Traumatic brain injury (TBI) remains a major cause of morbidity and mortality worldwide (Khellaf et al., 2019). Based on a survey, the population of TBI patients in China will exceed 1.39 billion, accounting for approximately 18% of the world population (Jiang et al., 2019). Higher rates of death and disability make TBI a global health challenge (Khellaf et al., 2019). The pathogenesis of TBI is a complex process including primary and secondary injuries, which make it incredibly challenging to treat (Jassam et al., 2017). Multitudinous studies have been conducted to search for the therapy of TBI. Disappointingly, these preclinical experimental researches have not been effectively converted to clinical treatment (Galgano et al., 2017) partly because of the comprehensive influence of various complicated secondary biochemical and pathophysiological cascade reactions happening at different times points (Lagraoui et al., 2017; Wang et al., 2018). Therefore, understanding complex pathophysiological and exploring optimal therapies after TBI has become meaningful.

Among the critical pathophysiological processes of TBI are the learning and memory deficits due to the injury (Wu et al., 2013). The hippocampus plays a crucial role in learning and memory while being extremely susceptible to TBI (Weston et al., 2021). Specifically, hippocampal volume reduction has been observed in TBI patients (Bae et al., 2020). The hippocampus is frequently discussed in brain subfields in TBI because of its vital parts in short-term and spatial-related memories (Shokouhi et al., 2020). Although multiple strategies have been explored to improve cognitive impairment after TBI (Yu et al., 2008; Patel and Sun, 2016), the current therapies are far from satisfactory (Konrad et al., 2011). It is necessary to consider the different natural therapeutic methods. Xuefu Zhuyu Decoction (XFZYD), a classical prescription, has been widely used in clinical to treat cardiac-cerebral vascular disease (Zhang et al., 2004; Yang T. et al., 2019; Fu et al., 2020; Wang et al., 2020a). Our previous studies reported that XFZYD could reduce neurological deficits after TBI via inflammatory inhibition (Xing et al., 2016) and improve the long-term prognosis post-TBI via synaptic regulation. (Zhu et al., 2018). Several scientists have explored the potential therapeutic effects of XFZYD, which may alter the protein and metabolites expression (Li et al., 2020; Zheng et al., 2020) in the hippocampus after TBI. Nevertheless, there might be a disconnection between the mRNAs’ expression and their resultant proteins (Gygi et al., 1999; Redell et al., 2009). The variations in mRNA and protein expression levels may be attributed to non-coding RNAs (ncRNAs) affection (Redell et al., 2009).

MicroRNAs (miRNAs), a class of small ncRNAs, govern a variety of physiological and pathological processes such as development, differentiation, metabolism, and apoptosis (Huang et al., 2011). miRNAs are essential ncRNAs abundant in the brain to regulate genes transcription and associated molecules expression (Yang Y. et al., 2019; Rahmati et al., 2021). What’s more, microRNAs (miRNAs), are a well-known diagnostic tool both in the clinical setting and in the medico-legal investigation (Sessa et al., 2019). Notably, investigations demonstrated that miRNAs levels are altered in the acute phase of TBI (Raghavendra Rao et al., 2003; Redell et al., 2009; Xiao et al., 2020). Previous medical investigations indicated that miRNAs may act as possible targets for disease progress evaluation and interference against TBI to alleviate impairment to the cerebrum (Martinez and Peplow, 2017). Integrated bioinformatics analysis has identified several molecules and pathways in rats’ hippocampus after TBI during the acute stage (Xiao et al., 2020). Nonetheless, no research discusses the miRNAs’ alteration in the hippocampus during the subacute phase of TBI. Unlike other diseases, TBI consists of a time-dependent range of events. The diverse alterations in the impaired area, including vascular injury, microglial polarization, neuronal death, and astrocyte activation, have been demonstrated to vary over time with different molecular expression modes (Algattas and Huang, 2013; Jassam et al., 2017). Thus, investigating miRNAs expression patterns in the subacute phase facilitates our knowledge of the underlying molecular mechanisms and the potential treatment after TBI.

In the current research, we explored the expression patterns of miRNAs in TBI and XFZYD-treated groups. First, the miRNAs-sequencing was applied to test the differential expression spectrum of miRNAs between the controlled cortical cortex (CCI) model and the animals treated with XFZYD. Next, we used bioinformatics analysis to investigate several differentially expressed miRNAs’ biological activities to uncover possible treatment pathways for XFZYD. The current study will provide unique insights into seeking the essential mechanisms in the XFZYD treating TBI.

To detect the potential therapeutic approach of XFZYD, we investigated the expression profiles of miRNAs in the hippocampus of TBI and XFZYD treated groups. In the present study, 16 miRNAs were XFZYD-treatment miRNAs. Furthermore, the bioinformatics analysis pointed out that the miRNAs could play roles by regulating cell migration, glutamatergic synapse, protein kinase binding, MAPK signaling pathway, etc. These findings make us deeply understand the XFZYD therapeutic targets and pharmacological mechanisms after TBI.

XFZYD, a traditional Chinese medicine, is recorded in Wang Qing ren’s “Yi Lin Gai Cuo.” Our previous work disclosed the traditional dosage of XFZYD significantly improved spatial learning and memory impairments (Zhu et al., 2018). Evidence-based investigations have manifested that XFZYD can ameliorate neurological recovery post-TBI (Zhou et al., 2017; Zhu et al., 2018). The mNSS of the XFZYD group decreased relative to that of the TBI group on the 3rd (p < 0.01), 7th (p < 0.05), and 14th days (p < 0.01) (Figure 2A). According to the results of the MWM test, The TBI mice exhibited significantly less dwell time in the target quadrant than a sham and XFZYD groups, indicating the retention of spatial and acquired memory was impaired (Figure 2B). A single-factor ANOVA revealed none of the groups’ swimming speeds significantly differed (Figure 2C). The result suggests differences of time in the target quadrant were not due to injury-induced motor impairment. HE staining states that the morphology of the sham group was the round and intact nucleus. In the TBI group, there was severe nuclear concentration, loose staining, and cell death. The pathological degree of the CA1 region in the XFZYD-therapy group was prominently reduced contrasted the TBI group (Figure 2D). Our results are consistent with the previous experiment, indicating the reliability of this study.

Despite animal models being unable to fully simulate the trauma to the human cerebrum, they remain the basis for comprehending the molecular and cellular mechanisms following TBI (Ma et al., 2019). The CCI model is a standard TBI animal model that takes advantage of TBI-0310 to induce damage to the exposed dura (Romine et al., 2014). This model confers duplicatable impairments and mimics numerous features of human trauma such as acute cerebral hemorrhage, blood-brain barrier breakdown, cortical tissue loss, intracranial hypertension, and axonal damage (Osier and Dixon, 2016; Siebold et al., 2018; Lu and Mao, 2019). The CCI model also results in a great many neural functions defects customary in human trauma patients, such as cognitive and motor complications (Adelson et al., 2013; Chen et al., 2014). We established a CCI mouse model as previously reported (Yang et al., 2010). The mNSS and MWM tests indicated that TBI induces neurological deficits in mice (Figures 2A–C). HE staining revealed that TBI leads to brain lesions (Figure 2D), which was quantitatively in good agreement with earlier reports (Xie et al., 2019). It was indicated that the animal model of CCI in our study was reliable.

Traumatic brain injury triggers multitudinous molecular and biochemical alterations during the whole of the central nervous system, including changed transcript expression, disturbed signal communication, affected cell process, and perturbed neurogenesis (Dash et al., 2004; Raghupathi, 2004; Richardson et al., 2007; Redell et al., 2009). Proteomics has been applied to investigate the pathophysiology of TBI. Nonetheless, there could be differences between mRNAs and their related protein products expression levels (Gygi et al., 1999; Redell et al., 2009). To some degree, the distinctions of mRNA and proteins could be ascribed to the roles of miRNAs (Redell et al., 2009). miRNAs, a class of small ncRNAs, control diverse biological action. miRNAs are particularly attractive due to their interactions with their target genes (Liu et al., 2014). Increasing evidence exhibited that miRNAs are engaged in neurological disorders, like Alzheimer’s disease (Gupta et al., 2017), stroke (Mirzaei et al., 2018), and TBI (Xiao X. et al., 2019). Thereby, discerning miRNAs, related targets genes, and their regulatory signaling pathways are crucial in knowing the common biological development of miRNAs and their actions in the disease process (Croce, 2009; Liu et al., 2014). Nowadays, scientists have explored the traditional Chinese Medicine (TCM) pharmacological mechanism and potential therapeutic targets via transcriptomics technologies (Zhang et al., 2017b; Xu, 2017; Li et al., 2018). miRNA-based therapeutics approaches have been assessed at the preclinical and clinical stages. However, trials found single-target strategy could not effectively hinder the development of diseases since additional miRNAs could also affect the target and interfere with the disease’s pathophysiological (Peng et al., 2019). Furthermore, scientists and clinicians regarded that combination therapies for multiple pathological processes might be more practical than single-target treatment in ameliorating neurobehavioral outcomes after TBI (Lagraoui et al., 2017). TCM performs efficient therapies through multiple targets (Li et al., 2020). In the past decades, related findings proved that TCM might affect multiple miRNAs simultaneously (Qian et al., 2013; Momtazi et al., 2016; Peng et al., 2019). Therefore, exploring differentially expressed miRNAs induced by XFZYD in the hippocampus of mice will provide a new direction into the TBI treatment.

RNA-seq supported a platform to analyze a lot of miRNAs simultaneously, comprehensively evaluating potential alterations in expression and generating miRNA expression characteristics for TBI. In this study, 16 differentially expressed miRNAs were found between TBI and XFZYD groups. miR-96-5p, miR-182-5p, miR-183-5p, miR-7015-5p, and miR-296-5p were the five most significant up-regulated miRNAs, while miR-383-5p, miR-142a-5p, miR-33-5p, miR-466d-5p, and miR-466n-5p were the five most significant down-regulated miRNAs (Table 3). Previous references demonstrated that miR-96-5p could regulate spinal cord injury through the NF-κB pathway (Huang et al., 2019) and decreased LPS-induced inflammatory responses (Chen et al., 2020). We identified that miR-96-5p was reduced in TBI and raised in XFZYD treatment. We found that the expression of miR-96-5p was down-regulated in BV-2 cells with scratch injury (Figure 9A). To further explore the functions of the three miRNAs in microglial, we transfected miR-96-5p mimic into the BV-2 microglia with scratch injury to further activate microglia. Subsequently, the pro-inflammatory factors IL-1β, IL-6, and TNF-α in cell culture supernatant were detected by ELISA. The results showed that expressions of IL-1β, IL-6, and TNF-α were significantly up-regulated in the microglia with scratch injury (p < 0.001) but down-regulated when microglia was transfected with miR-96-5p mimic (Figures 10A–C, p < 0.001). Lin et al. (Lin et al., 2017) have shown that miR-183-5p was raised after ischemic post-conditioning. In addition, the enhancement of miR-183-5p relives neuronal deficits after ischemia-reperfusion. Li et al. (Zhu et al., 2020) also verified that miR-183-5p expression was decreased in ischemic mice and reduced ischemic injury by negatively regulating PTEN. Wang et al. (Wang et al., 2020b) have proved that miR-183-5p decreased after intracerebral hemorrhage (ICH). What’s more, miR-183-5p hinders heme oxygenase-1 to improve neurological damage after ICH. Our study also found that XFZYD treatment could elevate the levels of miR-183-5p. BV-2 microglia transfected with miR-183-5p mimic suppressed the levels of IL-6 (Figure 10B, ^## p < 0.01) but did not affect changes of IL-1β (Figure 10A) and TNF-α (Figure 10C). miR-142a-5p, one of the miR-142 isoforms, is notably elevated in the context of autoimmune neuroinflammation (Talebi et al., 2017). Evidenced-based results demonstrated miR-142a-5p is associated with immune response. TBI induces brain injury itself and alters the immune response (Sharma et al., 2019). Expression of miR-142-5p was significantly increased in the frontal white matter from multiple sclerosis patients compared with white matter from non-multiple sclerosis controls. Increasing expression of miR-142 isoforms might be involved in the pathogenesis of autoimmune neuroinflammation by influencing T cell differentiation (Talebi et al., 2017). Our study also found miR-142a-5p was upregulated in TBI and downregulated by XFZYD intervention. BV-2 microglia transfected with miR-142a-5p inhibitor could eliminated the expressions of IL-1β (Figure 10A, p < 0.001) and IL-6 (Figure 10B, p < 0.001), but the expression of TNF-α in BV-2 cells was not affected by miR-142a-5p inhibitor (Figure 10C). However, we have not found any previous research investigating the miR-96-5p, miR-183-5p, and miR-142a-5p in TBI models of animals or humans. Their specific biological functions in TBI deserve to be further investigated. Further studies exploring the relevance of the above miRNAs in TBI will better understand TBI’s biological mechanisms and put insight into novel therapeutic targets for TBI.

Pathway analysis showed that the miRNAs were mainly enriched in the MAPK signaling pathway (path: mmu04010) (Figure 7 and Table 6). The mitogen-activated protein kinase (MAPK) signaling controls extensive biological processes, including growth, differentiation, oxidative stress, and neuroinflammation (Sun and Nan, 2016). Growth, inflammation, and stress response are processes triggered by TBI that are a crucial component of the overall pathophysiology. A large body of evidence suggested that the MAPK signaling pathway regulates inflammation response (Tao et al., 2018), cell apoptosis, and death (Zhang et al., 2020) in TBI. Previous research illustrated that the MAPK signaling pathway is involved in long-term memory (Walz et al., 1999). The recent study also indicated that TCM alleviated the learning and memory in Alzheimer’s disease through the MAPK pathway (Gao et al., 2019). However, the relationship between the MAPK pathway and XFZYD has not been documented. Further studies exploring the relevance of the MAPK in the TBI and XFZYD treatment will better understand the underlying mechanism of XFZYD.

RASA1 is a member of the RAS GTPase Activating Protein (RAS-GAP) family. The well-known oncoprotein RAS can be inactivated by binding to RAS-GAP members. Some studies have shown that mutation or loss of function of RASA1 leads to activation of the RAS-MAPK cascade in malignant tumors (Xiao W. et al., 2019). Dai et al. (Dai et al., 2019) have identified that lncRNA GAS5 served as a competing endogenous RNA (ceRNA) to upregulate Rasa1 via sponging miR-335 in the progression of TBI. Rasa1 was one of the potential targets of miR-96-5p (Figure 10D). We found that the level of Rasa1 was significantly up-regulated in BV-2 cells with scratch injury; while under the effects of miR-96-5p mimic, Rasa1 expression was down-regulated (Figure 10E).

The present study has several limitations. To begin with, the current study has only completed the functional predictions and expression profile; therefore, the next step is determining the roles of the alternative miRNAs in vitro-and-in vivo. Second, clinical samples and large sample sizes will be required to validate the current findings in future studies. More research is needed to explore the particular interactions and sites of binding between mRNAs and miRNAs. Besides, future studies should focus on the link between miR-96-5p and XFZYD treatment.

Herein, we explored the expression profiles of miRNAs in experimental TBI treated with XFZYD. In comparison to TBI, 16 miRNAs were considerably XFZYD therapy-related. miRNAs could be new therapeutic targets for XFZYD in treating TBI-induced cellular processes. The current study lays the groundwork for future research into the methods through which XFZYD protects against long-term neurological deficits following TBI.