Targeting autophagy by small molecule inhibitors of vacuolar protein sorting 34 (Vps34) improves the sensitivity of breast cancer cells to Sunitinib.

Resistance to chemotherapy is a challenging problem for treatment of cancer patients and autophagy has been shown to mediate development of resistance. In this study we systematically screened a library of 306 known anti-cancer drugs for their ability to induce autophagy using a cell-based assay. 114 of the drugs were classified as autophagy inducers; for 16 drugs, the cytotoxicity was potentiated by siRNA-mediated knock-down of Atg7 and Vps34. These drugs were further evaluated in breast cancer cell lines for autophagy induction, and two tyrosine kinase inhibitors, Sunitinib and Erlotinib, were selected for further studies. For the pharmacological inhibition of autophagy, we have characterized here a novel highly potent selective inhibitor of Vps34, SB02024. SB02024 blocked autophagy in vitro and reduced xenograft growth of two breast cancer cell lines, MDA-MB-231 and MCF-7, in vivo. Vps34 inhibitor significantly potentiated cytotoxicity of Sunitinib and Erlotinib in MCF-7 and MDA-MB-231 in vitro in monolayer cultures and when grown as multicellular spheroids. Our data suggests that inhibition of autophagy significantly improves sensitivity to Sunitinib and Erlotinib and that Vps34 is a promising therapeutic target for combination strategies in breast cancer.

Expression of the p53 target Wig-1 is associated with HPV status and patient survival in cervical carcinoma.

The p53 target gene WIG-1 (ZMAT3) is located in chromosomal region 3q26, that is frequently amplified in human tumors, including cervical cancer. We have examined the status of WIG-1 and the encoded Wig-1 protein in cervical carcinoma cell lines and tumor tissue samples. Our analysis of eight cervical cancer lines (Ca Ski, ME-180, MS751, SiHa, SW756, C-4I, C-33A, and HT-3) by spectral karyotype, comparative genomic hybridization and Southern blotting revealed WIG-1 is not the primary target for chromosome 3 gains. However, WIG-1/Wig-1 were readily expressed and WIG-1 mRNA expression was higher in the two HPV-negative cervical cell lines (C33-A, HT-3) than in HPV-positive lines. We then assessed Wig-1 expression by immunohistochemistry in 38 cervical tumor samples. We found higher nuclear Wig-1 expression levels in HPV-negative compared to HPV positive cases (p = 0.002) and in adenocarcinomas as compared to squamous cell lesions (p<0.0001). Cases with moderate nuclear Wig-1 staining and positive cytoplasmic Wig-1 staining showed longer survival than patients with strong nuclear and negative cytoplasmic staining (p = 0.042). Nuclear Wig-1 expression levels were positively associated with age at diagnosis (p = 0.023) and histologic grade (p = 0.034). These results are consistent with a growth-promoting and/or anti-cell death function of nuclear Wig-1 and suggest that Wig-1 expression can serve as a prognostic marker in cervical carcinoma.

CDK-mediated activation of the SCF(FBXO) (28) ubiquitin ligase promotes MYC-driven transcription and tumourigenesis and predicts poor survival in breast cancer.

SCF (Skp1/Cul1/F-box) ubiquitin ligases act as master regulators of cellular homeostasis by targeting key proteins for ubiquitylation. Here, we identified a hitherto uncharacterized F-box protein, FBXO28 that controls MYC-dependent transcription by non-proteolytic ubiquitylation. SCF(FBXO28) activity and stability are regulated during the cell cycle by CDK1/2-mediated phosphorylation of FBXO28, which is required for its efficient ubiquitylation of MYC and downsteam enhancement of the MYC pathway. Depletion of FBXO28 or overexpression of an F-box mutant unable to support MYC ubiquitylation results in an impairment of MYC-driven transcription, transformation and tumourigenesis. Finally, in human breast cancer, high FBXO28 expression and phosphorylation are strong and independent predictors of poor outcome. In conclusion, our data suggest that SCF(FBXO28) plays an important role in transmitting CDK activity to MYC function during the cell cycle, emphasizing the CDK-FBXO28-MYC axis as a potential molecular drug target in MYC-driven cancers, including breast cancer.

TRIM13 (RFP2) downregulation decreases tumour cell growth in multiple myeloma through inhibition of NF Kappa B pathway and proteasome activity.

Multiple myeloma (MM) is an incurable neoplasm caused by proliferation of malignant plasma cells in the bone marrow (BM). MM is characterized frequently by a complete or partial deletion of chromosome 13q14, seen in more than 50% of patients at diagnosis. Within this deleted region the tripartite motif containing 13 (TRIM13, also termed RFP2) gene product has been proposed to be a tumour suppressor gene (TSG). Here, we show that low expression levels of TRIM13 in MM are associated with chromosome 13q deletion and poor clinical outcome. We present a functional analysis of TRIM13 using a loss-of-function approach, and demonstrate that TRIM13 downregulation decreases tumour cell survival as well as cell cycle progression and proliferation of MM cells. In addition, we provide evidence for the involvement of TRIM13 downregulation in inhibiting the NF kappa B pathway and the activity of the 20S proteasome. Although this data does not support a role of TRIM13 as a TSG, it substantiates important roles of TRIM13 in MM tumour survival and proliferation, underscoring its potential role as a novel target for therapeutic intervention.

MiR-200c regulates Noxa expression and sensitivity to proteasomal inhibitors.

The pro-apoptotic p53 target Noxa is a BH3-only protein that antagonizes the function of selected anti-apoptotic Bcl-2 family members. While much is known regarding the transcriptional regulation of Noxa, its posttranscriptional regulation remains relatively unstudied. In this study, we therefore investigated whether Noxa is regulated by microRNAs. Using a screen combining luciferase reporters, bioinformatic target prediction analysis and microRNA expression profiling, we identified miR-200c as a negative regulator of Noxa expression. MiR-200c was shown to repress basal expression of Noxa, as well as Noxa expression induced by various stimuli, including proteasomal inhibition. Luciferase reporter experiments furthermore defined one miR-200c target site in the Noxa 3'UTR that is essential for this direct regulation. In spite of the miR-200c:Noxa interaction, miR-200c overexpression led to increased sensitivity to the clinically used proteasomal inhibitor bortezomib in several cell lines. This apparently contradictory finding was reconciled by the fact that in cells devoid of Noxa expression, miR-200c overexpression had an even more pronounced positive effect on apoptosis induced by proteasomal inhibition. Together, our data define miR-200c as a potentiator of bortezomib-induced cell death. At the same time, we show that miR-200c is a novel negative regulator of the pro-apoptotic Bcl-2 family member Noxa.

MiRNA-27a controls FBW7/hCDC4-dependent cyclin E degradation and cell cycle progression.

The F-box protein FBW7/hCDC4 is a tumor suppressor that acts as the substrate recognition component of an SCF ubiquitin ligase that targets numerous oncoproteins for proteasomal degradation. In this study, we investigated whether FBW7 is regulated by microRNAs, using a screen combining bioinformatic analysis, luciferase reporters and microRNA libraries. The ubiquitous miR-27a was identified as a major suppressor of FBW7 and in line with this, miR-27a prohibited ubiquitylation and turnover of the key FBW7 substrate cyclin E. Notably, we found that miR-27a only suppresses FBW7 during specific cell cycle phases, relieving its negative impact at the G1 to S-phase transition, prior to cyclin E protein degradation. We also demonstrate that attenuation of FBW7 by miR-27a overexpression leads to improper cell cycle progression and DNA replication stress, consistent with dysregulation of cyclin E expression. Finally, in the context of human cancer, miR-27a was discovered to be generally overexpressed in pediatric B-ALL and its expression to be inversely correlated with that of FBW7 in hyperdiploid cases of B-ALL. These data provide evidence for microRNA-mediated regulation of FBW7, and highlight the role of miR-27a as a novel factor fine-tuning the periodic events regulating cell cycle progression.

DLEU2, frequently deleted in malignancy, functions as a critical host gene of the cell cycle inhibitory microRNAs miR-15a and miR-16-1.

The microRNAs miR-15a and miR-16-1 are downregulated in multiple tumor types and are frequently deleted in chronic lymphocytic leukemia (CLL), myeloma and mantle cell lymphoma. Despite their abundance in most cells the transcriptional regulation of miR-15a/16-1 remains unclear. Here we demonstrate that the putative tumor suppressor DLEU2 acts as a host gene of these microRNAs. Mature miR-15a/miR-16-1 are produced in a Drosha-dependent process from DLEU2 and binding of the Myc oncoprotein to two alterative DLEU2 promoters represses both the host gene transcript and levels of mature miR-15a/miR-16-1. In line with a functional role for DLEU2 in the expression of the microRNAs, the miR-15a/miR-16-1 locus is retained in four CLL cases that delete both promoters of this gene and expression analysis indicates that this leads to functional loss of mature miR-15a/16-1. We additionally show that DLEU2 negatively regulates the G1 Cyclins E1 and D1 through miR-15a/miR-16-1 and provide evidence that these oncoproteins are subject to miR-15a/miR-16-1-mediated repression under normal conditions. We also demonstrate that DLEU2 overexpression blocks cellular proliferation and inhibits the colony-forming ability of tumor cell lines in a miR-15a/miR-16-1-dependent way. Together the data illuminate how inactivation of DLEU2 promotes cell proliferation and tumor progression through functional loss of miR-15a/miR-16-1.

The RBCC gene RFP2 (Leu5) encodes a novel transmembrane E3 ubiquitin ligase involved in ERAD.

RFP2, a gene frequently lost in various malignancies, encodes a protein with RING finger, B-box, and coiled-coil domains that belongs to the RBCC/TRIM family of proteins. Here we demonstrate that Rfp2 is an unstable protein with auto-polyubiquitination activity in vivo and in vitro, implying that Rfp2 acts as a RING E3 ubiquitin ligase. Consequently, Rfp2 ubiquitin ligase activity is dependent on an intact RING domain, as RING deficient mutants fail to drive polyubiquitination in vitro and are stabilized in vivo. Immunopurification and tandem mass spectrometry enabled the identification of several putative Rfp2 interacting proteins localized to the endoplasmic reticulum (ER), including valosin-containing protein (VCP), a protein indispensable for ER-associated degradation (ERAD). Importantly, we also show that Rfp2 regulates the degradation of the known ER proteolytic substrate CD3-delta, but not the N-end rule substrate Ub-R-YFP (yellow fluorescent protein), establishing Rfp2 as a novel E3 ligase involved in ERAD. Finally, we show that Rfp2 contains a C-terminal transmembrane domain indispensable for its localization to the ER and that Rfp2 colocalizes with several ER-resident proteins as analyzed by high-resolution immunostaining. In summary, these data are all consistent with a function for Rfp2 as an ERAD E3 ubiquitin ligase.

Loss of chromosome 13q is a frequently acquired event in genetic progression of soft tissue sarcomas in the abdominal cavity.

Soft tissue sarcomas (STSs) arising in the abdominal cavity constitute a group of aggressive tumours, typically of very large size and with a high recurrence rate in the affected patients. While some distinct genetic etiologies have been described, the genetic background of this tumour group is not well characterised. Here we have assessed gross chromosomal alterations in a series of such tumours obtained from 26 patients. CGH alterations were found in tumours from 23 of the patients (88%), the most frequent being loss of 13q21 (46%) and gain of 17p and/or q (46%). Furthermore, mutations of C-KIT exon 11 were demonstrated in five tumours from four patients, and the two myxoid liposarcomas exhibited a translocation t(12;16)(q13;p11). From the pattern of chromosomal alterations detected, a genetic progression of events was clearly evident in the tumours. Taken together with analysis of subsequent relapses from the same patients, the common CGH alteration +12q13 was suggested to be a relatively early event in the genetic progression, similar to t(12;16)(q13;p11) and C-KIT mutations. Moreover, -1p21-22, -13q21, -14q, -Xp22, +9q34, +17p, +17q, and +20q13 would all represent relative later events. The most consistent alteration was loss of 13q, that was found to target the 13q14-21 and 13q34 regions as determined by CGH and Southern blot analyses. Loss of 13q was identified independently of +12q13 and C-KIT mutation and the patient's sex, and was observed in all common subtypes of STS, suggesting that it is a general and late event in the genetic progression. The findings provide a starting point for further dissection of the target genes involved in development of STSs in the abdominal cavity.

Disruption of a novel ectodermal neural cortex 1 antisense gene, ENC-1AS and identification of ENC-1 overexpression in hairy cell leukemia.

Karyotypical alteration of chromosome 5 and in particular band 5q13 is a frequent finding in hairy cell leukemia (HCL). We have previously identified a number of candidate genes localized in close proximity to a constitutional inv(5)(p13.1q13.3) breakpoint in one HCL patient. These included beta-hexosaminodase HEXB, frequently mutated in the lysosomal storage disorder Sandhoff disease. We now report that the 5q13.3 breakpoint disrupts a novel evolutionary conserved alternative isoform of HEXB. This isoform directly overlaps, in a cis-antisense fashion, exon 1 of the gene for ectodermal neuronal cortex 1 ENC-1, and was thus named ENC-1AS. ENC-1 has previously been shown to be overexpressed in several malignancies, and is believed to play a critical regulatory role in malignant transformation of various tumors. Importantly, subsequent analysis of ENC-1 in purified primary HCL tumor cells revealed a striking upregulation of ENC-1 in all 26 patients examined, compared with normal peripheral blood lymphocytes from healthy donors. Upon further analysis of the ENC-1/ENC-1AS locus, we identified a complex 5' regulatory mechanism involving an inverse expression of the ENC-1 sense and the ENC-1AS transcripts in several tissues supporting the hypothesis that expression of ENC-1AS regulates ENC-1 levels. In addition, we have also found tissue-specific methylation of a 1.2 kb segment encompassing the overlapping ENC-1/ENC-1AS 5' exons, adding to the complexity of the regulation of this locus. Altogether, these results suggest that upregulation of ENC-1 contributes to the development of HCL and provides new information on the possible dysregulation of ENC-1 including expression of a novel antisense gene, ENC-1AS.

DLEU2 encodes an antisense RNA for the putative bicistronic RFP2/LEU5 gene in humans and mouse.

Our group previously identified two novel genes, RFP2/LEU5 and DLEU2, within a 13q14.3 genomic region of loss seen in various malignancies. However, no specific inactivating mutations were found in these or other genes in the vicinity of the deletion, suggesting that a nonclassical tumor-suppressor mechanism may be involved. Here, we present data showing that the DLEU2 gene encodes a putative noncoding antisense RNA, with one exon directly overlapping the first exon of the RFP2/LEU5 gene in the opposite orientation. In addition, the RFP2/LEU5 transcript can be alternatively spliced to produce either several monocistronic transcripts or a putative bicistronic transcript encoding two separate open-reading frames, adding to the complexity of the locus. The finding that these gene structures are conserved in the mouse, including the putative bicistronic RFP2/LEU5 transcript as well as the antisense relationship with DLEU2, further underlines the significance of this unusual organization and suggests a biological function for DLEU2 in the regulation of RFP2/LEU5.