The Effect of Laminin-111 Hydrogels on Muscle Regeneration in a Murine Model of Injury

Abstract

Volumetric muscle loss (VML) is characterized by a critical loss of muscle tissue that is accompanied by severe functional impairment and often long-term disability. Clinical therapies currently used in the treatment of VML are ineffective at regenerating lost muscle and restoring function. In this study, we used a novel hydrogel composed of fibrinogen and laminin-111, to promote regeneration and recovery of VML-traumatized muscle. Our previous study showed that these hydrogels exhibit a fibrous structure and Young's modulus of ∼2 kPa, while supporting C₂C₁₂ myoblast proliferation, myogenic marker expression, and proregenerative growth factor secretion in vitro. In a murine model of VML injury, the implantation of these hydrogels showed significant improvements in muscle weights and heightened infiltration of endothelial, hematopoietic, and immune cells at 2 weeks postinjury compared to the untreated muscles. At 4 weeks postinjury, the hydrogel-treated muscle showed increased myogenic activity, acetylcholine receptor clustering, and induction of the anti-inflammatory M2-like macrophage phenotype. However, improvements in muscle weights and force production were not observed at 4 weeks. An adjunct therapy such as physical rehabilitation or codelivery of stem cells may be required for the effective treatment of VML. Overall, these results will inform and guide the development of a successful tissue engineering strategy for the regeneration of skeletal muscle following trauma.

Impact Statement

Extremity injuries make up the most common survivable injuries in vehicular accidents and modern military conflicts. A majority of these injuries involve volumetric muscle loss (VML). The potential for donor site morbidity may limit the clinical use of autologous muscle grafts for VML injuries. Treatments that can improve the regeneration of functional muscle tissue are critically needed to improve limb salvage and reduce the rate of delayed amputations. The development of a laminin-111-enriched fibrin hydrogel will offer a potentially transformative and “off-the-shelf” clinically relevant therapy for functional skeletal muscle regeneration.

Introduction

While skeletal muscle has high regenerative potential, severe trauma is beyond its endogenous capacity for repair. Volumetric muscle loss (VML) injuries are common in both civilian and military populations and are characterized by functional impairment and long-term disability.¹ In military conflicts, including Operation Iraqi Freedom and Operation Enduring Freedom, a majority of injuries were extremity injuries involving severe musculoskeletal defects, which largely contribute to chronic injury.^2,3 These injuries are expensive due to the loss of trained military personnel and treatment costs.²

In mild injuries, skeletal muscle has a remarkable capacity for regeneration due to the resident stem cells, termed satellite cells. These stem cells reside between the basal lamina and sarcolemma and remain quiescent until prompted into the cell cycle through overstimulation or injury.⁴ In response to injury or overstimulation, the regenerative process proceeds through three sequential, yet, overlapping steps: (1) inflammatory response, (2) satellite cell activation, differentiation, and fusion, and (3) myofiber maturation and remodeling.⁴

Upon injury, myofiber necrosis takes place and proinflammatory mediators are generated by these injured, necrotic cells or from extracellular sources, activating the inflammatory response.⁵ Neutrophil infiltration followed by the recruitment of proinflammatory M1-like phenotype macrophages with a timely transition to the proregenerative M2-like phenotype macrophages encompass the first step of the regenerative process.⁴ Asymmetrical cell division of satellite cells then allows for the regenerative process to occur via myogenic progenitor cells, while the remaining satellite cells can return to quiescence to maintain the pool of stem cells necessary for long-term muscle integrity.⁶ This process results in fully functional, regenerated tissue with a restored structure-function relationship.^6,7

In cases of extreme trauma such as VML, endogenous methods of repair are dysregulated, resulting in impaired regeneration of lost tissue. The loss of myofibers, satellite cells, and the basal lamina are the primary reasons behind insufficient repair during VML.^8,9 This loss, as well as persistent inflammation from unresolved M1 phenotype macrophages, exacerbate muscle dysfunction by causing increased tissue damage and deposition of fibrotic tissue at the wound site and reducing the regenerative potential of the muscle.¹⁰ The remaining healthy tissue becomes damaged over time due to constant overloading of the muscle, which leads to further fibrosis of the tissue.¹

VML injuries are challenging to treat due to the high rate of complications, comorbidities, limited clinical options available, and varying levels of severity in each case. Furthermore, a successful treatment would be required to support the migration, proliferation, and differentiation of various cell populations, which moderate the complex processes of myofiber synthesis, immunomodulation, angiogenesis, neurogenesis, and fibrotic tissue deposition.

Decellularized scaffolds have been a popular choice of material to study in the treatment of VML injuries. Extracellular matrix (ECM) implantation in animal models has been shown to recruit stem cell populations, including perivascular stem cells,^11,12 pluripotent adult progenitors (Sox2⁺),¹³ Sca1⁺ cells,^3,14 CD133⁺ progenitor cells,^11,12 and neural stem cells.¹⁵ However, a lack of satellite and myogenic progenitor cell recruitment often results in an inability to recover function of the muscle when ECM scaffolds are implanted.^3,16 When incorporating minced muscle grafts into a decellularized urinary bladder matrix (MicroMatrix™; ACell, Inc.), Goldman and Corona found that, after 8 weeks, there was little to no tissue regeneration but improved functional capacity likely due to functional fibrosis.¹⁷ Furthermore, there have been contrary reports of the effect of the MicroMatrix has on the immune response.^{10,14,18–22} The variation in laboratory decellularization protocols and processing techniques can result in heterogeneity that makes it difficult to compare results among published work.

By using a biomaterial that encourages muscle regeneration, it may be possible to improve the functional recovery of the muscle and thus improve patient quality of life, while reducing the burden of cost on the health care system. This study uses a laminin (LM)-111-enriched fibrin hydrogel to attempt to fill the need for an effective, off-the-shelf treatment for VML. Fibrin is an excellent substrate for myoblast activity and vascularization and has been researched for muscle regeneration as microthread,²³ electrospun,²⁴ and hydrogel^25,26 scaffolds. In models of muscle disease and injury, administration of an embryonic isoform of LM, LM-111, has resulted in increased myofiber area and number, α7 integrin expression, number of satellite cells, and force production, as well as an overall decrease in inflammation.^27–30

In our previous study, LM-111-enriched fibrin hydrogels were fabricated and tested in vitro as a potential off-the-shelf therapy for VML injuries.²⁶ This fibrin-based biomaterial is biodegradable and enriched with an embryonic isoform of LM to encourage the activity of satellite cells in the remaining musculature. The structural and mechanical properties of the hydrogels were also shown to be favorable for muscle repair. Our previous study showed that a fibrin hydrogel enriched with 450 μg/mL of LM-111 significantly impacted the proliferation, myogenic protein expression, and myokine secretion of C₂C₁₂ myoblasts.²⁶ In this study, the hydrogel treatment was translated to use in a novel murine model of VML to determine in vivo effect and efficacy.

Materials and Methods

Hydrogel fabrication

LM-enriched fibrin hydrogels were created by combining bovine fibrinogen (20 mg/mL), thrombin (20 U/mL), calcium chloride (20 mM), protease inhibitor (1: 100; Sigma-Aldrich) and murine LM-111 (450 μg/mL; Trevigen) as described previously.²⁶

Implantation of hydrogel into a murine VML model

All animal works were performed in accordance with the Animal Welfare Act, the Animal Welfare Regulations, and following the Guide for the Care and Use of Laboratory Animals. The procedures were approved by the Saint Louis University Institutional Animal Care and Use Committee.

Adult (14–16 weeks old) male C57BL6 mice were purchased from Charles Laboratory and housed in an Association for Assessment and Accreditation of Laboratory Animal Care International accredited vivarium. Mice were provided with food and water ad libitum. Animals were weighed before surgery and anesthetized with 2.5% isoflurane in oxygen. Sustained release buprenorphine (1 mg/kg) was administered subcutaneously. The site of surgery was aseptically prepared. A lateral incision through the skin revealed the gastrocnemius-soleus (GAS) complex. A blunt probe was used to separate the GAS complex from surrounding muscles. A metal plate was inserted under the GAS complex, and a 3 mm biopsy punch was used to create a full thickness, ∼10% muscle mass defect. The injured muscles included both the medial and lateral heads of gastrocnemius muscle as well as the soleus muscle. Muscle biopsies were weighed for consistency (Table 1). A subset of animals was left untreated, while another subset received a 6 mm punch of the hydrogel. Hydrogels for implantation were created in a 48-well plate. Since hydrogels are water swollen structures, they tend to lose some water upon removal via a biopsy punch. Therefore, a 6 mm device was implanted to account for the shrinking of the gels. A final subset of animals was given a sham surgery as a control, where an incision was created and the muscle was blunt dissected away from the skin, but the GAS complex was left uninjured. Any bleeding was controlled by applying light pressure with a sterile swab, and the wound was closed with interrupted sutures. Animals were allowed to recover for 2 weeks or 1 month and euthanized via CO₂ asphyxiation. GAS muscles were collected, weighed, and processed appropriately for histological or biomolecular analyses (Table 1). At 2 weeks, three to six muscles per group were processed for histology and four to seven muscles were processed for westerns. At 4 weeks, eight muscles were processed for histology for untreated and hydrogel-treated groups, and n = 4–6 muscles were processed for westerns per group. A schematic representation of the experimental design can be seen in Figure 1.

Table 1.

Anatomical Measurements and Muscle Weights (Mean ± Standard Error of the Mean)

		Muscle weight (mg)		Body weight (g)
	Biopsy weight (mg)	2 weeks	4 weeks	2 weeks	4 weeks
Sham	n/a	170.1 ± 6.202	179.9 ± 4.214	30.2 ± 1.702	31.4 ± 1.105
Untreated	18.75 ± 0.547	147.5 ± 5.920	150.5 ± 2.677a	29.7 ± 1.277	30.3 ± 0.615
Hydrogel treated	18.56 ± 0.412	158.3 ± 10.535	153.7 ± 5.734	27.5 ± 2.436	30.6 ± 0.420

^a Significant difference compared to sham.

n/a, not applicable.

FIG. 1.

Schematic representation of the in vivo experimental design. The VML-injured hind limb muscles were either left untreated or implanted with laminin-111-enriched fibrin hydrogels. The sham group served as the control. The animals were allowed to recover, and muscle tissue was harvested at both 2 and 4 weeks. The peak isometric force was measured at 4 weeks. VML, volumetric muscle loss.

Histology

Upon collection, the GAS complex was weighed and frozen in 2-methyl butane (Fisher Scientific) that was supercooled in a liquid nitrogen bath. Cryosections (15 μm) of muscle cross-sections were stained with hematoxylin and eosin (H&E). Whole muscle H&E cross-sections were acquired (Olympus BX614S), and muscle fibers with centrally located nuclei (CLN) were counted manually from the entire muscle section (n = 3–6/group). The counts were normalized to the area of the muscle section that was manually measured using ImageJ.

Immunohistochemistry was performed on cross-sections using Collagen I (Abcam), MF20 (R&D Systems), F4/80 (Invitrogen), CD31 (R&D Systems), CD45 (Abcam), anti-LM γ1 (Millipore), LM (Abcam), and α-bungarotoxin Alexa Fluor 488 (Thermo Fisher). Appropriate fluorochrome-conjugated secondary antibodies (Invitrogen) were used as necessary. In brief, sections were fixed in ice-cold acetone, permeabilized in 0.1% Triton x-100 in phosphate-buffered saline (PBS), and blocked in 5% bovine serum albumin (BSA), 0.05% Tween-20, and 1:20 Fab Fragment in PBS. Primary antibodies were diluted 1:100 in PBS with 1% BSA and 0.1% Triton x-100 with the exception of LM (Abcam), which was diluted at 1:200 and CD31 (R&D systems) which were diluted at 1:50, overnight at 4°C. The appropriate fluorochrome-conjugated secondary antibodies were diluted 1:200 in PBS containing 1% BSA and 0.1% Triton x-100. Images were captured at 10 × and 20 × magnification using a Zeiss Axiocam microscope. All immunofluorescent images show the defect region. The defect site was identified by the high density of nuclei (DAPI⁺) and low density of myofibers. Quantitative analysis was performed using ImageJ as described previously.³ RGB channels were separated and then thresholded to remove background. The percentage area fraction of MHC (myosin heavy chain) were determined using nonoverlapping images of the defect sites in each muscle section in ImageJ.

Myofiber cross-sectional area measurements

Muscle sections stained with LM (Abcam) as described above were scanned to obtain a composite image of the entire muscle (Olympus BX614S). The fiber cross-sectional area (CSA) was analyzed in Fiji by converting the images to grayscale and thresholding to isolate individual muscle fibers. Fibers with a size range of 50–5000 μm² and circularity range of 0.125–1.0 were selected and sorted into categories of areas <500, 500–1000, >1000–2000, >2000–3000, >3000–4000, >4000–5000, and >5000 μm². The percentage of muscle fibers in each of these categories were determined based on the total number of fibers counted. The CSA data are reported as fold change over the CSA of the uninjured control (sham or cage control) muscles that served as control.

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis and western blotting

The harvested muscles were snap frozen in liquid nitrogen and homogenized on ice in RIPA lysis buffer with protease inhibitor cocktail (Sigma) for protein lysate isolation. Protein content was quantified using a Pierce BCA Protein Assay Kit (Thermo Scientific). Myogenic markers from protein lysates were quantified using western blotting as described previously.^3,31 In brief, equal amounts of reduced and denatured protein (60 μg) as determined from the bicinchoninic acid assay were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis using 4–20% gels (Bio-Rad). Protein was transferred onto nitrocellulose membranes with equal protein loading confirmed by Ponceau S staining. Membranes were probed using anti-MyoD (Millipore), anti-myogenin (Millipore), anti-α-actinin (Cell Signaling), antiheat shock protein-70 (anti-HSP-70; Cell Signaling), anti-inducible nitric oxide synthase (anti-iNOS; Abcam), anti-arginase (Cell Signaling), and anti-glyceraldehyde 3-phophate dehydrogenase (anti-GAPDH; Cell Signaling) primary antibodies as well as appropriate horseradish peroxidase-conjugated secondary antibodies. Quantification of band intensity was performed using ImageJ.

Muscle function testing

Peak isometric torque measurements (n = 8–11 muscles per group) of the hind limb posterior crural muscles were taken by the Aurora Scientific 3-in-1 whole animal system (Ontario, Canada). Mice were anesthetized under 2.5% isoflurane in oxygen. A small incision was made in the hind limb to sever the peroneal nerve. The incision was closed with Vetbond Tissue Adhesive (3M). The foot was secured to the foot pedal, and electrodes were inserted on either side of the sciatic nerve. A pin was used to secure the knee to minimize movement artifact. A custom-designed LabVIEW™ software was used. A twitch was induced to ensure proper placement of electrodes. Tetanic contractions (150 Hz, 0.3 ms pulse width, 330 ms train) were then elicited, and the average peak isometric torque was calculated from an average of three contractions. The peak isometric torque is reported as an absolute value as well as normalized to body weight.

Statistical analysis

Data are presented as a mean ± standard error of the mean. Analysis and graphing of data were performed using GraphPad Prism 6 for Windows. A one-way or two-way analysis of variance was used when appropriate to determine if there was a significant interaction or main effect between variables. A least significant difference post hoc comparison was performed to identify significance with p ≤ 0.05

Results

VML model analysis

Following 2 weeks of recovery, H&E and collagen and MHC staining were performed (Fig. 2A). A composite H&E-stained image of the muscle can be seen in Figure 2A-i and ii. The black box indicates the area of damage shown in higher magnification in the following H&E-stained panel (Fig. 2A-iii, iv). Collagen and MHC staining (Fig. 2A-v, vi) showed extensive damage and deposition of collagen in both treated and untreated groups. At 4 weeks, the damage persisted, and collagen bundles were still present in both groups (Fig. 2B-i–vi). The COL: MHC were significantly higher in the hydrogel-treated group compared with the uninjured control group at both 2 and 4 weeks (p = 0.0179 and p = 0.0321, respectively). There was no significant difference between the untreated group and either controls or hydrogel-treated group at either time point (Fig. 3A).

FIG. 2.

Untreated and hydrogel-treated muscles sections were stained with hematoxylin and eosin as well as collagen and MHC at both (A) 2 weeks and (B) 4 weeks after VML injury. The entire muscle section can be seen in (i, ii) for both 2- and 4-week time points. Images of the damaged area (indicated by the black box) can be seen in (iii, iv) at 10 × magnification. Collagen and MHC staining for untreated and hydrogel-treated muscles is shown in (v, vi) (n = 4–7 muscles/group) at 20 × magnification. MHC, myosin heavy chain. Color images available online at www.liebertpub.com/tea

FIG. 3.

(A) The muscle weights were normalized to body weights. VML injury resulted in significant deficits in muscle weights at both 2 and 4 weeks postinjury. A significant recovery in muscle weight was observed at 2 weeks postinjury with hydrogel treatment (n = 4–14 muscles/group). (B) The COL:MHC ratio was quantified from histological sections. No significant differences were observed at 2 or 4 weeks (n = 4–8 muscles/group). (C) The total number of myofibers with centrally located nuclei per mm² of muscle section significantly increased from 2 to 4 weeks in the VML-injured muscles (n = 4–8 muscles/group). (D) The mean CSA was determined for uninjured controls, untreated muscles, and hydrogel-treated muscles at 2 and 4 weeks. (E) A distribution of CSAs of fibers was determined and normalized to the sham at 2 weeks. ^#A difference from the 2-week time point of the same treatment and *a difference from the other treatment group at the same time point (n = 4–8 muscles/group). CSA, cross-sectional area. Color images available online at www.liebertpub.com/tea

The number of myofibers with CLN was manually counted from H&E-stained slides (Fig. 3C). There were no significant differences between untreated and hydrogel-treated groups at either time point, but the overall quantity of myofibers with CLN significantly increased at 4 weeks compared with 2 weeks (time effect, p = 0.0011).

Muscle weights of the GAS complex were recorded after animal euthanasia and normalized to body weight (Fig. 3A). At 2 weeks postinjury, untreated muscle weights were significantly lower than both the sham and hydrogel-treated muscle weights. At 4 weeks, sham muscle weights were significantly higher than the untreated muscles, but not the hydrogel-treated muscles.

The mean myofiber CSA was significantly lower in the untreated and hydrogel-treated muscles at 2 weeks postinjury (Fig. 3D). A distribution of myofiber CSA was determined at 2 and 4 weeks (Fig. 3E). Untreated muscles at 2 weeks showed a significantly higher percentage of small fibers (<500 μm²), compared with the 4-week untreated muscles (p < 0.0001). At 2 weeks, the hydrogel-treated muscles had a significantly lower quantity of small diameter fibers compared with the untreated muscles (p = 0.0353). There was a significant increase between 2 and 4 weeks in the percent of fibers with areas >2000–3000 μm² in untreated muscles (p = 0.0002) and hydrogel-treated muscles (p = 0.0114). There was a significant increase between 2 and 4 weeks in percent of fibers with areas >3000–4000 μm² in both untreated (p = 0.0003) and hydrogel-treated (p = 0.0071) muscles. There was a significant increase between 2 and 4 weeks in the percent of fibers with an area of >4000–5000 μm² in the untreated muscles (p = 0.0042), but not the hydrogel-treated muscles.

Muscle regeneration at 2 weeks postinjury

Figure 4 shows immunofluorescence staining performed at 2 weeks to determine the influx of endothelial cells (CD31⁺), hematopoietic cells (CD45⁺), and inflammatory cells (F4/80⁺). The hydrogel-treated group showed higher infiltration of endothelial, hematopoietic, and inflammatory compared with the untreated group.

FIG. 4.

Untreated and hydrogel-treated muscles sections (n = 4–7 muscles/group) obtained at 2 weeks were stained for the presence of (A) macrophages (F4/80⁺), (B) endothelial (CD31⁺) cells, (C) hematopoietic progenitors (CD45⁺), and (D) laminin gamma 1. *The remodeled hydrogel. Color images available online at www.liebertpub.com/tea

Skeletal muscle protein lysates from untreated and hydrogel-treated muscles were probed for myogenic and immunogenic markers at 2 weeks (Fig. 5). Representative bands of the proteins can be seen in Figure 5A. The myogenic markers MyoD, myogenin, and α-actinin were probed to determine the extent of regeneration in the muscle (Fig. 5B–D). The immunogenic markers HSP-70, iNOS, and arginase were also assessed to characterize the immunogenic response (Fig. 5E, F). HSP-70 was examined because it is essential in protecting cells from stress and plays a role in preparing cells for survival during challenges such as heightened immune responses.^32–34 The expression of iNOS and arginase was assessed to characterize the large number of F4/80⁺ macrophages that were seen infiltrating the defect site (Fig. 4).

FIG. 5.

Muscle protein lysates obtained at 2 weeks were analyzed via western blotting (n = 4–7 muscles/group). Representative bands of each protein can be seen (A). Myogenic markers such as (B) MyoD, (C) myogenin, and (D) α-actinin were quantified and normalized to GAPDH. Markers associated with immune response such as (E) HSP-70, (F) iNOS, and (G) arginase were also quantified and normalized to GAPDH. Hydrogel treatment resulted in significantly higher expression of myogenin and iNOS compared to the sham group. *p < 0.05. GAPDH, glyceraldehyde 3-phophate dehydrogenase; HSP, heat shock protein; iNOS, inducible nitric oxide synthase. Color images available online at www.liebertpub.com/tea

MyoD expression was significantly higher in both untreated and hydrogel-treated groups compared with the sham at 2 weeks postinjury (207.4% and 166.2% increase, respectively). Myogenin expression was significantly higher in hydrogel-treated muscles compared with the sham group (141.0% increase). No significant differences were present between groups in the expression of α-actinin or HSP-70. A significant increase in the expression of iNOS was observed between sham and hydrogel-treated groups at 2 weeks (350.6% increase). There were no significant differences in arginase levels at 2 weeks. No differences between the untreated and hydrogel-treated groups were found in expression of any proteins.

Muscle regeneration and recovery at 4 weeks postinjury

Four weeks after injury, protein lysates were run for the same myogenic and immunogenic panel of protein markers (Fig. 6). MyoD expression trended higher in hydrogel-treated muscles than in untreated muscles (p = 0.0510; 34.7% increase), but the difference was not significant (Fig. 6A). There were no differences between the expression of myogenin, α-actinin, HSP-70, or iNOS (Fig. 6B–E). Arginase expression was significantly higher in hydrogel-treated muscles than in sham muscles at 4 weeks postinjury with an increase of 922.4% (Fig. 6F).

FIG. 6.

Western blotting analysis of the protein lysates collected from the sham, untreated, and hydrogel-treated muscles at 4 weeks (n = 4–6/group). Myogenic markers such as (A) MyoD, (B) myogenin, and (C) α-actinin were quantified and normalized to GAPDH. Markers associated with immune response such as (D) HSP-70, (E) iNOS, and (F) arginase were also quantified and normalized to GAPDH. Hydrogel treatment resulted in significantly higher expression of arginase compared to the sham group, while the expression of MyoD trended higher (p = 0.0510). *p < 0.05. Color images available online at www.liebertpub.com/tea

At the 4 week time point, muscle sections were stained with α-bungarotoxin to assess the relative amount of acetylcholine receptor (AChR) clustering. It appears that the hydrogel-treated muscles showed more AChR clustering, suggesting that the fibers are primed for reinnervation (Fig. 7A).

FIG. 7.

Untreated and hydrogel-treated muscles sections obtained at 4 weeks were stained for the presence of (A) AchR clusters. The hydrogel-treated muscles showed increased AchR clustering. (B) Characteristic waveforms of the peak torque as obtained from MATLAB are shown. (C) Peak isometric torque was also measured at 4 weeks (n = 8–11 muscles/group). Significant deficits in force production were observed in response to VML injury in both untreated and hydrogel-treated muscles and deficits were maintained after normalizing force to the body weight. *p < 0.05. AchR, acetylcholine receptor. Color images available online at www.liebertpub.com/tea

Function testing performed 4 weeks postinjury showed that both untreated and hydrogel-treated muscles produced significantly lower peak isometric torque than the sham controls. There was no difference between the force produced by untreated and hydrogel-treated muscles. Characteristic waveforms of the peak isometric torque as acquired from MATLAB 8.5 can be seen in Figure 7B. Significant deficits were observed for peak isometric torque normalized to body weight (Fig. 7C). The deficit in force production between the sham and the untreated muscles was 22.6%, while the deficit between the sham and hydrogel-treated muscles was 17.7%.

Discussion

The implantation of LM-111-enriched fibrin hydrogels in a murine model of VML showed significant improvements in muscle weights and heightened infiltration of satellite, endothelial, hematopoietic, and immune cells at 2 weeks postinjury compared with the untreated muscles. The influx of these various cell types is important as it is often the early response that determines the success of the regenerative process.

We speculate that the LM-111 that was released from the hydrogels facilitated the increased cell recruitment. Several studies have found that other LMs in the basement membrane, especially LM-8 (α4ß1γ1), increase immune cell recruitment.^35,36

Since various cell populations are implicated in the repair and regeneration of skeletal muscle, increased cellular infiltration at the VML defect would suggest constructive remodeling.^3,37 Endothelial cells, for example, exert promitotic effects on progenitor cells and support the growth of myoblasts derived from satellite cells.³⁸ It has also been shown that CD45⁺ populations of hematopoietic cells can contribute nuclei to regenerating muscle fibers when injected intramuscularly to myofibers damaged by myotoxins.^39,40 Furthermore, signals from injured muscle can recruit hematopoietic cells and induce their progression down a myogenic lineage independently from Pax7, thus participating in muscle regeneration.⁴⁰ Macrophages, which infiltrate the site of injury within 24–72 h, secrete a multitude of growth factors and cytokines that can influence events such as myogenic activation, proliferation, and precursor cell fusion.^41,42

At 2 weeks postinjury, both untreated and hydrogel-treated groups had a significantly higher expression of MyoD compared with the sham group (Fig. 5B). Myogenin expression was similarly upregulated in both groups, although statistical significance was only observed with hydrogel treatment (Fig. 5C). At 4 weeks postinjury, it appears that there may be more myogenic activity occurring in the hydrogel-treated groups compared with the untreated as seen in the trending higher expression of MyoD (Fig. 6A).

Immunogenic markers were examined with western blotting as well (Fig. 5E, F). The expression of iNOS remained elevated at both 2 and 4 weeks postinjury, particularly in the hydrogel-treated group (Figs. 5F and 6E), suggesting a prolonged M1 macrophage response. At 4 weeks postinjury, the hydrogel-treated muscles showed significantly higher expression of arginase (Fig. 6F), suggesting a transition to a M2-like phenotype.^3,37 Therefore, at 4 weeks postinjury, a mixed macrophage phenotype response (consisting of both M1 and M2) is observed in the hydrogel-treated defect. It has been suggested that coordinated efforts by both pro-and anti-inflammatory macrophages may be required for effective muscle regeneration.^3,31 The presence of proinflammatory cells has been associated with recruitment and activation of myogenic precursor cells.⁴¹ For instance, while iNOS is considered to be a proinflammatory M1-like macrophage phenotype marker, it has also been shown to support the regenerative process by modulating inflammatory cell recruitment. The M1 phenotype macrophages which express that iNOS are able to “tag” debris with nitric oxide for removal by phagocytic cells.⁴³ It was found that in iNOS^−/− mice with muscle injuries, myogenic precursors did not proliferate or differentiate, suggesting that iNOS influences myogenesis.⁴⁴ These iNOS^−/− mice also expressed higher levels of chemokines, including macrophage inflammatory proteins and monocyte chemoattractant proteins, resulting in the persistence of neutrophils and macrophages at the injury site at later time points.⁴⁴ Another study found that applying a NOS inhibitor to decrease NO levels during muscle injury caused a decrease in muscle stem cells and increased collagen deposition.⁴⁵ Thus, even as a proinflammatory-associated marker, iNOS has a vital role in the skeletal muscle repair process.

The subsequent activation of anti-inflammatory cells eventually promotes the differentiation of satellite cells into myoblasts and supports myotube formation.⁴⁶ Therefore, it is possible that an increase in arginase expressing M2-like macrophages supported the increased myogenic activity indicated by a trend toward higher MyoD expression (Fig. 6A) at 4 weeks postinjury.

The fact that both untreated and hydrogel-treated animals produced significantly less torque than the animals subjected to a sham surgery highlights the severity of VML injuries. It has been found in several other preclinical models that VML defects that seem small (10–20% mass loss) disproportionately affect deficits in muscle strength (30–90%).^{1–3,47–49} Thus, even a 10% VML defect resulted in persistent deficits in muscle weights (Fig. 4A) and function, as evidenced in the 22.6% torque deficit in the untreated muscles and the 17.7% deficit in the hydrogel-treated muscles that is still observed at 4 weeks postinjury (Fig. 7C and Table 2). The lack of functional recovery would suggest an extensive loss of myofibers and contractile machinery.^50,51 In addition, the ratio of collagen:MHC remained constant over 4 weeks, and the number of myofibers with CLN increased with time (Fig. 3). Overall, these results would suggest sustained damage and ongoing regenerative events at the site of VML injury,⁵² which are characteristic of a traumatic muscle injury.

Table 2.

Functional Deficits at 4 weeks Postinjury

	Functional deficit (%)	Functional deficit normalized to body weight (%)
Sham vs. untreated	23.83	22.58
Sham vs. hydrogel treated	16.55	17.73

The mean CSA decreased significantly in the VML-injured muscles at 2 weeks but not at 4 weeks (Fig. 3D). Fiber size distribution shows an increase in the percentage of small diameter fibers (<500 μm²) at 2 weeks in the VML-injured muscles. The small diameter fibers could indicate both regenerating fibers and fibers undergoing atrophy due to damage and loss of innervation. The significant increase in the percentage of large diameter fibers (>2000 μm²) is likely due to the compensatory hypertrophy of the remaining muscle mass by 4 weeks.

Despite the appearance that hydrogel-treated muscle was better prepared for innervation and potentially muscle contraction due to higher clustering of AChRs, muscle function measurements did not show a significant difference in torque production between untreated and hydrogel-treated animals (Fig. 7B, C). AChRs cluster at the motor end plate and couple with ACh released from motor neurons. Clustering of AChRs indicates that a fiber is ready to be innervated by the motor neuron; it is a necessary step in the process of reinnervation following injury. It has been shown that LM-111 is able to induce the clustering of AChRs on rat myotubes at concentrations as low as 12 nM.⁵³ The LM-induced aneural clustering of AChRs has also been correlated with an increase in synaptogenesis, resulting in neuromuscular junction formation.⁵⁴ Since this study showed that AChR clustering was higher in LM-111-enriched fibrin hydrogel-treated muscles, yet did not translate into increased force production, it may be that other cues were lacking in the system or that the full reinnervation process may take several more weeks to be completed.⁵⁵ Reinnervation is an important and extensive process, which, if not accomplished successfully, can lead to denervation-induced atrophy, fibrosis, and fatty tissue infiltration in muscle.⁵⁶

We are only aware of one other study where LM-111 was delivered in a hydrogel to the site of VML injury. Goldman et al. implanted a hydrogel of 1% w/v thiol-modified hyaluronan, 1% polyethylene glycol diacrylate, and 50 μg/mL LM-111 to a VML-injured muscle; however, the hydrogel did not degrade which caused little to no cellular infiltration and resulted in impaired regeneration and functional recovery.⁴⁸ In this study, the LM-111-enriched fibrin hydrogels supported cellular infiltration, myogenic protein expression, and limited myofiber regeneration, but did not result in functional recovery. Our results were typical of several other studies that utilized acellular scaffolds to treat VML injuries and were unable to recover strength.^57,58 We propose that a completely acellular therapy may not be an effective approach to treat a VML injury due to the bulk loss of cells. Since the use of stem cell therapies is largely limited by availability, isolation and expansion time, survivability, and migration/regenerative potential following transplantation, cellular therapies are often combined with biomaterial scaffolds to improve outcomes in the treatment of VML.^59–61 Many different cell populations have been delivered in biomaterial scaffolds, including satellite cells or myoblasts,^{23–25,62–64} mesenchymal stem cells,^65–69 adipose-derived stem cells,^70–72 and even minced muscle grafts.^48,47,49 However, even though some of these therapies have shown improved function or reduced fibrosis, a successful, approved treatment for VML does not yet exist.

Several studies have had success using minced muscle autografts as part of their therapy to treat VML. Minced muscle grafts are bundles of single fibers that can be easily isolated using minimum technical training and expertise.³⁷ This treatment has resulted in improved muscle strength in both rodent^3,37 and porcine models.⁷³ It has also been shown that progenitor cells derived from the minced autograft directly contributed to fiber regeneration in a mouse tibialis anterior model of VML, promoted de novo muscle fiber regeneration, and reduced the appearance of chronic injury that was seen in the untreated animals.^37,74 A major disadvantage of using autologous muscle grafts includes the creation of a second injury where the patient is subject to donor site morbidity.

These challenges may be partially mitigated by volume expanding these minced autografts in the LM-enriched fibrin hydrogels. One study found that a 50% reduction in tissue used for minced autografts that were expanded in a collagen hydrogel had similar functional improvements to a 100% minced muscle graft.⁴⁷ However, the fibers regenerated with the collagen expanded minced autograft had smaller CSAs and approximately half of the number of fibers regenerated compared to the 100% graft repair. Based on the results of this study and our previous work,²⁶ we speculate that a LM-enriched fibrin hydrogel would serve as an excellent biomaterial for myoblast expansion. In addition, other adjunct therapies such as physical rehabilitation,^37,75 electrical/mechanical stimulation,^76,77 and codelivery of growth factors/signaling molecules^64,78 may also be considered for use with the LM-enriched fibrin hydrogels.

Future work would include a combination therapy of LM-111-enriched fibrin hydrogels with physical rehabilitation, autologous stem cells, or minced muscle autografts. The study would also be extended to 6–8 weeks to observe the hydrogel-mediated muscle regeneration and recovery. Another major limitation of this study is the use of a murine model. Future work will evaluate these hydrogels in a rat model of muscle injury. Overall, these data are instructive for the design and development of a biomaterial-based strategy that incorporates LM-111.

Disclosure Statement

No competing financial interests exist.