Akanksha Mahajan

Genomic comparison of early-passage conditionally reprogrammed breast cancer cells to their corresponding primary tumors

Conditionally reprogrammed cells (CRCs) are epithelial cells that are directly isolated from patients’ specimens and propagated in vitro with feeder cells and a Rho kinase inhibitor. A number of these cells have been generated from biopsies of breast cancer patients, including ductal carcinoma in situ and invasive carcinomas. The characterization of their genomic signatures is essential to determine their ability to reflect the natural biology of their tumors of origin. In this study, we performed the genomic characterization of six newly established invasive breast cancer CRC cultures in comparison to the original patients’ primary breast tumors (PBT) from which they derived. The CRCs and corresponding PBTs were simultaneously profiled by genome-wide array-CGH, targeted next generation sequencing and global miRNA expression to determine their molecular similarities in the patterns of copy number alterations (CNAs), gene mutations and miRNA expression levels, respectively. The CRCs’ epithelial cells content and ploidy levels were also evaluated by flow cytometry. A similar level of CNAs was observed in the pairs of CRCs/PBTs analyzed by array-CGH, with >95% of overlap for the most frequently affected cytobands. Consistently, targeted next generation sequencing analysis showed the retention of specific somatic variants in the CRCs as present in their original PBTs. Global miRNA profiling closely clustered the CRCs with their PBTs (Pearson Correlation, ANOVA paired test, P<0.05), indicating also similarity at the miRNA expression level; the retention of tumor-specific alterations in a subset of miRNAs in the CRCs was further confirmed by qRT-PCR. These data demonstrated that the human breast cancer CRCs of this study maintained at early passages the overall copy number, gene mutations and miRNA expression patterns of their original tumors. The further characterization of these cells by other molecular and cellular phenotypes at late cell passages, are required to further expand their use as a unique and representative ex-vivo tumor model for basic science and translational breast cancer studies.

Abstract 1940: Prenatal stress increases malignancy of neuroblastoma tumors in TH-MYCN animal model

Neuroblastoma (NB) is a pediatric malignancy arising due to defects in sympathetic neuron differentiation. NB is a heterogeneous disease, with phenotypes ranging from spontaneously regressing to highly aggressive, incurable tumors. This clinical variability cannot be explained solely by genetic aberrations. Even in families with hereditary NB the penetrance of the disease is incomplete and the same genetic mutation often results in tumors with phenotypes varying from differentiating ganglioneuromas to undifferentiated, highly aggressive NBs. Thus, other, perhaps non-genetic factors can contribute to the disease development and modify its phenotype. Strikingly, the two factors promoting de-differentiation of NB cells and their malignant phenotype, hypoxia and glucocorticoids, are elevated in the fetus during maternal stress, suggesting a role for prenatal stress in NB tumorigenesis. Previously, using TH-MYCN mice as a model of aggressive NB, we have shown that an increase in maternal corticosterone levels during pregnancy attained by inserting slow release pellets resulted in increased tumor frequency in TH-MYCN offspring. The goal of the current study was to determine the effect of prenatal stress on NB metastasis. To this end, pregnant mice carrying TH-MYCN hemizygous offspring were subjected to chronic stress at embryonic days 10-17, the time of sympathetic neuroblast proliferation and differentiation. Two established stress paradigms were used - chronic unpredictable stress, in which mice were subjected daily to various stressors, and chronic cold stress comprising of daily 30 min exposure to cold. The phenotypes of the disease and its dissemination were compared between offspring of control and stressed mothers. The offspring from both prenatally stressed groups presented with more malignant disease, as manifested by the presence of advanced lung metastases disseminating from small primary tumors ( 1,000 mm3). Although not common, lung metastases occur preferentially in NB patients with MYCN amplification and are associated with significantly worse prognosis, as compared to patients with metastatic disease, but no pulmonary involvement (14 vs 43% 3-year event-free survival, respectively). Thus, the profound pulmonary dissemination observed in prenatally-stressed TH-MYCN mice mimics one of the most malignant NB phenotypes observed in human disease. Altogether, our data implicate maternal stress during pregnancy as a potential environmental factor modifying the effects of genetic aberrations and promoting malignant phenotype of NB. Citation Format: Sung Hyeok Hong, Larissa Wietlisbach, Susana Galli, Akanksha Mahajan, Shiya Zhu, Jason Tilan, Yichien Lee, Olga Rodriguez, Chris Albanese, Joanna Kitlinska. Prenatal stress increases malignancy of neuroblastoma tumors in TH-MYCN animal model [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 1940. doi:10.1158/1538-7445.AM2017-1940

In Vivo Model for Testing Effect of Hypoxia on Tumor Metastasis

Hypoxia has been implicated in the metastasis of Ewing sarcoma (ES) by clinical observations and in vitro data, yet direct evidence for its pro-metastatic effect is lacking and the exact mechanisms of its action are unclear. Here, we report an animal model that allows for direct testing of the effects of tumor hypoxia on ES dissemination and investigation into the underlying pathways involved. This approach combines two well-established experimental strategies, orthotopic xenografting of ES cells and femoral artery ligation (FAL), which induces hindlimb ischemia. Human ES cells were injected into the gastrocnemius muscles of SCID/beige mice and the primary tumors were allowed to grow to a size of 250 mm3. At this stage either the tumors were excised (control group) or the animals were subjected to FAL to create tumor hypoxia, followed by tumor excision 3 days later. The efficiency of FAL was confirmed by a significant increase in binding of hypoxyprobe-1 in the tumor tissue, severe tumor necrosis and complete inhibition of primary tumor growth. Importantly, despite these direct effects of ischemia, an enhanced dissemination of tumor cells from the hypoxic tumors was observed. This experimental strategy enables comparative analysis of the metastatic properties of primary tumors of the same size, yet significantly different levels of hypoxia. It also provides a new platform to further assess the mechanistic basis for the hypoxia-induced alterations that occur during metastatic tumor progression in vivo. In addition, while this model was established using ES cells, we anticipate that this experimental strategy can be used to test the effect of hypoxia in other sarcomas, as well as tumors orthotopically implanted in sites with a well-defined blood supply route.

Abstract 2478: Tumor hypoxia promotes Ewing sarcoma metastases in a mouse xenograft model

Ewing sarcoma (ES) is a pediatric tumor induced by EWS-ETS fusion proteins, most often EWS-FLI1. While the presence of metastases is the single most powerful adverse prognostic factor for ES patients, the mechanisms underlying their development remain unclear. Tumor hypoxia is one of the few factors implicated in ES progression. In ES patients, the presence of nonperfused areas within tumor tissue was associated with poor prognosis. In vitro, hypoxia increases invasiveness of ES cells and triggers expression of pro-metastatic genes via changes in transcriptional activity of the EWS-FLI1 gene. However, despite this line of evidence, no direct proof for this hypoxia-induced ES progression and spread has been provided. Moreover, the mechanisms by which hypoxia could exert such effects are unknown. To fill this gap, we created an in vivo model of hypoxia in ES and tested its effect on tumor metastasis. SK-ES1 ES cells were injected into gastrocnemius muscles of SCID/beige mice. Once the tumors reached a volume of 250mm3, they were either excised (control) or subjected to femoral artery ligation (FAL) for 72h prior to excision, inducing ischemia of the lower hindlimb, thus creating tumor hypoxia. Then, the mice were monitored for metastases. The extent of the metastatic disease was assessed and compared between experimental groups based on periodic MRI, necropsy and histopathology findings. FAL resulted in profound tumor hypoxia, as evidenced by inhibition of primary tumor growth, severe tissue necrosis and positive staining for a hypoxyprobe, pimonidazole. However, despite the impaired growth of primary tumors, xenografts subjected to FAL were more metastatic. The involvement of hypoxic cells in metastases was evidenced by the accumulation of pimonidazole-positive cells (hypoxic at the time of FAL) in areas of tissue invasion and intravasation. Consequently, mice bearing FAL-treated tumors exhibited a decreased latency of metastases formation and an increase in their number from an average of 0.9 to 2.3 metastases per mouse in control and FAL groups, respectively. We also observed a change in the pattern of metastases, as FAL-treated tumors metastasized more often to distant organs (average of 0.3 organ metastases per mouse in control and 1.3 in FAL group). The hypoxia-induced metastases were most often observed in adrenal gland and spine (50% and 42% of mice in FAL group, respectively), while no such metastases were observed in the control group. Moreover, 100% of FAL-treated mice had signs of bone marrow invasion, while no tumor cells were detectable in bone marrow of control mice. This data provides the first-ever direct evidence for tumor hypoxia as a driver of ES metastases. Moreover, our model of tumor hypoxia in vivo provides an excellent opportunity to identify hypoxia-induced pathways involved in ES metastatic progression that subsequently may become novel therapeutic targets for this disease. Citation Format: Jason U. Tilan, Sung-Hyeok Hong, Susana Galli, Rachel Acree, Katherine Connors, Meredith Horton, Akanksha Mahajan, Larissa Wietlisbach, Yi-Chien Lee, Olga Rodriguez, Christopher Albanese, Joanna Kitlinska. Tumor hypoxia promotes Ewing sarcoma metastases in a mouse xenograft model. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2478.