Effects of controlling anterior tibial translation preserve sensory-motor function and mechanoreceptor morphology following anterior cruciate ligament injury in rats Authors: Shion Masuda, Momoko Nagai-Tanima, Asuka Kawaguchi, Kazuhiro Hayashi, Tomoki Aoyama Published: 2026 ABSTRACT AbstractAnterior cruciate ligament (ACL) injuries impair mechanical stability and proprioceptive signaling. Although conventional reconstruction techniques aim to restore joint stability, they often provide limited support for sensory recovery. We investigated whether controlling anterior tibial translation using an intra-articular suture-based stabilization model analogous to external bracing (controlled anterior translation of the tibia; CATT) could preserve sensorimotor function and mechanoreceptor morphology following ACL injury. Male Wistar rats were randomly assigned to sham, ACL transection (ACLT), and CATT groups. Sensorimotor performance was assessed using a balance beam test, and compound muscle action potentials were recorded to evaluate peripheral nerve function. Histological and immunofluorescence analyses were performed to quantify mechanoreceptors at the ACL attachment sites. The ACLT group exhibited impaired motor coordination and an increased number of atypical mechanoreceptors. By contrast, the CATT group demonstrated partial restoration of ligament continuity, improved sensorimotor outcomes, and preservation of typical mechanoreceptors. The significant correlation between mechanoreceptor count and behavioral performance supports the role of mechanoreceptor morphology in functional recovery. Anterior tibial translation control may help preserve the mechanoreceptor structure and sensorimotor coordination following ACL injury, providing a potential foundation for developing function-focused conservative strategies beyond structural repair.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1038/s41598-026-50092-2.Keywords: Anterior cruciate ligament injury, Sensorimotor function, Mechanoreceptors, Proprioception, Conservative treatment, Rat modelSubject terms: Musculoskeletal system, Ligaments Introduction The anterior cruciate ligament (ACL) plays a critical role in maintaining knee-joint stability by restricting anterior tibial translation and controlling rotational movements1–3. In addition to its mechanical function, the ACL is the central component of the proprioceptive system4. Proprioception, which encompasses joint position sensing and kinesthesia, is essential for postural control, neuromuscular coordination, and injury prevention. The ACL contains several types of mechanoreceptors5, including Ruffini endings, Pacinian corpuscles, and Golgi tendon organ-like receptors. These mechanoreceptors are primarily located in the sub-synovial layer near its femoral and tibial attachments6–8. These receptors detect the mechanical deformation of the ligament and transmit afferent signals to the central nervous system. ACL injury disrupts sensory pathways, resulting in impaired proprioceptive feedback, reduced neuromuscular control, increased joint instability, and an elevated risk of secondary complications, including reinjury and early-onset osteoarthritis9. Previous studies have demonstrated an association between ACL injury-induced proprioceptive deficits and a reduction in mechanoreceptor number4. However, the contributions of these morphological changes remain unclear. Interestingly, mechanoreceptors with atypical morphologies have never been observed in normal human ACLs5,10–12. However, they have been documented exclusively in injured ACLs8,13–15. Thus, these mechanoreceptors possibly originate from the degeneration or remodeling of morphologically normal mechanoreceptors. These structural alterations may reflect the pathological changes associated with disrupted sensory input. Therefore, elucidating the nature and progression of mechanoreceptor morphological changes may be crucial for understanding the mechanisms underlying proprioceptive impairment and developing strategies to restore neurosensory function after ACL injury. Surgical reconstruction remains the standard treatment for ACL injuries, particularly in young physically active individuals16. However, the native ACL has limited capacity for spontaneous healing, and conventional reconstruction techniques often involve complete removal of the injured ligament using autografts or allografts. This procedure eliminates native mechanoreceptors and restricts the potential for full proprioceptive recovery even after mechanical stabilization is achieved17. In recent years, remnant-preserving reconstruction techniques have emerged as promising alternatives for maintaining proprioceptive function by preserving the neural structures within the residual tissue18,19. These approaches have demonstrated potential benefits in terms of functional outcomes and reduced reinjury rates20. Despite not being the standard of care, conservative management is indicated in specific subgroups, including middle-aged and older individuals, patients with incomplete tears, and those with medical contraindications to surgery21. However, its effects on proprioceptive function remain unclear. Conventional conservative approaches have predominantly focused on compensatory strategies, such as muscle strengthening and neuromuscular training, rather than promoting intrinsic ligament healing21. Furthermore, the growing interest in remnant-preserving reconstruction techniques18–20 underscores the clinical relevance of understanding how mechanoreceptor integrity can be maintained even in the setting of ACL injury. Although preservation of the native ACL may facilitate maintaining mechanoreceptors and proprioceptive function, few studies have directly examined this possibility, and histological evaluations of mechanoreceptor morphology during conservative treatment are lacking. Recently, a novel preclinical model known as controlled anterior translation of the tibia (CATT), which uses brace-like external stabilization, has been developed and shown to promote spontaneous healing of a completely ruptured ACL22. However, whether the number and structure of mechanoreceptors, as well as the proprioceptive function, are preserved in the CATT model remains unclear. This study aimed to investigate the effects of the CATT model on sensorimotor function and mechanoreceptor morphology following ACL injury and to clarify the relationship between these two factors. We hypothesized that the CATT model preserves proprioceptive function after ACL injury by maintaining or restoring mechanoreceptor structures. Specifically, if proprioceptive function is preserved in the CATT model, we expected this to be reflected in the maintenance of typical mechanoreceptor numbers and the absence of a pathological increase in atypical mechanoreceptors at both the femoral and tibial attachment sites, consistent with structural preservation of the residual neural elements within the remnant tissue. We performed supplementary electrophysiological analyses to evaluate neural function after CATT intervention. These findings are useful for developing ACL treatment strategies that address not only mechanical stability, but also proprioceptive recovery. Results Evaluation of the sensorimotor function of the knee The experimental timeline is shown in Fig. 1. Rats were assigned to the sham, ACL transection (ACLT), or CATT groups and sacrificed at 1, 2, 4, or 8 weeks postoperatively. Behavioral, electrophysiological, and histological assessments were performed at each time point. Experimental design and surgical methods for the CATT model. (A) Experimental protocols. All rats were randomly assigned to three groups: sham-operated (Sham, n = 24), ACL transection (ACLT, n = 24), and controlled anterior translation of the tibia after ACL transaction (CATT, n = 24). All rats underwent 1 week of acclimatization (week − 1), followed by baseline behavioral measurements and surgical intervention at week 0. Each group was divided into four subgroups (n = 6 per time point) corresponding to 1, 2, 4, and 8 weeks postoperatively. Behavioral tests, electrophysiological analyses, and tissue sampling were conducted at 1, 2, 4, and 8 weeks postoperatively. Histological analysis was performed after sacrifice at each respective time point. (B) Surgical procedures for the CATT model. Detailed surgical steps used in this study. The medial capsule of the right knee joint was exposed (a), and the ACL was transected (b). A bone tunnel was created in the anterior region of the proximal tibia using a drill (c), and a 3 − 0 nylon suture was passed through the tunnel (d) and anchored to the posterior aspect of the distal femur (e). The nylon suture (f) applies a posteriorly directed traction force on the tibia to resist anterior translation relative to the femoral condyles (g). Note that intra-articular structures are not directly visible in panels (b) through (h), as the surgical procedure was performed through a medial capsular approach without direct intra-articular visualization, which is standard in rat ACLT models. CATT, controlled anterior translation of the tibia; ACL, anterior cruciate ligament; ACLT, ACL transection. In the balance beam test with a 2.4-cm beam width at 1, 2, and 4 weeks postoperatively, no significant difference in the normalized success rate in crossing the beam was observed among the three groups; however, at 8 weeks postoperatively, the normalized success rate of the ACLT group was significantly lower than that of the sham group (p = 0.0108; Fig. 2A). In the balance beam test with a 1.8-cm beam width at 1, 2, and 4 weeks postoperatively, no significant difference in the normalized success rate in crossing the beam was observed among the three groups; however, at 8 weeks postoperatively, the normalized success rate of the ACLT group was significantly lower than that of the sham group (p = 0.0412; Fig. 2B). In the balance beam test with a 2.4-cm beam width, the normalized crossing time of the ACLT group was significantly higher than that of the sham group at 1 and 8 weeks postoperatively (p = 0.0124 and p = 0.0005, respectively; Fig. 2C), and the normalized crossing time of the CATT group was significantly higher than that of the ACLT group at 2, 4, and 8 weeks postoperatively (p = 0.0296, p = 0.0412, and p = 0.0031, respectively; Fig. 2C). In the balance beam test with a 2.4-cm beam width, the normalized crossing time of the ACLT group was significantly higher than that of the sham group at 1 and 8 weeks postoperatively (at 1 and 8 weeks postoperatively, p = 0.0032 and p = 0.0386, respectively; Fig. 2D), and the normalized crossing time of the CATT group was significantly higher than that of the ACLT group at 2 and 8 weeks postoperatively (at 2 and 8 weeks postoperatively, p = 0.0077 and p = 0.0175, respectively; Fig. 2D). Sensorimotor performance on balance beam tests. Behavioral testing used a balance beam at 1, 2, 4, and 8 weeks postoperatively. (A) The normalized success rate in balance beam test on 2.4 cm beam width. (B) The normalized success rate in balance beam test on 1.8 cm beam width. (C) The normalized crossing time in balance beam test on 2.4 cm beam width. (D) The normalized crossing time in balance beam test on 1.8 cm beam width. Data are reported as means ± standard errors of the mean. Significantly different values (p < 0.05) are displayed. Compound muscle action potentials (CMAPs) CMAPs of the gastrocnemii were evaluated to assess the functional status of the nerves. The latency in the CATT group was significantly lower than that in the sham group at 4 weeks postoperatively (p = 0.0455), although no significant differences were observed among the three groups at 1, 2, and 8 weeks postoperatively (Fig. 3A–D). The duration (Fig. 3E–H) and amplitude (Fig. 3I–L) were not significantly different among the three groups at any time point. Electrophysiological assessment of gastrocnemius muscle function. Compound muscle action potentials (CMAPs) of the gastrocnemius muscle at 1, 2, 4, and 8 weeks postoperatively. Latency (ms) at 1 week (A), 2 weeks (B), 4 weeks (C), and 8 weeks (D) postoperatively, duration (ms) at 1 week (E), 2 weeks (F), 4 weeks (G), and 8 weeks (H) postoperatively, amplitude (mV) at 1 week (I), 2 weeks (J), 4 weeks (K), and 8 weeks (L) postoperatively. Data are reported as means ± standard errors of the mean. Significantly different values (p < 0.05) are displayed. Histological analysis At all time points, complete rupture was observed in the ACLT group (Fig. 4A-b, e, h, and k), and no healing response was observed in the remnants. In the CATT group, the remnants showed a healing response from 2 weeks postoperatively (Fig. 4A-c, f, i, l). Histological analysis of ACL healing and mechanoreceptor distribution at bone attachment sites. (A) Healing response between sham, ACLT, and CATT groups. Photomicrographs show longitudinal sections of the ACL stained with hematoxylin and eosin: (b, c, d, h, and k) a nonhealing ACL remnant (black arrows point to ligament remnants) and (f, i, and l) a healing ACL (white arrows point to the healing regions). ACLT, anterior cruciate ligament transection; CATT, controlled anterior translation of the tibia. Scale bars = 2000 μm (overview), 100 μm (magnified regions). (B) Morphological analysis of mechanoreceptor distribution at femoral and tibial bone attachment sites using hematoxylin–eosin staining. Rectangular frames indicate areas magnified to illustrate detailed ligament morphology at femoral and tibial attachments. Scale bars = 2000 μm (overview), 500 μm (magnified regions). The distribution of mechanoreceptors, as confirmed morphologically, was visualized in a schematic mapping plot, with black dots representing mechanoreceptor locations. Morphology and quantitative evaluation of mechanoreceptors Mechanoreceptor morphology and quantification were assessed at the 8-week time point, as significant differences in sensorimotor performance between groups were first detected at this time point in the balance beam test. Three types of standard corpuscles and atypical mechanoreceptors were identified at both femoral and tibial attachment sites (Fig. 5). The composition and total counts of the mechanoreceptor type at the femoral attachment sites were analyzed (Table 1). Although no significant difference in the number of standard mechanoreceptors was observed between the three groups, the number of atypical mechanoreceptors was significantly higher in the ACLT group than in the sham group (p = 0.0263) (Fig. 6A). At the tibial attachment sites, the composition and total counts of the mechanoreceptor types were similarly analyzed (Table 1). The number of standard mechanoreceptors was significantly lower in the ACLT group than in the sham group (p = 0.0340), whereas the number of atypical mechanoreceptors did not differ significantly among the three groups (Fig. 6B). Immunofluorescent identification of mechanoreceptor types at femoral and tibial bone attachments of ACL. Representative images of four types of mechanoreceptors as follows: type Ⅰ (Ruffini corpuscles) (A–D), type Ⅱ (Pacinian corpuscles) (E–H), type Ⅲ (Golgi-Mazzoni corpuscles) (I–L), and atypical mechanoreceptors (M–P). (A, E, I, M) Immunostaining for S100β (red). (B, F, J, N) Neurofilament markers (green). (C, G, K, O) Nuclear staining with 4′,6-diamidino-2-phenylindole (blue). (D, H, L, P) Merged images. White arrows indicate Schwann cells, and white arrowheads indicate nerve fiber endings. Scale bar = 50 μm. ACL, anterior cruciate ligament. Table 1Composition and total counts of mechanoreceptor types in femoral and tibial attachment sites.Attachment siteFemoral attachmentsProportion of each mechanoreceptor type (%)Total number of mechanoreceptorsMechanoreceptor typeShamACLTCATTShamACLTCATTFemoralRuffini corpuscles32.9 ± 4.914.7 ± 5.219.7 ± 5.5271015Pacinian corpuscles38.7 ± 3.827.4 ± 8.151.6 ± 5.3342142Golgi–Mazzoni corpuscles32.9 ± 4.917.0 ± 5.016.6 ± 3.2171213Atypical mechanoreceptors6.8 ± 3.141.0 ± 10.212.1 ± 4.862611Total mechanoreceptors846981TibialRuffini corpuscles27.9 ± 2.918.5 ± 2.436.6 ± 5.119617Pacinian corpuscles25.5 ± 6.812.6 ± 8.417.3 ± 6.61679Golgi–Mazzoni corpuscles37.2 ± 4.526.2 ± 6.528.5 ± 4.424813Atypical mechanoreceptors9.5 ± 4.942.7 ± 8.217.6 ± 5.871611Total mechanoreceptors663750Proportion data of each mechanoreceptor type (%) are reported as means ± standard errors of the means.Total mechanoreceptor counts represent the number of identified corpuscles per group.CATT, Controlled Anterior Translation of the Tibia; ACL, anterior cruciate ligament; ACLT, ACL transection. Composition and total counts of mechanoreceptor types in femoral and tibial attachment sites. Proportion data of each mechanoreceptor type (%) are reported as means ± standard errors of the means. Total mechanoreceptor counts represent the number of identified corpuscles per group. CATT, Controlled Anterior Translation of the Tibia; ACL, anterior cruciate ligament; ACLT, ACL transection. Quantitative analysis of mechanoreceptors at femoral and tibial attachment sites of ACL. (A) Immunofluorescent staining at femoral attachment sites for anterior cruciate ligament (a). Quantification of mechanoreceptor numbers at the femoral attachment, showing the distribution and count of standard and atypical mechanoreceptors (b). (B) Immunofluorescent staining at tibial attachment sites for anterior cruciate ligament (c). Quantification of mechanoreceptor numbers at the femoral attachment, showing the distribution and count of standard and atypical mechanoreceptors (d). The white arrows indicate mechanoreceptors. Mechanoreceptors are categorized into standard mechanoreceptors, including type I (Ruffini corpuscles), type II (Pacinian corpuscles), and type III (Golgi-Mazzoni corpuscles), and atypical mechanoreceptors. Scale bar = 500 μm. Data are reported as means ± standard errors of the mean. Significantly different values (p < 0.05) are displayed. ACL, anterior cruciate ligament. Correlation between mechanoreceptor counts and beam-walking performance The correlation between the mechanoreceptor counts and beam-walking performance was analyzed using Pearson’s correlation coefficient (Table 2). The total number of mechanoreceptors at the femoral and tibial attachments showed a significant negative correlation with normalized crossing time at a beam width of 2.4 cm (r = − 0.4799, p = 0.0439). By contrast, no significant correlations were observed between the total number of mechanoreceptors and the normalized success rate at either beam width (2.4 cm–1.8 cm). When analyzed separately for each attachment site, the total number of mechanoreceptors at the femoral and tibial attachments exhibited a significant negative correlation with normalized crossing time at a beam width of 1.8 cm (r = − 0.5672, p = 0.0141). Consistently, no significant correlations were observed between the total mechanoreceptor counts and normalized success rate at either beam width. The number of standard mechanoreceptors at both the femoral and tibial attachments showed significant negative correlations with normalized crossing time at a beam width of 2.4 cm (femoral attachment: r = − 0.6084, p = 0.0074; tibial attachment: r = − 0.7583, p = 0.0003). At the tibial attachment, the number of standard mechanoreceptors showed a significant positive correlation with the normalized success rate at a beam width of 2.4 cm (r = 0.4989, p = 0.0351) and a significant negative correlation with the normalized crossing time at a beam width of 1.8 cm (r = − 0.5385, p = 0.0211). In addition, the number of atypical mechanoreceptors at the femoral attachment showed significant negative correlations with normalized success rate at both beam widths (2.4 cm: r = − 0.6592, p = 0.0029; 1.8 cm: r = − 0.5429, p = 0.0199). By contrast, significant positive correlations were observed between the number of atypical mechanoreceptors and normalized crossing time (2.4 cm: r = 0.8061, p < 0.0001; 1.8 cm: r = 0.6431, p = 0.0040). At the tibial attachment, no significant correlations were identified between the number of atypical mechanoreceptors and either the normalized success rate or the normalized crossing time at both beam widths. Table 2Statistical analysis of the correlation between mechanoreceptor counts and balance beam performance.Femoral and tibial attachmentsFemoral attachmentsTibial attachmentsAll mechanoreceptorsAll mechanoreceptorsStandard mechanoreceptorsAtypical mechanoreceptorsAll mechanoreceptorsStandard mechanoreceptorsAtypical mechanoreceptorsNormalized success rate (beam width: 2.4 cm)r= 0.0737 ;p = 0.7714r = − 0.1055 ;p = 0.6769r = −0.2380 ;p = 0.3415r = − 0.6592 ;p = 0.0029*r = 0.2576 ;p = 0.3020r = 0.4989 ;p = 0.0351*r = − 0.4058 ;p = 0.0947Normalized crossing time (beam width: 2.4 cm)r= −0.4799 ;p = 0.0439*r = − 0.2149 ;p = 0.3917r = − 0.6084 ;p = 0.0074*r = 0.8061 ;p < 0.0001*r = − 0.5672 ;p = 0.0141*r = − 0.7583 ;p = 0.0003*r = 0.3118 ;p = 0.2078Normalized success rate (beam width: 1.8 cm)r = − 0.0444 ;p = 0.8611r = −0.2522 ;p = 0.3127r = 0.0436 ;p = 0.8637r = − 0.5429 ;p = 0.0199*r = 0.2335 ;p = 0.3511r = 0.4548 ;p = 0.0579r = − 0.3723 ;p = 0.1281Normalized crossing time (beam width: 1.8 cm)r = −0.2676 ;p = 0.2830r = − 0.0633 ;p = 0.8031r = − 0.3855 ;p = 0.1141r = 0.6431 ;p = 0.0040*r = − 0.3860 ;p = 0.1136r = − 0.5385 ;p = 0.0211*r = 0.2505 ;p = 0.3161Pearson’s correlation coefficients (r) and corresponding p-values are shown. *p < 0.05 was considered statistically significant. Statistical analysis of the correlation between mechanoreceptor counts and balance beam performance. r= 0.0737 ; p = 0.7714 r = − 0.1055 ; p = 0.6769 r = −0.2380 ; p = 0.3415 r = − 0.6592 ; p = 0.0029* r = 0.2576 ; p = 0.3020 r = 0.4989 ; p = 0.0351* r = − 0.4058 ; p = 0.0947 r= −0.4799 ; p = 0.0439* r = − 0.2149 ; p = 0.3917 r = − 0.6084 ; p = 0.0074* r = 0.8061 ; p < 0.0001* r = − 0.5672 ; p = 0.0141* r = − 0.7583 ; p = 0.0003* r = 0.3118 ; p = 0.2078 r = − 0.0444 ; p = 0.8611 r = −0.2522 ; p = 0.3127 r = 0.0436 ; p = 0.8637 r = − 0.5429 ; p = 0.0199* r = 0.2335 ; p = 0.3511 r = 0.4548 ; p = 0.0579 r = − 0.3723 ; p = 0.1281 r = −0.2676 ; p = 0.2830 r = − 0.0633 ; p = 0.8031 r = − 0.3855 ; p = 0.1141 r = 0.6431 ; p = 0.0040* r = − 0.3860 ; p = 0.1136 r = − 0.5385 ; p = 0.0211* r = 0.2505 ; p = 0.3161 Pearson’s correlation coefficients (r) and corresponding p-values are shown. *p < 0.05 was considered statistically significant. Discussion This study demonstrated that ACLT impaired sensorimotor function, as evidenced by the prolonged crossing time and reduced success rates in the balance beam test. CATT partially restored these deficits, particularly at 4 and 8 weeks postoperatively. The absence of significant differences at earlier time points may reflect compensatory neuromuscular mechanisms, such as increased muscular co-contraction and postural adaptations, which may mask underlying proprioceptive deficits23,24. However, electrophysiological assessments have revealed minimal changes in CMAP amplitude and duration, indicating that peripheral nerve conduction alone may not account for behavioral recovery. Histological analyses showed the preservation of ligament continuity and healing responses in the CATT group, supporting the notion that the intended mechanical stabilization provided by the CATT construct may have contributed to functional improvement, although direct biomechanical verification was not performed in this study. The nylon suture construct used in the CATT model applies a posteriorly directed traction force to resist anterior tibial translation without restricting knee flexion22. We propose that the superior outcomes in the CATT group result from two complementary mechanisms: first, preservation of remnant continuity restores mechanical stimulation of the ligament tissue, which may promote mechanosensitive gene expression and tissue homeostasis; second, this continuity maintains the tissue microenvironment surrounding the mechanoreceptors, potentially supporting their survival. In contrast, complete ligament disruption in the ACLT group eliminates these stimuli, likely promoting mechanoreceptor degeneration. However, gene expression analysis was not performed in this study and this interpretation remains speculative. It should also be noted that the intra-articular suture may have provided passive stabilization during beam walking, independent of biological healing; future studies should include a suture-only control group to disentangle these contributions Moreover, the present study clarified the distinct functional roles of standard and atypical mechanoreceptors. Standard mechanoreceptors, such as the Ruffini and Pacinian corpuscles, show strong negative correlations with motor impairment, indicating their critical involvement in coordinated motor function. By contrast, atypical mechanoreceptors were positively correlated with motor impairment, suggesting that they may reflect degenerative forms of mechanoreceptors that disrupt sensory feedback. Among the components of sensorimotor performance, the number of mechanoreceptors was significantly correlated with the crossing time rather than with the normalized success rate. Thus, mechanoreceptor degeneration after ACL injury might predominantly impair motor coordination and precision rather than simple success or failure in task completion. This observation is consistent with those reported in previous literature, which have indicated that proprioceptive deficits following ACL injury primarily affect fine motor control rather than basic functional outcomes23. The 1.8 cm beam was intended to place greater demands on proprioceptive coordination than the 2.4 cm beam. However, substantial body growth over the 8-week observation period likely altered the relative task difficulty of the narrower beam, potentially introducing variability that reduced its sensitivity as a measure of proprioceptive function across time points. This may account for the relatively weaker correlations observed between some mechanoreceptor counts and crossing time on the 1.8 cm beam compared with the 2.4 cm beam. Interestingly, our findings further differentiated between the roles of standard and atypical mechanoreceptors. Standard mechanoreceptors, including the Ruffini and Pacinian corpuscles, play well-documented roles in providing accurate proprioceptive feedback regarding joint position, motion, and load, thus facilitating precise neuromuscular control and joint stability13,25,26. Consistent with these known roles, standard mechanoreceptors at both the femoral and tibial attachments in our study showed strong negative correlations with normalized crossing time, indicating their critical contribution to efficient and coordinated movements. Conversely, atypical mechanoreceptors, which significantly increased in number in the ACLT group at the femoral attachment sites, showed negative correlations with normalized success rates and strong positive correlations with crossing time. However, the role of atypical mechanoreceptors in joint pathology remains unclear. Atypical mechanoreceptors have been reported not only in injured ACLs but also in other connective tissues of the knee joint, including the collateral ligaments and tendons27, suggesting that their formation may reflect a broader response to disrupted mechanical environments rather than being specific to ACL injury. Recent studies have suggested that atypical mechanoreceptors may represent degenerative mechanoreceptors that potentially provide inaccurate or abnormal sensory information that interfere with proper motor control27,28. Our results support and expand upon this hypothesis, showing that elevated counts of atypical mechanoreceptors are significantly associated with reduced sensorimotor performance. Thus, atypical mechanoreceptors may reflect a form of maladaptive plasticity following ACL injury, leading to disrupted proprioception and impaired motor control6,24,29. Moreover, in the present study, distinct changes in the mechanoreceptor profiles were observed between the femoral and tibial attachment sites following ACL transection. Specifically, increased atypical mechanoreceptors and a strong correlation between their number and motor dysfunction were noted at the femoral attachment, whereas decreased standard mechanoreceptors and their correlation with impaired function were observed at the tibial attachment. Previous human studies have reported no significant difference in mechanoreceptor counts between the femoral and tibial remnants following ACL injury30. However, our findings differ likely because of the use of a rat model and distinct evaluation criteria. Separate analyses of the femoral and tibial attachment sites may further elucidate the mechanoreceptor changes in the ACL. The differential mechanoreceptor changes between the femoral and tibial attachment sites may partly reflect variability in the precise ACL transection site. A transection closer to the femoral attachment could subject that region to greater mechanical disruption, potentially contributing to the higher incidence of atypical mechanoreceptors observed at this site. Future studies should standardize the transection site using image-guided or anatomical landmark-based approaches. Electrophysiological assessments using CMAPs have provided additional insights into peripheral nerve function after ACL injury and treatment. Although no significant differences in amplitude or duration were observed between the ACLT and CATT groups, the significantly lower latency in the CATT group compared with the sham group at 4 weeks warrants careful interpretation. ACL transection causes joint instability and local inflammatory changes6 that may exert mild mechanical or inflammatory effects on adjacent neural structures, potentially leading to a transient prolongation of latency in the absence of mechanical stabilization. The CATT construct, by restricting abnormal anterior tibial translation, may have attenuated this local inflammatory environment at 4 weeks, thereby preventing the mild latency prolongation that might otherwise occur. However, since this difference was observed only at a single time point and was not accompanied by changes in amplitude or duration—which are more sensitive indicators of axonal loss and demyelination, respectively—this finding should be interpreted with caution and may also represent a statistical artifact given the number of comparisons performed. Furthermore, the possibility that the intra-articular suture construct mechanically influenced peroneal or tibial nerve branches during early remodeling cannot be excluded without additional biomechanical and neuropathological assessments. However, the lack of meaningful electrophysiological differences implies that the improvements in motor performance are unlikely to be solely explained by efferent conduction. Given that CMAPs primarily reflect efferent motor pathways, the absence of between-group differences suggests a comparable peripheral motor output. Therefore, the preservation of motor function observed in the CATT group may reflect the attenuated degeneration of the afferent sensory pathways, as indicated by the immunohistochemical findings of mechanoreceptors. Additionally, central adaptations in response to altered sensory input following ACL injury may contribute to the observed behavioral outcomes. These interpretations are consistent with those of previous findings that electrophysiological measures alone may not fully capture the extent of functional impairment after ACL31–33. This study has some limitations. First, the assessment of sensorimotor function relied solely on balance beam testing, which may not comprehensively capture all aspects of sensorimotor performance. Second, whether peripheral sensory deficits caused the observed motor impairments remains unclear, as we did not administer regional anesthesia of the knee or evaluate its effects after the removal of CATT. Additionally, although manual anterior drawer testing confirmed ACL rupture and restriction of anterior tibial translation intraoperatively, quantitative measurement of tibial translation distance was not performed. Future studies incorporating instrumented laxity testing would further validate the biomechanical efficacy of the CATT model. A further limitation is that cartilage and osteoarthritic assessments were not performed in this study. As the primary focus was on neural and functional outcomes, and the histological sections used for mechanoreceptor analysis contained predominantly the ligament-bone interface, formal articular cartilage evaluation was not feasible in these sections. However, given that early-stage cartilage degeneration has been reported in rat ACLT models within as early as 2 weeks postoperatively34, the possibility of osteoarthritic changes in the ACLT group cannot be excluded and may represent a confounding factor. Future studies should incorporate Safranin-O staining or equivalent cartilage assessment to fully characterize the joint environment following ACL transection. Additionally, mechanoreceptor morphological analysis was limited to the 8-week time point. Longitudinal tracking of mechanoreceptor changes at earlier time points would be valuable for understanding the temporal progression of degeneration and recovery following ACL injury. Lastly, the assessment of neuromuscular function was restricted to the gastrocnemius muscle and thus, may not fully reflect the function of other muscles or nerves. Future studies should investigate the contribution of sensory input to motor behavior by using more specific electrophysiological assessments, including tests focused on sensory nerve function, to clarify the mechanisms underlying sensorimotor dysfunction following ACL injury. In conclusion, preserving the morphology and function of mechanoreceptors, particularly standard mechanoreceptors, at ACL attachment sites was essential for maintaining sensorimotor performance following ACL injury. The presence of atypical mechanoreceptors may reflect pathological remodeling processes that negatively contribute to functional recovery. Thus, therapeutic strategies aimed at preventing mechanoreceptor degeneration or selectively promoting the regeneration of functionally beneficial receptor types could be promising approaches to improve motor outcomes and potentially contribute to joint stability in clinical settings. Methods Animal and experimental groups This study was approved by the Institutional Animal Care and Use Committee, and adhered to the Kyoto University Animal Experimentation Regulations (MedKyo23077) and conducted in accordance with the ARRIVE guidelines (https://arriveguidelines.org). All methods were performed in accordance with the relevant guidelines and regulations. A total of 72 male Wistar rats (250–270 g body weight; Japan SLC, Hamamatsu, Japan) aged 12 weeks were used. All the rats were allowed to acclimatize to the environment for 1 week. Immediately after the acclimatization period, the surgical intervention was performed, and they were housed in plastic cages with sawdust bedding (1–2 animals per cage), maintained at room temperature (22 ± 2 °C) under a 12:12-hour light–dark cycle. The animals were allowed to move freely in their cages and were provided free access to food and water. The animals were randomly divided into three groups: the sham group (n = 24), ACLT group (n = 24), and CATT group (n = 24) and performed on the right hind limb. Animals were allocated to the following four time points: 1, 2, 4, and 8 weeks postoperatively for each group at each time point. These time points were selected to capture key phases of ligament healing and mechanoreceptor remodeling: 1 and 2 weeks represent the early inflammatory and proliferative phases of tissue repair, 4 weeks corresponds to the early remodeling phase during which initial structural recovery may become apparent, and 8 weeks allows evaluation of longer-term functional and histological outcomes. The 1- and 2-week time points correspond to those examined in the original CATT model study22, and the 4- and 8-week time points were additionally included to capture the progressive remodeling phase and to allow evaluation of longer-term functional and histological outcomes. Behavioral tests were performed using baseline measurements preoperatively and at 1, 2, 4, and 8 weeks postoperatively. Electrophysiological assessments were performed 1, 2, 4, and 8 weeks postoperatively. In addition, histological analysis to investigate mechanoreceptor dysfunction was performed at 1, 2, 4, and 8 weeks postoperatively in each group at each time point. After electrophysiological assessment, rats were placed under a surgical plane of anesthesia by intraperitoneal overdose of a triple-combination anesthetic (1.0 mL / 100 g body weight; 1.0 mg medetomidine hydrochloride, 0.4 mg midazolam, 0.5 mg butorphanol). Transcardial perfusion was then performed with phosphate-buffered saline (PBS) followed by 4% paraformaldehyde (PFA) in PBS. Cessation of heartbeat and respiration was verified at the end of perfusion to confirm death. Surgical procedure The animals were anesthetized with mixed anesthetics by intraperitoneal injection after inhalation anesthesia (1.0 ml/100 mg body weight). The right knee joint was exposed via the medial capsule without disrupting the patellar tendon (Fig. 1B-b), and the ACL was completely transected (Fig. 1B-c). A manual tibial anterior withdrawal test confirmed the ACL rupture. In the CATT group, a manual anterior drawer test was additionally performed after suture placement to confirm restriction of anterior tibial translation. Using a drill, a tibial tunnel was formed on the medial side of the rough tibial surface in the medial and lateral directions (Fig. 1B-d). In the ACLT group, the joint capsule and parapatellar fascia were closed using sutures (S15G04N-45; Bear Medic Corporation, Chiba, Japan) and the skin was closed. In the CATT group, based on a previous study22, sutures (S15G03N-45, Bear Medic Corporation, Chiba, Japan) were passed through the tibial tunnel (Fig. 1B-e) and twice through the posterior condyle of the distal femur (Fig. 1B-f) and then tied to restrict excessive anterior translation of the tibia (Fig. 1B-g). Finally, the skin was closed (Fig. 1B-h). The sham group underwent skin incision, capsular opening, and tibial tunnel creation without ACLT or CATT. Evaluation of the sensory-motor kunction of the Knee Knee sensory-motor function was evaluated using the balance beam test (BBT). A schematic diagram of the balance beam apparatus is provided in Supplementary Figure S1. In the BBT, rats crossing a beam 90 cm long and 1.8–2.4 cm wide were filmed from the right side in the direction of travel, and the number of hindlimb slips that occurred while crossing the beam was measured from the filmed images according to the method of Luong et al.35. A slip was defined as a complete fall of the hindlimb from balance or a fall of more than half of the hind tarsus from the beam. The mean of both procedures was calculated for each rat and normalized by dividing it by the baseline data before the surgical procedure. These three procedures cannot be performed consecutively. If the rat interrupted the crossing of the beam, the test was repeated and reassessed after a specific time interval. Electrophysiological analysis To assess peripheral neuromuscular function, CMAP recordings were conducted at 1, 2, 4, and 8 weeks postoperatively following motor functional analysis. The rats were anesthetized with mixed anesthetics (1.0 ml/100 g body weight), and electrophysiological assessments were performed using an electromyogram measuring system (Neuropack S1 MEB-9404; Nihon Kohden, Tokyo, Japan). Sciatic nerves were then stimulated proximal to the injury site, and CMAPs were recorded in the gastrocnemius muscle, which was innervated by the tibial nerve36,37. The latency, duration, and amplitude of CMAPs are expressed as the ratio of injured to non-injured individuals. Histological analysis The ACL healing response was evaluated histologically at 1, 2, 4, and 8 weeks postoperatively after perfusion fixation. At each time point, the knee constructions of the operated hind limb were collected, retrieved entirely, and fixed in 4% paraformaldehyde. After fixation, all tissues were infiltrated with 70% ethanol (30 min), 100% ethanol (overnight), and 70༅ ethanol (30 min), and were decalcified in a K-CX AT rapid demineralization solution (CS-6101, Falma, Japan) at 4 °C for 6–8 days. Tissues were embedded in paraffin at an optimal cutting temperature. Paraffin-embedded tissues were sliced to a thickness of 7 μm and stained with hematoxylin and eosin to assess the morphological features of the ACL. Immunofluorescence staining The sections were deparaffinized and heated at 62 °C for 20 min in HistoVT One (Nacalai Tesque, Inc., Kyoto, Japan; diluted 1:10) for antigen retrieval. The sections were washed twice for 5 min each with phosphate-buffered saline (PBS) and once for 5 min each with 0.1% Triton X-100 in PBS for permeabilization. Goat serum albumin solution (5%) was used to block non-specific reactions, and the preparation was allowed to react for 1 h at room temperature before washing twice for 5 min with PBS and once for 5 min with PBS with Tween® (PBST). Subsequently, the sections were incubated overnight at 4 ℃ with anti-neurofilament medium chain (anti-NFM) antibody (1:1000 dilution; RMO-270; Thermo Fisher Scientific), together with an anti-S100β antibody (1:1000 dilution; ab52642; Abcam). On the following day, the sections were washed twice for 5 min with PBS and once for 5 min with PBST. The secondary antibodies included Alexa Fluor 488 goat anti-rabbit IgG (H + L) (1:500 dilution; A-11001; Thermo Fisher Scientific), Alexa Fluor 594 goat anti-rabbit IgG (H + L) (1:500 dilution; A-11012; Thermo Fisher Scientific), and 4′,6-diamidino-2-phenylindole (1:5000 dilution; 340–07971; Fuji Film Wako Junyaku) which were incubated for 1 h at room temperature. Sections were washed thice for 5 min each in PBS and sealed with a cover glass using VECTASHIELD antifade mounting medium (H-1000; Vector Laboratories). Morphology and quantitative evaluation of mechanoreceptors Given that significant differences in behavioral test outcomes were observed at 8 weeks postoperatively, morphological and quantitative evaluations of the mechanoreceptors were conducted at this time point. Mechanoreceptors were identified based on the following criteria: (1) positive immunoreactivity for both NFM and S100β, markers specific to axons and Schwann-like glial cells38–40, and (2) 40 to 120 μm in diameter8,27,41,42. The morphological classification of the mechanoreceptors was determined by an examiner with expertise in assessment using the Freeman and Wyke43 morphological criteria. The mechanoreceptors in the ligaments include the Ruffini corpuscles (RCs), Pacinian corpuscles, and Golgi-Mazzoni corpuscles. The mechanoreceptors at the femoral and tibial attachment sites of the ACL were counted separately. Two consecutive ACL sections were prepared from each animal. Mechanoreceptors that appeared at the corresponding locations in both sections had the same structure and were counted only once. The total number of mechanoreceptors at each attachment site was recorded. Prior to embedding, each specimen was bisected in the sagittal plane along the medial border of the patella, and only the lateral portion was retained for sectioning. Sections were taken from the central portion of the ACL in the sagittal plane, and correct plane orientation was verified by the presence of both the femoral and tibial attachment sites within the same section. Atypical mechanoreceptors were defined as nerve endings that did not conform to any of the three standard morphological categories based on Freeman and Wyke’s classification43. Although previous studies have reported the presence of mechanoreceptors in the midsection of the human ACL5,44, no mechanoreceptor-like structures were observed in the midsection of the ACL in the histological sections evaluated in the present rat model. To reduce observer bias, all behavioral scoring, histological quantification, and electrophysiological data analyses were performed by investigators blinded to the experimental groups. Statistical analysis Statistical analyses were performed using JMP Pro17 (SAS Institute Inc., Cary, NC, USA) or R ( R Foundation for Statistical Computing, Vienna, Austria), and the following three groups were analyzed: sham, ACLT, and CATT groups. Normality was assessed using the Shapiro-Wilk test for each variable at each time point. For normally distributed data, descriptive statistics are displayed as mean ± standard error of the mean (SEM), and differences among the three groups were evaluated using analysis of variance with post hoc Tukey-Kramer or Games-Howell test. For non-normally distributed data, the Kruskal-Wallis test with post hoc Dunn’s test with Bonferroni correction was applied. For graphical presentation, all data are displayed as mean ± SEM, which reflects the precision of the estimated population mean and facilitates visual comparison of group differences45,46. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 2 Acknowledgements The authors thank Ms. Hiroe Sugiyama for her assistance with the experiments. We would like to thank Editage (www.editage.jp) for English language editing. Author contributions S.M., M.T., A.K., and K.H. conceived and designed the study. S.M. acquired and analyzed the data. S.M. drafted the manuscript. M.T. and K.H. provided technical support. M.T. provided the financial support. M.T. and T.A. supervised this study. All the authors interpreted the data and reviewed the manuscript. Funding This study was supported by a Grant-in-Aid from the JSPS KAKENHI Grant-in-Aid for Early-Career Scientists (21K17472) to M.T. Data availability The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Supplementary Material 2 Data Availability Statement The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.