Search

We investigated tumor-cell-intrinsic chromatin accessibility patterns of pancreatic ductal adenocarcinoma (PDAC) by ATAC-seq on EpCAM+ PDAC malignant epithelial cells, sorted from 54 freshly resected human tumors, and discovered a signature of 1092 chromatin loci displaying differential accessibility between patients with disease free survival (DFS) < 1 year and patients with DFS > 1 year. Analyzing transcription factor (TF) binding motifs within these loci, we identified two TFs (ZKSCAN1 and HNF1b) displaying differential nuclear localization between patients with short vs. long DFS. We further developed a novel chromatin accessibility microarray methodology termed ATAC-Array, an easy-to-use platform obviating the time and cost of next generation sequencing. Applying this novel methodology to the original ATAC-seq libraries as well as independent libraries generated from patient-derived organoids, we validated ATAC-array technology in both the original ATAC-Seq cohort as well as in an independent validation cohort. We conclude that PDAC prognosis can be predicted by ATAC-array, which represents a novel, low-cost, clinically feasible technology for assessing chromatin accessibility profiles.

Unlike other malignancies, therapeutic options in pancreatic ductal adenocarcinoma (PDAC) are largely limited to cytotoxic chemotherapy without the benefit of molecular markers predicting response. Here we report tumor-cell-intrinsic chromatin accessibility patterns of treatment-naïve surgically resected PDAC tumors that were subsequently treated with (Gem)/Abraxane adjuvant chemotherapy. By ATAC-seq analyses of EpCAM+ PDAC malignant epithelial cells sorted from 54 freshly resected human tumors, we show here the discovery of a signature of 1092 chromatin loci displaying differential accessibility between patients with disease free survival (DFS) 1 year. Analyzing transcription factor (TF) binding motifs within these loci, we identify two TFs (ZKSCAN1 and HNF1b) displaying differential nuclear localization between patients with short vs. long DFS. We further develop a chromatin accessibility microarray methodology termed “ATAC-array”, an easy-to-use platform obviating the time and cost of next generation sequencing. Applying this methodology to the original ATAC-seq libraries as well as independent libraries generated from patient-derived organoids, we validate ATAC-array technology in both the original ATAC-seq cohort as well as in an independent validation cohort. We conclude that PDAC prognosis can be predicted by ATAC-array, which represents a low-cost, clinically feasible technology for assessing chromatin accessibility profiles., ATAC-seq allows for the investigation of chromatin accessibility, a feature which can have important implications for gene regulation. Here the authors present ATAC-array to assess accessibility in human tumours and based on this predict disease prognosis.

SummaryExtensive transcriptional and ontogenetic diversity exists among normal tissue-resident macrophages, with unique transcriptional profiles endowing the cells with tissue-specific functions. However, it is unknown whether the origins of different macrophage populations affect their roles in malignancy. Given potential artifacts associated with irradiation-based lineage tracing, it remains unclear if bone-marrow-derived macrophages (BMDMs) are present in tumors of the brain, a tissue with no homeostatic involvement of BMDMs. Here, we employed multiple models of murine brain malignancy and genetic lineage tracing to demonstrate that BMDMs are abundant in primary and metastatic brain tumors. Our data indicate that distinct transcriptional networks in brain-resident microglia and recruited BMDMs are associated with tumor-mediated education yet are also influenced by chromatin landscapes established before tumor initiation. Furthermore, we demonstrate that microglia specifically repress Itga4 (CD49D), enabling its utility as a discriminatory marker between microglia and BMDMs in primary and metastatic disease in mouse and human.

Personalized therapy is the future of cancer care. Almost half of the cancer patients do not respond to chemotherapy. For instance, pancreatic ductal adenocarcinoma (PDAC) patients with the limited local disease and with no detectable metastasis typically have their primary tumor surgically resected, but the disease recurs in approximately 50% of cases within 1 year of surgery, in spite of adjuvant chemotherapy. Although gene expression signatures correlating prognosis have been described in PDAC, the therapeutic utility of these signatures has been limited based in part on a large number of genes displaying an altered expression. On the other hand, regulatory regions common to these genes might be amenable to collective epigenetic reprogramming using epigenetic drugs. We interrogated genome-wide chromatin accessibility using Assay for Transposase-Accessible Chromatin sequencing (ATAC-seq) on EpCAM+ PDAC malignant cells sorted from a cohort of 54 treatment-naïve resected tumors, in hopes of defining a tumor-intrinsic chromatin signature associated with recurrence. We discovered a signature of 1092 loci that were differentially accessible between recurrent (disease-free survival (DFS) < 1 year) and non-recurrent patients (DFS > 1 year). Through transcription factor (TF) binding motif analysis, we identified candidate TFs whose accessible motifs were differentially associated with recurrence. Nuclear localization of two such TFs, ZKSCAN1 and HNF1b, were assessed by immunostaining on tissue microarrays (TMA) representing 40 out of 54 patients. Nuclear staining of HNF1b was strong in tumor tissue from non-recurrent patients and weak or absent in recurrent patients, but ZKSCAN1 staining patterns were not significantly associated with recurrence. In a TMA representing an independent PDAC cohort (n=97) preselected for 52 long (OS 6 years)- and 45 short (OS 6 months)- term survivors, the number of nuclear positive cells for HNF1b was 52-fold higher in the long-term compared to the short-term survivors and that for ZKSCAN1 was 5.3-fold higher in the short-term compared to the long-term survivors. We further validated the 1092 chromatin accessibility signature by a novel microarray-based platform technology that we termed “ATAC-Array”, where the differentially accessible regions from the signatures were probed on a glass slide and then hybridized with fluorescent-labeled ATAC-libraries. This is a cost-effective, easy-to-use platform technology avoiding the time and cost of next-generation ATAC library sequencing. ATAC-array is the only microarray that reads chromatin accessibility. We have compared ATAC-array side-by-side with ATAC-seq (n=30) and found significant correlation (Pearson's median r= 0.64, range= 0.50- 0.77). By performing ATAC-array on the PDAC cohort (n=38), we have independently re-classified the patients who recurred early and the ones who did not (Gehan-Breslow-Wilcoxon test p=0.0076). ATAC-array, as the technology itself, has enormous potential for a wide range of applications, and we propose to develop it as a clinically validated theragnostic tool to predict and stratify cancer patients for epigenetic therapy. Citation Format: Surajit Dhara, Sagar Chhangawala, Himanshu Chintalapudi, Alexandra L. Massa, Victoria Aveson, Gokce Askan, Liguo Zhang, Remy Nicolle, Alvin P. Makohon-Moore, Smrita Sinha, Jiang Gui, Richard Moffitt, Kenneth H. Yu, Vinod Balachandran, Rohit Chandwani, Christina Leslie, Steven D. Leach. Pancreatic cancer prognosis is predicted by chromatin accessibility microarray [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr LB-263.

T-cell acute lymphoblastic leukemia (T-ALL) is an aggressive leukemia with high relapse rates compared to B-lineage ALL. We previously showed that HMGA1a transgenic mice develop aggressive T-ALL, indicating that HMGA1 causes leukemic transformation in vivo. HMGA1 is also highly expressed in embryonic stem cells, hematopoietic stem cells and diverse, refractory human cancers. Disruption of the CDKN2A tumor suppressor locus occurs in most cases of T-ALL and is thought to contribute to leukemic transformation. To determine whether loss of function of CDKN2A cooperates with HMGA1 in T-ALL, we crossed HMGA1a transgenics onto a Cdkn2a null background. We discovered that T-ALL is markedly accelerated in HMGA1a transgenic Cdkn2a null mice. In addition, these mice recapitulate salient clinical and pathologic features of human T-ALL. HMGA1 is also highly overexpressed in human T-ALL. These findings suggest that HMGA1 plays a causative role in T-ALL and could represent a rational therapeutic target.

Angiogenesis plays an important role in tumor growth and metastasis; therefore, inhibition of angiogenesis is a promising strategy for developing new anticancer drugs. Type 2 methionine aminopeptidase (MetAP2) protein is likely a molecular target of angiogenesis inhibitors.Nitroxoline, an antibiotic used to treat urinary tract infections, was identified from a high-throughput screen of a library of 175,000 compounds for MetAP2 inhibitors and from a parallel screen using the Johns Hopkins Drug Library to identify currently used clinical drugs that can also inhibit human umbilical vein endothelial cells (HUVEC) proliferation. To investigate the mechanism of action of nitroxoline, inhibition of MetAP2 activity and induction of senescence were assessed in HUVEC. To test the antiangiogenic activity of nitroxoline, endothelial tube formation in Matrigel and microvessel formation in Matrigel plugs in vivo were assessed. Antitumor efficacy of nitroxoline was evaluated in mouse models of human breast cancer xenograft (n = 10) and bladder cancer orthotopic xenograft (n = 11). Furthermore, the mechanism of action of nitroxoline was investigated in vivo.Nitroxoline inhibited MetAP2 activity in vitro (half maximal inhibitory concentration [IC(50)] = 54.8 nM, 95% confidence interval [CI] = 22.6 to 132.8 nM) and HUVEC proliferation (IC(50) = 1.9 μM, 95% CI = 1.54 to 2.39 μM). Nitroxoline inhibited MetAP2 activity in HUVEC in a dose-dependent manner and induced premature senescence in a biphasic manner. Nitroxoline inhibited endothelial tube formation in Matrigel and reduced microvessel density in vivo. Mice (five per group) treated with nitroxoline showed a 60% reduction in tumor volume in breast cancer xenografts (tumor volume on day 30, vehicle vs nitroxoline, mean = 215.4 vs 86.5 mm(3), difference = 128.9 mm(3), 95% CI = 32.9 to 225.0 mm(3), P = .012) and statistically significantly inhibited growth of bladder cancer in an orthotopic mouse model (tumor bioluminescence intensities of vehicle [n = 5] vs nitroxoline [n = 6], P = .045).Nitroxoline shows promise as a potential therapeutic antiangiogenic agent.

Background and aims: Pancreatic cancer is among the most dismal of human malignancies. Current therapeutic strategies are virtually ineffective in controlling advanced, metastatic disease. Recent evidence suggests that the Hedgehog signalling pathway is aberrantly reactivated in the majority of pancreatic cancers, and that Hedgehog blockade has the potential to prevent disease progression and metastatic spread. Methods: Here it is shown that the Hedgehog pathway is activated in the Pdx1-Cre;LsL-Kras G12D ;Ink4a/Arf lox/lox transgenic mouse model of pancreatic cancer. The effect of Hedgehog pathway inhibition on survival was determined by continuous application of the small molecule cyclopamine, a smoothened antagonist. Microarray analysis was performed on non-malignant human pancreatic ductal cells overexpressing Gli1 in order to screen for downstream Hedgehog target genes likely to be involved in pancreatic cancer progression. Results: Hedgehog inhibition with cyclopamine significantly prolonged median survival in the transgenic mouse model used here (67 vs 61 days; p = 0.026). In vitro data indicated that Hedgehog activation might at least in part be ascribed to oncogenic Kras signalling. Microarray analysis identified 26 potential Hedgehog target genes that had previously been found to be overexpressed in pancreatic cancer. Five of them, BIRC3, COL11A1, NNMT, PLAU and TGM2, had been described as upregulated in more than one global gene expression analysis before. Conclusion: This study provides another line of evidence that Hedgehog signalling is a valid target for the development of novel therapeutics for pancreatic cancer that might be worth evaluating soon in a clinical setting.

Hedgehog (Hh) signaling pathway is a valid therapeutic target in a wide range of malignancies. We focus here on glioblastoma multiforme (GBM), a lethal malignancy of the central nervous system (CNS). By analyzing RNA-sequencing based transcriptomics data on 149 clinical cases of TCGA-GBM database we show here a strong correlation (r = 0.7) between GLI1 and PTCH1 mRNA expression--as a hallmark of the canonical Hh-pathway activity in this malignancy. GLI1 mRNA expression varied in 3 orders of magnitude among the GBM patients of the same cohort showing a single continuous distribution-unlike the discrete high/low-GLI1 mRNA expressing clusters of medulloblastoma (MB). When compared with MB as a reference, the median GLI1 mRNA expression in GBM appeared 14.8 fold lower than that of the "high-Hh" cluster of MB but 5.6 fold higher than that of the "low-Hh" cluster of MB. Next, we demonstrated statistically significant up- and down-regulation of GLI1 mRNA expressions in GBM patient-derived low-passage neurospheres in vitro by sonic hedgehog ligand-enriched conditioned media (shh-CM) and by Hh-inhibitor drug vismodegib respectively. We also showed clinically achievable dose (50 μM) of vismodegib alone to be sufficient to induce apoptosis and cell cycle arrest in these low-passage GBM neurospheres in vitro. Vismodegib showed an effect on the neurospheres, both by down-regulating GLI1 mRNA expression and by inducing apoptosis/cell cycle arrest, irrespective of their relative endogenous levels of GLI1 mRNA expression. We conclude from our study that this single continuous distribution pattern of GLI1 mRNA expression technically puts almost all GBM patients in a single group rather than discrete high- or low-clusters in terms of Hh-pathway activity. That is suggestive of therapies with Hh-pathway inhibitor drugs in this malignancy without a need for further stratification of patients on the basis of relative levels of Hh-pathway activity among them.

Metastatic cancers adopt certain properties of normal cells in developing or regenerating organs, such as the ability to proliferate and alter tissue organization. We find here that activity of the Hedgehog (Hh) signalling pathway, which has essential roles in developmental patterning, is required for regeneration of prostate epithelium, and that continuous pathway activation transforms prostate progenitor cells and renders them tumorigenic. Elevated pathway activity furthermore distinguishes metastatic from localized prostate cancer, and pathway manipulation can modulate invasiveness and metastasis. Pathway activity is triggered in response to endogenous expression of Hh ligands, and is dependent upon the expression of Smoothened, an essential Hh response component that is not expressed in benign prostate epithelial cells. Monitoring and manipulating Hh pathway activity may thus offer significant improvements in diagnosis and treatment of prostate cancers with metastatic potential.

An early step in the carcinogenesis of hereditary non-polyposis colorectal cancer (HNPCC) and some sporadic colorectal cancers (CRCs) is the acquisition of a 'mutator phenotype' resulting from defects in DNA mismatch repair (MMR) genes, which normally maintain genomic stability. This mutator phenotype causes an approximately 100-1000-fold increase in base substitutions and small insertion/deletion mutations thereby driving carcinogenesis. It also causes genome-wide microsatellite instability (MSI) due to the inability to repair mutations within these small, hard to replicate, repetitive DNA elements. In contrast, less is known about the role of mutator phenotypes in microsatellite stable (MSS) CRC. In this report, we have measured the mutation rates in 11 MSS CRC cell lines to obtain an estimate of the prevalence of mutator phenotypes in MSS carcinogenesis. Of the 11 cell lines, three of them (27%) possess spontaneous hypoxanthine phosphoribosyltransferase mutation rates approximately 10-100-fold above background. When challenged with alkylating and oxidising agents, the degree of survival and apoptotic responses are different, indicating that these cell lines may represent more than one mutator phenotype. These data demonstrate that a significant portion of MSS CRC cell lines has increased mutation rates and that this may play a role in MSS CRC carcinogenesis.

Neoplastic cells of Glioblastoma multiforme (GBM) may or may not show sustained response to temozolomide (TMZ) chemotherapy. We hypothesize that TMZ chemotherapy response in GBM is predetermined in its neoplastic clones via a specific set of mutations that alter relevant pathways. We describe exome-wide enrichment of variant allele frequencies (VAFs) in neurospheres displaying contrasting phenotypes of sustained versus reversible TMZ-responses in vitro. Enrichment of VAFs was found on genes ST5, RP6KA1 and PRKDC in cells showing sustained TMZ-effect whereas on genes FREM2, AASDH and STK36, in cells showing reversible TMZ-effect. Ingenuity pathway analysis (IPA) revealed that these genes alter cell-cycle, G2/M-checkpoint-regulation and NHEJ pathways in sustained TMZ-effect cells whereas the lysine-II&V/phenylalanine degradation and sonic hedgehog (Hh) pathways in reversible TMZ-effect cells. Next, we validated the likely involvement of the Hh-pathway in TMZ-response on additional GBM neurospheres as well as on GBM patients, by extracting RNA-sequencing-based gene expression data from the TCGA-GBM database. Finally, we demonstrated TMZ-sensitization of a TMZ non-responder neurosphere in vitro by treating them with the FDA-approved pharmacological Hh-pathway inhibitor vismodegib. Altogether, our results indicate that the Hh-pathway impedes sustained TMZ-response in GBM and could be a potential therapeutic target to enhance TMZ-response in this malignancy.

The aim of this study was to analyse the expression of transcripts and proteins for vascular endothelial growth factor (VEGF) and placenta growth factor (PlGF) in different compartments of the early conceptus at primary implantation sites during lacunar (n = 6), early villous (n = 9) and villous placenta (n = 6) stages of gestation in the rhesus monkey. During the lacunar stage, VEGF expression was observed in the cytotrophoblast cells lining the extraembryonic cavity, but these cells did not express PlGF. With further development, cytotrophoblast cells lining villi, forming columns, and constituting anchoring villi, expressed both VEGF and PlGF during early villous and villous placenta stages. In addition, chorion, amnion and villous stromal cells expressed both VEGF and PlGF proteins and mRNA. During the lacunar stage, all epithelial cells in maternal endometrium generally expressed VEGF, while PlGF expression was observed in the plaque epithelium only. As gestation advanced, the expression of VEGF and PlGF from plaque cells decreased, and in surface and glandular epithelium the expression of VEGF increased, while the expression of PlGF remained unaltered. Decidual stromal cells expressed VEGF and PlGF only at low levels during the lacunar stage, while the expression of both increased during the early villous and the villous placenta stages of implantation. It appears from the present study that the expression of VEGF and PlGF are regulated in a temporal and spatial manner during early stages of implantation and that their concerted actions in placental and maternal compartments play a critical role in the evolving pregnancy in the rhesus monkey.

Although lung cancer is the leading cause of cancer death worldwide, the precise molecular mechanisms that give rise to lung cancer are incompletely understood. Here, we show that HMGA1 is an important oncogene that drives transformation in undifferentiated, large-cell carcinoma. First, we show that the HMGA1 gene is overexpressed in lung cancer cell lines and primary human lung tumors. Forced overexpression of HMGA1 induces a transformed phenotype with anchorage-independent cell growth in cultured lung cells derived from normal tissue. Conversely, inhibiting HMGA1 expression blocks anchorage-independent cell growth in the H1299 metastatic, undifferentiated, large-cell human lung carcinoma cells. We also show that the matrix metalloproteinase-2 (MMP-2) gene is a downstream target upregulated by HMGA1 in large-cell carcinoma cells. In chromatin immunoprecipitation experiments, HMGA1 binds directly to the MMP-2 promoter in vivo in large-cell lung cancer cells, but not in squamous cell carcinoma cells. In large-cell carcinoma cell lines, there is a significant, positive correlation between HMGA1 and MMP-2 mRNA. Moreover, interfering with MMP-2 expression blocks anchorage-independent cell growth in H1299 large-cell carcinoma cells, indicating that the HMGA1–MMP-2 pathway is required for this transformation phenotype in these cells. Blocking MMP-2 expression also inhibits migration and invasion in the H1299 large-cell carcinoma cells. Our findings suggest an important role for MMP-2 in transformation mediated by HMGA1 in large-cell, undifferentiated lung carcinoma and support the development of strategies to target this pathway in selected tumors. (Mol Cancer Res 2009;7(11):1803–12)

We synthesized YC-27 3 to provide a fluorescent imaging agent for the prostate-specific membrane antigen (PSMA), a marker for hormone-independent prostate cancer and tumor neovasculature, with suitable pharmacokinetics for use in vivo. Immediate precursor trifluoroacetate salt of 2-(3-{5-[7-(5-amino-1-carboxy-pentylcarbamoyl)-heptanoylamino]-1-carboxy-pentyl}-ureido)-pentanedioic acid 2 was conjugated with a commercially available near-infrared light-emitting dye (IRDye 800CW) to provide 3 in 72% yield. YC-27 3 demonstrated a PSMA inhibitory activity of 0.37 nM and was capable of generating target-to-nontarget ratios of at least 10 in PSMA-expressing PC3-PIP vs. PSMA-negative PC3-flu tumors in vivo. YC-27 3 may be useful for study of PSMA-expressing tissue in preclinical models or for intraoperative guidance.

Although HMGA1 (high-mobility group A1; formerly HMG-I/Y) is an oncogene that is widely overexpressed in aggressive cancers, the molecular mechanisms underlying transformation by HMGA1 are only beginning to emerge. HMGA1 encodes the HMGA1a and HMGA1b protein isoforms, which function in regulating gene expression. To determine how HMGA1 leads to neoplastic transformation, we looked for genes regulated by HMGA1 using gene expression profile analysis. Here, we show that the STAT3 gene, which encodes the signaling molecule signal transducer and activator of transcription 3 (STAT3), is a critical downstream target of HMGA1a. STAT3 mRNA and protein are up-regulated in fibroblasts overexpressing HMGA1a and activated STAT3 recapitulates the transforming activity of HMGA1a in fibroblasts. HMGA1a also binds directly to a conserved region of the STAT3 promoter in vivo in human leukemia cells by chromatin immunoprecipitation and activates transcription of the STAT3 promoter in transfection experiments. To determine if this pathway contributes to HMGA1-mediated transformation, we investigated STAT3 expression in our HMGA1a transgenic mice, all of which developed aggressive lymphoid malignancy. STAT3 expression was increased in the leukemia cells from our transgenics but not in control cells. Blocking STAT3 function induced apoptosis in the transgenic leukemia cells but not in controls. In primary human leukemia samples, there was a positive correlation between HMGA1a and STAT3 mRNA. Moreover, blocking STAT3 function in human leukemia or lymphoma cells led to decreased cellular motility and foci formation. Our results show that the HMGA1a–STAT3 axis is a potential Achilles heel that could be exploited therapeutically in hematopoietic and other malignancies overexpressing HMGA1a. [Cancer Res 2008;68(24):10121–7]

If molecular imaging is to prove clinically useful it will have to surpass current, primarily anatomic techniques in terms of sensitivity and the ability to detect minimal changes in tissue. One of the most important tests for molecular imaging is to determine whether it can image the metastatic potential of tumors. Like all predictive endeavors, the imaging of such "potential" is a daunting task, but one that only molecular imaging--rather than standard, anatomic techniques--is likely to solve. Although difficult, imaging of metastatic potential is also arguably the most important task for molecular imaging of cancer because it is generally the dissemination of malignant tissue, not its prolonged residence in an inopportune site, which kills the patient. Below are examples of uses of molecular imaging of metastases as well as of metastatic potential, the former being a far more developed area of clinical inquiry.

In the context of pancreatic cancer, metastasis remains the most critical determinant of resectability, and hence survival. The objective of this study was to determine whether Hedgehog (Hh) signaling plays a role in pancreatic cancer invasion and metastasis because this is likely to have profound clinical implications. In pancreatic cancer cell lines, Hh inhibition with cyclopamine resulted in down-regulation of snail and up-regulation of E-cadherin, consistent with inhibition of epithelial-to-mesenchymal transition, and was mirrored by a striking reduction of in vitro invasive capacity (P < 0.0001). Conversely, Gli1 overexpression in immortalized human pancreatic ductal epithelial cells led to a markedly invasive phenotype (P < 0.0001) and near total down-regulation of E-cadherin. In an orthotopic xenograft model, cyclopamine profoundly inhibited metastatic spread; only one of seven cyclopamine-treated mice developed pulmonary micrometastases versus seven of seven mice with multiple macrometastases in control animals. Combination of gemcitabine and cyclopamine completely abrogated metastases while also significantly reducing the size of “primary” tumors. Gli1 levels were up-regulated in tissue samples of metastatic human pancreatic cancer samples compared with matched primary tumors. Aldehyde dehydrogenase (ALDH) overexpression is characteristic for both hematopoietic progenitors and leukemic stem cells; cyclopamine preferentially reduced “ALDH-high” cells by ∼3-fold (P = 0.048). We confirm pharmacologic Hh pathway inhibition as a valid therapeutic strategy for pancreatic cancer and show for the first time its particular efficacy against metastatic spread. By targeting specific cellular subpopulations likely involved in tumor initiation at metastatic sites, Hh inhibitors may provide a new paradigm for therapy of disseminated malignancies, particularly when used in combination with conventional antimetabolites that reduce “bulk” tumor size. [Cancer Res 2007;67(5):2187–96]

Hedgehog pathway overactivity has been implicated in the development of a variety of human cancers. The Human Hedgehog interacting protein (HHIP), a negative regulator of hedgehog signaling, has been shown to be underexpressed in pancreatic cancers. In this study we determined if the HHIP gene is a target for genetic and epigenetic alterations. While no mutations of HHIP were identified, we found complete methylation of the HHIP promoter CpG island in three pancreatic cancer cell lines, and partial hypermethylation in 13/17 (80%) pancreatic cancer cell lines, 35/75 (46%) primary pancreatic cancers and 14/18 (78%) pancreatic cancer xenografts, but no methylation in 13 normal pancreata. In pancreatic cancer cell lines, complete methylation was associated with absent or reduced HHIP expression by real-time RT-PCR. HHIP expression could be restored in methylated cell lines using epigenetic modifier drugs. Restoring the expression of HHIP in pancreatic cancer cells by 5-aza-2'-deoxycytidine led to a decrease in Gli reporter activity, consistent with downregulation of Hedgehog signaling. These results indicate in some pancreatic adenocarcinomas that HHIP is epigenetically inactivated by promoter methylation, and its silencing could contribute to the increased Hedgehog signaling observed in pancreatic neoplasms.

Invasive pancreatic cancer is thought to develop through a series of noninvasive duct lesions known as pancreatic intraepithelial neoplasia (PanIN). We used cDNA microarrays interrogating 15,000 transcripts to identify 49 genes that were differentially expressed in microdissected early PanIN lesions (PanIN-1B/2) compared with microdissected normal duct epithelium. In this analysis, a cluster of extrapancreatic foregut markers, including pepsinogen C, MUC6, KLF4, and TFF1, was found to be up-regulated in PanIN. Up-regulation of these genes was further validated using combinations of real-time reverse transcription-PCR, in situ hybridization, and immunohistochemistry in a total of 150 early PanIN lesions from 81 patients. Identification of these gastrointestinal transcripts in human PanIN prompted assessment of other foregut markers by both semiquantitative and real-time reverse transcription-PCR, revealing similar up-regulation of Sox-2, Gastrin, HoxA5, GATA4/5/6, Villin and Forkhead 6 (Foxl1). In contrast to frequent expression of multiple gastric epithelial markers, the intestinal markers intestinal fatty acid binding protein, CDX1 and CDX2 were rarely expressed either in PanIN lesions or in invasive pancreatic cancer. Hedgehog pathway activation induced by transfection of immortalized human pancreatic ductal epithelial cells with Gli1 resulted in up-regulation of the majority of foregut markers seen in early PanIN lesions. These data show frequent up-regulation of foregut markers in early PanIN lesions and suggest that PanIN development may involve Hedgehog-mediated conversion to a gastric epithelial differentiation program.

Esophageal adenocarcinoma arises through well-defined precursor lesions (Barrett esophagus), although only a subset of these lesions advances to invasive adenocarcinoma. The lack of markers predicting progression in Barrett esophagus, typical presentation at advanced stage, and limitations of conventional chemotherapy result in90% mortality for Barrett-associated adenocarcinomas. To identify potential prognostic markers and therapeutic targets, we compared gene expression profiles from Barrett-associated esophageal adenocarcinoma cell lines (BIC1, SEG1, KYAE, OE33) and normal esophageal epithelial scrapings utilizing the Affymetrix U133_A gene expression platform. We identified 560 transcripts with3-fold up-regulation in the adenocarcinoma cell lines compared with normal epithelium. Utilizing tissue microarrays composed of normal esophageal squamous mucosa (n = 20), Barrett esophagus (n = 10), low-grade dysplasia (n = 14), high-grade dysplasia (n = 27), adenocarcinoma (n = 59), and node metastases (n = 27), we confirmed differential up-regulation of three proteins (Cdc2/Cdk1, Cdc5, and Igfbp3) in adenocarcinomas and Barrett lesions. Protein expression mirrored histologic progression; thus, 87% of low-grade dysplasias had at least focal surface Cdc2/Cdk1 and 20% had5% surface staining; 96% of high-grade dysplasias expressed abundant surface Cdc2/Cdk1, while invasive adenocarcinoma and metastases demonstrated ubiquitous expression. Esophageal adenocarcinoma cell lines treated with the novel CDC2/CDK1 transcriptional inhibitor, tetra-O-methyl nordihydroguaiaretic acid (EM-1421, formerly named M4N) demonstrated a dose-dependent reduction in cell proliferation, paralleling down-regulation of CDC2/CDK1 transcript and protein levels. These findings suggest a role for CDC2/CDK1 in esophageal adenocarcinogenesis, both as a potential histopathologic marker of dysplasia and a putative treatment target.

Background Advancements in experimental therapeutics for esophageal cancers have been hampered by the lack of a reliable preclinical model that recapitulates the biology of human cancer, including in vivo growth in an animal model. Methods Bilious reflux was induced by esophago-jejunostomy in Sprague–Dawley rats. Nine of 12 (75%) Sprague–Dawley rats developed squamous or adenosquamous cancers, and three cell lines were created by in vitro propagation of freshly resected tumors, JA and JB lines from one cancer, and the AMY cell line from another. We subsequently tested the ability of these cell lines to propagate long-term in vitro and form xenografts in vivo, both hallmarks of transformed cells. In addition, we determined the effects of small molecule inhibitors of two important oncogenic pathways—the epidermal growth factor receptor (EGFR) and Hedgehog (Hh) signaling pathways, in vitro, as a “proof of principle” of using these unique cell lines for developing targeted therapies for esophageal cancer. Mechanism-based growth inhibition was assessed by down-regulation of activated downstream targets of EGFR in the case of Iressa, and by Hh luciferase reporter activity with cyclopamine. Results JA, JB, and AMY cell lines were able to grow continuously in vitro and consistently form xenografts in vivo in athymic mice, both subcutaneously, as well as in the “orthotopic” location at the gastroesophageal serosal junction (n = 2 mice per line, six of six engrafted). By histology, the tumors grow in vivo as well-differentiated keratinizing squamous cell carcinomas. JB cells had the highest expression of EGFR protein and also the most profound response to Iressa (gefitinib), an EGFR inhibitor (IC50 m ). Growth inhibition by Iressa was mirrored functionally by down-regulation of activated targets of the EGFR pathway, phospho-ERK1/2 and phospho-MEK levels. AMY cells expressed ∼900-fold elevation of the Hh ligand, Indian Hh (Ihh), compared with normal esophageal epithelium, whereas expression of another Hh ligand, Sonic Hh (Shh), was not detected. On treatment with the specific Hh small molecule inhibitor cyclopamine, AMY cells demonstrated growth inhibition, which was accompanied by significant down-regulation of endogenous Hh luciferase reporter activity at 24 h and increased apoptosis in treated cells. Conclusions We have established a model of esophageal carcinogenesis, capable of long-term in vitro and in vivo passage, and demonstrated therapeutic potential of targeting the EGFR and Hh pathways in the cell lines created from the rodent cancers. These unique cell lines should provide a platform for rapid preclinical validation of novel therapeutics for esophageal cancers.

There are various cellular mediators which can affect the process of blastocyst implantation by regulating the proliferation and differentiation of conceptus and maternal endometrial cells. Insulin-like growth factors I (IGF-I) and II (IGF-II) are potent mitogenic and differentiation-promoting growth factors. However, the role of IGF peptides at implantation in primate species is not well understood. The objective of the present study was to immunohistochemically localize IGF-I and IGF-II peptides in trophoblast cells and maternal endometrial cells during lacunar and villous stages of placentation in the Rhesus monkey. Female animals (n = 10) were laparotomized on estimated days 13-16 after fertilization to collect primary implantation sites which were subjected to immunohistochemical staining for IGF-I and IGF-II peptides. Cell-type specificity for IGF-I and IGF-II was evident with a very low level of IGF-I peptide immunolocalized in trophoblast cells lining lacunae, and primary and secondary villi, while moderate to high amounts of IGF-II peptide were detected in lamellar syncytiotrophoblast cells lining lacunae, early villi and cell columns, as well as in migrating trophoblast cells in the extravillous compartment and in endovascular trophoblast cells. The observed presence of IGF-II peptide in differentiated lamellar syncytiotrophoblast cells during the very early stages of implantation and placentation in the Rhesus monkey may be important in their transition to this differentiated cell population. Maternal endometrial cells showed similar distribution profiles for IGF-I and IGF-II. In conclusion, we report differential distribution of IGF-I and IGF-II peptides in trophoblast cell populations at the feto-maternal interface during lacunar and villous stages of gestation in the Rhesus monkey.

The aim of the present study was to examine the cellular basis of the involvement of oestradiol and progesterone in blastocyst implantation in the primate. To this end, the cellular distribution of receptors for oestradiol (ER) and progesterone (PR) in fetal trophoblast cells and in endometrial compartments of timed lacunar (pre-villous) and villous stages of placentation in primary implantation sites collected on days 13-22 of gestation were investigated in rhesus monkeys. Both in pre-villous stage tissue and in villous stage tissue, cytotrophoblast cells and syncytiotrophoblast cells and other trophoblast derived cells were PR positive, while they were generally ER negative. Maternal endometrial cells were ER negative, while epithelial cells, stromal cells and vascular endothelial cells in maternal endometrium showed heterogeneous staining patterns for PR depending on their relative location; these patterns, however, correlated well with glandular hyperplasia and differentiation, stromal-decidual transformation and vascular response seen during blastocyst implantation.

A targeted therapy with lesser or no toxicity is an immediate need for glioblastoma multiforme (GBM). A group of Hedgehog (Hh) pathway inhibitor drugs is one among the most promising targeted therapies in a wide number of malignancies including medulloblastoma, a lethal childhood central nervous system (CNS) malignancy, about a third of which is presented with high Hh pathway activity. However, in case of GBM a great degree of ambiguity is reported in terms of this pathway activity. GBM is known to be highly heterogeneous at molecular level. It is indeed not clear whether any specific subgroup of GBM shows considerable Hh pathway activity so that could be targeted with appropriate Hh pathway inhibitory drugs. We estimated here the mRNA expression levels of 7 Hh pathway component genes — the ligand Sonic Hedgehog (SHH), trans-membrane receptors Patched (PTCH1) and Smoothened (SMO), Hh transcription factors Gli-1 (GLI1), Gli-2 (GLI2) and Gli-3 (GLI3), and also the Gli-1 target gene Snail (SNAI1), relative to the levels of β-actin expression, by quantitative reverse transcriptase polymerase chain reaction (qRT-PCR) method, in a cohort of GBM patients from the Eastern part of India. We have so far estimated the relative expression levels of these genes in 10 GBM, 8 low grade astrocytoma, 2 meningioma, 2 schwannoma, 1 medulloblastoma biopsy samples and 2 normal cerebellar tissue samples. CNS tumors with astrocytic origin were confirmed by Glial fibrillary acidic protein (GFAP) mRNA expression. We selected only the GBM cases for further analysis in the present study. We calculated the squared Pearson correlation coefficient between expression levels (2–ΔCt) of all the 7 genes and found the best correlations between three of them — SHH, GLI1 and SNAI1 respectively. High correlation of the expression of SHH with that of GLI1 (r2 = 0.9964) is suggestive of ligand-driven mechanism of this pathway activity in this disease. Although both PTCH1 and SNAI1 are bona fide targets of the Hh transcription factor GLI1, interestingly, we observed that the expression of GLI1 had a low correlation with that of PTCH1 (r2 = 0.0472) but was highly correlated with SNAI1 expression (r2 = 0.9986). Moreover, by using k-means clustering analysis three subgroups of patients were identified based on their expression patterns of all 7 Hh pathway component genes that were included in the present study. We have found detectable levels of SMO mRNA in all GBM patients included in the study so far, suggesting a potential benefit of the targeted therapy with smoothened antagonist drugs in this disease. Citation Format: Tapojyoti Das, Uttara Chatterjee, Samarendra Nath Ghosh, Sumit Deb, Suniti Kumar Saha, Puneet Gulati, Sarang Rote, Vikas Chandra, Ankur Mukherjee, Surajit Dhara. An objective clustering of GBM patients to identify clinically relevant subgroup with Hedgehog pathway activity. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 5587. doi:10.1158/1538-7445.AM2013-5587

Proceedings: AACR 104th Annual Meeting 2013; Apr 6-10, 2013; Washington, DC Glioblastoma multiforme (GBM) is the most prevalent central nervous system malignancy portending dismal prognosis. The median overall survival (OS) is 14.6 months with currently available standard care of surgery, radiotherapy and Temozolomide (TMZ) chemotherapy. TMZ, which is the best chemotherapeutic drug till date for this malignancy, gives a median OS advantage of only 2.5 months over surgery and radiotherapy alone. It is not clear how the benefit of TMZ is severely limited in a large majority of patients. To understand this we investigated genome wide sequence alterations of GBM neoplastic cells in response to TMZ. We present here two cases of GBM, [A49910][1] and [M45481][2], where the first one ([A49910][1]) showed response and the second one ([M45481][2]) showed no response to standard treatment in clinic. We isolated their primary tumor cells at the time of surgery and cultured them in vitro as neurospheres. When we exposed these two patient-derived neurospheres to clinically relevant dose of TMZ in vitro differential responses were observed among the two neurospheres which mirrored the clinical outcomes of the two patients respectively. Initially, at the end of 5 days of TMZ treatment, both [A49910][1] and [M45481][2] neurospheres showed 50% and 65% reduction in viable cell numbers respectively but after a 23 days of gap (“post TMZ-treatment recovery” from the 5 days long drug treatment), at 28th day of the treatment cycle, the total number of viable cells was 5% in TMZ-treated [A49910][1] with a stark contrast of 60% in the TMZ-treated [M45481][2] as compared to their respective DMSO-treated controls. Their growth curves, as measured by MTT assays, showed the exact reflection of this pattern, i.e., after 5 days of treatment both TMZ-treated [A49910][1] and TMZ-treated [M45481][2] showed equally retarded growth compared to their DMSO-treated controls, and at 28th day, only the TMZ-treated [A49910][1] but not the TMZ-treated [M45481][2] cells showed growth retardation. Moreover, almost every single cell of TMZ-treated [A49910][1] appeared bigger and stained intensely with SA-βgal confirming drug induced senescence at 28th day whereas this phenotype was completely absent in TMZ-treated [M45481][2], where no apparent sign of senescence was observed. However, apoptosis was two-fold higher in [M45481][2] than in [A49910][1] after 5 days of TMZ treatment. Whether this “drug induced cellular senescence (DICS)” phenotype is more beneficial to the patients claims further investigation with a larger cohort of GBM patients. For the proof of principle we sequenced the whole exome and also analyzed the genome wide single nucleotide polymorphisms (SNPs, intronic and exonic) in these two patient-derived neurospheres before and after TMZ treatment in vitro and detected a list of unique sequence alterations in both the cases, in response to TMZ. Citation Format: Neeta Sarkar Roy, Nidhan Biswas, Vikas Chandra, Tapojyoti Das, Ankita Chatterjee, Rabindra Narayan Bhattacharya, Laxminarayan Tripathy, Sunandan Basu, Arindam Maitra, Pryiadarshi Basu, Analabha Basu, Surajit Dhara. How important is the “post TMZ-treatment recovery” of GBM neoplastic cells. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 3430. doi:10.1158/1538-7445.AM2013-3430 [1]: /lookup/external-ref?link_type=GEN&access_num=A49910&atom=%2Fcanres%2F73%2F8_Supplement%2F3430.atom [2]: /lookup/external-ref?link_type=GEN&access_num=M45481&atom=%2Fcanres%2F73%2F8_Supplement%2F3430.atom

Introduction. Advancements in experimental therapeutics for esophageal cancers have been hampered by the lack of a reliable preclinical model that recapitulates the biology of human cancer, including in vivo growth and metastases. Methods. Bilious reflux was induced by esophago-jejunostomy in Sprague-Dawley rats. Eight of nine SD rats developed squamous or adenosquamous cancers, and three new cell lines were created by in vitro propagation, JA and JB lines from one, and AMY line from the other. We subsequently tested the ability of these cell lines to propagate long-term in vitro and form tumors in vivo. In addition, we determined the effects of small molecule inhibitors of two important oncogenic pathways---the epidermal growth factor receptor (EGFR) and hedgehog (Hh) signaling pathways, in vitro. Relative growth inhibition was measured as described (Berman et al Nature 2003) at 48 and 96 h, using standard MTS assay. Results. JA, JB, and AMY cell lines are able to grow continuously in vitro, with consistent in vivo growth in athymic mice, both as subcutaneous xenografts and as orthotopically implanted tumors at the gastroesophageal serosa (N = 2 mice per line, six of six engrafted each). By histology, the tumors grow in vivo as a well-differentiated keratinizing squamous cell carcinomas. Iressa (gefitinib), an EGFR inhibitor (5 μm), caused significant growth inhibition in all three cell lines (JA > JB > AMY) compared to mock-treated cells, mirrored functionally by down regulation of activated targets of the EGFR pathway, phospho-ERK1/2, and phospho-MEK levels, in the EGFR-inhibited cells. Cyclopamine, an Hh pathway antagonist (dose 6 μm), caused significant growth inhibition in the AMY cell line, but not JA or JB lines; growth inhibition was accompanied by significant down regulation of Hh reporter activity at 24 h in the AMY cells but not JA or JB lines, consistent with specific antagonism of the Hh signaling pathway. Conclusions. We have established a model of esophageal carcinogenesis, with ability for long-term in vitro and in vivo passage and exhibiting pathway-specific antagonistic effects.

Abstract 3969 Poster Board III-905 Because T-cell acute lymphoblastic leukemia (T-ALL) is an aggressive lymphoid leukemia with a high rate of relapse in both children and adults, research is urgently needed to discover molecular pathways that could be targeted in therapy. Recent studies have identified NOTCH1 activating mutations, genomic classifiers, and other molecular pathways that are disrupted in T-ALL, although these findings have not yet translated into better therapy or improved long-term survival. Thus, we are investigating molecular pathways that play a causative role in T-ALL and could be modulated by therapy. We previously showed that HMGA1a is overexpressed in human T-ALL. Further, HMGA1a transgenic mice develop aggressive T-ALL with complete penetrance, indicating that HMGA1a induces leukemogenic transformation in vivo. Disruption of the CDKN2A (INK4A/ARF) tumor suppressor locus has been reported in up to 90% of T-ALL, suggesting that this contributes to leukemic transformation in T-cells. To determine if loss of function of the INK4A/ARF tumor suppressor locus could cooperate with HMGA1a in T-ALL, we crossed our HMGA1a transgenics with mice that are null for the INK4A/ARF locus. We found that T-ALL was markedly accelerated in the HMGA1a transgenics that are also null at this tumor suppressor locus. These mice died significantly earlier than the HMGA1 transgenics that are wildtype at the INK4A/ARF locus (4.8 months+/- 1.0 month versus 10.5 +/- 2.5 months, p < 0.0001 by the student's t-test). The HMGA1a transgenic, INK4A/ARF null mice recapitulate salient clinical and pathologic features of human T-ALL. All mice appeared ill and exhibit marked organomegaly with extensive leukemic infiltration. Immunophenotyping by fluorescence activated cell sorting showed T-ALL in all cases with identical T-cell markers (CD3, CD8, and αβ-TCR) to those previously reported for the HMGA1a transgenics at 8-10 months. Interestingly, the HMGA1a-INK4A/ARF+/- mice also developed early lymphoid malignancy in 6/17 cases. The accelerated lymphoid malignancy in these mice suggests that there is loss of heterozygosity at the INK4A/ARF locus. Alternatively, haploinsufficiency of INK4A and/or ARF may be sufficient to promote lymphoid tumorigenesis in the setting of HMGA1a overexpression. To determine if HMGA1 overexpression and loss of INK4A/ARF expression occur in human T-ALL, we analyzed gene expression profile analysis in leukemic blasts from children and young adults with T-ALL. We found high levels of HMGA1 expression and low levels of INK4A/ARF expression in most patients. Taken together, our results indicate that overexpression of HMGA1 and loss of function at the INK4A/ARF (CDKN2A) tumor suppressor locus cooperate in T-cell leukemogenesis and these pathways could represent rational therapeutic targets. Further, our mice provide a novel preclinical model for T-ALL. Disclosures: No relevant conflicts of interest to declare.

Objectives The reasons for the increasing incidence of esophageal adenocarcinoma are not clear. A causal relation between gastroesophageal reflux disease and esophageal adenocarcinoma has been suggested. Support for this comes from the development of esophageal adenocarcinoma in the rat reflux model. However, to date, no systematic characterization of the tumors derived from this model has been reported. Methods We induced biliary reflux by creating esophagojejunal anastomoses in 12 Sprague–Dawley rats. The experiment was terminated at 9 months, and rat esophagi were harvested for histopathologic documentation of reflux-associated changes and evidence of tumor formation. Three cell lines were established from 2 of the reflux-associated tumors. We tested the ability of these cells to grow in vitro in tissue culture and in vivo as xenografts in an orthotopic location at the gastroesophageal junction. Furthermore, we performed a cytogenetic analysis and determined the array-based gene expression profiles of these 3 rodent carcinoma lines compared with normal esophageal mucosa. Results At 9 months, 12 of 12 rodents had histologic features of metaplastic columnar epithelium in the esophagus, with 7 having invasive carcinomas with glandular differentiation (either adenocarcinomas or adenosquamous carcinomas). The 3 cell lines established from 2 reflux-associated tumors were capable of sustained in vitro propagation and grew successfully as xenografts in both subcutaneous and orthotopic locations, confirming the tumorigenic nature of these lines. Despite their establishment from primary tumors with glandular features, the histology of the xenografts was that of well-differentiated squamous carcinomas. Karyotype analyses demonstrated cytogenetic heterogeneity and aneuploidy; furthermore, translocation (7:11) was present in all 3 lines. Array-based gene expression profiling confirmed upregulation of several cancer-related genes important in human esophageal cancer. Quantitative reverse transcription–polymerase chain reaction was used to confirm the differential expression of selected transcripts (vascular endothelial growth factor [ VEGF ], polo-like kinases [ PLK ], cyclin dependent kinase 4 [ CDK4 ], hypoxia-inducible factor 1α [ HIF1α ], and insulin-like growth factor 1 [ IGF-1 ]) in comparison with nonneoplastic esophageal mucosal scrapings. Conclusions The rodent reflux model is capable of inducing metaplastic epithelial changes simulating Barrett esophagus, as well as subsequent neoplastic transformation, at a high frequency. Cell lines have been established from these tumors that are capable of in vitro and in vivo passaging. The rodent reflux model should be a valuable model for studying therapy and chemoprevention efforts for Barrett esophagus, whereas the established cell lines provide a useful resource for drug discovery and other high-throughput studies.

While the oncogenic properties of the high mobility group A1 (HMGA1) gene are well established, the molecular pathways that mediate transformation by HMGA1 have not been clearly defined. HMGA1 is widely overexpressed in hematologic malignancies and other human cancers. Moreover, its overexpression portends a poor prognosis in some tumors. The HMGA1 gene encodes the HMGA1a and -A1b chromatin binding proteins. We previously showed that overexpression of HMGA1a or -A1b confers a transformed phenotype in cultured, human lymphoid cells. More recently, we developed transgenic mice overexpressing HMGA1a and all mice develop aggressive lymphoid malignancy. Because HMGA1 proteins function in regulating gene expression, we hypothesize that overexpression of HMGA1 leads to neoplastic transformation by altering expression of specific target genes involved in this process. To identify genes that are regulated by HMGA1 in transformation, we used microarray gene expression profile analysis in cells transfected to overexpress HMGA1a or -A1b. Ninety-seven genes were differentially regulated by ≥ 2-fold in cells overexpressing HMGA1a and 103 genes were differentially regulated by ≥ 2-fold in cells overexpressing HMGA1b. Fifty-five of these genes were common to both the HMGA1a and -A1b cells. From the list of genes up-regulated in cells overexpressing HMGA1a or A1b, the gene encoding the signaling molecule, signal transducer and activator of transcription or STAT3, was studied further because of its prominent role in malignancy. HMGA1 binds to the STAT3 promoter in human leukemia cells in vivo by chromatin immunoprecipitation experiments and enhances transcription of the STAT3 promoter in transfection experiments. Both total STAT3 and activated STAT3 proteins are increased in leukemia cells from HMGA1a transgenic mice compared to lymphoid cells from control mice. Moreover, activated STAT3 recapitulates the transforming activity of HMGA1a in cultured cells. Blocking STAT3 function preferentially induces apoptosis in leukemia cells from HMGA1a transgenic mice, but not in control cells. STAT3 mRNA is also increased in human acute lymphoid leukemia samples that overexpress HMGA1a, suggesting that this pathway is also involved in human lymphoid leukemia. Our results indicate that STAT3 is a direct HMGA1a target gene that is both necessary and sufficient for transformation induced by HMGA1a.