The superhealing MRL background improves muscular dystrophy
© Heydemann et al.; licensee BioMed Central Ltd. 2012
Received: 24 May 2012
Accepted: 8 October 2012
Published: 5 December 2012
Skip to main content
© Heydemann et al.; licensee BioMed Central Ltd. 2012
Received: 24 May 2012
Accepted: 8 October 2012
Published: 5 December 2012
Mice from the MRL or “superhealing” strain have enhanced repair after acute injury to the skin, cornea, and heart. We now tested an admixture of the MRL genome and found that it altered the course of muscle pathology and cardiac function in a chronic disease model of skeletal and cardiac muscle. Mice lacking γ-sarcoglycan (Sgcg), a dystrophin-associated protein, develop muscular dystrophy and cardiomyopathy similar to their human counterparts with limb girdle muscular dystrophy. With disruption of the dystrophin complex, the muscle plasma membrane becomes leaky and muscles develop increased fibrosis.
MRL/MpJ mice were bred with Sgcg mice, and cardiac function was measured. Muscles were assessed for fibrosis and membrane leak using measurements of hydroxyproline and Evans blue dye. Quantitative trait locus mapping was conducted using single nucleotide polymorphisms distinct between the two parental strains.
Introduction of the MRL genome reduced fibrosis but did not alter membrane leak in skeletal muscle of the Sgcg model. The MRL genome was also associated with improved cardiac function with reversal of depressed fractional shortening and the left ventricular ejection fraction. We conducted a genome-wide analysis of genetic modifiers and found that a region on chromosome 2 was associated with cardiac, diaphragm muscle and abdominal muscle fibrosis.
These data are consistent with a model where the MRL genome acts in a dominant manner to suppress fibrosis in this chronic disease setting of heart and muscle disease.
Murphy Roths Large (MRL) mice are an inbred mouse strain noted to have enhanced healing ability. This MRL strain was initially discovered because of its rapid ability to heal ear holes[1, 2]. The MRL strain’s capacity to rapidly recover from injury has been seen for both digit wounding and corneal scarring[3, 4]. The MRL strain has been reported to reduce scar formation after acute cardiac injury, including freeze injury and coronary artery ligation[5, 6]. However, other studies have suggested that larger scale acute cardiac injury cannot be overcome by the MRL strain’s healing capacity[7–11]. In those injury settings where the MRL background induces more rapid healing, multiple mechanisms have been implicated to explain this phenomenon including decreased scar formation, altered inflammatory response, reduced apoptosis, increased proliferation, improved remodeling and, in some settings, enhanced stem cell function[4, 6, 12–15]. Genetic data support that many different mechanisms account for enhanced healing since more than 40 different genetic loci have been associated with aspects of the healing phenotype[14, 16].
The dystrophin complex is composed of membrane-associated proteins that mediate membrane stability in heart and skeletal muscle. Mutations that disrupt expression of dystrophin or its associated proteins the sarcoglycans proteins cause progressive cardiac and skeletal muscle degeneration in humans and mouse models. At the cellular level, the loss of dystrophin or the sarcoglycans leads to a disrupted sarcolemmal membrane that is abnormally leaky. Membrane permeability results in increased intracellular calcium that triggers proteolysis and necrosis. An inflammatory response also contributes to muscle degeneration. Muscle contraction is thought to provoke submicroscopic disruption of the sarcolemma, and in the absence of a normal dystrophin complex, this leads to dysfunction and destruction of cardiomyocytes and skeletal myofibers. The reduction in contractile cells and the presence of fibrotic scar tissue within the heart led to reduced cardiac contractile function and congestive heart failure.
Mouse models of dystrophin or sarcoglycan mutations recapitulate the basic pathological defects seen in human forms of these genetic disorders. Mice lacking γ-sarcoglycan were engineered by removing the first coding exon of the Sgcg gene and model muscular dystrophy. We previously introduced the Sgcg allele into the DBA/2J genetic background and found that this background confers a more severe phenotype. The enhanced severity is seen as increased scar formation, measured as hydroxyproline (HOP) content because this modified amino acid is a marker of collagen. In skeletal muscle, increased Evans blue dye uptake is monitored to reflect membrane leakiness. Using these assays, we showed that the DBA/2J genetic background worsened the disease process.
Because of its role in wound healing and the basic similarities between common forms of cellular injury and what is seen in muscular dystrophy and cardiomyopathy, we introduced the MRL genome into the Sgcg model. We hypothesized that the indolent pace of cellular damage in this disorder could be abated by the MRL background. Sgcg mice were bred to MRL mice to generate Sgcg mice on a mixed genetic background with 50% contribution of the MRL strain. This breeding strategy produced Sgcg mice with 50% genetic background from the MRL strain and 50% background of the DBA/2J strain, and these mice are referred to Sgcg MRL/D2 . We found that a 50% contribution of the MRL genome reduced fibrosis in the heart and skeletal muscle, consistent with dominant genetic loci in the MRL background. Interestingly, the MRL genome did not consistently reduce membrane leak, suggesting that the MRL background does not exert its effect on myocyte membrane stability and instead acts downstream on remodeling. Sgcg MRL/D2 mice had improved cardiac function so that they were now similar to wild-type mice. We conducted a genome-wide scan using informative polymorphisms and identified a region on chromosome 2 that was associated with fibrosis in cardiac, diaphragm and abdominal muscles. These data demonstrate that genes within the MRL background are modifiers for cardiopulmonary involvement in muscular dystrophy.
Number of mice analyzed
Sgcg MRL/D2 , 32wk
Sgcg MRL/D2 , 32wk
Fiber size variability
Mice were sacrificed at either 8 or 32 weeks for analysis. Muscles were used for either the Evans blue dye assay or the HOP assay. Dye uptake assays were performed on triceps muscles, gastrocnemius/soleus muscles and gluteus/hamstrings groups of muscles. Dye uptake was performed on half of each quadriceps muscle and half of the abdominal muscles since these muscles are of sufficient size to perform both dye uptake and fibrosis assays. Hydroxyproline (HOP) assays were performed on the diaphragm muscle, the cardiac ventricles isolated as a single unit, half of the abdominal muscles and half of each quadriceps muscle. The amount of muscle assayed varied with the size of the muscle and ranged from 30 mg for the diaphragm muscle to 300 mg for the gluteus/hamstring muscle group that included the semimembranosus, semitendinosus and biceps femoris.
Evans blue dye (Sigma, E-2129) was performed as described[19, 21]. Evans blue dye (Sigma, E-2129) was dissolved in phosphate-buffered saline at 10 mg/ml and injected intraperitoneally at 5 μl/g body weight. Twenty to 40 h later, the tissues were harvested, finely minced, weighed and incubated at 55°C in 1 ml formamide for 2 h before spectrophotometric absorbance was measured at 620 nm[22, 23]. Results are reported as absorbance/mg tissue.
The hydroxyproline (HOP) assay was performed as described[19, 21, 24]. The tissue was minced, weighed and hydrolyzed overnight in 2 ml of 6 M hydrochloric acid at 110°C. Ten μl of this hydrolysate was mixed with 150 μl isopropanol, then 75 μl of 1.4% chloramine-T (Sigma, St Louis, MO) in citrate buffer and oxidized at room temperature for 10 min. One ml of a 3:13 solution of Ehrlich’s reagent (3 g of 4-(dimethylamino) benzaldehyde, Sigma, St Louis, MO; 10 ml ethanol; 675 μl sulfuric acid) to isopropanol was added, mixed and incubated for 30 min at 55°C followed by extinction measurement at 558 nm. A standard curve (0–5000 nM, trans-4-hydroxy-L-proline, Sigma, St Louis, MO) was included in each assay. Results are reported as nM HOP/mg tissue.
Tissues were flash frozen in liquid nitrogen-cooled isopentane and stored at 80°C; 7-μm sections were cut on a cryostat and fixed to slides in ice-cold 100% methanol. The following antibodies were used: dystrophin NCL-DYS2 (Novocastra/Leica), PH3 04–817 (Millipore), CD3 MON1003-1 (Monsanto), caspase 3 (BD Biosciences), MAC1 BD557395 (BD Biosciences) and eMHC F1.652 (ATCC). The TUNEL kit was from Millipore. Central nuclei and fiber size variability was determined blinded to genotype by analyzing ten randomly chosen fields of 40× magnification. Fiber size variability was compared using each animal’s coefficient of variability (standard deviation/mean).
Twelve-week animals were evaluated by echocardiography as described[25, 26]. Investigators were blinded to genotype. To avoid the stress associated with conscious restraint, anesthetized animals were studied. Anesthesia was induced by 1% isoflurane in a closed chamber (Ohmeda Fluotec 3; Matrix Medical, Orchard Park, NY) in 20% O2 delivered through a nose cone. Chest hairs were removed with a topical depilatory agent. Limb leads were attached for electrocardiogram gating, and the animals were imaged in the left lateral decubitus position with a Visual Sonics Vevo 770 machine using a 30-MHz high-frequency transducer. Body temperature was maintained using a heated imaging platform and warming lamps. Anesthesia was variably delivered to maintain heart rates throughout the procedure at a constant 380–420 beats per minute. Two-dimensional images were recorded in parasternal long- and short-axis projections, with guided M-mode recordings at the midventricular level in both views. LV cavity size and wall thickness were measured in at least three beats from each projection and averaged. LV wall thickness, interventricular septum (IVS) and posterior wall (PW) thickness, and internal dimensions at diastole and systole (LVIDd and LVIDs, respectively) were measured. LV fractional shortening [(LVIDd – LVIDs)/LVIDd] and relative wall thickness [(IVS thickness + PW thickness)/LVIDd] were calculated from the M-mode measurements.
Single nucleotide polymorphisms (SNPs, n = 1,701) informative between the parental DBA/2J and MRL/MpJ strains were genotyped in 80 Sgcg MRL/D2 F2-F4 Sgcg animals on the Illumina GoldenGate platform using the Mutation Mapping and Developmental Analysis Panel (MMDAP) and the Mouse Universal Genotyping Array (GeneSeek, Neogen Corp., Lansing, MI). R package QTLRel was used to perform whole-genome quantitative trait locus (QTL) mapping for each of the membrane permeability and fibrosis phenotypes and using sex as a covariate[29, 30]. Significance thresholds were determined by 1,000 permutation tests. The 1.5-LOD drop support interval was calculated using QTLRel. Tests of normality and other statistics were calculated in Prism (GraphPad). Build 37.1 was used for genomic analysis.
For HOP and dye uptake assays, data were analyzed by one-way, unpaired ANOVA with parametric methods followed by the Tukey multiple comparison post-test (Prism, Graphpad), and p < 0.05 was considered significant. Data from 197 Sgcg animals from an F3 intercross between the DBA/2J and 129T2/SvEmsJ backgrounds (Sgcg D2/129 ) was compared to the Sgcg MRL/D2 cohort.
We also assessed apoptosis using a TUNEL assay and caspase 3 staining and again found no differences between Sgcg MRL/D2 and Sgcg muscle, suggesting that gross differences in programmed cell death are unlikely to account for the improved healing of the MRL strain (Additional file1: Figure S2). We also characterized whether T cell infiltration or macrophage infiltration differed qualitatively between Sgcg MRL/D2 and Sgcg muscle, but we found no clear differences between Sgcg MRL/D2 and Sgcg muscle, suggesting other features contribute to the MRL background’s effect on muscular dystrophy (Additional file1: Figure S2).
Cardiac function in Sgcg mice
Sgcg D2 (n = 4)
33.87 ± 4.80 a
WT D2 (n = 5)
44.58 ± 7.65 a
Sgcg MRL/D2 (n = 5)
46.00 ± 14.4
WT MRL/D2 (n = 5)
48.37 ± 11.21
Ltbp4 genotype in Sgcg MRL/D2 mice
Ltbp4 D/D (n = 14)
3.90 ± 0.83
6.00 ± 1.81
Ltbp4 D/I (n = 34)
4.18 ± 0.84
8.05 ± 3.02
Ltbp I/I (n = 47)
4.00 ± 0.85
6.64 ± 2.15
Mutations in the dystrophin or sarcoglycan genes share a common pathological mechanism characterized by disruption of the plasma membrane of cardiomyocytes and skeletal myofibers. Membrane leak is an early step in the pathological process while fibrosis is thought to be a secondary response. Our study suggests that there are genetic pathways that target fibrosis without altering the sarcolemmal leak properties. In the heart, reduced fibrosis was correlated with functional benefit, although cardiac function may be improved from both cardiac intrinsic as well as cardiac extrinsic features. For example, improved skeletal muscle function may contribute to improved cardiac function, particularly when considering the contribution of the respiratory musculature.
The MRL strain typically shows the improved healing capability in its first 2–6 months of life. After 6 months of age, wild-type MRL mice develop autoimmune disorders; it is for this reason that we conducted studies using a 50% contribution of the MRL strain. Quantitative trait mapping has been used to define many different genetic regions associated with the healing properties. The autoimmune properties have been genetically separated from the healing properties since the four to six different “autoimmune” genetic loci do not overlap with those associated with improved healing. In our studies, only 50% of the MRL genome was capable of suppressing fibrosis and improving function. In the course of these studies, we never identified a single Sgcg MRL/D2 animal that had markedly elevated fibrosis. This stands in contrast to what is normally observed in the Sgcg animals where outliers with markedly elevated fibrosis are often seen. The maximum values for fibrosis in Sgcg MRL/D2 mice were all well below what was measured in Sgcg D2 mice. This is consistent with multiple, dominant genetic loci imparting the improved healing capabilities, similar to what has been observed for the ear hole repair properties in the MRL strain. The finding that improvement persists in the Sgcg MRL/D2 mice, with almost complete suppression of fibrosis at 32 weeks, suggests that a persistent healing effect can arise from the MRL strain.
When considering the overlapping region on chromosome 2 from the heart, diaphragm and abdominal fibrosis data, the interval contains 49 known genes. There are additional predicted and unnamed genes and genes encoding olfactory receptors that are not considered likely candidates. Within the interval is Jag1, encoding jagged the ligand for the Notch receptor. Loss of function of Notch 3 leads to muscle hyperplasia, especially when subjected to repetitive injury. Ex vivo activation of Notch signaling helps maintain donor cell engraftment during myoblast transfer. Another gene linked to growth and healing in the interval is Bmp2-encoding bone morphogenetic protein 2. The BMPs mediate musculoskeletal regeneration, and given the MRL background effect on multiple cells and tissues, BMP2 is well positioned to contribute to the MRL superhealing response. That said, this chromosome 2 interval has not previously been linked to other MRL healing properties. The strict criteria imposed by the QTLRel algorithm makes it unlikely that these results derive from relatedness of animals within the cohort.
The ability of the heart to recover after injury has been studied in the MRL strain using cryoinjury, left anterior descending ligation and ischemia reperfusion methods[5, 6, 8–10, 14, 15]. However, scar reduction has been noted after some forms of injury while not after others. In common to all these studies is an acute injury model where a normal heart was substantially damaged in a single setting. The MRL’s ability to heal the heart may be limited such that larger amounts of injury may be insurmountable, as proposed by Naseem. Our data support that lower, although persistent, levels of injury can be managed by the MRL strain where there is significant reduction in fibrosis and a corresponding functional improvement in cardiac function.
A number of mechanisms likely act in concert to achieve suppression of fibrosis and improvement of function. In skeletal muscle, where muscle stem cells robustly regenerate muscle after injury, there is indirect evidence of enhanced regeneration. Specifically, there is an increase in centrally nucleated myofibers, which is thought to reflect enhanced myoblast fusion. However, our data do not distinguish whether the MRL background exerts its effect on muscle regenerative cells, or by creating a more supportive extracellular matrix or both. MRL animals heal with embryonic characteristics with enhanced blastema formation and metabolic and gene expression features consistent with an earlier developmental state. Altered protease expression has also been noted in the MRL mouse in response to damage. Recent work characterizing the immune infiltrate in muscular dystrophy found that reducing osteopontin was effective at suppressing fibrosis. Together these data may favor matrix-associated modifications that can modulate muscle fibrosis and function.
Herein the MRL background was shown to reduce fibrosis in a chronic model of muscular dystrophy and cardiomyopathy, the Sgcg mouse. Although membrane leak was still evident in Sgcg mice sharing a portion of the MRL genome, the reduction in fibrosis was associated with improved cardiac function. The identification of gene(s) from the MRL genome will help identify pathways important for chronic repair of myopathic processes.
Bone morphogenetic protein
Duchenne muscular dystrophy
Latent TGFβ-binding protein
Murphy Roth Large
Quantitative trait loci
Single nucleotide polymorphism
Transforming growth factor β.
Supported by NIH R01HL61322 (EMM), R01HL102322 (AH) and NIH K08HL098565 (GK).
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.