A total of 62 LSCC patients who received no radiotherapy or chemotherapy and who underwent surgery between 2011 and 2015 at the First Affiliated Hospital of Wenzhou Medical University were examined in this retrospective research. All samples were from 61 males and 1 female with ages ranging from 43 to 79 years (mean, 64.5 years). The present study was approved by the Clinical Medicine Ethics Committee of The First Affiliated Hospital of Wenzhou Medical University, and consent for enrollment in the study was provided by the patients. The study protocol was carried out according to the principles of the Helsinki Declaration. In total, 50 formalin-fixed, paraffin-embedded tumour sections were obtained from the pathology department for immunohistochemical analysis, and 7 samples of histologically normal laryngeal tissues surrounding the tumours were treated as control tissues. Another 12 pairs of fresh-frozen LSCC tissues which were used for western blot analysis, including cancerous and matched adjacent non-cancerous tissues from the same patients, were maintained at −80°C. Patient follow-up was performed via telephone calls until October 2016. The outcome event was defined according to the date of death.

The human LSCC, Hep-2 cell line, was obtained from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). In brief, cells were cultivated in a complete medium containing RPMI-1640 medium, 10% fetal bovine serum (FBS) (both from HyClone, Logan, UT, USA) and 1% antibiotics (100 U/ml penicillin/streptomycin) in a humidified incubator at 37°C with 5% CO₂. Cisplatin was obtained from Qilu Pharmaceutical Co., Ltd. (Shandong, China).

Briefly, total protein was extracted and the concentration was measured with a bicinchoninic acid (BCA; Beyotime, Jiangsu, China) protein assay kit. For each sample, ~30–50 µg of protein were run on a 10% polyacrylamide gel, and then transferred to polyvinylidene fluoride (PVDF) membranes. The membranes were immunoblotted with the primary antibodies, followed by the secondary HRP-conjugated antibody. Immunoreactive proteins were visualized using ECL western blot detection kit (Advansta, Menlo Park, CA, USA) and quantified employing Image Lab software (Bio-Rad, Hercules, CA, USA). Information concerning the antibodies was as follows: rabbit anti-GAPDH polyclonal antibody (ab-p-r001; 1:1,000; Xianzhi, Zhejiang, China); rabbit anti-HSP47 monoclonal antibody (ab109117; 1:3,000; Abcam, Cambridge, UK); rabbit anti-Bcl-2 monoclonal antibody (#4223; 1:1,000); rabbit anti-Bax monoclonal antibody (#5023; 1:1,000); rabbit anti-PARP monoclonal antibody (#9532; 1:1,000); rabbit anti-cleaved caspase-7 monoclonal antibody (#8438; 1:1,000); rabbit anti-cleaved caspase-8 monoclonal antibody (#9496; 1:1,000); rabbit anti-cleaved caspase-9 monoclonal antibody (#9505; 1:1,000) (all from Cell Signaling Technology, Danvers, MA, USA).

Fifty formalin-fixed, paraffin-embedded tumour sections and 7 control tissues were stained with HSP47-specific antibody raised against the amino acid sequence of human HSP47 (ab109117; 1:300; Abcam). Briefly, the sections (5-µm) were deparaffinized in dimethylbenzene, dehydrated in graded alcohols, subjected to peroxidase blocking and then retrieved with citrate buffer. The sections were immunoblotted with the anti-HSP47 antibody at a dilution of 1:300 overnight at 4°C, followed by incubation with the secondary HRP-conjugated antibody. The sections were then counterstained with haematoxylin and observed under a biological imaging microscope (BX53; Olympus, Tokyo, Japan), which also was used to obtain images. All of the immunohistochemically stained sections were evaluated via a double-blind semi-quantitative analysis to avoid artificial errors. Cytoplasmic staining was regarded as positive. In total, 10 high-power fields (magnification, ×200) within the stained tumour cytoplasm were selected to analyse the expression of HSP47. In the present study, to evaluate prognosis, we divided the extent of staining into 2 groups: a high-score group (staining extent ≥50%) and a low-score group (staining extent <50%). These cut-off values have been used in past studies (20,21).

According to the protocol of the manufacturer, total RNA from the Hep-2 cells was extracted using Trizol (Invitrogen, Carlsbad, CA, USA). The first-strand cDNAs were synthesized using a reverse transcription kit (Takara, Dojindo, Japan). The synthesized cDNA (2 µl) was amplified (final volume, 20 µl) using a SYBR-Green PCR kit (Takara) and loaded on the ABI 7500 sequence detection system (Applied Biosystems, Foster City, CA, USA). DNA amplification conditions were as follows: 95°C for 30 sec and subsequent 40 cycles, 95°C for 5 sec, 60°C for 34 sec. The primer information was as follows: HSP47 (human forward) 5′-CACCGCCTTTGAGTTGGACAC-3′ and HSP47 (human reverse) 5′-GGCGCCCAATGAATAGCAG-3′; GAPDH (human forward) 5′-ATGAGCCCCAGCCTTCTCCAT-3′ and GAPDH (human reverse) 5′-GGTCGGAGTCAACGGATTTG-3′. The 2^−ΔΔCt analysis was used to calculate gene expression levels.

According to the reagent protocol, the cells were transiently transfected with plasmid DNA or small interfering RNA (siRNA) using Lipofectamine 3000 (Invitrogen). In regards to the plasmid transfection, Lipofectamine 3000 was diluted in Opti-MEM medium after the cells were seeded at 80–90% confluence. Meanwhile plasmid DNA and P3000 reagent were diluted in Opti-MEM medium. Each was incubated for 5 min, respectively, and then mixed together for 20 min. Finally, the DNA-lipid complex was added to the cells. Gene transfection of siRNA was the same, except for the addition of P3000 reagent when diluting the siRNA. Quantitative RT-PCR and western blotting were used to compare transfection efficiency. (pCDH)-HSP47, (pCDH)-control, siHSP47 and siControl were all obtained from GenePharma (Shanghai, China). Information regarding the siRNAs was as follows: siHSP47, 5′-GCAGCAAGCAGCACUACAATT-3′ and siControl, 5′-TTCTCCGAACGTGTCACGTTT-3′.

Cell viability was assessed using Cell Counting Kit-8 (CCK-8; Dojindo, Kumamoto, Japan). After transfection for 24 h, the cells were plated into 96-well plates at a density of 1×10³ cells/well, and then subjected to cell proliferation assay by CCK-8 at 24, 48, 72 and 96 h. For chemosensitivity analysis, the cells were transfected with (pCDH)-HSP47 or siHSP47 for 24 h, and then plated into 96-well plates at a density of 4×10³ cells/well. The cells were subsequently treated with cisplatin (9 µM/l) for 24–96 h, and subjected to cell proliferation assay by CCK-8 at 24, 48, 72 and 96 h. Each group contained 3 parallel wells. CCK-8 (10 µl) was added to the cell culture medium (90 µl) according to the manufacturers instructions. The cells were subsequently cultured for 2 h in an incubator at 37°C and the absorbance was measured at 450 nm.

The cells on the lower surface of the membrane were quantified to assess the invasive potential. As follows, Matrigel matrix gel (BD Biosciences, San Jose, CA, USA) was diluted with serum-free culture medium at 1:8, for coating of the 24-well transwell chambers (Corning Costar, USA) which were placed in an incubator at 37°C overnight. The upper chamber was filled with the transfected cells (1×10⁵ with 200 µl of serum-free culture medium), and meanwhile the lower chamber was filled with 600 µl of complete medium that contained 10% FBS. Following 18 h of cultivation, the cells in the upper chamber were wiped off with a cotton swab gently, and the invading cells that adhered to the lower membrane were fixed in paraformaldehyde for 15 min, followed by crystal violet staining for 1 h.

The relative distance of Hep-2 cells moving into a wounded space was used to evaluate cell migration ability. Briefly, cells (5×10⁵) were plated into 6-well plates in serum-free medium for 24 h. When achieving a monolayer, the confluent monolayers were scratched using a 10 µl pipette tip and the floating cells were removed by phosphate-buffered saline (PBS). Finally, fresh culture medium was added to the wells. Phase contrast microscopy was performed to measure the relative distance of the wound.

Cells (4×10⁵) were seeded into 6-well plates and transfected with the plasmid. After 48 h of culture, the cells were trypsinized, washed twice with PBS, and then either incubated with fluorescein isothiocyanate (FITC)-Annexin V/propidium iodide (PI) or only stained with PI (both from MultiSciences Biotech Co., Ltd., Zhejiang, China) in the dark according to the reagent protocol. A flow cytometer (FCM; Becton-Dickinson, San Jose, CA, USA) was used to calculated the percentages of apoptotic and necrotic cells and the DNA content.

All analyses were carried out using SPSS 16.0 software (SPSS, Inc., Chicago, IL, USA). Results are presented as mean values ± SD. The distinct expression of HSP47 protein between tumour tissues and controls was examined using paired-sample t-tests. The difference in overall survival (OS) between 2 groups was detected by the long-rank test. The prognostic value of HSP47 protein expression in LSCC patients was analysed by Kaplan-Meier survival curve. Tumour cell viability and invasion assays were assessed using independent t-tests. P<0.05 was considered to indicate a statistical significant result.

To explore the expression of HSP47 in LSCC, we quantified the protein levels of HSP47 in matched cancerous and adjacent non-cancerous tissues from 12 LSCC patients by immunoblot analysis. As shown in Fig. 1A and B, in contrast to the controls, the level of HSP47 was significantly decreased in the tumour tissues (P<0.05). Furthermore, immunohistochemistry was performed on 50 formalin-fixed, paraffin-embedded LSCC and 7 adjacent normal laryngeal tissues. Based on the histopathologic grade, we classified the samples into 4 groups: normal, high differentiation, medium differentiation and low differentiation. As shown in Fig. 1C, the expression of HSP47 protein in the normal control tissues was obviously higher than that in the LSCC tissues with low to high differentiation, and the HSP47 protein was mainly localized in the cytoplasm. Moreover, the average immunohistochemical scores for HSP47 expression were observed to decrease as the tumour pathological grade of LSCC increased (Fig. 1D). Consistent with the results of the western blotting, LSCC tissues showed lower expression of HSP47 protein compared with that noted in the adjacent normal laryngeal tissues (Fig. 1E; 71.0±7.1 vs. 45.7±4.0%, respectively; P<0.001).

We used a Kaplan-Meier survival curve to analyse the prognostic value of HSP47 protein expression in LSCC patients. We divided patients according to HSP47 protein expression levels into a low-score group (32.8±2.4; 95% CI, 27.9–37.7) and a high-score group (70.0±2.4; 95% CI, 65.0–85.0). As shown in Fig. 1F, the low-score group had a shorter OS time compared with the high-score group. The median OS time in the low-score group was 38±2.5 months, while in the high-score group was 46±2.9 months (P=0.001).

Taken together, these results indicate that HSP47 may be an important prognostic factor for LSCC patients.

To analyze the effects of HSP47 on Hep-2 cells, cell proliferation analysis was performed after transfection with either a plasmid encoding HSP-47 [(pCDH)-HSP47)] or HSP47-specific siRNA (siHSP47) to upregulate or knockdown HSP47 levels in Hep-2 cells. As shown in Fig. 2A, (pCDH)-HSP47-transfected cells showed a marked increase in HSP47 mRNA expression as confirmed by real-time RT-PCR, while siHSP47-transfected cells exhibited a significant decrease (Fig. 2D). After transfection for 24 or 48 h, HSP47 mRNA expression was significantly increased, by 10.3 and 4.6 times, respectively, in the (pCDH)-HSP47-transfected cells compared with the (pCDH)-control cells. In contrast, siHSP47-transfected cells respectively exhibited 68 and 69% lower expression than that noted in the siControl cells. The level of HSP47 protein expression was also detected by immunoblot analysis, which revealed an increase in the Hep-2-HSP47 cells (Fig. 2B) and a decrease in the Hep-2-siHSP47 cells (Fig. 2E).

Furthermore, we assessed the effect of HSP47 expression on cell viability using CCK-8 assay. Compared with the control, Hep-2 cells exhibited a marked reduction (Fig. 2C) in cell viability when HSP47 was upregulated, while there was a significant increase (Fig. 2F) when HSP47 was silenced. At 96 h after transfection, the cells with upregulation of HSP47 were 19.3% less viable when compared with the (pCDH)-control cells. Meanwhile, HSP47-knockdown cells were 16.4% more viable than the siControl cells. The present study indicated that upregulation of HSP47 inhibited the growth of Hep-2 cells, whereas HSP47 downregulation stimulated proliferation.

We also investigated whether HSP47 upregulation inhibits cell migration and invasive abilities by wound-healing and transwell invasion assays, respectively. HSP47 upregulation markedly diminished wound-healing migration in the Hep-2 cells (Fig. 3A), and the ability to migrate was decreased by 20% (Fig. 3B; P<0.05). Moreover, upregulation of HSP47 resulted in notably decreased migration (by 62.3%) into the lower chamber in triplicate independent assays (Fig. 3C and D; P<0.001). From these assays, we demonstrated that HSP47 upregulation suppressed Hep-2 cell mobility.

We next evaluated the potential effect on sensitivity to chemotherapy by HSP47 in LSCC cells. As shown in Fig. 4A, upregulation of HSP47 significantly increased the cell chemosensitivity to cisplatin, by 25.9%, at 48 h. In contrast, knockdown of HSP47 led to more resistance to cisplatin chemotherapy, with a 16.2% increase compared with the negative-control group at 48 h (Fig. 4B). These results revealed that HSP47 overexpression sensitized the Hep-2 cell line to cisplatin chemotherapy, whereas HSP47 silencing protected cells against cisplatin.

To investigate the mechanisms underlying the growth inhibition and chemosensitivity enhancement noted in the HSP47-upregulated cells, we assessed cell cycle distribution and apoptosis by flow cytometric assay at 48 h after transfection. As shown in Fig. 5A, upregulation of HSP47 resulted in marked G1 phase arrest, with an increase of 11.6%, compared with the cell cycle distribution of the control cells. This G1 phase arrest was accompanied by a 12.6% decrease in the percentage of S phase cells. Meanwhile, the percentage of cells in the lower right quadrant, which represent early apoptotic cells, increased by 221% compared with the control (Fig. 5B). However, no obvious difference in late apoptotic cells or necrotic cells was noted. These findings indicated that overexpression of HSP47 mainly arrested Hep-2 cells in the G1 phase and promoted early apoptosis.

To further elucidate the molecules downstream of HSP47, the expression of apoptosis-related biomarkers was detected via western blotting, including PARP, cleaved caspase-7/-8/-9, Bcl-2 and Bax. As shown in Fig. 5C, the above-mentioned biomarkers were markedly upregulated in the (pCDH)-HSP47-transfected cells compared with these levels noted in the (pCDH)-control cells, except for Bcl-2 protein, which was notably downregulated.