MIF downregulation attenuates neuroinflammation via TLR4/MyD88/TRAF6/NF-κB pathway to protect dopaminergic neurons in Parkinson’s disease model

Abstract

The progression of Parkinson’s disease (PD) is closely associated with neuroinflammatory responses and microglial activation. Consequently, research targeting microglia-mediated neuroinflammation has garnered increasing attention. Macrophage migration inhibitory factor (MIF), a multifunctional cytokine, is implicated in neurodegenerative pathologies, including PD. However, its precise regulatory mechanisms in PD-associated microglial activation and neuroinflammatory cascades remain incompletely characterized. In this study, we observe that MIF exacerbates the pathogenesis of PD through its pro-inflammatory effect, and downregulation of MIF could ameliorate motor behavior deficits, attenuate neuroinflammation, and protect midbrain dopamine (DA) neurons in PD mice. Mechanistically, MIF downregulation attenuates neuroinflammation and exerts neuroprotection against microglia-induced neuronal injury and degeneration by regulating the TLR4/MyD88/TRAF6 signaling axis. In conclusion, this study elucidates the pivotal role of MIF in regulating neuroinflammation associated with PD, suggesting that MIF may be a potential therapeutic target for intervening in PD progression, and providing new strategies for PD treatment.

Downregulation of MIF attenuates neuroinflammation by inhibiting MIF activation of the TLR4/MyD88/TRAF6/NF-κB pathway, which ultimately protects dopaminergic nerves from microglia activation-induced neurological injury and degeneration.

Introduction

Following Alzheimer’s disease, Parkinson’s disease (PD) is the second-most prevalent neurodegenerative disease¹. Approximately 1-2% of individuals over 65 are affected by PD, primarily affecting the motor nervous system². Motor and non-motor symptoms are commonly observed in individuals diagnosed with PD. Resting tremor, cogwheel rigidity, abnormal postural gait, and bradykinesia are examples of motor symptoms, which may be accompanied by non-motor symptoms like olfactory loss, cognitive decline, pain and fatigue, depression, constipation, and sleep disturbance³. As patients age, these symptoms emerge and deteriorate, significantly impacting their quality of life⁴. PD is characterized by the degeneration and demise of dopaminergic neurons in the substantia nigra pars compacta (SNpc), a notable decrease in the dopamine (DA) neuron levels in the striatum, and the occurrence of eosinophilic inclusion bodies (Lewy bodies) in the cytoplasm of remaining neurons in the substantia nigra⁵.

The development of PD and other neurodegenerative diseases is strongly associated with neuroinflammation^6,7. Previous research demonstrated that neuroinflammatory sources of oxidative stress and cytokine-dependent toxicity (cytokines, chemokines, and other inflammatory mediators can trigger microglia activation) may lead to degeneration of the nigrostriatal pathway and accelerate disease progression in patients with idiopathic PD. In postmortem examinations of the brain, researchers have observed activation of microglial cells and the complement system, infiltration of T-lymphocytes, higher levels of pro-inflammatory cytokines (e.g., IL-1β, IL-6, TNF-α, IFN-γ), and increased concentrations of nitric oxide synthase (iNOS) and reactive oxygen species (ROS) in the SNpc and striatum of PD patients compared to those who are healthy⁸. Furthermore, previous studies have demonstrated that PD patients exhibit elevated levels of pro-inflammatory cytokines, including IL-1β, IL-6, IL-2, TNF-α, and IFN-γ, in their plasma and serum⁹. Concurrently, increased macrophage migration inhibitory factor (MIF) levels have been observed in these patients¹⁰. Consistent results were obtained from cerebrospinal fluid (CSF) analyses, further corroborating the role of neuroinflammation in the onset and progression of PD and emphasizing its critical contribution to disease pathogenesis¹¹.

Microglia are mononuclear macrophages resident in the central nervous system, which are necessary for the normal development of the nervous system. In the early stage of brain development, extracellular signals such as transmitters and cytokines secreted by microglia can combine with specific receptors on neurons, improve the survival rate of neurons, save injured dopaminergic neurons¹² or promote tissue regeneration and repair, and remove cell debris and toxic substances through phagocytosis¹³, so as to regulate the number of neurons in the central nervous system. In neurodegenerative disorders like PD, the protective role of microglia is neutralized or even masked by pro-inflammatory effects, the balance is disturbed, microglia activate with reactive hyperplasia¹⁴, releasing pro-inflammatory cytokines (TNF-α, IL-1β, and IL-6) as well as several chemokines¹¹, such as superoxide and nitric oxide (NO), which produce neurotoxic effects promoting dopaminergic neuronal death¹⁵. It was found that disease-specific upregulation of microglia and microglia in the substantia nigra in a reactive state were observed in postmortem midbrain tissue of patients with idiopathic PD¹⁶. Upregulation of inflammatory factors and chemokines was found in the SNpc of postmortem PD patients, and microglia proliferation was also observed in areas where neuronal death had not yet been detected¹⁷, suggesting that microglial activation most likely occurs in the early stages of the development of PD and contributes to some degree to the death of dopaminergic neurons in the midbrain, exacerbating the progression of PD. He Y. and Wu D.C. et al. demonstrated that minocycline exerts neuroprotective effects in a PD mouse model by inhibiting microglia activation^18,19. Thus, targeting microglia-mediated neuroinflammation has gained increasing recognition.

Macrophage migration inhibitory factor (MIF), a type of cytokine that possesses various biological functions, serves as a crucial regulator in the immune system of the human body. MIF, a secreted protein with a molecular weight of 12.5 kDa, is extensively preserved and plays a role in numerous biological processes²⁰. It is widely expressed in various cells and tissues^21,22, with a particularly high expression in both the central and peripheral nervous systems²³. MIF, a neuroendocrine mediator, can induce pro-inflammatory effects by modulating signal transduction to enhance the release of several pro-inflammatory molecules^24,25, including TNF-α, IL-1β, and IL-6. Previous studies demonstrated that MIF upregulates the expression of the ERK1/2 pathway through CD74/CD44, activates downstream responses, promotes the upregulation of PGE2, and thus induces the expression of COX2, which plays a significant role in the inflammatory response^26,27. MIF can also positively regulate Toll-like receptor-4 (TLR4) and promote the body’s response to endotoxin-producing bacteria^28,29. In addition, the MKP-1/P38/AP-1 pathway can be regulated by targeted inhibition of intracellular MIF to play an anti-inflammatory role³⁰, which indicates that MIF can upregulate the inflammatory pathway and promote inflammation. Several clinical and preclinical studies have reported the role of MIF in PD and shown increased serum levels of MIF in PD patients compared to healthy controls, suggesting that serum MIF levels may be a potential diagnostic biomarker for PD^10,31.

In this study, we revealed the pathways involved in the pathological effects of MIF in PD and its mechanism. We investigated whether MIF could be a target for PD therapy, that is, whether downregulation of MIF could play a neuroprotective role by attenuating neuroinflammation through regulation of the Toll-like receptor pathway, thereby alleviating or ameliorating the course of PD.

Results

MIF expression and pro-inflammatory effects of MIF in the PD model

Neuroinflammation is known to be a contributing factor in PD. In vivo experiments have shown that microglia can adopt a pro-inflammatory phenotype associated with the release of cytotoxic molecules, resulting in increased neurodegeneration. MIF can induce the production and secretion of pro-inflammatory factors (IL-1β, TNF-α, and IFN-γ) and other inflammatory mediators (such as NO) and activate corresponding pro-inflammatory signaling pathways. Elevated MIF levels in PD patients may promote disease progression by amplifying neuroinflammatory responses¹⁰. However, the underlying mechanism of MIF-mediated neuroinflammation in PD is still unclear. To explore the expression of MIF in PD models, we detected the expression of MIF by Western blotting and RT-PCR, and we found that MIF and its mRNA expression were increased in the MPP⁺-induced PD cell model compared with controls (Fig. 1a–c). In addition, MIF and its mRNA were similarly increased in 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-induced PD mouse model compared with the control group (Fig. 1d–f).

Fig. 1. MIF expression and pro-inflammatory effects of MIF in the PD model.

MN9D cells were cultured and divided into a control group and MPP⁺ groups with different concentrations. The cells were treated with medium and MPP⁺ (100, 250, 500, 1000 μM) for 24 h and then collected. We used MPTP to construct a subacute PD mouse model. The experiment was divided into two groups: control group and MPTP group. a, d The expression level of MIF in cell lysates and mouse midbrain lysates was detected by Western blotting, and the band density was quantified. b, e Quantitative statistical analyses of band density in (a) and (d), respectively. c, f MIF mRNA levels in cells and mouse midbrain were detected by RT-qPCR. g The content of Iba1 was detected by immunofluorescence in tissue sections. The nuclei were labeled with DAPI, and microglia were labeled with anti-IBa1 antibody. h Iba1-positive cells were counted. i–k ELISA was used to detect the expression of serum pro-inflammatory cytokines (IL-6, TNF-α, and IL-1β). All cell experiments were repeated three times independently, n = 3 per group in animal experiments. Scale bar, 50 μm or 20 μm. All data are presented as mean ± SEM. Statistical analysis was performed using one-way ANOVA (NS not statistically significant, *p < 0.05, **p < 0.01).

The development of PD and other neurodegenerative diseases may be influenced by neuroinflammation^6,7. In PD, microglia are activated and accompanied by reactive proliferation, with Iba1 serving as a specific microglial marker. Therefore, we examined the immunoreactivity of Iba1 in the mouse midbrain with immunofluorescence of tissue sections, revealing an elevation in the quantity of Iba1-positive cells in the MPTP-induced PD mouse model when compared to the control group (Fig. 1g, h). Activated microglia in PD release pro-inflammatory cytokines (IL-6, TNF-α, and IL-1β) as well as various chemokines, such as superoxide and nitric oxide (NO), which have neurotoxic effects and promote dopaminergic neuron death^32,33. Next, we observed neuroinflammation in the PD model by measuring the expression of inflammatory markers in the mouse midbrain. Correspondingly, we found that all tested pro-inflammatory markers (IL-6, TNF-α, and IL-1β) were significantly increased in comparison with the control group (Fig. 1i–k).

Next, we used the mouse microglial cell line BV2 cells to further investigate the role of MIF in PD. BV2 cells were treated with different concentrations of MPP^+ 34, and the cell viability was detected by CCK-8 assay to determine the optimal concentration of MPP⁺ (Supplementary Fig. 1a). Western blotting revealed elevated MIF expression in MPP⁺-exposed BV2 cells compared to untreated controls(Supplementary Fig. 1b, c). Given that MIF can function intracellularly and extracellularly through autocrine or paracrine forms. We found that the content of MIF increased in the medium supernatant of BV2 cells exposed to MPP⁺, while MIF-siRNA decreased the content of MIF (Supplementary Fig. 1d). It is known that LPS can induce the activation of microglia and increase the secretion of inflammatory factors to mediate inflammatory damage³⁵. Therefore, we detected the mRNA levels of various pro-inflammatory factors and anti-inflammatory factors after MPP⁺ exposure by RT-qPCR. We found that the mRNA levels of pro-inflammatory factors (IL-6, TNF-α, and IL-1β) were upregulated, and the mRNA levels of anti-inflammatory factors (IL-4 and IL-10) were downregulated in BV2 cells exposed to MPP⁺ (Supplementary Fig. 1e–j).

Elevated MIF can lead to a series of pathological effects and activate related signaling pathways²⁰. TLR4, a crucial molecule involved in inflammatory signal transduction, has been documented to have a significant impact on PD. Additionally, MIF has been found to exert a positive regulatory effect on TLR4, as reported in studies^28,29. Therefore, we examined the expression of Toll-like receptor pathway-related proteins in BV2 cells exposed to MPP⁺ by Western blotting (Supplementary Fig. 1k–o). The results showed that the TLR4/MyD88/TRAF6 /NF-κB pathway was activated in the MPP⁺ group compared with the control group. We found that the TLR4/MyD88/TRAF6/NF-κB pathway was activated in the MPP⁺ group compared with the control group. In general, these results suggest an elevation in MIF expression in the PD model, and the PD model exhibits inflammatory effects caused by MIF. Simultaneously, MIF has the ability to trigger inflammatory pathways, resulting in pro-inflammatory impacts.

Downregulation or MIF inhibition exerts protective effect on the PD cell model

ISO-1 is a relatively common specific MIF inhibitor that selectively binds to the tautomerase active site of MIF and inhibits the enzyme activity, thereby inhibiting some of the biological functions of MIF^36,37. Current evidence indicates a significant positive correlation between the specific inhibition of MIF tautomerase activity by ISO-1 and the regulation of MIF pro-inflammatory activity, further elucidating the molecular mechanisms underlying the cytokine’s pro-inflammatory effects via targeting this catalytic site and highlighting its therapeutic potential³⁸. Multiple studies have established a functional link between MIF bioactivity and this active site^37,39. To verify whether downregulation or MIF inhibition had a protective effect on PD, we used MIF inhibitor ISO-1 pretreatment to inhibit MIF or transfected MIF-siRNA knockdown MIF in the PD cell model to detect cell viability. The knockdown effect of MIF-siRNA was verified by Western blotting, which examined the expression level of MIF after transfection with MIF-siRNA (Fig. 2a, b). We found that ISO-1 or MIF-siRNA had a significant protective effect on cell viability in the PD cell model and effectively suppressed the harmful influence of MPP⁺ on MN9D cells (Fig. 2c, d). Tyrosine hydroxylase (TH) is the rate-limiting enzyme of dopamine synthesis, which is mainly expressed in dopaminergic neurons. The decrease in TH content in neuronal cells indicates that the function of synthesizing and secreting dopamine is impaired, suggesting that PD lesions occur in cells. Therefore, the expression level of TH after treatment was detected by Western blotting. The finding indicated that the expression of TH was markedly decreased in the PD cell model, while the expression of TH was markedly increased after pretreatment with MIF inhibitor ISO-1 for 1 h or transfection with MIF-siRNA (Fig. 2e–h).

Fig. 2. Down-regulation or MIF inhibition has a protective effect on the PD cell model.

a MN9D cells were seeded into six-well plates and transfected with MIF-siRNA. The cells were divided into three groups: the NC group, siRNA1 group, and siRNA2 group. After 48–72 h transfection, the expression of MIF in cell lysates was detected by Western blotting. b Quantitative statistical analysis of the band density of (a). MN9D cells were seeded into 96-well plates, and three multiple wells were set in each group. c, d Cell viability was detected by CCK-8 assay. e, g Western blotting to detect TH expression in cell lysates. f, h Quantitative statistical analysis of band density in (e, g), respectively. All experiments were repeated three times independently. All data are presented as mean ± SEM. Statistical analysis was performed using one-way ANOVA (NS not statistically significant, *p < 0.05, **p < 0.01, ***p < 0.001).

Downregulation of MIF ameliorates behavioral deficits and pathological changes in the PD mouse model

To explore the therapeutic effect of downregulation of MIF on PD, we first screened shRNA with the best knockdown efficiency by unilateral stereotactic (ST) injection of adeno-associated virus (Supplementary Fig. 2a–c). Next, C57BL/6 mice were stereotactically injected with MIF shRNA or scrambled shRNA in the midbrain and infected for 4 weeks. Then, we constructed subacute PD mouse model with intraperitoneal injection of MPTP (30 mg/kg) once daily for 5 days (Fig. 3b). The colocalization of adeno-associated virus GFP and microglia marker Iba1 was detected by immunofluorescence, and we could observe the colocalization of GFP and Iba1, indicating that adeno-associated virus could infect microglia (Supplementary Fig. 2d).

Fig. 3. Downregulation of MIF ameliorates MPTP-induced behavioral deficits and pathological changes.

C57BL/6 mice were injected with MIF-shRNA or scrambled shRNA in the substantia nigra pars compacta (SNpc) and then received one i.p. injection of MPTP-HCl or saline daily for 5 days. a Schematic presentation of the experimental paradigm, created with BioGDP.com⁸¹. b Open-field test, c open-field test—total distance of mouse movement (mm), d open-field test—middle distance (mm), e open-field test—mean velocity (mm/s), f open-field test—central mean velocity(mm/s), g pole test, h hanging test, and i rotarod test. j The TH expression in the mouse midbrain was detected by Western blotting. k Quantitative statistical analysis of TH band density. l TH content was detected by immunofluorescence in tissue sections. Nuclei were labeled with DAPI, and dopaminergic neurons were labeled with anti-TH antibody. m TH-positive cells (dopaminergic neurons) were counted. n = 10 mice per group for the behavioral tests in (a–i). n = 3 per group in other animal experiments. Scale bar, 100 μm or 20 μm. All data are presented as mean ± SEM. Statistical analysis was performed using one-way ANOVA (NS: not statistically significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.001).

Following the modeling, various behavioral tests were performed to assess the motor function of mice, such as the pole test, hanging test, rotarod test, and open-field test⁴⁰. We found that the MPTP + AAV-MIF-NC group had no improvement effect in the pole test, open-field test, hanging test, and rotarod test compared with the MPTP-treated group (p > 0.05) (Fig. 3a, c–h), while the MPTP + AAV-MIF-shRNA group had significantly increased total distance of open-field movement, middle distance, mean velocity, and central mean velocity (Fig. 3b–f), a shortened pole climbing time (Fig. 3g), and a prolonged time of rotarod (Fig. 3i) compared with the MPTP-treated group. All of these results suggested that the motor ability and motor coordination of mice were significantly improved.

We employed western blotting to detect the effect of MIF knockdown on TH-immunopositive neurons in the midbrain. The TH immunoreactivity in SNpc of PD mice was significantly decreased, while the TH immunoreactivity in SNpc of PD mice injected with MIF shRNA was significantly increased (Fig. 3j, k). Next, we detected TH-positive cells in the SNpc via immunofluorescence. We found that the MIF-shRNA knockdown group significantly reversed MPTP-induced loss of TH-positive cells in the SNpc compared with vehicle treatment in MPTP-treated mice (Fig. 3l, m). In addition, we employed the immunohistochemical method to detect whether MIF knockdown can improve the pathological alterations of PD mice. Consistent with our predictions, MIF knockdown reversed the reduction in the number of TH-positive neurons in the SNpc (Supplementary Fig. 3a, c) and the TH-positive neuron density in the striatum of PD mice, compared with MPTP and scrambled RNA-treated mice (Supplementary Fig. 3b, d). Taken together, these findings suggested that downregulation of MIF had a notable alleviating effect on motor deficits in PD mice and that MIF knockdown had a protective effect on the nervous system of PD mice.

Effects of MIF knockdown on microglia activation and pro-inflammatory mediators in the PD model

To determine whether the protective effect of MIF knockdown on dopaminergic neurons is related to its anti-neuroinflammatory response, we studied the effect of MIF on MPTP-induced Iba1 expression in microglia of PD mice. We detected the expression of Iba1, a microglia-specific marker in the mouse midbrain, by Western blotting (Fig. 4a, b). We also measured the immunoreactivity of Iba1 in the mouse midbrain with immunofluorescence. The results showed a marked increase in the number of Iba1-positive cells in PD mice and a marked decrease in the number of Iba1-positive cells in MIF knockdown-treated mice (Fig. 4c, d). Activated microglia secrete pro-inflammatory mediators such as TNF-α, IL-1β, and IL-6, and their expression is increased in a variety of acute and chronic inflammatory diseases. We next determined the effects of MIF knockdown on pro-inflammatory mediators by measuring the expression of inflammatory indicators in the mouse midbrain. The expression of pro-inflammatory mediators (IL-6, TNF-α, IL-1β) in BV2 cells medium supernatant and mouse serum was detected (Fig. 4e–j). In conclusion, these results demonstrated that MIF knockdown could attenuate the activation of microglia in the PD model and reduce the production of pro-inflammatory mediators.

Fig. 4. Effects of MIF knockdown on microglia activation and neuroinflammation in PD mice.

a The expression of Iba1 in mouse midbrain lysates was detected by Western blotting. b Quantitative statistical analysis of Iba1 band density. c Iba1-positive cells were analyzed with immunofluorescence and imaged with a fluorescence microscope. The nuclei were labeled with DAPI, and microglia were labeled with anti-Iba1 antibody. d Iba1-positive cells were counted. e–g The expressions of inflammatory factors (IL-6, TNF-α, IL-1β) in BV2 cells were detected by ELISA, and h–j the expressions of inflammatory factors (IL-6, TNF-α, IL-1β) in the serum of mice were detected by ELISA. All cell experiments were repeated three times independently, n = 3 per group in animal experiments. Scale bar, 50 μm or 20 μm. All data are presented as mean ± SEM. Statistical analysis was performed using one-way ANOVA (NS not statistically significant, *p < 0.05, **p < 0.01, ***p < 0.001).

Effects of MIF on neuroinflammation

The pathologic development of PD is strongly associated with neuroinflammation, including chronic microglia proliferation⁴¹. Activated microglia promote the release of inflammatory factors, inducible nitric oxide synthase (iNOS) and cyclooxygenase-2 (COX2) are involved in the production of no and prostaglandin E2 (PGE2) in the process of inflammation, and further promote the development of neuroinflammation^42,43. Therefore, we used MIF inhibitors to explore the changes in pro-inflammatory factors and anti-inflammatory factors. The mRNA expression of pro-inflammatory factors and anti-inflammatory factors in BV2 cells treated with MIF inhibitor ISO-1 was detected by RT-qPCR. In our investigation, it was discovered that MIF inhibitor ISO-1 could reduce the expression of pro-inflammatory factors (IL-6, TNF-α, IL-1β, iNOS, COX2) and increase the expression of anti-inflammatory factors (IL-4, IL-10) (Supplementary Fig. 4a–g) in BV2 cells exposed to MPP⁺.

We found that the use of the MIF inhibitor ISO-1 reduced the production of pro-inflammatory factors and increased the production of anti-inflammatory factors, so we wondered whether transfection of siRNA to knock down MIF would also have consistent results. Therefore, RT-qPCR was used to detect the effect of transfection with MIF-siRNA. Expression of pro-inflammatory factors (IL-6, TNF-α, IL-1β, iNOS, COX2) and anti-inflammatory factors (IL-4, IL-10) mRNA were detected in BV2 cells exposed to MPP⁺ (Fig. 5a–g). Meanwhile, we examined the expression of pro-inflammatory mediators iNOS and COX2 (The lower one of the two stripes) by Western blotting. (Fig. 5h–j). At the same time, the mean fluorescence intensity of iNOS was detected by cellular immunofluorescence (Fig. 5k). These results indicated that downregulation or inhibition of MIF could alleviate neuroinflammation by decreasing the production of pro-inflammatory factors and increasing the production of anti-inflammatory factors.

Fig. 5. Effects of MIF on neuroinflammation.

Mouse microglial cell line BV2 cells were seeded into six-well plates. a–g The mRNA levels of pro-inflammatory factors (IL-6, TNF-α, IL-1β, iNOS, COX2) and anti-inflammatory factors (IL-4, IL-10) in BV2 cells were detected by RT-qPCR. h The expressions of iNOS and COX2 in cell lysates were detected by Western blotting. i, j Quantitative statistical analysis of band density of iNOS and COX2, respectively. l The fluorescence intensity of iNOS was detected by the immunofluorescence method. iNOS was labeled with green fluorescence, and the nucleus was displayed by blue fluorescence DAPI. k The content of iNOS was detected by immunofluorescence. All experiments were repeated three times independently. Scale bar, 20 μm or 10 μm. All data are presented as mean ± SEM. Statistical analysis was performed using one-way ANOVA (NS not statistically significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Knockdown of MIF reduces the production of pro-inflammatory mediators and attenuates neuroinflammation in the PD mouse model

Previous in vitro results suggested that downregulation of MIF might exert an anti-inflammatory effect by regulating the expression of pro-inflammatory factors and anti-inflammatory factors. To test this hypothesis in vivo, we first detected the expression of the pro-inflammatory mediators iNOS (The lower one of the two stripes) and COX2 by Western blotting. We found that MIF-shRNA knockdown significantly reduced pro-inflammatory mediators expression (Fig. 6a–c). In addition, we performed immunofluorescence double staining for the microglial marker Iba1 and the pro-inflammatory mediator iNOS (Fig. 6d), and, as expected, the number of Iba1⁺/iNOS⁺ double-labeled microglia cells increased after MPTP induction. In contrast, the number of double-labeled microglia cells significantly decreased after MIF knockdown (Fig. 6e). The majority of studies indicate that activated microglia exert neurotoxic effects on the central nervous system by secreting pro-inflammatory factors such as TNF-α, thereby exacerbating neuroinflammation. Consequently, we performed immunofluorescence co-localization analysis of the microglial marker Iba1 and the pro-inflammatory factor TNF-α (Fig. 6f). Our results demonstrated increased TNF-α expression in the MPTP group compared to the control group, with TNF-α co-localizing with activated microglia. Quantification revealed significantly elevated numbers of Iba1⁺/TNF-α⁺ double-positive cells in PD mice, suggesting TNF-α primarily originates from activated microglia. This indicates that activated microglia promote inflammatory progression via pro-inflammatory factors release. Notably, MIF knockdown markedly reduced Iba1 and TNF-α expression (Fig. 6g). Collectively, these findings indicated that MIF knockdown attenuates neuroinflammation by suppressing microglial activation and subsequent inflammatory factors release in PD mice.

Fig. 6. MIF knockdown attenuates neuroinflammation in the midbrain of PD mice.

C57BL/6 mice were injected with MIF-shRNA or scramble shRNA in the substantia nigra pars compacta (SNpc), followed by MPTP-HCL or saline once i.p. Injections were given for 5 days. a The expressions of iNOS and COX2 in the midbrain of mice were detected by Western blotting. b, c Quantitative statistical analysis of the band density of iNOS and COX2, respectively. d Microglia in the midbrain of mice were stained by immunofluorescence with microglia marker Iba1 and pro-inflammatory mediator iNOS, and representative images are shown. e Iba1⁺/iNOS⁺ double-labeled microglia cells were counted. f Fluorescence co-localization imaging of the microglial marker Iba1 and the pro-inflammatory factor TNF-α. g Iba1⁺/ TNF-α⁺ double-labeled microglia cells were counted. n = 3 per group in animal experiments. Scale bar, 50 μm or 20 μm. All data are presented as mean ± SEM. Statistical analysis was performed using one-way ANOVA (NS not statistically significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Downregulation of MIF alleviates neuroinflammation by inhibiting the TLR4/MyD88/TRAF6/NF-κB pathway

TLR4 is an important inflammatory signal transduction molecule that has been reported to play an important role in PD. MIF plays a positive regulatory role in TLR4^28,29, which is a key receptor regulating neuroinflammation. Moreover, TLR4 can trigger the MyD88-dependent pathway to induce an inflammatory response, namely the TLR4/MyD88/TRAF6/NF-κB pathway. MIF secreted by BV2 cells can function by activating relevant signaling pathways, and we verified that MIF-siRNA could reduce MIF secretion in our previous findings. Therefore, we examined the expression of Toll-like receptor pathway-related proteins and pro-inflammatory mediators (iNOS, TNF-α) in BV2 cells exposed to MPP⁺ by Western blotting (Fig. 7a–h). We found that MIF-siRNA reduced the expression of MPP⁺-induced pathway proteins and pro-inflammatory factors (iNOS, TNF-α). Based on the fact that the nuclear translocation of p65 is a key event in NF-κB-mediated inflammation, we detected the nuclear translocation of NF-κBp65 by cellular immunofluorescence assay and discovered that the nuclear expression of NF-κB was markedly increased in the MPP⁺ group and significantly decreased in the MPP⁺ + MIF-siRNA group (Fig. 7i, j). These findings further demonstrated that downregulation of MIF exerted anti-inflammatory effects by reducing the nuclear expression level of NF-κB. These results, in turn, suggested that downregulation of MIF attenuates neuroinflammatory damage by inhibiting the TLR4/MyD88/TRAF6/NF-κB pathway.

Fig. 7. MIF downregulation alleviates neuroinflammation by inhibiting TLR4/MyD88/TRAF6/NF-κB pathway.

Mouse microglial cell line BV2 cells were seeded into six-well plates. a The expressions of TLR4, MyD88, TRAF6, p-p65, p65, p-IκBα, IκBα, TNF-α, and iNOS in cell lysates were detected by Western blotting. b–h Quantitative statistical analysis of the band density of TLR4, MyD88, TRAF6, p-p65, p65, p-IκBα, IκBα, TNF-α, and iNOS, respectively. i Immunofluorescent staining and quantification of NF-κB p65. j The expression level of NF-κB in the nucleus was detected by the immunofluorescence method. NF-κB was labeled with red fluorescence, and the nucleus was displayed by blue fluorescence DAPI. All experiments were repeated three times independently. All experiments were repeated three times independently. Scale bar, 20 μm or 10 μm. All data are presented as mean ± SEM. Statistical analysis was performed using one-way ANOVA (NS not statistically significant, *p < 0.05, **p < 0.01, ***p < 0.001).

Next, we used the TLR4 signaling inhibitor TAK-242 to further verify the expression of TLR4 pathway proteins in BV2 cells transfected with MIF-siRNA with or without TAK-242 (1 μM) by Western blotting. We found that the expression of pathway proteins increased in MPP⁺ and MPP⁺ + NC groups compared with the control group. Western blotting results showed that the expression of pathway proteins was reduced in the MPP⁺+siRNA group compared with the MPP⁺ group, and the expression of pathway-related proteins was further decreased in the TAK-242 (1 μM) inhibitor-treated group compared with the MPP⁺+siRNA group (MPP⁺+siRNA+TAK-242 group) (Supplementary Fig. 5a–e). These findings suggested that the TLR4 inhibitor TAK-242 could further attenuate the activation effect of MIF on the TLR4/MyD88/TRAF6/NF-κB pathway, synergistically enhancing the anti-neuroinflammatory effects of MIF knockdown.

Our previous findings suggested that downregulation of MIF attenuated inflammation, so whether downregulation of MIF reduced dopaminergic neuronal injury and death was the next question we explored. We collected transfected and modeled microglia cell line BV2 cells in conditioned medium and cultured MN9D cells for 24 h. The experimental grouping was consistent with the BV2 cells grouping. We detected whether the viability of MN9D cells changed after treatment with different groups of conditioned media by CCK-8 assay, followed by Western blotting to detect the expression of anti-apoptotic (Bcl-2) and pro-apoptotic (Bax, caspase-8) proteins (Supplementary Fig. 6a–e). Meanwhile, we detected the average fluorescence intensity of the pro-apoptotic protein (caspase-3) by cellular immunofluorescence (Supplementary Fig. 6f, g). These results indicated that MIF-siRNA can protect MN9D cells from activated microglia-induced cytotoxicity and apoptosis.

MIF knockdown inhibits TLR4/MyD88/TRAF6/NF-κB pathway activation, thereby alleviating neuroinflammation

Subsequently, in order to verify the anti-inflammatory effect of MIF in PD mice, we examined the expression of brain TLR4 pathway proteins by Western blotting in PD mice. We found that the expression of TLR4 pathway proteins increased, and the pathway was activated in the midbrain of PD mice. However, when PD mice were injected with MIF-shRNA, the expression of the abovementioned indexes was restored to control levels (Fig. 8a–f). Taken together, these results demonstrated that MIF knockdown suppresses TLR4/MyD88/TRAF6/NF-κB pathway activation, thereby alleviating neuroinflammation.

Fig. 8. MIF knockdown inhibits its activation of the TLR4/MyD88/TRAF6/NF-κB pathway and alleviates neuroinflammation.

a The expressions of TLR4, MyD88, TRAF6, p-p65, p65, p-IκBαand IκBαin the midbrain of mice were detected by Western blotting. b–f Quantitative statistical analysis of the band density of TLR4, MyD88, TRAF6, p-p65/p65 and p-iκBα/IκBα, respectively. n = 3 per group in animal experiments. g Using specific inhibitor ISO-1 or MIF-siRNA, downregulation of MIF inhibits TLR4 receptor activation, thereby inhibiting the MyD88-dependent pathway-mediated pathway downstream of TLR4 and reducing nuclear translocation of the transcription factor NF-κBp65, thereby reducing pro-inflammatory factors (IL-1β, IL-6, TNF-α) transcriptional upregulation to alleviate neuroinflammation and ultimately protect dopaminergic nerves from microglia activation-induced neurological damage and degeneration. All data are presented as mean ± SEM. Statistical analysis was performed using one-way ANOVA (NS: not statistically significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Discussion

The pathogenesis of PD is complex and diverse. In the past few years, the molecular mechanisms related to PD neuroinflammation have attracted more and more attention. Moreover, the pathogenesis of PD is no longer considered to be single and limited; On the contrary, the interaction between various pathogenesis jointly leads to the occurrence and development of PD⁴⁴.

MIF is a widely expressed cytokine with multiple biological functions and exerts different effects by mediating different signaling pathways. It was found that the neuronal survival rate was enhanced and pathological effects were weakened in MIF-KO mice, which promotes the recovery of SCI motor function⁴⁵. MIF can regulate the inflammatory microenvironment by regulating MAPK and other inflammatory signaling pathways to induce the release of PGE₂⁴⁶. MIF can also inhibit p53-mediated apoptosis of tumor cells and promote tumor angiogenesis, which is closely associated with tumor invasion and metastasis⁴⁷. In addition, a recent study found that MIF can mediate cell death through unique nuclease activity that cleaves DNA and thus participates in the Parthanatos pathway, a novel death pathway dependent on poly-ADP-ribose polymerase-1 (PARP-1)⁴⁸. Therefore, inhibition of MIF nuclease activity or nuclear translocation of MIF attenuates DNA damage and cell degeneration⁴⁹, indicating that inhibition of PAAN/MIF nuclease activity prevents neurodegeneration of PD⁵⁰.

Matsunaga et al. found that MIF may participate in the detoxification of catecholamine products, thus participating in the protection of neurons⁵¹. By contrast, Cheng et al. demonstrated that inhibiting NLRP3 activation could rescue Atg5 deficiency-induced MIF enhancement and microglial activation, ultimately reducing the loss of neurons in vivo³¹. These findings indicated that MIF plays a somewhat contradictory role in neurological diseases, which may be beneficial or detrimental; therefore, we need to continue to explore it in depth. Increased levels of MIF expression in the plasma of patients with spinal cord injury, accompanied by the release of inflammatory factors, indicate that MIF is closely associated with the neuroinflammatory pathological effects of SCI⁵²; MIF also activates inflammatory pathways to increase the release of inflammatory factors and other neurotoxic factors, thereby inducing neuronal death^53,54. It has been proven that MIF concentrations are significantly increased in the cerebrospinal fluid of patients with Alzheimer’s disease⁵⁵ and that multiple biological effects of MIF are involved in the development and progression of Alzheimer’s disease, e.g., pro-inflammatory effects, promotion of toxic effects of Aβ aggregates^56,57, pro-astrocyte activation, and tau protein hyperphosphorylation⁵⁸.

Research has indicated that MIF can be expressed in different stages of various diseases and can regulate the pathological process of the disease and promote its development^59,60. MIF is a cytokine closely associated with neuroinflammation, which stimulates the release of specific pro-inflammatory factors (IL-6, TNF-α, IL-1β, IFN-γ), thereby inducing a pro-inflammatory response⁶¹. Higher MIF levels in some PD patients may promote the progression of neuroinflammation in PD by amplifying the neuroinflammatory response¹⁰. Hence, in this investigation, we assessed the MIF expression in the PD model. We observed that MIF expression was increased in the PD cell model and PD mouse model, while exerting pro-inflammatory effects. In order to investigate the pathological effects and potential mechanisms of MIF involvement in PD pathogenesis, we used two cell types, MN9D cells and BV2 cells, to construct a PD cell model and also constructed an MPTP mouse PD model for a series of experiments. Previous literature reported increased MIF concentration in the serum of PD patients detected by ELISA¹⁰, and our results also demonstrated increased MIF expression in the PD model. MIF is closely related to neuroinflammation and promotes disease progression. Consistently, our study also found that MIF exacerbates inflammatory pathological effects in the PD model, as well as activates the TLR4/MyD88/TRAF6/NF-κB inflammatory signaling pathway. Therefore, we subsequently used MIF inhibitors to inhibit the tautomerase activity of MIF or siRNA to knock down MIF in subsequent experiments to further discuss the pathological effects and potential mechanisms of MIF involvement in PD pathogenesis.

Previous research has indicated that in the MIF-deficient mouse model of atherosclerosis, there was a reduction in the degree of inflammation and plaque formation. Similarly, overall MIF deficiency facilitates atherosclerosis-protective autoantibody profiles, thereby reducing the inflammatory response and attenuating lesion formation⁶². The use of different MIF inhibitors can attenuate the inflammatory response caused by LPS in BV2 cells and C57BL/6 mice, exerting neuroprotective effects⁶³. In the previously obtained results, the inflammatory response was enhanced in BV2 cells exposed to MPP⁺, and we measured the levels of pro-inflammatory factors in culture medium supernatants and mouse serum using ELISA after transfection with siRNA and in mice injected with adeno-associated virus to knock down MIF, concluding that downregulation of MIF reduces the expression of pro-inflammatory factors in PD. Consequently, we demonstrated the protective effect of knocking down MIF via ISO-1 (a MIF inhibitor) and MIF-siRNA transfection in both PD mice and the cell model. In PD mice, knockdown of MIF ameliorated not only behavioral deficits but also pathological changes in MPTP-treated mice. Therefore, the protective effect of knockdown of MIF on neuronal death in PD may be related to its anti-inflammatory effect.

Our current study found that knockdown of MIF improved motor ability and motor coordination, upregulated TH expression, reduced microglial marker Iba1 expression, and prevented the neurodegeneration of PD mice. Microglia exert both pro-inflammatory and anti-inflammatory effects when stimulated by certain factors. As the disease progresses, the homeostatic balance of microglial anti-inflammatory function is disrupted, and the anti-inflammatory effect is much lower than the pro-inflammatory effect, which mainly focuses on the secretion of pro-inflammatory factors to mediate inflammatory injury^64,65. Meanwhile, the interaction between neuroglial cells can further promote the development of PD⁶⁶; studies have shown that activated microglia can induce neurotoxic reactive A1 astrocytes to promote neurodegeneration⁶⁷, and inhibition of microglia activation can reduce the formation of reactive A1 astrocytes, thereby attenuating neurological damage in PD^68,69. It was previously found that MIF-mediated inflammatory responses were attenuated by targeted inhibition of the MIF receptor CD74⁷⁰. Furthermore, in the advanced stages of morbid obesity, increased MIF secretion by M1 macrophages may further exacerbate the positive feedback loop of adipocyte inflammation and macrophage polarization, leading to increased release of pro-inflammatory factors and thus exacerbating the inflammatory pathological effects⁷¹. Therefore, in our study, we verified that downregulation or inhibition of MIF could attenuate the inflammatory response and subsequently reduce the damage to dopaminergic neurons.

Toll-like receptor pathway-mediated inflammatory responses are closely related to PD, and it has been reported that MIF positively regulates TLR4^28,29, a key receptor regulating neuroinflammation. MyD88 is an important downstream signaling adaptor of the TLR4 receptor, which can activate and regulate NF-κB signaling and promote the release of inflammatory factors. Research indicates that lipopolysaccharide (LPS) activates the TLR4 receptor via the MYD88-dependent pathway, subsequently triggering downstream effector molecules, this process is known to exacerbate neuroinflammation^72,73. It was noted that, in BV2 cells, MPP⁺ treatment activates the TLR4/NF-κB/cytokine signaling pathway, which provides new directions and insights into TLR4 signaling after MPP⁺-induced BV2 cell inflammation. Moreover, our study found that MIF promotes inflammatory pathological effects in the PD model and activates the TLR4/MyD88/TRAF6/NF-κB inflammatory signaling pathway. Therefore, we chose the TLR4/MyD88/TRAF6/NF-κB pathway to explore next, and our results showed that either cell transfection with MIF-siRNA or stereotaxic injection of adeno-associated virus AAV-MIF-shRNA could downregulate the expression of pathway-related proteins and reduce the amount of pro-inflammatory mediator iNOS and microglia marker Iba1 co-staining, thereby alleviating neuroinflammatory damage. To further verify this, we used the TLR4 inhibitor resatorvid (TAK-242) in our subsequent experiments, which is known to be a selective inhibitor of TLR4 signaling^74–76. Resatorvid downregulates the expression of MyD88 and TRAF6, which are downstream signaling molecules of TLR4, and it plays a crucial role in inflammatory conditions⁷⁶. We found that MIF activated the TLR4/MyD88/TRAF6/NF-κB pathway, whereas the TLR4 inhibitor TAK-242 attenuated the activation of the TLR4/MyD88/TRAF6/NF-κB pathway by MIF. This further suggests that MIF can promote neuroinflammation and ultimately lead to neurodegeneration through the TLR4/MyD88/TRAF6/NF-κB pathway. The interaction between neurons and glial cells is an important factor leading to nerve cell injury and death. It has been shown that conditioned medium for BV2 cells treated with anti-inflammatory therapy can attenuate dopaminergic neuronal death by reducing the release of inflammatory mediators⁷⁷. Additionally, Phenolic compounds from Moraiolo virgin olive oil (MVOO) can attenuate inflammatory responses through inflammatory signaling pathways, and the use of treated conditioned medium can protect SHSY-5Y neuroblastoma cell lines from microglia activation-induced cytotoxicity⁷⁸. Consistent with these findings, our results demonstrated that MIF-siRNA protects MN9D cells from activated microglia activation-induced cytotoxicity and apoptosis.

In summary, the expression of MIF is increased in the PD cell model and PD mouse model, mediating inflammatory pathological effects as well as activating the TLR4/MyD88/TRAF6/NF-κB pathway. The cell viability of the PD cell model increased, and the expression of TH was restored after MIF inhibition with ISO-1 and MIF-siRNA knockdown. Meanwhile, stereotactic injection of AAV-MIF-shRNA significantly improved the motor behavior disorder of PD mice, and MIF knockdown also upregulated the expression of TH in the midbrain of PD mice. Downregulation of MIF could also reduce the release of pro-inflammatory factors and attenuate the inflammatory response. Based on our in vivo and in vitro results, we suggest that down-regulation of MIF inhibits the TLR4/MyD88/TRAF6/NF-κB pathway, reduces the pro-inflammatory phenotype of BV2 cells (in vitro) and mouse midbrain microglia (in vivo), and regulates the release of anti-inflammatory and pro-inflammatory factors, thereby attenuating neuroinflammatory injury. Furthermore, TLR4 signal transduction inhibition confirmed that the TLR4 inhibitor TAK-242 could further attenuate the activation effect of MIF on the TLR4/MyD88/TRAF6/NF-κB pathway.

In conclusion, the present study shows that MIF exerts pathological effects in PD, and downregulation of MIF exerts protective effects in PD. Specifically, downregulation of MIF attenuates neuroinflammation by inhibiting MIF activation of the TLR4/MyD88/TRAF6/NF-κB pathway, which ultimately protects dopaminergic nerves from microglia activation-induced neurological injury and degeneration. Our findings may contribute to a better understanding of the relationship between MIF and neuroinflammation, and highlight MIF as a potential therapeutic target for PD, providing novel perspectives for the prevention and treatment of PD and drug development (Fig. 8g).

Methods

Animals and treatments

Eight-week-old, male, C57BL/6 mice were obtained from the Southern Medical University Laboratory Animal Center (Guangzhou, China). All the animal experiments were approved by the Experimental Animal Ethics Committee of Nanfang Hospital, Southern Medical University (Approval No. NFYY-2021-1117). We have complied with all relevant ethical regulations for animal use. The mice were kept in a 12-hour light/12-hour dark cycle, with unrestricted availability of food and water. The mice were divided into six groups in a random manner: control, control+AAV-MIF-NC, control+AAV-MIF-shRNA, MPTP, and MPTP + AAV-MIF-NC, MPTP + AAV-MIF-shRNA. According to their grouping, mice were first intracranially injected with MIF-shRNA or NC-shRNA (injection details below) and treated with MPTP or saline 7 days later. According to the method of previous studies, MPTP (Sigma-Aldrich, St. Louis, MO, USA)was administered intraperitoneally (i.p.) for five consecutive days at a dose of 30 mg/kg free base (MPTP-HCl) in saline.

Stereotaxic injection into the substantia nigra

The adeno-associated virus utilized in this research was acquired from Shanghai SunBio Medical Biotechnology Co., LTD. The sequence was CTCCACGTAGTGTTCTGTGTT. Before administering MPTP, either MIF-shRNA or NC-shRNA was injected into the SNpc of the left hemisphere using stereotaxic techniques(target coordinates from Bregma: AP, +1.2 mm; ML, −3.0 mm; DV, −4.3 mm). One day after the last injection of MPTP, behavioral experiments were conducted, and animals were euthanized to collect tissue.

Open field test

Assessment of mice motor behavior in a square open field (44 cm(l) × 44 cm(w) × 30 cm(h)). After being positioned in the middle of the chamber, the mice were monitored for a duration of 10 min. The behavioral parameters of the mice were recorded with a video camera and analyzed using the Shanghai JiLiang Animal Behavior software. (One test was performed for each mouse). At the conclusion of every experiment, the apparatus was sanitized using ethanol with a concentration of 75%.

Pole-climbing test

A wooden pole with a ball head (2 cm in diameter), 1 cm in diameter, and 50 cm in length was prepared, and gauze was wrapped around it to prevent slippage. The mice were positioned upside down on the wooden pole’s surface, encouraged to descend the pole naturally⁷⁹, and the time it took for the mice to go from the top to the bottom of the pole was recorded.

Hanging test

A solid foam box had a stainless steel wire attached to it, with a diameter of 1.5 mm and a length of 30 cm. The wire was fixed at a distance of 30 cm from the bottom of the box. Three times a day, with a 10 min interval, the recording was done for each mouse’s wire stomping or hind limb falling. A formal test was performed on the fourth day, and the time from the beginning to the fall of each mouse was recorded for statistical analysis.

Rotarod test

After setting the speed and turning on the power, the wheels can be automatically rotated at a fixed speed of 5 rpm for 5 min. Mice undergo this training three times daily for three consecutive days. On the fourth day, a formal test was performed in which the baton spinner was set to accelerate from 5 rpm to 40 rpm in 5 min. The time spent by each mouse on the baton spinner was recorded. Data were recorded for statistical analysis.

Inhibitor treatment

MIF inhibitor, (S, R)-3-(4-hydroxyphenyl)-4,5-dihydro-5-isoxazole acetic acid methyl ester (ISO-1, HY-16692) and Resatorvid (Toll-like receptor 4 inhibitor, HY-11109) were purchased from MedChemExpress. MN9D cells or BV2 cells were pretreated with ISO-1 (different concentration gradients) for 1 h, and BV2 cells were pretreated with TAK-242 (1 μM) for 2 h. Supplementary Table 1and Supplementary Table 2 contain the reagents and antibodies used in our study.

Cell culture

MN9D cells or BV2 cells were cultured in Dulbecco’s modified Eagle medium (DMEM; Gibco, NY, USA) supplemented with 10% fetal calf serum (Sigma-Aldrich, USA) in 5% CO₂ at 37 °C. And siRNA was purchased from Suzhou Genepharma Co., Ltd. MN9D cells or BV2 cells treated with or without MPP⁺ (Sigma-Aldrich, St. Louis, MO, USA) were transfected with MIF-siRNA and negative controls according to the manufacturer’s instructions provided by Genepharma. Groups with MPP⁺ treatment were added 24 h after transfection and continued to be cultured for 24 h.

Mif-Mus- siRNA1 sequence: sense (5‘-3’): GCCUAUGUUCAUCGUGAACTT; antisense(5‘-3’): GUUCACGAUGAACAUAGGCTT.

Mif-Mus- siRNA2 sequence: sense (5‘-3’): GCUCCACGUAGUGUUCUGUTT; antisense(5‘-3’): ACAGAACACUACGUGGAGCTT.

Cell-counting kit-8 assay

In each well of 96-well plates, 1 × 10⁴ MN9D cells or BV2 cells were seeded. Cells were pretreated with an inhibitor or siRNA transfection according to experimental needs, followed by incubation with MPP⁺ for 24 h. Cell viability was assessed using a CCK-8 assay, following the manufacturer’s guidelines, by employing 2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium mono-sodium salt (WST-8). Using a microplate reader (PerkinElmer, Hopkinton, MA, USA), the measurement of absorbance was taken at 450 nm.

ELISA

The concentrations of cytokines, including IL-6, TNF-α, and IL-1β, in BV2 cells’ medium supernatant and mouse serum were measured using ELISA kits, according to the manufacturer’s instructions.

Reverse transcription Real-Time Quantitative PCR (RT-qPCR)

We extracted total RNA from either total tissue or cells using a Trizol kit as per the instructions provided by the manufacturer (Life Technologies, Carlsbad, CA, U.S.A.). PrimeScript RT reagent kit (TaKaRa, Otsu, Japan) was utilized for cDNA synthesis with the use of Total RNA. Quantitative PCR was performed using a SYBR Premix Ex Taq kit (TaKaRa, Otsu, Japan). The relative expression value of the target gene was calculated as the ratio of target cDNA to β-actin. Supplementary Table 3 contained a list of the primers utilized in the real-time PCR.

Total protein extraction

Discard the medium, wash twice with 1 mL of pre-cooled PBS, add 1 mL of PBS to each well, scrape off the cells with a cell scraper, and transfer to a 2 mL EP tube. 10,000 rpm, centrifuge for 5 min, after centrifugation, discard the supernatant, re-centrifuge at 10,000 rpm × 5 min, after centrifugation, aspirate the supernatant. Prepare the cell lysate RIPA (P0013, Beyotime Biotechnology, Shanghai, China) and PMSF (ST505, Beyotime Biotechnology, Shanghai, China) in a ratio of 1:100, add the mixture according to the number of cells, and shake and mix well. Protein levels were determined using a BCA assay (P0012, Beyotime Biotechnology, Shanghai, China). Following centrifugation, the protein supernatant was transferred to a sterile EP tube for further analysis. The leftover protein solution was mixed with 5 × Loading buffer, heated at 100 °C for 10 min, cooled to ambient temperature, and preserved in a refrigerator at −80 °C for future experiments.

Western blotting

Proteins obtained from tissues or cells underwent separation through SDS-PAGE and were subsequently transferred onto a membrane made of polyvinylidene fluoride (PVDF). Next, according to the protein molecular weight of the target protein, the target band was cut, and the appropriate blocking solution (5% BSA or nonfat dry milk) was selected according to the detected band, and the band was blocked for 1.5 h–2 h at room temperature. The pictures were captured using a luminescent liquid (Clarity Western ECL Substrate, Bio-Rad, USA). The densities of protein bands were measured using the ImageJ software.

Immunohistochemistry

Three frozen sections were removed from each experimental group and allowed to thaw at room temperature for 10–15 min. Frozen sections are placed in a 40 °C oven and baked for 10 min before 1× PBS solution is used to wash the sections three times for 5 minutes each. The segments were immersed in 1× citrate buffer for antigen retrieval and subsequently allowed to cool down to room temperature. Break the membrane at room temperature for 30 min using 0.3% Triton-100. Expose the segments to a 3% solution of hydrogen peroxide for a duration of 10 minutes at ambient temperature. Add 5% BSA and block for 60 min. Following the blocking step, the primary antibody is introduced gradually (refer to the instructions for the appropriate antibody dilution ratio) and left to incubate overnight at a temperature of 4 °C. The following day, introduce the secondary antibody using the identical proportion and allow it to incubate for one hour at 37 degrees Celsius. The sections were visualized using 3,3-diaminobenzidine (DAB; KIT-0015, MXB Biotechnologies, Fujian, China) and hematoxylin solution. Pictures were obtained using a stereoscope. The Image-Pro Plus 6.0 analysis system (IPP 6.0; Media Cybernetics, Bethesda, MD, USA) was utilized to analyze the integrated optical density (IOD).

Immunofluorescent assay

The immunofluorescence staining procedure was conducted following the previously described method⁸⁰. Frozen sections were broken by using 0.3% Triton-100 for 30 min at room temperature, then placed in an incubator at 37 °C and blocked with 5% BSA for 1 h. Afterwards, the primary antibody was added dropwise and incubated overnight at 4 °C. The ratios of primary antibodies were all 1:200. For cell-climbing slices, the steps are the same as above, but the membrane does not need to be broken. The next day, take the wet box out of the refrigerator, equilibrate it at room temperature for 30 minutes to 1 h, add a drop of fluorescent secondary antibody, and incubate it in the incubator at 37 °C for 1 h. Next, the nuclei were stained with DAPI, and confocal photographs were taken. The immunostaining was subsequently analyzed with a Leica SP8 laser-scanning confocal microscope from Germany.

Statistics and reproducibility

All experimental data of this study were statistically analyzed using SPSS 27.0 software. All data were expressed as mean ± standard deviation (X ± S). Statistical methods included one-way analysis of variance (ANOVA) and Independent-Samples t-test, and the results of statistical analysis were analyzed by the LSD method if their variances were equal, and by Dunnett’s T3 post-hoc test if they were not. Dunnett’s T3 post-hoc test was used for analysis. The test level α = 0.05, *p < 0.05 considered the difference statistically significant, **p < 0.01 considered the difference statistically significant, ***p < 0.001 considered the difference and its significant statistical significance.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Author contributions

S.Q. designed the experiments, supervised the project, analyzed data, and wrote the manuscript. Y.H. performed the experiments, analyzed data, and wrote the manuscript. C.Z. performed experiments and analyzed data. All authors read and approved the final manuscript.

Peer review

Peer review information

Communications Biology thanks Rebecca Wallings and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editors: Ibrahim Javed and Benjamin Bessieres. A peer review file is available.

Data availability

All data generated or analyzed during this study are included in the main text and supplementary information. The source data behind the graphs in the main manuscript can be found in Supplementary Data 1. All uncropped and unedited blot images are included in the supplementary information. All other data are available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s42003-025-08997-7.