6318
D. Haebich et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6317–6318
Initially, conjugate 1a attracted our attention by reducing the
N
N
N
production of intracellular b-galactosidase in an E. coli transactiva-
tor assay.6,7 Prototype 1a showed no antibacterial activity in vitro
but exhibited efficacy in vivo in lethal murine sepsis models (infec-
tious dose: LD50–LD90) with Gram-negative and Gram-positive
bacteria (Table 1).
Incorporation of an additional N-methyl-group provided conju-
gate 1b which was more stable in mouse serum and showed a pro-
longed half-live in vivo.10 Conjugate 1b exhibited no antibacterial
a
HN
Me
+
Cl
NH
Me
HN
Me
N
Me
N
N
72%
N
2
3
4
5
O
NH
b
Cbz
N
N
Cbz
N
N
Me
H
95 %
NH Me
activity in vitro as well (MIC >250 lg/mL against E. coli and Staph-
Cbz
ylococcus aureus). Intraperitoneal (ip) treatment of lethal infections
by E. coli in mice with 30 mg/kg 0.5 h and 4 h post infection re-
sulted in 80% survival (Table 1). Notably, the observed efficacy de-
pended on the infective dose (inoculum).
c
1b
80%
In an E. coli sepsis, a 1.5 log reduction (7.8 ꢀ 104 to 2.7 ꢀ
103 CFU11/mL) of bacteria from blood was observed 5 h after ip
administration of a single dose of 50 mg/kg of 1b 0.5 h post infec-
tion (pi). Repeated administration at 0.5 h and 4 h pi resulted in a
more pronounced reduction of 2.5 log units (7.8 ꢀ 104 to
3.7 ꢀ 102 CFU/mL). No active metabolites were produced after
incubation of 1b in mouse serum and no antibacterial activity
could be detected in the serum of mice which were treated with
1b. In vivo efficacy in various sepsis models12 strongly depended
on the route of application. While intraperitoneal (ip) and subcuta-
neous (sc) application of 1b showed good efficacy, peroral (po) and
intravenous (iv) application produced only negligible effects. An
active support by the innate immune defense seemed mandatory,
since in immunocompromised mice 1b showed no activity at
all.13 Prophylactic efficacy in a lethal E. coli sepsis model after ip
treatment with 50 mg/kg of 1b could be demonstrated up to 3 h
before infection (Table 2).
Scheme 1. Exemplary synthesis of arginine–pyrimidine conjugate 1b. Reagents and
conditions: (a) 1,4-dioxane, 48 h, rt; (b) (Cbz)3Arg-OH, EDC, HOBt, CH2Cl2, 16 h, rt; (c)
HBr, HOAc, 4 h, rt; basic ion exchange resin M 600. Cbz: benzyloxycarbonyl, EDC: 1-
ethyl-3-(3-dimethylaminopropyl)carbodiimide, HOBt: 1-hydroxybenzotriazole.
In summary we have described the synthesis and the prelimin-
ary biological profile of novel arginine–pyrimidine conjugates 1
that exhibit therapeutic and prophylactic activity in lethal infec-
tions by Gram-positive and Gram-negative bacteria without show-
ing antibacterial activity in vitro. In the presence of conjugates 1
bacteria seemed no longer able to adapt to the host environment
and were eradicated by the host immune system. An interference
with the bacterial signal transduction pathway could be a possible
reason. Additional studies are required to fully understand the role
of conjugates 1 in host–pathogen interaction and to elucidate their
mode of action.
Up to 450 mg/kg (ip) no acute and no subchronic toxic effects
were observed in mice over 5 days. Neither an Ames-test nor broad
in vitro and in vivo screening of 1b provided hints for potential
pharmacological or toxicological side effects.
Acknowledgments
We thank K.-D. Bremm, W. Springer, and R. Endermann for
stimulating discussions, and A. DiBetta, M. Torner, and D. Wolter
for technical assistance.
An exemplary synthesis of 1b is shown in Scheme 1. Thus,
nucleophilic substitution of 2-chloropyrimidine 3 by N,N0-dimeth-
ylpropane-1,3-diamine (2) in dioxane gave diamine 4 that was
coupled with tris-benzyloxycarbonyl-protected L-arginine in the
References and notes
presence of EDC and HOBt to afford protected conjugate 5. Cleav-
age of the benzyloxycarbonyl groups by HBr in acetic acid and re-
moval of excessive acid by addition of basic ion exchange resin M
600 provided 1b in three steps and 55% overall yield (Scheme 1).
Structural variations of the guanidine, the pyrimidine and the
diamine spacer afforded less active congeners. However, reduction
of rotational bonds and integration of conformational constraints
in diketopiperazine 1c were tolerated.14
1. (a) Walsh, C. Nat. Rev. Microbiol. 2003, 1, 65; (b) Nathan, C. Nature 2004, 431,
899.
2. (a) Coates, A.; Hu, Y.; Bax, R.; Page, C. Nat. Rev. Drug Disc. 2002, 1, 895; (b) von
Nussbaum, F.; Brands, M.; Hinzen, B.; Weigand, S.; Häbich, D. Angew. Chem., Int.
Ed. 2006, 45, 5072; (c) Infectious Disease Society of America (IDSA): http://
3. (a) Liu, C.-I.; Liu, G. Y.; Song, Y.; Yin, F.; Hensler, M. E.; Jeng, W.-Y.; Nizet, V.;
Wang, A. H.-J.; Oldfield, E. Science 2008, 319, 1391; (b) Haebich, D.; von
Nussbaum, F. ChemMedChem 2008, 3, 1173.
4. (a) Brodsky, I. E.; Medzhitov, R. Nat. Cell Biol. 2009, 11, 521; (b) Ulvestad, E.
APMIS 2009, 117, 311.
5. (a) Fuqua, W. C.; Winans, S. C.; Greenberg, E. P. J. Bacteriol. 1994, 176, 269; (b)
Bassler, B. L. Curr. Opin. Microbiol. 1999, 2, 582.
Table 1
Therapeutic efficacy in lethal murine sepsis models after ip treatment with 30 mg/kg
0.5 h and 4 h post infectiona
6. Schmoll, T.; Ott, M.; Oudega, B.; Hacker, J. J. Bacteriol. 1990, 172, 5103.
7. Danese, P. N.; Snyder, W. B.; Cosma, C. L.; Davis, L. J. B.; Silhavy, T. J. Genes Dev.
1995, 9, 387.
Compound
Survivors (%)
S. aureus
E. coli
8. Keinänen, T. A.; Järvinen, A.; Uimari, A.; Vepsäläinen, J.; Khomutov, A. R.;
Grigorenko, N. A.; Hyvönen, M. T.; Cerrada-Gimenez, M.; Alhonen, L.; Jänne, J.
Mini Rev. Med. Chem. 2007, 7, 813.
9. Lührs, P.; Schmidt, W.; Kutil, R.; Buschle, M.; Wagner, S. N.; Stingl, G.;
Schneeberger, A. J. Immunol. 2002, 169, 5217.
1a
1b
40
60
40
20
40
80
60
40
Ciprofloxacin
Control
10. Conjugate 1b was completely stable in mouse plasma after 3 h at 37 °C.
a
Five mice per group.
Pharmacokinetic parameters, mouse: t1/2 = 1 h, cmax = 56
ip application of 100 mg/kg 1b).
lg/mL (30 min after
11. CFU = colony forming units.
12. Handbook of Animal Models of Infection; Zak, O., Sande, M. A., Eds.; Academic
Press: San Diego, USA, 1999.
13. Wound infection by Pseudomonas aeruginosa or muscle infection by
Streptococcus pyogenes in cyclophosphamide-treated (2 ꢀ 100 mg/kg ip on
day 4 and day 1 before infection) immunocompromised mice.
14. Compared to 1b, bacteria were more efficiently eliminated from the blood of
mice with diketopiperazine 1c. In preliminary studies a superior efficacy in
infection models (E. coli, S. aureus) was observed.
Table 2
Prophylactic efficacy in a lethal murine E. coli sepsis model after ip treatment with
50 mg/kg of 1b before infectiona
Time before infection (h)
Survivors (%)
3
2
1
0.5
80
Control
20
40
60
80
a
Five mice per group.