Full text of DOI:10.1089/108076801753266794

JOURNAL OF OCULAR PHARMACOLOGY
AND THERAPEUTICS
Volume 17, Number 5, 2001
Mary Ann Liebert, Inc.
Assessment of the Intraocular Pressure-Lowering
Activity of Bicyclic Derivatives of 1-Substituted
Benzyloctahydroisoquinoline
SHANKAR L. SAHA, IMMACULATA N. IGBO, CATHERINE A. OPERE,
GUILIN L. ZHAN, JANIS TANIYAMA, SUNNY E. OHIA,
and VICTORIA F. ROCHE
Department of Pharmaceutical and Administrative Sciences, School of Pharmacy and Allied Health
Professions, Creighton University, Omaha, Nebraska
ABSTRACT
In our study of IOP-lowering agents, we have synthesized several bicyclic analogs
of 1-benzyloctahydroisoquinoline. The target molecules were synthesized in an eleven-
step process. Structures were proved through spectrometry, elemental analysis and, in se-
lected cases, high resolution mass spectrometry. The final products were secondary or ter-
p
tiary amines containing a 1-benzyl moiety substituted at the -position with a methoxy,
methyl or chloro group. All target molecules were analyzed in 1% solution in distilled
water in normotensive rabbits. After topical administration, IOP was monitored in both
p
eyes for up to seven hours. The 1- -methoxybenzyl molecule 2 was the most active, and
6
5
caused a maximal IOP drop of 8.8 1.9 (n 7) mm Hg in the ipsilateral eye at 4 hours
post-administration,with only partial recovery at seven hours. All other compounds tested
either showed very weak activity (3–6) or were inactive (1). All compounds produced a
contralateral effect, and 5 induced rebound ocular hypertension. We conclude that selected
p
tertiary bicyclic 1- -methoxybenzyl-octahydroisoquinolines, particularly N-methylated
structures, exhibit a significant IOP-lowering effect in normotensive rabbits.
INTRODUCTION
We have previouslyreported on the synthesisand IOP-lowering action of a series of phenethylamine-
containing bicyclic derivatives of hexahydroaporphines(HHAs) (1). In a preliminary study, we showed
that these agents, particularly nor HHA (Fig. 1), enhance aqueous humor outflow facility without im-
pacting aqueous flow (2). Furthermore, we also found that prostaglandins may play a role in the pre- and
postjunctional effects of bicyclic HHAs in the bovine isolated iris (3). The 1-benzyloctahydroisoquino-
line precursor to the hexahydroaporphines,compound 1 (Fig. 2), also contains the phenethylamineadren-
ergic pharmacophore. Compound 1 and its N-methylated analog 2 have demonstrated a dose-dependent,
propranolol-reversible relaxant effect on isolated guinea pig trachea. Importantly, they had no stimula-
m
b
tory effect on atria at doses of up to 300 M, suggesting 2-adrenoceptor selectivity (4). EC50 values
m
m
obtained in the isolated tracheal studies were 15 M and 20.8 M for 1 and 2, respectively.
413

FIGURE 1. Structure of Nor HHA, an IOP Lowering Bicyclic Hexahydroaporphine
It was of interest to investigate whether these bicyclic 1-benzyloctahydroisoquinolines,with their
more flexible structure, could also lower IOP. In order to investigate the importance of both amine
p
substitution and the -methoxy group in any observed IOP lowering action, additional analogs were
synthesized (3–5, Fig. 2). Two of the new compounds were N-methylated tertiary amines (4 and 5),
p
and one contained the bulkier N- -methoxybenzyl amine substituent (3). Only 1 and 2 were expected
b
b
to be active as -adrenoceptor agonists, as the importance of hydrogen bonding of -adrenoceptor
via
SER residues to phenolic OH groups of ligands (generated from 1 and 2
oxidative O-dealkyla-
tion) has been known for some time (5).
MATERIALS AND METHODS
Synthesis of Bicyclic 1-Benzyloctahydroisoquinolines
The synthetic scheme used to produce bicyclic octahydroisoquinolines 1 and 2 has been reported
(6) and is summarized in Scheme 1. Briefly, commercially available bicyclo[2.2.2]oct-2-ene was hy-
FIGURE 2. Bicyclic 1-Benzyloctahydroisoquinolines Evaluated for IOP Lowering Activity
414

SCHEME 1. Synthesis of Bicyclic Octahydroisoquinolines 1 and 2
droborated, oxidized, and cyanomethylatedto produce nitrile 7. Reduction of 7 with lithium aluminum
hydride produced amine 8 which, upon treatment with triphenylphosphine dibromide, generated iso-
p
meric olefinic amines 9 and 10. Acylation of the enantiomeric olefinic amines with -methoxyphenyl-
acetyl chloride, and Bischler-Napieralski cyclization in phosphorus oxychloride provided the endo-
cyclic imine 11 which was reduced to secondary amine 1 with sodium borohydride. Analog 2 was
produced by formylating 1 with ethyl formate and reducing the resulting formamide with lithium alu-
p
minum hydride. Alkylation of 1 with -methoxybenzyl chloride produced analog 3.
A revised synthetic pathway was employed in the synthesis of compounds 3–5, which permit-
ted the generation of pure endocyclic olefin 8. In addition, Bischler-Napieralski cyclizations are now
conducted under milder conditions with phosphorus pentoxide, and potassium borohydride has re-
placed sodium borohydride in the subsequent reduction reaction that produces the octahydroiso-
quinoline ring system.
415

Proof of structure of the final products and all intermediates was conducted through standard
spectroscopic techniques (including 1H, 13C and two dimensional nmr, ir, and elemental analysis).
Measurement of Intraocular Pressure
Male Albino rabbits (New Zealand strain), weighing 2–2.5 kg, were purchased from Gary Smith,
Inc. (Harlan, Iowa) and conditioned on a 12 hour light-dark cycle. Animals were maintained and used
in accordance with current National Institute of Health Guidelines, and the Association for Research
in Vision and Ophthalmology (ARVO) Resolution in the Use of Animals in Research. IOP measure-
ments were conducted on restrained, awake animals using a calibrated Alcon pneumatonometer after
topical application of 0.5% proparacaine. Baseline IOPs were measured in each animal two days be-
fore each test, and immediately before evaluation of each compound. The baseline IOP values in an-
6
6
imals administered compound 2 ranged from 20.6 3.0 to 22.8 1.3 mm Hg (Table 1).
Each group of 5–8 rabbits received a single dose of bicyclic 1-benzylisoquinoline, and each dose
was tested only once in each animal. Right and left eyes were randomly assigned as ipsilateral and
contralateral, and these assignments were known to the investigators. All drugs were dissolved and
m
administered in distilled water. The ipsilateral eye of each animal was treated with 50 l of a 1% so-
m
lution of bicyclic ligand, while the contralateral (control) eye was treated with 50 l of distilled wa-
ter vehicle. On its own, vehicle had no effect on IOP in either ipsilateral or contralateral eyes. IOP
measurements were made in both eyes at 30, 60, and 90 minutes post-instillation, and every 60 min-
utes thereafter for a total of up to seven hours.
Data Analysis
The mean response to each dose of test ligand was compared against the mean baseline IOP mea-
sured immediately prior to the initiation of the experiment. Results are expressed as change in IOP
6
from baseline control values. Except where indicated otherwise, values given are means
S.E.M.
vs
vs
The significance of differences between 1) control test IOP values, and 2) test and control base-
line IOP values were evaluated using Student’s t-test (two-tailed) for unpaired observations. Differ-
,
ences with p values 0.05 were accepted as statistically significant.
RESULTS
The results of the IOP-lowering experiments conducted on compound 2 are shown in Figure 3.
Compounds 1 and 3–5 were also tested, but no significant impact on IOP was noted (data not shown).
None of the bicyclic octahydroisoquinolines 1–5 induced hyperemia or changes in pupillary diame-
p
in
ter at the doses tested. The fact that 1 and 2, which can both generate a -phenolic hydroxy group
vivo
b
, did not induce ocular vasodilation testifies to their low -adrenoceptor potency, which is pre-
416

6
FIGURE 3. IOP Lowering Activity Caused by Three Doses of 2. Values represent means S.E.M.
of 5–7 determinations. The * symbol indicates that values are significantly different from each other
,
,
at p 0.05. The f symbol indicates that values are significantly different from baseline IOP at p
0.05 or higher. Vehicle was administered to contralateral eyes.
417

m
dictable from the absence of the more critical -oxygen function. While secondary amine 1 had no
activity in this system, tertiary amine 2 demonstrated the ability to lower IOP in both treated and con-
trol eyes (Fig. 3) and induced a response that was significantly different from baseline. Specifically,
6
5
administration of one drop of a 1% solution of 2 produced a maximal 8.8 1.9 mm Hg (n 7) drop
6
in ipsilateral IOP four hours post-instillation, while the maximum contralateral IOP drop was 5.8
5
1.4 mm Hg (n 5) and occurred two hours post-instillation. IOP remained significantly below base-
line in the ipsilateral eye for four hours. A statistically significant drop in contralateral IOP from base-
line values was achieved at two and four hours post-instillation.
A dose-response study indicated that the drop in IOP produced by a 1% solution of 2 was greater
than that produced by either a 0.5% or a 0.1% solution in both ipsilateral and contralateral eyes. The
slow and steady drop in IOP produced in the test eye by a 1% solution of 2 was significantly differ-
ent from the shallow IOP drop produced by the other two concentrations at 4 and 6 hours post-in-
,
stillation (p 0.05). Significance between the 1% dose and the lower doses of 2 in the contralateral
,
eye was again observed at 2 and 4 hours post-instillation (p 0.01). There was no significant dif-
ference from baseline in the IOP-lowering action of the two lower doses of 2 in either eye, indicat-
ing a possible threshold effect of 2 at a dose between 0.5 and 1%. Doses higher than 1% were not
attempted due to the discomfort experienced by the animals when higher doses of the cyclized HHA
analogs were administered. The 1% dose induced a significant drop in IOP in the contralateral eye
at 2 and 4 hours post-administration, indicating the possibility of a centrally-mediated hypotensive
effect.
While the oxidized tetrahydroisoquinoline synthetic byproduct 6 (Fig. 4) was not a compound
specifically designed for study in this system, the ability of one drop of a 1% solution to produce a
slight decrease in IOP in both treated and control eyes proved an interesting observation (data not
shown).
DISCUSSION
The most potent IOP-lowering bicyclic 1-benzylisoquinoline, 2, possessed the reduced octahy-
p
droisoquinolinestructure, a tertiary amino nitrogen, and a -methoxy substituent on the 1-benzyl moi-
ety. In 1% doses, 2 provided an IOP lowering activity similar to that of its cyclized HHA congener
(1), indicating that both flexible and rigid bicyclic ligands containing the phenethylamine structure
are capable of reducing IOP. Given the very potent and prolonged activity of the secondary hexahy-
FIGURE 4. Structure of Compound 6, an Oxidized Tetrahydroisoquinoline Analog of 1
418

droaporphine nor HHA (Fig. 1), the lack of activity of the secondary bicyclic octahydroisoquinoline
precursor 1 was quite unexpected (1).
Lower doses of 2 (0.5% and 0.1%) dropped IOP to a lesser extent. At two time points in both
ipsilateral and contralateral eyes, the difference in IOP compared to that obtained from a 1% solution
,
was significant at p 0.05. Taken together, the data for all three doses of 2 indicate that a threshold
concentration of between 0.5% and 1% may be necessary for a significant lowering of IOP to be re-
alized.
p
Replacing the -methoxy group of 2 with the more lipophilicchloro or methyl group (compounds
4 and 5) provided significantly less potent IOP-lowering compounds, indicating that hydrophobicity
in this molecular area is not critical to activity. Hydrophobicity would facilitate CNS penetration,
however, and may be at least partially responsible for a prolonged, albeit weak, contralateral effect
exhibited by the more lipophilic of these agents, 4 (data not shown). The higher metabolic vulnera-
bility of the benzylic CH3 to oxidizing cytochrome P450 enzymes could explain the lack of a pro-
nounced and sustained IOP-lowering action by 5 in contralateral eyes. Because the chloro and methyl
substituents have differing electronic properties, and both 4 and 5 showed only weak IOP-lowering
p
action, the electron donating or withdrawing effect of the -substituent is also deemed non-critical to
IOP-lowering action. Finally, although a chlorine atom is larger than a methoxy group, a methyl group
is smaller, leading to the presumption that steric influences in this molecular area are also unimpor-
tant.
It is possible that hydrogen-bonding capabilities are important in holding these bicyclic ligands
to their receptors, and that the methoxy group found on the most potent compound is acting as a hy-
drogen acceptor in a hydrogen bond with a hydrogen-donating receptor residue (e.g., SER) (5). It is
in vivo
anticipated that O-dealkylation
would provide an active metabolite, still capable of hydrogen
bonding either as a hydrogen donating or accepting group. The slow, progressive drop in IOP pro-
duced by 2 may indicate that metabolic conversion to the phenol is an important component of its ac-
tivity.
Adding bulk to the tertiary nitrogen decreased IOP-lowering potency. Steric hindrance to bind-
p
ing is presumed responsible for the weak activity of the N- -methoxybenzyl analog 3. A similar in-
verse relationship between steric bulk of tertiary amines and IOP lowering activity was noted in hexa-
hydroaporphine analogs (1).
Aromatizing the nitrogen-containingring (6) also appears to induce an attenuating effect on IOP-
b
lowering action. This observation may not be surprising if the molecules are acting at -adrenocep-
tors, as a cationic nitrogen is known to be essential for anchoring drugs to this binding site (5). Ion-
b
ion anchoring is known to be essential in both agonist and antagonist ligands of -adrenoceptors. The
aromatic pyridine ring found in 6 is a much weaker base than the amino group of the saturated oc-
tahydroisoquinolines, and would be only weakly cationic at pH 7.4. This theoretical point, coupled
with the preliminary observation of low potency, indicated that 6 would not be a viable candidate for
further study.
p
In summary, a partially reduced bicyclic 1- -methoxybenzylisoquinoline significantly lowers in-
traocular pressure in the normotensive rabbit eye in doses of 1%. The N-methyl tertiary amine 2 was
the most active agent of the molecules studied, as it provided a distinct and sustained drop in IOP
without rebound ocular hypertension. Adding bulk to the tertiary amino nitrogen significantly atten-
p
uates the IOP-lowering effect. Neither the electronic nor lipophilic nature of the -substituent appears
p
p
critical to the IOP-lowering action of these compounds, as the -methyl and -chloro analogs were
p
also significantly less active than 2. A potential role for the -methoxy group as a hydrogen accep-
tor group in a hydrogen bond with a receptor surface is hypothesized.
While a mechanism of IOP-lowering action has not been established for either the bicyclic iso-
quinolines or HHAs, their structures contain a phenethylamine pharmacophore. Thus, they may be
allowed access to catecholaminergic receptors that mediate the inflow and outflow of aqueous hu-
mor. This hypothesis seems reasonable in view of their close structural relationship to the HHAs,
419

which have been shown to enhance outflow facility (2). Continuing studies to determine the receptor
affinity and selectivity of these unique IOP-lowering agents are planned.
ACKNOWLEDGMENTS
The authors gratefully acknowledge financial support from the Nebraska Department of Health
for the synthetic phase of this project. We also thank Dr. Michael Dalton for the generous donation
of some rabbits that were used in the latter part of this study. The pharmacological component of the
study was supported, in part, by an unrestricted grant from Alcon Laboratories, Inc.
REFERENCES
1. Saha, S., Zhan, G., Tang, L., Opere, C., Roche, E., Camras, C.B., Ohia, S.E., and Roche, V.F.
J. Ocul. Pharmacol.
Bicyclic derivates of hexahydroaporphine that lower intraocular pressure.
Ther
. 49:497–505, 1997.
2. Roche, V., Zhan, G.-L., Saha, S., Opere, C., Camras, C., and Ohia, S.E. Mechanism of intraoc-
Invest. Ophthalmol.
ular pressure-lowering activity of bicyclic analogs of hexahydroaporphine.
Vis. Sci.
(Abstr. No. 3780), 38:S812, 1997.
3. Roche, V.F., Saha, S., Opere, C.A., Ameh, F.A., Pendola, K., and Ohia, S.E. Pre- and postjunc-
Invest. Ophthalmol.
tional effects of a bicyclic hexahydroaporphine in the bovine isolated iris.
Vis. Sci.
(Abstr. No. 3710), 39:S801, 1998.
4. Shams, G., Fedyna, J., Romstedt, K.J., Adejare, A., Miller, D.D., Roche, V.F., and Feller, D.R.
Pharmacological properties of novel bicyclic isoquinolineanalogs in isolated guinea pig atria, tra-
Gen. Pharmacol.
chea, and in human platelets: Relationship to trimetoquinol.
1991.
22:1155–1163,
5. Strader, C.D., Candelore, M.R., Hill, W.S., Sigal, I.S., and Dixon, R.A.F. Identification of two
b
J. Biol. Chem.
serine residues involved in agonist activation of the -adrenergic receptor.
264:13572–13578, 1989.
6. Roche, V.F., Roche, E.G., Nagel, D.L., and McPhail, A.T. Studies in the synthesis of C ring
bridged morphinans 2. The synthesis and structural verification of a novel 3,11c-ethano-10-hy-
J. Org. Chem.
droxy-6-methyl-1,2,3,3a,11b,11c-hexahydroaporphine.
49:3881–3887, 1984.
Received: December 15, 2000
Accepted for Publication: March 19, 2001
Reprint Requests: Victoria F. Roche, Ph.D.
Department of Pharmaceutical and Administrative Sciences
School of Pharmacy and Allied Health Professions
Creighton University
Omaha, Nebraska 68178 U.S.A.
E-mail: roche@creighton.edu
420