Identification of a new family of pyrazolo[3,4-d]pyrimidine derivatives as multitarget Fyn-Blk-Lyn inhibitors active on B- and T-lymphoma cell lines

Article abstract of DOI:10.1016/j.ejmech.2019.07.048

Abnormal activation of B-cell receptor (BCR) signaling plays a key role in the development of lymphoid malignancies, and could be reverted by the simultaneous inhibition of Lyn, Fyn and Blk, three members of the Src family kinase (SFK). Fyn and Blk are al

Full text of DOI:10.1016/j.ejmech.2019.07.048

Accepted Manuscript
Identification of a new family of pyrazolo[3,4-d]pyrimidine derivatives as multitarget
Fyn-Blk-Lyn inhibitors active on B- and T-lymphoma cell lines
Anna Lucia Fallacara, Raffaele Passannanti, Mattia Mori, Giulia Iovenitti, Francesca
Musumeci, Chiara Greco, Emmanuele Crespan, Miroslava Kissova, Giovanni Maga,
Chiara Tarantelli, Filippo Spriano, Eugenio Gaudio, Francesco Bertoni, Maurizio
Botta, Silvia Schenone
PII:
S0223-5234(19)30669-5
DOI:
https://doi.org/10.1016/j.ejmech.2019.07.048
EJMECH 11545
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 8 May 2019
Revised Date: 11 July 2019
Accepted Date: 15 July 2019
Please cite this article as: A.L. Fallacara, R. Passannanti, M. Mori, G. Iovenitti, F. Musumeci, C.
Greco, E. Crespan, M. Kissova, G. Maga, C. Tarantelli, F. Spriano, E. Gaudio, F. Bertoni, M. Botta, S.
Schenone, Identification of a new family of pyrazolo[3,4-d]pyrimidine derivatives as multitarget Fyn-Blk-
Lyn inhibitors active on B- and T-lymphoma cell lines, European Journal of Medicinal Chemistry (2019),
doi: https://doi.org/10.1016/j.ejmech.2019.07.048.
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

ACCEPTED MANUSCRIPT
Identification of a new family of pyrazolo[3,4-d]pyrimidine derivatives as
multitarget Fyn-Blk-Lyn inhibitors active on B- and T- lymphoma cell
lines
Anna Lucia Fallacara a, Raffaele Passannanti a, Mattia Mori a, Giulia Iovenitti a, Francesca Musumeci b, Chiara Greco b,
Emmanuele Crespan c, Miroslava Kissova c, Giovanni Maga c, Chiara Tarantelli d, Filippo Spriano d, Eugenio Gaudio d,
Francesco Bertoni d, Maurizio Botta a, e, f, *, Silvia Schenone b
aDepartment of Biotechnology, Chemistry and Pharmacy, "Department of Excellence 2018-2022", University of Siena, Via Aldo Moro 2,
53100 Siena, Italy
bDepartment of Pharmacy, University of Genoa, Viale Benedetto XV, 3, 16132 Genoa, Italy
cIstituto di Genetica Molecolare, IGM-CNR, Via Abbiategrasso 207,I-27100 Pavia, Italy
dUniversità della Svizzera italiana (USI), Institute of Oncology Research (IOR), via Vela 6, 6500 Bellinzona, Switzerland.
eBiotechnology College of Science and Technology, Temple University, Biolife Science Building, Suite 333, 1900 N 12th Street,
Philadelphia, Pennsylvania 19122, United States
fLead Discovery Siena s.r.l., Via Vittorio Alfieri 31, 53019 Castelnuovo, Berardenga, Italy
ABSTRACT
Abnormal activation of B-cell receptor (BCR) signaling plays a key role in the development of
lymphoid malignancies, and could be reverted by the simultaneous inhibition of Lyn, Fyn and Blk,
three members of the Src family kinase (SFK).
Fyn and Blk are also promising targets for the treatment of some forms of T-cell non-Hodgkin
lymphoma which point to the druggability of SFKs for the treatment of these cancers. We recently
identified Si308 as a potent Fyn inhibitor, while preliminary data showed that it might also inhibit
Lyn and Blk. Here, molecular modelling studies were coupled with enzymatic assays to further
investigate the effect of Si308 on Lyn and Blk. A small library of pyrazolo[3,4-d]pyrimidines
structurally related to Si308 was synthesized and tested on human lymphoma cell lines.
Compound 2h emerged as a new multitarget inhibitor of Lyn, Fyn and Blk endowed with
remarkable antiproliferative effects on human B and T lymphoma cell lines. Its favorable ADME
properties make the compound suitable for further developments.
Keywords: Lymphoma, kinase inhibitor, pyrazolo[3,4-d]pyrimidine, Fyn, Lyn, Blk
Abbreviations1
1
BCR, B-cell receptor; SFK, Src family kinase; ADME, adsorption, distribution, metabolism, elimination;
WHO, World Health Organization; CTCL, cutaneous T-cell lymphoma; DLBCL, diffuse large B-cell
lymphoma; PTCL, peripheral T-cell lymphoma; HLs, Hodgkin lymphomas; NHLs, non-Hodgkin lymphomas;
FDA, Food and Drug Administration; PDB, Protein Data Bank; MD, molecular dynamics; RMSD, root-mean-
square deviation; GAFF, Generalized Amber Force Field; THF, tetrahydrofuran; TBAF, tetrabutylammonium
fluoride; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; SAR, structure-activity
relationship; HLM, human liver microsomes; PAMPA, parallel artificial membrane permeability assay; GI,
gastrointestinal; MR, membrane retention; PME, Particle Mesh Ewald; DMF, dimethylformamide; DMSO,
dimethyl sulphoxide; Pd(dppf)Cl2, [1,1′-bis(diphenylphosphino)ferrocene]dichloropalladium (II); DME, 1,2-
dimethoxyethane; MM-GBSA, molecular mechanics generalaized Born surface area;
1

ACCEPTED MANUSCRIPT
*Corresponding authors.
E-mail addresses: botta.maurizio@gmail.com (M. Botta)
1
INTRODUCTION
Lymphomas are neoplasms that affect the cells of the lymphoreticular or immune system [1]. The
World Health Organization (WHO) classifies these diseases with respect to the cells of origin in B-
cell, T-cell, and natural killer (NK-) cell lymphomas [2]. Each class includes different subtypes of
tumors which are defined on the basis of their peculiarities, such as cellular morphology, molecular
lesions and chromosomal alterations [1,3]. The WHO lymphoma classification constitutes the
guideline to the correct diagnosis and treatment of these neoplasms, and it is frequently updated
by the WHO according to the new genetic and molecular discoveries in the field [1,2]. Lymphomas
span from indolent to aggressive on the basis of their growth velocity. Follicular lymphoma and
cutaneous T-cell lymphoma (CTCL) are examples of indolent B- and T-cell lymphomas,
respectively, while diffuse large B-cell lymphoma (DLBCL) and peripheral T-cell lymphoma (PTCL)
are examples of aggressive lymphomas [2].
Lymphomas can be also classified into Hodgkin lymphomas (HLs) and non-Hodgkin lymphomas
(NHLs) through a histological analysis. Indeed, HLs are characterized by malignant B-lymphocytes
that might evolve into the pathognomonic binuclear Reed-Sternberg cells. In contrast, NHLs are a
group of neoplastic diseases characterized by the abnormal proliferation of lymphoid cells, either
B- or T-cells, in various sites, including lymph nodes, bone marrow, spleen, liver and the
gastrointestinal tract. NHLs are characterized by a great heterogeneity that is usually related to the
different stages of differentiation and maturation of these cells [4].
Overall, NHLs account for 90% of all malignant lymphomas, while HLs for the remaining 10%.
Evidence showed a dramatic increase in the diagnosis of NHL, which was surpassed only by lung
cancer in women and malignant melanoma in both men and women. Thus, much research effort is
being spent on understanding the causes of NHLs as well as in finding new therapeutic options [5].
Proven risk factors for NHLs include hereditary immunodeficiency disorders, states of strong
immunosuppression (HIV/AIDS, organ transplantation), infectious agents (Epstein-Barr Virus,
human T-cell lymphotrophic virus-1, hepatitis C virus, Helicobacter pylori) [6], autoimmune
disorders (rheumatoid arthritis, Sjögren's syndrome, systemic lupus erythematosus myositis,
Hashimoto's thyroiditis, celiac disease/dermatitis herpetiformis) [3].
2

ACCEPTED MANUSCRIPT
The deregulation of the signalling pathways downstream to the B- or T-cell receptors are frequently
observed in lymphomas derived from B- or T- cells, respectively [7,8]. Since different members of
the Src family kinases (SFKs) play a pivotal role in this signalling, the search for pharmacological
agents able to block this class of enzymes is very active. After activation from external factors,
BCR triggers multi-enzymatic cascade reactions which promote cell proliferation, growth and
survival (Figure 1). In B-cell lymphomas, the two SFKs Lyn and Blk have been recently shown to
be essential for the survival of cell lines derived from DLBCL of the germinal center B-cell (GCB)
type but not of the activated B cell-like (ABC) [9]. Simultaneous inhibition of the three SFK
Fig. 1. BCR signaling representation
.
members Lyn, Fyn and Blk with the multitarget SFK inhibitor masitinib leads to the complete block
of BCR oncogenic signals in GCB- and ABC-DLBCL cells [10]. This inhibition was found to be
effective on lymphoma cells and patient-derived xenograft models, which substantiates the
importance of the simultaneous blockage of these SFK members in the treatment of DLBCL, as
well as other forms of aggressive B-cell lymphomas [10].
Similar data have been reported by Scuoppo et al. who showed that the SFK inhibitor dasatinib,
thanks to the suppression of Fyn activity, possesses anti-tumor activity in both ABC- and GCB-
DLBCL cell lines [11]. Dasatinib resulted more active in ABC cell lines than in GCB-DLBC cells,
while two other SFK inhibitors (i.e. ponatinib and saracatinib) showed similar activity on both the
cell subtypes [12].
A central role for SFKs has also been shown in T-cells. Palomero et al. identified activating
mutations of Fyn in PTCL and demonstrated that the treatment of transformed Rat1A cells
expressing the Fyn Tyr531His mutant with dasatinib reduces cell growth, suggesting Fyn as a
3

ACCEPTED MANUSCRIPT
candidate target for PTCL treatment [13]. Furthermore, Krejsgaard et al. reported on the
involvement of Blk in CTCL [14], and have further corroborated these results by showing that
dasatinib inhibits CTCL proliferation in vitro and in vivo [15]. Moreover, the pharmacological
inhibition of SFKs also represents a way to overcome mechanisms of resistance to signaling
inhibitors currently in clinical use, such as the BTK inhibitor ibrutinib [10,11].
Our research group is extensively involved in the development of pyrazolo[3,4-d]pyrimidine
derivatives as SFK inhibitors [16–19]. Among them, compound Si308 (Figure 2A) has recently
shown a noticeable inhibitory effect against Fyn, as well as a promising inhibitory activity in a
preliminary assay on Lyn and Blk. Specifically, treatment with Si308 at a concentration of 10 µM
reduces the kinase activity of Lyn and Blk by 85% and 81%, respectively [17].
In the present work, the action of Si308 on Lyn and Blk is further investigated and elucidated by
means of molecular modelling studies and enzymatic evaluations. Since compound Si308
emerged as a candidate multitarget SFK inhibitor, a small library of pyrazolo[3,4-d]pyrimidines
structurally related to Si308 was designed and docked into the ATP binding site of Fyn, Lyn and
Blk. Finally, the library was synthesized and tested for anti-tumor activity in human cell lines
derived from DLBCL, PTCL and CTCL. An evaluation of in vitro ADME properties (PAMPA
permeation, water solubility, and metabolic stability) was also performed.
Overall, our efforts highlighted a number of new pyrazolo[3,4-d]pyrimidines endowed with good
anti-tumor activity against lymphoma cells. In particular, compound 2h proved to be a powerful
multitarget inhibitor of Fyn, Lyn and Blk with remarkable antiproliferative effects on human B- and
T- cell lymphoma cell lines.
2
RESULTS
2.1 Molecular modelling and enzymatic activity of the first generation compound Si308
In the present work, docking studies were carried out using Glide software [20] with the aim of
understanding the possible binding mode of Si308 to SFKs. The reliability of the program was
verified by docking compound 1a (Figure 2A), a structural precursor of Si308, in the ATP binding
site of the crystallographic structure of Fyn coded by PDB ID: 2DQT [21]. Results showed that the
binding mode of compound 1a within the catalytic domain of Fyn is highly comparable to that
previously published [17] (Figure 2B). It is worth noting that compound 1a, as well as most of the
pyrazolo[3,4-d]pyrimidine derivatives studied in this work, has a chiral centre. Nevertheless,
docking calculations were carried out with the R enantiomer, because in a previous
crystallographic study we obtained the complex of this enantiomer with the ATP-binding site of the
kinase c-Src, a SFK member [18]. However, the same study clearly showed that both enantiomers
have
comparable
efficacy
against
the
recombinant
protein.
4

ACCEPTED MANUSCRIPT
Fig. 2. Chemical structures of compounds 1a and Si308 are reported (A). Predicted binding mode of
compound Si308 and 1a in the ATP binding pocket of Fyn (B), Lyn (C), and Blk (D). Ligands are shown as
magenta and cyan sticks respectively. Ligand/protein polar contacts are highlighted by yellow dashed lines.
The amino acids in yellow sticks in (C) and (D) represent the residues of the ATP binding site of Lyn and Blk
that differ from Fyn.
The docking protocol was then used to dock Si308 in the ATP binding site of Fyn, Lyn and Blk that
share a very high percentage of identity within the catalytic domain (Figure 3). Notably, residues of
the ATP binding sites are highly conserved, which led us to hypothesize that the binding mode of
Si308 could be highly comparable in the three selected targets.
5

ACCEPTED MANUSCRIPT
Fig. 3. Sequence alignment of the catalytic domain of Fyn, Blk and Lyn (UniProtKB codes: P06241, P51451
and P07948, respectively). Bars below the sequence alignment correspond to the degree of amino acid
conservation among the sequences (full bar: residues identity; empty bar: completely different residues). The
red boxes show the residues of the ATP binding site.
The X-ray crystal structure of human Lyn kinase (PDB ID: 5XY1 [22]) was selected for docking
studies because of its highest resolution among available Lyn structures. In contrast, and to the
best of our knowledge, the 3D structure of the catalytic domain of Blk was not available in
structural databases at the time of this study. Thus, to carry out the structure-based computational
analysis, it was generated by homology modelling. To this aim, a BLAST search in the Protein Data
Bank (PDB) provided the human tyrosine kinase Hck as the sequence with the highest percentage
of identity (75%) with respect to the query, i.e. kinase domain of Blk. Among the structures of
human Hck available in the PDB, the crystallographic structure coded by PDB-ID 2HK5 was
selected as template for the construction of the homology model with the Prime software [23,24]
because it is in a DFG-in conformation. The preliminary homology model was first refined through
the serial loop sampling approach, while the default sampling algorithm was used to adjust the
conformation of side chains that do not derive from the template. The resulting model was finally
relaxed through energy minimization, and its quality was assessed by the analysis of
Ramachandran plots as well as by the ProSA-web tool [25,26].
As anticipated above, Si308 keeps the same binding mode in Fyn, Lyn and Blk (Figure 2) because
of the high percentage of sequence identity among these proteins. Overall, the aromatic amine of
Si308 establishes H-bonds with the backbone of the glutamate and the methionine of the hinge
region, while the C3 phenyl group fits into the ATP binding site hydrophobic region.
To evaluate the conformational stability of Si308 in complex with Fyn, Lyn and Blk and to gain
further insights into the ligand-protein interactions, molecular dynamics (MD) simulations were
carried out with Amber16 on docking complexes. In particular, binding poses described in Figure 2
were solvated in a rectilinear box of explicit water molecules and counter-ions up to charge
6

ACCEPTED MANUSCRIPT
neutralization, and relaxed through energy minimization. Following heating to 300 K and density
equilibration, unrestrained MD trajectories were produced for a total of 50 ns on each complex.
Figure 4A describes the evolution of the root-mean-square deviation (RMSD) calculated on all
heavy atoms of the complexes during MD simulation time with respect to the last frame of the
minimization, while the RMSD of the ligand with respect to the starting poses in each complex is
shown in Figure 4B. In all systems, we observed a slight initial increase of the RMSD that is
followed by a stabilization at values around 2 Å after few ns, which suggests that the binding mode
of Si308 is stable for the entire MD simulation time.
Fig. 4. (A) RMSD of the heavy atoms of the complexes (B) RMSD of
the ligand in the three complexes.
7

ACCEPTED MANUSCRIPT
Visual inspection of MD trajectories showed that Si308 is stabilized by a number of interactions
within the ATP binding site of Fyn, Lyn and Blk that were further highlighted by using the pairwise
decomposition of the delta energy of binding as implemented in the MM-GBSA approach (Table 1).
Table 1
Hot spots in ligand-protein interaction identified by MD simulations.
Residue in
delta energy of binding [Kcalmol-1]
Fyn(Lyn,Blk)
Si308/Fyn
-2.903
-2.245
-3.375
-0.304
-0.583
-1.201
-2.379
-3.200
-3.359
-3.694
-3.475
-0.586
-0.587
-0.082
Si308/Lyn
-2.861
-1.810
-3.116
-0.054
-0.634
-1.276
-2.221
-3.304
-3.175
3.736
Si308/Blk
-3.485
-1.836
-3.869
-0.599
-0.533
-0.942
-2.254
-3.050
-2.887
-3.900
-3.582
-0.677
-1.196
-0.116
Val285(*261,**255)
Ala297(*273,**267)
Lys299(*275,**269)
Glu314(*290,**284)
Met318(*294,**288)
Val327(*303,**297)
Thr342(*319,**312)
Glu343(*320,**313)
Tyr344(*321,**314)
Met345(*322,**315)
Leu397(*374,**367)
Ala407(*384,**377)
Asp408(*385,**378)
Phe409(*386,**379)
-4.919
-2.020
-0.859
-0.088
* numbering of the same amino acid in Lyn ** numbering of the same amino acid in Blk
Energy values related to the interactions made by the compound in the binding site of Fyn, Lyn and
Blk are comparable each other. Particularly, the most energetically-relevant interactions of Si308
are established with residues of the hinge region Glu343 and Met345, with the catalytic Lys299
and with two hydrophobic residues of the binding site Val285 and Leu397. Residue numbering
corresponds to Fyn sequence; the above residues in Lyn are Glu320, Met322, Lys275, Val261 and
Leu374, while in Blk they are Glu313, Met315, Lys269, Val255 and Leu367.
The MM-GBSA approach was also used to compute the theoretical delta energy of binding of the
complexes. Ligand theoretical affinity values are reported in Table 2, and show highly comparable
theoretical affinity of Si308 in tested kinases in agreement with docking results and with the
conformational stability observed in MD simulations (Figure 4).
Table 2
Delta energy of binding of the three complexes.
Complex
delta energy of binding [Kcalmol-1]
8

ACCEPTED MANUSCRIPT
Fyn-Si308
Lyn-Si308
Blk-Si308
-37.55 +/- 2.59
-37.84 +/- 2.70
-37.28 +/- 2.48
Overall, the computational results suggest that Si308 might bind Fyn, Lyn and Blk with a
comparable affinity, thus behaving as a candidate multitarget inhibitor acting on the SKF members
that are relevant to B- and T-cell lymphomas.
The inhibitory activity of Si308 against target SKFs was finally confirmed by enzymatic assays
carried out on recombinant Fyn, Lyn and Blk. The compound shows inhibition values in the sub-
micromolar range that are highly comparable each other (Table 3), and in agreement with
molecular modelling results.
Table 3
Inhibitory activities of compound Si308 toward Fyn, Lyn and Blk.
Fyn
Lyn
Blk
Ki [µM]
Ki [µM]
Ki [µM]
Si308
0.16 ± 0.01
0.19 ± 0.04
0.19 ± 0.05
2.2 Chemistry
Encouraged by these enzymatic results, we decided to synthesize a series of twelve derivatives of
Si308, 2a-l, with the aim of discovering new compounds with anti-tumor activity in lymphoma cell
lines. The C3 phenyl ring, which accommodates within the inner side of the ATP binding pocket,
was differently functionalized with point modifications in -para, -ortho and -meta positions, with the
aim of verifying whether these decorations could lead to an increase in inhibitory activity, following
the trend observed for Si308 compared to its precursor 1a in Fyn inhibition. In addition, some
compounds bearing modifications at the N1 2-chloro-2-phenylethyl side chain were synthesized, in
order to explore the relevance of modifications in the solvent exposed region in this family of
compounds (Figure 5).
9

ACCEPTED MANUSCRIPT
Fig. 5. General plan for the design of the target compounds. C3 phenyl ring point modifications (upper panel)
and N1 side chain modifications combined with C3 phenyl ring modifications (lower panel) are shown.
The synthesis of 3-susbtituted 1-(2-chloro-2-phenylethyl)-1H-pyrazolo[3,4-d]pyrimidin-4-amines 2a-
f was achieved following a route which starts with a three-component one-pot synthesis [27].
Sodium hydride was added in small batches to a solution of malononitrile in anhydrous THF
precooled at 0-5 °C. After 30 min, the suitable acyl chloride was added and the solution was stirred
at room temperature until the complete transformation of the acyl chloride into the corresponding
enol derivative, affording compounds 3a-f (not isolated). Then dimethylsulfate was added, and the
reaction was refluxed for 3-6 h, monitoring via TLC the formation of the
[methoxy(phenyl)methylene]malononitrile derivatives 4a-f (not isolated). Finally, 2-hydrazino-1-
phenylethanol 5 was prepared following a literature procedure [28] and added to the reaction in a
THF solution. After 4 h of reflux, the crude was purified by flash chromatography, affording the
intermediates 6a-f. The latter were heated in the presence of formamide at 190 °C for 3-4 h to
obtain the pyrazolo-pyrimidines 7a-f. Finally, the intermediates 7a-f were reacted with thionyl
chloride in anhydrous CH2Cl2 at room temperature for 12 h under nitrogen atmosphere to give the
10

ACCEPTED MANUSCRIPT
desired compounds 2a-f (Scheme 1).
R
OCH3
CN
OH
NC
CN
CN
CN
ib
ia
ic
N
N
CN
CN
H2N
R
R
3a-f
4a-f
6a-f
OH
R
R
NHNH2
OH
NH2
N
NH2
a: R = 4Br
b: R = 4I
c: R = 4C2H5
d: R = 4nC3H7
e: R = 2Cl
iii
ii
N
N
N
N
N
N
N
f: R = 3Cl
5
Cl
OH
7a-f
2a-f
Scheme 1. Syntesis of compounds 2a-f. Reagents and conditions: (i) a) NaH, an. THF, 0-5 °C, 30 min; then
suitable acyl chloride, rt, 1.5 h; b) Me2SO4, reflux, 3-6 h; c) 5, reflux, 4 h; (ii) formamide, 190 °C, 3-4 h; (iii)
SOCl2, an. CH2Cl2, rt, 12 h, N2 atmosphere.
Synthesis of compounds 2g-l was performed via a Suzuki cross-coupling route. For compound 2h
this synthetic approach was chosen since the reactivity of the primary alcohol prevented the use of
the one-pot strategy. Compound 2g was directly obtained from an intermediate of this route. For
compounds 2i-l, this approach was selected in order to introduce the variations at N1 and C3 of
the pyrazolo[3,4-d]pyrimidine scaffold only in the last two reactions, reducing the number of steps.
In detail, for the synthesis of compounds 2g,h, 3-iodo-1H-pyrazolo[3,4-d]pyrimidin-4-amine 8,
already prepared by us [17], was dissolved in anhydrous toluene in the presence of a solution of
tetrabutylammonium fluoride (TBAF) 1 M in THF. After heating at 70 °C for 1 h, 2-
bromoacetophenone was added and the reaction was stirred at 70 °C other 2 h affording 2-(4-
amino-3-iodo-1H-pyrazolo[3,4-d]pyrimidin-1-yl)-1-phenylethanone 9, which was purified by flash
chromatography. The intermediate 9 was reduced by reaction with sodium borohydride in MeOH at
0 °C for 3 h giving the corresponding alcohol 10. Then, compound 10 was reacted with SOCl2 in
anhydrous CH2Cl2 at room temperature for 12 h under nitrogen atmosphere to obtain the
chlorinated derivative 11. Finally, the Suzuki reaction between the intermediates 10 or 11 and 4-
(hydroxymethyl)phenylboronic acid, in the presence of a palladium catalyst and caesium carbonate
afforded the final compounds 2g and 2h (Scheme 2).
11

ACCEPTED MANUSCRIPT
NH2
N
NH2
I
NH2
NH2
I
I
I
N
N
N
N
ii
iii
i
N
N
N
N
N
N
N
N
N
N
N
H
8
9
10
11
Cl
OH
O
iv
iv
OH
OH
NH2
NH2
N
N
N
N
N
N
N
N
OH
Cl
2h
2g
Scheme 2. Synthesis of compounds 2g,h. Reagents and conditions: (i) TBAF 1 M in THF, an. toluene, 70 °C, 1
h; then 2-bromoacetophenone, 70 °C, 2 h; (ii) NaBH4, MeOH, 0 °C, 3 h; (iii) SOCl2, an. CH2Cl2, r.t., 12 h, N2
atmosphere; (iv) 4-(hydroxymethyl)phenylboronic acid, Cs2CO3, Pd(dppf)Cl2, DME, H2O, 90 °C, 14 h.
For the synthesis of compounds 2i-l, 3-iodo-1H-pyrazolo[3,4-d]pyrimidin-4-amine 8 was treated
with 2-bromo-1-phenylpropane or (2-bromoethyl)benzene in the presence of potassium carbonate
at 130 °C for 18 h, affording the 3-iodo-1H-pyrazolo[3,4-d]pyrimidin-4-amino derivatives 12a,b [17].
Finally, compounds 12a,b were reacted with an excess of the suitable boronic acid under Suzuki
conditions to give the desired compounds 2i-l (Scheme 3).
Compounds 2a-j, bearing a chiral center, have been synthesized as racemic mixtures, since the
enantiomers of analogue compounds of previous synthesis showed comparable in vitro activity
[18]. Further developments of this work could include the resolution of the racemic mixture for the
most promising compounds in order to evaluate the biological profile of each enantiomer.
R
NH2
NH2
N
NH2
I
I
2i: R = F, R1 = CH3
2j: R = NO2, R1 = CH3
2k:R = Cl, R1 = H
ii
i
N
N
N
N
N
N
N
N
N
N
N
2l: R = CH3, R1 = H
H
R1
R1
2i-l
8
12a: R1 = CH3
12b: R1 = H
Scheme 3. Synthesis of compounds 2i-l. Reagents and conditions: (i) 2-bromo-1-phenylpropane or (2-
bromoethyl)benzene, K2CO3, an. DMF, 130 °C, 18 h; (ii) suitable boronic acid, Cs2CO3, Pd(dppf)Cl2, an.
toluene, 90 °C, 14 h.
12

ACCEPTED MANUSCRIPT
2.3 Biology
The small library of compounds has been tested in tumor cell lines derived from GCB-DLBCL
(n.=4; DOHH2, FARAGE, SU-DHL-8 and SU-DHL-6) and T-cell lymphoma lines (CTCL, n.=3,
HUT-78, H9 and HH; PTCL, n.=1, FE-PD). The C-3 para-tolyl derivative 1b, previously reported by
our research group and endowed with a promising activity towards Fyn (Ki = 95 nM) [17], has been
added to the test set. We assessed the anti-proliferative activities of the compounds using the 3-
(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) cell viability assay, and the
biological data allowed us to carry out a preliminary structure-activity relationship (SAR) evaluation
on this class of derivatives.
Compound 2h was the most potent and promising compound, with IC50 values lower than 10 µM in
all cell lines. The compound induced slight G2/M arrest in most of the cell lines, but it was able
also to induce cell death, as indicated by sub-G0 accumulation, which was more evident in the in
FE-PD cells and less in HUT-78 and SU-DHL-8 (Figure 6). In term of SAR, compound 2h had a
hydroxymethyl group as C-3 substituent. Its analogue 2g, bearing an OH instead of the chlorine
atom on the N1 side chain, was slightly less active, confirming the importance of a chlorine atom
on this chain.
Compound Si308 showed an IC50 value of 8.87 µM in the GCB-DLBCL cell line SU-DHL-8 and IC50
values between 11-22 µM in all the other cell lines, except for one (HUT-78), which was completely
resistant. The previous generation compound 1b was quite active on this cell line, although its
activity profile results generally similar or lower than that of Si308 in the other cell lines.
Regarding the N1 side chain, the replacement of the chlorine atom with a methyl group
(compounds 2i and 2j) or a hydrogen atom (compounds 2k and 2l) was generally related to an
activity similar or slightly lower than that observed for the parent compound Si308. The only
exception was compound 2l, which showed an IC50 value of 9.74 µM in the GCB-DLBCL cell line
DOHH2 cell line. It should be also noted that the introduction of a nitro group (compound 2j) was
responsible for the total loss of activity.
The displacement of the chlorine atom from the para position to ortho or meta positions on the C3
phenyl ring (compounds 2e and 2f, respectively) was also characterized by a reduction of the
activity. On the other hand, the introduction of increasingly bulky substituents on the para position
of the C3 phenyl ring (compare the 4-chloro derivative Si308 with the corresponding halogenated
derivatives 2a and 2b, and the 4-methyl derivative 1b with the corresponding ethyl and n-propyl
derivatives 2c and 2d) did not affect the biologic activity. Indeed, the bulkier derivatives generally
possess an activity similar or higher than that of the previously synthesized compounds Si308 and
1b, but the trend was not maintained among all cell lines. Compound 2b presented anti-
proliferative activity in two cell lines (GCB-DLBCL cell line DOHH2, IC50 = 3.68 µM; CTCL cell line
HH, IC50 = 5.24 µM). Compound 2d was active in four cell lines, two derived from GCB-DLBCL cell
13

ACCEPTED MANUSCRIPT
lines (DOHH2, IC50 = 4.38; SU-DHL-8, IC50 = 6.72 µM) and two from CTCL (HUT-78, IC50 = 9.90;
H9, IC50 = 4.74 µM).
Table 4
Anti-tumor activity of compounds Si308, 1b and 2a-l on lymphoma cell lines.
R
NH2
N
N
N
N
R1
Cell line (IC50) [µM]
B-cell lymphoma cell line
FARAGE SUDHL8 SUDHL6
T-cell lymphoma cell line
Cpd
Si308
1b
2a
2b
2c
2d
2e
2f
R
R1
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
DOHH2
11.41
11.77
10.99
3.68
HUT-78
> 100
11.12
21.05
12.03
58.00
9.90
FEPD
15.76
> 100
21.7
H9
13.58
> 100
21.7
HH
4Cl
14.94
19.25
13.61
12.64
22.74
29.36
51.45
30.9
8.87
12.14
11.85
11.46
17.47
6.72
21.52
23.43
45.64
13.17
29.68
31.11
> 100
> 100
34.20
9.29
13.9
4CH3
4Br
14.82
12.44
5.24
4I
20.46
~ 21.53
29.15
38.86
24.89
26.49
10.84
> 100
> 100
31.1
11.61
24.17
4.74
4C2H5
4nC3H7
2Cl
21.18
4.38
21.08
48.53
> 100
43.12
17.31
2.40
26.58
27.35
12.09
2.44
18.39
13.69
27.18
6.14
> 100
> 100
> 100
4.50
38.86
~22.07
~ 22.26
3.42
3Cl
2g
2h
2i
4CH2OH OH
4CH2OH Cl
26.96
7.02
4F
CH3
CH3
H
21.26
> 100
11.81
9.74
22.8
12.08
> 100
35.96
37.03
~ 22.92
> 100
> 100
> 100
> 100
> 100
> 100
> 100
18.76
> 100
> 100
> 100
11.81
> 100
18.17
17.26
2j
4NO2
4Cl
> 100
18.6
2k
2l
4CH3
H
15.98
~ 21.54
14

ACCEPTED MANUSCRIPT
Fig. 6. Cell cycle analyses of lymphoma cell lines following treatment with compound 2h at the
concentrations of 5 and 10 µM for 72 hours. Following treatment, cells were fixed with 70% ethanol before
staining with 7AAD for cell cycle analysis. The percentage of cells in each phase of the cell cycle is shown
for each treatment type.
2.4 Molecular modelling and enzymatic activity of the second generation compound 2h
The possible binding mode of compound 2h in the ATP binding site of Fyn, Lyn and Blk was
investigated by molecular docking using the validated protocol described above. Overall, the
compound shows a comparable binding mode in the three kinases. Similarly to Si308, compound
2h interacts through H-bonds with the conserved glutamate and methionine residues of the hinge
region, and also forms a cation-π interaction with the conserved catalytic lysine residue. In
addition, compound 2h is able to make a H-bond in all three kinases with the catalytic glutamate of
the alpha c-helix, thanks to the hydroxyl group in para position of the ring in C3. Moreover, the
same hydroxyl group is also able to make a hydrogen bond with the aspartate of the DFG motif
15

ACCEPTED MANUSCRIPT
(Figure 7).
Fig. 7. Chemical structure of compound 2h is reported (A). Predicted binding mode of compound 2h in the
ATP binding pocket of Fyn (B), Lyn (C), and Blk (D). Ligand 2h is shown as orange sticks. Ligand/protein
polar contacts are highlighted by yellow dashed lines. The amino acids in yellow sticks in (C) and (D)
represent the residues of the ATP binding site of Lyn and Blk that differ from Fyn.
To further confirm the direct targeting of 2h to Fyn, Lyn and Blk enzymatic assays were carried out
showing inhibitory effects in vitro with Ki values in the sub-micromolar range. Notably, these values
are highly comparable to those observed for the parent Si308 (see Table 3 for comparison)
although Si308 showed a lower activity in vitro on cell lines.
16

ACCEPTED MANUSCRIPT
Table 5
Inhibitory activities of compound 2h toward Fyn, Lyn and Blk.
Fyn
Lyn
Blk
Ki [µM]
Ki [µM]
Ki [µM]
2h
0.12 ± 0.03
0.19 ± 0.02
0.17 ± 0.08
2.5 In vitro ADME studies
In vitro assays were performed to evaluate the ADME properties of the most promising compounds
Si308 and 2h, and to try to explain the different cellular profile of these molecules. In fact, 2h
proved to be ~1.5 to 5.8 fold more active than Si308 in cells even if the two compounds share a
highly similar inhibitory activity against recombinant enzymes in vitro. Aqueous solubility and
metabolic stability in the presence of human liver microsomes (HLM) were investigated, and the
parallel artificial membrane permeability assay (PAMPA) towards gastrointestinal (GI) membrane
was performed (Table 6).
Table 6
In vitro ADME properties of compound Si308 and 2h.
H2O Sol. µg.mL-1 a
Metabolic Stab.b,c
%
GIb,d Papp (10-6 cm.sec-1)
%(MR)e
Si308
0.01
0.09
85.2
85.0
9.95 (70.4)
2h
10.80 (15.2)
b
aDetermined by LC−MS-MS. Data represent mean values of at least three experiments. Determined by UV/LC−MS.
Data represent mean values of at least three experiments. cExpressed as percentage of unmodified drug.
dGastrointestinal parallel artificial membrane permeability assay. eMembrane retention (MR) expressed as percentage of
compound unable to reach the acceptor compartment.
Compound 2h possesses a water solubility of an order of magnitude higher than Si308, which is
explained by the presence of a polar hydroxyl group in the molecule. Moreover, both compounds
exhibit comparable metabolic stability in HLM. Regarding the passive permeability through the GI
membrane, 2h showed a slight enhancement with respect to Si308, but the most significant data is
related to the different MR values. Indeed, Si308 shows a MR percentage ~5 folds higher than 2h
(70.4% vs 15.2%). Taken together, these data could explain the reasons for the higher activity of
2h in cells compared to Si308.
3
CONCLUSIONS
Recent studies have highlighted the central role played by the three members of the SFK (Lyn, Fyn
and Blk) as potential molecular targets in B-cell lymphomas [9,10,13–15]. Moreover, other studies
have demonstrated the involvement of Fyn and Blk in the development of certain T-cell lymphomas
[13–15]. In this work, the Fyn inhibitor Si308 was submitted to in-depth molecular modelling studies
17

ACCEPTED MANUSCRIPT
flanked by enzyme inhibition assays to further investigate its potential as a multitarget SFK
inhibitor. We confirmed that this compound is able to target Lyn and Blk with an efficacy
comparable to that seen against Fyn. As a result, a small library of pyrazolo[3,4-d]pyrimidine
derivatives related to Si308 was synthesized and tested on B- and T- lymphoma cell lines. Among
the twelve synthesized derivatives, 2h was the most potent compound, with an IC50 value below 10
µM against different lymphoma cell lines, and Ki values in the sub-micromolar range against the
catalytic domain of recombinant Fyn, Lyn, and Blk kinases.
Compared to 2h, compound Si308 is less active on the cell lines, despite having Ki values
comparable to those of 2h. As evidenced by in vitro ADME studies, this discrepancy in biological
activity might be attributed to the higher percentage of Si308 that is trapped in the cell membrane
compared to 2h.
In summary, we identified a potent multitarget Fyn-Blk-Lyn inhibitor endowed with significant
antiproliferative activity on different human lymphoma cell lines and a good pharmacokinetic
profile. These results pave the way to further development of this class of pyrazolo[3,4-
d]pyrimidines as promising agents for the treatment of lymphomas.
4
MATERIALS AND METHODS
4.1 Computational details
4.1.1 Protein preparation.
The crystallographic structures of Fyn and Lyn (PDB ID: 2DQT [21] and 5XY1 [22], respectively)
were downloaded from the Protein Data Bank (www.rcsb.org/pdb) [29] and prepared using the
Protein Preparation Wizard [30] workflow implemented in Maestro Suite Version 11.2.013. In detail,
co-crystallized inhibitors and water molecules were removed, hydrogen atoms, side chains and
missing loops were added using Prime [23,24], and bond orders were assigned. In addition, an
optimization of the network of hydrogen bonds at pH 7.4 was carried out. Finally, the systems were
refined by performing a restrained energy minimization with the OPLS3 force field that was
stopped when the RMSD of the heavy atoms was below 0.30 Å.
4.1.2 Homology modelling.
The sequence of the catalytic domain of human Blk was obtained from UNIPROT database [31]
(entry P51451) and used for a sequence similarity search with BLAST in the PDB.
The crystallographic structure of the human kinase domain of HCK, PDB ID: 2HK5 [32] was
selected as template for the construction of the homology model by the software Prime [23,24]
implemented in Maestro Suite [33]. The loop refinement task was run, using the serial loop
sampling approach and choosing respectively “Default” as level of accuracy (recommended for
loop length up to 5 residues). In addition, the side chain conformation of residues that do not derive
from the template was refined using the "Default" sampling algorithm. Finally, the system was
energy minimized with the OPLS3 force field until the RMSD of the heavy atoms was below 0.30 Å.
18

ACCEPTED MANUSCRIPT
Model evaluation was performed through the analysis of Ramachandran plots (see Supplementary
materials, Figure S1). Moreover, the reliability of the model was verified using the tool ProSA-web
[25,26], which uses knowledge based potentials of mean force to evaluate the energy of the model.
In particular, the z-score obtained for the structure under consideration is a measure of the
deviation of the total energy of the structure with respect an energy derived from random
conformation.
In Supplementary materials Figure S2 the z-score of our model is reported in the form of a black
dot together with the z-scores of all experimentally determined proteins currently available in the
PDB (groups of structures from different sources are distinguished by different colours).
4.1.3 Docking studies.
Molecular docking simulations were carried out by the GLIDE software [20] (SP procedure). In
detail, the prepared protein systems were used to generate the receptor grids that were centred on
the corresponding crystallographic ligands. After grids preparation, compounds were drawn and
energy minimized using Maestro [33] and MacroModel [34], respectively. Finally, molecules were
docked, treating the proteins as rigid using the default values.
4.1.4 Molecular dynamics (MD).
Maestro 11 [33] was initially used to remove all hydrogens atom from the structures provided by
the Protein Preparation Wizard and homology modelling, to cap the N and C terminal residues of
the protein, and to extract the ligand from the binding site.
Antechamber [35] was used to assign atomic partial charges (AM1-BCC method) and atom types
to the ligands, while Parmchk2 was used to write the force field modification (frcmod) files
containing any force field parameters that are needed for the molecules but not supplied by
Generalized Amber Force Field (GAFF) [36] that was used to assign parameters to the ligands.
The parametrization of the proteins was performed by means of the leap module of Amber16 using
the ff14SB force field [37]. Hydrogen atoms were added to each complex, before solvation in a
TIP3P cuboid water box having a distance between the walls and the protein of 20 Å. An
appropriate number of counterions was added to neutralize the systems.
Before MD simulations, two steps of energy minimization were performed to remove bad contacts
and to relax the solvated complexes. First, the complex was kept fixed with a positional restraint
weight of 100 kcal/mol·Å2, allowing in this way the minimization of water molecules. 5000 steps of
energy minimization were run, of which the first 3000 with the steepest descent algorithm and the
last part with the conjugate gradient algorithm. Then, the entire system was energy minimized
using 10000 steps in which the first 7000 used the steepest descent algorithm and the last the
conjugate gradient algorithm.
MD with periodic boundary conditions at constant volume were run for 500 ps to heat the solvated
system to 300 K, using positional restrains with a weight of 2 kcal/mol·Å2 on the protein to prevent
too large fluctuations during the heating step.
19

ACCEPTED MANUSCRIPT
To achieve density equilibration, 1.5 ns of MD with periodic boundaries conditions at constant
pressure were performed at 300 K (by using the Berendsen barostat and the Langevin thermostat
to maintain the pressure and the temperature of our system constant) in two steps, the first of 500
ps and the second of 1000 ps. Also during these steps, positional restrains with a decreasing
weight of 2 kcal/mol and 0.5 kcal/mol·Å2 were used on the protein. Finally, 50 ns of unrestrained
MD simulation were run at constant temperature (300 K) and constant pressure (1 atm).
During MD simulations, bonds containing hydrogen atoms were restrained with the SHAKE
algorithm, and 2 fs time step was used. The electrostatic interactions were calculated using the
PME (Particle Mesh Ewald) method with a cutoff of 10 Å for nonbonded interactions.
The module cpptraj implemented in Amber16 was used to analyse trajectories.
4.1.5 MM-GBSA Analysis.
This method combining molecular mechanics calculations and continuum solvation models is used
to evaluate binding free energy but also to perform free energy decomposition into contributions
originating from different group of atoms [38].
In principle, the calculation of the binding free energy require three independent MD simulations
(for the complex, protein, and ligand separately), but typically is used the approximation that no
significant conformational changes occur upon binding [39], so we used only the simulations of the
complexes to obtain the snapshots for the other 2 species that are necessary for the calculation
(single trajectory approach).
For all the complexes 200 snapshots were used for the calculation. How is generally made when
only a comparison of states of similar entropy is desired, we did not perform the calculation of the
entropy term being computationally expensive and having a large margin of error; therefore, the
energy values produced in the work are attributable to ∆Hbin.
The MM-GBSA approach was also used to perform a pairwise energy decomposition in order to
map the interactions within the binding site that have a dominant role in the stabilization of the
complexes.
4.2 Chemistry
All commercially available chemicals and solvents were used as purchased (Merck-Sigma Aldrich,
Milan, Italy, or Alfa-Aesar, Karlsruhe, Germany). Anhydrous toluene was dried over sodium and
benzophenone as an indicator; anhydrous THF was obtained by passing the solvent over a column
of neutral alumina followed by storage over 3 Å molecular sieves; anhydrous CH2Cl2 and anhy-
drous dimethylformamide (DMF) were used as purchased. Chromatographic separations were per-
formed on columns packed with silica gel 220-440 mesh, 60 Å, for flash technique, or silica gel
0.060-0.200 mm, 40 Å. TLC (thin layer chromatography) was carried out using Merck TLC plates
silica gel 60 F254. Melting points (Mp) were determined with a Büchi B-540 apparatus and are un-
corrected. 1H NMR and 13C NMR spectra were recorded on a Bruker AC200F spectrometer (at 400
20

ACCEPTED MANUSCRIPT
1
1
MHz for H and 100 MHz for 13C) or on a Varian Gemini 200 (at 200 MHz for H and 50 MHz for
13C). Chemical shifts (δ scale) are reported in parts per million (ppm) with reference to solvent resi-
dues in DMSO-d6 (2.50 ppm for proton and 39.52 ppm for carbon), CDCl3 (7.26 ppm for proton
and 77.16 ppm for carbon). Data are shown as following: chemical shift, multiplicity (s = singlet, d =
doublet, t = triplet, q = quartet, sx = sextet, m = multiplet, br s = broad singlet), coupling constant
(J) in Hertz (Hz) and integration. Mass spectra (MS) were acquired using an Agilent 1100 LC-MSD
VL system (G1946C) by direct injection with a 0.4 mL/min flow rate using a binary solvent system
of 95/5 CH3OH/ H2O. UV detection was monitored at 254 nm. Mass spectra were acquired in posi-
tive mode scanning over the mass range 105-1500 m/z, using a variable fragmentor voltage of 10-
70 mV. IR spectra were measured in KBr with a Perkin-Elmer 398 spectrophotometer. Elemental
analysis for C, H and N was determined using Thermo Scientific Flash 2000 and results were with-
in ±0.4% of the theoretical value. All target compounds possessed a purity of ≥95% as verified by
elemental analyses by comparison with the theoretical values.
Compounds Si308, 1b, 8 and 12a have been already reported by us [17].
4.2.1 General procedure for the synthesis of derivatives 6a-f.
A 60% sodium hydride dispersion in mineral oil (1.21 g, 30.3 mmol) was added in small batches to
a solution of malonitrile (1.00 g, 15.1 mmol) in anhydrous THF (25 mL) precooled at 0-5 °C. After
30 min at 0-5 °C, the suitable acyl chloride (15.1 mmol) was added dropwise. The orange solution
was stirred at room temperature for 1.5 h, until the complete transformation of the acyl chloride to
the corresponding enol derivative (not isolated intermediates 3a-f, reaction monitored via TLC,
eluent Et2O/MeOH, 8:2). Then dimethylsulfate (1.75 mL, 18.2 mmol) was added dropwise and the
solution was refluxed for 3-6 h, monitoring via TLC (eluent Et2O/MeOH, 8:2) the formation of the
methoxy(phenyl)methylene]malononitrile derivatives 4a-f (not isolated). Finally, 2-hydrazino-1-
phenylethanol 5 (4.62 g, 30.2 mmol) dissolved in anhydrous THF (2 mL) was added and the
reaction was refluxed for 4 h (reaction monitored via TLC, eluent Et2O). After cooling to room
temperature, water (25 mL) and conc NH3 (5 mL) were added under stirring. After 15 min, THF was
removed under reduced pressure and the aqueous phase was extracted with CH2Cl2 (3 × 30 mL).
Organic phases were washed with water (15 mL) and brine (15 mL), dried (Na2SO4), and
evaporated under reduced pressure. The crudes were purified by flash chromatography (silica gel
0.060-0.200 mm, 40 Å) using a proper gradient elution from petroleum ether (PE) (bp 40-60
°C)/Et2O (8:2) to Et2O to afford compounds 6a-f.
4.2.1.1 5-Amino-3-(4-bromophenyl)-1-(2-hydroxy-2-phenylethyl)-1H-pyrazole-4-carbonitrile (6a).
1
Yield: 41%. White solid. Mp: 186-188 °C. H NMR (400 MHz, CDCl3): δ (ppm) 7.76-7.74 (m, 2H),
7.54-7.53 (m, 2H), 7.40-7.37 (m, 5H), 5.19-5.17 (m, 1H), 4.79 (br s, 1H), 4.27-4.24 (m, 1H), 4.13-
4.07 (m, 1H). 13C NMR (100 MHz, CDCl3): δ (ppm) 182.93, 153.29, 149.54, 140.06, 132.11,
130.89, 129.09, 128.84, 127.95, 125.94, 123.73, 115.22, 74.63, 55.85. IR (KBr) cm−1: 3500-3300
21

ACCEPTED MANUSCRIPT
(OH), 3398, 3329 (NH2), 2194 (CN). MS (ESI): m/z = 382.9 [M]+. Anal. calcd. for C18H15N4OBr: C
56.41, H 3.95, N 14.62; found: C 56.52, H 4.01, N 14.25.
4.2.1.2 5-Amino-1-(2-hydroxy-2-phenylethyl)-3-(4-iodophenyl)-1H-pyrazole-4-carbonitrile (6b).
1
Yield: 41%. White solid. Mp: 173-177 °C. H NMR (200 MHz, CDCl3): δ (ppm) 7.81-7.77 (m, 2H),
7.67-7.63 (m, 2H), 7.44-7.40 (m, 5H), 5.23-5.19 (m, 1H), 4.37-4.06 (m, 2H). 13C NMR (50 MHz,
CDCl3): δ (ppm) 180.87, 152.91, 152.00, 138.81, 137.83, 133.25, 128.92, 128.58, 127.89, 125.94,
115.54, 96.77, 72.51, 50.23. IR (KBr) cm−1: 3500-2800 (OH), 3398, 3329 (NH2), 2195 (CN). MS
(ESI): m/z = 431.7 [M+H]+. Anal. calcd. for C18H15N4OI: C 50.25, H 3.51, N 13.02; found: C 50.55,
H 3.56, N 13.30.
4.2.1.3 5-Amino-3-(4-ethylphenyl)-1-(2-hydroxy-2-phenylethyl)-1H-pyrazole-4-carbonitrile (6c).
1
Yield: 44%. White solid. Mp: 161-164 °C. H NMR (200 MHz, CDCl3): δ (ppm) 7.84-7.81 (m, 2H),
7.44-7.40 (m, 5H), 7.30-7.27 (m, 2H), 5.13-5.09 (m, 1H), 4.67 (br s, 2H), 4.16-4.03 (m, 2H), 2.71
(q, J = 7.4 Hz, 2H), 1.28 (t, J = 7.4 Hz, 3H). 13C NMR (50 MHz, CDCl3): δ (ppm) 152.91, 152.05,
146.89, 138.88, 133.23, 129.40, 128.93, 128.54, 125.96, 125.21, 123.62, 115.54, 72.56, 50.28,
40.11, 28.55. IR (KBr) cm−1: 3500-2800 (OH), 3393, 3320 (NH2), 2216 (CN). MS (ESI): m/z = 333.1
[M+H]+. Anal. calcd. for C20H20N4O: C 72.27, H 6.06, N 16.86; found: C 72.14, H 6.14, N 16.55.
4.2.1.4 5-Amino-1-(2-hydroxy-2-phenylethyl)-3-(4-propylphenyl)-1H-pyrazole-4-carbonitrile (6d).
1
Yield: 48%. White solid. Mp: 107-109 °C. H NMR (200 MHz, CDCl3): δ (ppm) 7.83-7.80 (m, 2H),
7.43-7.29 (m, 5H), 7.29-7.24 (m, 2H), 5.28-5.22 (m, 1H), 4.30-4.29 (m, 1H), 4.17-4.12 (m, 1H),
2.64 (t, J = 7.2 Hz, 2H), 1.68 (sx, J = 7.2 Hz, 2H), 0.97 (t, J = 7.2 Hz, 3H). 13C NMR (50 MHz,
CDCl3): δ (ppm) 152.91, 152.03, 151.47, 138.80, 133.23, 129.45, 128.96, 128.56, 125.99, 125.21,
122.98, 115.54, 72.54, 50.92, 37.61, 24.70, 12.43. IR (KBr) cm−1: 3500-2800 (OH), 3392, 3324
(NH2), 2217 (CN). MS (ESI): m/z = 347.3 [M+H]+. Anal. calcd. for C21H22N4O: C 72.81, H 6.40, N
16.17; found: C 72.68, H 6.72, N 16.45.
4.2.1.5 5-Amino-3-(2-chlorophenyl)-1-(2-hydroxy-2-phenylethyl)-1H-pyrazole-4-carbonitrile (6e).
1
Yield: 37%. White solid. Mp: 131-134 °C. H NMR (200 MHz, CDCl3): δ (ppm) 7.48-7.29 (m, 9H),
5.19-5.16 (m, 1H), 4.20-4.11 (m, 2H). 13C NMR (50 MHz, CDCl3): δ (ppm) 151.27, 148.70, 139.07,
132.20, 130.27, 129.43, 129.18, 128.32, 127.87, 127.62, 125.82, 125.16, 124.73, 113.47, 73.36,
54.70. IR (KBr) cm−1: 3450-2800 (OH), 3426, 3327 (NH2), 2218 (CN). MS (ESI): m/z = 339.8
[M+H]+. Anal. calcd. for C18H15N4OCl: C 63.81, H 4.46, N 16.54; found: C 63.88, H 4.72, N 16.38.
4.2.1.6 5-Amino-3-(3-chlorophenyl)-1-(2-hydroxy-2-phenylethyl)-1H-pyrazole-4-carbonitrile (6f).
1
Yield: 39%. White solid. Mp: 172-180 °C. H NMR (200 MHz, CDCl3): δ (ppm) 7.46-7.27 (m, 9H),
5.16-5.12 (m, 1H), 4.21-4.09 (m, 2H). 13C NMR (50 MHz, CDCl3): δ (ppm) 151.12, 147.89, 139.17,
132.65, 130.65, 129.55, 129.35, 128.10, 127.91, 127.82, 125.54, 125.19, 124.18, 113.51, 73.44,
55.81. IR (KBr) cm−1: 3450-2900 (OH), 3407, 3327 (NH2), 2218 (CN). MS (ESI): m/z = 339.7
[M+H]+. Anal. calcd. for C18H15N4OCl: C 63.81, H 4.46, N 16.54; found: C 63.86, H 4.56, N 16.83.
4.2.2 General procedure for the synthesis of derivatives 7a-f.
22

ACCEPTED MANUSCRIPT
A suspension of the suitable intermediate 6a-d (3 mmol) in formamide (18 mL, 450 mmol) was
heated at 190 °C for 3-4 h and then poured into water (40 mL). The crude solid was filtered,
washed with water, suspended in ethanol, and boiled with charcoal for 10 min. The solid dissolved
at the ethanol boiling point. After charcoal filtration, compounds 7a-f precipitated as pure solids.
4.2.2.1
2-[4-Amino-3-(4-bromophenyl)-1H-pyrazolo[3,4-d]pyrimidin-1-yl]-1-phenylethanol
(7a).
1
Yield: 65%. White solid. Mp: 198-201 °C. H NMR (400 MHz, CDCl3): δ (ppm) 8.28 (s, 1H), 7.73-
7.71 (m, 2H), 7.55-7.51 (m, 2H), 7.45-7.29 (m, 5H), 5.32-5.31 (m, 1H), 4.76-4.60 (m, 2H). 13C NMR
(100 MHz, CDCl3): 152.96, 147.21, 145.77, 140.25, 133.42, 131.50, 130.54, 129.83, 128.87,
127.07, 125.99, 125.15, 97.10, 73.00, 55.14. IR (KBr) cm−1: 3500-2900 (OH), 3406, 3293 (NH2).
MS (ESI): m/z = 411.9 [M+H]+. Anal. calcd. for C19H16N5OBr: C 55.62, H 3.95, N 17.07; found: C
55.81, H 4.06, N 16.74.
4.2.2.2 2-[4-Amino-3-(4-iodophenyl)-1H-pyrazolo[3,4-d]pyrimidin-1-yl]-1-phenylethanol (7b). Yield:
44%. White solid. Mp: 218-220 °C. 1H NMR (200 MHz, CDCl3): δ (ppm) 8.32 (s, 1H), 8.00-7.81 (m,
4H), 7.46-7.29 (m, 5H), 5.40-5.23 (m, 1H), 4.87-4.56 (m, 2H). 13C NMR (50 MHz, CDCl3): 159.51,
153.86, 153.45, 153.07, 138.89, 137.84, 132.61, 130.77, 128.92, 128.56, 127.83, 125.91, 96.72,
71.60, 54.28. IR (KBr) cm−1: 3500-2900 (OH), 3408, 3294 (NH2). MS (ESI): m/z = 458.6 [M+H]+.
Anal. calcd. for C19H16N5OI: C 49.91, H 3.53, N 15.32; found: C 50.13, H 3.53, N 15.27.
4.2.2.3 2-[4-Amino-3-(4-ethylphenyl)-1H-pyrazolo[3,4-d]pyrimidin-1-yl]-1-phenylethanol (7c). Yield:
1
60%. White solid. Mp: 54-59 °C. H NMR (400 MHz, DMSO-d6): δ (ppm) 8.19 (s, 1H), 7.55-7.53
(m, 2H), 7.37-7.20 (m, 7H), 5.56 (br s, 1H), 5.16-5.14 (m, 1H), 4.53-4.48 (m, 1H), 4.36-4.31 (m,
1H), 2.66 (q, J = 7.6 Hz, 2H), 1.22 (t, J = 7.6 Hz, 3H). 13C NMR (100 MHz, DMSO-d6): 158.71,
156.32, 154.92, 147.30, 144.81, 144.10, 143.61, 130.90, 128.75, 128.04, 127.91, 126.59, 97.82,
71.93, 54.18, 28.53, 16.06. IR (KBr) cm−1: 3600-2800 (OH), 3468, 3303 (NH2). MS (ESI): m/z =
360.3 [M+H]+. Anal. calcd. for C21H21N5O: C 70.17, H 5.89, N 19.48; found: C 70.05, H 6.18, N
19.77.
4.2.2.4
2-[4-Amino-3-(4-propylphenyl)-1H-pyrazolo[3,4-d]pyrimidin-1-yl]-1-phenylethanol
(7d).
Yield: 45%. White solid. Mp: 57-58 °C. 1H NMR (400 MHz, CDCl3): δ (ppm) 8.28 (s, 1H), 7.56-7.54
(m, 2H), 7.46-7.44 (m, 2H), 7.37-7.29 (m, 5H), 5.30-5.27 (m, 1H), 4.92 (br s, 1H), 4.70-4.65 (m,
2H), 2.66 (t, J = 7.6 Hz, 2H), 1.68 (sx, J = 7.6 Hz, 2H), 0.97 (t, J = 7.6 Hz, 3H). 13C NMR (100 MHz,
CDCl3): 155.88, 153.59, 151.04, 146.14, 145.18, 143.78, 140.60, 129.90, 129.52, 129.14, 128.69,
128.12, 126.08, 73.34, 55.27, 37.70, 24.32, 14.00. IR (KBr) cm−1: 3400-2900 (OH), 3471, 3304
(NH2). MS (ESI): m/z = 374.1 [M+H]+. Anal. calcd. for C22H23N5O: C 70.76, H 6.21, N 18.75;
found: C 70.62, H 6.40, N 18.64.
4.2.2.5
2-[4-Amino-3-(2-chlorophenyl)-1H-pyrazolo[3,4-d]pyrimidin-1-yl]-1-phenylethanol
(7e).
1
Yield: 46%. White solid. Mp: 126-129 °C. H NMR (200 MHz, CDCl3): δ (ppm) 8.35 (s, 1H), 7.53-
7.29 (m, 9H), 5.38-5.26 (m, 1H), 4.76-4.74 (m, 2H). 13C NMR (50 MHz, CDCl3): 156.34, 154.05,
153.61, 142.38, 139.60, 134.54, 133.37, 129.68, 128.50, 127.96, 127.56, 127.06, 125.33, 124.97,
23

ACCEPTED MANUSCRIPT
97.22, 72.26, 54.53. IR (KBr) cm−1: 3600-2800 (OH), 3319, 3183 (NH2). MS (ESI): m/z = 366.3
[M+H]+. Anal. calcd. for C19H16N5OCl: C 62.38, H 4.41, N 19.14; found: C 62.77, H 4.45, N 19.47.
4.2.2.6
2-[4-Amino-3-(3-chlorophenyl)-1H-pyrazolo[3,4-d]pyrimidin-1-yl]-1-phenylethanol
(7f).
Yield: 50%. White solid. Mp: 84-89 °C. 1H NMR (400 MHz, CDCl3): δ (ppm) 8.30 (s, 1H), 7.68-7.60
(m, 1H), 7.52-7.30 (m, 8H), 5.31-5.28 (m, 1H), 4.95 (br s, 1H), 4.73-4.62 (m, 2H). 13C NMR (100
MHz, CDCl3): 173.20, 155.62, 153.59, 151.04, 144.67, 140.60, 136.01, 133.72, 131.17, 130.15,
128.88, 128.75, 126.27, 126.08, 97.81, 73.36, 55.27. IR (KBr) cm−1: 3550-2800 (OH), 3319, 3180
(NH2). MS (ESI): m/z = 366.4 [M+H]+. Anal. calcd. for C19H16N5OCl: C 62.38, H 4.41, N 19.14;
found: C 62.19, H 4.78, N 19.13.
4.2.3 General procedure for the synthesis of compounds 2a-f.
SOCl2 (80 µL, 1.1 mmol) was added dropwise to a solution of the suitable intermediate 7a-f (0.5
mmol) in anhydrous CH2Cl2 (5 mL), and the reaction was stirred at room temperature for 12 h
under nitrogen atmosphere. Water (5 mL) and 1N NaOH (1 mL) were added with caution, and the
aqueous phase was extracted with CH2Cl2 (2 × 5 mL). Then the organic phase was washed with
water (5 mL) and brine (5 mL), dried (Na2SO4), and concentrated under reduced pressure. The
crudes were purified by column chromatography (silica gel 220-440 mesh, 60 Å) using a proper
mixture of CH2Cl2/MeOH (99:1 to obtain compounds 2a, 2c, 2e,f, 98:2 to obtain compound 2d, and
95:5 to obtain compound 2b) as the eluent to afford compounds 2a-f.
4.2.3.1 3-(4-Bromophenyl)-1-(2-chloro-2-phenylethyl)-1H-pyrazolo[3,4-d]pyrimidin-4-amine (2a).
1
Yield: 42%. White solid. Mp: 196-199 °C. H NMR (200 MHz, DMSO-d6): δ (ppm) 8.26 (s, 1H),
7.75-7.71 (m, 2H), 7.61-7.53 (m, 4H), 7.38-7.36 (m, 3H), 5.88-5.68 (m, 1H), 5.12-4.92 (m, 1H),
4.89-4.74 (m, 1H). 13C NMR (50 MHz, DMSO-d6): 159.57, 153.89, 153.41, 137.44, 132.64, 131.35,
130.78, 129.67, 128.96, 128.21, 127.73, 124.01, 101.74, 59.56, 49.37. IR (KBr) cm−1: 3455, 3307
(NH2). MS (ESI): m/z = 429.7 [M+H]+. Anal. calcd. for C19H15N5BrCl: C 53.23, H 3.53, N 16.34;
found: C 53.12, H 3.57, N 16.11.
4.2.3.2 1-(2-Chloro-2-phenylethyl)-3-(4-iodophenyl)-1H-pyrazolo[3,4-d]pyrimidin-4-amine (2b).
1
Yield: 53%. White solid. Mp: 215-218 °C. H NMR (200 MHz, DMSO-d6): δ (ppm) 8.28 (s, 1H),
7.94-7.89 (m, 2H), 7.55-7.38 (m, 7H), 5.87-5.63 (m, 1H), 5.13-4.98 (m, 1H), 4.89-4.76 (m, 1H). 13C
NMR (50 MHz, DMSO-d6): 159.43, 153.87, 153.41, 153.08, 137.85, 137.42, 132.61, 130.74,
128.91, 128.64, 127.89, 127.72, 96.72, 59.50, 49.37. IR (KBr) cm−1: 3450, 3303 (NH2). MS (ESI):
m/z = 476.5 [M+H]+. Anal. calcd. for C19H15N5ICl: C 47.97, H 3.18, N 14.72; found: C 47.99, H
3.35, N 14.66.
4.2.3.3 1-(2-Chloro-2-phenylethyl)-3-(4-ethylphenyl)-1H-pyrazolo[3,4-d]pyrimidin-4-amine (2c).
1
Yield: 53%. White solid. Mp: 228-229 °C. H NMR (200 MHz, DMSO-d6): δ (ppm) 8.27 (s, 1H),
7.60-7.56 (m, 3H), 7.42-7.37 (m, 6H), 5.87-5.69 (m, 1H), 5.11-4.99 (m, 1H), 4.91-4.77 (m, 1H),
2.70 (q, J = 7.2 Hz, 2H), 1.24 (t, J = 7.2 Hz, 3H). 13C NMR (50 MHz, DMSO-d6): 159.53, 153.87,
153.07, 146.81, 137.45, 132.69, 130.73, 129.45, 128.93, 128.61, 127.73, 125.21, 98.66, 72.56,
24

ACCEPTED MANUSCRIPT
49.36, 28.52, 17.04. IR (KBr) cm−1: 3453, 3305 (NH2). MS (ESI): m/z = 378.7 [M+H]+. Anal. calcd.
for C21H20N5Cl: C 66.75, H 5.33, N 18.53; found: C 66.91, H 5.40, N 18.56.
4.2.3.4 1-(2-Chloro-2-phenylethyl)-3-(4-propylphenyl)-1H-pyrazolo[3,4-d]pyrimidin-4-amine (2d).
1
Yield: 46%. White solid. Mp: 190-192 °C. H NMR (200 MHz, DMSO-d6): δ (ppm) 8.27 (s, 1H),
7.66-7.55 (m, 4H), 7.39-7.35 (m, 5H), 5.87-5.63 (m, 1H), 5.09-4.94 (m, 1H), 4.89-4.76 (m, 1H),
2.60 (t, J = 7.6 Hz, 2H), 1.63 (sx, J = 7.6 Hz, 2H), 0.93 (t, J = 7.6 Hz, 3H). 13C NMR (50 MHz,
DMSO-d6): 159.55, 153.83, 153.42, 153.07, 151.48, 137.45, 132.65, 129.48, 128.92, 128.63,
127.74, 125.26, 98.68, 72.58, 49.48, 28.50, 17.24, 12.53. IR (KBr) cm−1: 3454, 3303 (NH2). MS
(ESI): m/z = 392.5 [M+H]+. Anal. calcd. for C22H22N5Cl: C 67.42, H 5.66, N 17.87; found: C 67.63,
H 5.36, N 17.60.
4.2.3.5 3-(2-Chlorophenyl)-1-(2-chloro-2-phenylethyl)-1H-pyrazolo[3,4-d]pyrimidin-4-amine (2e).
1
Yield: 39%. White solid. Mp: 181-184 °C. H NMR (400 MHz, DMSO-d6): δ (ppm) 8.21 (s, 1H),
7.65-7.59 (m, 1H), 7.56-7.41 (m, 5H), 7.34-7.32 (m, 3H), 5.68-5.64 (m, 1H), 5.01-4.95 (m, 1H),
4.82-4.77 (m, 1H). 13C NMR (100 MHz, DMSO-d6): 158.25, 156.55, 154.77, 141.98, 138.43,
133.33, 132.45, 131.82, 131.44, 130.50, 129.39, 129.12, 128.07, 127.96, 99.34, 60.93, 53.07. IR
(KBr) cm−1: 3496, 3296 (NH2). MS (ESI): m/z = 385.1 [M+H]+. Anal . calcd. for C19H15N5Cl2: C
59.39, H 3.93, N 18.23; found: C 59.60, H 3.95, N 17.93.
4.2.3.6 3-(3-Chlorophenyl)-1-(2-chloro-2-phenylethyl)-1H-pyrazolo[3,4-d]pyrimidin-4-amine (2f).
1
Yield: 42%. White solid. Mp: 162-163 °C. H NMR (200 MHz, DMSO-d6): δ (ppm) 8.28 (s, 1H),
7.66-7.55 (m, 6H), 7.40-7.36 (m, 3H), 5.79-5.71 (m, 1H), 5.12-4.98 (m, 1H), 4.88-4.76 (m, 1H). 13C
NMR (50 MHz, DMSO-d6): 158.43, 156.66, 155.09, 142.12, 138.58, 133.43, 132.42, 131.78,
131.18, 130.56, 129.94, 129.15, 128.14, 122.28, 99.35, 61.03, 53.26. IR (KBr) cm−1: 3473, 3294
(NH2). MS (ESI): m/z = 385.9 [M+H]+. Anal. calcd. for C19H15N5Cl2: C 59.39, H 3.93, N 18.23;
found: C 59.54, H 4.00, N 18.46.
4.2.4 2-(4-Amino-3-iodo-1H-pyrazolo[3,4-d]pyrimidin-1-yl)-1-phenylethanone 9.
A solution of TBAF 1 M in THF (7.05 mL, 7.05 mmol) was added to a solution of 3-iodo-1H-
pyrazolo[3,4-d]pyrimidin-4-amine 8 (1.84 g, 7.05 mmol) in anhydrous toluene (40 mL), and the
mixture was heated at 70 °C for 1 h. Then, 2-bromoacetophenone (1.7 g, 8.46 mmol) was added
and the reaction was stirred at 70 °C for 2 h. After cooling to room temperature, toluene was
evaporated, H2O was added and the aqueous phase was extracted with CH2Cl2 (3 x 50 mL). The
organic phases were washed with water (3 x 50 mL), dried (Na2SO4), and concentrated under
reduced pressure. The crude was purified by column chromatography (silica gel 220-440 mesh, 60
Å) using a gradient elution from AcOEt/n-hexane (9:1) to AcOEt/MeOH (9:1) to afford the
1
compound 9. Yield: 47%. White solid. Mp: 227-231 °C. H NMR (400 MHz, DMSO-d6): δ (ppm)
8.15 (s, 1H), 8.05-8.03 (m, 2H), 7.72-7.68 (m, 1H), 7.59-7.55 (m, 2H), 5.96 (s, 2H). 13C NMR (100
MHz, DMSO-d6): 193.24, 158.24, 156.80, 154.64, 134.74, 129.47, 129.11, 128.51, 103.81, 90.39,
57.78. IR (KBr) cm−1: 3455, 3302 (NH2), 1706 (CO). MS (ESI): m/z = 380.6 [M+H]+. Anal. calcd.
25

ACCEPTED MANUSCRIPT
for C13H10N5OI: C 41.18, H 2.66, N 18.47; found: C 41.27, H 2.91, N 18.13.
4.2.5 2-(4-Amino-3-iodo-1H-pyrazolo[3,4-d]pyrimidin-1-yl)-1-phenylethanol 10.
NaBH4 (1.08 g, 28.48 mmol) was added to a solution of 2-(4-amino-3-iodo-1H-pyrazolo[3,4-
d]pyrimidin-1-yl)-1-phenylethanone 9 (600 mg, 1.58 mmol) in MeOH (150 mL), and the mixture was
stirred at 0 °C for 3 h. Water (40 mL) was added and the mixture was stirred for other 10 minutes.
Then MeOH was evaporated and the suspension was extracted with CH2Cl2 (3 x 50 mL), the
organic solution was washed with water (100 mL), dried (MgSO4), filtered, and concentrated under
reduced pressure. The crude oil was purified by column chromatography (silica gel 220-440 mesh,
60 Å) using CH2Cl2/MeOH (95:5) as the eluant to afford the pure product 10. Yield: 31%. White
solid. Mp: 186-188 °C. 1H NMR (400 MHz, CDCl3): δ (ppm) 8.08 (s, 1H), 7.24-7.09 (m, 5H), 5.01-
4.98 (m, 1H), 4.52-4.46 (m, 1H), 4.28-4.23 (m, 1H). 13C NMR (100 MHz, CDCl3): 157.07, 154.93,
154.05, 141.19, 128.71, 128.08, 126.19, 104.00, 86.98, 72.86, 55.85. IR (KBr) cm−1: 3500-300
(OH), 3465, 3294 (NH2). MS (ESI): m/z = 382.2 [M+H]+. Anal. calcd. for C13H12N5OI: C 40.96, H
3.17, N 18.37; found C 40.89, H 3.32, N 18.58.
4.2.6 1-(2-Chloro-2-phenylethyl)-3-iodo-1H-pyrazolo[3,4-d]pyrimidin-4-amine 11.
SOCl2 (230 µL, 3.2 mmol) was added dropwise to a solution of 2-(4-amino-3-iodo-1H-pyrazolo[3,4-
d]pyrimidin-1-yl)-1-phenylethanol 10 (370 mg, 0.97 mmol) in anhydrous CH2Cl2 (15 mL), and the
reaction was stirred at room temperature for 12 h under nitrogen atmosphere. Water (10 mL) and 1
N NaOH (2 mL) were added with caution and the aqueous phase was extracted with CH2Cl2 (3 x
15 mL). Then the organic phase was washed with water (30 mL), brine (30 mL), dried (Na2SO4)
and concentrated under reduced pressure. The crude product was purified by column
chromatography (silica gel 220-440 mesh, 60 Å) using CH2Cl2/MeOH (95:5) as the eluant to afford
the pure product 11. Yield: 33%. White solid. Mp: 204-209 °C. 1H NMR (200 MHz, CDCl3): δ (ppm)
8.34 (s, 1H), 7.57-7.29 (m, 5H), 6.43 (br s, 2H), 5.57-5.50 (m, 1H), 5.06-4.94 (m, 1H), 4.79-4.69
(m, 1H). 13C NMR (50 MHz, CDCl3): 156.96, 154.84, 154.03, 141.12, 128.65, 127.03, 126.09,
104.00, 85.97, 72.43, 55.37. IR (KBr) cm−1: 3452, 3301 (NH2). MS (ESI): m/z = 400.4 [M+H]+.
Anal. calcd. for C13H11N5ICl: C 39.07, H 2.27, N 17.53; found C 38.89, H 2.89, N 17.62.
4.2.7 General procedure for the synthesis of compounds 2g,h.
(Hydroxymethyl)phenylboronic acid (76 mg, 0.5 mmol), Cs2CO3 (122 mg, 0.375 mmol) and [1,1′-
bis(diphenylphosphino)ferrocene]dichloropalladium (II) (Pd(dppf)Cl2) (5 mg, 10 mol%) were added
to a solution of the suitable intermediate 10 or 11 (0.125 mmol) in 1,2-dimethoxyethane (DME) (2
mL) and H2O (0.3 mL). Then the mixture was stirred at 90° C for 14 h. After cooling to room
temperature, water (5 mL) was added and the aqueous suspension was extracted with AcOEt (3 x
5 mL). The organic layers were washed with water (15 mL), dried (MgSO4), filtered, and
concentrated under reduced pressure. The crude product was purified by column chromatography
(silica gel 220-440 mesh, 60 Å) using AcOEt/n-hexane (95:5) as the eluant to afford the pure
products 2g and 2h.
26

ACCEPTED MANUSCRIPT
4.2.7.1 2-{4-Amino-3-[4-(hydroxymethyl)phenyl]-1H-pyrazolo[3,4-d]pyrimidin-1-yl}-1-phenylethanol
1
(2g). Yield: 25%. White solid. Mp: 204-206 °C. H NMR (400 MHz, CDCl3): δ (ppm) 8.36 (s, 1H),
7.68-7.66 (m, 2H), 7.56-7.48 (m, 4H), 7.38-7.26 (m, 3H), 5.69 (br s, 2H), 5.30-5.28 (m, 1H), 4.80
(s, 2H), 4.74-4.60 (m, 2H). 13C NMR (100 MHz, CDCl3): 156.07, 153.50, 153.21, 137.69, 135.34,
132.61, 129.11, 128.82, 127.71, 127.35, 126.42, 121.43, 117.91, 100.25, 60.29, 54.10. IR (KBr)
cm−1: 3500-300 (OH), 3501, 3320 (NH2). MS (ESI): m/z = 362.4 [M+H]+. Anal. calcd. for
C20H19N5O2: C 66.47, H 5.30, N 19.38; found C 66.67, H 5.22, N 19.49.
4.2.7.2
{4-[4-Amino-1-(2-chloro-2-phenylethyl)-1H-pyrazolo[3,4-d]pyrimidin-3-yl]phenyl}methanol
1
(2h). Yield: 57%. White solid. Mp: 220-223 °C. H NMR (200 MHz, CDCl3): δ (ppm) 8.38 (s, 1H),
7.67-7.66 (m, 2H), 7.55-7.54 (m, 2H), 7.49-7.47 (m, 2H), 7.36-7.7.31 (m, 3H), 5.62-5.59 (m, 1H),
5.47 (br s, 2H), 5.08-5.03 (m, 1H), 4.82-4.80 (m, 3H). 13C NMR (50 MHz, CDCl3): 156.12, 153.55,
153.23, 137.72, 135.39, 132.62, 129.09, 128.84, 127.70, 127.39, 126.44, 121.46, 117.93, 100.35,
60.32, 54.08. IR (KBr) cm−1: 3500-300 (OH), 3500, 3319 (NH2). MS (ESI): m/z = 380.1 [M+H]+.
Anal. calcd. for C20H18N5OCl: C 63.24, H 4.78, N 18.44; found C 63.27, H 5.11, N 18.09.
4.2.8 3-Iodo-1-(2-phenylethyl)-1H-pyrazolo[3,4-d]pyrimidin-4-amine 12b.
K2CO3 (600 mg, 4.34 mmol) was added to a solution of 3-iodo-1H-pyrazolo[3,4-d]pyrimidin-4-
amine 8 (400 mg, 1.53 mmol) in dry DMF (5 mL), and the mixture was heated at 50-70 °C for 1 h.
Then (2-bromoethyl)benzene (426 mg, 2.30 mmol) was added, and the reaction was stirred at 130
°C for 18 h. After cooling to room temperature, water was added (20 mL) and the precipitated solid
was recrystallized from absolute ethanol to obtain 12b. Yield: 42%. Light brown solid. Mp: 192-193
°C. 1H NMR (400 MHz, CDCl3): δ (ppm) 8.28 (s, 1H), 7.25-7.20 (m, 5H), 6.11 (br s, 2H), 4.59 (t, J =
8.0 Hz, 2H), 3.20 (t, J = 8.0 Hz, 2H). 13C NMR (100 MHz, CDCl3): 159.51, 153.41, 153.07, 135.91,
130.75, 128.91, 128.72, 128.56, 93.36, 49.41, 31.10. IR (KBr) cm−1: 3481, 3358 (NH2). MS (ESI):
m/z = 365.9 [M+H]+. Anal. calcd. for C13H12N5I: C 42.76, H, 3.31 N 19.18; found: C 42.62, H 3.70,
N 18.91.
4.2.9 General procedure for the synthesis of compounds 2i-l.
The suitable boronic acid (1.08 mmol) was added to a suspension of the intermediates 12a or 12b
(0.27 mmol) in dry toluene (5 mL), and the mixture was stirred at room temperature under nitrogen
atmosphere for 10 min. Then Cs2CO3 (350 mg, 1.07 mmol) and [1,1′-
bis(diphenylphosphino)ferrocene]dichloropalladium (II) (Pd(dppf)Cl2) (20 mg, 10 mol%) were
added. The reaction was stirred at 90 °C for 14 h. After cooling to room temperature, water (10 mL)
was added and the aqueous suspension was extracted with AcOEt (3 × 10 mL). The organic phase
was washed with water (10 mL) and brine (10 mL), dried (Na2SO4), and concentrated under
reduced pressure. The crudes were purified by column chromatography (silica gel 0.060-0.200
mm, 40 Å) using a mixture of CH2Cl2/MeOH (95:5) as the eluent to afford compounds 2i-l.
4.2.9.1 3-(4-Fluorophenyl)-1-(2-phenylpropyl)-1H-pyrazolo[3,4-d]pyrimidin-4-amine (2i). Yield:
50%. Yellow solid. Mp: 210-212 °C. 1H NMR (400 MHz, CDCl3): δ (ppm) 8.34 (s, 1H), 7.65-7.61 (m,
27

ACCEPTED MANUSCRIPT
2H), 7.26-7.16 (m, 7H), 5.88 (br s, 2H), 4.57-4.54 (m, 2H), 3.63-3.57 (m, 1H), 1.28 (d, J = 6.8 Hz,
3H). 13C NMR (100 MHz, CDCl3): 163.43, 159.52, 153.86, 153.42, 153.01, 144.21, 132.68, 130.71,
128.95, 128.64, 127.73, 127.18, 115.93, 52.32, 41.67, 19.48. IR (KBr) cm−1: 3463, 3420 (NH2). MS
(ESI): m/z = 348.2 [M+H]+. Anal. calcd. for C20H18N5F: C 69.15, H 5.22, N 20.16; found: C 69.17,
H 5.37, N 20.39.
4.2.9.2 3-(4-Nitrophenyl)-1-(2-phenylpropyl)-1H-pyrazolo[3,4-d]pyrimidin-4-amine (2j). Yield: 23%.
Yellow solid. Mp: 255-257 °C. 1H NMR (400 MHz, CDCl3): δ (ppm) 8.33 (s, 1H), 7.84-7.82 (m, 2H),
7.26-7.22 (m, 7H), 6.21 (br s, 2H), 4.63-4.61 (m, 2H), 3.60-3.58 (m, 1H), 1.32 (d, J = 6.4 Hz, 3H).
13C NMR (100 MHz, CDCl3): 159.55, 153.84, 153.40, 152.93, 144.28, 132.72, 131.12, 130.74,
128.92, 128.61, 127.17, 127.18, 117.15, 52.35, 41.69, 19.52. IR (KBr) cm−1: 3470, 3312 (NH2),
1567, 1350 (NO2). MS (ESI): m/z = 375.0 [M+H]+. Anal. calcd. for C20H18N6O2: C 64.16, H 4.85, N
22.45; found: C 64.45, H 4.68, N 22.11.
4.2.9.3 3-(4-Chlorophenyl)-1-(2-phenylethyl)-1H-pyrazolo[3,4-d]pyrimidin-4-amine (2k) [40]. Yield:
1
48%. Light brown solid. Mp: 190-191 °C. H NMR (400 MHz, CDCl3): δ (ppm) 8.36 (s, 1H), 7.64-
7.62 (m, 2H), 7.53-7.51 (m, 2H), 7.27-7.20 (m, 5H), 5.49 (br s, 2H), 4.67 (t, J = 8.0 Hz, 2H), 3.27 (t,
J = 8.0 Hz, 2H). 13C NMR (100 MHz, CDCl3): 159.54, 153.82, 153.40, 152.97, 135.91, 135.72,
132.67, 130.74, 129.10, 128.92, 128.76, 128.68, 128.54, 49.43, 31.15. IR (KBr) cm−1: 3474, 3323
(NH2). MS (ESI): m/z = 350.0 [M]+. Anal. calcd. for C19H16N5Cl: C 65.24, H 4.61, N 20.02; found:
C 65.43, H 5.00, N 19.97.
4.2.9.4 3-(4-Methylphenyl)-1-(2-phenylethyl)-1H-pyrazolo[3,4-d]pyrimidin-4-amine (2l). Yield: 50%.
Yellow solid. Mp: 173-174°C. 1H NMR (400 MHz, CDCl3): δ (ppm) 8.34 (s, 1H), 7.58-7.56 (m, 2H),
7.35-7.33 (m, 2H), 7.29-7.18 (m, 5H), 5.63 (br s, 2H), 4.66 (t, J = 8.0 Hz, 2H), 3.27 (t, J = 8.0 Hz,
2H), 2.46 (s, 3H). 13C NMR (100 MHz, CDCl3): 159.58, 153.77, 153.43, 153.21, 139.82, 135.97,
132.64, 130.72, 129.33, 128.92, 128.76, 128.55, 125.22, 49.49, 31.17, 21.37. IR (KBr) cm−1: 3477,
3321 (NH2). MS (ESI): m/z = 330.0 [M+H]+. Anal. calcd. for C20H19N5: C 72.93, H 5.81, N 21.26;
found: C 72.69, H 5.59, N 21.25.
4.3 In vitro inhibition assay
Tyrosine kinases were purchased from Merck-Millipore. Reaction were performed according to
manufacturer’s instructions with minor modifications. All substrates were used at least at twice the
concentration of apparent Km. In details, Src and Fyn reactions were performed using 250 µM Src-
peptide (KVEKIGEGTYGVVYK). 100 µM ATP and 0.00087% NP-40. Lyn reaction was performed
using 200 µM Src-peptide, 40 µM ATP and 0.0013% NP-40. Blk reaction was performed using
200 µM Src-Peptide, 100 µM ATP and 0.0013 % NP-40. All reactions were performed using 0.1 µL
of γ−P33 ATP, 10-50 ng of enzyme, 10% DMSO in 10 µL at 30 °C for 10 min. To avoid peptide
adsorption to the plastic surface, protein low-binding tubes were used. Reactions were stopped by
adding 5 µl of phosphoric acid 0.8%. Aliquots (10 µL) were then transfer into a P30 Filtermat
28