Minimally Invasive Muscle Embedding Generates Donor-Cell-Derived Muscle Fibers that Express Desmin and Dystrophin

ABSTRACT

Introduction

The aim of this study was to quantify the extent of donor-cell-derived myogenesis achieved by a novel surgical technique known as Minimally Invasive Muscle Embedding (MIME).

Materials and Methods

Through MIME, we implanted a single extensor digitorum longus muscle from donor mice (N = 2) that expressed a red fluorescent protein (RFP), into the left tibialis anterior (TA) muscle of immunodeficient host mice (N = 4) that expressed a green fluorescent protein (GFP). Soon after MIME, we injected a myotoxin (barium chloride), into the host TA muscle, to trigger concerted muscle degeneration and regeneration. In lieu of MIME, we performed a SHAM procedure on the right TA muscle of the same set of animals.

Results

In MIME-treated muscles, 22% ± 7% and 78% ± 7% muscle fibers were RFP+ and GFP+, respectively (mean ± standard deviation); and all RFP+ fibers were positive for desmin and dystrophin. Conclusion. We conclude that MIME helps generate muscle fibers of donor origin, in host muscle.

INTRODUCTION

Skeletal muscle cells, better known as muscle fibers or myofibers, are postmitotic. However, skeletal muscle is capable of regeneration, because of the presence of tissue-resident myogenic (muscle-generating) cells known as satellite cells (SCs).^1,2 Skeletal muscle regeneration can be compromised due to changes in the extracellular matrix (eg, aging), loss of SCs (volumetric muscle loss from trauma), trophic changes (eg, Volkmann ischemic contracture), and exhaustion of regenerative potential because of recurring degeneration and regeneration (eg, muscle wasting in various muscular dystrophies).^3–6 In case of muscle loss from trauma and muscular dystrophies, providing a source of SCs might help regenerate some of the lost muscle tissue. Attempts to isolate SCs, expand them in culture to generate a large number of myoblasts, and inject myoblasts into host muscle have not yielded a large number of donor-cell-derived fibers.⁷ However, when whole muscle tissue or muscle fibers along with native SCs are grafted onto host muscles, donor-cell-derived myogenesis is better.^8–10 These results are consistent with recent regenerative biology literature, which shows that, when regenerative cells are seeded onto a tissue-like scaffold and then implanted into host tissue, donor-cell-derived regeneration is improved, when compared with delivering isolated cells alone.¹¹ In preparation for studies to test rehabilitative strategies for muscle loss, we have developed a novel surgical technique called Minimally Invasive Muscle Embedding (MIME). MIME involves creating a needle track in a host muscle and embedding the host muscle with a segment of donor muscle tissue.¹² Since the segment of implanted donor tissue contains SCs, MIME is capable of enabling donor-cell-derived myogenesis under suitable conditions. The experiments summarized in this article are designed to ascertain if MIME promotes donor-derived myogenesis, to verify that donor-derived muscle fibers are intact, and to quantify the extent of donor-derived myogenesis promoted by a single donor mouse extensor digitorum longus (EDL) muscle implanted into the TA muscle of a host mouse.

MATERIALS AND METHODS

Animal Models

All experiments with live animals were performed at the Wayne State University (Detroit, Michigan), according to protocols approved by the Institutional Animal Care and Use Committee. These protocols were in accordance with the Guide for the Care and Use of Laboratory Animals (1996, published by National Academy Press, 2101 Constitution Ave. NW, Washington, DC).

We performed MIME on the left tibialis anterior (TA) muscle of immunodeficient host mice that ubiquitously expressed a green fluorescent protein (GFP; nonobese diabetic severely combined immunodeficient gamma [NSG]-GFP mice; The Jackson Laboratory, Stk# 021937; N = 4).¹³ A step-by-step description on how MIME is performed has been published.¹² We implanted an EDL muscle from donor mice that ubiquitously expressed a red fluorescent protein (RFP; DsRed.T3 mice; The Jackson Laboratory, Stk# 006051; N = 2).¹⁴ Host and donor mice were 12-week-old males. After implanting donor tissue, we sealed the needle wounds in host mice with veterinary tissue adhesive (Vetbond Tissue Adhesive, 3 M, St. Paul, Minnesota). As a SHAM procedure, we created a needle track in the right TA of the host mouse but did not implant donor tissue. After the SHAM procedure, the needle wounds were closed with veterinary tissue adhesive, as above. About 5 minutes after MIME (or SHAM), we injected barium chloride (1.2% solution in distilled water, BACL, myotoxin) into the left and right TA muscles to induce concerted degeneration and regeneration of the host TA muscle (and embedded donor EDL muscle if MIME was performed).¹⁵ The MIME and SHAM procedures, as well as intramuscular BACL injection, were performed under general anesthesia (inhaled isoflurane; 1.5%–5% to effect) administered by a tabletop isoflurane vaporizer driven by medical oxygen (V-1 Tabletop Laboratory Animal Anesthesia System; VetEquip, Livermore, California). For pain relief, we administered carprofen (nonsteroidal anti-inflammatory drug; 5 mg/kg, q24h, SC) on the day of MIME (soon after anesthesia induction) and for 3 days thereafter.

At 14 days post-MIME, we euthanized the host mice by cervical dislocation under general anesthesia and collected their MIME- and SHAM-treated muscles.

Experimental Workflow

The workflow and rationale for our experiments is illustrated in (Fig. 1). At baseline, host NSG-GFP mice expressed GFP in all tissues including skeletal muscle (Fig. 1A–C). Through MIME, we implanted an EDL muscle from donor RFP mice into the TA muscle compartment of host mice, in order to provide a source of donor SCs to the host muscle (Fig. 1D–F). Soon after MIME, we injected BACL into the host muscle, in order to induce muscle damage and trigger host and donor SC activation (Fig. 1G–I). At 14 days post-MIME and BACL injection, we collected the TA muscle, to look for evidence of donor-cell-derived myogenesis (Fig. 1J–L). We performed a SHAM procedure that involved creating a needle track in the contralateral TA muscle, and injecting BACL, but not embedding donor tissue. SHAM TA muscles were also collected at 14 days postprocedure.

Figure 1.

Steps Involved in Performing MIME and Assessing Donor-Cell-Derived Myogenesis and Predicted Results. Through MIME, we implanted an entire EDL muscle from donor mice that expressed RFP into the TA muscle compartment of immunodeficient host mice that expressed GFP. Images of the donor muscle being implanted into a needle track created in the host muscle are shown sequentially in panels (A) and (D). Soon after MIME, we injected the host TA muscle along with the embedded donor tissue, with barium chloride (BACL, myotoxin), to induce concerted damage and trigger SC activation and myogenesis (G). At 14 days post-MIME and BACL injection, we collected the TA muscle to assess donor-cell-derived myogenesis (J; arrows point to TA muscle prior to harvesting, and inset shows TA muscle after harvesting). The middle set of panels (B), (E), (H), and (K) are illustrations of the host TA, at the whole muscle level, before MIME, immediately after MIME, after BACL injection, and then at the time of tissue collection, respectively. Our prediction was that if donor SCs contributed to myogenesis, then, at 14 days post-MIME and BACL injection, we will be able to detect RFP+ muscle fibers (K). Panels (C), (F), (I), and (L), respectively, are illustrations of TA muscle at the muscle fiber and SC level. Our prediction was that donor-cell-derived RFP+ fibers will have central nuclei, since muscle fibers arising from recent degeneration (induced by BACL) and regeneration have central nuclei. In panels (C), (F), (I), and (L), nuclei are depicted as blue ovoid structures, and SCs are depicted as spheroid structures sandwiched between muscle fibers.

Histological Studies

We weighed the harvested muscles, briefly dipped them in mineral oil for cryoprotection, blotted off the excess oil with laboratory wipes, snap-frozen the tissue in liquid nitrogen, and stored the samples at −80 °C.

We made 5-μm cross sections of MIME- and SHAM-treated muscles with a cryostat (HM525 NX with EC70 Blade Holder; Thermo Scientific, Waltham, Massachusetts) and collected the sections onto microscopy slides (Tissue Path Superfrost Plus Gold; Fisher Scientific, Pittsburgh, Pennsylvania). Without any additional processing, we imaged these sections under fluorescent optics (10× objective, Zeiss Axio Scope.A1; Carl Zeiss Microscopy, Pleasanton, California) to visualize both GFP and RFP. We captured digital images of multiple overlapping visual fields and generated tiled images of the entire TA muscle with the Photomerge function (File Menu → Automate → Photomerge) in Photoshop CS4 (Adobe Systems Inc., San Jose, California).¹⁶ We quantified donor-cell-derived myogenesis, by counting the numbers of host GFP+ fibers and donor RFP+ fibers, in MIME- and SHAM-treated muscles.

After imaging GFP and RFP, we performed immunofluorescent labeling on the same sections with antibodies to desmin, with a protocol that has been described earlier.¹⁷ We first applied a primary antibody against desmin (rabbit-anti-desmin IgG, RB9014P; Thermo Fisher, Waltham, Massachusetts), and then applied a secondary antibody conjugated to Alexa 647 dye (A21245, goat-anti-rabbit IgG; Thermo Fisher, Waltham, Massachusetts), in order to visualize desmin. Desmin, an intermediate filament protein associated with Z-disks, is commonly studied as a “muscle cell marker” and can be used to determine if fibers are intact or damaged.^18–20

On serial cross sections, collected as above, we performed immunofluorescent labeling with antibodies to dystrophin, similar to desmin labeling, as described.¹⁷ We first applied a primary antibody against dystrophin (rabbit-anti-dystrophin IgG, RB9024P; Thermo Fisher, Waltham, Massachusetts), and then applied a secondary antibody conjugated to Alexa 647 dye (A21245, goat antirabbit IgG; Thermo Fisher, Waltham, Massachusetts), in order to visualize dystrophin. Dystrophin is a sarcolemma-associated protein that is part of the dystrophin-associated protein complex and, like desmin, may also be used as a marker of muscle fiber integrity.^21,22

Sections that were labeled with antibodies to desmin or dystrophin were also labeled as described with 4′,6-diamidino-2-phenylindole (DAPI, 71-03-01; SeraCare Life Sciences, Milford, Massachusetts) to visualize myonuclei under fluorescence optics.²¹ Since central nucleation of fibers is a marker of muscle regeneration, we chose to assess this feature to determine if donor-cell-derived fibers had developed from myogenesis in the host muscle or through en bloc engraftment of donor fibers.²³ For quantitative analyses, centrally nucleated fibers (CNFs) were counted on sections labeled with DAPI and antibodies to dystrophin.

Statistical Methods

The GFP+ and RFP+ fiber counts were converted into percentages of total fibers counted in TA muscle cross sections. Weights of the MIME- and SHAM-treated TA muscles were recorded in milligrams and expressed as a ratio to body weight (recorded in grams). We analyzed GFP+ and RFP+ fiber counts by Student t test and muscle weights by Mann-Whitney rank sum test. To assess the expression of desmin and dystrophin, and the presence of central nuclei, we selected a region (400 μm × 400 μm) of predominantly RFP+ fibers and counted the total number of RFP+ fibers, desmin+ fibers, dystrophin+ fibers, and CNFs. We then expressed desmin+ fibers, dystrophin+ fibers, and CNFs, as a percentage of RFP+ fibers, and analyzed the data by Kruskal-Wallis one-way analysis of variance on ranks followed by pairwise comparisons with Student-Newman-Keuls correction. All cell counts were performed by blinded evaluators. All data are shown as mean ± standard deviation.

RESULTS

Our data confirm that, in all four animals, which were studied, MIME induced donor-cell-derived myogenesis (Fig. 2). This is evident from the presence of RFP+ muscle fibers in TA muscle cross sections (Fig. 2). In contrast, SHAM-treated muscles had no RFP+ fibers (Fig. 2). These data confirm that the RFP+ fibers seen at 14 days post-MIME and BACL injection are not due to a nonspecific red fluorescent signal caused by muscle damage and regeneration.

Figure 2.

MIME Consistently Facilitates Donor-Cell-Derived Myogenesis. Our data indicate that, at 14 days post-MIME and barium chloride (BACL, myotoxin) injection, 100% of MIME-treated muscles (4 out of 4 TA muscles) have RFP+ fibers, which is evidence of donor-cell-derived myogenesis. As expected, in SHAM-treated muscles from the same set of animals, we do not see any RFP+ muscle fibers. Under fluorescence optics, host fibers appear green since they express GFP, and donor-derived fibers appear red since they express RFP. The white arrows point to RFP+ donor-cell-derived muscle fibers that are present, even up to several hundreds of micrometers away from the region that contains predominantly RFP+ fibers. Scale bar = 100 μm.

Quantitative data analyses revealed that, of the total number of fibers counted in each MIME-treated TA muscle, ~ 20% were RFP+ and ~ 80% were GFP+ (Fig. 3G). Qualitatively, we found that RFP+ fibers were clustered together; however, there were a few RFP+ fibers located several hundreds of microns away from the cluster of RFP+ fibers (white arrows in MIME panels of Fig. 2).

Figure 3.

MIME Generates Donor-Cell-Derived Muscle Fibers That Express Desmin and Dystrophin. Panels (A–C) are images obtained by studying serial cross sections of the same TA muscle. In donor-cell-derived fibers, identified by the presence of RFP, the muscle fiber markers desmin and dystrophin are, respectively, present in the myoplasm and at the sarcolemma, as would be expected of mature and intact muscle fibers (B and C, respectively). High magnification images of the regions enclosed within the blue boxes in (A–C) (400 μm × 400 μm) are shown in panels (D–F), respectively. In panels (B), (C), (E), and (F), nuclei appear blue since they are stained with DAPI. The high magnification images show that most of the RFP+ fibers are CNFs, indicative of degeneration and subsequent regeneration (D–F, particularly visible in F). In panels (D–F), the white dashed line indicates an interface between fibers that are predominantly of donor origin (RFP+) and host origin (GFP+). Quantitative analyses of histological data from four separate TA muscles treated with MIME + BACL indicate that, ~20% fibers in the host muscle were RFP+ and are therefore of donor origin (G; ^* denotes P < 0.5 by Student t test). Quantitations also show that 100% of RFP+ fibers are positive for desmin and dystrophin and ~ 60% of RFP+ fibers are CNFs (H; ^* denotes P < 0.05 by Kruskal-Wallis one-way analysis of variance on ranks, followed by pairwise comparisons with Student-Newman-Keuls correction). Analysis of muscle weight to body weight ratio suggests that MIME and SHAM muscles do not differ in weight (I; P > 0.05 as analyzed by Mann-Whitney rank sum test). Data are reported as mean ± standard deviation. Scale bar = 100 μm.

Additional studies confirmed that the RFP+ structures with morphological resemblance to muscle fibers, seen in MIME-treated muscles, were indeed intact muscle fibers, since they expressed desmin and dystrophin (Fig. 3A–F). In a 400 μm × 400 μm region selected from each MIME-treated muscle, we found that all RFP+ fibers expressed desmin and dystrophin, and ~ 60% were CNFs (Fig. 3D–F and H). In panels D–F, the white dashed line indicates an interface between fibers that are of donor (RFP+) and host origin (GFP+).

At 14 days post-MIME and BACL injection, we found no evidence of muscle wasting caused by MIME, since there was no difference in weight between MIME- and SHAM-treated TA muscles.

Taken together, our data confirm that MIME facilitates the development of donor-cell-derived muscle fibers that are intact.

DISCUSSION

In this article, we provide proof of concept that MIME consistently facilitates donor-cell-derived myogenesis. In donor-cell-derived fibers, identified by the presence of a fluorescent reporter protein, the muscle fiber markers desmin and dystrophin are, respectively, present in the myoplasm and at the sarcolemma, as would be expected of mature and undamaged muscle fibers.^18–22 The evidence presented in this article forms the basis for ongoing and future experiments that will involve the MIME technique to implant myogenic donor tissue that is either natural or engineered, followed by the administration of rehabilitative, pharmacological, and genetic interventions, to improve donor-cell-derived myogenesis and muscle regeneration. Such studies would hopefully translate to effective clinical strategies to reverse muscle loss caused by muscle wasting syndromes (eg, muscular dystrophies), trauma, and age-related sarcopenia. Although it is unreasonable to expect that MIME can reverse muscle loss in all muscle groups affected by a muscular dystrophy, it is reasonable to hope that, in the future, a technique based on MIME, performed with suitable myogenic tissue, could help improve function in certain superficially located muscle groups that are necessary for basic activities of daily living (eg, the biceps brachii).²⁴ In a clinical situation, adjunctive therapies, such as exercise, and electromechanical stimulation would likely be used in lieu of a myotoxin-like BACL to trigger and optimize donor-cell-derived myogenesis.⁶

The MIME technique is a refinement of earlier methods that involved grafting whole or minced muscle tissue onto host muscles through an open surgical technique.²⁵ The ability of MIME to cause minimal disruption to host muscle architecture might be advantageous over open surgical methods to implant donor tissue. Although it is not possible to experimentally determine the exact percentage of donor myogenic cells that survive MIME and facilitate donor-cell-derived myogenesis, it is intriguing to note that the EDL muscle is roughly 20% to 25% of the weight of the TA muscle, and we observe that ~ 20% of the TA has been repopulated by RFP+ fibers at 14 days post-MIME and BACL injection (Fig. 3). We could therefore speculate that the majority of myogenic cells from the donor EDL have likely contributed to myogenesis. Because of the minimally invasive nature of MIME, it can be performed repeatedly on the same host muscle, if the goal is to repopulate a large portion of the host muscle with donor-cell-derived muscle fibers—this is useful in the generation of murine xenograft models of human diseases (eg, facioscapulohumeral muscular dystrophy).^9,26

The data presented in this article are from a small group of animals (N = 4) that were studied at a single time point (14 days post-MIME), thus limiting the scope of this study to only being proof of concept. Interim data published from ongoing studies suggest that donor-derived muscle fibers are present even 3 months post-MIME.²⁷ With clinical translation being the eventual goal, future directions for MIME include studying the ability of MIME to rejuvenate the regenerative potential of skeletal muscle, in experimental conditions where endogenous regenerative potential has been irreversibly compromised, in both small and large animal models.

CONCLUSION

The MIME technique donor-cell-derived muscle regeneration, and the donor-cell-derived muscle fibers express the muscle cell markers desmin and dystrophin.

FUNDING

Pilot Grant from the Alliance for Regenerative Rehabilitation Research and Training (AR³T), NIH 1R03HD091648-01 from NICHD, and a Faculty Startup Package and FRAP award from Wayne State University to J.A.R.