Discovery of potent, selective small molecule inhibitors of α-subtype of type III phosphatidylinositol-4-kinase (PI4KIIIα)

Article abstract of DOI:10.1016/j.bmcl.2015.05.093

Abstract The discovery and optimisation of novel, potent and selective small molecule inhibitors of the α-isoform of type III phosphatidylinositol-4-kinase (PI4Kα) are described. Lead compounds show cellular activity consistent with their PI4Kα potency inhibiting the accumulation of IP1 after PDGF stimulation and reducing cellular PIP, PIP2 and PIP3 levels. Hence, these compounds are useful in vitro tools to delineate the complex biological pathways involved in signalling through PI4Kα.

Full text of DOI:10.1016/j.bmcl.2015.05.093

Bioorganic & Medicinal Chemistry Letters 25 (2015) 3189–3193
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of potent, selective small molecule inhibitors of
of type III phosphatidylinositol-4-kinase (PI4KIII
a-subtype
a
)
a
b
a
a,c
Piotr Raubo a, , David M. Andrews , Jennifer C. McKelvie , Graeme R. Robb , James M. Smith
,
⇑
Martin E. Swarbrick c, Michael J. Waring a
a AstraZeneca, Mereside, Alderley Park, Macclesfield SK10 4TG, UK
b Cancer Research Technology, London BioScience Innovation Centre, 2 Royal College Street, London NW1 0NH, UK
c Cancer Research Technology, Jonas Webb Building, Babraham Research Campus, Cambridge CB22 3AT, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
The discovery and optimisation of novel, potent and selective small molecule inhibitors of the
of type III phosphatidylinositol-4-kinase (PI4K ) are described. Lead compounds show cellular activity
consistent with their PI4K potency inhibiting the accumulation of IP1 after PDGF stimulation and reduc-
ing cellular PIP, PIP2 and PIP3 levels. Hence, these compounds are useful in vitro tools to delineate the
complex biological pathways involved in signalling through PI4K
Ó 2015 Elsevier Ltd. All rights reserved.
a-isoform
Received 16 March 2015
Revised 28 May 2015
Accepted 29 May 2015
Available online 4 June 2015
a
a
a.
Keywords:
PI4K
a
In vitro tools
IP1
Oncology
Phosphatidylinositol-4-kinases (PI4Ks) catalyse the phosphory-
on cellular concentration of PI4P and PI(4,5)P2 and cellular prolifer-
ation in a panel of cancer cell lines.13 In this communication we
report the results of our work that led to the identification of
lation of phosphatidylinositol (PI) at the D4 position of the inositol
head-group to form phosphatidylinositol-4 phosphate (PI4P).1 PI4P
acts as a binding partner for a wide range of proteins with pleck-
strin homology (PH) or related lipid-binding domains.2,3 PI4P is
also an important signalling molecule as a precursor to other phos-
phoinositides such as phosphatidylinositol-4,5-diphosphate
PI(4,5)P2 and phosphatidylinositol-3,4,5-triphosphate PI(3,4,5)P3
(Scheme 1).3,4 Four PI4K isoforms have been identified in mammals
potent and selective small molecule inhibitors of PI4K
To optimise for suitable probe compounds that would allow us
to evaluate the effect of PI4K inhibition on the PI signalling
pathway, we first screened compounds against PI4K and b
Compounds active in PI4K were subsequently evaluated against
other related enzymes on the PI pathway, including PI3K /b/ /d
and PIP5K /b/ . Finally, with potent and selective PI4K inhibitors
a.
a
a
a
a
c
and are classified as type II (PI4KII
a
and b) and type III (PI4KIII
a
a
c
a
and b, hereafter referred to as PI4K
a
and b). PI4K is the main iso-
a
in hand, we assessed their effect on accumulation of inositol-1-
form responsible for PI4P generation at the plasma membrane and
is localised at the endoplasmic reticulum. PI4Kb is mainly associ-
ated with the Golgi complex where it functions in the generation
of Golgi-derived carriers.5 Recent reports suggest the implication
phosphate (IP1) in NIH3T3-PDGFRb cells.13
We identified the original PI4Ka-selective hit from a screen
against recombinant PI4Kb of approximately 100,000 compounds
selected from the AstraZeneca compound collection based on
structures known to have lipid kinase activity. Active compounds
from the screen then had their IC50 values determined against
of PI4Ka
and PI4P in viral replication.3,6
Pharmacological manipulation of cellular PI4P levels has been a
challenge due to the non-specificity and low potency of known
PI4K inhibitors such as wortmannin, LY294002 and phenylarsine
oxide7; although more recently, reports have emerged describing
PI4K inhibitors from several groups.8–12
PI4Kb and PI4Ka 14
From this screening campaign we identified a
.
moderately selective 2-aminobenzothiazole derivative 1 (Table 1)
as a promising starting point for further evaluation. To get more
detailed insight into structure activity relationships (SAR)
of 2-aminobenzothiazoles, we first investigated the effect of sub-
Recently, we disclosed two distinct series of potent and selec-
tive
a
- and b-subtype PI4K inhibitors and described their effect
stitution of the pyridine ring in 1 on PI4Ka potency and selectivity
(Table 1). Replacement of the ethoxy group with smaller
substituents such as 6-methoxy (2) or 6-methyl (10) led to
⇑
Corresponding author.
E-mail address: piotr.raubo@astrazeneca.com (P. Raubo).
reduction of PI4Ka potency and selectivity over PI4Kb and PI3Ks.
http://dx.doi.org/10.1016/j.bmcl.2015.05.093
0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.

3190
P. Raubo et al. / Bioorg. Med. Chem. Lett. 25 (2015) 3189–3193
R1
R1
O
O
O
OH
O
O
O
R2
OH
OH
O
P
R2
O
P
O
O
OH
OH
O
O
6
5
OH
O
PI4K
1
2
O
P
O
4
O
HO
O
3
O
HO
O
O
OH
PIP5K
R1
R1
O
O
O
O
P
O
O
O
OH
P
R2
O
O
O
O
O
P
O
O
O
O
O
PI3K
O
OH
P
P
R2
O
P
O
O
O
O
O
HO
O
O
O
O
O
O
O
O
HO
O
O
P
O
OH
O
R1=
2
n-C H
R =
17 35
Scheme 1. The PI4K cascade.
Table 1
Table 2
SAR of substituted 2-amino-6-(pyrid-3-yl)-benzothiazoles (1–18)17
SAR of substituted 2-amino-6-heteroaryl-benzothiazoles (19–35)17
N
R
S
N
NH2
R
S
NH2
N
19-35
No
R
PI4K
pIC50
a
PI4Kb
pIC50
PI3K
pIC50
a
/b/
c
/d
PIP5K
pIC50
c
1-18
No
R
PI4K
a
PI4Kb
pIC50
PI3K
pIC50
a
/b/
c
/d
PIP5K
pIC50
c
19 Ph
5.6
6.3
4.9/ND/ND/
ND
<4
pIC50
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
6-EtO–
6-MeO–
6-EtS–
6-MeOCH2–
6-MeNHCH2–
6-iPrO–
6.8
6.2
7.0
6.0
5.6
7.1
6.2
6.2
5.8
5.8
4.8
4.6
6.5
6.5
5.5
5.9
6.0
6.4
6.2
6.0
6.2
6.5
5.5
5.9
6.0
5.8
6.1
6.3
5.8
5.1
6.3
6.0
5.6
4.4
6.0
5.4
5.8/5.2/5.7/5.4
6.1/5.8/6.4/6.1
5.9/5.3/6.0/5.7
5.6/4.8/ND/5.2
4.2/4.1/4.3/4.2
5.3/4.5/4.9/4/9
5.7/4.9/5.4/5.2
5.2/4.4/5.4/4.9
5.5/ND/ND/ND
6.0/5.6/5.9/5.9
5.7/5.4/6.0/6.0
3.8/4.4/5.0/4.9
5.9/5.2/5.9/5.7
5.6/5.1/5.5/5.2
5.7/ND/ND/ND
4.4/4.1/4.7/4.2
6.3/5.9/6.6/5.9
6.6/6.6/6.9/6.5
<4
4.6
<4
<4
<4
<4
4.5
<4
<4
<4
<4
5.0
<4
<4
4.5
<4
4.5
<4
N
N
20
5.0
5.8
5.6
4.8
5.1
5.5
5.6
6.2
6.0
5.2
5.2
5.4
5.0/ND/4.9/4.1 <4
5.7/ND/ND/5.9 4.5
N
21
22
23
24
25
N
6-EtNH–
5.6/ND/ND/
ND
5.6
6-Et(Me)N–
6-HOCH2–
6-Me–
<4/ND/4.5/4.3
4.5
6-CF3–
N
6-MeSO2–
6-HO(CH2)2O–
6-MeO(CH2)2O–
6-MeNHC(O)–
6-Me2NC(O)–
5-i-PrO–
N
5.7/5.2/5.4/5.1 <4
4.6/4.3/5.0/4.5 4.4
N
O
HN
HN
5-MeSO2–
26
27
28
6.5
5.6
4.9
6.2
5.9
6.0
5.4/5.2/5.9/5.0 4.8
5.2/4.8/5.6/4.9 4.2
4.9/4.5/5.6/4.6 <4
O
N
N
N
The 6-ethylthio analogue 3 was found to have a similar profile to
the ethoxy compound 1. Increasing the size of the alkoxy sub-
stituent was tolerated in several cases (e.g., in 6, 13 and 14), but
this did not improve the selectivity profile for these compounds.
However, the 6-methoxymethyl derivative 4 was more potent at
N
N
N
29
30
31
5.6
5.5
5.8
5.9
5.8
5.7
5.1/4.8/5.6/5.1 <4
4.9/4.5/5.5/4.7 <4
4.9/4.9/5.8/4.9 4.6
PI4Kb than PI4Ka. Diminished selectivity was also observed with
N
N
6-aminopyridyl derivatives 7 and 8. The presence of electron with-
drawing groups such as a trifluoromethyl (11), methylsulfonyl (12)
and carbamoyl (15, 16) at the 6-position of pyridine ring gave
N
N
lower PI4Ka IC50 values for these compounds and hence, in most
cases, reduced their selectivity over PI4Kb and PI3K kinases. An
exception to this was the amide 16, which although less potent

P. Raubo et al. / Bioorg. Med. Chem. Lett. 25 (2015) 3189–3193
3191
less potent than 1 in PI4K
a
but showed equipotent PI4Kb activity
Table 2 (continued)
to 1. The corresponding phenyl derivative 19 showed a similar pro-
file. The replacement of the pyrid-3-yl substituent in 21 with other
No
R
PI4K
pIC50
a
PI4Kb
pIC50
PI3K
pIC50
a
/b/
c
/d
PIP5K
pIC50
c
6-membered heterocyclic rings generally led to reduction of PI4K
a
N
potency and selectivity. Interestingly, 2-hydroxypyrid-4-yl (2-pyri-
done) derivative 26 was about 10-fold more potent and selective
when compared to the corresponding pyrid-4-yl analogue 22.
The replacement of the pyrid-4-yl substituent in 22 with 1-
32
33
4.6
5.6
5.7
4.8/ND/5.4/4.7 4.2
5.0/5.0/6.0/5.4 <4
N
N
N
5.1
6.2
6.8
methyl-pyrazol-3-yl in 27 resulted in retention of PI4K
a potency
N
34
35
5.7
6.1
4.9/4.8/5.9/4.7 ND
5.1/4.6/5.7/4.7 <4
N
and selectivity. Isomeric pyrazoles 28 and 29 showed a comparable
profile to 27. However, increasing the size of the N-alkyl sub-
stituent in 29 with b-branching (relative to the N1 pyrazole nitro-
gen), as exemplified with the N-cyclopropylmethyl (34) and N-
N
Ph
N
benzyl (35) analogues, resulted in improved PI4K
a inhibitory
activity and enhanced selectivity against other kinases.
A similar effect on potency and selectivity was observed with N-
alkylated analogues of the 2-pyridone 26 (Table 3). While small
substituents such as methyl (36) or ethyl (37) revealed a similar
profile to the reference compound 26, the corresponding N-cyclo-
Table 3
Effect of N-substitution of the 2-pyridone in 26 on PI4K
a
potency and selectivity17
O
R
N
propylmethyl analogue 38 showed excellent potency at PI4K
a
S
(pIC50 8.2) and good selectivity over PI4Kb, PI3K and PIP5K iso-
forms (>100 fold). Moreover, 38 demonstrated a very good selec-
tivity profile when tested in the Millipore kinase panel,
consisting of 377 different kinases and related targets (Fig. 1),
showing activity against only 3 kinases (FGR 98%, ZIPK 72% and
NH2
N
26, 36-39
No
R
PI4K
pIC50
a
PI4Kb
pIC50
PI3K
a
/b/
c
/d
PIP5K
pIC50
c
pIC50
STK17A 68% of inhibition at 1
lM concentration). Pleasingly, 38
26
36
37
38
39
H
Me
Et
6.5
6.3
6.5
8.2
6.2
6.2
6.5
6.5
5.8
5.0
5.4/5.2/5.9/5.0
5.6/4.2/ND/5.3
5.2/4.3/5.5/5.0
5.2/<3.9/5.8/4.9
4.6/<3.7/5.0/<3.7
4.8
<4
<4
<4.1
<4
inhibited the accumulation of IP1 (IC50 0.52
lM) and reduced cellu-
lar PIP, PIP2 and PIP3 levels in NIH3T3-PDGFRb cells.13
Furthermore, 38 was tested for growth inhibition in the panel of
183 cancer cell lines inhibiting 91 cell lines with pGI50 >5.0.13
In order to understand the observed differences in selectivity
for these compounds, several compounds were docked into homol-
c-PrCH2
2-MeO(CH2)2O
ogy models of PI4K
the MOE16 software package based on available crystal structures
of PI3K
, as described previously.13 Ligand docking was also per-
a and PI4Kb. Homology models were built using
at PI4K
against PIP5K
a
showed an encouraging selectivity profile. Selectivity
/b/
was very good across the series.15
c
a
c
Next, we explored the replacement of the pyrid-3-yl substituent
at the 6-position of 2-aminobenzothiazole ring (Table 2). The
unsubstituted 3-pyridyl analogue 21 was an order of magnitude
formed in MOE with the two kinase-hinge hydrogen bonds set as
a pharmacophore restraint. An induced-fit protocol was employed
to allow limited movement of protein side-chains within 6 Å of the
Figure 1. Selectivity profile of 38, 40 and 43 in the kinase panel (% inhibition at 1 lM concentration).

3192
P. Raubo et al. / Bioorg. Med. Chem. Lett. 25 (2015) 3189–3193
Figure 2. (a) Compound 38 docked into the PI4K
a
homology model. (b) Compound 36 docked into the PI4Kb homology model. Notice that the subpocket occupied by the
cyclopropyl moiety of 38 is much smaller in b than in
a. Larger groups occupying this pocket confer isoform selectivity.
Table 4
N-Substituted analogues of 2-pyridone 3817
O
B
S
N
O
O
N
a
NH2
R
S
45
46
R
S
N
N
NH2
H
N
a, b
O
B
S
38, 40-44
PI4Kb PI3K
O
NHBoc
No
R
PI4K
a
a/b/
c
/
PIP5K
pIC50
c
IP1
pIC50
N
pIC50
pIC50
d pIC50
38
H
8.2
5.8
5.2/<3.9/
5.8/4.9
<4.1
6.3
6.1
6.2
Scheme 2. Reagents and conditions: (a) R-X (X = Br,Cl), Pd-118, K2CO3 or K3PO4,
MeCN–H2O, 80–100 °C, 1–2 h, MW; (b) 20% CF3COOH in CH2Cl2, room temperature,
1–2 h (overall yields: 8–53%).
<3.7/<3.7/
4.8/4.0
O
40
41
8.3
8.1
6.4
6.5
<4
Me
Me
<3.7/<3.7/
4.2/<3.7
<4
O
O
(R)
(S)
Br
S
N
a
NH2
<3.7/<3.7/
<3.7/<3.7
42
7.4
5.6
4.6
5.1
R
S
N
47
NH2
4.1/<3.7/
4.3/<3.7
4.0/<3.7/
5.0/<4.1
a, b
SO2
COOH
43
44
8.7
9.0
6.2
6.6
<4
6.2
6.5
Br
S
NHBoc
ND
N
48
Scheme 3. Reagents and conditions: (a) R-B(OH)2 or R-BPin, Pd-118, K2CO3 or
K3PO4, MeCN–H2O, 80–100 °C, 1–2 h, MW; (b) 20% CF3COOH in CH2Cl2, room
temperature, 1–2 h (overall yields: 4–64%).
ligand. Figure 2 shows the common binding mode of this series to
both PI4K isoforms. The amino-benzothiazole unit binds to the
‘hinge’ of the kinase, making two strong hydrogen-bonds. A third
hydrogen-bond is made to the catalytic lysine at the back of the
pocket (K1794, conserved between isoforms). We postulated that
the crucial difference in structure that determines selectivity of
38, compared to the unselective 36, is the presence of an arginine
(R1771) in beta, near the back of the pocket; the equivalent residue
in the alpha being aspartic acid (D1771), see Figure 2. The arginine,
being a bulkier residue, reduces the size of the subpocket near the
catalytic lysine. This, together with some small structural differ-
ences between isoforms, means that while there is a pocket in
Br
a
S
Br
S
N
NH2
NHBoc
NHBoc
N
47
48
b
b
O
B
O
B
a
S
S
O
O
PI4Ka large enough to bind the cyclopropyl moiety of 38 with high
NH2
N
N
affinity, no such pocket exists in PI4Kb. Compounds with smaller
substituents, such as 36, are accommodated in both structures
and as such bind with approximately equal potency to both iso-
forms. Docking of 38 in the b-isoform indicates that the ligand
shifts to accommodate the larger group in the smaller pocket and
the hydrogen-bond to K1794 cannot be made, hence lower PI4Kb
affinity.
The potent and selective probe 38 could be optimised further by
substitution of the NH2 group (Table 4). A number of substituents
were tolerated at this position as represented by the
46
45
Scheme 4. Reagents and conditions: (a) Boc2O, Et3N, DMAP, CH2Cl2, rt, 3 days, 70–
89%; (b) Pin2B2, Pd(dppf)2Cl2, AcOK, 1,4-dioxane, 95 °C, 5 h, 83–100%.
tetrahydropyranyl methyl derivative 40. The chiral (R) analogue
41 was equipotent with 40, whereas for the corresponding (S)
enantiomer 42 PI4Ka potency was reduced by 6-fold. However, a
gain in PI4K potency and further improvement in selectivity
a

P. Raubo et al. / Bioorg. Med. Chem. Lett. 25 (2015) 3189–3193
3193
O
N
O
Br
N
S
N
a, b
R
S
N
Cl
O
B
N
H
O
49
50
40-44
Scheme 5. Reagents and conditions: (a) R-NH2, Et3N, MeCN, 180 °C, 2 h, MW; (b) 50, Pd-118, K2CO3, MeCN–H2O, 120 °C, 1–2 h, MW (33–76% for 2 steps).
was achieved with the corresponding sulfone derivative 43 and the
acid 44. Benzothiazoles 40 and 43 were tested in the Millipore
kinase panel consisting of 124 different kinases (Fig. 1) showing
only weak activity against human PAK2 (49% and 53% inhibition
References and notes
1. Roth, M. G. Physiol. Rev. 2004, 84, 699.
2. Di Paolo, G.; De Camilli, P. Nature 2006, 443, 651.
3. Delang, L.; Paeshuyse, J.; Neyts, J. Biochem. Pharmacol. 2012, 84, 1400.
4. Graham, T. R.; Burd, C. G. Trends Cell Biol. 2011, 21, 113.
5. Sridhar, S.; Patel, B.; Aphkhazava, D.; Macian, F.; Santambrogio, L.; Shields, D.;
Cuervo, A. M. EMBO J. 2013, 32, 324.
6. Vaillancourt, F. H.; Pilote, L.; Cartier, M.; Lippens, J.; Luizzi, M.; Bethell, R.;
Cordingley, M. G.; Kukolj, G. Virology 2009, 387, 5.
7. Balla, T.; Balla, A. Trends Cell Biol. 2006, 16, 351.
8. LaMarche, M. J.; Borawski, J.; Bose, A.; Capacci-Daniel, C.; Colvin, R.; Dennehy,
M.; Ding, J.; Dobler, M.; Drumm, J.; Gaither, L. A.; Gao, J.; Jiang, X.; Lin, K.;
McKeeever, U.; Puyang, X.; Raman, P.; Thohan, S.; Tommasi, R.; Wagner, K.;
Xiong, X.; Zabawa, T.; Zhu, S.; Wiedmann, B. Antimicrob. Agents Chemother.
2012, 56, 5149.
9. Bianco, A.; Reghellin, V.; Donnici, L.; Fenu, S.; Alvarez, R.; Barhuffa, C.; Peri, F.;
Pagani, M.; Abrignani, S.; Neddermann, P.; De Francesco, R. PLoS Pathog. 2012,
8, e1002576.
at 1 lM concentration for 40 and 43, respectively). Additionally,
both 40 and 43 exhibit reduced STK17A inhibition (17% and 4%
inhibition for 40 and 43, respectively vs 68% inhibition for 38)
and FGFR1 activity was removed entirely. Compounds 40, 41, 43
and 44 displayed a similar inhibitory effect to 38 in the IP1 cellular
assay suggesting, that these compounds may also be considered as
useful in vitro tools to study the biological pathways involved in
signalling through PI4Ka and PI4P.
All compounds described in this communication were prepared
using, as a key step, Suzuki cross-coupling reaction between
selected heteroaryl halides or boronic acids/boronates and corre-
sponding 2-amino-benzothiazole derived bromides or boronate
esters (45, 46, 47 or 48) (Schemes 2 and 3).
10. Leivers, A. L.; Tallant, M.; Shotwell, J. B.; Dickerson, S.; Leivers, M. R.; McDonald,
O. B.; Gobel, J.; Creech, K. L.; Strum, S. L.; Mathis, A.; Rogers, S.; Moore, C. B.;
Botyanszki, J. J. Med. Chem. 2014, 57, 2091.
Required building blocks 45, 46 and 48 were synthesised from
the commercially available 2-amino-5-bromobenzothiazole 47 by
Boc protection of the amino group and the boronic ester were
introduced by prior or subsequent palladium catalysed coupling
(Scheme 4).
N-Substituted analogues of 38 (compounds 40–44) were made
from commercial 2-chloro-5-bromobenzothiazole 49 in a 2-step,
one-pot sequence that starts with nucleophilic substitution of the
chlorine at the 2-position of the benzothiazole ring with a corre-
sponding amine following by the Suzuki cross-coupling reaction
with the 2-pyridone boronate 50 (Scheme 5).
11. Keaney, E. P.; Connolly, M.; Dobler, M.; Karki, R.; Honda, A.; Sokup, S.; Karur, S.;
Britt, S.; Patnaik, A.; Raman, P.; Hamann, L. G.; Wiedmann, B.; LaMarche, M. J.
Bioorg. Med. Chem. Lett. 2014, 24, 3714.
ˇ
12. Mejdrová, I.; Chalupská, D.; Kögler, M.; Sala, M.; Placková, P.; Baumlová, A.;
Hrebabecky, H.; Procházková, E.; Dejmek, M.; Guillon, R.; Strunin, D.; Weber, J.;
Lee, G.; Birkus, G.; Mertlíková-Kaiserová, H.; Boura, E.; Nencka, R. J. Med. Chem.
2015, 58, 3767.
13. Waring, M. J.; Andrews, D.; Flemington, V.; Garcia, R.; Lenaghan, C.; McKelvie,
J.; Maman, S.; Raubo, P.; Robb, G.; Roberts, K.; Rowlinson, R.; Smith, J.;
Swarbrick, M.; Treinies, I.; Winter, J. G.; Wood, R. Chem. Commun. 2014, 5388.
14. Protocols for PI4 K
supplementary information of Ref. 13. Protocols for PI3 Kb/
reported for PI3 K
a
, PI4 Kb, PI3 K
a
, PIP5 K
c
and IP1assays are described in the
c
/d assays were as
a
.
15. Data presented for PIP5 Kc isoform are indicative of activity against PIP5 Kab
isoforms.
In conclusion, a novel series of potent and selective small mole-
16. Molecular Operating Environment (MOE), 2013.08; Chemical Computing
Group Inc., 1010 Sherbooke St. West, Suite #910, Montreal, QC, Canada, H3A
2R7, 2015.
17. pIC50s are mean values of up to nine independent experiments. ND means no
data available.
cule inhibitors of PI4K
a
was discovered. Several compounds
potency and selectivity against other
showed excellent PI4K
a
kinases. These compounds might serve as good quality biochemical
tools for evaluation and interpretation of complex biological cas-
cades involved in signalling through PI4Ka.