PEROXISOME-PROLIFERATOR- ACTIVATED RECEPTORS AND ...

REVIEWS

PEROXISOME-PROLIFERATORACTIVATED RECEPTORS AND CANCERS: COMPLEX STORIES Liliane Michalik, Béatrice Desvergne and Walter Wahli Peroxisome-proliferator-activated receptors (PPARs) are nuclear hormone receptors that mediate the effects of fatty acids and their derivatives at the transcriptional level. Through these pathways, PPARs can regulate cell proliferation, differentiation and survival, so controlling carcinogenesis in various tissues. But what are the links between each PPAR isotype and carcinogenesis and what is the relevance of these findings to human pathology and therapy? PEROXISOMES

Organelles that are found in all organisms and in various cell types. They are involved in many different functions, among which β-oxidation of long-chain fatty acids and H2O2-based respiration are the most prominent. LIPOXYGENASES

5-, 12-, and 15-lipoxygenases are enzymes that catalyse a key step in the production of leukotrienes in platelets, macrophages, mastocytoma cells and leukocytes.

Center for Integrative Genomics, NCCR Frontiers in Genetics, University of Lausanne, CH-1015 Lausanne, Switzerland. Correspondence to W.W. e-mail: [email protected] doi:10.1038/nrc1254

NATURE REVIEWS | C ANCER

Peroxisome-proliferator-activated receptors (PPARs) α, β/δ and γ (also known as NR1C1, NR1C2 and NR1C3 (REF. 1)) are ligand-inducible transcription factors that belong to the nuclear-hormone-receptor (NHR) family. They were identified in the early 1990s in rodents, Xenopus and humans and were initially described as the nuclear receptors for compounds that induce PEROXISOME proliferation in rodents — which explains their name. They have since been shown to be important sensors of cellular levels of fatty acids and fatty-acid derivatives that are mainly derived from the LIPOXYGENASE and CYCLOOXYGENASE (COX) pathways. Polyunsaturated fatty acids activate the three PPAR isotypes with relatively low affinity2, whereas fatty-acid derivatives show more binding selectivity (FIG. 1). Because of their hydrophobic nature, PPAR ligands are most probably delivered to the nucleus and to their receptor by cytoplasmic fatty-acid-binding proteins (FABPs), through a mechanism that remains to be elucidated3. PPARs function as heterodimers with their obligate partner — the retinoid receptor (RXR; FIG. 1). Like other NHRs, the PPAR–RXR heterodimer probably recruits cofactor complexes — either co-activators or co-repressors — that modulate its transcriptional activity4–7 (FIG. 1). An interaction between PPARα and nuclear-receptor corepressor has been described, but the functional relevance of this interaction is unknown8. So far, only PPARβ/δ has been implicated in transcriptional-repression functions and has been shown to repress the activity of PPARα or PPARγ target genes6,7.

Following activation by their ligands and heterodimerization with RXR, PPARs interact with the peroxisome-proliferator response element (PPRE) — usually 5′-AACT AGGNCA A AGGTCA-3′ — in the promoter of their target genes (FIG. 1). PPREs are found in various genes that are involved in lipid metabolism and energy homeostasis, including lipid storage or catabolism (βoxidation and ω-oxidation), fatty-acid transport, uptake and intracellular binding9. Recently, two new target genes were identified for PPARβ/δ, which code for integrin-linked kinase (ILK) and 3-phosphoinositide-dependent kinase-1 (PDK1) — both of which are involved in a pathway that is pivotal for cell survival, differentiation and proliferation10. Binding of PPAR– RXR to the PPRE that is present in the promoter of their target genes allows for adaptation of the physiological response of the organism to changes in lipid levels, by regulating transcription9. Each Ppar isotype shows a specific pattern of expression in rodents, which is similar overall during fetal development and in adults. Pparα is expressed in the liver, kidney, intestine, heart, skeletal muscle, brown adipose tissue, adrenal gland and pancreas. Pparβ/δ is expressed broadly and has been detected in all of the tissues tested, albeit with varied expression levels. The expression of Pparγ is much more restricted: it is expressed at high levels in the brown and white adipose tissues and at lower levels in the intestinal epithelium, the retina, skeletal muscle and lymphoid organs. Interestingly, the expression of the three isotypes peaks

VOLUME 4 | JANUARY 2004 | 6 1

REVIEWS

CYCLOOXYGENASES

(COXs). These catalyse a key step in the conversion of 20-carbon polyunsaturated fatty acids — such as arachidonic acid — to prostaglandins. COX1 is constitutively expressed in most tissues, but COX2 is induced by pathophysiological conditions such as tumorigenesis or inflammatory situations. COX2 is now recognized as an interesting target for the treatment of cancers. LIPOPROTEIN

Association of proteins with triglycerides, phospholipids and cholesterol. Lipoproteins are produced in the liver and serve as lipid carriers in the blood.

in the central nervous system at mid-gestation, but only the expression of the Pparβ/δ isotype remains high in that organ at postnatal stages11. Consistent with its distribution in tissues with high catabolic rates of fatty acids and peroxisomal metabolism, the main role of PPARα is the regulation of energy homeostasis. PPARα activates fatty-acid catabolism, stimulates gluconeogenesis and ketone-body synthesis and is involved in the control of LIPOPROTEIN assembly. PPARα also attenuates inflammatory responses and participates in the control of amino-acid metabolism and urea synthesis. PPARγ has a pivotal role in adipocyte differentiation, lipid storage in the white adipose tissue and energy dissipation in the brown adipose tissue. In addition, PPARγ is involved in the control of inflammatory reactions and in glucose metabolism through improvement of insulin sensitivity. PPARβ/δ functions were only recently identified, so remain less well characterized. It is required for placental development and is involved in the control of lipid metabolism, even though the underlying mechanisms still need investigation. The best-characterized function of PPARβ/δ is its important role in the control of cell proliferation, differentiation and survival, especially in 9,12 KERATINOCYTES . PPARα and PPARγ are linked to metabolic disorders and are interesting pharmaceutical targets (BOX 1). Among the synthetic ligands that activate these receptors, the fibrates — used for decades in the treatment of DYSLIPIDAEMIA — are hypolipidaemic compounds that activate PPARα. The thiazolidinediones (TZDs) — which selectively activate PPARγ — are hypoglycaemic molecules that are used to treat type II diabetes9. Growing evidence implicates each of the three isotypes in the development of tumours in various tissues, either being detrimental or beneficial. So, what mechanisms do PPARs use to regulate tumorigenesis? α PPARα

Peroxisome proliferators (PPs) are chemically nonrelated molecules that include naturally occurring steroids and lipids, as well as xenobiotics such as fibrates

(hypolipidaemic drugs that are used in the treatment of dyslipidaemias), industrial plasticizers, pesticides and solvents. Their main target is the liver, in which chronic administration induces a short-term pleiotropic response, especially in rodents13. This response includes liver hypertrophy and hyperplasia, peroxisome proliferation, transcriptional activation and increased activity of enzymes that are involved in fatty-acid metabolism (fatty-acid β-oxidation, fatty-acid transporters, cytoplasmic liver FABP (L-FABP)) and of the cytochrome p450 family9. This short-term response is followed by hepatocellular carcinomas in rats and mice14–16 (FIG. 2). Even though this hypothesis has not been verified, one cannot exclude the idea that some PPs might not be carcinogenic themselves, but could be metabolized into compounds that would in turn mediate the carcinogenic effects. Pparα was the first PPAR identified17 and was shown to mediate PP actions. Pparα-null mice are resistant to the acute effects that are induced by treatment with the PP clofibrate and Wy-14,643 (REF. 18), as well as to the development of hepatocellular carcinomas19, which confirms that Pparα acts exclusively to mediate these effects. The mechanisms that underlie the role of Pparα in these processes are still unclear, but several non-exclusive hypotheses have been proposed. PPs induce DNA replication and proliferation in hepatocytes20,21 in a Pparα-dependent manner19,22; however, no evidence for a direct transcriptional effect of Pparα on the expression of cell-cycle genes has been reported so far. PPs also repress spontaneous and induced hepatocyte apoptosis, in vitro and in vivo 23–27. The involvement of Pparα in the repression of apoptosis by PPs is indicated by the use of a Pparα dominantnegative protein, which abrogates this effect when transfected in rat primary hepatocytes28. Evidence indicates that the production of reactive oxygen species as by-products of β-oxidation reactions — which might damage DNA and promote hepatocyte proliferation — is also involved in Pparα-agonistmediated hepatocarcinogenesis29–32. However, the implication of Pparα in this PP-mediated effect

KERATINOCYTE

The main cell type of the epidermis, the uppermost layer of the skin. Basal keratinocytes are responsible for the renewal of the epidermis. The daughter keratinocytes then undergo a vectorial specific differentiation programme, while migrating to the top of the epidermis, where they finally die and desquamate. DYSLIPIDAEMIAS

The term lipidaemia refers to the circulating level of total lipids, including free fatty acids and lipoproteins. Dyslipidaemias are pathological disorders that affect normal lipidaemia, usually corresponding to high blood lipid levels that are a high risk factor for cardiovascular diseases.

62

Summary • Peroxisome-proliferator-activated receptors (PPARs) are ligand-activated transcription factors that belong to the nuclear-hormone-receptor family. Three isotypes have been identified — PPARα, PPARβ/δ and PPARγ. • PPARs are activated by endogenous ligands — fatty acids and fatty-acid derivatives, for which they act as intracellular sensors — which leads to transcriptional regulation of pathways that are involved in lipid and glucose metabolism. • PPARα is the therapeutic target of the fibrates, which are widely used as hypolipidaemic compounds in the treatment of dyslipidaemia. PPARγ is a non-exclusive target of the thiazolidinediones, a new class of compounds with hypoglycaemic properties that are used to treat type II diabetes. Because of its involvement in lipid metabolism and skin homeostasis, it is likely that PPARβ/δ will become a therapeutic target in the near future. • Each PPAR isotype is associated with pathways that relate to carcinogenesis. Long-term activation of PPARα by peroxisome proliferators induces the development of hepatocarcinomas in rodent liver, but not in humans. PPARγ is thought to have overall anti-carcinogenic effects in many different cell types, due to its anti-proliferation, prodifferentiation and pro-apoptotic properties. PPARβ/δ is involved in the control of cell proliferation, cell differentiation and apoptosis, but its involvement in the development of tumours is unclear at present. • In the future, PPARγ, and perhaps PPARβ/δ, might become interesting therapeutic targets for the treatment of tumours. However, further investigation is needed at both the scientific and clinical levels.

| JANUARY 2004 | VOLUME 4

www.nature.com/reviews/cancer

REVIEWS a

Lipoxygenase LTB4

8S-HETE

15-HETE

Arachidonic acid 15-dPGJ2

PGI2

Linoleic acid

9-HODE

13-HODE

oxLDL

9-HODE

13-HODE

Cyclooxygenase 12-,15-Lipoxygenase

b

CoRep? PPAR RXR L CoRep

CoAct

CoRep

CoAct

L

PPAR RXR

PPAR RXR

PPRE

PPRE

Figure 1 | Schematic representation of the PPAR signalling pathways. a | Endogenous agonists of peroxisome-proliferator-activated receptors (PPARs). PPARs are ligand-inducible receptors, which can be activated by fatty acids — such as arachidonic or linoleic acids — and their derivatives. The fatty-acid metabolites that activate PPARs are mainly derived from arachidonic or linoleic acids through the cyclooxygenase or the lipoxygenase pathways. The best characterized at the moment are leukotriene B4 (LTB4) and 8S-HETE (hydroxyeicosatetraenoic acid), which preferentially activate PPARα; 15-deoxy-prostaglandin J2 (15-dPGJ2) and 15-HETE, which are PPARγ-selective ligands; and the prostaglandin I2 (PGI2, also called prostacyclin), which is probably a PPARβ/δ natural ligand. PPARγ is also activated by 9-HODE (hydroxyoctadecadienoic acid) and 13-HODE, either derived from linoleic acid or as components of oxidized lowdensity lipoprotein (oxLDL). b | PPARs function as heterodimers with their obligate partner, retinoid receptor (RXR). The dimer probably interacts with co-regulators, either co-activators (CoAct) or co-repressors (CoRep). In the unliganded form, PPARβ/δ–RXR heterodimer, in contrast to PPARα–RXR and PPARγ–RXR heterodimers, recruits corepressors and represses the activity of PPARα and PPARγ target genes by binding to the peroxisome proliferator response element (PPRE) that is present in their promoters6,7. In their liganded form, the PPAR–RXR heterodimers interact with co-activators, bind to the PPRE that is present in the promoters of their target genes and activate their transcription.

COLITIS

Inflammatory state of the colon, possibly due to a single cause such as bacterial infection.


remains to be proven. So, growing evidence clearly shows that the activation of Pparα is required and causal to the adverse effects of PPs in rodent liver. The underlying mechanisms probably include perturbation of the cell cycle and the production of reactive oxygen species in the hepatocytes15,16 (FIG. 2). As many humans are almost constantly exposed to PPs — either through hypolipidaemic treatments or through contact with industrial plasticizers and pesticides — assessing the risk to humans from these compounds is of great importance. Species other than rats and mice,

including humans, do not display the full range of responses to PPs15,33,34. Epidemiological studies did not show significant peroxisome proliferation in the liver of patients who were treated with hypolipidaemic drugs15,16,35 and cell-culture studies indicate that human cells display a reduced transcriptional response to PPARα activation, compared with rats36. Several explanations for these species differences to PP exposure have been proposed. Human hepatocytes express only 1–10% of the amount of Pparα that has been quantified in rodent hepatocytes37,38. Moreover, the PPARα protein is less efficiently activated by some of the PPs39. Species specificity in the expression of PPAR cofactors4,5 or in the efficiency of PPREs found in PPARα human target genes40,41 could partially explain differences that are observed at the level of transcriptional activation by PPs. In addition to these molecular explanations, the bioavailability of some PPs seems to be reduced in humans42–44. Reports published so far convincingly show that humans are resistant to the adverse effects of the known molecules, but retain their beneficial effects. However, as the liver is a main site of PPARα action, it is important to maintain a constant vigilance when developing new therapeutic or industrial compounds. PPARγγ

The involvement of PPARγ in the development of tumours in various tissues is still debated. In many cell types, PPARγ activation has anti-tumorigenic effects, probably because of its anti-proliferative and prodifferentiation activities; on the contrary, in some other experimental settings or in vivo situations, its activation seems to have deleterious pro-carcinogenic consequences. Because of the numerous different models that are being investigated, the field remains complex. An association between the activity of Pparγ and tumour development has been best studied in rodent models of colon cancer, in which the expression of Pparγ is increased. In a xenograft model, the Pparγ agonist troglitazone inhibit the development of tumours that are derived from colon cancer cells45. Consistent with this, troglitazone fed to rodents reduces COLITIS, a risk factor for further colorectal cancer development in humans, and formation of aberrant crypt foci, an early step in the development of colon carcinoma46,47. However, in adenomatous polyposis coli (Apc)Min mice — a recognized model for human familial adenomatous polyposis — treatment with TZDs increases the number of colon tumors48,49. It has also been proposed that the increased incidence of colon cancer in individuals with a high-fat diet could be due to PPARγ activation by fatty acids50. A possible explanation for these contradictory results is that the consequence of the activation of PPARγ on the development of colon cancers depends on the presence of the APC mutation in the colon tissue51. Together, these reports indicate that an early treatment with the PPARγ agonists — before the first step of carcinogenesis occurs — might prevent tumour formation. Activation of PPARγ after tumour initiation — as in the Apc Min mice — might be inefficient or deleterious52.


REVIEWS

Box 1 | PPARs in health and diseases Cancer is certainly not the only pathological situation to which peroxisomeproliferator-activated receptors (PPARs) are linked106. Activating PPARα is useful in the treatment of dyslipidaemias and cardiovascular diseases. Fibrate treatments decrease the plasma levels of triglycerides and increase high-density lipoprotein (HDL) levels. In atherosclerosis — which is characterized by an inflammation of the arterial wall, proliferation of the vascular smooth-muscle cells, differentiation of macrophages into foam cells and, finally, rupture of the plaque — ligand-dependent activation of PPARα or PPARγ has beneficial effects. PPARα and PPARγ also have overall anti-inflammatory actions, which might be useful in the future in the treatment of inflammatory diseases. One of the best-known functions of PPARγ as a therapeutic target is in the treatment of type II diabetes, as its ligands, the thiazolidinediones, are used as insulin sensitizers. Human genetics has linked several mutations or polymorphisms in the PPARγ gene to some chronic metabolic disorders. The most frequent is a Pro12Ala substitution, which leads to a less active PPARγ that has been linked to a lower body-mass index and a higher insulin sensitivity107. A Pro115Gln mutation, resulting in increased PPARγ activity was identified in four very obese patients108. Finally, two different mutations, leading to the expression of dominant-negative forms of PPARγ, were identified in three individuals who were suffering from severe insulin resistance109.

Peroxisome proliferators

a Pparα activation

Rodent liver

d Mechanism Kupffer cells

?

Increased DNA replication Increased proliferation Decreased apoptosis Reactive oxygen species (DNA damage; proliferation)

b Short-term response

Studies of breast carcinogenesis are more homogenous. In breast cancer cells, TZDs induce phenotypic changes that correlate with a less malignant state53,54 and, in a murine organ culture, troglitazone efficiently prevents mammary carcinogenesis55. Consistent with this, troglitazone reduces the development of tumours in nude mice that are derived from an MCF-7 (breast cancer) cell line54 and GW 7845 — another PPARγ selective agonist — prevents chemically induced breast carcinogenesis in rats56. However, it is not yet clear whether all of these antitumour effects are dependent on PPARγ 57. PPARγ agonists also inhibit proliferation of prostate cancer cell lines and decrease the expression of prostate-specific antigen (PSA), a widely used marker, the serum level of which reflects the volume of the prostate tumour58,59. Interestingly, this last effect is maintained in vivo, as troglitazone treatment stabilizes PSA in patients with advanced prostate cancers59, in an androgen-dependent manner60. Consistent with a link between the activation of PPARγ and anti-carcinogenic consequences in many different models, various mutations that decrease PPARγ activity were found in the human PPARγ gene in colon tumours, and thyroid follicular carcinomas, whereas hemizygous deletions of the PPARγ gene were found in prostate cancers (TABLE 1). All of these mutations lead to a decrease in PPARγ activity, which indicates that PPARγ has anti-carcinogenic functions. A related finding is that the expression of PPARγ seems to be frequently decreased in thyroid carcinomas61. Although the implication of PPARγ mutations seems to be significant in some cases, they are relatively rare, overall, in the development of human tumours. Additionally, total loss of both PPARγ alleles has never been described in any tumour so far, an observation which indicates that PPARγ might not be a true tumour suppressor.

64


Transcriptional activation of genes that are involved in fatty-acid metabolism, in the cell cycle and in degradation of endogenous and exogenous compounds (cytochrome p450 family) Peroxisome proliferation Cell proliferation Liver hypertrophy

c Long-term response Hepatocellular carcinoma

α activation by PP in Figure 2 | Consequences of Pparα the liver and proposed underlying mechanisms. Longterm chronic activation of peroxisome-proliferator-activated receptor-α (Pparα) in the hepatocytes by its ligands (initial event; a) induces a short-term pleiotropic response (b) followed by hepatocellular carcinomas in both rats and mice (c). The short-term response includes transcriptional activation of enzymes that are involved in fatty-acid metabolism (fatty-acid β-oxidation, fatty-acid transporters and cytoplasmic liver fatty-acid-binding protein (L-FABP)), of genes that are involved in cell-cycle control and of genes coding for enzymes of the cytochrome p450 family (secondline events)14; peroxisome and cell proliferation (third-line events); and liver hypertrophy and hyperplasia (fourth-line events). The long-term consequence of these events is the development of hepatocellular carcinomas in rodents. d | Several underlying mechanisms are being debated15,16.Peroxisome proliferators (PPs) induce DNA replication and proliferation of hepatocytes in a Pparαdependent manner19,22. Furthermore, PPs repress spontaneous and induced hepatocyte apoptosis, in vitro and in vivo. As well as controlling of the cell cycle, the production of reactive oxygen species in response to Pparα agonists might damage DNA and promote hepatocyte proliferation, but the implication of Pparα in this effect remains to be proven. Additionally, non-hepatocyte cells, such as Kupffer cells, might participate in the short-term cascade of events by promoting hepatocyte proliferation31.


REVIEWS Table 1 | Mutations found in PPARγγ in human tumours Tissue

Frequency

Mutation

Sporadic colon tumors

4/55

Loss of function

112

Follicular thyroid carcinomas

5/8

Chromosomal translocation; PAX8–PPARγ fusion protein; dominant-negative inhibitor of PPARγ

113

Follicular thyroid carcinomas

5/9

Chromosomal translocation; PAX8–PPARγ fusion protein

114

Follicular thyroid adenomas

2/16

Dominant-negative inhibitor of PPARγ

Follicular thyroid carcinomas Follicular thyroid adenomas

13/33 1/23


115

Follicular thyroid carcinomas Follicular thyroid adenomas

6/17 6/11


116

Prostate tumours

8/38

Hemizygous deletion of PPARγ

71 cancer cell lines; 0 326 clinical cancers (colon, prostate, breast,lung cancers, osteosarcomas, leukaemias)

No mutation found

References

59 117

Mutations that decreased the activity of the receptor were described in the PPARγ gene in some human tumours of various origins. Overall, these mutations are rather rare, indicating that decreased activity of PPARγ might contribute to carcinogenesis, but is probably not causal to the pathology. For the sake of clarity (different approaches), each publication cited in the table is listed separately.

Several molecular mechanisms that could explain the anti-carcinogenic functions of PPARγ have been described. The data obtained in various cancer cells that express PPARγ indicate that activation of this receptor by an agonist either induces cell-cycle arrest — by inhibiting cell proliferation and stimulating cell differentiation — or promotes cell death (TABLE 2). Accordingly, the expression of several genes is modulated in parallel with PPARγ-agonist treatments. Even though the data remain controversial, PPARγ seems to regulate expression of the β-catenin gene49,51. However, whether PPARγ represses or activates the expression of this gene might depend on the existence of a functional APC–β-catenin pathway in the cell51. Most of the other putative targets of PPARγ that have been reported are implicated in the control of the cell cycle. The activation of PPARγ in a model of transformed adipogenic cells promotes cell-cycle withdrawal by inhibiting the DNA-binding activity of the E2F/DP transcription factor, which is required for cell proliferation62. In pre-adipocytes and pituitary adenoma cells, Pparγ activation correlates with a decrease in the phosphorylation state of the retinoblastoma susceptibility gene product Rb — its hyperphosphorylation is required for cell-cycle progression63,64. The expression of the cell-cycle inhibitors WAF1 (also known as CIP1 or p21) and INK4C (also known as p18) increases in parallel with adipogenesis in a PPARγ-liganddependent manner65, whereas the PPARγ ligand 15deoxy-PGJ2 increases the level of WAF1 in a breast cancer epithelial cell line66. Pparγ activation in Ras-transformed rat intestinal epithelial cells, pancreatic or breast cancer cell lines results in the inhibition of cell cycle and S-phase entry, most probably through a decrease in cyclin D1 expression or increased degradation of cyclin D1 (REFS 67–70). In a human colon cancer cell line, PPARγ activators


negatively regulate regenerating gene IA, which was initially identified in pancreatic cells and is involved in cell proliferation, and induce keratin 20, which is involved in colon-cell maturation71. In a similar model, the inhibition of cell growth due to treatment with rosiglitazone seems to be mediated through the increased expression of the transcriptional repressor TSC22 (REF. 72). The NF-κB pathway and the anti-apoptotic protein BCL2 might also be PPARγ targets in colon cancer cell lines, as the induction of apoptosis in these cells by 15-deoxy-PGJ2 treatment involves both of them73. In non-small-cell lung carcinoma cells, troglitazone mediates growth inhibitory effects through activation of GADD153. So far, however, it is not clear that GADD153 is a direct target of PPARγ 74. Treatment of human primary macrophages, colorectal, breast and pancreatic cancer cell lines with the PPARγ agonist rosiglitazone stimulates expression of the tumoursuppressor gene PTEN 75,76. This effect is probably mediated through the binding of PPARγ on the PPREs that are present in the promoter of that gene, indicating that PTEN is a direct PPARγ target gene. In addition to the consequences of its activation on cell-cycle arrest, PPARγ might also have anti-carcinogenic functions through inhibition of angiogenesis. PPARγ ligands inhibit proliferation and tube formation by human umbilical-vein endothelial cells in vitro and 15-deoxyPGJ2 was able to inhibit angiogenesis in vivo in a rat cornea model77. The effects of PPARγ ligands on the cell cycle should be considered carefully with regards to PPARγ activation though, as these molecules are known to induce PPARγ-independent effects (BOX 2). As reported above, exciting anti-carcinogenic effects of the PPARγ ligands have been described in many different cell lines. Importantly, these anti-carcinogenic effects were also observed in patients, although in a limited number of cases. Striking results were obtained in clinical trials that were set up to test PPAR-agonist efficiency in the treatment of liposarcomas. Consistent with the effect of the PPAR agonist pioglitazone in cell culture78, troglitazone efficiently induced differentiation and cell-cycle arrest of liposarcoma in three patients79. However, long-term monitoring of these patients would be required to determine whether PPARγ-agonist administration could be useful in the clinic. Several important observations should be kept in mind when considering the use of PPARγ agonists in the clinic. First, many of the anti-carcinogenic effects of the PPARγ agonists are described in cell culture and might not be clinically relevant. Second, and as mentioned above, some of these effects might be independent of PPARγ activation (BOX 2). Third, Pparγ activation in mice has opposite effects on the development of colon cancer, depending on the Apc genetic background. The importance of such genetic differences is twofold: APC mutations should be systematically searched for in patients suffering from type II diabetes who are treated with PPARγ agonists and in patients to whom an anticancer therapy with these molecules could be proposed. The administration of TZDs for anticancer therapy might also have side effects on the


REVIEWS Table 2 | Effects of PPARγγ agonists in various cell types Consequence of the treament Cell type Growth arrest

References

Human colorectal cancer Breast cancer Prostate cancer Myeloid leukaemia Human neuroblastoma Human hepatoma Vascular smooth muscle Human gastric cancer Thyroid carcinoma cells Uterine leiomyoma smooth-muscle cells Oligodendrocyte-like cells Pancreatic cell line

118 53,54 59 119 120 121 122 123 124 125 126 68

Cell differentiation

Colon cancer cell lines Macrophages Prostate cancer cells Non-small-cell lung carcinoma Liposarcomas Oligodendrocyte-like cells; rat spinal-cord-derived oligodendrocytes

45 127 58 74,128 78,79 126

Apoptosis

Human colon cancer cell line Breast cancer cells Macrophages B-lineage cells Human liver cancer cells Choriocarcinoma cells Adipocytes

129 54 130 131 132 133 134

Decrease of spontaneous immortalization

Li–Fraumeni-syndrome-derived human mammary epithelial cells

135

This table summarizes the effects of treatment with PPARγ agonists in various cell types, with regards to cell cycle. In most of the studies, treatment of the cells with a PPARγ agonist has anticarcinogenic consequences. Most of these results remain to be confirmed in vivo and their clinical relevance is not yet proven. Additionally, it is important to note that PPARγ agonists might also have PPARγ-independent effects (see Box 2) that were not separated from the PPARγ-dependent effects in most of the studies.

metabolism of patients because of their hypoglycaemic properties. Finally, several reports — particularly concerning the development of colon tumours as mentioned above — insist on the importance of an early treatment with the PPARγ agonists, limiting the benefits to the early steps of oncogenesis. So, successful treatment with PPARγ agonists might depend on monitoring the at risk population, which is realistic only for a limited number of tumours. For instance, people with a familial background of adenomatous polyposis or breast cancer could be monitored to

detect early carcinogenic lesions. An additional limitation is that PPARγ might not be expressed in some tumours80, therefore limiting the application of the treatment with PPARγ agonists. Despite these limitations, clinical assays have shown that, in the future, PPARγ might become an interesting target in the context of cancer therapy. However, this might be restricted to some specific tumours and patients under treatment should remain carefully monitored52,81. β/δδ PPARβ

This third PPAR isotype was called Pparβ when it was first isolated from a Xenopus oocyte library 82, but the mammalian Pparβ gene was not obviously homologous to the Xenopus gene, so it was named Pparδ when identified in the mouse83, Faar (fatty-acid-activated receptor) in the rat84 and NUC1 in humans85,86. It has now become clear that they are bona fide orthologues and, for clarity, it has been designated here as PPARβ/δ. Recent findings show that PPARβ/δ has key roles not only in embryonic development and lipid metabolism in peripheral tissues, but also in important cell functions such as adhesion, proliferation, differentiation and survival. Because PPARβ/δ is pivotal in the control of these functions, one can easily speculate that PPARβ/δ could be implicated in carcinogenesis and metastatic progression of tumours. So, what are the links between PPARβ/δ activity and the development of tumours? The first link between Pparβ/δ and carcinogenesis was established in the colon, in which it is expressed at high levels. In murine colorectal cells, the Apc– β-catenin tumour-suppressor pathway was shown to repress Pparβ/δ expression87. A recent report showed that Pparβ/δ is also a target of the Ras pathway, as its expression and activation are increased in rat intestinal cells by overexpression of activated oncogenic Kras88. Both studies propose that upregulated expression of Cox2 might modulate the activity of Pparβ/δ by producing Pparβ/δ activators such as Pgi2 (REFS 87–89). However, the consequence of a sustained and increased activation of Pparβ/δ in vivo is still unclear. Recently, two in vivo models were set up to analyse the functional consequence of a lack of PPARβ/δ expression

Box 2 | PPARγγ-independent functions for PPARγγ agonists Thiazolidinediones (TZDs) are peroxisome-proliferator-activated receptor-γ (PPARγ) activators that are known for their insulin-sensitizing and anticancer properties, whereas 15-deoxyPGJ2 is considered to be the main natural PPARγ agonist. It is generally assumed that the properties of these molecules are the result of PPARγ activation, but they also have effects that are independent of PPARγ. For instance, growth inhibition by TZDs is efficient in wild-type and in Pparγ-null mouse embryonic stem cells110. Troglitazone, a compound of the TZD family, upregulates early growth response-1 (EGR1) — a transcription factor that is thought to be a tumour suppressor — independently of PPARγ. Even though this does not exclude the effects of troglitazone mediated through PPARγ activation, it indicates that this compound also has PPARγ-independent anti-tumorigenic properties111. 15-DeoxyPGJ2 induces apoptosis in breast cancer cells, an effect that does not necessarily require PPARγ 57. Overall, the results of the treatments by PPARγ agonists on the cell cycle might be due to PPARγ activation or to other molecular mechanisms. The anticancer properties of a particular compound certainly remain interesting independently of the underlying molecular mechanism. However, it is worth discussing the fact that PPARγ is most probably not the unique target of TZD and 15-deoxyPGJ2, as this should unveil new anticancer molecular mechanisms, independent of PPARγ and, therefore, putative drug targets. The existence of PPARγ-null cells, PPARγ antagonists or tools such as RNA interference should now allow the distinction between PPARγ-dependent and PPARγ-independent pathways.

66



REVIEWS

a

b Pparβ/δ

Proliferation in keratinocytes

PPARγ

Resistance to apoptosis Migration Differentiation in inflammatory conditions

Differentiation Cell-cycle withdrawal

Preadipocytes

Liposarcoma Myeloid cell precursor

Clot Adipocyte

Resting macrophage

Adipocyte

Wound bed

Lipid-loaded macrophage

β/δδ and PPARγγ functions that relate to their carcinogenic properties. Figure 3 | PPARβ a | As demonstrated in a mouse-skin wound-healing model, Pparβ/δ inhibits keratinocyte proliferation and participates in inflammation-induced keratinocyte differentiation, which are anticarcinogenic actions. However, it also increases both migration and keratinocyte resistance to Tnf-α-induced apoptosis. b | PPARγ is implicated in the differentiation of pre-adipocytes to adipocytes and of monocytes to macrophages. In the presence of PPARγ and retinoid receptor (RXR) ligands, myeloid-cell precursors become resting macrophages, which can be turned to lipid-loaded macrophages, when PPARγ and RXR ligands are maintained. PPARγ can also withdraw liposarcoma-derived cells from cell division to trigger their differentiation to adipocytes.

on the development of colorectal carcinoma. In a PPARβ/δ-null colorectal cancer cell xenograft model, the absence of PPARβ/δ decreased tumorigenicity, indicating that decreased expression of PPARβ/δ suppresses tumorigenesis90. In the second model, an Apc MinPparβ/δ-null mouse was created, which showed that Pparβ/δ is dispensable for polyp formation91. However, increased activity of PPARβ/δ on an APC-null background might be one of the factors that leads to colorectal tumorigenesis. Indeed, PPARβ/δ expression is high in colorectal cancer and is decreased by APC in colorectal cancer cells. Whether increased PPARβ/δ expression is only a consequence of the lack of a functional APC pathway or whether it aggravates the carcinogenic process remains to be determined87. So, the involvement of PPARβ/δ in colorectal neoplasia in vivo is probably highly complex, as its expression and activity seem to be regulated by at least two pathways that are important for the normal intestinal cell cycle. PPARβ/δ expression and level of activity in tumours other than colorectal cancers is poorly documented. Increased expression of PPARβ/δ is reported in head and neck squamous carcinoma92, endometrial adenocarcinomas93 and human breast cancer cell lines94, even though in the latest case no correlation between the level of PPARβ/δ expression and tumorigenicity could be made. So, definitive evidence that PPARβ/δ is tightly linked to carcinogenesis is still lacking. Nevertheless, PPARβ/δ is ubiquitously expressed and is involved in the control of cell functions that have to be tightly balanced to avoid neoplasia. The involvement of PPARβ/δ in the control of the cell cycle has been reported in many different models that relate to either cell proliferation, differentiation or survival. In mouse keratinocytes, Pparβ/δ inhibits


proliferation and promotes cell survival and migration95–97. PPARβ/δ exerts its anti-apoptotic functions through increased expression of ILK and PDK1, which are important in signalling pathways that control cell adhesion, proliferation and survival. ILK and PDK1, in turn, phosphorylate and activate the survival factor AKT1 (REF. 10). Because all of these proteins are ubiquitously expressed, it is likely that PPARβ/δ promotes viability in other cell types. PPARβ/δ also consistently contributes to survival in kidney cells following hypertonic stress98. In contrast with these data, PGI2 was shown to promote apoptosis in a kidney cell line, most probably through PPARβ/δ activation99. Recently, it was proposed that PPARβ/δ could be a mediator of hepatic stellate-cell proliferation during liver inflammation100 and of vascular smooth-muscle cell proliferation101. PPARβ/δ might also modulate cell differentiation. Its expression is strongly increased following differentiation in human primary macrophages or in monocyte/macrophage cell lines102. Activation of Pparβ/δ using a selective agonist promotes oligodendrocyte differentiation in a mouse cell culture103, consistent with the myelination defect of the corpus callosum in Pparβ/δ-null mice97. Together with two members of the CCAAT/ enhancer-binding-protein (C/EBP) family, Pparβ/δ contributes to adipose-tissue differentiation, as underscored by the decrease in fat mass in Pparβ/δ-null mice 91,97. Taken together, the studies described above show that PPARβ/δ participates in the regulation of many cell functions that are involved in the development of tumours when uncontrolled. However, the final outcome of PPARβ/δ activation in vivo is impossible to predict based on current knowledge. Indeed, activating PPARβ/δ might have pro-carcinogenic consequences (resistance to apoptosis and increased migration properties), as well as anti-carcinogenic effects (decreased proliferation), depending on the cell- or tissue-specific context. So, further exploration needs to be made before PPARβ/δ becomes a therapeutic target, which will result in a better understanding of PPARβ/δ functions, as well as of the final outcome of its activation and/or repression in vivo, and identification of additional ligands. Therapeutic potential of PPAR modulators

PPARα and PPARγ are valuable therapeutic targets. The fibrates (which are widely used as hypolipidaemic compounds) are PPARα ligands, whereas the hypoglycaemic thiazolidinediones (which are used in the treatment of type II diabetes) are PPARγ agonists. The newly unveiled roles of PPARβ/δ in important basic cell functions — particularly well characterized in keratinocytes and, to a lesser extent, in lipid metabolism — certainly justify a further exploration of its potential as a therapeutic target in pathologies such as metabolic syndrome X, skin diseases and, perhaps, cancer. Probably linked to the therapeutic potential of PPARs in cancer therapy is the activity of COX2. COX2 catalyses the production of fatty-acid derivatives that are PPAR activators. Modulating its activity should influence the local availability of PPAR ligands, therefore indirectly modulating PPAR activity. So, using


REVIEWS modulators of PPARs and COX2 activity in parallel is certainly an interesting therapeutic avenue to investigate. Each PPAR isotype seems to be associated with carcinogenesis to a certain extent (FIGS 2,3). Sustained PPARα activation has carcinogenic consequences in the liver of rodents, but long-term usage of PPARα activators in the clinic, and epidemiological data, has proven that similar effects are unlikely to occur in humans. However, the risk might exist for newly developed compounds and, therefore, the marketing of new PPARα activators certainly deserves careful monitoring. PPARβ/δ and PPARγ might become

1. 2. 3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

68

A unified nomenclature system for the nuclear receptor superfamily. Cell 97, 161–163 (1999). Kersten, S. & Wahli, W. Peroxisome proliferator activated receptor agonists. EXS 89, 141–151 (2000). Tan, N. S. et al. Selective cooperation between fatty acid binding proteins and peroxisome proliferator-activated receptors in regulating transcription. Mol. Cell. Biol. 22, 5114–5127 (2002). This paper addresses the question of the role of fattyacid-binding proteins in the cytoplasmic transport and nuclear delivery of PPAR ligands, which are hydrophobic molecules, to their cognate nuclear receptor. Surapureddi, S. et al. Identification of a transcriptionally active peroxisome proliferator-activated receptor α-interacting cofactor complex in rat liver and characterization of PRIC285 as a coactivator. Proc. Natl Acad. Sci. USA 99, 11836–11841 (2002). Mueller, E. et al. Genetic analysis of adipogenesis through peroxisome proliferator-activated receptor γ isoforms. J. Biol. Chem. 277, 41925–41930 (2002). Krogsdam, A. M. et al. Nuclear receptor corepressordependent repression of peroxisome-proliferator-activated receptor δ-mediated transactivation. Biochem. J. 363, 157–165 (2002). Shi, Y., Hon, M. & Evans, R. M. The peroxisome proliferatoractivated receptor δ, an integrator of transcriptional repression and nuclear receptor signaling. Proc. Natl Acad. Sci. USA 99, 2613–2618 (2002). This work indicates that PPARβ/δ interacts with corepressors and competes with PPARα and PPARγ on PPAR response elements. PPARβ/δ could therefore act as a regulator of the expression of PPAR target genes, either through transcriptional activation or through competition with the two other PPAR isotypes and transcriptional repression. Dowell, P. et al. Identification of nuclear receptor corepressor as a peroxisome proliferator-activated receptor α interacting protein. J. Biol. Chem. 274, 15901–15907 (1999). Desvergne, B. & Wahli, W. Peroxisome proliferator-activated receptors: nuclear control of metabolism. Endocr. Rev. 20, 649–688 (1999). Di-Poi, N., Tan, N. S., Michalik, L., Wahli, W. & Desvergne, B. Antiapoptotic role of PPARβ in keratinocytes via transcriptional control of the Akt1 signaling pathway. Mol. Cell 10, 721–733 (2002). First description of PPARβ/δ target genes, showing that PPARβ/δ controls the AKT1 survival pathway in keratinocytes. This paper presents a model for the anti-apoptotic role of PPARβ/δ. Michalik, L. et al. PPAR expression and function during vertebrate development. Int. J. Dev. Biol. 46, 105–114 (2002). Michalik, L., Desvergne, B. & Wahli W. Peroxisome proliferator-activated receptors β/δ: emerging roles for a previously neglected third family member. Curr. Opin. Lipidol. 14, 129–135 (2003). Lock, E. A., Mitchell, A. M. & Elcombe, C. R. Biochemical mechanisms of induction of hepatic peroxisome proliferation. Annu. Rev. Pharmacol. Toxicol. 29, 145–163 (1989). Vanden Heuvel, J. P. et al. Comprehensive analysis of gene expression in rat and human hepatoma cells exposed to the peroxisome proliferator WY14,643. Toxicol. Appl. Pharmacol. 188, 185–198 (2003). Corton, J. C., Lapinskas, P. J. & Gonzalez, F. J. Central role of PPARα in the mechanism of action of hepatocarcinogenic peroxisome proliferators. Mutat. Res. 448, 139–151 (2000).


interesting therapeutic targets in the future. As discussed above, PPARγ agonists that are used in cancer treatment might have side effects on lipid and glucose metabolism, and this balance needs to be monitored. Moreover, early diagnosis and assessment of genetic predisposition will certainly be pivotal for the success of PPARγ-targeted therapy. Finally, the possible use of PPARβ/δ as a therapeutic target will largely depend on the development of specific PPARβ/δ agonists and antagonists and, of course, on the extent of our knowledge about PPARβ/δ functions in vivo, more particularly regarding its pro- and anti-carcinogenic functions.

16. Boitier, E., Gautier, J. C. & Roberts, R. Advances in understanding the regulation of apoptosis and mitosis by peroxisome-proliferator activated receptors in pre-clinical models: relevance for human health and disease. Comp. Hepatol. 2, 3 (2003). 17. Issemann, I. & Green, S. Activation of a member of the steroid hormone receptor superfamily by peroxisome proliferators. Nature 347, 645–650 (1990). 18. Lee, S. S. et al. Targeted disruption of the α isoform of the peroxisome proliferator-activated receptor gene in mice results in abolishment of the pleiotropic effects of peroxisome proliferators. Mol. Cell. Biol. 15, 3012–3022 (1995). This paper describes the phenotype of the Pparα-null mice, a clear demonstration that this receptor is the one mediating the effects of PP in rodent liver. 19. Peters, J. M., Cattley, R. C. & Gonzalez, F. J. Role of PPAR α in the mechanism of action of the nongenotoxic carcinogen and peroxisome proliferator Wy-14,643. Carcinogenesis 18, 2029–2033 (1997). 20. Marsman, D. S., Cattley, R. C., Conway, J. G. & Popp, J. A. Relationship of hepatic peroxisome proliferation and replicative DNA synthesis to the hepatocarcinogenicity of the peroxisome proliferators di(2-ethylhexyl)phthalate and [4-chloro-6-(2,3-xylidino)-2-pyrimidinylthio]acetic acid (Wy-14,643) in rats. Cancer Res. 48, 6739–6744 (1988). 21. Isenberg, J. S., Kolaja, K. L., Ayoubi, S. A., Watkins, J. B. 3rd & Klaunig, J. E. Inhibition of WY-14,643 induced hepatic lesion growth in mice by rotenone. Carcinogenesis 18, 1511–1519 (1997). 22. Peters, J. M. et al. Role of peroxisome proliferator-activated receptor α in altered cell cycle regulation in mouse liver. Carcinogenesis 19, 1989–1994 (1998). 23. Bayly, A. C., Roberts, R. A. & Dive, C. Suppression of liver cell apoptosis in vitro by the non-genotoxic hepatocarcinogen and peroxisome proliferator nafenopin. J. Cell Biol. 125, 197–203 (1994). 24. Oberhammer, F. A. & Qin, H. M. Effect of three tumour promoters on the stability of hepatocyte cultures and apoptosis after transforming growth factor-β 1. Carcinogenesis 16, 1363–1371 (1995). 25. Gill, J. H., James, N. H., Roberts, R. A. & Dive, C. The nongenotoxic hepatocarcinogen nafenopin suppresses rodent hepatocyte apoptosis induced by TGFβ1, DNA damage and Fas. Carcinogenesis 19, 299–304 (1998). 26. Gill, J. H., Brickell, P., Dive, C. & Roberts, R. A. The rodent non-genotoxic hepatocarcinogen nafenopin suppresses apoptosis preferentially in non-cycling hepatocytes but also elevates CDK4, a cell cycle progression factor. Carcinogenesis 19, 1743–1747 (1998). 27. Roberts, R. A. Non-genotoxic hepatocarcinogenesis: suppression of apoptosis by peroxisome proliferators. Ann. NY Acad. Sci. 804, 588–611 (1996). 28. Roberts, R. A., James, N. H., Woodyatt, N. J., Macdonald, N. & Tugwood, J. D. Evidence for the suppression of apoptosis by the peroxisome proliferator activated receptor α (PPAR α). Carcinogenesis 19, 43–48 (1998). 29. Reddy, J. K. & Rao, M. S. Oxidative DNA damage caused by persistent peroxisome proliferation: its role in hepatocarcinogenesis. Mutat. Res. 214, 63–68 (1989). 30. Yeldandi, A. V., Rao, M. S. & Reddy, J. K. Hydrogen peroxide generation in peroxisome proliferator-induced oncogenesis. Mutat. Res. 448, 159–177 (2000). 31. Rusyn, I., Rose, M. L., Bojes, H. K. & Thurman, R. G. Novel role of oxidants in the molecular mechanism of action of peroxisome proliferators. Antioxid. Redox Signal. 2, 607–621 (2000).

32. Rusyn, I. et al. Phthalates rapidly increase production of reactive oxygen species in vivo: role of Kupffer cells. Mol. Pharmacol. 59, 744–750 (2001). 33. Ashby, J. et al. Mechanistically-based human hazard assessment of peroxisome proliferator-induced hepatocarcinogenesis. Hum. Exp. Toxicol. 13, S1–S117 (1994). 34. Cattley, R. C. et al. Do peroxisome proliferating compounds pose a hepatocarcinogenic hazard to humans? Regul. Toxicol. Pharmacol. 27, 47–60 (1998). 35. Bentley, P. et al. Hepatic peroxisome proliferation in rodents and its significance for humans. Food Chem. Toxicol. 31, 857–907 (1993). 36. Scotto, C., Keller, J. M., Schohn, H. & Dauca, M. Comparative effects of clofibrate on peroxisomal enzymes of human (Hep EBNA2) and rat (FaO) hepatoma cell lines. Eur. J. Cell Biol. 66, 375–381 (1995). 37. Palmer, C. N., Hsu, M. H., Griffin, K. J., Raucy, J. L. & Johnson, E. F. Peroxisome proliferator activated receptor-α expression in human liver. Mol. Pharmacol. 53, 14–22 (1998). 38. Tugwood, J. D., Aldridge, T. C., Lambe, K. G., Macdonald, N. & Woodyatt, N. J. Peroxisome proliferator-activated receptors: structures and function. Ann. NY Acad. Sci. 804, 252–265 (1996). 39. Maloney, E. K. & Waxman, D. J. trans-Activation of PPARα and PPARγ by structurally diverse environmental chemicals. Toxicol. Appl. Pharmacol. 161, 209–218 (1999). 40. Woodyatt, N. J., Lambe, K. G., Myers, K. A., Tugwood, J. D. & Roberts, R. A. The peroxisome proliferator (PP) response element upstream of the human acyl CoA oxidase gene is inactive among a sample human population: significance for species differences in response to PPs. Carcinogenesis 20, 369–372 (1999). 41. Lambe, K. G., Woodyatt, N. J., Macdonald, N., Chevalier, S. & Roberts, R. A. Species differences in sequence and activity of the peroxisome proliferator response element (PPRE) within the acyl CoA oxidase gene promoter. Toxicol. Lett. 110, 119–127 (1999). 42. Lake, B. G., Phillips, J. C., Linnell, J. C. & Gangolli, S. D. The in vitro hydrolysis of some phthalate diesters by hepatic and intestinal preparations from various species. Toxicol. Appl. Pharmacol. 39, 239–248 (1977). 43. Rhodes, C. et al. Comparative pharmacokinetics and subacute toxicity of di(2-ethylhexyl) phthalate (DEHP) in rats and marmosets: extrapolation of effects in rodents to man. Environ. Health Perspect. 65, 299–307 (1986). 44. Anderson, W. A., Castle, L., Scotter, M. J., Massey, R. C. & Springall, C. A biomarker approach to measuring human dietary exposure to certain phthalate diesters. Food Addit. Contam. 18, 1068–1074 (2001). 45. Sarraf, P. et al. Differentiation and reversal of malignant changes in colon cancer through PPARγ. Nature Med. 4, 1046–1052 (1998). 46. Tanaka, T. et al. Ligands for peroxisome proliferatoractivated receptors α and γ inhibit chemically induced colitis and formation of aberrant crypt foci in rats. Cancer Res. 61, 2424–2428 (2001). 47. Osawa, E. et al. Peroxisome proliferator-activated receptor γ ligands suppress colon carcinogenesis induced by azoxymethane in mice. Gastroenterology 124, 361–367 (2003). 48. Saez, E. et al. Activators of the nuclear receptor PPARγ enhance colon polyp formation. Nature Med. 4, 1058–1061 (1998). 49. Lefebvre, A. M. et al. Activation of the peroxisome proliferator-activated receptor γ promotes the development of colon tumors in C57BL/6J-APCMin/+ mice. Nature Med. 4, 1053–1057 (1998).


REVIEWS 50. Wasan, H. S., Novelli, M., Bee, J. & Bodmer, W. F. Dietary fat influences on polyp phenotype in multiple intestinal neoplasia mice. Proc. Natl Acad. Sci. USA 94, 3308–3313 (1997). 51. Girnun, G. D. et al. APC-dependent suppression of colon carcinogenesis by PPARγ. Proc. Natl Acad. Sci. USA 99, 13771–13776 (2002). Paper that addresses the controversial role of Pparγγ activation in colon carcinogenesis. The data presented indicate that the consequence of the activation of Pparγ on the development of colon tumours might depend on the existence of a previous mutation event in the Apc–β-catenin pathway. 52. Girnun, G. D. & Spiegelman, B. M. PPARγ ligands: taking Ppart in chemoprevention. Gastroenterology 124, 564–567 (2003). 53. Mueller, E. et al. Terminal differentiation of human breast cancer through PPAR γ. Mol. Cell 1, 465–470 (1998). 54. Elstner, E. et al. Ligands for peroxisome proliferatoractivated receptor γ and retinoic acid receptor inhibit growth and induce apoptosis of human breast cancer cells in vitro and in BNX mice. Proc. Natl Acad. Sci. USA 95, 8806–8811 (1998). 55. Mehta, R. G., Williamson, E., Patel, M. K. & Koeffler, H. P. A ligand of peroxisome proliferator-activated receptor γ, retinoids, and prevention of preneoplastic mammary lesions. J. Natl Cancer Inst. 92, 418–423 (2000). 56. Suh, N. et al. A new ligand for the peroxisome proliferatoractivated receptor-γ (PPAR-γ), GW7845, inhibits rat mammary carcinogenesis. Cancer Res. 59, 5671–5673 (1999). 57. Clay, C. E., Monjazeb, A., Thorburn, J., Chilton, F. H. & High, K. P. 15-Deoxy-δ12,14-prostaglandin J2-induced apoptosis does not require PPARγ in breast cancer cells. J. Lipid Res. 43, 1818–1828 (2002). 58. Kubota, T. et al. Ligand for peroxisome proliferator-activated receptor γ (troglitazone) has potent antitumor effect against human prostate cancer both in vitro and in vivo. Cancer Res. 58, 3344–3352 (1998). 59. Mueller, E. et al. Effects of ligand activation of peroxisome proliferator-activated receptor γ in human prostate cancer. Proc. Natl Acad. Sci. USA 97, 10990–10995 (2000). 60. Hisatake, J. I. et al. Down-regulation of prostate-specific antigen expression by ligands for peroxisome proliferatoractivated receptor γ in human prostate cancer. Cancer Res. 60, 5494–5498 (2000). 61. Aldred, M. A. et al. Peroxisome proliferator-activated receptor γ is frequently downregulated in a diversity of sporadic nonmedullary thyroid carcinomas. Oncogene 22, 3412–3416 (2003). 62. Altiok, S., Xu, M. & Spiegelman, B. M. PPARγ induces cell cycle withdrawal: inhibition of E2F/DP DNA-binding activity via down-regulation of PP2A. Genes Dev. 11, 1987–1998 (1997). 63. Shao, D. & Lazar, M. A. Peroxisome proliferator activated receptor γ, CCAAT/enhancer-binding protein α, and cell cycle status regulate the commitment to adipocyte differentiation. J. Biol. Chem.272, 21473–21478 (1997). 64. Heaney, A. P., Fernando, M. & Melmed, S. PPAR-γ receptor ligands: novel therapy for pituitary adenomas. J. Clin. Invest. 111, 1381–1388 (2003). 65. Morrison, R. F. & Farmer, S. R. Role of PPARγ in regulating a cascade expression of cyclin-dependent kinase inhibitors, p18(INK4c) and p21(Waf1/Cip1), during adipogenesis. J. Biol. Chem.274, 17088–17097 (1999). 66. Clay, C. E., Atsumi, G. I., High, K. P. & Chilton, F. H. Early de novo gene expression is required for 15-deoxy-∆ 12,14prostaglandin J2-induced apoptosis in breast cancer cells. J. Biol. Chem. 276, 47131–47135 (2001). 67. Kitamura, S. et al. PPARγ agonists inhibit cell growth and suppress the expression of cyclin D1 and EGF-like growth factors in ras-transformed rat intestinal epithelial cells. Int. J. Cancer 94, 335–342 (2001). 68. Toyota, M. et al. Peroxisome proliferator-activated receptor γ reduces the growth rate of pancreatic cancer cells through the reduction of cyclin D1. Life Sci. 70, 1565–1575 (2002). 69. Qin, C. et al. Peroxisome proliferator-activated receptor γ agonists induce proteasome-dependent degradation of cyclin D1 and estrogen receptor α in MCF-7 breast cancer cells. Cancer Res. 63, 958–964 (2003). 70. Wang, C. et al. Inhibition of cellular proliferation through IκB kinase-independent and peroxisome proliferator-activated receptor γ-dependent repression of cyclin D1. Mol. Cell. Biol. 21, 3057–3070 (2001). 71. Gupta, R. A., Brockman, J. A., Sarraf, P., Willson, T. M. & DuBois, R. N. Target genes of peroxisome proliferatoractivated receptor γ in colorectal cancer cells. J. Biol. Chem. 276, 29681–29687 (2001). 72. Gupta, R. A. et al. Peroxisome proliferator-activated receptor γ and transforming growth factor-β pathways inhibit intestinal epithelial cell growth by regulating levels of TSC22. J. Biol. Chem. 278, 7431–7438 (2003).


73. Chen, G. G. et al. Apoptosis induced by activation of peroxisome-proliferator activated receptor-γ is associated with Bcl-2 and NF-κB in human colon cancer. Life Sci. 70, 2631–2646 (2002). 74. Satoh, T. et al. Activation of peroxisome proliferatoractivated receptor-γ stimulates the growth arrest and DNAdamage inducible 153 gene in non-small cell lung carcinoma cells. Oncogene 21, 2171–2180 (2002). 75. Patel, L. et al. Tumor suppressor and anti-inflammatory actions of PPARγ agonists are mediated via upregulation of PTEN. Curr. Biol. 11, 764–768 (2001). 76. Farrow, B. & Evers, B. M. Activation of PPARγ increases PTEN expression in pancreatic cancer cells. Biochem. Biophys. Res. Commun. 301, 50–53 (2003). 77. Xin, X., Yang, S., Kowalski, J. & Gerritsen, M. E. Peroxisome proliferator-activated receptor γ ligands are potent inhibitors of angiogenesis in vitro and in vivo. J. Biol. Chem. 274, 9116–9121 (1999). 78. Tontonoz, P. et al. Terminal differentiation of human liposarcoma cells induced by ligands for peroxisome proliferator-activated receptor γ and the retinoid X receptor. Proc. Natl Acad. Sci. USA 94, 237–241 (1997). 79. Demetri, G. D. et al. Induction of solid tumor differentiation by the peroxisome proliferator-activated receptor-γ ligand troglitazone in patients with liposarcoma. Proc. Natl Acad. Sci. USA 96, 3951–3956 (1999). Paper that describes the striking beneficial effects that are observed after PPARγγ agonists are given to treat patients suffering from liposarcoma. These data indicate that troglitazone is a promising drug in the treatment of certain tumours. 80. Theocharis, S. et al. Expression of peroxisome proliferator activated receptor-γ in non-small cell lung carcinoma: correlation with histological type and grade. Lung Cancer 36, 249–255 (2002). 81. Gupta, R. A. & Dubois, R. N. Controversy: PPARγ as a target for treatment of colorectal cancer. Am. J. Physiol. Gastrointest. Liver Physiol. 283, G266–G269 (2002). 82. Dreyer, C. et al. Control of the peroxisomal β-oxidation pathway by a novel family of nuclear hormone receptors. Cell 68, 879–887 (1992). 83. Kliewer, S. A. et al. Differential expression and activation of a family of murine peroxisome proliferator-activated receptors. Proc. Natl Acad. Sci. USA 91, 7355–7359 (1994). 84. Amri, E. Z., Bonino, F., Ailhaud, G., Abumrad, N. A. & Grimaldi, P. A. Cloning of a protein that mediates transcriptional effects of fatty acids in preadipocytes. Homology to peroxisome proliferator-activated receptors. J. Biol. Chem. 270, 2367–2371 (1995). 85. Matsuo, H. & Strauss, J. F. 3rd. Peroxisome proliferators and retinoids affect JEG-3 choriocarcinoma cell function. Endocrinology 135, 1135–1145 (1994). 86. Jow, L. & Mukherjee, R. The human peroxisome proliferatoractivated receptor (PPAR) subtype NUC1 represses the activation of hPPAR α and thyroid hormone receptors. J. Biol. Chem. 270, 3836–3840 (1995). 87. He, T. C., Chan, T. A., Vogelstein, B. & Kinzler, K. W. PPARδ is an APC-regulated target of nonsteroidal anti-inflammatory drugs. Cell 99, 335–345 (1999). Important report on the identification of PPARβ/δ as a potential target of the APC–β-catenin tumoursuppressor pathway. This paper also indicates that the activity COX2 in colon tumours might provide PPAR ligands. 88. Shao, J., Sheng, H. & DuBois, R. N. Peroxisome proliferatoractivated receptors modulate K-Ras-mediated transformation of intestinal epithelial cells. Cancer Res. 62, 3282–3288 (2002). 89. Gupta, R. A. et al. Prostacyclin-mediated activation of peroxisome proliferator-activated receptor δ in colorectal cancer. Proc. Natl Acad. Sci. USA 97, 13275–13280 (2000). 90. Park, B. H., Vogelstein, B. & Kinzler, K. W. Genetic disruption of PPARδ decreases the tumorigenicity of human colon cancer cells. Proc. Natl Acad. Sci. USA 98, 2598–2603 (2001). Based on a model of grafts using PPARβ/δ wild-type or null cells, this paper shows that PPARβ/δ activity can affect tumorigenesis and indicates that PPARβ/δ inhibitors could inhibit tumour growth. 91. Barak, Y. et al. Effects of peroxisome proliferator-activated receptor δ on placentation, adiposity, and colorectal cancer. Proc. Natl Acad. Sci. USA 99, 303–308 (2002). 92. Jaeckel, E. C. et al. Correlation of expression of cyclooxygenase-2, vascular endothelial growth factor, and peroxisome proliferator-activated receptor δ with head and neck squamous cell carcinoma. Arch. Otolaryngol. Head Neck Surg. 127, 1253–1259 (2001). 93. Tong, B. J. et al. Heightened expression of cyclooxygenase-2 and peroxisome proliferator-activated receptor-δ in human endometrial adenocarcinoma. Neoplasia 2, 483–490 (2000).

94. Suchanek, K. M., May, F. J., Lee, W. J., Holman, N. A. & Roberts-Thomson, S. J. Peroxisome proliferator-activated receptor β expression in human breast epithelial cell lines of tumorigenic and non-tumorigenic origin. Int. J. Biochem. Cell Biol. 34, 1051–1058 (2002). 95. Michalik, L. et al. Impaired skin wound healing in peroxisome proliferator-activated receptor (PPAR)α and PPARβ mutant mice. J. Cell Biol. 154, 799–814 (2001). 96. Tan, N. S. et al. Critical roles of PPARβ/δ in keratinocyte response to inflammation. Genes Dev. 15, 3263–3277 (2001). 97. Peters, J. M. et al. Growth, adipose, brain, and skin alterations resulting from targeted disruption of the mouse peroxisome proliferator-activated receptor β(δ). Mol. Cell. Biol. 20, 5119–5128 (2000). 98. Hao, C. M., Redha, R., Morrow, J. & Breyer, M. D. Peroxisome proliferator-activated receptor δ activation promotes cell survival following hypertonic stress. J. Biol. Chem. 277, 21341–21345 (2002). 99. Hatae, T., Wada, M., Yokoyama, C., Shimonishi, M. & Tanabe, T. Prostacyclin-dependent apoptosis mediated by PPAR δ. J. Biol. Chem. 276, 46260–46267 (2001). 100. Hellemans, K. et al. Peroxisome proliferator-activated receptor-β signaling contributes to enhanced proliferation of hepatic stellate cells. Gastroenterology 124, 184–201 (2003). 101. Zhang, J. et al. Peroxisome proliferator-activated receptor δ is up-regulated during vascular lesion formation and promotes post-confluent cell proliferation in vascular smooth muscle cells. J. Biol. Chem. 277, 11505–11512 (2002). 102. Vosper, H. et al. The peroxisome proliferator-activated receptor δ promotes lipid accumulation in human macrophages. J. Biol. Chem. 276, 44258–44265 (2001). 103. Saluja, I., Granneman, J. G. & Skoff, R. P. PPAR δ agonists stimulate oligodendrocyte differentiation in tissue culture. Glia 33, 191–204 (2001). 104. Dannenberg A. J. & DuBois, R. N. COX–2. A new target for cancer prevention and treatment (Karger, New Brunswick, 2003). 105. Purdue, P. E. & Lazarow, P. B. Peroxisome biogenesis. Annu. Rev. Cell Dev. Biol. 17, 701–752 (2001). 106. Kersten, S., Desvergne, B. & Wahli, W. Roles of PPARs in health and disease. Nature 405, 421–424 (2000). 107. Deeb, S. S. et al. A Pro12Ala substitution in PPARγ2 associated with decreased receptor activity, lower body mass index and improved insulin sensitivity. Nature Genet. 20, 284–287 (1998). 108. Ristow, M., Muller-Wieland, D., Pfeiffer, A., Krone, W. & Kahn, C. R. Obesity associated with a mutation in a genetic regulator of adipocyte differentiation. N. Engl. J. Med. 339, 953–959 (1998). 109. Barroso, I. et al. Dominant negative mutations in human PPARγ associated with severe insulin resistance, diabetes mellitus and hypertension. Nature 402, 880–883 (1999). 110. Palakurthi, S. S., Aktas, H., Grubissich, L. M., Mortensen, R. M. & Halperin, J. A. Anticancer effects of thiazolidinediones are independent of peroxisome proliferator-activated receptor γ and mediated by inhibition of translation initiation. Cancer Res. 61, 6213–6218 (2001). 111. Baek, S. J., Wilson, L. C., Hsi, L. C. & Eling, T. E. Troglitazone, a peroxisome proliferator-activated receptor γ (PPAR γ) ligand, selectively induces the early growth response-1 gene independently of PPAR γ. A novel mechanism for its anti-tumorigenic activity. J. Biol. Chem. 278, 5845–5853 (2003). 112. Sarraf, P. et al. Loss-of-function mutations in PPAR γ associated with human colon cancer. Mol. Cell 3, 799–804 (1999). 113. Kroll, T. G. et al. PAX8–PPARγ1 fusion oncogene in human thyroid carcinoma. Science 289, 1357–1360 (2000). 114. Marques, A. R. et al. Expression of PAX8–PPAR γ 1 rearrangements in both follicular thyroid carcinomas and adenomas. J. Clin. Endocrinol. Metab. 87, 3947–3952 (2002). 115. Nikiforova, M. N. et al. RAS point mutations and PAX8–PPAR γ rearrangement in thyroid tumors: evidence for distinct molecular pathways in thyroid follicular carcinoma. J. Clin. Endocrinol. Metab. 88, 2318–2326 (2003). 116. Cheung, L. et al. Detection of the PAX8–PPAR γ fusion oncogene in both follicular thyroid carcinomas and adenomas. J. Clin. Endocrinol. Metab. 88, 354–357 (2003). 117. Ikezoe, T. et al. Mutational analysis of the peroxisome proliferator-activated receptor γ gene in human malignancies. Cancer Res. 61, 5307–5310 (2001). 118. Brockman, J. A., Gupta, R. A. & Dubois, R. N. Activation of PPARγ leads to inhibition of anchorage-independent growth of human colorectal cancer cells. Gastroenterology 115, 1049–1055 (1998).


REVIEWS 119. Asou, H. et al. Growth inhibition of myeloid leukemia cells by troglitazone, a ligand for peroxisome proliferator activated receptor γ, and retinoids. Int. J. Oncol. 15, 1027–1031 (1999). 120. Han, S. W., Greene, M. E., Pitts, J., Wada, R. K. & Sidell, N. Novel expression and function of peroxisome proliferatoractivated receptor γ (PPARγ) in human neuroblastoma cells. Clin. Cancer Res. 7, 98–104 (2001). 121. Koga, H. et al. Involvement of p21(WAF1/Cip1), p27(Kip1), and p18(INK4c) in troglitazone-induced cell-cycle arrest in human hepatoma cell lines. Hepatology 33, 1087–1097 (2001). 122. Wakino, S. et al. Peroxisome proliferator-activated receptor γ ligands inhibit retinoblastoma phosphorylation and G1→S transition in vascular smooth muscle cells. J. Biol. Chem. 275, 22435–22441 (2000). 123. Takahashi, N. et al. Activation of PPARγ inhibits cell growth and induces apoptosis in human gastric cancer cells. FEBS Lett. 455, 135–139 (1999). 124. Ohta, K., Endo, T., Haraguchi, K., Hershman, J. M. & Onaya, T. Ligands for peroxisome proliferator-activated receptor γ inhibit growth and induce apoptosis of human papillary thyroid carcinoma cells. J. Clin. Endocrinol. Metab. 86, 2170–2177 (2001). 125. Houston, K. D. et al. Inhibition of proliferation and estrogen receptor signaling by peroxisome proliferator-activated receptor γ ligands in uterine leiomyoma. Cancer Res. 63, 1221–1227 (2003). 126. Roth, A. D. et al. PPAR γ activators induce growth arrest and process extension in B12 oligodendrocyte-like cells and terminal differentiation of cultured oligodendrocytes. J. Neurosci. Res. 72, 425–435 (2003).

70


127. Tontonoz, P., Nagy, L., Alvarez, J. G., Thomazy, V. A. & Evans, R. M. PPARγ promotes monocyte/macrophage differentiation and uptake of oxidized LDL. Cell 93, 241–252 (1998). 128. Chang, T. H. & Szabo, E. Induction of differentiation and apoptosis by ligands of peroxisome proliferator-activated receptor γ in non-small cell lung cancer. Cancer Res. 60, 1129–1138 (2000). 129. Yang, W. L. & Frucht, H. Activation of the PPAR pathway induces apoptosis and COX-2 inhibition in HT-29 human colon cancer cells. Carcinogenesis 22, 1379–1383 (2001). 130. Chinetti, G. et al. Activation of proliferator-activated receptors α and γ induces apoptosis of human monocytederived macrophages. J. Biol. Chem. 273, 25573–25580 (1998). 131. Padilla, J., Kaur, K., Cao, H. J., Smith, T. J. & Phipps, R. P. Peroxisome proliferator activator receptor-γ agonists and 15-deoxy-∆(12,14)(12,14)-PGJ(2) induce apoptosis in normal and malignant B-lineage cells. J. Immunol. 165, 6941–6948 (2000). 132. Toyoda, M. et al. A ligand for peroxisome proliferator activated receptor γ inhibits cell growth and induces apoptosis in human liver cancer cells. Gut 50, 563–567 (2002). 133. Keelan, J. A. et al. 15-Deoxy-∆(12,14)-prostaglandin J(2), a ligand for peroxisome proliferator-activated receptor-γ, induces apoptosis in JEG3 choriocarcinoma cells. Biochem. Biophys. Res. Commun. 262, 579–585 (1999). 134. Okuno, A. et al. Troglitazone increases the number of small adipocytes without the change of white adipose tissue mass

in obese Zucker rats. J. Clin. Invest. 101, 1354–1361 (1998). 135. Herbert, B. S. et al. A peroxisome proliferator-activated receptor-γ agonist and the p53 rescue drug CP-31398 inhibit the spontaneous immortalization of breast epithelial cells. Cancer Res. 63, 1914–1919 (2003).

Acknowledgements The authors wish to thank L. Gelman for critical reading of this manuscript. The work done in the authors’ laboratory was supported by the Swiss National Science Foundation (grants to W.W. and to B.D.), the Etat de Vaud and the Human Frontier Science Program Organization.

Competing interests statement The authors declare that they have no competing financial interests.

Online links DATABASES The following terms in this article are linked online to: Cancer.gov: http://cancer.gov/ breast cancer | colon cancer | endometrial cancer | head and neck cancer | non-small-cell lung cancer | prostate cancer | thyroid cancer LocusLink: http://www.ncbi.nlm.nih.gov/LocusLink/ AKT1 | Apc | β-catenin | cyclin D1 | GADD153 | ILK | INK4C | keratin 20 | NF-κB | PDK1 | Pgi2 | PPARα | PPARβ/δ | PPARγ | PTEN | Rb | RXR | TSC22 | WAF1 OMIM: http://www.ncbi.nlm.nih.gov/omim/ familial adenomatous polyposis
