Studies on the ATP Binding Site of Fyn Kinase for the Identification of New Inhibitors and Their Evaluation as Potential Agents against Tauopathies and Tumors

Article abstract of DOI:10.1021/acs.jmedchem.5b00140

Fyn is a member of the Src-family of nonreceptor protein-tyrosine kinases. Its abnormal activity has been shown to be related to various human cancers as well as to severe pathologies, such as Alzheimer's and Parkinson's diseases. Herein, a structure-based drug design protocol was employed aimed at identifying novel Fyn inhibitors. Two hits from commercial sources (1, 2) were found active against Fyn with K_i of about 2 μM, while derivative 4a, derived from our internal library, showed a K_i of 0.9 μM. A hit-to-lead optimization effort was then initiated on derivative 4a to improve its potency. Slightly modifications rapidly determine an increase in the binding affinity, with the best inhibitors 4c and 4d having K_is of 70 and 95 nM, respectively. Both compounds were found able to inhibit the phosphorylation of the protein Tau in an Alzheimer's model cell line and showed antiproliferative activities against different cancer cell lines. (Chemical Equation Presented).

Full text of DOI:10.1021/acs.jmedchem.5b00140

Subscriber access provided by UIC Library
Article
Studies on the ATP binding site of Fyn kinase for the
identification of new inhibitors and their evaluation
as potential agents against tauopathies and tumors
Cristina Tintori, Giuseppina La Sala, Giulia Vignaroli, Lorenzo Botta, Anna Lucia Fallacara,
Federico Falchi, Marco Radi, Claudio Zamperini, Elena Dreassi, Lucia Dello Iacono, Donata
Orioli, Giuseppe Biamonti, Mirco Garbelli, Andrea Lossani, Francesca Gasparrini, Tiziano
Tuccinardi, Ilaria Laurenzana, Adriano Angelucci, Giovanni Maga, Silvia Schenone, Chiara
Brullo, Francesca Musumeci, Andrea Desogus, Emmanuele Crespan, and Maurizio Botta
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b00140 • Publication Date (Web): 29 Apr 2015
Downloaded from http://pubs.acs.org on May 3, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Farmaceutiche
Crespan, Emmanuele; Istituto Genetica Molecolare,
Botta, Maurizio; University of Siena, Dipartimento Farmaco Chimico
Tecnologico; University of Siena, Dipartimento Farmaco Chimico
Tecnologico
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Studies on the ATP binding site of Fyn kinase for the identification of
new inhibitors and their evaluation as potential agents against
tauopathies and tumors
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1
1
1
1ꢀ2
Cristina Tintori, Giuseppina La Sala, Giulia Vignaroli, Lorenzo Botta, Anna Lucia Fallacara,
1
1
1
1
1
Federico Falchi, Marco Radi, Claudio Zamperini, Elena Dreassi, Lucia Dello Iacono, Donata
3
3
3
3
2
Orioli, Giuseppe Biamonti, Mirco Garbelli, Andrea Lossani, Francesca Gasparrini, Tiziano
5
6
7
3
8 *
Tuccinardi, Ilaria Laurenzana, Adriano Angelucci, Giovanni Maga, Silvia Schenone, Chiara
8
8
8
3*
1,9
Brullo, Francesca Musumeci, Andrea Desogus, Emmanuele Crespan, and Maurizio Botta
1
Dipartimento Biotecnologie, Chimica e Farmacia, Università degli Studi di Siena, Via A. De
2
Gasperi 2, Iꢀ53100 Siena, Italy. Dipartimento di Chimica e Tecnologie del Farmaco, Università La
3
Sapienza, Piazzale Aldo Moro 5, Iꢀ00185 Roma, Italy. Istituto di Genetica Molecolare, IGMꢀCNR,
4
Via Abbiategrasso 207, Iꢀ27100 Pavia, Italy. Dipartimento di Medicina Molecolare, Sapienza
5
Università di Roma, Piazzale Aldo Moro 5, 00185 Roma, Italy Dipartimento di Farmacia,
6
Università di Pisa, Via Bonanno 6, 56126 Pisa, Italy. IRCCSꢀCentro di Riferimento Oncologico
Basilicata (CROB), Laboratory of Preclinical and Translational Research, Via Padre Pio 1, Rionero
7
in Vulture 85028 Potenza, Italy. Dipartimento di Scienze Cliniche Applicate e Biotecnologiche,
8
Università dell’Aquila, Via Vetoio, 67100 Coppito, L’Aquila, Italy. Dipartimento di Farmacia,
9
Università di Genova, Viale Benedetto XV 3, Iꢀ16132 Genova, Italy. Biotechnology College of
th
Science and Technology, Temple University, Biolife Science Building, Suite 333, 1900 N 12
Street, Philadelphia, Pennsylvania 19122.
ACS Paragon Plus Environment

Page 3 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Abstract
Fyn is a member of the Srcꢀfamily of nonꢀreceptor proteinꢀtyrosine kinases. Its abnormal activity
has been shown to be related to various human cancers as well as to severe pathologies, such as
Alzheimer's and Parkinson's diseases. Herein, a structureꢀbased drug design protocol was employed
aimed at identifying novel Fyn inhibitors. Two hits from commercial source (1, 2) were found
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
active against Fyn with K of about 2 µM while derivative 4a, derived from our internal library,
i
showed a K of 0.9 µM. A hitꢀtoꢀlead optimization effort was then initiated on derivative 4a to
i
improve its potency. Slightly modifications rapidly determine an increase in the binding affinity,
i
with the best inhibitors 4c and 4d having K of 70 nM and 95 nM, respectively. Both compounds
were found able to inhibit the phosphorylation of the protein Tau in an Alzheimer’s model cell line
and showed antiproliferative activities against different cancer cell lines.
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Introduction
1
Fyn is a nonꢀreceptor tyrosine kinase belonging to the Src family kinases (SFKs). The nine
members of this family are grouped into subꢀclasses: the SrcA subfamily which includes Src, Yes,
Fyn and Fgr, the SrcB subfamily containing Lck, Hck, Blk and Lyn, and finally Frk in its own
subfamily. Fyn is a 59ꢀkDa protein comprising 537 amino acids, encoded by the Fyn gene, located
on chromosome 6q21. Three isoforms of Fyn are known: FynB mainly expressed in the brain, FynT
expressed in hematopoietic cells (Tꢀcells) and FynDelta7 which has been identified in peripheral
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
blood mononuclear cells. In vertebrates SFKs share a similar structure that comprises six distinct
functional domains: Src homology domain 4 (SH4), a unique domain, SH3 domain, SH2 domain, a
catalytic domain (SH1), and a Cꢀterminal regulatory region. SH4 domain is a region which contains
1
signals for modification with fatty acids. The unique domain is specific for each Src family protein
and is suggested to be responsible for peculiar interactions with particular receptors and protein
3
targets. SH2 and SH3 domains are involved in interactions with other proteins which regulate the
tyrosine kinase activity. The kinase domain, that catalyzes the transfer of the terminal phosphate
group of the ATP to a tyrosine residue of a protein substrate, shows a typical bilobed structure
formed by a small Nꢀterminal domain, involved in the ATP binding, and a larger Cꢀterminal lobe
which contains the activation loop (Aꢀloop) characterized by a tyrosine residue that is autoꢀ
4
phosphorylated in the active form of the enzyme. The Aꢀloop consists of 28 residues positioned
5
between two conserved tripeptide motifs, DFG (AspꢀPheꢀGly) and APE (AlaꢀProꢀGlu). Besides
sharing the same structure, the SFKs are also characterized by a common regulatory mechanism. In
fact, the activation or inhibition of kinase activity depends on intramolecular interactions between
SH2 and SH3 with the kinase domain and on phosphorylation/dephosphorylation of two critical
6
tyrosines, the first situated in the Aꢀloop and the second in the Cꢀterminal region. Fyn is able to
interact with almost 300 different proteins and, through these interactions, participates in many
7
ꢀ9
cellular pathways, both in physiological and pathological situations. Fyn is involved in the
3
ACS Paragon Plus Environment

Page 5 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
regulation of the immune system, and in Tꢀcell development and activation. It plays a crucial role in
the development of the central nervous system (CNS) where it is implied in myelination and
morphological differentiation associated with the formation of neurite in oligodendrocytes, synapse
10
formation and regulation, oligodendrocyte differentiation and memory formation. Recent
evidences suggest that Fyn hyperactivation/deregulation might contribute to Alzheimer’s disease
(AD) pathogenesis and other tauopathies. These diseases are characterized by the alteration in the
amount or the structure of the Tau protein, a microtubuleꢀassociated protein that constitutes a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11
fundamental component of the neurofibrillary tangles of AD. In normal neurons Tau is present in
the cytoplasm in an unphosphorylated form. On the contrary, Tau results phosphorylated at multiple
sites in AD. In particular, when associated to neurofibrillary tangles, Tau was found to be
1
2
phosphorylated at its amino terminus residue Tyr18, with Fyn being the solely kinase responsible
for such event in AD. Mounting evidences suggest that the phosphorylation of Tyr18 is an early
event in the pathophysiology of AD and leads to conformational changes in Tau, initiating its
13
14
fibrillarization. In addition, amyloidꢀbeta (Aβ) was found to activate Fyn. Moreover,
overexpression of Fyn accelerates synapse loss and the onset of cognitive impairment in a
15
transgenic AD mouse model, while inhibition of Fyn expression rescues synapse loss. The AD
therapeutic approaches currently in clinical trials are focused on Aβ clearance as well as on the
inhibition of its production or aggregation. Therefore, due to its central role in Aβ signal
9
transduction, Fyn represents a unique therapeutic target in AD. On the other hand, Fyn
overexpression has been shown to drive a morphologic transformation in normal cells, leading to
tumor development. In fact Fyn is overexpressed in various cancers, including glioblastoma
16
17
18
multiformae, squamous cell carcinoma of the head and neck, melanoma, breast, ovarian,
19
20
prostate, and pancreatic cancer. Recent studies have shown its involvement also in
21
22
mesothelioma. Lately, Singh and colleagues demonstrated that Fyn kinase activity plays a role in
the progression of chronic myeloid leukemia (CML), because it contributes to BcrꢀAbl induced
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
3
genomic instability, a feature of blast crisis CML. The terminal, blast crisis phase of the disease
remains a clinical challenge. Blast crisis CML is difficult to treat due to resistance to tyrosine kinase
inhibitors, increased genomic instability and acquired secondary mutations. Knockdown of Fyn
leads to decreased cell growth and proliferation in vitro and in vivo. Moreover, it has been
demonstrated that the complete loss of Fyn using genetic knockout models decreases the
proliferation and clonogenic potential of cells transduced with BcrꢀAbl, underscoring a dependency
upon Fyn for BcrꢀAbl mediated growth and clonogenicity. Additionally, using a cell line model of
blast crisis CML, they discovered that overexpression of constitutively active Fyn caused increased
aneuploidy and genomic alterations. Because of the involvement of Fyn in such disease, the search
for Fyn inhibitors represents an expanding field of study. Herein, molecular modelling studies were
combined with organic synthesis toward the development of novel Fyn kinase inhibitors. Both
commercial libraries of small molecules and a family of pyrazolo[3,4ꢀd]pyrimidines previously
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
4ꢀ32
identified by our research group
were virtually screened within the ATP binding site of Fyn in
order to select potential binding inhibitors. A hitꢀtoꢀlead optimization process was then carried out
on the most active Fyn inhibitor identified, resulting in the development of compounds active
against Fyn with K values in the nanomolar range. Selected compounds showed an interesting
i
antiproliferative activity profile against the human leukemia KUꢀ812, the human breast cancer
MDAꢀMDꢀ231 and the human glioblastoma multiforme cell lines. Furthermore they were found
able to inhibit the Fynꢀmediated phosphorylation of the protein Tau in an AD model cell line.
5
ACS Paragon Plus Environment

Page 7 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
RESULTS AND DISCUSSION
Molecular Modeling and Enzymatic Activities on isolated Fyn kinase.
A docking study was employed herein by means of the Gold software with the purpose of
identifying novel ATP competitive Fyn kinase inhibitors. First, a library of commercially available
compounds was analyzed to select the most promising binders of Fyn for biological investigation.
Later, the same computational methodology was applied to screen an in house library of
pyrazolo[3,4ꢀd]pyrimidines (consisting of ~ 300 structurally characterized compounds with > 98%
purity). Once the accuracy of the computational procedure was checked by reproducing the
experimental pose of the coꢀcrystallized staurosporine as well as those of the know active ligands
PP1 and PP2, the docking protocol was applied on the Asinex commercial database. A total of
703,200 entries were evaluated in two consecutive pruning steps to optimize the balance between
the quality of docking and the time required for calculations. Accordingly, the whole compound
collection was first analysed using the Number of Run parameter set to 10, thus allowing to speed
up the process and to handle a large number of ligands in a reasonable computing time. The top
5000 scored hits were selected for a second run of docking calculations for which the Number of
Run parameter was set to 100. The best ranked structures were visually inspected to evaluate the
complementarity between each bound ligand and the protein surface in terms of size, shape and
interactions. Six compounds were finally selected and submitted to biological testing to evaluate
their affinity toward Fyn in a cellꢀfree assay. Results are listed in Table 1. Compound 1 and 2 were
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
i
identified as efficient Fyn inhibitors being endowed with K values of 2.1 and 2.25 µM,
respectively. Figure 1 shows the binding mode of compounds 1 and 2 within the ATP binding site
of Fyn kinase as predicted by docking analysis. The quinolinone scaffold of 1 forms two Hꢀbonds
with the backbone of Met 345 while the 1,3ꢀbenzodioxole ring is directed toward an hydrophobic
cavity of the binding site where it forms a Hꢀbond with the nitrogen backbone of Asp408 and
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
interacts with Val327, Ala407 and the aliphatic chain of Lys299. Remarkably, the binding mode
identified for compound 1 is in perfect agreement to that found for a structurally closed Fyn
33
inhibitor (published as compound VS3) recently identified by us using a different computational
approach. Interestingly, many analogues of compound 1 are available from Asinex and 19 of them
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
3
have been previously selected as potential Fyn kinase inhibitors. Within the virtual screening
approach reported herein, such analogues showed docking dG ranging from ꢀ32 to ꢀ38 KJ/mol,
comparable to that of compound 1 (dG = ꢀ37.48 KJ/mol), further confirming the potentiality of this
chemical scaffold for the development of Fyn kinase inhibitors. With regard to compound 2, the
pyrazole ring makes hydrogen bonding with the NH of Met345 and the oxygen backbone of
Glu343, both belonging to the hinge region. The ethoxyphenyl group fills the bottom side of the
cavity where it establishes hydrophobic contacts with Val327, Ala407 and Lys299 as well as an Hꢀ
bond with the NH backbone of Asp408. The pyrrolidineꢀsulfonylꢀpiperazine side chain is directed
toward the solventꢀexposed region of the binding site. A similarity search for compounds 1 and 2
within Scifinder showed that the latter has never been reported as a protein kinase inhibitor (there
are no kinase ligands with a similarity score greater than 80).
24ꢀ32
On the other hand, an inꢀhouse library of pyrazolo[3,4ꢀd]pyrimidines
was virtually screened
against the active site of Fyn kinase using the docking protocol described before for the second run
of calculations. From a structural point of view, these compounds are related to the Src inhibitors
PP1 and PP2 but bear a different substitution pattern on the position 1, 3, 4, and 6 of the
pyrazolo[3,4ꢀd]pyrimidine nucleus. Seventeen compounds were selected for testing according to the
predicted score values and the binding poses observed (Table 2). Among them, some derivatives
29ꢀ32
(3aꢀi, 3p) have been already reported by us as BcrꢀAbl/Src inhibitors.
Other compounds have
been never reported before. They include derivatives 3j, 3k, 3l, 3o, that bear bulky thioꢀsubstituents
or alkyl group on C6, the N1 styryl derivatives 3m and 3n and the compound 4a, bearing a C3
phenyl ring and a C4 unsubstituted amino group. Three different binding modes emerged from
7
ACS Paragon Plus Environment

Page 9 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
docking studies. Compounds 3aꢀ3o, functionalized at positions 1, 4 and 6 and C3ꢀunsubstituted,
showed an interaction pattern similar to that previously found within the ATP binding site of cꢀSrc
28
(
herein referred to as BM A, Figure 2). The pyrazolo[3,4ꢀd]pyrimidine core occupies the adenine
region, the C4 substituent establishes hydrophobic interactions with Leu329, Phe409, Ile340,
Phe311, Met318, and Leu411, (hydrophobic region I), while the N1 side chain is involved in van
der Waals contacts with Tyr344, Met345, and Leu277 (hydrophobic region II). The C6 substituent
is directed toward the external region, though still involved in nonꢀpolar contacts with Val285 and
Lys299. Two hydrogen bond interactions were also found, one involving the C4 amino group and
the side chain of Thr342, and the other between the N2 of the pyrazolo[3,4ꢀd]pyrimidine nucleus
and the backbone NH group of Met345. Conversely, the binding mode B (BM B, Figure 2) was
identified for compound 4a which brings a substituent at the C3 position while is unsubstituted at
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
34
C6. 4a is a close analogue of PP1 and PP2, the first reported Fyn inhibitors. Accordingly, its
binding mode is consistent with that previously identified for PP1 and PP2 by crystallographic
experiments and reproduced herein by docking studies (compare the BM B in Figure 2 with the
35,36
binding mode of PP2 within the Fyn kinase active site visualised in Figure 3A).
In detail, the
exocyclic amine makes an Hꢀbond with the carbonyl backbone of Glu343, while the nitrogen at N5
position acts as Hꢀbond acceptor in the interaction with Met345. The C3 phenyl group binds to the
residues of hydrophobic region I through van der Waals interactions, while the N1 group is solvent
exposed. Finally, compound 3p which is unsubstituted at both C3 and C6 positions showed the
binding mode C visualized in Figure 2. According to our prediction, the ligand binds the Met345 of
the hinge region through two hydrogen bonds by means of the atom N5 and the exocyclic amino
group at C4. Moreover, the N1 side chain is located within the hydrophobic pocket I formed by
residues Leu329, Phe409, Ile340, Phe311, Met318, and Leu411 while the substituent at C4 interacts
with Tyr344 and Leu277 of hydrophobic region II. Finally, as a result from biological investigation
on isolated Fyn kinase (Table 2), 3b, 3n, 3p and 4a resulted to be the most active compounds being
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
endowed with K values of 1.4 µM, 1.15 µM, 3.5 µM and 0.9 µM, respectively. Derivatives 3f, 3g,
i
3
i, 3l, 3m and 3o exhibited moderate activity with K ranging from 7.5 µM to 16 µM. No activity
i
was detected for the remaining compounds.
Based on this finding, a hit to lead optimization campaign was then initiated on compound 4a to
improve its potency. Nine analogues were synthesized (4bꢀe, 5aꢀe) (Table 3). The C3 moiety was
differently functionalized, while little modifications were carried out on the 2ꢀchloroꢀ2ꢀphenylethyl
N1 side chain by replacing or removing the chlorine atom. Furthermore, with the aim of exploring
the effect caused by the introduction of a Sꢀmethyl moiety at the C6 position of the C3ꢀphenylꢀ
substituted pyrazole[3,4ꢀd]pyrimidine core, derivatives 6a and 6b, previously synthesized by our
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
7
group, were added to this set of compounds. The eleven compounds were tested in a cellꢀfree
assay to evaluate their affinity towards isolated Fyn kinase.
As it can be appreciated from Table 3, derivatives 4c and 4d showed potent in vitro inhibitory effect
i
against Fyn, with K values in the nanomolar range (70 nM and 95 nM, respectively). These
potencies were most likely evoked due to the contribution of a chlorine or methyl substituent in
para position of the C3 phenyl ring (compare 4c with 4a and 4d with 4a).
Interesting activities were also found for compounds 4b, 5b and 5c which exhibited submicromolar
affinities (0.36 µM, 0.78 µM and 0.995 µM, respectively). As a general trend, the substitution of
chlorine by methyl in the N1 side chain led to a considerable reduction of the affinity with about 10ꢀ
fold decreased activities (compare 5a with 4a, 5b with 4c and 5c with 4d). No activity was found
for compounds 6a and 6b bearing a methylthio group at the C6 position. In order to rationalize such
biological data, docking studies were carried out on the second generation series of compounds (4bꢀ
d, 5aꢀe, 6a,b) and the resulting binding modes were compared with that previously identified for the
starting hit 4a (BM B in Figure 2). All studied compounds adopted a pose in line with that of 4a
with the exception of the inactive derivatives 5d and 6a,b. With regard to compounds 6a,b, despite
they maintained an orientation similar to that of 4a within the ATP binding site of Fyn, the presence
9
ACS Paragon Plus Environment

Page 11 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
of the substitution in C6 position does not allow the ligands to properly interact with the residues of
the hinge region moving away the pyrazoloꢀpyrimidine scaffold from such amino acids. As a
consequence, the hydrogen bond between N5 and Met345 is lost in compound 6b (Figure 3B) while
both polar contacts with the hinge region were missing in the complex between Fyn and 6a.
Viceversa, the optimization of the van der Waals contacts within the hydrophobic region I of
ligands 4b, 4c, 4d, 5b and 5c, due to the introduction of an hydrophobic substituent such as F, Cl or
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CH at the para position of the C3 phenyl ring, rapidly determines an increase in the binding
3
affinity.
In order to assess the stability of the complexes and to gain further insight into the ligandꢀprotein
interactions, molecular dynamics (MD) simulations were carried out starting from the geometry
obtained by docking. The complexes between Fyn and derivatives 4a, 4b, 4c and 4d were selected
as the training set. Furthermore, the same computational protocol was applied on the complex
between Fyn and the know ligand PP2 for comparison purpose. Each simulation was carried out for
a total of 12 ns. Figure 4 shows the time evolution of the root mean square deviation (RMSD)
calculated on all heavy atoms during the time of simulation with respect to the starting structures.
For each analyzed complex we observed an initial increase due to the equilibration of the system,
followed by a stabilization of the RMSD value around 1.9 Å, 1.80 Å, 1.82 Å, 1.57 Å and 1.77 Å
with ligand 4a, 4b, 4c, 4d and PP2, respectively (RMSD average value was calculated from frame
1500 to 3000). The last 6 ns of all simulations were regarded as stable and used to extract 150
evenly spacedꢀout snapshots from each trajectory. Molecular Mechanics Poissonꢀ
Boltzmann/Generalized Born Surface Area (MMꢀPB/GBSA) approaches were then applied to
estimate the free energy of binding between protein and ligands. The computed values are reported
in Table 4. It must be noted that a better correlation between computed binding energies and
experimental free energies of binding for congeneric compounds 4aꢀd was observed for the GB
2
method for which the higher Pearson’s correlation coefficient was determined (r of 0.83 vs 0.47).
1
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Taking also into consideration that GB calculations are less timeꢀexpensive than PB calculations,
the former resulted to be the best choice for the future optimization of the pyrazolo[3,4ꢀ
d]pyrimidines as Fyn kinase inhibitors. Next, to go into the details of the interactions which play a
pivotal role in the ligandꢀprotein recognition process, a post processing analysis was conducted on
the complexes including hydrogenꢀbonds analysis and monitoring of the root mean square
fluctuations of the binding site residues.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Hꢀbond analysis. Hydrogenꢀbonding interactions between ligands and receptor as well as between
the main residues of the active site were determined and listed in Table 5 together with their
occupancy in the investigated time period. A weak hydrogenꢀbond (occupancy of about 30%) was
found between the nitrogen at position 5 (N5) of each ligand and the NH backbone of Met345.
Conversely, the carbonyl backbone of Glu343 established a stable hydrogenꢀbond contact
(occupancy higher than 60%) with the ligand exocyclic amine (N4). As a consequence of this
hydrogenꢀbond, the intermolecular interaction between the Thr342 side chain and the carbonyl
backbone of Glu343 is weakened. This effect is particularly pronounced for the most active
inhibitors 4c and 4d for which the Thr342ꢀGlu343 polar contact is totally lost as well as for PP2
ligand. Furthermore, no interaction was detected between N4 and the Thr342 side chain and this
could be due to the fact that the “gatekeeper” residue is stabilized in its position by favourable
interactions with Ala297 (Table 5). On the other hand, it has been previously demonstrated that the
10,16,17,19ꢀ23
active conformation of Fyn is overexpressed in pathological conditions.
In addition, the
main interactions for the stabilization of this conformation have been identified by means of
crystallography studies. Among them, the electrostatic network between the catalytic Lys299 and
the Glu314 belonging to the αC helix stabilizes the receptor in its active conformation. Taking into
account this information, we also monitored the Lys299ꢀGlu314 hydrogenꢀbonding interactions
during the MD simulations. As a result, the saltꢀbridge was rather stable for all complexes
suggesting the ability for the studied inhibitors of interacting and stabilizing the Fyn active
1
1
ACS Paragon Plus Environment

Page 13 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
conformation. Furthermore, a larger network of interactions emerged for residues Lys299 and
Glu314 in complexes 4bꢀ4d. In detail, in such systems Lys299 is also involved in a saltꢀbridge with
the side chain of Asp408 and, at the same time, the Glu314 participates in an Hꢀbond with the NH
backbone of Phe409. Such pool of interactions further contributes to the stabilization of the
complexes in the active conformation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RMSF. Root mean square fluctuations (RMSFs) of the αꢀcarbons of each residue were calculated
during the production phase of MD simulation of the analysed complexes and results were
compared with those coming from a MD simulation of Fyn kinase in its apo form (Figure 5). It is
interesting to note that some regions of the kinase domain which directly interact with compounds
4aꢀd and PP2, showed a lower flexibility in complexes with respect to the free receptor. In
particular, this corresponds to the loop between β3 strand and αC helix (residues from 300 to 310
highlighted in yellow in Figure 5), the Aꢀloop (residues from 416 to 431 highlighted in orange in
Figure 5) and the glycine rich loop or Gꢀloop (residues from 274 to 286 in cyan blue in Figure 5).
Focusing on Gꢀloop, the RMSFs values obtained for complexes 4aꢀd were in line with the
experimental activities. Indeed, lower fluctuations were found with the most active inhibitor 4c
suggesting that the stabilization of the Gꢀloop could plays a key role in determining the potency of a
ligand against Fyn.
1
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Chemistry
The synthesis of compounds 3jꢀn was performed starting from 1ꢀ(2ꢀhydroxyꢀ2ꢀphenylethyl)ꢀ6ꢀ
38
thioxoꢀ1,5,6,7ꢀtetrahydroꢀ4Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀone 7, previously reported by us.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Alkylation of the C6 thiocarbonyl group with the suitable alkyl bromide in anhydrous N,Nꢀ
dimethylformamide (DMF) at room temperature afforded the 6ꢀalkylthio derivatives 8aꢀc, in turn
treated with the Vilsmeier complex (POCl /DMF, 1:1) in CHCl to obtain the dichloroꢀderivatives
3
3
9
aꢀc. Finally, reaction with an excess of the appropriate amines in toluene at room temperature
afforded compounds 3j, 3k, 3r and 3q in good yields. Differently, compound 3l was obtained
reacting 9b with mꢀchloroaniline in absolute ethanol at reflux for 5 h. Compounds 3m and 3s have
been obtained by treatment of 3q and 3r with a 4N NaOH solution at reflux for 5 h (Scheme 1).
Oxidation of compound 3s with metaꢀchloroperoxybenzoic acid (mCPBA) in CHCl gave the 6ꢀ
3
methylsulfonyl derivative 10. Finally, 3n was obtained by nucleophilic substitution of the
methylsulfonyl group with 2ꢀaminoethanol in dimethylsulfoxide (DMSO) and butanꢀ1ꢀol at 90 °C
for 12 h in good yield (Scheme 2). Compound 3o was obtained starting from 2ꢀhydrazinoꢀ1ꢀ
39
40
phenylethanol 11. The synthesis of 12 and 13 was previously reported by us, but, in this study
compound 12 was prepared through an alternative method that uses microwave irradiation, with the
purpose of shortening the reaction times. 5ꢀAminoꢀ1ꢀ(2ꢀhydroxyꢀ2ꢀphenylethyl)ꢀ1Hꢀpyrazoleꢀ4ꢀ
carboxamide 13 was treated with sodium ethoxide and methyl phenylacetate in absolute ethanol at
reflux for 6 h to afford 6ꢀbenzylꢀ1ꢀ(2ꢀhydroxyꢀ2ꢀphenylethyl)ꢀ1,5ꢀdihydroꢀ4Hꢀpyrazolo[3,4ꢀ
d]pyrimidinꢀ4ꢀone 14. This last was in turn chlorinated with the Vilsmeier complex in CHCl
reflux for 12 h and gave the dichloro derivative 15, that by reaction with benzylamine, afforded 3o
Scheme 3). The synthesis of 3ꢀsusbtituted pyrazolo[3,4ꢀd]pyrimidines 4aꢀe was performed using a
3
at
(
41
three components oneꢀpot synthesis. Sodium hydride was added in small batches to a solution of
malononitrile 16 in dry THF precooled at 0/5 °C; after 30 minutes the suitable acyl chloride was
1
3
ACS Paragon Plus Environment

Page 15 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
added and the solution stirred at room temperature for 2ꢀ12 h. Then dimethylsulfate was added and
the solution was refluxed for 3ꢀ6 h. Finally, 2ꢀhydrazinoꢀ1ꢀphenylethanol 11 dissolved in dry THF
was added and the reaction was refluxed for 4 h to afford intermediates 17aꢀe, purified by flash
chromatography. Compounds 17aꢀe were suspended in formamide and the mixture was heated at
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
90 °C for 3ꢀ4 h to afford the pyrazoloꢀpyrimidines 18aꢀe, that were in turn reacted with thionyl
chloride in dry CH Cl at room temperature for 12 h under nitrogen atmosphere to give the final
2
2
41
compounds 4aꢀe (Scheme 4). We undertook a study, similar to that reported by Hanefeld et al., to
compare the stepꢀbyꢀstep synthesis to the oneꢀpot procedure. This study was aimed at evaluating if
the oneꢀpot reaction was convenient in terms of time, cost and green impact also for the synthesis of
our derivatives bearing a N1 phenyletanolic chain. Compound 17d, that was obtained with good
yield in the oneꢀpot synthesis, was selected for this study. In the stepꢀbyꢀstep method, intermediates
41
1
9
and
20
were
synthesized
and
isolated.
Finally,
[methoxy(4
methylphenyl)methylene]malononitrile 20 was reacted with 2ꢀhydrazinoꢀ1ꢀphenylethanol 11 in
absolute ethanol at reflux for 8 h, giving the desired compound 17d (Scheme 5). The yield of 17d
increases from 13% in the three steps procedure (Scheme 5A) to 42% in the one step procedure
(Scheme 5B). This allows to save time and solvents. Synthesis of compounds 5aꢀe was performed
via Suzuki crossꢀcoupling, since the oneꢀpot reaction previously described led to very low yields. 5ꢀ
4
2
Aminoꢀ1Hꢀpyrazoloꢀ4ꢀcarbonitrile 21, obtained by reaction of (ethoxymethylene)malononitrile
with hydrazine monohydrate, was cyclized by reaction with formamide at 200 °C for 1 h, affording
4
2
1
Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀamine 22. Reaction of 22 with Nꢀiodosuccinimide (NIS) in dry
DMF at 80 °C for 14 h under nitrogen atmosphere gave 3ꢀiodoꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀ
43
amine 23. This last was in turn treated with K CO and 1ꢀbromoꢀ2ꢀphenylpropane at 130 °C for 18
2
3
h to afford 3ꢀiodoꢀ1ꢀ(2ꢀphenylpropyl)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀamine 24 in good yield.
Finally, compound 24 was reacted with an excess of the suitable boronic acid in the presence of
Cs
2
CO
3
and PdCl
2
(dppf) in dry toluene at 90 °C for 14 h to give compounds 5aꢀe (Scheme 6).
1
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Compounds 6a and 6b, characterized by a phenylethyl and 2ꢀmethoxyꢀ2ꢀphenylethyl chain on the
N1 position and bearing a Sꢀmethyl moiety on C6, have been synthesized following the procedure
37
recently reported by us (Scheme 7). Starting from thiobarbituric acid 25, firstly the thiol moiety
3
was methylated to afford the intermediate 26 that was chlorinated with POCl giving compound 27.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The 4,6ꢀdichloroꢀ2ꢀ(methylthio)pyrimidine 27 was reacted with LDA and benzaldehyde to give the
alcohol 28, subsequently oxidized to the ketone 29. The latter was cyclized to afford the
pyrazole[3,4ꢀd]pyrimidine core (30) by using hydrazine monohydrate. Mitsunobu reaction with 2ꢀ
phenylethanol or 2ꢀmethoxyꢀ2ꢀphenylethanol was performed in order to introduce the N1
substituent, leading to compounds 31a and 31b. These were reacted with Nꢀmethylamine or
ammonia to afford the final compounds 6a and 6b, respectively.
1
5
ACS Paragon Plus Environment

Page 17 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Biology
Selectivity against a human kinase panel. To assess its specificity against Fyn, compound 4c was
tested against a panel of kinases including other Src family members (Hck, Blk, Fgr, Fyn, Src, Lck,
Lyn and Yes), tyrosine kinases (Abl, EGFR, IGF1R, JAK2, PDGFR, KDR), as well as some serineꢀ
threonine kinases. The percentage of enzymatic activity was measured for each kinase by using 4c
at 10 µM (Millipore); results are listed in Table 7. Remarkably, 4c proved to be more efficient
against Src family members than toward the other investigated kinases, confirming such a
compound as a useful probe to study Src family function. Furthermore, an high activity against Abl
was also detected as expected for the high structural similarity between Abl and the SFKs. Notably,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
c was not able to significantly inhibit any of the serineꢀthreonine kinases tested (Pimꢀ1, mTOR,
JNK, CDK5, Chl1), resulting not active also towards the serineꢀthreonine kinases DYRK1a and
4
4,45
GSK3ꢀβ, that were reported as implicated in AD.
Biological Evaluation in an Alzheimer’s disease model. In AD Fyn mediates the phosphorylation
of Tau on the Tyr18 residue, an early and crucial step in the disease progression, and is therefore
12
considered a promising therapeutic target. For this reason, the most interesting compounds
identified during in vitro inhibition assays, 4c and 4d, were evaluated for their ability to inhibit the
Fyn mediated phosphorylation of residue Tyr18 in Tau protein in an AD model cell line. To this
aim, neuroblastoma SHꢀSY5Y cells were differentiated to mature neurons with the administration
of retinoic acid, followed by brain derived neurotrophic factor, neuregulin β1, nerve growth factor,
4
6
and vitamin D3 treatment. Once differentiated, SHꢀSY5Y cells were treated with amyloid beta 1ꢀ
4
7,48
4
2 (Aβ42) oligomer/protofibril in order to induce ADꢀlike neurotoxicity.
Both compounds
significantly affected Aβ42 induced Tyr18ꢀTau phosphorylation with a similar degree and in a dose
dependent manner (Figure 6). Moreover, the inhibitory activity of 4c and 4d resulted constant over
1
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
time, being effective up to six hours after compound administration (Figure 6, compare panels A, B
with panels C, D).
Anti-cancer activity. The three most active compounds in terms of enzymatic activities, 4b, 4c, and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
d, were then evaluated for their antiproliferative activity on the human CML cell line K562 (Table
3
). Several members of the Src kinase family are upregulated and/or hyperactivated in CML, and
their activity regulates proliferation and differentiation of cancer cells. In K562 cells Fyn kinase
expression is under the direct control of BcrꢀAbl oncogene and its upregulation is fundamental in
2
3
sustaining K562 proliferation. The tested compounds showed an effective antiproliferative activity
that well correlates with the K values determined by in vitro inhibition assays. The most effective
i
compounds, 4c and 4d showed promising IC50 values in the submicromolar range for cell viability.
Moreover, antitumoral effect of 4c and 4d was evaluated through cell cycle analysis (Figure 7).
K562 cells were treated with increasing concentration of each compound (0.01ꢀ1 µM) and the
percentage of cells in each phase of cell cycle was evaluated by cytofluorimetric analysis of DNA
content. Both compounds determined a significant and doseꢀdependent accumulation of cells in the
G1 phase of cell cycle starting from 0.01 µM. In parallel we observed a progressive accumulation
of hypodiploid cells indicating the presence of apoptotic cells. Notably the treatment with 0.1 µM
4
d induced the apoptosis in about 50% of treated K562 cells. Compounds 4c and 4d were also
tested on early hematopoietic progenitors (CD34+) from Ph+ CML patient who developed
5
resistance to both Imatinib mesylate and Dasatinib. Cells were seeded at 2 x 10 cells/ml density
and treated with Imatinib 1 ꢁM, and 4cꢀ4d at 2 ꢁM concentration for 48h. The percentage of
apoptotic CD34+ after Imatinib treatment was found equal to control, confirming drug resistance.
By contrast, 4c and 4d increased apoptotic levels to 35% and 30% respectively (Figure 8). It is
important to note that these compounds, when tested in human normal fibroblasts, did not show any
sign of cell toxicity. In the same way, the viability of differentiated neuron, in the absence of Aβ42
treatment, resulted not affected by the administration of 4c and 4d (data not shown). On the other
1
7
ACS Paragon Plus Environment

Page 19 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
hand, we also evaluated our compound on two solid human tumor cell lines, MDAꢀMDꢀ231
(human breast cancer cell line) and U87 (human glioblastoma multiforme cell line), that have
previously described as dependent on SFKs for their proliferation. Cell lines were treated with
different dilutions of 4c and 4d and cell viability and proliferation were measured considering realꢀ
time impedance generated by cells growing in electric plates. Treated cells showed similar response
profiles with significant difference in cell growth starting from 20 h after treatment respect to
control cells and with a more evident inhibition of cell viability from 70 h after treatment (Figure 9).
U87 cells resulted particularly responsive to SFKs inhibitors with 4c and 4d treatment that resulted
in IC50 values of 0.074 µM and 0.78 µM, respectively. The tested 4c and 4d compounds also
showed an effective antiproliferative activity on the MDAꢀMDꢀ231 cell line with IC50s in the low
micromolar range (3.7 µM and 8.3 µM for compounds 4c and 4d, respectively).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In Vitro ADME Study
Compounds 4aꢀe were profiled in vitro for aqueous solubility, liver microsomal stability and
membrane permeability (Table 8). Their aqueous solubility (ꢀlogS ranging from 7.26 to 8.25) was
very weak and constituted a deleterious parameter in terms of suitable PK profile. However, passive
membrane permeability in a PAMPA assay indicated a good cell permeability for compounds 4aꢀe
ꢀ6
(ranging from 9.9 to 17.5 X 10 cm/s). Similarly, all of these compounds were found to cross the
blood brain barrier (BBB) in a specific PAMPAꢀBBB assay with good permeability values (ranging
ꢀ6
from 8.4 to 14.5 x 10 cm/s). Moreover, stability tests disclosed that all analogues showed good
(from 78.9 to 95.6%) metabolic stability in liver microsomes. Metabolite identification by LCꢀMS
analysis after incubation in human microsomes indicated the primary metabolites were actually the
products of an oxidative dehalogenation (Table 9) at the chlorine atom in the N1 side chain as
49
previously reported by us for similar compounds.
1
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Conclusions
Because of its central role in AD pathogenesis and in various human cancers, Fyn kinase could be
surely considered an interesting target for therapeutic intervention. In the present work, molecular
modeling calculations and organic synthesis were combined together towards the discovery of novel
and potent Fyn inhibitors targeting the ATP binding site. A structureꢀbased virtual screening
approach was initially applied on commercial libraries and led to the identification of two hit
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
compounds which showed K values in the low micromolar range. Remarkably, one of these Fyn
i
inhibitors was characterized by an original chemical scaffold, never discovered before in the field of
kinase. The same computational strategy based on the use of docking analysis on an internal library
of dual Src/Abl kinase inhibitors allowed as to identify a 3ꢀsubstituted pyrazolo[3,4ꢀd]pyrimidine
with a Ki of 0.9 µM. An increase in the in vitro binding affinity of such compound towards Fyn was
then rapidly obtained by the synthesis of a small family of analogues. The best identified inhibitors,
i
having nanomolar K , were then tested in cell based assays where showed interesting cell growth
inhibitory activities in an antiproliferative assay carried out on the K562 cell line of CML. Selected
compounds were also found active against a CML cell line derived from a patient resistant to both
Imatinib mesylate and Dasatinib. Interesting antiproliferative activities of SFKs inhibitors reported
herein were also detected on solid tumors and in particular against glioblastoma on which the best
compound showed an IC50 of 70 nM. On the other hand, the studied compounds showed
appreciable activities in an AD model cell line as a consequence of the inhibition of the Fynꢀ
mediated phosphorylation of the Tau protein. Taken together, the ability of these compounds to
potently inhibit the SFKs also in cell based assays, in addition to their promising toxicity profile,
suggest that 3ꢀsusbtituted pyrazolo[3,4ꢀd]pyrimidines could already be considered as promising
early agents to be further investigate for the treatment of cancer and AD. Furthermore, the high
penetration through the BBB is a favorable feature for the effective use of such compounds for the
treatment
of
AD
or
glioblastoma.
1
9
ACS Paragon Plus Environment

Page 21 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
EXPERIMENTAL SECTION
Computational details
Protein preparation. The Fyn three dimensional coordinates was selected from the Protein Data
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
50
Bank (PDB entry code 2DQ7) and was prepared by means of Protein Preparation Wizard
5
1
workflow implemented in Maestro suite. In particular, the inhibitor and the water molecules were
deleted, hydrogen atoms were added, and bond orders and charges were assigned; the orientation of
hydroxyl groups on Ser, Thr and Tyr, the side chains of Asn and Gln residues, and the protonation
state of His residues were optimized. Steric clashes were relieved by performing a small number of
minimization steps, not intended to minimize the system completely. In our study, the minimization
(
OPLS force field) was stopped when the RMSD of the nonꢀhydrogen atoms reached 0.30 Å.
Ligand preparation. A series of about 300 pyrazolo[3,4ꢀd]pyrimidines was collected from previous
5
1
work and were built with the Schrodinger Maestro 9.2 graphical interface. Compounds were then
52
processed with the Schrodinger LigPrep tool to generate separate files for all possible enantiomers
and protonation states at physiological pH. OPLS_2005 was used as force field. The same
procedure was used on the reference ligands staurosporine, PP1 and PP2.
Concerning the threeꢀdimensional structure of the commercially available compounds, we retrieved
53
them from ZINC database by selecting Asinex as vendor. Compounds analysed were 703200
belonging to both Gold and Synergy collections.
Docking simulation. Docking studies were performed within the ATP binding site of Fyn kinase
54
using the software package Gold 4.1. The binding pocket was inserted into a grid box centered on
the bound ligand Staurosporine and enclosing residues lying within about 8 Å from the ligand itself.
Chemscore was chosen as the fitness function for docking calculation. The genetic algorithm
parameter settings were employed using the search efficiency set at 100%, and different number of
runs were carried out for each ligand depending on the stage of VS (for the first step of the VS of
2
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the ASINEX database the value was set to 10, while for the second step and for the VS of
pyrazolo[3,4ꢀd]pyrimidine library the value was set to 100). Finally, results differing less than 1 Å
in ligandꢀall atom RMSD were clustered together. For each inhibitor, the first ranked solution as
well as the lowest energy conformation of the most populated cluster, were selected for further
analysis. The known ligands staurosporine, PP1 and PP2 were also docked to validate the docking
protocol. The binding mode predicted for such ligands was in perfect agreement with that
experimentally determined for the same molecules within the ATP binding site of Srcꢀfamily
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
50,55ꢀ56
kinases.
Molecular dynamic simulation (MD). Molecular dynamics simulations were performed starting
from the best docking poses of compounds 4aꢀd and PP2 within the ATP binding site of Fyn kinase.
57
The parameterization of ligands was performed by means of antechamber module of Amber 11.
58
The Generalized Amber Force Field (GAFF) was used to assign the atom type for each molecule,
59
while the charges were assigned using AM1ꢀBCC method. The parameterization of the protein
was performed by means of leap module of Amber 11 using the ff99SB force field. The parameters
60
set for phosphotyrosine of the activation loop were that published by Homeyer and collegues.
Hydrogens were added at each complex which, subsequently, was solvated in a TIP3P octaedric box
system of explicit water molecules with a 10 Å buffer. An appropriate number of counterꢀions were
added to neutralize the system. The SHAKE algorithm was used to treat hydrogenꢀcontaining
6
1
bonds. Before MD simulations, two steps of energy minimization were performed to remove bad
contacts. In the first stage, the protein was kept fixed with a constraint of 500 Kcal/mol, and only
the water molecules were minimized. In the second stage, the entire system was minimized
applying a constraint of 10 Kcal/mol on the αꢀcarbon. The two minimization stages consisted of
1
5000 steps in which the first 1000 were steepest descent and the last were conjugate gradient. MD
trajectories were run using the minimized structures as a starting point. Constant volume
2
1
ACS Paragon Plus Environment

Page 23 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
simulations were performed for 500 ps, during which the temperature was raised from 0 to 300 K
62,63
using the Langevin dynamics method with a collision frequency of γ = 1.0 ps ꢀ1.
Then 1500 ps of constantꢀpressure MD simulations were performed at 300 K in three steps of 500
ps each. During these steps a decreasing harmonic force constraint of 10, 5 and 1 Kcal/mol·Å was
applied respectively. Finally, a 10 ns MD simulations without restraints were run at constant
temperature of 300 K and a constant pressure of 1 atm. During the simulations, the Particle Mesh
Ewald method was employed to calculate the longꢀrange electrostatic interactions. A 10ꢀA˚ cutoff
value was used for the nonꢀbonded interactions, and a time step of 2 fs was used for the simulations.
The module Ptraj implemented in AMBER11 was used to analyze trajectories. In particular, RMSD
and RMSFs were calculated for the αꢀcarbons of each residue on the production stage. In particular,
snapshots have been extracted for the last 6 nanosecond of the production phase of the simulation;
one snapshot has been extracted every ten frames for a total of 150 structures. During the
simulation, the hydrogen bonds between ligands and kinase domain were detected when the
acceptorꢀdonor atom distance was lower than 3 Å and the occupancy was more than 10% in the
investigated time period. The same snapshots were evaluated to evaluate the binding free energy
between the preferred poses of ligands and the protein by means of MM/GBSA and MM/PBSA
approaches. The entropy term was considered negligible and it was thus not calculated in this work
being computationally expensive.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Chemistry
Starting materials were purchased from AldrichꢀItalia (Milan, Italy). Melting points were
determined with a Büchi 530 apparatus and are uncorrected. IR spectra were measured in KBr with
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
a PerkinꢀElmer 398 spectrophotometer. H NMR spectra were recorded in a CDCl
3
solution (unless
otherwise specified) on a Varian Gemini 200 (200 MHz) instrument or using a Bruker Avance
DPX400 (Bruker Biospin, Germany). Chemical shifts are reported as δ (ppm) relative to TMS as
1
the internal standard, J in Hz. H patterns are described using the following abbreviations: s =
singlet, d = doublet, t = triplet, q = quartet, m = multiplet, and br s = broad singlet. TLC was carried
out using Merck TLC plates silica gel 60 F254. Chromatographic purifications were performed on
columns packed with Merk 60 silica gel, 23ꢀ400 mesh, for flash technique. Analyses for C, H, N
and S were within ±0.3% of the theoretical value. Mass spectra (MS) data were obtained using an
Agilent 1100 LC/MSD VL system (G1946C) with a 0.4 mL/min flow rate using a binary solvent
system of 95:5 methanol/water. UV detection was monitored at 254 nm. MS were acquired in
positive and negative modes, scanning over the mass range 50−1500. The following ion source
parameters were used: drying gas flow, 9 mL/min; nebulizer pressure, 40 psig; drying gas
temperature, 350 °C. All target compounds possessed a purity of ≥ 95% as verified by elemental
analyses by comparison with the theoretical values.
29
64
30
31
32
65
24
29
40
Compounds 3aꢀc, 3d, 3e, 3fꢀi, 10, 3p, 3q, 3s, 7, 8a, 9a, , 3r, 8c and 9c, 12 and 13,
have been already reported by us.
General procedure for the synthesis of compounds 3j and 3k. The suitable amine (4 mmol) was
added to a solution of 4ꢀchloro derivative 9b (381 mg, 1 mmol) in anhydrous toluene (5 mL) and
the mixture was stirred at room temperature for 48 h. The organic phase was washed with water (2
4
x 10 mL), dried (MgSO ), filtered, and concentrated under reduced pressure. The crude oil was
2
3
ACS Paragon Plus Environment

Page 25 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
®
purified by column chromatography (Florisil , 100ꢀ200 mesh) using diethyl ether as the eluent. The
compounds crystallized by adding a 1:1 mixture of Et O/petroleum ether (PE) (bp 40ꢀ60 °C).
2
NꢀBenzylꢀ6ꢀ(secꢀbutylthio)ꢀ1ꢀ(2ꢀchloroꢀ2ꢀphenylethyl)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀamine
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
(
3j). White solid (271 mg, 60%); mp: 112ꢀ113 °C. H NMR: δ 1.05 (t, J = 7.2 Hz, 3H, CH
2
CH
3
),
1
4
.42ꢀ1.45 (m, 3H, CHCH
3
), 1.68ꢀ1.74 and 1.80ꢀ1.85 (2m, 2H, CH
2
CH
3
), 3.81ꢀ3.84 (m, 1H, SCH),
.68ꢀ4.88 (m, 4H, CH N + CH
2
2
Ar), 5.49ꢀ5.53 (m, 1H, CHCl), 7.24ꢀ7.41 (m, 10H Ar), 7.69 (s, 1H,
ꢀ1
+
Hꢀ3). IR (cm ): 3250 (NH). MS: m/z [M+1] 453. Anal. (C H N ClS) C, H, N, S.
24
26
5
6ꢀ(Secꢀbutylthio)ꢀ1ꢀ(2ꢀchloroꢀ2ꢀphenylethyl)ꢀNꢀ(2ꢀphenylethyl)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ
1
4
ꢀamine (3k). White solid (368 mg, 79%); mp: 97ꢀ98 °C. H NMR: δ 1.03 (t, J = 7.0 Hz, 3H,
CH
2
2
CH
3
), 1.34ꢀ1.50 (m, 3H, CHCH
3
), 1.59ꢀ1.83 (m, 2H, CH
2 3
CH ), 2.91 (t, J = 6.2 Hz, 2H,
CH
Ar), 3.70ꢀ3.88 (m, 3H, SCH + CH
2
NH), 4.60ꢀ4.90 (m, 2H, CH
2
N), 5.30 (br s, 1H, NH
ꢀ
disappears with D O), 5.41ꢀ5.54 (m, 1H, CHCl), 7.04ꢀ7.41 (m, 10H Ar), 7.63 (s, 1H, Hꢀ3). IR (cm
2
1
+
)
28 5
: 3255 (NH). MS: m/z [M+1] 467. Anal. (C25H N ClS) C, H, N, S.
6
ꢀ(Secꢀbutylthio)ꢀNꢀ(3ꢀchlorophenyl)ꢀ1ꢀ(2ꢀchloroꢀ2ꢀphenylethyl)ꢀ1Hꢀpyrazolo[3,4ꢀ
d]pyrimidinꢀ4ꢀamine (3l). mꢀChloroaniline (255 mg, 2 mmol) was added to a solution of the 4ꢀ
chloro derivative 9b (381 mg, 1 mmol) in absolute ethanol (5 mL), and the mixture was refluxed for
5 h. After cooling to room temperature, the obtained solid was filtered, washed with water, and
1
recrystallized from absolute ethanol. White solid (236 mg, 50%); mp: 213ꢀ214 °C. H NMR: δ 1.02
(
t, J = 7.0 Hz, 3H, CH
m, 1H, SCH), 4.75ꢀ4.80 and 4.87ꢀ4.93 (2m, 2H, CH
H Ar), 8.30 (s, 1H, Hꢀ3), 10.34 (br s, 1H, NH disappears with D O). IR (cm ): 2933 (NH). MS:
2
CH
3
), 1.40ꢀ1.44 (m, 3H, CHCH
3 2 3
), 1.65ꢀ1.80 (m, 2H, CH CH ), 3.80ꢀ3.85
(
2
N), 5.63ꢀ5.67 (m, 1H, CHCl), 7.14ꢀ7.66 (m,
ꢀ1
9
2
+
m/z [M+1] 473. Anal. (C H N Cl S) C, H, N, S.
23
23
5
2
2
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Nꢀ[2ꢀ(3ꢀChlorophenyl)ethyl]ꢀ6ꢀ(methylthio)ꢀ1ꢀ[2ꢀphenylvinyl]ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀ
amine (3m). A solution of 4N NaOH (2 mL) was added to a suspension of Nꢀ[2ꢀ(3ꢀ
chlorophenyl)ethyl]ꢀ1ꢀ(2ꢀchloroꢀ2ꢀphenylethyl)ꢀ6ꢀ(methylthio)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀ
amine 3r (458 mg, 1 mmol) in 95% ethanol (12 mL), and the mixture was refluxed for 5 h. After
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
cooling, the solid was filtered, washed with water, and recrystallized from absolute ethanol. White
1
solid (273 mg, 65%); mp: 104ꢀ106 °C. H NMR: δ 2.64 (s, 3H, SCH ), 2.98 (t, J = 5.0 Hz, 2H,
3
CH Ar), 3.87 (q, J = 5.0 Hz, 2H, CH NH), 5.52 (br s, 1H, NH disappears with D O), 7.09ꢀ7.50 (m,
2
2
2
ꢀ1
1
0H, 9Ar + CH=), 7.92 (s, 1H, Hꢀ3), 7.96 (d, Jtrans = 16.0 Hz, 1H, CH=). IR (cm ): 3269 (NH),
+
1
663 (C=C). MS: m/z [M+1] 423. Anal. (C22
H
20
N
5
ClS), C, H, N, S.
2
ꢀ(4ꢀBenzylaminoꢀ1ꢀstyrylꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ6ꢀylamino)ꢀethanol
(3n).
Ethanolamine (180 ꢁL, 3 mmol) was added to a suspension of 10 (405 mg, 1 mmol) in butanꢀ1ꢀol
16 mL) and DMSO (4 mL), and the mixture was heated at 90 °C for 12 h. After cooling to room
(
temperature, butanꢀ1ꢀol was removed under reduced pressure; then water (20 mL) was added and
the solution was extracted with ethyl acetate (2 x 20 mL); the organic phase was washed with water
(20 mL), dried (MgSO
4
) and evaporated under reduced pressure. The obtained solid was filtered
1
and recrystalized from absolute ethanol. White solid (255 mg, 66%); mp: 148ꢀ149 °C. H NMR: δ
3
.68 (q, J = 4.8 Hz, 2H, CH
2
2 2
), 3.88 (q, J = 4.8 Hz, 2H, CH ), 4.77 (d, J = 4.6 Hz, 2H, CH Ar), 5.63
(br s, 1H, NH, disappears with D
2
O), 7.22ꢀ7.54 (m, 11H, 10Ar + CH=), 7.78 (s, 1H, Hꢀ3), 7.82 (d,
ꢀ1
+
J
trans = 17.2 Hz, 1H, CH=). IR (cm ): 3281ꢀ3025 (OH + NH), 1657 (C=C). MS: m/z [M+1] 387.
22 6
Anal. (C22H N
O) C, H, N.
N,6ꢀDibenzylꢀ1ꢀ(2ꢀchloroꢀ2ꢀphenylethyl)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀamine
(3o).
Benzylamine (440 ꢁL, 4 mmol) was added to a solution of 4ꢀchloro derivative 15 (383 mg, 1 mmol)
in anhydrous toluene (5 mL) and the mixture was stirred at room temperature for 48 h. The organic
4
phase was washed with water (2 x 10 mL), dried (MgSO ), filtered, and concentrated under reduced
2
5
ACS Paragon Plus Environment

Page 27 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
pressure. The crude oil was crystallized by adding a 1:1 mixture of Et O/PE (bp 40ꢀ60 °C) to give
2
1
3
o. White solid (250 mg, 55%); mp: 125 °C. H NMR: δ 4.00 (s, 2H, CH Ar), 4.53ꢀ4.96 (m, 4H,
2
ꢀ1
CH
2
N + NHCH
2
Ar), 5.40ꢀ5.54 (m, 1H, CHCl), 7.00ꢀ7.73 (m, 15H Ar), 8.04 (s, 1H, Hꢀ3). IR (cm ):
+
3
255 (NH) MS: m/z [M+1] 455. Anal.(C27
H
24
5
N Cl) C, H, N.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
General procedure for the synthesis of 4aꢀe. SOCl
solution of the suitable intermediate 18aꢀe (0.5 mmol) in dry CH
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 CH Cl (2 x 5 mL). Then
the organic phase was washed with water (5 mL), brine (5 mL), dried (Na SO ) and concentrated
under reduced pressure. Compounds 4bꢀe were obtained as white solids adding a 1:1 mixture of
Et O/PE (bp 40ꢀ60 °C). Compound 4a, that resulted a yellow oil, was precipitated as hydrochloride
salt, by adding a saturated solution of HCl in dry Et O.
2
(80 ꢁL, 1.1 mmol) was added dropwise to a
2
Cl (5 mL) and the reaction was
2
,
2
2
2
4
2
2
1
ꢀ(2ꢀChloroꢀ2ꢀphenylethyl)ꢀ3ꢀphenylꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀamine hydrocloride (4a).
1
White solid (135 mg, 70%); mp: 129ꢀ132 °C. H NMR: δ 4.76ꢀ4.81 and 5.00ꢀ5.06 (2m, 2H, CH
2
N),
+
5
.50ꢀ5.54 (m, 1H, CHCl), 7.23ꢀ7.73 (m, 10H Ar), 9.49 (s, 1H, Hꢀ6). MS: m/z 387 [M+1] . Anal.
17 5 2
(C19H N Cl ) C, H, N.
1ꢀ(2ꢀChloroꢀ2ꢀphenylethyl)ꢀ3ꢀ(4ꢀfluorophenyl)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀamine
(4b).
N),
1
White solid (125 mg, 68%); mp: 203ꢀ206 °C. H NMR: δ 4.77ꢀ4.82 and 4.95ꢀ5.00 (2m, 2H, CH
2
ꢀ1
5
3
.68ꢀ5.70 (m, 1H, CHCl), 7.38ꢀ7.42 and 7.51ꢀ7.64 (m, 9H Ar), 8.22 (s, 1H, Hꢀ6). IR (cm ): 3477,
+
2 15 5
314 (NH ). MS: m/z 369 [M+1] . Anal. (C19H N ClF) C, H, N.
3
ꢀ(4ꢀChlorophenyl)ꢀ1ꢀ(2ꢀchloroꢀ2ꢀphenylethyl)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀamine
(4c).
N),
1
White solid (65 mg, 34%); mp: 150ꢀ151 °C. H NMR: δ 4.76ꢀ4.80 and 4.97ꢀ5.02 (2m, 2H, CH
2
2
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ꢀ
1
5.67ꢀ5.68 (m, 1H, CHCl), 7.34ꢀ7.36 and 7.51ꢀ7.61 (m, 9H Ar), 8.24 (s, 1H, Hꢀ6). IR (cm ): 3470,
+
3
301 (NH ). MS: m/z 385 [M+1] . Anal. (C H N Cl ) C, H, N.
2
19 15
5
2
1
ꢀ(2ꢀChloroꢀ2ꢀphenylethyl)ꢀ3ꢀ(4ꢀmethylphenyl)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀamine
(4d).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
White solid (90 mg, 49%); mp: 159ꢀ160 °C. H NMR: δ 2.36 (s, 3H, CH
3
), 4.75ꢀ4.80 and 4.95ꢀ5.01
(
2m, 2H, CH
2
N), 5.66ꢀ5.70 (m, 1H, CHCl), 7.24ꢀ7.39 and 7.50ꢀ7.63 (m, 9H Ar), 8.23 (s, 1H, Hꢀ6).
ꢀ1
+
IR (cm ): 3468, 3306 (NH
2
). MS: m/z 365 [M+1] . Anal. (C20
H
18
N
5
Cl) C, H, N.
1ꢀ(2ꢀChloroꢀ2ꢀphenylethyl)ꢀ3ꢀ(4ꢀmethyoxyphenyl)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀamine (4e).
1
White solid (146 mg, 77%); mp: 152ꢀ153 °C. H NMR: δ 3.89 (s, 3H, OCH
3
), 4.75ꢀ4.80 and 5.01ꢀ
5
2
.07 (2m, 2H, CH N), 5.58ꢀ5.62 (m, 1H, CHCl), 7.01ꢀ7.10 and 7.26ꢀ7.68 (m, 9H Ar), 8.36 (s, 1H,
ꢀ1
+
Hꢀ6). IR (cm ): 3470, 3305 (NH
2
). MS: m/z 381 [M+1] . Anal. (C20
H
18
5
N ClO) C, H, N.
General procedure for the synthesis of 5aꢀe. The suitable boronic acid (1.08 mmol) was added to
a suspension of 3ꢀiodoꢀ1ꢀ(2ꢀphenylpropyl)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀamine 24 (100 mg, 0.27
mmol) in dry toluene (5 mL), and the mixture was stirred at room temperature under nitrogen
atmosphere for 10 minutes. Then Cs
were added. The reaction was stirred at 90 °C for 14 h. After cooling to room temperature, water
70 mL) was added and the aqueous suspension was extracted with AcOEt (2 x 40 mL). The
organic phase was washed with water (40 mL) and brine (40 mL), dried (Na SO ) and concentrated
2 3 2
CO (350 mg, 1.07 mmol) and PdCl (dppf) (20 mg, 10% mol)
(
2
4
under reduced pressure to obtain a crude, which was purified by column chromatography (silica gel
0
2 2
.060ꢀ0.200 mm, 40 Å) using a mixture of CH Cl /MeOH (95:5) as the eluent to afford compounds
5
aꢀe.
3
ꢀPhenylꢀ1ꢀ(2ꢀphenylpropyl)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀamine (5a). White solid (24 mg,
1
2
7%); mp: 105ꢀ109 °C. H NMR: δ 1.28 (d, J = 6.8 Hz, 3H, CH
3
), 3.60ꢀ3.65 (m, 1H, CH), 4.56ꢀ
4.62 (m, 2H, CH
2
N), 5.58 (br s, 2H, NH
2
disappears with D
2
O), 7.17ꢀ7.27 and 7.46ꢀ7.67 (2m, 10H
2
7
ACS Paragon Plus Environment

Page 29 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
ꢀ
1
+
Ar), 8.40 (s, 1H, Hꢀ6). IR (cm ): 3476, 3298 (NH ). MS: m/z 330 [M+1] . Anal. (C H N ) C, H,
2
20 19
5
N.
3
ꢀ(4ꢀChlorophenyl)ꢀ1ꢀ(2ꢀphenylpropyl)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀamine (5b). Yellow
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
solid (42 mg, 43%); mp 153ꢀ154 °C. H NMR: δ 1.27 (d, J = 6.8 Hz, 3H, CH
3
), 3.46ꢀ3.62 (m, 1H,
CH), 4.50ꢀ4.60 (m, 2H, CH
2
), 5.74 (br s, 2H, NH
2
disappears with D
2
O), 7.18ꢀ7.25, 7.41ꢀ7.49 and
ꢀ1
+
7
.51ꢀ7.61 (3m, 9H Ar), 8.33 (s, 1H, Hꢀ6). IR (cm ): 3470, 3421 (NH
2
). MS: m/z 365 [M+H] . Anal.
(C H N Cl) C, H, N.
2
0
18
5
3ꢀ(4ꢀMethylphenyl)ꢀ1ꢀ(2ꢀphenylpropyl)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀamine
(5c).
Light
1
brown solid (30 mg, 32%); mp 152ꢀ155 °C. H NMR: δ 1.27 (d, J = 4.4 Hz, 3H, CHCH
3
), 2.43 (s,
3
H, CH
3 2 2
Ar), 3.47ꢀ3.66 (m, 1H, CH), 4.55ꢀ4.57 (m, 2H, CH ), 5.31 (br s, 2H, NH disappears with
ꢀ1
D
2
O), 7.18ꢀ7.20, 7.32ꢀ7.44 and 7.53ꢀ7.55 (3m, 9H Ar), 8.34 (s, 1H, Hꢀ6). IR (cm ): 3480, 3325
+
(NH ). MS: m/z 344 [M+H] . Anal. (C H N ) C, H, N.
2
21 21
5
1ꢀ{4ꢀ[4ꢀAminoꢀ1ꢀ(2ꢀphenylpropyl)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ3ꢀyl]phenyl}ethanone
(5d).
), 2.67 (s,
disappears with
1
Yellow solid (50 mg, 50%); mp 232ꢀ235 °C. H NMR: δ 1.29 (d, J = 6.8 Hz, 3H, CHCH
3
3
H, CH
3
CO), 3.58ꢀ3.65 (m, 1H, CH), 4.58ꢀ4.60 (m, 2H, CH
2
), 5.47 (br s, 2H, NH
2
ꢀ1
D O), 7.19ꢀ7ꢀ27, 7.77ꢀ7.79 and 8.11ꢀ8.13 (3m, 9H Ar), 8.37 (s, 1H, Hꢀ6). IR (cm ): 3478, 3315
2
+
(NH ). MS: m/z 372 [M+H] . Anal. (C H N O) C, H, N.
2
22 21
5
3ꢀ(1HꢀIndolꢀ5ꢀyl)ꢀ1ꢀ(2ꢀphenylpropyl)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀamine (5e). Light brown
1
solid (30 mg, 30%); mp 240ꢀ241 °C. H NMR: δ 1.26 (d, J = 7.2 Hz, 3H, CH
3
), 3.49ꢀ3.74 (m, 1H,
O), 6.65 (m, 1H, indole Hꢀ3),
.19ꢀ7.26, 7.27ꢀ7.32 and 7.49ꢀ7.56 (3m, 9H, 8H Ar and 1H, Hꢀ2 indole), 7.92 (s, 1H, NH), 8.35 (s,
2 2 2
CH), 4.56ꢀ4.59 (m, 2H, CH ), 5.46 (br s, 2H, NH disappears with D
7
1
ꢀ1
+
H, Hꢀ6). IR (cm ): 3465, 3309 (NH ). MS: m/z 369 [M+H] . Anal. (C H N ) C, H, N.
2
22 20
6
2
8
ACS Paragon Plus Environment