A2780 cells were obtained from the American Tissue Culture Collection (Manassas, VA, USA). Cells were grown in RPMI-1640 medium with L-glutamine (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA). The medium was supplemented with 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.) and antibiotic/antimycotic solution (100 U/ml penicillin, 100 µg/ml streptomycin and 0.25 µg/ml amphotericin B; Invitrogen; Thermo Fisher Scientific, Inc.). The cells were maintained under standard tissue culture conditions at 37°C and 5% CO₂/95% air atmosphere. The number of cells was determined using a ZF Coulter counter (Beckman Coulter, Inc., Brea, CA, USA), and the total cell viability was analyzed by staining with 0.15% eosin, followed by light microscopy.

HUVECs were isolated, cultured and characterized as previously described (31,32). Cells were cultured on gelatin-coated dishes in cM199 (M199 medium supplemented with 10% heat-inactivated human serum (PAA; GE Healthcare Life Sciences, Chalfont, UK), 10% heat-inactivated newborn calf serum (Cambrex Corporation, East Rutherford, NJ, USA), 150 µg/ml crude endothelial cell growth factor (ECGF), 5 IU/ml heparin, 100 IU/ml penicillin and 100 µg/ml streptomycin (Sigma-Aldrich, St. Louis, MO, USA) at 37°C under a 5% CO₂/95% air atmosphere. Endothelial cell cultures were washed with medium without ECGF and heparin 24 h prior to the experiments.

A2780 cells and HUVECs were washed twice with ice-cold phosphate-buffered saline (PBS), scraped into lysis buffer [Tris-HCl (pH 7.4), 0.1% SDS, 10% glycerol and 100X protease inhibitor cocktail (Sigma-Aldrich)] and incubated for 45 min. Then, lysates were homogenized by sonication on ice for 30 sec at 30 V (SONOPULS HD 2070; BANDELIN electronic GmbH & Co. KG, Berlin, Germany). After sonication, lysates were centrifuged at 10,000 × g for 10 min at 4°C, and the supernatants were transferred into 1.5-ml microcentrifuge tubes and quantified according to the Lowry protein assay protocol (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Lysates (100 µg each) were then subjected to 12% SDS-polyacrylamide gel electrophoresis (PAGE) and blotted onto a polyvinylidene fluoride (PVDF) membrane with transfer buffer [3.6 g Tris, 18 g glycine and 10% methanol (pH 7.4)]. The PVDF membrane was blocked with 3% non-fat milk, followed by 1-h incubation with 15 µg rhEpo (40,000 IU/ml; EPREX®; Janssen Biologics B.V., Leiden, The Netherlands) in 1X Tris-buffered saline (TBS) with 0.5% milk (pH 7.2–7.4). Then, the membrane was washed and incubated overnight with primary anti-Epo antibody (catalog no., MAB2871; 1:1,000; R&D Systems, Inc., Minneapolis, MN, USA) in 1% non-fat milk (pH 7.2–7.4). The following day, the membrane was washed two times for 4 min in TBS with 0.05% Tween 20 (TBST; pH 7.4), and incubated for 1 h with secondary horseradish peroxidase (HRP)-conjugated antibody [goat anti-mouse immunoglobulin (Ig)G F(ab')2 fragment; 1:2,000; catalog no., 31436; Pierce; Thermo Fisher Scientific, Inc.] at room temperature (RT). The antibody reactivity was visualized with Pierce ECL Western Blotting Substrate (Thermo Fisher Scientific, Inc.). Bioluminescent signals were detected with ChemiDoc™ XRS+ and Image Lab 3.0 software (Bio-Rad Laboratories, Inc.) or X-ray films (Roberts Technology Group, Inc., Chalfont, PA, USA). Traditional western blotting for the detection of EpoR with primary anti-Epo (1:1,000; catalog no., MAB2871) and primary goat anti-EpoR antibodies (1:1,000; catalog no., AF-322-PB; R&D Systems, Inc.) served as controls for far-western blotting detection.

To confirm the affinity between EpoR and rhEpo observed using far-western blotting, a second protein interaction assay where Epo was subjected to 12% SDS-PAGE and blotted onto a PVDF membrane was performed. The PVDF membrane with rhEpo was then incubated with A2780 or HUVEC cell lysates (220 µg each) for 1 h in 0.5% milk, followed by overnight incubation with primary anti-EpoR antibody (1:1,000; catalog no., AF-322-PB) in 1% milk. The following day, the membrane was washed two times for 4 min in TBST (pH 7.4), and then incubated for 1 h with secondary HRP-conjugated antibody (rabbit anti-goat IgG F(ab')2 fragment; Pierce; Thermo Fisher Scientific, Inc.) at RT. The antibody reactivity was visualized as mentioned above.

A total of 80 µl agarose beads (Protein A Agarose; Innova Biosciences Ltd., Cambridge, UK) were washed twice with 1 ml ice-cold radioimmunoprecipitation assay (RIPA) buffer (Sigma-Aldrich) and centrifuged for 30 sec at 8,200 × g at 4°C. Non-specific binding was blocked by incubation of the agarose beads with 2% bovine serum albumin for 1 h at RT, followed by two washes with ice-cold RIPA buffer and centrifugation as mentioned above. Both protein A agarose beads with 10 µg anti-Epo antibody (MAB 2871) and 22 µg rhEpo with the lysate of A2780 cells or HUVECs (250 µg of proteins each) were then incubated for 2 h on a rotary agitator (10 rpm, 4°C). After two-times washing, the protein A agarose beads/anti-Epo antibody complex was added to the rhEpo/cell lysate complex (EpoR and other proteins) and incubated together for another 1 h on a rotary agitator (10 rpm, 4°C). Upon washing and addition of sample buffer, the complex of protein A agarose beads/anti-Epo/rhEpo/EpoR was boiled for 5 min, centrifuged (10,000 × g at 24°C for 5 min) and subjected to SDS-PAGE (12%). The gel was washed two times in distilled (d)H₂O for 5 min, fixed (with 40% ethanol, 10% acetic acid and 50% dH₂O) for 1 h, washed again twice in dH₂O for 10 min, and stained by overnight incubation in 80% Colloidal Coomassie Blue (0.1% Coomassie Brilliant Blue G-250, 2% orthophosphoric acid and 10% ammonium sulfate) and 20% methanol at RT. The following day, the gel was washed in acetic acid (1%), and the potential EpoR proteins were detached from the gel and analyzed by matrix-assisted laser desorption/ionization (MALDI)-time-of-flight (TOF) mass spectrometry (MS). The detailed protocol of the pull-down assay has been previously described (33). Identification of the proteins observed in the gel was performed by peptide mass fingerprinting, as previously described (34). In short, protein bands were excised and digested as previously described (35). An aliquot of the digested solution was mixed with an aliquot of α-cyano-4-hydroxycinnamic acid (Bruker Corporation, Billerica, MA, USA) in 33% acetonitrile and 0.25% trifluoroacetic acid. This mixture was then deposited onto a 600-µm AnchorChip (Bruker Corporation), and allowed to dry. MALDI-MS data were obtained in an automated analysis loop using an Ultraflex TOF mass spectrometer (Bruker Corporation). Spectra were acquired in positive-ion mode at 50 Hz laser frequency, and 100–1,000 individual spectra were averaged. The selected precursor ions were subjected to fragment ion analysis in tandem TOF mode to obtain the corresponding MALDI-MS/MS spectra. Automated analysis of MS data was performed using the FlexAnalysis software version 3.0 (Bruker Corporation). MALDI-MS and MALDI-MS/MS data were combined through the BioTools software version 3.2 (Bruker Corporation) to search a non-redundant protein database in the National Center for Biotechnology Information website (www.ncbi.nlm.nih.gov/), using Mascot software (server version 2.4; Matrix Science, Ltd., London, UK). Each experiment was repeated three times.

Erk1/2 signaling was monitored in HUVECs only, since the same signaling pathway was previously described in A2780 cells by Solár et al (12). HUVECs were seeded into Petri dishes with M199 medium containing serum, and incubated under the standard conditions described above. HUVECs were washed with serum-free medium 24 h prior to the experiments, followed by the addition of 5 IU/ml rhEpo to the experimental group, and the addition of 20 ng vascular endothelial growth factor (VEGF) to the positive control group. After 5, 10, 30 or 60 min of incubation, cells were quickly washed twice with ice-cold PBS, and lysed as mentioned above. HUVEC lysates (30 µg each) were then subjected to SDS-PAGE (12%), blotted onto PVDF membranes and incubated overnight with primary anti-p44/42 mitogen-activated protein kinase (MAPK; #9102, 1:1,000; Cell Signaling Technology, Inc., Danvers, MA, USA) or anti-phospho-p44/42 MAPK (#9101, 1:1,000; Cell Signaling Technology, Inc.) antibodies at 4°C. The membrane was then reprobed with anti-β-actin primary antibody (AC-15; 1:1,000; catalog no., A5441; Sigma-Aldrich) to detect β-actin protein, which served as a loading control. The following day, the membrane was washed twice in TBST (pH 7.4) for 4 min, and incubated for 1 h with secondary HRP-conjugated antibody (goat anti-rabbit IgG F(ab')2 fragment; catalog no., 31461; Pierce; Thermo Fisher Scientific, Inc.) at RT. The antibody reactivity was visualized as mentioned above.

Far-western blotting avoids the problem of direct identification of EpoR with a potentially non-specific anti-EpoR antibody, and also confirms the interaction of rhEpo with EpoR. Whole proteins of ovarian adenocarcinoma A2780 and normal endothelial HUVECs (including EpoR) were separated and blotted on PVDF membranes, thus representing the antigen for the interaction with rhEpo. The expected EpoR antigen/rhEpo antigen complex was identified with an anti-Epo antibody, followed by traditional western blot detection (Fig. 1A). Two blots without rhEpo served as controls for far-western blot detection. In this regard, one blot was incubated with anti-Epo (Fig. 1B) and the second one with anti-EpoR antibodies (Fig. 1C). Indeed, far-western blotting revealed the same 57-kDa protein in the EpoR/rhEpo/anti-Epo-antibody complex blot and in the control blot with anti-EpoR antibody. In addition, the control blot with anti-Epo antibody detection did not exhibit any protein of 57 kDa in size (Fig. 1B).

Far-western blot detection of rhEpo/EpoR interaction was also confirmed by a second type of interaction assay. For that purpose, rhEpo was blotted onto PVDF membranes, followed by incubation with cell lysates (EpoR) and western blot detection using anti-EpoR antibody. Notably, rhEpo was detected not only by traditional immunoblotting with an anti-Epo antibody (Fig. 2C), but also through rhEpo/EpoR/anti-EpoR-antibody complex, as represented in Fig. 2A and B. In this regard, both A2780 cells (Fig. 2A) and HUVECs (Fig. 2B) exhibited a clear interaction of their EpoR protein with rhEpo.

For further verification of the aforementioned rhEpo/EpoR interaction, immunoprecipitation with protein A agarose beads was applied. Independently incubated anti-Epo antibody with protein A agarose beads and cell lysates with rhEpo were subsequently mixed and incubated together, and the resulting protein complex was separated on SDS-PAGE and stained with Coomassie Blue. While the control groups did not reveal any interaction between the protein A agarose beads and the proteins in the cell lysates or the added rhEpo (data not shown), the anti-Epo antibody captured on the protein A agarose beads did interact with 39-, 57- and >57-kDa proteins (Fig. 3). Only the 57-kDa protein was detached from the gel, subjected to MALDI-TOF analysis and verified as Epo-interacting EpoR (P<0.014), thus corresponding to the full-length EpoR. The details of MS analysis and Mascot search are as follows: number of peptides matched by MS, 6; number of peptides matched by MS/MS, 2; score, 73; sequence coverage, 53%; accession no. matched-gi|6137384 (www.ncbi.nlm.nih.gov/protein/gi%7C6137384).

Since EpoR signaling in A2780 cells has been previously published (12), the present study provided evidence of the functionality of the rhEpo/EpoR complex on HUVECs. These cells were starved overnight, and incubated with rhEpo for 5–60 min or with rhVEGF (positive control) for 10 min, and then processed for western blot analysis. rhEpo at a concentration of 5 IU/ml increased the phosphorylation of Erk1/2 at 30 min, and reached the maximum at 60 min after the addition of rhEpo. Furthermore, the phosphorylation of Erk1/2 at 60 min following rhEpo addition achieved the same intensity than phospho-Erk1/2 induced by recombinant VEGF (positive control) (Fig. 4).