FHL1B Interacts with Lamin A/C and Emerin at the Nuclear Lamina and is Misregulated in Emery-Dreifuss Muscular Dystrophy

Abstract

Background: Emery-Dreifuss muscular dystrophy (EDMD) is associated with mutations in EMD and LMNA genes, encoding for the nuclear envelope proteins emerin and lamin A/C, indicating that EDMD is a nuclear envelope disease. We recently reported mutations in FHL1 gene in X-linked EDMD. FHL1 encodes FHL1A, and the two minor isoforms FHL1B and FHL1C. So far, none have been described at the nuclear envelope.

Objective: To gain insight into the pathophysiology of EDMD, we focused our attention on the poorly characterized FHL1B isoform.

Methods: The amount and the localisation of FHL1B were evaluated in control and diseased human primary myoblasts using immunofluorescence and western blotting.

Results: We found that in addition to a cytoplasmic localization, this isoform strongly accumulated at the nuclear envelope of primary human myoblasts, like but independently of lamin A/C and emerin. During myoblast differentiation, we observed a major reduction of FHL1B protein expression, especially in the nucleus. Interestingly, we found elevated FHL1B expression level in myoblasts from an FHL1-related EDMD patient where the FHL1 mutation only affects FHL1A, as well as in myoblasts from an LMNA-related EDMD patient.

Conclusions: Altogether, the specific localization of FHL1B and its modulation in disease-patient’s myoblasts confirmed FHL1-related EDMD as a nuclear envelope disease.

Keywords

FHL1B emerin lamin A/C nuclear envelope Emery-Dreifuss muscular dystrophy

Abbreviations

four-and-a-half LIM domain 1

Emery-Dreifuss muscular dystrophy

nuclear envelope

inner nuclear membrane

outer nuclear membrane

proximity ligation assay.

INTRODUCTION

Emery-Dreifuss muscular dystrophy (EDMD) is characterized by early joint contractures, a slowly progressive muscular wasting and weakness of scapulo-peroneal distribution, and the development of cardiac disease in adults with a high risk of sudden death [1]. EMD and LMNA are responsible for X-linked and autosomal forms of the disorder, respectively [2, 3]. EMD encodes emerin, a type II integral membrane protein that is anchored with its C-terminal tail to the inner nuclear membrane (INM) [4], and LMNA encodes lamins A and C (hereafter referred to as lamin A/C), belonging to type V intermediate filament proteins, a major component of the nuclear lamina that lies adjacent to the INM [5], leading to the idea that EDMD is due to mutations in nuclear envelope (NE) proteins. However, the discovery that X-linked EDMD can be caused by mutations in four-and-a-half LIM domain 1 (FHL1) gene, encoding different FHL1 protein isoforms, raises the question of how these non-NE proteins can be linked to EDMD [6].

FHL1 protein isoforms are defined primarily by the structural motif of tandemly repeated LIM (LIN-11, Isl-1, Mec-3) domains with the consensus sequence (C-X₂-CX-_16 - 23-C/H-X_2 - 4-C/H/E)-X₂-(C-X₂-CX_16 - 21-C/H-X_1 - 3-C/H/D) [7]. FHL1A (SLIM/SLIM1/KyoT1) is the most abundant protein isoform in skeletal muscle in comparison to FHL1B (SLIMMER/KyoT3) and FHL1C (KyoT2). Recently, an additional isoform has been identified in human skeletal muscles [8] and in the heart of patients with various cardiac conditions. This isoform is highly similar to FHL1A with 16 additional amino acids in the N-terminal part [9]. Key roles in integrin-mediated cytoskeletal re-arrangement, myogenic differentiation, sarcomere assembly, and regulation of the skeletal muscle mass, development and homeostasis have been ascribed to FHL1A [8 , 10–13]. Compared to FHL1A, very few studies have focused on FHL1B or FHL1C. These minor isoforms are generated by the inclusion of exon 7 or the exclusion of exon 6, respectively, leading to frameshifts generating completely different C-terminal domains and stop codons. FHL1B has been found to localize both in the cytoplasm and the nucleus of murine myoblasts and to shuttle between the two compartments during cell cycle progression, while being mainly cytoplasmic in differentiated myotubes [14, 15]. In mouse skeletal muscle, localization of FHL1B is restricted to the Z-disc region and to the sarcolemma [14]. Furthermore, FHL1B has been reported to delay skeletal myoblast apoptosis via interaction with the pro-apoptotic protein Siva-1 and is likely to be implicated in cell cycle regulation via interaction with the β catalytic subunit of type 2A protein phosphatase (PP2A_Cβ) [14, 15].Through its inhibitory interaction with recombination signal binding protein for immunoglobulin kappa J region (RBP-Jκ), a downstream effector of the Notch signalling pathway, FHL1B (as FHL1C) is also potentially involved in MyoD-dependent myogenesis [16 –18].

To date, 17 different FHL1 mutations (missense, nonsense, splice site, deletion and insertion) have been identified in EDMD that affect the distal exons of FHL1 and leave the first one and a half LIM domain preserved [19 –22]. FHL1A is always affected and protein levels are reduced, while FHL1B and/or FHL1C can be spared by the mutations, with unknown consequences on protein levels [7].

We sought to further improve our understanding of the role played by FHL1B in human skeletal muscle. We show that beside its cytoplasmic and nucleoplasmic localization, FHL1B is also a NE protein like lamin A/C and emerin. By studying primary human myoblasts derived from patients with mutations in FHL1 and other genes encoding NE proteins, we reveal abnormal FHL1B expression in two patients, which could contribute to the development of the disease.

MATERIALS AND METHODS

Cells and biopsies

Muscle biopsies from healthy controls and patients harbouring different gene mutations encoding FHL1 and NE proteins, as indicated in Table S1, were taken with the informed consent of the donors and with approval of the local ethical boards (AC-2013-1868). Muscle biopsies were frozen in liquid nitrogen and cryosectioned for immunohistochemical techniques. Primary human myoblasts were isolated from muscle biopsies as described in [23]. All the procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national).

Control and patient myoblasts were used if possible at comparable cell divisions stage. Only human myoblasts that underwent no more than 20 cell divisions were studied. All myoblast cultures used in this study had a myogenic purity greater than 90%, as assessed by desmin expression. Cell selection using anti-CD56 antibody with MACS Microbeads (Miletnyi Biotec, Bergisch Gladbach, Germany) was used when needed to reach such percentage.

Primary human myoblasts were cultured in basal proliferation medium consisting of Medium 199 and high-glucose DMEM in a 1:4 ratio, supplemented with 50 μg/ml gentamicin and 20% fetal bovine serum (Life Technologies, Saint Aubin, France) supplemented with 0.2 μg/ml dexamethasone (Sigma-Aldrich, Saint Quentin-Fallavier, France), 5 ng/ml EGF, 0.5 ng/ml bFGF, 25 μg/ml fetuin and 5 μg/ml insulin (Life Technologies, Saint Aubin, France) for the patient myoblasts with FHL1 mutation, in order to improve the low proliferative capacity Differentiation of confluent myoblasts into myotubes was induced by replacing proliferation medium with differentiation medium consisting of high-glucose DMEM supplemented with 50 μg/ml gentamicin and 10 μg/ml insulin (Life Technologies, Saint Aubin, France). All primary cell cultures were incubated at 37°C in a humid atmosphere of 5% CO₂.

siRNA and leptomycin B treatment

Lamin A/C- and lacZ-specific siRNA duplexeswere obtained from Eurogentec (Paris, France). Thehuman lamin A/C siRNA sense sequence was asfollows: 5’-CUGGACUUCCAGAAGAACA(dTdT)-3’. The lacZ siRNA sense sequence was: 5’-GACUACACAAAUCAGCGAU(dTdT)-3’ and was used as a negative control. Myoblasts grown on glass coverslips were transfected at time point 0 h and 48 h using 100 nM siRNA and HiPerfect transfection reagent (Qiagen, Venlo, The Netherlands) according to the manufacturer’s instructions. Cells were fixed 96 h after initial transfection with 4% PFA before processing for immunostaining.

Human primary myoblasts were treated during 24 h with 20 mg/ml leptomycin B (Sigma-Aldrich, Saint Quentin-Fallavier, France) diluted in proliferation medium, while 3 day-old myotubes were treated daily with 20 mg/ml leptomycin B diluted in differentiation medium during 3 days before processed for immunofluorescence.

Immunofluorescence, immunohistochemistry and PLA

Myoblasts grown on glass-coverslips and myotubes on 1% gelatin-coated glass coverslips were processed for immunostaining. Cells were fixed in 4% PFA, permeabilized with either 0.5% Triton X-100 or 0.001% digitonin and blocked in 5% BSA. Cryosections (10-μm thick) of muscle biopsies were fixed in cold acetone and blocked in 3% BSA. Primary and secondary antibodies were both diluted in blocking solution. The coverslips were mounted on slides with Vectashield mounting medium containing DAPI (Vector Labs, Burlingame, CA) for cells, or Mowiol for sections.

Following primary antibodies were used: goat polyclonal anti-FHL1B (ab26072), rabbit monoclonal anti-muscle-actin (ab46805), and rabbit polyclonal anti-desmin (ab15200), all from, Abcam, Cambridge, UK; mouse monoclonal anti-lamin A/C (NCL-clone 636) and anti-emerin (NCL-clone 4G5), both from Novocastra, Newcastle, UK; mouse monoclonal anti-lamin A/C (MANLAC1) and anti-nesprin-1α (MANNES1E), generously provided by Glenn Morris, UK [24] mouse monoclonal anti-Pax7 (DSHB). Secondary antibodies were: Alexa-Fluor-488-conjugated chicken anti-mouse and chicken anti-rabbit, Alexa-Fluor-568-conjugated goat anti-mouse, goat anti-rabbit and donkey-anti goat (Life Technologies, Saint Aubin, France). Phalloidin-FluoProbes 488 (Interchim, Montluçon, France) was used to label F-actin. Alexa-Fluor-647-conjugated donkey anti-Goat antibody was used to detect FHL1B on human muscle sections.

Proximity ligation assay (PLA) was performed on myoblasts cultured on glass-coverslips using the Duolink i n situ red starter kit mouse/goat (Sigma-Aldrich, Saint Quentin-Fallavier, France). Myoblasts were fixed and processed for immunostaining as indicated. After the incubation step with primary antibodies, further steps were performed according to the manufacturer’s protocol.

Protein analysis

Proteins were extracted from pelleted myoblasts in total protein extraction buffer (50 mM Tris-HCl, pH 7.5; 2% SDS; 250 mM Sucrose; 75 mM Urea; 1 mM DTT) and with protease inhibitors (25 μg/ml Aprotinin; 10 μg/ml Leupeptin; 1 mM AEBSF and 2 mM Na₃VO₄). Quantification was performed using BCA protein assay kit (Life Technologies, Saint Aubin, France). Twenty μg for myoblast differentiation studies and 30 μg for all other studies were separated by electrophoresis on 10% SDS-PAGE, and electrophoretically transferred to 0.45 μm nitrocellulose membranes (Life Technologies, Saint Aubin, France). Membranes were blocked with 5% skimmed milk in PBS-0.05% Tween20 and hybridized with the following primary antibodies in blocking solution: mouse monoclonal anti-pan FHL1 (ab58067) and goat polyclonal anti-FHL1B (ab26072), both from Abcam, Cambridge, UK; mouse monoclonal anti-α-actinin (EA-53) and anti-vinculin (V9131), both from Sigma-Aldrich, Saint Quentin-Fallavier, France; rabbit polyclonal anti-lamin A/C (H110, Santa Cruz Biotechnology, Santa Cruz, CA, USA) and mouse monoclonal anti-NCL-emerin (NCL-clone 4G5, Novocastra, Newcastle, UK). Membranes were incubated with secondary antibodies coupled to horseradish peroxidase (Jackson ImmunoResearch, West Grove, PA, USA). Immunoblots were visualized with Immobilon Western Chemiluminescent HRP Substrate (Millipore, Molsheim, France) on aG-Box system with GeneSnap software (Ozyme, Saint Quentin, France).

Microscopy and image processing

Confocal images were taken with confocal laser scanning microscopy using a Leica SP2 inverted microscope (Leica Microsystems, Wetzlar, Germany) or Olympus FV-1000 microscope. Serial confocal sections were sequentially collected from the top to the bottom at a step size of 0.2 μm between each frame. Z-projection images were generated by the command “Median” in the ImageJ software (National Institutes of Health, Bethesda, Maryland, USA). Presented single confocal sections are sections of the middle of the nucleus.

RESULTS

FHL1B is a NE protein that interacts with lamin A/C and emerin

To define the subcellular distribution of FHL1B in primary human myoblasts, confocal immunofluorescent staining of FHL1B was performed. FHL1B protein displayed both a nuclear and a cytoplasmic localization (Fig. 1A). Furthermore, confocal section images showed a clear and strong nuclear rim labelling (Fig. 1B). To further characterize the localization of FHL1B at the NE, we treated myoblasts with digitonin (Fig. 1C). Unlike triton treatment that permeabilises all membranes, digitonin treatment allowed the selective permeabilisation of the plasma membrane, leaving the two nuclear membranes intact. Desmin staining confirmed permeabilisation of the plasma membrane, while lamin A/C was undetectable at the INM, indicating unaltered nuclear membranes. The faint emerin staining observed under these conditions corresponds to the small fraction of emerin at the outer nuclear membrane (ONM) and the connected endoplasmic reticulum [25]. Interestingly, no NE staining appeared for FHL1B. This result indicates that the nuclear rim staining observed for FHL1B corresponds to a localization of FHL1B at the INM.

This very intriguing observation raised the question of whether FHL1B co-localizes with other INM proteins. Double-labelling experiments showed co-localization of FHL1B with both lamin A/C and emerin (Fig. 2A). Analysis of the fluorescent signals of both fluorochromes used to visualize FHL1B-lamin A/C or FHL1B-emerin, shows a clear overlap at the NE, revealing co-localization of the three proteins (Fig. 2B). Using an in situ proximity ligation assay (PLA), which enables the detection of proteins in close proximity, compatible with interaction [26], we were able to detect a strong PLA signal when we used proximity probes against FHL1B and lamin A/C antibodies and an intermediate PLA signal using proximity probes against FHL1B and emerin antibodies (Fig. 2C). PLA signals were not restricted to the NE as some dots were also detectable in the nucleoplasm. As a negative control, we used myoblasts of a patient with a mutation in the EMD gene (EMD p.Gln219Argfs*20), resulting in the complete loss of emerin protein expression (Fig. 2D). In these cells, the PLA signal was abolished when we used proximity probes against FHL1B and emerin antibodies (Fig. 2C), demonstrating the specificity of the assay. We conclude that in myoblasts, the nuclear pool of FHL1B lies in close proximity with lamin A/C and emerin at the nuclear membrane and partially in the nucleoplasm.

FHL1B does not rely on lamin A/C, emerin or nesprin-1α for its NE localization

To determine whether FHL1B requires lamin A/C for its NE localization, we performed siRNA knockdown of lamin A/C in primary human myoblasts (Fig. 3A). In non-treated myoblasts (NT) and in myoblasts treated with control lacZ siRNA, lamin A/C staining was clearly visible, whereas it was reduced or abolished in knockdown myoblasts treated with siRNA specific to lamin A/C (Fig. 3A, arrow). Depletion of lamin A/C did not affect the localization of FHL1B to the NE. Thus, lamin A/C is not required for FHL1B localization to the NE.

We further investigated if other NE proteins were involved in the localization of FHL1B to the NE. Two patient myoblasts were analysed: myoblasts from the patient harbouring the mutation in EMD (EMD p.Gln219Argfs*20) leading to loss of emerin expression and causing EDMD (Fig. 2D) and myoblasts from a patient with a homozygous stop codon mutation in SYNE1 (SYNE1 p.Glu7854*) leading to absence of nesprin-1α expression and causing congenital muscular dystrophy (CMD) [27] (Fig. 3B and C, respectively). In the two patient’s myoblasts, FHL1B was correctly localized to the NE (Fig. 3B and C). From these results, we conclude that the NE localization of FHL1B is also independent of emerin or nesprin-1α.

FHL1B protein is downregulated during myoblast differentiation

We then studied the expression levels of FHL1B in proliferating myoblasts and in differentiating myotubes. While FHL1B expression was fairly unchanged during the first days of differentiation (Supplemental Figure 2), we found a marked reduction of FHL1B protein levels after 6 days of differentiation compared with proliferating myoblasts (Fig. 4A). Using immunofluorescence, we found that this drop mainly concerned the nuclear pool of FHL1B with almost no staining observed in the nucleoplasm and only a faint staining at the nuclear envelope compared with myoblasts. Moreover, co-staining with Phalloidin reveals a partial co-localization of FHL1B with actin cytoskeleton in myotubes (Fig. 4B). We performed immunostaining on adult muscle sections and found that FHL1B was mainly staining the cytoplasm and also accumulate in sub-sarcolemmal position (Supplemental Figure 1).

As the Leucine-rich nuclear export signal (NES) of FHL1B was previously shown to be important for the nuclear exclusion of FHL1B, we tested if the reduced FHL1B expression level in the nucleus is due to its nuclear exclusion via the exportin 1/Chromosome Region Maintenance 1 (CRM1)-dependent pathway by treating cells with the exportin 1-specific inhibitor leptomycin B (LMB) [28]. LMB treated myoblasts present with increased p53 in the nucleus, revealing the efficient inhibition of exportin 1 (Fig. 4C). However, such treatment was inefficient on FHL1B as both cytoplasmic and nuclear levels were similar in treated and untreated myoblasts. We thought that FHL1B might be only exported from the nucleus during myoblast differentiation. Hence, we treated differentiating myoblasts with LMB and analysed FHL1B localization in 6 day-old myotubes. Here again, we did not observed any accumulation of FHL1B in the nuclei, leading to the conclusion that the decrease of nuclear FHL1B in myotubes is not due to an exportin 1-dependent nuclear exclusion.

FHL1B expression levels are affected in EDMD patient myoblasts

As FHL1-related EDMD patients do not necessarily harbour mutations that affect the FHL1B isoform [6, 7], we were wondering whether FHL1B is involved in the pathological mechanisms causing EDMD. Hence, we checked for expression and localization of FHL1B, along with emerin and lamin A/C, in myoblasts of a patient with an EDMD-associated FHL1 mutation (FHL1 p.*230+50Gluext*53, corresponding to patient F997, previously reported as p.X281Glu (in [6]). This mutation is affecting the stop codon of FHL1A and is thus localized in the 3’ untranslated region of FHL1B and FHL1C, and hence is not supposed to have consequences on FHL1B and FHL1C isoforms [6]. Immunoblot analysis revealed a marked increase in FHL1B protein levels while lamin A/C and emerin protein levels were barely unchanged (Fig. 5A). Immunofluorescence analysis showed normal localization of FHL1B in FHL1-mutated myoblasts similar to age-matched control myoblasts (Fig. 5B). The increase in FHL1B staining was homogenous throughout the cell. This result shows that even if the FHL1 mutation has no impact on FHL1B per se, it affects the protein levels of FHL1B.

As tight regulation of FHL1B protein levels seems particularly important during myoblast differentiation (Fig. 4A and B), and having found elevated FHL1B protein expression in FHL1-related EDMD myoblasts (Fig. 5A and B), we looked whether FHL1B protein levels are also affected in myoblasts from patients with mutations in nuclear envelope proteins. We performed immunoblot analysis on the two patient myoblasts with mutations in EMD or SYNE1, previously analysed in this study (Fig. 3B and C) and two additional ones with mutations in LMNA that did not affect lamin A/C expression levels (Supplemental Figure 3). The first one is a short in phase deletion (LMNA p.Lys32del) causing EDMD or LMNA-related congenital muscular dystrophy (L-CMD) [29, 30] while the second one is a point mutation in LMNA (LMNA p.Leu380Ser), linked to L-CMD [29]. All patient myoblasts showed normal FHL1B expression, except patient myoblasts with LMNA p.Lys32del mutation which displayed up-regulated FHL1B protein levels (Fig. 5C). Immunofluorescence performed on these patient’s myoblasts revealed elongated nuclei with increased nucleoplasmic localization of lamin A/C, as reported previously (Fig. 5D) [30]. Interestingly, FHL1B staining seems to be more intense in the nucleoplasm as well.

DISCUSSION

Recent studies on understanding the role played by FHL1 in skeletal muscle have been conducted primarily on the main skeletal muscle isoform FHL1A using transgenic mouse models or cultured mouse muscle cells. In this study, we have chosen to focus on FHL1B and investigated its subcellular localization and expression in healthy and diseased human skeletal muscle cells. Our data are the first to demonstrate a link between FHL1 and the nuclear envelope. Moreover, we showed that FHL1B expression is strongly downregulated during myoblast differentiation, while it is increased in myoblasts of patients with FHL1- and LMNA-related EDMD that have reduced differentiation capacities [6].

In agreement with reported observations in Sol8 and C2C12 myoblasts, we detected both a cytoplasmic and a nuclear localization of FHL1B in primary human myoblasts [14, 31]. In the present study, we further described the nuclear FHL1B to be both in the nucleoplasm and to the NE. If the NE staining was not clearly mentioned previously, it was nonetheless present, although at much lower levels, in some C2C12 myoblasts using the same commercial antibody [14], but not in the first report using a custom-made antibody recognising the same epitope [31]. The strong NE staining in our experiments compared with the one published by Cottle might come from species differences. Alternatively, the NE staining in Sol8 and C2C12 myoblasts might have been underestimated due to the strong nucleoplasmic staining in these cells compared with primary human myoblasts.

Having found FHL1B localized to the NE was extremely exciting, given the fact that we had previously linked FHL1 mutations to EDMD [6], a muscular dystrophy that was so far linked to mutations of the NE proteins emerin and lamin A/C [2, 3]. Using in situ PLA we demonstrated interaction of FHL1B with both lamin A/C and emerin, with increased signal for FHL1B-lamin A/C probably reflecting the abundance of lamin A/C. In contrast to numerous proteins of the NE that require lamin A/C for their correct localization [32 –34], we found lamin A/C not essential for NE localization of FHL1B, suggesting the implication of other proteins. We also excluded emerin and nesprin-1α as putative candidates as FHL1B remained at the NE in myoblasts of patients with no emerin or nesprin-1α expression. One should consider the possibility that FHL1B is targeted to the NE via multiple interactions with proteins of the nuclear lamina. Indeed, LIM domains are known to generate a protein-binding interface allowing formation of multimeric protein complexes [35]. In addition, FHL1B could also be targeted to the NE through interaction with the chromatin and/or chromatin-linked proteins. One such candidate might be the transcription factor RBP-Jκ [18].

Myoblast differentiation is a multistep process that is tightly coordinated where committed myoblasts withdraw from the cell cycle before fusing with surrounding myoblasts to form multinucleated myotubes [36, 37]. It has been reported that FHL1B was excluded from the nucleus following Sol8 and C2C12 myoblast differentiation [14, 31]. In the present study, we showed a strong decrease in nuclear FHL1B staining in differentiated myotubes. Despite a previous study showing the role of the NES in the cytoplasmic translocation of FHL1B upon myogenic differentiation [31], we showed here that FHL1B nuclear export is not sensitive to LMB treatment, meaning that it is independent of exportin-1. Although we cannot exclude that FHL1B is exported through another pathway, since our results showed a strong decrease in FHL1B protein expression with myoblast differentiation, we hypothesize that the important decrease in the nuclear FHL1B fraction might be due to a specific degradation of this nuclear pool, rather than to its translocation to the cytoplasm.

What could be the biological significance of FHL1B at the NE in myoblasts and its degradation during differentiation? Early steps of myoblast differentiation correlate with massive changes in gene expression, where genes involved in cell cycle progression have to be repressed while genes involved in myogenesis have to be activated. During myoblast differentiation, genes that need to be repressed generally relocalize at the nuclear periphery and associated histones acquire inhibitory marks [38]. By homology with FHL1C, which interacts with the Polycomb Repressive Complex 1 (PRC1) [39, 40] to mediate transcriptional repression via histone modifications (for review see [41]), FHL1B may be required at the NE for these early steps of skeletal muscle differentiation. Interestingly, both emerin and lamin A/C have also been associated with functions in chromatin modification [42, 43], and interaction between lamin A/C and proteins from the Polycomb Repressive Complex 2 have been described [44].

Despite the NE localization of FHL1B, the absence of FHL1 mutations specifically affecting this isoform in several patients with an EDMD phenotype raised further questions concerning the direct involvement of FHL1B in the pathophysiology of the disease [6]. Hence, we investigated both the localization and expression levels of FHL1B in myoblasts from an EDMD patient harbouring a mutation in FHL1 that only affects FHL1A. While FHL1B localization was unaffected, we found increased levels of FHL1B protein, probably due to compensatory mechanisms. We also found elevated FHL1B protein levels in myoblasts from an LMNA-related EDMD patient. We think that elevated FHL1B expression is detrimental for myoblast differentiation and indeed myoblasts from the FHL1-patient included in this study displayed major myoblast differentiation defect [6]. As suggested for FHL1C in two recent publications reporting FHL1 mutations in EDMD, we hypothesize that increased FHL1B levels inhibit the expression of Notch target genes, such as DUSP1, via sequestration of RBP-Jκ [18 , 21]. DUSP1 encodes mitogen-activated protein kinase (MAPK) phosphatase 1 (MKP-1), which is involved in the deactivation of the extracellular signal-regulated kinase 1 and 2 (ERK1/2) [20, 45]. Enhanced activation of ERK1/2 is known to be detrimental at the early step of muscle differentiation by repression of myoblast determination protein (MyoD) expression and inhibiting nuclear accumulation of myocyte enhancer factor 2 (MEF2) [36], and has recently been reported in skeletal muscle of a transgenic mouse model for EDMD caused by mutation in the LMNAgene [46].

Altogether, this is the first study to reveal in human skeletal muscle colocalization of FHL1B, lamin A/C and emerin at the NE, the three proteins that are known to be implicated in the same disease. Moreover, we showed that FHL1B protein levels are modified in EDMD patient myoblasts, even in the absence of mutation in this specific isoform. We firmly believe that our work provides an important basis for future investigations of FHL1B functions under normal and pathogenic conditions.

CONFLICT OF INTEREST

The authors have no conflict of interest to report.

Footnotes

ACKNOWLEDGMENTS

We are grateful to Susana Quijano-Roy, Kathryn North, Thomas Voit, Mustafa A Salih and Norma B Romero for providing us with biological material from their patients, Rabah Ben Yaou for his helpful comments and discussion of the manuscript, and the Pitié-Salpêtriére Imaging Platform (PICPS) for microscopy assistance. This work was financially supported by the Institut National de la Santé et de la Recherche Médicale; the Université Pierre et Marie Curie Paris 06, the Centre National de la Recherche Scientifique, the Association Française contre les Myopathies, the Deutsche Forschungsgemeinschaft and the Université Franco-Allemand (as part of the MyoGrad International Graduate School for Myology GK 1631/1 and CDFA-06-11). E.Z received an UPMC-sponsored MyoGrad PhD fellowship and an AFM fellowship.

Appendix

References

Emery

AEH

. Emery-Dreifuss muscular dystrophy – a 40 year retrospective. Neuromusc Disord. 2000;10:228–32.

Bione

, Maestrini

, Rivella

, Manchini

, Regis

, Romei

, et al. Identification of a novel X-linked gene responsible for Emery-Dreifuss muscular dystrophy. Nature Genet. 1994;8:323–7.

Bonne

, Di Barletta

, Varnous

, Becane

, Hammouda

, Merlini

, et al. Mutations in the gene encoding lamin A/C cause autosomal dominant Emery-Dreifuss muscular dystrophy. Nature Genet. 1999;21(3):285–8.

Yorifuji

, Tadano

, Tsuchiya

, Ogawa

, Goto

, Umetani

, et al. Emerin, deficiency of which causes Emery-Dreifuss muscular dystrophy, is localized at the inner nuclear membrane. Neurogenetics. 1997;1(2):135–40.

Aebi

, Cohn

, Buhle

, Gerace

. The nuclear lamina is a meshwork of intermediate-type filaments. Nature. 1986;323:560–4.

Gueneau

, Bertrand

, Jais

, Salih

, Stojkovic

, Wehnert

, et al. Mutations of the FHL1 gene cause Emery-Dreifuss muscular dystrophy. Am J Hum Genet. 2009;85(3):338–53.

Cowling

, Cottle

, Wilding

, D’Arcy

, Mitchell

, McGrath

. Four and a half LIM protein 1 gene mutations cause four distinct human myopathies: A comprehensive review of the clinical, histological and pathological features. Neuromuscul Disord. 2011.

Domenighetti

, Chu

, Wu

, Sheikh

, Gokhin

, Guo

, et al. Loss of FHL1 induces an age-dependent skeletal muscle myopathy associated with myofibrillar and intermyofibrillar disorganization in mice. Hum Mol Genet. 2014;23(1):209–25.

Christodoulou

, Wakimoto

, Onoue

, Eminaga

, Gorham

, DePalma

, et al. 5’RNA-Seq identifies Fhl1 as a genetic modifier in cardiomyopathy. J Clin Invest. 2014;124(3):1364–70.

10.

Robinson

, Brown

, McGrath

, Coghill

, Gurung

, Mitchell

. Skeletal muscle LIM protein 1 regulates integrin-mediated myoblast adhesion, spreading, and migration. Am J Physiol Cell Physiol. 2003;284(3):C681–95.

11.

McGrath

, Cottle

, Nguyen

, Dyson

, Coghill

, Robinson

, et al. Four and a half LIM protein 1 binds myosin-binding protein C and regulates myosin filament formation and sarcomere assembly. J Biol Chem. 2006;281(11):7666–83.

12.

Cowling

, McGrath

, Nguyen

, Cottle

, Kee

, Brown

, et al. Identification of FHL1 as a regulator of skeletal muscle mass: Implications for human myopathy. J Cell Biol. 2008;183(6):1033–48.

13.

Lee

, Chien

, Lin

, Wu

, Wei

, Lee

, et al. Fhl1 as a downstream target of Wnt signaling to promote myogenesis of C2C12 cells. Mol Cell Biochem. 2012;365(1-2):251–62.

14.

Cottle

, McGrath

, Wilding

, Cowling

, Kane

, D’Arcy

, et al. SLIMMER (FHL1B/KyoT3) interacts with the proapoptotic protein Siva-1 (CD27BP) and delays skeletal myoblast apoptosis. J Biol Chem. 2009;284(39):26964–77.

15.

Wong

, Fung

, Ng

, Lee

, Waye

, Tsui

. LIM domain protein FHL1B interacts with PP2A catalytic beta subunit–a novel cell cycle regulatory pathway. FEBS Lett. 2010;584(22):4511–6.

16.

Taniguchi

, Furukawa

, Tun

, Han

, Honjo

. LIM protein KyoT2 negatively regulates transcription by association with the RBP-J DNA-binding protein. Mol Cell Biol. 1998;18(1):644–54.

17.

Kuroda

, Tani

, Tamura

, Minoguchi

, Kurooka

, Honjo

. Delta-induced Notch signaling mediated by RBP-J inhibits MyoD expression and myogenesis. J Biol Chem. 1999;274(11):7238–44.

18.

Liang

, Zhang

, Liang

, Niu

, Zhang

, Feng

, et al. KyoT3, an isoform of murine FHL1, associates with the transcription factor RBP-J and represses the RBP-J-mediated transactivation. Biochim Biophys Acta. 2008;1779(12):805–10.

19.

Gueneau

, Ben Yaou

, Demay

, Llense

, Deburgrave

, Leturcq

, et al. editors Looking for a third gene causing Emery-Dreifuss muscular dystrophy: Lessons and perspectives. Neuromusc Disord. 2006; 11th International Congress of the WMS. Bruges, Belgique, 4-7 October.

20.

Tiffin

, Jenkins

, Gray

, Cameron-Christie

, Eaton

, Aftimos

, et al. Dysregulation of FHL1 spliceforms due to an indel mutation produces an Emery-Dreifuss muscular dystrophy plus phenotype. Neurogenetics. 2013;14(2):113–21.

21.

Pen

, Nyegaard

, Fang

, Jiang

, Christensen

, Molgaard

, et al. A novel single nucleotide splice site mutation in FHL1 confirms an Emery-Dreifuss plus phenotype with pulmonary artery hypoplasia and facial dysmorphology. Eur J Med Genet. 2015;58(4):222–9.

22.

Roman

, Navarro

, Martinez

, Albert

, Polo

, Guardiola

, et al. Unclassifiable arrhythmic cardiomyopathy associated with Emery-Dreifuss caused by a mutation in FHL1. Clin Genet. 2016.

23.

Bigot

, Klein

, Gasnier

, Jacquemin

, Ravassard

, Butler-Browne

, et al. Large CTG repeats trigger p16-dependent premature senescence in myotonic dystrophy type 1 muscle precursor cells. Am J Pathol. 2009;174(4):1435–42.

24.

Manilal

, Randles

, Aunac

, Nguyen

, Morris

. A lamin A/C beta-strand containing the site of lipodystrophy mutations is a major surface epitope for a new panel of monoclonal antibodies. Biochim Biophys Acta. 2004;1671(1-3):87–92.

25.

Salpingidou

, Smertenko

, Hausmanowa-Petrucewicz

, Hussey

, Hutchison

. A novel role for the nuclear membrane protein emerin in association of the centrosome to the outer nuclear membrane. J Cell Biol. 2007;178(6):897–904.

26.

Soderberg

, Leuchowius

, Gullberg

, Jarvius

, Weibrecht

, Larsson

, et al. Characterizing proteins and their interactions in cells and tissues using the in situ proximity ligation assay. Methods. 2008;45(3):227–32.

27.

Voit

, Cirak

, Abraham

, Karakesisoglou

, Parano

, Pavone

, et al. editors. Congenital muscular dystrophy with adducted thumbs, mental retardation, cerebellar hypoplasia and cataracts is caused by mutation of Enaptin (Nesprin-1): The third nuclear envelopathy with muscular dystrophy. 12th International congress of the World Muscle Society; 2007; Taormina, Sicily, Italy.

28.

Fornerod

, Ohno

, Yoshida

, Mattaj

. CRM1 is an export receptor for leucine-rich nuclear export signals. Cell. 1997;90(6):1051–60.

29.

Quijano-Roy

, Mbieleu

, Bonnemann

, Jeannet

, Colomer

, Clarke

, et al. De novo lmna mutations cause a new form of congenital muscular dystrophy. Ann Neurol. 2008;64(2):177–86.

30.

Bertrand

, Ziaei

, Ehret

, Duchemin

, Mamchaoui

, Bigot

, et al. Cellular micro-environments reveal defective mechanosensing responses and elevated YAP signaling in LMNA-mutated muscle precursors. J Cell Sci. 2014.

31.

Brown

, McGrath

, Ooms

, Gurung

, Maimone

, Mitchell

. Characterization of two isoforms of the skeletal muscle LIM protein 1, SLIM1. Localization of SLIM1 at focal adhesions and the isoform slimmer in the nucleus of myoblasts and cytoplasm of myotubes suggests distinct roles in the cytoskeleton and in nuclear-cytoplasmic communication. J Biol Chem. 1999;274(38):27083–91.

32.

Libotte

, Zaim

, Abraham

, Padmakumar

, Schneider

, Lu

, et al. Lamin A/C-dependent localization of Nesprin-2, a giant scaffolder at the nuclear envelope. Mol Biol Cell. 2005;16(7):3411–24.

33.

Sullivan

, Escalante-Alcalde

, Bhatt

, Anver

, Bhat

, Nagashima

, et al. Loss of A-type Lamin Expression Compromises Nuclear Envelope Integrity Leading to Muscular Dystrophy. J Cell Biol. 1999;147(5):913–20.

34.

Vaughan

, Alvarez-Reyes

, Bridger

, Broers

JLV

, Ramaekers

FCS

, Wehnert

, et al. Both emerin and lamin C depend on lamin A for localization at the nuclear envelope. J Cell Sci. 2001;114(14):2577–90.

35.

Kadrmas

, Beckerle

. The LIM domain: From the cytoskeleton to the nucleus. Nat Rev Mol Cell Biol. 2004;5(11):920–31.

36.

Knight

, Kothary

. The myogenic kinome: Protein kinases critical to mammalian skeletal myogenesis. Skelet Muscle. 2011;1:29.

37.

Bentzinger

, von Maltzahn

, Rudnicki

. Extrinsic regulation of satellite cell specification. Stem Cell Res Ther. 2010;1(3):27.

38.

Robson

, de Las Heras

, Czapiewski

, Le Thanh

, Booth

, Kelly

, et al. Tissue-Specific Gene Repositioning by Muscle Nuclear Membrane Proteins Enhances Repression of Critical Developmental Genes during Myogenesis. Mol Cell. 2016;62(6):834–47.

39.

Qin

, Wang

, Liang

, Taniguchi

, Tanigaki

, Han

. RING1 inhibits transactivation of RBP-J by Notch through interaction with LIM protein KyoT2. Nucleic Acids Res. 2004;32(4):1492–501.

40.

Qin

, Du

, Zhu

, Li

, Feng

, Liang

, et al. The PcG protein HPC2 inhibits RBP-J-mediated transcription by interacting with LIM protein KyoT2. FEBS Lett. 2005;579(5):1220–6.

41.

van Kruijsbergen

, Hontelez

, Veenstra

. Recruiting polycomb to chromatin. Int J Biochem Cell Biol. 2015;67:177–87.

42.

Markiewicz

, Dechat

, Foisner

, Quinlan

, Hutchison

. Lamin A/C binding protein LAP2alpha is required for nuclear anchorage of retinoblastoma protein. Mol Biol Cell. 2002;13(12):4401–13.

43.

Koch

, Holaska

. Emerin in health and disease. Seminars in Cell & Developmental Biology. 2014;29:95–106.

44.

Cesarini

, Mozzetta

, Marullo

, Gregoretti

, Gargiulo

, Columbaro

, et al. Lamin A/C sustains PcG protein architecture, maintaining transcriptional repression at target genes. J Cell Biol. 2015;211(3):533–51.

45.

Kondoh

, Sunadome

, Nishida

. Notch signaling suppresses p38 MAPK activity via induction of MKP-1 in myogenesis. J Biol Chem. 2007;282(5):3058–65.

46.

Muchir

, Kim

, Reilly

, Wu

, Choi

, Worman

. Inhibition of extracellular signal-regulated kinase 1/2 signaling has beneficial effects on skeletal muscle in a mouse model of Emery-Dreifuss muscular dystrophy caused by lamin A/C gene mutation. Skelet Muscle. 2013;3(1):17.