BASP1 is upregulated in cervical cancer tissues

To investigate whether BASP1 regulates the development and progression of cervical cancer, we explored BASP1 expression in cervical cancer tissues and normal cervical epithelium tissues. Analysis of gene expression profiles (GSE9750, Fig. 1a) showed that BASP1 was significantly upregulated in cervical tissues compared with normal cervical tissue. Six paired cervical cancer tissues and matched adjacent normal cervical tissues were used to examine BASP1 protein levels. Western blotting showed that BASP1 levels were higher in cervical cancer tissues than in matched adjacent normal cervical tissues (Fig. 1b). Immunohistochemical analysis also revealed that BASP1 was upregulated in cervical cancer tissues (Fig. 1c). These results suggested that BASP1 is upregulated in cervical cancer tissues.

BASP1 levels correlate with clinical aggressiveness of cervical cancer

To determine the relationship between BASP1 levels and clinicopathological parameters, and whether BASP1 could serve as a new independent prognostic factor, we determined BASP1 levels in a cohort of 136 paraffin-embedded archived cervical cancer tissues using immunohistochemistry. Positive staining for BASP1 was observed in 98.5% (134/136) of samples, with low BASP1 staining in 59.6% (81/136) samples and high BASP1 staining in 40.4% (55/136) of the samples (Additional file 1: Table S2). BASP1 levels were higher in patients who had died during follow-up than in those had survived during follow-up (Fig. 2a). BASP1 levels were low in 67.3% (68/101) of surviving patients, and high in 32.7% (33/101). However, BASP1 levels were low in 37.1% (13/35) of the patients who died and high in 62.9% (22/35) (Fig. 2a). Kaplan–Meier survival curves demonstrated that the overall survival of the patients with low BASP1 expression was significantly longer than those with high BASP1 expression (Fig. 2b, p = 0.001). These results suggested high BASP1 levels might be an unfavorable factor for patients with cervical cancer.

BASP1 regulates tumor growth of cervical cancer

Brain abundant membrane attached signal protein 1 could serve as a new prognostic factor; therefore, we further determined whether BASP1 regulates tumor growth of cervical cancer. BASP1 levels were determined in cervical cancer cell lines and in two papillomavirus-immortalized normal cervical cell lines. BASP1 was upregulated in the cervical cancer cell lines (Fig. 3a), which was consistent with our previous results for BASP1 abundance in cervical cancer tissues (Fig. 1a, b). We determined the functional role of BASP1 in cervical cancer development by gain or loss of BASP1 in ME-180 and HT-3 cells. Western blotting showed that BASP1 levels were significantly upregulated when cells were transfected with pMSCV-BASP1, and were significantly downregulated when the cells were transfected with a short interfering RNA (siRNA) targeting BASP1 (siBASP1; Fig. 3b). The MTT assay indicated that overexpression of BASP1 promoted cervical cancer cell proliferation. A colony formation assay also showed that the cell number increased significantly when BASP1 was overexpressed (Fig. 3d). These results suggested that BASP1 promotes proliferation of cervical cancer cells. To further confirm that BASP1 regulates cell proliferation, we used cell cycle assays to determine the effect of BASP1 on cell cycle progression, overexpression of BASP1 increased the proportion of cells in the S phase from 34.52 to 50.78% in ME-180 cells, and from 31.31 to 52.58% in HT-3 cells, with a concomitant decrease in the proportion of cells in the G1/G2/M phases (Fig. 3e).

We confirmed these results by downregulating BASP1 in the indicated cervical cancer cell lines using siRNAs. The MTT assay showed decreased proliferation of cells transfected with siBASP1 compared to those transfected with a scrambled siRNA (Additional file 3: Figure S1A). The colony formation assay also showed that knockdown of BASP1 inhibited cellular proliferation (Additional file 3: Figure S1B). We also analyzed the effect of knockdown of BASP1 on cell cycle progression: The proportion of cells in the S phase decreased from 37.84 to 11.40% in ME-180 cells, and from 35.79 to 12.81% in HT-3 cells, with a concomitant increase in the proportion of cells in the G1/G2/M phases (Additional file 3: Figure S1C).

To determine the role of BASP1 in tumorigenicity, we used an anchorage-independent growth assay to determine the effect of BASP1 on tumorigenicity in vitro. BASP1 overexpression increased the anchorage-independent growth ability of indicated cells significantly, as shown by increased colony numbers and sizes (Fig. 4a). When BASP1 expression as knocked down, we observed a significant decrease the cells’ anchorage-independent growth ability (Fig. 4a). This suggested that BASP1 promoted the tumorigenicity of cervical cancer.

To confirm that BASP1 regulates cell proliferation and tumorigenicity, we determine the effect of BASP1 on tumor growth in vivo. We transplanted the indicated cells with BASP1 overexpression or knockdown into a subcutaneous area of nude mice. Overexpression of BASP1 promoted cervical tumor growth in the nude mice, and downregulation of BASP1 inhibited it (Fig. 4b, c). Ki67 is a marker for cell proliferation [10]; therefore, we determined Ki67 levels in the tumors grown in the nude mice, and found that Ki67 levels were upregulated in tumors overexpressing BASP1 and downregulated in BASP1 knockdown tumors (Fig. 4d). Together, these results suggested that BASP1 promotes tumor growth. This result also supported our clinical investigation, in which BASP1 levels correlated with T classification.

Cell culture and transfection

Human cervical cancer cell lines Ca Ski, MS751, ME-180, SiHa, HT-3, HeLa, C4-I and SW756 were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Natocor) supplemented with 10% fetal bovine serum (FBS) (GIBCO). Ectocervical Ect1/E6E7 and endocervical End1/E6E7 cells were cultured in Keratinocyte-Serum Free medium (GIBCO-BRL, 17005-042, USA) supplemented with 0.1 ng/mL human recombinant EGF, 0.05 mg/mL bovine pituitary extract (BPE, Sigma) and 0.4 mM CaCl₂ (Natocor). The cells were maintained in 5% CO₂ at 37 °C.

The full-length BASP1 cDNA was cloned into vector pMSCV-puro; pMSCV-empty vector was used as the negative control. pMSCV-BASP1 and pMSCV-empty vector were cotransfected with the IK packaging plasmid into 293T cells using the calcium phosphate transfection method. Supernatants were collected at 48 h after transfection, and infected with cervical cells for 12 h in the presence of polybrene (2.5 μg/mL). Puromycin was used to select the transfected cell lines.

For BASP1 knockdown experiments, two siRNAs for BASP1 and one scrambled siRNA were synthesized by Guangzhou RiboBio Co (Guangzhou, China). The sequences used to downregulate BASP1 were: siBASP1#1: 5′CGGGATCCATGGGA3′, siBASP1#2: 5′CGGAATTCTCACTCT3′. 50 nM of the siRNA was transfected into ME-180 and HT-3 cells using the Lipofectamine RNAiMAX Transfection Reagent (Life Technologies).

Patients and tissue samples

Six cervical cancer tissues and matched adjacent normal cervical epithelium tissues were obtained from the First Affiliated Hospital of Sun Yat-sen University. Guangzhou. These samples were snap-frozen immediately and stored in liquid nitrogen until use. To further analyze the relationship between BASP1 expression and the clinicopathological parameters, a cohort of 136 paraffin-embedded cervical cancer tissues was used. These tissues were diagnosed histopathologically and clinically at the First Affiliated Hospital of Sun Yat-sen University. For the use of above clinical samples for research purposes, prior patient’s consent and approval from the Institute Research Ethics Committee of the First Affiliated Hospital of Sun Yat-sen University were obtained. The detailed clinicopathological parameters are shown in Additional file 5: Table S1.

Western blotting and immunohistochemistry (IHC)

Western blotting was performed according to standard methods, as described previously [19], using anti-BASP1 (ab101855, Abcam), anti-Ki67 (sc-7846, Santa Cruz) antibodies. The membranes were stripped and reprobed with anti-β-actin antibodies as a loading control. The band intensity was analyzed using Image J software.

IHC was performed as described previously [20] using the anti-BASP1 antibody (ab101855, Abcam). The results of staining were scored independently by two pathologists blinded to the clinical outcome, and were based on both the proportion of positively stained tumors cells and the intensity of staining. The proportion of stained tumor cells was scored as follows: score 0, no positive cells; score 1, up to 25% positive cells; score 2, 26–50% positive cells; score 3, 51–75% positive cells; score 4, over 75% positive cells. The intensity of staining was determined as: 0 (no staining), 1 (light yellow = weak staining), 2 (yellow brown = moderate staining), and 3 (brown = strong staining). The staining index (SI) was determined as the product of staining intensity × percentage of positive tumor cells. Cutoff values for high and low expression of BASP1 were chosen based on a measurement of heterogeneity using the log-rank test with respect to overall survival. An SI score of greater than or equal to 6 was considered to be high expression, and an SI score of less than 6 was considered low expression.

Cell proliferation and cell cycle assay

To examine the role of BASP1 in the proliferation of cervical cancer cells, an MTT assay and colony formation assay were performed using previous described methods [21].

Cell cycle analysis was performed using a previously described method [21]. Briefly, cells were harvested and washed in cold PBS followed by fixation in 70% alcohol overnight at 4 °C. After washing with cold PBS three times, the cells were resuspended in PBS solution with 20 μg/mL propidium iodide (Sigma) and μg/mL RNase A for 30 min at 37 °C. Samples were then analyzed using a FACSCalibur cytometer (Becton–Dickinson).

Anchorage-independent growth assay

An anchorage-independent growth assay was performed according to a previously described method [21, 22]. Briefly, 500 cells were suspended in 2 mL of complete medium containing 0.3% agar (Sigma). The agar–cell mixture was plated on top of a solid bottom layer containing 1% complete medium agar mixture. After 10 days, colonies that contained more than 50 cells or were larger than 0.1 mm in diameter were counted. The experiment was performed in triplicate.

Growth of tumor xenografts in nude mice

Animal studies were performed according to institutional guidelines. BALB/c nude mice (4–5 weeks old) were used to make a xenograft model using ME-180 cervical cancer cell lines. Five mice were assigned randomly to each group and injected with ME-180 transfected with pMSCV-BASP1, pMSCV-empty, scramble siRNA, or BASP1 siRNA, and used to determine the role of BASP1 in tumor progression. 1 × 10⁶ cells were injected into the subcutaneous sites of nude mice. The tumor volume was calculated every 2 days for 1 month. The tumors were excised and subjected to protein extraction to determine Ki67 levels using western blotting.

Statistical analysis

All statistical analyses were performed using the SPSS 19.0 statistical software package. Results are presented as the mean ± standard deviation (SD) for at last three repeated individual experiments for each group. Chi square and Fisher’s exact tests were used to analyze the relationship between BASP1 levels and clinicopathological parameters. Bivariate correlations between variables were calculated using Spearman’s rank correlation coefficients. The survival curve was plotted using Kaplan–Meier survival analysis and compared using a log-rank test. Univariate and multivariate Cox regression analyses were used to estimate the significance of various variables for survival. A value of p < 0.05 was considered significant in all cases.