CD73 alleviates GSDMD-mediated pyroptosis in spinal cord injury through PI3K/AKT/Foxo1 signaling

Background Neuroinflammation-induced secondary injury is an important cause of sustained progression of spinal cord injury. Inflammatory programmed cell death pyroptosis executed by the pore-forming protein gasdermin D (GSDMD) is an essential step of neuroinflammation. However, it is unclear whether CD73, a widely accepted immunosuppressive molecule, can inhibit pyroptosis via mediating GSDMD. Methods CD73 deficient mice and LPS induced BV2 cells were respectively used to illustrate the effect of CD73 on macrophages/microglia pyroptosis in vivo and in vitro. A combination of molecular and histological methods was performed to assess pyroptosis and explore the mechanism both in vivo and in vitro. We showed molecular evidence for CD73 suppresses the activation of NLRP3 inflammasome complexes to reduce the maturation of GSDMD, leading to decreased pyroptosis in macrophages/microglia. Further analysis reveals that adenosine-A 2B adenosine receptor-PI3K-AKT-Foxo1 cascade is a possible mechanism of CD73 regulation. Importantly, we determine that CD73 inhibits the expression of GSDMD at the transcriptional level through Foxo1. What’s more, we confirm the accumulation of HIF-1α promotes the overexpression of CD73 after SCI, and the increased CD73 in turn upregulates the expression of HIF-1α, eventually forming a positive feedback regulatory loop. plasmid. BV2 cells were cotransfected with luciferase reporter plasmids, pRL-TK reporter plasmid (control reporter), and foxo1 plasmid (pc-foxo1). All these results suggest that A 2B AR is necessary for CD73 to regulate microglia pyroptosis.


Results
We showed molecular evidence for CD73 suppresses the activation of NLRP3 inflammasome complexes to reduce the maturation of GSDMD, leading to decreased pyroptosis in macrophages/microglia. Further analysis reveals that adenosine-A 2B adenosine receptor-PI3K-AKT-Foxo1 cascade is a possible mechanism of CD73 regulation. Importantly, we determine that CD73 inhibits the expression of GSDMD at the transcriptional level through Foxo1. What's more, we confirm the accumulation of HIF-1α promotes the overexpression of CD73 after SCI, and the increased CD73 in turn upregulates the expression of HIF-1α, eventually forming a positive feedback regulatory loop.

Conclusion
Our data reveal a novel function of CD73 on microphages/microglia pyroptosis, 4 suggesting a unique therapeutic opportunity for mitigating the disease process in SCI.

Background
Spinal cord injury (SCI) remains a devastating condition affecting millions of people worldwide, leading to severe dysfunction below injured segment [1]. About 240,000-337,000 people live with SCI in United States, and this figure is thought to grow by 17,000 annually [1]. Nearly 52% of these patients are paraplegic and 47% are considered quadriplegic [2]. Presently, no effective pharmacological therapy exists.
The pathogenesis of SCI is complicated and can be categorized into two phases [3].
After a transient initial mechanically inflicted trauma, a long-lasting second phase injury happens characterized in part by secretion of cytokines and chemokines produced at the lesion site, contributing largely to neurological damage [4]. Existing evidence demonstrates neuroinflammation exerts a key role in the secondary phase of SCI[5] , [6] , [7].
Traumatic injury to central nervous system (CNS) leads to the disruption of blood spinal barrier, triggering the invasion of cells and other components of immune system cause axonal destruction, neuronal loss and demyelination. Previous studies suggest activation of cytoplasmic inflammasome complex is essential for neuroinflammation post CNS trauma [8] , [9]. Inflammasomes are cytosolic multiprotein scaffolds assembled by particular pattern recognition receptors (PRRs), sensors of pathogen-associated molecular patterns (PAMPs) and damage-associated molecular patterns (DAMPs), enables the activation of pro-inflammatory caspase [10]. Presently, a number of inflammasome-associated sensors have been 5 discovered, such as NLRP1, NLRP3, AIM2 and pyrin [11]. Assembly of sensor proteins, scaffolding protein ASC and pro-inflammatory caspase (caspase-1 and -4/5 in humans and caspase-1 and -11 in mice) into inflammasome promotes autoactivation of caspase and subsequent proteolytic gasdermin D (GSDMD) cleavage, resulting in cell pyroptosis [10].
Pyroptosis is proinflammatory form of programmed cell death that relies on the activity of cytosolic GSDMD driven by inflammasomes [12]. Upon activation, GSDMD transfers to the plasma membrane and binds to the inner membrane lipids, oligomerizing to form membrane pores, resulting in local cell swelling, membrane rupture and extravasation of cytoplasmic DAMPs [13] , [14] , [15]. Released DAMPs will further recruit immune cells and aggravate inflammatory cascade [16].
Macrophages/microglia are vital mediators of innate immune responses following CNS injury [17]. These cells are also considered to be the main cells in CNS where pyroptosis occurs [10] , [18]. Pyroptosis of macrophages/microglia have been implicated in the pathogenesis of multiple CNS disease, including SCI and traumatic brain injury [19] , [20]. Although pyroptosis of macrophages/microglia is repeatedly mentioned in a variety of neuroinflammatory-related diseases, the mechanism of its occurrence is not well understood.
CD73, also known as ecto-5'-nucleotidase (NT5E), is an AMP hydrolase that is part of an extracellular enzymatic mechanism that regulates the conversion of extracellular ATP to adenosine [21]. Plenty of studies have point to CD73 as a key regulator in various pathophysiological processes, including maintaining immune homeostasis by regulating the balance between pro-inflammatory ATP and immunosuppressive adenosine to prevent excessive immune responses [22] , [23]. In addition, our 6 previous research determined CD73 has an anti-neuroinflammatory role in SCI, which is attributable to its regulation of macrophages/microglia polarization via adenosine-p38 cascade [24]. An increasing piece of evidence now clearly indicates CD73 is involved in activation of inflammasome [25] , [26]. Therefore, whether the immunosuppressive mechanism of CD73 after SCI contains regulation of macrophages/microglia pyroptosis is promising.
We hypothesized that CD73 could attenuate inflammasome activation and inhibit pyroptosis of macrophages/microglia, reducing neuroinflammation after spinal cord injury. In the current study, we first confirmed the role of CD73 and GSDMD in spinal cord injury through SCI patient blood samples. Then we explored the mechanism of CD73 regulating macrophages/microglia pyroptosis by in vivo and in vitro experiments. Finally, we investigated the upstream factor regulating the expression of cd73 after spinal cord injury.

Donor recruitment and blood sample preparation
From January 2019 to June 2019, 20 SCI patients from Huashan Hospital, Fudan University were enrolled. Inclusion criteria included the following: (1) a clear history of trauma, and there were no neurological abnormalities of spinal cord injury before injury; (2) existing neurological abnormalities such as limb sensation, motor abnormality, and dysfunction of the bowel and bladder; (3) MRI examination showed spinal cord compression and spinal cord signal changes. Patients with treatment of methylprednisolone before blood taking, with a history of brain disease, with a history of spinal surgery were excluded. Peripheral blood was collected from 20 patients with SCI and 20 healthy donors with similar gender and age distribution, 7 respectively. The Japanese orthopedic association (JOA) score, cervical dysfunction index (NDI) and American spinal injury association (ASIA) classification were used to assess the severity of SCI. The whole blood RNA was extracted using GeneJET Stabilized and Fresh Whole Blood RNA Kit (ThermoFisher, CA, USA) in accordance with the manufacturer's protocol.

Quantitative real-time PCR (RT-PCR) analysis
Total RNA was extracted using TRIzol reagent (Invitrogen, San Diego, CA, USA) according to the manufacturer's instructions, and quantification of mRNAs was performed using a 10-μl final reaction volume using SYBR Green PCR Master Mix (ThermoFisher, CA, USA). GAPDH mRNA was used as a housekeeping gene and relative expression levels of mRNAs were calculated using the comparative ΔΔCT method.

Establishment of SCI model and drugs treatment
Just as descripted in our previous study [24], mice were anesthetized with pentobarbital by intraperitoneal injection(35mg/kg). Each mouse was inflicted with spinal crush injury at the T8-T9 with Dumont-type forceps with a 0.2 mm spacer. We carried out the vertebrae laminectomies of T8-T9 with a pair of fine rongeurs, and the dura mater of mice was protected. The mice SCI model was made by lateral compression of the spinal cord with a depth of 0.2 mm for 20s. After operation, the 8 mice were given intramuscular injection of penicillin, 20,000 units once to resist infection. For further studied the effect of AKT and HIF-1α on SCI, SC79 (40mg/kg in DMSO) and BAY87-2243 (0.5mg/kg and 4mg/kg in DMSO) were injected intraperitoneally into mice every day after the establishment of mice SCI model. The mice in the control group were injected with the same volume of DMSO at the same time point.

Locomotion recover assessment
The locomotor behavior after SCI was detected by the Basso, Beattie, and

RNA sequencing and functional enrichment analysis
Total RNA was isolated from cells using the Trizol (Invitrogen Carlsbad CA, USA) according to the manufacturer's protocol. RNA integrity was evaluated using the Agilent 2200 TapeStation (Agilent Technologies, USA) and each sample had the RIN above 7.0. Subsequently, the purified RNAs were subjected to first strand and second strand cDNA synthesis following by adaptor ligation and enrichment with a low-cycle according to instructions of NEBNext® Ultra™ RNA Library Prep Kit for Illumina (NEB, USA). The purified l library products were evaluated using the Agilent 2200 TapeStation and Qubit®2.0(Life Technologies, USA) and then diluted to 10 pM for cluster generation in situ on the pair-end flow cell followed by sequencing (2×150 bp) HiSeq3000. The clean reads were obtained after removal of reads containing adapter, ploy-N and at low quality from raw data. HISAT2 was used to align the clean reads to the mouse reference genome mm10 with default parameters. HTSeq was subsequently employed to convert aligned short reads into read counts for each gene model. Differential expression was assessed by DEseq using read counts as input. The Benjamini-Hochberg multiple test correction method was enabled. Differentially expressed genes were chosen according to the criteria of fold change > 2 and adjusted p-value < 0.05. All the differentially expressed genes were used for heat map analysis and KEGG ontology enrichment analyses. For KEGG enrichment analysis, a P-value < 0.05 was used as the threshold to determine significant enrichment of the gene sets.

Enzyme-linked immunosorbent assay
In cell culture supernatants or mouse spinal cord homogenate, the measurement of protein levels of IL-1β, IL-6 and TNF-α was taken using the commercial ELISA kits from Sigma (Sigma-Aldrich, St. Louis, MO, USA) according to the manufacturer's instructions.

Cytotoxicity assay
The release of lactate dehydrogenase (LDH) was detected to determine the cytotoxicity using the LDH Cytotoxicity Assay Kit (Beyotime, Shanghai, China) following the manufacturer's instructions.

Western blot analysis
Protein of spinal cord tissue and BV2 cells was homogenized in radioimmunoprecipitation assay (RAPI) lysis buffer, and protein concentrations were determined using the BCA assay. Protein samples were fractionated using sodium

Immunohistochemical assessment
At the third day after SCI, different groups mice were deeply anesthetized with 10 % chloralic hydras (3.5 ml/kg, i.p.) and perfused with 0.9 % NaCl, followed by 4 % and CASP-1 (1:100, Abcam, ab1872) for 1 h, followed by incubation with HRPconjugated anti-rabbit secondary antibodies for 30 min. Bound antibodies were visualized by incubation with DAB for 10 min. All images were captured using a Nikon ECLIPSE Ti microscope (Nikon, Japan).

Immunofluorescence assessment
Spinal cord tissue samples were harvested as described above. ImmunoResearch, West Grove, PA). All images were acquired using Nikon ECLIPSE Ti microscope (Nikon, Japan).
After transfection for 24 hours, cells were harvested and measured using the toolVeritas 9100-002 (Turner BioSystems, Sunnyvale,CA, USA), and luciferase activity was divided by the Renilla luciferase activity to normalize for transfection efficiency.

Chromatin Immunoprecipitation assay
ChIP assay was performed using a ChIP assay kit (Abcam, Cambridge, UK) according to the manufacturer's protocol. In brief, primary antibodies of foxo1 (1:500, CST, 2880) or IgG (1:500, Abcam, ab172730) were used. DNA-protein cross-linking complexes were collected, and purified DNA was subjected to qPCR with SYBR Green PCR Master Mix (ThermoFisher, CA, USA).

Statistical analysis
All results are expressed as mean ± standard deviation.Student's unpaired t tests and two-way analysis of variance (ANOVA) followed by Dunnett's test were used to analyze data. A p value of less than 0.05 was considered to be statistically significant. All statistical analyses were done with the SPSS 14.0 software.

Expression of NLRP3/GSDMD genes in peripheral blood of patients with SCI
is positive correlated with the severity of injury.
To investigate whether NLRP3/GSDMD plays a role in neuroinflammation after SCI, 20 blood samples from SCI patients as well as 20 blood samples from normal people were subjected to RT-PCR assay. Then 20 SCI patients were divided into highexpression and low-expression groups based on whether NLRP3/GSDMD gene expression reached an average level. As showed in Table 1, various clinical parameters in SCI patients were compared between two groups. Specifically, 13 NLRP3/GSDMD high-expression group had a lower JOA score and a higher NDI index compared to NLRP3/GSDMD low-expression group. Strikingly, quantitative analysis revealed that the expression of NLRP3/GSDMD was significantly higher in the SCI patient samples in comparison to normal samples (Fig. 1A-AB). Furthermore, linear regression analysis confirmed a positive correlation between NLRP3/GSDMD expression and Japanses Orthopaedoc Assocation score (JOA score), while a negative correlation between CD73/GSDMD expression and Neck Disability Index (NDI index) (Fig.1C-1D). RCO curve demonstrated they have a good diagnostic significance for SCI (Fig.1E). Collectively, these results indicate the expression of CD73 and GSDMD in SCI patients correlate with the severity of injury.

CD73 deficiency facilitates NLRP3 inflammasome activation and pyroptosis of macrophages/microglia in vivo
To examine SCI inflammasome expression in a systematic manner, a wider panel of inflammasome genes was assessed in spinal cord tissue samples from mice with SCI or sham surgery. All inflammasome-associated genes examined were detectable in mice spinal cord tissue 3 days after surgery, with increased IL-1β, IL-18, CASP-1 and GSDMD transcript levels in SCI group ( Supplementary Fig.1A). Besides, all inflammasome sensors examined were also detectable, and NLRP3 and AIM2 transcript levels were upregulated ( Supplementary Fig.1B) after SCI. To investigate the influence of CD73 on inflammasome activation, the mRNA detection was performed once again in CD73 KO mice with SCI. Importantly, compared with WT mice, various inflammasome-associated genes (IL-1β, IL-18, CASP-1, NLRP3, ASC and GSDMD) and three inflammasome sensors (NLRP3, NLRP1 and AIM2) showed higher levels in CD73 KO mice 3 days post injury ( Fig.2A-2B). The release of proinflammatory factor (IL-1β, IL-6 and TNF-α) and LDH was also effectively 14 increased in CD73 KO mice after SCI (Fig.2C-2D). Western Bolt and immunochemistry revealed similar protein expression patterns of NLRP3, GSDMD, ASC and CASP-1 as their mRNA expression (Fig.2E-2F, 2H). In particular, levels of full length GSDMD and N-terminal GSDMD (GSDMD-N, an active form of GSDMD) were both increased (Fig.2E-2F). Immunofluorescence results showed increased expression of GSDMD coincides with CD68 (a marker of macrophages/microglia) (Fig.2G). Together, these results demonstrate that CD73 deficiency exacerbate macrophages/microglia pyroptosis caused by NLRP3 inflammasome activation in SCI.

CD73 alleviates LPS-induced NLRP3 inflammasome activation and pyroptosis in BV2 cells through A 2B adenosine receptor
Our previous studies have showed that CD73 regulates microglia polarization through the A 2B adenosine receptor A 2B AR . To determine whether A 2B AR mediates the CD73 ability to alleviates microglia pyroptosis, we constructed BV2 cells that either had downregulated or upregulated CD73 expression, exposed them with adenosine or A 2B AR antagonist MRS1706 after LPS pretreatment. Results of RT-PCR and ELISA showed that CD73 downregulation caused a profound induction of pyroptosis genes in transcript levels including NLRP3, ASC, CASP-1 and GSDMD, increased the release of IL-1β, IL-6, TNF-α and LDH, while CD73 upregulation was counterproductive ( Fig.3A-3C). Immunoblot and Immunofluorescence revealed similar changes in protein levels of NLRP3, GSDMD and ASC replying to CD73 change ( Fig.3D-3F). Notably, administration of adenosine or MRS1706 offsets CD73 downregulation or upregulation effect respectively in all above experiments (Fig.3A-3F). All these results suggest that A 2B AR is necessary for CD73 to regulate microglia pyroptosis.

CD73 attenuates microglia pyroptosis via PI3K/AKT/Foxo1 signal
To further investigate the mechanism by which CD73 regulates macrophage/microglia pyroptosis, mRAN sequencing were performed on CD73overexpressing BV2 cells after administrated with LPS. A total of 1649 mRNAs were differentially expressed between CD73 overexpressed BV2 cells and normal subjects (Fig.4A); 1081 were upregulated and 568 were downregulated; 55 differentially expressed mRNAs were enriched in PI3K/AKT pathway (Fig.4B). The mRNA and protein expression of PI3K were increased in CD73 over expressed BV2 cells, and the phosphorylated AKT and Foxo1 were also observed ( Fig.4C-4E). In addition, after pretreated with LPS, we utilized MRS1706 and an AKT inhibitor MK2206 to try to reverse the effect of CD73 upregulation. We found the mRNA levels of PI3K was decreased after exposed to MRS1706 in CD73 upregulated cells, and the phosphorylated AKT and Foxo1could be suppressed by MRS1706 and MK2206 ( Fig.4C-4E). MK2206 treatment also caused a remarkably increase in transcript levels including NLRP3, ASC, CASP-1 and GSDMD, increased the release of IL-1β, IL-6, TNF-α and LDH (Fig.4F-4H). Immunoblot demonstrated similar changes in protein levels of NLRP3, GSDMD and ASC replying to MK2206 (Fig.4I-4J). In vivo study, intraperitoneal injection of 40mg/kg SC79 (an AKT activator) was carried out every day after surgery in CD73 KO mice, and the results of immunohistochemistry indicated SC79 could decrease the expression of GSDMD and CASP-1 at the lesion site (Fig.4K). Furthermore, the BBB scoring in SC79 group was significantly superior to DMSO group at 28 th day after injury (Fig.4L). Taken together, we conclude that the CD73 inhibitory effect on microglial pyroptosis is mediated via the PI3K/AKT/Foxo1 pathway.

Foxo1 is a transcriptional activator in the promoter region of GSDMD gene
In order to analyze the molecular mechanisms involved in GSDMD regulation, we used the JASPAR database (http://jaspar.genereg.net) to identify potential binding sites of GSDMD promoter region for Foxo1. To measure the activity of potential 3 cis-acting elements and determine the minimum sequence required for activity, a series of 5 reporters constructs with progressively larger deletions from the 5′ end of the promoter was generated. Luciferase assay revealed an increased promoter activity of the pGL3-2000/+ 200bp as compared to the control group in BV2 cells, indicating the existence of Foxo1 binding site in the region of GSDMD gene (Fig.5A).
Consistently, the ChIP assay confirmed the binding of foxo1 to GSDMD promoter region (Fig.5B). When the promoter was deleted to position +50bp, the promoter activity with pGL3+50/+200bp decreased significantly compared with the pGL3−300/+200 (Fig.5A), this result demonstrated that positive regulatory elements were located in the −300/+50 region in BV2 cells. Further mutation of this binding site blocked the transcription of GSDMD (Fig.5C). Overexpression of Foxo1 in BV2 cells also increased the expression of GSDMD, while cotreatment of Foxo1 plasmid and CD73 plasmid inhibited the elevation of GSDMD mRNA, as well as reduced the release of IL-1β, IL-6, TNF-α and LDH (Fig.5D-5F). Thus, we conclude Foxo1 regulates the expression of GSMDM at transcriptional level.

HIF-1α mediates CD73 upregulation in microglia
Several reports have shown that HIF-1α can regulate CD73 expression. So, in this study, we exposed LPS activated BV2 cells with a HIF-1α small interference RNA to determine its regulatory role. LPS-induced pyroptosis could simultaneously promote the expression of HIF-1α and CD73 in BV2 cells, and the overexpression of CD73 was inhibited after interference of HIF-1α (Fig.6A-C). The results of in vivo experiment indicated SCI could increase the expression of HIF-1α along with the elevation of CD73 in damaged spinal cord tissue, while BAY87-2243 treatment, a HIF-1α inhibitor, reversed the expression pattern of CD73 (Fig.6D-G). Surprisingly, BAY87-2243 which was supposed to only affect the HIF-1α activity, also reduced HIF-1α expression both in mRNA and protein levels (Fig.6D-F). We suspect that CD73 reduction should be responsible for this phenomenon. As we thought, reduce the expression of CD73 in BV2 cells or knockout CD73 gene in mice blocked the increase of HIF-1α caused by LPS or SCI, respectively (Fig.7A-F). Moreover, Merighi et al.
reported a stimulatory effect of adenosine on HIF-1α by activation of p38 mitogenactivated protein kinases (MAPKs) phosphorylation via A 2B adenosine receptor [27].
Our previous study also determined that CD73 can regulate the polarization of macrophages/microglia through this pathway [24]. This prompted us to verify whether CD73 regulates HIF-1α expression is associated with this signal. In BV2 cells, the results of RT-PCR and immunoblots showed that both a A 2B adenosine receptor inhibitor MRS1706 and a p38 inhibitor SB203580 can reduce the increase in HIF-1α caused by overexpression of CD73 (Fig.7G-8I). The above studies confirmed that HIF-1α and CD73 can promote each other to accumulate and jointly regulate microglia pyroptosis.

Discussion
CNS trauma, either SCI or traumatic brain injury (TBI), can be divided into two stages (primary injury and secondary injury) according to its characteristics of pathological process [28]. Part of the secondary injury is characterized by persistence and diffuseness, including delayed glial and neuron cells death, leading to significant expansion of the damage site to higher segments and progressive neurodegeneration [29] , [30]. Neuroinflammation initiated by the innate immune response plays an important role in the pathology of secondary injury after CNS trauma, and inhibition of inflammation becomes a potential therapy for such injury[2] [6]. More recently, the scientific community has determined that the activation of cytoplasmic inflammasome complex leading to pyroptosis was an essential step of neuroinflammation in secondary CNS damage [7]. Although inflammasomes have been proposed in SCI, the effect of the pyroptosis-executing protein GSDMD on SCI was not clear. In addition, whether CD73, a 5'-ectonucleotidase which can attenuate neuroinflammation after SCI determined by our previous study, can affect the expression of GSDMD has not been studied. Here, we showed that NLRP3 and GSDMD were significantly upregulated in the blood samples from SCI patients and they were also correlated with the severity of this disease ( Fig.1A-1D). Moreover, RCO curve identified these genes have the potential to be a novel target for the diagnosis of human SCI (Fig.1E). These results confirmed our speculation about the involvement of NLRP3 and GSDMD in SCI.
In generally, inflammasome complexes consist of three parts: a cytosolic patternrecognition receptor, pro-inflammatory caspase and an adaptor protein ASC that facilitates the interaction between the two [11] , [31]. Among the multiple inflammasome complexes, NLRP3 appears to be one of the most relevant components in CNS trauma [32]. Results of the research carried out by Liu et al.
indicate NLRP3 inflammasome was overexpressed in microglia, neurons and astrocytes in TBI rats [33]. The experimental results of Wu et al demonstrate that pharmacologic suppression of NLRP3 inflammasome activation controls neuroinflammation, attenuates mitochondrial dysfunction in mice [19]. In addition, Adamczek and colleagues found AIM2 is expressed in cortical neurons and can be activated by TBI [34]. In this study, we systematically examined the expression of inflammasome-associated genes and found NLRP3 and AIM2 were overexpressed in spinal cord tissue after injury. What's more, the increase in NLRP3 was more pronounced, confirming that it may play a more important role after SCI, which was consistent with previous findings (Supplementary Fig.1). To estimate the effects of CD73 on inflammasome pathway in SCI, we compared inflammasome-related gene expression profile between CD73 KO mice and WT mice. We found that CD73 deficiency promoted expression of inflammasome genes, as well as increased the release of pro-inflammatory. Based on these results, we can conclude that CD73 plays a crucial role in the regulation of pyroptosis driven by NLRP3 inflammasome after SCI.
Macrophages/microglia are vital mediators of the innate immune response following CNS trauma and are essential for subsequent inflammatory responses[35] , [36].
There exists a substantial amount of evidence supporting that pyroptosis mainly occurs in macrophages/microglia in CNS neuroimmune diseases. A possible reason is that these cells express higher levels of PRRs which can recognize PAMPs and DAMPs, and initiate pyroptosis cascade [10] , [18]. We also observed intense GSDMD immunostaining in CD68-labeled macrophages/microglia and a profound increase in spinal cord lesions of CD73 KO mice (Fig.2G). Taken together, the present results defined a pyroptosis occurring in macrophages/microglia could be inhibited by CD73 in SCI.
CD73, a glycosylphosphatidylinositol (GPI) anchored cell surface protein, has a central role in the adenosinergic system and is considered as the rate-limiting enzyme in the generation if extracellular adenosine [37]. Kulesskaya et al. found about 85-95% of murine AMP-hydrolyzing capabilities are mediated by CD73 [38].
When the concentration of extracellular adenosine increase, it can activate P1 purinergic receptors (adenosine receptors) on target cells and stimulates a myriad of protective cellular responses restoring homeostasis [39] , [40]. The P1 G-proteincoupled receptor family consists of 4 distinct subtypes: A 1 R , A 2A R, A 2B R and A 3 R [41], however, these four receptors have different affinities with adenosine. A 1 R , A 2A R and A 3 R are categorized as high-affinity adenosine receptors because they can be activated by physiologic adenosine concentrations, while A 2B R only be activated under pathological conditions which is relevant to inflammatory events [42] , [43].
Experimental evidence revel that hypoxia can increase the expression of A 2B R [44], resulting in the activation of A 2B R signaling to protect rats lung injury [45] as well as myocardial ischemia [46]. Similarly, we confirmed CD73 attenuate SCI in mice by A 2B R in our previous research [24]. Taken together, we assumed that adenosine-A 2B R cascade may be one of the mechanisms by which CD73 regulates the pyroptosis of macrophage/microglia after SCI. In the subsequent study, our data showed that the effect of CD73 interference or overexpression in BV2 cells can be counteracted by adenosine or MRS1706 and these results are consistent with our hypothesis (Fig.3).
To further investigate the mechanism by which CD73 regulates macrophage/microglia pyroptosis, mRAN sequencing were performed on CD73overexpressing BV2 cells after administrated with LPS. KEGG analysis showed a large number of differential genes enriched in the PI3K/AKT pathway. PI3K/AKT signaling pathway play a central role in multiple cellular functions such as cell proliferation and survival [47]. More noteworthy is that the activation of this pathway is closely related to the inflammatory response. For example, the research 21 of Yin et al. demonstrated the expression of pro-inflammatory factors IL•12, TNF•α and IL•6 in human innate immune cells was increased by PI3K or AKT inhibitors, while the expression of anti-inflammatory factor IL•10 was decreased [48]. Another study showed that the activation of AKT inhibits inflammatory responses in LPSinduced sepsis mice and sepsis rabbits [49]. These studies reflect the immunosuppressive function of the PI3K/AKT pathway, which is consistent with the role of CD73. Foxo1 is a vital transcription factor downstream of AKT. When PI3K/AKT pathway is constitutively activated, Foxo1 can be phosphorylated by AKT in the nucleus, resulting in the translocation of Foxo1 to the cytoplasm and inactivate it [50]. A large number of studies have found that inhibition of Foxo1 can reduce inflammation in different cells [51] , [52]. To date, however, there are no reports about Foxo1 is related to CD73 or pyroptosis. In the present study, our results indicated that overexpression of CD73 alleviates the pyroptosis of macrophages/microglia in a PI3K/AKT dependent way both in vivo and in vitro ( Fig.4). Furthermore, the results of dual-luciferase reporter assay in BV2 cells demonstrated that there is a Foxo1 binding site between 200 bp upstream of the GSDMD gene and 50 bp downstream, and this discovery was also verified by ChIP assay (Fig.5). Based on the above findings, we believe that the PI3K/AKT/Foxo1 pathway is essential for CD73 to regulate pyroptosis.
The fact that CD73 was overexpressed after SCI has been confirmed repeatedly in the present study as well as our previous research. However, the biological mechanism behind this is still unclear. HIF-1 is a heterodimeric transcription factor composed of a HIF-1α subunit and HIF-1β subunit and functioned as a master regulator of oxygen homeostasis [53]. When in anoxic conditions, the HIF-1α subunit accumulates and then binds to HIF-1β ultimately activates HIF-1-target genes, whose products are regulating angiogenesis, glucose metabolism, cell survival, invasion and metastasis [54]. Numerous studies suggested that the pathogenesis of SCI involves the ischemia and hypoxia in lesion site, which increase the expression of HIF-1, enhancing the resilience of neuronal cells under hypoxia [55] , [56].
Karhausen et al. once reported activation of HIF-1α promoted CD73 transcription, resulting in attenuated loss of barrier during colitis in vivo [57]. Besides, Synnestvedt et al. examined the CD73 gene promoter and identified a binding site for HIF-1α, and inhibition of HIF-1α expression by antisense oligonucleotides resulted in significant decrease of hypoxia-inducible CD73 expression [58]. In this study, we also found that overexpression of CD73 after spinal cord injury is associated with HIF-1α. Interfering with the expression of HIF-1α or inhibiting the activity of HIF-1α can reduce the expression of CD73 after spinal cord injury (Fig.6).
Interestingly, we also found that the expression of HIF-1α in microglia was CD73 dependent. The adenosine-A 2B AR-p38 pathway, which has been confirmed regulated the function of macrophages/microglia after spinal cord injury in our previous study, probably the mechanism by which CD73 regulates HIF-1α expression (Fig.7). This finding is consistent with that reported by Merighi [27]. Therefore, we believe that a positive feedback regulatory loop can be formed between CD73 and HIF-1α after spinal cord injury, which is involved in the regulation of macrophages/microglia pyroptosis and inhibits neuroinflammatory response.

Conclusion
In summary, our present study demonstrates for the first time that CD73 has an anti-pyroptosis role in SCI, which is partly attributable to its inhibition of GSDMD via adenosine-A 2B AR-PI3K-AKT-Foxo1 signal. Moreover, our results showed that the 23 creation of a positive feedback loop between CD73 and HIF-1α may an important mechanism for reducing neuroinflammation after SCI (Fig.8). We propose that CD73 may be utilized as a target molecule for the development of novel therapeutic methods for SCI.