Preston, and J. R. Turner, Ninth International Congr. Chemother., M226, 1975) to be equivalent to cefamandole lithium by conventional agar dilution and disk plate susceptibility tests against a number of bacterial pathogens. It was also reported that a differential in activity between these two antibiotics was detectable against Bacillus subtilis based on a decreased growth rate in log-phase cultures, although only at concentrations below 0.1 ,ug/ml. We have shown that these apparently contradictory findings are the result of two basic differ-
ences in experimental conditions between the two tests: the duration of the experiments and the pH at which these experiments were performed. MATERIALS AND METHODS Antibiotics. Cefamandole lithium and cefamandole nafate (the O-formyl ester of cefamandole sodium) were obtained from the Lilly Research Laboratories, Eli Lilly and Co. Bacterial cultures. B. subtilis (ATCC 6633) was obtained from the American Type Culture Collection (ATCC). All other bacterial strains used were originally clinical isolates collected from various sources over several years and maintained in this laboratory. Disk susceptibility tests: effect of pH. Antibiotic activity at pH 6.0 or 8.0 was determined by the disk susceptibility test method using Lilly Biochem agar no. 6 (beef extract, 0.15%; yeast extract, 0.6%; peptone, 0.6%; and agar, 1.75%) or antibiotic medium 5
(Difco).
MICs. Minimal inhibitory concentrations (MICs) were determined by the agar dilution procedure described by Ericsson and Sherris (1). Inocula of approximately 104 bacteria were applied to each spot on the surface of antibiotic-containing Mueller-Hinton agar plates by use of a multipronged inoculating device. Growth medium and conditions for log-phase cultures. Minimal medium at pH 6.0 consisted of: (NH4)2S04, 0.2%; K2HPO4, 1.4%; KH2PO4, 0.6%; MgSO4-7H2O, 0.02%; sodium citrate, 0.1%; vitaminfree Casamino Acids (Difco), 0.02%; and glucose, 0.5%. Log-phase cultures were prepared by inoculating 50 ml of medium with a sample of a 16-h culture sufficient to give a reading of 5 Klett units. The cultures were incubated on a rotary shaker at 37°C 67
68
TURNER, PRESTON, AND WOLD
ANTmiCROB. AGENTs CHEMOTHER. TABLz 1. Agar dilution MIC of two forms of cefamandole for a variety of gram-positive and gramnegative bacteria
at 180 rpm. Antibiotic additions were made after log-phase growth haft been established. The antibiotic activity was quantitated in terms of the time required for lysis of the culture to occur. The measurement used for comparative purposes was termed "time to lysis" and was the point in time after addition of the antibiotic when the Klett reading decreased or remained constant. Assay of cefamandole and cefamandole nafate in culture medium. Uninoculated medium (agar or liquid culture medium) to which antibiotic had been added was incubated under the same conditions as the experimental cultures. Samples of these media were removed at appropriate time intervals, acidified with glacial acetic acid (final concentration, 25%), and frozen until assayed. The assay was accomplished by high-pressure liquid chromatogrphy as described by Wold et al. (Ninth International Congr. Chemother., M225, 1975). Experimental infections. Streptococcus pyogenes
Geometric mean
No. Bacteria Bested
Enterobacter aerogenes Enterobacter cloacae Escherichia coli Klebsiella pneumoniae Proteus mirabilis Proteus morganii Proteus rettgeri Proteus vulgaris Salmonella typhi Serratia marcescens Shigella flexneri Staphylococcus aureus
B. SUBTIUS z
ANTIBI
25-
C
B. SUBTIUS
MEDIUM
5
AT PH I
C
8 7 6 10 7 2 3 2 1 1 1 3
Nafate
1.2 2.7 1.3 1.0 1.3 0.7 1.6 2 0.25 64 0.13 0.63
1.1 2.7 1.3 0.9 1.0 0.5 1.0 1.4 0.13 64 0.25 0.51
/
ANTIBIOTIC MEDIUM 5 AT pH 8
CEFAMANOOLE NAFATE CEFAMANDOLE LITHIUM a0
MIC (Qg/ml) Lithium
10I~~~~~~~
O-)
F aKftl I.3
,2.5
..
I
I
10
I
I
I
15 20 25 30 ZONE DIAMETER {mm}
I
35
I
I
10
I
I
15 20 25 30 ZONE DIAMETER (mm}
I
35
50 U c
25
S. AUREUS
S. AUREUS
AGARNO. 6 ATPH
AAR NO. ATPH8
F
1.3 z
10
1.W
=2
2.515 20 25 30 ZONE DIAMETER (mml
10 e
35
10
15 20 25 30 ZONE DIAMETER Imm)
35
50-
-
c2
E. COU 2525
z Cl 9
II
/
E. COU
AGAR NO. 6 AT
AGAR NO. 6 AT PH 8
.
10
/ 5~~~~~~
2.5~~~~~ I 10
I I I I 20 25 30 15 ZONE DIAMETER (mm)
-
I
I
35
10
I
I
I
IA
15 20 25 30 ZONE DIAMETER (m)
35
FIG. 1. Assay of the antibiotic activity of two forms of cefamandole in agar media at pH 6.0 and 8.0.
69
CEFAMANDOLE, CEFAMANDOLE NAFATE ACTIVITY
VoL. 12, 1977
TABLE 2. Hydrolysis of cefamandole nafate to cefamandole in several bacteriological agar media Unchanged cefamandole nafate (%) remaining at. Agar medium
Mueller-Hinton agar (BBL) Antibiotic medium 5 (Difco), adjusted to low pH Antibiotic medium 5 (Difco), adjusted to high pH Lilly biochem agar no. 6, adjusted to low pH Lilly biochem agar no. 6, adjusted to high pH a Half-life of cefamandole nafate.
Initial
pH
5min
lh
2h
3h
4h
7.4 6.5
99.9 100+
48.9 91.9
11.2 73.9
3.5 57.7
0 49.4
8.1
67.0
0
0
0
0
6.0
100+
100+
91.5
85.7
78.4
8.15
72.9
0
0
0
0
and Streptococcus pneumoniae were grown overnight in brain heart infusion broth (BBL) supplemented with 5% defibrinated rabbit blood. Decimal dilutions of the test culture were made in brain heart infusion broth without blood to yield the previously determined infective concentration of organisms. McAllister strain Swiss white mice (13 to 15 g, random sex) were infected by intraperitoneal injection of 0.5-ml volumes of the desired bacterial suspension.
Staphylococcus aureus, Escherichia coli, and Klebsiella pneumoniae were grown for 18 h at 36°C on Trypticase soy agar (BBL) slants. The surface growth was suspended in saline to a density of 109 colony-forming units/ml and diluted to predetermined infective concentrations. The final dilution of these latter cultures was made in sterile 5% hog gastric mucin at pH 7.4. As with the streptococci, these bacterial suspensions were administered by intraperitoneal injection in 0.5-ml volumes. Groups of eight infected mice were treated subcutaneously with antibiotic (0.25-ml volumes) at 1 and 5 h after infection. Survivors and deaths were recorded daily for 7 days. The amount of antibiotic required to protect 50% of the infected mice (ED50) was estimated by the method of Reed and Muench (6). RESULTS
Comparison of the in vitro activity of cefamandole nafate with that of cefamandole as determined by agar dilution MIC, led to the conclusion that there was little, if any, activity difference between the two forms of cefamandole. Table 1 shows the geometric mean MICs for various genera. The MIC of both compounds was the same for 36 of the 51 bacterial isolates tested. There was one dilution difference in favor of cefamandole nafate with 13 isolates, and a one-dilution difference in favor of cefamandole lithium with 2 isolates. Similar results were obtained with the Food and Drug Administration standarized disk susceptibility test (2). Zone sizes, with 30-jg disks of the two compounds, against 18 isolates representing
T112 (h) 0.9 3.8
