Engineering a targeted daphnetin delivery system to enhance spinal cord injury treatment effectiveness

Abstract

Spinal cord injury (SCI) is a severe central nervous system disorder characterized by complex neuroinflammatory responses. While natural compounds such as daphnetin have demonstrated anti-inflammatory and neuroprotective effects, their clinical application is hindered by low bioavailability. To address this limitation, we developed a biofabricated nanoliposomal drug delivery system, designed to enhance daphnetin's targeted delivery to macrophages and improve its therapeutic efficacy in SCI repair. A biofabrication-driven approach was employed to synthesize and optimize targeted liposomal nanocarriers (Da@Lip-CRV) using phospholipid-based vesicles conjugated with CRVLRSGSC peptides for macrophage specificity. Their targeting and therapeutic efficiency were evaluated in vitro and in vivo. The biofabricated liposomal system exhibited stable physicochemical properties and enhanced macrophage targeting efficiency. Transcriptomic analyses revealed that Da@Lip-CRV suppressed NF-κB-mediated inflammation, inhibited macrophage pyroptosis, and promoted M2 polarization, thereby facilitating an anti-inflammatory microenvironment conducive to neural repair. Da@Lip-CRV showed superior recovery of hindlimb function, reduced glial scarring, and improved neuronal survival in vivo. This study presents a biofabrication-based liposomal drug delivery platform for targeted SCI therapy. The findings demonstrated that optimize targeted liposomal nanocarriers (Da@Lip-CRV) regulated macrophage polarization and inhibited the pyroptosis of macrophage so as to provide a promising strategy for the clinical treatment of SCI.

Graphical abstract

1. Introduction

Spinal cord injury (SCI) is a traumatic disease of the central nervous system with a high incidence and severe symptoms in recent years [1]. Due to the lack of effective clinical treatments to alleviate the paraplegia and excretion dysfunction caused by SCI, the disease seriously affects the physical condition and quality of life of patients [2]. The pathological changes of SCI are divided into two stages: primary injury and secondary injury [3,4], especially secondary injury. When secondary injury occurs, a large number of neurons, axons and glial cells die pathologically, and the spinal cord tissue is further damaged, which in turn affects sensory and motor functions [5,6]. This change in the pathological environment of injury largely determined the severity of the final injury, and also greatly affects the repair function of treatment measures [7,8]. Macrophages and microglia are immune cells in the central nervous system and play an important role in mediating neuroinflammatory responses [9,10]. When the spinal cord is injured, they are activated, produce many cytokines, and participate in cell signaling pathways [11,12]. These signaling pathways constitute a powerful pro-inflammatory response system that amplifies inflammation and aggravates secondary injuries. Therefore, controlling the activation of macrophages, regulating the polarization of macrophages, and the resulting inflammatory factors is helpful in the treatment of SCI. In addition, cell pyroptosis, also known as cell inflammatory necrosis, is a type of programmed cell death that depends on GSDMD protein-mediated, manifested by cell swelling, plasma membrane dissolution, chromatin fragmentation, and the release of inflammatory cytokines such as IL-18 and IL-1β, which in turn triggers a severe inflammatory response [13,14].

Studies have shown that a variety of natural compounds have a regulatory effect on neuroinflammation after SCI, such as neuroprotective effects of safranal by anti-inflammatory and repair effect of Tetramethylpyrazine on spinal cord injury via Akt/Nrf2/HO-1 pathway [15,16]. Daphnetin (7,8-dihydroxycoumarin, Daph) is an aromatic compound isolated from the herbal Daphne vulgaris, Legumin derivatives, existing naturally or being synthesized artificially [17]. It contains an oxygen-containing heterocycle and a characteristic benzo-α-pyrone skeleton and is biosynthesized from L-tyrosine and L-phenylalanine via the shikimate pathway [18]. A large number of in vivo and in vitro studies had proven that daphnetin had strong safety [19]. Furthermore, after receiving a daphnetin injection, animals did not have any severe local side effects, such as erythema, erosion, irritation, or mucosal ulcers [20]. Numerous investigations have demonstrated the possible anti-inflammatory and anti-hypoxic effects of daphnetin, and they have also proposed potential clinical uses for it [21]. Daphnetin treatment was reported to considerably reduce the extent of cerebral infarction in findings on the neuroprotective effect of daphnetin on cerebral artery occlusion/reperfusion mice model [22]. A large number of animal and cell experiments had shown that daphnetin exerted antioxidant and anti-inflammatory effects by regulating a variety of proteins including Nrf-2 and HO-1 [23,24]. Studies had shown that daphnetin inhibited the activation of the TLR-4/NF-κB signaling pathway, reduced the infiltration and aggregation of immune cells in the damaged area, and created conditions for nerve regeneration [25]. Therefore, daphnetin may have a regulatory effect on the activation of macrophages and subsequent inflammatory responses through the NF-κB pathway, thereby improving related neuropathology. However, due to the low bioavailability of daphnetin, it often requires a higher dose to exert its effect, which affects clinical transformation. Therefore, it is urgent to improve its bioavailability through some means.

Liposomes are composed of natural phospholipids, sphingolipids, cholesterol, and hydrophilic polymers. The liposomes used in medicine mainly consists of phosphatidylcholine (PC) and a little amount of phosphatidylethanolamine (PE) present within them both, with neutral charges under physiological pH [26]. Liposomes can be monolayer or multilayer vesicles, depending on how they are prepared. The film hydration method is the simplest and oldest preparation method of liposomes [27]. The lipids are dissolved in an organic solvent and then a lipid film is formed at the bottom of the round-bottomed flask using a rotary evaporator. The lipid film is hydrated to form a liposome dispersion. Hydration conditions affect the structure of liposome vesicles. For example, mild hydration can form large single-layer vesicles, intense agitation can form multi-layer vesicles with poor dimensional uniformity, and probes or water bath ultrasound can form small single-layer vesicles. The diameter of the liposomes can be controlled by continuous extrusion through a polycarbonate filter with a limited aperture [28]. The number of extrusion cycles is important to determine the homogeneity of the liposomes formed. The liposome size can vary between 20 nm and 2 μm, and both the size and number of bilayers determine the amount of the encapsulated drug [29]. Liposome is an excellent multifunctional drug delivery platform, which can wrap hydrophilic drugs in the interior of the water and the hydrophobic drugs in the hydrocarbon chain of the lipid bilayer and Liposomes have been used in a variety of clinical trials to deliver anti-inflammatory, antibiotics, antifungals, anticancer drugs, anesthetics and other drugs, as well as in gene therapy. Liposomes serve as an important drug delivery system, but unmodified liposomes still have significant limitations, such as lack of targeting selectivity, short blood circulation times, and in vivo instability. Therefore, small molecular ligands, peptides or monoclonal antibodies are coupled to the surface of liposomes to make them targetable. The combination of liposomes and some small molecules of peptides can achieve specific functions, such as penetrating the blood-brain barrier; Some small molecular peptides can be targeted to specific cells, such as microglia [30,31]. After entering the body, liposomes are quickly cleared from the blood by phagocytes. The biocompatible inert polymer polyethylene glycol (PEG) was incorporated into the liposomes to solve this problem. At the same time, PEG improved the stability and cycle half-life of liposomes due to steric hindrance.

In this study, we constructed a drug delivery platform with liposome coupled PEG and conjugated small molecule peptides, which could cross the blood-brain barrier and targeted spinal cord macrophages (microglia). The lipid nanoparticles increased blood drug concentration in the spinal cord and improved the bioavailability of drug carriers. Specifically, liposomes were synthesized by thin film hydration, and targeted peptides ((TGNYKALHPHNG-GGGG-CRVLRSGSC), TGNYKALHPHNG peptide crossed the blood-brain barrier [32]; CRVLRSGSC peptide targeted macrophages [33]) conjugated to the surface of liposomes. Daphnetin was loaded into the liposome, abbreviated as Da@Lip-CRV. In vitro and in vivo experiments were conducted to test its safety and effectiveness in repairing SCI, and to further explore its repair mechanism, laying a foundation for clinical transformation (Fig. 1).

Fig. 1.

Graphic abstract illustrating the underlying mechanism of Da@Lip-CRV in promoting polarization of macrophage into M2 and regulating the immune microenvironment to repair spinal cord injury.

2. Result

2.1. Functional construction and performance verification of targeted liposomes loaded with daphnetin (Da@Lip-CRV)

As we mentioned above, the characteristics of the daphnetin and liposomes, so we focused on the synthesis, identification and functional verification of targeted liposomes loaded with daphnetin (Fig. 2A). Firstly, we needed to determine whether the synthesized liposomes were stable. After the synthesis, by measuring the change in potential, we found that the potential change of the liposomes corresponding to various targeting peptides was stable and was between −30 mV and −40 mV (Supplementary Fig. 1A). Then, we continued to test the particle size and found that the diameters of various liposomes were mainly concentrated between 100 nm and 200 nm, indicating that the performance of the synthesized liposomes were stable (Supplementary Fig. 1B). And the liposome morphology could be seen by transmission electron microscopy (TEM), proving that the material was successfully synthesized (Fig. 2B).

Fig. 2.

Functional construction and performance verification of targeted liposomes loaded with daphnetin. A, Schematic illustration of the Da@Lip-CRV fabrication. B, TEM test liposome morphology. C, Immunofluorescence staining with FITC-labeled target liposomes (green) and DAPI-labeled macrophage nucleus (blue) to detect the liposomes could successfully target macrophages. n = 6. Scale bar: 50 μm. D, In vivo imaging test targeting effect of the Da@Lip-CRV-CY5 at different time spots. E, Quantification of the CY5 signal intensities of the Da@Lip-CRV-CY5 at different time spots. F, Immunofluorescence staining with CY-5 labeled Da@Lip-CRV-CY5 (red) and GFP labeled macrophage (green) to detect the targeting effect of the Da@Lip-CRV-CY5 in vivo. n = 6. Scale bar: 50 μm. (G) Quantification of co-localization between Da@Lip-CRV-CY5 and macrophage. Data are shown as mean ± SD. ∗∗∗P < 0.001, ∗∗∗∗P < 0.0001, n. s. means no statistical significance.

In order to verify the targeting of liposomes to macrophages, we used fragment CRVLRSGSC as a macrophage-binding peptide [33], and fluorescein isothiocyanate (FITC)-labeled liposomes Da@Lip-CRV-FITC (50 μmol/L), Da@Lip-Ctrl-FITC (50 μmol/L) and Da@Lip-CRV-FITC (50 μmol/L) + CRV (125 μmol/L) were co-cultured with RAW264.7 for 4 h. By confocal microscopy, we could find that the co-localization of liposomes with RAW264.7 cell nuclei in the experimental group was significantly more. However, after adding more CRV, the co-localization in the experimental group was significantly reduced, indicating that the liposomes could successfully target macrophages by CRV (Fig. 2C). Next, we assessed the effect of targeting liposomes in vivo. We firstly bought transgenic mice from Shanghai Model Organisms Center, Inc. by using CRISPR/Cas9 technology, and the Kozak-Creert2-Ires-EGFP expression frame was knocked into the Lyz2 gene start code site by homologous recombination and long-fragment PCR results identified that F0 generation transgenic mice were obtained successfully (Supplementary Fig. 1C). We divided the transgenic mice Lyz2-e (Kozak-CreERT2-IRES-EGFP), which macrophage expressed GFP, into two groups that the experimental group was injected with Da@Lip-CRV-CY5 through the tail vein, and the control group was injected with Da@Lip-Ctrl-CY5. In vivo imaging was performed 1 h, 2 h, 4 h, 8 h, and 24 h after administration and the CY5 signal intensity of the spinal cord was detected. The result showed that the CY5 signal intensities of the experimental group at different time spots were significantly stronger than that of the control group, indicating that the targeting effect of the experimental group (Da@Lip-CRV-CY5) was significantly better than the control group (Fig. 2D and E). Then we found that the experimental group showed more obvious co-localization of liposomes (CY5) and macrophages (GFP) in spinal cord of transgenic mice, indicating that the experimental group successfully targeted macrophages in spinal cord and its targeting was significantly better than that of the control group (Fig. 2F and G).

In addition, we used CCK8 kit to test different concentrations of targeted liposomes to HT22 and RAW264.7, and we could see that there was no significant difference in the survival rate of cells, indicating that the targeted liposomes had little toxicity to nerve cells and macrophages and had a certain degree of biosafety (Fig. 3A). Then we continued to assess biosafety in vivo, and Da@Lip-CRV-FITC were administered by tail vein injection for 7 days, and there were no obvious abnormalities in the mice's spirit, activity, hair, food, water, and defecation between the control group and the liposomes group and continued to gain weight (Fig. 3B). At the same time, we also evaluated the safety of liposomes through histopathological staining and the results showed that there was not tissue necrosis significant damage occurred after injection of targeted liposomes, indicating that the constructed targeted liposomes were safe for animals (Fig. 3C).

Fig. 3.

The toxicity and biosafety of targeted liposomes. A, CCK8 assay test different concentrations of Da@Lip-CRV to HT22 and RAW264.7 viability. B, Test the mice weight after administering Da@Lip-CRV-FITC by tail vein injection for 7 days. n = 16. C, H&E staining test the biosafety of Da@Lip-CRV in vivo. Scale bar: 500 μm.Data are shown as mean ± SD. ∗P < 0.05,n.s. means no statistical significance.

2.2. Targeted liposomes loaded with daphnetin promotes the recovery of hindlimb motor function after SCI

Previous studies have shown that daphnetin has anti-inflammatory, oxygen free radical scavenging and neuroprotective effects [23,24]. But there are few studies on its role in the field of SCI. Firstly, we divided C57BL/6 mice after SCI into three groups, namely, the SCI group, the daphnetin group and the daphnetin-loaded targeted liposome group (Da@Lip-CRV). Daphnetin treatment group: inject the daphnetin solvent (60 mg/kg/d) for 7 consecutive days; Daphnetin-loaded liposome treatment group: the liposome loaded with daphnetin (5 mg/kg/d) for 7 consecutive days. We used BMS test to evaluate the recovery of hind limb motor function after SCI. One day after SCI, the lower limbs of three groups were unable to move, the hind limb muscles had no voluntary contraction ability, and the muscle strength score was 0, which proved that the SCI model of mice was successfully established. As time progressed, the scores of the daphnetin treatment group and the liposome loaded with daphnetin group gradually increased. The scores of the two groups increased to a certain extent, and the increase in the daphnetin-loaded targeted liposome group was more obvious (Fig. 4A). In addition, six weeks after SCI, the lower limb function of each group was analyzed by Catwalk gait analysis to evaluate their motor function recovery. We found that compared with the daphnetin group, the mice in the liposome loaded with daphnetin group had a larger stride and better coordination (Fig. 4B, Supplementary Fig. 2). After a contusion injury, the forelimb's stride length was dramatically reduced before recovering with therapy in the SCI group, and the liposome loaded with daphnetin group was more effective (Fig. 4C). Additionally, following the liposome loaded with daphnetin and daphnetin treatment, functional restoration made us possible to quantify a number of metrics, including stand duration and maximum contact area, which were 60–90 % higher than those of the uninjured controls (Fig. 4D and E). This indicated that the liposome loaded with daphnetin group could more effectively promote the recovery of the motor function of SCI.

Fig. 4.

Targeted liposomes loaded with daphnetin(Da@Lip-CRV) promotes the recovery of hindlimb motor function after SCI. A, Motor function of different groups of mice was monitored once a week for 6 weeks by using BMS score. n = 10. B, Motor function recovery of different groups of mice 6 weeks after contusion detected by Catwalk gait system and representative footprint images of different groups. C, Stride length of the hindlimbs. D, Stands time of the hindlimbs. E, Max contact area of the hindlimbs. n = 10. Data are shown as mean ± SD. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

In addition to causing severe paralysis of the hindlimb, SCI also can impact urinary system functions, resulting in the loss of sphincter function, urinary retention and so on. H&E staining of the bladder tissues showed that the thinning and damage of the bladder wall tissue of the liposome loaded with daphnetin group was significantly improved than the control group, although the daphnetin group also showed some effect on repairing SCI (Fig. 5A and B). This further confirmed that the liposome loaded with daphnetin group had a better effect on the repair of SCI. Then we co-stained NeuN and GFAP and found that the fluorescence intensity of NeuN in the liposome loaded with daphnetin group was significantly higher than the daphnetin group and SCI group, while the fluorescence intensity of GFAP was lower (Fig. 5C and D), indicating that the neuroprotective effect of the liposome loaded with daphnetin was stronger and could more strongly inhibit the activation of astrocytes, thereby inhibiting the formation of glial scars and providing a good environment for neuronal survival. At the same time, we further performed H&E staining and Nissl Staining and showed that smaller injury size after treatment with liposome loaded with daphnetin, indicating that liposome loaded with daphnetin could promote to repair injury (Supplementary Fig. 4A–B).

Fig. 5.

Targeted liposomes loaded with daphnetin(Da@Lip-CRV) promotes the recovery after SCI at the histological level. A, H&E staining test the bladder wall tissue of different groups. n = 6. Scale bar: 500 μm. B, Quantification of thickness of bladder wall in different groups. C, Immunofluorescence staining with NeuN-labeled neuron (green) and GFAP-labeled Astrocyte (red) were performed to detect the functional recovery of SCI mice in different groups. n = 6. Scale bar: 50 μm. Data are shown as mean ± SD. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

2.3. Daphnetin inhibit macrophage pyroptosis and improves neuroinflammation

SCI usually results from an exogenous trauma that causes primary injury, damaging neurons and glial cells, and initiating a cascade of secondary injury, leading to more cell death and further damage to the spinal cord in the following weeks. In order to further clarify the mechanism by which daphnetin acts on macrophages to affect SCI, we treated macrophages with LPS and daphnetin, then performed RNA transcriptomics analysis, and verified it through biological experiments. Firstly, we tested the optimal concentration of daphnetin by MTT and the result showed that the median lethal concentration (IC50) of daphnetin is 166.4 μmol/L (Fig. 6A) and 40 μmol/L was a more appropriate concentration of daphnetin, which could more effectively inhibit the death of macrophages (Fig. 6B). After mRNA transcriptomics analysis, we found that there were 1427 differently expression genes (DEGs) between the LPS group and the control group, of which 663 were upregulated and 764 were downregulated (Fig. 6C, Supplementary Table 1); there were 3882 DEGs between the Daph group and the Control group, of which 2084 were upregulated and 1798 were downregulated (Fig. 6D, Supplementary Table 2); there were 2513 DEGs between the Daph group and the LPS group, of which 1465 were upregulated and 1048 were downregulated (Fig. 6E, Supplementary Fig. 3A and B, Supplementary Table 3). Then we performed GO enrichment analysis of DEGs and found that compared with the LPS group and the Control group, the BP analysis found that the differential genes were mainly enriched in cytokine-mediated signaling pathways, inflammatory response, innate immune response, defense response regulation, and cytokine production regulation; the MF analysis found that the differential genes were mainly enriched in transmembrane cell receptor activity, signal receptor activity, cytokine activity, and cytokine receptor binding; the CC analysis found that the differential genes were mainly enriched in the cell surface, the outer region of the cytoplasmic membrane, the cell membrane, and the lysosome. Compared with the Daph group and the Control group, the BP analysis found that the differential genes were mainly enriched in DNA repair, cell chemotaxis, cell cycle regulation, DNA metabolism regulation, and GTPase activity regulation; the MF analysis found that the differential genes were mainly enriched in microtubule binding, oxidized DNA binding, ubiquitin-like protein ligase binding, and ubiquitin protein ligase binding; the CC analysis found that the differential genes were mainly enriched in centrosome, condensed chromosomes, chromosome regions, and microtubule organization center. Compared with the LPS group and Daph group, the BP analysis found that the differential genes were mainly enriched in inflammatory response, leukocyte chemotaxis, cell chemotaxis, T cell chemotaxis, cytokine regulation, defense response to other organs, and adaptive immune response; the MF analysis found that the differential genes were mainly enriched in chemokine activity, cytokine activity, cytokine receptor binding, and chemical factor binding; the CC analysis found that the differential genes were mainly enriched in the extracellular membrane region, cell membrane part, and centrosome (Fig. 6F, Supplementary Tables 4–6). The results of KEGG analysis indicated that the differential genes in this study were primarily enriched in multiple pathways, of which the following were the more significant ones: NF-κB signaling pathway, NOD-like receptor signaling pathway, TNF signaling pathway, and C-type lectin receptor signaling pathway, P53 signaling pathway, RIG-like receptor signaling pathway, and so on (Fig. 6G, Supplementary Tables 7–9). Of note is that some of these are associated with inflammation enriched pathways, such as NF-κB signaling pathway, NOD-like receptor signaling pathway, TNF signaling pathway. And protein interaction network showed that Nfkbib, Nfkb1, Tnf, and Gsdmd which were related to the NF-κB signaling pathway, NOD-like receptor signaling pathway, and TNF signaling pathway were significant (Fig. 6H).

Fig. 6.

Identification of critical genes and enrichment pathway of macrophage post Daphnetin treatment. A, MTT test the median lethal concentration (IC50) of daphnetin to macrophage. B, MTT test the optimal concentration of daphnetin. C-E, Volcano plot of DEGs between different groups. F, GO enrichment analysis of differentially expressed genes in different groups. G, KEGG analysis. H, Protein interaction network (PPI). Data are shown as mean ± SD. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

The above sequencing results showed that Daph may affect macrophages through the NF-κB signaling pathway and the NOD-like receptor signaling pathway, and these signaling pathways were related to inflammation and pyroptosis. So we performed live-death staining and the result showed that compared with the control group, the proportion of dead cells in the LPS group increased significantly, while the proportion of dead cells in the Daph group was less than that in the LPS group, indicating that Daph could inhibit pyroptosis of macrophages (Fig. 7A and B). Then the Western Blot showed that the protein expressions of Caspase-11, GSDMD, p-IκB and p-P65 were significantly increased in LPS group, but those proteins were decreased in Daph group (Fig. 7C and D). And considering the more complicated micro-environment in vivo, we further verified above results in animal experiments. Extracting protein from SCI in each group, the results of Western Blot showed that the expression of Caspase-11, GSDMD, p-IκB, and p-65 proteins significantly increased in SCI group, while these proteins showed a decrease after daphnetin and Da@Lip-CRV treatment (Fig. 7E and F). The above findings showed that Daph could significantly inhibit LPS induced macrophage pyroptosis, in which the NF-κB signaling pathway played a key role.

Fig. 7.

Verification that Daphnetin inhibits macrophage pyroptosis and improves neuroinflammation. A, Live-dead staining test pyroptosis of macrophages in different groups. Scale bar:50 μm. B, Quantification of proportion of live cells in different groups. C, Western Blot test the expressions of Caspase-11, GSDMD, p-IκB and p-P65 in different groups in vitro. D, Quantification of Caspase-11, GSDMD, p-IκB and p-P65 expression of the Western Blot images. E, Western Blot test the expressions of Caspase-11, GSDMD, p-IκB and p-P65 in different groups in vivo. F, Quantification of Caspase-11, GSDMD, p-IκB and p-P65 expression of the Western Blot images. Data are shown as mean ± SD. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

2.4. Targeted delivery of daphnetin-loaded liposomes and daphnetin regulate macrophage polarization and pyroptosis

Polarization of macrophage is a very important process of repairing injury, and we had verified that daphnetin could significantly inhibit macrophage pyroptosis, so we needed to explore if daphnetin could regulate polarization of macrophage. Firstly, we used Flow Cytometry to test the polarization of macrophage after daphnetin treatment and we found that daphnetin could reduce the proportion of M1 macrophages induced by LPS and increased M2, indicating that daphnetin promoted the polarization of macrophage into M2 (Fig. 8A). Then we further test the effect of daphnetin on polarization of bone marrow-derived macrophages (BMDMs) by Immunofluorescence staining and the result showed that compared with the LPS group, the number of M1 (iNOS+) cells was decreased and the number of M2 (Arg-1+) cells was increased in the daphnetin group, indicating that daphnetin promoted the polarization of macrophage into M2 (Fig. 8B–D).

Fig. 8.

Targeted delivery of daphnetin-loaded liposomes(Da@Lip-CRV) and daphnetin regulate macrophage polarization and pyroptosis. A, Flow Cytometry to test the polarization of macrophage. B, Immunofluorescence staining with iNOS labeled M1 macrophage (red) and F4/80 labeled M0 macrophage (green) were performed to detect the polarization of macrophage in vitro in different groups. n = 6. Scale bar: 50 μm. C, Quantification of ratio of the iNOS + cells in the F4/80+ M0 macrophage. H, Immunofluorescence staining with Arg-1 labeled M2 macrophage (red) and F4/80 labeled M0 macrophage (green) were performed to detect the polarization of macrophage in vitro in different groups. n = 6. Scale bar: 50 μm. D, Quantification of ratio of the Arg-1+ cells in the F4/80+ M0 macrophage. F, Immunofluorescence staining with iNOS labeled M1 Microglia (green) and Iba-1 labeled all Microglia (red) were performed to detect the polarization of Microglia in different groups. n = 6. Scale bar: 50 μm. G, Quantification of ratio of the iNOS + cells in the Iba-1+ Microglias. H, Immunofluorescence staining with Arg-1 labeled M2 Microglia (green) and Iba-1 labeled all Microglia (red) were performed to detect the polarization of Microglia in different groups. n = 6. Scale bar: 50 μm. I, Quantification of ratio of the Arg-1+ cells in the Iba-1+ Microglias. Data are shown as mean ± SD. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

In addition, we further used Immunofluorescence staining to test the effect of Da@Lip-CRV on polarization of Microglia in vivo and we found that compared with the daphnetin group, the number of iNOS(+) Microglia was decreased and the number of Arg-1(+) Microglia was increased in the Da@Lip-CRV group, indicating that the liposome loaded with daphnetin could better promote the polarization of Microglia into M2 subtype, although daphnetin also promoted the polarization of Microglia into M2 compared to control group (Fig. 8F–I). The above results also proved that liposomes loaded with daphnetin could better promote the polarization of Microglia into M2 type.

3. Discussion

SCI is usually caused by exogenous trauma, which damages neurons and glial cells and initiates a cascade reaction to cause secondary injury, leading to further cell death in the following weeks. In the initial phases of injury, a strong inflammatory reaction combined with the breakdown of the blood-spinal cord barrier makes it easier for blood molecules and cells to pass through the barrier and into the damaged parenchyma, exacerbating spinal cord swelling and damage. After SCI, inflammation becomes more complicated and involves a variety of cells and inflammatory cytokines. This influences the inflammatory milieu of the injured spinal cord and hinders the regeneration of new nerve tissue [34].

Numerous natural substances have been found to have a regulating effect on neuroinflammation following SCI. Salvianolic acid A, which inhibits the miR-101/CUL3/Nrf2/HO-1 signaling pathway to restore the blood-spinal cord barrier, is one example of how some of them function by acting on the blood-spinal cord barrier [35]. Salvianolic acid B may increase the expression levels of tight junction proteins (ZO-1, occludin), and HO-1 while decreasing the permeability of the blood-spinal cord barrier and the water content of the spinal cord tissue via the MAPK pathway [36]. Protocatechuic acid has been shown to decrease matrix metalloproteinase-9 expression, enhance the expression of tight junction proteins (ZO-1, occludin), and decrease bleeding and exudation in spinal cord tissue [37]. Ligustrazine can dramatically increase the expression of anti-inflammatory factors (IL-10) while considerably inhibiting the expression of pro-inflammatory factors (TNF-α, IL-1β, and IL-18) and NF-κB [38,39]. Additionally, ligustrazine can increase the polarization of microglia from an M1 to an M2 phenotype, inhibit the NF-κB pathway, activate the STAT3/SOCS3 pathway, and so reduce inflammation following SCI [40]. In this study, we explored the effect of daphnetin on anti-inflammation and repair SCI, and we used a targeted liposomal nanocarriers system to enhance the effect of daphnetin.

In this study, we used daphnetin, a natural compound, coumarin derivatives isolated from Daphne koreana to regulate the inflammation activity to repair SCI [41,42]. Daphnetin interacts with different cell mediators to regulate related neural signaling pathways, involving antioxidant systems, reactive oxygen species (ROS), heat shock protein 70, neurotransmitters, JAK-STAT pathways, and TLR-4/NF-κB pathways to exert neuroprotective effects. Studies have shown that daphnetin has a protective effect on ischemia/reperfusion injury, impaired spatial memory caused by chronic unpredictable stress, NMDA-induced toxicity, and HT-22 cells and RAW264.7 cells after glutamate stimulation [25,43]. The Nrf-2/HO-1 signaling pathway plays an inhibitory role in the process of neuronal damage and apoptosis induced by oxidative stress and inflammatory stress [44]. A large number of studies also have shown that daphnetin plays a role in the pathophysiological process of neurodegenerative diseases by affecting Nrf-2 and its downstream mediators [23,45]. Furthermore, a number of investigations have shown that daphnetin decreased the expression of proinflammatory cytokines, Bax, and caspase-3 while increasing the expression of HSP-70, which could stop Parkinson's disease-related protein aggregation and neuronal death by preventing the activation of pro-apoptotic caspase-3 and caspase-9 [23,24]. In our study, we found some important and creative results by RNA sequencing that Daphnetin could inhibit macrophage pyroptosis and improve neuroinflammation by inhibiting NF-κB signaling pathway and the NOD-like receptor signaling pathway. Daphnetin reduced GSDMD protein by inhibiting caspase-1 activation, thereby inhibiting macrophages pyroptosis and reducing the secretion of pro-inflammatory cytokines, thereby improving the inflammatory environment in the damaged area, which is beneficial for nerve regeneration. Then we verified the result by different experiments in vitro and in vivo, such as Western Blot, which gave us a new view to treat SCI.

Following SCI, inflammatory signals and DAMPs activate astrocytes, which in turn control the expression of adhesion molecules like ICAM-1 and VCAM-1 in the blood-brain barrier/blood-stem cell barrier through released inflammatory signals [46,47]. This increases the expression of chemokines in and around the lesion site, which attracts lymphocyte infiltration and monocyte/microglia migration [48,49]. In order to control T cell differentiation and the ratio of Th1/Th2 in neuroinflammation, astrocytes can also emit inflammatory signals (INF-γ, IL-12, and IL-10) [50]. Together with microglia, they form an inflammatory microenvironment. After early astrocytes are activated, GFAP expression increases, and a large amount of extracellular matrix is produced to form glial scars, which play a neuroprotective role in the early stage and are conducive to neuronal survival. In the later stage, as the inflammatory environment further deteriorates, astrocytes overproliferate to form glial scars, build physical barriers and affect nerve regeneration. On the other hand, Galectin-9 secreted by astrocytes can promote microglia to secrete tumor necrosis factor, aggravate the inflammatory response and lead to nerve cell death [51]. Therefore, in the study we found that daphnetin could reduce astrocyte activation and expression of GFAP, reduce extracellular matrix secretion, reduce the formation of physical barriers that hinder nerve regeneration, and reduce the release of inflammatory-related cytokines of astrocytes and improve the inflammatory microenvironment formed by interaction with microglia, creating a relatively favorable environment for nerve regeneration. The more important point was daphnetin could increase the expression of Neun and protected the neuron to repair SCI.

Previous studies have shown that macrophages have different phenotypes after SCI, with pro-inflammatory (M1) and anti-inflammatory (M2) properties [52]. In the early stages of SCI, M1 type macrophages account for the majority [53], and the level of M2 is significantly increased in spinal cord tissue 2 weeks after SCI. During SCI, in addition to phagocytosis, polarized macrophages can also activate astrocytes [54], oligodendrocytes [55], and demyelinate [55,56]. A large number of studies have shown that secondary pathological processes of SCI, such as demyelination and nerve cell death, are closely related to macrophages [57,58]. During SCI, M1 macrophages clear foreign matter from the injury site by increasing phagocytosis and releasing pro-inflammatory cytokines, while M2 macrophages reduce the production of inflammatory cytokines and reactive oxygen species (ROS) and show tissue repair properties [59,60]. M1 macrophages have been found to directly mediate neuronal death in vitro studies [61]. Meanwhile, studies have found that M1 macrophages contain 17 times more chondroitin sulfate protein than M2, suggesting that M1 can also inhibit nerve regeneration after SCI [62,63]. In this study, we found that macrophages polarize to the M1 type after SCI, and daphnetin intervention could significantly inhibit the polarization of macrophages to the M1 type and induce them to polarize to the M2 type, thus leading to the secretion of inflammation-related cytokines reduced, and the secretion of anti-inflammatory related cytokines increased, which improved the local inflammatory microenvironment in the injured area and created a good environment for nerve regeneration. In addition, RNA sequencing revealed that the daphnetin might be related to the inhibition of macrophage pyroptosis by NF-κB signaling pathway to treat SCI.

As a drug delivery system, lipid nanoparticles have the advantages of simple synthesis and self-assembly, high biocompatibility and bioavailability, high payload, and adjustable biological properties [64,65]. Liposomes are usually constructed from phospholipids into a single or multilayer vesicle structure, which can load and transport hydrophilic, hydrophobic, and lipophilic drugs. By loading different types of compounds in the same system, the indications can be increased [66]. Liposomes are mostly absorbed by the reticuloendothelial system. After Surface pegylation, the circulation time and delivery capacity can be increased, which is more conducive to clinical application.

Due to the physical and biological barriers such as shear force, protein adsorption and rapid clearance in the body, it is difficult for liposomes to effectively achieve biodistribution and drug transport under physiological conditions, which limits the proportion of liposomes that reach the target treatment site. Under pathological conditions, these barriers change, making the drug transport efficiency of nanomaterials even worse [67,68]. In response to this situation, more and more scholars have optimized the design of nanoparticles. By using more complex technologies to improve their penetration, nanoparticles can circumvent these obstacles and achieve the purpose of drug delivery, but some limitations have also emerged. The physicochemical properties of nanoparticles, such as size, shape, charge, and surface modification, as well as the blood flow, secretion, phagocytic cells in the circulatory system, and the interaction between nanoparticles and local barriers can affect their stability and delivery efficiency. The blood-spinal cord barrier is a highly selective semipermeable membrane that separates the circulating blood from the spinal cord, preventing most molecules in the circulation from passing through, thereby maintaining the dynamic balance of the central nervous system. Due to the highly selective nature of the blood-spinal cord barrier, only a small fraction of therapeutic drugs can reach brain. In this study, by simultaneously adding a targeting peptide that can penetrate the blood spinal cord barrier and a macrophage targeting peptide to the liposome, it is easier to penetrate the blood spinal cord barrier and improve the specific distribution in SCI lesions, and then target macrophages in the lesions, significantly improving the neuroprotective effect of the loaded daphnetin.

4. Conclusion

Daphnetin exhibits neuroprotective properties by inhibiting macrophage pyroptosis and promoting macrophage polarization toward the M2 phenotype, with the NF-κB signaling pathway playing a crucial role in this process. Targeted liposomes encapsulating daphnetin effectively cross the blood-spinal cord barrier and selectively target activated macrophages following SCI. This targeted delivery system reduces the expression of pro-inflammatory factors while enhancing the expression of anti-inflammatory mediators, thereby significantly modulating the immune microenvironment in the injured spinal cord. Consequently, this approach promotes motor and neurological function recovery in SCI-affected mice, highlighting its potential as a promising therapeutic strategy for SCI treatment.

5. Materials and methods

5.1. Construction of targeted liposomes

The customized targeting peptide TGNYKALHPHNG-GGGG-CRVLRSGSC was used in the treatment group, where TGNYKALHPHNG functions as a blood-spinal cord barrier-penetrating peptide [32], and CRVLRSGSC serves as a macrophage-binding peptide [33]. To facilitate tracking, fluorescently labeled peptides were utilized.

• 1)FITC-TGNYKALHPHNG-GGGG-CRVLRSGSC (FITC-labeled, green fluorescence) was used for immunofluorescence co-localization in RAW264.7 macrophages to assess in vitro targeting.
• 2)FITC-TGNYKALHPHNG-GGGG-GGSGGSKG served as the control peptide for in vitro targeting validation.
• 3)CY5-TGNYKALHPHNG-GGGG-CRVLRSGSC (Cy5-labeled, red fluorescence) was used for in vivo tracking of liposomes in C57 mice through imaging and immunofluorescence microscopy of transgenic mouse tissue sections.
• 4)CY5-TGNYKALHPHNG-GGGG-GGSGGSKG functioned as the control for in vivo targeting validation.

Both the targeted peptide (CRV: TGNYKALHPHNG-GGGG-CRVLRSGSC) and the control peptide (Ctrl: TGNYKALHPHNG-GGGG-GGSGGSKG) were synthesized by GL Biochem (Shanghai) Ltd.

5.2. Preparation and characterization of daphnetin-loaded lipid nanovesicles

Lipid nanovesicles were prepared using the thin-film hydration method. A lipid mixture consisting of 65 mol% phosphatidylcholine (PC), 30 mol% cholesterol (Ch), and 5 mol% DSPE-PEG2000-NHS was dissolved in a solvent mixture of chloroform (0.4 mL) and methanol (0.1 mL) in a round-bottom flask. The organic solvent was removed by vacuum evaporation at 55 °C, forming a thin lipid film at the bottom of the flask. The lipid film was then hydrated with 1 mL PBS containing 1 mg of daphnetin and subjected to ultrasonic vibration in a water bath for 30 min to facilitate vesicle formation. The resulting nanovesicle suspension was filtered through a 0.22 μm pore size membrane for purification.

To functionalize the lipid nanovesicles, CRV or Ctrl peptides were added to 1 mL of the purified solution and incubated with stirring at 4 °C for 12 h, resulting in the formation of CRV- or Ctrl-conjugated lipid nanocapsules (Da@Lip-CRV and Da@Lip-Ctrl, respectively). Further purification was performed using ultrafiltration (Merck, MWCO: 30 kDa).

The morphology and size distribution of the lipid nanovesicles were characterized using Transmission Electron Microscopy (TEM, HHTC HT7800), while Dynamic Light Scattering (DLS, NICOMP 380/ZLS (PSS)) and ZPW388 software were used to analyze the average particle size and zeta potential.

5.3. Cell culture

The RAW264.7 mouse macrophages cell line and the HT22 mouse hippocampal neuronal cell line were rocured from the American Type Culture Collection (ATCC). The RAW264.7 and HT22 were cultured in Dulbecco's Modified Eagle Medium (DMEM) (Gibco; cat no., 11965092) supplemented with 10 % fetal bovine serum (FBS) (Gibco; cat no., A5256701) and 1 % penicillin-streptomycin (Solarbio; cat no., P1400) at 37 °C in a 5 % CO₂ incubator. When the cell density reached 80 % confluency, cells were passaged. The old culture medium was discarded, and the dish was washed three times with sterile phosphate-buffered saline (PBS, pH 7.4). Fresh culture medium (2 mL) was added, and cells were detached by gentle pipetting. The suspension was evenly distributed across four new culture dishes and returned to the incubator.

5.4. RNA extraction and detection

Cells were lysed using TRIzol reagent (1 mL per 10-cm dish) and incubated on ice for 5 min. The lysate was mixed with chloroform (1:5 ratio to TRIzol), vortexed for 15 s, and incubated on ice for 5 min. Samples were centrifuged at 12,000×g for 15 min at 4 °C, and the aqueous phase was transferred to a new 1.5-mL tube. RNA was precipitated with isopropanol (1:1 ratio), incubated at room temperature for 10 min, and centrifuged at 12,000×g for 10 min at 4 °C. The RNA pellet was washed twice with precooled 75 % ethanol, centrifuged at 7500 rpm for 5 min at 4 °C, and air-dried. The pellet was resuspended in RNase-free water and incubated at 37 °C for 10 min before measuring the RNA concentration using Nanodrop. The 260/280 value is between 1.8 and 2.0; the 260/230 value is better if it is above 2.0. The qualified samples were stored in a −80 °C refrigerator. The total amount and integrity of RNA were tested by bioanalyzer.

5.5. MTT test

The MTT assay was performed to assess cell viability and cytotoxicity based on the metabolic activity of viable cells. Mitochondrial dehydrogenases reduce 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) to formazan, an insoluble purple crystalline product. The amount of formazan formed is directly proportional to the number of metabolically active cells and was quantified by measuring absorbance at 490 nm using a microplate reader (SpectraMax iD5, Molecular Devices, USA).

Macrophages were seeded in 96-well plates at a density of 4 × 10³ cells per well in 100 μL of complete culture medium and incubated overnight at 37 °C in a 5 % CO₂ atmosphere to allow cell adhesion. To minimize edge effects, PBS was added to the outermost wells. Cells were then treated with DMSO, LPS, or LPS + Daph at defined concentrations and incubated for 24 h. Following treatment, the culture medium was removed, and 90 μL of fresh medium containing 10 μL of MTT solution (5 mg/mL in PBS) was added to each well. After 4 h of incubation, the medium was aspirated, and 100 μL of dimethyl sulfoxide (DMSO) was added to dissolve the formazan crystals. Absorbance at 490 nm was recorded, and cell viability was expressed as a percentage relative to untreated controls. All experiments were performed in triplicate.

5.6. Calcein-AM/PI staining

The live-dead staining kit was equilibrated to room temperature before use. A working solution was prepared by mixing 30 μL of 1.5 mmol/L PI and 5 μL of 4 mmol/L Calcein-AM with 10 mL of DPBS, followed by vortexing to ensure homogeneity. Meanwhile, the culture medium was aspirated, and adherent cells were trypsinized to achieve a single-cell suspension. After detachment, cells were resuspended in serum-containing medium, combined with the collected supernatant, and centrifuged at 1000 rpm for 5 min. The supernatant was discarded, and the cell pellet was washed twice with PBS to remove residual esterase activity. Cells were then resuspended in the prepared staining solution at a density of 1–5 × 10⁵ cells/mL, protected from light by covering with aluminum foil, and incubated at room temperature for 15 min. Following incubation, 10 μL of the stained cell suspension was transferred onto a slide and immediately visualized under a fluorescence microscope (Leica DMi8, Leica Microsystems, Germany). The number of viable (green) and dead (red) cells was quantified to calculate the live/dead cell ratio.

5.7. Extraction of BMDMs

Healthy C57BL/6 mice were selected, anesthetized, and then immersed in 75 % ethanol for disinfection. Wearing sterile gloves, use sterile scissors to cut the skin of the mouse legs, remove the muscles, remove the femurs and tibias on both sides, open the ends of the femurs and tibias, use RPMI1640 culture medium to flush the bone marrow cavity, and then filter the collected flushing fluid through a sterile sieve. After collecting the filtrate, centrifuge it, add 3 mL of red blood cell lysis solution to the precipitate with a pipette, mix it thoroughly, let it stand at room temperature for 3 min, and neutralize it with 3 mL of RPMI-1640 culture medium containing 10 % serum. Then rinse and centrifuge to discard the supernatant, and finally resuspend the precipitated cells with culture medium, inoculate them in a culture dish, place them in a cell culture incubator for 16 h, collect the cells that are not attached to the wall, and re-inoculate them in the well plate. RPMI-1640 culture medium containing M-CSF was added to the well plate for induction and then examined by flow cytometry.

5.8. Mice SCI model

For this investigation, 8–10 week old female C57BL/6 mice were employed. Vital River Laboratory Animal Technology Co., Ltd. was the source of the mice. The climate in which the animals were housed had a 12/12-h light-dark cycle, a temperature of 20–25 °C, and a humidity level of 40–60 %. Ad libitum food and water were provided to the mice. Weigh the mouse, turn on the gas anesthesia system, take enough isoflurane and pour it into the matching anesthesia syringe, and set up the gas anesthesia system. Set the weight in the system according to the mouse weight result, turn the three-way valve to face the induction anesthesia box, select the induction anesthesia in the system, set the flow rate to 2.5, and then prefill the induction anesthesia box. Set up the system and suck back the excess isoflurane in the induction anesthesia box. Open the lid of the induction anesthesia box and take out the mouse, place it in a prone position on the heating blanket, and fix its nose on the gas delivery nose mask. At the same time, turn the three-way valve to the direction of the nose mask, set the system to maintain anesthesia through the nose mask, set the flow rate to 1.5, and judge the breathing status of the mouse, and pay attention to adjust the anesthesia flow rate at any time. If the mouse breathes deeply and slowly, the flow rate needs to be reduced. If the mouse breathes shallowly and quickly, the flow rate needs to be increased until the breathing is regular and stable. Pinch the mouse's tail with the thumb and index finger of the right hand. If there is no reflex, it means that the anesthesia is successful, and further surgery can be performed. The impact parameters of Impactor Model III were set in advance, with the impact weight of 5 g and the height of 12.5 mm. Before modeling, the mouse eyes were covered with wet wipes to prevent eye infection. The surgeon wore disposable surgical gowns, surgical caps, masks and sterile gloves. After skin preparation with an electric hair clipper, hair removal was performed with depilatory cream. The skin preparation range was from the neck to the lower back, 1 cm on both sides of the midline. Then it was disinfected with iodine and covered with sterile drapes. The center of the operation was located at the eighth thoracic vertebra (T8). With the T8 spinous process as the center, the skin was cut layer by layer (about 1 cm long) along the longitudinal axis of the body with a scalpel, and sterile cotton balls were used to stop bleeding. With one hand, the mouse spine was fixed with toothed forceps, and with the other hand, the scalpel was used to cut along both sides of the spine and separate the muscles as much as possible, and then the T8 lamina on both sides were cut with ophthalmic scissors and the spinal canal was opened to expose the spinal cord. After the exposure is completed, the mouse is transferred to the operating table, and the EP tube soaked with isoflurane gauze is placed on the mouth and nose of the mouse to temporarily maintain anesthesia. Pay attention to adjusting the distance between the EP tube and the nose to prevent awakening or respiratory depression. Use the mechanical arm of the operating table to fix both ends of the spine in the injured area of the mouse. Then lower the impact head until the tip touches the spinal cord but without any pressure. At this time, reverse the rotation of the position adapter to raise the tip of the impact head. Press the impact button to impact and contusion the mouse, and record the impact force, displacement, speed and other values at the same time. The manifestation of a successful operation is spasm of both hind limbs. After local iodine disinfection, suture the injured area layer by layer according to the anatomical level and pay attention to sufficient hemostasis. At the same time, apply antibiotic ointment on the muscle. After suturing the skin with sutures, apply antibiotic ointment on the skin.

After surgery, the mice were sent back to the animal room for feeding, and each mouse was placed in a separate cage. Since the mice lost their appetite for mouse food and water after surgery, they should be given high-protein fluids and appropriate intraperitoneal fluid replacement. SCI in mice can cause urinary retention, and artificial assisted urination is required after surgery, three times a day, with an interval of 8 h, until the mice recover their autonomous urination function. If hematuria is found during assisted urination, the possibility of urinary tract infection should be considered, and penicillin anti-infection treatment is required.

5.9. Behavioral analysis

• (1)Basso Mouse Scale (BMS)

Using the recommendations of the Basso Mouse Scale, BMS scores were conducted prior to SCI, on day 1, 7, and then weekly to six weeks following SCI in order to assess the recovery of hindlimb locomotory behavior in mice. Prior to testing, mice were allowed to become used to the circular open field with a diameter of 1 m. A transparent glass plane encircled the circular open area, enabling viewers to watch mice's hindlimb locomotory activities. Two observers who were blind to the experimental groups conducted the experiments. Mice from several experimental groups were assessed and graded at random. Two researchers each had a single mouse in the wide field, and they watched it for 4 min. Each BMS score, which varied from 0 to 9, indicated a distinct level of hindlimb locomotory behavior. For instance, a hindlimb that did not move spontaneously received a score of 0, but a hindlimb that moved normally received a score of 9. Multiple hindlimb movement criteria, including joint movements, paw position, stepping pattern, coordination, trunk stability, and tail control, were taken into account while calculating the ratings.

• (2)Catwalk gait analysis

As previously mentioned, mice were used to objectively assess their footprints, locomotory behavior, and physical coordination using the CatWalk XT system and CatWalk XT software (version 10.6, Noldus, Wageningen, the Netherlands). The CatWalk XT system is made up of a high-speed camera that records actual footsteps and a transparent glass plate that is lit by a green LED light on a walking platform. The red LED background at the top of the walking platform allows users to see the mouse's body contour. The CatWalk tests were conducted in a silent, dark setting. Each mouse was trained to traverse the illuminated glass walkway three times or more prior to testing. Each mouse's footprint was captured by blowing glass with a digital video camera, and the CatWalk XT program examined the footage. A number of metrics, such as swing duration, base of support, stride length, stands, maximum contact area, and average speed, were measured and examined using this technique. The length of time the paw is not in touch with the glass plate is known as the swing duration. Base of support: The distance between the two front or hind paws measured perpendicularly called the base of support. The distance that the same paw traverses throughout a step cycle is known as the stride length. Stands: The length of time a paw makes contact with the glass plate. Average speed: Speed of the animal's body in the recorded run.

5.10. Perfusion and sectioning

The thoracic and abdominal cavities of the mice were opened under the xiphoid process, along both costal edges, after they had received a thorough anesthesia. First, pre-cooled PBS was transcardially perfused. Next, pre-cooled 4 % paraformaldehyde was transcardially perfused after the liver's color turned pale and the right atrial appendage's outflow fluid became clear. The spinal cord is readily visible when the bilateral lamina is severed with scalpel scissors. After being removed, the spinal cord was erect and preserved for an overnight period at 4 °C in 4 % paraformaldehyde. A sucrose solution with a concentration gradient was used to dry the tissue. The dehydrated spinal cords were sectioned at a thickness of 6 μm using a cryotome (Leica, CM3050S, Germany) after being frozen in OCT mounting solution.

5.11. Sequencing data quality assessment and correlation between samples

After sequencing is completed, since there may be machine errors in the sequencing process itself, the sequencing error rate distribution check can reflect the quality of the sequencing data. The base quality value of Illumina is represented by Qphred. The evaluation method includes sequencing error rate distribution, sequencing data filtering and base content distribution. The clean read used for subsequent analysis is obtained. The specific results are shown in Fig. 8. The results show that the base sequencing misalignment rate is 0.03 %, which meets the standard (Supplementary Fig. 3A and Supplementary Tables 1–3). The base content distribution can be used to determine whether there is base separation. The AT and GC curves have a good balance and a high degree of overlap, and are generally horizontal, indicating that the entire sequencing process is stable. An important indicator for testing the reliability of the experiment and whether the sample selection is reasonable is the correlation of gene expression levels between samples. The higher the similarity between samples, the closer the correlation coefficient is to 1. Ideally, it is recommended that the square of the Pearson correlation coefficient (R2) be greater than 0.92, and in actual operation, it is required to be at least 0.8. The correlation coefficients of samples within and between groups are calculated with reference to the FPKM values of all genes in each sample and drawn into a heat map to intuitively display the repetition of samples within the group and the difference between samples between groups. The square of the Pearson correlation coefficient (R2) in this study is above 0.85, indicating that the gene expression correlation between biological replicate samples in this subject is high (Supplementary Fig. 3B).

5.12. Statistics analysis

Statistical results were determined based on GraphPad Prism (Windows 8.02, GraphPad, San Diego, CA, USA). The experimental data was presented as mean ± standard deviation (SD). For in vitro model, the unpaired T test was used to calculate significance results. For vivo experiments, significant differences were determined by the one-way ANOVA with Tukey's post-hoc test. The criterion of P < 0.05 was considered to be statistically significant.

CRediT authorship contribution statement

Jun Shang: Writing – original draft, Visualization, Funding acquisition, Data curation, Conceptualization. Linhong Liu: Writing – original draft, Visualization, Software, Data curation. Jianping Zhang: Writing – original draft, Visualization, Software, Resources, Funding acquisition, Data curation. Bin Li: Resources, Funding acquisition. Pingping Zhang: Visualization, Funding acquisition, Data curation. Haifeng Wang: Funding acquisition, Data curation. Jincheng Lu: Visualization, Software, Data curation. Jinhong Fan: Validation, Data curation. Xinao Li: Methodology, Data curation. Zhenye Yan: Resources, Data curation. Huishuang Zou: Resources. Runtian Jiang: Software. Farnaz Ghorbani: Writing – review & editing, Validation. Chaozong Liu: Writing – review & editing, Supervision, Funding acquisition. Dachuan Wang: Writing – review & editing, Validation, Supervision, Project administration. Xiangyu Wang: Writing – review & editing, Validation, Supervision, Conceptualization.

Ethics statement

The animal study was reviewed and approved by the Animal Ethical and Welfare Committee (AEWC) of Shandong University Cheeloo College of Medicine (approval number: 24043) and performed according to the national guidelines for laboratory animal use and care. All methods were performed in accordance with the relevant guidelines and regulations.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

Data will be made available on request.