Calycosin-7- O -β-D-glucoside treatment promotes axonal 1 regeneration via Rho/ROCK pathway after ischemia/reperfusion 2 injury 3

6 Background: As a medical component in Astragalus(AR) ， Calycosin-7- O -β-D-glucoside 7 （ CG ） defends ischemia/reperfusion(I/R) injury in cerebral ischemia due to its anti- 8 oxidative and anti-inflammatory effects. However, whether CG can facilitate I/R injury by 9 stimulating neuroregeneration and its specific mechanism is remained to be elucidated. 10 Methods: In this study, an animal model of ischemic stroke was established by middle 11 cerebral artery occlusion (MCAO). Seven days after CG, triphenyltetrazolium chloride (TTC) 12 staining was performed to examine the ischemic volume, accompanied by behavioral tests to 13 assess neurological function. Nissl staining and Bielschowsky’s silver staining were used to 14 observe nerve cell damage and axonal loss, while immunofluorescence was used to evaluate 15 axonal regeneration. 16 Results: The expression of proteins associated with the Rho/ROCK pathway was detected by 17 using western blot (WB) and quantitative real-time polymerase chain reaction (qRT-PCR). 18 We showed that CG significantly reduced ischemic volume, facilitated axonal regeneration, 19 improved neurological function, and regulated expression of RGMa, Rho, ROCK, and 20 CRMP2. Conclusions: Our results suggested that CG promotes axonal regeneration by limiting 22 activation of the Rho/ROCK pathway to promote recovery after cerebral ischemia.


26
Ischemic stroke (IS) is a rapidly-developing brain disorder that accounts for the vast majority 27 of strokes [1]. In the early stages of stroke, brain damage is often accompanied by impaired 28 axonal atrophy resulting in neurotransmission defects. Therefore, stroke patients commonly 29 have disabilities that involve severe language, cognitive, and physical dysfunctions [2]. 30 Exercise rehabilitation training for IS patients or animal models can fix the nerve fibers that 31 control muscle movements, suggesting that promoting prolongation of atrophic axons can 32 significantly improve recovery of neurological function [3]. 33 2 The Ras homolog gene /Rho-associated protein kinase (Rho/ROCK) pathway is involved in 34 maintaining microtubule and cytoskeletal stability, which is important for growth cone 35 formation during axonal regeneration [4]. Surface receptors of the growth cone can recognize 36 various growth signals received in the neurons and guide the axon to extend in the correct 37 direction. Repulsive guidance molecule a (RGMa) is a potent inhibitor of axonal regeneration 38 that is expressed in the developing and mature central nervous system. After a stroke, up-39 regulated RGMa binds to the specific surface receptor on the growth cone and activates the 40 Rho/ROCK pathway, eventually destabilizing the cytoskeleton and causing the growth cone 41 to collapse [5]. Inhibiting RGMa or treatment with ROCK inhibitors can stimulate recovery 42 of motor function, which is possibly due to the promotion of axonal regeneration [6,7]. 43 Therefore, the inhibition of axonal regeneration is dependent on activation of this signaling 44 pathway. 45 As a basic herb in traditional Chinese medicine, AR has been proven to have various 46 bioactivities. Such as anti-inflammatory 、 anti-oxidative 、 immunoregulatory and certain 47 neuroregeneration effect et.al [8，9，10].. CG is a compound of Flavonoids [11], which is 48 mainly exist on Leguminosae sp. , as a medicinal ingredient of AR. Our previous researches 49 had confirmed that CG can reduce oxidative stress and neuronal apoptosis [12], and improved 50 expression of GAP43 in PC12 cell oxygen-glucose deprivation/reperfusion (OGD/R) model, 51 suggesting that CG can promote axonal regeneration. [13].Nevertheless, how CG exert 52 protection for I/R injury is unclear. This shows CG is a promising candidate in stimulating 53 neuroregeneration for I/R injury treatment. 54 In this study, we established an animal model of middle cerebral artery occlusion (MCAO) to 55 investigate the effects and mechanism of CG on improving axonal regeneration and 56 functional outcomes after I/R injury. We demonstrated that CG promoted axonal regeneration 57 and neurological recovery after IS by inhibiting Rho/ROCK pathway. Our study provides 58 insight into the mechanism for promoting stroke rehabilitation and empirical evidence for its 59 use in the clinic. 60 for each group according to a method published by Garcia[16]. The final score was the total 115 score of the six evaluation items, with a maximum score of 18 points, and a minimum score 116 of 3 points. A lower score is associated with more severe neurological damage. 117

Materials and Methods
For balance beam test (BBT), was aming to asses motor balance and coordination. The 118 narrow beam (80cm long wooden beam, 1.5cm wide ) was used for the experiment. The 119 beam was placed 15cm above the soft platform. Rats should be trained every day, enabling 120 them to pass the beam smoothly before making the model. Total score of BBT was 6-121 point [17], the higher the final score, the more severe the damge represented. 122 2.4. TTC staining 123 The rats were anesthetized and sacrificed, then brains were quickly removed and frozen at -124 20°C for 30 min. Each brain tissue was cut into 6 pieces which were then immersed in 2% 125 TTC solution in the dark at 37°C for 30 min. After the brain slices were removed from TTC 126 and washed with phosphate buffered saline (PBS, pH 7.4), the tail side of each slice was 127 taken for image analysis using Image J software (Rawak Software, Inc. Germany). The total 128 infarct volume and the infarct rate were calculated as described in a previous study [18], and 129 the formula for calculating the infarct rate is: (volume of non-infarcted hemispherevolume 130 of non-infarcted volume in infarcted hemisphere) / volume of non-infarcted hemisphere × 131 100% 132 133 2.5．Brain water content 134 After the last administration, the brain was quickly removed and was made into a coronal 135 section, each about 3 mm thick, and immediately weighed with an electronic balance to 136 obtain the wet weight of the brain. The brain slices were placed on the baked tin foil, dried in 137 an oven at 102°C for 12 h, and weighed to obtain the brain stem weight. The calculation 138 formula of brain water content is as follows: (wet weight-dry weight )/wet weight×100% 139 140 2.6. Pathological assessment 141 After 7 days of dosing, brain tissue and nerve cell damage were evaluated by HE staining and 142 Nissl staining. Briefly, the animals were anesthetized and perfused with cold saline followed 143 by 4% paraformaldehyde. The brain tissues were then quickly removed and frozen. loss of a few lesions containing less than 25% of tissue, 2 = deep axonal loss of lesions 156 including more than 25% of tissues, 3 = diffuse and extensive axonal loss including more 157 than 50% of tissues . 158

Immunofluorescence (IF) 159
Immunofluorescence staining of NF-200, MAP2 and GAP43 was used to assess 160 the growth of neuritis; GFAP and CSPGs to analyse the inhibition of axons. Frozen sections 161 were washed with PBS, followed by incubation with 0.30% Triton X-100. The brain sections 162 were treated with the following steps: washed by PBS, blocked with 10% goat serum, and 163 incubated Tween-20 (TBST) containing 5% skim milk, the membranes were incubated overnight at 4°C 178 with RGMa (1:15000), Rho (1:2500) and CRMP2 (1:5000) primary antibodies. β-actin 179 (1:1000) was used as the loading control. On the following day, the membranes were 180 incubated with IgG-HRP goat anti-rabbit secondary antibody (1:5000). Proteins on 181 membranes were visualized using an ECL reagent kit and a developer (Tanon, China). The 182 density of each band was analysed by Image J software. 183 2.10. Quantitative real-time polymerase chain reaction (qRT-PCR) 184 After 7 days of CG administration, total mRNA was extracted from brain tissues with TRIzol 185 and cDNA was generated with Glodenstar TM RT6 cDNA Synthesis Mix Kit. The qRT-PCR 186 reaction was performed using a 2× T5 Fast qPCR Mix (SYBR Green I) Kit. mRNA 187 expression levels were analysed using the relative quantification method (2 -△△ Ct ) with β-actin 188 as an internal control. Primer sequences are shown in Table 1. 189 190 (ANOVA) was used to determine significance between more than two groups of data, while 196 the student's t-test was used between two groups. The results were displayed as means ± 197 standard deviation (SD), and a P value less than 0.05 was considered statistically significant. 198 Brain edema was also remarkably reduced after 7 days of CG and HF administration. The 218

Results and Discussion
best anti-I/R effect was shown in the CG-H group with the minimum infarction rate 219 (15.63±1.63%) and cerebral water content (77.69±1.65%). The results demonstrated that CG 220 was protective against cerebral ischemic injury after stroke.

CG facilitates axonal regeneration after I/R injury 278
IF was used to observe the expression of proteins involved in axonal structure, 279 MAP2 and GAP43, reflexes of axonal regeneration. As shown in Figure 5B, expression was 280 easily detected in the sham group, but barely detectable in the I/R model group, suggesting 281 the axonal structure was seriously damaged after I/R injury. Protein expression was 282 significantly increased after 7 days of CG administration (Figure5B, ***P<0.001) in a dose-283 dependent manner. The results of both BSSM and IF analyses suggest that CG promotes 284 axonal regeneration and remodeling during the I/R recovery process.

CG promotes axonal regeneration via Rho/ROCK signaling pathway after I/R 303 injury 304
Our results show that CG can improve recovery after neurological damage caused by IS, 305 where axonal regeneration plays a vital role during this process. To address whether the 306 Rho/ROCK signaling pathway participates in this process, WB and qRT-PCR were carried 307 out to detect the expression of RGMa, Rho, ROCK, MLC2 and CRMP2. Semi-quantitative 308 analysis of WB (Figure7A-E) showed that protein expression of RGMa, Rho and ROCK 309 decreased dramatically in the CG and HF group compared with the I/R group, so do the 310 phosphorylation of MLC2 and CRMP2. In addition, a markedly reduction of RGMa, Rho, 311 ROCK, MLC2 and CRMP2 mRNA levels was presented in CG and HF group in contrast to 312 the I/R group ( Figure 7F-J). Taking the results of WB and qRT-PCR together, it reveals that 313 CG promoted axonal regeneration partly by inhibiting the Rho/ROCK signaling pathway 314 after IS. 315

320
In this study, we described the mechanism behind how CG can reduce cerebral infarction 321 volume and promote recovery of neurological function in IS rats. Necrosis cannot be reversed 322 in the central infarct area, even after a short period of time post-cerebral infarction [22]. 323 However, ischemic tissue surrounding the infarct centre, known as the ischemic penumbra 324 (IP), is short of electrical activity, but can still maintain normal transmembrane potential and 325 voltage. If the supply of blood to the brain can be repaired in time, IP can be recovered to 326 prevent further expansion of the infarct [23]. Therefore, timely and effective neuroprotection 327 of the IP is now the main method of ischemic stroke treatment [24]. We suggest that CG 328  Our study showed that CG downregulated the RGMa and Rho expression, but upregulated 381 expression of CRMP2 in MCAO rats. Therefore, we hypothesized that inhibition of the 382 Rho/ROCK pathway may be the potential underlying mechanism for CG-induced axonal 383 regeneration and improved neurological outcomes. 384

385
Our study indicated that CG can promote axonal regeneration and neural functional recovery 386 after the cerebra ischemia. In rats treated by CG, behavioral scores, nerve fiber fracture, and 387 axonal degeneration were significantly improved. Expression of NF-200, MAP2, and GAP43 388 were significantly upregulated. These beneficial mechanisms in neurons are dependent on the 389 Rho/ROCK signaling pathway inhibition after ischemic stroke. These findings extend our 390 understanding of CG in terms of neural regeneration and provides further evidence for 391 efficacy as a therapeutic treatment for ischemic stroke. 392 Availability of data and materials 393 The datasets used and/or analysed during the current study are available from the 394 corresponding author on reasonable request 395