ErbB4 isoforms selectively regulate growth factor induced Madin-Darby canine kidney cell tubulogenesis

Abstract

ErbB4, a member of the epidermal growth factor (EGF) receptor family that can be activated by heregulin beta1 and heparin binding (HB)-EGF, is expressed as alternatively spliced isoforms characterized by variant extracellular juxtamembrane (JM) and intracellular cytoplasmic (CYT) domains. ErbB4 plays a critical role in cardiac and neural development. We demonstrated that ErbB4 is expressed in the ureteric buds and developing tubules of embryonic rat kidney and in collecting ducts in adult. The predominant isoforms expressed in kidney are JM-a and CYT-2. In ErbB4-transfected MDCK II cells, basal cell proliferation and hepatocyte growth factor (HGF)-induced tubule formation were decreased by all four isoforms. Only JM-a/CYT-2 cells formed tubules upon HB-EGF stimulation. ErbB4 was activated by both HRG-beta1 and HB-EGF stimulation; however, compared with HRG-beta1, HB-EGF induced phosphorylation of the 80-kDa cytoplasmic cleavage fragment of the JM-a/CYT-2 isoform. HB-EGF also induced early activation of ERK1/2 in JM-a/CYT-2 cells and promoted nuclear translocation of the JM-a/CYT-2 cytoplasmic tail. In summary, our data indicate that JM-a/CYT-2, the ErbB4 isoform that is proteinase cleavable but does not contain a PI3K-binding domain in its cytoplasmic tail, mediates important functions in renal epithelial cells in response to HB-EGF.

Figure 1.

Schematic illustration of ErbB4 structure and isoforms. All the isoforms contain extracellular ligand binding domain (ECD), juxtamembrane (JM), transmembrane (TM) and a intracellular domain (ICD). ECD contains a signal sequence (SS), two ligand-interacting domains (L1, L2) and two cysteine-rich (CR) domains. JM contains alternative JM isoforms, with (JM-d, JM-a) or without (JM-b, JM-c) a proteolytic cleavage site between His651 and Ser652. ICD contains a tyrosine kinase domain (TK), a regulatory domain with autophosphorylation sites plus sequence changes attributable to the CYT-1 and CYT-2 isoforms (hatched box). CYT-2 isoform has a 16-amino acid deletion that contains a PI3K-binding (PI3K BD, underlined) and a WW domain binding (WW BD, bold) motif. Dashes represent deletions.

Figure 2.

Expression of ErbB4 in rat embryonic and adult kidney. (A) Western blot analysis of ErbB4 expression level. Homogenates from kidney tissue at different stages of development or adult kidney cortex or medulla were separated by 7.5% SDS-PAGE and immunoblotted with ErbB4 antibody. ErbB4 had the highest expression level in the metanephric embryonic kidney (E19). In adult kidney ErbB4 was mainly expressed in medulla. IP, immunoprecipitation. (B–D) Immunohistological staining showed that ErbB4 protein was localized predominantly in the derivatives of the ureteric buds, including developing medullary collecting ducts (arrows) and branching ducts in the cortex (arrowhead) in metanephros of embryonic kidney day 17.5 (B), some distal ducts (C), and the collecting ducts in adult kidney (D). Brown (DAB stain) indicates ErbB4 immunoreactivity; Blue (toluidine blue), nuclei.

Figure 3.

Quantitation of ErbB4 isoform mRNA level in rat development kidney by real-time PCR. ErbB4 JM isoforms (A) and CYT isoforms (B) were detected using specific primers and probes. Data were normalized to ubiquitin mRNA, which was used as internal control. Delta Ct was converted to arbitrary values by the formula 2^−dCt × 10³. Results were expressed as mean ± SE of three experiments.

Figure 4.

ErbB4 clone selection and phosphorylation induced by HRG-β1 and HB-EGF. (A) Representative clones with vector or ErbB4 JM-a/CYT-1, JM-a/CYT-2 or JM-b/CYT-1, or JM-b/CYT-2 expression levels by blot with ErbB4 c-18 antibody. Clones (marked by #) with relatively similar ErbB4 expression level were used for further studies. (B) ErbB4 isoform transfection was confirmed by RT-PCR using primers flanking human ErbB4 JM or CYT fragments. The expected size for the amplified ErbB4 JM-a isoform is 220 bp and JM-b is 190 bp. CYT-1 is 182 bp and CYT-2 is 134 bp. Neg, negative control. (C) Stimulation of ErbB4 phosphorylation by HRG-β1 or HB-EGF in MDCK II cells overexpressing ErbB4 isoforms. Cells were serum starved 24 h before treatment with HRG-β1 or HB-EGF for 10 min in 37°C. Immunoblotting was performed with anti-phospho-ErbB4 (p-ErbB4, top panel) and reblotted with ErbB4 antibody (bottom panel). (D) Relative protein phosphorylation level of the ErbB4 80-kDa cleavage fragment was calculated by correction for the total amount of the cleavage fragment and then compared with the control group. *p < 0.05 compared with control.

Figure 5.

Cyst formation and tubulogenesis in ErbB4-transfected MDCK cells grown in collagen. Experiments were performed as described in Materials and Methods. After 9–10 d of culture, cells were fixed with 3.7% formaldehyde and photographed. From top to bottom: no treatment and growth in the presence of HGF (20 ng/ml), HRG-β1 (50 ng/ml), or HB-EGF (50 ng/ml). These findings were representative of at least four independent experiments.

Figure 6.

ErbB4 expression inhibited MDCK II cell proliferation. Cell mitogenesis was determined by [³H]thymidine incorporation assays performed in a 3D collagen gel mixture as described in Material and Methods. (A) All four ErbB4 isoforms significantly inhibited MDCK II cell proliferation. Values are means ± SE of three independent experiments in triplicate wells. (B) Both HRG-β1– and HB-EGF–stimulated JM-a/CYT-1 and JM-a/CYT-2 cell proliferation. Interestingly, the JM-a/CYT-2 cells had significantly increased [³H]thymidine incorporation in response to HB-EGF compared with HRG-β1, whereas the opposite effect was noted in JM-a/CYT-1 cells. Growth factor–mediated cell proliferation was reduced by both the PI3K inhibitor (LY294002) and the MAPK inhibitor (U0126). Values represent ratios compared with controls (no treatment, 100%). CTR, control; LY, LY294002. *p < 0.05 versus corresponding control group; ^#p < 0.05 versus corresponding growth factor treatment group without inhibitors.

Figure 7.

Cell attachment and migration of ErbB4-transfected MDCK cells. (A) Effect of ErbB4 expression on attachment of MDCK II cells. Adhesion assays were performed on collagen I–coated plates as described in Materials and Methods. The results were the average of four independent experiments ± SE, each with three replications. JM-a/CYT-2 and JM-b/CYT-2 cells showed significantly increased cell–matrix adhesion compared with vector alone or JM-a/CYT-1 and JM-b/CYT-1 cells. (B) Expression of ErbB4 altered the migration of MDCK II cells in a modified Boyden chamber migration assay. MDCK II cells were placed in transwell filter chambers coated on the underside with collagen I. Cells were then allowed to migrate for 4 h and were quantitated as described in Materials and Methods. *p < 0.05 compared to vector alone cells.

Figure 8.

Effects of HRG-β1 and HB-EGF on phosphorylation of Akt, ERK 1/2, and STAT3. Quiescent cells were trypsinized and plated into 3D collagen gel mixture at 60,000 cells per 100 μl gel mixture with or without HRG-β1 or HB-EGF (50 ng/ml) treatment. Cells were lysed at the indicated time point, and cell lysates were resolved by 10% SDS-PAGE and subjected to immunoblotting with anti-phospho-Akt antibody, anti-phospho-ERK 1/2 antibody, or anti-phospho-STAT3 antibody. The membranes were then stripped and reprobed with anti-total ERK 1/2 antibody as a loading control.

Figure 9.

Cleavage and nuclear translocation of ErbB4 isoforms. (A) ErbB4 cleavage after PMA treatment. Cells were treated with 100 ng/ml PMA for 1 h. After SDS-PAGE and transfer to PVDF membrane, the samples were immunoblotted with anti-ErbB-4. (B and C) ErbB4 subcellular localization in response to PMA, HRG-β1, or HB-EGF treatment. ErbB4 isoform-transfected MDCK II cells were grown on coverslips, and cells were incubated with 10 ng/ml leptomycin B (LMB) for 12 h, an inhibitor of nuclear export, before treated with or without (as control) PMA (B) or HRG-β1 or HB-EGF (C) for 1 h. Subcellular localization of an intracellular ErbB4 epitope was determined by confocal microscopy. Green, ErbB4 staining; red, nuclei stained with TO-PRO-3.