Bivalent 5,8,9,13b-tetrahydro-6H-isoquino[1,2-a]isoquinolines and -isoquinolinium salts: Novel heterocyclic templates for butyrylcholinesterase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2010.03.011

Three different types of homobivalent compounds, 5,8,9,13b-tetrahydro-6H-isoqino[1,2-a]isoquinolines bearing tertiary N-atoms, their quaternary ammonium salts and their dibenzazecine analogues, connected by alkylene spacers of various lengths were synthes

Full text of DOI:10.1016/j.bmcl.2010.03.011

Bioorganic & Medicinal Chemistry Letters 20 (2010) 2946–2949
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Bivalent 5,8,9,13b-tetrahydro-6H-isoquino[1,2-a]isoquinolines
and -isoquinolinium salts: Novel heterocyclic templates for butyrylcholinest-
erase inhibitors
a
b
a
a,*
Maria Schulze , Oliver Siol , Michael Decker , Jochen Lehmann
a
Lehrstuhl für Pharmazeutische/Medizinische Chemie, Institut für Pharmazie, Friedrich-Schiller-Universität Jena, Philosophenweg 14, D-07743 Jena, Germany
Lehrstuhl für Pharmazeutische Biologie, Institut für Pharmazie, Friedrich-Schiller-Universität Jena, Semmelweisstrasse 10, D-07743 Jena, Germany
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Three different types of homobivalent compounds, 5,8,9,13b-tetrahydro-6H-isoqino[1,2-a]isoquinolines
bearing tertiary N-atoms, their quaternary ammonium salts and their dibenzazecine analogues, con-
nected by alkylene spacers of various lengths were synthesized. Compared to the therapeutically used
inhibitor galanthamine, some of the bivalent compounds showed much higher inhibitory activities at
both cholinesterases in the Ellman test. Surprisingly, not only the quaternary salts, but also the
uncharged tertiary compounds exhibited IC50 values at butyrylcholinesterase in the nanomolar range.
Selectivity toward BChE of up to 76-fold was observed.
Received 4 February 2010
Revised 2 March 2010
Accepted 2 March 2010
Available online 6 March 2010
Keywords:
Bivalent inhibitor
Acetylcholinesterase
Butyrylcholinesterase
Cytotoxicity
Ó 2010 Elsevier Ltd. All rights reserved.
Cholinesterase (ChE) is one major target in the current therapy
of Alzheimer’s disease (AD). Clinically approved inhibitors of ace-
tylcholinesterase (AChE) like rivastigmine, galanthamine and
donepezil are the most commonly applied drugs for symptomatic
treatment of cognitive deficits in AD. For the effectiveness of ther-
apeutic intervention by ChE inhibition, the importance of a second,
less specific cholinesterase in the human body, butyrylcholinester-
ase (BChE), is under current focus. While the treatment with selec-
tive AChE inhibitors is effective in the beginning of AD, their
effectiveness decreases with progression of the disease which
may be due in part to lower AChE-(10–15% of normal during AD
progression), but elevated BChE-levels.1 Positive effects of the
administration of the selective BChE inhibitor cymserine on learn-
In this regard, additionally bivalent compounds should be syn-
thesized and evaluated pharmacologically. In the last decade, the
bivalent ligand/inhibitor approach by combining covalently two
identical (homobivalent) or chemically related (heterobivalent)
drug molecules was successfully applied in various areas of thera-
peutic research (including AChE and BChE inhibitors) to yield
highly potent compounds with sometimes remarkably increased
5
–7
selectivity profiles.
Bivalent compounds of galanthamine were
investigated by Guillou et al. and revealed inhibitory potencies of
the homobivalent galanthamine derivatives equal to that of the
univalent compound galanthamine at AChE and improved inhibi-
8
tion by heterodimers with galanthaminium substructure. The im-
proved enzyme inhibition of the quaternary salt is not surprising,
since the natural substrate acetylcholine (ACh) itself is a quater-
nary ammonium salt. Hence, we used compound 2 as a potential
univalent inhibitor to design homobivalent molecules with alkyl-
ene chains of various lengths as spacer. A potential problem of
these compounds lies in the fact that it is unlikely for charged mol-
ecules to pass the blood–brain barrier which represents a basic
requirement for drugs against neurodegenerative disorders. Since
a competitive and reversible ChE inhibitor interaction with the cat-
alytic active site (CAS) is often mediated by a protonated tertiary
N-atom, we also tested the uncharged compounds for their inhib-
itory potency. Univalent structures are shown in Figure 1.
2
ing and b-amyloid peptide formation could be shown in rodents.
Compound 1 (3-methoxy-5,8,9,13b-tetrahydro-6H-isoquino-
[
1,2-a]isoquinoline) was originally synthesized as a precursor for
3
highly potent dopamine antagonists.
Some structural similarities of compound 1 to galanthamine
can be found (Fig. 1). We therefore considered it to be of interest,
if these structures might exhibit ChE-inhibiting properties, since
there are considerable efforts made to find novel structural tem-
4
plates for ChE inhibitors with improved therapeutic profiles.
The univalent precursor 4 was synthesized according to the
synthetic scheme described by Mohr et al., using (3-methoxy-
*
Corresponding author. Tel.: +49 3641 94 9800; fax: +49 3641 94 9802.
E-mail addresses: m.decker@qub.ac.uk (M. Decker), j.lehmann@uni-jena.de (J.
3
Lehmann).
phenyl)ethylamine and isochroman-1-one as starting materials
0
960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.03.011

M. Schulze et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2946–2949
2947
OH
O
-
O
+
I
H C
H3C
N
H C
H C
3
3
3
N
N
O
O
O
CH3
CH3
N
CH3
galanthamine
1
2
3
Figure 1. Tetracyclic compounds 1 and 2, tricyclic compound 3 and the AChE inhibitor galanthamine.
(
Scheme 1). Synthesis of homobivalent compounds 5a–f with ter-
Table 1
Inhibitory activities at AChE- and BChE and resulting selectivities toward BChE
tiary N-atoms was achieved in a Williamson ether synthesis by
deprotonation of phenolic hydroxy groups using sodium hydride
in DMF, followed by alkylation using half the molar amount of
the respective dibromoalkane. Quaternization was performed with
methyl iodide in acetonitrile/methanol. Compounds 7b and 7e
were obtained under Birch conditions using sodium in liquid
ammonia. Univalent structures were synthesized in the same
manner.
IC50 (AChE)a
(pIC50 ± SEM)
Galanthamine 0.64 (6.197 ± 0.052) 8.40 (5.076 ± 0.034)
IC50 (BChE)a
(pIC50 ± SEM)
Compound
(lM)
(lM)
IC50(AChE)/
IC50(BChE)
0.08
—
1
2
3
5
5
>10,000
>10,000
8.8 (5.057 ± 0.055)
>10,000
6.4 (5.197 ± 0.020)
>10,000
1.4
—
a
b
4.5 (5.347 ± 0.044)
1.1 (5.971 ± 0.076
0.15 (6.835 ± 0.051) 30.8
0.017 (7.768 ± 0.048) 76.2
We examined bivalent quaternary compounds 6a–f, their non-
charged precursors 5a–f, the bivalent dibenzoazecines 7b,e and
the univalent congeners 1, 2, and 3 in the colorimetric Ellman assay
at AChE (E.C. 3.1.1.7, type VI-S, from Electric Eel) and BChE (E.C.
5c
5d
0.65 (6.511 ± 0.178) 0.017 (7.773 ± 0.036) 38.3
0.69 (6.164 ± 0.158) 0.014 (7.846 ± 0.006) 48.9
5
5
6
6
e
f
a
b
1.2 (5.920 ± 0.071)
4.1 (5.385 ± 0.051)
0.45 (6.343 ± 0.121) 0.16 (6.806 ± 0.033)
0.68 (6.168 ± 0.047)
0.92 (6.035 ± 0.034)
3.0
4.2
5.8
3
.1.1.8, from equine serum). Although there are a number of spe-
0.39 (6.406 ± 0.076) 0.051 (7.293 ± 0.037) 8.8
0.28 (6.546 ± 0.036) 0.047 (7.326 ± 0.055) 5.9
0.15 (6.822 ± 0.020) 0.027 (7.570 ± 0.087) 5.6
cies-dependent differences between these enzymes and the human
ones, the enzymes used show sufficient homology in the amino acid
6c
6d
9
6e
0.28 (6.557 ± 0.137) 0.11 (6.954 ± 0.034)
2.5
sequence to the human enzymes (e.g., 88% for equine serum BChE).
6
7
f
b
0.12 (6.918 ± 0.055) 0.036 (7.452 ± 0.035) 3.4
Results of the cholinesterase assay are given in Table 1.
All compounds revealed higher inhibitory activities at both cho-
linesterases than the univalent congeners. As expected due to the
similarity with the natural ligand ACh, the quaternary ammonium
salts 6a–f showed better potency for AChE inhibition than the un-
charged molecules 5a–f.
The univalent tertiary compounds 3-methoxy-5,8,9,13b-tetra-
hydro-6H-isoquino-[1,2-a]isoquinoline (1) and 3-methoxy-7-meth-
yl-5,6,7,8,9,14-hexahydrodibenzo[d,g]azecine (3) did not show
activities at any ChE. The quaternary compound 2 is a micromolar
4.8 (5.317 ± 0.142)
2.0 (5.690 ± 0.106)
0.14 (6.847 ± 0.022) 34.0
0.21 (6.683 ± 0.036) 9.9
7e
a
IC50 values are means of at least three experiments.
olinium salts 6a–f, all of these compounds are submicromolar
inhibitors with inhibitory potencies from IC50(AChE) = 0.45
compound 6a with n = 3) to IC50(AChE) = 0.12 M (compound 6f
with n = 12). Despite the considerable differences in the lengths of
the alkylene spacers (from n = 3 to n = 12), the differences in poten-
cies are minor, not exceeding a factor of four. For each compound,
activities at BChE are even higher, ranging from IC50(BChE) =
lM
(
l
non-selective inhibitor of both ChEs (IC50(AChE) = 8.8
lM; IC50(B-
ChE) = 6.4 M). Interestingly, the bivalent compounds showed
l
0
.16 lM (compound 6a with n = 3) to IC50(BChE) = 0.027 lM (com-
pound 6d with n = 6).
much higher activities at both AChE and BChE. Regarding the
bivalent 7-methyl-5,8,9,13b-tetrahydro-6H-isoquino[1,2-a]isoquin-
O
NH2
H C
HO
O
O
n
3
N
N
N
b
a
4
+
0%
22 - 41%
4
5a - f
c
84 - 98%
O
O
O
O
n
O
O
n
H C
CH3
H3C
+
+ CH3
N
3
d
N
N
N
1
6 - 24%
-
-
I
I
7b,e
6a - f
a: n=3; b: n=4; c: n=5; d: n=6; e: n=8; f: n=12
Scheme 1. Synthesis of the test compounds. Reagents and conditions: (a) (i) 70 °C, 7d; (ii) POCl
reflux, 5 h; (b) (i) NaH, dry DMF, 0 °C to rt, 1 h; (ii) dibromoalkane, dry DMF, rt, 24 h; (c) methyl iodide, MeCN/MeOH, rt, 6 h; (d) Na, NH
3
, MeCN, 95 °C, 18 h; (iii) NaBH
4
/MeOH, 0 °C to rt, 1 h; (iv) HBr, glacial acid,
liq., À40 °C.
3

2
948
M. Schulze et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2946–2949
Despite the fact that the univalent tertiary compound 1 did not
quinolinium salts. Therefore docking results might be of limited
value. The increase of BChE-activity is not straightforwardly ex-
plained, either. Only a very few sets of compounds have been de-
scribed in which bivalency has led to pronounced increases in
show significant activities at either AChE or BChE, the respective
bivalent compounds 5a–f turned out to be potent inhibitors, espe-
cially at BChE. Activities at AChE generally lay in the micro- and
6
submicromolar range, with the lowest activity IC50(AChE) = 4.5
lM
BChE selectivity without modifying the spacer. In the huge major-
(
compound 5a with n = 3) and the most potent compound with an
IC50(AChE) = 0.65 lM (compound 5c with n = 5). In the contrary to
ity of cases an increased AChE selectivity resulted from bivalency.5
A related increase at BChE-activity and concomitant selectivity was
observed with hybrid molecules from ChE-inhibiting quinazolini-
the quaternary compounds 6a–f, the tertiary compounds 5a–f
showed a tether length dependent potency with pentylene and
hexylene chains as optimal spacers. This is in line with several
examples from the literature, which identified these spacer lengths
1
3
mines and lipoic acid. With these compounds kinetic measure-
ments were not able to prove an interaction with a second
distinct binding site additional to the CAS which would result in
non-competitive reversible interaction. Compound 5d was selected
for kinetic measurements at BChE, as it showed highest potencies
of all investigated compounds at BChE, high BChE selectivity, and
low cytotoxicity, while representing an uncharged molecule for
better passage of the blood–brain barrier. The mechanism of inhi-
bition was investigated by recording substrate–velocity curves
using various substrate concentrations at different concentrations
of 5d. The resulting Lineweaver–Burk plot (i.e., reciprocal velocities
vs reciprocal substrate concentrations) is shown in Figure 2:
5
,10
also as most favorable.
much higher ranging from IC50(BChE) = 0.92
with n = 12) down to IC50(BChE) = 0.014 M (compound 5d with
Most interestingly, activities at BChE are
lM (compound 5f
l
n = 6). At BChE a dependency of the activity on spacer length could
be observed: Only compounds with n = 4, 5, and 6, respectively,
showed activities from IC50(BChE) = 0.014–0.017 lM, whereas
n = 3 and also higher spacer lengths yielded compounds with 10
to 60-fold lower activities. Since the tertiary compounds 5a–f
exhibited lower activities at AChE and higher ones at BChE com-
pared to quaternary compounds 6a–f, BChE selectivity increased.
The most selective compound 5b (n = 4) showed a 76-fold selectiv-
ity towards BChE. Compounds 7b, e, also with tertiary nitrogen
atoms, but much less rigidized than compounds 5a–f, exhibited
lower activities than the other bivalent structures with the same
spacer length. In general, most bivalent compounds synthesized
showed similar or higher activities at AChE than the reference
galanthamine, and all novel compounds were superior at BChE.
Interestingly, a relationship between spacer length and biolog-
ical activity could be observed also in a cytotoxicity test on human
glia cells (MG-U87) and neuronal cells (SH-SY5Y), which was per-
formed prior to potential in vivo studies for compounds 5a–f (Ta-
ble 2). Therein cell viability was measured after substance
m
K values (i.e., the negative reciprocal of the X intercept) differ,
but Vmax values (i.e., the reciprocal of the Y intercept) do not signif-
icantly change with different inhibitor concentrations at BChE. This
indicates a reversible and competitive inhibition with the substrate
molecule, which means that the high inhibitory activity cannot be
explained by interaction of compound 5b with a second binding
site, which would lead to an altered kinetic profile.
In conclusion, we have synthesized two new kinds of bivalent
compounds, each with different spacer lengths (n = 3 up to
n = 12). Tertiary and quaternary bivalent structures showed greatly
enhanced inhibitory potencies compared to their univalent cong-
eners at both AChE and BChE. The charged quaternary ammonium
salts 6a-f revealed better inhibitory activities (two-digit nanomo-
lar IC50 values) at AChE and BChE than galanthamine, with moder-
11
administration according to previous described protocols to as-
sure innocuousness at effective concentrations in the brain. All
CC50 values (EC50 equivalent of cytotoxicity tests) are at micromo-
lar concentrations, with best ChE potency/toxicity ratio for com-
pound 5d. The univalent compound 1 showed much less effects
on cell viability than the bivalent compounds.
ate selectivity towards BChE. To potentially gain
a better
penetration of the blood–brain barrier we investigated also their
non-charged but protonable precursors 5a–f. Compounds 5c and
5d showed inhibitory activities at AChE comparable to galantha-
mine and BChE-inhibition in the low nanomolar range. For com-
pounds 5a–d a 30-fold up to 76-fold selectivity towards BChE
could be observed, what might be advantageous for the treatment
There was a statistically significant relationship between spacer
length and biological activity with the pentylene compound 5c
being the most cytotoxic one. In general though, much higher con-
centrations were necessary to observe cytotoxic effects than for
AChE and particularly BChE-inhibition.
2
,14
in the progressed forms of Alzheimer’s disease.
Inhibitory activi-
ties at cholinesterases, especially BChE, and cytotoxic effects were
spacer length dependent for these substances. The bivalent ligand
approach on 5,8,9,13b-tetrahydro-6H-isoquino[1,2-a]isoquinolines
and -isoquinolinium salts led to compounds with greatly enhanced,
submicromolar activities at both ChEs and selectivity towards BChE,
introducing a novel structural template for BChE inhibitors.
An increased activity at AChE of bivalent inhibitors could be ex-
plained by interaction of one part of the bivalent compound with
the catalytic active site (CAS), where ACh is hydrolyzed, and the
other part with the peripheral anionic binding site (PAS) at the out-
5
,12
er site of the gorge.
Due to the fact that quaternary compounds
6
a–f and the tertiary ones 5a–f do not show highly significant
Lineweaver-Burk plot BChE inhibition
correlation of inhibitory potency to spacer length at AChE, these
interactions with the PAS do not seem to be pronounced and highly
specific with tetrahydro-6H-isoqino[1,2-a]isoquinolines and -iso-
by compound 5d
7
5
5
0
[5d] = 40 nM
5d] = 20 nM
5d] = 10 nM
5d] = 8 nM
5d] = 5 nM
5d] = 0 nM
[
[
[
[
Table 2
Cell viability measured in a photometric MTT assay
Compound
CC50
(lM) glia cells
CC50 (lM) neuronal cells
[
1
774.9 ± 22.1
26.92 ± 1.93
19.51 ± 3.00
8.70 ± 2.94
20.73 ± 1.13
41.97 ± 6.86
46.63 ± 7.98
171.1 ± 41.70
8.94 ± 1.93
6.35 ± 1.30
—
7.40 ± 0.22
19.24 ± 1.85
—
25
5a
5b
5c
5d
5e
5f
-0.01
0.01
0.02
1/[S]
0.03
0.04
CC50 values are means of at least three experiments, each performed in
triplicate ± SD.
Figure 2. Lineweaver–Burk plot resulting from substrate–velocity curves at differ-
ent concentrations of compound 5d.

M. Schulze et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2946–2949
2949
3
4
.
.
Mohr, P.; Decker, M.; Enzensperger, C.; Lehmann, J. J. Med. Chem. 2006, 49,
110.
Loizzo, M. R.; Tundis, R.; Menichini, F.; Menichini, F. Curr. Med. Chem. 2008, 15,
Acknowledgment
2
The authors would like to thank Petra Wiecha for the reliable
technical assistance in performing the cholinesterase assay.
1209.
5. Du, D. M.; Carlier, P. R. Curr. Pharm. Des. 2004, 10, 3141.
6
7
.
.
Decker, M. J. Med. Chem. 2006, 49, 5411.
Peng, X.; Neumeyer, J. L. Curr. Top. Med. Chem. 2007, 7, 363.
Supplementary data
8. Guillou, C.; Mary, A.; Renko, D. Z.; Gras, E.; Thal, C. Bioorg. Med. Chem. Lett. 2000,
0, 637.
1
9
.
Savini, L.; Gaeta, G.; Fattorusso, C.; Catalanotti, B.; Campiani, G.; Chiasserini,
L.; Pellerano, C.; Novellino, E.; McKissic, D.; Saxena, A. S. J. Med. Chem.
2003, 46, 1.
Supplementary data (syntheses and spectral data of target com-
pounds, substrate–velocity curves and description of pharmacolog-
ical assays) associated with this article can be found, in the online
version, at doi:10.1016/j.bmcl.2010.03.011.
1
1
1
0. Holzgrabe, U.; Kapkova, P.; Alptuzun, V.; Scheiber, J.; Kugelmann, E. Expert
Opin. Ther. Targets 2007, 11, 161.
1. Schulze, M.; Siol, O.; Meise, W.; Lehmann, J.; Enzensperger, C. Arch. Pharm.
2
010, accepted.
2. Pang, Y. P.; Quiram, P.; Jelacic, T.; Hong, F.; Brimijoin, S. J. Biol. Chem. 1996, 271,
3646.
References and notes
2
1
1
3. Decker, M.; Kraus, B.; Heilmann, J. Bioorg. Med. Chem. 2008, 16, 4252.
4. O’Brien, K. K.; Saxby, B. K.; Ballard, C. G.; Grace, J.; Harrington, F.; Ford, G. A.;
O’Brien, J. T.; Swan, A. G.; Fairbairn, A. F.; Wesnes, K.; del Ser, T.; Edwardson, J.
A.; Morris, C. M.; McKeith, I. G. Pharmacogenetics 2003, 13, 231.
1
2
.
.
Giacobini, E.; DeSarno, P.; Clark, B.; McIlhany, M. Prog. Brain Res. 1989, 79, 335.
Greig, N. H.; Utsuki, T.; Ingram, D. K.; Wang, Y.; Pepeu, G.; Scali, C.; Yu, Q. S.;
Mamczarz, J.; Holloway, H. W.;Giordano, T.;Chen, D.; Furukawa, K.;Sambamurti,
K.; Brossi, A.; Lahiri, D. K. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 17213.