|
Advertisement | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
J. Biol. Chem., Vol. 282, Issue 23, 16713-16717, June 8, 2007
Fukutin-related Protein Associates with the Sarcolemmal Dystrophin-Glycoprotein Complex*
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
DG (6). Extracellular
DG and the transmembrane-spanning
DG bind to dystrophin, sarcoglycans, and other proteins to form the dystrophin-glycoprotein complex (DGC), which serves as a critical structural link between the cell cytoskeleton, the sarcolemma, and the extracellular basement membrane.
DG glycans, detected by antibody IIH6, mediate the interaction between the DGC and extracellular matrix proteins that contain laminin LG domains. Therefore, reduced
DG glycosylation may weaken the cell to matrix link, increasing structural instability and disease (6).
Previous studies have examined the localization of FKRP in a spectrum of cultured cells (e.g. COS7, SH-SY5Y, C2C12). The variable results suggested that FKRP is a resident of the Golgi, the rough endoplasmic reticulum, or perinuclear regions (711). In this study, we have examined FKRP protein complexes and their location in skeletal muscle of wild-type and dystrophic mice. We report that FKRP is localized at the muscle sarcolemma and that it co-fractionates with the DGC. Furthermore, disruption of the DGC (by alkaline treatment or genetic deletion) revealed that FKRP sedimentation and localization are dependent on the DGC. Overall, these data suggest that FKRP associates with the sarcolemmal DGC and that it plays a unique role in dystroglycanopathy muscle disease.
| EXPERIMENTAL PROCEDURES |
|---|
|
|
|---|
DG, Rbt 83
DG, and 21B5
SG have been described (1214). MANDRA1 dystrophin (Hybridoma Bank, University of Iowa), heparin sulfate proteoglycan (perlecan, Chemicon), FLAG (Sigma), and HA.11 (Covance) antibodies were used. Custom rabbit polyclonal antisera against mouse FKRP C terminus and FKRP 176189 peptides (supplemental Table 1) were developed (Sigma). Antisera were affinity-purified using peptide or recombinant FKRP. FKRP 176189 purifications were pooled and dialyzed.
MiceMouse procedures were Animal Care and Use Review Form-approved. C57BL/6 (Jax 000664), mdx (dystrophin-deficient, Jax 001801), MCK-Cre x T30 (floxed DG excision, MCK DG) (15),
SG null (16), myd (LARGE-deficient, Jax 000300) (17), dysferlin-null (18), 129sve (Taconic), and C57BL/10 (Jax 000665) strains were used. Skeletal muscle from 5-to-24-week mixed gender mice was frozen for immunofluorescence. Mice were 830 weeks old for biochemical analyses.
VirusesMouse FKRP or fukutin cDNAs (Open Biosystems) were used to create 3'-2XFLAG (CFLAG2FKRP) or 3XHA-(fukutin3XHA) tagged constructs (supplemental Table 2). cDNAs were sequenced (University of Iowa DNA Facility) and cloned into pacAd5CMV shuttle for virus generation (University of Iowa Gene Transfer Vector Core) (19). Five-to-seven-day-old C57BL/6 pups were injected in tibialis anterior or calf muscles with 1 µl of virus (29 x 1010 plaque-forming units/ml) mixed in 9 µl of saline (20).
BiochemistryDGC enrichment was adapted from previous protocols (14); for buffer constituents, see supplemental Table 3. Skeletal muscle (3 g) was solubilized and separated at 142,000 x g. Supernatants were enriched with WGA-agarose (Vector Laboratories). Pooled WGA elutions were adjusted to 50 mM NaCl, applied to DEAE resin (Whatman), washed, and eluted with 100, 150 (Fig. 2 only), 500, and 750 mM NaCl. Elutions (150 or 500 mM) were loaded onto 1030% sucrose gradients (Biocomp Instruments) and sedimented at 220,000 x g for 2 h; 14 fractions were collected. For alkaline experiments (21), DEAE elutions were titrated to pH 7.4 or pH 11, incubated for 1 h (22 °C), and sedimented on pH 7.4 or pH 11 sucrose gradients. SDS-PAGE (315%) and Western blotting on Immobilon-P polyvinylidene difluoride (Millipore) were adapted from standard protocols (22). Blocking and antibody incubations were done in Tris-buffered saline + 0.1% Tween 20 + 5% milk, low salt Tris-buffered saline + 0.1% Tween 20 (75 mM NaCl) + 5% milk (FKRP C terminus) or + 3% bovine serum albumin (FKRP 176189). Horseradish peroxidase-conjugated secondary antibodies were used (Chemicon, Roche). Chemiluminescence (Pierce) was digitally detected (Alpha Innotech).
ImmunofluorescenceSeven-µm cryosections were incubated overnight in primary antibody (4 °C) and for 45 min in Cy3-(Jackson ImmunoResearch) or Alexa Fluor 488-(Molecular Probes) conjugated secondary antibodies plus 4',6-diamidino-2-phenylindole dihydrochloride nuclear stain (Sigma), similar to previous procedures (14). For Fig. 1, sections were fixed and permeabilized (2% paraformaldehyde, 0.2% Triton X-100); for peptide competition, antibody and 1 mg/ml FKRP C terminus or control peptide (supplemental Table 1) were incubated 1 h and applied to a slide. Images were taken at x60 (Bio-Rad MRC600 or Olympus BX61 confocal) or x40 or x20 (Leica DMRXA) magnification. Image parameters were identical for x20 and x40 pictures of the same protein in the same panel.
Analyses and Digital ImagesWestern blot band intensity (the area under the peak) was autodetected (AlphaEase FC, Alpha Innotech), normalized to the maximum signal for each blot, and plotted versus the sucrose fraction number (Microsoft Excel). Western blot and fluorescence images were adjusted for size and signal strength in PhotoShop (Adobe). Adjustments were applied equally to the entire image; images using the same primary antibody in the same panel were adjusted identically.
| RESULTS |
|---|
|
|
|---|
DG post-translational modification. To begin to elucidate the role of FKRP in skeletal muscle, we tested mouse C57BL/6 quadriceps for FKRP expression using antibodies against mouse FKRP C-terminal (amino acids 481494) or internal (amino acids 176189) epitopes. Native FKRP immunofluorescence surrounded individual muscle fibers in a pattern consistent with sarcolemmal staining and was specific for the FKRP epitope as incubation with antigen peptide eliminated surface signal (Fig. 1A). Native FKRP was also detected at the sarcolemma in longitudinal muscle sections (Fig. 1B). FKRP signal was robust at neuromuscular junctions (supplemental Fig. 1); however, whether FKRP is targeted to this location or enhanced staining simply reflects additional membrane surface at the end plate is unclear. The FKRP sarcolemmal staining pattern was reproducibly observed for the C-terminal (Fig. 1) and internal FKRP epitopes (supplemental Fig. 1), although C-terminal detection was typically stronger. FKRP expression in quadriceps was representative of all muscles tested (hamstring, tibialis anterior, gastrocnemius, and soleus; data not shown).
|
The DGC is stably expressed at the sarcolemma and can be purified via multistep biochemical enrichment (12). As we detected FKRP at the membrane surface and this protein is implicated in
DG function, we hypothesized that FKRP may associate with the DGC. To test this, we assessed co-enrichment of FKRP with the DGC in C57BL/6 skeletal muscle. Like the DGC, a
50-kDa FKRP signal was enriched following WGA and ion exchange chromatography (Fig. 2A). When the sample was sedimented by sucrose-gradient fractionation, FKRP expression overlapped with that of all DGC components tested (dystrophin,
SG, and
- and
DG; Fig. 2, B and C, left), and FKRP and DG peak signals intensities were consistently detected in the same or adjacent fractions, demonstrating co-sedimentation (n = 3). When DGC-enriched samples were alkaline-treated to disrupt pH-dependent protein binding (21), FKRP and DGC components shifted, by varying degrees, to lighter fractions when compared with pH 7.4 (Fig. 2, B and C, right), indicating partial disruption of the complex. Dystrophin and
SG were predominately expressed in fractions 811 and fractions 610, respectively, whereas
DG,
DG, and FKRP were primarily concentrated in fractions 48 (96.7 ± 1.9% FKRP versus 94.2 ± 2.2%
DG signal in fractions 48, n = 3). Protein peaks in the sedimentation profile were detected in the same (n = 1) or in adjacent fractions (n = 2). The shift in FKRP sedimentation most closely mimics that of DG, suggesting that FKRP may link to the DGC via the DG subcomplex. However, as there is still some partial overlap with other DGC components, these experiments do not exclude the possibility of other FKRP-DGC interactions. Overall, these data suggest that FKRP is a novel protein that associates with the DGC.
|
SG-null, and
DG-hypoglycosylation (LARGEmyd) mice relative to wild-type and non-DGC dystrophic (dysferlin) mice. As expected, FKRP was localized at the muscle sarcolemma in wild-type and non-DGC dystrophic muscle (Fig. 3). Uniform FKRP expression was also observed in
SG-(albeit reduced) and LARGE-deficient muscle, indicating that neither the sarcoglycan complex nor
DG LARGE-dependent glycosylation is required for FKRP localization per se. In contrast, FKRP sarcolemmal staining was patchy or absent in dystrophin- or DG-deficient models. Notably, the few muscle fibers with FKRP expression were also positive for
- and
DG (Fig. 3, asterisks), suggesting that the sarcolemmal localization of FKRP is mediated by its interaction with DG.
|
- and
DG) shift to lighter fractions in dystrophin-deficient (mdx) muscle, where the DGC complex is incomplete (Fig. 4A). Similarly, a loss of
SG shifts these proteins to smaller molecular weight fractions (Fig. 4B). In both cases, FKRP expression appeared to be reduced. Genetic deletion of dystrophin or sarcoglycans reduces DGC components at the sarcolemma, presumably by destabilizing the complex, and in microsomal membrane preparations from these mice, the expression of DG and FKRP was also reduced (data not shown). Overall, these findings suggest that maintenance of FKRP at the membrane requires DG. In contrast, in non-DGC-related dysferlin null muscle (Fig. 4B) and in
DG hypoglycosylation myd muscle (data not shown), FKRP sedimentation was unchanged; it continued to co-migrate with both
DG and
DG. FKRP and DG showed strong co-fractionation in all models tested with only slight differences in distribution or location of peak signal, which may be attributed to experimental variation or a mild preference for FKRP to interact with some subpopulation of DG complexes. Combined, these data reveal a strong and consistent correlation between FKRP sarcolemmal expression and the presence of DG. Although any causative link between the loss of sarcolemmal FKRP and the concomitant loss of DG in these mouse models is currently unknown, it is possible that the absence of FKRP may contribute to pathogenesis in these dystrophies.
|
| DISCUSSION |
|---|
|
|
|---|
Cell surface and extracellular localization has been described for several glycosyltransferases and related enzymes (2327). Localization can be tissue-specific, may require proteolysis, and may be regulated by mechanisms such as alternative splicing or putative RNA editing (24, 26, 27). In this study, native skeletal muscle FKRP was detected as a smaller
50-kDa protein in skeletal muscle (predicted core protein
55 kDa). As the entire FKRP open reading frame is contained within one exon, alternative splicing is unlikely to regulate protein size. However, motif analysis (28) of the protein sequence detects putative signal peptide and protease recognition sites that could potentially generate
50-kDa FKRP.
The DGC, including dystrophin, DG, and sarcoglycans, resides at the plasma membrane, anchoring the extracellular basement membrane and cell surface to the cytoskeleton and acting as a scaffold for other proteins (e.g. neuronal nitric oxide synthase, growth factor receptor-bound protein 2 (Grb2)) (6). We have demonstrated co-enrichment and co-fractionation of FKRP with the DGC via an FKRP association with DG but not dystrophin or sarcoglycans. Although the putative glycosyltransferase LARGE has been shown to bind DG (20), its interaction is believed to be strictly intracellular and transitory as DG passes through the glycosylation pathway. In contrast, our data suggest that the FKRP-DGC association represents a stable, mature complex that is not disrupted by high salt or alkaline conditions. Our finding that FKRP is disrupted in mouse models of muscular dystrophy in which DG is destabilized or lost further suggests that FKRP localization and association may depend on DG. Interestingly, a recent clinical study has identified a novel FKRP mutation in two patients. These patients have reduced sarcolemmal DGC proteins rather than a selective loss of
DG glycosylation (29). Therefore, it is possible that FKRP and DGC expression are mutually codependent.
Although the precise activity of FKRP remains elusive, patient data support a role in the
DG glycosylation pathway. Although we did not detect native FKRP in intracellular compartments of skeletal muscle, it is possible that FKRP is enzymatically active as the DGC complex is assembled and processed. Alternatively, FKRP could regulate glycan modification, protein interactions, transit time through internal compartments, or DGC stability. FKRP as a molecular chaperone for DG processing and/or targeting, or vice versa, is an intriguing possibility. Whether FKRP activity is required at the sarcolemma for normal muscle function is unclear; however, known glycosyltransferases may possess enzymatic or lectin binding activity at the cell surface (2325, 30). Overall, our data indicate that FKRP is present at the muscle cell surface, associates with the DGC, and may have a unique role in the
DG processing pathway.
| FOOTNOTES |
|---|
The on-line version of this article (available at http://www.jbc.org) contains two supplemental figures and three supplemental tables. ![]()
1 Recipient of Alberta Heritage Foundation for Medical Research and Peter A. Getting Postdoctoral Award support. ![]()
2 An investigator of HHMI. To whom correspondence should be addressed: HHMI, Dept. of Molecular Physiology and Biophysics, University of Iowa, 4283 Carver Biomedical Research Bldg., 285 Newton Rd., Iowa City, IA 52242. Fax: 319-335-6957; E-mail: kevin-campbell{at}uiowa.edu.
3 The abbreviations used are: FKRP, fukutin-related protein; DG, dystroglycan;
DG,
-dystroglycan;
DG,
-dystroglycan; DGC, dystrophin-glycoprotein complex;
,
,
, or
SG,
-,
-,
-, or
-sarcoglycan; MCK, muscle creatine kinase promoter; LARGE, like-glycosyltransferase; HA, hemagglutinin epitope; WGA, wheat germ agglutinin; DEAE, DE52 diethylaminoethyl cellulose. ![]()
| ACKNOWLEDGMENTS |
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
L. U. Yamamoto, F. J. Velloso, B. L. Lima, L. L.Q. Fogaca, F. de Paula, N. M. Vieira, M. Zatz, and M. Vainzof Muscle Protein Alterations in LGMD2I Patients With Different Mutations in the Fukutin-related Protein Gene J. Histochem. Cytochem., November 1, 2008; 56(11): 995 - 1001. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Thornhill, D. Bassett, H. Lochmuller, K. Bushby, and V. Straub Developmental defects in a zebrafish model for muscular dystrophies associated with the loss of fukutin-related protein (FKRP) Brain, June 1, 2008; 131(6): 1551 - 1561. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |