HoxA5 stabilizes adherens junctions via increased Akt1

Abstract

Normal vascular development and angiogenesis is regulated by coordinated changes in cell-cell and cell-extracellular matrix (ECM) interactions. The Homeobox (Hox) family of transcription factors coordinately regulate expression of matrix degrading proteinases, integrins and ECM components and profoundly impact vascular remodeling. Whereas HoxA5 is down regulated in active angiogenic endothelial cells (EC), sustained expression of HoxA5 induces TSP-2 and blocks angiogenesis. Since HoxA5 is also lacking in EC in proliferating hemangiomas, we investigated whether restoring expression of HoxA5 could normalize hemangioma cell morphology and/or behavior. Sustained expression of HoxA5 in the murine hemangioma cell line (EOMA) reduced their growth in vivo and promoted branching morphogenesis in 3D BM cultures. Moreover, restoring HoxA5 expression increased the retention of beta-catenin in adherens junctions and reduced permeability. In addition we also show that the HoxA5 mediated increase in stability of adherens junctions requires Akt1 activity and introduction of constitutively active myr-Akt in EOMA cells also increased retention of beta-catenin in adherens junctions. Finally we show that HoxA5 increases Akt1 mRNA, protein expression and further enhances Akt activity via a coordinate down regulation of PTEN. Together these results demonstrate a central role for HoxA5 in coordinating a stable vascular phenotype.

Figure 1

HoxA5 and TSP-2 expression in endothelioma cells lines. (A) Semi-quantitative RT-PCR showing relative levels of HoxA5 mRNA in normal HMEC-1, bEND-1 and EOMA cell lines. (B) Semi-quantitative RT-PCR analysis showing relative levels of HoxA5 mRNA in control transfected EOMA cells and HoxA5-transduced EOMA cells (+HoxA5). The 800-bp product corresponds to HoxA5 mRNA. The relative RNA loading is shown by 488-bp band corresponding to 18S ribosomal RNA determined using competitive primers. (C) Upper panel shows semi-quantitative RT-PCR for TSP-2 mRNA and corresponding 18s RNA levels in control (EOMA) and HoxA5 transduced EOMA cells. Lower panel shows Western blot of TSP-2 protein expression in control EOMA and HoxA5 expressing EOMA cells.

Figure 2

HoxA5 blocks hemangioma growth in vivo. Photographs showing visible hemangiomas in mice receiving 2.5 × 10⁶ control transfected EOMA cells (A) or HoxA5 transduced EOMA cells (B). Upper panels show external appearance of the inoculation sites and lower panels show the tumors in the subcutaneous injection sites. (C) Quantitative analysis of tumor volume measured after seven days. Values are expressed as mean ± s.d. (**p < 0.001, n = 6). (D) Semi-quantitative RT-PCR shows corresponding mRNA levels of HoxA5 (upper panel) and TSP-2 (lower panel) in tissue harvested seven days after implantation of control (EOMA) or HoxA5 expressing EOMA cells (+HoxA5).

Figure 3

HoxA5 promotes branching of endothelial cells in 3D BM. Phase contrast micrographs shows morphology of (A) control transfected EOMA cells (B) EOMA cells stably expressing HoxA5 (C) and HMEC-1 cells 24 hours after culturing on 3D BM (Matrigel). Original magnification ×20 (n = 6). (D) Shows densitometric analysis of relative mRNA levels for HoxA5 or TSP-2 mRNA in HMEC-1 cells treated with control siRNA or HoxA5 siRNA. Photomicrographs showing capillary tubes in HMEC-1 treated with control siRNA (E) or HoxA5 siRNA (F), 24 hours after culturing on 3D BM. Original magnification 20x.

Figure 4

HoxA5 mediated branching is associated with increased retention of β-catenin in adherens junctions. Immunofluorescence staining of control transfected EOMA cells (A) or EOMA cells expressing HOXA5 (B). Upper panel shows staining for β-catenin (green) and lower panel shows merged image of the nuclear DAPI stain (blue) and β-catenin (green). Original magnification: ×60. (n = 6). (C) Representative Western blot shows distribution of β-catenin protein in the TX-100 soluble (Sol) and insoluble (Ins) fraction in lysates from control or HoxA5 expressing EOMA cells or HMEC-1. (D) Western blot analysis of EOMA or HoxA5 expressing EOMA cells following immunoprecipitation with an anti-VE-cadherin Ab (IP VE-cad. upper panel). Lower panel shows the same membrane reprobed with antibodies to β-catenin (IB β-cat). (E) Histogram showing relative β-catenin transcriptional activity in control (grey bar) or HoxA5 (black bar) expressing EOMA cells following transfection with pTOPFLASH or with pFOPFLASH luciferase reporters (white bars). CMV-βgal plasmid was cotransfected as internal control for transfection efficiency. Luciferase activity was measured at 72 hours after transfection and values are the mean triplicate samples of four experiments (**p < 0.001). (F) Histogram shows relative permeability of FITC-labelled Dextran in control or HoxA5 expressing HMEC-1 cells (n = 3, **p < 0.05).

Figure 5

HoxA5 mediated stabilization of junctions requires Akt activity. (A) Western blot showing levels of Ser473 phoshorylated AkT (upper panel) in EOMA cells expressing HoxA5 treated with DMSO (-) or 30 µM LY294002 (+) for four hours. Middle panel shows corresponding levels of total Akt protein and lower panel shows relative protein loading by western blotting for β-actin. (B) Upper panel shows western blot of β-catenin in Tx-100 soluble (Sol) and insoluble (Ins) fractions in HoxA5 expressing EOMA cells treated with DMSO (-) or 30 µM of LY294002 for four hours. Lower panel shows corresponding Western blot of β-catenin in Tx-100 soluble or insoluble fractions in normal HMEC-1 cells treated with DMSO (-) or 30 µM of LY294002 for four hours. Data shown are representative of three different experiments. (C) Western blot for β-catenin in Triton X-100 soluble and insoluble fractions of HoxA5 expressing EOMA cells after treatment with 1mM MG132 an in the presence or absence of LY294002. D) β-catenin transcriptional activity in control or HoxA5 expressing EOMA cells following treatment for four hours with DMSO or 30 µM of LY294002. Luciferase activity was measured at 72 hours after transfection with pTOPFLASH luciferase reporter constructs and four hour pretreatment with 1 mM MG132. Values represent the mean ± s.d. for triplicates samples in each of three separate experiments. (E) Cell lysates of control or HoxA5 expressing EOMA cells were subjected to immunoprecipitation with an anti-VE-cadherin antibody (IP VE-cad) in the presence or absence of LY294002. Immunoprecipitated proteins were western blotted with antibodies to VE-cadherin (IB VE-cad, upper panel), β-catenin (IB β-cat, middle panel) or total phosphotyrosine (IB P-Y lower panel). Histogram shows densitometric analysis of the average levels of tyrosine phosphorylated β-catenin relative to total β-catenin in each condition in each of three separate samples.

Figure 6

Constitutive Akt mimics HoxA5 induced retention of β-catenin in adherens junctions. (A) Western blot showing phospho-AKT activity in EOMA cells. Cell lysates of EOMA cells control (pWZL) or myr-Akt (myr-Akt) expressing cells were subjected to western blot and probed with antibodies to phospho-Akt (Ser 473). The lower band corresponds to phosphorylated myr-Akt. (B) Upper panel shows western blot of β-catenin in Tx-100 soluble and insoluble fractions in pWZL and myr-Akt expressing EOMA cells grown in 3D BM for 72 hrs. Lower panel shows the same blot reprobed for β-actin. (C) Relative levels of β-catenin to β-actin levels after densitometric analysis of the blot in (B) using ImageJ software (http://rsb.info.nih.gov). (D) Immunofluorescence staining for β-catenin (green) in control infected EOMA cells (pWZL, upper panels) and in EOMA cells transduced with a myr-Akt-expression virus myr-Akt (lower panels). Confocal photographs were taken 72 hours after infection and culturing on 3D BM. Original magnification ×60 (n = 3).

Figure 7

HoxA5 induces Akt mRNA and protein expression and increases Akt activity in endothelial cells. (A) Relative mRNA levels of Akt1 were determined by semi-quantitative RT-PCR analysis. Densitometric analysis and correction for total 18 s RNA levels in each sample is shown. Light bar represents control cells and dark bar represents HoxA5 expressing cells. Profiles shown are representative of six different experiments and error bars show s.d. **p < 0.001 compared values to control, p values calculated using Student's t test. (B) Western blot analysis showing levels of Akt (upper panel) relative to β-actin (middle panel). The membranes were stripped and reblotted with antibodies against phospho-Akt (Ser 473) (lower panel). Inset shows densitometric analysis of the relative activity of Akt by expressing the ration of phospho-Akt to total AKT in EOMA and EOMA cells expressing HoxA5. Profiles shown are representative from six experiments and error bars show s.d. *p < 0.01 and **p < 0.001 compared values to control, p values calculated using Student's t test. (C) Cell lysates were blotted with PTEN Ab (upper panel). The membranes were then stripped and treated with β Actin Ab (lower panel). Molecular mass standards (47 and 42 kDa) are indicated on the left side (n = 3). (D) Western blot for Foxo1a in nuclear (Nu) and cytoplasmic (Cy) extracts harvested from control transfected EOMA or HoxA5 expressing EOMA cells.