Tetrahydrobenzothiophene carboxamides: Beyond the kinase domain and into the fatty acid realm

Article abstract of DOI:10.1016/j.bmcl.2017.08.006

A series of tetrahydrobenzothiophene carboxamides, inspired by structural features present in kinase and SCD1 inhibitors, are presented here. Prototype compound 8 (MMDD13) modulates fatty acid elongase and desaturase indexes, lipid accumulation, while pre

Full text of DOI:10.1016/j.bmcl.2017.08.006

Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Tetrahydrobenzothiophene carboxamides: Beyond the kinase domain
and into the fatty acid realm
b
c
d
e
Sabin Llona-Minguez a, , Shabnam Fayezi , Alireza Alihemmati , Jordi Juárez-Jiménez , F. Javier Piedrafita ,
⇑
Thomas Helleday a
a Division of Translational Medicine and Chemical Biology, Science for Life Laboratory, Department of Medical Biochemistry and Biophysics, Karolinska Institutet, Stockholm 171
65, Sweden
b Department of Biology and Anatomical Sciences, Faculty of Medicine, Shahid Beheshti University of Medical Sciences, Tehran 1983969411, Iran
c Department of Anatomical Sciences, Faculty of Medicine, Tabriz University of Medical Sciences, Tabriz 5165665931, Iran
d EaStCHEM School of Chemistry, Joseph Black Building, The King’s Buildings, Edinburgh EH9 3JJ, United Kingdom
e Torrey Pines Institute for Molecular Studies, 3550 General Atomics Court, San Diego, CA 92121, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of tetrahydrobenzothiophene carboxamides, inspired by structural features present in kinase and
SCD1 inhibitors, are presented here. Prototype compound 8 (MMDD13) modulates fatty acid elongase
and desaturase indexes, lipid accumulation, while preserving kinase inhibitory activity. This chemotype
represents a stepping stone towards chemical probes to study the consequences of lipid metabolism
modulation through non-redundant pathways.
Received 13 June 2017
Revised 1 August 2017
Accepted 3 August 2017
Available online xxxx
Ó 2017 Published by Elsevier Ltd.
Text
pathogenesis of myeloproliferative neoplasms and other immune
conditions.14,15 Furthermore, JAK-STAT signalling has been linked
Long chain fatty acyl-coenzyme A esters (LCFA-CoAs) are
important organic building blocks used for energy production in
the human adipose tissue,1 and for the biosynthesis of phospho-
lipids, triglycerides, wax and cholesterol esters.2 Three mammalian
to adipogenesis,16 and loss of JAK2 function can arrest steatohep-
atitis development,17 impair lipolysis and improve fatty liver in
mice.18 Recent evidence has suggested that hepatic JAK2 deficiency
leads to elevated SCD1 expression and increased conversion of sat-
urated LCFA-CoAs to mono-unsaturated LCFA-CoA,19 linking JAK2
and SCD1 to LCFA-mediated pathology. Compounds that simulta-
neously inhibit SCD1 and JAK2 activity will be useful tools to probe
the consequences of decreased lipid metabolism through modula-
tion of non-redundant pathways. Furthermore, this dual inhibition
approach could exert the same pharmacological effect as a combi-
nation therapy (e.g. larger therapeutic window, long-lasting
response), while minimizing burdens associated with combination
therapies, such as complex clinical investigations, potential non-
mechanism-based toxicities and drug-drug interactions.
desaturases,
D5D, D6D, and D9D, also known as fatty acid desat-
urase 1 (FADS1) and 2 (FADS2) and stearoyl-CoA desaturase-1
(SCD1), respectively, are key enzymes in the conversion of satu-
rated LCFA-CoAs to mono-unsaturated LCFA-CoAs.3,4 Dysregula-
tion of cellular lipid metabolism has been implicated in diverse
pathologies,5 and re-balancing the LCFA-CoA pool via pharmaco-
logical modulation of desaturase or elongase activities, is an
approach of therapeutic interest.6–8 SCD1 is the desaturase enzyme
that has attracted more attention from the research community,
and pharmacological inhibition of SCD1 activity has been investi-
gated against diabetes, obesity, non-alcoholic steatohepatitis, hep-
atitis C, cancer, and acne.7,9 Unfortunately, partial eye closure and
progressive alopecia have precluded systemic SCD1 inhibition,10,11
Multiple SCD1 chemotypes have been identified since 2005.7,9
Xenon and Abbot’s SCD1 compounds (1 and 2, Fig. 1) are amongst
the first SCD1 inhibitors reported,20,21 and have served as a struc-
tural basis for the discovery of many other SCD1 inhibitors.
The new SCD1/JAK2 inhibitor chemotype, inspired by
GlaxoSmithKline and Takeda’s SCD1 inhibitors (3 and 4,
Fig. 1),22,23 and Merck’s JAK2 inhibitors (5a, Fig. 2),24 shared phar-
macophoric features with those three chemical series: a central
heterocyclic amide core (pink colour), flanked by two lipophilic
regions (green and orange colours), one of which, the benzene ring,
is linked to a hydrogen bond acceptor or donor (blue colour)
but organ-targeted SCD1 inhibition remains
a therapeutic
option.12,13
Janus kinase 2 (JAK2) activity and the JAK-STAT (Signal Trans-
ducer and Activator of Transcription) pathway are relevant in the
⇑
Corresponding author.
E-mail address: sabin.llona.minguez@scilifelab.se (S. Llona-Minguez).
http://dx.doi.org/10.1016/j.bmcl.2017.08.006
0960-894X/Ó 2017 Published by Elsevier Ltd.
Please cite this article in press as: Llona-Minguez S., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.08.006

2
S. Llona-Minguez et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
1, Xenon [21]
2, Abbott [20]
5a, JAK2 [24]
5c, IKK2 [34]
5f, PknG [37]
5b, pan-kinase [33]
5e, FLT3 [36]
5d, BCR-ABL [35]
3, GlaxoSmithKline [22]
4, Takeda [23]
Fig. 2. Chemical structures of thiophenecarboxamide kinase inhibitors, and their
putative pharmacological targets.
The required 4-(aminomethyl)benzoic acid was converted to the
correspondent acid chloride I-2, and subsequently coupled with
amino-thiophene intermediate I-1, to yield compounds 6, 7 and
11 (Table 1).
New chemotype presented in this work
Ester compounds could be further reduced with lithium boro-
hydride to the correspondent hydroxy-methyl derivatives 8, 9,
12, 13, 14, and 15 (Table 1), or hydrolysed, under basic conditions,
to the corresponding carboxylate 10 (Table 1). The basic pH in the
borohydride reductions of compounds 8, 12, 13, 14, triggered a
cyclodehydration reaction, forming the thienopyrimidinone by-
products 16, 17, 18 and 19 (Table 2), which were also isolated from
the reaction mixtures.27
Fig. 1. Chemical structures of SCD1 inhibitors.
(Fig. 1). The resulting thiophenecarboxamide core, a frequent scaf-
fold in the kinase inhibitor field (5b–f, Fig. 2), represents a new
avenue to explore kinase and non-kinase poly-pharmacology.25,26
The synthesis of compounds 6–19 followed the synthetic route
shown in Scheme 1. The tetrahydrobenzothiophene core (I-1) was
In order to assess compound effect on the cellular LCFA-CoA
pool saturation, HepG2 human hepatic cells, known for their high
furnished using
a Gewald condensation between 2-cyanoac-
etamide, sulphur, and the appropriate cyclohexanone derivative.24
Scheme 1. General synthetic route for tetrahydrobenzothiophene derivatives: (a) sulphur flowers (1 eq), morpholine (1 eq), EtOH, 85 °C, 18 h. (b) oxalyl chloride (2 eq), DMF
(cat.), DCM, rt, 2 h. (c) Et3N (1 eq), DCM, rt, 18 h. (d) LiBH4 (3 eq), THF/diethylether (2:1), 40 °C, 24 h. (e) LiOHÁH2O, 1,4-dioxane/H2O, rt, 18 h.
Please cite this article in press as: Llona-Minguez S., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.08.006

S. Llona-Minguez et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
3
Table 1
substrates offers a realistic assessment of inhibitor potency in the
cellular context. To further characterise compound effect on lipid
metabolism, HepG2 cells were exposed to stearic acid, in the pres-
ence or absence of inhibitors, and analysed using the Oil red O
assay.31
Structures of tetrahydrobenzothiophene derivatives, Part I.
Reference compounds 1 (hSCD1 IC50 = 26 nM) and 2 (hSCD1
IC50 = 41 nM),20,21 revealed
a
LCFA pool profile with
increased18:0/16:0 cellular elongase index, and decreased
16:1/16:0 and 18:1/18:0 desaturase indexes, at single concentra-
tion and in a concentration-dependent manner (Fig. 3), in agree-
ment with their potent SCD1 biochemical inhibition. In order to
assess SCD1-mediated compound activity, we set the goal to
identify compounds with a similar desaturation profile to 1 and
2. Analogues containing an un-substituted cyclohexane ring (6), a
short ester chain at the cyclohexane 6-position (7), or a hydrox-
yethyl chain at the cyclohexane 7-position (9) caused an opposite
effect to 1 and 2, decreasing the elongase index, as so did an ana-
logue missing the 4-(aminomethyl)benzoic acid moiety (15), and
the thienopyrimidinone analogues 17, 18 and 19. The reason for
this significant decrease in elongase activity is unclear. A car-
boxylic acid (10) or a phenol (11) substituent at the 6-position of
the cyclohexane ring did not affect the elongase activity, but a
hydroxymethyl group (8) elevated the elongase index, as seen with
1 and 2. Piperidine (12), dimethylamine (13) and thiomorpholine
1,1-dioxide (14) derivatives of 8 also showed a similar elongase
profile, and interestingly, so did the morpholine thienopyrimidi-
none analogue 16. The hydroxymethyl derivatives 8, 12, 13, 14
and 16 also showed a similar 16:1/16:0 and 18:1/18:0 desaturase
index profile to 1 and 2, at single concentration and in a concentra-
tion-dependent manner.
Cpd. No.
R
6
morpholine
7
morpholine
8
9
morpholine
morpholine
10
11
morpholine
morpholine
12
13
14
piperidine
Compound 8 was deemed as the most potent analogue, causing
a dramatic effect on the LCFA-CoA pool, even at 1 mM compound
concentration. In agreement with their potent inhibitory effect
on SCD1 activity, compounds 2 and 8 profoundly decreased cellular
lipid accumulation (Fig. 4), whereas the inactive analogue 18 did
not.
dimethylamine
thiomorpholine 1,1-dioxide
Compound 8 was subsequently tested at 10 mM in a small
kinase panel, including JAK2 and other serine/threonine and tyro-
sine kinases (Fig. 5).32 Despite retaining moderate JAK2 inhibitory
activity (45% inhibition), the structural modifications in 8 led to a
drop in potency when compared with the parent compound 5a.
As expected from the scaffold’s inherent kinase affinity, significant
inhibitory potency also remained against other kinase targets, such
as FLT-3, FMS and ABL wt kinases (>80% inhibition).
In conclusion, we have explored the pharmacological potential
of the tetrahydrobenzothiophene scaffold to simultaneously mod-
ulate SCD1 and JAK2 enzymatic activity. Lead compound 8 affected
cellular elongase and desaturase indexes, and cellular lipid accu-
mulation, analogously to reference SCD1 inhibitors, suggesting a
similar mode of action. The encouraging preliminary results pre-
sented, highlight the pharmacological potential of this chemical
desaturase and elongase activities,28 were exposed to the inhibi-
tors for 72 h. After extraction of the total cellular lipid content,
the LCFA pool was analysed by gas chromatography.29,30 Since
SCD1 desaturates stearoyl-CoA (18:0) to oleoyl-CoA (18:1) prefer-
entially over palmitoyl-CoA (16:0) to palmitoleoyl-CoA (16:1),5 the
18:1/18:0 and 16:1/16:0 ratios were monitored, together with the
18:0/16:0 ratio as the elongase activity readout. If the desaturation
capacity of the cell was inhibited by the compounds, the LCFA pool
would be more saturated. Unlike information on desaturase inhibi-
tion obtained using labelled substrates, estimation of the SCD
desaturase activity based on the ratio of endogenous products vs.
Table 2
Structures of tetrahydrobenzothiophene derivatives, Part II.
16, R = morpholine
17, R = piperidine
15
18, R = dimethylamine
19, R = thiomorpholine 1,1-dioxide
Please cite this article in press as: Llona-Minguez S., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.08.006

4
S. Llona-Minguez et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
a
b
Fig. 3. Effect of reference SCD1 inhibitors and tetrahydrobenzothiophene derivatives on the desaturatase and elongase activity indexes, 18:0/16:0 elongase index, 16:1/16:0
desaturase index, 18:1/18:0 desaturase index. (a) Compounds tested at 10 mM. (b) Compounds tested at different concentrations.a aAverage values of two or three replicates
per data point with standard deviation.
scaffold, and warrant further structural refinement towards JAK2
potency and selectivity.
Acknowledgments
We acknowledge Dr. Gaelle Cane for insightful manuscript dis-
cussions. This work was supported by the Swedish Research Coun-
cil, the Knut and Alice Wallenberg Foundation, the Göran
Gustafsson Foundation, the Swedish Pain Relief Foundation, the
Torsten Söderbergs Stiftelse, the Swedish Children’s Cancer Foun-
dation, and the Swedish Cancer Society. We would also like to
acknowledge the Faculty of Medicine at Tabriz University of Med-
ical Sciences for providing research facilities and ChemAxon
Fig. 4. Effect of compounds 2, 8 and 18 on cellular lipid accumulation.a aAverage
(http://www.chemaxon.com) for technical support.
values of three replicates per data point with standard deviation.
A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2017.08.
006.
References
1. Insull W, Bartsch GE. Am J Clin Nutr. 1967;20:13–23.
2. Miyazaki M, Gomez FE, Ntambi JM. J Lipid Res. 2002;43:2146–2154.
3. Dobrzyn A, Ntambi JM. Obes Rev. 2005;6:169–174.
4. Ntambi JM, Miyazaki M. Curr Opin Lipidol. 2003;14:255–261.
5. Ntambi JM, Miyazaki M. Prog Lipid Res. 2004;43:91–104.
6. Matsuzaka T, Shimano H, Yahagi N, et al. Nat Med. 2007;13:1193–1202.
7. Zhang Z, Dales NA, Winther MD. J Med Chem. 2014;57:5039–5056.
Fig. 5. Kinase activity profile for compound 8.a aAverage values of two or three
8. Lopez-Vicario C, Gonzalez-Periz A, Rius B, et al. Gut. 2014;63:344–355.
replicates per data point with standard deviation.
9. Uto Y. Chem Phys Lipids. 2016;197:3–12.
Please cite this article in press as: Llona-Minguez S., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.08.006

S. Llona-Minguez et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
5
10. Li CS, Belair L, Guay J, et al. Bioorg Med Chem Lett. 2009;19:5214–5217.
11. Leger S, Black WC, Deschenes D, et al. Bioorg Med Chem Lett. 2010;20:499–502.
12. Oballa RM, Belair L, Black WC, et al. J Med Chem. 2011;54:5082–5096.
23. Taniguchi T, Miyata K, Kubo O, Matsui J. WO2008044767A1; 2008.
24. Haidle AM, Zabierek AA, Childers KK, et al. Bioorg Med Chem Lett.
2014;24:1968–1973.
13. Meingassner JG, Aschauer H, Winiski AP, et al.
2013;133:2091–2094.
14. Quintas-Cardama A, Kantarjian H, Cortes J, Verstovsek S. Nat Rev Drug Discov.
2011;10:127–140.
J
Invest Dermatol.
25. Reddy AS, Zhang SX. Expert Rev Clin Pharmacol. 2013;6:41–47.
26. Liu X, Zhu F, Ma XH, et al. Curr Med Chem. 2013;20:1646–1661.
27. Konno S, Tsunoda M, Watanabe R, et al. Yakugaku Zasshi. 1989;109:464–473.
28. Dixon JL, Ginsberg HN. J Lipid Res. 1993;34:167–179.
15. Nefedova Y, Nagaraj S, Rosenbauer A, et al. Cancer Res. 2005;65:9525–9535.
16. Richard AJ, Stephens JM. Biochim Biophys Acta. 1842;2014:431–439.
17. Shi SY, Martin RG, Duncan RE, et al. J Biol Chem. 2012;287:10277–10288.
18. Nordstrom SM, Tran JL, Sos BC, et al. Mol Endocrinol. 2013;27:1333–1342.
19. Themanns M, Mueller KM, Kessler SM, et al. Sci Rep. 2016;6:34719.
20. Liu G, Lynch JK, Freeman J, et al. J Med Chem. 2007;50:3086–3100.
21. Zhang ZH, Sun SY, Kodumuru V, et al. J Med Chem. 2013;56:568–583.
22. Bouillot AMJ, Boyer T, Daugan AC-M, et al. WO2008104524A1; 2008.
29. Yousefi B, Darabi M, Baradaran B, et al. Bioimpacts. 2012;2:145–150.
30. Llona-Minguez S, Hoglund A, Ghassemian A, et al.
2017;60:4279–4292.
J
Med Chem.
31. Ramirez-Zacarias JL, Castro-Munozledo F, Kuri-Harcuch W. Histochemistry.
1992;97:493–497.
32. Llona-Minguez S, Hoglund A, Jacques SA, et al.
2016;59:1140–1148.
J
Med Chem.
Please cite this article in press as: Llona-Minguez S., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.08.006