1060
Journal ofkledicinal Chemistry, 1973, Vol. 16, No. 9
iVotes
Table I. N-Substituted Erythromycylamines 1 (R, = H) Minimum inhibitory concentration, pg/ml
Staph aureus R2
MP, "C
125-1 27 2" 181-183 C,H,CH,NH 2,4,6-Me3C,H,CH,NH 158-160 4-N02C,&CH, NH Amorphous Me,CHNH 136-1 39 146-150 Cyclohexyl NH 117-1 20 NCCH,CH,NH C,H,COCH,CH,NH 125-128 MeO,CCH,CH,NH 115-117 HOCH,CH,CH,NH 115-118 MeCH(0H)CH F H JVH 136-138 162-167 MeCONH 201-202 EtCONH C,H,CONH 153-160 C,H,SO,NH 134-1 35 MeNHCONH 166-170 C,H,NHCONH 196-200 C,H ,NHC SNH 174-178 EtNHCSNH 148-155 Erythromycin (for comparison)
Analyses C, H, N, 0 C, H, N, 0 C, H, N, 0 C, H, N, 0 C, H, N, 0 C, H, N , 0 C, H, N, 0 C, H, N, 0 C, H, N, 0 C, H. N, 0 C, H, N, 0 C, H, N, 0 C, H, N, 0 C, H, N, 0 , S C, H, N, 0 C, H, N, 0 C, H, N, 0 , S C, H, N, 0 , S
671 8Q
u12sb
os os
05
0 25 0 25 2 4 2 0 25
0 25 0 25
10
4 10 0 25
-7
10
2
os 32 25 6 >s12 4 >512 >s12 64 32 05
2 10 32 25 6 25 6 8 >512 >SI2 32 32 (1 5
1292OC
E coh AH 3
128 >512 64 64 128 128 25 6 64 25 6 64 64 >SI 2 >s12 >SI2 >512 >s12 >s12 >512 >512 >512
128 >512 25 6 512 512 256 25 6 64 2.5 6 51 2 64 > 512 > 512 > 512 > 512 > 512 512 > 512 > 512 > 512
Pro tells morgam D1
25 6 >SI2 25 6 >512 512 512 512 >512 2512 >512 15 6 >512 >512 >512
>SI2 X I 2 >5ll >512 >512 >512
aPenicillm and erythromycin sensitive bPenicillin resistant erythromycin sensitive CPenicillm and erythromycin resistant
with CH,CI2. The dried (MgS04) extracts were evaporated to an oil which was crystallized and recrystallized from aqueous EtOH giving 2.6 g (44%), mp 136-1 39", pure by tlc. The ir spectrum showed no C=N stretch. Anal. (C4,H11N20,2)C, H, N. N-Cyclohexylerythromycylaminewas prepared similarly. N-(2-Benzoylethyl)erythromycylamine. A solution of erythromycylamine (5.0 g, 6.8 mmol) and phenyl vinyl ketone (1.0 g, 7.6 mmol) in MeOH (40 ml) was heated under reflux for 30 min. TIC indicated complete conversion to a faster running compound. The solvent was evaporated and the resulting oil crystallized and recrystallized from CH2C12-Et20 giving 2.7 g (46%), mp 125-128", pure by tlc: ir (KBr) 1703 cm-' (CO). Anal. (C,,H7,N,0,,) C, H, N. N-(2-Carbomethoxyethyl)-and N-(2-cyanoethyl)erythromycylamine were prepared similarly from methyl acrylate and acrylonitrile, respectively. N-(3-Hydroxybutyl)erythromycylamine, To a solution of erythromycylamine (5.0 g, 6.8 mmol) in MeOH (30 ml) was added methylvinyl ketone (0.53 g, 7.5 mmol). After 1 hr, tlc showed virtually complete conversion to a faster running compound. Without isolation of this product, sodium borohydride (0.5 g, excess) was added in portions during 9 0 min. After a further 30 min, the bulk of the solvent was evaporated, water added, and the solution extracted three times with CH,Cl,. The dried (MgSO,) extracts were evaporated to an oil which was crystallized and recrystallized from Et,O giving 2.9 g (53%), mp 136-138", pure by tlc. Anal. (C,iH78,Oid C, H, N. N-( 3-Hydroxypropy1)erythromycylamine was similarly prepared from acrolein. N-Acetylerythromycylamine. A solution of erythromycylamine (5.0 g, 6.8 mmol) and Ac20 (0.76 g, 7.5 mmol) in MeOH (30 ml) was kept at room temperature for 10 min. Tlc showed complete conversion t o a faster running compound. Et,O was added and the solution washed with water after adjusting pH to 11 with 2 N NaOH. The ether layer was dried (MgSO,) and eventually deposited Nacetylerythromycylamine (1.3 g, 25%), mp 162-167", which was pure by tlc: ir (KBr) 1660, 1520 cm-I (CONH). Anal. (C,,H7zNzOi,) C, H, N. N-Ropionyl-, N-benzoyl-, and N- phenylsulfonylerythromycylamine were prepared similarly using (EtCO) ,O, BzCl, and PhSO,CI, respectively. N,N '-Erythromycylmethylurea. To a solution of erythromycylamine (5.0 g, 6.8 mmol) in CH,CI, was added methyl isocyanate (0.5 g, 8.8 mmol). After 10 min tlc showed complete conversion to a faster running compound. Et,O was added giving crystals (4.0 g) and recrystallized from EtOH-Et,O giving 3.2 g (59.5%), mp 166170", pure by tlc: ir (KBr) 1640 cm" (urea). Anal. (C,,H7,N,0,,) C, H, N. N,N'-Erythromycylphenylurea,ethylthiourea, and -phenylthiourea were prepared similarly by reaction with PhNCO, EtNCS, and PhNCS, respectively.
Acknowledgments. The authors wish to thank their colleagues at Eli Lilly & Co., Indianapolis, Ind., especially Drs. K. Gerzon and E. H. Massey, for making available the results of their early experiments with erythromycylamine. References (1) J. M. McGuire, R. L. Bunch, R. C. Andersen, H. E. Boaz, E. H. Flynn, H. M. Powell, and J . W. Smith, Antibiot. Chemother., 2,-218 (1952). ( 2 ) S. Diokic and 2 . Tamburdsev. Tetrahedron Lett., 1645 (1967). (3) M. $. Sigal, Jr., P. F. Wiley, K. Gerzon, E. H. Flynn, U. C. Quarck, and 0. Weaver, J. Amer. Chem. Soc., 78, 388 (1956). (4) E. H. Massey, B. Kitchell, L. D. Martin, K. Gerzon, and H. W. Murphy, Tetrahedron Lett., 157 (1970). (5) E. H. Massey and B. S. Kitchell, U.S. Patent 3,652,537 (1972); Chem. Abstr., 76, 140576m (1972). (6) G. H. Timms and E. Wildsmith, Tetrahedron Lett., 195 (1972); (7) E. Wildsmith, ibid., 29 (1972). (8) K. Gerzon and B. Kitchell, U. S. Patent 3,681,322 (1972); Chem. Abstr., 17,114272k (1972). (9) A. F. Cockerill, M. F. Ellis, D. M. Rackham, and E. Wildsmith, J. Chem. Soc., Perkin Trans. 2, 1973 (1973). (10) W. Szybalski, Science, 116, 46 (1952).
Antimicrobial Action of Isomeric Fatty Acids on Group A Streptococcus Jon J. Kabara,* Anthony J. Conley, Dennis M. Swieczkowski, Department of Biomechanics, College of Osteopathic Medicine, Michigan State University, East Lansing, Michigan 48823
I. A. Ismail, M. Lie Ken Jie, and Frank D. Gunstone Chemistry Department, University of St. Andrews, St. Andrews, Scotland. Received February 16, 19 73
Previous workers have shown the importance of unsaturation to the germicidal action of fatty acids on gram-positive micr~organisms.'-~The cis form of the unsaturated longchain fatty acid (oleic acid) is more active than the saturated compound (stearic acid), and toxicity against organisms increased with increase in the number of double bonds.5
Notes
Journal ofhiedicinal Chemistry, 1973, Vol. 16, No. 9
While these generalizations are substantially true, little is known concerning the effect that the position and kind of unsaturation has on fatty acid antimicrobial action. Recently, Jenkin, et a l , have presented patterns of utilization of isomeric octadecenoic acids in Leptospira6 and in monkey kidney cells.7 The purpose of the present report was to determine whether generalizations made for Leptospira were applicable to another organism, to evaluate the biological activity of the unsaturated dodecyl (short-chain) vs. octadecyl (long-chain) series, and to contrast the effectiveness of ethylenic vs. acetylenic unsaturation on biological activity.
Experimental Section Long-chain esters used in this study were synthesized and purified as previously r e p ~ r t e d . Because ~?~ previous indicated that the methyl esters of fatty acids were inactive, the synthesized esters were converted to soaps by mild saponification for 24 hr at room temperature. In order to measure any adverse effects or incompleteness of the saponification procedure, methyl laurate, methyl oleate, and linoleic acid were used as control compounds. Methyl esters (10 mg)were dissolved in approximately 0.25 ml of methanol and an equal volume of methanolic 0.5 M KOH was added. After standing for 24 hr at room temperature, 10 ml of sterile Trypticase soy broth (BBL) was added with vigorous stirring to each tube (pH 8.3). The same broth mixture was used to prepare dilutions of the fatty acid salts. Solutions containing various amounts of lipid were inoculated with 0.05 ml of an 18-hr broth culture (10' organisms) of a virulent strain of group A Streptococcus (isolated at Providence Hospital, Southfield, Mich. 1970). Lipid concentrations after the second dilution (pH 7 . 6 ) had no appreciable effect on the pH of the media. The minimal inhibitory concentration (MIC) in pmol/ml for each fatty acid was determined after an overnight incubation of the lipid and microorganism at 35" in a 5% CO, atmosphere. The MIC was determined by observing the highest dilution of the soap which inhibited growth when compared to an uninoculated chemical control tube.
-
6-
0
4 -
32,
I 0.8
-E
-, 0
0.6
---
0.4
-
0.30.2 -
i
--0.08 0.1
-
6-
o
ocetylenic ethylenic
4 -
32-
1 1 1 1 1 1 1 1 1 1 1 2 4 6 8 IO 12 14 I6 18 Bond Position Figure 1. The MIC values for unsaturated lauric acid derivatives are presented. All compounds are less toxic than lauric acid (MIC = 0.1 2 pmol/ml).
1061
Results The control methyl esters (12:O and 18:l) used to check the effect of the saponification procedure indicated complete hydrolysis of the ester since the formed soap and the free acid tested in the same manner had identical MIC values. Linoleic acid also was processed by the same method. No decomposition could be detected since chromatographing the treated and untreated fatty acid resulted in identical thin-layer patterns. Individual fatty acids, other than those used as model compounds, may be more or less susceptible to the saponification procedure. It was not possible because of limited quantities of material available to examine each test fatty acid as in the case of the control compounds. The limitation of test material also prevented us from screening the fatty acids against other microorganisms or other strains. However, in a previous study" it has been shown that fatty acids or their derivatives active against one strain retain a similar order of activity against a variety (four to six) of strains of that species. Thus, while the activity measured may have general application, the possibility of resistance among strains is always present. With these qualifications in mind, the following results on 12: 1 and 18: 1 ethylenic (cis) and acetylenic isomers are presented. Data for the 12: 1 isomers are presented in Figure 1. For the small number of acids compared, the dodecanoic acids were generally more toxic to the group A Streptococcus than the corresponding acetylenic isomers. The greatest difference between the two types of unsaturation is found with the A' isomer. For both types of unsaturation, the A'' isomer was the most toxic Clz acid, i.e., lowest MIC value. By comparison, the 18: 1 series of fatty acid salts were more toxic to group A Streptococcus than representatives of the 12:l family. These data, divided into two homologous series of isomers, odd vs. even carbon chain length, are presented in Figure 2. For each series the ethylenic and acetylenic are plotted together. The octadecenoates indicate less toxicity than acetylenic isomers. However, since the data for A6, A7, A", A", A1', A14, and A17 isomers represent only one tube difference in dilution, it is not considered significant. While the differences between ethylenic and acetylenic compounds at positions A4 and A1' are significant, they are of a reverse order. Consequently, little can be made of the difference in kind of unsaturation present. Peak inhibitory activity for both types of linkages is noted when unsaturation was located in the A' or A8 positions. The effect of a seccnd double bond and its positional isomers was examined by studying a number of 18:2 fatty acids. These lipids are more toxic than the 18:l series. For comparative purposes, individual monoenoic isomers (two) which have unsaturation in similar positions to the dienoic molecule are compared to the dienoic acid; i.e., the ASJz 18:2 is compared to the A9 18: 1 and the A1' 18:1. The differences are listed (Table I). In such a comparison, the activity of the 18:2 is always greater than the component 18: 1 isomers. Usually the more unsaturated fatty acids are 8-1 5-fold more toxic. Contrasted to the results found for the monoenoic fatty acids, the position of the second double bond in the six 18:2 isomers tested did not seem important in terms of enhancing antimicrobial activity.
Discussion In these experiments we have confirmed an earlier conclusion that the addition of one or two cis double bonds to stearic acid increases the toxicity to various microorgan-
1062
Journal of Medicinal Chemistry, 1973, Vol. 16, No. 9
Notes
EVEN 31
2t
i
a elhylene o ocelylene
2
4
6
Bond
8
IO
12
14
16
18
2
4
6
Bond
Position
8
IO
12
I4
I6
I8
Position
Figure 2. The MIC values for unsaturated C,, acids. In general, and opposite to shorter chain fatty acids, these unsaturated acids have lower MIC values than C 18:O (0 > 3.52 pmol/ml).
Table I. Comparison of Ethylenic 18:2 lis. 1 8 : l MIC's against G r o w A Strevtococcus Dienoic
Monoenoic
1\9"2 AI*
A9 AIZ A8 A7,12
AI2 A' A6312
A12 A6 ~ 5 4 2
A12
A5 A6912
A6 AI2 A6,ll
A6
AI' A640
A6 AI0
MIC, pmol/ml
0.01 0.18 0.18 0.02 0.18 0.04 0.02 0.18 0.09 0.01 0.18 0.09 0.02 0.18 1.77 0.01 0.09 0.18 0.02 0.09 0.18 0.01 0.09 0.09
Difference 0.17 0.17 0.16 0.02 0.16
0.07 0.17 0.08 0.16 1.75 0.08 0.17 0.07 0.16 0.08 0.08
Contrary to this accepted generalization for longchain fatty acids, unsaturation in a lower chain fatty acid does not always lead to greater activity. Lauric acid has a lower MIC (0.12 pmol/ml) than the most inhibitory ethylenic or acetylenic isomer (0.25 pmol/ml). Apparently, a critical structure of property of the fatty acid is reached with twelve carbon saturated and eighteen carbon unsaturated fatty acids. Compounds with greater (>18:0) or lesser (
