2486
T. Ullrich et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2483–2487
regarded as potential drug candidates, since our own
experience with similar compounds has taught us that
methyl esters (unless sterically hindered) are readily
cleaved by esterases in most organs and blood, partic-
ularly in rodents. Metabolic stability of such compounds
is usually so low that the obtained clearance values
would most likely not correlate with biological results.
hydroxyl group but still bears the methoxy substituent)
has a drastically improved metabolic profile for both
species (human: 55 lL/min/mg, rat: 123 lL/min/mg).
We have shown that the adverse metabolic stability
profile of lead compound 2a can be addressed by minor
modifications of its labile vanillyl substituent, but at the
expense of either in vitro or in vivo activity (or both).
Thus, elimination of the phenolic OH or its masking as
an acetal led to improved microsomal stability when
exposed to liver microsomes while sacrificing only little
of its cell-free activity but much of its cellular anti-
adhesive and in vivo activity. The discrepancy between
the in vitro and in vivo profiles of some compounds may
suggest species specificity of the ligands, as both the cell-
free and the cellular LFA-1 binding assays are based on
human proteins. It is also conceivable that the potent
N-vanillyl derivatives exhibit part of their anti-inflam-
matory activity in mice via a hitherto unknown, LFA-1
independent mechanism. Among the derivatives syn-
thesized, 2b emerged with the overall best activity and
stability profile: It shows IC50 values of 0.17 and
6.42 lM in the cell-free and cellular assay, respectively,
significantly inhibits inflammation in the mACD animal
model by 28% even at the low dose (0.1 mg/kg), and
shows in vitro clearance values of 383 lL/min/mg (hu-
man microsomes) and 100 lL/min/mg (rat microsomes).
None of the other amides with improved microsomal
stability are significantly active in the mACD model, not
even at the higher dose (1.0 mg/kg). Compound 2b, as
the only compound fulfilling our criteria, was selected as
a novel lead for further investigations.
Compounds 2e and 6b showed pronounced anti-
inflammatory activity in the mACD model (>40%
inhibition at 1.0 mg/kg; ꢀ30% at 0.1 mg/kg). In the in
vitro assays, however, both compounds were markedly
less active than 2a (cell-free assay: 16- and 20-fold, cel-
lular: 38- and 10-fold, respectively). Interestingly, 2e is
the only analogue, which features a phenolic group and
a stable phenethyl side chain. Acetal masking of the
phenol, as shown in benzodioxole 2b, led to retention of
activity in the cell-free LFA-1 assay (IC50 ¼ 0.17 lM),
compromising activity in the cellular assay
(IC50 ¼ 6.42 lM) and in vivo efficacy (ꢀ30% inhibition
at both doses). All the other test compounds (2c–h)
displayed a mediocre in vitro profile (cell-free assay: IC50
values >0.2 lM, cellular assay: IC50 values >5 lM). We
therefore assumed that the pronounced in vivo anti-
inflammatory activity was highly dependent on the
presence of the phenolic group (as in compounds 2a and
2e). Indeed, an attenuation of in vivo activity could be
observed when the phenol was masked, but could be
presumably ‘deprotected’ in vivo to restore the active
principle (as in acetals 2b and 6b). In the case of dif-
luorobenzodioxoles 2c and 6c, no such conversion was
expected,14 and these compounds showed strongly
reduced in vivo activity, like all other compounds
lacking a free or masked phenolic hydroxyl group.
Compounds with other functions on the benzylic side
chain, such as COOH (2g) or CONH2 (6h and 2h),
lacked activity in the LFA-1 dependent cellular adhesion
assay (IC50 values >30 lM) whereas submicromolar
activity in the cell-free binding assay was retained. In
general, it was observed that all derivatives of 2a were
more active in the cell-free binding assay than in the cell-
based adhesion assay (usually by a factor of >10),
highlighting a common issue with the substance class
that had previously been concealed by the good overall
profile of 2a. A comparison of the precursor compounds
(methyl esters 6a–h) with the methyl amides (2a–h)
implies that the majority of amides show a better profile
in the cell-free LFA-1/ICAM-1 binding assay than the
corresponding methyl esters; the cellular assay and
the in vivo data do not show a preference for either
substance class, respectively.
References and notes
1. For a thorough review on the subject, see: Weitz-Schmidt,
G. Trends Pharm. Sci. 2002, 23, 482.
2. Kallen, J.; Welzenbach, K.; Ramage, P.; Geyl, D.;
Kriwacki, R.; Legge, G.; Cottens, S.; Weitz-Schmidt, G.;
Hommel, U. J. Mol. Biol. 1999, 292, 1.
3. (a) Weitz-Schmidt, G.; Welzenbach, K.; Brinkmann, V.;
Kamata, T.; Kallen, J.; Bruns, C.; Cottens, S.; Takada, Y.;
Hommel, U. Nat. Med. 2001, 7, 687; (b) Shimaoka, M.;
Springer, T. A. Nat. Rev. 2003, 2, 703.
4. (a) Gahmberg, C. G.; Valmu, L.; Fagerholm, S.; Kotovu-
ori, P.; Ihanus, E.; Tian, L.; Pessa-Morikawa, T. Cell.
Mol. Life Sci. 1998, 54, 549; (b) Carlos, T. M.; Harlan,
J. M. Blood 1994, 84, 2068; (c) Hughes, P. E.; Pfaff, M.
Trends Cell Biol. 1998, 8, 359; (d) Dustin, M. L.; Springer,
T. A. Annu. Rev. Immunol. 1991, 9, 27; (e) Ni, H.-T.;
Deeths, M. J.; Li, W.; Mueller, D. L.; Mescher, M. F.
J. Immunol. 1999, 162, 5183.
Importantly, the masking or elimination of the phenolic
OH resulted in a pronounced improvement of the poor
metabolic stability observed for 2a (intrinsic clearance
CLi >500 lL/min/mg in both human and rat liver
microsomes). In comparison to 2a, compounds 2b and
2c (hydroxyl group masked) show markedly decreased
values when exposed to rat liver microsomes (100 and
207 lL/min/mg, respectively). The homovanillyl deriva-
tive 2e exhibits poor metabolic stability (human and rat:
>500 lL/min/mg), being indicative for the liability of the
vanillyl moiety. Compound 2d (which is devoid of the
5. Liu, G. Drugs Future 2001, 26, 767.
6. Liu, G. Exp. Opin. Ther. Patents 2001, 11, 1383.
7. Baenteli, R.; Bauer, W.; Cottens, S.; Ehrhardt, C.;
Hommel, U.; Kallen, J.; Meingassner, J. G.; Nuninger,
F.; Weitz-Schmidt, G. WO 00/48989.
8. Bauer, W.; Cottens, S.; Weitz-Schmidt, G.; Kallen, J.;
Hommel, U. WO 99/11258.
9. Welzenbach, K.; Hommel, U.; Weitz-Schmidt, G. J. Biol.
Chem. 2002, 277, 10590.
10. All tested compounds showed a chemical purity of >95%,
1
as determined by H NMR and HPLC. Further charac-