Synthesis and in vitro platelet aggregation and TP receptor binding studies on bicyclic 5,8-ethanooctahydroisoquinolines and 5,8-ethanotetrahydroisoquinolines

Article abstract of DOI:10.1016/S0968-0896(02)00101-3

Eighteen novel bicyclic 1-substituted benzyl octahydro- and tetrahydroisoquinolines were synthesized and evaluated for human thromboxane A₂/prostaglandin H₂ (TP) receptor affinity and antagonism of TP receptor-mediated platelet aggre

Full text of DOI:10.1016/S0968-0896(02)00101-3

Bioorganic & Medicinal Chemistry 10 (2002) 2779–2793
Synthesis and In Vitro Platelet Aggregation and TP Receptor
Binding Studies on Bicyclic 5,8-Ethanooctahydroisoquinolines and
5
,8-Ethanotetrahydroisoquinolines
a,y
a,
a,{
b
Shankar L. Saha, Victoria F. Roche, * Kathleen Pendola, Mark Kearley,
d
c
d,e
d,e
c
Longping Lei, Karl J. Romstedt, Mark Herdman, Gamal Shams, Vivek Kaisare
c,d
and Dennis R. Feller
aSchool of Pharmacy & Allied Health Professions, Creighton University, 2500 California Plaza, Omaha, NE68005, USA
bDepartment of Chemistry, Creighton University, Omaha, NE68005, USA
cDepartment of Pharmacology, National Center for Natural Products Research, School of Pharmacy, The University of Mississippi,
University, MS 38677, USA
d
Division of Pharmacology, College of Pharmacy, The Ohio State University, Columbus, OH 43210, USA
eDepartment of Biology, Capital University, Columbus, OH 43209, USA
Received 29 May 2001; accepted 24 January 2002
Abstract—Eighteen novel bicyclic 1-substituted benzyl octahydro- and tetrahydroisoquinolines were synthesized and evaluated for
human thromboxane A /prostaglandin H (TP) receptor affinity and antagonism of TP receptor-mediated platelet aggregation. In
2
2
both cases, potency depended more on the presence of methoxy groups on the 1-benzyl moiety than on nitrogen substitution or
extent of oxidation of the isoquinoline ring system. The most potent of the bicyclic compounds retained the 5,8-ethanooctahy-
droisoquinoline ring structure of the parent molecule (1) and required the 3,4,5-trimethoxybenzyl substitution pattern found in the
well-characterized tetrahydroisoquinoline antiplatelet agent trimetoquinol. Differences in nitrogen substituent SAR were noted
between the mono-methoxylated compounds and the 3,4,5-trimethoxybenzyl derivatives. # 2002 Elsevier Science Ltd. All rights
reserved.
À9
Introduction
importance.6 Several potential clinical advantages of
TP receptor antagonists over TXA synthase inhibitors
2
1
0
The pathogenesis of many cardiovascular and throm-
boembolic diseases can be related to metabolites of
arachidonic acid released from plasma membrane
phospholipids. A bioactive metabolite of particular
importance is the potent platelet aggregation inducer
have been suggested.
antagonists also block the effects of platelet aggregatory
prostaglandin precursors, such as PGH . In addition,
2
the TP receptor antagonists do not generate vasocon-
stricting prostaglandins such as PGF , and they are
2a
therapeutically useful after the in vivo generation of
Specifically, TP receptor
1À5
and vasoconstrictor thromboxane A2 (TXA ). The
2
development of compounds that selectively inhibit the
action of TXA at the thromboxane A /prostaglandin
H2 (TP) receptor level or by inhibiting its synthesis
would be of significant scientific and clinical
TXA . Unfortunately, few potent or selective TP recep-
2
tor antagonists have been identified, and those that have
8
2
2
,11À13
are primarily derivatives of prostanoic acid.
Recently cloned,14 the TP receptor is a 343 amino acid
protein coupled to a pertussis toxin-insensitive G pro-
q
tein, and has the characteristic helical structure common
1
5
*
Corresponding author. Tel.: +1-402-280-3191; fax: +1-402-280-
to members of this receptor family. In addition to the
G protein, there is evidence that the platelet TP recep-
1
148; e-mail: roche@creighton.edu
q
y
Current address: MediChem Res. Inc., 1700 S. Mt. Prospect Road,
Des Plaines, IL 60018, USA.
tor might also associate with a high molecular weight
transglutaminase protein known as G as well as to a
h
{
Current address: School of Pharmacy, University of California San
Francisco, S-926, San Francisco, CA 94143-0446, USA.
1
6,17
G13 protein.
0
968-0896/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00101-3

2780
S. L. Saha et al. / Bioorg. Med. Chem. 10 (2002) 2779–2793
Histidine and cysteine residues found in the first extra-
cellular loop of the TP receptor have been hypothesized
to be important in binding agonists to the TP receptor,
but the specific role of these residues has not been iden-
moiety.35 Affinity at both sites was directly related to
the lipophilic nature of the p-substituent and the molar
refraction of one m-substituent. The dependence of b2
receptor affinity on a low or negative resonance para-
meter value for the second m-substitutent indicated the
importance of electronic interactions.
1
8,19
tified.
Glycoslyation of asparagine residues at posi-
tions 4 and 16 of the protein is essential for high affinity
binding and specificity, although the addition of sugar
to only one of the residues will permit ligand binding.20
It is believed that two subtypes of the receptor exist, one
that regulates aggregatory/secretory responses and
another that controls calcium mobilization and shape
TMQ’s antiplatelet and b adrenoceptor agonist activ-
ities are stereoselective,
3
1,36À40
with the R(+) isomer
exerting the anti-platelet activity through the antagon-
ism of TXA and related endoperoxides at the receptor
2
5
,21
39,41,42
changes.
The platelet TP receptor is believed to be
level.
S(À)-TMQ is the most potent b adreno-
distinct from the binding site found in vascular smooth
muscle, although the two receptors are expressed by a
ceptor agonist, but is also capable of selectively inhibit-
ing platelet aggregation induced by non-prostanoids
such as phospholipase C, calcium ionophores (e.g.,
A23187) phorbol esters and low-dose thrombin (in the
5
,22
common gene.
3
9,43
Until recently, little was known about TP receptor
topography which would facilitate a strictly rational
approach to the design of specific antagonists, although
presence of aspirin).
In addition to anti-aggregatory
action, TMQ also exhibits lipolytic, bronchial relaxant,
hypotensive, cardiostimulant and uterine-stimulant
2
3,24
44
a few efforts at rational design have been attempted.
While the tertiary structure of the protein has not yet
been mapped, a 1993 molecular modeling study by
activities. Thus, while not selective itself, TMQ can
serve as a prototype benzylisoquinoline-based inhibitor
of prostaglandin-dependent and independent platelet
aggregation, and as a template for the design of selective
amine-containing antiplatelet agents.
2
4
Yamamoto et al. yielded a three-dimensional receptor
2
5
model. In the following year, Jin and Hopfinger iden-
tified a spatial pharmacophore for the four recognition
sites of the TP receptor which accommodates five
known carboxylate-containing TP receptor antagonists,
four of which were based on the structure of the endo-
Two derivatives of bicyclic 1-p-methoxybenzyl-5,8-
ethanooctahydroisoquinoline (1 and 2, Fig. 2) have
4
5
been synthesized in our laboratory and subsequently
shown to non-selectively antagonize platelet aggrega-
tion. Tertiary amine 2, the more potent of the two,
exhibited weak TP receptor antagonist action but was
up to 50-fold more potent than either TMQ or
N-methyl-TMQ as an inhibitor of prostaglandin-inde-
pendent phorbol ester (PMA) and A23187-induced pla-
genous agonist TXA . A subsequent four-dimensional
2
QSAR study by Albuquerque et al. provided structural
requirements and restrictions for TP receptor antago-
nists of the 7-oxabicyclo[2.2.1]heptane oxazole struc-
2
6
tural type. Early TP receptor antagonists suffered
from serious pharmacokinetic-based limitations.27
However, some members of the oxabicycloheptane class
of carboxylate-containing antiplatelet agents, such as
ifetroban sodium,2 have demonstrated good TXA2
antagonist potency with pharmacokinetic profiles con-
sistent with clinical utility. The aforementioned spatial
pharmacophore studies should prove useful in the
design of new, even more highly potent anionic TP
receptor antagonists.
4
6
telet aggregation. These compounds also possess a2
adrenoceptor antagonist activity.
8
Compound Design
In order to study the impact of (1) lipophilic and elec-
tronic character of the 1-benzyl substituent and (2)
steric bulk of the amine substitutent on the extent of TP
receptor antagonist action in bicyclic 5,8-ethanooctahy-
droisoquinoline molecules, compounds 3–12 (Fig. 2)
However, not all TP antagonists are modeled after
2
9
TXA2. Trimetoquinol (TMQ, Fig. 1), is an amine-
containing tetrahydroisoquinoline analogue that exhi-
bits both platelet aggregation inhibitory properties and
were synthesized. The p-CH and Cl substituents found
3
in compounds 4–9 and 14–15 are both more lipophilic
than the OCH found on the parent structures 1 and 2,
3
0À34
b-adrenoceptor stimulatory activity.
A
recent
3
QSAR study related binding affinity at both b adreno-
ceptors and TP receptors to physicochemical para-
meters of the aromatic substituents of the 1-benzyl
but differ in their p electron donating/withdrawing
character. Their relative TP receptor antagonist poten-
cies would shed light on the relative importance of
lipophilicity and electron density distribution within the
aromatic arm of the structure. The 3,4,5-trimethoxy
substitution pattern found in 10–12 and 16 mimics the
TMQ structural prototype and tests the hypothesis that
this structural arrangement is essential for high TP
receptor antagonism in the bicyclic ligands.
2
The aromatic analogues 13–16 (Fig. 2) were byproducts
of the Bischler–Napieralski cyclization/reduction step of
the synthetic pathway, and were also tested for anti-
platelet action. Since it was our hypothesis that the
bicyclic antagonists bind to the TP receptor through a
Figure 1. Structure of trimetoquinol (TMQ).

S. L. Saha et al. / Bioorg. Med. Chem. 10 (2002) 2779–2793
2781
Figure 2. 5,8-Ethanooctahydroisoquinolines and 5,8-ethanotetrahydroisoquinolines synthesized as TP receptor antagonists.
cationic amino nitrogen, non-ionizable 17 and perma-
nently cationic 18 (Fig. 2) were synthesized.
methods, and the mixture was subjected to acylation
and Bischler–Napieralski cyclization to produce target
compounds 1 and 2. In addition, some of the bicyclic
octane and octene precursors to the synthesis of the
amino alcohol were quite volatile and difficult to isolate
without significant product loss. The new route to the
desired bicyclic amine 25 described in Scheme 2 avoids
these problems, and represents a significant improve-
ment over the original pathway.
Structure–activity relationship studies have shown that
secondary amine-containing TMQ analogues provide
significantly greater TP receptor antagonist action than
4
4
any tertiary derivative. However, previous studies
with 1 and 2 have documented that secondary amine 1 is
much less active than its N–CH analogue 2 in inhibit-
3
ing platelet aggregation stimulated by the stable TP
receptor agonist U46619. In fact, 2 was consistently
more active than 1 in blocking aggregation in response
to mediators of both prostanoid-dependent and -inde-
The synthesis of key intermediate 25 began with a
Diels–Alder reaction with commercially available
4
8
cyclohexadiene and ethyl acrylate (Scheme 2). An
overnight reaction at 140 C in a sealed tube gave the
4
6
ꢀ
bicyclic ester 19 in 85% yield. Catalytic reduction of the
pendent aggregation pathways. Given this critical
SAR difference it was of interest to investigate whether
bulkier nitrogen substituents such as N-p-methoxy-
benzyl (among the more potent and platelet-selective of
nitrogen substituents in tertiary TMQ-based TP recep-
tor antagonists) also augment the antagonist potency of
the bicyclic 5,8-ethanooctahydroisoquinoline structures
4
9
olefin provided the known saturated analogue 20
which was subjected to lithium diisopropylamide-medi-
5
0
ated selenylation with phenylselenium bromide. Oxi-
dative deselenylation with 30% hydrogen peroxide
afforded 21, which had the requisite double bond in
4
7
50
(3, 6, 9, and 12).
desired position. The yield of 21 was routinely above
90%, and the NMR was consistent with that of the
5
1
known methyl ester. This conjugated ethyl ester was
purified by repeated column chromatography and frac-
tional distillation to single peak purity on a capillary gas
chromatrograph. However, an impurity that co-eluted
and co-distilled with the desired compound prevented
the percent carbon observed upon elemental analysis
from matching the anticipated value to within the
requisite Æ0.4%. High-resolution electron impact mass
spectrometry documented a molecular formula con-
sistent with the given structure.
Results
Chemistry
The syntheses of the bicyclic 5,8-ethanooctahydro-
4
5
isoquinolines 1 and 2 have been previously described.
A major drawback of the original pathway was the
generation of a mixture of olefinic isomers in the synth-
esis of the critical intermediate 2-(aminoethyl)bicyclo-
[
2.2.2]oct-2-ene (25). Compound 25 and its exocyclic
olefinic isomer were produced in equivalent amounts
when the precursor amino alcohol was subjected to
heat-catalyzed elimination after in situ conversion to the
bromide (Scheme 1). These isomeric olefins proved very
difficult to separate with conventional purification
Conversion of the ester 21 to the corresponding cyano-
methyl derivative 24 was achieved by the following
sequence of reactions (Scheme 2). Reduction of 21 to
the allylic alcohol 22 was accomplished with fresh di-
5
2
isobutylaluminum hydride (Dibal-H) in toluene.

2782
S. L. Saha et al. / Bioorg. Med. Chem. 10 (2002) 2779–2793
Scheme 1. Synthesis of 2-aminoethylbicyclo[2.2.2]octene isomers. (a) Ph
3 2
PBr , p-xylene, reflux.
ꢀ
Scheme 2. Synthesis of 2-aminoethylbicyclo[2.2.2]oct-2-ene (25): (a) 140 C, sealed tube, 85%; (b) H
ꢀ
2
, Pd, EtOH, rt, 89%; (c) (i) LDA, THF,
ꢀ
ꢀ
ꢀ
À78 C; (ii) PhSeBr, THF, À78 to 0 C; (iii) 30% H
2
O
2
,HOAc, 0 C to rt, 95%; DiBAL-H, CH Cl
2
2
, À78 C, 85%; acetyl chloride, pyridine, CH
2 2
Cl ,
ꢀ
rt, 90%; (d) Pd(Ph P) , CH
4
3
3
CN, toluene, reflux, 92%; (e) AlH
3
or LiAlH
4
, THF, 0 C, 79%.
Despite utilizing conditions that had produced quanti-
tative yields of cyclic allylic alcohols from esters in other
5
3
systems, the yield of this reaction was variable,
particularly upon scale-up. In small scale reactions,
yields of 85% were typical. Once isolated and purified,
2
2 was converted to its corresponding acetate 23 in 90%
5
4
yield with acetyl chloride and pyridine. The acetate
was reacted with trimethylsilyl nitrile in the presence of
palladium tetrakistriphenylphosphine in refluxing tolu-
5
5
ene to provide the nitrile 24 in 92% yield. All of these
small bicyclic intermediates suffered the same problem
as the conjugated ester 21 with respect to purification.
Despite repetitive column chromatography and frac-
tional distillation of the products to single peak purity
on the gas chromatograph, high resolution electron
impact GC/mass spectrometry was again required to
give unequivocable proof of molecular formula. Reduc-
tion of nitrile 24 with aluminum hydride generated in
situ from lithium aluminum hydride and aluminum
Figure 3. Benzoquinolizine rearrangement product (29).
oxide, followed by a basic workup and immediate
reduction of the intermediate imines with potassium
borohydride in methanol, provided a mixture of two
major products. These products were purified on a silica
gel column and characterized as the desired 5,8-ethano-
octahydroisoquinolines (4, 7, and 10) and aromatic
5,8-ethanotetrahydroisoquinoline (13–16) byproducts.
These octahydro and tetrahydro derivatives were rou-
tinely produced in 30–35 and 20–25% yields, respec-
tively. All N-p-methoxybenzyl analogues (3, 6, 9, and 12)
were synthesized in moderate to high yield by reacting the
secondary amines (1, 4, 7, and 10) with p-methoxybenzyl
5
6
chloride gave the pure endocyclic amine 25 in 79%
yield. A similar reaction using lithium aluminum
5
7
hydride alone provided the desired amine in 83%
yield.
The condensation of 25 with the appropriately sub-
stituted phenylacetyl chlorides 26 (which were generated
via reaction of the commercially available acid and
oxalyl chloride in dichloromethane) in the presence of
dry pyridine provided desired amides 27 in 64–75%
6
1
chloride in the presence of triethylamine.
N-Methylation of the p-chloro analogue 4 and the
p-methyl derivative 7 could be accomplished by reaction
5
8
62
yield after chromatographic purification (Scheme 3).
The p-methoxyphenylacetyl chloride was generated by
with 37% formaldehyde/formic acid. Under these
same conditions, however, the 3,4,5-trimethoxy con-
gener underwent intramolecular cyclization to produce
a benzoquinolizine product 29 (Fig. 3). This product
reacting the commercially available acid with thionyl
5
4
chloride, as previously described. Bischler–Napieralski
5
9,60
1
13
13
cyclization
of these amides with phosphorus pent-
was characterized by H and C NMR, and the
C

S. L. Saha et al. / Bioorg. Med. Chem. 10 (2002) 2779–2793
2783
Scheme 3. Synthesis of 5,8-ethanooctahydroisoquinolines and 5,8-ethanotetrahydroisoquinolines (3–16): (a) pyridine, THF, rt, 64–75%; (b) P
toluene, reflux, 20–25% of tetrahydro derivatives; (c) KBH , NH OH, MeOH, rt, 30–35% from 27; (c) H CO, HCOOH or H CO, NaCNBH
CH CN reflux, 25–78%; p-MeObenzyl chloride, Et N, CHCl , rt, 65–98%.
2
O
5
,
,
4
4
2
2
3
3
,
3
3
Pharmacology
resonance values found were in close agreement with a
highly similar dibenzo[a,g]quinolizine alkaloid, tetra-
The results of the evaluation of racemic bicyclic com-
pounds 1–18 as antagonists of U46619-induced platelet
aggregation and as competitive ligands for
6
3
hydropalmatine.
Methylation of polymethoxylated
analogue 10 to provide the desired tertiary amine 11 was
accomplished in 75–78% yield by reacting the second-
ary amine with 37% formaldehyde and sodium cyano-
borohydride in acetonitrile at room temperature.6
3
[ H]SQ29,548-labeled TP receptor sites in human plate-
lets are provided in Table 1.
4
Preliminary studies documented that none of the bicy-
clic 5,8-ethanooctahydroisoquinoline molecules caused
platelet aggregation in platelet rich plasma. With the
exception of 9, all bicyclic compounds were weak aggre-
gation antagonists with IC50 values against U46619-
induced platelet aggregation ranging from 5.3 to
550 mM. The most active platelet aggregation inhibitor
of the novel bicyclic structures was 10, which retained
the secondary amine and trimethoxybenzyl substitution
pattern of the prototypical isoquinoline TP receptor
antagonist TMQ. Compound 10 was approximately
seven times less active than TMQ in inhibiting throm-
boxane-induced platelet aggregation, and had about
one-third the affinity of TMQ for thromboxane binding
sites on platelets. The high significance (p<0.0001) of
the relationship between aggregation (pIC ) and dis-
The formamide derivative 17 was synthesized in 61%
yield from secondary amine 4 by reaction with ethyl
6
5
formate (Scheme 4). The NMR spectrum of this com-
pound was complicated by the presence of two isomeric
forms. A similar spectroscopic situation was noted in
4
5
the N-formylated analogue of 1. The quaternary
ammonium derivative 18 was synthesized in 76% yield
4
4
from tertiary amine 5 by reaction with methyl iodide
Scheme 4).
(
All bicyclic octahydro- and tetrahydroisoquinolines
1
13
were analyzed by H and C NMR spectrometry.
Characterization and evaluation of the purity of the
bicyclic octahydro- and tetrahydroisoquinoline hydro-
chloride or methanesulfonate salts by melting point was
difficult, as the compounds slowly sublimed upon heat-
ing. The melting points reported in the Experimental
represent the temperature at which a single crystal dis-
appeared.
5
0
3
placement of [ H]SQ29,548 binding to TP receptors
(pK ) indicates that the anti-aggregatory activities of
I
these compounds are through blockade of TP receptor
sites of human platelets (Fig. 4).

2784
S. L. Saha et al. / Bioorg. Med. Chem. 10 (2002) 2779–2793
ꢀ
Scheme 4. Synthesis of formylated and quaternary derivatives of a p-chloro-5,8-ethanooctahydroisoquinoline: (a) HCOOC
1%; (b) H CO, HCOOH, reflux, 40%; (c) MeI, KHCO , MeOH, rt, 76%.
H
2 5
, benzene, 50 C,
6
2
3
3
Table 1. Inhibitory effect of trimetoquinol (TMQ) versus 5,8-ethanooctahydroisoquinolines on U46619-induced aggregation and [ H]SQ29,548
binding to TP-receptors of human plateletsa
Compound
Platelet aggregation
pIC50b
Platelet binding
pK c
IC50 (mM)
P.R.d
1.0
K
i
(mM)
i
P.R.d
1.0
Trimetoquinol (TMQ)
0.81
6.09Æ0.09
0.62
6.21Æ0.13
2
4
3
p-OCH
1
3
series (R =R =H; R =OCH
R =H (1)
1
3
)
177.0
103.1
541.2
377.3
3.45Æ0.06
3.99Æ0.03
3.27Æ0.03
3.42Æ0.08
0.005
0.008
0.002
0.002
270.1
61.3
218.8
43.3
3.57Æ0.14
4.21Æ0.08
3.66Æ0.13
4.36Æ0.06
0.002
0.010
0.003
0.014
3
R =CH (2)
1
R =p-methoxybenzyl (3)
Tetrahydro analogue (13)
2
4
3
p-Cl series (R =R =H; R =Cl)
1
R =H (4)
1
128.8
124.0
374.4
553.8
530.9
180.0
3.89Æ0.10
3.91Æ0.12
3.43Æ0.04
3.62Æ0.09
3.28Æ0.08
3.77Æ0.10
0.006
0.010
0.007
0.001
0.002
0.005
69.7
40.7
4.16Æ0.08
4.39Æ0.74
3.65Æ0.20
3.99Æ0.05
4.02Æ0.03
4.17Æ0.07
0.009
0.015
0.003
0.006
0.006
0.010
R =CH
1
3
(5)
R =p-methoxybenzyl (6)
222.2
102.1
95.6
1
R =formyl (17)
1
R =N,N-dimethyl (18)
Tetrahydro analogue (14)
64.9
2
4
3
p-CH
1
3
series (R =R =H; R =CH
R =H (7)
1
3
)
154.9
77.6
>800
286.7
3.81Æ0.05
4.11Æ0.19
<3.10
3.54Æ0.08
0.005
0.010
<0.001
0.003
66.1
37.4
48.0
95.1
4.18Æ0.09
4.43Æ0.15
4.32Æ0.02
4.02Æ0.05
0.009
0.017
0.013
0.007
3
R =CH (8)
1
R =p-methoxybenzyl (9)
Tetrahydro analogue (15)
2
3
4
Tri-OCH
1
3
series (R =R =R =OCH
R =H (10)
1
3
)
5.3
13.8
66.4
133.5
5.28Æ0.27
4.86Æ0.05
4.18Æ0.08
3.87Æ0.06
0.15
1.8
8.1
6.7
9.3
5.74Æ0.05
5.09Æ0.09
5.17Æ0.08
5.03Æ2.03
0.34
R =CH
1
3
(11)
R =p-methoxybenzyl (12)
Tetrahydro analogue (16)
0.059
0.012
0.006
0.077
0.093
0.067
a
Data represent the meanÆSEM of n=3–6 experiments. U46619 (0.5–1.5 mM) was used as inducer for aggregation in the presence of 1 mM aspirin
ꢀ
3
2
incubated for 1 min at 37 C. [ H]SQ29,548 (1 or 5 nM) was used to label TXA sites; and nonspecific binding was defined by addition of 50 mM of
SQ29,548.
b
IC50=Àlog IC50, which is the inhibitory concentration reducing half of the maximal response to the inducer.
c
65
pK
i
=Àlog K
Potency ratio (P.R.)=(IC50 or K
experiment. TMQ is the reference compound.
i
, and was determined as IC50/(1+[ligand]/K
d d
), where K =3.1 nM is taken from previous data.
d
i
TMQ)/(IC50 or K ) bicycloisoquinolines, where TMQ value is the mean of individual control tested in each
i

S. L. Saha et al. / Bioorg. Med. Chem. 10 (2002) 2779–2793
2785
relationship is reversed in the 3,4,5-trimethoxy deriva-
tives 10 and 11 and TMQ, indicating that binding to the
TP receptor is influenced by the 1-benzyl moiety.
Conclusion
Eighteen novel bicyclic 1-substituted benzyloctahydro-
and tetrahydroisoquinolines were synthesized and eval-
uated for their ability to (1) bind to the human TP
receptor and (2) antagonize the activity of a stable
TXA analogue at that receptor surface. The activity of
2
the bicyclic ligands was compared to that of the well-
known tetrahydroisoquinoline platelet aggregation
inhibitor TMQ. The presence of at least one aromatic
methoxy group was crucial to the antiplatelet and TP
receptor binding activity of the bicyclic ligands. The
most potent of these compounds retained the 5,8-ethano-
octahydroisoquinoline ring structure of the parent
molecule (1) and required TMQ’s 3,4,5-trimethoxy sub-
stitution pattern on the 1-benzyl moiety. An interesting
inversion in nitrogen substituent SAR resulted when
three methoxy groups (as opposed to one) were present;
the bicyclic secondary amine was the most active mem-
ber of the trimethoxybenzyl family of compounds.
Figure 4. Correlation of anti-aggregatory activities (pIC50) and inhi-
3
i
bition of [ H]SQ29548 binding (pK ) to TP receptors in human plate-
lets by bicyclic octahydro- and tetrahydroisoquinolines (r=0.922,
p<0.0001).
Discussion
Analysis of structure–activity relationships for bicyclic
5
,8-ethanooctahydroisoquinolines clearly shows that the
presence of the TMQ-like trimethoxy substituent pat-
tern on the 1-benzyl group is important for optimal
anti-thromboxane activities of aggregation and TP
receptor binding. The work of Kumar and Singh sug-
gests that hydrophobicity of the p-substituent of TMQ
Adding bulk (beyond CH ) to the amino nitrogen of the
3
bicyclic compounds, or oxidizing the nitrogen-contain-
ing ring to a pyridine system, does not promote anti-
platelet activity or TP receptor affinity. Bicyclic 5,8-
ethanooctahydroisoquinolines containing these nitrogen
modifications were significantly less active in inhibiting
3
5
is important to TP receptor binding. Despite the
increase in lipophilicity compared to a methoxy group,
the lack of potency of all analogues of the p-chloro and
p-methyl series indicates that other physicochemical
parameters are operational in this bicyclic series. One
possibility is that hydrogen bonding in this molecular
area is essential for anti-platelet activity and TP recep-
tor affinity in bicyclic 5,8-ethanooctahydroisoquino-
lines, although additional analogues with varying steric,
electronic and hydrogen-bonding properties would need
to be synthesized and tested to confirm or refute the
hypothesis. The lack of appreciable activity of the
p-chloro analogues containing either the N,N-dimethyl
or the N-formyl groups precluded assessment of the
importance of the cationic nitrogen in binding these
compounds to the TP receptor. Studies will continue
with the trimethoxy series now that it has been identi-
fied as the most potent of the bicyclic receptor ligands.
3
U46619-induced platelet aggregation and [ H]SQ29,548
binding compared to N-methyl or secondary amine
derivatives. The most active bicyclic ligand (10) was
significantly less active than TMQ as both an inhibitor
of U46619-mediated platelet aggregation and in
3
displacing [ H]SQ29,548 from TP receptors.
Experimental
Chemistry
General. Melting points were determined on a Fisher–
1
Johns melting point apparatus and are uncorrected. H
1
3
and C NMR spectra were recorded on either a Varian
INOVA or Unity 300 MHz instrument. Chemical shifts
are reported in d units relative to CHCl (7.27 ppm).
3
Gas chromatography was conducted on both Hewlett
Packard HP 5988A and 5890 Series II chromatographs.
High resolution mass spectra were recorded in an EI
mode on a Micromass Autospec mass spectrometer at the
Nebraska Center for Mass Spectrometry (Lincoln, NE).
Microanalyses were performed by either Galbraith
Laboratories (Knoxville, TN) or Atlantic Microlabs, Inc.
(Norcross, GA). Infrared (IR) spectra were recorded on
either a Perkin-Elmer 1600 FT instrument or a Nicolet
In general, the nature of the amine substituent (H, CH3,
p-methoxybenzyl) or the extent of unsaturation in the
nitrogen-containing ring (octahydro versus tetrahydro
analogues) did not produce significant changes in the
inhibition of platelet aggregation and competitive
receptor binding. However, in the most active methoxy-
containing structures, the presence of multiple methoxy
groups on the aromatic ring appears to change the
preference of the receptor for the extent of substitution
on the amino nitrogen. In the original p-methoxy mol-
ecules, tertiary bicyclic ligand 2 was more active than
secondary analogue 1 as both an inhibitor of throm-
boxane-induced aggregation and in displacing the tri-
tiated TP receptor antagonist from its binding site. The
Avatar 380 FT-IR. R values were determined from TLC
f
performed on fluorescent, plastic-backed Selecto Scientific
silica gel 60 plates. Column chromatography was carried
out with Merck or Fisher silica gel 60 (200–425 mesh).
Solvents were dried by distillation after reflux with appro-
priate drying agents and stored over molecular sieve.

2786
S. L. Saha et al. / Bioorg. Med. Chem. 10 (2002) 2779–2793
À1
2-Ethoxycarbonylbicyclo[2.2.2]oct-5-ene (19). A thick-
walled reaction tube was charged with cyclohexadiene
(thin film) cm
stretch), 1711.0 (carbonyl stretch), 1260.3 (C–O stretch),
1081.1 (O–C–C stretch).
3052.6 (¼CH stretch), 2945.1 (CH
(
methylene blue (0.018 g, 0.05 mmol). The tube was
2 g, 25 mmol), ethyl acrylate (2.5 g, 25 mmol), and
ꢀ
sealed and the solution heated at 140 C for 12–18 h.
Distillation (76 C/<1 mm Hg; lit 106 C/15 mm Hg )
1
2-Hydroxymethylbicyclo[2.2.2]oct-2-ene (22). A round-
bottom flask was charged with 21 (16.3 g, 90 mmol) and
ꢀ
afforded 19 as a clear liquid (3.83 g, 85% yield). H
ꢀ
66
ꢀ
dry CH Cl (200 mL) and cooled to À78 C. To this was
2
2
NMR (CDCl ) d 6.3 (t, 1H, C¼CH), 6.13 (t, 1H,
added diisobutylaluminum hydride (180 mL of 1 M in
hexanes, 0.18 mol) and the reaction allowed to stir for
3
C¼CH), 4.1 (m, O–CH ), 2.93 (m, 1H, C2-H), 2.58 (m,
2
ꢀ
2
1
H, C1-H, C4-H), 1.8–1.4 (m, 4H, C3-H , C8-H ), 1.35–
2
2 h. The solution was then allowed to warm to À30 C
2
1
3
.25 (m, 5H, C7-H , CH ); C NMR with DEPT
3
and CH OH (30 mL) was added. After 20 min, 100 mL
3
2
(
(
(
CDCl ) d 175.6 (carbonyl carbon), 135.3 (C5), 131.6
of a 50:50 mixture of CH OH/H O was added and the
3
3
2
C6), 60.3 (O–CH ), 43.1 (C2), 32.7 (C1), 30.1 (C3), 29.6
C4), 25.5, 24.6 (C7, C8), 14.4 (CH3).
reaction allowed to warm to room temperature and
stirred for 30 min. The solution was filtered through
2
1
Celite , transferred to a separatory funnel, and the
layers separated. The aqueous layer was then extracted
with CH Cl (100 mL), the organic layers combined,
2
-Ethoxycarbonylbicyclo[2.2.2]octane (20). A pressure
bottle was charged with 19 (2.6 g, 15 mmol), 10% Pd/C
0.75 g), and 100% EtOH (50 mL). The bottle was
2
2
(
and dried over Na SO . The solution was filtered and
2 4
placed in the Parr hydrogenator and evacuated by
aspirator vacuum. The system was then filled with H2
the solvent removed in vacuo. Column chromatography
(R 0.22, 4.5:0.5 petroleum ether:ethyl acetate) afforded
f
(
filtered through Celite and the solvent removed in
50 psi) and allowed to shake for 18 h. The solution was
1
22 as a yellow oil (11.0 g, 88% yield). Distillation under
ꢀ
vacuum (96–97 C at <1 mm Hg) afforded 22 as a thick
1
1
vacuo to give 20 as a clear liquid (2.4 g, 89% yield). H
NMR (CDCl ) d 4.1 (q, 2H, O–CH ), 2.58 (m, 1H, C2-
colorless oil in 85% overall yield. H NMR (CDCl ) d
3
6.1 (d, 1H, C3-H), 4.2 (s, 2H, C9-H ), 2.5 (m, 2H, C1-H,
C4-H), 1.8 (s, 1H, OH), 1.6–1.2 (m, 8H, C5-H , C6-H ,
3
2
2
H), 2.0 (m, 2H, C1-H, C4-H), 1.78–1.4 (m, 13H, C3-H2,
C5-H , C6-H , C7-H , C8-H , CH ), 1.2 (t, 3H, CH );
2
2
1
3
C7-H , C8-H ); C NMR (CDCl ) d 146.2 (C2), 127.7
2
2
2
2
3
3
2
2
3
1
3
C NMR (CDCl ) d 176 (carbonyl carbon), 60.3 (O–
(C3), 64.6 (C9), 31.6 (C1), 30.0 (C4), 26.5, 26.1 (C5, C6,
C7, C8). Anal. calcd for C H O: C, 78.26; H, 10.14;
found: C, 77.32; H, 10.11. HRMS (EI) calcd for
3
CH ), 42.1 (C2), 28.4 (C1), 27.6 (C3), 26.5 (C4), 25.4,
2
2
9
14
5.4, 24.0, 22.1 (C5, C6, C7, C8), 14.5 (CH3).
+
C H O (m ) 138.1045. Found 138.1042. IR (thin film)
9
14
À1
2-Ethoxycarbonylbicyclo[2.2.2]oct-2-ene (21). A round-
bottom flask was charged with diisopropylamine (11.1 g,
cm 3316.8 (OH stretch), 2940.0 (CH stretch), 1451.1
(OH bend), 1045.2 (C–O stretch).
1
4.8 mL, 0.11 mol) and dry THF (300 mL). The solution
ꢀ
was then cooled to À78 C and to this was added drop-
wise 1.6 M n-butyllithium (75 mL, 0.12 mol). The yellow
solution was allowed to stir for 60 min. To this was
added 20 (17.3 g, 0.10 mol) dissolved in dry THF
2-Acetoxymethylbicyclo[2.2.2]oct-2-ene (23). A round-
bottom flask was charged with 22 (5.0 g, 36 mmol), dry
pyridine (6.8 g, 7.0 mL, 87 mmol), and dry CH Cl
2
2
ꢀ
(50 mL) and cooled to 0 C. To this was added acetyl
chloride (3.14 g, 2.8 mL, 40 mmol) dissolved in dry
CH Cl (10 mL). The reaction was allowed to stir at
0 C for 20 min and at room temperature for 2 h. To this
mixture was added H O (50 mL) and the solution
(
40 mL) and the reaction allowed to stir for 60 min at
ꢀ
bromide (25 g, 0.11 mol) dissolved in dry THF (50 mL).
À78 C. To this solution was added phenylselenium
2
2
ꢀ
ꢀ
to stir at 0 C for 10 min. H O (60 mL) followed by 30%
The reaction was allowed to warm to 0 C and allowed
ꢀ
H O (32 mL) and acetic acid (12 mL) were added and
2
transferred to a separatory funnel. The organic layer
was removed and the aqueous layer extracted with
CH Cl (25 mL) and the combined organic layers dried
2
2
2
the reaction allowed to slowly warm to room tempera-
ture. The solution was poured into a mixture of 10%
NaHCO (600 mL) and CH Cl (600 mL) and allowed
2
2
over Na SO . The drying agent was filtered off and the
2
4
solvent removed in vacuo. Column chromatography (Rf
0.59, 4.5:0.5 petroleum ether:ethyl acetate) afforded 23
as a yellow oil (3.9 g, 60% yield). Alternatively, vacuum
3
2
2
to stir for 15 min. The entire solution was transferred to
a separatory funnel and the layers separated. The
organic layer was washed with 1 M HCl (300 mL),
saturated NaCl (300 mL), and dried over Na SO . The
ꢀ
distillation (100 C at <1 mm Hg) afforded clean 23 as a
1
colorless liquid in 90% yield. H NMR (CDCl ) d 6.2
2
4
3
solution was filtered and the solvent removed in vacuo.
Column chromatography (R 0.61, 4.5:0.5 petroleum
(d, 1H, C3-H), 4.6 (s, 2H, C9-H ), 2.5 (m, 2H, C1-H,
C4-H), 2.1 (s, 3H, CH ), 1.6–1.2 (m, 8H, C5-H , C6-H ,
2
f
3
2
2
1
3
ether/diethyl ether) afforded 21 as a yellow oil (16.3 g,
9
C7-H , C8-H ); C NMR (CDCl ) d 171.2 (carbonyl
2
2
3
ꢀ
1 mm Hg) afforded 21 as a colorless liquid. H NMR
5% yield). Distillation under vacuum (93–94 C at
1
carbon), 141.2 (C2), 131.2 (C3), 65.9 (C9), 31.8 (C1),
30.1 (C4), 26.3, 25.9 (C5, C6, C7, C8), 21.2 (CH ). Anal.
<
3
(
CDCl ) d 7.3 (d, 1H, C3-H), 4.2 (q, 2H, C9-H ), 3.2 (s,
calcd For C H O : C, 73.33; H, 8.89; found: C, 71.76;
11 16
3
2
2
+
1
H ), 1.2–1.4 (m, 7H, C6-H , C8-H , C10-H ); C NMR
H, C1-H), 2.7 (d, 1H, C4-H), 1.6 (d, 4H, C5-H , C7-
1
H, 8.83. HRMS (EI) calcd for C H O (m ) 180.1150
2
2
11 16
3
À1
Found: 180.1145. IR (thin film) cm
stretch), 2940.0 (CH stretch), 1741.7 (carbonyl stretch),
1245.0 (C–O stretch), 1024.8 (O–C–C stretch).
3032.2 (¼CH
2
2
2
2
(
(
CDCl ) d 165.4 (carbonyl carbon), 145.6 (C3), 138.0
C2), 60.2 (O–CH ), 31.0 (C1), 29.2 (C4), 25.6, 25.3 (C5,
3
2
C6, C7, C8), 14.4 (CH ). Anal. calcd For C H O : C,
3
11 16
2
7
calcd for C H O (m ) 180.1150. Found 180.1149. IR
3.33; H, 8.89; found: C, 72.37; H, 8.88. HRMS (EI)
+
2-Cyanomethylbicyclo[2.2.2]oct-2-ene (24). A round-
bottom flask was charged with 23 (4.0 g, 23 mmol),
1
1
16
2

S. L. Saha et al. / Bioorg. Med. Chem. 10 (2002) 2779–2793
2787
1
tetrakis(triphenylphosphine)palladium (1.37 g, 1 mmol),
and dry toluene (75 mL). To this mixture was added
ide: H NMR (CDCl ) d 7.2 (s, 4H, aromatic), 4.20 (s,
3
2H, CH ), 2.40 (s, 3H, CH ). 3,4,5-Trimethoxyphenyl
1
2
3
(
allowed to reflux for 18 h. The reaction was cooled, fil-
CH ) SiCN (4.7 g, 6.4 mL, 48 mmol) and the reaction
acetyl chloride: H NMR (CDCl ) d 6.47 (s, 2H, aro-
3
3
3
matic protons), 4.08 (s, 2H, CH ), 3.87 (two s, 9H,
OCH3).
2
1
tered through Florisil , and the solvent removed in
vacuo. Column chromatography (R 0.58, 50:50 petro-
f
leum ether/CH Cl ) afforded 24 as a yellow oil (3.0 g,
2
[[[(Substituted benzyl)carbonyl]amino]ethyl]bicyclo[2.2.2]
oct-2-ene (27). The amides generated from the con-
densation of amine 25 with the substituted acid chlor-
ides 26 were prepared in the following manner. A
solution of 8.28 mmol of 26 and 4.87 mL dry pyridine,
2
ꢀ
9
<
2% yield). Distillation under vacuum (101 C at
1 mm Hg) provided 24 as a colorless liquid in 71%
1
yield. H NMR (CDCl ) d 6.25 (d, 1H, C3-H), 3.2 (s,
3
2
H, C9-H ), 2.6 (m, 1H, C4-H), 2.43 (s, 1H, C1-H), 1.7–
1
2
3
ꢀ
1
.2 (m, 8H, C5-H , C6-H , C7-H , C8-H ); C NMR
2
under nitrogen, was taken to 0 C. To this mixture was
2
2
2
with DEPT (CDCl ) d 134.6 (C2), 131.7 (C3), 117.7
added, in a dropwise fashion, a solution of 8.28 mmol
acid chloride 26 in 35 mL dry THF, and the solution
stirred at room temperature for 24 h. The solvent and
3
(
C8), 22.9 (C9). Anal. calcd For C H N: C, 81.63; H,
cyano), 33.9 (C1), 30.5 (C4), 26.3. 25.8 9 (C5, C6, C7,
1
0
13
8
.84; N, 9.52; found: C, 80.68; H, 8.85; N, 9.55. HRMS
+
excess pyridine were removed in vacuo, H O (50 mL)
2
(
EI) calcd for C H N (m ) 147.1048. Found:
1
and diethyl ether (100 mL) were added to the residue,
and the phases were separated. The aqueous phase was
extracted twice with diethyl ether, and the combined
0
13
À1
1
2
47.1044. IR (thin film) cm 3042.1 (¼CH stretch),
942.6 (CH stretch), 2249.8 (nitrile stretch).
organic extracts washed sequentially with H O, 10%
2
2
-(2-Aminoethyl)bicyclo[2.2.2]oct-2-ene (25). A round-
bottom flask was charged with LiAlH (1.22 g, 32 mmol)
HCl, and dried over anhydrous Mg SO . The crude
2
4
products were purified on a silica gel column. A gra-
dient solvent system of 25–66% diethyl ether in petro-
leum ether was used to purify the p-methoxy, p-methyl,
and p-chloro analogues. The trimethoxy analogue was
purified by eluting with petroleum ether/diethyl ether
(1:1) followed by 2–6% ethyl acetate in ethyl ether. The
chromatographed amides were obtained in 64–75%
4
ꢀ
and dry THF (35 mL) and cooled to 0 C. To this mix-
ture was added 24 (2.4 g, 16 mmol) dissolved in dry
THF (10 mL) and the reaction allowed to stir for 3 h.
The reaction was slowly quenched with H O (25 mL)
and 0.1 N NaOH (35 mL). The mixture was transferred
to a separatory funnel, extracted twice with ethyl ether
2
(
Na SO . The drying agent was filtered off and the sol-
75 mL), and the combined organic layers dried over
yield and were cyclized directly without further puri-
1
fication. p-Chlorobenzyl derivative: H NMR (CDCl ) d
2
4
3
vent removed in vacuo. Column chromatography (Rf
.70, 4.0:0.5:0.5 CHCl /CH OH/N(CH CH ) ) afforded
7.3 (d, 2H, aromatic, 7.15 (d, 2H, aromatic), 5.75 (d,
1H, C3-H), 5.3 (broad s, 1H, NH), 3.5 (s, 2H, CH CO),
3.3 (m, 2H, C10-H ), 2.5–2.2 (m, 2H, C1-H, C4-H), 2.15
2
0
3
3
3
2 3
2
1
2
5 as a yellow oil (1.85 g, 79% yield). H NMR (CDCl )
3
d 5.9 (d, 1H, C3-H), 2.8 (t, 2H, C10-H ), 2.5 (m, 1H,
(t, 2H, C9-H ), 1.6–0.8 (m, 8H, C5-H , C6-H , C7-H ,
2
2
2
2
2
1
C4-H), 2.3 (s, 1H, C1-H), 2.2 (t, 2H, C9-H ), 1.8–1.4 (m,
8
C8-H ). p-Methylbenzyl derivative: H NMR (CDCl ) d
2
2
3
H, C5-H , C6-H , C7-H , C8-H ). As the hydro-
2
6.90–7.20 (m, 4H, aromatic), 5.70–5.65 (d, 1H, C3-H),
5.45–5.35 (broad s, 1H, NH), 3.55 (s, 2H, CH CO),
2
2
2
+
3
bromide salt (CDCl ) d 8.0 (broad s, 3H, NH ), 6.1 (d,
3
2
1
2
H, C3-H), 3.2 (m, 2H, C10-H ), 2.61 (t, 2H, C9-H ),
.58 (d, 1H, C4-H), 2.4 (s, 1H, C1-H), 1.6–1.2 (m,
3.35–3.25 (q, 2H, C10-H ), 2.30 (s, 3H, CH ), 2.20–2.10
2
(m, 4H, C1-H, C4-H, C9-H ), 1.50–0.80 (m, 8H, C5-H ,
2 2
2
2
3
1
3
remaining aliphatic protons); C (CDCl ): d 140.3 (C2),
1
2
C6-H , C7-H , C8-H ). 3,4,5-Trimethoxybenzyl deriva-
2 2 2
3
1
30.4 (C3), 38.3 (C10), 33.7 (C1), 32.5 (C9), 30.0 (C4),
6.1, 25.8 (C5, C6, C7, C8).
tive: H NMR (CDCl ): d 6.40 (s, 2H, aromatic), 5.75–
3
5.65 (d, 1H, C3-H), 5.50 (broad s, 1H, NH), 3.80 (two s,
9H, OCH ), 3.50 (s, 2H, CH CO), 3.40–3.20 (q, 2H,
C10-H ), 2.40–2.00 (m, 4H, C1-H, C4-H, C9-H ), 1.50–
3 2
Substituted phenylacetyl chlorides (26). The acid chlor-
ides that were condensed with amine 25 were synthe-
sized from commercially available acids in the following
manner. Under nitrogen, 18.05 mmol substituted phe-
nylacetic acid and 27.08 mmol oxalyl chloride were
added to 30 mL dry CH Cl containing 1 drop of DMF.
The reaction was stirred for 18 h at room temperature
prior to evaporating solvent and excess oxalyl chloride.
Dry diethyl ether (30 mL) was added to the isolated
product until precipitation occurred. The precipitate
was filtered, and the filtrate concentrated. Liquid pro-
ducts were purified by vacuum distillation (2 mm Hg).
Purified yields approximated 90%. In the case of 3,4,5-
trimethoxyphenylacetyl chloride, a solid was isolated
when the crude acid chloride was allowed to stand in the
cold. This solid acid chloride was purified by recrys-
2
2
0.95 (m, 8H, C5-H , C6-H , C7-H , C8-H ).
2
2
2
2
5,8-Ethano-1-(p-chlorobenzyl)-1,2,3,4,5,6,7,8-octahydro-
isoquinoline (4) and 5,8-ethano-1-(p-chlorobenzyl)-5,6,
7,8-tetrahydroisoquinoline (14). A single necked flask
was purged with nitrogen, charged with 1.60 g
(5.31 mmol) of the p-chlorobenzyl amide (27) and
2
2
ꢀ
135 mL dry toluene, and heated to 130 C in an oil bath.
In small portions over 90 min, 15.07 g (106.13 mmol)
P O was added. The reaction mixture was refluxed for
2
5
2–3 h, cooled to room temperature, and treated with
3.7 mL concentrated HCl and ice. After brief stirring,
the toluene was removed in vacuo and the aqueous layer
basified to pH 7–8 with concentrated ammonium
hydroxide. The aqueous solution was extracted three
tallization from CCl and hexane, and isolated in 60–
4
times with 120 mL portions of CHCl , and the com-
3
1
7
0% purified yield. p-Chlorophenylacetyl chloride: H
bined organic extracts dried over anhydrous Na SO .
2
4
NMR (CDCl ) d 7.5 (d, 2H, aromatic), 7.2 (d, 2H, aro-
The 1.70 g (5.95 mmol) of crude red liquid imine 28
2
3
4
3
matic), 4.18 (s, 2H, CH ). p-Methylphenylacetyl chlor-
(R =R =H, R =Cl) which resulted upon solvent
2

2788
S. L. Saha et al. / Bioorg. Med. Chem. 10 (2002) 2779–2793
evaporation was dissolved in 20–30 mL MeOH. To this
solution was added 483 mg (8.93 mmol) KBH in small
remaining protons). This corresponds well with the
spectrum obtained on 1 free base synthesized by the
4
4
3 13
portions over 15–20 min, and the reaction was left to
stand overnight. The MeOH was then removed in
vacuo, and 30 mL H O and 100 mL ethyl acetate were
original method.
C NMR (CDCl ) d 158.02 (C15),
3
137.34 (C8a), 135.28 (C4a), 131.70 (C12), 130.03 (2C,
C13, C17), 113.90 (2C, C14, C16), 56.24 (C1), 55.20
(OCH ), 40.30 (C3), 38.40 (C11), 34.55 (C5), 31.12 (C8),
2
added. The organic layer was separated and the aqu-
eous layer extracted twice with 100 mL portions of ethyl
acetate. The combined organic extracts were washed
with 50 mL brine and dried over anhydrous Na SO .
3
27.62 (C4), 26.96, 26.80, 26.07, 25.94 (C6, C7, C9, C10).
13: mp hydrochloride 146–148 C. H NMR (CDCl ) d
ꢀ
8.37 (d, 1H, C3-H, J=5 Hz), 7.11 (d, 2H, C14-H, C16-
1
3
2
4
Concentration of the solvent provided 1.33 g crude
liquid product which was purified chromatographically
on a silica gel column. Compounds 4 and 14 were suc-
cessfully separated by eluting sequentially with petro-
leum ether, ethyl ether, and a 2–5% solution of 2%
H), 6.98 (d, 1H, C4-H, J=5 Hz), 6.79 (dd, 2H, C13-H,
C17-H, J=6 Hz), 4.17 (s, 2H, C11-H ), 3.75 (s, 3H,
2
OCH ), 3.24 (broad s, 1H, C8-H), 2.95 (broad s, 1H,
3
1
3
C5-H), 1.85–1.1 (m, 8H, remaining protons); C NMR
(CDCl ) d 157.65 (C15), 153.68, 153.45 (C1, C4a),
3
NH OH/MeOH in diethyl ether. A 40% yield (590 mg)
4
146.31 (C3), 137.32 (C8a), 132.34 (C12), 129.29 (C13,
C17), 117.55 (C4), 113.63 (2C, C14, C16), 55.0 (OCH3),
40.10 (C11), 33.59 (C5), 29.08 (C8), 25.20 (4C, C6, C7,
of 4 and a 20% yield (310 mg) of 14 were realized. Both
products were converted to the hydrochloride salt form
in the standard manner and recrystallized from MeOH/
.
C9, C10). Anal. calcd for C H NOCl H O: C, 68.37;
22
1
9
2
ꢀ
1
diethyl ether. 4: mp hydrochloride: 192—193 C.
H
H, 7.20; N, 4.20; Cl, 10.64; found: C, 68.62; H, 7.32; N,
4.13; Cl, 10.42.
NMR (CDCl ) d 7.29–7.14 (m, 4H, aromatic protons),
3
3
2
.55 (dd 1H, C1-H), 3.1–2.9 (m, 2H, C3-H, C11-H),
.85–2.75 (m, 1H, C3-H), 2.6–2.35 (m, 2H, C5-H, C11-
5,8-Ethano-1-(p-methylbenzyl)-1,2,3,4,5,6,7,8-octahydro-
isoquinoline (7) and 5,8-ethano-1-(p-methylbenzyl)-
5,6,7,8-tetrahydroisoquinoline (15). Utilizing a similar
procedure to that outlined for the synthesis of 4 and 13,
1.35 g (4.75 mmol) of the p-methylbenzyl amide and
13.49 g (95 mmol) P O were reacted to form the inter-
H), 2.25 (broad s, 1H, C8-H), 2.1 (t, 2H, C4-H ), 1.8–
2
1
tons); C NMR (CDCl ) d 138.48 (C12), 137.33 (C8a),
.6 (broad s, 1H, NH), 1.8–1.2 (m, 8H, remaining pro-
1
3
3
1
1
35.68 (C4a), 131.93 (C15), 130.45 (2C, C13, C17),
28.57 (2C, C14, C16), 56.17 (C1), 40.56 (C3), 38.94
2
2
5
4
3
(
2
C11), 34.64 (C5), 31.17 (C8), 27.75 (C4), 27.01, 26.84,
6.07, 25.96 (C6, C7, C9, C10). Anal. calcd for
C H NCl : C, 66.67; H, 7.10; N, 4.32; Cl, 21.91;
mediate imine 28 (R =R =H, R =CH ). Reduction of
3
the crude imine with KBH , followed by chromato-
4
graphic purification on a silica gel column with petro-
leum ether followed by (1) diethyl ether, (2) a 2–5%
solution of 2% NH OH/MeOH in diethyl ether, and (3)
1
8
23
2
found: C, 66.53; H, 7.03; N, 4.21; Cl, 22.23 (by differ-
ꢀ
1
ence). 14: mp hydrochloride 175–175.5 C. H NMR
CDCl ) d 8.4 (d, 1H, C1-H), 7.4–7.0 (m, 4H, C13-H,
C14-H, C16-H, C17-H), 6.9 (d, 1H, C4-H), 4.2 (s, 2H,
4
(
a 4–5% solution of 5% NH OH/MeOH in diethyl ether
4
3
as eluant provided a 31% yield (390 mg) of 7, along with
a 30% yield (380 mg) of 15. Both products were con-
verted to hydrochloride salts and purified by recrys-
tallization from MeOH/diethyl ether. 7: mp
C11-H ), 3.2 (broad s, 1H, C5a-H), 2.9 (broad s, 1H,
2
C8a-H), 2.0–1.1 (m, 8H, remaining protons); C NMR
1
3
(
(
CDCl ) d 153.76, 152.90 (C1, C4a), 146.60 (C3), 138.81
C12), 137.54 (C8a), 131.68 (C15), 129.85 (2C, C13,
3
ꢀ
1
hydrochloride 179–180 C. H NMR (CDCl ) assign-
3
C17), 128.44 (2C, C14, C16), 117.93 (C4), 40.40 (C11),
3.69 (C5), 29.25 (C8), 25.30, 25.26 (4C, C6, C7, C9,
C10). Anal. calcd For C H NCl : C, 67.50; H, 5.94;
ments verified by COZY: d 7.15 (s, 4H, aromatic pro-
tons), 3.6 (dd, 1H, C1-H), 3.15–3.0 (m, 2H, C3-H, C11-
H), 2.87 (s, 1H, NH), 2.86–2.78 (m, 1H, C3-H), 2.66–
2.58 (m, 1H, C11-H), 2.47 (broad s, 1H, C5-H), 2.37 (s,
3
1
8
19
2
N, 4.38; found: C, 67.50; H, 5.99; N, 4.30.
3H, Ar–CH ), 2.3 (broad s, 1H, C8-H), 2.24–2.18 (t,
2H, C4-H ), 1.62–1.25 (m, 8H, remaining protons);
3
1
3
5
,8-Ethano-1-(p-methoxybenzyl)-1,2,3,4,5,6,7,8-octahydro-
isoquinoline (1) and 5,8-ethano-1-(p-methoxybenzyl)-
,6,7,8-tetrahydroisoquinoline (13). Utilizing a similar
procedure to that outlined for the synthesis of 14, 3.50 g
C
2
NMR (CDCl ) d 137.37 (C15), 135.17 (C12), 136.57
3
5
(C8a), 135.47 (C4a), 129.09, 128.91 (4C, C13, C14, C16,
C17), 56.11 (C1), 40.23 (C3), 38.77 (C11), 34.48 (C5),
(
11.70 mmol) of the p-methoxybenzyl amide and 33.22 g
31.08 (C8), 27.59 (C4), 26.89, 26.75, 26.02, 25.90 (C6,
À1
(
mediate imine 28 (R =R =H, R =OCH ). Reduction
of the crude imine with KBH , followed by chromato-
4
graphic purification on a silica gel column with petro-
leum ether followed by a 5% solution of 5% NH OH/
MeOH in diethyl ether, provided a 29% yield (960 mg)
of 13, along with a 30% yield (1 g) of 1. Both products
were converted to hydrochloride or methanesulfonate
salts and purified by recrystallization. 1: mp methane-
233.97 mmol) P O were reacted to form the inter-
2
C7, C9, C10), 20.89 (Ar-CH ). IR (Thin film) cm
3
:
2
5
4
3
3100 (NH stretch), 2932 (CH stretch), 1108 (CH in
plane bend), 767.3 (CH out of plane bend). Anal. calcd
for C H NCl: C, 75.12; H, 8.57; N, 4.61; Cl, 11.70;
3
1
9
26
found: C, 75.08; H, 8.76; N, 4.45; Cl, 11.22; Karl
Fischer water=1.52%. 15: mp hydrochloride 198–
4
ꢀ
J=5 Hz), 7.05 (m, 4H, C13-H, C14-H, C16-H, C17-H),
1
199 C. H NMR (CDCl ) d 8.35 (d, 1H, C3-H,
3
6.85 (d, 1H, C4-H, J=5 Hz), 4.2 (s, 2H, C11-H ), 3.25
2
ꢀ
1
sulfonate 132–132.5 C. H NMR (CDCl ) d 7.15 (d,
(broad s, 1H, C5-H), 2.9 (broad s, 1H, C8-H), 2.2 (s,
1
3
3
2
(
H, Ar–H, J=8 Hz), 6.85 (d, 2H, Ar–H, J=8 Hz), 3.79
s, 3H, OCH ), 3.51–3.47 (m, 1H, C1-H), 3.1–2.8 (m,
3H, Ar–CH ), 1.75–1.0 (m, 8H, remaining protons);
3
C
NMR (CDCl ) d 153.54, 153.39 (C1, C4a), 146.29 (C3).
3
3
2
(
1
H, C3-H, C11-H), 2.85–2.65 (m, 2H, C3-H, NH), 2.52
m, 1H, C11-H), 2.40 (broad s, 1H, C5-H), 2.3 (broad s,
H, C8-H), 2.15 (t, 2H, C4-H ), 1.7–1.0 (m, 8H,
137.32, 137.10 (C15, C8a), 135.08 (C12), 128.86, 128.23
(C13, C14, C16, C17), 117.52 (C4), 40.54 (C11), 33.56
(C5), 29.06 (C8), 25.17 (4C, C6, C7, C9, C10), 20.81
2

S. L. Saha et al. / Bioorg. Med. Chem. 10 (2002) 2779–2793
2789
(
7
4
Ar–CH ). Anal. calcd for C H NCl: C, 76.13; H,
3
.35; N, 4.67; Cl, 11.85; found: C, 76.25; H, 7.43; N,
.57; Cl, 11.59; Karl Fischer water=0.62%.
remaining protons); 13C NMR (CDCl ) d 139.66 (C12),
136.56 (C8a), 134.52 (C4a), 131.34 (C15), 130.32 (2C,
19 22
3
C13, C17), 128.11 (2C, C14, C16), 63.75 (C1), 46.74
(C3), 42.33 (N–CH ), 37.68 (C11), 34.43 (C5), 32.27
3
5
,8-Ethano-1-(3,4,5-trimethoxybenzyl)-1,2,3,4,5,6,7,8-oc-
(C8), 26.80, 26.34 (2C), 25.94, (C6, C7, C9, C10), 23.59
.
(C4). Anal. calcd for C H Cl NO 0.5H O: C, 65.71;
19 27 2 2
tahydroisoquinoline (10) and 5,8-ethano-1-(3,4,5-trime-
thoxybenzyl)-5,6,7,8-tetrahydroisoquinoline (16).
Utilizing a similar procedure to that outlined for the
synthesis of 4 and 13, 3.15 g (8.77 mmol) of the 3,4,5-
trimethoxybenzyl amide and 24.90 g (175.38 mmol)
P O were reacted to form the intermediate imine 28
H, 7.78; N, 4.03; Cl, 20.46; found: C, 65.31; H, 8.05; N,
4.03; Cl, 20.44.
5,8-Ethano-2-methyl-1-(p-methylbenzyl)-1,2,3,4,5,6,7,8-
octahydroisoquinoline (8). Utilizing a similar procedure
to that outlined for the synthesis of 5, 190 mg
2
5
2
3
4
(R =R =R =OCH ). Reduction of the crude imine
3
with KBH , followed by chromatographic purification
(0.71 mmol) of 7, 130.69 mg (2.84 mmol) 88% HCO H,
2
4
on a silica gel column with diethyl ether followed by an
8
and 63.9 mg (2.31 mmol) 37% H CO were reacted to
2
–10% solution of 5% NH OH/MeOH in diethyl ether
4
form 170 mg crude 8 which was purified as the hydro-
chloride salt by recrystallization from MeOH/diethyl
ether. A 25% yield (50 mg) of pure 8 hydrochloride was
as eluant provided a 19% yield (580 mg) of 10, along
with a 17% yield (500 mg) of 16. Both products were
converted to methanesulfonate salts and purified by
ꢀ
1
obtained. Mp hydrochloride 189–190 C. H NMR
(CDCl ) d 7.0–7.2 (m, 4H, aromatic protons), 3.75 (m,
ꢀ
recrystallization. 10: mp methanesulfonate 177–178 C
1
3
H NMR (CDCl ) d 6.45 (s, 2H, aromatic protons), 3.85
1H, C1-H), 3.55–3.4 (m, 2H, C3-H, C11-H), 3.15 (t, 1H,
C3-H), 2.9–2.75 (dd, 1H, C11-H), 2.3 (s, 3H; NCH3),
3
(
s, 3H, OCH ), 3.83 (s, 3H, OCH ), 3.79 (s, 3H, OCH ),
3 3 3
3
1
.55 (m, 1H, C1-H), 3.0 (m, 2H, C11-H, C3-H), 2.75 (m,
H, C3-H), 2.6 (broad s, 1H, NH), 2.55–2.4 (m, 2H,
2.25 (s, 3H, Ar–CH ), 2.2 (broad s, 2H, C5-H, C4-H),
3
2.05 (broad s, 2H, C8-H, C4-H), 1.5–1.1 (m, 8H,
1
3
C11-H, C8-H), 2.35–2.0 (m, 3H, C5-H, C4-H ), 1.65–
2
remaining protons); C NMR (CDCl ) d 137.81 (C15),
3
1
3
1
1
1
6
.2 (m, 8H, remaining protons); C NMR (CDCl ) d
136.86 (C8a), 135.10 (C12), 133.97 (C4a), 128.82, 128.78
(4C, C13, C14, C16, C17), 63.90 (C1), 46.71 (C3), 42.24
(N–CH ), 38.01 (C11), 34.37 (C5), 32.26 (C8); 26.76,
3
26.31, 25.90 (2C) (C6, C7, C9, C10), 23.77 (C4), 21.00
.
(ArCH ). Anal. calcd for C H NCl 0.5H O: C, 73.50;
3 20 30 2
3
53.13 (2C, C14, C16), 136.85 (C8a), 136.0 (C12),
35.62 (C4a), 135.20 (C15), 105.83 (2C, C13, C17),
0.77 (OCH ), 56.02 (3C, C1,OCH ), 40.76 (C3), 39.82
3
3
(
2
C11), 34.53 (C5), 30.90 (C8), 27.43 (C4), 26.95, 26.78,
5.95, 25.87 (C6, C7, C9, C10). Anal. calcd for
C H NSO : C, 60.14; H, 7.52; N, 3.19; found: C,
H, 9.19; N, 4.29; Cl, 10.87; found: C, 73.38; H, 9.24; N,
4.20; Cl, 10.84.
2
2
33
6
6
1
0.04; H, 7.77; N, 3.11. 16: mp methanesulfonate 175–
ꢀ
1
76 C. H NMR (CDCl ) d 8.35 (d, 1H, C3-H,
5,8-Ethano-2-methyl-1-(3,4,5-trimethoxybenzyl)-1,2,3,4,-
5,6,7,8-octahydroisoquinoline (11). Utilizing a similar
procedure to that outlined for the synthesis of 5, 62 mg
3
J=5 Hz), 7.0 (d, 1H, C4-H, J=5 Hz), 6.4 (s, 2H, C13-
H, C17-H), 4.2 (s, 2H, C11-H ), 3.7 (s, 9H, OCH ), 3.3
2
3
(
(
broad s, 1H, C5-H), 2.95 (broad s, 1H, C8-H), 1.8–1.25
1
(0.18 mmol) 10, 25.9 mg (0.86 mmol) 37% HCO H and
2
3
m, 8H, remaining protons); C NMR (CDCl ) d
18 mg (0.29 mmol NaCNBH in 0.54 mL CH CN pro-
3
3
3
1
1
1
54.23 (C1); 153.40 (3C, C4a, C14, C16), 146.55 (C3),
37.88 (C8a), 136.85 (C12), 136.07 (C15), 118.07 (C4),
06.24 (2C, C13, C17), 60.97 (OCH ), 56.37 (2C,
3
vided 50 mg (78% yield) of crude 11, which was con-
verted to the hydrochloride salt and recrystallized from
MeOH/diethyl ether in the standard manner. Mp
ꢀ
1
OCH ), 41.30 (C11), 34.08 (C5), 29.64 (C8), 25.64, 25.55
3
hydrochloride 145–146 C. H NMR (CDCl ) d 6.5 (s,
3
(
6
3
4C, C6, C7, C9, C10). Anal. calcd for C H NSO : C,
2
0.69; H, 6.67; N, 3.22; found: C, 60.70; H, 6.94; N,
.18.
2H, aromatic protons), 3.9 (s, 6H, OCH ), 3.85 (s, 3H,
OCH ), 3.2–3.1 (m, 1H, C1-H), 3.05–2.85 (m, 2H, C3-
2
29
6
3
3
H, C11-H), 2.85–2.7 (t, 1H, C3-H), 2.7–2.55 (m, 1H,
C11-H), 2.35 (s, 3H, NCH ), 2.2 (broad s, 1H, C5-H),
3
5,8-Ethano-1-(p-chlorobenzyl)-2-methyl-1,2,3,4,5,6,7,8-
octahydroisoquinoline (5).
2.1 (broad s, 1H, C8-H), 1.85 (m, 1H, C4-H), 1.6–1.1
3
1
A
mixture of 200 mg
0.70 mmol) 4, 116.89 mg (2.54 mmol) 98% HCO H,
(m, 8H, remaining protons); C NMR (CDCl ) d
3
(
and 63 mg (2.10 mmol) 37% H CO was stirred for
152.76 (2C C14, C16), 136.80, 136.70 (C8a, C15), 135.96
(C12), 134.30 (C4a), 105.76, 105.56 (C13, C17), 63.44
(C1), 60.77 (OCH ), 55.94 (2C, OCH ), 46.88 (C3),
2
2
1
0 min at room temperature and then refluxed 24 h in an
ꢀ
3
3
oil bath at 110 C. The volatile materials were removed
in vacuo and the pH adjusted to 8 with a saturated
NaHCO solution. The emulsion was extracted three
42.32 (NCH ), 38.62 (C11), 34.38 (C5), 32.17 (C8),
3
26.74, 26.33, 26.29, 25.91 (C6, C7, C9, C10), 23.66 (C4).
.
Anal. calcd for C H NO Cl 1.33H O: C, 63.23; H,
22 32 3 2
3
times with 50 mL portions of CH Cl and dried over
2
8.30; N, 3.35; found: C, 63.43; H, 8.30; N, 3.10.
2
anhydrous Na SO . The crude product was purified by
2
4
recrystallization of the hydrochloride salt from MeOH/
diethyl ether to give 80 mg (40% yield) of crystalline
5,8-Ethano-1,2-di-(p-methoxybenzyl)-1,2,3,4,5,6,7,8-octa-
hydroisoquinoline (3). CHCl (0.73 mL) was added to
3
ꢀ
1
product. Mp hydrochloride 205–206 C. H NMR
CDCl ) d 7.4–7.0 (s, 4H, aromatic protons), 3.4–3.5 (m,
100 mg (0.35 mmol) 1, and to this solution was added
75 mg (0.70 mmol) Et N and 54.81 mg (0.35 mmol)
(
3
3
1
(
H, C1-H), 3.1–2.95 (m, 2H, C3-H, C11-H), 2.85–2.75
m, 1H, C3-H), 2.5–2.65 (m, 1H, C11-H), 2.30 (s, 3H,
NCH ), 2.2 (broad s, 2H, C5-H, C4-H), 2.1 (broad s,
p-methoxybenzyl chloride. The reaction was stirred at
room temperature for 20 h prior to dilution with 50 mL
CH Cl and 20 mL H O. The layers were separated and
3
2
2
2
1
H, C8-H), 2.1–1.9 (m, 1H, C4-H), 1.6–1.1 (m, 8H,
the aqueous layer extracted twice with 50 mL portions

2790
S. L. Saha et al. / Bioorg. Med. Chem. 10 (2002) 2779–2793
of CH Cl . The combined organic extracts were washed
2
with 20 mL each of H O and brine, and dried over
2
C24-H, J=8 Hz), 6.6 (d, 2H, C21-H, C23-H, J=8 Hz),
3.8 (s, 3H, OCH ), 3.5 (m, 3H, C1-H, C18-H ), 3.2 (m,
2
3
2
anhydrous Na SO . Purification was accomplished
2
1H, C3-H), 3.0 (t, 1H, C11-H), 2.7 (m, 3H, C3-H, C11-
H), 2.35 (s, 3H, Ar-CH ), 2.3 (broad s, 1H, C5-H), 2.15
4
through silica gel chromatography utilizing a 2% solu-
tion of 2% NH OH/MeOH in diethyl ether as eluant. A
3
(broad s, 1H, C8-H), 1.85–1.65 (m. 2H, C4-H ), 1.6–
2
4
1
3
7
8% yield (110 mg) of purified 3 was obtained. The
1.15 (m, 8H, remaining protons); C NMR (CDCl ) d
3
hydrochloride salt was formed in the standard manner
and recrystallized from MeOH/diethyl ether. Mp
158.47 (C22), 138.17 (C15), 137.48 (C8a), 135.06 (C12),
134.41 (C4a), 132.25 (C19), 130.02 (2C, C20, C24),
129.51 (2C, C13, C17), 128.76 (2C, C14, C16), 113.5
(2C, C21, C23), 61.47 (C1), 56.81 (C18), 55.44 (OCH3),
43.49 (C3), 39.27 (C11), 34.77 (C5), 32.82 (C8), 27.09,
26.74 (2C), 26.51 (4C, C6, C7, C9, C10), 22.26 (C4),
21.33 (ArCH ). Anal. calcd for C H NOCl: C, 76.50;
ꢀ
1
hydrochloride 90–91 C. H NMR (CDCl ) d 7.04 (d,
3
2
H, C20-H, C24-H, J=9 Hz); 6.92 (d, 2H, C21-H, C23-
H, J=9 Hz), 6.77 (d, 2H, C13-H, C17-H, J=9 Hz), 6.68
d, 2H, C14-H, C16-H, J=8 Hz), 3.80 (s, 3H, OCH3),
.76 (s, 3H, OCH ), 3.54–3.42 (dd, 3H, C18-H , C1-H,
(
3
3
2
3
27 34
J=13 Hz), 3.22–3.21 (m, 1H, C3-H), 2.99 (t, 1H, C11-
H), 2.75–2.69 (m, 2H, C3-H, C11-H), 2.25 (broad s, 1H,
C5-H), 2.15 (broad s, 1H, C5-H), 1.85–1.75 (m, 2H, C4-
H, 7.08; N, 3.30; Cl, 8.38; found: C, 76.10; H, 8.46; N,
3.01; Cl, 8.30; Karl Fischer water <1.19%.
1
3
H ), 1.65–1.1 (m, 8H, remaining protons); C NMR
2
5,8-Ethano-2-(p-methoxybenzyl)-1-(3,4,5-trimethoxyben-
zyl)-1,2,3,4,5,6,7,8-octahydroisoquinoline (12). Utilizing
a similar procedure to that outlined for the synthesis of
(
CDCl ) d 158.14 (C22), 157.61 (C15), 137.15 (C8a),
3
1
(
34.14 (C4a), 133.19 (C12), 132.05 (C19), 130.19, 129.67
4C, C13, C17, C20, C24), 113.19 (4C, C14, C16, C21,
C23), 61.13 (C1), 56.52 (C18), 55.23, 55.15 (OCH3),
3.25 (C3), 38.45 (C11), 34.46 (C5), 32.53 (C8), 26.78
C11), 26.40 (2C), 26.21 (4C, C6, C7, C9, C10), 21.94
3, 82 mg (0.24 mmol) 10, 59.21 mg (0.48 mmol) Et N,
3
and 45.22 (0.24 mmol) p-methoxybenzyl chloride in
0.56 mL CHCl3 gave crude 12. The compound was
purified by silica gel chromatography utilizing diethyl
ether/petroleum ether (1:1), diethyl ethyl, and 5%
4
(
(
.
C4). Anal. calcd for C H ClNO 0.75H O: C, 71.52;
2
2
7
34
H, 7.95; N, 3.09; Cl, 7.84; found: C, 71.52; H, 7.94; N,
2
MeOH in CHCl as eluant. A 65% yield (72 mg) of
3
.96; Cl, 7.86.
purified 12 was obtained. The compound was converted
to the methanesulfonate salt in the standard manner
and recrystallized from MeOH/diethyl ether. Mp
5
1
,8-Ethano -1-(p-chlorobenzyl)-2-(p-methoxybenzyl)-
,2,3,4,5,6,7,8-octahydroisoquinoline (6). Utilizing
ꢀ
1
a
methanesulfonate 150–151 C. H NMR (CDCl ) d 6.95
3
similar procedure to that outlined for the synthesis of 3,
1
(d, 2H, C20-H, C24-H, J=8 Hz), 6.70 (d, 2H, C21-H,
C23-H, J=8 Hz), 6.3 (s, 2H, C13-H, C17-H), 3.6–3.35
50 mg (0.53 mmol) 4 in 1 mL CHCl was reacted with
3
1
07.66 mg (1.06 mmol) Et N and 82.22 mg (0.53 mmol)
3
(m, 3H, C1-H, C18-H ), 3.3–3.1 (m, 1H, C3-H), 3.05–
2
p-methoxybenzyl chloride to produce 210 mg (98%
yield) of 6 which was purified as the hydrochloride salt
by recrystallization from MeOH/diethyl ether. Mp 152–
2.95 (m, 1H, C11-H), 2.85–2.6 (m, 2H, C3-H, C11-H),
2.2 (broad s, 1H, C5-H), 2.1 (broad s, 1H, C8-H), 1.9–
1.4 (m, 2H, C4-H ), 1.65–1.15 (m, 8H, remaining pro-
2
ꢀ
1
13
1
H, J=8 Hz); 7.0 (d, 2H, C13-H, C17-H, J=8 Hz), 6.8
53 C. H NMR (CDCl ) d 7.15 (d, 2H, C14-H, C16-
tons); C NMR (CDCl ) d 158.17 (C22), 152.50 (2C,
3
3
C14, C16), 136.88, 136.76 (C15, C8a), 136.0 (C12),
134.32 (C4a), 131.88 (C19), 129.76 (2C, C20, C24),
113.08 (2C, C21, C23), 106.10 (2C, C13, C17), 60.80,
(
C23-H, J=8 Hz), 3.8 (s, 3H, OCH ), 3.65–3.35 (m, 3H,
d, 2H, C20-H, C24-H, J=8 Hz), 6.6 (d, 2H, C21-H,
3
C1-H, C18-H ), 3.3–3.1 (m, 1H, C3-H), 3.0–2.85 (t, 1H,
2
60.58 (C1, OCH ), 56.47 (C18), 55.83, 55.08 (2C,
3
C11-H), 2.8–2.5 (m, 2H, C3-H, C11-H), 2.25 (broad s,
OCH ), 43.53 (C3), 39.61 (C11), 34.42 (C5), 32.44 (C8),
3
1
C4-H ), 1.4–1.1 (m, 8H, remaining protons); C NMR
H, C5-H), 2.15 (broad s, 1H, C8-H), 1.9–1.7 (m, 2H,
3
26.78, 26.44, 26.38, 26.17 (4C, C6, C7, C9, C10), 21.95
.
(C4). Anal. calcd for C H NSO 0.75H O: C, 71.52;
1
2
30 40
7
2
(
1
CDCl ) d 158.23 (C22), 139.52 (C12), 136.72 (C8a),
34.60 (C4a), 131.79 (C19), 131.14 (C15), 130.73 (2C,
H, 7.95; N, 3.09; S, 7.84; found: C, 71.52; H, 7.94; N,
2.96; S, 7.86.
3
C13, C17), 129.71 (2C, C20, C24), 127.74 (2C, C14,
C16), 113.20 (2C, C21, C23), 60.66 (C1), 56.49 (C18),
5,8-Ethano-2-formyl-1-(p-chlorobenzyl)-1,2,3,4,5,6,7,8-
octahydroisoquinoline (17). To a solution of 60 mg
(0.21 mmol) 4 in 2.6 mL benzene was added 0.60 mL
(544 mg, 7.35 mmol) ethyl formate, and the reaction
5
5.20 (OCH ), 43.45 (C3), 38.56 (C11), 34.50 (C5), 32.55
3
(
C8), 26.80, 26.48, 26.42, 26.24 (4C, C6, C7, C9, C10),
.
1.80 (C4). Anal. calcd for C H NOCl 0.5H O: C,
26 31 2 2
2
6
7
ꢀ
8.86; H, 7.28; N, 3.09; Cl, 15.67; found: C, 69.11; H,
.34; N, 3.11, Cl, 15.86.
mixture heated to 50 C for 24 h. The reaction was
cooled and the excess reactant and solvent removed in
vacuo. Diethyl ether (50 mL) was added to the residue,
the solution washed sequentially with 20 mL H O,
5
3
,8-Ethano-2-(p-methoxybenzyl)-1-(p-methybenzyl)-1,2,-
,4,5,6,7,8-octahydroisoquinoline (9). Utilizing a similar
2
20 mL 10% HCl, and 20 mL H O, and dried over
2
procedure to that outlined for the synthesis of 3, 60 mg
0.22 mmol) 7, 45.21 mg (0.45 mmol) Et N, and
anhydrous Na SO . When concentrated, 40 mg (61%
4
2
(
yield) of crude 17 was obtained. The compound was
purified via silica gel chromatography utilizing petro-
leum ether/diethyl ether (1:1) as eluant. A 45% purified
3
3
7.05 mg (0.22 mmol) p-methoxybenzyl chloride in
0
yield) of 9, which was purified as the hydrochloride salt
.44 mL CHCl were reacted to produce 60 mg (70%
3
ꢀ
1
yield (30 mg) was obtained. Mp 120–121 C. H NMR
by recrystallization from MeOH/diethyl ether. Mp
hydrochloride 164–165 C. H NMR (CDCl ) d 7.0 (s,
4
(CDCl ) d (two isomers) 7.95, 7.5 (s, 1H, CHO), 7.3–6.8
(m, 8H, aromatic protons), 4.95 (t, 1H, C1-H), 4.5 (dd,
1H, C1-H), 3.9 (broad d, 1H, C3-H), 3.5 (dd, 1H, C3-
3
ꢀ
H, C13-H, C14-H, C16-H, C17-H), 6.9 (d, 2H, C20-H,
1
3

S. L. Saha et al. / Bioorg. Med. Chem. 10 (2002) 2779–2793
2791
H), 3.1–2.4 (m, 6H, C11-H , C3-H ), 2.9 (t, 2H, C4-H ),
2
with 0.1 M sodium citrate (9:1=blood/sodium citrate).
Platelet-rich plasma (PRP) was prepared by centrifu-
gation of the blood sample at 200g (Sorvall RT 6000,
DuPont Company) for 15 min at room temperature,
and the supernatant transferred as PRP. An aliquot of
PRP (1 mL) was further centrifuged at 2000g for 2 min
in an Eppendorf microcentrifuge (5415C, Brinkmann
Instruments Inc., Germany), and the supernatant was
utilized as platelet-poor plasma (PPP). Aliquots
(900 mL) of PPP were added to 180 mL phosphate buffer
(10 mM sodium phosphate, 0.9% NaCl, pH 7.4) to
make diluted PPP. Diluted PPP (450 mL) was used to set
the 100% transmittance for experiments.
2
2
2
C8-H , C4-H ), 1.7–1.1 (m, 16H, remaining protons);
.45 (broad s, 2H, C5-H ), 2.35–2.2 (broad s and m, 4H,
2
2
2
1
3
C NMR (CDCl ) selected d (two isomers) 161.28,
3
1
61.22 (CHO), 136.34, 136.24 (C12), 135.22, 135.04
(
C8a), 134.84 (C4a), 132.68 (C15), 130.73, 130.56 (C13,
C17), 128.87, 128.27 (C14, C16), 58.42 (C1), 51.16 (C3),
0.63 (C11), 34.34, 33.80 (C5), 32.14, 31.91 (C8), 27.68
C4). Anal. calcd for C H NOCl: C, 72.26; H, 6.97; N,
4
(
1
9
22
4.43; found: C, 72.39; H, 7.22; N, 4.08. Karl Fischer
water=<3.19%.
5,8-Ethano-1-(p-chlorobenzyl)-2,2-dimethyl-1,2,3,4,5,6,7,
8-octahydroisoquinoline iodide (18). To a solution of
60 mg (0.20 mmol) 5 in 4 mL MeOH was added 0.2 mL
Protocol. Preliminary studies were conducted to deter-
mine whether each compound was an inducer of platelet
aggregation. Antagonist activities were determined by
incubating varying concentrations of each compound
with PRP for 1 min prior to addition of U46619.
Monophasic platelet aggregation by U46619, a TXA2
agonist, was measured in aspirin-treated human plate-
lets. Cuvettes contained a final volume 420 mL of
appropriate amount of phosphate buffer (10 mM),
aspirin (1 mM), PRP (350 mL) and U46619 (0.5–1.5 mM)
with a magnetic teflon stir bar spinning at 1000 rpm at
MeI and 200 mg KHCO . The mixture was stirred at
3
room temperature for 24 h and the excess solvent and
CH I removed in vacuo. After filtration and concen-
3
tration, 67 mg (76% yield) crude 18 was obtained. Pur-
ification was by recrystallization from MeOH/diethyl
ꢀ
aromatic protons, J=8 Hz), 7.29 (dd, 2H, aromatic pro-
1
ether. Mp 198–199 C. H NMR (CDCl ) d 7.50 (d, 2H,
3
tons), 4.66 (t, 1H, C1-H, J=7 Hz), 3.85 (broad m, 2H,
C3-H, C11-H), 3.6–3.45 (m, 4H, NCH , C3-H), 3.35 (s,
3
3
C4-H ), 2.45 (broad s, 1H, C5-H), 1.85 (broad s, 1H,
H, NCH ), 2.9–3.1 (m, 1H, C11-H), 2.7–2.5 (m, 2H,
3
ꢀ
37 C for 4 min.
2
1
3
C8-H), 1.6–1.0 (m, 8H, remaining protons); C NMR
(
CDCl ) d 134.55 (C12), 133.58, 133.29 (C8a, C4a),
Measurements. Aggregometers (Chrono-Log Corp.,
Model 560-VS and Model 660, Havertown, PA, USA)
were interfaced with personal computers to analyze
changes in aggregation. Changes in turbidity of the
samples in the aggregometer were acquired, monitored
and stored using the computer program (AGGRO/
LINK, 1993). As platelets aggregate, the turbidity of
PRP decreases with a corresponding increase in light
transmittance. Aggregation responses for each experi-
ment were measured and expressed as percentage of the
maximal light transmittance (Tmax) in the presence of
inducer.
3
1
C16), 71.68 (C1), 55.71 (C3), 52.23 (NCH ), 50.64
32.37 (C15), 131.11 (2C, C13, C17), 128.94 (2C, C14,
3
(
NCH ), 36.23 (C ), 33.56, 33.30 (C8 C5), 26.59, 25.58,
3 11 ,
2
for C H NClI: C, 53.90; H, 6.06; N, 3.15; found: C,
4.92 (2C) (C6, C7, C9, C10), 23.22 (C4). Anal. calcd
2
0
27
5
3.93; H, 5.94; N, 3.04; Karl Fischer water <1.44%.
1,4-Ethano-1,3,5,8,13,13a-hexahydro-9,10-dimethoxy-6H-
dibenzo[a,g]quinolizine (29).
A mixture of 90 mg
0.26 mmol) 10, 45.46 mg (0.95 mmol) H CO2 and
(
2
1
2
70.86 mg (0.76 mmol) 37% H CO was refluxed for
2
4 h. Volatile materials were removed in vacuo, the pH
adjusted to 8 with saturated NaHCO and the resulting
3
Data analysis. Values of IC50 and pIC50 for each
antagonist on aggregation were calculated using the
GraphPad Prism computer program. Data represent
the meanÆSEM of n=3–6 experiments.
emulsion extraced with CH Cl . After drying over
2
2
6
7
Na SO , the solvent was evaporated to give 30 mg
2
4
(0.0845 mmol) 29. The product was purified via silica gel
column chromatography and sequentially eluted with
the following: petroleum ether, 1:1 petroleum ether/di-
1
Platelet binding
ethyl ether, diethylether and 5% MeOH in CHCl . H
3
NMR (CDCl ) d 6.45 (s, 2H, C12-H ), 4.1–3.9 (2s, 2H,
Isolation of washed platelets. PRP was used in preparing
8
3
2
6
C8-H ), 3.85–3.75 (3s, 9H, OCH ), 3.45–3.3 (2s, 1H,
C14-H), 3.15–3.0 (m, 3H, C13-H, C6-H ), 2.75–2.6 (m,
washed platelets as described previously. PRP and
2
3
PGE (1 mM final) were gently mixed and spun at 1000g
2
1
1
H, C13-H), 2.6 (s, 1H, C5-H), 2.55–2.45 (m, 1H, C4-
H), 2.35 (broad s, 1H, C1-H), 2.1–1.95 (broad d, 1H,
(Clay-Adams, Inc.) for 4 min. The sediment was col-
lected, resuspended in 50 mM Tris–saline (7 g/L Trizma
HCl, 0.67 g/L Trizma base and 6.53 g/L NaCl, pH 7.4)
0
0
C5-H), 1.7–1.2 (m, 8H, C2-H , C2 -H , C3-H , C3 -H );
2
2
2
2
1
3
C NMR (CDCl ) d 151.7 (C9), 149.7 (C11), 139.8 (C10),
containing 5 mM EGTA and 1 mM PG E, and recen-
3
1
1
35.7 (C4a), 14.6 (C14a) 130.3 (C12a), 120.9 (C8a), 107.2
trifuged at 1000g for 4 min. Platelets were resuspended
in Tris–saline in the presence of 1 mM PG E and cen-
,
(C12), 60.8, 60.4 (C9-OCH , C11-OCH ), 58.2 (C10-
OCH ), 55.8 (C14), 53.1 (C8), 51.6 (C6), 34.3, 33.9 (C1,
C4), 29.8 (C5), 27.3, 26.8, 25.9, 25.8 (C2, C2 , C3, C3 ).
3
3
1
trifuged again. Platelet concentration was determined
with a hemocytometer and adjusted so the final suspen-
3
0
0
9
sion contained 1ꢁ10 intact platelets/mL in Tris–saline.
Platelet aggregation
Measurement of competitive equilibrium binding. The
3
Platelet preparation. Human blood (55–110 mL) was
drawn from medial cubital vein of donors, and mixed
specific binding of [ H]SQ29,548 (1 or 5 nM final, specific
activity 46 Ci/mmol, DuPont Company, Wilmington,

2792
S. L. Saha et al. / Bioorg. Med. Chem. 10 (2002) 2779–2793
8
DE, USA) to human platelets (2ꢁ10 platelets/mL final)
was conducted in a final volume of 0.5 mL at room
temperature. Nonspecifically bound ligand was defined
8. Cross, P. E.; Dickinson, R. P. Annu. Rep. Med. Chem.
1987, 22, 95.
9. Umetani, K.; Tamura, K.; Komori, S.; Watanabe, A.; Ish-
ihara, T.; Mochizuki, S.; Li, B.; Ijiri, H. Jpn. Circ. J. 1996, 60,
3
by 50 mM [ H]SQ29,548, and specific binding was
determined as the difference between total binding and
3
1
1
49.
0. Lefer, A. M.; Darius, H. Fed. Proc. 1987, 46, 144.
1. Joachim, K. D.; Klar, U.; Pletsch, A.; Rehwinkel, H.;
nonspecific binding. Specific binding varied between
9
4.5 and 98.1% of total binding. Inhibition of specifi-
Skuballa, W. Prostaglandins 1995, 50, 57.
2. Fukumoto, S.; Terashita, Z.; Ashida, y.; Terao, S.; Shir-
cally bound radioligand by bicyclic octahydroisoquino-
line analogues was determined using various
concentrations of the analogues after a 30 min incu-
bation. Samples were rapidly filtered by vacuum
through Whatman GF/C fiber filters on a Brandel cell
harvester (Model M 12-RI), and washed three times
6 mL each) with ice-cold 50 mM Tris–saline buffer (pH
.4). The filter disks were transferred to plastic vials
containing 10 mL of scintillation cocktail (Ultima
1
aishi, M. Chem. Pharm. Bull. Tokyo 1996, 44, 749.
13. Kawashima, Y.; Sato, M.; Yamamoto, S.; Shimazaki, Y.;
Chiba, Y.; Satake, M.; Iwata, C.; Hatayama, K. Chem.
Pharm. Bull. Tokyo 1995, 43, 1132.
14. Hirata, M.; Hayashi, Y.; Ushibubi, F.; Yakota, Y.;
Kageyama, R.; Nakanishi, S.; Narumiya, S. Nature 1991, 349,
(
7
6
1
17.
5. Shenker, A.; Goldsmith, P.; Unson, C. C.; Spiegel, A. M.
J. Biol. Chem. 1991, 266, 9309.
TM
Gold XR, Packard Instrument Company), and vor-
3
16. Vezza, R.; Habib, A.; FitzGerald, G. A. J. Biol. Chem.
1999, 274, 12774.
17. Djellas, Y.; Manganello, J. M.; Antonakis, K.; LeBreton,
texed. [ H] was measured in a liquid scintillation spec-
trometer (LKB Wallac Liquid Scintillation, Model 1219
Rackbeta).
G . C.J. Biol. Chem. 1999, 274, 14325.
8. Mayeux, P. R.; Morinelli, T. A.; Williams, T. C.; Hazard,
E. S.; Mais, D. E.; Oatis, J. E.; Baron, D. A.; Halushka, P. V.
1
Data analysis. Percent specific binding was calculated as
total bindingÀnon-specific binding)/total bindingꢁ
J. Biol. Chem. 1991, 266, 13752.
19. Dorn, G. W.; Liel, N.; Trask, J. L.; Mais, D. E.; Assey,
M. E.; Halushka, P. V. Circulation 1990, 81, 212.
(
1
half of the maximal responses (IC50 values) were deter-
00%. Individual inhibitory concentrations reducing
2
2
2
0. Chiang, N.; Tai, H. H. Arch. Biochem. Biophys. 1998, 352,
07.
1. Takahara, K.; Murray, R.; Fitzgerald, G. A.; Fitzgerald,
mined using GraphPad Prism. K values were obtained
I
using IC /(1+[ligand]/K ), where K (3.1 nM) was
d
taken from previously published work. Potency ratio
5
0
d
6
9
D. J. J. Biol. Chem. 1990, 265, 6836.
22. Halushka, P. V.; Mais, D. E.; Saussy, D. L., Jr. Fed. Proc.
1987, 46, 149.
(
or K ) bicyclohydroisoquinolines, where values for
I
P.R.) values were calculated as (IC50 or K TMQ)/(IC50
I
TMQ were used as reference standard for these studies.
Linear regression for correlation of aggregation (pIC )
and binding (pK ) was plotted and analyzed. Data
represent the meanÆSEM of n=3–6 experiments.
23. Ezumi, K.; Yamakawa, K.; Narisada, M. J. Med. Chem.
1990, 33, 1117.
5
0
2
1
2
3
2
4. Yamamoto, Y.; Kamiya, K.; Terao, S. J. Med. Chem.
993, 36, 820.
5. Jin, B.; Hopfinger, A. J. J. Chem. Inf. Comput. Sci. 1994,
4, 1014.
6. Albuquerque, M. G.; Hopfinger, A. J.; Barreiro, E. J.;
I
Acknowledgements
deAlencastro, R. B. J. Chem. Inf. Comput. Sci. 1998, 38, 925.
2
2
7. Ogletree, M. L. Fed. Proc. 1987, 46, 133.
8. Misra, R. J.; White, R. E.; Ogletree, M. L. Drugs Future
The authors gratefully acknowledge financial support
for this project from the Nebraska Department of
Health. The Nebraska Center for Mass Spectrometry,
which performed the high resolution analysis of selected
bicyclic molecules, is also acknowledged.
1994, 19, 107.
29. Hall, S. E. Med. Res. Rev. 1991, 11, 503.
30. Iwasawa, Y.; Kiyomoto, A. Jap. J. Pharmacol. 1967, 17, 143.
3
1. Mayo, J. R.; Navran, S. S.; Huzoor-Akbar Miller, D. D.;
Feller, D. R. Biochem. Pharmacol. 1981, 30, 2237.
2. Mukhopadhyay, A.; Navran, S.; Amin, H. M.; Abdel-
3
Aziz, S. A.; Chang, J.; Sober, D. J.; Miller, D. D.; Feller, D. R.
J. Pharmacol. Exp. Ther. 1985, 232, 1.
References and Notes
3
3. Shams, G.; Romstedt, K. J.; Gerhardt, M. A.; Harrold,
1. Vicari, A. M.; Margonato, A.; Macagni, A.; Luoni, R.;
Seveso, M. P.; Vicedomini, G.; Pozza, G. Clin. Cardiol. 1988,
M. W.; Miller, D. D.; Feller, D. R. Eur. J. Pharmacol. 1990,
184, 21.
1
2
1, 538.
. Hirsh, P. D.; Hillis, L. D.; Campbell, W. B.; Firth, B. G.;
34. Fraundorfer, P. F.; Fertel, R. H.; Miller, D. D.; Feller,
D. R. J. Pharmacol. Exp. Ther. 1994, 270, 665.
35. Kumar, R.; Singh, P. Ind. J. Biochem. Biophys. 1998, 35,
390.
36. Sato, M.; Yamaguchi, I.; Kiyomoto, A. Jpn. J. Pharma-
col. 1967, 17, 153.
37. Kiyomoto, A.; Sato, M.; Nagao, T.; Nakajima, H. Eur. J.
Pharmacol. 1969, 5, 303.
38. Dalton, C.; Crowley, H. J.; Czyzewski, L. B. Biochem.
Pharmacol. 1976, 25, 2209.
39. Ahn, C.-H.; Romstedt, K. J.; Wallace, L. J.; Miller, D. D.;
Feller, D. R. Biochem. Pharmacol. 1988, 37, 3023.
40. Fraudorfer, P. F.; Lezama, E. J.; Salazar-Bookaman,
M. M.; Fertel, R. H.; Miller, D. D.; Feller, D. R. Chirality
1994, 6, 76.
Willerson, J. T. N. Engl. J. Med. 1981, 304, 685.
. Nagata, T.; Uehara, Y.; Hara, K.; Igarashi, K.; Hazama,
H.; Kimura, K.; Goto, A.; Omata, M. Respirology 1998, 2,
3
2
4
4
5
1
6
83.
. Armstrong, R. A.; Wilson, N. H. Gen. Pharmacol. 1995, 26,
36.
. Sinead, M.; Miggin, B.; Kinsella, T. Biochim. Biophys. Acta
998, 1425, 543, and references therein.
. Coleman, R. A.; Collington, E. W.; Geisow, H. P.; Hornby,
E. J.; Humphrey, P. P. A.; Kennedy, I.; Levy, G. P.; Lumley, P.;
McCabe, P. J.; Wallis, C. J. Br. J. Pharmacol. 1981, 72, 524 P.
7. Mayo, J. R.; Navran, S. S.; Huzoor-Akbar Miller, D. D.;
Feller, D. R. Biochem. Pharmacol. 1981, 30, 2237.

S. L. Saha et al. / Bioorg. Med. Chem. 10 (2002) 2779–2793
2793
41. Shin, Y.; Romstedt, K. J.; Doyle, K.; Harold, M. W.;
Gerhardt, M. A.; Miller, D. D.; Patil, P. N.; Feller, D. R.
Chirality 1991, 3, 112.
Practical Organic Chemistry; John Wiley & Sons: New York,
1989; p 704.
55. Tsuhi, Y.; Yamada, N.; Tanaka, S. J. Org. Chem. 1993,
58, 16.
56. Nystrom, R. F. J. Am. Chem. Soc. 1955, 77, 2544.
57. Monkovic, I.; Wong, H.; Belleau, B.; Pachter, I. J.; Per-
ron, V. G. Can. J. Chem. 1975, 53, 2515.
58. Wissner, A.; Carroll, M. L.; Johnson, B. D.; Kerwar, S. S.;
Pickett, W. C.; Schaub, R. E.; Torley, L. W.; Trova, M. P.;
Kohler, C. A. J. Med. Chem. 1992, 35, 4779.
59. Bischler, A.; Napieralski, B. Ber. Deutsch. Chem. Ges.
1893, 26, 1903.
60. Kametani, T.; Kigasawa, K.; Hiiragi, M.; Satoh, F.; Sugi,
H.; Uryu, T. J. Heterocycl. Chem. 1972, 9, 1065.
61. Bruneau, P.; Delvare, C.; Edwards, M. P.; McMillan,
R. M. J. Med. Chem. 1991, 34, 1028.
4
2. Ahn, C.-H.; Wallace, L. J.; Miller, D. D.; Feller, D. R.
Thrombosis Res. 1988, 50, 387.
3. Huzoor-Akbar Navran, S. S.; Miller, D. D.; Feller, D. R.
Biochem. Pharmacol. 1982, 31, 886.
4. Adejare, A.; Miller, D. D.; Fedyna, J. S.; Ahn, C.-H.;
Feller, D. R. J. Med. Chem. 1986, 29, 1603.
5. Roche, V. F.; Roche, E. B.; Nagel, D. L.; McPhail, A. T.
J. Org. Chem. 1984, 49, 3881.
6. Shams, G.; Fedyna, J.; Adejare, A.; Romstedt, K. J.;
Miller, D. D.; Roche, V. F.; Feller, D. R. Gen. Pharmacol.
4
4
4
4
1
991, 22, 1155.
7. Harrold, M. W.; Grazyl, B.; Shin, Y.; Romstedt, K. J.;
4
Feller, D. R.; Miller, D. D. J. Med. Chem. 1988, 31, 1506.
4
1
4
7
5
8. Taber, D. F.; Louey, J. P.; Lim, J. A. Tetrahedron Lett.
993, 34, 2243.
9. Seka, R.; Tramposch, O. Ber. Deutsch. Chem. Ges. 1942,
5, 1379.
0. Reich, H. J.; Renga, J. M.; Reich, I. L. J. Amer. Chem.
62. Neumeyer, J. L.; Baindur, N.; Yuan, J.; Booth, G.; See-
man, P.; Niznik, H. J. Med. Chem. 1990, 33, 521.
63. Hughes, D. W.; Holland, H. L.; MacLean, D. B. Can. J.
Chem. 1976, 54, 2252.
64. Borch, R. F.; Hassid, A. I. J. Org. Chem. 1972, 37, 1673.
65. Schneider, O.; Hellerback, J. Helv. Chim. Acta 1951, 34,
2218.
66. Fickes, G. N.; Metz, T. E. J. Org. Chem. 1978, 43, 4057.
67. Motulsky, H. GraphPad, Intuitive Software for Science;
GraphPad Software, Inc.: San Diego, CA, April 1996.
68. Romstedt, K. J.; Lei, L.; Feller, D. R.; Witiak, D. T.;
Loiodice, F.; Tortorella, V. Il Farmaco (Italy) 1996, 51, 107.
69. Shin, Y.; Romstedt, K. J.; Markovich, K. M.; Miller,
D. D.; Feller, D. R. Pharmacol. Commun. 1992, 1, 303.
Soc. 1975, 97, 5434.
1. Al Holly, M. M.; Hasselgren, K.-H.; Nilsson, J. L. G.
Acta. Chem. Scand. 1973, 27, 1829.
5
52. Marshall, J. A.; Jenson, T. M. J. Org. Chem. 1984, 49,
1707.
53. Miller, A. E. G.; Biss, J. W.; Schwartzman, L. H. J. Org.
Chem. 1959, 24, 627.
4. Vogel, A. I.; Tatchell, A. R.; Smith, P. W.; Rogers, V.;
Hannaford, A. J.; Furniss, B. J. In Vogel’s Textbook of
5