The MDS cell line SKM-1 was provided by Professor Jianfeng Zhou of Tongji Medical College of Huazhong University of Science and Technology (Wuhan, China). Cells were cultured in RPMI-1640 basic medium (Corning Incorporated, Corning, NY, USA) supplemented with heat-inactivated 10% fetal bovine serum (Capricorn Scientific GmbH, Ebsdorfergrund, Germany) without antibiotic. All cells were cultured at 37°C in a CO₂ incubator with 5% CO₂ and 95% humidified air. The culture medium was changed every 2–3 days, and cells were maintained without any significant cell morphological change or biological response. LY294002 and z-VAD-fmk, purchased from Selleck Chemicals (Houston, TX, USA), were used to treat cells (5×10⁴ cells) at 20 µM for 1 h.

The SPAG6-targeting short hairpin RNA (shRNA) lentiviral vector (SPAG6-shRNA) and control non-specific lentiviral vector (NC-shRNA) were constructed as reported previously (23). The target sequence for SPAG6 shRNA was 5′-TGATGCTAAATTGAAGCAT-3′ and that for nonsense negative control (NC) was 5′-TTCTC CGAACGTGTCACGT-3′. Lentiviral particles were produced by Shanghai GeneChem Co., Ltd. (Shanghai, China) by transiently cotransfecting lentiviral vector and helper plasmids (Addgene, Inc., Cambridge, MA, USA) into 293T cells (American Type Culture Collection, Manassas, VA, USA), followed by concentrating by ultracentrifugation for 2 h at 20,000 × g at 4°C. Lentiviral plasmids were transfected into 293T cells (20 µg plasmids per 10 cm tissue culture dishes) via FuGENE^® HD Transfection reagent (Promega Corporation, Madison, WI, USA), following the protocol from Addgene. SKM-1 cells at the exponential growth stage were plated in 6-well plates (5×10⁴ cells per well) and infected with SPAG6-shRNA or NC-shRNA lentivirus at a multiplicity of infection of 20 in the presence of 5 µg/ml polybrene. All cells were washed and resuspended in complete medium at 10 h after infection. The infection efficiencies and apoptosis rates of SKM-1 cells were evaluated by flow cytometry by detecting the GFP and Annexin V staining on day 5 after lentiviral infection.

Lentivirus-infected SKM-1 cells were collected, washed twice with PBS, and then resuspended in 800 µl PBS. Cell apoptosis rates were detected using Annexin V and 7-aminoactinomycin D (7-AAD) double-staining, according to the manufacturer's instructions (BD Biosciences, San Jose, CA, USA). Flow cytometry was performed on a FACSCalibur machine (BD Biosciences) in the Life Science Department of Chongqing Medical University (Chongqing, China), and data were analyzed by FlowJo 7.6 software (FlowJo LLC, Ashland, OR, USA).

Total RNA was extracted with TRIzol^® reagent in accordance with the manufacturer's protocol (Takara Biotechnology Co., Ltd., Dalian, China). RNA (500 ng per sample) was reverse transcribed at 42°C for 1 h into cDNA using the PrimeScript™RT reagent kit (Takara Biotechnology Co., Ltd.). PCRs were performed using a CFX-Connect Real-Time PCR system and SYBRGreen (Bio-Rad Laboratories, Inc., Hercules, CA, USA) for a total of 40 cycles, following standard assay procedures. The cycling parameters were 95°C for 30 sec, then 40 cycles of 95°C for 5 sec and 60°C for 39 sec. The relative gene expression was calculated by the 2^−ΔΔCq method (24), and GAPDH was used as the endogenous control gene. The PCR primer pairs were designed based on the cDNA sequences (NCBI source: Homo sapiens), and the primer sequences were as follows: SPAG6 forward, 5′-CCTTTCAGCTCTCAGTCAGGTTTC-3′ and reverse, 5′-TCTTCACGTTTCATCCTTGTCCTT-3′ (accession number: NM_012443.3); PTEN forward, 5′-ACACGACGGGAAGACAAGTT-3′ and reverse, 5′-CTGGTCCTGGTATGAAGAATG-3′ (accession number: NM_000314.6); DNA methyltransferase 1 (DNMT1) forward, 5′-CAGGCAGTTCAACACCCTCATC-3′ and reverse, 5′-GCTGAAGAAGCCGTCCCACT-3′ (accession number: NM_001130823.2); GAPDH forward, 5′-CTTTGGTATCGTGGAAGGACTC-3′ and reverse, 5′-GTAGAGGCAGGGATGATGTTCT-3′ (accession number: NM_002046.6).

Cells were lysed on ice for 30 min using radioimmunoprecipitation assay lysis buffer (Beyotime Institute of Biotechnology, Haimen, China) supplemented with 1% phenylmethylsulfonyl fluoride (Beyotime Institute of Biotechnology). The protein concentrations were determined using the BCA Protein assay kit (Beyotime Institute of Biotechnology). Protein samples were denatured by boiling following mixing with loading buffer. A total of 50 µg of protein per lane was separated by 8–12% sodium dodecylsulfate-polyacrylamide gel electrophoresis (Bio-Rad Laboratories, Inc.) and transferred onto polyvinylidene difluoride membranes. Following blocking for 2 h with 5% skim milk, or with 5% bovine serum albumin (Biosharp, Hefei, China) for phosphorylated proteins, at room temperature, these membranes were incubated with specific antibodies overnight at 4°C. Primary antibodies against the following proteins were used: SPAG6, caspase-8, DNMT1 (cat. nos. ab155653, ab108333, and ab188453, respectively; 1:1,000 dilution; all from Abcam, Cambridge, UK); PTEN, AKT, phospho-AKT [P-AKT(S473)], apoptosis regulator Bcl-2 (cat. nos. 9188T, 4685T, 4060T, and 3498T, respectively; 1:1,000 dilution; Cell Signaling Technology, Inc., Danvers, MA, USA); induced myeloid leukemia cell differentiation protein Mcl-1 (Mcl-1), caspase-9, cytochrome c (cat. nos. sc-12756, sc-56076, and sc-13561, respectively; 1:500 dilution; Santa Cruz Biotechnology, Inc., Dallas, TX, USA); caspase-3 (cat. no. YT0656; 1:500 dilution; ImmunoWay Biotechnology Company, Plano, TX, USA); GAPDH (cat. no. AG019; 1:1,500 dilution; Beyotime Institute of Biotechnology). Subsequently, the membranes were washed with Tris-buffered saline Tween-20 (TBST) five times (5 min each) and incubated with a goat anti-rabbit or goat anti-mouse peroxidase-conjugated secondary antibody (cat. nos. A0216 and A0208, respectively; 1:3,000 dilution; Beyotime Institute of Biotechnology) for 1 h at room temperature. After washing with TBST another five times (5 min each), the membranes were processed to detect antigen signals with an enhanced chemiluminescence kit (Advansta, Inc., Menlo Park, CA, USA). The protein expression levels were analyzed using Vilber Fusion software (Fusion FX5 Spectra; Vilber Lourmat, Marne-la-Vallée, France).

Nonobese diabetic/severe combined immunodeficient mice (NOD/SCID; 4–5 weeks old, female, with the average weight of 17.4 g) were purchased from Beijing HFK Bioscience Co., Ltd. (Beijing, China) and housed in the animal breeding facilities at the Laboratory Animal Center of Chongqing Medical University (Chongqing, China). Mice were randomly divided into two groups (5 mice per group) and were inoculated subcutaneously in the right flank with SKM-1 cells (1×10⁷ cells per mouse) that were stably infected with the SPAG6-shRNA or the NC-shRNA lentivirus. A caliper was used for measuring the maximum lengths and widths of the xenografts, and the formula 0.523 × length × width² was used for calculating the tumor volumes, as reported previously (25). The mice were monitored every 3 days for signs of body weight loss and tumor growth. At 4 weeks after tumor cell inoculation, the mice were sacrificed by cervical vertebrae dislocation, and tumor tissues were collected for further analyses. All procedures involving experimental animals were approved by the University Committee on the Use and Care of Animals at Chongqing Medical University.

Parts of each tumor tissue were fixed with 4% polyoxymethylene for 8 h and embedded in wax, cut into 4-µm-thick slices. The immunohistochemical analyses were detected using an assay kit (Beyotime Institute of Biotechnology). Following deparaffinization (three times in xylene for 15 min, three times in ethanol for 5 min, and three times in PBS for 3 min) and permeabilization (freshly prepared 0.1% Triton X-100 in 0.1% sodium citrate for 2 min on ice), the sections were first put into boiled citrate solution (0.01 M, pH 6.0) for 15 min at 95°C. As the temperature cooled, the sections were washed five times (3 min each) in PBS and blocked with 10% goat serum (Beyotime Institute of Biotechnology) for 1 h at room temperature. Following incubation overnight with primary antibodies at 4°C, the sections were washed five times (5 min each) in PBS, and incubated with biotinylated secondary antibodies (1:1,000 dilution; cat. no. 610438, BD Biosciences) for 1 h at 4°C. Primary antibodies against the following proteins were used: SPAG6, poly(ADP-ribose) polymerase (PARP; cat. nos. ab155653 and ab32138, respectively; 1:200 dilution; Abcam); cleaved-caspase-8 (Asp391; 1:200 dilution; Cell Signaling Technology, Inc.); BH3-interacting domain death agonist (BID; cat. no. YT0656; 1:50 dilution; ImmunoWay Biotechnology Company). Then, the sections were washed five times (5 min each) and then incubated with a streptavidin-peroxidase complex (ready-to-use, Zymed; Thermo Fisher Scientific, Inc., Waltham, MA, USA) for 1 h at room temperature. Finally, the slides were counterstained with hematoxylin (3 min at room temperature) and detected using diaminobenzidine (5 min at room temperature) under a light microscope.

All data are expressed as the mean ± standard deviation, and the corresponding experiments were repeated with technical replicates at least three times. One-way analysis of variance was used to compare differences between multiple groups, and a least significant difference post hoc test was used for multiple comparisons for continuous variables. The Student's t-test was used to analyze differences between two groups. Statistical analysis was performed with SPSS 18.0 software (SPSS, Inc., Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant difference.

Initially, experiments were performed to demonstrate that the knockdown of SPAG6 expression was achieved by lentivirus-mediated delivery of shRNA against SPAG6 in SKM-1 cells. At 5 days after viral infection, the expression of green fluorescence protein and the high infection efficiencies in both SPAG6-shRNA and control (NC-shRNA) virus-transduced SKM-1 cells demonstrated that the lentiviral particles SPAG6-shRNA and NC-shRNA were able to infect SKM-1 cells effectively (data not shown), as previously reported (19). Subsequent RT-qPCR and western blot analyses also demonstrated that SKM-1 cells infected with SPAG6-shRNA virus had significantly decreased mRNA (Fig. 1A) and protein (Fig. 1B) levels of SPAG6, compared with NC-shRNA-infected cells. These results further indicate that the cell model with downregulated SPAG6 expression was successfully established. Consistent with our previous report (18), knockdown of SPAG6 substantially increased the expression of PTEN at the mRNA (Fig. 1A) and protein levels (Fig. 1B) in SKM-1 cells. These findings suggest that SPAG6 is involved in regulating PTEN expression in vitro.

PTEN affects cellular apoptosis by negatively regulating the mitochondrial apoptotic pathway via PI3K/AKT signaling (26). A positive association between PTEN expression and cell apoptosis was also observed in the SKM-1 cells. As reported in a previous study (19), the SPAG6-shRNA-transduced SKM-1 cells exhibited substantially higher apoptosis rates than the NC-shRNA-transduced SKM-1 cells by the Annexin V and 7-AAD staining assays. To determine whether SPAG6 regulates apoptosis through the PI3K/AKT pathway, the activation of AKT was detected. SKM-1 cells with significantly decreased SPAG6 expression had markedly reduced P-AKT, although the total AKT levels were not significantly different to control SKM-1 cells (Fig. 2A and B). Subsequently, the expression of factors associated with the mitochondrial apoptotic pathway was examined. As shown in Fig. 2C and D, the protein levels of Mcl-1 and Bcl-2 were lower in SPAG6-shRNA cells compared with control cells, while the levels of caspase-9 and cytochrome c were higher in the SPAG6-shRNA group than in the control group. Taken together, the results indicated that SPAG6 regulated apoptosis by activating the PTEN/PI3K/AKT-mediated mitochondrial apoptotic pathway.

To determine whether PTEN activation and AKT inactivation were sufficient for induction of apoptosis in SKM-1 cells, the PI3K inhibitor LY294002. As shown in Fig. 3A and B, the apoptosis rates of LY294002-treated and SPAG6-shRNA lentivirus-transduced SKM-1 cells were significantly higher than that in the control cells. Notably, cotreatment with LY294002 and SPAG6-shRNA markedly promoted apoptosis, and these SKM-1 cells displayed a higher apoptotic rate than any of the other groups. Western blot analysis indicated that the PTEN level in cells infected with SPAG6-shRNA lentivirus alone was higher than that in cells treated with LY294002 alone, although it was significantly lower than that in cells cotreated with LY294002 and SPAG6-shRNA lentivirus (Fig. 3C and D). Correspondingly, an opposing pattern of changes for AKT phosphorylation was observed, which was attributed to the negative regulation of AKT activity by PTEN (Fig. 3C and D). Furthermore, co-administration of LY294002 significantly enhanced SPAG6-shRNA lentivirus-mediated Mcl-1 downregulation, caspase-9 and caspase-3 activation, and cytochrome c release in SKM-1 cells (Fig. 3C and D). Taken together, these findings suggest that PTEN-mediated AKT inactivation has a critical role in enhancing apoptosis in SKM-1 cells with SPAG6 knockdown.

Apoptosis can be initiated through either death receptor activation or the mitochondrial apoptotic pathway. The cleavage/activation of caspase-8 and PARP also downregulate AKT activity and indirectly activate the mitochondrial apoptotic pathway (27). Thus, it is possible that PTEN activation followed by dephosphorylation of AKT in SKM-1 cells with SPAG6 knockdown is caused by caspase activation. To elucidate the role of caspase activation in PTEN activation, SKM-1 cells were treated with z-VAD-fmk, a cell-permeant pan-caspase inhibitor, to block the caspase-dependent apoptotic pathway prior to knockdown of SPAG6. Western blot analysis demonstrated that although z-VAD-fmk slightly inhibited the expression of total caspase-8/9, it did not block PTEN activation in SKM-1 cells with silenced SPAG6 (Fig. 4A and B). In the presence of z-VAD-fmk, knockdown of SPAG6 was still able to increase the level of total caspase-8/9/3 compared with z-VAD-fmk + NC-shRNA treatment, suggesting the progression of cell apoptosis. These data clearly indicate that caspase activation is a consequence of SPAG6 silencing-mediated AKT inactivation, but not the cause of PTEN activation in SKM-1 cells.

There is growing evidence indicating that DNA methylation can silence tumor suppressors through high expression of DNA methyltransferases (DNMTs) to facilitate tumorigenesis and tumor development (28). To explore whether DNMTs have a role in upregulating PTEN expression, the expression level of DNMT1 was determined by RT-qPCR and western blot analysis. Consistently, the mRNA and protein levels of DNMT1 were significantly decreased in SKM-1 cells infected with SPAG6-shRNA lentivirus, compared to that in the control cells (Fig. 4C and D). Thus, potentially a positive association between SPAG6 and DNMT1 expression may contribute to SPAG6 silencing-induced PTEN activation.

To determine whether the in vitro findings were applicable in vivo, NOD/SCID mice were inoculated with SKM-1 cells with or without SPAG6 knockdown by subcutaneous injection. Immunohistochemical analyses and western blotting were performed to evaluate the morphological changes and induction of apoptosis in tumor cells (Fig. 5). The tumor sizes and tumor weights of the mice inoculated with SKM-1 cells pre-infected by SPAG6-shRNA lentivirus were significantly less than those of the control mice (Fig. 5A). As reported in our previous study (18), tumor sections from the mice inoculated with SPAG6-shRNA-transduced SKM-1 cells showed an apparently necrotic morphology, infiltration of inflammatory cells, disordered and irregular tumor cell arrangement, and tumor cells with an increased nuclear cytoplasmic ratio. In addition, the number of dark brown-colored apoptotic cells from the SPAG6-shRNA lentivirus group was significantly higher than that from the NC-shRNA lentivirus group (18). Furthermore, tumor cells with SPAG6 knockdown exhibited evident signs of cell apoptosis, as indicated by significantly altered signals from the results of immunohistochemical staining for apoptosis-associated molecules, including BID, PARP, Bcl-2, cleaved-caspase-8, and cleaved-caspase-3 between these two groups (Fig. 5B). The western blot results further substantiated the upregulation of pro-apoptotic factors like cleaved-caspase-8, cleaved-caspase-3, PARP and BID, and downregulation of the anti-apoptotic factor Bcl-2 in tumor cells with SPAG6 knockdown (Fig. 5C and D). Collectively, these findings support that that SPAG6 knockdown induced apoptosis of SKM-1 cells via the extrinsic and intrinsic pathways during in vitro culturing and in the in vivo xenograft mouse model.