Dihydropyrimidinones--a new class of anti-staphylococcal antibiotics.

Article abstract of DOI:10.1016/S0960-894X(02)00880-6

We report the synthesis and pharmacological evaluation of new derivatives of natural dipeptide antibiotic TAN-1057 A, B. In the course of this program, we identified novel analogues of the natural product that display similar antibacterial activity and showed improved tolerability.

Full text of DOI:10.1016/S0960-894X(02)00880-6

Bioorganic & Medicinal ChemistryLetters 13 (2003) 241–245
Dihydropyrimidinones—A New Class of
Anti-Staphylococcal Antibiotics
Michael Brands, Rainer Endermann, Reinhold Gahlmann,
Jochen Kruger* and Siegfried Raddatz
BAYER AG, Business Group Pharma, Research, D-42096 Wuppertal, Germany
Received 23 July2002; revised 7 October 2002; accepted 9 October 2002
Abstract—We report the synthesis and pharmacological evaluation of new derivatives of natural dipeptide antibiotic TAN-1057 A,
B. In the course of this program, we identified novel analogues of the natural product that displaysimilar antibacterial activityand
showed improved tolerability.
# 2002 Elsevier Science Ltd. All rights reserved.
Introduction
In this communication, we summarize the synthesis
and pharmacologyof new TAN-1057 A, B analogues
focusing on alterations of the b-amino-acid side
chain.
Staphylococcus aureus is the most prominent nosoco-
mial pathogen causing severe infections such as sepsis,
pneumonia, or endocarditis. The appearance of multi-
resistant staphylococci (MRSA)1 increases the need for
novel antibiotics which effectivelycombat these resis-
tant pathogens.
Chemistry
At the outset of our program, we studied the influence of
the terminal guanidine moietyand its linkage to the
b-amino acid on the SAR. Thus, our first set of analo-
gues was structurallycloselyrelated to the natural pro-
duct, bearing functionalized guanidine or amidine
moieties differentlytethered to the amino acid side chain.
For the natural dipeptide antibiotic TAN-1057 A, B (1)
(Fig. 1) excellent Minimal InhibitoryConcentrations
(MIC) against staphylococci including MRSA were
reported.2 However, TAN-1057 A, B suffers from toxic
side effects (LD50 in the mouse: 50 mg/kg, ip) which in
our view is prohibitive for the therapeutic use in humans.
We accessed b-arginine derivative 5 (Scheme 1, Table 1)
starting from commerciallyavailable BOC- Z-b-ornithine
We initiated a chemical optimization program3 with the
goal to identifynovel analogues of the natural product
which exhibit good anti-staphylococcal activity and
possess a favorable toxicological profile.
Scheme 1. Synthesis of 5 and 9, reagents and conditions: (a) EDC,
HOBt, DMF, TAN-heterocycle; (b) HCl, dioxane; (c) 4, HgCl2, DMF,
Et3N, 50 ꢀC; (d) PdCl2, H2, MeOH; (e) DIC, DMAP, CH2Cl2,
2-SiMe3-ethanol; (f) 8, HgCl2, DMF, DIPEA, 60 ꢀC; (g) TBAF, THF;
(h) HATU, DIPEA, DMF, TAN-heterocycle; (i) HBr, acetic acid.
Figure 1. Natural dipeptide antibiotic TAN-1057 A, B.
*Corresponding author. Tel.: +49-202-362554; fax: +49-202-362566;
e-mail: joachim.krueger.jk2@bayer-ag.de
0960-894X/03/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00880-6

242
M. Brands et al. / Bioorg. Med. Chem. Lett. 13 (2003) 241–245
24 which was coupled with the racemic methylamino-
dihydropyrimidinone heterocyle (TAN-heterocycle)5
employing EDC/HOBt. Subsequently, the BOC-group
was cleaved off, the resulting primaryamine 3 was gua-
nylated using S-methylisothiourea 4 as electrophilic com-
ponent,6 and the protecting groups were finallyremoved
by hydrogenolysis to yield the b-arginine analogue 5.
pling with the TAN-heterocycle and deprotection to
provide 9, albeit in poor yield.
The synthesis of methylated TAN-1057 (10) and the
amidine analogue 11 as well as phenyl-bridged deriva-
tive 12 and homo-TAN-1057 (13) (cf. Table 1) have
been detailed before.3d
In analogy, the dihydroimidazole derivative 9 (cf.
Scheme 1, Table 1) was also obtained via a guanylation
protocol. The synthesis of 9 commenced with BOC-Z-b-
lysine4 (6) which was transformed into the corresponding
trimethylsilylethyl ester followed by BOC-deprotection
to give intermediate 7 in 73% yield. The installation of
the dihydroindole moiety proved problematic, and opti-
mized conditions led to the Z-protected guanidine inter-
The syntheses of b-lysine and b-homolysine analogues
14–17 (Table 1) were realized bya convergent synthesis
strategyusing a peptide coupling to fuse the TAN-het-
erocycle with the corresponding enantiomerically pure
bis-Z-protected (S)-b-amino acids. Hydrogenolytic
cleavage led to the target compounds.3d
The approach to keto- and hydroxy-substituted b-lysine
analogues 24 and 25 is depicted in Scheme 2, path A.
Accordingly, Z-protected 3-amino propionic acid 18
mediate in merely24% yield. The final steps towards 9
included liberation of the carboxylic acid, peptide cou-
Table 1. Pharmacological profile of novel TAN-1057 A, B analogues 1, 5, 9–17, 24, 25, 29, 33, 37–40, 42 and 46
Compd
Structures
MIC
(mg/mL) (mM), (mM),
S. aureus proc. euc.
transl. transl.
IC50
IC50 Selectivity
Compd
Structures
MIC
(mg/mL) (mM), (mM),
S. aureus proc. euc.
IC50
IC50 Selectivity
indexa
indexa
133
133
transl. transl.
24b
25c
29c
33c
37c
38b
39b
40b
42b
46c
3.2
3.0
3.0
1.5
1.0
2.4
n.d.
3.2
n.d.
n.d.
n.d.
1.2
1b
0.10
25.0
12.5
0.80
0.40
50
0.30 0.17
0.57
1.0
5b
1.2
9.4
1.2
9.4
0.40
6.3
9b
1.0
n.d.
1.2
10b
11b
12c
13b
14b
15b
16b
17b
0.50 4.5
9.1
0.40
3.2
2.4
1.0
0.10 0.60
0.17
0.52
5.3
100
>64
n.d.
n.d.
1.0
0.5
0.26
50
25
64
25
0.20
0.40
6.3
0.30 1.6
12.5
n.d.
n.d.
1.0
2.0
6.0
3.7
9.4
1.9
0.40
0.20 0.20
1.6
100
>64
n.d.
n.d.
0.05
0.80
0.50 2.8
5.3
4.0
7.6
1.9
aIn vitro selectivityindex (IC 50 eucaryotic transl./procaryotic transl.).
bCompounds were obtained as a 1:1 mixture of diastereomers.
cCompounds were obtained in racemic form as mixtures of diastereomers.

M. Brands et al. / Bioorg. Med. Chem. Lett. 13 (2003) 241–245
243
For the synthesis of branched b-lysine side chains, we
elaborated a general route which is depicted in Scheme
3. Accordingly, racemic pyrrolidinone 30 was Z-pro-
tected and subsequentlytreated with the Li-enolate
derived from ethyl acetate to furnish 31 in 48% yield
(Scheme 4, path A).8 In close resemblance to the chem-
istrydiscussed for Scheme 2, 31 was taken further to the
desired product 33. The synthesis of 37 as outlined in
Scheme 3, path B was conducted in close analogyto the
synthesis of 33.
Finally, we synthesized a variety of TAN-1057 deriva-
tives with non-basic end-groups. The constructions of
analogues 38, 39, and 40 (Table 1) have been reported
before,3d while the synthesis of thiourea 42 and alcohol
46 is outlined in Scheme 4.
Scheme 2. Synthesis of 24, 25 and 29, reagents and conditions: (a)
CDI, Mg-ethylmalonate, THF; (b) p-toluenesulfonic acid, ethanediol,
benzene; (c) CH2Cl2, KOSiMe3; (d) NH3, MeOH, NaCNBH3, HOAc;
(e) Z-Cl, CH2Cl2, Et3N; (f) separation of enantiomers bychiral HPLC;
(g) HATU, DIPEA, DMF, TAN-heterocycle; (h) aq HCl, THF; (i)
PdCl2, H2, MeOH; (j) NaBH4, THF, MeOH, À78 ꢀC, 95%.
The synthesis of the thiourea derivative 42 commenced
with BOC-Z-b-lysine (6) which was coupled with the
TAN-heterocycle followed by removal of the
BOC-group to give intermediate 41 in 67% yield for two
steps (Scheme 4, path A). The thiourea moietywas
installed byreaction of 41 with (FMOC)NCS9 followed
bybase-induced cleavage of the FMOC group. Finally,
hydrogenolytic deprotection gave rise to the formation
of 42.
was activated with carbonyldiimidazole and reacted with
Mg-ethylmalonate to yield ketoester 19 in 75% yield.7
The keto moietywas protected as a cyclic ketal, the ester
was saponified and the resulting carboxylic acid 20 was
transferred into the b-ketoester using the above men-
tioned Mg-malonate method. Subsequently, the amino
moietywas installed using a reductive amination proto-
col followed byZ-protection of the amino group. The
ester present in 21 was cleaved and the resulting car-
boxylic acid 22 was resolved into the single enantiomers
employing a chiral HPLC phase. The (S)-configurated
carboxylic acid was taken further in the sequence. Next,
the TAN-heterocycle was introduced by HATU-medi-
ated peptide coupling to yield the fully protected inter-
mediate 23. Stepwise deprotection led to 24 in 67%
yield, while ketal cleavage and NaBH4 reduction fol-
lowed byremoval of the remaining Z-groups provided
25 as a mixture of diastereomers in 44% yield.
For the construction of alcohol 46, we envisioned a
route involving the allyl-substituted b-lactam 4310 as
starting material (Scheme 4, path B). Thus, racemic
azetidinone 43 was opened under acidic conditions, and
the resulting b-amino moietywas Z-protected to yield
44 in 39%. Hydroboration, TBS-protection and sapo-
nification led to carboxylic acid 45 which was then
taken further to 46 as outlined in Scheme 4.
Pharmacology
The antibacterial as well as the cytotoxic properties of
TAN-1057 A, B and its derivatives originate from an
inhibition of the translation machinery.2a,3e Therefore,
we determined the in vitro inhibition of the procary-
otic- and eucaryotic translation using biochemical
translation assays.11 The therapeutic index was defined
as the ratio between the eucaryotic and the procaryotic
translation data. We consider this index as a helpful
tool for the preliminaryassessment of the toxicological
profile of the new derivatives. As a whole cell system
Racemic morpholine analogue 29 was obtained as fol-
lows: 26 was transformed into the b-ketoester 27, and
the amino moietywas introduced byreductive amina-
tion as outlined before. Finally, the heterocycle was
installed, and the Z-groups were removed byhdyro-
genolysis to give 29 as a mixture of diasteremoers.
Scheme 3. Synthesis of 33 and 37, reagents and conditions: (a) À78 ꢀC,
BuLi, THF, ZCl; (b) À78 ꢀC, LDA, EtOAc; (c) NH3, MeOH,
NaCNBH3, HOAc; (d) K2CO3, ZCl, dioxane/H2O; (e) CH2Cl2,
KOSiMe3; (f) HATU, DIPEA, DMF, TAN-heterocycle; (g) PdCl2,
H2, MeOH; (h) BOC2O, DMAP, MeCN; (i) BOC2O, Na2CO3, THF,
H2O; (j) HCl, dioxane.
Scheme 4. Synthesis of 42 and 46, reagents and conditions: (a) EDC,
HOBt, DMF, TAN-heterocycle; (b) HCl, dioxane; (c) (FMOC)NCS,
DIPEA, CH2Cl2; (d) piperidine, DMF; (e) PdCl2, H2, MeOH; (f) HCl,
EtOH; (g) ZCl, NEt3, CH2Cl2; (h) 9-BBN, THF then H2O2, Na2CO3;
(i) TBSCl, imidazole, DMF; (j) KOSiMe3, CH2Cl2; (k) HATU,
DIPEA, DMF, TAN-heterocycle; (l) HOAc, THF, H2O.

244
M. Brands et al. / Bioorg. Med. Chem. Lett. 13 (2003) 241–245
we selected a standard MIC assayagainst S. aureus
133. The pharmacological results are summarized in
Table 1.
Table 2. Cytotoxicity and in vivo activity for compounds 1, 10, 14
and 16
Compd
EC50 (mM)
Cytotoxicity
ED100 (mg/kg)
S. aureus 133
Not surprisingly, for the natural product 1 we found a
selectivitytowards the eucaroytic translation under-
scoring the toxic potential of TAN-1057 A, B (selectiv-
ityindex: 0.57). In turn, some of the closest analogues
revealed a more promising pharmacological profile. In
particular, methylated guanidine 10, homo-TAN-1057
13 and amidine 11 show a significantlyimproved ther-
apeutic index indicating that these derivatives should be
less toxic. At the same time, analogues 10, 11 and 13
display, compared to the natural product, similar activ-
ity in both the procaryotic translation and MIC assay.
1
0.25
5.50
25.0
6.0
0.25
2.0
0.20
0.25
10
14
16
As depicted in Table 2, we collected EC50 values for
cytotoxicity using a macrophage cell line (J774) which
has proven to be predictive for systemic toxicity of inhib-
itors of the protein biosynthesis. As we can conclude
from Table 2, our analogues 10, 14 and 16 were at least
20-fold less cytotoxic than the natural product. Addi-
tionally, all compounds in Table 2 show excellent ED100
values (murine sepsis model, route: iv) ranging between
0.2 and 2.0 mg/kg. Thus, compounds 10, 14 and 16 show
similar efficacies as TAN-1057 A, B against S. aureus 133
combined with a significantlyimproved tolerability.
In contrast, the antibacterial activitydrops dramatically
for the b-arginine derivative 5, for the modified
b-homoarginine 9 and for the phenyl-bridged derivative
12. Additionally, these analogues showed no favorable
selectivityindex.
As we can further conclude from Table 1, the guanidine
present in the natural product can be replaced byprimary
or secondaryamines with retention of the anti-staphylo-
coccal activity( 14–37). The most active molecules in this
series contain a b-lysine (14) or a b-homolysine (16) side
chain and small substituents such as methyl (33),
N-ethyl (17), or hydroxy (25) are well tolerated. More
expansive substitutions of the b-lysine moiety as found
in 29 or 37 led to less active compounds. Furthermore,
the antibacterial activityproved sensitive to the absolute
configuration of the b-amino stereocenter since the
(R)-configurated analogue 15 was bya factor of 15 less
active than its (S)-counterpart 14.
Summary
In summary, our chemistry program provided novel
analogues of TAN-1057 A,B which showed improved
tolerabilitycompared to the natural product with con-
comitant retention of the excellent anti-staphylococcal
activity.
Acknowledgements
We thank F.-U. Geschke, D. Habich and H. Labi-
schinski for helpful discussions.
Surprisingly, most of the active amino derivatives show
a favorable selectivityindex. This is especiallytrue for
compounds 14, 16 and 17. Moreover, these analogues
exhibit promising MIC values ranging from 0.04 mg/mL
to 0.8 mg/mL.
References and Notes
1. (a) Ehlert, K. Curr. Pharm. Des. 1999, 5, 45. (b) Marchese,
A.; Schito, G. C.; Debbia, E. A. J. Chemotherapy 2000, 12,
459. (c) Levy, S. B. Clin. Microbiol. Infect. 2000, 6, 101.
2. (a) Katayama, N.; Fukusumi, S.; Funabashi, Y.; Iwahi, T.;
Ono, H. J. Antibiot. 1993, 46, 606. (b) Funabashi, Y.; Tsubo-
tani, S.; Koyama, K.; Katayama, N.; Harada, S. Tetrahedron
1993, 49, 13.
3. For total synthesis of TAN-1057 A, B, see: (a) Williams,
R. M.; Yuan, C. J. Am. Chem. Soc. 1997, 119, 11777. (b)
Sokolov, V. V.; Kozhushkov, S. I.; Nikolskaya, S.; Belov,
V. N.; Es-Sayed, M.; de Meijere, A. Eur. J. Org. Chem. 1998,
777. For the synthesis of analogues, see: (c) Williams, R. M.;
Yuan, C.; Lee, V. J.; Chamberland, S. J. Antibiotics 1998, 51,
189. (d) Brands, M.; Es-Sayed. M.; Habich, D.; Raddatz, S.;
Kruger, J.; Endermann, R.; Gahlmann, R.; Kroll, H.-P.;
Geschke, F.-U.; de Meijere, A.; Belov, V.; Sokolov, V.; Koz-
hushkov, S.; Kordes, M. WO 00/12484. (e) Brands, M.;
Endermann, R.; Gahlmann, R.; Kruger, J.; Raddatz, S.;
Stoltefuß, J.; Belov, V. N.; Nizamov, S.; Sokolov, V. V.; de
Meijere, A. J. Med. Chem. 2002, 45, 4246.
Finally, we collected pharmacological data for analo-
gues with non-basic end groups. Replacement of the
basic guanidine in TAN-1057 A, B bya carbamate ( 38),
nitrile (39), primaryamide ( 40), or alcohol (46) resulted
in a dramatic loss of activitywhile thiourea 42 as the
onlynon-basic functionalitydisplayed a promising MIC
of 0.4 mg/mL. However, for 42 we observed no selectiv-
ityin the translation assays indicating that the thiourea
might have a toxic potential.
From the data in Table 1 we can deduce a recipe for the
construction of active TAN-1057 A, B analogues which
show favorable in vitro selectivityindices: the ( S)-con-
figurated b-amino-acid fragment has to be linked via a
C3–C4 spacer to a basic end-group which preferably
consists of a methylated guanidine, an amidine or an
amino group. Small substitutions such as methyl, oxo or
hydroxy are well tolerated in the side chain.
4. BOC-Z-b-ornithine and BOC-Z-b-lysine were purchased
from EMKA Chemical Enterprise, Ltd.
5. The racemic TAN-heterocylce was obtained according to
ref 3b.
6. (a) Levallet, C.; Lerpiniere, J.; Ko, S. Y. Tetrahedron 1997,
Finally, we selected three of the most promising candi-
dates from Table 1 for further pharmacological studies.

M. Brands et al. / Bioorg. Med. Chem. Lett. 13 (2003) 241–245
245
53, 5291. (b) Guo, Z.-X.; Cammidge, A. N.; Horwell, D. C.
Synth. Comm. 2000, 30, 2933.
9. Kearney, P. C.; Fernandez, M.; Flygare, J. A. J. Org.
Chem. 1998, 63, 196.
7. (a) Li, B.; Franck, R. W. Bioorg. Med. Chem. Lett.
1999, 9, 2629. (b) Gomez-Monterrey, I.; Gonzalez-Muniz, R.;
Herranz, R.; Garcia-Lopez, M. T. Tetrahedron Lett. 1993, 34,
3593.
8. Ezquerra, J.; de Mendoza, J.; Pedregal, C.; Ramirez, C.
Tetrahedron Lett. 1992, 33, 5589.
10. Tarling, C. A.; Holmes, A. B.; Markwell, R. E.; Pearson,
N. D. J. Chem. Soc., Perkin Trans. 1 1999, 12, 1695.
11. (a) Muller, M.; Blobel, G. Proc. Natl. Acad. Sci.
U.S.A. 1984, 81, 7737. (b) Escherichia coli S30 Extract
System for Circular DNA, Technical Bulletin 092, Promega
Corp.