783
J. Med. Chem. 1989,32, 783-788 were dried, and the radioactivity bound to the filters was measured by liquid scintillation spectrometry. Specific ['H]-&OH-DPAT binding was defined as the difference between binding in the abfience and Presence of 10 PM 5-HT. 1% and s l o p values from the competition assays were determined by nonlinear regression analysis by using the program PCNONLIN and the four-parameter logistic function described bv De Lean et al.18 I
Acknowledgment. The financial support from the Medical Research (14'-502), the (18) De Lean, A.; Muson, P. J.; Rodbard, D. Am. J. Physiol. 1978, 255, E97-E102.
Natural Science Research Council, and the U.S. National Institutes of Health (Grant Nos. NS16605 and NS01009) is gratefully acknowledged. We thank Ciba-Geigy Ltd., V&tra Frolunda, Sweden, for a generous gift of reserpine, Georgina Lambert and yu Hang for skillful assistance, and Dr. Urban Hoglund for help with the sta. . tlstlcs. Supplementary Material Available: Effects of the enantiomers of 1-4 on the NSD 1015 induced accumdation of & H m (brain stem, hemispheres) and DOPA (striatum, limbic system, brain stem, hemispheres) in rata (2 pages). Ordering information is given on any current masthead page.
Synthesis and Biological Evaluation of De(acetylglucosaminy1)didehydrodeoxy Derivatives of Teicoplanin Antibiotics Adriano Malabarba,* Aldo Trani, Giorgio Tarzia, Pietro Ferrari, Rosa Pallanza, and Marisa Berti Merrell Dow Research Institute, Lepetit Research Center, Via R. Lepetit 34, 21040, Gerenzano (Varese), Italy. Received July 5, 1988 A series of 34-de(acetylglucosaminyl)-34-deoxyderivatives of 3435- and 35,52-didehydro teicoplanin antibiotics have been synthesized from teicoplanin and its N-acetylglucosamine containing pseudoaglycons under basic conditions. The structures of these compounds have been determined by 'H NMR, UV, and FAB-MS. 35,524Jnsaturated derivatives maintained in vitro and in vivo antimicrobial activity to a different extent as well as the ability for binding to Ac+Lys-D-Ala-D-Ala, a bacterial cell-wall model for the site of action of glycopeptide antibiotics. In contrast, 34,3&unsaturated compounds were markedly less active and possessed a negligible affiiity for the synthetic tripeptide.
Teicoplaninl is a glycopeptide antibiotic produced by Actinoplanes teichomyceticus ATCC 31121.2 It belongs to the vancomycin-risketin family and it is active against Gram-positive b a ~ t e r i a . ~ Teicoplanin is a complex consisting (Figure 1)4of five major closely related factors (CTA) differing in the N-acyl aliphatic chain linked with 8-D-glucosamine at position 56 and of one pseudoaglycon (TB) deriving from CTA by loss of the above N-acylated amino sugar. Both CTA and TB contain one cy-Dmannose and one N-acetyl-8-D-glucosamineat the 42- and 34-positions? respectively, which are removed in that order by selective acidic hydrolysis, thus obtaining a second pseudoaglycon (TC) and the aglycon ETD).697 In the course of structural studies of teicoplanin, basic transformation products were also investigated. In particular, reaction with aqueous bicarbonate or methanolic amines resulted in the epimerization at C-3 (epiteicoplanins).* The relationship between the species involved in the acidic and basic treatments is outlined in Scheme I. Teicoplanin is the recommended INN of teichomycin. (a) parenti, Fa;Beretb, G.; Berti, M.; Arioli, v. J. Antibiot. 1978,31, 276-283. (b) Bardone, M. R.;Paternoster, M.; coronelli, C. Zbid. 1978.31, 170-177. (a) Pallanza, R.; Berti, M.; Goldstein, B. P.; Mapelli, E.; Randisi, E.; Arioli, V. J. Antimicrob. Chemother. 1983,11,419-425. (b) Somma, S.; Gastaldo, L.; Corti, A. Antimicrob. Agents Chemother. 1984,26,917-923. Teicoplanin and its acidic hydrolysis products were formerly named teichomycin Az complex, i.e. T-A2-1 to T-A2-5 (CTA) and T-A3-1 (TB), and T-A3-2 (TC), and T-aglycon (TD). Barna, J. C. J.; Williams, D. H.; Stone, D. J. M.; Leung, T.-W. 1984,106,4895-4902. C.; Doddrell, D. M. J. Am. Chem. SOC. Malabarba, A.; Strazzolini, P.; DePaoli, A.; Landi, M.; Berti, M.; Cavalleri, B. J. Antibiot. 1984, 37, 988-999. Malabarba, A.; Ferrari, P.; Gallo, G. G.; Kettenring, J.; Cavalleri, B. J. Antibiot. 1986, 39, 1430-1442. Barna, J. C. J.; Williams, D. H.; Strazzolini, P.; Malabarba, A.; Leung, T.-W. C. J. Antibiot. 1984,37, 1204-1208.
Scheme I base 2
* epi-CTA
CTA
b epi-TB
TB
1 J
I
acid 2
baoe I or 2
acid 2
b epi-TC
TC
1
ncid 3
acid 3
baae 2
+ epi-TD
TD
bane I- 32 aq. NaHC03; base 2- 20% methanolic n-C4HgNH2. a c i d I - 90% aq. TPA; a c l d 2- HCl (DHE);
a c i d 3- HC1 (TPE).
The glycosidic linkages of carbohydrates are generally hydrolyzed under acidic conditions but are relatively stable in diluted alkali? with the exception of a-D-mannOSyl phenyl glycosides.10 Sometimes, sugars attached to the hydroxyl group of serine can be removed by alkaline treatment through a 8-elimination mechanism." The 34-(N-acetyl-~-~-glucosamine) of teicoplanin is, in principle, amenable to undergo @-eliminationon treatment with alkalies to produce new pseudoaglycons or aglycon. (9) Overend, W. G. In Chemistry and Biochemistry of Carbohydrates; Academic Press: New York, 1972; Vol. lA, p 332. (10) K y d a , S.; Murata, S.; Tanaka, M. Chem. Phurm. Bull. 1983, 31,3902-3905. (11) Neurberger, A.; Gottschalk, A.; Marshall, R. D.; Spiro, G. In Glycoproteins; Their Composition, Structure, and Function, 2nd ed.;Gottschalk, A., Ed.; Elsevier: Amsterdam, 1972;p 450.
0022-26231891 1832-0783$01.5O/O 0 1989 American Chemical Society
Malabarba et al.
784 Journal of Medicinal Chemistry, 1989, Vol. 32, No. 4
Scheme I1 acid 1
1
@
ase 3
acid 2
2
base 3
3
base 3
5
/
base 4
acid I
CTA
b
acid 2
TB
w
acid 3
TC
W T D
base 3. methanolic KOH (DHF/DHSO. 3 / 2 ) ; base 6- 2% methsnolic NaOCH3 (DIIP); base 5- 0.1%NaHC03 (HeOH/H20, l / l ) . a c i d I = 90% aq. TFA; acid 2- H 2SO4 or HC1 (DHE or THF); acid 3- HC1 (TPB); acid 6- I N HC1.
a
OR,
R2
OH
b
OR1
1.
R 1 = N-(CIO-Cll)acyl- p -D-glucoraminyl;
2
R1
2
R 1 = Rt
-
R 2 = a-D-mannoryl.
8 ; R 2 = a-D-unnoryl.
-
H.
-0
-6 CTA
R
- -
0 -~-gl~co.aminyl;
~-.ectpi-
aminyl: R2
TB
R
TC
R
TD
R
-
R!
a-D-mannosyl.
N - ~ c e r y l - ~ - D - g l u c o ~ ~ m i " yRl ,; D - D ~ h c o B a m i n y l ; R,
N-acetyl-
R,
R2
I
-
OH N-(C
10 -c I I )acyl-
- - H; R 2 R2
-5
Figure 2. The structures of unsaturated derivatives. n
-~-g~ucos-
@-D-mannoayl.
H.
H.
Figure 1. The structures of teicoplanin A2 (CTA) and its pseudoaglycons T-A3-1 (TB), T-A3-2 (TC), and aglycon (TD). (a) IUPAC nomenclature. (b) Nomenclature adopted by Williams e t a1.697
Therefore, we decided to investigate this possibility by submitting teicoplanin antibiotics to basic treatment under conditions different from those leading to C-3 epimerization.
C h e m i s t r y . Teicoplanin A, (CTA) and its pseudoaglycon TB were respectively transformed (Scheme 11, Figure 2) into the corresponding 35,52-unsaturated derivatives 1 and 2 at room temperature in the presence of 5% methanolic KOH in DMF/DMSO (3/2). In contrast, the removal of N-acetyl-@-Dglucosamine from pseudoaglycon TC under the above conditions produced 34,35-unsaturated compound 4 through an intermediate isomer 5. The formation of 5 and its subsequent transformation into 4 was observed while the course of the reaction was monitored by HPLC. Compound 5 was the main product when sugar elimination was carried out with 2% methanolic NaOMe in DMF/DMSO (4/1). Unsaturated aglycons 4 and 5 differ in the structure of the peptidic linkage at position
Journal of Medicinal Chemistry, 1989, Vol. 32, No. 4 785
Derivatives of Teicoplanin Antibiotics Table I compd 1
yield, % 81
HPLC: t R , min d
potentiometry? PKMCS (EW) 5.0, 6.9 (935)
FAB-MS 1680 (M + Na)
formula (MW)
anal.*
(1656.5, 1658.5, 1672.5) C~H&l~N7022*HCl C, H, C1, N (1343.1 + 36.5) 3 80 10.06 4.7, 7.0 (670) 1180 CMH&l2N701,'HCl C, H, C1, N (1180.9 + 36.5) C, H, C1, N 4 41 11.18 4.8, 7.1 (650) 1180 CMH43C12N7017 (1180.9) 5 9 12.24 4.9, 6.9 (655) 1180 C68H4SC12N7017 C, H, C1, N (1180.9) OEquivalent weighta (EW) were corrected for solvent content and inorganic residue. Data for CTA were 5.0, 7.1 (965); for TD, 4.9, 6.9 (704). bAnalyticalresults were within 0.4% of the theoretical values. Inorganic residue, determined in O2 atmosphere at 900 OC, was always
