MDA-7/IL-24 functions as a tumor suppressor gene in vivo in transgenic mouse models of breast cancer.

Melanoma differentiation associated gene-7/Interleukin-24 (MDA-7/IL-24) is a novel member of the IL-10 gene family that selectively induces apoptosis and toxic autophagy in a broad spectrum of human cancers, including breast cancer, without harming normal cells or tissues. The ability to investigate the critical events underlying cancer initiation and progression, as well as the capacity to test the efficacy of novel therapeutics, has been significantly advanced by the development of genetically engineered mice (GEMs) that accurately recapitulate specific human cancers. We utilized three transgenic mouse models to better comprehend the in vivo role of MDA-7/IL-24 in breast cancer. Using the MMTV-PyMT spontaneous mammary tumor model, we confirmed that exogenously introducing MDA-7/IL-24 using a Cancer Terminator Virus caused a reduction in tumor burden and also produced an antitumor “bystander” effect. Next we performed xenograft studies in a newly created MMTV-MDA-7 transgenic model that over-expresses MDA-7/IL-24 in the mammary glands during pregnancy and lactation, and found that MDA-7/IL-24 overexpression delayed tumor growth following orthotopic injection of a murine PDX tumor cell line (mPDX) derived from a tumor formed in an MMTV-PyMT mouse. We also crossed the MMTV-MDA-7 line to MMTV-Erbb2 transgenic mice and found that MDA-7/IL-24 overexpression delayed the onset of mammary tumor development in this model of spontaneous mammary tumorigenesis as well. Finally, we assessed the role of MDA-7/IL-24 in immune regulation, which can potentially contribute to tumor suppression in vivo. Our findings provide further direct in vivo evidence for the role of MDA-7/IL-24 in tumor suppression in breast cancer in immune-competent transgenic mice.


INTRODUCTION
Breast cancer remains one of the most commonly diagnosed cancers among women and is the second leading cause of cancer-related death in the United States [American Cancer Society, Cancer Facts & Figures, 2015]. Increased awareness about breast cancer has aided early diagnosis of the disease in recent years; however patient prognosis is adversely affected once breast cancer metastasizes to distant areas in the body. Hence, there is an urgent need to delineate the mechanisms regulating breast cancer development, progression and metastasis in order to develop novel targeted therapeutics with enhanced potency. While in vitro tissue culture studies and in vivo xenograft studies in immunocompromised mice have expanded our understanding of breast cancer, transgenic mouse models harboring intact functional immune systems provide additional advantages and are indispensible tools for studying breast cancer [1][2]. By utilizing mammary-specific promoters that drive tumorpromoting genes, researchers have developed transgenic mouse lines that spontaneously develop mammary tumors [3][4]. Knockout and conditional knockout lines have similarly been developed to better understand the role of tumor suppressor genes in the mammary gland [2]. In these contexts, genetically modified mice have helped identify genes involved in tumor initiation, progression and metastasis and provide valuable tools to assess novel therapeutics for human breast cancer.
To evaluate the role of MDA-7/IL-24 in breast cancer in immune-competent mice and to gain a further understanding of the mechanism of action of MDA-7/ IL-24 in breast cancer, we performed in vivo experiments using three transgenic models -MMTV-PyMT mice, MMTV-MDA-7 mice and MMTV-MDA-7/MMTV-Erbb2 mice. Our results illustrate that MDA-7/IL-24 delayed tumor onset, suppressed tumor growth and also had antitumor "bystander" effects. Further, we also found that MDA-7/IL-24 mounted an antitumor immune response by increasing levels of infiltrating CD8 + T cells and the frequency of IFN-γ or granzyme B-producing CD8 + T cells in the mammary tumors. Accordingly, our findings confirm that MDA-7/IL-24 is a relevant therapeutic option even in immune competent mice, both in xenograft models and spontaneous tumor models, and can synergize with the immune system to directly target tumor cells for destruction.

MDA-7/IL-24 reduces tumor growth in MMTV-PyMT transgenic mice
To evaluate the relevance of MDA-7/IL-24 in suppression of tumor growth in immune competent mice, we initially utilized the MMTV-PyMT transgenic mouse model [4]. Female MMTV-PyMT transgenic mice develop tumors in all mammary glands within 2 to 3 months of age [4]. In this model, we used a tumor-specific conditionally replicating virus expressing MDA-7/IL-24 (designated a cancer terminator virus or CTV) [31]. A type 5 adenovirus (Ad5) was engineered to specifically replicate in tumor cells by expressing Ad5-E1A under the control of a cancerspecific promoter derived from progression elevated gene-3 [32][33] and also produces MDA-7/IL-24 (Ad5-CTV) [20,[23][24][29][30]. As a control, we included Ad5-E1A that also replicates in tumor cells, but lacks MDA-7/IL-24. Untreated mice served as an additional control to evaluate tumor development and progression in MMTV-PyMT mice. The mice were monitored for tumor onset and tumors were injected with the respective virus as described in the Materials and Methods once palpable tumors were observed. Mice that were injected intratumorally with Ad5-CTV showed a significant reduction in tumor burden as compared to mice treated intratumorally with Ad5-E1A or control untreated mice ( Figure 1A). Additionally, the total tumor burden at harvest was significantly reduced in mice treated with Ad5-CTV as compared to the tumor burden in mice treated with Ad5-E1A or untreated mice (Figure   Oncotarget  36930 www.impactjournals.com/oncotarget 1B to 1D). In order to evaluate both the tumor suppressive effects of MDA-7/IL-24 as well as the "bystander" antitumor effect of MDA-7/IL-24, approximately 50% of the tumors that formed were injected with the respective virus, while the remaining 50% were left uninjected. Interestingly, the tumors that were left untreated on mice with 50% tumors treated with Ad5-CTV also showed a reduction in tumor growth, indicating a "bystander" antitumor effect ( Figure 1E). Immunohistochemical staining showed that MDA-7/IL-24 was expressed in the Ad5-CTV injected tumors ( Figure 2) as well as the untreated tumors in the same mouse (Supplemental Figure 1A). As would be expected, MDA-7/IL-24 expression was higher in the injected tumors as compared to the uninjected tumors in the same mouse (Supplemental Figure 1A).

Tumor growth is reduced in MMTV-MDA-7 transgenic mice
Next we developed MMTV-MDA-7 transgenic mice that specifically express mda-7/IL-24 under the transcriptional control of the MMTV LTR (mouse mammary tumor virus long terminal repeat) promoter. Glucocorticoid, androgen and progesterone response elements are present within the MMTV LTR promoter and are regulated by the estrous cycle in the mammary glands. Transcription of genes under the control of the MMTV LTR promoter are highly up regulated during pregnancy and peak during lactation [34]. To validate expression of MDA-7/IL-24 within the mammary glands of MMTV-MDA-7 mice, we harvested mammary fat pads from pregnant and lactating mice, extracted protein and RNA, and evaluated the expression of MDA-7/IL-24. As would be expected with the MMTV promoter, MDA-7/IL-24 was robustly expressed during pregnancy and lactation both at the transcript and protein level ( Figure 3A and 3B). We did not observe an exact correlation between MDA-7/IL-24 expression at the transcript and protein level and this might be due to post translational modifications of the MDA-7/IL-24 protein. Regardless, we observed that MDA-7/IL-24 was up regulated during pregnancy and lactation. Next, we established MMTV-PyMT cells from MMTV-PyMT mammary tumors, a murine PDX Oncotarget 36931 www.impactjournals.com/oncotarget (patient derived xenograft) (mPDX) cell line, and stably transfected the cells with a luciferase-expressing plasmid in order to monitor tumor growth using bioluminescence (MMTV-PyMT luc cells) ( Figure 3C). Since MDA-7/ IL-24 is robustly expressed during lactation, MMTV-PyMT luc cells were introduced into the fourth mammary fat pad approximately 20 days after the birth of the first litter ( Figure 3D). Mice were monitored for tumor growth using bioluminescence imaging [33,35] (Figure 4A and 4B). In order to maintain MDA-7/IL-24 levels within the mammary glands, females were housed continuously with males to allow for normal pregnancy and lactation cycles. Non-transgenic littermate females were used as controls (MMTV-MDA-7 negative mice). The control mice showed tumor formation by 4 weeks following introduction of tumor cells into the mammary fat pad, however, MMTV-MDA-7 mice showed delayed tumor onset. The difference in tumors in MMTV-MDA-7 and control mice at 10 weeks was dramatic ( Figure 4). Immunohistochemical findings show that MDA-7/IL-24 was expressed in the MMTV-MDA-7 transgenic mouse tumors ( Figure 4D).

MMTV-MDA-7/MMTV-Erbb2 compound transgenic mice show delayed tumor onset
Finally, to determine the relevance of the presence of MDA-7/IL-24 on tumor onset and progression of a model of spontaneous mammary tumor development, we generated MMTV-MDA-7/MMTV-Erbb2 compound transgenic mice ( Figure 5A). MMTV-Erbb2 transgenic mice develop mammary tumors spontaneously over a 5-8 month period [3,36]. We decided to use the MMTV-Erbb2 model for our study because we wanted to determine the role of MDA-7/IL-24 in a model that develops tumors over a prolonged period of time. MMTV-PyMT transgenic mice are invaluable in understanding tumor onset and progression. However, because of the robust and aggressive nature of the tumors that form and the relatively short life span of these mice (about 4 months), we decided to use the MMTV-Erbb2 mouse model with comparatively slower tumor kinetics. In this model, both Erbb2 and mda-7/IL-24 transgene expression is dependent on pregnancy and lactation, so the females were again housed continuously with males to allow for normal pregnancy and lactation cycles. In this model, we observed a significant delay in tumor onset in the compound transgenic mice expressing both mda-7/IL-24 and Erbb2 transgenes as compared to the MMTV-Erbb2 littermate controls ( Figure 5B). Immunohistochemical staining showed that MDA-7/IL-24 was expressed in the MMTV-MDA-7/MMTV-Erbb2 compound mouse tumors ( Figure 5C).
In vitro studies showed that MDA-7/IL-24 induced G 2 /M cell cycle arrest in breast cancer cells via downregulation of AKT-GSK3β and upregulation of cyclin-dependent kinase inhibitor and apoptosis by activation of caspase-dependent signaling pathways and BAX signaling [21,[60][61]. However, normal breast cells were not adversely affected. MDA-7/IL-24 expression also inhibited tumor development in athymic xenograft mouse models [20][21]. Further studies showed that MDA-7/IL-24 could interact with BiP/GRP78 and activate p38 MAPK and GADD expression to selectively cause apoptosis in cancer cells [62][63]. Importantly, MDA-7/IL-24 also inhibited tumor growth of distant untreated tumors in the same athymic mouse, indicating MDA-7/IL-24 also had "bystander" anti-tumor activity [20]. Studies using rat mammary tumor models also showed that MDA-7/ IL-24 could efficiently suppress tumor development [18]. MDA-7/IL-24 induced the expression of growth arrestspecific gene-3 (GAS3). GAS3 interacts with β1 integrin and disrupts the interaction between β1 integrin and fibronectin, leading to suppression of mammary tumor growth [18]. Recent studies in breast cancer-initiating/ Oncotarget 36934 www.impactjournals.com/oncotarget stem cells showed that MDA-7/IL-24 induced apoptosis and endoplasmic reticulum stress and inhibited selfrenewal potential of breast cancer-initiating/stem cells by suppressing the Wnt signaling pathway [17]. Thus, while the mode of action and role of MDA-7/IL-24 in mammary tumors has been well studied, the relevance of MDA-7/ IL-24 in an immune competent animal is not well defined.
In our study, we evaluated the relevance of MDA-7/IL-24 in transgenic models with an intact immune system. The immune system plays a very important role in regulating tumor growth and progression and hence understanding the role of MDA-7/IL-24 in immune competent mice is important. Our first strategy was to introduce MDA-7/IL-24 by adenoviral injection into mammary tumors arising spontaneously in MMTV-PyMT transgenic mice. As expected, introduction of MDA-7/ IL-24 caused a reduction in the growth of the injected tumors. Importantly, MDA-7/IL-24 also showed anticancer "bystander" activity, since the growth of uninjected tumors was also suppressed in animals in which tumors were injected with the MDA-7/IL-24-expressing virus. When tumor sections were assessed for MDA-7/IL-24 expression by immunohistochemistry, we observed that both injected and uninjected tumors in the mice showed the presence of MDA-7/IL-24. Presence of MDA-7/IL-24 in the injected tumors validated that Ad5-CTV could replicate within the mammary tumors. Presence of MDA-7/IL-24 in the uninjected tumors, albeit at lower levels, suggests that MDA-7/IL-24 secreted by the injected tumors might have entered the circulation and localized at the distant tumor site, or more likely the secreted protein interacted with IL-20/IL-22 receptors on the distant tumor, inducing MDA-7/IL-24 production through a "paracrine/ autocrine" effect [22] thereby mediating tumor reduction. Alternatively, production of progeny virus might have entered the circulation and caused secondary infection of distant tumors, resulting in production of MDA-7/IL-24. This would appear less likely, since after several injections of the Ad5-CTV directly into tumors it is predicted that the newly released virus would enter the circulation and either be trapped in the liver or eliminated by the immune system (Supplemental Figure 1B).
Our second strategy was to xenograft cells isolated from mammary tumors developing in MMTV-PyMT (mPDX) mice, akin to PDX tumors isolated from a patient, into the mammary fat pads of naïve MMTV-MDA-7 mice. We introduced MMTV-PyMT luc (mPDX luc) cells into the mammary fat pads of mice after the MMTV-MDA-7 mice had given birth to litters to ensure adequate MDA-7/IL-24 expression. Additionally, by housing males continuously with females and allowing for further offspring, we were able to maintain MDA-7/ IL-24 expression within the mammary glands. This model showed a dramatic delay in tumor growth as compared to the other two models. This might be because we were able to control tumor development by xenografting mouse PDX tumor cells after MDA-7/IL-24 expression was present in the mammary glands. The mechanism underlying this suppression of PDX tumor growth in these mice and whether induction of apoptosis in a portion of the injected murine mPDX tumor cells contributed to the reduced final tumor volume requires further investigation.
Our third strategy was to co-express both a tumor promoting gene, Erbb2, and MDA-7/IL-24 endogenously in the mammary gland, and to follow tumor progression. Since the MMTV-PyMT mouse model is a very aggressive model and tumors form even in virgin mice [4], while MDA-7/IL-24 is optimally expressed during pregnancy Oncotarget 36936 www.impactjournals.com/oncotarget and lactation in MMTV-MDA-7 mice, we decided to use the MMTV-Erbb2 transgenic mouse model [4,64] instead of the MMTV-PyMT model to generate the double transgenic animals. The Erbb2 transgene expression is also regulated by pregnancy and lactation, and hence both Erbb2 and MDA-7/IL-24 expression would occur simultaneously. In this compound transgenic model, tumor onset was delayed by the presence of MDA-7/IL-24. Unlike previous results in athymic mice, we did not observe a complete regression of tumors in our models [20][21]. Given the current understanding of the role of MDA-7/IL-24, it was not expected that MDA-7/IL-24 alone would bring about a complete regression of tumors in transgenic mice as a single agent, since this effect would be dose-dependent even with viruses delivering higher amounts of MDA-7/IL-24. Additionally, in the absence of an intact immune system in athymic nude mice we anticipated enhanced efficacy of the administered Ad5-CTV, since it would be predicted that this conditionally replication competent virus would not be as efficiently cleared by the non-intact immune system in these animals. Of note, in preliminary unpublished studies combining Ad5-CTV with an Mcl-1 inhibitor (designated BI97-D6) resulted in enhanced anticancer activity in the MMTV-PyMT model (unpublished data and Supplemental Figure 2). Thus, our results represent important proofof-principle studies regarding the relevance of MDA-7/ IL-24 in immune competent animals and suggest that MDA-7/IL-24 might serve as an appropriate anti-cancer agent in combination with other therapeutics. Further, we assessed the role of MDA-7/IL-24 in engaging the immune system to mount an anti-tumor immune response against mammary tumors. Intratumoral injection of MDA-7/IL-24 caused an increase in tumor infiltrating, IFN-γ-producing CD8 + T cells. This enhanced immune activation was present in MDA-7/IL-24 treated tumors as well as non-treated tumors, indicating that a systemic antitumor immunity augmented by MDA-7/IL-24 may also contribute to its therapeutic activity in this breast cancer transgenic mouse model. Our findings, together with previous observations of MDA-7/IL-24 in promoting T-helper 1 (Th1) cytokines [11,[65][66], suggest that an immunostimulatory effect of MDA-7/IL-24 should be exploited for effective eradication of breast cancer.
The findings in an accompanying paper by Li et al. provide further evidence for the importance of MDA-7/ IL-24 in mammary tumorigenesis (67). Using a mammary gland specific, tet-inducible MDA-7/IL-24 transgenic mouse model crossed with MMTV-Her2/Neu transgenic mice, the authors show that MDA-7/IL-24 caused an inhibition of tumor development. Additionally using a pre-existing tumor model by implanting lineage-depleted tumor cells isolated from MMTV-rtTA:IL24 tet-on :MMTV-Her2/neu they show that MDA-7/IL-24 expression following doxycycline treatment significantly inhibited pre-existing tumor growth. They provided mechanistic insight into the signaling mechanism of MDA-7/IL-24 and show that the tumor suppressive effects observed in HER2 + breast cancer cells were mediated through PERP, a member of the GAS-3/PMP-22 family of tumor suppressors. The findings in the paper by Li et al. thus further establish the role of MDA-7/IL-24 in suppression of mammary tumors.
In conclusion, our study shows that MDA-7/IL-24 can delay tumor onset as well as tumor progression in transgenic mice and contributes to an immune response against mammary tumors. These important studies provide further in vivo evidence of the tumor suppressor function of this novel member of the IL-10 cytokine gene family [8,10,59]. Further studies are required to evaluate the combinatorial effect of MDA-7/IL-24 with immune components and other therapeutic agents, including chemotherapy, radiotherapy and antibodybased therapies in the context of animal models with intact immune systems. Studies using pure MDA-7/IL-24 protein will also be relevant in defining its anti-cancer activity. For studies using pure MDA-7/IL-24 protein, employing targeted delivery approaches, such as the use of polyfluorocarbonate microbubbles and ultrasound, as part of the ultrasound targeted microbubble destruction (UTMD) approach [28][29][30][68][69] are also worth exploring. Based on MDA-7/IL-24's broad spectrum selective activity toward cancer cells and its multiplicity of anti-cancer effects, including "bystander" antitumor activity, anti-angiogenesis activity, direct cancer-selective apoptosis-and toxic autophagy-induction, immune modulatory activity and synergy with other therapeutic modalities, MDA-7/IL-24 holds significant promise for developing efficacious approaches for the therapy of both primary and metastatic breast cancers, as well as other types of cancer [10,59].

Adenoviruses
The tumor-specific conditionally replicating type 5 adenovirus expressing MDA-7/IL-24 (designated cancer terminator virus -CTV) was generated as described previously [20,23,70]. Briefly, the minimally active region of the tumor-specific PEG (progression elevated gene-3) promoter drives expression of E1A and the CMV promoter drives expression of MDA-7/IL-24 to generate a tumor-specific conditionally replicating adenovirus.

Generation of transgenic mice
The Institutional Animal Care and Use Committee (IACUC) at Virginia Commonwealth University approved all animal studies. Female mice were used for Oncotarget 36937 www.impactjournals.com/oncotarget all experiments. Food and water was provided ad libitum.
MMTV-PyMT transgenic mice: MMTV-PyMT transgenic mice expressing the potent transforming protein Polyoma virus middle tumor antigen (PyMT) under the transcriptional control of the MMTV LTR (mouse mammary tumor virus long terminal repeat) promoter were originally described by Guy et al. [4], and were purchased from The Jackson Laboratory (Stock #002374). A colony was established and maintained in our animal facility by breeding MMTV-PyMT hemizygous male mice to wild type FVB/N female mice. Mice were genotyped by PCR using sense primer (5'-TCCACTACACGATGACTACTGGTC-3') and anti-sense primer (5'-ATGAGCTGGGGTACTTGTTCCTC-3').
MMTV-MDA-7 transgenic mice: To generate the MMTV-MDA-7 transgene construct, a 2.3-kb BamHI fragment containing the MMTV LTR promoter from pRD812 [71] was inserted into the unique BamHI site of the pBSpKCR3 vector [72], which contains the rabbit β-globin intron 2 flanked by a portion of exon 2 and all of exon 3, including the polyadenylation site. A 0.6-kb MfeI fragment containing the human mda-7/IL-24 cDNA was then inserted into the unique EcoRI site in β-globin exon 3. A 4.2-kb injection fragment was excised from the MMTV-MDA-7 construct with XhoI, and transgenic mice were generated by standard methods in an FVB/N genetic background [73]. Potential founders were screened for the presence of the MMTV-MDA-7 transgene by PCR analysis of genomic tail DNA using a rabbit β-globin intron 2 sense primer (5'-ACTACACCCTGGTCATCATCCTGC-3') and a human MDA-7/IL-24 cDNA anti-sense primer (5'-TGTGGACAAGGTAACAGCTCTCAG-3'). Amplification of DNA from mice carrying the transgene generated a 533-bp PCR product. Two independent founders were obtained, and line 2 was selected for use in the studies described here, based upon robust transgene expression in the mammary glands of pregnant and lactating female mice.
MMTV-MDA-7/MMTV-Erbb2 compound transgenic mice: MMTV-Erbb2 transgenic mice expressing the activated rat c-neu oncogene (Erbb2) under the transcriptional control of the MMTV LTR promoter were originally described by Muller et al. [3], and homozygous breeding pairs were obtained from The Jackson Laboratory (stock #005038) and maintained as homozygotes. To generate compound transgenic mice carrying both the MMTV-MDA-7 and MMTV-Erbb2 transgenes, hemizygous MMTV-MDA-7 (line 2) mice (male or female) were mated to homozygous MMTV-Erbb2 mice (male or female), so that all of the offspring were obligate hemizygotes for MMTV-Erbb2. Half of the offspring also carried the MMTV-MDA-7 transgene, while the other half were MMTV-MDA-7-negative. Female offspring were therefore genotyped for the presence of the MMTV-MDA-7 transgene.

Treatment of MMTV-PyMT mice with an adenovirus expressing MDA-7/IL-24 (Ad5-CTV)
MMTV-PyMT mice develop mammary tumors spontaneously in all the mammary glands within a period from 2-3 months of age. A control group of 5 mice were left untreated and tumors were monitored in these mice. To determine the potential tumor suppressive effects of MDA-7/IL-24, Ad5-CTV or Ad5-E1A (1 x 10 8 IU of the respective adenovirus in 50 µl) were injected intratumorally once a palpable tumor was observed in any mammary gland. As the mice developed tumors in the other mammary glands, at least 50% of the tumors that formed were injected with the respective adenovirus (e.g., 1 tumor was injected in mice with 1 or 2 tumors, 2 tumors were injected in mice with 3 or 4 tumors and so on). Care was taken to log the exact tumor that was injected to ensure repeated injection of the same tumors. Approximately 50% of the tumors within a particular mouse were left untreated to determine the bystander anti-tumor properties of MDA-7/IL-24. Each injected tumor received a maximum of 10 injections (depending on when palpable tumors were observed) over a 4-week period. Mice were sacrificed, tumors were harvested, formalin-fixed, paraffin-embedded and sectioned, and immunohistochemistry was performed following standard procedures.

MDA-7 transgenic mice
Mammary glands from pregnant and lactating female MMTV-MDA-7 transgenic mice were harvested and flash frozen in liquid nitrogen. Protein and RNA were extracted using standard procedures and the expression of MDA-7/IL-24 was assessed at the transcript and protein level. Real-time quantitative PCR was performed according to standard procedures as described previously [73]. MDA-7/IL-24 and Gapdh primer probes were obtained from Life Technologies. Western blotting was performed as described previously [74]. MDA-7/IL-24 antibody was obtained from GenHunter and EF1α antibody was obtained from EMD Millipore.

Generating MMTV-PyMT and MMTV-PyMT luc cells
Mammary tumors were harvested from MMTV-PyMT mice to develop mouse "patient-derived xenograft" (murine PDX); mPDX tumors, similar to human PDX tumors. The MMTV-PyMT tumors were cut into small pieces and digested using trypsin-EDTA to obtain single cells. The cells were washed in sterile PBS and then plated in DMEM media supplemented with 5% Pencillin/ www.impactjournals.com/oncotarget Streptomycin and 5% fetal bovine serum (FBS). The cells were allowed to attach and media was replenished to remove the unattached cells. The cells were passaged to obtain MMTV-PyMT (mPDX) tumor cells. The cells were injected in FVB mice to ensure that the cells retained their tumor-forming abilities. To enable tumor growth monitoring using bioluminescent imaging, MMTV-PyMT cells were transfected with a luciferaseexpressing construct (pGL4.50, Invitrogen). MMTV-PyMT cells stably expressing luciferase were selected using hygromycin and were designated MMTV-PyMT luc (mPDX luc) cells.

Tumor growth assessment in MMTV-MDA-7 transgenic mice
Each MMTV-MDA-7 (positive or negative/control/ non-transgenic littermate) female mouse was housed continuously with one male mouse. About 20 days after the birth of the first litter, MMTV-PyMT luc cells (1 x 10 6 cells in 50 µl) were injected into the 4 th mammary fat pad of 10 MMTV-MDA-7 transgenic mice and 10 control (non-transgenic littermate) mice. Tumor growth was monitored using bioluminescent imaging. Mice were sacrificed before tumors reached the maximum allowed limit. Tumors were harvested, formalin-fixed, paraffinembedded and sectioned, and immunohistochemistry was performed following standard procedures.

Tumor growth assessment in MMTV-MDA-7/ MMTV-Erbb2 compound transgenic mice
Each MMTV-MDA-7/MMTV-Erbb2 female mouse was housed continuously with one male mouse. MMTV-MDA-7 negative/MMTV-Erbb2 positive female littermates were used as controls and were also housed continuously with a male mouse. Ten female mice were assessed per group. The female mice were monitored for tumor onset and then tumor growth twice weekly. Mice were sacrificed before tumors reached the maximum allowed limit. Tumors were harvested, formalin-fixed, paraffinembedded and sectioned for immunohistochemistry.

Immunohistochemistry
Immunohistochemistry was performed according to standard protocols. Briefly, tumor sections were deparaffinized at 60˚C for 1 hour, followed by rehydration, and antigen retrieval using citrate buffer and heating. Avidin and biotin blocking kits and Vectastain ABC complex kits were obtained from Vector Laboratories. The MDA-7/IL-24 antibody utilized was from GenHunter and the E1A antibody was obtained from Thermo Scientific. Secondary antibodies were obtained from Jackson Immunoresearch. The slides were counter-stained using hematoxylin. Following staining, slides were dehydrated and mounted using Vectashield mounting media (Vector Laboratories).

Statistical analysis
Statistical analyses were performed using GraphPad Prism 5. Data is presented as mean ± SEM. Student's t-test and Kaplan Meier analysis were applied based on the statistical mandates or suggestions of each analysis.