Biochemical pathways mediated by KLK6 protease in breast cancer.

Kallikrein-related peptidase 6 (KLK6) is a serine protease normally expressed in mammary tissue and aberrantly regulated in breast cancer. At physiological levels, KLK6 functions as a suppressor of breast cancer, while its aberrant overexpression (> 50-fold higher than normal) is characteristic of a subset of breast cancers and has been linked to accelerated growth of primary breast tumors in severe combined immunodeficiency mice (Pampalakis et al. Cancer Res 2009, 69, 3779). Here, we investigated the molecular mechanisms underlying the concentration-dependent functions of KLK6 by comparing MDA-MB-231 stable transfectants expressing increasing levels of KLK6 in in vitro and in vivo tumorigenicity assays (soft agar, xenograft growth, tail vein metastasis). Quantitative proteomics was applied to identify proteins that are altered upon re-expression of KLK6 in MDA-MB-231 at normal or constitutive levels. Overexpression of KLK6 is associated with increased metastatic ability of breast cancer cells into lungs, increased expression of certain S100 proteins (S100A4, S100A11) and keratins (KRT), and downregulation of the apoptosis-related proteases CASP7 and CASP8, and RABs. On the other hand, KLK6 re-expression at physiological levels leads to inhibition of lung metastases associated with suppression of S100 proteins (S100A4, S100A10, S100A13, S100A16) and induced CASP7 and CASP8 expression. As this is the first report that KLK6 expression is associated with S100 proteins, caspases, RABs, and KRTs, we validated this finding in clinical datasets. By integrating proteomics and microarray data from breast cancer patients, we generated two composite scores, KLK6 + S100B-S100A7 and KLK6 + S100B-S100A14-S100A16, to predict long-term survival of breast cancer patients. We present previously unknown pathways implicating KLK6 in breast cancer. The findings promise to aid our understanding of the functional roles of KLK6 in breast cancer and may yield new biomarkers for the cancer types in which KLK6 is known to be aberrantly upregulated.

Insulin inhibition of apolipoprotein B mRNA translation is mediated via the PI-3 kinase/mTOR signaling cascade but does not involve internal ribosomal entry site (IRES) initiation.

Although insulin normally activates global mRNA translation, it has a specific inhibitory effect on translation of apolipoprotein B (apoB) mRNA. This suggests that insulin induces a unique signaling cascade that leads to specific inhibition of apoB mRNA translation despite global translational stimulation. Recent studies have revealed that insulin functions to regulate apoB mRNA translation through a mechanism involving the apoB mRNA 5' untranslated region (5' UTR). Here, we further investigate the role of downstream insulin signaling molecules on apoB mRNA translation, and the mechanism of apoB mRNA translation itself. Transfection studies in HepG2 cells expressing deletion constructs of the apoB 5' UTR showed that the cis-acting region responding to insulin was localized within the first 64 nucleotides. Experiments using chimeric apoB UTR-luciferase constructs transfected into HepG2 cells followed by treatment with wortmannin, a PI-3K inhibitor, and rapamycin, an mTOR inhibitor, showed that signaling via PI-3K and mTOR pathways is necessary for insulin-mediated inhibition of chimeric 5' UTR-luciferase expression. In vitro translation of chimeric cRNA confirmed that the effects observed were translational in nature. Furthermore, using RNA-EMSA we found that wortmannin pretreatment blocked insulin-mediated inhibition of the binding of RNA-binding factor(s), migrating near the 110 kDa marker, to the 5' UTR. Radiolabeling studies in HepG2 cells also showed that insulin-mediated control of the synthesis of endogenously expressed full length apoB100 is mediated via the PI-3K and mTOR pathways. Finally, using dual-cistronic luciferase constructs we demonstrate that apoB 5' UTR may have weak internal ribosomal entry (IRES) translation which is not affected by insulin stimulation, and may function to stimulate basal levels of apoB mRNA translation.

Translational control of apolipoprotein B mRNA via insulin and the protein kinase C signaling cascades: evidence for modulation of RNA-protein interactions at the 5'UTR.

The link between hepatic insulin signaling and apolipoprotein B (apoB) production has important implications in understanding the etiology of metabolic dyslipidemia commonly observed in insulin-resistant states. Recent studies have revealed important translational mechanisms of apoB mRNA involving the 5' untranslated region (5'UTR) and insulin-mediated translational suppression via an insulin-sensitive RNA binding protein. Here, we have investigated the role of the protein kinase C (PKCs) signaling cascade in the regulation of apoB mRNA translation, using a series of chimeric apoB UTR-luciferase constructs, in vitro translation of UTR-luciferase cRNAs, and metabolic labeling of intact HepG2 cells. The PKC activator, phorbol 12-myristate 13-acetate (PMA), increased luciferase expression of constructs containing the apoB 5' UTR whereas treatment with Bis-I, a general PKC inhibitor or Go6976, a more specific PKC alpha/beta inhibitor, decreased expression, under both basal and insulin-treated conditions. These effects were confirmed to be translational in nature based on in vitro translation studies of T7 apoB UTR-luciferase constructs transcribed and translated in vitro in the presence of HepG2 cytosol treated with insulin or signaling modulators. Mobility shift experiments using cytosol treated with either PKC inhibitor (Bis-I) or activator (PMA) showed parallel changes between translation of apoB 5'UTR-luciferase constructs and the binding of a protein(s) complex migrating around 110 kDa to the apoB 5' UTR. ApoB mRNA levels were unaltered under these conditions based on real-time PCR analysis. Bis-I and Go6976 were both able to significantly decrease newly synthesized apoB100 protein in the presence or absence of insulin. Overall, the data suggests that PKC activation may induce increased mRNA translation and synthesis of apoB100 protein through a mechanism involving the interaction of trans-acting factors with the apoB 5'UTR. We postulate potential links between PKC activation as seen in insulin-resistant/diabetic states, enhanced translation of apoB mRNA, and hepatic VLDL-apoB overproduction.

Insulin-mediated suppression of apolipoprotein B mRNA translation requires the 5' UTR and is characterized by decreased binding of an insulin-sensitive 110-kDa 5' UTR RNA-binding protein.

Insulin has been shown to acutely regulate hepatic apolipoprotein B (apoB) secretion at both translational and post-translational levels; however, mechanisms of apoB mRNA translational control are largely unknown. Recent studies of apoB untranslated regions (UTRs) revealed a potentially important role for cis-trans interactions at the 5' and 3' UTRs. In the present paper, deletion constructs of the UTR regions of apoB revealed that the 5' UTR was necessary and sufficient for insulin to inhibit synthesis of apoB15. Metabolic radiolabeling and in vitro translation experiments in the presence of protease inhibitors confirmed that the effect of insulin on the apoB 5' UTR was translational in nature. Using the nondenaturing electrophoretic mobility shift assay (EMSA), protein-RNA complexes were detected binding to the apoB 5' and 3' UTRs. Denaturing EMSA identified a 110-kDa protein interacting at the 5' UTR. Nondenaturing EMSA determined that insulin altered binding of large protein complexes to the 5' UTR. Binding specificity was determined by competition with both specific and nonspecific competitors. Insulin treatment decreased binding of the 110-kDa protein to the 5' UTR as visualized by EMSA. Absence of insulin increased binding of this trans-acting factor to the 5' UTR by 2-fold. Analysis of the 3' UTR showed no significant insulin-mediated changes in binding of trans-acting factors. We thus propose the existence of a novel RNA-binding insulin-sensitive factor that binds to the 5' UTR and may regulate apoB mRNA translation. Perturbations in hepatic insulin signaling as observed in insulin-resistant states may alter cis-trans interactions at the 5' UTR, leading to alterations in the rate of apoB mRNA translation, thus contributing to apoB-lipoprotein overproduction.

Translational control of apolipoprotein B mRNA: regulation via cis elements in the 5' and 3' untranslated regions.

Translational control of apolipoprotein B (apoB) mRNA has been previously documented; however, the molecular mechanisms that govern translation of apoB mRNA are unknown. We investigated the role of the untranslated regions (UTR) in the regulation of apoB mRNA translation first by analyzing apoB UTR sequences using M-fold, a program used to predict RNA secondary structure. M-fold analysis revealed hairpin-like elements within the 5'UTR and 3'UTR of apoB mRNA with potential to form stable secondary structure. Luciferase (LUC) reporter assays were conducted to assess the biological activity of the putative RNA motifs within the UTR sequences by transiently transfecting HepG2 cells with chimeric mRNAs containing the 5' and/or 3' apoB UTRs linked to a LUC reporter gene. We observed statistically significant increases in LUC activity for the 5'UTRpGL3 and 5'/3'UTRpGL3 constructs. LUC mRNA levels remained constant for all constructs, suggesting that increased LUC activity was likely posttranscriptional in nature. When RNA isolated from transfected cells was translated in vitro, parallel increases in translatable LUC activity were observed. We also examined the role of UTR sequences within the context of the apoB coding sequence, using constructs containing the N-terminal 15% of apoB (apoB15). We observed a 40% and 25% increase in total protein mass with the 5'UTR-apoB15 construct and the 5'UTR-apoB15-3'UTR, respectively, over the control construct with no apoB UTR, with only a slight stimulation observed for apoB15 3'UTR. Radiolabeling analysis of apoB15 synthetic rate showed a more striking 4.5-fold stimulation of protein synthesis by 5'UTR while addition of both UTRs caused a 3.1-fold stimulation over the control construct. Deletion mutant analysis revealed that the stimulatory effect of the 5'UTR on apoB mRNA translation may be dependent on specific hairpin elements formed within the 5'UTR secondary structure. Overall, our data suggest that putative 5'UTR motifs are important for optimal translation of the apoB message whereas the presence of the 3'UTR appears to attenuate wild-type expression. Potential cis-trans interactions of these motifs with putative RNA binding proteins/translational factors are likely to govern apoB mRNA translation and protein synthesis and may play an important role in dysregulation of atherogenic lipoprotein production in dyslipidemic states.