Synthesis and activity studies of analogues of the rat selective toxicant norbormide

Article abstract of DOI:10.1016/j.bmc.2007.02.012

Norbormide [5-(α-hydroxy-α-2-pyridylbenzyl)-7-(α-2-pyridylbenzylidene)-5-norbornene-2,3-dicarboximide] (NRB, 1), an existing but infrequently used rodenticide, is known to be uniquely toxic to rats but relatively harmless to other rodents and mammals. A series of NRB-related analogues were prepared to investigate the structural features responsible for, and the in vitro biological markers indicative of, in vivo lethality of the parent molecule in rats. Their synthesis and biological evaluation (vasoconstriction, vasodilation, mitochondrial dysfunction, cardiotoxicity and lethality) is described.

Full text of DOI:10.1016/j.bmc.2007.02.012

Bioorganic & Medicinal Chemistry 15 (2007) 2963–2974
Synthesis and activity studies of analogues of the rat
selective toxicant norbormide
David Rennison,a,* Sergio Bova,b Maurizio Cavalli,b Fernanda Ricchelli,c
Alessandra Zulian,b Brian Hopkinsd and Margaret A. Brimblea,*
aDepartment of Chemistry, University of Auckland, 23 Symonds Street, Auckland, New Zealand
bDepartment of Pharmacology and Anesthesiology, Pharmacology Division, University of Padova, Padova, Italy
cCNR Institute of Biomedical Technologies/Padova Unit, University of Padova, Padova, Italy
dLandcare Research, PO Box 40, Canterbury Agriculture and Science Centre, Gerald Street, Lincoln, New Zealand
Received 24 October 2006; revised 7 February 2007; accepted 8 February 2007
Available online 11 February 2007
Abstract—Norbormide [5-(a-hydroxy-a-2-pyridylbenzyl)-7-(a-2-pyridylbenzylidene)-5-norbornene-2,3-dicarboximide] (NRB, 1), an
existing but infrequently used rodenticide, is known to be uniquely toxic to rats but relatively harmless to other rodents and mam-
mals. A series of NRB-related analogues were prepared to investigate the structural features responsible for, and the in vitro bio-
logical markers indicative of, in vivo lethality of the parent molecule in rats. Their synthesis and biological evaluation
(vasoconstriction, vasodilation, mitochondrial dysfunction, cardiotoxicity and lethality) is described.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
tively harmless to other rodents and mammals2–4 NRB
displays unique species-specific constrictor activity that
is restricted to the peripheral arteries of the rat. In arter-
ies from all other species tested, as well as in rat aorta
and extravascular smooth muscle tissue, NRB exhibits
vasorelaxant properties at concentrations that induce
vasoconstriction in the rat peripheral arteries.5 Further-
more, detailed studies conducted upon the individual
stereoisomers of NRB, isolated from the endo rich ste-
reoisomeric mixture, found the parent compound’s
physiological effects to be strongly stereospecific. In rat
peripheral arteries only the endo isomers of NRB
retained the contractile activity elicited by the stereoiso-
meric mixture, with both the endo and exo isomers
exhibiting vasodilatory activity in rat aorta.6 In vivo
evaluation established that only the endo isomers of
NRB were toxic in rats.7 Also of note is that the stereo-
chemistry at the carbinol carbon (C-5) of NRB has a
significant effect on toxicity in that endo isomers which
bear a threo configuration at this centre are 10 times
as potent as their erythro counterparts.7
Rats cause substantial damage every year to agricultural
interests worldwide. Currently, there are a number
of toxicants on the market that are effective in controlling
rats, almost all being non-specific broad-spectrum roden-
ticides. To date, rodent control has been achieved through
the use of sub-chronic poisons (e.g., cholecalciferol, bro-
methalin), acute poisons (e.g., zinc phosphide), first-gen-
eration anticoagulants (e.g., warfarin, coumatetralyl,
diphacinone, chlorophacinone) and second-generation
anticoagulants (e.g., brodifacoum, bromadiolone, bro-
methalin, difethialone, difenacoum) with varying degrees
of success.1 Second-generation anticoagulants are the
most preferred products. Most of these, however, have
one common disadvantage in that they have secondary
non-target risks associated with them and are dangerous
not only to children, but also to domestic pets, wildlife
and livestock.
Norbormide, a vasoactive compound 1960s, was intro-
duced to the market as a rodenticide approximately 30
years ago. Found to be uniquely toxic to rats but rela-
The mechanisms involved in these divergent effects of
NRB have yet to be clarified. Available evidence
suggests that the vasoconstrictor effect may be mediated
by the stimulation of a number of signal transduction
pathways8 that lead to modulation of calcium
Keywords: Norbormide; Rodenticide; Rat toxicant.
*
Corresponding authors. Tel.: +64 9 373 7599x83881; fax: +64 9 373
7422; e-mail: d.rennison@auckland.ac.nz
influx, presumably mediated by phospholipase C
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.02.012

2964
D. Rennison et al. / Bioorg. Med. Chem. 15 (2007) 2963–2974
1c
N
N
N
OH
OH
2b
2c
N
N
4(C7)
3(C7)
OH
6( C5)
7( C5)
N
5 (C5/C7)
O
1b
12 (Imide)
NH
2a
O
N
N
5c
1a
C7
C5
O
O
O
NRB 1 (C5/C7/imide)
OH
O
NH
NH
NH
11 (endo-Imide)
NH
O
O
O
34
N
O
2 (C7/endo-imide)
5a
5b
3
4
O
O
O
O
NH
NH
NH
O
O
OH
OH
OH
OH
N
OH
O
O
O
NH
NH
+
+
O
O
N
N
N
N
9a (C5/endo-imide)
9b (C5/exo-imide)
8 (C5/imide)
10b (C5/exo-imide)
10a (C5/endo-imide)
Figure 1. Stepwise ‘pseudo-deconstruction’ of NRB through the synthesis of compounds 2–12.
(PLC)-coupled receptors expressed in rat peripheral ar-
tery myocytes,8 whereas the relaxant effect may be the
result of a reduction of Ca2+ entry through L-type
Ca2+ channels.9
further deconstruction of 2 to afford bicyclic C-7 ana-
logue 3; (1c) simplification of 3 to give C-7 analogue 4;
(2a) removal of the imide ring from NRB to reveal
C-5/C-7 analogue 5; (2b) further deconstruction of 5
to afford bicyclic C-5 analogue 6; (2c) simplification of
6 to give C-5 analogue 7; (3) preparation of C-5/imide
isostere 8; (4) replacement of the 2-pyridylbenzylidene
unit at C-7 with a phenylbenzylidene group to reveal
alternative C-5/imide analogues 9a and 9b; (5a) removal
of the group at C-7 from NRB to reveal C-5/imide ana-
logues 10a and 10b; (5b) further deconstruction of 10a
to afford bicyclic endo-imide analogue 11; (5c) simplifi-
cation of 11 to give imide analogue 12.
In this paper, we postulate three potential physiological
pathways of activity leading to death: (1) selective vaso-
constriction of rat small blood vessels, as previous stud-
ies indicate that NRB elicits divergent tissue responses,
causing selective vasoconstriction of small arteries and
vasodilation of large blood vessels in the rat, while dilat-
ing both small and large blood vessels of other species;5
(2) a previously unknown deleterious effect on mito-
chondrial function which was recently attributed to
stimulation of the mitochondrial permeability transition
(MPT);10 and (3) cardiotoxicity/coronaroconstriction as
proposed by Yelnosky and Lawlor.11 Using these in vitro
indicators as biological markers for lethality in vivo, a
focused analogue design programme12 was conducted
to explore the structural features responsible for each
of the individual physiological responses (vasoconstric-
tion, vasodilation, MPT and cardiotoxicity) elicited by
NRB.
2. Chemistry
Compound 213 was accessed via the Diels–Alder reac-
tion of fulvene 1414 and maleimide, with exclusive isola-
tion of the endo15 adduct (Scheme 1 and Table 1).
Fulvene 14 was selectively obtained over the related
2-fulvenylmethanol 1514 through careful control of the
reaction conditions, via the slow addition of 2-benzoyl-
pyridine (13) to an excess of cyclopentadienyl sodium
under high dilution, in an effort to limit a second con-
densation step.14 Difficulties associated with the prepa-
ration of compound 3 focused attention on the
synthesis of further simplified C-7 analogue 4,16 pre-
pared by the treatment of ketone 13 with methylmagne-
sium bromide to afford precursor alcohol 7,17 with
subsequent acid-promoted dehydration18 providing
compound 4 (Scheme 1).
Herein we report the stepwise ‘pseudo-deconstruction’
of NRB through the synthesis of a selection of pharma-
cological probes (2–12), prepared to establish the influ-
ence of the molecule’s three key structural features
(substituents at C-7, C-5 and the imide unit) on the ob-
served biological activity (Fig. 1). The following modifi-
cations were explored: (1a) removal of the group at C-5
from NRB to reveal C-7/endo-imide analogue 2; (1b)

D. Rennison et al. / Bioorg. Med. Chem. 15 (2007) 2963–2974
2965
at elevated temperatures, afforded precursor olefin 19,
which underwent regioselective hydrogenation of the
disubstituted olefin to afford 5 as a mixture of isomers
(Scheme 2 and Table 1).
OH
c
e
N
N
N
OH
7
13
6
O
Analogue 6 was directly accessed in a single step from
norbornene using Schlosser’s conditions (a mixture of
n-butyllithium and potassium tert-butoxide) to effect
formation of the vinyl lithium species,26 that was subse-
quently trapped with ketone 13 (Scheme 1 and Table 1).
d
a
N
b
N
N
O
At this stage it was assumed that compound 10 could
not be prepared, due to the synthetic inaccessibility of
the appropriate precursor diene 33. In light of this diffi-
culty an isostere of 10, namely 8, was prepared. Com-
pound 8 was accessed via the Diels–Alder reaction of
substituted furan 26 and maleimide to afford adducts
8a and 8b (endo:exo 2:98) (Scheme 3 and Table 1).
a-Bromoester 2327 was prepared in three steps from
maleic acid (22) via sequential esterification, bromina-
tion and dehydrobromination. Reduction28 of 23 using
diisobutylaluminium hydride afforded diol 24,28 which
underwent oxidative cyclization29 in the presence of
chromic acid (generated in situ) to give 3-bromofuran
(25).30 Substituted furan 26 was prepared via reaction
of the 3-lithio derivative of 25 with 2-benzoylpyridine
(13).31 A more efficient route to 26 proceeded via 3,4-
dibromofuran (29),30 prepared in two steps from 2-bu-
tyne-1,4-diol (27) via sequential dibromination32 (to give
2832) and oxidative cyclization.29,33 Subsequent forma-
tion of the mono-lithiated furan,30 using tert-butyllithi-
um at low temperature, followed by quenching with
water gave 3-bromofuran (25), which was used as
described previously.
4
14
NH
2
O
Scheme 1. Reagents and conditions: (a) i—cyclopentadiene, NaOEt,
EtOH, rt, 0.5 h, ii—13, EtOH, rt, 2.5 h; (b) maleimide, PhH, reflux,
18 h, 25% (two steps); (c) MeMgBr, THF, rt, 18 h, 95%; (d) H2SO4
(85% aq), 100 ꢁC, 15 min, 83%; (e) i—norbornene, KOtBu, nBuLi,
THF, ꢀ78 ꢁC, 15 min warming to ꢀ40 ꢁC, 1 h, ii—LiBr, THF, ꢀ70 ꢁC
warming to ꢀ50 ꢁC, 30 min, iii—13, THF, ꢀ70 ꢁC, 30 min then H2O,
20%.
Table 1. Isomeric ratios of NRB and compounds 2, 5, 6 and 8–10
Compound
Stereochemistry
Ratio
NRBa
cis-threo/trans-threo/cis-erythro/
trans-erythro
2:2:1:1
2b
5
cis/trans
1:1
1:1:1:1
cis-threo/trans-threo/cis-erythro/
trans-erythro
erythro/threo
6
8c
9ab
9bc
10ab
10bc
1:1
Exclusive
15:85
erythro
erythro/threo
erythro
Exclusive
1:2
55:45
erythro/threo
erythro/threo
Given that 8 was found to be an unsuitable model for
10a, due to the exo-orientation of the imide ring, com-
pound9a was proposed as an alternative C-5/endo-imide
analogue. This approach was based on the ability of the
phenylbenzylidene subunit at C-7 to ‘force’ the bicyclic
system to adopt an endo-conformation, and with endo-
imide compound 3434 (Fig. 1) having proven to be inac-
tive within previous studies of this type it was hoped that
the phenylbenzylidene group would have little direct
influence on the overall pharmacological behaviour of
compound 9. Analogue 9 was accessed via the Diels–Al-
der reaction of 2-fulvenylmethanol 3235 and maleimide
to afford adducts 9a and 9b (endo:exo 1.2:1), which were
subsequently separated by column chromatography
(Scheme 4 and Table 1). 2-Fulvenylmethanol 32 was
prepared in two steps through the reaction of benzophe-
none (30) with cyclopentadienylmagnesium bromide (to
give 31a and 31b), followed by a second condensation
step with 2-benzoylpyridine (13) and concurrent dehy-
dration to afford 32.35,36 Information gathered during
the preparation of compound 9, in particular the synthe-
sis of precursor 31,35 led us to believe that the prepara-
tion of compound 10, via diene 33,should be possible.
a >90% endo isomer.
b Exclusively endo isomer.
c Exclusively exo isomer.
Retrosynthetic analysis revealed 5 to be the cycloaddi-
tion product of 2-fulvenylmethanol 15 and ethylene
(Scheme 2 and Table 1). Due to the low reactivity of eth-
ylene as a dienophile, together with its explosive nature
under pressure, initial efforts focused on accessing 5 via
a Diels–Alder reaction of 15 with ethylene equivalent
phenyl vinyl sulfone.19 Subsequent desulfonation19 of
adduct 16, using sodium-mercury amalgam, successfully
afforded 5 albeit in low yield (Scheme 2, Route 1). A sec-
ond approach to 5 focused on the bis-decarboxylation of
vicinal dicarboxylic acid 18,20 prepared over two steps
from 15 via the hydrolysis of maleic anhydride adduct
1721 (Scheme 2, Route 2). Oxidative bis-decarboxyl-
ation22 of 18 using copper (I) oxide in quinoline at ele-
vated temperatures failed to afford olefin precursor 19.
Likewise, both lead tetra-acetate-promoted bis-decar-
boxylation23 of 18 and perester decomposition24 routes
to 19 also proved unsuccessful. Ultimately a modified
Diels–Alder approach employing acetylene equivalent
2125 provided access to 5 in three steps (Scheme 2, Route
3). Subsequent fluoride-induced elimination of
b-silylsulfone 20, using tetra-butylammonium fluoride
Compound 10 was thus accessed in a similar fashion via
Diels–Alder reaction of maleimide with substituted
cyclopentadiene 33,35 to afford adducts 10a and 10b
(endo:exo 6:1), which were subsequently separated by

2966
D. Rennison et al. / Bioorg. Med. Chem. 15 (2007) 2963–2974
N
ROUTE 1
SO2Ph
OH
b
16
N
a
N
N
N
N
15
SO2Ph
h
HO
OH
OH
OH
SiMe3
20
N
19
5
N
N
N
ROUTE 3
e
N
N
c
O
d
CO2H
OH
OH
CO2H
O
18
O
N
17
N
ROUTE 2
Scheme 2. Reagents and conditions: (a) phenyl vinyl sulfone, xylene, reflux, 18 h, 50%; (b) Na/Hg amalgam, THF/MeOH, rt, 18 h, 5%; (c) maleic
anhydride, PhH, reflux, 18 h, 80%; (d) NaOH (1 M aq), H2O, 60 ꢁC, 18 h, 90%; (e) Cu2O, quinoline, 2,20-dipyridyl, powdered glass, 180 ꢁC, 24 h,
reaction failed; (f) (E)-Me3SiCH@CHSO2Ph (21), PhMe, reflux, 48 h, ca. 22%; (g) TBAF, THF, reflux, 48 h, 20%; (h) H2 (1 atm), Pd/C, EtOAc, rt,
2 h, 100%.
column chromatography (Scheme 5 and Table 1). Com-
pounds 11 and 12 were sourced commercially.
(Table 2), in terms of the maximal developed tension
as a percentage of the maximal response to KCl, ana-
logues 2, 4 and 9a displayed levels of efficacy compara-
ble to NRB. From the data in Figure 2, it would appear
that the group at C-5 plays little or no direct role in the
vasoconstriction of NRB (C-5/C-7/imide), as illustrated
by compound 2, designed to mimic NRB with respect
to the orientation of the group at C-7 and the endo-
imide subunit.
Although the following biological evaluation of com-
pounds 2–12 deals primarily with isomeric mixtures (Ta-
ble 1), the stereochemical makeup of the analogues in
the main reflects that of the stereoisomeric NRB mixture
to which they are compared.
3. Results and discussion
Although clearly less potent than NRB, at maximum
doses analogue 2 retains similar levels of contractile
activity compared to NRB. Deconstructing 2 further re-
vealed C-7 analogue 4, again a low potency, high effica-
cy vasoconstrictor. Although at this stage it would
appear that the 2-pyridylbenzylidene subunit was
responsible for NRB’s vasoconstrictive action, further
studies confirmed compounds 2 and 4 to contract in a
non-NRB-type manner, concurrently confirming the
importance of the group at C-5 to NRB-type contractile
activity and in doing so dismissing 2 and 4 as trustwor-
thy biological markers for lethality within this study.
With NRB known to be a constrictor of rat peripheral
arteries and a relaxant in rat aorta,37 initial studies were
conducted in an effort to characterize the corresponding
analogues for their contractile and relaxing effects on rat
caudal artery and aortic vascular smooth muscle, respec-
tively. As depicted in Figure 2, of the compounds pre-
pared, only analogues 2, 4 and 9a retained the
contractile activity elicited by NRB. While these
compounds were found to be considerably less potent
than NRB, as reflected by their higher EC50 values

D. Rennison et al. / Bioorg. Med. Chem. 15 (2007) 2963–2974
2967
Br
Br
NRB
O
OH
g
f
HO
9a
2
29
200
150
100
50
OH
HO
Br
Br
Br
Br
27
28
4
h
O
O
O
O
Br
a, b, c
d
e
O
OH
OH
OMe
OH
OH
25
OMe
0
22
23
24
i
-7.0
-6.5
-6.0
-5.5
-5.0
-4.5
O
O
[Drug] (Log M)
O
NH
j
OH
N
Figure 2. Contractile activity of NRB, 2, 4, and 9a in rat caudal artery.
All other compounds (5–8, 9b, 10–12) assayed were found to be
inactive as vasoconstrictors. Cumulative concentration-response
curves (0.1–50 lM) were constructed by expressing the contractile
responses as a percentage of the contraction induced by 90 mM KCl.
Each point represents the mean SE of at least 4 rings.
OH
O
N
26
8a = endo, 8b = exo
Scheme 3. Reagents and conditions: (a) MeOH, cat. H2SO4, reflux,
18 h, 95%; (b) Br2, CH2Cl2, rt, 1 h, 94%; (c) Et3N, Et2O, rt, 18 h, 97%;
(d) DIBAL, CH2Cl2, 0 ꢁC, 2 h, 51%; (e) H2SO4 (7.5% aq), Na2Cr2O7,
H2O, reflux-steam distillation, 10%; (f) Br2, CH2Cl2, rt, 18 h, 85%; (g)
t
as described for step e, 40%; (h) BuLi, Et2O, ꢀ78 ꢁC, 30 min then
alpha-1 adrenergic receptor antagonist. Interestingly,
compound 4 was found to be species-selective (active
in the rat but not in the guinea pig), whereas compound
2 was not (data not shown).
H2O, 55%; (i) i—nBuLi, Et2O, ꢀ78 ꢁC then, ii—13, Et2O, ꢀ78 ꢁC for
30 min warming to rt, 18 h, 28%; (j) maleimide, PhH, rt, 18 h, 47%.
a
Compound 5, closely resembling NRB with respect to
the orientation of the groups at C-5 and C-7, that are
conformationally locked into a rigid norbormide-type
scaffold, clearly highlighted the critical role of the imide
group in NRB-type vasoconstriction, being devoid of all
contractile activity. Compound 5 was also found to be
inactive as an alpha adrenergic agent despite incorporat-
ing the same 2-pyridylbenzylidene subunit at C-7 as
present in analogues 2 and 4, potentially the result of
the group at C-5 not being tolerated in this instance.
Compound 10a, a C-5/imide analogue prepared to
probe the importance of the group at C-7, was also
found to be inactive as a vasoconstrictor. Indeed, of
all the analogues evaluated, only 9a was found to elicit
NRB-type constriction of rat caudal artery, a result
not only verifying that the 2-pyridylbenzylidene subunit
at C-7, though beneficial, was non-essential; but more
importantly confirming that, within this series of ana-
logues, groups at all three sites (C-7, C-5 and imide)
must be retained for NRB-type contractile activity.
OH
31a
OH
31b
O
30
b
c
O
HO
OH
32
NH
N
O
N
9a
9b
= exo
=
endo
,
Scheme 4. Reagents and conditions: (a) i—cyclopentadiene, MeMgBr,
PhH, reflux, 6 h, ii—then 30, Et2O, 0 ꢁC, 30 min, 33%; (b) i—31,
NaOEt, EtOH, 0 ꢁC, ii—then 13, Et2O, 0 ꢁC, 30 min, 41%; (c)
maleimide, PhH, rt, 18 h, 40%.
In terms of a relaxing effect, analogues 5, 6, 9a and 9b
were found to be vasodilators, as demonstrated by their
capacity to relax rat aorta precontracted with KCl
(Fig. 3). Indeed, compounds 5 and 6 were found to be
equipotent to NRB.
O
OH
b
a
N
N
NH
O
OH
N
O
13
33
With analogue 7 being devoid of these effects, com-
pound 6 would appear to represent our most simplified
analogue identified to date that retains such vasodilato-
ry activity, indicating that the norbornene-backbone
supporting the group at C-5 is the source of the parent
compound’s vasodilation action. Conversely, analogue
9a, at the same concentration that induced the maximal
contraction in the rat caudal artery, relaxes KCl con-
traction by only 65% of that of NRB.
10a = endo, 10b = exo
Scheme 5. Reagents and conditions: (a) i—cyclopentadiene, MeMgBr,
PhH, reflux, 6 h, ii—then 13, Et2O, 0 ꢁC, 30 min, 12%; (b) maleimide,
PhH, rt, 18 h, 93%.
Indeed 2 and 4 were found to exhibit alpha adrenergic
activity, since their vasoconstrictor effect was antago-
nized by nM concentrations of prazosin, a selective

2968
D. Rennison et al. / Bioorg. Med. Chem. 15 (2007) 2963–2974
Table 2. Vasoconstriction, vasodilation, mitochondrial permeability transition (stimulation of PTP opening), cardiotoxicity (coronaroconstriction)
and in vivo lethality of NRB and compounds 2–12
Compound
EC50 (lM)a
Maximum
contractile effectb
Maximum
relaxation effectc
Stimulation of
PTP openingd
Coronaroconstrictione
In vivo
lethality
f
NRB
2
1
209
193
185
—
64
—
—
60
68
—
—
42
31
—
—
—
—
++++
++++
++++
+++
+
Yes
No
No
No
No
No
No
Yes
No
No
No
No
No
Yes
No
7.8
11
—
—
—
—
4.2
—
—
—
—
—
4
No
No
5
6
—
—
No
7
+
NT
NT
Yesg,h
NT
NT
NT
NT
NT
8
—
+
9a
9b
10a
10b
11
12
185
—
++
++
+
—
—
+
—
—
—
—
a EC50 for the contractile effect.
b As a % of KCl contraction (rat caudal artery).
c As a % of KCl contraction (rat aorta).
d Stimulation of the MPT as measured by the Ca2+ retention capacity of rat liver mitochondria in the presence of 50 lM NRB and NRB analogues
(see Ref. 11), (++++ high, +++ moderate, ++ low, + negligible effect).
e As evidenced in Langendorff preparations in the presence of 1 lM NRB and NRB analogues.
f 20 mg/kg ip in rats—induced death after 1 h.
g 40 mg/kg ip in rats—induced death after 1 h.
h 40 mg/kg ip in mice—no apparent effect. NT, not tested.
69 4.3
Compounds 2, 4, 5, 6 and 9a, all of which elicited a po-
75
64 0.8
sitive result in vitro indicative of NRB-type activity,
were subsequently evaluated in vivo to establish whether
they induced death in rats (Table 2). Of these only 9a
(40 mg/kg ip) was found to induce death in rats, albeit
at higher dose than NRB (20 mg/kg ip). In terms of
using these in vitro indicators as biological markers
for lethality in vivo, neither vasodilation nor mitochon-
drial dysfunction appears to be directly linked to NRB-
induced death in rats, as supported by the in vivo data
for compound 6 and compounds 2 and 4, respectively.
60 1.3
50
25
0
42 1.7
31 2.3
In the absence of any clear correlation between either
the in vitro contractile activity on rat peripheral artery,
or coronaroconstriction, with in vivo lethality it is not
possible to make any definitive conclusion about the
relative roles of these factors in NRB-induced death,
only to speculate that these physiological pathways
may indeed be linked. Studies are currently under way
to further characterize the mode of action of NRB
through the design of either a pure constrictor of rat
peripheral artery or a pure coronaroconstrictant.
Figure 3. Vasodilation activity of NRB, 5, 6, 9a and 9b (at 50 lM) in
rat aorta. All other compounds (2, 4, 7, 8, and 10–12) assayed were
found to be inactive as vasodilators. Data expressed as percentage of
90 mM KCl-induced contraction. Each bar is the mean SE (to be
calculated and added to the figure) of at least 4 rings.
With regard to mitochondrial dysfunction, compounds
2 and 4 were found to activate the permeability transi-
tion (MPT) in liver mitochondria, at levels equipotent
to NRB (Table 2). Further studies confirmed compound
4 to be species-specific having no effect on either mice or
guinea pig mitochondria. Compound 5 was deemed
moderately active, while 9a and 9b were found to have
only a small effect on MPT activation. All other ana-
logues evaluated displayed negligible levels of activity.
Again it is the 2-pyridylbenzylidene subunit at C-7
which appears to be the key structural feature of NRB
responsible for the MPT-activating effects. In terms of
cardiotoxicity, only compound 9a was found to parallel
NRB’s coronaroconstriction activity (Table 2). All other
analogues were found to be inactive.
4. Experimental
4.1. General methods (Chemistry)
All reagents were used as supplied. Solvents were
purified by standard methods.38 Analytical thin layer
chromatography (TLC) was carried out on 0.20 mm
pre-coated silica gel plates (ALUGRAMꢂ SIL G/
UV254) and products were visualized by UV fluorescence.
Flash chromatography was performed using Scharlau 60
(40–60 lm mesh) silica gel. Melting points in degres
Celsius (ꢁC) were measured on an Electrothermalꢂ

D. Rennison et al. / Bioorg. Med. Chem. 15 (2007) 2963–2974
2969
melting point apparatus and are uncorrected. Nuclear
Magnetic Resonance (NMR) spectra were recorded on
a Bruker AVANCE DRX400 (1H, 400 MHz; 13C,
100 MHz) or a Bruker AVANCE 300 (1H, 300 MHz;
13C, 75 MHz) spectrometer at 298 K. For 1H NMR data,
chemical shifts are described in parts per million (ppm)
relative to tetramethylsilane (d 0.00) and are reported con-
secutively as position (dH), relative integral, multiplicity
(s = singlet, d = doublet, t = triplet, q = quartet,
dd = doublet of doublets, dt = doublet of triplets,
qd = quartet of doublets, m = multiplet), coupling con-
stant (J/Hz) and assignment. For 13C NMR data, chemi-
cal shifts (ppm) are referenced internally to CDCl3 (d 77.0)
and are reported consecutively as position (dC) and degree
of hybridization. Assignments were aided by DEPT135,
HSQC and HMBC experiments. When two sets of peaks
arise in the NMR spectra due to the presence of isomers,
the chemical shift for the minor isomer is asterisked (*).
Mass spectra were recorded on a VG-70SE mass spec-
trometer (EI, CI and FAB). High-resolution mass spectra
were recorded at a nominal resolution of 5000.
was allowed to cool to room temperature, poured onto
ice and basified through the careful addition of aqueous
sodium hydroxide (50% w/v). The mixture was extracted
with ethyl acetate and the combined organic phases
washed with brine, then water, dried over anhydrous
magnesium sulfate and the solvent removed in vacuo.
Purification by column chromatography (hexane/ethyl
acetate 3:1) afforded 4 as a thick colourless oil (2.0 g,
1
11.0 mmol, 83%). H NMR (400 MHz, CDCl3) d 5.60
(1H, d, J = 1.4 Hz, @CH), 5.99 (1H, d, J = 1.4 Hz,
@CH), 7.19 (1H, qd, J = 7.7, 4.5 and 1.0 Hz, H-50),
7.27 (1H, d, J = 7.7 Hz, H-30), 7.32–7.37 (5H, m, Ph),
7.63 (1H, dt, J = 7.7 and 1.6 Hz, H-40), 8.64 (1H, ddd,
J = 4.5, 1.6 and 0.7 Hz, H-60); 13C NMR (100 MHz,
CDCl3) d 117.7 (CH2), 122.4 (CH), 122.8 (CH), 127.8
(CH), 128.3 (CH), 128.4 (CH); 136.3 (CH), 140.3 (C),
149.2 (C), 149.4 (CH), 158.5 (C); m/z (CI+) 182 (MH+,
100%); (Found: MH+ 182.0972, C13H12N requires
182.0970).
4.1.3. cis, trans-erythro, threo-5-(a-Hydroxy-a-2-pyr-
idylbenzyl)-7-(a-2- pyridylbenzylidene)-5-norbornene (5).
A solution of 19 (8 mg, 0.02 mmol) and palladium-on-
carbon (5 mg, 10 wt%) in ethyl acetate (1 mL) was stir-
red at room temperature under a positive atmosphere
of hydrogen for 2 h. The mixture was filtered through
a cotton wool plug and the solvent removed in vacuo.
Purification by flash chromatography (chloroform/
methanol 100:1) afforded 5 as a mixture of cis-threo/
trans-threo/cis-erythro/trans-erythro isomers (1:1:1:1)
(thick colourless oil; 8 mg, 0.02 mmol, 100%). 1H
NMR (400 MHz, CDCl3) d: 1.23–1.95 (4H, m), 3.21
(0.25H, m), 3.41–3.45 (1H, m), 3.60 (0.25H, s), 3.67
(0.25H, m), 3.69 (0.25H, m), 5.54 (0.5H, dd, J = 3.3
and 1.0 Hz, H-6), 5.74 (0.25H, s, OH), 5.76 (0.25H, s,
OH), 6.00 (0.25H, dd, J = 3.3 and 1.1 Hz, H-6), 6.02
(0.25H, dd, J = 3.3 and 1.0 Hz, H-6), 6.16 (0.25H, s,
OH), 6.26 (0.25H, s, OH), 7.04–7.60 (16H, m), 8.48–
8.61 (2H, m, a-Pyr); m/z (EI+) 442 (M+, 70%), 230
(100); (Found: M+ 442.2054, C31H26N2O requires
442.2045).
4.1.1. cis,trans-endo-7-(a-20-Pyridylbenzylidene)-5-nor-
bornene-2,3-dicarboximide (2).34 To a solution of sodi-
um ethoxide (7.49 g, 110 mmol) in absolute ethanol
(100 mL) at room temperature was added freshly dis-
tilled cyclopentadiene (7.23 g, 110 mmol), followed by
the dropwise addition of 2-benzoylpyridine (13)
(4.00 g, 21.9 mmol) in absolute ethanol (150 mL) over
2 h. The mixture was stirred for a further 30 min. and
the solvent removed in vacuo. The crude residue was
redissolved in dichloromethane, washed with water,
the organic phase dried over anhydrous magnesium sul-
fate and the solvent removed in vacuo to afford 6-phen-
yl-6-(20-pyridyl)fulvene (14)14 as a brown oil (10 g)
which was used without further purification. A solution
of crude 14 (4.8 g) and maleimide (2.0 g, 20.6 mmol) in
benzene (50 mL) was heated at 80 ꢁC overnight. The
resulting white solid was collected by filtration and
washed with ice-cold benzene. Purification by column
chromatography (hexane/ethyl acetate 1:1) afforded
endo isomer 2 as a mixture of cis/trans isomers (1:1)
(white solid; 1.67 g, 5.1 mmol, 25% over two steps).
Mp 224–226 ꢁC [lit.34 mp 223–225 ꢁC]; 1H NMR
(400 MHz, CDCl3) d 3.47 (1H, dd, J = 7.7 and
4.8 Hz), 3.67 (1H, dd, J = 7.7 and 4.8 Hz), 3.86 (1H,
m), 4.39 (1H, m), 6.46–6.48 (2H, m, @CH), 6.96 (1H,
d, J = 7.7 Hz, H-30), 7.14 (2H, dd, J = 7.7 and 1.3 Hz,
Ph), 7.20 (1H, qd, J = 7.7, 4.9 and 1.0 Hz, H-50), 7.32–
7.41 (3H, m, Ph), 7.59 (1H, dt, J = 7.7 and 1.3 Hz, H-
40), 7.78 (1H, s, NH), 8.65 (1H, dd, J = 4.9 and
1.3 Hz, H-60); 13C NMR (100 MHz, CDCl3) d 45.6
(CH), 45.7 (CH), 46.0 (CH), 122.0 (CH), 122.5 (C),
124.5 (CH), 127.3 (CH), 128.2 (CH), 129.1 (CH), 134.0
(CH), 135.2 (CH), 136.4 (CH), 138.4 (C), 148.7 (CH),
154.8 (C), 158.2 (C), 178.1 (C), 178.3 (C); IR (CHCl3)
1709 (s); m/z (EI+) 328 (M+, 25%), 231 (100); (Found:
M+ 328.1211, C21H16N2O2 requires 328.1212).
4.1.4. erythro, threo-5-(a-Hydroxy-a-20-pyridylbenzyl)-5-
norbornene (6). Compound 6 was prepared by a proce-
dure similar to that of Mayo and Tam.26 To a solution
of potassium tert-butoxide (3.03 g, 27.0 mmol) in anhy-
drous tetrahydrofuran (40 mL) under nitrogen at
ꢀ78 ꢁC was added dropwise, via a cannula, norbornene
(5.0 g, 53.1 mmol) in anhydrous tetrahydrofuran
(12 mL), maintaining a temperature below ꢀ60 ꢁC.
n-Butyllithium (16.8 mL, 26.8 mmol, 1.6 M in hexanes)
was added slowly over 15 min. maintaining a tempera-
ture below ꢀ70 ꢁC. The mixture was allowed to warm
to ꢀ40 ꢁC and maintained at this temperature for 1 h.
Following cooling to ꢀ70 ꢁC a solution of anhydrous
lithium bromide (3.46 g, 39.8 mmol) in anhydrous tetra-
hydrofuran (10 mL) was added with vigorous stirring,
allowing the reaction to warm to ꢀ50 ꢁC where it was
held for a further 30 min. After cooling to ꢀ70 ꢁC, a
solution of 2-benzoylpyridine (7.29 g, 39.8 mmol) in
anhydrous tetrahydrofuran (5 mL) was added. After
30 min. the reaction was quenched with water and
extracted with ethyl acetate. The combined organic
4.1.2. 20-(1-Phenylvinyl)pyridine (4)16. Following the
procedure of Chang et al. a solution of 7 (2.65 g,
13.3 mmol) in aqueous sulfuric acid (10 mL, 85% v/v)
was heated at 100 ꢁC for 15 min. The reaction mixture

2970
D. Rennison et al. / Bioorg. Med. Chem. 15 (2007) 2963–2974
phases were washed with water, dried over anhydrous
magnesium sulfate and the solvent removed in vacuo.
Purification by column chromatography (hexane/ethyl
acetate 3:1) afforded 6 as a mixture of erythro and threo
isomers (50:50) (white solid; 2.2 g, 7.9 mmol, 20%). Mp
84–86 ꢁC; 1H NMR (400 MHz, CDCl3) d: 0.90–1.66
(6H, m), 2.79–2.84 (1.5H, m), 2.98 (0.5H, s), 5.24
(0.5H, d, J = 3.1 Hz, H-6), 5.73 (0.5H, J = 3.1 Hz, H-
6), 5.82 (0.5H, s, OH), 6.18 (0.5H, s, OH), 7.04–7.40
(6H, m), 7.46 (0.5H, dt, J = 7.7 and 1.7 Hz, H-40),
7.53–7.56 (1H, m), 7.60 (0.5H, dt, J = 7.7 and 1.7 Hz,
H-40), 8.45–8.48 (1H, m, H-60); 13C NMR (100 MHz,
CDCl3) d: 25.1 (CH2), 25.5 (CH2), 26.0 (CH2), 26.0
(CH2), 42.3 (CH), 42.5 (CH), 43.9 (CH), 44.0 (CH),
49.0 (CH2), 49.1 (CH2), 78.0 (C), 78.4 (C), 121.6 (CH),
121.8 (CH), 122.0 (CH), 122.1 (CH), 126.8 (CH), 126.9
(CH), 127.0 (CH), 127.1 (CH), 127.5 (CH), 127.9
(CH), 132.1 (CH), 133.3 (CH), 135.9 (CH), 136.2
(CH), 144.4 (C), 144.8 (C), 146.8 (CH), 147.5 (CH),
152.7 (C), 153.2 (C), 161.3 (C), 162.2 (C); IR (CHCl3)
3430 (br); m/z (EI+) 277 (M+, 100%); (Found: M+
277.1472, C19H19NO requires 277.1467).
(CH), 143.0 (C), 148.4 (CH), 155.5 (C), 160.3 (C), 176.6
(C), 176.7 (C); IR (CHCl3) 1713 (s); m/z (EI+) 348
(M+, 8%), 251 (100); (Found: M+ 348.1100,
C20H16N2O4 requires 348.1110).
4.1.7. endo, exo-erythro, threo-5-(a-Hydroxy-a-20-pyr-
idylbenzyl)-7-(a-phenylbenzylidene)-5-norbornene-2,3-
dicarboximide (9). A solution of 32 (0.83 g, 2.0 mmol)
and maleimide (0.19 g, 2.0 mmol) in benzene (5 mL)
was heated at reflux overnight. The solvent was removed
in vacuo to afford a mixture of endo 9a and exo 9b iso-
mers (1.2:1). Purification by column chromatography
(hexane/ethyl acetate 1:1) afforded endo isomer 9a as a
mixture of erythro*/threo isomers39 (15:85) (white solid;
225 mg, 0.44 mmol, 22%); and exo isomer 9b as the
erythro isomer39 (>95%) (white solid; 184 mg, 0.36
mmol, 18%). 9a: mp 210–212 ꢁC [lit.40 mp (mixed iso-
mers) 205–207 ꢁC]; 1H NMR (400 MHz, CDCl3) d:
3.44* (0.15H, dd, J = 7.9 and 4.5 Hz), 3.49 (0.85H, dd,
J = 7.8 and 4.5 Hz), 3.53–3.56 (1H, m), 3.74* (0.15H,
m), 3.89 (0.85H, m), 3.93* (0.15H, m), 4.08 (0.85H,
m), 5.64 (0.85H, dd, J = 3.3 and 1.2 Hz, H-6), 5.72*
(0.15H, s, OH), 5.83 (0.85H, s, OH), 6.11* (0.15H, dd,
J = 3.3 and 1.2 Hz, H-6), 6.86–7.54 (18H, m), 8.25
(1H, s, NH), 8.48–8.50 (1H, m, H-60); 13C NMR
(100 MHz, CDCl3) d: 45.6* (CH), 45.8 (CH), 46.1*
(CH), 46.5 (CH), 48.0 (CH), 48.4* (CH), 49.1 (CH),
49.3* (CH), 77.2* (C), 77.7 (C), 121.9 (CH), 122.5
(CH), 122.6* (CH), 123.2* (C), 127.0* (CH), 127.1
(CH), 127.2 (CH), 127.4 (CH), 127.6 (CH), 127.9*
(CH), 128.0 (CH), 128.1 (CH), 128.2 (CH), 129.1*
(CH), 129.2 (CH), 129.3 (CH), 129.8 (CH), 133.0*
(CH), 136.4 (CH), 136.5* (CH), 139.6 (C), 139.9 (C),
142.5 (C), 147.9 (CH), 148.2* (CH), 151.7 (C), 152.3*
(C), 155.1 (C), 160.7 (C), 176.4* (C), 176.6 (C), 176.9
(C); IR (CHCl3) 1710 (s); m/z (FAB+) 511 (MH+,
8%), 154 (100); (Found: MH+ 511.2013, C34H27N2O3 re-
quires 511.2022). 9b: mp 102–104 ꢁC [lit.40 mp (mixed
4.1.5. 1-Phenyl-1-(pyridine-20-yl)ethanol (7).17 To a solu-
tion of 2-benzoylpyridine (2.5 g, 13.7 mmol) in anhy-
drous tetrahydrofuran (30 mL) under nitrogen at room
temperature was added dropwise methylmagnesium
bromide (5.0 mL, 15.1 mmol, 3.0 M in diethyl ether),
and the mixture stirred for 18 h. The reaction was
quenched by pouring onto a mixture of ice and aqueous
hydrochloric acid (1 M), basified through the addition
of solid potassium carbonate and extracted with ethyl
acetate. The combined organic phases were washed with
brine, then water, dried over anhydrous magnesium sul-
fate and the solvents removed in vacuo. Purification by
column chromatography (hexane/ethyl acetate 3:1)
afforded 7 as a thick colourless oil (2.59 g, 13.0 mmol,
95%). 1H NMR (400 MHz, CDCl3) d 1.97 (3H, s,
CH3), 7.19–7.37 (5H, m), 7.51–7.54 (2H, m), 7.67 (1H,
dt, J = 7.8 and 1.6 Hz, H-40), 8.56 (1H, m, H-60); 13C
NMR (100 MHz, CDCl3) d 29.2 (CH3), 75.1 (C), 120.3
(CH), 122.0 (CH), 125.8 (CH), 126.9 (CH), 128.2
(CH), 137.0 (CH), 147.1 (C), 147.3 (CH), 164.7 (C);
IR (CHCl3) 3429 (br); m/z (EI+) 199 (M+, 100%);
(Found: M+ 199.0994, C13H13NO requires 199.0997).
1
isomers) 205–207 ꢁC]; H NMR (300 MHz, CDCl3) d:
2.93 (1H, dd, J = 7.4 and 0.7 Hz), 2.99 (1H, dd,
J = 7.4 and 0.7 Hz), 3.59 (1H, s), 3.82 (1H, m), 5.84
(1H, s, OH), 6.21 (1H, dd, J = 3.4 and 0.9 Hz, H-6),
6.89–6.94 (2H, m), 7.04–7.07 (2H, m), 7.13–7.30 (13H,
m), 7.49 (1H, dt, J = 7.7 and 1.7 Hz, H-40), 8.32 (1H,
s, NH), 8.50 (1H, m, H-60); 13C NMR (75 MHz, CDCl3)
d: 47.6 (CH), 49.1 (CH), 50.7 (CH), 78.6 (C), 121.8
(CH), 122.7 (CH), 127.0 (CH), 127.1 (CH), 127.2
(CH), 127.5 (CH), 128.0 (CH), 128.2 (CH), 128.8
(CH), 129.0 (CH), 134.7 (CH), 136.8 (CH), 139.7 (C),
139.8 (C), 142.7 (C), 144.9 (C), 148.2 (CH), 156.7 (C),
160.6 (C), 176.9 (C), 177.4 (C); IR (CHCl3) 1708 (s);
m/z (FAB+) 511 (MH+, 7%), 154 (100); (Found: MH+
511.2015, C34H27N2O3 requires 511.2022).
4.1.6. endo,exo-threo-7-Oxa-[5-(a-hydroxy-a-20-pyridylb-
enzyl)]-5-norbornene-2,3-dicarboximide (8). A solution of
26 (200 mg, 0.8 mmol) and maleimide (66 mg, 0.68 mmol)
in benzene (5 mL) was stirred at room temperature for 3
days. The solvent was removed in vacuo to afford a mix-
ture of endo 8a and exo 8b isomers (2:98). Purification by
column chromatography (hexane/ethyl acetate 1:1) affor-
ded exo isomer 8b exclusively as the threo isomer39 (thick
colourless oil; 112 mg, 0.32 mmol, 47%). 1H NMR
(400 MHz, CDCl3) d 2.97 (1H, d, J = 6.6 Hz), 3.09 (1H,
d, J = 6.6 Hz), 5.10 (1H, s), 5.30 (1H, d, J = 1.8 Hz),
5.87 (1H, s, OH), 6.15 (1H, d, J = 1.8 Hz, H-6), 7.28–
7.39 (6H, m), 7.47 (1H, m), 7.80 (1H, dt, J = 7.7 and
1.7 Hz, H-40), 8.60 (1H, m, H-60), 8.76 (1H, s, NH); 13C
NMR (100 MHz, CDCl3) d 49.3 (CH), 49.9 (CH), 76.9
(C), 82.1 (CH), 82.4 (CH), 121.4 (CH), 123.2 (CH),
126.5 (CH), 127.7 (CH), 128.2 (CH), 134.2 (CH), 137.4
4.1.8. endo, exo-erythro, threo-5-(a-Hydroxy-a-20-pyr-
idylbenzyl)-5-norbornene-2,3-dicarboximide (10). A solu-
tion of 33 (156 mg, 0.63 mmol) and maleimide (61 mg,
0.63 mmol) in benzene (3 mL) was stirred at room tem-
perature overnight. The solvent was removed in vacuo
to afford a mixture of endo 10a and exo 10b isomers
(6:1). Purification by column chromatography (hexane/
ethyl acetate 2:1) afforded endo isomer 10a as a mixture
of erythro*/threo isomers39 (1:2) (white solid; 173 mg,

D. Rennison et al. / Bioorg. Med. Chem. 15 (2007) 2963–2974
2971
1
0.5 mmol, 79%); and exo isomer 10b as a mixture of
erythro/threo*isomers39 (55:45) (white solid; 30 mg,
0.09 mmol, 14%). 10a: mp 196–198 ꢁC; 1H NMR
(400 MHz, CDCl3) d: 1.64* (0.33H, d, J = 8.7 Hz), 1.66
(0.66H, d, J = 8.7 Hz), 2.06* (0.33H, d, J = 8.7 Hz),
2.10 (0.66H, d, J = 8.7 Hz), 3.26–3.46 (3H, m), 3.58–
3.60 (1H, m), 5.45 (0.66H, d, J = 3.1 Hz, H-6), 5.64*
(0.33H, s, OH), 5.84* (0.33H, d, J = 3.1 Hz, H-6), 5.91
(0.66H, s, OH), 7.01 (0.66H, d, J = 7.5 Hz), 7.15
(0.66H, qd, J = 7.5, 5.0 and 1.1 Hz, H-50), 7.22–7.40
(4H, m), 7.51* (0.33H, d, J = 7.9 Hz), 7.55 (0.66H, dt,
J = 7.7 and 1.7 Hz, H-40), 7.60 (1.33H, dd, J = 8.8 and
1.6 Hz), 7.76* (0.33H, dt, J = 7.7 and 1.7 Hz, H-40),
8.10 (0.66H, s, NH), 8.38* (0.33H, s, NH), 8.50
(0.66H, m, H-60), 8.58* (0.33H, m, H-60); 13C NMR
(100 MHz, CDCl3) d: 44.3* (CH), 44.4 (CH), 47.7*
(CH), 47.8 (CH), 48.8 (CH), 48.9* (CH), 53.8 (CH2),
54.3* (CH2), 77.3 (C), 121.5* (CH), 121.7 (CH), 122.3
(CH), 122.6* (CH), 126.7* (CH), 127.1 (CH), 127.5
(CH), 127.8* (CH), 128.1 (CH), 130.3 (CH), 133.6*
(CH), 136.3 (CH), 136.7* (CH), 143.2 (C), 143.7* (C),
147.7 (CH), 148.2* (CH), 153.8* (C), 154.4 (C), 161.1
(C), 161.4* (C), 177.6 (C), 177.7 (C); IR (CHCl3) 1709
(s); m/z (EI+) 346 (M+, 100%); (Found: M+ 346.1321,
C21H18N2O3 requires 346.1317). 10b: mp 208–210 ꢁC;
1H NMR (400 MHz, CDCl3) d: 1.49–1.54 (1H, m),
1.76–1.85 (1H, m), 2.76* (0.45H, dd, J = 7.1 and 0.9
Hz), 2.80 (0.55H, d, J = 7.1 Hz), 2.91 (0.55H, m),
3.13–3.15 (1H, m), 3.28–3.32 (1H, m), 3.38* (0.45H,
m), 5.50* (0.45H, d, J = 3.1 Hz, H-6), 5.94 (0.55H, s,
OH), 5.95 (0.55H, d, J = 3.1 Hz, H-6), 6.24* (0.45H, s,
OH), 7.10* (0.45H, dd, J = 8.0 and 1.0 Hz, H-30),
7.21–7.38 (5.65H, m), 7.45–7.48 (0.9H, m), 7.61*
(0.45H, dt, J = 7.6 and 1.8 Hz, H-40), 7.75 (0.55H, dt,
J = 7.7 and 1.7 Hz, H-40), 8.17 (1H, s, NH), 8.53–8.57
(1H, m, H-60); 13C NMR (100 MHz, CDCl3) d: 43.2
(CH), 45.8 (CH), 46.0* (CH), 47.0 (CH), 47.2* (CH),
49.3 (CH), 49.5* (CH), 50.2 (CH2), 77.9* (C), 78.2 (C),
121.4 (CH), 122.1* (CH), 122.5* (CH), 122.8 (CH),
126.7 (CH), 126.8 (CH), 127.4* (CH), 127.6* (CH),
128.0 (CH), 128.4* (CH), 133.1* (CH), 135.4 (CH),
136.7* (CH), 137.0 (CH), 143.2 (C), 143.7* (C), 147.3*
(CH), 148.2 (CH), 155.5 (C), 156.0* (C), 160.4* (C),
161.1 (C), 177.9 (C), 178.0* (C), 178.1 (C); IR (CHCl3)
1708 (s); m/z (EI+) 346 (M+, 100%); (Found: M+
346.1319, C21H18N2O3 requires 346.1317).
0.16 mmol, 20%). H NMR (400 MHz, CDCl3) d: 3.95
(0.25H, m), 4.00 (0.25H, m), 4.06–4.08 (0.5H, m), 4.24
(0.25H, m), 4.26 (0.25H, m), 4.47 (0.25H, m), 4.54
(0.25H, m), 5.97 (0.25H, s, OH), 6.00 (0.25H, s, OH),
6.12 (0.25H, dd, J = 3.4 and 0.9 Hz, H-6), 6.14 (0.25H,
dd, J = 3.5 and 1.1 Hz, H-6), 6.33 (0.25H, dd, J = 3.5
and 1.1 Hz, H-6), 6.36 (0.25H, dd, J = 3.4 and 1.1 Hz,
H-6), 6.40 (0.25H, s, OH), 6.61 (0.25H, s, OH), 6.91–
7.52 (18H, m), 8.51–8.54 (2H, m, a-Pyr); m/z (FAB+)
441 (MH+, 9%), 154 (100); (Found: MH+ 441.19592,
C31H25N2O requires 441.1967).
4.1.10. R,S-3-(a-Hydroxy-a-20-pyridylbenzyl)furan (26).
To
a stirred solution of 3-bromofuran (0.5 mL,
5.56 mmol) in anhydrous tetrahydrofuran (8 mL) under
nitrogen at ꢀ78 ꢁC was added dropwise n-butyllithium
(3.5 mL, 5.56 mmol, 1.6 M in hexanes) and the temper-
ature maintained for 30 min. 2-Benzoylpyridine (1.12 g,
6.12 mmol) in anhydrous tetrahydrofuran (5 mL) was
added slowly and the reaction was allowed to warm to
room temperature overnight. The reaction was
quenched with a saturated aqueous solution of ammoni-
um chloride (1 mL), diluted with water and extracted
with diethyl ether. The combined organic phases were
washed with brine, then water, dried over anhydrous
magnesium sulfate and the solvent removed in vacuo.
Purification by column chromatography (hexane/ethyl
acetate 5:1) afforded 26 as a colourless oil (0.39 g,
1
1.55 mmol, 28%). H NMR (300 MHz, CDCl3) d: 6.24
(1H, s), 6.34 (1H, t, J = 0.7 Hz), 7.03 (1H, t,
J = 0.7 Hz), 7.11 (1H, dd, J = 7.6 and 5.0 Hz, H-50),
7.22–7.29 (4H, m), 7.34 (1H, t, J = 1.6 Hz), 7.40–7.43
(2H, m), 7.54 (1H, dt, J = 7.6 and 1.6 Hz, H-40), 8.47
(1H, d, J = 5.0 Hz, H-60); 13C NMR (75 MHz, CDCl3)
d: 75.6 (C), 110.2 (CH), 121.4 (CH), 122.1 (CH), 126.8
(CH), 127.0 (CH), 127.6 (CH), 131.4 (C) 136.3 (CH),
140.7 (CH), 142.9 (CH), 145.0 (C), 147.1 (CH), 162.2
(C); m/z (EI+) 251 (M+, 100%); (Found: M+ 251.0943,
C16H13NO2 requires 251.0946).
4.1.11. 2-(a-Hydroxy-a-2-phenylbenzyl)cyclopenta-1,3-
diene (31a) and 1-(a-hydroxy-a-2-phenylbenzyl)cyclopen-
ta-1,3-diene (31b)35. To a stirred solution of methylmag-
nesium bromide (6.6 mL, 20.0 mmol, 3.0 M in diethyl
ether) under nitrogen was added benzene (12.5 mL),
and the diethyl ether distilled off until the boiling point
of the mixture reached 63 ꢁC. Freshly distilled cyclo-
pentadiene (1.32 g, 20.0 mmol) was then added and the
mixture heated at reflux for 6 h. The resultant cyclopen-
tadienylmagnesium bromide solution was then cooled to
0 ꢁC and benzophenone (3.66 g, 20.0 mmol) in diethyl
ether (10 mL) added, while maintaining the temperature
at 0 ꢁC. After stirring at 0 ꢁC for 30 min., the mixture
was hydrolysed with an excess of ice water containing
glacial acetic acid (1.2 g) and extracted with diethyl
ether. The combined organic phases were washed with
an aqueous solution of sodium hydrogen carbonate,
dried over anhydrous magnesium sulfate and the solvent
removed in vacuo. Purification by column chromatogra-
phy (hexane/ethyl acetate 10:1) afforded a regioisomeric
mixture of 31a* and 31b (1:3) as a white solid (1.63 g,
4.1.9. cis, trans-erythro, threo-5-(a-Hydroxy-a-2-pyr-
idylbenzyl)-7-(a-2- pyridylbenzylidene)-2,5-norbornadiene
(19). A solution of 15 (2.07 g, 5.0 mmol) and (E)-1-ben-
zenesulfonyl-2-(trimethylsilyl)ethylene (21) (1.21 g,
5.0 mmol) in toluene (20 mL) was heated at reflux for
2 days. The solvent was removed in vacuo with partial
purification by column chromatography (hexane/ethyl
acetate 5:1) affording an isomeric mixture of 20 as a pale
brown solid (0.71 g, ca. 1.1 mmol, ca. 22%). A solution
of 20 (0.52 g, ca. 0.8 mmol) and tetra-butylammonium
fluoride hydrate (1.05 g, 4.0 mmol) in tetrahydrofuran
(4 mL) was heated at reflux for 2 days. The solvent
was removed in vacuo, with purification by column
chromatography (hexane/ethyl acetate 5:1) affording 19
as a mixture of cis-threo/trans-threo/cis-erythro/trans-
erythro isomers (1:1:1:1) (thick colourless oil; 70 mg,
1
6.57 mmol, 33%). H NMR (400 MHz, CDCl3) d: 2.85
(1H, s, OH), 2.98 (2H, s, H-5), 5.91* (0.25H, d,

2972
D. Rennison et al. / Bioorg. Med. Chem. 15 (2007) 2963–2974
J = 1.2 Hz, H-1), 6.06 (0.75H, d, J = 1.0 Hz), 6.33–6.36
(1.5H, m), 6.40* (0.25H, m), 6.44* (0.25H, m), 7.16–
7.36 (10H, m, Ph); 13C NMR (100 MHz, CDCl3) d:
40.9* (CH2), 41.5 (CH2), 79.0* (C), 79.9 (C), 125.7*
(CH), 126.7* (CH), 127.0 (CH), 127.1 (CH), 127.2
(CH), 127.7 (CH) 128.0* (CH), 129.7* (CH), 131.2
(CH), 133.1* (CH), 133.4 (CH), 133.8* (CH), 145.7*
(C) 146.4 (C), 152.1* (C), 153.6 (C); m/z (FAB+) 248
(M+, 25%), 154 (100); (Found: M+ 248.1192, C18H16O
requires 248.1201).
extraneous fatty and connective tissue under a dissection
microscope. All vessels were cut into rings 2 mm long,
mounted on a custom-built plexiglass support by means
of two intraluminal tungsten wires and placed in 20 mL
double-jacketed organ baths filled with Tyrode solution
of the following composition (mM): NaCl 125, KCl 5,
CaCl2 2.7, MgSO4 1, KH2PO4 1.2, NaHCO3 25 and glu-
cose 11, maintained at 37 ꢁC, pH 7.35, bubbled with
95% O2 and 5% CO2. The endothelium was removed
by gently rubbing the lumen of the rings with a round
wooden stick (aorta) or a very thin rough-surfaced tung-
sten wire (caudal artery). The mechanical activity of the
rings was detected by means of an isometric force trans-
ducer (Ugo Basile, Comerio, Italy) coupled to a pen
recorder (Ugo Basile, Comerio, Italy). Rings were pas-
sively stretched to impose a resting tension (1.5–2 g)
and were allowed to equilibrate for 60 min. After equil-
ibration, each ring was repeatedly stimulated with
10 lM phenylephrine until a reproducible response
was obtained. To verify the absence of the endothelium,
rings contracted with 1 lM phenylephrine were exposed
to 2 lM carbamylcholine. The absence of the endotheli-
um was revealed by the lack of carbamylcholine-induced
relaxation. Concentration-response curves (0.1–50 lM)
were produced for each compound to assess the contrac-
tile effect on caudal artery. The inibitory effect was eval-
uated in the aorta by measuring the contractile response
to 90 mM KCl in the absence and presence of maximal
concentrations of the compound, and was expressed by
the following formula: 100 · (C ꢀ AC)/C, where C and
AC are the contractile effects measured in milligrams
of KCl alone and in the presence of the antagonist,
respectively. In a second set of experiments dose-re-
sponse curves were made for all the compounds in pres-
ence of several antagonists, to assess norbormide-like
behaviour. In particular, prazosin, a selective alpha-1
adrenergic receptor antagonist, was used at concentra-
tions ranging from 1 to 10 nM.
4.1.12. Erythro, threo-6,6,a-Triphenyl-a-(20-pyridyl)-2-
fulvenylmethanol (32)35. To a stirred solution of 31a
and 31b (1.63 g, 6.57 mmol) and 2-benzoylpyridine
(1.2 g, 6.57 mmol) in absolute ethanol (100 mL) under
nitrogen at 0 ꢁC was added a solution of sodium ethox-
ide (0.45 g, 6.57 mmol) in absolute ethanol (13 mL), and
the mixture stirred at 0 ꢁC for 3 h. The resulting orange
solid was collected by filtration, washed with ethanol
and dried in vacuo to afford 32 (1.1 g, 2.66 mmol,
1
41%) which was used without further purification. H
NMR (400 MHz, CDCl3) d: 5.94 (1H, t, J = 2.0 Hz,
H-1), 6.17 (1H, s, OH), 6.32 (1H, dd, J = 5.4 and
2.0 Hz, H-4), 6.57 (1H, dd, J = 5.4 and 1.6 Hz, H-3),
7.19–7.37 (15H, m), 7.45 (2H, d, J = 7.7 and 1.5 Hz),
7.62 (1H, dt, J = 7.5 and 1.6 Hz, H-40), 8.56 (1H, m,
H-60); 13C NMR (100 MHz, CDCl3) d: 78.1 (C), 121.2
(CH), 122.2 (CH), 122.3 (CH), 125.3 (CH), 127.3
(CH), 127.5 (CH), 127.6 (CH), 128.0 (CH), 128.8
(CH), 131.9 (CH), 132.9 (CH), 136.5 (CH), 141.0 (C),
141.2 (C), 143.0 (C), 144.9 (C), 147.4 (CH), 151.8 (C),
162.1 (C); m/z (EI+) 413 (M+, 5%), 395 (100, M+
-
H2O); (Found: M+ 413.1777, C30H23NO requires
413.1780).
4.1.13. R,S-2-(a-Hydroxy-a-20-pyridylbenzyl)cyclopenta-
1,3-diene (33). A similar procedure35 to that described
above for the preparation of 31 was followed using
2-benzoylpyridine (3.66 g, 20 mmol) to afford a regioiso-
meric mixture of 1-(a-hydroxy-a-20-pyridylbenzyl)cyclo-
penta-1,3-diene and 33 (3:2). Purification by column
chromatography (hexane/ethyl acetate 20:1) afforded
33as a thick pale orange oil (0.61 g, 2.45 mmol, 12%).
1H NMR (400 MHz, CDCl3) d: 3.00 (2H, t,
J = 1.3 Hz, H-5), 5.85 (1H, t, J = 1.7 Hz, H-1), 6.20
(1H, s, OH), 6.44 (1H, m), 6.53 (1H, m), 7.14–7.31
(5H, m), 7.40–7.42 (2H, m), 7.58 (1H, dt, J = 7.8 and
1.8 Hz, H-40), 8.51 (1H, m, H-60); 13C NMR
(100 MHz, CDCl3) d: 41.0 (CH2), 77.6 (C), 122.1
(CH), 122.2 (CH), 127.1 (CH), 127.2 (CH), 127.8
(CH), 129.8 (CH), 133.2 (CH), 133.7 (CH), 136.4
(CH), 144.8 (C), 147.4 (CH), 151.4 (C), 162.1 (C); m/z
(EI+) 249 (M+, 35%), 231 (100); (Found: M+
249.1151, C17H15NO requires 249.1154).
4.2.2. Mitochondrial permeability transition (MPT).
Mitochondria from the liver of Albino Wistar rats were
prepared according to standard differential centrifuga-
tion procedures.10 The final pellet was suspended in
Tris/HCl buffer, pH 7.4, containing 0.25 M sucrose
and 0.1 mM EGTA to give a protein concentration of
80–100 mg/mL, as measured using the biuret method.
The functionality of mitochondria was established by
measuring the value of the respiratory control ratio
(RCR), the ratio between the rate of oxygen consump-
tion in state 3 (in the presence of 0.3 mM ADP) and
in state 4 (in the absence of ADP) in a thermostated
(T = 25 ꢁC), waterjacketed vessel, using a Clark elec-
trode connected to a recorder. The incubation buffer
contained 10 mM Tris–MOPS, 100 mM sucrose, 2 mM
MgCl2, 50 mM KCl, 10 mM KH2PO4, 1 mM EDTA
and 2 lM rotenone, pH 7.4. Succinate (5 mM) was used
as the energizing substrate. Only mitochondrial suspen-
sions with a RCR P 3.0 were considered. MPT was gen-
erally induced at 25 ꢁC in a medium containing 200 mM
sucrose, 10 mM Tris–MOPS, 10 lM EGTA–Tris, 1 mM
Pi, 0.5 lg/mL oligomycin, 5 mM succinate and 2 lM
rotenone, pH 7.4 (standard medium).41 Ca2+ was used
as MPT inducer. The MPT-stimulating effects of NRB
4.2. General methods (Pharmacology)
4.2.1. Rat caudal artery and aortic ring isolation and
recording of contractile force. Male Wistar rats (150–
250 g) were obtained from Charles River Italia (Milano,
Italy) and killed by decapitation. Arteries (thoracic aor-
ta and ventral caudal artery) were collected, placed in
Tyrode solution at room temperature and cleaned of

D. Rennison et al. / Bioorg. Med. Chem. 15 (2007) 2963–2974
2973
analogues were investigated by following the Ca2+ reten-
tion capacity (CRC) of mitochondria, i.e., the amount
of calcium that can be accumulated and retained by
mitochondria until the MPT occurs. The CRC of mito-
chondrial preparations was assessed fluorimetrically in
the presence of the Ca2+ indicator Calcium Green-5N
(Molecular Probes) with a Perkin Elmer LS50B spectro-
fluorimeter exactly as described previously.10
6. Brimble, M. A.; Muir, V. J.; Hopkins, B.; Bova, S. Arkivoc
2004, 1–11.
7. Poos, G. I.; Mohrbacher, R. J.; Carson, E. L.; Paragami-
an, V.; Puma, B. M.; Rasmussen, C. R.; Roszkowski, A.
P. J. Med. Chem. 1966, 9, 537–540.
8. Bova, S.; Trevisi, L.; Cima, L.; Luciani, S.; Golovina,
V.; Cargnelli, G. J. Pharm. Exp. Ther. 2001, 296, 458–
463.
9. Fusi, F.; Saponara, S.; Sgaragli, G.; Cargnelli, G.; Bova,
S. Br. J. Pharmacol. 2002, 137, 323–328.
10. Ricchelli, F.; Dabbeni-Sala, F.; Petronilli, V.; Bernardi, P.;
Hopkins, B.; Bova, S. BBA Bioenerg. 2005, 1708, 178–186.
11. Yelnosky, J.; Lawlor, R. Eur. J. Pharmacol. 1971, 16, 117–
119.
12. For earlier structure–activity studies on NRB, see Ref. 7.
13. Roszkowski, A. P.; Poos, G. I.; Mohrbacher, R. J. Science
1964, 144, 412–413.
14. Mohrbacher, R. J.; Paragamian, V.; Carson, E. L.; Puma,
B. M.; Rasmussen, C. R.; Meschino, J. A.; Poos, G. I.
J. Org. Chem. 1966, 31, 2149–2159.
15. endo/exo Assignments were made on the basis of coupling
constants (J1,2endo = 4.8–5.0 Hz). Davis, J. C., Jr.; Van
Auken, T. V. J. Am. Chem. Soc. 1965, 87, 3900–3905, and
Ref. 6.
16. Chang, C. J.; Kiesel, R. F.; Hogen-Esch, T. E. J. Am.
Chem. Soc. 1975, 97, 2805–2810.
4.2.3. Langendorff preparation. Male Wistar rats (150–
250 g) were heparinised and sacrificed by decapitation.
The hearts were quickly removed, incannulated through
the aorta and perfused at constant flow (10 mL/min)
with a Krebs–Henseleit solution of the following com-
position (mM): NaCl 118, KCl 4.7, CaCl2 2.5, MgSO4
1.2, KH2PO4 1.2, NaHCO3 25, glucose 11 and Na pyru-
vate 2, maintained at 37 ꢁC, pH 7.35, bubbled with 95%
O2 and 5% CO2. The heart was electrically driven at a
frequency of 6 Hz through platinum electrodes placed
on the left atrium. Left ventricular developed pressure
(LVDP) was measured by inserting in the left ventricle
a steel cannula connected to a pressure transducer (2B
instruments, Italy). Coronary pressure (CO) was record-
ed by means of a pressure transducer (2B instruments,
Italy) positioned just before the aortic cannula. Each
compound was tested by infusion at a final concentra-
tion of 5 lM. Hemodynamic data were collected by a
real-time digital acquisition and analysis system (model
MP-100, Biopac System, Santa Barbara (CA)) and dis-
played on a monitor during the experiment. LVDP
and CO were expressed as mmHg.
17. Gomez, I.; Alonso, E.; Ramon, D. J.; Yus, M. Tetrahe-
dron 2000, 56, 4043–4052.
18. Adamson, D. W.; Barrett, P. A.; Billinghurst, J. W.; Jones,
T. S. G. J. Chem. Soc. 1958, 312–324.
19. Carr, R. V. C.; Williams, R. V.; Paquette, L. A. J. Org.
Chem. 1983, 48, 4976–4986.
20. Mohrbacher, R. J.; Poos, G. I.; Roszkowski, A. P. French
Patent 1386735, 1965.
21. Mohrbacker, R. J.; Poos, G. I.; Roszkowski, A. P. Belgian
Patent 659610, 1965.
4.2.4. In vivo experiments. Four male Wistar rats (150–
250 g) were used to test the lethality of each com-
pound. Briefly, all the drugs were dissolved in
200 lM of DMSO to obtain a final concentration cor-
responding to 40 mg/kg and injected into the peritoneal
cavity. All the treated animals were housed in a quiet
room and monitored every 15 min to verify early signs
of toxicity. The concentration used was arbitrarily cho-
sen as the double of that of NRB. In parallel, four rats
were injected with the same volume of solvent alone as
a control.
22. Snow, R. A.; Degenhardt, C. R.; Paquette, L. A. Tetra-
hedron Lett. 1976, 49, 4447–4450.
23. Sheldon, R. A.; Kochi, J. K. Org. React. 1972, 19,
279.
24. Cain, E. N.; Vukov, R.; Masamune, S. Chem. Commun.
1969, 98.
25. Pillot, J.-P.; Dunogues, J.; Calas, R. Synthesis 1977, 7,
469.
26. Mayo, P.; Tam, W. Tetrahedron 2002, 58, 9527–9540.
27. Jones, G.; Fantina, M. E.; Pachtman, A. H. J. Org. Chem.
1976, 41, 329–333.
28. Efskind, J.; Benneche, T.; Undheim, K. Acta Chem.
Scand. 1997, 51, 942–952.
29. Kraus, G. A.; Wang, X. Synth. Commun. 1998, 28, 1093–
1096.
Acknowledgment
30. Alvarez-Ibarra, C.; Quiroga, M. L.; Toledano, E. Tetra-
hedron 1996, 52, 4065–4078.
31. Hartman, G. D.; Halczenko, W.; Phillips, B. T. J. Org.
Chem. 1986, 51, 142–148.
The authors wish to thank Landcare Research for their
financial support and assistance.
32. Akhrem, A. A. Bull. Acad. Sci. USSR Div. Chem. Sci.
1960, 654–661 (Eng. Transl.).
References and notes
33. Gorzynsky, M.; Rewicki, D. Leibigs Ann. Chem. 1986, 4,
625–637.
34. (a) Poos, G. I.; Lehman, M. M.; Landis, E. B.;
Rosenau, J. D. J. Med. Pharm. Chem. 1962, 5, 883–
896; (b) Mohrbacher, R. J.; Almond, H. R.; Carson, E.
L.; Rosenau, J. D.; Poos, G. I. J. Org. Chem. 1966, 31,
2141–2148.
35. Mohrbacher, R. J.; Poos, G. I. U.S. Patent 3,376,304,
1968.
36. HMBC studies confirmed the formation of compound 32
through a strong 3-bond coupling of the carbinol carbon
at C-5 with proton H-6 of the pyridyl ring.
1. Handbook of Pesticide Toxicology, Vol. 2, Agents, In 2nd
ed.; Krieger, R. I., Ed.; Academic Press, 2001.
2. Przyborowski, T.; Hillar, M. Biul. Inst. Med. Morsk
Gdansk 1968, 19, 211–216.
3. Rennison, B. D.; Hammond, L. E.; Jones, G. L.
J. Hygiene 1968, 66, 147–158.
4. Roszkowski, A. P. J. Pharm. Exp. Ther. 1965, 149, 288–
299.
5. Bova, S.; Cima, L.; Golovina, V.; Luciani, S.; Cargnelli,
G. Cardiovasc. Drug Rev. 2001, 19, 226–233.

2974
D. Rennison et al. / Bioorg. Med. Chem. 15 (2007) 2963–2974
37. Bova, S.; Trevisi, L.; Debetto, P.; Cima, L.; Furnari, M.;
Luciani, S.; Padrini, R.; Cargnelli, G. Br. J. Pharmacol.
1996, 117, 1041–1046.
resonates further downfield compared to H-6 threo.
erythro/threo designations were assigned on this basis.
See Ref. 6.
38. Perrin, D. D.; Armarego, W. L. F.;Perrin, D. R. Purification
of Laboratory Chemicals; Pergamon Press: Oxford, 1980.
39. The pyridyl group attached to the carbinol carbon can
deshield H-6 more strongly when the arrangement about
the carbinol carbon is erythro. Hence H-6 erythro
40. Mohrbacher, R. J.; Poos, G. I.; Roszkowski, A. P. U.S.
Patent 3,378,566, 1968.
41. Petronilli, V.; Cola, C.; Massari, S.; Colonna, R.;
Bernardi, P. J. Biol. Chem. 1993, 268, 21939–
21945.