Full text of DOI:10.1038/sj.bjp.0704332

British Journal of Pharmacology (2001) 134, 912 ± 920
ã
2001 Nature Publishing Group All rights reserved 0007 ± 1188/01 $15.00
www.nature.com/bjp
Dominant role of an endothelium-derived hyperpolarizing factor
(EDHF)-like vasodilator in the ciliary vascular bed of the bovine
isolated perfused eye
1
1Alister J. McNeish, William S. Wilson & *,1William Martin
1Division of Neuroscience & Biomedical systems, Institute of Biomedical & Life Sciences, West Medical Building, University of
Glasgow, Glasgow G12 8QQ, Scotland
1
The roles of the endothelium-derived nitric oxide, prostacyclin and endothelium-derived
hyperpolarizing factor (EDHF) in mediating vasodilator responses to acetylcholine and bradykinin
were assessed in the ciliary vascular bed of the bovine isolated perfused eye preparation.
2
Vasodilatation to acetylcholine or bradykinin was una€ected by the nitric oxide synthase
inhibitor, L-NAME (100 mM), or the cyclo-oxygenase inhibitor, ¯urbiprofen (30 mM), but was
virtually abolished following treatment with a high concentration of KCl (30 mM), or by damaging
the endothelium with the detergent, CHAPS (0.3%, 2 min).
3
Acetylcholine-induced vasodilatation was una€ected by glibenclamide (10 mM), an inhibitor of
ATP-sensitive K+ channels (K+ATP), but was signi®cantly attenuated by TEA (10 mM), a non-
selective inhibitor of K+ channels.
4
The small conductance calcium-sensitive K+ channel (SK+Ca) inhibitor, apamin (100 nM), and
the large conductance calcium-sensitive K+ channel (BK+Ca) inhibitor, iberiotoxin (50 nM), had no
signi®cant e€ect on acetylcholine-induced vasodilatation. In contrast, the intermediate (IK+Ca)/large
conductance calcium-sensitive K+ channel inhibitor, charybdotoxin (50 nM), powerfully blocked
these vasodilator responses, and uncovered a vasoconstrictor response.
5
The combination of apamin (100 nM) with a sub-threshold concentration of charybdotoxin
(10 nM) signi®cantly attenuated acetylcholine-induced vasodilatation, but the combination of
apamin (100 nM) with iberiotoxin (50 nM) had no e€ect.
6
In conclusion, blockade by a high concentration of KCl, by charybdotoxin, or by the
combination of apamin with a sub-threshold concentration of charybdotoxin, strongly suggests that
vasodilatation in the bovine isolated perfused eye is mediated by an EDHF.
British Journal of Pharmacology (2001) 134, 912 ± 920
Keywords: Bovine isolated perfused eye; endothelium; EDHF; L-NAME; nitric oxide; potassium channels; vasodilatation
Abbreviations: BK+Ca, large conductance calcium-sensitive potassium channel; COX, cyclo-oxygenase; CHAPS, 3-[(cholami-
dopropyl)dimethyl-ammonio]1-propanesulphonate; ChTX, charybdotoxin; EDHF, endothelium-derived hyper-
polarizing factor; FBP, ¯urbiprofen; IbTX, iberiotoxin; IK+Ca, intermediate conductance calcium-sensitive
potassium channel; K+ATP, ATP-sensitive potassium channel; K+IR, inwardly-rectifying potassium channel; L-
NAME, NG-nitro-L-arginine methyl ester; NO, nitric oxide; NOS, nitric oxide synthase; ODQ, 1H-
[1,2,4]oxadiazole[4,3-a]quinoxalin-1-one; SK+
TEA, tetraethylammonium
,
small conductance calcium-sensitive potassium channel;
Ca
Introduction
Many ocular diseases such as glaucoma (Henry et al., 1999)
and diabetic retinopathy (Schmetterer et al., 1997) are
associated with dysfunction of the vascular endothelium. In
systemic arteries, the vascular endothelium plays a major role
in the regulation of vasomotor tone through the release of the
vasodilators, nitric oxide (Moncada et al., 1991), prostacyclin
(Moncada & Vane, 1979) and the putative endothelium-
derived hyperpolarizing factor (EDHF, Chen et al., 1988).
There is now compelling evidence to support the view that
endothelium-derived nitric oxide is an important mediator of
vasomotor tone in the eye (for review see Koss, 1999). For
example, inhibition of nitric oxide synthase in vivo in a wide
range of species reduces basal ocular blood ¯ow, as measured
using radiolabelled microspheres (Nilsson, 1996; Hardy et al.,
1996; Seligsohn & Anders, 2000) or laser Doppler ¯owmetry
(Zagvazdin et al., 1996; Koss, 1998; Luksch et al., 2000). It
remains to be determined whether this tonic vasodilator
in¯uence of nitric oxide on the ocular vasculature in vivo is
derived from the endothelium as a result of basal release or
release stimulated by agonists (Benedito et al., 1991a; Gidday
& Zhu, 1995) or by ¯ow (Rubanyi et al., 1986), or is derived
from perivascular nitrergic nerves (Nilsson, 1996; Toda et al.,
1998; Zagvazdin et al., 1996). Nevertheless, in studies using
the porcine isolated perfused eye, where neural in¯uences can
be excluded (Meyer et al., 1993), and in isolated segments of
ophthalmic, ciliary, retinal and irideal arteries from a variety
of species including pig, cattle and rat (Benedito et al., 1991a;
Hae¯iger et al., 1993; Hill & Gould, 1995; Su et al., 1994;
Yao et al., 1991), where the e€ects of both ¯ow and nerves
*Author for correspondence; E-mail: W.Martin@bio.gla.ac.uk

A.J. McNeish et al
Vasodilatation in the bovine perfused eye
913
can be excluded, augmentation of vasoconstrictor tone by
inhibitors of nitric oxide synthase still provides evidence of
basal release of nitric oxide from the vascular endothelium.
In addition to this basal release, acetylcholine, bradykinin
and substance P have been shown to induce powerful,
endothelium-dependent vasodilator responses sensitive to
inhibition of nitric oxide synthase in many studies conducted
either in vivo (Gidday & Zhu, 1995; Kitamura et al., 1993) or
in vitro on the isolated eye (Meyer et al., 1993) or on isolated
ocular vessels (Benedito et al., 1991a; Gidday & Zhu, 1995;
Hae¯iger et al., 1993; Yao et al., 1991).
conducting experiments with vasodilators. Once this perfu-
sion pressure was established, vasodilator responses to
acetylcholine and bradykinin were assessed by adding 10 ml
volumes of varying doses with a Hamilton micro-syringe. In
some experiments full dose-response curves to acetylcholine
(1 pmol ± 100 nmol) or bradykinin (0.1 pmol ± 10 nmol) were
constructed, but in others, only a single dose was employed.
The endothelial dependence of vasodilator responses was
tested by infusing the detergent, CHAPS (0.3%, 2 min), to
selectively damage the endothelial cell layer (Randall &
Hiley, 1988).
In contrast to the overwhelming evidence supporting the
involvement of nitric oxide, most studies suggest a minor role
for prostacyclin in the regulation of ocular vasomotor tone
(Benedito et al., 1991b; Hardy et al., 1996 Meyer et al., 1993).
Furthermore, to the best of our knowledge, the potential role
of EDHF has not yet been investigated in the eye.
Accordingly, the purpose of this study was to determine the
respective roles of endothelium-derived nitric oxide, prosta-
cyclin and EDHF in mediating the vasodilator actions of
acetylcholine and bradykinin in the bovine isolated perfused
eye preparation. A preliminary account of these ®ndings has
already been published (McNeish et al., 2001).
The e€ects of a number of blocking drugs were examined
on vasodilator responses to acetylcholine and bradykinin.
These drugs were: the nitric oxide synthase inhibitor, L-
NAME (100 mM); the inhibitor of soluble guanylate cyclase,
ODQ (10 mM); the cyclo-oxygenase inhibitor, ¯urbiprofen
(30 mM); the non-selective K+ channel blocker, TEA
(10 mM); the ATP-sensitive K+ channel (K+ATP) blocker,
glibenclamide (10 mM); the non-selective, intermediate
(IK+Ca) and large conductance (BK+Ca) calcium-sensitive
K+ channel blocker, charybdotoxin (10 and 50 nM); the
selective BK+
channel blocker, iberiotoxin (50 nM); the
Ca
selective small conductance (SK+Ca) calcium-sensitive K+
channel blocker, apamin (100 nM); the inward recti®er
(K+IR) channel blocker, Ba2+ (30 mM); and the Na+/K+
ATPase inhibitor, ouabain (10 mM). In each case the blocking
drug was infused for at least 20 min before e€ects on
vasodilator responses were tested. In some experiments the
blocking drugs themselves (L-NAME, ODQ, TEA, high K+,
charybdotoxin and iberiotoxin) a€ected the U46619-induced
perfusion pressure and these e€ects are described in the
Results section.
Methods
Preparation of tissues
The ciliary vascular bed of the bovine eye was perfused using
the constant ¯ow perfusion method of Wilson et al. (1993). In
brief, bovine eyes obtained from a local abattoir within 1 h of
killing were cannulated through a long posterior ciliary artery
and perfused at 378C with Krebs solution containing (mM):
NaCl, 118; KCl, 4.7; CaCl2, 2.5; KH2PO4, 1.2; MgSO4, 1.2;
NaHCO3, 25; glucose, 11.5; and gassed with O2 containing
5% CO2. Flow was commenced at *0.2 ± 0.5 ml min71 and
was raised in 5 ± 10 increments to 2.5 ml min71 over a 50 min
period. After the ®nal ¯ow rate was achieved, eyes were
perfused for an equilibration period of at least 30 min.
Perfusion pressure was measured using Gould Statham P32
ID transducers via a side arm located immediately proximal
to the in¯ow cannula. Only eyes that had a basal perfusion
pressure of 20 ± 60 mmHg after the equilibration period were
used for further study. In some experiments a high K+
(30 mM KCl)-containing Krebs solution was used; in these a
proportionate reduction in the NaCl concentration was made
to maintain isotonicity.
Drugs and chemicals
Acetylcholine chloride, apamin (from bee venom), barium
chloride, CHAPS (3-[(cholamidopropyl)dimethyl-ammonio]1-
propanesulphonate), charybdotoxin (scorpion venom), L-
NAME (NG-nitro-L-arginine methyl ester), ouabain, TEA
(tetraethylammonium chloride) and U46619 (9,11-dideoxy-
11a,9a-epoxy-methanoprostaglandin F2a) were obtained from
Sigma (Poole, U.K.). Glibenclamide was a gift from Hoechst
(Hounslow, U.K.). Iberiotoxin (synthetic) was obtained from
Latoxan (Valence, France), ODQ (1H-[1,2,4]oxadiazole[4,3-
a]quinoxalin-1-one) was obtained from Alexis (Nottingham,
U.K.). Flurbiprofen was a gift from the Boots Pure Drug
company (Nottingham, U.K.). All drugs were dissolved in
0.9% saline except for glibenclamide (3 mM stock), which
was dissolved in ethanol, and ODQ (0.1 M stock), which was
dissolved in dimethylsulphoxide.
Experimental protocols
After the equilibration period, drugs were added either to
the Krebs reservoir for continuous infusion, or as bolus
doses immediately proximal to the cannula. The ®rst
experiments involved constructing cumulative concentra-
tion-response curves to the thromboxane A2-mimetic,
U46619 (1 nM ± 10 mM). In these experiments, vasoconstric-
tor responses to each concentration of U46619 were allowed
to stabilize before a higher concentration was added. From
these experiments, continuous infusion of U46619 at a
concentration of 100 ± 200 nM was chosen to achieve a sub-
maximal perfusion pressure (*130 mmHg) suitable for
Statistical analysis
Results are expressed as the mean+s.e.mean of n separate
observations. Vasoconstrictor responses are given in mmHg
and vasodilator responses are expressed as percentage (%)
reduction of U46619-induced perfusion pressure. Graphs
were drawn and statistical comparisons made (Student's t-
test, or one-way analysis of variance with Bonferroni's post-
test, as appropriate) using the computer package Prism
(GraphPad, San Diego, U.S.A.). A probability (P) less than
or equal to 0.05 was considered signi®cant.
British Journal of Pharmacology vol 134 (4)

914
A.J. McNeish et al
Vasodilatation in the bovine perfused eye
Results
bolus injections of acetylcholine (1 pmol ± 100 nmol) imme-
diately proximal to the preparation produced dose-dependent
falls in the vasoconstrictor-induced perfusion pressure
(pED50=9.42+0.16, maximum fall of 64.1+4.3%, Figures
1a and 2a). In experiments in the presence of L-NAME
(100 mM, 20 min) or ODQ (10 mM, 20 min) the concentration
of U46619 was adjusted such that the ®nal perfusion
pressures obtained were not signi®cantly di€erent to control
preparations (134.9+10.9 mmHg, n=9, and 116.4+17.3
mmHg, n=5, respectively). Under these conditions neither
L-NAME (Figures 1b and 2a) nor ODQ (Figure 2a), had any
e€ect on the vasodilator actions of acetylcholine. Further-
Basal and U46619-induced perfusion pressure
The basal perfusion pressure of the ciliary vascular bed of the
bovine isolated perfused eye preparation at a constant ¯ow of
2.5 ml min71 was 31.8+1.5 mmHg (n=105). Addition of the
inhibitor of nitric oxide synthase, L-NAME (100 mM), to the
perfusate had no e€ect on this basal perfusion pressure
(change of 70.6+3.3 mmHg, n=8). The inhibitor of soluble
guanylate cyclase, ODQ (10 mM), did, however, produce a
small but signi®cant rise in perfusion pressure of
14.6+3.6 mmHg (n=10, P50.05). The thromboxane A2-
mimetic, U46619 (1 nM ± 10 mM), produced a concentration-
dependent rise in perfusion pressure (pEC50=5.61+0.03,
maximum rise of 187.6+5.6 mmHg, n=6). Moreover, when
U46619 (*200 nM) was infused to obtain a sub-maximal rise
in perfusion pressure of 86.0+4.3 mmHg, (n=16), the
addition of L-NAME (100 mM, 20 min) produced a further
rise of 37.8+5.0 mmHg (n=16, Figure 1b).
Endothelial dependence of vasodilator responses to
acetylcholine and bradykinin and effects of L-NAME,
ODQ and flurbiprofen
Following infusion of U46619 (100 ± 200 nM), which in-
creased the perfusion pressure to 131.8+16.3 mmHg (n=8),
Figure 2 Control dose-response curves showing vasodilatation
induced by bolus doses of (a) ACh (1 pmol ± 100 nmol) and (b) BK
(100 fmol ± 10 nmol) on the bovine isolated perfused eye preparation.
The e€ects of treatment with the nitric oxide synthase inhibitor, L-
NAME (100 mM), the inhibitor of soluble guanylate cyclase, ODQ
(10 mM), or a high concentration of KCl (30 mM) are also shown.
Data represent mean+s.e.mean of ®ve or more observations.
*P50.05 and ***P50.001 indicate a di€erence from control.
Figure 1 Original traces showing acetylcholine-induced vasodilata-
tion in the bovine isolated perfused eye. The perfusion pressure was
increased using the thromboxane A2-mimetic, U46619 (200 nM), prior
to addition of graded doses of ACh. (a) Dose-dependent vasodilata-
tion produced by ACh (0.01 ± 100 nmol). (b) Addition of the nitric
oxide synthase inhibitor, L-NAME (100 mM), produced vasoconstric-
tion but had no e€ect on ACh-induced vasodilatation.
British Journal of Pharmacology vol 134 (4)

A.J. McNeish et al
Vasodilatation in the bovine perfused eye
915
more, the fall in perfusion pressure induced by 10 nmol
acetylcholine (45.6+3.2%, n=18) was una€ected by the
cyclo-oxygenase inhibitor ¯urbiprofen (30 mM, 20 min) nor
by the combination of ¯urbiprofen and L-NAME (Figure 3a).
Bradykinin (0.1 pmol ± 10 nmol) also produced dose-de-
pendent falls in U46619-induced perfusion pressure
(pED50=10.95+0.05, maximum fall of 68.8+1.1%, n=6,
Figure 2b). In the presence of L-NAME (100 mM, 20 min),
bradykinin-induced falls in perfusion pressure were com-
pletely una€ected (Figure 2b). In addition, the fall in
perfusion pressure produced by 10 pmol bradykinin
(47.0+4.4%, n=15) was una€ected by ¯urbiprofen or by
the combination of ¯urbiprofen and L-NAME (Figure 3b).
Infusion of the detergent CHAPS (0.3%, 2 min), in order
to damage the endothelium, had no e€ect on U46619-induced
perfusion pressure (160.0+16.4 and 170.2+16.6 mmHg,
n=13, before and after CHAPS, respectively), but resulted
in a profound reduction in the fall in perfusion pressure
induced by 10 nmol acetylcholine or 10 pmol bradykinin
(Figure 3a,b).
Effect of high K+ on vasodilator responses to
acetylcholine and bradykinin
The ability of acetylcholine and bradykinin to induce
vasodilator responses was examined in the presence of a
high concentration of KCl (30 mM). In these experiments
KCl itself produced a sustained increase in perfusion pressure
to 98.7+7.1 mmHg (n=12), but in order to examine
vasodilator responses at levels of pressure similar to control
experiments the pressure was increased to 137.6+12.3 mmHg
(n=12) with the infusion of U46619 (5 ± 10 nM). Under these
conditions, vasodilator responses to acetylcholine (1 pmol ±
100 nmol) and to bradykinin (0.1 pmol ± 10 nmol) were
powerfully inhibited (Figure 2a,b).
Effect of K+ channel blockers and ouabain on
acetylcholine-induced vasodilatation
A number of agents that block speci®c potassium channels
were infused for 20 min and their e€ects on acetylcholine
(10 nmol)-induced vasodilator responses were assessed. The
K+
channel blocker, glibenclamide (10 mM), had no e€ect
ATP
on U46619-induced perfusion pressure or on acetylcholine-
induced vasodilatation. In contrast, the non-selective K+
channel blocker, TEA (10 mM), produced a large transient
increase
in
U46619-induced
perfusion
pressure
(291.9+57.5 mmHg, n=4) as well as a signi®cant inhibition
of acetylcholine-induced vasodilatation (Figure 4a).
The K+ channel blocker, barium (30 mM), had no e€ect
IR
on acetylcholine-induced vasodilatation, but the Na+/K+
ATPase inhibitor, ouabain (10 mM), produced signi®cant
blockade (Figure 4b). The combination of barium and
ouabain produced
a greater blockade of acetylcholine-
induced vasodilatation than was seen with ouabain alone
(Figure 4b). Neither barium nor ouabain had any e€ect on
U46619-induced perfusion pressure.
The SK+
channel blocker, apamin (100 nM) had no
signi®cant e€ect on U46619-induced perfusion pressure but
Ca
the IK+Ca/BK+ inhibitor, charybdotoxin (50 nM; Figure 5)
and the BK+ channel inhibitor, iberiotoxin (50 nM), each
Ca
Ca
produced a large transient vasoconstriction (rise in perfusion
pressure of 339.6+18.8 mmHg, n=8, and 298.5+38.1
mmHg, n=7, respectively). Neither apamin (100 nM) nor
iberiotoxin (50 nM) alone had any signi®cant e€ect on the
acetylcholine (10 nmol)-induced vasodilatation (Figure 6b),
whereas charybdotoxin (50 nM) produced
a signi®cant
blockade of vasodilatation, and uncovered a vasoconstriction
(Figures 5 and 6a). A lower concentration of charybdotoxin
(10 nM) had no signi®cant e€ect on perfusion pressure or
acetylcholine-induced vasodilatation by itself, but signi®-
cantly attenuated the vasodilatation when combined with
apamin (100 nM, Figure 6a,b). In contrast the combination of
iberiotoxin (50 nM) and apamin (100 nM) had no e€ect on
acetylcholine-vasodilatation.
Figure 3 Histogram showing the e€ects of damaging the endothe-
lium with CHAPS (0.3%, 2 min), inhibiting cyclo-oxygenase with
¯urbiprofen (FBP, 30 mM) and combined treatment with ¯urbiprofen
and the inhibitor of nitric oxide synthase, L-NAME (100 mM), on
vasodilatation induced by (a) ACh (10 nmol) or (b) BK (10 pmol).
Data represent the mean+s.e.mean of ®ve or more observations.
***P50.001 indicates a di€erence from control.
British Journal of Pharmacology vol 134 (4)

916
A.J. McNeish et al
Vasodilatation in the bovine perfused eye
Figure 5 Original trace showing the e€ects of charybdotoxin
(ChTX, 50 nM) on perfusion pressure and on vasodilatation to
ACh (10 nmol). The addition of charybdotoxin resulted in a large
transient increase in perfusion pressure followed by reversal of ACh-
induced vasodilatation to vasoconstriction.
has also been reported in the porcine isolated perfused eye
preparation (Meyer et al., 1993).
Whether this `basal' nitric oxide activity results from a
tonic unstimulated release of nitric oxide or from release
stimulated by ¯ow-induced shear stress acting on the vascular
endothelium (Rubanyi et al., 1986) remains to be determined.
Nevertheless, studies using bovine isolated retinal rings
(Benedito et al., 1991a) and porcine ophthalmic and ciliary
arterial rings (Hae¯iger et al., 1993) in tissue baths where
shear stress is absent also indicate the presence of a tonic
vasodilator action of basal nitric oxide. Moreover, the ability
of inhibitors of nitric oxide synthase to reduce ocular blood
¯ow in species as diverse as rat (Koss, 1998), rabbit (Nilsson,
1996), dog (Kitamura et al., 1993) and human (Luksch et al.,
2000), demonstrates an important role for nitric oxide in
regulating blood ¯ow in the eye.
We wished to investigate the role of the vascular
endothelium in the vasodilator responses to acetylcholine
and bradykinin in the bovine isolated perfused eye. We found
that both drugs induced powerful dose-dependent vasodilator
responses. Moreover, infusion of the detergent, CHAPS,
which has previously been used to induce selective damage of
the vascular endothelium in the rat isolated perfused
Figure 4 Histogram showing the e€ects of potassium channel
blockers and of ouabain on vasodilatation induced by ACh
(10 nmol). (a) E€ects of the K+
channel blocker, glibenclamide
ATP
(10 mM), and of the non-selective K+ channel blocker, tetraethyl-
Ca
ammonium (TEA, 10 mM). (b) E€ects of the inward recti®er K
channel (KIR) blocker, barium (30 mM), and of the Na+/K+ ATPase
inhibitor, ouabain (10 mM), each alone and in combination. Data
represent mean+s.e.mean of ®ve or more observations. ***P50.001
indicates a di€erence from control.
mesentery preparation (Randall
& Hiley, 1988), almost
abolished vasodilator responses to acetylcholine and brady-
kinin in the bovine eye. Surprisingly, CHAPS failed to
reproduce the rise in U46619-induced perfusion pressure seen
when basal nitric oxide activity was abolished with L-NAME.
It is possible, however, that CHAPS might have induced
some slight additional damage to the smooth muscle which
obscured this expected rise. It seems reasonable to conclude,
however, that the vasodilator responses to acetylcholine and
bradykinin are endothelium dependent in this preparation.
Nevertheless, we found that vasodilator responses to
acetylcholine and bradykinin were completely una€ected
following the inhibition of nitric oxide synthase by L-NAME.
Moreover, the inhibitor of soluble guanylate cyclase, ODQ
also failed to inhibit vasodilator responses to acetylcholine.
Thus, our ®ndings suggest that nitric oxide does not
contribute at all to the vasodilator responses to acetylcholine
or bradykinin in the bovine eye. These ®ndings are surprising
and are in sharp contrast to those obtained in isolated rings
Discussion
The ®ndings of the present study demonstrate that in the
ciliary vascular bed of the bovine isolated perfused eye
preparation, the vascular endothelium plays an important
role in the regulation of perfusion pressure. Speci®cally,
following preconstriction with a submaximal concentration of
the thromboxane A2-mimetic, U46619, we found that
inhibition of nitric oxide synthase with L-NAME (100 mM)
resulted in further increase in perfusion pressure. This
suggested that a basal nitric oxide activity exerts a tonic
vasodilator e€ect in the resistance vessels of the bovine ciliary
vasculature. A similar tonic inhibitory action of nitric oxide
British Journal of Pharmacology vol 134 (4)

A.J. McNeish et al
Vasodilatation in the bovine perfused eye
917
newborn pigs is sensitive to inhibition of nitric oxide synthase
(Gidday & Zhu, 1995).
We also found that the vasodilator responses to acetylcho-
line and bradykinin in the bovine eye were completely
una€ected following treatment with the cyclo-oxygenase
inhibitor, ¯urbiprofen, or following combined treatment with
¯urbiprofen and L-NAME. It is therefore clear that neither
nitric oxide nor a cyclo-oxygenase product is responsible for
the powerful vasodilator responses to acetylcholine and
bradykinin in this preparation. To ®nd that acetylcholine-
and bradykinin-induced vasodilatation in the bovine eye is
completely independent of involvement of nitric oxide was
rather surprising, but similar ®ndings have been made in the
rabbit hindlimb (Mugge et al., 1991) and rat renal artery
(Jiang
& Dusting, 2001). In blood vessels where an
endothelium-derived hyperpolarizing factor (EDHF) report-
edly exists, such as in the rat mesenteric vascular bed
(McCulloch et al., 1997; Adeagbo & Triggle, 1993) or in
human coronary (Kato et al., 1997) or forearm resistance
vessels (Honing et al., 2000), it is more common to ®nd it
functioning together with nitric oxide.
The nature of the EDHF is still unresolved, but possible
candidates include a cytochrome P450 metabolite (Bauersachs
et al., 1994), perhaps an epoxyeicosatrienoic acid (EET;
Fisslthaler et al., 1999), an endogenous cannabinoid (Randall
et al., 1996) or potassium ions (Edwards et al., 1998).
Moreover, evidence has been reported suggesting that the
process of relaxation involves hyperpolarization of endothe-
lial cells (Edwards et al., 1998; 2000) and consequent gap
junctional transmission between endothelium and smooth
muscle, resulting in smooth muscle hyperpolarization and
relaxation (Davies et al., 1988; Beny, 1990; Chaytor et al.,
1998). Regardless of the con¯icting evidence for the proposed
candidates, there is consensus that EDHF responses are
mediated by an increase in K+ conductance and thus are
sensitive to depolarizing solutions of K+ (Adeagbo & Triggle,
1993; Corriu et al., 1996a) and by inhibitors of certain types
of K+ channels. The possibility that acetylcholine- and
bradykinin-induced vasodilator responses in the bovine eye
were produced by an EDHF was therefore investigated
further. Consistent with this possibility, we found that in the
presence of 30 mM KCl, vasodilator responses to acetylcho-
line and bradykinin were almost abolished.
Figure 6 Histograms showing the e€ect of calcium-sensitive
potassium channel inhibitors on vasodilatation induced by ACh
(10 nmol). (a) E€ects of the large/intermediate conductance (BK+
/
Ca
IK+Ca) calcium-sensitive K+ channel blocker, charybdotoxin (ChTX,
10 nM and 50 nM). (b) E€ects of the small conductance calcium-
sensitive K+ (SK+Ca) channel blocker, apamin (100 nM), alone and
in combination with a low concentration of ChTX (10 nM), and of
the large conductance calcium sensitive K+ (BK+
) channel
Ca
blocker, iberiotoxin (IbTX, 50 nM), alone and in combination with
apamin (100 nM). Data represent the mean+s.e.mean of ®ve or more
observations. *** P50.001 indicates a di€erence from control and
##P50.005 shows that the combination of iberiotoxin and apamin
was di€erent from iberiotoxin alone.
It has been suggested that K+ released from the
endothelium can act as an EDHF by stimulating the Na+/
K+ATPase or inwardly rectifying K+ channels (K+IR) on
smooth muscle of rat hepatic arteries (Edwards et al., 1998).
of porcine ophthalmic artery (Yao et al., 1991) or ciliary
artery (Hae¯iger et al., 1993; Zhu et al., 1997) and strips of
monkey ciliary artery (Toda et al., 1998), where vasodilata-
tion induced by either acetylcholine or bradykinin were
sensitive to inhibition of nitric oxide synthase. It is thus
possible that nitric oxide has more importance in regulating
vascular tone in larger vessels than in resistance vessels in the
bovine ciliary vasculature. Indeed, such a decreasing role for
nitric oxide with decreasing size of blood vessel has been
previously noted in the rat mesenteric vascular bed (Hwa et
al., 1994). However, in contrast to our ®ndings in the bovine
eye, bradykinin-induced vasodilator responses in the porcine
isolated perfused eye were powerfully blocked by L-NAME
(Meyer et al., 1993), suggesting an important role for nitric
oxide in such responses in this species. Furthermore,
vasodilatation to acetylcholine in the retinal circulation of
Therefore, we examined the e€ect of the K+ inhibitor,
IR
barium, and the Na+/K+ATPase inhibitor, ouabain, on
vasodilator responses to acetylcholine in the bovine perfused
eye preparation. Vasodilatation was una€ected by barium,
inhibited by ouabain and further inhibited by the combina-
tion of ouabain and barium. The ®nding that vasodilator
responses to acetylcholine in the bovine eye are sensitive to
inhibition of the Na+/K+ATPase and K+
could be
IR
consistent with the hypothesis that K+ ions can act as
EDHF. However, the possible role of K+ has been
challenged recently by the ®nding that EDHF-mediated
responses are una€ected by ouabain or barium in guinea-
pig carotid and porcine coronary artery (Quignard et al.,
1999) and rat hepatic artery (Andersson et al., 2000).
Moreover, in rat mesenteric (Lacy et al., 2000) and renal
British Journal of Pharmacology vol 134 (4)

918
A.J. McNeish et al
Vasodilatation in the bovine perfused eye
arteries (Jiang & Dusting, 2001), potassium-induced vasodi-
latation is abolished following removal of the endothelium,
thus casting doubt on a role for this ion as EDHF in these
vessels.
Moreover, apamin (100 nM) and iberiotoxin (50 nM) alone or
in combination had no signi®cant e€ect on the acetylcholine-
induced vasodilatation. In contrast, charybdotoxin (50 nM)
alone signi®cantly attenuated acetylcholine-induced vasodila-
tation, and uncovered a vasoconstrictor response. In previous
studies (Zygmunt & Hogestatt, 1996; Edwards et al., 1998;
1999) charybdotoxin alone did not attenuate acetylcholine-
induced vasodilatation as powerfully as was seen in the
bovine eye, and this may re¯ect the absence of the unusual
acetylcholine-induced vasoconstriction observed in our pre-
paration. Full characterization of this vasoconstrictor
response was beyond the scope of this study, but it appears
to be endothelium-dependent, since it was not observed when
the endothelium was functionally impaired with CHAPS.
Previous studies have reported anomalous, endothelium-
dependent vasoconstrictor responses in rat aorta (Franchi-
Micheli et al., 2000) and porcine renal interlobar artery
(Derkach et al., 2000). A lower concentration of charybdo-
toxin (10 nM) had no e€ect on perfusion pressure or
vasodilatation. Nevertheless, when this lower concentration
was combined with apamin, a signi®cant attenuation of
acetylcholine-induced vasodilatation was observed. Thus, the
attenuation of the acetylcholine-induced vasodilatations with
the combination of apamin and charybdotoxin, and the lack
of e€ect of the combination of apamin and iberiotoxin is
consistent with the possibility that the vasodilator responses
observed in the bovine perfused eye are mediated by an
EDHF.
Having established that high K+ powerfully inhibited
endothelium-dependent vasodilatation in the bovine eye, we
investigated the possible involvement of potassium channels.
K+
channels are not generally considered to be involved
in EDHF-mediated vasodilatation (Corriu et al., 1996b) and,
ATP
consistent with this, we found that addition of glibenclamide,
blocker of K+
a
channels, had no e€ect on the
ATP
vasodilatation induced by acetylcholine in the bovine eye.
The vasodilatation was, however, inhibited by TEA, which is
a
relatively non-selective calcium-sensitive K+ (K+
)
Ca
channel inhibitor (Cook & Quast, 1990), although it does
not block SK+Ca, even at concentrations up to 4 mM (Zima
et al., 2000). Our ability to block acetylcholine-induced
vasodilatation in the bovine eye with TEA is in keeping with
its ability to inhibit EDHF-mediated vasodilatation in human
forearm resistance vessels (Honing et al., 2000) and in rat
mesenteric arteries (Chen & Cheung, 1997).
Recent evidence suggests that the action of EDHF is
normally abolished by the combination of the small
conductance K+ (SK+Ca) channel blocker, apamin, with
Ca
the large and intermediate conductance K+ (BK+Ca/IK+
blocker, charybdotoxin, but not by the combination of
)
Ca
Ca
apamin and the large conductance K+ (BK+Ca) inhibitor,
iberiotoxin (Zygmunt & Hogestatt, 1996; Edwards et al.,
Ca
1998; 1999). On the basis of these results, the EDHF pathway
is thought to involve the opening of SK+ channels, and
In summary, a basal level of nitric oxide regulates
perfusion pressure in the cilliary vascular bed of the bovine
perfused eye. Nevertheless, despite acetylcholine- and brady-
kinin-induced vasodilatation being endothelium-dependent,
they do not involve nitric oxide or products of cyclo-
oxygenase. The blockade of acetylcholine-induced vasodilata-
Ca
IK+ channels but not BK+ channels on the endothelial
Ca
Ca
cell (Edwards et al., 1998; Ohashi et al., 1999). We found in
the bovine eye that apamin (100 nM) had no signi®cant e€ect
on U46619-induced perfusion pressure but charybdotoxin
(50 nM) and iberiotoxin (50 nM) both caused a large transient
vasoconstrictor response, similar to that seen with TEA (see
tion by 30 mM KCl, or by K+
channel inhibitors,
Ca
particularly charybdotoxin (50 nM), or the combination of
apamin (100 nM) with a low concentration of charybdotoxin
(10 nM), suggest involvement of an EDHF. To the best of
our knowledge, this is the ®rst report to suggest an important
role for an EDHF in the control of vasomotor tone in the
eye.
above). Inhibition of BK+ channels is known to inhibit
Ca
nitric oxide-induced vasodilatation (Demirel et al., 1994;
Champion & Kadowitz, 1997). We think it is unlikely,
however, that these powerful transient vasoconstrictions
re¯ect inhibition of basal nitric oxide activity, since the nitric
oxide synthase inhibitor, L-NAME, produced much smaller
rises in perfusion pressure which were maintained for many
hours. Perhaps our ®ndings indicate that BK+ channels are
Ca
We are grateful to the British Heart Foundation and Wellcome
Trust for support.
normally open under the conditions used in our study.
References
ADEAGBO, A.S. & TRIGGLE, C.R. (1993). Varying extracellular [K+]:
a functional approach to separating EDHF- and EDNO-related
mechanisms in perfused rat mesenteric arterial bed. J. Cardio-
vasc. Pharmacol., 21, 423 ± 429.
BENEDITO, S., PRIETO, D., NIELSEN, P.J. & NYBORG, N.C.B.
(1991a). Role of the endothelium in acetylcholine-induced
relaxation and spontaneous tone of bovine isolated retinal small
arteries. Exp. Eye Res., 52, 575 ± 579.
ANDERSSON, D.A., ZYGMUNT, P.M., MOVAHED, P., ANDERSSON,
T.L.G. & HOGESTATT, E.D. (2000). E€ects of inhibitors of small-
and intermediate-conductance calcium-activated potassium
channels, inwardly-rectifying potassium channels and Na+/K+
ATPase on EDHF relaxations in the rat hepatic artery. Br. J.
Pharmacol., 129, 1490 ± 1496.
BAUERSACHS, J., HECKER, M. & BUSSE, R. (1994). Display of the
characteristics of endothelium-derived hyperpolarizing factor by
a cytochrome P450-derived arachidonic acid metabolite in the
coronary microcirculation. Br. J. Pharmacol., 113, 1548 ± 1553.
BENEDITO, S., PRIETO, D., NIELSEN, P.J. & NYBORG, N.C.B.
(1991b). Histamine induces endothelium-dependent relaxation
of bovine retinal arteries. Invest. Ophthalmol. Vis. Sci., 32, 32 ±
38.
BENY, J.L. (1990). Endothelial and smooth-muscle cells hyperpolar-
ized by bradykinin are not dye coupled. Am. J. Physiol., 258,
H836 ± H841.
CHAMPION, H.C. & KADOWITZ, P.J. (1997). Vasodilator responses
to acetylcholine, bradykinin and substance P are mediated by a
TEA-sensitive mechanism. Am. J. Physiol., 273, R414 ± R422.
British Journal of Pharmacology vol 134 (4)

A.J. McNeish et al
Vasodilatation in the bovine perfused eye
919
CHAYTOR, A.Y., EVENS, W.H. & GRIFFITH, T.M. (1998). Central
role of heterocellular gap junctional communication in endothe-
lium-dependent relaxations of rabbit arteries. J. Physiol., 508,
561 ± 573.
JIANG, F. & DUSTING, G.J. (2001). Endothelium-dependent vas-
orelaxation independent of nitric oxide and K+ release in
isolated renal arteries of rats. Br. J. Pharmacol., 132, 1558 ± 1564.
KATO, M., SHIODE, N., YAMAGATA, T., MATSUURA, H.
&
CHEN, G. & CHEUNG, D.W. (1997). E€ect of K+-channel blockers in
acetylcholine-induced hyperpolarisation in rat mesenteric ar-
teries. Am. J. Physiol., 272, H2306 ± H2312.
KAJIYAMA, G. (1997). Bradykinin induced dilatation of human
epicardial and resistance coronary arteries in vivo: e€ect of
inhibition of nitric oxide synthesis. Heart, 78, 493 ± 498.
KITAMURA, Y., OKAMURA, T., KANI, K. & TODA, N. (1993). Nitric
oxide mediated retinal arteriolar and arterial dilatation induced
by substance P. Invest. Ophthalmol. Vis. Sci., 34, 2859 ± 2865.
KOSS, M.C. (1998). Role of nitric oxide in maintenance of basal
anterior choroidal blood ¯ow in rats. Invest. Ophthalmol. Vis.
Sci., 39, 559 ± 564.
KOSS, M.C. (1999). Functional role of nitric oxide in regulation of
ocular blood ¯ow. Eur. J. Pharmacol., 374, 161 ± 174.
LACY, P.S., PILKINGTON, G., HANVESAKUL, R., FISH, H.J., BOYLE,
J.P. & THURSTON, H. (2000). Evidence against potassium as an
endothelium-derived hyperpolarizing factor in rat mesenteric
arteries. Br. J. Pharmacol., 129, 605 ± 611.
CHEN, G., SUZUKI, H.
& WESTON, A.H. (1988). Acetylcholine
releases endothelium-derived hyperpolarizing factor and EDRF
from rat blood vessels. Br. J. Pharmacol., 95, 1165 ± 1174.
COOK, N.S. & QUAST, U. (1990). Potassium channel pharmacology.
In Potassium channels: structure, classi®cation, function and
therapeutic potential. pp 181 ± 255. Chichester: Ellis Hardwood.
CORRIU, C., FELETOU, M., CANET, E. & VANHOUTTE, P.M. (1996a).
Inhibitors of the cytochrome P450-mono-oxygenase and en-
dothelium-dependent hyperpolarizations in the guinea pig
isolated cartoid artery. Br. J. Pharmacol., 117, 607 ± 610.
CORRIU, C., FELETOU, M., CANET, E. & VANHOUTTE, P.M. (1996b).
Endothelium-derived factors and hyperpolarization of the
carotid artery of the guinea pig. Br. J. Pharmacol., 119, 959 ± 964.
LUKSCH, A., POLAK, K., BEIER, C., POLSKA, E., WOLZT, M.,
DORNER, G.T., EICHLER, H.G. & SCHMETTERER, L. (2000).
E€ects of systemic NO synthase inhibition on choroidal and optic
nerve head blood ¯ow in healthy subjects. Invest. Ophthalmol.
Vis. Sci., 41, 3080 ± 3084.
DAVIES, P.F., OLESEN, S.P., CLAPHAM, D.E., MORREL, E.M.
&
SCHOEN, F.J. (1988). Endothelial communication. State of the art
lecture. Hypertension, 11, 563 ± 572.
DEMIREL, E., RUSCO, J., LASKEY, R.E., ADAMS, D.J. & VAN
BREEMEN. (1994). TEA inhibits ACh-induced EDRF release:
endothelial Ca2+-dependent K+ channels contribute to vascular
tone. Am. J. Physiol., 267, H1135 ± H1141.
MCCULLOCH, A.I., BOTTRILL, F.E., RANDALL, M.D. & HILEY, C.R.
(1997). Characterization and modulation of EDHF-mediated
relaxations in the rat isolated superior mesenteric arterial bed.
Br. J. Pharmacol., 120, 1431 ± 1438.
MCNEISH, A.J., MARTIN, W. & WILSON, W.S. (2001). Characterisa-
tion of endothelium-dependent vasodilator responses in the
bovine isolated perfused eye. Br. J. Pharmacol., 133, 70P.
MEYER, P., FLAMMER, J. & LUSCHER, T.F. (1993). Endothelium-
dependent regulation of the ophthalmic microcirculation in the
perfused porcine eye: role of nitric oxide and endothelins. Invest.
Ophthalmol. Vis. Sci., 34, 3614 ± 3621.
MONCADA, S., PALMER, R.M.J. & HIGGS, E.A. (1991). Nitric oxide:
physiology, pathophysiology, and pharmacology. Pharmacol.
Rev., 43, 109 ± 142.
MONCADA, S. & VANE, J.R. (1979). Pharmacology and endogenous
roles of prostaglandins, endoperoxides, thromboxane A2 and
prostacyclin. Pharmacol. Rev., 30, 293 ± 331.
MUGGE, A., ANTONIO, J., LOPEZ, G., PIEGORS, D.J., BREESE, K.R. &
HEISTAD, D.D. (1991). Acetylcholine-induced vasodilatation in
rabbit hindlimb in vivo is not inhibited by analogues of L-
arginine. Am. J. Physiol., 260, H242 ± H247.
NILSSON, S.F.E. (1996). Nitric oxide as a mediator of parasympa-
thetic vasodilation in ocular and extraocular tissues in the rabbit.
Invest. Ophthalmol. Vis. Sci., 37, 2110 ± 2119.
OHASHI, M., SATOH, K. & ITOH, T. (1999). Acetylcholine-induced
membrane potential changes in endothelial cells of rabbit aortic
valve. Br. J. Pharmacol., 126, 19 ± 26.
DERKACH, D.N., HIRAND, K., NISHIMURA, J., TAKAHASHI, S. &
KANAIDE, H. (2000). Thrombin causes endothelium-dependant
biphasic regulation of vascular tone in the porcine renal
interlobar artery. Br. J. Pharmacol., 131, 1635 ± 1642.
EDWARDS, G., DORA, K.A., GARDENER, M.J., GARLAND, C.J. &
WESTON, A.H. (1998). K+ is an endothelium-derived hyperpo-
larizing factor in rat arteries. Nature, 396, 269 ± 272.
EDWARDS, G., GARDENER, M.J., FELETOU, M., BRADY, G.,
VANHOUTTE, P.M. & WESTON, A.H. (1999). Further investiga-
tion of endothelium-derived hyperpolarizing factor (EDHF) in
rat hepatic artery: studies using 1-EBIO and ouabain. Br. J.
Pharmacol., 128, 1064 ± 1070.
EDWARDS, G., THOLLON, C., GARDENER, M.J., FELETOU, M.,
VILAINE, J.-P., VANHOUTTE, P.M. & WESTON, A.H. (2000). Role
of gap junctions and EETs in endothelium-dependent hyperpo-
larization of porcine coronary artery. Br. J. Pharmacol., 129,
1145 ± 1154.
FISSLTHALER, B., POPP, R., KISS, L., POTENTE, M., HARDER, D.R.,
FLEMING, I. & BUSSE, R. (1999). Cytochrome P450 2C is an
EDHF synthase in coronary arteries. Nature, 401, 493 ± 497.
FRANCHI-MICHELI, S., FAILLI, P., MAZZETTI, L., BANI, D., CIUFFI,
M. & ZILLETTI, L. (2000). Mechanical stretch reveals di€erent
components of endothelial-mediated vascular tone in rat aortic
strips. Br. J. Pharmacol., 131, 1355 ± 1362.
GIDDAY, J.M. & ZHU, Y. (1995). Nitric oxide does not mediate
autoregulation of retinal blood ¯ow in newborn pig. Am. J.
Physiol., 269, H1065 ± H1072.
HAEFLIGER, I.O., FLAMMER, J. & LUSCHER, T.F. (1993). Hetero-
geneity of endothelium-dependent regulation in ophthalmic and
ciliary arteries. Invest. Ophthalmol. Vis. Sci., 34, 1722 ± 1730.
HARDY, P., NUYT, A.M., ABRAN, D., ST-LOUIS, J., VARMA, D.R. &
CHEMTOB, S. (1996). Nitric oxide in retinal and choroidal blood
¯ow autoregulation in newborn pigs: interactions with prosta-
glandins. Pediatr. Res., 39, 487 ± 493.
QUIGNARD, J.-F., DUHAULT, J., THOLLON, C., VILAINE, J.-P.,
DUHAULT, J. & VANHOUTTE, P.M. (1999). Potassium ions and
endothelium-derived hyperpolarising factor in guinea-pig cartoid
and porcine coronary arteries. Br. J. Pharmacol., 127, 27 ± 34.
RANDALL, M.D., ALEXANDER, S.P.H., BENNETT, T., BOYD, A.E.,
FRY, J.R., GARDINER, S.M., KEMP, P.A., MCCULLOCH, A.I. &
KENDALL, D.A. (1996). An endogenous cannabinoid as an
endothelium-derived vasorelaxant. Biochem. Biophys. Res.
Com., 229, 114 ± 120.
RANDALL, M.D. & HILEY, C.R. (1988). Detergent and methylene
blue a€ect endothelium-dependent vasorelaxation and pressure/
¯ow relations in rat blood perfused mesenteric arterial bed. Br. J.
Pharmacol., 95, 1081 ± 1088.
HENRY, E., NEWBY, D.E., WEBB, D.J.
& O'BRIEN, C. (1999).
Peripheral endothelial dysfunction in normal pressure glaucoma.
Invest. Ophthalmol. Vis. Sci., 40, 1710 ± 1714.
&
HILL, C.E.
GOULD, D.J. (1995). Modulation of sympathetic
RUBANYI, G.M., ROMERO, J.C. & VANHOUTTE, P.M. (1986). Flow-
induced release of endothelium-derived relaxing factor. Am. J.
Physiol., 250, H1145 ± H1149.
vasoconstriction by sensory nerves and nitric oxide in rat irideal
arterioles. J. Pharmacol. Exp. Ther., 273, 918 ± 926.
HONING, M.L.H., SMITS, P., MORRISON, P.J. & RABELINK, T.J.
(2000). Bradykinin-induced vasodilation of human forearm
resistance vessels is primarily mediated by endothelium-depen-
dent hyperpolarization. Hypertension, 35, 1314 ± 1318.
HWA, J.J., GHIBAUDI, L., WILLIAMS, P. & CHATTERJEE, M. (1994).
Comparison of acetylcholine-induced relaxation in large and
small arteries of rat mesenteric vascular bed. Am. J. Physiol., 266,
H952 ± H958.
SCHMETTERER, L., FINDL, O., FASCHING, P., FERBER, W.,
STRENN, K., BRIETENEDER, H., ADAM, H., EICHLER, H.G. &
WOLZT, M. (1997). Nitric oxide and ocular blood ¯ow in patients
with IDDM. Diabetes, 46, 653 ± 658.
SELIGSOHN, E.E.
&
ANDERS, B. (2000). E€ects of NG-nitro-L-
arginine methyl ester on the cardiovascular system of anesthe-
tized rabbit and on the cardiovascular response to thyrotropin-
releasing hormone. Br. J. Pharmacol., 109, 1219 ± 1225.
British Journal of Pharmacology vol 134 (4)

920
A.J. McNeish et al
Vasodilatation in the bovine perfused eye
SU, E.-N., ALDER, V.A., YU, D.-Y. & CRINGLE, S.J. (1994). Adrenergic
and nitrergic neurotransmitters are released by the autonomic
system of the pig long posterior ciliary artery. Curr. Eye Res., 13,
907 ± 917.
ZHU, P., BENY, J.L., FLAMMER, J., LUSCHER, T.F. & HAEFLIGER,
I.O. (1997). Relaxation by bradykinin in porcine ciliary artery:
role of nitric oxide and K+ channels. Invest. Ophthalmol. Vis.
Sci., 38, 1761 ± 1767.
TODA, N., TODA, M., AYAJIKI, K.
&
OKAMURA, T. (1998).
ZIMA, P.A.V., HARHUN, M.I. & SHUBA, M.F. (2000). Properties of the
apamin-sensitive component of Ca2+-dependent K+ current in
smooth muscle cells of the guinea pig taenia coli. Neurophysiol-
ogy, 32, 63 ± 69.
ZYGMUNT, P.M. & HOGESTATT, E.D. (1996). Role of potassium
channels in endothelium-dependent relaxation resistant to
nitroarginine in the rat hepatic artery. Br. J. Pharmacol., 117,
1600 ± 1606.
Cholinergic nerve function in monkey ciliary arteries innervated
by nitroxidergic nerve. Am. J. Physiol., 274, H1582 ± H1589.
WILSON, W.S., SHAHIDULLAH, M. & MILLAR, C. (1993). The bovine
arterially perfused eye: an in vitro method for the study of drug
mechanisms on IOP, aqueous humour formation and uveal
vasculature. Curr. Eye Res., 12, 609 ± 620.
YAO, K., TSCHUDI, M., FLAMMER, J. & LUSCHER, T.F. (1991).
Endothelium-dependent regulation of vascular tone of the
porcine ciliary artery. Invest. Opthalmol. Vis. Sci., 32, 1791 ±
1798.
(Received April 23, 2001
Revised July 2, 2001
Accepted July 8, 2001)
ZAGVAZDIN, Y.S., FITZGERALD, M.E.C., SANCESARIO, G.
&
REINER, A. (1996). Neural nitric oxide mediates Edinger-
Westphal nucleus evoked increase in choroidal blood ¯ow in
the pigeon. Invest. Ophthalmol. Vis. Sci., 37, 666 ± 673.
British Journal of Pharmacology vol 134 (4)