Human limbs not only have flexible and powerful motor functions, but also have sophisticated sensory functions. While the fingertips perceive information about the surrounding environment through superficial senses such as touchand pressure to issue commands to control limb activities, proprioceptive receptors such as muscles, tendons, and joint pikes and tendons regulate limb position and movement accuracy at all times through position, kinesthetic, and vibration senses [30–31]. However, conventional prostheses do not help amputees to perform precise movement commands, sense their surroundings, and perceive the position, speed, and direction of motion of the prosthesis in the same way as a normal limb [32].
Clite's team established the basic anatomical relationship of the agonist-antagonist myoneural by tandemly coupling the gastrocnemius-anterior tibialis muscle in mice, using an external device to provide a standard motor signal to the efferent nerve to contract the aortic muscle and activate the aortic proprioceptive receptors, while the coupled antagonist muscle diastole activates its own receptors, both of which provide proprioceptive afferent signals via the tibial and common peroneal nerves [33–35]. By re-establishing this basic motor synergy between the agonist-antagonist myoneural pair, the AMI technique uses native tissue receptors to convert prosthetic sensory information related to muscle stretch and tension into neural signals, and uses existing flexible interface technologies and physiologically relevant proprioceptive feedback from natural neural pathways to restore deep sensory functions such as motion position and vibration perception in the limb to provide amputees with real limb sensation [36]. Therefore, in this study, we investigated the effect of proprioceptive rehabilitation in rats after AMI surgery by establishing an effective method for the reconstruction of proprioceptive function in the missing limb using an animal model of limb disability.
To address the loss of limb proprioceptive function due to the disruption of the connection between proprioceptive signals and receptors after amputation, this study used the microscopic nerve anastomosis method to create a rat model of AMI in which the tibial and common peroneal nerves of the left hind limb of rats were disarticulated, and the two disarticulated ends of the nerves were terminated at the ventral part of the transplanted soleus muscle, and the transplanted tendons were microscopically anastomosed, and after a 2-month repair period, electrophysiological methods, gait analysis and expression of TIMP-1 and MMP-1 in the tendon were used to evaluate the postoperative muscle synergistic effects in rats, thus verifying the success of the model. The results showed that the gastrocnemius and tibialis anterior muscles in the AMI group (tendon anastomosis) and the control group (tendon dissection) showed significant atrophy compared with the healthy side (right side), and there was no significant difference in the degree of atrophy between the two groups. The absolute values of sciatic nerve function index, tibial nerve function index and common peroneal nerve function index, as well as the values of gait angle and body angle in the AMI group were significantly lower than those in the control group 2 months after surgery, suggesting that the repair status of injured nerve function and gait balance ability in the AMI group were better than those in the control group.
The proprioceptive is divided into conscious and non-conscious dominance modalities. The conscious proprioceptive pathway is mainly controlled by the integrated motor sensory areas of the flat cortex of the brain [37]. The control of proprioception is divided into high-level management by visual feedback combined with the involvement of the central nervous system in integrated information collection and processing, secondary management by the cerebellum and vestibule. Intermediate management, and primary management, which is low-level motor management in the conditioned reflex motor mode [38]. It is divided into limb position sensation, muscle kinesthesia and weight-bearing sensation [39]. In contrast, patients with lower limb amputation have a major loss of sensory information related to processing limb position and spatial movement due to the absence of limb muscle and tendon tissue or the loss of muscle and tendon proprioceptive receptors [40]. The use of lower limb prostheses requires higher requirements for trunk balance, postural adjustment, and motor control, and the basic control of the lower limb motor system needs to satisfy at least five points: (1) vertical support by gravity; (2) balance of the center of gravity; (3) postural stability; (4) control of the foot trajectory; (5) reduced speed of information transmission to higher centers [41–45].
First, the sarcomere tissue distributed in the skeletal muscles of the limb mainly encodes the analysis of muscle length signals and muscle extension changes, and this fast-conducting afferent feedback plays a key role in postural control regulation by regulating joint mobility in response to extension reflexes, etc. The tendon shuttle structures at the ends of the tendons, on the other hand, perceive changes in limb weight-bearing to assist in pedestrian motor load domination to move the limb forward, and the afferent input of kinesthetic information is significant for limb movement phase and pattern switching [46–48]. This pattern of muscle movement regulation that provides information to the higher nervous system about the relative position and movement of relevant muscles and joints is very delicate, and the loss of receptors and effectors caused by the absence of tissues such as muscles and joints in prosthetic patients directly disrupts the reflex loop [49]. Second, walking is not an unconscious process, but requires both "low-level" control of muscle and tendon shuttle receptors and "high-level" cognitive control, which is even more difficult for prosthetic users in terms of motor executive function and attentional control under walking conditions [50].
The walking swing duration is the maintenance time before the hindfoot touches the ground, and usually the faster the walking speed, the shorter the swing duration, while pain and lower limb motor function limitation caused by muscle and joint diseases can cause the swing duration to be prolonged, and the lack of kinesthesia and position sense will further prolong the swing duration before the hindfoot touches the ground [51]. The hindfoot starts to contact the ground after the swing phase, and the braking duration is from this time until the stage of maximum contact area between the hindfoot and the ground, when the limb is in the stage of deceleration and control of standing posture, and the prolongation of this stage may indicate that the body needs longer time to precisely distribute and control the standing load to ensure the balance of the limb. The propulsion duration, which is the time required to continue the forward motion, is from the maximum contact area between the hindfoot and the ground to the next. The duration of propulsion, the time required to continue forward motion, is the time between the maximum contact area of the hind foot with the ground and the next swing phase, which is another accelerated motion, and the shorter time may indicate greater trunk strength and control [52]. The stance duration is the whole process of hindfoot-ground contact and is generally positively correlated with walking speed. There are many influencing factors related to hindfoot-ground contact intensity, and rat body weight, plantar area, ground friction, locomotor status, and limb control may increase or decrease contact intensity [53]. The data from this study showed that the walking swing duration, braking duration, propulsion duration, and stance duration were significantly shorter in rats undergoing AMI surgery than in rats with tendon dissection after 2 months, and this result suggests that the intact neurofeedback loop established by AMI is effective in repairing the motor control of the lower limb in rats.
The stimulation-evoked electrical signal experiment showed that the CMAP signal of the peroneal nerve innervated graft (agonist muscle) was gradually enhanced with increased stimulation of the common peroneal nerve (agonist nerve) in the rats given AMI. At the same time, the CMAP signal of the tibial nerve innervated graft (antagonist muscle) was enhanced, and the CNAP signal of the tibial nerve (antagonist nerve) was increased. The results indicated that the co-motor unit of muscle-tendon pair-nerve could form a closed loop to complete the feedback of afferent and efferent signals, and the antagonist muscle and antagonist nerve of the control group with severed tendon had no obvious feedback effect on the stimulation signal. The motor unit of the common peroneal nerve conduction in the AMI group was significantly higher than that in the control group, indicating that the synergistic effect of the agonist and antagonist muscles of the lower limb was good after AMI surgery, and the motor function of the injured nerve was recovered significantly.
MMP-1 interacts with TIMP-1 to maintain the relative stability of tendon collagen, and together they promote extracellular interstitial collagen renewal, regulate tendon metabolism, and participate in the tendon remodeling process [54–55]. However, in this study, MMP-1 and TIMP-1 were lower in the control group than in the AMI group, and there was no significant difference between MMP-1/TIMP-1 and AMI group, which may be due to the fact that the dynamic regulation of MMP-1/TIMP-1 is more obvious in the early tendon repair. After 2 months postoperatively, the tendon had entered a chronic repair phase and MMP-1/TIMP-1 regulation did not continue to function. Therefore, later experiments could intermittently extract protein to observe the tendon repair process. Histological staining showed significant atrophy of both tendon and muscle tissue in the control group, with disrupted tissue structure leading to loss of function, whereas tendon healing in the AMI group was as expected, with intact collagen fiber structure in the AMI group as seen by Masson staining. This result indicated that this surgical approach was anatomically and morphologically feasible to reconstruct the basic motor relationship of the agonist-antagonist muscle.