Triquinane scaffolds: Shape and geometry as a function of saturation and bridgehead groups

Article abstract of DOI:10.1016/j.molstruc.2013.04.030

Polycyclic hydrocarbon compounds, also known as "cage compounds", are of interest in drug discovery due to their versatility as scaffolds. Derivatives of both pentacycloundecane-dione and triquinane-dione have been the focus of numerous investigations as multifunctional neuroprotective drugs where these compounds were used as novel drug scaffolds with the ability to cross the blood brain barrier. Here we present the synthesis, characterization and single crystal X-ray analysis for two triquinane synthons; tricyclo[6.3.0.0 ^2,6]undecane-4,9-diene-3,11-dione (compound 5 crystallizes in the monoclinic system, unit cell parameters are: a = 6.5876 (12) ?, b = 10.4204 (19) ?, c = 12.074 (2) ?; V = 825.4 (3) ?³ and Z = 4) and tricyclo[6.3.0.0^2,6]undecane-3,11-dione (compound 6 crystallizes in monoclinic system, unit cell parameters are: a = 7.5992 (7) ?, b = 10.7294 (10) ?, c = 10.8664 (10) ?; V = 884.04 (14) ?³ and Z = 4); as well as a triquinane derivative, N-(3-methoxybenzyl)-3,11-azatricyclo[6.3.0.0^2,6]undecane (compound 11 crystallizes in triclinic system, unit cell parameters are: a = 7.6714 (7) ?, b = 9.0100 (9) ?, c = 11.2539 (11) ?; V = 745.78 (12) ?³ and Z = 2). The size and geometrical conformation of the triquinane scaffolds were compared to tetra and pentacycloundecanes, revealing that tricyclo[6.3.0.0^2,6]undecane-3,11-dione experiences strain relief resulting in greater flexibility, a more asymmetric molecular shape and larger surface area. However, with the introduction of the aza-bridge in N-(3-methoxybenzyl)-3,11-azatricyclo[6.3.0.0^2,6]undecane, much of the flexibility and asymmetry is lost again. We also discuss the rearrangement mechanism for the observed retro cycloaddition and reversion, and utilized density functional theory calculations to discuss the photocyclization mechanism of this unique [2 + 2] Diels-Alder system.

Full text of DOI:10.1016/j.molstruc.2013.04.030

Accepted Manuscript
Triquinane scaffolds: Shape and geometry as a function of saturation and
bridgehead groups
Lois-May Young, Matthias Zeller, Werner J. Geldenhuys, Sarel F. Malan,
Cornelis J. Van der Schyf
PII:
S0022-2860(13)00326-8
DOI:
http://dx.doi.org/10.1016/j.molstruc.2013.04.030
MOLSTR 19685
Reference:
To appear in:
Journal of Molecular Structure
Received Date:
Revised Date:
Accepted Date:
27 November 2012
12 April 2013
12 April 2013
Please cite this article as: L-M. Young, M. Zeller, W.J. Geldenhuys, S.F. Malan, C.J. Van der Schyf, Triquinane
scaffolds: Shape and geometry as a function of saturation and bridgehead groups, Journal of Molecular Structure
(2013), doi: http://dx.doi.org/10.1016/j.molstruc.2013.04.030
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.

Triquinane scaffolds: Shape and geometry as a function of saturation
and bridgehead groups.
Lois-May Young,a Matthias Zeller,b Werner J. Geldenhuys,c Sarel F. Malan,d and
Cornelis J. Van der Schyfa,c*
aPharmaceutical Chemistry, School of Pharmacy, North-West University, Potchefstroom 2520,
South Africa, bDepartment of Chemistry, Youngstown State University, Youngstown, Ohio 44555,
USA, cDepartment of Pharmaceutical Sciences, Northeast Ohio Medical University,
Rootstown, Ohio 44272, USA, dSchool of Pharmacy, University of the Western Cape, Bellville,
7535, South Africa
*To whom correspondence should be addressed. Lois-May Young, Address: (present
address) 5/10 Victoria Parade, Manly, NSW 2095, Australia. E-mail:
y.loismay@gmail.com, phone: +61-416-358-280 and Cornelis J. Van der Schyf: Address,
Northeast Ohio Medical University, 4209 State Route 44, Rootstown, OH 44272; E-mail,
cvanders@neomed.edu, phone, +1-330-325-6467.
1

Abstract
Polycyclic hydrocarbon compounds, also known as “cage compounds”, are of
interest in drug discovery due to their versatility as scaffolds. Derivatives of both
pentacycloundecane-dione and triquinane-dione have been the focus of numerous
investigations as multifunctional neuroprotective drugs where these compounds were
used as novel drug scaffolds with the ability to cross the blood brain barrier. Here we
present the synthesis, characterization and single crystal X-ray analysis for two triquinane
synthons; tricyclo[6.3.0.02,6]undecane-4,9-diene-3,11-dione (compound 5 crystallizes in
the monoclinic system, unit cell parameters are: a = 6.5876 (12) Å, b = 10.4204 (19) Å, c
= 12.074 (2) Å; V = 825.4 (3) Å3 and Z = 4 ) and tricyclo[6.3.0.02,6]undecane-3,11-dione
(compound 6 crystallizes in monoclinic system, unit cell parameters are: a = 7.5992 (7)
Å, b = 10.7294 (10) Å, c = 10.8664 (10) Å; V = 884.04 (14) Å3 and Z = 4 ); as well as a
triquinane
derivative,
N-(3-methoxybenzyl)-3,11-azatricyclo[6.3.0.02,6]undecane
(compound 11 crystallizes in triclinic system, unit cell parameters are: a = 7.6714 (7) Å,
b = 9.0100 (9) Å, c = 11.2539 (11) Å; V = 745.78 (12) Å3 and Z = 2 ). The size and
geometrical conformation of the triquinane scaffolds were compared to tetra and
pentacycloundecanes, revealing that tricyclo[6.3.0.02,6]undecane-3,11-dione experiences
strain relieve resulting in greater flexibility, a more asymmetric molecular shape and
larger surface are. However, with the introduction of the aza-bridge in N-(3-
methoxybenzyl)-3,11-azatricyclo[6.3.0.02,6]undecane, much of the flexibility and
asymmetry is lost again. We also discuss the rearrangement mechanism for the observed
retro cycloaddition and reversion, and utilized density functional theory calculations to
discuss the photocyclization mechanism of this unique [2+2] Diels-Alder system.
2

Keywords: Polycyclic cage, pentacycloundecane, triquinane, crystal structure, L-type
calcium channel blockers.
1.
Introduction
Pentacyclo[5.4.0.02,6.03,10.05,9]undecane-8,11-dione (PCU, 4, scheme 1) is of
interest in drug discovery because of its unique cage-like structure that serves as a
lipophilic scaffold for synthesizing drugs with neuroprotective properties and the ability
to cross the blood brain barrier [1]. The polycyclic scaffold also has the ability to modify
and improve pharmacokinetic and pharmacodynamic properties of drugs [2, 3]. A well-
known
example
of
such
a
derivative
is
8-benzylamino-8,11-
oxopentacyclo[5.4.0.02,6.03,10.05,9]undecane (NGP1-01), widely reported for its activity as
a multifunctional neuroprotective drug (reviewed in [4]). NGP1-01 has been shown to be
a L-type calcium channel blocker in both neuronal and heart cells [5, 6], and also acts – in a
[7, 8]
dual mechanistic way – as an N-methyl-D-aspartate (NMDA) receptor antagonist
.
PCU (4) is also utilized as an intermediate in the synthesis of polyquinane natural
products [9], of which tricyclo[6.3.0.02,6]undecane-4,9-diene-3,11-dione (5) and
tricyclo[6.3.0.02,6]undecane-3,11-dione (6) and their derivatives are of particular interest
to our group. The triquinane scaffold was first synthesized by our group to test the
validity of the hypothesis that the pentacycloundecane skeleton serve only as a bulk
contributor to the biological activity of the polycyclic cage and its derivatives [10]
.
The triquinane scaffold and triquinylamine derivatives can be obtained via
thermal fragmentation of PCU and subsequent catalytic reduction [10]. These derivatives
3

showed suppression of the calcium action potential (AP) in guinea-pig papillary
muscle [10]. Aromatic triquinylamine derivatives were capable of completely suppressing
the AP while aliphatic derivatives suppressed the AP by ~ 50%. These compounds were
also able to significantly slow heart rate. It was concluded that ring opening of the cage
moiety did not diminish the calcium channel blocking activity, or the negative
chronotropic effects of these types of compounds [10]. These findings reiterated that the
triquinane system may afford a useful and novel synthon, and together with appropriate
side chain derivatization, it can be utilized to develop multifunctional compounds that
can be of use in the treatment of neurodegenerative diseases. The concept of utilizing L-
type calcium channel blockers in the treatment of neurodegenerative diseases has also
been explored by other authors [11, 12]. This study forms part of an extensive study in
which we utilized several pentacycloundecane and triquinane derivatives to determine
their IC50 values as L-type calcium channel blockers. Here we report the synthesis and X-
ray crystal structure investigations of the two scaffolds; tricyclo[6.3.0.02,6]undecane-4,9-
diene-3,11-dione (5) and tricyclo[6.3.0.02,6]undecane-3,11-dione (6) as well as a
substituted derivative, N-(3-methoxybenzyl)-3,11-azatricyclo[6.3.0.02,6]undecane (11).
This derivative (11) showed promising potential as an L-type calcium channel blocker in
a QSAR study (data to be published). Additionally, we elaborate on the reaction
mechanism of the cycloreversion of (4) and the retro cycloaddition of (5) utilizing
molecular modeling.
4

2.
Experimental
2.1. Synthesis
Pentacyclo[5.4.0.02,6.03,10.05,9]undecane-8,11-dione (4) was synthesized as
previously described (Scheme 3) [13], and the crystal structure for this compound has also
[13,
been reported
14]. Tricyclo[6.3.0.02,6]undecane-4,9-diene-3,11-dione (5) can be
obtained through thermal fragmentation (Scheme 1), by means of flash vacuum pyrolysis
of the saturated four-membered ring (4) [9]. The melting point of (5) was determined by
DSC to be 97.23 ºC. Catalytical hydrogenation of the thermolysis product (5) yielded
tricyclo[6.3.0.02,6]undecane-3,11-dione (6) [10]. The melting point of this compound was
determined by DSC to be 95.82 ºC. We have also encountered the reversible nature of the
[15]
cycloreversion
that yielded the cis-syn-cis linearly fused triquinane system (Scheme
3), in our attempts to polymerize (5) via a radical mechanism utilizing ultra violet (UV)
irradiation through a 450W UV lamp (quartz, EtOAc).
To obtain N-(3-methoxybenzyl)-3,11-azatricyclo[6.3.0.02,6]undecane (11)
(Scheme 2), tricyclo[6.3.0.02,6]undecane-3,11-dione (6) (3.01 g, 0.0169 mol) was
dissolved in anhydrous THF (30 ml), stirred and cooled to 5 C on an external ice bath.
An equimolar quantity of 3-methoxybenzylamine (2.32 g, 0.0169 mol) was slowly added,
while the mixture was stirred for approximately 6 h at reduced temperature. Upon
completion the solvent was removed in vacuo to yield the carbonylamine (9) as an oil.
This product was dehydrated under Dean Stark conditions for approximately 1 h using
40 ml anhydrous benzene. The benzene was removed in vacuo to afford Schiff base 10.
The resulting oil was dissolved in anhydrous MeOH (20 ml) and anhydrous THF (75 ml).
Reduction was carried out by the addition of NaBH3CN in one molar excess and stirring
5

overnight (18 h) at room temperature. The solvent was removed in vacuo to yield the
final product (11) as a dark colored oil. This oil was suspended in approximately 50 ml of
distilled water in a separation funnel. The product was extracted with DCM (4 × 20 ml)
and the combined DCM fractions were washed with distilled water (2 × 50 ml). The
organic phase was dried over anhydrous CaSO4 and filtered. The solvent was removed in
vacuo. The compound was purified using a two-step silica column chromatography
approach. The first purification step utilized PE:CHCl3:EtOAc, 10:6:1 as the eluant
mixture, and the second PE:THF, 5:1 at a lower temperature (5 °C). Purification afforded
a light yellow oil (yield: 0.4523 g, 0.002 mol, 9.46 %). From this oil the compound was
crystallized in EtOAc to obtain colorless crystals suitable for X-ray crystallographic
analysis.
2.2. NMR spectroscopy
1H-NMR and 13C-NMR spectra were obtained using a Varian Gemini 300
spectrometer at a frequency of 300.075 MHz and 75.462 MHz, respectively. This was
done in a 7 Tesla magnetic field and tetramethylsilane (TMS) was used as internal
1
standard. A bandwidth of 1,000 MHz at 24 kG was applied for H-13C-decoupling. All
chemical shifts are reported in parts per million (ppm) relative to the signal of TMS
(δ = 0) added to the deuterated solvent, CDCl3. Distorsionless Enhancement by
Polarisation Transfer (DEPT) experiments were also conducted for these compounds.
2.3. X-ray Crystallography Analysis
Diffraction data for (5), (6) and (11) were collected on a Bruker AXS SMART
6

APEX CCD diffractometer at 100(2) K using monochromatic Mo K radiation with
omega scan technique. The unit cells were determined using SMART and SAINT+ [16-18]
,
and the data were corrected for absorption using SADABS [17, 18]. The structures were
solved by direct methods and refined by full matrix least squares against F2 with all
reflections using SHELXTL [19, 20]. Refinement of an extinction coefficient was found to
be insignificant. All non-hydrogen atoms were refined anisotropically. Hydrogen atoms
were placed in calculated positions and were refined with an isotropic displacement
parameter 1.2 (C-H and CH2) or 1.5 (CH3) times that of the adjacent carbon atom. Crystal
data and refinement details are listed in Tables 2-6.
2.4. Molecular calculations
Ab initio and semi-empirical calculations were performed using the
SPARTAN `04 program (Wave function Inc. 18401 Von Karman Avenue, Suite 370
Irvine, CA92612). Geometry optimizations of initial structures were done using semi-
empirical AM1 minimization in vacuo to give a reasonable starting geometry for the
Hartree-Fock calculations. All ab initio minimizations were done using Hartree-Fock
calculations with 6-31G* basis set, and density functional theory (DFT) theory
calculations were conducted using the B3LYP functional and a 6-31G* basis sets.
3.
Results and Discussion
3.1. Spectroscopic Analysis
1H-NMR signals observed were used to identify compound 11 and suggest the
structure that was confirmed by X-ray crystallographic data from this study. Compounds
7

4, 5 and 6 have been previously synthesized and characterized by means of NMR
spectroscopy [10, 13, 21]. However, the crystal structure for 5 and 6 had never been reported.
For the novel compound (11), signals for both the „bridge‟ hydrogens (H-7a, 7b) were
observed at δH 1.34 and δH 1.16 ppm. The multiplet observed at 1.47 – 2.00 ppm that
integrated for eight protons was assigned to H-4a, 4b, 5a, 5b, 9a, 9b, 10a, 10b. The
multiplet observed at 2.35 – 2.62 ppm that integrated for two protons was assigned to H-6
and H-8. The multiplet observed at 2.66 – 2.88 ppm that integrated for two protons was
assigned to H-1 and H-2. The quartet at 3.05 – 3.30 ppm that integrated for two protons
was assigned to H-3 and H-11. The singlet at 3.57 – 3.69 ppm that integrated for two
protons was assigned to H-12a and H-12b and the three protons of the methoxy OCH3-19
correspond to the singlet at 3.74 – 3.93 ppm. We observed a doublet of doublets at 6.58 –
6.80 ppm that integrated for one proton and we assigned this to H-16. At 6.82 – 6.95 ppm
we observed a multiplet that integrated for two protons and we assigned this to H-14 and
H-18. The triplet at 7.06 – 7.23 ppm that integrated for one proton was assigned to H-17.
13
The C-NMR data are presented in Table 1 and show 14 resonances. The data obtained
from the spectroscopic analysis in this study are in full agreement with the X-ray
structure.
3.2.
X-ray Analysis
One important aspect in molecular recognition is shape. Overall size (i.e. volume)
and general geometry (e.g. the x-y-z aspect ratio of a molecule) play crucial roles as to
whether a molecule is recognized by a receptor, an active site of a protein, or other shape
driven interactions of a small compound with biological macromolecules. For
8

interactions of molecules with membrane ion channels, size, surface area, and
geometrical conformation are of particular importance and it is these properties that
determine the activity of a molecule. For polycyclic compounds, for example, the L-type
calcium channel blocking activity depends at least partially on the size and geometry of
the polycyclic scaffold [6]. Malan et al., (2000) determined that changes in the polycyclic
structure had a definitive effect on calcium channel blocking activity and that increased
inhibition of the calcium current was observed for structures in which the polycyclic
„cages‟ were enlarged. In this study we determined the single crystal structures of the two
triquinane scaffolds (5 and 6) and of the substituted derivative (11). This allowed us to
evaluate geometric feature such as the molecular size and shape. The ORTEP plots in
Figures 1(a) and 2(a) give perspective views of tricyclo[6.3.0.02,6]undecane-4,9-diene-
3,11-dione (5) and tricyclo[6.3.0.02,6]undecane-3,11-dione (6). The crystal packing for
the two compounds is represented in Figure 1(b & c) and 2(b, c & d). The cis-syn-cis
linear configuration for both 5 and 6 are clearly visible in the ORTEP plots. This is a
unique feature for these compounds as this configuration can only be obtained through
thermal fragmentation of the cage structure 4. Any other approach towards linearly fused
triquinanes would result in the cis-trans-cis tricyclopentanoid framework [22]. For
compound 5 the double bonds between C2 and C3 as well as C7 and C8 provide rigidity
to the structure and its framework does not have much flexibility as can also be seen in
the crystal packing. Compound 6 lacks the double bonds between C2 and C3 as well as
C7 and C8, and this allows for somewhat enhanced flexibility. When compared to related
pentacycloundecane counterparts [14], the two triquinane scaffolds 5 and 6 are both more
flexible and are also more asymmetric and less globular than the equivalent tetra or
9

pentacycloundecanes (Figure 3), with much larger x-y aspect ratios than the
pentacyclo(5.4.0.02,6.03,10.05,9)undecane-8,11-dione (4, CCDC entry code: FOBPAO), or
related compounds tetracyclo(6.3.0.04,11.05,9)undecane-2,7-dione (12, CCDC entry code:
TOHZUL) or pentacyclo(6.2.1.02,7.04,10.05,9)undecane-endo-endo-3,6-diol (13, CCDC
entry code: XIJPEL). An overlay of these compounds which all feature a C11O2 skeleton
is shown in Figure 3b. The triquinanes, 5 (red) and especially 6 (blue) are the by far the
most asymmetric molecules and are much wider than the tetra or pentacyclic compounds
4, 12 and 13. The largest C···C distances within one molecule, indicative of the length of
the molecule, are between 3.620 and 3.682 Å for the tetra and pentacyclic compounds.
Compounds (5 and 6), on the other hand, feature maximum C···C distances of 4.025 and
5.304 Å, respectively. The width of the molecules (measured as the distance between the
center of the two oxygen atoms and the methylene carbon atoms at the opposing end of
the molecules), on the other hand, are much more uniform and vary only between 4.661
and 4.158 Å. The smallest value of 4.148 Å is associated with compound (6), again
emphasizing the distorted elongated shape of this molecule. Contributing factors for this
observation are of course the absence of the bridging bonds between the carbon atoms in
5 and 6, and the associated release of strain. In 5 the presence of the sp2 hybridized
carbon atoms somewhat limits the conformational freedom of the carbon skeleton and
this molecule has an appearance not too different from that of the tetra or pentacyclic
compounds, but with the molecule expanded along the direction of the now missing C-C
bond. As a consequence, triquinane 5, as well as the tetra or pentacyclic compounds, are
mostly pseudo-mirror symmetric. The sp2 hybridization of the carbon atoms leads to a
highly strained cis-syn-cis linear configuration not too different from those seen for the
10

tetra or pentacyclic compounds. In 6, on the other hand, the substitution of the bridgehead
or sp2 hybridized C atoms by sp3 hybridized CH2 groups releases most of this strain
despite its cis-syn-cis linear configuration. Molecule 6 assumes a conformation
distinctively different from and much more asymmetric than that of the other four
molecules. It has no approximate local mirror symmetry, and the five membered rings are
substantially tilted away from each other. This asymmetry and greater flexibility also
affects the crystal packing of 6. The molecules are capable of enveloping each other as
can be seen in Figure 2d. Thus the intrinsic expansion by removal of a covalent bond and
associated asymmetry which further expands the molecule in one direction around its
girth, result in a larger surface area for compound (6). This observation was also made in
an article by Liebenberg et al., (1996) where the author drew comparison between the
solvent accessible surface area (SASA) of the pentacycloundecane and triquinane
scaffolds. The triquinane scaffold had a larger molecular volume (552 Å3) and SASA
(351 Å2) compared to the molecular volume (498 Å3) and SASA (325 Å2) of the
pentacycloundecane scaffold.
The increased flexibility and size (determined by molecular surface area and
volume) of 6 and to some extend its derivative 11 might be beneficial when designing
compounds that act as ion channel blockers due to better geometric conformational fit
within the ion-channel. The ORTEP plot in Figure 4 gives a perspective view of N-(3-
methoxybenzyl)-3,11-azatricyclo[6.3.0.02,6]undecane (11) which is a biologically active
derivative of 6. Its crystal structure confirms the results obtained by NMR spectroscopy
for this novel compound. From the crystal structure it is evident that the methoxybenzyl
side chain does not substantially fold back onto the triquinane scaffold. This is an
11

important observation that has an influence on the activity as an L-type calcium channel
blocker as it was described for the pentacycloundecylamines that compounds where the
side chain was able to fold back onto the scaffold had the lowest activity as L-type
calcium channel blockers [6]. Another aspect that can be seen from the structure of 11 is
that the introduction of the aza-bridge between the two halves of the molecule resulted in
some loss of the strain relieve experienced by compound 6, despite having ethylene
groups rather than double bonded sp2 carbon atoms in the structure. Molecule 11, when
compared to compound 5, is again much more rigid and the polycyclic part of the
molecule now exhibits again a pseudo-mirror plane (Figure 4). For future synthetic
approach the benefit of increased flexibility and size could be retained by introducing the
amine side chain without transannular cyclization, therefore not adding the aza- or oxa-
bridging bond.
3.3. Cyclization Reaction Mechanism
Although the retro cyclization of (5) has been reported [15], we wish to elaborate
on the mechanism that has been proposed for the somewhat unusual [π2s + 2s]
π
photochemical cycloaddition of this highly strained polycyclic system. In the synthesis of
4, the endo Diels-Alder adduct (3) was easily converted into the saturated cage isomer (4)
by photochemical cycloaddition [13]. The proposed mechanism (Scheme 3) for the
conversion can be described as excitation of electrons from the highest filled molecular
orbital (HOMO) of the bicycloheptene double bond to the lowest vacant molecular
orbitals (LUMO) of the enedionechromophore [13], thus allowing for [π2s + 2s]
π
12

[23]
photochemical cyclization to occur according to the Möbius-Hückeltheory
demonstrated by the DFT calculations in Figure 5.
as
Thermal cleavage or cycloreversion (Scheme 3) of 4 operates by the diradical
mechanism, the reverse of a [π2s + 2s] photochemical cycloaddition [23]. However it
π
should be noted that the product formed (5b) does not resemble the precursor (3) to the
cycloaddition as demonstrated by DFT calculations in Figure 5. This is due to the fact
that during the cycloreversion bonds 1, 7 and 2, 6 are cleaved instead of 1, 2 and 6, 7 that
formed during the cycloaddition step. The isomer that formed is also stereospecific with a
cis-syn-cis conformation which is unachievable by any other mechanism [9]. The
reversion of (5) back to (4) reaffirms the cis-syn-cis stereostructure which allows the
intramolecular [π2s + π2s] cycloaddition [9, 24, 25] to occur.
Explaining the mechanism for the [2+2] cycloaddition and cycloreversion can
also be aided by Frontier molecular orbital theory (FMO) which has been shown to be a
helpful tool in understanding chemical reactions [26]. The DFT level of theory has been
found to give satisfactory results for pericyclic reactions, as well as Diels-Alder
reactions [26] and was therefore chosen for the FMO evaluations as shown in this study.
Single point energies were calculated using the semi-empirical AM1 module. The
HOMO and LUMO were also calculated (Figure 5) using DFT and calculations were
conducted using the B3LYP functional and a 6-31G* basis. As can be seen from
Figure 5, for the photochemically initiated cycloaddition of 3 to give the cage structure 4,
the [2+2] system can easily form under photochemical conditions, to ensure phase
alignment of the HOMO-LUMO orbitals. Similarly, DFT calculations also show how the
retro-cycloaddition of 5b back to 4 is possible, as the energy in the form of UV light is
13

able to phase align the HOMO-LUMO orbitals for sufficient overlap to form a bond.
From the calculations, it can be seen that 3 is more likely to undergo cyclization with UV
light, as opposed to 5b, since less phase alignment is needed with 3 than 5b.
4.
Conclusions
Here we presented crystal structure data and explored some of the structural features for
two triquinane scaffolds (5 and 6) that are structurally related to the pentacycloundecane
scaffold (4) and useful in the design of multifunctional neuroprotective drugs with
activity on both the NMDA receptor channel and the L-type calcium channel. Least
squares overlays the two triquinane scaffolds (5 and 6) with the tetra (12) and
pentacycloundecane (4 and 13) allowed valuable insight into the differences in size and
geometric constrains, which are key features for L-type calcium channel activity of these
compounds. We found that substitution of the bridgehead or sp2 hybridized C atoms by
sp3 hybridized CH2 groups in triquinane 6 caused strain release, affecting the geometry of
the compound, and resulting in an asymmetric compound with greater flexibility and
larger surface are. These features enhance the activity of these compounds as L-type
calcium channel blockers. We were also able to obtain crystal structure data for 11, a
biological active derivative of 6, which has activity as an L-type calcium channel blocker.
It was observed that the introduction of the aza-bridge in this compound resulted in some
loss of the flexibility gained for its precursor, compound 6. Compound 11 have a
calculated log P value of 3.51 ± 0.39 (ACD log P® software), suggesting a more
lipophilic compound with a favorable, although not optimal, blood-brain barrier
permeability [1]. Activity for these compounds as L-type calcium channel blockers is
14

dominated by the lipophilic amine substituent. The triquinane scaffold, in addition to
contributing a larger surface area to the geometric conformation, are also a lipophilic
carrier with the ability to enhance the compound‟s pharmacokinetic/pharmacodynamic
properties to a favorable distribution in the central nervous system (CNS) [3]. These
features are of great benefit when designing drugs for treatment of neurodegenerative
disorders. We also explored the photocyclization and cycloreversion reaction
mechanisms and computational studies indicate that the cyclization of 3 to 4 is more
likely than from 5b to 4 due to higher activation energy and more suitable phase
alignment of the orbitals.
Supplementary crystallographic data for this paper are available from the
Cambridge Crystallographic Data Centre (CCDC) electronic archives (Reference: 894408
– 894410). This data can be obtained free of charge from the CCDC via
http://www.ccdc.cam.ac.uk/conts/retrieving.html
(or
from
the
Cambridge
Crystallographic Data Centre, 12, Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223
336033).
Acknowledgments: The X-ray diffractometer was funded by NSF Grant 0087210,
Ohio Board of Regents Grant CAP-491, and by Youngstown State University.
15

References
[1] J. Zah, G. Terre' Blanche, E. Erasmus and S. F. Malan, Bioorg. Med. Chem. 2003, 11:
3569-3579.
[2] W. J. Geldenhuys, S. F. Malan, J. R. Bloomquist, A. P. Marchand and C. J. Van der
Schyf, Med. Res. Rev. 2005, 25: 21-48.
[3] J. Joubert, W. J. Geldenhuys, C. J. Van der Schyf, D. W. Oliver, H. G. Kruger, T.
Govender and S. F. Malan, Chem. Med. Chem. 2012, 7: 375-384.
[4] W. J. Geldenhuys, M. B. H. Youdim, R. T. Carroll and C. J. Van der Schyf, Prog.
Neurobiol. 2011, 94: 347-359.
[5] C. J. Van der Schyf, G. J. Squier and W. A. Coetzee, Pharmacol. Res. Comm. 1986,
18: 407-417.
[6] S. F. Malan, J. J. Van der Walt and C. J. Van der Schyf, Arch. Pharm. (Weinheim).
2000, 333: 10-16.
[7] C. Kiewert, J. Hartmann, J. Stoll, T. J. Thekkumkara, C. J. Van der Schyf and J.
Klein, Neurochem. Res. 2006, 31: 395-399.
[8] W. J. Geldenhuys, S. F. Malan, J. R. Bloomquist and C. J. Van der Schyf, Bioorg.
Med. Chem. 2007, 15: 1525-1532.
[9] G. Mehta, A. Srikrishna, A. V. Reddy and M. S. Nair, Tetrahedron. 1981, 37: 4543-
4559.
[10] W. Liebenberg, P. H. Van Rooyen and C. J. Van der Schyf, Pharmazie. 1996, 51:
20-24.
16

[11] T. Yagami, K. Ueda, T. Sakaeda, N. Itoh, G. Sakaguchi, N. Okamura, Y. Hori and
M. Fujimoto, Biochem. Pharmacol. 2004, 67: 1153-1165.
[12] T. Kobayashi and Y. Mori, Eur. J. Pharmacol. 1998, 363: 1-15.
[13] R. C. Cooksen, E. Grundwell and J. Hudec, Chem. Ind. 1958, 1004-1003.
[14] A. Linden, J. Romanski, G. Mloston and H. Heimgartner, Acta Crystallogr. C. 2005,
C61: o221-o226.
[15] G. Mehta, D. S. Reddy and A. V. Reddy, Tetrahedron Lett. 1984, 25: 2275-2278.
[16] Bruker Advanced X-ray Solutions SMART for WNT/2000 (Version 5.628), Bruker
AXS Inc., Madison, Wisconsin: USA. 1997-2002.
[17] Bruker Advanced X-ray Solutions SAINT (Version 6.45), Bruker AXS Inc., Madison,
Wisconsin: USA. 1997-2003.
[18] Bruker Advanced X-ray Solutions SADABS in SAINT (Version 6.45), Bruker AXS
Inc., Madison, Wisconsin: USA. 1997-2003.
[19] Bruker Advanced X-ray Solutions SHELXTL (Version 6.10), Bruker AXS Inc.,
Madison, Wisconsin: USA. 2000.
[20] G. M. Shieldrick, Acta Crystallogr. A. 2008, A64: 112-122.
[21] G. Mehta and K. S. Rao, J. Org. Chem. 1985, 50: 5537-5543.
[22] G. Mehta and A. Srikrishna, Chem. Rev. 1997, 97: 671-720.
[23] M. B. Smith and J. March, March's advance organic chemistry, Wiley-Interscience
Publication, 2001, p. 2083.
[24] G. Mehta, A. V. Reddy, A. N. Murthy and D. S. Reddy, J. Chem. Soc. Chem.
Commun. 1983, 824.
[25] G. Mehta and A. V. Reddy, J. Chem. Soc. Chem. Commun. 1981, 756: 756-757.
17

[26] M. A. Silva, S. C. Pellegrinet and J. M. Goodman, J. Org. Chem. 2003, 68: 4059-
4066.
18

Figure 1a

Figure 1b

Figure 1c

Figure 2a

Figure 2b

Figure 2c

Figure 2d

Figure 3a

Fig 3b

Figure 4

Figure 5
3
LUMO
HOMO
5b
LUMO
HOMO
Figure 5.
Calculated HOMO & LUMO surfaces for compounds 3 and 5b.