The incidence of obesity and its associated disorders
is increasing markedly worldwide. Obesity predis-
poses individuals to an increased risk of developing
many diseases, including atherosclerosis, diabetes, non-
alcoholic fatty liver disease, certain cancers and some
immune-mediated disorders, such as asthma1–3. In addi-
tion to these associations between obesity and disease,
research in the past few years has identified important
pathways that link metabolism with the immune system
and vice versa. Many of these interactions between the
metabolic and immune systems seem to be orchestrated
by a complex network of soluble mediators derived from
immune cells and adipocytes (fat cells)1.
In mammals, adipose tissue occurs in two forms:
white adipose tissue and brown adipose tissue. Most
adipose tissue in mammals is white adipose tissue and
this is thought to be the site of energy storage. By con-
trast, brown adipose tissue is found mainly in human
neonates and is important for the regulation of body
temperature through non-shivering thermogenesis. In
addition to adipocytes, which are the most abundant cell
type in white adipose tissue, adipose tissue also contains
pre-adipocytes (which are adipocytes that have not yet
been loaded with lipids), endothelial cells, fibroblasts,
leukocytes and, most importantly, macrophages (FIG. 1).
These macrophages are bone-marrow derived and the
number of these cells present in white adipose tissue
correlates directly with obesity.
Adipose tissue is no longer considered to be an inert
tissue functioning solely as an energy store, but is emerg-
ing as an important factor in the regulation of many
pathological processes. Various products of adipose
tissue have been characterized, and some of the soluble
factors produced by this tissue are known as adipocy-
tokines. The term adipocytokine is used to describe
certain cytokines that are mainly produced by adipose
tissue, although it is important to note that they are not
all exclusively derived from this organ. Adiponectin,
leptin, resistin and visfatin are adipocytokines and are
thought to provide an important link between obesity,
insulin resistance and related inflammatory disorders1–6.
Adiponectin and leptin are the most abundant adipocy-
tokines produced by adipocytes. Various other products
of adipose tissue that have been characterized include:
certain cytokines, such as tumour-necrosis factor (TNF),
interleukin-6 (IL-6), IL-1 and CC-chemokine ligand 2
(CCL2; also known as MCP1); mediators of the clotting
process, such as plasminogen-activator inhibitor type 1;
and certain complement factors1,2 (FIG. 1). These products
have well-known roles in the immune system, and
although some of them are also produced by adipocytes,
they are not normally considered to be adipocytokines;
nonetheless, they have important roles at the interface
between the immune and metabolic systems.
Obesity is associated with a chronic inflammatory
response, which is characterized by abnormal cytokine
Christian Doppler Research
Laboratory for Gut
Department of Medicine,
Innsbruck Medical University,
6020 Innsbruck, Austria.
Correspondence to H.T.
22 September 2006
A chronic disorder of the
arterial wall characterized
by endothelial damage that
gradually induces deposits
of cholesterol, cellular debris,
calcium and other substances.
These deposits finally lead to
plaque formation and arterial
Adipocytokines: mediators linking
adipose tissue, inflammation and
Herbert Tilg and Alexander R. Moschen
Abstract | There has been much effort recently to define the role of adipocytokines, which
are soluble mediators derived mainly from adipocytes (fat cells), in the interaction between
adipose tissue, inflammation and immunity. The adipocytokines adiponectin and leptin have
emerged as the most abundant adipocyte products, thereby redefining adipose tissue as a
key component not only of the endocrine system, but also of the immune system. Indeed,
as we discuss here, several adipocytokines have a central role in the regulation of insulin
resistance, as well as many aspects of inflammation and immunity. Other adipocytokines,
such as visfatin, have only recently been identified. Understanding this rapidly growing family
of mainly adipocyte-derived mediators might be of importance in the development of new
therapies for obesity-associated diseases.
772 | OCTOBER 2006 | VOLUME 6
© 2006 Nature Publishing Group
Lean adipose tissue Obese adipose tissue
Pro-inflammatory cytokines and chemokines
Complement factors are
components of the
complement system. Activation
of these factors, which involves
proteolytic cleavage of serum
and cell-surface glycoproteins,
leads to the formation of a
terminal cell-lytic complex
inside the cell membrane of a
target cell. Complement
fragments such as C3a and
C5a have important pro-
inflammatory properties, such
as vasodilation, chemotaxis
production, increased synthesis of acute-phase reactants,
such as C-reactive protein (CRP), and the activation
of pro-inflammatory signalling pathways1. Although
there is no doubt that pro-inflammatory pathways are
activated in the adipose tissue itself in cases of obes-
ity, the relative contribution of adipocytes as a source
of the circulating and systemically active cytokines, adipo-
cytokines and chemokines remains unclear. The adipose
tissue of obese individuals also contains a large number
of macrophages, which are an additional source of solu-
ble mediators in the adipose tissue7,8 (FIG. 1). However,
although macrophages in adipose tissue seem to be the
main source of TNF, adipocytes contribute almost one
third of the IL-6 concentration in the circulation of
patients who are obese9. CCL2, produced by adipocytes,
has recently been identified as a potential factor contrib-
uting to macrophage infiltration into adipose tissue10.
Once macrophages are present and active in the adipose
tissue, they, together with adipocytes and other cell types
present in the adipose tissue, might perpetuate a vicious
cycle of macrophage recruitment and production of
In humans, adipocytokines function as hormones
to influence energy homeostasis and to regulate neuro-
endocrine function. As cytokines, they affect immune
functions and inflammatory processes throughout
the body. The field of adipocytokines has attracted
tremendous interest recently and the knowledge that
has accumulated might lead to the development of new
therapeutics. Here, we provide an overview of recent
advances in our view of the role of adipocytokines in
inflammation and immunity.
Insulin resistance: an inflammatory disease
Systemic chronic inflammation has been proposed to
have an important role in the pathogenesis of obesity-
related insulin resistance1,11. Biomarkers of inflamma-
tion, such as TNF, IL-6 and CRP, are present at increased
concentrations in individuals who are insulin resistant
and obese, and these biomarkers predict the develop-
ment of type 2 diabetes mellitus and cardiovascular
diseases (BOX 1).
The first link between obesity, an increase in expres-
sion of the pro-inflammatory cytokine TNF, and insulin
action was reported by Hotamisligil and colleagues11.
Their findings in rodents showed that adipocytes
directly express TNF and led to the concept of a role
for inflammation in obesity. These observations were
paralleled by human studies showing increased TNF
expression in the adipose tissue of individuals who were
obese, and decreased TNF expression after weight loss12.
Evidence supporting a key role for TNF in obesity-
related insulin resistance came from studies showing
that ob/ob mice (leptin-deficient mice with evidence of
insulin resistance) that were also deficient for TNF or
TNF receptors (TNFRs) had improved insulin sensi-
tivity in diet-induced obesity compared with TNF- and
TNFR-sufficient ob/ob mice13.
In searching for the mechanisms involved in
inflamma tion-induced insulin resistance, Yuan and
co-workers identified the inhibitor of nuclear factor-κB
(NF-κB) kinase-β (IKKβ) pathway of NF-κB activation
as a mediator of TNF-induced insulin resistance14. They
showed that overexpression of IKKβ in a human embry-
onic kidney cell line (the HEK293 cell line) attenuated
insulin signalling, and that ob/ob mice expressing only
one copy of the gene encoding IKKβ (Ikbkb) were pro-
tected against the development of insulin resistance14.
The JUN N-terminal kinase (JNK) family of serine/
threonine protein kinases, which are activated by many
inflammatory stimuli — including TNF and ligation of
Toll-like receptors (TLRs) — are also important regula-
tors of insulin resistance in mouse models of obesity15.
In both genetic and dietary animal models of obesity,
JNK activity is increased in the liver, muscle and adipose
tissue, and loss of JNK1 prevents insulin resistance15.
Another important mechanism involved in insulin
resistance is endoplasmic-reticulum (ER) stress16. ER stress
leads to suppression of signalling through the insulin
receptor in a rat hepatocyte cell line through activation
of JNKs and the subsequent serine phosphorylation of
insulin receptor substrate 1 (IRS1), which is one of the
main mediators of insulin signalling and thereby controls
Figure 1 | Adipose tissue: cellular components and molecules synthesized.
Expansion of the adipose tissue during weight gain leads to the recruitment of
macrophages through various signals, which might include chemokines synthesized
by adipocytes, such as CC-chemokine ligand 2 (CCL2). These macrophages are found
mainly around apoptotic adipocytes. Various mediators synthesized by adipocytes and
resident macrophages might contribute to local and systemic inflammation. The overall
‘adipocytokine–cytokine cocktail’ might favour a pro-inflammatory milieu. IL, interleukin;
TNF, tumour-necrosis factor.
NATURE REVIEWS | IMMUNOLOGY
VOLUME 6 | OCTOBER 2006 | 773
© 2006 Nature Publishing Group
Type 2 diabetes mellitus
A disorder of glucose
homeostasis that is
blood-glucose levels and
resistance of tissues to the
action of insulin. Recent
studies indicate that
inflammation in adipose tissue,
liver and muscle contributes to
the insulin-resistant state that
is characteristic of type 2
diabetes mellitus, and that the
anti-diabetic actions of
activated receptor-γ (PPARγ)
agonists result, in part, from
their anti-inflammatory effects
in these tissues.
Mice with a spontaneous
mutation in the gene encoding
leptin (chromosome 6) that
leads to decreased leptin
production. These mice are
severely obese and develop
(ER stress). A response by the
ER that results in the disruption
of protein folding and in the
accumulation of unfolded
proteins in the ER.
sensitivity to insulin. Furthermore, mice expressing only
a single copy of the gene encoding X-box-binding pro-
tein 1 (XBP1), which is a transcription factor that regu-
lates a large number of genes in the ER-stress response,
develop insulin resistance16. Other factors involved in
the regulation of insulin resistance are IL-6 and various
suppressor of cytokine signalling (SOCS) proteins17,18.
Therefore, several pro-inflammatory cytokines, SOCS
proteins, ER stress, the IKKβ pathway of NF-κB acti-
vation and JNK signalling pathways are all associated
with the development of insulin resistance, indicating
that various pro-inflammatory mediators released by
adipocytes, in addition to the initially described pro-
inflammatory cytokine TNF, link the immune system
with obesity-related insulin resistance.
By studying mice with conditional deletion of
Ikbkb in either hepatocytes or myeloid cells, Michael
Karin’s laboratory showed that mice lacking IKKβ in
hepato cytes retain insulin responsiveness in the liver
but develop insulin resistance in muscle and fat in
response to a high-fat diet, obesity or aging19. By con-
trast, mice lacking IKKβ in myeloid cells retain global
insulin sensitivity and are protected from insulin resist-
ance19. Therefore, this study indicates that myeloid
cells, probably macrophages, regulate systemic insulin
sensitivity and are involved in inflammation-associated
insulin resistance, whereas hepatic IKKβ expression is
required for insulin resistance in the liver. Selective
activation of NF-κB, causing continuous low-level
expression of IKKβ, in hepatocytes from a transgenic
mouse model leads to moderate systemic insulin resist-
ance20. In this study, insulin resistance was decreased by
systemic neutralization of IL-6, which also resulted in
decreased expression of SOCS1, SOCS2 and SOCS3
in the liver. Therefore, not only is a complex network of
mediators involved in the regulation of insulin resistance,
but various cell types in addition to adipocytes are also
involved, including hepatocytes, macrophages and
Although adiponectin is synthesized mainly by adipo-
cytes, it is also expressed by skeletal muscle cells, car-
diac myocytes and endothelial cells21–23 (FIG. 2a). It has
sequence homology with a family of proteins that are
characterized by an amino-terminal collagen-like region
and a carboxy-terminal, complement factor C1q-like
globular domain24–26. Adiponectin exists as a full-length
protein, as well as a proteolytic cleavage fragment,
consisting of the globular C-terminal domain (which
is known as globular adiponectin). It is thought that
a leukocyte elastase, secreted by activated monocytes
and/or neutrophils, mediates this cleavage process
and generates the globular fragment of adiponectin.
Globular adiponection can trimerize after cleavage, but
cannot oligomerize further27. Full-length adiponectin
can exist as: a trimer (known as low-molecular-weight
adipo nectin); a hexamer, which consists of two trimers
linked by a disulphide bond (known as middle-molecular-
weight adiponectin); and a high-molecular-weight
12- to 18-mer (FIG. 2a). Adiponectin circulates at high
concentrations in human serum (5–10 mg per ml,
compared with leptin, which circulates at a concen-
tration of a few ng per ml) and it has a wide range of
biological activities9. Serum levels of adiponectin are
markedly decreased in individuals with visceral obesity
and states of insulin resistance, such as non-alcoholic
fatty liver disease, atherosclerosis and type 2 diabetes
mellitus, and adiponectin levels correlate inversely with
insulin resistance28. Both trimers and other oligomers
of adiponectin are present in the circulation, whereas
the presence of the globular fragment in the serum in
humans has been questioned29,30. It has been suggested
recently that the ratio, and not the absolute amounts, of
high-molecular-weight and low-molecular-weight adi-
ponectin in the serum might be crucial in determining
insulin sensitivity31. In support of the importance of cir-
culating high-molecular-weight adiponectin in protecting
against insulin resistance, moderate weight loss leads to a
relative increase in the ratio of high-molecular-weight to
middle-molecular-weight adiponectin and a decrease
in the amount of low-molecular-weight adiponectin in
Two receptors for adiponectin have been identi-
fied recently (ADIPOR1 and ADIPOR2). ADIPOR1 is
expressed widely in mice, whereas ADIPOR2 is expressed
mainly in the liver33. Whereas globular adiponectin seems
to activate mainly ADIPOR1, ADIPOR2 engages mainly
with the full-length variant of adiponectin33. In addition,
T-cadherin, which is expressed by many cells including
endothelial cells and smooth muscle cells, seems to func-
tion as a receptor for middle-molecular-weight and high-
molecular-weight adiponectin, but not for the trimeric
(low-molecular-weight) and globular forms34.
TNF suppresses the transcription of adiponectin in
an adipocyte cell line, which might explain the lower
levels of serum adiponectin in individuals who are
Box 1 | Insulin resistance, obesity and inflammation
Insulin regulates the uptake, oxidation and storage of fuel in insulin-sensitive tissues,
such as the liver, skeletal muscle and adipose tissue, and also macrophages. Obesity,
in particular visceral obesity, which is the accumulation of adipose tissue inside the
abdominal cavity, is associated with resistance to the effects of insulin (insulin
resistance) on peripheral glucose and fatty-acid utilization, often leading to type 2
diabetes mellitus. With the recent trend for individuals to be more obese, a large
increase in the prevalence of insulin resistance in westernized countries is expected.
Insulin resistance, together with the associated hyperinsulinaemia and hyperglycaemia,
and the presence of pro-inflammatory mediators might lead to a state of vascular
endothelial dysfunction, an abnormal lipid profile, hypertension and vascular
inflammation, all of which promote the development of atherosclerotic cardiovascular
Subclinical, low-grade inflammation might have an important role in the pathogenesis
of insulin resistance and type 2 diabetes mellitus. Population studies show a strong
correlation between the levels of pro-inflammatory biomarkers, such as C-reactive
protein, interleukin-6 and tumour-necrosis factor, and perturbations in glucose
homeostasis, obesity and atherosclerosis. However, although epidemiological
correlations have been established, the exact cellular and molecular mechanisms that
link obesity and insulin resistance are unknown. Insulin resistance might be partly
precipitated or accelerated by an acute-phase reaction as part of the innate immune
response, in which large amounts of pro-inflammatory mediators and insufficient
amounts of anti-inflammatory mediators, such as adiponectin, are released from
774 | OCTOBER 2006 | VOLUME 6
© 2006 Nature Publishing Group
↑ GLUT4 translocation
↓ ACC (Malonyl-CoA)
obese35. Expression of adiponectin is also regulated by
other pro-inflammatory mediators such as IL-6, which
suppresses adiponectin transcription and translation in
an adipocyte cell line36. Weight loss is a potent inducer
of adiponectin synthesis37, as is activation of peroxisome-
proliferator-activated receptor-γ (PPARγ) by its ligands
thiazolidinediones, which are important for the treat-
ment of type 2 diabetes mellitus38,39. Circulating levels of
adiponectin, however, are affected by many other factors
including gender, age and lifestyle.
Role in innate and adaptive immunity. Early studies
indicated that adiponectin had an anti-inflammatory
effect on endothelial cells through the inhibition of
TNF-induced adhesion-molecule expression40. In
addition, adiponectin-deficient mice have higher lev-
els of expression of mRNA encoding TNF in adipose
tissue and higher TNF concentrations in plasma com-
pared with adiponectin-sufficient mice35. Adiponectin
inhibits NF-κB activation in endothelial cells and
interferes with the function of macrophages40,41
(FIGS 2b,3a); treatment of cultured macrophages with
adiponectin markedly inhibited their phagocytic activ-
ity and production of TNF in response to stimulation
with lipopolysaccharide (LPS)41.
Adiponectin also induces the production of
important anti-inflammatory cytokines, such as
IL-10 and IL-1 receptor antagonist (IL-1RA), by human
monocytes, macrophages (FIG. 3a) and dendritic cells
(DCs), and suppresses the production of interferon-γ
(IFNγ) by LPS-stimulated human macrophages42.
Through ADIPOR1, globular adiponectin suppresses
TLR-induced NF-κB activation43, indicating that adi-
ponectin negatively regulates macrophage responses to
TLR ligands, which is probably of relevance in innate
The presence of adiponectin in T-cell proliferation
assays resulted in a decreased ability to evoke an allo-
geneic T-cell response42 (TABLE 1), and adiponectin also
markedly reduced the phagocytic capacity of macro-
phages. However, stimulation of DCs with adiponectin
did not result in any changes in cell-surface marker
expression, phagocytic capacity or ability to stimulate
allogeneic T-cell proliferation, which might indicate that
adiponectin mainly affects the function of macrophages
and not DCs42.
There might, however, be certain situations in which
adiponectin has pro-inflammatory effects. In the pres-
ence of LPS, high-molecular-weight adiponectin was
shown to augment the translation of CXC-chemokine
ligand 8 (CXCL8; also known as IL-8) by human macro-
phages44. Low- and high-molecular-weight adiponectin
share some biological effects on monocytes, such as the
induction of apoptosis, the activation of AMP-activated
protein kinase (AMPK) and the suppression of scavenger-
receptor expression by macrophages45. However, in
this study45, high-molecular-weight adiponectin also
induced the secretion of IL-6 by human monocytes,
whereas only the low-molecular-weight form had anti-
inflammatory effects by decreasing IL-6 production in
response to LPS and inducing IL-10 synthesis. The exact
Figure 2 | Adiponectin: sources, structure and effects on pro- and anti-
inflammatory cytokines. a | Adiponectin is produced mainly by adipocytes, but other
cell types, such as skeletal and cardiac myocytes and endothelial cells, can also produce
this adipocytokine. Adiponectin exists as a full-length trimer (low-molecular-weight
form), as well as a proteolytic cleavage fragment (globular adiponectin). The full-length
trimer can dimerize to form a hexamer (middle-molecular-weight form), which can then
oligomerize to form a polymer (high-molecular-weight form). It has been proposed that
cleavage of full-length adiponectin by a leukocyte elastase secreted by activated
monocytes and/or neutrophils generates the globular fragment. b | Adiponectin
interacts with at least two known cellular receptors (ADIPOR1 and ADIPOR2). Activation
of ADIPOR1 and/or ADIPOR2 by adiponectin stimulates the activation of peroxisome-
proliferator-activated receptor-α (PPARα), AMP-activated protein kinase (AMPK) and
p38 mitogen-activated protein kinase33. Adiponectin regulates the expression of several
pro- and anti-inflammatory cytokines. Its main anti-inflammatory function might be
related to its capacity to suppress the synthesis of tumour-necrosis factor (TNF) and
interferon-γ (IFNγ) and to induce the production of anti-inflammatory cytokines such as
interleukin-10 (IL-10) and IL-1 receptor antagonist (IL-1RA). Activation of PPARs exerts
anti-inflammatory effects through inhibition of the transcriptional activation of pro-
inflammatory response genes. ACC, acetyl-CoA carboxylase; GLUT4, glucose transporter
type 4; IKK, inhibitor of nuclear factor-κB (IκB) kinase; NF-κB, nuclear factor-κB;
PPRE, peroxisome-proliferator response element; SREBP1C, sterol-regulatory-
element-binding protein 1C; TNFR, TNF receptor.
NATURE REVIEWS | IMMUNOLOGY
VOLUME 6 | OCTOBER 2006 | 775
© 2006 Nature Publishing Group
roles of the different full-length and globular forms of
adiponectin in inflammation and immunity remain to
Role in insulin resistance and inflammation. In obese
animals, treatment with adiponectin decreases hyper-
glycaemia and levels of free fatty acids in the plasma,
and improves insulin sensitivity9,46. Furthermore,
adiponectin-deficient mice develop diet-induced
insulin resistance on a high-fat, high-sucrose diet35.
Specific PPARγ agonists, such as thiazolidinediones,
improve insulin sensitivity by mechanisms that are
largely unknown. Circulating levels of adiponectin
are significantly upregulated in vivo after activation
of PPARγ38,39. Mice lacking adiponectin not only
have decreased hepatic insulin sensitivity but also
have reduced responsiveness to PPARγ agonists,
which indicates that adiponectin is an important
contributor to PPARγ-mediated improvements
in insulin sensitivity47. Adiponectin stimulates
β-oxidation in rat hepatocytes and downregulates
expression of sterol-regulatory-element-binding
protein 1C (SREBP1C), which is the main transcrip-
tion factor regulating expression of genes encoding
mediators of lipid synthesis (FIG. 2b). Sustained periph-
eral, ectopic expression of adiponectin decreases the
development of diet-induced obesity and improves
insulin sensitivity48. Together, these studies strongly
support a major role for adiponectin in regulating
Figure 3 | Effects of various adipocytokines on the
monocyte–macrophage system. Adipocytokines exert
different effects on the innate immune system and either
suppress or activate the monocyte–macrophage system.
a | Adiponectin, through interaction with its receptor
ADIPOR1 (and ADIPOR2), suppresses the nuclear factor-κB
(NF-κB)-dependent synthesis of tumour-necrosis factor
(TNF) and interferon-γ (IFNγ), and induces the production
of interleukin-10 (IL-10) and IL-1 receptor antagonist
(IL-1RA). Adiponectin also induces apoptosis of monocytes
and inhibits phagocytosis by macrophages. b | Leptin
signals through its receptor OBRb to induce activation of
the mitogen-activated protein kinases (MAPKs) p38 and
extracellular-signal-regulated kinase (ERK) and of signal
transducer and activator of transcription 3 (STAT3). This
results in production of the pro-inflammatory cytokines
TNF, IL-6 and IL-12. Leptin also induces the production
of nitric-oxide synthase 2 (NOS2) and, thereby, reactive
oxygen species (ROS), enhances macrophage
phagocytosis, and induces the activation, proliferation
and migration of monocytes. c | The receptor for resistin is
unknown, but this adipocytokine induces the activation
of p38, ERK and phosphatidylinositol 3-kinase (PI3K).
Resistin increases the production of TNF, IL-1β, IL-6 and
IL-12. Its effect on monocyte and macrophage functions
is not known. Whereas adiponectin can be considered an
anti-inflammatory strategy of the ‘adipose organ’, leptin
and resistin have dominant pro-inflammatory features.
IκB, inhibitor of NF-κB; LPS, lipopolysaccharide;
PPAR, peroxisome-proliferator-activated receptor;
PPRE, peroxisome-proliferator response element;
TLR4, Toll-like receptor 4.
776 | OCTOBER 2006 | VOLUME 6
© 2006 Nature Publishing Group
The amino-terminal domain
of adiponectin contains a signal
sequence that is followed by
a stretch of 22 collagen-like
repeats, consisting of 7 perfect
Gly-X-Pro repeats and
15 ‘imperfect’ Gly-X-Y repeats
(where X and Y are different
amino acids), which — similar
to procollagen — allows the
assembly of three full-length
adiponectin molecules to an
C1q-like globular domain
The carboxy-terminal globular
domain of adiponectin, which
has marked homology to
several other proteins,
including subunits of the
complement factor C1q.
The liver is one of the main insulin-sensitive tis-
sues and insulin resistance has an important role in
the development of non-alcoholic fatty liver disease.
Therefore, several studies have assessed the roles of
fat, inflamma tion and adiponectin in experimental
liver disease. Adiponectin has anti-inflammatory
effects in various animal models of liver inflammation.
Administration of adiponectin has beneficial effects in
both alcoholic and non-alcoholic fatty liver disease
in mice, by suppressing the expression of TNF in the
liver. Adiponectin also decreases hepatomegaly, steato-
sis and the levels of liver enzymes49. In addition, adipo-
nectin attenuates liver fibrosis in the carbon-tetrachloride
liver-fibrosis model50 and protects against endotoxin-
induced liver injury in another model of fatty-liver
disease, the KK-Ay obese mouse model51. Sennello and
colleagues studied concanavalin A (ConA)-induced
hepatotoxicity in lipodystrophic transgenic mice (which
constitutively express a truncated form of SREBP1c and
lack virtually all white adipose tissue) and lean, wild-
type control mice52. Serum adiponectin levels were low
in lipodystrophic mice compared with controls, and the
administration of adiponectin protected lipo dystrophic
mice from hepatotoxicity and protected primary
hepatocytes from TNF-induced cell death.
AMPK is an evolutionarily conserved sensor of
the energy status of a cell, and it has a crucial role in
controlling the systemic energy balance by regulating
food intake, body weight, and glucose and lipid homeo-
stasis53. AMPK increases the sensitivity of a cell to both
insulin and thiazolidinediones. Full-length adiponectin
stimulates AMPK phosphorylation and activation in
the liver, whereas globular adiponectin has this effect
in both skeletal muscle and liver tissue54. Each of these
effects of adiponectin could be inhibited by the use of a
dominant-negative AMPK mutant, further supporting
the model that glucose utilization occurs through activa-
tion of AMPK54. Adiponectin protects the myocardium
from injury by protecting cardiac muscle cells from
apoptosis through activation of AMPK signalling55,56.
Ischaemia–reperfusion in adiponectin-deficient mice
resulted in increased myocardial-infarct size, myocar-
dial apoptosis rate and TNF expression compared with
adiponectin-sufficient mice. All of these effects were
Table 1 | Effects of adipocytokines on the immune system and linked diseases
Adipocytokine Inflammatory effect Effects on immunity
↓ Endothelial adhesion
↓ IL-6(REF. 42)
↑ IL-10 (REF. 42)
↓ B-cell lymphopoiesis117
↓ T-cell responses42
• Insulin resistance and type 2 diabetes
• Experimentally induced liver disease: non-
alcoholic and alcoholic fatty liver disease49;
CCl4 liver fibrosis (REF. 50); LPS-treated
KK-Ay mice (REF. 51); and experimentally
induced hepatitis (ConA)52
• Cardiac injury55,56
• Inflammatory bowel disease112
• Rheumatoid arthritis115
↑ CXCL8 in presence of
↑ IL-6 (REF. 68)
↑ IL-12 (REF. 68)
↑ Neutrophil activation
↑ NK-cell function72
↑ Thymocyte survival73
↑ T-cell proliferation64
↑ TH1 response
(IL-2 and IFNγ)64
↓ TH2 response (IL-4)64
• Insulin resistance9
• Experimentally induced hepatitis
• EAE and antigen-induced arthritis4,77
• Experimentally induced colitis:
CD4+CD45RBhi T-cell transfer78; and IL-10-
• Insulin resistance (mice)5
• Type 2 diabetes mellitus (mice)80
• Rheumatoid arthritis86
• Non-alcoholic fatty liver disease118
• Chronic kidney disease94
↑ IL-6 (REF. 86)
↑ IL-12 (REF. 86)
↑ Endothelial adhesion
molecules (VCAM1 and
↑ IL-6 (REF. 119)
↑ IL-8 (REF. 119)
↓ Apoptosis of
• Insulin resistance and type 2 diabetes
• Acute lung injury97
• Sepsis98 X
CCl4, carbon tetrachloride; ConA, concanavalin A; CXCL, CXC-chemokine ligand; EAE, experimental autoimmune encephalomyelitis; ICAM, intercellular adhesion
molecule; IFNγ, interferon-γ; IL, interleukin; IL-1RA, IL-1 receptor antagonist; LPS, lipopolysaccharide; ND, not determined; NF-κB, nuclear factor-κB; NK, natural
killer; ROS, reactive oxygen species; TH, T helper; TNF, tumour-necrosis factor; VCAM, vascular cell-adhesion molecule.
NATURE REVIEWS | IMMUNOLOGY
VOLUME 6 | OCTOBER 2006 | 777
© 2006 Nature Publishing Group
Accumulation of adipose tissue
inside the abdominal cavity,
in particular at omental and
mesenteric regions, which are
drained by the portal vein and
therefore have direct access
to the liver.
A member of the cadherin
family of transmembrane
glycoproteins that mediate
(PPARγ). A nuclear receptor
that is a master transcriptional
regulator of metabolism and
fat-cell formation. The activity
of PPARγ can be modulated
by the direct binding of
small molecules —
thiazolidinediones. PPARγ has
by limiting the availability of
limited cofactors or blocking
promoters of pro-inflammatory
IL-1 receptor antagonist
(IL-1RA). A secreted protein
that binds to IL-1R, thereby
blocking IL-1R downstream
signalling. IL-1RA inhibits the
Intraperitoneal or oral
administration of hepatotoxic
carbon tetrachloride (CCl4)
to mice is a commonly used
model of both acute and
chronic liver injury. CCl4 causes
hepatocyte injury that is
characterized by centrilobular
necrosis followed by hepatic
KK-Ay obese mice
The spontaneous Ay mutation
(agouti signal protein; yellow)
was introduced onto the KK
strain background. KK-Ay
heterozygous mice have yellow
hair pigment and black eyes
and develop hyperglycaemia,
intolerance and obesity by
8 weeks of age.
decreased by the administration of adiponectin in both
adiponectin-deficient mice and wild-type mice56. These
studies indicate that several important adiponectin-
induced effects are mediated through the activation of
AMPK (FIG. 2b).
Adiponectin might have an important role in the
pathophysiology of atherosclerosis. Adiponectin-deficient
mice had a two-fold greater neointimal (that is, inner
vessel surface) formation in response to an external vas-
cular injury than did wild-type mice57. Furthermore, adi-
ponectin protected apolipoprotein E (APOE)-deficient
mice (mice lacking a key component in cholesterol
metabolism) from atherosclerosis58. This experimental
evidence is paralleled by several clinical reports that
support the observation that hypoadiponectinaemia is
associated with the development of atherosclerosis59–61.
So, adiponectin is an important mediator in the regu-
lation of insulin resistance and can suppress inflamma-
tion in various animal models. This adipocytokine also
has a crucial role in suppressing macrophage activity,
not only in adipose tissue but also in other tissues such
as the liver. Decreased synthesis of adiponectin, as is
observed in individuals who are obese, might lead to
dysregulation of the controls that inhibit the production
of pro-inflammatory cytokines, thereby leading to the
production of increased amounts of pro-inflammatory
mediators. One of the main challenges in understanding
the physiology of this adipocytokine will be to under-
stand why circulating levels decrease with the onset of
obesity. Also of great interest is how this decrease might
affect the cytokine–adipocytokine milieu, resulting in an
overwhelmingly pro-inflammatory state.
The role of leptin in modulating the immune response
and inflammation has become increasingly evident and
has been reviewed recently4. In addition to regulating
neuroendocrine function, energy homeostasis, haema-
topoiesis and angiogenesis, this adipocytokine is an
important mediator of immune-mediated diseases and
Similar to adiponectin, leptin is produced mainly
by adipocytes. However, unlike adiponectin, leptin is
considered to be a pro-inflammatory cytokine and it
has structural similarity to other pro-inflammatory
cytokines such as IL-6, IL-12 and granulocyte colony-
stimulating factor. The main function of leptin is con-
trol of appetite4. Indeed, mice with a mutation in the
gene encoding leptin (ob/ob mice) or the gene encoding
the leptin receptor (db/db mice) have obese phenotypes
and are used in many studies as mouse models of obes-
ity. However, these mice also have various defects in
cell-mediated and humoral immunity62–64.
Serum levels of leptin reflect the amount of energy
stored in the adipose tissue and are proportional to
overall adipose mass in both mice and humans4,65.
Serum levels are 2–3 times higher in women than in
men, even when adjusted for age and body-mass index
(BMI). In animal models, expression of leptin is
increased in conditions that are associated with the
release of pro-inflammatory cytokines, as induced
during acute inflammatory conditions such as sep-
sis66,67. An increase in leptin levels and a decrease in
expression of mRNA encoding the full-length isoform
b receptor (OBRb, which is one of at least six alterna-
tively spliced isoforms, each of which has a cytoplas-
mic domain of a different length) has been observed in
diet-induced obese rats4. In addition to adipose tissue,
leptin is produced by several other tissues, including
placenta, bone marrow, stomach, muscle and perhaps
the brain4. Therefore, pro-inflammatory mediators
and obesity seem to be the main factors responsible
for increased leptin synthesis.
Role in innate and adaptive immunity. In monocytes
and macrophages, leptin increases the production of
pro-inflammatory cytokines such as TNF, IL-6 and
IL-12 (REF. 68) (FIG. 3b). It also upregulates the expres-
sion of pro-inflammatory and pro-angiogenic factors,
such as CCL2 and vascular endothelial growth factor,
respectively, in human hepatic stellate cells69. This effect
in human stellate cells is mediated through activation of
NF-κB, as well as other signalling intermediates, includ-
ing the serine/threonine protein kinase AKT, which is
the main downstream target of phosphatidylinositol
3-kinase69. Leptin also activates neutrophils, as assessed
by increased expression of CD11b, and stimulates the
proliferation of human circulating monocytes in vitro
and upregulates the expression of activation markers
such as CD25 (also known as IL-2Rα) and CD71 (the
transferrin receptor) on these cells9,70. Leptin-induced
TNF production by murine peritoneal macrophages is
inhibited by globular adiponectin through suppression
of the phosphorylation of extracellular-signal-regulated
kinase 1 (ERK1), ERK2 and p38 (REF. 71). Furthermore,
leptin stimulates neutrophil chemotaxis and the produc-
tion of reactive oxygen species (ROS) by these cells, and
regulates natural killer (NK)-cell differentiation, pro-
liferation, activation and cytotoxicity72. Most of these
pro-inflammatory effects are mediated through the long
isoform of the leptin receptor (OBRb), which is expressed
mainly by endothelial cells and various leukocytes.
The effects of leptin on adaptive immunity have been
well studied4 (FIG. 4). Leptin induces the proliferation of
naive CD4+CD45RA+ T cells, but inhibits the proliferation
of memory CD4+CD45RO+ T cells in a mixed lymphocyte
reaction (MLR)64. In T-cell proliferation assays with
mouse cells, leptin increased production of the T helper
1 (TH1) cytokines IL-2 and IFNγ, and suppressed produc-
tion of the TH2 cytokine IL-4 (REF. 64). Administration of
leptin reversed the immunosuppressive effects of acute
starvation in mice64 and provided a survival signal for
thymocytes, thereby protecting the mice from starvation-
induced lymphoid atrophy. Leptin has also been shown
to increase thymic cellularity in ob/ob mice73. Despite this
evidence of a role for leptin in immune responses in vitro
and in mouse models, it is currently unclear whether
leptin influences immune responses in humans.
Role in inflammation. The role of leptin in inflamma-
tion is incompletely understood. Endogenous leptin
protects against TNF-mediated toxicity. Ob/ob mice and
778 | OCTOBER 2006 | VOLUME 6
© 2006 Nature Publishing Group
↓ B-cell number
↑ COX1/COX2 and
PGE2 in stromal cells
↑ IgG2a switch
↑ Thymocyte number
↑ CD4+CD8+ T cells
↑ CD4+CD8– T cells
Transgenic mice that express a
truncated, constitutively active
form of the sterol-regulatory-
element-binding protein 1C
(SREBP1C) transcription factor
under the control of the
adipose-specific aP2 promoter.
Lipodystrophic mice have
low plasma leptin levels,
(BMI). This is the most
frequently used method to
gauge an individual’s deviation
from ‘normal’ body weight.
The BMI is the quotient of
body weight (in kg) through
the square of height (m²).
Underweight: <20; ideal:
20–25; overweight: >25;
Mixed lymphocyte reaction
A tissue-culture technique that
is used for the in vitro testing
of the proliferative response
of T cells from one individual
to lymphocytes from another
The injection of tumour-
necrosis factor (TNF) into
animals, which results in acute
anorexia, weight loss, shock
and even death.
(EAE). An experimental model
of multiple sclerosis that is
induced by immunization
of susceptible animals with
such as myelin basic protein,
proteolipid protein or myelin
db/db mice, as well as mice treated with a leptin-receptor
antagonist, had increased sensitivity to the lethal effects
of TNF74. The addition of exogenous leptin protected
against TNF-mediated toxicity in ob/ob mice, but did
not increase the protective effect of endogenous leptin
in wild-type mice. Also, in various mouse models of
inflammation, such as ConA-induced hepatitis, leptin
deficiency has a protective effect by decreasing the pro-
duction of pro-inflammatory TH1 cytokines and shifting
the immune response towards a TH2-type response75.
Several studies have investigated the susceptibility
of ob/ob and db/db mice to experimentally induced
auto immune diseases4,9,76. Ob/ob mice are resistant to
experimental autoimmune encephalomyelitis, but become
susceptible to disease after leptin administration77. Leptin
administration also exacerbates disease in wild-type mice
through increased synthesis of TH1 cytokines77. Similar
data were obtained from models of intestinal inflamma-
tion, where CD4+CD45RBhi T cells from db/db mice had
a decreased capacity to cause colitis when transferred to
severe combined immunodeficient (SCID) mice, com-
pared with cells from wild-type mice78. Decreased IFNγ
production was observed early on in SCID mice that
received db/db T cells compared with wild-type T cells,
indicating that leptin affects the immune response, prob-
ably through its interaction with the long isoform of its
receptor (OBRb), which is expressed by CD4+CD45RBhi
T cells78. Leptin administration also increases disease
severity and accelerates autoimmune diabetes in female
non-obese diabetic (NOD) mice. Recent clinical reports
in patients with autoimmune diseases have shown that
high serum levels of leptin might be either a contributing
factor to or a marker of disease activity4,70.
Taken together, it seems clear that leptin has pro-
inflammatory effects. This could be detrimental in
many animal models of inflammatory and autoimmune
disease, but it might be protective in several infectious-
disease settings, such as during the early immune
response to pulmonary tuberculosis, following infec-
tion with Gram-negative pneumococci and during viral
myocarditis4,9. However, the direct interaction of the
two main adipocytokines adiponectin and leptin is not
well understood and this might have important implica-
tions in understanding the role of these adipocytokines
in obesity-associated disorders.
The adipocytokine resistin was discovered by three
independent groups79–81. Resistin (also known as
FIZZ3), which is a 114-amino-acid polypeptide, was
originally shown to induce insulin resistance in mice80.
It belongs to a family of cysteine-rich proteins, also
known as resistin-like molecules (RELMs), that have
been implicated in the regulation of inflammatory
processes79. Resistin was shown to circulate in two dis-
tinct forms: a more prevalent high-molecular-weight
hexamer and a substantially more bioactive, but less
prevalent, low-molecular-weight complex82.
mRNA encoding resistin can be found in mice and
humans in various tissues, including adipose tissue, the
hypothalamus, adrenal gland, spleen, skeletal muscle,
pancreas and gastrointestinal tract5. Although resis-
tin protein synthesis in mice seems to be restricted to
adipocytes, in humans, adipocytes, muscle, pancreatic
cells and mononuclear cells such as macrophages can
synthesize this protein. Expression levels of the gene
encoding resistin have been shown to be higher in
human peripheral-blood mononuclear cells (PBMCs)
than in adipocytes; however, comparative protein data
are not available5. So, it still remains to be shown which
cell type in humans is mainly responsible for systemic
production and for the high circulating levels of resistin.
Remarkably, at the protein level, human resistin is only
55% identical to its mouse counterpart, which indicates
that it might not be evolutionarily well conserved across
In human PBMCs, expression of resistin mRNA is
markedly increased by the pro-inflammatory cytokines
IL-1, IL-6 and TNF, and by LPS, whereas IFNγ and
leptin had no effect83. Similarly, stimulation of human
macrophages with LPS led to increased resistin mRNA
expression, and administration of LPS to humans
resulted in a marked increase in the level of resistin in
the serum84. The induction of resistin synthesis can be
Figure 4 | Effects of adipocytokines on adaptive
immunity. Only limited data are available to describe the
potential effects of adiponectin (and resistin) on adaptive
immunity. In terms of T-cell proliferation, adiponectin
suppresses the ability to evoke allogeneic T-cell responses.
Adiponectin affects T helper 1 (TH1)-cell immunity by
suppressing the production of interferon-γ (IFNγ) and
tumour-necrosis factor (TNF), and by inducing production
of the anti-inflammatory cytokine interleukin-10 (IL-10).
In terms of lymphopoiesis, adiponectin suppresses B-cell
development through the induction of prostaglandin
synthesis. Leptin is the best studied adipocytokine and has
many influences on adaptive immunity, such as inducing a
switch towards TH1-cell immune responses by increasing
IFNγ and TNF secretion, the suppression of TH2-cell
responses, and inducing the production of IgG2a by
B cells. Leptin promotes the generation, maturation and
survival of thymic T cells, and it increases the proliferation
of and IL-2 secretion by naive T cells. COX, cyclooxygenase;
PGE2, prostaglandin E2.
NATURE REVIEWS | IMMUNOLOGY
VOLUME 6 | OCTOBER 2006 | 779
© 2006 Nature Publishing Group
A localized dilation of a blood
vessel by more than 50%
of its diameter owing to
damage of the vessel wall.
attenuated by PPARγ agonists5,84. Accordingly, treatment
of patients with type 2 diabetes mellitus with the PPARγ
agonist pioglitazone decreased serum levels of resistin85.
In addition, several factors such as pituitary, steroid
and thyroid hormones, adrenaline, β3-adrenoreceptor
activation, endothelin-1 and insulin modulate resistin
Role in immunity. Resistin strongly upregulates the
expression of TNF and IL-6 by human PBMCs and
induces arthritis after injection into the joints of
healthy mice86. These pro-inflammatory properties
of resistin were abrogated by an NF-κB inhibitor,
showing the important role of NF-κB in resistin-
controlled inflammatory reactions. Resistin has also been
shown to accumulate in the inflamed joints of patients
with rheumatoid arthritis and its levels correlate with
markers of inflammation5. Human resistin stimulates
synthesis of the pro-inflammatory cytokines TNF, IL-1,
IL-6 and IL-12 by various cell types through an NF-κB-
dependent pathway83,87 (FIG. 3c). In further support of
its pro-inflammatory profile, resistin also upregulates
the expression of vascular cell-adhesion molecule 1
(VCAM1), intercellular adhesion molecule 1 (ICAM1)
and CCL2 by human endothelial cells and induces these
cells to release endothelin-1 (REF. 88).
Role in inflammation and insulin resistance. Resistin
has been implicated in the pathogenesis of obesity-
associated insulin resistance and type 2 diabetes mellitus
in mouse models80, whereas such a role in humans is
still debated89–91. Although a clear function for resistin in
humans is still lacking, its pro-inflammatory properties
indicate that it has a role in inflammatory processes5.
Macrophages infiltrating human atherosclerotic aneurysms
secrete resistin92. Resistin and adiponectin have reciprocal
effects on vascular endothelial cells: resistin induces the
expression of VCAM1, ICAM1 and pentraxin-3, whereas
adiponectin downregulates the expression of these mol-
ecules93. Increased levels of resistin in chronic kidney
disease are associated with impaired renal function and
inflammation, but not with insulin resistance94.
Therefore, this adipocytokine, at least in humans, has
many features of a pro-inflammatory cytokine and could
have a role in inflammatory diseases with or without
associated insulin resistance. These pro-inflammatory
effects, however, are based on a small number of studies
and much more information is required to characterize
resistin more fully in both mice and humans.
Visfatin and other new adipocytokines
Visfatin (also known as PBEF) has recently been identi-
fied as an adipocytokine that is secreted by adipocytes in
visceral fat and that decreases insulin resistance95. This
molecule binds to and activates the insulin receptor but
does not compete with insulin, which indicates that the
two proteins bind different sites on the insulin receptor.
Visfatin was originally identified as PBEF (pre-B-cell
colony-enhancing factor) more than 10 years ago and
since then, it has been linked to several inflammatory
disease states such as acute lung injury96,97. Furthermore,
expression of visfatin has been shown to be upregulated in
activated neutrophils and to inhibit the apoptosis of neu-
trophils98. Future studies of the cell biology of this natural
insulin mimetic and potential inflammation-regulating
adipocytokine should help to define its role in insulin
resistance and associated inflammatory disorders.
Hida and colleagues recently identified a new adi-
pocytokine termed VASPIN (visceral adipose-tissue-
derived serine protease inhibitor), which has similarities
to adiponectin in that it improves insulin sensitivity.
Preliminary studies indicate that VASPIN might also
have anti-inflammatory effects, as it suppresses the
production of TNF, leptin and resistin99.
Serum retinol-binding protein 4 (RBP4) is another
recently characterized adipocytokine100. Until recently,
the sole function of RBP4 was thought to be the deliv-
ery of retinol to tissues. However, in patients with type 2
diabetes mellitus, serum levels of RBP4 are increased.
Expression of RBP4 is also increased in adipose tissue
of mice lacking the glucose transporter GLUT4, and
consistent with this, expression of GLUT4 is selectively
decreased in adipocytes from obese individuals or indi-
viduals with type 2 diabetes. Treatment with fenretinide,
a synthetic retinoid that increases urinary excretion of
RBP4, normalizes serum levels of RBP4 and decreases
insulin resistance in mice with obesity induced by a
high-fat diet. Transgenic overexpression of human RBP4
or injection of recombinant RBP4 in normal mice causes
insulin resistance. Therefore, decreasing the concentra-
tion of RBP4 could be an interesting strategy for the
treatment of individuals with type 2 diabetes mellitus.
Role in cancer and immune-mediated diseases
In humans, the prevalence of cancer and some immune-
mediated diseases (such as asthma) is increased or disease
activity is more severe in individuals who are obese2,3.
Although there is currently only limited evidence for
this model, mainly derived from correlative studies, it
is proposed that adipocytokines could link obesity with
Adiponectin has anti-angiogenic effects, through the
inhibition of endothelial-cell proliferation and migra-
tion101. In a mouse tumour model, adiponectin markedly
inhibited primary tumour growth in a caspase-dependent
manner and it resulted in endothelial-cell apoptosis102.
Prostate cancer is also associated with obesity, and full-
length adiponectin inhibits the growth of prostate-cancer
cells at physiological concentrations103. Many studies have
investigated the effect of leptin on different cancer types
in experimental cellular and animal models104. Most of
the studies indicate that leptin can potentiate the growth
of cancer cells (breast, oesophageal, gastric, pancre-
atic, colorectal, prostate, ovarian and lung carcinoma
cell lines), whereas adiponectin seems to decrease cell
Epidemiological studies indicate that obesity is a
significant risk factor for the development of cancer,
although the exact mechanisms of this have not yet
been identified2. Recently, it has been shown that
patients with various types of cancer, including gas-
tric, endometrial, prostate and breast cancer, have low
780 | OCTOBER 2006 | VOLUME 6
© 2006 Nature Publishing Group
In patients with rheumatoid arthritis, serum leptin
levels correlate with BMI rather than with disease stage. In
addition, increased levels of adiponectin and resistin are
observed in the synovial fluid of patients with rheumatoid
arthritis compared with patients with osteoarthritis86,115.
Intra-articular injection of resistin also induces arthritis
in healthy mouse joints86. An association of obesity with
the presence and development of rheumatoid arthritis,
however, is less clear116.
Great strides have been made towards understanding
the molecules linking obesity, inflammation and immu-
nity and understanding why obesity leads to chronic
inflammation. It is now evident that there are prototypic
adipocytokines, such as adiponectin and leptin, that are
synthesized mainly in the fat tissue, circulate at high
concentrations (in particular, adiponectin), function in a
hormone-like manner and have many of the features of
classical cytokines. These two mediators have dominated
the field of adipocytokine research recently and there is
increasing evidence that they are involved in many diseases
and, under certain circumstances, might crossregulate
Other adipocytokines, such as resistin and visfatin,
are also produced by adipocytes but an important site of
synthesis might be outside the adipose tissue, in particu-
lar by monocytes and macrophages. Resistin, although
differing in several functions between mice and humans,
seems to be mainly a pro-inflammatory mediator. Other
adipocytokines, such as VASPIN and RBP4, have only
recently been identified. Understanding the mecha-
nisms that lead from obesity to inflammation will have
important implications for the design of new therapies to
reduce the morbidity and mortality of obesity.
levels of circulating adiponectin105–108. Another study
indicated an association of obesity and decreased adi-
ponectin serum concentrations with colorectal adeno-
mas and a higher risk of colorectal cancer109. Although
these first studies are only descriptive and do not allow
further conclusions to be drawn, adipocytokines could
be attractive candidates as the missing link between
obesity and cancer.
An association between obesity and increased
asthma incidence and severity has been reported in
some studies3. Administration of leptin increases air-
way hyperresponsivness and the production of TH2
cytokines in ovalbumin-sensitized mice110. High leptin
levels have been observed in asthmatic children com-
pared with a control group with similar BMI, indicat-
ing that in addition to BMI, other factors might affect
leptin levels in these patients9. However, the association
of obesity with asthma in this study might also be influ-
enced by other factors, such as the increased prevalence
of gastro-oesophageal reflux disease observed in the
Overexpression of adipocytokines, including adi-
ponectin, leptin and resistin, in the mesenteric adipose
tissue of patients with Crohn’s disease after ileocoecal
surgical resection has been reported111. In Crohn’s
disease, leptin and adiponectin are highly expressed in
the mesenteric fat tissue, indicating that both pro- and
anti-inflammatory adipocytokines are overexpressed in
this type of inflammation111,112. Leptin-deficient mice
are protected from inflammation in some experimen-
tal models of inflammatory bowel disease113. However,
there is no clear association between obesity and the
development of inflammatory bowel disease, although
obesity has been associated with increased disease
Wellen, K. E. & Hotamisligil, G. S. Inflammation,
stress, and diabetes. J. Clin. Invest. 115, 1111–1119
Calle, E. E. & Kaaks, R. Overweight, obesity and
cancer: epidemiological evidence and proposed
mechanisms. Nature Rev. Cancer 4, 579–591 (2004).
Mannino, D. M. et al. Boys with high body masses
have an increased risk of developing asthma: findings
from the National Longitudinal Survey of Youth (NLSY).
Int. J. Obesity (Lond) 30, 6–13 (2006).
La Cava, A. & Matarese, G. The weight of leptin in
immunity. Nature Rev. Immunol. 4, 371–379 (2004).
Kusminski, C. M., McTernan, P. G. & Kumar, S.
Role of resistin in obesity, insulin resistance and Type II
diabetes. Clin. Sci. (Lond) 109, 243–256 (2005).
Weisberg, S. P. et al. CCR2 modulates inflammatory
and metabolic effects of high-fat feeding. J. Clin.
Invest. 116, 115–124 (2006).
Weisberg, S. P. et al. Obesity is associated
with macrophage accumulation in adipose tissue.
J. Clin. Invest. 112, 1796–1808 (2003).
Xu, H. et al. Chronic inflammation in fat plays a crucial
role in the development of obesity-related insulin
resistance. J. Clin. Invest. 112, 1821–1830 (2003).
Fantuzzi, G. Adipose tissue, adipokines, and
inflammation. J. Allergy Clin. Immunol. 115, 911–919
10. Kanda, H. et al. MCP-1 contributes to macrophage
infiltration into adipose tissue, insulin resistance,
and hepatic steatosis in obesity. J. Clin. Invest.
116, 1494–1505 (2006).
11. Hotamisligil, G. S., Shargill, N. S. & Spiegelman, B. M.
Adipose expression of tumor necrosis factor-α: direct
role in obesity-linked insulin resistance. Science
259, 87–91 (1993).
12. Kern, P. A. et al. The expression of tumor necrosis
factor in human adipose tissue. Regulation by obesity,
weight loss, and relationship to lipoprotein lipase.
J. Clin. Invest. 95, 2111–2119 (1995).
This study (together with reference 9) shows for
the first time that the pro-inflammatory cytokine
TNF is a mediator of insulin resistance in obesity.
13. Uysal, K. T., Wiesbrock, S. M., Marino, M. W. &
Hotamisligil, G. S. Protection from obesity-induced
insulin resistance in mice lacking TNF-α function.
Nature 389, 610–614 (1997).
14. Yuan, M. et al. Reversal of obesity- and diet-induced
insulin resistance with salicylates or targeted
disruption of IKKβ. Science 293, 1673–1677 (2001).
The authors describe a central role for IKKβ β in the
pathogenesis of insulin resistance.
15. Hirosumi, J. et al. A central role for JNK in obesity
and insulin resistance. Nature 420, 333–336 (2002).
This is the first report that JNK is a mediator of
obesity and insulin resistance.
16. Ozcan, U. et al. Endoplasmic reticulum stress links
obesity, insulin action, and type 2 diabetes. Science
306, 457–461 (2004).
17. Rui, L., Yuan, M., Frantz, D., Shoelson, S. &
White, M. F. SOCS-1 and SOCS-3 block insulin
signaling by ubiquitin-mediated degradation of IRS1
and IRS2. J. Biol. Chem. 277, 42394–42398 (2002).
18. Croker, B. A. et al. SOCS3 negatively regulates IL-6
signaling in vivo. Nature Immunol. 4, 540–545 (2003).
19. Arkan, M. C. et al. IKK-β links inflammation to obesity-
induced insulin resistance. Nature Med. 11, 191–198
This paper provides evidence that myeloid cells
(macrophages) regulate systemic insulin resistance
in an IKKβ β-dependent manner.
20. Cai, D. et al. Local and systemic insulin resistance
resulting from hepatic activation of IKK-β and NF-κB.
Nature Med. 11, 183–190 (2005).
21. Pineiro, R. et al. Adiponectin is synthesized and
secreted by human and murine cardiomyocytes.
FEBS Lett. 579, 5163–5169 (2005).
22. Delaigle, A. M., Jonas, J. C., Bauche, I. B., Cornu, O. &
Brichard, S. M. Induction of adiponectin in skeletal
muscle by inflammatory cytokines: in vivo and
in vitro studies. Endocrinology 145, 5589–5597
23. Wolf, A. M. et al. Up-regulation of the anti-
inflammatory adipokine adiponectin in acute liver
failure in mice. J. Hepatol. 44, 537–543 (2006).
24. Scherer, P. E., Williams, S., Fogliano, M., Baldini, G. &
Lodish, H. F. A novel serum protein similar to C1q,
produced exclusively in adipocytes. J. Biol. Chem.
270, 26746–26749 (1995).
25. Maeda, K. et al. cDNA cloning and expression of
a novel adipose specific collagen-like factor, apM1
(AdiPose Most abundant Gene transcript 1).
Biochem. Biophys. Res. Commun. 221, 286–289
26. Hu, E., Liang, P. & Spiegelman, B. M. AdipoQ is a
novel adipose-specific gene dysregulated in obesity.
J. Biol. Chem. 271, 10697–10703 (1996).
References 24–26 report the cloning and
identification of adiponectin in mice and humans.
27. Waki, H. et al. Generation of globular fragment
of adiponectin by leukocyte elastase secreted
by monocytic cell line THP-1. Endocrinology
146, 790–796 (2005).
28. Arita, Y. et al. Paradoxical decrease of an adipose-
specific protein, adiponectin, in obesity. Biochem.
Biophys. Res. Commun. 257, 79–83 (1999).
NATURE REVIEWS | IMMUNOLOGY
VOLUME 6 | OCTOBER 2006 | 781
© 2006 Nature Publishing Group
29. Waki, H. et al. Impaired multimerization of human
adiponectin mutants associated with diabetes.
Molecular structure and multimer formation of
adiponectin. J. Biol. Chem. 278, 40352–40363
30. Fisher, F. F. et al. Serum high molecular weight
complex of adiponectin correlates better with glucose
tolerance than total serum adiponectin in Indo-Asian
males. Diabetologia 48, 1084–1087 (2005).
31. Pajvani, U. B. et al. Complex distribution, not absolute
amount of adiponectin, correlates with
thiazolidinedione-mediated improvement in insulin
sensitivity. J. Biol. Chem. 279, 12152–12162 (2004).
32. Bobbert, T. et al. Changes of adiponectin oligomer
composition by moderate weight reduction. Diabetes
54, 2712–2719 (2005).
33. Yamauchi, T. et al. Cloning of adiponectin receptors
that mediate antidiabetic metabolic effects. Nature
423, 762–769 (2003).
This is the first report of the isolation and
characterization of adiponectin receptors 1 and 2.
34. Hug, C. et al. T-cadherin is a receptor for
hexameric and high-molecular-weight forms of
Acrp30/adiponectin. Proc. Natl Acad. Sci. USA
101, 10308–10313 (2004).
35. Maeda, N. et al. Diet-induced insulin resistance
in mice lacking adiponectin/ACRP30. Nature Med.
8, 731–737 (2002).
36. Fasshauer, M. et al. Adiponectin gene expression
and secretion is inhibited by interleukin-6 in 3T3-L1
adipocytes. Biochem. Biophys. Res. Commun.
301, 1045–1050 (2003).
37. Bruun, J. M. et al. Regulation of adiponectin by
adipose tissue-derived cytokines: in vivo and in vitro
investigations in humans. Am. J. Physiol. Endocrinol.
Metab. 285, E527-E533 (2003).
38. Maeda, N. et al. PPARγ ligands increase expression
and plasma concentrations of adiponectin, an
adipose-derived protein. Diabetes 50, 2094–2099
39. Iwaki, M. et al. Induction of adiponectin, a fat-derived
antidiabetic and antiatherogenic factor, by nuclear
receptors. Diabetes 52, 1655–1663 (2003).
40. Ouchi, N. et al. Novel modulator for endothelial
adhesion molecules: adipocyte-derived plasma protein
adiponectin. Circulation 100, 2473–2476 (1999).
41. Yokota, T. et al. Adiponectin, a new member of the
family of soluble defense collagens, negatively regulates
the growth of myelomonocytic progenitors and the
functions of macrophages. Blood 96, 1723–1732
42. Wolf, A. M., Wolf, D., Rumpold, H., Enrich, B. & Tilg, H.
Adiponectin induces the anti-inflammatory cytokines
IL-10 and IL-1RA in human leukocytes. Biochem.
Biophys. Res. Commun. 323, 630–635 (2004).
43. Yamaguchi, N. et al. Adiponectin inhibits Toll-like
receptor family-induced signaling. FEBS Lett.
579, 6821–6826 (2005).
44. Saijo, S., Nagata, K., Nakano, Y., Tobe, T. &
Kobayashi, Y. Inhibition by adiponectin of IL-8
production by human macrophages upon coculturing
with late apoptotic cells. Biochem. Biophys. Res.
Commun. 334, 1180–1183 (2005).
45. Neumeier, M. et al. Different effects of adiponectin
isoforms in human monocytic cells. J. Leukocyte Biol.
79, 803–808 (2006).
46. Berg, A. H., Combs, T. P. & Scherer, P. E. ACRP30/
adiponectin: an adipokine regulating glucose and lipid
metabolism. Trends Endocrinol. Metab. 13, 84–89
47. Nawrocki, A. R. et al. Mice lacking adiponectin show
decreased hepatic insulin sensitivity and reduced
responsiveness to peroxisome proliferator-activated
receptor-γ agonists. J. Biol. Chem. 281, 2654–2660
48. Shklyaev, S. et al. Sustained peripheral expression
of transgene adiponectin offsets the development of
diet-induced obesity in rats. Proc. Natl Acad. Sci. USA
100, 14217–14222 (2003).
49. Xu, A. et al. The fat-derived hormone adiponectin
alleviates alcoholic and nonalcoholic fatty liver
diseases in mice. J. Clin. Invest. 112, 91–100 (2003).
50. Kamada, Y. et al. Enhanced carbon tetrachloride-
induced liver fibrosis in mice lacking adiponectin.
Gastroenterology 125, 1796–1807 (2003).
51. Masaki, T. et al. Adiponectin protects LPS-induced
liver injury through modulation of TNF-α in KK-Ay
obese mice. Hepatology 40, 177–184 (2004).
52. Sennello, J. A. et al. Regulation of T cell-mediated
hepatic inflammation by adiponectin and leptin.
Endocrinology 146, 2157–2164 (2005).
53. Kahn, B. B., Alquier, T., Carling, D. & Hardie, D. G.
AMP-activated protein kinase: ancient energy
gauge provides clues to modern understanding of
metabolism. Cell Metab. 1, 15–25 (2005).
54. Yamauchi, T. et al. Adiponectin stimulates glucose
utilization and fatty-acid oxidation by activating AMP-
activated protein kinase. Nature Med. 8, 1288–1295
55. Shibata, R. et al. Adiponectin stimulates angiogenesis
in response to tissue ischemia through stimulation of
AMP-activated protein kinase signaling. J. Biol. Chem.
279, 28670–28674 (2004).
56. Shibata, R. et al. Adiponectin protects against
myocardial ischemia–reperfusion injury through
AMPK- and COX-2-dependent mechanisms. Nature
Med. 11, 1096–1103 (2005).
57. Kubota, N. et al. Disruption of adiponectin causes
insulin resistance and neointimal formation. J. Biol.
Chem. 277, 25863–25866 (2002).
58. Okamoto, Y. et al. Adiponectin reduces atherosclerosis
in apolipoprotein E-deficient mice. Circulation
106, 2767–2770 (2002).
59. Kawano, T. et al. Close association of
hypoadiponectinemia with arteriosclerosis obliterans
and ischemic heart disease. Metabolism 54, 653–656
60. Maahs, D. M. et al. Low plasma adiponectin levels
predict progression of coronary artery calcification.
Circulation 111, 747–753 (2005).
61. Iglseder, B. et al. Plasma adiponectin levels and
sonographic phenotypes of subclinical carotid artery
atherosclerosis: data from the SAPHIR Study. Stroke
36, 2577–2582 (2005).
62. Mandel, M. A. & Mahmoud, A. A. Impairment of cell-
mediated immunity in mutation diabetic mice (db/db).
J. Immunol. 120, 1375–1377 (1978).
63. Zhang, Y. et al. Positional cloning of the mouse obese
gene and its human homologue. Nature 372, 425–
This study reports the cloning of the gene encoding
leptin in mice and humans.
64. Lord, G. M. et al. Leptin modulates the T-cell immune
response and reverses starvation-induced
immunosuppression. Nature 394, 897–901 (1998).
The authors show for the first time that leptin
modulates CD4+ T-cell responses.
65. Friedman, J. M. & Halaas, J. L. Leptin and the
regulation of body weight in mammals. Nature 395,
66. Grunfeld, C. et al. Endotoxin and cytokines induce
expression of leptin, the ob gene product, in hamsters.
J. Clin. Invest. 97, 2152–2157 (1996).
67. Sarraf, P. et al. Multiple cytokines and acute
inflammation raise mouse leptin levels: potential
role in inflammatory anorexia. J. Exp. Med. 185,
68. Gainsford, T. et al. Leptin can induce proliferation,
differentiation, and functional activation of
hemopoietic cells. Proc. Natl Acad. Sci. USA 93,
69. Aleffi, S. et al. Upregulation of proinflammatory
and proangiogenic cytokines by leptin in human
hepatic stellate cells. Hepatology 42, 1339–1348
70. Matarese, G., Moschos, S. & Mantzoros, C. S. Leptin
in immunology. J. Immunol. 174, 3137–3142 (2005).
71. Zhao, T. et al. Globular adiponectin decreases leptin-
induced tumor necrosis factor-α expression by murine
macrophages: involvement of cAMP-PKA and MAPK
pathways. Cell Immunol. 238, 19–30 (2005).
72. Tian, Z., Sun, R., Wei, H. & Gao, B. Impaired natural
killer (NK) cell activity in leptin receptor deficient mice:
leptin as a critical regulator in NK cell development
and activation. Biochem. Biophys. Res. Commun.
298, 297–302 (2002).
73. Howard, J. K. et al. Leptin protects mice from
starvation-induced lymphoid atrophy and increases
thymic cellularity in ob/ob mice. J. Clin. Invest. 104,
74. Takahashi, N., Waelput, W. & Guisez, Y. Leptin is an
endogenous protective protein against the toxicity
exerted by tumor necrosis factor. J. Exp. Med. 189,
75. Faggioni, R. et al. Leptin-deficient (ob/ob) mice are
protected from T cell-mediated hepatotoxicity: role of
tumor necrosis factor-α and IL-18. Proc. Natl Acad.
Sci. USA 97, 2367–2372 (2000).
76. Siegmund, B., Lear-Kaul, K. C., Faggioni, R. &
Fantuzzi, G. Leptin deficiency, not obesity, protects
mice from Con A-induced hepatitis. Eur. J. Immunol.
32, 552–560 (2002).
77. Matarese, G. et al. Requirement for leptin in the
induction and progression of autoimmune
encephalomyelitis. J. Immunol. 166, 5909–5916
78. Siegmund, B. et al. Leptin receptor expression
on T lymphocytes modulates chronic intestinal
inflammation in mice. Gut 53, 965–972 (2004).
79. Holcomb, I. N. et al. FIZZ1, a novel cysteine-rich
secreted protein associated with pulmonary
inflammation, defines a new gene family. EMBO J.
19, 4046–4055 (2000).
80. Steppan, C. M. et al. The hormone resistin links
obesity to diabetes. Nature 409, 307–312 (2001).
This study (together with references 79 and 81)
reports the cloning and identification of resistin,
a thiazolidinedione-regulated, adipocyte-derived
protein that mediates insulin resistance in mice.
81. Kim, K. H., Lee, K., Moon, Y. S. & Sul, H. S. A cysteine-
rich adipose tissue-specific secretory factor inhibits
adipocyte differentiation. J. Biol. Chem. 276,
82. Patel, S. D., Rajala, M. W., Rossetti, L., Scherer, P. E.
& Shapiro, L. Disulfide-dependent multimeric
assembly of resistin family hormones. Science 304,
83. Kaser, S. et al. Resistin messenger-RNA expression is
increased by proinflammatory cytokines in vitro.
Biochem. Biophys. Res. Commun. 309, 286–290
84. Lehrke, M. et al. An inflammatory cascade leading to
hyperresistinemia in humans. PLoS Med. 1, e45
85. Bajaj, M., Suraamornkul, S., Hardies, L. J.,
Pratipanawatr, T. & DeFronzo, R. A. Plasma resistin
concentration, hepatic fat content, and hepatic and
peripheral insulin resistance in pioglitazone-treated
type II diabetic patients. Int. J. Obes. Relat. Metab.
Disord. 28, 783–789 (2004).
86. Bokarewa, M., Nagaev, I., Dahlberg, L., Smith, U. &
Tarkowski, A. Resistin, an adipokine with potent
proinflammatory properties. J. Immunol. 174,
87. Silswal, N. et al. Human resistin stimulates the
pro-inflammatory cytokines TNF-α and IL-12 in
macrophages by NF-κB-dependent pathway.
Biochem. Biophys. Res. Commun. 334, 1092–1101
88. Verma, S. et al. Resistin promotes endothelial
cell activation: further evidence of adipokine–
endothelial interaction. Circulation 108, 736–740
89. Savage, D. B. et al. Resistin/Fizz3 expression in
relation to obesity and peroxisome proliferator-
activated receptor-γ action in humans. Diabetes 50,
90. McTernan, C. L. et al. Resistin, central obesity, and
type 2 diabetes. Lancet 359, 46–47 (2002).
91. Utzschneider, K. M. et al. Resistin is not associated
with insulin sensitivity or the metabolic syndrome in
humans. Diabetologia 48, 2330–2333 (2005).
92. Jung, H. S. et al. Resistin is secreted from
macrophages in atheromas and promotes
atherosclerosis. Cardiovasc. Res. 69, 76–85 (2006).
93. Kawanami, D. et al. Direct reciprocal effects of
resistin and adiponectin on vascular endothelial cells:
a new insight into adipocytokine–endothelial cell
interactions. Biochem. Biophys. Res. Commun. 314,
94. Axelsson, J. et al. Elevated resistin levels in chronic
kidney disease are associated with decreased
glomerular filtration rate and inflammation, but not
with insulin resistance. Kidney Int. 69, 596–604
95. Fukuhara, A. et al. Visfatin: a protein secreted by
visceral fat that mimics the effects of insulin. Science
307, 426–430 (2005).
This study identifies visfatin as a new
adipocytokine that is preferentially expressed in
visceral fat and that mimics insulin activity by
binding and activating the insulin receptor.
96. Samal, B. et al. Cloning and characterization of the
cDNA encoding a novel human pre-B-cell colony-
enhancing factor. Mol. Cell Biol. 14, 1431–1437
97. Ye, S. Q. et al. Pre-B-cell colony-enhancing factor
as a potential novel biomarker in acute lung injury.
Am. J. Respir. Crit. Care Med. 171, 361–370 (2005).
98. Jia, S. H. et al. Pre-B cell colony-enhancing factor
inhibits neutrophil apoptosis in experimental
inflammation and clinical sepsis. J. Clin. Invest. 113,
782 | OCTOBER 2006 | VOLUME 6
© 2006 Nature Publishing Group
99. Hida, K. et al. Visceral adipose tissue-derived serine Download full-text
protease inhibitor: a unique insulin-sensitizing
adipocytokine in obesity. Proc. Natl Acad. Sci. USA
102, 10610–10615 (2005).
100. Yang, Q. et al. Serum retinol binding protein 4
contributes to insulin resistance in obesity and type 2
diabetes. Nature 436, 356–362 (2005).
101. Wang, Y. et al. Adiponectin inhibits cell proliferation
by interacting with several growth factors in an
oligomerization-dependent manner. J. Biol. Chem.
280, 18341–18347 (2005).
102. Brakenhielm, E. et al. Adiponectin-induced
antiangiogenesis and antitumor activity involve
caspase-mediated endothelial cell apoptosis. Proc.
Natl Acad. Sci. USA 101, 2476–2481 (2004).
103. Bub, J. D., Miyazaki, T. & Iwamoto, Y. Adiponectin as a
growth inhibitor in prostate cancer cells. Biochem.
Biophys. Res. Commun. 340, 1158–1166 (2006).
104. Garofalo, C. & Surmacz, E. Leptin and cancer. J. Cell.
Physiol. 207, 12–22 (2005).
105. Ishikawa, M. et al. Plasma adiponectin and gastric
cancer. Clin. Cancer Res. 11, 466–472 (2005).
106. Miyoshi, Y. et al. Association of serum adiponectin
levels with breast cancer risk. Clin. Cancer Res.
9, 5699–5704 (2003).
107. Petridou, E. et al. Plasma adiponectin concentrations
in relation to endometrial cancer: a case–control
study in Greece. J. Clin. Endocrinol. Metab. 88,
108. Goktas, S. et al. Prostate cancer and adiponectin.
Urology 65, 1168–1172 (2005).
109. Wei, E. K., Giovannucci, E., Fuchs, C. S., Willett, W. C.
& Mantzoros, C. S. Low plasma adiponectin levels and
risk of colorectal cancer in men: a prospective study.
J. Natl Cancer Inst. 97, 1688–1694 (2005).
110. Shore, S. A. et al. Effect of leptin on allergic airway
responses in mice. J. Allergy Clin. Immunol. 115,
111. Barbier, M. et al. Overexpression of leptin mRNA in
mesenteric adipose tissue in inflammatory bowel
diseases. Gastroenterol. Clin. Biol. 27, 987–991
112. Yamamoto, K. et al. Production of adiponectin, an
anti-inflammatory protein, in mesenteric adipose
tissue in Crohn’s disease. Gut 54, 789–796 (2005).
113. Siegmund, B. et al. Development of intestinal
inflammation in double IL-10- and leptin-deficient
mice. J. Leukocyte Biol. 76, 782–786 (2004).
114. Blain, A. et al. Crohn’s disease clinical course and
severity in obese patients. Clin. Nutr. 21, 51–57
115. Schaffler, A. et al. Adipocytokines in synovial fluid.
JAMA 290, 1709–1710 (2003).
116. Symmons, D. P. Epidemiology of rheumatoid arthritis:
determinants of onset, persistence and outcome.
Best Pract. Res. Clin. Rheumatol. 16, 707–722
117. Yokota, T. et al. Adiponectin, a fat cell product,
influences the earliest lymphocyte precursors in bone
marrow cultures by activation of the cyclooxygenase–
prostaglandin pathway in stromal cells. J. Immunol.
171, 5091–5099 (2003).
118. Pagano, C. et al. Increased serum resistin in
nonalcoholic fatty liver disease is related to liver
disease severity and not to insulin resistance. J. Clin.
Endocrinol. Metab. 91, 1081–1086 (2006).
119. Ognjanovic, S. & Bryant-Greenwood, G. D. Pre-B-cell
colony-enhancing factor, a novel cytokine of human
fetal membranes. Am. J. Obstet. Gynecol. 187,
We gratefully acknowledge A. Kaser for helpful discussions
and critical reading of the manuscript. We are supported by
grants from the Austrian Science Foundation and the
Christian-Doppler Research Society (Austria).
Competing interests statement
The authors declare no competing financial interests.
The following terms in this article are linked online to:
adiponectin | ADIPOR1 | ADIPOR2 | APOE | CCL2 | CRP |
CXCL8 | GLUT4 | granulocyte colony-stimulating factor |
ICAM1 | IKKβ | IL-1 | IL-1RA | IL-4 | IL-6 | IL-10 | IL-12 | IFNγ |
IRS1 | JNK1 | leptin | OBR | pentraxin-3 | plasminogen-
activator inhibitor type 1 | PPARγ | RBP4 | resistin | SOCS1 |
SOCS2 | SOCS3 | SREBP1C | TNF | vascular endothelial
growth factor | VASPIN | VCAM1 | visfatin | XBP1
Access to this links box is available online.
NATURE REVIEWS | IMMUNOLOGY
VOLUME 6 | OCTOBER 2006 | 783
© 2006 Nature Publishing Group