Studies on calpain inhibitors. Synthesis of partially reduced isoquinoline-1-thione derivatives and conversion to functionalized 1-chloroisoquinolines

Article abstract of DOI:10.1016/j.tetlet.2008.02.023

Sequential treatment of a (3-substituted-1-thioxo-1,2,3,4-tetrahydroisoquinolin-4-ylidene)acetic acid with thionyl chloride and a nucleophile did not give the expected ester or amide, but a derivative of 2-(3-substituted-1-chloroisoquinolin-4-yl)acetic ac

Full text of DOI:10.1016/j.tetlet.2008.02.023

Available online at www.sciencedirect.com
Tetrahedron Letters 49 (2008) 2275–2279
Studies on calpain inhibitors. Synthesis of partially
reduced isoquinoline-1-thione derivatives and conversion
to functionalized 1-chloroisoquinolines q
a
a
b
a,
*
´
´
Roberto Chicharro , Mercedes Alonso , Vicente J. Aran , Bernardo Herradon
a Instituto de Quı´mica Orga´nica General, CSIC, Juan de la Cierva 3, 28006 Madrid, Spain
b Instituto de Quı´mica Me´dica, CSIC, Juan de la Cierva 3, 28006 Madrid, Spain
Received 10 November 2007; revised 31 January 2008; accepted 5 February 2008
Available online 9 February 2008
Abstract
Sequential treatment of a (3-substituted-1-thioxo-1,2,3,4-tetrahydroisoquinolin-4-ylidene)acetic acid with thionyl chloride and a
nucleophile did not give the expected ester or amide, but a derivative of 2-(3-substituted-1-chloroisoquinolin-4-yl)acetic acid, constituting
a simple procedure for the synthesis of functionalized 1-chloroisoquinolines, which can be useful synthetic intermediates. The different
reactivity between lactams and thiolactams has been computationally modelled. The activity as calpain inhibitors of both thiolactams
and chloroisoquinoline has been measured, finding that some of these compounds are inhibitors in the micromolar range.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Calpain inhibitor; Chloroisoquinoline; Lactam; Thiolactam
The calpain family of cysteine proteases catalyzes the
partial hydrolysis of a huge variety of proteins,1 constitut-
ing key regulators in a diversity of processes, such as signal
transduction, apoptosis, cell cycle, neuroplasticity and cell
motility.2 Although several isoenzymes are known, the
most relevant members are l-calpain and m-calpain, which
require, respectively, micromolar and millimolar amounts
of Ca2+ to be fully active. The overactivation of calpain
has been linked to several pathological conditions,3
especially degenerative diseases, constituting a current
pharmaceutical target.4
Our previous results have shown that the presence of
aromatic fragments in specific positions are essential for
high activity.
Recently we have found that simple esters and amides
(A) of the sec-butyl-substituted isoquinolinyl-acid 1 are
potent inhibitors of l-calpain;8 constituting hit compounds
for the development of pharmaceuticals.9 With the objec-
tive of obtaining knowledge on structure–activity relation-
ship, we intended to prepare thiolactams of type B. Since
the thiocarbonyl group has different size, polarizability
and capacity to form hydrogen bonds than the carbonyl
group,10 the biological activities of A and B might be quite
different.
Our contribution to this area has included the design
and synthesis of peptidic inhibitors derived from 3,6-dihy-
dro-2H-pyrans,5 biphenyl,6 and nitrogenated heterocycles;7
some of them with IC50 values in the picomolar range.6a
X
O
S
Cl
NH
NH
N
NH
COOH
COY
COY
COY
q
Taken in part from the Ph.D. of R.C. (UAM, Madrid, 2007) and the
A
C
B
projected Ph.D. of M.A.
*
1, X = O
2, X = S
Corresponding author. Tel.: +34 91 5615086; fax: +34 91 5644853.
E-mail address: herradon@iqog.csic.es (B. Herradon).
Y = OR, NHR
´
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.02.023

2276
R. Chicharro et al. / Tetrahedron Letters 49 (2008) 2275–2279
However, our synthetic expectations were not fulfilled;
However, compound 6 was not formed, but 1-chloroiso-
quinoline 7 was obtained as the only non-polar compound.
Although the isolated yield (18% after chromatography)
was modest, it must be taken into account that the trans-
formation of 2 into 7 involves six steps (formation of acyl
chloride, addition of chloride, elimination of a sulfur-con-
taining group, isomerization–aromatization and bis-
acylation).
Other nucleophiles (water, alcohol and amine) were also
used, obtaining higher isolated yields. Thus, acid 8 was pre-
pared in 67% yield by quenching the reaction with water,
the isopropyl ester 9 was synthesized in 60% yield by reac-
tion with 2-propanol, and amide 10 was obtained in 35%
yield by the reaction with the weakly nucleophilic 3-amino-
acetophenone (Scheme 2).
Analogous to the transformations from 2, we prepared
2-benzyl-1-chloroisoquinoline 13 (Scheme 3). The known13
racemic lactam 11 was transformed into thiolactam 12 by
Lawesson reaction (68% yield), which in turn was hydro-
lyzed (70% yield) to acid 13 (70% yield), that was submitted
to the sequential treatment with thionyl chloride and 2-pro-
panol to give 14 in 52% yield.
The different behaviour of lactams and thiolactams in
the reaction with thionyl chloride can be rationalized by
the higher nucleophilicity of sulfur in the thiolactam versus
thus, when we attempted to prepare thiolactams B, through
the acid chloride derived from 2, we obtained 1-chloroiso-
quinoline derivatives C. This interesting reactivity of thio-
amides has not been previously reported and it is quite
different from that of the corresponding amides.8 Addition-
ally, the obtained 1-chloroisoquinolines are not readily
available by other means and the method can be employed
for the synthesis of other chlorinated heterocycles. In this
Letter, we report the transformations of some thiolactams
to 1-chloroisoquinolines as well as a mechanistic insight
from computational modelling, and the biological activity
of thiolactams and 1-chloroisoquinoline as inhibitors of
l-calpain.
The methyl ester 4 was prepared in four steps from L-
isoleucine methyl ester hydrochloride (3), as previously
reported.8 Compound 4 was reacted with Lawesson’s
reagent (LR)11 to give the methyl ester 5, which, in turn,
was hydrolyzed to acid 2 in high overall yield (Scheme 1).12
In order to compare with the preceding biologically
active compounds, our target molecule was diester 6, which
was planned to be prepared by the sequential treatment of
acid 2 with thionyl chloride (to the acid chloride, not iso-
lated) and 0.4 M equiv of 4-(hydroxymethyl)benzyl alco-
hol, as previously reported with the lactam analogues.8
O
S
S
4 steps
43% yield
NH
(a)
NH
NH
(b)
+H3N CO2CH3
Cl-
ref. 8
CO2CH3
CO2CH3
CO2H
3
4
5
2
Scheme 1. Reagents and conditions: (a) LR [2,4-bis(4-methoxyphenyl)-2,4-dithioxo-1,3,2,4-dithiadiphosphetane], toluene, reflux, 1 h (91% yield); (b) 1 M
aq LiOH/THF, rt, overnight (96% yield).
S
S
NH
HN
Cl
N
O
O
O
O
(a, b)
6
CO2H
8; 66%
(a, c)
S
Cl
Cl
(a, b)
NH
N
N
O
O
O
O
Cl
N
CO2H
(a, e)
2
(a, d)
O
7
; 18%
Cl
N
O
O
HN
10; 35%
O
9; 60%
Scheme 2. Reagents and conditions: (a) SOCl2, CH2Cl2, 0 °C to rt, 0.5 h; (b) HOCH2-p-(C6H4)–CH2OH, CH2Cl2, rt, 1 h; (c) NaOH, 0.5 h; (d)
Me2CHOH, CH2Cl2, rt, 1 h; (e) NH2-m-(C6H4)–COMe, Et3N, CH2Cl2, rt, overnight.

R. Chicharro et al. / Tetrahedron Letters 49 (2008) 2275–2279
2277
X
S
Cl
NH
(b)
NH
(c)
N
Ph
CO2CH3
11, X = O
12
Ph
CO2H
Ph
O
13
14
O
(a)
, X = S
Scheme 3. Reagents and conditions: (a) LR, toluene, reflux, 1 h (68% yield); (b) 1 M aq LiOH/THF, rt, overnight (70% yield); (c) (i) SOCl2, CH2Cl2, 0 °C
to rt, 0.5 h; (ii) Me2CHOH, CH2Cl2, rt, 1 h (52% yield).
the oxygen of the lactam. To test this hypothesis, a compu-
tational study was carried out on model compounds 15–18
(Scheme 4), which included both esters and acids of both
oxygen- and sulfur-substituted heterocycles in the two
tautomeric forms. For the sake of shortening calculations,
the conformationally mobile sec-butyl group was changed
by methyl. We studied equilibria between tautomers, calcu-
lation of orbitals and reactivity descriptors. Since the
results of the corresponding methyl ester and acid are
essentially identical, we present only the results on the
acids.14
The starting geometries used in the calculations were
crystallographic structures of related oxo- and thioxo-di-
hydroisoquinolines.15 All the calculations were carried
out using GAUSSIAN 03 set of programs16 at the B3LYP/
6-31+G(d,p) theory level, and the geometries were opti-
mized at high convergence. The relative energies of the
two stable species and the transition states in the prototro-
pies causing equilibria between tautomers are indicated in
Figure 1. Each transition state has a single imaginary fre-
quency corresponding to the vibration of the H–X bond
(X = either O or S). It is observed that the keto-like form
is the most stable for both compounds, but the enol-like
form of the thio-analogue is kinetically and thermodynam-
ically more stable than the oxygenated compound.
Since thionyl chloride is an electrophile, the substituent
in position 1- of the starting isoquinoline has to act as
nucleophile. Therefore, some molecular characteristics
which gauge these electron-donor characteristics were com-
puted. The shapes of the HOMOs of 15a, 15b, 17a and 17b
(Fig. 2) show that the highest coefficient is placed on the
H
X
X
NH
N
CO2R
CO2R
15a
16a
15b
16b
, X = O, R = H
, X = O, R = H
, X = O, R = CH3
, X = O, R = CH3
17a, X = S, R = H
17b, X = S, R = H
18a
, X = S, R = CH3
18b
, X = S, R = CH3
Scheme 4. Equilibria between tautomers.
Fig. 1. B3LYP/6-31+G(d,p) optimized geometries and relative energies (kcal molÀ1) of the two tautomers of 15 and 17 and the corresponding transition
states (TS).

2278
R. Chicharro et al. / Tetrahedron Letters 49 (2008) 2275–2279
Table 2
Results of the inhibition of l-calpain
Compound
IC50 (lM)
2
5
100
38
7
97
8
23
9
100
42
Inactive
10
13
shown in Table 2 indicate that all the tested compounds are
either inactive or poor inhibitors of the enzyme.19
In conclusion, starting from 3-alkyl-4-(methoxy-
carbonyl)methylene-3,4-dihydro-2H-isoquinoline-1-thione,
we have prepared the derivatives of 4-(3-alkyl-1-chloroiso-
quinolinyl)acetic acid; which can be useful synthetic inter-
mediates via nucleophilic substitution, metallation or
transition-metal catalyzed transformations on the haloge-
nated position;20 as well as through the chemoselective
functionalization of the substituent at position 4. Addition-
ally, since the derivatives of 3-sec-butyl-1-chloroisoquino-
line bear a stereogenic center and they can be prepared in
both enantiomeric forms,21 these compounds can be useful
as catalysts or additives for asymmetric synthesis.
Fig. 2. HOMOs of the two tautomers of acids 15 and 17. The orbitals are
plotted on an isoelectronic density surface of 0.04 e bohrÀ3
.
sulfur atom of 17a. Additional clues are obtained from the
reactivity descriptors of acids 15a, 15b, 17a and 17b
(Table 1).17 It is observed that thiolactam 17a has the high-
est energy HOMO and the smallest HOMO–LUMO gap as
well as the highest polarizability a and the lowest ioniza-
tion potential I of the four compounds. All these results
explain the different reactivity of thionyl chloride with
either the thiolactam or the lactam as well as that the reac-
tion likely goes through the thiocarbonyl functionality. A
plausible reaction course is indicated in Scheme 5, where
the well-known transformation of COOH to COCl is not
detailed.
Acknowledgement
This work was financially supported by Spanish Minis-
try of Education and Science (Grants CTQ2004-01978
and CTQ2007-64891, and a fellowship to M.A.).
Some of the synthesized thiolactams (2, 5 and 13) and
chloroisoquinolines (7–10) were tested as inhibitors of l-
calpain using labelled casein as substrate,18 following the
experimental procedure previously reported.7a The biolog-
ical activity has been gauged by the IC50 values. The results
Supplementary data
Experimental details of synthetic procedures, spectro-
scopic and analytical data for new compounds and full
Table 1
Dipole moment (l in Debyes), HOMO energy (eHOMO in eV), LUMO energy (eLUMO in eV), HOMO–LUMO energy difference (DEHOMO–LUMO),
ionization potential (I in eV), electron affinity (A in eV), absolute hardness (g in eV), electrophilic global index (x in eV) and polarizability (a in au) of
compounds 15a, 15b, 17a and 17b
Compound
l
eHOMO
eLUMO
DEHOMO–LUMO
I
A
g
x
a
15a
15b
17a
17b
3.73
2.91
4.47
3.23
À6.99
À6.69
À5.95
À6.65
À2.58
À2.40
À2.76
À2.49
4.41
4.29
3.19
4.17
8.59
8.28
7.93
8.20
0.99
0.83
1.26
0.96
3.80
3.73
3.33
3.62
3.02
2.79
3.17
2.90
165
167
192
185
O
S
O
S
O
S
Cl-
Cl
S
S
Cl
Cl
N
S
N
Cl
Cl
Cl
H
H
R
Z
-[ClS(O)S-]
N
NH
R
Z
R
Z
R
Z
isomerization/
aromatization
O
O
O
O
Z = OH, Cl
Scheme 5. Reaction course for the transformation of thiolactam into 1-chloroisoquinoline.

R. Chicharro et al. / Tetrahedron Letters 49 (2008) 2275–2279
2279
´
´
9. Herradon, B.; Chicharro, R.; Aran, V. J.; Alonso, M. PCT Appli-
computational results. Supplementary data associated with
this article can be found, in the online version, at doi:
10.1016/j.tetlet.2008.02.023.
cation ES2005/070171.
10. Lee, H. J.; Choi, Y. S.; Lee, K. B.; Park, J.; Yoon, C. J. J. Phys.
Chem. A 2002, 106, 7010–7017.
11. Pedersen, B. S.; Scheibye, S.; Nilsson, N. H.; Lawesson, S. O. Bull.
Soc. Chim. Belg. 1978, 87, 223–228.
References and notes
12. The new compounds gave satisfactory analytical (C, H, N, 0.3%)
and spectroscopic (1H and 13C NMR, IR, MS) data. Detailed
experimental procedures and spectroscopic data are included as
Supplementary data.
13. Sanchez-Sancho, F.; Mann, E.; Herradon, B. Adv. Synth. Catal. 2001,
343, 360–368.
1. (a) Goll, D. E.; Thompson, V. F.; Li, H.; Wei, W.; Cong, J. Physiol.
Rev. 2003, 83, 731–801; (b) Tompa, P.; Buzder-Lantos, P.; Tantos, A.;
Farkas, A.; Szilagyi, A.; Banoczi, Z.; Hudecz, F.; Friedrich, P. J. Biol.
Chem. 2004, 279, 20775–20785; (c) MEROPS the peptidase database
(release 7.70); http://merops.sanger.ac.uk/ (last accessed: 28th Janu-
ary, 2008).
2. (a) Hood, J.; Brooks, W.; Roszman, T. Bioessays 2006, 28, 850–859;
(b) Croall, D. E.; Ersfeld, K. Genome Biol. 2007, 8, article 218; (c)
Hara, M. R.; Snyder, S. H. Ann. Rev. Pharmacol. Toxicol. 2007, 47,
117–141.
3. (a) Harris, F.; Chatfield, L.; Singh, J.; Phoenix, D. A. Mol. Cell.
Biochem. 2004, 261, 161–167; (b) Zatz, M.; Starling, A. New Engl. J.
Med. 2005, 352, 2413–2423.
4. (a) Neffe, A. T.; Abell, A. D. Curr. Opin. Drug Discovery Dev. 2005, 8,
684–700; (b) Carragher, N. Curr. Pharm. Des. 2006, 12, 615–638.
5. Montero, A.; Mann, E.; Herradon, B. Tetrahedron Lett. 2005, 46,
401–405.
6. (a) Montero, A.; Mann, E.; Chana, A.; Herradon, B. Chem. Biodiv.
2004, 1, 442–457; (b) Montero, A.; Albericio, F.; Royo, M.;
Herradon, B. Org. Lett. 2004, 6, 4089–4092; (c) Montero, A.; Alonso,
M.; Benito, E.; Chana, A.; Mann, E.; Navas, J. M.; Herradon, B.
Bioorg. Med. Chem. Lett. 2004, 14, 2753–2757; (d) Montero, A.;
Albericio, F.; Royo, M.; Herradon, B. Eur. J. Org. Chem. 2007, 1301–
1308.
7. (a) Mann, E.; Chana, A.; Sanchez-Sancho, F.; Puerta, C.; Garcia-
Merino, A.; Herradon, B. Adv. Synth. Catal. 2002, 344, 855–867; (b)
Salgado, A.; Mann, E.; Sanchez-Sancho, F.; Herradon, B. Hetero-
cycles 2003, 60, 57–71.
14. Complete computational data, including Cartesian coordinates of
minima and transition states, are collected as Supplementary data.
´
15. Chicharro, R. Ph.D. Thesis, Universidad Autonoma de Madrid,
2007.
16. Frisch, M. J., et al. GAUSSIAN 03 Revision B 04, Gaussian: Wallingford,
CT, 2004. The full reference is included as Supplementary data.
17. Parr, R. G.; Yang, W. Density Functional Theory of Atoms and
Molecules; Oxford University Press: New York, 1989; pp 70–101.
18. (a) Jiang, S.-T.; Wang, J.-W.; Chang, T.; Chen, C.-S. Anal. Biochem.
1997, 244, 233–238; (b) Thomson, V. F.; Saldana, S.; Cong, J.; Goll,
˜
D. E. Anal. Biochem. 2000, 279, 170–178.
19. The unique feature of calpain, that is activation by Ca2+, allows to
design inhibitors based on two distinct concepts: either electrophilic
inactivation of the cysteine at the active site or hampering activation
by Ca2+. As previously discussed (Refs. 6 and 8), it is likely that the
Michael acceptors 2 and 5 act by the former mechanism and the non-
electrophilic isoquinolines 7–10 by the latter one.
20. Nair, M. D.; Premila, M. S. In Isoquinolines; Kathawala, F. G.,
Coppola, G. M., Schuster, H. F., Eds.; The Chemistry of Heterocyclic
Compounds; Taylor, E. C., Ed.; John Wiley & Sons: New York, 1990;
Vol. 38, Part 2, pp 1–133.
21. From D-isoleucine or from L-allo-isoleucine. The enantiomeric purity
of 7–10 was assessed by HPLC (Nucleosil C18 as stationary phase and
100/0 to 10/90 CH3CN/H2O as mobile phase) with a UV/CD
detector.
8. Chicharro, R.; Alonso, M.; Mazo, M. T.; Aran, V. J.; Herradon, B.
ChemMedChem 2006, 1, 710–714.