Design, synthesis and evaluation of pentacycloundecane and hexacycloundecane propargylamine derivatives as multifunctional neuroprotective agents

Accepted Manuscript

Design, synthesis and evaluation of pentacycloundecane and hexacycloundecane

propargylamine derivatives as multifunctional neuroprotective agents

Frank T. Zindo, Sarel F. Malan, Sylvester I. Omoruyi, Adaze B. Enogieru, Okobi E.

Ekpo, Jacques Joubert

PII:

S0223-5234(18)31010-9

DOI:

https://doi.org/10.1016/j.ejmech.2018.11.051

EJMECH 10909

Reference:

To appear in: European Journal of Medicinal Chemistry

Received Date: 12 October 2018

Revised Date: 9 November 2018

Accepted Date: 21 November 2018

Please cite this article as: F.T. Zindo, S.F. Malan, S.I. Omoruyi, A.B. Enogieru, O.E. Ekpo, J. Joubert,

Design, synthesis and evaluation of pentacycloundecane and hexacycloundecane propargylamine

derivatives as multifunctional neuroprotective agents, European Journal of Medicinal Chemistry (2018),

doi: https://doi.org/10.1016/j.ejmech.2018.11.051.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to

our customers we are providing this early version of the manuscript. The manuscript will undergo

copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please

note that during the production process errors may be discovered which could affect the content, and all

legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Graphical Abstract

NH

O

NH

O

NGP1-01

12

P1

Neuroprotective

VGCC blocker

Neuroprotective

VGCC blocker

NMDAr antagonist

MAO-B inhibitor

ACCEPTED MANUSCRIPT

Design, synthesis and evaluation of pentacycloundecane and

hexacycloundecane propargylamine derivatives as multifunctional

neuroprotective agents

Frank T. Zindo^a, Sarel F. Malan^a, Sylvester I. Omoruyi^b, Adaze B. Enogieru^b, Okobi E.

Ekpo^b, Jacques Joubert^a*

^aPharmaceutical Chemistry, School of Pharmacy, University of The Western Cape, Private

Bag X17, Bellville 7535, South Africa

^bDepartment of Medical Biosciences, University of the Western Cape, Private Bag X17, Cape

Town, Bellville 7535, South Africa.

*Author for correspondence: jjoubert@uwc.ac.za; Tel.: +27-21-959-2195

Abstract

The multifactorial pathophysiology of neurodegenerative disorders remains one of the main

challenges in the design of a single molecule that may ultimately prevent the progression of

these disorders in affected patients. In this article, we report on twelve novel polycyclic

amine cage derivatives, synthesized with or without a propargylamine function, designed to

possess inherent multifunctional neuroprotective activity. The MTT cytotoxicity assay results

showed the SH-SY5Y human neuroblastoma cells to be viable with the twelve compounds,

particularly at concentrations less than 10 µM. The compounds also showed significant

neuroprotective activity, ranging from 31% to 66% at 1 µM, when assayed on SH-SY5Y

human neuroblastoma cells in which neurodegeneration was induced by MPP⁺. Calcium

regulation assays conducted on the same cell line showed the compounds to be significant

VGCC blockers with activity ranging from 26.6% to 51.3% at 10 µM; as well as significant

NMDAr antagonists, with compound 5 showing the best activity of 88.3% at 10 µM. When

assayed on human MAO isoenzymes, most of the compounds showed significant inhibitory

activity, with compound 5 showing the best activity (MAO-B: IC₅₀= 1.70 µM). Generally,

the compounds were about 3 – 52 times more selective to the MAO-B isoenzyme than the

MAO-A isoenzyme. Based on the time-dependency studies conducted, the compounds can be

defined as reversible MAO inhibitors. Several structure activity relationships were derived

from the various assays conducted, and the compounds’ possible putative binding modes

within the MAO-B enzyme cavity were assessed in silico.

Key words: Neurodegeneration, apoptosis, neuroprotection, multifunctional, polycyclic

amine, propargylamine, excitotoxicity, monoamine oxidase.

Introduction

Neurodegenerative disorders (NDs) refers to a cluster of neuronal diseases that are

characterised by transient and irreversible loss of neuronal cells in the brain due to apoptosis

[1]. Examples of these diseases include Alzheimer’s Disease (AD), Huntington’s Disease

(HD) and Parkinson’s Disease (PD). Though the pathogenesis is yet to be fully understood, it

is known that several factors are involved in the etiology of these disorders. Some of the

established theories of disease causation include, but are not limited to, disturbances in

ACCEPTED MANUSCRIPT

neurotransmitter systems such as the monoaminergic system [2,3], as well as excitotoxicity

resulting from excessive calcium influx into neuronal cells [4,5].

Patients with PD and AD have been shown to have age-related elevated levels of the B-type

monoamine oxidase isoenzyme (MAO-B) in the brain [6]. This enzyme does not only act

indirectly as a trigger to the apoptotic process in neuronal cells, but together with the A-type

isoenzyme (MAO-A), they form the major catabolic pathways of dopamine [7] and give rise

to some of the signs and symptoms associated with these disorders [8]. Both these

isoenzymes are therefore promising drug target sites for the treatment of NDs and as such,

monoamine oxidase enzyme inhibitors (MAOI’s) have been the mainstay therapy for the

management of NDs. By inhibiting the activity of the MAO isoenzymes, MAOI’s may exert

neuroprotective effects by inhibiting the formation of toxic by-products of MAO-catalyzed

oxidation of neurotransmitters [9]. Further to this, MAOI’s enhance dopaminergic

neurotransmission in the nigro-striatal pathway, thereby providing symptomatic relief in

patients with PD [10].

The first generation of MAOI’s such as tranylcypromine are nonspecific and irreversible

inhibitors of both isoforms of MAO. While these dual inhibiting compounds produce a

significant rise in dopamine levels and behavioural changes after administration [11], they

present a drawback of causing clinically significant potentiation of the ‘pressor response’, a

common adverse effect associated with ingesting tyramine containing foods together with

nonspecific MAOI’s. Potentiation of this adverse effect is mainly influenced by the

irreversibility of the inhibitor and the degree of MAO-A inhibition [12]. This has been shown

to be due to MAO-A being the major form of MAO in the liver and stomach. Reversible

inhibitors of MAO-A are therefore preferred to their irreversible counterparts as they show a

reduced hypertensive response [13].

A more widely explored strategy entails selectively inhibiting the MAO-B isoenzyme by

inhibitors such as rasagiline and selegiline (Figure 1). This approach is based on the fact that

the extrapyramidal region of the human brain has approximately 4-times more MAO-B

isoenzyme than the MAO-A isoenzyme [14]. Rasagiline and selegiline are second generation

propargylamine derivatives that irreversibly inhibit brain MAO-B, and have promising

neuroprotective activities [15]. Their activity can be attributed to the propargyl moiety [8],

that after oxidation, reacts with the flavin prosthetic group in the active site of the MAO-B

enzyme, forming a covalent adduct at the N5 position of the flavin adenine dinucleotide

(FAD) [16]. The propargyl moiety is also now known to play an important role in providing

neuronal and mitochondrial protective properties [17] as well as anti-apoptotic properties

[18]. This moiety has been particularly useful for incorporation into multi-target compounds

with inherent MAO-B inhibitory capacity [19].

While inhibiting the oxidative deamination reaction catalysed by the MAO isoenzymes is a

useful strategy in managing NDs, targeting a single enzymatic system or receptor has proved

to be insufficient for the treatment of these multifactorial diseases [20]. A more effective

therapy would result from the use of multi-target directed ligands (MTDLs) able to intervene

in the different pathological events implicated in the etiology of neuronal disorders [21].

ACCEPTED MANUSCRIPT

Calcium homeostasis in neuronal cells has also been implicated in the pathogenesis of NDs.

Given the crucial and widespread role calcium signalling has in excitable cells, it is not

surprising that alterations in calcium homeostasis is linked to several NDs [22,23].

Maintenance of intracellular calcium homeostasis is essential for the functioning and survival

of neurons and is a fundamental component of synaptic transmission for both pre- and

postsynaptic mechanisms [24]. Excessive influx of calcium can therefore overwhelm Ca²⁺-

regulatory mechanisms and lead to excitotoxicity and neuronal cell death [25]. The activation

of the postsynaptic N-methyl-D-aspartate receptors (NMDAr), 2-amino-3-(3-hydroxy-5-

methylisoxazol-4-yl)proprionate receptors and kainate receptors allows for opening of their

associated ion channels to allow the influx of Ca²⁺and Na⁺ions into the neuronal cells.

Calcium entry may also occur through L-type voltage gated calcium channels (VGCC) and

result in both calcium overload and mitochondrial disruption [25]. This mechanism of cell

death suggests these receptors and their associated calcium channels to be supplementary

drug target sites for the treatment of NDs.

N

NH

O

Rasagiline

NGP1-01

Selegiline

polycyclic scaffold

benzylamine

propargylamine

NH

OH

N

NH

O

12

P1

P2

Figure 1. The molecular structures of the selective MAO-B inhibitors selegiline and rasagiline, the VGCC

blocker and NMDAr antagonist NGP1-01, compounds from a previous study P1 and P2, and the new series

represented by compound 12, with the polycyclic scaffold, benzylamine and propargylamine moieties

highlighted.

In our previous work [26], we designed and synthesized polycyclic propargylamines,

represented by P1 and P2 (Figure 1), to serve as multifunctional drug ligands by inhibiting

MAO-B enzyme activity, and as regulators of cytosolic calcium entry predominantly

mediated by NMDAr and VGCC. While these compounds showed good in vitro anti-

apoptotic activity, they showed little to no MAO-B inhibitory activity and were inactive as

NMDAr antagonist and VGCC blockers. These findings suggested that P1 and P2 exhibit

their neuroprotective effect through some other mechanism(s) implicated in the complex

etiology of neurodegeneration unexplored in the study. We further postulated that the lack of

calcium regulatory activity in P1 and P2 was due to the absence of a benzylamine moiety

within their structures. This moiety, which is present in structure of NGP1-01, a molecule

known to have neuroprotective properties through VGCC blockade and NMDAr antagonistic

activity [27], seems to be vital for calcium modulatory activity in such polycyclic analogues.

ACCEPTED MANUSCRIPT

It is for this reason that we incorporated a benzylamine moiety in the design of a new series

of compounds reported in this current study (represented by compound 12, Figure 1).

Further investigation to clarify the lack of MAO-B inhibitory activity in these compounds

was carried out by performing computer-assisted simulated docking in the MAO-B enzyme

(PDB ID: 2V5Z) [28], utilizing Molecular Operating Environment (MOE) as previously

described [29]. The best-ranked docking solutions showed that compound P1 and P2 occupy

only the entrance cavity of the MAO-B enzyme and barely access the substrate cavity to form

the necessary binding interactions with the FAD cofactor, which could have resulted in

increased activity of these compounds (Figure 2). With the inclusion of a benzylamine moiety

in the structures of the new series of compounds, as seen in compound 12 (Figure 1 and

Figure 2), we envisaged better MAO-B inhibitory activity as this moiety significantly

elongates the compound allowing the adjacent propargylamine moiety to come in close

proximity to the FAD cofactor of the enzyme.

Figure 2. Schematics showing inactive compound P1 (left) and a representative of the new polycyclic

propargylamine derived series compound 12 (Right) in a computer simulated MAO-B enzyme cavity. The

compounds are shown in yellow and the FAD co-factor in green. Note that compound P1 merely occupies the

entrance cavity of the enzyme pocket and remains distant from the FAD co-factor rendering it inactive, while

the new series traverses deeper into the enzyme cavity to come in close proximity with the FAD co-factor

allowing for potential binding interactions which may result in improved MAO-B activity.

The findings from our previous work have thus inspired the design, synthesis and evaluation

of a new series of twelve pentacycloundecane and hexacycloundecane derivatives,

synthesized with or without a propargylamine function. The compounds in this series,

represented by compound 12 (Figure 1 and Figure 2), carry some or all of the following

features; (a) a polycyclic cage scaffold - for side-chain attachment as well as for improving

the drug’s lipophilicity [30] and enhance drug transport across cellular membranes, including

the selectively permeable blood–brain barrier, and also to increase drug affinity for lipophilic

regions in target proteins [30,31]; (b) a benzylamine moiety - as present in the structure of

NGP1-01, for improved VGCC blockade and NMDAr antagonism [27]; (c) a propargylamine

moiety - for inherent MAO inhibitory capacity and to provide neuronal and mitochondrial

protective properties [17], as well as anti-apoptotic properties [18]; (d) an elongated

orientation – resulting from the inclusion of a benzylamine moiety, to promote molecular

interaction between the propargylamine function and the FAD cofactor of the MAO enzyme.

Compounds that show such multi-mechanistic activity may have promising potential to

curtail the multifactorial etiology of neurodegenerative disorders.

The novel compounds were therefore synthesized and screened for cytotoxicity,

neuroprotection, NMDAr antagonism and VGCC blockade by performing in vitro assays on

ACCEPTED MANUSCRIPT

human neuroblastoma SH-SY5Y cells. Further to this, in vitro assays on the human MAO–A

and MAO–B enzymes were performed in order to determine their inhibitory potential on the

respective isoenzymes.

Chemistry

Each compound was synthesised to evaluate the activity and benefit of the presence of certain

functional groups in the molecule. These groups included the following; a terminal

propargylamine (5, 6, 9 and 12), a terminal secondary amine (1, 2, 4 and 10), as well as the

ketal (1 - 6), aza (7 - 9) and oxa (10 - 12) variations of the polycyclic cage scaffold. The

common starting Cookson’s diketone (a), was prepared according to the method of Cookson

et al. [32] From the Cookson’s diketone, the monoprotected ketal polycyclic cage (b) was

prepared according to the method of Dekker et al. [33] NGP1-01 was synthesized according

to an adaptation of the microwave assisted method (MWAM) described by Joubert et al. [34]

To synthesize the mono amine cage (c), debenzylation of NGP1-01 was carried out under

high-pressure catalytic hydrogenation as reported by Marchand et al. [35] (Scheme 1).

Compounds 1 – 12, were subsequently synthesized by direct conjugation of various

analogues to the three polycyclic cage scaffolds (a – c) using microwave irradiation.

MWAMs present several advantages which include; remarkable reduction of reaction time,

improved yields, cleaner reactions and reduction or elimination of hazardous solvents

compared to reactions performed under conventional thermal heating conditions [36,37]. As

such, MWAMs have since been adopted by several researchers as the preferred strategy for

the synthesis of these cage-derived organic compounds [34,38].

The synthesis of compounds 1, 2 and 3 was performed by conjugating their respective

primary diamines to the mono ketal polycyclic cage (b) through a direct amination reaction

using microwave irradiation (Scheme 2). An excess amount of p-xylylenediamine (5.0 equiv)

was used for the synthesis of compound 2 in order to promote the formation of a mono-

substituted amine, however, a significant percentage of the di-substituted derivative 3 formed

during the reaction and was successfully isolated giving respective yields of 32% and 13%.

The serendipitously synthesized compound 3 was also included in the series of compounds

for screening of potential activity. Propargyl chloride was conjugated to compound 2 using

MWAMs through an S_N2 nucleophilic substitution reaction, in the presence of K₂CO₃, to

produce compound 5. To synthesize compounds 4 and 6, reductive amination of the imines 2

and 5 was carried out using NaBH₄. A comparison of the biological activities of these four

compounds provides insight on the influence of the imine bond on their activity. The

reduction of the imines 1 and 2 with NaBH₄, followed by acid hydrolysed transanular

cyclization using HCl, gave the desired aza-bridged compounds 7 and 8 respectively.

Propargyl chloride was conjugated to compound 8 through a nucleophilic S_N2 substitution

reaction using MWAMs to yield compound 9 (Scheme 2).

Compound 10, which is an oxa analogue and structural isomer of 7, was synthesized by

conjugating 4-amino benzylamine to the Cookson’s diketone (a), followed by reduction with

NaBH₄(Scheme 3). A microwave assisted nucleophilic S_N2 substitution reaction was

employed to conjugate the mono amine cage (c) to p-xylylene dichloride to yield compound

ACCEPTED MANUSCRIPT

11 and to conjugate propargyl chloride to 11 and yield final compound 12. (Scheme 3)

Compounds 1, 7 and 10 were synthesized to serve as intermediates to potential subsequent

propargyl derivatives. However, due to the low reactivity of the aniline amine, reactions with

halopropargyls in the form of propargyl chloride and propargyl bromide could not yield the

desired derivatives. All compounds (1 – 12) were protected from light after synthesis and

were appropriately stored at -84 °C to prevent degradation.

O

i

O

+

NH₂

O

v

O

c

O

UV

iii & iv

ii

HO

O

H₂N

+

NH

OH

O

a

b

NGP1-01

Scheme 1: Reagents and conditions for the synthesis of diketone (a), ketal (b) and monoamine (c) polycyclic scaffolds: (i)

benzene, 0 °C, 1 h, 74%; (ii) benzene, p-TsOH (cat), Dean–Stark reflux, 5 h, 72%; (iii) ethanol, MW, 80 °C, 100 W, 100 psi,

2 h, quantitative yield; (iv) ethanol, NaBH₄,rt, 8 h, 62%; (v) ethanol, 10% Pd/C, H₂, 206 kPa, 50 °C, 14 h, 32%.

H₂N

ii

i (a or b)

NH₂

O

Cl

O

+

O

N

b

( )

1_{(n = 0)}, 2_{(n = 1)}

5

n

( )_nNH₂

NH

iii

iv

i (c)

O

NH

O

NH

N

4_{(n = 1)}

6

_nNH₂

( )

N

NH

3

O

v

vi

Cl

OH

N

OH

N

+

9

7

_{(n = 0)}, 8_{(n = 1)}

( )

n

NH

NH₂

Scheme 2: Reagents and conditions for the synthesis of ketal- and aza-polycyclic derivatives (compounds 1 – 9): (i, a: n = 0)

ethanol, MW, 80W, 150 psi, 100 °C, 30 min, 65%; (i, b: n = 1) ethanol, rt 1 h, then MW, 60 W, 80 psi, 100 °C, 3 h, 32%; (i,

c) ethanol, rt 1 h, then MW, 60 W, 80 psi, 100 °C, 3 h, 13%; (ii) acetonitrile, K₂CO₃, 60 °C, 4 h, 21%; (iii) ethanol, NaBH₄,

rt, 8 h, 72%; (iv) ethanol, NaBH₄, rt, 8 h, 69%; (v) acetone, 4 M HCl, rt, 12 h, 42% and 78%; (vi) acetonitrile, K₂CO₃, MW,

60 W, 60 °C, 2.5 h, 18%.

ACCEPTED MANUSCRIPT

i

ii

H₂N

OH

NH

+

O

NH₂

O

NH₂

O

10

NH₂

a

Cl

iv

iii

NH

+

NH₂

O

Cl

NH

Cl

12

11

c

Scheme 3: Reagents and conditions for the synthesis of oxa-polycyclic derivatives (compounds 10 – 12): (i) ethanol, MW,

100 W, 100 psi, 80 °C, 2 h; (ii) ethanol, NaBH₄, rt, 8 h, 32%; (iii) acetonitrile, K₂CO₃, MW, 150 W, 20 psi, 70 °C, 1 h, 19%;

(iv) acetonitrile, K₂CO₃, MW, 150 W, 20 psi, 70 °C, 1 h, 21%.

1

Characterization of all the compounds was carried out by means of H-, ¹³C-NMR, IR and

HREI-mass spectra. A common structural feature in all the synthesized compounds was the

polycyclic cage moiety. This structural moiety showed characteristic signal peaks on the ¹H-

NMR spectra which included a characteristic AB quartet signal, due to the two

unsymmetrical protons on the bridgehead. This signal appeared at a chemical shift in the

range of δ 1.10–1.91 ppm, with a coupling constant in the range of 10.6–10.8 Hz. The rest of

the protons making up the polycyclic cage appeared as multiplets or groups of multiplets,

apparent quartets, and apparent triplets in the range of δ 2.30–3.00 ppm. These signals

confirmed the presence of the polycyclic scaffold in the compound structures [38].

All ketal derivatives (1 – 6) showed a characteristic multiplet signal in the range of δ 3.70–

3.95 ppm, owing to the four protons on the ketal carbons. In compounds were the polycyclic

cage scaffold was linked to the rest of the molecule via a rigid imine bond (1, 2, 3 and 5), the

ketal moiety seemed to influence the signal of the two protons of the -N-CH₂- linker by virtue

1

of long-range coupling. Due to this coupling, the two protons appeared on the H-NMR

spectra as apparent doublet of doublets at about δ 4.30–4.60 ppm, with a coupling constant of

3.2 Hz. When the imine bond was reduced to a rotatable amine bond, in the case of 4 and 6,

this coupling effect was lost and the signal appeared as a multiplet at an upfield shift of about

δ 3.95 ppm.

A typical signal for the presence of the aza-cage moiety (7 – 9), was a triplet at a chemical

shift in the range of δ 3.21–3.95 ppm, with coupling constants in the range of 4.8–5.2 MHz.

This triplet corresponded to the single proton at the -CH-N- position, and its multiplicity is

attributed to the presence of two protons at the adjacent carbons atoms within the polycyclic

cage [38]. In the case of the oxa-cage moiety (10 – 12), this triplet signal was at a range of

about δ 4.65–4.73 ppm with a coupling constant in the range of 4.8–5.0 Hz. The downfield

shift of this methine hydrogen resulted from the deshielding effect of the adjacent oxygen

atom. This downfield shift, compared with that of its aza analogues, is attributed to the

electronegativity difference between nitrogen and oxygen, and thus, an increased deshielding

effect compared to the aza derivatives.

ACCEPTED MANUSCRIPT

The presence of a propargyl moiety in the structures of compounds 5, 6, 9 and 12 was

1

confirmed by a characteristic H-NMR triplet at a chemical shift in the range of δ 2.20–2.30

ppm, with a coupling constant of about 2.4 Hz. This triplet corresponds to the terminal

acetylene proton and its multiplicity is attributed to the neighbouring methylene protons

which influence it via long-range coupling. The terminal acetylene proton equally influenced

the two symmetrical methylene protons, which form part of the propargyl moiety, to appear

as a doublet in the range of δ 3.40–3.50 ppm, with a coupling constant of about 2.4 Hz.

13

Further characterization of these compounds was carried out by means of C-NMR, IR and

HREI-mass spectra, which all supplemented the ¹H-NMR spectra findings, further

confirming the structures of the compounds.

Cytotoxicity

Compounds 1-12 were assayed for cell viability on SH-SY5Y human neuroblatoma cells

using a standard 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide (MTT) assay

method [39]. The results are depicted in Figure 3 and Table 1. This assay, which measures

cell metabolic activity, was used to determine the cytotoxicity of the test compounds at 10

µM, 50 µM and 100 µM concentrations. The cells were exposed to test compounds for 48

hours, after which cell viability was determined spectrophotometrically. Vehicle control cells

were treated with dimethyl sulfoxide (DMSO, solvent for dissolving test compounds) and

served as a reference for 100% cell viability. NGP1-01, known to have neuroprotective

properties [27], was also used as a reference control for relative comparison.

150

100

*

10 µM

50 µM

100 µM

*

50

0

*

Figure 3. Percentage cell viability of compounds 1-12 assessed by means of measuring the metabolic activity of

SH-SY5Y neuroblatoma cells, after 48 h exposure to test compound, relative to a control of untreated viable

cells. Data are mean ± SEM (n = 3, four fields per repeat). Data were subjected to an ANOVA statistical

analysis and significance was defined as [(*) p < 0.05, (**) p < 0.001, (***) p < 0.0001].

The results from this assay suggested the SH-SY5Y human neuroblastoma cells to be viable

with the test compounds, particularly at 10 µM, as none of the compounds showed

ACCEPTED MANUSCRIPT

statistically significant cytotoxicity on the cell line at this concentration (Figure 3).

Percentage cell viability at this concentration ranged from 84% to 116% relative to the

untreated control. This was comparable to NGP1-01, which was found to be 86% viable at

the same concentration. The influence of the propargyl moiety was evident at higher

concentrations of test compound as seen when comparing the propargyl containing

compound 5, which was 99% viable at 100 µM, with its precursor devoid of this moiety,

compound 2, which was only 32% viable at the same concentration. This observation

suggests compound 2 to be 3-times more toxic than its propargyl containing derivative. The

same observation was made when comparing compound 6, which was 70% viable at 100 µM,

to its precursor compound 4, which was only 5% viable at the same concentration, suggesting

the inclusion of a propargyl moiety to be responsible for a 14-fold decrease in cytotoxicity.

Neuroprotection

The SH-SY5Y cytotoxicity assay results were used to determine the test concentration of

compounds 1-12 which would insignificantly affect the viability of the SH-SY5Y cells.

Based on the SH-SY5Y cytotoxicity analysis, it was decided to conduct the neuroprotection

studies at test concentrations between 1 µM and 10 µM, in order to maintain cell viability.

In this study, we employed a widely used cellular model in which neurotoxicity was induced

by 1-methyl-4-phenyl pyridinium (MPP⁺) in SH-SY5Y neuroblastoma cells [40,41]. MPP⁺is

highly toxic to neurons and has been widely used to induce neurodegeneration in various in

vitro and in vivo models [40,41]. Several signaling pathways have been suggested to be

responsible for MPP⁺-mediated neurotoxicity in SH-SY5Y cells, for instance, trigger of

oxidative stress [42], induction of apoptosis [43], and inactivation of pro-survival

phosphoinositide 3-kinase (PI3-K)/Akt cascade [44,45]. This assay was deemed appropriate

to test for initial neuroprotective ability of the compounds because of the multitude of

pathways involved in MPP⁺mediated neurotoxicity and the potential multifunctional

inhibitory abilities of the test compounds.

Briefly, the assay entailed treating the SH-SY5Y cell line with different concentrations (1

µM, 5 µM, 10 µM) of test compounds two hours prior to MPP⁺treatment. After 48 hours of

incubation, the neuroprotective effect of the compounds was assessed by means of the MTT

mitochondrial function assay which measured cell viability. Vehicle control cells were treated

with DMSO (solvent for dissolving test compounds) and served as a reference for 100% cell

viability. Percentage neuroprotection values were calculated as the difference between the

final percentage cell viability of the test compound treated cell line and that of the MPP⁺only

treated cell line. These values are presented in Figure 4 and Table 1. NGP1-01, known to

have neuroprotective properties [27], was also used as a reference control for relative

comparison.

ACCEPTED MANUSCRIPT

As illustrated in Figure 4, after the exposure to 1000 µM MPP⁺for 48 hours, the cell viability

declined significantly to around 45%. However, its cytotoxic effects were significantly

ameliorated in the presence of the test compounds at 1 µM, 5 µM and 10 µM concentrations.

All compounds in the series exhibited significant neuroprotective effects at a 1 µM

concentration as they managed to improve cell viability to values between 86% and 110%.

This was expected from these compounds as they all share molecular similarities with NGP1-

01 which is known to have neuroprotective properties as confirmed in this assay [27]. At 5

µM and 10 µM a slight decrease in cytoprotection was observed for most of the compounds.

This may indicate that the neurotoxin challenged state of the SH-SY5Y cells may be more

sensitive to the cytotoxic effect of the test compounds at higher concentrations.

150

*

* *

*

**

*

* *

*

**

*

100

50

0

*

**

*

MPP⁺

#

1 µM

5 µM

10 µM

Figure 4. The effects of compounds 1-12 on MPP⁺-induced (1000 µM) cytotoxicity in SH-SY5Y cells. The

viability of the untreated control was defined as 100%. MPP⁺without test compound showed a significant

decrease in cell viability relative to the control (#, p < 0.05). Data are mean ± SEM (n = 3, four fields per

repeat). The level of statistical significance for the test compounds is set at * p < 0.05 compared to the MPP⁺

only treated control. Tukey's multiple comparisons statistical analysis was performed on all raw data and

significant neuroprotective effect was defined as [(*) p < 0.05, (**) p < 0.01, (***) p < 0.001, (****) p <

0.0001].

VGCC assay

To investigate the test compounds’ potential VGCC blocking effect, we developed an assay

based on the methods described by Young et al. [46] and León et al. [47] Briefly, SH-SY5Y

human neuroblastoma cells loaded with the ratiometric fluorescent calcium indicator, Mag-

Fura-2/AM, were incubated at 37 ºC in the presence of the test compounds at a 10 µM

concentration for 30 mins, then stimulated with a concentrated solution of KCl to allow

opening of the calcium channels. The assay was performed at a 10 µM concentration as

determined by the cytotoxicity assay performed on the same cell line. Changes in

fluorescence as a consequence of an increase in cytosolic calcium elicited by high K⁺

ACCEPTED MANUSCRIPT

concentrations were measured in a Biotek^®fluorescent microplate reader. Two positive

controls were included in the VGCC assay; nimodipine, a commercially available

dihydropyridine L-type calcium channel blocker [48], and the prototype polycyclic amine

cage compound, NGP1-01 [49]. The percentage VGCC blockade of each compound was

calculated relative to the activity of nimodipine and the data is presented in Table 1.

Based on the assay results, all compounds, except 3, 6 and 10, showed significant VGCC

inhibitory activity; ranging from 26.6% to 51.3%, relative to a 10 µM solution of nimodipine.

This activity is comparable to that of NGP1-01, which inhibited calcium influx by 25.7% at

the same concentration. This general improvement in activity, compared to P1 and P2 which

were inactive as VGCC blockers at 100 µM [26], emphasizes the need to include a

benzylamine moiety in the structure of polycyclic amine cage derived compounds designed

for VGCC blockade as we previously postulated [26].

It seems that a primary terminal amine imparts the molecules with increased VGCC

inhibition. This was seen in compounds 1, 2, 4, 7 and 8, which showed activities of 40%,

35%, 43%, 36% and 51% respectively, a significant improvement to the activity noted for

NGP1-01. When this primary amine terminal was converted to a secondary amine, by

conjugating a propargyl moiety, in the case of compounds 5, 6 and 9, the VGCC inhibitory

activity notably decreased to 26%, 22.9% and 39% respectively, when compared to their

primary amine carrying precursors (2, 35%; 4, 43% and 8, 51%). A terminal primary amine

could be necessary for the formation of productive binding interactions between the

respective compounds and the binding site of the VGCC resulting in their blockade.

NMDA assay

To assess the compounds’ ability to block Ca²⁺influx via the NMDAr channels, a similar

method as in the VGCC assay was used. However, a concentrated NMDA/Glycine solution,

instead of KCl, was used to stimulate calcium flux through the NMDAr. Two positive

controls were included in this assay; MK-801, a commercially available non-

competitive antagonist of the NMDAr [50], and NGP1-01 [49]. All compounds were assayed

at 10 µM and the percentage NMDAr inhibition of each compound was calculated relative to

the activity of MK-801. The calculated percentage inhibition values are presented in Table 1.

Several compounds showed great NMDAr antagonism, with compound 5 showing the

highest activity of 88.3%. Compounds 1, 3 and 6 also showed good activity with percentage

inhibition values of 63.1%, 70.5% and 63.9% respectively. These four compounds, as well as

compound 2, which showed significant activity of 49.5%, showed NMDAr antagonism

comparable to NGP1-01, which exhibited 32.5% inhibition at the same concentration. A

common structural feature of these five compounds is a ketal moiety which seemed to be

important for increased NMDAr antagonism. This observation was more compelling when

considering that compounds 9 and 10, which were devoid of this functional moiety, were

both inactive; as well as compounds 7 and 11, which also lacked the ketal moiety, showed

statistically insignificant activity of 11.9% and 26.1% respectively. The ketal moiety could be

providing necessary bulk and charge which could allow for productive ligand-NMDAr

ACCEPTED MANUSCRIPT

binding interactions. Further mechanistic studies could provide further insight on the putative

binding mode of these compounds.

The inclusion of a propargyl moiety in the compound structures seemed to contribute to

improved NMDAr antagonism. This was evident when comparing the propargyl carrying

compounds 5 and 6, which were 88.3% and 63.9% active respectively, to their precursor

compounds 2 and 4, devoid of this moiety, which were 49.5% active and inactive

respectively. An imine bond between the polycyclic cage and the benzylamine moiety, as

seen in compounds 2 and 5, seemed to be ideal for NMDAr inhibitory activity compared to

the simple amine bond contained in the structures of compounds 4 and 6. When the imine

bond in compound 2 is reduced to a simple amine bond in compound 4, the NMDAr

antagonistic activity was completely lost. Similarly, there was about 25% decrease in activity

when the imine bond in compound 5 was reduced to afford compound 6. The rigid imine

bond seemed to be ideal for NMDAr inhibition. A probable explanation for this could be that

the rigid ligand confirmation allows the compounds to adequately traverse the ligand-

recognition region of the NMDAr subunit which is defined by two polypeptide segments, S1

and S2 [51].

Generally, these compounds seem to possess calcium regulatory potential comparable to the

prototype NGP1-01 by acting on both VGCC and NMDAr. However, the compounds

appeared to be more active as NMDAr antagonists than they are VGCC blockers.

MAO-A and MAO-B inhibition studies

The target compounds were investigated for their inhibitory activity against human MAO

(hMAO) by measuring the extent to which the test compounds reduce the oxidative catalysis

of kynuramine, a mixed MAO-A/B substrate, by the respective MAO enzymes [52]. The

fluorescence of the MAO generated 4-hydroxyquinoline in the supernatant fractions were

measured using a Biotek^®fluorescent microplate reader at an excitation wavelength of 310

nm and an emission wavelength of 400 nm. IC₅₀values were calculated as the compound

concentration that produces 50% enzyme activity inhibition. The inhibition potencies of the

test compounds and reference compounds, selegiline and clorgyline, are presented in Table 1.

Based on the results, most of the active compounds displayed higher inhibitory potencies

against the MAO-B compared to the MAO-A isoenzyme as suggested by the selectivity index

(SI) values. The only exception was compound 10, which showed selectivity to the MAO-A

isoenzyme. The ketal moiety seemed to be influential in conferring the molecules MAO-B

selectivity as observed from the high SI values of compounds 1, 2, 5 and 6, which all carry

this moiety. These compounds were about 3 to 52 times more selective to the MAO-B

compared to the MAO-A isoenzyme. Conversely, compounds 9 and 12, though devoid of the

ketal moiety, also showed selectivity to MAO-B with SI values of 14.8 and 3.6 respectively.

Several structure activity relationships could be derived from comparing the MAO IC₅₀

values of the various compounds in the series. Interesting to note, was that the

propargylamine carrying compounds 5, 6, 9 and 12 showed MAO-B inhibition values which

ranged between 1.7 µM and 36.31 µM (IC₅₀). This was a significant improvement from the

activities shown by the previously reported compounds P1 and P2, which were inactive as

ACCEPTED MANUSCRIPT

MAO-B inhibitors[26]. This improved activity can be attributed to the increased molecular

length of these compounds as a result of the incorporation of a benzylamine moiety into their

structures.

Table 1: The biological activity profiles of compounds 1-12 and related reference compounds.

Compound

Cell

Viability

%

Neuro-

VGCC

%

NMDA

%

MAO- A MAO-B

SI

protection

MAO-B^b

a

%

10 µM

1 µM

10 µM

IC₅₀µM IC₅₀µM

<10 nM

-

Selegiline

Clorgyline

MK-801

Nimodipine

-

<10 nM

-

Inactive

100.0

-

100.0

Inactive

-

86 ± 7.8

66 ± 6.9^****25.7 ± 9.2^*

32.5 ± 9.9^*

Inactive

>100

-

>100

Inactive

6%

NGP1-01

P1^[26]

Inactive

-

P2^[26]

1

18

4

97 ± 7.9

59 ± 7.2^****40.1 ± 3.1^***

63.1 ± 11.3^****

49.5 ± 6.0^**

70.5 ± 4.2^***

Inactive

>100

56.23

2.37^**

>100

1.91^**

13.80^*

100

>52.4

>7.2

-

96 ± 11.4 60 ± 1.3^****35.0 ± 10.2^**

2

3

4

5

6

7

8

9

88 ± 6.8

61 ± 5.9^****24.3 ± 15.7

84 ± 12.0 54 ± 8.5^****43.2 ± 3.9^***

>100

1.70^*

36.31^*

>100

3.80^*

100

-

114 ± 25.3 53 ± 3.5^****26.6 ± 3.9^*

88.3 ± 7.9^****

63.9 ± 13.4^**

11.9 ± 7.8

59.0 ± 4.7^****

0.6 ± 8.1

Inactive

>58.8

>2.8

-

38 ± 9.4^**

31 ± 13.0^*

116 ± 10.4

101 ± 6.0

107 ± 5.5

22.9 ± 5.5

36.5 ± 8.2^***

46 ± 13.2^***51.3 ± 10.4^****

-

98 ± 14.2 43 ± 9.2^***

104 ± 13.4 46 ± 1.2^****

39.4 ± 8.0^***

7.6 ± 1.3

14.8

0.02

7.08

3.55

10

11

12

98 ± 2.7

48 ± 4.1^****31.6 ± 1.7^**

26.1 ± 2.0

14.13^**

28.18^**

116 ± 18.8 54 ± 7.8^****29.1 ± 5.8^**

31.4 ± 6.8^*

^aPercentage neuroprotection values calculated as the difference between the final percentage cell viability of the

test compound treated cell line and that of the MPP⁺only treated cell line. Experimental values are a mean of at

least 9 independent experiments (n = 9); (b) = hMAO-B selectivity index calculated as: IC₅₀(hMAO-

A)/IC₅₀(hMAO-B); Statistical analysis was performed on raw data, with asterisks indicating significant activity

[(*) p < 0.05, (**) p < 0.01, (***) p < 0.001, (****) p < 0.0001].

The data from this study suggests the juxtapositioned benzylamine and propargyl moieties are

imperative in rendering the compounds better MAO-B inhibitory activity. This observation

was noted by comparing the activity of compounds 4 and 8, which both lacked the propargyl

moiety but carried the benzylamine moiety, to the activity of compounds 6 and 9, with both

moieties. Compounds 4 and 8 showed IC₅₀values of more than 100 µM, while compounds 6

and 9 showed values of 36.31 µM and 3.89 µM respectively. It is important to note that when

the benzylamine moiety is absent in the molecular structure of these compounds, as in the

case of P1 and P2, the propargylamine moiety alone fails to render significant activity on the

MAO-B enzyme, thus further supporting the need for both moieties in the structures of these

type of compounds for activity.

By comparing derivatives 5 and 6, whose IC₅₀values were 1.70 µM and 36.31 µM

respectively, the influence of the imine bond between the polycyclic scaffold and the

benzylamine moiety could be appreciated. This bond seemed to confer compound 5 a 16-fold

increase in activity when compared to compound 6, in which the imine bond was reduced to a

secondary amine. The same was true for the precursors of compounds 5 and 6, compounds 2

(imine) and 4 (amine), whose IC₅₀’s were 13.8 µM and >100 µM respectively. Perhaps this is

ACCEPTED MANUSCRIPT

because the imine bond, present in compounds 2 and 5, is more rigid than the rotatable amine

bond present in compounds 4 and 6, thus allowing compounds 2 and 5 to be confined to a

fixed region of the MAO-B enzyme active site allowing for more productive binding

interactions.

When compound 2 is di-substituted with the ketal-polycyclic moiety to give compound 3,

MAO-B inhibitory activity is lost. The di-substituted molecule appears to be bulky and

probably fails to make access into the enzyme active pocket, rendering it inactive. The

absence of a propargyl function within the structure of compound 3 further explains the lack

of MAO inhibition observed from this compound.

NGP1-01 showed little activity on both MAO-A and MAO-B enzymes, IC₅₀>100 µM on

both isoenzymes, but when para-substituted with a primary amine to give compound 10,

MAO-A inhibition improved (IC₅₀= 2.37 µM). This improved activity may be due to the

involvement of the amine group in the formation of binding interactions with amino acid

residues found in the substrate cavity of the MAO-A enzyme. Contrary to this, compound 4,

which was also para-substituted with a secondary amine, showed little activity on the MAO-

A isoenzyme. This could be due to the presence of the bulk ketal moiety attached to the

polycyclic cage scaffold, which may be hindering the entrance of the molecule into the

enzyme substrate cavity. It is important to note that all ketal derivatives 1 – 6 showed little

activity on the MAO-A isoenzyme of IC₅₀> 100 µM.

In general, compounds 1, 2, 5, 6, 9 and 12 showed the best overall MAO inhibitory activity

and appear to be selective for the MAO-B isoenzyme. Although the MAO-B inhibitory

activity herewith reported was a significant improvement in comparison to the previously

reported compounds P1 and P2, it is important to note that this activity was still about 1000-

fold weaker than selegiline. Further structural modifications of these compounds may

therefore be necessary to afford molecules which may potentially show improved MAO-B

inhibition.

MAO reversibility studies

To determine the binding mode of the studied compounds on MAO-A and/or -B, a time

dependency of enzyme inhibition was measured. If the compounds form a covalent adduct

with the enzyme, a time-dependent reduction of enzyme activity would be expected. In this

regard, the time dependent inhibition of MAO-A and-B by the active compounds 1, 5, 6, 9

and 12, were evaluated (Figure 5). Briefly, recombinant human MAO-A and MAO-B was

preincubated with the test compounds for periods of 0, 15, 30 and 60 min prior to starting the

enzyme reaction and the residual rates of the MAO-A and -B catalysed oxidation of

kynuramine were measured. For this purpose, the concentrations of the test compounds

chosen were approximately twofold the measured IC₅₀values for the inhibition of the

respective enzymes. Clorgyline, with known irreversible inhibition for MAO-A [53], and

selegiline, with known irreversible inhibition for the MAO-B [54], were used as a reference

compounds.

ACCEPTED MANUSCRIPT

10000

8000

6000

4000

2000

0

0 Mins

15 Mins

30 Mins

60 Mins

8000

6000

4000

2000

0

0 Mins

15 Mins

30 Mins

60 Mins

Figure 5. Time-dependent inhibition of the recombinant human MAO-A (top) and MAO-B (bottom) catalysed

oxidation of kynuramine by the selected active compounds. The enzymes were preincubated for various periods

of time (0–60 min, x-axis) Data are expressed as the mean RFU (relative fluorescent units) ± SEM of three

independent experiments.

As shown in Figure 5, only compound 9 showed a significant reduction in enzymatic activity

with increased preincubation time when assayed on the MAO-A isoenzyme, suggesting

compound 9 to be an irreversible inhibitor of MAO-A. MAO-B enzyme activity was not

reduced by compounds 1, 5, 6, 9 and 12 with increased preincubation time. This indicates

ACCEPTED MANUSCRIPT

that the selected compounds were reversible inhibitors of MAO-B. Preliminary MAO-A

molecular modelling studies performed on compound 9 (data not shown) showed potential

involvement of the free hydroxyl group attached to the polycyclic moiety in forming covalent

binding interactions with amino acid residue Gln215 of the MAO-A substrate cavity. This

interaction could be the explanation for the observed irreversible MAO-A inhibition

exhibited by compound 9. No covalent interactions were observed between compound 9 and

the MAO-B isoenzyme (see Figure 7), perhaps explaining the reversible MAO-B inhibition

observed for compound 9. It also was interesting to note that compounds 1 and 5 showed a

notable increase in enzymatic activity over time. This might be due to test compound

degradation in the aqueous medium as both compounds possess a ketal moiety as well as an

imine bond. The imine function is known to undergo hydrolysis in aqueous medium [55],

however, further degradation studies on these compounds are necessary to validate this

notion.

MAO-B molecular modelling studies

With most compounds showing selectivity for the MAO-B isoenzymes, the binding modes of

all twelve compounds in the MAO-B substrate cavity were examined in silico. Docking

simulations were performed on the human MAO-B crystal structure (PDB ID: 2V5Z) [28],

using the Molecular Operating Environment (MOE) software [29]. In general, the amino acid

residues between 120 and 220, as well as the FAD cofactor, are important in conferring

substrate selectivity of MAO-B [56], most importantly residues Ile-199, situated in the

entrance cavity, and Gln-206, situated in the substrate cavity. Interaction with these amino

acid residues may confer a compound MAO-B inhibitory activity as shown by safinamide

(Figure 6), a known MAO-B inhibitor.

Figure 6. The putative binding modes of safinamide within the human MAO-B enzyme active site. Safinamide

and the FAD cofactor are shown in the docking simulation on the left as bold lines (indicated in yellow and

green respectively), and the binding interactions with the amino acid residues shown on the right. Safinamide

shows productive interactions with amino acid residues Ile-199 and Gln-206.

The docking simulation results generally showed that the lipophilic polycyclic cage unit,

carried by all the test compounds, stabilized within the hydrophobic environment of the

enzymes entrance cavity, while the rest of the molecular structure traversed deep into the

substrate cavity (see Figure 7). Further analysis of the best-ranked docking solutions showed

all active compounds to be forming productive binding interactions with the MAO-B

substrate cavity. Interestingly, most of the active compounds, particularly compounds 1, 2, 5,

ACCEPTED MANUSCRIPT

9, 11 and 12, seemed to be forming binding interactions specifically with Ile-199, an

observation which could explain their noted in vitro activity. The electron-rich π system,

presented by the aromatic ring within these compounds’ structures, appeared to be vital for

the formation of ‘arene-hydrogen’ interactions with the amino acid Ile-199.

Figure 7: MOE generated binding interactions of compounds 1, 2, 5, 6, 9 and 12 with the respective amino acid

residues within the human MAO-B enzyme active site.

ACCEPTED MANUSCRIPT

In the case of compounds 1 and 2, the ketal moiety also seemed to be involved in the

formation of binding interactions with Ile-199. This interaction may be due to the

electronegative oxygen atom which creates a slightly positive environment on the adjacent -

CH₂group causing it to serve as a proton donor to the amino acid residue. While the aromatic

system and ketal moiety seemed to be important for the formation of productive interactions

with the enzyme cavity, it appeared as though these interactions were dependent on the

presence of an imine bond between the polycyclic cage and the benzylamine moiety. This

was evident when comparing binding interactions observed for compounds 1, 2 and 5, which

all carried this bond and showed interactions with Ile-199, to the binding interactions

observed for compounds 4 and 6, which lacked this bond and hence failed to interact with Ile-

199. This observation suggests the imine bond to be important for stabilizing the molecules’

benzylamine moiety within the MAO-B enzyme cavity, thus allowing for the formation of

productive binding interactions. The observation correlates with the observed in vitro activity

of compounds 1, 2 and 5, which all showed significant MAO-B inhibition of IC₅₀: 1.91 µM,

13.80 µM and 1.70 µM, respectively. While compound 6, shows no interaction with Ile-199

due to the absence of an imine linkage, the compound remains active due to the presence of a

propargylamine group which forms interactions with the FAD cofactor.

The inactive compounds 7, 8 and 10, the oxa- and aza-type derivatives with a terminal

primary amine group, showed no interaction with any amino acid residues within the MAO-B

active site. This observation may be the explanation to these compounds’ lack of activity.

Interesting to note was that when the secondary amine group was conjugated to a propargyl

function, as seen in compound 9, the resultant compound showed interactions with Ile-199

and the FAD cofactor. The role of the propargyl moiety seems to be particularly important for

MAO-B inhibitory activity as suggested by the in vitro activities of compounds 5, 6, 9 and 12

(IC₅₀: 1.70 µM, 36.31 µM, 3.80 µM and 28.18 µM, respectively). Molecular modelling

findings showed the propargyl moieties of these four compounds to either be in close

proximity to the FAD cofactor, in the case of compounds 5 and 12, or to be forming binding

interactions with the FAD cofactor, in the case of compounds 6 and 9.

Conclusion

We have successfully designed and synthesized twelve novel pentacycloundecane and

hexacycloundecane derived compounds which showed promising neuroprotective potential.

Regarding cytotoxicity, the SH-SY5Y human neuroblastoma cells seemed to be viable with

all the test compounds, particularly at 10 µM, as none of the compounds showed statistically

significant cytotoxicity at this concentration. At a 1 µM concentration, the compounds

significantly improved the viability of SH-SY5Y neuroblastoma cells, previously exposed to

the neurotoxin MPP⁺, from about 45% to values between 86% and 110%.

Most of the compounds, except 3, 6 and 10, showed statistically significant VGCC inhibitory

activity ranging from 26.6% to 51.3%, relative to a 10 µM solution of nimodipine. This was

comparable to that of NGP1-01, which inhibited calcium influx by 25.7% at the same

concentration. Compounds 1, 2, 3, 5, 6, 8 and 12, showed significant NMDAr inhibitory

activity ranging between 31.4% and 88.3%, relative to a 10 µM solution of MK-801.

ACCEPTED MANUSCRIPT

When assayed for MAO inhibition, most of the active compounds (1, 2, 5, 6, 9, 11 and 12)

displayed selectivity to the MAO-B isoenzyme with IC₅₀values ranging from 1.70 µM to

36.31 µM, with the only exception being compound 10, which showed selectivity to the

MAO-A isoenzyme. The most active MAO-A inhibitor, compound 10, as well as the MAO-B

inhibitors, compounds 1, 5, 6, 9 and 12, all appeared to be reversible MAO inhibitors as

defined by the time-dependency studies conducted. Compound 9 however, seemed to be an

irreversible inhibitor of the MAO-A isoenzyme, an undesirable type of inhibition which has

been shown to cause the unwanted hypertensive response in patients [13]. Further structural

modifications to compound 9 are therefore necessary in order to afford a resultant molecule

with the desired reversible MAO-A inhibitory potential.

Molecular modelling studies gave some insight on the compounds’ potential binding

interactions with the MAO-B isoenzyme. Based on the in silico studies, the benzylamine

moiety seemed to be a very important structural component in affording this series MAO-B

inhibitory activity. Not only was it important for elongating the molecules to allow for the

propargylamine moiety to interact with the isoenzyme’s FAD cofactor, but it also seemed to

be involved in most of the binding interactions observed in silico with the MAO-B substrate

cavity, particularly with Ile-199.

Various other structure-activity relationships were derived from the series of compounds, and

these include the influence of; (a) the propargyl moiety, in reduced cytotoxicity, improved

NMDAr antagonism and improved MAO-B inhibition, (b) a terminal primary amine, which

improved VGCC blockade (c) a benzylamine moiety for improved VGCC blockade as well

as MAO-B inhibition, (d) an imine bond between the polycyclic cage and the benzylamine

moiety, which was ideal for NMDAr antagonism and MAO-B inhibition, (e) a ketal moiety,

which seemed to be important for increased NMDAr antagonism and conferring the

compounds MAO-B selectivity.

When considering multifunctionality, compounds 1, 2, 5 and 12 showed the best overall

activity as viable neuroprotective agents with inherent calcium regulatory potential, by

blocking VGCC and NMDAr, and MAO-B inhibitory capacity. More importantly, due to the

inclusion of a propargyl moiety within their structures, compounds 5 and 12 are promising

leads to multi-mechanistic compounds the may be useful as drug agents for the treatment of

neurodegenerative disorders. Further in vitro and in vivo studies, which may include

mechanistic studies as well as elucidation of the actual binding mode of these compounds,

would provide better insight on their drugabilty. Also, these analogues may be screened for

potential neuroprotective activity on various other drug targets implicated in

neurodegeneration.

Acknowledgements

We are grateful to the University of the Western Cape and the National Research Foundation

of South Africa for financial support. Prof E. Antunes at the NMR facility of the University

of the Western Cape is also acknowledged.

ACCEPTED MANUSCRIPT

Supplementary material

Supplementary data to this article including experimental procedures, additional molecular

modelling data and NMR spectra can be found online at xxx.

References

[1] Dauer, W.; Przedborski, S. Parkinson's disease: mechanisms and models. Neuron. 39(6),

889-909 (2003).

[2] Baker, G.B.; Reynolds, G.P.; Biogenic amines and their metabolites in Alzheimer’s

disease: noradrenaline, 5-hydroxytryptamine and 5-hydroxyindole-3-acetic acid are depleted

in the hippocampus but not in substantia innominata. Neurosci. Lett. 100(1–3), 335–339

(1989).

[3] Cross, A.J. Serotonin in Alzheimer type dementia and other dementing illnesses. Ann. NY

Acad. Sci. 600, 405–415 (1990).

[4] Arundine, M.; Tymianski, M. Molecular mechanisms of calcium-dependent

neurodegeneration in excitotoxicity. Cell Calcium. 34(4-5), 325-37 (2013).

[5] Arundine, M.; Tymianski, M. Molecular mechanisms of glutamate-dependent

neurodegeneration in ischemia and traumatic brain injury. Cell Mol Life Sci. 61(6), 657-68

(2004).

[6] Nicotra, A.; Pierucci, F., Parvez, H.; Senatori, O. Monoamine oxidase expression during

development and aging. Neurotoxicology. 25(1–2), 155–165 (2004).

[7] Youdim, M.B.H.; Edmondson, D.; Tipton, K.F. The therapeutic potential of monoamine

oxidase inhibitors. Nat. Rev. Neurosci. 7(4), 295–309 (2006).

[8] Nicotra, A.; Pierucci, F.; Parvez, H.; Senatori, O. Monoamine oxidase expression during

development and aging. Neurotoxicology. 25(1-2), 155-65 (2004).

[9] Strydom, B.; Bergh, J.J.; Petzer, J.P. Inhibition of monoamine oxidase by phthalide

analogues. Bioorg Med. Chem. Lett. 23(5), 1269–1273 (2013).

[10] Youdim, M.B.; Collins, G.G.; Sandler, M.; Bevan Jones, A.B.; Pare, C.M.; Nicholson,

W.J. Human brain monoamine oxidase: multiple forms and selective inhibitors. Nature.

236(5344), 225-8 (1972).

[11] Green, A.R.; Mitchell, B.D.; Tordoff, A.F.; Youdim, M.B. Evidence for dopamine

deamination by both type A and type B monoamine oxidase in rat brain in vivo and for the

degree of inhibition of enzyme necessary for increased functional activity of dopamine and 5-

hydroxytryptamine. Br. J. Pharmacol. 60(3), 343–349 (1977).

[12] Zimmer, R. Relationship between tyramine potentiation and monoamine oxidase (MAO)

inhibition: comparison between moclobemide and other MAO inhibitors. Acta Psychiatr

Scand Suppl. 360, 81-3 (1990).

[13] Tipton, K.F. Monoamine oxidase inhibitors and pressor response to dietary amines. Vopr

Med Khim. 43(6), 494-503 (1997).

ACCEPTED MANUSCRIPT

[14] O’Carroll, A.M.; Fowler, C.J.; Phillips, J.P.; Tobbia, I.; Tipton, K.F. The deamination of

dopamine by human brain monoamine oxidase. Specificity for the two enzyme forms in

seven brain regions. Naunyn Schmiedebergs Arch. Pharmacol. 322(3), 198–202 (1983).

[15] Chen, J.J.; Swope, D.M.; Dashtipour, K. Comprehensive review of rasagiline, a second-

generation monoamine oxidase inhibitor, for the treatment of Parkinson's disease. Clin. Ther.

9, 1825-49 (2007).

[16] Edmondson, D.E.; Binda, C.; Mattevi, A. Structural insights into the mechanism of

amine oxidation by monoamine oxidases A and B. Arch. Biochem. Biophys. 464(2), 269-76

(2007).

[17] Maruyama, W.; Akao, Y.; Carrillo, M.C.; Kitani, K.; Youdium, M.B.H.; Naoi, M.

Neuroprotection by propargylamines in Parkinson’s disease: suppression of apoptosis and

induction of prosurvival genes. Neurotoxicol. Teratol. 24(5), 675–682 (2002).

[18] Youdim, M.B.; Weinstock, M. Molecular basis of neuroprotective activities of rasagiline

and the anti-Alzheimer drug TV3326 [(N-propargyl-(3R)aminoindan-5-YL)-ethyl methyl

carbamate]. Cell. Mol. Neurobiol. 21(6), 555–573 (2002).

[19] Zindo, F.T.; Joubert, J.; Malan, S.F. Propargylamine as functional moiety in the design

of multifunctional drugs for neurodegenerative disorders: MAO inhibition and beyond.

Future Med. Chem. 7(5), 609–629 (2015).

[20] Youdim, M.H.B.; Buccafusco, J.J. CNS Targets for multifunctional drugs in the

treatment of Alzheimer’s and Parkinson’s diseases. J. Neural. Transm. 112(4), 519-537

(2005).

[21] Bolea, I.; Gella, A.; Unzeta, M. Propargylamine-derived multitarget-directed ligands:

fighting Alzheimer’s disease with monoamine oxidase inhibitors. J. Neural. Transm. 120(6),

893-902 (2013).

[22] Mattson, M.P. Calcium and neurodegeneration. Aging Cell. 6(3), 337-50 (2007).

[23] Bezprozvanny, I. Calcium signaling and neurodegenerative diseases. Trends Mol Med.

15(3), 89-100 (2009).

[24] Berridge, M.J.; Lipp, P.; Bootman, M.D. The versatility and universality of calcium

signaling. Mol Cell Biol. 1(1), 11–21 (2000).

[25] Camps, P.; Duque, M.D.; Vázquez, S. et al. Synthesis and pharmacological evaluation of

several ring-contracted amantadine analogs. Bioorg. Med. Chem. 16(23), 9925-9936 (2008).

[26] Zindo, F.T.; Barber, Q.R.; Joubert, J.; Bergh, J.J.; Petzer, J.P.; Malan, S.F. Polycyclic

propargylamine and acetylene derivatives as multifunctional neuroprotective agents. Eur. J.

Med. Chem. 80, 122-134 (2014).

[27] Geldenhuys, W.J.; Malan, S.F.; Murugesan, T.; van der Schyf, C.J.; Bloomquist, J.R.

Synthesis and biological evaluation of pentacyclo[5.4.0.02,6.03,10.05,9]undecane derivatives

ACCEPTED MANUSCRIPT

as potential therapeutic agents in Parkinson’s disease. Bioorg. Med. Chem. 12(7), 1799-1806

(2004).

[28] Binda, C.; Wang, J.; Pisani, L.; Caccia, C.; Carotti, A.; Salvati, P.; Edmondson, D.E.;

Mattevi, A. Structures of human monoamine oxidase B complexes with selective noncovalent

inhibitors: safinamide and coumarin analogs. J Med Chem. 50(23), 5848-52 (2007).

[29] Molecular Operating Environment (MOE), Version 2018.01. http//www.

chemcomp.com.

[30] Joubert, J.; Geldenhuys, W.J.; Van der Schyf, C.J. et al. Polycyclic cage structures as

lipophilic scaffolds for neuroactive drugs. ChemMedChem 7(3), 375-84 (2012).

[31] Zah, J.; Terre’Blanche, G.; Erasmus, E.; Malan, S.F. Physicochemical prediction of a

brain-blood distribution profile in polycyclic amines. Bioorg. Med. Chem. 11(17), 3569-78

(2003).

[32] Cookson, R.C.; Grundwell, E.; Hudec, J. Synthesis of cage-like molecules by irradiation

of Diels-Alder adducts. Chem. Ind. (London) 1003-1004 (1958).

[33] Dekker, T.G.; Oliver, D.W. Synthesis of (D₃) trishomocuban-4-ol via carbenium

ion rearrangement of pentacyclo[5.4.0.0^2,6.0^3,10.0^5,9]undecan-8-ol. S. Afr. J. Chem. 32, 45-48

(1979).

[34] Joubert, J.; Sharma, R.; Onani, M.; Malan, S.F. Microwave-assisted methods for the

synthesis of pentacyclo[5.4.0.0^2,6.0^3,10.0^5,9]undecylamines. Tetrahedron Lett. 54, 6923-6927

(2013).

[35] Marchand, A. P.; Arney, B. E.; Dave, P. R.; Satyanarayana, N.; Watson, W. H.; Nagl, A.

Transannular cyclizations in the pentacyclo [5.4. 0.0^{2, 6}.0^{3, 1}0.0^{5, 9}] undecane-8, 11-dione

system. A reinvestigation. J. Org. Chem. 53, 2644-2647 (1988).

[36] Kappe, C.O.; Stadler, A. Microwaves in Organic and Medicinal Chemistry. Wiley-VCH.

Weinheim, 2005.

[37] Wathey, B.; Tierney, J.; Lidstrom, P.; Westman, J. The impact of microwave-assisted

organic chemistry on drug discovery. Drug Discov. Today. 7(6), 373-80 (2002).

[38] Sharma, R.; Joubert, J.; Malan, S.F. Synthesis and Biological Evaluations of NO-

Donating Oxa- and Aza-Pentacycloundecane Derivatives as Potential Neuroprotective

Candidates. Molecules. 23(2), 308 (2018).

[39] Mosmann, T. Rapid colorimetric assay for cellular growth and survival: application to

proliferation and cytotoxicity assays. J. Immunol. Methods. 65(1-2), 55-63 (1983).

[40] Cui, W.; Zhang, Z.; Li, W.; Hu, S.; Mak, S.; Zhang, H.; Han, R.; Yuan, S.; Li, S.; Sa, F.;

Xu, D.; Lin, Z.; Zuo, Z.; Rong, J.; Ma, E. D. L.; Choi, T. C.; Lee, S. M. Y.; Han, Y. The anti-

cancer agent SU4312 unexpectedly protects against MPP(+)-induced neurotoxicity via

selective and direct inhibition of neuronal NOS. Br. J. Pharmacol. 168(5), 1201-1214 (2013).

[41] Jantas, D.; Greda, A.; Golda, S.; Korostynski, M.; Grygier, B.; Roman, A.; Pilc, A.;

Lason, W. Neuroprotective effects of metabotropic glutamate receptor group II and III

ACCEPTED MANUSCRIPT

activators against MPP(+)-induced cell death in human neuroblastoma SH-SY5Y cells: the

impact of cell differentiation state. Neuropharmacology. 83, 36-53 (2014).

[42] Pyszko, J.; Strosznajder, J.B. Sphingosine kinase 1 and sphingosine-1-phosphate in

oxidative stress evoked by 1-methyl-4-phenylpyridinium (MPP+) in human dopaminergic

neuronal cells. Mol. Neurobiol. 50(1), 38-48 (2014).

[43] Wang, Y.; Gao, J.; Miao, Y.; Cui, Q. et al. Pinocembrin Protects SH-SY5Y Cells

Against MPP+-Induced Neurotoxicity Through the Mitochondrial Apoptotic Pathway. J.

Mol. Neurosci. 53(4), 537-545 (2014).

[44] Tasaki, Y.; Omura, T.; Yamada, T.; Ohkubo, T. et al. Meloxicam protects cell damage

from 1-methyl-4-phenyl pyridinium toxicity via the phosphatidylinositol 3-kinase/Akt

pathway in human dopaminergic neuroblastoma SH-SY5Y cells. Brain Res. 1344, 25-33

(2010).

[45] Zhu, G.; Wang, X.; Wu, S.; Li, Q. Involvement of activation of PI3K/Akt pathway in the

protective effects of puerarin against MPP+-induced human neuroblastoma SH-SY5Y cell

death. Neurochem. Int. 60(4), 400-408 (2012).

[46] Young, L.M.; Geldenhuys, W.J.; Domingo, O.C.; Malan, S.F.; Van der Schyf, C.J.

Synthesis and Biological Evaluation of Pentacycloundecylamines and Triquinylamines as

Voltage-Gated Calcium Channel Blockers. Arch Pharm (Weinheim). 349(4), 252-67 (2016).

[47] León, R.; de los Ríos, C.; Marco-Contelles, J. et al. New tacrine-dihydropyridine hybrids

that inhibit acetylcholinesterase, calcium entry, and exhibit neuroprotection properties.

Bioorg. Med. Chem. 16(16), 7759-69 (2008).

[48] Cohen, C.J.; McCarthy, R.T. Nimodipine block of calcium channels in rat anterior

pituitary cells. J Physiol. 387, 195-225 (1987).

[49] Van der Schyf, C.J.; Squier, G.J.; Coetzee, W.A. Characterization of NGP 1-01, an

aromatic polycyclic amine, as a calcium antagonist. Pharmacol. Res. 18(5), 407-17 (1986).

[50] Wong, E.H.; Kemp, J.A.; Priestley, T.; Knight, A.R.; Woodruff, G.N.; Iversen, L.L. The

anticonvulsant MK-801 is a potent N-methyl-D-aspartate antagonist. Proc Natl Acad Sci U S

A. 83(18), 7104-7108 (1986).

[51] Stern-Bach, Y.; Bettler, B.; Hartley, M.; Sheppard, P.O.; O'Hara, P.J.; Heinemann, S.F.

Agonist selectivity of glutamate receptors is specified by two domains structurally related to

bacterial amino acid-binding proteins. Neuron. 13(6), 1345-57 (1994).

[52] Novaroli, L.; Reist, M.; Favre, E.; Carotti, A.; Catto, M.; Carrupt, P. Human

recombinant monoamine oxidase B as reliable and efficient enzyme source for inhibitor

screening. Bioorganic Med. Chem. 13(22), 6212- 6217 (2005).

[53] Gleiter, C.H; Mühlbauer, B.; Schulz, R.M.; Nilsson, E.; Antonin, K.H.; Bieck, P.R.

Monoamine oxidase inhibition by the MAO-A inhibitors brofaromine and clorgyline in

healthy volunteers. J. Neural Transm. 95(3), 241-245 (1994).

ACCEPTED MANUSCRIPT

[54] Knoll, J. Pharmacological basis of the therapeutic effect of (-)deprenyl in age-related

neurological diseases. Med. Res. Rev. 12(5), 505-524 (1992).

[55] Layer, R.W. The Chemistry of Imines. Chem. Rev. 63(5), 489–510 (1963).

[56] Tsugeno, Y.; Hirashiki, I.; Ogata, F.; Ito, A. Regions of the Molecule Responsible for

Substrate Specificity of Monoamine Oxidase A and B: A Chimeric Enzyme Analysis. J.

Biochem. 118(5), 974-980 (1995).

ACCEPTED MANUSCRIPT

Highlights

•

In silico studies guided the design of twelve new polycyclic cage compounds.

All compounds showed significant neuroprotection between 31% and 61% at 1 µM.

The compounds have VGCC and NMDAr Ca²⁺regulatory potential similar to or

better than NGP1-01.

•

In vitro and in-silico studies suggest the compounds to be reversible MAO-B

inhibitors.

The inclusion of the propargylamine led to multi-mechanistic neuroprotective agents.

Article Doi

Design, synthesis and evaluation of pentacycloundecane and hexacycloundecane propargylamine derivatives as multifunctional neuroprotective agents

DOI: 10.1016/j.ejmech.2018.11.051

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.ejmech.2018.11.051

Full text of DOI:10.1016/j.ejmech.2018.11.051

Products guided by the article

R&D Labs maybe for 54353-48-9

Relevant to this article

Hot Product