Enhancer-dependent interaction between 5' and 3' splice sites in trans.

Splice-site selection and alternative splicing of nuclear pre-mRNAs can be controlled by splicing enhancers that act by promoting the activity of upstream splice sites. Here we show that RNA molecules containing a 3' splice site and enhancer sequence are efficiently spliced in trans to RNA molecules containing normally cis-spliced 5' splice sites or to normally trans-spliced spliced leader RNAs from lower eukaryotes. In addition, we show that this reaction is stimulated by (Ser + Arg)-rich splicing factors that are known to promote protein-protein interactions in the cis-splicing reaction. Thus, splicing enhancers facilitate the assembly of protein complexes on RNAs containing a 3' splice site, and this complex is sufficiently stable to functionally interact with 5' splice sites located on separate RNAs. This trans-splicing is mediated by interactions between (Ser + Arg)-rich splicing factors bound to the enhancer and general splicing factors bound to the 5' and 3' splice sites. These same interactions are likely to play a crucial role in alternative splicing and splice-site selection in cis.

Structural variety of arginine-rich RNA-binding peptides.

Arginine-rich domains are used by a variety of RNA-binding proteins to recognize specific RNA hairpins. It has been shown previously that a 17-aa arginine-rich peptide from the human immunodeficiency virus Rev protein binds specifically to its RNA site when the peptide is in an alpha-helical conformation. Here we show that related peptides from splicing factors, viral coat proteins, and bacteriophage antiterminators (the N proteins) also have propensities to form alpha-helices and that the N peptides require helical conformations to bind to their cognate RNAs. In contrast, introducing proline mutations into the arginine-rich domain of the human immunodeficiency virus Tat protein abolishes its potential to form an alpha-helix but does not affect RNA-binding affinity in vitro or in vivo. Based on results from several peptide-RNA model systems, we suggest that helical peptides may be used to recognize RNA structures having particularly wide major grooves, such as those found near loops or large bulges, and that nonhelical or extended peptides may be used to recognize less accessible grooves.

Triplet repeat expansion in myotonic dystrophy alters the adjacent chromatin structure.

Myotonic dystrophy is caused by an expansion of a CTG triplet repeat sequence in the 3' noncoding region of a protein kinase gene, yet the mechanism by which the triplet repeat expansion causes disease remains unknown. This report demonstrates that a DNase I hypersensitive site is positioned 3' of the triplet repeat in the wild-type allele in both fibroblasts and skeletal muscle cells. In three unrelated individuals with myotonic dystrophy that have large expansions of the triplet repeat, the allele with the triplet repeat expansion exhibited both overall DNase I resistance and inaccessibility of nucleases to the adjacent hypersensitive site. These results indicate that the triplet repeat expansion alters the adjacent chromatin structure, establishing a region of condensed chromatin, and suggests a molecular mechanism for myotonic dystrophy.

Accelerated evolution in the protein-coding regions is universal in crotalinae snake venom gland phospholipase A2 isozyme genes.

The nucleotide sequences of four genes encoding Trimeresurus gramineus (green habu snake, crotalinae) venom gland phospholipase A2 (PLA2; phosphatidylcholine 2-acylhydrolase, EC 3.1.1.4) isozymes were compared internally and externally with those of six genes encoding Trimeresurus flavoviridis (habu snake, crotalinae) venom gland PLA2 isozymes. The numbers of nucleotide substitutions per site (KN) for the noncoding regions including introns were one-third to one-eighth of the numbers of nucleotide substitutions per synonymous site (KS) for the protein-coding regions of exons, indicating that the noncoding regions are much more conserved than the protein-coding regions. The KN values for the introns were found to be nearly equivalent to those of introns of T. gramineus and T. flavoviridis TATA box-binding protein genes, which are assumed to be a general (nonvenomous) gene. Thus, it is evident that the introns of venom gland PLA2 isozyme genes have evolved at a similar rate to those of nonvenomous genes. The numbers of nucleotide substitutions per nonsynonymous site (KA) were close to or larger than the KS values for the protein-coding regions in venom gland PLA2 isozyme genes. All of the data combined reveal that Darwinian-type accelerated evolution has universally occurred only in the protein-coding regions of crotalinae snake venom PLA2 isozyme genes.

Mouse mammary tumor viruses with functional superantigen genes are selected during in vivo infection.

Mouse mammary tumor virus (MMTV) encodes a superantigen that is important for viral infectivity in vivo. To determine whether superantigen function was required for infection by milk-borne MMTV, we created HYB PRO/Cla transgenic mice. These mice produced a full-length, packaged viral RNA with a frameshift mutation that caused premature termination of the superantigen protein. Young HYB PRO/Cla mice showed no deletion of their cognate V beta 14+ T cells, although they shed virus in their milk. The nontransgenic offspring of the HYB PRO/Cla mice were infected with this virus, since transgene-specific viral transcripts were detected in their mammary glands. Surprisingly, these offspring demonstrated the progressive deletion of V beta 14+ T cells characteristic of exogenous MMTV (C3H) infection. Sequence analysis demonstrated that these newly acquired viruses had reconstituted superantigen open reading frames resulting from recombination between the HYB PRO/Cla and endogenous Mtv-1 proviral RNAs. Thus, there is selection during the infection process for MMTVs with functional superantigen genes.

NusA interferes with interactions between the nascent RNA and the C-terminal domain of the alpha subunit of RNA polymerase in Escherichia coli transcription complexes.

The effects of NusA on the RNA polymerase contacts made by nucleotides at internal positions in the nascent RNA in Escherichia coli transcription complexes were analyzed by using the photocrosslinking nucleotide analog 5-[(4-azidophenacyl) thio]-UMP. It was placed at nucleotides between +6 and +15 in RNA transcribed from the phage lambda PR' promoter. Crosslinks of analog in these positions in RNAs which contained either 15, 28, 29, or 49 nt were examined. Contacts between the nascent RNA and proteins in the transcription complex were analyzed as the RNA was elongated, by placing the crosslinker nearest the 5' end of the RNA 10, 23, 24, or 44 nt away from the 3' end. The beta or beta' subunit of polymerase, and NusA when added, were contacted by RNA from 15 to 49 nt long. When the upstream crosslinker was 24 nt from the 3" end of the RNA (29-nt RNA), alpha was also contacted in the absence of NusA. The addition of NusA prevented RNA crosslinking to alpha. When the crosslinker was 44 nt from the 3' end (49-nt RNA), alpha crosslinks were still observed, but crosslinks to beta or beta' and NusA were greatly diminished. RNA crosslinking to alpha, and loss of this crosslink when NusA was added, was observed in the presence of NusB, NusE, and NusG and when transcription was carried out in the presence of an E. coli S100 cell extract. Peptide mapping localized the RNA interactions to the C-terminal domain of alpha.

Catalytically critical nucleotide in domain 5 of a group II intron.

Domain 5 (D5) is a small hairpin structure within group II introns. A bimolecular assay system depends on binding by D5 to an intron substrate for self-splicing activity. In this study, mutations in D5 identify two among six nearly invariant nucleotides as being critical for 5' splice junction hydrolysis but unimportant for binding. A mutation at another site in D5 blocks binding. Thus, mutations can distinguish two D5 functions: substrate binding and catalysis. The secondary structure of D5 may resemble helix I formed by the U2 and U6 small nuclear RNAs in the eukaryotic spliceosome. Our results support a revision of the previously proposed correspondence between D5 and helix I on the basis of the critical trinucleotide 5'-AGC-3' present in both. We suggest that this trinucleotide plays a similar role in promoting the chemical reactions for both splicing systems.

A conserved family of elav-like genes in vertebrates.

A large family of genes encodes proteins with RNA recognition motifs that are presumed to bind RNA and to function in posttranscriptional regulation. Neural-specific members of this family include elav, a gene required for correct differentiation and maintenance of neurons in Drosophila melanogaster, and a related gene, HuD, which is expressed in human neuronal cells. I have identified genes related to elav and HuD in Xenopus laevis, zebrafish, and mouse that define a family of four closely related vertebrate elav-like genes (elrA, elrB, elrC, and elrD) in fish, frogs, and mammals. In addition to protein sequence conservation, a segment of the 3'-untranslated sequence of elrD is also conserved, implying a functional role in elrD expression. In adult frogs, elrC and elrD are exclusively expressed in the brain, whereas elrB is expressed in brain, testis, and ovary. During Xenopus development, elrC and elrD RNAs are detected by late gastrula and late neurula stages, respectively, whereas a nervous system-specific elrB RNA species is expressed by early tadpole stage. Additional elrB transcripts are detected in the ovary and early embryo, demonstrating a maternal supply of mRNA and possibly of protein. These expression patterns suggest a role for different elav-like genes in early development and neuronal differentiation. Surprisingly, elrA is expressed in all adult tissues tested and at all times during development. Thus, the widely expressed elrA is expected to have a related function in all cells.

Scrambling of the actin I gene in two Oxytricha species.

The DNA in a germ-line nucleus (a micronucleus) undergoes extensive processing when it develops into a somatic nucleus (a macronucleus) after cell mating in hypotrichous ciliates. Processing includes destruction of a large amount of spacer DNA between genes and excision of gene-sized molecules from chromosomes. Before processing, micronuclear genes are interrupted by numerous noncoding segments called internal eliminated sequences (IESs). The IESs are excised and destroyed, and the retained macro-nuclear-destined sequences (MDSs) are spliced. MDSs in some micronuclear genes are not in proper order and must be reordered during processing to create functional gene-sized molecules for the macronucleus. Here we report that the micronuclear actin I gene in Oxytricha trifallax WR consists of 10 MDSs and 9 IESs compared to the previously reported 9 MDSs and 8 IESs in the micronuclear actin I gene of Oxytricha nova. The MDSs in the actin I gene are scrambled in a similar pattern in the two species, but the positions of MDS-IES junctions are shifted by up to 14 bp for scrambled and 138 bp for the nonscrambled MDSs. The shifts in MDS-IES junctions create differences in the repeat sequences that are believed to guide MDS splicing. Also, the sizes and sequences of IESs in the micronuclear actin I genes are different in the two Oxytricha species. These observations give insight about the possible origins of IES insertion and MDS scrambling in evolution and show the extraordinary malleability of the germ-line DNA in hypotrichs.

Modulation of 5' splice site choice in pre-messenger RNA by two distinct steps.

Ser/Arg-rich proteins (SR proteins) are essential splicing factors that commit pre-messenger RNAs to splicing and also modulate 5' splice site choice in the presence or absence of functional U1 small nuclear ribonucleoproteins (snRNPs). Here, we perturbed the U1 snRNP in HeLa cell nuclear extract by detaching the U1-specific A protein using a 2'-O-methyl oligonucleotide (L2) complementary to its binding site in U1 RNA. In this extract, the standard adenovirus substrate is spliced normally, but excess amounts of SR proteins do not exclusively switch splicing from the normal 5' splice site to a proximal site (site 125 within the adenovirus intron), suggesting that modulation of 5' splice site choice exerted by SR proteins requires integrity of the U1 snRNP. The observation that splicing does not necessarily follow U1 binding indicates that interactions between the U1 snRNP and components assembled on the 3' splice site via SR proteins may also be critical for 5' splice site selection. Accordingly, we found that SR proteins promote the binding of the U2 snRNP to the branch site and stabilize the complex formed on a 3'-half substrate in the presence or absence of functional U1 snRNPs. A novel U2/U6/3'-half substrate crosslink was also detected and promoted by SR proteins. Our results suggest that SR proteins in collaboration with the U1 snRNP function in two distinct steps to modulate 5' splice site selection.

Distinct functions of SR proteins in recruitment of U1 small nuclear ribonucleoprotein to alternative 5' splice sites.

Alternative splicing of precursor messenger RNAs (pre-mRNAs) is an important mechanism for the regulation of gene expression. The members of the SR protein family of pre-mRNA splicing factors have distinct functions in promoting alternative splice site usage. Here we show that SR proteins are required for the first step of spliceosome assembly, interaction of the U1 small nuclear ribonucleoprotein complex (U1 snRNP) with the 5' splice site of the pre-mRNA. Further, we find that individual SR proteins have distinct abilities to promote interaction of U1 snRNP with alternative 5' splice junctions. These results suggest that SR proteins direct 5' splice site selection by regulation of U1 snRNP assembly onto the pre-mRNA.