4144
K. Ngu et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4141–4145
Table 3
were readily transformed to the pyrazolyl amines 7 by treatment
with hydrazine in the presence of acetic acid in methanol.
Selectivity data for selected compounds
Compound
Inhibitiona
5-LOb
The various analogs thus prepared were tested for in vitro activ-
ity against mammalian 15-LO isolated from rabbit reticulocyte
using either arachidonic acid (AA) or linoleic acid (LA) as sub-
strates. Inhibition was measured using a standard colorimetric as-
say monitoring the formation of the lipid hydroperoxide product of
either arachidonic acid or linoleic acid [15-hydroperoxyeicosatet-
raenoic acid (15-HPETE) or 13-hydroperoxyoctadecadienoic acid
(13-HPODE), respectively].13 Compound 8a displayed modest po-
15-LO IC50 (nM)
12-LOb (%)
8i
19
NT
68%
5.5 lM
60%
35
38
32
35
16
15f
15g
15h
15i
8.2
2.4
4.1
3.1
NT
NT = not tested.
a
Enzyme inhibition measured using arachidonic acid as substrate.
Displayed data represent IC50 values or % inhibition at 10 lM.
tency against 15-LO with an IC50 value of 3.98 lM using LA as
b
the substrate (Table 1). Replacement of the methyl group of 8a
with isobutyl generated compound 8b with submicromolar
potency. Further modification of isobutyl to phenyl resulted in a
twofold increase in potency (compound 8c, IC50 = 150 nM). Addi-
tional SAR was carried out by keeping the R2 group constant as
phenyl while modifications were made to the R1 group. The unsub-
stituted phenyl analog 8f (R1 = Ph) displayed superior potency
against 15-LO both with AA and LA as substrates compared to
analogs with methoxyphenyl groups (8c–e). While replacement
of the R2 phenyl with benzyl (8g) resulted in fivefold loss in po-
tency, introduction of a thien-2-yl group led to compound 8h
which was highly potent in both the LA and AA based-assays
(IC50 = 11 and 28 nM, respectively). The preference for a small
aromatic ring at R1 dictated an additional SAR investigation using
the thien-2-yl group as a constant while further modifications
were made to R2. Replacement of the R2 phenyl with heteroaryls
(e.g. isoxazolyl, thienyl, pyrazinyl) led to potent 15-LO inhibitors
with low double and single digit nanomolar enzyme inhibitory
activities. In particular, the symmetrical thien-2-yl analog 8i, with
4 and 19 nM inhibitory activities respectively in the LA and AA
based-assays, was determined to be the most potent analog within
this series.
of inhibitory activity in the cell based assay. For example, the
bis-thienyl sulfamide 15c with a c log P value of 5.09, displayed
79% inhibition at 1
than the corresponding sulfonamide 8i (c log P = 7.11, 54% inhibi-
tion at 10 M). Introduction of a more polar pyrazinyl ring (15e–
lM, and was significantly more potent in cells
l
15i) resulted in dramatic improvement in polarity and aqueous
solubility (e.g. compound 15e: c log P = 3.14, aqueous solubility =
79 lg/mL) as well as, in most cases, a significantly enhanced
inhibitory activity in the cells (Table 2). For example, the pyrazinyl
analog containing a 4-fluorophenyl group (compound 15i) with a
c log P value of 4.12 shows a marked improvement in aqueous sol-
ubility (129 lg/mL at pH 6.5) and significantly improved inhibitory
activity in CHO cells (IC50 = 51 nM).
In order to identify potential in vivo candidates, select com-
pounds were evaluated in a rabbit whole blood assay measuring
inhibition of 15-HETE production.16,17 Inhibitory activity was mea-
sured at 1, 3 or 10 lM concentrations and is outlined in Table 2. As
predicted by the CHO cell activity, the relatively hydrophobic and
poorly soluble n-pentlyphenyl analog 8i (c log P 7.11, aqueous solu-
bility < 1
lg/mL) displayed weak inhibition in the whole blood assay
Select compounds were tested in a cell based-assay using
Chinese hamster ovary (CHO) cells stably over expressing human
recombinant 15-LO.15 With the exception of compound 8i, which
(25% at 10
lM). In contrast, and not unexpectedly, the more polar
analogs (15a–15i) with generally improved CHO cell activity were
also significantly more potent in the whole blood assay. Among
the best was compound 15i: CHO cell IC50 of 51 nM and 73% inhibi-
exhibited only modest activity (54% inhibition at 10 lM, Table 1),
most compounds in Table 1 displayed relatively poor activity in
the cell-based assay. In addition, the various sulfonamides
described herein were exceedingly hydrophobic thus displaying
poor aqueous solubilities (e.g. compound 8i: pH 6.5, solubil-
tion of 15-HETE in the whole blood assay at 3 lM concentration.
Several compounds were evaluated for inhibition selectivities
versus the 5- and 12-lipoxygenase isoforms. As outlined in Table 3,
all compounds tested showed excellent selectivity with only mod-
est inhibition of 5-LO or 12-LO at relatively high concentrations
ity = 0.2 lg/mL, c log P = 7.1). Efforts to introduce more polar func-
tionalities centered around the optimization of thienyl analogs 8h
and 8i. In our most recent communication,15 we demonstrated that
replacement of the sulfonamide group with sulfamide functional-
ity resulted in compounds with lower c log P values and increased
aqueous solubilities. These sulfamides 15 were generated by deriv-
atization of amines 7 with 2-chloroethyl chlorosulfonylcarbamate
(13) in the presence of triethylamine to afford the oxazolidinones
14 (Scheme 3) which could readily be converted to the target sul-
famides 15 via treatment with various secondary amines.15
(10 lM).
In conclusion, we initially identified a novel series of pyrazoyl
sulfonamides as potent inhibitors of the mammalian 15-LO. While
the enzyme activity of this series could be significantly improved,
most sulfonamides displayed poor activity in cell-based assays.
Extensive SAR studies resulted in the discovery of the pyrrolidinyl
sulfamide series with significantly improved aqueous solubility
and enzyme and cell activity. Additionally, as demonstrated by
compound 15i (CHO cell IC50 = 51 nM, whole blood inhibition
Biological evaluation of compounds 15 revealed further
improvement in enzyme inhibitory activities. As outlined in Table
2, in addition to excellent potency in the LA based assay, most sul-
famides also displayed improved activity in the AA based-assay,
exhibiting similar single digit nanomolar potencies. The baseline
N-benzylpyrrolidine analog 15a displayed 3.6 and 7.5 nM inhibi-
tion respectively in the LA and AA based-screens. Introduction of
a 3-OMe group at the phenyl group (15b) resulted in a twofold
improvement in potency. Replacement of the R2 phenyl with hete-
roaryls (compounds 15c–15e) retained single digit nanomolar
enzyme activities. Introduction of small hydrophobic groups (Me,
F) at the benzylamine site was well tolerated. As hoped, introduc-
tion of the more polar pyrrolidinyl sulfamide group had a major
impact on the overall polarity (c log P) and aqueous solubility of
the target molecules. This resulted in a significant enhancement
73% at 3 lM), compounds with enhanced cell activity also dis-
played good inhibition in the whole blood assay. As such, many
of these analogs appeared to exhibit adequate in vitro potency
for testing in vivo. Unfortunately these analogs exhibited poor
pharmacokinetics in rodent models18 upon oral dosing and were
deemed not suitable for pharmacodynamic evaluation.
References and notes
1. Brash, A. R. J. Biol. Chem. 1999, 274, 23679.
2. Kuhn, H.; Thiele, B. J. FEBS Lett. 1999, 449, 7.
3. (a) Prigge, S. T.; Boyington, J. C.; Faig, M.; Doctor, K. S.; Gaffney, B. J.; Amzel, L.
M. Biochimie 1997, 79, 629; (b) Chen, X. S.; Funk, C. D. FASEB J. 1993, 7, 694;
Kuhn, H.; Walther, M.; Kuban, R. J. Prostaglandins Other Lipid Mediators 2002,
68–69, 263.
4. Ghosh, J.; Myers, C. E. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 13182.