Melatonin receptors and their regulation: biochemical and structural mechanisms.

Abstract

There is growing evidence demonstrating the complexity of melatonin's role in modulating a diverse number of physiological processes. This complexity could be attributed to the fact that melatonin receptors belong to two distinct classes of proteins, that is, the G-protein coupled receptor superfamily (MT(1), MT(2)) and the quinone reductase enzyme family (MT(3)) which makes them unique at the molecular level. Also, within the G-protein coupled receptor family of proteins, the MT(1) and MT(2) receptors can couple to multiple and distinct signal transduction cascades whose activation can lead to unique cellular responses. Also, throughout the 24-hour cycle, the receptors' sensitivity to specific cues fluctuates and this sensitivity can be modulated in a homologous fashion, that is, by melatonin itself, and in a heterologous manner, that is, by other cues including the photoperiod or estrogen. This sensitivity of response may reflect changes in melatonin receptor density that also occurs throughout the 24-hour light/dark cycle but out of phase with circulating melatonin levels. The mechanisms that underlie the changes in melatonin receptor density and function are still not well-understood, but data is beginning to show that transcriptional events and G-protein uncoupling may be involved. Even though this area of research is still in its infancy, great strides are being made everyday in elucidating the mechanisms that underlie melatonin receptor function and regulation. The focus of this review is to highlight some of these discoveries in an attempt to reveal the uniqueness of the melatonin receptor family while at the same time provide thought-provoking ideas to further advance this area of research. Thus, a brief overview of each of the mammalian melatonin receptor subtypes and the signal transduction cascades to which they couple will be discussed with a greater emphasis placed on the mechanisms underlying their regulation and the domains within the receptors essential for proper signaling.

Full text

Page 1

Minireview
Melatonin receptors and their regulation: biochemical and
structural mechanisms
Paula A. Witt-Enderbya,*, Jennifer Bennettb, Michael J. Jarzynkaa,
Steven Firestinea, Melissa A. Melanc
aDivision of Pharmaceutical Sciences, Duquesne University School of Pharmacy, 421 Mellon Hall, Pittsburgh, PA 15282, USA
bCarnegie Mellon University, Bone Tissue Engineering Center, 125 Smith Hall, 5000 Forbes Avenue,
Pittsburgh, PA 15213, USA
cDivision of Molecular Diagnostics, Department of Pathology, University of Pittsburgh Medical Center, S-738 Scaife Hall,
3550 Terrace Street, Pittsburgh, PA 15231, USA
Received 8 May 2002; accepted 11 December 2002
Abstract
There is growing evidence demonstrating the complexity of melatonin’s role in modulating a diverse number
of physiological processes. This complexity could be attributed to the fact that melatonin receptors belong to two
distinct classes of proteins, that is, the G-protein coupled receptor superfamily (MT1, MT2) and the quinone
reductase enzyme family (MT3) which makes them unique at the molecular level. Also, within the G-protein
coupled receptor family of proteins, the MT1and MT2receptors can couple to multiple and distinct signal
transduction cascades whose activation can lead to unique cellular responses. Also, throughout the 24-hour cycle,
the receptors’ sensitivity to specific cues fluctuates and this sensitivity can be modulated in a homologous fashion,
that is, by melatonin itself, and in a heterologous manner, that is, by other cues including the photoperiod or
estrogen. This sensitivity of response may reflect changes in melatonin receptor density that also occurs
throughout the 24-hour light/dark cycle but out of phase with circulating melatonin levels. The mechanisms that
underlie the changes in melatonin receptor density and function are still not well-understood, but data is
beginning to show that transcriptional events and G-protein uncoupling may be involved. Even though this area
of research is still in its infancy, great strides are being made everyday in elucidating the mechanisms that underlie
melatonin receptor function and regulation. The focus of this review is to highlight some of these discoveries in
an attempt to reveal the uniqueness of the melatonin receptor family while at the same time provide thought-
provoking ideas to further advance this area of research. Thus, a brief overview of each of the mammalian
melatonin receptor subtypes and the signal transduction cascades to which they couple will be discussed with a
0024-3205/03/$ - see front matter D 2003 Elsevier Science Inc. All rights reserved.
doi:10.1016/S0024-3205(03)00098-5
* Corresponding author. Tel.: +1-412-396-4346; fax: +1-412-396-5599.
E-mail address: wittp@duq.edu (P.A. Witt-Enderby).
www.elsevier.com/locate/lifescie
Life Sciences 72 (2003) 2183–2198

Page 2

greater emphasis placed on the mechanisms underlying their regulation and the domains within the receptors
essential for proper signaling.
D 2003 Elsevier Science Inc. All rights reserved.
Keywords: Melatonin; Melatonin receptors; G-protein coupled receptors; Molecular pharmacology; Signal transduction;
Receptor regulation; Desensitization; Neurites
Introduction
Melatonin, acting through melatonin receptors, is involved in numerous physiological processes
including circadian entrainment [19,42,47,72,73,82], blood pressure regulation [15,79], oncogenesis
[6,62,90], retinal physiology [18], seasonal reproduction [3,52,72], ovarian physiology [10,40], and
most recently in inducing osteoblast differentiation [69] (review by [41]). When one examines the
diversity of response of one class of receptors, the question that comes to mind is: How can a single
hormone produce such variable responses within the body? As one probes into the field of melatonin and
its receptors, the answer to this important question becomes a little more obvious.
Many factors contribute to the diversity of the melatonin response within the body. First of all,
melatonin levels fluctuate during the day and throughout the year [63]. For example, levels of melatonin
are lowest during the day and highest at night and persist for longer periods of time during the winter
Fig. 1. Schematic showing that when melatonin levels (.) are highest (during the hours of darkness), the affinity of melatonin
for its receptor (
), total 2-[125I]-iodomelatonin binding and the expression of melatonin receptor mRNA are lowest. In
addition, when cultured or recombinant cells expressing melatonin receptors are exposed to melatonin chronically, the potency
of melatonin at melatonin receptors decreases perhaps due to an uncoupling of the receptor from its effector, due to receptor
internalization and/or due to receptor down-regulation. However, when melatonin levels are lowest (during the hours of
daylight), the affinity of melatonin for 2-[125I]-iodomelatonin binding, total 2-[125I]-iodomelatonin binding and the expression
of melatonin receptor mRNA are highest. Data taken from references [28,30,55,73,78,87].
P.A. Witt-Enderby et al. / Life Sciences 72 (2003) 2183–2198
2184

Page 3

months as compared to the summer months (Fig. 1). These fluctuations in melatonin levels in vivo can
and do impact greatly on the functioning of the receptor. Melatonin can activate or inhibit signal
transduction cascades independent of receptors or through receptors. The ability of melatonin to act
independently from its receptors is attributed to its small and highly lipophilic nature and/or due to an
active uptake mechanism [5,21,49]. There are multiple receptor subtypes available to which melatonin
can bind and activate. Two of the melatonin receptor subtypes are G-protein coupled receptors [64,65]
while the third, which was recently affinity-purified, belongs to the family of quinone reductases [57].
Each of the G-protein coupled melatonin receptor subtypes, denoted MT1and MT2, can couple to
multiple signal transduction cascades whereas the signal transduction cascades mediating MT3responses
are still unclear. Finally, melatonin receptor expression and perhaps function can be regulated by
multiple cues including the light/dark cycle, scheduled arousal, an endogenous pacemaker, by melatonin
itself, and/or by other hormones.
The focus of this review is not to provide an exhaustive bibliography of melatonin receptors but rather
strives to present data that highlights key areas of interest and debate in this area of research. Thus, a
brief overview of each of the mammalian melatonin receptor subtypes and the signal transduction
cascades to which they couple will be discussed with a greater emphasis placed on the mechanisms
underlying their regulation and the domains within the receptors essential for proper signaling. For more
comprehensive summaries of melatonin receptors, please refer to these other excellent reviews
[45,80,83].
Melatonin Receptor Subtypes, Tissue Localization and Their Role in Physiological Processes
To date, three mammalian melatonin receptors have been either cloned, MT1[64] MT2[65] or
affinity-purified, MT3[57]. Two of these receptors are G-protein coupled receptors and are denoted as
MT1and MT2whereas the newly purified MT3protein belongs to the family of the quinone reductases.
The MT1and MT2melatonin receptors are classified as unique subtypes based on their molecular
structure and chromosomal localization [65,75,76].
MT1
The diversity of melatonin’s response within the body may be attributed to the fact that its
receptors are expressed in a wide variety of tissues. The MT1melatonin receptor (MT1R) which is
expressed in the suprachiasmatic nucleus of the hypothalamus (SCN) and cardiac vessels is involved
in modulating circadian rhythms [19,42] and constricting cardiac vessels [15]. Besides these specific
regions, the MT1R is expressed in other regions of the brain and peripheral tissues [10,62,71,92,93] as
reviewed [41,45,83]. Elucidating the physiological role of MT1Rs in many of these tissues is still
under investigation, however, the results from these studies are beginning to reveal some interesting
results.
Besides its vast tissue distribution, the fact that MT1Rs can couple to a wide variety of G-proteins
including Gia2, Gia3and Gaq[7,89], Gas, Gazand Ga16[9,34] may also explain its diversity of response
within the body. Even at the level of the cell, melatonin, acting through MT1Rs, can produce multiple
cellular responses. As shown in numerous studies, MT1Rs have been shown to produce inhibitory
responses on the cAMP signal transduction cascade [7,8,51,56,64,85], resulting in decreases in PKA
P.A. Witt-Enderby et al. / Life Sciences 72 (2003) 2183–2198
2185

Page 4

activity [52,87] and decreases in CREB phosphorylation [48,87]. This is a more generalized signaling
mechanism proposed for MT1Rs. However, in COS-7 cells, MT1Rs have also been shown to stimulate
cAMP probably through Gas[9]. Whether or not this is an artifact of the system needs to be supported or
refuted by further investigations in other cell models and/or tissues. Besides the cAMP-dependent
cascade, MT1Rs can couple to a stimulation of PLC-dependent signal transduction cascades directly
[7,34,44] or indirectly via Ghgsubunits [26,67] and can activate PKC [89]. MT1Rs can couple to
calcium activated potassium (BKCa
potassium (GIRK Kir 3) channels [36]. These receptors can also modulate the formation of arachidonic
acid [26], can stimulate c-Jun N-terminal kinase (JNK) activity [9] and also modulate MAP kinases
[9,31,88].
As stated above, MT1Rs can modulate the nuclear factor, CREB. However, others have shown that
activated MT1Rs can also inhibit the induction of c-fos and jun B mRNA and c-Fos translation induced
by forskolin [68]. Activator protein-1 (AP-1), a transcription factor formed by the immediate-early gene
products c-fos and c-jun, has been shown to be extensively regulated through the MAP Kinase pathway,
specifically ERK1 and ERK2, and JNK [84]. Thus, taken together, the findings that melatonin can
regulate these factors may provide a mechanism underlying melatonin’s ability to induce differentiation
of certain cells [6].
2 +) channels [24,25] and G-protein-activated inward rectifier
MT2
The role of MT2melatonin receptors (MT2R) in mammalian physiology as well as its signaling
properties is now becoming clearer with the recent development of MT2-selective ligands [18,77,86]. To
date what is known is that MT2Rs are involved in retinal physiology [18], in modulating circadian
rhythms [19], in dilating cardiac vessels [15] and are involved in inflammatory responses in the
microcirculation [43] as reviewed [41,45,83]. Unlike the MT1Rs, these receptors are more restricted in
their localization which includes the cerebellum, SCN of the hypothalamus, the retina, kidney, ovary,
cardiac vessels and various cancerous cell lines as reviewed [41,45,83]. Similar to the MT1R, MT2Rs
couple to an inhibition of cAMP formation in various transfected models [7,8,37,44,65] and to a
stimulation of PI hydrolysis [34,44]. However, unlike the MT1R, activation of the MT2Rs expressed in
HEK293 cells can also result in decreases in cGMP [59]. Perhaps the subtle differences in pharmacology
and signaling properties between the MT1R and MT2R may prove to be an essential component
underlying melatonin’s effects on specific physiological processes.
MT3
Recently, a protein that displays a binding profile similar to that of the ML2receptor [17,50] now
denoted MT3was affinity-purified from Syrian hamster kidney [57]. It was shown that this protein shares
95% homology to the human quinone reductase 2, an enzyme involved in detoxification [57]. This
protein and its associated activity, as revealed through radioligand binding and enzymatic assays, show
that it is expressed in the liver, kidney, brain, heart, brown adipose tissue, skeletal muscle, lung, intestine,
testis and the spleen of hamster, mouse, dog and monkey [58]. Very recently, though, it was shown that
the MT3protein may be involved in the regulation of intraocular pressure in rabbits [61] and in
inflammatory responses in the microvasculature [43]. Further investigation into this protein may reveal
other unique properties as well.
P.A. Witt-Enderby et al. / Life Sciences 72 (2003) 2183–2198
2186

Page 5

The amino acids involved in melatonin receptor activation and signaling
Melatonin receptors MT1 and MT2 are G-protein coupled receptors (GPCRs) that possess
structural motifs consisting of seven membrane spanning domains connected by a series of
extracellular and intracellular loops. The ability of melatonin to bind to, activate, and modulate
its own receptors depends on the interaction of melatonin with specific amino acids and/or domains.
This interaction may be with amino acids or domains unique to each receptor (i.e., MT1R- or
MT2R-specific) or similar between the melatonin receptors depending on the type of response being
elicited.
The analysis of melatonin binding to its receptor and the resulting changes that induce activation has
been aided by recent models of the melatonin receptors that are based upon the crystal structure of
rhodopsin [54]. Using this model, Navajas et al. [54] proposed a hypothetical melatonin binding site, in
which a valine and a histidine residue are responsible for recognition of the 5-methoxy group of
melatonin and a serine and alanine residue are critical for binding to the N-acetyl functional group [54].
This model was supported by analysis of binding of melatonin analogs and sequence alignments of
melatonin and biogenic amine receptors.
Given this foundation, studies of the mutagenesis of the MT1R were undertaken to support or
refute the melatonin receptor model. Mutagenesis of V208 in the ovine Mel1ahreceptor to either an
alanine or valine results in a 2–5 fold decrease in binding of 2[125I]-iodomelatonin as compared to
the wild-type receptor [11]. The mutants are functional in transiently transfected HEK293 cells and
their activities are in rank order to their binding affinities. These data suggest that these mutants
affect only binding, supporting a role for V208 in participating in the binding pocket of the
receptor. However, studies using the melatonin analogs N-[2-(1-napthyl) ethyl] acetamide (NEA) and
N-acetylserotonin failed to show a definitive interaction between V208 and the 5-position of
melatonin.
The role of histidine in the binding of melatonin is more interesting. Mutagenesis of H211 in Mel1ah
to either a phenylalanine or leucine result in a 6–7 fold decrease in the binding of 2-[125I]-iodomelatonin
as compared to the wild-type receptor [11]. Probing the binding of NEA and N-acetylserotonin to H211F
and H211L mutants indicate that analogs lacking either the methyl group of the 5-methoxy substituent or
a functional group at the 5-position bind as well or better to these mutants as compared to the wild-type
enzyme. The exact mechanism by which this occurs is under debate. One idea is that compensatory
changes are perhaps taking place between either phenylalanine or leucine with these hydrophobic
analogs. These studies clearly support a direct interaction between H211 and the 5-position of melatonin
[11]. Further analysis of H211F and H211L reveal that these mutants are not able to inhibit cAMP
production over the melatonin concentration range studied [11]. These functional data reveal that the
histidine residue, in addition to its role in binding, also plays a role in the activation of the receptor [11].
Mutagenesis of the corresponding histidine, a charged amino acid, in the human MT1R (i.e., H195) to a
neutral amino acid alanine, H195A, reveals that this mutation was functional in a yeast-based bioassay.
In this study, GPCR activation was measured via reporter genes coupled to a yeast mating pathway [38].
These data indicate that the charge of histidine does not play a large role in activity and instead suggest
that the size of the amino acid present at this position may be an important determinant in receptor
activation [38]. Interestingly, the yeast-based assay system also shows that the double mutant of H195
and V192T (the corresponding valine position, V208, in ovine Mel1ah) results in loss of function, again
supporting a role for these two residues in melatonin binding [38]. The effect of these mutations on the
P.A. Witt-Enderby et al. / Life Sciences 72 (2003) 2183–2198
2187