Duchenne muscular dystrophy (DMD) is an X chromosome-linked recessive disease, characterized by a progressive wasting of skeletal muscles (1 ). The gene associated with DMD consists of at least 79 exons spanning >2.4 Mb on the human X chromosome (2 ,3 ). The mRNA transcript of this gene in skeletal muscle measures 14 kb and encodes a 427 kDa rod-shaped, spectrin-like protein called dystrophin (4 ). Dystrophin is located on the cytoplasmic surface of the sarcolemma in normal skeletal muscle (5 -7 ). However, its precise cellular functions and the mechanisms by which its absence leads to myodegenerative disorders are unknown.
There are several tissue-specific and developmentally regulated dystrophin isoforms generated through differential promotor usage and alternative splicing at the 3'-end of the gene. Brain-, muscle- and Purkinje cell-specific promoters are located in the 5'-end of the gene, each of which directs transcription through a different 5' first exon (8 ,9 ). Four smaller isoforms, Dp71, Dp116, Dp140 and Dp260, are transcribed using internal promoters and also have unique first exons (10 -13 ). Dp71 is found in many non-muscle tissues, including the brain. Dp116 and Dp140 are exclusively expressed in the peripheral and central nervous system respectively. Finally, Dp260 is a retina-specific isoform located in the outer plexiform layer (OPL) (13 ,14 ).
There have been recent reports of abnormal retinal transmission in boys with either Duchenne or Becker muscular dystrophy (15 -19 ). Dark-adapted electroretinograms (ERG) for these boys recorded under scotopic testing conditions have revealed normal a-waves, but b-waves with reduced amplitude. Under dark-adapted conditions ON bipolar cells depolarize, whereas OFF bipolar cells hyperpolarize in response to light. These reactions cause an increase in extracellular potassium concentration in the OPL (20 ). Subsequently, Müller cells, located in the OPL, take in potassium and transport it through the cell body, later to release it into the vitreous humour. This potassium flux corresponds to a positive polarity when recorded as a b-wave in ERG (21 ,22 ). The reduced b-wave in some DMD and BMD patients implies that full-length dystrophin or its isoforms is involved in the generation of b-waves in ERG under dark-adapted conditions.
Dystrophin exon 52 knock-out mice were recently generated by Araki et al. (25 ). We examined retinal extract from wild-type, mdx and dystrophin exon 52 knock-out mice using the mouse monoclonal antibody MAB1694, specific for the last 17 amino acids of the C-terminal region of dystrophin (Fig. 1 ). In wild-type mice the antibody identified three distinct isoforms having estimated molecular weights of 427, 260 and 71 kDa. A single band at 427, a major band at 260 and two bands around 71 kDa corresponded to full-length dystrophin, Dp260 and Dp71 respectively. The doublet bands, identified as Dp71, resulted from alternative splicing around exons 71-78. We could not detect Dp140 and Dp116 in the retina of wild-type mice, though both isoforms were present in Western blots using a rabbit polyclonal antibody, which recognizes the last 17 amino acids of dystrophin (28 ). We also examined retinal extract using the antibody MANDRA1, raised against the C-terminal domain of dystrophin. Again, this antibody could detect neither Dp140 nor Dp116 in the retina of wild-type mice (data not shown). In the retina of mdx mice Dp260 and Dp71 were present, while full-length dystrophin was undetectable, in agreement with a previous report (13 ). Dp71 was also preserved in the retina of the dystrophin exon 52 knock-out mice, but both full-length dystrophin and Dp260 were absent in this case; these mice proved to be full-length dystrophin- and Dp260-deficient mutants.
Hematoxylin and eosin staining revealed the retina of the mutant mice to be normally organized in an orderly layered anatomical arrangement, similar to wild-type and mdx mice (Fig. 2 ). To determine the localization of dystrophin and its isoforms in the mutant, a series of retinal sections from wild-type, mdx and mutant mice was immunostained with a battery of antibodies against different regions of dystrophin. Using the monoclonal antibody MAB1692, which recognizes the middle part of the rod domain and thus identifies solely full-length dystrophin, we have achieved restricted and punctate staining in the OPL of wild-type mice, while the same staining was undetectable in the OPL of mdx and mutant mice (Fig. 3 a-c). These results indicate that full-length dystrophin is confined to the OPL. We then carried out immunostaining with the polyclonal antibody P-23a, raised against a part of the second half of the rod domain, which recognizes full-length dystrophin and Dp260. Staining in the OPL is present in wild-type and mdx mice, but not in the mutant mice (Fig. 3 d-f). This staining indicates that Dp260 is also confined to the OPL. Using the monoclonal antibody MAB1694, raised against the C-terminus last 17 amino acids of dystrophin, which recognizes all three dystrophin isoforms, we found staining in the OPL of wild-type and mdx mice, but not in the mutant mice (Fig. 3 g-i). Western blot analysis using the same antibody suggested that Dp71 is expressed in the retina of the mutant; however, there was no concentrated staining in the mutant OPL. Retinal blood vessels of wild-type, mdx and mutant mice were equally stained and this finding suggested expression of Dp71 around the blood vessels. We also immunostained sections with the monoclonal antibody MANDRA1, raised against the C-terminal domain of dystrophin, which recognizes the three dystrophin isoforms; the same pattern obtained with MAB1694 was found in the OPL of wild-type, mdx and mutant mice (Fig. 3 j-l). Expression of Dp140 was also disrupted in the mutant (25 ) and thus it is intriguing to know whether Dp140 is expressed in the mouse retina or not. In our Western blots, however, there was no trace of Dp140 or Dp116 in retinal extracts (Fig. 1 ) and expression of both isoforms has not been suggested in our immunohistochemical study.
To determine whether there is [beta]-dystroglycan expression in the mutant retina, we examined retinal sections with an antibody against [beta]-dystroglycan. Restricted and punctate staining of [beta]-dystroglycan was detected in the OPL of wild-type and mdx mice, but was absent in the mutant OPL (Fig. 4 a-c). This result revealed that full-length dystrophin or Dp260 is necessary for localization of [beta]-dystroglycan in the OPL and suggested that without full-length dystrophin or Dp260, dystrophin-associated protein cannot accumulate in the OPL. This antibody stained retinal blood vessels of wild-type, mdx and mutant mice, as did MAB1694 and MANDRA1.
Figure
To determine whether accumulation of mGluR6 is altered in the absence of Dp260, thus causing absence of the dystrophin-dystroglycan complex, we investigated expression of mGluR6 in the OPL (Fig. 4 d-f). Punctate immunoreactivity was observed in the OPL of wild-type, mdx and mutant mice. The pattern and amount of mGluR6 detected in the OPL were indistinguishable among these mice.
Table 1 .
To determine the influence of Dp260 depletion on visual synaptic transmission at the OPL, we compared the ERGs of wild-type mice with those of mutant mice (Fig. 5 ). The b-wave amplitude of the ERG of mutant mice did not differ from that of wild-type mice (Fig. 6 a). However, the implicit time of the b-wave of mutant mice was significantly increased across the entire range of stimulus intensity levels (Fig. 6 b). At a stimulus intensity level of -2.5 logNDF the amplitude and the implicit time of the a-wave were greater and longer in mutant mice than in wild-type mice (Table 1 ). Consequently, the b/a amplitude ratio was smaller in mutant mice compared with wild-type mice. (Table 1 ). We note that an increased amplitude and prolonged implicit time were not observed after relatively strong stimuli (Fig. 6 c and d). The ERGs of mdx mice, which are of B10 background, showed no difference in amplitude and implicit time of the a- and b-waves compared with control B10 mice.
Figure
Figure
In this study we confirmed localization of Dp260 in the OPL of the retina. Although previous immunohistochemical analysis of mdx mice suggested localization of Dp260 in the OPL (13 ), we obtained further evidence by comparing mdx mice with dystrophin exon 52 knock-out mice in another immunohistochemical analysis. The pattern in mutants having neither full-length dystrophin nor Dp260 indicates that the punctate pattern in the OPL of mdx mice was due to expression of Dp260.
Dystrophin exon 52 knock-out mice did not express [beta]-dystroglycan in the OPL, unlike wild-type and mdx mice. [alpha]- and [beta]-dystroglycan span the sarcolemma of skeletal muscle to provide a link between the subsarcolemmal cytoskeleton and the extracellular matrix; they form a large oligomeric complex with dystrophin (29 ,30 ). In the skeletal muscle of mdx mice and DMD patients the absence of dystrophin leads to a reduction in expression of [alpha]- and [beta]-dystroglycan. Dystrophin has been shown to form a complex in the retina. Co-localization of dystrophin and [beta]-dystroglycan has been discovered in the mouse and human OPL (28 ,31 ). The absence of [beta]-dystroglycan in the OPL of the mutant mice strongly suggests disruption of a dystrophin-dystroglycan complex and also implies that Dp260 allows localization of [beta]-dystroglycan at the OPL, although there is no full-length dystrophin in the region.
Compared with wild-type mice the ERGs of dystrophin exon 52 knock-out mice display increased implicit time of the b-wave without any difference in amplitude, larger amplitude and increased implicit time of the a-wave and a lower b/a amplitude ratio. A greater amplitude and increased implicit time of the a-wave might be explained as a secondary effect of delayed generation of the b-wave. This is feasible because no abnormality of the a-wave resulted after high intensity stimuli, which, nonetheless, cause a smaller b-wave. There also remains the possibility that the change in the a-wave stems from functional defects at the photoreceptor level. The delay in the b-wave implicit time indicates a molecular anomaly involving the release of L-glutamate from photoreceptors and the release of potassium ions from ON bipolar cells, all of which take place at the OPL (20 ).
Compared with mdxCv3 mice, which reveal delayed implicit time and reduced amplitude in the b-wave (24 ), dystrophin exon 52 knock-out mice show normal amplitude of the b-wave. The difference between the mdxCv3 mouse and this mutant mouse suggests that Dp71 plays an important role in generation of the b-wave in ERG. From our observations on Western blots and immunohistochemistry using a battery of dystrophin antibodies it is very difficult to determine the contribution of Dp140 and Dp116 in generation of the ERG b-wave. On the other hand, Dp71 was identified in the retina of the mutant mice using Western blot analysis. Expression, however, was not detected in the OPL by immunohistochemical analysis. The absence of [beta]-dystroglycan in the OPL strengthened the idea that Dp71 is not concentrated in large amounts in the mutant OPL. Jung et al. (32 ) examined the subcellular localization of Dp71 in the rat brain and found that it appeared mainly in synaptic plasma membranes and microsomes and not in the mitochondrial fraction. Dp71 has also been detected in cultured glial cells (33 ). Retinal glial cells which might express Dp71 include Müller cells and astrocytes. Positive dystrophin staining around retinal blood vessels might reflect expression of Dp71 in Müller cells and astrocytes (34 ). Interestingly, Müller cells have been suggested to be involved in generation of the b-wave in ERG (21 ,22 ).
Since DMD patients display markedly reduced amplitude of the b-wave, it is difficult to measure the precise implicit time of the b-wave. Dystrophin isoforms can be detected in the retina of DMD/BMD patients with abnormal ERG; however, it is assumed that expression of Dp260 is subject to loss, as evidenced by the deletion pattern of the DMD gene (18 ). The reason why similar genetic defects have different phenotypes in DMD/BMD patients and the dystrophin exon 52 knock-out mice of this study is unknown, but species differences between the mouse and human are likely to cause the divergent phenotypic expression of both. In the mdx mouse regenerated fibers, except those of the diaphragm, are resistant to further degradation. Regenerated fibers are, however susceptible to degradation in DMD patients (35 ).
It is possible that mGluR6 is also involved in generation of the ERG b-wave, since disruption of the gene causes disappearance of the b-wave (27 ). This study clearly indicates that accumulation of mGluR6 is independent of expression of full-length dystrophin/Dp260 in the OPL and of formation of the dystrophin-dystroglycan complex. In the neuromuscular junction utrophin, the autosomal homolog of dystrophin, is closely associated with the nicotinic acetylcholine receptor (nAChR) throughout development and in the adult, but it is not essential for nAChR clustering (36 ,37 ). Instead, other post-synaptic molecules, like rapsin (38 ), agrin (39 ) and MuSK (40 ), are important for this function. Similarly, accumulation of mGluR6 in the OPL may also be regulated by molecules other than dystrophin, but they have not yet been identified. Dystrophin exon 52 knock-out mice revealed b-wave abnormalities without an alteration in mGluR6, implying that full-length dystrophin/Dp260 affect generation of the b-wave independently of accumulation or maintenance of mGluR6.
Formation of the dystrophin-glycoprotein complex has been extensively investigated in skeletal muscle. In this report we have shown that accumulation of [beta]-dystroglycan depends on expression of full-length dystrophin and its isoforms, in particular Dp260, in the OPL through use of dystrophin exon 52 knock-out mice. The molecular background of the generation of ERG and especially the relationship between dystrophin and its isoforms and ERG can be clarified by further study.
The generation of dystrophin exon 52 knock-out mice was described in detail by Araki et al. (25 ). In brief, to construct the targeting vector a 2.5 kb HincII fragment containing exon 52 of the DMD gene was replaced with the neomycin resistance gene. The targeting vector was electroporated into CCE ES cells. Properly recombinated ES cells were cloned by positive-negative selection. Chimeric mice were obtained by injecting this clone into blastocysts of the C57BL/6J inbred strain. Heterozygous female F1 mice (Xmdx52/X) were crossed with C57BL/6J males to obtain dystrophin-deficient male mice. Littermate male mice of hemizygous mutant mice (Xmdx52/Y) were used as wild-type controls. ERGs for five mutants and four wild-type mice were performed. The average age and weight of the mice at the time of testing did not differ significantly among the two groups. The animals were kept on a 12 h light/dark cycle, then anesthetized by i.m. injection of a mixture of ketamine (11 mg/kg), xyladine (14 mg/kg) and urethane (500 mg/kg). All experiments conformed to the standards of the Association for Research in Vision and Ophthalmology Resolution on the use of animals in research.
To obtain samples of the mouse retina, the front hemisphere of the eyeball was removed so that the retina could be gently scraped off from the rear hemisphere with the help of a stereoscopic microscope. Protein fractions were then extracted using a reducing sample buffer (10% SDS, 70 mM Tris-HCl, 5% [beta]-mercaptoethanol, 10 mM EDTA). Aliquots of 20 µg/lane retinal extract were separated on 6 or 8% SDS-polyacrylamide gels. The proteins were subsequently transferred to a PVDF membrane (Millipore) employing an ampage of 242 mA for 2 h. The blot was later incubated with the anti-dystrophin monoclonal antibody MAB1694 (Chemicon), which recognizes the final 17 amino acids of dystrophin. The antibody was used at a 1/200 dilution and the signal was detected using the enhanced chemiluminescence (Amersham) method and a 1/5000 dilution of the anti-mouse secondary antibody (Amersham).
The eyes of mice were embedded in O.C.T. compound (Miles Inc.) and frozen in liquid nitrogen before 6 µm sections were cut and placed on poly(L-lysine)-coated slides. The slides were brought to room temperature, air dried for 1 h and acetone fixed for 10 min. For histological analysis the fixed sections were stained with hematoxylin and eosin. For immunohistochemistry the sections were blocked with 5% goat serum, 2% bovine serum albumin in 0.01 M phosphate-buffered saline at 37°C for 15 min, then incubated overnight at 4°C with the following primary antibodies, diluted in 0.01 M phosphate-buffered saline containing 5% goat serum and 2% bovine serum albumin: P-23a, a rabbit polyclonal antibody raised against the rod domain of dystrophin (amino acids 2360-2409), provided by Dr Yoshida (National Institute of Neuroscience, Tokyo) and used at a 1/50 dilution; MAB1694 (Chemicon), a hybridoma supernatant containing a monoclonal antibody to the last 17 amino acids of the dystrophin C-terminus, used at a 1/500 dilution; MAB1692 (Chemicon), a hybridoma supernatant containing a monoclonal antibody against the rod part of human dystrophin (amino acids 1181-1388), used at a 1/2 dilution; MANDRA1 (Sigma), ascites fluid containing a monoclonal antibody against amino acids 3558-3684 of human dystrophin, used at a 1/100 dilution; anti-[beta]-dystroglycan (Novocastra), a hybrydoma supernatant containing a monoclonal antibody, used at a 1/10 dilution; mGluR6 antibody, a rabbit polyclonal antibody against the C-terminal sequence of rat mGluR6, used at a 1/300 dilution. Following incubation with primary antibodies, the sections were washed three times for 30 min in phosphate-buffered saline containing 0.5% skim milk. The secondary antibodies used were as follows: for P-23a and mGluR6, FITC-conjugated goat antibody against rabbit (Tago Immunologicals); for monoclonal antibody MAB1692, MAB1694, MANDRA1 and [beta]-dystroglycan, FITC-conjugated rabbit antibody against mouse (Tago Immunologicals).
A cotton wick electrode placed on the cornea was partnered with another electrode attached s.c. to the nasal bone. The stimulus light was from a 150 W quartz-halogen lamp. The light was collected and focused onto a 3 mm diameter fiber optic bundle. The edge of the bundle was placed 0.5 cm from the corneal plane. Illuminance of the unattenuated stimulus on the surface of the eye was 140 000 lux and neutral density filters (NDF) were used to reduce the stimulus intensity. The 20 ms duration of the stimulus was controlled by an electromagnetic shutter. After 30 min dark adaptation, responses were recorded beginning with the lowest stimulus intensity. The responses were amplified by a pre-amplifier that had a band width between 1.5 and 100 Hz. Single flash ERGs were recorded. The a-wave amplitude was determined from the baseline to the bottom of the a-wave. The b-wave amplitude was determined from the bottom of the a-wave to the top of the b-wave. The implicit time of the a-wave and the b-wave was defined as the duration between the rise of the stimulus and the peak of the wave.
We thank Dr M.Yoshida for supplying the antibody P-23a. This work was partly supported by grants for Nervous and Mental Disorders from the Ministry of Health and Welfare. The authors thank Mr Daniel North for his review of this manuscript.
BMD, Becker muscular dystrophy; DMD, Duchenne muscular dystrophy; ERG, electroretinogramm; GluR6, metabotropic glutamate receptor subtype 6; nAChR, nicotinic acetylcholine receptor; NDF, neutral density filter; OPL, outer plexiform layer.
Human Molecular Genetics
Pages
Introduction
Results
Identification of dystrophin isoforms in the retina
Localization of dystrophin isoforms in the mutant retina
Localization of [beta]-dystroglycan and mGluR6 in the mutant retina
ERG findings
Discussion
Materials And Methods
Animals
Western blot analysis
Histology and immunohistochemistry
ERG
Acknowledgements
Abbreviations
References
Wild-type
Mutant
P-value
Amplitude (µV)
a-wave
300 ± 54.0
425 ± 58.6
0.013
b-wave
988 ± 153.3
955 ± 92.5
0.704
b/a ratio
3.36 ± 0.706
2.26 ± 0.201
0.049
Implicit time (ms)
a-wave
23.4 ± 3.89
30.1 ± 2.23
0.014
b-wave
46.7 ± 2.35
61.8 ± 0.201
0.0001
REFERENCES
This page is maintained by OUP admin. Last updated Sat Oct 18 13:29:50 BST 1997. Part of the OUP Journals World Wide Web service.
Copyright
Oxford University Press, 1997
