Introduction

Rotator cuff tears (RCTs) are a common cause of disability and pain in the adult population [1,2,3,4]. Tendon detachment can cause tendon retraction, fatty infiltration, muscle atrophy, and subsequently shoulder instability and dysfunction [3, 5, 6]. A critical hurdle of repairing massive RCTs is the high re-tear rate (up to 94%) after repair [3]. Several factors, such as patient age, tear size, muscle atrophy, delayed repair, and repair strategies, influence the rate of re-tear [3, 7,8,9,10,11,12].

One of the challenges of rotator cuff (RC) repair is the commonly observed delay between the time of injury and surgical repair [13]. The medical terms of acute, sub-acute, and chronic are being used as references to the delay between the time of injury and repair. Acute RC injury refers to the incision that will be repaired in a short time; in humans, it could be a few days or maybe weeks. The injury will be considered chronic if the torn RC is persisting for a prolonged period. On the other hand, sub-acute injuries indicate the conditions that have passed the acute phase but are not yet chronic.

Sub-acute and chronic RCTs are often associated with a set of pathological changes that include muscle atrophy, smaller muscle cross-sectional area, fatty infiltration, and fibrosis [14, 15]. In the initial days post-injury, the migration of inflammatory cells leads to apoptosis of muscle fibers [16, 17]. The intrinsic degenerative changes of the musculotendinous unit lead to muscle atrophy [18, 19]. Despite successful repair of the torn tendon, muscle atrophy is an issue that adversely affects RCTs healing in sub-acute and chronic tears [20]. The undesirable outcomes associated with current surgical techniques have motivated tissue regenerative engineering approaches for RC healing. Regenerative engineering is a novel, alternative approach using the convergence of advanced materials science, stem cell research, physical sciences, and areas of developmental biology and clinical translation that has the potential to address muscle degeneration in moderate injuries with minimal to no scar formation and may also offer a regeneration strategy in the case of severe injuries [21,22,23,24,25,26,27].

Biomimetic matrices with similar structures to native muscle can facilitate the regeneration of muscle [28]. Nanostructure-based matrices such as polymeric/non-polymeric nanofibers, nanotubes, and nanoparticles are considered promising materials in tissue regenerative engineering [29, 30]. Among different nanostructures, nanofibers are one of the most interesting materials owing to their continuous nanoscale long fibers [3, 4, 31,32,33,34]. The nanofiber-based structures showed the most similarity to native muscle tissue by providing similar structures and appropriate mechanical properties. Since muscle is an electrically excitable tissue, the combination of electroconductive material along with aligned nanofibers significantly improves interactions between materials and cells and facilitates myoblast communication, growth, and development [4, 28, 35,36,37].

So far, many studies evaluated the potential of electroconductive materials to promote muscle growth and differentiation in vitro [4, 38,39,40]. According to the literature, the regenerative potential of electroconductive matrices has not been extensively investigated for healing muscle atrophy after RCTs in an animal model. In this study, the efficacy of electrical and topographical cues on reducing muscle atrophy was investigated through implanting electroconductive nanofibrous matrix on supraspinatus muscle after successful RC repair. The aligned poly(ε-caprolactone) (PCL) nanofibers were fabricated by electrospinning technique; then, the electrospun layers were coated with poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) nanoparticles. The efficacy of the matrix was investigated in both acute and sub-acute RCT models. At 2 and 6 weeks post-repair, the supraspinatus muscle was harvested for histological analysis.

Materials and Methods

Matrix Preparation

Matrix design, fabrication, and characterization were published elsewhere (Fig. 1B) [4]. Briefly, the aligned nanofibers were fabricated by electrospinning using 15% (w/v) polycaprolactone (PCL, from Sigma-Aldrich) solution in ethanol and methylene chloride (15:85 ratio, from Sigma-Aldrich), with a 2.5 mL/min flow rate and 1 kV/cm potential at ambient temperature and humidity. Dopamine hydrochloride (2 mg/mL, average molecular weight 8000 Da, from Sigma-Aldrich) was coated onto the matrices for 24 h at pH 8.5 (DOPA/PCL) [41]. The modified DOPA/PCL matrices were washed, dried, and subsequently coated with 30 μL of 10% poly (3,4-ethylenedioxythiophene)-polystyrene sulfonate (PEDOT:PSS 1:2.5, from Sigma-Aldrich) in PBS (PEDOT:PSS/DOPA/PCL). The coated matrices were dried for 48 h under vacuum and sterilized using 70% ethanol and UV for half an hour/side.

Fig. 1
figure 1

(A) Diagram of experimental groups. For sub-acute groups, 4 and 6 rats were sacrificed 2 and 6 weeks after repair respectively; (B) schematic illustration of the rotator cuff surgery model. The aligned PCL nanofibers were fabricated by electrospinning technique. Following the electrospinning, dopamine was applied onto the PCL fibers to polymerize and form a polydopamine coating. DOPA/PCL surfaces were washed, dried, and coated with 10% PEDOT:PSS. Following the surgical repair of the torn supraspinatus tendon, the matrix was implanted on top of the supraspinatus muscle in both acute and sub-acute models

Full size image

Surgical Procedure

The surgical groups and numbers are listed in Fig. 1A. Acute and sub-acute models were developed using adult male Sprague–Dawley rats (Charles River Laboratories, Inc.) weighing between 400 and 450 g. All animal experiments were performed following protocols approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Connecticut (Protocol number: 101417-0919).

The rats were housed in a controlled environment on a light/dark cycle and fed with standard rodent chow and water. An open surgical approach was followed to create a full-thickness RCT of the supraspinatus tendon in the left shoulder. This was followed by a surgical repair with or without the insertion of PEDOT: PSS matrix. Anesthesia was induced with isoflurane (2–3%) and maintained through a facemask for both procedures. After skin incision, a longitudinal incision was made on the deltoid muscle to expose the RC tendon at the shoulder joint. The supraspinatus tendon was completely detached from the bone–tendon insertion on the humeral head, marked with a 5–0 Prolene suture (Ethicon, Johnson & Johnson Medical Ltd.) at the musculotendinous junction, and allowed to retract.

Acute Surgery Procedure

Immediately after the injury, a tunnel was created in the cortical bone under the insertion of the supraspinatus tendon and surgical reattachment of the tendon was performed. For the surgical repair with matrix groups (RM), a 0.6 cm (length) × 0.3 cm (width) electroconductive nanofibrous matrix was sutured on the belly of the supraspinatus muscle; then, the overlying tissue and skin were closed. Analgesia (buprenorphine, 0.05–0.1 mg/kg) and antibiotic (sulmethaoxole and trimethoprim 40 mg/mL and 8 mg/mL) were administered according to the protocol, and the rats were kept under heat lamps until they were transferred to their cages.

Sub-acute Surgery Procedure

The tendon tear was created as described above, and the surgical site closed. The second operation to reattach the tendon was performed 6 weeks after the injury as described in the acute surgery procedure.

Tissue Analysis

Tissue Harvest

Rats were sacrificed 6 weeks after surgery for acute groups and 2 and 6 weeks post-surgery for sub-acute groups. The supraspinatus muscles of both the surgical site and the control side (right shoulder) were harvested. The wet mass of the supraspinatus muscle was weighed immediately. After the measurement, the harvested muscle was used for paraffin embedding.

Wet Weight Measurement

The wet mass of the supraspinatus muscle was measured with the use of a digital scale immediately after dissection (Ohaus® Voyager Pro balance). The relative weight change was calculated to show the weight difference between the groups. The weight change percentage was calculated as follows:

$$ \mathrm{Relative}\ \mathrm{percentage}\ \mathrm{weight}\ \mathrm{loss}=\left(\frac{\mathrm{supraspinatus}\ \mathrm{weight}\ \mathrm{of}\ \mathrm{the}\ \mathrm{right}\ \mathrm{shoulder}-\mathrm{supraspinatus}\ \mathrm{weight}\ \mathrm{of}\ \mathrm{the}\ \mathrm{left}\ \mathrm{shoulder}}{\mathrm{supraspinatus}\ \mathrm{weight}\ \mathrm{of}\ \mathrm{the}\ \mathrm{left}\ \mathrm{shoulder}}\right)\times 100 $$

Immunochemistry Analysis

Following an existing protocol, 10-μm sections were rehydrated in phosphate-buffered saline (PBS), fixed in 4% paraformaldehyde, washed in PBS, and immersed in 5% normal donkey serum for 15 min [42]. Tissue sections were imaged and quantified using immunohistochemistry for the fast type II myosin heavy chain (MHC) isoform, slow type I MHC isoform (Novocastra™ Lyophilized Mouse Monoclonal Antibody Myosin Heavy Chain (fast and slow), 1:10; Leica Biosystems and CD31 (Abcam) using fluorescence labeling methods) [42].

Histological Analysis

The sections were stained by Masson’s trichrome staining, and hematoxylin and eosin to identify fibrotic tissue and muscle fiber outlines [42]. Images were captured and digitized (DM4000B Leica Upright Microscope; Leica Microsystems) at varying magnifications.

Muscle Cross-sectional Area Quantification

Sections were visualized at × 10 magnification (Leica Microsystems), and images were recorded using a digital camera. For Masson’s trichrome staining, the cross-sectional areas of all muscle fibers were counted. For fast muscle staining, the cross-sectional areas of fifty random muscle fibers per image (6 to 9 images per group) were selected and measured using ImageJ software (National Institutes of Health). Muscle samples were analyzed in a random order, random selection, and the investigator was blinded as to which muscle was being analyzed.

Muscle Gap Quantification

Gap areas were quantified between muscle fascicles. Nine representative images with 350 μm × 350 μm dimensions were selected from each group, and ImageJ analysis was used to quantify the muscle gap area.

Statistical Analysis

Mean and standard deviation were expressed as means ± SD. Power analysis was used to determine the number of animals. It is the first time the muscle atrophy in rotator cuff model was studied with matrix implantation. Preliminary data to see the statistical difference was based on volumetric muscle loss (VML) with matrix implantation. Differences between groups were analyzed using one-way ANOVA followed by the Tukey pairwise comparison test, and the difference of P < 0.05 was considered statistically significant.

Results

Figure 2 shows the supraspinatus muscle weight loss of acute and sub-acute groups 2 and 6 weeks after surgery. The TT/2w and TT/6w groups showed significant differences compared with sham, acute, and sub-acute repair groups 2 and 6 weeks after repair. There was no significant difference between the suture and matrix repair groups in both acute and sub-acute models. Moreover, no significant differences were observed between sub-acute repair groups after 2 weeks compared with the 6-week groups after repair.

Fig. 2
figure 2

Supraspinatus weight loss in (A) acute and (B) sub-acute repair groups 2 and 6 weeks after surgery (ns = P > 0.05, * = P ≤ 0.05, ** = P ≤ 0.01, *** = P ≤ 0.001, **** = P ≤ 0.0001)

Full size image

Figure 3A shows H&E staining of the supraspinatus muscle in both acute and sub-acute groups. The images demonstrated the formation of fat cells between muscle fibers in the TT/6w group after detachment. The significant fat formation can be observed in TT/6w compared with TT/2w showing the adverse effects of delayed repair after RCT. All repair groups presented muscle growth compared with the TT/6w group. According to Fig. 3, A-RM and SA-RM/6w exhibited similar muscle fiber integrity to the sham group.

Fig. 3
figure 3

Histology images of the supraspinatus muscle for both acute and sub-acute groups, (A) H&E, (B) Masson trichrome staining, and (C) immunohistochemistry of myosin heavy chain

Full size image

Consistent with H&E results, Masson’s trichrome staining (Fig. 3B) showed similar dense packing and integrity of muscle fibers in sham, A-RM, SA-RM/2w, and SA-RM/6w groups. The presence of fibrosis was observed in A-RS, SA-RS/2w, and SA-RM/2w, and TT group showed more fibrosis 6 weeks after detachment.

The immunohistochemistry of myosin heavy chain (MHC) was further evaluated for supraspinatus muscle to analyze muscle fiber size and cross-sectional area. According to MHC classification, MHC I and MHC II correspond to muscle fiber type I (slow muscle type) and type II (fast muscle type) [43]. Figure 3C shows the muscle fiber staining of all groups while MHC II fibers show brown color and MHC I fibers show white color. The TT groups presented the minimum quantity of slow muscle fibers. Based on the images, both muscle fibers were observed in all repair groups, but the majority of the muscle fibers were MHC II. The traces of fast muscle fibers ingrowth were observed along with the edges of the matrix, 2 and 6 weeks after implantation in both acute and sub-acute groups (Fig. 4A).

Fig. 4
figure 4

Cell infiltration into the matrix after repair. (A) Fast muscle fiber staining of acute and sub-acute groups 2 and 6 weeks post-repair; (B) H&E staining and Masson trichrome staining of acute group 6 weeks after repair; (C) CD31 staining of acute and sub-acute groups 6 weeks post-repair

Full size image

Figure 4 presents the neutrophils and mononuclear cell infiltration at the outer edges of the implanted matrix and muscle fibers growth in A-RM, SA-RM/2w, and SA-RM/6w post-repair. The CD31 staining images of the A-RM and SA-RM/6w groups are shown in Fig. 4C. Based on the images, the matrix induced revascularization 6 weeks after repair.

The quantification of muscle cross-sectional area of all groups was presented in Fig. 5. Consistent with Fig. 3, the TT/6w group showed smaller muscle cross-sectional area compared with TT/2w showing the muscle atrophy in the sub-acute model of RCT. According to Fig. 5A, the cross-sectional areas of the fast muscle fibers in sham/6w and TT/6w groups showed significant differences compared with acute and sub-acute/6w groups. There was no significant difference between acute repair groups (2430 ± 91.99 μm2 and 2537 ± 92.14 μm2, for A-RS and A-RM, respectively), and sub-acute repair groups 6 weeks post-repair (2618 ± 92.22 μm2 and 2724 ± 92.37 μm2, for SA-RS/6w with SA-RM/6w, respectively). However, SA-RM/2w showed a significant increase compared with SA-RS/2w, SA-RS/6w, A-RS, and A-RM.

Fig. 5
figure 5

(A) MHC II muscle fiber cross-sectional area of acute and sub-acute groups 2 and 6 weeks after repair. (B) Gap area between muscle fascicles of acute and sub-acute groups 6 weeks post-surgery (ns = P > 0.05, * = P ≤ 0.05, ** = P ≤ 0.01, *** = P ≤ 0.001, **** = P ≤ 0.0001; black star “★” denote **** = P ≤ 0.0001 significant difference compared with acute and sub-acute/6w)

Full size image

Figure 5B shows the quantification of intrafibrillar spaces, also termed gap area for all the groups. The sham and TT groups showed the lowest and the highest percentages of gap area respectively. The A-RM, SA-RS/6w, and SA-RM/6w showed significant differences compared with the TT group. However, there was no significant difference between the sham group and repair groups with matrices (A-RM and SA-RM/6w).

Discussion

Incorporation of electroactive biomaterials along with topographical cue can facilitate electrical signal transfer in electrically excitable tissues such as skeletal muscle. Studies suggested that electroconductive materials can significantly promote muscle regeneration even without applying external electrical stimulation [4, 44,45,46]. The potential of aligned PEDOT:PSS/PCL nanofibers to proliferate and differentiate myoblasts was investigated by our group and published elsewhere [4]. Further to the in vitro evaluation, the regenerative ability of the electroconductive matrix to treat RC muscle atrophy was examined in this study. Many patients with torn RC tendons do not visit a physician until several months or even years after the initial injury [47]. Muscle atrophy is a commonly observed challenge in sub-acute and chronic RCTs which can occur in the interim between injury and repair. In this study, the clinically relevant rat models of acute and sub-acute RCTs were developed following the tendon repair and implantation of the electroconductive matrix to treat muscle atrophy after successful RC repair. The efficacy of PEDOT:PSS matrix implantation was evaluated compared with the sham group, untreated group (TT), and repair group with suture alone in both acute and sub-acute models. Two and 6 weeks after the repair, the harvested samples were examined by the measurement of supraspinatus muscle weight loss, muscle cross-sectional area, intrafibrillar gap area, and histological changes. These are essential factors in characterizing RC muscle atrophy. Decreased muscle cross-sectional area and muscle mass limit the shoulder motion and muscle post-operative strength, and fail RC tendon repair [48].

The results demonstrated significant changes in supraspinatus muscle mass and muscle cross-sectional area 6 weeks after injury. The high percentage of muscle weight loss in the TT/6w group is consistent with literature showing muscle atrophy after RCT. In both acute and sub-acute models, repaired groups showed significantly less muscle weight loss compared with TT groups. Moreover, the histological results demonstrated the absence of fibrosis in A-RM and SA-RM/6w groups, which indicates the regeneration potential of the electroconductive nanofibrous matrix in both acute and sub-acute models. Moreover, our findings are consistent with clinical results showing that suture repair alone does not reverse muscle degeneration [49, 50].

The immunohistochemistry results demonstrated that MHC type II (fast muscle type) formed the majority of the RC muscle fibers (Fig. 3C). The TT groups showed no MHC I fibers (slow muscle type) 2 and 6 weeks post-surgery. The results were consistent with studies showing slow muscle type is more vulnerable to atrophy induced by external forces such as denervation as well as immobility because of their higher protein metabolic rate [51]. However, MHC II fibers are more sensitive to cachexia, chronic heart failure, chronic obstructive pulmonary disease, and aging [43]. In both acute and sub-acute repair models, the repair groups were able to recover more type I muscle post-surgery (Fig. 3C). Based on Fig. 5A, TT groups showed statistically significant differences compared with other groups suggesting the efficacy of RC repair in healing muscle atrophy. The study exhibited significantly greater MHC II cross-sectional areas and less muscle atrophy when the repair was augmented with the electroconductive matrix. Moreover, the high percentage of gap area between muscle fascicles in the TT group significantly reduced after matrix implantation.

In both acute and sub-acute groups, the implantation of the engineered matrix reduced the gap area and subsequently improved the muscle atrophy. The study showed that even when there is a delay in surgical repair the electroconductive fibrous matrix can successfully improve the muscle atrophy. The low percentage of gap area and larger muscle cross-sectional area were observed in matrix repaired groups in both acute and sub-acute models, where the matrix acted as a stabilizer for supraspinatus muscle and prevented the gap formation.

Studies reported the mononuclear cell infiltration as an initial response to matrix implantation, which peaks approximately 2 weeks post-surgery and gradually decreases afterward [42]. In our study, cell infiltration was observed 2 weeks post-surgery and continued to decrease by 6 weeks after surgery. At the same time, our results exhibited vascularization on implanted matrices 6 weeks after surgical repair in both acute and sub-acute groups, indicating the potential of the matrix for revascularization. Vascularization is an important step for skeletal muscle regeneration by providing nutrient delivery and maintaining metabolic conditions [2]. It was reported that RCTs can decrease the vascularity which subsequently correlates with the degree of muscle atrophy [52]. Therefore, improving vascularity may recover RC muscle growth.

The results of acute and sub-acute models demonstrated that the stimulating effects of topographical and electrical cues even without external electrical stimulation can prevent muscle gap formation, increase muscle cross-section area, and subsequently improve muscle atrophy compared with repair groups with suture alone.

Conclusion

In summary, the optimized electroconductive nanofibrous matrix provided synergistic effects of topographical and electrical cues for skeletal muscle tissue engineering. The results of the TT group showed muscle degeneration and atrophy 6 weeks after RCT. However, implantation of the matrix in both acute and sub-acute repairs significantly improved muscle atrophy. The histological results and quantification of muscle fiber cross-sectional area and gap area of repair groups with matrix showed significant improvement compared with TT group and repair groups with suture alone. The study demonstrated the efficacy of electroconductive aligned fibrous matrix in healing the RCT by improving the muscle atrophy in both acute and sub-acute models. Future studies can be conducted to enhance the regenerative potential of the matrix to treat muscle atrophy in RC chronic model with or without external electrical stimulation. Further, the current work only investigated the tissues at the histological level. Other analyses, such as tendon regeneration, biomechanical properties, and fatty infiltration, will be conducted in future studies.