Structure-activity relationships of polycyclic aromatic amines with calcium channel blocking activity

Article abstract of DOI:10.1002/(SICI)1521-4184(200001)333:1<10::AID-ARDP10>3.0.CO;2-5

8-Benzylamino-8,11- oxapentacyclo[5.4.0.0^2,6.03^3,10.0^5,9]undecane (1) inhibits the calcium current in L-type calcium channels. A series of nitrobenzylamines (2, 3, 4), methoxybenzylamines (5,6, 7), methylpyridines (8, 9m 10), and a phenylhydrazine derivative (11) of 8,11- oxapentacyclo[5.4.0.0^2,6.0^3,105,9]undecane was synthesized. By substituting the 8,11-oxapentacyclo- [5.4.0.0^2,6.0^3,10.0^5,9]undecane skeleton with 3-hydroxyhexacyclo- [6.5.0.0^3,7.0^4,12.0^5,10.0^9,10]tridecane (12), 8,13- dioxapentacyclo[6.5.0.0^2,6.0^5,10.0^3.11]tridecane-9-one (13), and pentacyclo[5.4.0.0^2,6.0^3,10.0^5,9]undecane (14), the effect of the polycyclic skeleton could also be investigated. Increased inhibition of calcium current was observed with aromatic substitution (especially ortho and meta substitution) in the pentacycloundecane series. The calcium channel activities of the methoxy compounds were slightly higher than those of the corresponding nitro compounds while a definite decrease in activity was observed for the phenylhydrazine and aminomethylpyridine derivatives. Increased inhibition of the calcium current was also observed for structures in which the polycyclic 'cages' were enlarged. Structure-activity relationships in this series of compounds therefore appear to be dominated by geometric or steric constraints.

Full text of DOI:10.1002/(SICI)1521-4184(200001)333:1<10::AID-ARDP10>3.0.CO;2-5

10
Malan, Van der Walt, and Van der Schyf
Structure-Activity Relationships of Polycyclic Aromatic Amines with
Calcium Channel Blocking Activity
a)
b)
a,c)
Sarel F. Malan *, J. Jurgens Van der Walt , and Cornelius J. Van der Schyf
a) Pharmaceutical Chemistry, Potchefstroom University for Christian Higher Education, Private Bag X6001, Potchefstroom, 2520, South Africa
b) Department of Physiology, Potchefstroom University for Christian Higher Education, Private Bag X6001, Potchefstroom, 2520, South Africa
c) The Harvey W. Peters Center, Departments of Chemistry and of Biomedical Sciences and Pathobiology, VA-MD Regional College of Veterinary Medicine,
Virginia Polytechnic Institute and State University, Blacksburg, VA 24061-0212, USA
Key Words: Polycyclic aromatic amines; calcium antagonism; structure-activity
[6]
potential dependent L-type calcium channel . 8-Benzyl-
Summary
2,6 3,10 5,9
amino-8,11-oxapentacyclo[5.4.0.0 .0 .0 ]undecane
[7,8]
8-Benzylamino-8,11-oxapentacyclo[5.4.0.02,6.03,10.05,9]undec-
ane (1) inhibits the calcium current in L-type calcium channels. A
series of nitrobenzylamines (2, 3, 4), methoxybenzylamines (5, 6,
7), methylpyridines (8, 9, 10), and a phenylhydrazine derivative
(11) of 8,11-oxapentacyclo[5.4.0.02,6.03,10.05,9]undecane was
synthesized. By substituting the 8,11-oxapentacyclo-
[5.4.0.02,6.03,10.05,9]undecaneskeleton with 3-hydroxyhexacyclo-
[6.5.0.0.3,7.04,12.05,10.09,13]tridecane (12), 8,13-dioxapenta-
cyclo[6.5.0.02,6.05,10.03,11]tridecane-9-one (13), and pentacyclo-
[5.4.0.02,6.03,10.05,9]undecane (14), the effect of the polycyclic
skeleton could also be investigated. Increased inhibition of calcium
current was observed with aromatic substitution (especially ortho
and meta substitution) in the pentacycloundecane series. The cal-
cium channel activities of the methoxy compounds were slightly
higher than those of the corresponding nitro compounds while a
definite decrease in activity was observed for the phenylhydrazine
and aminomethylpyridine derivatives. Increased inhibition of the
calcium current was also observed for structures in which the
polycyclic ‘cages’ were enlarged. Structure-activity relationships
in this series of compounds therefore appear to be dominated by
geometric or steric constraints.
(NGP 1-01, 1)
(Figure 1) is structurally unrelated to any
of the known calcium antagonists but was found to inhibit the
[9,10]
calcium current in L-type calcium channels
. This find-
ing provided a new tool for the study of drug action on ion
channels.
Figure 1. Structure of the lead compound NGP 1-01 (1).
As it was postulated that the pentacycloundecane skeleton
may serve only as a bulk contributor to the activity of NGP
1-01 , we decided to modify the molecule in the aromatic
amine side chain to study the structure-activity relationships
of this group of calcium channel antagonists. A series of
nitrobenzylamines (2, 3, 4), methoxybenzylamines (5, 6, 7),
methylpyridines (8, 9, 10), and a phenylhydrazine (11) was
synthesized for this purpose. By substituting the 8,11-oxa-
[9]
2,6 3,10 5,9
pentacyclo[5.4.0.0 .0 .0 ]undecane skeleton with
3,7 4,12 5,10 9,13
3-hydroxyhexacyclo[6.5.0.0.
0
.0 .0 ]tridecane,
2,6 5,10 3,11
8,13-dioxapentacyclo[6.5.0.0 .0 .0 ]tridecane-9-one,
and pentacyclo[5.4.0.0 .0 .0 ]undecane, the effect of
Introduction
2,6 3,10 5,9
the polycyclic skeleton could also be investigated. These
polycyclic ‘cages’ could be obtained by the boron tri-
fluoride etherate promoted reaction of pentacyclo-
Ion channels are a critical component of cellular function
and serve to translocate ions across cell membranes in re-
sponse to a variety of chemical and physical stimuli. Ion
channels are therefore a major class of cellular effectors
through which information received at receptors is
transduced into response . Channels sensitive to membrane
potential or voltage-gated ion channels are of particular inter-
2,6 3,10 5,9
[5.4.0.0 .0 .0 ]undecane-8,11-dione with ethyl
[11]
diazoacetate
tion
followed by subsequent decarbalkoxyla-
[12]
[13]
[1]
(Figure 2), by Baeyer-Villiger oxidation
of the
diketone (”Cookson’s cage”; Figure 3) and by the selective
decarboxylation (Huang-Minlon reaction) of the diketone
est as they are critical to the function of many excitable
[14]
(Figure 4)
.
systems including axonal conduction, neuronal firing, and
[1–4]
excitation-contraction coupling
. The chemical specific-
ity of these ion channels suggests that they may be considered
as pharmacological receptors. Therefore, all the properties
associated with conventional receptors have been found
+
+
2+
[1]
among the voltage-gated Na , K , and Ca channels . The
clinical application especially of the voltage-gated calcium
channels has been studied in great depth, particularly follow-
2+
ing the impact of Ca antagonists as cardiovascular
[5,6]
agents
.
Figure 2. Synthesis of 3-hydroxyhexacyclo[6.5.0.0.3,704,12.05,10.09,13]tride-
can-6-one.
Calcium channel antagonists are a chemically heterogene-
ous group of compounds and mostly exhibit activity on the
Arch. Pharm. Pharm. Med. Chem.
© WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0365-6233/00/0101/0010 $17.50 +.50/0

SAR of Polycyclic Aromatic Amines
11
receptor macromolecule for the ligand. To ensure effective
binding, the drug molecule needs to have specific geometri-
cal, electronic and/or steric characteristics. The physico-
chemical and electronic properties of any compound will thus
determine its biological activity. These properties therefore
are indicative of the pharmacodynamic as well as pharmaco-
kinetic properties of any drug.
Figure 3. Synthesis of 8,13-dioxapentacyclo[6.5.0.02,6.05,10.03,11]tridecane-
9,12-dione.
In the absence of a well-defined receptor structure for the
L-type calcium channel, a series of computer-based calcula-
TM
tions (Chemplus ) for the determination of physico-chemi-
cal properties (QSAR-data) for the test compounds (Table 1)
were performed. Quantum mechanical calculations to deter-
mine the minimum energy conformation of each molecule
were used. A minimum energy conformation was calculated
for each compound because it is assumed to be the most likely
[16]
conformation for interaction with the receptor
. The first
1 v
order valence connectivity index ( χ ), an intermediate non-
quantum mechanical calculation was also used to charac-
[17]
terize the test compounds
. Analysis of these data could
then give an indication of the receptor structure or of the
properties necessary for drug-receptor interaction within this
group of compounds.
Calcium channel antagonism was measured as the inhibi-
tion of calcium current after electric stimulation ofventricular
myocytes from the hearts of Duncan-Hartley strain guinea
Figure 4. Selective decarboxylation of the diketone to obtain 8-pentacy-
clo[5.4.0.02,6.03,10.05,9]undecanone.
pig. Whole cell voltage clamping was done using the chopped
[18]
clamp method
.
Reductive amination of these ketones with benzylamine
Results and Discussion
rendered
6-benzylamino-3-hydroxyhexacyclo-
3,7 4,12 5,10 9,13
[6.5.0.0 .0 .0 .0 ]tridecane (12), 12-benzylamino-
8,13-dioxapentacyclo[6.5.0.0 .0 .0 ]tridecan-9-one
At a test concentration of 10 µM the test compounds exhib-
ited inhibition of calcium current that ranged from 10–45%.
2,6 5,10 3,11
2,6 3,10 5,9
(13), and 8-benzylaminopentacyclo[5.4.0.0 .0 .0 ]un-
decane (14).
Figure 6. Inhibition of calcium current as measured with whole cell clamp-
ing.
The voltage clamping data indicate that aromatic substitu-
tion in the pentacycloundecane series results in increased
calcium current inhibition (compounds 2,3,5–7). This in-
crease in activity is more pronounced when substitution oc-
curs in the ortho and meta positions (2, 3, 5, 6). The activities
of para substituted compounds (4, 7) correlated with that of
the unsubstitutedleadcompound NGP 1-01(1). The activities
of the methoxy compounds (5–7) were slightly higher than
those of the corresponding nitro compounds (2–4) while a
Figure 5. Compounds included in the SAR study.
It is generally accepted that a certain complimentarity exists
between the receptor and its ligand in terms of the induced fit
[15]
theory
. This theory implicates a certain selectivity of the definitive decrease in activity was observed when the aro-
Arch. Pharm. Pharm. Med. Chem. 333, 10–16 (2000)

12
Malan, Van der Walt, and Van der Schyf
matic benzylamine was substituted with phenylhydrazine or matic ring pointed away from the polycyclic structure and
aminomethylpyridine (11, 8–10).
better calcium channel blocking activity was observed (3, 4,
6, 7, 12, 13; Figure 7).
Changes in the polycyclic structure also had a definitive
effect on calcium channel blocking activity. Better inhibition
of the calcium current was observed for structures in which
the polycyclic ‘cages’ were enlarged (12, 13). The calcium
channel activity of compound 14, in which the endocyclic
ether bond was removed, corresponded to that of the lead
compound.
The calculated physico-chemical and connectivity data
1 v
(log P; MR; α; VA; SA; VV; SV; H; and χ ; Table 1)
correlated well with one another and good linear regression
2
values (r > 0.9) between all the variables, with the exception
of log P, were obtained.
According to the induced fit theory
[15]
and complementar-
The structure-activity relationships observed in this series
of compounds therefore appear to be dominated by geometric
or steric constraints rather than by electronic considerations.
It is interesting to note that the structures that showed the
lowest activity were also those with the lowest calculated
energy minima (8–11; Figure 7) and that the side chain and
aromatic ring were folded back onto the polycyclic structure
in these cases. In the calculated minimum energy conforma-
tions of the remaining compounds, the side chain and aro-
ity between the receptor and ligand a correlation should exist
between the semi-empirical and theoretical data and the bio-
logical activity. Regression analysis of the calcium channel
blocking activity as obtained from the whole cell clamping
experiments against these calculated data was performed and
a correlation was found between activity and molar refractiv-
2
2
ity (r = 0.59), polarisability (r = 0.65), Van der Waals
2
2
surface area (r = 0.7), solvent accessible surface area (r =
2
0.69), Van der Waals surface bounded volume (r = 0.68),
Figure 7. Conformations and energy minima (in kcal mol–1) as obtained after
structure minimization with Hyperchem®.
Arch. Pharm. Pharm. Med. Chem. 333, 10–16 (2000)

SAR of Polycyclic Aromatic Amines
13
Table 1. Physico-chemical and connectivity data of test compounds.
————————————————————————————————————————————————————
Log P
MR
α
VA
SA
VV
SV
H
1χv
————————————————————————————————————————————————————
1
2
3
4
5
6
7
8
2.57
2.53
2.53
2.53
2.32
2.32
2.32
1.66
1.26
1.26
2.58
1.78
1.67
2.74
75.13
82.45
82.45
82.45
81.59
81.59
81.59
72.60
72.97
72.97
75.37
84.34
78.24
75.54
29.80
31.64
31.64
31.64
32.27
32.27
32.27
29.09
29.09
29.09
29.31
33.47
31.29
29.93
275.13
294.95
300.60
298.80
305.25
305.87
306.67
270.32
271.57
270.15
273.28
307.34
293.95
278.13
473.55
510.75
513.50
512.38
517.22
521.30
522.43
471.72
474.50
468.76
478.66
504.31
497.73
477.46
250.15
268.82
269.48
269.18
274.90
274.70
274.99
245.91
245.63
245.46
245.37
280.47
266.19
251.03
782.24
840.77
851.83
841.73
861.65
865.34
866.86
770.46
775.05
768.56
783.02
857.97
834.14
790.79
2.02
6.13
7.13
6.91
2.17
3.88
3.99
3.19
3.95
3.70
4.67
4.31
5.10
1.30
8.368
8.874
8.868
8.868
8.897
8.891
8.891
8.228
8.218
8.218
8.161
9.286
8.460
8.206
9
10
11
12
13
14
————————————————————————————————————————————————————
Where: log P = 1-octanol-water distribution coefficient [23,24]; MR (Å3) = molar refractivity [23,25]; α (Å3) = polarisability [26]
;
VA (Å2) = Van der Waals surface area (”grid”) [27,28]; SA (Å2) = solvent accessible surface area [25,26]; VV (Å3) = Van der Waals
surface-bounded molecular volume [25,26]; SV (Å3) = solvent accessible molecular volume [25,26]; H (kcal mol–1) = hydration energy[29]
,
and 1χv = valence first-order connectivity index [17,30]
.
Figure 9. Correlation between calcium channel antagonism, polarisability,
and solvent accessible molecular volume graphed three-dimensionally.
Figure 8. Correlation between calculated data and calcium channel antago-
nism (Activity = % inhibition of calcium current).
2
solvent accessible molecular volume (r = 0.73) as well as for
blocking activity. Conformations in which the side chain and
aromatic ring point away from the polycyclic structure (Fig-
ure 7) may therefore be preferred in the binding behavior of
these molecules to the channel receptor. The data, however,
also indicate that electronic characteristics (molar refractivity
and polarisability) influence – albeit to a lesser extent – the
affinity and/or activity of these compounds on calcium chan-
nel receptors and should always be accounted for.
2
the valence first order connectivity index (r = 0.57) (exam-
ples shown in Figure 8).
These examples show that an increase in calcium channel
blocking activity can be expected as molecular surface area
or volume is increased. This observation was also found to be
true for an increase in polarisability. Definitive clusters with
increased activity were observed at increased volume and/or
higher polarisability. In the series of compounds tested the
optimum polarisabilities and solvent accessible surface vol-
umes appear to be in the range found for two of the methoxy
compounds (compounds 5 and 6). Further increases in these
parameters resulted in diminished activity (Figure 9).
These data confirm the initial observation that the size and
geometrical conformation of the test molecules (surface area
and volume) are of particular importance for calcium channel
Experimental
Synthesis
Melting points were measured on a Gallenkamp melting point apparatus
and are uncorrected. The IR spectra were recorded on a Shimadzu FT IR
4200 spectrophotometer and the MS and high resolution MS on a VG 7070E
spectrometer (EI, 70 eV). The 1H and 13C NMR spectra were recorded on a
Arch. Pharm. Pharm. Med. Chem. 333, 10–16 (2000)

14
Malan, Van der Walt, and Van der Schyf
Varian VXR 300 spectrometer (1H at 300 MHz and 13C at 75 MHz). C, H
8-(2-Methoxybenzylamino)-8,11-pentacyclo[5.4.0.02,6.03,10.05,9]-
and N analyses were all within acceptable limits (± 0.4%).
undecane (5)
Obtained as light yellow oil from column chromatography on silica gel
with ethyl acetate:dichloromethane:petroleum ether 1:1:1 as mobile phase
(yield 49.5%).– HR-MS for C19H21NO2: calcd, 295.1572, found, 295.1571.–
IR (neat): νmax = 3330 cm–1, 2920, 1740, 1600, 1500, 1370, 1240.– 1H NMR
Synthesis of Benzylamine (1), Nitrobenzylamine (2–4), Methoxybenzylamine
(5–7), Aminomethylpyridine (8–10), and Phenylhydrazine (11) Derivatives
of Pentacyclo[5.4.0.02,6.03,10.05,9]undecane
Pentacyclo[5.4.0.02,6.03,10.05,9]undecane-8,11-dione (5 g, prepared ac-
cording to the method of Cookson et al.[19,20]) was dissolved in tetrahydro-
furan (THF, 50 ml) and cooled to ±5 °C in ice. An equimolar quantity of the
desired amine was slowly added under stirring. The white precipitate that
formed after about 10 min was filtered and washed with cold THF to render
the hydroxylamine. This product was dehydrated in dry benzene under
Dean-Stark conditions for 1 h or until no more water collected in the
container. Evaporation yielded the Shiff base as a yellowish oil and reduction
of this imine was done with sodium borohydride in dry methanol (30 ml) and
dry THF (150 ml) for 24 h at room temperature. The solvent was removed
under reduced pressure and water (100 ml) was added. The mixture was
extracted with dichloromethane (4 × 50 ml) and the combined organic frac-
tions washed with water before being dried over magnesium sulphate.
Evaporation yielded the desired products.
(CDCl3): δ = 7.33 (dd, 1H, JCD = 7.42, JBD = 1.76), 7.22 (2dd, 1H, JAB =
8.22, JBC = 7.42 JBD = 1.76), 6.92 (dt, 1H, JCD = 7.42, JAC = 1.07), 6.84 (dd,
1H, JAB = 8.22, JAC = 1.07), 4.67 (t, 1H, J = 5.27), 3.99 (AB-q, 2H, J = 13.60),
3.82 (s, 3H), 2.40–2.86 (4 × m, 9H), 1.73 (AB-q, 2H, J = 10.44).– 13C
(CDCl3): δ = 157.71 (s), 129.42 (d and s, 2C), 128.24 (d), 120.70 (d), 110.40
(d), 110.07 (s), 82.74 (d), 55.34 (k), 55.18 (d), 54.91 (d), 45.11 (d), 44.99 (d),
44.65 (d), 43.38 (t), 43.25 (d and t, 2C), 42.10 (d), 41.64 (d). –MS (70 eV):
m/z = 295 [M+], 266, 216, 136, 121, 91, 77, 65, 39, 28.
8-(3-Methoxy-benzylamino)-8,11-pentacyclo[5.4.0.02,6.03,10.05,9]-
undecane (6)
Obtained as light yellow oil from column chromatography on silica gel
with ethyl acetate:dichloromethane:petroleum ether 1:1:1 as mobile phase
(yield 44.8%).– HR-MS for C19H21NO2: calcd, 295.1572, found, 295.1576.–
IR (neat): νmax = 3340 cm–1, 2920, 1700, 1650, 1495, 1387.– 1H NMR
(CDCl3): δ = 7.23 (t, 1H, JAB = 8.10), 6.94 (m, 2H), 6.77 (m, 1H), 4.67 (t,
1H, J = 5.23), 3.98 (AB-q, 2H, J = 13.46), 3.80 (s, 3H), 2.41–2.86 (3 × m,
8H), 2.19 (br. s, 1H), 1.70 (AB-q, 2H, J = 10.44).– 13C NMR (CDCl3): δ =
160.01 (s), 142.82 (s), 129.60 (d), 120.37 (d), 113.52 (d), 112.57 (d), 109.89
(s), 82.67 (d), 55.37 (d), 55.28 (k), 54.89 (d), 47.85 (t), 44.97 (d, 2C), 44.67
(d), 43.37 (t), 43.26 (d), 42.11 (d), 41.64 (d).– MS (70 eV): m/z = 295 [M+],
267, 136, 121, 91, 77, 28.
8-Benzylamino-8,11-pentacyclo[5.4.0.02,6.03,10.05,9]undecane (1)
White needles obtained by recrystallisation from ethanol (yield 43.5%).–
HR-MS for C18H19NO: calcd, 265.1467, found, 265.1460.– Mp 77 °C.– IR
(KBr): νmax = 3300 cm–1, 2950, 1360, 1020.– 1H NMR (CDCl3): δ =
7.25–7.38 (m, 5H), 4.67 (t, 1H, J = 5.22 Hz), 4.0 (AB-q, 2H, J = 13.25 Hz),
2.5–2.85 (2 × m, 7H), 2.42 (t, 1H, J = 4.7 Hz), 2.20 (br. s, 1H), 1.72 (AB-q,
2H, J = 10.37 Hz).– 13C NMR (CDCl3): δ = 141.09 (s), 128.62 (d, 2C), 128.15
(d, 2C), 127.08 (d), 109.84 (s), 82.68 (d), 55.36 (d), 54.90 (d), 47.93 (t), 44.98
(d), 44.68 (d), 43.38 (t), 43.25 (d), 42.10 (d), 41.65 (d).– MS (70 eV): m/z =
265 [M+], 237, 186, 131, 91, 65, 28.
8-(4-Methoxybenzylamino)-8,11-pentacyclo[5.4.0.02,6.03,10.05,9
-
]undecane (7)
8-(2-Nitrobenzylamino)-8,11-pentacyclo[5.4.0.02,6.03,10.05,9]undecane (2)
Obtained as light yellow oil from column chromatography on silica gel
with ethyl acetate:dichloromethane:petroleum ether 1:1:1 as mobile phase
(yield 35.3%).– HR-MS for C19H21NO2: calcd, 295.1572, found, 295.1570.–
IR (neat): νmax = 3330 cm–1, 2970, 1700, 1616, 1517, 1460, 1355, 1242.– 1H
NMR (CDCl3): δ = 7.28 (dd, 2H, JAB = 6.62, JAC = 2.16), 6.85 (dd, 2H, JAB
= 6.62, JAC = 2.16), 4.67 (t, 1H, J = 5.22), 3.93 (AB-q, 2H, J = 12.92), 3.79
(s, 3H), 2.41–2.85 (3 × m, 8H), 2.08 (br. s, 1H), 1.72 (AB-q, 2H, J = 10.50).–
13C NMR (CDCl3): δ = 158.87 (s), 133.13 (s), 129.38 (d), 113.99 (d), 109.82
(s), 82.66 (d), 55.36 (d), 55.34 (k), 54.90 (d), 47.36 (t), 44.96 (d, 2C), 44.67
(d), 43.37 (t), 43.24 (d), 42.10 (d), 41.64 (d).– MS (70 eV): m/z 295 (M+),
267, 136, 121, 91, 77, 28.
Yellow needles obtained by recrystallisation from ethanol (yield 29.2%).–
HR-MS for C18H18N2O3: calcd, 310.1317, found, 310.1315.– Mp 87 C.– IR
(KBr): νmax = 3300 cm–1, 2960, 1525, 1350, 1010.– 1H NMR (CDCl3): δ =
7.92 (dd, 1H, JAB = 8.11, JAC = 1.37), 7.68 (dd, 1H, JCD = 7.76, JBD = 1.09),
7.57 (2dd, 1H, JAB = 8.11, JBC = 7.41, JAC = 1.37), 7.39 (2dd, 1H, JAB
=
8.11, JBC = 7.41, JAC = 1.37), 4.64 (t, 1H, J = 5.22), 4.28 (AB-q, 2H, J =
15.24), 2.40–2.86 (3 × m, 9H), 1.73 (AB-q, 2H, J = 10.55).– 13C NMR
(CDCl3): δ = 149.13 (s), 136.79 (s), 133.24 (d), 131.09 (d), 128.03 (d),
124.84 (d), 109.45 (s), 82.77 (d), 55.28 (d), 54.78 (d), 45.14 (t), 44.96 (d),
44.61 (d), 43.41 (t), 43.24 (d), 42.0 (d), 41.62 (d).– MS (70 eV): m/z = 310
[M+], 292, 276, 231, 159, 131, 91, 78, 44, 32, 28.
8-[(2-Aminomethyl)pyridine]-8,11-pentacyclo[5.4.0.02,6.03,10.05,9]-
8-(3-Nitrobenzylamino)-8,11-pentacyclo[5.4.0.02,6.03,10.05,9]undecane (3)
undecane (8)
Obtained as light yellow oil after column chromatography on silica gel
with ethyl acetate:dichloromethane:petroleum ether 1:1:1 as mobile phase
(yield 31.8%).– HR-MS for C17H18N2O: calcd, 266.1419, found 266.1410.–
IR (neat): νmax = 3330 cm–1, 2960, 1725, 1655, 1590, 1470, 1435, 1370.– 1H
NMR (CDCl3): δ = 8.55 (d, 1H, JAB = 5.43), 7.63 (m, 1H), 7.32 (d, 1H, JCD
= 7.91), 7.14 (t, 1H, JAB = 5.43), 4.66 (t, 1H, J = 5.22), 4.15 (AB-q, 2H, J =
14.86), 2.41–2.87 (3 × m, 9H), 1.72 (AB-q, 2H, J = 10.48).– 13C NMR
(CDCl3): δ = 160.17 (s), 149.39 (d), 136.63 (d), 122.16 (d), 121.98 (d), 109.74
(s), 82.67 (d), 55.51 (d), 54.91 (d), 49.15 (t), 44.94 (d, 2C), 44.66 (d), 43.38
(t), 43.24 (d), 42.14 (d), 41.62 (d).– MS (70 eV): m/z = 266 [M+], 238, 187,
131, 107, 93, 65, 28.
Yellow needles obtained by recrystallisation from ethanol (yield 40.5%).–
HR-MS for C18H18N2O3: calcd, 310.1317, found, 310.1314.– Mp 154 C.–
IR (KBr): νmax = 3320 cm–1, 2970, 1525, 1350, 1010.– 1H NMR (CDCl3): δ
= 8.28 (d, 1H, JAC = 1.64), 8.10 (m, 1H), 7.69 (dd, 1H, JAB = 7.83, JAC
=
1.64), 7.48 (t, 1H, JAB = 7.83), 4.67 (t, 1H, J = 5.27), 4.11 (AB-q, 2H, J =
14.87), 2.40–2.88 (4 × m, 8H), 2.36 (br. s, 1H), 1.75 (AB-q, 2H, J = 10.47).–
13C NMR (CDCl3): δ = 148.5 (s), 143.75 (s), 134.02 (d), 129.41 (d), 122.76
(d), 122.13 (d), 109.63 (s), 82.77 (d), 55.45 (d), 54.89 (d), 47.01 (t), 45.07
(d), 44.96 (d), 43.41 (t), 43.25 (d), 42.07 (d), 41.64 (d).– MS (70 eV): m/z =
310 [M+], 293, 231, 131, 91, 44, 32, 28.
8-(4-Nitrobenzylamino)-8,11-pentacyclo[5.4.0.02,6.03,10.05,9]undecane (4)
8-[(3-Aminomethyl)pyridine]-8,11-pentacyclo[5.4.0.02,6.03,10.05,9]-
Yellow needles obtained by recrystallisation from ethanol (yield 30.9%).–
HR-MS for C18H18N2O3: calcd, 310.1317, found, 310.1313.– Mp 162 C.–
IR (KBr): νmax = 3320 cm–1, 2970, 1510, 1340, 1010.– 1H NMR (CDCl3): δ
= 8.17 (d, 2H, JAB = 8.90), 7.54 (d, 2H, JAB = 8.90), 4.67 (t, 1H, J = 5.22),
4.12 (AB-q, 2H, J = 15.31), 2.42–2.88 (4 × m, 8H), 2.34 (br. s, 1H), 1.73
(AB-q, 2H, J = 10.59).– 13C NMR (CDCl3): δ = 149.28 (s), 147.20 (s), 128.46
(d, 2C), 123.81 (d, 2C), 109.63 (s), 82.74 (d), 55.45 (d), 54.88 (d), 47.23 (t),
45.01 (d), 44.95 (d), 44.69 (d), 43.39 (t), 43.26 (d), 42.04 (d), 41.64 (d).– MS
(70 eV): m/z 310 [M+], 282, 231, 131, 106, 91, 32, 28.
undecane (9)
Obtained as light yellow oil after column chromatography on silica gel
with ethyl acetate:dichloromethane:petroleum ether 1:1:1 as mobile phase
(yield 32.4%).– HR-MS for C17H18N2O: calcd, 266.1419, found 266.1409.–
IR (neat): νmax = 3280 cm–1, 2960, 1730, 1490, 1360, 1010.– 1H NMR
(CDCl3): δ = 8.60 (s, 1H), 8.49 (d, 1H, JAB = 3.90), 7.70 (d, 1H, JBC = 7.75),
7.25 (m, 1H), 4.67 (t, 1H, J = 5.22), 4.02 (AB-q, 2H, J = 14.12), 2.42–2.88
(4 × m, 9H), 1.73 (AB-q, 2H, J = 10.44).– 13C NMR (CDCl3): δ = 149.74
Arch. Pharm. Pharm. Med. Chem. 333, 10–16 (2000)

SAR of Polycyclic Aromatic Amines
15
(d), 148.59 (d), 136.49 (s), 135.81 (d), 123.55 (d), 109.70 (s), 82.73 (d), 55.42
(d), 54.89 (d), 45.35 (t), 44.94 (d, 2C), 44.68 (d), 43.38 (t), 43.21 (d), 42.04
(d), 41.63 (d).– MS (70 eV): m/z = 266 [M+], 238, 187, 131, 107, 92, 65, 28.
dried over magnesium sulphate. Evaporation yielded a yellow oil that con-
tained the product (13). Column chromatography with ethyl acetate:ethanol
9:1 gave (13) as a crystalline solid (Yield 18.9%).– HR-MS for C18H19NO3:
calcd, 297.1365, found, 297.1357.– Mp 178 C.– IR (KBr): νmax = 3550 cm–1
,
8-[(4-Aminomethyl)pyridine]-8,11-pentacyclo[5.4.0.02,6.03,10.05,9]-
undecane (10)
2950, 1680, 1455, 1370, 1255, 1185, 1100.– 1H NMR (CDCl3): δ = 7.30 (m,
5H), 4.97 (s + d, 2H, J = 15.66), 4.63 (m, 1H), 4.51 (dt, 1H, J = 7.72), 4.15
(d, 1H, J = 15.04), 3.97 (d, 1H, J = 4.95), 2.69–2.94 (m, 4H), 2.39–2.53 (2 ×
m, 2H), 1.74 (AB-q, 2H, J = 9.75).– 13C NMR (CDCl3): δ = 176.42 (s),
137.26 (s), 128.87 (d, 2C), 128.75 (d, 2C), 127.48 (d), 80.09 (d), 72.78 (d),
68.84 (d), 50.44 (d), 48.39 (d), 47.06 (t), 43.96 (t), 43.44 (d), 33.41 (d), 33.20
(d), 31.64 (d).– MS (70 eV): m/z = 297 [M+], 269, 228, 199, 173, 106, 91,
65, 40, 32, 28.
Off white needles obtained by recrystallisation from ethanol (yield
57.9%).– HR-MS for C17H18N2O: calcd, 266.1419, found, 266.1412.– Mp
110 °C.– IR (KBr): νmax = 3300 cm–1, 2980, 1605, 1560, 1505, 1415, 1375,
1240, 1010, 855.– 1H NMR (CDCl3): δ = 8.53 (d, 2H, JAB = 5.70), 7.31 (d,
2H, JAB = 5.70), 4.66 (t, 1H, J = 5.22), 4.02 (AB-q, 2H, J = 15.49), 2.38–2.87
(3 × m, 9H), 1.72 (AB-q, 2H, J = 10.47).– 13C NMR (CDCl3): δ = 150.63
(s), 150.00 (d, 2C), 122.78 (d, 2C), 109.68 (s), 82.72 (d), 55.44 (d), 54.86 (d),
46.66 (t), 44.98 (d, 2C), 44.66 (d), 43.38 (t), 43.25 (d), 42.05 (d), 41.62 (d).–
MS (70 eV): m/z = 266 [M+], 236, 187, 172, 159, 131, 91, 65, 40, 32, 28.
8-Benzylaminopentacyclo[5.4.0.02,6.03,10.05,9]undecane (14).
8-Pentacyclo[5.4.0.02,6.03,10.05,9]undecanone (1 g, prepared according to
the method of Oliver et al.[14]) was dissolved in a mixture of absolute ethanol
(8 ml) and benzylamine (0.9 g) and the reaction mixture was refluxed for
24 h. It was cooled in ice and a solution of sodium borohydride (1.89 g) in
cold water (20 ml) was slowly added. After stirring at room temperature for
5 h, the solvent was removed under vacuum. Water (50 ml) and dichlo-
romethane (50 ml) was added and the mixture was extracted with dichlo-
romethane (3 × 50 ml) and diethyl ether (2 × 50 ml). The combined organic
fractions were washed with water (2 × 100 ml), dried over magnesium
sulphate and evaporated to give a yellow oil. Column chromatography on
silica gel with ethyl acetate:dichloromethane:petroleum ether 1:1:1 asmobile
phase rendered 14 as a off white oil (Yield 7,6%).– HR-MS for C18H21N:
calcd, 251.1674, found, 251.1670.– 1H NMR (CDCl3): δ = 7.32 (m, 5H), 3.78
(s, 2H), 2.78 (t, 1H, J = 3.42), 2.16–2.68 (5 × m, 10H), 1.41 (AB-q, 2H, J =
10.25), 1.02 (dt, 1H, J = 3.68).– 13C NMR (CDCl3): δ = 141.25 (s), 128.50
(d, 2C), 128.24 (d, 2C), 126.96 (d), 61.50 (d), 53.45 (t), 47.32 (d), 44.69 (d),
44.15 (d), 42.07 (d), 41.92 (d), 40.91 (d), 37.70 (d), 36.34 (d), 34.74 (t), 28.83
(t).– MS (70 eV): m/z = 251 [M+], 104, 171, 160, 144, 106, 91, 79, 66, 39.
8-Phenylhydrazine-8,11-pentacyclo[5.4.0.02,6.03,10.05,9]undecane (11)
Off white needles obtained by recrystallisation from acetone (yield
62.8%).– HR-MS for C17H18N2O: calcd, 266.1419, found, 266.1418.– Mp
183 °C.– IR (KBr): νmax = 3273 cm–1, 2951, 1603, 1522, 1497, 1352, 1261,
1
1103, 739, 689.– H NMR (CDCl3): δ = 7.17 (m, 2H), 6.90 (d, 2H, JAB
=
7.63), 6.76 (t, 1H, JBC = 7.20), 5.68 (br. s, 1H), 4.66 (br. s, 1H), 4.35 (br. s,
1H,), 2.41–2.82 (m, 8H), 1.68 (AB-q, 2H, J = 10.44).– 13C NMR (CDCl3):
δ = 150.28 (s), 129.23 (d, 2C), 119.02 (d), 112.57 (d, 2C), 110.90 (s), 83.22
(d), 54.56 (d), 53.42 (d), 44.75 (d), 44.48 (d), 43.75 (t), 43.41 (d, 2C), 41.89
(d), 41.55 (d).– MS (70 eV): m/z = 266 [M+], 248, 159, 131, 105, 93, 77, 65,
44, 32, 28.
Benzylamine
Derivatives
of
3-Hydroxyhexacyclo-
[6.5.0.0.3,704,12.05,10.09,13]tridecan-6-one (12), 8,13-Dioxapentacyclo-
[6.5.0.02,6.05,10.03,11]tridecane-9,12-dione (13), and 8-Pentacyclo-
[5.4.0.02,6.03,10.05,9]undecanone (14), 6-Benzylamino-3-hydroxyhexacyclo-
[6.5.0.03,7.04,12.05,10.09,13]tridecane (12)
3-Hydroxyhexacyclo[6.5.0.0.3,704,12.05,10.09,13]tridecane-6-one (0.7 g;
prepared according to the methods described by Marchand and Arney[11] and
Krapcho and Lovey[12]) was dissolved in THF (20 ml) and benzylamine
(0.4 g) was slowly added with stirring at room temperature. After 2.5 h the
solution turned milky and the solvent was removed under vacuum after a
further 30 min to give a fine white precipitate. The hydroxyl amine formed
was dehydrated in dry benzene under Dean-Stark conditions. Evaporation of
the excess benzene yielded the Schiff base as a light yellow oil. Reduction
of this imine was done with sodium borohydride (1 g) in dry THF (50 ml)
and dry methanol (10 ml) for 24 h at room temperature. The solvent was
removed under reduced pressure and water (50 ml) and dichloromethane
(50 ml) was added. The mixture was extracted with dichloromethane (4 ×
50 ml) and the combined organic fractions washed with water before being
dried over magnesium sulphate. Evaporation yielded the desired product as
an off white oil. The white crystalline product (12) was obtained by recrys-
tallisation from cyclohexane (Yield, 84.3%).– HR-MS for C20H23NO: calcd,
292.1701, found, 292.1701.– Mp 123 C – IR (KBr): νmax = 3300 cm–1, 2950,
1485, 1455, 1330, 1318, 1130, 1063.– 1H NMR (CDCl3): δ = 7.33 (m, 5H),
3.82 (AB-q, 2H, J = 12.88), 3.37 (t, 1H, J = 3.36), 2.72 (dd, 1H, J = 3.02),
2.66 (t, 1H, J = 6.04), 2.49 (m, 2H), 2.36 (m, 3H), 2.17 (d, 1H, J = 5.43), 1.99
(m, 1H), 1.64 (AB-q, 2H, J = 11.33), 1.30 (AB-q, 2H, J = 9.61).– 13C NMR
(CDCl3): δ = 140.97 (s), 128.61 (d, 2C), 128.44 (d, 2C), 127.11 (d), 85.20
(s), 55.59 (d), 52.85 (t), 51.16 (d), 50.88 (d), 47.64 (d), 42.25 (d,), 40.85 (t),
38.86 (d), 38.52 (t), 37.32 (d), 35.25 (d), 32.92 (d), 31.94 (d).– MS (70 eV):
m/z = 292 [M+–1], 202, 186, 149, 106, 91, 69, 57, 43, 32, 28.
Electrophysiology
Ventricular myocytes from the hearts of Duncan-Hartley strain guinea-pig
were isolated by enzymatic dispersion of the cells with collagenase and
protease on a Langendorff apparatus [21]
.
Suction pipettes with resistance values between 2 and 3 MΩ were drawn
from borosilicate glass and heat polished. A GΩ seal was formed by light
suction. The patch was then ruptured and at least 15 min were allowed for
internal dialysis to reach equilibrium. Whole cell voltage clamping was done
using the chopped clamp method [18]. A concentration clamp and multibarrel
superfusion system that allowed changes in the bath solutions in less than 1 s
was used.
Potassium currents were eliminated by adding CsCl2 (20 mM) to the
Tyrode solution and TEA-Cl to the pipette solution while the effects of
sodium currents were eliminated by pre-clamping at –35 mV. Activation of
the calcium channel was done by voltage clamping from a holding potential
of –40 mV to a test potential of +10 mV for 400 ms. The pclamp 5.5®
program (Axon Instruments, Inc., 1989) was used for the control of the
experiment and storage of data. The calcium currents were measured as the
maximum negative (inward) current after subtraction of background cur-
rents. Test concentrations of 10 µM were used and calcium channel blocking
activity was expressed as the percentage inhibition of calcium current as
registered before and after inclusion of the test compound.
The composition (in mM) of the Tyrode solution for superfusion was: KCl
5.4; CaCl2 1.8; CsCl 20; MgCl2 0.5; NaCl 137.6; glucose 5 and 2-ethanesul-
phonic acid 11.6. The pipette solution (in mM) was: MgCl2 5; Na2ATP 5;
HEPES 10; CsCl 125; EGTA 15; TEA-Cl 20 and NaOH to pH = 7.2.
12-Benzylamino-8,13-dioxapentacyclo[6.5.0.02,6.05,10.03,11]tridecan-
9-one (13)
A mixture of 8,13-dioxapentacyclo[6.5.0.02,6.05,10.03,11]tridecane-9,12-
dione (2.06 g, prepared according to the method of Surapaneni and Gi-
lardi[13]), benzylamine (1.07 g) and sodium borohydride (0.6 g) was
dissolved in dry methanol and heated under reflux for 10 days. The solvent
was removed under reduced pressure and water (50 ml) and dichloromethane
(50 ml) was added. The mixture was extracted with dichloromethane (4 ×
50 ml) and the combined organic fractions washed with water before being
Computer Modeling and Calculation of Physico-Chemical Properties
A minimum energy conformation of each structure was obtained by
optimizing the structures with the Hyperchem® program (Release 4.5 for
Windows) using MM+ and AM1 semi empirical calculations and the Polak-
Ribiere minimizing procedure [22]
.
Arch. Pharm. Pharm. Med. Chem. 333, 10–16 (2000)

16
Malan, Van der Walt, and Van der Schyf
The physico-chemical data (QSAR data) of these minimized structures
[15] R.B. Silverman, The organic chemistry of drug design and drug action,
London, UK, Academic Press. 1992, 71–72, 422p.
were calculated using the ChemplusTM extension of the Hyperchem® pro-
gram. The valence first-order connectivity index (1χv) for each compound
was calculated[17] using the following equation symbols used as defined by
Kier and Hall [17]):
[16] L. Belvisi, S. Brossa, A. Salimbeni, C. Scolastico, R. Todeschini, J.
Comp.-aid. Molec. Des., 1991, 5, 571–584.
[17] L.B. Kier, L.H. Hall, Molecular connectivity in chemistry and drug
research, in G. DeStevens (Ed.), Medicinal chemistry. A series of
monographs, Vol. 14, New York, N.Y. Academic Press, 1976, 40–80.
Ne
1 ν
ν
ν
−1 ⁄ 2
χ
=
(δ δ )
i j s
∑
[18] O.P. Hamil, A. Marty, E. Neher, B. Sakman, F.J. Sigworth, Pflüg. Arch.
1981, 391, 85–100.
s = 1
[19] R.C. Cookson, E. Grundwell, J. Hudec, Chem. Ind. 1958, 1003–1004
References
[20] R.C. Cookson, E. Grundwell, R.R. Hill, J. Hudec, J. Chem. Soc. 1964,
[1] Y.W. Kwon, D.J. Triggle, Chirality 1991, 3, 393–404.
[2] D.J. Triggle, D.A. Langs, Annu. Rep. Med. Chem. 1990, 25, 225–234.
[3] K.E. Andersson, Acta Pharmacol. Toxicol. 1978, 43 (suppl 1), 5–14.
[4] E. Neher, Science 1992, 256, 498–502.
3062–3075.
[21] R. Mitra, M. Morad, Am.J. Physiol. 1985, 249, H1056–H1060.
[22] E. Polak, Computational methods in optimization, a unified approach,
New York, N.Y. Academic Press, 1971, 53–54.
[23] A.K. Ghose, A. Pritchett, G.M. Crippen,. J. Comput. Chem. 1988, 9,
[5] S. Hopkins, Manufacturing Chemist 1991, 62, 28–30.
[6] D.J. Triggle, Stroke (suppl. IV) 1990, 21, 49–58.
80–90.
[24] V.N. Viswanhadhan, A.K. Ghose, G.R. Revankar, R.K. Robins, J.
Chem. Inf. Comput. Sci. 1989, 29, 163–172.
[7] C.J. Van der Schyf, W. Liebenberg, R. Bornman, T.G. Dekker, P.H.
Van Rooyen, T.G. Fourie, E. Matthee, F.O. Snyckers, S. Afr. J. Chem.
1989, 42, 46–48.
[25] A.K. Ghose, G.M. Crippen, J. Chem. Inf. Comput. Sci. 1987, 27, 21–35.
[26] K.J. Miller, J. Am. Chem. Soc. 1990, 112, 8533–8542.
[8] A.P. Marchand, B.E. Arney, P.R. Dave, N. Satyanarayana, W.H. Wat-
son, A. Nagel, J. Org. Chem. 1988, 53, 2644–2647.
[27] N. Bodor, Z. Gabanyi, C-K. Wong, J. Am. Chem. Soc. 1989, 111,
[9] C.J. Van der Schyf, G.J. Squier, W.A. Coetzee, Pharmacol. Res.
Commun. 1986, 18, 407–417.
3783–3786.
[28] A. Gavezzotti,. J. Am. Chem. Soc. 1983, 105, 5220–5225.
[10] J.J. Van der Walt, C.J. Van der Schyf, J.M. Van Rooyen, J. de Jager,
[29] C.W. Still, A. Tempczyk, R.C. Hawley, T. Hendrickson, J. Am. Chem.
Soc. 1990, 112, 6127–6129.
M.N. Van Aarde, S. Afr. J. Sci. 1988, 84, 448–450.
[11] A.P. Marchand, B.E. Arney Jnr., J. Org. Chem. 1987, 52, 3455–3457.
[12] A.P. Krapcho, A.J. Lovey, Tetrahedron Lett. 1973, 12, 957–960.
[13] C.R. Surapaneni, R. Gilardi, J. Org. Chem. 1986, 51, 2382–2385.
[30] L.B. Kier, Molecular connectivity as a description of structure for SAR
analyses, in S.H. Yalkowsky, A.A. Sinkula, S.C. Valvani (Eds.), Physi-
cal chemical properties of drugs, New York, N.Y. Marcel Dekker,
1980, 227–319.
[14] D.W. Oliver, T.G. Dekker, F.O. Snyckers, Eur. J. Med. Chem. 1991,
26, 375–379.
Received: June 7, 1999 [FP401]
Arch. Pharm. Pharm. Med. Chem. 333, 10–16 (2000)