Feng Hua, Hao-Ran Wang, Yun-Feng Bai, Jin-Peng Sun, Wei-Shun Wang, Ying Xu, Ming-Shun Zhang, Jun Liu,*
AbstractIn response to spinal surgery, neurons secrete a large amount of substance P into the epidural area.Substance P is involved in macrophage differentiation and fibrotic disease.However, the specific roles and mechanisms of substance P in epidural fibrosis remain unclear.In this study, we established a mouse model of L1–L3 laminectomy and found that dorsal root ganglion neurons and the macrophages infiltrating into the wound area released sphingolipids.In vitroexperiments revealed that type 1 macrophages secreted substance P, which promoted differentiation of type 1 macrophages towards a type 2 phenotype.High-throughput mRNA-seq analysis revealed that the sphingolipid metabolic pathway may be involved in the regulation of type 2 macrophages by substance P.Specifically, sphingomyelin synthase 2, a component of the sphingolipid metabolic pathway, promoted M2 differentiation in substance P-treated macrophages,while treating the macrophages with LY93, a sphingomyelin synthase 2 inhibitor, suppressed M2 differentiation.In addition, substance P promoted the formation of neutrophil extracellular traps, which further boosted M2 differentiation.Blocking substance P with the neurokinin receptor 1 inhibitor RP67580 decreased the number of M2 macrophages in the wound area after spinal surgery and alleviated epidural fibrosis, as evidenced by decreased fibronectin,α-smooth muscle actin, and collagen I in the scar tissue.These results demonstrated that substance P promotes M2 macrophage differentiation in epidural fibrosis via sphingomyelin synthase 2 and neutrophil extracellular traps.These findings provide a novel strategy for the treatment of epidural fibrosis.
Key Words:dorsal root ganglion; epidural fibrosis; laminectomy; macrophage; mitochondria; neurokinin receptor 1; neutrophil extracellular traps;sphingomyelin synthase 2; substance P
Because of aging populations and poor lifestyle habits, the number of people with disc herniations is increasing.As a result, laminectomy is increasingly required to relieve spinal cord and nerve compression.Spinal surgery triggers sterile inflammation.Ideally, stromal cells and immune cells in the surgical area produce extracellular matrix and promote wound healing.However,excessive accumulation of extracellular matrix (Sorg et al., 2017) may cause tissue adhesion to the dura mater, potentially leading to epidural fibrosis,which is associated with failed back surgery syndrome.The pathogenesis of epidural fibrosis remains largely elusive.
Substance P (SP), encoded by theTAC1gene, belongs to the tachykinin family(Pavón-Romero et al., 2021).SP is a neuropeptide that is widely distributed in nerve fibers (Crosson et al., 2021) and is involved in pain transmission (Mathur et al., 2021).Unlike other common neurotransmitters, SP regulates various functions, including cell proliferation, differentiation, activation, and migration(Hodo et al., 2020).In addition to nerve cells, many non-nerve cells, including immune cells, smooth muscle cells, endothelial cells, and airway epithelial cells also secrete SP (Mashaghi et al., 2016).SP mainly exerts its functions through neurokinin receptor 1 (NK1-R), a G protein-coupled receptor (Jin et al., 2021).
As one of the most common types of innate immune cells, macrophages have a wide range of functions (Liang et al., 2021; Bautista and Krishnan, 2022;Mackie et al., 2022; Rosmus and Wieghofer, 2022).According to different activation states and functions, macrophages can be divided into classically activated macrophages (type 1 macrophages, M1) and alternatively activated macrophages (type 2 macrophages, M2) (Mills, 2015).M2 macrophages play an important role in the formation of fibrosis (Pakshir and Hinz, 2018; Chinju et al., 2022).After laminectomy, macrophages are quickly recruited to the surgical area in response to trauma signals (Das et al., 2015).At this point,appropriate differentiation of macrophages promotes anti-inflammatory and anti-bacterial activity and postoperative repair of the surgical site (Smigiel and Parks, 2018).In particular, M1 macrophages in the early phase may help remove injured cells, and M2 macrophages in the later phase may help with resolution of the inflammatory response and wound healing.Disruption of appropriate macrophage differentiation may have serious consequences,such as low back pain caused by the aggravation of epidural fibrosis (Wang et al., 2021).Macrophage differentiation is regulated by various factors.SP promotes M2 macrophages and favors fibrosis in the diabetic skin (Leal et al., 2015).In contrast, SP deficiency in the diabetic heart increases myocardia fibrosis (Widiapradja et al., 2021), suggesting the complicated roles of SP in macrophage differentiation and organ fibrosis.In the present study, we investigated whether SP regulates macrophage differentiation in the surgical area after spinal surgery and explored the mechanisms by which SP regulates macrophage differentiation during epidural fibrosis.
Clinical samples
To investigate the amount of SP present before or after laminectomy, we enrolled 12 patients diagnosed with lumbar disc herniation who underwent laminectomy at The Second Affiliated Hospital of Nanjing Medical University from September 2020 to December 2020.The patients’clinical characteristics are shown in Additional Table 1.The diagnosed criteria were as follows: (1)low back pain and numbness in lower extremities, (2) positive for the straight leg raise test, and (3) magnetic resonance imaging showed a herniated disc at the L1–L3 level (Gadjradj et al., 2022).None of the patients had cancer,autoimmune diseases, or serious infectious diseases.This study was approved by the Ethics Committee of the Second Affiliated Hospital of Nanjing Medical University (approval No.[2019]KY056) in 2019, and each subject signed an informed consent form.Peripheral blood was collected from the patients twice–before the operation and 3 days after the operation– and the blood samples were centrifuged at 500 ×gfor 5 minutes and stored at –80°C for further study.
Animals
Because male mice are more tolerant of surgery than female mice (Woosley et al., 2000), only male mice were used in this study.Male C57BL/6J mice(8 weeks old and 4 weeks old) were purchased from the Animal Center of Nanjing Medical University (license No.SYXK (Su) 2021-0023).The Nanjing Medical University Animal Protection and Use Committee approved the animal experiments (approval No.IACUC-1904052) in 2019.All experiments were designed and reported according to the Animal Research: Reporting ofIn VivoExperiments (ARRIVE) guidelines (Percie du Sert et al., 2020).
The 8-week-old mice were randomly assigned to three different groups: (1)the control group in which the lamina was not removed, (2) the operation group that underwent laminectomy, and (3) the treatment group that underwent laminectomy and was treated with RP67580.Each group contained at least six mice, and the groups were matched for mouse age and body weight.
Mice were intraperitoneally injected with 10 mg/kg xylazine (Medistar,Ascheberg, Germany) and 200 mg/kg ketamine hydrochloride (Riemser,Greifswald, Germany) in 100 μL normal saline.The spine was exposed from the posterior midline, and the L1–L3 laminae were removed with osteotomy forceps.In the control group, the lamina was exposed, and the skin was sutured without lamina removal.A gelatin sponge (Hushida, Nanchang,China) that matched the size of the wound (about 1.5 cm2) was soaked in NK1-R receptor inhibitor RP67580 (5 μM, Tocris, Minneapolis, MN, USA, Cat#1635) and inserted into the surgical site, which was then sutured closed.All surgeries were performed by the same person, and the size of the surgical wound was kept as consistent as possible in all of the mice to ensure that each mouse absorbed about the same amount of RP67580.In the control group, only a gelatin sponge (with no inhibitor) was used.After surgery, the mice were placed in a warm environment and returned to their cages after awakening from anesthesia.
To deplete macrophages, mice were intraperitoneally injected with 200 μL clodronate liposomes (5 mg/mL, Yeasen, Shanghai, China, Cat# 40338ES10)or blank liposomes (5 mg/mL, Yeasen, Cat# 40338ES05) 24 hours before the spinal surgery.
Macrophages and dorsal root ganglion cell culturing
After intraperitoneal injection with 10 mg/kg xylazine and 200 mg/kg ketamine hydrochloride in 100 μL normal saline, the 4-week-old mice were sacrificed by cervical dislocation.The tibia and femur were removed from the mice, and excess tissue was removed.The bone marrow was rinsed with cold phosphatebuffered saline (PBS).Cell precipitates were obtained by centrifugation at 4°C and 500 ×gfor 5 minutes in red blood cell lysis buffer (Thermo Fisher Scientific, Waltham, MA, USA).The obtained cells were washed twice and resuspended in Dulbecco’s modified Eagle medium (Sigma-Aldrich, Shanghai,China, D5030) containing 10% fetal bovine serum (Gibco, Waltham, MA, USA),100 U/mL penicillin, and 0.1 mg/mL streptomycin, at a concentration of 2 ×106cells/mL.Monocytes were then induced to adopt an M1 or M2 phenotype using 10 ng/mL granulocyte-macrophage-colony stimulating factor (Biolegend,San Diego, CA, USA, Cat# 576308) or macrophage-colony stimulating factor(Cat# 576408, Biolegend), respectively (Sehgal et al., 2021; Woo et al., 2021).The bone marrow-derived macrophage (BMDM) culture medium was changed on days 3 and 5.BMDMs were fully mature on day 7.BMDMs were identified by the F4/80 marker using flow cytometry, and the results showed that more than 95% of the cells were F4/80-positive.
To purify dorsal root ganglia (DRG) cells (Yang et al., 2022), six young male 4-week-old C57BL/6J mice, provided by the Animal Center of Nanjing Medical University, were placed supine in ice-cold PBS solution under aseptic conditions.After intraperitoneal injection with 10 mg/kg xylazine and 200 mg/kg ketamine hydrochloride in 100 μL normal saline, the mice were sacrificed by decapitation, and the chest and abdominal cavity were opened to remove the internal organs.Microscissors were inserted through the cervical spinal canal opening, and the spinal canal at the L1–L3 levels was cut open along both sides of the dorsal midline of the spine.After the spinal cord was removed and exposed, the DRG along both sides of the spine could be seen beside the foramina.The DRG were carefully extracted under an operating microscope, and the capsule was stripped.The isolated DRG tissue blocks were placed in 0.25% trypsin solution and digested at 37°C for 30 minutes.Digestion was terminated with 10 drops of fetal bovine serum (Gibco), and the digested tissues were centrifuged at 500 ×gfor 5 minutes.Then, the supernatant was removed and cultured with neurobasal medium containing 10% fetal bovine serum, 100 U/mL penicillin, 0.1 mg/mL streptomycin, B-27 supplement (1:50; Invitrogen, Carlsbad, CA, USA), and nerve growth factor (50 ng/mL; Invitrogen).When the DRG neurons reached 70% confluency in a 24-well culture plate, they were stimulated with 10 ng/mL high mobility group box 1 (HMGB1; Cat# 50913-M01H, SinoBiological, Beijing, China) for 24 hours.Flow cytometry was used to detect β3-tubulin, a marker of DRG neurons, and the results showed that more than 95% of the cells were β3-tubulin-positive.The growth medium was collected and centrifuged at 500 ×gfor 5 minutes at 4°C to remove cell debris, and the level of SP in the clarified supernatants was then measured using a commercial enzyme-linked immunosorbent assay(ELISA) kit (Cayman Chemical, Ann Arbor, MI, USA, Cat# 583751).
High-throughput sequencing of macrophage mRNA
BMDMs were treated with 10 μM SP for 1 hour.Total mRNA was extracted using TRIzol (Invitrogen) according to the manufacturer’s instructions.After testing RNA quality and integrity, the RNA was quantified and subjected to stranded RNA sequencing in collaboration with Seqhealth (Wuhan, China).RNA sequencing was performed using a Novaseq 6000 sequencer (Illumina,San Diego, CA, USA).For further analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis for differentially expressed genes was conducted using clusterProfiler (version 4.1.0, Yulab-SMU, Guangzhou, China)R package (Pvalue ＜ 0.05; false discovery rate ＜ 0.05).
Neutrophil extracellular trap purification and macrophage treatment
Neutrophils were isolated from the bone marrow of 4-week-old C57BL/6J mice (the same genetic background described above).The collected cells were passed through a 70-μm filter, and the filtrate was centrifuged at 500 ×gfor 5 minutes.The supernatant was discarded, and the cells were resuspended in 1 mL PBS.The following solutions were then added to a 15-mL tube in the order listed: 3 mL Histopaque 1119 (density 1.119 g/mL,11191, Sigma-Aldrich), 3 mL Histopaque10771 (density 1.0771 g/mL, 10771,Sigma-Aldrich), and 1 mL PBS containing bone marrow cells.Neutrophils stimulated with 10 μM SP for 6 hours were collected by centrifugation at 500 ×gfor 5 minutes.Then the supernatant was transferred to a 1.5-mL Eppendorf tube and centrifuged at 18,000 ×gfor 10 minutes.The neutrophil extracellular traps (NETs), which settled to the bottom of the tube, were then resuspended in 100 μL PBS.In some assays, NETs (1 μg/mL) were used to treat macrophages for 24 hours, and in other assays the elastase inhibitor Alvelestat (20 μg/mL, MCE, Shanghai, China, AZD9668) was used to block the effects of NETs on macrophages for 24 hours (Jin et al., 2020).
Enzyme-linked immunosorbent assay
SP in the blood samples collected from the patient’s upper extremities was measured using an SP ELISA kit (583751, Cayman Chemical) according to the manufacturer’s instructions.Both the macrophages and the DRG neurons from mice were stimulated with HMGB1 (10 ng/mL) or exogenous SP (1 nM).In some assays, macrophages were treated with neutralizing antibodies (10 ng/mL;SinoBiological, Beijing, China, Cat# 50349-RN023) against TNF-α.The growth medium of macrophages was placed in an Eppendorf tube and centrifuged at 500 ×gfor 5 minutes at 4°C, and the supernatants were collected.The SP concentration in the supernatants was measured using the same method as that used for the blood samples.The absorbance was measured at 450 nm using a microplate reader (Thermo Fisher Scientific, Cat# MK3), and the SP concentration was calculated according to a standard curve.
At 30 days after the operation or sham operation, the mice received intraperitoneal injection with 10 mg/kg xylazine and 200 mg/kg ketamine hydrochloride in 100 μL normal saline and were sacrificed by cervical dislocation.Then the tissue from the surgical site was harvested, homogenized in lysate buffer, and centrifuged at 4°C and 10,000 ×gfor 10 minutes.The supernatant was collected, and the collagen I content was determined using a collagen I ELISA kit (Beijing Long tian Technology Co., Ltd., Beijing, China).
The growth medium from macrophages stimulated with 1 nM SP was removed, diluted, and added to a tumor necrosis factor-α (TNF-α) ELISA kit(R&D Systems, Minneapolis, MN, USA, DY410).The ELISA results were then graphed to calculate the TNF-α content in the medium.
After macrophages were stimulated with 10 μM SP or treated with 1 μg/mL LY93 (a SGMS2 inhibitor; Li et al., 2019) for 24 hours, the growth medium was removed and diluted.The concentration of ceramide, which can synthesize sphingomyelin in the presence of SGMS2, was measured using a ceramide kit (Sbj Bio, Nanjing, China, SBJ-M0754) according to the manufacturer’s instructions.
After macrophages were stimulated with SP, the growth medium was removed and diluted, and the C-X-C motif chemokine ligand 1 (CXCL1) concentration was measured using a KC kit (R&D Systems, DY453-05) according to the manufacturer’s instructions.
Histological analysis
At 30 days after spinal surgery, after intraperitoneal injection with 10 mg/kg xylazine and 200 mg/kg ketamine hydrochloride in 100 μL normal saline, the mice were sacrificed via cervical dislocation.Injured spinal cord tissue from L1 to L3 was harvested, fixed in 4% paraformaldehyde, and embedded in paraffin.For immunofluorescence staining, the embedded tissue was cut into 5 μm–thick sections and incubated with anti-SP (mouse monoclonal, 1:100,Abcam, Cambridge, UK, Cat# AB14184, RRID: AB_300971), anti-β 3-tubulin(a DRG neuron marker, mouse monoclonal, 1:100, Abcam, Cat# AB18207,RRID: AB_444319), anti-F4/80 (a macrophage marker, rat monoclonal,1:100, Abcam, Cat# AB6640, RRID: AB_1140040), anti-CD206 (an M2 type macrophage marker, rabbit polyclonal, 1:100, Abcam, Cat# AB64693, RRID:AB_1523910), anti-myeloperoxidase (MPO) (a neutrophil marker, rabbit monoclonal, 1:100, Abcam; Cat# AB208670, RRID: AB_2864724), anticitrulline histone 3 (citH3) (a NET marker, rabbit polyclonal, 1:50, Abcam,Cat# Ab5103 RRID: AB_304752), or anti-α-SMA (rabbit monoclonal, 1:100,Abcam, Cat# AB124964, RRID: AB_11129103) for 24 hours at 4°C.Then, the slides were incubated with goat anti-mouse IgG (Alexa Fluor® 594) (1:100,Abcam, Cat# AB150116, RRID: AB_2650601), goat anti-mouse IgG (Alexa Fluor® 488) (1:100, Abcam, Cat# AB150113, RRID: AB_2576208), goat anti-rat IgG (Alexa Fluor® 594) (1:100, Abcam, Cat# AB150160, RRID: AB_2756445),goat anti-rabbit IgG (Alexa Fluor® 488) (1:100, Abcam, Cat# AB150077, RRID:AB_2630356), goat anti-rat IgG H&L (Alexa Fluor® 647) (1:100, Abcam, Cat#AB15016, RRID: AB_2864291), goat anti-rabbit IgG (Alexa Fluor® 594) (1:100,Abcam, Cat# AB150080, RRID: AB_2650602), rabbit polyclonal secondary anti-mouse IgG (HRP) (1:100, Abcam, Cat# AB6728, RRID: AB_955440), or goat anti-rabbit IgG (HRP) (1:100, Abcam, Cat# AB6721, RRID: AB_955447)for 2 hours at 27°C and viewed with a fluorescence microscope (model BX-53,Olympus Optical, Tokyo, Japan).
For histochemical staining, tissue sections from the injury site were incubated with hematoxylin-eosin staining solution (abs9217, Absin, Shanghai, China)and Masson trichrome solution (abs9347, Absin).Briefly, the sections were first fixed in 95% ethanol for 2 minutes, and then washed with water for 5 minutes.The sections were placed in hematoxylin solution for 3 minutes before placing them in eosin solution.To label the collagenous fiber, Masson trichrome solution was used.The sections were placed sequentially in Bouin’s solution for 30 minutes, Weigert’s hematoxylin solution for 10 minutes,Biebrich scarlet solution for 10 minutes, phosphotungstic acid-phospholimbic acid solution for 15 minutes, and aniline blue solution for 2 minutes.The sections were washed with water before each change in dye solution.Finally,sections were rinsed in a 1% acetic acid bath, treated with alcohol, and permeabilized with xylene.The slides were imaged using a fluorescence microscope.All histological slides were observed by two independent investigators.
Western blot analysis
Briefly, 24-well plates containing macrophages or neutrophils were placed on ice, and the cells were lysed in radioimmunoprecipitation assay buffer containing 1 mM phenylmethanesulfonyl fluoride.To detect proteins in tissue samples, we homogenized the harvested L1–L3 spinal cord tissues in lysis buffer, and then added 5× loading buffer.For each sample, the same amount of protein was separated via 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred to polyvinylidene fluoride membranes, and blocked with 5% bovine serum albumin at 27°C for 1 hour.After five rinses in PBS and Tween, the membranes were incubated overnight at 4°C with anti-nitric oxide synthase 2 (iNOS) (rabbit monoclonal, 1:1000, Abcam, Cat#AB178945, RRID: AB_2861417) to detect type M1 macrophages, anti-CD86(rabbit polyclonal, 1:1000, Abcam, Cat# AB112490, RRID: AB_10866571) to detect type M1 macrophages, anti-CD206 (rabbit polyclonal, 1:1000, Abcam,Cat# AB64693, RRID: AB_1523910) to detect type M2 macrophages, antiarginase-1 (Arg-1) (rabbit monoclonal, 1:1000, Cell Signaling Technology,Beverly, MA, USA, Cat# 93668, RRID: AB_2800207) to detect type M2 macrophages, anti-citH3 (rabbit polyclonal, 1:1000, Abcam, Cat# Ab5103,RRID: AB_304752) to detect NETs, anti-fibronectin (rabbit monoclonal, 1:1000,Abcam, Cat# AB268020, RRID: AB_2262874) to detect fibrosis, anti-α-SMA(rabbit monoclonal, 1:1000, Abcam, Cat# AB124964, RRID: AB_11129103)to detect fibrosis, and anti-β-actin (rabbit polyclonal, 1:1000, Abcam, Cat#AB5694, RRID: AB_2305186).After washing with phosphate-buffered saline and Tween, the membrane was finally incubated at 27°C for 1 hour with goat anti-rabbit IgG (HRP) (1:5000, Abcam, Cat# AB6721, RRID: AB_955447)diluted in phosphate-buffered saline and Tween.The protein bands were detected using enhanced chemiluminescence (ECL) (Cat# 36208ES60, Yeasen)high-signal reagents.The expression of all proteins was normalized by β-actin.
Quantitative polymerase chain reaction
A TRIzol kit was used to extract total RNA from BMDMs stimulated by SP for 24 hours, and the RNA was reverse transcribed into complementary DNA according to the kit manufacturer’s instructions.SP mRNA expression was measured using a StepOnePlus Real-Time PCR System (ABI, Bedford, MA,USA).Primers for real-time PCR were designed using Primerbank (https://pga.mgh.harvard.edu/primerbank) (Wang et al., 2012).The TAC1 primer sequences were (sense) 5′-AAG CGG GAT GCT GAT TCC TC-3′ and (antisense)5′-TCT TTC GTA GTT CTG CAT TGC G-3′, and β-actin primer sequences were(sense) 5′-CGT TGA CAT CCG TAA AGA CC-3′ and (antisense) 5′-AAC AGT CCG CCT AGA AGC AC-3′.β-Actin was used as the internal control.TAC1 and β-actin mRNA levels were determined by the 2–∆∆Ctmethod (Ma et al., 2021).Each sample was evaluated three times, and the results of at least three independent experiments are presented.
JC-1 staining and MitoTracker staining
Macrophages were washed with PBS for 10 minutes, treated with 20 μM JC-1(T3168, Thermo Fisher Scientific) or 20 μM MitoTracker (M7512, Thermo Fisher Scientific) solution at 27°C for 30 minutes and rinsed with PBS again.Then, the cells were observed under a fluorescence microscope (Hunt Optics and Imaging, Pittsburgh, PA, USA).
Transwell assay
Neutrophils (5 × 105) were added to the Transwell compartment (Corning,Beijing, China), and macrophages (5 × 105) or SP were added to the lower compartment of the 24-well plate.After 24 hours of co-culture, the inserts were removed, and the neutrophils attached to the lower layer of the inserts were fixed with methanol and stained with 0.1% crystal violet (Y0000418,Sigma-Aldrich) for 20 minutes.Then, the neutrophils were counted using a fluorescence microscope.
Oxygen consumption rate (OCR) and extracellular acidification rate (ECAR)
Macrophages were seeded into 24-well plates and then treated with 10 μM SP or 1 μM RP67580 or 1 μg/mL LY93 for 24 hours.To measure mitochondrial respiration, the cells were sequentially treated with 1.5 mM oligomycin (Selleck, Shanghai, China, S1478), 1.5 mM carbonyl cyanide-ptrifluoromethoxyphenylhydrazone (S8276, Selleck), and 1.0 mM rotenone/antimycin A (Sigma-Aldrich, R8875).To measure glycolytic activity, the cells were subsequently treated with 10 mM glucose (Sigma-Aldrich, NIST917C),1 μM oligomycin, and 50 mM 2-deoxy-D-glucose (Selleck, S4710).A Seahorse Xf96 analyzer (Seahorse Bioscience, Boston, MA, USA) was used for evaluation.
Statistical analysis
All data are presented as the mean ± standard error of mean (SEM).Statistical analyses were conducted using GraphPad Prism version 8.0.0 (GraphPad Software, San Diego, CA, USA, www.graphpad.com).An unpaired Student’st-test or one-way analysis of variance withpost hocTurkey’s honest significant difference test was used to determine statistically significant differences between groups.
SP levels are elevated in the injured tissues after laminectomy
To investigate the relationship between SP and epidural fibrosis, peripheral blood was obtained from patients before and 2 days after laminectomy.The SP content in postoperative blood samples was clearly higher than that in preoperative blood samples (Figure 1A), indicating that laminectomy may cause massive SP secretion.To verify this, we established a mouse model of spinal surgery (Figure 1B).Laminectomy caused injury to tissues surrounding the surgical area, including the spinal cord, muscles, and vertebral plates.Compared with the sham group, more SP-positive cells were observed at the wound site in the operated mice via histochemical staining (Figure 1C)and immunofluorescence staining (Figure 1D).Furthermore, increased SP expression was detected in the injured tissues by ELISA (Figure 1E).Collectively, these results suggest that spinal surgery increases SP expression in the injured tissues.
DRG neurons and macrophages at the surgical site secrete SP
DRG neurons are recognized as SP-producing cells (Chen et al., 2014).SP was detected in the DRG neurons (β3-tubulin-positive cells; Deng et al., 2021)isolated from epidural wound tissues harvested from mice that underwent laminectomy (Figure 2A).In vitro, we stimulated DRG neurons with HMGB1,a multifunctional protein that can induce inflammation (Lotze and Tracey,2005).The results showed that HMGB1 promoted SP secretion in DRG neurons (Figure 2B).In addition to DRG neurons, an increasing number of studies have shown that many non-nerve cells also secrete SP, including immune cells (Germonpre et al., 1999), smooth muscle cells (Warner et al.,2000), endothelial cells (Esteban et al., 2021), and airway epithelial cells(Esteban et al., 2009).Therefore, we investigated whether a portion of the SP at the wound site after laminectomy was derived from macrophages.At the wound site, we observed more macrophages that were doubly positive for F4/80 and SP in the operated group compared with the sham group (Figure 2C).Similarly, when HMGB1 was used to stimulate macrophages, the level of SP increased in the macrophage growth medium (Figure 2D).After using clodronate liposomes (Tian et al., 2017) to remove all macrophages in mice that underwent laminectomy, we observed a decrease in SP expression at the wound site, while the mice injected with blank liposomes did not show a decrease in SP in the epidural wound area (Figure 2E), indicating that macrophages play an important role in SP production in epidural wound tissues after spinal surgery.In summary, the SP that is present in epidural wound tissues post-spinal surgery may be produced by DRG neurons and macrophages.
Positive feedback associated with SP secretion by macrophages
Figure 1｜Spinal surgery increases SP expression in the injured tissues.
While investigating SP secretion by macrophages, we found that macrophageproduced SP promoted increased SP secretion by macrophages (Figure 3).A positive feedback loop regulating SP production was previously observed(Bae et al., 1999, 2002).We therefore speculated that SP promotes SP production by macrophages, which would explain the rapid production of SP at the epidural site in the early postoperative period.To test this, we stimulated macrophages with a small dose of exogenous SP and found that TAC1 expression increased (Figure 3A), as well as SP production (Figure 3B).To further understand the relationship between macrophages and SP,we cultured macrophages of two different phenotypes (M1 and M2) with granulocyte-macrophage-colony stimulating factor or macrophage-colony stimulating factor and stimulated them with SP.The results showed that M1 macrophages had a stronger ability to secrete SP than M2 macrophages(Figure 3C) and were better able to respond to the positive feedback stimulus provided by SP (Figure 3D).Treatment with the SP receptor NK1-R inhibitor RP67580 blocked the tendency of M1 macrophages to secrete SP (Figure 3E).We speculated that SP promotes inflammation in the short term, leading to an explosive increase in SP secretion by macrophages.To test this, we assessed secretion of TNF-α, a well-known inflammatory factor (Akdis et al.,2016), after stimulating macrophages with SP.The results showed an upward trend in TNF-α production over the 12 hours following stimulation (Figure 3F).This led us to hypothesize that TNF-α is the key factor influencing SP positive feedback.Therefore, we treated the cells with a TNF-α neutralizing antibody to observe whether it blocked the SP production positive feedback loop.As expected, treatment with the antibody interrupted the positive feedback loop (Figure 3G).Since most of the macrophages at the wound site in the early postoperative period are M1 macrophages (Ge et al., 2021), the responsiveness of M1 macrophages to SP positive feedback may contribute to excessive SP production at the epidural site in the early postoperative period.
Figure 3｜Type 1 macrophages (M1) produce SP.
SP induces M2 differentiation via the sphingolipid metabolic pathway
M2 differentiation of macrophages is important for fibrosis (Tang et al., 2019;Martin and García, 2021).Thus, we next examined whether SP can induce M2 differentiation after laminectomy.Western blot analysis showed that macrophages were induced to differentiate toward the M2 phenotype by SP in a dose- (Figure 4A) and time- (Figure 4B) dependent manner.Again, this M2 differentiation was prevented by treatment with the SP receptor inhibitor RP67580 (Figure 4C).
To explore the intracellular signaling changes in macrophages stimulated by exogenous SP, we conducted high-throughput bulk RNA sequencing of macrophages treated with SP or left untreated.Analysis of the differentially expressed genes (Figure 5A) and KEGG signaling pathway analysis (Figure 5B) highlighted the sphingolipid metabolic pathway as possibly being linked to macrophage differentiation.Sphingomyelin synthase 2 (SGMS2) is an essential component of the sphingolipid metabolic pathway (Pekkinen et al.,2019).Similar to the bulk-RNA seq results, western blotting confirmed that SP increased SGMS2 expression in macrophages (Figure 5C); furthermore,RP67580 blocked the effects of SP on SGMS2 expression levels (Figure 5D).To verify that SGMS2 is involved in SP-induced M2 differentiation, LY93, an inhibitor of SGMS2, was used (Li et al., 2019).The results showed that Arg-1,an M2 marker, was decreased, and iNOS, an M1 marker, was increased when SGMS2 was inhibited by LY93, suggesting that M2 differentiation in the SPtreated macrophages was largely suppressed (Figure 5E).Collectively, these results suggest that SMGM2 plays a key role in the SP-mediated promotion of M2 differentiation.SGMS2 is an enzyme that regulates ceramide and sphingomyelin levels(Ou et al., 2021).As expected, the ELISA results showed that an increase in SGMS2 expression caused a decrease in the concentration of ceramide(Figure 5F), a type of phospholipid that causes substantial damage to the mitochondria (Dany et al., 2016; Dadsena et al., 2019).Therefore, we examined mitochondrial function in SP-treated macrophages.Mitochondrial membrane potential was detected using JC-1 dye (Figure 5G) and MitoTracker Red dye (Figure 5H).Blocking SP with RP67580 or blocking SGMS2 with LY93 suppressed the effects of SP on the mitochondrial membrane potential in macrophages.Moreover, the seahorse glycolysis stress test was used to detect glycolysis and aerobic oxidation in macrophages.We found that SPtreated macrophages had a higher oxygen consumption rate, and both NK-1 inhibitor (RP67580) and SGMS2 inhibitor (LY93) blocked the elevation in oxygen consumption rate (Figure 5I), suggesting that SP/SGMS2 may regulate mitochondrial function.In the seahorse glycolysis stress test, we found that the ECAR level in SP-treated macrophages was not markedly different from that in untreated macrophages (Figure 5J), indicating that SP greatly enhanced mitochondrial function, while glycolysis did not change markedly.In summary, SP induces M2 differentiation via the sphingolipid metabolic pathway, especially SMGM2, which modulates mitochondrial metabolic function and reprograms macrophage metabolism, leading to M2 polarization.
Figure 4｜SP promotes M2 differentiation in vitro.
Figure 5｜ SP regulates metabolism in macrophages.
SP-induced NETs promote M2 differentiation
In addition to metabolic reprogramming, we found an indirect pathway through which SP induces M2 differentiation.When macrophages were stimulated with SP, we found that they secreted more CXCL1 (Figure 6A), a neutrophil chemokine (Drummond et al., 2019).In line with the elevated production of CXCL1, SP-treated macrophages promoted neutrophil migration in a Transwell assay (Figure 6B).We previously showed that NETs increase α-SMA and fibronectin production by macrophages (Jin et al., 2020), and indeed, more NETs (as detected by the immunofluorescence detection of the biomarker citH3) were observed at the injury site in operated mice than in sham mice (Figure 6C).Various stimuli can evoke NET production.We next asked whether SP is involved in NET production.As expected, SP increased the expression of citH3, a biomarker for NETs (Park et al., 2020) (Figure 6D).Moreover, a fibrous extracellular structure that stained positively for citH3 was observed surrounding SP-stimulated neutrophils (Figure 6E).Inhibiting SP with RP67580 diminished citH3 productionin vitro(Figure 6F) and NET formationin vivo(Figure 6G), further suggesting that SP directly triggers NET formation.NADPH oxidase is a key factor in NET generation.When NADPH was inhibited by treatment with apocynin, citH3 expression decreased in SPtreated neutrophils (Figure 6H), indicating that SP induces NET formation through NADPH.
Based on our previous work (Jin et al., 2020), we examined whether SPinduced NETs induce M2 macrophage differentiation.As shown in Figure 6I,SP-induced NETs increased Arg-1 expression but decreased iNOS expression,suggesting that SP-induced NETs may promote M2 differentiation.An elastase inhibitor rescued the changes in Arg-1 and iNOS expression in NETtreated macrophages (Figure 6J), indicating that elastase in NETs may be indispensable for M2 differentiation.
Blocking SP reduces M2 differentiation and alleviates epidural fibrosis
We further tried to block the SP pathway using RP67580in vivoto lessen epidural fibrosis in a mouse model of laminectomy (Figure 7A).The immunofluorescence results showed that a large number of F4/80+CD206+cells were recruited to the epidural wound site 3 days after spinal surgery,which was alleviated by treatment with RP67580 (Figure 7B).Moreover,RP67580 treatment decreased Arg-1 and CD206 expression levels in tissue from the injury site (Figure 7C).HE (Figure 7D) and Masson (Figure 7E)staining were performed on the spinal tissues 30 days after spinal surgery.Inflammatory factors and collagen deposition were increased after surgery,and this phenomenon was alleviated by treatment with RP67580.Finally, we used immunohistochemistry, western blotting, and ELISA analysis to detect the expression of α-SMA (Figure 7F and G), fibronectin (Figure 7G), and collagen I (Figure 7H), which suggest the formation of epidural fibrosis.We found that the increased fibronectin, α-SMA, and collagen I expression in the scar tissues caused by surgery could indeed be lessened by blocking the SP receptor by treatment with RP67580.Collectively, these results suggest that inhibiting SP with RP67580 reduced M2 differentiation in the injured tissues and alleviated epidural fibrosis after spinal surgery.
SP plays important roles in spinal cord injury (Wang et al., 2019) and neuropathic pain (Chen and Marvizon, 2020).Recently, SP was reported to be implicated in various fibrotic diseases (Peng et al., 2019).In the present study,we demonstrated for the first time that SP promotes epidural fibrosis after spinal surgery.In addition to DRG cells, which are well-recognized SP secretors,macrophages, especially M1 macrophages, were also found to secrete SP.Once bound to its receptor NK1-R, SP creates a positive feedback loop in macrophages, which leads to the production of more SP.M1 macrophage differentiation into M2 macrophages breaks the feedback loop, which ceasesthe production of SP.SGMS2, a component of the sphingolipid metabolic pathway, was found to be directly involved in SP-induced M2 differentiation.Inhibiting SGMS2 with Ly93 decreased M2 polarization in SP-treated macrophages.Our previous studies have demonstrated that NETs contribute to M2 differentiation and epidural fibrosis.As expected, SP promoted NET formation.In a murine model of epidural fibrosis, the NK1-R inhibitor RP67580 decreased M2 macrophages in the wound area and extracellular matrix deposition in scar tissues after spinal surgery, suggesting that SP promotes M2 macrophage differentiation in epidural wound fibrosis.Previous studies have found that SP regulates macrophage polarization by affecting the nuclear factor kappa-B (Ni et al., 2016) or PI3K-Akt (Lim et al., 2017)pathway.In our study, transcriptome sequencing revealed that SP regulates M2 differentiation in macrophages by recoding metabolism.SGMS2 is an enzyme that regulates ceramide and sphingomyelin levels (Ou et al., 2021).It transfers the phosphocholine head group of phosphatidylcholine to ceramide to form sphingomyelin and diacylglycerol as a byproduct (Gowda et al., 2011).In brief, it converts ceramide into sphingomyelin, thus reducing ceramide concentrations (Li et al., 2013).However, ceramides are known to play an important role in promoting mitochondrial apoptosis (Bi et al., 2014; Dany et al., 2016; Vaena et al., 2021; Vos et al., 2021).Therefore, we hypothesized that the effect of SP on macrophages was related to the mitochondrial metabolic changes caused by the SGMS2-ceramide pathway.A recent study found that metabolic transitions between glycolysis and mitochondrial oxidative phosphorylation are associated with macrophage polarization.The M1 phenotype primarily relies on glycolysis, while the M2 phenotype requires the tricarboxylic acid cycle and oxidative phosphorylation (Mouton et al., 2020).When the tricarboxylic acid cycle and oxidative phosphorylation are enhanced, macrophages polarize toward the M2 phenotype (Zhou et al.,2020).M1 macrophages generally have a lower mitochondrial membrane potential, and macrophages polarized toward the M2 phenotype have an increased mitochondrial membrane potential (Hinshaw et al., 2021).It is well known that there is a positive relationship between neuroinflammation and epidural fibrosis, but the specific mechanisms are not well understood.As the most abundant leukocytes in inflammation, neutrophils infiltrate the wound area after spinal surgery.Previously, we demonstrated that HGMB1, as an alarmin, triggers the formation of NETs, which promotes the differentiation of macrophages into myofibroblasts (Jin et al., 2020).In the present study, we found that SP promotes NET formation, which polarizes macrophages to the M2 phenotype in injured tissues after spinal surgery.These findings provide a deeper understanding of the link between neuroinflammation and epidural fibrosis.
Figure 6｜SP promotes neutrophil infiltration and NET formation.
Figure 7｜An NK1-R inhibitor (RP67580) alleviates epidural fibrosis.
As mentioned earlier, there is no genuinely effective therapy for epidural fibrosis in clinical practice.RP67580 has previously been reported to be an effective treatment for neuropathic pain (Chen and Marvizon, 2020)and myocardial infarction (Jeong et al., 2020) in animal disease models.SP inhibitors are safe for clinical application (Mathew et al., 2011).For the first time in this study, RP67580 was used to inhibit epidural fibrosis,and the results were satisfactory.Previously, we applied DNase combined with a temperature-sensitive hydrogel (Sun et al., 2022), which markedly reduced epidural scarring.We would like to try to combine SP inhibitor with temperature-sensitive hydrogel in the future, as a potential direction for further research in the clinical prevention of epidural fibrosis.The study had some limitations.First, the clinical sample size was small.The level of SP in the wound drainage post-spinal surgery was markedly increased compared with that in plasma isolated from peripheral blood.To confirm this marked increase in SP after spinal surgery, more patients should be tested.Second, the detailed mechanism by which SP participates in macrophage differentiation warrants further research.SP directly induced M2 macrophages, in which SP production was inhibited, and SP induced the formation of NETs, which promoted M2 macrophage polarization.Further experiments are required to determine whether SP plays a direct or indirect role inducing M2 macrophage polarization.Last but not least, in this mouse model of epidural fibrosis SP activity was inhibited by treatment with NK 1 receptor antagonist RP67580.In future studies, the roles of SP in epidural fibrosis should be validated in SP knockout or NK1-R knockout mice.
In summary, in this study we demonstrated that macrophages at the spinal injury site released SP, which promoted the differentiation of M1 macrophages into M2 macrophages and ceased the production of SP.SGMS2, a component of the sphingolipid metabolic pathway, was directly involved in the SPmediated promotion of macrophage differentiation.Indirectly, SP induced NET formation, which further boosted M2 macrophage differentiation.In a mouse model of epidural fibrosis, inhibiting SP decreased the number of M2 macrophages in the wound area after spinal surgery and alleviated epidural fibrosis.Our findings provide the basis for developing novel strategies for the treatment of epidural fibrosis.
Acknowledgments:We appreciate Dr.Lu Zhou from the School of Pharmacy,Fudan University, for the gift of LY93.
Author contributions:JL and MSZ conceived and designed the study.FH, HRW, YFB, JPS and WSW performed the experiments.FH drafted the manuscript.MSZ, YX and JL critically revised the manuscript.All authors approved the final version of this manuscript.
Conflicts of interest:The authors declare no conflicts of interest.
Data availability statement:All relevant data are within the paper and its Additional files.
Open access statement:This is an open access journal, and articles are distributed under the terms of the Creative Commons AttributionNonCommercial-ShareAlike 4.0 License, which allows others to remix, tweak, and build upon the work non-commercially, as long as appropriate credit is given and the new creations are licensed under the identical terms.
Additional file:Additional Table 1:Clinical characteristics of recruited patients with lumbar disc herniation.