X-ray diffraction, solution structure, and computational studies on derivatives of (3-sec-butyl-2,3-dihydro-1H-isoquinolin-4-ylidene)acetic acid: Compounds with activity as calpain inhibitors

Article abstract of DOI:10.1021/jo902091u

(Chemical Equation Presented) A thorough experimental and computational study of derivatives of (3-sec-butyl-2,3-dihydroisoquinolin-4-ylidene)acetic acid was performed. Some of these compounds are calpain inhibitors and could be useful as therapeutic agen

Full text of DOI:10.1021/jo902091u

pubs.acs.org/joc
X-ray Diffraction, Solution Structure, and Computational Studies on
Derivatives of (3-sec-Butyl-2,3-dihydro-1H-isoquinolin-4-ylidene)acetic
Acid: Compounds with Activity as Calpain Inhibitors
†
Mercedes Alonso, Roberto Chicharro, Carlos Miranda, Vicente J. Aran,
†
†
‡
ꢀ
§
Miguel A. Maestro, and Bernardo Herradon*
,†
ꢀ
†
‡
Instituto de Quımica Organica General, CSIC, Juan de la Cierva 3, 28006 Madrid, Spain, Instituto de
´
ꢀ
§
´
ꢀ
´
Quımica Medica, CSIC, Juan de la Cierva 3, 28006 Madrid, Spain, and Departamento de Quꢀıımica
~ ~
Fundamental, Facultad de Ciencias, Universidade da Coruna, Campus da Zapateira, 15071 A Coruna, Spain
herradon@iqog.csic.es
Received September 28, 2009
A thorough experimental and computational study of derivatives of (3-sec-butyl-2,3-dihydroisoqui-
nolin-4-ylidene)acetic acid was performed. Some of these compounds are calpain inhibitors and could
be useful as therapeutic agents, since this enzyme is a Ca2þ-dependent cysteine protease involved in a
wide variety of metabolic and physiological processes, whose over-activation is associated to several
pathological conditions. To gain a better understanding of the structure-activity relationships,
1
a structural analysis was carried out with H and 13C NMR spectroscopy and DFT calculations
together with the X-ray diffraction data of three compounds. The solid state structures showed that
the crystal packing as well as the intermolecular interactions depend on the substituent nature of the
COOR group. Also, the reactivity of the exocyclic double bond was theoretically evaluated, finding
that the more reactive compound is the most potent inhibitor of calpain (IC50 = 25 nM).
Introduction
unequivocally characterized, it has been found that a wide
variety of proteins suffer limited hydrolysis by the catalytic
action of calpain.3 Some of the substrates hydrolyzed by
calpain are cytoskeletal proteins, enzymes involved in signal
transduction, membrane receptors, and transcription fac-
tors; what this means is that calpain can be a key regulatory
factor in diverse physiological and metabolic pathways,
including apoptosis, cell cycle, neuron plasticity, cell moti-
lity, and so on.4 As a consequence, the over-activation of
The design of pharmaceuticals is firmly bound on the
availability of a large number of accurate data on biological
activity and structure, what allows the structure-activity
relationships to be established.1 In recent years, we have been
engaged in the preparation of calpain inhibitors, which is a
protease of the CA clan of the cysteine peptidase class.2
Although the natural substrate of this enzyme has not been
(1) (a) The Practice of Medicinal Chemistry, 2nd ed.; Wermuth, C. G., Ed.;
Elsevier/Academic Press: San Diego, CA, 2003. (b) Nantasenamat, C.; Isaran-
kura-Na-Ayudhya, C.; Naenna, T.; Prachayasittikul, V. EXCLI J. 2009, 8, 74–88.
(c) Salum, L.; Andricopulo, A. Mol. Diversity 2009, 13, 277–285.
(3) (a) Friedrich, P. Biochem. Biophys. Res. Commun. 2004, 323, 1131–
ꢀ
Banoczi, Z.; Hudecz, F.; Friedrich, P. J. Biol. Chem. 2004, 279, 20775–20785.
1133. (b) Tompa, P.; Buzder-Lantos, P.; Tantos, A.; Farkas, A.; Szilagyi, A.;
ꢀ ꢀ
ꢀ
(c) Friedrich, P.; Bozoky, Z. Biol. Chem. 2005, 386, 609–612.
(4) (a) Goll, D. E.; Thompson, V. F.; Li, H.; Wei, W.; Cong, J. Physiol.
(2) (a) McKerrow, J. H.; Caffrey, C.; Kelly, B.; Loke, P.; Sajid, M. Annu.
ꢀ
€
ꢀ
ꢀ
ꢀ
Rev. Pathol.: Mech. Dis. 2006, 1, 497–536. (b) Mittl, P. R.; Grutter, M. G.
Rev. 2003, 83, 731–801. (b) Janossy, J.; Ubezio, P.; Apati, A.; Magocsi, M.;
Tompa, P.; Friedrich, P. Biochem. Pharmacol. 2004, 67, 1513–1521.
(c) Franco, S. J.; Huttenlocher, A. J. Cell Sci. 2005, 118, 3829–3838.
(d) Wu, H.-Y.; Lynch, D. Mol. Neurobiol. 2006, 33, 215–236. (e) Croall, D.;
Ersfeld, K. Genome Biol. 2007, 8, 218.
Curr. Opin. Struct. Biol. 2006, 16, 769–775. (c) Turk, B. Nat. Rev. Drug
Discovery 2006, 5, 785–799. (d) MEROPS: the peptidase database. Rawlings,
N. D.; Morton, F. R.; Kok, C. Y.; Kong, J.; Barrett, A. J. Nucleic Acids Res. 2008,
36, D320-325. http://merops.sanger.ac.uk/ (last accessed Sept 24th, 2009).
342 J. Org. Chem. 2010, 75, 342–352
Published on Web 12/16/2009
DOI: 10.1021/jo902091u
r
2009 American Chemical Society

Alonso et al.
JOCArticle
CHART 1. Generic Structures of Calpain Inhibitors Based on
the Isoquinoline (A) and Pyrido[1,2-b]isoquinoline (B) Backbones
and Their Related Peptide Hybrids (C and D, respectively)
SCHEME 1a
CHART 2. Structures and Biological Activities of the
Isoquinolines 1-4 Studied in This Work
aReagents and conditions: (a) o-I-C6H4COCl, 1 M K2CO3, THF, 0 ꢀC
to rt (62% yield); (b) LiBH4, MeOH, THF, -10 ꢀC to rt (85% yield);
(c) (i) DMSO/(COCl)2, CH2Cl2, Et3N, -78 ꢀC, (ii) Ph3PdCHCO2CH3,
CH2Cl2, -78 ꢀC to rt (80% yield); (d) Pd(OAc)2, Ph3P, Et3N, CH3CN,
reflux (77% yield); (e) 1 M LiOH, THF/H2O (96% yield); (f) Lawesson’s
reagent, toluene, reflux (91% yield); (g) (i) SOCl2, CH2Cl2, reflux,
(ii) BuOH, CH2Cl2, rt (69% yield).
additional advantages of this compound are the following: (1) it
has a low molecular weight; (2) it is readily synthesized in
multigram scale in a few synthetic operations; and (3) it has a
variety of biological activity in cell assays.9
To expand our knowledge on the biological activity of
methyl (S,S,Z)-(3-sec-butyl-1-oxo-2,3-dihydro-1H-isoqui-
nolin-4-ylidene)acetate (1) we have synthesized closely re-
lated compounds to 1, finding that meanwhile the butyl ester
2 is a weak inhibitor of μ-calpain, and the acid 3 and the
thiolactam 4 are noninhibitors (Chart 2).
To deepen on the reasons of the quite different biological
activities of these related compounds, we have performed a
thorough structural study of 1-4, using experimental (NMR
and X-ray diffraction analysis) and computational meth-
odologies.
calpain causes an impairment in a diversity of essential
biological processes, and, therefore, it is involved in a variety
of diseases, including Alzheimer, Parkinson, muscular dys-
trophy, multiple sclerosis, arthritis, cataract, and other
degenerative and aging-related diseases.5
Previously, we reported that some partially reduced deriva-
tives of either isoquinoline (A) or pyrido[1,2-b]isoquinoline (B)
having an electrophilic olefin were inhibitors of μ-calpain
(Chart 1).6 However, our initial results indicated that the pre-
sence of peptidic fragments (e.g., C or D) was necessary in
order to have biological activity.7 On continuing our structure-
activity studies on isoquinoline derivatives, we were delighted to
find that the simple isoquinoline derivative 1 (Chart 2) was a
strong inhibitor of μ-calpain with an IC50 of 25 nM.8 Some
Results and Discussion
In the following discussion, the numbering of the atoms of
molecules 1-4 has been homogenized and it is based on the
X-ray structures as indicated in Chart 2. The numbering of
hydrogen atoms is the same as those atoms to which hydro-
gens are attached.
(5) (a) Gafni, J.; Ellerby, L. M. J. Neurosci. 2002, 22, 4842–4849. (b) Zatz,
M.; Starling, A. N. Engl. J. Med. 2005, 352, 2413–2423. (c) Selvakumar, P.;
Sharma, R. K. Int. J. Mol. Med. 2007, 19, 823–827. (d) Azuma, M.; Shearer,
T. R. Surv. Ophthalmol. 2008, 53, 150–163.
(6) For overviews of calpain inhibitors, see: (a) Neffe, A.; Abell, A. Curr.
Opin. Drug Discovery Dev. 2005, 8, 684–700. (b) Carragher, N. O. Curr.
Pharm. Des. 2006, 12, 615–638.
Synthesis. The synthesis of 1-4 (Scheme 1) has been
reported in previous works.8,10 Compound 1 was obtained
as a single stereoisomer and regioisomer by acylation
of (S,S)-isoleucine methyl ester hydrochloride with 2-iodo-
benzoyl chloride, reduction of the methoxycarbonyl group
ꢀ
(7) (a) Mann, E.; Chana, A.; Sanchez-Sancho, F.; Puerta, C.; Garcı
´
a-
Merino, A.; Herradon, B. Adv. Synth. Catal. 2002, 344, 855–867. (b) Salgado,
ꢀ
A.; Mann, E.; Sanchez-Sancho, F.; Herradon, B. Heterocycles 2003, 60,
57–71.
ꢀ
(8) Chicharro, R.; Alonso, M.; Mazo, M. T.; Aran, V. J.; Herradon, B.
ChemMedChem 2006, 1, 710–714.
ꢀ
(9) Herradon, B.; Chicharro, R.; Aran, V. J.; Alonso, M. PCT Appl.
ES2005/070171.
(10) Chicharro, R.; Alonso, M.; Aran, V. J.; Herradon, B. Tetrahedron
Lett. 2008, 49, 2275–2279.
ꢀ
J. Org. Chem. Vol. 75, No. 2, 2010 343

Alonso et al.
JOCArticle
FIGURE 1. Molecular solid state structure of 2 (A, showing the two molecules 2A and 2B in the asymmetric unit), 3 (B), and 4 (C) with atom
labeling scheme. Displacement ellipsoids are drawn at the 50% probability levels.
with LiBH4, followed by a sequential one-pot Swern-Wittig
reaction to give the corresponding N-acylated γ-amino-R,β-
unsaturated ester, and finally intramolecular cyclization
through the Heck reaction. The butyl ester derivative 2 was
prepared by hydrolysis of 1 with aqueous LiOH, yielding the
corresponding acid 3, and subsequent reaction of its acid
chloride with BuOH. On the other hand, compound 1 was
reacted with Lawesson’s reagent to give the analogous
thiolactam, whose methyl ester group was later hydrolyzed
to acid 4 in high overall yield.
A common structural feature to molecules 2-4 is that the
exocyclic double bond is not in the same plane of the phenyl
ring, which, although limiting electron delocalization, avoids
steric hindrance due to allylic strain13 between the olefinic
hydrogen H10 and the aromatic hydrogen at C6. Again, the
deviation from the coplanarity of the 2,3-dihydro-1H-iso-
quinolin-4-ylidene system is considerably smaller in 2B. It is
interesting to note that the bond length of C8-C7 in 2B
is 0.02 A shorter than the corresponding length in 2A. In all
of the molecules, it is observed that the carbonyl and
thiocarbonyl groups at C1 are noncoplanar with the aro-
matic ring and this nonplanarity is more accentuated in the
thiolactam 4.
Solid State Structures. The solid state structures of com-
pounds 2-4 were determined by single-crystal X-ray diffrac-
tion and they are presented in Figure 1A-C.11
A noteworthy structural difference between molecules 2A
and 2B and compounds 3 and 4 is the relative disposition of
the CO group (C10-C11 conformation) with respect to the
double bond. Whereas the carbonyl group adopts an s-cis
conformation in molecules 2A and 2B, it is s-trans in
compounds 3 and 4 (in the isoquinoline derivatives with a
carboxylic group instead of an ester group).14
The bulky sec-butyl substituent adopts the axial position
in all the structures with torsion angles C7-C8-C9-C12 of
75.2-77.3ꢀ in 2A, 3, and 4, and of 97.9ꢀ in 2B. Only H12 of
the sec-butyl group lies directly over the lactam ring in 2A, 3,
and 4, whereas both H13a and H13b point toward the lactam
ring in 2B. Therefore, the sec-butyl group is rotated about
180ꢀ in 2B [C8-C9-C12-C13 is -176.6ꢀ in 2A, -68.3ꢀ in
2B, 179.78ꢀ in 3, and 174.3ꢀ in 4]. In addition, extended
conformations are preferred for ethyl groups in 2B, 3, and 4,
showing C9-C12-C13-C14 dihedral angles from 167.2ꢀ to
171.8ꢀ. Only in 2A, where this value is -69.5ꢀ, is the ethyl
chain folded above the isoquinoline moiety.
The three compounds crystallize in the noncentrosym-
metric P2(1)2(1)2(1) space group, so only one enantiomer is
present in the crystals. The absolute configuration was
assigned as (S,S) on the basis of the synthetic sequence
starting from L-isoleucine. For compound 4, the S con-
figuration was confirmed by the Flack parameter,12 χ =
0.0(3), based on the anomalous dispersion of the sulfur atom
with Mo KR radiation. Compound 2 crystallizes with two
independent molecules in the asymmetric unit, differing
mainly in the dihedral angle of the heterocyclic ring and in
the disposition of the sec-butyl group. The rms deviation
between both structures is 0.652 A excluding hydrogen
atoms. Tables S1 and S2 of the Supporting Information
show the structural parameters of 2, 3, and 4.
The solid state conformation of the fused bicyclic system is
quite similar in the three compounds. As expected, the
benzenic rings are planar and aromatic within experimental
errors. However, the lactam ring adopts a distorted envelope
conformation with an axially oriented sec-butyl group:
the four sp2 carbon atoms are coplanar with the benzene
ring (allowing the extension of the delocalization of the
π-electrons), and the sp3 one and the pyramidalized nitrogen
atom are puckered out of this plane. The deviation of the N1
and C9 atoms from the plane defined by the four sp2 carbon
atoms C1, C2, C7, and C8 is collected in Table S2 of the
Supporting Information. It shows that the δ-lactam ring is
significantly flatter in 2B.
With respect to the O-n-butyl group, while in 2B the
aliphatic chain is fully extended showing a dihedral angle
C16-C17-C18-C19 of 176.3ꢀ, in 2A the terminal methyl
group is folded with a torsion angle of 77.2ꢀ.
(13) (a) Johnson, F. Chem. Rev. 1968, 68, 375–413. (b) Hoffmann, R. W.
Chem. Rev. 1989, 89, 1841–1860.
(14) A Cambridge Structural Database (CSD) search has been performed
with the program ConQuest 1.10 to identify the conformational preferences
of CdC;CdO torsion angle in R,β-unsaturated carboxylic acids and esters.
Dihedral angle statistics were analyzed with the program Vista (see Figures
S1 and S2, Supporting Information). Examination of 260 X-ray structures
containing R,β-unsaturated ester moiety led to 271 s-cis and 22 s-trans
conformations (ratio s-cis/s-trans: 12.3/1), whereas the acid search yielded
133 compounds, finding 107 s-cis and 64 s-trans conformers (ratio s-cis/s-
trans: 1.67/1).
(11) Despite attempting additional refinement, a better quality of X-ray
diffraction data for compound 4 was not achieved. Nevertheless, the crystal-
lographic data of this compound were good enough to be used as the starting
point for DFT calculations.
(12) Flack, H. Acta Crystallogr., Sect. A: Found. Crystallogr. 1983, 39,
876–881.
344 J. Org. Chem. Vol. 75, No. 2, 2010

Alonso et al.
JOCArticle
TABLE 1. HOMA Index Values Calculated for Different Conjugated
π-System Moieties of Compounds 2-4
compd
HOMAphenyl
HOMAlactam
HOMAext. π-system
-0.331a
2A
2B
3
0.974
0.953
0.981
0.868
-0.971
-0.738
-0.623
-0.251
0.116a
-0.048a
0.056b
4
aThe extended π-system contains O1, C1, C2, C7, C8, C10, C11, and
O2 atoms. bThe π-system extends over atoms S1, C1, C2, C7, C8, C10,
C11, and O1.
The molecular structures of 2-4 are stabilized by a non-
classical hydrogen bond (H-bond),15 C9-H9 O, formed
3 3 3
from the carbonyl oxygen of the butyl ester in 2A and 2B and
from the alcoholic oxygen of the carboxylic group in 3 and 4.
This intramolecular interaction contributes to further stabi-
lize the observed axial configuration of the sec-butyl group
(see Figure 1).
FIGURE 2. Crystal structure of compound 2 showing the dimer
formed by two nonequivalent molecules linked by N-H
hydrogen bonds.
O
3 3 3
The structural aromaticity descriptor HOMA (Harmonic
Oscillator Model of Aromaticity)16 has been calculated in
order to evaluate the aromaticity of the individual rings of
the isoquinoline ring system and the degree of π-electron
delocalization between the six adjacent sp2 carbon (C1, C2,
C7, C8, C10, and C11) together with the oxygen/sulfur atoms
of the carbonyl/thiocarbonyl groups. HOMA values, calcu-
lated from X-ray diffraction data, for the different moieties
of compounds 2-4 are listed in Table 1. According to
HOMA, molecule 2B, the flattest structure, has the most
extended π-electron delocalization.
TABLE 2. Potential H-Bonds in the Crystal Structure of 2 and Their
˚
Geometries (A, deg)
D-H
A
dD-H dH 3 3 3
dD 3 3 3
—
class
A
A
DHA
3 3 3
N1a-H1a O1b 0.86
1.94 2.768(4)
2.13 2.971(4)
2.23 2.956(4)
2.26 2.859(4)
160
167
130
118
intermolecular
intermolecular
intramolecular
intramolecular
3 3 3
N1b-H1b O1a 0.86
3 3 3
C9a-H9a O3a 0.98
3 3 3
C9b-H9b O2b 0.98
3 3 3
Although the solid state molecular structures of 2, 3, and 4
are quite similar, the different functionalities at the ends of
the molecules cause quite different supramolecular packing,
reflecting the importance of intermolecular interactions on
crystal packing. Whereas 3 and 4 form H-bonds through the
O-H of the carboxylic group and N-H of the lactam or
thiolactam ring, only 2 can assemble the second kind of
H-bonds.
In the crystal structure of 2, the molecules of 2A and 2B
are forming a dimer through N-H O intermolecular
FIGURE 3. View along the c-axis of the crystal packing of 2
showing the hydrogen bonds (black dotted line) and the CH
interactions (red dotted line). Molecules A are indicated in green
3 3 3
π
H-bonds (Figure 2 and Table 2). According to the graph
set theory notation,17 this dimer is classified as a R22(8) motif.
The crystal packing of 2 (Figure 3) shows that the dimers are
arranged in antiparallel layers along the a-axis, and are
maintained by arene-arene interactions.18 Pairs of phenyl
rings of nonequivalent molecules overlap with an interplanar
separation of 3.697 A and a centroid-centroid distance of
3.977 A in a parallel-displaced orientation. The angle be-
tween both rings is 8.9ꢀ. The nonequivalent molecules in-
volved in the π-π stacking interaction are parallel to each
other and they are rotated as ca. 77ꢀ. Therefore, the alternate
3 3 3
and B are shown in blue.
arrangement of molecules of 2A and 2B along the a-axis
π interaction between the
makes suitable
a
CH
3 3 3
C15a-H15a group and the phenyl ring of the lower molecule
2B (H15a π = 3.302 A, C15a π = 3.302 A, and
3 3 3
3 3 3
C15a-H15a π =81.8ꢀ).15 A list of geometrical data of
3 3 3
putative nonclassic H-bonds of 2 is indicated in Table 2.
The crystal structures of 3 and 4 are quite similar: each
molecule of either 3 or 4 is linked to two neighboring
molecules through two different H-bonds (Table 3), leading
to an infinite chain of molecules along the b-axis. Figures 4
and 5 show the two kinds of H-bonds present in the crystal
(15) (a) Levitt, M.; Perutz, M. F. J. Mol. Biol. 1988, 201, 751–754.
(b) Umezawa, Y.; Tsuboyama, S.; Takahashi, H.; Uzawa, J.; Nishio, M.
Tetrahedron 1999, 55, 10047–10056. (c) Nishio, M. CrystEngComm 2004, 6,
130–158. (d) Nishio, M. Top. Stereochem. 2006, 25, 255–302. (e) Nishio, M.;
Umezawa, Y.; Honda, K.; Tsuboyama, S.; Suezawa, H. CrystEngComm
2009, 11, 1757–1788.
(16) Kruszewski, J.; Krygowski, T. M. Tetrahedron Lett. 1972, 13, 3839–
3842.
(17) (a) Etter, M. C. Acc. Chem. Res. 1990, 23, 120–126. (b) Bernstein, J.;
Davis, R. E.; Shimoni, L.; Chang, N. L. Angew. Chem., Int. Ed. Engl. 1995,
34, 1555–1573.
structures of 3 and 4, respectively: O-H O/S bonds
3 3 3
between the carboxylic group and the carbonyl/thiocarbonyl
accepting group of the isoquinoline ring, and N-H O H-
3 3 3
bonds connecting the lactam ring with the carboxylic groups.
According to the graph set theory, both can be classified as a
chain C(4) motif and their combination generates a new
R22(8) pattern. This pattern is quite common in carboxylic
acids as noted by Etter17a in her insightful analysis
of hydrogen bonds in crystal structures. Additionally,
(18) (a) Hunter, C. A.; Lawson, K. R.; Perkins, J.; Urch, C. J. J. Chem.
Soc., Perkin Trans. 2 2001, 651–669. (b) Hunter, C. A.; Sanders, J. K. M.
J. Am. Chem. Soc. 2002, 112, 5525–5534. (c) Tsuzuki, S. Struct. Bonding 2005,
115, 149–193.
J. Org. Chem. Vol. 75, No. 2, 2010 345

Alonso et al.
JOCArticle
TABLE 3. Possible H-Bonds in the Crystal Structure of 3 and 4 and
As mentioned above, there are also intramolecular
H-bonds of the type C-H X (X = O, N, and S) in the
crystal structure of 2-4 that assist to stabilize its molecular
conformation in the solid state. In 4, there is an additional
˚
Their Geometries (A, deg)
3 3 3
D-H
A
dD-H dH 3 3 3
dD 3 3 3
—
class
A
A
DHA
3 3 3
3
intermolecular C15-H15 S interaction between two iden-
O3-H3A O1a
0.97
0.86
0.98
1.61
2.578(2)
2.865(2)
2.954(2)
2.930(3)
172
159
intermolecular
intermolecular
3 3 3
3 3 3
tical adjacent molecules from different chains in parallel
N1-H1A O2b
2.04
2.24
2.5
3 3 3
arrangement. The C-H X interactions that fulfill the
C9-H9A O3
129 intramolecular
106 intramolecular
3 3 3
3 3 3
geometrical criteria for the existence of H-bonds19 in the
crystal structures of 3 and 4 are indicated in Table 3.
Conformational Studies on 2-4. Experimental and Calcu-
lated NMR Spectra in Solution. 1H NMR experiments were
carried out in order to determinate the conformation of
compounds 2-4 in solution. Moreover, the 1H and 13C
magnetic shielding tensors of compounds 2-4 were com-
puted with the GIAO method20 in order to compare with the
experimental spectral data. The X-ray structures were pre-
viously optimized at the B3LYP/6-31G(d) level of theory to
refine hydrogen atom positions.
C13-H13B N1 0.97
3 3 3
4
O2-H2A S1c
1.13
0.86
2.36
2.01
3.195(5)
2.865(7)
128
173
intermolecular
intermolecular
intermolecular
3 3 3
N1-H1A O1d
3 3 3
C15-H15A S1 0.96
2.79 3.390(12) 121
3 3 3
C9-H9A O3
0.98
0.93
2.28
2.72
2.969(8)
3.101(7)
127 intramolecular
105 intramolecular
3 3 3
C3-H3 S1
3 3 3
C13-H13A S1 0.97
3 3 3
2.51 2.921(10) 105 intramolecular
aSymmetry transformations: -x, y þ 1/2, -z þ 3/2. bSymmetry
transformations: -x, y - 1/2, -z þ 3/2. cSymmetry transformations:
-x þ 1, y - 1/2, -z þ /2. dSymmetry transformations: -x þ 1, y þ /2,
1
1
1
-z þ /2. The four molecules of the unit cell of 3 and 4 are related by
1H NMR experimental chemical shifts of compounds 2-4
pseudoinversion center (symmetry) except for the chiral groups.
1
as well as H-1H coupling constants are summarized in
Table 4. From the experimental coupling constants 3JH1,H9
,
the H9-C9-N1-H1 dihedral angle can be calculated by
using the following Karplus equation:21
3J ¼ A cos2 j -B cos j þ C
ð1Þ
As reported elsewhere,22 there are three different equa-
tions to model the vicinal coupling between amide NH and
RCH protons with the following parameters (eq 1): A = 6.7,
B =1.3, C = 1.5; A = 6.4, B = 1.4, C = 1.9; and A = 9.4,
B =1.1, C = 0.4. According to these parameters, the range of
values has been obtained for the H9-C9-N1-H1 torsion
and they are shown in Table 5. These data indicate that the
three compounds have nearly the same conformation in the
lactam ring with the sec-butyl group axially oriented. It is
worth noting the close similarity between the values of
H9-C9-N1-H1 dihedral angles determined in solution
and in the solid state (see Table 5) despite the fact that the
accuracy of the position of the hydrogens in the X-ray
analysis is not high. It is especially remarkable that this
dihedral angle is nearly identical in solution and in molecule
2B of compound 2.
FIGURE 4. View along the a-axis of the crystal packing of 3
showing the antiparallel chains of molecules, held by N-H
O
3 3 3
intermolecular hydrogen bonds (black contacts). CH π interac-
3 3 3
tions are represented by red dotted lines.
Since a good agreement between experimental and theo-
retical geometrical parameters is found, we can confirm that
this level of theory offers a good description of geometries for
these kinds of compounds. In fact, the values of the
H9-C9-N1-H1 dihedral angle of the optimized structures
are much closer to the experimental ones (see Table 5).
(19) The H X distance is less than the sum of the corresponding van der
FIGURE 5. View along the a-axis of the crystal packing of 4
showing the antiparallel chains of molecules linked by intermole-
3 3 3
Waals radii and the C-H X angle is greater than 100ꢀ. (a) Desiraju, G. R.;
3 3 3
Steiner, T. The Weak Hydrogen Bond in Structural Chemistry and Biology;
Oxford University Press/International Union of Crystallography: Oxford, UK,
1999. (b) Jeffrey, G. A. Crystallogr. Rev. 2003, 9, 135–176. (c) Bialonska, A.;
Ciunik, Z. CrystEngComm 2006, 8, 66–74. (d) Buckingham, A. D.; Del Bene, J.
E.; McDowell, S. A. C. Chem. Phys. Lett. 2008, 463, 1–10.
cular hydrogen bonds (black dotted lines). CH π interactions are
indicated by red dotted lines.
3 3 3
a nonclassical weak CH π H-bond (involving molecules
3 3 3
(20) (a) Cheeseman, J. R.; Trucks, G. W.; Keith, T. A.; Frisch, M. J.
J. Chem. Phys. 1996, 104, 5497–5509. (b) Wolinski, K.; Hinton, J. F.; Pulay,
P. J. Am. Chem. Soc. 2002, 112, 8251–8260.
belonging to different and antiparallel chains) is present in
the crystal structures of 3 and 4: a H-C(Ph) bond is nearly
perpendicular to the midpoint of a CdC bond in another
phenyl ring [H5 C5 =2.847 A and C5-H5 C5= 149.0ꢀ
(21) (a) Ramachandran, G. N.; Chandrasekaran, R.; Kopple, K. D.
ꢀ
Biopolymers 1971, 10, 2113–2131. (b) Demarco, A.; Llinas, M.; Wuthrich,
K. Biopolymers 1978, 17, 637–650. (c) Wuthrich, K.; Billeter, M.; Braun, W.
€
€
3 3 3
3 3 3
J. Mol. Biol. 1984, 180, 715–740.
(H5 C4 = 2.857 A and C5-H5 C4 = 167.9ꢀ) in 3;
3 3 3 3 3 3
(22) Karolak-Wojciechowska, J.; Czylkowski, R.; Karczmarzyk, Z.;
Paluchowska, M. H.; Rys, B.; Szneler, E.; Mokrosz, M. J. J. Mol. Struct.
2002, 612, 39–47.
H4 C5 = 3.032 A and C4-H4 C5 = 157.2ꢀ (H5 C6 =
3.095 A and C5-H5 C4 = 3.928ꢀ) in 4].
3 3 3
3 3 3
3 3 3
3 3 3
346 J. Org. Chem. Vol. 75, No. 2, 2010

Alonso et al.
JOCArticle
TABLE 4. Experimental (δexp, DMSO-d6) and Calculated (δcalc) 1H
NMR Chemical Shifts (ppm) for 2-4a
TABLE 6. Experimental (δexp, DMSO-d6) and Calculated (δcalc
)
13C
NMR Chemical Shifts (ppm) for 2-4
2
2
δcalc
3
4
δcalc
3
4
atom
H3
δexp
2A
8.22 8.41 7.89g
dd, 1H
7.44 7.62 7.72g
dd, 1H
7.44 7.37 7.57
m, 2H
2B
δexp
δcalc
8.33 8.37i
dd, 1H
7.48 7.61
m, 1H
7.47 7.61
m, 2H
δexp
δcalc
atom
δexp
2A
2B
δexp
δcalc
δexp
δcalc
7.92b
dd, 1H
7.80b
dd, 1H
7.72
m, 2H
6.43
s, 1H
5.20c
dd, 1H
1.61
m, 1H
1.37
m, 1H
1.13
m, 1H
0.75
m, 3H
0.75
m, 3H
4.13d
t, 2H
1.61
m, 2H
1.37
m, 2H
0.90e
t, 3H
6.48f
8.79
C1
C11
C8
C7
C5
C4
C2
C3
C6
C10
C9
C12
C13
C15
C14
C16
C17
C18
C19
165.3
162.4
150.2
134.8
132.7
130.6
128.7
127.1
124.6
116.4
54.2
41.3
24.2
15.1
10.9
63.8
30.2
18.7
13.6
163.0
169.9
161.4
137.7
131.8
130.8
131.9
129.9
124.0
116.7
51.9
40.0
24.0
10.2
2.8
59.8
28.2
20.6
7.5
162.1
169.3
160.9
137.6
132.2
131.3
131.9
130.5
123.0
113.9
59.0
44.8
21.0
11.6
8.4
64.5
30.9
20.6
11.2
162.5
166.7
149.1
135.1
132.6
130.3
128.7
127.0
124.4
117.6
53.9
162.8
167.2
159.2
137.8
132.1
131.2
131.1
129.9
124.9
117.7
58.4
189.3
166.7
147.9
133.0
131.1
131.0
130.6
130.1
124.4
118.7
55.6
197.5
166.9
157.7
133.3
132.6
130.6
131.8
134.4
124.6
117.7
58.7
H6
7.41
7.41
6.01
4.90
1.09
1.71
0.76
0.54
0.62
H4/H5
H10
H9
6.16 6.33 6.34
s, 1H
6.09 6.38
s, 1H
5.88 5.65 5.24h
dd, 1H
4.92 5.34j
dd, 1H
H12
H13a
H13b
H14
H15
H16
H17
H18
H19
NHa
CO2Ha
1.43 1.46 1.34
m, 1H
1.48 1.05 1.46
m, 1H
1.08 0.62 1.02
m, 1H
0.65 0.61 0.74
m, 3H
1.04 1.37
m, 1H
1.68 1.49
m, 1H
41.2
24.3
15.0
10.9
43.3
25.3
9.9
8.9
40.3
24.9
14.8
10.9
42.9
26.0
9.6
8.7
0.70 1.12
m, 1H
0.61 0.76e
t, 3H
0.65 0.61 0.74
m, 3H
5.58 3.84
0.61 0.70k
d, 3H
rmsd
4.4
3.9
3.5
4.0
1.43 1.46
1.48 1.05
0.62 0.74
both the size of the isoquinoline derivatives and the high
conformational flexibility of the side chains, the overall
quantitative agreement expressed by the rmsd is excellent,
specially for compound 3, which proves the efficiency of the
computational method. The highest discrepancy between
theory and experiment was observed for the C8 atom in all
the examined structures, with an error of about 10 ppm.
It is interesting to note that although the experimental
chemical shifts of C12 and C13 are closer to the calculated
ones for 2A, the δexp value for C14 is more similar to δcalc of
2B, suggesting a preferred sec-butyl conformation as in 2A,
where the ethyl group would be fully extended.
8.60f
d, 1H
12.64
br s, 1H
11.03
d, 1H
12.77
br s
br s, 1H
rmsd
0.55 0.30
0.28
0.31
aThe interchangeable proton signals (NH and CO2H) were not
calculated because δ has a great dependence with the concentration of
the sample by the formation of clusters. bJ1 = 7.1, J2 = 1.5. cJ1 = 7.3,
J2 = 4.9. dJ1 = 6.6. eJ1 = 7.3. J1 = 4.9. gJ1 = 7.6, J2 = 1.2. hJ1 = 7.1,
f
J2 = 4.6. J1 = 7.8, J2 = 1.2. J1 = 8.1, J2 = 4.9. kJ1 = 6.8 (1H-1H
coupling constants in Hz).
i
j
These data suggest that the conformation 2B dominates in
solution. On the basis of these results, we can conclude that the
conformations of the isoquinoline derivatives 2-4 in DMSO
solution are quite similar to those found in the solid state.
Conformational Analysis of Methyl (S,S,Z)-(3-sec-Butyl-
1-oxo-2,3-dihydro-1H-isoquinolin-4-ylidene)acetate (1). Be-
cause it was not possible to obtain crystals suitable for
X-ray diffraction analysis of the isoquinoline 1, the most
potent calpain inhibitor of this series of compounds, we per-
formed its conformational analysis by a combination of
molecular dynamics (MD) simulations and DFT calcula-
tions. Since the only difference between compounds 1 and 2
concerns the substituent of the ester moiety, the two solid
state conformations of derivative 2 were modified and used
as starting geometries for MD simulations. In addition,
taking into account the results observed in 2, a third con-
formation (1C) was considered. This conformer is similar to
1A, but with the sec-butyl group totally extended as in 1B.
Three different average structures were obtained and
they were further optimized at the B3LYP/6-31G(d) level
of theory (Figure 6).
TABLE 5. Experimental and Theoretical H9-C9-N1-H1 Dihedral
Angles (in deg) obtained for compounds 2-4
crystal
solution
theoretical
2A
2B
3
22.2
41.0
24.25
24.53
36-41
36-41
38-43
36-41
39.4
46.0
37.2
30.1
4
Interestingly, the conformer 2B is more stable than 2A by
5.3 kJ mol-1 (ΔG = 5.7 kJ mol-1).
The calculated and experimental values of the 1H and 13
C
NMR chemical shifts are collected in Tables 4 and 6,
respectively. The qualitative agreement between the calcu-
lated and experimental values is very good in all the exam-
ined structures: the order of values of the chemical shifts is
essentially the same experimentally and computationally (see
rmsd values). Only the order of the chemical shifts for C1 and
C11 of compound 2 is interchanged.23 Taking into account
The analysis of the 10-ns MD simulations indicates that
the sec-butyl group is always in the axial position and,
therefore, ring inversion is not observed at 300 K (see the
(23) The experimental assignment was based on HMBC correlations:
Bax, A.; Summers, M. F. J. Am. Chem. Soc. 2002, 108, 2093–2094.
J. Org. Chem. Vol. 75, No. 2, 2010 347

Alonso et al.
JOCArticle
FIGURE 6. Optimized geometries of the three conformers of iso-
quinoline derivative 1: (a) 1A; (b) 1B; (c) 1C; and (d) superposition
of the three conformations. The theoretical distances between the
hydrogen atoms in which NOEs were observed are shown.
FIGURE 8. (A) C9-C12-C13-C14 dihedral angle of 1A and (B)
C8-C9-C12-C13 dihedral angle of 1B plotted against time to
illustrate their interconversion to 1C along MD simulations in
DMSO. (C) Time-dependent variation of the C8-C10-C11-C12
torsion angle of 1C showing the fast s-cis/s-trans equilibria.
FIGURE 7. Plot of relative energy vs. C7-C8-C9-C12 dihedral
angle of compound 2.
Supporting Information, Table S3). In fact, the calculated
energy barrier for the ring inversion of methyl ester 1 is
39.8 kJ mol-1 (Figure 7). Furthermore, the exocyclic double
other hand, qualitative NOE experiments (DMSO-d6) re-
vealed a strong enhancement between the olefinic proton
H10 and the aromatic protons, confirming the presence of
the Z isomer in solution. Further weaker NOE enhance-
ments were observed between the protons attached to the
methyl group C15 and the methyl ester, and between the
amidic proton and the methyl group of the ethyl moiety,
respectively. These NOE enhancements are consistent with a
major conformer 1A, in which the corresponding methyl
groups are pointing at the methyl ester and at the NH,
respectively. In conformer 1B, the distances between these
protons exceed 4.5 A.
3
bond is never coplanar with the phenyl ring along the MD
trajectories of all conformers, and thus, the less crowded face
of the electrophilic double bond is easily accessible for a
nucleophilic attack. A remarkable fact is that both the initial
structures of 1A and 1B converged to 1C, and all the
conformations underwent free rotation around the single
bond of the R,β-unsaturated ester system prevailing (more
than 90%) the s-cis configurations. Figure 8 shows a graphic
representation of the selected dihedral angles involved in
conformational exchange along MD trajectories. Similarly
to compound 2, the flattest conformation 1B is more stable
than 1A by 2.1 kJ mol-1 (ΔG = 3.3 kJ mol-1).
The experimental and calculated values of the 1H and 13
C
1H NMR and NOE experiments provide evidence that the
conformation of compound 1 in solution is quite similar to
the proposed ones. Thus, on one hand the range of values
obtained for the H9-C9-N1-H1 dihedral angle (3JH1,H9) is
33-40ꢀ, being closer to the torsion of -40ꢀ present in
conformer 1A (Supporting Information, Table S4). On the
NMR chemical shifts of the three conformers are summar-
ized in Table 7. A visual inspection of the data reveals that
1
the experimental H and 13C chemical shifts are consistent
with those calculated theoretically. As previously observed
with 2-4, only the order of the chemical shifts for C1 and
C11 of compound 1 is interchanged. The rmsd between the
348 J. Org. Chem. Vol. 75, No. 2, 2010

Alonso et al.
JOCArticle
TABLE 7. Experimental δexp and Calculated δcalc 1H and 13C NMR
Chemical Shifts (ppm) of Conformers 1A, 1B, and 1C
(Table S3). These data suggest that the s-cis rotamer is
the major one at equilibrium in solution for the esters
of (3-sec-butyl-1-oxo-2,3-dihydro-1H-isoquinolin-4-ylidene)-
acetic acid.
δcalc
atom
δexp
1A
1B
1C
Computational Support to the Biological Activity. A com-
mon mechanism for the inactivation of cysteine protease is
through the reaction of the SH-group of the cysteine residue
with an electrophilic functionality of the inhibitor.24
μ-Calpain is a particular cysteine protease whose distinctive
1H NMR
8.33
H3
7.92a
dd, 1H
7.81a
dd, 1H
7.62
m, 2H
6.43
s, 1H
5.19b
dd, 1H
3.71
s, 3H
1.62
m, 3H
0.75
m, 6H
8.37
7.52
7.42
6.32
5.63
3.45
1.01
0.63
8.31
7.41
7.42
6.23
5.62
3.44
1.18
0.61
H6
7.46
7.41
6.17
5.96
3.43
1.40
0.67
feature is the activation by Ca2þ 25
Despite this different
.
H4/H5
H10
mechanism of activation, it is well-known that the active site
of calpain is a pocket located in the major subunit with an
active cysteine residue, quite similar to papaine and other
related cysteine proteases. Therefore, the inhibition of
calpain by 1 and related compounds is likely through the
reaction with the SH-group of the cysteine residue of the
active center of calpain.26 The high activity of 1 as a calpain
inhibitor must arise from two facts: on one side, a suitable
recognition pattern by the active site of the enzyme,27 and on
the other hand, an adequate reactivity of the electrophilic
functionality of the inhibitor.
H9
H16
H12/H13
H15/H14
rmsd
0.37
0.35
0.33
The bond-forming step involves movement of electrons
from the HOMO of the nucleophile to the LUMO of the
unsaturated carbonyl compound. Thus, the inhibitory acti-
vity of these Michael acceptors must correlate with the
electron affinities (A) of the electrophilic double bond. The
higher A, the more reactive the Michael acceptor, and
consequently a greater inhibition capacity could be expected.
Figure 9 shows the LUMO surfaces of compounds 1-4. The
shape of the LUMO orbital is very similar in the four
molecules, even though the coefficients of the atomic orbital
vary from esters 1 and 2 to acids 3 and 4 (Table 8). In the case
of esters 1 and 2, the largest LUMO coefficient is at the
β-carbon of the R,β-unsaturated system (C8), suggesting that
the nucleophilic attack is directed toward this atom. It is
noteworthy that the highest coefficient at C8 is obtained for
isoquinoline derivative 1.
13C NMR
162.6
170.4
161.5
137.7
132.0
130.8
131.2
129.8
124.0
115.9
51.4
C1
C11
C8
C7
C5
C4
C2
C3
C6
C10
C9
C16
C12
C13
C15
C14
165.6
162.5
150.6
134.8
132.7
130.6
128.8
127.1
124.5
115.9
54.3
161.9
169.7
161.4
137.3
132.0
132.0
131.7
130.5
122.6
113.3
59.3
162.4
170.4
161.6
138.0
131.9
131.4
131.5
129.9
124.0
117.1
56.2
51.4
41.4
24.2
15.0
48.8
39.8
23.8
10.2
48.8
45.1
21.1
11.3
48.8
43.5
24.9
9.0
10.8
2.8
8.6
8.9
rmsd
4.46
4.28
4.25
Electron affinities were computed at the B3LYP/6-
31þþG(d,p) level of theory, since this method has been
demonstrated to be accurate in reproducing experimental
A values with an error of less than 0.1 eV.28 Furthermore,
with the aim to explain the differences observed in the
aJ1 = 7.08 Hz, J2 = 1.46 Hz. bJ1 = 7.08 Hz, J2 = 5.13 Hz.
experimental and calculated chemical shifts is very similar
for all the conformers. While the experimental NMR signals
for H9 and H10 are more similar to the estimated values of
conformers 1B and 1C, the sec-butyl moiety corresponds
better with a conformation type 1A. Furthermore, regarding
13C NMR chemicals shifts, the lowest rmsd value is found for
1C where δcalc for C14 is quite similar to δcalc for 1B.
All these data seem to support that the preferred confor-
mation is 1C, which is quite similar to 1A but with an
extended conformation of the ethyl moiety of the sec-butyl
group. DFT calculations on 1C indicate that this conforma-
tion is more stable than 1B and 1A with small energy
differences of 0.42 and 2.5 kJ mol-1, respectively.
Since the starting geometries of 1 were built by modifica-
tion of solid state structures of 2, the three conformations
obtained for compound 1 are also s-cis. To evaluate the
relative stability of the s-trans and the s-cis rotamers, we have
optimized the s-trans conformer of 1. According to the
calculations, the s-cis conformer is around 13 kJ mol-1 more
stable than the s-trans in all studied conformations. Even
though the cisoid-transoid conformational equilibrium is
observed along the MD trajectories of 1A-C, the s-trans
configurations are only detected below 10% of the time
(24) (a) Berti, P. J.; Storer, A. C. J. Mol. Biol. 1995, 246, 273–283.
(b) Otto, H.-H.; Schirmeister, T. Chem. Rev. 1997, 97, 133–172. (c) Leung-
Toung, R.; Li, W.; Tam, T. F.; Kaarimian, K. Curr. Med. Chem. 2002, 9,
979–1002. (d) Schirmeister, T.; Kaeppler, U. Mini Rev. Med. Chem. 2003, 3,
361–373. (e) Santos, M. M. M.; Moreira, R. Mini Rev. Med. Chem. 2007, 7,
1040–1050.
ꢀ
(25) (a) Bozoky, Z.; Alexa, A.; Tompa, P.; Friedrich, P. Biochem. J. 2005,
388, 741–744. (b) Croall, D. E.; Vanhooser, L. M.; Cashon, R. E. Biochim.
ꢀ ꢀ
ꢀ ꢀ
Biophys. Acta 2008, 1784, 1676–1686. (c) Toke, O.; Banoczi, Z.; Tarkanyi,
G.; Friedrich, P.; Hudecz, F. J. Pept. Sci. 2009, 15, 404–410. (d) This
mechanistic feature of activation is unique between protease, and it has been
advantageously used to design nonelectrophilic calpain inhibitors, see:
Montero, A.; Mann, E.; Chana, A.; Herradon, B. Chem. Biodiversity 2004,
1, 442–457. (e) Montero, A.; Albericio, F.; Royo, M.; Herradon, B. Org. Lett.
2004, 6, 4089–4092, and references cited therein (f) Although the electro-
philicity of the conjugate double bonds in the carboxylates is expected to be lower
than that in the corresponding esters, we evaluated both kinds of compounds as
calpain inhibitors since they can act through the nonelectrophilic mechanism
indicated above.
(26) Kunakbaeva, Z.; Carrasco, R.; Rozas, I. THEOCHEM 2003, 626,
209–216, and references cited therein
(27) (a) Hubbard, S. J.; Thornton, J. M.; Campbell, S. F. Faraday Discuss.
1992, 93, 13–23. (b) Tyndall, J. D. A.; Nall, T.; Fairlie, D. P. Chem. Rev. 2005,
105, 973–1000.
(28) Baranovski, V. I.; Denisova, A. S.; Kuklo, L. I. THEOCHEM 2006,
759, 111–115.
J. Org. Chem. Vol. 75, No. 2, 2010 349

Alonso et al.
JOCArticle
from the geometry of the neutral system, calculating the
electronic energies of the corresponding cation and anion.
The three conformations of methyl ester (1A-C) and the
two of butyl ester (2A,B) have positive values of A, being the
electron affinity values of 1 are slightly higher than those
of 2. So, these compounds react favorably with the cysteine
residue of the calpain active site. Likewise, the electrophilic
index (ω), a property that measures the capability of a
molecule to accept electrons, is higher in the isoquinoline
derivative 1. Since the electron affinity and electrophilic
index of 1 are higher, its inhibitory activity must be greater.
On the other hand, carboxylates 3 and 4 have highly negative
A values and small ω values, which explain the inactivity
toward calpain of these carboxylic acids or other derivatives
containing a carboxylic group close to the R,β-unsaturated
system.
Conclusions
A comprehensive structural analysis on derivatives of
(3-sec-butyl-2,3-dihydro-1H-isoquinolin-4-ylidene)acetic acid
has been performed with experimental (NMR and X-ray
diffraction) and computational methods. The results have
shown differences between ester and acid derivatives in both
molecular structure and crystal packing, these differences
steming from different functionalities present in the mole-
cules (amides, thioamides, ester, carboxylic acids, arene,
C-H donating groups) able to participate in a variety of
intermolecular interactions. Since these structural features
have been frequently used as components of tectons,30 the
experimental results can be useful in the field of crystal
engineering.31 While the molecules of ester 2 are arranged
in dimers by means of H-bonds, acids 3 and 4 are organized
in antiparallel chains. The conformational analysis and
theoretical study of the most potent calpain inhibitor 1 have
confirmed, apart from the dynamic mobility among confor-
mers, that the sec-butyl group in the axial position freezes the
ring inversion in these isoquinoline derivatives, leaving
accessible the less crowded face of the exocyclic double
bond. Moreover, we have further demonstrated a structure-
activity relationship for the calpain inhibition, based on the
reactivity of the double bond of this kind of compounds with
the SH-group of the cysteine residue of the calpain active site.
The acidic compounds 3 and 4 are noninhibitors since at
physiological pH the negative charge of the carboxylate
group prevents the approach of the sulfur anion. In addition,
the highest inhibitory activity of the ester derivatives is
related to both the shape of the LUMO orbital and the
highest values of electron affinity and electrophilic index.
In summary, the theoretical calculations model with high
accuracy the experimental results and we have proved that
the COOR group is a fundamental factor for inhibiting
calpain in isoquinoline derivatives.
FIGURE 9. LUMO orbital of compounds 1 (conformer A),
2 (conformer B), and carboxylates of 3 and 4.
TABLE 8. Calculated LUMO Coefficients for the Atoms of the R,β-
Unsaturated Carboxylic System for Derivatives 1-2 and Carboxylates
of 3-4
a
b
compd
C8
C10
C11
O2
O3
1A
1B
1C
2A
2B
3
0.2990
-0.2812
0.2875
-0.2862
-0.2753
-0.0731
-0.1189
-0.2426
0.2333
-0.2364
0.2380
0.2313
0.1923
0.1827
-0.2188
0.2020
-0.2143
0.2134
0.1971
0.0362
0.0925
0.2033
-0.1989
0.2001
-0.2014
-0.1950
-0.0671
-0.0749
0.1468
-0.1382
0.1434
-0.1444
-0.1359
-0.0808
-0.1038
4
aO1 in compound 4. bO2 in compound 4.
TABLE 9. B3LYP/6-31þþG(d,p) Reactivity Descriptors (values in
eV) for Derivatives 1-2 and Carboxylates 3-4
ΔELUMO-
HOMO
compd εHOMO εLUMO
A
I
μ
η
ω
1A
1B
1C
2A
2B
3
-6.85 -2.36
-6.74 -2.42
-6.88 -2.36
-6.83 -2.32
-6.72 -2.39
-2.17
-2.36
4.49
4.32
4.51
4.51
4.33
3.35
2.97
0.854 8.362 -4.608 3.754 2.828
0.914 8.250 -4.582 3.668 2.862
0.855 8.388 -4.621 3.767 2.835
0.841 8.313 -4.577 3.736 2.803
0.905 8.199 -4.552 3.647 2.841
-2.422 4.186 -0.882 3.304 0.118
-2.021 4.222 -1.101 3.121 0.194
1.18
0.61
4
biological activity, other electronic properties, apart from
the electron affinities, such as HOMO and LUMO orbital
energies (εHOMO and εLUMO), the energy difference between
frontier orbitals (ΔEHOMO-LUMO), as well as global reactiv-
ity indices (chemical potential, μ; chemical hardness, η; and
electrophilicity index, ω) derived from DFT have been
calculated (Table 9).29 Taking into account that carboxylic
acids 3 and 4 are deprotonated at pH values of the experi-
mental assays, the electronic properties have been calculated
from their corresponding carboxylates. Electron affinity (A)
and ionization potential (I) calculations were carried out
Experimental Section
Crystal Structure Analysis of Derivatives of (3-sec-Butyl-2,3-
dihydro-1H-isoquinolin-4-ylidene)acetic Acid. Suitable single
(30) Hosseini, M. W. Acc. Chem. Res. 2005, 38, 313–323.
(31) (a) Moulton, B.; Zaworotko, M. J. Chem. Rev. 2001, 101, 1629–1658.
(b) Braga, D. J. Chem. Soc., Chem. Commun. 2003, 2751–2754. (c) Price, S.
L.; Price, L. S. Struct. Bonding 2005, 115, 81–123. (d) Desiraju, G. R. Angew.
Chem., Int. Ed. 2007, 46, 8342–8356.
(29) For additional information about the theoretical background of the
reactivity indices see: Alonso, M.; Casado, S; Miranda, C.; Tarazona, J. V.;
Navas, J. M.; Herradon, B. Chem. Res. Toxicol. 2008, 21, 643–658.
350 J. Org. Chem. Vol. 75, No. 2, 2010

Alonso et al.
JOCArticle
crystals of either 2, or 3, or 4 were obtained at room temperature
by slow diffusion of methanolic solutions of the corresponding
derivative. Data collection for X-ray analysis was obtained with
use of an area detector single-crystal diffractometer with gra-
phite monochromated Mo KR radiation (λ = 0.71073 A)
operating at 50 kV and 30 mA. Data were collected over a φ
and ω scans hemisphere of the reciprocal space by a combina-
tion of the number of frames of intensity sets. Each frame
covered 0.3ꢀ in ω and the first 50 frames were recollected at
the end of data collection to monitor crystal decay. Absorption
corrections were applied with the SADABS program.32 The
structures were solved with the SHELXTL-PC software33 by
direct methods and refined by full-matrix least-squares methods
on F2. Treatment of hydrogen atoms was mixed, located in
density maps and included in calculated positions, and refined in
the riding mode. The details of the data collection and structure
refinement are summarized in Table S5 of the Supporting
Information. The structure drawings were prepared with the
programs MERCURY34 and ORTEP-3.35 The absolute con-
figurations at the stereogenic centers were assigned according to
the synthetic schemes that start from amino acids of known
stereochemistry.
31G* level of theory.37-39 Vibrational frequency calculations at
the same level of theory confirmed that all structures were
minima on the potential energy surface. For compounds 2-4,
the starting geometries for the calculations were those obtained
in the X-ray diffraction analysis. In the case of 2, the two
different conformations (2A and 2B) found in the crystallo-
graphic unit cell were considered. On the other hand, the starting
geometry for the quantum-chemical calculation of 1 was ob-
tained from the molecular dynamics (MD) simulations, using
MMFF94 force field40 and model charges as implemented in
Sybyl, version 7.0.41 The MD simulations were performed at
constant temperature (300 K) by coupling the system to a
thermal bath, using the Berendsen algorithm with a coupling
constant of 100 fs. The simulations were carried out in a vacuum
(with a distance-dependent dielectric constant ε = 1) for 10 ns,
using a time step of 1 fs, leaving 20 ps to equilibrate the system.
As the starting conformation, we used analogues of both
molecular solid state structures of compound 2, replacing the
butyl ester by a methyl ester and subsequently minimizing.
Additionally, the inversion barrier energy for the lactam ring
was estimated at the B3LYP/6-31G(d) level of theory for
compound 1 by changing the C7-C8-C9-C12 dihedral angle
by 10ꢀ steps, from 60ꢀ to 120ꢀ.
The structure of compound 2 was determined by single-
crystal X-ray diffraction at 100 and 298 K. At room tempera-
ture, due to the higher conformational flexibility of the aliphatic
chains, the molecular structure presents certain disorder in the
atomic positions of the sec-butyl group as well as in the butyl
ester of both molecules 2A and 2B.
1
The H and 13C magnetic shielding tensors of the B3LYP/
6-31G*-optimized structures were computed with the Gauge-
Independent Atomic Orbital (GIAO) method at the B3LYP/
6-311þG(2d,p) level.20 To compare isotropic shieldings with the
experimentally observed chemical shifts, the NMR parameters
for TMS were calculated at the same level and used as the
reference molecule. For compound 1, the shielding computa-
tions were also performed at the RHF/6-311þG(2d,p) level of
theory, finding a worse correlation with the experimental data
and a larger computational time.
NMR Measurements. Chemical shifts (δ) are reported in parts
per million and the coupling constants are indicated in hertz.
1
The H and 13C NMR spectra data were recorded in 2 mM
DMSO-d6 solutions. H NMR spectra were referenced to the
1
chemical shift of either TMS (δ = 0.00 ppm) or the residual
proton in the deuterated solvent. 13C NMR spectra were
referenced to the chemical shift of the deuterated solvent.
NOESY spectra were acquired by using the standard pulse
sequence.
Computational Details. DFT calculations were performed
with the Gaussian-03 suite of programs.36 Full geometry opti-
mizations of compounds 1-4 were performed at the B3LYP/6-
The HOMA values were calculated for both the solid state
molecular structures and the DFT-optimized geometries, using
eq 2:16,42
n
X
R
2
HOMA ¼ 1 -
ðRopt - RiÞ
ð2Þ
n
i ¼1
where n is the number of atoms taken into the summation,
and R is an empirical constant fixed to give HOMA = 0 for
a model nonaromatic system and HOMA = 1 for a system
with all bonds equal to an optimal value Ropt, assumed to
be realized for a fully aromatic system. Ri is the running bond
length.
Finally, the electron affinity (A), ionization potential (I), as
well as the global reactivity descriptors have been calculated at
the B3LYP/6-311G(d,p) level of theory, since this method has
been demonstrated to be accurate in reproducing experi-
mental A values with an error of less than 0.1 eV.43 The
theoretical basis for the reactivity descriptors has been amply
developed elsewhere.44 I and A are determined from the electro-
nic energies of the systems having N - 1, N, and N þ 1 electrons
at the geometry of the neutral system. Using I and A, the
chemical potential (μ), the chemical hardness (η), and the global
electrophilic index (ω) were calculated according to eqs 3-5,
respectively.
(32) Sheldrick, G. M. SADABS, Program for Absorption Corrections
€
€
Using Bruker CCC Detectors, University of Gottingen, Gottingen, Ger-
many, 1986.
(33) Sheldrick, G. M. SHELXTL-PC, Program for Crystal Structure
€
€
Solution, University of Gottingen, Gottingen, Germany, 1997.
(34) MERCURY, CSD 2.0, New Features for the Visualization and
Investigation of Crystal Structures; Macrae, C. F.; Bruno, I. J.; Chisholm, J.
A.; Edgington, P. R.; McCabe, P.; Pidcock, E.; Rodriguez-Monge, L.;
Taylor, R.; van de Streek, J.; Wood, P. A. J. Appl. Crystallogr. 2008, 41,
466-470. Web address: http://www.ccdc.cam.ac.uk/prods/mercury/.
(35) ORTEP-3, for Windows; Farrugia, L. J. J. Appl. Crystallogr. 1997,
30, 565-566. Web address: http://www.chem.gla.ac.uk/∼louis/software/or-
tep3/.
(36) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.;
Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci,
B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada,
M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.;
Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.;
Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain,
M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski,
J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision B.04;
Gaussian, Inc., Pittsburgh, PA, 2003.
ðI þ AÞ
μ ≈ -
ð3Þ
2
(40) (a) Halgren, T. A. J. Comput. Chem. 1996, 17, 490–641. (b) Halgren,
T. A. J. Comput. Chem. 1999, 20, 720–748.
(41) Sybyl molecular modeling system; Tripos Associated, St. Louis, MO.
(42) Krygowski, T. M. J. Chem. Inf. Comput. Sci. 1993, 33, 70–78.
(43) Baranovski, V. I.; Denisova, A. S.; Kuklo, L. I. THEOCHEM 2006,
759, 111–115.
(37) Becke, A. D. Phys. Rev. A 1988, 38, 3098–3100.
(38) Becke, A. D. J. Chem. Phys. 1993, 98, 1372–1377.
(39) Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. Rev. B 1988, 37, 785–789.
(44) Geerlings, P.; De Proft, F.; Langenaeker, W. Chem. Rev. 2003, 103,
1793–1874.
J. Org. Chem. Vol. 75, No. 2, 2010 351

Alonso et al.
JOCArticle
Supporting Information Available: Structural parameters of
2, 3, and 4 (Tables S1 ans S2), lowest, highest and average values
of dihedral angles involved in the conformational exchange of
1A-1C (Table S3), selected dihedral angles for conformers 1A,
1B, and 1C calculated at B3LYP/6- 31G* (Table S4), and crystal
data and structure refinement for 2-4 (Table S5); histogram
showing the distribution of CdC;CdO torsion angles (Figures
ðI -AÞ
η ≈
ð4Þ
ð5Þ
2
2
μ2
ðIþAÞ
ω ¼
¼
2η 4ðI -AÞ
Acknowledgment. Taken in part from the Ph.D. theses of
M.A and R.C. We thank the Spanish Ministry of Education
and Science (Projects CTQ2004-01978 and CTQ2007-64891/
BQU) for financial support and the Centro de Super-
1
S1 and S2), characterization data, and copies of H and 13C
NMR spectra of compounds 1-4; X-ray crystallographic files
(CIF) for derivatives 2-4; and computational data. This ma-
terial is available free of charge via the Internet at http://pubs.
acs.org. The CIF files are also available from the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge,
CB2 1EZ, U.K. [fax: þ44(0) 1223 336033 or e-mail deposit@
ccdc.cam.ac.uk], under CCDC reference nos. 748814 (2), 748815
(3), and 748816 (4).
ꢀ
computacion de Galicia (CESGA) for the use of the Compaq
HPC 320 supercomputer. M.A. thanks the Spanish Ministry of
Education and Science for a FPU fellowship. C.M. thanks the
Spanish National Research Council for a JAE-Doc contract.
ꢀ
We also thank Mr. Ivan Escabias for technical assistance.
352 J. Org. Chem. Vol. 75, No. 2, 2010