Both full-length myospryn and the TRIM region repress calcineurin activity in vitro. A) C2C12 myoblasts transfected with the 9X NFAT reporter construct plus calcineurin (+CN) alone, or plus calcineurin in combination with full length myospryn (+CN+FL) or TRIM (+CN+TRIM), or in combination with both (+CN+FL+TRIM). Results are shown as fold change over reporter alone, normalized to 1, from 3 independent experiments. Error bars = sem. B) C2C12 myoblasts transfected with the 9X NFAT reporter construct plus calcineurin alone, or plus calcineurin in combination with either the 3375–3739 or the Δ3 myospryn deletion constructs. Results are shown as fold change over reporter alone, normalized to 1. Error bars = sem. C) TRIM-like region does not inhibit the ability of full-length myospryn to interact with calcineurin. FLAG-CN was cotransfected into COS cells with Myc-full-length myospryn (Myc-FL; lane 2) or Myc-FL plus HA-TRIM (lane 3), immunoprecipitated with Myc antibody, and immunoblotted with FLAG antibody. Middle and bottom panels: immunoblotting with anti-FLAG and anti-HA to detect expression of FLAG-CN and HA-TRIM, respectively.

Myospryn TRIM region modulates calcineurin-mediated slow-fiber transformation in vivo. A) Real-time qRT-PCR analysis on 3 representative slow-fiber-specific genes. Pooled cDNA samples from each genotype (n≥3/genotype) were subjected to qRT-PCR using primers for myosin heavy chain type I (MYHC1), troponin I slow (TnI slow), and sarcoplasmic reticulum ATPase slow (SERCA2). Results are depicted as fold change relative to WT (normalized to 1). B) Transverse skeletal muscle sections were stained for type I fibers (green), laminin (red), and DAPI (blue) and imaged at ×100. C) Transverse sections taken from the midregion of the gastrocnemius and plantaris muscle were stained as in B, and total number of type I fibers was counted for each animal (n≥3/genotype; analysis included TRIM transgenic mice from lines 37 and 43). Error bars = sem. *P < 0.05, **< 0.01 vs. CN; ^##P < 0.01, ^###P < 0.001 vs. WT.

Myospryn and calcineurin interact in vitro and in vivo. A) Top panel: Myc-TRIM-like region of myospryn (TRIM, lane 2), an upstream region (1370–3098, lane 3), and full-length myospryn (FL, lane 4) were cotransfected into COS cells with FLAG-CN, immunoprecipitated with Myc antibody, and immunoblotted with FLAG antibody to detect the presence or absence of FLAG calcineurin. Middle and bottom panels: immunoblotting with anti-FLAG and anti-Myc to detect proper expression of the calcineurin and myospryn constructs, respectively. Expression of the full-length myospryn construct, which does not appear on this Western blot due to large size, has been confirmed previously (6). B) Fine mapping of the calcineurin binding sites of myospryn. Progressive Myc-tagged carboxy-terminal myospryn deletions were cotransfected with FLAG-CN and detected as described in A. C) Schematic diagram of Myc-myospryn constructs used to map the calcineurin binding domains, with interaction results at right. D) Myospryn TRIM domain interacts with full-length calcineurin. Myc-TRIM was cotransfected with FLAG-CN and detected as described in A. E) Endogenous myospryn and calcineurin are present in the same complex in neonatal rat ventricular myocytes (NRVMs). Protein extracts were immunoprecipitated with anti-calcineurin pan A or anti-GAL4 antibodies and immunoblotted with anti-myospryn. The 2 middle 10% input lanes were fractionated on the same SDS-PAGE gel as the immunoprecipitation lanes; image was cropped to remove blank wells between samples. Under these extraction conditions, myospryn is a labile protein. Therefore, the last 10% input lane on the right contains NRVM protein extract fractionated immediately after isolation on a separate gel and shows a more robust 10% input signal. F) GST pulldown assay. Protein extract from COS cells transfected with FLAG-CN was added to bacterially expressed GST or GST-TRIM bound to glutathione sepharose beads. Reactions were subjected to SDS-PAGE and immunoblotted with an anti-FLAG antibody. IB, immunoblotting.

Overexpression of TRIM in skeletal muscle does not alter normal muscle structure. A) Schematic depicting the FLAG-TRIM domain cloned downstream of the muscle creatine kinase (MCK) promoter to drive expression specifically in striated muscle. B) Four lines of MCK-TRIM transgenic mice were generated, and expression of the transgene was detected by Western blot. The highest-expressing line (line 37) was used for further analyses. C) General myofiber structure is unaltered in TRIM transgenic mice. Hematoxylin and eosin staining of transverse skeletal muscle sections was performed to visualize the myofiber cytoplasm (pink) and nuclei (blue-black arrows). View ×600. D) Immunostaining of transverse muscle sections reveals normal costamere localization patterns (white arrows) for FLAG-TRIM and all other proteins examined. View ×600. E) Anti-FLAG immunostaining of longitudinal muscle sections reveals a striated pattern of transgene expression (white arrows), similar to endogenous myospryn protein. F) PKA activity is unaltered in TRIM transgenic mice. PKA activity was measured in hindlimb muscle protein extracts from WT and TRIM transgenic mice (n=3/genotype). No significant difference was found between WT and TRIM transgenic mice.

MCK-TRIM mice display impaired regeneration in response to muscle injury. A) Hematoxylin and eosin staining for TA muscle sections. TRIM transgenic mice exhibit smaller regenerating fibers (pink) and increased nuclear infiltration (blue). View ×400. B) Transverse TA muscle sections were immunostained for laminin and used to quantify cross-sectional area (CSA) in C. View ×200. C) Myofiber CSA was quantified by measuring area of myofibers in laminin-stained images for both uninjured and injured TA muscles (>400 myofibers measured per animal), and is depicted as the recovery of the injured fiber size relative to the control uninjured fiber size (percentage) for 3 animals/genotype. D) Recovery of TA muscle mass is compromised in TRIM transgenic mice following muscle injury. Data are expressed as percentage of total weight recovered by the cardiotoxin-treated TA compared to the uninjured control TA weight from the same animal (average n=3/genotype). E) Reduction of MyoD and myogenin expression in TRIM regenerating muscle. Real-time qRT-PCR analysis was performed using pooled cDNA samples from each genotype (n=3); data are depicted as fold change of TRIM cardiotoxin-treated muscle relative to WT cardiotoxin-treated muscle (normalized to 1). F) Recovery of cross-sectional area in WT, TRIM, CN, and TRIM-CN double-transgenic mice. CSA was measured as in described in C. WT vs. TRIM, P < 0.01; TRIM-CN vs. CN, P < 0.05; TRIM-CN vs. TRIM, P = 0.057. Error bars = sem. *P < 0.05; **P < 0.01.