5302
B. Perry et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5299–5302
a 100
4.0
3.5
3.0
2.5
2.0
b
75
50
25
0
1
1
8g
10d
8g
10d
2.0 2.5 3.0 3.5 4.0
Predicted EC50 free
0.0 0.5 1.0 1.5 2.0
log [cpd] (mg/kg)
Figure 2. Efficacy of PI3K inhibitors in in vivo models. (a) Dose–response curves for
compounds 1, 8g and 10d in CD3-induced IL2 release in male Lewis rats. (b)
Correlation between in vitro and in vivo efficacy, expressed as EC50 (plasma free-
fraction for in vivo). The regression line and 95% confidence intervals are
represented as solid and dotted lines, respectively.8
sented using free compound exposure in plasma (EC50 free) for this
series, a good correlation is seen to the in vitro potency (Fig. 2b).8
This demonstrates the clear overall improvement in compound
properties of 8g and 10d, compared to 1.
In conclusion, we have demonstrated optimization of a series of
multi-isoform PI3K inhibitors. General cellular and enzyme activity
has been improved, and compounds with good solubility and ki-
nase selectivity have been identified, whilst retaining or improving
on the good in vivo pharmacological activity demonstrated by the
parent leads.
Figure 1. Relationship between (a) average fold drop-off in cellular activity (fMLP
IC50) versus PI3Kd enzyme IC50 (blue) and PI3Kc enzyme IC50 (yellow) relative logD
for compounds within the whole UCB series; (b) measured logD at pH 7.4 and
plasma protein binding (rat) for all compounds in Table 2; (c) in vitro hepatocytic
clearance and in vivo exposure (log scale) for compounds in Table 2.
Acknowledgments
Table 2
The authors acknowledge the significant contributions of the
DMPK, molecular biology and CADD groups at UCB Cambridge
and the medicinal chemistry group at UCB Slough in helping to car-
ry out this research.
In vivo PK analysis of key compounds orally dosed in Han-Wistar ratsa
Compound
Cmax (ng/mL)
AUC (ng h/mL)
% PPBb
1
2
1216
185
6162
2471
96.2
90.2
8c
8f
195
139
159
103
187
105
14
1005
591
495
455
807
345
78
89.4
98.7
86.6
98.0
86.6
88.7
90.4
References and notes
8g
8j
8q
10d
11e
1. Perry, B.; Alexander, R.; Bennett, G.; Buckley, G.; Ceska, T.; Crabbe, T.; Dale, V.;
Gowers, L.; Horsley, H.; James, L.; Jenkins, K.; Kitahara, K.; Kulisa, C.; Lightfoot,
H.; Lock, C.; Mack, S.; Morgan, T.; Nicolas, A.; Pitt, W.; Sabin, V.; Wright, S. Bioorg.
Med. Chem. Lett. 2008, 18, 4700.
2. (a) Alexander, R.; Batchelor, M; Balasundaram, A.; Brookings, D.; Kitahara,
K.; Kulisa, C; Turner, J.; Hutchinson, G.; Harris, S.; Wright, S. Bioorg. Med.
Chem. Lett. 2008, 18, 4316.; (b) Alexander, R.; Aujla, P.; Batchelor, M.;
Brookings, D.; Buckley, G.; Crepy, K.; Kulisa, C.; Turner, J. WO Patent
2006114606, 2006.
a
Dosed at 3 mg/kg po.
% Plasma protein bound in blood (male Han-Wistar rat).
b
3. (a) Cantley, L. C. Science 2002, 296, 1655; (b) Vanhaesebroeck, B.; Leevers, S. J.;
Panayotou, G.; Waterfield, M. D. Trends Biochem. Sci. 1997; (c) Crabbe, T.
Biochem. Soc. Trans. 2007, 35, 253; (d) Rommel, C.; Camps, M.; Ji, H. Nat. Rev.
Immunol. 2007, 7, 191; (e) Vanhaesebroeck, B.; Ali, K.; Bilancio, A.; Geering, B.;
Foukas, L. C. Trends Biochem. Sci. 2005, 30, 194; (f) Clayton, E.; Bardi, G.; Bell, S.
E.; Chantry, D.; Downes, C. P.; Gray, A.; Humphries, L. A.; Rawlings, D.; Reynolds,
H.; Vigorito, E.; Turner, M. J. Exp. Med. 2002, 196, 753; (g) Hirsch, E.; Katanaev, V.
L.; Garlanda, C.; Azzolino, O.; Pirola, L.; Silengo, L.; Sozzani, S.; Mantovani, A.;
Altruda, F.; Wymann, M. Science 2000, 287, 1049.
slightly lower Cmax and AUC than the pyrazole series, and direct-
linked amino-pyridazine 11e had very low exposure. Despite most
of these compounds having relatively similar profiles in the in vitro
rat and human microsomal clearance model, none demonstrated
levels of exposure in vivo comparable to parent compounds 1
and 2. Analysis of the link between in vitro hepatocytic clearance
and in vivo exposure suggests a logarithmic correlation between
these two factors, indicating that for this series of compounds
hepatocytic clearance may be key to achieving good in vivo expo-
sure (Fig. 1c).
Despite demonstrating significantly lower oral exposures rela-
tive to lead compounds 1 and 2, the increased free-fraction and im-
proved solubility and cellular activity of 8g and 10d led us to
investigate the activity of these compounds in vivo. Acute activa-
tion of rat T-cells by anti-CD3 antibody treatment causes release
of IL2, both in vitro and in vivo. As shown in Figure 2a, 8g and
10d inhibited CD3-induced IL2 release in male Lewis rats with
ED50s of 5 mg/kg and 20 mg/kg, respectively, compared with an
ED50 of 25 mg/kg for compound 1. When efficacy in vivo is repre-
4. Compound 1 (10 lM) demonstrated >50% inhibition against 2 kinases from a
panel of 50 (Pim-1 and SAPK2a).
5. Separation performed on chiralpack IA column using EtOH/heptane (60:40) as
eluant. Absolute configuration or 8p and 8q assigned through discreet synthesis
of 8q from reaction of 8a with (S)-4-methoxymethyl-1,2-dioxolan-2-one (NaOH,
DMF, 155 °C, 4 h, 49%, >98% ee).
6. Wolfe, John P.; Ahman, Jens; Sadighi, Joseph P.; Singer, Robert A.; Buchwald,
Stephen L. Tetrahedron Lett. 1997, 38, 6367.
7. Cellular activity determined through
a
PI3Kd/
c
driven assay monitoring
-primed
inhibition of superoxide production by fMLP-stimulation of TNF
a
human neutrophils, see Condliffe, A. M.; Davidson, K.; Anderson, K. E.; Ellson,
C. D.; Crabbe, T.; Okkenhaug, K.; Vanhaesebroeck, B.; Turner, M.; Webb, L.;
Wymann, M. P.; Hirsch, E.; Ruckle, T.; Campls, M.; Rommel, C.; Jackson, S. P.;
Chilvers, E. R.; Stephens, L. R.; Hawkins, P. T. Blood 2005, 106, 1432.
8. All data points in Figure 2b represent compounds from the morpholino- or
benzoxazino-dihydro-thiazolopyridinone series of PI3K inhibitors described in
Ref. 2b.
Perry, Benjamin
