Targeted deletion of the muscular dystrophy gene myotilin does not perturb muscle structure or function in mice

Abstract

Myotilin, palladin, and myopalladin form a novel small subfamily of cytoskeletal proteins that contain immunoglobulin-like domains. Myotilin is a thin filament-associated protein localized at the Z-disk of skeletal and cardiac muscle cells. The direct binding to F-actin, efficient cross-linking of actin filaments, and prevention of induced disassembly of filaments are key roles of myotilin that are thought to be involved in structural maintenance and function of the sarcomere. Missense mutations in the myotilin-encoding gene cause dominant limb girdle muscular dystrophy type 1A and spheroid body myopathy and are the molecular defect that can cause myofibrillar myopathy. Here we describe the generation and analysis of mice that lack myotilin, myo(-/-) mice. Surprisingly, myo(-/-) mice maintain normal muscle sarcomeric and sarcolemmal integrity. Also, loss of myotilin does not cause alterations in the heart or other organs of newborn or adult myo(-/-) mice. The mice develop normally and have a normal life span, and their muscle capacity does not significantly differ from wild-type mice even after prolonged physical stress. The results suggest that either myotilin does not participate in muscle development and basal function maintenance or other proteins serve as structural and functional compensatory molecules when myotilin is absent.

FIG. 1.

Generation of the myotilin-deleted allele (myo^−/−). A. The top diagram shows the targeting vector constructed by subcloning genomic myotilin fragments into the pDELBOY-3X vector. The exons are depicted by filled boxes, and the loxP sites are shown as filled arrows. Relevant restriction site positions are marked by letters as follows: B, BamHI; E, EcoRI; Bs, BssI. The second diagram shows partial genomic organization of the wild-type murine myotilin gene. NF3 and NR represents the relative positions of the genotyping primers used to detect the wild-type allele. The third diagram shows the organization of the conditional (floxed) allele after homologous recombination in ES cells. The solid bars below the intronic regions represent the 5′ and 3′ probes used for detection of correct recombination. The bottom diagram shows the final organization of conditionally deleted myotilin gene, missing exon 3. The deletion was obtained in vivo by crossing the mouse bearing the flox allele with a Cre-expressing mouse. The Neo and NR primers used for genotyping and detection of the deleted allele are shown schematically as arrows under the genomic locus. PGK-HSV-tk, herpes simplex virus type I thymidine kinase under the phosphoglycerine kinase promoter cassette; PGK-Neo, neomycin gene under the phosphoglycerine kinase promoter. The PKG-Neo cassette is flanked by two frt sites. B. Southern blot of BamHI-digested genomic DNA from three ES clones. The upper panel represents the autoradiography after hybridization with the 5′ probe, shown in panel A. The expected recombination events are observed in the left and right lane, where the wild-type allele of 20 kb and the floxed allele of 14 kb are indicated by arrows. The middle lane shows the control wild-type alleles. In the lower panel, detection of the recombination was done using the 3′ probe. The left lane shows correct recombination events by detection of the wild-type allele of 14 kb and the floxed allele of 9 kb. The middle lane shows only the wild-type allele, and the right lane depicts incorrect recombination events. C. The left panel shows a 350-bp fragment from the wild type (wt) and 450 bp from the floxed allele (f) generated in one PCR with primers NF3 and NR. The right panel shows the genotype of the mice after in vivo Cre-mediated recombination. Wild-type (+) and deleted (−) alleles were detected using two PCRs for each sample. Primers Neo and NR generate a 350-bp fragment from the wild type and 500 bp from the deleted allele (del). The right lane shows a wild-type (+) control reaction.

FIG. 2.

RNA analysis of wild-type and myo^−/− mice. A. Total skeletal muscle RNA was used as template for RT-PCR analysis. Exons 4 and 5 (126-bp fragment) were amplified in all samples. Exon 3 (192 bp) could be amplified from the myo^+/+ and myo^+/− mouse RNA, but not from myo^−/− mouse RNA. A 343-bp fragment flanking exon 3 was amplified form the myo^+/+ and heterozygous myo^+/− mouse RNA, whereas a 175-bp fragment was amplified from myo^−/− mice, further demonstrating the deletion of exon 3. The control lane on the left shows the RT-PCR in the absence of mouse RNA, and the bottom panel shows the actin RT-PCR as a control. B. Northern blot analysis of total RNA isolated from skeletal muscle of myo^−/−, myo^+/−, and myo^+/+ mice was performed with probes for exon 3 and for exons 4 to 6. The probes did not detect any myotilin transcript in myo^−/− mice. A β-actin probe is shown as a total RNA loading control.

FIG. 3.

Immunoblot and Northern blot analysis of muscle proteins. A. The presence of Z-disk proteins was investigated by immunoblotting using crude skeletal and cardiac muscle homogenates of myo^+/+, myo^−/−, and myo^+/− mice. The immunoblot with antimyotilin antibody 151 shows lack of reactivity in the myo^−/− muscle, indicating the absence of the protein. α-Actinin, ZASP (32-kDa and 78-kDa isoforms), FATZ-1, and α-tubulin are normally expressed. Instead, a markedly increased telethonin reactivity is observed in the skeletal muscle and a minor increase is detected in the cardiac muscle. Myosin heavy chain (MHC) detection using Coomassie staining of the postblotted gel served as a loading control. B. Quantification of telethonin content in skeletal and cardiac muscle of myo^−/− mice. Western blot intensity of the telethonin signal (n = 3) was normalized to the actin content detected with the monoclonal AC40 antibody. The data obtained from nine measurements are reported as means + SEM of the telethonin content in the myo^+/+ skeletal or cardiac muscles, which was assigned 100 (SEM = 14.26). C. Western blot analysis of an ∼200-kDa muscle palladin isoform in skeletal muscles. The upper panel shows equal amounts of palladin detected with E1a antibody. The lower panel shows Ponceau S-stained nitrocellulose membrane to illustrate an equal amount of protein loading. D. Northern blot analysis of myopalladin transcript in the skeletal muscle. The upper panel is the myopalladin probe, and the lower panel is a β-actin probe used as a loading control.

FIG. 4.

Morphological analysis of myo^+/+ and myo^−/− skeletal and cardiac muscle. A. The upper panel shows transversal cryosections of gastrocnemius from 12-month-old myo^+/+ and myo^−/− mice, stained with hematoxylin and eosin. The lower panel shows morphology of the heart. No morphological differences were seen in either muscle type. Bar, 20 μm. B. Ultrastructural analysis of the myo^−/− mouse skeletal and cardiac muscle shows an unaffected morphology of the sarcomere. The Z-disk, observed as a dark line, is not changed in the myo^−/− mice compared with myo^+/+ mice. Bar, 500 nm. C. Immunofluorescence staining of isolated myofibrils shows Z-disk myotilin staining in myo^+/+ mice but not in the myo^−/− muscles. Staining of α-actinin demonstrates normal Z-disk distribution in myo^−/− skeletal muscle. Bar, 20 μm.

FIG. 5.

Growth curves, grip strength measurements, and endurance of myo^+/+ and myo^−/− mice. A. Weights of myo^−/− and myo^+/+ mice from birth to postnatal day 25. The genotypes have identical birth weights and develop similarly. Each data point value depicts the mean value ± SEM. B. Postweaning weight curves are depicted separately for each genotype and sex until 38 weeks of age. The intensive growth phase and the adult period show gender differences but are not affected by genotype. Each data point depicts the mean value ± SEM. C. Forelimb grip strength of >10-month-old male myo^−/− mice (n = 7) was compared with that of a group of gender- and age-matched myo^+/+ mice. The grip strength from four trials, each consisting of 10 pulls, was normalized to the average body weight. The mean values are presented as column plots ± SEM. The myo^−/− mice showed an insignificant decrease of strength compared with the age- and sex-matched control group. D and E. Running performance of myo^−/− and myo^+/+ mice. Daily voluntary running of myo^−/− mice (n = 5) compared with the myo^+/+ (n = 4) mice is expressed as the average daily running time (D) and distance (E). myo^−/− mice showed a small but statistically insignificant reduction in running time and distance. The data are shown as the average ± SD.