Interleukin-24 Induces Expression of B4 Integrin but Suppresses Anchorage-Independent Growth of Rat Mammarytumor Cells Bya Mechanism That Is Independent of B4

Published OnlineFirst March 3, 2009; DOI: 10.1158/1541-7786.MCR-08-0252

Interleukin-24 Induces Expression of B4 Integrin but Suppresses Anchorage-Independent Growth of Rat MammaryTumor Cells bya Mechanism That Is Independent of B4

Wanli Xuan,2 You-Jun Li,1,3 Guodong Liu,2 Yaacov Ben-David,1,3 and Michael C. Archer1,2

Departments of 1Medical Biophysics and 2Nutritional Sciences, University of Toronto; 3Molecular and Cell Biology Research, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada

Abstract Wistar-Furth rats develop multiple mammary tumors following Wistar-Furth rats develop multiple mammary a single injection of methylnitrosourea, whereas Copenhagen adenocarcinomas following initiation with rats develop few, if any, tumors (1). To identify genes methylnitrosourea, whereas Copenhagen rats are associated with resistance, we infused a retrovirus containing resistant to the development of mammarytumors. v-Ha-ras directly into the main mammary ducts of resistant F1 We have previouslyisolated cell lines from tumors rats from a Copenhagen Â Wistar-Furth cross (2). We reasoned Â induced in resistant Copenhagen Wistar-Furth F1 that retroviral insertional inactivation of a resistance gene or rats byinfusion of a retrovirus harboring v-Ha- ras activation of a susceptibility gene in only one allele of the F1s directlyinto the main mammaryducts. Some of the may be sufficient to confer susceptibility. Approximately 50% cell lines were able to grow in soft agar, but a significant of infused glands developed tumors that were typical of rat number did not displayanchorage-independent mammary adenocarcinomas described previously in detail by growth. Here, we compared bymicroarrayanalysis Russo and Russo (3). Cell lines derived from these tumors genes that are differentiallyexpressed in these cell lines. displayed epithelial-like morphology and were clonal although

The expression of interleukin-24 (IL-24) and b4 integrin there was no common viral integration site (2). This suggested was highlycorrelated with the inabilityof cells to that resistance in the F1 rats was overcome by a high level of grow in soft agar. Ectopic expression of IL-24 in v-Ha-ras expression. anchorage-independent cells inhibited their growth in Because it is well established that expression of v-Ha-ras in monolayer culture, in soft agar, and in nude mice in vivo epithelial cells can overcome anchorage-dependent growth by and inhibited their abilityto migrate and invade in in vitro inhibiting suspension-induced apoptosis (anoikis; ref. 4), we assays. Furthermore, growth suppression by IL-24 decided to investigate the growth properties of our cell lines. was associated with the transcriptional up-regulation To our surprise, whereas some of the cell lines were able to of p27Kip1 via the activation of Stat3. We showed, for the grow in soft agar, a significant number did not display first time, that b4 integrin is a downstream target of IL-24. anchorage-independent growth, but these properties did not correlate with the level of v-Ha-ras expression (2). The growth However, B4 does not playa direct role in regulating the proliferative capacityof rat mammarytumor cells. characteristics were also reproduced when selected clones were Our results show that IL-24 suppresses the growth injected into nude mice. We hypothesized that our anchorage- of rat mammarycarcinoma cells and mayplaya role dependent and anchorage-independent cell lines may recapit- in the resistance of Copenhagen rats to mammary ulate the resistance and susceptibility of Copenhagen and carcinogenesis. (Mol Cancer Res 2009;7(3):433–42) Wistar-Furth rats, respectively, to mammary carcinogenesis that could facilitate the identification of breast cancer susceptibility genes. Introduction Furthermore, the ability of clones to grow in soft agar Although many strains of rats develop multiple mammary showed a striking positive correlation with the expression of adenocarcinomas when initiated with carcinogens such as cyclin D1, a gene that has been shown previously to stimulate methylnitrosourea, several strains are resistant. For example, anchorage-independent growth of a human mammary epithelial cell line (5). Transfection of some clones with cyclin D1 fully Received 5/29/08; revised 10/13/08; accepted 11/11/08; published OnlineFirst overcame resistance to anchorage-independent growth, whereas 3/3/09. the effect was partial in other clones. Transfection of cells with Grant support: Canadian Breast Cancer Research Alliance and Canadian Breast b-catenin produced a similar pattern of cyclin D1 up-regulation Cancer Foundation. The costs of publication of this article were defrayed in part by the payment of and growth in soft agar. Somatic fusion of an anchorage- page charges. This article must therefore be hereby marked advertisement in dependent to an anchorage-independent clone produced a accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Note: W. Xuan and Y-J. Li contributed equally to this work. hybrid that did not grow in soft agar. Requests for reprints: Michael C. Archer, University of Toronto, 150 College These various results suggested that a high level of cyclin Street, Toronto, Ontario, Canada M5S 3E2. Phone: 416-978-8195; Fax: 416-971- D1 expression is necessary, but not sufficient, for full 2366. E-mail: [email protected] Copyright D 2009 American Association for Cancer Research. anchorage-independent growth capacity in the mammary tumor doi:10.1158/1541-7786.MCR-08-0252 cell lines. This raises the possibility that loss of a tumor

Mol Cancer Res 2009;7(3). March 2009 433 Downloaded from mcr.aacrjournals.org on September 27, 2021. © 2009 American Association for Cancer Research. Published OnlineFirst March 3, 2009; DOI: 10.1158/1541-7786.MCR-08-0252

434 Xuan et al.

Table 1. MicroarrayAnalysisof Differential Gene Expression in FE1.2 and FE1.3 Cells

Fold change t test P Common name Gene description

159.50 2.45E-04 IL-24 Interleukin 24 94.12 6.64E-05 na Keratin complex 1, acidic, gene 18 92.07 1.34E-06 Ptprf Protein tyrosine phosphatase, receptor type, F 70.07 2.66E-06 Itgb4 Integrin b4 65.60 1.29E-04 Q7CTE2 (Q7CTE2) AGR_L_2024p, partial (7%) [TC555084] 63.64 1.39E-04 Cotl1_predicted Similar to chain A, solution structure of coactosin-like protein (cofilin family) from Mus musculus 48.05 1.52E-04 Krt1-19 Keratin complex 1, acidic, gene 19 46.66 2.67E-06 msln Mesothelin 37.61 2.52E-04 Col18a1 Collagen, type XVIII, a1 31.75 3.67E-05 na Hypothetical protein XP_347094 30.98 7.90E-05 Cdkn2a Cyclin-dependent kinase inhibitor 2A (p16, inhibits CDK4) 22.66 5.39E-05 na Keratin complex 1, acidic, gene 18 14.10 1.16E-04 Tmc4_predicted Similar to Tmc4 protein 13.71 1.71E-05 RABIN3 RAB3A interacting protein 12.57 1.52E-04 BF556648 UI-R-E0-dm-a-09-0-UI.r1 UI-R-E0 Rattus norvegicus cDNA clone UI-R-E0-dm-a-09-0-UI 5¶, mRNA sequence [BF556648] 9.17 8.19E-05 Src BI395724 EST531595 rat gene index, cDNA clone RGICL93 5¶ end, mRNA sequence [BI395724] 8.59 1.30E-04 Dctn2 Dynactin 2 6.88 6.61E-06 Ptpn6 Protein tyrosine phosphatase, nonreceptor type 6 6.38 4.00E-05 Cdk4 Cyclin-dependent kinase 4 6.26 5.64E-05 Ifngr2_predicted Similar to Interferon-c receptor 2 6.23 4.93E-09 na Similar to CG18661-PA 6.11 1.35E-04 AI454457 UI-R-C2p-qo-c-06-0-UI.s1 UI-R-C2p R. norvegicus cDNA clone UI-R-C2p-qo-c-06-0-UI 3¶, mRNA sequence [AI454457] 5.98 4.05E-05 Dusp22_predicted Similar to dual-specificity phosphatase TS-DSP2 5.87 2.95E-05 Zc3hdc1_predicted Similar to zinc finger CCCH-type domain containing protein 1 5.66 2.30E-05 Similar to proline dehydrogenase; PRODH (LOC287950), mRNA [XM_213584] 5.52 8.04E-10 Cct2 Chaperonin containing TCP1, subunit 2 (b) 5.49 5.53E-05 Prdx5 Peroxiredoxin 5 precursor 5.43 1.89E-08 Eya2 cDNA clone RGIEZ51 5¶ end, mRNA sequence [AW918854] 5.38 2.44E-06 Cdk4 Cyclin-dependent kinase 4 5.30 2.68E-04 na Similar to RIKEN cDNA 1190005P17 5.27 1.74E-04 MGC108808 Similar to RIKEN cDNA 2700038C09 (predicted)

NOTE: The analysis was done for five replicates using 44K Agilent Whole Rat Genome Arrays. Only values with a fold change >5 and t test P < 0.0005 are reported. All of the genes in the table have higher expression levels in FE1.3 than FE1.2 cells. Only one gene under these conditions was overexpressed in FE1.2 cells (enpp1).

suppressor gene is required to fully confer anchorage- (6), was shown to suppress growth of cells in soft agar and independent growth. Here, we have compared by microarray in vivo in nude mice. Furthermore, IL-24 induced the expres- Kip1 analysis genes that are differentially expressed in anchorage- sion of b4 integrin and, through activation of Stat3, p27 . dependent and anchorage-independent cell lines. Of several genes that are highly overexpressed in anchorage-dependent cells, interleukin-24 (IL-24), a known tumor suppressor gene Results Identification of Genes Differentially Expressed in Breast Cancer Cell Lines We performed differential microarray analysis to identify genes responsible for anchorage-independent and anchorage- dependent growth of the cell lines that we established from the mammary carcinomas induced by injection of a retrovirus containing v-Ha-ras directly into the main mammary ducts of Â resistant F1 (Copenhagen Wistar-Furth)rats (2).Five independent RNA isolates from the anchorage-dependent cell line FE1.3 and anchorage-independent cell line FE1.2 were subjected to microarray analysis using 44K Whole Rat Genome arrays (Agilent Technologies). This analysis resulted in the identification of 31 genes that are up-regulated at least 4-fold with t test P < 0.0005 in FE1.3 compared with FE1.2 cells (Table 1). Interestingly, three genes were in the keratin family, confirming our previous observations by immunohistochemis- try that FE1.2 and subclones stained negatively or only weakly FIGURE 1. Expression of genes that are up-regulated in anchorage- for keratin, whereas FE1.3 cells stained positively (2). dependent cells compared with anchorage-independent cells that were Surprisingly, only one gene, enpp1, was significantly down- identified by microarray analysis. A. Northern blot analysis of IL-24, ptpn6, regulated (3-fold)in FE1.3 cells (Table 1). msln, and b integrin in FE1.3 and FE1.2 cells. B and C. Expression of 4 We next performed Northern blot or reverse transcription- IL-24 (by Northern blot) and b4 integrin (by RT-PCR) in several anchorage-independent (asterisk) and anchorage-dependent cell lines. PCR (RT-PCR)analyses of four candidate genes that were

Mol Cancer Res 2009;7(3). March 2009 Downloaded from mcr.aacrjournals.org on September 27, 2021. © 2009 American Association for Cancer Research. Published OnlineFirst March 3, 2009; DOI: 10.1158/1541-7786.MCR-08-0252

IL-24 Regulates Growth of Rat Mammary Tumor Cells 435

FIGURE 2. IL-24 expression inhibits growth of cells in soft agar but does not affect cyclin D1 expression. A. Colo- nies formed in soft agar from untreated FE1.2 cells (a) and colonies formed in soft agar fromcells treated with recombinant IL-24 (magnification, Â10; b). B. Northern blot of IL-24 expression in FE1.2 and FE1.3 cells and FE1.2 cells infected with retrovirus containing IL-24 in the sense (two isolates: FE1.2 + IL-24 S1 and S2) or antisense (FE1.2 + IL-24 AS) orientation. C. Number of colonies formed in soft agar by FE1.2 + IL-24 AS, FE1.2 + IL-24 S1, and FE1.2 + IL-24 S2 cells. Mean F SE. *, P < 0.05, compared with FE1.2 + IL-24 AS cells. D. Western blot of cyclin D1 expression in the cells described herein.

identified in the microarray analysis that have been shown to IL-24 and b4 Integrin Expression Are Correlated with have diverse effects on cancer development. mda-7, cloned by Growth in Soft Agar subtractive hybridization from terminally differentiated mela- To examine whether ptpn6, msln, IL-24,orb4 integrin noma cells, is a tumor suppressor gene that selectively induces plays a role in anchorage-dependent growth, we examined the growth suppression and apoptosis in human cancer cells but expression of these genes in several our previously character- not in normal cells (7, 8). mda-7 is now recognized as IL-24,a ized anchorage-dependent and anchorage-independent cancer b member of the IL-10 gene family (8). 4 integrin associates cell lines (2). The expression of both IL-24 and b4 integrin with a6 integrin to form a laminin/kalinin receptor that is a directly correlated with the inability of the cell lines to grow in component of the hemidesmosome (9, 10). There is evidence soft agar. As shown in Fig. 1B and C, the anchorage- that b4 integrin may function as a suppressor gene for dependent cell lines FE1.3, FB0.1, FB1.1, and FB2.1 express epithelial tumors (11). msln is a differentiation antigen that is high levels of IL-24 and b4 integrin, whereas a negligible overexpressed in several human cancers, including pancreatic level of expression of these genes was detected in the and ovarian cancers and mesotheliomas (12), as well as anchorage-independent cell lines FE1.2, FB2.4, and FA2.1. methylnitrosourea-induced mammary tumors in Lewis rats The expression of msln and ptpn6, however, did not correlate (13). ptpn6, also known as shp-1, is primarily expressed in with the ability of these cells to grow in soft agar (data not hematopoietic cells and has been suggested to be a candidate shown). tumor suppressor gene in lymphomas, leukemias, and breast In view of these results, we decided to investigate a possible and other cancers through suppression of Jak kinases (14, 15). role for IL-24 and b4 integrin in anchorage-dependent and Northern blot analysis confirmed the microarray data showing anchorage-independent growth of our rat mammary tumor cells that IL-24 and b4 integrin were highly expressed in FE1.3 lines, although we recognize that several other genes in Table 1 cells with negligible expression in FE1.2 cells (Fig. 1A). Our may also be involved in regulating cell growth and perhaps microarray analysis did not identify a6 integrin as a gene that carcinogenesis. was differentially expressed in FE1.2 and FE1.3 cells. We confirmed this by RT-PCR, which showed that both cell lines IL-24 Inhibits Growth in Soft Agar do express a6 integrin but at approximately the same level The virtually exclusive expression of IL-24 in our (data not shown). msln and ptpn6 were both clearly expressed anchorage-dependent cells suggested a role for this cytokine in FE1.3 cells, although there was some expression in FE1.2 in determining their growth characteristics. For this reason, and cells (Fig. 1A). because IL-24 has been shown previously to suppress the

436 Xuan et al.

anchorage-independent growth of DU-145 human prostate FE1.3 cells (data not shown). This indicates that more than one carcinoma cells (16), we decided to investigate in more detail gene may be responsible for anchorage-dependent/independent a possible role for IL-24 in regulating anoikis in our cell lines. growth. Expression of IL-24, however, did not induce apoptosis First, we cultured anchorage-independent FE1.2 cells in soft in FE1.2 cells (data not shown). agar in the presence or absence of recombinant IL-24. The addition of this protein to the culture medium of the FE1.2 cells IL-24 Inhibits Cell Migration and Invasion In vitro significantly suppressed the number and size of the colonies As illustrated in Fig. 3B, FE1.2 cells infected with the (Fig. 2A). FE1.3 cells express high levels of IL-24, and we retrovirus containing IL-24 in the sense orientation were showed that medium in which these cells had been cultured significantly less able to migrate through polycarbonate filters also suppressed the growth of FE1.2 cells in soft agar (data compared with the same cells expressing antisense IL-24 or the not shown). parental FE1.2 cells. Similarly, using polycarbonate filters To examine further the importance of IL-24 in the coated with Matrigel, the invasive ability of cells expressing suppression of growth in soft agar, FE1.2 cells were infected IL-24 was reduced by 80% to 90% compared with those not with a retrovirus (MSCV2.1)expressing this gene (Fig. 2B).As expressing the gene (Fig. 3C). shown in Fig. 2C, two clones isolated from FE1.2 cells that overexpressed IL-24 following infection with the retrovirus containing IL-24 in the sense orientation gave rise to IL-24 Expression Suppresses the Growth of Breast significantly fewer colonies than cells infected with the Cancer Cells In vivo retrovirus containing IL-24 in the antisense orientation. We Because our previous studies showed a correlation between showed previously that the ability of cells to grow in soft agar the ability of cells to grow in soft agar and to form tumors in was positively associated with the expression of cyclin D1 (2). nude mice (2), we next examined the growth in nude mice of Here, we showed that the expression of cyclin D1 was not FE1.2 cells infected with the retrovirus expressing IL-24 in the affected by the expression of IL-24 in FE1.2 cells (Fig. 2D). sense or antisense orientation. Figure 3D shows that tumor growth was significantly inhibited in cells expressing IL-24 compared with cells expressing antisense IL-24 or the parental IL-24 Expression Inhibits Growth but Does Not Induce cell line. Apoptosis in Monolayer Culture Retroviral expression of IL-24 in FE1.2 cells reduced the ability of these cells to proliferate in monolayer culture IL-24 Induces Expression of b4 Integrin compared with cells infected with the retrovirus containing The strong correlation we observed between expression of IL-24 in the antisense orientation (Fig. 3A). FE1.2 cells b4 integrin and IL-24 raised the possibility that these genes are expressing IL-24 also became more rounded, although they did coregulated in our cell lines (Fig. 1A). Indeed, when FE1.2 cells not become morphologically identical to the epithelial-like that expressed no b4 integrin (Fig. 1A)were infected with the

FIGURE 3. IL-24 expression inhibits growth of FE1-2 cells in monolayer culture, inhibits cell migration and invasion in vitro, and inhibits growth in nude mice. A. Growth of cells in monolayer culture. B. Migration of cells through polycarbonate filters. C. Invasion of cells through polycarbonate filters coated with Matrigel. D. Growth of cells in nude mice. Tumors were weighed 2 weeks after injection of cells. Cell lines are described in Fig. 2. Mean F SE. *, P < 0.01, compared with FE1.2 or FE1.2 + IL-24 AS cells.

IL-24 Regulates Growth of Rat Mammary Tumor Cells 437

FIGURE 4. Up-regulation of IL-24 induces expression of b4 integrin in FE1.2 cells, whereas down-regulation of IL-24 inhibits expression of b4 integrin in FE1.2 cells and stimulates their growth in soft agar. A. RT-PCR of b4 integrin in FE1.2 and FE1.3 cells and FE1.2 cells infected with retrovirus containing IL-24 in the sense (four isolates: FE1.2 + IL-24 S1-S4) or antisense (FE1.2 + IL-24 AS) orientation. B. Western analysis of b4 integrin in indicated cells. C. Western analysis of IL-24 in FE1.3 cells, two isolates of FE1.3 cells infected with retrovirus expressing antisense IL-24 (FE1.3 + IL-24 AS1 and AS2), or FE1.3 cells infected with empty retrovirus (vector). D. Western analysis of b4 integrin in indicated cells. E. Growth in soft agar of cells described herein. Mean F SE. *, P < 0.05, compared with FE1.2 cells.

retrovirus expressing IL-24, b4 integrin was significantly up- higher in the anchorage-dependent cell lines compared with regulated as determined by RT-PCR (Fig. 4A). The expression those that grew in soft agar (Fig. 5A). To examine whether up- Kip1 of a6 integrin, however, was not different in FE1.2 cells regulation of p27 is induced by IL-24, we examined the infected with IL-24 in the sense or antisense orientations (data status of this cell cycle inhibitor in the anchorage-independent not shown). Up-regulation of b4 integrin by IL-24 was cell line FE1-2 infected with the retrovirus expressing IL-24.It reproduced in three different isolates of these cells is clear that two of the clones overexpressing IL-24 in the sense (FE1.2+IL-24 S1-4)and was confirmed by Western analysis orientation expressed higher levels of p27Kip1 than control cells (Fig. 4B). As shown previously, growth of FE1.2 cells infected with antisense IL-24 (Fig. 5B). RT-PCR analysis expressing the IL-24 construct was significantly inhibited showed that up-regulation of p27Kip1 by IL-24 occurred at the (Fig. 3). In contrast, when IL-24 was down-regulated in FE1.3 level of transcription (Fig. 5C). Furthermore, infection of cells using the antisense construct (two different isolates shown anchorage-dependent FE1-3 cells with antisense IL-24 resulted Kip1 in Fig. 4C), expression of h4 integrin was significantly inhibited in down-regulation p27 compared with control cells (Fig. 4D). These cells acquired the ability to grow in soft agar infected with the vector alone (Fig. 5D). (Fig. 4E), although growth was not as robust as FE1.2 cells. There are several reports that IL-24 activates Stat3 in a variety of cell types including breast cancer cells (6, 17-20). Because activation of Stat3 has been shown to induce p27Kip1 IL-24 Induces p27Kip1 Transcriptional Up-Regulation expression, we investigated Stat3 activation at both Tyr705 and through Activation of Stat3 Ser727 in our cell lines (21, 22). As shown in Fig. 6A to C, Stat3 Inhibition of growth by IL-24 in our breast cancer cell lines was phosphorylated at Tyr705 and Ser727 in FE1.3 but not FE1.2 suggested an alteration in the cell cycle machinery. Because a cells. However, Tyr705 and Ser727 were phosphorylated in recent report showed that treatment of human breast cancer FE1.2 cells infected with IL-24 in the sense orientation in cells with IL-24 led to the up-regulation of cell cycle inhibitor contrast to those infected with antisense IL-24 (Fig. 6A-C). p27Kip1 (17), we investigated the expression of this protein in To examine whether inhibition of Stat3 phosphorylation our cell lines. First, we showed that the levels of p27Kip1 were blocks p27Kip1 up-regulation, specific inhibitors of Stat3

438 Xuan et al.

tyrosine kinase (AG9 and AG490; refs. 23, 24)were added to ectopic expression of IL-24 in our anchorage-independent the medium of FE1.2 cells infected with IL-24.Both FE1.2 cells inhibited their growth in soft agar. These results are compounds completely inhibited Stat3 phosphorylation expres- in keeping with those of Su et al. who recently showed that sion and down-regulated p27Kip1 (Fig. 6A and B). Incubation immortalized normal human cells infected with a recombinant of the same cells with the mitogen-activated protein kinase adenovirus expressing IL-24 (Ad.mda-7)secreted IL-24 that phosphorylation inhibitor PD98059 that blocks phosphorylation suppressed the anchorage-independent growth of DU-145 of Ser727 in Stat3 (25)also resulted in a down-regulation of human prostate carcinoma cells (16). We have shown pre- p27Kip1 (Fig. 6C). Therefore, phosphorylation of both serine viously that a high expression level of cyclin D1 is necessary, and tyrosine sites in Stat3 is required for p27Kip1 regulation by although not sufficient, for anchorage-independent growth of IL-24. Interestingly, inhibition of Stat3 tyrosine and serine our breast cancer cell lines (2). Here, we show that expression kinase with AG9 had no effect on h4 integrin expression of cyclin D1 was not affected by the expression of IL-24, (Fig. 6D), suggesting that induction of h4 expression by IL-24 suggesting that suppression of the growth of FE1.2 cells in soft is independent of Stat3 up-regulation. agar by IL-24 is independent of cyclin D1. We have also shown that IL-24 expression in FE1.2 cells inhibited their growth in monolayer culture. Others have shown Discussion that ectopic expression of a transfected IL-24 gene induces We have shown by microarray analysis and confirmed by growth suppression in cell lines derived from a variety of Northern and RT-PCR analyses that IL-24 is highly overex- human tumors (ref. 17 and references therein). Although IL-24 pressed in rat mammary tumor cells that are unable to grow in expression inhibited growth of FE1.2 cells, it did not induce soft agar compared with cells that are anchorage independent. apoptosis. These cells clearly behave differently in this respect The differential expression of IL-24 in FE1.2 1 and FE1.3 cells from the human tumors in which IL-24 expression induced was somewhat surprising because both cell lines overexpress apoptosis (ref. 17 and references therein), although a recent v-Ha-ras (2)and expression of IL-24 has been shown to be report shows that recombinant IL-24 lacks apoptosis-inducing induced in Rat1 cells and rat intestinal epithelial cells by properties in human melanoma cells (27). Ramesh et al. (28) oncogenic Ha-ras (26). Our results clearly show that IL-24 have shown previously that ectopic production of IL-24 inhibits expression in mammary tumor cells is regulated differently. In invasion and migration of human lung cancer cells. Likewise, strong support of our microarray results, we have shown that we have shown that expression of IL-24 in FE1.2 cells inhibited

FIGURE 5. IL-24 up- regulates p27Kip1 through activation of Stat3. A. Western analysis of p27Kip1, Stat3, and Stat3 phosphorylation in several anchorage-dependent and anchorage-independent (asterisk) cell lines. B. Western analysis of p27Kip1,Stat3, and Stat3 phosphorylation in FE1.2 and FE1.3 cells and FE1.2 cells infected with retrovirus expressing IL-24 in the sense (two isolates: FE1.2 + IL-24 S1 and S2) or antisense (FE1.2 + IL-24 AS) orientation. C. RT-PCR of p27Kip1 in indicated cells. D. Western analysis of p27Kip1 in FE1.3 cells and cells infected with retrovirus expressing antisense IL-24 (two isolates: FE1.3 + IL-24 AS1 and AS2) and FE1.3 cells infected with empty retrovirus (vector).

IL-24 Regulates Growth of Rat Mammary Tumor Cells 439

FIGURE 6. Stat3 phosphorylation at Tyr205 and Ser727 in anchorage-dependent and anchorage-independent cell lines and effects on b4 integrin expression. A and B. Western analysis of effects of inhibitors of Stat3 tyrosine kinase (AG9 and AG490) on Stat3 phosphorylation and p27Kip1 levels in FE1.2 and FE1.3 cells and two isolates of FE1.2 cells infected with retrovirus expressing IL-24 in the sense orientation (FE1.2 + IL-24 S1 and S2). C. Western analysis of effect of inhibiting Stat3 serine kinase by mitogen- activated protein kinase phosphorylation inhibitor PD98059 on Stat3, Stat3 phosphorylation, and p27Kip1 levels in cells described herein. D. Western analysis of effect of Stat3 tyrosine kinase inhibitor AG9 on b4 integrin expression in cells indicated as well as FE1.2 cells infected with retrovirus expressing antisense IL-24 (FE1.2 + IL-24 AS).

both migration and invasion in in vitro assays, suggesting that independent. Although there is evidence that the h4 and a6 in vivo expression of this gene will suppress malignancy. integrin subunits can act as tumor suppressors (11, 30, 31), there Consistent with the action of IL-24 to inhibit growth of FE1.2 is also evidence that h4 enhances anchorage-independent growth cells in soft agar and in monolayer culture, the ability of FE1.2 of human breast carcinoma cells and stimulates tumor growth in cells to form tumors in nude mice in vivo was significantly nude mice (ref. 32 and references therein). Our results, therefore, inhibited in cells expressing IL-24 compared with those stand in contrast to these studies. In our rat mammary carcinoma expressing antisense IL-24 or the parental cells. These results cells, it is possible that the growth-inhibitory actions of the highly confirm those reported by Su et al. (29)who showed that overexpressed IL-24 override the growth-enhancing actions of infection of human breast cancer cells with Ad.mda-7 before h4. Furthermore, although h4 was overexpressed in FE1.3 cells, injection into nude mice inhibited tumor development. Overall, the expression of a6 was similar in the anchorage-dependent and our results show that the effects of IL-24 expression in rat anchorage-independent cells and was not induced by IL-24. It is mammary tumor cells are similar in most respects to its effects possible, therefore, that the growth-enhancing activity of the in human cancer cells. overexpressed h4 was limited by the expression of a6. In addition to IL-24, our microarray analysis also revealed IL-24-induced inhibition of cell proliferation was associated that b4 integrin is highly overexpressed in the cells that are with the transcriptional up-regulation of the cell cycle inhibitor unable to grow in soft agar compared with the anchorage- p27Kip1 mediated by Stat3 activation. Thus, our results confirm independent cells. To our surprise, ectopic expression of IL-24 in the association among p27Kip1, Stat3, and IL-24 that was shown FE1.2 cells led to a highly significant up-regulation of h4 recently in two human breast cancer cell lines (17), and we expression. Furthermore, infection of FE1.3 cells with the additionally show that up-regulation of both serine and tyrosine Kip1 antisense IL-24 construct led to a down-regulation of h4 phosphorylation of Stat3 by IL-24 is required for p27 expression and growth of the cells became partially anchorage- up-regulation. The inhibitory effect of p27Kip1 on tumor

440 Xuan et al.

proliferation is mainly through its interaction with the CDK- Microarray Analysis cyclin D1 complex (33). Indeed, we have shown previously FE1.2 and FE1.3 cells were grown in 10 cm dishes until lower levels of cyclin D1 expression in anchorage-dependent f70% confluent. Total RNA was isolated using TRIzol versus anchorage-independent cell lines despite expressing according to the manufacturer’s instructions (Invitrogen). A similar levels of Ha-ras (2)Our microarray analysis showed Low RNA Input Fluorescent Linear Amplification Kit (Agilent higher levels of Cdk4 expression in anchorage-dependent than Technologies)was used to amplify 200 ng RNA and label the in anchorage-independent cells, although cyclin D1 expression FE1.3 cell samples with Cy5 and the FE1.2 cell samples with was not affected by IL-24 expression. Interestingly, Cdkn2a Cy3. The cRNA was then fragmented and hybridized using a (p16INK4A)was highly up-regulated in the anchorage-dependent Gene Expression Hybridization Kit (Agilent Technologies). cells (Table 1), suggesting that stimulation of cell growth Microarray analyses were done at the University Health caused by Cdk4 up-regulation would have been inhibited. We Network (Toronto)Microarray Centre. Five 44K Whole Rat have recently observed overexpression of Cdk2 in the Genome arrays (Agilent Technologies)were hybridized with anchorage-dependent cells.4 Thus, inhibition of proliferation five different isolates of each of the two cell lines. The arrays by IL-24 may also be a result of increased amounts of the were hybridized for 17 h at 60jC according to the manufac- CDK2/p27Kip1 complex leading to inhibition of Cdk2 activity. turer’s recommendations. The arrays were then washed for Indeed, a recent study showed that oncostatin M, an IL-6 family 1 min in 6Â SSPE, 0.005% N-lauroylsarcosine and 1 min in cytokine, inhibits the growth of anchorage-dependent skeletal 0.06Â SSPE, 0.005% N-lauroylsarcosine. A final wash step was muscle cells through down-regulation of cyclin D1 and done for 30 s in Stabilization and Drying Solution (Agilent). The inhibition of CDK2/p27Kip1 (34). p27Kip1 likely functions as a arrays were then scanned in the Agilent G2565AA Microarray more potent inhibitor of cell proliferation in anchorage- scanner and quantified using Agilent Feature Extraction (version dependent cell lines. Because cyclin D1 expression in FE1.3 8.1). GeneSpring (version 7.5) was used to analyze the array cells is regulated through the h-catenin pathway (2), cooper- data using a lower signal threshold of 100. Significant ation between this cyclin and IL-24/Stat3 may be necessary to differences between the two cell lines were determined by both confer anchorage-dependent growth. signal fold differences of >2 and t test P values < 0.05. It has been shown that IL-24 exerts its effects on cell growth in cancer cells and not in normal cells (8). Our observations Infection and Transfection Experiments raise the possibility, however, that the expression of IL-24 is An IL-24-producing retrovirus was generated by inserting activated in the preneoplastic lesions of Copenhagen rats the entire 600-bp coding sequence of rat IL-24 in sense and leading to their failure to progress to cancer as we have antisense orientations into a unique EcoRI site in the retroviral described (35, 36). This notion is currently under investigation. expression vector MSCV2.1 (37). Replication-defective viruses In summary, using microarray analysis, we identified several were prepared by transfecting the viral plasmid into the helper- genes that are highly expressed in anchorage-dependent rat free packaging cell line GP+A (38)as described previously mammary tumor cells but not in anchorage-independent cells. (37). For viral infection, supernatants from the virus-producing One of these genes, IL-24, was shown to suppress proliferation, cells were used to infect FE1.2 cells plated at a density of anchorage-independent growth, migration, and invasion in vitro 2 Â 106. After 48 h G418 selection was started and after 10 d, and growth of cells in vivo. Furthermore, growth suppression Kip1 G418-resistant cells were pooled and subjected to subcloning by IL-24 was associated with up-regulation of p27 mediated by limiting dilution. Transfection was done using Lipofectamine by Stat3. These cellular and molecular effects of IL-24 in rat 2000 according to the manufacturer’s protocol (Invitrogen). mammary tumor cells are similar to its effects in human cancer cells. h4 Integrin was also highly expressed in the anchorage- dependent rat mammary tumor cells but not in the cells that Growth of Cells in Soft Agar To monitor growth of cells in soft agar, two layers of agarose grew in soft agar. We showed, for the first time, that h4 integrin is a downstream target of IL-24. However, h does not play a were used. The bottom layer contained 0.5% agarose and the 4 a direct role in regulating the proliferative capacity of the rat top layer contained 0.3% agarose both in -MEM. Triplicate mammary tumor cells. samples of 5,000 cells were seeded on 6 cm plates and incubated for 5 to 7 days. At the end of the incubation period, colonies with >25 cells were counted. FE1.2 cells were also grown in soft agar in the presence of 5 nmol/L IL-24-alkaline Materials and Methods phosphatase fusion protein (GenHunter). Cell Culture The establishment of the rat breast cancer cell lines was described previously (2). All cell lines were maintained in a- Growth of Cells in Nude Mice MEM supplemented with 10% fetal bovine serum, 10 ng/mL Four groups of six female athymic nude mice (CD1 nu/nu), epidermal growth factor, 1 Ag/mL hydrocortisone, and 1 Ag/mL ages 7 weeks, were purchased from Charles River Laboratories. 6 17h-estradiol. To examine growth rates, 104 cells were seeded At age 8 weeks, 10 FE1.2 or FE1.2 cells expressing IL-24 in per well in triplicate on 24-well plates and counted on each of the sense or antisense orientation, in a volume of 50 AL, were 4 consecutive days. injected through a 27-gauge needle into the right and left thoracic fat pads. Tumor size was measured twice weekly with vernier calipers by a single blinded observer and tumors were 4 Unpublished results. harvested 2 weeks after the mice were injected with the cells.

IL-24 Regulates Growth of Rat Mammary Tumor Cells 441

Cell Migration and Invasion Assays 3. Russo J, Russo IH. Atlas and histologic classification of tumors of the rat mammary gland. J Mammary Gland Biol Neoplasia 2000;5:187 – 200. Cell migration and invasion assays were done as described 4. Frisch SM, Francis H. Disruption of epithelial cell-matrix interactions induces by Chen and Thompson (39). apoptosis. J Cell Biol 1994;124:619 – 26. 5. Zhou Q, Wulfkuhle J, Ouatas T, et al. Cyclin D1 overexpression in a model of Western Blotting human breast premalignancy: preferential stimulation of anchorage-independent but not anchorage-dependent growth is associated with increased cdk2 activity. Western blotting was done as we have described previously Breast Cancer Res Treat 2000;59:27 – 39. (2). Primary murine monoclonal antibodies were obtained 6. Chada S, Sutton RB, Ekmekcioglu S, et al. MDA-7/IL-24 is a unique from the following sources: cyclin D1 from Neomarkers used cytokine-tumor suppressor in the IL-10 family. Int Immunopharmacol 2004;4: at a dilution of 1:500, h-actin from Sigma used at a dilution 649 – 67. 7. Sieger KA, Mhashilkar AM, Stewart A, et al. The tumor suppressor activity of of 1:50,000, Stat3 and phospho-Stat3 from Cell Signaling MDA-7/IL-24 is mediated by intracellular protein expression in NSCLC cells. used at a dilution of 1:500, mitogen-activated protein kinase Mol Ther 2004;9:355 – 67. and phospho-mitogen-activated protein kinase also from 8. Gupta P, Su ZZ, Lebedeva IV, et al. mda-7/IL-24: multifunctional cancer- Cell Signaling used at a dilution of 1:1,000, p27 from Santa specific apoptosis-inducing cytokine. Pharmacol Ther 2006;111:596 – 628. A B Cruz Biotechnology used at a dilution of 1:1,000, and h 9. Lee EC, Lotz MM, Steele GD, Jr., Mercurio AM. The integrin 6 4 is a 4 laminin receptor. J Cell Biol 1992;117:671 – 8. integrin from Chemicon International used at a dilution of 10. Niessen CM, Hogervorst F, Jaspars LH, et al. The A6B4 integrin is a receptor 1:1,000. for both laminin and kalinin. Exp Cell Res 1994;211:360 – 7.

Stat3 tyrosine kinase inhibitors AG9 and AG490 (Calbio- 11. Kim SY, Bachman NJ, Nair TS, et al. B4 Integrin transfection of UM-UC-2 chem)dissolved in DMSO were added to the cells to yield final (human bladder carcinoma)cells: stable expression of a spontaneous cytoplasmic truncation mutant with rapid loss of clones expressing intact B4. Cancer Res 1997; concentrations of 200 and 20 nmol/mL, respectively. Mitogen- 57:38 – 42. activated protein kinase phosphorylation inhibitor PD 098059 12. Hassan R, Bera T, Pastan I. Mesothelin: a new target for immunotherapy. (Sigma)dissolved in DMSO was added to cells at final Clin Cancer Res 2004;10:3937 – 42. concentration of 100 nmol/mL. Control cells were treated with 13. Uehara N, Unami A, Kiyozuka Y, Shikata N, Oishi Y, Tsubura A. Parous the same volumes of DMSO alone. Cells were harvested for mammary glands exhibit distinct alterations in gene expression and proliferation responsiveness to carcinogenic stimuli in Lewis rats. Oncol Rep 2006;15: Western blot analysis after 1 h of treatment. 903 – 11. 14. Wu C, Guan Q, Wang Y, Zhao ZJ, Zhou GW. SHP-1 suppresses cancer cell Northern Blotting and RT-PCR growth by promoting degradation of JAK kinases. J Cell Biochem 2003;90: 1026 – 37. Total RNA was prepared by using TRIzol. Electrophoresis 15. Wu C, Sun M, Liu L, Zhou GW. The function of the protein tyrosine on 1% formaldehyde denatured agarose gels (20 Ag RNA/lane) phosphatase SHP-1 in cancer. Gene 2003;306:1 – 12. was followed by transfer to Zeta-Probe Nylon membranes 16. Su Z, Emdad L, Sauane M, et al. Unique aspects of mda-7/IL-24 antitumor (Bio-Rad). The membranes were hybridized with 32P-labeled bystander activity: establishing a role for secretion of MDA-7/IL-24 protein by IL-24, msln, ptpn6,orcyclin D1 cDNA probes at 42jC over- normal cells. Oncogene 2005;24:7552 – 66. night. Hybridization buffer and wash procedures were as 17. Zheng M, Bocangel D, Doneske B, et al. Human interleukin 24 (MDA-7/ IL-24)protein kills breast cancer cells via the IL-20 receptor and is antagonized described previously (40). by IL-10. Cancer Immunol Immunother 2007;56:205 – 15. 18. Chada S, Bocangel D, Ramesh R, et al. mda-7/IL24 kills pancreatic cancer RT-PCR Analysis cells by inhibition of the Wnt/PI3K signaling pathways: identification of IL-20 receptor-mediated bystander activity against pancreatic cancer. Mol Ther 2005;11: The total RNA was extracted by TRIzol and treated with 724 – 33. DNase I. cDNA was created using the SuperScript II First- 19. Dumoutier L, Leemans C, Lejeune D, Kotenko SV, Renauld JC. Cutting Strand Synthesis System for RT-PCR (Invitrogen). PCR edge: STAT activation by IL-19, IL-20 and mda-7 through IL-20 receptor primers were as follows: p27 forward primer 5¶-AGGGCCAA- complexes of two types. J Immunol 2001;167:3545 – 9. ¶ ¶ 20. Chada S, Mhashilkar AM, Ramesh R, et al. Bystander activity of Ad-mda7: CAGAACAGAAGA-3 and reverse primer 5 -CCA- human MDA-7 protein kills melanoma cells via an IL-20 receptor-dependent but GATGGGGTGTCAGTTTT-3¶, GAPDH forward primer 5¶- STAT3-independent mechanism. Mol Ther 2004;10:1085 – 95. AACTTTGGCATTGTGGAAGG-3¶ and reverse primer 5¶- 21. de Koning JP, Soede-Bobok AA, Ward AC, et al. STAT3-mediated GGATGCAGGGATGATGTTCT-3¶, and b integrin forward differentiation and survival and of myeloid cells in response to granulocyte 4 colony-stimulating factor: role for the cyclin-dependent kinase inhibitor primer 5¶-GCACAGACGAGCTGTTCAAG-3¶ and reverse p27(Kip1). Oncogene 2000;19:3290 – 8. primer 5¶-TCGATAGAAAGCCTCCTCCA-3¶. 22. Shi X, Zhang H, Paddon H, Lee G, Cao X, Pelech S. Phosphorylation of STAT3 serine-727 by cyclin-dependent kinase 1 is critical for nocodazole-induced mitotic arrest. Biochemistry 2006;45:5857 – 67. Disclosure of Potential Conflicts of Interest 23. Bright JJ, Du C, Sriram S. Tyrphostin B42 inhibits IL-12-induced tyrosine No potential conflicts of interest were disclosed. phosphorylation and activation of Janus kinase-2 and prevents experimental allergic encephalomyelitis. J Immunol 1999;162:6255 – 62. Acknowledgments 24. Meydan N, Grunberger T, Dadi H, et al. Inhibition of acute lymphoblastic We thank Dr. Jianmin Chen for help with the cell migration and invasion assays. leukaemia by a Jak-2 inhibitor. Nature 1996;379:645 – 8. 25. Lim CP, Cao X. Serine phosphorylation and negative regulation of Stat3 by JNK. J Biol Chem 1999;274:31055 – 61. References 1. Wood GA, Korkola JE, Archer MC. Tissue-specific resistance to cancer 26. Zhang R, Tan Z, Liang P. Identification of a novel ligand-receptor pair development in the rat: phenotypes of tumor-modifier genes. Carcinogenesis constitutively activated by ras oncogenes. J Biol Chem 2000;275:24436 – 43. 2002;23:1 – 9. 27. Kreis S, Philippidou D, Margue C, et al. Recombinant interleukin-24 lacks 2. Li YJ, Song R, Korkola JE, Archer MC, Ben-David Y. Cyclin D1 is necessary apoptosis-inducing properties in melanoma cells. PLoS ONE 2007;2:e1300. but not sufficient for anchorage-independent growth of rat mammary tumor cells 28. Ramesh R, Ito I, Gopalan B, Saito Y, Mhashilkar AM, Chada S. Ectopic and is associated with resistance of the Copenhagen rat to mammary production of MDA-7/IL-24 inhibits invasion and migration of human lung carcinogenesis. Oncogene 2003;22:3452 – 62. cancer cells. Mol Ther 2004;9:510 – 8.

442 Xuan et al.

29. Su ZZ, Madireddi MT, Lin JJ, et al. The cancer growth suppressor gene 35. Korkola JE, Archer MC. Resistance to mammary tumorigenesis in mda-7 selectively induces apoptosis in human breast cancer cells and inhibits Copenhagen rats is associated with the loss of preneoplastic lesions. tumor growth in nude mice. Proc Natl Acad Sci U S A 1998;95:14400 – 5. Carcinogenesis 1999;20:221 – 7. 30. Clarke AS, Lotz MM, Chao C, Mercurio AM. Activation of the p21 pathway 36. Korkola JE, Wood GA, Archer MC. Cyclin D1 expression during rat of growth arrest and apoptosis by the B4 integrin cytoplasmic domain. J Biol mammary tumor development and its potential role in the resistance of the Chem 1995;270:22673 – 6. Copenhagen rat. Breast Cancer Res 1999;1:88 – 94. 31. Sager R, Anisowicz A, Neveu M, Liang P, Sotiropoulou G. Identification by 37. Bani MR, Rak J, Adachi D, et al. Multiple features of advanced melanoma differential display of A6 integrin as a candidate tumor suppressor gene. FASEB J recapitulated in tumorigenic variants of early stage (radial growth phase)human 1993;7:964 – 70. melanoma cell lines: evidence for a dominant phenotype. Cancer Res 1996;56: 3075 – 86. 32. Bertotti A, Comoglio PM, Trusolino L. B4 Integrin activates a Shp2-Src signaling pathway that sustains HGF-induced anchorage-independent growth. 38. Markowitz D, Goff S, Bank A. A safe packaging line for gene transfer: J Cell Biol 2006;175:993 – 1003. separating viral genes on two different plasmids. J Virol 1988;62:1120 – 4. 33. Chu IM, Hengst L, Slingerland JM. The Cdk inhibitor p27 in human cancer: 39. Chen J, Thompson LU. Lignans and tamoxifen, alone or in combination, prognostic potential and relevance to anticancer therapy. Nat Rev Cancer 2008;8: reduce human breast cancer cell adhesion, invasion and migration in vitro. Breast 253 – 67. Cancer Res Treat 2003;80:163 – 70. 34. Kim H, Jo C, Jang BG, Oh U, Jo SA. Oncostatin M induces growth arrest of 40. Chu W, Pak BJ, Bani MR, et al. Tyrosinase-related protein 2 as a mediator of skeletal muscle cells in G1 phase by regulating cyclin D1 protein level. Cell melanoma specific resistance to cis-diamminedichloroplatinum(II): therapeutic Signal 2008;20:120 – 9. implications. Oncogene 2000;19:395 – 402.

Interleukin-24 Induces Expression of β4 Integrin but Suppresses Anchorage-Independent Growth of Rat Mammary Tumor Cells by a Mechanism That Is Independent of β4

Wanli Xuan, You-Jun Li, Guodong Liu, et al.

Mol Cancer Res 2009;7:433-442. Published OnlineFirst March 3, 2009.

Updated version Access the most recent version of this article at: doi:10.1158/1541-7786.MCR-08-0252

Cited articles This article cites 40 articles, 13 of which you can access for free at: http://mcr.aacrjournals.org/content/7/3/433.full#ref-list-1

Citing articles This article has been cited by 1 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/7/3/433.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mcr.aacrjournals.org/content/7/3/433. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.