Fig 2: Effect of various ephrins on EphA4/GHR/JAK2/STAT5B signaling.(A) EphA4-HA was expressed in HEK293T cells, then stimulated with oligomerized ephrin-A1-Fc, ephrin-A5-Fc, ephrin-B2-Fc, or IgG-Fc alone as shown in the Materials and Methods for the time indicated. Cell lysates were fractionated with SDS-PAGE. pSTAT5B was detected by IB following IP using the antibodies shown, and EphA4-HA was detected with an anti-HA antibody. The same experiments were repeated 3 times, showing similar results.

Fig 3: Phosphorylation of GHR and JAK2 by ligand-activated EphA4.(A) GHR phosphorylation by EphA4 in the absence of JAK2. Jak2-KO mouse embryonic fibroblasts (MEF) retrovirally overexpressing EphA4 (EphA4 overexp) and those overexpressing EphA4 and Ghr (EphA4&GHR overexp) were pre-incubated with serum-free medium overnight and incubated with ephrin-A1 (0.5 µg/mL) for the time shown. Cell lysates were used for IP and/or IB studies with the indicated antibodies. Anti-pY: anti-phosphotyrosine antibody. The same experiments were repeated 3 times, showing the same results. (B) JAK2 phosphorylation by EphA4. HEK293T cells expressing JAK2(KD)-Flag with or without EphA4(WT)-HA were preincubated with serum-free medium for 6 h, then stimulated with ephrin-A1 (0.5 µg/mL) for the time shown. Cell lysates were detected by IB and/or IP with the indicated antibodies. The same experiments were repeated 3 times, showing the same results.

Fig 4: JAK2 activated by GH does not phosphorylate EphA4-bound or free STAT5B.(A) Structure of GHR WT and mutants. The numbers represent the amino acid numbers counting from the initiation codon. “Box1,” which is the proline-rich region of the cytoplasmic domain, is necessary for GH-dependent association with JAK2 and the following activation of JAK2 [32]. TM: transmembrane domain. The numbers represent the amino acid numbers counting from the one encoded by the initiation codon. (B) Binding of JAK2 with GHR (WT and mutants) and phosphorylation of STAT5B. GHR mutants tagged with HA were co-expressed with JAK2-myc in HEK293T cells and the cells were stimulated with GH for 15 min. GHR(464/2F) bound to and activated JAK2, but did not phosphorylate STAT5B. (C) Phosphorylation of STAT5B bound to kinase-deleted EphA4 by GH-activated JAK2. GHR(WT), (464/2F), or (?cyto) was co-expressed in HEK293T cells with EphA4(?JM,?K)-3Flag, which binds to STAT5B (Fig 2B), then the cells were stimulated with GH for 15 min. The expression of EphA4(?JM,?K)-3Flag did not alter the extent of STAT5B phosphorylation as compared with the control (vector). The cell lysates were used for IP and/or IB with the indicated antibodies. Arrowheads indicate the molecules expressed following transfection. The other bands are regarded as non-specific bands in comparison with the result of vector transfection (see also S1 Fig). GHR molecules are detected as double or diffuse bands with the upper bands representing carbohydrate-modified molecules. All the experiments in (B) and (C) were repeated twice, showing similar results.

Fig 5: Requirement of GHR for EphA4-mediated STAT5B phosphorylation.(A) Ghr mRNA expression in Ghr-knockdown (Ghr-k/d) Jak2-KO MEF. Jak2-KO MEF were transduced with lentivirus carrying either Ghr shRNA (Ghr-k/d) or control shRNA at an MOI of 10. The Ghr mRNA expression levels were measured in 8 independent MEF clones using quantitative real-time RT-PCR. Data are shown as mean ± standard deviation (SD) values of the 8 samples. (B) GHR protein expression levels. Cells from one of the MEF clones used for the following experiments were lysed and used for western blotting to quantify GHR protein expression. The results of 5 independent studies are shown in the upper panel, and their band densities are presented as mean ± SD values in the lower panel. (C) Effect of GHR knockdown on JAK2-independent STAT5B phosphorylation. Jak2-KO MEF with or without Ghr-k/d were pre-incubated in serum-free medium overnight (no ligands) and exposed to GH and/or ephrin-A1. The cells were exposed to ephrin-A1 (500 ng/mL) for 75 min, and to GH (500 ng/mL) for 30 min. In case of exposure to both GH and ephrin-A1, the cells were pre-exposed to ephrin-A1 for 45 min and then exposed to GH for 30 min. Cell lysates were used for IP and IB studies with the antibodies indicated in the figure. The same experiments were repeated 3 times, showing similar results. The amount of phosphorylated STAT5B (pSTAT5B) was reduced in the Ghr-k/d cells. (D) Effect of GHR KO on STAT5B phosphorylation. The same experiments as in (C) were carried out in Ghr-KO MEF. The same experiments were repeated 3 times, showing the same results. The complete absence of GHR (Ghr-KO) abolished STAT5B phosphorylation.