Melittin Regulates Iron Homeostasis as a Key Mediator of Macrophage Polarization in Rat Lumbar Spinal Stenosis


 BackgroundLumbar spinal stenosis (LSS) is defined as the narrowing of the spinal canal, which compresses the nerves traveling through the lower back into the legs. Inflammation is the most common cause of LSS. Chronic pain induced by nerve damage results from chronic inflammation, and the inflammation response worsens with elevated iron stores. Furthermore, macrophage polarization to the M1 (inflammatory) or M2 (anti-inflammatory) type is essential for controlling host defense or repairing tissues. However, the precise function of macrophage polarization in iron release or retention in LSS pathophysiology is not well-understood. Here, we introduce melittin to modulate macrophage polarization related to iron metabolism for LSS treatment.MethodsPrimary peritoneal macrophage were cultured in 200 or 500 ng/mL of melittin and FeSO4-containing medium for 24 h. Macrophage polarization was assessed by Immunofluorescence staining to CD86 or Arg1 antibodies. In an in vivo rat model of LSS, melittin were administered at 100 and 250 µg/kg, and in vivo effects of melittin on iron deposition-induced macrophage polarization was evaluated by immunochemistry, real time-PCR, western blot, and flow-cytometry. The locomotor functions were assessed by BBB, ladder scoring, and Von Frey test for up to 3 weeks. ResultsIn vitro experiments demonstrated that macrophages can be polarized toward an M2 phenotype after melittin treatment in iron-insulted primary macrophages. Treatment with 100 and 250 μg/kg melittin in a rat LSS model increased the proportion of M2 macrophages in the damaged spinal cord. Moreover, we found that melittin attenuated iron overload-induced M1 polarization via regulating iron metabolism-related genes in LSS rats. As a result, melittin improved locomotor recovery and stimulated axonal growth following LSS.ConclusionsMelittin can promote functional recovery in LSS models by activating M2 macrophages via controlling macrophage iron metabolism, suggesting the potential applications of melittin for treating LSS.

through the lower back into the legs leads to lower back pain and complex neurologic symptoms [3,4].
There are various reasons for the narrowing of the spinal canal, with hypertrophy of the lumbar facet joints or ligamentum ava being a major cause [5,6]. The ligamentum avum is a ligament that supports the facet joints and is composed of broblasts and extracellular matrix. Degenerative changes trigger in ammation and brosis, and the hardened and thickened ligamentum ava compress the nerves [7]. In ammation is a major factor that induces hypertrophy of the ligamentum ava, during which proin ammatory factors are released around the compressed nerves to exacerbate symptoms [8][9][10].
Particularly, LSS is a slow-progressing degenerative process compared to other conditions such as herniated disc. Ameliorating in ammation caused by the narrowing of the spinal canal to reduce pain has been proposed as a basic treatment for the condition rather than restoring the spinal canal to the original state [11,12]. Therefore, as discussed in numerous previous studies, in ammation is central to the mechanism that causes pain and degeneration of the lumbar spine. However, few studies have focused on how in ammatory mediators contribute to lower back pain in LSS.
According to a recent study, macrophages play a critical role in the initiation, maintenance, and resolution of in ammation, and excessive iron deposition induces in ammatory responses through M1 macrophage polarization [13]. In addition, the iron accumulated M1 macrophages in the damaged spinal cord tissues show increased expression of the in ammatory cytokine tumor necrosis factor [14].
High iron levels and iron-related virulence factors have been reported as risk factors for neurodegenerative disorders, including Alzheimer's disease, Parkinson's disease, and multiple system atrophy [16], indicating that regulation of iron levels in macrophages is an effective strategy for fundamental control of in ammation. Herein, we introduce the melittin to modulate the iron metabolism in macrophage polarization after LSS.
Bee venom is a complex substance that has been used clinically as a medicinal acupuncture for the treatment of intervertebral disc disease. Melittin, the major peptide component constituting 50% of bee venom, binds to various types of cells. It has anti-in ammatory and antibacterial, strong analgesic, and immune system-boosting effects [17,18]. Particularly, melittin has been known to induce antiin ammatory effects by preferentially binding to M2 macrophages [19,20]. Therefore, we hypothesized that melittin could control macrophage polarization by regulating the homeostasis of iron metabolism in macrophages in LSS and evaluated its potential therapeutic effects in terms of pain relief and functional recovery.

Methods
In vitro cultures for peritoneal macrophages Peritoneal macrophages were obtained from male Sprague-Dawley (SD) rats (10-12 weeks old; Daehan Bio Link, Chungju, Korea). All procedures were approved by the Jaseng Animal Care and Use Committee (JSR-2019-09-022-001). Rats were housed in standard cages at a constantly controlled temperature (23-25°C) and humidity (45-50%) with a 12-h light/dark cycle. All animals had free access to food and water.
Rats were injected intraperitoneally with 10 ml of 3% (w/v) thioglycollate (Becton Dickinson, Sunnyvale, CA, USA). Rats were euthanized from 4 days after injection and their peritoneal cavity was opened. The peritoneal uid was harvested from the peritoneal cavity by rinsing 40 mL of RPMI-1640 (Hyclone, South Logan, UT, USA), and then centrifuged at 4°C for 3 min at 1,500 rpm. The supernatant was discarded, and cell pellets were resuspended in 1 mL of red blood cell lysis buffer (Invitrogen, Carlsbad, CA, USA) at room temperature for 5 min and then washed in phosphate-buffered saline (PBS; Gibco, Grand Island, NY, USA).

Iron(II)sulfate heptahydrate (FeSO 4 ) and melittin treatment
Iron-mediated macrophage polarization was performed in an in vitro setting as described previously [21]. Brie y, primary macrophages were cultured in a poly-d-lysin coated plates for 24 h after cell seeding. For FeSO 4 (Sigma-Aldrich, St. Louis, MO, USA) treatment after 24 h, an iron stock solution was prepared using 278 mg of FeSO 4 dissolved in 10 mL PBS. After ltering using a 0.2 µm membrane, FeSO 4 was added to the cell culture media at a nal concentration of 50 μM for 30 min. After 30 min of incubation at 37°C in a CO 2 incubator, melittin was added to the FeSO 4 -containing medium at a concentration of 200 or 500 ng/mL. The cells were further incubated in 5% CO 2 at 37°C for 24 h, and samples were used for immunocytochemistry.

Immunocytochemistry
The macrophages were cultured in 200 or 500 ng/mL melittin, and the expression of macrophage M1 (CD86) and M2 (Arginase 1; Arg1) markers was detected by immunostaining after 24 h with ve biological replicates per condition. Brie y, the samples were xed with 4% paraformaldehyde (PFA; Biosesang, Seongnam, Korea) for 30 min and washed three times for 5 min each with PBS. The cells were permeabilized with 0.2% Triton X-100 in PBS for 5 min, washed two times and blocked with 2% normal goat serum (NGS) in PBS for 1 h. The following primary antibodies were used: Arg1 (1:200, Novus, Littleton, CO, USA), CD68 (1:500, Abcam, Cambridge, UK), CD86 (1:100, BD PharMingen, San Diego, CA, USA). The primary antibodies were diluted in 2% NGS and incubated overnight at 4°C. After washing with PBS three times for 5 min each, the samples were then incubated for 2 h with uorescein isothiocyanate (FITC) or rhodamine-conjugated secondary antibodies (goat anti-rabbit IgG or goat antimouse IgG; Jackson ImmunoResearch Laboratories, West Grove, PA, USA) diluted at 1:300 in 2% NGS. After 2 h of incubation at room temperature, the cells were washed three times for 5 min with PBS, mounted with uorescence mounting medium (Dako Cytomation, Glostrup, Denmark), and images were acquired by confocal microscopy (Eclipse C2 Plus; Nikon, Tokyo, Japan). For quanti cation of macrophage polarization, ve representative images were captured at 200 ×magni cation with xed acquisition parameters and analyzed using counting of ImageJ software. The numbers of CD86 + or Arg1 + /ED1 + cell was counted and the proportion of M1 or M2 double-positive cells was expressed as a percentage of cells that appear bright in response to CD86 or Arg1 by the number of ED1 + cells.

Rat LSS model and melittin administration
Male SD rats (7 weeks old, 230-250 g) were obtained from Daehan Bio Link (Chungju, Korea). All procedures were approved by the Jaseng Animal Care and Use Committee (JSR-2018-10-001-A). Rats were housed as described above. The animals were deeply anesthetized with 2-3% iso urane gas (Forane; BK Pham, Goyang, Korea), and a dorsal laminectomy was performed at L5 using ne rongeurs.
Next, a silicone block (80 kPa, 4 × 1 × 1 mm 3 ) was inserted at the L4 level using no. 5 ne forceps. Shamoperated rats underwent laminectomy only at the L5 level without silicone insertion. The spinal cord was covered with Surgicel® absorbable hemostat (Johnson and Johnson, Arlington, TX, USA) to help achieve hemostasis in the spinal cord. To prevent infection, all rats were injected intramuscularly with 40 mg/kg cefazolin sodium (Cefazolin®, Chong-Kun-Dang Pharm., Seoul, Korea) after suturing. All rats were also administered an oral dose of 10 mg/kg Children's Tylenol® (Janssen Korea, Inc., Seoul, Korea) after the anesthesia was resolved for pain management. Different concentrations of melittin (100 and 250 μg/kg) were administered once per day by subcutaneous injection. Injection was started at 30 min after LSS and continued for 1 or 3 weeks after LSS before sacri ce. The control group was administered the same volume (400 μL) of PBS.

Histology
Sham or LSS rats from each group were deeply anesthetized with 2-3% iso urane gas and perfused via cardiovascular system with 0.9% normal saline (Sigma-Aldrich) and 4% PFA (Biosesang) for histological staining and immunohistochemistry. The spinal cord at the silicone implantation site was dissected, postxed overnight in 4% PFA at 4°C and dehydrated with 30% sucrose in 0.1 M phosphate buffer for 3 days.
The samples were cryo-sectioned at 20 μm in the sagittal plane. Hematoxylin and eosin (H&E) staining was performed at the L4 level to evaluate the degree of damage to the spinal cord caused by implantation of the silicone block at 3 weeks. Brie y, the sectioned tissue slides were dipped into hematoxylin for 2 min, washed in running tap water for 2 min, and then stained with eosin for 5 s. The stained sections were dehydrated through a graded series of ethanol, cleared with xylene, mounted with washed three times with PBS and secondary antibodies ( uorescein isothiocyanate-coupled goat antimouse or anti-rabbit or rhodamine-conjugated goat anti-guinea pig, Jackson ImmunoResearch Laboratories, West Grove, PA, USA) were treated at a 1:300 dilution in 2% normal goat serum. Following incubation at room temperature for 2 h, the sections were washed three times with PBS and mounted with using Dako Mounting Medium (Dako Cytomation). The stained tissue sections were observed by confocal microscopy (Eclipse C2 Plus, Nikon). The uorescence intensity was analyzed for confocal images captured by the same acquisition settings. Background was subtracted using the subtract background tool in ImageJ software (1.37 v, National Institutes of Health, Bethesda, MD, USA). CD68positive macrophages were manually counted in the silicone implanted area using confocal microscopy images obtained at 100× magni cation. The result was expressed as a percentage or ratio. NF200labeled axons were quanti ed by using ImageJ. Brie y, NF200-labeled axons were captured at the implantation site under a 100× objective lens of a confocal microscope, and the number of pixels occupied by the NF200 bers was counted in three images; NF200-positive pixels were divided by the number of pixels in a eld.
Quantitative reverse-transcription PCR (qRT-PCR) was performed in triplicate using iQ SYBR Green Supermix with a CFX Connect Real-Time PCR Detection System (both from Bio-Rad, Hercules, CA, USA). Target gene expression was normalized to that of the β-actin gene and expressed as a fold-change relative to the control group.

Flow cytometry
Spinal cord tissues were isolated into single cells for FACS analysis. The tissue under silicone was dissociated using a Neural Tissue Dissociation kit (Miltenyi Biotec, Bergisch Gladbach, NRW, Germany) and gentleMACS Dissociator (Miltenyi Biotec). The dissociated cells were ltered in 20 μm cell strainer (FALCON) and centrifuged at 2,000 rpm for 3 min. Flow cytometry was performed to assess cell death following LSS. Apoptotic cell death was detected using a FITC Annexin V Apoptosis Detection Kit I (BD Bioscience, Franklin Lakes, N, USA). Brie y, the cells were collected and stained with 1 µL of Annexin Vphycoerythrin and 1 µL of propidium iodide in 200 µL of 1× binding buffer and directly analyzed by uorescence-activated cell sorting (Accuri C6 Plus Flow Cytometer, BD Biosciences). In addition, ferroportin expression was evaluated by ferroportin/SLC40A1 antibody (Novus) after xation with xation buffer (BD Bioscience). The mean positive cell values, as determined via ow cytometry, were expressed as a percentage relative to the control group.

Enzyme-linked immunosorbent assay
The expression levels of the pro-and anti-in ammatory markers IL-6 and IL-10 in a separated spinal cord, including at the implantation site, were evaluated by enzyme-linked immunosorbent assay (ELISA). The spinal cord tissues were homogenized in radioimmunoprecipitation assay (RIPA) buffer (GenDEPOT, Barker, TX, USA) containing a proteinase inhibitor (Millipore) using a Taco TM Prep Bead Beater (GeneReach, Taichung, Taiwan) and centrifuged at 1,000 rpm at 4°C for 3 min. Protein concentration was quanti ed using a bicinchoninic acid (BCA) protein assay kit (Thermo Fisher Scienti c, Waltham, MA, USA). The supernatants were examined using ELISA kits (BD Biosciences, Franklin Lakes, NJ, USA) according to the manufacturer's instructions. The nal concentrations of cytokines were calculated with respect to the amount of each protein.

Functional assessments
Locomotor function was assessed for 3 weeks using three tests after LSS. The Von Frey test was used to measure the foot response to pain. The rats were placed in acrylic cages for 15 min and then stimulated at the test area on the plantar surface. We measured the latency of paw withdrawal in response to mechanical stimulation using a Von Frey lament (Ugo Basile, Varese, Italy). Each value was automatically recorded from three or more measurements and then averaged. The BBB scale was assessed in an open eld for 4 min by two independent observers, and the BBB score was calculated from the average of hindlimb movements. The ladder walking test was also performed to con rm the balance and forelimb-hindlimb coordination. All rats walked the runway (2.5 cm interval) from left to right three times, and their movements were recorded using a digital camera and calculated as follows: ladder score (%) = erroneous steps of hind limb/total steps of hind limb × 100.

Results
Melittin promotes a phenotypic switch of FeSO 4 -induced M1 polarization in primary macrophages First, the phenotypic characterization of melittin-treated macrophages was investigated by immunocytochemical staining under in vitro conditions that simulate FeSO 4 -induced M1 polarization.
The immunocytochemical double staining for M1 (CD86 + ) or M2 (Arg1 + ) markers with macrophages (CD68 + ) was carried out ( Figure 1A). The population of M1 macrophages expressing CD86 marker was signi cantly increased by 35.21% after FeSO 4 treatment, while it signi cantly decreased from 17.05-12.71% in the melittin groups compared to that in the FeSO 4 group, showing a dose-dependent inhibition in its action of inducing M1 polarization ( Figure 1B). By contrast, the percentage of Arg1 + M2 macrophages was shown to increase signi cantly with increasing melittin dose in FeSO 4 -treated macrophages ( Figure 1C). In addition, macrophages treated only with 500 ng/mL melittin did signi cant increase in the percentage of Arg1 + M2 macrophages ( Figure S1). Our ndings demonstrate that melittin induces a phenotypic switch of macrophage polarization into an M2 anti-in ammatory state.
In vivo administration of melittin to rats with LSS reduces in ammatory response and apoptotic cell death To investigate the general histologic characteristics of the spinal cord after silicone implantation in the spinal canal, we performed H&E staining of the sagittal sections at 1 week after LSS. Compression of the spinal cord was observed in the stained sections ( Figure S2). Macrophage in ltration into the compressed spinal cord underneath the silicone block was detected by immunohistological staining. CD68-positive cells were not detected in the sham group, whereas CD68-positive cells were densely packed in the spinal cord after LSS and dramatically decreased after melittin administration in a dosedependent manner (Figure 2A). We further analyzed the presence of CD68-positive macrophages at 3 weeks after LSS. Overall, a small population of CD68-positive cells was detected in each group compared to those at 1 week. However, the signi cant pattern at 3 weeks is similar to that in animals that were administered melittin for 1 week after LSS induction and the number of macrophages positive for CD68 was also shown to decrease signi cantly in the melittin groups compared to that in the control group ( Figure 2B). Changes in IL-6 and IL-10 levels in the spinal cord were determined by ELISA. IL-6, a proin ammatory cytokine, is a key player leading to an in ammatory response and tissue damage in the spinal cord. The level of IL-6 was signi cantly higher in the control group than in the sham group at 1 week after LSS and was signi cantly and dose-dependently decreased after melittin administration in LSS rats ( Figure 2C). Additionally, the level of IL-10, an anti-in ammatory cytokine that was signi cantly lower in the control group than in the sham group at 1 week after LSS, was dose-dependently increased at 1 week after melittin administration in LSS rats ( Figure 2D). With regard to cell death at 1 week after melittin administration in LSS rats, we performed further analysis in spinal cord tissue using ow cytometric assay ( Figure 2E). The spinal cord cells in early apoptosis were stained with Annexin V + /PI − , whereas cells in late apoptotic stage were stained with Annexin V − /PI + . We found that most spinal cord cells at 1 week after LSS were early apoptotic, and a relatively high percentage of Annexin V + /PI − cells were observed in 25.5% of the control group compared with that in the sham group. When animals were administered with 100 and 250 µg/kg melittin for 1 week, the apoptotic population of isolated cells from the spinal cord was approximately 14.01% and 9.6%, respectively ( Figure 2F). These ndings reveal that melittin effectively reduced the in ammatory response and apoptotic cell death in the spinal cord after LSS.
Melittin ameliorates LSS-induced iron accumulation, which was released from activated macrophages in the spinal cord We next examined whether the CD68-positive macrophage population is linked to impaired iron storage after LSS. Macrophages play key roles in regulating systemic iron homeostasis, which is closely linked to macrophage polarization [22]. Iron overload is indicated by elevated ferritin levels, which is related to the amount of iron stored in cells. At 1 week after LSS, immunohistological staining revealed an increase in the number of ferritin-positive macrophages in the spinal cord after LSS, whereas their population signi cantly decreased in the melittin group ( Figure 3A). The expression levels of iron metabolism related genes, including TFRC, DMT1, ferritin heavy/light chain (FTH1/FTL), CP, HAMP, and ferroportin, were evaluated in the sham, control, and melittin groups at 1 week. qPCR analysis of the spinal cord was performed to con rm the changes in the expression of genes related to iron metabolism at 1 week after melittin administration in LSS rats. TFRC and DMT1 expression was signi cantly upregulated in the control group compared to that in the sham group and signi cantly downregulated after melittin administration. These ndings demonstrate that LSS induces an increase in iron intake by cells through TFRC and DMT1 and a decrease in iron intake after melittin administration ( Figure 3B, C). We also evaluated changes in the expression levels of the iron storage genes FTH1/FTL. FTH1 and FTL expression levels were signi cantly higher in the control group than in the sham group at 1 week. In contrast to that in the control group, melittin administration to LSS rats signi cantly downregulated FTH/FTL1 gene expression ( Figure 3D, E). We also investigated changes in the expression of CP and HAMP. CP expression was signi cantly downregulated in the control group compared to that in the sham group but was upregulated after melittin administration in LSS rats ( Figure 3F). Therefore, the upregulation of CP is mainly associated with elevated ferroxidase activity to export iron via FPN after melittin administration in LSS rats. We also con rmed that HAMP expression was signi cantly increased in the control group. In contrast, HAMP expression was signi cantly decreased after melittin administration in a dose-dependent manner ( Figure 3G). These results demonstrate that elevated HAMP expression can inhibit iron export by binding to FPN and inducing iron deposition within the spinal cord in LSS rats, whereas low expression of HAMP can induce iron e ux to maintain iron homeostasis after melittin administration in LSS rats. We further investigated the iron exporter ferroportin levels in spinal cord cell isolated from tissues of each group using ow cytometry ( Figure 3H). FPN positivity dramatically decreased owing to LSS induction, whereas administration of melittin with 100 or 250 µg/kg reversed this, with the positivity signi cantly increasing in LSS rats compared to that in the control ( Figure  3I). Thus, these ndings strongly suggest that melittin will be an effective treatment for maintaining iron homeostasis by regulating iron metabolism-related genes after LSS.
Melittin reduces M1 macrophage activation by suppressing iron accumulation Next, we investigated whether intracellular iron deposition induces M1-type macrophage polarization after LSS. M1 macrophages are rich in ferritin and prone to iron accumulation [23]. CD86 is widely used as a marker to identify M1 macrophages. We performed immunohistological staining using CD86 and ferritin antibodies ( Figure 4A). CD86/ferritin dual-stained cells were detected at 1 week after LSS. This nding suggests that LSS can induce M1 macrophage polarization with increased iron storage ferritin expression. However, CD86/ferritin dual-stained cells were signi cantly decreased in the 250 µg/kg melittin group ( Figure 4B). Furthermore, qPCR revealed increased gene expression of iNOS, COX-2, IL-1β, IL-6, and TNF-α in the spinal cord after LSS at 1 and 3 weeks, whereas the expression of these genes was signi cantly decreased in the melittin group, which was associated with a reduction in the in ammatory response owing to suppression of M1 macrophages activation ( Figure 4C-G). These ndings strongly suggest that melittin administration can reduce M1-type cell polarization by inhibiting iron deposition.

Melittin promotes M2 macrophage polarization and inhibits iron deposition
In our next set of experiments, we con rmed that melittin modulates macrophage polarization toward the M2 phenotype by inhibiting iron deposition. M2 macrophages are characterized by the expression of Arg1, a key enzyme produced by this type of macrophages, and the mannose receptor MRc1 (CD206). We performed immunohistological staining of Arg1 to identify M2-type cell polarization following melittin administration to LSS rats. In addition, we assessed ferritin expression in Arg1-positive M2 macrophages by double staining of ferritin and Arg1 ( Figure 5A). Interestingly, Arg1 expression was reduced in the damaged spinal cord at 1 week after LSS. However, ferritin expression was dramatically increased after LSS. In contrast, the number of M2 macrophages positive for Arg1 was signi cantly increased, whereas ferritin expression was decreased in the melittin groups ( Figure 5B). The mRNA expression of Arg1 was analyzed at 1-and 3-weeks post-operation. The expression level of Arg1 did not differ among the control and melittin groups but showed signi cant upregulation at 3 weeks in the melittin groups compared to that in the control group ( Figure 5C). The representative anti-in ammatory gene IL-10 was also signi cantly upregulated at 1 week in the melittin groups, whereas in 3 weeks after LSS, there was no signi cant differences between groups ( Figure 5D). Therefore, melittin can induce macrophage polarization to an M2 phenotype in LSS rats.

Melittin relieves neuropathic pain by inhibiting TRPV1 in LSS rats
To explore the antinociceptive effects of melittin on TRPV1 expression in dorsal root ganglionic (DRG) sensory neurons, we performed immunohistochemical staining by using TRPV1 from the rat spinal cord at 3 weeks after LSS. TRPV1 is a well-known pain-mediating ion channel expressed in sensory neurons, including DRG neurons, trigeminal ganglionic neurons, and vagal neurons. TRPV1 expression in DRG neurons was increased dramatically after LSS. In contrast, TRPV1 expression in NeuN-positive neurons was dramatically decreased after melittin administration in LSS rats ( Figure 6A). Fluorescence colocalization was quanti ed as the percentage of neurons expressing TRPV1 in NeuN-positive DRG sensory neurons at 3 weeks after LSS. A signi cant increase in the percentage of TRPV1-NeuN neurons in the DRG was observed in the control group at 3 weeks after LSS ( Figure 6B). In contrast, increased TRPV1 expression was dose-dependently inhibited by melittin administration to LSS rats. Moreover, the mean TRPV1 intensities showed similar expression patterns to those observed in the percentage of TRPV1positive neurons in the DRG ( Figure 6C). Therefore, melittin ameliorates neurotrophic pain by downregulating the expression of TRPV1 in the DRG of LSS rats.

Melittin promotes locomotor recovery by preventing axonal loss in LSS rats
We next examined the effect of melittin on axons damaged by implanted silicone by immunohistochemistry of NF200 (an axonal marker) in sagittal spinal cord sections from the implantation site. Axonal damage after LSS appeared to be induced by the implanted silicone. The damaged spinal cord showed improved axonal growth after administration of different concentrations of melittin ( Figure 7A). We also quanti ed the NF200-positive densities and intensities in the spinal cord at the implantation site, which were greater in the melittin groups than in the control group ( Figure 7B, C). Furthermore, the changes in several regeneration-related genes, including NF200, Nrf-2, Wnt3, and Wnt5a, were analyzed by qPCR at 1 and 3 weeks after LSS. NF200 gene expression differed between groups at 1 and 3 weeks after LSS ( Figure 7D). However, the mRNA level of Nrf-2 in the spinal cord was signi cantly higher in the 250 µg/kg melittin group than in the control group at 1 week, whereas there was no signi cant difference between groups at 3 weeks ( Figure 7E). The Nrf-2 pathway plays a critical role in protecting against oxidative stress and inducing regeneration of various tissues. In addition, Wnt signaling has been linked to axon regrowth and transport through the Wnt/β-catenin pathway (Wnt1 and Wnt3a) or Wnt/PCP pathway (Wnt5a). The expression levels of Wnt3 and Wnt5a were signi cantly and dose-dependently increased by melittin administration at 1 week ( Figure 7F, G). These ndings suggest that melittin can promote axonal growth in LSS rats.
Finally, locomotor functions were assessed for up to 3 weeks using three methods (BBB, ladder, and von Frey tests) to determine whether melittin improved functional recovery after LSS. The melittin groups showed increased BBB scores compared to those in the control group until 3 weeks. An average BBB score of 16 points was observed at 3 weeks in the control group. The melittin groups had an average higher BBB score for open eld locomotion at 3 weeks after LSS than the control group, with signi cant differences observed between the melittin groups and control group at 3 weeks; the scores in the melittin groups were approximately two points higher than that in the control group ( Figure 7H). The horizontal ladder test revealed more noticeable behavioral differences. The control group had a higher foot fault rate of approximately 10% for 3 weeks than the sham group, whereas the melittin groups showed a decreased foot fault frequency compared to that in the control group at 3 weeks. However, the 250 µg/kg melittin group showed a signi cant decrease in the foot fault rate for up to 3 weeks ( Figure 7I). In addition, we assessed the effect of melittin on mechanical allodynia using the von Frey test. The control group showed an average withdrawal latency of 4 s for 3 weeks. There was a signi cant difference between the sham and control groups for up to 3 weeks. In contrast, rats administered different concentrations of melittin showed signi cantly lower sensitivity than the control group at 3 weeks ( Figure 7J). Therefore, melittin improved locomotor functional recovery after LSS in rats by promoting axonal regrowth.

Discussion
In this study, we examined the in ammatory responses as the increase in the number of macrophages that penetrated the spinal cord tissues damaged by LSS. Speci cally, iron overload was mainly observed in M1 macrophages, whereas iron accumulation was not con rmed in M2 macrophages, which are known to secrete anti-in ammatory cytokines. In addition, administration of melittin, which promotes macrophage transformation from M1 to M2, decreased M1 macrophages and ferritin expression levels. In contrast, increases in M2 macrophages anti-in ammatory factor expression were observed. Moreover, these changes were closely associated with iron metabolism. We found that melittin could regulate iron balance by regulating of TFRC and DMT1, which transport iron into cells. Under the LSS condition, TFRC and DMT1 expression was elevated, and iron was mostly stored in the form of ferritin protein. Originally, to maintain a stable iron homeostasis, CP can be upregulated to increases the oxidation of ferrous iron to ferric iron and facilitate iron export through FPN. Hepcidin, another mediator of iron metabolism, can regulate iron transport by binding to FPN. However, CP was downregulated, and HAMP was upregulated after LSS. Meanwhile, melittin has been shown to be involved in LSS-induced changes in iron metabolism.
Aging involves the deterioration of cell functions and is a major cause of numerous diseases related to failed cellular homeostasis. Various neurodegenerative disorders such as Alzheimer's disease and Parkinson's disease are closely associated with aging, and a decrease in homeostasis due to the accumulation of oxidized proteins and reactive oxygen species (ROS) has been reported [24,25]. Cells possess an antioxidant system that effectively removes ROS to maintain the balance between oxidants and antioxidants as well as homeostasis [26]. In particular, ROS produced in large quantities by immune cells such as macrophages and neutrophils play an important role in removing foreign substances during infection or in ammatory responses. However, continuous and excessive ROS production and long-term accumulation causes damage to cells and tissues and promotes chronic in ammation and aging [24]. Therefore, failed homeostasis is the underlying cause of oxidative stress and in ammation in degenerative diseases, indicating that homeostasis can be regulated to treat diseases.
Iron is an essential trace metal in metabolism for maintaining homeostasis in humans, and many studies have focused on the relationship between the molecular mechanisms regulating iron metabolism and degenerative disorders related to aging [27]. Iron gradually accumulates in speci c cells and tissues during aging and accelerates the decline in mitochondrial functions [28]; iron overload acts as a catalyst in a reaction that produces ROS, thereby damaging tissues and causing diseases.
We con rmed macrophage polarization according to iron levels following LSS and the anti-in ammatory effect and functional recovery through the regulation of macrophage polarization. However, the LSS animal model in this study arti cially compresses the nerves when bio-silicone is implanted on the ligamentum ava of the spinal canal, and hypertrophy of the ligamentum ava was not directly induced.
Therefore, further studies are needed to investigate the direct relationship between changes in iron metabolism-related factors in hypertrophic ligamentum ava tissue or cells, oxidative stress, and in ammation due to the iron metabolism imbalance. Therefore, the changes in the regulatory factors related to iron metabolism, homeostasis following activation of the brosis of ligamentum ava, and therapeutic effect of iron homeostasis regulation should be further examined. Consent to participate: Not applicable.

Consent for publication
Not applicable.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The author declare that they have no competing interests.
(C) Quanti cation of uorescence intensity in TRPV1-positive neurons within DRG. Data are expressed as the means ± SEM. Signi cant differences are indicated as #### p < 0.0001 compared vs. the sham group, ***p < 0.001 and ****p < 0.0001 vs. the control group were analyzed via one-way ANOVA with Tukey's post-hoc test. Figure 7