Design, synthesis and biological activity of original pyrazolo-pyrido- diazepine, -pyrazine and -oxazine dione derivatives as novel dual Nox4/Nox1 inhibitors

Article abstract of DOI:10.1016/j.bmc.2011.10.016

Pyrazolo-pyrido-diazepine, -pyrazine and -oxazine dione derivatives are new chemical entities with good and attractive druglikeness properties. A series of pyrazolo-pyrido-diazepine dione analogs demonstrated to be particularly amenable to lead optimizati

Full text of DOI:10.1016/j.bmc.2011.10.016

Bioorganic & Medicinal Chemistry 19 (2011) 6989–6999
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Design, synthesis and biological activity of original pyrazolo-pyrido-diazepine,
-pyrazine and -oxazine dione derivatives as novel dual Nox4/Nox1 inhibitors
Francesca Gaggini, Benoît Laleu, Mike Orchard, Laetitia Fioraso-Cartier, Laurène Cagnon,
Sophie Houngninou-Molango, Angelo Gradia, Guillaume Duboux, Cédric Merlot, Freddy Heitz,
⇑
Cédric Szyndralewiez, Patrick Page
Genkyotex S.A., 16 Chemin des Aulx, CH-1228 Plan-Les-Ouates, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
Pyrazolo-pyrido-diazepine, -pyrazine and -oxazine dione derivatives are new chemical entities with good
and attractive druglikeness properties. A series of pyrazolo-pyrido-diazepine dione analogs demonstrated
to be particularly amenable to lead optimization through a couple of cycles in order to improve specificity
for isoforms Nox4 and Nox1 and had excellent pharmacokinetic parameters by oral route. Several mol-
ecules such as compound 7c proved to be highly potent in in vitro assays on human lung fibroblasts dif-
ferentiation as well as in curative murine models of bleomycin-induced pulmonary fibrosis with superior
efficiency over Pirfenidone. Pyrazolo-pyrido-diazepine dione derivatives targeting Nox4 and Nox1 iso-
forms appear highly promising therapeutics for the treatment of idiopathic pulmonary fibrosis.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 24 May 2011
Revised 4 October 2011
Accepted 7 October 2011
Available online 17 October 2011
Keywords:
Pirfenidone
IPF
NADPH oxidase
Nox4
Nox1
Inhibitor
Pyrazolo-pyrido-dione
Antifibrotic
1. Introduction
in isolated sheep liver microsomes in a dose-dependent manner.
While Pirfenidone is ineffective as a scavenger of superoxide radi-
Idiopathic pulmonary fibrosis (IPF) is a chronic progressive
parenchymal lung disease that is characterized by the formation
of scar tissue within the lungs in the absence of any known cause,
and which has a median survival of around 3 years once the condi-
tion has been diagnosed but the clinical course can be highly vari-
able.1,2 Pirfenidone is currently the only therapy specifically
approved in Japan and recently in Europe for the treatment of idio-
pathic pulmonary fibrosis. For many years, the symptoms of the
disease were often treated with anti-inflammatory drugs, despite
the treatment not producing significant benefit against progression
of the disease. One mechanism underlying both the anti-inflamma-
tory and the antifibrotic effects of Pirfenidone may be its ability to
suppress the induction and release of inflammatory cytokines. Oxi-
dative stress also enhances the release of cytokines from inflamma-
tory cells. Pirfenidone inhibits NADPH-induced lipid peroxidation
cals, it is a potent scavenger of hydroxyl radicals. The human ther-
apeutic dose of Pirfenidone used in clinical trials for the treatment
of IPF is very high, ranging from 1400 to 2800 mg per day. These
high doses are more in favor with a scavenging-based action of
Pirfenidone than a targeted drug mechanism of action. Thus, the
anti-fibrotic effects of Pirfenidone may be due at least in part to a
reduction of oxidative stress induced by toxic hydroxyl radicals.3
Increased oxidative stress or increased reactive oxygen species
(ROS) production has been implicated in a wide variety of pathol-
ogies, including idiopathic pulmonary fibrosis (IPF), kidney fibrosis,
diabetic nephropathy, atherosclerosis and neurodegenerative dis-
eases. The relationship between disease progression and over-
expression or activation of specific sub-units of Noxes has been
shown in many pathologies encouraging the recent development
of Nox specific inhibitors.4
Amongst the Nox family, the NADPH oxidase isoform 4 (Nox4)
has gained considerable and growing attention during the past
years. Increasing scientific evidence points to Nox4 as the key
source of reactive oxygen species (ROS) in the pathogenesis of idi-
opathic pulmonary fibrosis. It has been shown that Nox4 is upreg-
ulated in both mouse models of IPF and in lung fibroblasts of
human IPF patients. Moreover, genetic or pharmacologic silencing
Abbreviations: ADMET, absorption, distribution, metabolism, excretion and
toxicity; FITC, fluorescein isothiocyanate; hERG, human ether-a-go-go-related gene;
IPF, idiopathic pulmonary fibrosis; NADPH, nicotinamide adenine dinucleotide
phosphate; ROS, reactive oxygen species.
⇑
Corresponding author.
E-mail address: patrick.page@genkyotex.com (P. Page).
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.10.016

6990
F. Gaggini et al. / Bioorg. Med. Chem. 19 (2011) 6989–6999
of Nox4 abrogates fibrogenesis in two murine (bleomycin-induced
or FITC-induced) models of lung injury.5 These studies indicate a
crucial role of Nox4 in tissue fibrogenesis and provide proof of con-
cept for therapeutic targeting of Nox4 in fibrotic disorders.
Recently, we have published a large part of our investigation
regarding a new class of molecules so-called pyrazolo-pyridine
diones targeting the NADPH oxidase isoforms 4 (Nox4) and 1
(Nox1).6 In this Letter, we would like to complete our work by
highlighting sub-chemical series and more specifically new chem-
ical entities so-called pyrazolo-pyrido-diazepine dione derivatives
which appear highly promising therapeutics for the treatment of
idiopathic pulmonary fibrosis.
In general, functionalized diamines 2 (R = CH2Ar) were com-
mercially available or prepared through reductive amination.8
Starting from commercially available aromatic aldehydes and
1,3-propane diamine, the corresponding secondary amines 2 were
obtained in good to moderate yields. Isolation of compound 4 with
R = Ar was successfully achieved and its cyclization afforded tricy-
clic compounds 6 which gave compounds 7 after HCl treatment.
Instead, the synthesis of the piperazine analogs (n = 1 and
R = H) failed: different reaction conditions and methodologies
were used, but unsuccessfully. For example, any attempt of sta-
bilizing the NH-intermediates at all levels (type 4 as NH or 5 as
NBoc) gave messy reactions. No improvement was noticed by
using the mono Boc-ethylenediamine: chlorine displacement
never took place using different reaction conditions such as
tBuOK/THF or NaH/THF.
We suspected that the difficulties encountered for the synthesis
and isolation of these piperazine analogs might be due to their po-
tential instability. Indeed this was confirmed when analog 6 (n = 1,
R = CH2Ph) was finally made in only 11% yield after tremendous ef-
forts, and found to have poor stability.
A slightly different synthesis was optimized when ethanolamine
was used (pathway B, Scheme 1). Our starting building block 1 re-
acted with ethanolamine 8 in acetonitrile at 0 °C, as previously de-
scribed, allowing the isolation of the intermediate analog 9. This
enamine was directly converted into the final compound 11: the
nucleophilic attack of the alcohol was promoted by NaH/THF and
the morpholine–enamine 10 generated the pyrazolo-pyrido oxa-
zine dione derivatives in situ (pathway B, Scheme 1).
2. Results and discussion
2.1. Chemistry
Our further exploration of the pyrazolo-pyridine dione class7 is
described in the Scheme 1: in general, the starting material 1 as
chloroacetyl derivatives was easily obtained following the proce-
dure already published on the family of pyrazolo-pyridine dione
derivatives.6 Pyrazolone 1 was subsequently reacted in acetonitrile
at 0 °C with different primary diamines 2: a variety of final com-
pounds were generated according to carbon length (n = 1 and 2)
or substituents (R = H and Ar). Following the pathway A, the inter-
mediate 3 was never isolated (R = H,CH2Ar and n = 1,2) as it spon-
taneously gave the cyclized enamine 4 in situ. When 1,3-propane
diamines were used (n = 2), a variety of pyrazolo-pyrido-diazepine
were obtained. In particular, if R = H, the enamine intermediate 4
was trapped with Boc anhydride, yielding the resulting compound
5 that was easier purified than its NH precursor. The following
cyclization carried out in basic conditions formed the tricyclic
compound 6 and the Boc-deprotection was successfully achieved
with 4 M HCl solution, affording 7 as HCl salt.
2.2. Biological results
Based on our previous work identifying the need of specific Nox
inhibitors,6,7 we investigated a new class of molecules whose
pyrazolo-pyridine dione core is fused with another ring at the
northern part. The six- or seven-member rings closure was ex-
R
N
H
R
R
N
R
N
N
n
N
O
O
O
n
O
n
n
H
N
(d)
(c)
H
Cl
N
H
N
N
Ar
N
N
R
Ar
N
Ar
N
Ar
N
N
n
2
.HCl
N
N
N
H
O
O
H
H
CO2Me
CO2Me
(a)
A
7
4
6
3
OH
6a:
6b-o:
R = Boc
n = 2
7a:
R = H
n = 2
O
4a:
R = H
n = 2
R = CH2Ar n = 1, 2 7b-o: R = CH2Ar n = 1, 2
4b-n: R = CH2Ar n = 1, 2
Ar
Cl
(b)
N
N
5a:
R = Boc
n = 2
MeO2C
1
1a: Ar = o-OMe-phenyl
1b: Ar = o-Cl-phenyl
B
OH
Cl
O
O
N
(a)
N
O
OH
O
O
H2N
(e)
N
O
8
Ar N
Ar
N
Ar
N
N
H
N
N
CO2Me
CO2Me
11
9
10
Scheme 1. Reagents and conditions: (a) DIPEA, CH3CN, 0 °C-RT; (b) Boc2O, NaHCO3, MeOH; (c) MeOH/MeONa, (d) 4 M HCl, DCM; (e) NaH, THF, 0 °C-RT.

F. Gaggini et al. / Bioorg. Med. Chem. 19 (2011) 6989–6999
6991
pected to rigidify molecules leading to a diminished number of
conformation potentially improving affinity and some ADME
parameters such as permeability or metabolism. All compounds
were evaluated using our cell free assay of ROS production on hu-
man Nox membranes (Nox1, Nox2, Nox4 and Nox5 isoform inhibi-
tion data are reported in Table 1).
Encouragingly, the picolyl derivatives were found to be very potent
on hNox4, particularly in the case of ortho-substitution (Ki = 67 nM
for 7m). The beneficial effect of this o-picolyl group might be seen
as well with the slight improvement in potency observed when
moving from compound 7d (Ki = 144 nM) to 7n (Ki = 93 nM).
Among the selected compounds, the most active on Nox2 ap-
peared to be the one deprived of substituent on the northern part
of the molecule (e.g., 11 and 7a with Ki = 433 and 400 nM, respec-
tively). Besides having an o-methoxyphenyl-pyrazolone group on
the western part of the molecule seemed to increase the potency
on Nox2 (7d, 7n and 7o). All the compounds fall in the same range
regarding Nox1 potency (Ki ꢀ87–398 nM) with benzylchloro
diazepine derivatives being the most active along with 7c and
the one harboring an o-methoxyphenyl-pyrazolone group the least
potent molecules. Concerning Nox5, having a heteroaromatic ring
attached to the diazepine moiety brought a detrimental effect
(furan, pyridines) whereas molecules without northern substituent
were found to be the most active on this Nox isoform.
Similarly to pyrazolo pyridine dione derivatives described in
our previous paper,6 Nox1 and Nox4 functional potency seemed
also to be correlated in the pyrazolo-pyrido-diazepine, -pyrazine
and -oxazine dione chemical series, which is most likely due to
key common interactions related to the central core. Nox2/Nox4
and Nox5/Nox4 selectivity appeared to be acceptable for most
tested compounds. As a result these derivatives present relatively
similar Nox profiles and appeared to be dual Nox4/Nox1 inhibitors
with selectivity over the isoform Nox2. A counter-screening assay
Xanthine/Xanthine oxidase was performed on all the molecules
to rule out potential scavenging properties and proved to be nega-
tive for all of them.
First of all, the pyrazolone substituent was mostly restricted to
the o-chlorophenyl group toconcentrate thestructural investigation
on the northern part of the molecule. Initially, neither the size of the
cycle attached (six- or seven-membered rings) to the pyridine dione
moiety, nor the heteroatom within this ring (n or O atom) appeared
to have a strong positive impact on the Nox4 or Nox1 specific activ-
ities when compared to other pyrazolo-pyridine dione derivatives.
Pyrazolo-pyridine dione derivatives with six-membered rings moi-
ety showed interesting functional activity on Nox4 and Nox1 with
active compounds7b and 11 having Ki (hNox4) ꢀ142–156 nM. Com-
pound 7b showed also trends for a better selectivity over Nox2
(Ki = 2015 nM) and Nox5 (Ki = 1425 nM). As already described, the
inherent chemical difficulties to access easily to pyrazolo-pyridine
derivatives with six-membered rings led us to reorganize our strate-
gic chemical efforts.
Consequently we focused on the seven-membered ring deriva-
tives, discovering several molecules with nanomolar inhibition,
showing high potency on Nox1 and Nox4. The influence of the
substituent on the pyrazolone moiety (left part of the structure)
was summarily investigated (replacing the o-chlorophenyl by an
o-methoxyphenyl group) since our previous results demonstrated
the importance of an ortho aryl substituent on the pyrazolone
moiety to improve metabolic stability.
However, the o-methoxyphenyl substitution resulted in a loss
of affinity for all Nox enzymes: the Ki values for Nox1 and Nox4
are higher (i.e., Ki (hNox4) = 144 nM for 7d vs 72 nM for 7c, Ki
(hNox1) = 218 nM for 7d vs 101 nM for 7c) but selectivity towards
Nox2 and Nox5 is increased. This trend seemed to be corroborated
when comparing compounds 7o and 7n with 7i and 7m, respec-
tively although the differences in activities remained very modest.
The o-chlorophenyl group was thus confirmed as the substituent of
choice in order to build a series of pyrazolo-pyrido-diazepine dione
analogs by varying the northern part of the structure, namely the
group attached to the nitrogen atom of the diazepine. Interestingly,
removal of the benzyl group attached to the nitrogen atom in com-
pound 7a appeared to diminish the selectivity towards Nox2 while
keeping a comparable Nox4 potency (Nox2/Nox4 selectivity ra-
tio = 4.2 vs 18.6 for 7a and 7c, respectively). The introduction of
a Boc protecting group in compound 6a restored the Nox2/Nox4
selectivity ratio to 12.4. Overall, increasing steric hindrance by
introducing lipophilic groups on the nitrogen of the seven-mem-
bered rings seemed to increment selectivity towards Nox2.
On the other hand, substitution by a chlorine atom in o-, m-, or
p-position of the benzyl group attached to the diazepine moiety re-
vealed to be well tolerated with Ki (hNox4) = 76, 101 and 103 nM
for 7e, 7f and 7g, respectively. These compounds had also pretty
acceptable selectivity over Nox2 and Nox5.
No significant difference was noted when incorporating an
o-, m-, or p-methoxybenzyl group with Ki (hNox4) in the range
of 111–116 nM but higher selectivity towards Nox2 and Nox5
for compounds 7h, 7i and 7j (i.e., 7h vs 7e; 7j vs 7g; and 7i vs
7f). This showed as well that the nature of the substituent on
the benzyl group is not paramount. Indeed no dramatic differ-
ences in potency result from the introduction of a lipophilic en-
tity or hydrogen bond acceptor.
Having found that various benzylic substitutents led to potent
Nox4/Nox1 inhibitors, the introduction of heterocyclic moieties
was then considered for the northern diazepine part of the mole-
cule. Once again, this kind of substitution was tolerated: for in-
Further investigations did not allow to determine clear-cut opti-
mal substituent for the northern part of the molecule, the most ac-
tive compounds were then subjected to in vitro profiling, in
particular with respect to oxidative metabolism in human and
rat liver microsomes, Caco-2 cellular permeation model as well
as plasma protein binding assay (see Table 2). Compound 7a was
not selected because of some inappropriate physicochemical prop-
erties such as a very low logD = ꢁ0.38, predicting the compound to
be eliminated with a high renal clearance.
Good correlation between logD and metabolic clearance was
observed for all tested compounds. Most compounds were fairly
resistant to oxidative first-pass metabolism, demonstrating med-
ium clearance for both species studied: 635.0 and 664.0 lL/min/
mg protein in human and rat liver microsomes, respectively. How-
ever, notable exceptions were the three derivatives with a chloro-
benzyl group attached to the diazepine part: compounds 7e, 7f and
7g. Their low microsomal stability is presumably due to their high-
er lipophilicity (logD = 2.76) that renders them liable to phase I
oxidative metabolism, regardless of their substitution (o-, m- and
p-Cl).
Concerning the in vitro permeability in Caco-2 experiments,
compounds 7l, 7m and 7n harboring a pyridine group revealed
to have weak permeability properties associated with a high efflux
(Papp ꢀ1.0–1.8 ꢂ 10ꢁ6 cm/sec and efflux ratios ꢀ10.1–18.7). This
can be explained by the relatively low logD (<1.1) associated to
these compounds and also by the possible protonation of the basic
pyridine nitrogen atom, as second basic center in these molecules,
which can preclude them to cross readily the lipophilic mem-
branes. We have not investigated whether these compounds might
be substrate of transporters potentially providing explanation of
the observed efflux. Increasing the lipophilicity of the diazepine
substituent with compound 7i by 1.5–2.0 logP unit led to a suitable
permeability with a very modest efflux ratio. Interestingly, the
physicochemical properties of 7c allowed us to reach a high per-
meability range for this molecule (Papp 42.2 ꢂ 10ꢁ6 cm/sec) with
no efflux at all.
stance, Ki (hNox4) = 111 nM for 7k bearing
a furyl group.

6992
F. Gaggini et al. / Bioorg. Med. Chem. 19 (2011) 6989–6999
Table 1
Nox profile of compounds 6a, 7a–o and 11
Compound
Ki Nox1 (nM)a
Ki Nox2 (nM)a
Ki Nox4 (nM)a
Ki Nox5 (nM)a
O
O
N
O
N
148 18
1280 25
103 12
590 100
N
N
O
H
Cl
Cl
6a
7a
H
N
O
N
119 20
365
5
87
8
360 74
N
N
H
O
N
N
O
398 63
2015 155
142 20
1425 175
N
N
H
O
O
O
O
Cl
Cl
O
7b
7c
7d
N
O
101 10
1340 270
72
3
414 52
670 30
615 175
N
N
N
H
N
O
218 39
645 75
144 22
N
N
N
H
Cl
N
O
87
4
1160 340
76
3
N
N
N
H
Cl
7e
Cl
N
104
4
1150 500
101
7
475 65
O
N
N
N
H
O
Cl
7f
Cl
N
95
6
1260 490
103 10
585 225
O
N
N
N
H
O
Cl
7g

F. Gaggini et al. / Bioorg. Med. Chem. 19 (2011) 6989–6999
6993
Table 1 (continued)
Compound
Ki Nox1 (nM)a
Ki Nox2 (nM)a
Ki Nox4 (nM)a
Ki Nox5 (nM)a
O
N
133 15
1760 200
111 10
740 90
O
N
N
N
O
H
Cl
7h
O
N
113 21
1195 255
112 17
730 100
O
N
N
N
O
H
Cl
7i
O
N
184 24
780 170
116
9
690 10
O
N
N
N
O
H
Cl
7j
O
N
O
200 31
1020 190
111
6
7
3
955 145
1150 170
543 26
N
N
N
O
H
Cl
7k
N
N
208 46
1370 80
94
O
N
N
N
O
O
O
H
Cl
7l
N
N
O
113
1
1140 150
67
N
N
N
H
Cl
7m
N
N
O
217 18
510 20
93 11
610 40
N
N
N
H
O
7n
(continued on next page)

6994
F. Gaggini et al. / Bioorg. Med. Chem. 19 (2011) 6989–6999
Table 1 (continued)
Compound
Ki Nox1 (nM)a
Ki Nox2 (nM)a
Ki Nox4 (nM)a
Ki Nox5 (nM)a
O
N
212 51
820 150
131
6
690 60
O
N
N
N
O
O
H
O
7o
O
N
O
173 14
433 19
156 11
490 67
N
N
H
Cl
11
a
Cell free assay of ROS production on membranes over-expressing human Nox.
Table 2
In vitro profiling for selected pyrazolo-pyrido-diazepine dione derivatives
Compound
Cl (h-LM)a
lL/min/mg protein
Cl (r-LM)b
l
L/min/mg protein
Caco-2 (Papp)c(10ꢁ6 cm/sec)
Efflux ratio
logDd
Fue
7c
7e
7f
7g
7i
7l
7m
7n
35.0
87.1
81.4
77.3
35.0
21.0
13.9
15.7
64.0
164.0
153.0
152.0
60.1
31.1
27.5
32.1
42.2
—
—
0.48
—
—
2.23
2.76
2.76
2.76
1.99
0.92
1.10
0.33
12.0
—
—
—
—
—
22.3
1.0
1.8
1.4
1.91
18.7
10.1
16.8
90.1
2.9
2.6
6.9
a
b
c
Microsome stability experiment. h-LM, human liver microsomes.
Microsome stability experiment. r-LM, rat liver microsomes.
Caco-2 cell permeability assay. Papp, permeability coefficient.
Calculated at pH 7.4.
d
e
Fraction unbound, plasma protein binding assay in rat (in 50% plasma protein).
The results from these in vitro microsome stability and Caco-2
Remarkably, the pharmacokinetic profile of 7m was reasonably
good despite the questions raised by its relatively poor in vitro
permeability properties and a slightly low logD (1.10). This
compound presents indeed acceptable in vivo clearance, plasma
concentrations and oral bioavailability (FZ = 42%) and acceptable
volume of distribution into tissues (Vss = 0.47 L/kg after iv dosing).
Even more interestingly, compound 7c showed higher exposure pre-
sumably due to its good physicochemical properties and higher per-
meability ranking. Moreover this compound has a suitable oral
bioavailability (FZ = 53%) associated with a relatively low in vivo
clearance in line with in vitro data. Half-life of 2.71 and 2.37 h were
observed after intravenous and per os administrations, respectively,
as well as limited but acceptable distribution into tissues
(Vss = 0.28 L/kg).
The Nox specificity of 7c was confirmed in a counter-screening
assay for potential ROS scavenging (Table 4) and no potential prob-
lem was found regarding CYP450 inhibition. This molecule was not
mutagenic nor genotoxic according to mini Ames and in vitro
micronucleus tests. It showed as well low probability to generate
QT prolongation based on lack of inhibition of hERG. A NOAEL
(No Observed Adverse Effect Level) of 1000 mg/kg was determined
in mice after 2 weeks daily oral administration. Consequently, this
excellent early in vitro ADMET profile confers to this compound a
very large safety window.
permeability studies encouraged us to avoid chlorobenzyl or picolyl
groups attached to the diazepine moiety but using a simple benzyl
or methoxybenzyl group.
Concerning the in vitro plasma protein binding in rat species,
similar results were obtained for the very close structural analogs
7l and 7m: the free fractions recorded were 2.6% and 2.9%,
respectively in 50% plasma protein. Compound 7n had a higher
free fraction (6.9%) presumably due to its lower logD (0.33). On
the other hand, the unbound fraction of 7i revealed to be partic-
ularly high (90.1%) despite its average logD (1.99). Associated
with a medium metabolic clearance, this data did not encourage
us to push this compound forward. Fortunately, its analog 7c with
a logD in the same range (2.23) showed a reasonable free fraction
(Fu ꢀ 12%).
Finally, it is noteworthy to mention that no problem was
encountered regarding CYP inhibitions for all-tested compounds
having IC50 >25 lM (data not shown).
On the basis of the data gathered, two compounds were
selected for further investigation through their pharmacokinetic
properties in rat to help guide selection of a superior clinical can-
didate. In spite of disappointing in vitro results regarding Caco-2
permeability, compound 7m was chosen since it was the most po-
tent compound on hNox4 with no other problem spotted during its
in vitro profiling. Compound 7c embodied a good compromise
between microsomal metabolism, free fraction, permeability,
physical properties associated with high Nox4/Nox1 potency and
the highest Nox4/Nox2 selectivity ratio (18.6). The results are
combined in Table 3.
7c abrogates strongly lung fibrosis following a once-a-day oral
administration from days 10 to 25 after bleomycin intratracheal
administration in a curative model of bleomycin-induced pulmon-
ary fibrosis. Collagen deposition was inhibited by 64% at the dose
of 10 mg/kg po with 7c; the benchmarked Pirfenidone in the same

F. Gaggini et al. / Bioorg. Med. Chem. 19 (2011) 6989–6999
6995
Table 3
In vivo pharmacokinetic profile in rats for selected compounds
Compound
Route
Dose (mpk)
AUC (h ꢂ ng/mL)
Fz (%)
Cmax (ng/mL)
Tmax (h)
t1/2 (h)
Cl (L/kg/h)
Vss (L/kg)
1
7c
iv
po
iv
5
10
5
39266
41430
11819
9901
—
53
—
42336
14877
21827
11298
—
0.33
—
2.71
2.37
5.62
3.63
0.13
0.26
0.46
1.02
0.28
—
0.47
—
7m
po
10
42
0.25
Table 4
files led to the identification of the highly potent, orally bioavail-
able 7c. This compound benefits not only from a straightforward
synthetic procedure but also from an excellent pharmacological
and safety profile with Ki in the two digit nanomolar range on
Nox4. This compound proved to be highly potent in in vitro assays
on human lung fibroblast differentiation as well as in preventive
and curative murine models of bleomycin-induced pulmonary
fibrosis, where it demonstrates a superior efficiency to Pirfenidone
which is today the only drug accepted for the treatment of IPF in
Japan and Europe. 7c deserves clinical trials for the treatment of
idiopathic pulmonary fibrosis.
Early in vitro ADMET profile of compound 7c
IC50 (1A)a
>25
>25
>25
>25
>25
l
l
l
l
l
M
M
M
M
M
IC50 (2C19)a
IC50 (2C9)a
IC50 (2D6)a
IC50 (3A4)a
Ki Xanthine oxidaseb
hERGc
>30000 nM
8.7% at 1
Negative
Negative
lM
Mini AMES testd
In vitro micronucleuse
a
Inhibition of cytochrome P450 isoforms (see Section 4).
In vitro enzymatic assay.
Conventional patch-clamp.
Performed @ 100
Performed @ 500
b
c
d
e
4. Experimental section
4.1. Chemistry
l
l
M.
M.
model partially inhibited collagen deposition by 12% only at the
dose of 100 mg/kg po (see Fig. 1). A rough estimation of Human
Equivalent Dose (HED) by conversion of animal doses, based on
body surface area as per the FDA guidance for industry 2002,
predicts an orally pharmacologically active dose in human of
50–60 mg once-a-day.
All solvents used in these reactions were anhydrous and
obtained from commercial sources. All other reagents were used
as supplied unless otherwise stated. All reactions with moisture
sensitive compounds were carried out under nitrogen. Melting
points were measured with a Büchi Melting Point B-540 appara-
tus and were uncorrected. NMR-spectra were recorded on a
Bruker AMX-500 and 400 MHz spectrometers. Chemical shifts
were expressed in parts per million (d) and referenced with re-
spect to the residual solvent signal (DMSO: 2.50 ppm). Data were
reported as follow: chemical shift (d), integration, multiplicity
(s = singlet, bs = broad singlet, d = doublet, t = triplet, dd = doublet
of doublets, ddd = doublet of doublet of doublets, dt = doublet of
triplets, td = triplet of doublets, q = quadruplet and m = multiplet)
and coupling constants (J) in Hertz. UPLC (Ultra Performance
Liquid Chromatography) data provided were obtained using a
Waters Acquity SQ detector operated in positive or negative
electrospray. Analytical HPLC data were obtained using a Waters
3. Conclusion
We have identified a novel family of highly potent dual
Nox4/Nox1 inhibitors with selectivity over the isoform Nox2. This
investigation was devoted to the pyrazolo-pyrido-diazepine dione
derivatives with particular focus on the diazepine moiety. A variety
of compounds showed excellent potency in our cell free assay of
ROS production on human Nox4/Nox1 membranes. Further
improvement of the physical properties and pharmacological pro-
2690 separation module (Alliance) equipped with
Extend C-18 column (50 ꢃ 4.6 mm, 1.8 m). Detection was
conducted at 254 nm and full scan at 210–400 nm with
photodiode array detector. The mobile phase was a linear gradi-
ent with a flow rate of 0.8 mL/min using a 10:90 A/B to 100:0 A/
B mixture over 4 min. Solvent A was 0.1% formic acid in water,
and solvent B was 0.1% formic acid in acetonitrile. Silica flash
and reverse phase chromatography were done using the Armen
a Zorbax
l
#
a
120
§
100
80
Spot Flash instrument equipped with silica SI60 15–40
lM and
RP18 25–40 cartridges, respectively. For TLC (thin layer
l
M
chromatography) alumina sheets from Merck coated with silica
gel 60 F254 were used. Reported yields were not optimized,
because of our requirement for purity rather than quantity of
product.
60
40
4.2. Synthesis
NaCl
Bleomycin Bleomycin Bleomycin
+CMC/Tw +CMC/Tw +Pirfenidone + Compound 7c
Synthesis of starting materials 1a: methyl [4-(chloroacetyl)-5-
hydroxy-1-(2-methoxyphenyl)-2,3-dihydro-1H-pyrazol-3-yl]acetate
and 1b: methyl [4-(chloroacetyl)-1-(2-chlorophenyl)-5-hydroxy-
1H-pyrazol-3-yl]acetate) were obtained according to the general
procedure described in our previous publication.6
100mg/kg-po 10mg/kg-po
Figure 1. Curative model of Bleomycin for compound 7c. (a) Values are expressed
as mean SEM, #p <0.001 compared sham/control bleomycin, §p <0.05 compared
compound 7c/control bleomycin n = 8–13 animals.

6996
F. Gaggini et al. / Bioorg. Med. Chem. 19 (2011) 6989–6999
4.2.1. Synthesis of tert-butyl 2-(2-chlorophenyl)-1,5-dioxo-2,3,5,
8,9,11-hexahydro-1H-pyrazolo[40,30:3,4]pyrido[1,2-a][1,4]
diazepine-10(7H)-carboxylate (6a)
trated in vacuo until dryness. The residue was dissolved in
ethylacetate (10 mL), washed with saturated NaHCO3 solution
(5 mL ꢄ 2), dried over Na2SO4 and concentrated. The crude prod-
uct was used in the following step without further manipulation
(530 mg, 71% yield, 86% HPLC purity, LC–MS (M+H)+: 453.4).
The obtained enamine type 4 (530 mg, 1.2 mmol) was treated
with freshly prepared MeONa in MeOH, obtained from dissolution
of Na (55 mg, 2.4 mmol) in 3 mL of MeOH under N2. The solution
was stirred at room temperature until disappearance of the starting
enamine (1 h). The reaction mixture was concentrated in vacuum to
eliminate MeOH and the crude was dissolved in ethylacetate
(20 mL), extracted with water (20 mL ꢄ 3). Then the combined inor-
ganic layer was acidified to pH 6, extracted with ethylacetate
(30 mL ꢄ 3) and the combined organic layers were dried over
Na2SO4. The crude was purified by TLC yielding a beige solid
(142 mg, 28% yield, 97% HPLC purity).
The product 6 was dissolved in 2 mL MeOH and treated with 4 M
HCl in MeOH (0.082 mL). Then MeOH was removed, the residue was
triturated with Et2O. The combined yellow powder was dried in
vacuo (103 mg, HPLC: 97%). Mp 198–200 °C. 1H NMR (DMSO-d6,
400 MHz) d: 10.71 (bs, 1H), 7.66–7.62 (m, 1H), 7.59–7.55 (m, 1H),
7.49–7.47 (m, 2H), 7.32–7.25 (m, 4H), 7.25–7.21 (m, 1H), 5.73 (s,
1H), 4.70–4.55 (m, 2H), 4.55–4.39 (m, 2H), 3.64 (s, 2H), 2.83 (t,
J = 4.34 Hz, 2H), 1.69–1.61 (m, 2H). 13C NMR (DMSO-d6, 125 MHz)
d: 161.2, 159.8, 151.6, 151.4, 144.1, 133.9, 131.7, 131.3, 130.8,
130.3, 130.0, 129.6, 129.5, 128.9, 128.2, 103.6, 91.0, 56.7, 52.4,
48.6, 47.6, 40.6, 22.4. LC–MS (M+H)+: 421.1.
Pyrazolone 1b (0.94 mmol, 322 mg,) was dissolved in acetoni-
trile (5 mL) and cooled down to 0 °C. Propane 1,3-diamine
(0.85 mmol, 63 mg) was slowly added and the reaction was
stirred at 0 °C for 1 h. The solvent was concentrated in vacuo
and the crude was portioned between ethylacetate and a satu-
rated solution of NaHCO3. The combined organic phases were
dried over Na2SO4 and concentrated in vacuo yielding dark oil
which was used in the following step without further purification.
LC–MS (M+H)+: 363.1.
The oil deriving from the previous step was dissolved in
methanol, cooled down to 0 °C and NaHCO3 (71 mg, 0.85 mmol)
was added. Boc2O (185 mg, 0.85 mmol) was dissolved in dioxane
and dropwise added to the enamine at 0 °C. The reaction
mixture was stirred at room temperature overnight. Then the
solvent was concentrated in vacuo and the crude was filtered
through silica plug (10% petroleum ether/90% ethylacetate/0.1%
triethylamine) yielding 142 mg of a dark yellow solid LC–MS
(M+H)+: 463.2
Bocylated compound (142 mg, 0.31 mmol) was dissolved in
methanol and treated with a solution freshly prepared of MeOH/
MeONa (15 mg, 0.62 mmol of Na in 1 mL of MeOH). The resulting
solution was stirred at room temperature for 30 min. UPLC analysis
showed the cyclization went to completion. It was diluted with
H2O, neutralized with 1 M HCl and extracted with ethylacetate.
The organic layer was dried over Na2SO4 and concentrated in
vacuo. The crude was purified by silica gel chromatography (ethyl-
acetate/methanol gradient): final compound was obtained as a yel-
low solid (44 mg, 33% yield, 97% HPLC purity). 1H NMR (DMSO-d6,
500 MHz) d: 11.72 (bs, 1H), 7.66–7.60 (m, 2H), 7.51–7.40 (m, 3H),
5.66 (s, 1H), 5.19–5.11 (m, 2H), 4.50–4.41 (m, 2H), 3.64–3.55 (m,
2H), 1.98 (s, 9H), 1.72–1.66 (m, 2H). LC–MS M+H)+: 431.1.
Following the above procedure, and using the appropriate start-
ing materials, compounds 7d–7o were obtained.
4.3.2. 10-Benzyl-2-(2-methoxyphenyl)-2,3,8,9,10,11-hexahydro-
1H-pyrazolo[40,30:3,4]pyrido[1,2-a][1,4]diazepine-1,5(7H)-dione
(7d)
Isolated as a yellow solid (94% HPLC purity). 1H NMR (DMSO-d6,
400 MHz) d: 7.60–7.58 (m, 2H), 7.43–7.36 (m, 5H), 7.18–7.16
(d, J = 8 Hz, 1H), 7.04–7.00 (t, J = 7.2 Hz, 1H), 5.83 (s, 1H), 5.17 (s,
2H), 4.52 (s, 2H), 4.29 (s, 2H), 3.72 (s, 3H), 3.42 (s, 2H), 2.04 (s,
2H). LC–MS (M+H)+: 417.3.
4.2.2. Synthesis of 2-(2-chlorophenyl)-2,3,8,9,10,11-hexahydro-
1H-pyrazolo[40,30:3,4]pyrido[1,2-a][1,4]diazepine-1,5(7H)-dione
(7a)
The Boc protected compound 6a was dissolved in DCM and
4 M HCl in dioxane was added followed by one drop of water.
The reaction was stirred at room temperature for 30 min and
monitored by UPLC and HPLC. When the starting material was
completely consumed, the solvent was concentrated in vacuo
and the residue was precipitated from methanol and diethylether.
The yellow solid was filtered off and washed with small portions
of diethylether. The final compound was obtained as a yellow HCl
salt with 98% HPLC purity (91% yield). 1H NMR (DMSO-d6,
500 MHz) d: 11.07 (s,1H), 9.57 (BS, 1H), 7.70–7.67 (m, 1H),
7.57–7.54 (m, 1H), 7.53–7.49 (m, 2H), 5.89 (s, 1H), 5.04–4.98
(m, 2H), 4.61–4.51 (m, 2H), 3.47–3.37 (m, 2H), 1.95–1.88 (m,
2H). LC–MS (M+H)+: 331.1
4.3.3. 10-(2-Chlorobenzyl)-2-(2-chlorophenyl)-2,3,8,9,10,11-
hexahydro-1H-pyrazolo[40,30:3,4]pyrido[1,2-a][1,4]diazepine-
1,5(7H)-dione (7e)
Isolated as a brown solid (97% HPLC purity). 1H NMR (DMSO-d6,
400 MHz) d: 10.83 (bs, 1H), 7.63 (t, J = 2 Hz, 1H), 7.55–7.52 (m, 2H),
7.48–7.46 (m, 2H), 7.40–7.41 (m, 1H), 7.29–7.30 (m, 2H), 5.78 (s,
1H), 4.83–4.80 (m, 2H), 4.54–4.51 (m, 2H), 3.89–3.83 (m, 2H),
3.10–3.05 (m, 2H), 1.8–1.79 (m, 2H). LC–MS (M+H)+: 455.2.
4.3.4. 10-(3-Chlorobenzyl)-2-(2-chlorophenyl)-2,3,8,9,10,11-
hexahydro-1H-pyrazolo[40,30:3,4]pyrido[1,2-a][1,4]diazepine-
1,5(7H)-dione (7f)
Isolated as a brown solid (98% HPLC purity). 1H NMR (DMSO-
d6, 400 MHz) d: 11.01 (bs, 1H), 7.67–7.60 (m, 2H), 7.52 (t,
J = 4.8 Hz, 2H), 7.46 (t, J = 4.8 Hz, 2H), 7.45–7.40 (m, 2H), 5.84 (s,
1H), 5.10–4.90 (m, 2H), 4.52–4.48 (m, 2H), 4.23–4.20 (m, 2H),
3.18–3.13 (m, 2H), 2.00–1.96 (m, 2H). LC–MS (M+H)+: 455.0.
4.3. General procedure for the synthesis of hexahydro-1,4-
diazepine derivatives (n = 2)
4.3.1. Synthesis of 10-benzyl-2-(2-chlorophenyl)-2,3,8,9,10,11-
hexahydro-1H-pyrazolo[40,30:3,4]pyrido[1,2-a][1,4]diazepine-
1,5(7H)-dione (7c)
Pyrazolone 1b (643 mg, 1.88 mmol) was dissolved in CH3CN
(10 mL) and amine 2 N-benzylpropane-1,3-diamine was added
at 0 °C under N2. The reaction mixture was stirred at 0 °C for
3 h and monitored by LC–MS showing 40% of starting material
left. Then DIPEA (two drops) was added and the mixture was
stirred at RT for 1 h: LC–MS showed completely conversion of
starting material into desired product. The solvent was concen-
4.3.5. 10-(4-Chlorobenzyl)-2-(2-chlorophenyl)-2,3,8,9,10,11-
hexahydro-1H-pyrazolo[40,30:3,4]pyrido[1,2-a][1,4]diazepine-
1,5(7H)-dione (7g)
Isolated as a brown solid (98% HPLC purity). 1H NMR (DMSO-d6,
400 MHz) d: 7.66–7.62 (m, 3H), 7.55–7.53 (m, 1H), 7.49–7.46 (m,
4H), 5.90 (s, 1H), 5.15 (s, 2H), 5.17–5.13 (m, 2H), 4.56–4.53 (m,
2H), 4.33–4.31 (m, 2H), 4.32 (s, 2H), 3.34–3.31 (m, 2H), 2.10–2.05
(m, 2H). LC–MS (M+H)+: 455.3.

F. Gaggini et al. / Bioorg. Med. Chem. 19 (2011) 6989–6999
6997
4.3.6. 2-(2-Chlorophenyl)-10-(2-methoxybenzyl)-2,3,8,9,10,11-
hexahydro-1H-pyrazolo[40,30:3,4]pyrido[1,2-a][1,4]diazepine-
1,5(7H)-dione (7h)
4.98 (bs, 2H), 4.60–4.50 (m, 2H), 4.36 (s, 2H), 3.679 (s, 3H), 3.38
(t, J = 5.2, 2H), 2.00–1.90 (m, 2H). LC–MS (M+H)+: 418.6.
Isolated as a brown solid (98% HPLC purity). 1H NMR (DMSO-d6,
400 MHz) d: 11.17 (bs, 1H), 7.65 (d, J = 9.2 Hz, 1H), 7.57–7.56 (m,
1H), 7.51–7.49 (m, 3H), 7.23 (t, J = 7.2 Hz, 1H), 7.09 (d, J = 8.4 Hz,
1H), 6.98 (t, J = 7.4 Hz, 1H), 5.90 (s, 1H), 5.42 (bs, 1H), 5.04 (bs,
1H), 4.66 (bs, 2H), 4.29 (bs, 2H), 3.81 (s, 3H), 2.22–2.14 (m, 1H),
1.98–1.96 (m, 1H). LC–MS (M+H)+: 451.26.
4.3.13. 10-(3-Methoxybenzyl)-2-(2-methoxyphenyl)-2,3,8,9,10,
11-hexahydro-1H-pyrazolo[40,30:3,4]pyrido[1,2-a][1,4]
diazepine-1,5(7H)-dione (7o)
Isolated as a yellow solid (97% HPLC purity). 1H NMR (DMSO-d6,
400 MHz) d: 7.46–7.38 (m, 2H), 7.35–7.33 (m, 1H), 7.307–7.282 (m,
1H), 7.18 (dd, J = 1.2, 8.4 Hz, m, 2H), 7.04 (t, J = 7.6 Hz, 1H), 6.97
(dd, J = 2, 7.6 Hz, 1H), 5.85 (s, 1H), 5.19 (s, 2H), 4.55–4.54 (m,
2H), 4.275 (s, 2H), 3.77 (s, 3H), 3.75 (s, 3H), 3.181–3.163 (m, 2H),
2.15–1.98 (m, 2H). LC–MS (M+H)+: 447.1.
4.3.7. 2-(2-Chlorophenyl)-10-(furan-3-ylmethyl)-2,3,8,9,10,11-
hexahydro-1H-pyrazolo[40,30:3,4]pyrido[1,2-a][1,4]diazepine-
1,5(7H)-dione (7i)
Isolated as a brown solid (95% HPLC purity). Mp 100–102 °C. 1H
NMR (DMSO-d6, 400 MHz) d: 10.74 (s, 1H), 7.63 (t, J = 4.54 Hz, 1H),
7.55 (t, J = 4.8 Hz, 1H), 7.48–7.46 (m, 2H), 7.21 (d, J = 3.2 Hz,1H),
6.85–6.81 (m, 3H), 5.73 (s, 1H), 4.65–4.45 (m, 4H), 3.70 (s, 3H),
3.61 (s, 2H), 2.86–2.84 (m, 2H), 1.65 (s, 2H). LC–MS (M+H)+: 450.9.
4.4. General procedure for the synthesis of tricyclic compounds
with piperazine (n = 1) or hexahydro-1,4-diazepine (n = 2)
derivatives
4.4.1. Synthesis of 9-benzyl-2-(2-chlorophenyl)-2,3,7,8,9,10-
hexahydropyrazolo[40,30:3,4]pyrido[1,2-a]pyrazine-1,5-dione
(7b)
4.3.8. 2-(2-Chlorophenyl)-10-(4-methoxybenzyl)-2,3,8,9,10,11-
hexahydro-1H-pyrazolo[40,30:3,4]pyrido[1,2-a][1,4]diazepine-
1,5(7H)-dione (7j)
Pyrazolone 1b (1.88 mmol, 640 mg,) was dissolved in acetoni-
trile (10 mL) and cooled down to 0 °C. N-benzyl ethane 1,2-dia-
mine (1.69 mmol, 254 mg) was slowly added and the reaction
was stirred at 0 °C for 1 h. The solvent was concentrated in vacuo
and the crude was filtered through silica plug (ethyl acetate/0.1%
triethylamine) yielding a dark red oil (618 mg, 75% yield, 80% HPLC
purity) which was used without further manipulation and charac-
terization. LC–MS (M+H)+: 439.1.
Isolated as a brown solid (96% HPLC purity). 1H NMR (DMSO-d6,
400 MHz) d: 11.17 (bs, 1H), 7.75–7.63 (m, 1H), 7.59–7.59 (m, 1H),
7.60–7.58 (m, 4H), 7.06 (d, J = 4.4 Hz, 2H), 5.98 (s, 1H), 5.34–5.25
(m, 2H), 5.12–5.10 (d, 2H), 4.70–4.62 (m, 2H), 4.32 (s, 2H), 3.82
(s, 3H), 3.25–3.20 (m, 2H), 2.25–2.20 (m, 1H), 2.05–2.02 (m, 1H).
LC–MS (M+H)+: 451.2.
The enamine deriving from the previous step (618 mg,
1.42 mmol) was dissolved in MeOH (10 mL) and treated with a
freshly prepared solution of MeOH/MeONa obtained from the disso-
lution of Na (65 mg, 2.84 mmol, 2 equiv). The resulting solution was
stirred at room temperature for 30 min: UPLC analysis showed the
cyclization went to completion. It was diluted with H2O, neutralized
(pH 7) with 1 M HCl. Aqueous solution was loaded onto reverse
phase eluted with 100% H2O in gradient with CH3CN. The product
came off with 50% of CH3CN: lyophilization of aqueous solution gave
a beige solid (17 mg, 11% yield, 90% HPLC purity). 1H NMR (DMSO-
d6, 500 MHz) d: 10.71 (s, 1H), 7.63–7.61 (m, 1H), 7.53–7.50 (m, 1H),
7.47–7.44 (m, 2H), 7.37–7.34 (m, 4H), 7.30–7.26 (m, 1H), 5.62 (s,
1H), 4.10 (s, 2H), 3.83 (t, 2H, J = 5.6 Hz), 3.71 (s, 2H), 2.91 (t, 2H,
J = 5.5 Hz). LC–MS (M+H)+: 407.1
4.3.9. 2-(2-Chlorophenyl)-10-(furan-3-ylmethyl)-2,3,8,9,10,11-
hexahydro-1H-pyrazolo[40,30:3,4]pyrido[1,2-a][1,4]diazepine-
1,5(7H)-dione (7k)
Isolated as a brown solid (95% HPLC purity). 1H NMR (DMSO-
d6, 400 MHz) d: 11.11 (bs, 1H), 7.76 (s, 1H), 7.64–7.62 (m, 1H),
7.54–7.52 (m, 1H), 7.48–7.46 (m, 2H), 6.72 (d, J = 2.8 Hz,1H),
6.50 (d, J = 2.8 Hz, 1H), 5.85 (s, 1H), 4.99–4.96 (m, 2H), 4.35 (s,
2H), 3.34–3.33 (m, 4H), 2.10–1.90 (m, 2H). LC–MS (M+H)+:
411.2.
4.3.10. 2-(2-Chlorophenyl)-10-(pyridin-3-ylmethyl)-2,3,8,9,10,
11-hexahydro-1H-pyrazolo[40,30:3,4]pyrido[1,2-a][1,4]
diazepine-1,5(7H)-dione (7l)
Isolated as a yellow solid (98% HPLC purity). Mp 218–220 °C.
1H NMR (DMSO-d6, 400 MHz) d: 10.96 (s,1H), 8.98 (s, 1H), 8.78
(d, J = 4.8 Hz, 1H), 8.60 (d, J = 8 Hz, 1H), 7.92 (t, J = 2 Hz, 4H),
7.57–7.50 (m, 1H), 7.49–7.44 (m, 3H), 5.79 (s, 1H), 4.84–4.82
(m, 2H), 4.53–4.51 (m, 2H), 4.29–4.28 (m, 2H), 3.32–3.31 (m,
2H), 1.98–1.97 (m, 2H). LC–MS (M+H)+: 422.9.
4.4.2. Synthesis of 2-(2-chlorophenyl)-2,3,7,8-tetrahydro-1H-
pyrazolo[40,30:3,4]pyrido[2,1-c][1,4]oxazine-1,5(10H)-dione (11)
Pyrazolone 1b (1.88 mmol, 640 mg,) was dissolved in acetoni-
trile (10 mL) and cooled to 0 °C. Propanolamine (1.69 mmol,
103 mg) was slowly added and the reaction was stirred at 0 °C for
1 h. The solvent was concentrated in vacuo and the crude was fil-
tered through a silica plug (ethyl acetate/0.1% triethylamine) yield-
ing a yellow oil (521 mg, 80% yield, 86% HPLC purity) which was
used without further manipulation and characterization. LC–MS
(M+H)+: 386.1, 388.1.
After having washed NaH (60% mineral oil, 1.5 equiv, 53 mg)
with small volumes of pentane, it was suspended in THF (1 mL) un-
der N2 and cooled down to 0 °C. Enamine 9 from the previous step
(421 mg, 1.31 mmol, 1 equiv) was dissolved in THF (4 mL) and
dropwise added to the mixture of NaH in THF at 0 °C under N2.
The mixture was stirred for 10 min and then at RT. When the start-
ing material had completely reacted, the reaction was diluted with
water (5 mL) and neutralized with 1 M HCl. The aqueous solution
was extracted several times with ethyl acetate (50 mL ꢄ 4). The
combined organic phases were dried over Na2SO4 and after evapo-
ration of the solvent, the crude was dissolved in the minimum
amount of ethylacetate. The precipitate was filtered off, washed
with small portions of cold ethyl acetate and dried under vacuo
4.3.11. 2-(2-Chlorophenyl)-10-(pyridin-2-ylmethyl)-2,3,8,9,10,
11-hexahydro-1H-pyrazolo[40,30:3,4]pyrido[1,2-a][1,4]
diazepine-1,5(7H)-dione (7m)
Isolated as a brown solid (98% HPLC purity). Mp 93–95 °C. 1H
NMR (DMSO-d6, 400 MHz) d: 10.78 (s,1H), 8.52–8.51 (m, 1H),
7.85–7.84 (m, 1H), 7.63–7.60 (m, 1H), 7.52–7.44 (m, 4H),
7.35–7.34 (m, 1H), 5.76 (s, 1H), 4.72–4.72 (m, 2H), 4.51–4.49 (m,
2H), 3.95 (s, 2H), 3.08 (s, 2H), 1.76 (s, 2H). LC–MS (M+H)+: 422.9.
4.3.12. 2-(2-Methoxyphenyl)-10-(pyridin-2-ylmethyl)-2,3,8,9,
10,11-hexahydro-1H-pyrazolo[40,30:3,4]pyrido[1,2-a][1,4]
diazepine-1,5(7H)-dione (7n)
Isolated as a brown solid (94% HPLC purity). Mp 178–180 °C. 1H
NMR (DMSO-d6, 400 MHz) d: 8.66 (dd, J = 0.8, 1.6 Hz, 1H), 8.10 (td,
J = 1.6, 8.6 Hz, 1H), 7.74 (d, J = 7.6 Hz, 1H), 7.60 (dd, J= 0.8, 5.4 Hz,
1H), 7.43 (t, J = 8.8, 1.2, 1H), 7.31 (dd, J = 1.2, 7.6 Hz, 1H), 7.17 (dd,
J = 1.2 Hz, 8.4 Hz, 1H), 7.03 (td, J = 1.2, 7.6 Hz, 1H), 5.08 (s, 1H),

6998
F. Gaggini et al. / Bioorg. Med. Chem. 19 (2011) 6989–6999
yielding 21 mg of a yellowish solid (91% HPLC purity, 50% yield). 1H
NMR (DMSO-d6, 500 MHz) d: 10.81 (s, 1H), 7.66–7.63 (m, 1H),
7.55–7.52 (m, 1H), 7.50–7.46 (m, 2H), 5.67 (s, 1H), 5.13 (s, 2H),
4.08 (t, 2H, J = 5.3 Hz), 3.81 (t, 2H, J = 5.3 Hz). LC–MS (M+H)+: 318.1.
tion and represent the average of at least three individual experi-
ments performed in triplicate.
4.5.6. Oxidative metabolism
Pooled human liver microsomes (pooled male and female),
pooled rat liver microsomes (male Sprague Dawley rats) were used
to screen the metabolic instability resulting from phase I oxidation.
Microsomes (final concentration 0.5 mg/mL), 0.1 M phosphate buf-
fer pH 7.4, and test compound (final substrate concentra-
tion = 0.5 lM; final DMSO concentration = 0.25%) were added to
the assay plate and preincubated at 37 °C.
NADPH solution (final incubation concentration = 1 mM) was
added to initiate the reaction. Each compound is incubated for 0,
5, 15, 30 and 45 min. The control (minus NADPH) was incubated
for 45 min only. The reactions were stopped by the addition of
4.5. Biological assays
4.5.1. Plasmid construction
Human Nox1 (NM_007052.4) cDNA was cloned into the pcDNA5/
TO (Invitrogen), human p22 (NM_000101.2) was cloned into
pcDNA3.1/Zeo (+) (Invitrogen) and human NOXA1 (AY255769.1)
and human NOXO1 (AB097667) were cloned into the bi-cistronic
pVITRO1-neo-mcs plasmid (Invivogen).
4.5.2. Cell culture and stable transfected cell line generation
T-REx™-CHO cells (Invitrogen) were cultured in Ham’s F12
containing 4.5 g/l glucose supplemented with 10% fetal calf serum,
50 lL methanol containing internal standard at the appropriate
time points. The samples were centrifuged at 2500 rpm for
20 min at 4 °C to precipitate the protein. Samples were analyzed
by LC/MS–MS. The percentage of compound disappearance and
the in vitro intrinsic clearance was calculated according to the
literature.11
2 mM glutamine, penicillin (100 U/ml), streptomycin (100
lg/ml)
and blasticidin (5 g/ml). Cells stably expressing functional human
l
Nox1 (hNOX1 T-REx™-CHO) were obtained by co-transfecting
T-REx™-CHO with human NOX1, human p22, human NOXA1 and
human NOXO1 using the FuGENE 6 method (Roche, 11988387).
T-REx™-293 cell line expressing hNox4 (hNoxOX4 T-REx™-293)
was a gift from the Department of Immunology and Pathology,
University of Geneva.9 hNOX4 T-REx™-293 cells were cultured in
DMEM containing 4.5 g/l glucose supplemented with 10% Fetal calf
4.5.7. Cytochrome P450 inhibition
Six human recombinant cytochrome P450 isoenzymes
(BD Gentest, Woburn, MA) were tested (CYP1A2, CYP2C8, CYP2C9,
CYP2C19, CYP2D6 and CYP3A4). Test compound (0.1–25 lM) was
serum and penicillin (100 U/ml), streptomycin (100
in air with 5% CO2. Human Nox1 and human Nox4 expression were
l
g/ml) at 37 °C
incubated (final DMSO concentration = 0.3%) with specific species
liver microsomes and NADPH in the presence of a cytochrome
P450 isoform-specific probe substrate. For the CYP2C8, CYP2C9,
CYP2C19, CYP2D6 and CYP3A4 specific reactions, the metabolites
were monitored by mass spectrometry. CYP1A2 activity was
monitored by measuring the formation of a fluorescent metabo-
lite. A decrease in the formation of the metabolite compared to
the vehicle control was used to calculate an IC50 value.
induced with Tetracycline (1
brane preparation.
lg/ml) during 24 h prior to the mem-
4.5.3. Membrane preparation
Membranes from human polymorphonuclear (PMN) cells
(expressing high levels of Nox2) or from cells overexpressing
Nox1 or Nox4 or Nox5 were prepared as previously described.10
After resuspension in sonication buffer (11% sucrose, 120 mM NaCl,
1 mM EGTA in PBS, pH 7.4 for Nox4-expressing cells) or in relax
buffer (10 mM Pipes, 3 mM NaCl, 3.5 mM MgCl2, 0.1 M KCl, pH
7.4), cell were broken by sonication and centrifuged (200 g,
10 min). The supernatant was layered onto a 17/40% (w/v) discon-
tinuous sucrose gradient and centrifuged (150,000g for 30 min).
Membrane fractions were collected from the 17/40% interface
and were stored at ꢁ80 °C. Protein concentration was determined
with Bradford reagent.
4.5.8. Caco-2 permeability
Caco-2 cells were obtained from American type culture collec-
tion (Rockville, MD). The cells were seeded onto a Millipore Multi-
screen Caco-2 plates at 1 ꢂ 105 cell/cm2 and grown in Dulbecco’s
modified Eagle’s medium and media was changed every two or
three days. Permeability studies were performed with the mono-
layers cultured for 20 days. On day 20, the monolayers were pre-
pared by rinsing both basolateral and apical surfaces twice with
HBSS at 37 °C. Prior to all experiments, the cell monolayer integrity
was evaluated by trans epithelial electrical resistance; values
greater than 800 ohm ꢂ cm2 were used. Hank’s balanced salt solu-
tion buffer was then removed from either the apical compartment
(for apical to basolateral transport; A to B) or basolateral compart-
ment (for basolateral to apical transport; B to A) side of the mono-
layer and replaced with test compound dosing solutions. The
permeability studies were initiated by adding an appropriate vol-
ume of solutions made by diluting 10 mM test compound in DMSO
with HBSS pH 6.5 buffer to give a final test compound concentra-
4.5.4. ROS production measurement
Reactive oxygen species (ROS) production by membranes
expressing human Nox1, human Nox2 or human Nox4 or by
Xanthine oxidase was measured using the Amplex Red (AR) meth-
od following a slightly modified version of the manufacturer’s
instruction manual (Invitrogen). Briefly, membranes expressing
different Nox subunits or Xanthine oxidase were incubated in
PBS with Amplex Red, Horse Radish Peroxidase (HRP) and appro-
priate co-factors. ROS production was induced by addition of
NADPH to Nox expressing membranes or by addition of Xanthine
to Xanthine oxidase. Nonspecific signal was measured in the ab-
sence of membranes or in the absence of agonist. Antagonist activ-
ity of compounds was measured in the presence of increasing
tion of 10 lM (final DMSO concentration 1%). The monolayers were
then placed in an incubator at 37 °C. At the end of the incubation
time (2 h), samples were taken from both the apical and basolat-
eral compartments. Test and control compounds were quantified
by LC–MS/MS cassette analysis using a 5-point calibration with
appropriate dilution of the samples. The permeability coefficient
(Papp) was calculated according to the following equation:
Papp = (dQ/dt)/(Co x A), where dQ/dt is the rate of permeation of
the drug across the cell, Co is the donor compartment concentra-
tion at time zero and A is the area of the cell monolayer. Co is ob-
tained from analysis of donor and receiver compartments at the
end of the incubation period.
concentrations ranging from 1 nM to 100 lM. After 20 min. incu-
bation at 37 °C, ROS levels were measured using a BMG Labtech
microplate reader.
4.5.5. Data analysis
Data were analyzed using Prism (GraphPad Software Inc., San
Diego, CA). Ki values were calculated using the Cheng-Prusoff equa-

F. Gaggini et al. / Bioorg. Med. Chem. 19 (2011) 6989–6999
6999
4.5.9. In vivo pharmacokinetic studies in rat
4.6.4. Sircol assay for collagen quantification
These studies were conducted to estimate the plasma pharma-
cokinetic parameters and the oral bioavailability of compounds in
the male Sprague-Dawley rat. These studies were carried out using
a parallel design using six male rats. Three animals in group 1 were
given 5 mg/kg iv dose; three animals in group 2 were given 10 mg/
kg oral dose. Blood samples were collected from the carotid arterial
cannula at various times up to 24 h following administration of
compound. Plasma concentrations of compound were quantified
by LC/MS/MS (LLOQ = 1.00 ng/mL). Noncompartmental methods
were used for pharmacokinetic data analysis.
Lung collagen deposition was estimated by using a dye-binding
method (SIRCOL Assay, Biocolor Ltd.). Whole lung were carefully
removed and homogenized in 0.5 M acid acetic. Collagen was ex-
tracted overnight at 4 °C in a mechanical shaker. After centrifuga-
tion, 100 lL of diluted supernatant was mixed with 200 lL of
collagen concentrator and 1 mL of Sircol dye reagent for 30 min
at room temperature in a mechanical shaker. After centrifugation,
the pellet was suspended in 1 mL cuvette, with absorbance mea-
sured at 540 nm. Collagen concentrations were calculated based
on the amount of Alkali reagent and vortexed to release the dye
and 1 mL was transferred into a standard curve of known concen-
The vehicle used for per os administration was an aqueous solu-
tion of 0.5% CMC (carboxymethylcellulose) and 0.25% Tween 20.
The vehicle used for iv injection was an aqueous solution of EDPW
(10% EtOH, 10% DMA, 30% PG and 50% water).
tration of rat tail collagen. Data were expressed as
per lung.
lg of collagen
4.6.5. Histology
4.6. Murine curative model of bleomycine-induced lung fibrosis
Organs were dehydrated in successive baths containing alcohol,
xylol and paraffin at 70 °C for 18 h cycle in Histokinette. They were
then embedded in paraffin and cut using microtome (4 lm), then
4.6.1. Animals
The Veterinary Office of Geneva on the Use and Care of Animals
approved the animals protocols used in this study. Nine to 10-week
old males C57BL/6 mice weighing 25 g were obtained from Janvier
(Le Genest Saint Isle, French). Mice were housed under specific
pathogen-free conditions in plastic enclosed filter-top cages
(Techniplast system), five animals per cage. Animals were main-
tained in a 12 h light/dark cycle with ad libitum access to water
and rodent laboratory chow until euthanasia.
mounted on Superfrost slide. Sections were dried in a stove at
60 °C overnight and stained by hematoxylin and eosin or Masson’s
Trichrome for collagen staining.
Acknowledgments
We thank He Haiying, Cao Yafeng, Lei Jianguang (WuXi AppTec,
Shanghaï) and the Wuxi AppTec chemistry team for their participa-
tion to this work.
4.6.2. Study design and protocols. Instillation of BleomycinÒ
For induction of Bleomycin-induced pulmonary fibrosis, Bleo-
mycin Baxter (Pharmacie du Rondeau, Carouge, CH) was dissolved
in sterile 0.9% saline on the day of IT instillation. Prior to study start
animals were weighed and distributed randomly into cages. Ani-
mals were anesthetized using a solution of Ketamine/Xylazine
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2011.10.016.
(75 mg/kg and 8 mg/kg).
A single 40 lL aliquot containing
0.0125 U of BleomycinÒ diluted in normal saline was injected
intratracheally using a Tridak-stepper (Millian) and a 30-gauge
needle. Control mice received saline vehicle only.
Reference and notes
1. Schwartz, D. A.; Helmers, R. A.; Galvin, J. R.; Van Fossen, D. S.; Frees, K. L.;
Dayton, C. S.; Burmeister, L. F.; Hunninghake, G. W. Am. J. Respir. Crit. Care Med.
1994, 149, 450.
2. King, T. E., Jr.; Tooze, J. A.; Schwarz, M. I.; Brown, K. R.; Cherniack, R. M. Am. J.
Respir. Crit. Care Med. 2001, 164, 1171.
3. Misra, H. P.; Rabideau, C. Mol. Cell Biochem. 2000, 204, 119.
4. Kim, J. A.; Neupane, J. P.; Lee, E. S.; Jeong, B. S.; Park, B. C.; Thapa, P. Expert. Opin.
Ther. Pat. 2011 (Epub ahead of print).
5. Hecker, L.; Vittal, R.; Jones, T.; Jagirdar, R.; Luckhardt, T. R.; Horowitz, J. C.;
Pennathur, S.; Martinez, F. J.; Thannickal, V. J. Nat. Med. 2009, 15, 1077.
6. Laleu, B.; Gaggini, F.; Orchard, M.; Fioraso-Cartier, L.; Cagnon, L.; Houngninou-
Molango, S.; Gradia, A.; Duboux, G.; Merlot, C.; Heitz, F.; Szyndralewiez, C.;
Page, P. J. Med. Chem. 2010, 53, 7715.
4.6.3. Curative treatment
Animals were randomly assigned into groups of 8 to 13 animals
per group 10 days after the BleomycinÒ instillation. GKT137831
and Pirfenidone were administrated at the same dose from D10
to D35 by a daily oral administration of 10 mL/kg. Compound 7c
and Pirfenidone were dissolved into a volume of 3% in DMSO fol-
lowed by an aqueous solution of 0.5% CMC (carboxymethylcellu-
ose) and 0.25% Tween 20. Pirfenidone was uses as a benchmark
7. Page, P.; Gaggini, F.; Laleu, B. Pyrazoline dione derivatives as NADPH oxidase
inhibitors WO2011/036651 A1, 2011, Genkyotex.
8. Abdel-Magid, A. F.; Carson, K. G.; Harris, B. D.; Maryanoff, C. A.; Shah, R. D. J.
Org. Chem. 1996, 61, 3849.
9. Serrander, L.; Cartier, L.; Bedard, K.; Banfi, B.; Lardy, B.; Plastre, O.; Sienkiewicz,
A.; Fórró, L.; Sclegel, W.; Krause, K. H. Biochem. J. 2007, 406, 105.
10. Palicz, A.; Foubert, T. R.; Jesaitis, A. J.; Marodi, L.; McPhail, L. C. J. Biol. Chem.
2001, 76, 3090.
11. Lave, T. H.; Dupin, S.; Schmitt, C.; Chou, R. C.; Jack, D.; Coassolo, P. H. J. Pharm.
Sci. 1997, 86, 584.
molecule. Control groups received
10 mL/kg vehicle.
a daily administration of
At the end of the experiment on D35, mice were euthanized and
perfused via the right ventricle with 5 mL of saline. Lungs were
carefully removed and rinsed in PBS, dried and weighed. Five lungs
per group were placed in paraformaldehyde solution (PFA 4%) for
histology for 24 h. Lungs were rinsed twice in PBS and processed
for histology.
The remaining lungs were placed in a 1.5 mL eppendorf tubes,
frozen in liquid nitrogen and stored at ꢁ80 °C until the collagen
content was quantified by SIRCOL assay.