To investigate the expression of alpha–smooth muscle actin (ACTA2) in osteosarcoma tissues and its relationship with prognosis.
Prognostic analysis of lung metastasis–related genes in osteosarcoma using the TCGA database. Single-cell sequencing detected the expression of ACTA2 in 11 osteosarcoma tissues. Paraffin-embedded tissues of 74 osteosarcoma patients treated at the Sixth People’s Hospital of Shanghai Jiao Tong University from 2014 to 2019 were collected, and tissue microarrays were prepared. ACTA2 expression was detected and scored by immunohistochemistry. According to the median value of the ACTA2 histochemical score, 74 patients were divided into two groups, the high-expression group and low-expression group, and the relationship of expression with clinicopathological characteristics and prognosis was analyzed by Cox regression.
Through analysis of ACTA2 expression by single-cell sequencing of osteosarcoma samples together with an immuno-microarray, we found moderate ACTA2 expression. Upon analyzing the prognostic impact of ACTA2, CCL2, TGFBI, VEGFA, PDGFB, PDGFC, COL1A1, COL14A1, CXCL12, CXCL14, CSPP1, LUM, DES, MYL9, and SFRP2 on osteosarcoma patients using the TCGA database, we found that patients with high ACTA2 expression had a significantly better prognosis than those with low ACTA2 expression. Patients with high expression of ACTA2 in osteosarcoma lung metastases showed longer progression-free survival and overall survival than those with low expression.
High expression of ACTA2 in patients with osteosarcoma lung metastasis suggests a better prognosis.
Osteosarcoma is the most common primary malignant bone tumor. It occurs mostly during the growth spurt of adolescence and is most often seen in areas of rapid skeletal growth, such as the epiphyseal regions of long bones [
At the beginning of our study, analysis of the TCGA database for genes related to osteosarcoma lung metastasis, we found that patients with osteosarcoma with high ACTA2 expression had a better prognosis, which caught our attention.
In a previous study in our center, we found that ACTA2, CCL2, TGFBI, VEGFA, PDGFB, PDGFC, COL1A1, COL14A1, CXCL12, CXCL14, CSPP1, LUM, DES, MYL9, and SFRP2 were differentially expressed in osteosarcoma samples from different patients. We decided to perform a prognostic analysis of these factors in osteosarcoma. Therefore, we accessed The Atlas of the Human Genome (TCGA) homepage (
n (%) | |
---|---|
Race | |
White | 58 (65.9) |
Black or African-American people | 8 (9.0) |
Asian | 3 (3.4) |
Unmarked | 19 (21.5) |
Status of survival | |
Death | 16 (18.1) |
Survival | 72 (81.9) |
Age (years) | |
<60 | 88 (100) |
≥60 | 0 (0) |
Gender | |
Female | 41 (46.5) |
Male | 47 (53.5) |
Metastasis | |
M0 | 70 (79.5) |
M1 | 18 (20.5) |
Prognostic information of 88 patients with primary osteosarcoma was obtained from the TCGA osteosarcoma database. After testing the completeness of the genetic profile, we performed a multifactorial prognostic analysis on the 88 samples. First, we took the median values for the expression of transcriptomic ACTA2, CCL2, TGFBI, VEGFA, PDGFB, PDGFC, COL1A1, COL14A1, CXCL12, CXCL14, CSPP1, LUM, DES, MYL9, and SFRP2 genes. Values higher than the median value were classified as the high-expression group, and those lower than the median value were considered the low-expression group. Afterwards, we integrated the high- and low-expression groups of these genes with the clinical prognostic information of the patients, and used the “survival” package of R software to plot the overall survival curves and calculate the survival rates for the high- and low-expression groups of osteosarcoma patients. Receiver operating characteristic (ROC) curves were plotted using the “survival ROC” package in R software, and the area under the curve (AUC) was calculated to assess the efficacy of the prognostic gene model.
The survival status and survival time were extracted from the clinical data of osteosarcoma using Perl script, and combined with the gene corrected expression. The survival curve of each gene was plotted using the survival package in R software. The five-year survival rate was calculated for patients with high and low expression of each gene.
The fresh surgical tissues of 87 osteosarcoma patients treated at the Department of Oncology of Shanghai Sixth People’s Hospital from August 2014 to January 2019 were collected. Due to the loss during tissue preparation and the loss to follow-up, 13 cases were not included in the analysis, so the follow-up and data analysis were completed for 74 patients. Tissue microarrays were prepared from paraffin-embedded tissue of primary osteosarcoma: tissue chip 1 and tissue chip 2. This study was a retrospective analysis. It was approved by the hospital ethics committee, and all of the patients signed the informed consent form.
This was a retrospective study, the inclusion criteria of which were as follows: 1. patients aged 5–70 years; 2. were either male or female; 3. had osteosarcoma confirmed by pathological analysis; 4. had an expected survival time of more than 6 months; and 5. received standardized treatment for osteosarcoma. The exclusion criteria were: 1. patient could not be confirmed by pathological analysis; 2. patient failed to complete at least one survival follow-up or missed follow-up; 3. patient refused to receive standardized antineoplastic therapy; 4. patient had a history of other malignant tumors; and 5. patients had severe a chronic or immunodeficiency disease.
All 74 patients received pathological confirmation of the diagnosis of osteosarcoma. The patients included 37 males and 37 females, aged 6 to 65 years, with a mean age of 27.5 years. 65 cases were of common pathological type, while nine cases were of uncommon pathological type. There were 44 cases in the femur, 15 cases in the tibia, nine cases in the humerus, two cases in the fibula, and four cases in other skeletal sites. The necrosis rate after neoadjuvant chemotherapy was greater than 90% in 12 cases and lower than 90% in 44 cases, and no necrosis rate results were obtained in 18 cases.
Individual cells were encapsulated into emulsion droplets using a Chromium Controller (10 × Genomics). scRNA-seq libraries were constructed using a Chromium Single Cell 3′Library Kit and a Gel Bead & Multiplex Kit (10 × Genomics, V2 and V3) in accordance with the instructions. The cells were divided into Gel Beads in Emulsion in a ChromiumTM controller, lysed, and reverse-transcribed for RNA barcoding. Sequencing libraries were constructed using cDNA amplified by a Nextera XT DNA sample Pre-Kit (hc-131-1024, Illumina, Shanghai, China), and the final libraries of individual samples were evaluated on an Agilent Bioanalyzer using a high-sensitivity DNA kit (Agilent Technologies, Shanghai, China). The libraries from 75 kit runs were sequenced on an Illumina HiSeq X platform (Illumina, Shanghai, China) in 150 bp paired-end read mode. Seurat objects with gene expression data from individual samples were processed using the Read (10 × Genomics) function. For each sample, gene expression was calculated by determining the percentage of the expression of that gene multiplied by 10,000 and converting the value to the natural logarithm; 1 was added and the value was normalized to avoid taking the logarithm of 0. The top 3,000 highly variable genes (HVG) were filtered from the normalized expression matrix, centered, and scaled for principal component analysis (PCA). Based on the top 50 PCA components, the Harmony package for R (version 1.0) eliminated the batch effect.
Immunohistochemistry (IHC) was performed by the Strept Avidin-Biotin Complex (SABC) method for staining, and the steps were performed in accordance with the kit instructions. The mouse monoclonal antibody ACTA2 (ab7817, Abcam Company, Cambridgeshire, UK) was used at a concentration of 0.034 μg/mL. Heat-mediated antigen retrieval was performed with citrate buffer (pH 6) before starting IHC staining. Positive IHC results were scored as follows: the mean positive intensity of the measured areas was scored as 0, 1, 2, or 3, where negative staining was scored as 0; weakly positive pale-yellow staining was recorded as 1; moderately positive tan staining was recorded as 2’ and strongly positive tan staining was recorded as 3. The cell positivity ratio was scored as follows: 0% to 5% was scored as 0; 6% to 25% was scored as 1; 26% to 50% was scored as 2; 51% to 75% was scored as 3; and >75% was scored as 4. The composite score of positivity was determined as the staining intensity value × positive cell ratio score; a larger value indicated a stronger composite positive intensity.
After paraffin sections of patients with osteosarcoma had been dewaxed and hydrated, they were repaired with mouse monoclonal antibody ACTA2 (ab7817, Abcam Company, Cambridgeshire, UK) as the primary antibody and incubated at room temperature. After incubation with secondary antibodies, the slices were blocked by re-staining. After each round of staining, a 3D Histech Pannoramic MIDI digital section scanner (3DHISTECH, Budapest, Hungary) was used for fluorescence scanning.
The pseudo-color of the antibody marker was set before scanning, and here we labeled ACTA2 with cy3-labeled goat anti-mouse IgG secondary antibody (A0521, Beyotime Company, Shanghai, China). The CaseViewer image software was used to align and overlay the fluorescence images of each round to achieve the final multiplexed immunofluorescence staining.
Follow-up was conducted until progression or death by outpatient visits and telephone, with a final follow-up date of January 2022. Disease-free survival was defined as survival from the time of surgery until metastasis or recurrence at lung, bone, and other sites as evidenced by imaging. By January 2022, a total of 51 of 74 patients survived, and 23 died. Analysis of the course of the disease showed 30 patients with definite pulmonary metastases, 20 patients with local recurrence, 10 patients with bone metastases, and five patients with metastases to other sites. By the follow-up cutoff, the shortest survival period was 12 months, and the longest survival period was 60 months.
SPSS 20.0 software was used for data analysis. The measurement data are expressed as the mean ± standard deviation, and the median ACTA2 immunohistochemistry (IHC) score was used to divide the defined high- and low-expression groups. The patients’ sex, age, tumor size, pathological type, tumor site, tumor necrosis rate, lung metastasis, recurrence, bone metastasis, and metastasis at other sites were used as covariates, and multifactor ANOVA was used. The Kaplan–Meier method was used to calculate the survival rate, and analysis of variance was performed by the log-rank test. The test level was α = 0.05.
Based on the joint analysis of the TCGA’s osteosarcoma transcriptome database and clinical database, there was no strong correlation of CCL2, TGFBI, VEGFA, PDGFB, PDGFC, COL1A1, COL14A1, CXCL12, CXCL14, CSPP1, LUM, DES, MYL9, and SFRP2 with prognosis in 88 osteosarcoma samples, while the prognosis of patients with high expression of ACTA2 was significantly higher than that of patients with low expression (
The five-year survival rates of patients in the CCL2, TGFBI, VEGFA, PDGFB, PDGFC, COL1A1, COL14A1, CXCL12, CXCL14, CSPP1, LUM, DES, MYL9, and SFRP2 high-expression group were not significantly different from those in the low-expression group, while the five-year survival rates of patients in the ACTA2 high-expression group were significantly higher than those in the low-expression group (
Gene | Survival rate in the high-expression group (%) | Survival rate in the low-expression group (%) | |
---|---|---|---|
81.0 | 43.6 | 0.002 | |
74.1 | 55.6 | 0.088 | |
61.2 | 69.8 | 0.832 | |
54.2 | 71.8 | 0.248 | |
68.5 | 58.9 | 0.582 | |
62.8 | 68.9 | 0.577 | |
57.0 | 70.1 | 0.370 | |
68.0 | 61.2 | 0.518 | |
73.1 | 63.2 | 0.501 | |
67.8 | 68.2 | 0.620 | |
51.5 | 74.2 | 0.058 | |
65.6 | 59.4 | 0.236 | |
70.1 | 61.5 | 0.392 | |
78.2 | 53.8 | 0.051 | |
60.5 | 70.1 | 0.504 |
Based on the Harmony and Louvain algorithm, we identified the main subcluster cell groups in osteosarcoma tissues. Osteoblasts and chondroblasts are the most common pathological subtypes of conventional osteosarcoma. Through t-distributed stochastic neighbor embedding (t-SNE) analysis of malignant osteosarcoma cells, we identified seven subpopulations, including six osteoblast lines and one chondroblastic line. ACTA2 was mainly expressed in the third group of osteosarcoma cells, pericytes, fibroblasts, and mesenchymal stem cells, as reported in our previous study [Nature communications. 2020; 11: 6322] (
The tissue microarray scores of 74 primary osteosarcoma foci were calculated according to the scoring criteria of immunohistochemical results, and the median values were taken in line with the formula described in the methods. The patients were divided into the high-expression and the low-expression groups (47
Characteristic | ACTA2 expression | ||||
---|---|---|---|---|---|
High expression (n = 47) | Low expression (n = 27) | ||||
Gender | Male | 21 (44.6%) | 16 (59.2%) | 0.334 | |
Female | 26 (55.4%) | 11 (40.8%) | |||
Age | ≤14 years | 8 (17.0%) | 7 (25.9%) | 0.261 | |
>14 years | 39 (83.0%) | 20 (74.1%) | |||
Tumor size | ≤10 cm | 35 (74.5%) | 15 (55.6%) | 0.368 | |
>10 cm | 12 (25.5%) | 12 (44.4%) | |||
Type of pathology | Conventional | 44 (93.6%) | 21 (77.7%) | 0.102 | |
Not conventional | 3 (6.4%) | 6 (22.3%) | |||
Tumor site | Femur | 29 (61.7%) | 15 (55.6%) | 0.741 | |
Tibia | 10 (21.3%) | 5 (18.5%) | |||
Fibula |
1 (2.1%) |
1 (3.7%) |
|||
Tumor necrosis rate | ≤90% | 28 (59.6%) | 16 (59.3%) | 0.556 | |
>90% | 9 (19.1%) | 3 (11.1%) | |||
Not reported | 10 (21.3%) | 8 (29.6%) | |||
Lung metastases | Yes | 30 (63.8%) | 14 (51.9%) | 0.445 | |
No | 17 (36.2%) | 13 (48.1%) | |||
Relapse | Yes |
37 (78.7%) |
17 (63.0%) |
0.231 | |
Bone metastases | Yes |
38 (80.9%) |
23 (85.2%) |
0.747 | |
Other metastases | Yes |
45 (95.7%) |
24 (88.9) |
0.516 |
With multiple immunofluorescence staining, ACTA2 was labeled with pink fluorescence. The expression of ACTA2 in osteosarcoma tissue was clustered around the vascular wall and distributed in a ring shape. The specific distribution of ACTA2 suggests that there may be more normalized blood vessels in tumor tissue of patients with high expression of ACTA2. Tumor angiogenesis is significantly related to tumor proliferation, metastasis, and drug sensitivity, which may be related to the prognosis of patients (
The follow-up period of this study ranged from 36 to 60 months until January 2022, with 23 patients terminated due to death and the remaining 51 patients alive at the end of the follow-up period. Thirty patients had definite pulmonary metastases; 20 patients had local recurrence; 10 patients had bone metastases; and five patients had metastases other sites during the course of the disease. By the follow-up cutoff, the shortest survival was 12 months, and the longest survival was 60 months. The results of the Kaplan–Meier survival statistical analysis showed that among all 74 patients, the overall survival rate (OS) was significantly higher in the ACTA2 high-expression group than in the low-expression group (
Osteosarcoma is still a persistent disease that has not been overcome. The etiology is still unknown; the genomic heterogeneity is high; and no driver gene has yet been found. In addition, the five-year survival rate for patients with associated metastases is still ≤25% [
By analyzing the prognostic relationships of osteosarcoma-related genes in the TCGA database, including ACTA2, CCL2, TGFBI, VEGFA, PDGFB, PDGFC, COL1A1, COL14A1, CXCL12, CXCL14, CSPP1, LUM, DES, MYL9, and SFRP2, we found that the prognosis of patients with high ACTA2 expression was significantly higher than that of patients with low ACTA2 expression. Other molecules had no significant effect on patients’ prognosis. Li et al. [
Nontumor cells in the tumor microenvironment have an important impact on both tumor progression and prognosis. Among them, cancer-associated fibroblasts (CAFs) play an important role in the tumor microenvironment and are mainly characterized by the expression of actin alpha 2 (ACTA2). Communication between osteosarcoma cells and the surrounding tumor microenvironment (TME) is necessary for tumor growth and subsequent metastasis. The main influential components of the TME are tumor-associated macrophages (TAMs), which are immune cells involved in the inflammatory response and tissue homeostasis. Based on the analysis of TCGA database genes, we previously applied single-cell techniques to analyze the heterogeneity of osteosarcoma tissues and the immune microenvironment [
ACTA2 is an actin also known as α–smooth muscle actin (α-SMA). It is involved in the contraction of smooth muscle and is also a cytoskeletal protein expressed in myoepithelial cells [
ACTA2 is a class of microfilament-forming globular multifunctional proteins that are widely found in eukaryotic cells and function by interconnecting with ACTA2 and membranes through the adhesive linkage of calmodulin. ACTA2 is an essential component of the eukaryotic cytoskeleton and plays a role in rapid polymerization and depolymerization to achieve signal transduction by providing cellular support for the cells [
Conceptualization: ZW. Data curation: LT, HY, ZY, HJ, WG, YZ, JL, and ZW. Formal analysis: LT. Investigation: LT, HY, ZY, HJ, WG, YZ, JL, and ZW. Methodology: LT, ZW. Project administration: ZW. Resources: ZW. Supervision: ZW. Visualization: LT. Roles/writing-original draft: LT. Writing-review and editing: LT, HY, ZY, HJ, WG, YZ, JL, and ZW. All authors contributed to the article and approved the submitted version.
The trial was approved by the institutional review board of Shanghai Sixth People’s Hospital (China). Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin. The ethical approval code number is 2021-161.
The study was funded by
The authors declare that they have no conflicts of interest to report regarding the present study.