Tacrine-silibinin codrug shows neuro- and hepatoprotective effects in vitro and pro-cognitive and hepatoprotective effects in vivo

Article abstract of DOI:10.1021/jm300246n

A codrug of the anti-Alzheimer drug tacrine and the natural product silibinin was synthesized. The codrug's biological and pharmacological properties were compared to an equimolar mixture of the components. The compound showed potent acetyl- and butyrylcholinesterase inhibition. In a cellular hepatotoxicity model, analyzing the influence on viability and mitochondria of hepatic stellate cells (HSC), the toxicity of the codrug was markedly reduced in comparison to that of tacrine. Using a neuronal cell line (HT-22), a neuroprotective effect against glutamate-induced toxicity could be observed that was absent for the 1:1 mixture of components. In subsequent in vivo experiments in rats, in contrast to the effects seen after tacrine treatment, after administration of the codrug no hepatotoxicity and no induction of the cytochrome P450 system were noticed. In a scopolamine-induced cognitive impairment model using Wistar rats, the codrug was as potent as tacrine in reversing memory dysfunction. The tacrine-silibinin codrug shows high AChE and BChE inhibition, neuroprotective effects, lacks tacrine's hepatotoxicity in vitro and in vivo, and shows the same pro-cognitive effects in vivo as tacrine, being superior to the physical mixture of tacrine and silibinin in all these regards.

Full text of DOI:10.1021/jm300246n

Article
pubs.acs.org/jmc
Tacrine-Silibinin Codrug Shows Neuro- and Hepatoprotective Effects
in Vitro and Pro-Cognitive and Hepatoprotective Effects in Vivo
Xinyu Chen,† Katharina Zenger,† Amelie Lupp,‡ Beata Kling,† Jorg Heilmann,† Christian Fleck,*,‡
̈
Birgit Kraus,† and Michael Decker*,†
†Institut fur Pharmazie, Universitat Regensburg, Universitatsstraße 31, D-93053 Regensburg, Germany
̈
̈
̈
‡Institut fur Pharmakologie und Toxikologie, Universitatsklinikum Jena, Friedrich-Schiller-Universitat Jena, Drackendorfer Straße 1,
̈
̈
̈
D-07740 Jena, Germany
S
* Supporting Information
ABSTRACT: A codrug of the anti-Alzheimer drug tacrine and the natural product silibinin was synthesized. The codrug’s
biological and pharmacological properties were compared to an equimolar mixture of the components. The compound showed
potent acetyl- and butyrylcholinesterase inhibition. In a cellular hepatotoxicity model, analyzing the influence on viability and
mitochondria of hepatic stellate cells (HSC), the toxicity of the codrug was markedly reduced in comparison to that of tacrine.
Using a neuronal cell line (HT-22), a neuroprotective effect against glutamate-induced toxicity could be observed that was absent
for the 1:1 mixture of components. In subsequent in vivo experiments in rats, in contrast to the effects seen after tacrine
treatment, after administration of the codrug no hepatotoxicity and no induction of the cytochrome P450 system were noticed.
In a scopolamine-induced cognitive impairment model using Wistar rats, the codrug was as potent as tacrine in reversing memory
dysfunction. The tacrine−silibinin codrug shows high AChE and BChE inhibition, neuroprotective effects, lacks tacrine’s
hepatotoxicity in vitro and in vivo, and shows the same pro-cognitive effects in vivo as tacrine, being superior to the physical
mixture of tacrine and silibinin in all these regards.
Chart 1. Structures of the Anti-Alzheimer Drug and AChE
Inhibitor Tacrine (1) and the Hepatoprotective Natural
Product Silibinin (2)
INTRODUCTION
■
Alzheimer’s disease (AD) is the most prominent form of
dementia in the world. Despite huge efforts and numerous
successes in investigating the pathophysiology of AD, the
disease is still incurable.1 The number of approved drugs is
extremely limited to only three acetylcholinesterase (AChE)
inhibitors (the moderately active drugs rivastigmine, donepezil
and galantamine) and one NMDA antagonist (memantine).
The most potent AChE inhibitor tacrine (Chart 1), albeit
clinically effective like all AChE inhibitors,2 was withdrawn
from the market due to its dose-dependent hepatotoxicity.3
Because of the clinical effectiveness of AChE inhibitors in
general and the high potency of tacrine in particular, this
structure has been widely and successfully used in medicinal
chemistry for application in hybrid or multitarget compounds
in order to combine its potent AChE inhibition with other
pharmacological properties by covalently connecting tacrine to
other pharmacologically active structures,4 such as CB1 receptor
antagonists and an M1 agonist to name just two recent
examples.5,6
The hybrid approach was also applied by us and co-workers
to specifically target the problem of tacrine’s liver toxicity. For
this reason, we synthesized chemically stable hybrids with the
Received: February 23, 2012
Published: May 24, 2012
© 2012 American Chemical Society
5231
dx.doi.org/10.1021/jm300246n | J. Med. Chem. 2012, 55, 5231−5242

Journal of Medicinal Chemistry
Article
a
Scheme 1. Synthesis of Tacrine−Silibinin−Codrug 10
a
Reagents and conditions: (a) Et3N, THF, reflux, 16 h; (b) DIAD, Ph3P, anhydrous THF, N2, 16 h, 0 to 20 °C; (c) EtOH, 5% Pd/C, H2, 50 °C; (d)
POCl3, reflux, N2; (e) 1-heptanol, 140 °C; (f) EDCI, HOBt, anhydrous DMF, 16 h, rt.
natural product ferulic acid.7,8 These compounds were potent
AChE inhibitors and antioxidants, but in vivo improvement of
cognition assessed by scopolamine-induced cognitive impair-
ment in rats using a radial maze showed low activity.8b Another
approach used the combination of tacrine with nitric oxide
donating moieties: here, both the in vitro and in vivo properties
are promising, although the strong vasorelaxant properties of
NO-donors might cause problems in vivo.9
Driven by the positive results of combining tacrine with
other biologically active moieties to improve its overall
pharmacological properties, in this work we focused on the
natural product and flavonolignan silibinin in order to
specifically target tacrine’s hepatotoxicity. Silibinin (also
known as silybin; Chart 1) is one of the main components of
the silymarin complex, a standardized mixture obtained from
the fruits of Silybum marianum (L.) GAERTN. (syn. Carduus
marianus L., Asteraceae) commonly known as milk thistle. It
represents a diastereomeric mixture of silybin A (2R,3R,2′R,3′R)
and silybin B (2R,3R,2′S,3′S), only the structure of the latter is
used in this article for better illustration. Various pharmaco-
logical properties have been assigned to silibinin in recent years
including anti-inflammatory and anticancer activities.10 Inter-
estingly, also neuroprotective properties have been attributed to
silymarin and silibinin derivatives, which makes it an interesting
component for anti-Alzheimer drug candidates.11,12 The main
medicinal use of silymarin (since antiquity) and silibinin is
therapeutic intervention in liver diseases.13 Despite the fact that
in vitro results are generally positive, clinical studies are
controversial, probably because of the variable composition of
silymarin preparations applied leading in consequence to
different doses of the biologically active compounds.13
We aimed at synthesizing a compound consisting of a
tacrine-based AChE inhibiting part covalently connected to
silibinin. In contrast to previous work based on ferulic acid, the
compound should not contain a stable (amide) bond between
the components,7−9 but a more labile ester bond enabling the
AChE inhibiting part to penetrate the blood−brain barrier
(BBB) and therefore show central nervous system (CNS)
activity (Scheme 1), a codrug approach. Biological properties of
the codrug structure (in a hybrid structure the components are
connected in a chemically stable way14) should be compared to
an equimolar mixture of tacrine and silibinin in order to check
whether the stable connection provides additional benefits
(pharmacokinetically and kinetically) due to a chemical
connection as was recently shown for silibinin galloyl esters
with regard to antiangiogenic activity.15 The question whether a
chemical connection offers advantages over a physical mixture
of components or may actually lead to disadvantages is an open
5232
dx.doi.org/10.1021/jm300246n | J. Med. Chem. 2012, 55, 5231−5242

Journal of Medicinal Chemistry
Article
Table 1. Inhibition of AChE and BChE (IC50 and pIC50 Values SEM) by Codrug 10 and Tacrine Moiety-Containing
Intermediates and Putative Metabolites, Respectively, and Selectivity Expressed as the Ratio of Resulting IC50 Values
a
b
c
Data are the means of at least three determinations. pIC50 = −log IC50. AChE from electric eel. BChE from equine serum.
question that has to be worked out for every set of hybrid or
codrug molecules, although the number of examples with
additional benefits is overwhelming.4−6,14,16
tion). First, tacrine-spacer-succinamide was synthesized (so
connecting the tacrine part of the codrug with succinic acid in
the first step), but the amino acid-like structure made it a
poorly soluble compound even in THF and consequently led to
the failure of the Mitsunobu reaction with silibinin. Second,
silibinin (2) was partially protected by trityl groups according
to the method described in the literature.21 In the subsequent
reaction with either succinic anhydride or succinic acid, no
specific reaction at the primary hydroxyl group occurred
because of the poor reaction selectivity between aliphatic and
aromatic hydroxyl groups. The compound that formed during
the preceding protection step was actually 7,20-ditrityl silibinin
instead of 5,7-ditrityl silibinin reported in the literature as
confirmed by 1H NMR (see Supporting Information for
details).21
Cholinesterase Inhibition. The cholinesterase inhibition
of the codrug and its tacrine moiety-containing components
were evaluated using the colorimetric Ellman assay (Table 1).22
Two cholinesterases exist in the human body: acetylcholines-
terase (AChE; 3.1.1.7), the target of the approved anti-
Alzheimer drugs,2b and butyrylcholinesterase (BChE, 3.1.1.8),
the exact functions of which are not yet fully understood.23
Several findings indicate that BChE can compensate for the lack
of AChE appearing in later stages of AD. Cognitive
performance of aged rats was improved and the amount of β-
amyloid peptide was lowered by selective BChE inhibition
making its inhibition potentially therapeutically desirable.23b,24
We used AChE (EC 3.1.1.7, type VI−S, from electric eel)
and BChE (EC3.1.1.8, from equine serum) in the Ellman assay,
and selectivity was expressed as the ratio of IC50(AChE)/
IC50(BChE).
After the synthesis of a suitable codrug, several biological and
pharmacological properties were assessed in vitro: AChE
inhibition, cytotoxicity, and cytoprotection toward a neuronal
cell line, and toxicity toward a hepatic cell line (and therefore
putative resulting hepatoprotection). Additionally, the codrug,
the components, and the mixture, separately, were evaluated in
vivo in rats for hepatotoxicity or hepatoprotection and a
possible interaction with the biotransformation capacity of the
liver, as assessed by the influence on the cytochrome P450
(CYP) system. Experiments were performed at two different
dosages, the dosage used in the subsequent behavioral
pharmacology experiments and at a dosage equimolar to the
highest tolerated dose of tacrine. Furthermore, in vivo
procognitive effects of the compounds were evaluated. We
used a method applying scopolamine and an eight-arm radial
maze as an accepted protocol in behavioral pharmacology for
assessing the antidementive potential of compounds acting
directly or indirectly at the muscarinic acetylcholine receptor,
such as AChE inhibitors or allosteric modulators.6,8b,17
RESULTS AND DISCUSSION
■
Chemistry. The tacrine−silibinin codrug 10 is the
combination of two parts: silibinin hemisuccinate (6) and 6-
aminohexamethylene tacrine (N1-(1,2,3,4-tetrahydroacridin-9-
yl)hexane-1,6-diamine 9). The latter tacrine-spacer compound
(9) was synthesized by alkylation of 1,6-hexanediamine (8)
with 9-chlorotetrahydroacridine (7).7,18,19 To generate the
selective esterification of succinic acid (4) with the only primary
hydroxyl group of silibinin (2), a Mitsunobu reaction between
silibinin (2) and monobenzyl succinate (5) was applied.20 The
latter ester was synthesized using succinic acid (4) and benzyl
bromide (3) (Scheme 1). The chemical shift in 1H NMR of the
hydrogen atoms at position 23 changes upon esterification
according to what had been described in the literature.20
Two alternative synthetic routes were also investigated to
access the codrug structure, yet all failed due to the poor
solubility of both silibinin and tacrine derivatives (see
Supporting Information for detailed procedures and descrip-
As expected, silibinin (2) is not able to inhibit significantly
either AChE or BChE even at a concentration of 100 μM. The
codrug (10) loses quite a bit of inhibitory activity at BChE (16-
fold lower) and moderately at AChE (3.5-fold lower).
Nevertheless, the codrug is still a two-digit nanomolar inhibitor
leading to high inhibition concomitantly at both ChEs (Table
1). The two tacrine moiety-containing spacers were also tested.
With respect to the tacrine−hexamethylene amine (9),
analogues with lower and higher spacer lengths had been
synthesized and tested before and represent one-digit nano-
5233
dx.doi.org/10.1021/jm300246n | J. Med. Chem. 2012, 55, 5231−5242

Journal of Medicinal Chemistry
Article
molar inhibitors at AChE and at BChE.7,9 The primary amine
compound 9 shows an IC50 value of 39 nM at AChE and 2.0
nM at BChE, therefore acting at the same range as the parent
compound tacrine (1), and these values are in accordance with
the inhibition data of analogues with similar spacer lengths
(Table 1).7,9 The tacrine hexamethylene amine (9) connected
to succinic acid as a hemi amide (compound 11) shows much
weaker inhibition values with IC50(AChE) = 102 nM and
IC50(BChE) = 361 nM (Table 1). For this compound, some
minor selectivity toward AChE is achieved, but in general, for
all compounds tested selectivity differences are moderate.
Compound 11 is probably the initial hydrolysis product in vivo
through cleavage of the ester bond to silibinin. Despite the
inhibitory activity decrease, also this compound might at least
contribute to the codrug’s in vivo profile. All tacrine derivatives
investigated for their enzyme inhibition show at least similar
activity at AChE (ranging from 16 to 102 nM); therefore, no
indication can be given for the structure of the compound being
procognitively active in vivo (see Behavioral Studies).
can effectively prevent oxidative neuronal death in this cell
line.31
Figure 1A shows the influence of different doses of codrug 10
on the viability of hippocampal HT-22 cells measured by a
In summary, the codrug itself and its putative cleaved parts
(metabolites) are potent AChE and BChE inhibitors in the
submicromolar to the nanomolar range.
Neuroprotection. It is well established that oxidative stress,
the formation of reactive oxygen species (ROS), and
subsequent neurotoxicity are key processes in the pathophysi-
ology of AD.25 Silymarin, silibinin, and chemical derivatives
have been described to possess neuroprotective properties.11,12
The results obtained are not unambiguous though. Silymarin
increases glutathione (GSH) level and enhances superoxide
dismutase activity.12 However, it turned out that derivatives
(oxidized forms, alkenylated and amidated ones) are more
active than the parental silibinin in terms of radical scavenging
properties.11 Unambiguous structure−activity relationships for
silibinin derivatives are difficult to describe, but it seems that
lipophilicity increased the radical scavenging properties as well
as the cell viability of PC-12 cells (as a model for neuronal cell
death) measured in a MTT test, albeit the flavonol quercetin
used as a positive control was always more potent, especially
with regard to the parental silibinin.11a We therefore wanted to
investigate whether and to what extent tacrine, silibinin,
silibinin physically mixed with tacrine (equimolar mixture),
and the synthesized codrug (10) exhibit a cytotoxic or
especially cytoprotective potential in a neuronal cell line and
quantify putative differences.
The HT-22 mouse hippocampal cell line represents an
established in vitro model system for studying oxidative stress
induced neurotoxicity caused by glutamate.26 Two pathways of
glutamate toxicity have been described: an excitotoxic one
mediated by the ionotropic glutamate receptors via cell death
through Ca2+ influx and ROS accumulations and a nonreceptor
mediated “oxidative” pathway.27−29 The latter oxidative
glutamate toxicity is initiated by extracellular glutamate in
high concentrations that inhibits cystine (as oxidized form of
cysteine) uptake mediated by the cystine/glutamate antiporter
system leading to intracellular cysteine and therefore
glutathione depletion, which induces ROS accumulation and
cell injuries.28 HT-22 hippocampal cells lack these ionotropic
glutamate receptors, and therefore, the glutamate-mediated
excitotoxicity cell death pathway can be excluded.29 In this
neuronal oxidative stress induced toxicity (oxytosis), extra-
cellular glutamate in high concentrations decreases intracellular
glutathione and therefore leads to ROS production.30
Administration of antioxidants such as vitamin E or flavonoids
Figure 1. Evaluation of the neurotoxicity (A) and neuroprotection (B)
of codrug 10 and an equimolar mixture of tacrine (1) and silibinin (2)
against glutamate induced oxidative stress on HT-22 cells. Data were
subjected to one-way ANOVA followed by Dunnett’s multiple
comparison post-test using GraphPad Prism 4 Software (levels of
significance *p < 0.05; **p < 0.01; ***p < 0.001).
MTT test16 in comparison to the equimolar mixture of tacrine
and silibinin. The tacrine/silibinin mixture showed a significant,
albeit very moderate, neurotoxicity starting from a concen-
tration of 25 μM. Both tacrine and silibinin showed very weak
neurotoxicity starting from a concentration of 50 μM (cf.
Figure A in the Supporting Information). The mixture does not
show lower neurotoxicity compared to the individual
components; in contrast, the toxicities showed an additive
effect. For the codrug, even at the highest concentration tested
(50 μM), no neurotoxicity was observed (Figure 1A).
Figure 1B shows the effect of the tacrine/silibinin mixture
and the codrug on the reversal of glutamate-induced neuro-
toxicity at different concentrations using quercetin as the
(potent) positive control. All concentrations of the mixture
tested did not show any effect even at a concentration of 50
μM, the individual components did not show a neuroprotective
effect either (cf. Figure A in the Supporting Information).
Albeit intensively discussed in the literature, we could not
observe a neuroprotective effect of silibinin in this assay.11,12
A
weak, but similarly low effect was observed for silibinin on PC-
12 cells.11a In contrast to these data, the codrug showed a
significant increase in cell viability after exposure to glutamate
(Figure 1B). Although glutamate’s neurotoxic effect could not
be reversed to the extent reached by 25 μM Quercetin, already
5234
dx.doi.org/10.1021/jm300246n | J. Med. Chem. 2012, 55, 5231−5242

Journal of Medicinal Chemistry
Article
at a concentration of 10 μM of the codrug a significant increase
in cell viability was mediated (Figure 1B). Interestingly, this
value seems to represent the highest achievable increase in cell
viability since further increase of the concentration up to 50 μM
did not increase cell viability (Figure 1B). The mechanism of
neuroprotection seems to differ from the one of quercetin, or
alternatively, quercetin is able to activate several mechanisms of
antioxidant and neuroprotective activity.31
The codrug does not show the neurotoxic effects determined
by the unchanged viability of hippocampal HT-22 cells and
shows at low concentrations a significant neuroprotective effect
(albeit lower than the one of the flavonol quercetin) that could
not be observed for the physical mixture of tacrine and silibinin.
In Vitro Hepatoprotection. To investigate a possible
hepatoprotective effect of the codrug and the tacrine/silibinin
mixture compared to tacrine, the compounds’ effect on human
hepatic stellate cells (HSC) was investigated.32 First, the
influence of test compounds on the viability of HSC was
determined by a crystal violet assay. Crystal violet is a basic dye
and is used to stain cell nuclei. The photometrically measured
intensity of the dye directly correlates with the number of
cells.33 HSC were incubated with different concentrations of
the respective test compounds and stained with crystal violet,
and the absorbance was measured photometrically after
treatment. Whereas tacrine and the equimolar mixture had a
significant impact on the cell number starting at a concentration
of 75 μM, the codrug showed no significant toxicity up to 200
μM (cf. Figure B in the Supporting Information).
the codrug. At concentrations where tacrine is toxic, also the
mixture exhibits toxicity. Therefore, the codrug is greatly
superior to the equimolar mixture in being nontoxic even at the
highest concentration tested.
Additionally, we investigated the impact of the compounds
on the mitochondria of HSC, as mitochondrial dysfunction is
an important mechanism of hepatotoxicity34 and is discussed to
be involved in tacrine-induced liver toxicity.35 One such
mechanism is the onset of mitochondrial permeability
transition caused by opening of permeability transition pores
in the inner mitochondrial membrane. This pore opening
causes mitochondrial depolarization, uncoupling, and large
amplitude swelling and can lead to both necrotic and apoptotic
cell death.36,37 For mitochondrial analysis, mitochondria were
stained after the treatment of cells with different concentrations
of tacrine, tacrine/silibinin, and codrug with a mitochondria
specific dye that followed the determination of mitochondrial
fluorescence intensity (Figure 3 and Supporting Information,
Figure C).
For a more precise analysis, we applied fluorescence
microscopy and examined the influence on cell number and
mitochondria. Microscopical data confirmed the results of the
crystal violet assay, being even more sensitive. As can be seen in
Figure 2 (cf. also Figure C in the Supporting Information),
Figure 3. Fluorescence intensity of mitochondria after treatment of
HSC with different concentrations of tacrine, an equimolar mixture of
tacrine and silibinin, and codrug (10). Data were subjected to one-way
ANOVA followed by Tukey’s multiple comparison post-test using
GraphPad Prism 4 Software (levels of significance *p < 0.05; **p <
0.01; ***p < 0.001).
Mitochondrial fluorescence intensity dramatically increases
after the treatment of cells with tacrine at 50 μM to a value of
220% of the control (Figure 3). This increase in fluorescence
intensity indicates a swelling of mitochondria and therefore
mitochondrial dysfunction, which can lead to necrotic or
apoptotic cell death. This effect is significantly lower for the
tacrine/silibinin mixture at 50 and 75 μM. The codrug shows
the same effect as the mixture at these concentrations. In terms
of the hepatotoxicity indicated by increased mitochondrial
fluorescence intensity in HSC, both the physical mixture and
the codrug show greatly decreased toxicity compared to tacrine.
Regarding the effect of the codrug at 25 μM, both tacrine and
the tacrine/silibinin mixture show significantly higher fluo-
rescence intensity, again indicating a superior effect of the
codrug over the physical mixture of compounds (Figure 3).
Hepatotoxicity of tacrine and hepatoprotective activity of
silibinin have been demonstrated before on HepG2 cells.38 Our
studies revealed also pronounced cyto- and mitotoxicity of
tacrine on HSC. The mitotoxic effect is significantly lower for
both the physical mixture of tacrine/silibinin and the codrug
(Figure 3). Already at lower concentrations, the codrug’s
cytoprotective effect is more pronounced than the one of the
Figure 2. Comparison of cell numbers determined by microscopy after
treatment of HSC with different concentrations of tacrine, an
equimolar mixture of tacrine and silibinin, and codrug (10). Data
were subjected to one-way ANOVA followed by Tukey’s multiple
comparison post-test using GraphPad Prism 4 Software (levels of
significance *p < 0.05; **p < 0.01; ***p < 0.001).
tacrine decreases the amount of cells already starting at a
concentration of 50 μM with higher concentrations being more
hepatotoxic.
An interesting observation can be made when the influence
of the codrug is compared with the tacrine/silibinin mixture
(Figure 2): Statistically significant differences can already be
observed from a concentration of only 25 μM. The mixture
shows lower cell viability and therefore higher cytotoxicity than
5235
dx.doi.org/10.1021/jm300246n | J. Med. Chem. 2012, 55, 5231−5242

Journal of Medicinal Chemistry
Article
Figure 4. Histomorphological appearance of the liver of a male control rat (A) and of male rats 24 h after treatment with 6 μmol/100 g body weight
tacrine (B), tacrine plus silibinin (C) or codrug (10) (D), respectively. H&E; original magnification: ×200. V, central vein; P, portal field.
mixture (Figure 2). Here, the mixture was only moderately
superior to tacrine’s toxicity, whereas the codrug did not show
any toxicity even at the highest concentration tested.
tacrine plus silibinin, or codrug treatment. This effect was
dosage dependent and amounted to 10% or 15%, respectively.
However, the tissue content of oxidized glutathione was
elevated after the administration of tacrine or of tacrine plus
silibinin by 10% at the lower dosage (2 μmol/100 g b. wt.) and
by 15% at the higher dosage (6 μmol/100 g b. wt.),
respectively, but not following silibinin or codrug treatment.
Consequently, the ratio of reduced/oxidized glutathione was
significantly diminished after tacrine administration but
increased after codrug administration and at the higher dosage
also after silibinin treatment (Figure 5).
At the lower dosage in female rat treatment with silibinin,
tacrine plus silibinin, or codrug, a reduction in tissue content of
lipid peroxidation products of about 15% to 25% was observed.
After the higher dosage in male rats, a remarkable increase in
the values by about 140% was caused by tacrine treatment.
After simultaneous application of tacrine plus silibinin, this
effect was less pronounced. In contrast to these findings, no
increase in tissue content of lipid peroxidation products was
observed after silibinin or codrug treatment (Figure 6).
These results are in line with the histopathological findings
and point to an increased oxidative stress after administration of
tacrine that could partially be reversed by the simultaneous
treatment with silibinin. In line with the results from the in vitro
experiments and the histopathological findings, antioxidative
and hepatoprotective effects were observed not only with
silibinin but (at least to a similar extent) also with the codrug.
Behavioral Studies. In order to determine how the in vitro
cholinesterase inhibition translates into behavioral effects, we
applied a cognition assay using rats in an eight-arm radial
maze.6,8b,9b,43a In this model, deteriorated information acquis-
ition resulting in amnesia is reversibly induced by admin-
istration of the muscarinic acetylcholine receptor (M)
antagonist scopolamine.43b This effect can be reversed
In Vivo Studies on Hepatotoxicity/Hepatoprotection.
First, investigations were carried out on female rats and at a
dosage of the compounds of 2 μmol/100 g body weight, i.e., on
the same gender and at the same dosage used in the subsequent
behavioral pharmacology experiments. Second, experiments
were performed at a dosage equimolar to the highest tolerated
dose of tacrine, 6 μmol/100 g body weight, in male rats,
because of the higher cytochrome P450 content and activity in
the livers of male than of female rats [and thus a higher
biotransformation of tacrine (and probably also of the other
compounds) into (reactive) metabolites].
Histopathological Changes. In a histopathological exper-
imental setting in which doses of 2 μmol/100 g body weight
corresponding to a therapeutic dosage of tacrine and 6 μmol/
100 g body weight corresponding to the highest tolerated
dosage of tacrine were tested, livers of control and of silibinin
treated rats (results not shown) displayed normal histomor-
phology (Figure 4A). After tacrine administration, however,
especially after the higher dosage in male rats, liver cell
apoptosis and distinct fatty degeneration of the hepatocytes
were observed (Figure 4B). This hepatotoxic effect of tacrine is
well-known from the literature both from animal experiments
as well as from clinical experience in humans.3,39,40 As expected
from literature data on the hepatoprotective effects of silymarin
and silibinin,41,42 simultaneous administration of silibinin
prevented the toxic effects of tacrine (Figure 4C). In contrast
to tacrine, after treatment of the rats with the codrug at both
dosages, no adverse histomorphological changes were observed
(Figure 4D).
Liver Tissue Oxidative State. The tissue content of reduced
glutathione was significantly enhanced both after silibinin,
5236
dx.doi.org/10.1021/jm300246n | J. Med. Chem. 2012, 55, 5231−5242

Journal of Medicinal Chemistry
Article
Figure 5. Ratio of the tissue content on reduced glutathione (GSH)
and oxidized glutathione (GSSG) in the livers of female control rats
and of female rats 24 h after treatment with 2 μmol/100 g body weight
(A; therapeutic dosage) and in the livers of male control rats and of
male rats 24 h after treatment with 6 μmol/100 g body weight (B;
highest tolerated dosage of tacrine) silibinin, tacrine (1), tacrine/
silibinin, or codrug (10), respectively. For statistical analysis, the
Mann−Whitney test (p ≤ 0.05) was applied. Arithmetic means
SEM, n = 5−8; *, p ≤ 0.05 vs control; +, p ≤ 0.05 vs tacrine.
Figure 6. Tissue content on lipid peroxidation products as determined
by thiobarbituric acid reactive substances (TBARS) in the livers of
female control rats and of female rats 24 h after treatment with 2
μmol/100 g body weight (A; therapeutic dosage) and in the livers of
male control rats and of male rats 24 h after treatment with 6 μmol/
100 g body weight (B; highest tolerated dose of tacrine) silibinin,
tacrine, tacrine plus silibinin, or codrug (10), respectively. For
statistical analysis, the Mann−Whitney test (p ≤ 0.05) was applied.
Arithmetic means SEM, n = 5−8; *, p ≤ 0.05 vs control; +, p ≤ 0.05
vs tacrine.
effectively by the administration of M1 (ortho- and allosteric)
agonists or AChE inhibitors, such as tacrine.9b,17 In our setting,
various behavioral parameters can be investigated reflecting
cognitive deficits and their compound-induced attenuation,
such as the time to find food pellets or the number of pellets
not found as well as the number of errors made by the rats
while investigating the radial maze.6,9b,43
Figure 7 shows the effect of tacrine, silibinin, the equimolar
mixture, and the codrug on scopolamine-induced amnesia
measured as the number of food pellets not found plus the
number of errors made in an eight-arm radial maze.
Scopolamine induces cognitive deficits after 20 min, effects
that fade after 2 h due to metabolism and excretion of
scopolamine. This well-described effect is not altered by
silibinin, but coadministration of tacrine significantly reduces
the number of errors made which corresponds to an
improvement of working memory. As expected, the same
effect is observed for the equimolar mixture of tacrine and
silibinin. The codrug shows the same pro-cognitive effect as
tacrine (Figure 7), a statistically highly significant effect since 18
experiments were performed with 5 rats for each experiment.
Therefore, despite slightly lower in vitro AChE and BChE
inhibition and unknown metabolism of the codrug, the
compound’s in vivo efficacy is as high as that for tacrine.
Figure 7. Influence of tacrine, silibinin, equimolar mixture of tacrine
and silibinin, and codrug 10 (each 2 μmol in 1 mL/100 g b. wt.) on
scopolamine (0.05 mg/100 g b. wt.) induced impairment of working
memory in adult rats measured in an eight-arm radial maze. Arithmetic
means
SEM, n = 18 per group. Maximal run-time: 10 min.
Parameter: errors made plus pellets not found after 10 min. Student's t
test was used to assess significant differences (level of significance from
scopolamine *p > 0.05).
CONCLUSIONS
A codrug of the natural product silibinin and the AChE
inhibitor tacrine was synthesized, and its biological and
■
5237
dx.doi.org/10.1021/jm300246n | J. Med. Chem. 2012, 55, 5231−5242

Journal of Medicinal Chemistry
Article
pharmacological properties were investigated in vitro and in vivo
and compared to an equimolar mixture of these two
compounds. The codrug is a slightly less potent AChE and
BChE inhibitor than tacrine. It did not show neurotoxic effects
on a hippocampal cell line but exhibited a (saturable)
neuroprotective effect against glutamate induced oxidative
stress which could not be observed for the physical mixture.
Since tacrine is a known hepatotoxic agent (the reason for
withdrawal from the market), and silibinin was described as a
hepatoprotective agent, the influence of both the codrug 10 and
the physical mixture was compared to tacrine’s effects both in
vitro and in vivo. In hepatic stellate cells (HSC), the mixture did
not exhibit a significant improvement over tacrine with respect
to cell number; the codrug though was not cytotoxic even at
high concentrations. Increase of mitochondrial fluorescence
intensity, which was induced by tacrine, was used as an
indicator of mitochondrial dysfunction and therefore hepato-
toxic effect. Both the mixture and the codrug caused a
significantly lower increase in mitochondrial fluorescence
intensity than tacrine. Investigating the effects of tacrine,
silibinin, the physical mixture of both, and the codrug on rats in
vivo verified and enlarged the data obtained in vitro:
Histopathologically, tacrine’s toxicity was less pronounced for
the mixture and completely absent for the codrug. This finding
was further fostered by determination of the ratio GSH/GSSH,
which was lowered by tacrine, stayed the same for the mixture,
but was even increased for the codrug. Furthermore, the
amount of lipid peroxidation products formed after tacrine
administration was lowered at high tacrine dosage with
coadministration of silibinin; again the codrug showed even
superior effects. Co-administration of tacrine and silibinin
slightly diminished the induction of CYP isoforms observed for
tacrine, but the codrug exhibited no influence on the CYP
enzymes (cf. Supporting Information for detailed presentation
of results).
EXPERIMENTAL SECTION
■
Chemistry. General Methods for Synthesis. Melting points are
uncorrected and were measured in open capillary tubes, using a
Barnstead Electrothermal IA9100 melting point apparatus. 1H and 13
C
NMR spectral data were obtained from a Bruker Advance
spectrometer (300 and 75 MHz, respectively). TLC was performed
on silica gel on aluminum foils with a 254 nm fluorescent indicator
(Fluka) or aluminum oxide on TLC-PET foils with a 254 nm
fluorescent indicator (Fluka). For detection, iodine vapor or UV light
(254 nm) were used. ESI-MS samples were analyzed using
electrospray ionization ion-trap mass spectrometry in nanospray
mode using a Thermo Finnigan LCQ Deca. The CHN analyses
were undertaken using Perkin-Elmer Elemental Analyzer
PE2400CHNS. For column chromatography, silica gel 60, 230−400
mesh (Merck) was used. In addition to mass spectrometry and NMR,
purity was evaluated by high resolution mass and the following HPLC
system (confirming purity ≥95%).
System 1. Analytical HPLC using a VWR HITACHI L-2130 pump
coupled to a VWR HITACHI column oven L-2350, and L-2455 diode
array detector. The solvents were as follows: (A) water + 0.05%
trifluoroacetic acid and (B) acetonitrile + 0.05% trifluoroacetic acid;
flow, 0.4 mL/min. Column Hibra 125-4 Purospher STAR RP-18e (3
μm) at 20 °C, detecting at 247 nm; solvent A from 100 to 60% for 25
min, then 60% for 10 min, and 100% for 5 min.
System 2. Analytical HPLC using a VWR HITACHI L-2130 pump
coupled to a VWR HITACHI column oven L-2350, and L-2455 diode
array detector. The solvents were as follows: (A) water + 0.1% formic
acid and (B) 95% acetonitrile + 5% water + 0.1% formic acid; flow, 0.4
mL/min. Column Hibra 125-4 Purospher STAR RP-18e (3 μm) at 40
°C, detecting at 287 nm; solvent A from 80 to 0% for 30 min, then 0%
for 10 min, and 80% for 5 min.
N1-(1,2,3,4-Tetrahydroacridin-9-yl)hexane-1,6-diamine (9). The
mixture of 9-chloroacridine 77,18,19 (0.218 g, 1 mmol), hexamethyle-
nediamine (0.23 g, 2 mmol), and catalytic amount of sodium iodide in
1-pentanol (10 mL) was warmed to reflux in a 160 °C oil bath for 24
h. Then another portion of hexamethylenediamine (0.23 g, 2 mmol)
was charged, and the warming was continued for another 24 h during
which time the reaction mixture was protected by nitrogen. After the
reaction, the solvent was removed in vacuum, and then the crude
product was purified by column chromatography, using chloroform/
methanol/ammonia = 100:15:2 as the eluent system. The final product
was yielded as a brown oil (0.22 g, 73%). ESI-MS: 298.4 m/z [M +
In an in vivo-model investigating scopolamine-induced
deficits of working memory in rats, their memory was improved
by administration of both tacrine (1) and the codrug (10) to
the same extent.
1
H]+. H NMR (CDCl3, 300 MHz) δ: 1.18−1.62 (m, 8H), 1.82−1.88
(m, 4H, CH2(CH2)2CH2 (tacrine)), 2.59−2.64 (m, 4H,
CH2(CH2)2CH2 (tacrine)), 2.99−3.00 (m, 2H, NH(CH2)5CH2NH2),
3.41−3.44 (m, 2H, NHCH2(CH2)5NH2), 3.92 (br, 1H, NH), 7.26−
7.29 (m, 1H, arom), 7.46−7.50 (m, 1H, arom), 7.84−7.90 (dd, 2H, J =
8.50 Hz, arom) ppm. 13C NMR (CDCl3, 75 MHz) δ: 23.20, 23.42,
25.21, 27.10, 27.26, 32.20, 34.00, 34.38 (8C, NHCH2(CH2)4CH2NH2,
CH2(CH2)2CH2 (tacrine)), 42.48 (NH2CH2(CH2)5NH), 49.90
(NHCH2(CH2)5NH2), 116.24, 120.58, 123.27, 124.05, 128.78,
129.03, 147.74, 151.27, 158.76 ppm. HR-MS: C19H29N3 Calcd.:
298.2278 m/z [M + H]+. Anal.: 298.2276 m/z [M + H]+. HPLC
purity (system 1), 96.53%; retention time, 21.65 min.
Two important issues have not been addressed yet: the oral
bioavailabilty of the codrug, which might be lower than the one
of the components’ mixture, and the investigation about which
compounds/metabolites are actually responsible for hepato-
protection and the pro-cognitive effects and which time scale
describes their formation. Quite an intensive investigation into
stability especially in plasma is necessary for that, and with
regard to oral bioavailabilty, further modifications of the
chemical structure or pharmaceutical formulation might be
necessary, which has not been the scope of this work.
Nevertheless, this work has shown that a codrug represents a
chemical entity in its own regard with superior pharmacological
properties compared to the mixture of components.
Taken together, the silibinin−tacrine codrug exhibits
pronounced pro-cognitive effects just like tacrine, but it lacks
tacrine’s therapy-limiting hepatotoxic effects completely both in
vitro and in vivo, and additionally shows neuroprotective
properties. This superior behavior could only be observed for
the stable chemical connection within a codrug but was either
less pronounced or absent for the physical equimolar mixture of
tacrine and silibinin.
4-(Benzyloxy)-4-oxobutanoic Acid (5). A mixture of succinic acid 4
(8 g, 0.068 mol), benzyl bromide 3 (8.92 mL, 0.074 mol), and
triethylamine (10.4 mL, 0.074 mol) in tetrahydrofuran (20 mL) was
refluxed for 5 h. Ethyl acetate was added to the mixture and extracted
with 15% sodium carbonate aqueous solution (3 × 30 mL). The
combined aqueous phases were washed with diethyl ether (30 mL),
acidified with 4 N HCl to pH 1−2, and extracted with ethyl acetate (4
× 30 mL). The organic phases were combined, washed with brine (50
mL), dried over anhydrous sodium sulfate, and concentrated under
vacuum. Compound 5 was yielded as a white solid (9.3 g, 66%). M.P.:
56−58 °C. ESI-MS: 207.1 m/z [M − H]−. H NMR (CDCl3, 300
MHz) δ: 2.71 (s, 4H, CO(CH2)2CO2H), 5.18 (s, 2H, PhCH2O), 7.38
(s, 5H, arom) ppm. 13C NMR (CDCl3, 75 MHz) δ: 28.91, 28.95
(CO(CH2)2CO2H), 66.69 (PhCH2O), 127.71, 128.24, 128.33, 128.44,
1
5238
dx.doi.org/10.1021/jm300246n | J. Med. Chem. 2012, 55, 5231−5242

Journal of Medicinal Chemistry
Article
128.61 (arom CH), 140.64 (arom CCH2 ), 172.07
(CH2O2C(CH2)2CO2H), 178.26 (CO2H) ppm.
128.79, 128.94, 131.01, 136.36, 143.27, 143.42, 147.56, 147.59, 151.09,
151.90, 154.76, 159.89, 166.67, 176.26, 176.43 (2C,
O2CCH2CH2CONH), 199.81(C(4)) ppm. HR-MS: C48H51N3O12
Calcd.: 862.3546 m/z [M + H]+ Anal.: 862.3547 m/z [M + H]+.
HPLC purity (system 2), 96.06%; retention time, 15.46 min.
4-(((2S,3S)-3-(4-Hydroxy-3-methoxyphenyl)-6-((2R,3R)-3,5,7-tri-
hydroxy-4-oxochroman-2-yl)-2,3-dihydrobenzo[b][1,4]dioxin-2-yl)-
methoxy)-4-oxobutanoic Acid (6). To the precooled solution of
silibinin 2 (1 g, 2 mmol), compound 5 (0.8 g, 4 mmol), and
triphenylphosphine (1.4 g, 5.2 mmol) in dry tetrahydrofuran (10 mL)
was added dropwise the solution of diisopropyl azodicarboxylate (1.02
mL, 5.2 mmol) in dry tetrahydrofuran (10 mL) in an ice bath. The
mixture was then allowed to warm to room temperature gradually after
the addition, and it was stirred overnight at room temperature. The
mixture was concentrated after the reaction. The residue was dissolved
in ethanol (20 mL), and Pd/C (5%, 0.1 g) was added. The reaction
mixture was placed under hydrogen and stirred vigorously at 50 °C for
4 h. The catalyst was filtered off, and the filtrate was concentrated
under vacuum. The residue was purified via column chromatography
by using dichloromethane/methanol 7:1 as the eluent system.
Compound 6 was yielded as a colorless oil (500 mg, 40% over two
EXPERIMENTAL PROCEDURES FOR THE
PHARMACOLOGICAL INVESTIGATIONS
■
Acetyl- and Butyrylcholinesterase Inhibition Assay. AChE
(E.C.3.1.1.7, Type VI−S, from electric eel) and BChE (E.C.3.1.1.8,
from equine serum) were used, and the assay was performed as
previously described (cf. Supporting Information).6,16
Neurotoxicity and Protection Assay. Cell viability was
determined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium
bromide (MTT) assay according to previously described proce-
dures.16,44,45 Briefly, cells were seeded in 96-well plates at a density of
5 × 103 per well and cultured for 24 h. Subsequently, cells were
incubated for another 24 h either with medium, compounds, or solvent
only in the absence (neurotoxicity assay) or presence (neuro-
protection assay) of 5 mM glutamate (monosodium-L-glutamate,
Merck, Darmstadt, Germany). Quercetin (Sigma, Steinheim, Ger-
many) in a concentration of 25 μM served as the positive control in
the neuroprotection assay. All compounds were dissolved in DMSO
and diluted with fresh medium. DMSO concentration in final dilutions
was ≤0.1%. MTT (Sigma, Steinheim, Germany) solution (4 mg/mL in
PBS) was diluted 1:10 with the medium, and the mixture was added to
the wells after the removal of previous medium. The plates were then
incubated for further 3 h. Afterward, supernatants were removed, and
100 μL of sodium dodecyl sulfate (10%) was added to the wells.
Absorbance at 560 nm was determined on the next day with a
multiwell plate reader (Tecan, Crailsheim, Germany). Results of cell
viability are expressed as the percentage to untreated control cells.
Determination of Hepatotoxicity by Crystal Violet Assay.
Cells were seeded in 96-well plates at a density of 5 × 103 per well and
cultured for 24 h. Then cells were treated with medium, medium with
solvent (0.1% DMSO), or with test compounds, namely, silibinin
(PhytoLab, Vestenbergsgreuth, Germany), tacrine, an equimolar
mixture of silibinin and tacrine, respectively, and the codrug in
different concentrations (1−200 μM) for another 24 h. Subsequently,
the medium was carefully removed, and the cells were incubated with
50 μL of crystal violet solution (0.5% in 20% methanol) per well for 10
min at room temperature. Staining solution was aspirated, and cells
were washed three times with ultra pure water. After drying plates
overnight, 100 μL of sodium citrate buffer (EtOH + 0.1 M sodium
citrate (1:1, v/v)) was added to each well and absorbance of the
solution was determined at 560 nm using a multiwell plate reader
(Tecan, Crailsheim, Germany).
1
steps). ESI-MS: 581.1 m/z [M − H]−. H NMR (MeOD-d4, 300
MHz) δ: 2.53−2.56 (m, 4H, COCH2CH2CO2H), 3.82 (s, 3H, OCH3),
3.91−3.95 (m, 1H, C(10)H), 4.17−4.24 (m, 2H, C(23)H2O), 4.40−
4.44 (d, 1H, C(2)H, J = 11.61 Hz), 4.77−4.81 (m, 1H, C(11)H),
4.88−4.92 (d, 1H, C(3)H, J = 11.60 Hz), 5.88−5.92 (m, 2H, C(6)H,
C(8)H), 6.82−7.08 (m, 6H, arom) ppm. 13C NMR (MeOD-d4, 75
MHz) δ: 30.27, 30.43 (CH2O2C(CH2)2CO2H), 56.73 (OCH3), 64.33
(H2C(23)), 73.74 (C(3)), 77.23 (C(10)), 77.86 (C(11)), 84.53
(C(2)), 96.70 (C(8)), 97.65 (C(6)), 101.92 (C(4a)), 112.08 (C(13)),
116.57 (C(16)), 117.90 (C(18)), 118.08 (C(21)), 121.86 (C(15)),
122.49 (C(22)), 128.83 (C(14)), 131.75 (C(17)), 145.05 (C(19),
C(12a)), 148.46 (C(20)), 149.29 (C(16a)), 164.40 (C(8a)), 165.26
(C(5)), 168.78 (C(7)), 174.08 (CO2CH2), 177.05 (CO2H), 198.24
(O=C(4)) ppm.
((2S,3S)-3-(4-Hydroxy-3-methoxyphenyl)-6-((2R,3R)-3,5,7-trihy-
droxy-4-oxochroman-2-yl)-2,3-dihydrobenzo[b][1,4]dioxin-2-yl)-
methyl 4-oxo-4-(6-(1,2,3,4-tetrahydroacridin-9-ylamino)-
hexylamino)butanoate Hydrochloride (Codrug 10). The solution
of silibinin hemisuccinate 6 (400 mg, 0.69 mmol) in the mixture of
N,N-dimethylfomamide (10 mL) and toluene (20 mL) was
concentrated under vacuum until all toluene was removed. Then, N-
(3-dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (140
mg, 0.72 mmol) and 1-hydroxybenzotriazole (50 mg) were added to
the above solution. The resulting solution was stirred at room
temperature for 0.5 h. In the mean time, a solution of 9 (205 mg, 0.69
mmol) in a mixture of N,N-dimethylformamide (10 mL) and toluene
(20 mL) was also concentrated under vacuum until all toluene was
removed. The resulting solution was added to the above solution
dropwise at room temperature. The reaction solution was stirred
overnight under nitrogen. Afterward, the solvent was removed under
vacuum. The residue was dissolved in chloroform/methanol 6:1 (10
mL) and was acidified with HCl/iPrOH (5 M, 2 mL). The resulting
solution was concentrated. The residue was purified via column
chromatography using chloroform/methanol 6:1 as the eluent system.
The target compound 10 was yielded as yellow solid (350 mg, 59%).
M.P.: 135−138 °C. ESI-MS: 862.4 m/z [M + H]+. 1H NMR (CDCl3,
300 MHz) δ: 1.26−1.91 (m, 12H, CONHCH2(CH2)4CH2NH,
CH2(CH2)2CH2 (tacrine)), 2.42−2.47 (t, 2H, CH2(CH2)2CH2CN
(tacrine), J = 6.38 Hz), 2.58−2.65 (m, 4H, CH2O2C(CH2)2CONH),
2.85 (s, 1H, C(3)OH), 2.95−2.97 (m, 2H, CH2(CH2)2CH2CN
(tacrine), 3.07−3.12 (t, 2H, CONHCH2(CH2)5NH, J = 7.00 Hz),
3.50−3.66 (m, 2H, CONH(CH2)5CH2NH), 3.85 (s, 3H, OCH3),
3.92−3.97 (m, 1H, C(10)H), 4.20−4.27 (m, 2H, C(23)H2O2C),
4.35−4.39 (d, 1H, C(2)H, J = 11.5 Hz), 4.80−4.89 (m, 2H, C(3)H,
C(11)H), 5.82−5.89 (m, 2H, C(6)H, C(8)H), 6.82−7.08 (m, 6H,
arom (silibinin)), 7.47−7.51 (m, 1H, arom (tacrine)), 7.72−7.76
(arom (tacrine)), 8.22−8.24 (m, 1H, arom (tacrine)) ppm. 13C NMR
(CDCl3, 75 MHz) δ: 24.69, 25.76, 27.49, 29.96, 30.05, 32.48, 32.90,
33.15, 34.06, 34.33, 42.95, 51.97 (12C, CONH(CH2)6NH,
CH2(CH2)2CH2 (tacrine), O2CCH2CH2CONH)), 59.54 (OCH3),
66.92, 67.16, 76.25, 80.24, 86.88, 86.98, 99.37, 99.75, 103.98, 104.05,
114.30, 115.75, 119.22, 120.02, 120.47, 102.76, 123.83, 124.35, 125.17,
Fluorescent Microscopy and Image Analysis. For fluorescence
microscopic analysis, cells were seeded in 96-well plates at a density of
5 × 103 per well and cultured for 24 h. Then cells were treated for
another 24 h either with medium, medium with solvent, or with test
compounds. Then, the cells were incubated with a 1:2000 dilution of
MitoTracker Red CMXRos (Molecular Probes, Invitrogen) for 25 min
at 37 °C. Subsequently, nuclei were stained with Hoechst33342
(bisBenzimide H 33342 trihydrochloride, Sigma-Aldrich, 20 mM,
1:1200) for 5 min. Afterward, cells were fixed with 3%
paraformaldehyde for 30 min at room temperature and then shortly
treated with permeabilization buffer (PBS with 0.1% Triton X-100) to
improve signal-to-noise ratio.
Automatic image acquisition was carried out using a Carl Zeiss Axio
Observer (Carl Zeiss, Gottingen, Germany) with Software AxioVision
̈
4.8.1 (Carl Zeiss MicroImaging, Germany), motorized stage, AxioCam
HRm, a Plan-Neofluar 10× objective, and appropriated filters for the
fluorescent dyes.
Image analysis was done automatically by the ASSAYbuilder
Physiology Analyst software. Cell nuclei were identified by the
software and used to automatically detect and count the cells. The
mean fluorescence intensity of detected mitochondria spots was
measured within a defined ring mask over and around the nucleus.
5239
dx.doi.org/10.1021/jm300246n | J. Med. Chem. 2012, 55, 5231−5242

Journal of Medicinal Chemistry
Article
Statistical Analysis (in Vitro Hepatoprotection and Neuro-
protection). In the cell-based assays, results are presented as the
AUTHOR INFORMATION
Corresponding Author
■
mean
SD and refer to untreated control cells which were set as
*(C.F.) Tel: 0049-3641-9325551. Fax: 0049-3641-9325652. E-
mail: christian.fleck@mti.uni-jena.de. (M.D.) Tel: 0049-941-
943-3289. Fax: 0049-941-943-4990. E-mail: michael.decker@
chemie.uni-regensburg.de.
100% values. If not mentioned otherwise, experiments were carried out
with three parallels and repeated independently at least three times.
Statistical analysis was performed using GraphPad Prism 4 Software.
Levels of significance: p < 0.05 (*), p < 0.01 (**), and p < 0.001
(***).
Notes
In Vivo Studies on Hepatotoxicity/Hepatoprotection and on
the Interaction Capacity with the Cytochrome P450 System.
For the two experimental settings used (2 and 6 μmol/100 g body
weight), the rats were randomly assigned to different treatment groups
(n = 5−8 each): (1) control rats, receiving the solvent 0.9% NaCl i.p.,
(2) silibinin treated rats, (3) tacrine treated rats, (4) rats given the
mixture of tacrine and silibinin, and (5) codrug 10 treated rats. All
substances were given i.p. Twenty-four hours after drug admin-
istration, the animals were sacrificed, and the livers were removed for
histological and biochemical investigations. For details concerning the
biochemical and histological investigations, see Supporting Informa-
tion.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
M.D. thanks the Prorectorate for Research of Regensburg
University for financial support. A ‘Habilitationsstipendium’ of
Regensburg University for B.K. is gratefully acknowledged.
Appreciation is expressed to Mrs. Annegret Berthold for skillful
technical assistance and to Professor S. Elz for financial support
and scientific mentoring.
DEDICATION
Behavioral Studies. We examined whether the codrug could
improve spatial memory impairment induced by scopolamine using
performance in an eight-arm radial maze (RAM).43a
■
This article is dedicated to Professor John L. Neumeyer on the
occasion of his 82nd birthday.
Each arm (44 cm L, 30 cm H, 14 cm W) radiated from an octagonal
platform that served as a starting point. A food cup (3 cm diameter)
was located at the end of each arm. The entire arms and food cups
were painted gray and placed in a dark and calm room. Animal
behavior was monitored by a video camera. Image analysis and pattern
recognition from the monitor were performed by a VideoMot 2
program (video tracking, motion analysis, and behavior recognition
system) provided by TSE Systems (Bad Homburg, Germany). The
computerized recording systems were located in the same room.
After a short handling period in which the rats were in close contact
with the laboratory staff, three rats experienced free movement and
feeding in the RAM once a day for 5 days to adapt to the maze. The
baits (Dustless Precision Rodent Pellets, Bilaney Consultants Ltd.,
Sevenoaks, Kent, U. K.) were scattered in all arms. After handling and
adaptation of the rats to the maze, food was restricted to reduce the
rat’s body weight by 10%.
Training trials: Each rat was placed once daily in the center of the
RAM to visit all eight arms and eat all reward food baits in each food
cup. Each trial was performed until the rat entered and ate all the
pellets in the eight arms, or made 16 errors (re-entry into an arm that
has been previously visited), or 10 min elapsed.
Memory impairment by scopolamine was induced in rats trained to
the criteria of trials at three consecutive days. Before the
administration of substances, a control run was performed. Thus,
each animal was its own control. The following parameters were
registered 20, 60, and 120 min after scopolamine: number of errors
and time needed to visit all arms (total exploration time). The trial was
finished after either the rat had eaten all pellets, 10 min had elapsed, or
16 errors were made.
ABBREVIATIONS USED
■
ACh(E), acetylcholine(esterase); AD, Alzheimer’s disease;
BBB, blood−brain barrier; BChE, butyrylcholinesterase; CAS,
catalytic anionic site; CNS, central nervous system; CYP,
cytochrome P450; DCM, dichloromethane; DIAD, diisopropyl
azodicarboxylate; DMF, N,N-dimethylformamide; DMSO,
dimethyl sulfoxide; EDCI, 1-ethyl-3-(3-dimethylaminopropyl)
carbodiimide hydrochloride; GSH, reduced glutathione; GSSG,
oxidized glutathione; HOBt, hydroxybenzotriazole; HSC,
hepatic stellate cells; M receptor, muscarinic acetylcholine
receptor; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetra-
zolium bromide; ROS, reactive oxygen species; SAR,
structure−activity relationship; TBARS, thiobarbituric acid
reagible substances; TFA, trifluoroacetic acid; THF, tetrahy-
drofuran
REFERENCES
■
(1) Ballard, C.; Gauthier, S.; Corbett, A.; Brayne, C.; Aarsland, D.;
Jones, E. Alzheimer’s disease. Lancet 2011, 377, 1019−1031.
(2) (a) Knapp, M. J.; Knopman, D. S.; Solomon, P. R.; Pendlebury,
W. W.; Davis, C. S.; Gracon, S. I. A 30-week randomized controlled
trial of high-dose tacrine in patients with Alzheimer’s disease. J. Am.
Med. Assoc. 1994, 271, 985−991. (b) Takeda, A.; Loveman, E.; Clegg,
A.; Kirby, J.; Picot, J.; Payne, E. C. A systematic review of the clinical
effectiveness of donepezil, rivastigmine and galantamine on cognition,
quality of life and adverse events in Alzheimer’s disease. Int. J. Geriatr.
Psych. 2006, 21, 17−28.
(3) (a) Watkins, P. B.; Zimmermann, H. J.; Knapp, M. J.; Gracon, S.
I.; Lewis, K. W. Hepatotoxic effects of tacrine administration in
patients with Alzheimer’s disease. J. Am. Med. Assoc. 1994, 271, 992−
998. (b) Gracon, S. I.; Knapp, M. J.; Berghoff, W. G.; Pierce, M.;
DeJong, R.; Lobbestael, S. J.; Symons, J.; Dombey, S. L.; Luscombe, F.
A.; Kraemer, D. Safety of tacrine: clinical trials, treatment IND, and
postmarketing experience. Alzheimer Dis. Ass. Disord. 1998, 12, 93−
101.
Experimental groups: (1) 0.05 mg scopolamine/100 g b.wt. (scop);
5 min after scop administration coadministration of (2) 2 μmol
silibinin (sili), (3) 2 μmol tacrine (tac), (4) 2 μmol equimolar mixture
of tacrine and silibinin (equ. mix.), and (5) 2 μmol codrug (codrug)/
100 g b. wt. for each coadministration. Solutions were injected
intraperitoneally (i.p.).
ASSOCIATED CONTENT
■
(4) For reviews see (a) Tumiatti, V.; Minarini, A.; Bolognesi, M. L.;
Milelli, A.; Rosini, M.; Melchiorre, C. Tacrine derivatives and
Alzheimer’s disease. Curr. Med. Chem. 2010, 17, 1825−1838.
(b) Musial, A.; Bajda, M.; Malawska, B. Recent developments in
cholinesterases inhibitors for Alzheimer’s disease treatment. Curr. Med.
Chem. 2007, 14, 2654−2679.
S
* Supporting Information
Experimental and spectral data of synthetic precursors; detailed
information about cytochrome P450 isoform expression and
activity; and information on cell culture and cell lines, statistics,
information on animals, biochemical and histological proce-
dures, and statistics of behavioral studies. This material is
available free of charge via the Internet at http://pubs.acs.org.
(5) Lange, J. H. M.; Coolen, H. K. A. C.; van der Neut, M. A. W.;
Borts, A. J. M.; Sork, B.; Verveer, P. C.; Kruse, C. G. Design, synthesis,
5240
dx.doi.org/10.1021/jm300246n | J. Med. Chem. 2012, 55, 5231−5242

Journal of Medicinal Chemistry
Article
biological properties, and molecular modeling investigations of novel
tacrine derivatives with a combination of acetylcholinesterase
inhibition and cannabinoid CB1 receptor antagonism. J. Med. Chem.
2010, 53, 1338−1346.
tetrahydroacridines and 9-(pyrazol-1-yl)-tetrahydroacridines derived
from chiral cyclanones. Eur. J. Med. Chem. 2004, 39, 37−48.
(20) Wang, F.; Huang, K.; Yang, L.; Gong, J.; Tao, Q.; Li, H.; Zhao,
Y.; Zeng, S.; Wu, X.; Stockigt, J.; Li, X.; Qu, J. Preparation of C-23
̈
esterified silybin derivatives and evaluation of their peroxidation
inhibitory and DNA protective properties. Bioorg. Med. Chem. 2009,
17, 6380−6389.
(6) Fang, L.; Jumpertz, S.; Appenroth, D.; Fleck, C.; Mohr, K.;
Trankle, C.; Decker, M. Hybrid molecules from xanomeline and
̈
tacrine: Enhanced tacrine actions on cholinesterases and muscarinic
M1 receptors. J. Med. Chem. 2010, 53, 2094−2103.
(21) Lee, D. Y. W.; Zhang, X.; Ji, X. S. Preparation of tritium-labeled
silybin: A protectant for common liver diseases. J. Labelled Compd.
Radiopharm. 2006, 49, 1125−1130.
(7) Fang, L.; Kraus, B.; Lehmann, J.; Heilmann, J.; Zhang, Y.; Decker,
M. Design and synthesis of tacrine-ferulic acid hybrids as multi-potent
anti-Alzheimer drug candidates. Bioorg. Med. Chem. Lett. 2008, 18,
2905−2909.
(22) (a) Ellman, G. L.; Courtney, K. D.; Andres, V.; Featherstone, R.
M. A new and rapid colorimetric determination of acetylcholinesterase
activity. Biochem. Pharmacol. 1961, 7, 88−95. (b) Decker, M. Novel
inhibitors of acetyl- and butyrylcholinesterase derived from the
alkaloids dehydroevodiamine and rutaecarpine. Eur. J. Med. Chem.
2005, 40, 305−313.
(8) (a) Lupp, A.; Appenroth, D.; Fang, L.; Decker, M.; Lehmann, J.;
Fleck, C. Tacrine-NO and tacrine-ferulic acid hybrid molecules as new
anti-Alzheimer agents: Hepatotoxicity and influence on the
cytochrome P450 system in comparison to tacrine. Arzneim.-Forsch.
2010, 60, 229−237. (b) Fleck, C.; Appenroth, D.; Fang, L.; Schott, Y.;
Lehmann, J.; Decker, M. Investigation into the in vivo effects of five
newly synthesized tacrine/ferulic acid and β-carboline derivatives on
scopolamine-induced cognitive impairments in rats using radial maze
experiments. Arzneim.-Forsch. 2010, 60, 299−306.
̆
(23) (a) Çokugras,̧ A. N. Butyrylcholinesterase: Structure and
physiological function. Turk. J. Biochem. 2003, 28, 54−62.
(b) Guillozet, A. L.; Smiley, J. F.; Mash, D. C.; Mesulam, M. M.
Butyrylcholinesterase in the life cycle of amyloid plaques. Ann. Neurol.
1997, 42, 909−918.
(24) (a) 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.
Selective butyrylcholinesterase inhibition elevates brain acetylcholine,
augments learning, and lowers Alzheimer beta-amyloid peptide in
rodent. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 17213−17218.
(b) Mesulam, M. M.; Guillozet, A.; Shaw, P.; Levey, A.; Duysen, E.
G.; Lockridge, O. Acetylcholinesterase knockouts establish central
cholinergic pathways and can use butyrylcholinesterase to hydrolyze
acetylcholine. Neuroscience 2002, 110, 627−639.
(25) (a) Montiel, T.; Quiroz-Baez, R.; Massieu, L.; Arias, C. Role of
oxidative stress on β-amyloid neurotoxicity elicited during impairment
of energy in the hippocampus: protection by antioxidants. Exp. Neurol.
2006, 200, 496−508. (b) Palmer, A. M. Neuroprotective therapeutics
for Alzheimer’s disease: progress and prospects. Trends Pharmacol. Sci.
2011, 32, 141−147.
(26) (a) Koufaki, M.; Tsatsaroni, A.; Alexi, X.; Guerrand, H.; Zerva,
S.; Alexis, M. N. Isoxazole substituted chromans against oxidative
stress-induced neuronal damage. Bioorg. Med. Chem. 2011, 19, 4841−
4859. (b) Lee, K. Y.; Hwang, L.; Jeong, E. J.; Kim, S. H.; Kim, Y. C.;
Sung, S. H. Effect of neuroprotective flavonoids of Agrimonia eupatoria
on glutamate-induced oxidative injury to HT22 hippocampal cells.
Biosci. Biotechnol. Biochem. 2010, 74, 1704−1706. (c) Yang, H.; Sung,
S. H.; Kim, J.; Kim, Y. C. Neuroprotective diarylheptanoids from the
leaves and twigs of Juglans sinensis against glutamate-induced toxicity
on HT22 cells. Planta Med. 2011, 77, 841−845.
(9) (a) Fang, L.; Appenroth, D.; Decker, M.; Kiehntopf, M.; Roegler,
C.; Deufel, T.; Fleck, C.; Peng, S.; Zhang, Y.; Lehmann, J. Synthesis
and biological evaluation of NO-donor-tacrine hybrids as hepatopro-
tective anti-Alzheimer drug candidates. J. Med. Chem. 2008, 51, 713−
716. (b) Fang, L.; Appenroth, D.; Decker, M.; Kiehntopf, M.; Lupp,
A.; Peng, S.; Fleck, C.; Zhang, Y.; Lehmann, J. NO-Donating tacrine
hybrid compounds improve scopolamine-induced cognition impair-
ment and show less hepatotoxicity. J. Med. Chem. 2008, 51, 7666−
7669.
(10) Gazak, R.; Walterova, D.; Kren, V. Silybin and silymarin: new
̌
́ ́ ́
and emerging applications in medicine. Curr. Med. Chem. 2007, 14,
315−338 and references therein.
(11) (a) Yang, L. X.; Huang, K. X.; Li, H. B.; Gong, J. X.; Wang, F.;
Feng, Y. B.; Tao, Q. F.; Wu, Y. H.; Li, X. K.; Wu, X. M.; Zeng, S.;
Spencer, S.; Zhao, Y.; Qu, J. Design, synthesis, and examination of
neuron protective properties of alkenylated and amidated dehydro-
̌
́
silybin derivatives. J. Med. Chem. 2009, 52, 7732−7752. (b) Gazak, R.;
Svobodova,
́ ́ ̌ ́ ́
A.; Psotova, J.; Sedmera, P.; Prikrylova, V.; Walterova, D.;
̌
Kren, V. Oxidised derivatives of silybin and their antiradical and
antioxidant activity. Bioorg. Med. Chem. 2004, 12, 5677−5687.
(12) Nencinia, C.; Giorgia, G.; Micheli, L. Protective effect of
silymarin on oxidative stress in rat brain. Phytomedicine 2007, 14, 129−
135.
(13) Loguericio, C.; Festi, D. Silybin and the liver: From basic
research to clinical practice. World J. Gastroenterol. 2011, 17, 2288−
2301 and references therein..
(14) Decker, M. Hybrid molecules incorporating natural products:
Applications in cancer therapy, neurodegenerative disorders and
beyond. Curr. Med. Chem. 2011, 18, 1464−1475.
(27) Choi, D. W. Glutamate neurotoxicity and diseases of the
nervous system. Neuron 1988, 1, 623−634.
(28) Murphy, T. H.; Miyamoto, M.; Sastre, A.; Schnaar, R. L.; Coyle,
J. T. Glutamate toxicity in a neuronal cell line involves inhibition of
cystine transport leading to oxidative stress. Neuron 1989, 2, 1547−
1558.
(15) R. Gazak
M.; Medina, M. A.; Orborna,
antiangiogenic activity of new silybin galloyl esters. J. Med. Chem.
2011, 54, 7397−7407.
́
, R.; Valentova,
́
K.; Fuksova,
́
K.; Marhol, P.; Kuzma,
n, V. Synthesis and
̌
́
́
I.; Ulrichova, J.; Kre
́
̌
(29) Maher, P.; Davies, J. B. The role of monoamine metabolism in
oxidative glutamate toxicity. J. Neurosci. 1996, 16, 6394−6401.
(30) (a) Tan, S.; Wood, M.; Maher, P. Oxidative stress induces a
form of programmed cell death with characteristics of both apoptosis
and necrosis in neuronal cells. J. Neurochem. 1998, 71, 95−105.
(b) Tan, S.; Schubert, D.; Maher, P. Oxytosis: a novel form of
programmed cell death. Curr. Top. Med. Chem. 2001, 1, 497−506.
(31) (a) Behl, C. Vitamin E protects neurons against oxidative cell
death in vitro more effectively than 17-beta estradiol and induces the
activity of the transcription factor NF-kappaB. J. Neural. Transm. 2000,
107, 393−407. (b) Ishige, K.; Schubert, D.; Sagara, Y. Flavonoids
protect neuronal cells from oxidative stress by three distinct
mechanisms. Free Radical Biol. Med. 2001, 30, 433−446.
(16) Decker, M.; Kraus, B.; Heilmann, J. Design, synthesis and
pharmacological evaluation of hybrid molecules out of quinazolini-
mines and lipoic acid lead to highly potent and selective
butyrylcholinesterase inhibitors with antioxidant properties. Bioorg.
Med. Chem. 2008, 16, 4252−4261.
(17) Budzik, B.; Garzya, V.; Shi, D.; Walker, G.; Woolley-Roberts,
M.; Pardoe, J.; Lucas, A.; Tehan, B.; Rivero, R. A.; Langmead, C. J.;
Watson, J.; Wu, Z.; Forbes, I. T.; Jin, J. Novel N-substituted
benzimidazolones as potent, selective, CNS-penetrant, and orally
active M1 mAChR agonists. ACS Med. Chem. Lett. 2010, 1, 244−248.
(18) Hu, M. K.; Lu, C. F. A facile synthesis of bis-tacrine isosteres.
Tetrahedron Lett. 2000, 41, 1815−1818.
(32) Schnabl, B.; Choi, Y. H.; Olsen, J. C.; Hagedorn, C. H.; Brenner,
D. A. Immortal activated human hepatic stellate cells generated by
ectopic telomerase expression. Lab. Invest. 2002, 82, 323−33.
(19) Frideling, A.; Faure, R.; Galy, J.-P.; Kenz, A.; Alkorta, I.; Elguero,
J. Tetrahydroacridin-9-ones, 9-chlorotetrahydroacridines, 9-amino-
5241
dx.doi.org/10.1021/jm300246n | J. Med. Chem. 2012, 55, 5231−5242

Journal of Medicinal Chemistry
Article
(33) Gillies, R. J.; Didier, N.; Denton, M. Determination of cell
number in monolayer cultures. Anal. Biochem. 1986, 159, 109−113.
(34) Jaeschke, H.; Gores, G. G.; Cederbaum, A. I.; Hinson, J. A.;
Pessayre, D.; Lemasters, J. J. Mechanisms of hepatotoxicity. Toxicol.
Sci. 2002, 65, 166−176.
(35) Berson, A.; Renault, S.; Letter
Fau, D.; Le Bot, M. A.; Riche,
Feldmann, G.; Pessayre, D. Uncoupling of rat and human
mitochondria: a possible explanation for tacrine-induced liver
dysfunction. Gastroenterolgy 1996, 110, 1878−1890.
́
on, P.; Robin, M.-A.; Fromenty, B;
C.; Durand-Schneider, A.-M.;
́
(36) Lemasters, J. J.; Quian, T.; Bradham, C. A.; Brenner, D. A.;
Cascio, W. E.; Trost, L. C.; Nishimura, Y.; Nieminen, A.-L.; Herman,
B. Mitochondrial dysfunction in the pathogenesis of necrotic and
apoptotic cell death. J. Bioenerg. Biomembr. 1999, 31, 305−319.
(37) Rasola, A; Bernardi, P. The mitochondrial permeability
transition pore and its involvement in cell death and in disease
pathogenesis. Apoptosis 2007, 12, 815−833.
(38) Oh, H.; Kim, J.-S.; Song, E.-K.; Cho, H.; Kim, D.-H.; Park, S.-E.;
Lee, H.-S.; Kim, Y.-C. Sesquiterpenes with hepatoprotective activity
from Cnidium monnieri on tacrine-induced cytotoxicity in Hep G2
cells. Planta Med. 2002, 68, 748−749.
(39) Mimica, N.; Presecki, P. Side effects of approved anti-
dementives. Psychiatr. Danub. 2009, 21, 108−113.
(40) Sedky, K.; Nazir, R.; Joshi, A.; Kaur, G.; Lippmann, S. Which
psychotropic medications induce hepatotoxicity? Gen. Hosp. Psychiatry
2012, 34, 53−61.
(41) Letteron, P.; Labbe, G.; Degott, C.; Berson, A.; Fromenty, B.;
Delaforge, M.; Larrey, D.; Pessayre, D. Mechanism for the protective
effects of silymarin against carbon tetrachloride-induced lipid
peroxidation and hepatotoxicity in mice. Evidence that silymarin acts
both as an inhibitor of metabolic activation and as a chain-breaking
antioxidant. Biochem. Pharmacol. 1990, 39, 2027−2034.
(42) Raskovic, A.; Stilinovic, N.; Kolarovic, J; Vasovic, V.;
Yukmirovic, S.; Mikov, M. The protective effects of silymarin against
doxorubicin-induced cardiotoxicity and hepatotoxicity in rats.
Molecules (Basel) 2011, 16, 8601−8613.
(43) (a) Rawlins, J. N. P.; Deacon, R. M. J. Further Developments of
Maze Procedures. In Behavioural Neuroscience, Vol. I: A Practical
Approach.; Sahgal, A., Ed.; IRL Press: Oxford, U.K., 1993; pp 95−122;
(b) Appenroth, D.; Decker, M.; Lehmann, J.; Trankle, C.; Mohr, K.;
̈
Fleck, C. In vivo-investigations on the cholinesterase-inhibiting effects
of tricyclic quinazolinimines: Scopolamine-induced cognitive impair-
ments in rats are attenuated at low dosage and reinforced at higher
dosage. Pfluegers Arch. 2008, 455, 895−901.
(44) Mosmann, T. Rapid colorimetric assay for cellular growth and
survival: application to proliferation and cytotoxicity assays. J. Immunol.
Methods 1983, 65, 55−63.
(45) Kraus, B.; Wolff, H.; Elstner, E. F.; Heilmann, J. Hyperforin is a
modulator of inducible nitric oxide synthase and phagocytosis in
microglia and macrophages. Naunyn-Schmiedeberg's Arch. Pharmacol.
2010, 381, 541−553.
5242
dx.doi.org/10.1021/jm300246n | J. Med. Chem. 2012, 55, 5231−5242