REVIEWS

Targeting the endocannabinoid

system: to enhance or reduce?
Vincenzo Di Marzo

Abstract | As our understanding of the endocannabinoids improves, so does the awareness

of their complexity. During pathological states, the levels of these mediators in tissues
change, and their effects vary from those of protective endogenous compounds to those of
dysregulated signals. These observations led to the discovery of compounds that either
prolong the lifespan of endocannabinoids or tone down their action for the potential future
treatment of pain, affective and neurodegenerative disorders, gastrointestinal inflammation,
obesity and metabolic dysfunctions, cardiovascular conditions and liver diseases. When
moving to the clinic, however, the pleiotropic nature of endocannabinoid functions will
require careful judgement in the choice of patients and stage of the disorder for treatment.
Cannabinoids
Natural lipophilic products
from the flower of Cannabis
sativa, most of which have
a typical bicyclic or tricyclic
structure and a common
biogenetic origin from olivetol.

Cannabinoid receptors
G-protein-coupled receptors
for 9-tetrahydrocannabinol,
so far identified in most
vertebrate phyla. Two subtypes
are known: CB1 and CB2.

Endocannabinoids
Endogenous agonists of
cannabinoid receptors in
animals.

9-Tetrahydrocannabinol
The major psychotropic
component of Cannabis sativa,
and one of about 66
cannabinoids found in the
flowers of this plant.

Endocannabinoid Research
Group, Institute of
Biomolecular Chemistry,
National Research Council
(CNR), Via Campi
Flegrei 34, 80078,
Pozzuoli, Naples, Italy.
e-mail:
vdimarzo@icmib.na.cnr.it
doi:10.1038/nrd2553

Owing to its psychotropic and medicinal effects, Cannabis

sativa has been mankinds friend and foe for millennia1.
Yet, it was only in the second half of the twentieth century
that research finally cast light onto some of its mecha
nisms of action. First with the identification of its major
pharmacologically active components, the cannabinoids,
and then with the discovery in vertebrates of molecular
targets for some of these cannabinoids. These targets
include the Gprotein-coupled cannabinoid receptors and
their endogenous ligands, the endocannabinoids2,3. In the
past 10 years, it became clear that, at least in mammals,
the functions of this endocannabinoid signalling system
(BOX1; FIG.1) are not limited to the brain, but are exerted
in the whole organism. Endocannabinoids are generally
considered to be released from cells immediately after
biosynthesis, as no evidence exists for their storage
in secretory vesicles, and several of their biosynthetic
enzymes are found in the plasma membrane. Moreover,
endocannabinoids act on their receptors only locally,
possibly because of their high lipophilicity, and are
immediately inactivated under physiological conditions4.
However, the regulation of their levels by biosynthetic
and degradative enzymes, and their mode of action,
are becoming characterized by an increasing degree of
redundancy of pathways and promiscuity of molecular
targets, respectively4,5.
The realization of the complexity of endocannabinoid
regulation in physiological and pathological conditions
could have discouraged the development of new thera
peutics that target this system, other than preparations
based on 9-tetrahydrocannabinol and its synthetic analogues
for example, dronabinol (Elevat/Compassia/Marinol;

438 | may 2008 | volume 7

Solvay Pharmaceuticals) and nabilone (Cesamet; Eli Lilly)

which directly activate cannabinoid receptors and can
also produce central side effects. However, new genera
tions of endocannabinoid-manipulating drugs designed
almost entirely insilico or screened out of libraries of
compounds have been developed and are either being
tested in clinical trials or already on the market6,7. These
include compounds that inhibit endocannabinoid
degradation by fatty acid amide hydrolase (FAAH), for
example, SA47 and URB597, and compounds that act at
cannabinoid receptor 1 (CB1), for example, rimonabant
(Acomplia; SanofiAventis) and taranabant. These newer
drugs are thought to indirectly enhance or directly reduce
the functionality of endocannabinoid receptors only in
those tissues and cells in which there is an ongoing syn
thesis, release, action and degradation of endocannabi
noids. So, when it comes to adverse events, they should
be more selective and safer than direct agonists. This
Review article is aimed at critically discussing these new
developments, which are based on the emerging realiza
tion that alterations in the endocannabinoid system can
both counteract and participate in disease symptoms and
progress, thus opening the way to develop both endocan
nabinoid boosters (that is, indirect agonists) and curbers
(that is, antagonists) as therapeutics.

Ups and downs of endocannabinoids in disease

There are now several examples in almost each of the
major therapeutic areas of interest in which alterations in
the endocannabinoid system are associated with disease.
In particular, changes in tissue concentrations of two of
the most studied endocannabinoids anandamide and
www.nature.com/reviews/drugdisc

2008 Nature Publishing Group

REVIEWS
(2-AG) have been observed in
many disorders. These include pain and inflammation8,9;
immunological (autoimmune and allergic) disorders10;
neurological and neuropsychiatric conditions11; obesity
and metabolic12,13, and cardiovascular14 disorders; cancer15;
and gastrointestinal16 and hepatic17 disorders.
Importantly, apart from being affected by strictly
experimental variables (such as the time and method
used for tissue preparation and the analytical conditions),
the measured amounts of endocannabinoids in tissues
do not necessarily reflect extracellular, and hence can
nabinoid receptor-active, levels (intracellular 2AG, for
example, is an intermediate in phospholipid and glyceride
metabolism)4. Therefore, changes in endocannabinoid
concentrations associated with a given pathological con
dition in a certain tissue should be evaluated (and, hence,
their biological significance determined) as changes
from, rather than percent of, control physiological levels
in that tissue. In fact, it is conceivable that with mediators
that are not stored in secretory vesicles as is the case
with endocannabinoids disease-associated changes of
their tissue levels reflect corresponding changes in their
release from cells, and, therefore, in cannabinoid receptor
activation.
Intriguingly, for the same pathological condition,
there are often reports of both positive and negative
changes, and of both protective and worsening effects of
endocannabinoid receptor activation in the tissues and
organs involved in the disorder4,18. Additionally, changes
of endocannabinoid tone in the same direction often
accompany disorders with opposing symptoms; or levels
of anandamide and 2-AG change in different or even
opposing ways during the same condition. These appar
ently contradictory observations cannot be explained by
the use of different experimental protocols alone (such
as different animal models and species, different cohorts
of patients, different analytical procedures). Instead, it is
becoming increasingly clear that, within a certain tissue,
the endocannabinoid system is affected in more than just
one way by a given stressful or pathological stimulus,
depending on the nature and duration of this stimulus,
and subsequently leading to more than one functional
outcome. As this has recently been the subject of specific
comprehensive reviews4,18, I shall mention here only
some of the best-established examples (that is, arising
from work replicated in different laboratories) of this phe
nomenon, to give a general overview of endocannabinoid
regulation and function.
2arachidonoylglycerol

2-arachidonoylglycerol
(2-AG). The second moststudied endocannabinoid after
anandamide. It is thought to be
the most selective endogenous
agonist of cannabinoid 1 and 2
(CB1 and CB2) receptors, and
the one most often involved in
CB1-mediated retrograde
signalling.

Hyperphagia
A state characterized by an
exaggerated drive for food
consumption and subsequent
enhanced food-intake.

Gliosis
Proliferation of astrocytes in
damaged areas of the central
nervous system, often
associated with anoxic injury
and neuronal death, and found
in certain brain regions during
various neurodegenerative
disorders.

Eating disorders. In rodents, hypothalamic endocannabi

noid (particularly 2AG) levels first increase after ~18 hours
food deprivation and then decrease after food consump
tion19. The physiological meaning of these changes is
clear if one remembers that endocannabinoids act as
local pro-orexigenic mediators in the hypothalamus13.
This function is so important that rodents undergoing
prolonged dietary restrictions also exhibit reduced
hypothalamic 2AG levels, possibly to better cope with
the lack of food20, or, given the antilipolytic action of
CB1 receptors21, to allow the utilization of reserve energy
provided by fat. By contrast, obese rodents with defective

nature reviews | drug discovery

leptin signalling exhibit higher hypothalamic endocan

nabinoid levels that given the anorexic effects of CB1
antagonists like rimonabant (see below) and the foodintake stimulatory effects of intra-hypothalamic endo
cannabinoids22,23 are likely to contribute to hyperphagia.
In humans, blood levels of endocannabinoids are higher
both in obese female subjects with a binge-eating dis
order and in anorexic female subjects, possibly because
in both cases endocannabinoids escape from the tonic
inhibitory action of leptin24. Thus, changes in endocan
nabinoid levels seem to represent either an adaptive
response to induce the intake of food (or to cope with
the lack thereof) or a disrupted orexigenic mechanism
that participates in hyperphagia, fat accumulation and
obesity12,13. As will be discussed later, this latter phenom
enon is currently being exploited for the development of
CB1 antagonists (for example, rimonabant and tarana
bant) as new anti-obesity drugs, although there may be
some potential complications with this strategy.
Neurodegenerative disorders. In animals with lesions
of nigrostriatal dopaminergic neurons leading to
impaired striatal dopamine signalling and locomotion
similar to Parkinsons disease (PD), both increases2527
and decreases28,29 of striatal endocannabinoid levels
have been reported, even when using the same animal
model26,29. In my opinion, these opposing changes might
be explained by the impairment of possible tonic oppos
ing effects (inhibition and stimulation, respectively) of
dopamine D1 and D2 receptors on the level of endocan
nabinoids. The subsequent increased and decreased CB1
activity might contribute to locomotor impairment by
occurring at the level of medium spiny neurons partici
pating in different ways in locomotor control by the basal
ganglia (reviewed in ref.11) (FIG. 2). This might explain
why inhibitors of endocannabinoid degradation26,29 and
CB1 receptor antagonists27,28,30,31, despite their opposing
effects on CB1 activity, can both restore locomotion in
PD models.
Other examples of neurodegenerative conditions in
which endocannabinoid signalling can undergo oppos
ing changes and participate in contrary or similar ways
to the aetiology or symptoms are -amyloid-induced
cytotoxicity, multiple sclerosis and amyotrophic lateral
sclerosis. In -amyloid-induced cytotoxicity, hippo
campal 2AG levels are increased and anandamide levels
decreased 12 and 20 days following amyloid peptide
injection, respectively32. Moreover, both CB1 and CB2
receptor activation might either counteract or con
tribute to -amyloid-induced gliosis, neuronal damage
and memory-retention loss3335. These findings might
be relevant to the possible role of the endocannabi
noid system in Alzheimers disease (AD) and, together
with the finding of increased CB2 receptor and FAAH
expression in the post-mortem brain of patients with
AD36, suggest that both the symptoms and progress of
this disorder (depending on disease progression) might
be treated with either enhancers or blockers of the sys
tem. In rats and mice with acute experimental allergic
encephalomyelitis (EAE; a well-established model for
multiple sclerosis), brain endocannabinoid levels are
volume 7 | may 2008 | 439

2008 Nature Publishing Group

REVIEWS
Box 1 | Biosynthesis, action and inactivation of anandamide and 2-AG
Several pathways might exist for the formation and catabolism of anandamide and
2arachidonoylglycerol (2AG) (FIG.1). Anandamide originates from a phospholipid
precursor, Narachidonoyl-phosphatidyl-ethanolamine (NArPE), which is formed from
the Narachidoylation of phosphatidylethanolamine via both Ca2+-sensitive and
Ca2+-insensitive Nacyltransferases (NATs). NArPE is then transformed into anandamide
by four possible alternative pathways, the most direct of which (that is, direct
conversion) is catalysed by a Nacyl-phosphatidylethanolamine-selective
phosphodiesterase (NAPE-PLD)4,199. Other fatty-acid ethanolamides that do not
necessarily bind to cannabinoid receptors with high affinity and act on different
receptors for example, the anti-inflammatory compound palmitoylethanolamide
and the anorexic mediator oleoylethanolamide can also be formed through these
pathways5. 2AG, when serving as an endocannabinoid, is produced almost exclusively
by the hydrolysis of diacylglycerols (DAGs) via sn1-selective DAG lipases (DAGLs)
and . DAGL- is more abundant in adult nervous tissues, and DAGL- is more
abundant in developing nervous tissues4. However, redundancy might exist regarding
the routes through which DAGs serving as 2AG precursors are obtained, although
the one catalysed by phospholipase C seems to be the most widely used4.
After their cellular re-uptake, anandamide is metabolized via fatty acid amide
hydrolase (FAAH), and 2AG via monoacylglycerol lipase (MAGL)4. 2-AG is also
metabolized to some extent by other recently identified lipases, the -hydrolases 6
(ABH6) and 12 (ABH12), as well as FAAH200. The cellular re-uptake mechanism of
anandamide and 2-AG is yet to be characterized and is still controversial; however, it
appears to also mediate the release of denovo biosynthesized endocannabinoids164,165.
In FIG. 1, it is denoted as endocannabinoid membrane transporter (EMT), even though
it might not necessarily be uniquely mediated by plasma-membrane proteins.
Both anandamide and 2AG, possibly under conditions in which the activity of MAGL
or FAAH is suppressed, might become substrates for cyclooxygenase 2 (COX2) and
give rise to the corresponding hydroperoxy derivatives. The anandamide and 2-AG
hydroperoxy derivates can then be converted to prostaglandin ethanolamides
(prostamides) and prostaglandin glycerol esters, respectively, by various prostaglandin
synthases201. These metabolites are inactive at cannabinoid receptors but appear to act
at new binding sites, for which pharmacological, but no molecular, evidence exists201,202.
However, the physiological relevance of these pathways is still not fully understood.
Anandamide also interacts with several non-cannabinoid receptors150, the best
established of which is the transient receptor potential, vanilloid subtype 1 (TRPV1)
channel, to which the endocannabinoid binds at an intracellular site151. Recently,
anandamide and 2AG were reported by some authors203, but not by others204,
to activate GPR55, an orphan Gprotein-coupled receptor. Evidence for interaction of
endocannabinoids with peroxisome-proliferator-activating receptors (PPARs) and ,
although at high concentrations, has also been recently reviewed205. However,
cannabinoid 1 receptor (CB1) and CB2 are certainly the most-studied molecular targets
for anandamide and 2AG, which activate them with different affinity. Anandamide has
the highest affinity in both cases, whereas 2-AG has the highest efficacy in both cases.
Importantly, in the brain CB1 receptors are often expressed in presynaptic terminals so
that endocannabinoids synthesized from postsynaptic neurons can travel backwards
(retrograde signalling) and inhibit neurotransmitter release3,4,7. Apart from cannabinoid
receptor antagonists, so far specific blockers have only been developed for FAAH,
MAGL, DAGLs and the putative EMT (see main text and ref.206).

Retrograde signalling
A mechanism whereby a
chemical signal is released from
the postsynaptic neuron, travels
in the synaptic space and
activates presynaptic receptors
to modulate the release of
neurotransmitters, thereby
influencing synaptic plasticity.

either reduced37, unchanged, because of impaired bio

synthesis38, or enhanced39, although in all these cases
endocannabinoids are suggested to exert protective
effects against excitotoxicity and autoimmune reactions
via CB1 and CB2 receptors, respectively40,41. In a model
of amyotrophic lateral sclerosis, endocannabinoid levels
increase with time in both the brain and spinal cord of
superoxide dismutase 1(SOD1)-mutant mice. Moreover,
genetic knockout not only of FAAH but also of CB1
receptors, ameliorates symptoms and survival42, thus
suggesting, together with pharmacological experiments,
protective and counterprotective roles of CB2 and CB1
receptors, respectively43,44.

440 | may 2008 | volume 7

Anxiety and depression. CB1 agonists and antagonists

can produce both anxiogenic- and anxiolyticlike,
or anti-depressant- and pro-depressant-like effects.
These effects are dependent on the animal species, its
starting emotional state, the tests used to investigate
anxiety-like and depression-like behaviours, and the
dose of compounds used. Little is known on whether
endocannabinoid levels change during anxiogenicand depressive-like conditions in limbic areas of the
brain, although it is clear that, at least in rodents, the
system is activated during conditioned fear (in the amyg
dala and hippocampus45,46) and chronic stress (in the
hypothalamus47). That is, conditions that might lead to
anxiety and depression, respectively. A reduction (in the
hypothalamus47) or enhancement (in the periaqueductal
grey48) of endocannabinoid levels is also observed fol
lowing various acute stressors, and endocannabinoids
participate in some of the effects of long-term treatment
with tricyclic antidepressants49. However, inhibition of
endocannabinoid inactivation in some cases reduces
anxiety4951 and depression-like behaviour52 in rodents,
and in some cases it does not53, and CB1 receptor antago
nists can produce neurochemical effects that are similar
to those observed with antidepressant drugs (reviewed
in ref.54). Although anxiolytic and antidepressant
drugs used in the clinic are always taken as reference
compounds in these animal models, a note of caution is
always required when extrapolating the results of these
studies in which acute stressors are mostly used to
human anxiety and, particularly, depression, which are
instead chronic conditions. In this sense, the studies that
are probably more clinically relevant are those that sug
gest a protective function of CB1 receptors against the
consequences of stress55,56 and in the adaptation to new
stressful environmental conditions57, which both play a
major role in human affective disorders.
Pain and inflammation. Endocannabinoid levels in
the skin, spinal cord or peripheral nerves of rodents are
usually elevated following treatment with irritant and
inflammatory stimuli58. Endocannabinoid levels are also
increased in different models of neuropathic pain5961
and following the development of allergic contact der
matitis62. These alterations probably represent adaptive
reactions aimed at reducing pain and inflammation, as
suggested by the fact that pharmacological or genetic
inactivation of endocannabinoid degradative enzymes
usually counteract pain and inflammation6264. However,
depending on which of the tissues involved in pain trans
mission is analysed, decreases in endocannabinoid levels
have also been observed61,63. Emerging evidence also
suggests that CB1 and CB2 receptor antagonists can exert
analgesic65,66 and anti-inflammatory67,68 actions, possibly
due also to the fact that endocannabinoids can behave as
both pro- and anti-inflammatory mediators58,67.
Liver diseases and osteoporosis. These two types of disor
ders have been grouped together because both represent
typical examples of how CB1 and CB2 receptors, targeted
by elevated endocannabinoid levels, can be detrimental
and beneficial, respectively17,69.
www.nature.com/reviews/drugdisc

2008 Nature Publishing Group

REVIEWS
sn-1-arachidonate-containing
phospholipid

Phosphatidylethanolamine

Phosphatidic acid

NATs

PA phosphohydrolase
Arachidonoyl-CoA

N-arachidonoyl-phosphatidylethanolamine
PLC

Phospholipid

sn-2-arachidonatecontaining diacylglycerol

sn-1-lysophospholipid

ABH4 2

sPLA2

Phosphoanandamide

sn-2-lysophosphatidic acid

Glycerophosphoanandamide

sn-1 DAG lipases

2-lyso-N-arachidonoyl-phosphatidylethanolamide
lyso-PLD

Prostamide F2

NAPE-PLD
Phosphodiesterase
GDE1

PTPN22

COX2

Anandamide

FAAH

CB1, CB2, GPR55?

PLA1

PLC

PGF2
synthase

EMT

Prostaglandin E2 glycerol ester

PGE2
synthase

Prostamide
endoperoxide

Prostaglandin glycerol
ester endoperoxide

COX2

Arachidonate
+ ethanolamine

Arachidonate
+ glycerol

MAGL, ABH6,
ABH12, FAAH

FAAH

TRPV1

EMT

Lyso-PLC

MAGL, ABH6,
ABH12, FAAH
EMT

Anandamide

2-AG

CB1, CB2, GPR55?

EMT

Extracellular

2-AG

Figure 1 | Biosynthesis, action and inactivation of anandamide and 2arachidonoylglycerol (2-AG):

new targets for drug development. The biosynthetic pathways for anandamide and 2-AG are shown in blue, degradative
pathways are shown in pink. Thick arrows denote movement or action. For more information
please
refer
to Box
1.
Nature
Reviews
| Drug
Discovery
ABH4/6/12, -hydrolase 4/6/12; CB1/2, cannabinoid receptor 1/2; COX2, cyclooxygenase 2; DAG, diacylglycerol;
EMT, endocannabinoid membrane transporter; FAAH, fatty acid amide hydrolase; GDE1, glycerophosphodiester
phosphodiesterase 1; GPR55, G protein-coupled receptor 55; MAGL, monoacylglycerol lipase; NAPE-PLD, N-acylphosphatidylethanolamine-selective phosphodiesterase; NATs, N-acyltransferases; PA, phosphatidic acid; (s)PLA1/2,
(soluble) phospholipase A1/2; PLC, phospholipase C; PLC, phospholipase C; PLD, phospholipase D; PTPN22, protein
tyrosine phosphatase, non-receptor type 22; TRPV1, transient receptor potential, vanilloid subtype 1 receptor.
Superoxide dismutase 1
(SOD1). One of the enzymes
that converts the superoxide
anion in oxygen and hydrogen
peroxide. Gain-of-function
mutations in the Cu,Zn-SOD1
gene are implicated in
progressive motor neuron
death and paralysis in one
form of inherited amyotrophic
lateral sclerosis.

Conditioned fear
An animal defensive behaviour
(for example, immobility or
freezing) that is induced by
exposure to aversive stimuli
(for example, a non-noxious
electrical shock) coupled to a
non-aversive one (for example,
a light or an acoustic tone).
This behaviour can later
be reinstated by simply
re-exposing the animal to
the non-aversive stimulus.

Several hepatic pathological conditions, such as

fibrosis72,73 and ischaemia/
74
reperfusion injury , cause a chronic upregulation of
endocannabinoid levels. The subsequent activation of CB1
receptors then contributes to the symptoms (for example,
ectopic fat formation, fibrogenesis, hepatocyte death)70,73.
By contrast, activation of CB2 receptors, which may
become overexpressed during these disorders, often
counteracts these effects71,72,74. In the case of osteoporosis,
evidence is accumulating for a role of CB2 receptors in
enhancing endocortical osteoblast number and activity,
and restraining trabecular ostoeoclast formation75, which
would together reinforce bone structure. By contrast,
CB1 receptors reduce bone mass and contribute to
ovariectomy-induced bone loss in mice76. However, a large
part of the available data on the role of cannabinoid recep
tors in bone formation was obtained in CB1 or CB2 receptor
knockout mice with different genetic backgrounds, which
has given rise to some discrepant results75.
non-alcoholic steatosis70,71,

nature reviews | drug discovery

Cancer and gastrointestinal inflammation. Recent

data indicate a protective function of upregulated
endocannabinoid levels in cancer versus non-trans
formed cells, and in gastrointestinal tissues during
experimental colitis15,16,77,78. Both CB1 and CB2 agonists
were found to counteract carcinogenesis79, and the
growth and invasiveness of several types of cancer cells,
by modulating processes ranging from mitosis and
apoptosis to angiogenesis, cancer-cell migration and
metastasis15,77. However, examples of pro-proliferative
effects of endocannabinoids have also been reported,
although in most cases they do not seem to be mediated
by cannabinoid receptors. Indeed, the CB1 antagonist
rimonabant was also shown to inhibit cancer-cell pro
liferation by as yet unidentified molecular targets80. In
gastrointestinal inflammation, endocannabinoids exert
a homeostatic function via CB1 and CB2 receptors at the
level of visceral perception, gut motility, inflammation
and endothelial damage78,81,82. However, rimonabant
volume 7 | may 2008 | 441

2008 Nature Publishing Group

REVIEWS
b Hippocampus

CB1

CB1

Glutamate

c Prefrontal cortex

GABA

CB1

Acetylcholine

Excitotoxicity

Memory

CB1

TRPV1 CB1

Glutamate

GABA

Anxiety

d Cortex

a Periaqueductal grey

CB1

CB1

Striatum
Glutamate

TRPV1 CB1

Glutamate

CB1

Glutamate

Direct
pathway

Indirect
pathway

GABA
SnR

CB1

CB1

GPe

Antinociception
GABA

GABA

Locomotion

Non-alcoholic steatosis
Also known as non-alcoholic
fatty liver disease, this is the
inflammatory accumulation of
fat in the liver when this is not
due to excessive alcohol use.
It is related to insulin resistance.

Osteoblasts and osteoclasts

Osteoblasts are mononucleate
cells that are responsible
for bone formation. They
produce osteoid, which is
composed mainly of type I
collagen, and are responsible
for mineralization of the
osteoid matrix. Bones are
constantly being reshaped by
osteoblasts, which build bone,
for example in its endocortical
region, and osteoclasts, which
resorb bone, for example in its
trabecular region.

Figure 2 | Examples of the physiological roles of endocannabinoids and of the potential consequences
of their pathological dysregulation in central neurons. Endocannabinoids are normally produced to act only
on a selected population of neurons, usually by being released only from certain postsynaptic neurons to inhibit
Nature Reviews | Drug Discovery
neurotransmitter release from given presynaptic cannabinoid 1 receptor (CB1)-expressing neurons3,4,7. Under
pathological conditions, such as acute nociception, excitotoxicity and anxiety, endocannabinoids might act solely
on GABA (-aminobutyric acid)ergic CB1-expressing interneurons of the periaqueductal grey (a), in the former case85,
or on glutamatergic CB1-expressing terminals of hippocampal (b) and prefrontal cortex principal neurons (c), in the
latter two cases45,50,84,134. This causes descending analgesia, neuroprotection and reduced anxiety, respectively. In these
cases, enhancers of endocannabinoid action produce beneficial effects. However, with prolonged pathological stimuli,
endocannabinoid action might spread to glutamatergic or GABAergic terminals, respectively, thereby producing
opposite and undesired effects on the disorder. When this occurs, CB1 antagonists might produce beneficial effects.
In the hippocampus (b), a similar switch towards activation of CB1 receptors on cholinergic terminals might contribute
to memory retention loss during Alzheimers disease32,33. During Parkinsons disease (PD) (d), endocannabinoids might
be initially produced by neurons postsynaptic to GABAergic medium spiny neurons (MSNs) of the indirect pathway
(the former of which are located in the external layer of the globus pallidus (GPe)) to counteract GABA release. They might
also be produced from these MSNs to retrogradely counteract glutamate release from upstream corticostriatal terminals.
This would help restoring locomotion in both cases11. In an advanced phase of the disorder, endocannabinoids might
spread also to the GABAergic MSNs of the direct pathway, whose output terminals are located in the substantia nigra
reticulata (SnR). This second subset of MSNs are often very near to the aforementioned neurons of the indirect pathway,
and are also innervated by upstream corticostriatal terminals. However, unlike neurons of the indirect pathway, these
direct pathway MSNs are coupled to initiation of locomotion. Thus, spreading of endocannabinoid retrograde inhibitory
action to these neurons would produce effects opposite to those mentioned above for the indirect pathway, and
contribute to locomotor impairment11. Thus, inhibitors of endocannabinoid degradation might be beneficial in the early
phase of Parkinsons disease, whereas CB1 antagonists might be beneficial in the late phases of this disease. In some
brain areas (a,c), activation of transient receptor potential, vanilloid subtype 1 (TRPV1) channels by anandamide might
occur and produce effects on glutamate release opposite to those of CB1.

442 | may 2008 | volume 7

www.nature.com/reviews/drugdisc
2008 Nature Publishing Group

REVIEWS
was also found to exert paradoxical protective effects
via both CB1-mediated and CB1-independent effects in
gastrointestinal inflammation83.

Direct and indirect pathways

of locomotor control
Neuronal circuitries in the
basal ganglia involving medium
spiny GABA (-aminobutyric
acid)ergic neurons of the dorsal
striatum terminating onto
other GABAergic neurons in
either the substantia nigra
reticulata or external layer
of the globus pallidus, and
ultimately causing stimulation
or inhibition of locomotion,
respectively.

TRPV1
A six-transmembranedomain non-selective cation
channel that is activated by
either physical or chemical
stimuli. Stimuli include
thermosensation, sensory
transduction, taste, flowsensing, and the detection
of obnoxious and irritant
compounds.

Summary. The above examples indicate how, within its

general strategy of action of a local modulatory system,
endocannabinoid signalling can undergo several types of
changes during pathological conditions. Initial endocan
nabinoid responses to acute stressful or noxious stimuli
help to restore homeostasis with tight time-specific and
space-specific restrictions. By contrast, the continuation
of the pathological state dysregulates this system in such
a way that endocannabinoids act for a longer time, or
start activating the same receptors, or even different
receptor types, on cell populations that they were not
initially meant to target.
In neurodegenerative disorders (FIG.2), endocannabi
noids might be upregulated first to target only glutama
tergic CB1-expressing cells84 and to reduce excitotoxicity,
but then go on to act on neighbouring GABA (-amino
butyric acid)ergic neurons (which seem to require higher
concentrations of CB1 receptor ligands85), thus causing
the opposite effects. This might also be true for anxiety
and supraspinal descending antinociception, in which
evidence exists for opposing effects of CB1 receptor acti
vation in glutamatergic (anxiolytic/pro-nociceptive) or
GABAergic (anxiogenic/antinociceptive) terminals8,50,85.
In the basal ganglia (FIG.2), even a slightly spatially dys
regulated endocannabinoid signal might have dramati
cally different outcomes on nearby GABAergic afferent
neurons of the direct or indirect pathways of locomotor
control, with opposing effects on movement11. Activation
of CB2 receptors in the brain that are not strongly
expressed under normal conditions but are upregulated
in microglial and glial cells during neuroinflammatory
conditions, might cause both anti-inflammatory effects,
by reducing cytokine release, and pro-inflammatory
actions, by recruiting more immune-competent cells
from the blood when the bloodbrain barrier becomes
disrupted86,87 (FIG. 3). Also, during conditions of peripheral
hyper-reactivity, these two effects of endocannabinoids
might act in opposing ways on the outcome of the inflam
matory response, and the enhancement of CB2 receptormediated chemotaxis might produce opposing effects
depending on whether it occurs proximally or distally to
the site of inflammation (FIG.4).
Conversely, in the liver, elevated endocannabinoids
during fibrosis might first attempt to counteract the
growth of fibrogenic cells via CB2 receptors72 and then
end up contributing to this process via CB1 receptors73.
Finally, when the more promiscuous of the two moststudied endocannabinoids, anandamide, is upregulated,
other receptor types might also come into play. The
transient receptor potential, vanilloid subtype 1 (TRPV1)
receptor is activated by concentrations of anandamide
higher than those necessary to activate CB1 receptors.
Yet, this receptor can be sensitized by several noxious
and inflammatory stimuli, and participates in some
of the pernicious effects of anandamide, for example
during anxiogenic51,88, inflammatory89,90 or noxious
hypotensive9193 states and neuropathic pain94. Evidence

nature reviews | drug discovery

is emerging for the sensitizing effects of CB1 on TRPV1

receptors95,96 and this might underlie some of the para
doxical neuroprotective and analgesic/anti-inflamma
tory actions observed following chronic administration
of CB1 receptor blockers, although other mechanisms,
such as adaptive responses to prolonged CB1 receptor
inactivation, might also explain these effects.
In summary, it might be difficult to predict whether
indirect agonists or antagonists of the endocannabinoid
system will be beneficial for a given condition. There exist
to date only a few clear-cut cases in which the exclusive
use of either an enhancer or a reducer of endocannabi
noid action can be recommended for future clinical trials.
More often than not, studies in as many animal models as
possible, and then in the clinic especially with animals
and patients that are at different stages of the disorder
are required to provide a conclusive answer.

Therapeutic use of indirect agonists

Inhibitors of catabolism. At least two enzymes that
catalyse the degradation of anandamide and 2AG
(BOX1; FIG.1) have been cloned: FAAH, which recognizes
both compounds as substrates but is more selective
towards anandamide, and the monoacylglycerol lipases
(MAGLs), which are specific for 2AG. Selective and
potent inhibitors of FAAH only have been developed,
and tested invivo in several animal models of disease so
far (TABLE1). Three FAAH blockers in particular seem
promising for future clinical development: URB597
(REF. 97), arachidonoylserotonin (AA5-HT)98 and SA72
(Refs99,100).
URB597 is an irreversible inhibitor, still at the pre
clinical stage, with anxiety, depression and pain as the
most likely therapeutic targets101. It has potent analgesic
activity in models of neuropathic pain when adminis
tered orally102, and it has proved efficacious following
systemic administration in models of inflammatory
pain103 and inflammation104. However, high oral doses of
this compound (1050 mg per kg) are required to inhibit
neuropathic pain102, whereas lower intraperitoneal doses
are not effective against this type of pain103. Importantly,
the effects of oral102, but not intraplantar64, URB-597
against signs of neuropathic pain are accompanied by ele
vation of endocannabinoid levels in some of the tissues
involved in nociception. Other effects of URB597
administration include reduction of: blood pressure in
spontaneously hypertensive rats105; retinal damage after
high intraocular pressure-induced ischaemia in rats106;
morphine6-glucuronide-induced emesis in ferrets107
and lithium-induced nausea in rats108. In addition,
reduction of locomotor impairment in a PD model29
and of anxiety- and depression-like signs in animal
models50,97,109 have been observed. Importantly, and as
expected from the mechanism of action of indirect ago
nists, URB597 in not self-administered in animals110112
and has no general reinforcing action on drugs of abuse
except for the increase of alcohol self-administration in
rats113,114. Interestingly, URB597 is effective at increasing
anandamide levels in the brain, but not in the gut and
liver115, thus possibly preventing its application in gastro
intestinal and hepatic disorders.
volume 7 | may 2008 | 443

2008 Nature Publishing Group

REVIEWS
Microglia
CB2
CB2

CB2
CB2

Neuron

CB2

NK cell

Macrophage

CB2

Neuronal
inflammation
and toxicity

CB2

B cell
Bloodbrain barrier

CB2

CB2
CB2

Figure 3 | Physiological roles of endocannabinoids, and potential consequences of their dysregulation during
central neuroinflammation. Following the initial phases of several neurodegenerative disorders, cannabinoid 2 (CB2)
receptors are expressed in microglial cells, which become activated and start counteracting neuronal damage. As the
disorder progresses, the bloodbrain barrier becomes partly disrupted and blood macrophages,
lymphocytes
and
Nature BReviews
| Drug Discovery
natural killer (NK) cells start expressing CB2 receptors. The activation of these receptors stimulates the migration of these
cells into the nervous tissue and towards the endocannabinoids produced by both microglial and neuronal cells, thereby
initiating a neuroinflammatory response and causing gliosis, exaggerated microglial activity and neuronal death62,63,67,86,87.
Thus, both CB2 agonists and antagonists might be beneficial in counteracting the inflammatory consequences of
neurodegenerative disorders depending on the disease phase.

Theilers virus
Theilers murine encephalo-
myelitis virus (TMEV) is
a single-stranded RNA
picornavirus that persistently
infects the mouse central
nervous system, recently
reclassified into the cardiovirus
group. In the wild it produces
a gastrointestinal infection
that may be complicated by
concomitant infection of the
nervous system.

AA5-HT, on the other hand, increases gut endocan

nabinoid concentrations, at least following systemic
administration116, and, accordingly, was effective in an
animal model of ulcerative colitis82 and against azoxy
methane-induced precancerous lesions in the mouse
colon79. In both cases, colon endocannabinoid levels were
also increased. Furthermore, intratumour administration
of AA5-HT reduces cancer cell growth invivo in a model
of thyroid carcinoma117, inhibits both acute peripheral
and chronic neuropathic pain in rats and mice63, and,
like URB597, potentiates stress-induced analgesia118
and attenuates hyperlocomotion in an animal model of
hyperdopaminergia119. Interestingly, despite being less
potent as a FAAH inhibitor invitro98, AA5-HT often
exhibits analgesic efficacy similar to other inhibitors, a
property that might be attributed to its capability to also
antagonize TRPV1 receptors (see below)63.
Finally, less is known about SA47, which, together
with SA72, belongs to a series of carbamate FAAH
inhibitors described in patents held by SanofiAventis99.
Its very high selectivity100, however, prompts its testing
in clinical trials.
Inhibitors of cellular reuptake. Although the mechanism
of endocannabinoid cellular reuptake has not yet been
characterized from a molecular point of view and is still
controversial, increasingly selective inhibitors of this
process are being developed and found to be beneficial
in several preclinical studies (TABLE2). These compounds
can be divided into two classes: aromatic acylamide
derivatives and carbamoyl-tetrazoles.

444 | may 2008 | volume 7

Aromatic acylamide derivatives include the proto

typical AM404 (Ref.120), from which VDM11 (Ref.121),
OMDM1 and 2 (Ref.122) , UCM707 (Ref.123) and
AM1172 (Ref.124) were later developed. Carbamoyltetrazoles are developed by Eli Lilly, of which LY2183240
is the most potent invitro125,126. Apart from LY2183240,
which, however, is also a very potent FAAH inhibitor
(see below), all uptake inhibitors developed so far exhibit
low potency invitro (0.510 M). Clearly, the full char
acterization of the putative protein(s) responsible for
endocannabinoid reuptake will allow the design and
development of more potent and selective blockers of
this process.
The potential therapeutic targets for reuptake inhibi
tors can be predicted from animal studies carried out
with the above compounds, which were administered
systemically. These studies showed beneficial actions in
neuropathic and inflammatory pain in rodents125,127129;
conditioned fear130, stress-induced corticosterone release47
and anxiety- or depression-like signs109,131133 in rodents;
glutamate-mediated excitotoxicity134136, experimental
PD29,137 and amyloid-induced neurotoxicity and mem
ory retention loss in rodents32. Beneficial effects have
also been demonstrated in stress-induced suppression
of hippocampal cell proliferation138; low motor perform
ance and spasticity, respectively, in rats and mice with
EAE37,139142, but also neuroinflammation and demyeliniza
tion in Theilers virus-induced EAE143; thyroid epithelioma
growth in athymic mice117; experimental colitis in mice79;
hypertension in spontaneously hypertensive rats105; high
intraocular pressure in rabbits144; and emesis in ferrets and
www.nature.com/reviews/drugdisc

2008 Nature Publishing Group

REVIEWS

CB1

CB2

Histamine
cytokines
Keratinocyte
damage

Macrophage
CB2
Mast cell
Neuropathy,
inflammation

Oedema,
itch or pain

Neuron
TRPV1

Small vessel
plasma
extravasation

CB1
CGRP,
substance P,
NGF

Sensory
neuron
excitation

Figure 4 | Physiological roles of endocannabinoids and potential consequences of their dysregulation during
peripheral neuroinflammation. At the onset of neuropathic conditions or immune reactions against external agents,
cannabinoid 1 (CB1) receptors and transient receptor potential, vanilloid subtype 1 receptor (TRPV1) channels are
Nature Reviews | Drug Discovery
present in sensory neurons of the dorsal root ganglia, and CB2 receptors in eosinophils, mast cells and monocytes87.
Early activation of CB1 receptors counteracts the release of inflammatory and algesic peptides such as, nerve growth
factor (NGF), substance P and calcitonin gene-related peptide (CGRP). These would normally orchestrate the
inflammatory response together with inflammatory cytokines, whose release from immune cells is inhibited by CB2
receptors. However, in a later stage, CB2 and TRPV1 receptors are strongly upregulated, and 2-arachidonoylglycerol
(2AG) and anandamide might start enhancing the migration of eosinophils, mast cells and macrophages and
stimulating substance P and CGRP release, via these two receptors, respectively. This leads to neurogenic
inflammation, plasma extravasation and keratinocyte damage, and eventually to oedema, itch and pain, which can
be counteracted by TRPV1 and CB2 antagonists86,151. By contrast, CB1 agonists, endocannabinoid hydrolysis inhibitors
and, in some cases, CB1 antagonists (possibly via indirect desensitization of TRPV1 receptors) might be effective in
both the initial and late phase of these conditions.

Least shrews107,145. As in the case of URB597, OMDM2

and AM404 do not exhibit drug-reinforcing effects in
normal rodents, and hence are not likely to induce
dependence110,132,146. However, the concomitant effect
of uptake inhibitors on endocannabinoid levels, or the
intermediacy of CB1 or CB2 receptors in their effects,
were investigated in only a few of the above studies
(TABLES1,2). This leaves open the possibility that some of
these compounds might also act via mechanisms that are
different from those hypothesized. Furthermore, as most
of the inhibitors used in these studies inhibit FAAH at
high concentrations, even in those cases in which their
enhancing effect on tissue endocannabinoid levels was
demonstated32,79,105,107,117,145, it is not always clear whether
this effect is due to inhibition of the cellular uptake of
endocannabinoids alone or also of their catabolism.
Potential disadvantages of indirect agonists. Most
FAAH inhibitors (AA5-HT being perhaps the only
exception) have been designed from the chemical modi
fication of known classes of serine hydrolase inhibitors,
and therefore are likely to have off-target effects. Rapid
and accurate proteomic approaches have been devel
oped to identify possible off-targets for FAAH and
endocannabinoid reuptake inhibitors. They are based
on activity-based protein profiling, which uses active
nature reviews | drug discovery

site-directed chemical probes and allows the screening

of inhibitors against multiple enzymes in parallel. These
methods showed that URB-597 and other FAAH inhibi
tors (OL135, BMS1), as well as the inhibitor of the
putative endocannabinoid transporter LY2183240, but
not SA47 and PF750 (TABLE1), act on other carboxyl
esterases100,147149. However, even when extremely selec
tive, FAAH inhibitors are likely to prolong the lifespan
also of non-endocannabinoid substrates of this promis
cuous enzyme, including fatty acid ethanolamides such
as oleoylethanolamide and palmitoylethanolamide,
which act on non-cannabinoid receptors (BOX1). To
complicate things further, the increase of anandamide
levels induced by FAAH inhibitors might result in the
activation of non-cannabinoid receptor targets for this
compound150 and in particular of the TRPV1 recep
tor151. This has been demonstrated so far in animal
models of PD152,153; in the periaqueductal grey, follow
ing direct injection of the inhibitor in this area, with
subsequent influence on descending antinociception85;
and in mice being tested for anxiety-like behaviours154.
As TRPV1 stimulation triggers intracellular events that
are opposite to those produced by CB1 receptor activa
tion, selective FAAH inhibitors might cause effects that
are less efficacious than expected, as recently observed
with URB597 in models of anxiety-like behaviour53 and
volume 7 | may 2008 | 445

2008 Nature Publishing Group

REVIEWS
Table 1 | Endocannabinoid indirect agonists: inhibitors of FAAH and MAGL
Compound (FAAH/MAGL inhibitor)

H2N

H
N

O
O

URB-597 (FAAH)

OH

HN
NH

Potency in
vitro*

Off-targets

Effective doses

Potential indications tested in

animal models

IC50 of
109113 nM
by ABPP100,148
IC50 of 35 nM
by enzymatic
assay
with preincubation101

Carboxylesterase
inhibition100,148,208
in the liver
with IC50 of
2101,620 nM100
TRPA1 channels55

0.150.3 mg
Reduces neuropathic pain after
per kg101
systemic and oral administration102
150 mg
Inflammation104
per kg by oral
Blood pressure in spontaneously
administration102
hypertensive rats105
Glaucoma106
Emesis107-108
Locomotor impairment in a PD29
Anxiety- and depression-like
signs in animal models52,97

IC50 of
110 M by
enzymatic
assay,
with no preincubation98,208

Antagonistic
activity on TRPV1
with IC50 of
3740 nM63

0.35 mg per kg

Colitis82
Colon carcinogenesis79
Neuropathic pain63
Hyperactivity in
hyperdopaminergia119

IC50 of 2.1 nM
by ABPP148

Carboxylesterase
inhibition100,207

1.79.0 mg
per kg148,207

Neuropathic pain209

IC50 of 200 nM
by ABPP148
IC50 of 2 nM
by enzymatic
assay with preincubation210

Carboxylesterase
inhibition 100

20 mg per kg210

Neuropathic pain210

IC50 of
51,000 nM
by enzymatic
assay
with preincubation99

None so far100

2 mg per kg

Analgesia99

IC50 of
16.2595 nM
by enzymatic
assay and
depending on
pre-incubation
time207

None so far207

No study so far

No study so far

IC50of 28 M
by enzymatic
assay45

FAAH211

510 mg
per kg157

Inflammation157

O
CH3

AA-5-HT (FAAH)

N
O
N

OL-135 (FAAH)

H
N

O
O

BMS-1 (FAAH)
O

HN

NH

SA-47 (FAAH)

N
HN
N
O

PF-750 (FAAH)

H
N

URB-602 (MAGL)
*Potency in vitro, expressed here as the concentration necessary to exert half-maximal inhibition of anandamide or 2-arachidonoylglycerol hydrolysis (IC50).
Values vary according to the assay conditions, the animal species and the type of tissue or cells used to prepare the enzyme. All types of off-targets identified
so far are listed, even though the inhibitor might interact with them at concentrations higher than those required to inhibit fatty acid amide hydrolase (FAAH) or
monoacylglycerol (MAGL). Systemic (generally intraperitoneal) doses are shown unless otherwise stated. ABPP, activity-based protein profiling; PD, Parkinsons
disease; TRPA1, transient receptor potential, ankyrin-like type 1; TRPV1, transient receptor potential, vanilloid receptor subtype 1.

446 | may 2008 | volume 7

www.nature.com/reviews/drugdisc
2008 Nature Publishing Group

REVIEWS
PD153. Instead, in the case of supraspinal antinociception
and emesis, the capability of URB-597 to also indirectly
activate TRPV1 receptors seems to afford higher effi
cacy85,107. AA5-HT was recently found to also antagonize
TRPV1 receptors, a property that might explain the high
efficacy of this compound against neuropathic pain and

anxiety63,154, and thus prompt the development of further

therapeutically useful dual FAAH/TRPV1 blockers.
High concentrations of URB597, instead, were found
to activate TRPA1 channels155, which are also involved
in pain transduction, and this property might affect the
analgesic efficacy of high doses of this compound.

Table 2 | Endocannabinoid indirect agonists: inhibitors of endocannabinoid cellular uptake

Compound

Potency in
vitro*

Off-targets

Effective doses

Potential indications
tested in animal
models

OH

IC50 of
18 M120-121

TRPV1 receptor activation121,161

Fatty acid amide hydrolase158
CB1 receptors120

510 mg per kg

Neuropathic pain127-129
Anxiety132
Hypolocomotion in
Parkinsons disease137
Glaucoma144

OH

IC50 of
510 M121,158

Maybe fatty acid amide

hydrolase120,158

510 mg per kg

Thyroid carcinoma117
Colitis82
-amyloid-induced
neurotoxicity32

IC50 of
2.65.0 M122,158

None so far122,158

510 mg per kg

Multiple sclerosis
progress43
Spasticity in multiple
sclerosis141

IC50 of
0.830 M123,158

Fatty acid amide hydrolase158

CB2 receptors123

510 mg per kg

Hyperlocomotion in
Huntingtons disease142
Spasticity in multiple
sclerosis142

IC50 of
2.524 M124,158

CB1 and CB2 receptors124

Not tested yet

Not tested yet

IC50 of
0.27 nM125 or
IC50 of 15 nM126

Carboxylesterase inhibition100
Fatty acid amide hydrolase and
monoacylglycerol lipase149
Diacylglycerol lipase126

330 mg per kg125

Analgesia125

IC50 of
1.4 M140 or
IC50 of
2.8 M140

Not tested

130 mg per kg140

Spasticity in multiple
sclerosis140

N
H

AM-404
O
N
H

VDM-11
OH
O
OH

N
H

OMDM-1/OMDM-2
O
N
H
O

UCM-707

HO
H
N
O

AM1172
O
N
N
N

LY-2183240
O
R

R=

or
Cl

OH

O-3246

OH

O-3262

*Potency in vitro, expressed here as the concentration necessary to exert half-maximal inhibition of anandamide uptake (IC50). Values vary according to the assay
conditions and the type of cells used to study the cellular uptake of radiolabelled anandamide. Only off-targets for which the inhibitor interacts at concentrations
comparable with those required to inhibit the cellular uptake of anandamide are listed here. Systemic (generally intraperitoneal) doses are shown unless otherwise
stated. CB1/2, cannabinoid receptor 1/2; TRPV1, transient receptor potential vanilloid receptor subtype 1.

nature reviews | drug discovery

volume 7 | may 2008 | 447

2008 Nature Publishing Group

REVIEWS
As the MAGLs, unlike FAAH, only recognize 2AG
as a substrate, and this compound, unlike anandamide,
activates cannabinoid receptors selectively over other
targets, inhibitors of these enzymes should produce a
more specific indirect activation of these receptors.
Although, to date, no specific and potent MAGL inhibi
tor suitable for use invivo has been developed, the benefi
cial effects observed with the relatively weak URB-602
(Ref.48) against inflammatory pain156,157 raise optimistic
expectations also in this direction.
Although they inhibit the inactivation of endocannabi
noids selectively versus other FAAH substrates158, uptake
inhibitors also have off-target effects. All such compounds
described so far in the literature, except for OMDM1 and
OMDM-2 (Ref.122) and O3246 and O3262 (Ref.140),
also inhibit FAAH at concentrations that, depending on
the assay conditions, might be similar to those necessary
to inhibit anandamide reuptake158,159. AM404, which is
also a product of the invivo metabolism of paracetamol160,
potently activates TRPV1 receptors121,161. Non-endocan
nabinoid-related invitro effects of VDM11 and AM404
were also reported162,163. Furthermore, evidence exists
demonstrating that uptake inhibitors also block endocan
nabinoid release from cells164,165. Therefore, the effect of
these compounds, particularly if they are administered
before the disease-induced biosynthesis and release of
protective endocannabinoids, might be to reduce endo
cannabinoid signalling rather than enhancing it. The
timing of administration was, although for different
reasons, a crucial issue in a study of VDM-11. This com
pound, in order to be effective against amyloid-induced
neurotoxicity and loss of memory retention, had to be
administered within 3 days from the insult, whereas, if
administered later, it worsened the effect of amyloid32.
One further possible problem with indirect agonists
is represented by the high degree of redundancy of path
ways and enzymes through which endocannabinoids are
inactivated4,5 (BOX1; FIG. 1). However, because of their
nature as local mediators, it is predicted that endocan
nabinoids are biosynthesized and degraded via cellspecific pathways, and that different enzymes come into
play only in different tissues and organs, thus making it
difficult for alternative pathways to compensate for those
that have been inhibited. This might explain why FAAH
and uptake inhibitors have proved to be effective in so
many animal models of diseases.

Therapeutic use of inverse agonists/antagonists

CB1 receptor antagonists. Several CB1 receptor antago
nists have been developed to be used therapeutically and
are already being tested in several clinical trials (TABLE3).
The first such compound, rimonabant, successfully
completed four PhaseIII trials166169, and is currently on
the market in the European Union and in several other
countries. Rimonabant is used as an aid to diet and exer
cise for body-weight reduction in obese patients (body
mass index >30), or to reduce associated risks (such as
dyslipidaemia and type2 diabetes) in overweight patients
(body mass index >27). It also improves the odds of suc
cess in individuals who are willing to quit smoking170,
and is currently being tested in no less than 15 PhaseIIIb
448 | may 2008 | volume 7

and IV clinical trials (as determined from searches in

ClinicalTrials.gov database) aimed at directly assessing its
efficacy against atherogenic dyslipidaemia, cardiometa
bolic risk factors, cardiovascular events and type2 dia
betes. These ongoing trials are based on recent evidence
showing that CB1 stimulation enhances lipogenesis and
inhibits glucose and fatty-acid oxidation by acting at the
level of adipocytes, hepatocytes, endocrine pancreas and
skeletal muscle21,171,172. Moreover, an overactive endocan
nabinoid system in the visceral adipose tissue, liver and
pancreas might contribute to glucose intolerance and
dyslipidaemia in a direct way and hence independently
from its effects on food intake and weight gain21,171,172.
Accordingly, in subsets of treated and untreated obese
patients that lost the same amount of weight, rimonabant
always reduced the plasma levels of glycated haemoglobin
(a marker of type2 diabetes), and increased those of adi
ponectin (an insulin-sensitizing adipokine) significantly
more than placebo167,173. Subsequently, rimonabant and
other CB1 antagonists were found to stimulate lipid oxi
dation, glucose clearance and energy expenditure, and
hence to ameliorate the metabolic consequences of obesity
in lean and obese rodents, also independently from their
effects on food intake174176.
Other CB1 antagonists that are being tested against
obesity and related co-morbidities include tarana
bant (developed by Merck177), for which the results
of a 12-week weight-loss PhaseII trial have just been
published178, and which is soon due to complete a
PhaseIII trial. Taranabant is also being tested for
smoking cessation and for weight maintenance after
weight loss (TABLE3). Additional CB1 antagonists in the
pipeline include CP945,598 (otenabant) and BMS646256, developed by Pfizer and BristolMyers Squibb,
respectively, as anti-obesity drugs; and surinabant and
AVE1625, also developed by SanofiAventis 176,179.
BMS-646256 and AVE1625 are in clinical trials for
the treatment of nicotine-dependence and/or obesity;
AVE1625 is also being tested in patients with mild to
moderate AD, as well as against cognitive impairment
in schizophrenia (TABLE3). Indeed, as outlined above,
and apart from AD 32 and schizophrenia 180,181, there
are several other pathological conditions in which
elevated endocannabinoid levels acting at CB1 receptors
might contribute to disorder symptoms and progress.
Accordingly, thus far limited to animal models, various
CB1 antagonists were found to ameliorate locomotion,
alone or together with lDOPA, and/or to improve
lDOPA-induced dyskinesia, in rodent and primate
PD models25,27,30,31,182; improve memory retention in
amyloid-treated mice33; reduce both alcohol and
nicotine self-administration183,184 and reinstatement of
heroin185 and cocaine186 self-administration in rats. In
addition, CB1 antagonists also produce antidepressantlike effects in rodents187,188; inhibit inflammatory pain
and inflammation189; prevent bone loss in a mouse
model of osteoporosis76; retard progression of liver
fibrosis in mice73; reduce hypotension and cardiopathy
in cirrhosis190192; and inhibit breast cancer cell growth
invivo80. Therefore, preclinical bases exist now to test
CB1 receptor antagonists in clinical trials targeting at
www.nature.com/reviews/drugdisc

2008 Nature Publishing Group

REVIEWS
Table 3 | CB1 receptor antagonists/inverse agonists in clinical trials*
Compound

Cl
O
N

N NH

Cl

Potential indications tested in

animal models

Most important clinical

trial results

Ki of 2 nM
in binding
assays179

0.310 mg per kg

Reduces body weight,

dyslipidaemia, hyperglycaemia
and steatohepatitis, preserves
renal function70,174176
Nicotine and alchol abuse,
relapse of heroin and cocaine
abuse183186
Hypotension, cardiopathies,
encephalopathy and liver fibrosis
in cirrhosis73,190,192,213
Parkinsons disease25,27,3031
Alzheimers disease33
Osteoporosis76
Paradoxical beneficial effects in
inflammatory and neuropathic
pain, anxiety and depression187189

Five Phase III trials

completed; at 20 mg per
day it reduces body weight
and cardiovascular risk
factors in overweight/
obese patients with or
without dyslipidaemia
and with or without type 2
diabetes166169
Increases the odds
of quitting cigarette
smoking170
Adverse events leading to
discontinuation include
transient diarrhoea,
dizzines, nausea, signs of
depression and anxiety

Ki of 0.56 nM
in binding
assays179

3.8 mg per kg

Reduces ethanol or sucrose

consumption in mice and rats,
and food intake in fasted and
non-deprived rats179
Has an anti-alcohol profile
in selectively bred Sardinian
alcohol-preferring rats184

Phase II, smoking cessation

(NCT00432575)||
Phase II, obesity178
(NCT00239174)||
Results not yet disclosed
Adverse events not yet
disclosed

IC50 of 25
nM in a
functional
assay176

330 mg per
kg by oral
administration176

Induces lipolysis from fat tissue

and glycogenolysis from the liver
when acutely administered to
Wistar rats176

Three ongoing Phase II

trials, for the treatment
of obesity with
atherogenic dyslipidaemia
(NCT00345410)||,
Alzheimers disease
(NCT00380302)|| and
cognitive impairmenent
(NCT00439634)||
Results not yet disclosed
Adverse events not yet
disclosed

Ki of 0.13 nM
in binding
assays177

0.31 mg per
kg177

Produces significant weight

loss in rats with diet-induced
obesity177

Phase III trial ongoing

for the reduction of
body weight in obese
patients (NCT00131391)||,
and smoking cessation
(NCT00109135)||
Phase II trial in obesity
reduced dose-dependently
body weight
Adverse events leading
to discontinuation include
gastrointestinal and
psychiatric effects178

Not yet
disclosed

Not yet disclosed

One Phase II study

completed, in comparison
with sibutramine
( NCT00134199)||
Reduces body weight and
cardiovascular risk factors
in overweight/obese
patients
Adverse events not yet
disclosed

Cl

Br
O
N

Effective doses
in animals

Rimonabant

N NH

Affinity or
potency
in vitro

N
Cl

Cl

Surinabant
Cl

F
N
N

Cl

O
C

AVE-1625

O
O

N
H

Taranabant

Cl

N
N
Cl

F
F

NH
O

NH2

N
Cl

Otenabant

*A Phase II clinical trial with the CB1 antagonist/inverse agonist BMS-646256 has not been completed yet (NCT00388609). Affinity, usually expressed here as
affinity constant (Ki). Values vary according to the assay conditions and the type of cells used to study and the animal species. The Ki for cannabinoid receptor 2
(CB2) is not shown, and is usually at least 50-fold higher than the Ki values shown here. Systemic (generally intraperitoneal) doses are shown unless otherwise
stated. ||Identifying code of trial from ClincalTrials.gov database (http://clinicaltrials.gov/ct2/home).

nature reviews | drug discovery

volume 7 | may 2008 | 449

2008 Nature Publishing Group

REVIEWS
Table 4 | CB2 receptor antagonists/inverse agonists in preclinical studies
Compound

Cl
N

Affinity or
potency
in vitro*

Effective doses
in animals

Potential indications tested

in animal models

Ki of 0.6 nM
in binding
assays194

0.35 mg
per kg; 10 mg
per kg by oral
administration

Blocks the 12-O-tetradecanoylphorbol-13-acetate-induced ear

swelling196
Attenuates the recruitment of
eosinophils and ear swelling in
chronic contact dermatitis induced
by repeated challenge with
oxazolone58
Inhibits leukocyte recruitment in
a murine model of delayed-type
hypersensitivity67

N
NH

SR-144528 O

O
HN

Ki of
0.110 mg
0.381.55 nM per kg by oral
in binding
administration
assays193

Inhibits carrageenin-induced
mouse paw oedema193
Inhibits dinitrofluorobenzeneinduced ear swelling195
Alleviates dermatitis symptoms
in a mouse model of contact
dermatitis58

Ki of 0.3 nM
(human)67,86

This compound and its congeners

are potent modulators of immune
cell mobility in vivo and of bone
damage in antigen-induced
mono-articular arthritis and
in experimental autoimmune
encephalomyelitis67,86

HN
O

JTE-907

O
O

S O
O
S
O

O
HN

310 mg per
kg by oral
administration

Sch.336
*Affinity, usually expressed here as affinity constant (Ki). Values vary according to the assay conditions and the type of cells used to
study and the animal species. The Ki for cannabinoid receptor 1 (CB1) is not shown, and is usually at least 50-fold higher than the Ki
values shown here.

least some of these disorders. However, although these

compounds exhibit an overall relatively safe profile
of side effects, some concern is raised by the fact that
their chronic use in obese patients is associated with
increased risk of developing signs of depression173,178.
CB2 receptor antagonists. Several CB2 antagonists/
inverse agonists have also been developed (TABLE4). The
following compounds in particular received attention
by virtue of their anti-inflammatory effects: JTE907
(Ref.193) SR144528 (Ref.194), and Sch.336, Sch.036 and
Sch.414319 (Refs67,86). Iwamura etal.193 showed that
oral administration of JTE907 and SR144528 inhibit
carrageenin-induced paw oedema in mice. The same
group also reported that orally administered JTE907
and SR144528 significantly inhibit dinitrofluoroben
zene-induced ear swelling195. SR144528 blocked 12Otetradecanoylphorbol13-acetate-induced ear swelling,
while decreasing leukotriene B4 production and neu
trophil infiltration into the ear196. Ear swelling was also
suppressed by administration of SR144528 in oxazoloneinduced contact dermatitis58. This CB2 antagonist also
suppressed pro-inflammatory cytokine mRNA expres
sion and attenuated eosinophil recruitment into the
450 | may 2008 | volume 7

treated ear, which suggests that CB2 antagonists have

high potential against allergic states such as contact
dermatitis. When the effects of Sch.414319 were inves
tigated on different types of inflammatory disorders, the
compound reduced bone damage in antigen-induced
monoarticular arthritis and counteracted the clinical
signs of EAE in the Lewis rat strain. The authors sug
gested that these effects can all result from the control
of inflammatory cell migration, possibly involving
lplastin phosphorylation86. Despite these promising
preclinical results, however, clinical trials of JTE907,
SR144528 and Sch.336 or Sch.414319 in allergic contact
dermatitis or autoimmune disorders do not seem to be
in the pipeline.
Potential disadvantages of cannabinoid receptor
antagonists. Similar to indirect agonists, antagonists
should exert pharmacological effects preferentially in
those tissues and organs where the receptor they target
is tonically activated by locally acting endogenous ago
nists. However, all CB1 and CB2 receptor antagonists
tested so far in preclinical and clinical experimentation
behave in invitro assays of functional activity not
as neutral antagonists but as inverse agonists. That is,
www.nature.com/reviews/drugdisc

2008 Nature Publishing Group

REVIEWS
they appear to destabilize a receptor conformation
that preferentially couples to the Gprotein also in the
absence of endogenous agonists. This might raise the
possibility that these compounds exert effects (opposite
to those of agonists) also on organs where endocannabi
noid levels are not aberrantly elevated, with subsequent
loss of specificity. Yet, this does not often seem to be
the case, as cannabinoid receptor antagonists have been
shown so far to be significantly more efficacious invivo
under physiological or pathological conditions in which
local endocannabinoid levels are elevated (see above).
However, it is almost impossible to determine whether
or not inverse agonism occurs invivo. The finding of a
pharmacological profile similar to those of cannabinoid
receptor antagonists also with inhibitors of endocan
nabinoid biosynthesis might be used in the future as
indirect evidence that tonic activity of receptors occurs
because of endogenous ligands and not of pre-coupling
to G-proteins. Unfortunately, the only endocannabinoid
biosynthesis inhibitors developed so far, although selec
tive, are not suitable for systemic invivo use197.
Another potential limitation of cannabinoid recep
tor antagonists targeting a certain disorder to whose
symptoms or progress dysregulated endocannabinoids
might contribute, is that they might interfere also with
other concomitant disorders in which endocannabinoids
might instead be playing a protective effect. Indeed, the
use of CB2 receptor antagonists against allergic contact
dermatitis or multiple sclerosis might be limited by
the fact that endocannabinoids do not only contribute
to inflammation with chemotaxic effects, but can also
inhibit inflammatory cytokine release198, to the point
that both CB2/CB1 agonists produce beneficial effects
in the same model of allergic dermatitis in which CB 2
antagonists do58,62. Likewise, while being used against
obesity and metabolic disorders, CB1 antagonists might
interfere with endocannabinoid-mediated adaptation to
new stressful conditions, thus explaining the anxiogenic
and pro-depressant effects observed in obese patients
treated with these compounds166169,178. However, one
does not expect cannabinoid receptor antagonists to
be capable perse of inducing adverse events, but only
to interfere with some of the protective effects of endo
cannabinoids that might arise when these compounds
are produced denovo during a new acute pathological
condition. It is also clear that specific cannabinoid recep
tor antagonists produce fewer adverse events than those
expected from the many protective actions postulated for
endocannabinoids. This is because, in the pathological
conditions that they target, CB1 or CB2 sometimes play
opposing functions and hence blockade of only one of

1.

2.

3.

Russo, E. & Guy, G. W. A tale of two cannabinoids:

the therapeutic rationale for combining
tetrahydrocannabinol and cannabidiol.
Med. Hypotheses 66, 234246 (2006).
Mechoulam, R. Discovery of endocannabinoids and
some random thoughts on their possible roles in
neuroprotection and aggression. Prostaglandins
Leukot. Essent. Fatty Acids 66, 9399 (2002).
Pertwee, R. G. Cannabinoid pharmacology: the first 66
years. Br. J. Pharmacol. 147 (Suppl. 1), 163171
(2006).

4.
5.

6.

these receptors does not prevent the beneficial effects of

the other. Finally, in view of the high specificity of endo
cannabinoid protective role, the tissue-specific distribu
tion/accumulation/targeting of non-receptor-saturating
doses of these drugs might prevent them from acting in
the wrong place. In this respect, it is interesting to note
that doses of taranabant that are clinically efficacious at
reducing body weight in humans only occupy 3040%
of human brain CB1 receptors178.

Concluding remarks
Perhaps no other signalling system discovered during
the past 15 years is raising as many expectations for the
development of new therapeutic drugs, encompass
ing such a variety of pathological conditions, targeting
so many different organs and tissues, and using such a
wide range of potential strategies for treatment, as the
endocannabinoid system. The articles published over
the past 4 years have shown that both direct or indirect
agonists and antagonists of cannabinoid receptors can
produce beneficial effects, sometimes even in the same
condition, in agreement with the pleiotropic homeo
static function of this system and with its unfortunate
tendency to become dysregulated. While this might look
attractive for drug developers, it can also be a drawback
when the time comes to go from preclinical to clinical
studies. Nevertheless, we know from the experience of
dronabinol, nabilone, tetrahydrocannabinol/cannabidiol,
rimonabant and of several other compounds currently
in the clinical pipeline, that good drugs can be made by
either increasing or decreasing the tone of the endocan
nabinoid system, while keeping at bay most side effects.
Therefore, perhaps, while we wait to understand more
about the physiological function of this system, it is just a
matter of trying what is the best endocannabinoid-based
drug for a certain condition. One, however, needs to be
cautious as the possibility exists that, when someone
is being treated with a selective enhancer or blocker of
endocannabinoid action for long periods of time, a comorbidity develops that is worsened or counteracted by
such action, respectively, and this might cause problems.
However, experience has shown that collateral events of
these drugs can be controlled, in both clinical trials and
in the medical practice, by using the appropriate dosage,
selecting the right patient and making the most of clinical
surveillance. In conclusion, for those who are engaged in
developing new therapeutics by targeting the endocan
nabinoid system, this task can be described by Giuseppe
Verdis definition (in La Traviata) of love as Croce e
Delizia: a series of painstaking, and sometimes frustrating,
efforts alternating with immense gratifications.

Di Marzo, V. & Petrosino, S. Endocannabinoids and

the regulation of their levels in health and disease.
Curr. Opin. Lipidol. 18, 129140 (2007).
Alexander, S. P. & Kendall, D. A. The complications
of promiscuity: endocannabinoid action and
metabolism. Br. J. Pharmacol. 152, 602623
(2007).
Di Marzo, V., Bifulco, M. & De Petrocellis, L.
The endocannabinoid system and its therapeutic
exploitation. Nature Rev. Drug Discov. 3,
771784 (2004).

nature reviews | drug discovery

7.
8.
9.

Piomelli, D. The endocannabinoid system: a drug

discovery perspective. Curr. Opin. Investig. Drugs.
6, 672679 (2005).
Hohmann, A. G. & Suplita, R. L. 2nd. Endo
cannabinoid mechanisms of pain modulation. AAPS J.
8, e693e708 (2006).
Jhaveri, M. D., Richardson, D. & Chapman, V.
Endocannabinoid metabolism and uptake:
novel targets for neuropathic and inflammatory
pain. Br. J. Pharmacol. 152, 624632
(2007).

volume 7 | may 2008 | 451

2008 Nature Publishing Group

REVIEWS
10. Lambert, D. M. Allergic contact dermatitis and
the endocannabinoid system: from mechanisms
to skin care. ChemMedChem 2, 17011702 (2007).
11. Bisogno, T. & Di Marzo, V. Short- and long-term
plasticity of the endocannabinoid system in
neuropsychiatric and neurological disorders.
Pharmacol. Res. 56, 428442 (2007).
12. Cota, D. CB1 receptors: emerging evidence for central
and peripheral mechanisms that regulate energy
balance, metabolism, and cardiovascular health.
Diabetes Metab. Res. Rev. 23, 507517 (2006).
13. Matias, I. & Di Marzo, V. Endocannabinoids and the
control of energy balance. Trends Endocrinol. Metab.
18, 2737 (2007).
14. Ashton, J. C. & Smith, P. F. Cannabinoids and
cardiovascular disease: the outlook for clinical
treatments. Curr. Vasc. Pharmacol. 5, 175185
(2007).
15. Bifulco, M., Laezza, C., Gazzerro, P. & Pentimalli, F.
Endocannabinoids as emerging suppressors of
angiogenesis and tumor invasion. Oncol. Rep. 17,
813816 (2007).
16. Storr, M. A. & Sharkey, K. A. The endocannabinoid
system and gutbrain signalling. Curr. Opin.
Pharmacol. 7, 575582 (2007).
17. Mallat, A., Teixeira-Clerc, F., Deveaux, V. &
Lotersztajn, S. Cannabinoid receptors as new targets
of antifibrosing strategies during chronic liver
diseases. Expert Opin. Ther. Targets 11, 403409
(2007).
18. Pacher, P., Batkai, S.. & Kunos, G. The endocannabinoid
system as an emerging target of pharmacotherapy.
Pharmacol. Rev. 58, 389462 (2006).
19. Kirkham, T. C., Williams, C. M., Fezza, F. &
Di Marzo, V. Endocannabinoid levels in rat limbic
forebrain and hypothalamus in relation to fasting,
feeding and satiation: stimulation of eating by
2arachidonoyl glycerol. Br. J. Pharmacol. 136,
550557 (2002).
20. Hanus, L. et al. Short-term fasting and prolonged
semistarvation have opposite effects on 2AG levels in
mouse brain. Brain Res. 983, 144151 (2003).
21. Matias, I. etal. Regulation, function, and
dysregulation of endocannabinoids in models of
adipose and -pancreatic cells and in obesity and
hyperglycemia. J. Clin. Endocrinol. Metab. 91,
31713180 (2006).
22. Jamshidi, N. & Taylor, D. A. Anandamide
administration into the ventromedial hypothalamus
stimulates appetite in rats. Br. J. Pharmacol. 134,
11511154 (2001).
23. Di Marzo, V. etal. Leptin-regulated endocannabinoids
are involved in maintaining food intake. Nature 410,
822825 (2001).
24. Monteleone, P. et al. Blood levels of the
endocannabinoid anandamide are increased in
anorexia nervosa and in binge-eating disorder, but not
in bulimia nervosa. Neuropsychopharmacology 30,
12161221 (2005).
25. Di Marzo, V., Hill, M. P., Bisogno, T., Crossman, A. R.
& Brotchie, J. M. Enhanced levels of endogenous
cannabinoids in the globus pallidus are associated
with a reduction in movement in an animal model of
Parkinsons disease. FASEB J. 14, 14321438
(2000).
26. Gubellini P etal. Experimental parkinsonism alters
endocannabinoid degradation: implications for
striatal glutamatergic transmission. J. Neurosci. 22,
69006907 (2002).
27. van der Stelt, M. et al. A role for endocannabinoids in
the generation of parkinsonism and levodopainduced dyskinesia in MPTP-lesioned non-human
primate models of Parkinsons disease. FASEB J. 19,
11401142 (2005).
28. Ferrer, B., Asbrock, N., Kathuria, S., Piomelli, D. &
Giuffrida, A. Effects of levodopa on endocannabinoid
levels in rat basal ganglia: implications for the
treatment of levodopa-induced dyskinesias. Eur. J.
Neurosci. 18, 16071614 (2003).
29. Kreitzer, A. C. & Malenka, R. C. Endocannabinoidmediated rescue of striatal LTD and motor deficits in
Parkinsons disease models. Nature 445, 643647
(2007).
30. Fernandez-Espejo, E. etal. Cannabinoid CB1
antagonists possess antiparkinsonian efficacy only in
rats with very severe nigral lesion in experimental
parkinsonism. Neurobiol. Dis. 18, 591601 (2005).
31. Gonzalez S. et al. Effects of rimonabant, a selective
cannabinoid CB1 receptor antagonist, in a rat model
of Parkinsons disease. Brain Res. 10731074,
209219 (2006).

32. van der Stelt, M. etal. Endocannabinoids and

amyloidinduced neurotoxicity invivo: effect of
pharmacological elevation of endocannabinoid levels.
Cell. Mol. Life Sci. 63, 14101424 (2006).
33. Mazzola, C., Micale, V. & Drago, F. Amnesia induced
by -amyloid fragments is counteracted by
cannabinoid CB1 receptor blockade. Eur. J.
Pharmacol. 477, 219225 (2003).
34. Esposito, G. et al. Opposing control of cannabinoid
receptor stimulation on amyloidinduced reactive
gliosis: invitro and invivo evidence. J. Pharmacol.
Exp. Ther. 322, 11441152 (2007).
35. Ramirez, B. G., Blazquez, C., Gomez del Pulgar, T.,
Guzman, M. & de Ceballos, M.L. Prevention of
Alzheimers disease pathology by cannabinoids:
neuroprotection mediated by blockade of microglial
activation. J. Neurosci. 25, 19041913 (2005).
36. Benito, C. et al. Cannabinoid CB2 receptors and fatty
acid amide hydrolase are selectively overexpressed in
neuritic plaque-associated glia in Alzheimers disease
brains. J. Neurosci. 23, 1113611141 (2003).
37. Cabranes, A. etal. Decreased endocannabinoid levels
in the brain and beneficial effects of agents activating
cannabinoid and/or vanilloid receptors in a rat model
of multiple sclerosis. Neurobiol. Dis. 20, 207217
(2005).
38. Witting, A. etal. Experimental autoimmune
encephalomyelitis disrupts endocannabinoidmediated neuroprotection. Proc. Natl Acad. Sci. USA
103, 63626367 (2006).
39. Centonze, D. etal., The endocannabinoid system is
dysregulated in multiple sclerosis and in
experimental autoimmune encephalomyelitis. Brain
130, 25432553 (2007).
40. Maresz, K. etal. Direct suppression of CNS
autoimmune inflammation via the cannabinoid
receptor CB1 on neurons and CB2 on autoreactive
Tcells. Nature Med. 13, 492497 (2007).
The first study that really clarifies the distinct
beneficial roles of the two cannabinoid receptor
types in autoimmune neuroinflammation.
41. Croxford, J. L. etal. Cannabinoid-mediated
neuroprotection, not immunosuppression, may be
more relevant to multiple sclerosis. J. Neuroimmunol.
193, 120129 (2007).
42. Bilsland, L. G. etal. Increasing cannabinoid levels by
pharmacological and genetic manipulation delay
disease progression in SOD1 mice. FASEB J. 20,
10031005 (2006).
43. Kim, K., Moore, D. H., Makriyannis, A. & Abood, M. E.
AM1241, a cannabinoid CB2 receptor selective
compound, delays disease progression in a mouse
model of amyotrophic lateral sclerosis. Eur. J.
Pharmacol. 542, 100105 (2006).
44. Shoemaker, J. L., Seely, K. A., Reed, R. L., Crow, J. P.
& Prather, P. L. The CB2 cannabinoid agonist
AM1241 prolongs survival in a transgenic mouse
model of amyotrophic lateral sclerosis when initiated
at symptom onset. J. Neurochem. 101, 8798
(2007).
45. Marsicano, G. etal. The endogenous cannabinoid
system controls extinction of aversive memories.
Nature 418, 530534 (2002).
Perhaps the first example of the site- and timespecific activation of the endocannabinoid system
following a stressor and with a protective function
in the adaptation to new environmental conditions.
46. Kamprath, K. etal. Cannabinoid CB1 receptor
mediates fear extinction via habituation-like
processes. J. Neurosci. 26, 66776686 (2006).
47. Patel, S., Roelke, C. T., Rademacher, D.J., Cullinan, W. E.
& Hillard, C. J. Endocannabinoid signaling negatively
modulates stress-induced activation of the
hypothalamicpituitaryadrenal axis. Endocrinology
145, 54315438 (2004).
An important study showing how the
endocannabinoids and CB1 receptors can
be involved in the control of stress.
48. Hohmann, A. G. etal. An endocannabinoid
mechanism for stress-induced analgesia. Nature 435,
11081112 (2005).
Another elegant example of the site- and
time-specific activation of the endocannabinoid
system following a stressor, and of the therapeutic
exploitation of specific inhibitors of
endocannabinoid degradation.
49. Hill, M. N. et al. Involvement of the endocannabinoid
system in the ability of long-term tricyclic
antidepressant treatment to suppress stress-induced
activation of the hypothalamicpituitaryadrenal axis.
Neuropsychopharmacology 31, 25912599 (2006).

452 | may 2008 | volume 7

50. Moreira, F. A., Kaiser, N., Monory, K. & Lutz, B.

Reduced anxiety-like behaviour induced by genetic
and pharmacological inhibition of the
endocannabinoid-degrading enzyme fatty acid amide
hydrolase (FAAH) is mediated by CB1 receptors.
Neuropharmacology 54, 141150 (2007).
51. Rubino, T. etal. Role in anxiety behavior of the
endocannabinoid system in the prefrontal cortex.
Cereb Cortex 5 Oct 2007 (doi:10.1093/cercor/
bhm161).
52. Gobbi, G. etal. Antidepressant-like activity and
modulation of brain monoaminergic transmission by
blockade of anandamide hydrolysis. Proc. Natl Acad.
Sci. USA 102, 1862018625 (2005).
53. Naidu, P. S. et al. Evaluation of fatty acid amide
hydrolase inhibition in murine models of emotionality.
Psychopharmacology (Berl.) 192, 6170 (2007).
54. Degroot, A. & Nomikos, G. G. Invivo neurochemical
effects induced by changes in endocannabinoid
neurotransmission. Curr. Opin. Pharmacol. 7, 6268
(2007).
55. Steiner, M. A. etal. Impaired cannabinoid receptor
type1 signaling interferes with stress-coping behavior
in mice. Pharmacogenomics J. 7 Aug 2007
(doi:10.1038/sj.tpj.6500466).
56. Rademacher, D. J. et al.Effects of acute and repeated
restraint stress on endocannabinoid content in the
amygdala, ventral striatum, and medial prefrontal
cortex in mice. Neuropharmacology 54, 108116
(2008).
57. Lutz, B. The endocannabinoid system and extinction
learning. Mol. Neurobiol. 36, 92101 (2007).
58. Oka, S. etal. Involvement of the cannabinoid CB2
receptor and its endogenous ligand
2arachidonoylglycerol in oxazolone-induced contact
dermatitis in mice. J. Immunol. 177, 87968805
(2006).
59. Mitrirattanakul, S. etal. Site-specific increases in
peripheral cannabinoid receptors and their
endogenous ligands in a model of neuropathic pain.
Pain 126, 102114 (2006).
60. Petrosino, S. etal. Changes in spinal and supraspinal
endocannabinoid levels in neuropathic rats.
Neuropharmacology 52, 415422 (2007).
61. Agrawal, N. etal. Cannabinoids mediate analgesia
largely via peripheral type1 cannabinoid receptors in
nociceptors. Nature Neurosci. 10, 870879 (2007).
62. Karsak, M. etal. Attenuation of allergic contact
dermatitis through the endocannabinoid system.
Science 316, 14941497 (2007).
The first study that suggests that inhibition of
endocannabinoid degradation can be used against
allergic contact dermatitis.
63. Maione, S. etal. Analgesic actions of Narachidonoyl
serotonin, a fatty acid amide hydrolase inhibitor with
antagonistic activity at vanilloid TRPV1 receptors.
Br. J. Pharmacol. 150, 766781 (2007).
64. Jhaveri, M. D., Richardson, D., Kendall, D. A.,
Barrett, D. A. & Chapman, V. Analgesic effects of fatty
acid amide hydrolase inhibition in a rat model of
neuropathic pain. J. Neurosci. 26, 1331813327
(2006).
65. Costa, B. et al. Effect of the cannabinoid CB1 receptor
antagonist, SR141716, on nociceptive response and
nerve demyelination in rodents with chronic
constriction injury of the sciatic nerve. Pain 116, 5261
(2005).
66. Saez-Cassanelli, J. L., Fontanella, G. H., DelgadoGarcia, J. M. & Carrion, A. M. Functional blockage of
the cannabinoid receptor type1 evokes a
opiatedependent analgesia. J. Neurochem. 103,
26292639 (2007).
67. Lunn, C. A. etal. A novel cannabinoid peripheral
cannabinoid receptor-selective inverse agonist blocks
leukocyte recruitment invivo. J. Pharmacol. Exp.
Ther. 316, 780788 (2006).
68. Croci, T. & Zarini, E. Effect of the cannabinoid CB1
receptor antagonist rimonabant on nociceptive
responses and adjuvant-induced arthritis in obese
and lean rats. Br. J. Pharmacol. 150, 559566
(2007).
69. Kunos, G., Osei-Hyiaman, D., Batkai, S. & Gao, B.
Cannabinoids hurt, heal in cirrhosis. Nature Med. 12,
608610 (2006).
70. Gary-Bobo, M. etal. Rimonabant reduces obesityassociated hepatic steatosis and features of metabolic
syndrome in obese Zucker fa/fa rats. Hepatology 46,
122129 (2007).
71. Mendez-Sanchez, N. etal. Endocannabinoid receptor
CB2 in nonalcoholic fatty liver disease. Liver Int. 27,
215219 (2007).

www.nature.com/reviews/drugdisc
2008 Nature Publishing Group

REVIEWS
72. Julien, B. etal. Antifibrogenic role of the cannabinoid
receptor CB2 in the liver. Gastroenterology 128,
742755 (2005).
73. Teixeira-Clerc, F. etal. CB1 cannabinoid receptor
antagonism: a new strategy for the treatment of liver
fibrosis. Nature Med. 12, 671676 (2006).
An important study showing how CB1 and CB2
receptors play opposing roles in liver fibrosis, and
how CB1 antagonists might be used to treat this
disorder.
74. Batkai, S. etal. Cannabinoid2 receptor mediates
protection against hepatic ischemia/reperfusion injury.
FASEB J. 21, 17881800 (2007).
75. Ofek, O. etal. Peripheral cannabinoid receptor, CB2,
regulates bone mass. Proc. Natl Acad. Sci. USA 103,
696701 (2006).
76. Idris, A. I. etal. Regulation of bone mass, bone loss
and osteoclast activity by cannabinoid receptors.
Nature Med. 11, 774779 (2005).
77. Guzman, M. Cannabinoids: potential anticancer
agents. Nature Rev. Cancer 3, 745755 (2003).
78. Di Marzo, V. & Izzo, A. A. Endocannabinoid
overactivity and intestinal inflammation. Gut 55,
13731376 (2006).
79. Izzo, A. A. etal. Increased endocannabinoid levels
reduce the development of precancerous lesions in the
mouse colon. J. Mol. Med. 86, 8998 (2008).
80. Sarnataro, D. etal. The cannabinoid CB1 receptor
antagonist rimonabant (SR141716) inhibits human
breast cancer cell proliferation through a lipid raftmediated mechanism. Mol. Pharmacol. 70,
12981306 (2006).
81. Massa, F. etal. The endogenous cannabinoid system
protects against colonic inflammation. J. Clin. Invest.
113, 12021209 (2004).
82. DArgenio, G. et al. Up-regulation of anandamide
levels as an endogenous mechanism and a
pharmacological strategy to limit colon inflammation.
FASEB J. 20, 568570 (2006).
The first study showing that endocannabinoid
enhancers can be as effective as well-established
drugs against experimental colitis.
83. Croci, T., Landi, M., Galzin, A. M. & Marini, P. Role of
cannabinoid CB1 receptors and tumor necrosis factor-
in the gut and systemic anti-inflammatory activity of
SR 141716 (rimonabant) in rodents. Br. J. Pharmacol.
140, 115122 (2003).
84. Monory, K. etal., The endocannabinoid system
controls key epileptogenic circuits in the hippocampus.
Neuron 51, 455466 (2006).
An elegant study showing how the
endocannabinoids and CB1 receptors work in a
neuron-type-specific way to dampen excitotoxicity.
85. Maione, S. etal. Elevation of endocannabinoid levels
in the ventrolateral periaqueductal grey through
inhibition of fatty acid amide hydrolase affects
descending nociceptive pathways via both
cannabinoid receptor type1 and transient receptor
potential vanilloid type1 receptors. J. Pharmacol.
Exp. Ther. 316, 969982 (2006).
86. Lunn, C. A. etal. Biology and therapeutic potential of
cannabinoid CB2 receptor inverse agonists. Br. J.
Pharmacol. 153, 226239 (2007).
87. Miller, A. M. & Stella, N. CB2 receptor-mediated
migration of immune cells: it can go either way. Br. J.
Pharmacol. 153, 299308 (2007).
88. Marsch, R. etal. Reduced anxiety, conditioned fear,
and hippocampal long-term potentiation in transient
receptor potential vanilloid type1 receptor-deficient
mice. J. Neurosci. 27, 832839 (2007).
89. Dinis, P. etal. Anandamide-evoked activation of
vanilloid receptor 1 contributes to the development of
bladder hyperreflexia and nociceptive transmission to
spinal dorsal horn neurons in cystitis. J. Neurosci. 24,
1125311263 (2004).
90. Singh Tahim, A., Santha, P. & Nagy, I. Inflammatory
mediators convert anandamide into a potent activator
of the vanilloid type1 transient receptor potential
receptor in nociceptive primary sensory neurons.
Neuroscience 136, 539548 (2005).
91. Orliac, M. L., Peroni, R., Celuch, S. M. & AdlerGraschinsky, E. Potentiation of anandamide effects in
mesenteric beds isolated from endotoxemic rats.
J. Pharmacol. Exp. Ther. 304, 179184 (2003).
92. Domenicali, M. etal. Increased anandamide induced
relaxation in mesenteric arteries of cirrhotic rats:
role of cannabinoid and vanilloid receptors. Gut 54,
522527 (2005).
93. Moezi, L. et al. Anandamide mediates hyperdynamic
circulation in cirrhotic rats via CB1 and VR1 receptors.
Br. J. Pharmacol. 149, 898908 (2006).

94. Horvath, G., Kekesi, G., Nagy, E. & Benedek, G.

The role of TRPV1 receptors in the antinociceptive
effect of anandamide at spinal level. Pain 134,
277284 (2007).
95. Hermann, H. et al. Dual effect of cannabinoid CB1
receptor stimulation on a vanilloid VR1 receptormediated response. Cell. Mol. Life Sci. 60, 607616
(2003).
96. Evans, R. M., Scott, R. H. & Ross, R. A. Chronic
exposure of sensory neurones to increased levels of
nerve growth factor modulates CB1/TRPV1 receptor
crosstalk. Br. J. Pharmacol. 152, 404413 (2007).
97. Kathuria, S. etal. Modulation of anxiety through
blockade of anandamide hydrolysis. Nature Med. 9,
7681 (2003).
First study demonstrating the potential use of
endocannabinoid boosters against anxiety.
98. Bisogno, T. etal. Arachidonoylserotonin and other
novel inhibitors of fatty acid amide hydrolase.
Biochem. Biophys. Res. Commun. 248, 515522
(1998).
99. Abouabdellah, A. et al. Derivatives of dioxane2alkyl
carbamates, preparation thereof and application
thereof in therapeutics. Patent US20050182130
A1(2005).
100. Zhang, D. etal. Fatty acid amide hydrolase inhibitors
display broad selectivity and inhibit multiple
carboxylesterases as off-targets. Neuropharmacology
52, 10951105 (2007).
101. Piomelli, D. etal. Pharmacological profile of the
selective FAAH inhibitor KDS4103 (URB597).
CNS Drug Rev. 12, 2138 (2006).
102. Russo, R. etal. The fatty acid amide hydrolase
inhibitor URB597 (cyclohexylcarbamic acid
3carbamoylbiphenyl3-yl ester) reduces neuropathic
pain after oral administration in mice. J. Pharmacol.
Exp. Ther. 322, 236242 (2007).
103. Jayamanne, A. et al. Actions of the FAAH inhibitor
URB597 in neuropathic and inflammatory chronic pain
models. Br. J. Pharmacol. 147, 281288 (2006).
104. Holt S., Comelli, F., Costa, B. & Fowler, C. J. Inhibitors
of fatty acid amide hydrolase reduce carrageenaninduced hind paw inflammation in pentobarbitaltreated mice: comparison with indomethacin and
possible involvement of cannabinoid receptors.
Br. J. Pharmacol. 146, 467476 (2005).
105. Batkai, S. etal. Endocannabinoids acting at
cannabinoid1 receptors regulate cardiovascular
function in hypertension. Circulation. 110, 19962002
(2004).
A complete study showing how endocannabinoids
and CB1 receptors are produced to play a
protective function during certain types of
hypertension, and how this might be treated by
endocannabinoid boosters.
106. Nucci, C. etal. Involvement of the endocannabinoid
system in retinal damage after high intraocular
pressure-induced ischemia in rats. Invest. Ophthalmol.
Vis. Sci. 48, 29973004 (2007).
First complete study suggesting the potential use
of endocannabinoid enhancers against glaucoma.
107. Sharkey, K. A. etal. Arvanil, anandamide and
Narachidonoyldopamine (NADA) inhibit emesis
through cannabinoid CB1 and vanilloid TRPV1
receptors in the ferret. Eur. J. Neurosci. 25,
27732782 (2007).
108. Cross-Mellor, S. K., Ossenkopp, K. P., Piomelli, D. &
Parker, L. A. Effects of the FAAH inhibitor, URB597,
and anandamide on lithium-induced taste reactivity
responses: a measure of nausea in the rat.
Psychopharmacology (Berl.) 190, 135143 (2007).
109. Patel, S. & Hillard, C. J. Pharmacological evaluation
of cannabinoid receptor ligands in a mouse model of
anxiety: further evidence for an anxiolytic role for
endogenous cannabinoid signaling. J. Pharmacol.
Exp. Ther. 318, 304311 (2006).
110. Vlachou, S., Nomikos, G. G. & Panagis, G. Effects of
endocannabinoid neurotransmission modulators on
brain stimulation reward. Psychopharmacology (Berl.)
188, 293305 (2006).
111. Solinas, M., Justinova, Z., Goldberg, S. R. & Tanda, G.
Anandamide administration alone and after inhibition
of fatty acid amide hydrolase (FAAH) increases
dopamine levels in the nucleus accumbens shell in
rats. J. Neurochem. 98, 408419 (2006).
112. Solinas, M. etal. The endogenous cannabinoid
anandamide produces 9tetrahydrocannabinol-like
discriminative and neurochemical effects that are
enhanced by inhibition of fatty acid amide hydrolase
but not by inhibition of anandamide transport.
J. Pharmacol. Exp. Ther. 321, 370380 (2007).

nature reviews | drug discovery

113. Hansson, A. C. etal. Genetic impairment of

frontocortical endocannabinoid degradation and high
alcohol preference. Neuropsychopharmacology 32,
117126 (2007).
114. Vinod, K. Y., Sanguino, E., Yalamanchili, R.,
Manzanares, J. & Hungund, B. L. Manipulation of fatty
acid amide hydrolase functional activity alters
sensitivity and dependence to ethanol. J. Neurochem.
104, 233243 (2007).
115. Fegley, D. etal. Characterization of the fatty acid
amide hydrolase inhibitor cyclohexyl carbamic acid
3carbamoylbiphenyl3yl ester (URB597): effects
on anandamide and oleoylethanolamide deactivation.
J. Pharmacol. Exp. Ther. 313, 352358 (2005).
116. Capasso, R. etal. Fatty acid amide hydrolase controls
mouse intestinal motility invivo. Gastroenterology
129, 941951 (2005).
117. Bifulco, M. et al. A new strategy to block tumor
growth by inhibiting endocannabinoid inactivation.
FASEB J. 18, 16061608 (2004).
The first study showing that endocannabinoid
enhancers can retard cancer growth in vivo.
118. Suplita, R. L. 2nd, Farthing, J. N., Gutierrez, T. &
Hohmann, A. G. Inhibition of fatty-acid amide
hydrolase enhances cannabinoid stress-induced
analgesia: sites of action in the dorsolateral
periaqueductal gray and rostral ventromedial medulla.
Neuropharmacology 49, 12011209 (2005).
119. Tzavara, E. T. etal. Endocannabinoids activate
transient receptor potential vanilloid 1 receptors to
reduce hyperdopaminergia-related hyperactivity:
therapeutic implications. Biol. Psychiatry. 59,
508515 (2006).
120. Beltramo, M. et al. Functional role of high-affinity
anandamide transport, as revealed by selective
inhibition. Science 277, 10941097 (1997).
121. De Petrocellis, L., Bisogno, T., Davis, J. B., Pertwee, R. G.
& Di Marzo, V. Overlap between the ligand recognition
properties of the anandamide transporter and the VR1
vanilloid receptor: inhibitors of anandamide uptake
with negligible capsaicin-like activity. FEBS Lett. 483,
5256 (2000).
122. Ortar G., Ligresti, A., De Petrocellis, L., Morera, E. &
Di Marzo, V. Novel selective and metabolically stable
inhibitors of anandamide cellular uptake. Biochem.
Pharmacol. 65, 14731481 (2003).
123. Lopez-Rodriguez, M. etal. Design, synthesis and
biological evaluation of new endocannabinoid
transporter inhibitors. Eur. J. Med. Chem. 38,
403412 (2003).
124. Fegley, D. etal. Anandamide transport is
independent of fatty-acid amide hydrolase activity
and is blocked by the hydrolysis-resistant inhibitor
AM1172. Proc. Natl Acad. Sci. USA 101,
87568761 (2004).
125. Moore, S. A. etal. Identification of a high-affinity
binding site involved in the transport of
endocannabinoids. Proc. Natl Acad. Sci. USA 102,
1785217857 (2005).
126. Ortar, G. etal. Carbamoyl tetrazoles as inhibitors of
endocannabinoid inactivation: a critical revisitation.
Eur. J. Med. Chem. 43, 6272 (2008).
127. Costa, B. etal. AM404, an inhibitor of anandamide
uptake, prevents pain behaviour and modulates
cytokine and apoptotic pathways in a rat model of
neuropathic pain. Br. J. Pharmacol. 148, 10221032
(2006).
128. La Rana, G. etal. Modulation of neuropathic and
inflammatory pain by the endocannabinoid
transport inhibitor AM404 [N(4-hydroxyphenyl)
eicosa5,8,11,14-tetraenamide]. J. Pharmacol.
Exp. Ther. 317, 13651371 (2006).
129. Mitchell, V. A., Greenwood, R., Jayamanne, A. &
Vaughan, C. W. Actions of the endocannabinoid
transport inhibitor AM404 in neuropathic and
inflammatory pain models. Clin. Exp. Pharmacol.
Physiol. 34, 11861190 (2007).
130. Chhatwal, J. P., Davis, M., Maguschak, K. A.
& Ressler, K. J. Enhancing cannabinoid
neurotransmission augments the extinction of
conditioned fear. Neuropsychopharmacology 30,
516524 (2005).
131. Hill, M. N. & Gorzalka, B. B. Pharmacological
enhancement of cannabinoid CB1 receptor activity
elicits an antidepressant-like response in the rat
forced swim test. Eur. Neuropsychopharmacology 15,
593599 (2005).
132. Bortolato, M. etal. Anxiolytic-like properties of
the anandamide transport inhibitor AM404.
Neuropsychopharmacology 31, 26522659
(2006).

volume 7 | may 2008 | 453

2008 Nature Publishing Group

REVIEWS
133. Braida, D., Limonta, V., Malabarba, L., Zani, A.
& Sala, M. 5HT1A receptors are involved in the
anxiolytic effect of 9-tetrahydrocannabinol and AM
404, the anandamide transport inhibitor, in SpragueDawley rats. Eur. J. Pharmacol. 555, 156163
(2007).
134. Marsicano, G. etal. CB1 cannabinoid receptors and
on-demand defense against excitotoxicity. Science
302, 8488 (2003).
135. Karanian, D. A., Brown, Q. B., Makriyannis, A.,
Kosten, T. A. & Bahr, B. A. Dual modulation of
endocannabinoid transport and fatty acid amide
hydrolase protects against excitotoxicity. J. Neurosci.
25, 78137820 (2005).
136. Wettschureck, N. etal. Forebrain-specific
inactivation of Gq/G11 family G proteins results
in age-dependent epilepsy and impaired
endocannabinoid formation. Mol. Cell Biol. 26,
58885894 (2006).
137. Fernandez-Espejo, E. etal. Experimental
parkinsonism alters anandamide precursor synthesis,
and functional deficits are improved by AM404:
a modulator of endocannabinoid function.
Neuropsychopharmacology 29, 11341142 (2004).
138. Hill, M. N., Kambo, J. S., Sun, J. C., Gorzalka, B. B.
& Galea, L. A. Endocannabinoids modulate stressinduced suppression of hippocampal cell proliferation
and activation of defensive behaviours. Eur. J.
Neurosci. 24, 18451849 (2006).
139. Baker, D. etal. Endocannabinoids control spasticity in
a multiple sclerosis model. FASEB J. 15, 300302
(2001).
140. Ligresti, A. etal. New potent and selective inhibitors
of anandamide reuptake with antispastic activity in a
mouse model of multiple sclerosis. Br. J. Pharmacol.
147, 8391 (2006).
141. de Lago, E. etal. Invivo pharmacological actions of
two novel inhibitors of anandamide cellular uptake.
Eur. J. Pharmacol. 484, 249257 (2004).
142. de Lago, E. etal. UCM707, an inhibitor of the
anandamide uptake, behaves as a symptom control
agent in models of Huntingtons disease and multiple
sclerosis, but fails to delay/arrest the progression of
different motor-related disorders. Eur.
Neuropsychopharmacol. 16, 718 (2006).
143. Mestre, L. etal. Pharmacological modulation of the
endocannabinoid system in a viral model of multiple
sclerosis. J. Neurochem. 92, 13271339 (2005).
144. Laine, K. et al. Effects of topical anandamide-transport
inhibitors, AM404 and olvanil, on intraocular pressure
in normotensive rabbits. Pharm. Res. 18, 494499
(2001).
145. Darmani, N. A. et al. Cisplatin increases brain
2arachidonoylglycerol (2-AG) and concomitantly
reduces intestinal 2AG and anandamide levels in the
Least shrew. Neuropharmacology 49, 502513
(2005).
146. Solinas, M. et al. Cannabinoid agonists but not
inhibitors of endogenous cannabinoid transport or
metabolism enhance the reinforcing efficacy of heroin
in rats. Neuropsychopharmacology 30, 20462057
(2005).
147. Leung, D., Hardouin, C., Boger, D. L. & Cravatt, B. F.
Discovering potent and selective reversible inhibitors
of enzymes in complex proteomes. Nature Biotech.
21, 687691 (2003).
148. Lichtman, A. H. et al. Reversible inhibitors of fatty acid
amide hydrolase that promote analgesia: evidence for
an unprecedented combination of potency and
selectivity. J. Pharmacol. Exp. Ther. 311, 441448
(2004).
149. Alexander, J. P. & Cravatt, B. F. The putative
endocannabinoid transport blocker LY2183240 is a
potent inhibitor of FAAH and several other brain serine
hydrolases. J. Am. Chem. Soc. 128, 96999704
(2006).
150. Oz, M. Receptor-independent effects of endo
cannabinoids on ion channels. Curr. Pharm. Des. 12,
227239 (2006).
151. Starowicz, K., Nigam, S. & Di Marzo, V. Biochemistry
and pharmacology of endovanilloids. Pharmacol. Ther.
114, 1333 (2007).
152. Lee, J., Di Marzo, V. & Brotchie, J. M. A role for
vanilloid receptor 1 (TRPV1) and endocannabinnoid
signalling in the regulation of spontaneous and lDOPA
induced locomotion in normal and reserpine-treated
rats. Neuropharmacology 51, 557565 (2006).
153. Morgese, M. G., Cassano, T., Cuomo, V. & Giuffrida, A.
Anti-dyskinetic effects of cannabinoids in a rat model
of Parkinsons disease: Role of CB1 and TRPV1
receptors. Exp. Neurol. 208, 110119 (2007).

154. Micale, V. etal. Anxiety-like behaviors in wild-type and

dopamine D3 receptor knockout (KO) mice: role of
endocannabinoids and vanilloid TRPV1 receptors.
Program No. 501.25/JJ25. Society for Neuroscience
Meeting Planner web site [online], <http://www.
abstractsonline.com/viewer/?mkey=%7BFF8B70E5%
2DB7F9%2D4D07%2DA58A%2DC1068FDE9D25%
7D> (2007).
155. Niforatos, W. etal. Activation of TRPA1 channels
by the fatty acid amide hydrolase inhibitor
3carbamoylbiphenyl3-yl cyclohexylcarbamate
(URB597). Mol. Pharmacol. 71, 12091216
(2007).
156. Guindon, J., Desroches, J. & Beaulieu, P.
The antinociceptive effects of intraplantar injections of
2arachidonoyl glycerol are mediated by cannabinoid
CB2 receptors. Br. J. Pharmacol. 150, 693701
(2007).
157. Comelli, F., Giagnoni, G., Bettoni, I., Colleoni, M. &
Costa, B. The inhibition of monoacylglycerol lipase by
URB602 showed an anti-inflammatory and antinociceptive effect in a murine model of acute
inflammation. Br. J. Pharmacol. 152, 787794
(2007).
158. Hillard, C. J., Shi, L., Tuniki, V. R., Falck, J. R. &
Campbell, W. B. Studies of anandamide accumulation
inhibitors in cerebellar granule neurons: comparison
to inhibition of fatty acid amide hydrolase. J. Mol.
Neurosci. 33, 1824 (2007).
159. Dickason-Chesterfield, A. K. etal. Pharmacological
characterization of endocannabinoid transport and
fatty acid amide hydrolase inhibitors. Cell. Mol.
Neurobiol. 26, 407423 (2006).
160. Hogestatt, E. D. etal. Conversion of acetaminophen to
the bioactive Nacylphenolamine AM404 via fatty acid
amide hydrolase-dependent arachidonic acid
conjugation in the nervous system. J. Biol. Chem. 280,
3140531412 (2005).
161. Zygmunt, P. M., Chuang, H., Movahed, P., Julius, D. &
Hogestatt, E. D. The anandamide transport inhibitor
AM404 activates vanilloid receptors. Eur. J.
Pharmacol. 396, 3942 (2000).
162. Jonsson, K. O. et al. AM404 and VDM 11 nonspecifically inhibit C6 glioma cell proliferation at
concentrations used to block the cellular accumulation
of the endocannabinoid anandamide. Arch. Toxicol.
77, 201207 (2003).
163. Kelley, B. G. & Thayer, S. A. Anandamide transport
inhibitor AM404 and structurally related compounds
inhibit synaptic transmission between rat
hippocampal neurons in culture independent of
cannabinoid CB1 receptors. Eur. J. Pharmacol. 496,
3339 (2004).
164. Ronesi, J., Gerdeman, G. L. & Lovinger, D. M.
Disruption of endocannabinoid release and striatal
long-term depression by postsynaptic blockade of
endocannabinoid membrane transport. J. Neurosci.
24, 16731679 (2004).
165. Ligresti, A. etal. Further evidence for the existence
of a specific process for the membrane transport of
anandamide. Biochem. J. 380, 265272
(2004).
166. Van Gaal, L. F., Rissanen, A. M., Scheen, A. J.,
Ziegler, O. & Rossner, S. Effects of the cannabinoid1
receptor blocker rimonabant on weight reduction and
cardiovascular risk factors in overweight patients:
1year experience from the RIO-Europe study. Lancet
365, 13891397 (2005).
167. Despres, J. P., Golay, A. & Sjostrom, L. Effects of
rimonabant on metabolic risk factors in overweight
patients with dyslipidemia. N. Engl. J. Med. 353,
21212134 (2005).
A report on an important clinical trial
demonstrating the potential use of rimonabant in
obese patients not just as a therapeutic aid to
reduce body weight.
168. Pi-Sunyer, F. X., Aronne, L. J., Heshmati, H. M.,
Devin, J. & Rosenstock, J. Effect of rimonabant,
a cannabinoid1 receptor blocker, on weight and
cardiometabolic risk factors in overweight or obese
patients: RIO-North America: a randomized controlled
trial. JAMA 295, 761775 (2006).
169. Scheen, A. J., Finer, N., Hollander, P., Jensen, M. D. &
Van Gaal, L. F. Efficacy and tolerability of rimonabant in
overweight or obese patients with type2 diabetes: a
randomised controlled study. Lancet 368, 16601672
(2006).
170. Cahill, K. & Ussher, M. Cannabinoid type1 receptor
antagonists (rimonabant) for smoking cessation.
Cochrane Database Syst. Rev. 4, CD005353
(2007).

454 | may 2008 | volume 7

171. Osei-Hyiaman, D. etal. Endocannabinoid activation at

hepatic CB1 receptors stimulates fatty acid synthesis
and contributes to diet-induced obesity. J. Clin. Invest.
115, 12981305 (2005).
Another important example of a possible role of
CB1 receptors in the aetiology of a liver disorder,
and of a potential therapeutic use of CB1
antagonists.
172. Pagano, C. etal. The endogenous cannabinoid
system stimulates glucose uptake in human fat cells
via PI3-kinase and calcium-dependent mechanisms.
J. Clin. Endocrinol. Metab. 92, 48104819 (2007).
173. Van Gaal, L., Pi-Sunyer, X,, Despres, J. P., McCarthy, C.
& Scheen, A. Efficacy and safety of rimonabant for
improvement of multiple cardiometabolic risk factors
in overweight/obese patients: pooled 1year data from
the Rimonabant in Obesity (RIO) program. Diabetes
Care. 31, S229S240 (2008).
174. Poirier, B. etal. The anti-obesity effect of rimonabant
is associated with an improved serum lipid profile.
Diabetes Obes. Metab. 7, 6572 (2005).
175. Janiak, P. etal. Blockade of cannabinoid CB1 receptors
improves renal function, metabolic profile, and
increased survival of obese Zucker rats. Kidney Int. 72,
13451357 (2007).
176. Herling, A. W. etal. CB1 receptor antagonist
AVE1625 affects primarily metabolic parameters
independently of reduced food intake in Wistar rats.
Am. J. Physiol. Endocrinol. Metab. 293, e826e832
(2007).
177. Fong, T. M. etal. Antiobesity efficacy of a novel
cannabinoid1 receptor inverse agonist, N[(1S,2S)
3(4-chlorophenyl)2(3-cyanophenyl)1
methylpropyl]2methyl2[[5-(trifluoromethyl)pyridin
2yl]oxy]propanamide (MK0364), in rodents.
J. Pharmacol. Exp. Ther. 321, 10131022 (2007).
178. Addy, C. etal. Safety, tolerability, pharmacokinetics,
and pharmacodynamic properties of taranabant,
a novel selective cannabinoid1 receptor inverse
agonist, for the treatment of obesity: results from a
double-blind, placebo-controlled, single oral dose
study in healthy volunteers. J. Clin. Pharmacol.
7 Feb 2008 (doi:10.1177/0091270008314467).
179. Rinaldi-Carmona, M. etal. SR147778 [5(4bromophenyl)1(2,4-dichlorophenyl)4ethylN
(1-piperidinyl)1Hpyrazole3carboxamide],
a new potent and selective antagonist of the CB1
cannabinoid receptor: biochemical and
pharmacological characterization. J. Pharmacol.
Exp. Ther. 310, 905914 (2004).
180. De Marchi, N. et al. Endocannabinoid signalling in the
blood of patients with schizophrenia. Lipids Health
Dis. 2, 5 (2003).
181. Giuffrida, A. etal. Cerebrospinal anandamide
levels are elevated in acute schizophrenia and are
inversely correlated with psychotic symptoms.
Neuropsychopharmacology 29, 21082114 (2004).
182. Cao, X. etal. Blockade of cannabinoid type1
receptors augments the antiparkinsonian action
of levodopa without affecting dyskinesias in
1methyl4phenyl1,2,3, 6tetrahydro
pyridinetreated rhesus monkeys. J. Pharmacol.
Exp. Ther. 323, 318326 (2007).
183. Cohen, C., Perrault, G., Griebel, G. & Soubrie, P.
Nicotine-associated cues maintain nicotine-seeking
behavior in rats several weeks after nicotine
withdrawal: reversal by the cannabinoid (CB1)
receptor antagonist, rimonabant (SR141716).
Neuropsychopharmacology 30, 145155 (2005).
184. Gessa, G. L., Serra, S., Vacca, G., Carai, M. A. &
Colombo, G. Suppressing effect of the cannabinoid
CB1 receptor antagonist, SR147778, on alcohol
intake and motivational properties of alcohol in
alcohol-preferring sP rats. Alcohol Alcohol. 40, 4653
(2005).
185. Spano, M. S. et al. CB1 receptor agonist and heroin,
but not cocaine, reinstate cannabinoid-seeking
behaviour in the rat. Br. J. Pharmacol. 143, 343350
(2004).
186. De Vries, T. J. etal. A cannabinoid mechanism
in relapse to cocaine seeking. Nature Med. 7,
11511154 (2001).
187. Griebel, G., Stemmelin, J. & Scatton, B. Effects of
the cannabinoid CB1 receptor antagonist rimonabant
in models of emotional reactivity in rodents.
Biol. Psychiatry. 57, 261267 (2005).
188. Steiner, M. A. etal., Antidepressant-like behavioral
effects of impaired cannabinoid receptor type1
signaling coincide with exaggerated corticosterone
secretion in mice. Psychoneuroendocrinology 33,
5467 (2008).

www.nature.com/reviews/drugdisc
2008 Nature Publishing Group

REVIEWS
189. Costa, B. Rimonabant: more than an anti-obesity
drug? Br. J. Pharmacol. 150, 535537 (2007).
190. Batkai, S. etal. Endocannabinoids acting at vascular
CB1 receptors mediate the vasodilated state in
advanced liver cirrhosis. Nature Med. 7, 827832
(2001).
191. Gaskari, S. A. et al. Role of endocannabinoids in the
pathogenesis of cirrhotic cardiomyopathy in bile ductligated rats. Br. J. Pharmacol. 146, 315323 (2005).
192. Batkai, S. et al. Endocannabinoids acting at CB1
receptors mediate the cardiac contractile dysfunction
invivo in cirrhotic rats. Am. J. Physiol. Heart Circ.
Physiol. 293, H1689H1695 (2007).
193. Iwamura, H., Suzuki, H., Ueda, Y., Kaya, T. & Inaba, T.
Invitro and invivo pharmacological characterization
of JTE907, a novel selective ligand for cannabinoid
CB2 receptor. J. Pharmacol. Exp. Ther. 296, 420425
(2001).
194. Rinaldi-Carmona M etal. SR 144528, the first potent
and selective antagonist of the CB2 cannabinoid
receptor. J. Pharmacol. Exp. Ther. 284, 644650
(1998).
195. Ueda, Y., Miyagawa, N., Matsui, T,, Kaya, T, &
Iwamura, H. Involvement of cannabinoid CB2
receptor-mediated response and efficacy of
cannabinoid CB2 receptor inverse agonist, JTE907,
in cutaneous inflammation in mice. Eur. J. Pharmacol.
520, 164171 (2005).
196. Oka, S. etal. Evidence for the involvement of the
cannabinoid CB2 receptor and its endogenous ligand
2arachidonoylglycerol in 12Otetradecanoyl
phorbol13acetate-induced acute inflammation in
mouse ear. J. Biol. Chem. 280, 1848818497
(2005).
Perhaps the most convincing study of the possible
use of CB2 antagonists against inflammation.

197. Bisogno, T. etal. Development of the first potent and

specific inhibitors of endocannabinoid biosynthesis.
Biochim. Biophys. Acta 1761, 205212 (2006).
198. Klein, T. W. Cannabinoid-based drugs as antiinflammatory therapeutics. Nature Rev. Immunol. 5,
400411 (2005).
199. Liu, J. etal., Multiple pathways involved in the
biosynthesis of anandamide. Neuropharmacology 54,
17 (2008).
200. Blankman, J. L., Simon, G. M. & Cravatt, B. F.
A comprehensive profile of brain enzymes that
hydrolyze the endocannabinoid 2arachidonoylglycerol.
Chem. Biol. 14, 13471356 (2007).
201. Rouzer, C. A. & Marnett, L. J. Non-redundant functions
of cyclooxygenases: oxygenation of endocannabinoids.
J. Biol. Chem. 283, 80658069 (2008).
202. Woodward, D. F., Liang, Y. & Krauss, A. H. Prostamides
(prostaglandin-ethanolamides) and their pharmacology.
Br. J. Pharmacol. 153, 410419 (2008).
203. Ryberg, E. etal. The orphan receptor GPR55 is a
novel cannabinoid receptor. Br. J. Pharmacol. 152,
10921101 (2007).
204. Oka, S., Nakajima, K., Yamashita, A,, Kishimoto, S.
& Sugiura, T. Identification of GPR55 as a
lysophosphatidylinositol receptor. Biochem. Biophys.
Res. Commun. 362, 928934 (2007).
205. OSullivan, S. E. Cannabinoids go nuclear: evidence
for activation of peroxisome proliferator-activated
receptors. Br. J. Pharmacol. 152, 576582 (2007).
206. Saario, S. M. etal. Characterization of the sulfhydrylsensitive site in the enzyme responsible for hydrolysis
of 2arachidonoylglycerol in rat cerebellar
membranes. Chem. Biol. 12, 649656 (2005).
207. Ahn, K. etal., Novel mechanistic class of fatty acid
amide hydrolase inhibitors with remarkable
selectivity. Biochemistry. 46, 1301913030 (2007).

nature reviews | drug discovery

208. Fowler, C. J. etal. Inhibition of fatty acid

amidohydrolase, the enzyme responsible for the
metabolism of the endocannabinoid anandamide,
by analogues of arachidonoyl-serotonin. J. Enzyme
Inhib. Med. Chem. 18, 225231 (2003).
209. Chang, L. etal. Inhibition of fatty acid amide
hydrolase produces analgesia by multiple
mechanisms. Br. J. Pharmacol. 148, 102113
(2006).
210. Sit, S. Y. etal. Novel inhibitors of fatty acid amide
hydrolase. Bioorg. Med. Chem. Lett. 17, 32873291
(2007).
211. Vandevoorde, S. etal. Lack of selectivity of URB602
for 2oleoylglycerol compared to anandamide
hydrolysis invitro. Br. J. Pharmacol. 150, 186191
(2007).

Acknowledgements

The author wishes to thank A. Ligresti and S. Petrosino at the

Endocannabinoid Research Group for their help with preparing
the manuscript. This article is dedicated to the memory ofthe
highly esteemed scientists and friends Professor Santosh
Nigam, who passed away on 2 October 2007, and Professor
Michael J. Walker, who passed away 5 January 2008.

Competing interests statement

The author declares competing financial interests: see web

version for details.

DATABASES
IUPHAR Receptor Database:
http://www.iuphar-db.org/index.jsp
CB1 | CB2
All links are active in the online pdf

volume 7 | may 2008 | 455

2008 Nature Publishing Group

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.