Structure-based discovery of novel 4,5,6-trisubstituted pyrimidines as potent covalent Bruton's tyrosine kinase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2016.05.014

A series of novel 4,5,6-trisubstituted pyrimidines were designed as potent covalent Bruton's tyrosine kinase (BTK) inhibitors based on the structure of ibrutinib by using a ring-opening strategy. Among these derivatives, compound I₁ exhibited the most potent inhibitory activity with an IC₅₀ value of 0.07 μM. The preliminary structure-activity relationship was discussed and the primary amino group at the C-4 position of pyrimidine was crucial for maintaining BTK activity. Furthermore, molecular dynamics simulations and binding free energy calculations were performed for three inhibitor-BTK complexes to determine the probable binding model, which provided a comprehensive guide for further structural modification and optimization.

Full text of DOI:10.1016/j.bmcl.2016.05.014

Accepted Manuscript
Structure-based discovery of novel 4,5,6-trisubstituted pyrimidines as potent
covalent Bruton’s tyrosine kinase inhibitors
Yi Zou, Jianhu Xiao, Zhengchao Tu, Yingyi Zhang, Kun Yao, Minghao Luo,
Ke Ding, Yihua Zhang, Yisheng Lai
PII:
S0960-894X(16)30498-X
http://dx.doi.org/10.1016/j.bmcl.2016.05.014
BMCL 23872
DOI:
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
19 February 2016
19 April 2016
4 May 2016
Please cite this article as: Zou, Y., Xiao, J., Tu, Z., Zhang, Y., Yao, K., Luo, M., Ding, K., Zhang, Y., Lai, Y.,
Structure-based discovery of novel 4,5,6-trisubstituted pyrimidines as potent covalent Bruton’s tyrosine kinase
inhibitors, Bioorganic & Medicinal Chemistry Letters (2016), doi: http://dx.doi.org/10.1016/j.bmcl.2016.05.014
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.

Structure-based discovery of novel 4,5,6-trisubstituted pyrimidines
as potent covalent Bruton’s tyrosine kinase inhibitors
Yi Zoua,‡, Jianhu Xiaoa,‡, Zhengchao Tub,‡, Yingyi Zhanga, Kun Yaoa, Minghao Luoa
Ke Dingb,*, Yihua Zhanga, Yisheng Laia,*
aState Key Laboratory of Natural Medicines, Jiangsu Key Laboratory of Drug Discovery for Metabolic
Diseases, Center of Drug Discovery, China Pharmaceutical University, Nanjing 210009, PR China
bKey Laboratory of Regenerative Biology and Institute of Chemical Biology, Guangzhou Institutes of
Biomedicine and Health, Chinese Academy of Sciences, Guangzhou 510530, PR China
ABSTRACT
A series of novel 4,5,6-trisubstituted pyrimidines were designed as potent covalent Bruton’s tyrosine
kinase (BTK) inhibitors based on the structure of ibrutinib by using a ring-opening strategy. Among these
derivatives, compound I1 exhibited the most potent inhibitory activity with an IC50 valve of 0.07 μM. The
preliminary structure-activity relationship was discussed and the primary amino group at the C-4 position
of pyrimidine was crucial for maintaining BTK activity. Furthermore, molecular dynamics simulations
and binding free energy calculations were performed for three inhibitor-BTK complexes to determine the
probable binding model, which provided a comprehensive guide for further structural modification and
optimization.
Keywords: BTK, covalent kinase inhibitor, pyrimidine, structure-activity relationship, molecular
dynamics simulation
‡
*Corresponding author. E-mail: yslai@cpu.edu.cn (Yisheng Lai); ding_ke@gibh.ac.cn (Ke Ding)
2016 Elsevier Ltd. All rights reserved.

Bruton’s tyrosine kinase (BTK), a member of the Tec family of non-receptor tyrosine kinases, plays
a vital role in the B-cell signaling pathway linking cell surface B-cell receptor (BCR) stimulation to
downstream intracellular responses. The expression and activity of BTK are critical to several key steps
in the life cycle of B-lineage cells including proliferation, development, differentiation, survival, and
apoptosis.1,2 Ample evidence has indicated that the dysregulation of BTK is closely associated with the
pathogenesis and development of various B-cell malignancies and autoimmune diseases, including
rheumatoid arthritis, systemic lupus erythematosis, multiple sclerosis, B-cell lymphomas and leukemias,
such as mantle cell lymphoma (MCL) and chronic lymphocytic leukemia (CLL).3,4 Therefore, BTK has
been considered as a potential therapeutic target for treating these diseases.
In recent years, many BTK inhibitors have been reported, which can be divided into two major
classes, reversible and covalent irreversible ones, and some of them have been investigated in clinical
trials.5 Among these, covalent BTK inhibitors are a unique class of drugs, which form a covalent bond
with a noncatalytic cysteine (Cys481) located at the rim of the ATP-binding pocket of BTK. This type of
irreversible inhibitors has distinct characteristics compared with traditional reversible one, such as rapid
onset of inhibition, greater potency, and longer duration of drug action.6 Ibrutinib is a first-in-class
irreversible inhibitor with subnanomolar against BTK (IC50 = 0.5 nM),7 and has been approved for the
treatment of MCL,8 CLL,9 and Waldenström's macroglobulinemia (WM).10 Subsequently, other fused
pyrimidine-based covalent inhibitors of BTK, such as ONO-4059,11 TM-71224,12 and acalabrutinib,13
were also quickly advanced into clinical trials for the treatment of B-cell malignancies and autoimmune
disorders (Figure 1). Inspired by the pioneering work on these fused pyrimidines, we initiated docking
simulations to investigate the interactions between ibrutinib and BTK, and found that the pyrazole
moiety of ibrutinib didn’t appear to make any visible interactions with the ATP-binding pocket directly,
which provided new direction for development of novel covalent BTK inhibitors by the structural
modification of this skeleton.

Figure 1. Representative fused pyrimidine-based covalent BTK inhibitors.
The protein structure, cocrystallized with an ibrutinib analogue B43,14 was firstly carefully
analyzed (PDB code: 3GEN). Besides, we performed docking simulations using Glide15 in Schrodinger
Suite with the default setting to investigate the interactions between ibrutinib and BTK. As predicted,
docking simulations suggested that ibrutinib interacted with the active site in a fashion similar to
compound B43 (Figure 2). Briefly, the primary amine NH2 formed two hydrogen bonds with the
gatekeeper Thr474 hydroxyl and the backbone carbonyl of Glu475, the N-3 nitrogen of the pyrimidine
ring interacted with the backbone NH of Met477 at the hinge region, and the diphenyl ether moiety
occupied the hydrophobic pocket behind the Thr474 gatekeeper residue and displayed an edge-to-face
aromatic interaction with Phe540. The sulfhydryl group of Cys481 of BTK was about 6 Å away from the
acrylamide moiety at the head region of ibrutinib, which bound covalently to a cysteine residue proximal
to the ATP-binding pocket of the kinase catalytic domain by Michael addition reaction in vivo.
Meanwhile, we speculated that the pyrazolyl group of ibrutinib might act the role of maintaining its
bioactive conformation merely as it didn’t seem to make any visible interactions with the ATP-binding
site.
Thr474
Glu475
Phe540
Met477
Cys481
Figure 2. Docking mode of ibrutinib with BTK. (A) Chemical structure of B43. (B) Superposed docking
poses of B43 (yellow) and ibrutinib (green). (C) Docking pose of ibrutinib (green) with BTK. PDB ID:

3GEN.
Based on the above analysis, we designed a series of novel pyrimidine-based BTK inhibitors by
evolved the pyrazole ring of ibrutinib to 5-carbonyl and 6-amino using a ring-opening strategy (Figure
3).16,17 We hoped that the active conformation could be retained through the formation of a pseudo-ring
which was formed by intramolecular hydrogen bonding between carbonyl and amino group. The
intramolecular hydrogen bond plays a unique role in drug design and discovery, as it could control the
conformation of the molecules.18,19 Two main factors were considered in the whole process of compound
designing: the binding mode of the designed compounds should be overlapped well to that of ibrutinib,
and their distances between Michael receptor and the sulfhydryl group of BTK Cys481 had to be similar
to each other (around 6 Å).
Figure 3. Design of 4,5,6-trisubstituted pyrimidine-based BTK inhibitors
After opening the pyrazole ring, compound I1 was firstly designed and synthesized (Figure 3). As
can be seen from Figure 4A, docking simulations of the proposed compound I1 within the ATP binding
site of BTK displayed almost the same binding mode to ibrutinib and the distance between Michael
receptor and the sulfhydryl group of Cys481 of BTK was 5.5 Å. The preparation of compounds I1-9 was
exemplified in Scheme 1. The commercially available pyrimidine-4,6-diol was subjected to Vilsmeier
reaction with POCl3 to give the aldehyde 1, followed by oxidation with NaH2PO4 and NaClO2 to obtain
the corresponding carboxylic acid 2 in quantitative yields. Subsequent compound 2 was reacted with
oxalyl chloride in anhydrous THF to give the corresponding acyl chloride which was used directly to
assemble the key intermediate 3 by utilizing an intermolecular Friedel-Crafts acylation. The
intermediates 3a-b were then subjected to nucleophilic attack by (R)-1-Boc-3-aminopiperidine and
substituted aliphatic amines to furnish compounds 5a-g, respectively. After removing the Boc-protecting
group of 5a-g, the newly formed compounds were treated with acryloyl chloride to give the target

compounds I2-6 and I8-9. Meanwhile, the intermediates 3a-b were condensed successively with ammonia
and (R)-1-Boc-3-aminopiperidine to generate compounds 8a-b, followed by removal of the
Boc-protecting group to give 9a-b as intermediates. Finally, the desired products I1 and I7 were prepared
via the reaction of 9a-b and acryloyl chloride under an atmosphere of nitrogen.
Scheme 1. General synthetic route to the target compounds I1-9. Reagents and conditions: (a) POCl3, DMF, 0
oC, 1 h; rt, 30 min; reflux, 3 h; (b) NaH2PO4, NaClO2, t-BuOH/H2O (3/1, V/V), 0 oC, 1 h; (c) (COCl)2, DMF,
anhydrous THF, rt, 4 h; (d) AlCl3, anhydrous CH2Cl2, reflux, 3 h; (e) (R)-1-Boc-3-aminopiperidine, DIPEA,
EtOH, rt, 24 h; (f) DIPEA, EtOH, reflux, 6 h; (g) TFA, anhydrous CH2Cl2, rt, 12 h; (h) Acryloyl chloride,
anhydrous CH2Cl2, rt, 2 h; (i) NH3·H2O, EtOH, rt, 24 h; (j) (R)-1-Boc-3-aminopiperidine, DIPEA, EtOH,
reflux, 72 h; (k) TFA, anhydrous CH2Cl2, rt, 12 h; (l) Acryloyl chloride, anhydrous CH2Cl2, rt, 2 h.
Visual inspection of docking result also revealed that the para position of substituted piperidine ring
of compound I1 was closer to Cys481 when compared to the meta position (6.5 Å and 6.0 Å, respectively)

(Figure 4B). We anticipated that the addition of Michael acceptor extending from the para position of this
piperidine ring would yield substituents which also could directly interact with the sulfhydryl group of
Cys481. Thus, compounds II1-8 were designed for this goal and the docking results indicated that the
distance between Michael receptor and the sulfhydryl group of BTK Cys481 was 4.7 Å (Figure 4C). The
general synthetic routes for the preparation of title compounds II1-8 were illustrated in Scheme 2 and the
procedures were similar to that described above.
Scheme 2. General synthetic route to the target compounds II1-8 Reagents and conditions: (a)
1-Boc-4-aminopiperidine, DIPEA, EtOH, rt, 24 h; (b) DIPEA, EtOH, rt, 24 h; (c) DIPEA, EtOH, reflux, 6 h;
(d) 1-Boc-4-aminopiperidine, DIPEA, EtOH, reflux, 72 h; (e) TFA, anhydrous CH2Cl2, rt, 12 h; (f) Acryloyl
chloride, anhydrous CH2Cl2, rt, 2 h.
Subsequently, compounds III1-8 with tertiary amino group at C-6 were designed to inspect whether
the whole active conformation could be remained when they cannot form the intramolecular hydrogen
bond as we predicted (Figure 4D). Modeling also exhibited a good overlay of compound III1 with
ibrutinib and the distance between Michael receptor and the sulfhydryl group of BTK Cys481 was still
within the ideal range (5.5 Å). As shown in Scheme 3, compounds 14a-h were obtained via nucleophilic
attack by substituted aliphatic amine or ammonia and piperazine in sequence, followed by the treatment
with acryloyl chloride to furnish III1-8 as the target compounds.

Scheme 3. General synthetic route to the target compounds III1-8. Reagents and conditions: (a) RNH2, IPA, 0
oC, 1 h; (b) Piperazine, EtOH, reflux, 4 h; (c) Acryloyl chloride, anhydrous CH2Cl2, rt, 2 h.
Figure 4. Docking mode of the target compounds with BTK. (A) Superimposition of the docking models of
ibrutinib (green) and compound I1 (yellow); (B) The distances from the para and meta position of substituted
piperidine ring of compound I1 to the sulfhydryl group of the targeted cysteine; (C) Superimposition of the
docking models of ibrutinib (green) and compound II1 (yellow); (D) Superimposition of the docking models
of ibrutinib (yellow) and compound III1 (pink).

The target compounds I-III were evaluated for their inhibitory activity against BTK by Z-Lyte
fluorescence resonance energy transfer method and ibrutinib was served as the reference compounds. As
shown in Table 1, compounds with a primary amino group at the C-4 position turned out to be more
effective for the BTK kinase inhibition without exception. When secondary amino, tertiary amino and
even halogeno groups were installed at the C-4 position respectively, the resulting 4,5,6-trisubstituted
pyrimidine derivatives displayed significantly lower potency against BTK kinase. These results indicated
that the amino group at the C-4 position was critical to its inhibitory potency against BTK. Meanwhile,
the linker moiety between Michael receptor and the pyrimidine scaffold had an impact on the distance to
Cys481, which also played an important role in inhibition against BTK enzyme. It was observed that
compound I1 with a 3-aminopiperidinyl linker displayed better activity compared to compounds II1
(4-aminopiperidinyl-linker) and III1 (piperazinyl-linker). It was worth mentioning that compounds III1
and III5 were nearly equivalent to compound II1 with IC50 values of 0.88 μM, 1.32 μM, and 1.07 μM,
respectively. The results indicated that the lack of intramolecular hydrogen bonding may have little effect
on the activity. This was most likely because its bioactive conformation was still maintained by the
increase of the rigidity of the entire molecule compare to compounds I-II (Figure 4D). Furthermore, the
R1 group of the target compounds extended into the hydrophobic pocket was investigated. Surprisingly,
replacement of the phenyl group in compound I1 by an ethyl group resulted in a substantial reduction in
potency (compound I7 was 25-fold less active than I1). This was consistent with modeling studies from
which it was evident that the phenyl ring sits in a hydrophobic pocket of BTK, forming an edge-to-face
interaction with Phe540. These results indicated that the diphenyl ether group was optimal for BTK
activity. Mass spectrometry experiment was further employed to assess the ability of a small molecule to
form a covalent adduct with BTK. In this method, compound I1 (molecular mass, 443.2 Da) with the
highest inhibitory activity against BTK was used to incubate with an aqueous solution of the catalytic
domain of BTK (393-657). After analyzing with a Waters LC-MS system, we found that the peak was
fully shifted compared with the mass of apo BTK kinase domain (molecular mass, 32814.8 Da), which
obviously indicated that compound I1 was a covalent inhibitor of BTK (Figure S1 in the Supplementary
Data).

Table 1. Activities of 4,5,6-trisubstituted pyrimidines I-III.
b
Compd.
I1
R1
Ph
R2
Inhibitory rate a (%)
100.51
6.66
IC50
NH2
0.07 μM
Ph
NHCH3
nd
nd
I2
Ph
NHCH2CH3
NHCH(CH3)2
N(CH3)2
18.24
14.56
11.42
15.67
52.88
10.27
0
I3
Ph
nd
I4
Ph
nd
I5
Ph
Piperidinyl
NH2
nd
I6
CH2CH3
CH2CH3
CH2CH3
Ph
1.80 μM
nd
I7
NHCH3
I8
NHCH2CH3
NH2
nd
I9
78.14
10.61
0
1.07 μM
nd
II1
Ph
NHCH3
II2
Ph
NHCH2CH3
NHCH(CH3)2
NHCH3
nd
II3
Ph
1.20
nd
II4
CH2CH3
CH2CH3
CH2CH3
CH2CH3
Ph
0
nd
II5
NHCH2CH3
NHCH(CH3)2
N(CH3)2
6.81
nd
II6
8.69
nd
II7
9.80
nd
II8
NH2
93.21
8.14
0.88 μM
nd
III1
III2
III3
III4
III5
III6
III7
III8
Ibrutinib
Ph
NHCH3
Ph
NHCH2CH3
NHCH(CH3)2
NH2
17.53
48.54
87.67
11.87
6.43
nd
Ph
nd
CH2CH3
CH2CH3
CH2CH3
CH2CH3
1.32 μM
nd
NHCH3
NHCH2CH3
NHCH(CH3)2
nd
24.04
nd
100.13
0.64 nM
a Inhibitory rate at 10 μM.
b Enzyme IC50 values are averages of two or more experiments. nd: not determined

Compared with the standard docking studies, molecular dynamics (MD) is a computer simulation
method for studying the physical movements of atoms and molecules, giving a view of the dynamical
evolution of the system.20 To investigate the effect on activity made by our ring-opening operation, three
prime-refined complexes, derived directly from the docking-based modeling of BTK and its binding with
ibrutinib, compound I1 and compound III5, were selected to conduct the molecular dynamics simulation,
respectively. Binding free energy calculation and two distance parameters analysis (the distance between
two non-hydrogen atoms that can form hydrogen bond and the distance between Michael receptor of the
inhibitor and the sulfhydryl group of BTK Cys481) will provide the detailed information about the
differences between each complex. Root-mean-square deviation (RMSD) of the backbone atoms of these
complexes and root-mean-square fluctuations (RMSF) were calculated based on MD trajectory frames.
For each BTK-inhibitor complex, after the MD simulation was finished, 10 snapshots from the
equilibrium period of the MD trajectory with one snapshot for every 100 ps were extracted and subjected
to binding free energy calculation.
The evolution of RMSD of each system during 10 ns and 15 ns MD simulations was profiled in
Figure 5. For each complex, the protein-ligand system reached the converged stage early at about 2 ns.
The backbone of receptor fluctuated around 0.25 nm, and the ligand changed in the range of 0.05-0.1 nm
during the remaining simulation time, which showed the protein-ligand system reached the equilibrated
and stable stage (Figure 5A, 5C and 5E). The macroscopic variables, including density, energy, pressure,
and temperature, also showed the protein-ligand system reached the equilibrated and stable stage (Figure
S2, S4 and S6 in the Supplementary Data). The RMSF plots were made to determine the flexibility of
residues in the binding pocket during the simulation. Little changes were found on the conserved amino
acid residues in the process of the simulation, like the hinge region (Thr474, Glu475 and Met477) and the
hydrophobic region (Phe540), indicating that the ligand binding site was also kept in a stable
conformation (Figure 5B, 5D and 5F). Residue Cys481 was kept at a moderate value and its distance to
ibrutinib (vinyl β-C) was 0.53 ± 0.07 nm, which bound covalently to ibrutinib actually. Vinyl β-C of
Compound I1 had the shortest distance to residue Cys481 (0.42 ± 0.05 nm), while compound III5 showed
the longest distance to it (0.63 ± 0.04 nm), which probably accounted for the significant difference in
inhibitory activity against BTK between them. Besides the differences made by the tendency for covalent
binding, hydrogen bond interactions with the binding pocket also played an important role. From the
results of hydrogen bond distance analysis, all three compounds formed stable hydrogen bonds with the

hinge region residues, which were considered to be the key residues for binding with the inhibitors. The
primary amino group of these compounds were strongly hydrogen-bonded with the carbonyl oxygen at
the backbone of residue Glu475 and the distance increased gradually from 0.31 ± 0.01 nm for
BTK-ibrutinib and BTK-compound I1 binding, to 0.33 ± 0.02 nm for BTK-compound III5 interaction.
The average distance between the 3-nitrogen atom on the pyrimidine skeleton of the inhibitor and the
nitrogen atom on the backbone of residue Met477 was 0.32 ± 0.01 nm for BTK-ibrutinib binding, 0.32 ±
0.01 nm for BTK-compound I1 binding and 0.31 ± 0.04 nm for BTK-compound III5 binding. Compared
to the BTK-compound I1 (0.49 ± 0.06 nm) and BTK-compound III5 complexes (0.51 ± 0.05 nm), the
BTK-ibrutinib complex (0.37 ± 0.05 nm) had the shortest distance between the nitrogen atom on the
amine group of the inhibitor to the side chain of residue Thr474, as tracked from the MD trajectories.
These results suggested that even though the hydrogen-bond interaction between the inhibitor and
residue Thr474 was relatively weaker than two other critical residues, it was of critical effect upon the
inhibitory activity against BTK (Figure 6).

Figure 5. Stability examination for MD simulations. Tracked positional root-mean square deviation (RMSD)
for backbone atoms of the binding structure of BTK binding with ibrutinib (A), compound I1 (C) and
compound III5 (E) along with MD trajectories. Atomic fluctuations (RMSF) of BTK residue Cα atoms
during MD trajectories (B) (D) (F).
B
A
C
Thr474
Thr474
Thr474
Phe540
Phe540
Glu475
Glu475
Glu475
Phe540
Met477
Met477
Met477
Cys481
Cys481
Cys481
Figure 6. Intermolecular interactions in the prime-refined BTK-inhibior binding structure. Residues from
BTK within 0.5 nm of inhibitor are labeled and shown in stick style (turquoise). The dashed line represents
the averaged distance between non-hydrogen atoms of key residues of BTK and inhibitor based on the MD
trajectory. (A) Ibrutinib is colored in green; (B) compound I1 is colored in purple; (C) compound III5 is
colored in yellow.
Binding free energy was calculated for these compounds using MM/PBSA method21,22 which was
considered to reflect the binding affinity of the ligands (Figure 7). Overall, the calculated effective
binding energies of the complex showed significant negative values, indicating the favorable
protein-ligand interaction. Ibrutinib got more favorable ΔGbind values than compounds I1 and III5,
indicating that ibrutinib got higher binding affinity with binding pocket. This was consistent with the
activity differences between these three compounds and, thus, may be valuable for future computational
design of new and potent BTK inhibitors.

400
200
0
ib ru tin ib -B T K
co m p o u n d I1 -B T K
co m p o u n d III5 -B T K
-200
-400
r
r
e
d
w
E e l
o
o
n
l
E v d
l
i
o
o
b
p
p
G
n
Δ
G
Δ
o
Δ
n
Δ
G
Δ
B in d in g E n e rg y C o n trib u tio n
Figure 7. Binding free Energy calculated using MM-PBSA. Data are shown as mean ± standard error of the
mean.
Furthermore, we utilized Cov ock, a covalent docking program developed by chr dinger,23 to
investigate the effect of covalent bonding during molecular modeling. This method consists of
conventional noncovalent docking of the prereactive species, heuristic formation of the covalent
attachment, and structural refinement of the covalently bound protein-ligand complex. We have
extensively investigated the conformational differences of ibrutinib and compound I1 between covalent
and non-covalent docking performances. The pose prediction assessment was based on the heavy atom
RMSD between the covalent docking structure and non-covalent docking structure. Our results
demonstrated that ibrutinib and compound I1 showed relatively lower RMSD values (0.61 Å and 0.53 Å,
respectively). As can be seen from Figure 8, the fragments occupied at the hinge and hydrophobic region
experienced little conformational change between covalent and non-covalent docking structure, so most
of the important interactions between inhibitor and BTK still retained after covalent docking. The
difference of RMSD value was prevailingly derived from the conformational change at the head region
of ibrutinib and compound I1, which might adjust their orientation to facilitate reaction for forming
covalent bond to BTK protein.

Figure 8. The comparison of the binding mode between the covalent (white) and non-covalent (green)
docking structure. Residues from BTK within 0.5 nm of inhibitor are labeled and shown in stick style
(turquoise). The green dotted line represents the hydrogen bonding interaction between inhibitors and
BTK. (A) ibrutinib; (B) compound I1.
In conclusion, we have designed and synthesized three kinds of novel 4,5,6-trisubstituted
pyrimidine derivatives based on the structure of ibrutinib using a ring-opening strategy in the present
work. Among them, compound I1 showed the most potent inhibitory activity with an IC50 valve of 0.07
μM and its covalent binding mechanism was proved by mass spectrometry experiment. The preliminary
structure-activity relationship was disclosed and the primary amino group at the C-4 position of
pyrimidine was crucial for maintaining its inhibitory potency against BTK. The subsequent 10 ns or 15 ns
molecular dynamic simulations of the formerly mentioned protein-ligand complexes, viz., ibrutinib,
compound I1 and compound III5 were done to probe the binding mode and the stability of predicted
protein-ligand complexes, and assess the contributions to ligand binding and activity against BTK. The
results of this study will facilitate the research and accelerate the discovery of novel BTK inhibitors with
high activity and selectivity.
Acknowledgments
This research was supported by grants from the National Natural Science Foundation of China (No.
21472244), the Natural Science Foundation of Jiangsu Province (No. BK2011626), the Six Major Talent
Peak Project of Jiangsu Province (No. 2011-YY-004), and the Fundamental Research Funds for the
Central Universities (No. 2015021).

Supplementary data
Supplementary data (chemistry methods, characterization, bioactivity test methods, molecular
docking and molecular dynamic methods) associated with this article can be found, in the online version,
at http://dx.doi.org/10.1016/j.bmcl.2016.xx.xxx.
References and notes
1. Rokosz, L. L.; Beasley, J. R.; Carroll, C. D.; Lin, T.; Zhao, J.; Appell, K. C.; Webb, M. L. Expert
Opin. Ther. Targets 2008, 12, 883.
2. Kurosaki, T.; Hikida, M. Immunol. Rev. 2009, 228, 132.
3. De Weers, M.; Verschuren, M. C. M.; Kraakman, M. E. M.; Mensink, R. G. B.; Schuurman, R. K. B.;
Dongen, J. J. M.; Hendriks, R. W. Eur. J. Immunol. 1993, 23, 3109.
4. 4Katz, F. E.; Lovering, R. C.; Bradley, L. A.; Rigley, K. P.; Brown, D.; Cotter, F.; Chessells J. M.;
Levinsky R. J.; Kinnon C. Leukemia 1994, 8, 574.
5. Akinleye, A.; Chen, Y.; Mukhi, N.; Song, Y.; Liu, D. J. Hematol. Oncol. 2013, 6, 59.
6. Singh, J.; Petter, R. C.; Baillie, T. A.; Whitty, A. Nat. Rev. Drug Discovery 2011, 10, 307.
7. Pan, Z.; Scheerens, H.; Li, S. J.; Schultz, B. E.; Sprengeler, P. A.; Burrill, L. C.; Mendonca, R. V.;
Sweeney, M. D.; Scott, K. C.; Grothaus, P. G.; Jeffery, D. A.; Spoerke, J. M.; Honigberg, L. A.;
Young, P. R.; Dalrymple, S. A.; Palmer, J. T. ChemMedChem 2007, 2, 58.
8. Byrd, J. C.; Furman, R. R.; Coutre, S. E.; Flinn, I. W.; Burger, J. A.; Blum, K. A.; Grant, B.;
Sharman, J. P.; Coleman, M.; Wierda, W. G.; Jones, J. A.; Zhao, W.; Heerema, N. A.; Johnson, A. J.;
Sukbuntherng, J.; Chang, B. Y.; Clow, F.; Hedrick, E.; Buggy, J. J.; James, D. F.; O'Brien, S. N.
Engl. J. Med. 2013, 369, 32.
9. Wang, M. L.; Simon, R.; Peter, M.; Andre, G.; Rebecca, A.; Kahl, B. S.; Wojciech, J.; Advani, R. H.;
Romaguera, J. E.; Williams, M. E. N. Engl. J. Med. 2013, 369, 507.
10. Treon, S. P.; Tripsas, C. K.; Meid, K.; Warren, D.; Varma, G.; Green, R.; Argyropoulos, K. V.;
Yang, G.; Cao, Y.; Xu, L.; Patterson, C. J.; Rodig, S.; Zehnder, J. L.; Aster, J. C.; Harris, N. L.;
Kanan, S.; Ghobrial, I.; Castillo, J. J.; Laubach, J. P.; Hunter, Z. R.; Salman, Z.; Li, J.; Cheng, M.;
Clow, F.; Graef, T.; Palomba, M. L.; Advani, R. H. N. Engl. J. Med. 2015, 372, 1430.
11. Yasuhiro, T.; Yoshizawa, T.; Birkett, J. T. P.; Kawabata, K. Blood 2013, 122, 5151.
12. Park, J.K.; Park, J.A.; Lee, Y.J.; Song, J.; Oh, J.I.; Lee, Y.-M.; Suh, K.H.; Son, J.; Lee, E.B. Ann.
Rheum. Dis. 2014, 73, 355.
13. Byrd, J. C.; Harrington, B.; O'Brien, S.; Jones, J. A.; Schuh, A.; Devereux, S.; Chaves, J.; Wierda, W.
G.; Awan, F. T.; Brown, J. R.; Hillmen, P.; Stephens, D. M.; Ghia, P.; Barrientos, J. C.; Pagel, J. M.;
Woyach, J.; Johnson, D.; Huang, J.; Wang, X.; Kaptein, A.; Lannutti, B. J.; Covey, T.; Fardis, M.;
McGreivy, J.; Hamdy, A.; Rothbaum, W.; Izumi, R.; Diacovo, T. G.; Johnson, A. J.; Furman, R. R. N.

Engl. J. Med. 2016, 374, 323.
14. Marcotte, D. J.; Liu, Y. T.; Arduini, R. M.; Hession, C. A.; Miatkowski, K.; Wildes, C. P.; Cullen, P.
F.; Hong, V.; Hopkins, B. T.; Mertsching, E.; Jenkins, T. J.; Romanowski, M. J.; Baker, D. P.; Silvian,
L. F. Protein Sci. 2010, 19, 429.
15. Glide, version 5.9, Schrödinger, LLC, New York, NY, 2013.
16. Furet, P.; Caravatti, G.; Guagnano, V.; Lang, M.; Meyer, T.; Schoepfer, J. Bioorg. Med. Chem. Lett.
2008, 18, 897.
17. Zhang, W.; Zhang, D.; Stashko, M. A.; DeRyckere, D.; Hunter, D.; Kireev, D.; Miley, M. J.;
Cummings, C.; Lee, M.; Norris-Drouin, J.; Stewart, W. M.; Sather, S.; Zhou, Y.; Kirkpatrick, G.;
Machius, M.; Janzen, W. P.; Earp, H. S.; Graham, D. K.; Frye, S. V.; Wang, X. J. Med. Chem. 2013,
56, 9683.
18. Harter, W. G.; Albrect, H.; Brady, K.; Caprathe, B.; Dunbar, J.; Gilmore, J.; Hays, S.; Kostlan, C. R.;
Lunney, B.; Walker, N. Bioorg. Med. Chem. Lett. 2004, 14, 809.
19. Furet, P.; Bold, G.; Hofmann, F.; Manley, P.; Meyer, T.; Altmann, K. H. Bioorg. Med. Chem. Lett.
2003, 13, 2967.
20. Rapaport, D. C. (1996) ''The Art of Molecular Dynamics Simulation''. ISBN 0-521-44561-2.
21. Kumari, R.; Kumar, R.; Open Source Drug Discovery Consortium, Lynn, A. J. Chem. Inf. Model.
2014, 54, 1951.
22. Hou, T.; Wang, J.; Li, Y.; Wang, W. J. Chem. Inf. Model. 2010, 51, 69.
23. Zhu, K.; Borrelli, W. K.; Greenwood, J. R.; Day, T.; Abel, R.; Farid, R. S.; Harder, E. J. Chem. Inf.
Model. 2014, 54, 1932.

Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Structure-based discovery of novel 4,5,6-trisubstituted
pyrimidines as potent covalent Bruton’s tyrosine
kinase inhibitors
Leave this area blank for abstract
info.
a
a,
a,
b,
a
a
b,*
‡
‡
‡
Yi Zou , Jianhu Xiao , Zhengchao Tu , Yingyi Zhang , Kun Yao , Minghao Luo , Ke Ding ,
Yihua Zhanga, Yisheng Laia,*
aState Key Laboratory of Natural Medicines, Jiangsu Key Laboratory of Drug Discovery for
Metabolic Diseases, Center of Drug Discovery, China Pharmaceutical University, Nanjing 210009,
PR China
bKey Laboratory of Regenerative Biology and Institute of Chemical Biology, Guangzhou Institutes
of Biomedicine and Health, Chinese Academy of Sciences, Guangzhou 510530, PR China
Fonts or abstract dimensions should not be changed or altered.