Trichostatin A (TSA), Ponceau S, 4ʹ,6-diamidino-2-phenylindole (DAPI), and fluorescence mounting solution were purchased from Sigma-Aldrich (St. Louis, MO, USA). Dulbecco’s modified Eagle’s medium (DMEM) was purchased from Welgene (Seoul, Korea). Fetal bovine serum, antibiotics, Opti-MEM, methanol-free paraformaldehyde, and transfection reagent, Lipofectamine LTX with Plus were purchased from Thermo Fisher Scientific (Waltham, MA, USA). Antibodies against FLAG and β-actin were purchased from Sigma-Aldrich. Two different APE1/Ref-1 antibodies were used. A monoclonal antibody for the detection of APE1/Ref-1 (N-terminal 80–100 aa) was obtained from Novus (Littleton, CO, USA). A polyclonal rabbit APE1/Ref-1 antibody was generated by immunization with recombinant human APE1/Ref-1 (rh APE1/Ref-1) as previously reported [11].

Bladder tissue biospecimens were obtained from patients with operable bladder cancer (n =10). All samples were obtained from the archives of the Department of Urology, Chungnam National University, between 2010 and 2011. This study was approved by the Institutional Review Board of Chungnam National University Hospital (approval number: 1101-16). To identify human APE1/Ref-1 variants in coding DNA sequences, reverse transcription polymerase chain reaction was performed; Polymerase chain reaction (PCR) primers were designed based on the GenBank reference sequences (accession number. NM_001641). The coding DNA sequence of human APE1/Ref-1 cDNA was amplified using the following primers set: forward primer 5ʹ-TGC CCA AGG GCG TTC GTA ACG GGA -3ʹ, reverse primer 5ʹ-GCT CAA AGT GAT TTA GGG GTG GTG -3ʹ. Amplified cDNA was inserted into the pGEM-T easy vector system (Promega, Madison, WI, USA) and transformed into DH5α Escherichia coli. Finally, APE1/Ref-1 variants were identified with DNA sequencing analysis. Furthermore, to evaluate the function of APE1/Ref-1 variants, the identified human APE1/Ref-1 variants were subcloned into a pCMV-Tag2A vector between EcoRI and HindIII restriction sites or a pEGFP-C1 vector between EcoRI and KpnI restriction sites. The subcloning of each cDNA was confirmed by DNA sequencing.

The human embryonic kidney epithelial 293T (HEK293T) cell line was purchased from the American Type Culture Collection (Manassas, VA, USA). Cells were grown in DMEM (Welgene) containing 10% fetal bovine serum (Thermo Fisher Scientific) and 1% antibiotics. Cells were maintained in a humidified atmosphere of 95% air and 5% CO₂, at 37°C. For transient transfection, HEK293T cells were seeded in plates at a density of 2.5×10⁵ cells/mL and incubated overnight at 37°C. HEK293T cells were transfected with each APE1/Ref-1 construct using Lipofectamine LTX (Thermo Fisher Scientific) according to the manufacturer’s protocol.

Cell viability was determined using propidium iodide staining and an automatic cell counter (ADAM-MC), as previously described [9].

A sandwich enzyme-linked immunosorbent assay (ELISA) was used to quantify the level of secreted APE1/Ref-1 in the culture medium (CM), as previously described [11].

HEK293T cells were grown to confluence on a 6-well plate (5×10⁵ cells/well) and then treated with 1μM TSA for 1 hour after changing the CM. The CM was collected at different time points and centrifuged at 1,200 rpm for 3 minutes to remove the residual cells. The supernatant was again centrifuged at 3,000 rpm for 3 minutes. The APE1/Ref-1 in the CM was immunoprecipitated using anti-APE1/Ref-1 antibody as previously reported [10]. The immune complexes were mixed with sample buffer and subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by immunoblotting.

Whole cell lysates were prepared as previously described [11]. The cell lysates were cleared by centrifugation at 14,000×g for 15 minutes, and the supernatant fraction was used for immunoblotting. Proteins were resolved by SDS-PAGE and transferred onto a polyvinylidene difluoride membrane. After blocking with 5% nonfat dry milk in Tris-buffered saline (TBS) containing 0.05% Tween 20, the membrane was incubated with the desired primary antibody overnight at 4°C. The membrane was then washed and probed with the appropriate secondary antibody, and the immunoreactive bands were visualized by an enhanced chemiluminescence method. Each membrane was stripped and reprobed with antiactin antibody to ensure equal protein loading.

A subcellular localization of each APE1/Ref-1 variant was observed in cells transfected with EGFP-tagged APE1/Ref-1 variants as previously described [12]. Cells were grown on glass coverslips and transiently transfected with 1.6 μg of plasmid from the pEGFP-tagged APE1/Ref-1 variant plasmids. Cells were fixed with 4% paraformaldehyde, and DAPI was used for 10 minutes to stain nucleus. Coverslips were mounted on microscopic slides using a fluorescence mounting solution and green fluorescence protein (GFP) signals were visualized with a fluorescence microscope (Carl Zeiss, Jena, Germany).

All data are expressed as the mean±standard error of the mean. Statistical significance of differences in measured variables was determined using one-way analysis of variance followed by Bonferroni test for multiple comparisons. Differences were considered significant at P<0.05.

Nucleotide variants of APE1/Ref-1 cDNA were evaluated in 10 patients with bladder cancer. DNA sequencing analysis data showed the identification of four different APE1/Ref-1 variants: a single mutation with Asp/Glu at codon 148 of APE1/Ref-1 (D148E), and 3 different double mutations with Ile/Val at codon 64 (I64V), Trp/Arg at codon 67 (W67R), and Glu/Gly at codon 86 (E86G) with D148E (Table 1). Furthermore, APE1/Ref-1 variants with nucleotide substitutions 318 T>A or 257 A>G were observed without amino acid changes in 2 specimens. Noticeably, the genotype with D148E was observed in 6 specimens with single or double mutations (Table 1).

To examine the association of APE1/Ref-1 variants in relation to its expression in cells, we constructed four expression vectors containing full APE1/Ref-1 cDNA with D148E, I64V/D148E, W67R/D148E, and E86G/D148E. The level of APE1/Ref-1 protein expression was determined in 293T cells, which were transiently transfected with the substitution mutant constructs. As shown in Fig. 1A, the APE1/Ref-1 proteins were observed without any fragmentation in cells expressing each substitution mutant. Lower APE1/Ref-1 protein expression compared to substitution mutants was only detected in W67R/D148E expressing cells, although the differences in protein levels were normalized against the actin loading control.

The effect of APE1/Ref-1 variants on cellular localization was further investigated using GFP-tagged substitution mutant constructs. Representative images for GFP-APE1/Ref-1, and DAPI staining in 293T cells expressing each substitution mutant was shown in Fig. 1B. Consistent with immunoblot results, each APE1/Ref-1 mutant exclusively existed in the nuclei as evident from the colocalization of green (APE1/Ref-1 variants) and blue fluorescence (DAPI staining). Remarkably, the W67R/D148E double mutant was distributed as random-dots in the nucleus, indicating a possible alteration in the localization of APE1/Ref-1 caused by the substitution mutation.

We proposed earlier that the secretory APE1/Ref-1 is a novel biomarker because the level of APE1/Ref-1 protein is elevated in the serum and urine samples of patients with bladder cancer [7,8]. Indeed, many reports have corroborated the extracellular existence and dynamic translocalization of APE1/Ref-1 from the nuclei to cytosol [11,12]. Here, we determined whether the substitution mutant of APE1/Ref-1 was secreted into CM. The secretion level of each APE1/Ref-1 mutant in CM was estimated by comparison with a recombinant APE1/Ref-1 protein standard curve. As shown in Fig. 2A, the cell viability in basal or TSA-stimulated conditions was not changed in 293T cells transfected with APE1/Ref-1 variants. The secretory activity of APE1/Ref-1 variants was evaluated using the newly developed ELISA [7] in basal and TSA-stimulated conditions. Basal secretion of APE1/Ref-1 in cells transfected with substitution mutants (I64V/D148E, E86G/D148E, and D148E) was significantly increased compared to that of wild-type APE1/Ref-1 (Fig. 2B). We previously reported that the acetylation of cellular proteins by histone deacetylase (HDAC) inhibitors, TSA, and acetylating agents induced the extracellular release of APE1/Ref-1 [9]. Interestingly, all variants with D148E, except double mutant W67R/D148E, showed increased APE1/Ref-1 secretion in response to TSA.

Immunoprecipitation using specific anti-APE1/Ref-1 antibody followed by immunoblot was performed to visualize the secreted APE1/Ref-1 in the CM. Minimal secreted APE1/Ref-1 was detected in native culture supernatants including wild-type APE1/Ref-1. The maximum secretion level of APE1/Ref-1 variants with the D148E mutation, although it was in its nonacetylated form. The expression of secreted APE1/Ref-1 in all variants was remarkably increased with exposure of cells to TSA treatment. However, the secretion of APE1/Ref-1 in double mutant W67R/D148E was not measured. Taken together, these data indicate that APE1/Ref-1 secretion is obviously caused by mutation at D148E of APE1/Ref-1, implying a correlation between D148E mutation with increased secretory activity and its role as a biomarker for bladder cancer.

Given that the secretion of D148E in APE1/Ref-1 was clearly distinct from other secretions, we directly compared the secreted amount of wild-type and D148E variants in CM for 1 hour after changing media, as represented in Fig. 3A (upper). The secretion level of the wild-type APE1/Ref-1 protein reached 0.15 ng/mL in 1 hour, whereas that of the D148E mutants showed a drastic increase to 0.65 ng/mL in the same amount of time, suggesting a high secretory activity of D148E mutants even in the absence of any stimulus (Fig. 3A bottom). Next, degradation kinetics of secreted APE1/Ref-1 D148E mutants were studied and evaluated for 6 hours in cell-free medium after stimulation with TSA for 1 hour. Nonlinear fit regression analysis showed that the half-life for degradation of the secreted wild-type and D148E was not significantly different (half-life of 1.6 hours for wild-type and 1.8 hours for D148E) in cell-free CM (Fig. 3B). Collectively, these results suggest that the increased levels of APE1/Ref-1 in the D148E variant might be due to increased secretory activity, rather than the delayed degradation of D148E.