Molecular dissection of the dysferlin protein complex in skeletal muscle

muscle

Huang, Y.

Citation

Huang, Y. (2006, September 26). Molecular dissection of the dysferlin protein complex in

skeletal muscle. Gildeprint Drukkerijen, Enschede. Retrieved from

https://hdl.handle.net/1887/4573

Version: Corrected Publisher’s Version

License: Licence agreement concerning inclusion of doctoral thesis in the_{Institutional Repository of the University of Leiden}

4

Cal

pai

n

3

proteol

ysi

s

regul

ates

AHNAK

turnover

and

prevents

i

ts

i

nteracti

on

wi

th

dysferl

i

n

and

m yoferl

i

n

Yanchao Huang1, Kate Bushby2, Rune R.Frants1, Johan den Dunnen1, Silvère M .van der M aarel1

1 Center for Human and ClinicalGenetics, Leiden University M edicalCenter, Leiden, The Netherlands

2 Institute of Human Genetics, InternationalCentre for Life,

Newcastle-upon-Tyne, UK

Abstract

Mutations in calpain 3 (CAPN3) cause limb-girdle muscular dystrophy type 2A (LGMD 2A). The exact function of this Ca2+-dependent cysteine protease in skeletal muscle is largely unknown but CAPN3 was shown to proteolyze cytoskeletal components and actin binding proteins. Previously, we showed that CAPN3 and AHNAK coexist in the dysferlin protein complex. As AHNAK also interacts with actin and is cleaved by unknown proteases in skeletal muscle, we hypothesized that AHNAK undergoes proteolysis by CAPN3. Here, we demonstrate that CAPN3 and AHNAK colocalize in skeletal muscle and that CAPN3 directly binds to- and cleaves several domains of AHNAK in cell culture. W e also show that AHNAK is downregulated in cells expressing active CAPN3 while being upregulated in patients with a calpainopathy. Moreover, this cleavage of AHNAK by CAPN3 prevents its interaction with dysferlin and myoferlin, suggesting that CAPN3 mediates remodelling of cytoskeleton-sarcolemmal interactions during regeneration and membrane repair in skeletal muscle.

Introducti

on

LGMD2A (MIM# 253600) or calpainopathy is considered the most frequent form of recessive LGMD in most populations investigated [1-3]. The gene responsible for LGMD2A coding for CAPN3 (OMIM# 114240) was localized by linkage analysis to the chromosomal region 15q15.1–15.3 and subsequently identified by positional cloning [3]. To date, >150 distinct pathogenic mutations in the CAPN3 gene have been described, according to

the Leiden Muscular Dystrophy database

(http://www.dmd.nl/capn3_home.html). CAPN3 is a skeletal

muscle-specific member of the calpain superfamily of non-lysosomal, Ca2+ -dependent cysteine proteases [4]. The precise physiological function of CAPN3 and its substrates remains largely unknown. Several cytoskeleton components were identified as partners and substrates for CAPN3 linking its function to the regulation of cytoskeleton structure [5;6]. It was suggested that mutations in CAPN3 lead to incorrect processing of these substrates, eventually resulting in muscular dystrophy [7]. In addition, CAPN3 has been shown to interact with dysferlin by coimmunoprecipitation using VHH antibody fragments against dysferlin [8].

AHNAK1 (meaning 'giant' in Hebrew, also called desmoyokin, or DY, MW

11q12–13 [9]. AHNAK1 contains three main structural regions: the amino-terminal 251 amino acids, large central region of ~4300 amino acids with multiple repeated units, most of which are 128 amino acids in length, and the carboxyl-terminal 1002 amino acids [10]. A second AHNAK1-like protein, AHNAK2 (MW 600 kDa), located on chromosome 14q32, was recently identified by a search for homologous sequences in the human and mouse genome [11]. AHNAK1-deficient mice show no obvious phenotype, and it is speculated that AHNAK2 can compensate for the loss of AHNAK1 [11]. The exact biological function of AHNAK1 is largely unknown. In vitro, the carboxyl-terminal region of AHNAK1 (aa 5262–5643) binds to G-actin and cosediments with F-G-actin suggesting a role for AHNAK in the maintenance of the structural and functional organization of the subsarcolemmal cytoarchitecture in cardiomyocytes [12]. Additionally, our previous studies showed a physical association between AHNAK and dysferlin (Huang et al., submitted).

Haase and colleagues [13] analyzed AHNAK expression in rat cardiac and skeletal muscle preparations using the KIS antibody against the central domain of AHNAK and the Tail antibody against a serine-rich epitope located near the C terminus of AHNAK. They observed a high molecular mass protein of the expected size of 700 kDa and a faint 500 kDa protein on Western blots using the KIS antibody. In addition, a lower molecular weight isoform of AHNAK was also detected by the Tail-antibody. These results suggested that AHNAK is cleaved by potential proteases but to date, the identity of the proteases involved in posttranslational processing of AHNAK is unknown. Moreover, the biological relevance of this cleavage is not understood.

Materials and Methods

Antibodies

The following antibodies were used in this study. The monoclonal anti-dystrophin antibody NCL-DYS2 (Novocastra, Newcastle, UK) was used in a dilution of 1:10 for immunofluorescence microscopy. The mouse monoclonal anti-CAPN3 antibody NCL-12A2 (Novocastra) was used in a dilution of 1:10 for Western blotting. Affinity-purified mouse anti-VSV (P5D4) (gift from Dr. J. Fransen, Nijmegen, Netherlands) was used in a dilution of 1:1,000 for Western blot analysis and in a dilution of 1:250 for immunofluorescent microscopy. The monoclonal anti-HA tag antibody (Upstate Cell signaling solution, VA, USA) was used in a dilution of 1:2,000 for Western blot analysis and in a dilution of 1:100 for immunofluorescence microscopy. Secondary antibodies goat anti-mousealexa488 (Molecular Probes, Eugene, OR), goat anti-rabbitalexa594 (Molecular Probes) and rabbit anti-mouseHRP (DakoCytomation, Glostrup, Denmark) were diluted 1:250, 1:1,000 and 1:1,000, respectively. KIS and CQL anti-AHNAK polyclonal antibodies were obtained from Dr. Jacques

Baudier and had been described previously [14]. Mouse anti-T7 HRP

(Novagen) was diluted 1:10,000 for Western blot analysis. Patient mutation analysis

The unrelated muscular dystrophy case was a patient with genetically confirmed FSHD as indicated by the presence of 2 D4Z4 repeat units on chromosome 4. CAPN3 patient 1 is heterozygous for a c.550delA (p.Thr184ArgfsX36) and a c.1913A>C (p.Gln638Pro) substitution. CAPN3 patient 2 is heterozygous for a c.1342C>T (p.Arg448Cys) substitution and a c.1981delA (p.Ile661X). CAPN3 patient 3 carries a homozygous c.724dupA in exon 5 (p.Arg242LysfsX5).

Fusion protein constructs and expression in prokaryotic vector

parts by BamHI/SacI digestion (designated C1-DY1, from residue

4646-5145) and by SacI/XhoI digestion (designated C2-DY1, from residue

5146-5643) and ligated in BamHI/SacI-digested, or SacI/XhoI-digested pET28c-GST, respectively.

To generate the carboxy-terminus of AHNAK2, C1-DY2 (residue

4428-5145) and C2-DY2 (residue 5146-5637), 2169 bp and 1487 bp fragments

were PCR amplified from cDNA clone (clone ID: DKFZp686J02145,

RZPD, Berlin, Germany) with forward primer

(5'-GGTCGACGTGGAGGTGTCTCTGC-3') and reverse primer

(5'-GCGGCCGCTCCTTTGGAT-3'), forward primer

(5'-GGGTCGACCCTCTCCCTTTTCAGA-3') and reverse primer

(5'-GCGGCCGCTCAGCCTTCATT-3'), respectively, and cloned into TOPO blunt vector (Invitrogen Ltd, Paisley, UK). Subsequently, the fragment was digested with SalI/NotI and ligated in the SalI/NotI-digested prokaryotic expression vector pGEX 4T-3 (Amersham biosciences, Uppsala, Sweden). Unfused GST protein was used as control.

To generate full length CAPN3 (aa 1-821), digestion by SalI/NotI was performed from the plasmid containing CAPN3 in pBLUESCRIPT SK+ and ligated in the SalI/NotI-digested prokaryotic expression vector pET28c. Site-directed mutagenesis was performed by use of the QuickChange site-directed mutagenesis kit (Stratagene, La Jolla, USA). CAPN3 C129S was obtained by replacing the cysteine at position 129 with serine. All sequences were confirmed by automated sequencing (LGTC, Leiden, the Netherlands). The expression plasmid in pGEX 4T-1 vector (encoding the first 124 aa of human dysferlin) and the expression plasmid in pGEX 4T-1 vector (encoding the first 125 aa of human myoferlin) were a kind gift of Professor E. M. McNally (Department of Human Genetics, University of Chicago, USA).

Gluthione-Sepharose 4B (Amersham), incubated at room temperature for 30 minutes, and then centrifuged at 500g for 5 minutes. The supernatant was removed and the Glutathione-Sepharose was washed three times with binding buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 0.2% Triton X-100). The GST/GST-fusion protein affinity beads were ready for pull-down assay.

Fusion protein constructs and expression in eukaryotic vector

DNA fragments encoding the protein fragments N-DY1, M-DY1, C1-DY1 and C2-DY1 were amplified from cDNA clones (Keio University School of

Medicine) with forward primer

CGCGGATCCGCGCCTGGGATAAAG-3') and reverse primer CCGCTCGAGCGGAGGTTTCTGAATAATCA-3'), forward primer CGCGGATCCGCGTCTTTGCCAGATGTT-3') and reverse primer CCGCTCGAGCGGGTGCGTCTGTATATTCA-3'), forward primer

(5'-CGCGGATCCGCGCGTCTGGATTTC-3') and reverse primer

(5'-CCGCTCGAGCGGTTTGGGAAGTTTAAT-3'), forward primer

CGCGGATCCGCGGCTCCTGATCTAA-3') and reverse primer (5'-CCGCTCGAGCGGCTCTTTCTTTGTGGAA-3'), respectively.

A pSG8-VSV eukaryotic expression vector was modified by insertion of a Heamaglutinin (HA) tag at XhoI/BglII sites using forward primer (5'-TCGAGTATCCATATGATGTTCCAGATTATGCTTGAA-3') and reverse primer (5'-GATCTTCAAGCATAATCTGGAACATCATATGGATAC-3'). All PCR products were digested with BamHI/XhoI and ligated in BamHI/XhoI -digested pSG8-VSV-HA vector.

The full length CAPN3 cDNA was amplified from the plasmid containing

CAPN3 in pBLUESCRIPT KS+ with forward primer

CCGCTCGAGCGGTGCCATGCCGAC-3') and reverse primer

(5'-CGCGGATCCGCGATTCGGCATACATGGT-3'). Subsequently, the

fragment was digested with BamHI/XhoI and ligated in BamHI/XhoI -digested eukaryotic expression vector pEGFP N2 (Clontech, CA, USA). CAPN3 C129S was obtained by replacing the cysteine at position 129 with serine by use of the QuickChange site-directed mutagenesis kit (Stratagene). Pull down assays

To determine the association of CAPN3 and AHNAK1, GST-AHNAK1

proteins were used to pull down T7-tagged fusion proteins of CAPN3 C129S

and GST-C2A-myoferlin were used to pull down COS-1 cell lysates which were co-transfected with active CAPN3 and C2-AHNAK1.

Lysates were centrifuged for 15 min at 4ºC and the supernatants were pre-cleared with Gluthione-Sepharose 4B (Amersham) for 1 h at 4ºC. 500 µl aliquots of precleared supernatant were incubated with 10 µg of purified GST fusion protein affinity beads in binding buffer supplemented 1% BSA for o/n at 4ºC, or with unfused GST protein as control. After binding, the beads were washed three times in binding buffer without BSA and 1 time in 50 mM Tris-HCl pH 7.4, then eluted by boiling in 50 µl of 2×SDS-PAGE sample buffer for 5 min. 10 µl was loaded on 12% SDS-PAGE gels to detect proteolytic fragments of CAPN3 C129S. After separation, proteins were transferred to PVDF membranes and dried o. n. Then, blots were incubated with anti T7 HRP (1: 10,000) for 1.5 h at RT for detection of recombinant

CAPN3C129S protein. ECL plus (Amersham) was used for visualization.

Cell culture and transfection

COS-1 (monkey kidney cell line) and 3T3 Fibroblast cells were maintained in Dulbecco's modified Eagle's medium (Gibco-BRL, Grand Island, NY) supplemented with 10% fetal bovine serum (Gibco-BRL), 1% Pen/Strep (100IU/100UG/ML, Gibco-BRL). For plasmid transfections, cells were harvested, plated at 50% subconfluence (200,000 cells per well in 6-well microtiter plates), and allowed to grow for 24 h. 1 µg of each construct and 6 µl of FuGENE 6 transfection reagent (Roche Applied Science) were used to transfect cells. Two days after transfection, cells were lysed in buffer 2 (50 mM Tris, pH 7.4, 0.15 M NaCl, 0.2 % Triton X-100, plus 1× protease inhibitor cocktail (Roche Molecular) for immunoblotting and GST pull-down assay. For immunoblotting, protein concentrations were determined using a BCA protein assay kit (Pierce Science, the Netherlands) according to manufacturer' instructions and an equal amounts of lysates were loaded per well of the SDS-PAGE gel. After separation, proteins were transferred to PVDF membranes and dried o/n. Then, blots were incubated with mouse anti VSV antibody (1:1,000), mouse anti HA antibody (1:2,000) and rabbit anti actin antibody (1:200) for 1.5 h at RT. Rabbit anti mouse HRP and swine anti rabbit HRP were used for detection of cleaved DY1 fusion protein fragments and actin, respectively. ECL plus (Amersham) was used for visualization.

Immunostaining

(PBS) containing 0.1% triton X for 30 min, following by pre-incubation with PBS 4% skimmed milk at room temperature for 2 h. The sections were next incubated with primary antibody fragments o/n at 4˚C, and subsequently by incubation of fluorescein-labeled secondary antibody for 40 min at RT. Background staining was performed by omitting the primary antibody from the first step. The sections were washed with PBS, dehydrated with 70, 90, 100% ethanol and mounted in a DAPI (50 ng/µl)/

VectashieldTM mounting medium (Burlingame, CA). Final preparations

were analysed with a Leica Aristoplan fluorescence microscope and images were obtained using Cytovision (Applied imaging) digital system.

For immunocytochemical examinations, two days after transfection, cells were fixed in 3.7% formaldehyde containing 0.1% triton X for 10 min and then blocked in 1% BSA for 30 min. Primary polyclonal rabbit KIS antibody for detection of endogenous AHNAK was incubated for 2 h at a dilution of 1: 100 and secondary goat anti-rabbitalexa594 antibody (Molecular probes) was incubated for 1 h at a dilution of 1: 1,000.

Results

AHNAK1 and 2 interact with CAPN3

To determine whether CAPN3 directly interacts with both AHNAK proteins and which domains of AHNAK are responsible for the specific binding, a series of recombinant GST-AHNAK fusion proteins representing the amino-terminal domain of AHNAK (N-DY1: aa 2-252), a central repeat unit representing the repeat domain (M-DY1: aa 821-1330), and two

carboxyterminal domains (C1-DY1: aa 4646-5145 and C2-DY1: 5146-5643;

C1-DY2: aa 4428-5145 and C2-DY2: aa 5146-5637) were applied in a

GST-pull down assay (Fig. 1). Due to the rapid autolysis of CAPN3 [6], the proteolytic inactive form CAPN3C129S was used for the GST-pull down assay. Recombinant T7-tagged CAPN3C129S was expressed in BL21 cells and protein extracts were used as prey. Western blot analysis of the pull

down fractions using Mouse anti-T7HRP showed strong interaction between

C1-, C2-DY1 and full-length CAPN3C129S and its degradation fragments

(Fig. 2A). M-DY1 and C2-DY2 were also able to weakly bind to 60 kDa and

Figure 1: Schematic representation of full length AHNAK and the different constructs used in these studies. On the left, the ability to bind and/or act as a substrate for CAPN3 is indicated for each AHNAK fragment.

Figure 2: Identification of the interaction sites between AHNAK and inactive CAPN3C129S. GST fusion proteins representing different domains of AHNAK1 (A) and AHNAK2 (B) or unfused GST lysates were used in pull-down assays with lysates of T7-tagged CAPN3C129S fusion protein as described in Materials and Methods. Bound proteins were separated by SDS-PAGE and analyzed by immunoblotting with anti-T7 HRP_{. In panel A and B, lanes 1-4 represent uninduced fusion proteins,}

induced fusion proteins, soluble fusion proteins, and precleared fusion proteins, respectively. Lane 9 in A and lane 8 in B represent GST alone pull-down fractions. Lane 5-8 in panel A represent GST-N-DY1, GST-M-GST-N-DY1, GST-C1-DY1 and GST-C2-DY1 pull-down fractions. As shown, GST-C1-DY1

and GST-C2-DY1 pulled down T7-tagged full length CAPN3C129S fusion protein and its degradation

products (arrowheads). GST-M-DY1 pulled down weakly a 60 kDa fragment of T7-tagged CAPN3C129S (panel A; lane 6). In panel B, lanes 6 and 7 represent GST-C1-DY2 and GST-C2-DY2

while lane 5 represents GST-C2-DY1, used as positive control. Only GST-C2-DY2 precipitated a 40

kDa fragment of T7-tagged CAPN3C129S. These results demonstrate a specific and direct interaction between M-DY1, C1-DY1, C2-DY1, C2-DY2 and CAPN3C129S. Unfused GST did not interact with

fusion proteins. Arrows indicate full length fusion proteins. A molecular weight marker is indicated on the left.

CAPN3 cleaves endogenous AHNAK

CAPN3 and constitutive inactive CAPN3C129S were transfected individually into COS-1 cells and proteolysis was evaluated by a change in molecular weight of AHNAK on Western blots. No transfection was performed in control cells. Cells were harvested 48 h after transfection and lysates were analyzed by immunoblotting using rabbit polyclonal KIS (Fig. 3A) and CQL (Fig. 3B) antibodies to detect potential proteolytic cleavage products of AHNAK. The KIS antibody detected a 700 kDa protein and high molecular mass proteins in non-transfected control COS-1 cells, wt CAPN3 expressing cells and CAPN3C129S expressing cells. The CQL antibody showed a 700 kDa and a 125 kDa protein in all three samples. Additionally, the CQL antibody revealed that active CAPN3 cleaved endogenous AHNAK to a ~120 kDa fragment. No cleavage was detected in non-transfected control cells or cells non-transfected with inactive CAPN3C129S.

Active CAPN3 and CAPN3C129S were cloned into pEGFP expression vector;

therefore the GFP tag increased CAPN3 molecular weight from 94 kDa to approximately 120 kDa. In Fig. 3C, we show that wt CAPN3 and CAPN3C129S were sufficiently and correctly expressed in COS-1 cells according to the molecular weight of full length protein and their specific proteolytic fragments. Therefore, from these experiments, we conclude that active CAPN3 can cleave AHNAK in COS-1 cells.

Figure 3: Endogenous AHNAK is cleaved by wt CAPN3 in COS-1 cells. In all three panels, total proteins lysates from non-transfected control COS-1 cells (lane 1), COS-1 cells expressing wt CAPN3 (lane 2) and COS-1 cells expressing CAPN3C129S (lane 3) were analyzed by immunoblotting using rabbit polyclonal KIS (Fig. 3A) and CQL (Fig. 3B) antiserum to detect potential proteolytic cleavage products. In panel A, the KIS antibody detected high molecular mass AHNAK proteins in non-transfected control COS-1 cells, wt CAPN3 expressing cells and CAPN3C129S expressing cells. The same lysates were also analyzed by the CQL antibody (Panel B). As shown, AHNAK proteins of 700 kDa and 125 kDa were detected in all three samples. Additionally, the CQL antibody revealed that wt CAPN3 cleaved endogenous AHNAK to a ~120 kDa fragment (arrowhead). No cleavage was detected in non-transfected control cells or cells transfected with inactive CAPN3C129S. In panel C, the CAPN3 antibody 12A2 was used to detect overexpressed CAPN3 protein. Wt CAPN3 and CAPN3C129S _{were sufficiently and correctly expressed in COS-1 cells according to the molecular}

CAPN3 cleaves N-, C1, C2-AHNAK1, but not M-AHNAK1

In Fig. 2A, it was shown that AHNAK1 directly interacts with CAPN3 in a GST-pull down assay. To investigate whether and where CAPN3 can cleave AHNAK1, four fusion proteins representing the N-terminus (N),

repeated domain (M), and C-terminus (C1, and C2) of AHNAK1,

respectively, were constructed to identify specific domains that can be cleaved by CAPN3. Fusion proteins were cloned to pSG8-VSV-HA eukaryotic expression vector fusing the AHNAK fragments N-terminally to a VSV tag and C-terminally to a HA tag. These fusion constructs were individually transfected and cotransfected with various forms of CAPN3

(active GFP-CAPN3, GFP-CAPN3C129S, and unfused GFP vector as control)

and expressed in COS-1 cells. Antibodies recognizing the N-terminal VSV tag and C-terminal HA tag were used to detect proteolytic cleavage products on Western blots. As shown in Fig. 4.1, HA and VSV antibody detection revealed that N-DY1 was cleaved from 35 kDa to 17 and 18 kDa protein

fragments. VSV antibody detection revealed that C1-DY1 was cleaved from

65 kDa to 30 kDa (Fig. 4.3) but HA antibody could not detect cleaved

C-terminal fragments. VSV antibody detection also showed that C2-DY1 was

cleaved from 60 kDa to 36 and 22 kDa fragments and the HA antibody detected complementary fragments of 24 kDa and 38 kDa (Fig. 4.4), respectively. These cleavage products were observed only when wt CAPN3 was transfected, but not in CAPN3C129S, unfused GFP vector or fusion protein transfected cells. No cleavage was detected for M DY1 (Fig. 4.2). Thus, CAPN3 cleaves the N-terminus of AHNAK twice and the C-terminus three times, of which the N-terminal cleavage sites are at close distance. Cleavage of AHNAK prevents the binding of AHNAK to dysferlin or myoferlin

GST-C2A-dysferlin or GST-C2A-myoferlin. No binding was observed for equivalent amount of the control unfused GST fusion.

Figure 4: Specific domains of AHNAK1 are cleaved by wt CAPN3 in cell culture. Four AHNAK1 fusion proteins representing N (Fig. 4.1), M (Fig. 4.2), C1(Fig. 4.3), and C2 (Fig. 4.4) of AHNAK1,

respectively, were expressed in the presence of active or inactive CAPN3 to asses whether AHNAK1 can act as a substrate for CAPN3. The AHNAK fusion proteins were N-terminally tagged with a VSV tag and C-terminally tagged with a HA tag to detect N-terminal and C-terminal AHNAK cleavage products. Actin detection was used for equal loading control (panels A). Antibodies recognizing the N-terminal VSV tag (panels B) and C-terminal HA tag (panels C) were used to detect proteolytic cleavage products on Western blots. In all panels, lane 1-5 represent lysates of non-transfected COS-1 cells, co-transfection of AHNAK fusion protein with wt CAPN3, co-transfection of AHNAK fusion protein with CAPN3C129S, co-transfection of AHNAK fusion protein with unfused GFP vector, and single transfection of AHNAK fusion protein, respectively. As shown in Fig. 4.1, HA antibody and VSV antibody detection revealed that N DY1 was cleaved from 35 kDa to 17 and/or 18 kDa fragments and complementary fragments were observed with the HA antibody; VSV antibody detection revealed that C1-DY1 was cleaved from 65 kDa to 30 kDa (Fig. 4.3) but HA antibody could

not detect cleaved C-terminal fragments. VSV antibody detection also showed that C2 DY1 was

cleaved from 60 kDa to 36 and 22 kDa fragments and the HA antibody detected complementary fragments of 24 kDa and 38 kDa (Fig. 4.4), respectively. These cleavage products were observed only when wt CAPN3 was transfected, but not in CAPN3C129S, unfused GFP vector or fusion protein transfected cells. No cleavage was detected for M DY1 (Fig. 4.2). Red circles represent cleaved fragments detected by VSV antibody and blue circles represent cleaved fragments detected by HA antibody. Below each panel a schematic representation of each AHNAK fusion construct with its CAPN3 cleavage sites is presented.

Cleavage of AHNAK by CAPN3 downregulates AHNAK

immunocytochemically examined wt CAPN3 and CAPN3C129S expressing cells using the KIS antibody to detect AHNAK. Interestingly, we observed the absence of endogenous AHNAK in wt CAPN3 overexpressing COS-1 and 3T3 cells (Fig. 6A and 6A’), compared to nontransfected cells. By contrast, normal or even increased AHNAK signals were observed in CAPN3C129S expressing cells (Fig. 6B and 6B’). Taken together, these results suggest that proteolytic cleavage of AHNAK by wt CAPN3 causes the downregulation of AHNAK in COS-1 and 3T3 cells. In contrast,

CAPN3C129S leads accumulation of AHNAK when normal turnover is

disrupted.

Figure 5: CAPN3-cleaved AHNAK1 fragments lose their ability to bind to dysferlin or myoferlin. GST-C2A-dysferlin and GST-C2A-myoferlin fusion proteins were used in pull-down assays with cell lysates which were co-transfected with active CAPN3 and C2-DY1 as described in Materials and

Methods. Bound proteins were separated by SDS-PAGE and analyzed by immunoblotting with antibodies recognizing the N-terminal VSV tag (Fig. 5A) and C-terminal HA tag (Fig. 5B). In panel A and B, lanes 1-5 represent soluble fractions of cell lysates, precleared lysates, GST-C2A-dysferlin, GST-C2A-myoferlin and GST alone pull-down fractions, respectively. Only full-length C2-DY1

(indicated by arrowheads) was pulled down by GST-C2A-dysferlin (lane 3) and GST-C2A-myoferlin (lane 4). Unfused GST did not interact with C2-DY1 (lane 5). These results demonstrate a specific and

direct interaction between full-length C2-DY1 and C2A-dysferlin and C2A-myoferlin.Cleaved C2

-DY1 fragments (22 kD and 38 kD) lose their binding affinity for dysferlin and myoferlin. Red circles represent cleaved fragments (22 kD) detected by VSV antibody and blue circles represent cleaved fragments (38 kD) detected by HA antibody. A molecular weight marker is indicated on the left.

AHNAK is upregulated in CAPN3 patients

matrix proteins (data not shown). Dystrophin was used as a membrane staining control. We observed clear upregulation of AHNAK at the sarcolemma and blood vessels in LGMD 2A patients, compared to FSHD and normal control (Fig. 7).

Figure 6: Cleavage of AHNAK by CAPN3 downregulates AHNAK in vivo.

AHNAK colocalizes with CAPN3 in longitudinal skeletal muscle fibers Many studies highlighted the diverse localization of AHNAK in a variety of tissues [14;15]. We were particularly interested in the localization of AHNAK in skeletal muscle fibers. Due to the direct interaction of CAPN3 and AHNAK and the cleavage of AHNAK by CAPN3, we expected CAPN3 to colocalize with AHNAK. Previous studies have already shown that CAPN3 colocalizes with its substrates at the myotendinous junctions

and the A-I junction region of the titin molecule [6;16].

To investigate whether AHNAK also localizes in these areas, we performed double immunostaining using the KIS antibody and antibody T11, staining on I-band near the A-I junction region of the titin molecule, an antibody for vinculin staining on the myotendinous junctions, and an anti Į-actinin antibody, staining on the Z-line, respectively. In parallel, double staining was performed for CAPN3 for comparison. We observed that (i) CAPN3 and AHNAK colocalize at I-band near the A-I junction (Fig. 8A). (ii) CAPN3 and AHNAK are enriched in most of MTJ (Fig. 8B) and, (iii) Į-actinin was detected in the middle of each CAPN3 and AHNAK doublet (Fig. 8C).

Discussion

Mutations in the muscle-specific, non-lysosomal cysteine protease CAPN3 cause LGMD 2A. The function of CAPN3 and its role in LGMD2A pathogenesis is still largely unknown. Towards understanding the function of CAPN3 in skeletal muscle, we focus on studying proteins that are in complex with CAPN3 and which may be substrates for CAPN3. Our previous studies demonstrated that CAPN3 and AHNAK are in complex with dysferlin. Moreover, CAPN3 showed the ability to cleave several cytoskeletal components and actin binding proteins. As AHNAK also associates with actin and is proposed to be cleaved by yet unidentified proteases in skeletal muscle, we hypothesized that AHNAK is a substrate for CAPN3.

This suggests that AHNAK1 has a tighter association with CAPN3 than AHNAK2. The identification of both AHNAK proteins as binding partners for CAPN3 open up a novel marcromolecular complex of CAPN3 interacting proteins, including titin [17], filamin C (FLNC) [5] and AHNAK.

In previous studies, Western blots of total protein lysates from rat heart and skeletal muscle revealed the presence of 700 kDa, 500 kDa, 340 kDa, 170 kDa in addition to low molecular mass AHNAK protein fragments by using the C-terminal Tail-antibody. Moreover, a band of approximately 120 kDa was also detected by the CQL antibody in lung and kidney tissues [14]. It was suggested that this band, corresponding to a C-terminal AHNAK proteolytic product, was cleaved by either µ-calpain or m-calpain, but to date, the identity of the proteases that cleave AHNAK is unknown. We analyzed COS-1 cells transiently transfected with active and inactive CAPN3 using the CQL antibody and revealed a 120-kDa fragment only in cells expressing active CAPN3 by Western blotting. Thus, this demonstrates that CAPN3 can cleave AHNAK in a cell culture system.

Due to the strong interaction of AHNAK1 and CAPN3, we subsequently further investigated AHNAK1 cleavage by CAPN3 in COS-1 cells. Four domains of AHNAK1 were fused to a N-terminal VSV and C-terminal HA tag and analyzed in COS-1 cells in the presence of active or inactive CAPN3. N-, C1- and C2-AHNAK1 fusion proteins were clearly cleaved in

cells expressing wt CAPN3. In contrast, specific cleavage products of AHNAK1 were not observed in cells transfected with inactive CAPN3, or unfused GFP vector. No cleavage was also observed for the repeat domain of AHNAK (M-AHNAK1). Taken together, we conclude that C1- and C2

-AHNAK1 directly interact with full-length CAPN3 while N-, C1-, and C2

-AHNAK1 are cleaved by CAPN3.

turnover is regulated by CAPN3 activity in skeletal muscle and that this turnover is perturbed in patients with a calpainopathy.

Although the exact biological function of AHNAK is unknown, AHNAK has been implicated in several essential biological functions based on its interaction with other proteins. In resting neuronal PC12 cells, AHNAK is localized within the lumen of specific vesicles called ‘enlargesomes’, and is redistributed to the external surface of the plasma membrane in response to large increases in Ca2+ [18]. These enlargosomes have been proposed to participate in cell membrane enlargement, differentiation and repair. In cardiomyocytes, AHNAK interacts specifically with the ȕ2 subunit of

cardiac L-type calcium channels at the plasma membrane [12], and is suggested to play a role in protein kinase A (PKA)-mediated signal transduction pathway [19]. In vitro, the carboxyl-terminal AHNAK region (aa 5262–5643) binds to G-actin and cosediments with F-actin [12]. These observations suggest a role for AHNAK in the maintenance of the structural and functional organization of the subsarcolemmal cytoarchitecture. Based on the proteolysis of AHNAK mediated by CAPN3, we suggest that in healthy muscle CAPN3 works as a regulatory enzyme, modulating the homeostasis for AHNAK and that the disruption of this balance may result in the direct and/or indirect regulation of the biological function of AHNAK and its interacting proteins, eventually leading to yet unidentified in vivo biological malfunctions. This has been well demonstrated for the CAPN3-dependent proteolysis of IțBĮ leading to its accumulation in the cytoplasm and nucleus, and subsequent indirectly regulating NF-țB dependent expression of survival genes [20].

It has been shown that CAPN3 cleaves the C-terminus of FLNC [5]. Cleavage of FLNC by CAPN3 inhibits FLNC’s ability to interact with the N-termini of both į- and Ȗ-sarcoglycans, suggesting that CAPN3 proteolytic activities may be physiologically relevant in regulating protein-protein interactions within the cell [5]. We showed that C-terminal AHNAKs can directly interact with C2A-dysferlin and myoferlin and redistributes to sarcolemma with dysferlin during muscle regeneration (Huang, et al., submitted). Analogous to FLNC we observed that cleaved AHNAK fragments also lose their affinity for dysferlin and myoferlin by a GST pull-down assay. This ability of CAPN3 to mediate AHNAK’s interactions with dysferlin and myoferlin may be important for remodelling of cytoskeleton-sarcolemma interactions during muscle regeneration and membrane repair process.

Acknowledgments

We are grateful to Dr. Takashi Hashimoto, Keio University School of Medicine, for providing us with cDNA clones of human AHNAK. We also thank Prof. E. M. McNally, University of Chicago for providing us with GST-C2A-dysferlin and GST-C2A-myoferlin cDNA and Dr. I. Ginjaar, Leiden University Medical Center, for providing muscle cryosections. We thank Dr. J. Baudier for providing us rabbit polyclonal anti-AHNAK KIS and CQL antibodies. This work was supported by grants from SenterNovem (IOP-Genomics IGE01019) and the National Institutes of Health (NIH-NIAMS R21-AR48327-01).

References

1. Dincer P, Leturcq F, Richard I, Piccolo F, Yalnizoglu D, de Toma C, Akcoren Z, Broux O, Deburgrave N, Brenguier L, Roudaut C, Urtizberea JA, Jung D, Tan E, Jeanpierre M, Campbell KP, Kaplan JC, Beckmann JS, Topaloglu H. A biochemical, genetic, and clinical survey of autosomal recessive limb girdle muscular dystrophies in Turkey. Ann Neurol 1997; 42: 222-229.

3. Richard I, Broux O, Allamand V, Fougerousse F, Chiannilkulchai N, Bourg N, Brenguier L, Devaud C, Pasturaud P, Roudaut C, . Mutations in the proteolytic enzyme calpain 3 cause limb-girdle muscular dystrophy type 2A. Cell 1995; 81: 27-40.

4. Goll DE, Thompson VF, Li H, Wei W, Cong J. The calpain system. Physiol Rev 2003; 83: 731-801.

5. Guyon JR, Kudryashova E, Potts A, Dalkilic I, Brosius MA, Thompson TG, Beckmann JS, Kunkel LM, Spencer MJ. Calpain 3 cleaves filamin C and regulates its ability to interact with gamma- and delta-sarcoglycans. Muscle Nerve 2003; 28: 472-483.

6. Taveau M, Bourg N, Sillon G, Roudaut C, Bartoli M, Richard I. Calpain 3 is activated through autolysis within the active site and lyses sarcomeric and sarcolemmal components. Mol Cell Biol 2003; 23: 9127-9135.

7. Ono Y, Sorimachi H, Suzuki K. Structure and physiology of calpain, an enigmatic protease. Biochem Biophys Res Commun 1998; 245: 289-294. 8. Huang Y, Verheesen P, Roussis A, Frankhuizen W, Ginjaar I, Haldane F, Laval S, Anderson LV, Verrips T, Frants RR, de Haard H, Bushby K, den Dunnen J, van der Maarel SM. Protein studies in dysferlinopathy patients using llama-derived antibody fragments selected by phage display. Eur J Hum Genet 2005; 13: 721-730.

9. Kudoh J, Wang Y, Minoshima S, Hashimoto T, Amagai M, Nishikawa T, Shtivelman E, Bishop JM, Shimizu N. Localization of the human AHNAK/desmoyokin gene (AHNAK) to chromosome band 11q12 by somatic cell hybrid analysis and fluorescence in situ hybridization. Cytogenet Cell Genet 1995; 70: 218-220.

10. Shtivelman E, Cohen FE, Bishop JM. A human gene (AHNAK) encoding an unusually large protein with a 1.2-microns polyionic rod structure. Proc Natl Acad Sci USA 1992; 89: 5472-5476.

11. Komuro A, Masuda Y, Kobayashi K, Babbitt R, Gunel M, Flavell RA, Marchesi VT. The AHNAKs are a class of giant propeller-like proteins that associate with calcium channel proteins of cardiomyocytes and other cells. Proc Natl Acad Sci USA 2004; 101: 4053-4058.

12. Hohaus A, Person V, Behlke J, Schaper J, Morano I, Haase H. The carboxyl-terminal region of ahnak provides a link between cardiac L-type Ca2+ channels and the actin-based cytoskeleton. FASEB J 2002; 16: 1205-1216.

domain induces actin bundling and stabilizes muscle contraction. FASEB J 2004; 18: 839-841.

14. Gentil BJ, Delphin C, Benaud C, Baudier J. Expression of the giant protein AHNAK (desmoyokin) in muscle and lining epithelial cells. J Histochem Cytochem 2003; 51: 339-348.

15. Gentil BJ, Benaud C, Delphin C, Remy C, Berezowski V, Cecchelli R, Feraud O, Vittet D, Baudier J. Specific AHNAK expression in brain endothelial cells with barrier properties. J Cell Physiol 2005; 203: 362-371. 16. Keira Y, Noguchi S, Minami N, Hayashi YK, Nishino I. Localization of calpain 3 in human skeletal muscle and its alteration in limb-girdle muscular dystrophy 2A muscle. J Biochem (Tokyo) 2003; 133: 659-664. 17. Sorimachi H, Kinbara K, Kimura S, Takahashi M, Ishiura S, Sasagawa N, Sorimachi N, Shimada H, Tagawa K, Maruyama K, . Muscle-specific calpain, p94, responsible for limb girdle muscular dystrophy type 2A, associates with connectin through IS2, a p94-specific sequence. J Biol Chem 1995; 270: 31158-31162.

18. Borgonovo B, Cocucci E, Racchetti G, Podini P, Bachi A, Meldolesi J. Regulated exocytosis: a novel, widely expressed system. Nat Cell Biol 2002; 4: 955-962.

19. Haase H, Podzuweit T, Lutsch G, Hohaus A, Kostka S, Lindschau C, Kott M, Kraft R, Morano I. Signaling from beta-adrenoceptor to L-type calcium channel: identification of a novel cardiac protein kinase A target possessing similarities to AHNAK. FASEB J 1999; 13: 2161-2172.

20. Richard I, Roudaut C, Marchand S, Baghdiguian S, Herasse M, Stockholm D, Ono Y, Suel L, Bourg N, Sorimachi H, Lefranc G, Fardeau M, Sebille A, Beckmann JS. Loss of calpain 3 proteolytic activity leads to muscular dystrophy and to apoptosis-associated IkappaBalpha/nuclear factor kappaB pathway perturbation in mice. J Cell Biol 2000; 151: 1583-1590. 21. Kramerova I, Kudryashova E, Venkatraman G, Spencer MJ. Calpain 3 participates in sarcomere remodeling by acting upstream of the ubiquitin-proteasome pathway. Hum Mol Genet 2005; 14: 2125-2134.