- Email: [email protected]
Differential expression and developmental regulation of a novel α-dystrobrevin isoform in muscle
Gene 238 (1999) 479–488 www.elsevier.com/locate/gene
Differential expression and developmental regulation of a novel a-dystrobrevin isoform in muscle k Rebecca E. Enigk, Margaret M. Maimone * Department of Anatomy and Cell Biology, State University of New York Health Science Center, 750 East Adams Street, Syracuse, NY 13210, USA Received 20 April 1999; received in revised form 15 July 1999; accepted 6 August 1999; Received by A.J. van Wijnen
Abstract a-Dystrobrevin is a dystrophin-related protein expressed primarily in skeletal muscle, heart, lung and brain. In skeletal muscle, a-dystrobrevin is a component of the dystrophin-associated glycoprotein complex and is localized to the sarcolemma, presumably through interactions with dystrophin and utrophin. Alternative splicing of the a-dystrobrevin gene generates multiple isoforms which have been grouped into three major classes: a-DB1, a-DB2, and a-DB3. Various isoforms have been shown to interact with a variety of proteins; however, the physiological function of the a-dystrobrevins remains unknown. In the present study, we have cloned a novel a-dystrobrevin cDNA encoding a protein (referred to as a-DB2b) with a unique 11 amino acid C-terminal tail. Using RT–PCR with primers specific to the new isoform, we have characterized its expression in skeletal muscle, heart, and brain, and in differentiating C2C12 muscle cells. We show that a-DB2b is expressed in skeletal muscle, heart and brain, and that exons 12 and 13 are alternatively spliced in a-DB2b to generate at least three splice variants. The major a-DB2b splice variant expressed in adult skeletal muscle and heart contains exons 12 and 13, while in adult brain, two a-DB2b splice variants are expressed at similar levels. This is consistent with the preferential expression of exons 12 and 13 in other a-dystrobrevin isoforms in skeletal muscle and heart. Similarly, in a-DB1 the first 21 nucleotides of exon 18 are preferentially expressed in skeletal muscle and heart relative to brain. We also show that the expression of alternatively spliced a-DB2b is developmentally regulated in muscle; during differentiation of C2C12 cells, a-DB2b expression switches from an isoform lacking exons 12 and 13 to one containing them. We demonstrate similar developmental upregulation of exons 12, 13, and 18 in a-DB1 and of exons 12 and 13 in a-DB2a. Finally, we show that a-DB2b protein is expressed in adult skeletal muscle, suggesting that it has a functional role in adult muscle. Together, these data suggest that alternatively spliced variants of the new a-dystrobrevin isoform, a-DB2b, are differentially expressed in various tissues and developmentally regulated during muscle cell differentiation in a fashion similar to that previously described for a-dystrobrevin isoforms. © 1999 Elsevier Science B.V. All rights reserved. Keywords: Alternative splicing; C2C12 cells; cDNA cloning; Dystrophin associated protein; RT–PCR
1. Introduction Dystrobrevin is a dystrophin-related protein that was originally identified as an acetylcholine receptor (AChR) Abbreviations: Ab, antibody; AChR, acetylcholine receptor; a-DB, a-dystrobrevin; cDNA, DNA complementary to mRNA; DMEM, Dulbecco’s Modified Eagle Medium; dNTP, deoxyribonucleoside triphosphate; DTT, dithiothreitol; EtdBr, ethidium bromide; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; nt, nucleotide(s); PAGE, polyacrylamide gel electrophoresis; PCR, polymerase chain reaction; RT–PCR, reverse transcriptase polymerase chain reaction; u, unit(s); UTR, untranslated region; vr, variable region. k The nucleotide sequences reported in this paper have been submitted to GenBank@ with accession numbers: AF143542, AF143543, and AF143544. * Corresponding author. Tel.: +1-315-464-8526; fax: +1-315-464-8535. E-mail address: [email protected] (M.M. Maimone)
-associated protein present in postsynaptic membranes of Torpedo electric organ, and chick and rat endplates (Carr et al., 1989). Cloning of a-dystrobrevin from Torpedo ( Wagner et al., 1993), human (Sadoulet-Puccio et al., 1996), and mouse (Blake et al., 1996) revealed multiple transcripts generated by alternative splicing of the a-dystrobrevin gene (Ambrose et al., 1997; SadouletPuccio et al., 1997a). More recently, a second gene encoding a closely related protein, b-dystrobrevin, was identified in mouse and human (Peters et al., 1997b; Blake et al., 1998; Puca et al., 1998) and shown to be expressed in a broader range of tissues than a-dystrobrevin. The mouse a-dystrobrevin gene consists of 24 coding exons (Ambrose et al., 1997). The alternative usage of three exons containing stop codons, exons 24, 18, and 11, generates transcripts encoding three major classes of
0378-1119/99/$ – see front matter © 1999 Elsevier Science B.V. All rights reserved. PII: S0 3 7 8 -1 1 1 9 ( 9 9 ) 0 0 35 8 - 3
480
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
a-dystrobrevin proteins: a-dystrobrevin1 (a-DB1), a-dystrobrevin2 (a-DB2), and a-dystrobrevin3 (a-DB3), respectively (Ambrose et al., 1997; Nawrotzki et al., 1998). Alternative splicing within the transcripts encoding a-DB1 and a-DB2 at three different sites generates additional variability within the a-dystrobrevin family of proteins. Variable region 1 (vr1), encoded by exon 9, is predominantly expressed in brain, while exons 12 and 13 of variable region 3 (vr3) are predominantly expressed in skeletal muscle (Blake et al., 1996). More recently, the expression of exons 12 and 13 was shown to increase during myotube differentiation in the mouse muscle cell line H-2Kb-tsA58 (Holzfeind et al., 1999). While it is clear that multiple isoforms of a-dystrobrevin are expressed in skeletal muscle, the physiological function of the various isoforms remains unknown. Early studies in skeletal muscle showed that a-dystrobrevin was localized to the sarcolemma and neuromuscular junction (Carr et al., 1989) and that it was a component of the dystrophin-associated protein complex, presumably through its interactions with dystrophin, utrophin and/or syntrophins (Butler et al., 1992; Ahn and Kunkel, 1995; Dwyer and Froehner, 1995; Suzuki et al., 1995; Peters et al., 1997a; SadouletPuccio et al., 1997b). Subsequently, immunofluorescence localization studies with isoform-specific antibodies showed that a-DB2 is localized to the sarcolemma and neuromuscular junction, whereas a-DB1 is more restricted to the neuromuscular junction (Balasubramanian et al., 1998; Peters et al., 1998). In addition, a-DB1 and a-DB2 differ in their ability to interact with other proteins; a-DB2 binds dystrophin, whereas a-DB1 binds both dystrophin and utrophin (Peters et al., 1998) and associates more tightly with syntrophin than does a-DB2 (Balasubramanian et al., 1998). a-DB1 contains two major sites of tyrosine phosphorylation in its unique C-terminal tail (Balasubramanian et al., 1998), suggesting that its activity at the neuromuscular junction may be modulated by tyrosine phosphorylation. However, a-DB1 is not phosphorylated in response to treatment with agrin (Nawrotzki et al., 1998), the nerve-derived signaling molecule that initiates the formation of the neuromuscular junction through tyrosine phosphorylation (Glass and Yancopoulos, 1997; Colledge and Froehner, 1998). These findings suggest that alternative splicing generates multiple a-dystrobrevin isoforms that differentially localize in skeletal muscle and that potentially have different physiological functions. In the present study, we have cloned a novel splice variant of a-DB2 from a mouse muscle cell line library. The new isoform, referred to as a-DB2b, contains a unique 11 amino acid C-terminal tail encoded by a novel 3∞ exon, referred to as exon Z, that is spliced into the transcript after the first 21 nt of exon 18. Using RT– PCR and primers specific for the new isoform, we show that a-DB2b is expressed in skeletal muscle, heart, and
brain, and that it is alternatively spliced at vr3 to generate at least three isoforms. We demonstrate that the expression of a-DB2b and its alternative splicing at vr3 are both developmentally regulated in the C2C12 muscle cell line, as are other a-dystrobrevin isoforms. In addition, we show that a polyclonal antibody specific for a-DB2b recognizes a protein of the expected molecular weight in skeletal muscle. Taken together, our data provide support for a general principle that alternatively spliced variants of a-dystrobrevin are differentially expressed in tissues and developmentally regulated in muscle cells.
2. Materials and methods 2.1. Cloning and sequencing To isolate a-dystrobrevin cDNA clones, a BC3H1 mouse muscle cell line cDNA library ( Frail et al., 1988) was screened with a full-length Torpedo dystrobrevin cDNA ( Wagner et al., 1993) probe using the Genius System (Boehringer Mannheim, Indianapolis, IN ), and 19 positive clones were analyzed by restriction mapping and partial sequencing. Two clones, 11.2 and 9.2.3, which contained consensus polyA addition signals near their 3∞ ends were subcloned into pBluescriptII(+) (Stratagene, La Jolla, CA) and completely sequenced using Sequenase ( United States Biochemicals, Cleveland, OH ). Clone 11.2 contained an apparently full-length copy of the mouse a-DB3 mRNA, while clone 9.2.3 encoded part of an a-DB2 isoform with a unique C-terminus compared to the previously published a-DB2 sequence (Blake et al., 1996). We use the terms a-DB2a and a-DB2b to distinguish the previously published and new isoforms, respectively. The BC3H1 library was rescreened with a 550 bp fragment of the longest clone (7.2), and six additional positive clones were isolated and analyzed. One clone (16.1), which was subcloned and sequenced as described above, contained an almost full-length copy of the a-DB2a mRNA including a consensus poly A additional signal near the 3∞ end. Cycle sequencing of the a-DB2b cDNA generated by RT–PCR was performed by the DNA Sequencing Facility of the BioResource Center at Cornell University in Ithaca, NY. 2.2. C2C12 cell culture C2C12 cells ( Yaffe and Saxel, 1977; Blau et al., 1985) were maintained at subconfluence on gelatin (0.2%) -coated tissue culture dishes in growth medium consisting of DMEM (Life Technologies, Gaithersburg, MD) supplemented with 20% fetal bovine serum (Atlanta Biologicals, Norcross, GA), 50 u/ml penicillin and 50 mg/ml streptomycin (Sigma Chemicals, St. Louis,
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
MO). To prepare cells at various stages of differentiation for RNA isolation, myoblasts were grown to confluence (3–4 days) and then the growth medium was replaced with differentiation medium consisting of DMEM supplemented with 5% horse serum (Life Technologies), 50 u/ml penicillin and 50 mg/ml streptomycin. RNA was isolated from cells as described in Section 2.3 at specific time points (0 h, 6 h, 12 h, 24 h, 2 days, 3 days, 4 days, and 5 days) after switching to differentiation medium.
2.3. RNA isolation Total RNA was prepared from mouse tissues and C2C12 cells using TRIzol reagent (Life Technologies) as recommended by the manufacturer. Briefly, hindlimb muscles, hearts, and brains were dissected from 8 week old Balb/CJ mice, frozen in liquid nitrogen, and stored at −70 °C. Frozen tissue was ground in a mortar under liquid nitrogen; 1 ml of TRIzol was added per 50– 100 mg of tissue and the solution was homogenized with a Polytron homogenizer at room temperature. For C2C12 cells, 18 ml of TRIzol was added to each 6 cm dish of cells. These TRIzol-containing solutions were extracted with chloroform and precipitated with isopropyl alcohol. The RNA pellet was then washed with 70% ethanol and resuspended in RNAse-free dH O. RNA 2 integrity was confirmed by electrophoresis of 1 mg on a 1% agarose gel, and the RNA concentration was determined by absorbance at 260 nm.
2.4. RT–PCR
481
Fig. 1. Schematic diagram of a-dystrobrevin isoforms generated by alternative splicing. Four classes of a-dystrobrevin isoforms generated by the usage of four different stop codons are presented, with the new a-DB2b isoform denoted with an asterisk. Large and small boxes represent coding and non-coding regions, respectively, and shaded boxes denote alternatively spliced exons at the three variable regions designated vr1 (exon 9), vr2 (exon 18), and vr3 (exons 11, 12, 13). The approximate positions of primers used for RT–PCR are designated by lower-case letters (refer to Table 1 for primer sequences). Table 1 Oligo primers used for RT–PCR Primera
Sequence
a b c d e f g h i j GAPDH-c GAPDH-nc AChRa-c AChRa-nc
5∞-CGTCGTGGAAATCACCTGC-3∞ 5∞-GTTCCCTCCTCAGGAAGTCC-3∞ 5∞-CCTGGTGATGAAAGGACTTC-3∞ 5∞-CTCCGGCTCCTCAGACAG-3∞ 5∞-GGCAGATGCTGAACGGATG-3∞ 5∞-CACTGTTAGTTAAGACCTGCAG-3∞ 5∞-GGAATTCC/TCATGTTCTCTCCTCAAGG-3∞ 5∞-GTCTTCCATCCAGTTGAGTG-3∞ 5∞-GGAACAGGAAGTCATGTTATCC-3∞ 5∞-TGTGTGCTCTAGATGAACTG-3∞ 5∞-ACCCAGAAGACTGTGGATGG-3∞ 5∞-GATCCACGACGGACACATTG-3∞ 5∞-CGCAGACGGCGACTTTGC-3∞ 5∞-GAAGTAGAGGGGCAGGCG-3∞
a c designates coding primer; nc designates non-coding primer.
First-strand cDNA was synthesized from 2 mg of total RNA using an oligo dT adapter primer (5∞-TGAGCTCGAGTCGACATCGA( T ) V-3∞) and 17 Superscript II Reverse Transcriptase (Life Technologies) as recommended by the manufacturer, and then stored at −70 °C. Each 50 ml PCR reaction contained 2 ml of cDNA, 40 pmol of each primer (see Fig. 1 for location of primers and Table 1 for sequence of primers), 1.5 units of Platinum Taq DNA polymerase (Life Technologies), Platinum Taq PCR buffer (20 mM Tris–HCl, pH 8.4, 50 mM KCl ), 0.2 mM dNTPs, and 1.5 mM MgCl and 2 was overlaid with 2 drops of mineral oil. All reactions were incubated for 2 min at 94°C to activate the Platinum Taq enzyme, and then cycled 29 times through 30 s at 94°C, 30 s at the appropriate annealing temperature and 3 min at 72°C. Note: for the a-DB2b C2C12 time course amplification, the reactions were incubated for 5 min at 94°C, 2 min at 58°C and 40 min at 72°C, and then the reaction was cycled 29 times through 40 s at 94°C, 2 min at 58°C and 3 min at 72 °C, followed by a final incubation of 15 min at 72°C.
2.5. Polyclonal antibody and immunoblotting A polyclonal antibody that recognizes the unique C-terminal tail of aDB2-b (Ab 6224) was generated in rabbits by GeneMed (San Francisco, CA) using the synthetic peptide C-HEIIPLEERT-COOH. Crude serum was used in immunoblots of mouse skeletal muscle fractions prepared according to the method of Kramarcy et al. (1994). Briefly, frozen skeletal muscle tissue (255 mg) from 8 week old Balb/CJ mice was ground under liquid nitrogen with a mortar and pestle, homogenized with a Polytron homogenizer for 30 s in 10 volumes of ice-cold homogenization buffer (10 mM sodium phosphate, 5 mM ethylenediaminetetraacetic acid, 0.4 M NaCl, pH 7.8 and protease inhibitors: aprotinin and leupeptin at 5 mg/ml each, pepstatin A at 0.05 mg/ml, and 2 mM phenylmethylsulfonyl fluoride), and centrifuged for 10 min at 12 000×g. The pellet was resuspended in homogenization buffer and recentrifuged for 10 min at 12 000×g. This pellet was then solubilized
482
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
in homogenization buffer containing 1% Triton X-100 for 15 min on ice and centrifuged for 10 min at 12 000×g. Sample buffer was added to both the supernatant (soluble) and pellet (insoluble) fractions to give a final concentration of 80 mM Tris (pH 6.8), 2% SDS, 15% glycerol, 0.01% bromophenol blue, and 100 mM DTT. Protein concentrations of the samples were determined by the Bio-Rad Protein Assay (Bio-Rad, Hercules, CA.). Two microliters representing 2% of the soluble and insoluble fractions (2.0 mg and 14.5 mg total protein, respectively) were then subjected to SDS–PAGE on a 12% polyacrylamide gel and transferred electrophoretically to nitrocellulose. Immunoblotting was performed according to the manufacturer’s directions for enhanced chemiluminescence detection (Amersham, Piscataway, NJ ). Pre-immune and immune sera were used at a dilution of 1:3000, while the horseradish peroxidase-conjugated goat anti-rabbit IgG (BioRad ) was used at a dilution of 1:6000.
3. Results 3.1. Cloning of a-dystrobrevin isoforms Screening of a mouse muscle cell line cDNA library yielded many a-dystrobrevin clones corresponding to the a-DB2 and a-DB3 isoform classes (Fig. 1) (Ambrose et al., 1997). One clone, 9.2.3, contained a partial cDNA insert that appeared to be in the a-DB2 class based on its length; however, the sequence at the 3∞ end, including the last 13 codons, diverged from the previously published a-DB2 sequence (Blake et al., 1996) after the first 21 nt of exon 18 (Fig. 1). As shown previously for a-DB1 (Ambrose et al., 1997), the GT at nt 22 and 23 of exon 18 can function as a cryptic 5∞ splice donor site for the new exon, which we have designated exon Z since we do not know its location in the gene relative to the other exons. In addition to the highly conserved GT at the 5∞ end of the intron, the cryptic splice site in exon 18, GGA//GTAAGT, conforms to the 5∞ splice site consensus sequence, (A/C )AG//GTRAGT, at six of nine bases. On average, mammalian 5∞ splice sites match the consensus at seven of nine bases (reviewed in Horowitz and Krainer, 1994). To determine whether this partial cDNA clone represented a true mRNA present in muscle, we carried out RT–PCR on total RNA isolated from mouse skeletal muscle using primers
483
in the 5∞-UTR and in the unique 3∞ coding region (primers j and g in Fig. 1) (data not shown). Cycle sequencing of the resulting 1.9 kb PCR product showed that it was identical to the previously reported a-DB2 sequence, except that it lacked exon 9 at vr1, included exons 12 and 13 at vr3, and diverged into unique sequence after nt 21 of exon 18. This indicated that the mRNA represented by clone 9.2.3 was indeed expressed in skeletal muscle and differs from a-DB2 at its 3∞ end. We therefore refer to the new isoform as a-DB2b and the previously described isoform as a-DB2a. Although clone 9.2.3 contained a polyA addition signal near its 3∞ end, its 5∞ end began in the middle of the a-DB2 coding region, so we combined the sequences of clone 9.2.3 and the PCR product to generate the a-DB2b sequence shown in Fig. 2A. The cDNA library screening also yielded clones containing the complete coding sequence and 3∞-UTR for a-DB2a and a-DB3. The sequence of our a-DB2a clone ( Fig. 2A) differed from that previously published (Blake et al., 1996) in that it lacked the brain-specific exon 9, but contained exons 12 and 13. The 5∞-UTR of our a-DB2a cDNA was larger and consisted of exons C and D instead of exons B and D (Ambrose et al., 1997), and the 3∞-UTR of our a-DB2a cDNA was 992 bp longer and included a polyA addition signal near its 3∞ end. In addition, there were two single nucleotide differences; our sequence contained a G instead of C at nt 721, consistent with the sequence of exon 7 of the a-dystrobrevin gene (Ambrose et al., 1997), and four As instead of three beginning at nt 1843. Sequencing of our a-DB3 cDNA ( Fig. 2A) showed that the 5∞-UTR consisted of exons C and D, while the coding region consisted of exons 1–11, excluding exon 9. In addition, the 3∞-UTR was identical to that previously described for a-DB3 (Nawrotzki et al., 1998), except that it was 14 bp longer and had four single nt differences: four Ts instead of three at nt 1167, one A instead of two at nt 1217, C instead of T at nt 1318, and C instead of T at nt 1323. A comparison of the proteins encoded by our three full-length dystrobrevin cDNAs ( Fig. 2B) shows that each has a unique C terminus generated by alternative splicing. 3.2. Expression of a-dystrobrevin isoforms in muscle, heart, and brain To determine whether a-DB2b was expressed in heart and brain as well as skeletal muscle, as are the other
Fig. 2. Sequence analysis of a-DB2b. (A) Nucleotide sequence of a-DB2b compared with a-DB2a and a-DB3. Dots represent sequence derived from the same exons, while upper- and lower-case letters represent coding and non-coding regions, respectively. Vertical lines denote selected exon boundaries, with exon numbers indicated on either side of the lines (exon Z contains the stop codon for a-DB2b). The alternatively spliced exons 12 and 13 are shaded. Boxes denote nucleotide differences from the previously published a-DB2a and a-DB3 sequences, and the polyA additional signals are underlined. (B) Amino acid sequence of a-DB2b compared with a-DB2a and a-DB3. Dots represent regions encoded by the same exons. Boxes denote amino acids that are encoded by alternatively spliced exons with the exon numbers indicated adjacent to each; exZ denotes the exon that encodes the unique 3∞ end of a-DB2b.
484
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
a-dystrobrevin isoforms, we performed RT–PCR on RNA isolated from the three tissues using a common primer in exon 10, and an a-DB2b-specific primer in the unique 3∞ end (primers b and g in Fig. 1). Agarose gel analysis of the products ( Fig. 3A) showed the expected 671 bp band in skeletal muscle, heart, and brain, as well as two other products whose sizes were consistent with a-DB2b cDNA splice variants lacking exon 12 (593 bp) and lacking both exons 12 and 13 (500 bp). An RT– PCR reaction using skeletal muscle RNA and primers a and g (see Fig. 1) produced similar products, and sequencing of the smaller product showed that it did indeed lack exons 12 and 13 as predicted (data not shown). Clearly, the major a-DB2b isoform expressed in skeletal muscle and heart contains the alternatively spliced exons 12 and 13 of vr3; in contrast, two a-DB2b splice variants, one containing exons 12 and 13 and one presumably lacking exon 12, are expressed at approximately equal levels in brain. The a-DB2b isoform lacking both exons 12 and 13 (500 bp band) was expressed only at very low levels in all three tissues. These data show that the splicing of exons 12 and 13 into a-dystrobrevin transcripts is not restricted to skeletal muscle and heart.
Fig. 3. RT–PCR analysis of a-dystrobrevin isoform expression in mouse skeletal muscle, heart and brain. Total RNA from the indicated tissues was reverse transcribed into cDNA, which was then amplified by PCR using primers flanking the three variable regions of a-dystrobrevin. Primers b and g, b and f, b and e, d and e, a and c, and h and i were used in panels A–F, respectively (see Table 1 and Fig. 1 for sequence and location of primers). The specificity of each primer pair and the encompassed variable region is indicated at the left of each panel. GAPDH-specific coding and non-coding primers were used in panel G to amplify GAPDH as a cDNA synthesis control. PCR products were separated on either a 1% agarose gel (panels A– C, F and G) or a 4% Agarose 1000 gel (panels D and E ) (Life Technologies) and visualized by EtdBr staining. Product sizes are indicated on the right (in bp). The sizes of the middle-sized bands observed in panels A, B, and C suggest that they most likely represent the expression of exon 13 alone at vr3.
To directly compare the alternative splicing of exons 12 and 13 in a-DB2b to that in a-DB2a and a-DB1 in all three tissues, we repeated the RT–PCR using the same common primer in exon 10 (primer b in Fig. 1) and an a-DB2a or a-DB1-specific primer (primers f or e in Fig. 1). In agreement with data discussed by Blake et al. (1996), our data show that the primary a-DB1 and a-DB2a transcripts in muscle and heart contain exons 12 and 13 ( Fig. 3B, C ), as was also observed for a-DB2b (Fig. 3A). In contrast to skeletal muscle and heart, previously described RT–PCR experiments suggested that the major a-dystrobrevin transcript in brain lacks exons 12 and 13, although rare transcripts containing the exons were also detected (Ambrose et al., 1997). We have now extended these results to address the isoform specificity of the alternative splicing in brain. We found that the three a-dystrobrevin classes, a-DB2b, a-DB2a, and a-DB1, differ in their expression patterns in brain. The major a-DB2a splice variant expressed in brain lacks exons 12 and 13 ( Fig. 3B), while three variants of a-DB1 were readily detected in brain ( Fig. 3C ), including a variant containing exons 12 and 13. These data indicate that, in brain, a-DB2a transcripts containing exons 12 and 13 are extremely rare; however, these exons are not rare in a-DB1 transcripts. Our data extend previous analyses of exons 12 and 13, and suggest that the splicing of these exons into mRNA in brain is differentially regulated for the three isoforms (a-DB1, a-DB2a and a-DB2b). Current evidence suggests that in addition to vr3, alternative splicing also occurs at two other sites in a-dystrobrevin, vr1 (exon 9) and vr2 (exon 18) (Blake et al., 1996; Ambrose et al., 1997). Exon 9, which is expressed in at least three classes of a-dystrobrevin (a-DB1, a-DB2a, and a-DB3), appears to be restricted to brain since it has only been found in cDNA clones isolated from brain (Blake et al., 1996; Nawrotzki et al., 1998). In contrast, the alternative splicing of the first 21 nt of exon 18 has only been reported in a-DB1 (Ambrose et al., 1997), and the tissue specificity of this expression has not been examined. To determine whether the alternative splicing at these other two sites shows tissue specificity or selectivity, RT–PCR was performed on RNA from skeletal muscle, heart and brain using primers that flanked either vr2 or vr1. Amplification of the vr2 region of a-DB1 (primers d and e in Fig. 1) in skeletal muscle and heart yielded two products whose sizes are consistent with the presence and absence of the alternatively expressed 21 nt of exon 18 ( Fig. 3D). In contrast, a single product of the smaller size was obtained from brain RNA (Fig. 3D), indicating that the first 21 nt of exon 18 are not spliced into a-DB1 mRNA in this tissue. Amplification of the vr1 region using primers that are common to all four classes of a-dystrobrevin (primers a and c in Fig. 1) yielded a single product in skeletal muscle and heart whose size
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
is consistent with the absence of exon 9 ( Fig. 3E), confirming that exon 9 is not expressed in muscle. In contrast, amplification of the vr1 region from brain RNA yielded three products, two of which were consistent with the presence (224 bp) and absence (215 bp) of exon 9 (Fig. 3E ). The third product, which is approximately 230 bp in length, was unexpected though consistently present in every preparation of brain RNA analyzed. This product may contain a previously unidentified exon in place of exon 9, or some other alternative splicing event that has not yet been described. Taken together, these results demonstrate that expression of the first 18 nt of exon 18 in a-DB1 is restricted to skeletal muscle and heart, while expression of exon 9 is restricted to brain. Thus, at least some of the alternative splicing that occurs in a-dystrobrevin appears to be tissue specific. To complete the analysis of a-dystrobrevin expression in skeletal muscle, heart and brain, RT–PCR was performed with a primer pair specific for a-DB3 (primers h and i in Fig. 1). As shown in Fig. 3F, a-DB3 is most abundant in skeletal muscle, but is also expressed in heart and brain, albeit at much lower levels. Early studies failed to detect a-DB3 in a northern blot of human brain (Nawrotzki et al., 1998); however, a more recent study showed very weak expression of a-DB3 in a Northern blot of mouse brain (Holzfeind et al., 1999), in agreement with our results. We previously showed that amplification of a region encompassing vr1 yielded three splice variants in brain ( Fig. 3E), but the primers used did not distinguish between the four a-dystrobrevin classes. To determine whether the a-DB3 expressed in brain was restricted to a single splice variant or consisted of multiple splice variants, we repeated the RT–PCR with a primer pair (primers a and i in Fig. 1) that would yield a much smaller product size to facilitate detection of exon 9. The reaction yielded three products (data not shown) whose sizes were consistent with the products obtained in brain with the common primers ( Fig. 3E), suggesting that three splice variants of a-DB3 are expressed in brain. Based on these data, we conclude that a-DB3 is predominantly, but not exclusively, expressed in skeletal muscle. 3.3. Expression of a-dystrobrevin isoforms during C2C12 muscle cell differentiation As shown in Fig. 3A, adult skeletal muscle and heart selectively express an a-DB2b isoform that contains exons 12 and 13 of vr3. Previous reports have suggested that alternative splicing of exons 12 and 13 are developmentally regulated in muscle (Nawrotzki et al., 1998; Holzfeind et al., 1999); therefore, we examined the expression of exons 12 and 13 in a-DB2b during the differentiation of the C2C12 muscle cell line. RNA was isolated from C2C12 cell cultures at various times after
485
switching the myoblast cultures to differentiation medium, and RT–PCR was performed using a common forward primer and an a-DB2b-specific reverse primer (primers b and g in Fig. 1). Expression of a-DB2b was first observed after 2 days in differentiation medium ( Fig. 4A), at which time splice variants containing and lacking exons 12 and 13 were expressed at similar levels. By 3 days, the splice variant containing exons 12 and 13 clearly predominates, and this uneven distribution is maintained through the remainder of the time course ( Fig. 4A). Another experiment performed using the same common primer and an a-DB2a-specific primer (primers b and f in Fig. 1) showed similar late expression of a-DB2a with subsequent upregulation of the splice variant containing exons 12 and 13 (Fig. 4B). In contrast to a-DB2b, the a-DB2a splice variant lacking exons 12 and 13 was down-regulated. Our results are similar to those obtained for a-DB2a in H-2Kb-tsA58 muscle cells ( Holzfeind et al., 1999), except that a-DB2a was expressed about 1 day earlier in the differentiation time
Fig. 4. RT–PCR analysis of a-dystrobrevin isoform expression during C2C12 muscle cell differentiation. Total RNA, isolated from C2C12 cells at eight different time points (0, 6, 12, 24 h, 2, 3, 4, and 5 days) after addition of differentiation medium, was reverse transcribed into cDNA and then amplified by PCR with primer pairs flanking vr3 or vr2 and specific for a single isoform class as indicated to the left of the panels. Primers b and g, b and f, b and e, d and e, and h and i were used in panels A–E, respectively (see Table 1 and Fig. 1 for sequence and location of primers). AChR a subunit-specific coding and non-coding primers were used in panel F as an early marker of differentiation, and GAPDH-specific coding and non-coding primers were used in panel G as a cDNA synthesis control. A different set of cDNA preparations was used in panel C, but the GAPDH control for this set was similar to that shown. PCR products were separated on a 1% agarose gel, except for the products in panel D which were separated on a 4% Agarose 1000 gel (Life Technologies), and visualized by EtdBr staining. Product sizes are indicated on the right (in bp).
486
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
course of H-2Kb-tsA58 cells. In addition, our results with a-DB2a are consistent with a-DB2a protein expression in C2C12 cells beginning at about day 4 (Nawrotzki et al., 1998), assuming that there is a delay between expression of the mRNA and expression of the protein. In contrast to the a-DB2 isoforms, a-DB1 transcripts were expressed earlier in C2C12 cells, beginning about 6 h after switching to differentiation medium ( Fig. 4C, D). However, the transition from transcripts lacking exons 12 and 13 to transcripts containing both exons occurred around days 2 and 3 (Fig. 4C ), as observed for the a-DB2 isoforms (Fig. 4A, B), and by day 4, the predominant a-DB1 splice variant present contained both exons 12 and 13 (Fig. 4C ). These results are similar to those obtained in H-2Kb-tsA58 muscle cells (Holzfeind et al., 1999), and are consistent with a-DB1 protein expression at day 1 in C2C12 cells with a switch to a larger protein (presumably containing the amino acids encoded by exons 12 and 13) at day 3 (Nawrotzki et al., 1998). We also found that the a-DB1 splice variant containing exon 13 alone at vr3 is upregulated (Fig. 4C ), but this protein isoform was not visible in Western blots (Nawrotzki et al., 1998). Interestingly, when we examined the developmental regulation of vr2 in a-DB1, we found that the expression of the splice variant containing the first 21 nt of exon 18 is also upregulated at day 2 of C2C12 cell differentiation, with a concurrent down regulation of the splice variant lacking exon 18 ( Fig. 4D). Finally, analysis of a-DB3 during C2C12 cell differentiation showed that it is upregulated during differentiation, with a time course similar to that of a-DB1 (Fig. 4E), consistent with the results obtained in H-2Kb-tsA58 muscle cells (Holzfeind et al., 1999). In conclusion, it is clear that alternative splicing of the various a-dystrobrevin transcripts is regulated during muscle cell differentiation, suggesting that the amino acids encoded by exons 12 and 13 at vr3 and the first 21 nt of exons 18 at vr2 may have important functions in mature myotubes. 3.4. Expression of a-DB2b protein in mouse skeletal muscle To show that the a-DB2b transcript detected by RT– PCR is actually translated into protein, we generated a polyclonal antibody, Ab 6224, that specifically recognizes the unique C-terminal tail encoded by exon Z (see Fig. 2B). Immunoblotting of soluble and insoluble fractions of skeletal muscle extracts with Ab 6224 revealed a single band of appropriate molecular weight (approx. 65 kDa) in both fractions (Fig. 5, right panel ), demonstrating that the a-DB2b protein is indeed expressed in skeletal muscle. No bands were detected in either fraction by the pre-immune serum ( Fig. 5, left panel ), indicating that the immunoreactivity detected by Ab 6224 was specific. This expression of a-DB2b protein in
Fig. 5. Immunoblot of skeletal muscle extract with a-DB2b-specific antibody. Adult mouse skeletal muscle homogenate was extracted with Triton X-100 and centrifuged to generate soluble (S) and insoluble (P) fractions. Two percent of each fraction was run on a SDS–PAGE gel and transferred to nitrocellulose. Identical samples were immunoblotted with an a-DB2b-specific polyclonal antibody, Ab 6224 (right panel ) or pre-immune serum ( left panel ). A single band of immunoreactivity was detected in both fractions by Ab 6224, but not by pre-immune serum, demonstrating the specificity of the antibody binding. The positions of molecular weight markers are indicated on the left (in kDa), and the single a-DB2b band is denoted with an arrowhead on the right.
mouse skeletal muscle suggests that it plays a functional role in this tissue.
4. Discussion In the present study, we cloned an a-dystrobrevin cDNA encoding a novel a-dystrobrevin isoform, a-DB2b, and provide evidence that alternatively spliced variants of the new isoform, as well as the other a-dystrobrevin isoforms, are differentially expressed in skeletal muscle, heart and brain, and developmentally regulated during muscle cell differentiation. In addition, we demonstrate that a-DB2b protein is expressed in mouse skeletal muscle, which suggests that it plays a functional role in this tissue. The data show that skeletal muscle and heart preferentially express a-dystrobrevin splice variants that contain exons 12 and 13 at vr3 and lack exon 9 at vr1. In contrast, brain expresses a variety of vr1 and vr3 splice variants within each a-dystrobrevin class, with no consistent splicing preference for all the classes. For example, at vr3, brain shows expression of two a-DB2b splice variants, one a-DB2a splice variant and three a-DB1 splice variants. The data also show that alternative splicing of the first 21 nt of exon 18 at vr2 in a-DB1 is restricted to skeletal muscle and heart. Expression studies in C2C12 muscle cells show that expression of the a-dystrobrevins is upregulated during differentiation, with a-DB1 and a-DB3 expressed earliest and a-DB2a and a-DB2b expressed later, and that alternative splicing of exons 12 and 13 at vr3 and exon 18 at vr2 is also upregulated during differentiation. The upregulation of specific exons in differentiating muscle
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
cells and their continued expression in adult muscle suggests that these exons encode amino acids that are important for the function of the a-dystrobrevins in mature muscle. Although the physiological function of a-dystrobrevin is unknown, its association with the sarcolemmal proteins utrophin, dystrophin, and syntrophins (Butler et al., 1992; Wagner et al., 1993; Ahn and Kunkel, 1995; Dwyer and Froehner, 1995; Suzuki et al., 1995; Peters et al., 1997a; Sadoulet-Puccio et al., 1997b; Peters et al., 1998) suggest that it may play a structural or signaling role at the cytoplasmic face of the sarcolemma. The utrophin/dystrophin binding site within a-DB1 has been mapped to a coiled coil domain encoded within exons 15 and 16 (SadouletPuccio et al., 1997b), while the syntrophin binding site has been mapped to residues 375–426 of Torpedo dystrobrevin (Dwyer and Froehner, 1995), corresponding to exons 13 and 14 of mouse a-DB1. The residues encoded by the alternatively spliced exons (12, 13 and 18a) are on either side of these protein binding domains, suggesting that these residues may modulate the ability of a-DB1 to interact with these proteins. The dystrophin and utrophin binding sites mapped in a-DB1 are also present in a-DB2; however, a-DB2a binds dystrophin, but not utrophin (Peters et al., 1998). It is possible that the unique C-terminal tail of a-DB2a prevents its interaction with utrophin, or conversely, that the unique C-terminal domain of a-DB1 confers the ability to interact with utrophin. The three syntrophin isoforms, a1, b1, and b2, which are encoded by different genes, are thought to function as modular adapters recruiting signaling proteins to the sarcolemma via their interaction with the dystrophin complex. Studies in mouse skeletal muscle have shown that different syntrophin isoforms preferentially associate with different dystrophin family members; a1 and b1 syntrophins preferentially associate with dystrophin and dystrobrevin, while b1 and b2 syntrophins preferentially associate with utrophin (Peters et al., 1997a). The antibodies used in that study did not distinguish between a-DB1 and a-DB2 or between their splice variants, so it remains to be determined which syntrophins associate with which a-dystrobrevin splice variants in skeletal muscle. We hypothesize that the a-dystrobrevins lacking the amino acids encoded by exons 12 and 13 would be unable to associate with syntrophins, since the syntrophin binding site is partly encoded by exon 13. If so, then syntrophin association with a-dystrobrevin would not occur until late in differentiation when transcripts containing exons 12 and 13 are expressed. It is also tempting to speculate that the alternative C-termini of a-DB2a and a-DB2b could determine which syntrophins are associated with each. Immunofluorescence studies in mouse skeletal muscle with isoform-specific antibodies have shown that a-DB1
487
is restricted to the neuromuscular junction, while a-DB2a is localized to both the sarcolemma and the neuromuscular junction (Peters et al., 1998). Within the neuromuscular junction, a-DB1 appears to be restricted to the crests of the junctional folds, while a-DB2a appears to be restricted to the troughs (Peters et al., 1998). This localization parallels that of utrophin and dystrophin which are restricted to the crests (Bewick et al., 1996) and troughs (Byers et al., 1991; Sealock et al., 1991) of the junctional folds, respectively. Interestingly, a-DB2a is retained at the neuromuscular junction in dystrophin/utrophin double knockouts, even though it preferentially associates with dystrophin (Peters et al., 1998). This suggests that there may be additional interactions with other molecules that could be facilitated by the unique C-terminus of a-DB2a. Although we have shown that a-DB2b protein is present in adult skeletal muscle, its subcellular localization remains to be determined. Its homology to a-DB2a and its similar expression pattern in developing muscle cells would suggest a similar localization to the sarcolemma and neuromuscular junction, but its unique C-terminal tail could potentially mediate a different subcellular localization. Studies are now underway to investigate the subcellular localization of the new isoform, a-DB2b, in skeletal muscle, and its biochemical interactions with other sarcolemmal proteins. The expression of a single variant of the a-DB2b mRNA and its encoded protein in mature skeletal muscle suggests that it may play an important structural or signaling role at the sarcolemma and/or neuromuscular junction.
Acknowledgements We thank Drs. Barry Knox, Mary Lou Vallano, and James McCasland for critically reviewing the manuscript, and we thank Christine Valway for excellent technical assistance in sequencing the a-dystrobrevin clones. This research was supported by grants from the National Institutes of Health (NS34070) and the American Heart Association, New York State Affiliate to M.M. Maimone.
References Ahn, A.H., Kunkel, L.M., 1995. Syntrophin binds to an alternatively spliced exon of dystrophin. J. Cell Biol. 128, 363–371. Ambrose, H.J., Blake, D.J., Nawrotzki, R.A., Davies, K.E., 1997. Genomic organization of the mouse dystrobrevin gene: comparative analysis with the dystrophin gene. Genomics 39, 359–369. Balasubramanian, S., Fung, E.T., Huganir, R.L., 1998. Characterization of the tyrosine phosphorylation and distribution of dystrobrevin isoforms. FEBS Lett. 432, 133–140. Bewick, G.S., Young, C., Slater, C.R., 1996. Spatial relationships of utrophin, dystrophin, beta-dystroglycan and beta-spectrin to ace-
488
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
tylcholine receptor clusters during postnatal maturation of the rat neuromuscular junction. J. Neurocytol. 25, 367–379. Blake, D.J., Nawrotzki, R., Loh, N.Y., Go´recki, D.C., Davies, K.E., 1998. b-Dystrobrevin, a member of the dystrophin-related protein family. Proc. Natl. Acad. Sci. USA 95, 241–246. Blake, D.J., Nawrotzki, R., Peters, M.F., Froehner, S.C., Davies, K.E., 1996. Isoform diversity of dystrobrevin, the murine 87-kDa postsynaptic protein. J. Biol. Chem. 271, 7802–7810. Blau, H.M., Pavlath, G.K., Hardeman, E.C., Chiu, C.P., Silberstein, L., Webster, S.G., Miller, S.C., Webster, C., 1985. Plasticity of the differentiated state. Science 230, 758–766. Butler, M.H., Douville, K., Murnane, A.A., Kramarcy, N.R., Cohen, J.B., Sealock, R., Froehner, S.C., 1992. Association of the Mr 58,000 postsynaptic protein of electric tissue with Torpedo dystrophin and the Mr 87,000 postsynaptic protein. J. Biol. Chem. 267, 6213–6218. Byers, T.J., Kunkel, L.M., Watkins, S.C., 1991. The subcellular distribution of dystrophin in mouse skeletal, cardiac, and smooth muscle. J. Cell Biol. 115, 411–421. Carr, C., Fischbach, G.D., Cohen, J.B., 1989. A novel 87,000-Mr protein associated with acetylcholine receptors in Torpedo electric organ and vertebrate skeletal muscle. J. Cell Biol. 109, 1753–1764. Colledge, M., Froehner, S.C., 1998. Signals mediating ion channel clustering at the neuromuscular junction. Curr. Opin. Neurobiol. 8, 357–363. Dwyer, T.M., Froehner, S.C., 1995. Direct binding of Torpedo syntrophin to dystrophin and the 87 kDa dystrophin homologue. FEBS Lett. 375, 91–94. Frail, D.E., McLaughlin, L.L., Mudd, J., Merlie, J.P., 1988. Identification of the mouse muscle 43,000-dalton acetylcholine receptor-associated protein (RAPsyn) by cDNA cloning. J. Biol. Chem. 263, 15602–15607. Glass, D.J., Yancopoulos, G.D., 1997. Sequential roles of agrin, MuSK and rapsyn during neuromuscular junction formation. Curr. Opin. Neurobiol. 7, 379–384. Holzfeind, P.J., Ambrose, H.J., Newey, S.E., Nawrotzki, R.A., Blake, D.J., Davies, K.E., 1999. Tissue-selective expression of alphadystrobrevin is determined by multiple promoters. J. Biol Chem. 274, 6250–6258. Horowitz, D.S., Krainer, A.R., 1994. Mechanisms for selecting 5∞ splice sites in mammalian pre-mRNA splicing. Trends Genet. 10, 100–106. Kramarcy, N.R., Vidal, A., Froehner, S.C., Sealock, R., 1994. Association of utrophin and multiple dystrophin short forms with the
mammalian M(r) 58,000 dystrophin-associated protein (syntrophin). J. Biol. Chem. 269, 2870–2876. Nawrotzki, R., Loh, N.Y., Ruegg, M.A., Davies, K.E., Blake, D.J., 1998. Characterisation of a-dystrobrevin in muscle. J. Cell Sci. 111, 2595–2605. Peters, M.F., Adams, M.E., Froehner, S.C., 1997a. Differential association of syntrophin pairs with the dystrophin complex. J. Cell Biol. 138, 81–93. Peters, M.F., O’Brien, K.F., Sadoulet-Puccio, H.M., Kunkel, L.M., Adams, M.E., Froehner, S.C., 1997b. b-Dystrobrevin, a new member of the dystrophin family — identification, cloning and protein associations. J. Biol. Chem. 272, 31561–31569. Peters, M.F., Sadoulet-Puccio, H.M., Mark Grady, R., Kramarcy, N.R., Kunkel, L.M., Sanes, J.R., Sealock, R., Froehner, S.C., 1998. Differential membrane localization and intermolecular associations of alpha-dystrobrevin isoforms in skeletal muscle. J. Cell Biol. 142, 1269–1278. Puca, A.A., Nigro, V., Piluso, G., Belsito, A., Sampaolo, S., Quaderi, N., Rossi, E., Di Iorio, G., Ballabio, A., Franco, B., 1998. Identification and characterization of a novel member of the dystrobrevin gene family. FEBS Lett. 425, 7–13. Sadoulet-Puccio, H.M., Feener, C.A., Schaid, D.J., Thibodeau, S.N., Michels, V.V., Kunkel, L.M., 1997a. The genomic organization of human dystrobrevin. Neurogenetics 1, 37–42. Sadoulet-Puccio, H.M., Khurana, T.S., Cohen, J.B., Kunkel, L.M., 1996. Cloning and characterization of the human homologue of a dystrophin related phosphoprotein found at the Torpedo electric organ post-synaptic membrane. Hum. Mol. Gen. 5, 489–496. Sadoulet-Puccio, H.M., Rajala, M., Kunkel, L.M., 1997b. Dystrobrevin and dystrophin: an interaction through coiled-coil motifs. Proc. Natl. Acad. Sci. USA 94, 12413–12418. Sealock, R., Butler, M.H., Kramarcy, N.R., Gao, K.X., Murnane, A.A., Douville, K., Froehner, S.C., 1991. Localization of dystrophin relative to acetylcholine receptor domains in electric tissue and adult and cultured skeletal muscle. J. Cell Biol. 113, 1133–1144. Suzuki, A., Yoshida, M., Ozawa, E., 1995. Mammalian alpha1- and beta1-syntrophin bind to the alternative splice-prone region of the dystrophin COOH terminus. J. Cell Biol. 128, 373–381. Wagner, K.R., Cohen, J.B., Huganir, R.L., 1993. The 87K postsynaptic membrane protein from Torpedo is a protein–tyrosine kinase substrate homologous to dystrophin. Neuron 10, 511–522. Yaffe, D., Saxel, O., 1977. Serial passaging and differentiation of myogenic cells isolated from dystrophic mouse muscle. Nature 270, 725–727.
Differential expression and developmental regulation of a novel a-dystrobrevin isoform in muscle k Rebecca E. Enigk, Margaret M. Maimone * Department of Anatomy and Cell Biology, State University of New York Health Science Center, 750 East Adams Street, Syracuse, NY 13210, USA Received 20 April 1999; received in revised form 15 July 1999; accepted 6 August 1999; Received by A.J. van Wijnen
Abstract a-Dystrobrevin is a dystrophin-related protein expressed primarily in skeletal muscle, heart, lung and brain. In skeletal muscle, a-dystrobrevin is a component of the dystrophin-associated glycoprotein complex and is localized to the sarcolemma, presumably through interactions with dystrophin and utrophin. Alternative splicing of the a-dystrobrevin gene generates multiple isoforms which have been grouped into three major classes: a-DB1, a-DB2, and a-DB3. Various isoforms have been shown to interact with a variety of proteins; however, the physiological function of the a-dystrobrevins remains unknown. In the present study, we have cloned a novel a-dystrobrevin cDNA encoding a protein (referred to as a-DB2b) with a unique 11 amino acid C-terminal tail. Using RT–PCR with primers specific to the new isoform, we have characterized its expression in skeletal muscle, heart, and brain, and in differentiating C2C12 muscle cells. We show that a-DB2b is expressed in skeletal muscle, heart and brain, and that exons 12 and 13 are alternatively spliced in a-DB2b to generate at least three splice variants. The major a-DB2b splice variant expressed in adult skeletal muscle and heart contains exons 12 and 13, while in adult brain, two a-DB2b splice variants are expressed at similar levels. This is consistent with the preferential expression of exons 12 and 13 in other a-dystrobrevin isoforms in skeletal muscle and heart. Similarly, in a-DB1 the first 21 nucleotides of exon 18 are preferentially expressed in skeletal muscle and heart relative to brain. We also show that the expression of alternatively spliced a-DB2b is developmentally regulated in muscle; during differentiation of C2C12 cells, a-DB2b expression switches from an isoform lacking exons 12 and 13 to one containing them. We demonstrate similar developmental upregulation of exons 12, 13, and 18 in a-DB1 and of exons 12 and 13 in a-DB2a. Finally, we show that a-DB2b protein is expressed in adult skeletal muscle, suggesting that it has a functional role in adult muscle. Together, these data suggest that alternatively spliced variants of the new a-dystrobrevin isoform, a-DB2b, are differentially expressed in various tissues and developmentally regulated during muscle cell differentiation in a fashion similar to that previously described for a-dystrobrevin isoforms. © 1999 Elsevier Science B.V. All rights reserved. Keywords: Alternative splicing; C2C12 cells; cDNA cloning; Dystrophin associated protein; RT–PCR
1. Introduction Dystrobrevin is a dystrophin-related protein that was originally identified as an acetylcholine receptor (AChR) Abbreviations: Ab, antibody; AChR, acetylcholine receptor; a-DB, a-dystrobrevin; cDNA, DNA complementary to mRNA; DMEM, Dulbecco’s Modified Eagle Medium; dNTP, deoxyribonucleoside triphosphate; DTT, dithiothreitol; EtdBr, ethidium bromide; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; nt, nucleotide(s); PAGE, polyacrylamide gel electrophoresis; PCR, polymerase chain reaction; RT–PCR, reverse transcriptase polymerase chain reaction; u, unit(s); UTR, untranslated region; vr, variable region. k The nucleotide sequences reported in this paper have been submitted to GenBank@ with accession numbers: AF143542, AF143543, and AF143544. * Corresponding author. Tel.: +1-315-464-8526; fax: +1-315-464-8535. E-mail address: [email protected] (M.M. Maimone)
-associated protein present in postsynaptic membranes of Torpedo electric organ, and chick and rat endplates (Carr et al., 1989). Cloning of a-dystrobrevin from Torpedo ( Wagner et al., 1993), human (Sadoulet-Puccio et al., 1996), and mouse (Blake et al., 1996) revealed multiple transcripts generated by alternative splicing of the a-dystrobrevin gene (Ambrose et al., 1997; SadouletPuccio et al., 1997a). More recently, a second gene encoding a closely related protein, b-dystrobrevin, was identified in mouse and human (Peters et al., 1997b; Blake et al., 1998; Puca et al., 1998) and shown to be expressed in a broader range of tissues than a-dystrobrevin. The mouse a-dystrobrevin gene consists of 24 coding exons (Ambrose et al., 1997). The alternative usage of three exons containing stop codons, exons 24, 18, and 11, generates transcripts encoding three major classes of
0378-1119/99/$ – see front matter © 1999 Elsevier Science B.V. All rights reserved. PII: S0 3 7 8 -1 1 1 9 ( 9 9 ) 0 0 35 8 - 3
480
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
a-dystrobrevin proteins: a-dystrobrevin1 (a-DB1), a-dystrobrevin2 (a-DB2), and a-dystrobrevin3 (a-DB3), respectively (Ambrose et al., 1997; Nawrotzki et al., 1998). Alternative splicing within the transcripts encoding a-DB1 and a-DB2 at three different sites generates additional variability within the a-dystrobrevin family of proteins. Variable region 1 (vr1), encoded by exon 9, is predominantly expressed in brain, while exons 12 and 13 of variable region 3 (vr3) are predominantly expressed in skeletal muscle (Blake et al., 1996). More recently, the expression of exons 12 and 13 was shown to increase during myotube differentiation in the mouse muscle cell line H-2Kb-tsA58 (Holzfeind et al., 1999). While it is clear that multiple isoforms of a-dystrobrevin are expressed in skeletal muscle, the physiological function of the various isoforms remains unknown. Early studies in skeletal muscle showed that a-dystrobrevin was localized to the sarcolemma and neuromuscular junction (Carr et al., 1989) and that it was a component of the dystrophin-associated protein complex, presumably through its interactions with dystrophin, utrophin and/or syntrophins (Butler et al., 1992; Ahn and Kunkel, 1995; Dwyer and Froehner, 1995; Suzuki et al., 1995; Peters et al., 1997a; SadouletPuccio et al., 1997b). Subsequently, immunofluorescence localization studies with isoform-specific antibodies showed that a-DB2 is localized to the sarcolemma and neuromuscular junction, whereas a-DB1 is more restricted to the neuromuscular junction (Balasubramanian et al., 1998; Peters et al., 1998). In addition, a-DB1 and a-DB2 differ in their ability to interact with other proteins; a-DB2 binds dystrophin, whereas a-DB1 binds both dystrophin and utrophin (Peters et al., 1998) and associates more tightly with syntrophin than does a-DB2 (Balasubramanian et al., 1998). a-DB1 contains two major sites of tyrosine phosphorylation in its unique C-terminal tail (Balasubramanian et al., 1998), suggesting that its activity at the neuromuscular junction may be modulated by tyrosine phosphorylation. However, a-DB1 is not phosphorylated in response to treatment with agrin (Nawrotzki et al., 1998), the nerve-derived signaling molecule that initiates the formation of the neuromuscular junction through tyrosine phosphorylation (Glass and Yancopoulos, 1997; Colledge and Froehner, 1998). These findings suggest that alternative splicing generates multiple a-dystrobrevin isoforms that differentially localize in skeletal muscle and that potentially have different physiological functions. In the present study, we have cloned a novel splice variant of a-DB2 from a mouse muscle cell line library. The new isoform, referred to as a-DB2b, contains a unique 11 amino acid C-terminal tail encoded by a novel 3∞ exon, referred to as exon Z, that is spliced into the transcript after the first 21 nt of exon 18. Using RT– PCR and primers specific for the new isoform, we show that a-DB2b is expressed in skeletal muscle, heart, and
brain, and that it is alternatively spliced at vr3 to generate at least three isoforms. We demonstrate that the expression of a-DB2b and its alternative splicing at vr3 are both developmentally regulated in the C2C12 muscle cell line, as are other a-dystrobrevin isoforms. In addition, we show that a polyclonal antibody specific for a-DB2b recognizes a protein of the expected molecular weight in skeletal muscle. Taken together, our data provide support for a general principle that alternatively spliced variants of a-dystrobrevin are differentially expressed in tissues and developmentally regulated in muscle cells.
2. Materials and methods 2.1. Cloning and sequencing To isolate a-dystrobrevin cDNA clones, a BC3H1 mouse muscle cell line cDNA library ( Frail et al., 1988) was screened with a full-length Torpedo dystrobrevin cDNA ( Wagner et al., 1993) probe using the Genius System (Boehringer Mannheim, Indianapolis, IN ), and 19 positive clones were analyzed by restriction mapping and partial sequencing. Two clones, 11.2 and 9.2.3, which contained consensus polyA addition signals near their 3∞ ends were subcloned into pBluescriptII(+) (Stratagene, La Jolla, CA) and completely sequenced using Sequenase ( United States Biochemicals, Cleveland, OH ). Clone 11.2 contained an apparently full-length copy of the mouse a-DB3 mRNA, while clone 9.2.3 encoded part of an a-DB2 isoform with a unique C-terminus compared to the previously published a-DB2 sequence (Blake et al., 1996). We use the terms a-DB2a and a-DB2b to distinguish the previously published and new isoforms, respectively. The BC3H1 library was rescreened with a 550 bp fragment of the longest clone (7.2), and six additional positive clones were isolated and analyzed. One clone (16.1), which was subcloned and sequenced as described above, contained an almost full-length copy of the a-DB2a mRNA including a consensus poly A additional signal near the 3∞ end. Cycle sequencing of the a-DB2b cDNA generated by RT–PCR was performed by the DNA Sequencing Facility of the BioResource Center at Cornell University in Ithaca, NY. 2.2. C2C12 cell culture C2C12 cells ( Yaffe and Saxel, 1977; Blau et al., 1985) were maintained at subconfluence on gelatin (0.2%) -coated tissue culture dishes in growth medium consisting of DMEM (Life Technologies, Gaithersburg, MD) supplemented with 20% fetal bovine serum (Atlanta Biologicals, Norcross, GA), 50 u/ml penicillin and 50 mg/ml streptomycin (Sigma Chemicals, St. Louis,
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
MO). To prepare cells at various stages of differentiation for RNA isolation, myoblasts were grown to confluence (3–4 days) and then the growth medium was replaced with differentiation medium consisting of DMEM supplemented with 5% horse serum (Life Technologies), 50 u/ml penicillin and 50 mg/ml streptomycin. RNA was isolated from cells as described in Section 2.3 at specific time points (0 h, 6 h, 12 h, 24 h, 2 days, 3 days, 4 days, and 5 days) after switching to differentiation medium.
2.3. RNA isolation Total RNA was prepared from mouse tissues and C2C12 cells using TRIzol reagent (Life Technologies) as recommended by the manufacturer. Briefly, hindlimb muscles, hearts, and brains were dissected from 8 week old Balb/CJ mice, frozen in liquid nitrogen, and stored at −70 °C. Frozen tissue was ground in a mortar under liquid nitrogen; 1 ml of TRIzol was added per 50– 100 mg of tissue and the solution was homogenized with a Polytron homogenizer at room temperature. For C2C12 cells, 18 ml of TRIzol was added to each 6 cm dish of cells. These TRIzol-containing solutions were extracted with chloroform and precipitated with isopropyl alcohol. The RNA pellet was then washed with 70% ethanol and resuspended in RNAse-free dH O. RNA 2 integrity was confirmed by electrophoresis of 1 mg on a 1% agarose gel, and the RNA concentration was determined by absorbance at 260 nm.
2.4. RT–PCR
481
Fig. 1. Schematic diagram of a-dystrobrevin isoforms generated by alternative splicing. Four classes of a-dystrobrevin isoforms generated by the usage of four different stop codons are presented, with the new a-DB2b isoform denoted with an asterisk. Large and small boxes represent coding and non-coding regions, respectively, and shaded boxes denote alternatively spliced exons at the three variable regions designated vr1 (exon 9), vr2 (exon 18), and vr3 (exons 11, 12, 13). The approximate positions of primers used for RT–PCR are designated by lower-case letters (refer to Table 1 for primer sequences). Table 1 Oligo primers used for RT–PCR Primera
Sequence
a b c d e f g h i j GAPDH-c GAPDH-nc AChRa-c AChRa-nc
5∞-CGTCGTGGAAATCACCTGC-3∞ 5∞-GTTCCCTCCTCAGGAAGTCC-3∞ 5∞-CCTGGTGATGAAAGGACTTC-3∞ 5∞-CTCCGGCTCCTCAGACAG-3∞ 5∞-GGCAGATGCTGAACGGATG-3∞ 5∞-CACTGTTAGTTAAGACCTGCAG-3∞ 5∞-GGAATTCC/TCATGTTCTCTCCTCAAGG-3∞ 5∞-GTCTTCCATCCAGTTGAGTG-3∞ 5∞-GGAACAGGAAGTCATGTTATCC-3∞ 5∞-TGTGTGCTCTAGATGAACTG-3∞ 5∞-ACCCAGAAGACTGTGGATGG-3∞ 5∞-GATCCACGACGGACACATTG-3∞ 5∞-CGCAGACGGCGACTTTGC-3∞ 5∞-GAAGTAGAGGGGCAGGCG-3∞
a c designates coding primer; nc designates non-coding primer.
First-strand cDNA was synthesized from 2 mg of total RNA using an oligo dT adapter primer (5∞-TGAGCTCGAGTCGACATCGA( T ) V-3∞) and 17 Superscript II Reverse Transcriptase (Life Technologies) as recommended by the manufacturer, and then stored at −70 °C. Each 50 ml PCR reaction contained 2 ml of cDNA, 40 pmol of each primer (see Fig. 1 for location of primers and Table 1 for sequence of primers), 1.5 units of Platinum Taq DNA polymerase (Life Technologies), Platinum Taq PCR buffer (20 mM Tris–HCl, pH 8.4, 50 mM KCl ), 0.2 mM dNTPs, and 1.5 mM MgCl and 2 was overlaid with 2 drops of mineral oil. All reactions were incubated for 2 min at 94°C to activate the Platinum Taq enzyme, and then cycled 29 times through 30 s at 94°C, 30 s at the appropriate annealing temperature and 3 min at 72°C. Note: for the a-DB2b C2C12 time course amplification, the reactions were incubated for 5 min at 94°C, 2 min at 58°C and 40 min at 72°C, and then the reaction was cycled 29 times through 40 s at 94°C, 2 min at 58°C and 3 min at 72 °C, followed by a final incubation of 15 min at 72°C.
2.5. Polyclonal antibody and immunoblotting A polyclonal antibody that recognizes the unique C-terminal tail of aDB2-b (Ab 6224) was generated in rabbits by GeneMed (San Francisco, CA) using the synthetic peptide C-HEIIPLEERT-COOH. Crude serum was used in immunoblots of mouse skeletal muscle fractions prepared according to the method of Kramarcy et al. (1994). Briefly, frozen skeletal muscle tissue (255 mg) from 8 week old Balb/CJ mice was ground under liquid nitrogen with a mortar and pestle, homogenized with a Polytron homogenizer for 30 s in 10 volumes of ice-cold homogenization buffer (10 mM sodium phosphate, 5 mM ethylenediaminetetraacetic acid, 0.4 M NaCl, pH 7.8 and protease inhibitors: aprotinin and leupeptin at 5 mg/ml each, pepstatin A at 0.05 mg/ml, and 2 mM phenylmethylsulfonyl fluoride), and centrifuged for 10 min at 12 000×g. The pellet was resuspended in homogenization buffer and recentrifuged for 10 min at 12 000×g. This pellet was then solubilized
482
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
in homogenization buffer containing 1% Triton X-100 for 15 min on ice and centrifuged for 10 min at 12 000×g. Sample buffer was added to both the supernatant (soluble) and pellet (insoluble) fractions to give a final concentration of 80 mM Tris (pH 6.8), 2% SDS, 15% glycerol, 0.01% bromophenol blue, and 100 mM DTT. Protein concentrations of the samples were determined by the Bio-Rad Protein Assay (Bio-Rad, Hercules, CA.). Two microliters representing 2% of the soluble and insoluble fractions (2.0 mg and 14.5 mg total protein, respectively) were then subjected to SDS–PAGE on a 12% polyacrylamide gel and transferred electrophoretically to nitrocellulose. Immunoblotting was performed according to the manufacturer’s directions for enhanced chemiluminescence detection (Amersham, Piscataway, NJ ). Pre-immune and immune sera were used at a dilution of 1:3000, while the horseradish peroxidase-conjugated goat anti-rabbit IgG (BioRad ) was used at a dilution of 1:6000.
3. Results 3.1. Cloning of a-dystrobrevin isoforms Screening of a mouse muscle cell line cDNA library yielded many a-dystrobrevin clones corresponding to the a-DB2 and a-DB3 isoform classes (Fig. 1) (Ambrose et al., 1997). One clone, 9.2.3, contained a partial cDNA insert that appeared to be in the a-DB2 class based on its length; however, the sequence at the 3∞ end, including the last 13 codons, diverged from the previously published a-DB2 sequence (Blake et al., 1996) after the first 21 nt of exon 18 (Fig. 1). As shown previously for a-DB1 (Ambrose et al., 1997), the GT at nt 22 and 23 of exon 18 can function as a cryptic 5∞ splice donor site for the new exon, which we have designated exon Z since we do not know its location in the gene relative to the other exons. In addition to the highly conserved GT at the 5∞ end of the intron, the cryptic splice site in exon 18, GGA//GTAAGT, conforms to the 5∞ splice site consensus sequence, (A/C )AG//GTRAGT, at six of nine bases. On average, mammalian 5∞ splice sites match the consensus at seven of nine bases (reviewed in Horowitz and Krainer, 1994). To determine whether this partial cDNA clone represented a true mRNA present in muscle, we carried out RT–PCR on total RNA isolated from mouse skeletal muscle using primers
483
in the 5∞-UTR and in the unique 3∞ coding region (primers j and g in Fig. 1) (data not shown). Cycle sequencing of the resulting 1.9 kb PCR product showed that it was identical to the previously reported a-DB2 sequence, except that it lacked exon 9 at vr1, included exons 12 and 13 at vr3, and diverged into unique sequence after nt 21 of exon 18. This indicated that the mRNA represented by clone 9.2.3 was indeed expressed in skeletal muscle and differs from a-DB2 at its 3∞ end. We therefore refer to the new isoform as a-DB2b and the previously described isoform as a-DB2a. Although clone 9.2.3 contained a polyA addition signal near its 3∞ end, its 5∞ end began in the middle of the a-DB2 coding region, so we combined the sequences of clone 9.2.3 and the PCR product to generate the a-DB2b sequence shown in Fig. 2A. The cDNA library screening also yielded clones containing the complete coding sequence and 3∞-UTR for a-DB2a and a-DB3. The sequence of our a-DB2a clone ( Fig. 2A) differed from that previously published (Blake et al., 1996) in that it lacked the brain-specific exon 9, but contained exons 12 and 13. The 5∞-UTR of our a-DB2a cDNA was larger and consisted of exons C and D instead of exons B and D (Ambrose et al., 1997), and the 3∞-UTR of our a-DB2a cDNA was 992 bp longer and included a polyA addition signal near its 3∞ end. In addition, there were two single nucleotide differences; our sequence contained a G instead of C at nt 721, consistent with the sequence of exon 7 of the a-dystrobrevin gene (Ambrose et al., 1997), and four As instead of three beginning at nt 1843. Sequencing of our a-DB3 cDNA ( Fig. 2A) showed that the 5∞-UTR consisted of exons C and D, while the coding region consisted of exons 1–11, excluding exon 9. In addition, the 3∞-UTR was identical to that previously described for a-DB3 (Nawrotzki et al., 1998), except that it was 14 bp longer and had four single nt differences: four Ts instead of three at nt 1167, one A instead of two at nt 1217, C instead of T at nt 1318, and C instead of T at nt 1323. A comparison of the proteins encoded by our three full-length dystrobrevin cDNAs ( Fig. 2B) shows that each has a unique C terminus generated by alternative splicing. 3.2. Expression of a-dystrobrevin isoforms in muscle, heart, and brain To determine whether a-DB2b was expressed in heart and brain as well as skeletal muscle, as are the other
Fig. 2. Sequence analysis of a-DB2b. (A) Nucleotide sequence of a-DB2b compared with a-DB2a and a-DB3. Dots represent sequence derived from the same exons, while upper- and lower-case letters represent coding and non-coding regions, respectively. Vertical lines denote selected exon boundaries, with exon numbers indicated on either side of the lines (exon Z contains the stop codon for a-DB2b). The alternatively spliced exons 12 and 13 are shaded. Boxes denote nucleotide differences from the previously published a-DB2a and a-DB3 sequences, and the polyA additional signals are underlined. (B) Amino acid sequence of a-DB2b compared with a-DB2a and a-DB3. Dots represent regions encoded by the same exons. Boxes denote amino acids that are encoded by alternatively spliced exons with the exon numbers indicated adjacent to each; exZ denotes the exon that encodes the unique 3∞ end of a-DB2b.
484
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
a-dystrobrevin isoforms, we performed RT–PCR on RNA isolated from the three tissues using a common primer in exon 10, and an a-DB2b-specific primer in the unique 3∞ end (primers b and g in Fig. 1). Agarose gel analysis of the products ( Fig. 3A) showed the expected 671 bp band in skeletal muscle, heart, and brain, as well as two other products whose sizes were consistent with a-DB2b cDNA splice variants lacking exon 12 (593 bp) and lacking both exons 12 and 13 (500 bp). An RT– PCR reaction using skeletal muscle RNA and primers a and g (see Fig. 1) produced similar products, and sequencing of the smaller product showed that it did indeed lack exons 12 and 13 as predicted (data not shown). Clearly, the major a-DB2b isoform expressed in skeletal muscle and heart contains the alternatively spliced exons 12 and 13 of vr3; in contrast, two a-DB2b splice variants, one containing exons 12 and 13 and one presumably lacking exon 12, are expressed at approximately equal levels in brain. The a-DB2b isoform lacking both exons 12 and 13 (500 bp band) was expressed only at very low levels in all three tissues. These data show that the splicing of exons 12 and 13 into a-dystrobrevin transcripts is not restricted to skeletal muscle and heart.
Fig. 3. RT–PCR analysis of a-dystrobrevin isoform expression in mouse skeletal muscle, heart and brain. Total RNA from the indicated tissues was reverse transcribed into cDNA, which was then amplified by PCR using primers flanking the three variable regions of a-dystrobrevin. Primers b and g, b and f, b and e, d and e, a and c, and h and i were used in panels A–F, respectively (see Table 1 and Fig. 1 for sequence and location of primers). The specificity of each primer pair and the encompassed variable region is indicated at the left of each panel. GAPDH-specific coding and non-coding primers were used in panel G to amplify GAPDH as a cDNA synthesis control. PCR products were separated on either a 1% agarose gel (panels A– C, F and G) or a 4% Agarose 1000 gel (panels D and E ) (Life Technologies) and visualized by EtdBr staining. Product sizes are indicated on the right (in bp). The sizes of the middle-sized bands observed in panels A, B, and C suggest that they most likely represent the expression of exon 13 alone at vr3.
To directly compare the alternative splicing of exons 12 and 13 in a-DB2b to that in a-DB2a and a-DB1 in all three tissues, we repeated the RT–PCR using the same common primer in exon 10 (primer b in Fig. 1) and an a-DB2a or a-DB1-specific primer (primers f or e in Fig. 1). In agreement with data discussed by Blake et al. (1996), our data show that the primary a-DB1 and a-DB2a transcripts in muscle and heart contain exons 12 and 13 ( Fig. 3B, C ), as was also observed for a-DB2b (Fig. 3A). In contrast to skeletal muscle and heart, previously described RT–PCR experiments suggested that the major a-dystrobrevin transcript in brain lacks exons 12 and 13, although rare transcripts containing the exons were also detected (Ambrose et al., 1997). We have now extended these results to address the isoform specificity of the alternative splicing in brain. We found that the three a-dystrobrevin classes, a-DB2b, a-DB2a, and a-DB1, differ in their expression patterns in brain. The major a-DB2a splice variant expressed in brain lacks exons 12 and 13 ( Fig. 3B), while three variants of a-DB1 were readily detected in brain ( Fig. 3C ), including a variant containing exons 12 and 13. These data indicate that, in brain, a-DB2a transcripts containing exons 12 and 13 are extremely rare; however, these exons are not rare in a-DB1 transcripts. Our data extend previous analyses of exons 12 and 13, and suggest that the splicing of these exons into mRNA in brain is differentially regulated for the three isoforms (a-DB1, a-DB2a and a-DB2b). Current evidence suggests that in addition to vr3, alternative splicing also occurs at two other sites in a-dystrobrevin, vr1 (exon 9) and vr2 (exon 18) (Blake et al., 1996; Ambrose et al., 1997). Exon 9, which is expressed in at least three classes of a-dystrobrevin (a-DB1, a-DB2a, and a-DB3), appears to be restricted to brain since it has only been found in cDNA clones isolated from brain (Blake et al., 1996; Nawrotzki et al., 1998). In contrast, the alternative splicing of the first 21 nt of exon 18 has only been reported in a-DB1 (Ambrose et al., 1997), and the tissue specificity of this expression has not been examined. To determine whether the alternative splicing at these other two sites shows tissue specificity or selectivity, RT–PCR was performed on RNA from skeletal muscle, heart and brain using primers that flanked either vr2 or vr1. Amplification of the vr2 region of a-DB1 (primers d and e in Fig. 1) in skeletal muscle and heart yielded two products whose sizes are consistent with the presence and absence of the alternatively expressed 21 nt of exon 18 ( Fig. 3D). In contrast, a single product of the smaller size was obtained from brain RNA (Fig. 3D), indicating that the first 21 nt of exon 18 are not spliced into a-DB1 mRNA in this tissue. Amplification of the vr1 region using primers that are common to all four classes of a-dystrobrevin (primers a and c in Fig. 1) yielded a single product in skeletal muscle and heart whose size
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
is consistent with the absence of exon 9 ( Fig. 3E), confirming that exon 9 is not expressed in muscle. In contrast, amplification of the vr1 region from brain RNA yielded three products, two of which were consistent with the presence (224 bp) and absence (215 bp) of exon 9 (Fig. 3E ). The third product, which is approximately 230 bp in length, was unexpected though consistently present in every preparation of brain RNA analyzed. This product may contain a previously unidentified exon in place of exon 9, or some other alternative splicing event that has not yet been described. Taken together, these results demonstrate that expression of the first 18 nt of exon 18 in a-DB1 is restricted to skeletal muscle and heart, while expression of exon 9 is restricted to brain. Thus, at least some of the alternative splicing that occurs in a-dystrobrevin appears to be tissue specific. To complete the analysis of a-dystrobrevin expression in skeletal muscle, heart and brain, RT–PCR was performed with a primer pair specific for a-DB3 (primers h and i in Fig. 1). As shown in Fig. 3F, a-DB3 is most abundant in skeletal muscle, but is also expressed in heart and brain, albeit at much lower levels. Early studies failed to detect a-DB3 in a northern blot of human brain (Nawrotzki et al., 1998); however, a more recent study showed very weak expression of a-DB3 in a Northern blot of mouse brain (Holzfeind et al., 1999), in agreement with our results. We previously showed that amplification of a region encompassing vr1 yielded three splice variants in brain ( Fig. 3E), but the primers used did not distinguish between the four a-dystrobrevin classes. To determine whether the a-DB3 expressed in brain was restricted to a single splice variant or consisted of multiple splice variants, we repeated the RT–PCR with a primer pair (primers a and i in Fig. 1) that would yield a much smaller product size to facilitate detection of exon 9. The reaction yielded three products (data not shown) whose sizes were consistent with the products obtained in brain with the common primers ( Fig. 3E), suggesting that three splice variants of a-DB3 are expressed in brain. Based on these data, we conclude that a-DB3 is predominantly, but not exclusively, expressed in skeletal muscle. 3.3. Expression of a-dystrobrevin isoforms during C2C12 muscle cell differentiation As shown in Fig. 3A, adult skeletal muscle and heart selectively express an a-DB2b isoform that contains exons 12 and 13 of vr3. Previous reports have suggested that alternative splicing of exons 12 and 13 are developmentally regulated in muscle (Nawrotzki et al., 1998; Holzfeind et al., 1999); therefore, we examined the expression of exons 12 and 13 in a-DB2b during the differentiation of the C2C12 muscle cell line. RNA was isolated from C2C12 cell cultures at various times after
485
switching the myoblast cultures to differentiation medium, and RT–PCR was performed using a common forward primer and an a-DB2b-specific reverse primer (primers b and g in Fig. 1). Expression of a-DB2b was first observed after 2 days in differentiation medium ( Fig. 4A), at which time splice variants containing and lacking exons 12 and 13 were expressed at similar levels. By 3 days, the splice variant containing exons 12 and 13 clearly predominates, and this uneven distribution is maintained through the remainder of the time course ( Fig. 4A). Another experiment performed using the same common primer and an a-DB2a-specific primer (primers b and f in Fig. 1) showed similar late expression of a-DB2a with subsequent upregulation of the splice variant containing exons 12 and 13 (Fig. 4B). In contrast to a-DB2b, the a-DB2a splice variant lacking exons 12 and 13 was down-regulated. Our results are similar to those obtained for a-DB2a in H-2Kb-tsA58 muscle cells ( Holzfeind et al., 1999), except that a-DB2a was expressed about 1 day earlier in the differentiation time
Fig. 4. RT–PCR analysis of a-dystrobrevin isoform expression during C2C12 muscle cell differentiation. Total RNA, isolated from C2C12 cells at eight different time points (0, 6, 12, 24 h, 2, 3, 4, and 5 days) after addition of differentiation medium, was reverse transcribed into cDNA and then amplified by PCR with primer pairs flanking vr3 or vr2 and specific for a single isoform class as indicated to the left of the panels. Primers b and g, b and f, b and e, d and e, and h and i were used in panels A–E, respectively (see Table 1 and Fig. 1 for sequence and location of primers). AChR a subunit-specific coding and non-coding primers were used in panel F as an early marker of differentiation, and GAPDH-specific coding and non-coding primers were used in panel G as a cDNA synthesis control. A different set of cDNA preparations was used in panel C, but the GAPDH control for this set was similar to that shown. PCR products were separated on a 1% agarose gel, except for the products in panel D which were separated on a 4% Agarose 1000 gel (Life Technologies), and visualized by EtdBr staining. Product sizes are indicated on the right (in bp).
486
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
course of H-2Kb-tsA58 cells. In addition, our results with a-DB2a are consistent with a-DB2a protein expression in C2C12 cells beginning at about day 4 (Nawrotzki et al., 1998), assuming that there is a delay between expression of the mRNA and expression of the protein. In contrast to the a-DB2 isoforms, a-DB1 transcripts were expressed earlier in C2C12 cells, beginning about 6 h after switching to differentiation medium ( Fig. 4C, D). However, the transition from transcripts lacking exons 12 and 13 to transcripts containing both exons occurred around days 2 and 3 (Fig. 4C ), as observed for the a-DB2 isoforms (Fig. 4A, B), and by day 4, the predominant a-DB1 splice variant present contained both exons 12 and 13 (Fig. 4C ). These results are similar to those obtained in H-2Kb-tsA58 muscle cells (Holzfeind et al., 1999), and are consistent with a-DB1 protein expression at day 1 in C2C12 cells with a switch to a larger protein (presumably containing the amino acids encoded by exons 12 and 13) at day 3 (Nawrotzki et al., 1998). We also found that the a-DB1 splice variant containing exon 13 alone at vr3 is upregulated (Fig. 4C ), but this protein isoform was not visible in Western blots (Nawrotzki et al., 1998). Interestingly, when we examined the developmental regulation of vr2 in a-DB1, we found that the expression of the splice variant containing the first 21 nt of exon 18 is also upregulated at day 2 of C2C12 cell differentiation, with a concurrent down regulation of the splice variant lacking exon 18 ( Fig. 4D). Finally, analysis of a-DB3 during C2C12 cell differentiation showed that it is upregulated during differentiation, with a time course similar to that of a-DB1 (Fig. 4E), consistent with the results obtained in H-2Kb-tsA58 muscle cells (Holzfeind et al., 1999). In conclusion, it is clear that alternative splicing of the various a-dystrobrevin transcripts is regulated during muscle cell differentiation, suggesting that the amino acids encoded by exons 12 and 13 at vr3 and the first 21 nt of exons 18 at vr2 may have important functions in mature myotubes. 3.4. Expression of a-DB2b protein in mouse skeletal muscle To show that the a-DB2b transcript detected by RT– PCR is actually translated into protein, we generated a polyclonal antibody, Ab 6224, that specifically recognizes the unique C-terminal tail encoded by exon Z (see Fig. 2B). Immunoblotting of soluble and insoluble fractions of skeletal muscle extracts with Ab 6224 revealed a single band of appropriate molecular weight (approx. 65 kDa) in both fractions (Fig. 5, right panel ), demonstrating that the a-DB2b protein is indeed expressed in skeletal muscle. No bands were detected in either fraction by the pre-immune serum ( Fig. 5, left panel ), indicating that the immunoreactivity detected by Ab 6224 was specific. This expression of a-DB2b protein in
Fig. 5. Immunoblot of skeletal muscle extract with a-DB2b-specific antibody. Adult mouse skeletal muscle homogenate was extracted with Triton X-100 and centrifuged to generate soluble (S) and insoluble (P) fractions. Two percent of each fraction was run on a SDS–PAGE gel and transferred to nitrocellulose. Identical samples were immunoblotted with an a-DB2b-specific polyclonal antibody, Ab 6224 (right panel ) or pre-immune serum ( left panel ). A single band of immunoreactivity was detected in both fractions by Ab 6224, but not by pre-immune serum, demonstrating the specificity of the antibody binding. The positions of molecular weight markers are indicated on the left (in kDa), and the single a-DB2b band is denoted with an arrowhead on the right.
mouse skeletal muscle suggests that it plays a functional role in this tissue.
4. Discussion In the present study, we cloned an a-dystrobrevin cDNA encoding a novel a-dystrobrevin isoform, a-DB2b, and provide evidence that alternatively spliced variants of the new isoform, as well as the other a-dystrobrevin isoforms, are differentially expressed in skeletal muscle, heart and brain, and developmentally regulated during muscle cell differentiation. In addition, we demonstrate that a-DB2b protein is expressed in mouse skeletal muscle, which suggests that it plays a functional role in this tissue. The data show that skeletal muscle and heart preferentially express a-dystrobrevin splice variants that contain exons 12 and 13 at vr3 and lack exon 9 at vr1. In contrast, brain expresses a variety of vr1 and vr3 splice variants within each a-dystrobrevin class, with no consistent splicing preference for all the classes. For example, at vr3, brain shows expression of two a-DB2b splice variants, one a-DB2a splice variant and three a-DB1 splice variants. The data also show that alternative splicing of the first 21 nt of exon 18 at vr2 in a-DB1 is restricted to skeletal muscle and heart. Expression studies in C2C12 muscle cells show that expression of the a-dystrobrevins is upregulated during differentiation, with a-DB1 and a-DB3 expressed earliest and a-DB2a and a-DB2b expressed later, and that alternative splicing of exons 12 and 13 at vr3 and exon 18 at vr2 is also upregulated during differentiation. The upregulation of specific exons in differentiating muscle
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
cells and their continued expression in adult muscle suggests that these exons encode amino acids that are important for the function of the a-dystrobrevins in mature muscle. Although the physiological function of a-dystrobrevin is unknown, its association with the sarcolemmal proteins utrophin, dystrophin, and syntrophins (Butler et al., 1992; Wagner et al., 1993; Ahn and Kunkel, 1995; Dwyer and Froehner, 1995; Suzuki et al., 1995; Peters et al., 1997a; Sadoulet-Puccio et al., 1997b; Peters et al., 1998) suggest that it may play a structural or signaling role at the cytoplasmic face of the sarcolemma. The utrophin/dystrophin binding site within a-DB1 has been mapped to a coiled coil domain encoded within exons 15 and 16 (SadouletPuccio et al., 1997b), while the syntrophin binding site has been mapped to residues 375–426 of Torpedo dystrobrevin (Dwyer and Froehner, 1995), corresponding to exons 13 and 14 of mouse a-DB1. The residues encoded by the alternatively spliced exons (12, 13 and 18a) are on either side of these protein binding domains, suggesting that these residues may modulate the ability of a-DB1 to interact with these proteins. The dystrophin and utrophin binding sites mapped in a-DB1 are also present in a-DB2; however, a-DB2a binds dystrophin, but not utrophin (Peters et al., 1998). It is possible that the unique C-terminal tail of a-DB2a prevents its interaction with utrophin, or conversely, that the unique C-terminal domain of a-DB1 confers the ability to interact with utrophin. The three syntrophin isoforms, a1, b1, and b2, which are encoded by different genes, are thought to function as modular adapters recruiting signaling proteins to the sarcolemma via their interaction with the dystrophin complex. Studies in mouse skeletal muscle have shown that different syntrophin isoforms preferentially associate with different dystrophin family members; a1 and b1 syntrophins preferentially associate with dystrophin and dystrobrevin, while b1 and b2 syntrophins preferentially associate with utrophin (Peters et al., 1997a). The antibodies used in that study did not distinguish between a-DB1 and a-DB2 or between their splice variants, so it remains to be determined which syntrophins associate with which a-dystrobrevin splice variants in skeletal muscle. We hypothesize that the a-dystrobrevins lacking the amino acids encoded by exons 12 and 13 would be unable to associate with syntrophins, since the syntrophin binding site is partly encoded by exon 13. If so, then syntrophin association with a-dystrobrevin would not occur until late in differentiation when transcripts containing exons 12 and 13 are expressed. It is also tempting to speculate that the alternative C-termini of a-DB2a and a-DB2b could determine which syntrophins are associated with each. Immunofluorescence studies in mouse skeletal muscle with isoform-specific antibodies have shown that a-DB1
487
is restricted to the neuromuscular junction, while a-DB2a is localized to both the sarcolemma and the neuromuscular junction (Peters et al., 1998). Within the neuromuscular junction, a-DB1 appears to be restricted to the crests of the junctional folds, while a-DB2a appears to be restricted to the troughs (Peters et al., 1998). This localization parallels that of utrophin and dystrophin which are restricted to the crests (Bewick et al., 1996) and troughs (Byers et al., 1991; Sealock et al., 1991) of the junctional folds, respectively. Interestingly, a-DB2a is retained at the neuromuscular junction in dystrophin/utrophin double knockouts, even though it preferentially associates with dystrophin (Peters et al., 1998). This suggests that there may be additional interactions with other molecules that could be facilitated by the unique C-terminus of a-DB2a. Although we have shown that a-DB2b protein is present in adult skeletal muscle, its subcellular localization remains to be determined. Its homology to a-DB2a and its similar expression pattern in developing muscle cells would suggest a similar localization to the sarcolemma and neuromuscular junction, but its unique C-terminal tail could potentially mediate a different subcellular localization. Studies are now underway to investigate the subcellular localization of the new isoform, a-DB2b, in skeletal muscle, and its biochemical interactions with other sarcolemmal proteins. The expression of a single variant of the a-DB2b mRNA and its encoded protein in mature skeletal muscle suggests that it may play an important structural or signaling role at the sarcolemma and/or neuromuscular junction.
Acknowledgements We thank Drs. Barry Knox, Mary Lou Vallano, and James McCasland for critically reviewing the manuscript, and we thank Christine Valway for excellent technical assistance in sequencing the a-dystrobrevin clones. This research was supported by grants from the National Institutes of Health (NS34070) and the American Heart Association, New York State Affiliate to M.M. Maimone.
References Ahn, A.H., Kunkel, L.M., 1995. Syntrophin binds to an alternatively spliced exon of dystrophin. J. Cell Biol. 128, 363–371. Ambrose, H.J., Blake, D.J., Nawrotzki, R.A., Davies, K.E., 1997. Genomic organization of the mouse dystrobrevin gene: comparative analysis with the dystrophin gene. Genomics 39, 359–369. Balasubramanian, S., Fung, E.T., Huganir, R.L., 1998. Characterization of the tyrosine phosphorylation and distribution of dystrobrevin isoforms. FEBS Lett. 432, 133–140. Bewick, G.S., Young, C., Slater, C.R., 1996. Spatial relationships of utrophin, dystrophin, beta-dystroglycan and beta-spectrin to ace-
488
R.E. Enigk, M.M. Maimone / Gene 238 (1999) 479–488
tylcholine receptor clusters during postnatal maturation of the rat neuromuscular junction. J. Neurocytol. 25, 367–379. Blake, D.J., Nawrotzki, R., Loh, N.Y., Go´recki, D.C., Davies, K.E., 1998. b-Dystrobrevin, a member of the dystrophin-related protein family. Proc. Natl. Acad. Sci. USA 95, 241–246. Blake, D.J., Nawrotzki, R., Peters, M.F., Froehner, S.C., Davies, K.E., 1996. Isoform diversity of dystrobrevin, the murine 87-kDa postsynaptic protein. J. Biol. Chem. 271, 7802–7810. Blau, H.M., Pavlath, G.K., Hardeman, E.C., Chiu, C.P., Silberstein, L., Webster, S.G., Miller, S.C., Webster, C., 1985. Plasticity of the differentiated state. Science 230, 758–766. Butler, M.H., Douville, K., Murnane, A.A., Kramarcy, N.R., Cohen, J.B., Sealock, R., Froehner, S.C., 1992. Association of the Mr 58,000 postsynaptic protein of electric tissue with Torpedo dystrophin and the Mr 87,000 postsynaptic protein. J. Biol. Chem. 267, 6213–6218. Byers, T.J., Kunkel, L.M., Watkins, S.C., 1991. The subcellular distribution of dystrophin in mouse skeletal, cardiac, and smooth muscle. J. Cell Biol. 115, 411–421. Carr, C., Fischbach, G.D., Cohen, J.B., 1989. A novel 87,000-Mr protein associated with acetylcholine receptors in Torpedo electric organ and vertebrate skeletal muscle. J. Cell Biol. 109, 1753–1764. Colledge, M., Froehner, S.C., 1998. Signals mediating ion channel clustering at the neuromuscular junction. Curr. Opin. Neurobiol. 8, 357–363. Dwyer, T.M., Froehner, S.C., 1995. Direct binding of Torpedo syntrophin to dystrophin and the 87 kDa dystrophin homologue. FEBS Lett. 375, 91–94. Frail, D.E., McLaughlin, L.L., Mudd, J., Merlie, J.P., 1988. Identification of the mouse muscle 43,000-dalton acetylcholine receptor-associated protein (RAPsyn) by cDNA cloning. J. Biol. Chem. 263, 15602–15607. Glass, D.J., Yancopoulos, G.D., 1997. Sequential roles of agrin, MuSK and rapsyn during neuromuscular junction formation. Curr. Opin. Neurobiol. 7, 379–384. Holzfeind, P.J., Ambrose, H.J., Newey, S.E., Nawrotzki, R.A., Blake, D.J., Davies, K.E., 1999. Tissue-selective expression of alphadystrobrevin is determined by multiple promoters. J. Biol Chem. 274, 6250–6258. Horowitz, D.S., Krainer, A.R., 1994. Mechanisms for selecting 5∞ splice sites in mammalian pre-mRNA splicing. Trends Genet. 10, 100–106. Kramarcy, N.R., Vidal, A., Froehner, S.C., Sealock, R., 1994. Association of utrophin and multiple dystrophin short forms with the
mammalian M(r) 58,000 dystrophin-associated protein (syntrophin). J. Biol. Chem. 269, 2870–2876. Nawrotzki, R., Loh, N.Y., Ruegg, M.A., Davies, K.E., Blake, D.J., 1998. Characterisation of a-dystrobrevin in muscle. J. Cell Sci. 111, 2595–2605. Peters, M.F., Adams, M.E., Froehner, S.C., 1997a. Differential association of syntrophin pairs with the dystrophin complex. J. Cell Biol. 138, 81–93. Peters, M.F., O’Brien, K.F., Sadoulet-Puccio, H.M., Kunkel, L.M., Adams, M.E., Froehner, S.C., 1997b. b-Dystrobrevin, a new member of the dystrophin family — identification, cloning and protein associations. J. Biol. Chem. 272, 31561–31569. Peters, M.F., Sadoulet-Puccio, H.M., Mark Grady, R., Kramarcy, N.R., Kunkel, L.M., Sanes, J.R., Sealock, R., Froehner, S.C., 1998. Differential membrane localization and intermolecular associations of alpha-dystrobrevin isoforms in skeletal muscle. J. Cell Biol. 142, 1269–1278. Puca, A.A., Nigro, V., Piluso, G., Belsito, A., Sampaolo, S., Quaderi, N., Rossi, E., Di Iorio, G., Ballabio, A., Franco, B., 1998. Identification and characterization of a novel member of the dystrobrevin gene family. FEBS Lett. 425, 7–13. Sadoulet-Puccio, H.M., Feener, C.A., Schaid, D.J., Thibodeau, S.N., Michels, V.V., Kunkel, L.M., 1997a. The genomic organization of human dystrobrevin. Neurogenetics 1, 37–42. Sadoulet-Puccio, H.M., Khurana, T.S., Cohen, J.B., Kunkel, L.M., 1996. Cloning and characterization of the human homologue of a dystrophin related phosphoprotein found at the Torpedo electric organ post-synaptic membrane. Hum. Mol. Gen. 5, 489–496. Sadoulet-Puccio, H.M., Rajala, M., Kunkel, L.M., 1997b. Dystrobrevin and dystrophin: an interaction through coiled-coil motifs. Proc. Natl. Acad. Sci. USA 94, 12413–12418. Sealock, R., Butler, M.H., Kramarcy, N.R., Gao, K.X., Murnane, A.A., Douville, K., Froehner, S.C., 1991. Localization of dystrophin relative to acetylcholine receptor domains in electric tissue and adult and cultured skeletal muscle. J. Cell Biol. 113, 1133–1144. Suzuki, A., Yoshida, M., Ozawa, E., 1995. Mammalian alpha1- and beta1-syntrophin bind to the alternative splice-prone region of the dystrophin COOH terminus. J. Cell Biol. 128, 373–381. Wagner, K.R., Cohen, J.B., Huganir, R.L., 1993. The 87K postsynaptic membrane protein from Torpedo is a protein–tyrosine kinase substrate homologous to dystrophin. Neuron 10, 511–522. Yaffe, D., Saxel, O., 1977. Serial passaging and differentiation of myogenic cells isolated from dystrophic mouse muscle. Nature 270, 725–727.