ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, June 1978, p. 930-938
0066-4804/78/0013-0930$02.00/0 Copyright © 1978 American Society for Microbiology
Vol. 13, No. 6
Printed in U.S.A.
Azlocillin and Mezlocillin: New Ureido Penicillins KWUNG P. FU AND HAROLD C. NEU*
Departments of Medicine and Pharmacology, College of Physicians and Surgeons, Columbia University, New York, New York 10032
Received for publication 24 June 1977
The activity of azlocillin and mezlocillin, new semisynthetic ureido penicillins, and compared with that of other known /8-lactam antibiotics. At of 25 ,ug/ml, azlocillin inhibited 74% of Enterobacter, 97% of Proteus mirabilis, 64% of Citrobacter, 91% of Pseudomonas aeruginosa, and 82% of Bacteroides strains tested. Mezlocillin inhibited 86% of Shigella, 96% of Enterobacter, 80% of indole-positive Proteus, 88% of Bacteroides, and 63% of Pseudomonas strains tested. Azlocillin was more active against Pseudomonas than was ticarcillin, carbenicillin, or mezlocillin. Mezlocillin was more active than carbenicillin and ampicillin against Escherichia coli, Klebsiella, Enterobacter, Citrobacter, Acinetobacter, Serratia, and Bacteroides. Azlocillin and mezlocillin were less active than cefazolin against ,B-lactamase-producing E. coli and Klebsiella strains but more active than cefazolin against Enterobacter, indole-positive Proteus, Acinetobacter, Citrobacter, and Serratia strains. Both compounds showed activity equivalent to that of cefoxitin against Bacteroides isolates. Both agents were destroyed by many of the ,B-lactamases from gram-negative organisms. was investigated a concentration
Although a number of new penicillins have been introduced in recent years, there is still a need for agents with an enlarged spectrum of activity. The discovery of carbenicillin added Pseudomonas aeruginosa, Enterobacter cloacae, and less commonly encountered species such as Acinetobacter calcoaceticus and the indole-positive Proteus to the spectrum of organisms inhibited by penicillins. Ticarcillin has a two- to fourfold greater activity than carbenicillin against Pseudomonas, but it is no more active than carbenicillin against members of the Enterobacteriaceae (7). Neither carbenicillin nor ticarcillin is active against Klebsiella, and their activity against Serratia is extremely variable. Strains of Escherichia coli possessing TEM-type /8-lactamases are resistant to all of the commercially available penicillins. Furthermore, the serum and tissue levels of carbenicillin and ticarcillin needed to inhibit many hospitalacquired strains of Pseudomonas are high, and hence the doses of the agents used are also high, resulting in a greater incidence of side effects (1, 5, 6). We therefore were interested in evaluating the activities of azlocillin and mezlocillin, new broad-spectrum penicillins, and comparing their activity with that of other 83-lactam antibiotics. (These findings were presented at the 16th Interscience Conference on Antimicrobial Agents and Chemotherapy, 27-29 October 1976,
Chicago, Ill.)
MATERIALS AND METHODS Azlocillin and mezlocillin were obtained from Delbay Pharmaceuticals. Other antibiotics were gifts from their respective manufacturers. Bacterial strains were those isolated from patients hospitalized at the Columbia-Presbyterian Medical Center. The isolates came from blood, sputum, and urine specimens and did not represent the same strains insofar as could be determined by epidemiological means, general antibiograms, and bacteriocin typing. Susceptibility tests. The antimicrobial activity was measured by agar dilution or broth dilution methods as previously described in more detail (3). The minimum inhibitory concentration (MIC) in agar was determined with a replicating device by using a final inoculum on the plate of 105 colony-forming units (CFU). Standard medium used was Mueller-Hinton (BBL). Broth dilution was performed with MuellerHinton broth, with a final inoculum in the assay of 105 CFU. The MIC was the concentration at which there was no visible growth, and the minimum bactericidal concentration (MBC) was the concentration which on subculture yielded fewer than five colonies after 24 h of incubation. ,B-Lactamase preparations and assays. Partially purified 8?-lactamases were prepared by published methods (8). A modification of the Novick microiodometric assay was used (10).
RESULTS The overall activity of azlocilhn against 524 bacterial isolates is shown in Table 1. We have included a large number of isolates that possess 930
VOL. 13, 1978
AZLOCILLIN AND MEZLOCILLIN
931
TABLE 1. Activity of azlocillin against gram-positive and gram-negative organisms No. of Percentage inhibited by an MIC (ug/ml) equivalent to:' Organisn isolates
Staphylococcus aureus ..... 16 Streptococcus faecalis ...... 15 S. pyogenes ................ 30 Haemophilus ........ 10 Neisseria 5 Streptococcus pneumoniae 5 Escherichia coli ........... 61
Shigella
14
-1.6
3.1
69 87 100 100 100 100 3
94 100
16 14
6.3
12.5
25
50
100
200
31
36
49 86 54 90 20
67
65 97 24
94
53 76
88
100
79
82
33 92 77 94
53 94
>400
70
100 100 100 100 100
100
25
30
29
Klebsiella ................. 48 2 4 6 13 40 52 Enterobacter .............. 31 3 15 74 58 84 Serratia 50 2 4 8 18 Salmonella 15 7 47 40 Citrobacter 25 4 12 24 64 68 Proteus, indole positive ..... 28 14 54 68 75 P. mirabilis ............... 31 69 87 94 97 Providencia ............... 15 7 13 20 27 Pseudomonas aeruginosa 95 5 15 29 78 91 Acinetobacter 13 15 23 62 38 Bacteroides ............... 17 29 53 82 88 a MICs were determined by agar dilution with Mueller-Hinton agar (BBL).
,8-lactamases. The E. coli, Salmonella, and Shigella isolates tested are those which would be encountered in a hospital setting rather than in a community, since this agent would only be administered parenterally in the hospital. Of strains of beta-hemolytic streptococci (groups A, B, C, and G) tested, 95% were inhibited by 0.1 jig/ml. Strains of Staphylococcus aureus which lack ,B-lactamases were inhibited by 0.1 to 0.5 jig/ml, and even most ,B-lactamase-producing isolates were inhibited by 3.1 ,ig/ml. Azlocillin inhibited 87% of Streptococcus faecalis strains at a concentration of 1.6 ,ig/ml and 100% at 3.1 jig/ml. All of the small number of Haemophilus influenzae and Neisseria gonorrhoeae isolates, including several f8-lactamases-producing, ampicillin-resistant strains, were inhibited by 0.1 to 0.5 ,ug/ml. Strains of Streptococcus pneumoniae were inhibited by 0.1 to 0.5 ,ug/ml, but penicillin G was twofold more active against these isolates. The activity of azlocillin against Enterobacteriaceae varied from family to family. E. coli strains that are ampicillin susceptible were inhibited by 1.6 to 12.5 jig/ml. A small percentage of isolates was inhibited by 25 to 100 ,ug/ml, but the MICs for most ,B-lactamase-containing strains were 200 jig or more per ml. The activity of azlocilhin against Shigella sonnei and Salmonella was similar to its activity against E. coli and correlated with the presence of 18-lactamases. Azlocillin inhibited 40% of Klebsiella pneumoniae strains at a concentration of 25 jAg/ ml, but the MICs for 35% of isolates were above 200,ug/mI. Seventy-four percent of Enterobacter strains were inhibited by 25 ,ug/ml, and 90% were
400
93
38
100
98 85
100 100 100 100 100
100
inhibited by 100iLg/ml. Serratia, on the other hand, was extremely resistant, with only 18% of strains inhibited by 25 jig/ml. Azlocillin inhibited 75% of indole-positive Proteus strains, mainly P. morganii and P. vulgaris, at 25 jig/ ml, but Proteus rettgeri and Providencia stuartii were much more resistant; for these the MICs were greater than 200 jig/ml. Eighty-two percent of Bacteroides fragilis strains were inhibited by 25,ug/ml. However, the most impressive activity was seen with strains of Pseudomonas aeruginosa: 91% of isolates were inhibited by 25 ,ug/ml, and 78% were inhibited by 12.5 jig/ml. The in vitro activity of mezlocillin against gram-positive cocci and gram-negative bacilli is shown in Table 2. The majority of fB-hemolytic streptococci (groups A and B) as well as Streptococcus pneumoniae were inhibited by less than 0.5 jg/ml. Staphylococcus aureus and S. epidermidis that do not produce fi-lactamase were inhibited by less than 0.5 jig/ml, whereas the MICs for fl-lactamase-producing staphylococci were up to 50 jug/ml. All isolates of ,Bhemolytic streptococci (groups A, B, F, and G) tested were inhibited by 0.05 to 0.1 jig/il. H. influenzae was inhibited by 0.1 to 0.2 ,ug/nil, as was N. gonorrhoeae. The activity of mezlocillin against E. coli was trimodal. Non-hospital-acquired strains that lack 8?-lactamases were inhibited by 0.8 to 6.3 ,ug/ml. These included a small number of E. coli strains that were inhibited by concentrations of 25 to 100,ug/ml. Most ampicillin-resistant strains of E. coli were resistant to mezlocillin; the MICs were 400 jig or more per ml. The activity of
932
FU AND NEU
ANTIMICROB. AGENTS CHEMOTHER.
mezlocillin against Shigella and Salmonella was hibited 63% of Pseudomonas aeruginosa isosimilar to its activity against E. coli, with a lates at a concentration of 25 ,ag/ml. However, bimodal distribution of susceptible and resistant 97% of P. aeruginosa were inhibited by 100 strains. Citrobacter isolates were susceptible to ,ug/ml, and 20% of Klebsiella were not inhibited mezlocillin, with 80% inhibited by 6.3 ,ug/ml. by 400 ,tg/ml. Sixty-five percent of B. fragilis Over 97% of Proteus mirabilis and 70% of indole- were inhibited by 12.5 ,ug/ml, and 100% were positive Proteus, including P. morganii, P. vul- inhibited by 100 ,ug/ml. garis, and P. rettgeri, were inhibited by 1.6 jig Comparison of the MICs determined by the or less per ml. In contrast, Providencia ranged agar plate technique and by the broth dilution in susceptibility from 1.6 ,ug/ml to greater than method yielded similar results with the strains 400 ,ug/ml. Enterobacter cloacae and E. aero- of E. coli, Klebsiella, Enterobacter, Citrobacter, genes were inhibited by 50 ,tg or less per ml, and and Proteus morganii tested; only a single di80% were inhibited by 12.5 ,tg or less per ml; the lution difference was seen with an occasional MICs for 50% of Serratia were greater than 100 isolate. The MICs and MBCs were identical or ytg/ml. The majority of Acinetobacter were in- differed by only one dilution for members of the hibited by 50 ,ug or less per ml. Mezlocillin in- Enterobacteriaceae (Table 3). However, the TABLE 2. Activity of mezlocillin against gram-positive and gram-negative organisms No. of isolates c1.6
Organism
Staphylococcus aureus ........ 18 Streptococcus epidermidis ..... 35 S. faecalis .................... 15 Haemophilus ................. 10 Neisseria gonorrhoeae ........ 5
72 46 100 100 100
Streptococcus, 8-hemolytic
100
Serratia Salmonella Citrobacter
100 25 14 6 23 4 22 24 71 97 13 7 22
30 Streptococcus pneumoniae ..... 10 Escherichia coli .............. 44 14 Shigella Klebsiella .................... 31 Enterobacter ................. 31 49 15 25 28 31 15 76 13 17
Percentage inhibited by an MIC (,ug/ml) of:a 3.1
6.3
12.5
94 74
100 80
83
30 21 23 61 19 47 58 76
34 43 43 72 23 73 79
39 86 56 81 27
25
50
200
400
91
43 71 96 35
48
62 87 85 94
100
52
65
77
100 100
96
100 84
88
100 100 100 100
80 54 63 56 88
100 100
57
83 100 38
>400
100
45 93
88
Proteus, indole positive ........ P. mirabilis .................. Providencia .................. 31 39 46 Pseudomonas aeruginosa ...... 12 17 36 Acinetobacter 31 44 Bacteroides fragilis ........... 29 65 a MIC determined by agar dilution with Mueller-Hinton agar.
100
92 99 100
97
100
TABLE 3. Comparison of MICs and MBCs of aziocillin and mezlocillin
Organism
No. tested
No. of isolates with MICa and MBC identical
No. of isolates with MBC greater than MIC (fold)
2
Ab
4
M M A A A Escherichia coli ...... 6 2 3 4 3 Enterobacter ......... 5 2 2 3 3 Klebsiella ............ 6 2 4 4 1 Proteus morganii ..... 3 1 2 2 1 Citrobacter ........... 2 1 2 1 Serratia .............. 2 2 2 Pseudomonas ........ 10 1 4 1 3 a MIC values were determined by broth dilution method in Mue11er-Hinton broth. b
A, Azlocillin.
C M,
Mezlocilin.
8
216
Me
M
A
M
6
3
1
2
VOL. 13, 1978
MIC for a number of Pseudomonas strains was at least eightfold greater than the MIC of both azlocillin and mezlocillin. Table 4 illustrates the effect of medium used to assay the activity of azlocillin and mezlocillin. The MICs and MBCs varied over a twofold dilution range without a consistent pattern, with brain heart infusion broth and Trypticase soy broth not yielding significantly higher MICs than those obtained in Mueller-Hinton broth. Of interest is the fact that nutrient broth, which has a low conductivity, 1.6 mS compared with 12 mS of Mueller-Hinton broth, and also a low osmolality, 60 mosmol compared with 440 mosmol, did not yield lower MICs. There was some pH effect, as shown by the lower MICs and MBCs in Mueller-Hinton broth at pH 8 compared with those in broth prepared at pH 6. This was noted with both Enterobacteriaceae and Pseudomonas. The effect of the inoculum size on the azlocillin and mezlocillin MICs and MBCs is shown in Table 5. With 103 to 106 CFU, the MICs and MBCs for the Enterobacteriaceae and the Pseudomonas strains tested were similar, with only a twofold increase in the concentrations. However, with 107 CFU of most organisms tested, the MICs rose above 400 Ag/ml. Those Pseudomonas isolates with which there was only a slight rise in the MICs with 107 CFU showed MBCs of >400 ,ug/ml. The presence of 50% normal human serum in the assay increased the MICs required for inhibition of some organisms twofold. MBCs were increased two- to eightfold. For example, the MICs and MBCs of mezlocillin against E. coli 2371 were 6.3 and 12.5 ,g/ml, respectively, in broth and were 12.5 and 50 ,ug/ml, respectively, in 50% serum. In contrast, with Klebsiella 2104 the MICs and MBCs were the same: 3.1 ,tg/ml in broth and 3.1 and 6.3 ,ug/ml in serum. Comparison of the activity of azlocillin and mezlocillin with other ,8-lactam compounds. The activity of azlocillin and of mezlocillin against various gram-negative bacilli was compared with the activitv of ampicillin, carbenicillin, and one of the available cephalosporins (Table 6). Azlocillin and mezlocillin were similar in activity to ampicillin and carbenicilhin against E. coli isolates lacking a fl-lactamase, but mezlocillin and azlocillin were more active at high concentrations, i.e., at 100 to 200 ,ug/ml, than was carbenicillin. Cefazolin was twice as active as either compound against these hospital isolates of E. coli. Mezlocillin was much more active than ampicillin, azlocilHin, or carbenicillin against Klebsiella; 77% of isolates were inhibited by 25 ,ug of mezlocillin per ml, whereas none were inhibited by carbenicillin. Although azlo-
AZLOCILLIN AND MEZLOCILLIN
933
cillin inhibited many of the Klebsiella strains, cefazolin showed much greater overall activity against Klebsiella than did any ofthe penicillins. The activity of azlocillin and of mezlocillin against P. mirabilis was equivalent to that of ampicillin and carbenicillin. Ampicillin was the most active agent; 50% of isolates were inhibited by 0.4 ,ig/ml. Mezlocillin was twofold more active than carbenicillin against Proteus morganii and P. vulgaris and was the most active of the five compounds tested. Mezlocillin and carbenicillin had equivalent activity against Providencia stuartii; 50% of isolates were inhibited by 12.5 ,ug/ml. Azlocillin was twofold less active than mezlocillin or carbenicillin. Mezlocillin was appreciably more active against Enterobacter than was carbenicillin or azlocilhin or even cefamandole; 88% of Enterobacter isolates were inhibited by 12.5 Ag of mezlocillin per ml, compared with 53% inhibited by carbenicillin and 46% inhibited by cefamandole. Mezlocillin was twofold more active than azlocillin and carbenicillin against Citrobacter, but less active than cefamandole. Against Acinetobacter, azlocillin and mezlocillin were similar in activity to carbenicillin, but both were more active than carbenicillin against Serratia. Against B. fragilis, 25-,ug/ml concentrations of antibiotics inhibited the following percentages: ampicillin, 33%; carbenicillin, 28%; cefazolin, 5%; cefoxitin, 83%; azlocillin, 83%; and mezlocillin, 88%. Azlocillin was the most active penicillin tested against Pseudomonas aeurginosa. Twice as many isolates were inhibited by azlocillin as by ticarcillin at a concentration of 6.3 ,tg/ml. Mezlocillin and carbenicillin at low concentrations had equivalent activity against Pseudomonas, but more Pseudomonas isolates were inhibited by mezlocillin in concentrations of 50 to 200 ,ug/ml than by carbenicillin. Comparison of the activity of azlocillin, mezlocillin, ampicillin, and carbenicillin against susceptible and resistant Shigella sonnei showed that mezlocillin was consistently two- to fourfold more active. Comparison of the activity of ampicillin, carbenicillin, and mezlocillin against Salmonella showed that mezlocillin was more active against selected isolates, but there was no consistent pattern of increased activity. Table 7 illustrates the comparative activity of mezlocillin as compared with carbenicillin and ampicillin against various gram-positive organisms. The percentage of Staphylococcus epidermidis strains inhibited by mezlocillin at c1.6 ,ug/ml was twofold greater than that observed with carbenicillin. Mezlocillin was less active against Staphylococcus aureus than was carbenicillin and approximately as active as ampicillin. Against streptococci, ampicillin was the most
934
FU AND NEU
ANTIMICROB. AGENTS CHEMOTHER.
C-4
~
0
CO
0
(D
CQC~~ C~~
~
CO0
CO
LO Lo .6 C
o'~(-~ 8QE 8 qq00o
LoCq
-
-
Lfq
C~~
_6 ci
~c6
vv
L0L
-~~~
CO
C
cI
LoC
-
-V
V
-
0_
-zo
ult~ U:
LO
LO
r-4
C.)
la
6.5 0
E
=
4)
14)
Lc
C6
ciC
l ~6 6ciC l
_6 v V___---cfii e6 C6 V
cqi-cli C- C'V
COqt
C cli
c
cC-q
4
cl
L
Cr 0 LOCO 6C Oc co LO
Ct
V) V c0
V-
L C' C> L
ci
z.4
D
.E
t ci _C' IL6s f C' vs_ MCL
Pcs
8
_ 0 CO
Lo CO
-
*N
4 o_L _ cO
_A
c-
O
6V 6
C
6 C.)1 *,4 0
00
_D Mo Lo Lo
c: o
0
g
Lo cO
0.
0
1-
6 6c'i
e u
'.
-4 c~Lo m __LoL O_ _co 5 ,VVj v Lo N "-
CY
0
41) 4)
.0
.0
CY m m D
N .)
CO LO _O
x
a CO
C COQO O OeO
U)
C)
C
e
.)
0 N
_
0 U)
~C6 Q
Nt
O
.2 cD
L C
OO
0
'QCIQ
at
4) 4)
-4) ,
LOe
CD
LO CSi
O"1 S: CS-4
U)
C14
C CDi C14C
-I Cq ,-4 ->C-4 O C14
CSn4
C.
C.N
Cq
C L LO OO LO CD _ O C : ONLOSO
VV
A
O O
__ A_
_
AA
_ --4
0
o
--
C4
s O Lo
c14
Q
.t46)
0
o oqo cq C-
C
AO AO
-CO q 0e 0~ 0- 0-
L N
0L
lt 0 n
C.)
i
AA A
-
A m
A~ A
A
-
A
LO
A
C.)
4l)
c-i _ o cq
IC .
0
0r
8 8IRr 8lw 8I, 8 8rv88 v v A A A
A
A
A
8v 8"W A
A
A AA
A
A
A
A
A
1-4
H)
A
0 -4
B
mI oo
l
o C1
1
cl
oso Iqo 1
t
m
CSl C.)
._c 4-J0
0 .0
._ ¢9
¢ 04CO 00 00 CO
* C.) ° ¢~ g 0C.) 0 ¢2¢ C.)Qs Cz ¢.) *s *gQ~*g
C.
= c; c; _ Q m cJ =~~1= = c; z cz _>0~0"~0~0" cl N c) sro
c
-0
LO~~~~~~~~0 0 cli
.40
=
-sb e
c
0 .3E -z,~
*
s
CO
r. 0
--4
r-
0
I.. 4
m
b
s
H
~
S SX0 0 C.) X 0 -U6C.
U.)
O .)
0 4)
&
_
r-
UE
S.-4 4.
i)
*~ U
..
0
0
*
~4
VOL. 13, 1978
AZLOCILLIN AND MEZLOCILLIN
935
TABLE 6. Activity of azlocillin and mezlocillin compared with that of known penicillins and cephalosporins Organism (no. of isolates)
Percentage of strains inhibited by an MIC' (pg/ml)' of five penicillins of:
Antibiotic :51.6
3.1
6.3
12.5
25
50
29 15 23 28 29 36 Azlocillin 24 30 38 Mezlocillin 15 25 30 Ampicillin 18 28 30 44 Carbenicillin 18 42 62 88 Cefazolin 80 82 Klebsiella (21) Azlocillin 4 8 20 46 65 Mezlocillin 4 23 46 62 77 Ampicillin 28 67 Carbenicillin 5 Cefazolin 62 81 95 Enterobacter (47) Azlocilin 24 53 68 72 Mezlocillin 16 60 72 88 96 100 2 Ampicillin 4 8 13 19 Carbenicillin 12 20 49 53 60 68 Cefazolin 2 Cefamandole 22 33 41 46 54 61 Proteus mirabilis (31) Azlocilin 96 Mezlocillin 97 Ampicillin 97 Carbenicilin 97 Cefazolin 32 52 72 96 Proteus, indole positive (35) Azlocillin 19 57 64 72 Mezlocillin 67 76 Ampicillin 2 26 Carbenicilin 39 46 58 67 74 Cefazolin 5 14 16 Providencia (15) Azlocllin 7 13 20 27 Mezlocillin 13 31 39 46 54 62 Ampicillin Carbenicillin 29 36 50 Citrobacter (37) Azlocillin 1 3 4 12 30 55 Mezlocillin 4 12 24 64 68 Ampicillin 8 24 38 Carbenicillin 14 22 40 Cefazolin 34 37 39 53 63 79 81 Cefamandole 84 92 94 Acinetobacter (17) Azlocillin 34 50 Mezlocillin 38 62 Ampicillin 9 Carbenicillin 9 36 45 Cefazolin Serratia (23) Azlocillin 5 8 16 29 Mezlocillin 9 5 18 32 Ampicillin 9 13 17 22 26 Carbenicillin 13 17 Cefazolin Pseudomonas aeruginosa Azlocillin 6 19 37 70 82 86 (76) Mezlocillin 6 12 16 35 63 87 24 27 33 Ticarcillin 47 73 87 Carbenicillin 11 14 17 30 38 68 Bacteroides (17) Azlocillin 29 53 83 88 Mezlocillin 29 65 88 94 Ampicillin 33 78 28 Carbenicillin 83 Cefazolin 33 6 72 Cefoxitin 77 83 a MICs were determined by the agar dilution method in Mueller-Hinton agar.
Escherchia coli (50)
100
200
400
>400
48 60 34 48 90
67 76 36
72 82 58
96 80 92 81 38
92
100 100 100 100 100 100
86 52
100 100 100
100 78
92
100
32 68 13 74
47 70 19 82
62 28 89
80
85
55 82 51 53
77 84 81
76
42 23 33
92 61 76 59 51 60 100 63 77 36 55
83 88 67 73 86
55 9
40 47
92 97 99 78 94 100 83 100 72 94
95 100 86 100 89 100 100
57 100 100 78 83 89
100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100
100 100 100
82 85 64 63 64 50 54 34 26 9 100 99
100 100 100 100 100 100 100 100 100 100
95
100
100
100
936
FU AND NEU
ANTIMICROB. AGENTS CHEMOTHER.
active agent. Mezlocillin was the most active agent against Streptococcus faecalis. At a con-
by ,B-lactamases. Partially purified ,B-lactamases of E. coli, Citrobacter, Acinetobacter, Providencia, and Pseudomonas were examined for ability to hydrolyze azlocillin and mezlocillin, and the rates of hydrolysis were compared with
centration of 0.8 ,ug or less per ml, it inhibited all the isolates. The activity of azlocillin and mezlocillin against some 8)-lactamase-producing organisms is shown in Table 8. Azlocillin and mezlocillin were more active than ampicillin and carbenicillin against 8-lactamase-producing Shigella, Klebsiella, Enterobacter, Serratia, and Acinetobacter. However, cefoxitin and cefazolin were the most active against the /i-lactamase-producing strains of E. coli, Shigella, and Providencia tested. Twelve Pseudomonas aeruginosa isolates resistant to carbenicillin were tested for their susceptibility to azlocillin, mezlocillin, and ticarcillin (Table 9). Azlocilhin was more active than ticarcillin against all but one of the carbenicillinresistant Pseudomonas isolates tested, whereas mezlocillin was more active than ticarcillin against some strains. Hydrolysis of azlocillin and mezlocillin
TABLE 9. Activity of azlocillin and mezlocillin against 10 carbenicillin-resistant strains of Pseudomonas aeruginosa MIC (ug/ml) of: Strain no.
Carbenicillin
Azlo-
cillin
2800 2800
2131 2137 2699 2711 2960 3330 3331 3467 3423 3408
Mezlocillin
Ticarcillin
.-800 100 100 25 25 50 100 25 200 200
2800 2800
2800 200 25 12.5 100
200 200
2:800 200 200 2800 400 400
12.5 12.5 50 25 50
100 100
2800 100 50
2800 200 400
TABLE 7. Comparative activity of mezlocillin and carbenicillin against some gram-positive organisms Antibiotic
Organism (no. of isOlate8)
s0.8
Staphylococcus epi- Carbenicillin dermidis (35) Mezlocilhin S. aureus (30) CarbeniciHlin MezlociUin AmpiciNlin Streptococcus fae- Carbenicillin calis (31) Mezlocillin AmpiciLin Beta-hemolytic Carbenicillin streptococci (32) Mezlocillin
17 31 33 30 60 3 100 97 91 97 100
Ampicillin
Percentage of strains inhibited by an MIC (yg/ml) of: 3.1 6.3 12.5 25 50 100 200 400 51 29 43 91 80 86 46 74 80 83 91 63 83 100 43 73 83 93 97 100 73 77 87 90 97 93 100 6 65 100
1.6
100 94
97 100
2800 100 100
100
TABLE 8. Comparative activity of aziocillin and mezlocillin against JI-lactamase-producing organisms MIC (Ag/nm) of: Organism
Escherichia coli 3083 E. coli 3138 Shigella sonnei 12-31 Klebsiella 2762 Klebsiella 3001 Enterobacter 2249 E. cloacae Serratia 2788 Serratia 2773 Providencia 2395 Pseudomonas 2983 Acinetobacter 30 Proteus morganii 1635 P. rettgeri Pseudomonas 3647
Azlocllin
Mezlocllin
2800
12.5 2800
6.3 6.3 12.5 25 12.5 25 100
2800 12.5 200
50 25 100 3.1 25 12.5 100 400 25 200 50
2800 50
25
Amnpi cillin
Carbenicllin
400
50
2800 2800
2800
100
400 200
2800 100
2800 25
400 200
2800 2800 2800 2800 2800 2800 2800
2:800
12.5 2800 2800 25 2800 400 2800 2800
Cefazolin
Cefoxitin
3.1 25 1.6
3.1 1.6 1.6 100 12.5
2800 6.3
2800 2800 2800 2800 2800 2800 2800 800
2800 2800
2800 2800 25 100 0.8
2800 2800 800 3.1
2800
VOL. 13, 1978
AZLOCILLIN AND MEZLOCILLIN
937
TABLE 10. Comparative ,3-lactamase stability of azlocillin and mezlocillin Relative rate of hydrolysisa
Organism
Ampicillin
Carbenicillin
Escherichia coli 3337 43 95 Citrobacter 2732 250 42 Citrobacter 2836 28 30 Enterobacter 2647 180 62 Pseudomonas aeruginosa 3901 76 26 Proteus mirabilis 3378 84 71 Acinetobacter 2284 107 110 Providencia 2395 66 53 aActivity relative to the rate of hydrolysis of penicillin G.
those of penicillin G, ampicillin, carbenicillin, and ticarcillin (Table 10). Both compounds were hydrolyzed at rates comparable to the rates for ticarcillin, although there were selected differences. Protein binding. The binding of mezlocillin to human protein was determined at several concentrations of antibiotic by the agar diffusion assay. Protein binding was 20% at both low (10 ,ug/ml) and high (100 ,g/ml) concentrations. Azlocillin was less protein bound (12% by the agar diffusion assay).
DISCUSSION Azlocillin and mezlocillin were shown in this study to have a broad spectrum of antimicrobial activity against both gram-positive and gramnegative organisms. The in vitro activity of these compounds against streptococci and staphylococci is comparable to or greater than that reported earlier for carbenicillin and ticarcillin (7, 8). The activity of both compounds against Enterobacteriaceae varied with the isolates studied. Azlocillin and mezlocillin were slightly more active than carbenicillin and ticarcilhin against E. coli, Salmonella, Shigella, and Citrobacter strains lacking f8-lactamases. It is against the family Klebsielleae that mezlocillin has extended the range of activity of this type of compound. For example, at a concentration of 25 ,ug/ml, readily achieved in man (preliminary studies by H. C. Neu), 77% of Klebsiella, 96% of Enterobacter, and 32% of Serratia isolates were inhibited. This is a major increase in activity compared with carbenicillin. However, at that concentration cefazolin would inhibit 95% of Klebsiella and cefamandole would inhibit both 95% of Klebsiella and 54% of Enterobacter. Mezlocillin was twofold more active than carbenicillin against Proteus morganii and P. vulgaris. Against B. fragilis, both azlocillin and mezlocillin were considerably more active than carbenicillin at concentrations of 25 to 50 ,ug/ml, which could be readily achieved in man. Their activity against B. fragilis was similar to
Ticarcillin 45
Azlocillin
75 21 118 22 95 90 50
25
Mezlocillin
50 77 53 55
69
10 66 89 52 60 67
the activity of cefoxitin. Azlocillin was the most active penicillin tested against Pseudomonas. Its activity against Enterobacteriaceae was only slightly better than that of carbenicillin or ticar-
cillin.
It is clear that there is a difference between the MICs and MBCs of these ureido penicillins against both Enterobacteriaceae and Pseudomonas. With Pseudomonas, this difference in MICs was often 64-fold or greater. Furthermore, with large inocula of the Pseudomonas strains tested, both the MICs and MBCs were higher than those usually encountered with carbenicillin or ticarcillin (7, 8). The clinical meaning of these in vitro observations is not established, but may be important for the development of future resistance. The fact that most of the carbenicillin-resistant Pseudomonas isolates were susceptible to azlocillin suggests that azlocillin may be useful in the treatment of Pseudomonas infections. This study also demonstrated that azlocillin and mezlocillin were more stable to f,-lactamase tested in vitro than were penicillin G and ampicillin; the clinical benefit of this stability remains to be seen. It is clear from these studies that azlocillin and mezlocillin are effective new broad-spectrum agents. Their role in clinical medicine can only be evaluated by in vivo studies in animals and man. LITEPlATURE CITED 1. Brown, C. H., E. A. Natelson, W. Bradshaw, T. W. Williams, and C. P. Alfrey. 1974. The hemostatic defect produced by carbenicillin. N. Engl. J. Med. 291:265-270.
2. Buchanan, C. E., and J. L. Strominger. 1976. Altered penicillin-binding components in penicillin-resistant mutants of Bacillus subtilis. Proc. Natl. Acad. Sci. U.S.A. 73:1816-1820. 3. Fu, K. P., and H. C. Neu. 1976. In vitro study of netilmicin compared with other aminoglycosides. Antimicrob. Agents Chemother. 10:526-534. 4. Fu, K. P., and H. C. Neu. 1976. In vitro synergistic effect of netilmicin, a new aminoglycoside antibiotic. Antimicrob. Agents Chemother. 10:511-518. 5. Hoffman, T. A., R. Cestero, and W. E. Bullock. 1970. Pharmacodynamics of carbenicillin in hepatic and renal
938
FU AND NEU
failure. Ann. Intern. Med. 173:173-178. 6. Klastersky, J., B. Vanderkelen, D. Daneau, and M. Mathieu. 1973. Carbenicillin and hypokalemia. Ann. Intern. Med. 78:774-775. 7. Neu, H. C., and G. J. Garvey. 1975. Comparative in vitro activity and clinical pharmacology of ticarcillin and carbenicillin. Antimicrob. Agents Chemother. 8:457-462. 8. Neu, H. C., and E. B. Winshell. 1970. Purification and characterization of penicillinase from Salmonella ty-
ANTIMICROB. AGENTS CHEMOTHER. phimurium and Escherichia coli. Arch. Biochem. Biophys. 139:278-290. 9. Neu, H. C., and E. B. Winshell. 1971. In vitro studies of a semisynthetic penicillin, 6-[D(-)-a-carboxy-3-thienylacetamido] penicillanic acid (BRL 2288), active against Pseudomonas, p. 385-389. Antimicrob. Agents Chemother. 1970. 10. Sykes, R. B., and K. Nordstrom. 1972. Microiodometric determination of J?-lactamase activity. Antimicrob. Agents Chemother. 1:94-99.
0066-4804/78/0013-0930$02.00/0 Copyright © 1978 American Society for Microbiology
Vol. 13, No. 6
Printed in U.S.A.
Azlocillin and Mezlocillin: New Ureido Penicillins KWUNG P. FU AND HAROLD C. NEU*
Departments of Medicine and Pharmacology, College of Physicians and Surgeons, Columbia University, New York, New York 10032
Received for publication 24 June 1977
The activity of azlocillin and mezlocillin, new semisynthetic ureido penicillins, and compared with that of other known /8-lactam antibiotics. At of 25 ,ug/ml, azlocillin inhibited 74% of Enterobacter, 97% of Proteus mirabilis, 64% of Citrobacter, 91% of Pseudomonas aeruginosa, and 82% of Bacteroides strains tested. Mezlocillin inhibited 86% of Shigella, 96% of Enterobacter, 80% of indole-positive Proteus, 88% of Bacteroides, and 63% of Pseudomonas strains tested. Azlocillin was more active against Pseudomonas than was ticarcillin, carbenicillin, or mezlocillin. Mezlocillin was more active than carbenicillin and ampicillin against Escherichia coli, Klebsiella, Enterobacter, Citrobacter, Acinetobacter, Serratia, and Bacteroides. Azlocillin and mezlocillin were less active than cefazolin against ,B-lactamase-producing E. coli and Klebsiella strains but more active than cefazolin against Enterobacter, indole-positive Proteus, Acinetobacter, Citrobacter, and Serratia strains. Both compounds showed activity equivalent to that of cefoxitin against Bacteroides isolates. Both agents were destroyed by many of the ,B-lactamases from gram-negative organisms. was investigated a concentration
Although a number of new penicillins have been introduced in recent years, there is still a need for agents with an enlarged spectrum of activity. The discovery of carbenicillin added Pseudomonas aeruginosa, Enterobacter cloacae, and less commonly encountered species such as Acinetobacter calcoaceticus and the indole-positive Proteus to the spectrum of organisms inhibited by penicillins. Ticarcillin has a two- to fourfold greater activity than carbenicillin against Pseudomonas, but it is no more active than carbenicillin against members of the Enterobacteriaceae (7). Neither carbenicillin nor ticarcillin is active against Klebsiella, and their activity against Serratia is extremely variable. Strains of Escherichia coli possessing TEM-type /8-lactamases are resistant to all of the commercially available penicillins. Furthermore, the serum and tissue levels of carbenicillin and ticarcillin needed to inhibit many hospitalacquired strains of Pseudomonas are high, and hence the doses of the agents used are also high, resulting in a greater incidence of side effects (1, 5, 6). We therefore were interested in evaluating the activities of azlocillin and mezlocillin, new broad-spectrum penicillins, and comparing their activity with that of other 83-lactam antibiotics. (These findings were presented at the 16th Interscience Conference on Antimicrobial Agents and Chemotherapy, 27-29 October 1976,
Chicago, Ill.)
MATERIALS AND METHODS Azlocillin and mezlocillin were obtained from Delbay Pharmaceuticals. Other antibiotics were gifts from their respective manufacturers. Bacterial strains were those isolated from patients hospitalized at the Columbia-Presbyterian Medical Center. The isolates came from blood, sputum, and urine specimens and did not represent the same strains insofar as could be determined by epidemiological means, general antibiograms, and bacteriocin typing. Susceptibility tests. The antimicrobial activity was measured by agar dilution or broth dilution methods as previously described in more detail (3). The minimum inhibitory concentration (MIC) in agar was determined with a replicating device by using a final inoculum on the plate of 105 colony-forming units (CFU). Standard medium used was Mueller-Hinton (BBL). Broth dilution was performed with MuellerHinton broth, with a final inoculum in the assay of 105 CFU. The MIC was the concentration at which there was no visible growth, and the minimum bactericidal concentration (MBC) was the concentration which on subculture yielded fewer than five colonies after 24 h of incubation. ,B-Lactamase preparations and assays. Partially purified 8?-lactamases were prepared by published methods (8). A modification of the Novick microiodometric assay was used (10).
RESULTS The overall activity of azlocilhn against 524 bacterial isolates is shown in Table 1. We have included a large number of isolates that possess 930
VOL. 13, 1978
AZLOCILLIN AND MEZLOCILLIN
931
TABLE 1. Activity of azlocillin against gram-positive and gram-negative organisms No. of Percentage inhibited by an MIC (ug/ml) equivalent to:' Organisn isolates
Staphylococcus aureus ..... 16 Streptococcus faecalis ...... 15 S. pyogenes ................ 30 Haemophilus ........ 10 Neisseria 5 Streptococcus pneumoniae 5 Escherichia coli ........... 61
Shigella
14
-1.6
3.1
69 87 100 100 100 100 3
94 100
16 14
6.3
12.5
25
50
100
200
31
36
49 86 54 90 20
67
65 97 24
94
53 76
88
100
79
82
33 92 77 94
53 94
>400
70
100 100 100 100 100
100
25
30
29
Klebsiella ................. 48 2 4 6 13 40 52 Enterobacter .............. 31 3 15 74 58 84 Serratia 50 2 4 8 18 Salmonella 15 7 47 40 Citrobacter 25 4 12 24 64 68 Proteus, indole positive ..... 28 14 54 68 75 P. mirabilis ............... 31 69 87 94 97 Providencia ............... 15 7 13 20 27 Pseudomonas aeruginosa 95 5 15 29 78 91 Acinetobacter 13 15 23 62 38 Bacteroides ............... 17 29 53 82 88 a MICs were determined by agar dilution with Mueller-Hinton agar (BBL).
,8-lactamases. The E. coli, Salmonella, and Shigella isolates tested are those which would be encountered in a hospital setting rather than in a community, since this agent would only be administered parenterally in the hospital. Of strains of beta-hemolytic streptococci (groups A, B, C, and G) tested, 95% were inhibited by 0.1 jig/ml. Strains of Staphylococcus aureus which lack ,B-lactamases were inhibited by 0.1 to 0.5 jig/ml, and even most ,B-lactamase-producing isolates were inhibited by 3.1 ,ig/ml. Azlocillin inhibited 87% of Streptococcus faecalis strains at a concentration of 1.6 ,ig/ml and 100% at 3.1 jig/ml. All of the small number of Haemophilus influenzae and Neisseria gonorrhoeae isolates, including several f8-lactamases-producing, ampicillin-resistant strains, were inhibited by 0.1 to 0.5 ,ug/ml. Strains of Streptococcus pneumoniae were inhibited by 0.1 to 0.5 ,ug/ml, but penicillin G was twofold more active against these isolates. The activity of azlocillin against Enterobacteriaceae varied from family to family. E. coli strains that are ampicillin susceptible were inhibited by 1.6 to 12.5 jig/ml. A small percentage of isolates was inhibited by 25 to 100 ,ug/ml, but the MICs for most ,B-lactamase-containing strains were 200 jig or more per ml. The activity of azlocilhin against Shigella sonnei and Salmonella was similar to its activity against E. coli and correlated with the presence of 18-lactamases. Azlocillin inhibited 40% of Klebsiella pneumoniae strains at a concentration of 25 jAg/ ml, but the MICs for 35% of isolates were above 200,ug/mI. Seventy-four percent of Enterobacter strains were inhibited by 25 ,ug/ml, and 90% were
400
93
38
100
98 85
100 100 100 100 100
100
inhibited by 100iLg/ml. Serratia, on the other hand, was extremely resistant, with only 18% of strains inhibited by 25 jig/ml. Azlocillin inhibited 75% of indole-positive Proteus strains, mainly P. morganii and P. vulgaris, at 25 jig/ ml, but Proteus rettgeri and Providencia stuartii were much more resistant; for these the MICs were greater than 200 jig/ml. Eighty-two percent of Bacteroides fragilis strains were inhibited by 25,ug/ml. However, the most impressive activity was seen with strains of Pseudomonas aeruginosa: 91% of isolates were inhibited by 25 ,ug/ml, and 78% were inhibited by 12.5 jig/ml. The in vitro activity of mezlocillin against gram-positive cocci and gram-negative bacilli is shown in Table 2. The majority of fB-hemolytic streptococci (groups A and B) as well as Streptococcus pneumoniae were inhibited by less than 0.5 jg/ml. Staphylococcus aureus and S. epidermidis that do not produce fi-lactamase were inhibited by less than 0.5 jig/ml, whereas the MICs for fl-lactamase-producing staphylococci were up to 50 jug/ml. All isolates of ,Bhemolytic streptococci (groups A, B, F, and G) tested were inhibited by 0.05 to 0.1 jig/il. H. influenzae was inhibited by 0.1 to 0.2 ,ug/nil, as was N. gonorrhoeae. The activity of mezlocillin against E. coli was trimodal. Non-hospital-acquired strains that lack 8?-lactamases were inhibited by 0.8 to 6.3 ,ug/ml. These included a small number of E. coli strains that were inhibited by concentrations of 25 to 100,ug/ml. Most ampicillin-resistant strains of E. coli were resistant to mezlocillin; the MICs were 400 jig or more per ml. The activity of
932
FU AND NEU
ANTIMICROB. AGENTS CHEMOTHER.
mezlocillin against Shigella and Salmonella was hibited 63% of Pseudomonas aeruginosa isosimilar to its activity against E. coli, with a lates at a concentration of 25 ,ag/ml. However, bimodal distribution of susceptible and resistant 97% of P. aeruginosa were inhibited by 100 strains. Citrobacter isolates were susceptible to ,ug/ml, and 20% of Klebsiella were not inhibited mezlocillin, with 80% inhibited by 6.3 ,ug/ml. by 400 ,tg/ml. Sixty-five percent of B. fragilis Over 97% of Proteus mirabilis and 70% of indole- were inhibited by 12.5 ,ug/ml, and 100% were positive Proteus, including P. morganii, P. vul- inhibited by 100 ,ug/ml. garis, and P. rettgeri, were inhibited by 1.6 jig Comparison of the MICs determined by the or less per ml. In contrast, Providencia ranged agar plate technique and by the broth dilution in susceptibility from 1.6 ,ug/ml to greater than method yielded similar results with the strains 400 ,ug/ml. Enterobacter cloacae and E. aero- of E. coli, Klebsiella, Enterobacter, Citrobacter, genes were inhibited by 50 ,tg or less per ml, and and Proteus morganii tested; only a single di80% were inhibited by 12.5 ,tg or less per ml; the lution difference was seen with an occasional MICs for 50% of Serratia were greater than 100 isolate. The MICs and MBCs were identical or ytg/ml. The majority of Acinetobacter were in- differed by only one dilution for members of the hibited by 50 ,ug or less per ml. Mezlocillin in- Enterobacteriaceae (Table 3). However, the TABLE 2. Activity of mezlocillin against gram-positive and gram-negative organisms No. of isolates c1.6
Organism
Staphylococcus aureus ........ 18 Streptococcus epidermidis ..... 35 S. faecalis .................... 15 Haemophilus ................. 10 Neisseria gonorrhoeae ........ 5
72 46 100 100 100
Streptococcus, 8-hemolytic
100
Serratia Salmonella Citrobacter
100 25 14 6 23 4 22 24 71 97 13 7 22
30 Streptococcus pneumoniae ..... 10 Escherichia coli .............. 44 14 Shigella Klebsiella .................... 31 Enterobacter ................. 31 49 15 25 28 31 15 76 13 17
Percentage inhibited by an MIC (,ug/ml) of:a 3.1
6.3
12.5
94 74
100 80
83
30 21 23 61 19 47 58 76
34 43 43 72 23 73 79
39 86 56 81 27
25
50
200
400
91
43 71 96 35
48
62 87 85 94
100
52
65
77
100 100
96
100 84
88
100 100 100 100
80 54 63 56 88
100 100
57
83 100 38
>400
100
45 93
88
Proteus, indole positive ........ P. mirabilis .................. Providencia .................. 31 39 46 Pseudomonas aeruginosa ...... 12 17 36 Acinetobacter 31 44 Bacteroides fragilis ........... 29 65 a MIC determined by agar dilution with Mueller-Hinton agar.
100
92 99 100
97
100
TABLE 3. Comparison of MICs and MBCs of aziocillin and mezlocillin
Organism
No. tested
No. of isolates with MICa and MBC identical
No. of isolates with MBC greater than MIC (fold)
2
Ab
4
M M A A A Escherichia coli ...... 6 2 3 4 3 Enterobacter ......... 5 2 2 3 3 Klebsiella ............ 6 2 4 4 1 Proteus morganii ..... 3 1 2 2 1 Citrobacter ........... 2 1 2 1 Serratia .............. 2 2 2 Pseudomonas ........ 10 1 4 1 3 a MIC values were determined by broth dilution method in Mue11er-Hinton broth. b
A, Azlocillin.
C M,
Mezlocilin.
8
216
Me
M
A
M
6
3
1
2
VOL. 13, 1978
MIC for a number of Pseudomonas strains was at least eightfold greater than the MIC of both azlocillin and mezlocillin. Table 4 illustrates the effect of medium used to assay the activity of azlocillin and mezlocillin. The MICs and MBCs varied over a twofold dilution range without a consistent pattern, with brain heart infusion broth and Trypticase soy broth not yielding significantly higher MICs than those obtained in Mueller-Hinton broth. Of interest is the fact that nutrient broth, which has a low conductivity, 1.6 mS compared with 12 mS of Mueller-Hinton broth, and also a low osmolality, 60 mosmol compared with 440 mosmol, did not yield lower MICs. There was some pH effect, as shown by the lower MICs and MBCs in Mueller-Hinton broth at pH 8 compared with those in broth prepared at pH 6. This was noted with both Enterobacteriaceae and Pseudomonas. The effect of the inoculum size on the azlocillin and mezlocillin MICs and MBCs is shown in Table 5. With 103 to 106 CFU, the MICs and MBCs for the Enterobacteriaceae and the Pseudomonas strains tested were similar, with only a twofold increase in the concentrations. However, with 107 CFU of most organisms tested, the MICs rose above 400 Ag/ml. Those Pseudomonas isolates with which there was only a slight rise in the MICs with 107 CFU showed MBCs of >400 ,ug/ml. The presence of 50% normal human serum in the assay increased the MICs required for inhibition of some organisms twofold. MBCs were increased two- to eightfold. For example, the MICs and MBCs of mezlocillin against E. coli 2371 were 6.3 and 12.5 ,g/ml, respectively, in broth and were 12.5 and 50 ,ug/ml, respectively, in 50% serum. In contrast, with Klebsiella 2104 the MICs and MBCs were the same: 3.1 ,tg/ml in broth and 3.1 and 6.3 ,ug/ml in serum. Comparison of the activity of azlocillin and mezlocillin with other ,8-lactam compounds. The activity of azlocillin and of mezlocillin against various gram-negative bacilli was compared with the activitv of ampicillin, carbenicillin, and one of the available cephalosporins (Table 6). Azlocillin and mezlocillin were similar in activity to ampicillin and carbenicilhin against E. coli isolates lacking a fl-lactamase, but mezlocillin and azlocillin were more active at high concentrations, i.e., at 100 to 200 ,ug/ml, than was carbenicillin. Cefazolin was twice as active as either compound against these hospital isolates of E. coli. Mezlocillin was much more active than ampicillin, azlocilHin, or carbenicillin against Klebsiella; 77% of isolates were inhibited by 25 ,ug of mezlocillin per ml, whereas none were inhibited by carbenicillin. Although azlo-
AZLOCILLIN AND MEZLOCILLIN
933
cillin inhibited many of the Klebsiella strains, cefazolin showed much greater overall activity against Klebsiella than did any ofthe penicillins. The activity of azlocillin and of mezlocillin against P. mirabilis was equivalent to that of ampicillin and carbenicillin. Ampicillin was the most active agent; 50% of isolates were inhibited by 0.4 ,ig/ml. Mezlocillin was twofold more active than carbenicillin against Proteus morganii and P. vulgaris and was the most active of the five compounds tested. Mezlocillin and carbenicillin had equivalent activity against Providencia stuartii; 50% of isolates were inhibited by 12.5 ,ug/ml. Azlocillin was twofold less active than mezlocillin or carbenicillin. Mezlocillin was appreciably more active against Enterobacter than was carbenicillin or azlocilhin or even cefamandole; 88% of Enterobacter isolates were inhibited by 12.5 Ag of mezlocillin per ml, compared with 53% inhibited by carbenicillin and 46% inhibited by cefamandole. Mezlocillin was twofold more active than azlocillin and carbenicillin against Citrobacter, but less active than cefamandole. Against Acinetobacter, azlocillin and mezlocillin were similar in activity to carbenicillin, but both were more active than carbenicillin against Serratia. Against B. fragilis, 25-,ug/ml concentrations of antibiotics inhibited the following percentages: ampicillin, 33%; carbenicillin, 28%; cefazolin, 5%; cefoxitin, 83%; azlocillin, 83%; and mezlocillin, 88%. Azlocillin was the most active penicillin tested against Pseudomonas aeurginosa. Twice as many isolates were inhibited by azlocillin as by ticarcillin at a concentration of 6.3 ,tg/ml. Mezlocillin and carbenicillin at low concentrations had equivalent activity against Pseudomonas, but more Pseudomonas isolates were inhibited by mezlocillin in concentrations of 50 to 200 ,ug/ml than by carbenicillin. Comparison of the activity of azlocillin, mezlocillin, ampicillin, and carbenicillin against susceptible and resistant Shigella sonnei showed that mezlocillin was consistently two- to fourfold more active. Comparison of the activity of ampicillin, carbenicillin, and mezlocillin against Salmonella showed that mezlocillin was more active against selected isolates, but there was no consistent pattern of increased activity. Table 7 illustrates the comparative activity of mezlocillin as compared with carbenicillin and ampicillin against various gram-positive organisms. The percentage of Staphylococcus epidermidis strains inhibited by mezlocillin at c1.6 ,ug/ml was twofold greater than that observed with carbenicillin. Mezlocillin was less active against Staphylococcus aureus than was carbenicillin and approximately as active as ampicillin. Against streptococci, ampicillin was the most
934
FU AND NEU
ANTIMICROB. AGENTS CHEMOTHER.
C-4
~
0
CO
0
(D
CQC~~ C~~
~
CO0
CO
LO Lo .6 C
o'~(-~ 8QE 8 qq00o
LoCq
-
-
Lfq
C~~
_6 ci
~c6
vv
L0L
-~~~
CO
C
cI
LoC
-
-V
V
-
0_
-zo
ult~ U:
LO
LO
r-4
C.)
la
6.5 0
E
=
4)
14)
Lc
C6
ciC
l ~6 6ciC l
_6 v V___---cfii e6 C6 V
cqi-cli C- C'V
COqt
C cli
c
cC-q
4
cl
L
Cr 0 LOCO 6C Oc co LO
Ct
V) V c0
V-
L C' C> L
ci
z.4
D
.E
t ci _C' IL6s f C' vs_ MCL
Pcs
8
_ 0 CO
Lo CO
-
*N
4 o_L _ cO
_A
c-
O
6V 6
C
6 C.)1 *,4 0
00
_D Mo Lo Lo
c: o
0
g
Lo cO
0.
0
1-
6 6c'i
e u
'.
-4 c~Lo m __LoL O_ _co 5 ,VVj v Lo N "-
CY
0
41) 4)
.0
.0
CY m m D
N .)
CO LO _O
x
a CO
C COQO O OeO
U)
C)
C
e
.)
0 N
_
0 U)
~C6 Q
Nt
O
.2 cD
L C
OO
0
'QCIQ
at
4) 4)
-4) ,
LOe
CD
LO CSi
O"1 S: CS-4
U)
C14
C CDi C14C
-I Cq ,-4 ->C-4 O C14
CSn4
C.
C.N
Cq
C L LO OO LO CD _ O C : ONLOSO
VV
A
O O
__ A_
_
AA
_ --4
0
o
--
C4
s O Lo
c14
Q
.t46)
0
o oqo cq C-
C
AO AO
-CO q 0e 0~ 0- 0-
L N
0L
lt 0 n
C.)
i
AA A
-
A m
A~ A
A
-
A
LO
A
C.)
4l)
c-i _ o cq
IC .
0
0r
8 8IRr 8lw 8I, 8 8rv88 v v A A A
A
A
A
8v 8"W A
A
A AA
A
A
A
A
A
1-4
H)
A
0 -4
B
mI oo
l
o C1
1
cl
oso Iqo 1
t
m
CSl C.)
._c 4-J0
0 .0
._ ¢9
¢ 04CO 00 00 CO
* C.) ° ¢~ g 0C.) 0 ¢2¢ C.)Qs Cz ¢.) *s *gQ~*g
C.
= c; c; _ Q m cJ =~~1= = c; z cz _>0~0"~0~0" cl N c) sro
c
-0
LO~~~~~~~~0 0 cli
.40
=
-sb e
c
0 .3E -z,~
*
s
CO
r. 0
--4
r-
0
I.. 4
m
b
s
H
~
S SX0 0 C.) X 0 -U6C.
U.)
O .)
0 4)
&
_
r-
UE
S.-4 4.
i)
*~ U
..
0
0
*
~4
VOL. 13, 1978
AZLOCILLIN AND MEZLOCILLIN
935
TABLE 6. Activity of azlocillin and mezlocillin compared with that of known penicillins and cephalosporins Organism (no. of isolates)
Percentage of strains inhibited by an MIC' (pg/ml)' of five penicillins of:
Antibiotic :51.6
3.1
6.3
12.5
25
50
29 15 23 28 29 36 Azlocillin 24 30 38 Mezlocillin 15 25 30 Ampicillin 18 28 30 44 Carbenicillin 18 42 62 88 Cefazolin 80 82 Klebsiella (21) Azlocillin 4 8 20 46 65 Mezlocillin 4 23 46 62 77 Ampicillin 28 67 Carbenicillin 5 Cefazolin 62 81 95 Enterobacter (47) Azlocilin 24 53 68 72 Mezlocillin 16 60 72 88 96 100 2 Ampicillin 4 8 13 19 Carbenicillin 12 20 49 53 60 68 Cefazolin 2 Cefamandole 22 33 41 46 54 61 Proteus mirabilis (31) Azlocilin 96 Mezlocillin 97 Ampicillin 97 Carbenicilin 97 Cefazolin 32 52 72 96 Proteus, indole positive (35) Azlocillin 19 57 64 72 Mezlocillin 67 76 Ampicillin 2 26 Carbenicilin 39 46 58 67 74 Cefazolin 5 14 16 Providencia (15) Azlocllin 7 13 20 27 Mezlocillin 13 31 39 46 54 62 Ampicillin Carbenicillin 29 36 50 Citrobacter (37) Azlocillin 1 3 4 12 30 55 Mezlocillin 4 12 24 64 68 Ampicillin 8 24 38 Carbenicillin 14 22 40 Cefazolin 34 37 39 53 63 79 81 Cefamandole 84 92 94 Acinetobacter (17) Azlocillin 34 50 Mezlocillin 38 62 Ampicillin 9 Carbenicillin 9 36 45 Cefazolin Serratia (23) Azlocillin 5 8 16 29 Mezlocillin 9 5 18 32 Ampicillin 9 13 17 22 26 Carbenicillin 13 17 Cefazolin Pseudomonas aeruginosa Azlocillin 6 19 37 70 82 86 (76) Mezlocillin 6 12 16 35 63 87 24 27 33 Ticarcillin 47 73 87 Carbenicillin 11 14 17 30 38 68 Bacteroides (17) Azlocillin 29 53 83 88 Mezlocillin 29 65 88 94 Ampicillin 33 78 28 Carbenicillin 83 Cefazolin 33 6 72 Cefoxitin 77 83 a MICs were determined by the agar dilution method in Mueller-Hinton agar.
Escherchia coli (50)
100
200
400
>400
48 60 34 48 90
67 76 36
72 82 58
96 80 92 81 38
92
100 100 100 100 100 100
86 52
100 100 100
100 78
92
100
32 68 13 74
47 70 19 82
62 28 89
80
85
55 82 51 53
77 84 81
76
42 23 33
92 61 76 59 51 60 100 63 77 36 55
83 88 67 73 86
55 9
40 47
92 97 99 78 94 100 83 100 72 94
95 100 86 100 89 100 100
57 100 100 78 83 89
100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100
100 100 100
82 85 64 63 64 50 54 34 26 9 100 99
100 100 100 100 100 100 100 100 100 100
95
100
100
100
936
FU AND NEU
ANTIMICROB. AGENTS CHEMOTHER.
active agent. Mezlocillin was the most active agent against Streptococcus faecalis. At a con-
by ,B-lactamases. Partially purified ,B-lactamases of E. coli, Citrobacter, Acinetobacter, Providencia, and Pseudomonas were examined for ability to hydrolyze azlocillin and mezlocillin, and the rates of hydrolysis were compared with
centration of 0.8 ,ug or less per ml, it inhibited all the isolates. The activity of azlocillin and mezlocillin against some 8)-lactamase-producing organisms is shown in Table 8. Azlocillin and mezlocillin were more active than ampicillin and carbenicillin against 8-lactamase-producing Shigella, Klebsiella, Enterobacter, Serratia, and Acinetobacter. However, cefoxitin and cefazolin were the most active against the /i-lactamase-producing strains of E. coli, Shigella, and Providencia tested. Twelve Pseudomonas aeruginosa isolates resistant to carbenicillin were tested for their susceptibility to azlocillin, mezlocillin, and ticarcillin (Table 9). Azlocilhin was more active than ticarcillin against all but one of the carbenicillinresistant Pseudomonas isolates tested, whereas mezlocillin was more active than ticarcillin against some strains. Hydrolysis of azlocillin and mezlocillin
TABLE 9. Activity of azlocillin and mezlocillin against 10 carbenicillin-resistant strains of Pseudomonas aeruginosa MIC (ug/ml) of: Strain no.
Carbenicillin
Azlo-
cillin
2800 2800
2131 2137 2699 2711 2960 3330 3331 3467 3423 3408
Mezlocillin
Ticarcillin
.-800 100 100 25 25 50 100 25 200 200
2800 2800
2800 200 25 12.5 100
200 200
2:800 200 200 2800 400 400
12.5 12.5 50 25 50
100 100
2800 100 50
2800 200 400
TABLE 7. Comparative activity of mezlocillin and carbenicillin against some gram-positive organisms Antibiotic
Organism (no. of isOlate8)
s0.8
Staphylococcus epi- Carbenicillin dermidis (35) Mezlocilhin S. aureus (30) CarbeniciHlin MezlociUin AmpiciNlin Streptococcus fae- Carbenicillin calis (31) Mezlocillin AmpiciLin Beta-hemolytic Carbenicillin streptococci (32) Mezlocillin
17 31 33 30 60 3 100 97 91 97 100
Ampicillin
Percentage of strains inhibited by an MIC (yg/ml) of: 3.1 6.3 12.5 25 50 100 200 400 51 29 43 91 80 86 46 74 80 83 91 63 83 100 43 73 83 93 97 100 73 77 87 90 97 93 100 6 65 100
1.6
100 94
97 100
2800 100 100
100
TABLE 8. Comparative activity of aziocillin and mezlocillin against JI-lactamase-producing organisms MIC (Ag/nm) of: Organism
Escherichia coli 3083 E. coli 3138 Shigella sonnei 12-31 Klebsiella 2762 Klebsiella 3001 Enterobacter 2249 E. cloacae Serratia 2788 Serratia 2773 Providencia 2395 Pseudomonas 2983 Acinetobacter 30 Proteus morganii 1635 P. rettgeri Pseudomonas 3647
Azlocllin
Mezlocllin
2800
12.5 2800
6.3 6.3 12.5 25 12.5 25 100
2800 12.5 200
50 25 100 3.1 25 12.5 100 400 25 200 50
2800 50
25
Amnpi cillin
Carbenicllin
400
50
2800 2800
2800
100
400 200
2800 100
2800 25
400 200
2800 2800 2800 2800 2800 2800 2800
2:800
12.5 2800 2800 25 2800 400 2800 2800
Cefazolin
Cefoxitin
3.1 25 1.6
3.1 1.6 1.6 100 12.5
2800 6.3
2800 2800 2800 2800 2800 2800 2800 800
2800 2800
2800 2800 25 100 0.8
2800 2800 800 3.1
2800
VOL. 13, 1978
AZLOCILLIN AND MEZLOCILLIN
937
TABLE 10. Comparative ,3-lactamase stability of azlocillin and mezlocillin Relative rate of hydrolysisa
Organism
Ampicillin
Carbenicillin
Escherichia coli 3337 43 95 Citrobacter 2732 250 42 Citrobacter 2836 28 30 Enterobacter 2647 180 62 Pseudomonas aeruginosa 3901 76 26 Proteus mirabilis 3378 84 71 Acinetobacter 2284 107 110 Providencia 2395 66 53 aActivity relative to the rate of hydrolysis of penicillin G.
those of penicillin G, ampicillin, carbenicillin, and ticarcillin (Table 10). Both compounds were hydrolyzed at rates comparable to the rates for ticarcillin, although there were selected differences. Protein binding. The binding of mezlocillin to human protein was determined at several concentrations of antibiotic by the agar diffusion assay. Protein binding was 20% at both low (10 ,ug/ml) and high (100 ,g/ml) concentrations. Azlocillin was less protein bound (12% by the agar diffusion assay).
DISCUSSION Azlocillin and mezlocillin were shown in this study to have a broad spectrum of antimicrobial activity against both gram-positive and gramnegative organisms. The in vitro activity of these compounds against streptococci and staphylococci is comparable to or greater than that reported earlier for carbenicillin and ticarcillin (7, 8). The activity of both compounds against Enterobacteriaceae varied with the isolates studied. Azlocillin and mezlocillin were slightly more active than carbenicillin and ticarcilhin against E. coli, Salmonella, Shigella, and Citrobacter strains lacking f8-lactamases. It is against the family Klebsielleae that mezlocillin has extended the range of activity of this type of compound. For example, at a concentration of 25 ,ug/ml, readily achieved in man (preliminary studies by H. C. Neu), 77% of Klebsiella, 96% of Enterobacter, and 32% of Serratia isolates were inhibited. This is a major increase in activity compared with carbenicillin. However, at that concentration cefazolin would inhibit 95% of Klebsiella and cefamandole would inhibit both 95% of Klebsiella and 54% of Enterobacter. Mezlocillin was twofold more active than carbenicillin against Proteus morganii and P. vulgaris. Against B. fragilis, both azlocillin and mezlocillin were considerably more active than carbenicillin at concentrations of 25 to 50 ,ug/ml, which could be readily achieved in man. Their activity against B. fragilis was similar to
Ticarcillin 45
Azlocillin
75 21 118 22 95 90 50
25
Mezlocillin
50 77 53 55
69
10 66 89 52 60 67
the activity of cefoxitin. Azlocillin was the most active penicillin tested against Pseudomonas. Its activity against Enterobacteriaceae was only slightly better than that of carbenicillin or ticar-
cillin.
It is clear that there is a difference between the MICs and MBCs of these ureido penicillins against both Enterobacteriaceae and Pseudomonas. With Pseudomonas, this difference in MICs was often 64-fold or greater. Furthermore, with large inocula of the Pseudomonas strains tested, both the MICs and MBCs were higher than those usually encountered with carbenicillin or ticarcillin (7, 8). The clinical meaning of these in vitro observations is not established, but may be important for the development of future resistance. The fact that most of the carbenicillin-resistant Pseudomonas isolates were susceptible to azlocillin suggests that azlocillin may be useful in the treatment of Pseudomonas infections. This study also demonstrated that azlocillin and mezlocillin were more stable to f,-lactamase tested in vitro than were penicillin G and ampicillin; the clinical benefit of this stability remains to be seen. It is clear from these studies that azlocillin and mezlocillin are effective new broad-spectrum agents. Their role in clinical medicine can only be evaluated by in vivo studies in animals and man. LITEPlATURE CITED 1. Brown, C. H., E. A. Natelson, W. Bradshaw, T. W. Williams, and C. P. Alfrey. 1974. The hemostatic defect produced by carbenicillin. N. Engl. J. Med. 291:265-270.
2. Buchanan, C. E., and J. L. Strominger. 1976. Altered penicillin-binding components in penicillin-resistant mutants of Bacillus subtilis. Proc. Natl. Acad. Sci. U.S.A. 73:1816-1820. 3. Fu, K. P., and H. C. Neu. 1976. In vitro study of netilmicin compared with other aminoglycosides. Antimicrob. Agents Chemother. 10:526-534. 4. Fu, K. P., and H. C. Neu. 1976. In vitro synergistic effect of netilmicin, a new aminoglycoside antibiotic. Antimicrob. Agents Chemother. 10:511-518. 5. Hoffman, T. A., R. Cestero, and W. E. Bullock. 1970. Pharmacodynamics of carbenicillin in hepatic and renal
938
FU AND NEU
failure. Ann. Intern. Med. 173:173-178. 6. Klastersky, J., B. Vanderkelen, D. Daneau, and M. Mathieu. 1973. Carbenicillin and hypokalemia. Ann. Intern. Med. 78:774-775. 7. Neu, H. C., and G. J. Garvey. 1975. Comparative in vitro activity and clinical pharmacology of ticarcillin and carbenicillin. Antimicrob. Agents Chemother. 8:457-462. 8. Neu, H. C., and E. B. Winshell. 1970. Purification and characterization of penicillinase from Salmonella ty-
ANTIMICROB. AGENTS CHEMOTHER. phimurium and Escherichia coli. Arch. Biochem. Biophys. 139:278-290. 9. Neu, H. C., and E. B. Winshell. 1971. In vitro studies of a semisynthetic penicillin, 6-[D(-)-a-carboxy-3-thienylacetamido] penicillanic acid (BRL 2288), active against Pseudomonas, p. 385-389. Antimicrob. Agents Chemother. 1970. 10. Sykes, R. B., and K. Nordstrom. 1972. Microiodometric determination of J?-lactamase activity. Antimicrob. Agents Chemother. 1:94-99.
