P. Eastwood et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5270–5273
5273
vresulting in poor plasma levels. The difluoro compound 12 also
had a relatively high volume of distribution and this coupled with
a low plasma clearance led to a compound with an exceptionally
long half life—however due to the observed cytotoxicity (vide supra)
this compound was not subjected to further profiling. Encourag-
ingly, improved oral pharmacokinetic profiles over that seen with
1 were obtained for the tetrahydropyran analogue 6 and the aza
analogue 24. Both compounds had low clearance and low-moderate
volumes of distribution resulting in moderate residence times and
in addition both showed good oral bioavailability and plasma levels.
However, both compounds were not progressed any further in the
screening cascade for distinct reasons; for compound 6 there were
particular concerns over the possible clinical consequences of the
hydrolytic instability in human microsomes (see Table 1) and for
compound 24 the discrepancy between the enzymatic and whole
blood potencies was cause for concern.10 In addition, upon selectiv-
ity profiling both compounds were found to be reasonably potent
inhibitors of the kinase c-Raf (h) (6: IC50 = 120 nM, 24:
IC50 = 105 nM).
O2N
BnO2C
Br
H
N
O2N
Cl
Br
Br
(i)
(ii)
O
N
N
N
CO2Bn
30
29
31
Scheme 3. Reagents and conditions: (i) NaH, DME, CH(CO2Bn)2, rt, 80%; (ii) Fe, HCl/
EtOH, reflux, 60%.
H
N
H
N
H
N
D
Cl
(ii)
Cl (i)
D
Cl
D
O
O
A
A
A
Br
Br
32
33
34
A, D = CH or N
Scheme 4. Reagents and conditions: (i) PyHBr.Br2, n-BuOH, 30 °C, 90–95%; (ii) Zn,
AcOH, rt, 80–95%.
Preparation of aza-oxindole 31 (Scheme 3) was achieved via a
two step sequence: a SNAr reaction of chloropyridine 29 with the
anion of dibenzylmalonate11 provided 30 which, upon treatment
with Fe/HCl, gave 31 via a one-pot multi-step sequence involving
nitro group reduction, removal of the benzyl protecting groups,
cyclisation and decarboxylation. The remaining isomeric nuclei of
type 34 were accessed in high yield via a published sequence12
(Scheme 4) which involved reaction of aza-indoles 32 with pyrid-
inium tribromide to give dibromo intermediates of type 33 which
were then subjected to a Zn/acetic acid-mediated reduction.
It can be seen from Table 2 that the heterocyclic analogues, in
general, maintain good potencies in the primary enzymatic assay.
However potencies in the whole blood assay were somewhat dis-
appointing with compounds typically showing differences of at
least an order of magnitude compared to the enzymatic assay de-
spite all compounds showing good permeability through a Caco-2
cell line (data not shown). Nevertheless, it is interesting to note
that the increased polarity of these compounds resulted in both a
decrease in the in-vitro cytotoxicity when compared to the de-
aza counterparts and a significant reduction in all forms of metab-
olism; of particular interest was the observed reduction of the
hydrolytic metabolism in human microsomes.
In summary, further optimisation of the indolin-2-one series of
p38a inhibitors has led to potent compounds with reduced cyto-
toxicity and improved stability towards metabolic oxidation lead-
ing to compounds with improved oral bioavailabilities in the rat
compared to the lead compound 1. It has also been demonstrated
that the species-specific amide hydrolysis upon incubation with
human microsomes can be diminished by incorporation of partic-
ular substituents and/or preparation of heterocyclic analogues. The
further advancement of this series to give metabolically stable
compounds with demonstrated in-vivo activity will be the subject
of a following publication.
References and notes
1. Kumar, S.; Boehm, J.; Lee, J. C. Nat. Rev. Drug Disc. 2003, 2, 717.
2. Eastwood, P.; González, J.; Gómez, E.; Vidal, B.; Caturla, F.; Roca, R.; Balagué, C.;
Orellana, A.; Domínguez, M. Bioorg. Med. Chem. Lett. 2011, 21, 4310.
3. Data from MS/MS studies was used in the structural identification of all
metabolites.
4. Full details for all biological assays can be found in Ref. 2 or references cited
therein.
5. For further discussion see: (a) Waring, M. J. Expert Opin. Drug Disc. 2010, 5, 235;
Young, R. J.; Green, D. V. S.; Luscombe, C. N.; Hill, A. P. Drug Discovery Today
6. Full synthetic details can be found in the following patent: Eastwood, P. R.;
Gonzalez, J.; Vidal, B.; Aguilar, N. PCT Int. Appl. WO2009124692, 2009.
7. For metallation and trapping of indolin-2-ones see: (a) Stevens, F. C.;
Bloomquist, W. E.; Borel, A. G.; Cohen, M. L.; Droste, C. A.; Heiman, M. L.;
Kriauciunas, A.; Sall, D. J.; Tinsley, F. C.; Jesudason, C. D. Bioorg. Med. Chem. Lett.
2007, 17, 6270; (b) Fensome, A.; Bender, R.; Cohen, J.; Collins, M. A.; Mackner,
V. A.; Miller, L. L.; Ullrich, J. W.; Winneker, R.; Wrobel, J.; Zhang, P.; Zhang, Z.;
Zhu, Y. Bioorg. Med. Chem. Lett. 2002, 12, 3487.
The pharmacokinetic profiles in rat were determined for several
analogues and the results are given in Tables 3 and 4.
The basic piperidine analogue 4 had both a very high volume of
distribution and a high clearance (>85% hepatic blood flow)
Table 3
Intravenous pharmacokinetic profiles of selected derivatives in rat (1 mg/kg)a
Compd Terminal Cmax
Cl
AUC0–1
Vss (L/kg) MRT (h)
t½ (h)
(ng/mL) (L/h/kg) (ng h/mL)
8. Gonczi, C.; Csikos, E.; Hermecz, I.; Heja, G.; Illar, A.; Nagy, L.; Santa Csutor, A.;
Simon, A.; Simon, K.; Smelko Esek, A.; Szomor, T.; Szvoboda, G., PCT Int. Appl.
WO2001005760, 2001.
9. For discussions on the stereoselective reduction of similar spirocyclic ketones
see: (a) Venkatesan, H.; Davis, M. C.; Altas, Y.; Snyder, J. P.; Liotta, D. C. J. Org.
Chem. 2001, 66, 3653; (b) Halász, J.; Podányi, B.; Sánta-Csutor, A.; Böcskei, Z.;
Simon, K.; Hanusz, M.; Hermecz, I. J. Mol. Struct. 2003, 654, 187.
1
4
6
5.3
5.1
2.7
14.4
3.9
2002
133
1532
436
0.4
4.4
0.4
0.2
0.6
2532
226
2772
5017
1848
0.8
19.1
0.8
4.2
1.2
2.1
4.3
2.2
21.0
2.2
12
24b
1245
a
Formulations: compds 1, 4, 12: 40% PEG; 6: 40% PEG + DMSO; 24: 20%
PEG + 0.4% HCl.
10. This whole blood assay was used in many p38
a programmes to compare in-
vitro potency against potency in a cellular system in a whole blood medium
which takes into account other factors such as cell permeability (not likely to
be a significant problem for the majority of compounds in this article as
assessed by Caco-2 permeability experiments) and off-target binding including
plasma protein binding. In regard to the latter factor, recent opinion would
suggest that extreme caution should be taken when prioritizing compounds
based on results from in-vitro assays which are adversely affected by the
inclusion of plasma proteins because unbound plasma concentrations in-vivo
are determined not by plasma protein binding but by intrinsic clearance after
oral dosing. For further discussion see: (a) Liu, X.; Chen, C.; Hop, C. E. C. A. Curr.
Top. Med. Chem. 2011, 11, 450; (b) Smith, D. A.; Di, L.; Kerns, E. H. Nat. Rev. Drug.
Disc. 2010, 9, 929.
b
Pharmacokinetic studies were performed with the HCl salt.
Table 4
Oral pharmacokinetic profiles of selected derivatives in rat (10 mg/kg)a
Compd
Cmax (ng/mL)
AUC0–1 (ng h/mL)
F (%)
1
4
367
134
1310
4625
2644
786
10367
12013
10
35
40
65
6
24b
11. Daisley, R. W.; Hanbali, J. R. Synth. Commun. 1981, 11, 743.
12. Marfat, A.; Carta, M. P. Tetrahedron Lett. 1987, 28, 4027.
a
Formulations: compds 1, 4, 24: 0.5% methylcellulose + 0.1% Tween 80; 6: 40%
PEG + DMSO.
b
Pharmacokinetic studies were performed with the HCl salt.