Expression of dystrophin in the mouse myenteric neurones

Neuroscience Letters 300 (2001) 120±124

www.elsevier.com/locate/neulet

Expression of dystrophin in the mouse myenteric neurones Maria-Giuliana Vannucchi a, Letizia Corsani a, Maria-Grazia Giovannini b, Maria-Simonetta Faussone-Pellegrini a,* a

Department of Human Anatomy, Histology and Forensic Medicine, Section of Histology ªE. Allaraº, University of Florence, Viale G. Pieraccini 6, 50139 Florence, Italy b Department of Preclinical and Clinical Pharmacology, University of Florence, Viale G. Pieraccini 6, 50139 Florence, Italy Received 10 January 2001; accepted 15 January 2001

Abstract Dystrophin, a membrane-associated protein, plays relevant roles in cell functions. Its lack or trunkated expression results in Duchenne muscular dystrophy (DMD), a pathology associated with alterations in gastrointestinal motility considered to be neural in origin. No data are available on the presence of dystrophin in myenteric neurones. We labelled mouse myenteric neurones with DYS1-, DYS2-, DYS3-antibodies; staining was located on the perikarya and processes, with no differences in distribution or intensity among the antibodies; the western immunoblot analysis indicated that myenteric neurones express several dystrophin isoforms; anti-dystrophins/anti-neuronal speci®c enolase double-labeling con®rmed that all neurones express dystrophin. Dystrophin in myenteric neurones might play a role in cytoskeletal organization, axonal transport and signal pathways; its lack might cause the intestinal motor abnormalities reported in DMD patients. q 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Dystrophin; Myenteric neurones; Immunocytochemistry; Mouse; Western blot

Dystrophin is a large membrane-associated protein belonging to the spectrin superfamily [12]. It is encoded by a gene located on the X chromosome which contains seven independent promoters responsible for the transcription of several mRNA and related dystrophin isoforms with a molecular weight (MW) ranging from 427 to 45 kDa. In respect to full-length dystrophin (427 kDa), the shorter isoforms are avoided by the N-terminal actin binding domain and by most of the long spectrin-like rod region, while retaining the cysteine-rich and the highly conserved C-terminal domains [1,8]. Dystrophin has been found in different tissues such as skeletal, cardiac and smooth muscle [5,17], and in some brain and sympathetic neurones [7,9,11,15,19]. In muscle tissue the full-length dystrophin is the most represented isoform [5,19]; conversely, several dystrophin isoforms are present in brain neurones and the shorter ones are particularly abundant, among them Gdystrophin (71 kDa) shows the highest level [9,11,19]. Lack or truncated expression (deletions at the C-terminal domain) of dystrophin in the sarcolemma of skeletal muscle * Corresponding author. Tel.: 139-055-4271389; fax: 139-0554271385. E-mail address: [email protected]®.it (M.-S. FaussonePellegrini).

results in a fatal genetic disease, Duchenne muscular dystrophy (DMD) [10]. The disease is characterized by a rapidly progressive myopathy, usually fatal by the third decade of life. DMD patients, as well as mdx mice, a mutant strain lacking in the dystrophin gene, also show a signi®cant reduction in or absence of dystrophin content and cellular expression in some brain areas [15,19]. These ®ndings have been considered a possible basis for the mild and nonprogressive mental impairment constantly reported in DMD patients. DMD is also associated with clinical manifestations of abnormal gastric and colonic motor activities [4,18] and degenerative neuropathy of the myenteric plexus [21]. Motor abnormalities of the gastric and colonic wall have also been described in the mdx strain and neuropathogenetic involvement has been hypothesized [2,3,16]; however, to date, no data are available on the presence of dystrophin in the enteric neurones. With the aim to verify whether these cells express dystrophin, an immunohistochemical study was carried out in the myenteric neurones of the mouse stomach and colon using antibodies against the three main regional domains of the protein, the central rod, the carboxy terminus and the amino terminus domains. Western immunoblot analysis was performed to ascertain which isoform(s) is (are) expressed

0304-3940/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S0 30 4- 39 40 ( 01) 0 15 55- 5

M.-G. Vannucchi et al. / Neuroscience Letters 300 (2001) 120±124

Fig. 1. Dystrophin immunolabeling (DYS-IR) in mouse myenteric neurones (A±C) and skeletal muscle (D±F). (A) and (D) DYS1-IR; (B) and (E) DYS2-IR; (C) and (F), DYS3-IR. Confocal microscope. Bar, (A±C) 10 mm, (D±F) 20 mm.

by these cells. Finally, double-labeling using dystrophin antibodies and a widely used neuronal marker (neuronal speci®c enolase antibody, NSE) were used to verify whether all myenteric neurones express dystrophin. Immunohistochemistry: 1 month-old male CD1 mice (six animals) were used. The animals were killed by cervical dislocation and segments of the stomach and colon were removed promptly. After excision, specimens were cleaned of digestive material with saline, embedded in OCT compound (Miles, Elkhart, IN, USA) and immediately frozen at 2808C. Transverse sections (12 mm thick) were cut with a cryostat and collected on 0.1% in distilled water polylysine (Sigma, St. Louis, MO, USA) coated slides and incubated with the primary antisera. Some sections that were to be used for co-localization experiments were post®xed with cool acetone for 3 min. Three region-speci®c dystrophin mouse monoclonal antibodies were used: DYS1 (mid rod domain), DYS2 (carboxy terminus), DYS3 (amino terminal domain) (Medac Diagnostika, Wedel, Germany). To label the myenteric neurones, rabbit polyclonal antibody NSE (g, g 0 -enolase) (Af®niti Research Ltd., Nottingham, UK) was used. Sections were preincubated in phosphate buffer saline (PBS) 0.1 M, (pH 7.4) containing 3% Normal Goat Serum (NGS) and 0.5% Triton X-100 (v/v) (BDH, Poole, UK). All primary antisera were diluted in PBS added with NGS 1.5% and Triton X-100 0.5%. Dystrophin antibodies (1:2) were applied for 2 h in a moist chamber at room temperature. None of the antibodies used recognizes utrophin, a protein with a high sequence homology with full-length dystrophin and coded by an autosomical gene. For co-localization experiments, some sections, previously incubated with anti-DYS1 or -DYS2 or -DYS3 antibodies, were then incubated overnight at 48C in the presence of the NSE antibody (1:1000). Nega-

121

tive controls were performed by omitting the primary antibodies or substituting them with a non-immune rabbit or mouse serum. The speci®city of the dystrophin antibodies used was also evaluated by the immunohistochemical labeling of skeletal muscle (Fig. 1D±F). At the end of incubation, sections were rinsed three times in 10 min washes in the same buffer as above. After the ®nal wash, the monoclonal primary antisera were revealed by incubating the sections in the presence of rhodamine anti-mouse IgG (Fab speci®c) conjugate (Sigma Immuno Chemicals, St Louis, MO, USA) secondary antibody, and the polyclonal primary antiserum was revealed by incubating the sections in the presence of ¯uorescein (DTAF)-conjugated pure goat antirabbit IgG (H 1 L Jackson Immuno-Research, West Grove, PA, USA) secondary antibody, diluted 1:40 for 2 h at room temperature. Sections were then mounted in an aqueous medium (Gel Mount, Biomeda Corp., Foster City, CA, USA). The immunoreaction products were observed under an epi¯uorescence Zeiss Axioskop microscope and/or under a Bio-Rad 1024 confocal laser scanning microscope (Cambridge, MA, USA) with laser beam excitation at 488 and 568 nm wave length, and 10±15 optical sections were taken at 0.40 mm intervals. Images of 512 £ 512 pixels were obtained, processed using Adobe Photoshop 5.0 (Adobe Systems Mountain View, CA, USA) and printed using a HP DeskJet 970CX1 Printer. Western immunoblot analysis: 100 ml of ice-cold lysis buffer were added to each tube containing specimens of skeletal muscle (gastrocnemius), gastric and colonic muscle coat. Attempts to obtain laminae containing only muscle tissue, either from the colon or stomach, failed since routine histological control (cresyl violet) revealed the presence of some neurones in all of the samples. The specimens were homogenised on ice directly in the tube (20 strokes, 1 stroke/sec) using a motorised Potter±Elvehjem homogenise. The lysis buffer had the following composition (in mM): 10 Tris±HCl (pH 7.4), 10 NaCl, 5 ethylenebis(oxonitrilo)tetraacetate (EGTA), 0.25% Triton X-100, 1 phenylmethylsulfonyl ¯uoride (PMSF), 20 mg/ml leupeptin, 30 mg/ml aprotinin. Protein determination was performed using the Bradford method with a Bio-Rad Protein Assay reagent (Bio-Rad, Hercules, CA, USA). An appropriate volume of 6£ loading buffer was added to 5 ml of skeletal muscle homogenates and 10 ml of colon and stomach muscle coat homogenates and samples were boiled for 5 min. The samples (74 mg of proteins/well) were loaded on a 5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gel and resolved by standard electrophoresis. The gels were then transferred electrophoretically onto nitrocellulose membrane (Hybond-C extra, Amersham, Arlington Heights, IL, USA). The membranes were blocked overnight at 48C with the following blocking buffer (BB): 5% non-fat dry milk in PBS containing 0.1% Tween 20 (PBS-T). Then, the membranes were probed for 2 h at room temperature using primary antibodies (1:200). All primary antibodies were dissolved in BB. In all the immu-

122

M.-G. Vannucchi et al. / Neuroscience Letters 300 (2001) 120±124

noblot protocols, the blots were washed extensively in PBST after incubation with primary and secondary antibodies (typically three washes, 10 min each). The membranes were incubated with horseradish peroxidase-conjugated antirabbit or anti-mouse IgG (1:1000) and proteins were visualised using chemiluminescence (Amersham ECL Western Blotting Analysis System; Arlington Heights, IL, USA). The results obtained by immunohistochemistry demonstrated that the three region-speci®c dystrophin antibodies used (DYS1, mid rod domain, DYS2, carboxy terminus, DYS3, amino terminus) stained myenteric neurones as well as smooth muscle cells of both gastric and colonic muscle coat. In neurones the labeling was very intense and distributed both in somata and processes. No difference in labeling distribution or intensity was seen among the three antibodies tested (Fig. 1A±C) and intensity of labeling in the neurones was similar to that of skeletal muscle (Fig. 1D±F). Immunoreactivity (IR) was always distributed at the cell periphery (Fig. 1A) and in some neurones the entire or part (Fig. 1A±C) of the cytoplasm were also intensely stained. The nucleus was never immunoreactive. Under the confocal microscope, the peripheral IR appeared patchily distributed on the cell contour (Fig. 2A±C), while staining in the cytoplasm was diffuse (Fig. 2B,C). IR in smooth muscle cells appeared as faintly stained patches, exclusively distributed on the cell contour (Fig. 3). No staining was detected in glial cells. By double-labeling, all the NSE-IR neurones were DYS1-, DYS2- and DYS3-IR (Fig. 2A±C). Since NSE-IR was uniformly distributed throughout the cytoplasm and DYS-IR mainly on the cell periphery, the overlapping between the two IRs was not always complete (Fig. 2C); however, the DYS-IR patches located on the cell contour never overlapped with NSE-IR (Fig. 2A±C) in all the neurones and independently of the antibody used. Western immunoblot analysis revealed the presence of several dystrophin isoforms, both in gastrointestinal and

Fig. 2. Dystrophin and neurone speci®c enolase double-labeling (DYS/NSE-IR) in mouse myenteric neurones. DYS-IR in red, NSEIR in green, DYS/NSE-IR in yellow/orange. (A±C) all the neurones have peripherally distributed DYS-IR patches. In (A) the inner cytoplasmic portion is almost exclusively NSE-IR. In (B) the DYS/NSE-IR overlapping is complete, while in (C) this overlapping occurs over a limited portion of the cytoplasm and the remaining portion is exclusively NSE-IR. Confocal microscope.

Fig. 3. Dystrophin immunolabeling (DYS-IR) in myenteric neurones is more intense than in smooth muscle cells. Longitudinal muscle layer (LM) on the left; circular muscle layer (CM) on the right and the intensely stained neurones in the middle. DYS1-IR. Confocal microscope. Bar, 5 mm.

skeletal muscle (Fig. 4). The specimens containing the myenteric plexus plus the complete muscle coat and those containing the circular muscle layer plus remnants of the myenteric plexus gave the same results. DYS1 and DYS2 antibodies recognized two bands common to stomach, colon and skeletal muscle. One was above 400 kDa and the second was at ~71 kDa. DYS1 labelled also other three bands, whose MW ranged from ~360 to ~160 kDa, which were present in the colon and stomach only. DYS3 antibody did not work. The present ®ndings provide the ®rst evidence for the presence of dystrophin in the myenteric neurones of the mouse stomach and colon. Similar to brain neurones

Fig. 4. Western immunoblot analysis of dystrophin in mouse tissues. Electrophoresis on 5% acrylamide gels. col, colon; st, stomach; sk, skeletal muscle. Lanes 1, 2, 3: DYS1-IR. Five bands are stained in colonic and gastric muscle coat, one above 400 kDa, three from ~360 to ~160 kDa, and one at ~71 kDa. Two bands are stained in skeletal muscle, one above 400 kDa and one at ~71 kDa. Lanes 4, 5, 6: DYS2-IR. Two bands are stained in all specimens, one above 400 kDa and one at ~71 kDa.

M.-G. Vannucchi et al. / Neuroscience Letters 300 (2001) 120±124

[15,19], dystrophin-IR was distributed in the perikarya and processes. Using a confocal microscope it was possible to distinguish a very peripheral and punctate IR in the perikaryon, suggestive for dystrophin accumulations at post synaptic sites [7,11,14,15], and a deep and diffuse IR, suggestive for a labeling of cytoskeletal- and organelle-linked dystrophin [7]. It is noteworthy that labeling intensity in the neurones was higher than in smooth muscle cells with all antibodies used. Finally, NSE/DYS-IR co-localization showed that all myenteric neurones express dystrophin. Altogether, these ®ndings indicate that dystrophin is highly expressed and plays common roles in myenteric neurones. Immunoblot analysis revealed the presence of ®ve bands that should correspond to as many dystrophin isoforms: the band at the highest MW to the full-length dystrophin, the one at the lowest MW to the short-length (G-)dystrophin and the bands at intermediate MW to intermediate-length dystrophins. We consider it reasonable to hypothesize that all the isoforms identi®ed are expressed by mouse myenteric neurones. Indeed, G-dystrophin is the most represented isoform in nerve tissue [9,13,19] and full- and intermediate-length isoforms are present in other neurones in this same animal species (brain [15], sympathetic ganglia [7]) and in man (brain, [19]). Interestingly, labeling distribution in myenteric neurones was the same with all the antibodies used, suggesting that each dystrophin isoform does not have a preferential site of accumulation. The presence of the 71 kDa band obtained with the DYS1 antibody, although unexpected, is in agreement with the results obtained by others using the same antibody [19]. DYS2 antibody results were incomplete when compared with those obtained with DYS1 antibody. On the other hand, contradictory results are frequent when dystrophin antibodies are used in western blot analysis [5,9,15,19]. It is likely that these unpredicted results depend on limits dystrophin antibodies have when used in western blotting. However, we believe that, despite these limits, our immunoblot results are consistent in indicating the presence of several dystrophin isoforms in myenteric neurons. Finally, we cannot exclude that smooth muscle cells express all the dystrophin isoforms presently identi®ed since several isoforms have been found in these cells, among which the full-length one is the most represented [5,19]. The presence of dystrophin in non-contractile cells such as neurones is intriguing. Contrary to skeletal muscle, only hypotheses on the role(s) this molecule plays in neurones can be put forward. For example, the full-length dystrophin, that in skeletal muscle binds to a-actinin-2 through the Nterminal region [20] and regulates contractile apparatus organization, in neurones might be related to cytoskeleton organization and axonal transport. Through its C-terminus and the adjacent cysteine domain binding to integral membrane proteins, dystrophin in the neurones might have a very speci®c role in synaptic activity and signal pathways, anchoring receptors and ion channel proteins at the post synaptic sites. Morphological ®ndings ®t well with

123

the latter hypothesis: a punctate immunolabeling of perikaryal contour was found under ¯uorescence microscope both in brain [15] and myenteric (present data) neurones, and dystrophin accumulations at the post synaptic sites were con®rmed under the electron microscope [7,14]. Besides, in the neurones, similar to skeletal muscle [6], the C-terminal region might bind to nitric oxide synthase (NOS) intervening in signal pathways mediated by nitric oxide (NO) released by a constitutive NOS common to all neurones (i.e. eNOS, personal data). The ®nding of dystrophin in the myenteric neurones may present an opportunity to perform direct functional studies of neuronal dystrophins since these cells are located outside the blood±brain barrier and, therefore, accessible to both in vivo and ex vivo studies. Deletions or lack of one or another dystrophin isoform might correlate to a malfunction of enteric neurones and, consequently, to the intestinal motor abnormalities reported for DMD patients and mdx mice. The ®nancial support of Telethon-Italy (Grant no 1134 to M-S F-P) is gratefully acknowledged. [1] Ahn, A.H. and Kunkel, L.M., The structural and functional diversity of dystrophin, Nat. Genet., 3 (1994) 238±291. [2] Azzena, G.B. and Mancinelli, R., Nitric oxide regenerates the normal colonic peristaltic activity in mdx dystrophic mouse, Neurosci. Lett., 261 (1999) 9±12. [3] Baccari, M.C., Romagnani, P. and Calamai, F., Impaired nitrergic relaxations in the gastric fundus of dystrophic (mdx) mice, Neurosci. Lett., 282 (2000) 105±108. [4] Bensen, E.S., Jaffe, K.M. and Tarr, P.I., Acute gastric dilatation in Duchenne muscular dystrophy: a case report and review of the literature, Arch. Phys. Med. Rehabil., 77 (1996) 512±514. [5] Byers, T.J., Kunkel, L.M. and Watkins, S.C., The subcellular distribution of dystrophin in mouse skeletal, cardiac, and smooth muscle, J. Cell Biol., 115 (1991) 411±421. [6] Chang, W.J., Iannaccone, S.T., Lau, K.S., Masters, B.S.S., McCabe, T.J., McMillan, K., Padre, R.C., Spencer, M.J., Tidball, J.G. and Stull, J.T., Neuronal nitric oxide synthase and dystrophin-de®cient muscular dystrophy, Proc. Natl. Acad. Sci. USA, 93 (1996) 9142±9147. [7] De Stefano, M.E., Zaccaria, M.L., Cavaldesi, M., Petrucci, T.C., Medori, R. and Paggi, P., Dystrophin and its isoforms in a sympathetic ganglion of normal and dystrophic mdx mice: immunolocalization by electron microscopy and biochemical characterization, Neuroscience, 80 (1997) 613±624. [8] Fabbrizio, E., Nudel, U., Hugon, G., Robert, A., Pons, F. and Mornet, D., Characterization and localization of a 77 kDa protein related to the dystrophin gene family, Biochem. J., 299 (1994) 359±365. [9] Gorecki, D.C. and Barnard, E.C., Speci®c expression of Gdystrophin (Dp71) in the brain, NeuroReport, 6 (1995) 893± 896. [10] Hoffman, E.P., Brown, R.H. and Kunkel, L.M., Dystrophin: the protein product of the Duchenne muscular dystrophy locus, Cell, 51 (1987) 919±928. [11] Jancsik, V. and Hajos, F., Differential distribution of dystrophin in postsynaptic densities of spine synapses, NeuroReport, 9 (1998) 2249±2251. [12] Koenig, M., Monaco, A. and Kunkel, L.M., The complete

124

[13]

[14]

[15]

[16]

M.-G. Vannucchi et al. / Neuroscience Letters 300 (2001) 120±124 sequence of dystrophin predicts a rod-shaped cytoskeletal protein, Cell, 53 (1988) 219±228. Lederfein, D., Levy, Z., Augier, N., Mornet, D., Fuchs, O., Yaffe, D. and Nudel, U., A 71kd protein is a major product of the Duchenne muscular dystrophy gene in brain and other non muscle tissue, Proc. Natl. Acad. Sci. USA, 89 (1992) 5346±5350. Lidov, H.G.W., Byers, T.J., Watkins, S.C. and Kunkel, L.M., Localization of dystrophin to postsynaptic regions of central nervous system cortical neurons, Nature, 348 (1990) 725±728. Lidov, H.G.W., Byers, T.J. and Kunkel, L.M., The distribution of dystrophin in the murine central nervous system: an immunocytochemical study, Neuroscience, 54 (1993) 167± 187. Mule, F., D'Angelo, G., Tabacchi, A., Amato, A. and Serio, R., Mechanical activity of small and large intestine in normal and mdx mice: a comparative analysis, Neurogastroenterol. Motil., 11 (1999) 133±139.

[17] North, A.J., Galazkiewicz, B., Byers, T.J., Glenney, J.R. and Small, J.V., Complementary distributions of vinculin and dystrophin de®ne two distinct sarcolemma domains in smooth muscle, J. Cell Biol., 120 (1993) 1159±1167. [18] Nowak, T.V., Ionasescu, V. and Anuras, S., Gastrointestinal manifestations of the muscular dystrophies, Gastroenterology, 82 (1982) 800±810. [19] Uchino, M., Teramoto, H., Naoe, H., Yoshioka, K., Miike, T. and Ando, M., Localisation and characterisation of dystrophin in the central nervous system of controls and patients with Duchenne muscular dystrophy, J. Neurol. Neurosurg. Psych., 57 (1994) 426±429. [20] Watkins, S.C., Cullen, M.J., Hoffman, E.P. and Billington, L., Plasma membrane cytoskeleton of muscle: a ®ne structural analysis, J. Micr. Res. Tech., 48 (2000) 131±141. [21] Yoshida, M.M., Krishnamurthy, S., Wattchow, D.A., Furness, J.D. and Schuf¯er, M.D., Megacolon in myotonic dystrophy caused by a degenerative neuropathy of the myenteric plexus, Gastroenterology, 95 (1988) 820±827.