Discovery of Novel Naphthylphenylketone and Naphthylphenylamine Derivatives as Cell Division Cycle 25B (CDC25B) Phosphatase Inhibitors: Design, Synthesis, Inhibition Mechanism, and in Vitro Efficacy against Melanoma Cell Lines

Article abstract of DOI:10.1021/acs.jmedchem.9b00632

CDC25 phosphatases play a critical role in the regulation of the cell cycle and thus represent attractive cancer therapeutic targets. We previously discovered the 4-(2-carboxybenzoyl)phthalic acid (NSC28620) as a new CDC25 inhibitor endowed with promising anticancer activity in breast, prostate, and leukemia cells. Herein, we report a structure-based optimization of NSC28620, leading to the identification of a series of novel naphthylphenylketone and naphthylphenylamine derivatives as CDC25B inhibitors. Compounds 7j, 7i, 6e, 7f, and 3 showed higher inhibitory activity than the initial lead, with K_i values in the low micromolar range. Kinetic analysis, intrinsic fluorescence studies, and induced fit docking simulations provided a mechanistic understanding of the activity of these derivatives. All compounds were tested in the highly aggressive human melanoma cell lines A2058 and A375. Compound 4a potently inhibited cell proliferation and colony formation, causing an increase of the G2/M phase and a reduction of the G0/G1 phase of the cell cycle in both cell lines.

Full text of DOI:10.1021/acs.jmedchem.9b00632

Subscriber access provided by Nottingham Trent University
Article
Discovery of Novel Naphthylphenylketone and Naphthylphenylamine
Derivatives as Cell Division Cycle 25B (CDC25B) Phosphatase
Inhibitors: Design, Synthesis, Inhibition Mechanism
and in Vitro Efficacy against Melanoma Cell Lines
Carmen Cerchia, Rosarita Nasso, Matteo Mori, Stefania Villa, Arianna Gelain,
Alessandra Capasso, Federica Aliotta, Martina Simonetti, Rosario Rullo, Mariorosario
Masullo, Emmanuele De Vendittis, Maria Rosaria Ruocco, and Antonio Lavecchia
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b00632 • Publication Date (Web): 11 Jul 2019
Downloaded from pubs.acs.org on July 16, 2019
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 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of Novel Naphthylphenylketone and Naphthylphenylamine
Derivatives as Cell Division Cycle 25B (CDC25B) Phosphatase Inhibitors:
Design, Synthesis, Inhibition Mechanism and in Vitro Efficacy against
Melanoma Cell Lines
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
Carmen Cerchia,1,+ Rosarita Nasso,2,4,+ Matteo Mori,3+ Stefania Villa, Arianna Gelain ,
3
3
2
2
2
2,5
Alessandra Capasso, Federica Aliotta, Martina Simonetti, Rosario Rullo, Mariorosario
4
2
2,*
1,*
Masullo, Emmanuele De Vendittis, Maria Rosaria Ruocco, Antonio Lavecchia
1
Department of Pharmacy, "Drug Discovery" Laboratory, University of Naples Federico II,
Via D. Montesano, 49, 80131 Naples, Italy
2
Department of Molecular Medicine and Medical Biotechnology, University of Naples
Federico II, Via S. Pansini 5, 80131 Naples, Italy
3
Department of Pharmaceutical Sciences, University of Milan, Via Mangiagalli, 25, 20133
Milan, Italy
4
Department of Movement Sciences and Wellness, University of Naples "Parthenope", 80133
Naples, Italy
5
Institute for the Animal Production Systems in the Mediterranean Environment, Via Argine
1085, 80147 Naples, Italy
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 90
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
CDC25 phosphatases play a critical role in the regulation of the cell cycle and thus represent
attractive cancer therapeutic targets. We previously discovered the 4-(2-
carboxybenzoyl)phthalic acid (NSC28620) as a new CDC25 inhibitor endowed with
promising anticancer activity in breast, prostate and leukemia cells. Herein, we report a
structure-based optimization of NSC28620, leading to the identification of a series of novel
naphthylphenylketone (NPK) and naphthylphenylamine (NPA) derivatives as CDC25B
inhibitors. Compounds 7j, 7i, 6e, 7f and 3 showed higher inhibitory activity than the initial
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
lead, with K values in the low micromolar range. Kinetic analysis, intrinsic fluorescence
i
studies and induced fit docking simulations provided a mechanistic understanding of the
activity of these derivatives. All compounds were tested in the highly aggressive human
melanoma cell lines A2058 and A375. Compound 4a potently inhibited cell proliferation and
colony formation, causing an increase of the G2/M phase and a reduction of G0/G1 phase of
the cell cycle in both cell lines.
2
ACS Paragon Plus Environment

Page 3 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
The cell division cycle 25 phosphatases (CDC25s) are members of the family of dual-
specificity phosphatases (DSPs) and regulate cyclin-dependent kinase (Cdk) complexes,
which are key participants in the cell cycle.1, 2 CDC25s are also critical components of the
cell cycle checkpoints for handling DNA damage and are inactivated or degraded to induce
cell cycle arrest. Misregulation of these processes can cause genomic instability3, 4 and
ultimately lead to cancer. In mammalian cells three variants of CDC25 were identified:
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
5
6
CDC25A, -B and -C. CDC25A dephosphorylates Cdk4-Cyclin D and Cdk6-Cyclin D
7
8, 9
complexes, as well as Cdk1-CyclinB, Cdk2-Cyclin A and Cdk2-Cyclin E complexes,
controlling both G1/S and G2/M progression. CDC25B regulates the G2/M transition by
activating Cdk1-Cyclin B complex at the centrosome which is subsequently completely
1
0
activated by CDC25C in the nucleus at the onset of mitosis. Furthermore, CDC25B was
reported to dephosphorylate and activate also Cdk2-Cyclin A and Cdk2-Cyclin E
complexes.11, 12 CDC25C regulates the G2/M transition by targeting Cdk1-Cyclin B
complex.13, 14 However, it is believed that all three variants of CDC25 are involved in the
1
5
regulation of both G1/S and G2/M transitions. CDC25A and CDC25B overexpression is
frequently observed in various human cancers, and often associated with more aggressive
tumors and poor clinical outcome.10, 16 Concerning CDC25C, only a few studies showed an
overexpression of this form in cancers.17, 18 However, growing evidence suggests that the
overexpression of CDC25C could be underrated because of the non-consideration of its
alternative splicing.19, 20 The overexpression of CDC25s in many human cancers supports
their clinical significance and has encouraged the pursuit of specific small molecule
inhibitors.
21
The crystal structures of the catalytic domain of CDC25A (PDB: 1C25) and CDC25B
2
2
(
PDB:1QB0) have been solved at 2.3 Å and 1.9 Å resolution, respectively. Despite the close
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 90
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
structural similarity between the two proteins with regard to the catalytic domain, CDC25A
fails to bind oxyanions in its catalytic site, whereas CDC25B readily binds tungstate and
sulfate. Moreover, the C-terminal region is well ordered in the crystal structure of CDC25B
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
(residues 531−547), and folded into a well-defined α-helix, while in the CDC25A crystal
structure this region is undefined. The active-site region of CDC25A appears in a flat
conformation and exposed to bulk solvent, whereas the structure of the catalytic domain of
CDC25B reveals a shallow active site cleft, positioned near a large cavity formed by highly
hydrophilic moieties in the C-terminal residues of CDC25B, which was referred to as the
“
swimming pool” due to the abundant presence of many well organized water molecules.23
Therefore, the structure-based design of inhibitors and the derivation of significant structure-
activity relationships (SAR) are really challenging. Nevertheless, several CDC25s inhibitors
with different structural features were discovered over the past years, the most potent of
which are quinonoid-based derivatives.24-27 These latter mainly act via irreversible inhibition
of CDC25s by electrophilic modification28, 29 or oxidation of the critical cysteine residue in
the catalytic domain by reactive oxygen species (ROS).30, 31 Furthermore, quinone-containing
agents could potentially trigger a range of unrelated events in cells because ROS may oxidize
2
6
other phosphatases, as well as unrelated cysteine-based enzymes, raising questions about
their potential toxicity and limiting their therapeutic applications.32
Among the CDC25 inhibitors reported to date, very few display reversible inhibition
3
3
34
kinetics. The indolylhydroxyquinone 8L, the naphthofurandione 5169131, the pyrazolone
derivative EK-6136,35 the thiazolepyrimidine 4436 and the aminoisoquinolinone 1337 in
Figure 1 are some representatives. However, most of the reversible inhibitors have potency
within the micromolar range with no currently defined clinical utility. Thus, the identification
of non-quinoid reversible inhibitors that do not oxidize cysteine residues or generate ROS
4
ACS Paragon Plus Environment

Page 5 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
would have great potential for the development of pharmaceutical agents and useful
biological tools.
Insert Figure 1
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
We have previously reported the discovery of a new CDC25 inhibitor with cytotoxic
activity, the 4-(2-carboxybenzoyl)phthalic acid (NSC28620) by means of a structure-based
high-throughput virtual screening.38 This compound displayed reversible, competitive
inhibition kinetics with K values of 2.3 and 5.3 µM for CDC25A and -B, respectively,
i
arrested cells at the G0/G1 and G2/M phases of the cell cycle, increased Cdk1
hyperphosphorylation in K562 leukemia cells and significantly inhibited the growth of
human MCF-7 breast, PC-3 prostate and K562 leukemia cancer cell lines.
Metastatic melanoma is one of the most aggressive and therapy-resistant human cancers.
So, much effort is being directed at discovering new, effective agents for the treatment of this
cancer, by targeting various signaling pathways and attacking elements that underlie the
39
tumor's propensity for growth and chemoresistance. Previous studies have reported that
Cdk2, Cdk4, Cdk6, cyclins D1, E, and D3 and CDC25A were significantly overexpressed in
metastatic melanomas as compared to nevus tissue.40-42 Moreover, other literature data have
indicated that phosphorylation and inhibition of CDC25B cause delayed progression from G2
into mitosis in melanoma cells.43 Finally, we have recently demonstrated that a small
molecule CDC25 inhibitor was able to arrest melanoma cells in G2/M phase through an
2
7
intrinsic apoptotic pathway. Taken together, these evidences emphasize that CDC25 could
be considered as a possible target in treating melanoma.
Herein we present the results of our lead optimization program, pursued on the CDC25
inhibitor lead NSC28620. Guided by preliminary docking studies, we rationally explored
different substitution patterns with the purpose of improving the inhibitory profile of the
starting lead and optimizing its potential as antitumor drug. A total of 31
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 90
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
naphthylphenylketone (NPK) and naphthylphenylamine (NPA) derivatives were synthesized
(Table 1) and tested according to a SAR investigation by a step-by-step modification strategy.
Among them, twelve compounds (3, 3b, 6b,d,e, 7d,f,i,j, 8, 9 and 12) displayed an inhibition
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
potency similar or even higher compared to that exerted by the lead compound NSC28620
(Table 1), 7j being 7-fold more active than NSC28620; the remaining nineteen derivatives
conserved a lower but measurable inhibition. The biological effect of these compounds was
evaluated in a cellular context, using the highly aggressive human melanoma cell lines A2058
and A375. The data showed that, when tested at a very low concentration, only compound 4a
was by far the most effective one in the inhibition of cell proliferation and colony formation,
as emerging from the cytotoxicity tests. Furthermore, compound 4a affected the cell cycle
progression of either A2058 and A375 cells with a significant increase of the G2/M phase
and a reduction of G0/G1 phase.
RESULTS AND DISCUSSION
Structure-Guided Optimization of Lead NSC28620 to Target the CDC25B Phosphatase.
We used the binding mode of the lead compound NSC28620 into the CDC25B catalytic
3
8
domain as predicted by docking experiments (Figure 2A) for the rational design of new
analogues to study the SAR and to achieve further increases in potency. Four regions in
NSC28620 were explored for modifications (Figure 2B) to design new compounds based
upon structural data. First, we explored additional optimization of the phthalic moiety that
binds to the swimming pool pocket of CDC25B, a protein region near to the catalytic site and
composed by both hydrophobic (F543, L445, M483, W550, P444 and L545) and polar
residues (Y428, R479, R482, E446, R544, T547 and S549).
Insert Figure 2
6
ACS Paragon Plus Environment

Page 7 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
As depicted in Figure 2A, the two carboxylate groups of the phthalic moiety (positions 1
and 2) of NSC28620 protrude into the swimming pool region and form salt-bridges with both
R482 and R544 side chains. In order to explore new potential interactions (hydrophobic
and/or hydrophilic) within the swimming pool aimed at improving the inhibition power, we
thoroughly replaced the phthalic system of NSC28620 with a bulkier naphthalene ring
system; this moiety was undecorated (compounds 1-7) or decorated with hydroxy, methoxy
or amino groups linked to one out of the positions 3, 4 and 5, corresponding to substituents
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
1
2
3
R , R and R , respectively (compounds 1a-c, 3a,b, 4a,b, 6a,b,d,e,f, 7a,b,d,e,f,i,j, 8, 9, 10,
1
1, 12) (Table 1). In the next step, we performed the optimization of the benzoyl substituent
interacting with the catalytic site of CDC25B. In the complex of CDC25B with NSC28620,
the carboxylate group of the benzoyl moiety binds deep in the catalytic site, making a salt-
bridge with the R479 guanidinium group and a H-bond with the catalytic backbone NHs of
R479 and E478. This moiety occupies the same location as the sulfate ion bound to the
22
catalytic site of the CDC25B crystal structure. Therefore, we sought to explore the role of
4
5
the carboxylic acid function at position 2 or 3 of the benzoyl moiety (substituents R and R ,
respectively) in order to establish more favorable interactions within the catalytic domain
(compounds 1, 1a-c, 6, 6a,b,d,e,f, 7, 7a,b,d,e,f,i,j and 12). Moreover, according to the well-
4
4
known Caco-2 uptake studies , demonstrating the enhanced permeability of esters compared
to their corresponding acids, the carboxylic acidic moiety was modified via esterification
with the aim of increasing the lipophilicity and enhancing the cellular uptake (compounds 2,
3
, 3a,b, 4, 4a,b, 8, 9, 10, 11). Finally, the NSC28620•CDC25B complex revealed that the
carbonyl group of the inhibitor accepts a H-bond from the Y428 OH group (Figure 2A). On
the basis of the structural model, we replaced the carbonyl group (position X) with an amino
group with the overall goal to explore the H-bond acceptor or donor capability at this position
(compounds 3, 3a,b, 4, 4a,b, 5, 6, 6a,b,d,e,f, 7, 7a,b,d,e,f,i,j, 8, 9, 10, 11, and 12).
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 90
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
Synthesis of Rationally Designed NPK and NPA Derivatives. The general synthetic
strategy followed for the preparation of NPK derivatives is outlined in Scheme 1 and Scheme
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
. Grignard’s reagent was prepared in situ starting from the commercially available 1-
bromonaphthalene in presence of magnesium turnings. The organomagnesium halide was
4
5
added to a solution of phthalic anhydride in dry tetrahydrofuran to afford compound 1. The
subsequent Fischer esterification of the carboxylic group in ethanol with sulfuric acid as
catalyst provided the final ester 2 (Scheme 1).46
Insert Scheme 1
Otherwise, in order to obtain the same class of compounds previously described but
bearing a hydroxyl group on the naphthalene ring, a second synthetic approach was
developed (Scheme 2).
The 1-aminonaphthalene (24), by halogenation with bromine in acetic acid, provided the
intermediate 13 whose treatment with sodium nitrite led to the formation of the diazonium
4
7
salt 14, which was isolated and characterized. The conversion of 14 into the 4-bromo-2-
hydroxynaphthalene 15a was performed with sodium borohydride. This latter or the
commercially available 4-bromo-1-hydroxynaphthalene (15b) were subsequently protected as
methyl ether in the presence of iodomethane to provide the intermediates 16a,b. Finally, the
exchange of the halogen atom with lithium by treatment with a solution of n-butyllithium (n-
BuLi) dry THF at -78°C, followed by the formylation with N,N-dimethylformamide, led to
the naphthaldehydes 17a,b (Scheme 2).
On the other hand, the key intermediate 2048 was obtained by condensation of the
commercially available 2-bromobenzoic acid with 2-amino-2-methyl-1-propanol to afford the
intermediate 18. This latter was reacted with thionyl chloride in diethyl ether to give the
8
ACS Paragon Plus Environment

Page 9 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
oxazoline 19, which underwent a metal-halogen exchange in the presence of the
organolithium species n-BuLi to afford 20.
Compound 20 reacted with the appropriate substituted naphthaldehydes 17a,b and the acid-
catalyzed deprotection of the carboxylic moiety led to the desired lactones 21a,b in alkaline
condition. The oxidation of these latter with potassium permanganate and 25% aqueous
solution of KOH in pyridine afforded the 2-(methoxy-1-naphthyl)benzoic acids 1a and 1b.
Finally, the cleavage of the methyl ether of 1a with a 1M solution of boron tribromide in
dichloromethane provided the final product 1c (Scheme 2).
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
Insert Scheme 2
In parallel, the synthesis of NPA derivatives was carried out (Schemes 3 and 4). The
appropriate substituted aryl halide and aryl amine underwent a palladium-catalyzed
4
9
Buchwald amination reaction (compounds 3, 3a-c, 3h, 4, 4a-c, 4g,h and 5) using a mixture
of palladium acetate (Pd(OAc) ) and racemic 2,2’-bis(diphenylphosphino)-1,1’-binaphthalene
2
(
BINAP) as catalysts, in the presence of Cs CO . Since the yield of the aryl amination
2 3
reaction was critically hindered by the electron deficiency of the amine, the combination of
the more nucleophilic amine and more electrophilic aryl bromo derivatives was used
(procedures C1 and C2, Scheme 3).
The appropriate bromo naphthylamines for the synthesis of 4g,h were N-protected with di-
5
0
tert-butyl dicarbonate. The protection of the substituted benzoic acids to obtain ethyl-2(3)-
bromobenzoate (22, 23) was performed by Fischer esterification.51
Insert Scheme 3
A basic hydrolysis of the ester group was then performed, leading to the formation of the
5
2
corresponding acids 6, 6a-c, 7, 7a-c, 7g,h and 12 (Scheme 4) . The cleavage of methoxy
protection was performed with a 1M solution of boron tribromide in dichloromethane to
afford compounds 6d-f and 7d-f (Scheme 4). The N-Boc protecting group of compounds 7g
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 90
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
and 7h, was hydrolyzed with HCl (4M in 1,4-dioxane) to give compounds 7i,j while by using
trifluoroacetic acid products 10 and 11 were obtained (Scheme 4). From the deprotected
intermediate 10, using the same conditions, product 12 was obtained (Scheme 4). The esters 8
and 9 were achieved by esterification of compounds 6f and 7f respectively, in acid conditions
by microwave irradiation (Scheme 4).
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
Insert Scheme 4
Effect of NPK and NPA Derivatives on the Phosphatase Activity of Purified
Recombinant Form of CDC25B. The evaluation of the inhibition properties exhibited by
the NPK and NPA derivatives was carried out through kinetic measurements of the
phosphatase activity sustained by the recombinant form of CDC25B. In particular, the
hydrolysis rate of the synthetic substrate 3-O-methylfluorescein phosphate (OMFP) triggered
by CDC25B was measured at different substrate concentrations, in the absence or in the
presence of various inhibitor concentrations. The resulting data of initial velocity (v ) were
i
analyzed in Lineweaver-Burk plots, thus allowing an inspection on the inhibition mechanism
possessed by the novel compounds. As reported in Table S1 (Supporting information), the
affinity for the substrate OMFP measured in the absence of inhibitor (K = 2.7  0.2 µM),
M
remained essentially unvaried in the presence of some inhibitors, decreased with other
inhibitors or slightly increased with the remaining ones. On the other hand, the maximum
velocity of OMFP hydrolysis measured in the absence of inhibitor (Vmax = 323  10 AU/min)
significantly decreased in the presence of the indicated concentrations of all synthesized
compounds. These findings differ from the data already reported for NSC28620, because this
compound caused an increase of the K without any variation of the Vmax, a typical behavior
M
3
8
for a competitive inhibition of CDC25B. Therefore, none of the analyzed compounds may
be ranked as a competitive inhibitor of CDC25B; probably, the addition of a bulky benzene
1
0
ACS Paragon Plus Environment

Page 11 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
ring in all derivatives causes a variation in the binding mode of the inhibitors to CDC25B.
However, this different interaction drives in some cases to a stronger inhibition of CDC25B
(Table 1). An inspection of the kinetic data allowed a putative assignment of the compounds
to different categories of inhibition mechanism. Hence, inhibitors were classified as non-
competitive (Figure S1 of Supporting Information), un-competitive (Figure S2 of Supporting
Information) or mixed (Figure S3 of Supporting Information) on the basis of the comparison
of the Lineweaver-Burk plots obtained in the presence of different inhibitor concentrations
with that obtained in the absence of inhibitor. For instance, as depicted in Figure 3A, 7j
appears as a typical non-competitive inhibitor, because the progressive decrease of the Vmax at
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
increasing inhibitor concentration occurs without a significant change of the K for OMFP
M
(see also Table S1). On the other hand, 3 exhibits a kinetic behaviour approaching that of an
un-competitive inhibitor, because the progressive decrease of Vmax appears to be concomitant
to a roughly similar decrease of the K (Figure 3B and Table S1).
M
Insert Figure 3
This kinetic approach allowed the calculation of the inhibition constant (K ) for each
i
derivative on the basis of the varied kinetic parameters resulting from the inhibition. The Ki
values (Table 1) are comprised in a large interval ranging from 0.8 µM to 69 µM, thus
indicating that the inhibition properties of this series of compounds are dependent on the
modifications realized through the optimization program. In particular, some derivatives
(such as 7j, 7i, 6e, 7f or 3) are endowed with a significantly stronger inhibition potency
compared to that of the initial lead NSC28620 (K = 5.3 µM), whereas the remaining
i
compounds display a comparable or lower inhibition.
Insert Table 1
Most of the CDC25B inhibitors, including the lead compound NSC28620, are also active
against CDC25A and CDC25C.27, 38 Hence, an evaluation of the inhibition properties of some
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 90
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
NPA derivatives towards the recombinant forms of both CDC25A and -C was considered. In
particular, we have checked the inhibition potency of the most potent non-competitive and
un-competitive inhibitor, i.e. 7j and 3, respectively. Using the same approach and
experimental conditions adopted for CDC25B, the kinetic parameters K and V measured
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
M
max
for CDC25A in the absence of inhibitors were 2.0 µM and 217 AU/min, respectively; these
values were quite similar to those obtained for CDC25B. On the other hand, the
corresponding K and V
measured for CDC25C were 16 µM and 182 AU/min, thus
max
M
indicating that this CDC25 form exhibits a lower catalytic efficiency, essentially due to a
lower affinity for the synthetic substrate OMFP. The kinetic measurements obtained in the
presence of the inhibitors 7j and 3 indicated that both compounds were active also on
CDC25A and -C. In particular, the K of 7j towards CDC25A and -C was 0.6 µM and 4.0
i
µM, respectively, whereas the corresponding K of 3 was 0.6 µM and 0.4 µM, respectively. In
i
the complex, the analysis of the inhibition parameters related to the three CDC25 forms could
reflect slight differences in the binding mode of substrate and/or inhibitor, as already reported
for other CDC25 inhibitors.27, 38
SAR Investigation. The biochemical evaluation of the inhibitory effects of the thirty-one
compounds against CDC25B allowed a comprehensive description of the SAR for both NPK
and NPA scaffolds. Derivatives bearing a carbonyl group as linker were all less active than
the lead compound NSC28620 (1, 1a, 1c and 2), with only compound 1b showing a nearly
comparable activity. Indeed, the replacement of the carbonyl group with a secondary amino
group in the X position caused an overall improvement of the inhibition power of this series
of derivatives, as emerging from the comparison of the K values of 1 vs 6 (69 vs 22 µM), 1a
i
vs 6a (36 vs 16.9 µM), 1c vs 6d (34 vs 5.1 µM), or 2 vs 3 (54 vs 2.8 µM).
1
2
ACS Paragon Plus Environment

Page 13 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Then, we analyzed the effects observed for the compounds bearing the carboxylic acid
5
moiety at R (derivatives 7 and 7a,b,d-f,i,j). The unsubstituted derivative 7, as well as
1
2
compounds bearing the methoxy group at R (7a) or R (7b) showed poor inhibitory activity
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
(
K 30, 55 or 45 µM in 7, 7a or 7b, respectively). Instead, the introduction of polar
i
substituents caused an interesting increase in inhibition potency. Indeed, the presence of the
1
2
3
hydroxyl group at R (7d), R (7e), and even more at R position (7f) significantly enhanced
the inhibition power, as reflected by the decrease of the K value in the order: 30 µM in 7 >>
i
5
.8 µM in 7d ∼ 6.4 µM in 7e > 2.9 µM in 7f. Furthermore, the introduction of the amino
2
3
group at R in 7i and even more at R in 7j provoked a further increase of the inhibition
power (1.4 µM in 7i and 0.8 µM in 7j), leading to the most potent compound of the whole
series of inhibitors. Moreover, 7j was 7-fold more active than the initial lead NSC28620.
Therefore, the presence of one polar group (hydroxyl or amino) on the naphthalene system of
5
inhibitors carrying a carboxylic group at R significantly increased their capability of
inhibiting the CDC25B phosphatase activity.
A somehow similar trend was observed when the free carboxylic moiety was moved at
4
position R (derivatives 6, 6a,b,d-f, 12). The unsubstituted derivative 6 had a low inhibition
activity (K 22 µM), similar to that obtained after the introduction of the methoxy group at
i
1
2
position R in 6a (K 16.9 µM); however, when this group was moved at R in 6b, a
i
remarkable increase of the inhibition potency was observed (K 6.1 µM). Instead, the
i
2
presence of the hydroxy group at R in 6e conferred the highest potency in this series (K 2.7
i
1
3
µM), but shifting this group at R in 6d or even more at R in 6f caused a decrease in potency
3
(
K 5.1 µM in 6d and 13.4 µM in 6f). Instead, derivative 12, where position R was occupied
i
by an amino group, showed an improved inhibition (K 6.5 µM) compared to 6d.
i
4
5
In some derivatives the esterification of the carboxylic group at R (3, 3a-b, 8, 10) or R
(
4, 4a-b, 9, 11) position was performed in order to increase the lipophilicity of the compound.
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 90
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
4
Among the ester series at R , the unsubstituted derivative 3 showed the highest potency (K
i
2
.8 µM). No improvement of the inhibition power was obtained with the introduction of a
1
2
methoxy group at position R in 3a or R in 3b, or with the presence of a polar substituent,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
such as hydroxyl or amino group at position R in 8 or 10, respectively. Instead, the K value
i
of the substituted derivatives raised to 12.5 µM in 3a, or slightly increased to 6.9 µM in 3b,
7
.4 µM in 8 or 9.9 µM in 10. Therefore, the activity of these compounds is not improved by
the presence of different substituents on the naphthalene moiety.
5
Moving to the ester function at R , in four out of five derivatives the inhibition power
4
was essentially similar to that exhibited by the corresponding ester analogues at R . In
particular, the unsubstituted derivative 4 displayed a slightly lower potency (K 7.3 µM) than
i
1
2
the analogue 3. Derivatives bearing the methoxy group at R (4a) or R (4b) showed potency
in the same range (8.5 and 11.9 µM, respectively) of the analogues 3a and 3b. A slight
improvement of the potency (K 4.9 µM) was obtained upon the introduction of the hydroxyl
i
3
group at R in 9; in contrast, the presence of the amino group at this position was,
surprisingly, detrimental (K 36 µM in 11).
i
In conclusion, the presence of the NPA scaffold appears to be sufficient to attain the best
4
5
inhibition, when the carboxylic group at R or R was esterified.
Intrinsic Fluorescence of the CDC25B Catalytic Domain Upon the Addition of Typical
NPA Derivatives. The amino acid sequence of the catalytic domain of CDC25B includes one
tryptophan (W550) and eleven tyrosine residues (Y382, Y400, Y428, Y430, Y432, Y497,
2
2
Y501, Y502, Y506, Y512 and Y528). In particular, W550 is located in the protein region
known as “swimming pool” for the great number of water molecules found in the crystal
structure.22 These findings prompted an investigation on the intrinsic fluorescence of the
purified recombinant CDC25B; Figure 4 shows the emission spectrum obtained at the
1
4
ACS Paragon Plus Environment

Page 15 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
excitation wavelength of 280 nm (black line). The emission maximum at 357 nm is likely due
to the intrinsic fluorescence of W550, whereas the shoulder around 310 nm is probably due to
the emission of the eleven tyrosine residues. This hypothesis was confirmed by the emission
spectrum recorded at 295-nm excitation wavelength. Under this condition, a drastic reduction
of the tyrosine emission was obtained and, as predicted, the shoulder at 310 nm essentially
disappeared, whereas the emission maximum at 357 nm was still present, although reduced in
quantum yield (not shown). These findings indicate a clear distinction between tryptophan
and tyrosine emission; furthermore, the position of the emission peak at 357 nm is typical of
the great exposure of W550 to the water environment, thus suggesting that this residue lays
on the edge of the swimming pool.
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
The effect of the inhibitors on the intrinsic fluorescence of CDC25B was investigated
by adding some NPA derivatives causing a measurable inhibition of the CDC25B
phosphatase activity. In particular, compounds 7j and 7i, typical non-competitive inhibitors,
were added at 1.5 µM and 3 µM respectively, whereas the un-competitive inhibitors 3 and 4a
were added at 4 µM and 10 µM, respectively. The chosen concentrations reflected the
different inhibition potency of the various inhibitors. As shown in Figure 4, 7j (purple line)
and 7i (green line) caused an evident quenching of the fluorescence yield, without shifting the
emission maximum at 357 nm. On the other hand, the addition of 3 (red line) or 4a (blue line)
was ineffective in altering the CDC25B emission maximum. A possible explanation for the
different behavior of the inhibitors is hereafter reported. Indeed, the non-competitive
inhibitors 7j and 7i could directly bind to CDC25B and this interaction caused a
conformational change in the region containing W550. On the other hand, the un-competitive
inhibitors 3 and 4a could be unable to directly bind to the enzyme, because binding of these
compounds should require the preformed enzyme•substrate complex. Obviously, the
substrate OMFP was omitted in fluorescence experiments, owing to its fast hydrolysis.
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 90
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
Insert Figure 4
Docking Studies. Among the thirty-one synthesized derivatives, compounds 7j and 3 are
representative members of a different binding interaction with CDC25B, as emerging from
the kinetic and fluorescence studies. Extensive efforts trying to co-crystallize CDC25B with
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
7
j and 3 or to soak 7j and 3 into crystals of CDC25B were unsuccessful. We next carried out
2
2
docking experiments into the CDC25B crystal structure (PDB ID: 1QB0) in an attempt to
determine the binding mode of these two NPA derivatives and to rationalize the observed
SAR data. Under this concern, we have the information of the kinetic data, indicating that 7j
and 3 are non-competitive and un-competitive inhibitor, respectively. This means that 7j and
3 must, by definition, be able to bind in a site distinct from the substrate active site and leave
the active site unblocked. However, 7j should equally react with free enzyme or the
enzyme•substrate complex, whereas 3 should react only with the complex. In addition,
several CDC25B inhibitors identified by other groups have been proposed to bind into the
swimming pool,23, 34, 53-55 that is a deep and concave pocket between the C-terminal helix A
and the P-loop and does not face the catalytic site directly. Another useful information for
docking experiments comes from the fluorescence studies, suggesting that 7j directly binds to
CDC25B in a region affecting the emission of W550, the residue located in the C-terminal
helix lining the swimming pool. Vice versa, 3 seems to be unable to directly bind to the free
enzyme. Hence, our docking studies were focused on the swimming pool site.
Analysis of the available crystal structures of the native catalytic domain of CDC25B
(
Figure S4 of Supporting Information) showed that several residues of the swimming pool
W550, R548, R544, F543, L540, N532, Y428, E446, R482), as well as residues that
(
surround the active site cleft exhibit much higher B-factors, which is indicative of their
higher flexibility. In particular, R482, which plays an important role in the catalytic activity
5
6
of the enzyme, showed the highest degree of mobility. Thus, 7j and 3 were docked to the
1
6
ACS Paragon Plus Environment

Page 17 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
structure of CDC25B using the Induced Fit Docking (IFD) protocol of the Schrödinger
suite,57, 58 that is intended to circumvent the inflexible binding site and accounts for the side
chain (Figure S4 of Supporting information) or backbone movements or both upon ligand
binding. Docking results were validated with the metadynamics method described by Clark et
al., which allows accurate and reliable predictions of protein•ligand binding poses at a
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
feasible computational cost.59
As illustrated in Figure 5A, the top-ranked solution for the most potent non-competitive
inhibitor 7j was deeply inserted into the swimming pool, with the carboxylic acid moiety
forming two strong salt-bridges with R482 and R544. This latter formed an additional cation-
π interaction with the naphthalene moiety of the ligand. Notably, both arginine residues are
involved in the recognition of the Cdk phosphate group and were previously described to be
critical for the binding of CDC25 inhibitors by both site-directed mutagenesis60 and
computational studies.55, 61 In addition, the participation of R482 and R544 in the binding of
CDC25B inhibitors was also revealed by the crystal structure of CDC25B in complex with a
6
2
modified peptide. The secondary amino group at position X of the NPA derivative engaged
3
a H-bond with the side chain of E446, whereas the amino group at position R formed
2
2
contacts with C426 (the “back-door cysteine”) through a H-bond with the CO backbone.
The inhibitor established further hydrophobic interactions with the side chains of M483,
Y428 and W550, as also confirmed through the fluorescence studies. In summary, 7j lies
within the swimming pool pocket and does not interfere with the substrate binding in the
active site, thus explaining the non-competitive nature of the inhibition, as determined
through kinetic analyses. Therefore, the interaction of the carboxylic acid moiety in 7j and its
congeners with R482 and R544 may explain the increased inhibitory potency towards
CDC25B. This model also suggests that the differences in potencies observed when the polar
amino group was moved to another position on the naphthalene moiety (7i), or replaced by
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 90
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the hydroxyl group in various positions (7d,e,f), could be due to a less efficient interaction
with residue C426.
Insert Figure 5
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
Next, we went on to characterize the binding mode of the most potent un-competitive
inhibitor 3. Taking into account that such inhibition mechanism implies the binding of the
inhibitor to the enzyme•substrate complex, a model of the artificial substrate OMFP bound to
CDC25B was first generated. Therefore, OMFP was docked to the structure of CDC25B
using the IFD protocol and the binding pose metadynamics was employed to select the most
stable pose consistent with the catalytic mechanism. As depicted in Figure 5B, the
phosphoryl group of the artificial substrate was found to bind within the active site pocket of
CDC25B, where it formed interactions with the residues of the P-loop containing the catalytic
cysteine (C473). The orientation of the phosphoryl group mimicked exactly the structure of
the bound sulfate, mimicking a bound phosphorylated substrate, in the CDC25B crystal
22
structure. The xanthene moiety of OMFP protruded into the swimming pool and formed H-
bonds with Y428 and T547 residues as well as a -cation interaction with R544, whereas the
spirobenzofuranone system was engaged in H-bonds with the side chain of R544. The
interaction between aromatic rings of OMFP and the M531, Y428, F543 and W550 side
chains further contributed to the binding through hydrophobic stabilization.
The newly generated CDC25B•OMFP complex was employed for further docking of 3.
The resulting top-ranked binding mode, which places 3 directly in the swimming pool near
the bound substrate OMFP, is consistent with its un-competitive mode of inhibition. In
particular, the naphthalene ring of 3 was sandwiched between R482 and the xanthene moiety
of OMFP, with this latter forming a -stacking interaction. In addition, 3 engaged further
hydrophobic interactions with M483 and L445, whereas the phenyl ring holding the ester
function at position 2 also stacked against the spirobenzofuranone ring of OMFP. The linker
1
8
ACS Paragon Plus Environment

Page 19 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
amino group and the carbonyl ester of the inhibitor was engaged in H-bonds with the CO
backbone and the side chain of S549, respectively, while the ethyl-ester group interacted with
the non-polar side chain carbons of R548.
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
Therefore, the upper wall of the swimming pool together with the substrate OMFP
provide a hydrophobic environment critical for binding of the bulky NPA scaffold of 3.
Indeed, the inhibition of CDC25B by 3 seems to occur when the binary enzyme•OMFP
complex is formed. Moreover, as evidenced by SAR data, the binding affinity of this series of
compounds is less influenced by the nature and position of substituents on the naphthalene
ring. Indeed, the unsubstituted 3 tightly fits in the hydrophobic cleft lined by the swimming
1
pool and OMFP, whereas the introduction of a methoxy group at R (3a) is disfavored due to
steric clashes with E478 and R479.
The lower affinity of 3 and its congeners compared to 7j seems to be due to the presence
of the ester function in place of the acidic one: in the first case the ester function forms only
weak hydrophobic interactions with the enzyme, while in the second case the free acidic
group is involved in strong salt-bridges with the key arginine residues within the swimming
pool. For these reasons, 3 adopts a different binding mode that generates a side-chain
reorientation into the swimming pool pocket, likely disrupting the network of interactions
critical for the catalytic activity of the enzyme.
Effects of the NPK and NPA Derivatives on Cell Growth Rate of Melanoma Cells. We
previously documented that the lead NSC28620 affected the cell cycle, increased the levels of
3
8
the inactive p-Cdk1, as well as reduced the cell viability of some cancer cell lines. We next
evaluated the effects of the novel NPK and NPA derivatives in the metastatic human
melanoma cell line A2058, a highly resistant and aggressive cancer type. In a preliminary
screening, the effect of these compounds, added at 50 μM, was assessed after different times
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 90
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
of treatment, using the MTT assay. After 24 h, only 4a caused an evident cytotoxic activity,
but when the analysis was prolonged, also another compound, i.e. 4, resulted cytotoxic after
48-h treatment (not shown).
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
To check the minimum concentration of 4 and 4a leading to a reduction of the cell
growth rate, the time-dependent effect of these compounds on A2058 melanoma cells was
tested at lower concentrations, i.e. 2.5 and 5 µM (Figure 6, panels A and B). The data indicate
that only 4a exerted a progressive reduction of cell viability. Indeed, after 48-h incubation,
the cell growth was significantly reduced in the presence of 5 µM 4a and, after 72-h
treatment, the reduction of cell viability was significant even at 2.5 µM (Figure 6B). We have
then evaluated if compound 4a exerted a cytotoxic effect even in normal cells, such as the
non-malignant human fibroblast cell line BJ-5ta (Figure 6C). The MTT assay, performed at
different concentrations and prolonged up to 72 h, clearly indicates that compound 4a was
not cytotoxic for this non-tumor cell line, even when present at 10 µM. It is known that
fibroblasts are consistently and ubiquitously present in any body tissue; interestingly, the lack
of cytotoxicity of 4a in BJ-5ta suggests that this compound could not cause unspecific side-
toxic effects in a normal cellular context.
Insert Figure 6
The cytotoxic effect of 4a in melanoma cells prompted us to extend the evaluation of its
inhibition power towards the recombinant forms of CDC25A and -C, besides that of -B.
Under this concern, the values of K determined for CDC25A and -C were 3.0 µM and 1.9
i
µM, respectively. These data indicate that 4a is active towards the three different forms of
CDC25. In the light of all these observations, compound 4a was chosen for further biological
studies to better investigate its cytotoxicity in melanoma cell lines. Notably, the melanoma, in
its late stage, represents one of the most aggressive tumors, characterized by a low
responsiveness to conventional anti-tumoral therapies.63 Therefore, 4a is particularly
2
0
ACS Paragon Plus Environment

Page 21 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
attractive as it is active at low concentrations in melanoma cells. Under this concern, we
should point out that the biological activity of the initial lead NSC28620 was observed only
at much higher doses, i.e. 200 µM, and in more responsive tumor cell lines.38
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
Insert Figure 6
To evaluate the long-term cell growth inhibitory effect, we determined if 4a could
suppress colony formation of melanoma cells. To this aim, A2058 and A375 melanoma cell
lines were exposed to increasing concentrations of 4a (0-10 µM) and colony formation was
monitored. As shown in Figure 7, 4a affected colony formation in a progressive and
remarkable manner. Indeed, the number of cell clones progressively decreased with 0.625,
1.25 and 2.5 µM 4a and was almost absent in the presence of 5 or 10 µM inhibitor in both
cell lines. Hence, the capability to form cell clones, an intrinsic characteristic of tumors
6
4
cells, was reduced when melanoma cells were treated with very low concentration of 4a.
Insert Figure 7
Effect of 4a on Cell Cycle Progression and p-Cdk1. A further investigation on the
mechanisms mediating the toxic action of 4a in melanoma cells was carried out. It is
noteworthy that CDC25 enzymes play a pivotal role in the regulation of cell cycle, as well as
1
5
in the control of DNA damage. Hence, the inhibitory potential of 4a towards the CDC25B
phosphatase activity suggests that this compound could lead to a cell cycle arrest in cultured
cells. To evaluate the effect of 4a on cell cycle progression, melanoma cells were exposed to
1
0 µM 4a for 8- and 16-h and the cell cycle phase distribution was cytofluorimetrically
monitored after PI staining for either A2058 (Figure 8) or A375 (Figure 9) cells. Compared to
untreated cells, a significant increase of the G2/M phase associated to a reduction of G0/G1
phase was detected after 8-h treatment with 4a in both A2058 (Figure 8A) and A375 (Figure
9
A) cells. When the incubation was prolonged up to 16-h, a somehow different response
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 90
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
emerged from the two cell lines. In A2058 cells, a further increase of cell percentage arrested
in G2/M was evident, as well as an ulterior decrease of cells in G0/G1 phase, together with a
small not-significant increase of the sub G1 phase (Figure 8B). On the other hand, in A375
cell lines the number of cells arrested in G2/M phase was similar between treated and
untreated cells, the reduction of cell in G0/G1 phase was still evident but less pronounced
compared to 8 h treatment, but interestingly, the increase of sub G1 phase became greatly
significant, compared to untreated cells (Figure 9B). These results could indicate that the
arrest in G2/M represents a cellular response to the toxic effect of 4a, perhaps mediated by
the CDC25 inhibition; however, this effect could evolve in cell death after a prolonged
treatment, as suggested by the behavior of A375 cells. The significant increase of a sub G1
phase observed in A375 melanoma cells after 16-h treatment is probably due to the derivation
of this cell line from a primary tumor, whereas A2058 is a metastatic tumor cell line; this
different histological origin could explain the greater responsiveness of A375 cells to the
drug treatment.
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
Insert Figures 8 and 9
The arrest in G2/M of both melanoma cell lines during treatment with 4a could be
related to the inhibitory capacity of this molecule toward CDC25 phosphatase activity also in
a cellular system. It is known that CDC25 enzymes regulate cell cycle progression by
modulating the activity of Cdks, which represent their main targets.65, 66 In particular, the
phosphorylation of different substrates by Cdk1 is responsible for the modulation of many
biochemical events involved in the entry to mitosis. Under this concern, a critical event for
the activation of Cdk1 is the dephosphorylation of its inactive form, p-Cdk1, mediated by
CDC25.67-69 Therefore, we verified if the treatment of melanoma cells with 4a could affect
the protein levels of p-Cdk1. Indeed, as shown in Figure 10, an increase of p-Cdk1 became
evident in both melanoma cell lines after a short incubation time (8 h) with 4a. Hence, the
2
2
ACS Paragon Plus Environment

Page 23 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
higher protein levels of p-Cdk1 found in melanoma with respect to untreated cells could
represent a further evidence of the inhibitory effect exerted by 4a on CDC25 enzymatic
activity in these cellular systems.
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
Insert Figure 10
CONCLUSIONS
Over the last years, many efforts have been devoted to the discovery of CDC25 inhibitors
featuring novel chemotypes; however, only a limited number of literature reports have
described the identification of inhibitors based on non-quinoid structures, although these
compounds could be more attracting because intrinsically endowed with lower side-toxic
effects.
In the present study we described the design, the synthesis, the inhibition mechanism, as well
as the biological effects of novel NPK and NPA derivatives as inhibitors of CDC25B, that
were designed starting from the lead compound NSC28620, recently reported by us. Different
synthetic pathways allowed the exploration of the effects of various substituents on the
naphthalene moiety, as well as the role of the carboxylic function, leading to the
identification of the main structural requirements for the inhibitory activity. Interestingly, the
structural modifications had a profound impact on the inhibition mechanism: indeed, while
the lead compound NSC28620 showed a competitive inhibition, the newly designed
inhibitors displayed a non-competitive, or un-competitive, or mixed behavior. Some
derivatives (7j, 7i, 6e, 7f and 3) showed higher inhibitory activity than the initial lead
NSC28620, with the most potent inhibitor 7j being 7-fold more active (K of 0.8 µM). IFD
i
simulations helped to rationalize the molecular basis of the activity of this series of
derivatives, which bind within the swimming pool pocket, as also demonstrated by kinetic
analysis and intrinsic fluorescence studies. Moreover, the interaction of the carboxylic acid
moiety of 7j and its congeners with the key residues R482 and R544 accounts for the
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 90
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
increased inhibitory potency towards CDC25B. On the other hand, the un-competitive
derivatives, such as 3, tightly fit in the hydrophobic cleft formed by the substrate OMFP and
the swimming pool. These compounds were characterized for their cellular activity in the
highly aggressive human melanoma cell lines A2058 and A375. In particular, compound 4a
was the most effective one in the inhibition of cell proliferation and colony formation, also
causing an increase of the G2/M phase and a reduction of G0/G1 phase of the cell cycle in
both cell lines. Importantly, 4a induced an increase of p-Cdk1, providing a further evidence
of the inhibitory effect on CDC25B. Furthermore, it is important to note that this compound
exhibits a cytotoxic effect at low concentration in melanoma cells, an important issue for the
possible reduction of undesired side effects. Our findings not only demonstrate the feasibility
of exploiting CDC25s as a therapeutic target for melanoma, but also suggest that the NPA
scaffold represents an attractive starting point for further structural elaboration to achieve
higher affinity, and more specific inhibition.
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
4
ACS Paragon Plus Environment

Page 25 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
EXPERIMENTAL SECTION
Reagents and Chemicals. Reagents and solvents were purchased from Sigma-Aldrich
and used without further purification. Some reactions involving air-sensitive reagents were
performed under nitrogen atmosphere and anhydrous solvents were used when necessary.
The Biotage Initiator 2.0 microwave synthesizer was used. Reactions were monitored by thin
layer chromatography analysis on aluminum-backed Silica Gel 60 plates (70-230 mesh,
Merck), using an ultraviolet fluorescent lamp at 254 nm and 365 nm. Visualization was aided
by opportune staining reagents. Purification of intermediates and the final compounds was
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
1
13
performed by flash chromatography using Geduran® Si 60 (40-63 µm, Merck). H- and C-
NMR spectra were recorded at room temperature on a Varian Oxford 300 MHz instrument
using TMS as internal standard. CDCl , CD OD, acetone-d and DMSO-d were used as
3
3
6
6
deuterated solvents for all the spectra run. Chemical shift are expressed as δ (ppm).
Multiplicity is reported as s (singlet), br s (broadened singlet), d (doublet), t (triplet), q
(quartet), m (multiplet), dd (doublet of doublets), dt (doublet of triplets). The coupling
constants (J-values) are given in Hertz (Hz). All spectroscopic data match the assigned
structures. The melting points were determined on a Buchi Melting Point B540 instrument.
MS analyses were carried out with a Thermo Finnigan LCQ Advantage system, equipped
with a quaternary pump, a Diode Array Detector (working wavelength: 254 nm) and a MS
spectrometer, with an Electrospray ionization source and an Ion Trap mass analyzer
(ionization: ESI positive or ESI negative; capillary temperature: 250°C; source voltage: 5.50
kV; source current: 4.00 µA; multipole 1 and 2 offset: -5.50 V and -7.50 V, respectively;
intermultipole lens voltage: -16.00 V; trap DC offset voltage: -10.00 V). Elemental analyses
(CHN) are within ± 0.40% of theoretical values. The results of elemental analyses, carried
out with a EurovectorEuro EA 3000 model analyzer, indicated that the purity of all tested
compounds was higher than 95%.
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 90
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
General Procedure A. To a stirred mixture of the proper lactone (0.172 mmol) in 25%
KOH (1 mL) and pyridine (0.5 mL), powdered KMnO (43.5 mg, 0.275 mmol) was added,
4
and the mixture was heated at reflux for 5 hours. The hot mixture was filtered and the residue
was washed with water. Then, the filtrate was extracted with diethyl ether (2×2 mL) to
remove the unreacted lactone. The aqueous phase was acidified with 3N HCl (2 mL),
extracted with ethyl acetate (3×2 mL), washed with water and the organic layer was dried
over anhydrous sodium sulfate. The organic solvent was evaporated under vacuum to give a
crude oil, which was purified by flash chromatography eluting firstly with cyclohexane/ethyl
acetate (80:20) and then (10:90) to afford the desired compounds.
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
General Procedure B. To a solution of the suitable bromobenzoic acid (2.487 mmol) in
ethanol (1 mL), a catalytic amount of concentrated sulfuric acid was added dropwise. The
resulting mixture was kept under stirring at reflux overnight and, after cooling, the solvent
was evaporated under vacuum. Ethyl acetate (3 mL) was added, and the organic layer was
firstly washed with a saturated solution of sodium bicarbonate (1×2 mL) and then with water
(
2×2 mL). The organic solvent was dried over anhydrous sodium sulfate and evaporated.
Purification with flash chromatography (eluent: cyclohexane/ethyl acetate 90:10) was
performed.52
General Procedure C. Procedure C1. A dried flask was purged with argon, charged
with (±)-BINAP (3% mmol), and then toluene (2.7 mL) was added. The mixture was heated
to 80°C under stirring until the BINAP was dissolved. The solution was cooled to room
temperature, and Pd(OAc) (2% mmol) was added. The mixture was stirred at room
2
temperature for 1 minute, then a solution of the proper 1-bromonaphthalene (1.0 mmol) and
the corresponding ethyl aminobenzoate (0.8 mmol) in toluene (0.5 mL) was added. Finally,
Cs CO (1.14 mmol) and other toluene (2 mL) were added to the reaction mixture. The
2
3
solution was heated at 80°C under stirring; then, it was cooled to room temperature, diluted
2
6
ACS Paragon Plus Environment

Page 27 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
with diethyl ether, filtered and concentrated. The obtained crude product was purified by
column chromatography to provide the desired adduct.49
Procedure C2. A dried flask was purged with argon, charged with (±)-BINAP (3%
mmol), and then toluene (2.7 mL) was added. The mixture was heated to 80°C under stirring
until the BINAP was dissolved. The solution was cooled to room temperature, and Pd(OAc)2
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
(2% mmol) was added. After stirring the mixture at room temperature for 1 minute, a solution
of the suitable ethyl bromobenzoate (1.0 mmol) with the appropriate naphthylamine (0.8
mmol) in toluene (0.5 mL) was added. Cs CO (1.14 mmol) was introduced within the
2
3
mixture and toluene (2 mL) was added. The mixture was heated to 80°C under stirring; then,
it was cooled to room temperature, diluted with diethyl ether, filtered, and concentrated. The
crude product was purified by column chromatography to obtain the desired adduct.49
General Procedure D. To a solution of the suitable ethyl-(naphthylamino)-benzoate
(0.15 mmol) in 2 mL of THF/EtOH (1:1), 1M NaOH (0.375 mmol) was added. The resulting
mixture was refluxed for a different period of time, depending on the substrate. The solvent
was evaporated, and then water (2 mL) was added. The aqueous phase was firstly washed
with diethyl ether in order to purify the reaction mixture from the unreacted starting material,
and then treated with 1N HCl (0.5 mL). The acid was extracted with ethyl acetate (3×1 mL),
the organic phase was dried over anhydrous sodium sulfate and the solvent was evaporated
under reduced pressure. The crude product was purified by flash chromatography eluting
with dichloromethane/methanol (95:5) to afford the desired derivatives.50
General Procedure E. To a solution of the proper methoxynaphthalene (0.180 mmol) in
dry dichloromethane (2 mL), BBr (1M in dichloromethane, 0.945 mmol) was added
3
dropwise under N . The reaction was carried out at room temperature for all the synthesized
2
compounds. The starting solution was firstly cooled at -78°C and BBr was carefully added
3
dropwise to the reaction mixture. After 20 minutes, the resulting solution was warmed to
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 90
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
room temperature and the stirring was continued for further 3 hours. The reaction was then
quenched at 0°C with water (4 mL), and extracted with ethyl acetate (3×2 mL). The
combined organic phases were washed with brine (3×1 mL) and dried over anhydrous
sodium sulfate. The subsequent filtration and evaporation of the organic solvent provided a
crude product, which was purified by flash chromatography eluting firstly with
dichloromethane/methanol (95:5) and then with dichloromethane/ methanol (80:20) to afford
the final naphthol.49
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
General Procedures F. Procedure F1. To an ice-cold solution of the proper tert-butyl-
naphthylcarbamate (1.67 mmol) in dry 1,4-dioxane (1 mL), HCl (4M in 1,4-dioxane, 1 mL)
was added dropwise, under N . The reaction was stirred at room temperature for 24 hours.
2
After completion of the reaction, monitored by TLC, diethyl ether was added to the reaction
mixture and it was stirred. After cooling at 0°C, the formed precipitate was collected by
filtration, washed with cold diethyl ether and dried to afford the hydrochloride salt of
compound 7i or 7j.
Procedure F2. Trifluoroacetic acid (0.65 mmol) was added to a solution of the suitable
tert-butyl-naphthylcarbamate (0.13 mmol) in dichloromethane (1 mL) and the reaction was
stirred at room temperature for 60 min. The solvent was removed in vacuo; then, a 2M
sodium hydroxide solution (2 mL) was added and extracted with ethyl acetate (3×1 mL). The
organic layers were dried over anhydrous sodium sulfate and the solvent was evaporated
under reduced pressure. Products 10 and 11 were isolated as hydrochloride salts by treatment
with 1.25 M HCl in methanol.
General Procedure G. To a solution of the opportune benzoic acid (2.487 mmol) in
ethanol (1 mL), a catalytic amount of concentrated sulfuric acid was added dropwise. The
resulting mixture was stirred under microwave irradiation for 90 min at 80°C, and then the
solvent was evaporated under reduced pressure. Ethyl acetate (3 mL) was added, and the
2
8
ACS Paragon Plus Environment

Page 29 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
organic layer was firstly washed with a saturated solution of sodium bicarbonate (1×2 mL)
and then with water (2×2 mL). The organic solvent was dried over anhydrous sodium sulfate
and evaporated under vacuum. The crude product was purified by flash chromatography.52
General Procedure H. To a solution of the required bromonaphthol (0.672 mmol) in dry
N,N-dimethylformamide (1 mL) sodium hydride (60% dispersion in mineral oil, 29.6 mg,
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
0
.739 mmol) was added under N . After the evolution of H , iodomethane (0.17 mL, 2.688
2 2
mmol) was added, and the reaction mixture was stirred at room temperature for 3 hours. After
the evaporation of N,N-dimethylformamide, the resulting residue was extracted with ethyl
acetate (3×2 mL). The collected organic phase was dried over anhydrous sodium sulfate and
the solvent was evaporated under reduced pressure to give the desired derivatives. The crude
oils were purified by flash chromatography (cyclohexane/ethyl acetate 90:10) to provide
compounds 16a and 16b.
General Procedure I. To a solution of the proper bromomethoxynaphthalene (0.422
mmol) in anhydrous THF (4 mL) cooled to -78°C under N was added n-BuLi (2.7M in
2
heptane, 0.39 mL, 1.055 mmol) and the resulting mixture was stirred for 1 hour. Anhydrous
N,N-dimethylformamide (0.20 mL, 2.532 mmol) was added, and the reaction was stirred for a
further 1 hour at -78°C. The reaction was quenched by a slow addition of cold water (0.5
mL), in order to neutralize the unreacted n-BuLi, and THF and DMF were evaporated. The
residue was poured into 3N HCl (3 mL) and extracted with diethyl ether (6×2 mL). The
combined organic layers were washed firstly with water, then with a saturated sodium
hydrogen carbonate solution and, finally, with a saturated sodium chloride solution. The
organic phase was dried, filtered and concentrated. The crude oils were purified by flash
chromatography (cyclohexane/ethyl acetate 9:1) to provide compounds 17a and 17b.
General Procedure J. To a solution of compound 19 (0.787 mmol) in dry diethyl ether
(1 mL) at -78°C was added dropwise 0.58 mL of a 2.7 M solution of n-BuLi in heptane
2
9
ACS Paragon Plus Environment