Use of 2-[125I]iodomelatonin to Characterize Melatonin Binding Sites in Chicken Retina

Abstract

2-[125I]Iodomelatonin binds with high affinity to a site possessing the pharmacological characteristics of a melatonin receptor in chicken retinal membranes. The specific binding of 2-[125I]iodomelatonin is stable, saturable, and reversible. Saturation experiments indicated that 2-[125I]iodomelatonin labeled a single class of sites with an affinity constant (Kd) of 434 +/- 56 pM and a total number of binding sites (Bmax) of 74.0 +/- 13.6 fmol/mg of protein. The affinity constant obtained from kinetic analysis was in close agreement with that obtained in saturation experiments. Competition experiments showed a monophasic reduction of 2-[125I]iodomelatonin binding with a pharmacological order of indole amine affinities characteristic of a melatonin receptor: 2-iodomelatonin greater than 6-chloromelatonin greater than or equal to melatonin greater than or equal to 6,7-dichloro-2-methylmelatonin greater than 6-hydroxymelatonin greater than or equal to 6-methoxymelatonin much greater than N-acetyltryptamine greater than N-acetyl-5-hydroxytryptamine greater than 5-methoxytryptamine greater than 5-hydroxytryptamine (inactive). The affinities of these melatonin analogs in competing for 2-[125I]iodomelatonin binding sites were correlated closely with their potencies for inhibition of the calcium-dependent release of [3H]dopamine from chicken and rabbit retinas, indicating association of the binding site with a functional response regulated by melatonin. The results indicate that 2-[125I]iodomelatonin is a selective, high-affinity radioligand for the identification and characterization of melatonin receptor sites.

Full text

Proc.
Natl.
Acad.
Sci.
USA
Vol.
84,
pp.
3916-3920,
June
1987
Neurobiology
Use
of
2-[1251]iodomelatonin
to
characterize
melatonin
binding
sites
in
chicken
retina
(receptor/radioligand/N-acetyltryptamine/indole
ethylamine)
MARGARITA
L.
DUBOCOVICH*t
AND
JOSEPH
S.
TAKAHASHIt
*Department
of
Pharmacology,
Northwestern
University
Medical
School,
303
East
Chicago
Avenue,
Chicago,
IL
60611;
and
tDepartment
of
Neurobiology
and
Physiology,
Northwestern
University,
Hogan
Hall,
Evanston,
IL
60201
Communicated
by
Colin
S.
Pittendrigh,
February
12,
1987
ABSTRACT
2-['25I]Iodomelatonin
binds
with high
affinity
to
a
site
possessing
the
pharmacological
characteristics
of
a
melatonin
receptor
in
chicken
retinal
membranes.
The
specific
binding
of
2-[125I]iodomelatonin
is
stable,
saturable,
and
re-
versible.
Saturation
experiments
indicated
that
2-[125I]iodo-
melatonin
labeled
a
single
class
of
sites
with
an
affinity
constant
(Kd)
of
434
±
56
pM
and
a
total
number
of
binding
sites
(B.,)
of
74.0
±
13.6
fmol/mg
of
protein.
The
affinity
constant
obtained
from
kinetic
analysis
was
in
close
agreement
with
that
obtained
in
saturation
experiments.
Competition
experiments
showed
a
monophasic
reduction
of
2-[1251]iodomelatonin
bind-
ing
with
a
pharmacological
order
of
indole
amine
affinities
characteristic
of
a
melatonin
receptor:
2-iodomelatonin
>
6-chloromelatonin
:
melatonin
,
6,7-dichloro-2-methylmela-
tonin
>
6-hydroxymelatonin
>
6-methoxymelatonin
>
N-
acetyltryptamine
>
N-acetyl-5-hydroxytryptamine
>
5-meth-
oxytryptamine
>>>
5-hydroxytryptamine
(inactive).
The
af-
fmities
of
these
melatonin
analogs
in
competing
for
2-[1251]iodo-
melatonin
binding
sites
were
correlated
closely
with
their
potencies
for
inhibition
of
the
calcium-dependent
release
of
[3H]dopamine
from
chicken
and
rabbit
retinas,
indicating
association
of
the
binding
site
with
a
functional
response
regulated
by
melatonin.
The
results
indicate
that
2-[1251]iodo-
melatonin
is
a
selective,
high-affinity
radioligand
for
the
identification
and
characterization
of
melatonin
receptor
sites.
A
body
of
literature
suggests
that
the
hormone
melatonin
(N-acetyl-5-methoxytryptamine)
regulates
a
number
of
phys-
iological
processes
in
vertebrates.
These
include
the
regula-
tion
of
reproduction
in
photoperiodic
mammals
(1, 2),
the
control
of
circadian
rhythms
in
birds
and
reptiles
(3-5),
and
the
modulation
of
retinal
physiology.
Melatonin
has
been
found
in
the
retina
of
several
species
(6-10).
Retinal
melatonin
has
been
implicated
in
photoreceptor
outer
seg-
ment
disc
shedding
and
phagocytosis
(11,
12),
melanosome
aggregation
in
pigment
epithelium
(13),
and
cone
photorecep-
tor
retinomotor
movement
(14).
Because
so
many
important
biological
processes
are
regulated
by
melatonin,
the
mecha-
nism
by
which
this
hormone
acts
is
of
great
interest.
Under-
standing
of
this
mechanism
has
been
hampered
by
several
technical
difficulties,
perhaps
the
most
important
of
which
is
the
lack
of
a
soundly
based
pharmacology.
Dubocovich
(15-17)
has
shown
that
picomolar
concentrations
of
melatonin
selectively
inhibit
the
calcium-dependent
release
of
dopamine
from
rabbit
and
chicken
retina
through
activa-
tion
of
a
site
possessing
the
pharmacological
and
functional
characteristics
of
a
receptor.
Although
binding
sites
for
[3H]melatonin
have
been
reported
in
cytosolic
and
membrane
fractions
of
several
central
nervous
system
tissues
including
the
retina
of
lower
vertebrates,
these
sites
have
not
been
associated
with
biological
effects
of
melatonin
and
related
indoles
(10,
18-21).
Further,
the
relatively
low
specific
activity
of
[3H]melatonin,
compared
to
that
of
radioiodinated
ligands,
may
have
hindered
detection
of
high-affinity
binding
sites
in
tissues
with
a
low
density
of
receptors
(10).
To
enhance
the
ability
to
detect
melatonin
receptor
sites
in
vertebrate
retina,
we
have
utilized
the
radioiodinated
ligand
2-['25I]iodomelatonin
(22).
Here
we
report
the
characteristics
of
binding
of
2-[125I]iodomelatonin
to
chicken
retinal
mem-
branes
and
demonstrate
a
pharmacological
correlation
be-
tween
the
binding
site
labeled
by
2-['25I]iodomelatonin
and
a
functional
response
regulated
by
melatonin
in
the
chicken
and
rabbit
retina.
MATERIALS
AND
METHODS
Materials.
2-['25I]Iodomelatonin
was
synthesized
by
a
modification
(J.S.T.,
S.
S.
Nikaido,
and
M.L.D.,
unpub-
lished)
of
the
method
of
Vakkuri
et
al.
(22).
The
specific
activity
of
the
radioligand
was
1800-2175
Ci/mmol
(1
Ci
=
37
GBq)
and
was
stable
for
60
days.
Purity
of
the
radioligand
was
checked
by
TLC
and
was
>95%.
2-Iodomelatonin
was
synthesized
by
the
method
of
Vakkuri
et
al.
(23)
and
purified
by
silica
gel
chromatography.
The
purity
of
this
compound
was
shown
to
be
greater
than
98%
by
HPLC.
6-Chloromel-
atonin,
6,7-dichloro-2-methylmelatonin,
and
N-acetyltrypt-
amine
were
donated
by
J.
Clemens
(Eli
Lilly);
N-acetyl-5-
methoxykynurenamine
by
D.
E.
Clark
(College
of
Pharma-
cy,
University
of
Houston,
TX);
and
6-methoxymelatonin
by
D.
C.
Klein
(Laboratory
of
Developmental
Endocrinology,
National
Institutes
of
Health,
Bethesda,
MD).
Other
drugs
were
obtained
from
commercial
sources
or
the
pharmaceu-
tical
company
of
origin.
Membrane
Preparation.
Chickens
(4-6
weeks
old)
main-
tained
in
a
controlled
lighting
regime
(14
hr
light/10
hr
dark)
were
decapitated
during
the
light
phase.
Retinas
were
dis-
sected
free
of
pigment
epithelium
and
homogenized
in
ice-
cold
50
mM
Tris
HCl
buffer,
pH
7.5
(250C)/0.1%
ascorbic
acid
with
a
Brinkmann
Polytron
PT-S
at
setting
5
for
10
sec.
The
homogenate
was
centrifuged
at
50,000
x
g
for
10
min
at
40C.
The
pellet
was
washed
once
by
resuspension
and
centrifugation
in
Tris-HCl
buffer.
The
retinal
membrane
pellet
was
resuspended
by
homogenization
at
a
concentration
of
500
,ug
of
protein
per
ml.
Where
indicated,
retinal
homogenates
were
fractionated
by
differential
centrifugation
at
40C
to
yield
a
crude
nuclear
pellet
(P1,
1200
x
g
for
10
min),
a
crude
mitochondrial
pellet
(P2,
27,000
x
g
for
20
min),
and
a
crude
microsomal
pellet
(P3,
100,000
x
g
for
60
min).
The
pellets
were
resuspended
by
homogenization
in
ice-cold
50
mM
Tris-HCl
buffer,
pH
7.5/0.1%
ascorbic
acid
and
used
in
binding
assays.
Binding
Assays.
For
binding
assays,
2-[1251]iodomelatonin
was
diluted
in
Tris
HCl
buffer
with
0.01%
bovine
serum
albumin,
and
drugs
were
dissolved
in
1
mM
HCl
with
0.1%
tTo
whom
reprint
requests
should
be
addressed.
3916
The
publication
costs
of
this
article
were
defrayed
in
part
by
page
charge
payment.
This
article
must
therefore
be
hereby
marked
"advertisement"
in
accordance
with
18
U.S.C.
§1734
solely
to
indicate
this
fact.

Proc.
Natl.
Acad.
Sci.
USA
84
(1987)
3917
bovine
serum
albumin.
Binding
was
initiated
by
addition
of
220-Al
aliquots
of
membranes
resuspended
in
the
Tris
HCl
buffer
to
tubes
containing
20
A.l
of
appropriate
2-[l251]iodo-
melatonin
concentrations
and
20
1.l
of
drugs
or
vehicle.
Unless
otherwise
indicated
the
binding
of
2-['251]iodomela-
tonin
was
routinely
measured
in
duplicate
after
incubation
at
0C
for
5
hr
in
the
dark.
Reactions
were
terminated
by
addition
of
5
ml
of
ice-cold
Tris
HCl
buffer,
and
the
contents
were
immediately
filtered
through
glass-fiber
filters
(Schleicher
&
Schuell
no.
30)
soaked
in
0.5%
(vol/vol)
polyethylenimine
solution.
Each
filter
was
washed
twice
with
5
ml
of
the
cold
buffer.
Radioactivity
was
determined
in
a
gamma
counter.
Nonspecific
binding,
unless
otherwise
indi-
cated,
was
defined
as
binding
in
the
presence
of
3
AM
6-chloromelatonin.
Specific
binding
of
2-[125I]iodomelatonin
was
calculated
by
subtracting
nonspecific
binding
from
total
binding
and
expressed
as
fmol/mg
of
protein.
In
a
typical
experiment,
total
2-[125I]iodomelatonin
(67
pM)
binding
was
2542
±
361
cpm
(n
=
3),
and
the
nonspecific
binding
defined
with
3
AM
6-chloromelatonin
was
350
±
35
cpm
(n
=
3).
The
total
radioactivity
bound
to
filters
represented
166
±
10
cpm
(n
=
3).
Identity
of
Bound
Ligand.
Immediately
following
a
filtra-
tion
binding
assay,
filters
were
removed
and
placed
in
1
ml
of
methanol.
The
methanol
extract
was
evaporated
under
ni-
trogen
to
a
volume
of
100
,Al.
Melatonin
(10
,4g)
was
added
as
a
carrier
and
50
jul
was
applied
to
a
silica
gel
TLC
plate
(Kodak
13179).
Chromatograms
were
developed
with
ethyl
acetate.
The
TLC
plate
was
cut
into
0.5-cm
segments
and
radioactivity
was
quantified.
The
peak
of
radioactivity
was
compared
with
the
position
of
authentic
2-[125I]iodomelato-
nin.
Calculations.
Kinetic
data
were
analyzed
by
the
method
of
Bennett
and
Yamamura
(24),
using
pseudo-first-order
con-
ditions
to
estimate
the
association
rate
constant
(kj)
and
addition
of
a
1000-fold
molar
excess
of
unlabeled
ligand
to
estimate
the
dissociation
rate
constant
(kL).
Data
for
satu-
ration
and
competition
experiments
were
analyzed
by
use
of
the
EBDA/LIGAND
program
(25).
Ki
values
were
calculated
from
IC50
values
by
the
method
of
Cheng
and
Prusoff
(26).
[3H]Dopamine-Release
Experiments.
Retinas
from
chickens
maintained
on
14
hr
light/10
hr
dark
cycle
were
dissected
free
of
pigment
epithelium
and
prepared
for
superfusion
experi-
ments
as
described
(17,
27).
In
brief,
pieces
of
chick
retina
were
labeled
in
vitro
with
[3H]dopamine
(specific
activity
28.4
Ci/mmol)
in
Krebs'
solution
with
1.3
mM
CaCl2
and
then
were
superfused
at
a
rate
of
1
ml/min
until
the
spontaneous
outflow
of
radioactivity
leveled
off.
Tritium
release
was
elicited
twice
in
each
experiment
by
electrical
stimulation
at
3
Hz
for
2
min
(20
mA,
2-msec
duration).
The
first
(Sl)
and
the
second
(S2)
periods
of
stimulation
were
applied
60
min
and
100
min
after
the
end
of
the
incubation
with
[3H]dopa-
mine.
Results
were
calculated
as
the
percentage
of
the
total
tissue
radioactivity
released
in
each
sample
(17).
RESULTS
Characterization
of
2-[125I]Iodomelatonin
Binding.
Binding
of
2-[125I]iodomelatonin
was
determined
in
washed
chicken
retinal
membranes
in
50
mM
Tris
HCl
buffer
(pH
7.5)
containing
0.1%
ascorbic
acid,
following
5
hr
of
incubation
at
0°C.
The
highest
concentration
of
2-[125I]iodomelatonin
bind-
ing
sites
was
found
in
membranes
from
nuclear
(43%)
and
mitochondrial
(47%)
subcellular
fractions.
In
subcellular
fractions,
specific
binding
defined
with
3
,uM
6-chloromela-
tonin
was
85-90%
of
total
2-[125I]iodomelatonin
binding.
Because
we
could
not
detect
any
differences
in
the
binding
of
2-[1251]iodomelatonin
to
the
crude
mitochondrial
pellet
(P2)
and
the
total
particulate
fraction,
we
used
the
latter
for
routine
binding
studies.
The
binding
of
2-[125I]iodomelatonin
(30
pM)
to
chicken
retinal
membranes
was
linear
between
25
and
150
pug
of
protein.
Incubation
of
retinal
membranes
for
5
min
in
a
boiling
water
bath
prior
to
the
binding
assay
completely
abolished
specific
binding;
nonspecific
binding
increased
from
350
±
35
cpm
to
553
±
32
cpm
(n
=
3).
Extraction
of
the
bound
radioligand
after
a
binding
assay
showed
that
the
bound
radioligand
comigrated
on
TLC
with
authentic
2-[125I]iodomelatonin;
no
other
radiolabeled
bands
were
observed.
Thus,
2-[125I]iodomelatonin
is
stable
during
the
course
of
a
standard
binding
assay
at
00C.
Specific
2-[1251]iodomelatonin
binding
to
washed
chicken
retinal
mem-
branes
was
not
affected
by
MgCl2
(0.1-10
mM),
KCl
(5-200
mM),
or
CaCl2
(0.1-10
mM).
However,
NaCl
(100-200
mM)
inhibited
binding
in
a
concentration-dependent
manner
[63
+
4.3%
inhibition
(n
=
3)
for
200
mM
NaCl].
The
binding
of
2-[1251]iodomelatonin
to
chicken
retinal
membranes
was
reversible
and
temperature-dependent.
At
0C
the
binding
of
2-[125I]iodomelatonin
(40
pM)
reached
a
steady
state
by
3
hr
and
was
stable
for
10
hr
(Fig.
1).
The
association
rate
constant
(kj)
determined
from
the
pseudo-
first-order
equation
was
5.5
x
107
M-1
min'.
Specific
2-[125I]iodomelatonin
binding
was
reversed
(t1/2
=
36.6
min)
by
the
addition
of
excess
competing
ligand
(3
/iM
6-
chloromelatonin).
The
rate
constant
for
dissociation
(kL1)
was
0.0188
min'.
The
kinetic
dissociation
constant
(Kd)
for
2-[1251]iodomelatonin
calculated
from
the
ratio
kL1/k,
was
342
pM.
The
total
binding
of
2-[1251]iodomelatonin
at
equilibrium
was
similar
at
0°,
25°,
and
370C;
however,
a
steady
state
was
reached
at
30
min
and
at
9
min
in
binding
assays
at
25TC
and
370C,
respectively.
Concentration-dependent
binding
of
2-[125I]iodomelatonin
(0.05-1.5
nM)
to
chicken
retinal
membranes
was
saturable
and
resulted
in
linear
Scatchard
plots
suggesting
binding
to
a
single
class
of
sites
(Fig.
2).
The
apparent
Kd
for
2-
[1251]iodomelatonin
from
the
Scatchard
analysis
was
434
±
56
pM
(n
=
5)
and
the
total
number
of
binding
sites
was
74.0
+
13.6
fmol/mg
of
protein
(n
=
5).
The
Kd
values
determined
in
kinetic
(342
pM)
and
saturation
(434
pM)
experiments
were
in
close
agreement.
The
pharmacological
characterization
of
2-[125I]iodomela-
tonin
binding
to
chicken
retinal
membranes
was
carried
out
with
tracer
concentrations
of
radioligand
(30-60
pM).
N-
Acetyltryptamines,
which
are
potent
inhibitors
of
calcium-
dependent
release
of
[3H]dopamine
from
rabbit
retina
(17),
were
most
effective
in
competing
for
2-[125I]iodomelatonin
binding
sites
(Table
1
and
Fig.
3).
All
four
N-acetyltrypt-
amines
shown
in
Fig.
3
inhibited
the
binding
of
2-[125I]iodo-
melatonin
to
the
same
extent.
N-Acetyl-5-methoxykynuren-
amine,
an
endogenously
occurring
metabolite
of
melatonin
possessing
a
non-indolic
structure,
was
as
potent
as
6-
methoxymelatonin
in
competing
for
2-[1251]iodomelatonin
binding
sites
in
chicken
retinal
membranes.
N-Acetyltrypt-
amine,
a
putative
melatonin
receptor
antagonist
in
chicken
retina
(16,
17),
and
N-acetyl-5-hydroxytryptamine
were
less
.s
o
Q
4)-
Im
la
0
00
l-
o
,
E
t4
0
IL
'Or
5
I
T
F~
~
i
2
4 6
8
10
Time,
hr
FIG.
1.
Reversible
binding
of
2-[1251]iodomelatonin
to
chicken
retinal
membranes
at
0C.
Specific
binding
of
2-[125I]iodomelatonin
(40
pM)
was
reversible
upon
addition
of
3
puM
6-chloromelatonin
(arrow).
Values
are
means
+
SEM
of
three
determinations.
Neurobiology:
Dubocovich
and
Takahashi

3918
Neurobiology:
Dubocovich
and
Takahashi
'E
c
60-
ct
v
cS~
0
v
-
m.
E'-40-
-
0
E0
-
..
v
2
-0I
3
p
O
-
-6
2-[125
I]Iodomelatonin,
nM
10
10
-
X
5
0
1
0
20
40
2-[
1251]
lodomelatonin
specifically
bound,
fmol/mg
of
protein
Table
1.
Pharmacological
profile
of
the
melatonin
binding
site
of
chicken
retina
Ki
(nM)
for
IC50
(nM)
for
2-['251]iodomelatonin
[3H]dopamine
Inhibitor
binding*
releaser
60
FIG.
2.
Scatchard
analysis
of
2-[1251]iodomelatonin
binding
to
chicken
retinal
membranes.
Membranes
were
incubated
with
various
concentrations
of
2-[1251]iodomelatonin
(0.05-1.2
nM)
for
5
hr
at
00C.
Nonspecific
binding
(v)
was
measured
in
the
presence
of
3
AiM
6-chloromelatonin.
Specific
binding
(s)
is
defined
as
total
binding
(a)
minus
nonspecific
binding.
Values
shown
are
means
from
a
repre-
sentative
experiment
performed
in
triplicate.
(Left)
Saturation
curve.
(Right)
Transformation
of
the
saturation
curve
by
the
method
of
Scatchard;
this
analysis
gave
a
dissociation
constant
(Kd)
of
0.26
nM
and
a
total
number
of
binding
sites
(Bmax)
of
55.6
fmol/mg
of
protein.
potent
than
the
corresponding
5-methoxyindoles
(Table
1).
Replacement
of
the
acetamido
group
of
melatonin
with
a
primary
amino
group,
as
in
5-methoxytryptamine,
with
a
hydroxyl
group,
as
in
5-methoxytryptophol,
or
with
a
tertiary
amino
group,
as
in
5-methoxy-N,N-dimethyltryptamine,
sig-
nificantly
reduced
the
inhibition
of
2-[125I]iodomelatonin
binding
(Table
1
and
ref.
17).
Antagonists
of
serotonin
and
dopamine
receptors
and
of
a-
and
P-adrenergic
receptors
were
fairly
poor
inhibitors
of
2-[125I]iodomelatonin
binding
(Table
1).
2-[1251]Iodomelatonin
binding
was
not
affected
by
inhibitors
of
the
neuronal
uptake
of
dopamine
or
norepineph-
rine
(Table
1).
Correlation
Between
Ability
to
Inhibit
2-[251]Iodomelatonin
Binding
and
Ability
to
Inhibit
the
Calcium-Dependent
Release
of [3H]Dopamine
from
Retina.
Melatonin
and
related
indoles
inhibited
the
calcium-dependent
release
of
[3H]dopamine
from
chicken
retina
with
a
similar
order
of
potency
to
that
found
in
rabbit
retina
(Table
1
and
ref.
17).
The
iodinated
ligand
is
biologically
active,
since
2-iodomelatonin
inhibited
the
calcium-dependent
release
of
[3H]dopamine
from
chicken
retina
in
vitro
(Table
2).
2-lodomelatonin
was
about
10
times
more
potent
than
melatonin
in
inhibiting
the
calcium-depen-
dent
release
of
[3H]dopamine
and
did
not
affect
the
sponta-
neous
release
of
radioactivity
from
chicken
retina.
These
results
demonstrate
that
iodination
of
melatonin
in
the 2
position
of
the
indole
ring
did
not
reduce
the
biological
activity
of
the
molecule
and
suggest
that
2-[1251]iodomelato-
nin
may
bind
to
a
melatonin
receptor
site
in
the
retina
(15-17).
In
the
chicken
retina,
the
melatonin
analogs
that
were
the
most
potent
inhibitors
of
[3H]dopamine
release
were
those
possessing
a
methoxy
group
on
carbon
5
of
the
indole
nucleus
and
an
acetamidoethyl
group
on
the
same
position
as
in
melatonin.
The
relative
potencies
(IC50)
of
5-methoxyindoles
in
inhibiting
[3H]dopamine
release
from
chicken
retina
were
as
follows:
2-iodomelatonin
:
6-chloromelatonin
:
mela-
tonin
>
6-hydroxymelatonin
>
6-methoxymelatonin
>
5-
methoxytryptamine
(Table
1).
This
order
of
potencies
for
inhibition
of
[3H]dopamine
release
is
identical
to
the
one
reported
by
Dubocovich
(17)
for
rabbit
retina.
N-Acetyltrypt-
amine,
a
putative
melatonin
receptor
antagonist,
competi-
tively
antagonized
the
inhibitory
effect
of
melatonin
in
chicken
retina
(16,
17).
In
support
of
an
association
of
the
melatonin
binding
site
labeled
by
2-[1251]iodomelatonin
with
a
functional
melatonin
receptor,
a
highly
significant
correlation
was
found
between
the
potency
of
indoles
to
inhibit
the
calcium-dependent
5-Methoxyindoles
2-Iodomelatonin
6-Chloromelatonin
Melatonin
6,7-Dichloro-2-methyl-
melatonin
6-Hydroxymelatonin
6-Methoxymelatonin
5-Methoxytryptamine
5-Methoxytryptophol
5-Methoxy-N,N-di-
methyltryptamine
5-Methoxytryptophan
5-Methoxyindole-3-acetic
acid
5-Methoxyindole
5-Hydroxyindoles
N-Acetyl-5-hydroxy-
tryptamine
5-Hydroxytryptophol
5-Hydroxytryptamine
5-Hydroxytryptophan
5-Hydroxyindole-3-acetic
acid
Indoles
N-Acetyltryptamine
Tryptamine
L-Tryptophan
N-Acetyltryptophan
Indole-3-acetic
acid
Indomethacin
Miscellaneous
compounds
N-Acetyl-5-methoxy-
kynurenamine
6-Methoxyharmalan
Methysergide
Methiothepine
Yohimbine
Bufotenine
Fluphenazine
Nomifensine
Desipramine
Propranolol
Colchicine
2.5
4.0
6.3
10
74
460
4,600
46,400
>100,000
>100,000
>100,000
>100,000
3,000
30,000
>100,000
>100,000
0.1
0.5
1
30
100
200
>10,000
>10,000
>10,000
>10,000
300
>10,000
>100,000
1,600
>100,000
>100,000
>100,000
>100,000
>100,000
>1,000*
500
1,600
6,300
6,300
16,000
20,000
30,000
>100,000
>100,000
>100,000
>100,000
*K1
values
were
calculated,
from
IC50
values
obtained
from
compe-
tition
curves,
by
the
method
of
Cheng
and
Prusoff
(26).
Inhibition
of
specific
binding
of
2-['251]iodomelatonin
(30-60
pM)
in
chicken
retinal
membranes
was
determined
for
11
concentrations
of
com-
peting
drugs.
Results
are
mean
values
of
3-8
independent
deter-
minations.
tEach
indole
compound
was
tested,
at
three
to
seven
concentrations,
for
its
effect
on
the
[3H]dopamine
release
evoked
by
electrical
stimulation
(3
Hz,
2
min)
of
chicken
retina.
The
IC50
value
is
the
concentration
of
drug
required
to
inhibit
the
calcium-dependent
release
of
[3H]dopamine
by
50%.
The
IC50
values
were
determined
graphically
from
concentration-effect
curves.
tKB
value
for
the
putative
melatonin
receptor
antagonist
N-
acetyltryptamine
is
33
nM
(16,
17).
release
of
[3H]dopamine
from
chicken
and
rabbit
retina
and
the
potency
to
compete
for
2-[1251]iodomelatonin
binding
in
chicken
retinal
homogenates
(Fig.
4).
This
correlation
strong-
ly
suggests
that
2-[1251]iodomelatonin
labels
a
site
with
the
pharmacological
and
functional
characteristics
of
a
melatonin
receptor.
At
this
time,
however,
we
cannot
exclude
the
Proc.
Natl.
Acad.
Sci.
USA
84
(1987)

Proc.
Natl.
Acad.
Sci.
USA
84
(1987)
3919
C
-
v-
-C
-
Co
C
L-
C-
100
-
c/
I,-
50
c
d
\
\
...
\i
\
e
n
I,,,
I
I
x
12
10
8
6
4
-log[inhibitorl
(M)
FIG.
3.
Competition
curves
for
inhibition
of
2-[1251]iodomelatonin
binding
by
various
melatonin
agonists
in
chicken
retinal
membranes.
Washed
chicken
retinal
membranes
were
incubated
with
30-60
pM
2-[1251]iodomelatonin
and
various
concentrations
of
5-hydroxytrypt-
amine
(curve
a,
*),
N-acetyl-5-hydroxytryptamine
(curve
b,
v),
6-hydroxymelatonin
(curve
c,
*),
6-chloromelatonin
(curve
d,
0),
or
melatonin
(curve
e,
*).
Values
are
means
of
3-8
independent
determinations.
possibility
that
2-[1251]iodomelatonin
may
label
melatonin
receptor
binding
sites
regulating
other
processes
in
retina,
such
as
photoreceptor
outer
segment
disc
shedding
and
phagocytosis
(11,
12)
and
cone
retinomotor
movements
(14),
in
addition
to
those
sites
modulating
[3H]dopamine
release.
DISCUSSION
Our
results
strongly
suggest
that
2-[1251]iodomelatonin
is
a
selective,
high-affinity
ligand
for
the
identification
and
char-
acterization
of
melatonin
receptor
sites.
The
specific
binding
of
2-[1251]iodomelatonin
fulfills
all
the
criteria
for
binding
to
a
receptor
site,
being
stable,
reversible,
saturable,
and
of
high
affinity.
2-[1251]Iodomelatonin
appears
to
label
a
single
class
of
sites
in
membranes
isolated
from
chicken
retina.
The
apparent
affinity
constant
derived
from
kinetic
analysis
was
in
close
agreement
with
the
Kd
obtained
from
saturation
experiments.
Competition
curves
were
monophasic
and
Table
2.
Effect
of
melatonin
and
2-iodomelatonin
on
the
calcium-
dependent
release
of
[3H]dopamine
from
chicken
retina
%
total
tissue
Drug
present
radioactivityt
during
S2*
n
S1
S2
S2/Sj
ratio:
None
(control)
8
1.54
±
0.28
1.25
±
0.16
0.87
±
0.06
Melatonin
(10
nM)
8
1.45
±
0.26
0.56
±
0.14§
0.36
±
0.06§
2-lodomelatonin
(2
nM)
11
1.13
±
0.09
0.51
±
0.080
0.43
±
0.05$
*Drugs
were
added
to
the
medium
20
min
before
the
second
period
of
stimulation
(S2).
In
the
controls,
the
spontaneous
outflow
of
radioactivity
calculated
as
the
percentage
of
total
tissue
radioac-
tivity
released
during
the
4
min
preceding
the
first
period
of
stimulation
(Sl)
was
1.27%
+
0.10%
(n
=
8).
(S)-Sulpiride
(0.1
/LM)
was
present
in
every
experiment
from
40
min
before
S1.
Melatonin
or
2-iodomelatonin,
present
in
the
superfusion
medium
from
20
min
before
S2,
did
not
modify
the
spontaneous
outflow
of
radioactivity
at
the
concentrations
indicated.
Radioactivity
retained
by
the
control
tissue
after
120
min
of
superfusion
was
62.4
±
12.7
nCi
per
chamber
(n
=
8).
tPercentage
of
the
total
tissue
radioactivity
released
above
the
spontaneous
levels
elicited
by
field
stimulation
at
3
Hz
(20
mA,
2
msec)
during
a
2-min
period.
tRatio
of
the
percentage
of
total
tissue
radioactivity
released
during
S2
to
that
released
during
S1.
§P
<
0.01
and
¶P
<
0.001
when
compared
with
corresponding
control
(Student's
t
test).
A
_-
';
-
c
4
-C
4
C_
-
E
'be
6
_
*,.
r-
1_
CI
1()
I
11
9
7
5
B
k
12
10
8
6
-log
WCs(,
(M)
for
inhibition
of
1HIdoparmine
release
FIG.
4.
Correlation
between
the
affinities
of
compounds
for
2-['25I]iodomelatonin
binding
sites
in
chicken
retina
and
their
ability
to
inhibit
[3H]dopamine
release
from
chicken
retina
(A)
and
rabbit
retina
(B).
Ki
values
for
inhibition
of
2-[1251]iodomelatonin
binding
in
chicken
retinal
membranes
and
IC50
values
for
inhibition
of
the
calcium-dependent
release
of
[3H]dopamine
from
chicken
retina
were
obtained
from
Table
1.
IC50
values
for
inhibition
of
the
calcium-dependent
release
of
[3H]dopamine
from
rabbit
retina
were
obtained
from
ref.
17
and
unpublished
observations.
Linear
regres-
sion
of
a
logarithmic
transformation
of
the
data
yielded
a
slope
of
0.952
and
a
correlation
coefficient
of
0.968
(P
<
0.01,
n
=
7)
in
A
and
a
slope
of
0.767
and
a
correlation
coefficient
of
0.969
(P
<
0.01,
n
=
11)
in
B.
Compounds
are
indicated
as
follows:
a,
2-iodomelatonin;
b,
6-chloromelatonin;
c,
melatonin;
d,
6-hydroxymelatonin;
e,
6-methoxy-
melatonin;
f,
N-acetyl-5-hydroxytryptamine;
g,
5-methoxytryptamine;
h,
6,7-dichloro-2-methylmelatonin;
i,
N-acetyl-5-methoxykynuren-
amine;
j,
N-acetyltryptamine;
k,
5-methoxytryptophol.
Scatchard
plots
were
linear.
The number
of
2-[1251]iodomel-
atonin
binding
sites
found
in
chicken
retina
(Bmax
=
74.0
+
13.6
fmol/mg
of
protein)
was
almost
identical
with
the
number
of
sites
reported
for
[3H]melatonin
in
the
frog
retina
(10).
However,
the
affinity
of
the
melatonin
binding
site
of
frog
retina,
determined
by
use
of
[3H]melatonin,
was
<0.001
times
(10)
the
affinity
of
the
melatonin
binding
site
of
chicken
retina
determined
by
use
of
2-[1251]iodomelatonin.
2-[1251I]Iodomelatonin
offers
distinct
advantages
for
the
characterization
of
melatonin
binding
sites.
The
binding
of
2-[1251]iodomelatonin
to
chicken
retinal
membranes
was
90%
specific,
even
with
protein
concentrations
as
low
as
30
jig
per
assay,
allowing
detection
of
melatonin
binding
sites
in
small
tissue
samples.
The
pharmacological
characterization
of the
melatonin
receptor
site
labeled
by
2-[1251]iodomelatonin
in-
dicated
that
the
radioligand
selectively
binds
to
a
site
with
the
pharmacological
characteristics
of
a
melatonin
receptor,
since
melatonin
and
related
indoles
inhibited
2-[1251]iodomel-
atonin
binding
with
the
same
order
of
potency
found
for
inhibition
of
release
of
[3H]dopamine
from
chicken
(Table
1)
and
rabbit
(17)
retina.
In
rabbit
retina,
2-[1251]iodomelatonin
also
appears
to
label
a
melatonin
receptor
site,
since
a
highly
significant
correlation
was
obtained
between
the
relative
potencies
of
melatonin
analogs
in
inhibiting
2-[1251]iodomel-
atonin
binding
and
in
inhibiting
calcium-dependent
release
of
[3H]dopamine
from
rabbit
retina
(r
=
0.799,
P
<
0.01,
n
=
8;
ref.
28).
[3H]Melatonin
binding
sites
have
been
demonstrated
in
several
tissues.
Saturable
binding
of
[3H]melatonin
has
been
shown
in
bovine
hypothalamic
membranes
(18)
and
in
cyto-
solic
fractions
of
brain
from
several
species
(19,
21).
[3H]Melatonin
also
binds
to
membranes
of
bovine
pineal
gland
(29)
and
trout
and
frog
retinas
(10,
20).
In
these
studies
the
binding
site
labeled
by
[3H]melatonin
was
of
relatively
low
affinity
and
the
abilities
of
the
melatonin-related
indoles
to
compete
for
this
binding
site
were
not
correlated
with
any
functional
response
to
melatonin.
Moreover,
the
order
of
potency
of
melatonin
and
related
indoles
on
the
[3H]melato-
nin
binding
site
of
bovine
hypothalamus
(18)
differs
consid-
erably
from
the pharmacological
characteristics
of
the
mel-
atonin
receptor
in
rabbit
and
chicken
retina
(ref.
17;
present
Neurobiology:
Dubocovich
and
Takahashi
VI

3920
Neurobiology:
Dubocovich
and
Takahashi
results).
The
most
distinct
difference
is
that
of
6-hydroxy-
melatonin,
which
is
very
potent
in
inhibiting
[3H]dopamine
release
and
competing
for
the
receptor
labeled
by
2-[125I]-
iodomelatonin
in
rabbit
and
chicken
retina
(17,
28)
but
is
a
poor
inhibitor
of
[3H]melatonin
binding
in
bovine
hypothal-
amus
(18).
In
contrast,
indoles
such
as
5-methoxytryptophol,
5-methoxyindole-3-acetic
acid,
5-hydroxytryptamine
(sero-
tonin),
and
5-hydroxytryptophol,
which
are
inactive
or
poor
agonists
on
the
melatonin
receptor
of
chicken
and
rabbit
retina
(refs.
17
and
28;
present
results),
are
among
the
most
potent
competitors
of
[3H]melatonin
binding
in
bovine
hy-
pothalamus
(18).
Similar
discrepancies
in
the
order
of
poten-
cy
of
melatonin
and
related
indoles
on
the
melatonin
binding
site
are
observed
between
our
results
obtained
with
2-
['25I]iodomelatonin
and
those
reported
for
retinas
of
lower
vertebrates
with
[3H]melatonin
(10,
20).
In
contrast,
we
have
recently
found
that,
in
hamster
brain
membranes,
2-[125I]_
iodomelatonin
labels
a
melatonin
binding
site
whose
phar-
macological
characteristics
are
almost
identical
to
those
found
for
the
binding
site
on
retinal
membranes
(30).
The
differences
in
the
pharmacological
characteristics
of
the
sites
labeled
by
[3H]melatonin
and
2-['25I]iodomelatonin
in
mem-
branes
prepared
from
central
nervous
system
tissues
appear
to
reflect
different
binding
sites.
Using
a
functional
measurement
of
melatonin
activity,
Dubocovich
(17)
showed
that
the
efficacy
of
melatonin
and
related
indoles
in
inhibiting
the
calcium-dependent
release
of
[3H]dopamine
from
rabbit
retina
is
determined
by
the
moiety
(methoxy)
on
carbon
5
of
the
indole
nucleus,
whereas
the
affinity
for
the
receptor
is
determined
primarily
by
the
moiety
(acetamidoethyl)
on
carbon
3.
In
support
of
this
conclusion,
we
have
found
that
the
most
potent
inhibitors
of
2-['1251]io-
domelatonin
binding
in
retina
were
the
N-acetyltryptamines.
On
the
basis
of
both
the
2-[1251]iodomelatonin
binding
and
the
functional
responses
to
indole
amines,
we
suggest
that
a
melatonin
receptor
is
characterized
by
the
following
phar-
macological
order
of
affinities:
2-iodomelatonin
>
6-chloro-
melatonin
:
melatonin
:
6,7-dichloro-2-methylmelatonin
>
6-hydroxymelatonin
:
6-methoxymelatonin
>
N-acetyltryp-
tamine
>
N-acetyl-5-hydroxytryptamine
>
5-methoxy-
tryptamine
>>>
5-hydroxytryptamine
(inactive).
In
addition
to
its
role
in
the
retina,
melatonin
has
been
implicated
in
the
regulation
of
seasonal
reproduction
in
mammals
(1,
2)
and
in
the
control
of
circadian
rhythms
in
birds
and
reptiles
(3-5).
Although
it
is
generally
assumed
that
melatonin
acts
centrally
upon
the
hypothalamopituitary
axis
in
mammals
(31)
the
anatomical
loci
of
melatonin
receptors
in
the
central
nervous
system
have
not
been
identified.
We
have
recently
characterized
melatonin
binding
sites
labeled
by
2-[1251]iodomelatonin
in
hamster
brain
membranes
(30).
This
binding
site
exhibits
pharmacological
characteristics
similar
to
those
of
the
melatonin
receptor
of
chicken
and
rabbit
retina
(17,
28).
In
summary,
the
radioligand
2-[125I]iodomelatonin
should
provide
a
useful
probe
for
the
localization
and
characterization
of
central
melatonin
receptors,
as
well
as
the
elucidation
of
the
mechanism
of
action
of
melatonin.
We
thank
Patrick
Rita
for
technical
assistance,
Selene
S.
Nikaido
for
preparing
the
2-iodomelatonin,
and
Dr.
Marilyn
J.
Duncan
for
critical
comments
on
the
manuscript.
This
work
was
supported
by
Public
Health
Service
Grants
EY04788,
RR05470,
and
MH39592;
by
National
Science
Foundation
Grant
DCB-8451642;
and
by
Searle
Scholars
Award
85-H-107.
1.
Goldman,
B.
D.
(1983)
in
Pineal
Research
Reviews,
ed.
Reiter,
R.
J.
(Liss,
New
York),
pp.
145-182.
2.
Reiter,
R.
J.
(1982)
in
Frontiers
in
Neuroendocrinology,
eds.
Ganong,
W.
F.
&
Martini,
L.
(Raven,
New
York),
pp.
287-316.
3.
Menaker,
M.
(1982)
in
Vertebrate
Circadian
Systems,
eds.
Aschoff,
J.,
Daan,
S.
&
Gross,
G.
A.
(Springer,
Berlin),
pp.
1-12.
4.
Takahashi,
J.
S.
&
Menaker,
M.
(1984)
in
Biological
Regula-
tion
and
Development,
eds.
Goldberger,
R.
F.
&
Yamamoto,
K.
R.
(Plenum,
New
York),
pp.
285-303.
5.
Underwood,
H.
&
Harless,
M.
(1985)
Physiol.
Behav.
35,
267-270.
6.
Hamm,
H.
E.
&
Menaker,
M.
(1980)
Proc.
Natl.
Acad.
Sci.
USA
77,
4998-5002.
7.
Yu,
H.
S.,
Pang,
S.
F.
&
Tang,
P.
L.
(1981)
J.
Endocrinol.
91,
477-481.
8.
Reiter,
R.
J.,
Richardson,
B.
A.,
Matthews,
S.
A.,
Lane,
S.
J.
&
Ferguson,
B.
N.
(1983)
Life
Sci.
32,
1229-1236.
9.
Dubocovich,
M.
L.,
Lucas,
R.
C.
&
Takahashi,
J.
S.
(1985)
Brain
Res.
335,
321-325.
10.
Weichmann,
A.
F.,
Bok,
D.
&
Horwitz,
J.
(1986)
Invest.
Ophthalmol.
Vis.
Sci.
27,
153-163.
11.
Besharse,
J.
C.
&
Dunnis,
D.
A.
(1983)
Science
219,
1341-1343.
12.
Ogino,
N.,
Matsumura,
M.,
Shirakawa,
H.
&
Tsukahara,
I.
(1983)
Ophthalmic
Res.
15,
72-89.
13.
Pang,
S.
G.
&
Yew,
D.
T.
(1979)
Experientia
35,
231-233.
14.
Pierce,
M.
E.
&
Besharse,
J.
E.
(1985)
J.
Gen.
Physiol.
86,
671-689.
15.
Dubocovich,
M.
L.
(1983)
Nature
(London)
306,
782-784.
16.
Dubocovich,
M.
L.
(1984)
Eur.
J.
Pharmacol.
105,
193-194.
17.
Dubocovich,
M.
L.
(1985)
J.
Pharmacol.
Exp.
Ther.
234,
395-401.
18.
Cardinali,
D.
P.,
Vacas,
M.
I.
&
Estevez
Boyer,
E.
(1979)
Endocrinology
105,
437-441.
19.
Niles,
L.
P.,
Wong,
Y.-W.,
Mishra,
R.
K.
&
Brown,
G.
M.
(1979)
Eur.
J.
Pharmacol.
55,
219-220.
20.
Gem,
W.
A.,
Gorell,
T.
A.
&
Owens,
D.
W.
(1981)
in
Melatonin-Current
Status
and
Perspectives:
Advances
in
the
Biosciences,
eds.
Birau,
N.
&
Schloot,
W.
(Pergamon,
Ox-
ford),
Vol.
29,
pp.
223-233.
21.
Brown,
G.,
Grota,
L.,
Bubenik,
G.,
Niles,
L.
&
Tsui,
H.
(1981)
in
Melatonin-Current
Status
and
Perspectives:
Advances
in
the
Biosciences,
eds.
Birau,
N.
&
Schloot,
W.
(Pergamon,
Oxford),
Vol.
29,
pp.
95-112.
22.
Vakkuri,
O.,
Leppaluoto,
J.
&
Vuolteenaho,
0.
(1984)
Acta
Endocrinol.
106,
152-157.
23.
Vakkuri,
O.,
Lamsa,
E.,
Rahkamaa,
E.,
Ruotsalainen,
H.
&
Leppaluoto,
J.
(1984)
Anal.
Biochem.
142,
284-289.
24.
Bennett,
J.
P.
&
Yamamura,
H.
I.
(1985)
in
Neurotransmitter
Receptor
Binding,
eds.
Yamamura,
H.
I.,
Enna,
S.
J.
&
Kuhar,
M.
J.
(Raven,
New
York),
2nd
Ed.,
pp.
61-89.
25.
McPherson,
G.
A.
(1983)
Computer
Prog.
Biomed.
17,
107-114.
26.
Cheng,
Y.-C.
&
Prusoff,
W.
H.
(1973)
Biochem.
Pharmacol.
22,
3099-3108.
27.
Dubocovich,
M.
L.
&
Weiner,
N.
(1981)
J.
Pharmacol.
Exp.
Ther.
219,
701-707.
28.
Dubocovich,
M.
L.,
Nikaido,
S.
S.
&
Takahashi,
J.
S.
(1986)
Soc.
Neurosci.
Abstr.
12,
995.
29.
Vacas,
M.
I.
&
Cardinali,
D.
P.
(1980)
Hormone
Res.
13,
121-131.
30.
Duncan,
M.
J.,
Takahashi,
J.
S.
&
Dubocovich,
M.
L.
(1986)
Soc.
Neurosci.
Abstr.
12,
994.
31.
Glass,
J.
D.
&
Lynch,
G.
R.
(1981)
Science
214,
821-823.
Proc.
Natl.
Acad.
Sci.
USA
84
(1987)