Abstract X-linked myotubular myopathy (XLMTM) is a severe congenital disorder in male infants that leads to generalized skeletal muscle weakness and is frequently associated with fatal respiratory failure. XLMTM is caused by loss-of-function mutations in the MTM1 gene, which encodes myotubularin, the founder member of a family of 15 homologous proteins in mammals. We recently demonstrated the therapeutic efficacy of intravenous delivery of rAAV vectors expressing MTM1 in animal models of myotubular myopathy. Here, we tested whether the closest homologues of MTM1, MTMR1, and MTMR2 (the latter being implicated in Charcot-Marie-Tooth neuropathy type 4B1) are functionally redundant and could represent a therapeutic target for XLMTM. Serotype 9 recombinant AAV vectors encoding either MTM1, MTMR1, or MTMR2 were injected into the tibialis anterior muscle of Mtm1-deficient knockout mice. Two weeks after vector delivery, a therapeutic effect was observed with Mtm1 and Mtmr2, but not Mtmr1; with Mtm1 being the most efficacious transgene. Furthermore, intravenous administration of a single dose of the rAAV9-Mtmr2 vector in XLMTM mice improved the motor activity and muscle strength and prolonged survival throughout a 3-month study. These results indicate that strategies aiming at increasing MTMR2 expression levels in skeletal muscle may be beneficial in the treatment of myotubular myopathy. AAV vector, Mouse model, MTM1, MTMR1, MTMR2, Myotubular myopathy, Myotubularin INTRODUCTION Centronuclear myopathies are a group of rare congenital skeletal muscle disorders classified together because of the presence of centrally localized nuclei within hypotrophic myofibers (1). The most severe and frequent form of these myopathies, X-linked myotubular myopathy (XLMTM, OMIM 310400) is a life-threatening disease affecting 1 in 50 000 male infants. XLMTM is characterized by severe hypotonia, generalized muscle weakness and respiratory failure at birth (2). Even though intensive care measures including ventilation and feeding support ameliorate the clinical status beyond the postnatal period, the disease is often fatal within the first years of life (2). Myotubular myopathy is a hereditary monogenic disease caused by loss-of-function mutations in the MTM1 gene (3). Myotubularin, the encoded protein, is a phospholipid phosphatase that dephosphorylates the D3′ position of the inositol ring of phosphatidylinositol-3-phosphate (PI3P) and phosphatidylinositol-3, 5-bisphosphate [PI(3, 5)P2] (4–6). Although myotubularin is ubiquitously expressed (3), the pathology caused by its deficiency is mainly muscular. Myogenesis takes place, but muscle fibers are hypotrophic and weaker (7–10). They present profound structural abnormalities, such as mislocalization of nuclei and mitochondria (7–11), disruption of the intermediate filament network (12), triad disorganization (8–10, 13), and a reduction in the number and proliferation capacity of satellite cells (14). From a functional point of view, deficiencies in neuromuscular transmission (15, 16), excitation-contraction coupling (8, 9), and mitochondrial function (12) were previously suggested to account for muscle weakness and fatigability. Although myotubular myopathy has no curative treatment in patients to date, several experimental therapies are currently being explored (15–20). We recently demonstrated that systemic AAV-mediated MTM1 gene therapy could largely improve the XLMTM phenotype and prolong survival in mouse and dog models of the disease (19, 21). Interestingly, myotubularin is the archetypical and founding member of a family of 15 homologous proteins, the myotubularin-related proteins (MTMR) (22). Like myotubularin, some MTMR proteins (MTMR1–4, MTMR6–8, MTMR14) display lipid phosphatase activity toward PI3P and PI(3, 5)P2 and contain the conserved C(X)5 R active motif in the catalytic site, whereas others are inactive due to amino acid substitutions within this catalytical motif (MTMR5, MTMR9–13) (22). The MTMR proteins present some degree of functional redundancy (22, 23), suggesting that overexpression of a homologous protein might rescue myotubularin deficiency. This approach could define a novel molecular target for drug development in myotubular myopathy with the goal of increasing the expression level of the endogenous gene, or by exogenous delivery. Additionally, these proteins would offer alternative gene therapy strategies for situations where myotubularin replacement is not well tolerated (which may occur in patients who express no myotubularin protein). Amongst the MTMR proteins, myotubularin-related protein 1 (MTMR1) and 2 (MTMR2) appear as the most relevant for that application because the 2 proteins belong to the same phylogenetic subgroup than myotubularin (24); they share the highest level of structural similarity and display 3-phosphatase activity towards PI3P (4, 5, 23, 25–29) and PI(3, 5)P2 (6, 26, 28–30). In the present study, we compared the therapeutic effect of these candidates by gene transfer and demonstrate that intramuscular delivery of a rAAV9 vector encoding MTMR2 rescues the muscle pathology in a murine model of myotubular myopathy, while the benefit of MTMR1 overexpression remains only marginal. Furthermore, whole body treatment of Mtm1 knockout (KO) mice by intravenous injection of a single dose of a serotype 9 recombinant adeno-associated viral vector (rAAV9) expressing the Mtmr2 gene ameliorated muscle function and prolonged lifespan, indicating that MTMR2 can compensate myotubularin deficiency in skeletal muscle. MATERIALS AND METHODS Generation and Titration of Recombinant AAV Vectors Murine Mtmr1, Mtmr2, and Mtm1 coding sequences were cloned downstream of the human desmin promoter in an AAV2 plasmid backbone by PCR amplification. Pseudo-typed recombinant rAAV2/9 (rAAV9) viral preparations were generated by packaging AAV2-inverted terminal repeat (ITR) recombinant genomes into AAV9 capsids. Briefly, the cis-acting plasmids encoding one of the transgenes pAAV2-Des-Mtm1, -Mtmr1, or Mtmr2, a trans-complementing rep-cap9 plasmid encoding the proteins necessary for the replication and structure of the vector and an adenovirus helper plasmid were cotransfected into HEK293 cells. Vector particles were purified through 2 sequential cesium chloride gradient ultra-centrifugations and dialyzed against sterile phosphate-buffered saline (PBS). DNAse I resistant viral particles were treated with proteinase K. Viral titers were quantified by a TaqMan real-time PCR assay (Applied Biosystems, Foster City, CA), with primers and probes specific for the ITR2 region (31), and expressed as viral genomes per ml (vg/ml). The primer pairs and TaqMan MGB probes used for ITR2g amplification were: Forward 5′-CTCCATCACTAGGGGTTCCTTGTA-3′; reverse 5′-TGGCTACGTAGATAAGTAGCATGGC-3′; and MGB/taqman probe 5′-GTTAATGATTAACCC-3′. Animal Care and Experimental Design All mice were handled according to the European guidelines for the humane care and use of experimental animals and procedures were approved by the institutional ethical committee (number CE 12-006). The Mtm1-KO murine strain obtained by constitutive deletion of Mtm1 exon 4 was previously described (7), and maintained in a 129SvPasIco background (9). Wild-type (WT) littermates were used as control animals. For the intramuscular administration study, a volume of 10 µl of rAAV9-Mtmr1 (3 × 1010 vg per injection), rAAV9-Mtmr2 (2.4 × 1010 vg), rAAV9-Mtm1 (3 × 1010 vg), or PBS was injected into the tibialis anterior (TA) muscle of 2-week-old male KO animals. WT male littermates were treated with an equivalent volume of PBS. Two weeks after injection, TA muscles were sampled, quickly frozen in isopentane cooled with liquid nitrogen and stored at −80°C. For the intravenous administration, a dose of 2.4 × 1014 vg/kg of body weight of rAAV9-Mtmr2 was injected in the tail vein of 2-week-old male KO animals. Age-matched WT animals received an equivalent volume of PBS. Analyses were performed 3 months after the injections. The number of animals in each experimental procedure and subgroup are included in the figure legends. Quantitative RT-PCR Total RNA was purified from whole TA muscles using TRIzol reagent (Life Technologies, Saint Aubin, France), according to the manufacturer’s instructions. RNA concentration was determined by measure of the optical density at 260 nm and RNA integrity was confirmed by electrophoresis using an agarose gel containing GelRed as a nucleic acid stain (Biotium, Hayward, CA). Genomic DNA was degraded with a DNAse treatment (Thermo Scientific, Rockford, IL), and cDNA was synthesized from RNA using RevertAid H minus Reverse Transcriptase (Thermo Scientific, Waltham, MA) in the presence of Random Primers (Thermo Scientific). Real-time PCR was performed using SYBR green (Life Technologies) as a nucleic acid stain and an ABI Prism 7900 apparatus (Applied Biosystems). The primers used are: Mtmr1 Forward 5′(GGCACAGATGGAAGAAGCTC)3′, reverse 5′ (CCACTCTGCTGATCACTCCA)′; Mtmr2 Forward 5′(TGCTGACAAGGTCGAGTCTG)3′, reverse 5′(CTGCGTCAGCATGGTTCTTA)3′; Chrn-α1 Forward 5′(GAATCCAGATGACTATGGAG)3′, reverse 5′(GACAATGATCTCACAGTAGC)3′; Chrn-δ Forward 5′(AACGTGTGGATAGATCATGC)3′, reverse 5′(CATAGACAAGCACATTGCAGG)3′; Chrn-γ Forward 5′(ATCCGGCACCGACCGGCTAA)3′, reverse 5′(CATTTCTGCCCGCCCGCCTT)3′. The ubiquitous acidic ribosomal phosphoprotein P0 (Rplp0) was used to normalize the data across samples, and primers are: Forward 5′(CTCCAAGCAGATGCAGCAGA)3′, reverse 5′(ATAGCCTTGCGCATCATGG)3′. Western Blot For intramuscular administration studies, thirty micrometer of muscle cryo-sections were sliced and proteins were extracted using 50 µl of a lysis buffer containing 10 mM Tris HCl pH 7.4, 150 mM NaCl, 5 mM EGTA, 2 mM sodium orthovanadate, 100 mM NaF, 4 Mm Na pyrophosphate, 1% Triton X-100, 0.5% IGEPAL, and Protease Inhibitor Cocktail used according to the manufacturer’s instruction (Roche Applied Science, Meylan, France). For intravenous administration studies, muscle proteins were extracted in the same buffer with a ratio of 200 µl per 30 mg of frozen muscle and homogenized with a FastPrep-24 sample preparation system (MP Biomedicals, Illkirch, France). In each case, samples were incubated on ice with occasional mixing for 30 minutes. After centrifugation at 12 000g for 10 minutes at 4°C, the supernatants were recovered for Western blot analysis. Total protein concentration was determined by the Bradford methodology according to the manufacturer’s instruction (BioRad, Marnes-la-Coquette, France). Fifty to 80 µg of total proteins were denatured for 5 minutes at 95°C in a buffer containing 125 mM Tris HCl pH 6.8, 4% SDS, 0.2 M DTT, 50% glycerol, and bromophenol blue. Protein samples were submitted to sodium dodecyl sulfate-polyacrylamide gel electrophoresis in 10% acrylamide gels and transferred onto PVDF membranes (GE Healthcare, Aulnay-Sous-Bois, France) by the application of an electric field (100 V, 1 hour) at 4°C. The membranes were first incubated for 60 minutes at room temperature in a blocking solution composed of 50% Tris-buffered saline (TBS), 0.1% Tween 20, and 50% Odyssey blocking buffer (LI-COR, Lincoln, NE) and then probed successively with a custom-made rabbit polyclonal antibody raised against the C-terminal extremity of murine MTMR1 (peptide CGSSPTHSATPVHTSV), MTMR2 (peptide CASSPAQCVTPVQTVV), and MTM1 (17) (all used at a dilution of 1:2000), and a mouse monoclonal antibody specific for α-tubulin (dilution 1:2000; Millipore, Billerica, MA). Antibody incubations were carried out for 60 minutes at room temperature or overnight at 4°C in TBS, 0.1% Tween 20, and 1% milk. Detection was performed with a secondary goat antirabbit antibody coupled to IRDye 800 nm (dilution 1:10 000; LI-COR) for MTMR1, MTMR2 and MTM1, and with a secondary goat antimouse antibody coupled to AlexaFluor 680 nm (dilution 1:10 000; Life Technologies) for α-tubulin and the membranes were exposed to the Odyssey infrared imaging system (LI-COR) for detection and quantification of the fluorescent signal (Odyssey software, version 1.2, 2003). Measurement of PI3P Level In intramuscular studies, the level of PI3P was measured by immuno-staining on 8-µm-thick TA slices. The tissue was fixed for 20 minutes at -20°C in cold acetone and epitopes were retrieved by boiling the sections for 10 minutes in PBS. After 15-minute permeabilization in 0.1% Triton in TBS, unspecific sites were blocked by incubating the sections for 45 minutes at 37°C in 10% donkey serum diluted in 0.1% Triton in TBS. PI3P staining was performed with a PI3P-specific mouse monoclonal antibody (1:250 in 0.1% Triton in TBS; Echelon Biosciences Incorporated, Salt Lake City, UT) for 1 hour at 37°C. After TBS washes, the sections were incubated 2 hours at room temperature with an Alexa Fluor 594-conjugated goat antimouse antibody (1:500 in TBS; Life Technologies), glass slides were mounted with Fluoromount reagent (SouthernBiotech, Birmingham, AL); the staining was visualized with a 63× 1.4NA immersion objective on a Leica confocal microscope TCS-SP2 (Leica Microsystems GmbH, Wetzlar, Germany). On 4 randomly selected fields, intracellular subsarcolemmal regions of interest (ROIs) were manually drawn inside 10 representative fibers and the mean value of pixel intensity measured using ImageJ (Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda, MD, https://imagej.nih.gov/ij/, 1997–2016). In intravenous studies, lipids were extracted from whole soleus muscles and PI3P was measured using a proprietary PI3P-specific ELISA method (Echelon Biosciences, Inc.). Briefly, frozen soleus were powdered and incubated in 150 µl of Western blot lysis buffer for 45 minutes at 4°C. Lipids were extracted by the addition of 900 µl of a chloroform/methanol/0.1 M HCl mixture (1/1/1). After vigorous shaking, the lower organic phase was separated by a 2-minute 10 000g centrifugation, sampled, and a second extraction carried out with the addition of 240 µl of methanol/1 M HCl mix (1/1). After settling, the lower phase was dried using a speed-vac concentrator system (Thermo Fisher Scientific, Villebon Sur Yvette, France), and kept at -80°C until resuspension and PI3P assay according to the manufacturer’s instructions. Histology, Morphometry, and Immuno-Stainings Histology and Morphometry Serial 8-µm-thick transverse cryo-sections were prepared from frozen muscles. Hematoxylin and eosin (H&E) staining was performed using standard procedures. The proportion of internalized nuclei was quantified from H&E-stained sections using the Histolab software (Microvision, Evry, France). Nicotinamide adenine dinucleotide tetrazolium reductase (NADH-TR) coloration was performed in order to visualize the localization of mitochondria and endoplasmic reticulum. Muscle cryo-sections were incubated for 10 minutes at 57°C in a solution composed of 50 mM Tris HCl, pH 7.3, 1.2 mM Nitroblue Tetrazolium (Sigma-Aldrich, Lyon, France) and 0.6 mM β-NADH (Sigma-Aldrich), washed, and mounted in Eukitt (Sigma-Aldrich). For determination of the number and minimal diameter of fibers, laminin immunostaining (antibody) was performed to outline each fiber. Briefly, endogenous peroxidases were neutralized by incubation in H2O2 for 30 minutes. Sections were incubated for 30 minutes with 10% goat serum in PBS (DAKO, Trappes, France) in order to block unspecific sites and then overnight in a 1:1000 dilution of antilaminin rabbit polyclonal antibody (DAKO) at room temperature. After PBS washes, sections were incubated with a goat antirabbit secondary antibody conjugated with horseradish peroxidase (Kit En Vision Rabbit HRP, DAKO) for 30 minutes at room temperature, and the revelation was performed after additional PBS washes by incubation in 3, 3′-diaminobenzidine (DAB) for 2 minutes. Myofibers minor diameters were automatically measured on digital images of the sections using the Ellix software (Microvision). Pax 7 Immunostaining Pax 7 immunostaining was performed on sections of TA muscles fixed 5 minutes in 4% paraformaldehyde (PFA) in PBS, permeated 6 minutes in methanol at -20°C and treated for protein refolding in citric acid at 100°C for 10 minutes. Endogenous peroxidases were inhibited, and nonspecific sites were blocked using 5% BSA for 2 hours and Mouse-On-Mouse ([MOM], Vector Laboratories, Burlingame, CA) blocking reagent for 30 minutes. Sections were incubated overnight in murine antibodies specific for Pax7 (DSHB, Iowa City, IA; dilution 1:20 in 4% BSA in PBS), and 45 minutes in biotinylated goat antimouse antibodies (1:1000 in 4% BSA in PBS; SouthernBiotech). Biotin was revealed using the peroxidase-conjugated Vectastain system (Vectastain ABC system, Vector Laboratories) and its DAB substrate. Muscle sections were counterstained with eosin, mounted and pax7-positive fibers were quantified with the Histolab software. Immunofluorescence Transverse muscle sections were fixed for 10 minutes by incubation either in PBS at 100°C for the dihydropyridin receptor alpha (DHPRα1) staining or for 7 minutes in 4% PFA in PBS at 4°C for the ryanodine receptor 1 (Ryr1) and desmin staining, respectively. Nonspecific antigens were blocked at room temperature for 1 hour with PBS containing either 3% bovine serum albumin (DHPRα1 and desmin detection) or MOM blocking reagent (Ryr1 detection). Sections were incubated in mouse primary monoclonal antibodies directed against DHPRα1 (dilution 1:400, overnight at 4°C; Thermo Scientific) or Ryr1 (dilution 1:100, 1 hour at room temperature; Abcam, Cambridge, MA) or in goat polyclonal antibodies specific for desmin (1:100, 1 hour at room temperature; Santa Cruz Biotechnology, Heidelberg, Germany). After extensive PBS washes, sections were incubated with donkey antigoat antibodies conjugated with Alexa Fluor 488 (dilution 1:1000; Life Technologies) for desmin staining, or with biotinylated goat antimouse antibodies (dilution 1:200; SouthernBiotech) and, after additional washes, with Alexa Fluor 488-bound streptavidin (1:1000; Life Technologies) for DHPRα1 and Ryr1 stainings, respectively. Glass slides were mounted with FluorSave reagent (Calbiochem, Merck Millipore, Darmstadt, Germany) and visualized with a Leica confocal microscope TCS-SP2 (Leica Microsystems GmbH). Electron Microscopy Electron microscopy was performed as previously described (9, 18) for the evaluation of sarcotubular architecture. Briefly, glutaraldehyde fixed skeletal muscle specimens were processed at the Medical College of Wisconsin (MCW) EM Core facility for the evaluation of triad morphology. After confirming appropriate tissue quality on scout sections, longitudinally oriented 30-nm sections were stained with 2% uranyl acetate and Reynold’s lead citrate, and electron microscopy was performed using a Hitachi H600 transmission electron microscope. Longitudinally oriented muscle fibers were evaluated and photographed, with photography of the single best-oriented region from each myofiber in the specimen at 10 000×, 20 000×, and 30 000× magnification. Ten fibers per specimen were evaluated and photographed in this manner. The number of transverse (T)-tubules, longitudinal (L)-tubules, and triad structures were then manually quantified by a blinded, board-certified neuropathologist using the 20 000× magnification image. The values for all images in a given treatment condition were pooled together to generate average numbers of T-tubules, L-tubules and triads in that condition. Measurement of Muscle Function Actimeter Test Murine spontaneous locomotor activity was measured with the LE 8811 IR motor activity monitor (Bioseb, Chaville, France). Briefly, mice were placed in an open field bounded with 16 × 16 horizontal photoelectric Infrared beams able to measure 3-dimensional movements of the animals. The field is shielded from external noise and light, illuminated with a white fluorescent light and fully ventilated. The distance crossed was recorded for the 90 minute-course of the test. Escape Test The global strength of mice was evaluated by the “escape test” (32). Briefly, mice were placed on a platform facing the entrance of a 30-cm-long tube. A cuff wrapped around the tail was connected to a fixed force transducer and the mice were induced to escape within the tube in the direction opposite from the force transducer by a gentle pinching of the tail. A short peak of force was induced by this flight forward and the average of the 5 highest force peaks normalized by body weight were analyzed. EDL Contractile Force Measurements of isometric contractile properties of extensor digitorum longus (EDL) were performed in vitro according to methods previously detailed (33). Animals were anesthetized by intraperitoneal injection of pentobarbital (50 mg/kg). EDLs were surgically excised and soaked in an oxygenated Krebs solution (95% O2 and 5% CO2) containing 118 mM NaCl, 25 mM NaHCO3, 5 mM KCl, 1 mM KH2PO4, 2.5 mM CaCl2, 1 mM MgSO4, and 5 mM glucose maintained at a temperature of 20°C. Muscles were connected at one end to an electromagnetic puller and at the other end to a force transducer, and stimulation was delivered through electrodes running parallel to the muscle. Twitch and tetanic (125 Hz, 300 mseconds) isometric contractions were recorded at Lo (the length at which maximal tetanic isometric force is observed). For comparative purposes, normalized isometric force was assessed. Isometric tension was calculated by dividing the force by the estimated cross-sectional area of the muscle. TA Contractile Force Skeletal muscle function was evaluated by the measure of muscle contraction in situ, as previously described (34). Animals were anesthetized by intraperitoneal injection of Ketamine (100 mg/kg) and Xylazine (10 mg/kg) and supplemental doses were administered as required to maintain deep anesthesia. The knee and foot were fixed with clamps and stainless steel pins. The TA muscle was exposed and the distal tendon was cut and attached to a force transducer (Aurora Scientific, Dublin, Ireland). The sciatic nerves were proximally crushed and distally stimulated by a bipolar silver electrode using supramaximal square wave pulses of 0.1 msecond duration. Absolute maximal forces were determined at optimal length (length at which maximal tetanic tension was observed). The specific maximal force was calculated by normalizing the total muscle force with the muscle mass. Statistical Analysis Data are presented as means ± SE. Data were analyzed for normality using the Kolmodorov-Smirnov test and individual means were compared using the student parametric test or the Mann-Whitney nonparametric test. Differences were considered to be statistically significant at p < 0.05, p < 0.01, and p < 0.001. Video A video showing mice 3 weeks after intravenous injection of rAAV9-Mtmr2 is included as Supplementary Data: WT mouse, untreated Mtm1-KO mouse (KO + PBS), and rAAV9-Mtmr2-treated Mtm1-KO mouse (KO + AAV). RESULTS Effect of Mtmr1 and Mtmr2 Gene Delivery in the Skeletal Muscle of Mtm1-KO Mice In order to test whether overexpression of MTMR1 and MTMR2 proteins could rescue the pathological signs of myotubular myopathy, rAAV9 vectors carrying either Mtmr1 (3 × 1010 viral genomes [vg]) or Mtmr2 (2.4 × 1010 vg) were injected into the TA muscle of 2-week-old Mtm1-KO mice (see viral titration data in Supplementary Data Fig. S1A). A vector encoding myotubularin (3 × 1010 vg/TA) was used as a positive control, as we previously showed the efficiency of this transgene in correcting the XLMTM phenotype (17, 19). PBS was administrated in the contralateral muscle of KO animals and WT littermates as additional controls. At the time of treatment, the mRNA levels of Mtmr1 and Mtmr2 did not show significant changes in Mtm1 KO muscles (Supplementary Data Fig. S2). Two weeks after vector injection, the mice were killed and TA muscles were sampled, weighed and processed for analyses. The vector copy number (viral genomes per diploid genome) in injected muscles was nearly identical amongst the various treatment conditions (4–6 vector copies, Supplementary Data Fig. S1B), and MTMR1 and MTMR2 protein levels, although undetectable in WT-PBS and KO-PBS TA, were strongly increased in AAV-treated KO muscles (Fig. 1A). FIGURE 1. View largeDownload slide Effect of intramuscular rAAV9-mediated Mtmr1 and Mtmr2 gene transfer in Mtm1 knockout mice. (A) Representative immunoblots of MTMR1 (top panel), MTMR2 (middle panel) and myotubularin (lower panel) protein levels (in green) in tibialis anterior of untreated (KO-PBS) and AAV-treated (KO-AAV) Mtm1-KO mice, and in saline-injected control muscle (WT-PBS). The immunodetection of α-tubulin was used as loading control (in red). (B) Weight of TA muscle from mutant mice 2 weeks after intramuscular delivery of rAAV9-Mtmr1, -Mtmr2 or -Mtm1, and appropriate controls (KO-PBS and WT-PBS). Results are also shown as a percentage of WT values (WT-PBS n = 41; KO-PBS n = 44; KO-Mtmr1 n = 8; KO-Mtmr2 n = 22; KO-Mtm1 n = 16). (C) Mean diameter of myofibers in PBS-, Mtmr1-, Mtmr2-, and Mtm1-injected tibialis anterior muscle (WT-PBS n = 13; KO-PBS n = 33; KO-Mtmr1 n = 7; KO-Mtmr2 n = 9; KO-Mtm1 n = 8). ***, p < 0.001 versus WT-PBS; Δ, ΔΔ, ΔΔΔ, p < 0.05, 0.01, or 0.001 versus KO-PBS; ♦♦, ♦♦♦, p < 0.01 or 0.001 versus KO-Mtm1. (D) PI3P level in TA muscles, A.U. arbitrary unit (WT-PBS n = 6; KO-PBS n = 5; KO-Mtmr1 n = 5; KO-Mtmr2 n = 6; KO-Mtm1 n = 6). *, **, ***, p < 0.05, 0.01 or 0.01 versus WT-PBS; Δ, ΔΔΔ, p < 0.05 or 0.001 versus KO-PBS; ••, p < 0.01 versus KO-Mtmr1; ♦, p < 0.05 versus KO-Mtm1. FIGURE 1. View largeDownload slide Effect of intramuscular rAAV9-mediated Mtmr1 and Mtmr2 gene transfer in Mtm1 knockout mice. (A) Representative immunoblots of MTMR1 (top panel), MTMR2 (middle panel) and myotubularin (lower panel) protein levels (in green) in tibialis anterior of untreated (KO-PBS) and AAV-treated (KO-AAV) Mtm1-KO mice, and in saline-injected control muscle (WT-PBS). The immunodetection of α-tubulin was used as loading control (in red). (B) Weight of TA muscle from mutant mice 2 weeks after intramuscular delivery of rAAV9-Mtmr1, -Mtmr2 or -Mtm1, and appropriate controls (KO-PBS and WT-PBS). Results are also shown as a percentage of WT values (WT-PBS n = 41; KO-PBS n = 44; KO-Mtmr1 n = 8; KO-Mtmr2 n = 22; KO-Mtm1 n = 16). (C) Mean diameter of myofibers in PBS-, Mtmr1-, Mtmr2-, and Mtm1-injected tibialis anterior muscle (WT-PBS n = 13; KO-PBS n = 33; KO-Mtmr1 n = 7; KO-Mtmr2 n = 9; KO-Mtm1 n = 8). ***, p < 0.001 versus WT-PBS; Δ, ΔΔ, ΔΔΔ, p < 0.05, 0.01, or 0.001 versus KO-PBS; ♦♦, ♦♦♦, p < 0.01 or 0.001 versus KO-Mtm1. (D) PI3P level in TA muscles, A.U. arbitrary unit (WT-PBS n = 6; KO-PBS n = 5; KO-Mtmr1 n = 5; KO-Mtmr2 n = 6; KO-Mtm1 n = 6). *, **, ***, p < 0.05, 0.01 or 0.01 versus WT-PBS; Δ, ΔΔΔ, p < 0.05 or 0.001 versus KO-PBS; ••, p < 0.01 versus KO-Mtmr1; ♦, p < 0.05 versus KO-Mtm1. The histopathological signs of the disease were already present in myotubularin-deficient mice at the time of vector administration (9), and became more pronounced when mice were 4 weeks of age, as evidenced by an ∼55% reduction of TA muscle weight (Fig. 1B). Expression of Mtmr1, Mtmr2, and Mtm1 transgenes resulted in an increase of muscle weight 2 weeks after vector injection (+14.4% for Mtmr1, +18.5% for Mtmr2, +86.7% for Mtm1, Fig. 1B). Detailed morphometric analysis of Mtmr1-treated muscles showed a small but significant increase in myofiber mean diameter (15.3 µm in rAAV9-Mtmr1 vs 13.5 µm in PBS-injected KO TA, p < 0.05), whereas the number of large fibers (diameter > 20 µm) remained unchanged (Fig. 1C). On the contrary, MTMR2 overexpression resulted in an increase in both muscle fiber diameter (mean 16.6 µm, i.e. +23%), and the amount of large myofibers (35% of fibers in rAAV9-Mtmr2 vs 15% in PBS-injected TA), and was therefore more efficient than MTMR1 in compensating myotubularin deficiency. MTMR2 but Not MTMR1 Restores PI3P Levels in Myotubularin-Deficient Muscle Increased levels of PI3P in muscle were shown to play a critical role in the pathogenesis of XLMTM, as normalization by pharmacological inhibition of phosphatidylinositol 3-kinase resulted in histopathological amelioration and prolonged survival of Mtm1-KO mice (35, 36). We therefore quantified the level of this phosphoinositide in TA muscle 2 weeks after the injection of either PBS or the Mtmr1, Mtmr2, or Mtm1 transgenes to see if vector response correlated with changes in PI3P levels. As already reported (8), the amount of PI3P was significantly increased in Mtm1-KO muscles (2.1-fold over WT-PBS values, p < 0.01; Fig. 1D). Interestingly, both Mtmr2 and Mtm1 gene transfer normalized PI3P levels in mutant muscle, whereas Mtmr1 overexpression had no significant effect (2.6-fold over WT-PBS level, p < 0.05). These results suggest that exogenous MTMR1 is inefficient in dephosphorylating critical PI3P subpools in muscle whereas MTMR2 is able to cross-correct myotubularin enzymatic deficiency in vivo. MTMR2 Overexpression Ameliorates the Internal Architecture of Muscle Fibers in Mtm1-KO Mice To evaluate the effect of the treatment on other histopathological hallmarks of the disease further, H&E, NADH-TR, and immunofluorescence stainings were carried out on TA cross-sections of 4-week-old Mtm1 mutant mice. In addition to hypotrophy, mutant muscle fibers contain internal nuclei and altered organization of intermyofibrillar mitochondria and endoplasmic reticulum, which accumulate in central areas or in the subsarcolemmal region (Fig. 2A) (37). Despite the presence of occasional internal nuclei, the architecture of myotubularin-deficient myofibers was largely ameliorated after AAV-mediated Mtmr2 gene transfer, whereas it remained unchanged by Mtmr1 overexpression. The percentage of fibers presenting internal nuclei, which was increased in PBS-injected KO TA (5.9-fold over WT-PBS values, p < 0.001), was equally reduced in presence of either MTMR2 or myotubularin 2 weeks after vector injection (respective diminution of 43% with MTMR2 and 35% with myotubularin; Fig. 2B). Given the absence of substantial therapeutic benefit from MTMR1 overexpression in Mtm1-KO muscles, only rAAV9-Mtmr2-treated muscles were further analyzed. FIGURE 2. View largeDownload slide MTMR2 but not MTMR1 overexpression improves the histopathological features of myotubularin-deficient muscle. (A) H&E and NADH-TR stainings of TA cross-sections from Mtm1-KO and WT mice 2 weeks after intramuscular delivery of PBS or rAAV9 vectors carrying either Mtmr1, Mtmr2, or Mtm1 transgenes. Nuclei internalization: Black stars on H&E; cytoplasmic and sub-sarcolemmal accumulations of mitochondria: White star and white arrow on NADH-TR staining, respectively. Scale bar = 20 µm. (B) Quantification of the percentage of myofibers containing internalized nuclei after rAAV9-Mtmr2 and rAAV9-Mtm1 treatment, and appropriate untreated controls. (WT-PBS n = 6; KO-PBS n = 13; KO-Mtmr2 n = 9; KO-Mtm1 n = 8). **, ***, p < 0.01 or 0.001 versus WT-PBS; Δ, p < 0.05 versus KO-PBS. FIGURE 2. View largeDownload slide MTMR2 but not MTMR1 overexpression improves the histopathological features of myotubularin-deficient muscle. (A) H&E and NADH-TR stainings of TA cross-sections from Mtm1-KO and WT mice 2 weeks after intramuscular delivery of PBS or rAAV9 vectors carrying either Mtmr1, Mtmr2, or Mtm1 transgenes. Nuclei internalization: Black stars on H&E; cytoplasmic and sub-sarcolemmal accumulations of mitochondria: White star and white arrow on NADH-TR staining, respectively. Scale bar = 20 µm. (B) Quantification of the percentage of myofibers containing internalized nuclei after rAAV9-Mtmr2 and rAAV9-Mtm1 treatment, and appropriate untreated controls. (WT-PBS n = 6; KO-PBS n = 13; KO-Mtmr2 n = 9; KO-Mtm1 n = 8). **, ***, p < 0.01 or 0.001 versus WT-PBS; Δ, p < 0.05 versus KO-PBS. Other distinctive features of myotubularin deficiency in skeletal muscle fibers include aberrant localization of 2 triadic proteins, the DHPR1α (T-tubule marker) and the Ryr1 (junctional sarcoplasmic reticulum marker), and desmin, an intermediate filament protein interacting with MTM1 (9, 12). These proteins were abnormally distributed within the sarcoplasm and subsarcolemmal region in Mtm1-KO myofibers (Fig. 3A). AAV-mediated MTMR2 or myotubularin expression restored normal localization of DHPR1α, RYR1, and desmin with comparable efficacies, suggesting restructuring of the sarcotubular and intermediate filament networks. FIGURE 3. View largeDownload slide Restoration of XLMTM muscle biomarkers by Mtmr2 gene transfer. (A) Localization of desmin (left panels), DHPR1α (middle panels), and Ryr1 (right panels) proteins in tibialis anterior cross sections from Mtm1-KO and WT mice 2 weeks after intramuscular delivery of either rAAV9-Mtmr2, rAAV9-Mtm1, or saline. Scale bar = 20 µm. (B) Transcript levels of AChR subunits (Chrn-α1, Chrn-δ, and Chrn-γ) in untreated and AAV-treated TA muscle (WT-PBS n = 7; KO-PBS n = 8; KO-Mtmr2 n = 8; KO-Mtm1 n = 3). (C) The number of Pax7-positive cells per myofiber was quantified in TA of wild-type and Mtm1-KO mice 2 weeks after PBS, rAAV9-Mtmr2 or rAAV9-Mtm1 administration (WT-PBS n = 10; KO-PBS n = 9; KO-Mtmr2 n = 9; KO-Mtm1 n = 5). *, **, ***, p < 0.05, 0.01 or 0.001 versus WT-PBS; Δ, ΔΔ, ΔΔΔ, p < 0.05, 0.01 or 0.001 versus KO-PBS; ♦, ♦♦, ♦♦♦, p < 0.05, 0.01 or 0.001 versus KO-Mtm1. FIGURE 3. View largeDownload slide Restoration of XLMTM muscle biomarkers by Mtmr2 gene transfer. (A) Localization of desmin (left panels), DHPR1α (middle panels), and Ryr1 (right panels) proteins in tibialis anterior cross sections from Mtm1-KO and WT mice 2 weeks after intramuscular delivery of either rAAV9-Mtmr2, rAAV9-Mtm1, or saline. Scale bar = 20 µm. (B) Transcript levels of AChR subunits (Chrn-α1, Chrn-δ, and Chrn-γ) in untreated and AAV-treated TA muscle (WT-PBS n = 7; KO-PBS n = 8; KO-Mtmr2 n = 8; KO-Mtm1 n = 3). (C) The number of Pax7-positive cells per myofiber was quantified in TA of wild-type and Mtm1-KO mice 2 weeks after PBS, rAAV9-Mtmr2 or rAAV9-Mtm1 administration (WT-PBS n = 10; KO-PBS n = 9; KO-Mtmr2 n = 9; KO-Mtm1 n = 5). *, **, ***, p < 0.05, 0.01 or 0.001 versus WT-PBS; Δ, ΔΔ, ΔΔΔ, p < 0.05, 0.01 or 0.001 versus KO-PBS; ♦, ♦♦, ♦♦♦, p < 0.05, 0.01 or 0.001 versus KO-Mtm1. In murine models of myotubular myopathy, the muscle pathology has also been linked to defects in the morphology and function of the neuromuscular junction (16) and in satellite cells (14), leading to alterations in the expression of several biomarkers such as various subunits of the acetylcholine receptor (AChR) and Pax7. In our experiments, we found that MTMR2 overexpression did not correct the increase in AChR subunit α1 (Chrn-α1), γ (Chrn-γ), and δ (Chrn-δ) mRNA levels observed in the contralateral KO-PBS muscle, while myotubularin did (Fig. 3B). With regards to the reported deficiency in muscle progenitor cells, we quantified the number of Pax7-positive cells in AAV-treated and untreated TA muscles, and found a significant partial restoration of this biomarker after expression of either Mtmr2 or Mtm1 transgene (2- and 6.7-fold increase of satellite cells in MTMR2 and myotubularin-treated muscles, p < 0.01 and p < 0.001, respectively; Fig. 3C). Finally, we quantified the number of triads, transverse (T)-tubules, and longitudinal (L)-tubules in muscle preparations (Supplementary Data Fig. S3). Among these parameters, we found that the amount of L-tubules in rAAV9-Mtmr2-treated TA, which is significantly increased in KO-PBS animals as previously described (9), was decreased. The findings suggest a partial improvement of the sarcotubular architecture, but should be interpreted cautiously because the T-tubule and triad numbers did not significantly improve. Altogether, these results show that MTMR2 overexpression in myotubularin-deficient skeletal muscle ameliorates pathological and molecular signs of the disease, even though less efficiently than MTM1 itself. Mtmr2 Gene Transfer Increases Muscle Strength in Mtm1-Deficient Muscles XLMTM mice display progressive muscle weakness starting in the hind limbs at early phases of the disease (7, 9). Indeed, the total and specific tetanic forces of TA muscle dropped significantly in 4-week-old Mtm1-KO mice by 90% and 80%, respectively, versus the WT animals (p < 0.001, Fig. 4). We quantified the effect of transgene expression in the strength of TA muscle 2 weeks after vector injection. MTMR2 overexpression led to a 2.9- and 2.5-fold increase in the total and specific forces, respectively, with respect to values in KO-PBS muscles (Fig. 4). Albeit partial compared to myotubularin replacement, the effect of MTMR2 on the specific force clearly showed an improvement of the contractile quality of Mtm1-KO myofibers. FIGURE 4. View largeDownload slide Intramuscular delivery of rAAV9-Mtmr2 increases muscle strength in Mtm1-KO mice. Total tetanic (left graph) and specific (right graph) force of TA muscle from KO mice after intramuscular delivery rAAV9-Mtmr2 and -Mtm1 vectors, and appropriate controls (WT-PBS n = 16; KO-PBS n = 10; KO-Mtmr2 n = 10; KO-Mtm1 n = 7). ***, p < 0.001 versus WT-PBS; ΔΔΔ, p < 0.001 versus KO-PBS; ♦♦♦, p < 0.001 versus KO-Mtm1. FIGURE 4. View largeDownload slide Intramuscular delivery of rAAV9-Mtmr2 increases muscle strength in Mtm1-KO mice. Total tetanic (left graph) and specific (right graph) force of TA muscle from KO mice after intramuscular delivery rAAV9-Mtmr2 and -Mtm1 vectors, and appropriate controls (WT-PBS n = 16; KO-PBS n = 10; KO-Mtmr2 n = 10; KO-Mtm1 n = 7). ***, p < 0.001 versus WT-PBS; ΔΔΔ, p < 0.001 versus KO-PBS; ♦♦♦, p < 0.001 versus KO-Mtm1. Systemic Delivery of rAAV9-Mtmr2 Prolongs Survival and Improves Muscle Function of XLMTM Mice Myotubularin-deficient mice survive on average less than 2 months (median survival 51 days; Fig. 5A). We therefore assessed the efficacy of rAAV9-Mtmr2 treatment at the whole body level of mutant mice over a 3-month observation period. We first tested an intravenous dose of 8 × 1013 vg/kg, which is close to the efficacious dose of rAAV8-Mtm1 that was administrated in our previous gene replacement study in XLMTM mice (15), but did not observe a therapeutic benefit (not shown). We therefore increased the dose of rAAV9-Mtmr2 and injected 2.4 × 1014 vg/kg of vector into the tail-vein in Mtm1-KO mice at 2 weeks of age. All mutant mice treated at this higher dose remained viable and gained body mass during the study period (Fig. 5A, B and Supplementary Data Video). At the time of death, transgene expression was analyzed at the mRNA and protein level in a panel of representative muscles throughout the body (Fig. 5C, D). Mtmr2 mRNA expression was 50–100 times higher than the endogenous level in skeletal muscles (Fig. 5C), reaching a 250-fold increase in heart (Supplementary Data Fig. S4), and resulted in high levels of MTMR2 protein in transduced muscles (Fig. 5D). We also analyzed PI3P levels in skeletal muscle and, similar to the effect observed after local delivery of rAAV9-Mtmr2 in mutant muscle (Fig. 1D), systemic administration of this vector reduced PI3P levels in Mtm1-deficient muscle 7 weeks after vector injection (Fig. 5E). FIGURE 5. View largeDownload slide Intravenous administration of rAAV9-Mtmr2 results in muscular MTMR2 overexpression and prolongs the lifespan of myotubularin-deficient mice. (A, B) Effect of intravenous rAAV9-Mtmr2 administration in Mtm1-KO mice over a 3-month observation period. Lifespan (A) and body mass (B) of untreated and Mtmr2-treated mutant mice, compared to WT animals (WT-PBS n = 9; KO-PBS n = 10; KO-Mtmr2 n = 6). (C) RT-qPCR analysis shows that Mtmr2 mRNA transcripts are 50–100 times higher than endogenous levels in various skeletal muscles (WT-PBS n = 6; KO-Mtmr2 n = 4). (D) Representative Western blot images of MTMR2 in skeletal muscles (TA: tibialis anterior, GA: gastrocnemius, QUA: quadriceps, TRI: triceps, BI: biceps brachii) (◀ in green) and α-tubulin (in red) 3 months after vector injection. MTMR2 is overexpressed in muscles scattered throughout the body. (E) PI3P level in soleus muscle of Mtm1-KO mice 7 weeks after rAAV9 vector injection compared to WT littermates, and untreated mutants at 6 weeks of age. **, p < 0.05 versus WT-PBS. FIGURE 5. View largeDownload slide Intravenous administration of rAAV9-Mtmr2 results in muscular MTMR2 overexpression and prolongs the lifespan of myotubularin-deficient mice. (A, B) Effect of intravenous rAAV9-Mtmr2 administration in Mtm1-KO mice over a 3-month observation period. Lifespan (A) and body mass (B) of untreated and Mtmr2-treated mutant mice, compared to WT animals (WT-PBS n = 9; KO-PBS n = 10; KO-Mtmr2 n = 6). (C) RT-qPCR analysis shows that Mtmr2 mRNA transcripts are 50–100 times higher than endogenous levels in various skeletal muscles (WT-PBS n = 6; KO-Mtmr2 n = 4). (D) Representative Western blot images of MTMR2 in skeletal muscles (TA: tibialis anterior, GA: gastrocnemius, QUA: quadriceps, TRI: triceps, BI: biceps brachii) (◀ in green) and α-tubulin (in red) 3 months after vector injection. MTMR2 is overexpressed in muscles scattered throughout the body. (E) PI3P level in soleus muscle of Mtm1-KO mice 7 weeks after rAAV9 vector injection compared to WT littermates, and untreated mutants at 6 weeks of age. **, p < 0.05 versus WT-PBS. All analyzed skeletal muscles from vector-treated mice gained weight compared to untreated mutant mice at 5 weeks of age, consistent with the overall increase of body weight (Fig. 6A). This amelioration correlated with an increase in myofiber size (mean diameter was 15.9 µm in TA and 17 µm in biceps brachii of 5-week-old Mtm1-KO animals vs 31.9 µm and 27.4 µm, respectively, 3 months after vector injection), and a quasi-normalization in mitochondrial and nuclear distribution within myofibers (Fig. 6B–D). In the hearts of rAAV9-Mtmr2-treated KO mice, we found small focal patches of fibrosis generally associated with cellular infiltrates, lesions that were clinically well-tolerated (Supplementary Data Fig. S4). FIGURE 6. View largeDownload slide Whole-body Mtmr2 gene transfer corrects muscle hypotrophy and other histopathological hallmarks of the disease. (A) Weight of various skeletal muscles of untreated (KO-PBS) and Mtmr2-treated (KO-Mtmr2) mutant mice 3 months after vector administration (KO-PBS n = 4; KO-Mtmr2 n = 6; WT-PBS n = 9) (TA: tibialis anterior, EDL: extensor digitorum longus, SOL: soleus, GA: gastrocnemius, QUA: quadriceps, TRI: triceps, BI: biceps brachii). Weight is expressed as a percentage of age-matched WT-PBS animals (◀). (B) Morphometric analysis of the mean myofiber diameter in tibialis anterior (upper graph) and biceps brachii (lower graph) muscles of untreated Mtm1-KO mice at 5 weeks of age, and in Mtmr2-treated mice 3 months after treatment, compared to normal littermates (WT-PBS n = 6 [5 weeks of age] and n = 5 [3 months of age]; KO-PBS n = 4; KO-Mtmr2 n = 4). (C) H&E and NADH-TR stainings of tibialis anterior and biceps muscle cross sections from Mtm1-KO mice 3 months after Mtmr2-treatment (scale bar = 50 µm) (D) and quantification of the percentage of myofibers with internal nuclei in TA and BI of myotubularin-deficient mice 3 months after rAAV9-Mtmr2 injection (WT-PBS n = 5; KO-Mtmr2 n = 4). (E) Electron microscopy photographs of TA muscles from WT-PBS, KO-PBS, and KO-Mtmr2 show the position of triads (white arrows) along sarcomeres. The inset is from the same sample but a different fiber. The 2 black arrows point to L-tubules, scale bar = 500 nm (250 nm within the inset). Quantification of the number of triads, T-tubules and L-tubules in TA shows an improvement of the structure 10 weeks after AAV-Mtmr2 injection. *, **, p < 0.05 or 0.01 versus 5-week-old WT-PBS; •, ••, •••, p < 0.05, 0.01 or 0.001 versus 3-month-old WT-PBS; Δ, ΔΔ, p < 0.05 or 0.01 versus 5-week-old KO-PBS. FIGURE 6. View largeDownload slide Whole-body Mtmr2 gene transfer corrects muscle hypotrophy and other histopathological hallmarks of the disease. (A) Weight of various skeletal muscles of untreated (KO-PBS) and Mtmr2-treated (KO-Mtmr2) mutant mice 3 months after vector administration (KO-PBS n = 4; KO-Mtmr2 n = 6; WT-PBS n = 9) (TA: tibialis anterior, EDL: extensor digitorum longus, SOL: soleus, GA: gastrocnemius, QUA: quadriceps, TRI: triceps, BI: biceps brachii). Weight is expressed as a percentage of age-matched WT-PBS animals (◀). (B) Morphometric analysis of the mean myofiber diameter in tibialis anterior (upper graph) and biceps brachii (lower graph) muscles of untreated Mtm1-KO mice at 5 weeks of age, and in Mtmr2-treated mice 3 months after treatment, compared to normal littermates (WT-PBS n = 6 [5 weeks of age] and n = 5 [3 months of age]; KO-PBS n = 4; KO-Mtmr2 n = 4). (C) H&E and NADH-TR stainings of tibialis anterior and biceps muscle cross sections from Mtm1-KO mice 3 months after Mtmr2-treatment (scale bar = 50 µm) (D) and quantification of the percentage of myofibers with internal nuclei in TA and BI of myotubularin-deficient mice 3 months after rAAV9-Mtmr2 injection (WT-PBS n = 5; KO-Mtmr2 n = 4). (E) Electron microscopy photographs of TA muscles from WT-PBS, KO-PBS, and KO-Mtmr2 show the position of triads (white arrows) along sarcomeres. The inset is from the same sample but a different fiber. The 2 black arrows point to L-tubules, scale bar = 500 nm (250 nm within the inset). Quantification of the number of triads, T-tubules and L-tubules in TA shows an improvement of the structure 10 weeks after AAV-Mtmr2 injection. *, **, p < 0.05 or 0.01 versus 5-week-old WT-PBS; •, ••, •••, p < 0.05, 0.01 or 0.001 versus 3-month-old WT-PBS; Δ, ΔΔ, p < 0.05 or 0.01 versus 5-week-old KO-PBS. To assess whether Mtmr2 gene therapy ameliorates the excitation-contraction coupling machinery in Mtm1-KO animals, the number of triads and the morphology of tubules (transverse -T- or longitudinal -L-) were quantified in TA samples 7 weeks after vector injection. Our results show that the abnormal reduction of triads and presence of L-tubules that are observed in myotubularin-deficient muscle were both largely corrected by systemic administration of the rAAV9-Mtmr2 vector (7 triads per ROI with 5% of L-tubules in KO-PBS TA vs 11.1 triads per ROI with 1% of L-tubules in KO-Mtmr2 muscle), consistent with a partial restoration of the triad architecture in skeletal muscles upon treatment (Fig. 6E). Finally, to measure the effect of gene therapy on global muscle function, the open field actimeter and escape tests were used. Mutant mice were very weak (whole body tension, 0.07 ± 0.01 vs 0.15 ± 0.005 N/g in healthy animals) and covered less than half the distance explored by WT littermates at 5 weeks of age (Fig. 7A, B). Importantly, mice treated with rAAV9-Mtmr2 had a significant improvement in muscle function with values close to normal at 3 months postinjection (86% recovery of whole body strength, 0.16 ± 0.02 N/g). Accordingly, the specific tetanic force of isolated EDL (Fig. 7C) and soleus (data not shown), muscles was almost normalized 3 months after treatment. Altogether, our results demonstrate that the reduced lifespan and muscle impairment associated with myotubularin deficiency in mice can be rescued by systemic administration of a vector expressing the homologous MTMR2 protein. FIGURE 7. View largeDownload slide Systemic rAAV9-Mtmr2 improves mobility, strength and myofiber contractility in XLMTM mice. (A) Whole body spontaneous mobility of normal (WT-PBS), mutant (KO-PBS), and AAV-treated mutant (KO-Mtmr2) mice 3 months after saline or vector injection. The distance covered over the 90-minute test was assessed using an open field actimeter. (B) Global strength developed by mice during the escape test. (C) Specific force of isolated EDL muscles from 5-week-old untreated KO and WT mice, and 3 months after vector delivery in mutant animals compared to control littermates. WT-PBS, n = 6 at 5 weeks of age and n = 11 at 3 months of age, KO-PBS, n = 4 at 5 weeks of age and KO-Mtmr2, n = 6 at 3 months of age. *, **, p < 0.05 or 0.01 versus 5-week-old WT-PBS; Δ, ΔΔ, p < 0.05 or 0.01 versus 5-week-old KO-PBS. FIGURE 7. View largeDownload slide Systemic rAAV9-Mtmr2 improves mobility, strength and myofiber contractility in XLMTM mice. (A) Whole body spontaneous mobility of normal (WT-PBS), mutant (KO-PBS), and AAV-treated mutant (KO-Mtmr2) mice 3 months after saline or vector injection. The distance covered over the 90-minute test was assessed using an open field actimeter. (B) Global strength developed by mice during the escape test. (C) Specific force of isolated EDL muscles from 5-week-old untreated KO and WT mice, and 3 months after vector delivery in mutant animals compared to control littermates. WT-PBS, n = 6 at 5 weeks of age and n = 11 at 3 months of age, KO-PBS, n = 4 at 5 weeks of age and KO-Mtmr2, n = 6 at 3 months of age. *, **, p < 0.05 or 0.01 versus 5-week-old WT-PBS; Δ, ΔΔ, p < 0.05 or 0.01 versus 5-week-old KO-PBS. DISCUSSION In the present study, we investigated the potency of both MTMR1 and MTMR2 to compensate myotubularin deficiency in skeletal muscle and demonstrate that only rAAV9-mediated expression of MTMR2 led to major improvements in the structure and mechanical properties of muscle fibers from Mtm1-KO mice. A short-term local treatment rescued muscle contractility and many of the histopathological hallmarks of the disease, and systemic administration of the vector prolonged the lifespan and ameliorated the motor activity of XLMTM mice. Among the 14 myotubularin-related proteins, MTMR1 and MTMR2 were chosen for the study because they have the closest homology to MTM1. The 3 proteins belong to the same phylogenetic subgroup (24), have similar primary structures (65% and 59% of amino acid identity for MTMR2 and MTMR1 compared to MTM1, respectively, Supplementary Data Fig. S5), bear the same functional domains (22), and exert a lipid phosphatase activity towards PI3P and PI(3, 5)P2 (6, 26, 28–30). Moreover, MTMR2 prevents muscle dysfunction in drosophila (38) and zebrafish (8) models of myotubularin deficiency, and MTMR1 has a similar efficacy in zebrafish embryos (8). Surprisingly, our study shows that under similar conditions only MTMR2 overexpression was capable of rescuing the muscle phenotype of a murine model of XLMTM. The poor therapeutic efficacy of MTMR1 could be the consequence of a lower enzymatic activity in muscle because MTMR1 PI3P-specific activity was reported to be 30 times lower than that of MTM1 and MTMR2 (25). Accordingly, the increased levels of PI3P in skeletal muscles of Mtm1-KO mice were normalized upon AAV-mediated expression of myotubularin and MTMR2, but not MTMR1. Therefore, this effect could be directly linked to differences in the enzymatic activity of these proteins and/or access to PI3P subpools in vivo, but it could also result from distinct protein-protein interactions. For example, MTMR12 (also known as 3-PAP), an MTM1 protein interactor that regulates its stability and localization in muscle cells, is able to interact with MTMR2 but not MTMR1 (39, 40). Interestingly, we found that MTMR2 and myotubularin rescued common pathological features in skeletal muscle, suggesting that these 2 proteins can regulate similar cellular pathways. MTMR2 overexpression in Mtm1-deficient muscle doubled the population of satellite cells, which indicates a function similar to myotubularin on the turnover and/or survival of muscle stem cells (14), and corrected desmin and mitochondria subcellular localizations, suggesting also an overlapping role in the maintenance of the intermediate filament and mitochondrial networks (12). MTMR2-driven repositioning of mitochondria in myofibers could also account for muscle force recovery, as the alteration of mitochondrial localization/function was suggested to promote muscle weakness in Mtm1-KO muscle (12). The reason why MTMR2 does not compensate naturally the XLMTM phenotype in vivo is unclear but it may be related to its low level of expression in skeletal muscle. The specific force of Mtmr2 transduced myofibers improved independently from major ultrastructural changes in triads 2 weeks after vector injection (only the number of L-tubules was reduced), suggesting that the defective excitation-contraction coupling associated with myotubularin deficiency may not be entirely due to morphological alterations in triads. Functional defects in calcium release from the sarcoplasmic reticulum linked to a depression in RYR1 receptor activity may also contribute to muscle weakness in the disease, as these channels are known to be regulated by phosphoinositides (9, 36, 41). Our results show that PI3P levels were rapidly restored upon Mtmr2 and Mtm1 gene delivery, which suggests that the treatment had an effect at the functional level in the excitation-contraction coupling machinery. The present study demonstrates that functional compensation can be achieved by overexpression of an MTM1 homologous protein in a murine model of myotubular myopathy. Similar compensatory studies were performed for other muscular diseases in the past, such as Duchenne muscular dystrophy (42, 43), and limb-girdle muscular dystrophy type 2C (44), for which utrophin and epsilon-sarcoglycan overexpression were shown to ameliorate dystrophin and α-sarcoglycan deficiencies, respectively, in animal models. An extensive search for molecules that upregulate utrophin expression resulted in the identification of several pharmacological agents and a phase 1 clinical trial in pediatric Duchenne muscular dystrophy patients (45, 46). Likewise, upregulation of MTMR2 represents a promising new therapeutic target for myotubular myopathy. This could be achieved by the use of small chemical compounds or RNAi-mediated knock-down of genes leading to increased MTMR2 transcript and/or protein levels, or by direct protein replacement. Given our results, MTMR2 overexpression in skeletal muscles by rAAV-mediated gene delivery could also be a translationally relevant approach for XLMTM. The feasibility of gene replacement therapy for myotubular myopathy was previously reported in mouse and dog models of the disease by administration of rAAV vectors expressing MTM1 (17, 19). Intravenous injection of a rAAV8 vector led to generalized muscle transduction and long-term full phenotypic correction in these mutant animals, supporting a ongoing clinical trial in XLMTM patients. In the present study, we assessed whole body treatment of Mtm1-KO mice by administrating a rAAV9 vector that expresses MTMR2 under the muscle-specific desmin promoter. All treated mice responded well to the treatment and survived until the end of the 3-month study period, with an amelioration in skeletal muscle histology and strength. The potency of exogenous MTMR2 appeared, however, lower than MTM1, which is likely due to functional differences between these proteins. Accordingly, while preparing this manuscript, a study (47) reported the expression in skeletal muscle of a shorter MTMR2 isoform lacking the N-terminal part of the protein (571 amino acids) that displays an MTM1-like activity in yeast cells, in addition to the initially described longer isoform (643 amino acids) that was used in our study (48). They compared the effect of intramuscular administration of rAAV1 vectors expressing the 2 MTMR2 variants in the TA muscle of Mtm1 mutant mice, and showed that the short MTMR2 isoform (571 amino acids) had a better rescuing effect than the longer variant, although it remains unclear whether similar expression levels of the 2 MTMR2 proteins were achieved after local rAAV transduction. Therefore, it will be interesting to compare the effect of these MTMR2 isoforms by intravenous delivery, and explore whether a shorter MTMR1 protein variant(s) exists in skeletal muscle and could also display an increased potency in myotubularin-deficient muscle upon overexpression. In conclusion, our study demonstrates that systemic administration of a rAAV vector expressing Mtmr2 can ameliorate the severe generalized muscle disease in Mtm1 mutant mice, and identifies MTMR2 as a surrogate to re-establish muscle function, paving the way for studies modulating its expression as a treatment approach for myotubular myopathy. ACKNOWLEDGMENTS We are grateful to the platforms of Genethon for their excellent technical expertise and contribution to this work, Laurine Buscara and Karine Poulard for mouse colony management and genotyping, Fanny Collaud for help in vector production, and Jérémie Cosette for assistance in imaging analysis. We also thank Nadia Messaddeq for help with tissues for electron microscopy and Jean-Louis Mandel for support in the initial phases of this project. REFERENCES 1 Romero NB. Centronuclear myopathies: A widening concept. Neuromuscul Disord 2010; 20: 223– 8 http://dx.doi.org/10.1016/j.nmd.2010.01.014 Google Scholar CrossRef Search ADS PubMed 2 McEntagart M, Parsons G, Buj-Bello Aet al. , Genotype-phenotype correlations in X-linked myotubular myopathy. Neuromuscul Disord 2002; 12: 939– 46 http://dx.doi.org/10.1016/S0960-8966(02)00153-0 Google Scholar CrossRef Search ADS PubMed 3 Laporte J, Hu LJ, Kretz Cet al. , A gene mutated in X-linked myotubular myopathy defines a new putative tyrosine phosphatase family conserved in yeast. Nat Genet 1996; 13: 175– 82 http://dx.doi.org/10.1038/ng0696-175 Google Scholar CrossRef Search ADS PubMed 4 Blondeau F, Laporte J, Bodin Set al. , Myotubularin, a phosphatase deficient in myotubular myopathy, acts on phosphatidylinositol 3-kinase and phosphatidylinositol 3-phosphate pathway. Hum Mol Genet 2000; 9: 2223– 9 http://dx.doi.org/10.1093/oxfordjournals.hmg.a018913 Google Scholar CrossRef Search ADS PubMed 5 Taylor GS, Maehama T, Dixon JE. Inaugural article: Myotubularin, a protein tyrosine phosphatase mutated in myotubular myopathy, dephosphorylates the lipid second messenger, phosphatidylinositol 3-phosphate. Proc Natl Acad Sci U S A 2000; 97: 8910– 5 http://dx.doi.org/10.1073/pnas.160255697 Google Scholar CrossRef Search ADS PubMed 6 Tronchere H, Laporte J, Pendaries Cet al. , Production of phosphatidylinositol 5-phosphate by the phosphoinositide 3-phosphatase myotubularin in mammalian cells. J Biol Chem 2004; 279: 7304– 12 http://dx.doi.org/10.1074/jbc.M311071200 Google Scholar CrossRef Search ADS PubMed 7 Buj-Bello A, Laugel V, Messaddeq Net al. , The lipid phosphatase myotubularin is essential for skeletal muscle maintenance but not for myogenesis in mice. Proc Natl Acad Sci U S A 2002; 99: 15060– 5 http://dx.doi.org/10.1073/pnas.212498399 Google Scholar CrossRef Search ADS PubMed 8 Dowling JJ, Vreede AP, Low SEet al. , Loss of myotubularin function results in T-tubule disorganization in zebrafish and human myotubular myopathy. PLoS Genet 2009; 5: e1000372 Google Scholar CrossRef Search ADS PubMed 9 Al-Qusairi L, Weiss N, Toussaint Aet al. , T-tubule disorganization and defective excitation-contraction coupling in muscle fibers lacking myotubularin lipid phosphatase. Proc Natl Acad Sci U S A 2009; 106: 18763– 8 http://dx.doi.org/10.1073/pnas.0900705106 Google Scholar CrossRef Search ADS PubMed 10 Beggs AH, Bohm J, Snead Eet al. , MTM1 mutation associated with X-linked myotubular myopathy in Labrador Retrievers. Proc Natl Acad Sci U S A 2010; 107: 14697– 702 http://dx.doi.org/10.1073/pnas.1003677107 Google Scholar CrossRef Search ADS PubMed 11 Bevilacqua JA, Bitoun M, Biancalana Vet al. , “Necklace” fibers, a new histological marker of late-onset MTM1-related centronuclear myopathy. Acta Neuropathol 2009; 117: 283– 91 Google Scholar CrossRef Search ADS PubMed 12 Hnia K, Tronchere H, Tomczak KKet al. , Myotubularin controls desmin intermediate filament architecture and mitochondrial dynamics in human and mouse skeletal muscle. J Clin Invest 2011; 121: 70– 85 http://dx.doi.org/10.1172/JCI44021 Google Scholar CrossRef Search ADS PubMed 13 Toussaint A, Cowling BS, Hnia Ket al. , Defects in amphiphysin 2 (BIN1) and triads in several forms of centronuclear myopathies. Acta Neuropathol 2011; 121: 253– 66 http://dx.doi.org/10.1007/s00401-010-0754-2 Google Scholar CrossRef Search ADS PubMed 14 Lawlor MW, Alexander MS, Viola MGet al. , Myotubularin-deficient myoblasts display increased apoptosis, delayed proliferation, and poor cell engraftment. Am J Pathol 2012; 181: 961– 8 http://dx.doi.org/10.1016/j.ajpath.2012.05.016 Google Scholar CrossRef Search ADS PubMed 15 Robb SA, Sewry CA, Dowling JJet al. , Impaired neuromuscular transmission and response to acetylcholinesterase inhibitors in centronuclear myopathies. Neuromuscul Disord 2011; 21: 379– 86 http://dx.doi.org/10.1016/j.nmd.2011.02.012 Google Scholar CrossRef Search ADS PubMed 16 Dowling JJ, Joubert R, Low SEet al. , Myotubular myopathy and the neuromuscular junction: A novel therapeutic approach from mouse models. Dis Model Mech 2012; 5: 852– 9 http://dx.doi.org/10.1242/dmm.009746 Google Scholar CrossRef Search ADS PubMed 17 Buj-Bello A, Fougerousse F, Schwab Yet al. , AAV-mediated intramuscular delivery of myotubularin corrects the myotubular myopathy phenotype in targeted murine muscle and suggests a function in plasma membrane homeostasis. Hum Mol Genet 2008; 17: 2132– 43 http://dx.doi.org/10.1093/hmg/ddn112 Google Scholar CrossRef Search ADS PubMed 18 Lawlor MW, Armstrong D, Viola MGet al. , Enzyme replacement therapy rescues weakness and improves muscle pathology in mice with X-linked myotubular myopathy. Hum Mol Genet 2013; 22: 1525– 38 http://dx.doi.org/10.1093/hmg/ddt003 Google Scholar CrossRef Search ADS PubMed 19 Childers MK, Joubert R, Poulard Ket al. , Gene therapy prolongs survival and restores function in murine and canine models of myotubular myopathy. Sci Transl Med 2014; 6: 220ra210 Google Scholar CrossRef Search ADS 20 Cowling BS, Chevremont T, Prokic Iet al. , Reducing dynamin 2 expression rescues X-linked centronuclear myopathy. J Clin Invest 2014; 124: 1350– 63 http://dx.doi.org/10.1172/JCI71206 Google Scholar CrossRef Search ADS PubMed 21 Mack DL, Poulard K, Goddard MAet al. , Systemic AAV8-mediated gene therapy drives whole-body correction of myotubular myopathy in dogs. Mol Ther 2017; 25: 839– 54 http://dx.doi.org/10.1016/j.ymthe.2017.02.004 Google Scholar CrossRef Search ADS PubMed 22 Hnia K, Vaccari I, Bolino A, Laporte J. Myotubularin phosphoinositide phosphatases: Cellular functions and disease pathophysiology. Trends Mol Med 2012; 18: 317– 27 http://dx.doi.org/10.1016/j.molmed.2012.04.004 Google Scholar CrossRef Search ADS PubMed 23 Laporte J, Liaubet L, Blondeau Fet al. , Functional redundancy in the myotubularin family. Biochem Biophys Res Commun 2002; 291: 305– 12 http://dx.doi.org/10.1006/bbrc.2002.6445 Google Scholar CrossRef Search ADS PubMed 24 Laporte J, Bedez F, Bolino Aet al. , Myotubularins, a large disease-associated family of cooperating catalytically active and inactive phosphoinositides phosphatases. Hum Mol Genet 2003; 12 (Spec No. 2): R285– 92 Google Scholar CrossRef Search ADS PubMed 25 Kim SA, Taylor GS, Torgersen KMet al. , Myotubularin and MTMR2, phosphatidylinositol 3-phosphatases mutated in myotubular myopathy and type 4B Charcot-Marie-Tooth disease. J Biol Chem 2002; 277: 4526– 31 http://dx.doi.org/10.1074/jbc.M111087200 Google Scholar CrossRef Search ADS PubMed 26 Berger P, Bonneick S, Willi Set al. , Loss of phosphatase activity in myotubularin-related protein 2 is associated with Charcot-Marie-Tooth disease type 4B1. Hum Mol Genet 2002; 11: 1569– 79 http://dx.doi.org/10.1093/hmg/11.13.1569 Google Scholar CrossRef Search ADS PubMed 27 Buj-Bello A, Furling D, Tronchere Het al. , Muscle-specific alternative splicing of myotubularin-related 1 gene is impaired in DM1 muscle cells. Hum Mol Genet 2002; 11: 2297– 307 http://dx.doi.org/10.1093/hmg/11.19.2297 Google Scholar CrossRef Search ADS PubMed 28 Schaletzky J, Dove SK, Short Bet al. , Phosphatidylinositol-5-phosphate activation and conserved substrate specificity of the myotubularin phosphatidylinositol 3-phosphatases. Curr Biol 2003; 13: 504– 9 http://dx.doi.org/10.1016/S0960-9822(03)00132-5 Google Scholar CrossRef Search ADS PubMed 29 Rohde HM, Tronchere H, Payrastre Bet al. , Detection of myotubularin phosphatases activity on phosphoinositides in vitro and ex vivo. Methods Mol Biol 2009; 462: 265– 78 Google Scholar PubMed 30 Tsujita K, Itoh T, Ijuin Tet al. , Myotubularin regulates the function of the late endosome through the gram domain-phosphatidylinositol 3, 5-bisphosphate interaction. J Biol Chem 2004; 279: 13817– 24 http://dx.doi.org/10.1074/jbc.M312294200 Google Scholar CrossRef Search ADS PubMed 31 Rohr UP, Wulf MA, Stahn Set al. , Fast and reliable titration of recombinant adeno-associated virus type-2 using quantitative real-time PCR. J Virol Methods 2002; 106: 81– 8 http://dx.doi.org/10.1016/S0166-0934(02)00138-6 Google Scholar CrossRef Search ADS PubMed 32 Carlson CG, Makiejus RV. A noninvasive procedure to detect muscle weakness in the mdx mouse. Muscle Nerve 1990; 13: 480– 4 http://dx.doi.org/10.1002/mus.880130603 Google Scholar CrossRef Search ADS PubMed 33 Fougerousse F, Gonin P, Durand Met al. , Force impairment in calpain 3-deficient mice is not correlated with mechanical disruption. Muscle Nerve 2003; 27: 616– 23 http://dx.doi.org/10.1002/mus.10368 Google Scholar CrossRef Search ADS PubMed 34 Vignaud A, Cebrian J, Martelly Iet al. , Effect of anti-inflammatory and antioxidant drugs on the long-term repair of severely injured mouse skeletal muscle. Exp Physiol 2005; 90: 487– 95 http://dx.doi.org/10.1113/expphysiol.2005.029835 Google Scholar CrossRef Search ADS PubMed 35 Sabha N, Volpatti JR, Gonorazky Het al. , PIK3C2B inhibition improves function and prolongs survival in myotubular myopathy animal models. J Clin Invest 2016; 126: 3613– 25 http://dx.doi.org/10.1172/JCI86841 Google Scholar CrossRef Search ADS PubMed 36 Kutchukian C, Lo Scrudato M, Tourneur Yet al. , Phosphatidylinositol 3-kinase inhibition restores Ca2+ release defects and prolongs survival in myotubularin-deficient mice. Proc Natl Acad Sci U S A 2016; 113: 14432– 7 Google Scholar CrossRef Search ADS PubMed 37 Lawlor MW, Beggs AH, Buj-Bello Aet al. , Skeletal muscle pathology in X-linked myotubular myopathy: Review with cross-species comparisons. J Neuropathol Exp Neurol 2016; 75: 102– 10 http://dx.doi.org/10.1093/jnen/nlv020 Google Scholar CrossRef Search ADS PubMed 38 Ribeiro I, Yuan L, Tanentzapf Get al. , Phosphoinositide regulation of integrin trafficking required for muscle attachment and maintenance. PLoS Genet 2011; 7: e1001295 Google Scholar CrossRef Search ADS PubMed 39 Nandurkar HH, Layton M, Laporte Jet al. , Identification of myotubularin as the lipid phosphatase catalytic subunit associated with the 3-phosphatase adapter protein, 3-PAP. Proc Natl Acad Sci U S A 2003; 100: 8660– 5 http://dx.doi.org/10.1073/pnas.1033097100 Google Scholar CrossRef Search ADS PubMed 40 Gupta VA, Hnia K, Smith LLet al. , Loss of catalytically inactive lipid phosphatase myotubularin-related protein 12 impairs myotubularin stability and promotes centronuclear myopathy in zebrafish. PLoS Genet 2013; 9: e1003583 Google Scholar CrossRef Search ADS PubMed 41 Rodriguez EG, Lefebvre R, Bodnar Det al. , Phosphoinositide substrates of myotubularin affect voltage-activated Ca(2)(+) release in skeletal muscle. Pflugers Arch 2014; 466: 973– 85 Google Scholar CrossRef Search ADS PubMed 42 Tinsley JM, Potter AC, Phelps SRet al. , Amelioration of the dystrophic phenotype of mdx mice using a truncated utrophin transgene. Nature 1996; 384: 349– 53 http://dx.doi.org/10.1038/384349a0 Google Scholar CrossRef Search ADS PubMed 43 Tinsley J, Deconinck N, Fisher Ret al. , Expression of full-length utrophin prevents muscular dystrophy in mdx mice. Nat Med 1998; 4: 1441– 4 http://dx.doi.org/10.1038/4033 Google Scholar CrossRef Search ADS PubMed 44 Imamura M, Mochizuki Y, Engvall Eet al. , Epsilon-sarcoglycan compensates for lack of alpha-sarcoglycan in a mouse model of limb-girdle muscular dystrophy. Hum Mol Genet 2005; 14: 775– 83 http://dx.doi.org/10.1093/hmg/ddi072 Google Scholar CrossRef Search ADS PubMed 45 Guiraud S, Squire SE, Edwards Bet al. , Second-generation compound for the modulation of utrophin in the therapy of DMD. Hum Mol Genet 2015; 24: 4212– 24 http://dx.doi.org/10.1093/hmg/ddv154 Google Scholar CrossRef Search ADS PubMed 46 Ricotti V, Spinty S, Roper Het al. , Safety, tolerability, and pharmacokinetics of SMT C1100, a 2-arylbenzoxazole utrophin modulator, following single- and multiple-dose administration to pediatric patients with Duchenne muscular dystrophy. PLoS One 2016; 11: e0152840 Google Scholar CrossRef Search ADS PubMed 47 Raess MA, Bertazzi DL, Kretz Cet al. , Expression of the neuropathy-associated MTMR2 gene rescues MTM1-associated myopathy. Hum Mol Genet 2017; 26: 3736– 48 http://dx.doi.org/10.1093/hmg/ddx258 Google Scholar CrossRef Search ADS PubMed 48 Bolino A, Muglia M, Conforti FLet al. , Charcot-Marie-Tooth type 4B is caused by mutations in the gene encoding myotubularin-related protein-2. Nat Genet 2000; 25: 17– 9 http://dx.doi.org/10.1038/75542 Google Scholar CrossRef Search ADS PubMed © 2018 American Association of Neuropathologists, Inc. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
Journal of Neuropathology & Experimental Neurology – Oxford University Press
Published: Apr 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera