The mouse dystrophin muscle enhancer-1 imparts skeletal muscle, but

Human Molecular Genetics, 2004, Vol. 13, No. 22 doi:10.1093/hmg/ddh305 Advance Access published on September 22, 2004

2853–2862

The mouse dystrophin muscle enhancer-1 imparts skeletal muscle, but not cardiac muscle, expression onto the dystrophin Purkinje promoter in transgenic mice Yves De Repentigny1, Philip Marshall1, Ronald G. Worton1,2,4 and Rashmi Kothary1,3,4,* 1

Ottawa Health Research Institute, University of Ottawa Center for Neuromuscular Disease, Ottawa, Ontario, Canada K1H 8L6, 2Department of Biochemistry, Microbiology and Immunology, 3Department of Cellular and Molecular Medicine and 4Department of Medicine, University of Ottawa, Ottawa, Ontario, Canada K1H 8M5

Received July 26, 2004; Revised and Accepted September 15, 2004

A subset of patients harboring mutations in the dystrophin gene suffer from X-linked dilated cardiomyopathy (XLCM), a familial heart disease that is not accompanied by any clinical signs of skeletal muscle myopathy. As the muscle (M) isoform of dystrophin is not expressed in these patients, the absence of skeletal muscle symptoms has been attributed to expression of the brain (B) and cerebellar Purkinje (CP) isoforms of dystrophin in skeletal, but not cardiac, muscles of XLCM patients. The compensatory mechanism of dystrophin B and CP promoter upregulation is not known but it has been suggested that the dystrophin muscle enhancer from intron 1, DME-1, may be important in this activity. Previous studies have shown that the presence of the DME-1 is essential for a significant increase in dystrophin B and CP promoter activity in skeletal muscle cells in culture. Here, we demonstrate that the mouse dystrophin CP promoter drives expression of a lacZ reporter gene specifically to the cerebellar Purkinje cell layer but not to skeletal or cardiac muscle of transgenic mice. However, if the mouse counterpart of DME-1 is present in the transgene construct, the dystrophin CP promoter is now activated in skeletal muscle, but not in cardiac muscle. Our findings provide in vivo evidence for the importance of the dystrophin muscle enhancer sequences in activating the dystrophin CP promoter in skeletal muscle. Furthermore, they provide support for the model in which muscle enhancers, like DME-1, activate the dystrophin B and CP promoters in skeletal muscle, but not in cardiac muscle, of XLCM patients.

INTRODUCTION Duchenne and Becker muscular dystrophies (DMD/BMD) are muscle degenerative diseases that are caused by deficiency of the cytoskeletal protein dystrophin in skeletal and cardiac muscles (1,2). The gene encoding dystrophin spans 2.9 Mb and is located on the short arm of the X-chromosome. It consists of 79 exons and is expressed as a 14 kb transcript in muscle cells (3,4). To date there have been at least eight promoters identified driving transcription at the DMD locus and producing several tissue-specific isoforms (reviewed in 5). Of these, only the muscle (M), brain (B) and the cerebellar Purkinje (CP) promoters produce transcripts encoding a full-length 427 kDa dystrophin protein (6 – 10).

Within the larger group of patients with mutations in the dystrophin gene, there exists a subgroup that suffers from a progressive myocardial disease resulting in a dilated cardiomyopathy (11,12). Although these patients do have a skeletal muscle pathology (high creatine kinase and reduction of dystrophin expression in skeletal muscle), they do not display any skeletal muscle symptoms (11,12), indicating that the residual dystrophin in skeletal muscle is sufficient for escaping the dystrophic phenotype. This myocardial disease has been termed X-linked dilated cardiomyopathy (XLCM) and amongst the first few patients studied, several had a mutation affecting a regulatory sequence within the dystrophin gene (12 – 14). The lack of skeletal muscle symptoms in some XLCM patients appears to be due to an increase in the expression of nonmuscle dystrophin isoforms in the skeletal muscles of these

*To whom correspondence should be addressed at: Ottawa Health Research Institute, 501 Smyth Road, Ottawa, Ontario, Canada K1H 8L6. Tel: þ1 6137378707; Fax: þ1 6137378803; Email: [email protected]

Human Molecular Genetics, Vol. 13, No. 22 # Oxford University Press 2004; all rights reserved

2854

Human Molecular Genetics, 2004, Vol. 13, No. 22

patients (15 –17), highlighting the importance of understanding dystrophin gene regulation. Dystrophin expression and function has largely been studied in muscle tissues, whereas in the brain the regulation of the dystrophin B and CP promoters is not well understood. Early studies on the dystrophin M promoter indicated that the first 150 bp are required for muscle-specific expression in cultured myogenic cells (18) and that the muscle promoter is regulated by the serum response factor, the dystrophin promoter-bending factor and the YY1 transcription factor (19,20). In addition, a transcriptional enhancer has been identified 6.5 kb downstream of the M promoter within muscle intron 1 of the human dystrophin gene (21). This enhancer, referred to as dystrophin muscle enhancer-1 (DME-1), increases transcription from the dystrophin M promoter in both myoblasts and myotubes (22). Similarly, we have characterized an intron 1 enhancer element located 8.5 kb downstream of the mouse dystrophin M promoter (23). This enhancer shows 65% sequence identity with its human counterpart and we, therefore, refer to it as the mouse dystrophin muscle enhancer-1 (mDME-1). Although there are some differences between the human and mouse dystrophin muscle enhancers in the sequence elements that confer enhancer function, both have similar effects on promoter activity. Indeed, we have demonstrated the muscle specificity of mDME-1 in cultured myogenic cell lines (23). A 3 kb fragment harboring the mDME-1 increases transcription from the mouse dystrophin muscle promoter by a factor of 20 in G8 myotubes and by a factor of 6 in cardiac muscle-derived H9C2 myotubes (24). In vivo, the dystrophin muscle promoter alone targets expression of a reporter lacZ gene only to the right ventricle of the hearts of transgenic mice (24,25), suggesting the need for additional regulatory elements to promote expression in skeletal muscle and the rest of the heart. Addition of the mDME-1-containing fragment to the dystrophin muscle promoter-lacZ transgene was sufficient for expression in skeletal muscle as well as in other compartments of the heart (24). Thus, it appears that the dystrophin muscle promoter, although essential, is dependent on the enhancer sequence to target both skeletal and heart muscles. The molecular mechanism by which XLCM patients escape a skeletal muscle dystrophic phenotype is largely unknown, but analysis of the deletion breakpoints of two XLCM patients revealed the loss of the dystrophin M promoter, whereas the dystrophin B and CP promoters were left intact (26). Furthermore, the latter study showed that the dystrophin B and CP promoters were essentially inactive in muscle cell lines unless the dystrophin muscle enhancer was present in the expression constructs. They also showed that the DME-1 induced activation of the dystrophin non-muscle promoters occurred selectively in skeletal muscle cells in culture. These results suggested the possible role of DME-1 in the induction of dystrophin B and CP promoters within the skeletal muscle of XLCM patients, thus helping to explain the lack of symptoms in this tissue. In the present study, we have assessed whether the mouse dystrophin Purkinje promoter is capable of driving the expression of a lacZ reporter gene in the muscle of transgenic mice. Our results demonstrate that this promoter is specifically

activated in the cerebellar Purkinje cell layer and is not expressed in skeletal or cardiac muscles. We also report that the mDME-1 can impart skeletal muscle expression, but not cardiac muscle expression, onto the dystrophin Purkinje promoter in transgenic mice, providing support for the model in which the DME-1 sequence activates the dystrophin B and CP promoters in skeletal muscle, but not in cardiac muscle, of XLCM patients.

RESULTS Mouse dystrophin muscle enhancer activates transcription from the mouse dystrophin Purkinje promoter in the G8 myogenic cell line Our previous results have demonstrated that the mDME-1 is capable of increasing the expression of a reporter gene under the control of the mouse dystrophin muscle promoter in both myogenic cell lines and transgenic mice (23,24). Other studies have shown that the human dystrophin CP promoter is inactive in skeletal muscle (16). To determine whether the mDME-1 has the potential to affect transcription from the dystrophin CP promoter, we have purified by polymerase chain reaction (PCR)-amplification a 1.1 kb region just upstream of the start of the dystrophin CP transcript from mouse genomic DNA (Fig. 1). Previous analysis of this sequence has failed to identify any CpG islands, has noted the presence of a weak match to a consensus TATA box and has identified two putative E-box elements that might serve as sites for MyoD binding (27; Fig. 1). The activity from the mouse dystrophin Purkinje promoter (DE3 construct) was minimal after transfection into G8 myogenic cells. In sharp contrast, the enhancer-containing fragment (DE4 construct) increased transcription from the mouse dystrophin Purkinje promoter by 23-fold in G8 myoblasts and 28-fold in G8 myotubes (Fig. 2). These results suggest that the mouse dystrophin Purkinje promoter alone is not sufficient to impart expression in muscle cell cultures, but is capable of this activity in the presence of the mDME-1. The mouse dystrophin Purkinje promoter drives expression of a linked lacZ reporter gene specifically to the cerebellar Purkinje cell layer in transgenic mice Earlier reports have raised the possibility that the dystrophin Purkinje promoter is active in skeletal muscle based on the detection of the Purkinje isoform-specific transcripts in normal heart and skeletal muscles (10,27). To determine whether the 1.1 kb of the mouse dystrophin Purkinje promoter has any potential for muscle activity in vivo, we constructed a lacZ transgene under the regulation of this promoter (BE5 construct, Fig. 3A). This construct was microinjected into one-cell mouse embryos (see Materials and Methods) and three founder mice were obtained, which harbored the transgene. These mice were sacrificed and various tissues assessed for lacZ activity. In all three founder mice, we observed a Purkinje cell-specific b-galactosidase staining in the cerebellum (summarized in Table 1; Fig. 3B and C). However, examination of other tissues, notably skeletal and cardiac muscles, revealed an absence of any b-galactosidase staining (data not


2855

Figure 1. The CP promoter sequence from within the mouse dystrophin gene. The 1.1 kb region just upstream of the start (designated with an arrow) of the dystrophin P transcript was used in subsequent expression constructs. The shaded area highlights an imperfect TATA box and the boxed regions depict potential E-box elements that might serve as sites for MyoD binding (27).

Figure 2. Mouse dystrophin muscle enhancer activates transcription from the mouse dystrophin Purkinje promoter in G8 myogenic cells. The mDME-1containing fragment of 3 kb was inserted upstream of the mouse 1.1 kb dystrophin CP promoter driving the expression of the firefly luciferase gene. Constructs DE3 (no enhancer) and DE4 (with 3 kb enhancer) were transfected into mouse skeletal muscle-derived G8 myoblasts (G8 mb). Co-transfection with a plasmid expressing b-galactosidase served as a control for transfection efficiency. Myoblasts were harvested after 24 h or induced to differentiate into mature myotubes (G8 mt) over 4 days. Cells were harvested and extracts assayed for luciferase and b-galactosidase activities. Activation of transcription from dystrophin Purkinje promoter sequences was determined by comparing luciferase activity obtained from the promoter/enhancer with the activity obtained from the dystrophin Purkinje promoter alone, after correction for transfection efficiency. Experiments were carried out in triplicate between two and three times. The values obtained are expressed as mean+SD.

shown). Thus, the mouse dystrophin Purkinje promoter used in this study is not capable of imparting expression to muscle in vivo when used on its own. The mDME-1 selectively activates the dystrophin Purkinje promoter in skeletal muscle, but not in cardiac muscle, of transgenic mice As our experiments with the mouse dystrophin Purkinje promoter alone indicated that it was not sufficient to impart muscle expression in transgenic mice, we tested whether presence of the muscle enhancer, mDME-1, would trigger

activation in muscle. Previously, we had demonstrated that mDME-1 could in fact target expression of the dystrophin muscle promoter in skeletal and cardiac muscles of transgenic mice (24). Therefore, the 3 kb fragment containing the mDME1 was cloned downstream of the dystrophin Purkinje promoter in the context of the lacZ reporter gene (BE4 construct, Fig. 4A). This construct was microinjected into one-cell mouse embryos (see Materials and Methods) and eight founder mice were obtained, which harbored the transgene. Five of the eight founder mice were sacrificed at 1 month of age and the remaining three were bred to establish independent transgenic lines. Various tissues were assessed for lacZ activity.

2856


#288), the lacZ expression in muscle was weaker and was limited to biceps femoral and tibialis anterior (data not shown). In contrast to the relatively high rate of expression of the BE4 transgene in skeletal muscle, it was found to be inactive in all regions of the heart (left and right ventricles, septum and left and right atria) examined from all eight different transgenic strains (summarized in Table 2; Fig. 5E and F). Thus, although the dystrophin Purkinje promoter was promiscuous for expression in skeletal muscle in the presence of mDME-1, it continued to be inactive in cardiac tissue. Developmental profile of BE4 expression in skeletal muscle of transgenic mice

Figure 3. The mouse dystrophin Purkinje promoter directs expression of a lacZ reporter specifically in the cerebellar Purkinje cell layer of transgenic mice. (A) Schematic representation of the BE5 transgene construct in which lacZ expression is driven by the mouse dystrophin Purkinje promoter (yellow box) as indicated by the arrow. Transgenic mice harboring this construct were generated and characterized for lacZ activity. (B and C) Photomicrographs showing coronal sections through the cerebellum of a 1-monthold non-transgenic wild-type (wt) mouse (B) and a 1-month-old transgenic mouse (C, line #171) after b-galactosidase staining. Purkinje cells (pc) between the granular layer (gl) and the molecular layer (ml) of the transgenic cerebellum are positive for b-galactosidase staining. The boxed areas represent magnified views of a part of the photomicrographs.

Table 1. Expression of the lacZ gene in transgenic mice that harbor the BE5 construct Lines

Skeletal muscle

Cardiac muscle

Cerebellum

171 (1 m) 182 (1 m) 179 (1 m)

2 2 2

2 2 2

þþ þ þ

þ: Presence and relative intensity of b-galactosidase stained cells; 2: no b-galactosidase staining.

In all eight independent transgenic insertions, we observed a Purkinje cell-specific b-galactosidase staining in the cerebellum (summarized in Table 2; Fig. 4B). This indicates a remarkable fidelity for the dystrophin Purkinje promoter because all 11 out of 11 transgenic preparations (includes both BE3 and BE4 constructs) harboring this promoter displayed a striking cerebellar expression pattern. Of considerable note was our observation that four of the eight BE4 transgenic strains (#386, #283, #291 and #288) expressed the lacZ reporter in hind limb skeletal muscle of 1 month or older mice (Fig. 5). The muscle types examined included biceps femoris, extensor digitorum longus and tibialis anterior. Our histological observations revealed that the intensity of the staining in the skeletal muscle tissues varied between lines (Table 2). From amongst the four BE4 lines that displayed lacZ activity in skeletal muscle, lines #386 and #291 exhibited the more intense b-galactosidase staining (represented in Fig. 5). Coincidently, line #386 also displayed the highest level of b-galactosidase staining in the cerebellar Purkinje cells. In the remaining two BE4 transgenic lines (#283 and

An important question to address was whether the expression of the BE4 construct in transgenic mice came under developmental regulation. We collected E15.5 stage embryos and skeletal muscle from P0, P30 and P60 stage mice from the highest expressing line (#386) and subjected them to b-galactosidase staining. There was no staining observed in the E15.5 stage embryos (data not shown). For the post-natal stages, skeletal muscle from P0 mice was also negative for b-galactosidase activity (Fig. 6A). Only at P30 (Fig. 6C and D) and P60 (Fig. 6B) we were able to detect b-galactosidase staining. This result suggests that the positive effect of the mDME-1 element on the mouse Purkinje promoter is restricted to stages after birth and is not developmentally active. In many cases, the expression pattern in skeletal muscle was mosaic in which some fibers from the gastrocnemius and flexor digitorum longus muscles stained blue and some fibers remained unstained (Fig. 6C and D). Primary cultures of myoblasts from BE4 transgenic mice preferentially express the lacZ reporter gene but the cardiomyocytes do not To determine whether the differential pattern of lacZ expression observed in vivo is maintained once primary cultures are established, we derived myoblasts and cardiomyocytes from the skeletal and cardiac muscles of BE4 transgenic mice (line #386). As expected, the myoblasts cultured from these animals expressed the lacZ transgene as determined by b-galactosidase activity (Fig. 7A). The occasional myotube in differentiating cultures also stained positive for b-galactosidase (Fig. 7B). Interestingly, not all myogenic cells expressed the transgene, paralleling our observations in the intact skeletal muscles of the transgenic mice (discussed earlier). In contrast to the skeletal muscle-derived myogenic cells, ventricular myocytes harboring the BE4 transgene did not display any b-galactosidase activity (Fig. 7C), once again faithfully representing the in vivo situation.

DISCUSSION Here, we have demonstrated that the region spanning nucleotides 21100 to þ1 of the mouse dystrophin CP promoter is relatively inactive in a skeletal muscle-derived cell line. Similar to the human promoter studies (26), the mouse dystrophin CP promoter was induced to express in skeletal


2857

Figure 4. The mouse dystrophin muscle enhancer (mDME-1) does not compromise the activity of the Purkinje promoter with regard to its cerebellar Purkinje cell layer expression in transgenic mice. (A) Schematic representation of the BE4 transgene construct in which lacZ expression is driven by a cartridge that contains the mouse dystrophin Purkinje promoter (yellow box) and the 3 kb muscle enhancer containing fragment (blue box). Transgenic mice harboring this construct were generated and characterized for lacZ activity. (B) Photomicrograph showing coronal sections through the cerebellum of a 1-month-old transgenic mouse (line #386) after b-galactosidase staining. The section was counterstained with nuclear fast red. Purkinje cells (pc) are positive for b-galactosidase staining and are readily observed between the granular (gl) and the molecular (ml) layers. Magnification, 10.

muscle cells after addition of mDME-1 to the transgene construct. The activation of the promoter was observed in both myoblasts and myotubes in culture. Thus, the mouse and human dystrophin CP promoters respond in a similar fashion to the presence of dystrophin muscle enhancers. An earlier study has described the presence of several putative musclespecific regulatory motifs within the human dystrophin CP promoter (26) and raised the possibility that these elements may contribute to dystrophin transcription in skeletal muscle. However, the collective data imply that musclespecific regulatory elements are not present in the dystrophin CP proximal promoter region itself but that this promoter is responsive to activation in muscle cells if a dystrophin muscle enhancer is brought in close proximity. This capability of the dystrophin CP promoter to respond to the muscle enhancer sequence is restricted to skeletal muscle-derived cells. In the present study, we have also provided evidence that the 1.1 kb mouse dystrophin CP promoter can faithfully drive expression of a lacZ reporter gene to the Purkinje cell layer of the cerebellum of transgenic mice. Despite the presence of putative E-box sequence elements within the mouse dystrophin CP promoter region used, we did not observe any

lacZ activity in skeletal or cardiac muscles of these mice. This is similar to studies in humans that showed that the dystrophin CP promoter is predominantly active in cerebellar Purkinje cells and was subject to developmental regulation (10,28). However, studies on the expression of the human dystrophin CP isoform in other cell types have yielded mixed results. In one study, there was no detectable expression of the dystrophin CP isoform in skeletal muscle and heart of normal individuals (16), whereas in other studies, CP isoform transcripts were detected in both skeletal and cardiac muscles (10,27). Our results also show that the mDME-1 placed adjacent to the mouse dystrophin CP promoter altered the expression profile to yield transgenic mice with lacZ expression in the Purkinje cell layer and also in skeletal muscle, but not in cardiac muscle. Primary myoblasts from these mice expressed the lacZ transgene, whereas cardiomyocytes did not. Taken together, our results in transgenic mice essentially recapitulate what has been observed on the dystrophin CP promoter in cell culture experiments—that the CP promoter alone is relatively inactive in muscle cells, but can be induced by mDME-1 to express in skeletal muscle, but not in cardiac muscle.

2858


Table 2. Expression of the lacZ gene in transgenic mice that harbor the BE4 construct Lines 5645 nb 1m 2m 390 nb 1m 2m 386 nb 1m 2m 283 1m 291 1m 288 1m 278 1m 282 1m

Skeletal muscle

Cardiac muscle

Cerebellum

2 2 2

2 2 2

þ þ þ

2 2 2

2 2 2

þ þ þ

2 þþ þþ

2 2 2

þ þþ þþ

þ

2

þ

þþ

2

þ

þ

2

þ

2

2

þ

2

2

þ

þ: Presence and relative intensity of b-galactosidase stained cells; 2: no b-galactosidase staining; nb: new born mice.

patients are spared from disease, but the hearts are not. One possibility is that XLCM patients upregulate their dystrophin B and CP promoters in trans via some form of transcriptional enhancement. The best candidate for this activity is the one well-characterized transcriptional enhancer, DME-1 (21,22,26). This element has already been suggested to play a role in the activation of the non-muscle isoforms in the skeletal muscles of XLCM patients (26). We have taken this a step further by demonstrating through transgenesis that the mouse counterpart of the DME-1 is capable of activating the dystrophin CP promoter in skeletal muscle, but not in cardiac muscle, in vivo. Our findings, therefore, provide support for the model in which muscle enhancers, like the DME-1 sequence, activate the dystrophin B and CP promoters in the skeletal muscle, but not in cardiac muscle, of XLCM patients. Clearly, whether DME-1 acts alone or in conjunction with other muscle enhancers from within the dystrophin gene, like the recently identified DME-2 (30), in this compensatory activation remains to be determined. Nonetheless, these studies provide us with a better understanding of how the dystrophin gene is regulated, and should prove useful in the development of molecular therapies targeted for the treatment of some forms of muscular dystrophies caused by mutations in dystrophin.

MATERIALS AND METHODS Luciferase constructs

The muscle expression pattern observed in our transgenic mice displayed several important features. First, expression was restricted to skeletal muscle of adult mice and was not observed in embryonic or neonatal muscles. Second, expression was not induced in cardiac muscle, in contrast to the effect mDME-1 had on the dystrophin M promoter in transgenic mice in which lacZ reporter activity was observed in hearts (24). Third, the mosaic nature of the lacZ expression pattern in skeletal muscle suggests that the mosaic expression pattern is a reflection of the choice of the reporter gene and the position-effect of the site of integration of the transgene. A similar mosaic pattern of expression was seen in primary cultures of myoblasts derived from these transgenic mice. Admittedly, it is always problematic when comparing mouse studies with human disease. Differences in gene regulation between species may be present and might complicate this comparison. Nevertheless, there are intriguing parallels that can be drawn from our transgenic mouse data and the subset of patients with XLCM. In these patients, the corresponding mutations in the dystrophin gene have been identified. Three separate studies have reported deletions of the genomic region harboring the dystrophin M promoter and muscle exon 1 in XLCM patients (13,14,26), whereas another study has reported a point mutation within a splice site at muscle dystrophin exon 1 (29). All of these mutations lead to a complete loss of expression of the muscle isoform of dystrophin, suggesting that abolishing of this isoform may represent the major event switching the dystrophin B and CP isoform upregulation. Whatever this as yet undefined compensatory mechanism is, there is upregulation of dystrophin B and CP isoform expression in skeletal muscle, but not in cardiac muscle, of XLCM patients (15–17,29). It is through this compensatory mechanism that skeletal muscles of XLCM

The pGL3 vector (Promega) in which the firefly luciferase gene is present was used to build our recombinant plasmids. To generate the DE3 construct (Fig. 2), a 1.1 kb Bgl II fragment that contains the mouse dystrophin Purkinje promoter was inserted into the Bgl II site of the pGL3B vector (Promega). The mouse dystrophin Purkinje promoter was obtained by Bgl II cleavage of a fragment that was amplified by PCR from mouse genomic DNA. To obtain the DE4 construct (Fig. 2), a 3.0 kb Sac I–Xho I fragment harboring mDME-1 was excised from a construct previously described (23) and inserted into the DE3 construct between the Sac I and Xho I sites. LacZ constructs To generate the BE4 construct, the C1 construct of a previous study (23) was cleaved with Hind III to release a small fragment that contains the SV40 promoter. The larger fragment containing the rest of the plasmid was self-ligated, and the resulting construct was digested with Hind III and Bam HI to release the luciferase gene. The 3.0 kb fragment that was missing the luciferase gene, but still harbored the dystrophin muscle enhancer, was ligated to a Hind III – Bam HI fragment that contains the lacZ gene followed by the SV40 late polyadenylation signal. Finally, we linearized the resulting construct at the Hind III site and ligated to it a Hind III fragment containing 1.1 kb of the mouse dystrophin Purkinje promoter to obtain the recombinant plasmid BE4. A pGL3B vector (Promega) was used to generate the BE5 construct. The vector was cleaved with Hind III and Sal I. A fragment of the pGL3B without the luciferase gene was ligated to a Hind III– Sal I fragment of 3.3 kb that contains the lacZ gene followed by the SV40 late polyadenylation signal. Following


2859

Figure 5. The mouse dystrophin muscle enhancer selectively activates the dystrophin Purkinje promoter in skeletal muscle, but not in cardiac muscle, of BE4 transgenic mice. (A) A digital photograph showing in ventral view a comparative macroscopical observation of the left hind limbs from 2-month-old non-transgenic wild-type (left) and transgenic mice from line #386 (right). Note the intense b-galactosidase staining observed in the different muscles of the transgenic preparation. (B) A digital photograph of a teased extensor digitorum longus muscle from 1-month-old wild-type (top) and transgenic mice from line #291 (bottom). Once again, the b-galactosidase staining in the transgenic muscle is readily apparent. (C) Photomicrograph of a longitudinal section taken from the thigh muscle of a 1-month-old transgenic mouse (line #386) showing intense b-galactosidase staining. Magnification, 10. (D) Photomicrograph of a longitudinal section taken from the biceps femoris muscle of a 1-month-old transgenic mouse (line #291) showing strong b-galactosidase staining. Magnification, 4. (E) Photomicrograph of a histological section taken from the heart of a BE4 transgenic mouse (line #386). Note the complete absence of lacZ activity in every region of the heart. ra ¼ right atrium, la ¼ left atrium, rv ¼ right ventricle, lv ¼ left ventricle. Magnification, 2.5. (F) Photomicrograph demonstrating that the BE4 transgene is not activated in the cardiac muscle of a second transgenic line (#291). Magnification, 10.

that, the construct was cleaved using the enzyme HindIII and ligated to a Hind III fragment that was obtained by PCR and that contains 1.1 kb of the mouse dystrophin Purkinje promoter. Transfections and biochemical assays Transfection of recombinant plasmids into myogenic G8 cells was performed by using liposomes as described previously (23). Plasmid DNA, including pGKb-galactosidase, was incubated for 10 min with DODAC – DOPE liposomes (Inex Pharmaceuticals, Inc.) in 0.9% NaCl. Subsequent to this, a-minimal

Eagle’s medium with 25% fetal bovine serum was added to the solution and the whole mixture was added to 2 105 G8 cells and left to incubate for 3 h at 378C. Following that, the mix was discarded and replaced with fresh medium containing 10% fetal bovine serum in which 10% horse serum was added, and the G8 cells were incubated for 20 h at 378C. One group of myoblasts were collected after a period of 24 h in luciferase lysis buffer and another group was induced to differentiate into mature myotubes in medium containing 1% horse serum for .4 days and then harvested in luciferase lysis buffer. Protein assays, b-galactosidase activity and luciferase activities were measured as described previously (23).

2860


Figure 6. The positive effect of the dystrophin muscle enhancer on the activation of the dystrophin Purkinje promoter is most apparent in 1 month or older mice. BE4 transgenic mice from line #386 were analyzed at P0 (A), P30 (C and D) and P60 (B) for b-galactosidase staining in their skeletal muscles. (A) Photomicrograph of a cross-section through the hind limb demonstrates that the lacZ transgene is not active at this stage. f ¼ fibula, t ¼ tibia. Magnification, 2.5. (B and C) Photomicrographs of cross-sections through the gastrocnemius muscle demonstrate b-galactosidase staining at P60 and P30. A mosaic expression pattern in which a considerable number of fibers are lacZ positive was noted. Magnification, 10 for (B) and 4 for (C). (D) This mosaic nature of the transgene is readily observed in longitudinal section where some fibers are clearly positive for b-galactosidase staining, whereas others remain unstained. Magnification, 20.

Transgenic mice To generate transgenic mice, we used the hybrid C57BL/ 6 –C3H F1 mice (produced by crossing C57BL/6 female mice with C3H male mice, obtained from Charles River) as donors for fertilized one-cell embryos. Pronuclear microinjection of BE4 and BE5 constructs was performed at a concentration of 3 ng/ml. Zygotes were cultured overnight at 378C in M16 medium under oil. The following day, two-cell stage embryos were subjected to oviduct transfers in pseudopregnant female CD-1 mice. Transgenic BE4 and BE5 mice were identified by PCR amplification of genomic tail DNA. Care and use of experimental mice followed the guidelines established by the Canadian Council on Animal Care (CCAC).

Histological analysis Transgenic and wild-type mice at birth, 1 month (P30) and 2 months (P60) of age were anesthetized by intraperitoneal

injection of tribromoethanol (avertin). Each mouse was perfused transcardially with 5– 10 ml of PBS followed by 10 – 20 ml of 0.2% glutaraldehyde fixative (1.25 mM EGTA pH 7.3, 2 mM MgCl2 in PBS). This process was particularly important to obtain efficient b-galactosidase staining in adult mice. Heart, skeletal muscle, brain and cerebellum were collected from these mice. Tissues were kept in the fix for 4 – 5 h at 48C with agitation using a rotary shaker, followed by three washes in PBS for 15 min each at 48C. Samples were incubated in lacZ stain [1 mg/ml X-gal (5-bromo-4chloro-3-indolyl-b-D -galactoside) in N,N-dimethylformamide, 5 mM K3Fe(CN)6, 5 mM K4Fe(CN)6.H2O, 2 mM MgCl2, 0.02% NP-40 in 0.1 M phosphate buffer, pH 8.0] overnight at 308C. The following day, the samples were washed three times in PBS for 5 min each, and post-fixed in 0.2% glutaraldehyde (1.25 mM EGTA pH 7.3, 2 mM MgCl2 in PBS). The samples were processed with automated tissue processor and embedded into paraffin. The sections were cut at 20 mm thickness, deparaffinized in toluene and rehydrated in alcohol gradient followed by deionized water. The sections were


2861

Figure 7. BE4 transgenic mouse-derived primary myoblast cultures preferentially express the lacZ reporter gene in contrast to the primary cardiomyocytes. (A) Photomicrograph showing the presence of b-galactosidase activity in primary myoblasts derived from the hind limb skeletal muscle of BE4 mice (line #386), Magnification, 20. (B) The occasional myotube in differentiating cultures also displayed b-galactosidase activity, Magnification, 20. (C) Photomicrograph of primary ventricular myocytes on a bed of fibroblasts derived from the hearts of BE4 mice (line #386). b-galactosidase activity was never detected in these cultures, Magnification, 32.

counterstained with nuclear fast red for 1 min, washed for 5 min with water, dehydrated in alcohol gradient and finally cleared with toluene and cover-slipped using Permount (Fisher). Sections were examined under a stereomicroscope or by light microscopy using a Zeiss Axioplan microscope and photographed. Isolation of primary myoblasts and cardiomyocytes Satellite cell-derived primary myoblasts were isolated from the hind limb muscles of 9-day-old stage BE4 transgenic pups (line #386) essentially as described before (31). When cells were confluent, they were subjected to b-galactosidase staining. Some of the primary myoblasts were induced to differentiate by replacing the growth medium with a low serum medium (1% fetal horse serum). Primary ventricular cardiomyocytes were freshly isolated from BE4 transgenic P4 neonate hearts (ventricles only) and maintained as previously described (32). Beating myocytes were observed within 2 –3 days of culture at which point they were analyzed for b-galactosidase activity. For assessing b-galactosidase activity in situ, cells were fixed in 0.2% glutaraldehyde for 5 min. Following rinsing in PBS, the cells were incubated with lacZ stain (discussed earlier) overnight at 378C and photographed using a Zeiss Axioplan microscope or a Zeiss Axiovert S100 inverted microscope.

ACKNOWLEDGEMENTS We are grateful to Dr Dennis Bulman for critical reading of the manuscript, and the rest of the Kothary laboratory for helpful discussions. We thank Danica Albert and Carrie Anderson for help in maintaining the transgenic mouse colonies, Jennifer Knudson for help in establishing cardiomyocyte primary cultures and Louise Pelletier for preparation of tissue sections. Thanks to the Jesse’s Journey Foundation for Gene and Cell Therapy for their generous support of our research program. This project was funded by grants from the Canadian Institutes of Health Research and the Muscular Dystrophy Association (USA) to R.K., and from the Canadian Genetic Diseases Network and the Heart and Stroke Foundation of Ontario to R.G.W.

REFERENCES 1. Emery, A.E. (1993) Duchenne Muscular Dystrophy. Oxford University Press, Oxford. 2. Worton, R.G., Molnar, M.J., Brais, B. and Karpati, G. (2000) The muscular dystrophies. In Scriver, C.R., Beaudet, A.L., Valle, D., Sly, W.S., Childs, B., Kinzlew, K.W. and Vogelstein, B. (eds.), The Metabolic and Molecular Basis of Inherited Diseases, 8th edn. McGraw-Hill, New York. 3. Burghes, A.H., Logan, C., Hu, X., Belfall, B., Worton, R.G. and Ray, P.N. (1987) A cDNA clone from the Duchenne/Becker muscular dystrophy gene. Nature, 328, 434– 437.

2862


4. Koenig, M., Hoffman, E.P., Bertelson, C.J., Monaco, A.P., Feener, C. and Kunkel, L.M. (1987) Complete cloning of the Duchenne muscular dystrophy (DMD) cDNA and preliminary genomic organization of the DMD gene in normal and affected individuals. Cell, 50, 509–517. 5. Sadoulet-Puccio, H.M. and Kunkel, L.M. (1996) Dystrophin and its isoforms. Brain Pathol., 6, 25–35. 6. Hoffman, E.P., Brown, R.H., Jr and Kunkel, L.M. (1987) Dystrophin: the protein product of the Duchenne muscular dystrophy locus. Cell, 51, 919 –928. 7. Koenig, M., Monaco, A.P. and Kunkel, L.M. (1988) The complete sequence of dystrophin predicts a rod-shaped cytoskeletal protein. Cell, 53, 219 –226. 8. Boyce, F.M., Beggs, A.H., Feener, C. and Kunkel, L.M. (1991) Dystrophin is transcribed in brain from a distant upstream promoter. Proc. Natl Acad. Sci. USA, 88, 1276–1280. 9. Gorecki, D.C., Monaco, A.P., Derry, J.M., Walker, A.P., Barnard, E.A. and Barnard, P.J. (1992) Expression of four alternative dystrophin transcripts in brain regions regulated by different promoters. Hum. Mol. Genet., 1, 505– 510. 10. Holder, E., Maeda, M. and Bies, R.D. (1996) Expression and regulation of the dystrophin Purkinje promoter in human skeletal muscle, heart, and brain. Hum. Genet., 97, 232–239. 11. Berko, B.A. and Swift, M. (1987) X-linked dilated cardiomyopathy. N. Engl. J. Med., 316, 1186–1191. 12. Towbin, J.A., Hejtmancik, J.F., Brink, P., Gelb, B., Zhu, X.M., Chamberlain, J.S., McCabe, E.R. and Swift, M. (1993) X-linked dilated cardiomyopathy. Molecular genetic evidence of linkage to the Duchenne muscular dystrophy (dystrophin) gene at the Xp21 locus. Circulation, 87, 1854–1865. 13. Muntoni, F., Cau, M., Ganau, A., Congiu, R., Arvedi, G., Mateddu, A., Marrosu, M.G., Cianchetti, C., Realdi, G., Cao, A. et al. (1993) Brief report: deletion of the dystrophin muscle-promoter region associated with X-linked dilated cardiomyopathy. N. Engl. J. Med., 329, 921–925. 14. Yoshida, K., Ikeda, S., Nakamura, A., Kagoshima, M., Takeda, S., Shoji, S. and Yanagisawa, N. (1993) Molecular analysis of the Duchenne muscular dystrophy gene in patients with Becker muscular dystrophy presenting with dilated cardiomyopathy. Muscle Nerve, 16, 1161–1166. 15. Nakamura, A., Ikeda, S., Yazaki, M., Yoshida, K., Kobayashi, O., Yanagisawa, N. and Takeda, S. (1997) Up-regulation of the brain and Purkinje-cell forms of dystrophin transcripts, in Becker muscular dystrophy. Am. J. Hum. Genet., 60, 1555–1558. 16. Muntoni, F., Melis, M.A., Ganau, A. and Dubowitz, V. (1995) Transcription of the dystrophin gene in normal tissues and in skeletal muscle of a family with X-linked dilated cardiomyopathy. Am. J. Hum. Genet., 56, 151 –157. 17. Muntoni, F., Wilson, L., Marrosu, G., Marrosu, M.G., Cianchetti, C., Mestroni, L., Ganau, A., Dubowitz, V. and Sewry, C. (1995) A mutation in the dystrophin gene selectively affecting dystrophin expression in the heart. J. Clin. Invest., 96, 693–699. 18. Klamut, H.J., Gangopadhyay, S.B., Worton, R.G. and Ray, P.N. (1990) Molecular and functional analysis of the muscle-specific promoter region of the Duchenne muscular dystrophy gene. Mol. Cell. Biol., 10, 193 –205.

19. Galvagni, F., Lestingi, M., Cartocci, E. and Oliviero, S. (1997) Serum response factor and protein-mediated DNA bending contribute to transcription of the dystrophin muscle-specific promoter. Mol. Cell. Biol., 17, 1731–1743. 20. Galvagni, F., Cartocci, E. and Oliviero, S. (1998) The dystrophin promoter is negatively regulated by YY1 in undifferentiated muscle cells. J. Biol. Chem., 273, 33708–33713. 21. Klamut, H.J., Bosnoyan-Collins, L.O., Worton, R.G., Ray, P.N. and Davis, H.L. (1996) Identification of a transcriptional enhancer within muscle intron 1 of the human dystrophin gene. Hum. Mol. Genet., 5, 1599–1606. 22. Klamut, H.J., Bosnoyan-Collins, L.O., Worton, R.G. and Ray, P.N. (1997) A muscle-specific enhancer within intron 1 of the human dystrophin gene is functionally dependent on single MEF-1/E box and MEF-2/AT-rich sequence motifs. Nucl. Acids Res., 25, 1618– 1625. 23. Marshall, P., Chartrand, N. and Worton, R.G. (2001) The mouse dystrophin enhancer is regulated by MyoD, E-box-binding factors, and by the serum response factor. J. Biol. Chem., 276, 20719–20726. 24. Marshall, P., Chartrand, N., De Repentigny, Y., Kothary, R., Pelletier, L., Mueller, R. and Worton, R.G. (2002) Mouse dystrophin enhancer preferentially targets lacZ expression in skeletal and cardiac muscle. Dev. Dyn., 224, 30–38. 25. Kimura, S., Abe, K., Suzuki, M., Ogawa, M., Yoshioka, K., Kaname, T., Miike, T. and Yamamura, K. (1997) A 900 bp genomic region from the mouse dystrophin promoter directs lacZ reporter expression only to the right heart of transgenic mice. Dev. Growth Differ., 39, 257–265. 26. Bastianutto, C., Bestard, J.A., Lahnakoski, K., Broere, D., De Visser, M., Zaccolo, M., Pozzan, T., Ferlini, A., Muntoni, F., Patarnello, T. et al. (2001) Dystrophin muscle enhancer 1 is implicated in the activation of non-muscle isoforms in the skeletal muscle of patients with X-linked dilated cardiomyopathy. Hum. Mol. Genet., 10, 2627–2635. 27. Abdulrazzak, H., Noro, N., Simons, J.P., Goldspink, G., Barnard, E.A. and Gorecki, D.C. (2001) Structural diversity despite strong evolutionary conservation in the 50 -untranslated region of the P-type dystrophin transcript. Mol. Cell. Neurosci., 17, 500–513. 28. Torelli, S., Ferlini, A., Obici, L., Sewry, C. and Muntoni, F. (1999) Expression, regulation and localisation of dystrophin isoforms in human foetal skeletal and cardiac muscle. Neuromuscul. Disord., 9, 541–551. 29. Milasin, J., Muntoni, F., Severini, G.M., Bartoloni, L., Vatta, M., Krajinovic, M., Mateddu, A., Angelini, C., Camerini, F., Falaschi, A. et al. (1996) A point mutation in the 50 splice site of the dystrophin gene first intron responsible for X-linked dilated cardiomyopathy. Hum. Mol. Genet., 5, 73 –79. 30. Bastianutto, C., De Visser, M., Muntoni, F., Klamut, H.J. and Patarnello, T. (2002) A novel muscle-specific enhancer identified within the deletion overlap region of two XLDC patients lacking muscle exon 1 of the human dystrophin gene. Genomics, 80, 614 –620. 31. Boudreau-Larivie`re, C. and Kothary, R. (2002) Differentiation potential of primary myogenic cells derived from skeletal muscle of dystonia musculorum mice. Differentiation, 70, 247–256. 32. Argentin, S., Ardati, A., Tremblay, S., Lihrmann, I., Robitaille, L., Drouin, J. and Nemer, M. (1994) Developmental stage-specific regulation of atrial natriuretic factor gene transcription in cardiac cells. Mol. Cell. Biol., 14, 777–790.