Q4

Q1

Q2

The American Journal of Pathology, Vol. -, No. -, - 2013

123456789

1011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162

ajp.amjpathol.org

6364656667686970717273747576777879

COMMENTARYAnticancer Activities of PPARg in Breast Cancer AreContext-DependentJin-Tang Dong

8081

From the Winship Cancer Institute, Emory University School of Medicine, Atlanta, Georgia

Accepted for publication March 15, 2013.

828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123

In this issue of The American Journal of Pathology, Nakleset al1 used mouse mammary glands with null Brca1 andhaplo-insufficient Trp53 as a tumor model to evaluate theeffect of PPAR agonist efatutazone on mammary tumori-genesis. Mice were treated at age 4 months, so the findingscould be more relevant to treatment rather than prevention.The authors found that, although efatutazone did not havea significant effect on the incidence of invasive tumors, itsignificantly reduced the incidence of non-invasivemammarytumor and promoted more differentiated histology in thetumors that formed. Efatutazone also reduced the prevalenceof dense preneoplastic lesions including lobular growth andthe number of hyperplastic alveolar nodules (HANs). Tumorsfrom the efatutazone group had reduced expression ofphosphor-AKT but not total AKT, and CDK6 was the onlycell cycle regulator that was down-regulated. These findingsnot only provide insight but also strengthen several issues inan area that could lead to new therapy in the treatment ofbreast cancer.

Targeting PPARg

The peroxisome proliferator-activated receptor (PPAR) be-longs to the nuclear receptor (NR) superfamily that includesestrogen receptors (ERs), retinoic acid receptors (RARs),retinoid X receptors (RXRs), and the vitamin D receptor(VDR). PPARg is one of the PPARs that are expressed indifferent tissues. On the binding of its ligand, PPARg heter-odimerizes with RXR to form a complex that translocates intothe nucleus, where the complex and its coactivators bind togene promoters to regulate the transcription of PPARg targetgenes. Known coactivators and corepressors of PPARginclude the histone acetyltransferase p300 (CBP), SRC-1,TEF2, Drip205 (Med220), and PGC-1. Binding of ligand toPPARg usually leads to its dissociation from corepressors andassociation with coactivators.2 PPARg regulates differentbiological processes including adipocyte differentiation,

Copyright ª 2013 American Society for Investigative Pathology.

Published by Elsevier Inc. All rights reserved.

http://dx.doi.org/10.1016/j.ajpath.2013.03.005

DIS 5.1.0 DTD � AJPA1336_proof

glucose homeostasis, cell proliferation and differentiation,inflammation, apoptosis, angiogenesis, development, andcarcinogenesis.3,4 The diverse functions of PPARg aremediated by different group of genes that are known toregulate these processes, including those involved in cellcycle arrest, apoptosis, and DNA damage response.5,6

After the widely used anti-diabetic thiazolidinedione(TZD) drugs were found to be ligands for PPARg,7 somenatural ligands for PPARg have been identified and used toelucidate the role of PPARg in cellular functions both in vitroand in vivo.8 Derivatives of TZD have also been developedwith higher specificities and activities. The agonist used inthis study, efatutazone (also called CS-7017 or RS5444), isa selective high-affinity TZD-class agonist of PPARg. Efa-tutazone induces PPARg-dependent transactivation but hasno effect on the activation of either PPARa or PPARD.

In human tissues, the PPARG gene is expressed in normalcells and can undergo mutations in some cancers. Forexample, PPARGmutation has been detected in colon cancerand chromosomal translocation that produces an oncogenicfusion protein involving PPARg; ie, PAX8ePPARg, occursin thyroid follicular carcinoma,9 which suggests that PPARgcould be oncogenic. In breast cancer, however, the PPARGgene is not mutated. PPARg is expressed in both normal andmalignant mammary tissues and its expression level is oftenhigher in cancer tissues.10 However, an increased expressionin cancer cells does not necessarily mean an oncogenic role intumor development. Therefore, there is no genetic evidencefor either a tumor suppressor or an oncogenic function ofPPARg in breast cancer.

PPARg as a Tumor Suppressor?

The key finding of Nakles et al1 is that the activation ofPPARg by its agonist reduced the incidence of non-invasive

124

� 22 April 2013 � 1:53 pm � EO: AJP13_0147

J.-T. Dong

125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186

187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247

cancer and promoted more differentiated histology withoutaffecting the incidence and histology of invasive cancers.These results clearly indicate a tumor suppressor functionfor PPARg in breast cancer and confirm the complexity ofPPARg functions in carcinogenesis. Targeting the signalingpathways transduced by NRs is a valid approach for thetreatment and prevention of different diseases includingcancer. One well known example is the commonly pre-scribed tamoxifen, an ERa antagonist, for the treatment ofERa-positive breast cancers. As a crucial regulator of cellproliferation, differentiation, and apoptosis,11 PPARg hasbeen evaluated as a target for therapeutic development, andagonists for PPARg have been developed for the treatmentof different diseases.2

However, PPARg appears to have contradicting functionsin tumorigenesis, which is likely determined by the molecularcontexts of cells. First of all, a number of studies haveshowed a suppressive function of PPARg in cultured breastcancer cells, including both ER-positive (eg, MCF-7) andER-negative cells (eg, MDA-MB-231). The suppressiveactivities include inhibition of cell proliferation, induction ofapoptosis, and promotion of differentiation.12 The effectshave been detected in both ER-positive and ER-negativebreast cancer cells, and the findings among different studiesare generally consistent. However, findings in PPARgfunctions from animal studies are less consistent and some-times contradicting, which is somewhat parallel to the dualexpression trends of PPARg in human tumors. On one hand,there are plenty of animal studies showing a tumor suppressorfunction of PPARg. For example, TZD suppresses thetumorigenesis of MCF-7 human breast cancer cells in nudemice,13 knockout of PPARg increases tumor incidence andmetastasis in DMBA-induced tumorigenesis,14 and PPARgagonist shows both tumor suppressive and chemoprotectiveeffects on tumorigenesis induced by the combined treatmentof excess hormone and the DMBA carcinogen.15 In addition,expression of the dominant-negative Pax8-PPARg fusiongene in mouse mammary epithelial cells, although nottumorigenic itself, makes cells highly susceptible to DMBA-induced tumorigenesis.16 The paper by Nakles et al1 adds tothe list of studies.

On the other hand, an oncogenic role has also beendetected for PPARg, although the mechanism for such anopposite function is not really understood. For example,overexpression of PPARg increases the risk of mammarytumor in mice that are prone to developing the disease.17

Using the MMTV-Her-2/neu transgenic mouse model ofmammary tumorigenesis, Saez et al18 found that PPARgoverexpression did not prevent, but instead greatly accel-erated tumorigenesis, although bigenic tumors are moresecretory and differentiated. Expression of a constitutivelyactive form of PPARg also accelerated mammary tumori-genesis in the transgenic MMTV-PyV mice, which are proneto mammary cancer development.17 Contradicting findingshave also been reported in other types of cancers includingcolon cancer.17

2DIS 5.1.0 DTD � AJPA1336_proo

Clinical studies have also been inconsistent for the effectsof PPARg agonists on human breast cancer. Use of PPARgagonists TZD is associated with better survival for diabeticswith HER2-positive breast cancer,19 but patients withmetastatic breast cancer do not appear to show any clinicalbenefits of TZD use.20 There are at least several possibilitiesfor discrepancies in PPARg function in breast cancer amongdifferent studies, and Nakles et al1 provide some insight.Better understanding of how PPARg functions and interactswith other signaling pathways should help clarify the issue,thereby improving the design and outcome of clinical trials.For example, if PPARg is only relevant to more differen-tiated ER-positive breast cancer as suggested by Nakleset al1 and other studies, it would not make sense to useadvanced breast cancers, which are often metastatic and ER-negative, in a clinical trial for PPARg agonists.

The Role of Other Pathways

One possibility for inconsistent functions of PPARg is thatPPARg functions in a context-dependent manner. Forexample, certain molecular pathways need to be present orabsent for PPARg to function. Nakles et al1 used mice thatundergo a mammary epithelial cell targeted deletion of Brca1in somatic cells coupled with germline Trp53 haplo-insufficiency. These mice provide a model that is relevantto the human disease because both genes are functionallyinactivated in some human breast cancer. The findings ofa tumor suppressor function for PPARg thus suggest thatbreast cancers with abnormal Brca1 and TP53would respondto PPARg agonists. There are reported links between PPARgfunction and BRCA1 and TP53. For TP53, TZD rosiglita-zone has been shown to increase the expression of TP53 andits effector p21 (WAF1/Cip1).21 In addition, mammaryepithelial cells from mice with TP53 insufficiency undergomore frequent spontaneous immortalization, and treatmentwith rosiglitazone, another agonist of PPARg, reduces thefrequency of spontaneous immortalization.22 Taken togetherwith Nakles et al,1 it is likely that proper function of PPARgneeds functional TP53. If that be the case, efatutazonetreatment could increase TP53 expression, which was notexamined in the paper. A relevance of PPARg to BRCA1 isalso possible, as PPARg agonists induce the expression ofBRCA1 in both cultured cells and mammary tissues.23

Another proliferation-related pathway suppressed byPPARg agonist is Akt/PTEN, as TZDs increase PTEN ex-pression in MCF-7 breast cancer cells and decrease Aktphosphorylation.24 Consistently, interruption of PPARg byexpressing the PAX8/PPAR fusion gene reduces Ptenexpression and increases AKT activation in both normal andcancer cells.16Nakles et al1 confirm that efatutazone decreasedthe expression of p-AKT without affecting the total level ofAKT expression, although Pten expression was not evaluatedin the model. There is evidence for PPARg signaling tointeract with other signaling pathways, including STAT5B,NF-kB, TGF-b, and Ras/MAPK.16 Ligand activation of

ajp.amjpathol.org - The American Journal of Pathology

248

f � 22 April 2013 � 1:53 pm � EO: AJP13_0147

3

Commentary

249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310

311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371

PPARg has been shown to potentiate Wnt signaling in anin vivo model of mouse mammary tumorigenesis.17 Thesepathways were not evaluated in the Nakles study,1 so it is notclear whether the effect of efatutazone on tumorigenesis in thediscussed paper is associated with any of these pathways.Improved understanding of molecular signatures associatedwith different effects of PPARg agonists should help us todevelop PPARg-based therapies in the treatment andprevention of breast cancer.

The connection with ER is another important aspect in thefunction of PPARg signaling. In the Nakles study,1 efatutazonetreatment led to the formation ofmore differentiatedER-positivetumors, which suggests a positive association between PPARgfunction and ER expression. On the other hand, inhibition ofPPARg function via expressing the dominant-negative Pax8-PPARg fusion gene made DMBA-induced mouse mammarytumors also ER-positive, and subsequently highly responsive tothe ER antagonist fulvestrant in tumor prevention.16 Based onthis study and Nakles et al,1 it becomes interesting to knowwhether combined use of efatutazone and an ER antagonist hasany additive or synergistic (or even neutralizing) effect onmammary tumorigenesis induced by Brca1 deficiency andTrp53 insufficiency.

A direct connection between PPARg function and ERtranscription has also been reported, although in an oppositemanner. Inmousemammary organ cultures, althoughDMBA-induced mammary alveolar lesions (MAL) were inhibitedby either an agonist or an antagonist of PPARg, it is theantagonist and not the agonist that induced ER and PRexpression in the presence of estrogen.25 Further insights intothe relationship between PPARg signaling and ER signalingcould lead to better intervention of breast cancer usinga combined use of PPARg agonists and ER antagonists.

Future Directions

An interesting observation by Nakles et al1 is that efatutazonelacks an effect on invasive tumors. It is possible that this groupof cancers originated from cells with distinct molecular alter-ations. It is also possible, as the authors discuss, that invasivecancers come from efatutazone-resistant cancer or cancerprogenitor cells that are not clinically well characterized.

Although not addressed in the study, it would be inter-esting to compare invasive cancers to the non-invasive, moredifferentiated, and ER-positive cancers for the molecularpathways associated with PPARg signaling. Moleculardifferences between the two groups of cancers should helpclarify what molecular pathways and signatures are requiredor altered by PPARg activation and what are specific toinvasive cancers that are not responsive to PPARg activation.

Another outstanding question is how both enhanced andattenuated PPARg activities induce ER-positive tumors inthe mammary gland. As nuclear receptors, it is not impossiblethat PPARg and ER have direct interactions. A speculationis that the pro-differentiation function of PPARg restrainsthe pro-proliferative/tumorigenic function of ER via direct

The American Journal of Pathology - ajp.amjpathol.orgDIS 5.1.0 DTD � AJPA1336_proof

interaction, and the inactivation of PPARg could thus releaseor enhance the pro-proliferative/tumorigenic function of ER,leading to the transformation of luminal cells that are alreadyER-positive to tumor cells that remain ER-positive. Thispossibility is worth testing in future studies. It is also possiblethat progenitor cells, which are ER-negative, have acquiredtumor-initiating events while differentiating into ER-positivecells under the influence of pro-differentiation function ofPPARg, and eventually become fully transformed cells thatare ER positive.

Finally, efatutazone was administered at four month of age,which leaves several questions unaddressed. For example,could efatutazone treatment modify some early cancer orprogenitor cells toward a more differentiated phenotype?Could early administration of efatutazone prevent the occur-rence of the more differentiated ER-positive cancer? Is theeffect of efatutazone reversible?

Concluding Remarks

In summary, the results produced byNakles et al1 have severalimplications: i) PPARg activation indeed interferes withtumor development, ii) the tumor suppressor function ofPPARg is more likely associated with its pro-differentiationfunction, and iii) PPARg activation is only effective inspecific cells that could be at a specific stage of malignanttransformation or have a distinct set of molecular alterations.Clarification of molecular basis for the different effects ofefatutazone on tumor development and histology should helpthe development of PPARg agonists as therapeutic and/orpreventive reagents in breast cancer and other types of cancers.

References

1. Nakles RE, Kallakury BVS, Furth PA: Efatutazone, a PPARg agonist,increases the spectrum of well-differentiated mammary cancersubtypes initiated by loss of full-length BRCA1 in association with p53haploinsufficiency. Am J Pathol 2013. In press Q

2. Han S, Roman J: Peroxisome proliferator-activated receptor gamma:a novel target for cancer therapeutics? Anticancer Drugs 2007, 18:237e244

3. Lehrke M, Lazar MA: The many faces of PPARgamma. Cell 2005,123:993e999

4. Sporn MB, Suh N, Mangelsdorf DJ: Prospects for prevention andtreatment of cancer with selective PPARgamma modulators(SPARMs). Trends Mol Med 2001, 7:395e400

5. Yu C, Markan K, Temple KA, Deplewski D, BradyMJ, Cohen RN: Thenuclear receptor corepressors NCoR and SMRT decrease peroxisomeproliferator-activated receptor gamma transcriptional activity andrepress 3T3-L1 adipogenesis. J Biol Chem 2005, 280:13600e13605

6. Van Laere SJ, Van den Eynden GG, Van der Auwera I,Vandenberghe M, van Dam P, Van Marck EA, van Golen KL,Vermeulen PB, Dirix LY: Identification of cell-of-origin breast tumorsubtypes in inflammatory breast cancer by gene expression profiling.Breast Cancer Res Treat 2006, 95:243e255

7. Lehmann JM, Moore LB, Smith-Oliver TA, Wilkison WO,Willson TM, Kliewer SA: An antidiabetic thiazolidinedione is a highaffinity ligand for peroxisome proliferator-activated receptor gamma(PPAR gamma). J Biol Chem 1995, 270:12953e12956

3

372

� 22 April 2013 � 1:53 pm � EO: AJP13_0147

J.-T. Dong

373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411

412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450

8. Belvisi MG, Hele DJ, Birrell MA: Peroxisome proliferator-activatedreceptor gamma agonists as therapy for chronic airway inflammation.Eur J Pharmacol 2006, 533:101e109

9. Lui WO, Foukakis T, Liden J, Thoppe SR, Dwight T, Hoog A,Zedenius J, Wallin G, Reimers M, Larsson C: Expression profilingreveals a distinct transcription signature in follicular thyroid carci-nomas with a PAX8-PPAR(gamma) fusion oncogene. Oncogene 2005,24:1467e1476

10. Ditsch N, Vrekoussis T, Lenhard M, Ruhl I, Gallwas J,Weissenbacher T, Friese K, Mayr D, Makrigiannakis A, Jeschke U:Retinoid X receptor alpha (RXRalpha) and peroxisome proliferator-activated receptor gamma (PPARgamma) expression in breastcancer: an immunohistochemical study. In Vivo 2012, 26:87e92

11. Sertznig P, Seifert M, Tilgen W, Reichrath J: Peroxisome proliferator-activated receptors (PPARs) and the human skin: importance of PPARsin skin physiology and dermatologic diseases. Am J Clin Dermatol2008, 9:15e31

12. Kim KY, Cheon HG: Antiangiogenic effect of rosiglitazone is medi-ated via peroxisome proliferator-activated receptor gamma-activatedmaxi-K channel opening in human umbilical vein endothelial cells.J Biol Chem 2006, 281:13503e13512

13. Elstner E, Muller C, Koshizuka K, Williamson EA, Park D,Asou H, Shintaku P, Said JW, Heber D, Koeffler HP: Ligands forperoxisome proliferator-activated receptorgamma and retinoic acidreceptor inhibit growth and induce apoptosis of human breastcancer cells in vitro and in BNX mice. Proc Natl Acad Sci U S A1998, 95:8806e8811

14. Nicol CJ, Yoon M, Ward JM, Yamashita M, Fukamachi K, Peters JM,Gonzalez FJ: PPARgamma influences susceptibility to DMBA-induced mammary, ovarian and skin carcinogenesis. Carcinogenesis2004, 25:1747e1755

15. Yin Y, Russell RG, Dettin LE, Bai R, Wei ZL, Kozikowski AP,Kopelovich L, Glazer RI: Peroxisome proliferator-activated receptordelta and gamma agonists differentially alter tumor differentiation andprogression during mammary carcinogenesis. Cancer Res 2005, 65:3950e3957

16. Yin Y, Yuan H, Zeng X, Kopelovich L, Glazer RI: Inhibition ofperoxisome proliferator-activated receptor gamma increases estrogenreceptor-dependent tumor specification. Cancer Res 2009, 69:687e694

4DIS 5.1.0 DTD � AJPA1336_proo

17. Saez E, Rosenfeld J, Livolsi A, Olson P, Lombardo E, Nelson M,Banayo E, Cardiff RD, Izpisua-Belmonte JC, Evans RM: PPARgamma signaling exacerbates mammary gland tumor development.Genes Dev 2004, 18:528e540

18. Yee LD, Young DC, Rosol TJ, Vanbuskirk AM, Clinton SK: Dietary(n-3) polyunsaturated fatty acids inhibit HER-2/neu-induced breastcancer in mice independently of the PPARgamma ligand rosiglitazone.J Nutr 2005, 135:983e988

19. He X, Esteva FJ, Ensor J, Hortobagyi GN, Lee MH, Yeung SC:Metformin and thiazolidinediones are associated with improved breastcancer-specific survival of diabetic women with HER2þ breast cancer.Ann Oncol 2012, 23:1771e1780

20. Burstein HJ, Demetri GD, Mueller E, Sarraf P, Spiegelman BM,Winer EP: Use of the peroxisome proliferator-activated receptor(PPAR) gamma ligand troglitazone as treatment for refractory breastcancer: a phase II study. Breast Cancer Res Treat 2003, 79:391e397

21. Bonofiglio D, Aquila S, Catalano S, Gabriele S, Belmonte M,Middea E, Qi H, Morelli C, Gentile M, Maggiolini M, Ando S:Peroxisome proliferator-activated receptor-gamma activates p53 genepromoter binding to the nuclear factor-kappaB sequence in humanMCF7 breast cancer cells. Mol Endocrinol 2006, 20:3083e3092

22. Herbert BS, Pearce VP, Hynan LS, LaRue DM, Wright WE,Kopelovich L, Shay JW: A peroxisome proliferator-activated receptor-gamma agonist and the p53 rescue drug CP-31398 inhibit the spon-taneous immortalization of breast epithelial cells. Cancer Res 2003, 63:1914e1919

23. Subbaramaiah K, Howe LR, Zhou XK, Yang P, Hudis CA,Kopelovich L, Dannenberg AJ: Pioglitazone, a PPARgamma agonist,suppresses CYP19 transcription: evidence for involvement of 15-hydroxyprostaglandin dehydrogenase and BRCA1. Cancer Prev Res(Phila) 2012, 5:1183e1194

24. Kim KY, Kim SS, Cheon HG: Differential anti-proliferative actions ofperoxisome proliferator-activated receptor-gamma agonists in MCF-7breast cancer cells. Biochem Pharmacol 2006, 72:530e540

25. Mehta RG, Peng X, Roy S, Hawthorne M, Kalra A, Alimirah F,Mehta RR, Kopelovich L: PPARgamma antagonist GW9662 inducesfunctional estrogen receptor in mouse mammary organ culture:potential translational significance. Mol Cell Biochem 2013, 372:249e256

ajp.amjpathol.org - The American Journal of Pathologyf � 22 April 2013 � 1:53 pm � EO: AJP13_0147