We demonstrate that four different proteins from calf thymus are able to restore splicing in the same splicing-deficient extract using several different pre-mRNA substrates. These proteins are members of a conserved family of proteins recognized by a monoclonal antibody that binds to active sites of RNA polymerase II transcription. We purified this family of nuclear phosphoproteins to apparent homogeneity by two salt precipitations. The family, called SR proteins for their serine-and arginine-rich carboxy-terminal domains, consists of at least five different proteins with molecular masses of 20, 30, 40, 55, and 75 kD. Microsequencing revealed that they are related but not identical. In four of the family members a repeated protein sequence that encompasses an RNA recognition motif was observed. We discuss the potential role of this highly conserved, functionally related set of proteins in pre-mRNA splicing.

Mammals normally maintain their core body temperature (CBT) despite changes in environmental temperature. Exceptions to this norm include suspended animation-like states such as hibernation, torpor, and estivation. These states are all characterized by marked decreases in metabolic rate, followed by a loss of homeothermic control in which the animal's CBT approaches that of the environment. We report that hydrogen sulfide can induce a suspended animation-like state in a nonhibernating species, the house mouse (Mus musculus). This state is readily reversible and does not appear to harm the animal. This suggests the possibility of inducing suspended animation-like states for medical applications.

Alternative splicing of precursor messenger RNAs (pre-mRNAs) is a common mechanism of regulating gene expression. SR proteins are a family of pre-mRNA splicing factors that are structurally related and evolutionarily conserved. Any member of the SR family can complement a splicing-deficient extract that lacks the entire family of SR proteins. Here it is demonstrated that particular SR proteins have distinct functions in alternative pre-mRNA splicing in vitro. In addition, SR proteins are differentially expressed in a variety of tissues. These results suggest a fundamental role for SR proteins in the regulation of alternative splicing.

Abstract. An antibody was identified previously that recognizes sites of polymerase II transcription on lampbrush chromosomes, puffs on polytene chromosomes, and many small granules in the nucleoplasm of all cells tested . This antibody binds a conserved family of phosphorylated polypeptides in vertebrate and invertebrate cells . We developed a method for purifying these proteins that involves differential solubility in MgC12. We N ASCENT transcripts are bound by proteins, concomitant with synthesis by polymerase II . This was demonstrated in observations of ribonuclease treated amphibian lampbrush chromosomes (Gall and Callan, 1962) where it is possible to see nascent protein-bound transcripts by light microscopy, and was later confirmed by a number of studies in other systems (Malcolm and Sommerville, 1974;Lamb and Daneholt, 1979). Electron microscopic studies of transcription units using spreading techniques developed by Miller and colleagues confirmed this observation (Miller and Bakken, 1972).Because of their role in the elaboration ofgenetic information, there has been a major effort to identify transcriptbinding proteins. To date, two sets have been identified : hnRNP and snRNP proteins. The hnRNP proteins, which number at least 24, have been defined by biochemical cofractionation with radiolabeled hnRNA (Samarina et al., 1966;Beyer et al ., 1977) and more recently by immunochemical analysis (Pifiol-Roma et al., 1988) . Several genes encoding hnRNP proteins have been cloned and at least one protein has been shown to interact directly with RNA in vitro (Merrill et al., 1988) . Twelve snRNP proteins were identified by co-immunoprecipitation with snRNAs using an antibody that recognized a common epitope found on these proteins (Lerner and Steitz, 1979).We show here that a mAb 104 (mAb104), described as binding to lateral loops on amphibian lampbrush chromosomes (Roth et al., 1990) and to puffs on Drosophila polytene chromosomes (unpublished observation) binds a previously undescribed conserved family of nuclear phosphoproteins . Antigens recognized by mAbl04 do not coprecipitate with anti-hnRNP antibodies (Roth et al., 1990). We have cloned a gene that encodes a mAb104 immunoreactive protein from Drosophila melanogaster The encoded protein shares sequence homology with proteins involved in pre-mRNA splic-© The Rockefeller University Press, 0021-9525/91/11/587/10 $2 .00 The Journal of Cell Biology, Volume 115, Number 3, November 1991587-596 isolated a Drosophila cDNA encoding one of the proteins using information obtained from microsequencing . In vivo expression studies show that this protein is concentrated on sites of polymerase II transcription and that it is highly phosphorylated . The protein shares a high degree of homology with proteins involved in alternative splicing of pre-mRNA suggesting the possibility that this protein plays a role in pre-mRNA splicing .ing. We show that heterogeneously-sized RNA-but not snRNÁs-coprecipitate using mAb104 . Based on these experiments as well as in vivo expressio...

Some metazoans have evolved the capacity to survive severe oxygen deprivation. The nematode, Caenorhabditis elegans, exposed to anoxia (0 kPa, 0% O 2 ) enters into a recoverable state of suspended animation during all stages of the life cycle. That is, all microscopically observable movement ceases including cell division, developmental progression, feeding, and motility. To understand suspended animation, we compared oxygen-deprived embryos to nontreated embryos in both wild-type and hif-1 mutants. We found that hif-1 mutants survive anoxia, suggesting that the mechanisms for anoxia survival are different from those required for hypoxia. Examination of wild-type embryos exposed to anoxia show that blastomeres arrest in interphase, prophase, metaphase, and telophase but not anaphase. Analysis of the energetic state of anoxic embryos indicated a reversible depression in the ATP to ADP ratio. Given that a decrease in ATP concentrations likely affects a variety of cellular processes, including signal transduction, we compared the phosphorylation state of several proteins in anoxic embryos and normoxic embryos. We found that the phosphorylation state of histone H3 and cell cycle-regulated proteins recognized by the MPM-2 antibody were not detectable in anoxic embryos. Thus, dephosphorylation of specific proteins correlate with the establishment and/or maintenance of a state of anoxia-induced suspended animation.

Hydrogen sulfide (H2S) is naturally produced in animal cells. Exogenous H 2S has been shown to effect physiological changes that improve the capacity of mammals to survive in otherwise lethal conditions. However, the mechanisms required for such alterations are unknown. We investigated the physiological response of Caenorhabditis elegans to H 2S to elucidate the molecular mechanisms of H 2S action. Here we show that nematodes exposed to H2S are apparently healthy and do not exhibit phenotypes consistent with metabolic inhibition. Instead, animals exposed to H 2S are thermotolerant and long-lived. These phenotypes require SIR-2.1 activity but are genetically independent of the insulin signaling pathway, mitochondrial dysfunction, and caloric restriction. These studies suggest that SIR-2.1 activity may translate environmental change into physiological alterations that improve survival. It is interesting to consider the possibility that the mechanisms by which H 2S increases thermotolerance and lifespan in nematodes are conserved and that studies using C. elegans may help explain the beneficial effects observed in mammals exposed to H 2S.

Abstract. An mAb library was produced against proteins from the germinal vesicle (GV) of the frog Xenopus/aev/s; mAb 104 was selected from this library on the basis of its immunofluorescent staining of lampbrush chromosome loops. Chromosomes from several species of frogs and salamanders stained equally well. The antibody also stained the surface of numerous small granules in the GV nucleoplasm. The interior of the same granules was stained by antibodies against small nuclear ribonucleoproteins (snRNPs). mAb 104 also stained somatic nuclei from many vertebrate and invertebrate species, usually in a finely punctate pattern similar to that described for anti-snRNP and other antinuclear antibodies. The staining of somatic nuclei was much stronger during the mitotic stages than during interphase. Immunoblot analysis showed that mAb 104 recognizes a phosphorylated epitope.AMPBRUSH chromosomes from amphibian oocytes provide a useful system for studying proteins associated with nascent RNA transcripts. Each chromosome has a central axis consisting of transcriptionally inactive chromatin (chromomeres) from which loops of active chromatin extend laterally. The bulk of a loop consists of nascent transcripts with associated ribonucleoprotein (RNP)~ still attached to the DNA template. The RNP matrix of a loop is so abundant that individual transcription units are readily visible by light optical microscopy (Scheer et al., 1976;Gall et al., 1983). Some loops consist of a single transcription unit, whereas many contain two or more units in various orientations.Antibodies have been used to identify proteins associated with the nascent transcripts on lampbrush loops (Scott and Sommerville, 1974; Sommerville et al., 1978;Martin and Okamura, 1981;Lacroix et al., 1985;Roth and Gall, 1987;Gall and Callan, 1989;Pifiol-Roma et al., 1989). The majority of loops are morphologically similar and contain a set of common proteins, including both heterogeneous nuclear RNPs (hnRNPs) and small nuclear RNPs (snRNPs). A few "landmark" loops are morphologically distinct (reviewed in Callan, 1986), and in several cases their protein composition is known to be unusual.We have produced a number of mAbs from mice injected with germinal vesicle (GV) proteins from the frog Xenopus laevis, and the newt Notophthalmus viridescens. Several of these bind strongly to lampbrush chromosome loops (Roth 1. Abbreviations used in this paper: GV, germinal vesicle; hn, heterogeneous nuclear; RNP, ribonucleoprotein; sn, small nuclear.and or to other intranuclear structures. Most are relatively species-specific; for instance, most mAbs raised against Xenopus react with Xenopua and other anurans (e.g., Rana) but not with urodeles or other vertebrates. One mAb, designated 104, has proved of unusual interest because it cross-reacts with a wide variety of vertebrate and invertebrate species. In immunofluorescence assays mAb 104 stains most larnpbrush loops and numerous small granules in the GV nucleoplasm. In somatic cells it stains similar but smaller nuclear granules i...

The cardiac troponin T pre-mRNA contains an exonic splicing enhancer that is required for inclusion of the alternative exon 5. Here we show that enhancer activity is exquisitely sensitive to changes in the sequence of a 9-nucleotide motif (GAGGAAGAA) even when its purine content is preserved. A series of mutations that increased or decreased the level of exon inclusion in vivo were used to correlate enhancer strength with RNA-protein interactions in vitro. Analyses involving UV cross-linking and immunoprecipitation indicated that only four (SRp30a, SRp40, SRp55, and SRp75) of six essential splicing factors known as SR proteins bind to the active enhancer RNA. Moreover, purified SRp40 and SRp55 activate splicing of exon 5 when added to a splicing-deficient S100 extract. Purified SRp30b did not stimulate splicing in S100 extracts, which is consistent with its failure to bind the enhancer RNA. In vitro competition of SR protein splicing activity and UV cross-linking demonstrated that the sequence determinants for SR protein binding were precisely coincident with the sequence determinants of enhancer strength. Thus, a subset of SR proteins interacts directly with the exonic enhancer to promote inclusion of a poorly defined alternative exon. Independent regulation of the levels of SR proteins may, therefore, contribute to the developmental regulation of exon inclusion.Pre-mRNA splicing involves two transesterification reactions mediated by the spliceosome (11, 21). The mechanisms by which exon-intron borders are first defined within complex pre-mRNAs remain largely unknown. Although splice donor and acceptor sequences are fairly well conserved among premRNAs, these elements are not sufficient to distinguish bona fide splice sites from unused cryptic splice sites present in many metazoan pre-mRNAs. Because splicing is so precise, additional controls for splice site selection must be present elsewhere in the pre-mRNA.Results with vertebrate and invertebrate experimental systems have demonstrated that sequences within exons, exclusive of splice junctions, can play a role in splice site selection (13). One positively acting splicing element is a 13-nucleotide repeat in the female-specific exon of the Drosophila doublesex (dsx) gene. Two proteins, transformer-2 and transformer, promote utilization of an upstream 3Ј splice site via direct interactions with the repeats (12,15,22). Purine-rich splicing elements, called splicing enhancers, have recently been identified in several vertebrate exons (1,6,16,19,28,31,33,35,36) and in the dsx female-specific exon (20). Splicing enhancers are required for efficient splicing of the resident exon, and several have been shown to enhance splicing of heterologous exons (16,19,33,35,36). With one exception (16), splicing enhancers appear to activate splicing of the upstream intron (35). The enhancer in the fibronectin ED1 exon has been shown to promote binding of the U2 small nuclear ribonucleoprotein particle (snRNP) to the upstream branch site (19), which is consistent with this observ...