This article is mentioned in:

Introduction

Osteosarcoma (OS), the most frequent primary bone malignancy affects adolescents and young adults, is characterized by occurrence at the extremities of long bones and originating from primitive osteogenic mesenchymal cells (1). OS is most common in pediatric bone malignancies, accounting for ~5% of pediatric tumors (2). Currently, chemotherapeutic treatments combined with surgical resection is one of the predominant treatments for patients with OS (3,4). Despite considerable progress in the diagnosis and treatment of OS, the overall clinical efficacy of OS treatment remains unsatisfactory (5,6). However, pulmonary metastasis is the predominant site of osteosarcoma recurrence and the most general cause of mortality (7). Thus, it is necessary to look for new strategies for treatment of OS.

MicroRNAs (miRNAs) are a kind of regulatory RNA, which are small endogenous non-coding RNAs consisting 18–25 nucleotides, and negatively regulate translation of the specific target gene via directly binding to 3′UTR region of target mRNAs (8). Increasing evidence has demonstrated that miRNAs display important roles in modulating the progression and development of various tumors (9–11). Accumulating evidence confirms that dysregulated miRNAs contribute to multiple physiological and pathological processes in different malignancies (including osteosarcoma), such as apoptosis, cell proliferation, and autophagy (12–14). Extensive research has shown that microRNA-429 (miR-429) deregulates and plays major roles in many tumors including melanoma, esophageal squamous cell carcinoma, renal cell carcinoma, and OS (15–18). Therefore, determining the exact role of miR-93 in OS carcinogenesis might contribute to improving the diagnosis and prognosis of patients with this tumor.

HOXA9, is a homeobox (HOX) gene, reported to regulate several malignant diseases. For instance, increasing expression of HOXA9 inhibited lung cancer cell invasion and migration (19). Makabe et al (20) found that HOXA9 was involved in endometrial carcinogenesis. Furthermore, HOXA9 was reported to promote leukemogenesis (21). In addition, HOXA9 was reported as the direct target of some miRNAs in regulation of tumor development and progression. A previous study reported that HOXA9 acted as the target of miR-873 in regulating the progression of OS (22). It was also demonstrated as the target of miR-182 in regulating gastric cancer cell proliferation and migration (23). However, whether HOXA9 is the target of miR-429 in modulating OS cell viability, invasion and migration has not been reported.

Wnt/β-catenin pathway is well known to participate in tumorigenesis. Previous studies have stated that HOXA9 could regulate Wnt/β-catenin pathway in glioma cell growth (24) and regulate OS cell proliferation and apoptosis (25). In this study, we aimed to test miR-429 functional role in OS progression and to confirm whether miR-429 modulate OS progression by targeting HOXA9 through Wnt/β-catenin pathway.

Materials and methods

Samples

Fifty-six paired OS tissues and adjacent normal tissues were collected from the patients who underwent surgery in Weifang People's Hospitall (Weifang, China) between April 2011 and October 2017. The mean age of all patients was 9.4 years, and the age range 5–17 years. None of the patients received any treatment prior to surgery. All the tissues were collected in the same condition and were histopathologically verified carcinoma. The fresh OS tissues were confirmed by pathologists and then stored at −80°C for further experimental use. This study was conducted in accordance with the Declaration of Helsinki. The Ethics Committee of Weifang People's Hospital approved tissue sample collection and use protocols. Written informed consent was obtained from each patient and their parents before collecting the specimens. Tables I and II show the demographic features and clinicopathological data.

Table I. Association between miR-429 expression and clinicopathological characteristics of patients with OS.

Table I.

Association between miR-429 expression and clinicopathological characteristics of patients with OS.

		miR-429
Characteristics	Cases	Low	High	P-value
Age, years				0.798
≥18	20	14	6
<18	36	24	12
Sex				0.176
Female	30	18	12
Male	26	20	6
Tumor size, cm				0.022a
≤5	31	25	6
>5	25	13	12
TNM stage				0.012a
I	26	22	4
II/III	30	16	14
Distant metastasis				0.013a
No	29	24	5
Yes	27	14	13
Location				0.022a
Femur/Tibia	28	23	5
Elsewhere	28	15	13

{ label (or @symbol) needed for fn[@id='tfn1-ol-0-0-11766'] } Statistical analyses were performed by the χ² test.

a P<0.05 was considered significant. OS, osteosarcoma.

Table II. Association between HOXA9 expression and clinicopathological characteristics of patients with OS.

Table II.

Association between HOXA9 expression and clinicopathological characteristics of patients with OS.

		HOXA9
Characteristics	Cases	High	Low	P-value
Age, years				0.472
≥18	30	18	12
<18	26	18	8
Sex				0.264
Female	28	20	8
Male	28	16	12
Tumor size, cm				0.012a
≤5	32	25	7
>5	24	11	13
TNM stage				0.018a
I	34	26	8
II/III	22	10	12
Distant metastasis				0.003a
No	26	22	4
Yes	30	14	16
Location				0.439
Femur/Tibia	29	20	9
Elsewhere	27	16	11

{ label (or @symbol) needed for fn[@id='tfn3-ol-0-0-11766'] } Statistical analyses were performed by the χ² test.

a P<0.05 was considered significant. OS, osteosarcoma; HOXA9, homeobox A9.

Cell culture

The OS cell lines (MG-63, U2OS, Saos-2) and normal human fetal osteoblast (hFOB 1.19) cells were purchased from Tianjin Sai'er Biotechnology Co., Ltd. The cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Gibco; Thermo Fisher Scientific, Inc.) and DMEM/F-12 medium, respectively. The media contained 10% fetal bovine serum (FBS; HyClone; GE Healthcare Life Sciences), 100 IU/ml penicillin and 100 mg/ml streptomycin (both from Baomanbio). The cells were maintained at 37°C with 5% CO2.

Cell transfection

Overexpression of miR-429 in MG-63 cells and knockdown of miR-429 in Saos-2 cells due to its expression was lower in Saos-2 cells than in MG-63 cells. miR-429 mimic, miR-429 inhibitor, control mimic and control inhibitor were used for increasing or decreasing miR-429, and was obtained from Shanghai GenePharma Co., Ltd. HOXA9 siRNA was synthesized by Shanghai GeneChem Co., Ltd. When the cells reached growth phase, they were collected for transfection. Transfections were finished by using the Lipofectamine™ 2000 reagent (Invitrogen; Thermo Fisher Scientific, Inc.). miR-429 mimic, forward, 5′-UAAUACUGUCUGGUAAAACCGU-3′ and reverse, 5′-GGUUUUACCAGACAGUAUUAUU-3′; NC mimic, forward, 5′-UUCUCCGAACGUGUCACGUUT-3′ and reverse, 5′-ACGUGACACGUUCGGAGAATT-3′; miR-429 inhibitor, 5′-GCTGATTTAAAGGCTTAG-3′; NC inhibitor, 5′-CAAATGTAGGTAGAGGA-3′; HOXA9 siRNA, 5′-ACGGCAUUACCAGACAGUAUUA-3′; NC siRNA, 5′-AGCGUGUAGCUAGCAGAGG-3′. The efficiency of knockdown and upregulation was analyzed using qRT-PCR.

Reverse transcription-quantitative polymerase chain reaction (RT-qPCR)

TRIzol Reagent (Invitrogen; Thermo Fisher Scientific, Inc.) was applied for extracting total RNA from OS clinical tissues and cell lines. TaqMan MicroRNA Reverse Transcription kit (Takara) was applied for performing reverse transfection. TaqMan miRNA qRT-PCR Kit (Takara) was used for performing quantitative RT-PCR. U6 and GAPDH were applied for normalizing miR-429 level and HOXA9, respectively. The thermocycling conditions were as follows: Initial denaturation at 95°C for 3 min and 35 cycles of denaturation at 95°C for 15 sec and annealing/elongation at 60°C for 30 sec. A melting curve analysis was performed to detect products. The relative expression was analyzed using the 2−ΔΔCq method (26). The primer sequences are shown in Table III.

Table III. Primer sequences for real-time fluorescence quantification PCR.

Table III.

Primer sequences for real-time fluorescence quantification PCR.

Gene	Primer sequences
GAPDH	F: 5′-GCACCGTCAAGGCTGAGAAC-3′
	R: 5′-ATGGTGGTGAAGACGCCAGT-3′
U6	F: 5′-CTCGCTTCGGCAGCACA-3′
	R: 5′-AACGCTTCACGAATTTGCGT-3′
miR-429	F: 5′-GTCGTATCCAGTGCAGGGTCCGAGGTA-3′
	R: 5′-AGTGCAGGGTCCGAGGTATT-3′
HOXA9	F: 5′- CAACAAAGACCGAGCAAA-3′
	R: 5′- CAACAAAGACCGAGCAAA-3′

[i] F, forward; R, reverse; HOXA9, homeobox A9.

MTT assay

3-(4, 5-Dimethylthiazolyl-2)- 2,5-diphenyltetrazolium bromide (MTT) assay was carried out for assessing cell viability. The cells treated with miR-429 mimic, miR-429 inhibitor, control mimic and control inhibitor were seeded in 96-well plates at a density of 2,000 cells/well with three duplicate wells per group and cultured for 1, 2, 3, 4 days. MTT solution (20 µl) was added to each well and incubated for another 4 h at 37°C. Then, the medium was removed and dimethyl sulfoxide (DMSO) was added to dissolve formazan crystals by swirling gently. Finally, the absorbance was measured at 490 nm using a spectrophotometric plate reader.

Transwell assays

Transwell assays were applied for detection of cell migration and invasion. The steps performed for the migration and invasion were similar, except for the top chamber coating with or without Matrigel. The Transwell chambers (8 µm pore filter) were coated with or without 100 ml Matrigel (5 mg/ml) (BD Biosciences). In brief, cells were added into the top chamber and DMEM medium containing 20% FBS was added to the lower chamber. The upper and lower chamber was separated by polycarbonate film. After incubation for 24 h at 37°C, the cells were invaded or migrated to the lower chamber. The non-migrated and non-invaded cells on the upper surface of the membrane were carefully scraped off using a cotton swab. Migrated and invaded cells were fixed with methanol and stained with 0.5% crystal violet for 10 min at room temperature. Then, the cells in the lower chamber were fixed, stained and finally counted with a microscope (Olympus Corporation).

Western blot analysis

Radio immunoprecipitation assay (RIPA) lysis buffer (Sigma-Aldrich; Merck KGaA) was applied for lysing OS cells. The protein concentration was determined using the BCA assay, the proteins (10 µg) were separated by 15% SDS-PAGE, followed by transferred to NC membranes. Then, the membranes were blocked for 1 h at 37°C with 5% skimmed milk overnight. Subsequently incubated with primary antibodies (HOXA9, ab140631, 1:1,000; β-catenin, ab32572, 1:1,000; c-myc, ab32072, 1:1,000; p-c-Jun, ab32385, 1:1,000; GAPDH, ab181602, 1:1,000; all from Abcam) at room temperature for 3 h, and then incubated with sheep anti-mouse or donkey anti-rabbit horseradish peroxidase-conjugated (HRP) antibody (GENXA931-1ML and GENA934-1ML, Sigma-Aldric; Merck KGaA). The enhanced chemiluminescence reagent (ECL; Pierce Biotechnology, Inc.) was used to conduct autoradiography.

Dual-luciferase reporter assay

The putative targets of miR-130a-3p were predicted using the TargetScan (http://www.targetscan.org/vert_72/). The wild-type (WT) or mutant type (MuT) of HOXA9 3′UTR fragments were sub-cloned into the pMIR-REPORT vector. Then, the MG-63 and Saos-2 cells were co-transfected with miR-429 mimic, miR-429 inhibitor, control mimic and control inhibitor along with the 3′UTR of HOXA9 using Lipofectamine 2000 reagent. After transfection for 48 h, the luciferase activity was measured by a luciferase reporter assay system (Promega Corporation), and Renilla luciferase activity was used to normalize the data.

Statistical analysis

All experimental conditions were repeated in duplicate. Results are presented as mean ± SD. Statistic analysis was performed using SPSS v.19.0 software (SPSS, Inc.). Unpaired Student's t-test or Tukey's post hoc test of one-way analysis of variance (one-way ANOVA) was applied for comparing the differences between two groups or more than two groups. The clinical association was analyzed with Chi-square test. P<0.05, was considered statistically significant.

Results

miR-429 expression is decreased and HOXA9 expression is increased in OS

According to previous studies, miR-429 and HOXA9 were dysregulated in multiple tumors. Here, the aim was to test miR-429 and HOXA9 expression in OS tissues and cell lines. As illustrated in Fig. 1A, the expression of miR-429 was downregulated in OS tissues compared with that in normal tissues. Subsequently, the expression of miR-429 expression was decreased in OS cell lines compared with that in normal human fetal osteoblasts (hFOB 1.19) as shown in Fig. 1B. Moreover, miR-429 expression was lowest in MG-63 cells compared with that in U2OS and Saos-2 cells (Fig. 1B). Furthermore, the findings also showed that HOXA9 was increased in OS tissues (Fig. 1C) and cell lines (Fig. 1D). Due to the opposite expression of miR-429 and HOXA9 in OS, we detected the correlation between miR-429 and HOXA9. Results showed that their relationship was negative (Fig. 1E). To confirm their clinical importance, we divided the subgroups (high/low) according to the median value as a cutoff of miR-429 and HOXA9. In addition, the expression of miR-429 (Table I) and HOXA9 (Table II) were notably associated with tumor size, TNM stage and distant metastasis. Collectively, the above results indicated that miR-429 might serve as a tumor suppressor while HOXA9 serves as an oncogene in OS.

Figure 1. Downregulation of miR-429 and upregulation of HOXA9 in OS. (A) Measurement of miR-429 expression in OS tissues (n=56). (B) Measurement of miR-429 expression in OS cells. (C) HOXA9 expression measured in OS tissues (n=56). (D) HOXA9 expression measured in OS cell lines. (E) Relationship between miR-429 and HOXA9 (r=−8583; P<0.0001). *P<0.05, **P<0.01. HOXA9, homeobox A9; OS, osteosarcoma.

miR-429 suppresses OS cell viability, invasion and migration

As stated, the expression of miR-429 was highest in Saos-2 cells, the expression of miR-429 was lowest in MG-63 cells. Thus, miR-429 inhibitor was transfected in Saos-2 cells and miR-429 mimic in MG-63 cells. Results of qRT-PCR revealed that miR-429 expression was increased by miR-429 mimic in MG-63 cells and reduced by miR-429 inhibitor in Saos-2 cells (Fig. 2A). MTT assay was applied for testing MG-63 cell viability after treated with miR-429 mimic or Saos-2 cells after treated with miR-429 inhibitor. The results of MTT assay showed that miR-429 overexpression suppressed MG-63 cell proliferation, while miR-429 knockdown remarkably promoted Saos-2 cell proliferation (Fig. 2B). Furthermore, Transwell analysis showed that miR-429 overexpression reduced MG-63 cell migration and miR-429 knockdown showed the opposite effect on Saos-2 cells (Fig. 2C). For invasion, upregulating of miR-429 expression inhibited MG-63 cell invasion, while downregulating of miR-429 expression enhanced Saos-2 cell invasion (Fig. 2D). These findings demonstrated that miR-429 inhibited OS cell proliferation, invasion and migration.

Figure 2. Impeding effect of miR-429 on OS progression. (A) miR-429 expression tested in MG-63 cells after increasing miR-429, and miR-429 expression tested in Saos-2 cells after repressing miR-429. (B) MG-63 cell viability detected after increasing miR-429 expression, Saos-2 cell viability detected after decreasing miR-429 expression. (C) MG-63 cell migration measured after increasing miR-429 expression, Saos-2 cell migration tested after decreasing miR-429 expression. (D) MG-63 cell invasion after treated with miR-429 mimic, Saos-2 cell invasion after treated with miR-429 inhibitor. *P<0.05, **P<0.01, ***P<0.001.

HOXA9 is the target of miR-429

As we found that HOXA9 mRNA in tissues was inversely correlated with miR-429 expression, we further investigated whether HOXA9 was the target of miR-429. By searching both TargetScan (http://www.targetscan.org) and Sanger (http://microrna.sanger.ac.uk) databases, it was found that the 3′UTR of HOXA9 contains conserved miR-429-binding sites (Fig. 3A). Then, luciferase reporter assay was applied for confirmation. The findings revealed that overexpression of miR-429 significantly suppressed firefly luciferase reporter activity of the WT HOXA9 3′UTR, while miR-429 knockdown increased its luciferase activity, however, they did not affect the luciferase activity of MuT-HOXA9 3′UTR (Fig. 3B). Next, qRT-PCR and Western blot analysis were carried out for detecting the effect of miR-429 on HOXA9 expression in mRNA and protein level, respectively. The findings demonstrated that miR-429 reversely regulated the mRNA and protein expression of HOXA9 in OS cells (Fig. 3C and D). Taken together, the above results confirmed the inverse relationship between miR-429 and HOXA9.

Figure 3. HOXA9 is the direct target of miR-429. (A) The binding site of miR-429 with HOXA9 3′UTR. (B) Detection of HOXA9 3′UTR luciferase activity in MG-63 and Saos-2 cells after increasing or decreasing miR-429. (C) Detection of HOXA9 protein level in MG-63 and Saos-2 cells after increasing or decreasing miR-429. (D) HOXA9 mRNA expression in MG-63 and Saos-2 cells after increasing or decreasing miR-429. **P<0.01. HOXA9, homeobox A9.

Knockdown of HOXA9 reverses the effects of miR-429 on OS cells

The effect of HOXA9 on OS cell proliferation, invasion and migration were measured. As HOXA9 was overexpressed in OS tissues and cell lines compared with normal tissues and human fetal osteoblasts (hFOB 1.19), HOXA9 siRNA was tranfected into OS cells to inhibit HOXA9 expression. As expected (Fig. 4A), HOXA9 expression was decreased after transfection with HOXA9 siRNA. The results of MTT assay found that knockdown of HOXA9 expression inhibited OS cell viability, while decreasing miR-429 expression promoted cell expression. Furthermore, HOXA9 siRNA reversed the promotion effect of miR-429 inhibitor on OS cell viability (Fig. 4B). In addition, similar effects were observed on the migration and invasion capacity of OS cells (Fig. 4C and D). These findings demonstrated that HOXA9 is involved in miR-429-mediated regulation of OS cells.

Figure 4. Inhibitory effect of HOXA9 siRNA on OS progression. (A) Measurement of HOXA9 mRNA expression in Saos-2 cells after inhibition of HOXA9. (B) Measurement of Saos-2 cell viability after treated with miR-429 inhibitor, HOXA9 siRNA, or miR-429 along with HOXA9 siRNA. (C) Measurement of Saos-2 cell migration after treated with miR-429 inhibitor, HOXA9 siRNA, or miR-429 along with HOXA9 siRNA. (D) Measurement of Saos-2 cell invasion after treated with miR-429 inhibitor, HOXA9 siRNA, or miR-429 along with HOXA9 siRNA. *P<0.05, **P<0.01. HOXA9, homeobox A9.

miR-429 regulates Wnt/β-catenin pathway in vitro

To further examine the mechanism of miR-429 in modulating OS progression, the expression of the downstream genes of Wnt/β-catenin pathway was tested by Western blot analysis. As shown in Fig. 5A and B, miR-429 overexpression suppressed the expression level of β-catenin, c-myc and p-c-Jun, as well as HOXA9 in MG-63 cells, while knockdown of miR-429 exhibited the opposite effect on these levels. Taken together, the findings indicated that miR-429 inhibited OS progression by targeting HOXA9 and suppressing Wnt/β-catenin pathway.

Figure 5. miR-429 regulates activation of Wnt/β-catenin in vitro. (A) Detection of the expressional level of HOXA9, β-catenin, c-myc and p-c-Jun in MG-63 cells after miR-429 overexpression and in Saos-2 cells after miR-429 silencing. (B) Detection of relative β-catenin protein level in MG-63 cells after re-expression of miR-429 and in Saos-2 cells after silencing miR-429. (C) Detection of relative p-c-Jun level in MG-63 cells after increasing miR-429 and in Saos-2 cells after decreasing miR-429. *P<0.05, **P<0.01. HOXA9, homeobox A9.

Discussion

OS is the most frequent primary bone malignancy, it affects adolescents and young adults, and is characterized by high incidence, great potential for metastasis and rapid progression (27,28). Although years of continuous progress have been obtained in OS therapy, the metastasis rates and mortality of OS are still very high, and the clinical effect of OS treatment remains unsatisfactory (29). Therefore, it is urgent to identify targets for the development of therapeutics for OS. The molecular mechanisms of OS carcinogenesis, especially regarding alterations of miRNAs in OS have attracted much attention in recent decades.

Increasing evidence has indicated that miRNAs play vital roles as diagnostics biomarkers and therapeutic targets in human malignancies, and expression of miRNAs are dysregulated, including OS (30,31). miR-429, in combination with the other four members (miR200a, miR-200b, miR-200c and miR-141), forms the miR-200 family (32). miR-429 mostly took part in the development of many tumors as a tumor suppressor. For example, miR-429 expression was downregulated in human thyroid cancer and it inhibited cell growth and promoted cell apoptosis (33). Moreover, miR-429 served as a suppressor in cervical cancer (34) and in pancreatic cancer (35). However, studies concerning miR-429 expression and its functional effect in OS are still rare and it is therefore necessary to investigate the underlying mechanisms of miR-429 in OS. Herein, it was shown that miR-429 was decreased in OS tissues and cells, which was in line with the studies reported by Liu et al and Deng et al (18,36). Based on previous studies, we further found that miR-429 suppressed OS cell proliferation, invasion and migration. The low expression of miR-429 was obviously associated with adverse clinical pathological issues of OS patients, including OS tumor size, TNM stage and clinical stage.

Combined with the bioinformatic analysis tool and dual-luciferase activity reporter assay, HOXA9 was validated a putative target for miR-429. HOXA9 is a member of the HOX gene family reported to have crucial roles in regulating hematopoiesis (37). Previous studies have shown that HOXA9 was the target of miRNAs in acute myeloid leukemia (38), in cutaneous squamous cell carcinoma (39) and in ovarian cancer (40). Therefore, we investigated whether HOXA9 was a mediator of miR-429 tumor inhibition in OS cells. In this study, it is first displayed that HOXA9 is a direct target of miR-429 in OS progression. Moreover, it was found that HOXA9 acted as an oncogene in OS development. These findings are consistent with previous research that HOXA9 level is increased markedly in ovarian cancer patients (41) and it was upregulated in colon cancer (42). The higher expression of HOXA9 was associated with OS tumor size, TNM stage and clinical stage. Furthermore, we showed that inhibiting HOXA9 suppressed OS cell viability, invasion and migration and it could overturn miR-429 inhibitor effect on OS progression. The above findings suggested miR-429 may act as a tumor suppressor by increasing HOXA9 expression in OS progression. HOXA9 overexpression reverses the inhibition effects of miR-429 on OS cells.

Increasing evidence shows that Wnt/β-catenin axis was regulated, for example, by miR-25, −376c, −21, −34a miRNAs in OS progression (43–45). Wnt/β-catenin signaling pathway modulates various genes that in turn regulate a diversity of cell functions such as proliferation, differentiation and morphogenesis, and β-catenin was proved to enhance progression, invasion and tumorigenesis in cancers (46). Increasing evidence has suggested that the signaling pathway was activated in OS cells and the aberrant regulation of the signaling pathway results in tumorigenesis of OS cells (47–49). To confirm the involvement of Wnt/β-catenin signaling pathway in regulating migration of OS cells, Western blotting was performed to analyze the nuclear fractions of β-catenin, c-myc and p-c-Jun. In this study, results show that miR-429 upregulation deactivated the Wnt/β-catenin pathway, while suppressing miR-429 promoted the activation of this pathway.

In conclusion, miR-429 was underexpressed in OS and repressed OS progression, whereas, HOXA9 was overexpressed in OS and facilitated OS progression. Moreover, HOXA9 was shown as the direct target of miR-429 in OS. Taken together, miR-429 suppressed OS progression by targeting HOXA9 via Wnt/β-catenin signaling pathway.

Acknowledgements

Not applicable.

Funding

No funding was received.

Availability of data and materials

The datasets used and/or analyzed during the present study areavailable from the corresponding author on reasonable request.

Authors' contributions

LS and LW contributed to the conception of the study. SL contributed significantly to the data analysis and study preparation. YJ and QW performed the data analyses and wrote the study. LS helped perform the analysis with constructive discussions. All authors read and approved the final study.

Ethics approval and consent to participate

The study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of Weifang People's Hospital (Weifang, China). Written informed consent was obtained from each patient and their parents before collecting the specimens.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References