Bing-Qiang He, Ai-Cheng Li, Yu-Xuan Hou, Hui Li, Xing-Yuan Zhang, Hui-Fei Hao, Hong-Hua Song, Ri-Xin Cai, Ying-Jie Wang, Yue Zhou, Yong-Jun Wang,

Abstract The low intrinsic growth capacity of neurons and an injury-induced inhibitory milieu are major contributors to the failure of sensory and motor functional recovery following spinal cord injury. Heat shock transcription factor 1 (HSF1), a master regulator of the heat shock response, plays neurogenetic and neuroprotective roles in the damaged or diseased central nervous system. However, the underlying mechanism has not been fully elucidated. In the present study, we used a gecko model of spontaneous nerve regeneration to investigate the potential roles of gecko HSF1 (gHSF1) in the regulation of neurite outgrowth and inflammatory inhibition of macrophages following spinal cord injury. gHSF1 expression in neurons and microglia at the lesion site increased dramatically immediately after tail amputation. gHSF1 overexpression in gecko primary neurons significantly promoted axonal growth by suppressing the expression of suppressor of cytokine signaling-3, and facilitated neuronal survival via activation of the mitogen-activated extracellular signal-regulated kinase/extracellular regulated protein kinases and phosphatidylinositol 3-kinase/protein kinase B pathways. Furthermore, gHSF1 efficiently inhibited the macrophagemediated inflammatory response by inactivating IkappaB-alpha/NF-kappaB signaling. Our findings show that HSF1 plays dual roles in promoting axonal regrowth and inhibiting leukocyte inflammation, and provide new avenues of investigation for promoting spinal cord injury repair in mammals.

Key Words: apoptosis; gecko; heat shock factor 1; inflammation; neuron; regeneration; spinal cord; suppressor of cytokine signaling-3

Introduction

Traumatic spinal cord injury (SCI) leads to the innervation loss of sensory and motor neurons, and recovery of this innervation is minimal (Eckert and Martin, 2017; Eldahan and Rabchevsky, 2018; Fan et al., 2018; Zipser et al., 2022). The initial impact results in direct disruption of axons, blood vessels, and cell membranes and is followed by secondary pathologic events consisting of ischemia, inflammation, glial scar formation, and delayed apoptotic cell death, leading to further damage (Ahuja et al., 2017; Liu et al., 2022). The failure of spinal cord regeneration in adult mammals is attributed, to a large extent, to the low intrinsic regrowth capacity of neurons and to an inhibitory milieu composed of the glial scar and various other factors (Steketee et al., 2014; Sofroniew, 2018). Several lines of evidence suggest that activation of regeneration-associated genes (RAGs) and/or interference of environmental inhibitory factors can confer regenerative ability to the spinal cord following injury (Ma and Willis, 2015; Ineichen et al., 2017; Li et al., 2022; Muller et al., 2022). Interestingly, many of these RAG genes are regulated by hub molecules that either promote or inhibit their transcription (Ma and Willis, 2015; Cheng et al., 2022). For example, Krüppel-like factor 7 (KLF7) and signal transducer and activator of transcription 3 (STAT3) positively regulate axonal growth in the corticospinal tract, and suppressor of cytokine signaling-3 (SOCS3) and phosphatase and tensin homolog (PTEN) negatively regulate axonal growth in the corticospinal tract, by influencing growth-associated targets (Sun et al., 2011; Blackmore et al., 2012; Lang et al., 2013; Gallaher and Steward, 2018). However, the mechanisms underlying the failure of RAG upregulation after SCI and the simultaneous formation of a non-permissive milieu for axonal regeneration remain unclear. Insights gained from spinal cord regenerative models are expected to identify informative approaches to enhancing spinal cord repair.

Heat shock transcription factor 1 (HSF1) is a member of a family of DNAbinding proteins that activate the transcription of heat shock proteins (HSPs) to promote proper folding, trafficking, and degradation of misfolded proteins (Zhong et al., 1998; Gomez-Pastor et al., 2018). In an unstressed state, the inactive HSF1 monomer forms a complex with HSPs 40, 70, and 90, as well as the cytosolic chaperonin TCP1 ring complex, in the cytoplasm (Zhong et al., 1998). Upon stimulation by high temperature or tissue injury, the transcription factor dissociates from the complex to form a homologous trimer and translocates into the nucleus, where it binds to heat shock elements and promotes the transcription of HSPs and other downstream genes (Yue et al., 2016; Ran et al., 2018). HSF1 is biologically important to the central nervous system (CNS) because it helps promote neuronal survival and inhibit excessive inflammation following insult (Xie et al., 2002; Gadani et al., 2015; Yue et al., 2016; Li et al., 2021). Loss of HSF1 function leads to increased activation of microglia and apoptotic cells, which is associated with demyelination, astrogliosis, and impaired motor activity (Verma et al., 2014; Hassannejad et al., 2018; Szyller and Bil-Lula, 2021). Therefore, HSF1 represents a potential target for SCI therapy. To clarify the importance of HSF1 in improving spinal cord function following injury, investigating the roles it plays in spontaneous spinal cord regeneration is indispensable.

Several adult fish, amphibians, and reptiles can regenerate their spinal cords following injury (Diaz Quiroz and Echeverri, 2013; Zottoli et al., 2021). These vertebrates retain conserved regenerative processes and regulatory mechanisms, as has been observed at the anatomic and molecular levels (Chernoff et al., 2002). Lizards belonging to the amniotic clade, which exhibit a wide range of structural diversity, are capable of CNS repair (Romero-Aleman et al., 2004; Alibardi, 2019, 2020, 2022). Following transection of the lumbar spinal cord, in these animals, a limited number of axons regenerate across the bridge region between the rostral and caudal stumps (Alibardi, 2022). However, this injury evokes a milder inflammatory response in lizards than it does in their mammalian counterparts (Zhang et al., 2020; He et al., 2021a). Gekko japonicus has a remarkable ability to regenerate cartilage, muscles, and spinal nerves following tail amputation (Wang et al., 2012; Zhou et al., 2013; Szarek et al., 2016). Thus, in this study we used the gecko as a model to investigate the role of HSF1 in regulation of nerve regeneration and inflammation following SCI and identify novel treatment strategies for CNS repair in mammals.

Methods

Animals

Adult geckos (G. japonicus: 10–15 months old, body weight 4–6 g) were obtained from the Experimental Animal Center of Nantong University (License No. SYXK (Su) 2020-0029). The animals were housed in standard cages (10 geckos in each cage) in a room with a controlled temperature (25–28°C), an ambient humidity level of 40–50%, and a 12-hour light-dark cycle. The geckos were fed mealworms and hadad libitumaccess to water. Tail amputation was performed on equal numbers of male and female animals at the sixth caudal vertebra, identified by the special tissue structure present at that position (McLean and Vickaryous, 2011), by tying a slipknot in nylon thread, placing the thread around the tail at the designated point, and pulling gently until the tail was detached to mimic a natural injury.

All experiments were conducted in accordance with the guidelines of the National Institutes of Health Guide for the Care and Use of Laboratory Animals (8thed, National Research Council, 2011), and were approved by the Animal Care and Use Committee of Nantong University (approval No. S20190420-405) on April 20, 2019. All geckos were anesthetized prior to sacrifice by intraperitoneal injection of an anesthetic mixture containing 170 mg/kg chloral hydrate (Richjoint, Shanghai, China), 84.8 mg/kg magnesium sulfate (MgSO4) (YONGHUA CHEMICAL Co., Ltd., Shanghai, China), 35.4 mg/kg sodium pentobarbital (Toronto Research Chemicals, Toronto, Canada), 0.57 ml/kg absolute alcohol (Xilong Scientific Co., Ltd. Shantou, China), and 1.35 mL/kg 1,2-propanediol (Xilong Scientific Co., Ltd., Shantou, Guangdong Province, China).

Drug administration

Adult geckos were randomly assigned to the vehicle group (n= 9) or the 17-allylamino-17-demethoxygeldanamycin (17-AAG) (MilliporeSigma, Burlington, MA, USA) group (n= 9). The stock solution of 17-AAG, an HSF1 agonist, was prepared by dissolving it in dimethyl sulfoxide (MilliporeSigma) to a concentration of 100 mM before use. After tail amputation, the 17-AAG solution was diluted with 0.01 M PBS to a concentration of 10 mM, and 8 μL of 10 mM 17-AAG was slowly injected intraperitoneally every 2 days. The vehicle group received an equal volume of 10% dimethyl sulfoxide diluted with 0.01 M PBS (pH 7.4). Tissue samples (0.5-cm spinal segments) were collected from the injured site. The regenerated spinal tissue was imaged by stereomicroscopy (SZ61; Olympus, Tokyo, Japan) immediately after tail amputation and 7 and 14 days following tail amputation.

Multiple alignment and phylogenetic analysis of amino acid sequences

The amino acid sequences of HSF1 proteins from representative vertebrates were obtained from the National Center for Biotechnology Information (NCBI). The CLUSTAL program (MegAlign, DNAStar, Madison, WI, USA) was used to align the multiple protein sequences. The neighbor-joining method (PHYLIP package 3.5c, http://evolution.genetics.washington.edu/phylip.html) was used for phylogenetic analyses.

Western blotting

Protein was extracted from the 0.5-cm spinal cord segments taken from the lesion site (rostral to the amputation plane,n= 6 for each time point) 0, 1, 3, 7, and 14 days following tail amputation, or from the human neuroblastoma cell line SH-SY5Y (Shanghai Institute of Biochemistry and Cell Biology (SIBCB), Shanghai, China, Cat# SCSP-5014, RRID: CVCL_0019) and the murine macrophage cell line RAW 264.7 (SIBCB, Cat# SCSP-5036, RRID: CVCL_0493) after they had been subjected to various treatments, using a buffer containing 1% sodium dodecyl sulfatepolyacrylamide (SDS), 100 mM Tris-HCl, 1 mM phenylmethanesulfonyl fluoride (PMSF), and 0.1 mM β-mercaptoethanol. The protein concentration of each sample was quantified using a Bradford protein assay kit (Beyotime, Shanghai, China, Cat# P0006) to ensure equal loading. Protein extracts were heat-denatured at 95°C for 5 minutes, and 20 μg of each sample was electrophoretically separated on 10% SDS-PAGE gel (Beyotime, Cat# P0012AC), followed by transfer onto a polyvinylidene difluoride (PVDF) membrane (BioRad, Hercules, CA, USA). The membrane was blocked with 5% skim milk (Beyotime) in Tris-buffered saline (Beyotime) containing 0.1% Tween-20 for 1 hour, followed by overnight incubation with primary antibodies at 4°C. After washing three times with TBST for 10 minutes each, the membrane was incubated with secondary antibodies for 2 hours at room temperature. The HRP activity was detected using an enhanced chemiluminescence (ECL) kit (Beyotime, Cat# P0018). Images were taken with a GS800 Densitometer Scanner (Bio-Rad, Hercules, CA, USA), and the data were analyzed using PDQuest 7.2.0 software (Bio-Rad, Hercules, CA, USA). β-actin expression was used as an internal control. The primary and secondary antibodies used are shown in Table 1.

Immunofluorescence staining

Spinal cord segments (0.5 cm rostral to the amputation plane) were harvested from six experimental animals at each time point and fixed with 4% paraformaldehyde in 0.01 M PBS (pH 7.4) at 4°C overnight. Then, the tissues were dehydrated in 10%, 20%, and 30% sucrose solutions in 0.01 M PBS (pH 7.4) for 24 hours at each concentration. Finally, they were embedded in Tissue-Tek Optimal Cutting Temperature (OCT) compound, frozen, and sectioned 10 μm thick slices using a cryostat (CM3050S, Leica, Weztlar, Germany). The sections were blocked with 0.01 M PBS containing 3% BSA (BioFroxx, Guangzhou, China, Cat# 4240GR100), 0.1% Triton X-100 (Cat# ST797, Beyotime), and 10% normal goat serum (Bioss, Beijing, China, Cat# C-0005) for 1 hour at 37°C, incubated overnight at 4°C with primary antibodies, and further incubated overnight at 4°C with secondary antibodies. After a final incubation with Hoechst 33342 (1:4000, MilliporeSigma, Cat# 14533) for 15 min, the sections were observed under a fluorescence microscope (ZEISS, axio image M2). Immunostaining of cells was carried out according to the same procedure, with the exception of taking frozen sections. Quantification of neurite length was performed using Image-Pro Plus 6.0 software (Media Cybernetics, Rockville, MD, USA). Experiments were performed in triplicate, and 50 images were analyzed for each time. The primary and secondary antibodies used are shown in Table 1.

Table 1 ｜The primary and secondary antibodies used in western blot assay

Construction of the HSF1-overexpression adenovirus

The HSF1-overexpression adenovirus (GV345-HSF1) was constructed by Genechem Co. Ltd. (Shanghai, China). Briefly, the open reading frame (ORF) of gecko HSF1 was cloned into a GV345 vector via the AgeI and BamHI sites. The EF-1α promoter drove the expression of HSF1, and Cherry expression was driven by the CMV promoter. The 293T cells (SIBCB, Cat# SCSP-502; RRID: CVCL_0063) were used to produce HSF1-overexpression adenovirus (Sena-Esteves and Gao, 2018) to a titer of 2 × 1010PFU/mL.

Cell culture conditions and treatments

Primary neurons from the adult gecko cerebral cortex were isolated and cultured according to previously described methods (He et al., 2021b). Briefly, the geckos were euthanized following injection of anaesthetic mixtures, and sterilized with 75% alcohol before harvesting the entire brain. The brain was placed in a precooled physiological solution comprising a 1:1 mixture of Dulbecco’s modified Eagle’s medium and Ham’s F-12 (MilliporeSigma) medium, 100 U/mL penicillin, and 0.1 mg/mL streptomycin (penicillinstreptomycin solution, Beyotime). Then, the cerebral cortex was isolated and digested with 1 ml of 0.2% collagenase (Solarbio) for 15–18 minutes in an incubator, with shaking every 5 minutes, followed by digestion with 4 mL of 0.25% trypsin (Gibco, Billings, MT, USA) for 20–25 minutes at 30°C, with shaking every 5 minutes. The cell dissociation reaction was stopped by adding an equal amount of DMEM/F12 medium supplemented with 10% fetal bovine serum (FBS, Gibco) and penicillin-streptomycin solution, and the suspension was then filtered through a 200 mesh sieve. The filtered solution was collected and centrifuged at 168 ×gfor 10 minutes, the supernatant was discarded, and the cells were resuspended in DMEM/F12 medium supplemented with 10% FBS and penicillin-streptomycin. The cells were then plated into culture dishes (35 × 10 mm2) coated with PDL (MilliporeSigma) at a density of 2 × 106cells/mL and maintained in a humidified atmosphere containing 5% CO2for 24 hours at 30°C, followed by culture in Neurobasal medium (Gibco) supplemented with 2% B27 (STEMCELL Technologies, Vancouver, BC, Canada, Cat# 05711), penicillin-streptomycin solution, 0.5 μg/mL NGF (Cat# 556-NG, RDSYSTEMS, Minneapolis, MN, USA), and 1% glutamine (Beyotime). Five days later, the neurons were transfected with GV345-vector or GV345-HSF1 adenovirus for 72 hours, and the immunofluorescence was observed.

The human neuroblastoma cell line SH-SY5Y, which is commonly used as a neuronal model because of its similar physiological properties to neurons, and the murine macrophage cell line RAW 264.7, which exhibits immune functions similar to those of peripheral macrophages, were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) - high glucose medium (Millipore, Sigma) containing 10% fetal bovine serum (Gibco) and penicillinstreptomycin solution (Beyotime) in a 37°C humidified incubator (Thermo Fisher Scientific, Waltham, MA, USA) with 5% CO2. The RAW 264.7 cells were transfected with GV345-vector or GV345-HSF1 adenovirus for 48 hours and then treated with 0.5 μg/mL lipopolysaccharide (LPS) for 6 hours. The SH-SY5Y cells were transfected with GV345-vector or GV345-HSF1 adenovirusor for 48 hours, followed by glucose deprivation (GD) for 24 hours. The GD insult was delivered as described previously (Ferretti et al., 2016; Alherz et al., 2021).

Luciferase reporter gene assay

The assay was performed using a Dual-Luciferase reporter System (firefly and Renilla luciferase assays, Promega, Madison, WI, USA) following the manufacturer’s instructions. The gecko SOCS3 (gSOCS3) promoter sequence (–2704 to +1) was PCR-amplified genomic gecko DNA as a template and then digested with Hind III and Kpn I before cloning into the plasmid pGL3-Basic to yield pGL3-gSOCS3. The gecko HSF1 (gHSF1) amplified from cDNA of spinal cord was digested with BamH I and EcoR I and cloned into the pcDNA3.1 plasmid to yield pcDNA3.1-gHSF1. The plasmids were transfected into 293T cells in 24 well dishes at 70% confluency using X-tremeGENE HP (2 μL for 1 μg plasmids). The Renilla and firefly luciferase activities were measured 48 hours after transfection following the manufacturer’s instructions. Luciferase activity was normalized to Renilla luciferase activity.

Cell viability assay

SH-SY5Y cell viability after GD treatment for 3, 6, 12, and 24 hours was determined using a Cell Counting Kit-8 (CCK-8, Dojindo, Kyushu Island, Japan) in accordance with the manufacturer’s protocol. Absorbance at 450 nm was measured using a microplate reader (Bio Tek Synergy2, Winooski, VT, USA).

Apoptosis assay

SH-SY5Y cell apoptosis was analyzed using an Annexin V-Fluorescein isothiocyanate (FITC) Apoptosis Detection Kit (Beyotime) in accordance with the manufacturer’s protocol. In brief, following treatment the cells were incubated with Annexin V-FITC solution (Cat# C1062, Beyotime) for 20 minutes at room temperature in the dark. The nuclei were counterstained with Hoechst 33342 (MilliporeSigma, Cat# 14533). Cells were imaged with a Zeiss fluorescence microscope, and the data were analyzed using Image-Pro Plus 6.0 software (Media Cybernetics, Rockville, Md, USA). Briefly, the Choose Region of Interest (ROI) Manager tool was used, measurements were carried out using the Analyze > Analyze Particles function, and the results were automatically recorded and displayed in the results panel.

Enzyme linked immunosorbent assay

Tumor necrosis factor (TNF)-α, interleukin (IL)-1β, and IL-6 expression levels were quantified by two-site sandwich enzyme linked immunosorbent assay (ELISA; Dominguez et al., 2019). RAW 264.7 cells were transfected with adenovirus transfection for 48 hours, then incubated with 0.5 μg/mL LPS for 6 hours. The supernatant was harvested, and the cells were lysed in buffer containing 100 mM Tris-HCl, 1 mM PMSF, 1% SDS, and 0.1 mM β-mercaptoethanol. The lysates were centrifuged at 13,800 ×gfor 15 minutes, and then the TNF-α, IL-1β, and IL-6 expression levels were measured using the appropriate ELISA kits (MULTI SCIENCES, Hangzhou, China). A multifunctional enzyme marker was added (Biotek Synergy2, Winooski, VT, USA), and the plates were read at a 450-nm wavelength.

Statistical analysis

No statistical methods were used to predetermine sample sizes; however, our sample sizes are similar to those reported in previous publications (Zhang et al., 2019; Kim et al., 2021; Zhao et al., 2022a). The gross observation of the regenerating tail was conducted by investigators who were blinded to the experimental and treatment conditions. GraphPad Prism 8 software (GraphPad Software, San Diego, CA, USA, www.graphpad.com) was used for statistical analysis. All data were presented as mean ± SEM. Comparisons between two groups were analyzed by two-tailed unpaired Student’st-test. Differences between multiple groups were analyzed by one way or two-way analysis of variance followed by followed by Dunnett’s or Tukey’spost hoctest.P< 0.05 was considered statistically significant.

Results

Characterization of the gHSF1 amino acid sequence

To shed light on the structural characteristics of gHSF1 (GenBank accession number XP_015273988), the deduced gHSF1 amino acid sequence was aligned with HSF1 sequences from other vertebrates. The results demonstrated that gHSF1 is 522 amino acids in length and contains the conserved DNA-binding domain, leucine zippers 1–4, the regulatory domain (RD) related to gHSF1 posttranslational modification, and the activation domain (AD) (Gomez-Pastor et al., 2018). gHSF1 shares 64.5–84.9% identity with human, rat, mouse, chicken, lizard, frog, and zebrafish HSF1 (Additional Figure 1A and B). Phylogenetic analysis showed that gHSF1 clusters with amniote HSF1 sequences rather than with frog and fish HSF1 sequences (Additional Figure 1C), suggesting that the physiological function of HSF1 was considered during the evolution of amniotes.

Tail amputation induces gHSF1 expression in neurons and microglia in a gecko model of spinal cord injury

HSF1 has been shown to protect cells against various stresses (Li et al., 2017). To elucidate the potential roles of gHSF1 in the injured spinal cord in a gecko model, 0.5-cm cord segments proximal to the amputation plane were collected 0, 1, 3, 7, and 14 days after tail amputation. Western blot analysis demonstrated that SCI significantly increased gHSF1 levels from day 1 onwards, with a peak at 3 days that returned to the control level at 14 days (P= 0.0013; Figure 1A and B). To understand how gHSF1 mediates cellular events within neurons and microglia, immunostaining was performed to observe its colocalization with markers of the two cell types following tail amputation. The results showed that gHSF1 colocalized with NeuN-positive neurons (Figure 1C) and OX42-positive microglia (Figure 1D), but not GFAP-positive astrocytes (Additional Figure 2A), before and after SCI. Notably, gHSF1 primarily localized to the nuclei, as determined by co-staining with Hoechst 33342 (Additional Figure 2B). These findings indicate that gecko SCI results in a significant increase in gHSF1 expression within neurons and microglia, and that this phenomenon is involved in successful nerve regeneration.

Figure 1｜Dynamic changes in gHSF1 expression in the injured gecko spinal cord.

gHSF1 promotes axonal elongation by suppressing gSOCS3 expression

To determine the effects of gHSF1 on gecko neurons after SCI, primary neurons were isolated from the cortex and cultured as previously described (He et al., 2021b). When the primary neurons were cultured for 1 dayin vitro(DIV1), most of the cells adhered to the bottom of the cell culture dish. At DIV3, however, sprouting neurites were visible on some cells (Figure 2A–D). To test the effect of gHSF1 overexpression on nerve regeneration, neurons over 95% pure, as evaluated by β-III-tubulin staining (Additional Figure 3A and B). Immunostaining showed that gecko neurons constitutively expressed gHSF1 (Additional Figure 3C). The neurons at DIV5 were transfected with gHSF1 adenovirus (Ad-gHSF1) for 72 hours, and structural analysis showed that this overexpression facilitated axonal elongation in comparison with the control cells (Figure 2E–G).

To explore how gHSF1 mediates neuronal growth, we looked for HSF1-binding elements in the promoters of several genes encoding axonal growth–related inhibitory molecules. As expected, the promoter of SOCS3, which is a critical axonal growth inhibitor (Sun et al., 2011; Gallaher and Steward, 2018), contained 12 putative HSF1-binding elements, as predicted by Jaspar analysis (https://jaspar.genereg.net/) (Additional Figure 3D and E). Therefore, the gSOCS3 promoter (–2074 to –1) was cloned into a dual-luciferase reporter gene construct. The luciferase reporter gene assay revealed that gHSF1 markedly decreased gSOCS3 expression (Figure 2H). Taken together, these findings indicate that gHSF1 promotes axonal elongation in geckos by binding to the gSOCS3 promoter and inhibiting its expression.

Figure 3｜Effects of gHSF1 on neuronal apoptosis.

gHSF1 overexpression attenuates glucose deprivation-induced neuronal apoptosis

Neuronal apoptosis occurs after SCI, and HSF1 can protect neurons from death caused by the accumulation of misfolded proteins (Verma et al., 2014; Hassannejad et al., 2018). To further elucidate the pathophysiological roles of increased gHSF1 expression in gecko neurons following SCI, glucose deprivation (GD) was used to induce neuronal apoptosis (Ferretti et al., 2016). SH-SY5Y cells were cultured in glucose-free medium for 3–24 hours, which led to a marked decrease in cell viability, as determined by CCK8 assay (Additional Figure 4). Annexin V-FITC staining showed that GD for 24 hours increased the number of apoptotic cells (Figure 3A and B). However, when the cells were transfected with Ad-gHSF1 for 48 hours and then subjected to GD for 24 hours, the number of apoptotic cells was markedly reduced (Figure 3C and D). These findings indicate that gHSF1 protects neurons from apoptosis.

Figure 2｜Effects of gHSF1 overexpression on gecko neuron neurite growth.

gHSF1-mediated neuronal survival involves activation of intracellular MEK/ERK and PI3K/AKT pathway

To clarify the mechanism of gHSF1-mediated protection from apoptosis, we transfected SH-SY5Y cells with Ad-gHSF1 for 48 hours, followed by cell culture in glucose-free medium for 24 hours, and assessed apoptosis-related signaling pathways. The transfection efficiency was over 80% (Additional Figure 5A–D). Western blot analysis demonstrated that caspase3 activity was significantly attenuated by gHSF1 overexpression. In addition, the expression of Bcl-XL, which is an anti-apoptotic member of the Bcl-2 family, was markedly increased under the same conditions (Figure 4A–D). Because the MEK/ERK and PI3-K/AKT pathways regulate Bcl-XL’s anti-apoptotic activity in different cell types (Ramljak et al., 2003; He et al., 2021c), we assessed the levels of both phosphorylated ERK1/2 and phosphorylated AKT in SH-SY5Y cells following transfection with Ad-gHSF1. The results showed that both ERK1/2 kinase and AKT kinase were significantly activated by gHSF1 overexpression in comparison with the control cells (Figure 4A, E and F). These findings indicate that gHSF1-mediated protection of neurons from apoptosis involves activation of MEK/ERK and PI3K/AKT signaling.

gHSF1 suppresses LPS-induced macrophage-mediated inflammation through inactivation of IκB-α/NF-κB signaling

HSF1 has been shown to antagonize acute inflammation through transcriptional repression of cytokine gene expression (Xie et al., 2002). To determine the effects of gHSF1 on the macrophage-mediated inflammatory response, RAW 264.7 macrophages were transfected with Ad-gHSF1 for 48 hours, followed by treatment with 0.5 μg/mL LPS for 6 hours. The transfection efficiency was over 80% (Additional Figure 5E–H). ELISA analysis showed that TNF-α, IL-1β, and IL-6 levels in the cell supernatants and lysates after stimulation with LPS was markedly attenuated by gHSF1 overexpression (Figure 5A–F), indicating that gHSF1 is able to suppress LPS-mediated inflammatory activation of macrophages.

To elucidate the mechanism underlying gHSF1-mediated suppression of the macrophage inflammatory response, RAW 264.7 macrophages were transfected with Ad-gHSF1 for 48 hours, followed by stimulation with 0.5 μg/mL LPS for 6 hours. Western blot analysis showed that HSP70, which is a downstream regulatory molecule of HSF1 involved in inhibition of inflammation (Krause et al., 2015; Sevin et al., 2015), was not affected by gHSF1 overexpression (Figure 6A–C). However, phosphorylated IκB-α and p65NF-κB expression levels were markedly decreased by Ad-gHSF1 transfection (Figure 6A, D and E). These findings indicate that gHSF1 suppresses LPS-induced macrophage inflammation by inactivating the IκB-α/NF-κB pathway.

Figure 4｜gHSF1 activates the mitogen-activated extracellular signal-regulated kinase/extracellular regulated protein kinases (MER/ERK) and phosphatidylinositol 3-kinase/protein kinase B (PI3K/AKT) signaling pathways in SH-SY5Y cells.

Figure 5｜gHSF1 suppresses LPS-induced inflammation in RAW 264.7 macrophages.

HSF1 activation has no effect on gecko tail regeneration

To understand the effects of gHSF1 overexpression on gecko tail regeneration, the animals were intraperitoneally injected with 8 μL of 10 mM HSF1 agonist 17-AAG or vehicle following tail amputation. Western blot analysis demonstrated that gSOCS3 levels at the lesion site were markedly decreased at 7 days after amputation in the agonist group compared with the vehicle group (Figure 7A–C). Gross observation of the regenerating tail at 0, 7, and 14 days after amputation revealed that 17-AAG treatment had no effect on the regenerating tail in comparison with vehicle (Figure 7D and E). These findings indicate that excessive HSF1 activation has no detectable effect on tail regeneration in geckos.

Figure 6｜gHSF1 inhibits IkappaB-alpha/NF-kappaB (IκB-α/NF-κB) signaling.

Figure 7｜Effects of HSF1 agonist on gecko tail regeneration.

Discussion

Adult mammals have a limited capacity for spinal cord regeneration following injury, and currently, no effective method is available to restore the disrupted motor function below the injury site (O’Shea et al., 2017; Zhao et al., 2022b). Investigation of several vertebrate species that are capable of spinal cord regeneration shows that axonal regeneration, inflammation and the immune response, and neurogenesis contribute to their high regenerative capacity (Zhou et al., 2013; He et al., 2021c; Alibardi, 2022; Donato and Vickaryous, 2022). Therefore, identifying the critical factors involved in regulating spinal cord regenerative capacity may lead to novel approaches to SCI repair. Since the transcription factor HSF1 was first discovered, it has been found to play multiple physiological and pathological roles, especially in promoting cell survival, inhibiting the inflammatory response, mediating tumorigenesis, and protecting cells from protein misfolding (Zelin and Freeman, 2015; Dai, 2018; Carpenter and Gokmen-Polar, 2019; Ahmed et al., 2020; Levi-Galibov et al., 2020; Pincus, 2020). Although endotherms have evolved to have CNSs that are well-protected against external stimuli, stress-induced HSF1 expression also plays a protective role in response to CNS insults (Hashimoto-Torii et al., 2014; Pignataro, 2019; Yao et al., 2019; Liu et al., 2020). Unexpectedly, HSF1 is not activated in mammals until 4 days after SCI (Li et al., 2021). By that time, neuronal apoptosis and excessive inflammation have already occurred (Zhang and Gensel, 2014). This explains, at least in part, why endogenous HSF1 does not play a considerable protective role against pathogenesis following mammalian SCI. Comparatively, gHSF1 expression was immediately induced within gecko neurons and leukocytes after SCI, which might favor nerve regeneration by protecting neurons from apoptosis and limiting excessive inflammation.

SOCS3 is a key negative regulator of axonal regrowth and cell survival following SCI (Liu et al., 2015). One study has summarized the critical roles of the SOCS3-dependent JAK/STAT pathway in neuronal loss after SCI (Park et al., 2014). Suppression of SOCS3 or of the JAK/STAT pathway in the context of axonal damage can promote axonal regeneration (Kretz et al., 2005; Miao et al., 2006; Sekine et al., 2018; Lindborg et al., 2021). Therefore, SOCS3 is recognized as a negative regulator of RAGs (Smith et al., 2009). In the present study, gHSF1 was found to promote neuronal survival and neurite growth by suppressing transcription of gSOCS3, suggesting a novel regulatory mechanism for axonal regeneration. However, SOCS3 is most widely known for the role it plays in inhibiting inflammatory activation (Yoshimura et al., 2012). Here we demonstrated that gHSF1 overexpression in macrophages inhibited, rather than activated, the inflammatory response. One reason for this may be that gHSF1-mediated inhibition of SOCS3 is only minimally involved in activating inflammation-related signaling pathways, whereas gHSF1-mediated inactivation of the IκB-α/NF-κB pathway is the main antiinflammatory mechanism.

The role of HSF1 in chaperone regulation is dependent on HSF1 trimerization. Increasing evidence indicates that monomeric HSF1 also has important functions (Huang et al., 2018). For example, monomeric HSF1 protects cultured neurons from heat shock even in the absence of chaperone induction (Verma et al., 2014). As such, both amyotrophic lateral sclerosis drug riluzole, anti-oxidative and anti-inflammatory agent methylene blue were approved by Food and Drug Adinistration (FDA) to protect damaged cells by increasing the amount of the monomeric HSF1 (Yang et al., 2008; Huang et al., 2017). In the present study, we found that gHSF1 overexpression promoted neuronal survival, similar to monomeric HSF1.

The activity of HSF1 in suppressing the leukocyte-mediated inflammation is apparent at multiple levels. HSF1 can directly inhibit TNF-α and IL-1β transcription by binding to their promoter regions (Singh et al., 2002; Xie et al., 2002). Furthermore, it can repress NLRP3 inflammasome activation and IL-6 production (Takii et al., 2010; Yue et al., 2016). HSF1-mediated inhibition of NLRP3 involves activation of β-catenin (Yue et al., 2016), which in turn decreases nuclear translocation of NF-κB, the hub regulator for innate immunity (Bethea et al., 1998; Bouwmeester et al., 2004; Hu et al., 2021). In addition, HSF1 has been found to affect activity of the inhibitory κB kinase signalosome by negatively regulating NFκB signaling under heat shock stress conditions (Paszek et al., 2020). In the present study, we found that gHSF1 inhibits macrophage-mediated inflammation via decreasing phosphorylated IκB-α and p65NF-κB levels, suggesting the importance of HSF1-mediated NFκB signaling in suppression of the inflammatory response. As activation of inflammation influences pathological progress or spinal cord regeneration in both mammals and regenerative models (Alibardi, 2014; Gadani et al., 2015; He et al., 2021a), gHSF1-mediated suppression of macrophage inflammation may indirectly contribute to axonal regeneration following gecko SCI. Blocking p65NF-κB also promotes axon regeneration in the CNS, independent of its role in the inflammatory cascades elicited in macrophages (Haenold et al., 2014). Therefore, the multiple effects of HSF1-mediated NF-κB signaling on axonal regeneration remain to be elucidated.

Although gHSF1 was shown to promote neuronal survival and neurite elongation, as well as inhibit macrophage inflammation following gecko SCI, it failed to facilitate tail regeneration. Therefore, the limitation of this study is that it did not explore the function of gHSF1 in oligodendrocytes and astrocytes in the gecko spinal cord.

In conclusion, gHSF1 expression was inducibly elevated in the neurons and microglia of the injured gecko spinal cord following tail amputation. gHSF1 overexpression facilitated neurite outgrowth and neuronal survival and suppressed macrophage-mediated inflammation through inactivation of NF-κB signaling. The results from our nerve regeneration model reveal a dual role for HSF1 in promoting neurite outgrowth and inhibiting leukocyte inflammation, providing new avenues for exploring functional recovery of the injured spinal cord.

Acknowledgments:The authors acknowledge that the graphical abstract was created with BioRender.com.

Author contributions:Study design: YJunW; manuscript writing: YJunW; experiment implementation: BQH; data analysis: YJunW, BQH, ACL, YXH, HL, XYZ, HFH, YZ, and HHS; data discussions: RXC and YJieW; manuscript revision: YJunW, BQH. All authors approved the final version of the manuscript.

Conflicts of interest:The authors have no conflict of interest to declare.

Data availability statement:All relevant data are within the paper and its Additional files.

Open access statement:This is an open access journal, and articles are distributed under the terms of the Creative Commons AttributionNonCommercial-ShareAlike 4.0 License, which allows others to remix, tweak, and build upon the work non-commercially, as long as appropriate credit is given and the new creations are licensed under the identical terms.

Additional files:

Additional Figure 1: Multiple alignment and phylogenetic analysis of gHSF1 with those of representative vertebrates.

Additional Figure 2: Colocalization of gHSF1 (red) with GFAP (green) and Hoechst (blue) in the injured spinal cord.

Additional Figure 3:Primary culture of gecko neurons and gHSF1 binding sites at gecko SOCS3 promoter.

Additional Figure 4: CCK-8 assay of cell viability following SH-SY5Y cell culture in a glucose-free medium for 3, 6, 12, and 24 hours, respectively.

Additional Figure 5: Transfection efficiency of Ad-gHSF1 on SH-SY5Y cells or RAW 264.7 macrophages.