Advancements in our understanding of circular and long non-coding RNAs in spinal cord injury.

Spinal cord injury (SCI), either from trauma or degenerative changes, can result in severe disability and impaired quality of life. Understanding the cellular processes and molecular mechanisms that underlie SCI is imperative to identifying molecular targets for potential therapy. Recent studies have shown that non-coding RNAs, including both long non-coding RNAs (lncRNAs) and circular RNAs (circRNAs), regulate various cellular processes in SCI. In this review, we will describe the changes in lncRNA and circRNA expression that occur after SCI and how these changes may be related to SCI progression. Current evidence for the roles of lncRNAs and circRNAs in neuronal cell death and glial cell activation will also be reviewed. Finally, the possibility that lncRNAs and circRNAs are novel modulators of SCI pathogenesis will be discussed.

Introduction

Spinal cord injury (SCI) is a devastating insult to the central nervous system. It not only causes severe disturbances in sensorimotor and autonomic functions, but also entails a high economic burden associated with numerous morbidities (Badhiwala et al., 2021). Patients with SCI typically undergo two phases of pathology: (1) the primary phase incited by a traumatic injury, followed by (2) a secondary phase that involves hypoxia, inflammation, and apoptosis (Fouad et al., 2021). Severe sensorimotor deficits occur as a result of these cellular changes, and recovery is challenging because of the poor regenerative capacity of nerve cells (Heller et al., 2021). Understanding the molecular and pathophysiological processes involved in SCI is therefore paramount, as there may be druggable targets that can be exploited to help overcome these challenges.

Long non-coding RNAs (lncRNAs) are non-coding RNAs that are more than 200 nucleotides in length and do not have a protein-coding function (Khan et al., 2021). LncRNAs are transcribed from exons, promoters, and intergenic regions, which produce different kinds of lncRNAs (Zhang et al., 2021). There are mammalian orthologs of only 5.1% of zebrafish lncRNAs, suggesting poor overall conservation compared with protein-coding genes (Hadjicharalambous and Lindsay, 2019). LncRNAs act as “sponges” for miRNAs and RNA-binding proteins (RBPs) (Yang et al., 2021). For example, the lncRNA Neat1 functions as a molecular sponge for miR-124 to induce activation of Wnt/β-catenin signaling in spinal cord neural progenitor cells (Cui et al., 2019). LncRNAs also control gene expression by different mechanisms, such as transcriptional repression, translational repression, and transcript degradation (Dykes and Emanueli, 2017). In fact, lncRNAs have been implicated in the regulation of many molecules and cellular processes associated with neurons, microglial cells, and astrocytes (Wang et al., 2015, 2020; Jiang and Zhang, 2018). Circular RNAs (circRNAs) are another type of non-coding RNA with a stable cyclic structure. In contrast to linear RNAs, which terminate in 5′ caps and 3′ tails, circRNAs form closed loops based on covalent binding of the 3′ and 5′ ends. The roles of circRNAs have come to be more appreciated recently, and many circRNAs have been identified (He et al., 2021). CircRNAs are produced from exons (exonic circRNA), introns (intronic circRNA), and intergenic regions. Unlike lncRNAs, most circRNAs are stable, specific, and highly conserved (Kristensen et al., 2019). CircRNAs also function as sponges for microRNAs and RBPs (Wawrzyniak et al., 2020). Importantly, differential expression of circRNAs has been observed in SCI, suggesting that they may serve as promising therapeutic targets (Zhao et al., 2020a). In this review, we will describe recent research progress regarding the role of lncRNAs and circRNAs in SCI.

Database Search Strategy

The literature search for this narrative review was performed using the Web of Science, MEDLINE, and Google Scholar databases (searching all content up until June 2021). The following combinations of keywords were used to retrieve articles: neuronal cells, neuronal glia, lncRNAs, circRNAs, spinal cord, regulation, function, pathogenesis. The results were further screened by title and abstract to exclude those studies that had relatively low relevance to our topic.

Alterations and Regulation in LncRNAs and CircRNAs after Spinal Cord Injury

The treatment regimen and overall prognosis for SCI are based heavily on the patient’s clinical status (Griffin and Bradke, 2020). LncRNA and circRNA microarrays, RNA sequencing (RNA-seq), and bioinformatics analyses have been utilized to study the altered expression and regulatory functions of lncRNAs and circRNAs in patients with SCI in an attempt to find any clinical correlation. In fact, numerous lncRNAs and circRNAs are found in the spinal cord after SCI, suggesting that their expression levels change in response to SCI. Additionally, studies have recently reported alterations in the expression profiles of lncRNAs and circRNAs in rats with SCI (Table 1 and ).

Table 1

Differentially expressed long non-coding RNAs (lncRNAs) and circular RNAs (circRNAs) after SCI

Reference	Animal	Model	Level	Sampling	Methods	LncRNA/circRNA expression changes	The correlation of the RNA expression with SCI progress
Qin et al., 2018	Adult female SD rats	SCI induced by an Infinite Horizon Impactor (10 g, 12.5 mm)	T10	3 d post-SCI	Microarray	CircRNA: 1676 (415 up, 1261 down)	--
Zhou et al., 2018a	Adult female SD rats	Contusive SCI using an NYU impactor (10 g, 25 mm)	T10	2 h post-SCI	Microarray	LncRNA: 772 (528 up, 244 down)	--
Zhang et al., 2018c	Adult male SD rats	Chronic SCI using dried water-absorbing polyurethane polymer sheets coated with a sustained-release membrane	C5	28 d post-CSCI	Microarray	LncRNA: 1266 (738 up, 528 down)	--
Shi et al., 2019	Adult female SD rats	Contusive SCI using an NYU impactor (10 g, 50 mm)	T10	2 d post-SCI	Microarray	LncRNA: 3193 (1332 up, 1861 down)	--
Wu et al., 2019	Adult male SD rats	SCI through spinal cord hemisection	T9	1, 3, 7, 14, 21, or 28 d post-SCI	RNA sequencing	360 circRNAs were differentially expressed at 1, 3, 7, 14, 21, or 28 d post-SCI 94% of circRNAs decreased from 3 d onward	Knockdown of circRNA_01477 significantly inhibited astrocyte proliferation and migration
Zhou et al., 2019	Adult male SD rats	SCI using MASCIS Impactor weight drop Device (10 g, 25 mm)	T9	6 h after SCI	RNA sequencing	CircRNA: 150 (99 up, 51 down)	--

SD: Sprague-Dawley; SCI: spinal cord injury.

Diagram showing the differential expression of lncRNAs and circRNAs after SCI.

Increases and decreases in lncRNA and circRNA expression after SCI are indicated by up arrows and down arrows, respectively. The detection time and methods are also shown. circRNA: circular RNA; lncRNA: long non-coding RNA; SCI: spinal cord injury

The competing endogenous RNA (ceRNA) hypothesis explains how RNA transcripts are regulated by microRNAs (miRNAs). Changes in the expression of one miRNA can alter the number of unbound miRNAs and their subsequent activities. There are many molecular models supporting this hypothesis, one of which includes lncRNAs as targets for altered expression in response to changes in miRNA levels. Wang et al. identified differentially expressed lncRNAs (DE lncRNAs) in SCI based on the ceRNA hypothesis. They utilized the Gene Expression Omnibus database of the National Center for Biotechnology Information to study the roles of lncRNAs and described relationships between lncRNAs and miRNAs (Wang et al., 2019). They built a network consisting of 13 lncRNAs, 93 mRNAs, and nine miRNA nodes, with a total of 202 edges. Three lncRNA nodes were identified based on the network that were associated with autophagy, extracellular communication, and transcription factor networks. Importantly, one of the three lncRNAs (i.e., XR_350851) regulating autophagy was suggested as a novel biomarker and therapeutic target for SCI.

Several studies have shown the prevalence of lncRNAs in SCI. Shi et al. (2019) noted altered expression of DE lncRNAs and differentially expressed mRNAs (DE mRNAs) in a rat model at 2 days post-SCI. They identified 3193 DE lncRNAs and 4308 DE mRNAs between the SCI and control groups. The ten core genes discovered by protein-protein interaction network analysis included CCNA2, CCNB1, CDC20, CDK1, IL6, ITGAM, MYC, POLE, TNF, and TOP2A (Shi et al., 2019). Zhou et al. (2018a) examined the expression of lncRNAs and mRNAs in a rat model. They identified 772 DE lncRNAs and 992 DE mRNAs at 2 hours post-SCI by performing microarray analyses on SCI and control samples. This study also suggested that the CCL2, CSF2, FOS, FGF2, IL6, JUN, MBOAT4, MYC, STAT3, and TNF loci contribute to the pathogenesis of the primary phase of SCI. Zhang et al. (2018c) found 1266 DE lncRNAs and 847 DE mRNAs in animals with chronic SCI. This study demonstrated that expression of Col6a1 and miR-330-3p, which are both regulated by lncRNA6032, were significantly increased and decreased, respectively.

The prevalence of circRNAs in SCI has been demonstrated by several studies. Zhou et al. (2019) performed RNA sequencing of spinal cord tissue from rats with induced SCI to analyze the circRNA expression profile and found 150 DE circRNAs. Out of these 150 DE circRNAs, 99 were up-regulated, and two (i.e., circRNA_07079 and circRNA_01282) were found to be associated with SCI. These molecules could play a vital role in SCI through the circRNA-targeted miRNA-mRNA axis (Zhou et al., 2019). Qin et al. (2018) described a total of 1676 DE circRNAs in the contused spinal cord using circRNA microarrays; 1261 circRNAs were significantly down-regulated, while 415 were up-regulated. Additionally, RT-qPCR analysis showed that rno_circRNA_005342, rno_circRNA_015513, rno_circRNA_002948, rno_circRNA_006096, and rno_circRNA_013017 were down-regulated. Wu et al. (2019) reported DE circRNAs at various timepoints (0, 1, 3, 7, 14, 21, and 28 days) after performing a right-sided spinal cord hemisection in rats. This study identified 360 DE circRNAs and found that astrocyte migration and proliferation were associated with circRNA-01477, which acts as a sponge to regulate miR-423-5p expression. This study noted that the circRNA-01477/miR-423-5p axis could be an important regulator of neuronal regeneration during recovery from SCI (Wu et al., 2019).

Functions of LncRNAs and CircRNAs – Neuronal Apoptosis

Our knowledge of DE circRNAs and lncRNAs in SCI is rapidly expanding. Regulation of glial cell activation and neuronal apoptosis by lncRNAs and circRNAs are two particular areas that are being intensively investigated (Figures , and ).

LncRNAs and circRNAs involvement in SCI.

The probable mechanisms of the effects exerted by lncRNAs on SCI. Several lncRNAs and circRNAs have been reported to play different roles, primarily in neuroinflammation, neuronal apoptosis, and oxidative stress, following SCI. The lncRNA PTENP1 promoted neuronal apoptosis in SCI by inhibiting miR-21 and miR-19b expression and up-regulating PTEN expression. A decrease in LINC00707 expression was shown to reduce lipopolysaccharide-induced inflammation and apoptosis in PC-12 cells by targeting miR-30a-5p/Neurod 1. Overexpression of the lncRNA SNHG16 reduced H2O2-induced cell injury through the AMPK and ERK1/2 signaling pathways by up-regulating miR-423-5p. Up-regulation of the lncRNA ANRIL was associated with SCI severity and involved targeting the miR-499a/PDCD4 axis and regulating PI3K/Akt/mTOR/p70s6K signal transduction. Overexpression of the lncRNA DGCR5 inhibited neuronal apoptosis by suppressing PRDM5 in hypoxia-induced SCI. Knockdown of the lncRNA BDNF-AS showed anti-apoptotic effects mediated by the miR-130b-5p/PRDM5 axis. Decreased expression of the lncRNA Sox2ot reduced H2O2-induced oxidative injury in SCI by altering miR-211 expression and Akt/mTOR/p70S6K pathway activation, thereby relieving oxidative damage. Overexpression of the lncRNA MALAT1 and Bcl-2 exerted anti-apoptotic effects in the neurocytes of a rat model of spinal cord ischemia reperfusion injury by regulating miR-204. Treatment with hydrogen sulfide protected the spinal cord by increasing expression of the lncRNA CasC7, which targets miR-30c. Overexpression of the lncRNA-Map2k4 and FGF1 promoted neuronal cell proliferation in SCI by modulating miR-199a expression. Up-regulated circ-HIPK3 protected neurocytes from apoptosis by regulating the miR-588/DPYSL5 axis in SCI. CircTYW1 induced FGF9 up-regulation by inhibiting miR-380 expression, which reduced neuronal cell apoptosis in SCI rats. Knockdown of the lncRNA XIST appeared to reduce neuronal apoptosis in a rat model of SCI by regulating the miR-494/PTEN/Akt signaling axis. Down-regulation of the lncRNA TUG1 inhibited TLR4 signaling pathway-mediated inflammatory damage via suppression of TRIL expression. Overexpression of the lncRNA MALAT1 activated the IKKβ/NF-κB signaling pathway via modulation of miR-199b expression and increased the expression of proinflammatory cytokines. Overexpression of the lncRNA TUSC7 inhibited microglial activation and inflammatory factor expression in microglia cells by regulating PPAR-γ expression through miR-449a. The lncRNA SNHG5 increased the viability of microglia and astrocytes by upregulating KLF4. Lastly, down-regulation of the lncRNA SCIR1 promoted astrocyte migration and proliferation by regulating Bmp7 and Adm expression. Adm: Adrenomedullin; Akt: serine/threonine kinase; AMPK: adenosine 5′-monophosphate (AMP)-activated protein kinase; ANRIL: antisense non-coding RNA in the INK4 locus; Bcl-2: B-cell lymphoma-2; Bmp7: bone morphogenetic protein 7; circRNA: circular RNA; DPYSL5: dihydropyrimidinase like 5; ERK1/2: extracellular signal-regulated kinase 1/2; FGF: fibroblast growth factor; H2O2: hydrogen peroxide; IKKβ: IκB kinase β; Klf4: Krüppel-like factor; lncRNA: long non-coding RNA; mTOR: mammalian target of rapamycin; Neurod1: neurogenic differentiation factor 1; NFκB: nuclear factor kappa-B; p70s6K: ribosome S6 protein kinase; PDCD4: programmed cell death 4; PI3K: phosphatidylinositol 3-kinase; PPARγ: peroxisome proliferator-activated receptor gamma; PRDM5: PR domain zinc finger protein 5; PTEN: phosphatase and tensin homolog; PTENP1: PTEN pseudogene 1; SCI: spinal cord injury; TLR4: Toll-like receptor 4.

Differentially expressed lncRNAs and circRNAs regulate target genes, resulting in multiple outcomes after SCI.

Disrupted expression of lncRNAs and circRNAs during SCI results in multiple pathological outcomes, including neuronal apoptosis, blood-brain barrier disruption, tissue infarction, microglial and astrocyte activation, and more. Left panel: Increased lncRNA expression lead to down-regulation of target genes, resulting in poorer outcomes after SCI. The lincRNAs-target gene pairs include lncRNA PILNP1:PTEN, lnc00707:neurod1, lncANRIL:PI3K/Akt/mTOR/p70S6K, lncRNABDNF-AS:PRDM5, lncRNASox2ot: Akt/mTOR/p70S6K, lncRNAXist:PTEN/AKT, lncRNATUG1:TLR4/NFκB/IL1β, Lnc SNHG5: Klf4, and others. Right panel: Decreased lincRNAs and circRNAs lead to up-regulation of target genes, resulting in detrimental outcomes after SCI. The lincRNA-target gene pairs include lncSNGH16:AMPK/ERK1/2, lncDGCR5:PRDM5, lncRNA MALAT1:Bcl2(IKKβ/NFκB), lncRNACasC7:Beclin1, lncRNA Map2k4:FGF1, lncTUSC7:PPARγ, and lncSCIR1:Bmp7/Adm, and the circRNA-target gene pairs include circHIPK3:DPYSL5, circTYW1:FGF9, and others. Adm: Adrenomedullin; Akt: serine/threonine kinase; AMPK: adenosine 5′-monophosphate (AMP)-activated protein kinase; ANRIL: antisense non-coding RNA in the INK4 locus; BBB: blood-brain barrier; Bcl-2: B-cell lymphoma-2; Bmp7: bone morphogenetic protein 7; circRNA: circular RNA; DPYSL5: dihydropyrimidinase like 5; ERK1/2: extracellular signal-regulated kinase 1/2; FGF: fibroblast growth factor; IKKβ: IκB kinase β; Klf4: Krüppel-like factor; lncRNA: long non-coding RNA; mTOR: mammalian target of rapamycin; Neurod1: neurogenic differentiation factor 1; NFκB: nuclear factor kappa-B; p70s6K: ribosome S6 protein kinase; PDCD4: programmed cell death 4; PI3K: phosphatidylinositol 3-kinase; PPARγ: peroxisome proliferator-activated receptor gamma; PRDM5: PR domain zinc finger protein 5; PTEN: phosphatase and tensin homolog; PTENP1: PTEN pseudogene 1; SCI: spinal cord injury; TLR4: Toll-like receptor 4.

Neuronal apoptosis is an essential element of SCI pathophysiology. Because neurons are the core functional component of spinal cord signaling, it is important to understand SCI regulatory mechanisms (Zhu et al., 2020). LncRNAs are associated with neuronal survival after SCI (Shi et al., 2018; Yuan et al., 2019; Kong et al., 2020). In particular, the roles of the lncRNA phosphatase and tensin homolog pseudogene 1 (PTENP1) in the pathogenesis of SCI have been revealed (e.g., SCI-induced neural cell apoptosis results in up-regulation of PTEN). Interestingly, elevated miR-21/miR-19b levels could inhibit neuronal apoptosis by down-regulating PTEN expression. Additionally, the lncRNA PTENP1 promotes neuronal apoptosis in SCI by suppressing miR-21 and miR-19b expression (Wang et al., 2020). Zhu et al. (2019) showed that LINC00707, a lncRNA, was up-regulated in response to lipopolysaccharide (LPS)-induced inflammation, which mimics the neuronal injury environment after SCI. The study showed that LINC00707 reduced LPS-induced inflammation and apoptosis in LPS-treated PC-12 cells by targeting miR-30a-5p/Neurod 1, suggesting that LINC00707 could be a druggable target for SCI (Zhu et al., 2019). Another study demonstrated that the lncRNA SNGH16 reduces H2O2-induced oxidative injury in PC-12 cells by increasing miR-423-5p expression, thereby altering AMPK and ERK1/2 signaling, suggesting that inhibiting miR-423-5p could eliminate SNGH16’s protective effect on PC-12 cells (Liu et al., 2019).

LncRNAs have been identified that modulate the activity of various molecular pathways after SCI. Guo et al. (2019) showed that the lncRNA ANRIL regulates the rate of PI3K/Akt/mTOR/p70s6K signal transduction by targeting the miR-499a/PDCD4 axis; increased ANRIL expression was associated with the severity of SCI. A few studies have indicated that the lncRNA BDNF-AS/miR-130b-5p/PRDM5 axis could be a promising molecular target for SCI therapies (Zhang et al., 2018a). The lncRNA DGCR5 was found to be down-regulated in rat neurons after hypoxia-induced SCI. When the lncRNA DGCR5 was overexpressed, neuronal apoptosis was inhibited, as it suppressed PRDM5 (Zhang et al., 2018b). In addition, the lncRNA BDNF-AS was shown to function as a ceRNA by compartmentalizing miR-130b-5p in neurons after acute SCI; in fact, PRDM5 was found to be a target of miR-130b-5p (Zhang et al., 2018a). Yin et al. (2018) suggested that reduced expression of the lncRNA Sox2ot could protect PC-12 cells against H2O2-induced oxidative injury in SCI. The mechanism was based on the miR-211/MCL-1 isoform 2 axis, in which MCL-1 isoform 2 activated the Akt/mTOR/p70S6K pathway to mediate oxidative damage (Yin et al., 2018). Other studies have found that the lncRNA MALAT1, which is regulated by miR-204, protects neurocytes from apoptosis by upregulating the expression of Bcl-2 (Qiao et al., 2018), and that treatment with hydrogen sulfide protects the spinal cord by increasing expression of the lncRNA CasC7, which is regulated by miR-30c (Liu et al., 2018). It has been reported that the lncRNA Map2k4 regulates neuronal growth through a pathway involving FGF1 and miR-199a (Lv, 2017). Lastly, the XIST/miR-494/PTEN/Akt signaling axis was noted to play a key role in SCI; knockdown of the lncRNA XIST appeared to reduce neuronal apoptosis in a rat model of SCI (Gu et al., 2017). CircRNAs continue to attract interest in spinal cord injury research. Circ-HIPK3 is expressed at lower levels in SCI rat models than in control rats, and studies have shown that circ-HIPK3 functions as a sponge for miR-558 to upregulate DPYSL5. Circ-HIPK3 alleviates neuronal apoptosis by modulating the miR-588/DPYSL5 axis in SCI (Zhao et al., 2020a). Sun et al. (2021) reported that circTYW1 was down-regulated in rats after SCI. Also, circTYW1 induced FGF9 up-regulation by repressing miR-380, which reduced neuronal cell apoptosis and promoted the recovery of neurological function (Sun et al., 2021). These findings may provide new opportunities for the clinical treatment of SCI.

Functions of LncRNAs and CircRNAs – Glial Activation

Neuroglia regulate neuronal apoptosis, demyelination, and inflammation. Glial cells are activated within a day of SCI (microglial activation) and persist for months to years (astrogliosis) following the injury (DePaul et al., 2017; Bellver-Landete et al., 2019). Glia undergo various cellular and molecular changes that provide targets for potential therapies (Li et al., 2019). Microglia are involved in the inflammatory response, and astrocytes are the main components of glial scars after SCI (Zhang et al., 2019b; Xie et al., 2020). Knockdown of the lncRNA TUG1 inhibits TLR4 signaling (i.e., IL-1β/NF-κβ inflammatory pathways), TRIL-mediated proinflammatory cytokines, and microglial activation after spinal cord ischemia-reperfusion injury (Jia et al., 2019). These inflammatory changes result in decreased brain-spinal cord barrier integrity, as well as neurological deficits. Reducing expression of the lncRNA MALAT1 leads to functional recovery after SCI by inhibiting the inflammatory microglial response. Decreased expression of the lncRNA MALAT1 is associated with a reduced microglial inflammatory response (Zhou et al., 2018b). Overexpression of the lncRNA tumor suppressor candidate 7 (TUSC7) inhibits the inflammatory microglial response via the miR-449a/PPAR-γ pathway. The USC7/miR-449a/PPAR-γ axis induces microglial activation and inflammation. This axis provides novel therapeutic targets for alleviating neuropathic pain, which could be useful in cervical spondylotic radiculopathy (Yu et al., 2018).

Astrocytes are the most prominent glial cell type in the central nervous system. In response to SCI, astrocytes proliferate and undergo reactive gliosis to form a glial scar. The glial scar tissue serves as a biochemical and physical barrier to neuronal plasticity and regeneration and impedes functional recovery (Zarei-Kheirabadi et al., 2019). Researchers have investigated the effects of lncRNAs on astrocyte proliferation and reactive gliosis. Jiang and Zhang (2018) found that the lncRNA SNHG5 directly interacts with KLF4 to upregulate eNOS expression. This increases the viability of astrocytes and microglia and worsens SCI. Wang et al. (2015) found that the expression level of the lncRNA SCIR1 is inversely related to the expression levels of Bmp7 and Adm, both of which promote astrogliosis in the spine, by knocking down SCIR1 in cultured astrocytes. They concluded that down-regulation of the lncRNA SCIR1 could promote astrocyte migration and proliferation and play a role in the pathophysiology of SCI. Our review may have involved incomplete article retrieval or reporter bias, so further review should be performed in the future.

Conclusion and Clinical Perspectives

Microarray and RNA-seq data have helped to study lncRNA and circRNA expression changes in response to SCI and how these RNA species may be involved in the intricate cellular and molecular processes involved in SCI pathology (Zhang et al., 2018c; Zhou et al., 2018a, 2019; Shi et al., 2019; Wu et al., 2019). These findings shed new light on the mechanisms of SCI and could lead to the development of novel therapeutics for SCI. To further investigate the potential role of lncRNAs and circRNAs in SCI, however, more studies are needed to analyze the regulatory mechanisms of the differentially expressed lncRNAs and circRNAs.

Neuronal apoptosis after SCI is an important pathological process and can cause secondary insults that are concomitant with axonal degeneration (Zhu et al., 2020). Microglia are sensitive to secondary injury and exhibit an exaggerated inflammatory response in SCI; in addition, reactive astrocytes play a key role in wound healing and functional recovery after SCI (Zhang et al., 2019b). Studies suggest that ncRNAs, including lncRNAs and circRNAs, function as essential regulators in SCI. However, there are several limitations to these studies. The aforementioned research results are mainly based on animal and cellular models of SCI. More knowledge of these ncRNAs, their regulatory functions, and signaling pathway activities needs to be acquired based on bioinformatic analyses, clinical samples, and data regarding the multidimensional interactions of RNA species (Gu et al., 2017; Zhou et al., 2018b; Zhang et al., 2019a). A comprehensive analysis of ncRNA expression profiles could help clarify the interactions between lncRNAs and circRNAs. Additionally, the regulatory roles of lncRNAs have not been thoroughly investigated, although dysregulation of lncRNAs has been noted after SCI (Zhang et al., 2018a, b; Guo et al., 2019). The role of circRNAs in regulating SCI pathophysiology is still largely unknown. There are a few hurdles to overcome before circRNAs and lncRNAs can be used successfully as therapeutic options. For instance, an effective delivery system is needed. Studies have indicated that some circRNAs are conserved between two species (Rybak-Wolf et al., 2015; Baskozos et al., 2019); however, lncRNAs are poorly conserved among species (Zhang et al., 2021). This suggests that the most effective way to deliver these RNA species to the desired cellular and molecular targets could be through the use of an appropriate vector.

LncRNAs and circRNAs are increasingly recognized as important regulatory molecules in SCI. CircRNAs and lncRNAs could serve as promising biomarkers for SCI diagnosis, prognosis, and treatment. Future studies exploring the detailed molecular mechanisms and regulatory roles of lncRNAs and circRNAs may bring us closer to developing promising therapies for SCI.

Additional file: .

Table 2

Long non-coding RNAs (lncRNAs) and circular RNAs (circRNAs) involved in spinal cord injury (SCI)

Reference	Animal	Model	Level	lncRNAs and circRNAs	Tissues or cells of lncRNAs and circRNAS	The correlation of the RNA expression with SCI progress
Wang et al., 2015	Adult female SD rats	The contusive spinal cord injury was performed at T10 spinal cord using an NYU impactor (10 g, 12.5 mm).	T10	LncRNA SCIR1	Spinal cord tissue and primary astrocyte	LncRNA SCIR1 knockdown could promote astrocyte proliferation and migration in vitro, which might play a detrimental role in the pathophysiology of traumatic SCI.
Gu et al., 2017	Adult male SD rats	Spinal cord was subjected to impact trauma by compression at an interval of 12.5 mm to produce severe injury.	T10	LncRNA XIST	Spinal cord tissue	LncRNA-XIST knockdown may limit neuronal apoptosis in SCI.
Lv, 2017	–	–		LncRNA-Map2k4	Primary neuron	LncRNA-Map2k4 could promote spinal cord neuron growth.
Jiang and Zhang, 2018	Adult male SD rats	The dorsal surface of the T10 level was subjected to a 25 g/cm impact.	T10	LncRNA SNHG5	Spinal cord tissue and primary astrocytes and microglia	LncRNA SNHG5 could enhance astrocytes and microglia viability.
Liu et al., 2018	–	The balloon was inflated with 0.05 mL of distilled water to induce spinal cord ischemia. 12 hours after ischemia, the balloon was deflated.	–	Lnc RNA CasC7	Spinal cord tissue and SH5Y-SY cells	Upregulation of lncRNA-CasC7 could reduce neuronal cell apoptosis in SCII.
Qiao et al., 2018	Adult male SD rats	SCII was induced by clamped for 14 min.	–	LncRNA MALAT1	Spinal cord tissue and AGE1.HN and P12 neuronal cell lines	Overexpression of LncRNA MALAT1 could reduce neuronal cell apoptosis in SCII.
Yin et al., 2018	–	–	–	Lnc RNA Sox2ot	PC12 neuronal cells	Knockdown of lncRNA -Sox2ot could protect PC 12 cells from H2O2-exposed injury in SCI
Yu et al., 2018	Adult male SD rats	Spinal Cord compression Model: Ligamentum flavumin of C5–C6 and C6–C7 were resected, the periosteum of C6 lamina was removed.	C6	LncRNA TUSC7	Spinal cord tissue and microglia cell line HAPI	Upregulation of LncRNA TUSC7 could repress the inflammation induced by microglia activation in SCI.
Zhang et al., 2018a	Adult male SD rats	Spinal cord crush was performed on the T10 spinous process for 20 s.	T10	LncRNA DGCR5	Spinal cord tissue and PC12 neuronal cells	LncRNA DGCR5 could suppress neuronal apoptosis and improve ASCI.
Zhang et al., 2018b	Adult male SD rats	ASCI model was induced by extradural compression.	T10	LncRNA BDNF-AS	Spinal cord tissue and AGE1.HN and PC12 neuronal cells	Reduction of lncRNA BDNF-AS could inhibit neuronal cell apoptosis in ASCI.
Zhou et al., 2018b	Adult male SD rats	ASCI was induced by the weight drop (10 g) from 2.5 cm height.	T10	lncRNA MALAT1	Spinal cord tissue and primary microglia and microglia cell lines N9, BV2	LncRNA MALAT1 knockdown could attenuate ASCI by inhibiting inflammatory response of microglia.
Guo et al., 2019	–	–	–	LncRNA ANRIL	PC12 neuronal cells	A high level of lncRNA -ANRIL could aggravate H2O2-disposed injury of PC12 cells.
Jia et al., 2019	Adult male SD rats	Spinal cord ischemia was induced by cross-clamping the descending aorta that was just distal to the left subclavian artery for 14 min.	–	LncRNA TUG1	Spinal cord tissue	Reduction of lncRNA TUG1 repressed inflammatory damage after SCII
Liu et al., 2019	–	–	–	LncRNA SNGH16	PC12 neuronal cells	LncRNA SNGH16 could reduce H2O2-evoked cell injury in PC12 cells.
Zhu et al., 2019	–	–	–	LINC00707	PC12 neuronal cells	Inhibition of LINC00707 could alleviate lipopolysaccharide-induced inflammation and apoptosis of PC12 cells.
Wang et al., 2020	Adult male SD rats	SCI induced by an impactor (10 g, 25 mm).	T9/10	LncRNA PTENP1	PC12 neuronal cells	Decrease of lncRNA PTENP1 expression could inhibit neuronal apoptosis.
Zhao et al., 2020a	Adult male SD rats	Spinal Cord Impactor (Precision Systems and Instrumentation) was used for delivering the contusion injury of 200 kdyn (2 × 10–3 kN).	T10	circ-HIPK3	Spinal cord tissue and AGE1.HN and P12 neuronal cell lines	Circ-HIPK3 could relieve neuronal cell apoptosis in SCI.
Sun et al., 2021	Adult male SD rats	A syringe needle was utilized to stimulate the injury, which was released from a height of 12.5 mm above the surface of the cord.	T8	circTYW1	Spinal cord tissue and P12 neuronal cell lines	CircTYW1 could promote neurological recovery in SCI rats and inhibit neuronal cell apoptosis.

ASCI: Actue spinal cord injury; circRNA: circular RNA; lncRNA: Long non-coding RNA; SCII: spinal cord ischemia-reperfusion injury; SD: Sprague-Dawley.