Abstract
We used in vitro and in vivo models to characterize the physiological role of the novel protein encoded by C6ORF105. This gene's expression is androgen-responsive, and the encoded protein is predicted to be palmitoylated and membrane multi-spanning. Previously we showed that C6ORF105 expression co-regulates with tissue factor pathway inhibitor (TFPI)in human endothelial cells (EC); hence we named this protein "androgen-dependent TFPI-regulating protein" (ADTRP). Using in vitro cell-based TOP-Flash reporter assay we identified ADTRP as a negative regulator of canonical Wnt signaling in human cells. Overexpressing ADTRP in HEK293T cells inhibited the activity of beta-catenin/TCF-dependent transcriptional reporter, while silencing ADTRP increased the expression of Wnt target genes LEF-1, AXIN-2, IL-8 and DKK-2 in EA.hy926 EC line and HUVEC. Addition of LiCl showed that the effect of ADTRP was upstream of GSK3, therefore we focused the investigations on the Wnt signalosome proteins. ADTRP expression in HEK293T cells led to decreased phosphorylation of Wnt co-receptor LRP6, suggesting that ADTRP can affect this critical membrane-located event of Wnt signaling. Furthermore, ADTRP expression in reporter cells transfected with a constitutively phosphorylated form of LRP6 (LRP6DN mutant) inhibited Wnt3a- induced signaling, which suggests that ADTRP can interfere with events downstream of LRP6 phosphorylation, such as Axin-2 binding. Altogether, these data indicate that the Wnt signaling inhibitory activity of ADTRP takes place at the plasma membrane level. Site directed mutagenesis of the predicted palmitoylation site Cys61 showed that Wnt inhibitory effects of ADTRP require palmitoyl-mediated anchoring, highlighting the importance of proper membrane location of ADTRP for Wnt pathway inhibition. In vivo morpholino-based knockdown of adtrp in zebrafish embryos produced aberrant angiogenesis, defective branching and ruptured vessels, hemorrhage spots, pericardial edema and slow heart-beat, all reminiscent of defects caused by activation of canonical Wnt signaling. Indeed, adtrp knock down increased Wnt mediated lef-1 and pax-2a as well as mmp2 and mmp9 mRNA expression. Co-injection of ADTRP mRNA partially recovered the adtrp morpholino- induced morphologic abnormalities. Also, knock down of adtrp in a Wnt reporter zebrafish showed increased expression of ectopic Wnt signaling. Furthermore, our recently established Adtrp-/- mice also display some typical Wnt-mediated vascular defects, including: (i) abnormal patterning, increased capillary tortuosity, abnormal branching and increased density of the capillary network; (ii) dilated vessels, especially venules and veins; (iii) increased leakeage of permeability tracers (Evans blue and fluorescent dextran) without evident changes in endothelial junctions; (iv) hemorrhage spots in the skin, meningeal layers, heart, bladder and kidneys; (v) intravascular and interstitial fibrin deposition in the lung, liver and kidney. ADTRP deficiency decreased plasma TFPI antigen by ~2-times. Furthermore, TFPI antigen and anticoagulant activity in lung extracts and isolated lung EC were similarly decreased, which confirms our previous in vitro data. We aslo noticed increased tail bleeding time (>500 sec vs. 200 sec in WT littermates) and blood volume loss, which likely was caused by increased dilation of the tail vein. Gene expression analysis of whole organs showed upregulation of Wnt target genes involved in vascular contractility (Nos3), and extracellular matrix remodeling (Mmp2). Similarly, skin fibroblasts and lung EC isolated from Adtrp-/- mice showed increased expression of Wnt target genes (Lef-1, Cyclin D, Dkk2, c-Myc), which indicates constitutive activation of canonical Wnt signaling. In conclusion, we used genetic animal models and cell culture systems to show for the first time that the novel protein ADTRP plays major roles in vascular development and function. Lack of, or low levels of ADTRP associate with activation of coagulation and vascular development defects, which may be due, at least in part, to intrinsic high levels of ectopic canonical Wnt signaling.
Disclosures
No relevant conflicts of interest to declare.
EHMT2 epigenetically suppresses Wnt signaling and is a potential target in embryonal rhabdomyosarcoma
