Abstract

Objective

   Osteosarcoma is the most malignant and common primary bone tumor with a
   high rate of recurrence that mainly occurs in children and young
   adults. Therefore, it is vital to facilitate the development of novel
   effective therapeutic means and improve the overall prognosis of
   osteosarcoma patients via a deeper understanding of the mechanisms of
   chemoresistance in osteosarcoma progression.

Methods

   In this research, the relationship between ITGB3 and the clinical
   characteristics of patients was detected through analysis of publicly
   available clinical datasets. The expression of ITGB3 was analysis in
   collected human osteosarcoma tissues. In addition, the potential
   functions of ITGB3 in the cisplatin resistance of osteosarcoma cells
   were investigated in vitro and in tumor xenotransplantation. Finally,
   the molecular mechanism of ITGB3 in the progression and recurrence of
   osteosarcoma were explored via transcriptome analysis.

Results

   ITGB3 was identified as a potential regulator of tumorigenicity and
   cisplatin resistance in relapsed osteosarcoma. Furthermore, the
   decreased osteosarcoma cell proliferation and migration ability in
   ITGB3 knockout osteosarcoma cells were related to increased apoptosis
   and slowing cell cycle progression. In addition, ITGB3 had a positive
   correlation with cisplatin resistance in cells and tumor xenografts in
   mice. Accordingly, ITGB3 performed the functions of proliferation and
   cisplatin resistance in osteosarcoma through the MAPK and VEGF
   signaling pathways.

Conclusion

   Our results will contribute to a better understanding of the function
   and mechanism of ITGB3 in osteosarcoma cisplatin resistance and provide
   a novel therapeutic target to decrease cisplatin resistance and tumor
   recurrence in osteosarcoma patients.

   Keywords: apoptosis, cisplatin, CRISPR/Cas9, ITGB3, osteosarcoma,
   transcriptome
     __________________________________________________________________

   In the present work, ITGB3 knockout osteosarcoma cells had the
   decreased proliferation and migration ability, increased apoptosis and
   slowing cell cycle progression in vitro, and reduced cisplatin
   resistance in vivo. Further analyses suggested that MAPK and VEGF
   signaling pathways may involved in these process.

   graphic file with name CAM4-12-8452-g002.jpg

1. INTRODUCTION

   Osteosarcoma is one of the most common primary and malignant bone
   tumors and mainly affects children and adolescents.[42] ^1 The survival
   rate of osteosarcoma patients has been improved by treatment with a
   combination of surgery and perioperative neoadjuvant chemotherapy, in
   which cisplatin, doxorubicin, and methotrexate are commonly used
   chemotherapeutic drugs.[43] ^2 However, in patients who respond poorly
   to these drugs, the rate has remained roughly constant over the past
   three decades.[44] ^3 , [45]^4 , [46]^5 Even though many patients who
   use additional doses or drugs still experience distant metastasis and
   relapse, the five‐year survival rate is only approximately 20%.[47] ^6
   , [48]^7 Notably, poor prognosis is closely related to drug resistance.
   Therefore, to develop novel effective treatment means and dramatically
   improve osteosarcoma patient prognosis, it is imperative to explore the
   molecular mechanisms of osteosarcoma chemotherapy resistance.

   β3 integrin (ITGB3), also called GP3A or CD61, is one of the most
   extensively studied members of the integrin family. ITGB3 plays diverse
   important functions in the progression of malignant tumors and the
   molecular reprogramming of the microenvironment.[49] ^8 In recent
   years, some research has reported that there is a strong relationship
   between ITGB3 and drug resistance.[50] ^9 , [51]^10 , [52]^11 Naik et
   al. indicated that the NRP1‐ITGB3 axis also mediates the
   chemoresistance response of breast cancer cells.[53] ^12 Other evidence
   suggests that inhibiting ITGB3 expression increases the antitumor
   activity of ALK inhibitors in ALK‐rearranged non‐small cell lung
   cancer.[54] ^13 The overexpression of ITGB3 is also involved in the
   resistance to EGFR inhibition, mechanistically due to the complex
   formed by ITGB3/KRAS/RalB and the activation of TBK1 and NFκB that the
   complex mediates.[55] ^14 , [56]^15 However, the relationship between
   ITGB3 expression and clinical prognosis and recurrence in osteosarcoma
   remains unclear.

   In this research, we detected the relationship between ITGB3 and the
   clinical characteristics of patients and analyzed the expression of
   ITGB3 in human osteosarcoma tissues. In addition, the potential
   functions of ITGB3 in the cisplatin resistance of osteosarcoma cells
   were investigated in vitro and in tumor xenotransplantation. Finally,
   we further explored the molecular mechanism of ITGB3 in the progression
   and recurrence of osteosarcoma.

2. MATERIALS AND METHODS

2.1. Patients and tissue samples

   The raw count data of RNA‐sequencing (RNA‐Seq) and the relevant
   clinical information of 95 osteosarcoma samples in this study were
   obtained from the TARGET (Therapeutically Applicable Research to
   Generate Effective Treatments) dataset
   ([57]https://ocg.cancer.gov/programs/target), in which the requirements
   and application procedures complied with relevant protocols and
   policies. For Kaplan–Meier (KM) curves, the survival analysis with p
   values, hazard ratio with 95% confidence interval, and log‐rank test
   were chosen to compare the difference of overall survival (OS) and
   disease‐free survival (DFS) between two groups: ITGB3 high and
   low‐expression groups (Table [58]S1). Time‐series receiver operating
   characteristic analysis was performed to compare the predictive
   accuracy of the genes and their risk score. All R packages and
   analytical procedures were executed using R software (v4.0.3). Primary
   and recurrent osteosarcoma specimens assayed in this research were
   gathered from the Department of Spinal Surgery, The Affiliated
   Traditional Chinese Medicine Hospital of Southwest Medical University
   and informed consent was obtained from all study patients. This
   research was approved by the Ethics Committee. All procedures performed
   in this study were approved by the Ethics Committee of The Affiliated
   Traditional Chinese Medicine Hospital of Southwest Medical University
   (KY2022021) and followed the ethical standards and principles of the
   Helsinki Declaration and its later amendments.[59] ^16

2.2. Cells and culture medium

   The osteosarcoma cell lines (143B) were purchased from the National
   Infrastructure of Cell Line Resource (NICLR, China) and kept in our
   laboratory and were cultured in Dulbecco's modified Eagle's medium
   (Gibco) plus 10% fetal bovine serum (HyClone). The cell lines used in
   this research were cultured at 37°C with 5% CO[2]. When the cells
   occupied 80% confluence in the dish or plates, they were dissociated
   with 0.05% trypsin (Gibco) at 37°C for 3 min and harvested for
   passaging or used for subsequent experiments. The cell numbers were
   detected using a Rigel cytometer in this research (Countstar).

2.3. Cell counting kit‐8 assay

   The doubling time of cells was estimated by dividing the cell number by
   the culture time, and the cisplatin‐resistance effect was detected by
   using a cell counting kit‐8 (CCK‐8) assay kit (CK04‐01; DOJINDO). For
   cisplatin resistance analysis, the cells were seeded into 96‐well
   plates at a density of 1000 cells/well. At the indicated times, 10 μl
   of the CCK‐8 solution was then added to each well followed by 2 h of
   culture. The absorbance at 450 and 600 nm was measured by using a
   Synergy H1 multifunctional microplate reader (Biotek).

2.4. Immunohistochemistry and immunofluorescence

   Immunohistochemistry (IHC) and immunofluorescence (IF) were performed
   according to the procedures designed in previous studies.[60] ^17
   Briefly, cells and tumor tissues were fixed and permeabilized.
   Afterward, the cells were treated with the primary antibody, followed
   by incubation with the corresponding secondary antibody. The primary
   antibodies used in this study were anti‐Ki67 (Abcam; ab15580) and
   anti‐ITGB3 (Cell Signaling Technology; 13166), and the cell nuclei were
   labeled with 4′,6‐diamidino‐2‐phenylindole. IHC and IF images were
   obtained by using an FV3000 confocal fluorescence microscope (Olympus).
   To compare gene expression differences in different samples, the
   standard procedures described in previous studies[61] ^17 , [62]^18
   were used to perform quantitative immunofluorescence (QIF) analysis.
   The relative value of QIF was quantified using ImageJ software.

2.5. CRISPR/Cas9‐mediated ITGB3 knockout in 143B cells

   To knock out ITGB3 in 143B cells, sgRNA (single guide RNA) sequences
   spanning exon 3 and exon 4 of ITGB3 were designed to decrease the
   potential off‐target points by using a web‐based tool
   ([63]http://www.e‐crisp.org/E‐CRISP/). Then, the two sequences of each
   point were annealed and ligated into the vector pX459 (Addgene; #62988)
   to generate two plasmid constructs, pX459‐ITGB3‐gRNA1 and
   pX459‐ITGB3‐gRNA2. These two vectors were coelectroporated into 143B
   cells by using the Neon Transfection System (Thermo). After 24 h of
   cultivation, 1 μg/ml puromycin was added and selected for 48 h, and
   then the electroporated cells were seeded on 10 cm dishes at a very low
   density of approximately 2000 cells/dish. On day 14, we manually picked
   approximately 100 puromycin‐resistant colonies into 96‐well plates and
   then expanded them separately. Genomic DNA was purified from these
   picked cell colonies, and the homozygous mutant cells were confirmed
   via PCR amplification and Sanger sequencing by using specific primers
   that spanned the knockout fragments. The oligo sequences of the two
   gRNAs and the detected primers were as follows:

   ITGB3‐gRNA1: TCACTCAAGTCAGTCCCCAG

   ITGB3‐gRNA2: GGTGAGCTTTCGCATCTGGG

   ITGB3‐KO‐F: ATAGCAGGGGTTTTCGAGGG

   ITGB3‐KO‐R: GCCATAGCTCTGATTGCTGG

2.6. Real‐time quantitative reverse transcription PCR

   Total RNA was purified using a TRIzol reagent (1–5 × 10^6 cells/1 ml
   TRIzol) according to the manufacturer's procedure (Invitrogen). One
   microgram of total RNA was reverse transcribed into first strand cDNA
   with the HiScript^III first Strand cDNA Synthesis Kit (Vazyme). Then,
   real‐time quantitative reverse transcription PCR (RT–qPCR) procedures
   were conducted in triplicate using Taq Pro Universal SYBR qPCR Master
   Mix (Vazyme). GAPDH was used as a loading control for each gene, and
   comparative CT analysis was used to quantify the gene expression. At
   least three experiments were repeated. The primers used in this study
   were as follows:

   hITG‐E6‐F: CCTAATGACGGGCAGTGTCA

   hITG‐E9‐R: TCACGCACTTCCAGCTCTAC

   hITG‐E5‐F: AGCAGAGTGTGTCACGGAAC

   hITG‐E7‐R: TCAGTCATCAGCCCCAAAGAG

2.7. Tumor xenotransplantation in vivo

   In the subcutaneous tumor xenotransplantation experiment, two cell
   lines, 143B and ITGB3‐KO cells, were washed and resuspended in
   phosphate‐buffered saline, and 1 × 10^6 cells (100 μl) were
   subcutaneously injected into 6‐week‐old NOD/SCID mice (Charles River).
   From the seventh day of inoculation, they were treated with a control
   solution (normal saline) and cisplatin (4 mg/kg). All groups received
   their own treatments every 3 days for a total of seven treatments. The
   long (l) and short (s) diameters of the tumor were measured weekly, and
   the tumor volume was calculated according to the following formula:
   volume = (l × s ^2)/2. Twenty‐eight days after cell injection, the mice
   were euthanized, and the tumors were collected for subsequent assays.
   All animal care and mouse testing procedures were approved by the
   Institutional Animal Care and Use Committee at Hospital (T.C.M.)
   Affiliated to Southwest Medical University.

2.8. RNA‐Seq and analysis

   All RNA‐sequencing (RNA‐Seq) reads of 143B and ITGB3‐KO were mapped to
   the human genome GRCh38 (Hg38) using Bowtie2 (version 2.2.4). The read
   counts per gene were detected by using feature counts (Version 1.6.4),
   and differentially expressed gene analysis was constructed by using
   DESeq2.[64] ^19 The Venn diagram was created using the Venny online
   tool ([65]http://bioinfogp.cnb.csic.es/tools/venny). Kyoto Encyclopedia
   of Genes and Genomes (KEGG) pathway enrichment analysis was performed
   using the Cytoscape software (Cytoscape 3.9.1) STRING module (Search
   Tool for the Retrieval of Interacting Genes/Proteins version 10.0).

3. RESULTS

3.1. ITGB3 stimulated proliferation in osteosarcoma carcinoma

   The RNA‐Seq raw count data and related clinical information of ITGB3
   were obtained from the TARGET dataset (Table [66]S1). Then, the risk
   score of every patient was calculated, the median cut‐off point was
   obtained, and the patients were divided into a high‐risk group (n = 47)
   and a low‐risk group (n = 48) based on ITGB3 expression
   (Figure [67]1A). KM survival analysis with the auto‐select best cut‐off
   was performed and revealed no significant difference between the
   high‐risk group and the low‐risk group for OS time (Figure [68]1B), but
   osteosarcoma patients with higher expression of ITGB3 had a shorter DFS
   time (Figure [69]1C). Since ITGB3 has an obvious causal or prognostic
   effect on osteosarcoma recurrence, we collected 18 paired osteosarcoma
   tissues (primary: n = 9, recurrence: n = 9) and detected the expression
   level of ITGB3 by IHC staining assays (Figure [70]1D). According to the
   expression level evaluation, ITGB3 was markedly overexpressed in
   recurrence tissues compared with primary tissues. Next, we assessed the
   cell division marker Ki67, which can indicate tumor cell growth, to
   confirm whether increased ITGB3 expression was associated with tumor
   cell proliferation (Figure [71]1D). The ratios of ITGB3‐positive
   (ITGB3^+), Ki67‐positive (Ki67^+), and ITGB3^+/Ki67^+ double‐positive
   (Figure [72]1E) osteosarcoma cells in the recurrence groups were
   significantly higher than those in the primary groups, which was in
   agreement with our prediction. Based on the above results, the
   increased expression of ITGB3 may facilitate cell proliferation and
   promote human osteosarcoma growth.

FIGURE 1.

   FIGURE 1
   [73]Open in a new tab

   ITGB3 overexpression was associated with disease‐free survival and
   stimulated proliferation in osteosarcoma tissues. (A) Prognostic
   analysis of the ITGB3 signature in the TARGET datasets, where the top
   represents the scatter diagram of the gene expression of the median
   risk score from low to high and divides the patients into low‐risk
   (blue) and high‐risk (red) groups. The middle panel represents the
   scatter plot distribution of survival status and times responding to
   the expression of ITGB3 in different patients' tumor samples. At the
   bottom, the heatmap of the expression of ITGB3 is displayed. All the
   x‐axes of the top, middle, and bottom of the three figures represent
   the 95 patients' tumor samples, and the order was consistent. (B)
   Kaplan–Meier curve showing OS in the ITGB3 high‐expression and
   low‐expression groups. Among them, different groups were tested by the
   log‐rank. HR (high EXP) represents the risk coefficient of the
   high‐expression groups compared with the low‐expression groups. If
   HR > 1, this gene is a risk factor, and if HR < 1, this gene is a
   protective factor. The 95% CL represents the HR confidence interval;
   the median time represents the corresponding time (median survival
   time) of the high‐expression and low‐expression groups, and the unit is
   years. (C) Kaplan–Meier curve revealing the DFS in the ITGB3
   high‐expression and low‐expression groups. The patients were grouped as
   shown in (B). (D) IHC staining was used to assess endogenous Ki67 and
   ITGB3 expression in paired primary and recurrent osteosarcoma tissues.
   IHC images were obtained by using a standard confocal fluorescence
   microscope (FV3000). Scale bar: 50 μm. (E) Histogram graph revealing
   the ratio of double‐positive ITGB3^+/Ki67^+ cells in primary and
   recurrent osteosarcoma samples. Student's t test was used to calculate
   the p‐value. CI, confidence interval; DFS, disease‐free survival; HR,
   hazard ratio; IHC, immunohistochemistry; OS, overall survival.

3.2. Knockout of ITGB3 reduced osteosarcoma cell proliferation

   To further validate the potential functions of ITGB3 on osteosarcoma
   cell growth and migration in vitro, we used the CRISPR/Cas9‐mediated
   disruption system to generate the ITGB3‐knockout 143B cell line
   (Figure [74]2A). Genomic PCR products (Figure [75]2B) and Sanger
   sequencing (Figure [76]2C) confirmed that both alleles were mutated in
   ITGB3‐KO mice. RT–qPCR (Figure [77]2D), western blot (Figure [78]2E),
   and IF (Figure [79]2E) assays were used to detect ITGB3 expression and
   to clarify the knockout efficiency in ITGB3‐KO mutant cell lines. The
   growth of ITGB3 knockout 143B cells was markedly inhibited
   (Figure [80]2F). Furthermore, to determine whether ITGB3 affected cell
   migration, scratch wound healing experiments were performed and
   revealed that the scratch wound areas in ITGB3‐KO cells were
   significantly greater than those in 143B cells, which indicated that
   ITGB3 may play important functions in osteosarcoma motility
   (Figure [81]2G,H). Collectively, these results indicated that ITGB3
   promotes osteosarcoma cell proliferation in vitro.

FIGURE 2.

   FIGURE 2
   [82]Open in a new tab

   Knockout of ITGB3 reduced osteosarcoma cell proliferation. (A) Strategy
   to generate ITGB3‐KO cells by using the CRISPR/Cas9 system in the 143B
   cell line. In the upper panel, the exon information is shown. The
   target site and sgRNA sequences of exons 3 and 4 on ITGB3 are labeled
   in red. (B) Identification of the band of ITGB3^−/− (ITGB3‐KO) cells
   via gel electrophoresis of genomic PCR products. WT: wild‐type 143B
   cells. (C) Both alleles of targeted sequences were detected by Sanger
   sequencing in one mutant clone. A 28‐bp exon 3 and 140‐bp exon 4 were
   excised in one allele, and 33‐bp exon 3 and 138‐bp exon 4 were excised
   in another allele. (D) The expression levels of ITGB3 were detected by
   real‐time quantitative reverse transcription PCR analysis in the mutant
   ITGB3‐KO cell line. The expression levels of ITGB3 were normalized to
   those of GAPDH. Two paired primers and three replicated experiments
   were utilized, and the p‐values were calculated by using Student's t
   test. (E) Western blot analysis was used to detect the expression of
   ITGB3 in WT and ITGB3‐KO cell lines. GAPDH was used as an internal
   loading control. (F) The ITGB3 knockout efficiency in ITGB3‐KO cells
   was assessed by IF staining. 4′,6‐diamidino‐2‐phenylindole staining
   served as the internal loading control. (G) The number of ITGB3‐KO and
   143B cells was counted on the indicated days in culture to evaluate the
   proliferation rate. (H) The scratch wound‐healing experiments showed
   the scratch region of 143B and ITGB3‐KO cells 24 h later. (I) Analysis
   of the percentage of the scratch area of 143B and ITGB3‐KO cells. The
   scratched area of wound healing was normalized to the initial area at
   0 h. All data in these figures were detected by three independent
   experiments, and the student's t test was used and is shown as the
   mean ± standard deviation. IF, immunofluorescence.

3.3. ITGB3 reduces osteosarcoma cell sensitivity to cisplatin

   Notably, ITGB3 knockout decreased cell viability after treatment with
   10, 20, and 30 μM cisplatin (Figure [83]3A). We then tested the
   functions of ITGB3 in the cisplatin resistance of osteosarcoma cells by
   using CCK‐8 assays and 143B cell knockout for ITGB3. ITGB3 knockout
   inhibited the viability of 143B cells at 24 h after treatment with 15
   and 25 μM cisplatin (Figure [84]3B). We also analyzed the functions of
   ITGB3 in cisplatin‐induced apoptosis. ITGB3 KO significantly increased
   the rate of cisplatin‐induced apoptosis (25.3% vs. 12.86%, p = 0.0006)
   (Figure [85]3C,D). The cell cycle was analyzed by flow cytometry, which
   showed a decreased S‐phase ratio and an increased G1‐phase ratio after
   ITGB3 knockout. These results revealed that ITGB3 affects cell cycle
   progression (Figure [86]3E) and that the knockout of ITGB3 was
   negatively related to the cisplatin resistance of osteosarcoma cells in
   vitro. To summarize, ITGB3 expression was positively related to
   increased cisplatin resistance and decreased apoptosis in osteosarcoma
   progression.

FIGURE 3.

   FIGURE 3
   [87]Open in a new tab

   ITGB3 knockout reduced cisplatin resistance and promoted apoptosis. (A)
   Light microscopic images of 143B and ITGB3‐KO cells treated with
   different concentrations of cisplatin. (B) Graphical presentation of
   the viability of 143B and ITGB3‐KO cells treated with cisplatin. (C)
   Annexin V‐FITC/propidium iodide staining and FACS analysis were used to
   detect the percentage of apoptotic cells in 143B and ITGB‐KO cells with
   or without cisplatin treatment for 24 h. (D) The bar graph shows the
   percentage of Annexin V‐positive apoptotic cells between WT and
   ITGB3‐KO cells. WT: 143B cells. (E) The percentage of each cell phase
   among WT and ITGB3‐KO cells is shown in a histogram chart. All data
   were detected by three independent experiments, and Student's t test
   was used and is shown as the mean ± standard deviation. ns, not
   significant.

3.4. ITGB3 knockout reduced tumor cisplatin resistance in vivo

   To study whether ITGB3 can affect osteosarcoma cisplatin resistance in
   vivo, we subcutaneously injected ITGB3‐KO and 143B cells into NOD/SCID
   mice (day 0). From the seventh day of inoculation, they were treated
   with a control solution (normal saline) and cisplatin (4 mg/kg). All
   groups received their own treatments every 3 days for a total of seven
   treatments. The long and short diameters of the tumors were measured
   once a week. At day 28, the tumor tissues were isolated from the
   tumor‐bearing NOD/SCID mice (Figure [88]4A,B). When comparing the
   ITGB3‐KO group and the 143B group, the tumor weights were decreased
   (Figure [89]4C), and tumor volumes were diminished (Figure [90]4D),
   which indicated that ITGB3 knockout could suppress tumor growth in
   vivo. Further analysis revealed that treatment with cisplatin
   significantly decreased the weights (Figure [91]4C) and volumes
   (Figure [92]4D) of the tumors derived from the ITGB3‐KO cell line,
   which was in accordance with our in vitro data. Therefore, these
   findings support that ITGB3 knockout could potentiate osteosarcoma
   sensitivity to cisplatin in vivo.

FIGURE 4.

   FIGURE 4
   [93]Open in a new tab

   ITGB3 knockout reduced cisplatin resistance in vivo. (A) 143B and
   ITGB3‐KO cells were injected subcutaneously into the right side of the
   abdomen of NOD/SCID mice (1 × 10^6 cells/site, n = 6). On day 28, the
   xenograft tumor‐bearing mice were anesthetized, sacrificed, and
   photographed. Scale bar: 1 cm. A representative picture of the tumor is
   shown in this figure, and other tumor samples are shown in
   Figure [94]S1. (B) NOD/SCID mice were inoculated with 143B and ITGB3‐KO
   cells (1 × 10^6 cells/site, n = 6). On day 7, treatment was initiated.
   In the cisplatin group, cisplatin (4 mg/kg) in 100 μl of 0.9% saline
   solution was administered via intraperitoneal injection. In the control
   group, the control solution (200 μl of 0.9% saline) was injected. All
   groups were administered their own treatments every 3 days for a total
   of seven times. A representative picture of the tumor is shown in this
   figure, and other tumor samples are shown in Figure [95]S1. (C) The
   tumor volume was observed and measured once a week in WT and ITGB3‐KO
   mice with or without cisplatin treatment (each group, n = 6). (D) The
   weights of the tumors were weighed when the mice were sacrificed (each
   group, n = 6).

3.5. Transcriptome signatures associated with ITGB3 in osteosarcoma

   To elucidate the underlying mechanism of the anti‐cisplatin effect of
   ITGB3 in osteosarcoma, we collected total RNA and performed
   transcriptome analysis on 143B (WT), ITGB3‐KO (KO), cisplatin‐treated
   143B (Cis_WT) and ITGB3‐KO (Cis_KO) cells. The correlation heatmap
   (Figure [96]5A,B) and PCA (Figure [97]S1) revealed that KO and WT
   showed small differences, while Cis_KO and Cis_WT showed significant
   differences. There was a different expression pattern when comparing
   Cis_WT with WT cells and Cis_KO with WT cells, which was demonstrated
   via a volcano plot analysis (Figure [98]5C,D). There were 1221
   upregulated genes (Cis_WT_up) and 1115 downregulated genes
   (Cis_WT_down) in Cis_WT compared with WT, and there were 842
   upregulated genes (Cis_KO_up) and 1348 downregulated genes
   (Cis_KO_down) in Cis_KO compared with KO (Figure [99]5C,D;
   Table [100]S2). The number of up and downregulated genes between the
   Cis_KO and Cis_WT groups is shown (Figure [101]5E; Table [102]S2). To
   investigate the mechanism by which ITGB3 exerts cisplatin resistance in
   osteosarcoma, we analyzed the genes that were only upregulated and
   downregulated in the Cis_KO group using the computational tool
   Cytoscape and STRING enrichment analysis. It is interesting to note
   that KEGG pathway enrichment was observed only in the downregulated
   group (Figure [103]5F; Table [104]S3). Among these enrichments, some
   signaling pathways were commonly related to cisplatin resistance in
   osteosarcomas, such as the mitogen‐activated protein kinase (MAPK)
   pathway[105] ^20 , [106]^21 , [107]^22 , [108]^23 , [109]^24 and
   vascular endothelial growth factor (VEGF).[110] ^25 , [111]^26 ,
   [112]^27 Additionally, 10 hub genes were identified, and their
   associated signaling pathways were demonstrated (Figure [113]5G). These
   hub genes also appeared to play a regulatory role in reducing cisplatin
   resistance when ITGB3 was knocked out (Figure [114]5G). ICAM1, PTGS2,
   HRAS, and NFKBIA. Accordingly, these analyses suggested that ITGB3 may
   perform the functions of cisplatin resistance in osteosarcoma through
   these signaling pathways.

FIGURE 5.

   FIGURE 5
   [115]Open in a new tab

   The transcriptome of ITGB3‐KO cells. (A) Pearson correlation heatmap
   associating WT, ITGB3‐KO, Cis_WT, and Cis_KO. Colors from blue to red
   indicate weak to strong correlations. KO: ITGB3‐KO cell line; WT: 143B
   cell line. (B) A heatmap shows the different levels of RNA expression
   in WT, KO, Cis_WT, and Cis_KO. (C) A volcano plot displaying the DEGs
   when comparing Cis_KO to KO. Red represents the upregulated genes
   (log2FoldChange >1, −log10 adjusted p value >2) in Cis_KO compared with
   KO. Green indicates the downregulated genes (log2FoldChange <−1, −log10
   adjusted p value >2) in Cis_KO compared with KO. Gray represents
   nonsignificant DEGs. (D) A volcano plot displaying the DEGs when
   comparing Cis_WT to WT. Red represents the upregulated genes
   (log2FoldChange >1, −log10 adjusted p value >2) in Cis_WT compared with
   WT. Green indicates downregulated genes (log2FoldChange <−1, −log10
   adjust p value >2) in Cis_WT compared with WT. Gray represents
   nonsignificant DEGs. (E) The Venn diagram on the left shows the overlap
   of genes between Cis_KO_down and Cis_WT_down, and the right shows the
   overlap of genes between Cis_KO_up and Cis_WT_up. (F) The enriched KEGG
   pathways of Cis_KO_down were generated based on the STRING database in
   Cytoscape. (G) The 10 hub genes and real genes of the KEGG pathways of
   Cis_KO_down are shown in (F) with the MCODE algorithm in Cytoscape. (H)
   Heatmap showing the expression of ITGB3 and 10 hub genes in WT, KO,
   Cis_WT, and Cis_KO. The raw data of (B–E) are shown in Table [116]S2,
   and the raw data of (F) are shown in Table [117]S3. DEG, differentially
   expressed gene; KEGG, Kyoto Encyclopedia of Genes and Genomes.

4. DISCUSSION

   Long‐term clinical drug therapy may cause drug resistance, followed by
   osteosarcoma recurrence and poor clinical outcomes. Cisplatin, one of
   the most common first‐line chemotherapy drugs, is the most active drug
   for osteosarcoma treatment.[118] ^28 Therefore, a better understanding
   of the underlying molecular signaling networks that regulate cisplatin
   resistance may represent more feasible options to effectively decrease
   drug resistance and improve patient OS. In the present work, ITGB3 was
   identified as a potential regulator of tumorigenicity and cisplatin
   resistance in osteosarcoma through public clinical dataset (TARGET
   database) analysis, clinical sample detection, and functional
   experimental model testing. First, using a clinical database and
   clinical samples, we found that ITGB3 was more highly expressed in
   recurrent osteosarcoma patients and was associated with tumor
   proliferation. Furthermore, the decreased osteosarcoma cell
   proliferation and migration ability in ITGB3 knockout osteosarcoma
   cells were also confirmed. The decreased cell growth was related to
   increased apoptosis and slowing cell cycle progression. We further
   discovered that ITGB3 had a positive correlation with cisplatin
   resistance through tumor xenografts in mice.

   Next, through Cytoscape and STRING analysis of cisplatin‐treated or
   nontreated WT and ITGB3‐KO cell‐related RNA‐Seq data, we found that
   ITGB3 plays a role in promoting osteosarcoma progression through
   multiple signaling pathways. The combined analysis of hub genes and
   enriched signaling pathways (Figures [119]S2–S4) revealed that hub
   genes were regulated by these signaling pathways, including the MAPK,
   VEGF, C‐type lectin receptor, and cell adhesion molecule signaling
   pathways. Among them, the key hub genes were ICAM1 (intercellular
   adhesion molecule 1), HRAS (Ras family small GTP binding protein
   H‐Ras), PTGS2 (prostaglandin‐endoperoxide synthase 2, also known as
   COX‐2), and FOS (FBJ osteosarcoma oncogene). Interleukin‐6 (IL‐6)
   activated the expression of ICAM‐1 and contributed to the migration of
   human osteosarcoma cells through integrin‐linked kinase (ILK)/Akt/AP‐1
   pathways.[120] ^29 Osteosarcoma cells colonize the lungs more quickly
   by upregulating ICAM1 expression in response to tumor‐derived IL‐6,
   which facilitates glycolytic metabolism in tumor cells by activating
   the MEK/ERK/HIF‐1α pathway.[121] ^30 On the surface of the endothelium,
   cisplatin upregulates the expression of intercellular adhesion
   molecules, which causes increased infiltration of monocytes or
   leukocytes. Cisplatin upregulated endothelial ICAM‐1 expression through
   an NF‐κB‐dependent mechanism ([122]31). HRAS belongs to the Ras
   oncogene family, whose members are associated with mammalian sarcoma
   retroviruses. In osteosarcoma cells, a dominant‐negative H‐RAS mutant
   (17N) partially suppressed ERK activation and delayed apoptosis induced
   by chelerythrine.[123] ^32 Notably, gallium(III) complex 1, which has
   higher toxicity toward osteosarcoma cells grown than cisplatin,
   downregulates COX‐2 expression and kills osteosarcoma cells in a
   COX‐2‐dependent manner.[124] ^33 ERK/JNK and c‐JUN/c‐FOS, as upstream
   activators of FOXP1, drive osteosarcoma development.[125] ^34 FOS also
   efficiently initiates key stem cell proliferation, migration, division,
   and differentiation.[126] ^35 FOS can regulate prognostic genes related
   to osteosarcoma metastasis.[127] ^36 These results indicated that
   ICAM1, HRAS, PTGS2, and FOS, as downstream genes, may predict the
   potential of cisplatin treatment. Thus, targeted suppression of the
   MAPK and VEGF pathways in combination with cisplatin treatment may be a
   promising method for overcoming osteosarcoma chemoresistance. In
   conclusion, a thorough investigation of the molecular mechanism by
   which ITGB3 participates in osteosarcoma cisplatin resistance via the
   involvement of the signaling pathway should promote treatment effects.

   Taken together, our data showed that ITGB3 may be a critical regulator
   of cisplatin resistance in osteosarcoma. Our results will contribute to
   a better understanding of the function and mechanism of ITGB3 in
   osteosarcoma cisplatin resistance and provide a novel therapeutic
   target to decrease cisplatin resistance and tumor recurrence in
   osteosarcoma patients.

AUTHOR CONTRIBUTIONS

   Qian Li: Data curation (equal); methodology (equal). Guangyou Chen:
   Conceptualization (equal); data curation (equal). Huachai Jiang: Data
   curation (equal); software (equal). Haoping Dai: Methodology
   (supporting). Dongdong Li: Data curation (supporting); software
   (supporting). Kai Zhu: Methodology (supporting). Kaiquan Zhang:
   Methodology (supporting). Huarui Shen: Funding acquisition
   (supporting); writing – original draft (equal). Houping Xu: Funding
   acquisition (equal); writing – original draft (equal); writing – review
   and editing (equal). Sen Li: Funding acquisition (lead); writing –
   review and editing (lead).

CONFLICT OF INTEREST STATEMENT

   All the authors declare no competing financial interests.

Supporting information

   Table S1.
   [128]Click here for additional data file.^ (15.6KB, xlsx)

   Table S2.
   [129]Click here for additional data file.^ (323.4KB, xlsx)

   Table S3.
   [130]Click here for additional data file.^ (189.7KB, xlsx)

   Figure S1.

   Figure S2.

   Figure S3.

   Figure S4.

   Figure S5.

   Figure S6.
   [131]Click here for additional data file.^ (2.3MB, pdf)

ACKNOWLEDGMENTS