Development of isoform selective PI3-kinase inhibitors as pharmacological tools for elucidating the PI3K pathway

Article abstract of DOI:10.1016/j.bmcl.2012.07.042

Using a parallel synthesis approach to target a non-conserved region of the PI3K catalytic domain a pan-PI3K inhibitor 1 was elaborated to provide alpha, delta and gamma isoform selective Class I PI3K inhibitors 21, 24, 26 and 27. The compounds had good cellular activity and were selective against protein kinases and other members of the PI3K superfamily including mTOR and DNA-PK.

Full text of DOI:10.1016/j.bmcl.2012.07.042

Bioorganic & Medicinal Chemistry Letters 22 (2012) 5445–5450
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Development of isoform selective PI3-kinase inhibitors as pharmacological
tools for elucidating the PI3K pathway
⇑
Ian Bruce , Mohammed Akhlaq, Graham C. Bloomfield, Emma Budd, Brian Cox, Bernard Cuenoud,
Peter Finan, Peter Gedeck, Julia Hatto, Judy F. Hayler, Denise Head, Thomas Keller, Louise Kirman,
Catherine Leblanc, Darren Le Grand, Clive McCarthy, Desmond O’Connor, Charles Owen, Mrinalini S. Oza,
Gaynor Pilgrim, Nicola E. Press, Lilya Sviridenko, Lewis Whitehead
Novartis Institutes for Biomedical Research, Respiratory Disease Area, Wimblehurst Road, Horsham, West Sussex RH12 5AB, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Using a parallel synthesis approach to target a non-conserved region of the PI3K catalytic domain a pan-
PI3K inhibitor 1 was elaborated to provide alpha, delta and gamma isoform selective Class I PI3K inhib-
itors 21, 24, 26 and 27. The compounds had good cellular activity and were selective against protein
kinases and other members of the PI3K superfamily including mTOR and DNA-PK.
Ó 2012 Elsevier Ltd. All rights reserved.
Received 6 June 2012
Revised 8 July 2012
Accepted 9 July 2012
Available online 17 July 2012
Keywords:
PI3-Kinase inhibitors
Pharmacological tools
Aminothiazoles
Parallel synthesis
Isoform selectivity
Phosphatidylinositol 3-kinases (PI3K’s), a family of enzymes
which catalyse the phosphorylation of the 30–OH of the inositol
ring, play a central role in regulating a wide range of cellular
processes including metabolism, survival, motility and cell activa-
tion.1 These lipid kinases are divided into 3 major classes, I, II & III,
according to their structure and in vitro substrate specificity.2 The
most widely understood class I family is further subdivided into
subclasses IA and IB. Class IA PI3K’s consist of an 85 kDa regula-
tory/adapter protein and three 110 kDa catalytic subunits
4002, have proven to be invaluable tools with which to
investigate the functional relevance of PI3-kinase inhibition
although they inhibit multiple isoforms.
Progress in the field led to a second generation of structurally
diverse PI3-kinase inhibitors (Fig. 1) which have shown efficacious
effects in animal models.7 Whilst selectivity over protein kinases
appears to be readily achievable, many second generation com-
pounds were either pan-PI3K inhibitors (e.g., PIK-90)8 or were only
moderately selective, often displaying additional inhibition of the
PIKK’s such as mTOR and DNA-PK which seems to favour flat
hydrophobic scaffolds.9 A number of examples do show specificity
(p110
kinase system whilst class IB consists of a single p110
which is activated by G protein-coupled receptors. The three mem-
bers of class II PI3K (C2 , C2b and C2 ) and single member of class
a, p110b and p110d) which are activated in the tyrosine
c
isoform
for single isoforms, e.g. AS-605240 (p110c
),10 TGX-221 (p110b),11
a
c
and CAL101 (p110d),12 however, reports of selective inhibitors
such as these remain uncommon.
III PI3K (Vps34) are less well understood. In addition there are also
four PI4K’s and several PI3-kinase related protein kinases (termed
PIKK’s or class IV) including DNA-PK, mTOR, ATM and ATR, all of
which have a similar catalytic domain.3
There is currently an intense focus by pharmaceutical compa-
nies to develop PI3K inhibitors as therapy for cancer, cardiovascu-
lar, respiratory, autoimmune and inflammatory diseases.4–6 The
first generation of PI3-Kinase inhibitors, Wortmannin and LY29
The situation is often not clear as to which isoform selectivities
would provide the optimum balance between a desired therapeu-
tic effect and appropriate safety profile. Indeed, it has been postu-
lated that, for some targets, inhibition of more than one isoform
may be a requirement for efficacy.13 There is a continuing demand
by researchers for suitable isoform selective class I PI3K inhibitors
to use as pharmacological tools, especially PI3Kc inhibitors, to
further enhance the biological validation of these isoforms as drug
targets.
In this paper we describe progression from a pan-PI3K lead mol-
ecule to isoform selective inhibitors with good cellular activity. All
⇑
Corresponding author. Tel.: +44 (0) 1403 323080.
E-mail address: ian.bruce@novartis.com (I. Bruce).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.07.042

5446
I. Bruce et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5445–5450
Figure 1. Second generation PI3-Kinase inhibitors
compounds were screened for in vitro PI3-kinase activity against
p110 , p110b, p110
and p110d isoforms.14 For secondary cellular
primarily a PI3Kc
and PI3Kd dependent process15 whilst B-cell
a
c
proliferation16 is a PI3Kd dependent process; selected results are
shown in Tables 1 and 3.
assays the inhibitory effect of the compounds on respiratory burst
in human neutrophils and B cell proliferation was determined.
For the respiratory burst assay, the chemotactic peptide f-Met-
Leu-Phe (fMLP) is used to mimic the action of N-formyl peptides
released during bacterial infection which contribute to neutrophil
recruitment and release of highly reactive oxygen species. This is
Aminothiazole 1 (Fig. 2), identified from an HTS campaign using
a PI3Kc biochemical assay, was selected as a suitable lead for opti-
mization as it lacked activity against a panel of protein kinases.
Table 2
Isoform selective PI3K inhibitors
Table 1
Compd Thiazole
precursor
Amine
precursor
PI3-Kinase Inhibition (Ki,
l
M)a
p110d p110b
Inhibition of PI3K by 1, 7a–f, 11 and 15
P110
c
p110a
Compd
PI3-Kinase Inhibition (Ki,
l
M)a
fMLP
18
19
20
21
22b
23
24
25
26
27
6a
6d
6f
10
—
6d
10
6d
14
6d
A
A
A
A
A
B
B
C
D
E
5.81
3.06
1.14
0.53
1.19
0.12
1.02
0.02
0.187
0.050
0.020
0.005
0.04
0.11
0.15
2.13
1.89
3.62
2.90
1.67
0.22
0.54
0.003
0.004
0.38
0.74
0.47
>10
>10
>10
P110
c
p110
a
p110d
p110b
IC50 (lM)
1
0.11
0.23
0.11
0.20
0.21
0.19
1.43
1.36
1.30
0.55
1.41
0.76
0.87
0.41
1.94
2.6
2.0
7a
7b
7c
7d
7e
7f
1.93
1.93
0.34
0.22
0.30
0.29
0.32
1.76
2.94
0.86
0.34
0.09
0.25
0.32
0.057
0.032
0.183
0.062
0.061
0.110
0.260
0.035
0.011
0.122
0.045
0.028
0.030
0.190
0.079
0.061
0.247
0.079
0.074
0.019
0.260
0.088
0.036
1.40
11
15
a
Values are means of at least two experiments.
Values quoted for 22 are IC50’s measured in a Kinase–Glo luminescent assay.
b
a
Values are means of at least two experiments.

I. Bruce et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5445–5450
5447
Table 3
In vitro cellular and selectivity data for compounds 21, 24, 26 and 27
Compd
fMLP IC50
Ma
,
BCP IC50
l
,
PI3K related assays.
IC50
Ma
Vps34
PI4 Kb
l
Ma
, l
Human
Mouse
Human
mTOR
DNA-PK
21
24
26
27
0.93
0.72
0.22
0.27
0.37
0.014
0.76
0.59
0.32
0.005
1.16
1.61
>9.1
>9.1
>9.1
>9.1
>10
>10
>10
>10
>9.1
>9.1
>9.1
>9.1
0.92
0.04
0.24
1.33
a
Values are means of at least two experiments
The interior of the ATP binding site is highly conserved between
class I PI3K isoforms so the lack of selectivity with compound 1
was not unexpected (Table 1). However, residues lining the
binding pocket are less conserved. We hypothesized that isoform
selectivity might be achievable by extending or replacing the acyl
moiety with diverse substituents to exploit these differences.19 In
an attempt to improve cellular potency by reducing the Polar Sur-
face Area of 1 (PSA = 102 Å) the primary sulfonamide was replaced
with
methylsulfone
(7a,
PSA = 76.13 Å)
pyrazole
(11,
PSA = 62.28 Å) and methyl ketone (15, PSA = 59.06 Å).
Synthesis of p-methylsulfone analogues from p-fluorobenzalde-
hydes is shown in Scheme 1. The methyl sulfone moiety was intro-
duced by nucleophilic substitution using the sodium salt of
methane sulfinic acid to provide analogues 3.20 Reaction with
nitroethane gave nitrostyrenes which were reduced with iron in
acetic acid to give benzyl methyl ketones 4.21 Bromination was
effected by slow addition of bromine to a rapidly stirred solution
of 4 in dioxane to give bromo derivatives 5. Reaction with thiourea
in ethanol gave aminothiazoles 6 which were acetylated to give
products 7a–f.
Figure 2. HTS hit 1 from screening against a PI3K
Docking of 1 in the p110 crystal structure is shown.
c enzyme inhibition assay.
c
The pyrazole analogue 11 (Scheme 2) was prepared from 8 by a
similar route. Attempted bromination of ketone 9 in this case gave
a complex mixture but aminothiazole 10 could be accessed from 9
in a one-pot procedure using iodine and thiourea in pyridine.22
Acetylation of 10 as before gave 11.
Ketone analogue 15 (Scheme 3) was prepared by Suzuki cou-
pling of the commercial boronic acid with protected bromoamino-
thiazole 13 followed by deprotection and acetylation.
Replacing the primary sulfonamide of 1 by methylsulfone to
give 7a led to a reduction in binding activity and a minor improve-
ment in cellular potency in the fMLP stimulated neutrophil assay
(Table 1). Introducing electron withdrawing groups ortho to the
sulfone (compounds 7b–f) was found to restore potency in the
biochemical assay by up to 8 fold and further improve cellular
Docking studies of 1 and the kinase domain of p110c showed the
methyl group nicely occupies a small hydrophobic pocket in the
ATP binding site formed by Tyr-867, Val-882 and Ile-963 with
hydrophobic residues Ile-879 and Ile-576 located above and below
the phenyl ring.17 The thiazole nitrogen and amide hydrogen form
a bidentate donor–acceptor pair with the NH and carbonyl of
Val-882 whilst the sulfonamide oxygen appears to form a hydro-
gen bond to the side chain of Lys-883. The acyl substituent points
to a channel leading to solvent space. This binding mode is consis-
tent with the reported X-ray co-crystal structure of a related
aminothiazole PIK-93 bound to p110c 18
.
Scheme 1. Reagents and conditions: (a) MeSO2Na, DMSO, 70 °C; (b) EtNO2. NH4OAc, reflux; (c) Fe, AcOH, 100 °C; (d) Br2, dioxane, 5–10 °C; (e) Thiourea, EtOH, reflux; (f) Ac2O,
70 °C, 10 min.

5448
I. Bruce et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5445–5450
Scheme 2. Reagents and conditions: (a) EtNO2. NH4OAc, reflux, 18 h (95%); (b) Fe,
AcOH, 100 °C, 2 h (91%); (c) I2, pyridine, thiourea, 80 °C, 8 h (26%); (d) Ac2O, 70 °C,
10 min (83%).
Figure 3. Amines RR’NH, which when combined with selected aminothiazole
scaffolds to give ureas, provided Class 1 isoform-selective PI3-kinase inhibitors.
crude reaction mixture through a plug of polymer supported
isocyanate (Argonaut Technologies), to remove the excess amine,
or by preparative HPLC.
This method was particularly amenable to automation; using a
Bohdan Synthesizer enabled us to generate hundreds of final com-
pounds per week. Since it was not clear how to achieve selectivity
for any particular isoform the first iteration utilised a diverse set of
80 amines (MW <200), mostly from commercial sources, leading to
approximately 400 isolated examples. For compounds showing a
trend towards isoform selectivity the process was repeated using
the appropriate aminothiazoles and smaller more focused subsets
of amines derived from compound archives or synthesis. Results
for selected examples are shown in Table 2. Whilst the majority
of final compounds were inactive or weak pan-PI3K inhibitors,
three types of amines were identified which led to potent isoform
selective inhibitors (Fig. 3). In particular, (S)-pyrrolidine carboxam-
Scheme 3. Reagents and conditions: (a) Ph2C = NH, toluene, reflux, 68 h; (b) NBS,
AcOH, rt, 1 h (96%); (c) 4-acetylphenylboronic acid, (PPh3)4Pd, Cs2CO3, dioxane–
water, reflux, 6 h (100%); (d) 2 M HCl, THF, rt, 1 h (82%); (e) Ac2O, 70 °C, 10 min.
ide (A) conferred PI3Ka selectivity on the aminothiazole core scaf-
folds used herein as exemplified by compounds 18–21 (Table 2) as
well as a related benzothiazole core to give analogue 22 (Fig. 4).23
The binding of A-66, a related aminothiazole taken from a Novartis
patent24 and which contains the (S)-pyrrolidine carboxamide sub-
stituent, has recently been characterized by others using molecular
modelling25 and in vitro mutagenesis.26 Results suggest that the
Scheme 4. Reagents and conditions: (a) CDI, CH2Cl2, reflux, 5 h; (b) RR’NH, Et3N, rt
to 70 °C, 30 min – 24 h.
potency. Pyrazole and ketone analogues (11 and 15) gave similar
results to the sulfones 7a–f. All examples were least potent against
PI3Kb.
With cellularly active compounds available we required an effi-
cient procedure for generating a large number of analogues to test
our hypothesis that substituents extending out to the non-con-
served region of the binding site might result in isoform selective
inhibitors. Our approach was to synthesize urea libraries from
aminothiazoles using a parallel synthesis strategy. The general
2-step protocol is shown in Scheme 4. Reacting aminothiazoles
6a–f, 10 and 14 with carbonyl di-imidazole in dichloromethane
gave intermediates 16 which were either isolated crude by
removing the solvent in vacuo or pure by filtration of the precipi-
tate from the cooled reaction. Intermediates 16 were then allowed
to react with an excess of an amine in tetrahydrofuran or dimeth-
ylformamide. The final products were isolated by filtering the
Figure 4. Isoform selective PI3-kinase inhibitors

I. Bruce et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5445–5450
5449
Table 4
prove valuable in dissecting the relative contributions of individual
isoforms in PI3K signaling pathways.
PK profiles of isoform selective PI3K inhibitors
Compda Doseb
Tmax Cmax
AUC
F
CL
T1/2 Vss
(mg/
kg)
(h)
(l
M)
(l
M.h) (%) (ml/min/
kg)
(h)
(L/
kg)
Acknowledgments
21
24
26
27
10
10
10
5
4
8
05
1
0.28
0.05
1.72
0.49
4.1
0.69
10.8
4.3
21 21
18
57 23
24 4.0
0.7 1.29
0.6 0.4
0.8 1.4
1.8 0.6
We would like to thank Doriano Fabbro for screening com-
pounds in the PIKK assays, Roger Williams of the University of
Cambridge for the PI3Kc crystal structure, Greg Hollingworth for
3
assistance in preparation of the manuscript and our colleagues in
synthetic development and analytical services for support.
a
Results are mean of n = 4 experiments for 21 and 22 and n = 3 experiments for
26 & 27.
b
Dosed p.o. as a 1% suspension in carboxymethyl cellulose (CMC) to female
Wistar rats.
References and notes
1. Vanhaesebroeck, B.; Leevers, S.; Ahmadi, K.; Timms, J.; Katso, R.; Driscoll, P. C.;
Woscholski, R.; Parker, P. J.; Waterfield, M. D. Annu. Rev. Biochem. 2001, 70, 535.
2. Wymann, M.; Pirola, L. Biochim. Biophys. Acta 1998, 1436, 127.
3. Abraham, R. T. DNA Repair 2004, 3, 883.
4. Stein, R. C.; Waterfield, M. D. Mol. Med. Today 2000, 6, 347.
5. Ward, S.; Sotsios, Y.; Dowden, J.; Bruce, I.; Finan, P. Chem. Biol. 2003, 10, 207.
6. Thomas, M.; Owen, C. Cur. Opin. Pharmacol. 2008, 8, 267.
7. (a) Marone, R.; Cmiljanovic, V.; Giese, B.; Wymann, M. P. Biochim. Biophys. Acta
Proteins Proteomics 2008, 1784, 159; (b) Shuttleworth, S. J.; Silva, F. A.; Cecil, A.
R. L.; Tomassi, C. D.; Hill, T. J.; Raynaud, F. I.; Clarke, P. A.; Workman, P. Curr.
Med. Chem. 2011, 18, 2687.
8. Shimada, M.; Murata, T.; Fuchikami, K.; Tsujishita, H.; Omori, N.; Kato, I.; Miura,
M.; Urbahns, K.; Gantner, F.; Bacon, K. WO 2004029055, 2004. See also
reference 18 which describes structural studies with PIK-90.
9. Hollick, J. J.; Rigoreau, L. J. M.; Cano-Soumillac, C.; Cockcroft, X.; Curtin, N. J.;
Frigerio, M.; Golding, B. T.; Guiard, S.; Hardcastle, I. R.; Hickson, I.;
Hummersone, M. G.; Menear, K. A.; Martin, N. M. B.; Matthews, I.; Newell, D.
R.; Ord, R.; Richardson, C. J.; Smith, G. C. M.; Griffin, R. J. J. Med. Chem. 1958,
2007, 50.
10. Camps, M.; Ruckle, T.; Ji, H.; Ardissone, V.; Rintelen, F.; Shaw, J.; Ferrandi, C.;
Chabert, C.; Gillieron, C.; Francon, B.; Martin, T.; Gretener, D.; Perrin, D.; Leroy,
D.; Vitte, P. A.; Hirsch, E.; Wymann, M. P.; Cirillo, R.; Schwartz, M. K.; Rommel,
C. Nat. Med. 2005, 11, 936.
carboxamide group makes favorable contacts with residues in
p110a .
, especially the non-conserved residue Gln859
Aminopropionic acid derivatives led to either delta or gamma
selective compounds depending on the exact nature of the substi-
tuent. For PI3Kd selectivity, screening aryl and heteroaryl amides
led to optimized fragment (B) which was used to provide
compounds 23 and 24. These were both low nM PI3Kd inhibitors
showing >30-fold selectivity over the other three class 1 iso-
forms.27 Moderately selective PI3K
c inhibitors were produced
from alkyl aminopropionic esters; maximum potency and selectiv-
ity was achieved when the t-butyl ester (C) was combined with
aminothiazole 6d to give 25.
To address the potential hydrolytic and metabolic instability of
25 a series of ester replacements was investigated. Fragments D
and E provided a favourable compromise between potency, selec-
tivity and PK profile, particularly when combined with aminothiaz-
oles 6d and 14 to give urea analogues 26 and 27.28,29 Both PI3K
c
11. Robertson, A. D.; Jackson, S.; Kenche, A.; Yaip, C.; Parbaharan, H.; Thompson, P.
WO 200153266, 2001.
12. Herman, S. E.; Gordon, A. L.; Wagner, A. J.; Heerema, N. A.; Zhao, W.; Flynn, J.
M.; Jones, J.; Andritsos, L.; Puri, K. D.; Lannutti, B. J.; Giese, N. A.; Zhang, X.; Wei,
L.; Byrd, J. C.; Johnson, A. J. Blood. 2010, 116, 2078.
13. (a) Chaisuparat, R.; Hu, J.; Jham, B. C.; Knight, Z. A.; Shokat, K. M.; Montaner, S.
Cancer Res. 2008, 68, 8361; (b) Rommel, C. Curr. Top. Microbiol. Immunol. 2010,
1, 279.
inhibitors showed good selectivity (>20 fold) against PI3K
a
although were somewhat less selective against the d and b iso-
forms. Interestingly no PI3Kb selective examples were identified
from this process. Molecular modeling and structural studies indi-
cated that d and
c isoform selectivity of compounds containing
fragments (B) and (C) is most likely derived from interactions be-
tween the terminal functional groups of the urea side chain and
non-conserved amino acids at the outer edge of the binding site.
These findings will be disclosed in due course.
14. The inhibitory activities of compounds against the recombinant PI3-kinase
enzymes were determined by using
a high throughput assay based on
Amersham Pharmacia Biotech’s Scintillation Proximity Assay (SPA) that
measures the transfer of the terminal phosphate of adenosine triphosphate
to phosphatidylinositol. Each well contains 10
dimethylsulphoxide and 20 l assay mix (40 mM Tris, 200 mM NaCl, 2 mM
ethyleneglycol-aminoethyl-tetraacetic acid (EGTA), 15 g/ml
phosphatidylinositol, 12.5 M adenosine triphosphate (ATP), 25 mM MgCl2
0.1
Ci [33P]ATP). The reaction is started by the addition of 20
(40 mM Tris, 200 mM NaCl, 2 mM EGTA containing recombinant GST-p110
ll test compound in 5%
Cellular assay data for selected examples is shown in Table 3.
l
The PI3K
cellular assays; this result is thought be accounted for by activity
against other isoforms (PI3K & PI3Kd) at higher concentrations.
a selective inhibitor 21 was moderately potent in all three
l
l
,
l
ll of enzyme mix
c
c
).
The PI3Kd selective inhibitor 24 was very potent in both mouse
and human B-cell proliferation (BCP) assays which correlates well
with the p110d binding data. The potent PI3Kc inhibitors 26 and 27
The plate is incubated at room temperature for 60 min and the reaction
terminated by the adding 150 l of WGA-bead stop solution (40 mM Tris,
200 mM NaCl, 2 mM EGTA, 1.3 mM ethylene diamine tetraacetic acid (EDTA),
2.6 M ATP and 0.5 mg of Wheat Germ Agglutinin–SPA beads (Amersham
l
l
showed good activity in the oxidative burst (fMLP) assay.
Compounds 21, 24, 26 and 27 were inactive against an in-house
panel of 20 protein kinases. In addition, 27 was also screened
against an external panel of 230 protein kinases and showed
minimal activity on all kinases in the panel with the exception of
two (CLK1, IC50 = 118 nM, and CLK2, IC50 = 117 nM); the functional
relevance of this is unclear. When screened against other members
of the PI3K superfamily no significant inhibition was observed
against mTOR, DNA-PK or Vps34 for the compounds in Table 3
although all showed activity in the PI4KIIIb assay.
The in vivo rat pharmacokinetic profiles for selected compounds
are shown in Table 4. The oral bioavailability (%F) of compound 24
was low as anticipated from its low aqueous solubility (<0.002 g/L
@ pH 6.8). By contrast, the other examples showed greater plasma
exposure.
Biosciences) to each well. The plate is sealed, incubated at room temperature
for 60 min, centrifuged at 1200 rpm and then counted for 1 min using a
scintillation counter. Total activity is determined by adding 10
dimethylsulphoxide (DMSO) and non-specific activity is determined by adding
10 l 50 mM EDTA in place of the test compound.
ll of 5%
l
15. 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.; Camps, M.; Rommel, C.; Jackson, S. P.; Chilvers, E. R.;
Stephens, L. R.; Hawkins, P. T. Blood 2005, 106, 1432.
16. Julius, M. H.; Heusser, C. H.; Hartmann, K. H. Eur. J. Immunol 1984, 14, 753.
17. Walker, E. H.; Perisic, O.; Reid, C.; Stephens, L.; Williams, R. L. Nature 1999, 402,
313.
18. Knight, Z. A.; Gonzalez, B.; Feldman, M. E.; Zunder, E. R.; Goldenberg, D. D.;
Williams, O.; Loewith, R.; Stokoe, D.; Balla, A.; Toth, B.; Balla, T.; Weiss, W. A.;
Williams, R. L.; Shokat, K. M. Cell 2006, 125, 733.
19. Knight, Z. A.; Chiang, G. G.; Alaimo, P. J.; Kenski, D. M.; Ho, C. B.; Coan, K.;
Abraham, R. T.; Shokat, K. M. Bioorg. Med. Chem. 2004, 12, 4749.
20. Ulman, A.; Urankar, E. J. Org. Chem. 1989, 54, 4691.
21. Sargent, T., III; Shulgin, A. T.; Mathis, C. A. J. Med. Chem. 1984, 1071, 27.
22. King, L. C.; Hlavacek, R. J. J. Am. Chem. Soc. 1950, 72, 3722.
23. Ni, Z-J.; Pecchi, S.; Burger, M.; Han, W.; Smith, A.; Atallah, G.; Bartulis, S.;
Frazier, K.; Verhagen, J.; Zhang, Y.; Iwanowicz, E.; Hendrickson, T.; Knapp, M.;
Merritt, H.; Voliva, C.; Wiesmann, M.; Le Grand, D. M.; Bruce, I.; Dale, J.; Lan, J.;
Levine, B.; Costales, A.; Liu, J.; Pick, T.; Menezes, D. WO 2007095588, 2007.
In conclusion we have described progress towards isoform
selective PI3K inhibitors using an automated parallel synthesis
strategy. The compounds in Table 430 are pharmacological tools
exhibiting good isoform selectivity and cellular activity that should

5450
I. Bruce et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5445–5450
24. R. H. Fairhurst, R. H.; P. Imbach, P. WO 2009080705, 2009.
25. Jamieson, S.; Flanagan, J. U.; Kolekar, S.; Buchanan, C.; Kendall, J. D.; Lee, W. J.;
Rewcastle, G. W.; Denny, W. A.; Singh, R.; Dickson, J.; Baguley, B. C.; Shepherd,
P. R. Biochem. J. 2011, 438, 53.
26. Zheng, Z.; Amran, S. I.; Zhu, J.; Schmidt-Kittler, O.; Kinzler, K. W.; Vogelstein, B.;
Shepherd, P. R.; Thompson, P. E.; Jennings, I. Biochem. J. 2012, 444, 529.
27. Bruce, I.; Weitz, G.S. WO 2008000421, 2008.
29. Bloomfield, G. C.; Bruce, I.; Hayler, J. F.; Leblanc, C.; Le Grand, D. M.; McCarthy,
C. WO 2005021519, 2005.
30. Detailed synthetic procedures for compounds 21, 24, 26 & 27 are provided in
the folowing patents; Compound 24 is Example 2 in WO2008000421 (Ref.27).
compound 21 is prepared as described for compound 24 using (S)-pyrrolidine
carboxamide in the final step. Compound 22 is Example 2–33 in
WO2007095588 (Ref.23). Compound 26 is Example
1
in WO2007068473
28. Budd, E.; Hatto, J. D. I.; Hayler, J. F.; Le Grand, D. M.; Valade, B. WO 2007068473,
2007.
(Ref.28) and compound 27 is Example 110 in WO2005021519 (Ref.29).