Design, synthesis, and biological evaluation of thieno[3,2-d]pyrimidine derivatives as potential simplified phosphatidylinositol 3-kinase alpha inhibitors

Article abstract of DOI:10.1111/cbdd.13425

A series of thieno[3,2-d]pyrimidine derivatives as phosphatidylinositol 3-kinase (PI3K) inhibitors was designed using the combination strategy. The synthesis and biological evaluation of the derivatives demonstrated their potent inhibition of PI3K, culminating in the discovery of 7 and 21. Determination of a co-crystal structure of 7 complexed with PI3Kα provided the structural basis for the high enzymatic activity. Furthermore, cellular investigation of compounds 7 and 21 revealed that they efficiently suppressed cancer cell lines proliferation through inhibition of intracellular PI3K/AKT/mammalian target of rapamycin pathway. The results provided potent simplified inhibitors of PI3K with a promising overall profile and a chemical series for further optimization to progress into vivo experiments.

Full text of DOI:10.1111/cbdd.13425

PROFESSOR JIAN ZHANG (Orcid ID : 0000-0002-6558-791X)
Article type : Research Article
Design, Synthesis, and Biological Evaluation of Thieno[3,2-d]pyrimidine Derivatives as
Potential Simplified Phosphatidylinositol 3-Kinase alpha (PI3Kα) inhibitors
Xiuyan Yang a,||, Meng Deng a,||, Xi Zhang b,||, Kun Song a, Ruan Cong a, Linghua Meng b,*,
Jian Zhang a,c,*
a Department of Pathophysiology, Key Laboratory of Cell Differentiation and Apoptosis of Chinese Ministry of Education, Shanghai
JiaoTong University, School of Medicine, Shanghai 200025, China
b Division of Anti-tumor Pharmacology, State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica, Chinese Academy of
Sciences, Shanghai 201203, China
c Basic Clinical Research Center, Renji Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, 200127, China.
* Correspondence to: Jian Zhang (Phone: +86-21-63846590-776922; Fax: +86-21-64154900; Email:
jian.zhang@sjtu.edu.cn) or Linghua Meng (Email: lhmeng@simm.ac.cn)
|| X. Yang, M. Deng, and X. Zhang contributed equally to this work.
Abstract
A series of thieno[3,2-d]pyrimidine derivatives as phosphatidylinositol 3-kinase (PI3K) inhibitors was designed
using the combination strategy. The synthesis and biological evaluation of the derivatives demonstrated their potent
inhibition of PI3K, culminating in the discovery of 7 and 21. Determination of a co-crystal structure of 7 complexed
with PI3Kα provided the structural basis for the high enzymatic activity. Furthermore, cellular investigation of
compounds 7 and 21 revealed that they efficiently suppressed cancer cell lines proliferation through inhibition of
intracellular PI3K/AKT/mTOR pathway. The results provided potent simplified inhibitors of PI3K with a promising
overall profile and a chemical series for further optimization to progress into vivo experiments.
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process, which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1111/cbdd.13425
This article is protected by copyright. All rights reserved.

Keywords: PI3Kα inhibitors, thieno[3,2-d]pyrimidine derivatives, proliferation inhibition
Running head: Discovery of Potent PI3Kα inhibitors
Introduction
The phosphatidylinositol 3-kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR) signaling pathway plays
1
crucial roles in fundamental cellular processes, such as cell growth, survival, migration, and metabolism . Aberrant
activation of the PI3K/AKT/ mTOR pathway has been associated with many human cancer types 2, 3. The activation
begins with class IA PI3Ks, which can phosphorylate phosphatidylinositol 4,5-bisphosphate (PIP2) to
phosphatidylinositol 3,4,5-trisphosphate (PIP3). PIP3 recruits AKT to the membrane and then AKT is phosphorylated
at Ser473 by mTOR, which increases its kinase activity and leads to enhance cell survival and growth. This process is
negatively regulated by the well-studied tumor suppressor PTEN, which converts PIP3 back into PIP2 as a
phosphatase 4-8. As this critical role in cancer development, PI3K and mTOR have been identified as a tumor-specific
drug target. Multiple categories of small molecule inhibitors have recently been reported, including isoform-specific
PI3K inhibitor, pan-PI3K inhibitor, ATP competitive mTOR inhibitor, allosteric mTOR inhibitor and PI3K/mTOR
dual inhibitor 9-14. A number of these are currently undergoing clinical trials such as GSK2126458 (GlaxoSmithKline),
XL-765 (Exelixis), GDC-0941 (Genentech), and BEZ235 (Novartis) (Figure 1) 15-19
.
Our previously complex structure of PI3Kα-9d revealed that the amino-substituted phenol moiety of 9d induced
conformational change of the Lys802 side chain in the PI3Kα, which led to an additional space at the catalytic site for
further modification of more potent inhibitors against PI3Kα based on compound 9d 20. Meanwhile, studies on
analogues of GDC-0941 have shown that thieno[3,2-d]pyrimidine backbone could be explored as replacement for
pyrido[3׳,2׳:4,5]furo-pyrimidine to simplify the structure of PI3K inhibitors 21, 22. So we adopted the combination
principle by attaching the amino-substituted phenol moiety of 9d to a simple thieno[3,2-d] pyrimidine backbone of
GDC-0941 to discovery a series of thieno[3,2-d]pyrimidine derivatives with low molecular weight (Figure 2). We also
modified the substituents at 3-position and 5-position of the phenyl ring and further explored the structure-activity
relationship (SAR). The synthesis and biological evaluation of the derivatives demonstrated their potent inhibition of
PI3K. Of the 15 tested compounds, compound 7 showed comparable bioactivity against PI3Kα with IC50 value of 31.6
± 9.4 nM and compound 21 displayed better inhibitory potency against PI3Kα with IC50 value of 12.5 ± 2.8 nM
compared to PI103. Moreover, determination of co-crystal structure of PI3Kα-7 illustrated molecular interactions and
established the structural basis for further optimization. Finally, the effect of inhibitors 7 and 21 was measured on
This article is protected by copyright. All rights reserved.

SKOV3 and Rh30 cell lines, shown to efficiently suppress proliferation through inhibition of intracellular PI3K
phosphorylation.
Material and Method
1. Chemistry
The general synthetic routes used to prepare thieno[3,2-d]pyrimidine derivatives are outlined in the following
schemes.
In Scheme 1, commercially available 4 reacted with morpholine in methanol to produce key intermediate 5
according to a similar method reported previously 23. Then, 5 was treated with (3-hydroxy-5-nitrophenyl)boronic acid
under Suzuki conditions to provide 6 24. Reduction of 6 with iron powder and aqueous acidified ethanol gave the
corresponding aniline 7. Treatment of 7 with the different aldehyde using standard reductive amination conditions
provided the desired derivatives 8-11. To prepare the 3-fluoro derivate 13, key intermediate 5 was treated with (3-
fluoro-5-nitrophenyl)boronic acid under Suzuki conditions to provide 12. Reduction of 12 with iron powder and
aqueous acidified ethanol provided the desired compound 13.
Scheme 2 described the route used for the preparation of thieno[3,2-d]pyrimidine derivatives 20-24. Palladium
mediated treatment of 3-bromo-5-hydroxybenzaldehyde 14 with bis(pinacolato)diboron yielded 3-hydroxy-5-(4,4,5,5-
tetramethyl-1,3,2-dioxaborolan-2-yl) benzaldehyde 15. Reaction of 15 with 5 under Suzuki conditions yielded 16.
Acetylation with acetyl chloride resulted in 17. Reduction of 17 with sodium borohydride yielded 18. Halogenation of
18 with N-bromosuccinimide (NBS) yielded compound 19. Finally, 19 was converted into the desired compound 20
by alkaline hydrolysis with 1N NaOH. 3-hydroxy-5-(4-morpholinothieno[3,2-d]pyrimidin-2-yl) benzaldehyde 16,
which was a more versatile intermediate for the preparation of thieno[3,2-d]pyrimidine derivatives, was reacted with a
series of different substrates, affording the final derivatives 21-24.
2. Protein Preparation and Crystallization
To generate human wild-type full-length p110α and the nSH2 and iSH2 domains (hereafter termed “niSH2”) of
p85α that would be suitable for crystallography, a fusion protein of 6*his-TEV-p85α (318-615)-linker-p110α was
expressed in the Bac-to-Bac Baculovirus expression system (Invitrogen) using a construct containing the iSH2 domain
(467–568) of p85α fused at the N-terminus of the full-length p110α, as described previously.10 To produce a
stoichiometric complex to enhance crystallization, analogous niSH2–p110 fusion constructs were designed,
incorporating an additional linker between the iSH2 domain and the p110α in place of the thrombin cleavage site,
which is prone to non-specific proteolytic cleavage and accompanying heterogeneity. Using this method, a high purity
This article is protected by copyright. All rights reserved.

of the 6*his-TEV-p85α (318-615)-linker-p110α homogeneous fusion protein was obtained via Ni-NTA, Sepharose Q
and gel filtration 25-27. The resulting complex contained all five p110α domains [an amino-terminal adaptor-binding
domain (ABD), residues 1 to 108; a Ras-binding domain (RBD), residues 190 to 291; a C2 (protein-kinase-C
homology-2) domain, residues 330 to 480; a helical domain, residues 525 to 696; and a carboxyl-terminal kinase
domain, residues 697 to 1068], as well as the nSH2 (residues 318 to 430) and iSH2 (residues 431 to 615) domains of
p85α. Through a sparse matrix screening of 1400 conditions and seeding optimization, a diffraction-quality crystal
was produced for data collection 28, and an apo crystal was also incubated with 10 mM 7. Diffraction data to a
resolution of 2.97 Å were obtained for the 7-p110α 20
.
3. In vitro PI3K Kinase Assays
PI3K inhibitors were dissolved as stock solutions at 10 mM in 100% dimethylsulfoxide (DMSO) and stored in
aliquots at -20 °C. The compounds were diluted to the desired concentrations immediately before each experiment.
The kinase activity of the purified PI3Ks was determined by the PI3-Kinase Homogeneous Time-Resolved
Fluorescence Assay (HTRF Assay) (Millipore). The assays were performed according to the manufacturer’s protocol.
Briefly, the EC80 concentration of each enzyme was incubated in the assay buffer, containing 10 μM PIP2, in a white
384-well plate (Perkin Elmer). After incubation at room temperature for 30 min, the reaction was initiated by adding
ATP and then terminated by adding the stop solution and the detection mix. The final concentrations of ATP were 5
μM, 20 μM, 50 μM, and 20 μM for p110α, β, δ and γ, respectively. The plate was then sealed and incubated at room
temperature overnight 29. The intensity of the light emission was measured by an EnVision Multilabel Reader
(PerkinElmer) in TR-FRET mode (excitation at 320 nM and emission at 665 nM) 30. The IC50 values were calculated
by fitting data to a logistic curve using GraphPad Prism 6 software 31
.
4. Cell Proliferation Assay
The human prostate cancer PC-3 cells were obtained from the American Type Culture Collection (Manassas, VA)
and maintained in Ham’s F-12 medium (Invitrogen, Carlsbad, CA). The human ovarian cancer SKOV3 cells were
obtained from the Japanese Foundation of Cancer Research (Tokyo, Japan) and maintained in Dulbecco’s modified
Eagle’s medium (Invitrogen). The Rhabdomyosarcoma Rh30 cells were from St. Jude Children’s Hospital (Memphis,
TN) and were maintained in RPMI medium 1640 (Invitrogen) with HEPES (10 mM, pH 7.4). All the media were
supplemented with 10% FBS (Invitrogen), penicillin (100 IU/ml) and streptomycin (100 μg/ml). All cells were
cultured in a humidified atmosphere of 95% air and 5% CO2 at 37 °C 32
.
This article is protected by copyright. All rights reserved.

Cell proliferation was evaluated by a Sulforhodamine B (Sigma-Aldrich) assay as described previously 12, 33. In
brief, cells seeded in 96-well plates were treated with test compounds in triplicate for 72 h 34. The OD value was
measured at 560 nM with a multiwall spectrophotometer (VersaMax), and the IC50 values were determined by a four-
parameter logit method of the curve-fitting program from SoftMax Pro Software (version 5.4.1, Molecular Devices,
Menlo Park, CA) 20, 35
.
5. Western Blot Analysis
Cells seeded in a twelve-well plate were exposed to the test compounds at the desired concentrations for 1 h. Cells
were harvested and subjected to standard Western blot analysis using antibodies against phosphorylated AKT at
20, 35
Ser473 and total AKT (Cell signaling Technology). β-actin (Sigma-Aldrich) was used as a loading control
.
Results and Discussion
1.1. Inhibition of PI3Kα activity and SAR
All newly prepared thieno[3,2-d]pyrimidine derivatives were assessed in vitro in terms of biological activity against
the PI3Kα isoform by using the PI3-Kinase HTRF Assay. PI103 was assigned as positive controls (Table 1).
Using the combination strategy, we successfully obtained the simplified compound 7, which showed comparable
bioactivity against PI3Kα to PI103 with IC50 value of 31.6 ± 9.4 nM. With the long-chain substituted amino-group, 8,
9, 10, and 11 all showed significant loss in activity. The depression of activity became gradually significant with the
carbon chain being lengthened and was about 31-fold (IC50 = 0.99 ± 0.2 µM), 51-fold (IC50 = 1.6 ± 0.3 µM), 128-fold
(IC50 = 4.1 ± 0.8 µM) and 316-fold (IC50 > 10 µM) compared to 7. This data illustrated the long substituent in amino-
group could hardly accommodate into the induced pocket and optimization of the chain length was critical to
enzymatic activity.
Thus, we began our optimization of the amino-group from 7 with shorter substituents. Replacement of the amino
group of 7 with a (E)-hydrazonomethyl group or a (E)-carboxyvinyl group resulted in 24 (IC50 > 1 µM) or 23 (IC50 > 1
µM), which apparently reduced the enzymatic activity. In contrast, the bromomethyl and (E)-(hydroxyimino) methyl
derivatives, 20 (IC50 = 61.9 ± 1.1 nM) and 22 (IC50 = 59.2 ± 9.0 nM) regained its activity, suggesting that the induced
pocket only could accommodate smaller substituents. Further optimization with the hydroxymethyl group resulted to
the most potent compound 21, which was about 3-fold more potent than 7. However, introducing a nitro-group or a
formyl group on the phenol group yielded 6 (IC50 = 2.7 ± 0.8 µM) and 16 (IC50 > 1 µM), which was > 31-fold less
potent than 7. These data suggested again importance of the chain length and hydroxymethyl proved to be the optimal
length for enzymatic activity.
This article is protected by copyright. All rights reserved.

In a next step, we explored the influence of the hydroxyl group at the R1 position and found that the fluoro-
substitution derivative 13 (IC50 = 6.3 ± 0.1 µM) and acetyl-substituted derivative 19 (IC50 > 1 µM) showed a distinct
decrease in potency. It suggested that the hydroxyl group was critical for activity. We speculated that there was a
hydrogen bond between the hydroxyl group and its nearby residue located at the binding pocket of PI3Kα.
1.2. Crystal Structure of 7 bound to PI3Kα
To gain structural insights for the high binding affinity, we attempted to determine co-crystal structure for 7 in a
complex with PI3Kα. The data collection and refinement statistics were summarized in Table 2. As shown in Figure
3A, the crystal structure of PI3Kα-7 verified that compound 7 bound to the ATP-binding pocket of the PI3Kα kinase
catalytic domain and the conformation of 7 with PI3Kα was similar to that of the PI3Kα–9d. In detail, the morpholine
moiety of 7 was in close proximity to the hinge region of the kinase with the oxygen, forming a hydrogen bond to
amide of V851 and the carbon atoms packing against the side chain of V850 and M922 (Figure 3B). The hydrophobic
thienopyrimidine core of the compound packed against the side chain of M772, W780, I932, and I800. The phenol
moiety pointed toward a pocket formed by residues Y836, D810, I848, and I932. Two hydrogen bonds were formed
between the hydroxyl group of the phenol moiety and the Asp810 carboxyl side chain and Tyr836 hydroxyl group.
That explained why the hydroxyl was so critical for activity and precious attempts to replace the hydroxyl group on
related templates resulted in a significant loss of activity. Finally, the amino group extended toward the sidechain of
L807, D805, K802 and I848, forming a hydrogen bond to the amino of K802. Residue L807, D805, K802, and I848
formed a modest pocket, which may explain the sensitive activity to the chain length.
1.3. Selectivity of other isoforms of PI3Ks
To check selectivity for PI3Kα, compound 7, 20, 21, 22, and PI103 were evaluated against other PI3K isoforms
(Table 3). 7 was equipotent against PI3Kα and PI3Kδ, while displaying about 3-fold selectivity for PI3Kα over PI3Kβ
and 6-fold selectivity for PI3Kα over PI3Kγ. 20 showed about 7-fold selectivity for PI3Kα over PI3Kβ and 5-fold
selectivity for PI3Kα over PI3Kγ and PI3Kδ. 21 exhibited about 13-fold selectivity for PI3Kα over PI3Kγ and 4-fold
selectivity for PI3Kα over PI3Kβ and PI3Kδ. 22 showed about 18-fold selectivity for PI3Kα over PI3Kβ and 6-fold
more selectivity for PI3Kα over PI3Kγ and PI3Kδ. 7 and 21 were promising dual PI3Kα/δ inhibitors. 20 and 22 were
PI3Kα inhibitors with modest isoform selectivity. Since mTOR and PI3K conserved a typical ATP site despite their
little kinase fold resemblance, we also evaluated the inhibitory effect of our compounds on mTOR. As it turns out, 7,
20, 21, and 22 also displayed good nanomolar inhibitions against mTOR. These experiments demonstrated that
increasing the length of R2 substituents could improve the selectivity for PI3Kα but decrease the enzymatic activity of
PI3Kα.
This article is protected by copyright. All rights reserved.

1.4. Antiproliferative Activity in Selected Cell Lines
To investigate the functional consequence of inhibiting PI3Kα in cells, three cell lines with PIK3CA activating
mutations, SKOV3 ovarian cancer cell line (H1047R), Rh30 rhabdomyosarcoma cell line (high activation) and PC3
human prostate cancer cell line (PTEN deletion), were treated with 7, 20, 21, 22 and growth was monitored. All these
PI3K-activated cell lines were growth inhibited by exposure to compounds for 96h. As shown in Table 4, 7 and 21
potently inhibited the proliferation of all of the cell lines examined, irrespective of the different triggers for PI3K
activation in these cell lines, with GI50 values in the nanomolar range, which were lower than those obtained with
PI103. However, 20 and 22 were less potent than PI103 in inhibiting cell proliferation. These results suggested that, 7
and 21, represented lead compounds for further optimization of cancer therapy targeting class I PI3K.
1.5. 7 and 20 inhibits PI3K Signaling
Compounds 7 and 20 as PI3Ka inhibitors potently blocked phosphorylation of Akt in SKOV3 ovarian cancer cell
and Rh30 rhabdomyosarcoma cell (Figure 4). They also blocked activation of the mTORC1 pathway as measured by
phosphorylation of the mTORC1 target 4E-BP1. This result indicated that 7 and 20 suppressed SKOV3 cell and Rh30
cell proliferation by inhibition of intracellular PI3K/AKT/mTOR pathway.
Conclusions
A series of thieno[3,2-d]pyrimidine derivatives was designed using the combination strategy based our previous
potency compound 9d and GDC-0941. The synthesis and biological evaluation of the derivatives demonstrated their
nanomole enzymatic and cellular activities against PI3K with acceptable kinase selectivity, suggesting the successful
of our combination strategy. Further determination of complex PI3Kα-7 provided accurate structural insights for their
high-affinity interactions with PI3Kα and a structural basis for further design and optimization. The simplified
thieno[3,2-d]pyrimidine derivatives was potency PI3Kα inhibitor with low molecular weight. A particular focus will
be made on bioavailability evaluation and ADME-toxicity studies. Further optimization may ultimately yield a
chemical series to progress into vivo experiments.
Acknowledgments
This work was funded by the National Basic Research Program of China (973 Program) (2015CB910403), the
National Natural Science Foundation of China (81721004 to J.Z. and 21702137 to X.Y), and the Shanghai Sailing
Program (17YF1410600 to X.Y).
This article is protected by copyright. All rights reserved.

Conflict of Interest
The authors have declared no conflict of interest.
Supplementary Material
Experimental section: Synthetic details, characterization data for all compounds.
NMR Spectrum: 1H- and 13C-NMR spectrum of new compounds 7, 20-22.
Figure legend
Figure 1. Structures of PI3K inhibitors undergoing clinical trials.
Figure 2. Design strategy of the thieno[3,2-d]pyrimidine derivatives based on GDC-0941 and 9d.
Figure 3. (A) X-ray complex of 7 (blue, sticks) to PI3Kα (grey, surface, PDB ID: 5XGI). (B) The interactions
between 7 (blue, sticks) and PI3Kα (pink, carton). The key binding site residues of PI3Kα are shown as sticks.
Hydrogen bonds between 7 and the protein are shown as a red dashed line.
Figure 4. Inhibition of PI3K-mediated signaling by test compounds in SKOV3 and Rh30 cells. SKOV3 cells were
incubated with test compounds at the desired concentrations for 1 h. Cells were collected for Western blot to analyze
the level of phosphorylated AKT at S473 and 4E-BP1. Representative images of three independent experiments are
shown.
References
1. A. Arcaro, A. S. Guerreiro, Current Genomics 2007, 8, 271-306.
2. D. Gyori, T. Chessa, P. T. Hawkins, L. R. Stephens, Cancer 2017, 9, 24.
3. D. A. Sabbah, J. Hu, H. A. Zhong, Current Topics in Medicinal Chemistry 2016, 16, 1413-1426.
4. P. Chen, L. Zhang, J. M. Ding, J. Zhu, Y. Li, S. G. Duan, H. T. Yan, Y. W. Huan, J. H. Dong, Biochemical and
Biophysical Research Communications 2006, 342, 887-893.
5. J. A. Engelman, Cancer 2009, 9, 550-562.
6. P. Liu, H. Cheng, T. M. Roberts, J. J. Zhao, Nature Reviews Drug Discovery 2009, 8, 627-644.
7. V. Ludovini, F. Bianconi, L. Pistola, R. Chiari, V. Minotti, R. Colella, D. Giuffrida, F. R. Tofanetti, A. Siggillino,
A. Flacco, E. Baldelli, D. Iacono, M. G. Mameli, A. Cavaliere, L. Crino, Journal of Thoracic Oncology 2011, 6, 707-
615.
8. S. J. Rogers, C. Box, K. J. Harrington, C. Nutting, P. Rhys-Evans, S. A. Eccles, Expert Opinion on Therapeutic
Targets 2005, 9, 769-790.
9. W. A. Denny, Expert Opinion on Therapeutic Patents 2013, 23, 789-799.
10. Z. Zheng, S. I. Amran, J. Zhu, O. Schmidt-Kittler, K. W. Kinzler, B. Vogelstein, P. R. Shepherd, P. E. Thompson,
I. G. Jennings, The Biochemical Journal 2012, 444, 529-535.
This article is protected by copyright. All rights reserved.

11. J. R. Garlich, P. De, N. Dey, J. D. Su, X. Peng, A. Miller, R. Murali, Y. Lu, G. B. Mills, V. Kundra, H. K. Shu, Q.
Peng, D. L. Durden, Cancer Research 2008, 68, 206-215.
12. L. C. Foukas, I. M. Berenjeno, A. Gray, A. Khwaja, B. Vanhaesebroeck, Proceedings of the National Academy of
Sciences of the United States of America 2010, 107, 11381-11386.
13. C. E. Edling, F. Selvaggi, R. Buus, T. Maffucci, P. Di Sebastiano, H. Friess, P. Innocenti, H. M. Kocher, M.
Falasca, Clinical Cancer Research 2010, 16, 4928-4937.
14. G. A. Morales, J. R. Garlich, J. Su, X. Peng, J. Newblom, K. Weber, D. L. Durden, Journal of Medicinal
Chemistry 2013, 56, 1922-1939.
15. T. P. Heffron, L. Salphati, B. Alicke, J. Cheong, J. Dotson, K. Edgar, R. Goldsmith, S. E. Gould, L. B. Lee, J. D.
Lesnick, C. Lewis, C. Ndubaku, J. Nonomiya, A. G. Olivero, J. Pang, E. G. Plise, S. Sideris, S. Trapp, J. Wallin, L.
Wang, X. Zhang, Journal of Medicinal Chemistry 2012, 55, 8007-8020.
16. Y. Samuels, Z. Wang, A. Bardelli, N. Silliman, J. Ptak, S. Szabo, H. Yan, A. Gazdar, S. M. Powell, G. J. Riggins,
J. K. Willson, S. Markowitz, K. W. Kinzler, B. Vogelstein, V. E. Velculescu, Science 2004, 304, 554.
17. S. Velho, C. Oliveira, A. Ferreira, A. C. Ferreira, G. Suriano, J.S. Schwartz, A. Duval, F. Carneiro, J. C. Machado,
R. Hamelin, R. Seruca, European Journal of Cancer 2005, 41, 1649-1654.
18. D. A. Fruman, C. Rommel, Cancer Discovery 2011, 1, 562-572.
19. M. Hayakawa, H. Kaizawa, H. Moritomo, T. Koizumi, T. Ohishi, M. Yamano, M. Okada, M. Ohta, S. Tsukamoto,
F. I. Raynaud, P. Workman, M. D. Waterfield, P. Parker, Bioorganic & Medicinal Chemistry Letters 2007, 17, 2438-
2442.
20. Y. Zhao, X. Zhang, Y. Chen, S. Lu, Y. Peng, X. Wang, C. Guo, A. Zhou, J. Zhang, Y. Luo, Q. Shen, J. Ding, L.
Meng, J. Zhang, ACS Medicinal Chemistry Letters 2014, 5, 138-142.
21. M. Hayakawa, H. Kaizawa, H. Moritomo, T. Koizumi, T. Ohishi, M. Okada, M. Ohta, S. Tsukamoto, P. Parker, P.
Workman, M. Waterfield, Bioorganic & Medicinal Chemistry 2006, 14, 6847-6858.
22. M. Nacht, L. Qiao, M. P. Sheets, T. St Martin, M. Labenski, H. Mazdiyasni, R. Karp, Z. Zhu, P. Chaturvedi, D.
Bhavsar, D. Niu, W. Westlin, R. C. Petter, A. P. Medikonda, J. Singh, Journal of Medicinal Chemistry 2013, 56, 712-
721.
23. N. Srimongkolpithak, S. Sundriyal, F. L. Li, M. Vedadi, M. J. Fuchter, Medicinal Chemistry Communication
2014, 5, 1821-1828.
24. W. F. Zhu, C. Chen, C. Y. Sun, S. Xu, C. J. Wu, F. Lei, H. Xia, Q. D. Tu, P. W. Zheng, European Journal of
Medicinal Chemistry 2015, 93, 64-73.
25. J. Li, L. Sun, C. Xu, F. Yu, H. Zhou, Y. Zhao, J. Zhang, J. Cai, C. Mao, L. Tang, Y. Xu, J. He, Acta Biochimica et
Biophysica Sinica 2012, 44, 300-306.
26. S. Chen, T. Hu, J. Zhang, J. Chen, K. Chen, J. Ding, H. Jiang, X. Shen, The Journal of Biological Chemistry 2008,
283, 554-564.
27. S. Lu, A. Banerjee, H. Jang, J. Zhang, V. Gaponenko, R. Nussinov, The Journal of Biological Chemistry 2015,
290, 28887-28900.
28. W. Huang, G. Wang, Q. Shen, X. Liu, S. Lu, L. Geng, Z. Huang, J. Zhang, Bioinformatics 2015, 31, 2598-600.
29. C. Han, J. Zhang, L. Chen, K. Chen, X. Shen, H. Jiang, Bioorganic & Medicinal Chemistry Letters 2007, 15, 656-
662.
30. J. Li, J. Zhang, J. Chen, X. Luo, W. Zhu, J. Shen, H. Liu, X. Shen, H. Jiang, Journal of Combinatorial Chemistry
2006, 8, 326-337.
31. H. Jiang, R. Deng, X. Yang, J. Shang, S. Lu, Y. Zhao, K. Song, X. Liu, Q. Zhang, Y. Chen, Y. E. Chinn, G. Wu,
This article is protected by copyright. All rights reserved.

J. Li, G. Chen, J. Yu, J. Zhang, Nature Chemical Biology 2017, 13, 994-1001.
32. Q. Shen, F. Cheng, H. Song, W. Lu, J. Zhao, X. An, M. Liu, G. Chen, Z. Zhao, J. Zhang, The American Journal of
Human Genetics 2017, 100, 5-20.
33. C. Han, J. Zhang, L. Chen, K. Chen, X. Shen, H. Jiang, Bioorganic & Medicinal Chemistry 2007, 15, 656-662.
34. X. Hou, J. Du, J. Zhang, L. Du, H. Fang, M. Li, Journal of Chemical Information and Modeling 2013, 53, 188-200.
35. X.Y. Yng, X. Zhang, M. Huang, K. Song, X. F. Li, M. L. Meng, J. Zhang, Scientific Reports 2017, 7, 14572.
Figure 1. Structures of PI3K inhibitors undergoing clinical trials.
This article is protected by copyright. All rights reserved.

Figure 2. Design strategy of the thieno[3,2-d]pyrimidine derivatives based on GDC-0941 and 9d.
Figure 3. (A) X-ray complex of 7 (blue, sticks) to PI3Kα (grey, surface, PDB ID: 5XGI). (B) The interactions
between 7 (blue, sticks) and PI3Kα (pink, carton). The key binding site residues of PI3Kα are shown as sticks.
Hydrogen bonds between 7 and the protein are shown as a red dashed line.
This article is protected by copyright. All rights reserved.

Figure 4. Inhibition of PI3K-mediated signaling by test compounds in SKOV3 and Rh30 cells. SKOV3 cells were
incubated with test compounds at the desired concentrations for 1 h. Cells were collected for Western blot to analyze
the level of phosphorylated AKT at S473 and 4E-BP1. Representative images of three independent experiments are
shown.
Scheme 1a
a Reagents and conditions: (a) Morpholine, MeOH; (b) (3-hydroxy-5-nitrophenyl)boronic acid, Pd(PPh3)4, K2CO3; (c)
Fe, NH4Cl, EtOH; (d) Sodium triacetoxyborohydride, THF; (e) (3-fluoro-5-nitrophenyl)boronic acid, Pd(PPh3)4,
K2CO3; (f) Fe, NH4Cl, EtOH.
This article is protected by copyright. All rights reserved.

Scheme 2a
a Reagents and conditions: (a) Bis(pinacolato)diboron, Pd(dppf)Cl2, K2CO3; (b) Pd(PPh3)4, K2CO3; (c) CH3COCl,
CH2Cl2; (d) NaBH4, THF ; (e) NBS, PPh3; (f) 1N NaOH, MeOH; (g) 21: Sodium triacetoxyborohydride, THF; 22:
Hydroxylamine hydrochloride, THF; 23: Malonic acid, toluene; 24: Hydrazine hydrate, THF.
This article is protected by copyright. All rights reserved.

Table 1. Kinase Inhibition against PI3Kα
Compound
6
7
R1
OH
OH
R2
NO2
NH2
PI3Kα IC50 a (nM)
2723.5 ± 860.5
31.6 ± 9.4
8
OH
OH
OH
OH
998.6 ± 170.3
1643.5 ± 357.1
4112.5 ± 794.1
＞10,000
9
10
11
13
16
F
NH2
6301 ± 108.9
OH
CHO
＞1,000
19
OAc
CH2Br
＞1,000
20
21
OH
OH
CH2Br
CH2OH
61.9 ± 1.1
12.5 ± 2.8
22
23
OH
OH
OH
59.2 ± 9.0
> 1000
24
> 1000
PI-103
17.0 ± 2.2
a IC50 values shown are the average of least two independent experiments in duplicate with typical variations of less
than 20%.
This article is protected by copyright. All rights reserved.

Table 2. Data collection and Refinement Statistic for the Crystal Structure
PDB ID code
PI3K/7, 5XGI
Data Collection
Resolution (Å)*
Space group
Cell dimensions
a (Å)
50–2.56 (2.65–2.56)
γ (˚)
90.00
310146
P212121
Total reflections
Unique reflections
Completeness (%)*
Multiplicity*
47786
71.348
135.825
150.245
90.00
99.8 (99.7)
6.5 (6.2)
12.7(2.2)
14.4 (97.8)
b (Å)
c (Å)
Average I/σ(I)*
Rmerge (%)*
α (˚)
β (˚)
90.00
Refinement
Rwork (%)
Rfree (%)
23.9
29.0
Average B Value (Ų)
Protein Mean B Value (Ų)
Ligand Mean B Value (Ų)
Water Mean B Value (Ų)
Ramachandran Statistics
Most favored regions
77.083
77.261
49.368
55.038
RMSD in Bond Lengths (Å)
RMSD in Bond Angles (°)
Number of Atoms
Total
0.007
1.036
10818
10647
23
91.9%
8.0%
0.1%
0.0%
Protein
Additional allowed regions
Generously allowed regions
Disallowed regions
Ligand
Water
102
B factor Statistics
* Values in parentheses are for highest-resolution shell.
This article is protected by copyright. All rights reserved.

Table 3. Biochemical Selectivity of Compound 7, 20, 21, and 22 a
IC50 (nM) b
Compounds
PI3Kα
PI3Kβ
PI3Kγ
PI3Kδ
mTOR
27.1±1.4
PI-103
7
17.0±2.2
173.3±50.6
571.4±133.4
215.0±101.9
21.6±6.0
57.5±16.6
12.5±3.8
68.2±15.7
31.6±9.4
61.9±1.1
12.5±2.8
59.2±9.0
101.5±26.9
428.1±50.8
70.2±12.5
243.3±81.8
297.8±8.6
33.4±3.8
344.6±80.1
47.2±7.5
20
21
276.3±151.8
654.5±52.5
22
1082.0±295.6
365.0±105.5
a Effects of test compounds on the kinase activity of class I PI3K and mTOR were assessed as described in the
Experimental Section.
b IC50 values shown are the average of least two independent experiments in duplicate with typical variations of less
than 20%.
Table 4. The test compounds 7, 20, 21, and 22 inhibit proliferation of cancer cell a
Proliferation IC50 b (μM)
Genetic
Cell line
Cancer type
Prostate
PI103
7
20
21
22
mutation
PC-3
PTEN deletion
0.45±0.29
0.55±0.31
0.27±0.17
0.29±0.04 0.83±0.08 0.21±0.02
0.24±0.01 0.72±0.10 0.26±0.02
0.11±0.01 0.38±0.08 0.15±0.01
0.96±0.09
0.86±0.07
0.33±0.04
Rh30
Rhabdomyosarcoma High activation
Ovary cancer cells PI3KCA mutation
SKOV-3
a Cell proliferation was assessed by an SRB assay as described in the Experimental Section. b IC50 values shown are the
average ± SD of at least three independent experiments performed in triplicate.
This article is protected by copyright. All rights reserved.

Graphical Abstract
This article is protected by copyright. All rights reserved.