Positive and Negative Intronic Regulatory Elements Control Muscle-Specific Alternative Exon Splicing of Drosophila Myosin Heavy Chain Transcripts

Corresponding author: Charles P. Emerson, Jr., Department of Cell and Developmental Biology, 245 Anatomy and Chemistry Bldg., University of Pennsylvania School of Medicine, Philadelphia, PA 19104-6058., emersonc{at}mail.med.upenn.edu (E-mail)

Communicating editor: A. J. LOPEZ

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Alternative splicing of Drosophila muscle myosin heavy chain (MHC) transcripts is precisely regulated to ensure the expression of specific MHC isoforms required for the distinctive contractile activities of physiologically specialized muscles. We have used transgenic expression analysis in combination with mutagenesis to identify cis-regulatory sequences that are required for muscle-specific splicing of exon 11, which is encoded by five alternative exons that produce alternative "converter" domains in the MHC head. Here, we report the identification of three conserved intronic elements (CIE1, -2, and -3) that control splicing of exon 11e in the indirect flight muscle (IFM). Each of these CIE elements has a distinct function: CIE1 acts as a splice repressor, while CIE2 and CIE3 behave as splice enhancers. These CIE elements function in combination with a nonconsensus splice donor to direct IFM-specific splicing of exon 11e. An additional cis-regulatory element that is essential in coordinating the muscle-specific splicing of other alternative exon 11s is identified. Therefore, multiple interacting intronic and splice donor elements establish the muscle-specific splicing of alternative exon 11s.

THE regulated splicing of alternative exons in pre-mRNAs is a widely utilized genetic mechanism for generating tissue-specific protein isoforms (CHABOT 1996 ). Muscle makes extensive use of alternative splicing to produce the muscle-specific contractile protein isoforms in different muscle types to specialize their contractile properties (GEORGE et al. 1989 ; NADAL-GINARD et al. 1991 ; BANDMAN 1992 ; STANDIFORD et al. 1997 ). Consequently, muscle has been useful as a model cell type to investigate molecular mechanisms that regulate tissue-specific alternative exon splicing and has provided detailed information about these processes. For instance, molecular and biochemical studies of the mechanisms that regulate the alternative splicing of the chicken cardiac troponin T gene have shown that muscle-specific intronic splicing enhancers function in combination with a general exonic splicing enhancer (ESE; XU et al. 1993 ; RYAN and COOPER 1996 ) to direct the inclusion of exon 5 in the embryo. Similarly, the splicing of exon 2 of -tropomyosin (-TM) in smooth muscle cells requires purine-rich exonic splicing enhancers (DYE et al. 1998 ), but the proper use of this exon also requires the repression of the default splicing of exon 3 (MULLEN et al. 1991 ; GOODING et al. 1994 , GOODING et al. 1998 ). Studies of the chicken ß-tropomyosin gene transcript have demonstrated a role for secondary structure in alternative exon selection (CLOUET D'ORVAL et al. 1991 ; LIBRI et al. 1991 ), whereas the rat gene is regulated through 5' donor competition (CHEN et al. 1999 ). Together, these studies have demonstrated a diversity of roles played by cis-acting sequence information to provide either positive or negative interactions that ensure the proper tissue-specific use of alternative exons.

These same models have also been instrumental in defining the factors that interact with cis-acting elements to mediate the proper recognition of the alternative exons. For instance, the cTNT ESE has been shown to interact with several members of the SR family of splicing factors (RAMCHATESINGH et al. 1995 ), while the study of - and ß-TM transcripts has shown the involvement of hnRNP proteins as regulators of alternative splicing (PATTON et al. 1991 ; LIN and PATTON 1995 ; PEREZ et al. 1997 ; CHEN et al. 1999 ; LOU et al. 1999 ; SOUTHBY et al. 1999 ).

The analysis of muscle-specific alternative splicing has also contributed to the discovery that interactions between cis-acting elements and trans-acting factors can be influenced by cell or tissue-specific conditions. For instance, polypyrimidine tract binding protein is constitutively expressed, yet can specifically antagonize the use of -tropomyosin alternative exon 3 in smooth muscle (PEREZ et al. 1997 ). Similar events occur in other tissues and studies have shown that splice site selection can be differentially modified through tissue-specific phosphorylation of SR proteins (DU et al. 1998 ; XIAO and MANLEY 1998 ) or through variations in the abundance of splicing factors that act antagonistically (GALLEGO et al. 1997 ; LABOURIER et al. 1999 ) or have different strengths (STARK et al. 1999 ), demonstrating that the exact tissue-specific environment is likely to play a large role in directing alternative exon use. Because of this complexity, it is unlikely that in vitro approaches to the study of alternative splicing will be sufficient to fully replicate the conditions that exist to separate and define the specific pattern of alternative splicing of transcripts among different tissues.

We have developed an in vivo model for studying the mechanism that regulates muscle-specific alternative splicing that is based on the analysis of Drosophila skeletal muscle myosin heavy chain (MHC) gene transcripts. Drosophila has a single Mhc gene consisting of 19 exons, 5 of which are represented as 2 to 5 alternatively spliced exons encoding related sequences (Fig 1A; GEORGE et al. 1989 ). The alternative exons in each group are spliced in a mutually exclusive fashion, but the combinatorial use of alternative exons in different groups allows, in theory, for the expression of 480 MHC isoforms. However, the splicing of these exons is subject to precise muscle-specific regulation that allows individual specialized muscles typically to produce only a single MHC protein isoform (HASTINGS and EMERSON 1991 ). For example, Mhc exon 11 contains five alternatives, and transcripts expressed in the functionally distinct indirect flight muscle (IFM) and jump muscle tergal depressor of the trochanter (TDT) differ essentially only by the selection of exon 11e in the IFM and exon 11b in the TDT. Exon 11 encodes the MHC "converter" domain, which is thought to play a critical role in determining key functional properties of myosin (BERNSTEIN and MILLIGAN 1997 ; DOMINGUEZ et al. 1998 ), and the fact that exon 11e is utilized only in the highly specialized IFM suggests that normal flight function is dependent on the expression of this particular myosin isoform. This requirement indicates that IFM-specific splicing regulation is subject to intense evolutionary selection and, for this reason, we have pursued a detailed analysis of the cis-regulatory elements that control exon 11e splicing.

View larger version (30K): In this window In a new window Download PPT slide   Figure 1. (A) Diagram of the Drosophila 36B Mhc gene, showing the position of the 13 common, 5 alternative, and single differentially spliced exons. Also shown are the MHC domains that are encoded by the exons, including the myosin motor domain, the light chain binding region, and the MHC rod domain. The myosin rod protein (Mrp) gene is integral to Mhc and begins in Mhc intron 12. (B) Diagrams of the parental Mhc minigene construct used in this study. The gD1048 contains the entire exon 11 domain, with five alternatives that are flanked by common exons 10 and 12. Mhc exons 1 and 2, which encompass the Mhc promoter domain, are fused to the 5' end of exon 10 and the intron 12 and a portion of exon 13 fused to the LacZ reporter are included at the 3' end of the construct. Splicing products are assayed with RT-PCR using primer sets that specifically amplify the minigene transcript (E2/E12) across the exon 11 domain to identify correctly spliced products based on size. The exon 11 interval contains CIEs and those studied here are shown in the gD1048 transgene. CIE1 and CIE2 are 3' of the exon 11 donor in the exon 11e intron, while the large CIE3 element is distally positioned 200 nt 5' of the exon 12 acceptor. (C) List of conserved intronic elements important for exon 11e splicing in the IFM. CIE1 is identical between D. melanogaster and D. virilis and consists of a repeated ATGTACC element. The CIE2 element is a single element located within 24 nt of CIE1 in D. virilis and D. melanogaster. CIE3 is a larger element with several domains of high conservation (shaded). CIE3 also contains three ATCC repeats and an A/T-rich domain.

In studies preliminary to this report, we developed Mhc transgene reporters to investigate the regulatory sequences required for the alternative splicing of exon 11 alternatives in specialized muscles (Fig 1B; STANDIFORD et al. 1997 ). Our findings established that Mhc exon 11 contains sufficient cis- information to direct the normal, muscle-specific splicing of alternative exons. We also showed that nonconsensus exon 11 splice donors are essential for muscle-specific alternative exon splicing and, further, that conserved intronic elements located in the exon 11 interval are required components of the alternative splicing mechanism. In this report, we have focused on the mechanism used by the IFM to direct the inclusion of alternative exon 11e into its processed transcripts. Specifically, we use directed mutagenesis of transgenic Mhc exon 11 minigenes and RT-PCR analysis to show that the IFM has a strict requirement for exon 11e and defaults to an exon-skipping pathway in its absence. Further, we find that three conserved intronic elements are essential components of the mechanism that directs the utilization exon 11e in the IFM, and the in vivo analysis of these unique elements shows that they have either positive or negative activities that regulate the selection of exon 11e in the IFM. These data support a model in which there is a specific and unique alternative splicing mechanism in the IFM that is directed toward the inclusion of Mhc exon 11e in this muscle.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Constructs:
The construction of the gD1048, gD1060, gD1120, gD1168, gD1105, gD1090, gD1177, and gD1222 Mhc minigene constructs has been described previously (STANDIFORD et al. 1997 ). The gD1254, gD1256, and gD1265 constructs were made using the QuickChange procedure (Stratagene, La Jolla, CA) to mutagenize the conserved intronic elements (CIE) sites. Mutagenesis was conducted against an isolated 4-kb EcoRI fragment from gD1060, which was inserted into BlueScript SK- (Stratagene) and followed the manufacturer's protocol. The sense and antisense mutagenic oligos for CIE1 are CCGTCCTTCTCCACGTACCAGAAATCAATTC andGTGGAGAAGGACGGACAACGAGGATCAACG, respectively, and GTTTCGAACTATGGCTTAGGTGTATCTCCG and CCTAAGCCATAGTTCGAAACTGGAATTGAT for the CIE2 mutation. Mutagenized constructs were identified by sequence analysis and a EagI (exon 11e)-PacI (intron 12) fragment from the mutant subclone was used to replace the same fragment in gD1060 to introduce the CIE mutations. The double CIE mutant was made by sequential rounds of QuickChange mutagenesis using the same CIE1 and CIE2 mutagenic oligos. The gD1275, gD1276, and gD1320 mutants were made by QuickChange mutagenesis and the same CIE1 or CIE2 mutagenic oligos, but mutagenesis was done in a 1-kb EagI-KpnI fragment from gD1048. These same sites were used to introduce the mutagenized sequences into either gD1048 (gD1275 and gD1276) or gD1120 (gD1320). gD1334 was made with QuickChange using a mutant oligo to replace the exon 11e donor with that of exon 11b, which was done in a CIE2 mutagenized background. The gD1223 construct was generated by amplifying Mhc nucleotides (nt) 12010–12300 (numbering from GenBank accession no. M61229) using oligo-containing flanking AflII sites (sense-GCATCACTTTAAGACCAGGTTGATAAGTC; antisense-GCATCACTTAAGTTTCATTTGTGGATGC), which contain exon 11b and its native flanking intronic sequence, digesting this with AflII and placing it into the gD1060 construct that was linearized and filled at the AflII site. The gD1242 transgene was constructed by amplifying a section from gD1048 that included exon 2 to Mhc nt 12301 (intron 11b), using the exon 2 primer listed below and an Mhc exon 11b intron antisense primer (GCCGGCTCGAGGAAGAACCGCTTAAGCATAACG), which contains an XhoI restriction site at its 5' end. A second fragment was generated from gD1048 using a LacZ specific antisense primer and an 11d intron specific primer (GCGCGCTCGAGTTTGTATTTCATTTGTGGATGC) beginning at Mhc nt 13941, which also contains an XhoI site. Both fragments were digested with XhoI and were ligated to each other. This product was subjected to an additional PCR reaction to amplify between exon 2 and LacZ to produce a fragment that now contains an exon 11 interval that is deleted for exon 11c through CIE3. This fragment was digested with SacII and PacI and used to replace the same in gD1048, resulting in the gD1242 construct. The gD1152 gD1356, gD1357, and gD1362 constructs were made by amplifying across the CIE3 and inserting this fragment in the appropriate background.

Drosophila methods:
All Drosophila cultures were maintained at 25° on standard cornmeal molasses medium (ASHBURNER 1989 ). Transgenic flies were generated by standard methods. The strain y^{1 w67c23(2)} was used as the injection strain and as a "wild-type" reference strain in all of the experiments. A minimum of three independently derived transgenic lines was generated for each construct. Histochemical staining of flies for ß-galactosidase activity was performed as described by ASHBURNER 1989 .

Reverse transcriptase-dependent PCR:
RT-PCR methods were essentially those followed in STANDIFORD et al. 1997 . Briefly, for adult muscle, newly eclosed flies were collected and then soaked for a few minutes in petri plates containing 100% ethanol at room temperature. After this treatment the IFM, TDT, and other muscles become stiff and can be easily dissected away from each other. Separated IFM and TDT muscles were placed into 1.5-ml microfuge tubes containing 100% ethanol on ice until 10–15 flies per genotype were dissected. The tubes containing the muscles were spun at 2000 rpm, at 4°, and the ethanol was removed. The muscles were resuspended in 100 µl of H₂O, and 0.5 ml of RNAzol reagent (Sigma, St. Louis) was immediately added along with 20 µg of yeast transfer RNA (Sigma) as carrier. Total RNA was isolated with the RNAzol reagent using the manufacturer's instructions. Following the isopropanol precipitation the RNA pellet was resuspended in 100 µl reverse transcriptase buffer with 100 µM dNTPs, 40 units RNAsin (Promega, Madison, WI), 200 units SuperScript II reverse transcriptase (GIBCO-BRL, Gaithersburg, MD), and an antisense oligonucleotide primer to Mhc exon 12. Reverse transcription reactions were at 42° for 1 hr. For the primary PCR reaction, 5 µl of the reverse transcriptase reaction were used in a 100-µl reaction under the following conditions: 1 cycle at 95° for 5 min; 30 cycles of 95° for 30 sec, 55° for 1 min, 72° for 1 min; and 1 cycle of 74° for 4 min. Five microliters of the primary PCR reaction was used for the secondary PCR reaction and the same PCR cycle profile was followed. RNAs from larval muscles were prepared identically, except that 10–15 first instar larvae of each genotype were directly homogenized in RNAzol prior to RT-PCR analysis. The sequences for all of the oligonucleotide primers used in these experiments are presented below:

• Exon 2 sense: GACTCGAAGAAGTCTTGCTG

• Exon 12 inner antisense: CTGACCCAGGACACCGGCGCG

• Exon 12 outer antisense: GGACATGATCTTGCCCAGACGC

• Skip-specific primer: CTACGCCACAAGAAGGCG

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Disruption of exon 11e splicing leads to exon 10–12 skip splicing in the IFM:
Previously, we established that a Mhc minigene containing the entire exon 11 domain (gD1048; Fig 1B) has sufficient regulatory sequences to direct the correct alternative slicing of exon 11 alternatives in their appropriate muscles (STANDIFORD et al. 1997 ). Specifically, RT-PCR and LacZ reporter gene analysis were used to show that the selection of alternative exons from transcripts arising from the gD1148 minigene exactly matched the use of exons from the endogenous Mhc gene transcripts. For instance, exon 11e from the gD1048 minigene was used exclusively in the IFM and the use of no other exon was detectable in this muscle, which conforms to the precise, IFM-specific use of the endogenous exon 11e. Extensive mutational analysis on the Mhc minigene revealed that several classes of alterations could significantly disrupt the normal use of exon 11e in the IFM: (1) the removal of exon 11e itself, (2) the conversion of the nonconsensus exon 11e splice donor to consensus, and (3) the removal of a conserved intronic sequence (CIE3; Fig 1; STANDIFORD et al. 1997 ). Together, these results indicate that there is a IFM-specific splicing regulatory mechanism that is designed to promote exon 11e splicing in this muscle.

A key observation from these earlier studies was that the deletion of exon 11e from the gD1048 transgene does not lead to default splicing of other alternative exon 11s in the IFM nor does this deletion effect the splicing of any other alternative exon 11 in their specific muscle types (STANDIFORD et al. 1997 ). Here, we have defined the molecular response of the IFM-specific splicing apparatus to the loss of exon 11e using an RT-PCR assay that shows that a truncated transcript results in the IFM when exon 11e is removed from the minigene (gD1168; Fig 2). Sequence analysis shows that the truncated transcript results from the skipping of all other alternatives and the direct splicing of constitutive exons 10 and 12 in the IFM. Using a PCR primer designed to specifically assay for exon 10–12 skip splicing, we also find that the gD1168 transgene produces a skip-spliced transcript only in the IFM and not in larval muscles where exon 11e is not normally utilized. Thus, in the absence of sequence information required to make the correct alternative exon selection in the IFM, the splicing machinery defaults to an exon 10–12 skip-splicing pathway, further supporting our data that exon 11e splicing is controlled by IFM-specific splicing machinery.

View larger version (47K): In this window In a new window Download PPT slide   Figure 2. Exon skipping occurs in the absence of IFM-specific exon 11e sequences. When the IFM-specific exon 11e alternative is deleted in gD1168, RT-PCR using the exon 2–12 primer set detects only a truncated transcript in the IFM, although this analysis shows the expected size of the transcript in the IFM of flies expressing the wild-type gD1048 transgene. The size of this transcript is consistent with a skip-spliced product, which was confirmed by sequencing (skip product). A PCR primer (skip primer) that specifically amplifies skipped products was able to detect the skip product in the IFM of the gD1168 flies, but not in larval muscles or in the IFM of flies expressing the wild-type gD1048 minigene, showing the initiation of the skip-splicing pathway in the IFM in the absence of required cis-acting sequence information.

Distal intronic sequences have IFM-specific and global regulatory activities in exon 11e splicing:
Exon 10–12 skip splicing provides a sensitive method for monitoring disruptions in IFM-specific splicing and we have applied this assay in combination with a series of mutated minigene constructs to identify further cis-acting elements required for the IFM-specific splicing of exon 11e. A large conserved intronic domain in the exon 11 domain (CIE3; Fig 1) located between exon 11d and exon 12 was examined earlier for its role in directing IFM-specific splicing of exon 11e (ICR; STANDIFORD et al. 1997 ), and deletion of this element did not alter splice choice specification of any other alternative exon when deleted. However, ß-galactosidase expression, which is a function of the normal processing of the transgene, was found to be suppressed in the absence of CIE3, indicating a loss of processing efficiency. An examination of this effect here with the use of the skip-splicing assay shows that the observed loss of splicing efficiency in a CIE3 (gD1120) deletion background actually results from the activation of skip splicing and a routing of transcripts into this pathway (Fig 3). Further, the loss of CIE3 disrupts splicing only in the IFM, and transcripts arising from gD1120 in the TDT, which utilizes exon 11b, are normally spliced and no skip product is detected. Because our previous results showed that no other alternative is aberrantly included in the IFM in the absence of CIE3 (STANDIFORD et al. 1997 ), these data indicate that CIE3 is required to specially enhance exon 11e splicing in the IFM.

View larger version (30K): In this window In a new window Download PPT slide   Figure 3. Sequences distal to exon 11e have IFM specific and global regulatory activities. Although the complete Mhc exon 11 interval faithfully directs exon 11e splicing in the IFM in the gD1048 transgene and no skip products are detected, a deletion of the distal conserved intronic element, CIE3, in gD1120 induces exon skipping. This effect is specific for the IFM, and RNAs collected from the TDT that are assayed by RT-PCR do not contain any skip product. The gD1242 transgene is unable to direct normal splicing of alternative exons in all muscles, including the IFM and larval muscles. Normal splicing is restored in the IFM in the further deleted gD1060 construct, although exon skipping occurs in all other muscles including larval muscles (see Fig 5). The relative level of skip splicing induced by each construct is presented at the right of each construct diagram for this and subsequent experiments, where (---) equals no skip splicing is detected, while (+++) indicates strong induction of skip splicing.

Interestingly, while the loss of CIE3 appears to disrupt the splicing of only exon 11e, a larger 3' deletion that removes 11c through CIE3 was found to have more global effects on exon 11 alternative splicing (gD1242; Fig 3). This deletion disables the normal processing of all remaining exons (11e, 11a, 11b) and only skip-spliced products are detected in the IFM and larval muscles. In contrast, a further deletion (gD1060) that leaves exon 11e as the only alternative is normally spliced in the IFM, but is skipped in all other muscles (see Fig 6). Thus, sequences deleted by the gD1242 construct appear to be important in coordinating the selection of alternatives in a multi-exon context. We refer to this activity as the splicing coordinator (SC).

View larger version (28K): In this window In a new window Download PPT slide   Figure 4. Specific exons and nonconsensus splice donors are not required for IFM exon 11e splicing. To examine the role of conserved splice donors and exonic sequences, transcripts arising in the IFM from constructs that contained an exon 11e to exon 11b swap (gD1222), an exon 11e to exon 11b splice donor conversion (gD1105), or an exon 11e splice donor conversion to consensus (gD1090) were analyzed with RT-PCR and the skip specific primers. In each case, however, only the normally processed transcript was detected.

View larger version (36K): In this window In a new window Download PPT slide   Figure 5. (A) IFM-specific regulatory elements have context-dependent activities. When CIE1 or CIE2 substitution mutations were introduced into the gD1048 transgene (gD1175 and gD1176, respectively), there was no effect on exon 11e use in the IFM, as measured by the detection of the exon 11 spliced transcripts and the absence of exon skipping. The combination of the CIE2 and CIE3 mutations (gD1320) did not enhance skip splicing over that seen in the CIE3 deletion alone (gD1120; Fig 3). (B) CIE1 and CIE2 have IFM-specific regulatory functions. The truncated Mhc minigene, gD1060, faithfully directs the inclusion of exon 11e in the IFM and contains CIE1 and CIE2. The replacement of CIE2 with random sequence in gD1254 results in a strong induction of skip splicing as measured both by the exon 2–12 and skip-specific primers. In contrast, the substitution of CIE1 in the gD1256 did not effect normal splicing or induce skip splicing. The combination of CIE1 and CIE2 mutations in the gD1265 transgene resulted in the return of normal splicing to near wild-type levels. (C) The single or combined loss of CIE1 or CIE2 does not lead to the activation of 11e splicing in larval muscles, indicating that these elements function specifically in the IFM.

View larger version (38K): In this window In a new window Download PPT slide   Figure 6. CIE3 is a IFM-specific splicing enhancer for exon 11e. Transcripts arising from the gD1152 construct are properly regulated for the specific inclusion of exon 11e in the IFM and no skipping is detected. The removal of CIE2 when CIE3 is intact in the gD1357 transgene does not alter the normal use of exon 11e in the IFM, nor does the concurrent loss of CIE1 and CIE2 in the gD1356 construct, demonstrating that the CIE3 element can act as an IFM-specific splicing enhancer for exon 11e. Further, when only the 5' half of CIE3 is included in the gD1362 construct, some skipping occurs but at a level below that found in gD1254 (Fig 5), where CIE2 and CIE3 are removed, suggesting that the CIE3 splice-enhancing activity is localized to the 5' domain of the element.

Exon 11e splice selection is controlled exclusively through intronic elements:
As seen in Fig 3, exon 11e selection in the IFM is specifically promoted by the CIE3 splicing enhancer and its loss leads to the activation of skip splicing in the IFM. However, alternative spliced exons are often regulated through the use of ESEs (LIU et al. 1998 ; NAGEL et al. 1998 ; ZHENG et al. 1998 ; GERSAPPE and PINTEL 1999 ). The possible function of such regulatory elements in exon 11 alternative splicing has been examined previously, but only with respect to splice choice specificity. These earlier results showed that swapping alternative exons also converted their use (STANDIFORD et al. 1997 ), indicating that exon position and not its specific sequence determined its muscle-specific selection. Further, although all alternative exon donor sites are nonconsensus, but well conserved and unique to each alternative, they were found to not influence exon selection. However, while these exonic elements are not providing specificity information, it is possible that they are important as splicing enhancers, similar to the CIE3 element. To test this possibility, we used the exon 10–12 skip-splicing assay to examine the effects of exon and donor swaps on IFM-specific splicing. As shown in Fig 4, we found that, in the context of the complete exon 11 domain, an exon swap involving replacement of exon 11e with 11b (gD1222) does not promote exon 10–12 skip splicing in the IFM, although 11b is now selected for inclusion in the IFM (STANDIFORD et al. 1997 ). The replacement of the highly conserved exon 11e donor with that from exon 11b (gD1105) or a consensus donor (gD1090) also does not induce the skipping reaction or change the use of exon 11e in the IFM (STANDIFORD et al. 1997 ). Thus, consistent with our earlier findings, neither exon 11e nor its unique nonconsensus splice donor are required for the efficient selection and splicing of this exon in the IFM. Therefore, the sequence information required to direct IFM-specific exon 11e splicing must reside in the exon 11 intronic sequence. In addition to CIE3, we identified two other conserved elements within this intronic domain, CIE1 and CIE2 (STANDIFORD et al. 1997 ), and these sequences have been subjected to further mutagenesis experiments to identify their role as potential splicing regulatory elements.

CIE1 and CIE2 elements have distinct negative and positive regulatory functions for IFM-specific, alternative splicing of exon 11e:
The CIE1 element is proximal to exon 11e and consists of a twice-repeated ATGTACC sequence in D. melanogaster and D. virilis (Fig 1C), but is represented as a single element in D. hydei (MIEDEMA et al. 1994 ). CIE2 is a complex element with the consensus AGTGCTGT^G/_CT, which is separated from CIE1 by 20 nt in D. melanogaster and D. virilis, but is contiguous with CIE1 in D. hydei. To determine whether CIE1 and CIE2 elements are required for exon 11e splicing in the IFM, mutations were introduced into CIE1 and CIE2 in the gD1048 background, and transcripts from IFM of transgenic flies were tested for exon 11e splicing and skip splicing using RT-PCR. As shown in Fig 5A, the substitution of a random linker in CIE1 (CCGTCCTTCTCCAC; gD1175) or CIE2 (GAACTATCCC; gD1176) does not promote exon 11e skip splicing or lead to the misexpression of any other alternative in the IFM (not shown), suggesting that these CIE elements do not participate in splicing regulation in this context or are redundant with other elements in the gD1048 transgene.

However, when further examined using the truncated gD1060 background, both CIE1 and CIE2 were found to be critical for the proper use of exon 11e in the IFM. When the same CIE2 mutation from gD1176 was placed in the truncated context (gD1254; Fig 5B), there was a significant loss of exon 11e splicing in the IFM, with 50% of the transcripts undergoing exon 10–12 skip splicing, while the substitution of CIE1 with random sequence did not alter the IFM-specific splicing of exon 11e or induce any detectable skip splicing (gD1256; Fig 5B). However, when the CIE1 and CIE2 mutations were combined in the gD1265 transgene, skip splicing was greatly reduced from that observed in the CIE2 substitution alone. These results indicate that CIE2 is required to promote exon 11e use in the IFM and that CIE1 can act to repress this activity. The substitutions of CIE1 and CIE2 made here do not result in the activation of splicing in larval muscles (Fig 5C) or in the TDT (not shown), showing that these substitutions do not themselves confer gain-of-function effects, and sequence analysis of the spliced products revealed products that resulted from only exon 11e to 12 (normal) or exon 10 to 12 (skip) splice reactions, suggesting that no new splice sites were added or activated by the CIE1 and CIE2 substitutions. Thus, these mutations appear to specifically affect the activity of these elements.

While CIE2 appears required for the normal utilization of exon 11e, particularly in the presence of CIE1, these experiments also show that normal splicing can occur in its absence, suggesting either that the alternative splicing regulatory mechanism also includes additional positive-acting regulatory elements found elsewhere in the gD1060 interval or that a default splicing mechanism that includes any alternative exon is engaged in the absence of CIE1 and CIE2. This second possibility, however, is not favored since a construct that is similar to gD1060 except that it contains exon 11b and its flanking introns is completely skipped in the IFM (not shown).

CIE3 is a positive regulator of IFM-specific exon 11e splicing:
CIE3 is required in the context of the full-length exon 11 to properly promote the inclusion of exon 11e in the IFM. The ability of this element to function in a truncated context was determined in Fig 6 to test whether CIE3 acts directly with exon 11e or whether additional elements within the domain are also needed for its function. When placed into a gD1060 background where both CIE1 and CIE2 are intact (gD1152), no misexpression or skipping of exon 11e can be detected. When CIE3 is present in the absence of the CIE2 splicing enhancer (gD1356), exon 11e is normally spliced in the IFM and no exon 10–12 skipping is detected, indicating that CIE3 can functionally substitute for CIE2 in this context. Consistent with this observation is the fact that CIE3 also directs the inclusion of exon 11e into the IFM in the absence of both CIE1 and CIE2 (gD1357). Further, when the 3' half of CIE3 is deleted in combination with the CIE2 substitution (gD1362), some skipping occurs, but this appears less than that seen in the gD1254 construct (Fig 5), where CIE3 is fully deleted. Thus, CIE3 appears to enhance the ability of the splicing apparatus to recognize exon 11e in the IFM in the absence of CIE1 and CIE2 and independently of any additional information contained in the interval removed by the gD1060 deletion. Interestingly, the 5' domain of the CIE3 is well conserved and contains several purine-rich elements that could serve as sites for interactions with SR proteins (SCHAAL and MANIATIS 1999 ).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In this study we have utilized transgenic analysis and mutagenesis to identify the intronic regulatory elements in Mhc transcripts that control the muscle-specific splicing of the alternative exon 11, which encodes the "converter domain" of Drosophila MHC isoforms (BERNSTEIN and MILLIGAN 1997 ). Experiments have been directed specifically to the analysis of regulatory elements that control the splice choice specificity of exon 11e, which is spliced exclusively in the highly specialized IFM (HASTINGS and EMERSON 1991 ). These regulatory elements have been identified using RT-PCR assays that detect the spliced products of transcripts arising from wild-type and mutated transgenic Mhc exon 11 minigenes in the IFM. The results of our analyses reveal that, generally, the IFM-specific use of exon 11e depends on both positive and negative interactions that appear to act either locally on exon 11e or globally across the entire exon 11 domain. Specifically, we have identified four exon 11 splicing regulatory elements that have distinct functions. Two elements, CIE1 and CIE2, are located immediately 3' of exon 11e and act locally on IFM-specific exon 11e splicing in the absence of the other four alternative exon 11s. Independent of CIE1 and CIE2 are two distal elements, CIE3, which also positively directs the use of exon 11e in the IFM, and a SC, which appears necessary to facilitate alternative splicing of exon 11e when in the context of multiple alternative exon 11s. Therefore, a hierarchy of regulatory mechanisms controls the IFM-specific splicing of exon 11e. These data extend our previous findings to provide compelling evidence that the IFM splicing machinery is exquisitely specific for splicing of exon 11e in the IFM.

Exon 11e splicing involves a IFM-specific mechanism:
In contrast to other known alternatively spliced transcripts, where competitive interactions involving splice site strengths or the abundance and state of various binding proteins can influence the balance of exon selection within a specific tissue, alternative splicing of Mhc exon 11e appears to involve discrete, muscle-specific interactions that ensure the exact and exclusive use of the correct exon. This property is clearly shown in the splicing behavior of the gD1168 transgene, where exon 11e alone is removed (Fig 2). In this case, the IFM splicing machinery does not select any other exon 11 in the transgene transcript but instead completely defaults to the skip-splicing mode, showing that the IFM has an intrinsic requirement for a specific alternative exon and cannot replace this with another in its absence. The same also appears true for other non-IFM muscles, which do not utilize exon 11e in the transcripts from the gD1060 transgene, but instead default to skip splicing.

Exon 11e is positively selected in the IFM:
The basis for the IFM exclusivity of exon 11e splicing likely involves exon 11e-specific splicing factors that are uniquely present in the IFM and absent in all other muscles. We hypothesize that such IFM-specific factors would interact with the CIE elements and potentially other intronic regulatory sequences in the exon 11 domain to promote exon 11e splicing. Alternatively, the IFM specificity of exon 11e splicing could involve additional negative regulatory factors that repress the splicing of other exon 11s in the IFM and/or repress exon 11e splicing in non-IFM muscles. At this time we do not favor this possibility since, to date, we have not identified mutations in regulatory elements that induce the misexpression of exon 11e in non-IFM muscles or misexpression of other exon 11s in the IFM. An exception to this conclusion is the finding that the conversion of the nonconsensus exon 11e splice donor to a consensus donor leads to the exclusive selection of exon 11e in all muscles. This finding demonstrates the intrinsic weakness of the normal nonconsensus 5' splice donor sequences and suggests that the activation of the donor is the regulatory focus of the alternative splicing mechanism. Thus, our model of exon 11 alternative splicing regulation is predicated upon the muscle-specific activation of donor sites to effect the splicing of the appropriate alternative exon.

Conserved intronic elements are specific regulators of exon 11e:
The ability of CIE1 to repress the selection of exon 11e occurs only in the IFM, and the removal of this element does not lead to the activation of exon 11e splicing in other muscles even in situations where exon 11e is the only alternative (gD1256, Fig 5). Typically, splicing repressors function to prevent the use of alternative exons in inappropriate tissue, and what role the IFM-specific repression of exon 11e splicing plays is an intriguing question. Given the length of the exon 11 domain and number of alternatives it contains, it is possible that this component of the mechanism is required to regulate and coordinate the timing of exon 11e splicing during the period when other alternatives in the domain are being transcribed, which serves to prevent the inappropriate definition of the exon 11e donor until the strong exon 12 splice acceptor becomes available (ROBBERSON et al. 1990 ). Interestingly, mutual exclusivity is stringently maintained among the exon 11 alternatives, and none of the mutations that we have examined lead to the splicing of multiple alternative exons. The ability to repress exon definition until the entire interval is transcribed is potentially part of this mechanism and could be a feature of all exon 11 alternatives.

CIE2 has an activity reciprocal to that of CIE1 and likely positively regulates exon 11e splicing. This model is consistent with the observation that exon 11e is not efficiently spliced in the absence of CIE2 in the gD1254 transgene, resulting in exon 10–12 skip splicing in the IFM. We find it particularly significant that the CIE1 and CIE2 sites are positioned immediately adjacent to one another in D. hydei and are only 24 nt apart in D. melanogaster and D. virilis, consistent with the possibility of interactions between these elements. However, the fact that IFM-specific use of 11e still occurs in the absence of CIE1 and CIE2 indicates that these elements do not encompass all the information for 11e activation, although our current data show that these will also reside in the intronic sequence found in gD1060. Further, we have not directly tested the role of secondary structure in exon 11e splicing and do not exclude the possibility that it plays a role in CIE1 and CIE2 function in particular or in exon 11 alternative splicing generally. However, stable RNA structures are not predicted to form within or between CIE1 and CIE2, suggesting that these elements do not themselves play a direct role in aspects of exon 11e splicing that may be regulated through secondary structure.

CIE3 is a large element located at the 3' end of the exon 11 domain that is required to direct the efficient use of exon 11e when the entire intron is present. Our analysis of the truncated exon 11 background demonstrates that this element also acts as a splicing enhancer to direct exon 11e use in the IFM and can substitute for CIE2 in this context. However, while both CIE2 and CIE3 appear to act as exon 11e-specific splicing enhancers, their dissimilar composition and positions within the primary transcript suggest that these two elements function through separate mechanisms. For instance, while CIE2 appears to be positioned where it can interact with the splice donor of exon 11e, the proximity of CIE3 to the putative branchpoint (AUCUAAC) and polypyrimidine tract for exon 12 might indicate that it acts to influence the activity of these elements in the IFM.

As seen in Fig 3, the loss of the domain containing exon 11c through CIE3 (gD1242) results in the failure of either remaining alternative (11e and 11b) to be processed and the induction of skip splicing in all muscles. Since a larger deletion that also removes exon 11b (gD1060) is permissive for exon 11e splicing in the IFM, these findings suggest that information removed in gD1242 is important for splicing regulation when multiple alternatives are present. How such a SC might function is unclear, but it may act to repress competition or other interactions among multiple alternatives that prevent efficient recognition of any alternative, which, in turn, promotes exon skipping. Experiments are in progress to localize the SC sequence element to define its function in Mhc exon 11 alternative splicing.

It is clear that a complex and interconnected regulatory hierarchy is required to ensure the proper selection of exon 11e and that perturbations in this regulation lead to alternative exon skipping. With this consideration in mind, alternative exon skipping could reflect disruptions in two classes of splicing regulatory processes: (1) fidelity mechanisms that control the specificity of alternative exon splicing (e.g., interaction of muscle-specific splicing factors, activators, and repressors with alternative exon-specific regulatory sequences) and (2) alternative/constitutive exon selection mechanisms that control the differential splicing of alternative exons over the splicing between flanking constitutive exons, which could involve active processes such as repressors that suppress exon skipping between constitutive exons or kinetic processes that favor formation of spliceosomes on alternative exons over formation of spliceosomes for flanking constitutive exons. The results of our mutagenesis studies of IFM-specific exon 11e splicing regulation have identified splicing regulatory elements that appear to fall into both the fidelity and differential selection classes of regulatory processes. A model to account for the role of the elements described here is presented in Fig 7.

View larger version (16K): In this window In a new window Download PPT slide   Figure 7. CIE-mediated regulation of Mhc exon 11e in the IFM. A model for the role of intronic elements in the regulation of exon 11e alternative splicing in the IFM is based on the selective enhancement in the strength of the exon 11e nonconsensus donors (see text). CIE2 acts as a positive regulator of this event and is predicted to bind factors that will either directly or indirectly attract the U1 snRNP to the 11e donor. The CIE1 element has an IFM-specific repressor activity perhaps required to coordinate the timing of exon 11e donor utilization relative to other splicing processes. CIE1 and CIE2 are positioned close to each other, and factors that recognize CIE2 might inactivate CIE1 repression. CIE3 is required for high-fidelity use of exon 11e in the presence of other exon 11s and can substitute for CIE2 in the truncated "gD1060" background. This element might directly affect the strength of the exon 11e splice donor, but its proximal position relative to the presumptive exon 12 branch point sequence (BPS) suggests that the activity of CIE3 promotes interactions between the 11e donor and the exon 12 acceptor. Deleting the exon 11c–CIE3 interval results in complete loss of muscle-specific alternative exon 11 splicing, as evidenced by skip splicing in all muscles. Thus, this domain appears to contain an SC activity that is required to promote the selection of individual exons and to suppress skip splicing when multiple alternative exon 11s are present.

Multiple strategies exist to direct Mhc alternative splicing:
The cis-mechanism that directs the alternative splicing for several different Mhc alternative exons has been examined (HODGES and BERNSTEIN 1992 ; STANDIFORD et al. 1997 ; DAVIS et al. 1998 ; HODGES et al. 1999 ), and from these studies it has become clear that a variety of mechanisms exist to regulate the alternative splicing of exons from different alternative groups. For instance, our work here indicates that exon 11 selection is mediated through 5' donor enhancement, while the selection of exon 7 alternatives appears to proceed through a mechanism that is directed at strengthening the 3' acceptor (DAVIS et al. 1998 ), and the inclusion of exon 18 appears to require a specific conserved polypyrimidine tract to enhance the recognition of weak donor and acceptor sites (HODGES et al. 1999 ). The apparent diversity of alternative splicing mechanisms may underlie the observation that there appear to be no requirements for the coexpression of alternatives from different exon groups (HASTINGS and EMERSON 1991 ) and may be important in conferring a greater flexibility to generate MHC isoform diversity.

A surprising result from our study is the stringent requirement by the IFM for exon 11e and the discrete shift to a skip-splicing pathway when required regulatory elements such as CIEs are replaced or removed. The single response mode of exon skipping when confronted with a number of different regulatory defects provides a unique opportunity to define multiple components of the alternative splicing apparatus in the IFM using genetic screens designed to identify skip splicing. Having defined the specific intronic and splice donor elements that are required for IFM-specific exon 11e splicing, the challenge now is to identify the splicing factors that determine the precise muscle specificity of alternative exon 11 splicing in Drosophila.

	ACKNOWLEDGMENTS

We thank Beth Bucher (University of Pennsylvania) for advice and comments on the manuscript and members of the laboratory for their interest and discussion throughout the course of this project. This work was supported by a U.S. Army Breast Cancer Initiative Postdoctoral Fellowship (to D.M.S.), a Muscular Dystrophy Association Postdoctoral Fellowship (to M.B.D.), an American Cancer Society grant NP 841 (to C.P.E.), and a National Institutes of Health R01-AR42363 grant (to C.P.E.).

Manuscript received May 2, 2000; Accepted for publication September 28, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ASHBURNER, M., 1989 Drosophila: A Laboratory Handbook. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

BANDMAN, E., 1992  Contractile protein isoforms in muscle development. Dev. Biol. 154:273-283[Medline].

BERNSTEIN, S. and R. MILLIGAN, 1997  Fine tuning a molecular motor: the location of alternative domains in the Drosophila myosin head. J. Mol. Biol. 217:1-6.

CHABOT, B., 1996  Directing alternative splicing: cast and scenarios. Trends Genet. 12:472-478[Medline].

CHEN, C. D., R. KOBAYASHI, and D. M. HELFMAN, 1999  Binding of hnRNP H to an exonic splicing silencer is involved in the regulation of alternative splicing of the rat beta-tropomyosin gene. Genes Dev. 13:593-606[Abstract/Free Full Text].

CLOUET D'ORVAL, B., Y. D'AUBENTON-CARAFA, J. M. BRODY, and E. BRODY, 1991  Determination of an RNA structure involved in splicing inhibition of a muscle-specific exon. J. Mol. Biol. 221:837-856[Medline].

DAVIS, M. B., J. DIETZ, D. M. STANDIFORD, and C. P. EMERSON, JR., 1998  Transposable element insertions respecify alternative exon splicing in three Drosophila myosin heavy chain mutants. Genetics 150:1105-1114[Abstract/Free Full Text].

DOMINGUEZ, R., Y. FREYZON, K. M. TRYBUS, and C. COHEN, 1998  Crystal structure of a vertebrate smooth muscle myosin motor domain and its complex with the essential light chain: visualization of the pre-power stroke state. Cell 94:559-571[Medline].

DU, C., M. E. MCGUFFIN, B. DAUWALDER, L. RABINOW, and W. MATTOX, 1998  Protein phosphorylation plays an essential role in the regulation of alternative splicing and sex determination in Drosophila. Mol. Cell 2:741-750[Medline].

DYE, B. T., M. BUVOLI, S. A. MAYER, C. H. LIN, and J. G. PATTON, 1998  Enhancer elements activate the weak 3' splice site of alpha-tropomyosin exon 2. RNA 4:1523-1536[Abstract].

GALLEGO, M. E., R. GATTONI, J. STEVENIN, J. MARIE, and A. EXPERT-BEZANCON, 1997  The SR splicing factors ASF/SF2 and SC35 have antagonistic effects on intronic enhancer-dependent splicing of the beta-tropomyosin alternative exon 6A. EMBO J. 16:1772-1784[Medline].

GEORGE, E. L., M. B. OBER, and C. P. EMERSON, JR., 1989  Functional domains of the Drosophila melanogaster muscle myosin heavy-chain gene are encoded by alternatively spliced exons. Mol. Cell. Biol. 9:2957-2974[Abstract/Free Full Text].

GERSAPPE, A. and D. J. PINTEL, 1999  CA- and purine-rich elements form a novel bipartite exon enhancer which governs inclusion of the minute virus of mice NS2-specific exon in both singly and doubly spliced mRNAs. Mol. Cell. Biol. 19:364-375[Abstract/Free Full Text].

GOODING, C., G. C. ROBERTS, G. MOREAU, B. NADAL-GINARD, and C. W. SMITH, 1994  Smooth muscle-specific switching of alpha-tropomyosin mutually exclusive exon selection by specific inhibition of the strong default exon. EMBO J. 13:3861-3872[Medline].

GOODING, C., G. C. ROBERTS, and C. W. J. SMITH, 1998  Role of an inhibitory pyrimidine element and polypyrimidine tract binding protein in repression of a regulated -tropomyosin exon. RNA 4:85-100[Abstract].

HASTINGS, G. A. and C. P. EMERSON, JR., 1991  Myosin functional domains encoded by alternative exons are expressed in specific thoracic muscles of Drosophila. J. Cell Biol. 114:263-276[Abstract/Free Full Text].

HODGES, D. and S. I. BERNSTEIN, 1992  Suboptimal 5' and 3' splice sites regulate alternative splicing of Drosophila melanogaster myosin heavy chain transcripts in vitro. Mech. Dev. 37:127-140[Medline].

HODGES, D., R. M. CRIPPS, M. E. O'CONNOR, and S. I. BERNSTEIN, 1999  The role of evolutionarily conserved sequences in alternative splicing at the 3' end of Drosophila melanogaster myosin heavy chain RNA. Genetics 151:263-276[Abstract/Free Full Text].

LABOURIER, E., E. ALLEMAND, S. BRAND, M. FOSTIER, and J. TAZI et al., 1999  Recognition of exonic splicing enhancer sequences by the Drosophila splicing repressor RSF1. Nucleic Acids Res. 27:2377-2386[Abstract/Free Full Text].

LIBRI, D., A. PISERI, and M. Y. FISZMAN, 1991  Tissue-specific splicing in vivo of the beta-tropomyosin gene: dependence on RNA secondary structure. Science 252:1842-1845[Abstract/Free Full Text].

LIN, C. H. and J. G. PATTON, 1995  Regulation of alternative 3' splice site selection by constitutive splicing factors. RNA 1:234-245[Abstract].

LIU, H. X., M. ZHANG, and A. R. KRAINER, 1998  Identification of functional exonic splicing enhancer motifs recognized by individual SR proteins. Genes Dev. 12:1998-2012[Abstract/Free Full Text].

LOU, H., D. M. HELFMAN, R. F. GAGEL, and S. M. BERGET, 1999  Polypyrimidine tract-binding protein positively regulates inclusion of an alternative 3'-terminal exon. Mol. Cell. Biol. 19:78-85[Abstract/Free Full Text].

MIEDEMA, K., H. HARHANGI, S. MENTZEL, M. WILBRINK, and A. AKHMANOVA et al., 1994  Interspecific sequence comparison of the muscle-myosin heavy-chain genes from Drosophila hydei and Drosophila melanogaster. J. Mol. Evol. 39:357-368[Medline].

MULLEN, M. P., C. W. SMITH, J. G. PATTON, and B. NADAL-GINARD, 1991  Alpha-tropomyosin mutually exclusive exon selection: competition between branchpoint/polypyrimidine tracts determines default exon choice. Genes Dev. 5:642-655[Abstract/Free Full Text].

NADAL-GINARD, B., C. W. SMITH, J. G. PATTON, and R. E. BREITBART, 1991  Alternative splicing is an efficient mechanism for the generation of protein diversity: contractile protein genes as a model system. Adv. Enzyme Regul. 31:261-286[Medline].

NAGEL, R. J., A. M. LANCASTER, and A. M. ZAHLER, 1998  Specific binding of an exonic splicing enhancer by the pre-mRNA splicing factor SRp55. RNA 4:11-23[Abstract].

PATTON, J. G., S. A. MAYER, P. TEMPST, and B. NADAL-GINARD, 1991  Characterization and molecular cloning of polypyrimidine tract-binding protein: a component of a complex necessary for pre-mRNA splicing. Genes Dev. 5:1237-1251[Abstract/Free Full Text].

PEREZ, I., C. H. LIN, J. G. MCAFEE, and J. G. PATTON, 1997  Mutation of PTB binding sites causes misregulation of alternative 3' splice site selection in vivo. RNA 3:764-778[Abstract].

RAMCHATESINGH, J., A. M. ZAHLER, K. M. NEUGEBAUER, M. B. ROTH, and T. A. COOPER, 1995  A subset of SR proteins activates splicing of the cardiac troponin T alternative exon by direct interactions with an exonic enhancer. Mol. Cell. Biol. 15:4898-4907[Abstract].

ROBBERSON, B. L., G. J. COTE, and S. M. BERGET, 1990  Exon definition may facilitate splice site selection in RNAs with multiple exons. Mol. Cell. Biol. 10:84-94[Abstract/Free Full Text].

RYAN, K. J. and T. A. COOPER, 1996  Muscle-specific splicing enhancers regulate inclusion of the cardiac troponin T alternative exon in embryonic skeletal muscle. Mol. Cell. Biol. 16:4014-4023[Abstract].

SCHAAL, T. D. and T. MANIATIS, 1999  Selection and characterization of pre-mRNA splicing enhancers: identification of novel SR protein-specific enhancer sequences. Mol. Cell. Biol. 19:1705-1719[Abstract/Free Full Text].

SOUTHBY, J., C. GOODING, and C. W. SMITH, 1999  Polypyrimidine tract binding protein functions as a repressor to regulate alternative splicing of alpha-actinin mutually exclusive exons. Mol. Cell. Biol. 19:2699-2711[Abstract/Free Full Text].

STANDIFORD, D., M. DAVIS, W. SUN, and J. C. EMERSON, 1997  Splice-junction elements and intronic sequences regulate alternative splicing of the Drosophila myosin heavy chain gene transcript. Genetics 147:725-741[Abstract].

STARK, J. M., T. A. COOPER, and M. B. ROTH, 1999  The relative strengths of SR protein-mediated associations of alternative and constitutive exons can influence alternative splicing. J. Biol. Chem. 274:29838-29842[Abstract/Free Full Text].

XIAO, S. H. and J. L. MANLEY, 1998  Phosphorylation-dephosphorylation differentially affects activities of splicing factor ASF/SF2. EMBO J. 17:6359-6367[Medline].

XU, R., J. TENG, and T. A. COOPER, 1993  The cardiac troponin T alternative exon contains a novel purine-rich positive splicing element. Mol. Cell. Biol. 13:3660-3674[Abstract/Free Full Text].

ZHENG, Z. M., M. HUYNEN, and C. C. BAKER, 1998  A pyrimidine-rich exonic splicing suppressor binds multiple RNA splicing factors and inhibits spliceosome assembly. Proc. Natl. Acad. Sci. USA 95:14088-14093[Abstract/Free Full Text].

This article has been cited by other articles:

				J. M. Kreahling and B. R. Graveley The iStem, a Long-Range RNA Secondary Structure Element Required for Efficient Exon Inclusion in the Drosophila Dscam Pre-mRNA Mol. Cell. Biol., December 1, 2005; 25(23): 10251 - 10260. [Abstract] [Full Text] [PDF]

				Z.-P. HUANG, H. ZHOU, H.-L. HE, C.-L. CHEN, D. LIANG, and L.-H. QU Genome-wide analyses of two families of snoRNA genes from Drosophila melanogaster, demonstrating the extensive utilization of introns for coding of snoRNAs RNA, August 1, 2005; 11(8): 1303 - 1316. [Abstract] [Full Text] [PDF]

				S. L. Hooper and J. B. Thuma Invertebrate Muscles: Muscle Specific Genes and Proteins Physiol Rev, July 1, 2005; 85(3): 1001 - 1060. [Abstract] [Full Text] [PDF]

				A. Borchert, N. E. Savaskan, and H. Kuhn Regulation of Expression of the Phospholipid Hydroperoxide/Sperm Nucleus Glutathione Peroxidase Gene. TISSUE-SPECIFIC EXPRESSION PATTERN AND IDENTIFICATION OF FUNCTIONAL CIS- AND TRANS-REGULATORY ELEMENTS J. Biol. Chem., January 17, 2003; 278(4): 2571 - 2580. [Abstract] [Full Text] [PDF]

				A. M. Celotto and B. R. Graveley Alternative Splicing of the Drosophila Dscam Pre-mRNA Is Both Temporally and Spatially Regulated Genetics, October 1, 2001; 159(2): 599 - 608. [Abstract] [Full Text] [PDF]