4452
R. Ringom et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4449–4452
Table 2. Replacement of the CF3 group and the effect on A-FABP
bindinga
tions with the ligands (Fig. 3). These side chains of
Ile104, Val115, and Cys117 are not conserved in the re-
lated protein H-FABP, and are all replaced by leu-
cines.13 Alignment of the very similar A-FABP and H-
R'
N
˚
FABP protein structures gave an rms distance of 0.8A
HO
N
N
H
over all alpha-carbons.14 It is evident that the leucine
side chains extend further into the binding pocket and
obstruct the binding of this type of A-FABP inhibitors.
Especially, the CF3 group clearly collides with Leu104
of H-FABP.
Cl
Compound
R0
IC50 (lM)
15a
15b
15c
15d
Me
Et
>100
47
Phenyl
2-Fluorophenyl
>100
>100
In summary, a novel structural class of human A-FABP
inhibitors based upon a benzylamino-6-(trifluorometh-
yl)pyrimidine-4(1H)-ones is presented. These com-
pounds were shown to have moderate to high
inhibitory activity. The compounds display a good selec-
tivity for A-FABP over H-FABP and represent poten-
tial leads for human A-FABP inhibition.
a In vitro binding data are reported as the mean of triplicate experi-
ments performed on the same dilution.
binding mode for this series similar to compound 1, the
benzyl groups of 9e and 9a could fit well into a hydro-
phobic pocket that accommodates the piperidine of 1.
Modeling studies suggest that the phenyl group stacks
on Phe16, and the smaller para-substituents fill a minor
pocket defined by Tyr19, Met20, Val23, and Val25.
Bulkier substituents such as p-phenyl become too large,
and this explains the drop in activity for 9h.
References and notes
1. Storch, J.; Thumser, A. E. A. Biochim. Biophys. Acta 2000,
1486, 28–44.
2. Hertzel, A. V.; Bernlohr, D. A. TEM 2000, 11, 175–
180.
3. Hotamisligil, G. S.; Johnson, R. S.; Distel, R. J.; Ellis, R.;
Papaioannou, V. E.; Spiegelman, B. M. Science 1996, 274,
1377–1379.
4. Robl, J. A.; Parker, R. A.; Biller, S. A.; Jamil, H.;
Jacobson, B. L.; Kodukula, K. WO 200015229 and WO
200015230.
The CF3 group of the most potent amino analog 9e was
exchanged for a number of alkyl and phenyl groups. As
shown in Table 2, all replacements gave rise to inactive
compounds except for the ethyl group that retained
some activity. It is obvious from the crystal structure
that the larger substitutions such as phenyl will collide
with Ile104. The size of the CF3 group is well adjusted
to allow the positioning of the rest of the molecule as
a fatty acid mimic. The impact of the CF3 group on
binding as compared to equally sized but inactive methyl
analog 15a must have additional explanations. It seems
to keep the network of water molecules intact, by the
formation of at least one hydrogen bond. Finally, it
may have an effect on the acidity of the pyrimidine hyd-
roxyl group, which could be of importance for the polar
interactions with Tyr128 and Arg126 (Fig. 2).
5. Binas, B.; Danneberg, H.; McWhir, J.; Mullins, L.; Clark,
A. J. FASEB J. 1999, 13, 805–812.
6. For experimental details see: Van Dongen, M. J. P.;
˚
Uppenberg, J.; Svensson, S.; Lundba¨ck, T.; Akerud, T.;
Wikstro¨m, M.; Schultz, J. J. Am. Chem. Soc. 2002, 124,
11874–11880, The fluorescence polarization assay de-
scribed herein was used with minor modifications.
7. Performed as described in: Lehmann, F.; Haile, S.; Axen,
E.; Medina, C.; Uppenberg, J.; Svensson, S.; Lundba¨ck,
T.; Rondahl, L.; Barf, T. Bioorg. Med. Chem. Lett. 2004,
8. (a) Xu, Z. H.; Bernlohr, D. A.; Banaszak, L. J. J. Biol.
Chem. 1993, 268, 7874–7884; (b) Reese-Wagoner, A.;
Thompson, J.; Banaszak, L. Biochim. Biophys. Acta––
Mol. Cell. Biol. Lipids 1999, 1441, 106–116.
A selection of compounds was also evaluated for human
H-FABP inhibitory activity (Table 3). An inherent selec-
tivity for A-FABP against H-FABP is observed for all
tested 6-trifluoromethylpyrimidine analogs, exemplified
by the approximately 30-fold selectivity obtained with
2 and 9e. There are three side chains on one face of
the binding pocket that have close hydrophobic interac-
¨
9. See for instance: Svensson, S.; Ostberg, T.; Jacobsson, M.;
Norstro¨m, C.; Stefansson, K.; Hallen, D.; Climent-Jo-
hansson, I.; Zachrisson, K.; Ogg, D.; Jendeberg, L.
EMBO 2003, 22, 4625–4633.
10. Kaiser, C. J. Org. Chem. 1959, 24, 113–114.
11. Hirayama, F.; Koshio, H.; Katayama, N.; Kurihara, H.;
Taniuchi, Y.; Sato, K.; Hisamichi, N.; Sakai-Moritani, Y.;
Kawasaki, T.; Matsumoto, Y.; Yanagisawa, I. Bioorg.
Med. Chem. 2002, 10, 1509–1523.
12. Elokdah, H. M.; Friedrichs, G. S.; Chai, S.-Y.; Harrison,
B. L.; Primeau, J.; Chlenov, M.; Crandall, D. L. Bioorg.
Med. Chem. Lett. 2002, 12, 1967–1971.
13. The H-FABP coordinates were taken from pdb entry
1HMS: Young, A. C.; Scapin, G.; Kromminga, A.; Patel,
S. B.; Veerkamp, J. H.; Sacchettini, J. C. Structure 1994, 2,
523–534.
14. The alignment was made with the program O: Jones, T.
A.; Zou, J. Y.; Cowan, S. W.; Kjeldgaard, M. Acta
Crystallogr. A 1991, 47, 110–119.
Table 3. Binding affinities for selected compounds for human A-FABP
and human H-FABPa
Compound
A-FABP IC50 (lM)
H-FABP IC50 (lM)
2
0.6
3.9
2.9
4.0
24
17
>100
>100
>100
>100
9a
9e
9k
9l
a In vitro binding data are reported as the mean of triplicate experi-
ments performed on the same dilution.