Sitagliptin improves functional recovery via GLP‐1R‐induced anti‐apoptosis and facilitation of axonal regeneration after spinal cord injury

Abstract Axon growth and neuronal apoptosis are considered to be crucial therapeutic targets against spinal cord injury (SCI). Growing evidences have reported stimulation of glucagon‐like peptide‐1 (GLP‐1)/GLP‐1 receptor (GLP‐1R) signalling axis provides neuroprotection in experimental models of neurodegeneration disease. Endogenous GLP‐1 is rapidly degraded by dipeptidyl peptidase‐IV (DPP4), resulting in blocking of GLP‐1/GLP1R signalling process. Sitagliptin, a highly selective inhibitor of DPP4, has approved to have beneficial effects on diseases in which neurons damaged. However, the roles and the underlying mechanisms of sitagliptin in SCI repairing remain unclear. In this study, we used a rat model of SCI and PC12 cells/primary cortical neurons to explore the mechanism of sitagliptin underlying SCI recovery. We discovered the expression of GLP‐1R decreased in the SCI model. Administration of sitagliptin significantly increased GLP‐1R protein level, alleviated neuronal apoptosis, enhanced axon regeneration and improved functional recovery following SCI. Nevertheless, treatment with exendin9‐39, a GLP‐1R inhibitor, remarkably reversed the protective effect of sitagliptin. Additionally, we detected the AMPK/PGC‐1α signalling pathway was activated by sitagliptin stimulating GLP‐1R. Taken together, sitagliptin may be a potential agent for axon regrowth and locomotor functional repair via GLP‐1R‐induced AMPK/ PGC‐1α signalling pathway after SCI.


| INTRODUC TI ON
Traumatic spinal cord injury (SCI) results in impaired neurologic function, such as severe motor and sensory dysfunction, that for many individuals is permanent and significantly impacts health, function, quality of life and life expectancy. 1 Currently, there is no effective treatment for injured spinal cord, and the mechanisms that underlie recovery after SCI have rarely been definitively established. In a traumatic spinal cord injury, an external force causes immediate damage to neurons and axons at the injury site, resulting in a secondary neuropathological process that further aggravates neuronal damage. 2,3 Neuronal apoptosis is recognized as one of the most crucial secondary neuropathological processes underlying SCI, which eventually leads to significant expansion of the damage. 1 Improving the neuronal intrinsic regenerative ability of central nervous system (CNS) by stabilizing microtubules can promote the axon regeneration and locomotor functional recovery after SCI. 4,5 Therefore, improving axon regrowth and neuronal survival remain the major prerequisite for regeneration and eventually functional recovery.
Glucagon-like peptide-1 (GLP-1) is considered an incretin hormone when secreted by intestinal enteroendocrine L-cell and a neuropeptide when released in the neural cells. 6 GLP-1 works by activating its specific receptor, GLP-1 receptor (GLP-1R), which belongs to the G protein-coupled receptor family with seven-transmembrane spanning. 6 The GLP-1/GLP-1R signalling axis is best known for its effects on glucose homeostasis and facilitation of insulin signalling. 7 GLP-1 and GLP-1R also are found in CNS. Beside the potential ability to regulate diabetes, the stimulation of GLP-1R has been demonstrated to be neurotrophic and neuroprotective against neurodegenerative disease, such as Alzheimer's disease and traumatic injury as well as beneficial for neurogenesis in the CNS. [8][9][10][11] Evidences have supported that the activation of GLP-1R has beneficial effects on neuronal cells survival, proliferation and differentiation, neurite outgrowth. 12 Nevertheless, native GLP-1 is rapidly degraded by the ubiquitously expressed enzyme dipeptidyl peptidase-4 (DPP-4) into the inactive forms which have very low affinity for the classical GLP-1R. 6,17 The enzymatic activity of DPP-4 is a major determinant of the biological activity of GLP-1. Therefore, inhibiting the action of DPP-4 is one way of increasing endogenous GLP-1 levels and stimulating GLP-1R to exert multiple effects in SCI.
Sitagliptin, a selective DPP-4 inhibitor for increasing the half-life of endogenous GLP-1 and stimulating GLP-1 receptor, 18 is approved in more than 130 countries worldwide for the treatment of adult patients with type 2 diabetes (T2D). 19 Sitagliptin is efficacious and safe drug for patients with type 2 diabetes mellitus because of its lower hypoglycaemic effect and low incidence of gastrointestinal complaints and reduced cardiovascular risk. 20,21 Besides its anti-hyperglycaemia effect in T2D, sitagliptin can also exert roles of anti-oxidative stress, anti-inflammation and anti-apoptosis in cardiovascular diseases, 4 neurodegenerative disease and neural injury. [22][23][24][25][26] Recently, increasing attention has been paid to the protective roles of GLP-1R stimulation by GLP-1 analogue or DPP-4 inhibitors in SCI repairing. Actually, activation of GLP-1R with GLP-1 analogue exendin-4 can significantly promote locomotion recovery in rats after SCI through the induction of autophagy and inhibition of apoptosis. 27 Moreover, our previous studies have shown that another GLP-1 analogue liraglutide improved locomotor functional recovery through stimulating autophagy and suppression of apoptosis via GLP-1R after rat SCI. 21 However, the potential neuroprotective benefits of sitagliptin have not been illustrated in SCI. In the present study, we aimed to assess the effects of sitagliptin on neurobehavioral outcome, axon regeneration and neuronal apoptosis in vivo and in vitro after SCI.
We also examined the effects of sitagliptin on GLP-1R. Finally, we explored whether the AMPK/PGC1α signalling pathway was involved in sitagliptin-induced functional recovery following SCI.

| Animal usage and ethic statement
Adult female Sprague Dawley rats (6-8 weeks old, 250-300 g) were obtained from the Animal Center of the Chinese Academy of Science (Shanghai, China). The care and use of all animals is in line with the guidelines set by the National Institutes of Health of China. All the rats were reared under controlled conditions. Operations and K E Y W O R D S AMPK signalling pathway, axon regeneration, glucagon-like peptide-1 receptor, neurite outgrowth, sitagliptin, spinal cord injury F I G U R E 1 Sitagliptin protects neurons against apoptosis after SCI. A-D, Western blot and the quantitative analysis of cleaved caspase-3, Bax and Bcl-2 in rats of each group at 7 d after SCI. GAPDH was used as a loading control and for band density normalization. E-F, TUNEL staining (green) and the quantitative assessment of TUNEL-positive cells from six independent sections from injured spinal cord in each group at 7 dpi, scale bar = 100 μm. G-H, Co-immunofluorescence for cleaved caspase-3 (green) and NeuN (red) and the quantitative assessment of the fluorescence intensity of cleaved caspase-3 in neurons in each group, scale bar = 20 μm. All data are presented as the mean values ± SEM, n = 6 per group. (The 'n' in the whole text means the number (or replication) of samples (rats or cell treatment) in each group). Sham is used as a control. ***P < 0.001, ****P < 0.0001 vs SCI group

| SCI model and drug administration
SCI protocols in adult female Sprague Dawley rats under the aseptic conditions were as described previously. Briefly, rats were sedated with 8% (w/v) chloral hydrate (3.5 mL/kg, ip). Skin along the midline of back was cleaned and incised. Muscles were expanded, and a laminectomy was performed at T9-T10 level. After laminectomy, the spine was stabilized and a moderate contusion on the thoracic spinal cord was performed using MACSIS/NYU Impactor (10 g forces, 3 cm height).
Sham group being used as a control, received no damage after laminectomy. Afterwards, animals were orally treated with sitagliptin (10 mg/ kg, saline) and then daily for up to 28 days after SCI. Another group was administrated with sitagliptin (10 mg/kg/d) and exendin9-39 (1 μg/kg body weight, ip). Sham group was given equal saline. Subsequently, the rats were killed at 1, 3, 5, 7, 14 and 28 days after injury.

| Tissue preparation
Animals were anesthetized by 8% (w/v) chloralic hydras (3.5 mL/ kg, ip) at corresponding time points following SCI. Then, the hearts were perfused with normal saline.

| Western blot
Proteins from animals or primary cortical neurons were first quantified with BCA reagents, and 80 μg proteins were separated on 10% (w/v) gels and transferred onto PVDF membrane (Bio-Rad, Hercules, CA, USA). Membranes were blocked with 5% (w/v) non-fat milk (Bio-Rad)

| Histology and immunofluorescence assay
Sections mounted on slides were prepared as described previously.
Transverse sections for histological analysis were treated by HE staining and Nissl staining following the manufacturer's instructions. Bright field images were acquired using light microscopy. To assess acetylated, Next, the sections were washed with PBST for three times, then incubated with DAPI for 7 minutes, washed with PBST and sealed with a cover slip. For visualization of F-actin, 4 U/mL rhodamine-coupled phalloidin was used (Yeason, Shanghai, China). The images were captured by a confocal fluorescence microscope (Nikon, Japan). Three slides from different rats per group were used for histology and immunofluorescence assay, and five fields of visions were captured from a defined region of interest. And the positive area quantification has F I G U R E 2 Sitagliptin stabilizes microtubule and facilitates axon restoration after SCI. A-J, Representative Western blot analysis and quantitative assessment of Ace-tubulin, Tau, MAP2 and GAP43 expressions at 7 dpi and 14 dpi. K-L, Immunofluorescence staining and quantitative analysis of MAP2 expression in each group. Scale bar = 50 μm. M, Immunofluorescence staining for Ace-tub (green) and GFAP (red) in each group at 14 dpi. Scale bar = 1000 μm; scale bar (enlarged) = 200 μm. All data are presented as the mean value ± SEM, n = 6 per group; significant differences among the different groups are indicated by **P < 0.01, ***P < 0.001, ****P < 0.0001. Sham is used as a control | 8693 been normalized for tissue area analysed. The quantification of each density was performed by the Image J software.

| ATP assay
The ATP-GloBioluminometric Cell Viability Assay (Biotium, Hayward, CA, USA) was used to assess cellular ATP levels according to the manufacturer's protocol. Data were collected from five replicate wells for each experiment.

| The terminal deoxynucleotidyltransferase (TdT) dUTP nick end labelling (TUNEL) assay
To measure apoptotic level, sections (4-5 mm caudal and rostral) were obtained and TUNEL staining was performed at 7 days after SCI following the instructions. All images were captured using a Nikon ECLIPSE Ti microscope (Nikon, Japan). TUNEL-positive cells were counted from 20-30 random sections from three rats to get an average number in each group.

| Locomotion recovery assessment
Basso-Beattie-Bresnahan (BBB) score and footprint analysis were performed to evaluate the restoration of hindlimb locomotor function of rats. In brief, BBB score was conducted at 0, 3, 7, 14 and 28 days. Rats were placed in an open experimental field and allowed to move freely for 5 minutes. The movement of rats hindlimbs was recorded by five trained investigators who were blind to the experimental conditions. Crawling ability was assessed by the BBB score ranging from 0 (complete paralysis) to 21 (normal locomotion). At the time point of 14 days after injury (dpi), the footprint analysis was performed by dipping the animal's posterior limb with red dye and fore limb with blue dye. All of the rats were tested in a confined walkway (about a 10 cm wide by 100 cm long white paper), and then the hindlimb motion trajectories printed on the white runway were collected.

| Statistical analysis
All data were expressed as the means ± standard error of the mean (SEM). Differences between groups in BBB scores were analysed with one-way analysis of variance (ANOVA) followed by Tukey's multiple comparison test. Statistical analyses of other data were performed using one-way ANOVA. All statistical analyses were conducted using statistical software GraphPad Prism version 5 for Windows. Differences were suggested to be statistically significant when P < 0.05.

| Sitagliptin promotes functional recovery and reverses neurological deficit after SCI
To assess the therapeutic role of sitagliptin in SCI repairing, rats were randomly assigned to three groups, including sham group, SCI group and SCI+ sitagliptin group. BBB score as well as footprint analysis were applied to evaluate locomotor recovery after SCI. The results showed that BBB scores of both of SCI surgery group were distinctly decreased compared with sham group, but with no significant difference between the two SCI group at early stage (1, 3, 5 days after injury, dpi). However, compared with SCI group, sitagliptin treatment group exhibited a significant functional improvement at 7, 14, 28 dpi

| Sitagliptin protects against neuronal apoptosis after SCI
To confirm the significance of sitagliptin after SCI and to explore the mechanism of neuroprotection of sitagliptin following SCI, we designed a timeline as shown in Figure S1A. Cell apoptosis occurs at the beginning of the injury and sustains in the entire recovery process. To determine whether sitagliptin has ability to decrease apoptosis caused by acute SCI, Western blot analyses for Bax, Bcl-2 and F I G U R E 3 Sitagliptin reverses SCI-induced mitochondria dysfunction in vivo. A, ATP content was assessed by ATP-GloBioluminometric Cell Viability Assay. (B-C) Immunofluorescence double-labelled staining for co-localization of Drp-1 with Tom20 (red: Drp-1, green: Tom 20, bar: 10 μm). D, The quantitative analysis of fluorescence intensity of Tom20 from (B). E, Mitochondrial membrane potential was detected by Mitotracker and the nuclei were stained with DAPI. (bar = 10 μm). F, The quantitative analysis of Mitotracker fluorescence intensity from (E). G-H, Western blot and the quantitative analysis of the protein levels of Tom20 in PC12 cells treated as indicated. All data represent mean ± SEM (n = 6 per group). ***P < 0.001, ****P < 0.0001 vs SCI group (or H 2 O 2 treated group). Sham is used as a control | 8695 cleaved caspase-3 were carried out at 7 dpi. Compared with the sham group, acute SCI increased the pro-apoptotic proteins Bax and cleaved caspase-3 levels and reduced the anti-apoptotic protein Bcl-2, which were significantly reversed by sitagliptin ( Figure 1A-D). These results were verified by TUNEL staining and immunofluorescence staining.
The number of TUNEL-positive cells apparently rose after SCI, while administration of sitagliptin resulted in a clear fall in apoptotic activity ( Figure 1E-F). Similarly, immunofluorescence staining illustrated a prominent increase of the cleaved caspase-3-positive neurons in the SCI group, but that reduced with sitagliptin administration (Figure 1G-H). Collectively, these data indicate that sitagliptin can inhibit neurons apoptosis after SCI.

| Sitagliptin reverses SCI-induced mitochondria dysfunction in vitro
Previous studies have proved that SCI leads to mitochondrial dysfunction, we therefore explored whether sitagliptin could alleviate the damage of SCI to mitochondria. H 2 O 2 (100 µmol/L/mL) was employed to simulate SCI microenvironment in PC12 cells, and the optimal concentration of sitagliptin (1 mmol/L) was determinate using Cell Counting Kit-8 (CCK8) (Figure S2A-B). The result indicated that H 2 O 2 treatment increased cellular ATP depletion, while sitagliptin treatment partially restored cellular ATP level in H 2 O 2 treated cells ( Figure 3A). Tom20 is a mitochondria outer membrane receptor and Drp-1 mediates mitochondria fission, which are two typical protein markers of mitochondria integrity. Similar to the findings, the results of double-labelled immunofluorescence staining of Drp-1 and Tom20 displayed changes in co-localization ratio. As shown in Figure 3B  (G-H) Co-immunofluorescence staining for GLP-1R (green) with TUJ-1 (red, bar = 20 μm) and quantitative analysis of GLP-1R in PC12 cells in each group. All data are presented as the mean ± SEM, n = 6 per group. Significant differences among the different groups are indicated by **P < 0.01, ***P < 0.001, ****P < 0.0001. Sham is used as a control | 8697

| GLP-1R plays a key role in sitagliptin-induced locomotor functional recovery
GLP-1R has been approved to play an important neuroprotective role in CNS. 10 To confirm whether GLP-1R is key to neuroprotection of sitagliptin following SCI, a selective GLP-1R inhibitor exendin9-39 was ip administrated at a dose of 1 μg/kg body weight once daily.
As shown in Figure 5A

| Sitagliptin-stimulated GLP-1R mediates the enhancement of nerve outgrowth and axon sprout
To determine the effect of GLR-1R stimulated by sitagliptin on neurite growth in vitro, PC12 cells and primary cortical neurons were exposed to H 2 O 2 (1 μmol/L/μL) to simulate SCI microenvironment. As shown in immunofluoresence staining for Ace-tubulin in PC12 cells ( Figure 6A

| Stimulation of GLP-1R by sitagliptin attenuates neuronal apoptosis and activates AMPK/ PGC-1α signalling pathway
To establish whether sitagliptin alleviates neuronal apoptosis by regulating GLP-1R under SCI in vivo and in vitro, rats and PC12 cells treated as above, and the apoptotic activity was detected.

| D ISCUSS I ON
At early stages of injury, rampant neuronal apoptosis arises, which is a key obstacle to the recovery from the stage of secondary injury. F I G U R E 5 GLP-1R plays a key role in sitagliptin-induced locomotor functional recovery. A, The BBB locomotion scores of each group at 1 dpi, 3 dpi, 5 dpi, 7 dpi, 14 dpi and 28 dpi. (B-D) The statistical analysis of BBB scores at 7 dpi, 14 dpi and 28 dpi from (A), n = 12 per group. E, Footprint analysis in the different groups at 14 dpi (blue: forelimbs; red: hind limbs), n = 6 per group. F, Nissl staining of tissue from each group to evaluate the survival of neurons at 28 dpi. G, The quantification of VMN from (F), n = 6 per group. H, HE staining of spinal cord sections from each group at 28 dpi and (I) the quantification of the percentage of necrotic tissue at each interval, n = 6 per group. All data are presented as the mean ± SEM. Significant differences among the different groups are indicated by **P < 0.01, ***P < 0.001, ****P < 0.0001 | 8699 Besides, more and more studies pay attention to the intrinsic axon regeneration capacity. Growing evidences indicate that microtubule assembly and stabilization improves growth cone initiation and axon regeneration after SCI. 4,5,28,29 In the present study, we  14,22,23 Evidences have revealed that sitagliptin acts its roles, such as reducing blood glucose level, anti-inflammation, anti-oxidative stress and anti-apoptosis in a GLP-1/GLP-1R dependent way. 18,[32][33][34] It is generally believed that neuronal apoptosis is a crucial process that is responsible for neurological impairment after SCI. Increasing evidences show that inhibiting neuronal apoptosis is an effective approach to facilitate neural restoration and functional recovery after CNS injury. In this study, we found that sitagliptin improved locomotor functional recovery and reverses neurological deficit following rat SCI. We further used Bcl-2, Bax and cleaved caspase 3 as markers to measure apoptotic activation level after SCI. Bax is released upon initiation of apoptotic process, cleaved caspase 3 mediates cleavage of cellular components, and Bcl-2 prevents apoptosis. 35,36 Our results showed that sitagliptin administration significantly Increased oxidative stress and decreased ATP synthesis cause mitochondrial dysfunction following SCI, 39 which has been suggested to be crucial for secondary injury and subsequent neuronal cell death. 40 In our study, acute SCI gave rise to mitochondria dysfunction, measured as decrease of Tom20 protein and ATP product and induction of Drp-1. However, mitochondria dysfunction caused by SCI was reversed by sitagliptin. Many axon regeneration processes require energy; hence, maintenance of mitochondria function and localization of mitochondria to injured axons both are essential for axon migration and regeneration. 41 As a result, our data reveal that sitagliptin facilitates axon regeneration and neural functional recovery also possibly by alleviation of mitochondria dysfunction.
AMPK, a conserved serine/threonine kinase, when activated, AMPK triggers a switch from ATP-consuming anabolic pathways to ATP-producing catabolic pathways, mediating cellular energy homeostasis. 42,43 In CNS, activation of AMPK is found to reduce apoptosis and confer neuroprotection in ischaemic stroke. 44,45 Furthermore, up-regulation of the AMPK/PGC1α signalling pathway F I G U R E 6 Sitagliptin-stimulated GLP-1R mediates the enhancement of nerve outgrowth and axon sprout. (A and B) Immunofluorescence staining for Ace-tub (green) and the quantitative analysis of axon length in PC12 cells subjected to H 2 O 2 , sitagliptin and exendin9-39 treatment, scale bar = 50 μm, n = 3 per group. (C-F) Representative western blot analysis and quantitative assessment of Ace-tubulin, MAP2 and GAP43 expressions at 7 dpi, n = 6 per group. (G and H) Immunofluorescence staining for Tuj1 and quantification of axon length in primary cortical neurons, scale bar = 100 μm, n = 3 per group. J, Co-immunofluorescence staining for Ace-tub (green) and Tyr-tub (red) in primary cortical neurons in each group after five days culture, scale bar = 50 μm. I, The quantification of the ratio of fluorescence intensity of Ace-tubulin/Tyr-tubulin from (J), n = 3 per group. K, Co-immunofluorescence staining of TUJ-1 (green) and F-actin (red) to identify the growth cones of primary cortical neurons in each group after three days culture. Scale bar = 50 μm; scale bar (enlarge) = 10 μm, n = 3 per group. All data are presented as the mean value ± SEM; significant differences among the different groups are indicated by *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Sham is used as a control | 8701 HAN et Al. leads to neurite outgrowth and axon regeneration. 46 The AMPK/ PGC1α pathway is regulated by many upstream signalling molecular. Several lines of evidence support the involvement of AMPK in mediating the beneficial effects of GLP-1 or GLP-1R. 9,44,47,48 Our results demonstrated that sitagliptin could activate AMPK/PGC1α pathway. Moreover, we detected that increased GLP-1R protein level was accompanied by down-regulation of apoptosis and enhancement of AMPK activity as well as PGC1α protein level when rats treated with sitagliptin after SCI (Figure 7). Thus, these data indicate that GLP-1R stimulated by sitagliptin facilitating neurons survival is involved in up-regulation of the AMPK/PGC1α pathway following SCI.
In conclusion, we have provided a combined evidence to demon-

ACK N OWLED G EM ENTS
This study was partially supported by a research grant from National Natural Science Foundation of China (81722028, 81802251, 81801233), Zhejiang Provincial Natural Science Foundation (LQ18H090008, LQ18H150003, LY19H150001).

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data were used to support the findings of this study are available from the corresponding author upon request.