Effects Of Locomotor Training On Plasticity Of Excitatory Interneurons Located In CPG In Rats With Spinal Cord Injury

Background and objectsSpinal cord injury (SCI),which with high disability, refers to as the damage of spinal cord function (motor, sensory, sphincter and autonomic nerve function) under the injury level caused by a variety of pathogenic factors(trauma,inflammation,tumor,etc). Spinal cord injury has a significant impact on quality of life,and over 50% of the spinal cord injury patient suffer from the dysfunction of both lower extremities. There are no fully restorative therapies for spinal cord injury as yet,while locomotor training is considered as an effective therapeutic interventions enhancing the recovery of motor function and it’s regareded as a conventional clinical rehabilitation measures. Among numerous experimental studies,scientists suggest that locomotor training promote the animal models’ hind limbs motor function through the improvement of blood supply in spinal cord injury site, relieving of inflammatory reaction and the expression of neurotrophic factors.Furhermore, this improvement occurs because,after spinal cord injury, the spinal circuitry below the lesion site does not become silent but maintains active and functional neuronal properties, and can respond to peripheral input from below the level of the injury.The basic motor patterns underlying rhythmic limb movements during locomotion are generated by the intrinsic interneuronal networks located within the spinal cord. These n6tworks are called Central Pattern Generators (CPGs) and it’s longitudinal distribution extend from the lower thoracic (Thll-Thl3) spinal cord and throughout the lumbar region (L1-L6):the same region that contains the motor neurons controlling leg muscles. These networks are composed of several types of excitatory and inhibitory interneurons which innervated with each other. The initiation of locomotion is mediated by the neurons in the mesencephalic locomotor region (MLR) that project to neurons in the medial reticular formation (RF) in the lower brainstem. Neurons in the medial reticular formation project to the locomotor CPG in the spinal cord that executes locomotion.Training-dependent plasticity is one of the most important characteristics for CPG,that is,even in the absence of supraspinal descending signal after a complete spinal cord injury,the potential for the spinal circuitry to generate oscillating coordinated motor patterns remains, and peripheral input continues to flow into the spinal cord and be processed, although in a modified manner. While receiveing persistent and rhythmic stimulation,spinal automaticity in than comprised of CPG circuitry become stronger and the excitatory and inhibitory interneurons become activation combined with sensory input from the periphery. After all,CPG generate both the rhythm as well as the correct patterns of activities.The mainly excitatory interneurons located in CPG region are divided into three types according to the production of different neurotransmitters.The excitatory neurotransmitters include glutamate, noradrenaline and dopamine,etc.These subpopulation of excitatory interneurons play indispensable role in the self-initiated activation of CPQas well as they maintain and coordinate the rhythm-generation and pattern-generation in locomotion.Until now,we can not identify the most important excitatory interneurons in CPG region due to lack of the research compare these three types neurons at the same time.Combining non-invasive positron emission tomography (PET) with [18F]fluoro-2-deoxy-D-glucose (18F-FDG), a widely used PET tracer, has allowed in vivo imaging of specific biological pathways, such as increased glucose utilization in tumour cells, the high uptake of inflammatory cellular elements and blood supply in the clinic. In laboratory studies, 18F-FDG micro-PET-CT (computed tomography) imaging, due to its inherent imaging characteristics, has allowed scientists to investigate whole body metabolic activity and acquire images reflecting quantitative metabolic information in regions of interests (ROIs) in small-animal models of different diseases.Since spinal cord is one part of the central nerve system,it gets the vast majority of its energy from glucose.In the mean time,several fundamental researchs observe the change of glucose metabolism in part of spinal cord using micro-PET/CT. In this research,we adopted micro-PET/CT to detect the metabolism change in lumbar spinal cord and figure out the activated state of CPG after spinal cord injury and locomotor training.In this research,we used T10 segment imcomplete spinal cord injury rat model and the models received 6 weeks treadmill locomotor training.The evaluation indicators include BBB score, number and the morphological change of motor end plate in musculus gastrocnemius, metabolism change in the lumbar segment and quantitative variance of 3 subpopulation of intemeurons.The object of this research is to observe the effect of locomotor training on CPG on the aspects of ethology、 histology、 immunochemistry, metabolism and explore the dominant interneuron thereby providing a new way to improve the motor function recovery in patients with spinal cord injury.MethodsI.Objects:AIl 45 Sprague Dawley rats were randomly divided into three groups,trained group(n=15),untrained group(n=15) and sham group(n=15)(just remove the vertebral plate). Moderate T10 segment imcomplete spinal cord injury was executed using NYU method.2.Methods:After the model group was made, the locomotor training group was intervened according to the preset program,control group and sham operation group were fed without any intervention. BBB score were observed periodically, micro-PET/CT scanning、 immunofluorescence staining、nissl staining and acetylcholinesterase staining were performed postoperative 7 weeks.(1) Interventions:Treadmill training conducted lweek after surgery and last for 6 weeks. Training parameters:5 days per week,30 minutes per day, break for 5 minutes. Walking speed increased gradually from 5m/min to 15m/min.Monitored SD rats’mental states and food/water intake.(2) Taking materials:Fresh specimens and slices of gastrocnemius muscle: removed the rat right triceps 7 weeks postoperative and placed in-80 ℃ for 1h.Then performed 10mm routine frozen serial sections, extract 1 specimen from every 5 pieces; rat perfusion, fresh specimens and slices of spinal cord:7 weeks after surgery we infused rat with saline and 4% poly formaldehyde through left ventricle until trunk and limbs of rats appeared white and stiff. Located the lumbar 1-2 segment of the rat’s spinal cord and taked it out. Fixed, dehydrated and embedded tissue according to conventional method. Performed 10mm and 20mm coronal position frozen serial sections, extract 1 specimen from every 5 pieces.Placed all the sections in-80℃.(3) Observation index and methods1)Behavioral evaluation:BBB score was evaluated before and 7、14、21、28、 35、42 days after surgery using double-blind method.2)CPG metabolism level.The 18F-FDG micro-PET/CT examination was performed right after the treadmill training(n=5, in each group).Weighting SD rats, the 18F-FDG (37kBq (1 μCi)/g) was injected via the tail vein.Forty minutes following the 18F-FDG injection,the rats were fixed in a prone position and scanned the whole spinal cord with a small-animal PET/CT scanner using Inveon software. Co-registration of the PET images was used to ensure consistency in anatomic localization.In horizontal, sagittal and coronal scan,three-dimensional ROIs were measured.By using Inveon research workplace 4.1 software, the 18F-FDG absorption rate was expressed as the percent of the injected dose per gram of body weight (%ID/g).Compare the different nuclide absorption among three groups.3)Counting of total neurons in L1-2:Nissle staining,dehydration and vitrification of lumbar spinal cord cryo section in each group. Randomly selected five views under 100 X microscopy, count the total number of interneurons located in lumbar spinal lamina VII,VIII and X using Image Pro Plus6.0.4) Counting and expression of 3 types of interneurons in LI-2: Immunofluorescent stainingCode the cryosection A、B、C according to the different target interneurons in each group.①Glutamatergic interneurons:Using number A cryosection. Place at room temperature for 1h. Using 0.3%Triton X-100 to punch the cell membrane for 30min and than sections were blocked using 10% goat serum for 1h.Respectively, added the mouse anti NeuN antibody (1:100) and rabbit anti glutamine synthtase antibody (1:100),4 ℃ for the night.After rinse every load slides, added 200ul of mixed goat anti-rabbit-DyLight 488(1:200) and goat anti-mouse-DyLight 594(1:200) and incubated in 37℃ for lh.Washed each slide and added 200ul DAPI for 5min.Mount preparations.② Noradrenergic interneurons:Using number B cryosection.The immunofluorescent staining following the same steps mentioned above. The primary antibody include:mouse anti NeuN antibody (1:100) and rabbit anti dopamine-β-hydroxylase (1:100)③ Dopaminergic interneurons:Using number C cryosection.The immunofluorescent staining following the same steps mentioned above.The primary antibody include:mouse anti NeuN antibody (1:100) and rabbit anti tyrosine hydroxylase (1:100)④ Shooting photos using confocal laser scanning microscopy. Randomly selected five views under 100×microscopy, count the total number of interneurons located in lumbar spinal lamina VII,VIII and X. Randomly selected five views under 600 X microscopy, count the gray value as the mean fluorescence intensity using Image Pro Plus6.0.5) Expression of motor end plate in musculus gastrocnemius:After fresh frozen section of the gastrocnemius muscle of rats, the slices were incubated in the solution in 37℃ for 1.5h until the color of the slices was changed to light brown. Take out slices, distilled water washed for 5min. Hematoxylin reagent stained at room temperature for l-2min staining. Immediately rinse with distilled water when nuclei were pale purple. Neutral gum sealing piece.Randomly selected five views under 100 X microscopy, count the total number of motor end plate and the average optical density using Image Pro Plus6.0.7. Statistical analysis:The statistics program used for the analysis in this study was SPSS 19.0 software (Chicago Inc., IL, USA). Data were expressed as mean ± standard deviations.Tests of normality and homoscedasticity were performed to examine if parameters were normally distributed. Results were evaluated by analysis of variance (One-way ANOVA) (followed by post hoc LSD or Dunnett’T3 test). The relationship between bivariate was determined by the correlation coefficient method according to the spearman correlation analysis. The relationship between BBB score and multivariate was calculated using multiple linear regression analysis.A p value of <0.05 was considered to be statistically different.Results1. BBB scoreThe SD rats in sham group can perform a regular gait right after anesthesia recorvery.One day after surgery, the rats’ motor function recovery as preoperative states and the BBB score are 20-21.At the early stage after spinal cord injury, spinal cord injury rats suffered from severe hindlimbs dysfuntion and the BBB score is 0-1 points.2 weeks after spinal cord injury,the SD rats in trained group began to show the muscle contraction and hip, knee and ankle joint movement intermittently while the BBB score is 5.67±1.03 which is significant higer than untrained group(T<0.05).In untrained group,the rats contract muscles and move hip or knee joint occasionally and the BBB score is 2.83±0.75.Over time, the recovery degree of hindlimbs motor function in trained group is higher than rats in untrained group at each time points.The BBB score of trained group at each time point are all significant higher than untrained group(P<0.05).7 weeks after injury,rats in trained group show the weight-bearing locomotion with coordinated upper and lower limbs movements and the BBB score is 16.08±0.80 which is significant higher than untrained group(14.08±0.66).2. CPG metabolism levelFifty days after spinal cord injury,the results of the micro-PET/CT imaging showed an increased focal 18F-FDG uptake in the lumbar ROI in trained group (0.284±0.053%ID/g),but the similar accumulation was not found in the lumbar ROI in sham group (0.139±0.026%ID/g) and untrained group (0.145 ±0.081%ID/g) (P<0.05). And we found no significant difference between sham group and untrained group (P>0.05)Correlation analysis:Correlation analysis between the average uptake value of 18F-FDG accumulated in lumbar segment and BBB score in untrained group and trained group.The result showed that there was a significant correlation between the average uptake value of 18F-FDG accumulated in lumbar segment and BBB score according to different intervention methods.It suggest that on account of locomotor training, SCI rats’hindlimb showed motor function’s improvement due to the enhancing of metabolic activity in lumbar segment.3. Counting of total neurons in L1-2Counting result:The number of interneurons located in lumbar lamina VII,VIII and X in untrained group (94.2±9.86) is significantly reduced compared to sham group (123.2±11.00) (P<0.05).While treadmill locomotor training can increase the interneurons’numbers(131.2±10.03) and is significant higher than untrained group (P<0.05)4. Counting and expression of 3 types of interneurons in L1-2Glutamatergic interneurons:Counting:In sham group,the number of glutamatergic INs is 118.67±2.30,after spinal cord injury,the number of glutamatergic INs is 118.67±4.21 and has no significant difference compared with sham group (P>0.05).The number of glutamatergic INs in trained group is 135.80 ±6.46 and is significant higher than sham group and untrained group (P<0.05)。 MFI:In sham group,the MFI is 0.0780±0.0172, after spinal cord injury,the MFI is increase to 0.1975±0.0193 and is significant higher than sham group (P<0.05).6 weeks after training.the MFI is more higher than untrained group,the quantitative value is 0.3559±0.0027 (P<0.05)Noradrenergic interneurons:In sham group,the number of noradrenergic INs is 83.80±4.71,after spinal cord injury.the number of noradrenergic INs is 83.60±5.22 and has no significant difference compared with sham group(P>0.05).The number of noradrenergic INs in trained group is 89.60±3.21 and has no significant difference compared with sham group and untrained group (P>0.05)。 MFI:In sham group,the MFI is 0.1120±0.0236, after spinal cord injury,the MFI is increase to 0.1588± 0.0114 and is significant higher than sham group (P<0.05).6 weeks after training,the MFI is more higher than untrained group,the quantitative value is 0.3172±0.0165 (P<0.05)Dopaminergic interneurons:In sham group,the number of dopaminergic INs is 81.00±5.34,after spinal cord injury,the number of dopaminergic INs is 80.20±4.66 and has no significant difference compared with sham group (P>0.05).The number of dopaminergic INs in trained group is 86.20±3.27 and has no significant difference compared with sham group and untrained group (P>0.05)o MFI:In sham group,the MFI is 0.0564±0.0060, after spinal cord injury,the MFI is increase to 0.1466± 0.0296 and is significant higher than sham group (P<0.05).6 weeks after training,the MFI is more higher than untrained group,the quantitative value is 0.2134±0.0257 (P<0.05)5. The correlation between the expression of glutamatergic, noradrenergic, dopaminergic neurons in the lumbar region and the BBB scoreRegard the count and MFI of 3 subpopulations of INs as independent variables, and the BBB score as dependent variable and we started the stepwise linear regression analysis.The result showed that equation was statistically significant,the R square is 0.931,the adjusted R square is 0.922, glutamatergic INs number can count into the equation.The number of glutamatergic INs show positive correlation with BBB score, namely the more count of glutamatergic INs,the higher BBB Score we may see.6.Counting and expression of motor end plate in musculus gastrocnemiusThe gastrocnemius motor end plate in sham group was dark brown, oval shaped, regular shape, and arranged more in a curved line beside 1/3 outer of the long axis of the gastrocnemius.The shape,coloring degree and the arrangement of motor end plate in trained group were similar as sham group,and the differences between the total number of motor end plate in trained group and sham group were not significant (P>0.05). In the control group, the motor end plate of the gastrocnemius muscle of the rats was relatively light brown, and the center was almost transparent, but it was oval in shape and small in size.While the location is similar to sham group,but in control group the density is much lower,and the count was significant reduced compared to sham group and trained group(P<0.05).Analysis of the average optical density showed that staining degree of motor end plate in trained group is similar with sham group(P>0.05),both of which were significant higher than the coloring degree of motor end plate in untrained group(P<0.05).Correlation analysis:Correlation analysis between the number of MEP and BBB score and also the analysis between the AGD of MEP and BBB score in untrained group and trained group.The result showed that there was a significant correlation between the number of MEP and BBB score according to different intervention methods.And there was also a significant correlation between between the AOD of MEP and BBB score.It suggest that on account of locomotor training, SCI rats’ hindlimb showed motor function’s improvement due to the increasing of number and AOD of MEP.Conclusiou1. Treadmill locomotor training can effectively promote the recovery of the motor ability of the hind limbs of SCI rats,promote the expression of glutamatergic, noradrenergic,dopaminergic excitatory interneurons in CPQincrease level of CPG metabolism and maintain motor endplate function.2. Glutamatergic interneurons is the main excitatory interneurons participate in the formation of locomotor-rhythm.3. Locomotor training improve the motor function of the hind limbs in SCI rats through enhancing the plasticity of the glutamatergic, noradrenergic,dopaminergic excitatory interneurons and increasing the number and the expression of the motor end plates of the gastrocnemius muscle.