FGF1 improves functional recovery through inducing PRDX1 to regulate autophagy and anti‐ROS after spinal cord injury

Abstract Fibroblast growth factor 1 (FGF1) is thought to exert protective and regenerative effects on neurons following spinal cord injury (SCI), although the mechanism of these effects is not well understood. The use of FGF1 as a therapeutic agent is limited by its lack of physicochemical stability and its limited capacity to cross the blood‐spinal cord barrier. Here, we demonstrated that overexpression of FGF1 in spinal cord following SCI significantly reduced tissue loss, protected neurons in the ventricornu, ameliorated pathological morphology of the lesion, dramatically improved tissue recovery via neuroprotection, and promoted axonal regeneration and remyelination both in vivo and in vivo. In addition, the autophagy and the expression levels of PRDX1 (an antioxidant protein) were induced by AAV‐FGF1 in PC12 cells after H2O2 treatment. Furthermore, the autophagy levels were not changed in PRDX1‐suppressing cells that were treated by AAV‐FGF1. Taken together, these results suggest that FGF1 improves functional recovery mainly through inducing PRDX1 expression to increase autophagy and anti‐ROS activity after SCI.

autophagy, chaperone-mediated autophagy (CMA), was reported to be triggered by resisting ROS-induced motoneuronal death during spinal cord destruction. 12,13 These results suggest that autophagy may play an important role in acute trauma.
PRDX1 is expressed mainly in the cytosol and participates in the regulation of several ROS-dependent signalling pathways and is considered a key intracellular factor maintaining the balance of cell survival and apoptosis. 14,15 Most studies of PRDX1 have focused on its effect in cancer, but there have been only few investigating its role in acute injury.
Fibroblast growth factors (FGFs) are expressed in several systems during cell generation, differentiation and migration. 16 Fibroblast growth factor 1 (FGF1) plays a crucial role in neuroprotection and axon regeneration in the nervous system, and its use has been examined for safety and feasibility by clinical trials. 17 However, administration of FGF1 by either subcutaneous or intravenous is ineffective in the treatment of SCI because FGF1 is a macromolecular protein with poor penetrability of the blood-spinal cord barrier (BSCB). Therefore, it is essential to find an effective delivery route for FGF1 with sustained release. Adenoassociated virus (AAV) can be used to remarkably promote gene expression with efficiency and longevity. 18 Here, the results showed that PRDX1 and autophagy are both augmented after up-regulation of FGF1, which indicate that FGF1 can influence ROS and autophagy to promote functional recovery after SCI. Additionally, PRDX1 may play a key role in mediation of the interaction between ROS and autophagy.

| AAV vector construction and virus production
The AAV-2 particles of FGF1 and lentivirus particles of PRDX1 were constructed and synthesized as previous study described. 19 Spinal cord microinjection of the recombinant AAV vector was performed with a stereotaxic instrument.

| Cell culture and virus transfection
PC12 cells were obtained from the Cell Storage Center of Wuhan University (Wuhan, China). Cells were maintained under appropriate conditions for proliferation using RMPI 1640 containing heat-inactivated 10% FBS, and were cultured in a humidified atmosphere of 5% CO 2 and 95% air at 37°C. AAV-FGF1 (1 9 10 9 TU) and LV-PRDX1 (MOI = 20) were added to PC12 cells at 24 hours after plating for 12 hours.

| Spinal cord injury model and virus injection
Spinal cord injury protocols were carried out in adult female SD rats as previously described. 20 Immediately after SCI, 10 lL of AAV was orthotopically injected at a dose of 1 9 10 9 TU using a microsyringe. The muscle and skin were sutured layer by layer and then sterilized by iodine. Post-operative monitoring included manual bladder emptying 3 times each day. Subsequently, the animals were killed at 7 or 14 days after SCI.

| Locomotion recovery assessment
Locomotion recovery was assessed using the Basso-Beattie-Bresnahan (BBB) locomotion scale and footprint analysis at 0, 1, 3, 5, 7 and 14 days. Rats were placed in an open experimental field and allowed to move freely for 5 minutes. Crawling ability was assessed by the BBB scale ranging from 0 (no limb movement or weight support) to 21 (normal locomotion). The footprint analysis was performed by dipping the animal's posterior limb with red dye and the forelimb with blue dye. Evaluations were performed by 5 independent examiners who were blinded to the experimental conditions.

| Tissue preparation
At specific time-points following SCI, the rats were anaesthetized with 10% chloral hydrate (ip, 3.5

| Histology and immunofluorescence staining
Longitudinal or transverse sections mounted on slides were prepared as previously described. 21 The sections for histopathological examination were subjected to H&E staining and LFB staining according to manufacturer's instructions.

| Statistical analysis
All data were expressed as means AE SEM. Differences between groups in BBB scales were analysed with 1-way ANOVA followed by Tukey's multiple comparison test. Statistical analysis of the other data was performed with 1-way analysis of variance (ANOVA). All statistical analyses were conducted using GraphPad Prism 5 for Windows. Differences were considered statistically significant when P < .05.

| Feasibility of AAV vector to mediate the overexpression of FGF1 for SCI treatment
In this study, AAV-FGF1 (5 9 10 11 pfu/kg) was injected into the lesion area of rat immediately after SCI. As shown in Figure 1A, the sham-operated group had an average score of 21, representing normal motor function. There was no significant difference between the groups (SCI group, the SCI + AAV-FGF1 group and the SCI + AAV-CON group) in the early stage (1, 3 and 5 dpi). However, with a longer time of observation, SCI + AAV-FGF1 group showed better posterior limb motor function and scored twofold to threefold higher than rats in the SCI group and the SCI + AAV-CON group at 7 dpi (days post-injury) and 14 dpi ( Figure 1A   showed that phosphorylation of mTOR was observably decreased after treatment with AAV-FGF1 following SCI. In contrast, the levels of ATG7, Beclin1 and LC3 II/I were significantly increased in the SCI + AAV-FGF1 group compared with the levels in the injured group and the control vector group. The expression level of P62, a biomarker of mature autophagic vesicles, was markedly decreased in the SCI + AAV-FGF1 group compared with that of the injured group and the control group. Overall, these data suggest that administration of FGF1 could enhance the expression of autophagy-associated proteins to stimulate the level of autophagy after injury. It seems reasonable that this effect may be related to the therapeutic effect of FGF1.

| FGF1 protects the cytoskeleton and promotes cell polarization in vitro
To study the effect of AAV-FGF1 in vivo, the AAV-FGF1 virus was transferred to PC12 cells (50 pfu/cell). To simulate the microcirculation of SCI, we provoked PC12 cells with H 2 O 2 (100 lmol/L) for 8 hours. We examined levels of ace-tubulin, an indicator protein of the stabilized cytoskeleton, and Tau protein, a microtubule-associated protein that plays a crucial role in microtubule assembly.
Immunostaining of ace-tubulin in PC12 cells was performed to evaluate the effect of FGF1 on cell morphology ( Figure 5D). We

| FGF1 enhanced autophagy in vitro
In order to investigate the effect of FGF1 on autophagy in PC12 cells, the expression levels of autophagy-associated proteins were measured after exposure to H 2 O 2 and treatment with AAV-FGF1.
Interestingly, compared with the H 2 O 2 group and the H 2 O 2 + LV-CON group, the phosphorylation of mTOR protein in the H 2 O 2 + AAV-FGF1 group was significantly decreased (Figure 6A,B).
As shown in Figure 6A  The result showed the up-regulation of FGF1 mediated by AAV vector markedly enhanced the expression degree of PRDX1 at 7 and F I G U R E 5 FGF1 protects the cytoskeleton and promotes cell polarization in vitro. A, Immunofluorescence images show ace-tubulin (green) in PC12 cells subjected to H 2 O 2 (100 lmol/L) and treated with AAV-FGF1. Scale bar = 50 lm. B, Quantification of fluorescence (axonal length of PC12 cells) from A, ns means "no significant difference"; *P < .05. C, Representative Western blot results of microtubule protein for each group of PC12 cells. D-E, Quantification of Western blot data from C, ***P < .001; **P < .01 14 dpi, an effect different from that seen in the SCI group and SCI + CON group. These data revealed a protectional role of FGF1 in SCI. Based on these results, PRDX1 and autophagy may counter the detrimental effect of ROS to promote recovery of SCI.

| FGF1 regulates the level of autophagy via PRDX1 in vitro
To study the precise relationship between PRDX1 and autophagy, we detected the expression degree of autophagy and PRDX1 after treatment with 3-methyladenine (3-MA) and lentivirus-PRDX1 (LV-PRDX1, carrying siRNA against PRDX1) to inhibit autophagy and PRDX1, respectively. As shown in Figure 8A-F, no significant difference was found in comparison with p-mTOR/mTOR because 3-MA inhibits Akt pathway, but not mTOR. The autophagy-associated proteins, ATG7, Beclin1 and LC3 II/I, were markedly decreased in the 3-MA group, suggesting that the intracellular autophagy was effectively inhibited by 3-MA. Comparison of the PRDX1 levels revealed no significant difference between the control group and the 3-MA group. In addition, as shown in Figure 8A,G-K, after the expression of PRDX1 was apparently reduced by LV-PRDX1, the level of p-mTOR/mTOR was obviously decreased and the expression levels of ATG7, Beclin1 and LC3 II/I were significantly decreased.
Considering these data, we concluded that the down-regulation of PRDX1 may inhibit autophagy. In order to examine whether the upregulation of autophagy by FGF1 was via PRDX1 to affect the mTOR signal pathway, lentivirus-PRDX1 (LV-PRDX1) was used to interrupt the expression of PRDX1 after infecting AAV-FGF1 in PC12 cells. The result showed that the autophagy-associated proteins, ATG7, Beclin1 and LC3 II/I, were significantly decreased in the AAV-FGF1 + LV-PRDX1 group compared to the AAV-FGF1 group after exposure to H 2 O 2 ( Figure 9A-E). Consistent with the previous results, we concluded that the overexpression of FGF1 enhanced the expression of PRDX1 and then increased autophagy to counter ROS to allow recovery of SCI.

| DISCUSSION
According to previous studies, the vital neuroprotective and neuroregenerative effects of FGF1 may be through the PI3K/Akt and MAPK/ERK pathways. [22][23][24] However, there are still a few obstacles According to previous studies, the molecular mechanism of FGF1 to repair damage of SCI is associated with PI3K/Akt, MAPK/ERK and endoplasmic reticulum stress. 30 However, the exact mechanism was undefined. Our previous study showed that activation of autophagy was involved in the restoration of SCI. 20,31 Here, it was first to demonstrate that expression levels of autophagy-associated proteins, such as ATG7, Beclin1 and LC3 II/I, were obviously increased in SCI at 7 days of post-injury due to the release of FGF1 by cells infected by the AAV. Additionally, ROS is a crucial factor in secondary damage of SCI and impacts diverse processes, such as inflammation, apoptosis and autophagy. However, few studies have investigated the influence of autophagy on ROS. We detected a remarkable decrease in ROS in vitro and an increase in PRDX1, an important protein in anti-ROS. These data indicate that FGF1 can regulate the level of ROS and autophagy during SCI, and suggest an interactional relationship.
Previous studies of PRDX1 have been concentrated in the field of cancer, 14,32 with few studies examining the role of PRDX1 in the field of CNS injury, such as SCI. In this present study, the results showed a special relationship between PRDX1 and autophagy. In PC12 cells, we detected autophagy-associated proteins and PRDX1 after using 3-MA and lentivirus-PRDX1 to inhibit autophagy and PRDX1. Interestingly, our results showed that autophagy-associated proteins, such as ATG7, Beclin1 and LC3 II/I, were significantly decreased after the inhibition of PRDX1, but no obvious alteration of PRDX1 was found after the inhibition of autophagy. Additionally, the results showed that the level of autophagy decreased after using lentivirus-PRDX1 (interruption), AAV-FGF1 (overexpression) and H 2 O 2 . Taken together, we conclude that PRDX1 regulation by FGF1 can impact autophagy and exert an efficient anti-ROS effect.
To summarize, the results suggest that the beneficial effect of FGF1 induced is via an increase in PRDX1. The increase in PRDX1 stimulates the up-regulation of autophagy and counters ROS. FGF1 facilitates neuroprotection, axon regeneration and remyelination.
However, there are unanswered questions that need to be resolved in further studies. For example, the exact target and the route for PRDX1 action on the autophagy pathway remain elusive. To conclude, our results identify an effective delivery strategy for FGF1 in SCI therapy and explored the precise mechanism of autophagy and ROS after SCI. This work provides the theoretical basis of therapeutic effect of FGF1 in acute and chronic patients with spinal cord injury.

CONFLI CT OF INTEREST
The authors declare that they have no conflicts of interest.