Benzylamide antagonists of protease activated receptor 2 with anti-inflammatory activity

Article abstract of DOI:10.1016/j.bmcl.2015.12.048

Activation of protease activated receptor 2 (PAR2) has been implicated in inflammatory and metabolic disorders and its inhibition may yield novel therapeutics. Here, we report a series of PAR2 antagonists based on C-terminal capping of 5-isoxazolyl-l-cyclohexylalanine-l-isoleucine, with benzylamine analogues being effective new PAR2 antagonists. 5-Isoxazolyl-l-cyclohexylalanine-l-isoleucine-2-methoxybenzylamine (10) inhibited PAR2-, but not PAR1-, induced release of Ca²⁺(IC₅₀0.5 μM) in human colon cells, IL-6 and TNFα secretion (IC₅₀1-5 μM) from human kidney cells, and was anti-inflammatory in acute rat paw inflammation (ED₅₀5 mg/kg sc). These findings show that new benzylamide antagonists of PAR2 have anti-inflammatory activity.

Full text of DOI:10.1016/j.bmcl.2015.12.048

Bioorganic & Medicinal Chemistry Letters xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Benzylamide antagonists of protease activated receptor 2 with anti-
inflammatory activity
Mei-Kwan Yau a, Ligong Liu a, Junxian Lim a, Rink-Jan Lohman a, Adam J. Cotterell a, Jacky Y. Suen a,
David A. Vesey b, Robert C. Reid a, David P. Fairlie a,
⇑
a Division of Chemistry and Structural Biology, Institute for Molecular Bioscience, The University of Queensland, Brisbane, Queensland 4072, Australia
b Centre for Kidney Disease Research, Department of Medicine, The University of Queensland, Translational Research Institute, Princess Alexandra Hospital, Brisbane, Queensland
4102, Australia
a r t i c l e i n f o
a b s t r a c t
Article history:
Activation of protease activated receptor 2 (PAR2) has been implicated in inflammatory and metabolic
disorders and its inhibition may yield novel therapeutics. Here, we report a series of PAR2 antagonists
Received 14 October 2015
Accepted 15 December 2015
Available online xxxx
based on C-terminal capping of 5-isoxazolyl-
logues being effective new PAR2 antagonists. 5-Isoxazolyl-
benzylamine (10) inhibited PAR2-, but not PAR1-, induced release of Ca2+ (IC50 0.5
colon cells, IL-6 and TNF secretion (IC50 1–5 M) from human kidney cells, and was anti-inflammatory
L
-cyclohexylalanine-
L
-isoleucine, with benzylamine ana-
-isoleucine-2-methoxy-
M) in human
L-cyclohexylalanine-
L
l
Keywords:
Benzylamine
Antagonist
Agonist
Protease activated receptor 2 (PAR2)
Inflammation
a
l
in acute rat paw inflammation (ED50 5 mg/kg sc). These findings show that new benzylamide antagonists
of PAR2 have anti-inflammatory activity.
Ó 2015 Elsevier Ltd. All rights reserved.
Protease activated receptor 2 (PAR2) was the second member
discovered of the protease activated receptors, a unique group of
class A G protein coupled receptors.1–4 PAR2 is expressed in
immune and inflammatory cells including T-cells, monocytes,
macrophages as well as in many types of cancer cells.3–6 There
has been some controversy about PAR2 activation being pro-
and/or anti-inflammatory.4,7 However PAR2 activation is clearly
associated with inflammatory diseases such as arthritis8,9 and
inflammatory bowel disease,10–12 stimulating growth and invasion
with a morpholine group produced a weak antagonist (3, IC50
57
M, iCa2+, HT29 cells, Fig. 1).16
Structure–activity relationship (SAR) studies on the PAR2
antagonist GB88 (4) suggested that the isoxazolyl, cyclohexyl-ala-
nine and isoleucine residues were intolerant of other substitution
for antagonist potency (unpublished results). Our PAR2 homology
model (derived from ORL-1 receptor crystal structure 4EA3)18 also
suggested that the C-terminal spiro[indene-1,40-piperidine] ring of
l
4 may be involved in a p-stacking interaction with the receptor in a
of pro-inflammatory and cancer cells, producing TNF-
a
and IL-6,
hydrophobic pocket created by Y1563.33 and F3006.48. As the struc-
tures of 2 and 4 are similar, it is possible that the phenyl ring of 2
binds to the same hydrophobic pocket as the spiroindene phenyl
ring of 4. Furthermore, the benzylamine core of 2 replaces the
piperidine of 4, while presenting a phenyl ring closer to the amide
linker than in 4. The ready availability of benzylamine derivatives
provided easy access to a focused library around the common core
of 2–4.
and inducing joint swelling, mechanical and thermal hyper-alge-
sia.9,13,14 Effective PAR2 antagonists could be beneficial for probing
PAR2 function in vivo and for treating disease.
GPCR antagonists have often been derived from their corre-
sponding endogenous agonists through minor changes.15,16 2f-
LIGRLO-NH2 (1) and GB110 (2) have been reported as equipotent
PAR2 agonists (Fig. 1),16,17 for example, in activating human colon
adenocarcinoma (HT29) cells to release intracellular calcium
Truncation of agonist 2 by removing the piperidine methy-
lamine gave the parent benzylamine derivative 5 (Table 1). Diverse
functional groups were attached to the benzyl ring with varying
electronic and steric properties. The benzylamine/alkylamine tem-
plates were either commercially available or synthesized in-house.
The antagonist/agonist potency of all compounds on HT29 cells
was assessed using an intracellular calcium (iCa2+) mobilization
assay. Compounds that showed agonist activity ꢀ50% or less at
(iCa2+, EC50 0.2 M).16 Compound 1 is unstable in vivo due to pro-
l
teolysis, but the non-peptidic agonist 2 is a promising template for
elaborating to a plasma stable PAR2 antagonist. We previously
reported that replacement of the C-terminal primary amine in 2
⇑
Corresponding author. Fax: +61 733462990.
E-mail address: d.fairlie@uq.edu.au (D.P. Fairlie).
10 lM were screened for potential antagonist activity. Compound
http://dx.doi.org/10.1016/j.bmcl.2015.12.048
0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Yau, M.-K.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.12.048

2
M.-K. Yau et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
Figure 1. Chemical structures of PAR2 agonists (1, 2) and antagonists (3, 4).
Table 1
Ca2+ release induced or inhibited in HT29 cells by isoxazolyl-L-cyclohexylalanine-L-isoleucine-NR1R2
Entry
% Activation (at 10
l
M)a
% Inhibition (at 1 l
M)b
4
5
6
16
23
40
54
76
93
7
8
28
28
74
82
9
41
20
33
10
100
11
39
32
12
13
87
83
—
—
14
15
16
86
43
42
—
85
57
Please cite this article in press as: Yau, M.-K.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.12.048

M.-K. Yau et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
3
Table 1 (continued)
Entry
% Activation (at 10
l
M)a
% Inhibition (at 1 l
M)b
17
44
18
18
19
20
45
7
94
7
20
27
21
1
0
22
23
24
5
6
2
17
0
12
25
26
1
0
26
29
a
Calcium efflux (%) in HT29 cells induced by compounds alone at 10
lM relative to 100% induced by 100
lM 2f-LIGRLO-NH2 (data at 30 and 100
lM also reported in
Fig. S1, supporting information).
b
% Inhibition by compounds (1 lM) of the calcium efflux elicited in HT29 cells by 1 lM 2f-LIGRLO-NH2 (data at 10 lM of ligands also reported in Fig. S1).
activities were first assessed at fixed concentrations, with promis-
ing ligands further investigated over larger concentration ranges.
Representative single concentration data are shown in Tables 1
and 2.
with an increase in agonist activity. Incorporation of heteroaromat-
ics, such as pyridines (12, 13) or benzimidazole (14), resulted in
full agonists. Electron-withdrawing substituents, such as chloro,
nitro, trifluoromethyl and trifluoromethoxy (15–18), also dis-
A methoxy group at the ortho position (10) showed more potent
antagonist activity than close analogues 5–9. Compound 10 was 2-
played agonist responses up to 40% at 10 lM.
We have observed in a separate study that removal of a poten-
tial H-bond donor at the C-terminus can switch an agonist to an
antagonist, so N-methylated derivatives were prepared (19 vs 5
and 20 vs 10). However, such a modification substantially reduced
antagonist activity (19, 20) relative to parent compounds (5, 10),
possibly due to conformational change or loss of the NH as a H-
bond donor. Therefore, the NH of the amide was retained in later
compounds to maintain antagonist potency.
fold more potent (IC50 0.5
lM, pIC50 6.3 0.1, Fig. 2a) than 4 (IC50
1.0
l
M, Fig. 2a) in inhibiting iCa2+ release in HT29 human colon
cells. It was speculated that the methoxy group on the benzyl ring
of 10 might make a hydrogen bond interaction with the receptor or
fit into a small hydrophobic pocket in PAR2. It is also possible that
the electron-donating methoxy group enhances
p-stacking
between the benzyl ring and an aromatic residue in the receptor.
However, this compound did show some agonist activity at much
higher concentrations (up to 100 lM, Supporting information, SI).
Introduction of flexibility, while increasing the distance between
5-isoxazolyl-Cha-Ile and the aromatic ring, led to compound 11
The size of the o-methoxybenzylamine group in compound 10
is smaller than the spiroindene-[1,4]-piperidine in 4. Therefore,
bulky groups such as ortho, meta, para biphenyl and naphthalene
compounds were synthesized (21–24) to map the hydrophobic
Please cite this article in press as: Yau, M.-K.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.12.048

4
M.-K. Yau et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
Table 2
interaction with receptor, possibly projecting their bulk into a dif-
ferent binding site than for spiro[indene-1,40-piperidine] of 4.
Since compound 10 displayed promising antagonist activity, its
regioisomers were examined with a methoxy group at the meta or
para position. The ortho- (10), meta- (27) and para-methoxy (28)
compounds had similar antagonist potency in the Ca2+ assay
(Table 2), but the para-methoxy (28) analogue was an agonist.
Since the ortho- and meta-methoxy analogues were tolerated,
two methylenedioxyphenyl analogues (29, 30) were synthesized,
both showing similar antagonist potencies to 27 but also with par-
tial agonist activity. To determine whether the oxygen from ortho-
methoxybenzyl in 10 was involved in a H-bond interaction, the
methyl derivative 31 was assessed and it showed a similar antag-
onist profile as the methoxy analogue (10, Fig. 2a). This indicated
that the oxygen from the methoxy group did not participate in a
H-bonding interaction. Therefore, the methoxy or methyl group
at the ortho position must only fill space in the corresponding bind-
ing site in PAR2, which was too small to accommodate an ortho-
phenyl substituent (21) (Table 1). Thus, short aliphatic sub-
stituents, such as O-ethyl, O-propyl, O-butyl and branched isomers
(32–36) were screened for optimal van der Waals contacts, while
locking the receptor into an antagonist-binding conformation.
The more potent antagonists from Table 2 were further evalu-
ated over a wider concentration range to quantify their functional
potencies. Compounds 10 (o-OMe), 27 (p-OMe), 31 (o-Me) and 32
Ca2+ release induced or inhibited in HT29 cells by isoxazolyl-L-cyclohexylalanine-L-
isoleucine-NHR1
% Activation
% Inhibition
(at 1
Entry
27
(at 10
l
M)a
l
M)b
32
98
28
51
33
95
84
29
30
46
29
90
74
31
32
29
1
71
34
8
33
34
(o-OEt) were more potent (IC50 0.4–0.6 lM) than the reference 4
(Fig. 2a), while 35 (sec-butoxy) had similar antagonist potency to
4 (Fig. 2b). The n-butoxy substituent (34) almost abolished activity,
while n-propoxy (33) and sec-butoxy (35) substituents led to
antagonists with negligible agonist activity (Table 2). This SAR
study suggested that the binding pocket around the ortho position
of the benzyl ring was relatively small and that a methyl, methoxy
or ethoxy group were optimal.
1
35
6
53
67
36
38
a,b
Defined as in Table 1.
Figure 2. Selected PAR2 antagonists inhibited calcium release in HT29 cells,
induced by 1 M 2f-LIGRLO-NH2 (100% response), in a concentration-dependent
manner. (a) Compounds 10 (j), 27 (r), 31 (d) and 32 (N) were more potent (IC50
0.5 M, pIC50 6.3 0.1) than reference compound (4, ., IC50 1.0 M, pIC50 5.9 0.1).
(b) Compound 35 (s, IC50 1.0 M, pIC50 5.9 0.2) showed similar antagonist
potency as 4 (., IC50 1.0 M, pIC50 5.9 0.1) and was less potent than 10 (j). Each
l
l
l
l
l
data point represents mean SEM (n = 3).
Figure 3. Compound 10 selectively inhibited PAR2 activation and had anti-
inflammatory effects in in vitro and in vivo. (a) Trypsin (100 nM, 100% response)
induced iCa2+ release in HT29 cells was inhibited by 10 (j, IC50 1.9
l
M, pIC50
M, pIC50 5.0 0.2). (b) 4 and 10
are selective for PAR2 over PAR1 as they did not inhibit (at 100
M) iCa2+ release
triggered by PAR1 agonists ( -thrombin at 10 U/mL or TFLLR at 50 M) in PC3 cells.
(c and d) Secretion of inflammatory cytokines (TNF- , IL6) induced by 2f-LIGRLO-
NH2 (1 M) was reduced in HTEC by treating with PAR2 antagonist 10 at 5 and
10
M (**p <0.01, ***p <0.001).
5.7 0.1) with higher potency than 4 (., IC50 9.6
l
space around the benzyl group. Also, two representatives with
bulky aliphatic substituents (25–26) were prepared. All of these
variations led to significant loss of activity (21–26, Table 1), consis-
l
a
l
a
tent with the benzylamine aromatic ring being important for
p-
l
l
Please cite this article in press as: Yau, M.-K.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.12.048

M.-K. Yau et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
5
The pharmacokinetic profile and anti-inflammatory activity of
compounds and 10 were comparatively evaluated in rats
4
(Fig. 4). Compounds 4 and 10 reached a maximum plasma concen-
tration at 2.5–3 h after oral administration at 10 mg/kg, with a
maximum plasma concentration of 2.0 and 0.5 lM, respectively.
Compound 10 gave a lower plasma concentration after 6 h than
4. These compounds were compared (10 mg/kg, p.o.) for their
capacity to inhibit rat paw swelling induced 2 h later by intraplan-
tar (i.pl.) administration of 2f-LIGRLO-NH2 (Fig. 4). Compound 4
has been reported as an orally active anti-inflammatory agent in
inhibiting PAR2-induced paw inflammation.20,21 Similarly com-
pound 10 is orally active in reducing the severity of paw oedema
following oral administration at 10 mg/kg, but had lower potency
than 4 which was attributed to its lower oral bioavailability
(Fig. 4a). However, when both compounds were compared after
subcutaneous (sc) administration at 5 mg/kg (in saline), 10 was
more potent than 4 in suppressing paw swelling (Fig. 4b).
All compounds reported in this study were synthesized in par-
allel using standard solution phase peptide synthesis protocols and
Boc-protected amino acids (Scheme 1).22 All compounds were
characterized by HRMS, 1H NMR spectroscopy and rpHPLC (SI).
The spectral data for the most potent compound in this study
(10) is reported under reference and notes section.23
In conclusion, several benzylamine derivatives are reported as
novel PAR2 antagonists with superior or equal potency to 4. Com-
pound 10 was able to inhibit intracellular calcium release induced
by PAR2 agonists, the hexapeptide 2f-LIGRLO-NH2 (1) and trypsin,
but not by the PAR1 agonist thrombin. Compound 10 showed anti-
inflammatory activity in a rat paw model of acute inflammation, as
well as inhibition of the secretion of pro-inflammatory cytokines.
Compound 10 was a more effective anti-inflammatory agent than
4 when given sc to rats. These findings offer new clues for develop-
ing PAR2 antagonists for cellular and animal models of disease.
Figure 4. Anti-inflammatory activity of
Compound 10 (Tmax 158 45 min; Cmax 0.51 0.25
not as effective as 4 (Tmax 167 18 min; Cmax 1.97 0.38
in reducing paw oedema following oral administration at 10 mg/kg. Paw oedema in
rats were induced through intraplantar (i.pl.) administration of 2f-LIGRLO-NH2
4
versus 10 in rat paw oedema. (a)
l
M; AUC6h 444 ngꢁh/mL) was
l
M; AUC6h 3420 ngꢁh/mL)
(350 lg per paw in 100 lL saline control). Compounds were given orally in olive oil
120 min prior to 2f-LIGRLO-NH2. (b) Compound 10, at 5 mg/kg subcutaneous
injection, was more effective in reducing paw oedema compared to 4. Each data
point represents mean SEM (n = 3–5). Samples were compared to 2f-LIGRLO-NH2
treated control, **p <0.01, ***p <0.001; significant differences as indicated.
Compound 10 was able to inhibit iCa2+ release induced by two
PAR2 agonists, the PAR2 selective peptide agonist 1 (Fig. 2) and tryp-
sin (Fig. 3a) in HT29 cells. This compound was also tested in a com-
petitive binding assay on human PAR2-transfected CHO-cells (SI,
Acknowledgments
We thank the National Health and Medical Research Council of
Australia (NHMRC) for a Senior Principal Research Fellowship
(APP1027369) to DPF and grants from the NHMRC (APP1084083,
APP1083131, APP1047759), the Australian Research Council
(DP130100629, CE140100011) and the Queensland Government
(CIF grant), and acknowledge the Centre for Inflammation and Dis-
ease Research, University of Queensland for supporting the research.
Fig. S2) where it showed binding affinity (IC50 4.5 lM) for PAR2 by
competing with Europium-labeled 2f-LIGRLO-NH2 (1). To evaluate
the selectivity of 10 for PAR2 over PAR1, a human prostate cancer
cell line (PC3) was used as it has higher PAR1 expression than
HT29. Compound 10 was found to be selective for PAR2 over PAR1,
since it had no effect in blocking the release of calcium in PC3 cells
stimulated with PAR1 agonists (a-thrombin or TFLLR; Fig. 3b).
Previously our group reported that antagonist 4 (GB88) inhib-
ited PAR2-induced pro-inflammatory cytokines released in pri-
mary human kidney tubule epithelial cells (HTEC).19 ELISA
Supplementary data
Supplementary data (synthetic procedures, compound charac-
terization data (1H NMR, HRMS, HPLC) and assay results of all final
compounds) associated with this article can be found, in the online
version, at http://dx.doi.org/10.1016/j.bmcl.2015.12.048.
results here indicate that 10 (5 and 10 lM) also showed inhibitory
effects in attenuating secretion of inflammatory cytokines TNF-
a
and IL6 induced by PAR2 agonist 1 in HTEC (Fig. 3c and d).
Scheme 1. General synthesis of benzylamine/alkylamine derivatives.
Please cite this article in press as: Yau, M.-K.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.12.048

6
M.-K. Yau et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
18. Perry, S. R.; Xu, W.; Wirija, A.; Lim, J.; Yau, M. K.; Stoermer, M. J.; Lucke, A. J.;
References and notes
Fairlie, D. P. J. Chem. Inf. Model. 2015, 55, 1181.
19. Suen, J. Y.; Cotterell, A.; Lohman, R. J.; Lim, J.; Han, A.; Yau, M. K.; Liu, L.; Cooper,
M. A.; Vesey, D. A.; Fairlie, D. P. Br. J. Pharmacol. 2014, 171, 4112.
20. Lohman, R. J.; Cotterell, A. J.; Barry, G. D.; Liu, L.; Suen, J. Y.; Vesey, D. A.; Fairlie,
D. P. FASEB J. 2012, 26, 2877.
21. Suen, J. Y.; Barry, G. D.; Lohman, R. J.; Halili, M. A.; Cotterell, A. J.; Le, G. T.;
Fairlie, D. P. Br. J. Pharmacol. 2012, 165, 1413.
1. Nystedt, S.; Emilsson, K.; Wahlestedt, C.; Sundelin, J. Proc. Natl. Acad. Sci. U.S.A.
1994, 91, 9208.
2. Nystedt, S.; Emilsson, K.; Larsson, A. K.; Strombeck, B.; Sundelin, J. Eur. J.
Biochem. 1995, 232, 84.
3. Barry, G. D.; Le, G. T.; Fairlie, D. P. Curr. Med. Chem. 2006, 13, 243.
4. Adams, M. N.; Ramachandran, R.; Yau, M. K.; Suen, J. Y.; Fairlie, D. P.;
Hollenberg, M. D.; Hooper, J. D. Pharmacol. Ther. 2011, 130, 248.
5. Steinhoff, M.; Buddenkotte, J.; Shpacovitch, V.; Rattenholl, A.; Moormann, C.;
Vergnolle, N.; Luger, T. A.; Hollenberg, M. D. Endocr. Rev. 2005, 26, 1.
6. Rothmeier, A. S.; Ruf, W. Semin. Immunopathol. 2012, 34, 133.
7. De Campo, B. A.; Henry, P. J. Curr. Drug Targets 2006, 7, 629.
8. Ferrell, W. R.; Lockhart, J. C.; Kelso, E. B.; Dunning, L.; Plevin, R.; Meek, S. E.;
Smith, A. J.; Hunter, G. D.; McLean, J. S.; McGarry, F.; Ramage, R.; Jiang, L.;
Kanke, T.; Kawagoe, J. J. Clin. Invest. 2003, 111, 35.
9. Kelso Elizabeth, B.; Lockhart John, C.; Hembrough, T.; Dunning, L.; Plevin, R.;
Hollenberg Morley, D.; Sommerhoff Christian, P.; McLean John, S.; Ferrell
William, R. J. Pharmacol. Exp. Ther. 2006, 316, 1017.
10. Hansen, K. K.; Sherman, P. M.; Cellars, L.; Andrade-Gordon, P.; Pan, Z.; Baruch,
A.; Wallace, J. L.; Hollenberg, M. D.; Vergnolle, N. Proc. Natl. Acad. Sci. U.S.A.
2005, 102, 8363.
11. Tanaka, Y.; Sekiguchi, F.; Hong, H.; Kawabata, A. Biochem. Biophys. Res.
Commun. 2008, 377, 622.
12. Lohman, R. J.; Cotterell, A. J.; Suen, J.; Liu, L.; Do, A. T.; Vesey, D. A.; Fairlie, D. P.
J. Pharmacol. Exp. Ther. 2012, 340, 256.
13. Vergnolle, N.; Wallace, J. L.; Bunnett, N. W.; Hollenberg, M. D. Trends Pharmacol.
Sci. 2001, 22, 146.
14. Xue, M.; Chan, Y. K.; Shen, K.; Dervish, S.; March, L.; Sambrook, P. N.; Jackson, C.
J. Arthritis Rheum. 2012, 64, 88.
22. Compounds were synthesized in solution phase using Boc-protected amino
acids on 100–200 mg scale. Firstly, the amino acid (1.2–1.5 equiv) was
activated with HBTU (1.5 equiv) and DIPEA (1.5 equiv) as 0.2–0.5 M solution
in DMF for 10 min. Then the solution was added to an amino group bearing C-
terminal moiety (R1R2NH) and the mixture was stirred for a minimum of 1 h at
room temperature. The reaction mixture was diluted with EtOAc and washed
with saturated NaHCO3 (2ꢂ). The organic extracts were dried over MgSO4,
filtered and evaporated in vacuo. The crude product was then treated with 20%
TFA in DCM and stirred for 1–2 h to remove the Boc group. TFA was removed
by evaporating the reaction mixture under a stream of N2. The residue was
dissolved in DCM and washed with saturated NaHCO3 (2ꢂ). DCM phase was
dried with MgSO4, filtered and evaporated in vacuo. Subsequent N-Boc-amino
acids and amines were sequentially coupled under the same conditions. Each
coupling reaction was monitored by ESMS, with most reactions going to
completion overnight. All final compounds were purified on rpHPLC (P97% by
analytical HPLC) and fully characterized by NMR and HRMS (yields between
30% and 40%).
23. Spectral data of compound 10: 1H NMR (600 MHz, CDCl3): d 0.82–0.86 (m, 7H),
0.88–1.00 (m, 2H), 1.13 (m, 1H), 1.25–1.37 (m, 2H), 1.43 (m, 1H), 1.59–1.73 (m,
7H), 1.77–1.84 (m, 2H), 3.86 (s, 3H), 4.21 (t, 1H, J = 7.2 Hz), 4.39 (dd, 1H, J = 6.0,
14.4 Hz), 4.48 (dd, 1H, J = 6.0, 14.4 Hz), 4.63 (m, 1H), 6.20 (t, 1H, J = 5.6 Hz),
6.51 (d, 1H, J = 8.8 Hz), 6.87–6.93 (m, 3H, Ar-H), 7.00 (d, 1H, J = 7.2 Hz), 7.23–
7.29 (m, 2H), 8.33 (d, 1H, J = 1.6 Hz). 13C NMR (150 MHz, CDCl3), d 11.2, 15.2,
25.0, 25.9, 26.1, 26.3, 32.7, 33.6, 34.1, 37.6, 39.7, 40.1, 51.2, 55.3, 57.9, 106.8,
110.2, 120.6, 125.6, 129.1, 129.8, 151.0, 155.5, 157.5, 162.4, 170.3, 171.1;
HRMS: calcd for C27H39N4O5+ 499.2915 [MH]+, found 499.2915.
15. Underwood, D. J.; Strader, C. D.; Rivero, R.; Patchett, A. A.; Greenlee, W.;
Prendergast, K. Chem. Biol. 1994, 1, 211.
16. Barry, G. D.; Suen, J. Y.; Le, G. T.; Cotterell, A.; Reid, R. C.; Fairlie, D. P. J. Med.
Chem. 2010, 53, 7428.
17. McGuire, J. J.; Saifeddine, M.; Triggle, C. R.; Sun, K.; Hollenberg, M. D. J.
Pharmacol. Exp. Ther. 2004, 309, 1124.
Please cite this article in press as: Yau, M.-K.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.12.048