Long non-coding RNA HOXA11 antisense RNA upregulates spermatogenesis-associated serine-rich 2-like to enhance cisplatin resistance in laryngeal squamous cell carcinoma by suppressing microRNA-518a.

Long noncoding RNAs (LncRNAs) are closely associated with the chemoresistance of laryngeal squamous cell carcinoma (LSCC). Previous studies indicated that HOXA11-AS could function as a vital regulator in human cancers. However, the regulatory mechanisms of HOXA11-AS in the chemoresistance of LSCC remain unclear. In this study, it was found that HOXA11-AS expression was upregulated in cisplatin (CDDP)-resistant LSCC tissues and cells. Loss-of-function assays revealed that HOXA11-AS knockdown inhibited the viability, migration, and invasion, but promoted the apoptosis of CDDP-resistant LSCC cells. Meanwhile, we identified miR-518a as a downstream gene of HOXA11-AS in LSCC, and miR-518a silencing reversed the promotive effect of HOXA11-AS knockdown on CDDP sensitivity of LSCC cells. In addition, miR-518a could inhibit spermatogenesis-associated serine-rich 2-like (SPATS2L) expression by direct interaction, and upregulation of SPATS2L abrogated the inhibitory effect of HOXA11-AS silencing or miR-518a overexpression on CDDP resistance of CDDP-resistant LSCC cells. In sum, our results demonstrated that HOXA11-AS enhanced CDDP resistance of LSCC via miR-518a/SPATS2L axis, which might offer novel therapeutic strategies for CDDP-resistant LSCC.

Introduction

Laryngeal squamous cell carcinoma (LSCC), as the major pathological type of laryngeal cancer, is one of the most common malignant tumors of the head and neck [1,2]. LSCC arises from the squamous epithelium of the larynx and is associated with high metastasis rates [3]. Despite the improvement of therapeutic strategies for LSCC, the 5-year overall survival rate of LSCC remains less than 50% [4]. Currently, chemotherapy has been widely used for LSCC treatment in clinical practice, but its clinical efficacy is limited due to drug resistance [5]. Therefore, investigation of the underlying mechanism of chemoresistance of LSCC may develop novel treatment strategies to overcome the chemoresistance in LSCC.

Long noncoding RNAs (lncRNAs) are a type of RNA longer than 200 nt that lack the protein-coding ability [6,7]. Previous studies have indicated that lncRNAs play a vital role in the chemoresistance of various tumors, including LSCC. For example, FGD5-AS1 increased cisplatin (CDDP) resistance by sponging miR-497-5p to regulate SEPT2 expression in LSCC [8]. H19 sensitized LSCC to CDDP via miR-107/HMGB1 axis in vitro and in vivo [9]. FOXD2-AS1 enhanced chemotherapeutic resistance of LSCC via STAT3 activation [10]. Hence, it is important to verify the chemoresistance-related lncRNAs and explore their potential molecular mechanisms in the chemoresistance of LSCC.

Homeobox (HOX) A11 antisense lncRNA (HOXA11-AS), a highly conserved lncRNA [11], has been reported to be associated with the drug resistance of different cancers. For example, HOXA11-AS enhanced CDDP resistance of human lung adenocarcinoma cells via regulating miR-454-3p/Stat3 axis [12]. Silencing HOXA11-AS promoted cell apoptosis and enhanced the CDDP sensitivity of nasopharyngeal carcinoma cells through inhibiting the c-Met/AKT/mTOR pathway [13]. Moreover, Wang et al reported that HOXA11-AS increased cell viability and CDDP resistance in oral squamous cell carcinoma by downregulating miR-214-3p [14]. Nevertheless, whether HOXA11-AS is involved in the chemoresistance of LSCC remains unclear.

In this study, we hypothesized HOXA11-AS might be involved in the chemoresistance of LSCC, and the current study aimed to clarify its role in CDDP resistance and explore the molecular mechanism by which HOXA11-AS regulated CDDP resistance in LSCC. These findings might provide novel treatment strategies to overcome CDDP resistance in LSCC.

Materials and methods

Tissue collection

The LSCC patients with cancer progression or recurrence within 6 months after chemotherapy were defined as CDDP-resistant group and those with recurrence beyond 6 months or without recurrence were defined as the CDDP-sensitive group. A total of 60 tumor tissues (27 CDDP-resistant LSCC and 33 CDDP-sensitive LSCC tissues) were obtained at the affiliated Hospital of Hebei Engineering University. Written informed consent was obtained from each participant, and this study was approved by the Ethics Committee of Affiliated Hospital of Hebei Engineering University Hospital.

Cell culture

The human LSCC cell lines (TU686 and TU177) were purchased from Chinese Academy of Sciences (Shanghai) and cultured in DMEM with 10% FSB at 37°C and 5% CO2.

The CDDP-resistant LSCC cells (TU686/CDDP and TU177/CDDP) were established by a stepped exposure of TU686 and TU177 cells to increasing concentrations of CDDP for more than 6 months. The CDDP-resistant cells were cultured in DMEM with 1 μg/mL CDDP to maintain the drug resistance.

Cell transfection

Short hairpin (sh)RNAs targeting HOXA11-AS (shHOXA11-AS#1: 5’- CUACCAUCCCUGAGCCUUA-3’; shHOXA11-AS#2: 5’- UGACAUCCGAGGAGACUUC-3’) or control (shNC), miR-518a mimics or inhibitor with its control (NC mimics or inhibitor), and pcDNA3.1/HOXA11-AS were obtained from GenePharm (Shanghai). Transfection was conducted using Lipofectamine 2000 reagent (Invitrogen).

RT-qPCR

The RNA was extracted from tissues and cell lines using TRIzol reagent (Invitrogen). Total RNA was reverse-transcribed into cDNA using the Reverse Transcription kit (Takara). Then cDNA was synthesized by using Prime Script RT reagent kits (Takara). The PCR amplification was applied using SYBR Green Real-Time PCR kit (TaKaRa) on ABI 7500 Real-Time PCR System (Applied Biosystems).

CCK-8 assay

TU686/CDDP and TU177/CDDP cells were seeded into 96-well plates for various periods (0, 24, 48, and 72 h) or treated with different concentrations of CDDP (0, 2, 4, 6, 8, and 20 μg/mL) for 24 h. Afterward, cells were incubated with 10uL CCK-8 reagent (Beyotime,) for 4 h. The absorbance at 450 nm was measured on a microplate reader [15].

Transwell assay

The invasion ability was analyzed using transwell chambers (EMD Millipore) [16]. Transfected cells in serum-free DMEM were plated into the upper chamber coated with Matrigel (BD Biosciences). Then, 600 ul DMEM containing 10% FBS was added to the lower chamber. After 48 h, invasive cells in the lower chamber were fixed and dyed with crystal violet. The invaded cells were counted using a microscope. The migration ability was measured with the upper layer of the filter not coated with Matrigel according to the aforementioned method.

Luciferase reporter assay

The wild-type (wt) and mutant (mut) sequences of HOXA11-AS and SPATS2L were cloned into the pmirGLO vector (Promega Corporation). Then, these vectors were co-transfected with miR-518a mimics or NC mimics into CDDP-resistant LSCC cells. The luciferase activities were analyzed using a Dual-Luciferase Reporter Assay System (Promega Corporation) [15].

TUNEL assay

The apoptosis of CDDP-resistant LSCC cells was evaluated via TUNEL Apoptosis Assay kit [17]. After rising twice with PBS, cells were fixed by 4% paraformaldehyde followed by permeabilization with Triton-X 100. The cells were incubated in TUNEL assay solution (Invitrogen). Subsequently, TUNEL-stained cells were washed with PBS and counterstained with DAPI (Beyotime). Finally, TUNEL-positive cells were counted using a cell counter (BD Biosciences).

Western blot

Total protein was extracted using RIPA lysis buffer (Beyotime), separated using SDS-PAGE, and transferred to the PVDF membrane. The membranes were incubated with primary antibodies against SPATS2L (1:1,000, cat. no. ab161869; Abcam) or GAPDH (1:1,000, cat. no. ab9485; Abcam). The membranes were and then incubated with secondary antibodies. The protein bands were visualized using an ECL reagent (GE Healthcare) [18].

Statistical analysis

Statistical analysis was performed using SPSS 19.0 (IBM). Data are displayed as mean ± standard deviation (SD) of three independent experimental repeats. Comparisons between groups were analyzed using student’s t-test or one-way ANOVA. Pearson’s correlation analysis was used to determine the correlation between miR-518a and HOXA11-AS or SPATS2L expression levels. P < 0.05 indicated statistically significant.

Results

The current study explored the biological role and molecular mechanism of HOXA11-AS in CDDP resistance of LSCC. Our data indicated that HOXA11-AS decreased CDDP sensitivity of CDDP-resistant LSCC through the miR-518a/SPATS2L axis. These findings might offer a novel strategy for overcoming CDDP resistance in LSCC.

HOXA11-AS was upregulated in CDDP-resistant LSCC

Firstly, RT-qPCR indicated that HOXA11-AS level was highly expressed in CDDP-resistant LSCC tissues (Figure 1a). To detect HOXA11-AS expression in CDDP-resistant LSCC in vitro, CDDP-resistant LSCC cells were successfully constructed and the IC50 value of CDDP in TU686/CDDP and TU177/CDDP cells were increased (Figure 1b). RT-qPCR revealed that HOXA11-AS expression was upregulated in TU686/CDDP and TU177/CDDP cells (Figure 1c). Moreover, patients with high HOXA11-AS levels exhibited a shorter overall survival time than those with low HOXA11-AS expression (Figure 1d). The above results indicated that HOXA11-AS was upregulated in CDDP-resistant LSCC tissues and cells.

Figure 1.

HOXA11-AS was upregulated in CDDP-resistant LSCC. (a) RT-qPCR showed the relative expression of HOXA11-AS in CDDP-resistant LSCC tissues (n = 27) compared with CDDP-sensitive LSCC tissues (n = 33). (b) Successfully constructed TU686/CDDP and TU177/CDDP cells with increased IC50 values. (c) RT-qPCR revealed HOXA11-AS expression was upregulated in TU686/CDDP and TU177/CDDP cells. (d) Kaplan-Meier curve for overall survival according to high HOXA11-AS expression level compared with low HOXA11-AS expression. *p < 0.05 and **p < 0.01.

Knockdown of HOXA11-AS increased CDDP sensitivity of CDDP-resistant LSCC cells

To further analyze the role of HOXA11-AS in CDDP resistance of LSCC, TU177/CDDP and TU686/CDDP cells were transfected with shHOXA11-AS#1 and shHOXA11-AS#2, and the knockdown efficiencies were confirmed (Figure 2a). CCK-8 assay revealed that cell viability was markedly suppressed by silencing of HOXA11-AS in TU177/CDDP and TU686/CDDP cells (Figure 2b). Transwell assays uncovered that the interference of HOXA11-AS suppressed cell migration and invasion (Figure 2c and d). TUNEL assay revealed cell apoptosis rate was increased in HOXA11-AS knockdown group (Figure 2e). These data suggested that HOXA11-AS silencing suppressed CDDP resistance of LSCC.

Figure 2.

Knockdown of HOXA11-AS increased CDDP sensitivity of CDDP-resistant LSCC cells. (a) Reduced HOXA11-AS expression after transfection of shHOXA11-AS#1 and shHOXA11-AS#2 in TU686/CDDP and TU177/CDDP cells. (b and c) CCK-8 assay evaluated the cell proliferation of TU686/CDDP and TU177/CDDP cells transfected with shNC, shHOXA11-AS#1, and shHOXA11-AS#2. (c and d) The migration and invasion abilities of shNC, shHOXA11-AS#1 or shHOXA11-AS#1-transfected TU686/CDDP and TU177/CDDP cells were assessed via transwell assays. (e) The increased apoptosis rates of TU686/CDDP and TU177/CDDP cells transfected with shNC, shHOXA11-AS#1, and shHOXA11-AS#2 were detected through TUNEL assay. ** p  < 0.01 and *** p  < 0.001.

MiR-518a was a downstream gene of HOXA11-AS

Through Starbase website, binding sites of miR-518a on HOXA11-AS were predicted (Figure 3a). miR-518a mimics decreased the luciferase activity of HOXA11-AS-wt in CDDP-resistant LSCC cells, while no impacts were observed in HOXA11-AS-mut group (Figure 3b). Moreover, miR-518a was lowly expressed in CDDP-resistant tissues and cells (Figure 3c and d). In addition, miR-518a was upregulated in CDDP-resistant cells transfected with shHOXA11-AS, and downregulated in pcDNA3.1/HOXA11-AS group (Figure 3e). Furthermore, miR-518a expression was revealed to be inversely correlated with HOXA11-AS expression in CDDP-resistant LSCC tissues (Figure 3f). These findings revealed that HOXA11-AS directly bound with miR-518a.

Figure 3.

MiR-518a was a downstream gene of HOXA11-AS. (a) StarBase prediction of HOXA11-AS as a binding target of miR-518a. (b) Luciferase reporter assay showed luciferase activity in TU686/CDDP and TU177/CDDP cells transfected with NC mimics and miR-518a mimics. (c and d) RT-qPCR showed the relative miR-518a expression in CDDP-resistant tissues and cells. (e) RT-qPCR showed the relative miR-518a expression in TU686/CDDP and TU177/CDDP cells transfected with shNC, shHOXA11-AS#1, pcDNA3.1, or pcDNA3.1/ HOXA11-AS. (f) Pearson’s correlation analysis was used to assess the correlation between HOXA11-AS and miR-518a expression. *p < 0.05 and **p < 0.01.

Knockdown of HOXA11-AS decreased CDDP resistance of CDDP-resistant LSCC cells through miR-518a

To investigate whether HOXA11-AS regulated the CDDP resistance of LSCC through miR-518a, TU177/CDDP and TU686/CDDP cells were transfected with shNC, shHOXA11-AS#1, and shHOXA11-AS#1+ miR-518a inhibitor. RT-qPCR indicated that miR-518-3p inhibition partially restored the promoting effect of HOXA11-AS silencing on miR-518a expression (Figure 4a). CCK-8 and Transwell assays showed that HOXA11-AS silencing suppressed the proliferation and metastasis of CDDP-resistant LSCC cells, which was reversed by inhibition of miR-518a (Figure 4b-d). Moreover, TUNEL assay revealed that cell apoptosis was increased by HOXA11-AS knockdown, while such effect was restored by the downregulation of miR-518-3p (Figure 4e). The above data suggested that HOXA11-AS reduced CDDP sensitivity of CDDP-resistant LSCC cells by regulating miR-518a.

Figure 4.

Knockdown of HOXA11-AS decreased CDDP resistance of CDDP-resistant LSCC cells through miR-518a. (a) RT-qPCR indicated that relative miR-518a expression in TU686/CDDP and TU177/CDDP cells transfected with shNC, shHOXA11-AS#1, and shHOXA11-AS#1+ miR-518a inhibitor. (b-e) CCK-8 and Transwell assays showed the cell viability metastasis, and apoptosis of TU686/CDDP and TU177/CDDP cells transfected with shNC, shHOXA11-AS#1, and shHOXA11-AS#1+ miR-518a inhibitor. *p < 0.05 and **p < 0.01.

SPATS2L was directly targeted by miR-518a

Bioinformatics analysis was performed using PITA, miRmap and microT databases, and 10 potential genes (TBC1D10B, TSN, ZNF282, C1orf43, ZBTB7A, SPATS2L, PTPRU, TIRAP, TFE3, TBC1D10B) were predicted as downstream genes of miR-518a (Figure 5a). RT-qPCR revealed that only SPATS2L was significantly downregulated by miR-518a overexpression, indicating that miR-518a could negatively regulate SPATS2L expression (Figure 5b). The potential binding sequences between SPATS2L and miR-518a were presented in Figure 5c. Furthermore, luciferase activity of SPATS2L-wt group was decreased by miR-518a overexpression, while no difference was found in SPATS2L-mut groups (Figure 5d). In addition, the SPATS2L expression was remarkably upregulated in CDDP-resistant LSCC tissues and cells (Figure 5e-g). Pearson’s correlation analysis showed that miR-518a was negatively correlated with SPATS2L expression in CDDP-resistant LSCC tissues (Figure 5h). The above data demonstrated that miR-518a negatively regulated SPATS2L expression.

Figure 5.

SPATS2L was directly targeted by miR-518a. (a) StarBase website predicted 10 potential targets of miR-518a. (b) RT-qPCR revealed that relative SPATS2L expression in CDDP-resistant LSCC cells transfected with NC mimics or miR-518a mimics. (c) The binding sequences between miR-518a and SPATS2L were presented. (d) Luciferase reporter assay revealed the luciferase activities of SPATS2L-wt or SPATS2L-mut reporters in TU686/CDDP and TU177/CDDP cells transfected with NC mimics or miR-518a mimics. (e) RT-qPCR assay showed the expression of SPATS2L in in CDDP-resistant LSCC tissues (n = 27) compared with CDDP-sensitive LSCC tissues (n = 33). (f and g) RT-qPCR and Western blot assays showed the mRNA and protein level of SPATS2L in TU686/CDDP and TU177/CDDP cells. (h) Pearson’s correlation analysis showed the correlation between miR-518a and SPATS2L expression in CDDP-resistant LSCC tissues. *p < 0.05 and **p < 0.01.

HOXA11-AS/miR-518a axis regulated CDDP resistance of LSCC via SPATS2L

To further investigate whether HOXA11-AS or miR-518a regulated chemoresistance of LSCC via SPATS2L, rescue assays were conducted. RT-qPCR revealed that SPATS2L overexpression restored the inhibitory effects of HOXA11-AS knockdown or miR-518-3p overexpression on SPATS2L expression (Figure 6a and b). As shown in Figure 6c-f, HOXA11-AS downregulation or miR-518a upregulation suppressed the cell proliferation, migration, and invasion, and facilitated cell apoptosis of CDDP-resistant LSCC cells, while these effects were counteracted by the upregulation of SPATS2L. In addition, silencing of HOXA11-AS reduced SPATS2L expression, while this effect was restored by the inhibition of miR-518a (Figure 6g). These data revealed that SPATS2L was a vital regulator of HOXA11-AS or miR-518a in CDDP-resistant LSCC cells.

Figure 6.

HOXA11-AS/miR-518a axis regulated CDDP resistance of LSCC via SPATS2L. (a and b) RT-qPCR revealed that relative SPATS2L expression in TU686/CDDP cells transfected with shNC, shHOXA11-AS#1, shHOXA11-AS#1+ pcDNA3.1/SPATS2L and TU177/CDDP transfected with NC mimics, miR-518a mimics, miR-518a mimics+pcDNA3.1/SPATS2L. (c-f) CCK-8, Transwell and TUNEL assays showed the viability, migration, and invasion abilities and apoptosis rate in different groups. (g) RT-qPCR showed relative SPATS2L expression levels in TU686/CDDP and TU177/CDDP cell lines transfected with shNC, shHOXA11-AS#1, and shHOXA11-AS#1+ miR-518a inhibitor. *p < 0.05, **p < 0.01.

Discussion

Epidemiologic data indicated that epigenetic and genetic alterations, and environmental and lifestyle-related factors, such as tobacco, alcohol, and viral infection contributed to the pathogenesis of LSCC [19]. Chemoresistance is one of the major obstacles in the chemotherapeutic treatment of LSCC [20,21], and accumulating studies have demonstrated that dysregulation of lncRNAs is implicated with chemoresistance in cancers [22,23]. For instance, SNHG6 promoted chemoresistance of colorectal cancer through ULK1-induced autophagy by regulating miR-26a [24]. Upregulation of SNHG16 inhibited 5-fluorouracil resistance of hepatocellular carcinoma by downregulating miR-93 [25]. KCNQ1OT1 enhanced the chemoresistance of colon cancer by regulating the miR-34a/ATG4B axis [26]. In our study, HOXA11-AS was upregulated in CDDP-resistance LSCC and knockdown of HOXA11-AS inhibited CDDP resistance of CDDP-resistant LSCC, as indicated by the attenuated cell viability, migration, and invasion, and enhanced cell apoptosis.

LncRNA could act as a competing endogenous RNA (ceRNA) by competitively binding targeted miRNAs, which further participated in the regulation of carcinogenesis of human tumors, including LSCC. For example, MNX1-AS1 promoted LSCC progression and served as a ceRNA to target FoxM1 by sponging miR-370 [27]. TRPM2-AS facilitated cell metastasis by decreasing miR-138 and increasing SOX4 expression in LSCC [28]. HOXA11-AS-mediated ceRNA network has also been reported to be involved in tumorigenesis or chemoresistance. Liu et al reported that HOXA11-AS contributed to cell proliferation and EMT in hepatocellular carcinoma by regulating the miR-506-3p/Slug axis [29]. HOXA11-AS increased CDDP resistance by modulating the miR-98/PBX3 axis in nasopharyngeal carcinoma [30]. In this study, miR-518a was verified as a target of HOXA11-AS. The biological role and function of miR-518a have been identified in several cancers. For example, miR-518a overcame CDDP resistance in colon cancer by targeting MDMA [31]. Moreover, Li et al revealed that HOXA11-AS enhanced cell viability and invasion via miR-518a in oral squamous cell carcinoma [32]. Herein, we found that miR-518a inhibition reversed the effect of HOXA11-AS knockdown on cell proliferation, migration, invasion, and apoptosis of CDDP-resistant LSCC.

Through a series of experiments, spermatogenesis-associated serine-rich 2-like (SPATS2L) was confirmed as a downstream gene of miR-518a. SPATS2L is ubiquitously expressed in multiple tissues and its overexpression is related to the carcinogenesis of glioma and hepatocellular carcinoma [33,34]. Our results demonstrated that SPATS2L was upregulated in CDDP-resistant LSCC, and miR-518a inhibited SPATS2L expression by direct interaction. Moreover, the inhibitory effect of HOXA11-AS knockdown and miR-518a overexpression on cell proliferation and metastasis were abolished by SPATS2L upregulation, suggesting that HOXA11-AS1 and miR-518-3p regulated the resistance of LSCC to CDDP via SPATS2L.

Conclusion

Our data demonstrated that HOXA11-AS promoted CDDP resistance of LSCC by acting as a ceRNA of miR-518a to regulate SPATS2L, shedding light on the development of promising therapeutic strategy to overcome CDDP resistance in LSCC patients. In the future, other potential downstream genes, may also be important in HOXA11-AS-regulated phenotypes of LSCC, which need to be investigated.