Identification of Four Genes and Biological Characteristics Associated with Acute Spinal Cord Injury in Rats Integrated Bioinformatics Analysis

570

Original Article Page 1 of 13

Identification of four genes and biological characteristics associated with acute spinal cord injury in rats integrated bioinformatics analysis

Qiang Li1,2,3#, Bo Li1,4#, Bo Tao1,2#, Chenxi Zhao1,2#, Baoyou Fan1,2, Qi Wang1,2,5, Chao Sun1,2, Huiquan Duan1,2, Yilin Pang1,2, Xuanhao Fu1,2, Shiqing Feng1,2

1Department of Orthopedics, Tianjin Medical University General Hospital, Tianjin, China; 2International Science and Technology Cooperation Base of Spinal Cord Injury, Tianjin Key Laboratory of Spine and Spinal Cord Injury, Tianjin, China; 3Department of Orthopedics, Shanxi Bethune Hospital, Shanxi Academy of Medical Sciences, Taiyuan, China; 4Department of Orthopedics, Sun Yat-sen Memorial Hospital of Sun Yat-sen University, Guangzhou, China; 5Department of Orthopedics, Tianjin Hospital of ITCWM Nankai Hospital, Tianjin, China Contributions: (I) Conception and design: S Feng, Q Li; (II) Administrative support: S Feng; (III) Provision of study materials or patients: B Li, B Tao, S Feng; (IV) Collection and assembly of data: Q Li, B Li, B Tao, C Zhao; (V) Data analysis and interpretation: Q Li, B Li; (VI) Manuscript writing: All authors; (VII) Final approval of manuscript: All authors. #These authors contributed equally to this work. Correspondence to: Shiqing Feng. Department of Orthopedics, Tianjin Medical University General Hospital, No. 154 Anshan Road, Heping District, Tianjin 300052, China. Email: [email protected].

Background: Spinal cord injury (SCI) is a serious condition that can cause physical disability and sensory dysfunction. Cytokines play an extremely important role in the acute phase of SCI. Clarifying the cytokine expression profile is of great importance. Methods: Cytokine array analysis was used to explore the changes in 67 different proteins at 0 hours, 2 hours, 1 day, 3 days, and 7 days after acute SCI in rats. The differentially expressed cytokines in the various periods were analyzed and compared. The biological processes related to the differentially expressed proteins were examined using Gene Ontology (GO) analysis. Results: Immediately after SCI (0 hours), only ciliary neurotrophic factor (CNTF) was slightly upregulated, while 23 other proteins were down-regulated. At 2 hours after SCI, there were 3 upregulated and 21 downregulated proteins. At 1 day after SCI, there were 5 upregulated and 6 downregulated proteins. At 3 days after SCI, there were 6 upregulated and 4 downregulated proteins. At 7 days after SCI, there were 4 upregulated and 9 downregulated proteins. Erythropoietin (EPO) and Fms related tyrosine kinase 3 ligand (Flt-3L) were downregulated at all time points. CD48 was decreased at 2 hours to 7 days after SCI. Monocyte chemotactic protein-1 (MCP-1) was the only protein that was upregulated at 2 hours to 7 days. The GO and pathway analyses revealed that the cytokine-related pathways, cell death, and proliferation might play a key role during secondary SCI. Conclusions: This study identified 3 downregulated proteins during SCI, that being EPO, Flt-3L, and CD48. MCP-1 was the only upregulated protein, and its expression was upregulated till day 7 following SCI. These 4 identified genes may be potential therapeutic targets for the treatment of SCI.

Keywords: Spinal cord injury (SCI); inflammation; erythropoietin; Fms related tyrosine kinase 3 ligand (Flt-3L); CD48; monocyte chemoattractant protein-1 (MCP-1)

Submitted Jan 11, 2021. Accepted for publication Mar 05, 2021. doi: 10.21037/atm-21-603 View this article at: http://dx.doi.org/10.21037/atm-21-603

© Annals of Translational Medicine. All rights reserved. Ann Transl Med 2021;9(7):570 | http://dx.doi.org/10.21037/atm-21-603 Page 2 of 13 Li et al. A cytokine array analysis in acute spinal cord injury

Introduction neuroprotective agent (16). Monocyte chemoattractant protein-1 (MCP-1) is a member of the chemotactic Spinal cord injury (SCI) is a traumatic disease that causes cytokine family that functions to recruit monocytes and exercise, sensory, or autonomic dysfunction, thereby directly modulate T-cell function (17). Fms related tyrosine kinase affecting the quality of life of patients and imposing a heavy 3 ligand (Flt-3L) causes dendritic cell and lymphocyte financial burden on society and families (1). It is therefore proliferation (18). CD48, a member of the signaling crucial to clarify the pathophysiological processes associated lymphocyte activation molecule family, participates in with SCI. After SCI, secondary injury such as inflammation, adhesion and activation of immune cells (19). At different oxidative stress, and apoptosis, can lead to even more time points after SCI, the cytokine expression trends are serious consequences. Ultimately, secondary injury induces different, indicating that different key cytokines appear at cell death, demyelination, and axonal degeneration at the different time points. Targeting key cytokines at precise epicenter of the injury and the surrounding regions (2,3). time points may exert maximum treatment efficacy. Thus, Inflammatory reactions mediated by cytokines and other clarifying the cytokine expression profiles at different factors plays an important role in the development of times following SCI will benefit the understanding of the secondary injury following SCI (4,5). Inflammatory factors underlying mechanisms and aid in the development of are important bridges between immune cells and other effective therapeutic strategies. cells and are important mediators in regulating the immune While previous studies have focused on several major response. These cytokines include pro-inflammatory factors, inflammatory factors, the expression of cytokines after SCI anti-inflammatory factors, chemokines, and neurotrophic has not been fully analyzed. Jiang et al. identified 5 hub factors. Studies have shown that the critical balance between genes with high-throughput sequencing after compression these processes plays an important role in the progress and SCI (20). In this study, we established a weight-drop SCI outcome of the neurodegenerative process (5-7). model, it has advantages of simple performance, good The role of inflammation in the process of SCI is controllability and repeatability compare with compression complex, with certain beneficial aspects as well, such SCI model. As the protein expression of genes is the results as the removal of cellular debris, a number of studies of transcription and translation. The protein expression have suggested that inflammatory responses spread the could be regulated by miRNA, lncRNA and circRNA at damage to surrounding tissue, induce apoptotic cell death, post-transcription level. Proteins is the main undertaker and impair spontaneous regeneration and functional of life activities. Thus, we chose to use protein array recovery (3). The infiltration of leukocytes and activation of technology to compare and analyze the expression of glial cells can aggravate tissue damage by releasing proteases, various cytokines, rather than mRNA. The protein array reactive oxygen intermediates, lysosomal enzymes, and technology is more conductive to reflect the occurrence inflammatory cytokines (8). Studies found that cytokines and the development of SCI. These results will provide new may be sensitive markers of disease. Cytokines network, sights into the underlying mechanisms and contribute to but not a single cytokine, responses to environmental the design of novel therapeutic methods. We present the hazards. High-throughput detection of cytokines provides following article in accordance with the ARRIVE reporting a global view of changes cytokine profiles and helps us checklist (available at http://dx.doi.org/10.21037/atm-21- identify many vital differentially expressed cytokines in 603). the development of SCI. Inflammatory factors include those that are pro-inflammatory and those that are anti- inflammatory. Pro-inflammatory factors such as interleukin Methods (IL)-1β, IL-6, and tumor necrosis factor (TNF)-α play a Ethics statement and animal grouping vital role in the secondary injury of SCI, and blockers or knockout of related genes improve spinal cord function A total of thirty 10-week-old female Wistar rats (250±25 g) (9-13). Studies have also shown that IL-6 facilitates spinal were purchased from the Tianjin Institution of Radiation cord repair (14). Anti-inflammatory factors, such as Medicine. The rats were housed under continuous humidity granulocyte-macrophage colony-stimulating factor (GM- conditions in 12-hour day/night cycles. Rats were randomly CSF), IL-10, and IL-13, promote the recovery of nerve separated into two groups, with 5 rats in the sham group and function (5,15). Erythropoietin (EPO), the erythropoietic 25 rats in the injury group. Spinal cord tissue was collected hormone is a leading therapy in neonatology as a immediately after SCI (0 hour) and at 4 other time points,

© Annals of Translational Medicine. All rights reserved. Ann Transl Med 2021;9(7):570 | http://dx.doi.org/10.21037/atm-21-603 Annals of Translational Medicine, Vol 9, No 7 April 2021 Page 3 of 13 namely, 2 hours, 1 day, 3 days, and 7 days (5 rats/time point). expressed proteins (DEPs) were investigated using GO All experimental procedures were approved by the Ethics analysis, the Fisher test, and cluster Profiler data package Committee of Tianjin Institute of Radiation Medicine from R/Bioconductor. When more than 5 differentially (approval number: IRM-DWLL-2019039) and were expressed proteins were identified in a particular GO term performed according to the National Institutes of Health in and P<0.05, the results were considered significant. the Guide for the Care and Use of Laboratory Animals (NIH Publications no. 85-23, revised 1996). Statistical analysis

Experimental data were analyzed using SPSS 20.0 software Establishing the SCI model and expressed as the mean ± standard error (SD). The The SCI model was established using the New York statistical differences in protein expression between two University Impactor device (NYU, New York, NY, USA) groups were evaluated by Student’s unpaired t-test. For as described previously (21). Briefly, rats were anesthetized more than two group, statistical significance was determined with chloral hydrate (4%, 3 mL/kg) and the spinal cord with one-way analysis of variance (ANOVA) test, followed was exposed after T10 laminectomy. The spinal cord was by Dunnett’s LSD post hoc testing to calculate the confused with a 10 g ×25 mm free dropped node. When the longitudinal differences between groups. A value of P<0.05 spinal cord was injured by the node, the tail was twisted and was considered statistically significant. the hindlimbs were shaken immediately. The incision was then sutured, and the rats were kept on a warm pad until they regained consciousness. The spinal cord samples were Results extracted from 5 mm around the epicenter of damage at the Hierarchical clustering was used to analyze the profiles of indicated times. the differentially expressed proteins at different time points after SCI

Protein array detection To identify the differentially expressed proteins at 0 hour, The proteins in the spinal cord samples were extracted and 2 hours, 1 day, 3 days, and 7 days after SCI, hierarchical detected using the Rat Cytokine Array 67 (Raybiotech, clustering was conducted. As showed in Figure 1A, there Norcross, GA) according to the manufacturer’s instructions. were more down-regulated proteins at the 0-hour and Briefly, the proteins in spinal cord samples were extracted 2-hour time points, and more up-regulated proteins with cell lysis buffer and a protease inhibitor cocktail at 1, 3, and 7 days after SCI. These 67 proteins were (Raybiotech, Norcross, GA). After determining the divided into 3 clusters, namely a high expression group protein concertation of the samples with the BCA protein (>1,000 pg/mL), a moderate expression group (expression assay kit (Raybiotech, Norcross, GA), the samples were levels between 100 and 1,000 pg/mL), and a low expression incubated with protein array glass slides. A 100 μL sample group (≤100 pg/mL). Among the 67 proteins, at all (500 μg/mL) was added to each cell of the glass slide and timepoints low expression proteins were the most abundant, incubated at 4 overnight. The slides were then washed moderate expression proteins ranked the second, while high with Thermo Scientific Well Wash Versa (Thermo expression proteins were the least abundant. At the 0-and ℃ Scientific, USA) and incubated with a biotinylated antibody 2-hour time points, the number of proteins in the moderate cocktail and Cy3 equivalent Dye-Streptavidin. Thereafter, cluster decreased and the number of proteins in the low the signals were visualized using a laser scanner. Serial expression cluster increased (Figure 1B). The proteins that dilutions of a predetermined calibration standard mix were sustained its expression in the same cluster at all time points used to generate a standard curve for each cytokine. The were shown in Figure 1C. The moderate expression cluster cytokine concentrations in the samples were calculated had more proteins than the other two clusters. according to the standard curve.

The protein expression patterns following SCI Gene Ontology (GO) analyses A total of 6 different expression patterns were observed The biological processes related to the differentially for the 67 differentially expressed proteins and these have

© Annals of Translational Medicine. All rights reserved. Ann Transl Med 2021;9(7):570 | http://dx.doi.org/10.21037/atm-21-603 Page 4 of 13 Li et al. A cytokine array analysis in acute spinal cord injury

A Sham 0 h 2 h 1 d 3 d 7 d VEGF TWEAK R IL-1a RAGE 50 High IL-2 TIMP-1 B MCP-1 Middle CINC-2 CINC-3 Low PDGF-AA 40 FGF-BP IL-1b Galectin-3 Notch-2 Cinc-1 30 Neuropilin-1 JAM-A GFR alpha-1 Notch-1 Gas 1 20 RANTES IL-1 ra iIL-22 IL-13 ICAM=-1 10 B7-2 number (n) Protein Nope IL-10 IL-2 Ra Adlponectin 0 IL-4 EphA5 Sham 0 h 2 h 1 d 3 d 7 d IL-17F IL-1 R6 IFNg C GM-CSF Gp130 Gene symbols INFa TCK-1 TIMP-2 Fit-3L High CNTF gp130 CTACK IL-1 R6 Galectin-1 TCK-1 Prolactin R L-Selectin Fractalkine TIM-1 CNTF Neuropilin-2 LIX TIMP-2 Flt-3L IL-1a B7-2 P-Cadherin LIX 12 b-NGF 11 b-NGF VEGF SCF RANTES IL-2 Ra CINC-3 SCF MIP-1a 10 Eotaxin 9.49 IL-1b IL-7 IL-3 GM-CSF L-Selectin TIM-1 ACTIVIN a 8.40 Orolactin 7.31 Middle IL-13 IL-4 RAGE CD48 CINC-1 Prolactin CTACK 6.23 P-Cadherin B7-1 5.14 TNFa IFNg IL-2 Eotaxin MIP-1a IL-17F 4-1BB 4.05 IL-7 2.96 TREM-1 Erythropoietin 1.88 IL-3 0.79 CD48 −0.30 HGF Galectn-1 −1.39 Decorin −2.47 Decorin Activin A Neuropilin-1 Neuropilin-2 B7-1 PDGF-AA Low Adiponectin Fractalkine Nope Galectin-3 Notch-1

Figure 1 The profiles of the differentially expressed proteins at different time points following spinal cord injury. (A) The heat-map shows the upregulated genes (red) and downregulated genes (blue) between the sham and injury groups at 0 hours, 2 hours, 1 day, 3 days, and 7 days after SCI (n=5 in the injury group at all time points except day 3; n=4 in the injury group at day 3; n=11 in the sham group). (B) The distribution of proteins in the high, moderate, and low expression clusters. (C) The proteins within the same cluster with sustained expression throughout all time points. SCI, spinal cord injury; CNTF, ciliary neurotrophic factor; gp130, Glycoprotein 130; CTACK, human chemokine (C-C Motif) ligand 27; TCK-1, thymus chemokine-1; LIX, LPS-induced CXC chemokine; TIMP-2, tissue inhibitor of metalloproteinases-2; F1t-3L, Fms related tyrosine kinase 3 ligand; IL-1a, interleukin-1Alpha; B7-2, cluster differentiation 86; P-Cadherin, placental cadherin; b-NGF, beta-nerve growth factor; VEGF, vascular endothelial growth factor; SCF, Stem cell factor; RANTES, regulated upon activation normal T cell expressed and secreted factor; IL-2 Ra, interleukin-2 receptor; CINC-3, cytokine induced neutrophil chemoattractant-3; IL-1b, interleukin-1 beta; IL-7, interleukin-7; IL-3, interleukin-3; GM-CSF, Granulocyte-macrophage colony stimulating factor; L-Selectin, leucocyte-selectin; TIMP-1, tissue inhibitor of metalloproteinases-1; IL-13, interleukin-13; IL-4, interleukin-4; RAGE, receptor for advanced glycosylation end products; CD48, cluster differentiation 48; CINC-1, cytokine induced neutrophil chemoattractant-1; TNF-α, tumor necrosis factor-α; IFNg, Interferon Gamma; IL-2, interleukin-2; MIP-1a, macrophage inflammatory protein-1a; IL-17F, interleukin-17F; TREM-1, triggering receptor expressed on monocytes-1.

been denoted as follows: decrease (D pattern, observed The differentially expressed proteins at different time in 12 proteins, Figure 2), decrease-increase (D-I pattern, points after SCI observed in 19 proteins, Figure 3), decrease-increase- To better demonstrate the differentially expressed proteins decrease (D-I-D pattern, observed in 24 proteins, at different time points, volcano-plot analyses were Figure 4), increase (In pattern, observed in 2 proteins, conducted. The proteins were taken into consideration Figure 5, upper row), increase-decrease (In-D pattern, when the fold change (FC) was over 1.2 or less than 0.83. observed in 8 proteins, Figure 6). and irregularity (Ir At 0 hours after SCI, only ciliary neurotrophic factor pattern, observed in 3 proteins, Figure 5, lower row). (CNTF) was slightly up-regulated, while 23 proteins were Among the 6 patterns, the D-I-D, D-I, and D patterns were down-regulated, with EPO being the most downregulated. the most abundant, indicating that decreased expression At 2 hours after SCI, there were 3 upregulated and 21 was the main trend for most of the differentially expressed downregulated proteins. MCP-1 was the most upregulated proteins. and EPO was the most down-regulated protein. At 1 day

500 B7-1 1000 Adiponectin 800 Activin A

400 800 600 300 600 400 200 400

100 200 200

Protein expression (pg/mL) expression Protein 0 Protein expression (pg/mL) expression Protein 0 (pg/mL) expression Protein 0 Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d 1500 4-1BB 1.0 IL-4 40 P-Cadherin

0.8 30 1000 0.6 20 0.4 500 10 0.2 Protein expression (pg/mL) expression Protein 0 (pg/mL) expression Protein 0.0 (pg/mL) expression Protein 0 Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d

600 Erythropoietin 200 EphA5 40 Eotaxin

150 30 400

100 20 200 50 10 Protein expression (pg/mL) expression Protein Protein expression (pg/mL) expression Protein 0 (pg/mL) expression Protein 0 0 Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d 30 SCF 3000 Gp130 200 IL-10

150 20 2000 100

10 1000 50 Protein expression (pg/mL) expression Protein (pg/mL) expression Protein Protein expression (pg/mL) expression Protein 0 0 0 Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d

Figure 2 The differentially expressed proteins with a decreased pattern of expression. Protein expression analysis revealed 12 proteins with a decreased expression trend in the injury group at 0 hours, 2 hours, 1 day, 3 days, and 7 days after SCI compared to the sham group. SCI, spinal cord injury; B7-1, cluster differentiation 80; 4-1BB, cluster differentiation137; IL-4, interleukin-4; SCF, stem cell factor.

after SCI, there were 5 upregulated and 6 downregulated the acute phase of SCI, Venn plots were used to show proteins. At 3 days after SCI, there were 6 upregulated the differentially expressed proteins at more than three and 4 downregulated proteins. At 7 days after SCI, there time points (Figure 8A,B). EPO and c were differentially were 4 upregulated and 9 downregulated proteins. Tissue expressed at all time points. CD48 and MCP-1 were inhibitor of metalloproteinases-1 (TIMP-1) and MCP-1 differentially expressed from 2 hours to 7 days. Nope and were the 2 most upregulated proteins and EPO was still Notch-1 were differentially expressed from 0 hours to the most downregulated protein at day 1, day 3, and day 7 1 day, while TIMP-1 was differentially expressed from 1 to (Figure 7A,B,C,D,E). 7 days. The expression profiles of EPO, Flt-3L, CD48, and MCP-1 at all time points are shown in Figure 8C,D. MCP-1 was the only protein that was upregulated at all time points, Analysis of the key proteins in the acute phase of SCI while Flt-3L, CD48, and EPO were downregulated at all To determine which key proteins play a pivotal role in time points. The latter 3 proteins had similar expression

© Annals of Translational Medicine. All rights reserved. Ann Transl Med 2021;9(7):570 | http://dx.doi.org/10.21037/atm-21-603 Page 6 of 13 Li et al. A cytokine array analysis in acute spinal cord injury

10 TIMP-2 80 IL-7 25 Fit-3L 200 HGF 8 60 20 150 6 15 40 100 4 10 20 50 2 5

Protein expression (pg/mL) expression Protein 0 0 Protein expression (pg/mL) expression Protein 0 Protein expression (pg/mL) expression Protein Protein expression (pg/mL) expression Protein 0 Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d 1000 Neuropilin-1 25 IL-2 Ra 150 JAM-A 800 Notch-1 800 20 600 100 600 15 400 400 10 50 200 200 5

0 (pg/mL) expression Protein Protein expression (pg/mL) expression Protein 0 0 0 Protein expression (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d 1500 IL-1 Ra 6 IL-3 80 B7-2 200 Gas 1

60 150 1000 4 40 100 500 2 20 50

0 0 Protein expression (pg/mL) expression Protein Protein expression (pg/mL) expression Protein 0 Protein expression (pg/mL) expression Protein 0 Sham 0h 2h 1d 3d 7d Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d Nope 800 2000 ICAM-1 80 CD48 16000 Galectin-1

600 1500 60 14000

400 1000 40 12000

200 500 20 10000 0

0 0 (pg/mL) expression Protein Protein expression (pg/mL) expression Protein 8000 Protein expression (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d 100 Prolactin 600 Decorin 8000 CTACK 80 400 6000 60 4000 40 200 20 2000

Protein expression (pg/mL) expression Protein 0 0 Protein expression (pg/mL) expression Protein 0 Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d

Figure 3 The differentially expressed proteins showing a decrease-increase pattern. Protein expression analysis identified 19 proteins with a decrease-increase expression trend in the injury group at 0 hours, 2 hours, 1 day, 3 days, and 7 days after SCI compared to the sham group. SCI, spinal cord injury; TIMP-2, tissue inhibitor of metalloproteinases-2; IL-7, interleukin-7; F1t-3L, Fms related tyrosine kinase 3 ligand; JAM-A, junction adhesion molecule-A; CTACK, human chemokine (C-C Motif) ligand 27. patterns. Flt-3L was expressed at a low level in normal rats GO molecular functions detected. “Cell surface” and but was not detectable in SCI rats at 0 hours to 3 days, “intrinsic component of plasma membrane” were the top 2 followed by a slightly increased expression at 7 days. EPO enriched GO cellular components detected. The top 10 GO was also the most downregulated protein over the course of biological processes related to the differentially expressed the experiment. proteins were “regulation of cell death”, “regulation of cell proliferation”, “defense response”, “regulation of programmed cell death”, “regulation of apoptotic process”, The gene ontology and pathway analysis of the “response to cytokine”, “cellular response to cytokine differentially expressed proteins stimulus”, “positive regulation of cell proliferation”, To better clarify the biological processes of the DEPs at “cytokine-mediated signaling pathway”, and “response to different time points after SCI, GO and pathway analyses wounding” (Figure 9). The top 4 pathways were “cytokine were performed. “Signaling-receptor binding” and signaling in immune system”, “cytokine-cytokine receptor “molecular function regulators” were the 2 most enriched interaction”, “extracellular matrix (ECM) and associated

IL-1b TIM-1 20 RANTES 10 4 150 TNFα 8 3 15 100 6 10 2 4 50 1 5 2 0 0 (pg/mL) expression Protein Protein expression (pg/mL) expression Protein (pg/mL) expression Protein

Protein expression (pg/mL) expression Protein 0 0 Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d 300 TWEAK R 50 TREM-1 450 PDGF-AA 1000 Galectin-3 40 800 200 400 30 600

20 400 100 350 10 200 0 Protein expression (pg/mL) expression Protein 0 (pg/mL) expression Protein 300

Protein expression (pg/mL) expression Protein 0 Sham 0 h 2 h 1 d 3 d 7 d (pg/mL) expression Protein Sham 0h 2h 1d 3d 7d Sham 0h 2h 1d 3d 7d Sham 0 h 2 h 1 d 3 d 7 d IL-6 400 4 RAGE 50 IL-2 80 CINC-3

300 3 45 60 40 200 2 40 35 100 1 30 20 Protein expression (pg/mL) expression Protein Protein expression (pg/mL) expression Protein 0 0 25 0 Protein expression (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d IL-1a IL-1 R6 100 110 FGF-BP 800 Neuropilin-2 5000 80 4000 100 600 60 90 3000 400 40 80 2000 200 20 70 1000

0 (pg/mL) expression Protein 0 Protein expression (pg/mL) expression Protein 60 (pg/mL) expression Protein 0

Sham 0 h 2 h 1 d 3 d 7 d (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d MIP-1a IL-22 CINC-1 2.0 IFNg 40 500 45 1.8 400 30 40 1.6 300 1.4 20 35 200 1.2 10 100 30 1.0

0.8 0 (pg/mL) expression Protein 0 25 Protein expression (pg/mL) expression Protein Protein expression (pg/mL) expression Protein Protein expression (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d 4 TIM-1 50 TIM-1 4000 TIM-1 1500 TIM-1

3 45 3000 1000 2 40 2000 500 1 35 1000

Protein expression (pg/mL) expression Protein 0 30 (pg/mL) expression Protein Protein expression (pg/mL) expression Protein 0 0 Sham 0 h 2 h 1 d 3 d 7 d (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d Figure 4 The differentially expressed proteins with a decrease-increase-decrease pattern. Protein expression analysis identified 24 proteins with a decreased-increase- decreased expression trend in the injury group at 0 hours, 2 hours, 1 day, 3 days, and 7 days after SCI compared to the sham group. SCI, spinal cord injury; TIM-1, tissue inhibitor of metalloproteinases-1; TNF-α, tumor necrosis factor-α; RANTES, regulated upon activation normal T cell expressed and selected factor; TWEAK R, tumor necrosis factor receptor superfamily member 12A; PDGF-AA, platelet-derived growth factor-AA; RAGE, receptor for advanced glycosylation end products; FGF-BP, fibroblast growth factor binding protein; CINC-2, cytokine induced neutrophil chemoattractant-2; GM-CSF, granulocyte-macrophage colony stimulating factor; GFR alpha-1, GDNF family receptor alpha-1.

© Annals of Translational Medicine. All rights reserved. Ann Transl Med 2021;9(7):570 | http://dx.doi.org/10.21037/atm-21-603 Page 8 of 13 Li et al. A cytokine array analysis in acute spinal cord injury

15 LIX 10 b-NGF

8 10 6

4 5 2 0 Protein expression (pg/mL) expression Protein 0 Protein expression (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d 4000 CNTF 8 IL-13 10 IL-17F

8 3000 6 6 2000 4 4

1000 2 2

0 (pg/mL) expression Protein 0 0 Protein expression (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d Figure 5 The differentially expressed proteins with an increased pattern (upper row) and an irregular pattern (lower row). Protein expression analysis showed 2 proteins with an increased expression trend and 3 proteins with an irregular expression trend in the injury group at 0 hours, 2 hours, 1 day, 3 days, and 7 days after SCI compared to the sham group. SCI, spinal cord injury; LIX, LPS-induced CXC chemokine; b-NGF, beta-Nerve Growth Factor; CNTF, ciliary neurotrophic factor; IL-13, interleukin-13; IL-17F, interleukin-17F.

3000 Prolactin R 30 L-Selectin 800 CINC-2

600 3000 20 400 1000 10 200

0 0 0 Protein expression (pg/mL) expression Protein Protein expression (pg/mL) expression Protein Protein expression (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d TCK-1 TIMP-1 11000 0.6 VEGF 6000

10000 0.4 4000 9000 0.2 2000 8000

7000 0.0

Protein expression (pg/mL) expression Protein 0 Protein expression (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d Sham 0 h 2 h 1 d 3 d 7 d

400 Fractalkine 800 MCP-1

300 600

200 400

100 200

0 0 Protein expression (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d (pg/mL) expression Protein Sham 0 h 2 h 1 d 3 d 7 d

Figure 6 The differentially expressed proteins with an increase-decrease pattern. Protein expression analysis showed 8 proteins with an increase-decrease expression trend in the injury group at 0 hours, 2 hours, 1 day, 3 days, and 7 days after SCI compared to the sham group. SCI, spinal cord injury; CINC-2, cytokine induced neutrophil chemoattractant-2; TCK-1, Tilletia controversa Kühn-1; VEGF, vascular endothelial growth factor; TIMP-1, tissue inhibitor of metalloproteinases-1; MCP-1, monocyte chemoattractant protein-1.

A Injury 0h vs. Sham B Injury 2h vs. Sham 8 8

Notch-1 MCP-1

FIt-3L 6 FIt-3L Nope 6

JAM-A GFR alpha-1 TNFa Notch-1

IL-17F Notch-2 Notch-2 Neuropilin-1 Gp 130 4 4 Gas 1 Nope GFR alpha-1 Gas 1 Neuropilin-1 Erythropoietin TREM-1 Fractalkine IL-2 Ra TREM-1 IL-1 ra Neuropilin-2 Adiponectin RANTES EphA5 IL-2 Ra L-1 R6 Erythropoietin 2 SCF EphA5 CNTF −log10P Value 2 IL-6 HGF P-Cadherin B7-1 –log10P Value Prolactin TWEAK R TIMP-2 IL-7 CD48 TIMP-2 IL-17F IL-7 0 0 −10 −5 0 5 −6 −4 −2 0 2 4 LogFC LogFC

C Injury 1d vs. Sham D Injury 3d vs. Sham E Injury 7d vs. Sham 20 15 15

MCP-1 MCP-1 15 MCP-1 10 10 10 CINC-2 FIt-3L TIMP-1 FIt-3L TIMP-1 5 5 TIMP-1 FIt-3L 5 CINC-2 Notch-1

−log10P Value CD48 −log10P Value −log10P Value EphA5 CD48 Eotaxin ICAM-1 Erythropoietin Fraxtalkine Erythropoietin CD48 Erythropoietin Nope TWEAK R P-Cadherin CNTF 4-166 HGF ICAM-1 TWEAK R P-Cadherin IL-17F CNTF 0 0 0 −10 −5 0 5 10 −10 −5 0 5 10 −10 −5 0 5 10 LogFC LogFC LogFC

Figure 7 Volcano-plot showing the differentially expressed proteins at different time points after spinal cord injury (SCI). (A-E) The volcano-plot shows upregulated genes (red) and downregulated genes (green) in the sham group and the injury group at 0 hours, 2 hours, 1 day, 3 days, and 7 days after SCI. The X-axis represents the logFC (fold change), and the two dotted lines at the X-axis are at X=1 and X=−1 (presenting 2-fold upregulation and 2-fold downregulation, respectively). The Y-axis represents the –log10P value, and the dotted line at the Y-axis is at Y=1.30103 (P=0.05). The upper right red dots represent the differentially expressed proteins with FC >2 and P<0.05, and the upper left green dots represent the differentially expressed proteins with FC <2 and P<0.05.

proteins”, and “regulators and secreted factors”. These immediately following SCI, and the acute phase is one results indicated that cytokine-related pathways, cell death, week after SCI. The inflammatory reaction, which is the and proliferation may play a key role during secondary SCI. main mechanism of acute SCI (13,22,23), contributes greatly to secondary SCI and is a complex process that involves invading immune cells, such as neutrophils and Discussion macrophages, and astrocyte and microglia (23). The SCI results in traumatic injury to the central nervous immune system is a major contributor to the pathogenesis system leading to locomotor and sensory function loss. of spinal cord injury. T and B lymphocytes that respond Novel therapeutic strategies such as anti-inflammatory to the same antigen migrate to the spinal cord injury site, targets, anti-axon growth inhibitory molecules, stem cell resulting in a multifaceted adaptive immune response (24). transplantation, and biomaterial transplantation have been Lymphocyte’s infiltration occurs in the first week after SCI developed in animal studies and clinical trials. However, to and maintained for a long time. In genetic mice which lack date, there are no effective treatment strategies to promote T-cells could find functional and tissue improvement and recovery of spinal cord functional in the clinical setting. The B-cell knockout mice have improved BMS scores associated mechanisms of secondary SCI are still not fully understood with decreased lesion following SCI (25). This observation and this presents a significant barrier to developing may be due to an enhanced regenerative response in effective treatment strategies. Thus, unravelling the precise the absence of lymphocytes, or a reduction in immune- mechanisms underlying SCI is of great importance to the mediated tissue injury when T- and B-cells are removed. development of effective treatment strategies. The cytokines secreted from immune cells, astrocyte, SCI is divided into the super acute phase, the acute and microglia are rapidly expressed within minutes after phase, the subacute phase, and the chronic phase depending SCI (26). Investigating the cytokine expression profile after on time since injury. The super acute phase is the minutes SCI is of great importance in elucidating the underlying

© Annals of Translational Medicine. All rights reserved. Ann Transl Med 2021;9(7):570 | http://dx.doi.org/10.21037/atm-21-603 Page 10 of 13 Li et al. A cytokine array analysis in acute spinal cord injury

A 2h B 5 Names Total Proteins 0h 0 9 0h 2h 1d 3d 7d 2 Erythropoietin Flt-3L 0 2 6 2h 1d 3d 7d 2 CD48 MCP-1 0 2 0 0 2 0h 2h 1d 2 Nope Notch-1 0 0 0 0h 2h 7d 2 EphA5 IL-17F 2 2 0 1d 0h 1d 3d 1 TWEAK R 0 1 0 0 2 0 1 0h 3d 7d 2 CNTF P-Cadherin 0 1 0 0 1d 3d 7d 1 TIMP-1 1 3 0 7d 3d Protein expression (pg/mL) C 10 D 80 800 Erythropoietin and MCP-1

5 60 600

0 40 400

Fold Change −5 20 200 Flt-3L and CD48

−10 Protein expression (pg/mL) expression Protein 0 0 0h 2h 1d 3d 7d Sham 0h 2h 1d 3d 7d Erythropoietin FIt-3L CD48 MCP-1 Erythropoietin FIt-3L CD48 MCP-1 Figure 8 An analysis of the key proteins in the acute phase of spinal cord injury. (A) The Venn diagram shows co-dysregulated proteins at 0 hours, 2 hours, 1 day, 3 days, and 7 days after SCI. (B) The proteins that are co-dysregulated at more than 3 time points are shown. (C,D) The expression profiles of the proteins that are co-dysregulated at more than 4 time points are shown. SCI, spinal cord injury; Flt-3L, Fms related tyrosine kinase 3 ligand; MCP-1, monocyte chemoattractant protein-1; IL-17F, interleukin-17F; TWEAK R, Tumor necrosis factor receptor superfamily member 12A; CNTF, ciliary neurotrophic factor; TIMP-1, tissue inhibitor of metalloproteinases-1.

pathophysiological processes of SCI. Unlike polymerase differentially expressed cytokines in the serum of patients chain reactions (PCRs), enzyme linked immunoassays with chronic SCI (>12 months), with both IL-2 and TNF-α (ELISAs), or Western blot analyses, which can only detect levels upregulated (29). In our study, 67 differentially one marker one time (27), cytokine array is a sensitive, high expressed proteins were detected and analyzed, which is specificity technique that only requires a small sample and a much larger panel of proteins identified compared to can detect a multitude of proteins (up to 500 markers). previous studies. The upregulated proteins such as the pro- Hence, this is an effective method to study protein inflammatory factors, were most likely derived from the expression profiles following SCI(28) . activated immune cells in the blood or leaked proteins due Some studies have simultaneously detected multiple to damage in the blood spinal cord barrier. The differentially protein expressions after SCI. Lee et al. detected expressed proteins in spinal cord tissues at different time 9 differentially expressed proteins using Real-time points could explain the pathophysiological processes after quantitative polymerase chain reaction (RT-qPCR) and injury. Thus, detecting the differentially expressed proteins immunohistochemistry within 21 days post-SCI. TNF-α in the injured spinal cord at different time points is of great mRNA was upregulated 1 hour post-injury and decreased importance to clarify the events in SCI. to baseline by 3 days post-injury. MCP-1 and macrophage In this study, the Rat Cytokine Array 67 was used inflammatory protein (MIP)-1α were also detected after SCI. to demonstrate the shifts in cytokine expression at MCP-1 mRNA increased 1 hour after SCI and decreased 0 hours, 2 hours, 1 day, 3 days, and 7 days after SCI. The to baseline after 14 days (4). Hong et al. detected level- 67 cytokines were separated into high, moderate, and low specific differences in vascular endothelial growth factor expression clusters according to their expression level. The (VEGF), leptin, and IL10 expression in SCI rat plasma used low expression cluster had the most proteins, while the high-throughput ELISA panels. Another study detected 6 moderate expression cluster had the most stably expressed

GO: Cellular Component GO: Molecular Function C-fiberg Erythropoietin receptor binding MAML1-RBP-Jkappa-ICN1 complex Monophenol monooxygenaseactivity Rough endoplasmic reticulum Growth factor activity Neuronal cell body Leaflet of membrane bilayer Cytokine receptor binding Side of membrane Receptor regulator activity Anchoring junction Signaling receptor activator activity Adherens junction Receptor ligand activity External side of plasma membraney Cell body Cytokine activity Intrinsic component of plasma membrane Molecular function regulator Cell surface Signaling receptor binding 0 2 4 6 8 0 2 4 6 8 Gene in common (n) Gene in common (n) Pathway GO:Biological Process IL-17 signaling pathway Regulation of cell death lnterleukin-10 signaling Regulation of cell proliferation IL23-mediated signaling events Defense response lnterleukin-4 and 13 signaling Regulation of programmed cell death Signaling by Interleukins Regulation of apoptotic process Genes encoding secreted soluble factors Response to cytokine Cytokine Signaling in lmmune system Cellular response to cytokine stimulus Cytokine-cytokine receptor interaction Positive regulation of cell proliferation Extracellular matrix and -associated proteins Cytokine-mediated signaling pathway ECM-associated proteins, regulators and secreted factors Response to wounding 0 2 4 6 8 0 2 4 6 8 Gene in common (n) Gene in common (n) Figure 9 The Gene Ontology (GO) and pathway analysis of the differentially expressed proteins. The ordinate represents the GO and pathway terms, and the abscissa represents the number of genes related to the GO and pathway terms. proteins at all time points. There were 6 protein expression result, CD48 was expressed normally in sham rats, but its patterns among the 67 proteins. The 3 most common expression decreased during the acute phase of SCI. This patterns according to protein numbers were decrease- suggested that CD48 may have a potential regulatory role increase-decrease, decrease-increase, and decrease. There and may act as a beneficial receptor in SCI. Consistent with were 4 proteins with expression changes at more than other studies, MCP-1 was upregulated after SCI (33). The 4 time points with statistical significance. Of these expression of MCP-1 can be induced by many inflammatory 4 proteins, EPO, Flt-3L, and CD48, were downregulated factors such as IL-1β and TNF-α (33). It is a chemokine till 7 days after SCI, and MCP-1 was the only upregulated that binds to the CC chemokine receptor-2 of inflammatory protein at all time points. The differential expression of cells. Reports have demonstrated that anti-MCP-1 these 4 proteins suggested that they may have a regulatory antibodies have a beneficial effect on tissue protection role in secondary SCI. EPO may protect tissues via its anti- following SCI (33). Thus, inhibiting MCP-1 expression at oxidation, anti-apoptosis, and neuroprotection abilities. the early stages of SCI may be an effective method for the It protects the spinal cord from ischemia and reperfusion treatment of SCI. injury (30). In fact, EPO has been developed as a therapeutic drug to protect against cell death (7). In our study, EPO Conclusions was normally expressed in rats given a sham operation, but its downregulation was sustained in rats following SCI. In conclusion, this study identified 3 downregulated The downregulated EPO expression indicated that the proteins, EPO, Flt-3L, and CD48, and 1 upregulated injured spinal cord could not resist secondary injury. Thus, protein, MCP-1 that remained upregulated until day 7 upregulation of EPO as soon as possible after SCI may following SCI. Regulation of these 4 proteins, separately or be beneficial for functional recovery of the injured spinal together, at the early stages of SCI may be a novel strategy cord. Flt-3L is one of the growth factors that is significantly for the repair of SCI. decreased in stroke patients (6). Application of Flt-3L decreases the amount of latent virus in infected neurons (31). Acknowledgments The downregulation of Flt-3L in our study also suggested poor anti-injury and regenerative ability. CD48 is thought Funding: This work was supported by the Key Projects to be a receptor expressed on hematopoietic cells and serves for Science and Technology Support (19YFZCSY00660), as an adhesion and costimulatory molecule (32). In our Tianjin Research Innovation Project for Postgraduate

© Annals of Translational Medicine. All rights reserved. Ann Transl Med 2021;9(7):570 | http://dx.doi.org/10.21037/atm-21-603 Page 12 of 13 Li et al. A cytokine array analysis in acute spinal cord injury

Students (2019YJSB106) and Key Projects of Tianjin inflammation after spinal cord injury. Inflamm Regen Natural Science Foundation (19JCZDJC34900). 2016;36:20. 4. Lee YL, Shih K, Bao P, et al. Cytokine chemokine expression in contused rat spinal cord. Neurochem Int Footnote 2000;36:417-25. Reporting Checklist: The authors have completed the 5. Ma SF, Chen YJ, Zhang JX, et al. Adoptive transfer STROBE reporting checklist. Available at http://dx.doi. of M2 macrophages promotes locomotor recovery in org/10.21037/atm-21-603 adult rats after spinal cord injury. Brain Behav Immun 2015;45:157-70. Data Sharing Statement: Available at http://dx.doi. 6. Li X, Lin S, Chen X, et al. The Prognostic Value of Serum org/10.21037/atm-21-603 Cytokines in Patients with Acute Ischemic Stroke. Aging Dis 2019;10:544-56. Conflicts of Interest: All authors have completed the ICMJE 7. Wang P, Xie ZD, Xie CN, et al. AMP-activated uniform disclosure form (available at http://dx.doi. protein kinase-dependent induction of autophagy by org/10.21037/atm-21-603). The authors have no conflicts erythropoietin protects against spinal cord injury in rats. of interest to declare. CNS Neurosci Ther 2018;24:1185-95. 8. Ren Y, Young W. Managing inflammation after spinal cord Ethical Statement: The authors are accountable for all injury through manipulation of macrophage function. aspects of the work in ensuring that questions related Neural Plast 2013;2013:945034. to the accuracy or integrity of any part of the work are 9. Zong S, Zeng G, Wei B, et al. Beneficial effect of appropriately investigated and resolved. All experimental interleukin-1 receptor antagonist protein on spinal cord procedures were approved by the Ethics Committee of injury recovery in the rat. Inflammation 2012;35:520-6. Tianjin Institute of Radiation Medicine (approval number: 10. Boato F, Rosenberger K, Nelissen S, et al. Absence of IRM-DWLL-2019039) and were performed according to IL-1beta positively affects neurological outcome, lesion the National Institutes of Health in the Guide for the Care development and axonal plasticity after spinal cord injury. and Use of Laboratory Animals (NIH Publications no. 85- J Neuroinflammation 2013;10:6. 23, revised 1996). 11. Mukaino M, Nakamura M, Yamada O, et al. Anti-IL-6- receptor antibody promotes repair of spinal cord injury by Open Access Statement: This is an Open Access article inducing microglia-dominant inflammation. Exp Neurol distributed in accordance with the Creative Commons 2010;224:403-14. Attribution-NonCommercial-NoDerivs 4.0 International 12. Lacroix S, Chang L, Rose-John S, et al. Delivery of License (CC BY-NC-ND 4.0), which permits the non- hyper-interleukin-6 to the injured spinal cord increases commercial replication and distribution of the article with neutrophil and macrophage infiltration and inhibits axonal the strict proviso that no changes or edits are made and the growth. J Comp Neurol 2002;454:213-28. original work is properly cited (including links to both the 13. Biglari B, Swing T, Child C, et al. A pilot study on formal publication through the relevant DOI and the license). temporal changes in IL-1beta and TNF-alpha serum levels See: https://creativecommons.org/licenses/by-nc-nd/4.0/. after spinal cord injury: the serum level of TNF-alpha in acute SCI patients as a possible marker for neurological remission. Spinal Cord 2015;53:510-4. References 14. Yang P, Wen H, Ou S, et al. IL-6 promotes regeneration 1. Wu Y, Gao Q, Zhu S, et al. Low-intensity pulsed and functional recovery after cortical spinal tract injury ultrasound regulates proliferation and differentiation by reactivating intrinsic growth program of neurons and of neural stem cells through notch signaling pathway. enhancing synapse formation. Exp Neurol 2012;236:19-27. Biochem Biophys Res Commun 2020;526:793-8. 15. Thompson CD, Zurko JC, Hanna BF, et al. The 2. Mortezaee K, Khanlarkhani N, Beyer C, et al. therapeutic role of interleukin-10 after spinal cord injury. J Inflammasome: Its role in traumatic brain and spinal cord Neurotrauma 2013;30:1311-24. injury. J Cell Physiol 2018;233:5160-9. 16. Khalid K, Frei J, Aboouf MA, et al. Erythropoietin 3. Okada S. The pathophysiological role of acute Stimulates GABAergic Maturation in the Mouse

Hippocampus. eNeuro 2021;8:ENEURO.0006-21.2021. expression in contused rat spinal cord. Neurochem Int 17. Deshmane SL, Kremlev S, Amini S, et al. Monocyte 2000;36:417-25. chemoattractant protein-1 (MCP-1): an overview. J 27. Zhao X, Chen C, Wei Y, et al. Novel mutations of Interferon Cytokine Res 2009;29:313-26. COL4A3, COL4A4, and COL4A5 genes in Chinese 18. Guermonprez P, Gerber-Ferder Y, Vaivode K, et al. Origin patients with Alport Syndrome using next generation and development of classical dendritic cells. Int Rev Cell sequence technique. Mol Genet Genomic Med Mol Biol 2019;349:1-54. 2019;7:e653. 19. McArdel SL, Terhorst C, Sharpe AH. Roles of CD48 28. Amo-Aparicio J, Martinez-Muriana A, Sanchez-Fernandez in regulating immunity and tolerance. Clin Immunol A, et al. Neuroinflammation Quantification for Spinal 2016;164:10-20. Cord Injury. Curr Protoc Immunol 2018;123:e57. 20. Jiang C, Li Z, Wu Z, et al. Integrated Bioinformatics 29. Hayes KC, Hull TC, Delaney GA, et al. Elevated Analysis of Hub Genes and Pathways Associated with a serum titers of proinflammatory cytokines and CNS Compression Model of Spinal Cord Injury in Rats. Med autoantibodies in patients with chronic spinal cord injury. J Sci Monit 2020;26:e927107. Neurotrauma 2002;19:753-61. 21. Duan HQ, Wu QL, Yao X, et al. Nafamostat mesilate 30. Simon F, Floros N, Ibing W, et al. Neurotherapeutic attenuates inflammation and apoptosis and promotes potential of erythropoietin after ischemic injury locomotor recovery after spinal cord injury. CNS Neurosci of the central nervous system. Neural Regen Res Ther 2018;24:429-38. 2019;14:1309-12. 22. Brennan FH, Ruitenberg MJ. Targeting acute 31. Smith JR, Thackray AM, Bujdoso R. Reduced herpes inflammation to complement spinal cord repair. Neural simplex virus type 1 latency in Flt-3 ligand-treated mice Regen Res 2015;10:1596-8. is associated with enhanced numbers of natural killer and 23. Beck KD, Nguyen HX, Galvan MD, et al. Quantitative dendritic cells. Immunology 2001;102:352-8. analysis of cellular inflammation after traumatic spinal 32. McArdel SL, Brown DR, Sobel RA, et al. Anti- cord injury: evidence for a multiphasic inflammatory CD48 Monoclonal Antibody Attenuates Experimental response in the acute to chronic environment. Brain Autoimmune Encephalomyelitis by Limiting the 2010;133:433-47. Number of Pathogenic CD4+ T Cells. J Immunol 24. Jones TB. Lymphocytes and autoimmunity after spinal 2016;197:3038-48. cord injury. Exp Neurol 2014;258:78-90. 33. Yang L, Ge D, Chen X, et al. miRNA-544a Regulates 25. Tanabe S, Yamashita T. B lymphocytes: Crucial the Inflammation of Spinal Cord Injury by Inhibiting contributors to brain development and neurological the Expression of NEUROD4. Cell Physiol Biochem diseases. Neurosci Res 2019;139:37-41. 2018;51:1921-31. 26. Lee YL, Shih K, Bao P, et al. Cytokine chemokine

Cite this article as: Li Q, Li B, Tao B, Zhao C, Fan B, Wang Q, Sun C, Duan H, Pang Y, Fu X, Feng S. Identification of four genes and biological characteristics associated with acute spinal cord injury in rats integrated bioinformatics analysis. Ann Transl Med 2021;9(7):570. doi: 10.21037/atm-21-603