Hongying Hao

Identifying mRNA, MicroRNA and Protein Profiles of Melanoma Exosomes

Background Exosomes are small membranous vesicles secreted into body fluids by multiple cell types, including tumor cells, and in various disease conditions. Tumor exosomes contain intact and functional mRNAs, small RNAs (including miRNAs), and proteins that can alter the cellular environment to favor tumor growth. Molecular profiling may increase our understanding of the role of exosomes in melanoma progression and may lead to discovery of useful biomarkers. Methodology/Principal Findings In the present study, we used mRNA array profiling to identify thousands of exosomal mRNAs associated with melanoma progression and metastasis. Similarly, miRNA array profiling identified specific miRNAs, such as hsa-miR-31, -185, and -34b, involved in melanoma invasion. We also used proteomic analysis and discovered differentially expressed melanoma exosomal proteins, including HAPLN1, GRP78, syntenin-1, annexin A1, and annexin A2. Importantly, normal melanocytes acquired invasion ability through molecules transported in melanoma cell-derived exosomes. Conclusions/Significance Our results indicate that melanoma-derived exosomes have unique gene expression signatures, miRNA and proteomics profiles compared to exosomes from normal melanocytes. To the best of our knowledge, this is the first in-depth screening of the whole transcriptome/miRNome/proteome expression in melanoma exosomes. These results provide a starting point for future more in-depth studies of tumor-derived melanoma exosomes, which will aid our understanding of melanoma biogenesis and new drug-targets that may be translated into clinical applications, or as non-invasive biomarkers for melanoma.

Adenovirus E1B55K Region Is Required To Enhance Cyclin E Expression for Efficient Viral DNA Replication

ABSTRACT Adenoviruses (Ads) with E1B55K mutations can selectively replicate in and destroy cancer cells. However, the mechanism of Ad-selective replication in tumor cells is not well characterized. We have shown previously that expression of several cell cycle-regulating genes is markedly affected by the Ad E1b gene in WI-38 human lung fibroblast cells (X. Rao, et al., Virology 350:418-428, 2006). In the current study, we show that the Ad E1B55K region is required to enhance cyclin E expression and that the failure to induce cyclin E overexpression due to E1B55K mutations prevents viral DNA from undergoing efficient replication in WI-38 cells, especially when the cells are arrested in the G0 phase of the cell cycle by serum starvation. In contrast, cyclin E induction is less dependent on the function encoded in the E1B55K region in A549 and other cancer cells that are permissive for replication of E1B55K-mutated viruses, whether the cells are in the S phase or G0 phase. The small interfering RNA that specifically inhibits cyclin E expression partially decreased viral replication. Our study provides evidence suggesting that E1B55K may be involved in cell cycle regulation that is important for efficient viral DNA replication and that cyclin E overexpression in cancer cells may be associated with the oncolytic replication of E1B55K-mutated viruses.

Melanoma cell-derived exosomes promote epithelial-mesenchymal transition in primary melanocytes through paracrine/autocrine signaling in the tumor microenvironment.

The tumor microenvironment is abundant with exosomes that are secreted by the cancer cells themselves. Exosomes are nanosized, organelle-like membranous structures that are increasingly being recognized as major contributors in the progression of malignant neoplasms. A critical element in melanoma progression is its propensity to metastasize, but little is known about how melanoma cell-derived exosomes modulate the microenvironment to optimize conditions for tumor progression and metastasis. Here, we provide evidence that melanoma cell-derived exosomes promote phenotype switching in primary melanocytes through paracrine/autocrine signaling. We found that the mitogen-activated protein kinase (MAPK) signaling pathway was activated during the exosome-mediated epithelial-to-mesenchymal transition (EMT)-resembling process, which promotes metastasis. Let-7i, an miRNA modulator of EMT, was also involved in this process. We further defined two other miRNA modulators of EMT (miR-191 and let-7a) in serum exosomes for differentiating stage I melanoma patients from non-melanoma subjects. These results provide the first strong molecular evidence that melanoma cell-derived exosomes promote the EMT-resembling process in the tumor microenvironment. Thus, novel strategies targeting EMT and modulating the tumor microenvironment may emerge as important approaches for the treatment of metastatic melanoma.

E2F-1 induces melanoma cell apoptosis via PUMA up-regulation and Bax translocation

BackgroundPUMA is a pro-apoptotic Bcl-2 family member that has been shown to be involved in apoptosis in many cell types. We sought to ascertain whether induction of PUMA plays a crucial role in E2F-1-induced apoptosis in melanoma cells.MethodsPUMA gene and protein expression levels were detected by real-time PCR and Western blot in SK-MEL-2 and HCT116 cell lines after Ad-E2F-1 infection. Activation of the PUMA promoter by E2F-1 overexpression was detected by dual luciferase reporter assay. E2F-1-induced Bax translocation was shown by immunocytochemistry. The induction of caspase-9 activity was measured by caspase-9 colorimetric assay kit.ResultsUp-regulation of the PUMA gene and protein by E2F-1 overexpression was detected by real-time PCR and Western blot analysis in the SK-MEL-2 melanoma cell line. In support of this finding, we found six putative E2F-1 binding sites within the PUMA promoter. Subsequent dual luciferase reporter assay showed that E2F-1 expression could increase the PUMA gene promoter activity 9.3 fold in SK-MEL-2 cells. The role of PUMA in E2F-1-induced apoptosis was further investigated in a PUMA knockout cell line. Cell viability assay showed that the HCT116 PUMA-/- cell line was more resistant to Ad-E2F-1-mediated cell death than the HCT116 PUMA+/+ cell line. Moreover, a 2.2-fold induction of the PUMA promoter was also noted in the HCT116 PUMA+/+ colon cancer cell line after Ad-E2F-1 infection. Overexpression of a truncated E2F-1 protein that lacks the transactivation domain failed to up-regulate PUMA promoter, suggesting that PUMA may be a transcriptional target of E2F-1. E2F-1-induced cancer cell apoptosis was accompanied by Bax translocation from the cytosol to mitochondria and the induction of caspase-9 activity, suggesting that E2F-1-induced apoptosis is mediated by PUMA through the cytochrome C/Apaf-1-dependent pathway.ConclusionOur studies strongly demonstrated that E2F-1 induces melanoma cell apoptosis via PUMA up-regulation and Bax translocation. The signaling pathways provided here will further enhance insights on the mechanisms of E2F-1-induced cancer cell apoptosis as a strategy for cancer therapy.

Adenovirus-Mediated Expression of Truncated E2F-1 Suppresses Tumor Growth In Vitro and In Vivo

Adenovirus (Ad)‐mediated E2F‐1 gene transfer induces apoptosis in cancer cells in vitro and in vivo, but clinical application of E2F‐1 in cancer gene therapy remains controversial because of the oncogenic potential of E2F‐1. This barrier can be circumvented by using the truncated form of the E2F‐1 gene (E2Ftr) (amino acids 1 through 375), which lacks the E2F‐1 transactivation domain and cell cycle‐promoting effects.

E2F-1- and E2Ftr-mediated apoptosis: the role of DREAM and HRK.

E2F‐1‐deleted mutant, ‘truncated E2F’ (E2Ftr, E2F‐1[1–375]), lacking the carboxy‐terminal transactivation domain, was shown to be more potent at inducing cancer cell apoptosis than wild‐type E2F‐1 (wtE2F‐1; full‐length E2F‐1). Mechanisms by which wtE2F‐1 and E2Ftr induce apoptosis, however, are not fully elucidated. Our study demonstrates molecular effects of pro‐apoptotic BH3‐only Bcl‐2 family member Harakiri (Hrk) in wtE2F‐1‐ and E2Ftr‐induced melanoma cell apoptosis. We found that Hrk mRNA and Harakiri (HRK) protein expression was highly up‐regulated in melanoma cells in response to wtE2F‐1 and E2Ftr overexpression. HRK up‐regulation did not require the E2F‐1 transactivation domain. In addition, Hrk gene up‐regulation and HRK protein expression did not require p53 in cancer cells. Hrk knockdown by Hrk siRNA was associated with significantly reduced wtE2F‐1‐ and E2Ftr‐induced apoptosis. We also found that an upstream factor, ‘downstream regulatory element antagonist modulator’ (DREAM), may be involved in HRK‐mediated apoptosis in response to wtE2F‐1 and E2Ftr overexpression. DREAM expression levels increased following wtE2F‐1 and E2Ftr overexpression. Western blotting detected increased DREAM primarily in dimeric form. The homodimerization of DREAM resulting from wtE2F‐1 and E2Ftr overexpression may contribute to the decreased binding activity of DREAM to the 3′‐untranslated region of the Hrk gene as shown by electromobility shift assay. Results showed wtE2F‐1‐ and E2Ftr‐induced apoptosis is partially mediated by HRK. HRK function is regulated in response to DREAM. Our findings contribute to understanding the mechanisms that regulate wtE2F‐1‐ and E2Ftr‐induced apoptosis and provide insights into the further evaluation of how E2Ftr‐induced apoptosis may be used for therapeutic gain.

Developing adenoviral vectors encoding therapeutic genes toxic to host cells: Comparing binary and single-inducible vectors expressing truncated E2F-1

Adenoviral vectors are highly efficient at transferring genes into cells and are broadly used in cancer gene therapy. However, many therapeutic genes are toxic to vector host cells and thus inhibit vector production. The truncated form of E2F-1 (E2Ftr), which lacks the transactivation domain, can significantly induce cancer cell apoptosis, but is also toxic to HEK-293 cells and inhibits adenovirus replication. To overcome this, we have developed binary- and single-vector systems with a modified tetracycline-off inducible promoter to control E2Ftr expression. We compared several vectors and found that the structure of expression cassettes in vectors significantly affects E2Ftr expression. One construct expresses high levels of inducible E2Ftr and efficiently causes apoptotic cancer cell death by activation of caspase-3. The approach developed in this study may be applied in other viral vectors for encoding therapeutic genes that are toxic to their host cells and/or inhibit vector propagation.

Induction of Apoptosis Signal-Regulating Kinase 1 by E2F-1 May Not Be Essential for E2F-1-Mediated Apoptosis in Melanoma Cells

Objectives: In the present study, we investigate the role of apoptosis signal-regulating kinase 1 (ASK1) mitogen-activated protein (MAP) kinase signal pathways in E2F-1-mediated apoptosis. Methods: A gene expression profile in response to E2F-1 overexpression was performed by cDNA microarray analysis and confirmed by real-time reverse-transcription polymerase chain reaction. Kinase activities were assayed by Western blot analysis or kinase assay. Apoptosis was assessed by morphologic inspection and flow-cytometric analysis. Cytotoxicity was monitored by MTT assay. Results:E2F-1 upregulated the expression of ASK1 8-fold compared to the Ad-LacZ-infected control in SK-MEL-2 melanoma cells, which was confirmed by reverse-transcription polymerase chain reaction. Sequence analysis showed that there are 2 putative E2F-1 DNA binding sites in the ASK1 promoter region. Truncated E2F-1 protein, which lacks the transactivation domain, failed to upregulate ASK1, suggesting that ASK1 was regulated at the transcriptional level by E2F-1. E2F-1 overexpression resulted in the transient activation of c-Jun N-terminal kinase (JNK); however, dominant negative mutant ASK1 had no effect on E2F-1 cytotoxicity and JNK activation. p38 was not activated by E2F-1, and inhibition of p38 had no effect on E2F-1-mediated cell death. The ASK1 kinase assay showed that ASK1 activity was not upregulated in response to E2F1 overexpression. The inhibition of ASK1 upstream kinase-AKT can enhance E2F-1-mediated cell death. Moreover, an adenovirus expressing truncated E2F-1 keeps the ability of inducing apoptosis in melanoma cells. Conclusions:ASK1 expression is upregulated by E2F-1 at the transcription level, but the upregulation of ASK1 expression by E2F-1 was not coordinated with an increased ASK1 activity. The ASK1-JNK/p38 pathway does not appear to play a crucial role in E2F-1-induced apoptosis.

Chemosensitization of Tumor Cells: Inactivation of Nuclear Factor-Kappa B Associated with Chemosensitivity in Melanoma Cells After Combination Treatment with E2F-1 and Doxorubicin

Combination chemotherapy has been shown to be more effective than single-agent therapy for many types of cancer, but both are known to induce drug resistance in cancer cells. Two major culprits in the development of this drug resistance are nuclear factor-kappaB (NF-kappaB) and the multidrug resistance (MDR) gene. For this reason, chemogene therapy is emerging as a viable alternative to conventional chemotherapy combinations. We have shown that transduction of the E2F-1 gene in melanoma cells markedly increases cell sensitivity to doxorubicin, thereby producing a synergistic effect on melanoma cell apoptosis. Our microarray results show that the NF-kappaB pathway and related genes undergo significant changes after the combined treatment of E2F-1 and doxorubicin. In fact, inactivation of NF-kappaB is associated with melanoma cell apoptosis induced by E2F-1 and doxorubicin, providing a link between the NF-kappaB signaling pathway and the chemosensitivity of melanoma cells after this treatment.

Unique Genes in Tumor-Positive Sentinel Lymph Nodes Associated with Nonsentinel Lymph Node Metastases in Melanoma

BackgroundCurrent risk assessment tools to estimate the risk of nonsentinel lymph node metastases after completion lymphadenectomy for a positive sentinel lymph node (SLN) biopsy in cutaneous melanoma are based on clinical and pathologic factors. We identified a novel genetic signature that can predict non-SLN metastases in patients with cutaneous melanoma staged with a SLN biopsy.MethodsRNA was collected for tumor-positive SLNs in patients staged by SLN biopsy for cutaneous melanoma. All patients with a tumor-positive SLN biopsy underwent completion lymphadenectomy. A 1:10 case:control series of positive and negative non-SLN patients was analyzed by microarray and quantitative RT-PCR. Candidate differentially expressed genes were validated in a 1:3 case:control separate cohort of positive and negative non-SLN patients.ResultsThe 1:10 case:control discovery set consisted of 7 positive non-SLN cases matched to 70 negative non-SLN controls. The cases and controls were similar with regards to important clinicopathologic factors, such as gender, primary tumor site, age, ulceration, and thickness. Microarray and RT-PCR identified six potential differentially expressed genes for validation. In the 40-patient separate validation set, 10 positive non-SLN patients were matched to 30 negative non-SLN controls based on gender, ulceration, age, and thickness. Five of the six genes were differentially expressed. The five gene panel identified patients at low (7.1%) and high risk (66.7%) for non-SLN metastases.ConclusionsA novel, non-SLN gene score based on differential expressed genes in a tumor-positive SLN can identify patients at high and low risk for non-SLN metastases.