Discovery and antiplatelet activity of a selective PI3Kβ inhibitor (MIPS-9922)

Article abstract of DOI:10.1016/j.ejmech.2016.06.010

A series of amino-substituted triazines were developed and examined for PI3Kβ inhibition and anti-platelet function. Structural adaptations of a morpholine ring of the prototype pan-PI3K inhibitor ZSTK474 yielded PI3Kβ selective compounds, where the selectivity largely derives from an interaction with the non-conserved Asp862 residue, as shown by site directed mutagenesis. The most PI3Kβ selective inhibitor from the series was studied in detail through a series of in?vitro and in?vivo functional studies. MIPS-9922, 10 potently inhibited ADP-induced washed platelet aggregation. It also inhibited integrin α_IIbβ₃activation and α_IIbβ₃dependent platelet adhesion to immobilized vWF under high shear. It prevented arterial thrombus formation in the in?vivo electrolytic mouse model of thrombosis without inducing prolonged bleeding or excess blood loss.

Full text of DOI:10.1016/j.ejmech.2016.06.010

Accepted Manuscript
Discovery and antiplatelet activity of a selective PI3Kβ inhibitor (MIPS-9922)
Zhaohua Zheng, Jo-Anne Pinson, Simon J. Mountford, Stephanie Orive, Simone
M. Schoenwaelder, David Shackleford, Andrew Powell, Erin M. Nelson, Justin R.
Hamilton, Shaun P. Jackson, Ian G. Jennings, Philip E. Thompson
PII:
S0223-5234(16)30487-1
10.1016/j.ejmech.2016.06.010
EJMECH 8671
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 16 March 2016
Revised Date: 6 June 2016
Accepted Date: 8 June 2016
Please cite this article as: Z. Zheng, J.-A. Pinson, S.J. Mountford, S. Orive, S.M. Schoenwaelder, D.
Shackleford, A. Powell, E.M. Nelson, J.R. Hamilton, S.P. Jackson, I.G. Jennings, P.E. Thompson,
Discovery and antiplatelet activity of a selective PI3Kβ inhibitor (MIPS-9922), European Journal of
Medicinal Chemistry (2016), doi: 10.1016/j.ejmech.2016.06.010.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Discovery and antiplatelet activity of a selective PI3Kβ inhibitor
(MIPS-9922)
Zhaohua Zheng1,2, Jo-Anne Pinson1, Simon J. Mountford1, Stephanie Orive2, Simone M.
Schoenwaelder2, David Shackleford3, Andrew Powell3, Erin M. Nelson2, Justin R. Hamilton2,
Shaun P. Jackson2, Ian G. Jennings1 and Philip E. Thompson1*
1Medicinal Chemistry, Monash Institute of Pharmaceutical Sciences, Monash University,
Parkville, Victoria 3052, AUSTRALIA
2Australian Centre for Blood Diseases, Monash University, Alfred Medical Research and
Education Precinct (AMREP), 89 Commercial Road, Prahran, Victoria 3004 AUSTRALIA
3Centre for Drug Candidate Optimisation, Faculty of Pharmacy and Pharmaceutical
Sciences, Monash University, Parkville, Victoria 3052, AUSTRALIA
1

ACCEPTED MANUSCRIPT
Abstract
A series of amino-substituted triazines were developed and examined for PI3Kβ inhibition
and anti-platelet function. Structural adaptations of a morpholine ring of the prototype pan-
PI3K inhibitor ZSTK474 yielded PI3Kβ selective compounds, where the selectivity largely
derives from an interaction with the non-conserved Asp862 residue, as shown by site directed
mutagenesis. The most PI3Kβ selective inhibitor from the series was studied in detail through
a series of in vitro and in vivo functional studies. MIPS-9922, 10 potently inhibited ADP-
induced washed platelet aggregation. It also inhibited integrin αIIbβ3 activation and αIIbβ3
dependent platelet adhesion to immobilized vWF under high shear. It prevented arterial
thrombus formation in the in vivo electrolytic mouse model of thrombosis without inducing
prolonged bleeding or excess blood loss.
Table of Contents Graphic
Highlights
•
•
•
•
Potent and selective inhibitors of PI3Kβ.
Distinct mechanisms underpinning selectivity compared to known inhibitors
Anti-platelet activity shown in platelet aggregation, activation and adhesion in vitro
Anti-thrombotic activity in vivo in mouse electrolytic injury model without effects on
haemostasis.
2

ACCEPTED MANUSCRIPT
1. Introduction
The phosphoinositide 3-kinases (PI3Ks) are a family of lipid kinases which phosphorylate the
3-hydroxyl group of the inositol ring of phosphatidylinositols. In vivo, Class I PI3Ks convert
substrate PIP2 to PIP3 which acts as a second messenger in cellular signal transduction. Class
I PI3Ks regulate various vital cellular functions including cell adhesion, migration, survival,
proliferation and growth [1]. All Class I PI3K isoforms are expressed in platelets but of the
four isoforms, PI3Kβ has been shown to play an important role in thrombotic processes and
thus represents a potential target for anti-thrombotic therapy [2-6]. The PI3Kβ selective
inhibitor TGX-221 (Figure 1) was the first pharmacological agent to show the importance of
PI3Kβ blockade in platelets [3]. TGX-221 blocks platelet responses to activation with weak
agonists such as ADP, collagen and TRAP, inhibiting integrin αIIbβ3 activation and platelet
aggregation. TGX-221 also showed effective blockade of shear induced platelet adhesion and
activation. TGX-221 was an effective anti-thrombotic in animal models of thrombosis in rat
and rabbit and critically showed little increase in bleeding time at therapeutic doses[7, 8].
This latter point is pivotal to the prospect of developing anti-thrombotic agents superior to
aspirin, clopidogrel or the newest P2Y12 inhibitors, prasugrel and ticagrelor [9].
Subsequently, an analogue of TGX-221, AZD6482 (Figure 1) which shows largely parallel
effects to TGX-221 was progressed through studies in dogs and then elevated to studies
human subjects [10, 11]. AZD6482 has a rapid onset of action and a short half-life in rats,
dogs, and humans, so sustained platelet inhibition requires continuous infusion of the
compound. Other PI3Kβ inhibitors that have been progressed for oncology indications,
including the AZD8186, SAR260301 and GSK2636771.
The need for further investigation of PI3Kβ inhibition as a mechanism for anti-platelet
therapy has been emphasized by recent studies in PI3Kβ-null mice which showed that
thrombi in these animals had reduced stability, warning of potential embolism-related
3

ACCEPTED MANUSCRIPT
secondary ischemic events.[12] It is therefore important to identify structurally distinct
PI3Kβ inhibitors that can be utilized to confirm their likely utility as anti-thrombotic drugs.
Figure 1. Structures of reported PI3Kβ inhibitors
We recently communicated a series of PI3K inhibitors of general structure I, (Figure 1)
derived from the pan-PI3K inhibitor ZSTK474 that included examples that were strongly
PI3Kβ selective [13]. This group of inhibitors, which share little structural similarity to the
TGX-221 class, achieves affinity and selectivity for PI3Kβ in a demonstrably different
manner to TGX-221 and other recently reported PI3Kβ selective inhibitors [14]. Where all
other PI3Kβ selective inhibitors engage a cryptic pocket in the P-loop of the kinase active site,
4

ACCEPTED MANUSCRIPT
these compounds interact with a single non-conserved glutamate residue (Asp862) of the
PI3Kβ binding site for selectivity as demonstrated by site-directed mutagenesis studies [13].
Here we report the first investigation of the compound class as potential anti-thrombotic
agents. We have examined the structure activity relationships governing the pharmacophoric
aminoacylpiperazine moiety and identified potent inhibitors of human platelet aggregation in
vitro consistent with PI3Kβ inhibitory potency. Focussing on compound 10, now named
MIPS-9922, we found that it is effective in blocking PI3K mediated activation of platelet
glycoprotein αIIbβ3 activation and platelet adhesion in vitro. MIPS-9922 also prevents clot
formation in a mouse model of thrombosis without causing excessive bleeding or blood loss
in mice. Finally, MIPS-9922 also shows promising ADME properties contributing to a good
in vivo half-life and oral bioavailability in rats. The results strengthen the validation of PI3Kβ
as an anti-thrombosis target. Moreover these results encourage further development of MIPS-
9922 or analogues in the setting of thrombosis or other diseases in which PI3Kβ is implicated.
2. Results
2.1. Chemistry
In our previous communication we reported upon the synthesis of aminoacylpiperazine
derivatives I (Figure 1),[13] some of which (3, 4, 6, 7, 10, 11, 15, 18) are included in this
study. Here we have extended our study of that class to include variants with unusual amino
acids and others with modifications to the piperazine portion, such as N-linked 1-piperidinyl
and C-linked 4-piperidinyl derivatives. The compounds were designed to test the principle
that an appropriately placed pendant amino group was critical to PI3Kβ potency and
selectivity with the aim of identifying a potent, selective inhibitor for pharmacological studies.
The synthetic routes to those compounds are summarized in Scheme 1. The N-linked
compounds – piperazine (3-18), were each synthesized via Boc-piperazine substitution of the
chloro-substituted precursor, 1 to give 2 as previously described,[13] with subsequent
5

ACCEPTED MANUSCRIPT
elaboration via Boc- or Fmoc-amino acid couplings and then deprotection. Similarly
piperidine (19-26) and aminopiperidine (27-28) analogues were prepared by reaction of 1
with corresponding Boc-protected amines and deprotection to give amines 19, 22, 24, 25 and
27. The C-linked piperidines (31-34) were prepared from 1 by Suzuki coupling using Boc-
protected (1,2,3,6-tetrahydropyridin-4-yl)boronic acid to give 29 followed by reduction to 30
and then similarly elaborated by amide coupling to give 31-34.[15]
Scheme 1. Synthesis of triazine PI3K inhibitors. i, amine, K2CO3, DMF, mw, 20 min; ii,
TFA/DCM, 30 min.; iii Boc- or Fmoc-amino acid, HCTU, DIPEA, DMF, 2h; iv, N-Boc-1,2,3,
6

ACCEPTED MANUSCRIPT
6-tetrahydropyridine-4-boronic acid pinacol ester, PdCl2(dppf), Na2CO3, dioxan, reflux, 2h; v,
H2, Pd-C, H-cube.
2.2 Biochemical Assays
The compounds were all tested for inhibition of PI3Kβ activity as the primary screen. In the
first series, data for 8 compounds previously published is included for SAR comparison. In
the first group of compounds (3 – 18) bearing a piperazine ring, the data show the quite
profound effect of substituents on the pendant amino acid. While the compounds containing
glycine (3), β-ala (4), and γ-aminobutyric acid (5) chains all retain high potency for PI3Kβ,
the introduction of branching groups has a dramatic effect with just compound 6, retaining
strong activity from the methyl substituted analogues. The requirement for an L-α amino
acid appears to carry through with the aromatic amino acids. The L-Phe substituted
compound 10 has an IC50 of 63nM, but the D-Phe (11), β3hPhe (12) and β3hPhg (13)
analogues are all poor inhibitors. One feature originally identified was the role chirality
played in determining potency for PI3Kβ (L-enantiomers).[13] However, the cyclic D-Pro
containing compound 16 retained very high potency. The cyclic γ-amino acid derivative 17
was much less active.
Of the other compounds, the aminoethylpiperidine, 24 and the C-linked piperidine 31 showed
the strongest inhibitory potency, but not to the level of the piperazine compounds, indicating
that the changes in conformation related to swapping a piperazine nitrogen for a carbon have
a deleterious impact on binding. In the series 19-26, neither the positional substitution nor the
spacing the amino group provides compounds of strong potency. The insertion of an amino
group as in 27 and 28 is deleterious to activity also. The same is true for compounds 31-34;
the comparison of these to compounds 3, 6, 7 and 15 shows that this change of piperazine to
7

ACCEPTED MANUSCRIPT
piperidine costs an order of magnitude in affinity. Collectively, it appears that the less active
compounds, for which the IC50 values sit around 1 ꢀM are not readily able to make the
positive interactions with the binding site which sustain the high affinity of the L-α-
aminoacylpiperazines.
We extended our evaluation of selected compounds to the PI3Kδ isoforms, for which we had
previously reported compounds 4 and 7 to potently inhibit.[13] The inhibitory potency
against PI3Kδ (and therefore PI3Kβ selectivity) varied markedly from the most PI3Kβ
selective compound 10 (35-fold) to several PI3Kδ selective compounds 7-8. Compound 16
bearing a D-Proline substituent was a potent inhibitor of both PI3Kβ and PI3Kδ. Compound
24 was a poor inhibitor of PI3Kδ so retained a degree of PI3Kβ selectivity, while for 31 the
selectivity had swapped to PI3Kδ. Interestingly the introduction of the chiral L-Ala in 32 was
tolerated for PI3Kβ and not PI3Kδ comparable to the 3 and 6, but in contrast the D-Ala
substitution 33 was disfavoured for both isoforms.
Table 1 IC50 values for compounds of Scheme 1 against PI3K activity.
PI3K isoform IC50 (nM)a
Compound
PI3Kβ
6
PI3Kδ
ZSTK-474
TGX-221
52
3b
4b
5
35
105
58
98
81
77
6b
7b
8
31
490
70
1000
1263
1709
109
1251
9
8

ACCEPTED MANUSCRIPT
10 (MIPS-9922)b
63
> 10000
7052
1797
> 10000
26
2200
3600
2841
17000
9100
390
11b
12
13
14
15b
16
17
18b
19
20
21
22
23
24
25
26
27
28
31
32
33
34
36
12
3210
50
330
670
c
955
-
1328
521
-
-
811
-
428
-
190
1000
554
-
1210
2114
1060
360
-
-
-
120
1090
740
-
540
3200
2000
aIC50 values are means of at least two duplicate experiments (Kinase-Glo). Standard errors were all
b
c
within 25% of the mean. Assays were performed using 10 µM ATP. Data from Ref. 12 - = not
determined
We have previously described the use of in vitro mutagenesis in order to assess the
underpinning mechanisms of PI3K isoform selectivity,[14, 16] and have shown in our
previous communication that the key residue for PI3Kβ selectivity in the case of compounds
9

ACCEPTED MANUSCRIPT
3, 6 and 10 was Asp862 and mutation of that residue to a glutamine, caused an 80-fold
change drop in potency for compound 10.[13] We also noted that mutation of the nearby
Glu858 to histidine impacted upon binding of 3. Glu858 of PI3Kβ corresponds to an acidic
residue Asp832 in PI3Kδ and the capacity to interact with that residue may be the reason that
PI3Kβ selectivity over PI3Kδ is difficult to achieve. We extended our examination to the
compounds 24 and 31, which have comparable structures to compound 3 against our panel of
PI3Kβ mutant forms noting that compound 31 has a preference for PI3Kδ. The inhibition of
the in vitro mutants by 24 showed fundamentally the same profile as compound 3 (Gly)
losing most binding at D862Q but also at E858H;. In contrast, 31 showed a markedly
different profile, with the compound most impacted upon by mutation of E858H and also
Y778I, although reduced inhibitory potency of D862Q was still observed. This result shows
that the two analogues (3 and 31) that differ by a single CH to N modification have quite
different interactions with the binding site. The influence of tyrosine mutant mutation
observed for Y778I is more reminiscent of TGX221 and the other reported PI3Kβ inhibitors
and that from the perspective of the enzyme 31 may have more in common with the
“propeller style” inhibitors.[14]
10

ACCEPTED MANUSCRIPT
Figure 2. Effect of PI3K isoform mutations (β-α) on compound inhibition profile at selected
concentrations: (A) 3, 70nM, (B) 24 200nM, (C) 31, 200nM.
In summary, from our study emerged a series of compounds that showed promising levels of PI3Kβ
inhibitory potency. In addition, we had previously shown that compounds 3 and 10 were potent
inhibitors of AktS473 phosphorylation in intact MDA-MB-468 cells indicative of cell permeability.[13]
2.3. Pharmacology
The hallmarks of PI3Kβ inhibition in relation to platelet function have been established
around the studies of TGX-221, which stands as a benchmark compound for comparison in
our studies. TGX-221 blocks platelet responses to activation with weak agonists such as ADP
11

ACCEPTED MANUSCRIPT
(at 5ꢀM)[17] and TRAP (1 ꢀM) and blocked shear induced platelet adhesion (at 0.5 ꢀM) and
integrin αIIbβ3 activation. TGX-221 abolished stable platelet adhesion and thrombus
formation on a vWf substrate at elevated shear rates (1,800 s–1). TGX-221 was an effective
anti-thrombotic in animal models of thrombosis and critically, showed little increase in
bleeding time at therapeutic doses[7, 8].
2.3.1 Inhibition of platelet function in vitro.
The effect of selected compounds 6, 10, 15, 16, 18 and 24 on platelet function was evaluated
by aggregation assays using ADP and TRAP as agonists of washed human platelets using a
96 well-plate based method.[18] Compounds 10 (Figure 1), 15 and 16 all strongly inhibited
platelet aggregation induced by threshold concentrations of ADP (5 µM) in a concentration-
dependent manner with IC50 values of approximately 30 nM. Compounds 24 and 6 showed
comparable and potent inhibition but could not surmount the aggregatory response
completely. The poorest inhibitor was compound 18. (Figure 3) Against the PAR-1 agonist,
TRAP (1 ꢀM) comparable profiles were observed (Supplementary Figure 1). It was also
noted that 18 invoked a strong pro-aggregatory response to sub-threshold ADP induced
aggregation, even at very low concentrations of inhibitor (data not shown) which may explain
the poor inhibitory potency observed.
Figure 3. Dose response curves for inhibition of ADP-induced washed platelet aggregation.
12

ACCEPTED MANUSCRIPT
With the platelet level responses identified consistent with PI3Kβ inhibition, we focussed our
attention upon compound 10, coded as MIPS-9922. It displayed the best PI3Kβ selectivity
profile, and compares favourably to TGX-221 - 75-fold selectivity over PI3Kα, 35-fold over
PI3Kδ and is inactive against PI3Kγ. Moreover, as described above MIPS-9922 showed
potent blockade of Akt phosphorylation in MDA-MB-468 cells and high selectivity over other
kinases in closely related pathways when screened against a panel of 96 kinases. [13]
A
B
Figure 4. Dose-response curve of MIPS-9922 inhibition of ADP induced platelet aggregation, (a) in
washed platelets, (b) in human PRP.
We examined the ability of MIPS-9922 to inhibit platelet aggregation in a conventional
platelet aggregometer. MIPS-9922 potently inhibited ADP induced washed platelet
aggregation with an IC50 value of 56 nM. (Figure 4a) While MIPS-9922 at a concentration of
0.5 μM was sufficient to block aggregation induced by 5 and 10 μM ADP, at 25 µM ADP the
same concentration of MIPS-9922 had a marginal effect with approximately 15% inhibition
of aggregation observed (Supplementary Fig. 2) which matches what was observed for TGX-
221. When tested in human PRP, MIPS-9922 again blocked ADP-induced aggregation in a
concentration-dependent manner (Figure 4b). The concentrations required to cause the same
level of reversibility were higher in PRP. The IC50 value of MIPS-9922 in PRP was 2.8 µM,
which is a 50-fold increase as compared to washed platelets. It was asserted that protein
13

ACCEPTED MANUSCRIPT
binding by MIPS-9922 might explain the difference between the washed platelet and PRP
IC50 values.
As described above, effect of PI3Kβ inhibition on platelet aggregation is thought to be due to
inhibition of αIIbβ3 activation. Analysis of αIIbβ3 activation in the presence of MIPS-9922 was
performed using flow cytometry as measured by Oregon-Green labelled fibrinogen (OG-FNG)
binding[6]. After platelet stimulation by threshold concentration of ADP (5 µM) or TRAP
(0.5 µM), integrin αIIbβ3 activation was measured by FACS analysis of OG-FNG binding to
the platelets. MIPS-9922 effectively inhibited αIIbβ3 activation in a concentration-dependent
manner for both ADP and TRAP (Fig. 5a and 5b). The inhibition was marked at 100 nM,
with a maximum response at 1 ꢀM of MIPS-9922 consistent with PI3Kβ inhibitory potency
and matching that seen with high dose LY294002 (25ꢀM) or ZSTK-474 (0.25 ꢀM) (data not
shown) and consistent with that reported for TGX-221 in mouse platelets[6]. There is also
good correlation between inhibition of integrin αIIbβ3 and inhibition of platelet aggregation by
MIPS-9922, suggesting the effect of this inhibitor is relayed along the PI3Kβ signalling
pathway [19].
The anti-platelet effect of MIPS-9922 was further evaluated by flow-based adhesion studies
using anticoagulated whole blood. Under high shear (1800 s-1) MIPS-9922 inhibited platelet
adhesion to a vWf matrix in a concentration-dependent manner. Maximal reduction was
achieved with 1 µM (P < 0.01) and 5 µM MIPS-9922 (P < 0.001), reducing platelet adhesion
up to 60% (Fig. 5c). It was also found that MIPS-9922 did not affect platelet tethering or
translocation, but changed platelet status from stationary to rolling (data not shown),
suggesting that the inhibition was limited to integrin αIIbβ3 [20]. This inhibitory effect on
stationary adhesion was identical to that for TGX-221 (2 ꢀM) under the same conditions.
14

ACCEPTED MANUSCRIPT
A
B
C
Figure 5. Concentration-dependent inhibition of MIPS-9922 on integrin αIIbβ3 activation.
OG-FNG binding was measured by Flow Cytometry. DMSO control (0.25%) was set as
100%. (A) Washed platelets were stimulated by (a) ADP (5 µM) (n = 3, mean ± SEM) and (b)
TRAP (between 0.5 µM and 1 µM depending on donor responses) (n = 4-7, mean ± SEM).
(C) Stationary platelet adhesion following human whole blood perfusion at 1800s-1 for 4 min.
Statistical analysis was performed using a t test comparing MIPS-9922 at 0.001ꢀM (Fig 5A
and B) or 0.1ꢀM (Fig 5C) versus MIPS-9922 at selected concentration. ***P < 0.005
2.3.2 Studies of MIPS-9922 in vivo efficacy and bleeding time in mouse
15

ACCEPTED MANUSCRIPT
Studies of arterial thrombosis were performed using the electrolytic mouse model[6]. MIPS-
9922 showed sufficient aqueous solubility to deliver an intravenous dose of 2.5 mg/kg in
saline. In vivo, MIPS-9922 effectively prevented the formation of arterial thrombi (Fig. 6a
and b). After electrolytic injury, the rate and volume of blood flow was stable and maintained
at pre-injury levels for the duration of the treatment indicating that no thrombotic occlusion
had occurred. The profile was again consistent with that reported for TGX-221 (2 - 3mg/kg;
noting the variations in model in terms of species and injury model).[3, 7, 8] Treatment with
MIPS-9922 at the same dose did not prolong bleeding time (Fig. 6c) or cause excess blood
loss (Fig. 6d), indicating normal haemostasis was preserved.
A
B
C
D
Figure 6. In vivo efficacy of MIPS-9922. (a) Effects of saline (n = 8) or MIPS-9922
(2.5mg/kg; n = 8) on carotid artery blood flow (ml/min per 100g body weight) after
electrolytic injury in mice (mean±S.E.M.). (b) Effects of saline or MIPS-9922 on carotid
16

ACCEPTED MANUSCRIPT
artery blood flow volume (area under curve over the 60-min period) after electrolytic injury
in mice (***P<0.005). (c) Tail bleeding time in mice treated with either vehicle (0.01%
DMSO; n = 8) or MIPS-9922 (2.5mg/kg; n = 8) (P=0.9390). (d) Blood loss from mice
undergone tail bleeding assay (measured by absorbance at 575nm of haemoglobin)
(P=0.4993).
2.3.3. Pharmacokinetics of MIPS-9922 in rats
MIPS-9922 showed moderate aqueous solubility and on intravenous dosing showed an
apparent half-life in rats of approximately 3 h, the volume of distribution was high and
clearance was moderate (Fig. 7). In vitro MIPS-9922 showed relatively high levels of protein
binding (~ 97%), good stability in plasma and moderate stability against liver microsomes
(Table S1, Supplementary Information).
In consideration of potential oral administration, the apparent oral bioavailability at 20 mg/kg
ranged from 25-28% and maximum plasma concentrations were observed 1-4 h post-dose.
The data compare closely to that seen for TGX-221 after intravenous infusion (serum
concentration of 2 and 0.2 ꢀM at 5 and 60 min respectively)[3]. No adverse reactions or
compound-related side effects were observed following single-dose administration of MIPS-
9922 via the intravenous or oral routes. The PK parameters indicate that both with oral and
intravenous dosing, plasma concentrations (> 0.1ꢀM) could be sustained at a level to achieve
a therapeutic effect.
17

ACCEPTED MANUSCRIPT
Figure 7. Plasma concentration versus time profiles for MIPS-9922 in Sprague Dawley rats
(average of n=2) following intravenous administration at 3 mg/kg (closed symbols) and oral
administration at 20 mg/kg (open symbols).
3. Discussion
We have previously reported a small series of novel PI3Kβ inhibitors that are potent and
show good isoform selectivity.[13] These compounds have a distinct chemical structure and
achieve PI3Kβ selectivity through a different mechanism as compared to reported PI3Kβ
inhibitors. Like other PI3Kβ inhibitors, the compounds exert cellular activity through
inhibition of Akt phosphorylation downstream of PI3Kβ. When tested in a panel of cancer
cell lines, the compounds showed efficacious cytotoxicity across the board with superior cell
killing profile against p110β dependent PTEN-deficient cell lines such as MDA-MB-468
cells.[13] These data were consistent with those for other PI3Kβ inhibitors.[21, 22]
The PI3Kβ selectivity in particular appears to derive from the presence of a terminal amino
function, which hinders interactions with the PI3Kα and PI3Kγ isoforms in particular, and
offers altered selectivity against PI3Kδ. With a view to extending these observations, we
prepared a range of analogues bearing modifications with respect to the aminoacyl group,
18

ACCEPTED MANUSCRIPT
including a variety of unusual amino acids as well as modification to the piperazinyl linkage
point, swapping it for aminopiperazines and piperidines. While none of these changes offered
major improvements over the initial series in terms of PI3Kβ potency, we did observe a
marked change in the pattern of selectivity and the sensitivity of ligands to mutation, showing
that the different linkers result in changes to the engagement with the enzyme binding site.
PI3Kβ has also been identified as a potential target in thrombosis,[3] where the possibility of
separating anti-thrombotic activity from bleeding potential is attractive. The archetype of
PI3Kβ inhibition in the context of thrombosis has been the interruption of integrin αIIbβ3
signalling, to prevent stable adhesion of platelets under shear stress, the effective inhibition of
platelet aggregation induced by weak platelet agonists, and in vivo the effective reduction in
thrombus formation without impacting on normal haemostasis (as measured by bleeding
time).[3]
When a selection of compounds were tested for their ability to inhibit agonist induced platelet
aggregation, all of them were found to have anti-platelet activity. The most effective of these
had IC50’s of ~ 50nM against ADP- and TRAP-induced platelet aggregation.
MIPS-9922 shows very similar in vitro and in vivo activity to the benchmark inhibitor TGX-
221 in the series of assays described here. These data further strengthen the validation of
PI3Kβ as a target in thrombosis, which to date have only been shown for the TGX-221-like
compounds. MIPS-9922 is generated from a different chemical scaffold and has a different
selectivity mechanism.[13, 14, 23] AZD6482 showed effective anti-thrombotic activity in
human subjects, but had a short in vivo half-life. Workers at AstraZeneca also recently
described an inhibitor (designated S-21) [24] that also showed a promising profile: selective
inhibition of PI3Kβ (except against PI3Kδ); effective inhibition of ADP-induced aggregation
in vitro and ex vivo in dogs; a minimal effect of bleeding time and reduced evidence of
19

ACCEPTED MANUSCRIPT
insulin-resistance related effects. However no data pertaining to αIIbβ3 activation or shear
dependence was reported.
With regard to the development of MIPS-9922 as a potential therapeutic, the aqueous
solubility, albumin binding, plasma stability, microsomal stability (in human and rat liver
microsomes), in vivo half-life, volume of distribution, clearance are all consistent with the
observed anti-thrombotic activity in the bolus intravenous dose reported above. While not
directly comparable to studies of other PI3Kβ inhibitors[24-27], the ADME properties of
MIPS-9922 do not highlight any fundamental liabilities that would limit further development
of this compound, although for effective oral antiplatelet therapy more extended exposure
would be desirable.
In conclusion, MIPS-9922 represents a potent and selective inhibitor of PI3Kβ that shows
robust anti-platelet activity in a range of pharmacological assays of platelet function and
experimental thrombosis.
4. Experimental
4.1. General
N-Fmoc-protected and N-Boc-protected amino acids, piperidine, TFA and O-(1H-6-
chlorobenzotriazol-1-y1)-N,N,N’,N’-tetramethyluronium hexafluorophosphate (HCTU) were
purchased from Auspep and ChemImpex. DMF and dichloromethane were purchased from
Merck. N-Boc-1,2,3,6-tetrahydropyridine-4-boronic acid pinacol ester was purchased from
Combi-Blocks. Dioxane, DIPEA, triethylamine and acetic anhydride were purchased from
Sigma–Aldrich. PdCl2(dppf) was purchased from Boron Molecular. All chemicals were used
1
without further purification. H NMR spectra were routinely recorded at 400 MHz using a
1
400MHz Bruker Ultrashield–Advance III NMR spectrometer. The H NMR spectra refer to
solutions in deuterated solvents as indicated. The residual solvent peaks have been used as an
internal reference, with each resonance assigned according to the following convention:
chemical shift (δ) measured in parts per million (ppm) relative to the residual solvent peak.
LC/MS was performed on an Agilent 6120 Single Quad LC/MS coupled with an Agilent
1260 Infinity HPLC fitted with a Luna C8(2) 5ꢀL 50 X 4.6mm 100Å column and samples
20

ACCEPTED MANUSCRIPT
were run in 0.1% formic acid in acetonitrile at a gradient of 5-100% over 10 minutes. High
resolution mass spectometry (HRMS) analyses were collected on an Agilent 6224 TOF
LC/MS Mass Spectrometer coupled to an Agilent 1290 Infinity. Analytical thin layer
chromatography (TLC) was performed on Merck aluminum sheets coated in silica gel 60
F254 and visualization accomplished with a UV lamp. Column chromatography was carried
out using silica gel 60 (Merck). The purity of compounds (≥95%) was established by reverse
phase HPLC. Hydrogenations were performed on a ThalesNano H-Cube continuous-flow
hydrogenation reactor equipped with a 10% Pd/C cartridge. The reaction mixture was passed
through the cartridge at a rate of 1 mL/min at 0 bar and 50°C on the Full hydrogen mode.
General Method A:
A mixture of 4-(4-chloro-6-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-1,3,5-triazin-2-
yl)morpholine (1 equiv.), the appropriate amine (2 equiv.), and K2CO3 (2 equiv.) in DMF (2
mL) in a sealed tube was exposed to microwave irradiation (110°C) for 20 min. The reaction
mixture was filtered and the DMF removed in vacuo. The resulting residue was taken up in
EtOAc (30 mL) and washed with H2O (3 x 15 mL), brine (15 ml), dried (Na2SO4), filtered
and solvent removed in vacuo to give the protected crude product.
General Method B: Protected Amino Acid coupling
To a solution of the N-protected amino acid (2 equiv.) in DMF (1 mL) was added HCTU (2
equiv.) followed by DIPEA (4 equiv.). A suspension of the intermediate amine (1 equiv.) in
DMF (1 mL) was added and the solution stirred at ambient temperature for 1.5 - 2 h. A 5%
aqueous NaHCO3 solution (20-30 mL) was added and the resulting suspension stirred
vigorously for 15 - 30 min. The precipitate was collected, washed with H2O and dried in
vacuo to give the crude product. (NB: in some cases the reaction mixture was dried and
directly chromatographed or crystallized but this gave inferior yields)
General Method C: Boc deprotection
The crude product was dissolved in TFA:DCM (1:1) and the solution stirred at ambient
temperature for 30 min. The volatile components were removed under a stream of N2. The
residue was taken up in H2O, basified with an aqueous solution of sat. NaHCO3 and extracted
with DCM (3 x 20 mL). The combined organic extract was dried (Na2SO4), filtered and
solvent evaporated in vacuo to give the crude product.
General Method D: Fmoc deprotection
The crude product was dissolved in 10% piperidine in DMF (1 mL) and stirred at ambient
temperature for 1 h. The reaction mixture was quenched with 5% aqueous NaHCO3 solution
21

ACCEPTED MANUSCRIPT
(20 mL) and extracted with EtOAc (3 x 20 mL). The combined organic layer was washed
with H2O, brine, dried (MgSO4), filtered and solvent removed in vacuo to give the crude
product.
4.2. Chemical Syntheses of Selected Individual Compounds
Compounds 3-18 were prepared by coupling 4-(4-(2-(difluoromethyl)-1H-benzo[d]imidazol-
1-yl)-6-(piperazin-1-yl)-1,3,5-triazin-2-yl)morpholine (2) with Boc-protected amino acids
according to method B followed by deprotection according to Method C, or using Fmoc-
protected amino acids followed by deprotection according to Method D as described
previously. Compounds 3, 4, 6, 7, 10, 11, 15 and 18 have been previously reported [13].
Details of compounds 5, 8, 9, 12, 13, 14, 16 and 17 are provided in the Supplementary
Information.
4.2.1 (S)-2-amino-1-(4-(4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-
1,3,5-triazin-2-yl)piperazin-1-yl)-3-phenylpropan-1-one (MIPS9922, 10)
4-(4-(2-(difluoromethyl)-1Hbenzo[d]imidazol-1-yl)-6-(piperazin-1-yl)-1,3,5-triazin-2-
yl)morpholine (2) (0.075 g, 0.18 mmol) was treated with Boc-L-Phe-OH (0.095 g, 0.36 mmol)
according to Method B to give a white solid (0.121 g, quant.), LCMS (ESI): (m/z) = 664.2
[M+H]+; HPLC (MeCN): 9.79 min. The Boc-protected product was deprotected according to
method C. The crude product was chromatographed through silica, eluting with 96:4
CHCl3/MeOH, and the fractions concentrated then lyophilized to give 10
as a white powder (0.074 g, 75%); 1H NMR (401 MHz, CDCl3) δ 8.28 (s, 1H), 7.91 – 7.86 (m,
1H), 7.50 (t, J = 53.7 Hz, 1H), 7.46 – 7.37 (m, 2H), 7.34 – 7.28 (m, 2H), 7.25 – 7.20 (m, 3H),
4.03 (br. s, 1H), 3.86 (s, 5H), 3.78 (br. s, 6H), 3.66 – 3.55 (m, 2H), 3.46 – 3.36 (m, 1H), 3.22
– 3.13 (m, 1H), 3.10 (br. s, 1H), 3.02 – 2.88 (m, 2H). 13C NMR (101 MHz, CDCl3) δ 165.1,
165.0, 162.1, 146.3, 146.0, 142.1, 137.5, 133.7, 129.5, 128.9, 127.2, 125.9, 124.6, 121.5,
115.9, 108.6 (t, J = 239.9 Hz), 66.7, 52.7, 45.0, 44.9, 44.2, 43.4, 43.1, 41.8, 41.7. LCMS
(ESI): (m/z) = 564.2 [M+H]+; HR-MS calculated for C28H31N9O2F2 [M+H]+: 564.2642;
found 564.2647; HPLC (MeCN): 6.94 min.
4.2.2 1-(4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-triazin-2-
yl)piperidin-4-amine (19)
22

ACCEPTED MANUSCRIPT
The title compound was prepared via General Method A from 4-(4-chloro-6-(2-
(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-1,3,5-triazin-2-yl)morpholine (1) (150 mg, 0.41
mmol), 4-(N-Boc-amino)piperidine (164 mg, 0.82 mmol), and K2CO3 (113 mg, 0.82 mmol)
in DMF (2 mL) to give an off-white solid (168 mg). Boc deprotection was achieved
according to Method C. Purification by column chromatography (SiO2,
EtOAc:MeCN:MeOH:H2O, 60:15:15:15) yielded the product as a white solid (157 mg, 89%).
HRMS (ESI+): Found: m/z 431.2123 (M + H)+, C20H25F2N8O requires m/z 431.2114. 1H NMR
(400 MHz, CDCl3) δ 8.35 (d, J = 7.6 Hz, 1H), 7.89 (d, J = 7.3 Hz, 1H), 7.59 (t, J = 53.7 Hz,
1H), 7.45 – 7.35 (m, 2H), 4.75 – 4.60 (m, 2H), 3.94 – 3.84 (m, 4H), 3.84 – 3.73 (m, 4H), 3.19
– 2.97 (m, 3H), 2.02 – 1.88 (m, 2H), 1.43 – 1.28 (m, 2H). LC-MS (ESI+) m/z: 431.2 (M + H)+
(100%).
4.2.3 2-amino-N-(1-(4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-
triazin-2-yl)piperidin-4-yl)acetamide (20)
The title compound was prepared via Method B from 1-(4-(2-(difluoromethyl)-1H-
benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-triazin-2-yl)piperidin-4-amine (18) (50 mg, 0.12
mmol) and Boc-Gly-OH (41 mg, 0.23 mmol). Boc deprotection was achieved according to
Method C. Purification by column chromatography (SiO2, EtOAc:MeCN:MeOH:H2O,
60:15:15:15) yielded the product as a white solid (45 mg, 79%).
1
HRMS (ESI+): Found: m/z 488.2340 (M + H)+, C22H28F2N9O2 requires m/z 488.2329. H
NMR (400 MHz, CDCl3) δ 8.34 (d, J = 7.3 Hz, 1H), 7.89 (d, J = 7.3 Hz, 1H), 7.58 (t, J =
53.7 Hz, 1H), 7.47 – 7.37 (m, 2H), 7.34 (d, J = 8.3 Hz, 1H), 4.76 – 4.63 (m, 2H), 4.19 – 4.07
(m, 1H), 3.96 – 3.83 (m, 4H), 3.83 – 3.75 (m, 4H), 3.36 (s, 2H), 3.27 – 3.07 (m, 2H), 2.07 (d,
J = 10.6 Hz, 2H), 1.54 – 1.43 (m, 2H). LC-MS (ESI+) m/z: 488.2 (M + H)+ (100%).
4.2.4 N-(1-(4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-triazin-2-
yl)piperidin-4-yl)acetamide (21)
1-(4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-triazin-2-
yl)piperidin-4-amine (18) (40 mg, 0.09 mmol) was dissolved in CH2Cl2 (1 mL) and treated
with acetic anhydride (11 mg, 0.11mmol) and triethylamine (19 mg, 0.18 mmol). The mixture
was stirred at ambient temperature for 4 h. The reaction was quenched with sat. NaHCO3 (5
mL) and extracted with CH2Cl2 (3 x 5 mL). The combined organic extract was dried
(Na2SO4), filtered and solvent evaporated in vacuo to give a white solid (42 mg). Purification
23

ACCEPTED MANUSCRIPT
by column chromatography (SiO2, CH2Cl2:MeOH, 98:2) yielded the product as a white solid
(36 mg, 82%).
1
HRMS (ESI+): Found: m/z 473.2228 (M + H)+, C22H27F2N8O2 requires m/z 473.2220. H
NMR (400 MHz, CDCl3) δ 8.33 (d, J = 7.5 Hz, 1H), 7.89 (d, J = 7.2 Hz, 1H), 7.57 (t, J =
53.6 Hz, 1H), 7.47 – 7.37 (m, 2H), 5.36 (d, J = 7.6 Hz, 1H), 4.73 (bs, 2H), 4.18 – 4.03 (m,
1H), 3.96 – 3.74 (m, 8H), 3.22 – 3.00 (m, 2H), 2.09 (dd, J = 12.4, 2.5 Hz, 2H), 2.00 (s, 3H),
1.40 (bs, 2H). LC-MS (ESI+) m/z: 473.2 (M + H)+ (100%).
4.2.5 (1-(4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-triazin-2-
yl)piperidin-4-yl)methanamine (22)
The title compound was prepared via Method A from 4-(4-chloro-6-(2-(difluoromethyl)-1H-
benzo[d]imidazol-1-yl)-1,3,5-triazin-2-yl)morpholine (150 mg, 0.41 mmol), 4-(Boc-
aminomethyl)piperidine (175 mg, 0.82 mmol), and K2CO3 (113 mg, 0.82 mmol) in DMF (2
mL) to give an white solid (154 mg). Purification by column chromatography (SiO2,
CH2Cl2:MeOH, 9:1 to 4:1) yielded the product as a white solid (144 mg, 79%).
1
HRMS (ESI+): Found: m/z 445.2281 (M + H)+, C21H27F2N8O requires m/z 445.2270. H
NMR (400 MHz, CDCl3) δ 8.35 (d, J = 7.6 Hz, 1H), 7.89 (d, J = 7.6 Hz, 1H), 7.60 (t, J =
53.6 Hz, 1H), 7.46 – 7.35 (m, 2H), 4.81 (t, J = 14.1 Hz, 2H), 3.83 (dt, J = 9.1, 4.3 Hz, 8H),
3.06 – 2.85 (m, 2H), 2.66 (s, 2H), 1.89 (t, J = 12.2 Hz, 2H), 1.73 – 1.63 (m, 1H), 1.22 (ddd, J
= 16.4, 12.4, 4.8 Hz, 2H). LC-MS (ESI+) m/z: 445.2 (M + H)+ (100%).
4.2.6 2-amino-N-((1-(4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-
triazin-2-yl)piperidin-4-yl)methyl)acetamide (23)
The title compound was prepared via Method B from (1-(4-(2-(difluoromethyl)-1H-
benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-triazin-2-yl)piperidin-4-yl)methanamine (23) (50
mg, 0.11 mmol) and Boc-Gly-OH (39 mg, 0.22 mmol). Boc deprotection was achieved
according to Method D. Purification by column chromatography (SiO2, CH2Cl2:MeOH, 9:1)
yielded the product as a white solid (44 mg, 77%).
1
HRMS (ESI+): Found: m/z 502.2498 (M + H)+, C23H30F2N9O2 requires m/z 502.2485. H
NMR (400 MHz, CDCl3) δ 8.34 (d, J = 7.6 Hz, 1H), 7.89 (d, J = 7.4 Hz, 1H), 7.59 (t, J =
53.7 Hz, 1H), 7.53 – 7.47 (m, 1H), 7.44 – 7.36 (m, 2H), 4.79 (t, J = 13.0 Hz, 2H), 3.94 – 3.74
(m, 8H), 3.39 (s, 2H), 3.32 – 3.17 (m, 2H), 3.07 – 2.84 (m, 2H), 1.96 – 1.78 (m, 3H), 1.35 –
1.17 (m, 2H). LC-MS (ESI+) m/z: 502.2 (M + H)+ (100%).
24

ACCEPTED MANUSCRIPT
4.2.7 2-(1-(4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-triazin-2-
yl)piperidin-4-yl)ethan-1-amine (24)
The title compound was prepared via General Method A from 4-(4-chloro-6-(2-
(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-1,3,5-triazin-2-yl)morpholine (1) (100 mg, 0.27
mmol), 4-(2-Boc-aminoethyl)piperidine (125 mg, 0.55 mmol), and K2CO3 (75 mg, 0.55
mmol) in DMF (2 mL) to give a white solid (105 mg). The residue was deprotected following
Method C. Purification by column chromatography (SiO2, CH2Cl2:MeOH 4:1) yielded the
product as a white solid (90 mg, 72%).
1
HRMS (ESI+): Found: m/z 459.2429 (M + H)+, C22H29F2N8O requires m/z 459.2427. H
NMR (400 MHz, MeOD) δ 8.41 (d, J = 7.6 Hz, 1H), 7.78 (d, J = 7.3 Hz, 1H), 7.69 (t, J =
53.2 Hz, 1H), 7.50 – 7.45 (m, 1H), 7.45 – 7.39 (m, 1H), 4.83 – 4.70 (m, 2H), 3.92 – 3.82 (m,
4H), 3.82 – 3.70 (m, 4H), 3.09 – 2.87 (m, 2H), 2.73 (t, J = 7.4 Hz, 2H), 1.87 – 1.78 (m, 2H),
1.77 – 1.66 (m, 1H), 1.47 (dd, J = 14.8, 7.0 Hz, 2H), 1.26 – 1.12 (m, 2H). LC-MS (ESI+) m/z:
13
459.2 (M + H)+ (100%). C NMR (101 MHz, CDCl3) δ 165.30, 164.59, 162.15, 146.40,
146.13, 145.87, 142.10, 133.77, 125.78, 124.38, 121.38, 116.01, 110.96, 108.58, 106.20,
66.81, 44.29, 44.08, 40.34, 39.47, 33.90, 32.17.
4.2.8 1-(4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-triazin-2-
yl)piperidin-3-amine (25)
The title compound was prepared via Method A from 4-(4-chloro-6-(2-(difluoromethyl)-1H-
benzo[d]imidazol-1-yl)-1,3,5-triazin-2-yl)morpholine (150 mg, 0.41 mmol), 3-(Boc-amino)
piperidine (164 mg, 0.82 mmol), and K2CO3 (113 mg, 0.82 mmol) in DMF (2 mL) to give an
white solid (110 mg). Boc deprotection was achieved according to Method C.
Purification by column chromatography (SiO2, CH2Cl2:MeOH, 95:5) yielded the product as a
white solid (97 mg, 55%).
1
HRMS (ESI+): Found: m/z 431.2120 (M + H)+, C20H25F2N8O requires m/z 431.2114. H
NMR (400 MHz, CDCl3) δ 8.35 (d, J = 7.7 Hz, 1H), 7.89 (d, J = 7.5 Hz, 1H), 7.59 (t, J =
53.6 Hz, 1H), 7.45 – 7.36 (m, 2H), 4.63 – 4.30 (m, 2H), 3.98 – 3.74 (m, 8H), 3.30 – 3.13 (m,
1H), 3.13 – 2.84 (m, 2H), 2.10 – 1.98 (m, 1H), 1.93 – 1.80 (m, 1H), 1.49 – 1.35 (m, 2H). LC-
MS (ESI+) m/z: 431.2 (M + H)+ (100%).
4.2.9 2-amino-N-(1-(4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-
triazin-2-yl)piperidin-3-yl)acetamide (26)
25

ACCEPTED MANUSCRIPT
The title compound was prepared via Method B from 1-(4-(2-(difluoromethyl)-1H-
benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-triazin-2-yl)piperidin-3-amine (21) (50 mg, 0.12
mmol) and Boc-Gly-OH (41 mg, 0.23 mmol). Boc deprotection was achieved according to
Method D. Purification by column chromatography (SiO2, CH2Cl2:MeOH, 9:1) yielded the
product as a white solid (45 mg, 79%).
HRMS (ESI+): Found: m/z 488.2339 (M + H)+, C22H28F2N9O2 requires m/z 488.2329. 1H
NMR (401 MHz, MeOD) δ 8.50 – 8.38 (m, 1H), 7.78 (d, J = 7.6 Hz, 1H), 7.71 (td, J = 52.9,
6.0 Hz, 2H), 7.53 – 7.46 (m, 1H), 7.42 (t, J = 7.2 Hz, 1H), 4.48 – 4.35 (m, 1H), 4.31 – 4.13
(m, 1H), 4.05 – 3.72 (m, 9H), 3.57 – 3.33 (m, 2H), 3.26 (d, J = 3.7 Hz, 2H), 2.07 – 1.97 (m,
1H), 1.94 – 1.80 (m, 1H), 1.76 – 1.60 (m, 2H). LC-MS (ESI+) m/z: 488.2 (M + H)+ (100%).
4.2.10 4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-N-(piperidin-4-yl)-
1,3,5-triazin-2-amine (27)
The title compound was prepared via Method A from 4-(4-chloro-6-(2-(difluoromethyl)-1H-
benzo[d]imidazol-1-yl)-1,3,5-triazin-2-yl)morpholine (150 mg, 0.41 mmol), 4-amino-1-Boc-
piperidine (164 mg, 0.82 mmol), and K2CO3 (113 mg, 0.82 mmol) in DMF (2 mL) to give an
white solid
(166
mg). Purification
by column
chromatography (SiO2,
EtOAc:MeCN:MeOH:H2O, 60:15:15:15 followed by CH2Cl2:MeOH:NH4OH, 9:0.9:0.1)
yielded the product as a white solid (149 mg, 85%).
1
HRMS (ESI+): Found: m/z 431.2123 (M + H)+, C20H25F2N8O requires m/z 431.2114. H
NMR (400 MHz, CDCl3) δ 8.41 (dd, J = 7.2, 4.8 Hz, 1H), 7.89 (t, J = 7.2 Hz, 1H), 7.64 (td, J
= 53.7, 5.5 Hz, 1H), 7.47 – 7.36 (m, 2H), 5.24 (t, J = 8.4 Hz, 1H), 4.08 – 3.94 (m, 1H), 3.94 –
3.72 (m, 8H), 3.19 (d, J = 12.4 Hz, 2H), 2.86 – 2.72 (m, 2H), 2.10 (d, J = 12.1 Hz, 2H), 1.51
(qd, J = 12.0, 3.7 Hz, 2H). LC-MS (ESI+) m/z: 431.2 (M + H)+ (100%).
4.2.11 2-amino-1-(4-((4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-
triazin-2-yl)amino)piperidin-1-yl)ethan-1-one (28)
The title compound was prepared via Method B from 4-(2-(difluoromethyl)-1H-
benzo[d]imidazol-1-yl)-6-morpholino-N-(piperidin-4-yl)-1,3,5-triazin-2-amine (27) (50 mg,
0.12 mmol) and Boc-Gly-OH (41 mg, 0.23 mmol). Boc deprotection was achieved according
to Method C. Purification by column chromatography (SiO2, EtOAc:MeCN:MeOH:H2O,
60:15:15:15) yielded the product as a white solid (39 mg, 68%).
1
HRMS (ESI+): Found: m/z 488.2341 (M + H)+, C22H28F2N9O2 requires m/z 488.2329. H
NMR (400 MHz, CDCl3) δ 8.39 (d, J = 7.0 Hz, 1H), 7.90 (t, J = 7.6 Hz, 1H), 7.61 (td, J =
26

ACCEPTED MANUSCRIPT
53.5, 13.5 Hz, 1H), 7.45 – 7.35 (m, 2H), 5.22 (dd, J = 23.7, 7.6 Hz, 1H), 4.59 (t, J = 14.5 Hz,
1H), 4.14 (bs, 1H), 3.98 – 3.68 (m, 9H), 3.50 (s, 2H), 3.17 (t, J = 12.4 Hz, 1H), 2.92 (t, J =
12.1 Hz, 1H), 2.24 – 2.08 (m, 2H), 1.55 – 1.39 (m, 2H). LC-MS (ESI+) m/z: 488.2 (M + H)+
(100%).
4.2.12 tert-Butyl 4-(4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-
triazin-2-yl)-3,6-dihydropyridine-1(2H)-carboxylate (29)
A mixture of N-Boc-1,2,3,6-tetrahydropyridine-4-boronic acid pinacol ester (1.0 g, 2.7 mmol),
4-(4-chloro-6-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-1,3,5-triazin-2-yl)morpholine
(1.0 g, 3.3 mmol), PdCl2(dppf) (0.2 g, 0.27 mmol) were suspended in 2M aqueous Na2CO3 (4
mL) and dioxane (20 mL) and heated to reflux under a N2 atmosphere for 2 h. The dioxane
was removed under vacuum and the product extracted into CH2Cl2 and dried to give a brown
residue. Purification by column chromatography (SiO2, CH2Cl2:EtOAc, 95:5) yielded the
product 29 as a white solid (1.13 g, 81%).
1H NMR (400 MHz, CDCl3) δ 8.45 (d, J = 7.6 Hz, 1H), 7.92 (d, J = 7.6 Hz, 1H), 7.65 (t, J =
53.6 Hz, 1H), 7.51 – 7.46 (m, 1H), 7.46 – 7.40 (m, 1H), 7.38 (bs, 1H), 4.24 (d, J = 2.9 Hz,
2H), 4.08 – 3.92 (m, 4H), 3.88 – 3.79 (m, 4H), 3.65 (t, J = 5.6 Hz, 2H), 2.69 (bs, 2H), 1.51 (s,
9H). LC-MS (ESI+) m/z: 514.2 (M + H)+ (100%).
4.2.13 4-(4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-(piperidin-4-yl)-1,3,5-triazin-
2-yl)morpholine (30)
A solution of 29 (70 mg, 0.14 mmol) in MeOH (27 mL) (0.005M) was hydrogenated over 10%
Pd/C at 50°C for 1.5 h using the H-cube. The solvent was removed in vacuo and crystallised
from ether to give a white solid 61 mg, 87%.
1Η ΝΜΡ (400 ΜΗz, CDCL3) δ 8.45 (d, J = 8.0 Ηz, 1Η), 7.91 (d, J = 7.6 Ηz, 1Η), 7.68 (t, J =
53.7 Ηz, 1Η), 7.51 – 7.40 (m, 2Η), 4.24 (br s, 2Η), 4.05 – 3.89 (m, 4Η), 3.89 – 3.77 (m, 4Η),
2.97 – 2.77 (m, 3Η), 2.05 (d, J = 12.8 Ηz, 2Η), 1.80 (dd, J = 12.9, 3.9 Ηz, 2Η), 1.49 (s, 9Η).
LR-MS (ESI+) m/z: 516.2 (Μ + Η)+ (60%), 460.2 (100%).
4.2.14 2-amino-1-(4-(4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-
triazin-2-yl)piperidin-1-yl)ethan-1-one (31)
The title compound 31 was prepared via Method D then Method B from 30 (59 mg, 0.11
mmol) and Boc-Gly-OH (40 mg, 0.23 mmol). Boc deprotection of the crude intermediate was
27

ACCEPTED MANUSCRIPT
then achieved according to Method D. Purification by column chromatography (SiO2,
CHCl3:MeOH, 9:1) yielded the product as a white solid (34 mg, 63%).
1
HRMS (ESI+): Found: m/z 473.2230 (M + H)+, C22H27F2N8O2 requires m/z 473.2220. H
NMR (400 MHz, CDCl3) δ 8.42 (d, J = 7.8 Hz, 1H), 7.90 (d, J = 7.5 Hz, 1H), 7.65 (t, J =
53.6 Hz, 1H), 7.50 – 7.40 (m, 2H), 4.71 (d, J = 13.1 Hz, 1H), 4.04 – 3.90 (m, 4H), 3.90 –
3.76 (m, 5H), 3.61 (bs, 2H), 3.17 (t, J = 12.1 Hz, 1H), 2.93 (tt, J = 11.4, 3.7 Hz, 1H), 2.84 (t,
J = 11.8 Hz, 1H), 2.14 (d, J = 12.7 Hz, 2H), 1.98 – 1.74 (m, 2H). LC-MS (ESI+) m/z: 473.2
13
(M + H)+ (100%). C NMR (101 MHz, CDCl3) δ 181.38, 164.89, 162.35, 146.46, 146.20,
145.94, 142.10, 133.58, 126.34, 124.95, 121.59, 116.25, 111.06, 108.67, 106.28, 66.67,
66.56, 44.70, 44.23, 44.18, 42.17, 30.09, 29.60.
4.2.15 (S)-2-amino-1-(4-(4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-
1,3,5-triazin-2-yl)piperidin-1-yl)propan-1-one (32)
The title compound was prepared via Method E from tert-butyl 4-(4-(2-(difluoromethyl)-1H-
benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-triazin-2-yl)piperidine-1-carboxylate (25) (50
mg, 0.1 mmol) and Boc-L-Ala-OH (37 mg, 0.19 mmol). Boc deprotection of the crude
intermediate was then achieved according to Method D. Purification by column
chromatography (SiO2, CH2Cl2:MeOH, 9:1) yielded the product as a white solid (33 mg,
70%).
1
HRMS (ESI+): Found: m/z 487.2384 (M + H)+, C23H29F2N8O2 requires m/z 487.2376. H
NMR (401 MHz, MeOD) δ 8.53 (d, J = 7.9 Hz, 1H), 7.81 (d, J = 7.5 Hz, 1H), 7.78 (t, J =
53.1 Hz, 1H), 7.55 – 7.43 (m, 2H), 4.60 (d, J = 9.9 Hz, 1H), 4.09 (d, J = 13.4 Hz, 1H), 4.05 –
3.90 (m, 5H), 3.87 – 3.75 (m, 4H), 3.38 – 3.25 (m, 1H), 3.06 (tt, J = 11.4, 3.8 Hz, 1H), 2.97 –
2.85 (m, 1H), 2.29 – 2.11 (m, 2H), 1.99 – 1.71 (m, 2H), 1.27 (dd, J = 11.8, 6.8 Hz, 3H). LC-
MS (ESI+) m/z: 487.2 (M + H)+ (100%).
4.2.16 (R)-2-amino-1-(4-(4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-
1,3,5-triazin-2-yl)piperidin-1-yl)propan-1-one (33)
The title compound was prepared via Method E from tert-butyl 4-(4-(2-(difluoromethyl)-1H-
benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-triazin-2-yl)piperidine-1-carboxylate (25) (50
mg, 0.1 mmol) and Boc-D-Ala-OH (37 mg, 0.19 mmol). Boc deprotection of the crude
intermediate was then achieved according to Method D. Purification by column
chromatography (SiO2, CH2Cl2:MeOH, 9:1) yielded the product as a white solid (36 mg,
77%).
28

ACCEPTED MANUSCRIPT
1
HRMS (ESI+): Found: m/z 487.2390 (M + H)+, C23H29F2N8O2 requires m/z 487.2376. H
NMR (401 MHz, MeOD) δ 8.53 (d, J = 7.9 Hz, 1H), 7.81 (d, J = 7.5 Hz, 1H), 7.78 (t, J =
53.1 Hz, 1H), 7.54 – 7.43 (m, 2H), 4.60 (d, J = 9.8 Hz, 1H), 4.09 (d, J = 13.2 Hz, 1H), 4.05 –
3.90 (m, 5H), 3.87 – 3.73 (m, 4H), 3.38 – 3.24 (m, 1H), 3.06 (tt, J = 11.3, 3.8 Hz, 1H), 2.98 –
2.84 (m, 1H), 2.29 – 2.12 (m, 2H), 1.99 – 1.71 (m, 2H), 1.27 (dd, J = 11.9, 6.8 Hz, 3H). LC-
MS (ESI+) m/z: 487.2 (M + H)+ (100%).
4.2.17 (R)-4-(4-(2-(difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-(1-prolylpiperidin-4-yl)-
1,3,5-triazin-2-yl)morpholine (34)
The title compound was prepared via Method E from tert-butyl 4-(4-(2-(difluoromethyl)-1H-
benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-triazin-2-yl)piperidine-1-carboxylate (25) (50
mg, 0.1 mmol) and Fmoc-D-Pro-OH (65 mg, 0.19 mmol). Fmoc deprotection of the crude
intermediate was then achieved according to Method D. Purification by column
chromatography (SiO2, CH2Cl2:MeOH, 9:1) yielded the product as a white solid (19 mg,
38%).
1
HRMS (ESI+): Found: m/z 513.2542 (M + H)+, C25H31F2N8O2 requires m/z 513.2533. H
NMR (400 MHz, MeOD) δ 8.54 (d, J = 8.4 Hz, 1H), 7.82 (d, J = 7.9 Hz, 1H), 7.78 (t, J =
53.1 Hz, 1H), 7.55 – 7.43 (m, 2H), 4.58 (d, J = 13.2 Hz, 1H), 4.14 – 4.05 (m, 2H), 4.05 –
3.91 (m, 4H), 3.87 – 3.75 (m, 4H), 3.38 – 3.26 (m, 1H), 3.25 – 3.16 (m, 1H), 3.13 – 3.02 (m,
1H), 3.00 – 2.82 (m, 2H), 2.33 – 2.13 (m, 3H), 1.97 – 1.67 (m, 5H). LC-MS (ESI+) m/z: 513.2
(M + H)+ (100%).
4.3. Biochemical Assays
Biochemical assays were performed as previously described and are described in brief
here.[13]
4.3.1 Generation of baculovirus containing p110 mutant DNA
The methods used here have been described previously [28, 29] with the pFastBacTM system
(Invitrogen, U.S.A.) used to generate recombinant baculovirus. In brief, mutant plasmids
were generated using the appropriate primer pair and Pfu DNA polymerase (Promega, U.S.A.)
with the template DNA being pFastBacTM wild-type p110 isoform as appropriate. Mutant
plasmids were then transformed into DH10Bac E.coli for transposition into the bacmid.
Recombinant bacmid DNA was then transfected, using lipofectin (Invitrogen, U.S.A.), into
Sf21 cells and high titre virus stock was produced through two cycles of infection.
29