Oxidative Stress in Desminopathies and Myotilinopathies: A Link between Oxidative Damage and Abnormal Protein Aggregation

Abstract

Myotilinopathies and desminopathies are subgroups of myofibrillar myopathies (MFM) caused by mutations in myotilin and desmin genes, respectively. They are characterized by the presence of protein aggregates in muscle cells. As oxidation of proteins facilitates their aggregation and makes them more resistant to proteolysis, the present study was geared to analyze oxidative stress in MFM. For this purpose, markers of glycoxidation, lipoxidation and nitration were examined with gel electrophoresis and Western blotting, single immunohistochemistry, and double‐ and triple‐labeling immunofluorescence and confocal microscopy in muscle biopsies from patients suffering from myotilinopathy and desminopathy. Increased levels of glycation‐end products (AGEs), N‐carboxymethyl‐lysine (CML) and N‐carboxyethyl‐lysine (CEL), malondialdehyde‐lysine (MDAL), 4‐hydroxynonenal (HNE) and nitrotyrosine (N‐tyr) were found in MFM. Furthermore, aberrant expression of AGE, CML, CEL, MDAL and HNE, as well as of neuronal, inducible and endothelial nitric oxide synthases (nNOS, iNOS, eNOS), and superoxide dismutase 2 (SOD2), was found in muscle fibers containing protein aggregates in myotilinopathies and desminopathies. AGE, ubiquitin and p62 co‐localized in several muscle fibers in MFM. As oxidized proteins are vulnerable to misfolding and are resistant to degradation by the UPS, the present observations support a link between oxidative stress, protein aggregation and abnormal protein deposition in MFMs.

INTRODUCTION

Free radical production is a widespread phenomenon occuring under physiological aerobic metabolism in eukaryotic cells. In a healthy organism, a balance between oxidation and reduction reactions serves to maintain the physiological redox status; however, when oxidative stress occurs this can may cause deleterious metabolic effects, when the generation of reactive oxygen species exceeds the level of antioxidant responses. Skeletal muscle is not an exception, and muscle cells are particularly prone to accumulating oxidative damage to DNA, lipids and proteins over time (22, 30, 38). In addition, evidence for free radical toxicity has been found in several pathological conditions including inclusion‐body myositis (IBM) and myopathies (39, 40), mitochondrial myopathies (5) and dystrophinopathies (25), as well as inflammatory myopathies (14, 32).

The term myofibrillar myopathy (MFM) encompasses a group of muscle disorders characterized morphologically by focal dissolution of the myofibrils and accumulation of protein aggregates (4, 17, 29). Protein aggregates in MFM contain desmin, myotilin and other cytoskeletal proteins, chaperones, phosphorylated tau and β‐amyloid, plectin, gelsolin, clusterin, and ubiquitin (4, 10, 19, 29). MFMs are caused by mutations in different genes, including desmin (11, 12, 16), αB‐crystallin (34), selenoprotein N (6), myotilin (19, 27), ZASP (28) and filamin (35). Little is known about the putative role of oxidative stress in the pathogenic cascade occurring in MFM, although overexpression of semicarbacide‐sensitive amine oxidase, which has been considered as an indicator of oxidative stress in different tissues, has been found in muscle fibers containing protein aggregates in MFM (18).

In order to investigate oxidative stress in MFM, the present study focused on the expression of markers of glycoxidation, lipoxidation and nitration, as well as on nitric oxide (NO)‐producing enzymes and superoxide dismutase (SOD2), in muscle biopsy samples of patients suffering from myotilinopathy and desminopathy. With this purpose, antibodies to advanced glycation end products (AGE), N‐carboxymethyl‐lysine (CML) and N‐carboxyethyl‐lysine (CEL), as glycoxidation markers; antibodies to malondialdehyde‐lysine (MDAL) and 4‐hydroxynonenal (HNE) adducts, as lipoxidation markers; and antibodies to nitrotyrosine (N‐tyr) have been used for gel electrophoresis and Western blotting, and immunohistochemistry. In addition, immunohistochemistry to neuronal, endothelial and inducible isoforms of nitric oxide synthase (nNOS, eNOS and iNOS) was carried out to investigate expression of enzymes linked with NO production. Finally, antibodies to RAGE were used to recognize the expression of AGE receptors, and antibodies to SOD2 to reveal antioxidant responses in muscle cells. Double and triple immunofluorescence and confocal microscopy was performed to further refine oxidative damage with protein aggregates and proteasome markers.

MATERIALS AND METHODS

Patients and muscle biopsies. Muscle biopsies from five patients suffering from myotilinopathy and four patients suffering from desminopathy were obtained following informed consent and in accord with the guidelines of the Ethics Committee of the Hospital Universitari de Bellvitge. Patients suffering from myotilinopathy were two women and three men aged from 52 to 80 years of age (mean age: 71 years). Muscle samples were taken from the lateral gastrocnemius in one case, vastus lateralis in two cases, and deltoid muscle in the remaining two. A detailed description of these patients has been reported elsewhere (19). Desminopathy cases were three women and one man aged from 27 to 49 years of age (mean age 37.5 years). Muscle biopsies in these patients were taken from the lateral gastrocnemius in two cases and from the biceps brachii in the other two. A detailed description of three of these patients has been reported elsewhere (20). Finally, muscle biopsy specimens from five age‐matched controls were used for comparative purposes. Samples were taken from the vastus lateralis in two cases, deltoid in two cases and the biceps braquii in the remaining one. A summary of the clinical characteristics of all the patients included in the present study is provided in Table 1.

Table 1.

Summary of cases included in the present study.

Patient	Age/gender	Site of muscle biopsy	Mutation
1	52/M	Lateral gastrocnemius	MYOT Ser55Phe
2	70/M	Vastus lateralis	MYOT Ser60Cys
3	80/M	Deltoid	MYOT Ser60Phe
4	73/F	Deltoid	MYOT Ser55Phe
5	80/F	Vastus lateralis	MYOT Lys36Glu
6	49/F	Lateral gastrocnemius	DES Pro419Ser
7	46/F	Biceps brachii	DES Leu392 Pro
8	27/M	Lateral gastrocnemius	DES Ile367Phe
9	28/F	Biceps brachii	DES Arg406Trp
10	78/M	Biceps brachii
11	52/F	Deltoid
12	64/F	Deltoid
13	33/M	Vastus lateralis
14	29/F	Vastus lateralis

Western blotting. Roughly 50 mg of frozen muscle samples was homogenized with a manual glass homogenizer in ice with 1 mL of homogenizer buffer (75 mM Tris‐HCl pH 6.8, 0.001% (w/v) bromophenol blue, 15% (w/v) SDS, 20% (v/v) glycerol, 5% (v/v) β‐mercaptoethanol) and a mix of protease inhibitors containing 1 mM PMSF (phenylmethylsulfonylfluoride), 1 µg/mL pepstatin A, 10 µg/mL leupeptin and 10 µg/mL aprotinin. Total homogenates were boiled at 94°C for 4 minutes and centrifuged at 9500 g for 5 minutes. Pellets were discarded and 25 µL aliquots were stored at –80°C.

For Western blot studies 5 µL of each sample was loaded in an 8% SDS‐PAGE electrophoresis (20 mA/gel for 80 minutes at 4°C) and then transferred to nitrocellulose membranes (20 V and 40 mA/gel for 75 minutes) in a Semi‐Dry Transfer System (Bio‐rad, Madrid, Spain). Once the membranes were stained with Ponceau Solution (Sigma, Madrid, Spain) as a transfer quality control, they were immediately incubated with 5× Western Blocking Reagent (Roche, Mannheim, Germany) at 4°C overnight. Then, membranes were incubated with one of the primary antibodies diluted in TBS‐T (100 mM Tris base, 1.4 M NaCl and 0.1% (v/v) Tween 20, pH 7.4) with 1% Western Blocking Reagent (Roche) at 4°C overnight. Details of the antibodies are shown in Table 2. Subsequently, the membranes were washed with TBS‐T and then incubated with the corresponding secondary antibody labeled with horseradish peroxidase (Dako, Barcelona, Spain) at a dilution of 1:1000 or 1:2000 in the same buffer (TBS‐T with 0.5% Blocking Reagent) for 45 minutes at room temperature. After washing the membranes with TBS‐T, the protein bands were detected by chemiluminescence ECL method (Amersham Biosciences, Little Chalfont, UK). The myosin band of 205 kDa stained with Coomassie Brilliant Blue R (Sigma) in the post‐transfer gel was used as a control of protein loading.

Table 2.

Antibodies used in the present study. Abbreviations: IHQ = immunohistochemistry; IF = immunofluorescence; WB = Western blotting.

Antigen	Epitope	Host	Dilution		Manufacturer
			IHQ IF	WB
SOD2	Polyclonal	Rabbit	1:500	1:2000	Stressgen
N‐tyr	Monoclonal	Mouse	–	1:1000	Zymed
nNOS	Polyclonal	Rabbit	1:1000	–	Calbiochem
iNOS	Polyclonal	Rabbit	1:500	–	Chemicon
eNOS	Polyclonal	Rabbit	1:50	–	Chemicon
CEL	Monoclonal	Mouse	1:16	1:300	TransGenic
CML	Monoclonal	Mouse	1:25	1:500	TransGenic
AGE	Monoclonal	Mouse	1:200–250	1:500	TransGenic
RAGE	Polyclonal	Goat	1:500	–	Serotec
HNE	Polyclonal	Rabbit	1:1000	1:1000	Alexis
MDAL	Polyclonal	Goat	1:10	1:300	Academy Bio‐Medical Company
Myotilin	Monoclonal	Mouse	1:100	–	Novocastra
Myotilin	Polyclonal	Goat	1:10	–	Santa Cruz
Desmin	Monoclonal	Mouse	1:15	–	Dako
Desmin	Polyclonal	Rabbit	1:200	–	GeneTex
p62 C‐terminal	Polyclonal	Guinea pig	1:100	–	Progen
Ubiquitin	Polyclonal	Rabbit	1:100	–	Dako

Novocastra, Newcastle, UK; Stressgen, Bionova Científica, Madrid, Spain; Zymed; San Francisco, CA, USA; Calbiochem, San Diego, CA, USA; Chemicon, Barcelona, Spain; TransGenic, Kobe, Japan; Serotec, Bloomington, MN, USA; Alexis, Lausen, Switzerland; Academic BioMedical Company, Houston, TX, USA; Santa Cruz, Quimigen, Madrid, Spain; Progen, Heidelberg, Germany; GeneTex, San Antonio, TX, USA; Dako, Barcelona, Spain; Sigma, Madrid, Spain.

The densitometric quantification of Western blot bands was carried out with Total Lab v2.01 software, and the data obtained were analyzed using Statgraphics Plus v5.1 software. Differences between control and diseased samples were analyzed with ANOVA: *P < 0.05 or **P < 0.01.

Immunohistochemistry. Cryostat sections, 8 µm‐thick, were incubated with 1% hydrogen peroxide for 5 minutes followed by 10% goat or 3% horse normal serum for 2 h, and then incubated overnight with one of the primary antibodies listed in Table 2. After washing, the sections were processed with Super Sensitive Link‐Label IHC Detection System (BioGenex, San Ramon, CA, USA) or with LSAB+ System‐HRP (Dako) following the instructions of the manufacturer. The immunoreaction was visualized as a dark blue precipitate using NH₄NiSO₄ (0.05 M) diluted in phosphate buffer (0.1 M) with 0.03% diaminobenzidine (DAB), 0.04% NH₄Cl and 0.001% hydrogen peroxide. Sections processed only with the secondary antibody were used as negative controls. Serial consecutive sections from each case were stained with hematoxilin and eosin.

Double‐ and triple‐labeling immunofluorescence and confocal microscopy. Cryostat sections 8‐µm‐thick were blocked for 30 minutes at room temperature with 10% fetal bovine serum diluted in 1 × PBS in order to avoid unspecific binding reactions. For double‐labeling immunofluorescence, the sections were incubated overnight at 4°C with different combinations of two primary antibodies as follows: (i) mouse monoclonal anti‐AGE antibody, and goat polyclonal anti‐myotilin or rabbit polyclonal anti‐desmin antibodies; (ii) goat polyclonal anti‐RAGE antibody, and mouse monoclonal anti‐myotilin or mouse monoclonal anti‐desmin antibodies; and (iii) rabbit polyclonal anti‐eNOS antibody and mouse monoclonal anti‐myotilin or mouse monoclonal anti‐desmin antibodies. For triple‐labeling immunofluorescence, some sections were incubated with mouse monoclonal anti‐AGE, rabbit polyclonal anti‐ubiquitin and guinea‐pig polyclonal anti‐p62 (C‐terminus) antibodies. Details of the primary antibodies are shown in Table 2.

After washing, the sections were incubated with the corresponding combination of secondary antibodies Alexa 488 anti‐rabbit (green), Alexa 546 anti‐mouse (red) or Alexa 647 (blue) (all from Molecular Probes, Leyden, The Netherlands), at a dilution of 1:400 for 3 h at room temperature. Subsequently, the nuclei were stained using To‐pro^®‐3‐iodide (Molecular Probes) at a dilution of 1:1000 for 20 minutes at room temperature. Sections were mounted with Fluorescent Mounting Medium (Dako Cytomation), sealed and dried overnight at 4°C. Sections were examined with a Leica TCS‐SL confocal microscope. Sections incubated only with the secondary antibodies were used as controls.

RESULTS

Western blotting

Glycoxidation and lipoxidation markers

Western blots of total homogenates revealed that several bands from 45 to 115 kDa were immunostained with anti‐AGE, anti‐CML and anti‐CEL antibodies. Although variable from one case to another, the bands of 60 kDa had higher intensity in myotilinopathies and desminopathies when compared with controls. The density of the band of about 60 kDa was higher in MFM cases (Figure 1). Densitometry and quantification of this band as revealed with anti‐AGE antibodies showed significant differences in desminopathies and myotilinopathies when compared with controls (P < 0.01 and P < 0.05, respectively). Similarly, differences were significant regarding CEL immunoreactivity between MFM cases and controls (P < 0.05). Western blots to MDAL and HNE also showed multiple bands in control and diseased cases. Yet differences were not significant between control and diseased cases (Figure 1).

Figure 1.

Western blots to AGE, CEL, MDAL and HNE in myotilinopathies (m) and desminopathies (d) compared with controls (c) show multiple bands of molecular weights between 45 and 110 kDa. A band of about 60 kDa is more prominent in diseased cases than in controls. Coomassie blue staining is used to control protein loading. Densitometric studies show significant differences between control and diseased cases regarding AGE and CEL immunostaining (*P < 0.05, **P < 0.01). Differences in the intensity of MDAL and HNE bands of about 60 kDa between control and diseased cases were not significant.

Nitrated proteins (N‐tyr)

Western blots to nitrotyrosine revealed a similar pattern of nitrated proteins in myotilinopathies, desminopathies and control cases. However, a band of 60 kDa showed a significantly higher density in myotilinopathies (P < 0.05) and desminopathies (P < 0.01) when compared with controls (Figure 2).

Figure 2.

Western blots to nitrotyrosine (N‐Tyr) in myotilinopathies (m) and desminopathies (d) compared with controls (c). Several bands are seen in control and diseased muscles. A band of about 60 kDa is more prominent in diseased muscles compared with controls. Significant differences in the intensity of the band of 60 kDa were seen between control and desminopathy (**P < 0.05) and between controls and myoytilinopathy (*P < 0.01) cases.

Immunohistochemistry

Control muscles

Glycoxidation markers In control samples, CML immunoreactivity was observed in the cytoplasm of muscle fibers. A mosaic pattern with two populations of fibers was noted depending on the degree of the immunoreaction. Faint CEL and AGE immunoreactivity was also observed in the cytoplasm of normal muscle fibers, whereas RAGE immunoreactivity was absent. The wall of intramuscular vessels displayed CML, CEL, AGE and RAGE immunoreactivity.

Lipoxidation markers Faint HNE and moderate MDAL immunoreactivity was observed in the cytoplasm of normal muscle fibers.

nNOS, eNOS and iNOS Strong nNOS and moderate eNOS immunoreactivity was observed at the sarcolemma of muscle fibers in control cases, whereas faint iNOS immunoreactivity was seen within the cytoplasm. In addition, strong eNOS and faint iNOS immunoreactivity was seen in the wall of intramuscular vessels and capillaries as well as in the nuclei of myocytes.

SOD2 Faint SOD2 immunoreactivity was present in the cytoplasm of normal fibers. Two populations of fibers could be distinguished according to differing intensities of the immunoreaction. SOD2 immunolabeling was also observed in the nuclei of muscle fibers.

Representative sections of muscle fibers stained with anti‐oxidation and nitration markers is shown in Figure 3.

Figure 3.

Representative sections of AGE (A), CEL (B), CML (C), HNE (D), MDAL (E), eNOS (F), nNOS (G), iNOS (H) and SOD 2 (I) immunoreactivty in control muscles. Cryostat sections without counterstaining. Bar = 50 microns.

Myotilinopathy

Glycoxidation markers Faint CEL and CML, but strong AGE immunoreactivity, were observed in abnormal muscle fibers in myotilinopathies. AGE expression was particularly strong in muscle fibers containing spheroid bodies (Figure 4) In addition, RAGE immunoreactivity was observed in the sarcolemma, as well as within the cytoplasm of abnormal muscle fibers (data not shown). The number of fibers immunostained with CEL, CML, AGE and RAGE varied from one case to another, but all the abnormal fibers were immunostained with the antibodies recognizing glycoxidation markers.

Figure 4.

Muscle biopsy from one patient suffering from myotilinopathy showing a large fiber, and two atrophic fibers containing eosinophilic cytoplasmic inclusions (A). Strong AGE immunoreactivity decorating collections of spheroid bodies in two atrophic fibers, as well as moderate AGE immunostaining in focal areas of the cytoplasm in other fibers (B). Faint CML immunoreactivity in abnormal fibers (C). MDAL immunoreactivity is observed at the periphery of the inclusions, but not within them (D). eNOS (E) and SOD2 (F) are found in abnormal fibers. Cryostat sections without counterstaining (except A). Bar = 50 microns.

Lipoxidation markers Increased HNE expression was found in the sarcolemma of the majority of fibers in myotilinopathies when compared with controls. Furthermore, faint HNE immunoreactivity was observed within the cytoplasm in muscle fibers containing cytoplamic inclusions. Strong MDAL immunoreaction appeared at the periphery of the fibers containing cytoplasmic inclusions, but, curiously, not within the cytoplasmic inclusions themselves (Figure 4). The number of fibers immunostained with HNE and MDAL antibodies varied from one case to another, but always correlated with the number of fibers containing cytoplasmic inclusions.

nNOS, eNOS and iNOS Strong eNOS and nNOS immunoreactivity was observed at the sarcolemma as well as within the cytoplasm of abnormal fibers. Strong iNOS immunostaining was seen in the cytoplasm but not in the sarcolemma of abnormal fibers. In each specimen, all the muscle fibers containing cytoplasmic inclusions were immunostained with the three anti‐NOS antibodies (Figure 4).

SOD2 Strong SOD2 immunoreaction was observed within the cytoplasm in all the muscle fibers containing cytoplasmic inclusions (Figure 4).

Desminopathy

Glyoxidation markers Faint to moderate AGE, CML, CEL and RAGE immunoreactivity was observed in the cytoplasm of muscle fibers in desminopathies. In addition, a small proportion of fibers, corresponding to those containing dense hyaline inclusions displayed strong AGE immunoreactivity. The number of fibers immunostained with AGE, CML, CEL and RAGE varied from case to case, depending on the number of muscle fibers containing cytoplasmic or subsarcolemmal inclusions (Figure 5).

Figure 5.

Muscle biopsy from one patient suffering from desminopathy. A muscle fiber containing a subsarcolemmal inclusion is seen in (A). Serial consecutive sections demonstrating strong AGE (B), MDAL (C), iNOS (D), eNOS (E) and SOD2 (F) immunoreactivity in the abnormal fiber. Cryostat sections without counterstaining (except A). Bar = 50 microns.

Lipoxidation markers Strong HNE immunostaining was observed in the sarcolemma in the vast majority of fibers in desminopathies when compared to controls. Furthermore, HNE adducts occurred in a subset of fibers containing dense hyaline cytoplasmic inclusions. Areas devoid of MDAL immunoreactivity were observed in muscle fibers in desminopathy when compared to controls. These negative areas corresponded to abnormal fiber regions containing non‐hyaline inclusions (Figure 5).

nNOS, eNOS and iNOS Strong eNOS, iNOS and nNOS immunoreactivity was observed in the sarcolemma of muscle fibers. Furthermore, eNOS, iNOS and nNOS immunostaining decorating fiber regions containing inclusions was observed in every case (Figure 5).

SOD2 Increased SOD2 immunoreactivity was also noted in abnormal fibers in desminopathies (Figure 5).

Double and triple‐labeling immunofluorescence and confocal microscope

Myotilinopathy

Markers of glycoxidation in muscle fibers containing myotilin aggregates Double‐labeling immunofluorescence confirmed the existence of glycoxidation immunoreactive products in muscle fibers containing protein aggregates. Double labeling immunofluorescence revealed the coexistence of AGE (red) and myotilin (green) in the same fibers. AGE and myotilin co‐localized in muscle fibers containing small, confluent, punctate aggregates, as well as at the periphery of larger aggregates. In general, the areas showing AGE immunofluoresence exceeded those displaying myotilin immunofluoresence (Figure 6).

Figure 6.

Double‐labeling immunofluorescence and confocal microscopy to myotilin (green) and AGE (red) in myotilinopathy. Myotilin deposition occurs as multiple small confluent granular aggregates, as diffuse deposits in atrophic fibers or decorating the periphery of abnormal muscle fibers (A,D). AGE immunoreactivity is observed as large dense deposits in abnormal fibers (B,D). Co‐localization of myotilin and AGE (merge, yellow) is found in some aggregates, whereas AGE immunoreactivity is localized inside myotilin aggregates in other fibers (C,F). One section of the same case stained only with the secondary antibodies is used as a negative control (G–I). Nuclei are visualized with To‐pro®‐3‐iodide (blue).

In contrast, sections double‐immunostained with RAGE (green) and myotilin (red) revealed almost complete co‐localization of RAGE and myotilin in affected fibers (Figure 7).

Figure 7.

Double‐labeling immunofluorescence and confocal microscopy to myotilin (green, A) and RAGE (red, B) showing almost complete co‐localization of RAGE and myotilin (merge, yellow, C). One section of the same case stained only with the secondary antibodies is used as a negative control (D–F). Nuclei are visualized with To‐pro®‐3‐iodide (blue).

Lack of localization of MDAL in myotilin aggregates In line with single labeling immunohistochemistochemical observations, faint MDAL immunofluoresence was observed surrounding myotilin aggregates but no MDAL immunostaining was found within myotilin aggregates (data not shown).

eNOS and myotilin double‐labeling Double‐labeling to eNOS (green) and myotilin (red) revealed complete co‐localization in protein aggregates. However, partial co‐localization was observed in atrophic fibers exhibiting diffuse myotilin immunostaining (Figure 8A–C).

Figure 8.

Double‐labeling immunofluorescence and confocal microscopy to eNOS (green, A,D) and myotilin (red, B) in myotilinopathy (A–C) and with desmin (red, E) in desminopathy (D–F). Co‐localization of eNOS and myotilin is observed in some fibers in myotilinopathy (merge, C), whereas atrophic fibers displaying diffuse myotilin immunofluorescence show partial co‐localization. Partial co‐localization of e‐NOS (D) and desmin (E) is observed mainly at the sarcolemma of some fibers in desminopathy. One section stained only with the secondary antibodies is used as a negative control (G–I). Nuclei are visualized with To‐pro®‐3‐iodide (blue).

Triple‐labeling to AGE, ubiquitin and p62 Confocal microscopy of triple immunolabeling revealed co‐localization of AGE (green), ubiquitin (red) and p62 (blue) in several but not all aggregates. Some AGE products were not immunolabeled with ubiquitin or with anti‐p62 antibodies (Figure 9).

Figure 9.

Triple‐labeling immunofluorescence and confocal microscopy to AGE (green, A,E), ubiquitin (red, B,F) and p62 (blue, C,G) in myotilinopathy showing co‐localization of AGE adducts with ubiquitin and p62, in several but not all aggregates (merge, yellow, D,H). Yet some AGE products are not imunolabeled with ubiquitin or with anti‐p62 antibodies. One section stained only with the secondary antibodies is used as a negative control (I–L). Nuclei are visualized with To‐pro®‐3‐iodide (blue).

Desminopathy

Markers of glycoxidation in muscle fibers containing desmin aggregates AGE immunofluorescence revealed the presence of immunoreactive adducts within some muscle fibers, whereas desmin immunofluorescence revealed desmin deposition in the subsarcolemma as well as within the cytoplasm of selected fibers. Double‐labeling to AGE (green) and desmin (red) revealed the coexistence of AGE adducts and desmin aggregates in the same fibers. Yet AGE‐immunostained products were usually found at the center of desmin aggregates (Figure 10).

Figure 10.

Double‐labeling to AGE (green, A,D) and desmin (red, B,E) in desminopathy. AGE deposition is observed in several fibers, even some fibers which do not contain desmin aggregates. Partial co‐localization of AGE and desmin is present in some fiber regions (merge, yellow, C,F). One section of the same case stained only with the secondary antibodies is used as a negative control (G–I). Nuclei are visualized with To‐pro®‐3‐iodide (blue).

MDAL and desmin double‐labeling Faint MDAL immunofluoresence was seen in the subsarcolemma in a small proportion of fibers. No MDAL immunofluorescence was present in fiber regions containing desmin aggregates (data not shown).

eNOS and desmin double‐labeling eNOS (green) immunofluoresence was widely distributed in the sarcolemma, as well as within the cytoplasm of selected fibers. eNOS co‐localized with desmin in some areas but not in others (Figure 8D–F).

Triple‐labeling to AGE, ubiquitin and p62 Confocal microscopy of triple immunolabeling revealed co‐localization of AGE (green), ubiquitin (red) and p62 (blue) in some fiber regions. However, AGE co‐localized with ubiquitin but not with p62 in other fibers (Figure 11).

Figure 11.

Triple‐labeling immunofluorescence and confocal microscopy to AGE (green, A,E), ubiquitin (red, B,F) and p62 (blue, C,G) in desminopathy showing co‐localization of AGE adducts with ubiquitin and p62, in some fiber regions (merge, yellow, D,H). In other fibers, AGE co‐localized with ubiquitin but not with p62. One section stained only with the secondary antibodies is used as a negative control (I–L). Nuclei are visualized with To‐pro®‐3‐iodide (blue).

DISCUSSION

Basal levels of glycoxidized, lipoxidized and nitrated protein species were found in skeletal muscles of controls. This is not surprising as protein oxidation is a normal phenomenon occurring as a consequence of aerobic metabolism. Yet increased levels of CEL, CML and AGE, mainly glycoxidative adducts, as well as HNE and MDAL lipoxidized proteins have been noted in muscle samples in myotilinopathies and desminopathies. Furthermore, increased levels of N‐Tyr in diseased muscles points to extensive oxidative/nitrosative damage in MFMs. Combined gel electrophoresis and Western blotting, and immunohistochemistry, have shown not only an increase but also a re‐localization of oxidative products in association with abnormal protein aggregates in MFMs. Differences do exist between myotilinopathies and desminopathies, but these are largely related to the characteristics of the protein deposits rather than to the specificity of the oxidative lesions. Similarly, enhanced immunostaining to eNOS, iNOS and nNOS is found in myotilinopathies and desminopathies; and RAGE and SOD2 expression is activated in both groups of MFMs.

However, glycoxidative and lipoxidative markers do not match in a particular fiber with abnormal protein aggregates. Although AGE co‐localizes with protein aggregates, MDAL immunoreactivity is largely located at the external borders of protein deposits. This is further demonstrated with double‐labeling immunofluorescence and confocal microscopy, thus suggesting that different proteins are substrates for glycoxidation, lipoxidation or both. It is also interesting to demonstrate re‐localization of NO‐producing enzymes in association with abnormal protein deposits in myotilinopathies and desminopathies. Similar sub‐cellular re‐distribution of NOS has been previously demonstrated in dystrophinopathies (1, 23, 36), and it is likely related to the production of NO as a mediator to reduce cellular susceptibility to metabolic stress (24, 31).

Characterization of targets to oxidation and nitration is a means of identifying particular cytoskeletal proteins and enzymes, but preliminary studies have shown desmin to be the main target of glycoxidation and nitration in myotilinopathies and desminopathies (15). The present findings show that other proteins, in addition to desmin, may be putative substrates of oxidation and nitration in MFMs. In this line, double‐labeling immunofluorescence and confocal microscopy revealed co‐localization of AGE and myotilin, and of AGE and desmin in damaged muscle fibers in myotilinopathies and desminopathies, respectively. Interestingly, AGE adducts were not only present in muscle fibers containing myotilin aggregates in myotilinopathies, but also in fiber areas devoid of myotilin deposits. AGE adducts were located at the center of the desmin deposits in desminopathies.

Oxidized proteins appear to be degraded by the 20S proteasome (2). Yet excessive oxidative stress may render the proteolytic capacity of this system insufficient, and thereby facilitate the accumulation of abnormal and often disabled proteins, through covalent cross‐linking reactions and increased surface hydrophobicity (2, 3, 13). Previous studies have shown abnormal ubiquitin‐proteasome system (UPS) in MFM, basically characterized by abnormal expression of several subunits of the 19S and 20S proteasome, up‐regulation of the immunoproteasome and enzymatic proteolitic activity (7). This is accompanied by increased clusterin and immunoreactivity in association with abnormal protein deposits, and expanded aggresome as revealed by increased immunoreactivity to γ‐tubulin in damaged fibers (8). Furthermore, mutant ubiquitin (UBB⁺¹) has recently been found to accumulate in muscle fibers in myotilinopathies and desminopathies (21). Finally, p62 was demonstrated in association with protein deposits in MFMs (21). p62 is a protein for which a role in protein aggregation and degradation has recently been attributed; p62 is able to self‐aggregate and it has high affinity for multi‐ubiquitin chains (9, 23, 26, 33, 37).

In the present study, AGE co‐localized p62 and ubiquitin in several abnormal protein aggregates, thus establishing a link between protein oxidation and protein aggregation in myotilinopathies and desminopathies. As the oxidation of proteins not only facilitates their aggregation, but also makes them resistant to proteolytic machinery, the present observations suggest that oxidative damage added to protein mutation is an important contributory factor to muscle fiber damage in myotilinopathies and desminopathies.