4-Substituted boro-proline dipeptides: Synthesis, characterization, and dipeptidyl peptidase IV, 8, and 9 activities

Article abstract of DOI:10.1016/j.bmcl.2012.07.033

The boroProline-based dipeptidyl boronic acids were among the first DPP-IV inhibitors identified, and remain the most potent known. We introduced various substitutions at the 4-position of the boroProline ring regioselectively and stereoselectively, and incorporated these aminoboronic acids into a series of 4-substituted boroPro-based dipeptides. Among these dipeptidyl boronic acids, Arg-(4S)-boroHyp (4q) was the most potent inhibitor of DPP-IV, DPP8 and DPP9, while (4S)-Hyp-(4R)-boroHyp (4o) exhibited the most selectivity for DPP-IV over DPP8 and DPP9.

Full text of DOI:10.1016/j.bmcl.2012.07.033

Bioorganic & Medicinal Chemistry Letters 22 (2012) 5536–5540
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
4-Substituted boro-proline dipeptides: Synthesis, characterization, and
dipeptidyl peptidase IV, 8, and 9 activities
Wengen Wu, Yuxin Liu, Lawrence J. Milo Jr., Ying Shu, Peng Zhao, Youhua Li, Iwona Woznica,
Gengli Yu, David G. Sanford, Yuhong Zhou, Sarah E. Poplawski, Beth A. Connolly,
⇑
James L. Sudmeier, William W. Bachovchin, Jack H. Lai
Tufts University School of Medicine, Department of Biochemistry, 136 Harrison Ave., Boston, MA 02111, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The boroProline-based dipeptidyl boronic acids were among the first DPP-IV inhibitors identified, and
remain the most potent known. We introduced various substitutions at the 4-position of the boroProline
ring regioselectively and stereoselectively, and incorporated these aminoboronic acids into a series of 4-
substituted boroPro-based dipeptides. Among these dipeptidyl boronic acids, Arg-(4S)-boroHyp (4q) was
the most potent inhibitor of DPP-IV, DPP8 and DPP9, while (4S)-Hyp-(4R)-boroHyp (4o) exhibited the
most selectivity for DPP-IV over DPP8 and DPP9.
Received 1 May 2012
Revised 4 July 2012
Accepted 6 July 2012
Available online 14 July 2012
Keywords:
4-Substitued boroPro
BoroHyp
Ó 2012 Elsevier Ltd. All rights reserved.
DPP-IV
DPP8
DPP9
Dipeptidyl peptidase IV (DPP-IV, E.C. 3.4.14.5), a serine protease
that degrades the incretin hormones GLP-1 and GIP, is now a vali-
dated target for the treatment of type 2 diabetes. The Xaa-boroPro
class of dipeptidyl boronic acid inhibitors was among the first, and
remain among the most potent known inhibitors of DPP-IV. Typical
Ki values for boroPro-based dipeptides are in the picomolar range,
making them 100- to 1000-fold more potent than DPP-IV inhibi-
tors in clinical use or currently known to be in development.1
Extensive studies have been done using large compound libraries
with variations in the P2 position of Xaa-boroPro dipeptides to
examine the substrate specificity of DPP-IV.2 However, as of yet,
there are very limited reports of P1 proline modification and their
effects on specificity.3 Although the known Xaa-boroPro dipeptides
exhibit excellent inhibition against DPP-IV, they provide very poor
selectivity for DPP-IV over DPP8 and DPP9, a potentially crucial
property conferring safety to these inhibitors.4 Thus, continued
exploration of modifications of these lead compounds is still neces-
sary. Compounds with substitutions in the 4-position of proline
residues, like 4-hydroxyproline (Hyp), play very important roles
in the chemistry, structure, physiological properties, and develop-
ment of amino acid and peptide-based drugs containing such
modified amino acid residues. For example, recently a 4-fluoro-2-
cyanopyrrolidine dipeptide derivative was reported to exhibit
better DPP-IV inhibitory activity than the corresponding
compound lacking substitution at the 4-position.5 In light of the
importance of such a substitution, we attempted to introduce
substitutions at the 4-position of the boroPro pyrrolidine ring to
obtain more potent and selective inhibition with respect to DPP-IV.
Regioselectively and stereoslectively introducing a substituent to
the 4-position of boroPro is challenging. Sunose et al.,6 reported that
N-Boc-3-hydroxypyrrolidine can undergo a directed C-lithiation at
C-5. The resulting dilithiated intermediate can then be trapped
by an eletrophile to regioselectively give the 2-substituted-4-
hydroxypyrrolidine. This strategy was utilized in the lithiation of
N-Boc-(S)-(+)-3-pyrrolidinol, which was then trapped with an alkyl
borate to regioselectively introduce the boronic acid group at the
2-position of the pyrrolidine ring. The (+)-pinanediol protecting
group was subsequently introduced as a chiral auxiliary to enable
the C-2 diastereomers to be resolved by diastereomeric crystalliza-
tion, following which 1a (cis-N-Boc-L-boroHyp-pn) was obtained as
a single, pure diastereomer. This key intermediate allowed for the
facile introduction of other substituents at the 4-postion of the
boroPro pyrrolidine ring.
As outlined in Scheme 1 the commercially available N-Boc-
(S)-(+)-3-pyrrolidinol was lithiated using sec-BuLi (2.2 equiv) in
THF–TMEDA at À78 °C and the solution was then allowed to warm
to À46 °C for 2 h to complete the lithiation. The reaction mixture
was re-cooled to À78 °C and triisopropyl borate was added. The
reaction was allowed to warm to room temperature and stirred
overnight. After sequential aqueous work-ups with NaOH and then
⇑
Corresponding author. Tel.: +1 617 636 2443; fax: +1 617 636 2409.
E-mail address: jack.lai@tufts.edu (J.H. Lai).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.07.033

W. Wu et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5536–5540
HO
5537
HO
HO
c
a, b
O
O
B
O
N
H
B
O
N
HCl.
Boc
N
Boc
N-Boc-(S)-(+)
-3-pyrrolidinol
1a
2a
d
OH
OH
OH
O
N
f
N
e
H2N
N
B
BocHN
B
OH
OH
O
O
N
OH
HO
H2
B
O
O
Cyclo-4a
4a
3a
Scheme 1. Reagents and conditions: (a) s-BuLi, TMEDA, (i-PrO)3B, THF, À78 °C, then NaOH, HCl work-ups; (b) (+)-Pinanediol, Et2O, 35% over two steps; (c) 4 N HCl in dioxane,
98%; (d)
-Boc-Ala-OH, HATU, DIEA, DMF, 94%; (e) BCl3, CH2Cl2, À78 °C, 75%; (f) D2O, pH 8.8.
L
HCl2 the crude boronic acid converted to the (+)-pinanediol ester
by addition of 1.0 equiv of (+)-pinanediol in ethyl ether. A mixture
of cis-2, 4-disubstituted (1a) and trans-2,4-disubstituted adducts
was yielded in a ratio of 7:3, where the desired cis-2,4-disubsti-
tuted was the majority. Pure, 1a was separated successfully from
yield. Removal of the tert-butoxycarbonyl (Boc) group with 4 N HCl
in dioxane afforded 2a as a hydrochloride salt in 98% yield. Com-
pound 2a was then coupled with N-Boc-L-Ala-OH in the presence
of HATU and DIPEA to give 3a, after which the Boc and pinanediol
groups were simultaneously removed by treatment with BCl3 to af-
ford dipeptide boronate 4a in 70% yield over two steps.
the L, D-mixture by crystallization from ethyl acetate in 35% overall
Compound 4a, in addition to being the desired target com-
pound, served to facilitate the determination of both the regio-
chemistry and stereochemistry of the introduced boronate
substituent in 1a and its derivatives. Conversion of 1a to 4a and
subsequently to the corresponding cyclic species (Cyclo-4a) pro-
vides a means by which the stereochemistry and regiochemsitry
can be unambiguously assigned. Compound 1a itself is unsuitable
for stereochemistry and regiochemistry determination, as the (+)-
pinanediol group obfuscates the 2D COSY and NOESY NMR spectra.
Furthermore, removal of the Boc and (+)-pinanediol groups results
in a compound that provides less clear spectral data for this pur-
pose than Cyclo-4a. Compound 4a has similar difficulties as the
deprotected compound, as the same protons would be used for ste-
reochemistry and regiochemistry determination. However, there is
an added complication in that compound 4a has cis and trans iso-
mers that complicate the spectra. The rigid bicyclic structure of Cy-
clo-4a eliminates the complication of cis/trans isomerization, and
provides added NOEs for stereochemical assignment.
C(Hβ)3
O
H
OH
B
H1D
H1B
OH
N
N
H
Hα
HO
HA
H2D
H2B
HG
NOE
NOE
cyclo-4a
2D 1D
G
A
2B
1B
α
β
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
/
α β
Previous studies have demonstrated that dipeptidyl boronic
acids undergo a pH-dependent cyclization through the formation
of a dative B?N bond. This cyclic species is favored over the linear
species above physiological pH.7 Compound 4a was converted to
Cyclo-4a via incubation in D2O at pH 8.8. Cyclo-4a has a very
similar 1D 1H NMR spectrum with regard to the boroProline ring
as cyclo(Xaa-boroPro), and thus aided in our assignments.8 Stereo-
specific assignment of all the protons of Cyclo-4a is shown in Fig-
G/1B
G/2B
2B/1B
A/1B
A/2B
ure 1 (1H–1H COSY spectrum) and Figure
2
(1H–1H NOESY
2D/1D
G/1D
G/2D
spectrum). Cross-peaks between the G proton and each of the 1B,
2B, 1D, and 2D protons were observed in the COSY spectrum; how-
ever, there was no cross-peak observed between G and A, clearly
indicating that the boronic acid and hydroxyl groups are 2, 4-
disubstituted, not 2, 3-disubstituted on the pyrrolidine ring. The
stereochemistry of the carbon alpha to the boronic acid group
was established as R (or L) by the presence of cross-peaks between
G and A protons as well as between the A and
a protons in the
1H–1H 2D-NOESY spectrum.
4.5
4.0
3.5
3.0
ppm
2.5
2.0
1.5
1.0
Having synthesized and confirmed the structure of key interme-
diate compound 1a, we were then able to introduce other substit-
uents at the 4-position of the boroPro pyrrolidine ring and
Figure 1. Presaturated 1H–1H COSY spectrum of 0.03 M Cyclo-4a in D2O at pH 8.8
and 25 °C.

5538
W. Wu et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5536–5540
groups, 4b was obtained in similar yield as 4a. The (4S)-hydroxyl
intermediate (3a) was treated with tosyl chloride in pyridine and
then refluxed with TBAF in THF to give the (4R)-fluoro derivative.
This (4R)-fluoro compound was then deprotected to afford 4c10
in 20% total yield. It should be noted that several other fluorination
reagents including KF and DAST were tried, but proved unsuccess-
ful. Swern oxidation worked well in converting (4S)-hydroxyl
derivative 1a to 5, but it failed to convert 3a or 3b to their corre-
sponding 4-oxo derivatives enroute to the synthesis of 4d. The ob-
tained 4-oxo derivative 5 was converted to 4d by a series of
reactions similar to those used in the preparation of 4a from 1a.
The intermediate (4R)-hydroxyl derivative 3b was converted to
(4S)-chloride using Ph3P/CCl4, which was then deprotected to give
4e.
Finally, we also varied the P2 residues and coupled these to
either (4S)-boroHyp or (4R)-boroHyp, giving 4f–q (Xaa = Glu, Tle,
Pro, (4R)-Hyp, (4S)-Hyp, Aad and Arg), as outlined by Scheme 3.
IC50 values of a series of 4-substituted boro-proline dipeptides
against DPP-IV, DPP8 and DPP911 were determined to facilitate a
further understanding of the structure-activity relationships and
selectivity preferences of each of these enzymes, in particular with
respect to compounds containing various 4-substituted pyrrolidine
P1 aminoboronic acids (Table 1). In contrast to the unsubstituted
analogue, Ala-boroPro (6)4, introducing a substituent at position
4 of the boroProline ring generally resulted in a reduction in
DPP-IV, DPP8 and DPP9 inhibitory activity, while (4S)-hydroxyl
(4a) and (4R)-fluoro derivatives (4c) still retained most of their
2B
β
G
α 2D
1B
1D
A
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
/
α β
2B/1B
G/1B
G/2B
A/1B
A/2B
1D/1B
2D/2B
/A
α
G/A
2D/A
2D/1D
G/1D
G/2D
4.5
4.0
3.5
3.0
ppm
2.5
2.0
1.5
1.0
Figure 2. Presaturated 1H–1H NOESY spectrum of 0.03 M Cyclo-4a in D2O at pH 8.8
and 25 °C.
incorporate these novel aminoboronic acids into dipeptide deriva-
tives. The (4R)-boroHyp derivative, 1b, was obtained in 62% yield
by inverting the configuration at the C-4 atom of 1a via the Mits-
unobu reaction9 (Scheme 2). Removal of the tert-butoxycarbonyl
(Boc) group in 1b with 4 N HCl in dioxane gave another important
intermediate: the trans-boroHyp pinanediol ester, 2b. After cou-
DPP-IV inhibitory activity. The inhibitory activity 4a (IC50
:
2.3 nM) was only fivefold less potent than 6 (IC50: 0.46 nM), while
a similar change on the 2-cyanopyrrolidine dipeptide derivative
was reported to result in a 250-fold decrease.5 However, introduc-
tion of a keto to the 4-possition (4d) indeed greatly decreased the
DPP-IV inhibitory activity (IC50: 200 nM; 430-fold less potent than
6). The stereochemical configuration of the 4-hydroxyl substituent
pling with Boc-
L
-Alanine, followed by the removal of the protecting
HO
HO
O
O
B
O
N
H
HCl.
c
B
O
N
Boc
2b
1b
a, b
d
HO
OH
OH
O
B
O
N
N
e
N
H2N
Boc
BocHN
O
B
O
B
OH
O
3b
HO
4b
1a
O
c, d
g
f, e
OH
O
N
Cl
BocHN
O
O
B
B
O
O
3a
N
N
Boc
O
H2N
B
O
5
OH
HO
h, i, e
4e
c, d, e
F
O
N
O
N
H2N
H2N
B
B
O
OH
OH
HO
HO
4d
4c
Scheme 2. Reagents and conditions: (a) DEAD, Ph3P, p-NO2-PhCO2H, THF, 62%; (b) LiOH, THF–H2O, 93%; (c) 4 N HCl in dioxane, 95–98%; (d)
L-Boc-Ala-OH, HATU, DIEA, DMF,
92–95%; (e) BCl3, CH2Cl2, À78 °C, 70–85%; (f) CCl4–PPh3, 72%; (g) (COCl)2, DMSO, NEt3, CH2Cl2, À78 °C to rt, 76%; (h) TsCl, Py, 85%; (i) TBAF, THF, reflux, 35%.

W. Wu et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5536–5540
5539
OH
125 nM against DPP9) was about 40-fold less potent than 4a
((4S)-boroHyp, IC50: 9.2 nM against DPP8, 3.5 nM against DPP9)
for both DPP8 and DPP9. In the case of the Pro-boroHyp and
Hyp-boroHyp compound pairs, the 4R-boroHyp compounds 4k
HO
4
*
4
*
a, b
O
Boc-Xaa
+
N
B
O
N
Xaa
H
HCl.
B
OH
HO
(IC50: 110 nM against DPP8, 98 nM against DPP9) and 4m (IC50
:
500 nM against DPP8, 200 nM against DPP9) were about 20-fold
less potent than the corresponding 4S-boroHyp compounds 4j
4f-q
2a
2b
or
Xaa = Glu, Tle, Pro, (4R)-Hyp, (4S)-Hyp, Aad, Arg
(IC50: 5.9 nM against DPP8, 2.1 nM against DPP9) and 4l (IC50
:
27 nM against DPP8, 9.3 nM against DPP9) with respect to DPP8
and DPP9 inhibition.
Scheme 3. Reagents and conditions: (a) HATU, DIEA, DMF, 90–95%; (b) BCl3,
CH2Cl2, À78 °C, 70–85%.
Of all the compounds evaluated herein, 4o (IC50: 16 nM against
DPP-IV, 530 nM against DPP8, 240 nM against DPP9) exhibited the
greatest selectivity for DPP-IV over DPP8 and DPP9 (33-fold and
15-fold, respectively). As expected, variation of the P2 amino acid
residue resulted in a wide range of inhibitory activities. The great-
est inhibition against each of DPP-IV, DPP8 and DPP9 was observed
for 4q (IC50: 0.3 nM against DPP-IV, 1.2 nM against DPP8, 0.31 nM
against DPP9), which had the basic amino acid, arginine, in the P2
position. When the P2 amino acid was Glu (4f) or Aad (4p)—both
containing acidic P2 side-chains, only moderate inhibitory activity
was observed (20–120 nM against DPP-IV, DPP8, and DPP9).
Increasing the size of the alkyl group of the P2 residue from a
on the boroHyp residue generally does not significantly affect the
DPP-IV inhibitory activity, except in the case of 4a (4S-boroHyp
derivative), which demonstrated 10-fold greater potency than 4b
(4R-boroHyp derivative). However, this configuration does play
an important role in the inhibitory activity of these compounds
against DPP8 and DPP9. In general, (4R)-boroHyp derivatives were
much less potent than the corresponding (4S)-boroHyp derivatives
against DPP8 and DPP9, thus exhibiting good selectivity for DPP-IV
over DPP8 and DPP9. For example, with the Ala-boroHyp com-
pound pair, 4b ((4R)-boroHyp, IC50: 480 nM against DPP8,
Table 1
Inhibition of dipeptidases by 4-substituted boro-proline dipeptides
R
*
4
N
Xaa
B
OH
HO
Compound
Xaa
Ala
R
DPP-IV inhibition IC50 (nM)
2.3
DPP8 inhibition IC50 (nM)
9.2
DPP9 inhibition IC50 (nM)
3.5
4a
–OH
(4S)
–OH
(4R)
–F
4b
4c
Ala
Ala
23
2.1
480
23
125
11
(4R)
=O
–Cl
4d
4e
Ala
Ala
200
510
l
180
18
l
Ma
29
17
53
59
27
Ma
6.9
35
86
6.4
14
2.1
98
9.3
200
9.5
l
Ma
(4S)
–OH
(4S)
–OH
(4R)
–OH
(4S)
–OH
(4R)
–OH
(4S)
–OH
(4R)
–OH
(4S)
–OH
(4R)
–OH
(4S)
–OH
(4R)
–OH
(4R)
–OH
(4S)
H
4f
Glu
54
4g
4h
4i
Glu
320
17
Tle
Tle
43
4j
Pro
6.2
6.5
20
5.9
4k
4l
Pro
110
27
Hyp(4R)
Hyp(4R)
Hyp(4S)
Hyp(4S)
Aad
4m
4n
4o
4p
4q
48
15
16
17
500
19
530
114
240
40
Arg
0.30
1.2
0.31
6
Ala
—
0.46
6.0b
13
2.8
MK-043112
—
11
l
Mb
20l
Mb
a
The inhibitor stock solutions (1 mg/ml) for enzymatic assays were all prepared at pH 2.0 and equilibrated overnight except 4e which behaved strangely in the enzyme
assays at this pH. The listed data for 4e was thus obtained at pH 7.8.
b
This is in-house data (the assay has been conducted under the same conditions as for 4a etc.).

5540
W. Wu et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5536–5540
methyl group (4a or 4b) to the bulkier tert-butyl group (4h or 4i)
reduced the inhibitory potency against DPP-IV, but this tendency
was not consistent with regard to DPP8 and DPP9, as evidenced
by the fact that 4i (IC50: 43 nM against DPP8, 14 nM against
References and notes
1. Van der Veken, P.; Haemers, A.; Augustyns, K. Curr. Top. Med. Chem. 2007, 7,
621.
2. Coutts Simon, J.; Kelly Terence, A.; Snow Roger, J., et al J. Med. Chem. 1996, 39,
2087.
3. Heiser, Ulrich; Niestroj, Andre J.; Gaertner, Ulf-Torsten; Demuth, Hans-Ulrich.
U.S. Patent 0 293 618, 2008; Chem. Abstr. 2008, 150, 5897.
4. Connolly Beth, A.; Sanford David, G.; Chiluwal Amrita, K.; Healey Sarah, E.;
Peters Diane, E.; Dimare Matthew, T.; Wengen, Wu; Yuxin, Liu; Hlaing, Maw;
Yuhong, Zhou; Youhua, Li; Zhiping, Jin; Sudmeier James, L.; Lai Jack, H.;
Bachovchin William, W. J. Med. Chem. 2008, 51, 6005.
5. Fukushima, Hiroshi; Hiratate, Akira; Takahashi, Masato.; Saito, Masako;
Munetomo, Eiji; Kitano, Kiyokazu.; Saito, Hidetaka; Takaoka, Yuji; Yamamoto,
Koji Bioorg. Med. Chem. 2004, 12, 6053.
6. Sunose, Mihiro; Peakman Torren, M.; Charmant Jonathan, P. H.; Gallagher,
Timothy; Charmant Jonathan, P. H.; Macdonald Simon, J. F. Chem. Commun.
1998, 16, 1723.
DPP9) was more potent against DPP8 and DPP9 than 4b (IC50
:
480 nM against DPP8, 125 nM against DPP9). Replacement of the
P1 Pro residue (4j, 4k) with the more polar Hyp residue (4l, 4m,
4n, 4o) also slightly decreased the inhibition with respect to
DPP-IV, DPP8, and DPP9. However, the configuration of the 4-hy-
droxy group of Hyp does not appear to drastically affect either
the activity or the selectivity of these compounds for DPP-IV,
DPP8, and DPP9.
In summary, we have introduced a hydroxyl group regioselec-
tively and stereoselectively into the 4-position of boroProline ring,
and obtained cis-N-Boc-
L
-BoroHyp-pn (1a) as a single, pure diaste-
7. Snow Roger, J.; Bachovchin William, W.; Barton Randall, W.; Campbell Scot, J.;
Coutts Simon, J.; Freeman Dorothy, M.; Gutheil William, G.; Kelly Terence, A.;
Kennedy Charles, A., et al J. Am. Chem. Soc. 1994, 116, 10860.
reomer following crystallization. By converting 4a to a cyclic spe-
cies (Cyclo-4a) the regiochemistry and stereochemistry of 1a
were then further characterized with 2D NMR. This key intermedi-
ate enabled the successful design and synthesis of a series of 4-
substituted boroPro, including boroHyp-containing dipeptides.
The inhibitory activity of these dipeptides against DPP-IV, DPP8
and DPP9 was determined, and the structure-activity relationships
ascertained. Among the compounds examined, Arg-(4S)-boroHyp
(4q) demonstrated the most potent inhibitory activity against
DPP-IV, DPP8 and DPP9; while (4S)-Hyp-(4R)-boroHyp (4o) exhib-
ited the greatest selectivity for DPP-IV over DPP8 and DPP9.
8. (a) Sudmeier James, L.; Gunther Ulrich, L.; Gutheil William, G.; Coutts Simon, J.;
Snow Roger, J.; Barton Randall, W.; Bachovchin William, W. Biochemistry 1994,
33, 12427; (b) James, Sudmeier; Zhou, Yuhong; Lai Jack, H.; Maw Hlaing, H.;
Wu, Wengen; Bachovchin William, W. J. Am. Chem. Soc. 2005, 127, 8112.
9. Hodges Jonathan, A.; Raines Ronald, T. J. Am. Chem. Soc. 2003, 125, 9262.
10. Characterization data for 4c: 1H NMR (D2O, 300 MHz) d 1.46 (d, J = 6.9 Hz, 3H,
CH3CHNH2), 1.80–2.05 (m, 1H, CH2CHB), 2.30–2.45 (m, 1H, CH2CHB), 3.29 (dd,
J = 6.6 Hz, 6.6 Hz; 1H, CH2CHB), 3.60–4.00 (m, 2H, NCH2CHF), 4.29 (q, J = 6.9 Hz,
2
1H, H2NCHCO), 5.47 (d, JH–F = 52 Hz, 1H, NCH2CHF). LC–MS (ESI+) m/z (rel
intensity): 373.2, ([2 Â (MÀH2O)+H]+, 100); 227.3, ([M+Na]+, 18). HRMS: calcd
for C7H13BFN2O2, [MÀH2O+H]+, 187.1054, found, 187.1054.
11. Enzymatic assays for human DPP-IV, DPP8 and DPP9 were determined at
ambient temperature with the chromogenic substrate Ala-Pro-p-nitroanilide
hydrochloride salt (Bachem) in 50 mM Phosphate Buffer, pH 7.8. Inhibitor
stock solutions (1 mg/ml) were prepared in 0.01 N HCl, pH 2.0. Approximately
1 nM of human DPP-IV (R&D Systems), DPP8, or DPP9 (both supernatant from
transfected HEK 293T cells) enzyme was incubated with concentrations of
inhibitor ranging from 10À4 to 10À11 M for 10 min prior to the addition of the
Acknowledgment
We are grateful to Arisaph Pharmaceuticals, Inc. for their
financial support.
substrate (substrate concentration was 25 lm in DPP-IV assay and 60 lM in
DPP8/9 assays). The mixture was then incubated for 30 min and measured the
absorbance at 410 nM. IC50 values were determined by a nonlinear regression
fit of the data to a variable-slope sigmoidal dose–response curve using Prism
5.0.4 (Graphad, Inc.).
Supplementary data
12. Kim, Dooseop; Wang, Liping; Beconi, Maria; Eiermann George, J.; Fisher
Michael, H.; He, Huaibing; Hickey Gerard, J.; Kowalchick Jennifer, E.; Leiting,
Barbara; Lyons, Kathryn; Marsilio, Frank; McCann Margaret, E.; Patel Reshma,
A.; Petrov, Aleksandr; Scapin, Giovanna; Patel Sangita, B.; Roy, Ranabir Sinha;
Wu Joseph, K.; Wyvratt Matthew, J.; Zhang Bei, B.; Zhu, Lan; Thornberry Nancy,
A.; Weber Ann, E. J. Med. Chem. 2005, 48, 141.
Supplementary data (experimental procedures and compounds
characterization data) associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2012.07.
033.