Graft alignment impacts the regenerative response of skeletal muscle after volumetric muscle loss in a rat model

Graft alignment impacts the regenerative response of skeletal muscle after volumetric muscle loss in a rat model

ARTICLE IN PRESS JID: ACTBIO [m5G;February 1, 2020;7:37] Acta Biomaterialia xxx (xxxx) xxx Contents lists available at ScienceDirect Acta Biomate...

6MB Sizes 0 Downloads 0 Views

ARTICLE IN PRESS

JID: ACTBIO

[m5G;February 1, 2020;7:37]

Acta Biomaterialia xxx (xxxx) xxx

Contents lists available at ScienceDirect

Acta Biomaterialia journal homepage: www.elsevier.com/locate/actbio

Full length article

Graft alignment impacts the regenerative response of skeletal muscle after volumetric muscle loss in a rat model John Kim a, Ben Kasukonis a, Kevin Roberts a,b, Grady Dunlap a, Lemuel Brown b, Tyrone Washington b, Jeffrey Wolchok a,∗ a

Department of Biomedical Engineering, College of Engineering, University of Arkansas, Fayetteville, AR, United States Department of Health, Human Performance, and Recreation, College of Education and Health Professions, University of Arkansas, Fayetteville, AR, United States

b

a r t i c l e

i n f o

Article history: Received 6 August 2019 Revised 15 January 2020 Accepted 16 January 2020 Available online xxx

a b s t r a c t A key event in the etiology of volumetric muscle loss (VML) injury is the bulk loss of structural cues provided by the underlying extracellular matrix (ECM). To re-establish the lost cues, there is broad consensus within the literature supporting the utilization of implantable scaffolding. However, while scaffold based regenerative medicine strategies have shown potential, there remains a significant amount of outcome variability observed across the field. We suggest that an overlooked source of outcome variability is differences in scaffolding architecture. The goal of this study was to test the hypothesis that implant alignment has a significant impact on genotypic and phenotypic outcomes following the repair of VML injuries. Using a rat VML model, outcomes across three autograft implant treatment groups (aligned implants, 45° misaligned, and 90° misaligned) and two recovery time points (2 weeks and 12 weeks) were examined (n = 6–8/group). At 2 weeks post-repair there were no significant differences in muscle mass and torque recovery between the treatment groups, however we did observe a significant upregulation of MyoD (2.5 fold increase) and Pax7 (2 fold increase) gene expression as well as the presence of immature myofibers at the implant site for those animals repaired with aligned autografts. By 12 weeks post-repair, functional and structural differences between the treatment groups could be detected. Aligned autografts had significantly greater mass and torque recovery (77 ± 10% of normal) when compared to 45° and 90° misaligned autografts (64 ± 10% and 61 ± 11%, respectively). Examination of tissue structure revealed extensive fibrosis and a significant increase in non-contractile tissue area fraction for only those animals treated using misaligned autografts. When taken together, the results suggest that implant graft orientation has a significant impact on in-vivo outcomes and indicate that the effect of graft alignment on muscle phenotype may be mediated through genotypic changes to myogenesis and fibrosis at the site of injury and repair. Statement of Significance A key event in the etiology of volumetric muscle loss injury is the bulk loss of architectural cues provided by the underlying extracellular matrix. To re-establish the lost cues, there is broad consensus within the literature supporting the utilization of implantable scaffolding. Yet, although native muscle is a highly organized tissue with network and cellular alignment in the direction of contraction, there is little evidence within the field concerning the importance of re-establishing native architectural alignment. The results of this study suggest that critical interactions exist between implant and native muscle alignment cues during healing, which influence the balance between myogenesis and fibrosis. Specifically, it appears that alignment of implant architectural cues with native muscle cues is necessary to create a pro-myogenic environment and contractile force recovery. The results also suggest that misaligned cues may be pathological, leading to fibrosis and poor contractile force recovery. © 2020 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.



Corresponding author at: Department of Biomedical Engineering, University of Arkansas, John A. White, Jr. Engineering Hall, Suite 120, Fayetteville, AR. E-mail address: [email protected] (J. Wolchok).

https://doi.org/10.1016/j.actbio.2020.01.024 1742-7061/© 2020 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article as: J. Kim, B. Kasukonis and K. Roberts et al., Graft alignment impacts the regenerative response of skeletal muscle after volumetric muscle loss in a rat model, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2020.01.024

JID: ACTBIO 2

ARTICLE IN PRESS

[m5G;February 1, 2020;7:37]

J. Kim, B. Kasukonis and K. Roberts et al. / Acta Biomaterialia xxx (xxxx) xxx

1. Introduction

2. Methods

When provided the appropriate regenerative cues, skeletal muscle has a robust capacity for repair. Following mild muscle damage (e.g. strains, contusions, and lacerations) cells are damaged but the underlying extracellular matrix (ECM) is largely intact and regeneration at the injury site is robust [1,2]. However, when significant muscle volume is lost (trauma, infection, or surgical resection) the cues provided by the ECM are missing, and the injury site is instead replaced with non-contractile scar tissue [3]. Termed volumetric muscle loss (VML), the bulk loss of skeletal muscle tissue overwhelms the body’s capacity for regeneration, resulting in a permanent functional deficit [3–6]. Current soft tissue repair techniques and traditional rehabilitation have not been able to reverse the pathological changes that occur following VML injury. At present there are unfortunately no surgical guidelines that effectively address the treatment of VML injuries. The differential response to mild versus severe muscle damage (regeneration vs. scarring) suggests a key event in the etiology of VML injury is the bulk loss of architectural cues provided by the ECM. To re-establish these lost cues, a number of groups including our own, have been exploring the development of “off the shelf” scaffolding materials for the treatment of VML injury [7–16]. Native skeletal muscle is a hierarchical tissue defined by an ECM network and myofiber cellular population aligned to the direction of contractile force production [17,18]. Though the importance of restoring architectural cues as a means of enhancing post VML in-vivo muscle healing has been discussed in the literature [10,14,19], there has been scant in-vivo examination. It has been well-documented that cellular behavior is influenced by topographical features [20]. Of these features, alignment is one of the most studied patterns in-vitro, and evidence suggests it promotes a high degree of muscle cell alignment and enhances their fusion into myofibers [21,22]. Studies have also demonstrated that mechanical stimulation of aligned myoblasts acts as a cell cycle regulator and influences the expression of myogenic regulatory factors [23,24]. Despite strong in-vitro evidence suggesting an important role for alignment during muscle regeneration, VML scaffolding architecture has differed widely between in-vivo studies, ranging from gels and unaligned tissue ECM networks [25– 28] to aligned fiber scaffolds and muscle derived tissue implants [10,29–31]. We recognize that differences in compositional cues also play an important role in determining cell fate [32], which is likely why ECM scaffolds receive the majority of research attention. Yet the role of alignment cues as a driver of VML repair site myogenesis is poorly understood, and addressing this gap in knowledge provides a valuable opportunity to advance the field. Towards this end, this study was designed to test the hypothesis that restoration of architectural cues congruent to the direction of muscle contraction are necessary to trigger a pro-myogenic environment with contractile recovery following VML injury repair. Additionally, we examined whether cue misalignment impedes myogenesis and potentially contributes to the fibrotic wound healing response that is observed in response to untreated VML injury. To examine the role of cue congruency we utilized a muscle plug autograft repair strategy in which autograft implant alignment could be systematically varied in relation to the surrounding muscle contractile direction. Using this model the effect of graft alignment and misalignment on short-term genotypic changes (2 weeks post-repair) and longer-term (12 weeks) functional and structural recovery following VML repair was examined using a small animal (rat) in-vivo model.

2.1. Animal implantation Fischer 344 rats (Harlan, Indianapolis, IN), weighing approximately 300–325 g were used as the animal model for all implantation studies. All animal procedures were performed in accordance with protocols approved by the University of Arkansas Institutional Animal Care and Use Committee (AUP# 19044) and guided by published methods [33,34]. Anesthesia was induced using isoflurane (4%) in oxygen. The implant site was surgically exposed through a 1–2 cm incision running parallel to the tibia. The tibialis anterior muscle (TA) was identified and a sterile surgical pen was used to draw an orientation line marking the native alignment of the TA myofibers. A partial thickness circular VML defect (8 mm diameter x 3 mm deep) was created using a sterile biopsy punch (Fig. 1A–D). The muscle tissue removed from the defect site was weighed (Average defect weight = 93.4 mg). Muscle defect mass values (20% of TA mass) were based on pilot study TA muscle mass measurements (average TA mass = 470 ± 17 mg). Animals were randomly assigned to one of three treatment groups. Treatment 1 (Aligned 0°): VML defects repaired using an autograft aligned to the surrounding TA muscle. Treatment 2 (Misaligned 45°): VML defect repaired using a misaligned autograft rotated 45° with respect to the surrounding tissue. Treatment 3 (Misaligned 90°): VML defect repaired using a misaligned autograft rotated 90° with respect to the surrounding tissue. Muscle defect plugs were immediately implanted in the defect site and sutured in place with 6–0 polypropylene sutures (Fig. 1E) (Redilene, MYCO Medical, Cary, NC). The contralateral limb was left uninjured to serve as an internal comparative control. The deeper fascia and surface skin layers were separately closed using an interrupted stich with a 5–0 absorbable suture (Vicryl, Ethicon, Summerville, MA). A single surgeon performed all implantation procedures. Post-operative analgesia consisted of 0.1 mg/kg buprenorphine administered subcutaneously via injection twice daily for two days. Access to anti-inflammatory medication (Carprofen) via a dietary gel cup (Medigel CPF, ClearH2O, Westbrook, ME) was made available to each cage for one week following surgery. Animals were individually housed in standardsized rat cages with unrestricted movement and were allowed to bear weight on the operative extremity as tolerated. All animals were housed for a 2-week (n = 6/treatment group) or 12-week (n = 8/treatment group) recovery period. 2.2. Isometric torque measurement At the completion of the assigned 2- or 12-week recovery period, the peak tetanic contractile torque was measured in-vivo and guided by published methods [35]. Animals were anesthetized and the lower limb was stabilized at 90° of knee flexion (tibia parallel to the benchtop) using a custom-made alignment jig. The ankle was flexed to 90° and the foot was secured to the lever arm of a commercial muscle physiology system (Aurora Scientific, Ontario, Canada,). To reduce the contribution of the TA during torque measurement, distal tenotomies were performed on the extensor digitorum longus (EDL) and extensor hallucis longus (EHL). TA peak isometric tetanic torque was measured by stimulating the peroneal nerve with the aid of a physiological stimulator (Grass; S88). Optimal voltage (2–5 V) was determined using a series of tetanic contractions (150 Hz, 0.1 ms pulse width, 400 ms train).

Please cite this article as: J. Kim, B. Kasukonis and K. Roberts et al., Graft alignment impacts the regenerative response of skeletal muscle after volumetric muscle loss in a rat model, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2020.01.024

JID: ACTBIO

ARTICLE IN PRESS

[m5G;February 1, 2020;7:37]

J. Kim, B. Kasukonis and K. Roberts et al. / Acta Biomaterialia xxx (xxxx) xxx

3

(Aurora Scientific) as an additional measure of TA muscle function/kinetics. At the conclusion of electrophysiological testing, hind limb muscles were harvested and all animals were euthanized through a two-part procedure: 4% isoflurane to induce anesthesia followed by immediate carbon dioxide inhalation in accordance with guidelines provided by the 2013 AVMA Panel on Euthanasia of Animals. 2.3. Tissue histology Repaired and contralateral uninjured TA muscles were harvested and flash frozen in liquid nitrogen-chilled isopentane. Frozen TA tissue samples (n = 4/experimental group) were sectioned (8 μm) transversely through the mid-portion of the defect site with the aid of a cryostat. Sections were blocked in PBS containing 4% goat serum and 0.05% sodium azide for 1 h at room temperature prior to incubation in primary antibodies including monoclonal (IgG1 ) mouse-anti-collagen I (1:500, Sigma-Aldrich), rabbit polyclonal anti-collagen III (1:10 0 0, Abcam), and mouse monocloncal (IgG2B ) anti-myosin heavy chain (MF-20, 1:10, Developmental Studies Hybridoma Bank, Iowa City, IA) for 2 h at room temperature. Following PBS washes, slides were incubated in the appropriate fluorescently-labeled secondary antibodies (AlexaFluor, 1:500, Life Technologies) for 30 min at room temperature. Additional tissue sections were stained using a commercial Masson’s Trichrome kit or hematoxylin and eosin (H&E) following the manufacturer’s guidelines (Sigma-Aldrich). All sections were mounted onto microscope slides and digitally imaged with the aid of a microscope (Nikon Ci-L). 2.4. Image analysis

Fig. 1. (Surgical Procedure and Growth): Graphical schematic of surgical procedure (A). A ~1.5 cm incision parallel to tibia was made in the left hind limb of Fischer 344 rats (B). The skin and fascia was dissected to reveal the TA and an alignment mark was drawn to indicate the direction of myofiber alignment (C). An 8 mm biopsy punch was used to create a circular VML defect in the TA to a depth of 3 mm (D). The defect muscle plug was removed, weighed, and re-implanted either aligned to the surrounding muscle (0°), rotated 45°, or rotated 90° with respect to the surrounding muscle (90° misaligned pictured) (E). The autograft was sutured in place using polypropylene sutures to maintain muscle plug orientation. The wound was closed using double layer closure of the fascia and skin utilizing absorbable sutures with interrupted stitches. Animal growth rate scatter plots with group means ± SD are presented for each treatment group (0, 45, and 90° graft rotation) at both 2 weeks and 12 weeks (F). Data was analyzed using a one-way ANOVA with Tukey’s post-hoc; n = 6/group at 2 weeks and n = 8/group at 12 weeks.

Raw peak tetanic contractile torque (N mm) was recorded from both the treated and contralateral control limbs of each animal. Peak tetanic torque for each animal was determined using the average of 5 contractions. All contractions were separated by one minute of rest. Peak torque data for each treatment group (0°, 45°, or 90°) was normalized to animal weight (N mm/kg) and also expressed relative to the uninjured contralateral control limb (%uninjured). Contraction time and relaxation time was also calculated (seconds) from torque recordings using supplier software

Muscle fibrosis and non-contractile tissue area was quantified within normal and treated muscle tissue sections at the site of repair using measures of collagen I (Sigma-Aldrich, St. Louis, MO) immunoreactivity with myosin heavy chain as a counter stain. Representative tissue sections were imaged (100X), converted to 8-bit greyscale, and a uniform threshold was applied across all samples to isolate collagen type I-positive tissue regions from the surrounding tissue within each section. From each image, tissue immunoreactivity to collagen I as a percentage of total tissue area was calculated with the aid of image analysis software (ImageJ, NIH) and is reported in the results as percent non-contractile tissue (% NCT). Similar image analysis methods were used to measure fiber cross-sectional area (μm2 ) from magnified (100X) collagen III immunostained images. The tissue region bounded by collagen III immunoreactivity was used to calculate the area of individual fiber cross sections. Approximately 50–100 fibers were captured and measured within each magnified image. Three non-consecutive sections were imaged per animal. Representative tissue sections collected from four animals per treatment group were used for all %NCT and fiber area calculations. A total of twelve images (4 animals x 3 sections/animal) were analyzed for each treatment group. In addition to the quantification of %NCT and fiber size, the fiber size frequency distribution was calculated using bins in increments of 500 μm2 . Lastly, sections were qualitatively evaluated for myofiber formation and organization as well as the appearance and size of VML site repair tissue for comparison between aligned and misaligned treatment groups. 2.5. Gene expression Real-time PCR was performed using the protocol described in Washington et al. [36]. Tissue samples (approximately 30 mg) collected from the defect/repair site (n = 4/sample group/time point) were homogenized with Trizol (Ambion, Carlsbad, CA), chloroform

Please cite this article as: J. Kim, B. Kasukonis and K. Roberts et al., Graft alignment impacts the regenerative response of skeletal muscle after volumetric muscle loss in a rat model, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2020.01.024

JID: ACTBIO 4

ARTICLE IN PRESS

[m5G;February 1, 2020;7:37]

J. Kim, B. Kasukonis and K. Roberts et al. / Acta Biomaterialia xxx (xxxx) xxx

(Sigma Aldrich, St. Louis, MO), and treated with DNase (Invitrogen, Carlsbad, CA). RNA was extracted using the RNeasy kit (Invitrogen, Carlsbad, CA). RNA concentration and purity was determined by UV spectrophotometry. RNA with a 260-to-280-nm ratio ≥ 1.8 was used for subsequent analysis. cDNA was reverse transcribed from 1 μg of total RNA using the Superscript Vilo cDNA synthesis kit (Life Technologies, Carlsbad, CA, USA). cDNA was amplified in a 25 μL reaction containing appropriate primer pairs and TaqMan Universal Mastermix (Applied Biosystems, Grand Island, NY). Commercially available TAQMAN primers (Invitrogen, Carlsbad, CA) for MyoD, Pax7, Collagen I, Collagen III, TGF-β 1, TIMP1, Nestin, Robo2, VEGF, and CD31, with 18 s ribosomal housekeeping were used to quantify the expression of desired myogenic, angiogenic, axonal, matrix, and matrix regulatory genes. Experimental group samples were normalized to 18 s and then referenced to the contralateral normal limb. Gene expression levels are reported as fold change using the 2−(Ct) method. The full transcriptome of a representative 2-week aligned and 90° misaligned tissue samples was analyzed using RNA-Seq. The 2week time point was selected in order to capture early post-repair transcriptional activity. Total RNA was isolated using the Purelink RNA Mini Kit (ThermoFisher). RNA concentration was determined by nanodrop spectrophotometry, with RNA quality (28S/18 s > 2, RIN >7) confirmed using a fragment analyzer (Advanced Analytical). cDNA libraries were sequenced on the NextSeq500 platform (Illumina) to a mean depth of 20 million 75 bp reads per library. RNA sequencing reads were mapped to the Rattus norvegicus genome (RGSC build 6.0) using Tophat 2.1, followed by quantification of reads and analysis of differential expression in Cufflinks 2.2. 2.6. Data analysis All data are presented as scatter plots with mean with standard deviation unless noted. Data were tested for normality using the Shapiro–Wilks Test. The effect of treatment (aligned, 45° misaligned, and 90° misaligned) or time points (2 and 12 weeks) on each of the outcome measures (peak torque, muscle mass, fiber size, %NCT, and gene expression) was evaluated using ANOVA. Post hoc comparisons were made using Tukey’s HSD. A Chi-square test was used to compare fiber area frequency histograms. Comparisons between 2 and 12 week time points were conducted using a Student’s t-test. All data analyses were performed using commercial statistical analysis software (Prism 8 and JMP 13). A standard p < 0.05 level of significance was used for all statistical tests. 3. Results 3.1. Growth rate All treatment groups tolerated the implantation surgery well. At one-week post-implantation, all animals were fully ambulatory with no visually discernable gait differences between groups and reached the assigned 2- or 12-week study endpoints without complications. All animals gained weight during the study period. Aligned 0° repair animals gained an average of 7.6 ± 0.3 g/week over the 12-week recovery period. Misaligned 45° and 90° animals gained 7.4 ± 0.6 g/week and 6.6 ± 0.3 g/weeks respectively. There was no significant main effect of graft alignment on animal growth rate at either the 2-week (p = 0.79) or 12-week (p = 0.08) time point (Fig. 1F). 3.2. Peak isometric torque At both 2 and 12 weeks following repair, contractile torque recordings for all TA muscles were characterized by a sharp rise in torque at the initiation of stimulation, followed by a stable and

Fig. 2. (Electrophysiology): Representative in vivo isometric tetanic torque waveforms (A: 150 Hz; 400 msec train; 0.1 msec pulse width). Animal weight normalized (Nmm/kg) and relative (% uninjured) peak torque values were calculated for each animal tested. Scatter plots of weight normalized and relative peak torque data at 2 (B) and 12 weeks (C) are presented for each treatment group (0, 45, and 90° graft rotation). Twelve-week muscle contraction time and relaxation time (sec) are also presented (D). Data was analyzed using a one-way ANOVA with Tukey’s post-hoc; n = 6/group at 2 weeks and n = 8/group at 12 weeks. ∗ (p < <0.05) distinguish significant differences.

detectable plateau during peak torque production, finishing with a rapid return to a no-torque resting state at the completion of stimulation (Fig. 2A). At 2 weeks post-repair we did not detect a significant main effect of alignment on either weight normalized (p = 0.58) or %uninjured (p = 0.74) peak contractile torque values (Fig. 2B). Aligned 0°, misaligned 45°, and misaligned 90° peak TA torque fell to 48 ± 8%, 51 ± 11%, and 53 ± 11% of contralateral

Please cite this article as: J. Kim, B. Kasukonis and K. Roberts et al., Graft alignment impacts the regenerative response of skeletal muscle after volumetric muscle loss in a rat model, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2020.01.024

JID: ACTBIO

ARTICLE IN PRESS

[m5G;February 1, 2020;7:37]

J. Kim, B. Kasukonis and K. Roberts et al. / Acta Biomaterialia xxx (xxxx) xxx

5

uninjured values respectively. Alternatively, at 12 weeks postrepair a significant main effect of alignment to increase both the weight normalized (p = 0.02) and %uninjured (p = 0.01) peak torque was detected. Aligned 0° % uninjured peak torque values reached 77 ± 10% at 12 weeks post-repair, while misaligned 45° and misaligned 90° reached 64 ± 10% and 61 ± 11% of uninjured mean values respectively. The difference at 12 weeks between the aligned 0° repair group and both misaligned repair groups was significant (45°; p = 0.04 and 90°; p = 0.01). No significant differences (p = 0.85) were detected at 12 weeks between the misaligned 45° and misaligned 90° groups (Fig. 2C). Comparison between 2 week and 12 week peak torque data revealed a similar main effect of alignment to increase recovery (p < <0.001). At 12 weeks, the aligned 0° repair group had on average recovered 60% of the peak torque lost at 2-weeks (recovery to 77% from 48%). Misaligned 45° and 90° repair recovery from 2-week values were more modest, recovering 25%, and 15% of the lost torque respectively. The difference in peak torque between 2 and 12 weeks was significant for the aligned repair (p = 0.001), but was not significant for either of the misaligned repair groups (45°; p = 0.37 and 90°; p = 0.71). We did not detect a significant effect of alignment on contraction time (p = 0.35). The effect of aligned repair to decrease relaxation time approached, but did not reach, significance (p = 0.07) (Fig. 2D) 3.3. TA histology and mass Uninjured contralateral TA muscles across each repair group were consistently characterized by a distinctive teardrop morphology, with a distal to proximal length (ankle to knee) that was approximately twice the medial to lateral width. The anterior surface was unremarkable, with no discernable variations or disruptions in surface appearance. TA muscles harvested at 2 weeks following repair had observable morphological features that differed between the treatment groups (Fig. 3A and B). Muscles repaired using misaligned 90° autografts exhibited notable graft degeneration within the repair site. Alternatively, both aligned 0° and misaligned 45° autograft repairs retained their volume and remained well approximated to the surrounding normal muscle tissue, although the site of repair was easily discerned from the surrounding normal tissue. At 12 weeks post-repair, we saw no evidence of implant site infection or gross deformity at the treatment site for any treatment group. TA muscles repaired with either aligned 0° or misaligned 45° autografts appeared to be less atrophied than misaligned 90° repairs, and better resembled the distinctive teardrop shape of normal TA muscles. Across all treatment groups, the VML repair site was less discernable from the surrounding tissue when compared to the appearance at 2 weeks (Fig. 3C). Similar to the response detected within the peak torque data, the main effect of alignment on TA mass was not significant at 2 weeks (p = 0.75), but did reach significance (p = 0.03) at 12 weeks (Fig. 3D). At 12 weeks, aligned 0° repair TA muscle mass was on average 6% and 11% larger than the misaligned 45° and 90° repair groups respectively, although only the difference between the aligned 0° and misaligned 90° rose to the level of significance (p = 0.02). Comparison between the 2- and 12-week time points revealed varying levels of TA mass recovery for each group. Aligned 0° repair % uninjured TA mass increased from 81 ± 5% at 2 weeks post repair to 93 ± 6% at 12 weeks, a recovery of 63%, which is similar to the level of recovery detected for peak torque. The increase in % uninjured TA mass between 2 and 12 weeks was significant (p = 0.006) for the aligned 0° repair group. Alternatively, TA mass recovery (2 weeks versus 12 weeks) in response to misaligned repair was not significant for either the 45° (p = 0.12) or 90° (p = 0.25) repair group. Mean misaligned 45° TA mass increased from 83 ± 9% at 2-weeks to 88 ± 5% at 12 weeks, while

Fig. 3. (TA Morphology and Mass): A representative VML repair site for each of the graft orientation treatment groups is shown intra-operatively (t = 0), illustrating alignment (0°) and misalignment (45°, or 90°) of the autograft muscle plug following excision and re-implantation (A). The repaired anterior surface of representative TA muscles shown at both 2 weeks (B) and 12 weeks (C) following VML repair. The repair sites with sutures (black arrows) located at the periphery remain distinguishable at both timepoints. Scatter plots of relative TA mass (% uninjured) for all treatment groups (0, 45, and 90° graft rotation). at both 2 weeks and 12 weeks is presented (D). Data was analyzed using a one-way ANOVA with Tukey’s post-hoc; n = 6/group at 2 weeks and n = 8/group at 12 weeks. ∗ (p < <0.05) distinguish significant differences. Scale bar = 1 cm.

mean misaligned 90° TA mass increased from 79 ± 10% at 2-weeks to 84 ± 7%, recoveries of approximately 25% for both groups. 3.4. 2-Week repair site histology At 2 weeks post-implantation, Masson’s Trichrome staining revealed detectable collagen deposition at the site of repair for all treatment groups when compared to uninjured contralateral TA muscle sections (Fig. 4A–D). Within the VML repair site, MHC+ myofiber islands surrounded by collagen I immuno-reactive regions were observed for each of the treatment groups (Fig. 4A–D). However, the aligned 0° repair site tissue had notably larger myofiber islands and less collagen I immunoreactivity

Please cite this article as: J. Kim, B. Kasukonis and K. Roberts et al., Graft alignment impacts the regenerative response of skeletal muscle after volumetric muscle loss in a rat model, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2020.01.024

JID: ACTBIO 6

ARTICLE IN PRESS

[m5G;February 1, 2020;7:37]

J. Kim, B. Kasukonis and K. Roberts et al. / Acta Biomaterialia xxx (xxxx) xxx

Fig. 4. (2-week VML Site Histology): Representative sections of the VML repair site of the uninjured (A), aligned 0° (B), misaligned 45° (C), and misaligned 90° (D) groups at 2 weeks. Sections stained were stained with Masson’s Trichrome and immunostained against collagen I (green) and counterstained against MHC (red). Scale bar = 100 μm unless noted. Arrow indicates anterior direction. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3.5. Gene expression At 2 weeks post-VML repair, muscle tissue collected from the aligned 0° repair site revealed a statistically significant increase in the expression of both of Pax7 and MyoD compared to uninjured controls, with fold changes of 1.8 ± 0.3 and 2.7 ± 0.8, respectively. Alternatively, the expression of Pax7 and MyoD was not significantly increased when measured in tissue samples collected from either the misaligned 45° or misaligned 90° repair sites (Fig. 5A). A significant decrease in the VEGF (p = 0.001) gene expression, but not CD31 (p = 0.3) was detected for each of the aligned and misaligned treatment groups. VEGF expression was on average decreased 3–4 fold compared to uninjured normal tissue expression across all repair groups. We did not detect significant differences in gene expression for either of the axon related genes examined (Nestin; p = 0.9 and Robo2; p = 0.98). VML repair site tissue collected at 2 weeks post-implantation revealed a significant main effect of repair to increase the expression of several key ECM and ECM regulatory genes (Fig. 5B). Significantly elevated expression levels for collagen I (p = 0.03) were detected within each of the VML repair groups when compared to uninjured normal tissue. Aligned, misaligned 45°, and misaligned 90° collagen I expression was approximately 10-fold higher than uninjured muscle tissue expression. While collagen I expression

was elevated within each of the repair groups, no significant differences were detected between the aligned 0° and either misaligned repair group. Similarly, collagen III (p = 0.01) and TGF-β 1 (p = 0.001) expression was not significantly different between the repair groups, but was significantly elevated within each of the repair groups when compared to uninjured. The increase in collagen III expression ranged from a high of 9-fold for samples collected from aligned 0° repair site tissue to 7 fold increases for both the misaligned 45° and misaligned 90° repair groups. Mean TGF-β 1 expression levels within the aligned and misaligned repair groups were lower (3–4 fold increase compared to uninjured) than those detected for either collagen I or III. Lastly, TIMP-1 expression was similarly increased (p = 0.01) within each repair group when compared to uninjured, yet did not significantly vary between the repair groups. When examined at 12 weeks, both aligned and misaligned repair group gene expression values were not significantly different from uninjured tissue values for any of the genes examined, suggesting a return to baseline levels (Fig. 5C and D). 3.6. RNA-Seq Transcriptome profiling of aligned and misaligned 90° defects at 2 weeks post-injury demonstrated a cumulative total 104 unique differentially expressed genes (DEGs) across both groups relative

Please cite this article as: J. Kim, B. Kasukonis and K. Roberts et al., Graft alignment impacts the regenerative response of skeletal muscle after volumetric muscle loss in a rat model, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2020.01.024

JID: ACTBIO

ARTICLE IN PRESS

[m5G;February 1, 2020;7:37]

J. Kim, B. Kasukonis and K. Roberts et al. / Acta Biomaterialia xxx (xxxx) xxx

7

Fig. 5. (Gene Expression): 2 week (A and B) and 12 week (C and D) expression of myogenic (Pax7, and MyoD), angiogenic (VEGF and CD31), axonal (Nestin and Robo2), ECM (Col I and Col III), and ECM regulatory (TGF-β 1, and TIMP1) genes were assessed via RT-qPCR using muscle tissue harvested at the repair site. Scatter plots with group means + SD are presented for each treatment group (uninjured, 0, 45, and 90° graft rotation). Data was analyzed using one-way ANOVA with Tukey’s post-hoc; n = 3–4/group. ∗ (p < <0.05) and ∗ ∗ (p < <0.01) distinguish significant differences.

to uninjured contralateral muscles (Fig. 6A). 23% of these DEGs are uniquely expressed in the aligned group while 47% are unique to the 90° group, with 26% of DEGs being shared between the two (Fig. 6B). The full list of DEG’s is presented as supplementary data (Supplementary Table 1). Gene set enrichment analysis against the Gene Ontology Biological Process database indicated that anatomical structure development, negative regulation of biological process, and regulation of response to stimulus were the top three overrepresented categories for both the aligned and 90° groups. The query did not return any results related to skeletal muscle development, inflammation, or neurogenesis. Ingenuity Pathway Analysis revealed that Inhibition of Matrix Metalloproteases (MMPs) and Fibrosis signaling pathways are significantly upregulated in both aligned and 90° groups. The aligned and 90° groups shared the expression of various MMPs (MMP9, MMP13) and collagen (COL5A1, COL18A1) transcripts related to both the inhibition of MMP and fibrosis pathways. However, only the aligned group exhibited an overrepresentation of genes related to axon guidance signaling (Fig. 6C). Specifically, sequences for EPHA3 (p = 0.01), MMP10 (p = 0.01), MMP13 (p = 0.01), MMP9 (p = 0.01), RHOD (p = 0.003), SEMA3A (p = 0.01), and TUBB6 (p = 0.01) were found to be significantly upregulated in the aligned group as it relates to axon guidance signaling. 3.7. 12-Week repair site histology At 12 weeks, the VML repair site was less discernable from the surrounding TA muscle tissue as compared to its appearance at 2 weeks post-repair (Fig. 7A–C). Masson’s Trichrome and H&E stains where characterized by tissue healing with extensive fibrosis at the site of repair for both of the misaligned repair groups. Alternatively, fibrosis was not apparent within sections prepared from aligned group repair sites. Collagen immunoreac-

tivity (I and III) examined at the site of repair revealed structural differences between the aligned and misaligned repair groups (Fig. 8A and B). Repair site collagen I and III immunoreactivity was markedly elevated for both the 45° and 90° misaligned repair groups, but was qualitatively similar to uninjured TA muscle when examined within aligned repair tissue sections. Mean %NCT values were significantly elevated within both 45° (p = 0.005) and 90° (p = 0.02) misaligned repair groups when compared to uninjured TA muscle tissue (Fig. 8C). Alternatively, aligned repair site mean %NCT values were not significantly different (p = 0.99) from uninjured tissue values. The mean fiber diameter frequency distribution was not significantly different (p = 0.35) between the treatment groups, although a significant increase in the frequency of moderately sized fibers (size range = 10 0 0–1250um2 ) was detected within the 90° misaligned frequency data when compared to normal muscle (Fig. 8D and E). 4. Discussion The results from this study suggest that graft orientation has a significant effect on contractile torque recovery following repair of VML injury. Study findings provide significant in-vivo insights regarding the potential importance of re-establishing native muscle alignment cues to improve post VML repair torque recovery. The effect of graft orientation on torque recovery appears to be mediated through the balance of myogenesis and fibrosis during healing at the site of injury and repair. Specifically, the implantation of aligned grafts was associated with the increased expression of genetic markers of satellite cell activation (Pax7) and myogenesis (MyoD) during the early stages of muscle healing. While at later timepoints, aligned repair resulted in significantly less fibrosis at the VML repair site. These findings in the presence of graft alignment cues are consistent with recent in vitro findings utilizing

Please cite this article as: J. Kim, B. Kasukonis and K. Roberts et al., Graft alignment impacts the regenerative response of skeletal muscle after volumetric muscle loss in a rat model, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2020.01.024

JID: ACTBIO 8

ARTICLE IN PRESS

[m5G;February 1, 2020;7:37]

J. Kim, B. Kasukonis and K. Roberts et al. / Acta Biomaterialia xxx (xxxx) xxx

Fig. 6. (RNASeq): Visualization of differentially expressed gene intersections across the aligned and 90° misaligned groups (A) with the top canonical pathways for each group identified by Ingenuity Pathway Analysis (B). All differential expression data was calculated relative to uninjured controls.

aligned substrates to promote myogenesis [37-41] and provide new evidence suggesting that orientation cues provided by the graft during healing may play a positive role in the balance between myogenesis and fibrosis during VML healing. This finding is important because the role of scaffold architecture as a driver of muscle regeneration has yet to be explored in-depth. We believe that sensitivity to differences in scaffolding architecture may explain the outcome variability that has been observed within the field [34,42–47]. We acknowledge that the inclusion of gene expression data for late myogenic differentiation markers such as myogenin and myosin heavy chain would strengthen the supposition that scaffold alignment improved myogenesis and should be considered in future investigations. Furthermore, we recognize that the inclusion of an unrepaired VML control group would have allowed for a direct comparison between repaired and unrepaired recovery outcomes. However, previous studies from our group [34,48] using this VML model (20% loss of TA mass) have reported a 40–50% loss of muscle strength when the injury remains unrepaired. We believe this prior data provides an acceptable reference for assessment of the aligned and misaligned VML treatment strategies explored in this study. Comparison to that data suggests that aligned graft repair restored approximately half of the muscle strength lost to VML in-

jury, while misaligned repair recovery was similar to unrepaired VML outcomes. In this study we examined the role of implant alignment cues through manipulation of muscle autograft orientation with respect to the surrounding skeletal muscle tissue. While these findings are intended to provide practical insights to help guide the development of engineered VML repair scaffolding, we chose to utilize an autograft repair model as the VML repair strategy for this study in order to precisely restore the TA muscle’s native pennate alignment. We recognize that the autograft VML repair has notable differences from a scaffold repair strategy, which limits our interpretation of results. Yet in the end, we believe the autograft strategy had several experimental strengths that motivated its application and outweighed the concerns during this initial set of experiments. Namely, we were apprehensive about the level of orientation consistency that could be achieved using the decellularized skeletal muscle scaffold examined previously by our group [13]. Furthermore, the degree of consistency between material composition and cellular delivery for the autograft implant would have been difficult to obtain with a more traditional regenerative medicine approach utilizing a cell-seeded scaffold. We surmised that if we observed no benefit from the restoration of graft orientation in the autograft VML repair model we were unlikely to detect an effect using a scaffolding approach. The detection of an effect with the autograft implant motivates further examination of scaffolding alignment cues using clinically relevant state of the field VML implants including anisotropic scaffolds [49–51] with orientation cues similar to skeletal muscle [52], as well as isotropic scaffolds, such as porcine sub-intestinal submucosa and bladder wall tissue [53,54]. The contractile torque values measured from the misaligned graft groups were similar to that which was previously reported by our group [34] for unrepaired VML defects of the same size (20% of TA weight) and geometry (8 mm diameter by 3 mm deep). This finding was unexpected and notable because it suggests that outcome expectations following VML repair using misaligned grafts may be no better than untreated injuries. While we hypothesized that the implantation of aligned grafts into the VML injury site would better enhance contractile torque recovery when compared to misaligned grafts, we still expected the implantation of misaligned graft to result in at least some amount of torque recovery. This a priori expectation was based on the positive VML repair results reported by the Corona group using a minced muscle repair strategy [55]. The autogenic minced muscle implants explored by Corona contain compositional and cellular cues that are similar to the intact autografts we employed in this study, except the mincing of the muscle to create the implantable muscle paste largely eliminates the long-range alignment cues that are present within the intact graft. Comparable composition combined with deficient alignment cues would suggest similar outcome expectations for both misaligned grafts and minced muscle paste VML repair strategies, yet this was not observed. We suggest that the combination of Corona’s findings with our results could be interpreted as suggesting that the absence of alignment cues may be preferable to the delivery of misaligned cues. We suspect that the implantation of misaligned scaffolds at the VML injury sites creates a significant barrier to cell migration and subsequent muscle, nerve, and vessel regeneration. A recent study by Webster et al. utilizing 3D intravital imaging to visualize muscle regeneration in live mice revealed that extracellular matrix remnants from injured myofibers governed myogenic progenitor cell behavior [56]. Namely, the “ghost fibers” facilitated progenitor cell migration and proliferation within the natively aligned matrix but re-orienting the ghost fibers negatively impacted migration and resulted in disorganized de novo fiber formation. Our findings within the in-vivo VML wound healing environment suggests that alignment cues need to be utilized

Please cite this article as: J. Kim, B. Kasukonis and K. Roberts et al., Graft alignment impacts the regenerative response of skeletal muscle after volumetric muscle loss in a rat model, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2020.01.024

JID: ACTBIO

ARTICLE IN PRESS

[m5G;February 1, 2020;7:37]

J. Kim, B. Kasukonis and K. Roberts et al. / Acta Biomaterialia xxx (xxxx) xxx

9

Fig. 7. (12-week VML Site Histology): Representative sections of aligned 0° (A), misaligned 45° (B), and misaligned 90° (C) groups at 12 weeks post-repair. Transverse sections were stained with Masson’s Trichrome and hematoxylin and eosin. Anterior edge of the sections are indicated by a black arrow. Scale bar = 100 μm unless noted. Centrally located nuclei are indicated by black arrows, Intramuscular connective tissue bands are indicated by white arrows.

with caution as the restoration of misaligned orientation cues has potentially pathological consequences. While we did not detect any significant differences between the treatment groups for the angiogenic or axonal genes examined in this study, suggesting a lack of genotypic differences across the treatment groups, we cannot rule out the potential role of residual vessels and nerves as a factor in the improved recovery observed in the aligned repair group. Autograft implants were coarsely secured to the surrounding TA muscle at four suture sites, thereby re-establishing a mechanical linkage but not an intimate approximation of the implant borders to the surrounding muscle tissue. Yet while the small vessels and nerves were not surgically reconnected, one might expect that regeneration originating from the surrounding skeletal tissue would have an increased likelihood of reestablishing contact with residual vessel and nerve counterparts within the aligned autograft implants as compared to ro-

tated grafts, resulting in improved innervation and perfusion of the aligned grafts. In the future, the effect of graft alignment on broader measures of muscle tissue regeneration, including repair site cell infiltration (satellite, endothelial and Schwann cells), vessel and nerve regeneration, and graft integration with the surrounding tissue should be probed to provide a more comprehensive regeneration picture. In particular, the histological examination of peripheral nerve regeneration could address the conflicting findings we observed in the rt-PCR and RNA-Seq findings. While it is not possible to fully recapitulate the complex cellular and chemical milieu of the autograft muscle plugs examined in this study, we believe the findings are still relevant to the design of engineered scaffolds. Specifically, our findings suggest that muscle regeneration could be enhanced through the development of multi-modal scaffolds that recapitulate not only the ECM alignment cues that enhance myogenesis, but also ECM alignment cues

Please cite this article as: J. Kim, B. Kasukonis and K. Roberts et al., Graft alignment impacts the regenerative response of skeletal muscle after volumetric muscle loss in a rat model, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2020.01.024

JID: ACTBIO 10

ARTICLE IN PRESS

[m5G;February 1, 2020;7:37]

J. Kim, B. Kasukonis and K. Roberts et al. / Acta Biomaterialia xxx (xxxx) xxx

Fig. 8. (12-week Morphometric Measurements): Magnified images (100X) of transverse tissue sections from the VML repair site of uninjured, aligned, misaligned 45°, and misaligned 90° groups immunostained for collagen I (green) and counterstained with MHC (red) (A) and collagen III (red) (B). Sections were quantified for collagen I area fraction (C), fiber cross-sectional area (D), and fiber area frequency distribution (E). Scatter plots with group means ± SD are presented for each treatment group (0, 45, and 90° graft rotation). Collagen I area fraction and fiber cross sectional area data was analyzed using a one-way ANOVA with Tukey’s post-hoc. Fiber area frequency distribution data was analyzed using the Chi-square test. n = 4/group. ∗ (p < <0.05) and ∗ ∗ (p < <0.01) distinguish significant differences. Scale bar = 100 μm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

that direct nerve and vessel ECM regrowth. Scaffolds derived from native tissues would be expected to preserve these cues, while engineered scaffolds could be printed or spun using fibers that differentially target muscle, nerve, and vessel tissue [57–61]. Towards this end, recent work from Nakayama et al. has demonstrated that it is possible to develop engineered constructs to promote regeneration in multiple tissue types [44]. Specifically, engineered muscle tissues composed of aligned nanofibrillar collagen scaffolds populated by C2C12 myoblasts and endothelial cells were found to significantly improve scaffold integration with the host tissue, microvasculature formation, and vascular perfusion following implan-

tation. As cellular filopodia are highly sensitive to nanotopographical features which regulate their guidance mechanisms [62] as well as the activation of mechanically-induced signaling pathways [63], the group leveraged this in vitro to achieve highly aligned and contractile myotubes prior to implantation. Adding to the value of the aligned scaffolds, after implantation an increase in the secretion of myo-angiogenic cytokines (angiogenin, IGFBP-3, and VEGFA) as well as myoblast differentiation and fusion promoting factors were detected. Also, when compared to randomly-oriented scaffolds, the aligned scaffolds recapitulated the orientation of native skeletal muscle and capillary organization to a higher degree [44].

Please cite this article as: J. Kim, B. Kasukonis and K. Roberts et al., Graft alignment impacts the regenerative response of skeletal muscle after volumetric muscle loss in a rat model, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2020.01.024

ARTICLE IN PRESS

JID: ACTBIO

J. Kim, B. Kasukonis and K. Roberts et al. / Acta Biomaterialia xxx (xxxx) xxx

These findings highlight the importance of scaffold alignment in promoting regeneration that more resembles the native processes of skeletal muscle as well as the promising potential to target other tissues such as nerves with similar co-culture methodologies employed in the study. The use of a freshly excised autograft plug as the implant material was used to not only examine recovery sensitivity to graft orientation but was also expected to provide unique insights into the limits of implant material performance. We hypothesized that re-implantation of the very tissue removed from the VML injury site would provide an ideal implant material. It is difficult to imagine an engineered replacement material with better structural, compositional, and cellular components than the tissue removed from the site of injury. The recovery outcomes using injury site tissue autografts were informative and concerning. Immediate re-implantation of the tissue plug was capable of restoring on average half of the contractile torque lost to VML injury. The results could be interpreted to suggest that tissue engineering strategies that focus on scaffold and cellular implantation alone may have a recovery limit. In fact, recovery of half the torque lost to injury is consistent with that observed across differing repair strategies and models, further suggesting that VML recovery limits may exist. Opinion leaders in the field have suggested that a potential barrier to the current muscle regenerative medicine schemes is the poor rate of vascularization and innervation of regenerating muscle tissue [16,64], an observation that is consistent with our gene expression findings. Moving forward, VML regenerative medicine strategies may need to consider the introduction of wound healing cues that accelerate angiogenesis and neurogenesis [65-68]. 5. Conclusions The key findings of this study suggest that: 1. VML injury repair with aligned autografts resulted in a significant recovery of peak tetanic torque when compared to defects repaired using misaligned autografts. 2. At two weeks post repair, the expression of key pro-myogenic genes (MyoD & Pax7) was significantly elevated within aligned autograft repair site tissue, while myogenic gene expression in the misaligned repair site tissue was unchanged. 3. At two weeks post-repair, tissue collected from both aligned and misaligned treatment groups had increased expression levels for several key ECM and ECM regulatory genes (Col I, Col III, TGF-β , MMP2). 4. Repair using misaligned grafts resulted in prominent fibrosis and a significant increase in the presence of non-contractile tissue. Declaration of Competing Interest No competing financial interests exist for any of the authors. Acknowledgments Research reported in this publication was supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health under Award Number 1R15AR064481, 1R15AR073492, and the Arkansas Biosciences Institute. Supplementary materials Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.actbio.2020.01.024.

[m5G;February 1, 2020;7:37] 11

References [1] M. Hill, A. Wernig, G. Goldspink, Muscle satellite (stem) cell activation during local tissue injury and repair, J. Anat. 203 (1) (2003) 89–99. [2] A. Mauro, Satellite cell of skeletal muscle fibers, J. Biophys. Biochem. Cytol. 9 (1961) 493–495. [3] N. Terada, S. Takayama, H. Yamada, T. Seki, Muscle repair after a transsection injury with development of a gap: an experimental study in rats, Scand. J. Plast. Reconstr. Surg. Hand Surg. 35 (3) (2001) 233–238. [4] A. Aurora, J.L. Roe, B.T. Corona, T.J. Walters, An acellular biologic scaffold does not regenerate appreciable de novo muscle tissue in rat models of volumetric muscle loss injury, Biomaterials 67 (2015) 393–407. [5] B.T. Corona, J.C. Wenke, C.L. Ward, Pathophysiology of volumetric muscle loss injury, Cells Tissues Organs 202 (3–4) (2016) 180–188. [6] B.J. Hurtgen, C.L. Ward, K. Garg, B.E. Pollot, S.M. Goldman, T.O. McKinley, J.C. Wenke, B.T. Corona, Severe muscle trauma triggers heightened and prolonged local musculoskeletal inflammation and impairs adjacent tibia fracture healing, J. Musculoskelet. Neuronal. Interact. 16 (2) (2016) 122–134. [7] V.J. Mase Jr., J.R. Hsu, S.E. Wolf, J.C. Wenke, D.G. Baer, J. Owens, S.F. Badylak, T.J. Walters, Clinical application of an acellular biologic scaffold for surgical repair of a large, traumatic quadriceps femoris muscle defect, Orthopedics 33 (7) (2010) 511. [8] B.M. Sicari, J.P. Rubin, C.L. Dearth, M.T. Wolf, F. Ambrosio, M. Boninger, N.J. Turner, D.J. Weber, T.W. Simpson, A. Wyse, E.H. Brown, J.L. Dziki, L.E. Fisher, S. Brown, S.F. Badylak, An acellular biologic scaffold promotes skeletal muscle formation in mice and humans with volumetric muscle loss, Sci. Transl. Med. 6 (234) (2014) 234ra58. [9] J.E. Valentin, N.J. Turner, T.W. Gilbert, S.F. Badylak, Functional skeletal muscle formation with a biologic scaffold, Biomaterials 31 (29) (2010) 7475–7484. [10] M.T. Wolf, K.A. Daly, J.E. Reing, S.F. Badylak, Biologic scaffold composed of skeletal muscle extracellular matrix, Biomaterials 33 (10) (2012) 2916–2925. [11] L. Hou, C. Gong, Y. Zhu, In vitro construction and in vivo regeneration of esophageal bilamellar muscle tissue, J. Biomater. Appl. 30 (9) (2016) 1373–1384. [12] Z. Shen, S. Guo, D. Ye, J. Chen, C. Kang, S. Qiu, D. Lu, Q. Li, K. Xu, J. Lv, Y. Zhu, Skeletal muscle regeneration on protein-grafted and microchannel-patterned scaffold for hypopharyngeal tissue engineering, Biomed. Res. Int. 2013 (2013) 146953. [13] B.M. Kasukonis, J.T. Kim, T.A. Washington, J.C. Wolchok, Development of an infusion bioreactor for the accelerated preparation of decellularized skeletal muscle scaffolds, Biotechnol Prog 32 (3) (2016) 745–755. [14] E.K. Merritt, D.W. Hammers, M. Tierney, L.J. Suggs, T.J. Walters, R.P. Farrar, Functional assessment of skeletal muscle regeneration utilizing homologous extracellular matrix as scaffolding, Tissue Eng. Part A 16 (4) (2010) 1395–1405. [15] E.K. Merritt, M.V. Cannon, D.W. Hammers, L.N. Le, R. Gokhale, A. Sarathy, T.J. Song, M.T. Tierney, L.J. Suggs, T.J. Walters, R.P. Farrar, Repair of traumatic skeletal muscle injury with bone-marrow-derived mesenchymal stem cells seeded on extracellular matrix, Tissue Eng. Part A 16 (9) (2010) 2871– 2881. [16] M. Quarta, M. Cromie, R. Chacon, J. Blonigan, V. Garcia, I. Akimenko, M. Hamer, P. Paine, M. Stok, J.B. Shrager, T.A. Rando, Bioengineered constructs combined with exercise enhance stem cell-mediated treatment of volumetric muscle loss, Nat. Commun. 8 (2017) 15613. [17] L. Teodori, A. Costa, R. Marzio, B. Perniconi, D. Coletti, S. Adamo, B. Gupta, A. Tarnok, Native extracellular matrix: a new scaffolding platform for repair of damaged muscle, Front. Physiol. 5 (2014) 218. [18] C. Fuoco, L.L. Petrilli, S. Cannata, C. Gargioli, Matrix scaffolding for stem cell guidance toward skeletal muscle tissue engineering, J. Orthop. Surg. Res. 11 (1) (2016) 86. [19] B. Perniconi, A. Costa, P. Aulino, L. Teodori, S. Adamo, D. Coletti, The pro-myogenic environment provided by whole organ scale acellular scaffolds from skeletal muscle, Biomaterials 32 (31) (2011) 7870–7882. [20] A.S. Curtis, The mechanism of adhesion of cells to glass. A study by interference reflection microscopy, J. Cell Biol. 20 (1964) 199–215. [21] D.J. Evans, S. Britland, P.M. Wigmore, Differential response of fetal and neonatal myoblasts to topographical guidance cues in vitro, Dev. Genes Evol. 209 (7) (1999) 438–442. [22] P. Clark, D. Coles, M. Peckham, Preferential adhesion to and survival on patterned laminin organizes myogenesis in vitro, Exp. Cell Res. 230 (2) (1997) 275–283. [23] M. Brosig, J. Ferralli, L. Gelman, M. Chiquet, R. Chiquet-Ehrismann, Interfering with the connection between the nucleus and the cytoskeleton affects nuclear rotation, mechanotransduction and myogenesis, Int. J. Biochem. Cell Biol. 42 (10) (2010) 1717–1728. [24] W. Kuang, J. Tan, Y. Duan, J. Duan, W. Wang, F. Jin, Z. Jin, X. Yuan, Y. Liu, Cyclic stretch induced miR-146a upregulation delays C2C12 myogenic differentiation through inhibition of Numb, Biochem. Biophys. Res. Commun. 378 (2) (2009) 259–263. [25] A.S. Salimath, A.J. Garcia, Biofunctional hydrogels for skeletal muscle constructs, J. Tissue Eng. Regen. Med. 10 (11) (2016) 967–976. [26] Y. Morimoto, M. Kato-Negishi, H. Onoe, S. Takeuchi, Three-dimensional neuron-muscle constructs with neuromuscular junctions, Biomaterials 34 (37) (2013) 9413–9419. [27] M. Li, C.E. Dickinson, E.B. Finkelstein, C.M. Neville, C.A. Sundback, The role of fibroblasts in self-assembled skeletal muscle, Tissue Eng. Part A 17 (21–22) (2011) 2641–2650.

Please cite this article as: J. Kim, B. Kasukonis and K. Roberts et al., Graft alignment impacts the regenerative response of skeletal muscle after volumetric muscle loss in a rat model, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2020.01.024

JID: ACTBIO 12

ARTICLE IN PRESS

[m5G;February 1, 2020;7:37]

J. Kim, B. Kasukonis and K. Roberts et al. / Acta Biomaterialia xxx (xxxx) xxx

[28] D.W. van der Schaft, A.C. van Spreeuwel, H.C. van Assen, F.P. Baaijens, Mechanoregulation of vascularization in aligned tissue-engineered muscle: a role for vascular endothelial growth factor, Tissue Eng. Part A 17 (21–22) (2011) 2857–2865. [29] T.L. Jenkins, D. Little, Synthetic scaffolds for musculoskeletal tissue engineering: cellular responses to fiber parameters, NPJ Regen. Med. 4 (2019) 15. [30] E.M. Reece, S.N. Oishi, M. Ezaki, Brachioradialis flap for coverage after elbow flexion contracture release, Tech. Hand Up Extrem. Surg. 14 (2) (2010) 125–128. [31] T. Okano, S. Satoh, T. Oka, T. Matsuda, Tissue engineering of skeletal muscle. Highly dense, highly oriented hybrid muscular tissues biomimicking native tissues, ASAIO J 43 (5) (1997) M749–M753. [32] M. De Lisio, T. Jensen, R.A. Sukiennik, H.D. Huntsman, M.D. Boppart, Substrate and strain alter the muscle-derived mesenchymal stem cell secretome to promote myogenesis, Stem Cell Res. Ther. 5 (3) (2014) 74. [33] X. Wu, B.T. Corona, X. Chen, T.J. Walters, A standardized rat model of volumetric muscle loss injury for the development of tissue engineering therapies, Biores. Open Access 1 (6) (2012) 280–290. [34] B. Kasukonis, J. Kim, L. Brown, J. Jones, S. Ahmadi, T. Washington, J. Wolchok, Codelivery of infusion decellularized skeletal muscle with minced muscle autografts improved recovery from volumetric muscle loss injury in a rat model, Tissue Eng. Part A 22 (19–20) (2016) 1151–1163. [35] E.L. Mintz, J.A. Passipieri, D.Y. Lovell, G.J. Christ, Applications of in vivo functional testing of the rat tibialis anterior for evaluating tissue engineered skeletal muscle repair, J. Vis. Exp. (116) (2016). [36] T.A. Washington, J.P. White, J.M. Davis, L.B. Wilson, L.L. Lowe, S. Sato, J.A. Carson, Skeletal muscle mass recovery from atrophy in IL-6 knockout mice, Acta Physiol. (Oxf) 202 (4) (2011) 657–669. [37] M. Costantini, S. Testa, P. Mozetic, A. Barbetta, C. Fuoco, E. Fornetti, F. Tamiro, S. Bernardini, J. Jaroszewicz, W. Swieszkowski, M. Trombetta, L. Castagnoli, D. Seliktar, P. Garstecki, G. Cesareni, S. Cannata, A. Rainer, C. Gargioli, Microfluidic-enhanced 3D bioprinting of aligned myoblast-laden hydrogels leads to functionally organized myofibers in vitro and in vivo, Biomaterials 131 (2017) 98–110. [38] Y. Guo, J. Gilbert-Honick, S.M. Somers, H.Q. Mao, W.L. Grayson, Modified cell-electrospinning for 3D myogenesis of C2C12s in aligned fibrin microfiber bundles, Biochem. Biophys. Res. Commun. 516 (2) (2019) 558– 564. [39] S.H. Cha, H.J. Lee, W.G. Koh, Study of myoblast differentiation using multi-dimensional scaffolds consisting of nano and micropatterns, Biomater. Res. 21 (2017) 1. [40] H. Hasmad, M.R. Yusof, Z.R. Mohd Razi, R.B. Hj Idrus, S.R. Chowdhury, Human amniotic membrane with aligned electrospun fiber as scaffold for aligned tissue regeneration, Tissue Eng. Part C Methods 24 (6) (2018) 368–378. [41] D. Browe, J. Freeman, Optimizing C2C12 myoblast differentiation using polycaprolactone-polypyrrole copolymer scaffolds, J. Biomed. Mater. Res. A 107 (1) (2019) 220–231. [42] B.N. Brown, S.F. Badylak, Extracellular matrix as an inductive scaffold for functional tissue reconstruction, Transl. Res. 163 (4) (2014) 268–285. [43] I.T. Swinehart, S.F. Badylak, Extracellular matrix bioscaffolds in tissue remodeling and morphogenesis, Dev. Dyn. 245 (3) (2016) 351–360. [44] K.H. Nakayama, M. Quarta, P. Paine, C. Alcazar, I. Karakikes, V. Garcia, O.J. Abilez, N.S. Calvo, C.S. Simmons, T.A. Rando, N.F. Huang, Treatment of volumetric muscle loss in mice using nanofibrillar scaffolds enhances vascular organization and integration, Commun. Biol. 2 (2019) 170. [45] K.H. Patel, A.J. Dunn, M. Talovic, G.J. Haas, M. Marcinczyk, H. Elmashhady, E.G. Kalaf, S.A. Sell, K. Garg, Aligned nanofibers of decellularized muscle ECM support myogenic activity in primary satellite cells in vitro, Biomed. Mater. 14 (3) (2019) 035010. [46] M.J. McClure, D.J. Cohen, A.N. Ramey, C.B. Bivens, S. Mallu, J.E. Isaacs, E. Imming, Y.C. Huang, M. Sunwoo, Z. Schwartz, B.D. Boyan, Decellularized muscle supports new muscle fibers and improves function following volumetric injury, Tissue Eng. Part A 24 (15–16) (2018) 1228–1241. [47] M.A. Machingal, B.T. Corona, T.J. Walters, V. Kesireddy, C.N. Koval, A. Dannahower, W. Zhao, J.J. Yoo, G.J. Christ, A tissue-engineered muscle repair construct for functional restoration of an irrecoverable muscle injury in a murine model, Tissue Eng. Part A 17 (17–18) (2011) 2291–2303.

[48] J. Kim, B. Kasukonis, G. Dunlap, R. Perry, T. Washington, J. Wolchok, Regenerative repair of volumetric muscle loss injury is sensitive to age, Tissue Eng. Part A (2019). [49] S. Jana, A. Cooper, M. Zhang, Chitosan scaffolds with unidirectional microtubular pores for large skeletal myotube generation, Adv. Healthc. Mater. 2 (4) (2013) 557–561. [50] V. Kroehne, I. Heschel, F. Schugner, D. Lasrich, J.W. Bartsch, H. Jockusch, Use of a novel collagen matrix with oriented pore structure for muscle cell differentiation in cell culture and in grafts, J. Cell Mol. Med. 12 (5A) (2008) 1640–1648. [51] H.W. Kang, S.J. Lee, I.K. Ko, C. Kengla, J.J. Yoo, A. Atala, A 3D bioprinting system to produce human-scale tissue constructs with structural integrity, Nat. Biotechnol. 34 (3) (2016) 312–319. [52] K. Wilson, A. Terlouw, K. Roberts, J.C. Wolchok, The characterization of decellularized human skeletal muscle as a blueprint for mimetic scaffolds, J. Mater. Sci. Mater. Med. 27 (8) (2016) 125. [53] N.J. Turner, J.S. Badylak, D.J. Weber, S.F. Badylak, Biologic scaffold remodeling in a dog model of complex musculoskeletal injury, J. Surg. Res. 176 (2) (2012) 490–502. [54] T.W. Gilbert, A. Nieponice, A.R. Spievack, J. Holcomb, S. Gilbert, S.F. Badylak, Repair of the thoracic wall with an extracellular matrix scaffold in a canine model, J. Surg. Res. 147 (1) (2008) 61–67. [55] C.L. Ward, B.E. Pollot, S.M. Goldman, S.M. Greising, J.C. Wenke, B.T. Corona, Autologous minced muscle grafts improve muscle strength in a porcine model of volumetric muscle loss injury, J. Orthop. Trauma 30 (12) (2016) e396–e403. [56] M.T. Webster, U. Manor, J. Lippincott-Schwartz, C.M. Fan, Intravital imaging reveals ghost fibers as architectural units guiding myogenic progenitors during regeneration, Cell Stem Cell 18 (2) (2016) 243–252. [57] E. Soliman, F. Bianchi, J.N. Sleigh, J.H. George, M.Z. Cader, Z. Cui, H. Ye, Engineered method for directional growth of muscle sheets on electrospun fibers, J. Biomed. Mater. Res. A 106 (5) (2018) 1165–1176. [58] D.G. Han, C.B. Ahn, J.H. Lee, Y. Hwang, J.H. Kim, K.Y. Park, J.W. Lee, K.H. Son, Optimization of electrospun poly(caprolactone) fiber diameter for vascular scaffolds to maximize smooth muscle cell infiltration and phenotype modulation, Polymers (Basel) 11 (4) (2019). [59] Y. Li, C. Liao, S.C. Tjong, Electrospun polyvinylidene fluoride-based fibrous scaffolds with piezoelectric characteristics for bone and neural tissue engineering, Nanomaterials (Basel) 9 (7) (2019). [60] S. Laternser, H. Keller, O. Leupin, M. Rausch, U. Graf-Hausner, M. Rimann, A novel microplate 3D bioprinting platform for the engineering of muscle and tendon tissues, SLAS Technol. 23 (6) (2018) 599–613. [61] C.B. Pinnock, E.M. Meier, N.N. Joshi, B. Wu, M.T. Lam, Customizable engineered blood vessels using 3D printed inserts, Methods 99 (2016) 20–27. [62] M.J. Dalby, M.O. Riehle, H. Johnstone, S. Affrossman, A.S. Curtis, Investigating the limits of filopodial sensing: a brief report using SEM to image the interaction between 10 nm high nano-topography and fibroblast filopodia, Cell Biol. Int. 28 (3) (2004) 229–236. [63] I.C. Liao, J.B. Liu, N. Bursac, K.W. Leong, Effect of electromechanical stimulation on the maturation of myotubes on aligned electrospun fibers, Cell Mol. Bioeng. 1 (2–3) (2008) 133–145. [64] B.T. Corona, B.E. Henderson, C.L. Ward, S.M. Greising, Contribution of minced muscle graft progenitor cells to muscle fiber formation after volumetric muscle loss injury in wild-type and immune deficient mice, Physiol. Rep. 5 (7) (2017). [65] B. Egan, J.R. Zierath, Exercise metabolism and the molecular regulation of skeletal muscle adaptation, Cell Metab. 17 (2) (2013) 162–184. [66] B.T. Corona, K.E. Flanagan, C.M. Brininger, S.M. Goldman, J.A. Call, S.M. Greising, Impact of volumetric muscle loss injury on persistent motoneuron axotomy, Muscle Nerve 57 (5) (2018) 799–807. [67] S.M. Greising, C.L. Dearth, B.T. Corona, Regenerative and rehabilitative medicine: a necessary synergy for functional recovery from volumetric muscle loss injury, Cells Tissues Organs 202 (3–4) (2016) 237–249. [68] J.A. Passipieri, G.J. Christ, The potential of combination therapeutics for more complete repair of volumetric muscle loss injuries: the role of exogenous growth factors and/or progenitor cells in implantable skeletal muscle tissue engineering technologies, Cells Tissues Organs 202 (3–4) (2016) 202–213.

Please cite this article as: J. Kim, B. Kasukonis and K. Roberts et al., Graft alignment impacts the regenerative response of skeletal muscle after volumetric muscle loss in a rat model, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2020.01.024