The Pyrazolobenzothiazine Core as a New Chemotype of p38 Alpha Mitogen-Activated Protein Kinase Inhibitors

Article abstract of DOI:10.1111/cbdd.12516

The identification, synthesis, biological activity, and binding mode prediction of a series of pyrazolobenzothiazines as novel p38α MAPK inhibitors are reported. Some of these compounds showed interesting activity in both p38α MAPK and TNF-α release assays. Derivative 6 emerged as the most interesting compound with IC₅₀ (p38α) = 0.457 μm, IC₅₀ (TNF-α) = 0.5 μm and a promising kinase selectivity profile. The obtained results strongly indicate the pyrazolobenzothiazine core as a new p38α inhibitor chemotype worthy of future chemical optimization efforts directed toward identifying a new generation of anti-inflammatory agents.

Full text of DOI:10.1111/cbdd.12516

Received Date : 15-Sep-2014
Revised Date : 19-Dec-2014
Accepted Date : 05-Jan-2015
Article type
: Research Article
The pyrazolobenzothiazine core as a new chemotype of p38 alpha
mitogen-activated protein kinase inhibitors
Stefano Sabatini1, Giuseppe Manfroni1,*, Maria Letizia Barreca1,*, Silke M. Bauer2, Marco
Gargaro3, Rolando Cannalire1, Andrea Astolfi1, Jose Brea4, Carmine Vacca3, Matteo Pirro5,
Serena Massari1,Oriana Tabarrini1, Maria Isabel Loza4, Francesca Fallarino3, Stefan A.
Laufer2, and Violetta Cecchetti1
1 Department of Pharmaceutical Sciences, University of Perugia, Via A. Fabretti, 48, 06123
Perugia (Italy)
2 Department of Pharmaceutical & Medicinal Chemistry, Institute of Pharmacy, Eberhard-
Karls University Tuebingen, Auf der Morgenstelle 8, 72076 Tuebingen (Germany).
3 Department of Experimental Medicine, University of Perugia, Piazzale Gambuli, 06100
Perugia (Italy)
4 CIMUS Research Center, University of Santiago de Compostela, Avda de Barcelona s/n,
Planta 3. Despacho 1, 15782 Santiago de Compostela (Spain)
5 Department of Medicine, University of Perugia, Piazzale Gambuli, 06100 Perugia (Italy)
*corresponding authors: giuseppe.manfroni@unipg.it, marialetizia.barreca@unipg.it
Abstract: The identification, synthesis, biological activity and binding mode prediction of a
series of pyrazolobenzothiazines as novel p38α MAPK inhibitors are reported. Some of
these compounds showed interesting activity in both p38α MAPK and TNF-α release assays.
Derivative 6 emerged as the most interesting compound with IC50 (p38α) = 0.457 µM, IC50
(TNF-α) = 0.5 µM and a promising kinase selectivity profile. The obtained results strongly
indicate the pyrazolobenzothiazine core as a new p38α inhibitor chemotype worthy of future
chemical optimization efforts directed toward identifying a new generation of anti-
inflammatory agents.
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process which may
lead to differences between this version and the Version of Record. Please cite this article as
an 'Accepted Article', doi: 10.1111/cbdd.12516
This article is protected by copyright. All rights reserved.

Keywords: pyrazolobenzothiazines; p38α MAPK inhibitors; anti-TNF agents; molecular
docking; ligand efficiency.
Introduction
p38 Mitogen-activated protein kinases (MAPKs) are a class of evolutionary conserved
serine/threonine kinases that link extracellular signals to the intracellular machinery
modulating a plethora of cellular processes (e.g. cell cycle, cell death, differentiation, and
senescence) (1-4). It is well known that p38 MAPKs play an essential role in regulating the
production of cytokines as tumor necrosis factor alpha (TNF-α), interleukins 1β and 6 which
in turn represent the major mediators in inflammatory autoimmune diseases such as
rheumatoid arthritis, psoriasis, lupus erythematosus, bowel diseases, asthma, chronic
obstructive pulmonary disease (COPD), and type-1 diabetes (2-4). More recently, some
other consolidated roles of p38 MAPKs in the pathogenesis of Parkinson’s disease (5),
multiple sclerosis (6,7), cancer (8-10), and Alzheimer’s disease (11,12) are emerging from
literature data. There are four isoforms of p38 MAPK (α, β, γ, δ also called MAPK14,
MAPK11, MAPK12, MAPK13, respectively) that are encoded by separate genes and show ≥
60% overall sequence similarity and >90% identity in catalytic domain (13,14). The four
isoforms have different tissue distribution with p38α MAPK and, to less extent, p38β MAPK
being ubiquitous. The latters are overexpressed during the activation of the inflammatory
process and regulate TNF-α and cytokines production (1,2,13,14) designating p38α MAPK a
valid target for therapeutic intervention (15).
In the past two decades, tremendous efforts have been made by both pharmaceutical
companies and academic groups to identify and develop new potent and selective p38α
MAPK inhibitors, some of which have reached different stage of clinical trials. Furthermore,
their mechanisms of action, kinase selectivity, crystallographic and computational studies,
and pharmacological profiles have been extensively reviewed in literature (15-21).
Notwithstanding these promising results, multiple attempts to generate clinically useful p38α
MAPK inhibitors have generally failed, either due to toxicity or inadequate efficacy. However,
several first generation p38α MAPK inhibitors are still in clinical trials for asthma, COPD,
neuropathic pain, etc (22).
Particular interest has been recently focused on the identification of a new generation of
p38α MAPK inhibitors able to overcome both the low in vivo efficacy and safety. Toward this
direction, many pharmaceutical companies have recently filed several patents reporting the
discovery and potential clinical use of new potent molecules especially devoted to the
treatment of asthma or COPD (23). In parallel, the identification of new chemotypes able to
inhibit p38α MAPK function is of a great relevance.
This article is protected by copyright. All rights reserved.

In this context, we herein describe the identification of a novel class of p38α MAPK inhibitors.
In the past, some of us reported that the pyrazolobenzothiazine core proved to be a
promising scaffold for anti-inflammatory activity, although at that time it was not possible to
investigate the mechanism of action (24). With this history as backdrop, a series of 34
pyrazolobenzothiazines (1-6, Scheme 1, 2 and Table 1) was subsequently designed and
synthesized by merging the structures of meloxicam and celecoxib in the search of selective
inhibitors of cycloxygenase isoform two (COX-2). Unfortunately, when tested for their ability
to inhibit the prostaglandin E2 (PGE2) production in a biochemical assay employing both
COX-1 and COX-2 isoenzymes, all compounds resulted inactive (data not shown). Anyway,
we felt confident enough about the potentiality of the pyrazolobenzothiazine scaffold for anti-
inflammatory activity to decide to test series 1-6 against another target well-known to be
involved in the inflammatory process, that is p38α MAPK. This target selection was also
supported by a somehow structural similarity between our compounds and SB203580 (25-
29), one of the best studied p38α MAPK inhibitors. Taking into account the bioisosteric
relationship between the imidazole (SB203580) and the pyrazole (in house compounds)
nucleus, two chemical similarity hypothesis could be proposed depending on the
functionalization at the N-1 atom in the pyrazolobenzothiazine scaffold (Figure 1). The main
difference between the two assumptions consisted in the pyrazolobenzothiazine functional
group corresponding to the para-pyridinyl substituent of the known inhibitor. In derivatives 1-
3, the phenyl ring of the benzothiazine core replaced the pyridine moiety of SB203580 whilst
in derivatives 4-6 we focused the attention on the hydrogen-bond acceptor character of the
pyridinyl nitrogen atom, which could be mimicked by the sulfone group.
This article is protected by copyright. All rights reserved.

Figure 1. General structure of pyrazolobenzothiazines 1-6 and SB203580. Keeping constant
the pyrazole ring as bioisosteric replacer of imidazole ring of SB203580, two chemical
similarities hypothesis are proposed. The different colours highlight the chemical similarities
between the two chemotypes.
In this paper, we report the synthesis, the biological evaluation, the structure-activity
relationship (SAR) and computational studies of pyrazolobenzothiazine derivatives 1-6.
Material and Methods
General Chemistry
All starting materials were commercially available, unless otherwise indicated. Reagents and
solvents were purchased from common commercial suppliers and were used as such.
Organic solutions were dried over anhydrous Na2SO4 and evaporated to dryness with a
rotary evaporator at low pressure. All reactions were routinely checked by thin-layer
chromatography (TLC) on silica gel 60F254 (Merck) and visualized by using UV or iodine.
Column chromatography separations were carried out on Merck silica gel 60 (mesh 70-230),
flash chromatography on Merck silica gel 60 (mesh 230-400). Melting points were
determined in capillary tubes (Büchi Electrotermal model 9100) and are uncorrected. Yields
were of purified products and were not optimized. The reactions conducted under
microwaves irradiation were carried out employing a microwave reactor Biotage InitiatorTM
2.0 version 2.3 build 6250. Ultrasonic mediated reactions were carried out in VWR Ultrasonic
Bath (HF45kHz80W) made by VWR, Malaysa. 1H NMR spectra were recorded at 200 or 400
MHz (BrukerAvance DRX-200 or 400, respectively) while 13C NMR spectra were recorded at
100 MHz (BrukerAvance DRX-400). Chemical shifts are given in ppm (δ) relative to TMS.
Spectra were acquired at 298 K. Data processing was performed with standard Bruker
software XwinNMR and the spectral data are consistent with the assigned structures. All
compounds were ≥95% pure as determined by LC/MS using an Agilent 1290 Infinity System
machine equipped with DAD detector from 190 to 640 nm. With exception of derivative 5,
that was analyzed at 254 4 nm, the purity was revealed at 270.4 4 nm. For compounds 1a-
k, 2a-k, 3a-c, 3f-k, 4, and 6 an Phenomenex AerisWIDEPORE C18 (2.1 mm × 100 mm, 1.7
µm particle size column) reverse phase was used with gradient of 0-100% acetonitrile with
0.1% formic acid (channel B) in water with 0.1% formic acid (channel A) for 20 min. at 0.3
mL/min. Injection volume was 0.5 µL and column temperature of 40 °C. Peaks retention time
are given in minutes. Compound 5 was instead analyzed using an Phenomenex
AerisWIDEPORE C4 All-ion (4.6 mm × 100 mm, 3.6 µm particle size column) employing the
same parameters as used for previous compounds. Detection was based on electrospray
This article is protected by copyright. All rights reserved.

ionization (ESI) in negative polarity using Agilent 1290 Infinity System equipped with a MS
detector Agilent 6540UHD Accurate Mass Q-TOF.
General Procedure for Preparation of Pyrazolobenzothiazes 1a-i (Method A). In a
microwave oven tube, the key synthon 7 (30) (1 mmol) was dissolved in DMF (2.5 mL) and
the appropriate phenylhydrazine hydrochloride (1.5 mmol) was added. The solution was
irradiated at 100 °C from 5 to 75 min, employing the following experimental parameters:
pressure 5 bar, cooling on, FHT on, prestirring 30 s, very high absorption. The solution was
then poured into ice-water and acidified with 2N HCl (pH = 4) to give a precipitate which was
filtered, dried and crystallized by EtOH.
Experimental details and analytical data for compounds 1a-i are reported in Supporting
Information.
[4-(3-methyl-5,5-dioxidopyrazolo[4,3-c][1,2]benzothiazin-1(4H)-yl)phenyl]amine (1j). A
stirred solution of the nitro derivative 1i (0.52 g, 1.4 mmol) in DMF (30 mL), was
hydrogenated over catalytic amount of Raney nickel at room temperatuere and atmospheric
pressure for 30 min. The mixture was filtered over Celite and the filtrate was evaporated to
dryness; the residue was crystallized by EtOH to give target amino derivative 1j (0.30 g,
1
65%) as pale brown solid: mp 287-288 °C. H NMR (200 MHz, DMSO-d6): δ 2.25 (s, 3H,
CH3), 5.80 (brs, 2H, NH2), 6.70-6.80 (m, 2H, H-3’ and H-5’), 6.90-7.25 (m, 3H, H-9, H-2’, and
H-6’), 7.45-7.70 (m, 2H, H-7 and H-8), 7.75-8.00 (m, 1H, H-6), 10.25 (brs, 1H, NH). HRMS
(ESI) m/z [M+H]+ calcd for C16H14N4O2S: 327.0910, found: 327.0916; LC-MS: ret. time 6.44
min.
In a similar manner, compound 2j was prepared starting from 2i (See Supporting
Information)
N-[4-(3-methyl-5,5-dioxidopyrazolo[4,3-c][1,2]benzothiazin-1(4H)-
yl)phenyl]methanesulfonamide (1k). To a solution of compound 1j (0.3 g, 0.92 mmol) in a
1:1 dry CH2Cl2/pyridine mixture (50 mL) and under nitrogen flux, methanesulfonyl chloride
(0.4 mL, 4.5 mmol) was added dropwise at 0 °C. The mixture was then stirred at 50 °C for
24h and the solvent was concentrated to one third of the volume. Water was added and the
mixture was acidified with 2N HCl (pH=3) and extracted with EtOAc (5 times). The organic
phases were collected and washed with brine, dried and evaporated to dryness. The crude
residue was purified by flash column chromatography eluting with CH2Cl2/MeOH (95:5) to
1
give target compound 1k (0.20 g, 54%) as pale brown solid: mp 298-300 °C. H NMR (200
MHz, DMSO-d6): δ 2.30 (s, 3H, CH3), 3.15 (s, 3H, SO2CH3), 6.95-7.05 (m, 1H, H-9), 7.30-
7.70 (m, 6H, H-7, H-8, H-2’, H-3’, H-5’, and H-6’), 7.95-8.00 (m, 1H, H-6), 10.25 (brs, 1H,
This article is protected by copyright. All rights reserved.

SO2NH), 10.55 (brs, 1H, benzothiazine SO2NH). HRMS (ESI) m/z [M+H]+ calcd for
C17H16N4O4S2: 405.0692, found: 405.0686; LC-MS: ret. time 7.66 min.
In a similar manner, compound 2k was prepared starting from 2j (See Supporting
Information)
General Procedure for the Preparation of Compounds 2b-g and 2i. Method B. To a
suspension of 60% NaH (2.5 mmol) in dry DMF (3 mL) cooled at 0 °C and under nitrogen
flux, a solution of the appropriate compound (1b-g and 1i) (1 mmol) in dry DMF (5 mL), was
added dropwise. After 30 min, MeI (0.62 mL, 10 mmol) diluted in dry DMF (1 mL), was
added dropwise and the mixture was stirred for additional 1h at room temperature. The
mixture was then poured into ice/water and extracted several time with EtOAc, the combined
organic layers were washed with brine and dried. The crude residue was purified as reported
below for each compound to give the target derivatives 2b-g and 2i.
Experimental details and analytical data for compounds 2b-g and 2i are reported in
Supporting Information.
4-(3,4-Dimethyl-5,5-dioxidopyrazolo[4,3-c][1,2]benzothiazin-1(4H)-
yl)benzenesulfonamide (2h). To a suspension of K2CO3 (0.021 g, 0.15 mmol) in dry DMF
(1 mL), a solution of compound 1h (0.060 g, 0.15 mmol) in dry DMF (3 mL) was added
dropwise and the mixture was stirred at room temperature for 1h. MeI (0.011 mL, 0.17 mmol)
diluted in dry DMF (1 mL) was then added dropwise and the mixture was stirred at room
temperature for 15h. The mixture was poured into ice/water and the formed precipitated was
filtered to give a solid which was crystallized by MeOH to give compound 2h (0.054 g, 89%)
as white solid: mp 275-276 °C. 1H NMR (200 MHz, CDCl3): δ 2.45 ( s, 3H, CH3), 3.15 (s, 3H,
NCH3), 4.95 (s, 2H, SO2NH2), 7.10 (dd, J = 1.2 and 7.4 Hz, 1H, H-9), 7.40-7.65 (m, 2H, H-7
and H-8), 7.70 (m, 2H, H-2’ and H-6’), 7.95-8.10 (m, 3H, H-6, H-3’ and H-5’). HRMS (ESI)
m/z [M+H]+ calcd for C17H16N4O4S2: 405.0692, found: 405.0687; LC-MS: ret. time 8.87 min.
2-[2-(4-Fluorophenyl)-2-oxoethyl]-1,2-benzisothiazol-3(2H)-one 1,1-dioxide (8) (31). To
a solution of saccharin sodium salt (9.40 g, 46 mmol) in dry DMSO (30 mL) a solution of 2-
bromo-1-(4-fluorophenyl)ethanone (5.00 g, 23.0 mmol) in dry DMSO (10 mL) was added
dropwise and the reaction was stirred at 70 °C for 3h. The mixture was then poured into
ice/water to give a precipitate which was filtered to give 8 (6.50 g, 88%) as crude product
which was used as is in the next reaction step: mp 177-178 °C. 1H NMR (200 MHz, DMSO-
d6): δ 5.45 (s, 2H, CH2), 7.25-7.45 (m, 2H, H-3’ and H-5’), 7.90-8.20 (m, 5H, H-5, H-6, H-7,
H-2’, and H-6’), 8.30 (d, J = 8.1 Hz, 1H, H-4).
This article is protected by copyright. All rights reserved.

(4-Fluorophenyl)(4-hydroxy-1,1-dioxido-2H-1,2-benzothiazin-3-yl)methanone (9) (31).
Shiny metallic sodium (0.60 g, 26 mmol) was added to absolute EtOH (80 mL), the mixture
was stirred at room temperature until the sodium was completely dissolved. Then, a dried
powder of intermediate 8 (3.30 g, 10.3 mmol) was added at once and the red solution was
stirred at 60 °C for 1h. The mixture was poured into ice/water and slowly acidified with 12N
HCl (pH = 2) obtaining a precipitate that was filtered to give compound 9 (2.60 g, 78%) as
1
yellow crude solid which was used as is in the next reaction step: mp 188-189 °C. H NMR
(200 MHz, DMSO-d6): δ 7.30-7.50 (m, 2H, H-3’ and H-5’), 7.80-7.95 (m, 3H, H-5, H-6, and
H-7), 8.00-8.25 (m, 3H, H-8, H-2’, and H-6’), 9.90 (brs, 1H, NH).
3-(4-Fluorophenyl)-1,4-dimethyl-1,4-dihydropyrazolo[4,3-c][1,2]benzothiazine
5,5-
dioxide (4) and 3-(4-Fluorophenyl)-2,4-dimethyl-2,4-dihydropyrazolo[4,3-
c][1,2]benzothiazine 5,5-dioxide (5). In a microwave oven tube, the benzothiazinone 9
(0.60 g, 1.9 mmol) was dissolved in EtOH (10 mL) and methylhydrazine (0.30 mL, 5.7 mmol)
was added. The solution was irradiated at 100 °C for 30 min, employing the following
experimental parameters: pressure 5 bar, cooling on, FHT on, prestirring 30 s, very high
absorption. The solvent was distilled off and the residue was chromatographed eluting with
CHCl3/MeOH (99:1) in order to remove impurities. A 1:1 mixture (0.4 g) of regioisomers 3-(4-
fluorophenyl)-1-methyl-1,4-dihydropyrazolo[4,3-c][1,2]benzothiazine 5,5-dioxide (10) and 3-
(4-fluorophenyl)-2-methyl-2,4-dihydropyrazolo[4,3-c][1,2]benzothiazine 5,5-dioxide (11) was
obtained and dissolved in dry DMF (7.0 mL). This solution was added dropwise to a stirred
suspension of 60% NaH (0.12 g, 3.03 mmol) in dry DMF (2 mL) at 0 °C. The mixture was
kept at 0 °C under stirring for additional 30 min. The MeI (0.75 mL, 12.1 mmol) diluted in dry
DMF (2.0 mL) was added dropwise and the mixture was stirred at room temperature for 1h
then poured into ice/water and extracted with EtOAc (3 times). The combined organic layers
were washed with brine, dried and filtered. The solvent was removed under vacuum to give
a crude product formed by the two methylated regioisomers 4 and 5. After flash
chromatographic separation eluting with cyclohexane/EtOAc (50:50) the N-1 methyl
regioisomer 4 (Rf< by TLC) (0.04 g) and the N-2 methyl regioisomer 5 (Rf> by TLC) (0.07 g)
were obtained as white solids. A fraction of the two regioisomers enriched by compound 4
was further crystallized by MeOH to give additional 0.07 g of 4.
1
Compound 4 (0.11 g, 52%): mp 217-219 °C. H NMR (400 MHz, DMSO-d6): δ 2.85 (s, 3H,
N-4 CH3), 4.30 (s, 3H, N-1 CH3), 7.30-7.40 (m, 2H, H-3’ and H-5’), 7.80 (t, J = 7.6 Hz, 1H, H-
7), 7.90-8.05 (m, 4H, H-6, H-8, H-2’, and H-6’), 8.10 (d, J = 7.8 Hz, 1H, H-9). 13C NMR (100
MHz, DMSO-d6): δ 38.83, 40.65, 116.47 (d, JC-F = 22.0 Hz, C-3’ and C-5’), 122.46, 124.23,
125.59, 125.90, 127.93 (d, JC-F = 3.0 Hz, C-1’), 128.23 (d, JC-F = 8.2 Hz, C-2’ and C-6’),
1
130.23, 130.39, 130.65, 134.06, 140.85, 172.45 (d, JC-F = 140.0 Hz, C-4’). 2D H NMR
This article is protected by copyright. All rights reserved.

NOESY spectra showed two relevant NOE cross-peaks: N-1 CH3→H-9; N-1 CH3→H-8.
HRMS (ESI) m/z [M+H]+ calcd for C17H14FN3O2S: 344.0870, found: 344.0863; LC-MS: ret.
time 11.93 min.
1
Compound 5 (0.07 g, 33%): mp 212-215 °C. H NMR (400 MHz, DMSO-d6): δ 2.80 (s, 3H,
N-4 CH3), 3.90 (s, 3H, N-2 CH3), 7.25-7.55 (m, 2H, H-3’ and H-5’), 7.65-7.75 (m, 3H, H-7, H-
2’, H-6’), 7.85 (t, J = 7.5 Hz, 1H, H-8), 7.90 (d, J = 7.8 Hz, 1H, H-6), 8.10 (d, J = 7.6 Hz, 1H,
H-9). 13C NMR (100 MHz, DMSO-d6): δ 38.82, 39.37, 116.80 (d, JC-F = 22.0 Hz, C-3’ and C-
5’), 123.09, 123.88, 123.92, 124.74, 128.02, 129.68, 131.84, 131.95 (d, JC-F = 9.0 Hz, C-2’
and C-6’), 133.96, 135, 74, 136.73, 165.04 (d, JC-F = 140.0 Hz, C-4’). HRMS (ESI) m/z
[M+H]+ calcd for C17H14FN3O2S: 344.0870, found: 344.0867; LC-MS: ret. time 12.08 min.
(4-Fluorophenyl)(4-hydroxy-2-methyl-1,1-dioxido-2H-1,2-benzothiazin-3-yl)methanone
(12). A mixture of intermediate 9 (0.39 g, 1.2 mmol) in 1N NaOH/EtOH (1:4) (10 mL) was
stirred at room temperature untill a red solution was obtained. Then, MeI (0.23 mL, 3.7
mmol) was added dropwise and the solution was stirred for 3h at room temperature and after
diluted with 0.5 N HCl (20 mL) giving a precipitate which was filtered and dried. Compound
12 was obtained (0.33 g, 82%) as yellow solid and used as is for the next reacion step: mp
161-163 °C. 1H NMR (200 MHz, DMSO-d6): δ 2.70 (s, 3H, CH3), 7.40-7.55 (m, 2H, H-3’ and
H-5’), 7.90-8.05 (m, 3H, H-5, H-6, and H-7), 8.10-8.25 (m, 3H, H-8, H-2’, and H-6’).
3-(4-Fluorophenyl)-4-methyl-1,4-dihydropyrazolo[4,3-c][1,2]benzothiazine 5,5-dioxide
(6). In neat conditions compound 12 (0.37 g, 1.1 mmol) and hydrazine monohydrate (0.27
mL, 5.5 mmol) were heated at 70 °C under sonication for 20 min. The excess of hydrazine
monohydrate was removed under vaccum distillation and the residue was poured into
ice/water and acidified with 2N HCl (pH=3) to give an orange precipitate which was filtered,
dried and purified by flash column chromatogrphy eluting with CHCl3/MeOH (99:1) to give
derivative 6 (0.20 g, 55%), as pale yellow solid: mp 195-197 °C. 1H NMR (200 MHz, DMSO-
d6): δ 2.80 (s, 3H, CH3), 7.40-7.50 (m, 2H, H-3’ and H-5’), 7.60-8.10 (m, 6H, H-6, H-7, H-8,
H-9, H-2’, and H-6’). HRMS (ESI) m/z [M+H]+ calcd for C16H12FN3O2S: 330.0713, found:
330.0707; LC-MS: ret. time 10.62 min.
Biological assays
p38α MAPK assay (32) and human whole blood (HWB) assay (33) were performed
following a procedure previously reported. Experimental details of the two methods
employed are reported in the Supporting Information.
This article is protected by copyright. All rights reserved.

The kinase functional assays were performed in 384-well plates using a 30 μL reaction
volume. The reactions were run in reaction buffer. The reaction was initiated by combination
of kinase (Carna Biosciences) and 1.5 μM peptide substrate in the presence of ATP. The
reaction mixture was analyzed on the Caliper EZ Reader (Perkin Elmer) by the
electrophoretic mobility shift of the fluorescent substrate and phosphorylated product.
Inhibition data were calculated by comparison to no enzyme control reactions for 100%
inhibition and DMSO-only reactions for 0% inhibition. Dose-response curves were generated
to determine the concentration required to inhibit 50% of kinase activity (IC50).
Staurosporine was included as a positive control of each enzymatic assay. Compounds 1d
and 6 were also evaluated with this method on p38α MAPK yielding to an IC50 value of 0.31
µM and 1.9 µM, respectively, consistent with data reported in Table 2.
The concrete conditions for each enzymatic reactions are reported in the Supporting
Information (Table S1).
Computational methods
Docking studies were performed using the software AutoDock 4.2 (AD4.2) (34). The crystal
structure of p38α MAPK complexed with inhibitor SB203580 (PDB code 3ZS5) (35) was
retrieved from the RCSB Protein Data Bank and used as target structure for our modeling
studies.
Before the docking run, the complex was submitted to Schrödinger's Protein Preparation
Wizard (36,37): water molecules were deleted, hydrogen atoms were added, bond orders
and charges were then assigned, the orientation of hydroxyl groups on Ser, Thr and Tyr, the
side chains of Asn and Gln residues, and the protonation state of His residues were
optimized. The inhibitor SB203580 was then extracted from the complex and its coordinates
were transformed in order to exclude any biasing toward the experimental protein-bound
conformation.
Pyrazolobenzothiazine derivatives 1d, 1k, 2d, 2k, 3k and 6 were built using the Schrödinger
Maestro Interface (38) and then submitted to Polak−Ribiere conjugate gradient minimization
[0.0005 kJ/(Å mol) convergence].
Prior to docking, particular attention was directed to the ionization state of the sulfonamide
nitrogen of derivatives 1d, 1k and 3k, and the possible tautomeric state of the pyrazole ring
in derivative 6. These inhibitors were thus processed by MoKa (39,40), a software able to
accurately predict pKa values and tautomerism in the aqueous medium. The in silico
analysis indicated pKa values of 4.64 (1d and 1k) and 4.16 (3k) for the sulfonamide group;
at pH=7-7.5, this moiety thus exists only in the anionic form (98% abundance). Therefore,
This article is protected by copyright. All rights reserved.

we considered the sulfonamide moiety of compounds 1d, 1k and 3k as bearing negatively
charged nitrogen. Furthermore, the ligand pyrazole moiety of 6 was predicted to exist in
solution as an equilibrium between two tautomeric forms (1H- and 2H-pyrazole), which were
both taken in consideration for the computational study.
Ligands and receptor structures were converted to AD4 format files using AutoDockTools
(41), and then the Gesteiger−Marsili partial charges were then assigned. The dimensions of
the grid were 60 × 60 × 60 with grid points separated by a 0.375 Å. The grid was centered
on the ATP binding site, using the SB203580 crystallographic position as reference. The
Lamarckian genetic algorithm local search method was used, and for each compound the
docking simulation was composed of 100 runs. Clustering of docked conformations was
performed on the basis of their root-mean-square deviation (rmsd) (tolerance = 2.0 Å), and
the results were ranked based on the estimated free energy of binding. The cluster analysis
revealed a predominant ligand orientation for all studied compounds (89, 74, 74, 51, 74, 100
and 67 conformations in the first ranked cluster for 1d, 1k, 2d, 2k, 3k, and 1H- and 2H-
pyrazole tautomers of 6, respectively), and the most energetically favorable conformation for
each ligand was chosen for further analysis. Figures 3 and 4A-6A were generated using the
Ligand Interaction Diagram tool of Maestro GUI (38), whereas Figure 4B-6B were prepared
using the software PyMOL (42).
The program VolSurf+ (43,44) was used to calculate the physicochemical and ADME
properties of the analyzed compounds.
Results and Discussion
Synthesis of pyrazolobenzothiazines 1a-k, 2a-k, and 4-6.
The synthetic pathways utilized in the preparation of the pyrazolobenzothiazines 1a (45), 1b-
k, 2a (45), 2b-k and 4-6 are outlined in Schemes 1 and 2, whilst compounds belonging to
the subset 3 (Table 1) were already described and prepared according to the literature (46).
N-1 phenylpyrazolobenzothiazines 1a (45) and 1b-i were obtained through a regioselective
condensation (47) of appropriate phenylhydrazine hydrochloride with the key synthon 7 (30)
(Scheme 1). Initially, this reaction step was carried out employing the procedure described
for the preparation of derivatives 3 (46), which entails the condensation of intermediate 7
with phenylhydrazines hydrochloride using classical conditions such as EtOH at reflux with
traces of H2SO4 as dehydrating agent. This procedure suffers from very long time (several
days) and moderate to low yields, depending on the utilized substrate. Herein, we report a
more advantageously method to obtain compounds 1a-i. In particular, microwave (MW)
irradiation of a mixture of benzothiazine 7 and appropriate phenylhydrazine hydrochloride in
This article is protected by copyright. All rights reserved.

DMF at 100 °C afforded in good yield (60-70%) and very short time (5-75 min) the desired
pyrazolobenzothiazines 1a-i. Target nitro derivative 1i was then hydrogenated under
catalytic conditions to amino derivative 1j, which was further elaborated to
sulfonamidomethyl derivative 1k by a mesylation reaction.
On the other hand, the target derivatives 2a (45), 2b-g and 2i (Scheme 1) were obtained
from 1a-g and 1i, respectively by methylation in DMF using MeI and NaH as base.
Scheme 1. Synthesis of pyrazolobenzothiazine subsets 1 and 2. Reagents and conditions:
a) ArNH2NH2·HCl, DMF, 100 °C, MW; b) H2, Raney-Ni, rt, atm pressure; c) MsCl,
CH2Cl2/pyridine (1:1), 50 °C; d) MeI, NaH, dry DMF, rt (for 2a-g and 2i) or e) MeI, K2CO3, dry
DMF, rt (for 2h).
Methylation of compound 1h to give derivative 2h was instead realized replacing NaH with
K2CO3 in order to avoid the formation of exocyclic N-methylated sulfonamide group. In
addition, starting from derivative 2i and employing the same conditions as used for the
preparation of 1j and 1k, compounds 2j and 2k were obtained.
This article is protected by copyright. All rights reserved.

The preparation of target derivatives 4-6 (Scheme 2) started from the commercially available
saccharin sodium salt, which was reacted with 2-bromo-1-(4-fluorophenyl)ethanone in
DMSO at 70 °C to give benzoisothiazole 8 (31) in good yield.
It was then expanded to benzothiazine 9 (31) in EtOH and using NaOEt at 60 °C.
Intermediate 9 was reacted with methylhydrazine in EtOH under MW irradiation at 100 °C to
give a nearly equimolecular ratio of N-1-methyl- and N-2-methyl-pyrazolobenzothiazine
regioisomers 10 and 11. This mixture resulted very difficult to separate by flash
chromatography, so that it was used as is for the methylation reaction. This latter reaction
was conducted in DMF in presence of MeI and NaH affording N-4-metylated targets 4 and 5,
respectively, which were separated by flash chromatography. The molecular structure of 4
was in depth analyzed performing a 2D NMR spectrum which showed two diagnostic
interactions observed between the H-8 and the H-9 protons of pyrazolobenzothiazine
nucleus with the N-1 methyl group. No interactions were found between the N-1 methyl and
the H-2’ and H-6’ of the C-3 phenyl ring (Figure 2).
Figure 2. NOESY experiments for 4 showed two main interactions: N-1- Me→H-8 and N-1-
Me→H-9.
On the other hand, derivative 9 was also selectively methylated at N-4 position to give
intermediate 12 employing a mixture of 1N NaOH/EtOH and MeI at room temperature as
reported for similar compounds (48). Finally, intermediate 12 was condensed with hydrazine
monohydrate at 70 °C under sonication, following the procedure reported for other
pyrazolobenzothiazines (49), to give the target compound 6.
This article is protected by copyright. All rights reserved.

Scheme 2. Synthesis of pyrazolobenzothiazines 4-6. Reagents and conditions: a)
BrCH2CO(4-FC6H5), dry DMSO, 70 °C; b) NaOEt, EtOH, 55 °C; c) MeNHNH2, EtOH, 100 °C,
MW; d) MeI, NaH, dry DMF, rt; e) MeI, 1N NaOH, EtOH, rt; f) NH2NH2 H2O, 70 °C, sonication.
Biological evaluation of compounds 1-6
Target compounds 1a-k, 2a-k, 3a-c, 3f-k and 4-6 were evaluated for their ability to inhibit
both p38α MAPK activity and TNF-α release in lipopolysaccharide (LPS) stimulated in
human whole blood (HWB), and the results are reported in Table 1.
The biochemical test determines the ability of a compound to compete with ATP for its
binding at the p38α MAPK catalytic domain. The phosphorylation of activating transcription
factor-2 (ATF-2) was determined with an anti-phospho-ATF-2 antibody. The degree of
phosphorylation inversely correlated with the inhibitory activity of the tested compound (32).
In this assay, compounds were considered primary actives only when the inhibition of the
p38α MAPK phosphorylation activity was reduced at least of 65% employing 10 µM
compound concentration. The compounds selected as primary actives were tested in dose
response assay to calculate the 50% inhibitory concentration (IC50) values. SB203580 was
included as an internal reference and inhibited p38α MAPK with an IC50 value of 0.05
0.001 µM, consistent with previously reported data (25).
The HWB assay was used to determine the anti-inflammatory activity of the tested
compounds. This assay is commonly used also because it takes into account the behavior of
a compound in the physiological conditions (e.g. water solubility at physiological pH, plasma
protein binding, cell permeability, ATP concentration, cell metabolism, etc.). Reduction in the
This article is protected by copyright. All rights reserved.

TNF-α release was determined by ELISA method after incubation of HWB with LPS (33).
The IC50 were determined as the concentration required to reduce of the 50% the TNF-α
concentration.
Results from the enzymatic assay confirmed the initial hypothesis that p38α MAPK could be
effectively targeted by the pyrazolobenzothiazine chemotype. In fact, derivatives 1d, 1f, 1i,
1k, 3a-c, 3f, 3h, 3i, 3k, and 4-6 emerged as effective inhibitors with IC50 values ranging from
0.43 to 7.73 µM (Table 1).
Table 1. p38α MAPK and TNF-α inhibition of pyrazolobenzothiazines 1-6.
% inh.
IC50 (µM)
IC50 (µM)
cpds
1a
R1
R3
R4
H
H
H
H
H
H
p38αa
p38αb
TNF-α HWBc
Me
Me
Me
Me
Me
Me
26.4
NCd
92.3 5.6
1b
1c
52
NCd
>100
46.8
91
NCd
>100
1d
1e
0.936 0.296
NCd
18 3.5
>100
24.5
74.7
1f
0.967 0.137
>100
1g
1h
1i
Me
Me
Me
Me
Me
Me
H
H
41.3
60.1
67.8
30
NCd
NCd
>100
>100
>100
>100
20 2.6
>100
H
2.784 0.694
NCd
1j
H
1k
2a
H
85.1
14.4
0.593 0.099
NCd
Me
2b
2c
2d
Me
Me
Me
Me
Me
Me
19.5
36.4
27.7
NCd
NCd
NCd
>100
>100
31 1.8
This article is protected by copyright. All rights reserved.

2e
2f
Me
Me
Me
Me
35.6
16.8
NCd
NCd
>100
>100
2g
2h
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
H
26.6
15
NCd
NCd
25 2.1
>100
>100
>100
>100
>100
2i
28.2
25.5
33.9
65.8
NCd
2j
NCd
2k
NCd
3a(46)
7.727 1.196
3b(46)
3c(46)
3f(46)
H
H
H
79.5
87
2.863 0.169
1.949 0.314
3.332 0.168
52 2.5
43 2.8
>100
75.3
3g(46)
3h(46)
3i(46)
3j(46)
3k(46)
4
H
H
51.5
78.8
76.5
50
NCd
82.7 5.2
>100
2.749 0.294
3.529 0.119
NCd
H
51 3.2
>100
H
H
81.7
96
1.124 0.357
0.431 0.053
32.4 2.8
>100
Me
Me
5
6
82
95
2.424 0.154
0.457 0.054
12 1.8
0.5 0.1
H
Me
[
a] The percentage of inhibition was determined employing a 10 µM compound solution and evaluating the
phosphorylation state of the ATF-2 substrate. The degree of phosphorylation inversely correlates with the
inhibitory activity of the tested compound. Only compounds that were able to inhibit the kinase activity by a
percentage ≥ 65% were further analyzed to determine the IC50. [b] IC50 represents the compound
concentration necessary to reduce by 50% the phosphorylation degree of ATF-2. Data are from three
separate experiments SD. [c] IC50 represents the compound concentration required to reduce by 50% the
TNF-α levels employing ELISA method. Data are from three separate experiments SD. [d] NC= not
calculated.
The p38α MAPK inhibition data analysis suggested a few SAR trends. Focusing the
attention on the subset 1, it emerged that compounds 1d, 1f, 1i, and 1k, bearing a N-4 acidic
endocyclic sulfonamide function, showed promising activity with IC50 values going from 0.59
This article is protected by copyright. All rights reserved.

to 2.78 µM. When this N-4 nitrogen atom was methylated, as in subset 2, the inhibitory effect
on the p38α MAPK activity dramatically decreased (compare 2d, 2f, 2i, and 2k vs 1d, 1f, 1i,
and 1k, respectively). Overall, derivatives belonging to the subset 2 showed only a very
weak effect at 10 µM concentration with percentages of inhibition ranging from 14 to 36. At
the same time, with the exception of two compounds, pyrazolobenzothiazines belonging to
subset 3 showed good p38α MAPK inhibitory activity. Although they differed for the
substituent at C-3 position of the pyrazolobenzothiazine scaffold, compounds of subset 3
had the same acidic sulfonamide feature as in subset 1, indicating that the presence of the
N-4 acidic sulfonamide moiety appears to be very important to ensure p38α MAPK inhibitory
activity.
When comparing subsets 1 and 3, the influence of the substituents placed at C-3 position
was not well attributable because either the methyl group or the para-chlorophenyl ring gave
active compounds. In some cases the presence of the C-3 para-chlorophenyl ring permitted
a good activity to be recovered with respect to the parent C-3 methyl derivatives of subset 1
(i.e. 3a-c and 3h vs 1a-c, and 1h, respectively). Independently from the C-3 substituent, the
presence of a meta-trifluoromethyl, para-nitro and para-sulfonamidomethyl groups at N-1
phenyl ring granted active compounds in both 1 and 3 subsets.
A different behaviour was observed in subsets 4-6 characterized by the absence of the N-1
phenyl ring in the pyrazolobenzothiazine nucleus coupled with the presence of a C-3 para-
fluorophenyl group and a methyl group at the N-4 position. The good inhibitory activity
obtained in the absence of the acidic endocyclic sulfonamide moiety was in contrast with the
SAR observed for subsets 1-3. This distinctive behaviour suggested that
pyrazolobenzothiazines 4-6 could present different interacting residues and/or binding
modes within the p38α MAPK ATP binding site compared to derivatives of subsets 1 and 3.
It is worth noting that compounds 4 and 6 displayed the best IC50 values among the whole
set of pyrazolobenzothiazine derivatives.
The HWB assay revealed that derivative 6 showed the best activity, with an IC50 value of 0.5
µM, in perfect agreement with its p38α inhibition potency. Although derivatives 1d, 1k, 3b,
3c, 3i, 3k and 5 exhibited promising inhibitory activity in the p38α MAPK assay, they were
only moderately active in reducing TNF-α release (HWB assay). Additionally, no correlation
was observed between the potency obtained in the enzyme and in the TNF-α release
assays for compounds 1f, 1i, 3a, 3f, 3h, and 4. This behaviour could be explained either by
the relative low potency of some compounds in p38α MAPK assay or by considering that the
HWB assay differs from the in vitro enzyme assay by additional factors, such as plasma
protein binding, cell permeability, ATP concentration, various cell types and possible
metabolism, all influencing heavily the activity of the tested compounds in cells. The results
of the HWB assay also demonstrated that compounds 2d and 2g were moderate inhibitors
This article is protected by copyright. All rights reserved.

of TNF-α release. The absence of activity of these two derivatives against the isolated p38α
suggested that other targets are probably involved in their ability to block TNF-α production.
Overall, the data indicated derivative 6 (IC50s of 0.457 µM and 0.5 µM on biochemical and
HWB assays, respectively) as effective p38α MAPK inhibitor within the explored series of
pyrazolobenzothiazines. Although compound 6 was less potent than SB203580, it turned out
to be an efficient inhibitor, as highlighted by the corresponding ligand efficiency (LE) and
lipophilic efficiency (LipE) values (Table 2). In fact, derivative 6 had LE value greater than
0.3, that is the generally accepted lower limit of efficiency in lead discovery and optimization.
Moreover, although LipE greater than 5 is usually considered optimal for a promising drug
candidate, mean LipE values of 2.5 and 3.6 have been reported for collection of HTS hits
and leads, respectively (50).
Table 2. LE values, and predicted LipE, physicochemical and ADME parameters
Cpds
SB203580
6
LEa
0.37
0.38
LipEb MWc
PSAd
80.2
75.2
LgD7.5e LgS7.5f
PBg
92.8
74.4
CACO2h MetStabi
4.1
377.4
3.2
-5.2
0.7
47.1
3.4
329.3
3.0
-3.9
0.8
57.0
[a] LE: ligand efficiency in units of kcalmol-1/heavy atom. LE= 1.4pIC50/N; N is the number of heavy atoms (27 for
SB203580 and 23 for 6, respectively) [b] LipE = pIC50 − predicted logD7.5. [c] MW: molecular weight. [d] PSA: polar
surface area. [e] LgD7.5: logarithm of the octanol/water partition coefficient at pH=7.5. [f] LgS7.5: logarithm of the aqueous
solubility at pH=7.5 [g] PB: % of plasma protein binding. [h] CACO2: Caco-2 cell (human intestinal epithelial cell line
derived from a colorectal carcinoma) permeability. This value is qualitative, and the compounds are classified in
penetrating (CACO2+) or having little if any ability to penetrate the epithelial cells (CACO2-). [i] MetStab is the percentage
of the remaining compound after incubation with human CYP3A4 enzyme. Values greater than 50 indicate stable
behaviour.
It is also worth noting that the emerged hit compound showed satisfactory in silico
determined physicochemical and ADME properties (43,44), comparable to those of the
reference compound SB203580 (Table 2); this analysis further validated compound 6 as a
good starting point for future hit-to-lead optimization work.
Based on the results of the biological evaluations, the next step of this research included
computational studies aimed at both suggesting a plausible binding mode for this new class
of inhibitors and providing some insights to support the SAR analysis. The in silico results
are reported in the following section.
Binding Mode Analysis
Compound SB203580 is the prototype of the class of p38α MAPK pyridinylimidazole
inhibitors, which have been shown to act as ATP-competitors (28). Crystal structures of
This article is protected by copyright. All rights reserved.

SB203580 bound to p38α MAPK showed that the pyridine ring of the inhibitor occupied the
adenine-binding region, with the nitrogen atom accepting a hydrogen bond from the NH
backbone of Met109 in the hinge region (residues 106-110) (35,51-53) (Figure 3). The ligand
imidazole core showed π-π interaction with Phe169, with its N-3 atom interacting with the
Lys53 side chain. The para-fluorophenyl ring was buried into the hydrophobic region I (HRI),
which is mainly defined by residues Ala51, Leu75, Iso84, Leu86, Val105, Leu104 and
Thr106. Of note, the small gatekeeper residue Thr106 in p38α MAPK provides access to this
pocket, thus playing a key role in kinase inhibitor selectivity, as related kinases possess
bulkier residue at the same position.
Finally, the para-methylsulfinylphenyl group occupied the phosphate-binding region, with the
phenyl ring making π-π interaction with Tyr35 (Figure 3).
Figure 3. Schematic representation of the interactions between the pyridinylimidazole
inhibitor SB203580 and p38α MAPK ATP binding site (PDB ID 3ZS5 as representative
complex). Protein residues lying within a distance of 4 Å from the bound ligand are shown.
Based on the results of the biological evaluations, the most interesting pyrazolobenzothazine
derivatives (i.e., 1d, 1k, 3k, and 6) were selected to investigate the possible ligand-p38α
MAPK interactions by means of automated docking experiments. The N-4-methylated
inactive derivatives 2d and 2k were analysed as well to try to explain the key role of the N-4
acidic sulfonamide moiety in subset 1 compared to subset 2.
Self-docking calculations of SB203580 reproduced the experimental binding conformation
(RMSD equal to 0.5 Å) as a single solution (100 conformations in one cluster), confirming
the good performance of the AutoDock 4.2 software; notably, the predicted inhibition
This article is protected by copyright. All rights reserved.

constant (Ki-pred) value (0.024 μM) was in perfect agreement with the experimental IC50 value
(0.05 μM).
The docking poses of 1d (Ki-pred=0.3 μM) and 1k (Ki-pred=4 μM) were comparable and
perfectly aligned to SB203580, as shown in Figure 4 using 1d as representative compound.
Similarly to the imidazole core of the known inhibitor, the pyrazole nucleus was able to
establish a hydrogen-bonding with Lys53 and stacking interaction with the phenyl side chain
of Phe169. Key intermolecular interactions were also identified between the inhibitor meta-
chlorophenyl moiety and HRI residues.
Figure 4. (A) Simplified 2D representation of the predicted interactions between derivative
1d and the p38α MAPK ATP binding site. (B) Docking pose of compound 1d (violet) together
with the experimental position of SB203580 (green). The hinge region (orange) and the
residues (wheat) establishing hydrogen bond interactions (black dashed lines) with the
ligand are highlighted.
The docking results for the inactive N-methylated derivatives 2d and 2k suggested that the
presence of a methyl group at N-4 position could prevent these compounds by establishing
the crucial intermolecular interactions needed to inhibit p38α MAPK enzymatic activity. In
fact, compounds 2d and 2k only partially occupied the ATP-binding site, and differently from
subset 1, they lacked the key interactions with residues defining HRI (see Supporting
Information).
The Autodock-conformation of active compound 3k (Ki-pred=0.71 μM, Figure 5) was
somewhat similar to that of SB203580, but different with respect to the one observed for the
subset 1. In fact, whilst in the latter compounds the HRI was occupied by the phenyl
substituent at the N-1 position, in subset 3 this hydrophobic pocket accommodated the para-
chlorophenyl substituent at the C-3 position, which also showed π-cation interaction with
This article is protected by copyright. All rights reserved.

Lys53. Furthermore, one of the sulfone oxygen atoms served as H-bond acceptor to the
main chain nitrogen of the hinge region residue Met109 in a manner analogous to the
pyridine nitrogen of the known p38α MAPK inhibitor SB203580, whereas the pyrazole ring
established aromatic stacking interaction with Phe169. An additional polar contact was
engaged between the exocyclic sulfonamide fragment and the nitrogen atom of the Gly170
backbone.
Figure 5. (A) Schematic representation of the predicted interactions between compound 3k
and the p38α MAPK residues. (B) Predicted binding mode of 3k (violet) into the ATP binding
site. Experimental position of SB203580 is depicted in green sticks as reference.
Intermolecular hydrogen bonds are shown as black dashed lines, whereas the hinge region
is illustrated in orange.
Finally, the docking solutions for the two tautomers of 6 showed a conserved ligand pose
superimposable to that observed for derivative 3k (Figure 6). In particular, one of the sulfone
oxygens of the ligand interacted with Met109, the para-fluorophenyl group was located in the
HRI, and the pyrazole ring formed a stacking interaction with Phe169. Similarly to the N-3
atom of the SB203580 imidazole nucleus, the N-2 atom of 1H-pyrazole tautomer of
derivative 6 (Ki-pred=0.67 µM) formed an anchoring H-bond with Lys53; this contact could not
be observed in the docking conformation of 2H-pyrazole tautomer (Ki-pred=1.89 µM) where
the same nitrogen atom was protonated.
Interestingly, the lack of interaction with Lys53 might be also responsible of the lower activity
of the N-2-methylated derivative 5 (IC50=2.43 µM) compared to the strict N-unsubstituted
analogue 6 (IC50=0.455 µM) (see Supporting Information).
This article is protected by copyright. All rights reserved.

Figure 6. (A) Simplified 2D representation of the predicted interactions between derivative 6
(tautomer 1H-pyrazole as example) and the p38α MAPK residues. (B) Docking pose of
compound 6 (violet) together with the experimental position of SB203580 (green). Residues
defining the hinge region are shown in orange, while the hydrogen bonds are represented as
dashed black lines.
Kinase selectivity assessment
Spurred by the interesting biological results and with the aim to further validate the potential
of the pyrazolobenzothiazines as a new class of p38α-MAPK inhibitors, we decided to
evaluate the best compounds 1d and 6 against a panel of 14 kinases available in our hands
(Table 3). This hit verification assay is critical for highlighting potential undesirable side
effects due to the inhibition of other kinases, thus leading to the identification of potential off-
targets.
Data from inhibition of other MAPKs showed that compounds 1d and 6 did not inhibit Erk1,
p38δ and p38γ, whereas they were able to block the catalytic activity of p38β. These results
could be explained considering that, differently from p38α and p38β, the other three MAPKs
show a gatekeeper residue larger than Thr, which could prevent the accommodation of the
meta-chlorophenyl (1d, Figure 4) and para-fluorophenyl (6, Figure 6) rings in a suitable
hydrophobic region. Interestingly, derivatives 1d and 6 did not inhibit the activity of
structurally distinct CDC2, CDK5, JAK2, LCK, SRC, and PKAC-α kinases. Among all the
This article is protected by copyright. All rights reserved.

kinases evaluated, only JAK3 seemed to be significantly inhibited by compound 1d. On the
contrary, derivative 6 was unable to inhibit the same kinase over 70% at 100 µM. A weak
activity was instead observed for both derivatives against PDGFRb and GSK3-β kinases,
although no complete inhibition of the enzymatic activity was observed.
Overall, we can conclude that derivative 6 has a reasonable kinase selectivity profile within
the explored set of kinases.
Table 3. Kinase selectivity profile of derivatives 1d and 6
kinase
1d
6
%inh. @100 µMa IC50 (µM)b %inh. @100 µMa IC50 (µM)b
CDC2
CDK5
Erk1
JAK2
JAK3
21 1
23 1
30 1
33 1
79 5
1 1
ND
ND
ND
27 3
30 4
20 4
23 3
52 1
2 2
ND
ND
ND
ND
ND
ND
2.5c
ND
LCK
ND
SRC
6 1
ND
16
ND
3 2
ND
1.7
ND
p38-β
p38-δ
p38-γ
PDGFRb
GSK3-β
PKAC-α
100 2
22 3
51 1
82 2
74 4
7 2
100 2
25 1
13 1
100 1
78 3
26 4
ND
ND
82.9c
24.9c
ND
58.1c
48.4c
ND
[a] a dose-response curve was generated only for compounds reaching 70%
inhibition of kinase activity. [b] IC50 represents the concentration required to
reduce by 50% the kinase activity expressed as the ability to phosphorylate
a specific substrate in presence of ATP. [c] The compound does not
completely inhibit the enzymatic activity. IC50 value extrapolated by the
analysis software might be not accurately estimated.
Conclusion and future directions
The availability of in-house pyrazolobenzothiazine-based compounds led to the identification
of a novel chemical series of p38α MAPK inhibitors.
Some pyrazolobenzothiazines exhibited promising biological activity in both enzyme and
HWB assays, with 6 resulting the most potent compound. The low molecular weight, ligand
efficiency values and kinase selectivity profile of pyrazolobenzothiazine 6 make this
derivative an attractive starting point for chemistry efforts aimed at further potency
This article is protected by copyright. All rights reserved.

optimization. For instance, it is well known that the potency and selectivity of a moderate
p38α inhibitor can be improved by establishing additional interactions with the less-
conserved solvent-exposed front region (hydrophobic region II) of the protein.(23). Therefore,
taking into account the above-described binding mode, we figure it could be possible to
optimize the hit compound 6 by adding suitable substituents on the phenyl ring of the
pyrazolobenzothiazine scaffold.
Acknowledgements
We are grateful to Dr. Davide Martino for assisting with HWB assay and to Dr. Nunzio Iraci
for technical assistance in computational studies and critical reading of the manuscript.
References
1. Schieven, G. L. (2005) The biology of p38 kinase: a central role in inflammation. Curr Top
Med Chem 5: 921-928.
2. Saklatvala, J. (2004) The p38 MAP kinase pathway as a therapeutic target in
inflammatory disease. Curr Opin Pharmacol 4: 372-377.
3. Feng, Y. J. Li, Y. Y. (2011) The role of p38 mitogen-activated protein kinase in the
pathogenesis of inflammatory bowel disease. J Dig Dis 12: 327-332.
4. Patterson, H. Nibbs, R. McInnes, I. Siebert, S. (2014) Protein kinase inhibitors in the
treatment of inflammatory and autoimmune diseases. Clin Exp Immunol 176: 1-10.
5. Wang, G. Pan, J. Chen, S. D. (2012) Kinases and kinase signaling pathways: potential
therapeutic targets in Parkinson's disease. Prog Neurobiol 98: 207-221.
6. Namiki, K. Matsunaga, H. Yoshioka, K. Tanaka, K. Murata, K. Ishida, J. Sakairi, A. Kim, J.
Tokuhara, N. Shibakawa, N. Shimizu, M. Wada, Y. Tokunaga, Y. Shigetomi, M. Hagihara,
M. Kimura, S. Sudo, T. Fukamizu, A. Kasuya, Y. (2012) Mechanism for p38α-mediated
experimental autoimmune encephalomyelitis. J Biol Chem 287: 24228-24238;
7. Krementsov, D. N. Thornton, T.M. Teuscher, C. Rincon, M. (2013) The emerging role of
p38 mitogen-activated protein kinase in multiple sclerosis and its models. Mol Cell Biol
33: 3728-3734.
8. Xu, Y. Li, N. Xiang, R. Sun, P. (2014) Emerging roles of the p38 MAPK and
PI3K/AKT/mTOR pathways in oncogene-induced senescence. Trends Biochem Sci 39:
268-276.
This article is protected by copyright. All rights reserved.

9. Hong, B. Li, H. Zhang, M. Xu, J. Lu, Y. Zheng, Y. Qian, J. Chang, J. T. Yang, J. Yi, Q.
(2014). p38 MAPK inhibits breast cancer metastasis through regulation of stromal
expansion. Int. J. Cancer. doi: 10.1002/ijc.28958.
10. Koul, H. K. Pal, M. Koul, S. (2013) Role of p38 MAP Kinase Signal Transduction in Solid
Tumors. Genes Cancer 4: 342-359.
11. Zhou, W. Hu, W. (2013) Anti-neuroinflammatory agents for the treatment of Alzheimer's
disease. Future Med Chem 5: 1559-1571.
12. Munoz, L. Ammit, A. J. (2010) Targeting p38 MAPK pathway for the treatment of
Alzheimer's disease. Neuropharmacology 58: 561-568.
13. Coulthard, L. R. White, D. E. Jones, D. L. McDermott, M. F. Burchill, S. A. (2009)
p38(MAPK): stress responses from molecular mechanisms to therapeutics. Trends Mol
Med 15: 369-379
14. Cuadrado, A. Nebreda, A. R. (2010) Mechanisms and functions of p38 MAPK signaling.
Biochem J 429: 403-417.
15. Kumar, S. Boehm, J. Lee, J. C. (2003) p38 MAP kinases: key signaling molecules as
therapeutic targets for inflammatory diseases. Nature Rev Drug Discov 2: 717-726.
16. Wroblesky, S. T. Doweyko, A. M. (2005) Structural comparison of p38 inhibitor-protein
complexes: a review of recent p38 inhibitors having unique binding interactions. Curr Top
Med Chem 5: 1005-1016.
17. Wagner, G. Laufer, S. (2006) Small molecular anti-cytokine agents. Med Res Rev 26: 1-
62.
18. Margutti, S. Laufer, S. A. (2007) Are MAP kinases drug targets? Yes, but difficult ones.
ChemMedChem 2: 1116-1140.
19. Goldestein, D. M. Kuglstatter, A. Lou, Y. Soth, M. J. (2010) Selective p38alpha inhibitors
clinically evaluated for the treatment of chronic inflammatory disorders. J Med Chem 53:
2345-2353.
20. Schnieders, M. J. Kaoud, T. S. Yan, C. Dalby, K. N. Ren, P. (2012) Computational
insights for the discovery of non-ATP competitive inhibitors of MAP kinases. Curr Pharm
Des 18: 1173-1185.
21. Kong, T.-T. Zhang, C-M. Liu, Z.-P. (2013) Recent developments of p38α MAP kinase
inhibitors as antiinflammatory agents based on the imidazole scaffolds. Curr Med Chem
20: 1997-2016.
22. https://clinicaltrials.gov/ct2/results?term=p38&Search=Search (accessed September, 12,
2014).
23. Bühler, S. Laufer, S. A. (2014) p38 MAPK inhibitors: a patent review (2012-2013).
Expert Opin Ther Patents 24: 1-20.
This article is protected by copyright. All rights reserved.

24. Cecchetti, V. Fravolini, A. Schiaffella, F. De Regis, M. Orzalesi, G. Volpato, I. (1983)
Synthesis and Pharmacological Activity of Pyrazolo-1,2-benzothiazine- and Isoxazolo-
1,2-benzothiazinecarboxamides. Il Farmaco 38: 35−44.
25. Gallagher, T. F. Seibel, G. L. Kassis, S. Laydon, J. T. Blumenthal, M. J. Lee, J. C. Lee,
D. Boehm, J. C. Fier-Thompson, S. M. Abt, J. W. Soreson, M. E. Smietana, J. M. Hall, R.
F. Garigipati, R. S. Bender, P. E. Erhard, K. F. Krog, A. J. Hofmann, G. A. Sheldrake, P. L.
McDonnell, P. C. Kumar, S. Young, P. R. Adams, J. L. (1997) Regulation of stress-
induced cytokine production by pyridinylimidazoles; inhibition of CSBP kinase. Bioorg
Med Chem 5: 49-64.
26. Cuenda, A. Rouse, J. Doza, Y. N. Meier, R. Cohen, P. Gallagher, T. F. Young, P. R. Lee,
J. C. (1995) SB 203580 is a specific inhibitor of a MAP kinase homologue which is
stimulated by cellular stresses and interleukin-1. FEBS Letters 364: 229-233.
27. Tong, L. Pav, S. White, D. M. Rogers, S. Crane, K. M. Cywin, C. L. Brown, M. L. Pargelli,
C. A. (1997) A highly specific inhibitor of human p38 MAP kinase binds in the ATP pocket.
Nat Struct Biol 4: 311-316.
28. Young, P. R. McLaughlin, M. M. Kumar, S. Kassis, S. Doyle, M. L. McNulty, D.
Gallagher, T. F. Fisher, S. McDonnel, P. C. Carr, S. A. Huddleston, M. J. Seibel, G. Porter,
T. G. Livi, G. P. Adams, J. L. Lee, J. C. (1997) Pyridinyl imidazole inhibitors of p38
mitogen-activated protein kinase bind in the ATP site. J Biol Chem 272: 12116-12121.
29. Gum, R. J. McLaughlin, M. M. Kumar, S. Wang, Z. Bower, M. J. Lee, J. C. Adams, J. L.
Livi, G. P. Goldsmith, E. J. Young, P. R. (1998) Acquisition of sensitivity of stress-
activated protein kinases to the p38 inhibitor, SB 203580, by alteration of one or more
amino acids within the ATP binding pocket. J Biol Chem 273: 15605-15610.
30. Zinnes, H. Comes, R. A. Zulesky, F. R. Caro, A. N. Shavel, J. Jr. (1965) 1,2-
Benzothiazines. II. The preparation and sodium borohydride reduction of 3-acyl-2H-1,2-
benzothiazin-4(3H)-one 1,1-dioxides. J Org Chem 30: 2241-2246.
31. Hasegawa, T. Motoizumi, M. Kobayashi M (Sankyo, Co, Ltd, Japan) (2007)
Benzothiadiazine derivatives. Japanese Pat. JP 2007197369A.
32. Goettert, M. Graeser, R. Laufer, S. A. (2010) Optimization of a nonradioactive
immunosorbent assay for p38alpha mitogen-activated protein kinase activity. Analyt
Biochem 406: 233-234.
33. Koch, P. Baeuerlein, C. Jank, H. Laufer, S. (2008) Targeting the ribose and phosphate
binding site of p38 mitogen-activated protein (MAP) kinase: synthesis and biological
testing of 2-alkylsulfanyl-, 4(5)-aryl-, 5(4)-heteroaryl-substituted imidazoles. J Med Chem
51: 5630-5640.
This article is protected by copyright. All rights reserved.

34. Morris, G. M. Huey, R. Lindstrom, W. Sanner, M. F. Belew, R. K. Goodsell, D. S. Olson,
A. J. (2009) Autodock4 and AutoDockTools4: automated docking with selective receptor
flexiblity. J Comput Chem 16: 2785-91.
35. Azevedo, R. van Zeeland, M. Raaijmakers, H. Kazemier, B. de Vlieg, J. Oubrie, A.
(2012) X-ray structure of p38α bound to TAK-715: comparison with three classic inhibitors.
Acta Cryst D68: 1041-1050.
36. Protein Preparation Wizard 2013-3; Epik version 2.4, Schrödinger, LLC, New York, NY,
2013; Impact version 5.9, Schrödinger, LLC, New York, NY, 2013; Prime version 3.2,
Schrödinger, LLC, New York, NY, 2013.
37. Sastry, G. M. Adzhigirey, M. Day, T. Annabhimoju, R. Sherman, W. (2013) Protein and
ligand preparation: parameters, protocols, and influence on virtual screening enrichments.
J Comput Aided Mol Des 27: 221-234.
38. Maestro, version 9.6; Schrödinger, LLC: New York, NY, 2013.
39. Milletti, F. Storchi, L. Sforna, G. Cruciani, G. (2007) New and original pKa prediction
method using grid molecular interaction fields. J Chem Inf Model 47: 2172−2181.
40. Cruciani, G. Milletti, F. Storchi, L. Sforna, G. Goracci, L. (2009) In silico pKa prediction
and ADME profiling. Chem Biodivers 6:1812−1821.
41. Sanner, M. F. (1999) Python: a programming language for software integration and
development. J Mol Graphics Mod 17: 57-61
42. The PyMOL Molecular Graphics System, version 1.5.0.4; Schrödinger, LLC: New York,
2013.
43. VolSurf+. http://www.moldiscovery.com/software/vsplus
44. Cruciani, G. Crivori, P. Carrupt, P. A. Testa, B. Molecular fields in quantitative
structure−permeation relationships: the VolSurf approach. (2000) J Mol Struct
(Theochem) 503: 17−30.
45. Shavel, J. Jr. Zinnes H. (1967) Pyrazolo [4,3-c] [1,2] benzothiazine 5, 5-dioxides and
process for their production (Warner-Lambert Pharmaceutical Co; Rockaway, New Jersey,
USA), US Pat. US 3346572.
46. Sabatini, S. Gosetto, F. Serritella, S. Manfroni, G. Tabarrini, O. Iraci, N. Brincat, J. P.
Carosati, E. Villarini, M. Kaatz, G.W. Cecchetti, V. (2012) Pyrazolo[4,3-
c][1,2]benzothiazines 5,5-dioxide: a promising new class of Staphylococcus aureus NorA
efflux pump inhibitors. J Med Chem 55: 3568-3572.
47. Barreca, M. L. Manfroni, G. Leyssen, P. Winquist, J. Kaushik-Basu, N. Paeshuyse, J.
Krishnan, R. Iraci, N. Sabatini, S. Tabarrini, O. Basu, A. Danielson, U. H. Neyts, J.
Cecchetti, V. (2013) Structure-based discovery of pyrazolobenzothiazine derivatives as
inhibitors of hepatitis C virus replication. J Med Chem 56: 2270-2282.
This article is protected by copyright. All rights reserved.

48. Shapira, C. B. Perillo, I. A. (1993) 13C NMR Analysis of some 4-hydroxy-2H-1,2-
benzothiazine 1,1-dioxides. J Het Chem 30: 1051–1056.
49. Ahmad, M. Siddiqui, H. L. Zia-ur-Rehman, M. Parvez, M. (2010) Anti-oxidant and anti-
bacterial activities of novel N'-arylmethylidene-2-(3, 4-dimethyl-5, 5-dioxidopyrazolo[4,3-
c][1,2]benzothiazin-2(4H)-yl) acetohydrazides. Eur J Med Chem 45: 698-704.
50. Tarcsay, A. Nyiri, K. Keserű, G. M. (2012) Impact of Lipophilic Efficiency on Compound
Quality. J Med Chem 55: 1252−1260.
51. Simard, J. R. Getlik, M. Grütter, C. Pawar, V. Wulfert, S. Rabiller, M. Rauh, D. (2009)
Development of a fluorescent-tagged kinase assay system for the detection and
characterization of allosteric kinase inhibitors. J Am Chem Soc 131: 13286–13296.
52. Wang, Z. Canagarajah, B. J. Boehm, J. C. Kassisà, S. Cobb, M. H. Young, P. R. Abdel-
Meguid, S. Adams, J. L. Goldsmith, E. J. (1998) Structural basis of inhibitor selectivity in
MAP kinases. Structure 6: 1117-1128.
53. Bukhtiyarova, M. Karpusas, M. Northrop, K. Namboodiri, H. V. M. Springman, E. B.
(2007) Mutagenesis of p38alpha MAP kinase establishes key roles of Phe169 in function
and structural dynamics and reveals a novel DFG-OUT state. Biochemistry 46: 5687–
5696
This article is protected by copyright. All rights reserved.