Page 1

ABBREVIATIONS:

NO, nitric oxide; ODQ,

iH-[i

,2,4]oxadiazolo[4,3-ajquinoxalin-i

-one; aCSF, artificial

cerebrospinal

fluid; AMPA,

a-amino-3-

hydroxy-5-methyl-4-isoxazolepropionate;

EPSP, excitatory

postsynaptic

potential;

SNP, sodium

nitroprusside;

ANF, atrial natriuretic

factor;

NMDA, N-methyl-D-aspartate;

HEPES, 4-(2-hydroxyethyl)-i-piperazineethanesulfonic

acid.

184

0026-895X/95/020184-05$3.00/O

Copyright

© by The

American

Society

for Pharmacology

and Experimental

Therapeutics

All rights

of reproduction

in any

form

reserved.

MOLECULAR

PHAR.MACOWGY,

48:184-188

(1995).

ACCELERATED

COMMUNICATION

Potent and Selective Inhibition of Nitric Oxide-Sensitive

Guanylyl

Cyclase by 1H-[1 ,2,4]Oxadiazolo[4,3-a]quinoxalin-1

-one

JOHN

GARTHWAITE,

ERIC

SOUTHAM,

CAROLINE

L. BOULTON,

ERIK

B. NIELSEN,

KURT

SCHMIDT,

and

BERND

MAYER

Weilcome

Research

Laboratories,

Beckenham,

Kent BR3 3BS, UK (J.G., E.S., C.L.B.), Pharmaceuticals

Division,

Novo Nordisk

AIS, 2760

M#{226}l#{248}v,

Denmark

(E.B.N.), and Institut f#{252}rPharmakologie

und Toxikologie,

Karl-Franzens-Universitat,

Universitdt

Graz, A-8010

Gra.z, Austria

(KS., B.M.)

Received April 26, i 995; Accepted

May i 5, i 995

SUMMARY

In brain

and other tissues,

nitric

oxide

(NO) operates

as a

diffusible

second

messenger

that stimulates

the soluble

form of

the guanylyl

cyclase

enzyme

and

so elicits

an accumulation

of

cGMP

in target cells. Inhibitors

of NO synthesis

have been used

to implicate

NO in a wide spectrum

of physiological

and patho-

physiological

mechanisms

in the nervous

system

and else-

where.

The

function

of cGMP

in most tissues,

however,

has

remained

obscure.

We have now

identified

a compound,

i H-

[i ,2,4]oxadiazolo[4,3-a}quinoxalin-i

-one

(ODQ),

that potently

and selectively

inhibits

NO-stimulated

guanylyl

cyclase

activity.

In incubated

slices

of cerebellum,

ODQ

reversibly

inhibited

the

NO-dependent

cGMP

response

to glutamate

receptor

agonists

(IC50 20 nM)

but did not affect NO synthase

activity.

The

compound

did not affect synaptic

glutamate

receptor

function,

as assessed

in hippocampal

slices, nor did it chemically

mac-

tivate

NO. ODQ did, however,

potently

inhibit cGMP

generation

in response

to NO-donating

compounds.

An

action

on

NO-

stimulated

soluble

guanylyl

cyclase

was confirmed

in studies

with

the

purified

enzyme.

ODQ

failed

to inhibit

NO-mediated

macrophage

toxicity,

a phenomenon

that is unrelated

to cGMP,

nor

did it affect the activity

of particulate

guanylyl

cyclase

or

adenylyl

cyclase.

ODQ is the first inhibitor

that acts selectively

at the level of a physiological

NO “receptor”

and, as such, it is

likely to prove useful for investigating

the function

of the cGMP

pathway

in NO signal transduction.

The diffusible

molecule

NO has been

implicated

in a wide

range

of physiological

functions,

including

endothelium-

dependent

relaxation

of blood

vessels,

chemical

communica-

tion between

peripheral

nerves

and smooth

muscle,

and long

term

modifications

in the efficacy

of central

synapses

(1, 2).

In the

central

nervous

system,

NO

was

originally

identified

as the endogenous

intercellular

messenger

that

is generated

in a Ca2-dependent

manner

after

glutamate

(particularly

NMDA)

receptor

activation

and

that

elicits

cellular

cGMP

accumulation

in target

cells by stimulating

soluble

guanylyl

cyclase

(3). The

subsequent

function

of cGMP

in this

and

many

other

tissues

has remained

mysterious.

Moreover,

NO

This work was supported

by grants

from the European

Community,

Medical

Research

Council

(UK),

and the International

Human

Frontiers

Science

Pro-

gramme

Orgamsation

(J.G.)

and

by Grant

P 10098

from

the

Fonds

zur

For-

derung

der Wissenschaftlichen

Forschung

in Osterreich

(B.M.).

may

have

effects

that

are

unrelated

to soluble

guanylyl

cy-

clase

stimulation.

We have

now

identified

a unique

and

pow-

erful

tool with

which

to investigate

the function

of the NO-

cGMP

pathway,

namely

a potent

and

selective

inhibitor

of

NO-sensitive

guanylyl

cyclase.

Materials

and Methods

cGMP

levels

and

NO synthase

activity

in brain

slices.

Cer-

ebellar

slices

(400-.tm

thick

in the

sagittal

plane)

were

cut from

immature

(8-day-old)

or adult

(4-6-week-old)

Wistar

rats

(of either

gender)

(4). In some

experiments,

transverse

hippocampal

slices

(400-tm

thick)

from

adult

rats were

used

(5). The slices

were

incu-

bated

at 37#{176}in aCSF

containing

120 mM NaC1,

2 mM KC1, 2 mM

CaCl2, 26 mii NaHCO3,

1.19 mM MgSO4,

1.18 mist KH2PO4,

and

11

mM

glucose.

cGMP

responses

to the

glutamate

receptor

agonists

NMDA

and

AMPA

and

to NO

donor

drugs

were

determined

as

described

previously

(4, 5). When

tested,

the compound

ODQ

(syn-

(A)

120

L100

U,

80

a)

a-

60

C.,

It

40

20

0

(B)

120

100

.

80

l

60

It

40

20

0

1OOtMNMDA

+100nMODQ

-IIINOS

activity

-.-i

00)CM NMDA

--100tM

AMPA

-+--300)IM

SNP

(C)

250

__200

C

a)

0

0.

150

0)

E

.

100

a-

ta

0

o

io

i#{243}o

[ODQ] (nM)

control

+

1M

ODQ

-I.

.1

I

SIN-i

SNOG

SNAP

1mM

300tM

300tM

Inhibitor

of NO-Sensitive

Guanylyl

Cyclase

I 85

thesized

by Novo

Nordisk)

(see

Fig. 1A, inset,

for structure)

was

added

15-30

mm before

the effector

agent,

because

pilot experiments

demonstrated

that its effects

were

maximal

with this preincubation

period.

NO synthase

activity

was

measured

by monitoring

the ni-

troarginine

( 100

f.LM)-sensitive

conversion

of L-[ ‘4Clarginine

to

L-[’4C]citrulline

(6). Slices

were

incubated

for 30

mm

in aCSF

con-

taming

20 .tM

L-[14C]arginine

(150,000

dpm/ml)

and

1 mM L-citrul-

line, with or without

ODQ, and then

100 iM

NMDA

was added.

After

5 mm

the slices

were

withdrawn,

plunged

immediately

into boiling

50 mM

Tris/4

mM

EDTA

buffer,

pH

7.6, homogenized,

and,

after

removal

of L-[14Clarginine

by using

50% (by volume)

Dowex

50W ion

exchange

resin

(sodium

form), the amount

of L-[’4Cjcitrulline

in the

supernatant

was quantified

using

liquid

scintillation

counting.

The

same

slices

were

also assayed

for their

cGMP

and protein

contents.

Hippocampal

slice

electrophysiology.

Transverse

hippocam-

pal slices

(400-Mm

thick)

were

prepared

from

adult

(4-6-week-old)

rat brains

using

standard

techniques,

placed

in a submersion

cham-

ber held

at 24#{176},and perfused

with

oxygenated

aCSF.

Field

EPSPs,

evoked

at a frequency

of 0.033

Hz, were

recorded

from

stratum

radiatum

of CAl after electrical

stimulation

of the Schaffer

collater-

al-commissural

pathway

(7). Stable

EPSPs

(which,

under

these

con-

ditions,

are mediated

by AMPA

receptors)

were recorded

for at least

20 mm before

drug

application

and were quantified

by measurement

of their

slopes.

NMDA

receptor-mediated

EPSPs

were evoked

in the

presence

of 10

M

2,3-dihydroxy-6-nitro-7-sulfamoylbenzo(F)qui-

noxaline

and 50 .tM picrotoxin.

Soluble

guanylyl

cyclase

assay.

Purified

soluble

guanylyl

cy-

clase

(150

ng)

from

bovine

lung

(a kind

gift from

the late

Eycke

Bbhme,

Free

University

of Berlin,

Germany)

was incubated

at 37#{176}

for 10 mm

in 0.1 ml of 50 mM triethanolamine-HC1

buffer,

pH 7.4,

containing

10 .tM [a-32P]GTP

(200,000-300,000

cpm), 1 mM

2

mM glutathione,

and

1 mM unlabeled

cGMP.

Reactions

were

stopped

by ZnCO3

precipitation,

and [32P]cGMP

was isolated

(8).

Cyclic

nucleotides

in endothelial

cells.

Porcine

aortic

endo-

thelial

cells were

isolated

by enzymatic

treatment

with 0. 1% collag-

enase

and were cultured

in Petri dishes,

for up to three

passages,

in

Opti-MEM

medium

(Gibco-BRL)

containing

3% fetal calf serum

and

antibiotics.

Before

the experiments

cells

were

subcultured

in six-well

plastic

plates,

and

confluent

monolayers

(approximately

106

cells’

dish)

were

used

for cGMP/cAMP

measurements.

Endothelial

cells

were

washed

twice

with

isotonic

HEPES

buffer,

pH 7.4,

containing

2.5 nmt CaC12 and 1 mM MgC12 and were

preincubated

for 15 mm at

37#{176}in

1.4

ml

of the

same

buffer

containing

1 mM

isobutylmethyl-

xanthine

and 1 .tM indomethacin,

in the

absence

or presence

of ODQ.

The incubation

was started

by addition

ofO.1

ml ofa solution

of ANF

(human

A1.28),

S-nitrosoglutathione,

or forskolin

and

was

stopped

after 4 mm

by removal

of the incubation

medium

and addi-

tion of 1 ml of0.01

N HC1. Within

1 hr, intracellular

cGMP

and cAMP

were

completely

released

into the supernatant

and were

measured

by radioimmunoassay.

Macrophage

toxicity.

RAW

264.7

macrophages

were cultured

in

Petri dishes

in Opti-MEM

medium

(Gibco-BRL)

supplemented

with

3% fetal calf serum

and

antibiotics.

The cells were

subcultured

in

24-well

plates

and confluent

monolayers

were

incubated

at 37#{176}with

L-methylarginine

or

ODQ,

in the absence

or presence

of 2 pg/ml

bacterial

lipopolysaccharide

and

50 units/ml

interferon-y.

After

24

hr, the culture

media

were assayed

for lactate

dehydrogenase

activ-

ity (to quantifr

cell death)

and

nitrite

ion concentration

(to measure

NO formation)

(9).

NO autoxidation.

Autoxidation

of NO was determined

with

an

NO-sensitive

electrode

(World

Precision

Instruments,

Mauer,

Ger-

many)

as described

recently

(10). Briefly,

3 pJ ofa saturated

solution

of NO gas (-2

mM)

were

added

to 1.8 ml of air-saturated

triethano-

lamine-HC1

buffer

(50 mM; pH

7.0).

The

output

signal

was

recorded

at a sampling

rate

of 3 Hz.

Results

and Discussion

In cerebellar

slices,

ODQ

inhibited

the cGMP

response

to

the glutamate

receptor

agonist

NMDA,

with

an IC50 of about

20 nt

(Fig.

1A). This

effect

of ODQ

was

fully

reversible,

because

the

response

to NMDA

in slices

exposed

to 100

nM

ODQ

for 30 mm

and

then

maintained

for a 1-hr

washout

period

was

101

±

15%

of controls

(mean

±

standard

error,

four

experiments).

The

loss

of response

produced

by ODQ

was

unlikely

to be due

to NMDA

receptor

blockade

or to

inhibition

of NO

synthase,

because

the

cGMP-enhancing

ef-

cGMP

Fig. I .

ODQ potently inhibits NO-stimulated

cGMP accumulation,

in cerebellar slices, induced by exposure to NMDA, AMPA, or exogenous

NO

but does not affect NO synthase

activity.

A, ODQ (structure

shown

in inset)

concentration-dependently

inhibited

cGMP

responses

in immature

cerebellar

slices exposed

to NMDA

(2 mm), AMPA

(0.5 mm), or the NO donor SNP (5 mm). The amplitudes

of the control responses

were as follows:

NMDA,

i44

±

10 pmol/mg

of protein;

AMPA,

ii

±

i pmol/mg

of protein;

SNP, 72 ±

i pmol/mg

of protein.

Basal cGMP

levels were 0.7 ± 0.1

pmol/mg

of protein.

Each data point represents

the mean

± standard

error of four slices. B, Incubation

of immature

cerebellar

slices with ODQ

had no effect on NO synthase

(NOS)

activity

in slices

stimulated

for 5 mm with NMDA

(control

value,

17 ±

3 pmoVmg

of protein/mm;

five

experiments),

whereas,

in the same slices, it reduced

cGMP

responses

(control

value, i 7i ±

20 pmoVmg

of protein).

C, ODQ inhibited

cGMP

responses

to three

different

NO donors

(5-mm

exposure)

in adult cerebellar

slices

(four slices).

SIN-i,

3-morpholinosydnonymine;

SNOG,

S-nitrosoglutathione;

SNAP,

S-nitrosoacetylpenicillamine.

(A)

600

I-.--

3&M

I --100MSNP

(C)

.

400

E

x

E 300

a-

( 200

0

E

C

.

control

.

0.3tM

ODQ

.

C

E

x

E

x

a-

0

C.)

0

E

C

100

0

0

8

7

6

5

C

E

x

0)

E

>(

a-

0

C.)

0

E

>

[ODQ] (-log M))

[NO-donor]

(-log M)

i/[GTP]

(jiM1)

Fig. 2. ODQ inhibits

purified

soluble

guanylyl

cyclase. A, Concentration-response

curves for ODQ-mediated

inhibition

of soluble

guanylyl cyclase

activity

induced

by i 00 .tM

SNP

or 3 .tM S-nitrosoglutathione

(SNOG).

The data

points

are means

±

standard

errors

of three

separate

determinations.

B, Concentration-response

curves for S-nitrosoglutathione

and SNP in the absence

and presence

of 0.3 M

ODQ. The data points

are means

± standard

errors of three experiments.

C, Double-reciprocal

plot of soluble

guanylyl

cyclase

activity

in the presence

of i 00 pM

SNP

and i 0-500

,.tM GTP. The data points are mean values of four separate

determinations.

After weighting

of the data by a standard

procedure

(i 9),

Km

values

in the absence

and presence

of 0.3 tM

ODQ were calculated

to be 3i and 35 .tM,

respectively;

the

corresponding

Vm,

values

were

3.0 and i .3 p.mol/mg

of protein/mm.

7

6

5

4

3

0.00

0.02

0.04

186

Garthwaite

eta!.

1 B. Mayer,

unpublished

observations.

feet of the non-NMDA

receptor

agonist

AMPA

was

similarly

inhibited

(Fig.

1A), whereas

NMDA-stimulated

NO

synthase

activity

was

not

modified

(Fig.

1B). In confirmation,

at the

highest

concentration

tested

(10

,.tM),

ODQ

did not affect

the

activity

of purified

brain

NO

synthase.’

As

in

cerebellum,

cGMP

accumulation

in

hippocampal

slices

after

exposure

to NMDA

(300

M,

for 2 mm)

or AMPA

(100

p.M,

for 30 see)

was

nullified

by

1 .tM

ODQ

(data

not

shown);

specific

tests

for an

action

of ODQ

on synaptically

activated

glutamate

receptors

were

carried

out using

these

tissues.

After

30-40-mm

exposure

to ODQ

at the

highest

concentration

examined

(10 p.M),

AMPA

and NMDA

receptor-

mediated

synaptic

potentials

were

not significantly

changed.

The values

of the initial

slopes

were

104 ± 6%

and

107

± 5%

of controls,

respectively

(values

measured

over

a 10-mm

pe-

riod

immediately

before

the ODQ

application;

means

± stan-

dard

errors,

four

to six experiments).

These

findings

indicate

that

ODQ

acts

downstream

from

the glutamate

receptor-NO

synthase

pathway.

In support

of

this

conclusion,

ODQ

was

able

to inhibit

cGMP

accumulation

in cerebellar

slices

exposed

to the

NO

donor

SNP

(300

.tM),

with

a potency

similar

to that

shown

against

NMDA

and

AMPA

(Fig.

1A). Responses

to other

NO

donors

were

also

sensitive

to the compound

(Fig.

1C), suggesting

either

that

it

directly

inactivates

NO

or that

it inhibits

stimulation

of

guanylyl

cyclase

by NO.

The

first

possibility

was

eliminated

by direct

tests

using

a

NO

electrode.

In accordance

with

previous

reports

(11),

the

third-order

rate

constant

for NO

autoxidation

was

found

to

be about

9 x 106 M2

X sec’.

ODQ

(0.3

nmt final

concentra-

tion)

affected

neither

the

amplitude

of the

response

of the

electrode

to NO

nor

the

autoxidation

kinetics

of NO

(two

experiments).

In addition,

at concentrations

up

to 10

.LM,

ODQ

did not generate

superoxide

anions

(which

rapidly

react

with

NO) either

in the presence

or in the absence

ofthiols,

as

determined

by superoxide

dismutase-inhibitable

reduction

of

cytochrome

c.’

Experiments

to directly

test

the

effect

of ODQ

on soluble

guanylyl

cyclase

were

performed

with

purified

enzyme.

ODQ

inhibited

the

enzyme

activity

evoked

by matched

concentra-

tions

of two

different

NO

donors

(SNP

and

S-nitrosogluta-

thione),

with

identical

inhibition

curves

(Fig.

2A).

Dose-

response

curves

for both

NO

donors

were

shifted

to the right

in the

presence

of 0.3

j.M

ODQ,

with,

at least

in the

case

of

SNP,

a clear

reduction

in

the

maximum

effect

(Fig.

2B).

Further

studies

are

needed

to more

accurately

classify

the

type

of inhibition

with

respect

to NO. Kinetic

analysis

sug-

gested

that

the inhibition

was

noncompetitive

with

respect

to

the

substrate

GTP

(Fig.

2C).

To

determine

whether

ODQ

acts

selectively

on

soluble

guanylyl

cyclase

or whether

it can

also

inhibit

the

mem-

brane-associated,

or particulate,

type

ofguanylyl

cyclase

that

transduces

natriuretic

peptide

receptor

activation,

its effects

on

ANF-induced

cGMP

accumulation

in

endothelial

cells

were

explored.

ODQ

reduced

basal

cGMP

levels

with

a po-

tency

in the

nanomolar

range,

similarly

to that

observed

against

the

NO

donor

S-nitrosoglutathione

(Fig.

3A).

How-

ever,

the

increase

in cGMP

levels

induced

by 100

nvt ANF

persisted

in the

presence

of ODQ

at concentrations

of up to

100

p.M

(Fig.

3B). cGMP

responses

to lower

ANF

concentra-

tions

were,

likewise,

insensitive

to ODQ

(Fig.

3B). Adenylyl

cyclase

activity,

assayed

by measuring

cAMP

accumulation

induced

by forskolin,

was

unaffected

by high

ODQ

concen-

trations

(Fig.

3C).

Evidence

that

ODQ

does

not block

actions

of NO

that

are

(B)

5

(A)

20

15

U)

a)

U

0

; 10

0

E

0.

a-

0

C.)

-U-

control

-.-

10&M ODQ]

I 87

4

U)

a)

C.)

, 3-

0

E

0.

Inhibitor

of NO-Sensitive

Guanylyl

Cyclase

(C)

20

I -U-

1 O0M forskolin

I

Or-

0

4-.

5

0

0

9

8

7

6

5

4

0J r-/I----

0

9

e

7

[ODQ]

(-log M)

[ANfl

(-log M)

Fig. 3. ODQ does not inhibit

ANF-stimulated

particulate

guanylyl

cyclase

or adenylyl

cyclase.

A, In endothelial

cells,

increasing

concentrations

of ODQ

reduced

basal cGMP

levels and cGMP

accumulation

induced

by 100 .tM S-nitrosoglutathione

(SNOG),

without

affecting

the cGMP

response

to i 00 nM ANF. B, Dose-cGMP

response

curves for ANF in the absence

and presence

of 10 M

ODQ are shown. C, In a wide range of

concentrations,

ODQ failed to affect cAMP

accumulation

induced

by 100 .tM forskolin.

The data points

are means

± standard

errors

of three

to

five experiments.

6

5

4

[ODQ] (-log M)

unrelated

to guanylyl

cyclase

activation

was

obtained

by

examining

the

cytotoxicity

of activated

macrophages.

Expo-

sure

of macrophages

to bacterial

lipopolysaccharide

and/or

cytokines

results

in the

time-dependent

expression

of the

inducible

isoform

of NO

synthase,

and

the

resulting

NO

ultimately

causes

cell death

through

mechanisms

that,

al-

though

not yet clearly

defined,

include

inhibition

of metabolic

enzymes

(1). In our experiments,

ODQ

failed

to influence

the

cytotoxicity

of

activated

macrophages

under

conditions

where

the NO

synthase

inhibitor

L-methylarginine

was effec-

tive

(Table

1). Furthermore,

in

the

same

experiments,

L-

methylarginine

simultaneously

inhibited

NO

formation,

whereas

ODQ

did

not,

suggesting

that

the

inducible

NO

TABLE

i

Effects

of L-methylarginine

and ODQ on NO-induced

macrophage

cytotoxicity

Values

are means

± standard

errors (three experiments).

Lactate dehydrogenase

release

Nitrite

concentration

milliunits/mi

p.M

Untreated

macrophages

Control

65 ± 5

i.4 ± 0.6

L-Methylargmnmne,

iOO .tM

7i ± 4

0.9 ± 0.1

ODQ

i0.tM

67±4

2.6±0.2

100 .tM

71 ± 5

2.7 ± 0.2

Induced

macrophages

Control

229 ± 1 7

24 ± 2

L-Methylarginine

i0j.tM

2ii±30

24±4

iOOj.tM

84±3

17±3

1mM

51±4

11±1

ODQ

ip.M

247±iO

24±2

i0iM

249±8

24±3

iOOp.M

302±1

27±i

synthase,

like

the

neuronal

Ca2-dependent

isoform,

is in-

sensitive

to ODQ.

These

results

demonstrate

that

ODQ

is a potent

inhibitor

of NO-stimulated

soluble

guanylyl

cyclase

activity,

without

actions

on particulate

guanylyl

cyclase

or on

adenylyl

cy-

clase.

In

brain

slices

stimulated

with

glutamate

receptor

agonists,

ODQ

does

not appear

to interfere

with

any

of the

steps

leading

to NO

synthesis;

moreover,

it does

not inhibit

constitutive

or inducible

NO

synthase

activity

and

it does

not

inactivate

NO.

Two

other

compounds

are

often

referred

to, and

used,

as

guanylyl

cyclase

inhibitors,

namely

LY-83583

and

methylene

blue.

In reality,

LY-83583

was

originally

reported

to stimu-

late

guanylyl

cyclase

activity

but,

paradoxically,

to lower

tissue

cGMP

levels

nonspecifically

and

by an unknown

mech-

anism

(12). Although

under

certain

assay

conditions

guany-

lyl cyclase

inhibition

has

been

observed

(and

attributed

to a

metabolite),

LY-83583

more

potently

inhibits

NO

release

(13)

and

has

been

shown

to have

several

other

effects

(14-16)

that,

collectively,

make

any

results

obtained

with

this

chem-

ical very

difficult

to interpret.

Methylene

blue

is actually

a

weak

guanylyl

cyclase

inhibitor

that

more

effectively

gener-

ates

superoxide

anions

and

inhibits

NO

synthase

(17).

ODQ

thus

represents

the first

inhibitor

acting

on the NO

“receptor,”

soluble

guanylyl

cyclase,

although

further

work

will

be required

to define

more

precisely

its mechanism

of

action.

In the

brain,

both

the

NO

synthetic

machinery

and

NO-stimulated

cGMP

accumulation

display

a widespread

and

correlated

distribution,

suggesting

a general

functional

partnership

between

the

two

(18).

The

function

of cGMP,

which

has been

known

for a quarter

ofa century

to be present

in the

brain,

has

remained

elusive.

ODQ

represents

a pow-

erful

and

unique

tool that

we anticipate

will play

a major

role

in further

elucidating

the

physiological

significance

of the

NO-cGMP

pathway,

not only in the nervous

system

but also

188

Garthwaite

et a!.

in

many

other

organs

in

which

this

signaling

system

is

present.

Acknowledgments

The technical

assistance

ofEva

Leopold

and Margit

Rehn

(both

at

Universit#{228}t

Graz)

is gratefully

acknowledged.

References

1. Moncada,

S., R. M. J. Palmer,

and

E. A. Higgs.

Nitric

oxide:

physiology,

pathophysiology,

and

pharmacology.

Pharmacol.

Rev.

43:109-142

(1991).

2. Garthwaite,

J. Glutamate,

nitric

oxide

and cell-cell

signalling

in the ner-

vous

system.

Trends

Neurosci.

14:60-67

(1991).

3. Garthwaite,

J., S. L. Charles,

and

R. Chess-Williams.

Endothelium-

derived

relaxing

factor

release

on activation

of NMDA

receptors

suggests

role as intercellular

messenger

in the

brain.

Nature

(Lond.)

336:385-388

(1988).

4. Southam,

E., S. J. East,

and J. Garthwaite.

Excitatory

amino

acid recep-

tars

coupled

to the

nitric

oxide:cyclic

GMP

pathway

in rat

cerebellum

during development.

J. Neurochem.

56:2072-2081

(1991).

5. East, S. J., and J. Garthwaite.

NMDA

receptor

activation

in rat hippocam-

pus

induces

cyclic

GMP

formation

through

the

L-arginine-nitric

oxide

pathway.

Neurosci.

Lett.

123:17-19

(1991).

6. Bredt,

D. S., and

S. H. Snyder.

Nitric

oxide

mediates

glutamate-linked

enhancement

ofcGMP

levels

in the

cerebellum.

Proc.

Natl.

Acad.

Sci. USA

86:9030-9033

(1989).

7. Boulton,

C. L., A. J. Irving,

E. Southam,

B. Potier,

J. Garthwaite,

and

G.

Collingridge.

The

nitric

oxide-cyclic

GMP

pathway

and

synaptic

depres-

sion

in rat

hippocampal

slices.

Eur.

J. Neurosci.

6:1528-1535

(1994).

8. Schultz,

G., and

E. Bohme.

Guanylate

cyclase:

GTP

pyrophosphate-lyase

(cyclizing),

EC 4.6.1.2.,

in Methods

of Enzymatic

Analysis

(H. U. Berg-

meyer,

J. Bergmeyer,

and

M. Graill,

eds.).

Verlag

Chemie,

Weinheim,

Germany,

379-389

(1984).

9. Green,

L. C., D. A. Wagner,

J. Glogowski,

P. L. Skipper,

J. S. Wishnok,

and

S. R. Tannenbaum.

Analysis

of nitrate,

nitrite

and

[‘5Njnitrate

in biolog-

ical fluids.

Anal.

Biochem.

126:131-138

(1982).

10. Mayer,

B., P. Klatt,

E. R. Werner,

and

K. Schmidt.

Kinetics

and

mecha-

nism

of tetrahydrobiopterin-induced

oxidation

of nitric

oxide.

J. Biol.

C/rem.

270:655-659

(1995).

11. Ford,

P. C., D. A. Wink,

and

D. M. Stanbury.

Autoxidation

kinetics

of

aqueous

nitric

oxide.

FEBS

Lett.

326:1-3

(1993).

12. Schmidt,

M. J., B. D. Sawyer,

L. L. Truex,

W. S. Marshall,

and

J. H.

Fleisch.

LY83583:

an

agent

that

lowers

intracellular

levels

of cyclic

guanosine

3’,S’-monophosphate.

J. Pharmacol.

Exp.

Ther.

232:764-769

(1985).

13. Mulsch,

A., R. Busse,

S. Liebau,

and U. Forstermann.

LY83583

interferes

with

the

release

of endothelium-derived

relaxing

factor

and

inhibits

solu-

ble guanylate

cyclase.

J. Pharmacol.

Exp.

Ther.

247:283-288

(1988).

14. Barbier,

A. J. M., and

R. A. Lefebvre.

Effect

of LY83583

on

relaxation

induced

by non-adrenergic

non-cholinergic

nerve

stimulation

and

exoge-

nous

nitric

oxide

in the

rat

gastric

fundus.

Eur.

J. Pharmacol.

219:331-

334

(1992).

15. Kontos,

H. A., and

E. P. Wei.

Hydroxyl

radical-dependent

inactivation

of

guanylate

cyclase

in

cerebral

arterioles

by

methylene

blue

and

by

LY83583.

Stroke

24:427-434

(1993).

16. Luond,

R. M., J. H. McKie,

and

K. T. Douglas.

A direct

link

between

LY83583,

a selective

repressor

of cyclic

GMP

formation,

and

glutathione

metabolism.

Biochem.

Pharmacol.

45:2547-2549

(1993).

17. Mayer,

B., F. Brunner,

and

K Schmidt.

Inhibition

ofnitric

oxide

synthesis

by methylene

blue.

Biochem.

Pharmacol.

45:367-374

(1993).

18. Southam,

E., and

J. Garthwaite.

The

nitric

oxide-cyclic

GMP

signalling

pathway

in rat brain.

Neuropharmacology

32:1267-1277

(1993).

19. DeLevie,

R. When,

why

and

how

to use

weighted

least

squares.

J. C/rem.

Educ.

63:10-15

(1986).

Send

reprint

requests

to: John

Garthwaite,

Weilcome

Research

Laborato-

ries,

Langley

Court,

South

Eden

Park

Road,

Beckenham,

Kent

BR3

3BS, UK