Gramine Derivatives Targeting Ca2+ Channels and Ser/Thr Phosphatases: A New Dual Strategy for the Treatment of Neurodegenerative Diseases

Article abstract of DOI:10.1021/acs.jmedchem.6b00478

We describe the synthesis of gramine derivatives and their pharmacological evaluation as multipotent drugs for the treatment of Alzheimer's disease. An innovative multitarget approach is presented, targeting both voltage-gated Ca²⁺ channels, cl

Full text of DOI:10.1021/acs.jmedchem.6b00478

Subscriber access provided by UNIV OF NEBRASKA - LINCOLN
Article
2+
Gramine derivatives targeting Ca channels and Ser/Thr phosphatases:
a new dual strategy for the treatment of neurodenerative diseases
Rocío Lajarín-Cuesta, Carmen Nanclares, Juan Alberto Arranz-Tagarro, Laura González-
Lafuente, Raquel L Arribas, Monique Araújo de Brito, Luis Gandia, and Cristóbal de los Ríos
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00478 • Publication Date (Web): 09 Jun 2016
Downloaded from http://pubs.acs.org on June 15, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
+
Gramine derivatives targeting Ca channels and
Ser/Thr phosphatases: a new dual strategy for the
treatment of neurodegenerative diseases
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
†
†
†
Rocío Lajarín-Cuesta, Carmen Nanclares, Juan-Alberto Arranz-Tagarro, Laura González-
‡
,
†
§
†
∥
Lafuente, Raquel L. Arribas, Monique Araujo de Brito, Luis Gandía, Cristóbal de los
,
†,‡
Ríos*
†
Instituto Teófilo Hernando and Departamento de Farmacología y Terapéutica, Facultad de
Medicina, Universidad Autónoma de Madrid, C/ Arzobispo Morcillo, 4, 28029 Madrid, Spain
§
Programa de Pós Graduação em Ciências Aplicadas a Produtos Para a Saúde, Faculdade de
Farmácia, Universidade Federal Fluminense, Niterói, RJ, Brasil
‡
Servicio de Farmacología Clínica, Instituto de Investigación Sanitaria, Hospital Universitario de
la Princesa, C/ Diego de León, 62, 28006 Madrid, Spain
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
We describe the synthesis of gramine derivatives and their pharmacological evaluation as
multipotent drugs for the treatment of Alzheimer´s disease. An innovative multitarget approach
2
+
is presented, targeting both voltageꢀgated Ca
channels, classically studied for
neurodegenerative diseases, and Ser/Thr phosphatases, which have been marginally aimed, even
despite their key role in protein τ dephosphorylation. Twenty five compounds were synthesized
and mostly their neuroprotective profile exceeded that offered by the head compound gramine. In
2
+
general, these compounds reduced the entry of Ca through VGCC, as measured by Fluoꢀ4/AM
2
+
and patch clamp techniques, and protected in Ca overloadꢀinduced models of neurotoxicity,
like glutamate or veratridine exposures. Furthermore, we hypothesize that these compounds
decrease τ hyperphosphorylation based on the maintenance of the Ser/Thr phosphatase activity
and their neuroprotection against the damage caused by okadaic acid. Hence, we propose this
multitarget approach as a new and promising strategy for the treatment of neurodegenerative
diseases.
ACS Paragon Plus Environment

Page 3 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Introduction
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Over a hundred years after Alois Alzheimer reported his observations describing a new
1
neurodegenerative disease, the battery of clinically approved drugs available to treat
Alzheimer’s disease (AD) patients is scarce and inefficient. Nowadays, AD patients are
addressed with three cholinesterases (ChE) inhibitors (galantamine, donepezil, and rivastigmine),
2
and memantine, a Nꢀmethylꢀ
D
ꢀaspartate (NMDA) glutamate receptor blocker. The former
augments the levels of the neurotransmitter acetylcholine at the cholinergic synapsis, under the
2
+
3
rationale of the soꢀcalled “cholinergic hypothesis”. Memantine avoids the altered Ca uptake by
neurons through the NMDAꢀsensitive ionic channelꢀcoupled receptor, relying on the theory that
2
+
3
all neurodegenerative diseases present an imbalanced Ca homeostasis, so the reduction of the
2+
cytosolic Ca concentration in neurons could slow down the progression of such diseases.
4
Attempts to discover better drugs acting on any of these therapeutic targets have failed.
Targeting the bestꢀdocumented pathophysiological hallmarks of AD, that are senile plaques,
5
mainly formed by the amyloid β peptide (Aβ), and neurofibrillary tangles (NFT) generated by
6
the aggregation of the hyperphosphorylated microtubuleꢀstabilizing τ protein, neither have
supplied more successful therapies. To face amyloidogenesis, several strategies have been
studied, e.g. compounds inhibiting the synthesis of Aβ (αꢀsecretase activators, both βꢀ and γꢀ
7
secretase inhibitors, among others) or its aggregation. As far as the τꢀinduced NFT formation,
less work, but not negligible, has been carried out, mainly devoted to the preparation of kinase
enzymes inhibitors, with the goal of hindering the hyperphosphorylation of τ, as the more
8ꢀ9
phosphorylated τ protein is, the more susceptibility to its selfꢀaggregation into NFT.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Nevertheless, few of these approaches have supplied drugs eligible to be studied in clinical trials
for AD. Many reasons could explain the lack of promising drugs for AD, but it seems quite
likely that, as other neurodegenerative diseases, AD features a multifactorial origin, where
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
0
various physiopathological events participate in the neuronal damage that leads to AD. Such
hypothesis gave rise to the multiꢀtarget approach in drug discovery, which postulates that a drug,
or the association of drugs, acting on two or more different biological targets that operate in the
pathological process, could present a more beneficial effect on AD patients than a highly potent
1
1
singleꢀtargeting drug. Hence, during the last decade, a plethora of multiꢀtarget drugꢀbased
strategies have been described, by combining pharmacological interventions over ChE enzymes,
1
2ꢀ15
Aβ peptides or the τꢀinduced NFT formation, among other therapeutic targets. In this regard,
our research group has reported several families of multiꢀtarget drugs since 2000, with the
2
+
16ꢀ19
control of neuronal Ca homeostasis as the common target. Many evidences support the idea
2
+
that pharmacological modulation of Ca homeostasis within neurons could mitigate the
20
neurotoxicity present in a neurodegeneration scenario, as that neurons having NFT present high
2
+
2+
21
activity of Ca ꢀdependent protease enzymes, or that the presence of Aβ peptides elevates Ca
22ꢀ23
concentrations in resting neurons, sensitizing them for apoptosis phenomena. There is
2
+
preclinical and clinical evidence that the blockade of voltageꢀgated Ca channels (VGCC) may
2
+
attenuate dementia, since an excessive and prolonged Ca entry into the cytosol leads to
24ꢀ26
neurodegeneration and neuronal loss. The VGCC blockade, both at presynaptic (P/Qꢀ and Nꢀ
type VGCC) and at postsynaptic sites (Lꢀtype VGCC), has shown a neuroprotective effect in
different animal models of neurodegeneration and excitotoxicity, such as the blockade of ωꢀ
2
7
agatoxin IVA in a rat ischemia model (P/QꢀVGCC blocker), the blockade by the synthetic
version of ωꢀconotoxin MVIIA in a traumatic diffuse brain injury model in rats (Nꢀtype VGCC
ACS Paragon Plus Environment

Page 5 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
28
29
blocker), that of verapamil in a focal ischemia model in rats (Lꢀtype VGCC blocker), or those
of nivaldipine and nitrendipine (Lꢀtype VGCC blockers) in an Aβꢀinduced toxicity model in
30
mice. Some VGCC antagonists have been tested in clinical trials for dementia and, according to
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
the Cochrane report published in 2002, nimodipine can be of some efficacy for the treatment of
31
AD.
On the other hand, during the latest years, we have paid attention to the neurodegeneration
32
elicited by the hyperphosphorylated τ protein and the subsequent intracellular NFT. The degree
9
of τ phosphorylation depends on the enzymatic activity of both kinases and phosphatases.
Tremendous scientific efforts have been invested in the development of kinase enzymes
33ꢀ34
inhibitors, e.g. GSK3β inhibitors, with scarce success. However, rather few contributions
35
have aimed to promote the dephosphorylation process carried out by phosphatase enzymes,
mainly performed by the phosphoprotein phosphatases (PPP) 2A and 1 (PP2A and PP1,
3
6
respectively). PPP enzymes catalyze the hydrolysis of phosphate esters at Ser and Thr residues,
being PP1 and PP2A responsible of the 90% of the cellular Ser/Thr phosphatase activity within
36
healthy human brains. In AD brains, total phosphatase activity is reduced by half, with PP2A,
36ꢀ37
PP1, and PP5 activities decreased by 50%, 20%, and 20%, respectively.
Regarding to PP2A,
35
it is the most efficient τ phosphatase (accounts for over 70% of τ dephosphorylation). The
relationship between the ADꢀtriggered neurodegeneration and the downregulation of PP2A
activity has been widely demonstrated: in postmortem brains from AD patients, PP2A mRNA
PP2A
3
8ꢀ39
and protein levels are decreased, and the endogenous PP2A inhibitors I1
(nuclear) and
PP2A
40ꢀ41
I2
(cytoplasmatic) are increased by 20%; it exists an augmented inactivation of the PP2A
42
catalytic subunit (PP2Ac) through the phosphorylation at Y307 residue; the PP2A methylation
43
rates at L309, responsible for the PP2A assembly to its regulatory subunits, is decreased; there
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
is also a reduced expression of B55α, the main PP2A regulatory subunit mediating
44ꢀ46
dephosphorylation of τ, in frontal and temporal cortex. Moreover, PP2A dephosphorylates
2+
NMDA receptors at several NRx subunits, evoking a lesser Ca influx through these receptors,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
7ꢀ48
what entails a neuroprotective effect against excitotoxicity, similarly to that of memantine.
Although the leading role of PP2A is indisputable, other PPP enzymes, like PP1 and PP5, are
49
also implicated in the pathology of AD.
Our research group described the neuroprotective properties of a multiꢀtarget drug that was able
50
to inhibit cholinesterase enzymes and to maintain the PP2Aꢀcontrolled phosphatase activity.
Confirming the correlation between NMDA and PP2A, ethyl 5ꢀaminoꢀ2ꢀmethylꢀ6,7,8,9ꢀ
tetrahydrobenzo[b][1,8]naphthyridineꢀ3ꢀcarboxylate (ITH12246), protected rat hippocampal
slices against the damage elicited by glutamate through a mechanism implicating PPP, as
okadaic acid (OA), a selective inhibitor of PP1 and PP2A, reversed its neuroprotective signal. In
addition, it reduced the loss of memory in mice subjected to scopolamine administration, as
51
monitored by the object placement test.
For all of these reasons, we considered worthwhile to keep deepening in the study of PPP
enzymes as a part of a multiꢀtarget strategy to find new drugs for AD, remaining the control of
2
+
cytosolic Ca as the other main therapeutic target. In this work, we describe the findings
observed with a family of indole derivatives, analogues to the natural alkaloid gramine (1, Figure
52
1), which has shown promising pharmacological activities.
ACS Paragon Plus Environment

Page 7 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Chemical structure of the alkaloid gramine (3ꢀ(dimethylaminomethyl)ꢀ1Hꢀindole) (1).
2
+
53
Indeed, some of its brominated derivatives have been described as Ca channels blockers. In
our laboratory, we observed that 1 protected SHꢀSY5Y neuroblastoma cells against the toxic
stimulus generated by OA in a wide range of concentrations. We wondered whether such huge
neuroprotective effect was, at least in part, due to the activation of PPP enzymes. This prompted
2
+
us to design a new family of 1 derivatives and evaluate their pharmacological activity as Ca
channel blockers and Ser/Thr phosphatase activators, as well as assessing their neuroprotective
profile in several in vitro models of neurodegeneration. The results reported in this paper support
the idea that scientific community devoted to the search of multiꢀtarget drugs must continue its
efforts to offer original proposals for the discovery of new drugs for neurodegenerative diseases.
Results and Discussion
2+
Chemistry. The design of 1 derivatives focused on favoring a potential Ca channel blocking
activity as well as the interaction with the principal dephosphorylating enzymes of τ, that are the
Ser/Thr phosphatases. As two marineꢀderived bromoꢀsubstituted 1 analogs have been described
5
3
to block VGCC, we reasoned that our first family of synthesized 1 derivatives should possess a
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
bromine atom at C5. Regarding to the binding to PP2A, previous studies of our research group
50
detected a binding site close to the catalytic core in the PP2Ac subunit, where specific ligands
would not compromise the PP2A enzymatic activity, but alternatively would avoid the arrival of
direct inhibitors to the catalytic machinery of PP2A. By preliminary computational studies, if
new indoleꢀbased ligands present a tiny hydrophobic substituent at C5, as well as a long
hydrophobic moiety linked to the indole nitrogen, their affinity to this binding site would
improve. Thus, a second family of 1 derivatives having a methyl group at C5 was designed,
where the members of both families are differentially substituted at the indole nitrogen. The Nꢀ
alkylation was carried out by nucleophilic substitution of alkyl halides in presence of NaH as a
base (Scheme 1). Alkyl, benzyl or propargyl bromides and iodides suffered the nucleophilic
attack of 5ꢀsubstituted indoles in excellent yields, except for ethyl bromobutyrate, as only 5ꢀ
bromoindole was able to lead to the formation of the ethyl indolylbutanoate 6 in low yields.
Attempts to optimize this substitution, by using other bases such as KOH and solvents (DMF,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CH CN, THF or acetone), phaseꢀtransfer catalysts, among other experimental conditions, were
3
not satisfactory. The nucleophilic substitution over 1ꢀchloroꢀ4ꢀiodobutane, leading to the Nꢀ(4’ꢀ
chlorobutyl)ꢀindoles 9 and 10, allowed us to incorporate piperidine as a pending substituent,
through a further nucleophilic attack of piperidine in acetonitrile, using K CO as a base,
2
3
obtaining the 5ꢀbromo or 5ꢀmethylꢀNꢀsubstituted derivatives 11 and 12, respectively (Scheme 1).
Polycations, such as polylysine and protamine, as well as highly basic compounds, specifically
54
histones, were the first compounds found to activate PP2A in vitro. This fact, together with
other recently discovered positively charged modulators, e.g. memantine, led to establish basic
55ꢀ56
cationic compounds as a family of PP2A activators. The head compound 1 possesses a basic
amino group that is positively charged at physiological pH, so it could be classified as a cationic
ACS Paragon Plus Environment

Page 9 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
activator. In order to keep this property, all 1 derivatives conserve this basic amino group at C3,
either as dimethylamine or piperidine group.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
Scheme 1. Preparation of 1 derivatives
a
Reagents and Conditions: (a) (1) NaH, DMSO, rt, 1–2 h; (2) alkyl halide, rt, 1–12 h. (b)
piperidine, MeCN, K CO , rt to 60 ºC, 12 h. (c) Me NH or piperidine, HCHO (aq), AcOH, 0 ºC
2
3
2
to rt, 2 to 5 h.
The further Mannich type reaction furnished the desired C3ꢀsubstituted 1 derivatives 20–38 in
which the order of addition of reagents played an essential role, allowing us to reach excellent
yields. 5ꢀBromoindole reacted with dimethylamine or piperidine at these conditions to form the
Nꢀunsubstituted 1 derivatives 16 and 17 in good yields (Scheme 1). However, the same protocol
proposed for the reaction of 5ꢀmethylindole with piperidine, derived in a complex mixture of
products. To sort this out, we selected an alternative method conducted through the generation of
the Nꢀoxide intermediate 14, from the commercially available 5ꢀmethylgramine 18, which is
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
further treated with neat piperidine to form compound 19 in quantitative yield. The same strategy
was proposed to prepare the piperidineꢀbearing analogue of 1, that is compound 15, synthesized
and pharmacologically evaluated for comparative purposes (Scheme 2).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Scheme 2. Alternative route for the preparation of N-unsubstituted 1 derivatives
a
Reagents and conditions: (a) H O (30% aq), EtOH, rt, 2 h. (b) piperidine, reflux, 3 h.
2
2
At this point, it is worthwhile mentioning that, despite the huge literature in both Organic and
Medicinal Chemistry regarding to the syntheses and uses of 3ꢀaminoethylindoles, backed by the
57ꢀ
Chemistry of the tryptophan, serotonin and melatonin, among other soꢀcalled biogenic amines,
59
the study of the 3ꢀaminomethylindoles analogs has been comparatively inappreciable. Indeed,
1 derivatives have been extensively used in Organic Synthesis as starting material for the
60ꢀ61
preparation of serotonineꢀlike natural products and potential drugs. Among the reasons for
this lack of interest, their delicate chemical stability in aqueous acid media seems to be the main
1
62
reason. We confirmed by HꢀNMR that 3ꢀmethylenꢀ3Hꢀindoles are formed as byꢀproducts after
routine procedures such as silica gelꢀbased flash chromatography purification or in the NMR
analysis when using CDCl as deuterated solvent. These pitfalls were avoided by executing the
3
workꢀup at pH 14, neutralizing the organic solvents, mainly CH Cl , with basic alumina and,
2
2
when necessary, purification with flash chromatography was carried out in basic alumina instead
ACS Paragon Plus Environment

Page 11 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
of silica gel, and NMR spectra were made in acetoneꢀd . After confirming the structure and
6
purity of final compounds 15–38, with the aim of facilitating their manipulation and biological
evaluation, they were salinized to hydrochloride or oxalate salts, by treatment with HCl in ether
or oxalic acid in ethyl acetate solutions, respectively. Solvents for the salinization solutions were
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
previously dried over MgSO to remove possible traces of water. We then confirmed structure
4
and purity of the salinized compounds, and their chemical stability at physiological pH by
1
3
1
monitoring C and HꢀNMR spectra in D O, not detecting any modification in the NMR signals
2
after several days.
2+
Pharmacological Evaluation. Effect of 1 derivatives 15–38 on the Ca entry through
VGCC. To evaluate the blocking activity of the new 1 derivatives, we stimulated SHꢀSY5Y
neuroblastoma cells with an extracellular solution containing KCl 70 mM, leading to membrane
6
3
depolarization and VGCC opening. The SHꢀSY5Y neuroblastoma cell line is a reliable model
2
+
to monitor cytosolic Ca fluctuations in excitable cells, as it possesses VGCC and intracellular
2
+
64ꢀ65
Ca ꢀbuffering organelles similarly to regular primary neuronal cultures. Consequently, the
2+
SHꢀSY5Y cells, charged with the Ca ꢀsensitive fluorescent probe Fluoꢀ4, suffered a rise in the
2
+
cytosolic Ca that was modulated by the presence of these potential VGCC blockers, tested at 1
+
2+
ꢀ
M (Table 1). Eleven out 24 compounds reduced the high K ꢀevoked cytosolic Ca increase in a
statistically significant manner. Considering the data expressed in Table 1, some structureꢀ
activity relationships can be inferred. Most of the compounds that block VGCC bear a
piperidinylmethyl substituent at C3; compound 20 (X = Br, R = Bn) was the only
2
+
dimethylaminomethylꢀsubstituted compound, together with 1, that reduced the Ca increase. As
2
+
far as the Nꢀalkylation, the benzyl moiety affords the best blockade of Ca increase, but also
butyl, chlorobutyl, and propargyl groups are optimal substituents, whose results are fairly better
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
when compounds present additionally a bromine atom at C5 and a piperidine at C3, instead of
methyl and dimethylamine, respectively. It is already known in literature that bromoꢀsubstitution
2
+
53
at the indole core provides compounds with Ca antagonist activity. The blocking effect of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
compounds that present a bulky substituent at the indole nitrogen (27, 28, 35–38) substantially
66
dropped to below 20%. The wellꢀknown VGCC blocker nifedipine, used as a standard at 3 ꢀM,
2+
2+
reduced Ca increase by 73% in these experiments. With the goal of deepening into their Ca ꢀ
antagonist effect, we carried out electrophysiological techniques, performing patchꢀclamp
2
+
experiments in the wholeꢀcell configuration. We studied the effect on Ca (I ) currents of 1, the
Ca
best VGCC blockers of the family (21, 25, 30, and 33), and three more compounds (20, 23, and
26) that showed important neuroprotective and/or proꢀphosphatases properties (see below).
Current vs. voltage (I/V) curves, from –60 to +60 mV as depolarizing potentials, were initially
performed in absence of compounds, to select the depolarizing pulse that featured the highest
ICa,. Then, bovine chromaffin cells were subjected to 50 ms depolarizing pulses every 20 s at the
potential at which maximal I was obtained, usually 0 mV; after reaching a stable response,
Ca
cells were perfused with the different compounds until a maximal blockade degree was achieved.
Figure S1 of supporting info reflects a typical register of I dealing with a control situation.
Ca,
Under these experimental conditions, compounds 1, 21, 23, 25, and 26 blocked I in a
Ca
statistically significant manner (Figure 2), confirming their VGCC blocking properties.
2
+
+
Table 1. Blockade by compounds 15–38 of the Ca entry elicited by K 70 mM-evoked
2
+
a
depolarization in SH-SY5Y cells, measured by the Ca -sensitive probe Fluo-4/AM
ACS Paragon Plus Environment

Page 13 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Comp.
X
R
R’
% Blockade
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
b
***
Nifedipine
1
ꢀ
H
ꢀ
H
ꢀ
73 ± 5
*
*
*
CH
25 ± 7
30 ± 4
3
*
15
16
17
H
H
ꢀ(CH
2
)
5
5
5
5
5
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ns
Br
H
CH
3 ± 3
3
ns
Br
H
ꢀ(CH
CH
)
7 ± 7
2
ns
ns
18
CH
CH
Br
3
H
3
10 ± 5
16 ± 6
38 ± 6
45 ± 3
18 ± 6
22 ± 9
14 ± 6
63 ± 4
21 ± 6
13 ± 6
16 ± 3
10 ± 5
42 ± 3
14 ± 6
25 ± 7
57 ± 6
35 ± 5
14 ± 6
19 ± 7
16 ± 7
19
3
H
ꢀ(CH
)
2
***
20
Bn
Bn
Bn
Bn
nꢀBu
nꢀBu
nꢀBu
CH
3
***
ns
*
21
Br
ꢀ(CH
)
2
22
CH
CH
Br
3
CH
3
23
3
ꢀ(CH
)
2
ns
***
*
24
CH
3
25
Br
ꢀ(CH
2
)
5
ꢀ
ꢀ
26
CH
3
ꢀ(CH
)
2
5
ns
ns
ns
27
Br ꢀ(CH
Br ꢀ(CH
Br
2
)
3
3
CO
2
2
Et
Et ꢀ(CH
CH
3
CH
3
28
2
)
CO
2
)
5
ꢀ
ꢀ
29
30
31
32
33
34
35
36
37
38
CH C≡CH
2
***
Br
CH
CH
Br
ꢀ(CH
)
2
CH C≡CH
5
2
ns
CH
3
3
CH C≡CH
3
2
*
*
*
*
ꢀ(CH
ꢀ(CH
ꢀ(CH
)
)
)
ꢀ
ꢀ
ꢀ
CH C≡CH
2
5
5
5
2
**
**
ꢀ(CH
ꢀ(CH
)
)
Cl
Cl
2
4
2
2
CH
Br
3
2
4
ns
*
ꢀ(CH
ꢀ(CH
ꢀ(CH
ꢀ(CH
)
)
)
)
Pi
Pi
Pi
Pi
CH
3
2
2
2
2
4
Br
ꢀ(CH
CH
)
ꢀ
4
4
4
2
5
ns
CH
CH
3
3
ns
ꢀ(CH
)
ꢀ
4 ± 3
3
2
5
a
+
2+
Data are mean of at least six experiments ± SEM comparing with the K ꢀinduced Ca increase
***
**
*
in absence of compounds. Compounds assayed at 1 ꢀM. p < 0.001, p < 0.01, p < 0.05, and
b
ns = not significant. Tested at 3 ꢀM. Pi = piperidinyl.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2+
Figure 2. Depolarizationꢀevoked Ca currents (I ) in bovine chromaffin cells. Wholeꢀcell I
Ca
Ca
2+
were recorded using 2 mM Ca as charge carrier and –80 mV as holding potential. Averaged
results, of at least 7 cells of independent batches, representing normalized remaining ICa in
***
**
*
presence of selected 1 derivatives at the concentration of 1 ꢀM. p < 0.001, p < 0.01, p <
0.05, comparing with control in absence of compounds (C, white bar).
Of note is the differential behavior of these compounds between the two experimental
procedures realized, i.e. the fluorescentꢀbased assay and the patchꢀclamp techniques. This could
be interpreted in base of the different expression of VGCC subtypes between SHꢀSY5Y cells,
64
where 50% of them belong to the Lꢀsubtype, and bovine chromaffin cells, which only present a
6
7
2
0% of the Lꢀsubtype. Such hypothesis raised an important point, as compound 23 blocked I
Ca
in bovine chromaffin cells by 40% (Figure 2), but it only reduced the fluorescenceꢀmonitored
2+
Ca increase in SHꢀSY5Y by 22% (Table 1), so we wondered whether compound 23 could be
ACS Paragon Plus Environment

Page 15 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
selectively blocking the non Lꢀsubtypes of VGCC, what it would be of interest taking into
account the lack of efficient pharmacological tools available to selectively interact to such nonꢀL
68
VGCC, either Nꢀ or P/Q subtypes. For this reason, we further registered the VGCC blocking
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
effect of compound 23 in presence of the selective Lꢀtype VGCC blocker nifedipine, as well as
69
in presence of the N and P/Q type VGCC blocker ωꢀconotoxin MVIIC, as shown in Figure 3.
Compound 23 exerted an additional blockade of the I in the presence of nifedipine (Figure 3A
Ca
and 3C), but on the contrary, the I observed in the presence of ωꢀconotoxin MVIIC remained
Ca
constant when compound 23 was administered (Figure 3B and 3C). Therefore, we can conclude
that compound 23 is acting as a selective nonꢀL VGCC blocker.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Page 17 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2+
Figure 3. Compound 23 partially blocks nonꢀL type Ca currents. Bovine chromaffin cells,
voltageꢀclamped at –80 mV, were challenged with 50 ms depolarizing pulses to 0 mV at 20 s
2
+
2+
intervals. Ca 2 mM was used as charge carrier. Panel A shows typical Ca currents (I ) time
Ca
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
curse in a cell perfused with nifedipine 3 ꢀM (Lꢀtype VGCC blocker) alone and in the presence
of compound 23 at 1 ꢀM. Panel B shows a typical ICa time curse in a cell perfused with ωꢀ
2
+
conotoxin (ωꢀctx) MVIIC 2 ꢀM (Nꢀ and P/Qꢀtype Ca VGCC blocker) alone and in the
presence of compound 23 at 1 ꢀM. Panel C shows the averaged peak amplitude of ICa with
respect to control (white bar) in the presence of either nifedipine or MVIIC alone (black
columns) or concomitantly with compound 23 (stripped columns). Data are means ± SEM of 16
***
cells for each treatment from 3 different cell cultures. p < 0.001, respect to cells depolarized in
#
##
absence of drugs (control group), p < 0.001, ns = not significant.
Effect of 1 derivatives 15–38 on the Ser/Thr phosphatase activity. As mentioned in the
introduction, the Ser/Thr phosphatase activity of PP1 and PP2A is impaired in neurodegeneration
36
scenarios, leading to the augmentation of the phosphorylation rate in τ protein, and thus
favoring its selfꢀaggregation to form the NFT. This reduction in the phosphatase activity is
mainly ascribed to the concourse of endogenous inhibitors that bind at the catalytic site of PPP,
40
hindering the hydrolysis of phosphate esters at Ser or Thr amino acids. The use of the PP1 and
PP2 inhibitor OA is a reliable model of these pathological processes, as it has been described that
70
it tightly binds catalytic subunits of both enzymes. Indeed, the administration of OA to rodents
71
mimicked the morphological hallmarks of AD. Hence, to appraise the effect of compounds 15–
38 on the activity of PPP enzymes, we subjected SHꢀSY5Y cells to the exposure of OA at 15
72ꢀ73
nM, an adequate concentration to inhibit both PP1 and PP2A, in the presence of the testing
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
compounds, estimating the resulting phosphatase activity by the colorimetric method of the pꢀ
nitrophenylphosphate (pNPP). The pNPP is a colorless substance that turns yellow (absorbance
74
4
05 nm) when is hydrolyzed to p-nitrophenol by the action of phosphatase enzymes. Under
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
these experimental conditions, OA diminished the total phosphatase activity by a 32%, and 12
out of 25 compounds reduced the deprival of total phosphatase activity in a statistically
significant manner, being the best compounds able to maintain such enzymatic activity a 40%
higher than with OA in absence of compounds (Table 2). Although OA interacts exclusively
with the Ser/Thr phosphatases PP1 and PP2A, due to the Tyr phosphatase activity is much more
extended in nervous system than the Ser/Thr phosphatase activity, we decided to study the effect
75
of the compounds in presence of the Tyr phosphatase inhibitor sodium orthovanadate,
administered to the SHꢀSY5Y cells at 1 mM, so the phosphatase activity recorded by the pNPP
probe would be only derived from Ser/Thr phosphatase enzymes. Pretty much similar data were
found; half of the compounds counteracted the OAꢀinduced reduction of the Ser/Thr phosphatase
activity in a statistically significant manner (Table 2). Regarding to structureꢀactivity
relationships, either no substitution at the indole nitrogen (R = H), benzyl group or a butyl group,
seem to be the most optimal alternatives as R substituents, as hit compound 1 offers a
satisfactory proꢀphosphatase activity, similarly to compounds 15, 16, 18, 20, 23, 24, 25, and 31.
Collectively, compounds 15, 23, and 25, together with the hit compound 1, featured a multitarget
2
+
activity, as they demonstrated both proꢀphosphatase and Ca channels blocking activities, in all
the experimental protocols implemented. In parallel, we evaluated the capability of the 1
derivatives 15–38 to protect against neuronal damage in in vitro models of neurodegeneration
resembling AD.
ACS Paragon Plus Environment

Page 19 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 2. Effect of 1 derivatives 15–38 on the Ser/Thr phosphatase activity depressed by
a
okadaic acid (OA)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
%
Total
%Ser/Thr
phosphatase
activity
Comp.
X
R
R’
phosphatase
activity
100
Control
OA
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
100
#
##
###
68 ± 2
61 ± 3
*
*
*
***
Memantine
ꢀ
H
ꢀ
H
ꢀ
77 ± 3
80 ± 5
76 ± 4
80 ± 4
74 ± 4
79 ± 4
71 ± 3
81 ± 4
79 ± 4
73 ± 3
77 ± 4
81 ± 6
78 ± 3
69 ± 2
76 ± 7
73 ± 2
78 ± 5
*
1
CH
3
76 ± 5
*
15
H
H
ꢀ(CH
CH
2
)
5
5
5
5
5
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
79 ± 8
**
ns
*
*
16
Br
H
3
76 ± 6
ns
17
Br
H
ꢀ(CH
CH
2
)
67 ± 3
*
1
1
2
2
8
9
0
1
CH
CH
Br
3
H
3
77 ± 4
ns
ns
H
ꢀ(CH
CH
)
69 ± 5
3
2
**
*
Bn
Bn
Bn
Bn
nꢀBu
nꢀBu
nꢀBu
78 ± 5
3
*
ns
Br
ꢀ(CH
CH
)
70 ± 7
2
ns
*
ns
22
CH
CH
Br
3
3
66 ± 5
*
2
2
2
2
3
4
5
6
3
ꢀ(CH
CH
2
)
78 ± 6
*
*
76 ± 4
3
*
**
Br
ꢀ(CH
ꢀ(CH
)
ꢀ
ꢀ
79 ± 4
2
5
ns
ns
ns
ns
CH
)
2
66 ± 6
3
5
ns
27
Br (CH
Br (CH
2
)
)
3
CO
CO
2
Et CH
3
66 ± 4
*
2
2
8
9
2
3
2
Et ꢀ(CH
2
)
5
ꢀ
ꢀ
71 ± 3
ns
ns
Br
CH
70 ± 5
77 ± 3
80 ± 4
71 ± 3
64 ± 4
CH C≡CH
3
2
**
ns
30
31
32
Br
CH
CH
ꢀ(CH
CH
ꢀ(CH
2
)
5
72 ± 5
CH C≡CH
2
*
*
74 ± 9
3
3
CH C≡CH
3
2
ns
ns
)
ꢀ
66 ± 7
CH C≡CH
2
5
2
ns
ns
ns
ns
33
Br
CH
Br
ꢀ(CH
2
)
4
Cl ꢀ(CH
2
2
)
)
5
ꢀ
ꢀ
68 ± 3
71 ± 4
73 ± 5
62 ± 7
ns
34
3
ꢀ(CH
2
)
4
Cl ꢀ(CH
5
62 ± 6
ns
35
ꢀ(CH
2
)
4
Pi
CH
3
68 ± 5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
*
ns
ns
*
3
6
Br
CH
CH
ꢀ(CH
ꢀ(CH
ꢀ(CH
2
2
2
)
)
)
4
4
4
Pi ꢀ(CH
2
)
5
5
ꢀ
ꢀ
73 ± 2
74 ± 3
72 ± 3
70 ± 4
68 ± 5
70 ± 5
ns
ns
37
3
3
Pi
CH
3
38
Pi ꢀ(CH
2
)
a
SHꢀSY5Y cells were treated with compounds at 0.1 ꢀM in presence of the PPP inhibitor
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
okadaic acid (OA) at 15 nM. Data are mean of at least four experiments ± SEM, expressed as
percentage respect to control situation, where cells were not treated with compounds neither OA.
#
##
***
**
*
p < 0.001, respect to control; p < 0.001, p < 0.01, p < 0.05, and ns = not significant,
respect to cells only treated with OA. Phosphatase activity was evaluated by the method of the
pNPP. To estimate Ser/Thr phosphatase activity, the Tyr phosphatase inhibitor sodium
orthovanadate at 1 mM was applied to the experiments. Pi = piperidinyl.
Docking studies. We selected compound 23 to exemplify the interaction between the main τ
protein phosphatase, i.e. PP2A, with these 1 derivatives, because of its significant phosphatase
activity (Table 2), as well as its neuroprotective activity counteracting the damage provoked by
OA and Glu (Table 3 and Figure 5, see below). The complex of PP2A and compound 23
revealed van der Waals interactions between the piperidine ring of compound 23 and the side
chain of residues Ile123 (d = 3.85 Å) and Val189 (d = 1.76 Å) (Figure 4B). The indole ring of
compound 23 also interacts with both the imidazole ring of His191 (d = 3.06 Å) and the indole
ring of the Trp200 (d = 4.65 Å), probably in a hydrophobic πꢀ π stacking contact (Figure 4B).
7
0
OA also interacted with Ile123, His191 and Trp200, as previously reported by Xing et al., 2006.
Also, the nitrogen of the piperidine ring of compound 23 probably makes a strong hydrogen
bond with the hydroxyl group of Ser120 (d = 1.66 Å) (Figure 4C), which contributes to the
affinity of the ligand in this binding site of PP2A (Figure 4A).
ACS Paragon Plus Environment

Page 21 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(C)
Ser120
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
.66Å
2
3
Figure 4. Docking of the compound 23–PP2A complex. (A) Pose of compound 23 (yellow) in
the PP2Ac binding site (main residues in green). (B) Van der Waals contacts (Val189 and
Ile123) and hydrophobic π interactions (His191 and Trp200) of compound 23 (yellow). (C)
Hydrogen bond between compound 23 and Ser120 (Energy scores provided in Table S1 of
supporting information).
Neuroprotection experiments. Since we had found some bioactive compounds, the raised
2+
question was whether these 1 derivatives, by regulating the neuronal Ca homeostasis and the
Ser/Thr phosphatase activity, would be able to protect neurons against neurotoxic stimuli related
to AD. We monitored cell viability with the method of the 3ꢀ(4,5ꢀdimethylthiazolꢀ2ꢀyl)ꢀ2,5ꢀ
2
+
76
diphenyltetrazolium bromide (MTT) reduction, in in vitro neuronal models subjected to Ca
ACS Paragon Plus Environment

Page 23 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
overload or τ hyperpolarization. Since VGCC and Ser/Thr phosphatases, targeted by our 1
derivatives, present highlighted physiological roles, we firstly assessed the potential per se
toxicity of compounds 15–38 in the neuronal models object of study, i.e. cortical motor neurons
and SHꢀSY5Y neuroblastoma cells. As graphed in Figures S2 and S3 of supporting info, none of
the compounds, incubated for 48 h, reduced the cell viability of SHꢀSY5Y cells at concentrations
lower than 10 ꢀM. In the case of cortical motor neurons, compounds 15 and 36 marginally
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
reduced cell viability when incubated for 48 h at 1 ꢀM (Figures S4 and S5, supporting info).
2+
In order to emulate a Ca overload, cortical motor neurons were exposed to veratridine 20 ꢀM, a
+
voltageꢀgated Na channels (VGNC) ligand that delays their inactivation, leading to a
2+
7
7
depolarizationꢀelicited Ca entry. After 24 h preꢀincubation of the compounds, at 1 ꢀM,
neurons were coꢀincubated for another 24 h with veratridine (20 ꢀM), inducing a loss of viability
of about 50% (Table 3). More than a half of the compounds, but not 1, reduced the veratridineꢀ
induced damage in a significant manner. Compounds 23, 25, and 26 stood out by rescuing the
neuronal viability by a 50% or more. In these experiments, the presence of piperidine at C3
instead of dimethylamine seems to favor this neuroprotective activity. Again, benzyl and butyl
groups branched to the indole nitrogen afforded the best neuroprotection (Table 3). The VGNC
78
blocker tetrodotoxin (TTX) was used as standard in these experiments.
In parallel, we tested the neuroprotective activity of compounds 15–38 against a model of τ
hyperphosphorylation, by stimulating SHꢀSY5Y neuroblastoma cells with the PPP selective
inhibitor OA. SHꢀSY5Y cells suffered a fall of 35% in their viability when exposed for 20 h to
OA 20 nM. Under this situation, most of compounds noticeably mitigated the reduction in the
cell viability caused by OA, and 10 of them did it by 50% or more, being the best compound 25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
79
(
Table 3). Melatonin was used as standard in these experiments. Among all the structureꢀ
activity relationships considered, we appreciated outstanding activities when the compounds
present benzyl, butyl or no substitution at the indole nitrogen.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 3. %Neuroprotection of 1 derivatives 15–38 on the cell viability of cortical motor
neurons and SH-SY5Y neuroblastoma cells damaged with veratridine and okadaic acid,
a
respectively
b
c
Comp.
TTX
X
R
R’
Veratridine Okadaic acid
*
**
ꢀ
ꢀ
ꢀ
77 ± 1
ꢀ
ꢀ
*
**
Melatonin
1
ꢀ
H
ꢀ
H
ꢀ
43 ± 3
45 ± 2
47 ± 5
47 ± 4
60 ± 5
59 ± 4
38 ± 3
56 ± 3
54 ± 6
44 ± 4
56 ± 5
66 ± 5
71 ± 3
46 ± 5
58 ± 2
43 ± 5
ns
***
**
**
**
**
***
***
**
*
CH
0 ± 4
3
ns
1
1
1
5
6
7
H
H
ꢀ(CH
CH
2
)
5
5
5
5
5
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
18 ± 7
11 ± 5
19 ± 3
30 ± 5
ns
Br
H
3
*
*
Br
H
ꢀ(CH
CH
)
2
18
CH
CH
Br
3
H
3
ns
1
2
2
2
9
0
1
2
3
H
ꢀ(CH
CH
2
)
7 ± 3
*
**
Bn
Bn
Bn
Bn
nꢀBu
nꢀBu
nꢀBu
34 ± 3
36 ± 6
35 ± 9
51 ± 6
34 ± 5
72 ± 7
60 ± 7
42 ± 6
36 ± 7
3
*
*
*
*
Br
ꢀ(CH
CH
)
2
CH
CH
Br
3
3
*
**
**
***
*
23
3
ꢀ(CH
)
2
**
24
CH
3
*
*
*
*
**
25
Br
ꢀ(CH
ꢀ(CH
2
)
5
ꢀ
ꢀ
**
2
2
2
6
7
8
CH
3
)
2
5
**
*
Br (CH
Br (CH
)
)
CO
CO
Et CH
3
2
3
2
Et ꢀ(CH
)
ꢀ
2
3
2
2
5
ns
ns
29
Br
CH
3
17 ± 5
12 ± 4
CH C≡CH
2
ACS Paragon Plus Environment

Page 25 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
*
**
ns
ns
3
3
3
0
1
2
Br
CH
CH
ꢀ(CH
CH
ꢀ(CH
2
)
5
ꢀ
ꢀ
24 ± 4
47 ± 3
12 ± 3
42 ± 8
CH C≡CH
2
ns
14 ± 10
3
3
CH C≡CH
3
2
ns
)
5
19 ± 5
CH C≡CH
2
2
*
*
3
3
3
4
Br
CH
Br
ꢀ(CH
2
)
4
Cl ꢀ(CH
2
2
)
)
5
ꢀ
ꢀ
39 ± 6
36 ± 5
21 ± 9
48 ± 8
56 ± 6
31 ± 3
61 ± 4
37 ± 1
26 ± 3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
**
ns
*
ꢀ(CH
)
4
Cl ꢀ(CH
3
2
5
**
***
***
**
35
ꢀ(CH
ꢀ(CH
ꢀ(CH
ꢀ(CH
2
2
2
2
)
4
)
4
)
4
)
4
Pi
CH
3
ns
ns
ns
3
3
3
6
7
8
Br
Pi ꢀ(CH
2
)
5
ꢀ
ꢀ
0 ± 5
4 ± 6
1 ± 5
CH
CH
Pi
CH
3
3
Pi ꢀ(CH
)
5
3
2
a
Cell viability was measured by the method of the MTT reduction. Data are expressed as the
percentage of cell viability recovered in cells treated with the toxic stimulus plus compounds,
comparing with viability of cells only exposed to veratridine or OA; mean of at least 5
*
**
**
*
experiments ± SEM in triplicate. p < 0.001, p < 0.01, p < 0.05, and ns = not significant,
b
respect to cells only treated with veratridine or OA. Cortical motor neurons exposed to
veratridine 20 ꢀM in presence of compounds at 1 ꢀM, except for compound 36 which was tested
c
at 0.1 ꢀM due to its toxicity at 1 ꢀM (see Figure S5 supporting info). SHꢀSY5Y neuroblastoma
cells exposed to OA 20 nM in presence of compounds at 0.1 ꢀM. Pi = piperidinyl.
Gathering all the pharmacological data obtained, we proposed that these new 1 derivatives are
multitarget neuroprotective compounds able to interact with two pathophysiological events
2
+
involved in AD, i.e. Ca dishomeostasis and Ser/Thr phosphatases inhibition, and in turn
entailing a neuroprotective profile, according to the in vitro models of neurodegeneration we
have performed. Overall, it can be extrapolated that the presence of benzyl or butyl groups at the
indole nitrogen along with a piperidine at C3, preferentially offer privileged structures of 1
derivatives, according to the pharmacological activities object of study, and their consequent
neuroprotective properties over ADꢀresembling neurodegeneration models. However,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
neuroprotective compounds historically fail in their pharmacological actions when they are
evaluated in more physiological models of experimental study. Therefore, we carry out a more
physiological model of biological evaluation, defined by the use of rat hippocampal slices
subjected to toxic stimuli related to AD, recording the ability of selected compounds to
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
7, 50, 63, 80
counteract the tissue damage.
In this preparation, neurons are surrounded by their natural
environment, e.g. glia cells, extracellular matrix, etc. Rat hippocampi were subjected to a
sustained exposition to the excitatory neurotransmitter glutamate, which is considered a close
81
model to the excitotoxicity occurred in neurodegeneration. In addition, recent contributions
2+
have hypothesized that PP2A can downꢀregulate the Ca influx through the NMDA receptors, as
dephosphorylation of NR1A subunit by this phosphatase accelerates the desensitization rate of
4
8, 82ꢀ83
the NMDA receptor.
Hence, this model has proven to be an excellent experimental protocol
2+
to test the neuroprotective activities of either Ca ꢀantagonists or Ser/Thr phosphatase activators.
Tested compounds were selected according to optimal biological activities and neuroprotective
properties in vitro, as well as chemical diversity (Figure 5).
ACS Paragon Plus Environment

Page 27 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. Effect of selected 1 derivatives on the viability of rat hippocampal slices damaged with
glutamate (Glu) 1 mM, measured through the MTT reduction. %Protection respect to the
viability of hippocampal slices treated with Glu in absence of compounds. Data are expressed as
***
**
*
mean ± SEM of at least 4 different tissue batches in quadruplicate. p < 0.001, p < 0.01, p <
0.05, respect to slices only treated with Glu.
2
+
Compounds 21, 23, and 30 that had evinced regulating activities of both Ca entry and Ser/Thr
phosphatases, and protected neurons in both veratridine and OAꢀbased models of
neurodegeneration, avoided the glutamateꢀinduced loss of viability of the hippocampal slices by
30% or more, however the multitarget and in vitro neuroprotectant compound 25 failed to protect
such preparations. Compounds 16, 18, and 24, behaved as Ser/Thr phosphatase activators (Table
2), and had a proper in vitro neuroprotection profile. Nevertheless, in this model, only
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 72
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
compounds 16 and 24 reasserted that neuroprotective properties, as they reduced the damage by
glutamate by 40% or more (Figure 5). Compound 34 also protected hippocampi from the
glutamate excitotoxicity, according to its in vitro neuroprotection properties (Table 3), although
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2+
2+
it had only targeted the cell Ca signal (Table 1). Compound 27 protected against both Ca
overload and τ hyperphosphorylation, but it did not show any activity on either depolarizationꢀ
2+
evoked Ca entry or Ser/Thr phosphatase. In addition, it substantially protected hippocampal
slices against the glutamate exposure (Figure 5); one possibility is that compound 27 would be
affecting other pathways involved in the neuronal death/survival scenarios.
Conclusions
This work illustrates an alternative approach for a multitarget strategy for the treatment of AD,
2
+
by devising a double interaction with neuronal Ca signaling and Ser/Thr phosphatase activity.
2
+
Regulation of cytosolic Ca concentrations by different VGCC blockers has been a
pharmacology target widely studied for several neurodegenerative diseases, but few
contributions has selected it as a part of a multitargetꢀbased drug development, which have
mostly inclined towards the complement of cholinesterases inhibition with Aβ downꢀregulation
or τ hyperphosphorylation reduction strategies, affording unfruitful results. Regarding to the τ
hyperphosphorylation, the research works have almost exclusively focused on the development
of protein τ kinase inhibitors. We propose that the activation, or the reduction of inhibitory
actions, over protein τ phosphatase enzymes, deserves more attention, overcoming the historical
32
prejudice of their unspecificity, currently rebutted. Of note is that, when we recorded the
phosphatase activity by the method of the pNPP (Table 2), a substantial decrease of 70% was
ACS Paragon Plus Environment

Page 29 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
ꢀ
computed in cells exposed to the Tyr phosphatase inhibitor VO , what confirms that Tyr
3
phosphatase activity is much more extended than Ser/Thr phosphatase activity in nervous
system. The fact that τ protein is mainly dephosphorylated by PP1 and PP2A entails the
advantage that Ser/Thr phosphatase activators would not present such sizeable offꢀtarget
pharmacology as Tyr phosphatase activators would. To finish up, we hypothesize that compound
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
3 is an eligible lead compound to further investigate its pharmacological activities in preclinical
models of AD, as it demonstrated significant activity over VCGG and Ser/Thr phosphatases,
accompanied by a wideꢀspectrum neuroprotective profile. The fact that compound 23 exerts Ca
blockade preferentially through nonꢀL type VGCC is in agreement to the rising interest of
2
+
68, 84
developing both Nꢀ and P/Q type VGCC blockers for dementia,
what supposes a
2
+
breakthrough in the study of Ca antagonists for central nervous system diseases because they
2
5ꢀ
could bypass the cardiovascular side effects resulting from peripheral Lꢀtype channel blockade,
26
Hence, the derivative 1ꢀbenzylꢀ5ꢀmethylꢀ3ꢀ(piperidinꢀ1ꢀylmethyl)ꢀ1Hꢀindole (compound 23,
which we have named ITH12657) features an innovative multitargetꢀbased mechanism of action,
2
+
aiming neuronal Ca control and Ser/Thr phosphatase activity.
Experimental Section
General Procedures. 5ꢀBromoindole, 5ꢀmethylindole, 1, compound 18, and general reagents
were purchased from SigmaꢀAldrich (Madrid, Spain). Solvents were purchased from VWR
(
Barcelona, Spain). Reactions were monitored by thin layer chromatography aluminum oxide on
TLCꢀPET foils (SigmaꢀAldrich). Detection was made with UV light at 254 nm. All the reactions
were subjected to Schlenk conditions (vacuum purges and argon atmosphere). Preꢀcharged basic
ACS Paragon Plus Environment