Degradation versus inhibition: Development of proteolysis-targeting chimeras for overcoming statin-induced compensatory upregulation of 3-hydroxy-3-methylglutaryl coenzyme a reductase

Article abstract of DOI:10.1021/acs.jmedchem.0c00339

3-Hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR) is an eight-pass transmembrane protein in the endoplasmic reticulum (ER) and a classical drug target to treat dyslipidemia. Statins including the well-known atorvastatin (Lipitor; Pfizer) have been w

Full text of DOI:10.1021/acs.jmedchem.0c00339

Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE
Article
Degradation Versus Inhibition: Development of Proteolysis-
Targeting Chimeras for Overcoming Statin-Induced Compensatory
Upregulation of 3-Hydroxy-3-methylglutaryl Coenzyme A Reductase
Mei-Xin Li, Yiqing Yang, Qiuye Zhao, Yue Wu, Lei Song, Haiyan
Yang, Ming He, Hongying Gao, Baoliang Song, Jie Luo, and Yu Rao
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c00339 • Publication Date (Web): 22 Apr 2020
Downloaded from pubs.acs.org on April 24, 2020
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 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 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.
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
Degradation Versus Inhibition: Development of Proteolysis-Targeting
Chimeras for Overcoming Statin-Induced Compensatory Upregulation of 3-
Hydroxy-3-methylglutaryl Coenzyme A Reductase
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
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
†
, #
‡, #
‡
‡
⊥
§
‡
Mei-Xin Li , Yiqing Yang , Qiuye Zhao , Yue Wu , Lei Song , Haiyan Yang , Ming He , Hongying
‡
†
†,
‡,
Gao , Bao-Liang Song , Jie Luo *, Yu Rao *
†Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, Wuhan University, Wuhan
4
30072, P.R. China.
‡MOE Key Laboratory of Protein Sciences, School of Pharmaceutical Sciences, MOE Key
Laboratory of Bioorganic Phosphorus Chemistry & Chemical Biology, Tsinghua University, Beijing
1
00084, P.R. China.
⊥State Key Laboratory of Proteomics, Beijing Proteome Research Center, National Center for
Protein Sciences PHOENIX Center, Beijing Institute of LifeOmics, Beijing, 102206, P.R. China
§School of Life Sciences, Tsinghua University, Beijing 100084, P.R. China.
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
ABSTRACT
3
-Hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR) is an eight-pass transmembrane
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
protein in the endoplasmic reticulum (ER) and a classical drug target to treat dyslipidemia. Statins
including the well-known atorvastatin (Lipitor®; Pfizer) have been widely used for the prevention and
treatment of cardiovascular disease for decades. However, statins can elicit a compensatory
upregulation of HMGCR protein and cause adverse effects including skeletal muscle damage. They
are ineffective for patients with statin intolerance. Inspired by the recently emerging proteolysis-
targeting chimeras (PROTACs), we set out to eliminate HMGCR protein using PROTAC-mediated
degradation. One PROTAC designated as P22A was found to reduce HMGCR protein level and block
cholesterol biosynthesis potently with less compensatory upregulation of HMGCR. To the best of our
knowledge, HMGCR is the first ER-localized, polytopic transmembrane protein successfully degraded
by the PROTAC technique. This finding may provide a new strategy to lower cholesterol levels and
treat the associated diseases.
1
ACS Paragon Plus Environment

Page 3 of 76
Journal of Medicinal Chemistry
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
3
-Hydroxy-3-methylglutaryl (HMG) coenzyme A (HMG-CoA) reductase (HMGCR) is the rate-
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
limiting enzyme catalyzing the conversion of HMG-CoA to mevalonate in the cholesterol biosynthetic
pathway. It is localized in the endoplasmic reticulum (ER) and composed of a membrane-spanning
domain and a catalytic domain protruding into the cytosol. HMGCR serves as the target of statins, a
class of cholesterol-lowering drugs most commonly prescribed for preventing and treating
1
-5
cardiovascular disease (CVD) . By possessing the HMG-like moieties, statins bind competitively to
the active site of the enzyme, blocking generation of mevalonate and downstream derivatives including
®
cholesterol. Among all types of statins on the market, atorvastatin (Lipitor ; Pfizer) has demonstrated
superior efficacy in lowering plasma low-density lipoprotein (LDL) cholesterol levels and reducing
6
-9
adverse cardiovascular outcomes .
Although statins can blunt cholesterol biosynthesis via inhibiting HMGCR activity, a compensatory
increase in protein expression has often been encountered in cultured cells10-13, as well as in the livers
of several animal species including humans14-22. This statin-induced HMGCR increment is not
surprising, as the reductions in cholesterol and other products of the mevalonate pathway can mitigate
normal feedback inhibition of the enzyme in a multivalent manner23-24. It, however, does limit the
maximal effectiveness of the drug, provoking more intensive therapies associated with higher risks of
side effects25-29. Skeletal muscle damage is the most alarming adverse effect of statins. In fact, about
2
7
1%-5% of patients receiving statin treatments develop myopathy . Hence, strategies that can ablate
both activity and abundance of HMGCR protein are urgently needed.
3
0
In the native process of ER-associated degradation (ERAD) , HMGCR binds to the ER membrane
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
proteins Insigs via its membrane domain. The E3 ubiquitin ligases gp78, TRC8 and RNF145 that
associate with Insigs then catalyze ubiquitination of two lysine residues in the HMGCR membrane
domain31-33. Ubiquitinated HMGCR is recognized by the AAA-ATPase, extracted across the ER
membrane, and degraded by proteasome. Chemical knockdown by small-molecule proteolysis
targeting chimeras (PROTACs, Figure 1A), heterobifunctional compounds consisting of a binding
moiety for the protein of interest, a linker and another moiety for a specific E3 ubiquitin ligase, has
gained widespread attention as a promising therapeutic approach to induce degradation of disease-
causing proteins via the ubiquitin–proteasome system (UPS)34-41. Moreover, PROTACs containing
small molecule inhibitors coupled to cereblon (CRBN)-recruiting immunomodulatory drugs might
elicit more pronounced downstream biological effects than conventional inhibitors alone that last even
after washout. Such encouraging results have spurred our interests in investigating whether PROTACs
can be effectively used to eliminate HMGCR protein and thus, inhibit cholesterol production. In
particular, we are very curious whether PROTACs can overcome statin-induced compensatory
upregulation of HMGCR.
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
In this study, we designed and synthesized a battery of PROTAC molecules where atorvastatin was
conjugated to ligands of the E3 ligase CRBN (Figure 1). We found that compound P22A was most
potent in promoting degradation of endogenous HMGCR protein in Chinese hamster ovary (CHO)
cells deficient in Insig-1 and Insig-2 (referred as SRD15 cells). P22A induced ubiquitination of
HMGCR in a CRBN-dependent manner, and P22A-mediated HMGCR degradation required formation
of the HMGCR/PROTAC/CRBN complex. Importantly, the increase in HMGCR protein induced by
P22A was much less than that by atorvastatin in human hepatocarcinomatous Huh7 cells. P22A was
also able to activate the sterol regulatory element-binding protein (SREBP) pathway and block
3
ACS Paragon Plus Environment

Page 5 of 76
Journal of Medicinal Chemistry
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
cholesterol biosynthesis. To our knowledge, HMGCR represents the first ER-localized, polytopic
transmembrane protein successfully degraded by the PROTAC technique. Our study provides a proof-
of-concept for PROTAC-induced HMGCR degradation as a new and potential strategy for treating
hypercholesterolemia and CVD.
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
RESULTS AND DISCUSSION
Design and synthesis of a HMGCR-PROTAC library. We first sought to construct a HMGCR-
targeting PROTAC library (Figure 1B). Atorvastatin, which recognizes the binding pocket of HMGCR,
was used as the moiety for HMGCR recruitment (Figure 1B). The ligands of CRBN successfully used
3
4
in previous PROTACs were examined in this work. We installed the azide-tagged polyethylene
glycol (PEG) on the amine group of pomalidomide, a ligand of CRBN, and the alkyne tags on a
position at the benzene ring of atorvastatin that is exposed to solvent (indicated by the red arrow in
Figure 1C). The azide- and alkyne- tagged intermediates were conjugated via click chemistry to
synthesize HMGCR-PROTACs (Figure 1B). Immunoblotting was conducted to evaluate the
degradative potency of each PROTAC. We found that the CRBN-based PROTACs could efficiently
induce the degradation of HMGCR, which encouraged us to prepare more CRBN-based PROTACs
and evaluate their biological activity carefully in the following study.
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
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
Figure 1. Design of a HMGCR-PROTACs library. (A) Schematic presentation of PROTAC-mediated protein
degradation. (B) Construction of PROTACs via click chemistry. (C) The binding pocket of HMGCR in complex with
atorvastatin (PDB code: IHWK).
To optimize the potency and study the structure-activity relationship, we further designed and
synthesized different HMGCR-PROTACs (Scheme 1). In brief, fluorobenzene was acylated with acyl
chloride (compound 1) by Lewis acid to produce compound 2. Compound 3 had a bromide atom
installed at the α position of carbonyl group of compound 2. After bromide atom was substituted by 4-
methyl-3-oxo-N-phenylpentanamide, the important 1,4-dicarbonyl intermediate 4 was generated.
Then the pyrrole ring was constructed from the condensation of 4 and primary amine via Paal-Knorr
pyrrole synthesis. The C-B bond in 6 was built from C-Br bond in 5 via Miyaura borylation reaction
and transformed to the C-O bond in 7 subsequently. Then the phenolic hydroxyl group in 7 was used
5
ACS Paragon Plus Environment

Page 7 of 76
Journal of Medicinal Chemistry
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
as a handle to attach the alkyne tags through substitution. Compounds 9 were collected after the
deprotection of 8 and coupled together with different azide-tagged CRBN ligands 10 using click
chemistry to afford the final HMGCR-PROTACs (Scheme 1, 11-14). The alkyne groups in compounds
14 could be further reduced to compounds 15 by catalytic hydrogenation.
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
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
Scheme 1. Synthesis of HMGCR-PROTACs.
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
Reaction conditions: (a) SOCl , DMF, DCM. (b) AlCl . (c) Br in AcOH, Br , CHCl . (d) K CO , Acetone. (e) t-
2
3
2
2
3
2
3
BuCOOH, PhMe, THF, Cyclohexane. (f) Pd(dppf) Cl .DCM, KOAc, Bis(pinacoloto)diboron, DMF. (g) H O in
2
2
2
2
H O, NaOH, THF. (h) K CO , MeCN. (i) HCl, MeOH, NaOH. (j) CuSO , sodium ascorbate, THF, H O. (h) Pd/C,
2
2
3
4
2
H , MeOH.
2
7
ACS Paragon Plus Environment

Page 9 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
Evaluation of PROTAC-induced HMGCR degradation. It is known that high sterol levels can
induce the membrane domains of HMGCR protein to bind ER-resident Insig-1 and Insig-2 proteins,
resulting in rapid degradation of the enzyme via the UPS. To exclude any interference of sterol-
regulated, Insig-mediated HMGCR degradation when screening for PROTACs, we utilized SRD15
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
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
11
cells that lack both Insig-1 and Insig-2 . Compared with the parental CHO cells, SRD15 cells
exhibited much higher levels of endogenous HMGCR protein when grown in normal culture medium
containing 5% FBS (Figure 2A, compare lanes 1 and 4), which, however, remained relatively
unchanged regardless of sterol depletion or oxysterol repletion (Figure 2A, lanes 5-6). In addition,
atorvastatin elicited a dose-dependent increase of HMGCR protein in CHO cells but not in SRD15
cells (Figure 2B), further supporting the notion that decreased protein degradation primarily accounts
for HMGCR accumulation induced by statins. Together, these results suggest that SRD15 cells are
suitable to evaluate PROTAC-mediated HMGCR degradation in vitro.
Figure 2. Selection of SRD15 cells to examine PROTAC-mediated HMGCR degradation. (A) CHO and SRD15 cells
were either maintained in normal culture medium containing 5% FBS (lanes 1 and 4), or depleted of lipids by
incubating in the medium supplemented with 5% lipoprotein-deficient serum, 1 M atorvastatin and 50 M
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
mevalonate for 16 h (lanes 2, 3, 5 and 6). Cells were then treated without or with 25-hydroxycholesterol (25-HC) for
6 h and harvested for immunoblotting. (B) CHO and SRD15 cells were depleted of lipids for 16 h, incubated with
1
atorvastatin at the indicated concentrations for 16 h and harvested for immunoblotting. The densitometry of HMGCR
was normalized to that of actin. The normalized value in DMSO-treated group is defined as 1.
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
We then tested the degradative potency of HMGCR-PROTACs on SRD15 cells. Firstly, varying
concentrations of pomalidomide-based PROTACs modified with the PEG linkers of different lengths
were evaluated by immunoblotting (Figure 3A-3B and Figure S1 and S2). In a series of PROTACs
that covered a wide range of the linker length (Figure 3B), those with short linkers (compounds 11a-
1
1b, also named P13A and P14A) failed to induce the degradation of HMGCR (Figure S1). Extending
the linker of compound 11b with just one more atom (compound 11c, also named P15A) caused
moderate HMGCR degradation half degraded concentration (DC = 3 μM) (Figure 3B and Figure S1).
5
0
Further increasing the linker length (compounds 11d-11i, also named P16A-P21A) did not enhance
degradation potency considerably (Figure 3B and Figure S1 and S2), probably due to improper folding
or stretched conformations of the ternary complex. When the linker length reached 22 atoms
(compound 11j, also named P22A) and above (compounds 11l-11n, also named P23A-P25A),
PROTACs could induce obvious degradation of HMGCR (Figure 3B and Figure S2), suggesting that
the increased rotatable bonds might contribute to more degree of freedom and suitable conformations.
Compound 11j (P22A) with a DC value of 0.1 μM was most potent in promoting degradation of
5
0
endogenous HMGCR. However, longer linkers as in compounds 11o (also named P26A) and 11p (also
named P27A) seemed to lessen the potency in HMGCR degradation (Figure 3B and Figure S2).
9
ACS Paragon Plus Environment

Page 11 of 76
Journal of Medicinal Chemistry
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
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
Figure 3. Potency of HMGCR-PROTACs. (A) The structures of HMGCR-PROTACs. (B-F) The degradative activity
of HMGCR-PROTACs. DC values, the compound concentration leading to 50% protein loss relative to DMSO as
5
0
determined by the semi-quantitative immunoblotting, were calculated by fitting the data to a dose-response curve
using Graphpad Prism. The DC values as the means of 3 independent experiments are indicated by asterisks. n.d.,
5
0
not determined. Corresponding blots are shown in the Supporting Information: Figure S1 and S2 for (B), Figure S3
for (C), Figure S4 for (D), Figure S5 for (E), Figure S6 for (F).
10
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
Given the heteroatomic property of the oxygen atom in the PEG linker could change its conformation,
we next replaced it with the carbon atom and evaluated the degradative potency of this new series
(compounds 12a-12g, also named P13A1-P19A1) (Figure 3A and C). The carbon substitution also
confers hydrophobicity that might facilitate the membrane permeability of PROTACs. As shown in
Figure 3C and Figure S3, the homogeneous linkers markedly promoted HMGCR degradation
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
(
compare compounds 11a-11g with 12a-12g). However, the DC values of this carbon-substituted
50
series showed no improvement compared to compound P22A. Other CRBN ligands including meta-
substituted pomalidomide or alkyne-substituted lenalidomide derivatives have been found to
significantly affect the potency of PROTACs in degrading BET38-40. In our study, compounds that
harbor meta-substituted pomalidomide (compounds 13a-13f, also named P19AM-P24AM) were less
capable of inducing HMGCR degradation than P22A (Figure 3D and Figure S4). Among these, the
compound that also contains a linker of 22 atoms in length (compound 13d, also named P22AM)
showed strongest degradative potency. A possible hook effect was detected at the concentration of 3
μM (Figure S4). For the PROTACs with alkyne-substituted lenalidomide (compounds 14a-14f, also
named L13A-L18A), which is a more rigid CRBN ligand, those encompassing long linkers
(compounds 14d-14f) were superior to those with short linkers (compounds 14a-14c) in degrading
HMGCR (Figure 3E and Figure S5), probably because the rigidity of CRBN ligands confers the linkers
less flexibility so that compounds in this series are more comparable to compounds containing shorter
linkers in other series. Alkyne reduction (compounds 15a-15f, also named L13AR-L18AR) slightly
lessened the degradative potency (Figure 3F and Figure S6). We believed that, in contrast to
compounds 12a-12f, the oxygen atoms in the PEG linkers of 11a -11f and 15a-15f cause very different
11
ACS Paragon Plus Environment

Page 13 of 76
Journal of Medicinal Chemistry
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
conformations. It is noteworthy that, while the linkers apparently influence the potency of PROTACs,
the precise effects are difficult to predict owing to their large degree of freedom.
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
P22A promotes HMGCR degradation by hijacking the CRBN-E3 ubiquitin ligase complex.
P22A induced degradation of HMGCR in a dose-dependent manner, and as high as 70% of HMGCR
protein was degraded by P22A at the concentration of 1 μM (Figure 4A and Figure S2). No further
degradation was detected when higher concentrations of P22A were used (Figure S7A). To determine
whether P22A-induced HMGCR degradation occurs via the UPS, we incubated SRD15 cells with
P22A, the proteasome inhibitor MG132 and the neddylation inhibitor MLN-4924 in various
combinations. Addition of MG132 or MLN-4924 completely blocked P22A-induced HMGCR
degradation (Figure 4B-4C). HMGCR degradation was also completely reversed by atorvastatin at 0.3
μM (Figure 4D) or pomalidomide at 1 μM (Figure 4E) via competition with P22A. Notably,
atorvastatin tagged with alkyne (compounds 9a-9d, Scheme 1) failed to elicit HMGCR degradation
(Figure S7B), indicating that the E3 ligand was required for degradation. Attachment of an ethyl group
to the piperidine-2,6-dione ring of pomalidomide (compound 11k, also named P22A-Et, Figure 3B)41
impaired the pomalidomide-CRBN interaction and abolished HMGCR degradation by P22A (Figure
4
F).
12
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
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
Figure 4. P22A is an effective HMGCR degrader. (A) SRD15 cell were treated with the indicated concentrations of
P22A for 16 h and harvested for immunoblotting. Blots shown here are one of the three repeats in Figure S2. The
densitometry of HMGCR was normalized to that of actin. The normalized value in DMSO-treated group is defined
as 1. (B-C) SRD15 cell were treated with 1 µM P22A for 16 h, and then with 10 μM MG132 (B) or the indicated
concentrations of MLN-4924 (C) for 3 h. (D-F) SRD15 cell were depleted of lipids for 16 h and treated with 1 µM
P22A and atorvastatin (D) or pomalidomide (E) at the indicated concentrations, or with 1 µM P22A-Et for 16 h (F).
We next transfected SRD15 cells with T7-tagged HMGCR together with Flag-tagged CRBN or Flag-
tagged CRBNYW/AA, a mutant defective for pomalidomide binding42-43. Overexpression of wild-type
CRBN drastically accelerated P22A-induced degradation of transfected HMGCR protein (Figure 5A,
compare lanes 2 and 4), whereas of CRBNYW/AA failed to do so (Figure 5A, compare lanes 2 and 6).
Importantly, P22A did not alter Hmgcr mRNA levels at any of the concentrations tested (Figure 5B),
suggestive of a post-transcriptional modulation of HMGCR protein by P22A. Depletion of Crbn using
siRNA-mediated knockdown greatly diminished the potency of P22A in eliminating endogenous
13
ACS Paragon Plus Environment

Page 15 of 76
Journal of Medicinal Chemistry
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
HMGCR protein (Figure 5C and 5D). We next evaluated the ubiquitination level of HMGCR
immunoprecipitated from cells treated with or without P22A. High-molecular-weight smears were
detected upon addition of P22A and MG132 (Figure 5E). To directly demonstrate P22A mediates the
interaction between HMGCR and CRBN, we performed an immunoprecipitation experiment in which
HEK293T cells were co-transfected with Myc-tagged HMGCR and Flag-tagged CRBN followed by
treatments with or without P22A for 16 h. The overexpressed HMGCR protein was only co-
immunoprecipitated with CRBN in the presence of P22A (Figure 5F). Addition of atorvastatin (Figure
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
5
G) or pomalidomide (Figure 5H) blocked HMGCR immunoprecipitation by CRBN. To further
demonstrate that CRBN participates in HMGCR ubiquitination by P22A, we transfected SRD15 cells
with plasmids encoding T7-tagged HMGCR, Flag-tagged CRBN and HA-tagged ubiquitin. Cells were
then incubated with or without P22A for 16 h, and the transfected HMGCR protein was
immunoprecipitated and subjected to immunoblotting with the anti-ubiquitin antibody. P22A
markedly increased ubiquitination of the transfected HMGCR protein (Figure 5I), which, however,
was competitively inhibited by atorvastatin (Figure 5J) and pomalidomide (Figure 5K). Taken together,
these results suggest that P22A induces degradation of HMGCR protein via UPS.
14
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
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
Figure 5. P22A induces degradation of HMGCR via UPS. (A) SRD15 cells were transfected with the indicated
plasmids and treated with DMSO or 1 µM P22A for 16 h. Cells were harvested for immunoblotting. (B) SRD15 cells
15
ACS Paragon Plus Environment

Page 17 of 76
Journal of Medicinal Chemistry
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
were treated with 1 µM P22A at the indicated concentrations for 16 h and harvested for quantitative real-time PCR.
Data are presented as mean  s.d. (n = 3 independent experiments). (C-D) SRD15 cells were transfected with the
indicated siRNAs for 48 h and treated with 1 µM P22A for 16 h. Cells were harvested for quantitative real-time PCR
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
(C) to measure the knockdown efficiency of siRNA duplexes against Crbn and immunoblotting (D). NC, negative
control. Data are normalized to control cells and presented as mean  s.d. (n = 3 independent experiments). (E)
SRD15 cells were treated with DMSO or 1 µM P22A together with 10 µM MG132 for 4 h. Cells were then harvested
and probed with P4D1 (anti-ubiquitin) antibodies. (F-H) SRD15 cells were transfected with the indicated plasmids
and treated with 1 µM P22A, 1 µM atorvastatin or 10 µM pomalidomide together with10 µM MG132 as indicated
for 5 h. Cell lysates were immunoprecipitated with the anti-Flag coupled agarose. (I-K) SRD15 cells were transfected
with the indicated plasmids and treated with 1 µM P22A, 1 µM atorvastatin or 10 µM pomalidomide together with
10µM MG132 as indicated for 5 h. Cell lysates were immunoprecipitated with the anti-Myc coupled agarose.
P22A is capable of blocking cholesterol biosynthesis. We next sought to monitor the real-time
degradation of endogenous HMGCR in live cells. A construct encoding HMGCR tagged with GFP
was prepared and its ER targeting was confirmed by the colocalization with RFP-tagged KDEL (Lys-
Asp-Glu-Leu), a well-established ER retention motif (Figure 6A). We then established a HMGCR-
GFP stable cell line by infecting SRD15 cells with lentivirus expressing HMGCR-GFP followed by
fluorescence-activated cell sorting of GFP positive cells. The HMGCR-GFP stable cells were
incubated with either DMSO or P22A for 16 h and examined under a confocal microscope. Consistent
with the earlier immunoblotting results (Figure 4A and Figure S2), P22A completely eliminated GFP
intensity that was normally visible in DMSO-treated cells (Figure 6B). It is known that statins by
16
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
inhibiting HMGCR-mediated cholesterol biosynthesis promote proteolytic processing of SREBP44-48
,
generating the cleaved N terminus that translocates into the nucleus to activate the transcription of
target gene including HMGCR and LDL receptor (LDLR). We reasoned P22A, by inhibiting HMGCR
activity via the atorvastatin moiety while inducing HMGCR degradation via the pomalidomide moiety,
would further decrease sterol levels in the cell and thus activate the SREBP pathway in a greater extent
than atorvastatin alone did. Indeed, the compensatory increases in HMGCR protein were markedly
attenuated in Huh7 cells treated with P22A relative to those with atorvastatin (Figure 6C). We next
examined the expression of Hmgcr and Ldlr, two SREBP2 targets, in mouse primary hepatocytes
receiving side-by-side treatments of atorvastatin and P22A. P22A increased slightly, if not
significantly, the mRNA abundance of Hmgcr and Ldlr relative to atorvastatin at both 0.3 and 1 μM
(Figure 6D). In addition to P22A, we also examined the protein levels of HMGCR, SREBP2 and LDLR
in Huh7 cells treated with compounds 12b-12e, and found these PROTAC molecules showed similar
activities as P22A (Figure S8). To directly compare the effects of atorvastatin and P22A on cholesterol
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
4
biosynthesis, we pulse-labeled Huh7 cells pre-treated with P22A or atorvastatin with C-acetate. The
radiolabeled lipids were fractionated by thin-layer chromatography (TLC) and visualized using a
phosphorimager. The radioactivity of 14C-labeled cholesterol in P22A-treated cells was dose-
dependently decreased in an extent similar to that in atorvastatin-treated cells (Figure 6E). As excess
cholesterol can be converted to cholesteryl esters that store with triglycerides in the lipid droplets, we
also examined the amounts of lipid droplets in Huh7 cells treated with P22A and found no significant
changes in Oil Red O staining relative to cells exposed to DMSO (Figure S9). These results suggest
that, although de novo cholesterol synthesis is decreased by P22A, the total neutral lipids (cholesteryl
ester and triglyceride) are not altered under these conditions.
17
ACS Paragon Plus Environment

Page 19 of 76
Journal of Medicinal Chemistry
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
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
Figure 6. P22A can inhibit cholesterol biosynthesis. (A) SRD15 cells were transfected with HMGCR-GFP (green)
and RFP-KDEL (red). After 48 h, cells were fixed and counterstained with Hoechst (blue). Scale bar, 50 μm. (B)
SRD15 cells stably expressing HMGCR-GFP were treated with DMSO or 1 µM P22A for 16 h. Cells were fixed and
counterstained with Hoechst (blue). Scale bar, 50 μm. (C) Huh7 cells were treated with atorvastatin or P22A as
indicated for 16 h and harvested for immunoblotting. (D) Mouse primary hepatocytes were treated with atorvastatin
or P22A as indicated for 16 h and harvested for quantitative real-time PCR. Data are presented as mean  s.d. (n = 3
independent experiments). (E) Huh7 cells were depleted of sterols and incubated with varying concentrations of
18
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
1
4
atorvastatin or P22A for 14 h. Cells were then pulse-labeled with 1 μCi/μL C-acetate for 2 h and the radiolabeled
lipids were fractionated by TLC and visualized using a phosphorimager. The value in DMSO-treated group is defined
as 1.
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
P22A and atorvastatin intervene differently on the proteomic level. Then we performed a
proteomic study to compare the expressional differences in a broad spectrum of proteins between
inhibition by atorvastatin and degradation by P22A. SRD15 cells were treated with DMSO,
atorvastatin or P22A and whole proteome was analyzed quantitatively. The quality of acquired data
was validated by their normal distribution pattern and Pearson correlation coefficient (Figure S10A
and S10B). The triplicates of DMSO, atorvastatin and P22A groups were clustered within very
different regions in the plot of principal component analysis (PCA, Figure 7A). The ANOVA analysis
of the whole proteome revealed a total of 122 proteins with distinct expression levels that could be
further divided into 4 clusters based on their normalized abundance (Figure 7B, Cluster 1-4 and Table
S1). The HMGCR abundance was approximately doubled in the inhibition group but reduced to half
in the degradation group, consistent with the immunoblotting results (Figure 4A and Figure S11).
Another key player in cholesterol homeostasis, Niemann-Pick type C2 protein, was found to increase
following both atorvastatin and P22A treatments by proteomic and immunoblotting analysis (Figure
S11 and Table S2), further confirming the legitimacy of our proteomic data. Intriguingly, we observed
4
9
that TXNDC11, an uncharacterized disulphide reductase involved in the ERAD pathway , was
increased by 108 fold (Table S2, p = 0.17) in the inhibition group and 7 fold (Table S2, p < 0.01) in
the degradation group, suggesting it could be a factor that responds to atorvastatin stimulation. As the
19
ACS Paragon Plus Environment

Page 21 of 76
Journal of Medicinal Chemistry
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
degradation mechanism of HMGCR via the Insig-mediated ERAD pathway is very different from the
PROTAC-induced process50-52, it would be interesting to investigate the relationship between statins
and TXNDC11 in the future.
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
Figure 7. P22A and atorvastatin affect proteomic level differently. (A) The PCA plot of DMSO, atorvastatin
(
inhibition) and P22A (degradation). (B) The heatmap from ANOVA for whole proteomic data. Blue means low
abundance and red means high abundance after normalization. (C) The volcano plots of fold change for atorvastatin
inhibitor, upper panel) and P22A (degrader, lower panel) compared with DMSO.
(
CONCLUSIONS
20
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
In the process of traditional drug discovery, small molecules have been designed to affect protein
activity by binding to the active and allosteric sites of target proteins, or disrupting their protein-protein
interactions. Instead of protein activity, modulators of protein abundance could function as promising
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
therapeutic tools for complementing the deficiency of traditional drug discovery53-54. RNA interference
57
(
RNAi)55-56 and CRISPR-Cas9 are two representative techniques in this respect. However, RNA-
based therapeutics are limited by RNA instability, immunogenicity, and the requirement for a vehicle
for efficient transport into the target cells, whereas CRISPR-Cas9 remains untested in clinic yet56-57
.
By contrast, PROTACs that combine the pharmacokinetic properties of small molecules and the
advantages of genetic knockdown techniques could have more promising opportunities in the future,
especially for the anti-cancer field36,58
.
Dyslipidemia is a well-established risk factor for CVD characterized by high levels of cholesterol2,59
,
and is prevalent in both developed and developing countries. Atorvastatin under the trade name Lipitor
60
was the world's best-selling drug with 130 billion dollars in sales from 1996 to 2012 . However, few
options are available for patients with statin intolerance, CVD after maximal dosage of statin, or severe
6
1
dyslipidemia . In particular, statin-induced compensatory upregulation of HMGCR unavoidably
limits their utilization, urging for the development of additional therapy strategies. The proprotein
6
2
convertase subtilisin-kexin type 9 (PCSK9) monoclonal antibodies are a milestone in the field . Small
63
interfering RNAs targeting PCSK9 have also been developed . Antibody drugs may always be
limited by their price, immunogenicity and route of administration, while the disadvantages of RNA-
based methods are discussed above. Several ubiquitin ligases have been reported to mediate
degradation of HMGCR via the UPS under physiological conditions, and this native process is valuable
for developing new therapies for diseases caused by high cholesterol31,45,64. Therefore, the artificial
21
ACS Paragon Plus Environment

Page 23 of 76
Journal of Medicinal Chemistry
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
atorvastatin-pomalidomide conjugates such as HMGCR-PROTACs may be applicable for the
treatment of dyslipidemia treatment, CVD prevention and other diseases.
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
In comparison with the native Insig-mediated HMGCR degradation pathway, PROTACs utilize
different E3 ligases to degrade HMGCR and are able to induce degradation of both endogenous and
ectopically expressed HMGCR protein. Particularly, the P22A compound has an atorvastatin moiety
and a pomalidomide moiety. On one hand, the statin group of P22A inhibits HMGCR activity,
decreases the downstream sterols and then activates the SREBP pathway, thereby causing
compensatory increases in HMGCR protein in the cells with Insigs. On the other hand, the
pomalidomide moiety recruits the E3 ubiquitin ligase CRBN to degrade HMGCR. In SRD15 cells that
lack Insig-1 and -2, the SREBP pathway has been constitutively activated, and statins cannot further
increase HMGCR expression. The basal level of HMGCR is very high. Therefore, P22A-induced
HMGCR degradation through CRBN is very dramatic. In Huh7 cells, the two effects of P22A exist
and atorvastatin only increases HMGCR. Hence, the HMGCR level in P22A-treated cells were much
less than that in atorvastatin-treated cells (Figure 6C), indicating that P22A does promote HMGCR
degradation via CRBN. Notably, P22A inhibits cholesterol biosynthesis with the similar potency to
atorvastatin (Figure 6E).
Together, our study establishes a proof-of-concept for HMGCR degradation by PROTACs and
suggests a new potential strategy for treating hypercholesterolemia and CVD. Our work also reveals
that, in contrast to conventional inhibition by small molecule inhibitors (such as statins), PROTACs
may potentially be used as a general way to solve compensatory upregulation of proteins that
contributes to illness, adverse effects and drug resistance65-67
.
22
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
EXPERIMENTAL SECTION
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
Chemistry. All commercial chemical materials (Energy, Ouhe, Aladdin, J&K Chemical Co. Ltd.)
1
13
were used without further purification. All solvents were analytical grade. The H-NMR and C-NMR
spectra were recorded on a Bruker AVANCE III 400 MHz spectrometer in CDCl , CD OD, using
3
3
1
3
tetramethylsilane (TMS) or solvent peak as a standard. All C NMR spectra were recorded with
complete proton decoupling. Low-resolution mass spectral analyses were performed with a Waters
ACQUITY UPLC/MS. High-resolution mass spectral analyses were performed with a Thermo
Scientific OrbiTrap Q-Exactive. Analytical thin-layer chromatography (TLC) was performed on
Yantai Chemical Industry Research Institute silica gel 60 F254 plates, and flash column
chromatography was performed on Qingdao Haiyang Chemical Co. Ltd. silica gel 60 (200−300 mesh).
A BUCHI Rotavapor R-3 was used to remove solvents by evaporation. The purity of all the final tested
compounds was more than 95% as confirmed by NMR and UPLC. Atorvastatin was from Energy
Chemical (E1201400010), Pomalidomide was from Bide Pharmatech Ltd (BD235626). Compounds
P13A-P23A were synthesized and dissolved in DMSO at a concentration of 10 mM and stored at -
20°C. For cell culture experiments drugs were diluted at least by 1:1,000 so that the final DMSO
concentration was 0.1% or lower. Sterols were obtained from Steraloids. MG-132 was obtained from
14
Sigma. C-acetic acid sodium salt was obtained from Perkin Elmer.
2-(3-Bromophenyl)acetyl chloride (1). To a 100 ml round-bottom flask was added 2-(3-
bromophenyl)acetic acid (4.2 g, 20 mmol), thionyl chloride (4 ml, 60 mmol), DMF (1 drop) and DCM
(30 ml) as solvent. The mixture was stirred at room temperature overnight. Then the solvent was
23
ACS Paragon Plus Environment

Page 25 of 76
Journal of Medicinal Chemistry
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
removed under reduced pressure. The residual was used directly in the next step without purification.
2
-(3-Bromophenyl)-1-(4-fluorophenyl)ethan-1-one (2). To a 100 ml round-bottom flask containing
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
AlCl (3.2 g, 24 mmol) and fluorobenzene (9 ml, 100 mmol) at ice bath was added 1 dropwise. After
3
o
stirring at ice bath for 20 min, it was stirred at 50 C for another 7 h. The mixture was diluted by 2N
HCl solution and extracted by EtOAc. The organic layer was purified by silica gel column with hexane:
1
EtOAc = 20: 1 as eluent to get 2 (Isolated yield = 93 %). H-NMR (400 MHz, CDCl , ppm): 8.02 –
3
1
3
7
.99 (m, 2H), 7.40 – 7.38 (m, 2H), 7.21 – 7.11 (m, 4H), 4.42 (s, 2H). C-NMR (100 MHz, CDCl ,
3
ppm): 195.3, 167.3, 164.7, 136.6, 132.9, 132.9, 132.6, 131.4, 131.3, 130.3, 128.3, 122.8, 116.1, 115.9,
+
44.9. LR−MS: calculated for C H BrFO [M+H] , 292.99; found, 293.15.
1
4
11
2
-Bromo-2-(3-bromophenyl)-1-(4-fluorophenyl)ethan-1-one (3). Compound 2 (2.9 g, 10 mmol)
was dissolved in CHCl to which 30 % HBr in AcOH (1 ml) was added at room temperature. A solution
3
of bromine (about 600 uL) in CHCl was added dropwise. The addition continued until slight bromine
3
coloration remined. Then a saturated Na SO solution was added and extract with EtOAc. The organic
2
3
layer was purified by silica gel column with hexane: EtOAc = 20: 1 as eluent to get 3 (Isolated yield
1
=
70 %). H-NMR (400 MHz, CDCl , ppm): 8.03 (dd, J = 8.88 Hz, J = 5.24 Hz, 2H), 7.69 (t, J = 1.72
3
Hz, 1H), 7.46 (t, J = 7.08 Hz, 2H), 7.25 (t, J = 7.88 Hz, 1H), 7.14 (t, J = 8.48 Hz, 2H), 6.22 (s, 1H).
1
3
C-NMR (100 MHz, CDCl , ppm): 189.2, 167.5, 164.9, 137.8, 132.5, 132.3, 132.1, 132.0, 130.6,
3
+
1
30.4, 130.4, 128.0, 122.9, 116.4, 116.2, 48.8. LR−MS: calculated for C H Br FO [M+H] , 370.90;
1
4
10
2
found, 371.34.
2
-(1-(3-Bromophenyl)-2-(4-fluorophenyl)-2-oxoethyl)-4-methyl-3-oxo-N-phenylpentanamid (4).
A mixture of compound 3 (2.2 g, 5.9 mmol), 4-Methyl-3-oxo-N-phenylpentanamide (1.2 g, 5.9 mmol)
24
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
o
and K CO (1.2 g, 8.9 mmol) in 30 ml acetone was refluxed at 60 C overnight. Then dilute with brine
2
3
solution and extract with EtOAc. The organic layer was purified by silica gel column with hexane:
1
EtOAc = 7: 1 as eluent to get 4 (Isolated yield = 55 %). H-NMR (400 MHz, CDCl , ppm): 7.98 (dd,
3
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
J = 8.84 Hz, J = 5.36 Hz, 2H), 7.45 (t, J = 1.56 Hz, 1H), 7.37 – 7.04 (m, 12H), 5.36 (d, J = 9.52 Hz,
1
H), 4.51 (d, J = 9.56 Hz, 1H), 3.01 – 2.95 (m, 1H), 1.24 (d, J = 5.92 Hz, 3H), 1.15 (d, J = 5.84 Hz,
1
3
3H). C-NMR (100 MHz, CDCl , ppm): 209.6, 196.0, 167.2, 164.9, 164.8, 137.6, 136.7, 132.0, 132.0,
3
1
1
31.9, 131.8, 131.6, 131.2, 131.1, 129.2, 127.5, 125.3, 123.5, 120.6, 116.1, 115.9, 64.30, 54.0, 41.3,
+
8.7, 18.2. LR−MS: calculated for C H BrFNO [M+H] , 496.08; found, 496.27.
2
6
24
3
Tert-butyl
2-((4R,6R)-6-(2-(3-(3-bromophenyl)-2-(4-fluorophenyl)-5-isopropyl-4-
(
phenylcarbamoyl)-1H-pyrrol-1-yl)ethyl)-2,2-dimethyl-1,3-dioxan-4-yl)acetate (5). To a 50 ml
round-bottom flask containing 5 ml toluene, 5 ml THF and 10 ml cyclohexane was added 4 (2.3 g, 4.6
mmol), tert-butyl 2-((4R,6R)-6-(2-aminoethyl)-2,2-dimethyl-1,3-dioxan-4-yl)acetate (2.5g, 9.2 mmol)
o
and t-BuCOOH (470mg, 4.6mmol). The mixture was refluxed at 115 C for 40 h and extracted with
EtOAc. The organic layer was purified by silica gel column with hexane: EtOAc = 15: 1 as eluent to
1
get 5 (Isolated yield = 82%). H-NMR (400 MHz, CDCl , ppm): 7.30 (s, 1H), 7.26 – 7.16 (m, 7H),
3
7.05 – 7.01 (m, 5H), 6.86 (s, 1H), 4.17 – 4.13 (m, 1H), 4.09 – 4.02 (m, 1H), 3.84 – 3.77 (m, 1H), 3.72
–
3.65 (m, 1H), 3.51 – 3.44 (m, 1H), 2.39 (dd, J = 15.24 Hz, J = 7.00 Hz, 1H), 2.24 (dd, J = 15.24 Hz,
J = 6.12 Hz, 1H), 1.69 – 1.62 (m, 2H), 1.52 (s, 3H), 1.50 (s, 3H), 1.43 (s, 9H), 1.36 (s, 4H), 1.30 (s,
1
3
3H), 1.08 – 1.03 (m, 1H). C-NMR (100 MHz, CDCl , ppm): 170.30, 165.0, 163.8, 161.3, 141.1,
3
1
38.2, 136.9, 133.3, 133.2, 133.1, 129.8, 129.4, 129.1, 129.0, 128.9, 127.9, 127.9, 124.0, 122.2, 120.3,
120.0, 115.8, 115.6, 98.8, 80.8, 66.4, 66.0, 42.5, 40.9, 38.2, 36.0, 30.0, 28.2, 26.2, 22.0, 21.8, 19.8.
+
LR−MS: calculated for C H BrFN O [M+H] , 733.26; found, 733.36.
40 47
2
5
25
ACS Paragon Plus Environment

Page 27 of 76
Journal of Medicinal Chemistry
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
Tert-butyl 2-((4R,6R)-6-(2-(2-(4-fluorophenyl)-5-isopropyl-4-(phenylcarbamoyl)-3-(3-(4,4,5,5-
tetramethyl-1,3,2-dioxaborolan-2-yl)phenyl)-1H-pyrrol-1-yl)ethyl)-2,2-dimethyl-1,3-dioxan-4-
yl)acetate (6). To a 50 ml round-bottom flask was added 5 (2.5 g, 3.4 mmol), Pd(dppf) Cl .DCM (90
2
2
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
mg, 0.1 mmol), KOAc (700 mg, 6.8 mmol), Bis(pinacoloto)diboron (1 g, 3.75 mmol) and 30 ml DMF
o
as solvent under the protection of argon. The mixture was stirred at 80 C overnight and extracted with
EtOAc. The organic layer was purified by silica gel column with hexane: EtOAc = 10: 1 as eluent to
1
get 6 (Isolated yield = 68%). H-NMR (400 MHz, CDCl , ppm): 7.68 (s, 1H), 7.68 (d, J = 5.12 Hz,
3
1
H), 7.20 – 7.14 (m, 6H), 7.07 (d, J = 7.84 Hz, 2H), 7.00 – 6.94 (m, 3H), 6.88 (s, 1H), 4.17 – 4.13 (m,
1H), 4.09 – 4.02 (m, 1H), 3.87 – 3.81 (m, 1H), 3.72 – 3.65 (m, 1H), 3.60 – 3.57 (m, 1H), 2.39 (dd, J =
5.24 Hz, J = 6.96 Hz, 1H), 2.24 (dd, J = 15.48 Hz, J = 6.24 Hz, 1H), 1.68 – 1.65 (m, 2H), 1.53 (s,
1
1
3
3H), 1.52 (s, 3H), 1.43 (s, 9H), 1.36 (s, 4H), 1.30 (s, 3H), 1.23 (s, 12H), 1.07 – 1.02 (m, 1H). C-NMR
(
100 MHz, CDCl , ppm): 170.3, 164.91, 163.6, 161.1, 141.7, 138.5, 137.4, 134.0, 133.4, 133.4, 133.3,
3
1
32.7, 129.0, 128.9, 128.6, 128.4, 128.4, 127.8, 123.6, 121.9, 120.0, 115.4, 115.2, 115.2, 99.7, 83.7,
80.7, 66.5, 66.0, 42.5, 40.9, 38.1, 36.0, 29.9, 28.2, 26.13, 24.8, 24.8, 21.8, 21.6, 19.7. LR−MS:
+
calculated for C H BFN O [M+H] , 781.43; found, 781.45.
4
6
59
2
7
Tert-butyl
phenylcarbamoyl)-1H-pyrrol-1-yl)ethyl)-2,2-dimethyl-1,3-dioxan-4-yl)acetate (7). To a 50 ml
round-bottom flask containing 20ml THF was added 6 (1.8 g, 2.3mmol), 30% H O in H O (2ml) and
2-((4R,6R)-6-(2-(2-(4-fluorophenyl)-3-(3-hydroxyphenyl)-5-isopropyl-4-
(
2
2
2
NaOH (200 mg, 4.6 mmol) at ice bath. After stirring at ice bath for 15 min, it was warmed up to room
temperature for another 15 min. Extract with EtOAc and wash by brine. The organic layer was purified
1
by silica gel column with hexane: EtOAc = 5: 1 as eluent to get 7 (Isolated yield = 84%). H-NMR
(
400 MHz, CDCl , ppm): 7.26 – 7.08 (m, 6H), 7.03 – 6.94 (m, 5H), 6.65 – 6.63 (m, 3H), 4.19 – 4.14
3
26
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
(m, 1H), 4.08 – 4.02 (m, 1H), 3.85 – 3.81 (m, 1H), 3.79 – 3.67 (m, 1H), 3.60 – 3.56 (m, 1H), 2.38 (dd,
J = 15.28 Hz, J = 7.08 Hz, 1H), 2.23 (dd, J = 15.28 Hz, J = 6.96 Hz, 1H), 1.66 – 1.63 (m, 2H), 1.50
13
(
s, 3H), 1.49 (s, 3H), 1.43 (s, 3H), 1.35 (s, 3H), 1.29 – 1.20 (m, 4H), 1.07 – 1.02 (m, 1H). C-NMR
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
(100 MHz, CDCl , ppm): 170.5, 165.3, 163.6, 161.1, 156.4, 141.6, 138.2, 136.1, 133.2, 133.2, 129.6,
3
1
6
28.9, 128.8, 128.3, 128.2, 123.9, 122.5, 121.7, 120.3, 117.5, 115.5, 115.3, 115.1, 114.0, 98.8, 80.9,
6.5, 66.0, 42.5, 40.9, 38.1, 36.0, 30.0, 28.2, 26.1, 24.8, 21.8, 21.6, 19.7. LR−MS: calculated for
+
C H FN O [M+H] , 671.34; found, 671.32.
4
0
48
2
6
Tert-butyl 2-((4R,6R)-6-(2-(2-(4-fluorophenyl)-5-isopropyl-4-(phenylcarbamoyl)-3-(3-(prop-2-
yn-1-yloxy)phenyl)-1H-pyrrol-1-yl)ethyl)-2,2-dimethyl-1,3-dioxan-4-yl)acetate (8a). To a 10 ml
sealed tube containing 3ml MeCN was added 7 (670 mg, 1 mmol), K CO (540mg, 4 mmol) and 3-
2
3
o
bromopropyne (4 mmol). The mixture was stirred at 70 C for 20 h and extracted with EtOAc. The
organic layer was purified by silica gel column with hexane: EtOAc = 10: 1 as eluent to get 8a (Isolated
1
yield = 88%). H-NMR (400 MHz, CDCl , ppm): 7.20 – 7.16 (m, 4H), 7.13 – 7.06 (m, 3H), 7.05 –
3
6.96 (m, 4H), 6.73 – 6.69 (m, 3H), 4.44 (s, 2H), 4.18 – 4.13 (m, 1H), 4.09 – 4.03 (m, 1H), 3.86 – 3.77
(m, 1H), 3.69 – 3.68 (m, 1H), 3.57 – 3.53 (m, 1H), 2.38 (dd, J = 15.20 Hz, J = 6.80 Hz, 1H), 2.26 –
2.21 (m, 3H), 1.68 – 1.65 (m, 2H), 1.54 (s, 3H), 1.52 (s, 3H), 1.43 (s, 9H), 1.37 – 1.32 (m, 4H), 1.30
1
3
(
m, 3H), 1.09 – 1.01 (m, 1H). C-NMR (100 MHz, CDCl , ppm): 170.3, 164.8, 163.6, 161.2, 157.5,
3
1
41.6, 138.6, 136.9, 133.3, 133.2, 129.5, 128.9, 128.8, 128.3, 124.0, 123.6, 121.6, 119.7, 116.4, 115.6,
115.4, 115.4, 114.2, 98.8, 80.8, 78.4, 75.5, 66.5, 66.0, 55.8, 42.6, 40.9, 38.2, 36.1, 30.0, 28.2, 26.2,
+
2
1.8, 21.7, 19.8. LR−MS: calculated for C H FN O [M+H] , 709.36; found, 709.46.
4
3
50
2
6
Tert-butyl 2-((4R,6R)-6-(2-(2-(4-fluorophenyl)-5-isopropyl-3-(3-(pent-4-yn-1-yloxy)phenyl)-4-
27
ACS Paragon Plus Environment

Page 29 of 76
Journal of Medicinal Chemistry
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
(phenylcarbamoyl)-1H-pyrrol-1-yl)ethyl)-2,2-dimethyl-1,3-dioxan-4-yl)acetate (8b). It was
1
synthesized as compound 8a. H-NMR (400 MHz, CDCl , ppm): 7.19 – 7.16 (m, 4H), 7.13 – 7.06 (m,
3
3
H), 7.00 – 6.96 (m, 4H), 6.73 – 6.69 (m, 3H), 4.18 – 4.13 (m, 1H), 4.09 – 4.03 (m, 1H), 3.86 – 3.77
(m, 3H), 3.69 – 3.68 (m, 1H), 3.57 – 3.53 (m, 1H), 2.38 (dd, J = 15.20 Hz, J = 6.80 Hz, 1H), 2.26 –
.21 (m, 3H), 1.91 (t, J = 1.04 Hz, 1H), 1.81 – 1.78 (m, 2H), 1.68 – 1.65 (m, 2H), 1.54 (s, 3H), 1.52
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
1
3
(s, 3H), 1.43 (s, 9H), 1.37 – 1.32 (m, 4H), 1.30 (m, 3H), 1.09 – 1.01 (m, 1H). C-NMR (100 MHz,
CDCl , ppm): 170.3, 165.0, 163.6, 161.1, 158.8, 141.5, 138.6, 136.0, 133.3, 133.2, 129.4, 128.8, 128.8,
3
1
28.4, 128.4, 123.6, 123.0, 121.8, 119.7, 115.9, 115.6, 115.4, 114.0, 98.8, 83.5, 80.8, 68.9, 66.5, 66.1,
66.0, 53.5, 42.5, 40.9, 38.2, 36.1, 30.0, 28.2, 28.0, 26.2, 21.9, 21.7, 19.8, 15.1. LR−MS: calculated for
+
C H FN O [M+H] , 737.39; found, 737.36.
4
5
54
2
6
Tert-butyl 2-((4R,6R)-6-(2-(2-(4-fluorophenyl)-3-(3-(hex-5-yn-1-yloxy)phenyl)-5-isopropyl-4-
(phenylcarbamoyl)-1H-pyrrol-1-yl)ethyl)-2,2-dimethyl-1,3-dioxan-4-yl)acetate (8c). It was
1
synthesized as compound 8a. H-NMR (400 MHz, CDCl , ppm): 7.20 – 7.17 (m, 4H), 7.12 – 7.06 (m,
3
3H), 7.03 – 6.97 (m, 3H), 6.94 (s, 1H), 6.71 – 6.69 (m, 3H), 4.18 – 4.03 (m, 2H), 3.86 – 3.78 (m, 1H),
3
.71 – 3.64 (m, 3H), 3.57 – 3.51 (m, 1H), 2.38 (dd, J = 15.28 Hz, J = 6.80 Hz, 1H), 2.23 (dd, J = 15.28
Hz, J = 7.06 Hz, 1H), 2.16 – 2.12 (m, 2H), 1.93 (t, J = 1.04 Hz, 1H), 1.73 – 1.63 (m, 4H), 1.58 – 1.49
1
3
(
m, 8H), 1.44 (s, 9H), 1.36 (s, 3H), 1.31 – 1.27 (m, 4H), 1.09 – 1.01 (m, 1H). C-NMR (100 MHz,
CDCl , ppm): 170.3, 165.0, 163.6, 161.2, 158.9, 141.5, 138.6, 136.0, 133.3, 133.2, 129.4, 128.8, 128.8,
3
128.4, 128.4, 123.7, 122.9, 121.8, 119.6, 115.9, 115.6, 115.5, 115.4, 114.0, 98.8, 84.2, 80.8, 68.7, 67.2,
6
6.5, 66.0, 42.6, 40.9, 38.2, 36.1, 30.0, 28.2, 28.1, 26.2, 25.0, 21.9, 21.7, 19.8, 18.1. LR−MS:
+
calculated for C H FN O [M+H] , 751.40; found, 751.37.
4
6
56
2
6
28
ACS Paragon Plus Environment