(Journal of Leukocyte Biology. 2002;71:388-400.)
© 2002
by Society for Leukocyte Biology
The role of PPARs in inflammation and immunity
Robert B. Clark
Division of Rheumatic Diseases, Department of Medicine, University of Connecticut Health Center, Farmington
Correspondence: Robert B. Clark, M.D., Division of Rheumatic Diseases, Department of Medicine, University of Connecticut Health Center, 263 Farmington Avenue, Farmington, CT 06030. E-mail: rclark{at}nso2.uchc.edu
 |
ABSTRACT
|
|---|
The family of transcription factors termed peroxisome
proliferator-activated receptors (PPARs) has recently been the focus of
much interest for their possible role in the regulation of inflammation
and immune responses. PPAR
and PPAR
have been implicated in the
regulation of macrophage and endothelial cell inflammatory responses.
Although PPAR activation has generally been shown to have
anti-inflammatory effects, opposite effects have been noted, and
results often appear to depend on the ligands being used and the
inflammatory parameters being measured. Recently, my laboratory and
others have described a role for PPAR in the responses of T
lymphocytes. Ligands for PPAR have been found to inhibit
proliferation of activated T cells, and this appears to involve
inhibition of IL-2 secretion and/or the induction of apoptosis.
However, one problem in the interpretation of many of the studies of
PPAR, inflammation, and immunity is that ligands thought to be
specific for PPAR may have regulatory effects on inflammatory
parameters that are PPAR-independent. Future studies of the role of
the PPARs in inflammatory and immune responses should include further
studies of T cells, T-cell subsets, and dendritic cells but will have
to re-examine the issue of PPAR specificity of the ligands being used.
This may require further knockout studies and technology, together with
the identification of endogenous and perhaps more specific synthetic
PPAR ligands.
Key Words: T cells • macrophages • endothelial cells • thiazolidinediones
|
INTRODUCTION
|
|---|
Until recently, the family of transcription factors termed
peroxisome proliferator-activated receptors (PPARs) was believed to
regulate genes predominantly associated with lipid and glucose
metabolism. However, understanding the role played by PPARs in the
regulation of inflammation and immunity is evolving rapidly. In this
review, we will examine the role of PPARs in inflammation and immunity,
focusing primarily on the role played by PPAR and the newly
described role of PPAR in T-cell biology.
|
THE PPAR FAMILY OF NUCLEAR RECEPTORS
|
|---|
Peroxisomes are subcellular organelles found in most plant and
animal cells that perform diverse metabolic functions including
H2O2-based respiration, ß-oxidation of fatty
acids, and cholesterol metabolism. In rodents, a large class of
structurally diverse industrial and pharmaceutical chemicals including
herbicides, industrial solvents, and hypolipidemic drugs leads to
significant increases in the size and number of peroxisomes in liver,
liver hypertrophy, liver hyperplasia, hepatocarcinogenesis, and
transcription of genes encoding peroxisomal enzymes. The structurally
diverse compounds that induce these effects are termed peroxisome
proliferators. In 1990, a nuclear receptor that was transcriptionally
activated by peroxisome proliferators was identified
[1
]. Because the peroxisome proliferators were initially
demonstrated to activate but not to directly bind this receptor, the
receptor was named PPAR.
It is now known that PPARs are a family of transcription factors
belonging to the nuclear receptor superfamily. PPARs regulate numerous
genes through ligand-dependent transcriptional activation and
repression, and until recently, the genes regulated were believed to be
those predominantly associated with lipid metabolism
[2
3
4
]. Three different members of the PPAR family have
been identified, encoded by separate genes: PPAR, PPAR
(also
called beta or NUC-1), and PPAR [5
6
7
8
]. These three
isotypes exhibit distinct patterns of tissue distribution and differ in
their ligand-binding domains. PPAR is expressed ubiquitously and
binds some of the same ligands as PPAR, however its function remains
unclear. PPAR and PPAR will be discussed in greater detail below.
|
GENERAL PHYSIOLOGY OF PPARs
|
|---|
Like other members of the nuclear receptor superfamily, PPARs
possess a central DNA-binding domain that recognizes DNA sequences,
termed PPAR-response elements (PPREs), in the promoter regions of their
target genes. PPARs heterodimerize with another member of the
nuclear-receptor superfamily, the retinoid X receptors (RXR), and the
transcription regulation of target genes by PPARs is actually achieved
through the binding of these PPAR-RXR heterodimers to PPREs
[2
, 3
, 9
, 10
].
RXR also exists in multiple isoforms, RXR, ß, and , and like
the PPARs, have a variable tissue distribution [11
,
12
]. RXR isoforms are activated by 9-cis retinoic acid
[2
]. It is not known if any one of the particular RXR
isoforms preferentially binds one or more of the PPAR isoforms.
RXR also forms heterodimers with other members of the nuclear receptor
superfamily, and these interactions influence the PPAR-regulated
transcriptional activation because of the competition among various RXR
heterodimerization partners for RXR [4
]. In the presence
of ligands for PPAR, the PPAR:RXR heterodimer does not require that
9-cis retinoic acid be present for transcriptional activation. However,
when combined as PPAR:RXR heterodimer, PPAR ligands and 9-cis retinoic
acid can act synergistically on PPAR responses [2
]. The
different heterodimers of RXR (e.g., PPAR:RXR) allow for specific
responses by binding to highly specific sequences in the promoter
regions of the genes they transactivate [10
].
Although the PPAR:RXR dimer is the focus for determining specific gene
transcription on ligand activation, transactivation of a particular
gene actually requires a large complex of proteins [13
,
14
]. Thus, the regulation of PPAR-regulated
transcriptional activation is made more complex by the involvement of
coactivators and corepressors [15
, 16
]. In
the inactivated state, the PPARs are believed to be in complexes bound
with corepressor proteins. In this state, in some but not all cell
types, PPARs may have a cytoplasmic rather than a nuclear location
[17
, 18
]. Upon ligand activation, PPARs
dissociate from corepressors and recruit coactivators, including the
PPAR-binding protein [15
] and the steroid receptor
coactivator-1 [16
], and can translocate from the
cytoplasm to the nucleus [18
].
|
PPAR—GENERAL PHYSIOLOGY
|
|---|
PPAR was the first member of the family identified, and it is
believed to be solely responsible for the effects of the peroxisomal
proliferators described above [1
]. PPAR expression is
relatively high in hepatocytes, heart, enterocytes, muscle, and the
kidney. PPAR regulates genes involved in the ß-oxidation of fatty
acids and lipoprotein metabolism. Using PPAR-deficient mice, this
receptor was shown to be involved in high-density lipoprotein and
triglyceride metabolism and in hepatic regulation of apolipoprotein and
fatty acid ß-oxidation enzyme expression [19
,
20
]. In addition to the structurally diverse compounds
described above as peroxisomal proliferators, fibrates (used as drugs
for the reduction of high triglyceride levels), specific fatty acids,
and eicosanoids also can act as ligands for PPAR
[21
]. In studies involving PPAR and
inflammation/immune responses, the relevant ligands studied most
frequently have been the naturally occurring ligand, leukotriene
B4 (LTB4), and the synthetic ligands,
fenofibrate and Wyeth-14,643.
|
PPAR—GENERAL PHYSIOLOGY
|
|---|
PPAR was characterized originally as a key regulator of
adipocyte differentiation and lipid metabolism [7
,
22
23
24
]. PPAR expression is directed by different
promoters, leading to three PPAR isoforms [25
,
26
]. Until recently, it had been believed that PPAR1
isoform expression was restricted to liver, adipocytes, and a few other
cell types and that the PPAR2 isoform was expressed predominantly in
adipocytes [7
, 22
]. It is now clear that
PPAR is also found in other cell types including fibroblasts,
myocytes, breast cells, the white and red pulp of rat spleen, human
bone-marrow precursors, and macrophages/monocytes [27
,
28
]. In addition, PPAR has been shown in macrophage
foam cells in atherosclerotic plaques [29
,
30
]. An important role for PPAR in glucose metabolism
was identified when it was demonstrated that a class of antidiabetic
drugs, the thiazolidinediones, were high-affinity PPAR ligands
[31
]. The thiazolidinediones were developed originally
for the treatment of type-2 diabetes on the basis of their ability to
lower glucose levels (and levels of circulating fatty acids) in rodent
models of insulin resistance. The finding that the thiazolidinediones
mediate their therapeutic effects through direct interactions with
PPAR established PPAR as a key regulator of glucose and lipid
homeostasis [32
]. Despite being described initially as a
regulator of lipid and glucose metabolism, PPAR has also been
demonstrated recently to have a role in cell proliferation and
malignancy. Ligands for PPAR have been shown to mediate positive and
negative effects on cell proliferation and malignancy
[16
, 33
34
35
36
37
38
39
].
In addition to the thiazolidindione class of antidiabetic drugs, a
variety of nonsteroidal anti-inflammatory drugs also can function as
PPAR ligands, although the latter have relatively low affinity
[40
]. The prostaglandin D2
(PGD2) dehydration product PGJ2 was the first
endogenous ligand discovered for PPAR [41
,
42
]. The additional PGD2 dehydration product,
15-deoxy-
12,14-PGJ2 (15d-PGJ2),
is also a naturally occurring substance that binds directly to PPAR
and is a potent ligand for PPAR activation [41
,
42
]. In addition, components of oxidized low-density
lipoproteins (OxLDL), including 9- and 13-hydroxyoctadecadienoic acid
(HODE), have been described as endogenous PPAR activators
[43
]. 12- and 15-Hydroxyeicosatetaenoic acid (HETE) are
also PPAR ligands [44
]. Although many naturally
occurring fatty acids and their metabolites can activate PPAR, they
bind with relatively low affinities and must be added to cells at high
concentrations to stimulate transcription. It has been difficult
therefore to establish the physiological relevance of any of these
substances as actual in vivo regulators of PPAR [45
].
For a more detailed review of the structure and physiology of PPARs,
see Ricote et al. [46
].
|
PPAR IN INFLAMMATION AND IMMUNE RESPONSES
|
|---|
In vivo studies of inflammatory/immune functions of PPAR
In vivo studies of the inflammatory/immune-related functions of
PPAR have yielded a somewhat muddled picture. In vivo evidence for a
role of PPARs in inflammation was first suggested in studies using
PPAR knockout mice. PPAR knockout mice are viable but lack
responses to appropriate ligands (i.e., no peroxisome proliferation, no
gene activation, no hepatomegaly) and exhibit abnormalities in
triglyceride and cholesterol metabolism [47
,
48
]. In an early study, inflammatory responses were
induced in vivo in the ears of mice using LTB4, the
phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA), or
arachidonic acid, and the responses of wild-type mice were
compared with that of PPAR null mice [49
].
Inflammation induced by LTB4 or arachidonic acid, but not
TPA, was prolonged in PPAR null mice. In light of the fact that
anything affecting the degradation rate of LTB4 can affect
the extent and duration of an inflammatory reaction, and given that
direct interaction between PPAR and LTB4 induces enzymes
for fatty acid degradation and as such increases the catabolism of
LTB4, the investigators proposed that LTB4
induces its own catabolism via PPAR activation [49
].
In contrast to the anti-inflammatory effects of PPAR activation
suggested by the former study, another in vivo study involved the
treatment of CD-1 mice with fenofibrate or Wy14,643. Such treated mice
had fivefold higher lipopolysaccharide (LPS)-induced tumor necrosis
factor a (TNF-) plasma levels and a significantly lower, 50% lethal
dose than control mice [50
]. Using PPAR null mice,
this result was confirmed to be mediated by PPAR
[50
]. In these same studies, however, peritoneal
macrophages from wild-type mice treated with Wy14,643 showed modestly
decreased (rather than increased) TNF expression in vitro. Thus, the
systemic effects of the PPAR agonists represent a complicated
mixture of effects [50
]. Finally, a further in vivo
study of the inflammatory/immune relevance of PPAR involved the
aging process [51
]. Previously, it had been shown that
in aged mice, oxidative stress-induced, redox-regulated transcription
of nuclear factor-
B (NF-B) becomes constitutive in many tissues.
It was then demonstrated that the administration of PPAR agonists to
aged mice restored the cellular-redox balance and resulted in an
elimination of the constitutively active NF-B and in a resultant
loss in spontaneous inflammatory cytokine production
[51
]. Furthermore, aged PPAR null mice did not show
these changes after administration of PPAR agonists. Finally, aged
C57BL/6 mice were found to express reduced PPAR transcripts that
increased with the addition of PPAR ligands. Thus, these
investigators suggested that PPAR may play a role in the evolution
of the oxidative stress excesses observed in aging [51
].
In vitro studies of inflammatory/immune functions of PPAR
The majority of the investigations of the role of PPAR in
inflammation/immunity involves in vitro investigations. It has been
demonstrated that PPAR is expressed in murine [50
,
52
] and human monocyte-macrophages [53
].
Also relevant to its function in inflammation/immunity, PPAR has
been found to be expressed in various type of human endothelial cells
(ECs) [54
55
56
]. However, PPAR is not expressed in
murine-dendritic cells [57
], in human mast cells
[58
], or in murine bone marrow-derived mast cells
[59
].
Monocytes-macrophages
Using the activated RAW264.7 murine-macrophage cell line,
proinflammatory and anti-inflammatory effects of PPAR ligands have
been demonstrated [52
]. It was found that the two
naturally occurring PPAR ligands, LTB4 and 8(S)-HETE,
stimulated nitric oxide synthase (NOS) activity. However, the synthetic
ligand Wy14,643 inhibited NOS activity [52
]. The
difference between the natural and synthetic ligands was postulated to
be a result of the low specificity (and thus multiple effects) of the
natural compounds compared with the more selective effects of the
synthetic compounds [60
]. In human monocytes, PPAR
has been shown to be constitutively present in undifferentiated
monocytes and to reside in the cytoplasm [53
]. Ligands
for PPAR were not able to induce apoptosis of inactivated
(differentiated) macrophages but did cause apoptosis of activated
macrophages [53
]. An additional study demonstrating the
anti-inflammatory effects of PPAR ligation involved the THP-1 human
monocytic leukemia cell line [61
]. In this study,
fenofibrate did not affect the LPS-induced secretion of interleukin
(IL)-8 but induced significant down-regulation of LPS-induced secretion
of matrix metalloprotein-9 (MMP-9).
ECs
Several studies have suggested a role for PPAR in the
down-regulation of EC-inflammatory responses. An early event in
inflammation is the expression of specific adhesion molecules on the
surface of ECs, which subsequently bind leukocytes. PPAR was found
to be expressed in human aortic ECs (HAECs), and WY14,643 was
demonstrated to partially inhibit the phorbol 12-myristate 13-acetate
(PMA)- and LPS-induced expression of vascular cell-adhesion molecule-1
(VCAM-1) on HAECs and had no effect on the expression of intercellular
adhesion molecule-1 (ICAM-1), E-selectin, or neutrophil-like HL-60
cell-binding to activated HAECs [54
]. In a confirmatory
study, human carotid artery ECs were found to express PPAR
[56
]. When these ECs were precultured with Wy14,643 or
fenofibrate, TNF--induced VCAM-1 expression was inhibited in a time-
and concentration-dependent manner (an effect not seen with PPAR
agonists) [56
]. PPAR has also been found to be
expressed in human aortic smooth-muscle cells, and in these cells,
PPAR ligands inhibit IL-1-induced production of IL-6 and PG and
inhibit the expression of cyclooxygenase-2 (COX-2) [62
].
In hyperlipidemic patients, fenofibrate decreases the plasma
concentrations of IL-6, fibrinogen, and C-reactive protein
[62
]. Confirmatory experiments demonstrated that aortic
explants from PPAR null mice stimulated with LPS secreted increased
amounts of IL-6 and that fibrates decreased IL-6 mRNA levels in
LPS-stimulated aortas from wild-type but not from PPAR null mice
[63
]. In human aortic smooth-muscle cells, fibrates were
found to inhibit IL-1-induced IL-6 gene expression [63
].
In contrast to the majority of studies that demonstrate
anti-inflammatory, regulatory effects of PPAR in ECs, another study
of HAECs supported a proinflammatory role for PPAR. This
proinflammatory effect was seen in the mediation of EC production of
monocyte-chemotactic activity in response to lipids
[55
]. In addition, Wy14,643 was found to stimulate HAECs
to synthesize IL-8 and monocyte-chemoattractant protein-1 (MCP-1).
Mechanisms of PPAR effects
The mechanisms of the anti- or proinflammatory effects of PPAR
have been studied extensively. As recently summarized by Delerive et
al. [64
], many mechanisms have been described possibly
underlying the anti-inflammatory effects of PPAR seen in
macrophage/monocytes and ECs. For example, direct protein-protein
interactions between PPAR and activated protein-1 (AP-1) and NF-B
proteins have been invoked as mechanisms of negative regulation of
inflammatory responses [63
]. In addition, by
up-regulating antioxidant enzyme activities, PPAR ligands reduce the
oxidative stress and thus may inhibit NF-B activation
[51
]. Finally, Delerive et al. [64
] have
demonstrated that fibrates induce IB expression in a
PPAR-dependent manner. This induction results in an inhibition of
NF-B DNA binding, leading to a sharp reduction of the p65-mediated
gene activation.
Conclusion—PPAR effects on inflammation/immune responses
Based on in vivo and in vitro studies with different cell types,
PPAR ligands appear to have a largely anti-inflammatory effect.
However, the effects have been found to vary, and in some studies,
proinflammatory effects have been demonstrated (Table 1
). It seems that in monocyte/macrophages, the major effects appear
to be apoptotic for activated macrophages and an inhibition of a subset
of macrophage proinflammatory cytokines. In ECs, the major effects
appear to be a decrease in the vascular expression of VCAM-1, but in
contrast, a proinflammatory role for PPAR has been shown in the
mediation of EC production of monocyte-chemotactic activity in response
to lipids. In vivo, PPAR appears to play an anti-inflammatory role
in LTB4-mediated inflammation, perhaps through a role in
the degradation of the LTB4, while other in vivo studies
reveal proinflammatory effects of PPAR ligands, including an
elevation of plasma TNF- and a decrease in LD50 in response to LPS.
Finally, PPAR may play a role in the regulation of aging-related
abnormalities in inflammation. The majority of these effects on
inflammation and immunity appear to be mediated through effects on
NF-B, but other pathways are also likely involved.
|
PPAR IN INFLAMMATION AND IMMUNITY
|
|---|
Until recently, evidence for PPAR expression and function in
the immune system had been limited. Greene et al. [28
]
screened a human bone-marrow cDNA library and found that peripheral
blood lymphocytes expressed only a truncated PPAR transcript, which
the authors believed could not encode all of the PPAR functional
domains. Braissant et al. [27
] studied PPAR expression
in rat tissues using in situ hybridization and immunohistochemistry and
described the expression of PPAR in the white and red pulp of the
rat spleen as well as in the Peyer’s patches of the rat. Finally, in
relating PPARs to inflammation, Gilroy et al. [65
],
studying COX-2 inhibitors in a rat model of carrageenin-induced pleural
inflammation, presented evidence for an anti-inflammatory role of
PGD2 and 15d-PGJ2, suggesting a possible role
for PPAR in inflammation.
Role of PPAR in inflammatory and immune functions of
monocytes/macrophages
One of the earliest findings associating PPARs and macrophages was
that PPAR was highly expressed in macrophage-derived foam cells of
human and murine atherosclerotic lesions [29
,
30
, 66
]. Subsequently, it has been
demonstrated that PPAR is expressed in human and murine
monocytes/macrophages. Functionally, PPAR has been shown to play a
role in the differentiation and activation of monocytes and in the
regulation of inflammatory activities [43
,
46
, 66
67
68
] (Table 2
).
Expression of PPAR in monocytes/macrophages
Many studies have suggested a link between the state of
macrophage/monocyte differentiation or activation and PPAR
expression. In the mouse, PPAR is expressed at low levels in
nonactivated monocytes/macrophages, and higher levels are expressed in
activated peritoneal macrophages [30
]. A similar
relationship between the state of differentiation and activation and
PPAR expression has been described for human peripheral blood
monocytes [53
]. Furthermore, activation of
monocyte/macrophages or monocyte lines with phorbol esters, OxLDL,
macrophage colony-stimulating factor (M-CSF), and
granulocyte-macrophage CSF (GM-CSF) induced an increased expression of
PPAR [30
, 53
]. In recent studies, which
have begun to link monocyte/macrophage PPAR to the adaptive immune
system, Huang et al. [69
] investigated the effects of
cytokines on macrophage PPAR expression and function. They found
that IL-4 strongly induced PPAR1 expression in resident peritoneal
macrophages and human peripheral blood monocytes and that IL-4 not only
up-regulated PPAR expression in macrophages but also enhanced the
activation of PPAR via the production of the endogenous PPAR
ligands, 13-HODE, 12-HETE, and 15-HETE [69
]. As such,
this study may represent an important documentation of the in vivo
function of endogenous PPAR ligands. In addition to PPAR
expression being up-regulated after macrophage activation, evidence
suggests that PPAR activation can itself lead to monocyte
differentiation in a cell line [66
]. Finally, given the
relevance of dendritic cells in the initiation, and perhaps in the
regulation, of adaptive immune responses, it is important to note that
one study has demonstrated the expression of PPAR (but not PPAR)
in immature and mature murine, spleen-derived, dendritic cells
[57
]. Furthermore, a thiazolidinedione (rosiglitazone)
did not interfere with the in vitro maturation of dendritic cells nor
did it modify their ability to activate naive T cells in vivo. However,
PPAR activators were found to down-modulate the CD-40-induced
secretion of IL-12 [57
]. Additional studies of PPAR
expression and function in dendritic cells will now be needed to
confirm these results.
Role of PPAR in monocytes/macrophage inflammatory responses
Many studies have demonstrated that PPAR ligands inhibit
macrophage-inflammatory responses. Previously, this subject has been
well reviewed [46
, 60
] and will be
summarized briefly here. The anti-inflammatory effects of PPAR
activation have been demonstrated with human and murine
monocyte/macrophages and monocyte/macrophage lines. Activation of
macrophages normally leads to the secretion of several different
proinflammatory mediators. Treatment with 15d-PGJ2 or
thiazolidinediones has been found to inhibit the secretion of many of
these mediators (including gelatinase B, IL-6, TNF-, and IL-1ß)
and also to reduce the induced expression of inducible NOS (iNOS) and
the transcription of the scavenger receptor-A gene [67
,
68
]. In addition, Azuma et al. [70
]
demonstrated that dPGJ2 as well as 13-HODE inhibited
LPS-induced IL-10 and IL-12 production by macrophages. Finally,
Chinetti et al. [53
] have demonstrated that in human
monocyte/macrophages, ligand activation of PPAR (but not PPAR)
resulted in apoptosis of nonactivated macrophages and that both PPAR
and PPAR ligands induced apoptosis of macrophages that had been
activated.
Although many studies have demonstrated inhibitory effects, others
suggest that PPAR ligands lead to a more complex pattern of
macrophage-inflammatory responses. PPAR ligands have been found to
stimulate the expression of the proinflammatory receptors CD14 and
CD11b/CD18 and to increase expression of class B scavenger receptors
(CD36 and SR-B1) [66
, 71
, 72
].
Although human peripheral blood monocyte-stimulated production of
TNF- and IL-6 was inhibited by 15d-PGJ2, four other
high-affinity PPAR ligands failed to affect cytokine production
[73
]. In vivo studies with db/db mice challenged with
LPS and treated with a thiazolidinedione showed no suppression of
cytokine production and, in fact, showed higher blood levels of TNF-
and IL-6 than the controls [73
]. Other examples of the
complexity of the effects shown for PPAR activation include the
finding that rosiglitazone (and fibrates) failed to modulate
LPS-induced secretion of IL-8 while down-regulating MMP-9 in a human
monocytic line [61
] and that 15d-PGJ2
induced IL-8 gene expression and suppressed MCP-1 but did not affect
expression of RANTES (regulated on activation, normal T expressed and
secreted) in human monocyte/macrophages [74
]. The latter
investigators also showed that 15d-PGJ2 potentiated
LPS-induced but suppressed PMA-induced expression of IL-8 mRNA. The
effects of PPAR ligands on monocyte/macrophage inflammatory
responses are obviously not simple and among other parameters, appear
to depend on the PPAR ligands used, the mode by which
monocytes/macrophages are activated, and the inflammatory responses
measured. Finally, there have been no studies describing the effects of
PPAR activation on the antigen-presenting function of
macrophage/monocytes, and such studies could prove valuable in
understanding the role of PPAR in the adaptive immune response.
PPAR and ECs
The other major cell type studied relating PPAR to inflammation
and immunity is the EC. ECs play an important role in homing
inflammatory and immune-relevant cells and thus in the localization of
inflammatory and immune responses. ECs from various sources have been
shown to express PPAR, and agonists have been shown to mediate
effects on cell survival, surface-protein expression, and cytokine and
chemokine expression. As with the studies of monocyte/macrophages, the
results of these endothelial studies, when taken together, do not
present an easily unified picture (Table 3
). Bishop-Bailey and Hla [75
] demonstrated that ECs
expressed PPAR, , and and that 15d-PGJ2 and a
thiazolidinedione induced endothelial apoptosis. Anti-inflammatory
effects shown include a thiazolidinedione-induced decrease in the
levels of IL-8 and MCP-1 in HAECs [55
] and an inhibition
by PPAR ligands of the interferon- (IFN-)-induced expression
of CXC chemokines but not a CC chemokine [76
]. The
latter investigators suggested that PPAR ligands might thus
attenuate the recruitment of activated T cells at sites of T-helper
cell type 1 (Th1)-mediated inflammation [76
]. However, a
dual effect of troglitazone on human vascular ECs was demonstrated in
which there was an increase in basal ICAM-1 expression but an
inhibition of TNF--induced ICAM-1 expression [77
].
Jackson et al. [54
] showed that in HAECs, ICAM-1
induction by PMA was unaffected by any of the PPAR ligands. They found,
however, that PPAR and activators (although not BRL49653)
partially inhibited the induced expression of VCAM-1. In contrast, Marx
et al. [56
] reported that PPAR agonists did not
inhibit expression of TNF--induced VCAM-1 as did PPAR agonists.
Overall, a consistent and integrated picture of the effect of PPAR
ligands on the inflammatory/immune aspects of endothelial function
awaits further study.
|
PPAR AND INFLAMMATION/IMMUNITY IN DISEASE AND DISEASE MODELS
|
|---|
The relevance of PPAR has been studied in several human
autoimmune diseases and animal models of autoimmune diseases. Kawahito
et al. [78
] demonstrated that synovial tissue expressed
PPAR in patients with rheumatoid arthritis (RA). PPAR was found
to be highly expressed in macrophages, and modest expression was noted
in synovial-lining fibroblasts and ECs. Activation of PPAR by
15d-PGJ2 and troglitazone induced RA synoviocyte apoptosis
in vitro [78
].
Alzheimer’s disease (AD) is characterized by the extracellular
deposition of ß-amyloid fibrils within the brain and the activation
of microglial cells associated with the amyloid plaque. The activated
microglia subsequently secrete a diverse range of inflammatory products
[79
]. Kitamura et al. [80
] assessed the
occurrence of COX-1, COX-2, and PPAR in normal and AD brains using
specific antibodies and found increased expression of these moieties in
AD brains. Nonsteroidal, anti-inflammatory drugs (NSAIDs) have been
shown to be efficacious in reducing the incidence and risk of AD and in
delaying disease progression [81
]. Combs et al.
[79
] demonstrated that NSAIDs, thiazolidinediones, and
PGJ2, all of which are PPAR agonists, inhibited the
ß-amyloid-stimulated secretion of inflammatory products by microglia
and monocytes. PPAR agonists were shown to inhibit the
ß-amyloid-stimulated expression of the genes for IL-6 and TNF- and
the expression of COX-2 [79
]. Finally, Heneka et al.
[82
] demonstrated that microinjection of LPS and IFN-
into rat cerebellum induced iNOS expression in cerebellar granule cells
and subsequent cell death. Coinjection of PPAR agonists (including
troglitazone and 15d-PGJ2) reduced iNOS expression and cell
death, whereas coinjection of a selective COX inhibitor had no effect.
Overall, work in AD seems to suggest that PPAR agonists can modulate
inflammatory responses in the brain and that NSAIDs may be helpful in
AD as a result of their effect on PPAR.
Several studies have investigated the role of PPAR ligands in
modifying animal models of autoimmune diseases. Su et al.
[83
] showed that in a mouse model of inflammatory bowel
disease, thiazolidinediones markedly reduced colonic inflammation.
These authors proposed that this effect might be a result of a
direct effect on colonic epithelial cells, which express high levels of
PPAR and can produce inflammatory cytokines. Kawahito et al.
[78
] demonstrated that intraperitoneal administration of
the PPAR ligands, 15d-PGJ2 and troglitazone, ameliorated
adjuvant-induced arthritis with suppression of pannus formation and
mononuclear cell infiltration in rats. Nino et al. [84
]
examined the effect of a thiazolidinedione on experimental allergic
encephalomyelitis and found that this treatment attenuated the
inflammation and decreased the clinical symptoms in this mouse model of
multiple sclerosis. Finally, Reilly et al. [85
]
demonstrated that renal glomerular mesangial cells are important
modulators of the inflammatory response in lupus nephritis, secreting,
when activated, inflammatory mediators including NO and cyclooxygenase
products, thus perpetuating the local inflammatory response. Mesangial
cells isolated from the lupus-prone MRL/lpr mice or control Balb/c mice
were stimulated, and it was found that the MRL/lpr mesangial cell
cultures did not increase PGJ2 production as did the cells
from Balb/c mice. This suggested an abnormality in MRL/lpr mice in a
normally present endogenous, anti-inflammatory pathway mediated by
PGJ2, perhaps working through PPAR. NO production from
the mesangial cells of both mouse strains was found to be blocked by
PGJ2 and a thiazolidinedione. Given the above studies, the
relevance of PPARs and the utility of treatment with PPAR agonists in
diseases with an inflammatory or autoimmune pathogenesis will likely
continue to remain a research focus.
|
THE ROLE OF PPAR IN T CELLS
|
|---|
It is noteworthy that even in the studies of autoimmune disease
models, the question of PPAR expression and function in T cells had
not been raised. Recently, my laboratory was the first to describe the
expression and function of PPAR in T lymphocytes
[86
]. We demonstrated for the first time that murine T
cells express PPAR1, but not PPAR2. Using reverse
transcriptase-polymerase chain reaction (RT-PCR) and
immunohistochemistry, SJL-derived Th1 clones [clones MM4 and B48,
specific for bovine myelin basic protein (BMBP)] were found to express
significant levels of PPAR1. Using RT-PCR, we have also found that
freshly isolated T-cell-enriched splenocytes from SJL mice express
PPAR1 mRNA but not PPAR2. The PPAR expressed by the T-cell
clones and by freshly isolated C57BL/6 T-cell-enriched splenocytes was
shown to be of functional significance. This functional significance
was demonstrated through the use of two PPAR ligands,
15d-PGJ2 and a thiazolidinedione, ciglitazone.
My laboratory demonstrated that 15d-PGJ2 and ciglitazone
mediate significant inhibition of the antigen (BMBP)-stimulated
responses of the T-cell clones and mediate significant inhibition of
the anti-CD3 antibody-stimulated proliferative responses of the T-cell
clones and the freshly isolated T-cell-enriched splenocytes
(Tables 4 5
6
7
). This inhibition was directed at the level of the T cell, given that
T-cell stimulation with immobilized anti-CD3 antibody is a
macrophage/antigen-presenting cell (APC)-independent response, and for
the T-cell clones, this inhibition was observed in the absence of
monocytes/macrophages. The inhibition of the responses of the
T-cell-enriched splenocytes (Table 6)
suggests that PPAR is
functionally relevant in freshly isolated T cells or becomes
functionally relevant early in activation.
In these studies, it was also demonstrated that the two ligands for
PPAR mediated inhibition of IL-2 secretion by the T-cell clones and
did not inhibit IL-2-induced proliferation of such clones (Tables 4
5
and 7)
. This lack of inhibition of the IL-2-induced proliferation is
noteworthy for two reasons. First, it demonstrates that the inhibitory
effects noted were not a result of toxic effects of
15d-PGJ2 or ciglitazone at the concentrations studied.
Second, it suggests that PPAR-ligand interaction may affect
signaling pathways involved in the response to T-cell receptor
stimulation but not pathways involved after IL-2-receptor ligation.
Soon after this study, Yang et al. [87
] showed that
PPAR was also expressed in human peripheral blood T cells. In
agreement with my findings in murine T cells, these investigators found
that a thiazolidinedione, troglitazone (20–40 µM), and
15d-PGJ2 (1–10 µM) but not a PPAR ligand inhibited
phytohemagglutinin (PHA)-induced proliferation, IL-2 production, and
IL-2 mRNA expression in human peripheral blood T cells in a
dose-dependent manner. Then, they transfected PPAR2 cDNA into Jurkat
cells and found that transfected but not wild-type Jurkat cells (which
express little detectable PPAR mRNA) were inhibited in IL-2
secretion by PPAR ligands. This effect was shown to be at least
partially a result of PPAR effects on IL-2 promoter activity. A
PPAR ligand did not inhibit IL-2 secretion, suggesting that PPAR
could not mediate this inhibition in Jurkat cells. Given the recent
studies demonstrating that 15d-PGJ2 and the
thiazolidinediones appear to mediate macrophage anti-inflammatory
effects that are PPAR-independent (see below), these T-cell
transfection studies are important. They may indicate that in T cells,
thiazolidinediones mediate their effects only through PPAR-dependent
pathways. Finally, Yang et al. [87
] demonstrated that
the activated PPAR physically associates with the transcription
factor nuclear factor of activated T cells (NFAT), thus blocking its
DNA-binding and transcriptional activation of the IL-2 promoter. The
activation and function of NFAT are known to be absolute requirements
for IL-2 transcription [87
].
More recently, Harris and Phipps [88
] confirmed the
expression of PPAR in murine T cells. They demonstrated that naive
and PMA-activated, ovalbumin-specific T cells from T-cell-receptor
transgenic mice expressed PPAR1 mRNA and protein. The investigators
also found that their T cells did not express mRNA for PPAR2.
Immunohistochemical analysis revealed that PPAR staining in naive
cells was predominantly cytoplasmic with some perinuclear staining. In
activated T cells, staining also included a nuclear component and was
more intense overall than in naive T cells. When T cells were
stimulated (with PMA and ionomycin or antigen and APCs) in the presence
of 15d-PGJ2 or troglitazone, these investigators also noted
an inhibition of proliferation. However, they found that this
inhibition of proliferation was accompanied by significant decreases in
cell viability. The latter was demonstrated to be a result of apoptosis
of the T cells and occurred only when cells were treated with PPAR
but not PPAR agonists. 15d-PGJ2 mediated these effects
in a dose range of 3–100 µM. The thiazolidinediones, ciglitazone and
troglitazone, were effective in dose ranges of 25–100 µM. Finally,
Harris and Phipps [88
] also suggest that PPAR may
play a role in T-cell development, given that thymic-stromal cells
express COX-1 and COX-2 enzymes, that inhibition of these enzymes
interferes with positive selection [89
,
90
], and that PGD synthase is produced by
thymic-dendritic cells, suggesting that PGD2 and the
J-series PGs may be present [91
].
The demonstration of the expression and function of PPAR in human
and murine T cells now greatly expands the possible immunoregulatory
role for PPAR. Taken together with the effects of PPAR ligands on
macrophage function, the T-cell findings suggest that PPAR may play
a significant role in the regulation of the innate and adaptive immune
response. (However, see the discussion below concerning the PPAR
specificity of the macrophage effects noted.) Finally, although the
T-cell studies agree on the functional role of PPAR activation in
inhibiting activated T-cell proliferation, the mechanism(s) of this
inhibition is not yet totally clear. Our studies [86
]
and those of Yang et al. [87
] suggest an inhibition of
IL-2 secretion, and the studies of Harris and Phipps
[88
] suggest apoptosis as the mode of PPAR-mediated
T-cell regulation. However, our studies using IL-2-dependent murine
T-cell clones may also involve cell death as a mode of regulation.
Preliminary studies in my laboratory indicate that the addition of IL-2
at different times relative to PPAR ligation may lead to different
outcomes. These preliminary studies indicate that if the PPAR
ligands are removed, and exogenous IL-2 is added two days after rather
than simultaneously with the addition of the higher concentrations of
15d-PGJ2 (5 µM) or ciglitazone (40 µM), our T-cell
clones cannot be recovered and undergo cell death. Future studies in my
laboratory will examine further the relationship between apoptosis and
PPAR-mediated T-cell effects.
Many other issues now await study regarding the relevance of PPAR to
T-cell biology. These issues include the mechanism(s) by which PPAR
ligands mediate the T-cell effects, as well as the possible
differential expression and function of PPAR in the many T-cell
types and subtypes. Given the varied patterns of signaling pathways and
cytokine secretion found in different subsets of T cells, I believe it
is likely that PPAR ligation will have pleiomorphic effects in
different T-cell subsets. In vivo, it may be that PPAR ligation
plays an immunoregulatory role early in the initiation of T-cell immune
responses as well as in inhibiting the recruitment of naive T cells
into an ongoing immune response. It is possible that as additional
endogenous ligands become known, they will be found to be involved in
normal T-cell immunoregulatory processes. Finally, as discussed below,
it will be crucial to document the true specificity of any ligands used
in PPAR studies in T cells, and this will likely require
T-cell-specific knockout technology.
|
SPECIFICITY OF 15d-PGJ2 for PPAR
|
|---|
As reviewed by Spiegelman [92
], one problem with
many of the experiments described above is that they used
15d-PGJ2 as a PPAR-specific ligand. Many studies have
suggested that 15d-PGJ2 may not be specific for PPAR
[93
94
95
96
]. Another difficulty with many of these studies
is that when more selective ligands (e.g., thiazolidinediones) were
used, high concentrations—far exceeding those required to bind the
receptor—were necessary to achieve the results described
[97
].
Recently, the issue of the specificity of 15d-PGJ2 for
PPAR has been at least partially clarified. NF-B is a critical
activator of genes involved in inflammation and immunity
[98
]. In this activation, the IB kinase complex (IKK)
phosphorylates the NF-B inhibitors (IB proteins) leading to their
conjugation with ubiquitin and subsequent degradation. This then allows
freed NF-B dimers to translocate to the nucleus and induce target
genes [98
]. Rossi et al. [98
]
demonstrated that the cyclopentenone PGs, including
15d-PGJ2, directly inhibit and modify the IKK2 subunit of
IKK. This, in turn, prevents the phosphorylation of the inhibitory
IB proteins that then target these proteins for ubiquitin
conjugation and degradation [99
]. This then prevents the
activation of NF-B. Similarly, Castrillo et al. [99
]
showed that in RAW 264.7 macrophage cells treated with LPS and IFN-,
incubation with 15d-PGJ2 resulted in a significant
inhibition of IKK2 activity and an inhibition of the degradation of the
inhibitory IB proteins. This, in turn, caused a partial inhibition
of NF-B activity and an impaired expression of genes requiring
NF-B activation—such as type-2 NOS and cyclooxygenase 2
[99
]. Finally Straus et al. [100
] also
confirmed these findings. These investigators demonstrated that
15d-PGJ2 inhibits NF-B-dependent transcription by two
PPAR-independent pathways: covalent modifications of critical
cysteine residues in IKK and the DNA-binding domains of NF-B
subunits. These studies have now clearly demonstrated that
15d-PGJ2 cannot be used to document effects that are
mediated solely through PPAR.
|
STUDIES USING PPAR NULL MACROPHAGES
|
|---|
Attempts to generate a PPAR-deficient mouse have resulted in
embryonic lethality. Recently, to overcome this, investigators have
used two approaches to study PPAR null macrophages: 1) using
homologous recombination to create embryonic stem cells that are
homozygous for a null mutation in the PPAR gene, together with in
vitro differentiation of embryonic stem cells into macrophages and 2)
assessment (via PCR) of mature PPAR null-macrophage representation
in chimeric mice that were generated by the injection of
PPAR-deficient embryonic stem cells into wild-type blastocysts
[97
, 101
]. These studies have allowed for
new insights into the role played by PPAR in macrophage
differentiation and function (including in inflammatory responses) and
for new insights into the specificity of PPAR ligands in
macrophages. Studying PPAR null macropages, Moore et al.
[101
] have demonstrated that PPAR is neither
essential for macrophage differentiation nor for such mature macrophage
functions as phagocytosis and inflammatory cytokine production. PPAR
was found to be required for basal expression of CD36 but not for
expression of the other major scavenger receptor, SR-A, responsible for
uptake of modified lipoproteins. The absence of PPAR was found to
significantly reduce the cellular uptake and degradation of OxLDL.
These authors also confirmed the requirement for PPAR in the
IL-4-induced increase in macrophage CD36 expression. However,
PPAR-deficient macrophages showed no difference from wild-type
macrophages in expression of CD14 or other macrophage-specific surface
markers and produced similar levels of TNF- and IL-6 after LPS
stimulation. This suggests a lack of PPAR ligand involvement in the
normal regulation of macrophage-cytokine secretion. PPAR null
macrophage-phagocytic activity was also similar to that seen in the
wild-type macrophages. In addition, the effect of PPAR ligands on
PMA stimulation showed that in wild-type macrophages and the cell line
RAW264.7, troglitazone increased IL-6 production, but this increase was
not observed in PPAR-deficient macrophages. Furthermore,
15d-PGJ2 inhibited PMA-induced IL-6 production in wild-type
macrophages as well as in PPAR-deficient macrophages, indicating
that this inhibitory response is not dependent on PPAR.
Chawla et al. [97
], using PPAR null macrophages,
demonstrated that PPAR is neither essential for nor substantially
affects the development of the macrophage lineage in vitro and in vivo.
In contrast, they showed that it is an important regulator of the
scavenger receptor CD36. Most important, these investigators
demonstrated that 15d-PGJ2 and thiazolidinediones have
anti-inflammatory effects that are independent of PPAR. They showed
that PPAR is required for positive effects of its ligands in
modulating macrophage-lipid metabolism, but the inhibitory effects on
cytokine production and inflammation may be PPAR-independent. When
macrophages were stimulated with LPS in the context of PPAR ligands,
the LPS-induced level of TNF- and IL-6 did not differ between
wild-type and PPAR-deficient macrophages, and 15d-PGJ2
and thiazolidinediones inhibited the secretion of both cytokines
equally in wild-type and deficient macrophages. To further confirm
this, they stimulated macrophages with IFN- and evaluated gene
expression for iNOS and COX-2. The increase in mRNAs was equal between
wild-type and PPAR-deficient macrophages. Finally, the expression of
these proinflammatory genes was inhibited by both ligands in wild-type
and PPAR-deficient macrophages.
Although there are plausible alternative interpretations for the
results of these studies [102
], and although it is
possible that the PPAR-independent effects of the thiazolidinediones
may not be seen in cell types other than macrophages, caution will now
be needed in interpreting all results in which 15d-PGJ2 or
the thiazolidinediones are used as putative PPAR-specific ligands.
|
OVERALL CONCLUSIONS
|
|---|
It is 
TOP
ABSTRACT
INTRODUCTION
