Differentially Expressed Genes Reveal the Biomarkers and Molecular Mechanism of Osteonecrosis

Ma, Huanzhi; Zhang, Wei; Shi, Jun

doi:https://doi.org/10.1155/2022/8684137

Journal of Healthcare Engineering

On this page

Abstract Introduction Materials and Methods Results Discussion Data Availability Conflicts of Interest Acknowledgments References Copyright Related Articles

Special Issue

Developments in Optimization Algorithms for Smart Healthcare

View this Special Issue

Research Article | Open Access

Volume 2022 | Article ID 8684137 | https://doi.org/10.1155/2022/8684137

Differentially Expressed Genes Reveal the Biomarkers and Molecular Mechanism of Osteonecrosis

Huanzhi Ma,¹Wei Zhang,¹and Jun Shi¹

Academic Editor: Kalidoss Rajakani

Received03 Nov 2021

Revised06 Dec 2021

Accepted07 Dec 2021

Published07 Jan 2022

Abstract

Osteonecrosis is one of the most refractory orthopedic diseases, which seriously threatens the health of old patients. High-throughput sequencing (HTS) and microarray analysis have confirmed as an effective way for investigating the pathological mechanism of disease. In this study, GSE7716, GSE74089, and GSE123568 were obtained from Gene Expression Omnibus (GEO) database and used to identify differentially expressed genes (DEGs) by R language. Subsequently, the DEGs were analyzed with Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment. Moreover, the protein-protein interaction (PPI) network of DEGs was analyzed by STRING database and Cytoscape. The results showed that 318 downregulated genes and 58 upregulated genes were observed in GSE7116; 690 downregulated genes and 1148 upregulated genes were screened from 34183 genes in GSE74089. The DEGs involved in progression of osteonecrosis involved inflammation, immunological rejection, and bacterial infection-related pathways. The GO enrichment showed that osteonecrosis was related with extracellular matrix, external encapsulating structure organization, skeletal system development, immune response activity, cell apoptosis, mononuclear cell differentiation, and serine/threonine kinase activity. Moreover, PPI network showed that the progression of osteonecrosis of the femoral head was related with CCND1, CDH1, ESR1, SPP1, LOX, JUN, ITGA, ABL1, and VEGF, and osteonecrosis of the jaw is related with ACTB, CXCR4, PTPRC, IL1B, CXCL8, TNF, JUN, PTGS2, FOS, and RHOA. In conclusion, this study identified the hub factors and pathways which might play important roles in progression of osteonecrosis and could be used as potential biomarkers for diagnosis and treatment of osteonecrosis.

1. Introduction

Osteonecrosis, characterized by eventual collapse of the femoral head, is a common bone-related disease with high incidence, especially in old people [1, 2]. The osteonecrosis of the femoral head and jaw is common typical disease of osteonecrosis, and the surgical intervention has always been used for relieving the symptom of the patients [3]. However, the pathogenic mechanism of osteonecrosis is very complex, and timely accurate diagnosis and conservative therapeutic methods may be more suitable than surgery for early patients [4]. Thus, due to limitation of special drugs, even with current intervention strategy, the outcome of the patients with osteonecrosis remains unsatisfactory. Recently, targeted drugs have been recognized as an effective way for the treatment of difficult and miscellaneous diseases. Increasing studies also confirmed that targeted therapy may also be a promising way for osteonecrosis [5].

High-throughput sequencing (HTS) is one of the most important technologies in biological research in recent years [6]. Microarray analysis based on the datasets of HTS has been confirmed as a promising and efficient strategy for the investigation of the diagnosis methods and pathogenic mechanisms of disease [7, 8]. Increasing studies have focused on illustrating the potential biomarkers and molecular mechanisms via delving global gene profiling of patients with multiple diseases ranging from epidemic to nonepidemic [9]. For osteonecrosis, several studies have also revealed the difference of the patients and normal subjects [10]. With the current biological technology, the gene profiling of the patients could be easily obtained and analyzed, and then the related strategies for the diagnosis and treatment could be developed on the basis of the body indicators of the normal persons [11].

This study attempted to (i) investigate the observed difference of expressed genes of normal and osteonecrosis patients via excavating the biological information of the chips in GEO database; (ii) identify critical factors of osteonecrosis via analyzing the gene network; (iii) and reveal the related pathological processes and signaling pathways by enrichment analysis.

2. Materials and Methods

2.1. Data Resource

The expression microarray datasets of osteonecrosis were obtained from the GEO database by R language. The microarray data of GSE74089 (GPL13497), GSE7116 (GPL570), and GSE123568 (GPL15207) were collected from GEO database in National Center for Biotechnology information (NCBI).

2.2. Differential Genes Analysis

For data analysis, the package of GEOquery and limma was preinstalled in R language, and then the datasets were downloaded from GEO database. For data analysis, the dataset was corrected and normalized to obtain the standardized data with high quality. After that, the difference in mRNA level of the dataset was obtained by the lmFit and eBayes function in limma package. Finally, the genes with significantly different expression () were obtained. The corresponding value of gene symbols following a t-test was defined as the adjusted p value; log2 fold change >2 and were considered to be the cut-off criteria for differentially expressed genes (DEGs).

2.3. Enrichment Analysis for Related Function and Pathways of Genes

The gene enrichment analysis of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) was performed with cluster Profiler Package of Bioconductor and David Online Database (https://david.ncifcrf.gov/). The genes in pathological group exhibiting significant difference with control () were selected for pathway enrichment. P < 0.05 was set as the cut-off for enrichment analysis.

2.4. Protein-Protein Interaction (PPI) Network Analysis

The protein-protein interaction (PPI) network analysis of the DEGs was performed for finding key factors of the diseases. The DEGs of the datasets were analyzed with Search Tool for the Retrieval of Interacting Genes/Proteins (STRING; http://string-db.org) to obtain the data of PPI. After that, the data of the PPI were further analyzed and displayed with Cytoscape (Version 3.7.1).

3. Results

3.1. DEGs Identification in Osteonecrosis Samples

To analyze the expression difference of the normal subjects and the patients with osteonecrosis, 318 downregulated genes and 58 upregulated genes were observed in GSE7116; 690 downregulated genes and 1148 upregulated genes were screened from 34183 genes in GSE74089; 243 downregulated genes and 36 upregulated genes were screened from GSE123568 genes (Figure 1). Moreover, the top 30 genes with high differential expression in three datasets were structured with heat map (Figure 2).

(a)

(b)

(c)

(a)

(b)

(c)

3.2. KEGG Enrichment Analysis

To explore the regulatory pathways of the aberrant genes, the clusterProfiler, org.Hs.eg.db, and topGO were preloaded in R studio, and the KEGG enrichment was performed for analyzing the relative pathways of the DEGs of three datasets. The results showed that 47 relative pathways were obtained according to the DEGs of GSE74089, 45 pathways were obtained according to the DEGs of GSE7116, and 16 pathways were obtained according to the repetitive genes of GSE74089, GSE7116, or GSE123568 (Figures 3(a) and 3(b)). Moreover, it was also found that the pathways involved inflammatory, immune response, and infection-related pathways (Figures 3(a) and 3(b)). For repetitive genes of GSE74089, GSE7116, or GSE123568, the enriched pathways also involved the inflammation, immunological rejection, and bacterial infection (Figure 3(c)). Those proofs suggested that the progression of osteonecrosis involved inflammation, immunological rejection, and bacterial infection-related pathways.

(a)

(b)

(c)

3.3. GO Enrichment Analysis

To investigate the biological function of DEGs in osteonecrosis development, the DEGs of the datasets were analyzed with GO enrichment. The results showed that 164 molecular functions were found in GSE7116 and 23 molecular functions in GSE74089. Moreover, the DEGs in intersection of three datasets were also analyzed by GO enrichment, and 52 molecular functions were also found to involve the repeat genes in GSE74089 and GSE7116 (Figure 4). The GO enrichment revealed that osteonecrosis development was related with the change in extracellular matrix, external encapsulating structure organization, skeletal system development, and cellular development in GSE7116 and immune response activity, cell apoptosis, mononuclear cell differentiation, and serine/threonine kinase activity (Figure 4).

(a)

(b)

3.4. PPI Network Analysis

To reveal the interaction of the proteins in development of osteonecrosis, the DEGs of the GSE7116 and GSE74089 were analyzed with STRING databases and their interactive networks were structured with Cytoscape. According to the results, the networks with top three scores of GSE74089 were selected for next analysis including cluster 1 with 22 nodes and 412 edges, culster 2 with 68 nodes and 548 edges, and cluster 3 with 61 nodes and 490 edges. The networks with top three scores of GSE7116 were obtained, including cluster 1 with 24 nodes and 178 edges, cluster 2 with 14 nodes and 82 edges, and cluster 3 with 11 nodes and 50 edges. For GSE74089, CCND1, CDH1, ESR1, SPP1, LOX, JUN, ITGA, ABL1, and VEGF were hub genes which had multiple connection to other genes (Figures 5(a)–5(c)). For GSE7116, ACTB, CXCR4, PTPRC, IL1B, CXCL8, TNF, JUN, PTGS2, FOS, and RHOA were hub genes (Figures 5(d)–5(f)). Moreover, the common DEGs in GSE74089 and GSE7116 were also analyzed by PPI network, and the study showed that PTPRC and JUN were selected as hub nodes (Figure 5(g)).

(a)

(b)

(c)

(d)

(e)

(f)

(g)

4. Discussion

Osteonecrosis has high incidence in middle-aged and elderly age people, which seriously impacts the body functions of the patients [12, 13]. Although some studies have investigated the pathological mechanism and therapeutic methods in recent years, the systemic mechanism remains unclear. Therefore, systemic and intensive researches for key virulence factors of osteonecrosis are still necessary. In this study, the biological information of pathological samples of the subjects was obtained from the GEO database. For exploring pathological mechanism of osteonecrosis, the gene profiling datasets of the patients with osteonecrosis of the femoral head and the femoral head jaw including GSE74089, GSE123568, and GSE7116 were used to observe the expression difference of the subjects with osteonecrosis and the subjects without osteonecrosis. Moreover, a broad range of analysis strategies was performed for delving the related molecular mechanism of osteonecrosis, such as KEGG enrichment, GO enrichment, and PPI network analysis.

Osteonecrosis of the femoral head and osteonecrosis of the jaw may have certain similarity. In this study, the DEGs of GSE74089 and GSE7116 were analyzed and compared, and all the patients exhibited significant difference in some genes compared with those of the subjects without osteonecrosis, and GSE123568 including the patients steroid-induced osteonecrosis of the femoral head was also used to compare with GSE74089 and GSE7116 for revealing the key factors of the disease. The gene in the intersection of the datasets suggested that osteonecrosis of the femoral head and osteonecrosis of the jaw might involve some common core factors.

Pathway enrichment analysis is an effective strategy for biological researching, which is widely used for revealing the molecular mechanism of disease [14]. In this study, the DEGs of common factors in GSE74089 and GSE7116 were enriched in KEGG pathway and found that the development of osteonecrosis involved multiple pathway. The pathological mechanism of osteonecrosis of the femoral head and osteonecrosis of the jaw was related with the activity change of multiple pathways. Moreover, the DEGs of the patients with osteonecrosis of the femoral head or osteonecrosis of the jaw exhibited high connection with the pathways involved inflammation immune response, infection, and cancer-related pathways. Inflammation has been contributed as a major reason causing the progression of osteonecrosis. The study by Wu et al. suggested that inhibiting the TNF-α expression and improving the inflammatory level could effectively inhibit the development of the osteonecrosis [15]. Moreover, high levels of inflammatory factors such as IL-6 and IL-21 could also aggravate the symptom of osteonecrosis [16, 17]. The dysfunction of the immune system is the high incidence event for the patients with osteonecrosis, while the immune system sustains the normal progression of bone remodeling and tissue repair [18]. The tissues of the patients with low immune level may be susceptible to invasion by bacteria and viruses, and aberrant cells may not be timely removed by immune cells [19, 20]. The DEGs of the datasets were also analyzed with GO function enrichment. In this study, it was found that osteonecrosis development involved with the functional change of the tissues including extracellular matrix, external encapsulating structure organization, skeletal system development, and cellular development in GSE7116 and immune response activity, cell apoptosis, mononuclear cell differentiation, and serine/threonine kinase activity. Moreover, the modular functions of immune response activity were also observed in clinical data of the patients with osteonecrosis of the femoral head or osteonecrosis of the jaw.

PPI network analysis has been also used for tracking the interactions of the genes and then directly illustrated the molecular mechanism [21]. In this study, the PPI network of DEGs in the datasets was also analyzed. For osteonecrosis of the femoral head, aberrant expression of CCND1 has been also found in the patients with osteonecrosis of the femoral head induced by systemic lupus erythematosus, and ESR1 may involve the osteonecrosis development of the children with acute lymphoblastic leukemia [22, 23]. Moreover, abnormal level of VEGF was also found in gene profiling of the patients [24]. VEGF serves as key role for vascularization, and ischemia could also enhance osteonecrosis. For osteonecrosis of the jaw, it was found that TNF-α, ACTB, IL1B, CXCL8, PTGS2, and RHOA play hub roles in PPI network. Tumor Necrosis Factor-α (TNF-α) serves as an important inflammatory factor which could inhibit the malignant progressions of tumors and promote the inflammatory reaction of the patients [25]. Several studies have indicated that increased TNF-α involves the progression of osteonecrosis of the femoral head the femoral head. The study by Wu et al. has found that TNF-α is significantly upregulated in necrotic zone of the rats, and TNF-a-mediated alteration of M1/M2 macrophage polarization contributed to the pathogenesis of steroid-induced osteonecrosis [25]. The stability of ACTB is significantly downregulated in bone marrow-derived mesenchymal stem cells of the osteonecrosis patients [26]. Increased IL1B serves as close relationship with the progression of jaw osteonecrosis of mice, and several studies have indicated that IL1B is abnormally expressed in the serum [27]. The study by Pavlova et al. has identified CXCL8 as a potential key factor for osteonecrosis induced by Gaucher disease [28]. Chen et al. [29] has indicated that the level of PTGS2 is increasingly upregulated with the progression of osteonecrosis of the femoral head. Chen et al. have also proved that RHOA may play a hub role in progression of osteonecrosis via PPI network analysis [30]. JUN is a transcription factor, and several studies have found that c-JUN dysfunction could promote the femoral head necrosis of the rats [31]. The PPI network showed that C-JUN played a hub role in the DEGs of both patients with osteonecrosis of the femoral head or osteonecrosis of the jaw.

In conclusion, this study identified the hub factors and pathways which might play important roles in progression of osteonecrosis and could be used as potential biomarkers for diagnosis and treatment of osteonecrosis.

Data Availability

The data used to support the findings of this study are available on reasonable request from the corresponding author.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This research was supported by Natural Science Foundation of Shandong Province (ZR2020MH085).

References

O. O. Adesina and L. D. Neumayr, “Osteonecrosis in sickle cell disease: an update on risk factors, diagnosis, and management,” Hematology, vol. 2019, no. 1, pp. 351–358, 2019 Dec 6.
View at: Publisher Site | Google Scholar
J. Chang, A. E. Hakam, and L. K. McCauley, “Current understanding of the pathophysiology of osteonecrosis of the jaw,” Current Osteoporosis Reports, vol. 16, no. 5, pp. 584–595, 2018 Oct.
View at: Publisher Site | Google Scholar
C. Chang, A. Greenspan, and M. E. Gershwin, “The pathogenesis, diagnosis and clinical manifestations of steroid-induced osteonecrosis,” Journal of Autoimmunity, vol. 110, Article ID 102460, 2020 Jun.
View at: Publisher Site | Google Scholar
A. Derache, J. Chauvelot, A. Bellot et al., “Osteonecrosis of the jaw induced by arsenic trioxide therapy in a leukemia patient: a rare case report and literature review,” Oral Oncology, vol. 109, Article ID 104725, 2020 Oct.
View at: Publisher Site | Google Scholar
V. Fusco, D. Santini, G. Armento, G. Tonini, and G. Campisi, “Osteonecrosis of jaw beyond antiresorptive (bone-targeted) agents: new horizons in oncology,” Expert Opinion on Drug Safety, vol. 15, no. 7, pp. 925–935, 2016 Jul.
View at: Publisher Site | Google Scholar
T. Neumann, V. A. Herzog, M. Muhar et al., “Quantification of experimentally induced nucleotide conversions in high-throughput sequencing datasets,” BMC Bioinformatics, vol. 20, no. 1, p. 258, 2019 May 20.
View at: Publisher Site | Google Scholar
N. Komarova, D. Barkova, and A. Kuznetsov, “Implementation of high-throughput sequencing (HTS) in aptamer selection technology,” International Journal of Molecular Sciences, vol. 21, no. 22, Article ID 8774, 2020 Nov 20.
View at: Publisher Site | Google Scholar
D. Brabbing-Goldstein, A. Reches, R. Svirsky, A. Bar-Shira, and Y. Yaron, “Dilemmas in genetic counseling for low-penetrance neuro-susceptibility loci detected on prenatal chromosomal microarray analysis,” American Journal of Obstetrics and Gynecology, vol. 218, no. 2, pp. 247.e1–247.e12, 2018 Feb.
View at: Publisher Site | Google Scholar
I. Monier, A. Receveur, V. Houfflin-Debarge et al., “Should prenatal chromosomal microarray analysis be offered for isolated fetal growth restriction? a French multicenter study,” American Journal of Obstetrics and Gynecology, vol. 225, no. 21, pp. 00603–00607, 2021 May 29.
View at: Google Scholar
K.-T. Peng, K.-C. Huang, T.-W. Huang et al., “Single nucleotide polymorphisms other than factor V Leiden are associated with coagulopathy and osteonecrosis of the femoral head in Chinese patients,” PLoS One, vol. 9, no. 8, Article ID e104461, 2014 Aug 13.
View at: Publisher Site | Google Scholar
M. Ma, P. Li, L. Liu et al., “Integrating transcriptome-wide association study and mRNA expression profiling identifies novel genes associated with osteonecrosis of the femoral head,” Frontiers in Genetics, vol. 12, Article ID 663080, 2021 Jun 7.
View at: Publisher Site | Google Scholar
C. Ma, G. Andre, D. Edwards, and H. K. W. Kim, “A rat model of ischemic osteonecrosis for investigating local therapeutics using biomaterials,” Acta Biomaterialia, vol. 132, pp. 260–271, 2021 Sep 15.
View at: Publisher Site | Google Scholar
B. M. DeFeo, S. C. Kaste, Z. Li et al., “Long-term functional outcomes among childhood survivors of cancer who have a history of osteonecrosis,” Physical Therapy, vol. 100, no. 3, pp. 509–522, 2020 Mar 10.
View at: Publisher Site | Google Scholar
M. V. Slizen and O. V. Galzitskaya, “Comparative analysis of proteomes of a number of nosocomial pathogens by KEGG modules and KEGG pathways,” International Journal of Molecular Sciences, vol. 21, no. 21, p. 7839, 2020 Oct 22.
View at: Publisher Site | Google Scholar
R. Sacco, S. Shah, R. Leeson et al., “Osteonecrosis and osteomyelitis of the jaw associated with tumour necrosis factor-alpha (TNF-α) inhibitors: a systematic review,” British Journal of Oral and Maxillofacial Surgery, vol. 58, no. 1, pp. 25–33, 2020 Jan.
View at: Publisher Site | Google Scholar
M. Yousef, E. Ülgen, and O. Uğur Sezerman, “CogNet: classification of gene expression data based on ranked active-subnetwork-oriented KEGG pathway enrichment analysis,” PeerJ Computer Science, vol. 7, p. e336, 2021 Feb 22.
View at: Publisher Site | Google Scholar
Z. Zhou, C. Pan, N. Wang et al., “A high-fat diet aggravates osteonecrosis through a macrophage-derived IL-6 pathway,” International Immunology, vol. 31, no. 4, pp. 263–273, 2019 Mar 28.
View at: Publisher Site | Google Scholar
B. Chen, Y. Liu, and L. Cheng, “IL-21 enhances the degradation of cartilage through the JAK-STAT signaling pathway during osteonecrosis of femoral head cartilage,” Inflammation, vol. 41, no. 2, pp. 595–605, 2018 Mar.
View at: Publisher Site | Google Scholar
R. Elsayed, Z. Kurago, C. W. Cutler et al., “Role of dendritic cell‐mediated immune response in oral homeostasis: a new mechanism of osteonecrosis of the jaw,” The FASEB Journal, vol. 34, no. 2, pp. 2595–2608, 2020 Feb.
View at: Publisher Site | Google Scholar
P. Wang, X. Liu, J. Xu et al., “Deep learning for diagnosing osteonecrosis of the femoral head based on magnetic resonance imaging,” Computer Methods and Programs in Biomedicine, vol. 208, Article ID 106229, 2021 Sep.
View at: Publisher Site | Google Scholar
U. Ayub, I. Haider, and H. Naveed, “SAlign-a structure aware method for global PPI network alignment,” BMC Bioinformatics, vol. 21, no. 1, p. 500, 2020 Nov 4.
View at: Publisher Site | Google Scholar
P. Kang, Z. Wu, Y. Zhong et al., “A network pharmacology and molecular docking strategy to explore potential targets and mechanisms underlying the effect of curcumin on osteonecrosis of the femoral head in systemic lupus erythematosus,” BioMed Research International, vol. 2021, Article ID 5538643, 2021 Sep 13.
View at: Publisher Site | Google Scholar
D. French, L. H. Hamilton, L. A. Mattano et al., “A PAI-1 (SERPINE1) polymorphism predicts osteonecrosis in children with acute lymphoblastic leukemia: a report from the Children’s Oncology Group,” Blood, vol. 111, no. 9, pp. 4496–4499, 2008 May 1.
View at: Publisher Site | Google Scholar
Y. Luo, D. Li, X. Xie, and P. Kang, “Porous, lithium-doped calcium polyphosphate composite scaffolds containing vascular endothelial growth factor (VEGF)-loaded gelatin microspheres for treating glucocorticoid-induced osteonecrosis of the femoral head,” Biomedical Materials, vol. 14, no. 3, Article ID 035013, 2019 Apr 5.
View at: Publisher Site | Google Scholar
L. Yu, Y. Xu, H. Qu et al., “Decrease of MiR‐31 induced by TNF‐α inhibitor activates SATB2/RUNX2 pathway and promotes osteogenic differentiation in ethanol‐induced osteonecrosis,” Journal of Cellular Physiology, vol. 234, no. 4, pp. 4314–4326, 2019 Apr.
View at: Publisher Site | Google Scholar
X. Wu, X. Feng, Y. He et al., “administration exerts preventive effects via suppression of underlying inflammation and TNF-α-induced apoptosis in steroid-induced osteonecrosis,” Osteoporosis Internationa, vol. 27, no. 5, pp. 1827–1837.
View at: Publisher Site | Google Scholar
E. V. Pavlova, P. B. Deegan, J. Tindall et al., “Potential biomarkers of osteonecrosis in Gaucher disease,” Blood Cells, Molecules, and Diseases, vol. 46, no. 1, pp. 27–33, 2011 Jan 15.
View at: Publisher Site | Google Scholar
I. F. Domingos, D. A. Pereira-Martins, M. J. V. C. Sobreira et al., “High levels of proinflammatory cytokines IL-6 and IL-8 are associated with a poor clinical outcome in sickle cell anemia,” Annals of Hematology, vol. 99, no. 5, pp. 947–953, 2020 May.
View at: Publisher Site | Google Scholar
G. Chen, L. Zhong, Q. Wang et al., “The expression of chondrogenesis-related and arthritis-related genes in human ONFH cartilage with different Ficat stages,” PeerJ, vol. 7, Article ID e6306, 2019 Jan 16.
View at: Publisher Site | Google Scholar
D. Chen, G. Zhang, Y. Li et al., “Up-regulation of urinary exosomal hsa-microRNA-200b-3p and hsa-microRNA-206 in patients of steroid-induced osteonecrosis of femoral head,” American Journal of Tourism Research, vol. 13, no. 7, pp. 7574–7590, 2021 Jul 15.
View at: Google Scholar
P. Peng, Z. Nie, F. Sun, and H. Peng, “Glucocorticoids induce femoral head necrosis in rats through the ROS/JNK/c‐Jun pathway,” FEBS Open Bio, vol. 11, no. 1, pp. 312–321, 2021 Jan.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2022 Huanzhi Ma et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies

Views

478

Downloads

659

Citations