Inhibitors of dipeptidyl peptidase 8 and dipeptidyl peptidase 9. Part 2: Isoindoline containing inhibitors

Article abstract of DOI:10.1016/j.bmcl.2008.05.079

To obtain selective and potent inhibitors of dipeptidyl peptidases 8 and 9, we synthesized a series of substituted isoindolines as modified analogs of allo-Ile-isoindoline, the reference DPP8/9 inhibitor. The influence of phenyl substituents and different P2 residues on the inhibitors' affinity toward other DPPs and more specifically, their potential to discriminate between DPP8 and DPP9 will be discussed. Within this series compound 8j was shown to be a potent and selective inhibitor of DPP8/9 with low activity toward DPP II.

Full text of DOI:10.1016/j.bmcl.2008.05.079

Bioorganic & Medicinal Chemistry Letters 18 (2008) 4159–4162
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Inhibitors of dipeptidyl peptidase 8 and dipeptidyl peptidase 9.
Part 2: Isoindoline containing inhibitors
b
a
Sebastiaan Van Goethem a, Pieter Van der Veken a, , Véronique Dubois , Anna Soroka ,
*
Anne-Marie Lambeir b, Xin Chen c, Achiel Haemers a, Simon Scharpé b, Ingrid De Meester b,
Koen Augustyns a,
*
a Laboratory of Medicinal Chemistry, University of Antwerp (UA), Universiteitsplein 1, B-2610 Antwerp, Belgium
b Laboratory of Medical Biochemistry, University of Antwerp (UA), Universiteitsplein 1, B-2610 Antwerp, Belgium
c Division of Biotechnology and Pharmaceutical Research, National Health Research Institutes, Zhu Nan, Miaoli 350, Taiwan, ROC
a r t i c l e i n f o
a b s t r a c t
Article history:
To obtain selective and potent inhibitors of dipeptidyl peptidases 8 and 9, we synthesized a series of
substituted isoindolines as modified analogs of allo-Ile-isoindoline, the reference DPP8/9 inhibitor. The
influence of phenyl substituents and different P2 residues on the inhibitors’ affinity toward other DPPs
and more specifically, their potential to discriminate between DPP8 and DPP9 will be discussed. Within
this series compound 8j was shown to be a potent and selective inhibitor of DPP8/9 with low activity
toward DPP II.
Received 24 April 2008
Revised 19 May 2008
Accepted 20 May 2008
Available online 24 May 2008
Keywords:
DPP8
Ó 2008 Elsevier Ltd. All rights reserved.
DPP9
DPP IV
DPP4
DPP II
DPP2
Inhibitors
Allo-Ile-isoindoline
Isoindoline
Selectivity
Proline-selective dipeptidyl peptidases (DPPs) are a group of
phylogenetically related serine proteases.1 Many peptide hor-
mones, chemokines, and neuropeptides contain one or more pro-
line residues that could function as regulatory elements for
proteolytic processing. The biological significance of the DPPs is
generally deemed to reside in the hydrolytic modification of pep-
tides containing such ‘checkpoint’-residues, a property suggesting
potential as targets for drug development.2,3 Dipeptidyl peptidase
IV (DPP IV, CD26, EC 3.4.14.5) has become the best-characterized
DPP since it was shown that inhibition of the enzyme leads to
an increased half-life of the incretin hormone glucagon-like
peptide-1.
During the last decade, DPP8 and DPP9 have been described as
two members of the S9b subfamily of serine proteases, in which
also DPP IV is categorized. As compared to the latter, however, rel-
atively little is known about DPP8 and DPP9. Neither the tertiary
structures nor in vivo substrates and corresponding physiological
functions of these two enzymes are currently known.1,4 Similar
to DPP8, DPP9 is also ubiquitously expressed and the catalytic effi-
ciencies of both enzymes toward model substrates were found to
be very similar.5
DPP8 and DPP9 have garnered much attention following re-
search results reported by Lankas et al. suggesting that the use of
the DPP8/9 inhibitor allo-Ile-isoindoline is associated with severe
toxicity in animal models.6 Whether the observed toxicity was
DPP8/DPP9 related, or solely due to compound mediated off-target
effects, remains nonetheless to be unambiguously established.1 In
accordance with this remark, a recent study suggests that DPP8/9
inhibition alone cannot account for the gastrointestinal toxic
symptoms.7 Secondly, another report claims the observation that
DPP8 and DPP9 are upregulated in experimentally induced asthma
and that these peptidases specifically respond to the inflammatory
stimulus.8 Finally, in an in vitro model of T-cell activation, inhibi-
tion of DPP8/9 activity attenuated T-cell proliferation.6 This finding
suggests that the previously assumed DPP IV mediated effect on
the immune function could at least partially be attributed to
DPP8/9.1 In order to verify whether the reported toxicity observa-
tions are related to DPP8/9 inhibition and to further study the
clinical relevance of potential enzyme involvement in pathologies
* Corresponding authors.
E-mail address: pieter.vanderveken@ua.ac.be (P. Van der Veken).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.05.079

4160
S. Van Goethem et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4159–4162
or T-cell function, there is an urgent need for new inhibitors of
these enzymes.
The inhibitors in this report were developed starting from allo-
Ile-isoindoline 1 as a lead (Fig. 1). Two types of structural modifi-
cations were investigated in order to obtain compounds with a
maximally optimized DPP8/9 potency-selectivity profile and to ex-
plore the possibility of discerning between DPP8 and DPP9 with
aminoacyl isoindoline derivatives.
a-
The first modification type consists of introducing substituents
on the phenyl ring of isoindoline in order to allow for a steric and
electronic scan of the P1 pocket of DPPs. Secondly, since we could
not conclude from literature data that the P2 allo-Ile residue in
compound 1 had already been optimized, analogs with different
Scheme 2. Synthesis of 6,7-dihydro-5H-pyrrolo[3,4-b]pyridine compound 14. Re-
agents and conditions: (a) benzylamine followed by thermal dehydration (95–99%);
(b) metallic tin, acetic acid, HCl (37–67%); (c) LiAlH4, THF (13%); (d) H2, Pd/C; (e)
Et3N, Boc2O, DCM (d + e 45%); (f) 1—TFA, DCM; 2—Ile-Boc, TBTU, Et3N, DCM; (1 + 2
56%); 3—TFA, DCM (96%).
P2 residues were prepared. Isoleucine, lysine, and e-N-Z-protected
lysine, three amino acids that were identified elsewhere in this is-
sue as useful building blocks for DPP8/9 inhibitors; along with 4-
aminoproline were selected for this purpose.9 The latter can be
viewed as a conformationally constrained mimic of dibasic
no acids.
a-ami-
For the preparation of structurally modified isoindoline analogs,
mainly three synthetic strategies were followed (Schemes 1–3).
Target compounds 8a–l were synthesized starting from com-
mercially available phthalic anhydrides, outlined in Scheme 1.
The route starts with the amidation of substituted phthalic anhy-
drides 2, using aqueous ammonia in tetrahydrofuran (THF). By
thermal dehydration, a ring closure was performed yielding the
corresponding phthalimides.10 Compounds 3 were reduced using
borane in THF forming isoindolines 4 followed by an N-Boc protec-
tion step to allow for chromatographic purification.11 Acidolytic
cleavage using trifluoroacetic acid (TFA) was followed by coupling
intermediates 6 with N-Boc protected isoleucine or allo-isoleucine.
Final deprotection of these intermediates with TFA afforded com-
pounds 8a–l.
Scheme 3. Synthesis of 5-nitro-, 5-iodo-, and 5-azido-substituted isoindoline inh-
ibitors (20a–c). Reagents and conditions: isoindoline was protected with a N-trifl-
uoroacetyl protecting group using TFA anhydride and Et3N yielding 15 (70%); (a)
nitronium tetrafluoroborate, ACN (70%); (b) H2, Pd/C, methanol (81%); (c) p-TosOH,
ACN; NaNO2/KI, H2O, 15 °C (65%); (d) NaNO2/NaN3; TFA (52%); (e) 1—K2CO3, H2O
(61–99%); 2—Et3N, Ile-Boc, TBTU, DCM (28–35%); 3—TFA, DCM (95–99%).
Due to the incompatibility of 14, 16, and 19 with the reaction
conditions used in the synthesis of the substituted isoindolines,
two other synthetic routes were elaborated.
Preparation of the 6,7-dihydro-5H-pyrrolo[3,4-b]pyridine 14, as
illustrated in Scheme 2, began with protection of the imide by
reacting the starting product with benzylamine followed by ther-
mal dehydration. The protected pyrrolopyridine-5,7-dione 10 was
reduced in two steps to the corresponding amine 12 using metallic
tin in acetic acid followed by reduction of the amide with lithium
aluminum hydride.12,13 The benzyl group was removed via cata-
lytic hydrogenation yielding the free amine. Coupling to isoleucine
Figure 1. Target compounds derived from lead structure allo-Ile-isoindoline (1).
Table 1
IC50 values of substituted isoindolines (8a–h, 20a–c, 14)
Compounda
R
IC50
(
lM)
DPP8
DPP9
DPP IV
DPP II
8a
8b
8c
8d
8e
8f
H
4-F
4-CH3
5-F
5-CH3
5-t-But
5-CF3
5-Br
5-I
5-NO2
5-N3
H
1.3 0.3
7.3 1.3
>100
1.2 0.1
4.6 0.4
>100
>100
4.9 0.5
>25
4.3 0.1
8.6 0.4
>100
0.54 0.02
7.6 0.6
>100
>100
4.1 0.4
>50
115 11
>100
>100
80.1 4.6
>100
>100
>100
>100
>100
>50
43 11
12.2 1.6
>100
23 2.5
>100
>100
8g
13.0 1.3
7.4 0.4
2.6 0.2
8h
20a
20b
20c
14
>50
>100
56.9 7.4
>25
66
5
Scheme 1. Synthesis of substituted isoindoline inhibitors (8a–l). Reagents and
conditions: (a) NH4OH, THF followed by thermal dehydration (95–99%); (b) borane,
THF; (c) Et3N, Boc2O, DCM (b + c 21–33%); (d) TFA, DCM (95–99%); (e) Ile-Boc
(8a–h) or allo-Ile-Boc (8i–l) TBTU, Et3N, DCM (38–80%; (f) TFA, DCM (95–99%).
96
8
>100
>100
>50
>100
82.8 8.7
a
X = N (14), X = CH (8a–h, 20a–c).

S. Van Goethem et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4159–4162
4161
Table 2
ent at the 5-position (8d, 8e, 8j). Further, a general trend in these
data consists of a decrease in potency toward DPP8 and 9 with
increasing substituent sizes for both DPP8 and DPP9, but not for
DPP II. Halogenated inhibitors 8d, 8h, and 20a might serve as illus-
trative examples thereof. The iodinated isoindoline residue in 20a
can even be envisaged as a novel, selectivity conferring building
P1-block for the development of dipeptide-derived DPP II
inhibitors.
The sterically most demanding t-butyl and azide substituents
(8f and 20c), however, do not seem to be accommodated efficiently
by DPP II either. Next, allo-isoleucine clearly outperforms isoleu-
cine as a P2 fragment in inhibitors of both DPP8 and 9: upon com-
paring, for example, compound 8a (Table 1) with Ref.1 (Table 2), a
10-fold increase in DPP8/9 potency and selectivity toward DPP II
can be observed. Finally, the most active compound prepared in
this part of the study was found to be 8j, bearing a P1 5-fluoroiso-
indoline. Although its inhibitory potential toward DPP8 is similar
to that of parent structure 1, there is a modest, fourfold increase
in potency toward DPP9 compared to the latter. In addition, the
presence of a fluorine atom might lead to improved biopharmaceu-
tical properties. Nevertheless optimization of the P2 residue is re-
quired to obtain inhibitors which are able to discriminate between
DPP8 and DPP9.
IC50 values of substituted isoindolines (1, 8i–l)
Compound
R
IC50 (lM)
DPP8
DPP9
DPP IV
90
>100
>100
>100
>100
DPP II
29
7.1 0.8
21.5 3.2
47 4.4
1
H
4-F
5-F
5-CH3
5-CF3
0.12 0.01
0.8 0.1
0.16 0.016
2.8 1.4
>100
0.29 0.02
1.2 0.05
0.07 0.04
1.1 0.1
>100
4
1
8i
8j
8k
8l
12.6 2.5
with TBTU was followed by acidolytic cleavage of the Boc-func-
tionality, furnishing target compound 14.
Scheme 3 summarizes the synthesis of the P1 5-nitro-substi-
tuted isoindoline residue, obtained by regioselective nitration of
N-protected-isoindoline using nitronium tetrafluoroborate.14
Hydrogenation of 16 followed by iodination or transfer azidation
with sodium azide resulted in compounds 18 and 19, respec-
tively.15,16 After deprotection, the isoindoline derivatives were
elaborated into final products 20 using identical protocols as
shown for intermediates 6.
Compounds 21, 22, and 23 bearing a substituted lysine or a
substituted 4-aminoproline at the P2 position are listed in Tables
3–5. These inhibitors were synthesized by coupling isoindoline to
substituted lysines (21) or substituted 4-amino-prolines (22–23),
respectively. Naphthyl and benzyl substituents were introduced
relying on a reductive amination protocol optimized earlier.18
Substituted 4-aminoproline building blocks were synthesized fol-
lowing a literature procedure.19
All inhibitors were tested for their inhibitory activity against
DPP II, DPP IV, DPP8, and DPP9.17
Tables 1 and 2 document the SAR of inhibitors containing 4- or
5-substituted isoindolines as the P1 fragment (8a–l, 20a–c). In gen-
eral, the compounds tested showed very limited affinity for DPP IV,
comparable to what was observed for reference compound 1. In
terms of DPP8/9 potency, introduction of a 4-substituent (8b, 8c,
8i) proved to be less favorable than introducing the same substitu-
As discussed in our preceding paper the side chains of Ile, allo-
Ile, Lys, and Lys(Z) can be selected as useful P2 fragments for dipep-
Table 3
IC50 values of substituted lysine isoindolines (21a–f)
Compound
R1
R2
IC50
(l
M)
DPP8
DPP9
DPP IV
161
245 85
>50
>62.5
178 14
>31.25
DPP II
21a
21b
21c
21d
21e
21f
H–
H–
Bn–
Bn–
2-Naphthyl
2-Naphthyl
H–
Z–
H–
Bn–
H–
2-Naphthyl
0.20 0.01
0.20 0.02
0.111 0.006
0.666 0.064
0.043 0.002
0.479 0.045
0.5 0.03
n.a.
0.2 0.006
2.39 0.14
0.169 0.014
3.67 0.27
8
1.78 0.08
2.5 0.3
0.117 0.007
0.365 0.021
0.316 0.025
0.416 0.021
Table 4
IC50 values of substituted 4-aminoproline isoindolines (22a–e)
Compound
R1
R2
IC50
(l
M)
DPP8
DPP9
DPP IV
DPP II
22a
22b
22c
22d
22e
Bn–
H–
H–
1-Naphthyl
H–
2-Naphthyl
2.9 0.2
0.71 0.03
13.1 1.3
>250
4.97 0.45
n.i.
>100
n.a.
3.4 0.4
0.86 0.02
1.7 0.3
1.3 0.2
>100
1-Naphthyl
1-Naphthyl
2-Naphthyl
2-Naphthyl
41
3
12
2
16.1 1.1
5.77 0.34
6.69 0.6
0.91 0.1
10.2 1.4

4162
S. Van Goethem et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4159–4162
Table 5
IC50 values of substituted 4-aminoproline isoindolines (23a–d)
Compound
R1
R2
IC50 (lM)
DPP8
DPP9
DPP IV
DPP II
23a
23b
23c
23d
H–
Bn–
Bn–
2-Naphthyl
H–
H–
Bn–
H–
24 0.6
5.9 0.3
13 0.7
67.8 3.7
29.8 2.38
>25
>500
>1000
>500
>250
41 2.2
3.5 0.6
4.5 0.9
0.38 0.04
0.29 0.02
2.65 0.27
tide-derived DPP8 or DPP9 inhibitors.9 However, substitution of
allo-Ile in compound 1 for lysine (21a) or Z-protected lysine
(21b) is shown not to be beneficial, as it reduces selectivity toward
this article can be found, in the online version, at doi:10.1016/
j.bmcl.2008.05.079.
DPP II. Introduction of mono- or di- e-substitution such as benzyl
(21c,d) or naphthyl (21e,f) increased potency for DPP8/9; however,
References and notes
significant DPP II inhibition was also observed.
1. Van der Veken, P.; Soroka, A.; Brandt, I.; Chen, Y.-S.; Maes, M.-B.; Lambeir,
A.-M.; Chen, X.; Haemers, A.; Scharpe, S.; Augustyns, K.; De Meester, I. Curr.
Top. Med. Chem. 2007, 7, 621.
2. Lee, H.-J.; Chen, Y.-S.; Chou, C.-Y.; Chien, C.-H.; Lin, C.-H.; Chang, G.-G.; Chen, X.
J. Biol. Chem. 2006, 281, 38653.
3. Rosenblum, J. S.; Kozarich, J. W. Curr. Opin. Chem. Biol. 2003, 7, 496.
4. Lambeir, A.-M.; Durinckx, C.; Scharpé, S.; De Meester, I. Crit. Clin. Lab. Sci. 2003,
40, 194.
5. Bjelke, J. R.; Christensen, J.; Nielsen, P. F.; Branner, S.; Kanstrup, A. B.;
Wagtmann, N.; Rasmussen, H. B. Biochem. J. 2006, 396, 391.
6. Lankas, G. R.; Leiting, B.; Roy, R. S.; Eiermann, G. J.; Beconi, M. G.; Biftu, T.; Chan,
C. C.; Edmondson, S.; Feeney, W. P.; He, H. B.; Ippolito, D. E.; Kim, D.; Lyons, K.
A.; Ok, H. O.; Patel, R. A.; Petrov, A. N.; Pryor, K. A.; Qian, X. X.; Reigle, L.; Woods,
A.; Wu, J. K.; Zaller, D.; Zhang, X. P.; Zhu, L.; Weber, A. E.; Thornberry, N. A.
Diabetes 2005, 54, 2988.
7. Rosemblum, J. S.; Minimo, L. C.; Liu, Y.; Wu, J.; Kozarich, J. W. Abstract of Posters,
National Meeting of the ADA; Diabetes, 2007; Abstract 138.
8. Schade, J.; Stephan, M.; Schmiedl, A.; Wagner, L.; Niestroj, A. J.; Demuth, H.-U.;
Frerker, N.; Klemann, C.; Raber, K. A.; Pabst, R.; von Hörsten, S. J. Histochem.
Cytochem. 2008, 56, 147.
9. Van der Veken, P.; De Meester, I.; Dubois, V.; Soroka, A.; Van Goethem, S.; Maes,
M.-B.; Brandt, I.; Lambeir, A.-M.; Chen, X.; Haemers, A.; Scharpé, S.; Augustyns,
K. Bioorg. Med. Chem. Lett. 2008, 18, 4154.
As depicted in Tables 4 and 5 substitution of the allo-Ile moiety
by a 4-aminoprolyl residue afforded some compounds with high
nM activity against DPP II (22b and 23d). Unfortunately the activ-
ity against DPP9 decreased about 10 times for the most active com-
pound 23d compared to 1. No correlation between conversion in
stereochemistry at the 4-amino group and change in potency or
selectivity was observed.
In conclusion, fluorinated compound 8j, (2S,3R)-2-amino-1-(5-
fluoroisoindolin-2-yl)-3-methylpentan-1-one was found to exhibit
the best balance of potency and selectivity. The introduction of sub-
stituents larger than fluorine results in a negative effect on potency
and selectivity. Further optimization of the P2 position might pro-
vide selective DPP8/9 inhibitors that will be able to discriminate be-
tween these two closely related dipeptidyl dipeptidases. Finally, this
study attests the potential of substituted isoindolines as useful
building blocks for structurally novel, selective DPP II inhibitors.
10. Noyes, W. A.; Porter, P. K. Org. Synth. 1941, 1, 457.
11. Gawley, R. E.; Chemburkar, S. R.; Smith, A. L.; Anklekar, T. V. J. Org. Chem. 1988,
53, 5381.
Acknowledgments
12. Gadekar, S. M.; Frederick, J. L.; Semb, J.; Vaughan, J. R. J. Am. Chem. Soc. 1961, 26,
468.
13. Barnes, R. A.; Godfrey, J. C. J. Am. Chem. Soc. 1957, 22, 1043.
14. Dwyer, C. L.; Holzapfel, C. W. Tetrahedron 1998, 54, 7843.
15. Krasnokutskaya, E. A.; Semenischeva, N. I.; Filimonov, V. D.; Knochel, P.
Synthesis 2007, 1, 81.
Pieter Van der Veken is indebted to the Flemish Fund for Scien-
tific research (FWO-Vlaanderen) for financial support. We are
grateful to Mrs. Nicole Lamoen and Mr. Willy Bollaert for their
excellent technical assistance.
16. Mendenhall, G. D.; Smith, A. S. Org. Synth. 1973, 5, 829.
17. Dubois, V.; Lambeir, A.-M.; Van der Veken, P.; Augustyns, K.; Creemers, J.; Chen,
X.; Scharpé, S.; De Meester, I. Front. Biosci. 2008, 13, 3558.
18. Senten, K.; Van der Veken, P.; De Meester, I.; Lambeir, A.-M.; Scharpé, S.;
Haemers, A.; Augustyns, K. J. Med. Chem. 2004, 47, 2906.
19. Curran, T. P.; Marques, K. A.; Bilimoria, S. I. Lett. Org. Chem. 2005, 2, 15.
Supplementary data
Enzymatic assay conditions are described in the supplementary
information of this article. Supplementary data associated with