Design, Synthesis, and Physicochemical and Pharmacological Profiling of 7-Hydroxy-5-oxopyrazolo[4,3-b]pyridine-6-carboxamide Derivatives with Antiosteoarthritic Activity in Vivo

Article abstract of DOI:10.1021/acs.jmedchem.0c00595

The hallmark of joint diseases, such as osteoarthritis (OA), is pain, originating from both inflammatory and neuropathic components, and compounds able to modulate the signal transduction pathways of the cannabinoid type-2 receptor (CB2R) can represent a helpful option in the treatment of OA. In this perspective, a set of 18 cannabinoid type-2 receptor (CB2R) ligands was developed based on an unprecedented structure. With the aim of improving the physicochemical properties of previously reported 4-hydroxy-2-quinolone-3-carboxamides, a structural optimization program led to the discovery of isosteric 7-hydroxy-5-oxopyrazolo[4,3-b]pyridine-6-carboxamide derivatives. These new compounds are endowed with high affinity for the CB2R and moderate to good selectivity over the cannabinoid type-1 receptor (CB1R), associated with good physicochemical characteristics. As to the functional activity at the CB2R, compounds able to act either as agonists or as inverse agonists/antagonists were discovered. Among them, compound 51 emerged as a potent CB2R agonist able to reduce pain in rats carrying OA induced by injection of monoiodoacetic acid (MIA).

Full text of DOI:10.1021/acs.jmedchem.0c00595

Subscriber access provided by Uppsala universitetsbibliotek
Article
Design, Synthesis, Physicochemical and Pharmacological
Profiling of 7-Hydroxy-5-oxopyrazolo[4,3-b]pyridine-6-
carboxamide Derivatives with Anti-Osteoarthritic Activity in Vivo
Claudia Mugnaini, Magdalena Kostrzewa, Marta Bryk, Ali Mokhtar Mahmoud, Antonella Brizzi, Stefania
Lamponi, Gianluca Giorgi, Francesca Ferlenghi, Federica Vacondio, Paola Maccioni, Giancarlo
Colombo, Marco Mor, Katarzyna Starowicz, Vincenzo Di Marzo, Alessia Ligresti, and Federico Corelli
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c00595 • Publication Date (Web): 09 Jun 2020
Downloaded from pubs.acs.org on June 10, 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 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design, Synthesis, Physicochemical and Pharmacological Profiling of 7-Hydroxy-5-
oxopyrazolo[4,3-b]pyridine-6-carboxamide Derivatives with Anti-Osteoarthritic Activity in
Vivo
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
Claudia Mugnaini,†,¶,* Magdalena Kostrzewa,‡,#,¶ Marta Bryk, Ali Mokhtar Mahmoud, Antonella
§
‡
†
†
†
||
||
Brizzi, Stefania Lamponi, Gianluca Giorgi, Francesca Ferlenghi, Federica Vacondio, Paola


||
§
‡
Maccioni, Giancarlo Colombo, Marco Mor, Katarzyna Starowicz, Vincenzo Di Marzo, Alessia
‡
,*
†
Ligresti, and Federico Corelli
†Department of Biotechnology, Chemistry and Pharmacy, University of Siena, Via Aldo Moro 2,
3100 Siena, Italy
5
‡Endocannabinoid Research Group, Institute of Biomolecular Chemistry, National Research Council
of Italy, Via Campi Flegrei 34, 80078 Pozzuoli (Napoli), Italy
#Institute of Genetics and Biophysics, National Research Council of Italy, Via Pietro Castellino 111,
8
0131 Napoli, Italy
§Department of Neurochemistry, Institute of Pharmacology, Polish Academy of Sciences, ul. Smetna
1
2, 31-343 Cracow, Poland
||Department of Food and Drug, University of Parma, Parco Area delle Scienze 27/A, 43124 Parma,
Italy
Institute of Neuroscience, National Research Council of Italy, S.S. 554, km 4,500, 09042 Monserrato
(Cagliari), Italy
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 68
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
The hallmark of joint diseases, such as osteoarthritis (OA), is pain, originating from both
inflammatory and neuropathic components, and compounds able to modulate the signal transduction
pathways of the cannabinoid type-2 receptor (CB2R) can represent a helpful option in the treatment
of OA. In this perspective, a set of 18 cannabinoid type-2 receptor (CB2R) ligands was developed
based on an unprecedented structure. With the aim of improving the physicochemical properties of
previously reported 4‑hydroxy-2-quinolone-3-carboxamides, a structural optimization program led to
the discovery of isosteric 7-hydroxy-5-oxopyrazolo[4,3-b]pyridine-6-carboxamide derivatives.
These new compounds are endowed with high affinity for the CB2R and moderate to good selectivity
over the cannabinoid type-1 receptor (CB1R), associated with good physicochemical characteristics.
As to the functional activity at the CB2R, compounds able to act either as agonists or inverse
agonists/antagonists were discovered. Among them, compound 51 emerged as a potent CB2R agonist
able to reduce pain in rats carrying OA induced by injection of monoiodoacetic acid (MIA).
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
Keywords
CB2 receptor
CB2-Selective ligands
Osteoarthritis
Pain
Isosterism
2
ACS Paragon Plus Environment

Page 3 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Painful and disabling pathologies shared by most of the people worldwide include rheumatic and
1
joint diseases, whose prevalence is expected to rise mainly because the world population grows older
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
and medical advances lengthen life expectancy. Osteoarthritis (OA), a prevalent disease accompanied
by chronic, debilitating pain, is the most common form of arthritis and a leading cause of disability
2
for older individuals in developed countries, where the increasingly higher incidence of obesity is an
important factor besides ageing. It is estimated that 25% of individuals affected by osteoarthritis are
3
prevented from performing daily activities. OA-induced pain is related to activation and release of
local pro-inflammatory mediators and is accompanied by the destruction of tissue. However,
clinically there is often disparity between the degree of pain perception and the extent of joint changes
in subjects with OA.4
Nowadays, a growing body of evidence shows that the development of damage to joints and bones,
as well as joint-related functional impairment, involve both local and, even more, systemic
5
6
inflammation. The hallmark symptom of joint diseases is pain, which originates from both
inflammatory and neuropathic components, the latter seemingly being the major determinant of the
intensity and quality of pain, although specific nerve lesions have not yet been identified in these
6
pathologies. Whilst it is well established that osteoarthritis is a heterogeneous condition with a
7
variety of pathophysiologic drivers leading to multiple phenotypes, the pathogenesis of this disease
is an intricate multifactorial process not fully clarified yet. In this scenario, the existence of a robust
cross-talk between cartilage and subchondral bone as well as the occurrence of alterations of all
junction structures (cartilage, bone, synovium, matrix, nerve endings, and blood vessels) is widely
8
recognized. In joint diseases, dysregulation and persistent activation of mast cells play a role at
different levels, by altering the balance between cartilage and subchondral bone, promoting
9
degenerative processes, and favoring the development of spinal neuroinflammation. Among the
inflammatory mediators involved in this pathological mechanism, interleukin- (IL-) 1 has been
recognized as an important initiator of OA pathophysiological development.10
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 68
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
1
The therapeutic options currently available for treatment of osteoarthritis remain limited. The
traditional pharmacological intervention addressing chronic pain in this condition basically relies on
symptomatic use of paracetamol and nonsteroidal anti-inflammatory drugs. There is an urgent need
for effective osteoarthritis pharmacotherapy, able to prevent joint degeneration, to delay disease
progression, and to restore the organ functionality in affected individuals.12 Accordingly, a
convincing approach to address effectively this problem focuses on the development of disease-
modifying drugs, and compounds targeting cannabinoid receptors and peroxisome proliferator-
activated receptors (PPARs) represent a potential answer to this therapeutic need.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
4
5
6
7
8
9
0
Type-1 cannabinoid receptors (CB1R) and type-2 cannabinoid receptors (CB2R) are the main
receptors of the endocannabinoid system that have been identified so far.13 Given their wide
distribution in the body, these CBRs are implicated in several physiological and pathological
processes and can be regarded as intriguing targets for potential pharmacotherapies in a number of
1
4
disorders. In this context, CBRs draw special attention, as they play a key role in bone formation,
resorption, and growth. In particular, both CBRs are present in the skeleton, with CB1R expressed
on nerve endings, whereas CB2R is found in osteoblasts, osteocytes, and osteoclasts.15,16 Under
1
7
pathological conditions, hypertrophic chondrocytes express both receptors, while synovial tissue
18
fibroblasts of individuals affected by rheumatoid arthritis and immune system cells like mast cells
mainly express CB2R. Both CB1 and CB2 receptor agonists play a protective role in joint diseases,
as these receptors are involved in age-related bone remodeling and bone loss; in addition, due to its
localization on cells of the immune system, the CB2R mediates immunomodulatory and anti-
inflammatory effects as well. As a result, the development of CB2R agonists, which are devoid of the
unwanted psychotropic effects ensuing from CB1R activation in the central nervous system (CNS),
may be a promising avenue to novel, disease-modifying drug candidates for the treatment of the
1
9
osteoarthritis process and chronic pain. Accordingly, many efforts were made by academic and
industrial laboratories to identify novel CB2R selective agonists because of their therapeutic potential
4
ACS Paragon Plus Environment

Page 5 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
in the treatment of a number of pathological conditions, such as neurodegenerative diseases, pain,
2
0
ischemic stroke, inflammation, autoimmune diseases, osteoporosis, and cancer. As a result of this
endeavor, a significant number of CB2R agonists with even excellent in vitro activities have been
discovered, and some of them have successfully passed the preclinical evaluation to enter clinical
trials (selected examples are reported in Chart 1),19,20a also by virtue of balanced physicochemical
properties and favorable pharmacokinetic characteristics.
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
Chart 1. Selected CB2 Agonists in Clinical Development
During the last decade, we have been actively investigating new cannabinoid ligands based on the
quinolone scaffold and our endeavor has led to the discovery of several compounds (e.g. 1, Chart 2)
showing remarkable affinity and selectivity at CB2R.21
Chart 2. Representative Examples of Quinolones as Potent and Selective CB2R Ligands
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 68
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
Compound 1 (usually referred to as COR167) behaved as a full agonist and was further assayed in in
vitro, ex vivo, and in vivo tests, where it displayed: i) potent immunomodulatory activity on immune
22
cells from healthy subjects and patients with multiple sclerosis, ii) protective effects on rat brain
tissues toward ischemia and reperfusion-induced injury,23 and iii) anti-nociceptive properties,24
thereby validating CB2R as a therapeutic target and becoming a useful pharmacological tool for
further investigations. Very recently, we and other groups have introduced significant modifications
to the quinolone scaffold of CB2R ligands aimed at improving their general physicochemical
properties. Thus, on the one hand, our approach relied on the modification of the 4-quinolones into
2
5
4
-hydroxy-2-quinolones, such as compound 2, in order to get slightly acidic compounds with pKa
2
6
values typically in the range 4.2-5.0 and hence expected to show better aqueous solubility at
physiological pH. On the other hand, Aghazadeh Tabrizi et al. opted for a scaffold hopping strategy,
by replacing the benzene ring of 4-quinolones with pyrazole, giving rise to two isomeric series,
27
namely pyrazolo[3,4-b]pyridine-5-carboxamides (e.g. 3, Chart 2), and pyrazolo[1,5-a]pyrimidine-
2
8
6
-carboxamides. Both works led to some advances in terms of physicochemical optimization of the
target cannabinoid ligands, associated with preservation of binding affinity and receptor selectivity.
As a further attempt to improve the hydrophilic/lipophilic balance of CB2R ligands, we conceived a
more marked structural modification of the original 4-quinolone nucleus, stemming from the
combination of both approaches mentioned above and leading to molecules based on the new scaffold
7
-hydroxy-5-oxopyrazolo[4,3-b]pyridin-6-carboxamide (Chart 3).
6
ACS Paragon Plus Environment

Page 7 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Chart 3. Design of Novel 7-Hydroxypyrazolo[4,3-b]pyridine-5-one Derivatives
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
Here, we describe the synthesis, drug-like and physicochemical properties, and in vitro
pharmacological profiling of novel CB2R selective ligands, as well as the in vivo evaluation of a
selected CB2R agonist in a rat model of osteoarthritis.
RESULTS AND DISCUSSION
Chemistry. The chemical synthesis to attain the target pyrazolo[4,3-b]pyridine derivatives started
with the alkylation of commercially available methyl 4-nitro-1H-pyrazole-3-carboxylate (Scheme 1)
with either iodomethane or 2-bromoethyl acetate to introduce a first element of chemical diversity
into the final compounds.
Scheme 1a
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 68
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
a
Reagents and conditions: (a) CH I, K CO , acetone, 50 °C, 2 h; (b) 2-bromoethyl acetate, KI, K CO ,
3
2
3
2
3
acetone, 50 °C, 24 h
Both reactions produced a mixture of regioisomers 4/5 and 6/7 which were separated by column
chromatography on silica gel. Their structure was assigned by X-ray crystallographic analysis (see
Supporting Information), that unambiguously showed as in both cases the more polar isomers were
4
and 6, alkylated at the 2-position, while the faster eluting isomers 5 and 7 were 1-substituted
pyrazole derivatives. Regioisomers 4 and 6 were also obtained in slightly higher percentage (4:5 =
.3:1 and 6:7 = 2:1).
Compounds 4-7 were the precursors of four different series of pyrazolo[4,3-b]pyridine derivatives
Scheme 2).
1
(
Scheme 2a
8
ACS Paragon Plus Environment

Page 9 of 68
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
a
Reagents and conditions: (a) Fe, CaCl , EtOH, H O, 60 °C, 3 h; (b) appropriate alkyl iodide or
2
2
bromide, dry DMF, 50 °C, 6-8 h; (c) 3-chloro-3-oxopropionate, NEt , dry DCM, rt, 2 h.
3
Reduction of the nitro group was accomplished using Fe/CaCl in ethanol/water and amines 8-11
2
were subjected to alkylation with different haloalkyl derivatives. This reaction was best performed in
DMF as solvent and in the absence of any added base to prevent the formation of products doubly
substituted on the amino group, likely due to higher nucleophilic reactivity of secondary amines
compared to primary ones. Being also the most basic species present in solution, the initially formed
monoalkylated compounds preferentially acted as scavenger for produced HX, making possible the
selective conversion of the primary amines 8-11 into the corresponding secondary amines 12-22.
After all, this procedure allowed to obtain the expected compounds 12-22 as the main reaction
products, albeit in only moderate yield (30-44% after chromatographic purification), as well as to
enhance their chemical diversity. Subsequent acylation of secondary amines 12-22 with methyl 3-
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 68
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
chloro-3-oxopropionate led to the corresponding amides 23-33, which in turn were subjected to
Dieckmann condensation (Scheme 3) under classical conditions (NaH/cat. MeOH/THF) to give the
pyrazolo[4,3-b]pyridine esters 34-44.
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
Scheme 3a
aReagents and conditions: (a) NaH, MeOH (cat.), dry THF, 60 °C, 2 h; (b) 1-aminoadamantane or 3-
aminoadamantan-1-ol, THF, toluene, reflux, 1-2 h; (c) 10% NaOH, H O, reflux, 1 h; (d) AcCl, NEt ,
2
3
dry DCM, rt, 2.5 h.
Due to their remarkable polarity and very low solubility in most organic solvents, these compounds
were directly used in the next reaction. Treatment of 34-44 with 1-adamantylamine or 3-amino-1-
1
0
ACS Paragon Plus Environment

Page 11 of 68
Journal of Medicinal Chemistry
1
2
3
adamantanol in refluxing toluene, with azeotropic removal of formed methanol, provided in good
yield the final adamantylamides 45-60, which could be easily purified by chromatography and/or
crystallization. This unexpectedly fast and high-yielding conversion of the ester group into an amide
functionality without using specific catalysts can be rationalized by taking into account an
elimination-addition equilibrium, leading to the formation of a highly electrophilic ketene
intermediate, with the azeotropic removal of methanol driving the reaction to completion (Scheme
4), as already reported by Jansson et al. for similar substrates.29
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
Scheme 4.
Interestingly enough, during this reaction, diesters 41-44 underwent concurrent deacetylation on the
pyrazole side chain so as to yield directly the final compounds 56-60. Conversely, the acetoxy moiety
of 37 outlasted those reaction conditions and provided a still acetylated compound 50, that was then
converted into 61 by basic hydrolysis. On the other hand, deacetylated compound 56 was reacetylated
to 62 in order to evaluate the influence of this structural modification on receptor binding affinity in
in vitro assays.
1
13
1
All the final compounds were characterized by MS, H NMR, and C NMR. In particular, H NMR
1
spectra revealed the presence of double signals, as shown as an example by H NMR spectrum of
compound 51 (Supporting Information). The interested signals are those concerning N1 methyl, C3–
H, amide NH, and enolic OH, while the alkyl side chain at N4 position and adamantane moiety do
not give rise to double signals. This outcome may account for possible tautomeric equilibria in
solution, such as that highlighted in Figure 1S (Supporting Information) which involves two most
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 68
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
likely tautomers, although we cannot rule out the possibility that other tautomeric forms might be
1
present as well. Based on the H NMR spectra registered in CDCl , we were able to estimate the ratio
3
between tautomers present in solution for the four families of final amides, which resulted to be
roughly 99:1 for 45, 3:1 for 51, 8:1 for 57, and 1.3:1 for 60.
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
Affinity/Selectivity for Cannabinoid Receptors and Functional Activity. The binding affinities
(
K values) of compounds 45-62 for human recombinant CB1R and CB2R are reported in Table 1.
i
Table 1. CB1R and CB2R Affinity Values for Compounds 45-62 in Comparison with the
a,b
Selective CB2R Ligand SR144528.
CB1Rc
CB2Rd
Compd
Structure
SIe
f
f
K (nM)
K (nM)
i
i
45
231 ± 76
2.5 ± 0.2
92
4
4
6
7
>1,000
10.6 ± 1.1
>94
532 ± 8
>1,000
2.1 ± 0.2
6.2 ± 0.1
253
48
>161
1
2
ACS Paragon Plus Environment

Page 13 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4
5
9
0
33.5 ± 0.9
>1,000
0.18 ± 0.01
17.3 ± 3.9
167
>57
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
51
69.0 ± 3.3
86.9 ± 14.9
39.3 ± 0.8
1.8 ± 0.4
19.3 ± 0.6
1.4 ± 0.2
38
5
5
2
3
4
28
5
5
4
5
161 ± 55
56.0 ± 15.8
> 1,000
18.9 ± 0.8
0.90 ± 0.01
159 ± 13
8
62
6
56
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 68
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
5
5
7
8
>1,000
108 ± 8
18.4 ± 1.9
4.8 ± 0.7
>54
22
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
6
6
9
0
1
2
492 ± 40
>1,000
> 1,000
>1,000
57.1 ± 2.3
27.1 ± 2.9
14.3 ± 0.8
186 ± 18
9
>37
>70
5
SR144528g,h
116 ± 22
1.8 ± 0.5
64
aData represent mean values ± SD for at least three separate experiments performed in duplicate and
b
are expressed as K (nM). For both receptor binding assays, the new compounds were tested using
i
3
membranes from HEK cells transfected with either CB1R or CB2R and [ H]-(–)-cis-3-[2-hydroxy-4-
1
4
ACS Paragon Plus Environment

Page 15 of 68
Journal of Medicinal Chemistry
1
2
3
1,1-dimethylheptyl)phenyl]-trans-4-(3-hydroxypropyl)cyclohexanol ([ H]CP-55,940). cCB1R:
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
d
e
human cannabinoid type 1 receptor. CB2R: human cannabinoid type 2 receptor. SI: selectivity index
f
for CB2R, calculated as K (CB1R)/K (CB2R) ratio. K : inhibitor constant, that is, the concentration
i
i
i
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
of the competing ligand that will bind to half the binding sites at equilibrium in the absence of
g
h
radioligand or other competitors. CB2 reference compound. The binding affinities of reference
compounds were evaluated in parallel with test compounds under the same conditions.
30
The compounds were evaluated in parallel with compound SR144528 as reference CB2R ligand, as
2
1
previously described. All eighteen compounds tested displayed significant binding affinity for
CB2R, with K values spanning three orders of magnitude (0.18 to 187 nM), and, with the only
i
exception of compounds 56, 59 and 62, fifteen of them proved to be potent CB2R ligands with Ki
values in the range 0.18 – 27.1 nM. Only compounds 49, 51, 52, 53 and 55 showed also significant
affinity for CB1R, with K values in the medium nanomolar range (33.5 – 86.9 nM). Nevertheless,
i
only compound 52 exhibited low receptor selectivity, while the other four pyrazolopyridine
derivatives 49, 51, 53 and 55 had selectivity index values between 28 and 167. A higher SI (37 – 253)
was found for compounds with low CB1R affinity, except for compound 62 (SI = 5), which resulted
the poorest cannabinoid receptor ligand of the series.
We selected the most potent and selective compounds (i.e. 45, 47, 48, 50, 51, 53, 54, 58, 60, 61) for
additional in vitro pharmacological evaluation to assess their ability to functionally activate the
CB2R. We performed the cAMP Hunter™ assay, to measure intracellular cAMP levels, with enzyme
fragment complementation technology in NKH-477-stimulated CHO cells optimized to overexpress
CB2R. As shown in Figure 1, compounds (51, 53, 54, 55, and 60) activated the receptor with typical
agonist behavior by reducing cAMP levels induced by the NKH-477 (a water-soluble analog of
forskolin) as expected for a Gi protein-coupled receptor agonist.
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 68
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
1
50
00
150
51
53
NKH
NKH
EC = 109.7 nM
100
50
0
EC = 21.92 nM
50
5
0
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
0
0
-10
-9
-9
-9
-8
-8
-8
-7
-6
-5
-4
-10
-9
-9
-9
-8
-8
-8
-7
-6
-5
-4
log [M]
log [M]
1
1
50
00
150
100
50
5
4
55
NKH
NKH
EC = 71.79 nM
EC = 252.2 nM
50
50
5
0
0
0
-10
-10
-7
-6
-5
6
-4
-7
-6
-5
-4
log [M]
log [M]
1
1
50
150
100
50
0
JWH-133
NKH
NKH
00
EC = 628.2 nM
EC = 391.1 nM
50
5
0
5
0
0
0
-10
-10
-7
-6
-5
-4
-7
-6
-5
-4
log [M]
log [M]
Figure 1. Concentration-response curves of compounds in cAMP Hunter™ assay enzyme fragment
complementation chemiluminescent detection kit. The curves show the effect of increasing
concentrations of compounds on NKH-477-induced cAMP levels in stable CHO cells expressing the
human CB2R. Effect of JWH-133 is reported as reference compound. Data are reported as mean ±
SEM of three independent experiments conducted in triplicate and were normalized considering the
NKH-477 stimulus alone as 100% of the response as indicated in the experimental session. EC was
50
1
6
ACS Paragon Plus Environment

Page 17 of 68
Journal of Medicinal Chemistry
1
2
3
determined by Graph Pad Prism 5 as the concentration that provokes the response halfway between
the baseline (Bottom) and maximum response (Top) of the fitted dose-response.
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
Conversely, N2 substitution on the pyrazolopyridine moiety (compounds 45, 47, 48, 49, 50, 58, 61)
promoted a switch to an antagonist (48 and 58) or inverse agonist (compounds 45, 47, 49, 50, 61)
behavior, the latter being suggested by the ability of these compounds to increase the levels of NKH-
477 induced cAMP (not shown). To further confirm this hypothesis, compounds 45, 47, 48, 50, 58
and 61 were also tested in the presence of an EC concentration of a CB2-ligand (4 µM of JWH-
8
0
1
33). The compounds were able to antagonize JWH-133-induced inhibition of NKH-477 induced
cAMP formation and, with the exception of 48 and 58, up to levels higher than those induced by
NKH-477 alone (considered as 100% response), thus confirming their predicted behavior (Figure 2).
Structure-Activity Relationship. Based on our previous findings,21,25 in all compounds the
adamantane moiety was retained, with either no modification or the one modification represented by
the insertion of a hydroxy group at the 3’-position. This change resulted in some decrease of affinity
at CB2R, as evident by comparing couples of compounds (45/46, 47/48, 51/52, and 58/59) differing
from each other only in this respect.
A major impact on receptor affinity was exerted by the alkyl chain at N4. The stepwise lengthening
of this chain from 3 (56) to 4 (57) or 5 (58) carbon atoms caused a progressive improvement in
CB2R binding, but the chain was not elongated further to avoid an excessive increase of lipophilicity.
Conversely, we took into consideration the frequently used strategy of incorporating fluorine into
drug molecules, aimed at improving potency and optimizing physicochemical and ADME
3
1
properties. This approach has ensured beneficial effects during lead optimization programs, when
attention was paid to the corresponding changes in a number of parameters of the drug candidate.32
Accordingly, we opted for the replacement with fluorine of a hydrogen atom of the terminal methyl
group in the 4-carbon atom chain, by evaluating at the same time the benefits of this modification in
terms of binding affinity and lipophilic ligand efficiency (vide infra).
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 68
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
EC = 0.27 µM
4
5
EC = 0.37 µM
47
50
4
3
2
1
00
00
00
00
0
50
400
300
NKH
NKH
JWH-133
JWH-133
2
1
00
00
0
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
-10
-9
-9
-9
-9
-8
-7
-6
-5
-4
-4
-4
-4
-10
-9
-9
-9
-8
-7
-6
-5
-4
-4
-4
-4
log [M]
log [M]
4
8
EC = 0.59 µM
49
EC = 1.18 µM
50
4
3
2
1
00
00
00
00
0
50
400
300
200
NKH
NKH
JWH-133
JWH-133
1
00
0
-10
-8
-7
-6
-5
-10
-8
-7
-6
-5
log [M]
log [M]
50
EC = 2.08 µM
58
EC = 0.88 µM
50
4
3
2
00
00
00
50
400
300
200
100
0
NKH
NKH
JWH-133
JWH-133
100
0
-10
-8
-7
-6
-5
-10
-8
-7
-6
-5
log [M]
log [M]
EC = 26.69 µM
6
1
SR144528
50
EC = 1.45 µM
4
3
2
1
00
00
00
00
0
400
300
200
100
0
50
EC = 3.15 µM
NKH
AM630
NKH
JWH-133
50
JWH-133
-10
-8
-7
-6
-5
-10
-9
-8
-7
-6
-5
log [M]
log [M]
Figure 2. Concentration-response curves of compounds in cAMP Hunter™ assay. Curves show the
effect of increasing concentrations of compounds on NKH-477-induced cAMP levels and in presence
1
8
ACS Paragon Plus Environment

Page 19 of 68
Journal of Medicinal Chemistry
1
2
3
of an EC CB2-ligand challenge (4 µM of JWH-133), in stable CHO cells expressing the human
8
0
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
CB2. Data are reported as mean ± SEM of three independent experiments conducted in triplicate, and
were normalized to the maximal and minimal response as indicated in the experimental session. EC50
was determined by Graph Pad Prism 5 as the concentration that provokes the response halfway
between the baseline (Bottom) and maximum response (Top) of the fitted dose-response.
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
Comparison between 45/47, 46/48, 51/53, and 52/54 clearly shows that no substantial change in
CB2R affinity occurred as a consequence of hydrogen/fluorine exchange; in all cases, a very slight
increase in affinity could be measured instead, associated with an improvement in receptor selectivity.
On the other hand, the insertion at the end of the alkyl chain of a hydroxyl (61) or an acetoxy (50)
group led to a 6/7–fold decrease in binding affinity at CB2R with concurrent loss of affinity for the
other receptor subtype.
A longer chain with an H-bond donor/acceptor hydroxyl was also substituted for the N2-methyl group
to get more information on the structure-activity relationship (SAR). Compounds 56-59 exhibited
significantly lower CB2R affinity than their methylated counterparts; similar result was also yielded
by compound 60 compared to its N1-methyl analog 51.
The structure-function relationship appears to be very well defined and strictly dependent on N1 or
N2 substitution. N1-Methyl substituted derivatives 51, 53, 54, 55, 60 show the ability to activate the
CB2R, while the corresponding compounds with the methyl substituent at N2 position behave as
inverse agonists (45, 47, 49, 50, 61) or neutral antagonists (48 and 58) at the same receptor.
Modification of other structural features, such as the N4 chain or adamantane moiety, were irrelevant
in this regard.
Estimated Druglike Properties.
The molecular properties of tested compounds 45-62 were estimated using the free web tool
SwissADME [http://www.swissadme.ch] and compared to those of compounds 1 and 2 (Table 1S,
Supporting Information), that previously proved to be very potent and selective CB2R ligands, but
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 68
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
did not exhibit completely satisfactory physicochemical characteristics. Overall, the new
pyrazolo[4,3-b]pyridine derivatives were in line with the general druglikeness requirements,
possessing better virtual profiles than compounds 1 and 2 according to the Lipinski’s “rule-of-five”
3
3
34
35
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
(
Ro5), Veber’s parameters, and lipophilic ligand efficiency (LLE) . In particular, the calculated
Log P (CLogP) and TPSA values for many of the compounds 45-62 fell within the optimal range for
oral bioavailability and blood–brain barrier (BBB) permeability (CLogP = 2–4, TPSA = 50–90 Å2),36
2
while the higher polarity (TPSA = 109–129 Å ) estimated for some of them prompt to presume that
these compounds might be predominantly peripheral CB2R binders, as their ability to penetrate the
3
7
CNS may be expected to be lower. Sixteen of the eighteen compounds were characterized by a
multi-parameter score LLE within the optimal range for suitable drug candidates (5 < LLE < 7), since
the reduction in overall lipophilicity, as predicted by CLogP values, did not affect the CB2R affinity
values (pKiCB2).
In Vitro Physicochemical Characterization. The aqueous solubility of compounds 45-62 was then
2
5
assessed measuring their kinetic solubility starting form DMSO stock solutions. This approach was
preferred to measurements of equilibrium solubility from powder, because these are significantly
affected by the nature of the solid state (amorphous, crystalline, presence of different polymorphs,
solvates), which, generally, is not thoroughly investigated at these stages of drug development.
In Table 2 the values of kinetic solubility, at pH 1.0 and 7.4, expressed as µg/mL, are reported.
Table 2. Experimental and Calculated Solubility of Compounds 45-62 in Comparison with
Compounds 1 and 2
Kinetic
Solubility
pH 1.0
Kinetic
Solubility
pH 7.4
Kinetic
Solubility
pH 7.4
Calculated
Calculated
Solubility
Solubility
Compd
ACD/Labsb
ALOGPSc
(
µg/mL)a
(µg/mL)a
Log S
Log S
Log S
2
0
ACS Paragon Plus Environment

Page 21 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4
4
4
4
4
5
5
6
7
8
9
0
0.23  0.03
15  2
1.9  0.3
–5.32
–4.32
–5.73
–4.06
–5.35
–5.54
–5.96
-4.95
–5.52
–4.45
–5.58
–5.03
–5.30
–5.80
–4.49
–6.26
–5.11
–5.61
–5.64
–5.47
–4.38
–3.36
–4.51
–3.49
–4.65
–4.26
–4.32
–3.32
–4.46
–3.45
–4.65
–3.95
–4.22
–4.49
–3.46
–4.20
–3.61
–4.28
–6.04
–3.64
–2.76
–3.72
–2.72
–4.01
–3.81
–3.67
–2.79
–3.77
–2.78
–4.02
–2.99
–3.32
–3.66
–2.89
–3.29
–3.06
–3.77
–6.15
–4.99
20.9  0.8
0.8  0.1
37.6  1.3
1.9  0.2
1.3  0.2
0.49  0.06
4.6  0.3
1.5  0.4
15.1  0.2
1.08  0.04
3.9  0.1
2.1  0.4
0.70  0.02
15  1
0.6  0.1
39.9  0.2
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
2.9  0.2
d
51
0.13  0.01
2.1  0.6
5
2
53
0.7  0.2
5
5
5
5
5
5
4
5
6
7
8
9
9.1  0.1
0.4  0.1
3.4  0.6
1.6  0.2
0.38  0.05
9.70  0.02
0.11  0.03
60
0.24  0.04
3.2  0.1
1.1  0.1
1.0  0.1
1.4  0.1
6
1
2.53  0.06
d
62
1e
2e
0.21 0.02
0.23 0.06
–4.47
a
b
Reported are mean values together with Standard Deviation (n = 4). ACD calculated solubility
c
values (ACD/Labs I-Lab 2.0: http://www.ilab.acdlabs.com). ALOGPS 2.1 calculated solubility
values (http://www.vcclab.org). Reported are the mean solubility values of the neutral species
calculated taking into account the different tautomeric forms reported in Figure 2S (Supporting
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 68
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
d
Information). Not stable in the test time period (> 25% hydrolysis of the side chain ester group in 4
e
h). See also reference 25.
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 corresponding Log S values, in Log (mol/L) scale are compared to calculated Log S by ACD/Labs
and ALOGPS software. In general, experimental Log S at acidic pH (pH 1.0) ranged between –4.03
of the most soluble 48 and –6.59 of the least soluble 60 and between –4.06 (48) and –6.26 (60) at
physiological pH, spanning over two orders of magnitude. The values of log S calculated by the two
software, ACD/Labs and ALOGPS respectively, were mutually correlated (Log SACD = 0.97(0.08)
2
Log SALOGPS –0.77(0.27); n = 18; r = 0.905; s = 0.15; F = 152), but tended to overestimate solubility
in comparison with experimental results. Correlation between experimental and calculated values was
2
significantly lower, with ACD/Labs (Log Sexp = 1.01(0.19) Log SACD –1.15(0.77); n = 18; r =
0.641; s = 0.37; F = 28.5) performing better than ALOGPS (Log Sexp = 0.98(0.21) Log SALOGPS
–
2
1
.93(0.71); n = 18; r = 0.580; s = 0.40; F = 22.1).
Considering experimental results, in the subset of N2-methyl derivatives 45-50 and 61, hydroxylation
of the adamantyl group (46) led to a 1-log unit increase in solubility at both pHs, if compared to the
parent compound 45. Fluorination of the butyl side chain at N4 (47) increased the solubility at acidic
pH (3 times vs. 45), but led to a slight decrease at physiological pH. The combination of both
hydroxylation and fluorination gave the best result, with a solubility for compound 48 170 times (pH
1
.0) and 20 times (pH 7.4) higher than that of 45. Within the subset of N2-methyl substituted
derivatives, shifting from n-butyl to n-pentyl side chain at N4 (49) did not significantly affect kinetic
solubility, especially at physiological pH, whereas hydroxylation of the side chain (61) led to an
increase in the solubility at both pHs, even if the magnitude of the effect was not comparable to that
obtained with 3’-hydroxylation at the adamantyl ring. Acetylation of the hydroxyl group in the side
chain (50) did not significantly affect solubility at pH 7.4 and led to a stability issue owing to
hydrolysis of this ester group at acidic pH. N1-Methyl derivatives 51-55 were in general less soluble
at both pHs than the corresponding N2-methyl ones. For instance, parent compound 51 was twice
2
2
ACS Paragon Plus Environment

Page 23 of 68
Journal of Medicinal Chemistry
1
2
3
less soluble at pH 1.0 and four times less soluble at pH 7.4 than corresponding 45. Nevertheless, 3’-
hydroxylation at the adamantyl group (52), fluorination on the n-butyl side chain at N4 (53) and the
combination of both substitutions (54) paralleled the effects on kinetic solubility observed for the 46-
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
4
8 subset. Compound 54 was the most soluble of N1-methyl derivatives with a Log S at physiological
pH around –4.45.
Finally, the combined effect on kinetic solubility of introducing a hydroxyethyl side chain at N1 (60)
or N2 (56-59) of the 5-oxopyrazolo[4,3-b]pyridine nucleus and of elongating the side chain at N4
was explored, looking for the right combination of both solubility and activity. As previously
observed, the N1-substituted derivative (60) was much less soluble than the corresponding N2-
substituted one (57) and also the pharmacological activity and selectivity for CB2R favored the latter.
Elongation of the alkyl side chain at N4 led to a 1-log unit progressive decrease in kinetic solubility
at both pHs (56-58). Compound 57 combined good selectivity for CB2R (SI > 54) with a Log S = –
5
.30. The ionization constants (pKa) and experimental lipophilicity (log P ) of compounds 45, 46,
oct
4
8, 51, and 52 were also measured, by means of a UV-metric and potentiometric approach (see
Experimental Section and Table 3).38
Table 3. Ionization Constants (pKa) and Lipophilicity (Log P ) of Compounds 45, 46, 48, 51,
oct
and 52
Compd
pKaa,b
8.6  0.1
7.9  0.1
7.97  0.07
8  0.1
Log Poctb,c
5.3  0.1
4
4
4
5
5
6
8
1
4.08  0.05
3.15  0.01
5.3  0.1
52
7.56  0.05
4.85  0.01
a
Aqueous pKa values extrapolated from MeOH/0.15 M ionic strength adjusted water solutions (see
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 68
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
b
Experimental Section for details). Reported are mean values together with their Standard Deviations
c
38
(
n=3). Measured by the potentiometric pH-metric approach.
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
For all compounds, aqueous pKa values were extrapolated starting from solutions at different
percentages of co-solvent (MeOH). In all the cases, the slope of the extrapolation curve was typical
of acidic compounds. Measured pKa values ranged from 7.56 (52) to 8.56 (45). In both subsets of
N2-methyl (45, 46) and N1-methyl (51, 52) substituted derivatives, the introduction of the hydroxyl
group at 3’ on the adamantyl ring led to a significant decrease in pKa value, while no effect was seen
for fluorination (48). The partition coefficients (log P ) were determined evaluating the shifts
oct
produced in the titration curves by different ratios of partition solvent (n-octanol) and 0.15 M ionic
strength adjusted water. Remarkably, in the N2-methyl derivatives (45, 46, 48) a 2-log unit decrease
of lipophilicity was observed when 3’-hydroxylation on adamantane or N4-side chain fluorination
took place. Experimental log P values differed from calculated ones by a ∆(log P – CLogP) ranging
oct
between 1.2 and 1.6, except for compound 52, which rather surprisingly showed an experimental log
P value significantly higher than the corresponding N2-substituted isomer 46. However, the
experimental values paralleled the calculated ones and compounds 45, 46, 48, 51, and 52 ranked in
the same order based on both theoretical and experimental lipophilicity. Therefore, the calculation of
druglikeness was useful to predict how structural modifications of compounds would affect their main
physicochemical properties.
In Vitro Cytotoxicity. In vitro cytotoxicity and cytocompatibility testing provides a crucial means
of ranking compounds for consideration in drug discovery. Although all the molecules of the series
were tested for cytotoxicity, only compounds which showed low cytotoxicity were reported.
To check the potentiality of these compounds, in vitro cell toxicity up to 24 hours towards either
NIH3T3 fibroblasts or Normal Human Articular Chondrocytes (NHAC-kn) was assessed as a
function of concentration, and results (IC values) are reported in Table 4.
50
2
4
ACS Paragon Plus Environment

Page 25 of 68
Journal of Medicinal Chemistry
1
2
3
Table 4. Cytotoxicity Values of Selected Compounds 45-48, 50-54, 57-61 towards NIH3T3
Fibroblasts and Normal Human Articular Chondrocytes (NHAC-kn)a
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
Compd
NIH3T3 (IC , µM)
NHAC-kn (IC , µM)
5
0
50
4
4
4
4
5
5
5
5
6
7
8
0
1
2
12.5
7.5
12.5
25
7.5
2.5
7.5
12.5
7.5
25
25
12.5
7.5
25
53
54
12.5
25
12.5
7.5
5
5
5
6
6
7
8
9
0
1
12.5
7.5
7.5
12.5
25
2.5
2.5
2.5
2.5
12.5
aCell viability measured by the Neutral Red Uptake (NRU) test. All compounds were tested at
increasing concentrations ranging from 2.5 to 50 µM. The assay was carried out three times in three
replicates for each tested concentration. The standard deviation is less than 5%.
All the test compounds showed good cell compatibility and demonstrated to inhibit cell viability at
micromolar concentrations: from 7.5 to 25 µM for NIH3T3 fibroblasts and from 2.5 to 12.5 µM for
NHAC-kn.
Three compounds in particular (48, 51 and 61) showed low cytotoxicity (IC >10 µM) against both
5
0
cell types. Based on its functionality at CB2R and drug-like properties, we selected 51 as the lead
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 68
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
compound, which was further investigated for its long lasting cytotoxicity and capability of blocking
the in vitro dedifferentiation process of human chondrocytes. Cell viability of NHAC-kn exposed to
7.5 µM, i.e. the highest not toxic concentration, up to 14 days was assessed by the Neutral Red Uptake
assay. As shown in Table 5, no toxic effect towards chondrocyte viability was observed as cell density
was not statistically different (p<0.05) in comparison with vehicle at 1, 7, and 14 days, demonstrating
that compound 51 does not interfere with NHAC-kn cell viability.
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
Table 5. Percentage of Viable NHAC-kn in Contact with Compound 51 (7.5 µM) Evaluated by
the Neutral Red Uptake Test as a Function of Incubation Time
Percentage (± SD) of viable NHAC-kn as a function
of incubation time
1
day
7 days
14 day
Negative control
Growth Medium)
47±6
73±8
100±11
(
Compound 51
*51±8
*68±6
*94±7
(7.5 µM)
Data are means ± SD of three replicates for each tested concentration or incubation
time. *Data are not statistically different in comparison to negative control, p<0.05.
NHAC-kn in contact with 7.5 µM compound 51 were analysed for their content of major
cartilagineous extra-cellular matrix (ECM) components, i.e. sulphated GAG and total collagen
(HYP), in order to assess the ability of the test compound to prevent chondrocytes dedifferentiation.
In fact, human chondrocytes dedifferentiate rapidly during monolayer culture, acquiring a fibroblast-
2
6
ACS Paragon Plus Environment

Page 27 of 68
Journal of Medicinal Chemistry
1
2
3
like phenotype and decreasing the production of cartilaginous tissue. As reported in Table 6,
compound 51 statistically increased sGAG levels compared to the negative control, i.e. growth
medium not supplemented with growth factors able to decelerate the dedifferentiation process,39
already after 7 days and its effect persisted after 14 days of incubation. The increase in sGAG levels
is a significant result, as it shows that chondrocytes are not differentiated.
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
Hydroxyproline (HYP) is a non-proteinogenic amino acid formed by the post-translational
hydroxylation of proline. HYP is a major component of collagen, where it serves to stabilize the
helical structure. Because HYP is largely restricted to collagen, the measurements of HYP levels can
be used as an indicator of collagen content.
The results obtained regarding HYP production by human chondrocytes in contact with negative
control and 7.5 µM compound 51, are reported in Table 6.
Table 6. Extracellular Matrix Production in NHAC-kn Cultures in the Presence of Growth
Medium (Negative Control) or Compound 51 (7.5 µM) as a Function of Incubation Time.
Extracellular matrix production
sGAG (µg ± SD)
HYP (µg ± SD)
sGAG/HYP ratio
Incubation
time
Negative Compound Negative Compound Negative Compound
control
9.3±0.2
51
control
9.9±0.3
51
control
0.9±0.1
0.8±0.2
0.9±0.2
51
1
day
8.9±0.4
10.1±0.2
12.9±0.3
13.9±0.3
0.9±0.2
1±0.2
1±0.2
7
days
11.2±0.4
12.0±0.4
*12.6±0.3
*14.2±0.3
13.2±0.3
14.2±0.4
1
4 days
Data are means±SD of three replicates for each incubation time.
Data are significantly different in comparison to the negative control. p<0.05.
*
The data demonstrate that the amount of HYP increased by increasing the incubation time for both
the negative control and test compound of the same extent. The increasing of both sGAG and HYP
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 68
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
levels, and the value of the sGAG/HYP ratio, calculated as an indicator for hyaline or fibrocartilage
nature of the produced matrix (Table 6), demonstrated that negative control and compound 51 showed
the same trend and that the sGAG/HYP ratio remained constant by increasing the incubation time.
These results indicate that, although specific collagen was not measured, the produced extracellular
matrix was hyaline. In addition, these results demonstrate the good ability of the compound 51 to
stimulate the production of hyaline ECM and to prevent the in vitro dedifferentiation of primary
human chondrocytes (HC) confirming its chondrogenic capacity.
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
Effect of CB2R Agonist Systemic Administration on Joint Hypersensitivity and Walking
Pattern in MIA-Treated Rats. To evaluate the analgesic properties of the lead compound in vivo,
we induced osteoarthritis in rats by intraarticular injection of sodium monoiodoacetate (MIA), which
prompts rapid pain-like responses in the ipsilateral limb and morphological changes of the articular
cartilage. All changes of the micro-architecture of subchondral bone in terms of bone volume,
trabecular thickness and trabecular number lead to bone disruption and cause knee joint
hypersensitivity. Along with pain phenotype development, joint damage and subchondral bone
disruption, MIA injection induces weight-bearing deficits that are measurable and quantifiable
particularly 21 and 28 days post injection.40a Moreover, osteoarthritis (OA) development is
accompanied by dynamic changes of CB2 expression in dorsal root ganglion (DRG) and cartilage in
MIA rats.41a,b Therefore, 21 days after OA induction we assessed pain-related behaviors performing
two different in vivo tests, test pressure applied measurement (PAM)40b,c and the kinetic weight
bearing (KWB) tests.40a Two sensory systems were evaluated by joint hypersensitivity (PAM) and
gait analysis (KWB) to well define the in vivo effect of compound 51 in OA-induced animals. In the
PAM test, compound 51 triggered a statistically significant analgesic effect at the 1 mg/kg dose 60,
1
20 and 180 min after administration (p < 0.05 and p < 0.01, Figure 3).
2
8
ACS Paragon Plus Environment

Page 29 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
VEHICLE
6
4
0
0
COMPOUND 51 (1 mg/kg)
COMPOUND 51 (1 mg/kg) + AM-630 (1 mg/kg)
#
**
*
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
**
20
0
0
0
0
8
0
0
6
2
4
0
1
1
2
3
Time post drug administration (mins)
Figure 3. Time course and effect of acute, systemic treatment with compound 51 alone or in
combination with the CB2 receptor antagonist AM-630 on hind limb joint hypersensitivity in
osteoarthritic rats (PAM test). 21 or 28 days post injection with 1 mg of MIA, rats received compound
5
1 (1 mg/kg) intraperitoneally (i.p.), and the hind limb withdrawal threshold was assessed during a
00-min period. Compound 51 triggered a statistically significant analgesic effect (* p < 0.05 and **
3
p < 0.01) compared to the vehicle group at 60, 120, and 180 min post administration recordings. A
systemic pre-administration of AM-630 (1 mg/kg, i.p.) reduced the analgesic effect of compound 51
with statistical significance (# p < 0.05) 120 min post injection. Data are presented as maximal
possible effect (%MPE) ± SEM. Male Wistar rats were used in the study and divided into four
experimental groups with n =17 for vehicle, n =10 for compound 51, n =10 for AM-630 + compound
5
1, n = 6 for AM-630. Statistical analysis was performed using analysis of variance (ANOVA)
followed by Tukey post-hoc test. The values with p * < 0.05 were considered significant. * denotes
significant differences vs vehicle, # denotes significant differences vs compound 51 (1 mg/kg).
2
9
ACS Paragon Plus Environment