Skip to main content
NCBI home page
Molecular Biology of the Cell logoLink to Molecular Biology of the Cell
. 2009 Feb 1;20(3):834–845. doi: 10.1091/mbc.E08-07-0753

A Myopathy-linked Desmin Mutation Perturbs Striated Muscle Actin Filament Architecture

Gloria M Conover 1,*, Syerra N Henderson 1,†, Carol C Gregorio 1,✉
Editor: Robert G Parton
PMCID: PMC2633376  PMID: 19005210

Abstract

Desmin interacts with nebulin establishing a direct link between the intermediate filament network and sarcomeres at the Z-discs. Here, we examined a desmin mutation, E245D, that is located within the coil IB (nebulin-binding) region of desmin and that has been reported to cause human cardiomyopathy and skeletal muscle atrophy. We show that the coil IB region of desmin binds to C-terminal nebulin (modules 160-164) with high affinity, whereas binding of this desmin region containing the E245D mutation appears to enhance its interaction with nebulin in solid-phase binding assays. Expression of the desmin-E245D mutant in myocytes displaces endogenous desmin and C-terminal nebulin from the Z-discs with a concomitant increase in the formation of intracellular aggregates, reminiscent of a major histological hallmark of desmin-related myopathies. Actin filament architecture was strikingly perturbed in myocytes expressing the desmin-E245D mutant because most sarcomeres contained elongated or shorter actin filaments. Our findings reveal a novel role for desmin intermediate filaments in modulating actin filament lengths and organization. Collectively, these data suggest that the desmin E245D mutation interferes with the ability of nebulin to precisely regulate thin filament lengths, providing new insights into the potential molecular consequences of expression of certain disease-associated desmin mutations.

INTRODUCTION

Striated muscle cells maintain their shape, drive contraction, and generate mechanical force through the efficient integration of microfilament, microtubule, and intermediate filament (IF) function. Actin-thin filaments anchored at the Z-discs interdigitate with the myosin-thick filaments in the middle of the sarcomere, forming the basic repeating contractile units of striated muscle (for review see Clark et al., 2002). Giant nebulin molecules span the entire length of the thin filaments, with their N-terminal region interacting with the pointed end of the filaments and their C-terminal region anchored at the Z-discs. Previously, a yeast two-hybrid study reported that the C-terminal region of nebulin interacts with the desmin, illustrating a direct link between sarcomeres and IFs (Bang et al., 2002).

Like other IF proteins, desmin is composed of three major structural domains (for review see Herrmann and Aebi, 2004). At its N-terminal end is a nonconserved, non-α-helical head domain; in the middle is a conserved amphipathic α-helical coiled-coiled rod domain that contains three linker regions, and at its C-terminal end is a nonconserved tail domain (Geisler and Weber, 1982). Desmin filaments form a 3D cytoplasmic scaffold linking individual myofibrils laterally at the periphery of the Z-discs connecting the sarcomere to the sarcolemma, T-tubules, mitochondria, and nucleus (Lazarides and Hubbard, 1976; Lazarides, 1978). Desmin is also enriched at intercalated discs/desmosomes, myotendinous junctions, and in Purkinje fibers and is inconsistently detected at the M-line (Tokuyasu et al., 1983; Thornell et al., 1985; O'Neill et al., 2002). Together, the multiple cellular localizations of the desmin filaments are thought to maintain the structural integrity of myocytes and contribute to the force transmission and longitudinal load bearing (e.g., Shah et al., 2004). Desmin −/− mice develop cardiomyopathies characterized by mitochondrial abnormalities, extensive cardiomyocyte death, fibrosis, calcification, and eventual heart failure (Weisleder et al., 2004). Skeletal and cardiac muscle from these desmin −/− mice exhibit misaligned sarcomeres and disintegrated myofibrils, as well as accumulation of mitochondria (Milner et al., 1996; Li et al., 1997). Collectively, these studies highlight the importance of desmin IF in structurally integrating the striated muscle cytoskeleton.

Nebulin is a highly modular protein composed of a unique N-terminal domain, a central region containing ∼35 residue actin-binding repeats organized into single modules (M) (M1-8 and M163-185) and into sets of seven-module super-repeats (M9-162) and a C-terminal end region containing a serine-rich linker and SH3 domain (e.g., Labeit and Kolmerer, 1995; Wang et al., 1996). Each single nebulin repeat is predicted to bind an actin monomer with the potential of nebulin binding ∼200 actin monomers along its length (Pfuhl et al., 1996; Wang et al., 1996). Nebulin represents ∼2–3% of total myofibrillar proteins in skeletal muscle (Wang and Wright, 1988), whereas it is present at much lower levels in cardiac muscle (Fock and Hinssen, 2002; Kazmierski et al., 2003; Bang et al., 2006). Several studies have indicated that nebulin likely plays an important role in thin filament regulation, together with actin filament capping proteins, to maintain actin filament lengths (for reviews see Fowler et al., 2006; Horowits, 2006; Pappas et al., 2008). In particular, reduction of nebulin via siRNA approaches resulted in longer thin filament lengths in rat cardiac myocytes, whereas knockout of nebulin in mice resulted up to a ∼15% reduction in the lengths of the skeletal muscle thin filaments (McElhinny et al., 2005; Bang et al., 2006; Witt et al., 2006). The nebulin −/− mice died within 1–3 wk after birth and had defects in contractile activity and in active tension production. Misaligned sarcomeres, myofibril splitting with moderately wider Z-discs, undefined M-lines, absence of H-zones, and accumulation of mitochondria and nemaline bodies were also reported in nebulin −/− skeletal muscles (Bang et al., 2006; Witt et al., 2006). Remarkably, some of the structural abnormalities described in the nebulin −/− mice are similar to those described for desmin −/− mice (Milner et al., 1996; Li et al., 1997).

Single missense mutations in nebulin or desmin can lead to heterogeneous human muscle disorders that typically follow an autosomal dominant pattern of inheritance. Nebulin mutations can result in nemaline myopathy, a disease of the thin filament that is characterized by the presence of nemaline bodies between muscle fibers that contain aberrantly arranged Z-disc and I-band proteins (for review see Wallgren-Pettersson et al., 2007). Missense mutations in desmin can result in dilated and restrictive cardiomyopathies, as well as desmin-related myopathy (DRM), a distinct form of myofibrillar myopathy (Taylor et al., 2007). DRM is characterized by cytoplasmic desmin-positive aggregates and streaming of Z-bands (for review see Goldfarb et al., 2004). Many mutations (>40 thus far) scattered throughout desmin have been found to result in DRM (e.g., Vrabie et al., 2005). Earlier studies proposed that the DRM mutations affected IF network assembly. However, this idea does not explain the pathology observed in many DRM cases. For example, it was found that six DRM-associated desmin rod mutants (A213V, E245D, A360P, N393I, Q389P, and D399Y) formed seemingly normal IFs in in vitro assembly assays (Bar et al., 2005). Likewise, a number of desmin tail mutant proteins integrated into preexisting IF networks in C2C12 mouse skeletal myoblasts and formed extended filamentous arrays in in vitro assembly assays (Bar et al., 2007). These studies suggest that the cellular defects that lead to DRM are complex and involve many cellular processes, many of which are poorly understood.

Here, we sought to understand the role of the interaction of the desmin IF network with the integral sarcomeric component, nebulin. Nebulin M160-164 binds to desmin coil IB with high affinity, and the addition of the DRM-associated E245D mutation in this desmin fragment alters its interaction with nebulin. Expression of the E245D desmin mutant results in reduction in the assembly of endogenous desmin and nebulin at the Z-disc and cytoplasmic aggregate formation, as seen in biopsies of diseased DRM muscle. Expression of the desmin E245D mutant also results in two striking alterations in sarcomeric actin filament structure. First, most sarcomeres contained nonuniform, nonstriated phalloidin staining (i.e., absence of gaps in staining in the H-zone) indicating thin filament elongation. Second, phalloidin staining also revealed the presence of actin filaments with reduced lengths in cardiac myocytes. Hence, we propose that alterations in the desmin–nebulin interaction, and in particular, alterations in nebulin function may contribute to the development of DRM in human muscle.

MATERIALS AND METHODS

Cloning and Site-directed Mutagenesis

The sequences of the primers used in this study are listed in Supplemental Table S1. To generate an N-terminal glutathione S-transferase (GST)-tagged recombinant nebulin M160-170 fragment, the fragment was amplified by PCR from mouse heart cDNA (XM_130232) and cloned into pGEX4T-1 (GE Healthcare, Piscataway, NJ). The N-terminal His-tagged nebulin fragments M160-170, M165-167, and M160-164 were subcloned from pGST-nebulin M160-170 into the pET-11 vector (Novagen, Madison, WI) as described (Pappas et al., 2008). The desmin coil IB region was amplified from human skeletal cDNA (MN_001927) and cloned into pGEX4T-1.

The following N-terminal green fluorescent protein (GFP)-desmin constructs were cloned into pEGFP-C1 (Clontech, Mountain View, CA): enhanced GFP (EGFP)-desmin containing residues 2-470, EGFP-head containing residues 1-103, EGFP-coil IB containing residues 151-263, EGFP-coil IIB containing residues 304-412, and EGFP-tail containing residues 411-469. The E245D mutation was introduced into GST-desmin and GST-coil IB using a PCR-based QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA). PCR products were digested with DpnI, and transformed into XL1-blue super competent Escherichia coli. All constructs were verified by sequencing.

Expression, Purification, and Biotinylation of Recombinant Protein Fragments

GST and 6X His-tagged constructs were transformed into BL21-Codon Plus (DE3)-RIL competent E. coli (Stratagene) for protein production. The bacteria were induced with 1 mM IPTG for 2–3 h at 37°C, and pellets were resuspended in 1% Triton X-100 PBS for GST-fusion proteins or in lysis buffer (50 mM NaH2PO4, 300 mM NaCl, and 10 mM imidazole, pH 8.0) for His-tagged fusion proteins. The protease inhibitors, 0.5 mM Pefabloc, 10 μg/ml leupeptin, 10 μg/ml pepstatin A, and 1 μg/ml aprotinin (Calbiochem, San Diego, CA), were added to all buffers used for protein purification. Samples were sonicated on ice and centrifuged, and the supernatants were incubated on glutathione Sepharose 4B resin (GE Healthcare), or for His-tagged proteins on 50% Ni-NTA slurry in a 15-ml conical tube (Qiagen, Valencia, CA) for 1–2 h at 4°C. The beads with the bound proteins were placed in 10-ml disposable columns and were washed either in 1% Triton X-100 PBS for GST-fusion proteins or in 50 mM NaH2PO4, 300 mM NaCl, and 20 mM imidazole, pH 8.0, for His-fusion proteins. GST-fusion proteins were eluted with 10 mM glutathione, whereas His-fusion proteins were eluted with 50 mM NaH2PO4, 300 mM NaCl, and 250 mM imidazole, pH 8.0. Samples were analyzed on SDS-PAGE gels, immediately dialyzed into binding buffer (20 mM HEPES, 80 mM KCl, 2 mM MgCl2, pH 7.4), snap-frozen in liquid nitrogen, and stored at −80°C until use. Recombinant protein fragments were biotinylated at a target ratio of 1:10 wt/wt biotin to protein fragment according to the manufacturer's instructions (Zymed, San Francisco, CA). A Coomassie-stained 12% SDS-PAGE gel of each of the recombinant proteins used in this study is shown (Supplemental Figure S2).

Blot Overlay Assays

For blot overlay analysis, flash-frozen rat psoas and heart muscles were pulverized with a mortar and pestle and solubilized in SDS-sample buffer. Samples were incubated at 70°C for 5 min before being resolved on a 12% SDS-PAGE gel. Proteins were transferred to nitrocellulose membranes (0.2 μm, PerkinElmer Life and Analytical Sciences, Boston, MA) and stained with Ponceau-S. Membrane strips were blocked with 5% milk in binding buffer with 0.05% Tween for 1 h at RT and incubated with 0.5–1.0 μg/ml GST-desmin coil IB or GST alone overnight at 4°C. Strips were washed and incubated with HRP-conjugated GST antibodies (GE Healthcare) diluted 1:8000 in binding buffer for 1 h at RT and then incubated in SuperSignal chemiluminescent substrate (Pierce, Rockford, IL) and exposed to BioMax MR film (Eastman Kodak, Rochester, NY). To detect the mobility of endogenous desmin on the blots, the nitrocellulose membranes were stripped at RT with 0.1 M glycine, pH 2.5, in PBS for 15 min and washed in PBS for 15 min, followed by an incubation in 1 M NaCl in PBS for 15 min. After a 15-min wash in PBS, stripped blots were probed with monoclonal anti-desmin antibodies (DE-U-10: Sigma-Aldrich, St. Louis, MO) at 0.8 μg/ml in 5% milk/0.05% Tween/PBS for 1 h at RT. Primary antibody binding was detected with HRP-conjugated anti-mouse antibodies diluted 1:20,000 (Jackson ImmunoResearch, West Grove, PA).

ELISA Assays

Ninety-six-well microtiter ELISA plates were coated with 100 μl of recombinant His-tagged nebulin fragments (40 nM in Figure 3C or 150 nM in Figure 3, A and B) in 0.1 M carbonate buffer (pH 9.6) overnight at 4°C, following a modified protocol from McElhinny et al. (2001). Wells were blocked for 45 min with 0.2% BSA 0.05% Tween in binding buffer for 1 h at RT, followed by 1-h incubation with biotinylated GST-tagged desmin fragments in binding buffer at the indicated concentrations. After washes with binding buffer, wells were incubated with alkaline phosphatase-conjugated streptavidin (Pierce). After washes with binding buffer, the p-nitrophenyl phosphate (pNPP, Sigma-Aldrich; N9389) substrate (1 mg/ml in 0.1 M glycine, 1 mM MgCl2, 1 mM ZnCl2, pH 10.4) was added to the wells and incubated for 30 min at 37°C. Binding of desmin to nebulin fragments was determined by reading the absorbance at 405 nm. No significant binding was detected when the desmin fragments were tested for binding with GST protein alone. The values shown were obtained by subtracting the values from obtained wells containing binding buffer plus 0.2% BSA and the appropriate biotinylated protein fragment in solution (i.e., nonspecific binding). The estimated dissociation constant (Kd), determined from the saturation curves, was calculated from nonlinear regression curves fitted with a one-site binding equation using Prism software (GraphPad, San Diego, CA). Values shown are the average of triplicate wells ± SD.

Figure 3.

Figure 3.

Open in a new tab

Desmin coil IB specifically binds to nebulin M160-164 in ELISAs. (A) Desmin coil IB interacts with recombinant nebulin M160-170. The interaction of nebulin M2 with Tmod was used as positive control, and α-actinin probed with desmin coil IB was used as a negative control. (B) GST alone and His-tagged nebulin fragments representing modules M160-170, M160-164, M165-168, M169-171, M171-177, and M182-C-terminal were plated onto microtiter plates and incubated with biotinylated GST-coil IB. A strong signal was only detected when desmin coil IB was added to nebulin M160-164. (C) In this example, desmin coil IB binds to nebulin M160-164 with an estimated Kd of ∼35 nM with a Bmax of ∼0.9 (○), whereas desmin coil IB E245D mutant fragment binds with a Kd of ∼30 nM with a Bmax of ∼3.3 (●). A molar ratio of 1:1 was used in A and B.

Cell Culture and Transfections Protocols

Primary cultures of skeletal myocytes were isolated from pectoralis muscle day-11 chick embryos as previously described (Ojima et al., 1999; McElhinny et al., 2005). Myoblasts were plated at a density of ∼4.5 × 105 cells per 35-mm culture dish coated with BD matrigel (BD Biosciences, San Jose, CA) following the manufacturer's specifications and were maintained for 96 h in growth medium after isolation. Cardiomyocytes were isolated from rat embryonic day 18 hearts, plated at a density of ∼5 × 105 cells per 35-mm culture dish, and maintained in DMEM supplemented with 10% FBS and 1% penicillin and streptomycin antibiotics as described previously (Gustafson et al., 1987). Myocytes were transfected with Effectene reagent (Qiagen) ∼24 h after plating according to the manufacturer's recommendations. Cardiac cultures were incubated for an additional 6 d before fixation.

Immunofluorescence Microscopy and Antibodies

To examine the localization of various proteins, myocytes were stained as previously described (McElhinny et al., 2005). All cells were treated with relaxing buffer (150 mM KCl, 5 mM MgCl2, 10 mM MOPS, 1 mM EGTA, 4 mM ATP, pH 7.4) before fixation in 3% paraformaldehyde. Cells were permeabilized with 0.2% Triton X-100 in PBS and blocked in 2% BSA/1% donkey serum in PBS. Myocytes were stained with rabbit anti-desmin (1:5: Biomeda, Foster City, CA), anti-Tmod1 (10 μg/ml), anti-N-terminal titin Z1-Z2 (1:100), anti-C-terminal nebulin M160-164 (10 μg/ml), and anti-C terminal nebulin M176-181 (1:50; kindly provided by Dr. Siegfried Labeit, Universitatsklinikum Mannheim, Germany) antibodies, and monoclonal anti-myosin F59 (1:10 of culture supernatant: kindly provided by F. Stockdale, Stanford University, Stanford, CA), anti-myomesin (1:100: kindly provided by E. Ehler, King's College London, United Kingdom, and J.-C. Perriard, ETH, Zurich, Switzerland) and anti-sarcomeric α-actinin (1:10,000: clone EA-53, Sigma-Aldrich) antibodies. When the GFP-tagged desmin fusion proteins were expressed in fibroblasts (contaminating the primary myocytes cultures), they were not detected with the anti-desmin antibodies, suggesting that this antibody recognizes the endogenous protein only (and not the GFP-tagged recombinant proteins) in transfected myocytes (data not shown). Secondary antibodies purchased from Jackson ImmunoResearch and Invitrogen (Carlsbad, CA) included Alexa Fluor 350–conjugated goat anti-mouse IgG (1:250), and Texas Red–conjugated donkey anti-rabbit IgG (1:600). Texas-Red phalloidin (Sigma-Aldrich) was used to label the actin filaments (1:1500).

Images of myocytes were acquired with a Deltavision deconvolution microscope (Applied Precision, Issaquah, WA) with a 100× objective (1.3 NA) using a CoolSnap HQ charged-couple device camera (Photometrics, Tucson, AZ). The images were processed for presentation with Adobe Photoshop CS (San Jose, CA), and statistical analysis was performed using Microsoft Excel (Microsoft, Redmond, WA), and proFit 6.1.2 software (QuantumSoft, Zurich, Switzerland). Intensity plots were performed on three adjacent sarcomeres of Texas Red-phalloidin–stained myocytes (boxes 6.57 × 1.96 μm) and analyzed with ImageJ (http://rsb.info.nih.gov/ij/; National Institutes of Health, Bethesda, MD). Sarcomere lengths were determined from α-actinin staining by measuring the distance between the peaks using the intensity plot tool in ImageJ. Actin filament lengths determined from Texas-Red phalloidin staining were measured using Softworks software from pointed end to pointed end in two independent experiments. Thin filament lengths as determined by staining for Tmod1 (from pointed end to pointed end) were measured using ImageJ. Both sets of measurements (phalloidin and Tmod1) were divided by 2 to obtain thin filament lengths from barbed to pointed ends (half sarcomeres). The mean and SD of phalloidin measurements were computed using proFit 6.1.2 and by fitting the distribution of individual measurements with a Gaussian cumulative distribution function. The statistical significance of the difference in the mean value of the phalloidin measurements was obtained for GFP-desmin coil IB and GFP-desmin coil IB E245D by performing a nonparametric Kolmogorov-Smirnov test.

To visualize the distributions of individual measurements (see Figure 9B), the histograms of the data were plotted and normalized so that the sum of all bins equaled 1 (i.e., the number of measurements falling in each histogram bin was divided by the total number of measurements). A Gaussian curve with the mean and SD as calculated from the individual measurements was overlaid on each histogram. Only the amplitude of the Gaussian curve was fitted to the binned data. This fit was performed using the Levenberg-Marquardt algorithm in proFit 6.1.2.

Figure 9.

Figure 9.

Open in a new tab

Alterations in actin filament lengths in cells expressing desmin E245D. (A) Bar graph displaying the percentage of myofibrils with nonstriated (i.e., lack of H-zone) phalloidin staining in cells expressing the indicated GFP-desmin fragments. Student's t test was used to determine significance (** p < 0.01) of the indicated desmin fragment, compared with GFP alone. (B) Actin filament lengths were measured in sarcomeres containing H-zones in phalloidin staining in cells expressing GFP-coil IB or GFP-coil IB E245D. Normalized measurements are displayed as histograms with Gaussian fit to the data. (C) Pixel intensity profiles of phalloidin staining in three adjacent sarcomeres in cells expressing desmin GFP-coil IB (a) or GFP-coil IB E245D (b) with discernable gaps at the H-zone. No major differences in sarcomere lengths (distance between Z-lines) were detected; however, the Z-disc profile was broader, and nonuniform in cells expressing coil IB E245D.

RESULTS

Desmin Interacts with Nebulin Modules M160-170

The interaction of desmin with nebulin was first identified by a yeast two-hybrid study using C-terminal nebulin M160-183 as a bait (Bang et al., 2002). To confirm the interaction of desmin with nebulin and to begin to narrow down the interacting sites, we performed blot overlay assays on lysates of rat psoas and heart muscle. After transfer to nitrocellulose membranes, lysates were probed with recombinant GST-nebulin M160-170 or GST alone, followed by incubation with HRP-conjugated anti-GST antibodies (Figure 1, lanes 1 and 3). GST-nebulin M160-170 bound to several bands including a protein with the same molecular weight as endogenous desmin (∼52 kDa, Figure 1, lane 1), as determined by Western blot analysis (Figure 1, lane 2). In addition, GST-nebulin M160-170 also bound with a strong signal to two other myofibrillar proteins, with mobility of ∼100 and ∼34 kDa, probably corresponding to α-actinin and CapZ, respectively. Similar signals were obtained when rat heart muscle lysates were probed with GST-nebulin M160-170 (data not shown).

Figure 1.

Figure 1.

Open in a new tab

Nebulin M160-170 binds to a protein with the same apparent molecular weight as desmin in a blot overlay assay of rat psoas muscle (lane 1). Blots were probed with GST-nebulin M160-170, followed by HRP-conjugated anti-GST antibodies. Nebulin M160-170 binds to a band of the same molecular weight as desmin (∼52 kDa). Western blots showed that the band detected in the blot overlay assay was desmin (lane 2). No signal was detected when blots were probed with GST (lane 3) or 2° antibodies alone (data not shown). Star marks desmin.

GFP-Desmin Coil IB Assembles at the Z-discs in Cardiac Myocytes

To determine which regions of desmin have the propensity to localize to the Z-discs, we generated several GFP-tagged desmin constructs and expressed them in primary cultures of rat cardiac myocytes. Six days after transfection, the myocytes were relaxed, fixed, and costained for α-actinin to mark the Z-discs (Figure 2). Rat cardiac myocytes were used because of our strong interest in understanding the molecular mechanisms contributing to certain forms of cardiomyopathies associated with mutations in desmin. The localization of each of the GFP-desmin fragments analyzed in this study is described (Supplemental Table S2).

Figure 2.

Figure 2.

Open in a new tab

Desmin coil IB assembles in clear striations at the Z-discs. The ability of various GFP-tagged desmin fragments to assemble at the Z-disc was assessed by their colocalization with α-actinin (red, in insets) in cardiomyocytes. In contrast to the diffuse distribution pattern observed in cells expressing GFP alone (a), GFP-desmin (b) was mostly found at Z-discs or aggregated in the cytoplasm (Figure 4B). GFP-head (c) was found aggregated near the Z-discs, whereas GFP-coil IIB (e) and GFP-tail (f) assembled in striations at the Z-discs and M-lines. GFP-coil IB assembled in a clear striated pattern at the Z-discs (d). Z-disc (arrow), M-line (asterisk), aggregates (arrowhead), short filaments (long arrows) and N (nucleus). Bar, 10 μm.

Full-length GFP-desmin localized to multiple subcellular sites: surrounding Z-discs, intercalated discs and cytoplasmic “aggregates” and as short filament-like structures interconnecting Z-discs (Figures 2b and 4Aa: data not shown for Z-sections that were obtained to determine the 3D nature of the filament arrangement around the Z-discs, and localization of intercalated discs; see below for discussion of aggregates), closely resembling the distribution pattern of endogenous desmin (see Figure 5a). GFP-desmin head was most often observed diffusely distributed, but it also was found localized in small aggregates or short filaments-like structures, often in close proximity to the Z-discs (Figure 2c). GFP-coil IB assembled into clear striations at the Z-discs (Figure 2d). GFP-coil IIB also localized at the Z-discs, although the stained striations were not as prominent as the GFP-coil IB striations (Figure 2e). We also observed that nearly all of the cells expressing GFP-coil IIB exhibited a bright nuclear-associated staining (Figure 2e, “N”). Furthermore, both GFP-coil IB and IIB localized to the M-lines, albeit less obviously than to the Z-discs and to intercalated discs (Figure 2, d and e, and data not shown). Finally, GFP-tail was consistently localized in a striated pattern at both the Z-discs and M-lines (Figure 2f), although the Z-disc localization was not as prominent as for GFP-desmin coil IB.

Figure 4.

Figure 4.

Open in a new tab

Expression of desmin E245D mutant results in loss of Z-disc localization and enhanced aggregate formation. (A) The assembly of wild-type GFP-desmin (a), GFP-desmin E245D (b), GFP-coil IB (c), and GFP-coil IB E245D (d). Arrows, striations; arrowheads, cytoplasmic aggregates. Bar, 10 μm. (B) Quantification of the Z-disc and aggregate assembly patterns of exogenous desmin fragments in cells expressing GFP-desmin, n = 418 and GFP-desmin E245D, n = 224. (C) Quantification of the Z-disc and aggregate assembly patterns in cells expressing GFP-coil IB, n = 765, and GFP-coil IB E245D, n = 713. Values ± SD were obtained from three to six independent experiments. Significance was determined using Student's t test comparing full-length desmin (or coil IB) with the mutants (*p < 0.05 and **p < 0.01).

Figure 5.

Figure 5.

Open in a new tab

Expression of desmin E245D resulted in perturbation of endogenous desmin localization and formation of desmin-rich aggregates. Merged deconvolved images of cells expressing GFP-desmin fragments (green) that were costained for endogenous desmin (red). Most cells expressing GFP alone showed striated endogenous desmin organization at the Z-discs (a). GFP-desmin and GFP-coil IB assembled in a striated pattern at the Z-discs colocalizing with striated endogenous desmin (arrows in b and d). Cells expressing GFP-desmin E245D (c), or GFP-coil IB E245D (e) contained a striking decrease in the distribution of endogenous desmin at the Z-discs as well as an increase in intracellular aggregates enriched for desmin (arrowheads in c and e). Bar, 10 μm.

GFP-coil IB was also more easily visualized at the Z-disc than any of the other GFP-desmin fragments when expressed in chick skeletal myocytes (Supplemental Figure S1b). Taken together, the prominent localization of desmin coil IB at the Z-disc in both heart and skeletal muscle is consistent with the interaction of this fragment with the C-terminal region of nebulin M160-170 and suggests that the coil IB region of desmin may directly mediate the interaction of desmin with nebulin at the Z-discs.

Desmin Coil IB Binds with High Affinity to Nebulin M160-164

Because C-terminal nebulin was described in a yeast two-hybrid assay to be important for the desmin–nebulin interaction (Bang et al., 2002), our blot overlay experiments suggested that endogenous desmin interacts with M160-170 (Figure 1), and our expression studies demonstrated that desmin coil IB prominently localizes to the Z-discs (Figure 2d), we hypothesized that this region of desmin contains the major nebulin-binding site. Thus, a His-tagged nebulin M160-170 fragment was immobilized onto a microtiter plate and incubated with biotinylated GST-tagged desmin coil IB. Our results showed that desmin coil IB interacted with nebulin M160-170, but not with α-actinin that served as a negative control (Figure 3A). To further define the desmin coil IB binding site within nebulin M160-170, several recombinant His-tagged nebulin fragments within this region were generated and tested for binding to desmin coil IB in ELISAs. We determined that desmin coil IB bound similarly to nebulin M160-164 as to nebulin M160-170, whereas significantly less signal was detected with the other C-terminal nebulin fragments encompassing modules 165-167, 169-171, 171-177, and 182-C terminus (Figure 3B). Note, nebulin M178-M181 was not tested because it could not be cloned; these modules are variably spliced (Millevoi et al., 1998). In conclusion, our results show that desmin coil IB appears to be sufficient for binding to nebulin M160-164, although we cannot rule out the possibility that other binding sites within full-length nebulin and desmin, not represented in these fragments, may contribute to the interaction.

Addition of the E245D Mutation in Desmin Coil IB Alters Its Binding to Nebulin In Vitro

To search for mutations that alter the desmin–nebulin interaction as a step toward deciphering the physiological significance of the interaction and to test our hypothesis that the nebulin–desmin interaction may contribute to the disease processes of certain myopathies, we searched the literature for missense disease-linked mutations located within the coil IB region of desmin (151–263), the region of desmin that interacts with nebulin (Figure 3A and Bang et al., 2002). We focused our attention on a dominantly inherited heterozygous Glu245Asp (E245D) mutation that causes muscle weakness and cardiomyopathy in humans (Vrabie et al., 2005), but was shown to not interfere with the filament assembly properties of desmin in vitro (Bar et al., 2005).

To estimate a Kd for the interaction of desmin coil IB (and the desmin coil IB E245D mutant) with nebulin M160-M164 (the smallest fragment that we found to bind coil IB), another solid-phase binding assay was utilized. Nebulin M160-M164 (40 nM) was immobilized on a microtiter plate and incubated with increasing concentrations of biotinylated desmin coil IB. Although there was some variability in the Kds measured in each individual experiment, our data consistently showed that the interaction of desmin coil IB E245D resulted in increased avidity (e.g., Bmax which indicates the binding capacity) to nebulin M160-164. For example, in the experiment shown, the interaction of nebulin M160-164 with both wild-type desmin coil IB and mutant desmin coil IB E245D was saturable, with a Kd of ∼35 nM and Kd of ∼30 nM, respectively (Figure 3C). However, the estimated Bmax was significantly higher for desmin coil IB E245D mutant compared with wild-type desmin coil IB (Bmax of ∼0.9 and ∼3.3, respectively, Figure 3C). Hence, our data show that a single mutation in desmin can alter its interaction with nebulin in vitro.

Expression of the Desmin E245D Mutation Results in Loss of Z-disc Localization and Promoted Cytoplasmic Aggregate Formation

To investigate whether the localization of full-length GFP-desmin or GFP-coil IB was perturbed when the E245D mutation was present, we assessed their distribution in cardiac myocytes. Although GFP alone was never found to be distributed in aggregates, wild-type GFP-desmin and to a lesser extent GFP-coil IB were observed to assemble into intracellular aggregates of varying sizes. Full-length GFP-desmin containing the E245D mutation, however, showed a significant decrease in Z-disc localization with a concomitant increase in its distribution as aggregates, compared with wild-type GFP-desmin (Figure 4A, a and b, and B; compare 62 ± 7% aggregate assembly for GFP-desmin with 88 ± 1% for GFP-desmin E245D). Similar trends were obtained when the distribution patterns of GFP-coil IB containing E245D mutation were observed (Figure 4A, c and d, and C; compare 43 ± 3% aggregate assembly for GFP-coil IB with 65 ± 4% for GFP-coil IB E245D). All together, these data revealed that expression of the E245D mutation in desmin decreased the Z-line distribution of this molecule. Remarkably, the intracellular accumulations observed are strikingly similar to the aggregates observed in cardiac and skeletal muscle biopsies from patients carrying the desmin E245D mutation and also frequently seen in patients suffering from all forms of desmin-related myopathies (e.g., Goldfarb et al., 2004; Vrabie et al., 2005).

To evaluate if expression of desmin containing the E245D mutation affected the functional link of desmin with nebulin in cardiac myocytes, we sought to determine whether the expression of E245D mutation would alter the organization of the endogenous desmin and/or nebulin filament networks. As described above, full-length GFP-desmin or GFP-coil IB containing E245D mutation assembled at a higher frequency in cytoplasmic aggregates of varying sizes than their unmutated counterparts (Figure 4). Furthermore, we observed that endogenous desmin often colocalized with aggregates of the GFP-fusion proteins (see arrowheads in Figure 5, c and e). To see if we could identify other components of the aggregates, we stained myocytes expressing the E245D desmin mutants for several sarcomeric proteins. Approximately 15% of the aggregates of GFP-coil IB E245D colocalized with endogenous C-terminal nebulin (see Supplemental Figure S3). No colocalization of the mutant coil IB fragments with N-terminal titin, α-actinin, myosin, actin, or myomesin in aggregates was detected (e.g., Supplemental Figure S4).

The Z-disc localization of endogenous desmin was strikingly decreased in cardiomyocytes expressing the E245D mutation (Figures 5 and 6, A and B). Specifically, quantitative analysis showed that expression of GFP-desmin E245D mutant led to a significant decrease (23 ± 6%) compared with cells expressing wild-type GFP-desmin (36 ± 6%) or GFP alone (41 ± 3%; Figure 6A). Interestingly, cells expressing GFP-coil IB increased the ability of endogenous desmin to assemble at the Z-discs compared with GFP alone (76 ± 5 and 52 ± 10%, respectively), whereas a significant reduction of Z-disc assembly was detected in cells expressing GFP-coil IB E245D (31 ± 4%; Figure 6B). In contrast, the endogenous desmin Z-disc distribution in cells expressing a random desmin mutation (K190A) within GFP-coil IB was not significantly different (data not shown).

Figure 6.

Figure 6.

Open in a new tab

Disruption of the Z-disc localization of endogenous desmin and nebulin upon expression of desmin E245D mutant. Cardiac myocytes expressing full-length GFP-desmin or GFP-coil IB with or without the E245D mutation were scored for the percentage of total cells with striated endogenous desmin (A and B) or C-terminal nebulin (C and D) at the Z-discs. (A) Z-disc associated endogenous desmin in cells expressing GFP, n = 203; GFP-desmin, n = 277; and GFP-desmin E245D, n = 255. (B) Quantification of Z-disc associated endogenous desmin in cells expressing GFP, n = 403; GFP-coil IB, n = 355; and GFP-coil IB E245D, n = 328. (C) Z-disc associated nebulin in cells expressing GFP, n = 298; GFP-desmin, n = 335; and GFP-desmin E245D, n = 187. (D) Z-disc–associated nebulin in cells expressing GFP n = 367; GFP-coil IB, n = 344; and GFP-coil IB E245D, n = 198. Values ± SD were obtained from two to three separate experiments. Student's t test was used to determine significance (* p < 0.05, ** p < 0.01) of GFP-desmin proteins, compared with GFP alone.

A possible effect on the Z-disc localization of C-terminal nebulin was also investigated using two polyclonal anti-nebulin–specific antibodies generated against modules 160-164 or modules 176-181 (data shown for anti-nebulin M176-181 antibodies: Figure 6, C and D, Supplemental Figure S3, A and B). Expression of GFP-desmin E245D resulted in a significant decrease in the localization of nebulin at the Z-disc (33 ± 6%) compared with cardiac myocytes expressing GFP-desmin (69 ± 10%) or GFP alone (57 ± 3%; Figure 6C). Likewise, the Z-disc localization of nebulin was significantly decreased in cells expressing GFP-coil IB E245D (40 ± 4%), compared with cells expressing GFP-coil IB (82 ± 6%) or GFP alone (59 ± 5%; Figure 6D). Consistent with these findings, expression GFP-coil IB E245D mutation also resulted in displacement of C-terminal nebulin Z-disc assembly in skeletal myocytes (Supplemental Figure S3, C and D). Taken together, our results indicate that expression of the desmin E245D mutation results in a significant loss of desmin and C-terminal nebulin Z-disc localization.

Expression of the Desmin Coil IB E245D Mutation Disrupts the Distribution of Myosin

To examine whether the integrity of other sarcomeric components was affected upon expression of the desmin E245D mutant protein, we stained cells expressing the mutant for the thick filament protein myosin, the M-band protein myomesin, titin, and the major Z-disc protein, α-actinin (Figure 7). Our results showed a disruption in the distribution of myosin in cells expressing GFP-coil IB E245D (Figure 7i), suggesting that the desmin mutant influences the distribution of the thick filaments. No obvious perturbations of myomesin, titin, and α-actinin were observed in cells expressing GFP-coil IB E245D, compared with cells expressing GFP alone or desmin GFP-coil IB (Figure 7, b–d, f–h, and j–l). See Supplemental Figure S1 for the corresponding images of the assembly of the desmin GFP-tagged fragments.

Figure 7.

Figure 7.

Open in a new tab

The distribution of myosin was perturbed in cells expressing GFP-coil-IB E245D, whereas the distribution of M-band myomesin, Z-disc titin, and α-actinin was unaffected. Myosin (a, e, and i), myomesin (b, f, and j), titin Z1-Z2 (c, g, and k), and α-actinin (d, h, and l) were properly organized in myocytes expressing GFP alone, GFP-coil IB, or GFP-coil IB E245D. Cells expressing GFP-coil IB E245D (i) had disrupted myosin organization; whereas no effect was observed for cells expressing GFP alone (a) or GFP-coil IB (e). See Supplemental Figure S4 for corresponding images of the GFP fusion protein localization. Bar, 10 μm.

Abnormal Actin Organization and Lengths in Myocytes Expressing Desmin E245D

Because nebulin is believed to have a role in thin filament architecture, we next sought to determine whether sarcomeric actin filaments were affected by the expression of the E245D desmin mutation. In this experiment, cardiac myocytes expressing the desmin mutants were stained for filamentous actin (F-actin) with Texas Red–conjugated phalloidin and costained with anti-α-actinin antibodies to assess Z-disc structure and spacing (Figure 8). The majority of cells expressing GFP alone or desmin GFP-coil IB contained well organized actin filaments with uniform staining along the lengths of the filaments with bright distinct staining at the Z-line and the absence of staining at the M-line (Figure 8, b and e). This staining pattern indicated uniformity of thin filament lengths at both the barbed and pointed ends. Interestingly, actin filament morphology was significantly altered in cells expressing the GFP-coil IB E245D. Notably, the phalloidin staining at the Z-discs in these mutant cells was often irregular (nonuniform) and broader (see pixel intensity plots in Figure 9B). Additionally, the phalloidin staining at the pointed ends of the filaments was often not well defined, and in many instances gaps at the M-line were not visible (Figures 8h and 9A). These observations suggest thin filament misalignment and/or alterations in filament lengths at both ends of the filaments (Figure 8, compare b and e to h). Remarkably, expression of GFP-coil IB E245D appeared to have no significant effect on the localization of the major Z-disc protein α-actinin, as well as Z-disc spacing (i.e., sarcomere lengths) as evidenced by a striated mature staining pattern for α-actinin in the identical myofibrils that displayed altered phalloidin staining (Figure 8, compare c and f to i). Similar alterations (i.e., lack of gaps in phalloidin staining at the M- line) in actin filament architecture were observed in nearly all myofibrils in chick skeletal myocytes expressing GFP-coil IB E245D (Supplemental Figure S6).

Figure 8.

Figure 8.

Open in a new tab

Abnormal actin filament architecture in cells expressing desmin coil IB E245D. Cardiac myocytes were transfected with GFP (a), GFP-coil IB (d), or GFP-coil IB E245D (g), fixed in relaxing buffer and stained with Texas Red-phalloidin and for α-actinin. Cells expressing GFP alone (b) and GFP-coil IB (e) displayed uniform striated phalloidin staining, with distinct, bright Z-disc staining and intact H-zones. In contrast, the phalloidin staining pattern was mostly nonuniform, lacking gaps at the M-lines in cells expressing GFP-coil IB E245D (h); nonstriated staining suggests thin filament elongation from their pointed ends. α-Actinin was unaffected by expression of GFP alone (c), wild-type desmin coil IB (f), or mutant desmin coil IB (i). Arrows mark the Z-lines, and asterisks mark the M-lines. Insets, 4× magnification. Bar, 10 μm.

To further analyze the actin filaments in cells expressing the desmin mutant fragment, we quantified the percentage of myofibrils containing “nonstriated” (no H-zones in staining) versus “striated” (discernable H-zones in staining) phalloidin staining (Figure 9A). In cells expressing GFP alone or unmutated GFP-coil IB, ∼13 and ∼18% of the myofibrils, respectively, showed nonstriated phalloidin staining. However, in cells expressing GFP-coil IB E245D, ∼60% of the myofibrils displayed nonstriated phalloidin staining (Figure 9A). Similarly, even though ∼49% of the myofibrils in cells expressing wild-type GFP-desmin displayed nonstriated phalloidin staining, expression of GFP-desmin E245D resulted in ∼87% of myofibrils lacking distinct gaps at the M-lines (Figure 9A, Supplemental Figure S5). Because the sarcomere lengths (distance between Z-discs) as assessed by staining for α-actinin were not altered (i.e., contracted; Figure 8), the presence of nonstriated phalloidin staining suggests that the thin filaments elongated from their pointed ends in cells expressing the E245D mutation and to a lesser extent in cells overexpressing wild-type full-length desmin.

We measured phalloidin staining (actin filament lengths) in individual sarcomeres containing discernable gaps at the M-line. This analysis revealed that the actin filaments that could be measured (from pointed end to pointed end) in cells expressing GFP-coil IB E245D were significantly shorter (1.38 ± 0.12 μm) than those found in myocytes expressing GFP (1.58 ± 0.13 μm), or GFP-coil IB (1.56 ± 0.11 μm). We also stained cells with anti-Tmod1 antibodies to mark the pointed ends of the actin filaments; this analysis also demonstrated a significant decrease in actin filament lengths in cells expressing desmin coil IB E245D (Supplemental Table 3). Lastly, the actin filament measurements obtained were analyzed as histograms with curves showing a Gaussian fit to the data, and by performing a two-sample Kolmogorov-Smirnov test (Figure 9C, Supplemental Figure S7). These analyses showed that the distribution of actin filament lengths in cells expressing GFP-coil IB E245D was shifted to the left, indicating reduced thin filament length distributions in cells with the desmin mutant compared with cells expressing desmin GFP-coil IB. Taken together, our data, reveal a novel function for the desmin IFs in regulating actin filament morphology and lengths and suggest that alterations in the desmin–nebulin interaction may contribute to certain DRM disease pathologies.

DISCUSSION

Thirty years ago, morphological evidence revealed that the thin filaments and desmin intermediate filaments comprise a “collar” surrounding the Z-discs of adjacent myofibrils (Granger and Lazarides, 1978). Biochemical support for this became available when it was shown that the giant actin filament–binding protein, nebulin, interacts with desmin at the periphery of the Z-discs (Bang et al., 2002). Nevertheless, the functional significance of the desmin–nebulin interaction in live myocytes and the potential role of this interaction in the context of muscle disease was previously unexplored.

In this study, we established a functional link between desmin and nebulin. Our results revealed that expression of a disease-associated point mutation at amino acid 245 (E245D) within the coil-IB region of desmin resulted in disruption of the endogenous desmin and nebulin Z-disc localization and promoted cytoplasmic aggregate formation, mirroring a major histological feature found in human DRM muscle biopsies. Furthermore, our data demonstrated that expression of the desmin E245D mutation dramatically altered actin filament lengths in myocytes. Our findings support a model in which the functional properties of nebulin associated with thin filament length regulation and/or maintenance appear to be, at least in part, modulated by its interaction with desmin. Thus, our data give novel insights into potential molecular consequences underlying DRMs.

Our targeting experiments showed that of all the GFP desmin regions tested, the coil IB region of desmin was consistently the most efficient at localizing to the Z-discs in cardiac and skeletal myocytes, in agreement with its interaction with C-terminal nebulin (Figure 2, Supplemental Figure S1). Because of this result as well as our observation that overexpression of the full-length molecule itself alters thin filament lengths and result in the accumulation of cytoplasmic aggregates, we utilized the wild-type coil IB fragment (which displays little, or no, detectable cellular perturbations) in most of our studies to focus on the interaction of desmin at the Z-disc. A possible explanation for the observed cellular effects on full-length desmin overexpression is that it likely altered the stoichiometric ratios of desmin to nebulin and/or it may have influenced interactions with other desmin-binding partners within myocytes. Interestingly, expression of the wild-type coil IB fragment alone was found to “recruit” endogenous desmin to the Z-disc, suggesting that it has the ability to self-associate with the full-length molecule. Additionally, we observed clear M-line targeting of both the coil IIB and tail regions of desmin. This raises the exciting possibility that desmin is indeed a component of the transverse M-line associated filaments connecting adjacent myofibrils observed in electron micrographs and rarely visualized in confocal studies (Price, 1984; O'Neill et al., 2002); these filaments are thought to contribute to the transverse stability of myofibrils during muscle contraction-relaxation cycles (e.g., Wang and Ramirez-Mitchell, 1983). Our data also suggest that the coil IIB and tail regions of desmin can target efficiently to the M-line in cultured cells because they are missing other regions of full-length desmin that contain interacting sites for non-M-line components that can mostly outcompete for M-line assembly in vivo. This interpretation is consistent with the near absence of GFP-desmin staining at the M-line in cultured myocytes. Based on the findings that the M-band protein EH-myomesin (skelemin), the major embryonic vertebrate heart isoform of myomesin, copurified with desmin and containing two “intermediate filament binding motifs,” it has been proposed that it may bind to desmin (Price, 1984; Steiner et al., 1999; Agarkova et al., 2000). The significance of the M-line localization of desmin and identification of its M-line interacting partner(s) remain to be fully elucidated.

Our biochemical evidence revealed that the coil IB region of desmin is important for its interaction with the C-terminal end of nebulin, because desmin coil IB binds with high affinity to nebulin M160-164 with a dissociation constant of Kd ∼35 nM (Figure 3C). Notably, the affinity of desmin coil-IB and nebulin M160-164 is similar to that reported for other nebulin-protein interactions (e.g., Tmod1 with nebulin, CapZ with nebulin: McElhinny et al., 2005; Pappas et al., 2008). The dissociation constants were not significantly different between wild-type coil IB versus mutant coil IB E245D in ELISAs. However, the addition of the desmin coil IB E245D mutation reproducibly increased its binding capacity for nebulin (the Bmax was >4-fold higher for the mutant: Figure 3C). In agreement with these findings, a competition assay revealed that the coil IB E245D mutant was a more efficient competitor than nonmutated coil IB for the binding of nonmutated coil IB to nebulin M160-164 (the IC50 was ∼13-fold lower for the mutant: data not shown). One possibility is that the increase in binding capacity of the mutant may alter its binding affinities to other sarcomeric components. Together, these experiments show that both wild-type and mutant desmin coil-IB bind with high affinity to nebulin M160-164 in vitro. Interestingly with respect to our data, based on binding data, Pappas et al., (2008) showed that nebulin modules M160-164 must be present at the edge of the Z-disc and not projecting out from the Z-disc as previously reported. Thus, we propose a model in which desmin maintains the alignment of the myofibrils by directly binding to a segment of nebulin M160-164 facing the periphery of the Z-disc.

Here, we report that overexpression of full-length desmin itself or the coil IB region of desmin containing the E245D mutation results in a marked removal of endogenous desmin from the Z-disc with a concomitant increase in endogenous desmin in intracellular aggregates containing the mutant proteins (Figures 4 and 5). These aggregates appeared similar to those reported in muscle biopsies taken from the patients with the desmin E245D mutation. These findings suggest that our approach of overexpressing the desmin E245D mutation in a model of primary cultures of myocytes is valid, as it resembles the genetic background found in humans, because this particular desmin mutation is heterozygous with an autosomal dominant pattern of inheritance (Vrabie et al., 2005). Interestingly, we also found that expression of the desmin E245D mutation resulted in striking alterations in actin filament organization and lengths as evidenced by 1) nonuniformity in phalloidin staining at the Z-disc, suggesting disorganization and/or filament elongation at their barbed ends; 2) significant increase in nonstriated (i.e., no visible gaps/absence of H-zones) phalloidin staining (without a decrease in sarcomere lengths), suggesting thin filament elongation from their pointed ends; and 3) thin filament shortening in cardiac sarcomeres that contained H-zones in phalloidin staining (Figures 8 and 9 and Supplemental Figure S6). Intriguingly, the alterations in thin filament lengths observed from expression of the E245D desmin mutant are reminiscent of the changes in actin filament lengths observed in nebulin −/− mouse models and in nebulin knockdown studies. For example, in nebulin −/− mouse models, thin filament lengths were significantly shorter (Bang et al., 2006; Witt et al., 2006), whereas nebulin knockdown by siRNA led to both nonstriated (longer) and shorter actin filaments (McElhinny et al., 2005; Pappas et al., 2008). These data, together with the fact that no major changes in the distribution of most (but not all, see below) sarcomeric components were observed as a result of expressing the desmin E245D mutant (Figure 7), are consistent with our prediction that nebulin's actin filament length regulation functions are compromised in cells expressing the desmin E245D. This adds a new functional role of desmin IFs in regulating (at least some of) nebulin's length-regulating properties.

The significance of our findings is particularly noteworthy because perturbations in actin organization and/or lengths would be expected to directly influence the relative range of lengths over which the sarcomeres can generate force effectively. Interestingly, desmin does not require nebulin to localize to the Z-disc as observed in nebulin −/− mice (Bang et al., 2006). However the interaction appears to be important for force transmission, based on the observations reported in nebulin −/− models in which the missing linkage of desmin to the Z-disc to nebulin results in lowered stress generation in the nebulin −/− mice (Bang et al., 2006; Witt et al., 2006). Thus, an interesting question arises as what is the mechanical relationship between desmin and nebulin and how nebulin might aid in transmission of force from the thin filament to the intermediate filament system via its association with desmin at the Z-discs. Consistent with this idea, the observed perturbations in the organization of myosin upon expression of DRM-associated desmin E245D mutation could be influenced by suboptimal interactions with actin filaments of altered lengths (Figure 7i). Another interpretation is that the observed changes in actin filament morphology may result from a nebulin-based misregulation of actin filament–capping protein dynamics (Tmod1 and/or CapZ) at the ends of the thin filaments. A conundrum is that although there appears to be much lower levels of nebulin in cardiac myofibrils, in comparison to skeletal myofibrils (e.g., Kazmierski et al., 2003; Bang et al., 2006), we observed similar thin filament perturbations as a result of the expression of desmin E245D mutant. We, as well as others, speculate that many sarcomeric components contribute to the regulation of thin filament lengths, which may also be affected by the expression of the desmin E245D mutant. This prediction is supported by the fact that invertebrate organisms such as Caenorhabditis elegans and Drosophila (do not have nebulin or desmin orthologues), yet still have uniform thin filament lengths. It is clear that more studies will be required to resolve how the levels of nebulin influence its thin filament length–regulating properties.

In conclusion, our results provide new evidence that the desmin–nebulin interaction plays an important role in muscle physiology and that variations in actin filament lengths may directly contribute to pathogenic molecular consequences underlying certain desmin-related myopathies.

Supplementary Material

[Supplemental Materials]
E08-07-0753_index.html (790B, html)

ACKNOWLEDGMENTS

We thank Joseph Bahl, Alexandria Lau, Verena Mentrup, and Samantha Whitman for preparing the rat cardiac and chick skeletal myocyte cultures; Christopher Pappas for providing His-tagged nebulin constructs; Lucas Macri for computer assistance and statistical analysis; Takehiro Tsukada and Paul St John for help in ELISA analysis; and Takehiro Tsukada, Anke Zieseniss, Christopher Pappas, and Chinedu Nworu for critical review of the manuscript. This work was funded by a National Institutes of Health (NIH) Minority Supplement (HL57461) to G.M.C., NIH Grants HL57461 and HL083146 and American Heath Association Grant 0655637 to C.C.G., and the Undergraduate Biology Research Program (HHMI to University of Arizona) to S.N.H.

Abbreviations used:

IF

intermediate filaments

DRM

desmin-related myopathies.

Footnotes

This article was published online ahead of print in MBC in Press (http://www.molbiolcell.org/cgi/doi/10.1091/mbc.E08-07-0753) on November 12, 2008.

REFERENCES

  1. Agarkova I., Auerbach D., Ehler E., Perriard J. C. A novel marker for vertebrate embryonic heart, the EH-myomesin isoform. J. Biol. Chem. 2000;275:10256–10264. doi: 10.1074/jbc.275.14.10256. [DOI] [PubMed] [Google Scholar]
  2. Bang M. L., Gregorio C., Labeit S. Molecular dissection of the interaction of desmin with the C-terminal region of nebulin. J. Struct. Biol. 2002;137:119–127. doi: 10.1006/jsbi.2002.4457. [DOI] [PubMed] [Google Scholar]
  3. Bang M. L., Li X., Littlefield R., Bremner S., Thor A., Knowlton K. U., Lieber R. L., Chen J. Nebulin-deficient mice exhibit shorter thin filament lengths and reduced contractile function in skeletal muscle. J. Cell Biol. 2006;173:905–916. doi: 10.1083/jcb.200603119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bar H., et al. Conspicuous involvement of desmin tail mutations in diverse cardiac and skeletal myopathies. Hum. Mutat. 2007;28:374–386. doi: 10.1002/humu.20459. [DOI] [PubMed] [Google Scholar]
  5. Bar H., Mucke N., Kostareva A., Sjoberg G., Aebi U., Herrmann H. Severe muscle disease-causing desmin mutations interfere with in vitro filament assembly at distinct stages. Proc. Natl. Acad. Sci. USA. 2005;102:15099–15104. doi: 10.1073/pnas.0504568102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Clark K. A., McElhinny A. S., Beckerle M. C., Gregorio C. C. Striated muscle cytoarchitecture: an intricate web of form and function. Annu. Rev. Cell Dev. Biol. 2002;18:637–706. doi: 10.1146/annurev.cellbio.18.012502.105840. [DOI] [PubMed] [Google Scholar]
  7. Fock U., Hinssen H. Nebulin is a thin filament protein of the cardiac muscle of the agnathans. J. Muscle Res. Cell Motil. 2002;23:205–213. doi: 10.1023/a:1020909902462. [DOI] [PubMed] [Google Scholar]
  8. Fowler V. M., McKeown C. R., Fischer R. S. Nebulin: does it measure up as a ruler? Curr. Biol. 2006;16:R18–20. doi: 10.1016/j.cub.2005.12.003. [DOI] [PubMed] [Google Scholar]
  9. Geisler N., Weber K. The amino acid sequence of chicken muscle desmin provides a common structural model for intermediate filament proteins. EMBO J. 1982;1:1649–1656. doi: 10.1002/j.1460-2075.1982.tb01368.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Goldfarb L. G., Vicart P., Goebel H. H., Dalakas M. C. Desmin myopathy. Brain. 2004;127:723–734. doi: 10.1093/brain/awh033. [DOI] [PubMed] [Google Scholar]
  11. Granger B. L., Lazarides E. The existence of an insoluble Z disc scaffold in chicken skeletal muscle. Cell. 1978;15:1253–1268. doi: 10.1016/0092-8674(78)90051-x. [DOI] [PubMed] [Google Scholar]
  12. Gustafson T. A., Bahl J. J., Markham B. E., Roeske W. R., Morkin E. Hormonal regulation of myosin heavy chain and alpha-actin gene expression in cultured fetal rat heart myocytes. J. Biol. Chem. 1987;262:13316–13322. [PubMed] [Google Scholar]
  13. Herrmann H., Aebi U. Intermediate filaments: molecular structure, assembly mechanism, and integration into functionally distinct intracellular scaffolds. Annu. Rev. Biochem. 2004;73:749–789. doi: 10.1146/annurev.biochem.73.011303.073823. [DOI] [PubMed] [Google Scholar]
  14. Horowits R. Nebulin regulation of actin filament lengths: new angles. Trends Cell Biol. 2006;16:121–124. doi: 10.1016/j.tcb.2006.01.003. [DOI] [PubMed] [Google Scholar]
  15. Kazmierski S. T., Antin P. B., Witt C. C., Huebner N., McElhinny A. S., Labeit S., Gregorio C. C. The complete mouse nebulin gene sequence and the identification of cardiac nebulin. J. Mol. Biol. 2003;328:835–846. doi: 10.1016/s0022-2836(03)00348-6. [DOI] [PubMed] [Google Scholar]
  16. Labeit S., Kolmerer B. The complete primary structure of human nebulin and its correlation to muscle structure. J. Mol. Biol. 1995;248:308–315. doi: 10.1016/s0022-2836(95)80052-2. [DOI] [PubMed] [Google Scholar]
  17. Lazarides E. Comparison of the structure, distribution and possible function on desmin (100 A) filaments in various types of muscle and nonmuscle cells. Birth Defects Orig. Artic. Ser. 1978;14:41–63. [PubMed] [Google Scholar]
  18. Lazarides E., Hubbard B. D. Immunological characterization of the subunit of the 100 A filaments from muscle cells. Proc. Natl. Acad. Sci. USA. 1976;73:4344–4348. doi: 10.1073/pnas.73.12.4344. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Li Z., Mericskay M., Agbulut O., Butler-Browne G., Carlsson L., Thornell L. E., Babinet C., Paulin D. Desmin is essential for the tensile strength and integrity of myofibrils but not for myogenic commitment, differentiation, and fusion of skeletal muscle. J. Cell Biol. 1997;139:129–144. doi: 10.1083/jcb.139.1.129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. McElhinny A. S., Kolmerer B., Fowler V. M., Labeit S., Gregorio C. C. The N-terminal end of nebulin interacts with tropomodulin at the pointed ends of thin filaments. J. Biol. Chem. 2001;276:583–592. doi: 10.1074/jbc.M005693200. [DOI] [PubMed] [Google Scholar]
  21. McElhinny A. S., Schwach C., Valichnac M., Mount-Patrick S., Gregorio C. C. Nebulin regulates the assembly and lengths of the thin filaments in striated muscle. J. Cell Biol. 2005;170:947–957. doi: 10.1083/jcb.200502158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Millevoi S., Trombitas K., Kolmerer B., Kostin S., Schaper J., Pelin K., Granzier H., Labeit S. Characterization of nebulette and nebulin and emerging concepts of their roles for vertebrate Z-discs. J. Mol. Biol. 1998;282:111–123. doi: 10.1006/jmbi.1998.1999. [DOI] [PubMed] [Google Scholar]
  23. Milner D. J., Weitzer G., Tran D., Bradley A., Capetanaki Y. Disruption of muscle architecture and myocardial degeneration in mice lacking desmin. J. Cell Biol. 1996;134:1255–1270. doi: 10.1083/jcb.134.5.1255. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. O'Neill A., Williams M. W., Resneck W. G., Milner D. J., Capetanaki Y., Bloch R. J. Sarcolemmal organization in skeletal muscle lacking desmin: evidence for cytokeratins associated with the membrane skeleton at costameres. Mol. Biol. Cell. 2002;13:2347–2359. doi: 10.1091/mbc.01-12-0576. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Ojima K., Lin Z. X., Zhang Z.Q., Hijikata T., Holtzer S., Labeit S., Sweeney H. L., Holtzer H. Initiation and maturation of I-Z-I bodies in the growth tips of transfected myotubes. J. Cell Sci. 1999;112(Pt 22):4101–4112. doi: 10.1242/jcs.112.22.4101. [DOI] [PubMed] [Google Scholar]
  26. Pappas C. T., Bhattacharya N., Cooper J. A., Gregorio C. C. Nebulin interacts with CapZ and regulates thin filament architecture within the Z-disc. Mol. Biol. Cell. 2008;19((5)):1837–1847. doi: 10.1091/mbc.E07-07-0690. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Pfuhl M., Winder S. J., Castiglione Morelli M. A., Labeit S., Pastore A. Correlation between conformational and binding properties of nebulin repeats. J. Mol. Biol. 1996;257:367–384. doi: 10.1006/jmbi.1996.0169. [DOI] [PubMed] [Google Scholar]
  28. Price M. G. Molecular analysis of intermediate filament cytoskeleton—a putative load-bearing structure. Am. J. Physiol. 1984;246:H566–H572. doi: 10.1152/ajpheart.1984.246.4.H566. [DOI] [PubMed] [Google Scholar]
  29. Shah S. B., Davis J., Weisleder N., Kostavassili I., McCulloch A. D., Ralston E., Capetanaki Y., Lieber R. L. Structural and functional roles of desmin in mouse skeletal muscle during passive deformation. Biophys. J. 2004;86:2993–3008. doi: 10.1016/S0006-3495(04)74349-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Steiner F., Weber K., Furst D. O. M band proteins myomesin and skelemin are encoded by the same gene: analysis of its organization and expression. Genomics. 1999;56:78–89. doi: 10.1006/geno.1998.5682. [DOI] [PubMed] [Google Scholar]
  31. Taylor M. R., et al. Prevalence of desmin mutations in dilated cardiomyopathy. Circulation. 2007;115:1244–1251. doi: 10.1161/CIRCULATIONAHA.106.646778. [DOI] [PubMed] [Google Scholar]
  32. Thornell L. E., Eriksson A., Johansson B., Kjorell U., Franke W. W., Virtanen I., Lehto V. P. Intermediate filament and associated proteins in heart Purkinje fibers: a membrane-myofibril anchored cytoskeletal system. Ann. NY Acad. Sci. 1985;455:213–240. doi: 10.1111/j.1749-6632.1985.tb50414.x. [DOI] [PubMed] [Google Scholar]
  33. Tokuyasu K. T., Dutton A. H., Singer S. J. Immunoelectron microscopic studies of desmin (skeletin) localization and intermediate filament organization in chicken cardiac muscle. J. Cell Biol. 1983;96:1736–1742. doi: 10.1083/jcb.96.6.1736. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Vrabie A., Goldfarb L. G., Shatunov A., Nagele A., Fritz P., Kaczmarek I., Goebel H. H. The enlarging spectrum of desminopathies: new morphological findings, eastward geographic spread, novel exon 3 desmin mutation. Acta Neuropathol. 2005;109:411–417. doi: 10.1007/s00401-005-0980-1. [DOI] [PubMed] [Google Scholar]
  35. Wallgren-Pettersson C., Lehtokari V. L., Kalimo H., Paetau A., Nuutinen E., Hackman P., Sewry C., Pelin K., Udd B. Distal myopathy caused by homozygous missense mutations in the nebulin gene. Brain. 2007;130:1465–1476. doi: 10.1093/brain/awm094. [DOI] [PubMed] [Google Scholar]
  36. Wang K., Knipfer M., Huang Q. Q., van Heerden A., Hsu L. C., Gutierrez G., Quian X. L., Stedman H. Human skeletal muscle nebulin sequence encodes a blueprint for thin filament architecture. Sequence motifs and affinity profiles of tandem repeats and terminal SH3. J. Biol. Chem. 1996;271:4304–4314. doi: 10.1074/jbc.271.8.4304. [DOI] [PubMed] [Google Scholar]
  37. Wang K., Ramirez-Mitchell R. A network of transverse and longitudinal intermediate filaments is associated with sarcomeres of adult vertebrate skeletal muscle. J. Cell Biol. 1983;96:562–570. doi: 10.1083/jcb.96.2.562. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Wang K., Wright J. Architecture of the sarcomere matrix of skeletal muscle: immunoelectron microscopic evidence that suggests a set of parallel inextensible nebulin filaments anchored at the Z line. J. Cell Biol. 1988;107:2199–2212. doi: 10.1083/jcb.107.6.2199. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Weisleder N., Taffet G. E., Capetanaki Y. Bcl-2 overexpression corrects mitochondrial defects and ameliorates inherited desmin null cardiomyopathy. Proc. Natl. Acad. Sci. USA. 2004;101:769–774. doi: 10.1073/pnas.0303202101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Witt C. C., Burkart C., Labeit D., McNabb M., Wu Y., Granzier H., Labeit S. Nebulin regulates thin filament length, contractility, and Z-disk structure in vivo. EMBO J. 2006;25:3843–3855. doi: 10.1038/sj.emboj.7601242. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

[Supplemental Materials]
E08-07-0753_index.html (790B, html)
E08-07-0753_1.pdf (916.6KB, pdf)

Articles from Molecular Biology of the Cell are provided here courtesy of American Society for Cell Biology

RESOURCES