PKC Theta Ablation Improves Healing in a Mouse Model of Muscular Dystrophy

Inflammation is a key pathological characteristic of dystrophic muscle lesion formation, limiting muscle regeneration and resulting in fibrotic and fatty tissue replacement of muscle, which exacerbates the wasting process in dystrophic muscles. Limiting immune response is thus one of the therapeutic options to improve healing, as well as to improve the efficacy of gene- or cell-mediated strategies to restore dystrophin expression. Protein kinase C θ (PKCθ) is a member of the PKCs family highly expressed in both immune cells and skeletal muscle; given its crucial role in adaptive, but also innate, immunity, it is being proposed as a valuable pharmacological target for immune disorders. In our study we asked whether targeting PKCθ could represent a valuable approach to efficiently prevent inflammatory response and disease progression in a mouse model of muscular dystrophy. We generated the bi-genetic mouse model mdx/θ−/−, where PKCθ expression is lacking in mdx mice, the mouse model of Duchenne muscular dystrophy. We found that muscle wasting in mdx/θ−/− mice was greatly prevented, while muscle regeneration, maintenance and performance was significantly improved, as compared to mdx mice. This phenotype was associated to reduction in inflammatory infiltrate, pro-inflammatory gene expression and pro-fibrotic markers activity, as compared to mdx mice. Moreover, BM transplantation experiments demonstrated that the phenotype observed was primarily dependent on lack of PKCθ expression in hematopoietic cells. These results demonstrate a hitherto unrecognized role of immune-cell intrinsic PKCθ activity in the development of DMD. Although the immune cell population(s) involved remain unidentified, our findings reveal that PKCθ can be proposed as a new pharmacological target to counteract the disease, as well as to improve the efficacy of gene- or cell- therapy approaches.


Introduction
Duchenne muscular dystrophy (DMD) is one of the most common X-linked lethal diseases, and results from the mutation within the gene encoding dystrophin, a large cytoskeletal protein, whose ablation leads to membrane instability [1]. Therapies based on the restoration of dystrophin expression or the administration of dystrophin +ve stem cells are promising, but still in the preclinical phase [2][3][4]. In this context, one of the barrier to successful gene therapy has been recently identified in cellular immunity [5]. Thus, the monitoring of cellular immune responses should be a priority for any experimental therapy designed to increase the number of dystrophin-positive myofibers in patients with Duchenne's muscular dystrophy. In addition, although mechanical injury and membrane defects are important factors promoting dystrophic pathology, increasing evidences highlight aberrant intracellular signalling cascades that regulate inflammatory and immune processes, as key contributors to the degenerative process [1,6]. Up-regulated inflammatory gene expression and activated immune cell infiltrates are evident during early disease stages in dystrophic muscle, and the identification of specific targets for anti-inflammatory therapies is one of the ongoing therapeutic options. Indeed, glucocorticoids, which have anti-inflammatory properties, are being used to treat DMD with some success; however, the side effects of these drugs often outweigh their benefit [7,8]. Numerous other anti-inflammatory therapies have been proposed to improve healing [9][10][11][12][13]. In this context, it has been recently shown that rapamycin treatment reduced dystrophic phenotype in mdx mice and that this effect was associated to a significant reduction in infiltration of T eff cells in skeletal muscle tissue, while T reg cells were preserved [14]. Indeed, a role of lymphocytes activity in the progression of muscular dystrophy is long known, as antibody-or genetic-mediated lymphocyte depletion improved the disease in mdx, however, not all studies had produced definitive results, and the implication of lymphocytes and their subtypes in the disease is still to be clearly defined. Among the possible targets, genetic and pharmacological evidences suggest that protein kinase C (PKC) isotypes have essential functions in promoting both early T-cell activation and sustained T-cell adhesion and are proposed as drug targets in adaptive immunity [15]. Among the isotypes, PKCh is unique in its ability to translocate to the immunological synapses upon T-cell receptor (TCR) activation, regulating NFkB, AP-1, and NFAT transcriptional activity [16]. PKCh can also directly associate to chromatin, regulating T-cell-specific inducible gene expression program and microRNAs [17]. It is worth noting that PKCh has been recently shown to be required for full T eff activation while inhibiting T regmediated suppression [18]. Indeed, PKCh is being proposed as a particularly attractive target for developing ways to selectively manipulate T eff cell functions that are relevant to pathogenesis of different diseases, including asthma, rheumatoid arthritis, multiple sclerosis and colitis [15,[19][20][21], without imparting a severe immunosuppression. Interestingly, PKCh is the PKC isoform predominantly expressed also in skeletal muscle, where it mediates various cellular responses [22][23][24][25][26][27]. Although most of these studies demonstrated that PKCh is actually required for complete histogenesis, differentiation and homeostasis of skeletal muscle, we asked which would be its prevalent role in a context of chronic inflammation, where immune cells activity is a key determinant as in muscular dystrophy. We here demonstrate that immune-cell intrinsic PKCh activity plays a crucial role in the progression of muscular dystrophy, and targeting PKCh can be proposed as a valuable therapeutic strategy for the disease.

Lack of PKCh in mdx mice reduces muscle degeneration and inflammation
To verify whether inhibition of PKCh may improve healing in muscular dystrophy, we crossed the PKCh knock out model (PKCh 2/2 ) with mdx, the mouse model of DMD. We first observed, by Western blot analysis, that PKCh was highly expressed in both WT and mdx hindlimb muscle, but a significantly higher portion of it was phosphorylated, as a feature of its activation [16], in the mdx, as compared to WT (Fig. 1A). As expected, no dystrophin expression was detectable in the mdx and mdx/h 2/2 muscle, and no PKCh immunoreactivity in the mdx/ h 2/2 muscle (Fig. 1A). Both mutants were healthy at birth. We thus evaluated myofiber degeneration in diaphragm (as being diaphragm one of the most affected muscle group in mdx) and in tibialis anterior (TA) of 2 mo old mice, as Evan's blue dye (EBD) uptake. As shown in Figure 1B and C (a-b), myofiber degeneration was significantly reduced in both muscles derived from mdx/h 2/2 in respect to mdx. As expected, those degenerating areas were sites of a robust inflammatory response, as shown by anti-mouse IgG immunofluorescence (Fig. 1C, c-d). Indeed, while degenerating fibers in mdx muscle were surrounded by many mononucleated cells, degenerating fibers in mdx/h 2/2 were almost avoided of surrounding mononucleated cells, and mouse IgGs were strictly localized on the fiber itself (Fig. 1D). Accordingly, Western Blot analysis revealed that the IgG content in the protein extract from mdx/h 2/2 muscle was greatly lower than that from mdx (Fig. 1E). Hematoxilin/Eosin (H/E) staining of TA muscle sections, showed that lack of PKCh resulted in significant reduction in cell infiltrate as compared to mdx ( Fig. 2A), leading to an overall maintenance of muscle structure; however, the myofibers variability and the percentage of centro-nucleated myofibers over the total number of fibers (as features of dystrophic muscle) were similar between the two genotypes ( Fig. 2B-C). As being macrophage infiltration the most prominent immune feature observed in mdx [6], TA cryosections were analysed for esterase activity. As shown in Figure 2D, the areas of macrophages Representative western blot analysis of total protein fraction of TA muscles derived from 2 mo old WT, PKCh 2/2 , mdx and mdx/h 2/2 mice, as indicated. The blot was incubated with the indicated primary antibodies. GAPDH expression level is shown in the bottom for equal loading. PKCh activation in muscle derived from mdx (black bar) mice, expressed as fold induction in respect to WT (white bar, assumed as 1), is shown as the ratio of pPKCh/PKCh (right panel), as determined by densitometric analysis from three independent experiments (*p,0.05). (B) EBD uptake in diaphragm derived from 2 mo old mdx or mdx/h 2/2 mice, as indicated, shown under light (a, c) and epifluorescence (b, d) microscopy. (C) EBD uptake in TA muscle derived from 2 mo old mdx (a) or mdx/h 2/2 (b) mice, as indicated; immunofluorescence analysis of IgG accumulation in mdx (c) or mdx/h 2/2 (d) mice; bar = 200 mm. (D) Mononuclear cells accumulation, revealed as Hoechst staining, around single degenerating fiber, detected as EBD uptake (a-b) and IgG immunofluorescence (c-d), in TA muscles from mdx (a and c) or mdx/ h 2/2 (b and d). (E) Representative western Blot analysis of IgG accumulation in TA muscles from mdx or mdx/h 2/2 (two mice/ genotype), as indicated. Densitometric analysis is shown in the bottom (mdx, black bar; mdx/h 2/2 , grey bar, *p,0.05). doi:10.1371/journal.pone.0031515.g001 infiltrate in 2 mo old mdx/h 2/2 muscle were strongly reduced; accordingly, FACS analysis of CD45/Mac-3 co-expressing cells, revealed that the level of macrophages was reduced by <50% in 2 mo old mdx/h 2/2 muscle, as compared to mdx (Fig. 2E).
Macrophages are known to store and produce matrix metalloproteinase 9 (MMP-9) in response to different stimuli, such as oxidative stress from necrotic tissue, and represent the major source of MMP-9 [28]. Indeed, the high level of MMP9 activity observed in mdx, was strongly prevented in mdx/h 2/2 muscle, as shown by zymography (Fig. 2F). Moreover, as shown in Figure 3, lack of PKCh strongly prevented the hyper-expression/activation of pro-inflammatory signalling pathways; in fact, both the level of expression and of phosphorylation of the p65 subunit of NFkB and of JNK in mdx/h 2/2 TA muscle were much lower than in mdx, very similar to the level observed in WT. The p65 subunit of NFkB is the NFkB subunit mostly hyperactive in mdx muscle [11] and JNK is the upstream regulator of AP1 signalling pathway, which is also hyperactive in mdx [1].

Lack of PKCh in mdx mice improves muscle regeneration
To verify whether the observed reduction in muscle wasting and cell infiltrate came along with improvement in muscle repair, muscle regeneration was analysed in the mutant mice by both morphological and biochemical analyses. Immunofluorescence analysis of embryonic myosin (eMyHC) expression, a marker of regenerating myofibers, revealed that lack of PKCh resulted in an increase on eMyHC expressing myofibers along with a reduction in EBD positive myofibers (Fig. 4A). Indeed, quantitative analysis revealed that the extension of ''regenerating area'', including eMyHC expressing fibers, was significantly higher in mdx/h 2/2 , as compared to mdx muscle, parallel to a significant decrease in ''degenerating area'', including EBD stained fibers (Fig. 4A). Accordingly, the level of myogenin expression, a marker of differentiating myoblasts, was strongly increased in mdx/h 2/2 , as compared to mdx muscle (Fig. 4B). Intriguingly, we previously showed that PKCh is actually required for myofiber growth both in vivo and in vitro, as being an upstream regulator of the expression of pro-fusion genes [27]. To unravel the apparent contradiction with our current results, in vitro differentiation of primary myoblasts derived from mdx and from mdx/h 2/2 hindlimb muscle was compared. As shown in Figure 4C, by 48 hours in DM, mdx myoblasts had formed elongated myotubes containing a large number of nuclei; by contrast, mdx/h 2/2 myoblasts formed thinner myotubes with reduced number of myonuclei, according to our previous observation in PKCh 2/2 myoblasts. Taken together, these results suggest that lack of PKCh in mdx makes a more favourable environment for muscle precursor cells to differentiate, rather than enhancing their activation/differentiation ability. Indeed, though the regenerating area in vivo was more extensive in mdx mice lacking PKCh, as compared to mdx, cross- Percentage of centrally nucleated myofibers in TA muscles derived from 2 mo old mdx and mdx/h 2/2 , expressed as percentage over the total number of myofibers (n = 3/genotype). (D) Esterase histochemical staining of TA cryosections derived from 2 mo old mdx and mdx/h 2/2 mice, as indicated. Arrows indicate cell infiltrates, arrows indicate neuromuscular junctions. Bar = 200 mm (E) FACS analysis of CD45 +ve / Mac3 +ve mononucleated cells isolated from TA muscle derived from 2 mo old mdx and mdx/h 2/2 mice, as indicated, expressed as percentage of the total number of cells examined. The percentage of reduction in mdx/h 2/2 muscle, in respect to mdx, is also shown (n = 3/ genotype). (F) Gel zymography of MMP9 activity in TA muscle derived from 2 mo old mdx and mdx/h 2/2 mice, as indicated; media collected from differentiating muscle cell cultures was used as positive control (+). doi:10.1371/journal.pone.0031515.g002 sectional area of individual eMyHC, regenerating myofibers was medially reduced by <15%, as a result of PKCh.ablation.

Lack of PKCh in mdx mice preserves exercise performance
To verify whether the improvement in muscle maintenance and regeneration resulted in better performance, a treadmill endurance test was performed and the number of times the mice stopped during the 30 min running, each day of the test, was recorded. As expected, mdx mice stopped increasing times during individual running (not shown) and, medially, many more times than WT mice, in each day examined (Fig. 5). Importantly, mdx/h 2/2 mice behaved very similar to WT, both during the running than in all days examined (Fig. 5).
Rescue of PKCh expression in hematopoietic cells in mdx/h 2/2 mice, by bone marrow transplantation, partly restores mdx mice phenotype To verify whether the obtained phenotype was actually due to alterations in inflammatory cells activity, 1 mo old mdx bone marrow (BM) was transplanted into irradiated, age and sex matching mdx/h 2/2 mice. In these mice, hereafter called mdx/h 2/2BMmdx , PKCh was thus expressed only in the BMderived cells. As control, another group of age and sex matching mdx/h 2/2 mice was transplanted with BM derived from mdx/h 2/2 , hereafter called mdx/h 2/2BMmdx/h2/2 . The mice were sacrificed 6 weeks after transplantation and BM repopulation was ensured by genomic PCR analysis, as the appearance of the PKCh WT PCR product (Fig. 6A). Western blot analysis revealed that PKCh expression was largely rescued in thymus derived from mdx/h 2/2BMmdx transplanted mice, while the expression of other PKC isoforms was unaltered (Fig. 6B), and immunofluorescence analysis revealed that PKCh expressing cells were detectable in the spleen (Fig. 6C), demonstrating that engrafted hematopoietic cells repopulated also mature compartments. H/E staining of TA muscle crysections derived from mdx/h 2/2BMmdx revealed a significant increase of infiltrating cells (Fig. 7A, e), as compared to those derived from mdx/h 2/2BMmdxh2/2 (Fig. 7A, a), which were mostly macrophages, as shown by esterase histochemistry (Fig. 7A, f). Moreover, EBD uptake revealed an increase in degenerating myofibers in the mdx/h 2/2BMmdx (Fig. 7A, g), as compared to mdx/h 2/2BMmdxh2/2 (Fig. 7A, c), parallel to a reduction in eMyHC expressing fibers (Fig. 7A, h) to levels comparable to mdx, as shown by quantitative analyses (Fig. 7B). Accordingly, iNOS expression, as a marker of macrophage infiltration, increased, as well as both NFkB and JNK expression and activity (Fig. 7C). However, the observed worsen phenotype was not translated in worsened exercise performance, as both mdx/h 2/2BMmdx and mdx/h 2/2BMmdxh2/2 behaved similar to WT mice in a treadmill endurance test (Fig. 7D).

Discussion
We show in this article that lack of PKCh in mdx prevents muscle wasting while improving muscle structure, regeneration and performance. This phenotype is associated to, and, probably, dependent on, reduced inflammation, which could make a more favourable environment for muscle precursor cells to differentiate.
The mdx mouse strain is the most widely used animal model for DMD; although it presents a milder phenotype compared to DMD, the immune cell populations in their muscles resemble those seen in DMD patients [6]. As first, we found that PKCh is highly phosphorylated/active in mdx muscle as compared to WT muscle, and its lack significantly reduced muscle necrosis evaluated as EBD positive fibers in both DIA and TA. Interestingly, not only the percentage of degenerating myofibers was reduced, but also were sites of IgG accumulation. As PKCh deficient mice have been shown to mount a significantly reduced lung inflammation response to antigen challenge and exhibit reduced inflammation in rodent models of arthritis as well as in autoimmune disease, the possibility that the improved phenotype was dependent on its activity on promoting inflammation is thus reasonable. Indeed, we show that lack of PKCh in mdx resulted in significantly reduced macrophages infiltration, as revealed by both IHC and FACS analysis. Macrophages are the primary immune cells evident in skeletal muscle of mdx mice together to T-cells and neutrophils, as being macrophage infiltration the most prominent immune feature observed [6]. The reduction in macrophages is probably responsible for the reduction in the MMP-9 accumulation observed. Matrix metalloproteinases (MMPs), a family of zincdependent endopeptidases, have been shown to play an important role in ECM degradation, inflammation, fibrosis, and activation of latent cytokines and cell adhesion molecules in different pathophysiological conditions, including muscular dystrophies [28]. In particular, the expression of MMP-9 is increased in dystrophic muscle and its inhibition, either genetic or pharmacological, considerably reduces inflammatory response, fibrosis, and enhances the myofiber regeneration in mdx mice [28]. Moreover, we show that lack of PKCh prevented the hyper-activation of the proinflammatory pathways NFkB and AP1, known to be up-regulated in mdx. It is worth noting that chronic activation of NF-kB signalling is required for DMD pathology by acting both on immune cells and damaged skeletal muscles to promote inflammation and to inhibit myogenic differentiation of muscle precursors [29]. Genetic and pharmacological studies demonstrated that inhibition of NFkB signalling reduced inflammation and necrosis enhancing regeneration and implicated the NFkB signalling pathway as a potential therapeutic target for this disease [11,30]. Also AP-1 activity is increased early in mdx muscle, and is supposed to be implicated in the activation of many inflammatory cytokines and chemokines [31]. Thus, the observed reduction in these pathways in mdx muscle lacking PKCh may contribute to favour muscle healing. It is well known that the inflammatory response negatively contributes to the limited ability of dystrophic muscle to regenerate. Indeed, muscle regeneration is the initial response to muscle damage, but, in Duchenne muscular dystrophy, muscle progenitor cells activation/differentiation is limited [32,33]. This fact depends mainly to the exhaustion of muscle progenitor cells pool, due to the continuous cycles of degeneration/regeneration [32,33], but also chronic inflammation contributes to make an unfavourable environment for their activation/differentiation. As we show here, lack of PKCh in mdx actually favoured muscle regeneration, as demonstrated by the increase in eMyHC expressing fibers and by myogenin upregulation. In this context, sustained myogenin expression in mdx/ h 2/2 muscle may theoretically contribute to myofiber metabolic changes, which, in turn, may be involved in muscle wasting prevention and physiology improvement. In fact, several observations suggested that myogenin may participate in at least a part of a fast-to-slow fiber-type transition [34,35]. Although further Extension of regenerating, eMyHC +ve , area (c) and of necrotic, EBD +ve , area (f) in mdx/h 2/2 , expressed as the percentage in respect to the respective areas in mdx (assumed as 1); *p,0.01, n = 3/genotype. Bar = 200 mm. (B) Representative Western Blot analysis of total protein fraction of TA muscles derived from 2 mo old WT, mdx and mdx/h 2/2 mice, as indicated, incubated with the a-myogenin antibody; Red Ponceau staining of the membrane is shown for equal loading. Up-regulation of myogenin expression in mdx/h 2/2 , in respect to mdx (assumed as 1), muscles, as determined by densitometric analysis of three independent experiments is shown in the bottom (n = 3/genotype) (C) Representative Wright staining of mdx-and mdx/h 2/2 -muscle derived cells, as indicated, cultured in DM for 48 hrs. The mean number of nuclei contained within each myotube is shown, as well as the percentage of reduction in mdx/h 2/2 in respect to mdx, as determined from three independent experiments. doi:10.1371/journal.pone.0031515.g004 analyses are required, no alterations in the level of metabolic enzymes were observed in mdx/h 2/2 muscle, as compared to mdx (Madaro et al. unpublished observation). The observed myogenin up-regulation should thus primarily reflect the increase in differentiating satellite cells, as a feature of regenerating muscle, rather than a sustained myogenin expression. These results may appear in contrast with our previous observation showing that PKCh 2/2 regenerating muscle displayed the characteristics of delayed regeneration, as compared to time-matching regenerating WT muscle. However, those effects were not due to an impairment in muscle precursors activation/differentiation, rather to an impairment of the late phases of growth/regeneration process, delaying further addition of myonuclei to newly formed myofibers, due to reduction in the expression of the so-called profusion genes. Indeed, we show in this paper that primary cells derived from mdx/h 2/2 muscle formed thinner myotubes with reduced number of myonuclei in vitro, as we observed in PKCh 2/2 muscle cells, when compared to those derived from mdx muscle. It is thus conceivable that in mdx chronic inflammation contributes significantly in preventing activation of muscle precursor cells, which however are able to efficiently differentiate, as they do ex vivo. By contrast, lack of PKCh, though delaying the fusion process itself, makes a more favourable environment for muscle precursors cells to differentiate by reducing inflammation, thus allowing muscle repair. Indeed, mdx/h 2/2 regenerating myofiber CSA was reduced, as compared to mdx. As a result, muscle tissue structure is preserved, though the myofiber CSA variability and the percentage of centrally-nucleated myofibers were similar to those in mdx. Importantly, lack of PKCh preserved exercise performance in mdx, suggesting that the improved phenotype correlates to improved functionality of the muscle. Taken together, these results demonstrate that lack of PKCh in mdx mice dramatically reduces cell infiltration and inflammation, improving muscle regeneration and performance, suggesting a hitherto unrecognized crucial role of PKCh in promoting immune response in muscular dystrophy. Consistent with this hypothesis, PKCh expressing BM derived cells partly counteracted improvement in inflammation, muscle wasting and regeneration in mdx/h 2/2 mice, as shown by BM transplantation experiments. Intriguingly, exercise performance was not impaired, instead. This fact may depend on the time required for muscle to decline its functionality: it is conceivable that 6 weeks upon BM transplantation were sufficient to partly restore immune response, which however was still less than in age-matching mdx, but not to worsen muscle performance. Although the specific cell population(s) involved is not clear yet, an attractive hypothesis would be that lack of PKCh prevents T eff cells activity, while maintaining T reg cells activity, which in turn may modulate macrophages phenotype and activity. However, as PKCh lacks in all cells in the mdx/h 2/2 model, the possible contribution of other tissue components, in particular skeletal muscle, in the observed phenotype cannot be ruled out. In any case, the results obtained demonstrate that PKCh expression/activity in immune cells is required for the robust inflammatory response in mdx, which, in turn, exacerbates the muscle pathology. Further studies are needed to verify long term efficacy, as well as the effect of targeting PKCh in older animals, when pathological features are already established. Moreover, as being muscular dystrophy a multifactorial disease, in which inflammation plays a crucial role, the possibility to combine PKCh targeting as anti-inflammatory therapy, to other gene-or cell-based therapeutic approaches to significantly improve and optimize the therapeutic efficacy, should be explored [5]. The availability, and the ongoing development, of  specific PKCh inhibitors, some of which are already in clinical trials for immune disorders, thus opens new perspectives for pharmacological approach to muscular dystrophy [15,19].

Mice
PKCh 2/2 mice were gently provided by Prof Dan Littman [36]. Mdx mice were purchased from Charles River. The mice were cross-bread to generate Mdx/h 2/2 double mutant mice. C57BL10 WT mice (Charles River) were used as control. The animals were housed at the Histology Unit accredited animal facility, in individual cages in an environmentally controlled room (23uC, 12-h light-dark cycle) and provided food and water ad libitum. All the procedures were approved by the Italian Ministry for Health and conducted according to the US National Institutes of Health guidelines.

Antibodies
The anti-PKC-h and the anti-phospho Thr538 PKC-h, the anti-p65NFkB, the anti-phospho p65NFkB, the anti-JNK and the antiphospho JNK rabbit polyclonal antibodies were purchased from Cell Signalling Inc., Danvers, MA, USA; the anti-iNOS mouse monoclonal antibody was from BD Bioscience, CA, USA, while the anti-dystrophin was from Leica Microsystems, Germany; the anti-myogenin F5D and the anti-embryonic myosin heavy chain F1.652 mouse monoclonal antibodies were from Developmental Studies Hybridoma Bank, Iowa City, Iowa, USA.

Primary muscle cell cultures
Primary cultures were prepared from total limb muscles of mdx or mdx/h 2/2. mice, as previously described [27]. Muscle derived cells were grown on collagen-coated dishes, in growth medium, GM (Dulbecco's modified Eagle's medium, D-MEM containing 20% HS, 3% EE, all from Gibco Invitrogen, Carlsbad, CA, USA) in a humidified 5% CO 2 atmosphere at 37uC. Differentiation was induced by replacing the medium with medium containing lower serum and EE concentration, DM (D-MEM containing 5% HS, 0,75% EE). The cells were fixed after 48 hrs in DM and stained with Wright's solution. The mean number of nuclei contained within each myotube was determined as previously described [27]. Approximately 100 myotubes were counted per dish.

Histological and immunohistochemical analyses
Muscle cryosections were fixed in 4% paraformaldehyde (Sigma-Aldrich, MO, USA) on ice. Myofiber necrosis was evaluated on muscle cryosections prepared from mice intraperitoneally injected with a 1% Evan's Blue Dye solution (EBD, Sigma-Aldrich, MO, USA), at 1% volume relative to body mass, between 16 and 24 h prior to tissue sampling. The samples were analyzed under an epifluorescence Zeiss Axioskop 2 Plus microscope. For histological analysis, muscle cryosections were stained with Hematoxylin/Eosin solution (Sigma-Aldrich, MO, USA). The muscle fiber mean cross sectional area (CSA) was determined by measuring CSA of fibres in the entire section, using Scion Image 4.0.3.2 software (NIH, Bethesda, MA, USA). Myofiber variability was determined by multiplying the standard deviation of all measurements by 1,000 and dividing it by the mean fiber diameter [37]. Esterase localization and activity was evaluated on cryosections by esterase staining (a-naphthyl butyrate/hexazotized pararosaniline).
Immunofluorescence analysis on crysections was performed as previously described [27]. Nuclei were counterstained with Hoechst 33342 (Fluka, WI, USA) and the samples were analyzed under an epifluorescence Zeiss Axioskop 2 Plus microscope.

Western blot analysis
Tissue samples were homogenized in ice-cold buffer containing 20 mM Tris (pH 7.5), 2 mM EDTA, 2 mM EGTA, 250 mM sucrose, 5 mM DTT, 200 mg/ml leupeptin, 10 mg/ml Trasylol, 1 mM PMSF, and 0.1% Triton X-100, as previously described [27]. An equal amount of protein from each sample was loaded onto 10% SDS-polyacrylamide gels and transferred to a nitrocellulose membrane (Schleicher and Schuell, Dassel, Germany). The membranes were incubated with the appropriate primary antibodies, and processed as described [27]. Densitometric analysis was performed using the Aida 2.1 ImageH software.
Flow Cytometry Analysis. Mononuclear cell population was isolated from dissected muscles by enzymatic digestion. 10 6 cells were incubated on ice with 1 mg of the anti-Mac-3 FITC-labelled and of the anti-CD45 PE-labelled antibodies (BD Biosciences, CA, USA) and analysed with a FacsStar Plus cytofluorimeter. Nonspecific fluorescent emission was determined with the specific labelled isotypes.

Treadmill exercise performance
Mice were first acclimated to the treadmill (LE 8710, PanLab S.L.U., Barcelona, Spain) before running by placing them on an Figure 7. Rescue of PKCh expression in hematopoietic cells in mdx/h 2/2 mice restores mdx mice phenotype. (A) H/E (a and e) and esterase (b and f, arrows indicate cell infiltrates, asterisks indicate neuromuscular junctions) as well as EBD uptake (c and g) and eMyHC immunofluorescence (d and h, Hoechst was used to counterstain the nuclei) of TA cryosections derived from mdx/h 2/2BMmdx/h2/2 (a-d) and from mdx/h 2/2BMmdx (e-h) mice. Bar = 200 mm. (B) Extension of regenerating, eMyHC +ve , area and of necrotic, EBD +ve , area in mdx/h 2/2BMmdx/h2/2 and in mdx/h 2/2BMmdx expressed as the percentage in respect to the respective areas in mdx (assumed as 1), *p,0.05, n = 3/genotype; (C) Western Blot analysis of iNOS expression, and of NFkB and JNK expression and activation (phosphorylation) in protein extracts from TA muscle derived from mdx/ h 2/2BMmdx/h2/2 and from mdx/h 2/2BMmdx ; red Ponceau staining of the membrane is shown (loading) for equal loading. Representative blots are shown. The level of expression or of the activation of the above molecules was determined by densitometric analysis and expressed as fold induction in mdx/h 2/2BMmdx in respect to mdx/h 2/2BMmdx/h2/2 , assumed as 1, evaluated from 3 separate experiments (bottom). *p,0.05 (D) Treadmill exercise test performed on mdx/h 2/2BMmdx/h2/2 (squares, dotted grey line) and mdx/h 2/2BMmdx (circles, grey.line) mice for a 30 min running, twice a week, for 3 weeks, as above, starting 6 weeks after transplantation (n = 3/genotype). Performance of mdx (triangles, black line) and mdx/h 2/2 (rhombi, dotted black line) mice is also reported for comparison. doi:10.1371/journal.pone.0031515.g007 unmoving treadmill for 10 min. The treadmill was set at a speed of 15 cm/sec. The test ran for 30 min, and the number of times at which mice failed to keep running, was recorded. Each test was performed twice a week for three weeks, and the results were averaged for each mouse.
Bone Marrow transplantation 1 mo old mdx/h 2/2 mice were X-irradiated (8 Gy per mouse) by a Siemens linear accelerator operating at 10 MV, at a rate of 3 Gy/min. At 2 h postirradiation, the mice were intravenously injected with bone marrow cell suspension derived from agematching mdx or mdx/h 2/2 mice. Mice were sacrificed 6 weeks after transplantation, and bone marrow, thymus and spleen were evaluated for exogenous cells accumulation. Different muscles were dissected for morphological and molecular analyses.

PCR genotyping
Genomic DNA PCR, for PKCh 2/2 mutation, was conducted as previously described [36]. A 700 bp PCR fragment is expected for the mutated PKCh gene, while a 400 bp PCR fragment for the WT one.

Statistical analysis
Quantitative data are presented as means 6 SEM or 6 SD (as specified) of at least three independent experiments. Statistical analysis to determine significance was performed using paired Student's t tests. Differences were considered to be statistically significant at the p,0.05 level.

Aknowledgments
We thank Prof. DA. Littman, NYU, NY, USA, for providing the PKCh mutant mice; we also thank Prof. Antonio Musarò, for advice in functional studies, Prof. Mario Molinaro, for continuous discussion and critical reading of the manuscript, and Dr. Fabrizio Padula, for FACS analysis, all from DAHFMO, Histology Unit.