The structure of mammalian small nuclear ribonucleoproteins. Identification of multiple protein components reactive with anti-(U1)ribonucleoprotein and anti-Sm autoantibodies

The Sm small nuclear ribonucleoproteins (snRNPs) from mammalian cells have been characterized as containing U1, U2, U4, U5, and U6 RNA associated with some subset of at least 10 distinct polypeptides (called 68K, A, A', B, B', C, D, E, F, and G) that range in molecular weight from 68,000 to 11,000. Whereas this entire collection of snRNP particles is precipitated by patient anti-Sm autoantibodies, anti-(U1)RNP autoantibodies specifically recognize U1 snRNPs. Here, we have performed immunoblots using the sera from 29 patients and a mouse anti-Sm monoclonal antibody to identify which HeLa cell snRNP proteins carry anti-Sm or anti-(U1)RNP antigenic determinants. Strikingly, every serum surveyed, as well as the monoclonal antibody, recognizes determinants on two or more snRNP protein components. The three proteins, 68K, A, and C, that uniquely fractionate with U1 snRNPs are specifically reactive with anti-(U1)RNP sera in blots. Anti-Sm patient sera and the mouse monoclonal antibody react with proteins B, B', D, and sometimes E, one or more of which must be present on all Sm snRNPs. The blot results combined with data obtained from a refined 32P-labeled RNA immunoprecipitation assay reveal that, in our collection of the sera from 29 patients, anti-Sm rarely exists in the absence of equal or higher titers of anti-(U1)RNP; moreover, (U1)RNP sera often contain detectable levels of anti-Sm. Our findings further define the protein composition of the Sm snRNPs and raise intriguing questions concerning the relatedness of snRNP polypeptides and the mechanism of autoantibody induction.     

The association of the U1-specific 70K and C proteins with U1 snRNPs is mediated in part by common U snRNP proteins.

The U1 small nuclear ribonucleoprotein particle (snRNP)-specific 70K and A proteins are known to bind directly to stem-loops of the U1 snRNA, whereas the U1-C protein does not bind to naked U1 snRNA, but depends on other U1 snRNP protein components for its association. Focusing on the U1-70K and U1-C proteins, protein-protein interactions contributing to the association of these particle-specific proteins with the U1 snRNP were studied. Immunoprecipitation of complexes formed after incubation of naked U1 snRNA or purified U1 snRNPs lacking their specific proteins (core U1 snRNP) with in vitro translated U1-C protein, revealed that both common snRNP proteins and the U1-70K protein are required for the association of U1-C with the U1 snRNP. Binding studies with various in vitro translated U1-70K mutants demonstrated that the U1-70K N-terminal domain is necessary and sufficient for the interaction of U1-C with core U1 snRNPs. Surprisingly, several N-terminal fragments of the U1-70K protein, which lacked the U1-70K RNP-80 motif and did not bind naked U1 RNA, associated stably with core U1 snRNPs. This suggests that a new U1-70K binding site is generated upon association of common U1 snRNP proteins with U1 RNA. The interaction between the N-terminal domain of U1-70K and the core RNP domain was specific for the U1 snRNP; stable binding was not observed with core U2 or U5 snRNPs, suggesting essential structural differences among snRNP core domains. Evidence for direct protein-protein interactions between U1-specific proteins and common snRNP proteins was supported by chemical crosslinking experiments using purified U1 snRNPs. Individual crosslinks between the U1-70K and the common D2 or B'/B protein, as well as between U1-C and B'/B, were detected. A model for the assembly of U1 snRNP is presented in which the complex of common proteins on the RNA backbone functions as a platform for the association of the U1-specific proteins.     

Monospecific antibodies reveal details of U2 snRNP structure and interaction between U1 and U2 snRNPs.

Monospecific antibodies directed against several U small nuclear ribonucleoprotein (U snRNP) particle proteins were affinity purified from a patient's anti-(U1,U2)RNP serum. These were used to demonstrate that: (i) proteins equivalent to the mammalian U2 snRNP-specific A' and B" proteins are present in Xenopus laevis oocytes; (ii) both proteins A' and B" have the same structural requirements for binding to U2 snRNA; (iii) proteins B, B' and D have the same structural requirement for binding to U2 snRNA; (iv) using very high specific activity RNA probes it is possible to detect a fraction of either U1 or U2 snRNA precipitable by antibodies directed against proteins specific for the other U snRNP, indicating an interaction between U1 and U2 snRNPs. The structural requirements of this interaction were studied for the U2 snRNP. All changes made to U2 snRNA or snRNP structure resulted in loss of the interaction with U1 snRNP.     

Autoantibodies to the U2 small nuclear ribonucleoprotein in a patient with scleroderma-polymyositis overlap syndrome

Autoantibodies directed against the U2 small nuclear ribonucleoprotein (snRNP) have been found in the serum of a patient with scleroderma-polymyositis overlap syndrome. This specificity, called anti-(U2)-RNP, is distinct from all previously described autoantibodies, including those that precipitate related snRNPs: anti-Sm antibodies, which react with the entire set of U1, U2, U4, U5, and U6 snRNPs, and anti-(U1)RNP antibodies, which recognize only U1 snRNPs. From HeLa cell extracts, anti-(U2)RNP immunoprecipitates predominantly one 32P-labeled RNA species, identified as U2 small nuclear RNA, and six [35S]methionine-labeled protein bands, A' (Mr = 32,000), B (Mr = 28,000), D (Mr = 16,000), E (Mr = 13,000), F (Mr = 12,000), and G (Mr = 11,000). Protein blot analysis reveals that the A' protein carries (U2)RNP antigenic determinant(s) and therefore represents a polypeptide unique to the U2 snRNP; the B protein associated with U2 snRNPs may also be unique. Like U1 and the other Sm snRNPs, U2 snRNPs occupy a nuclear, non-nucleolar location and are antigenically conserved from insects to man. An antibody specific for the U2 snRNP will be useful in deciphering the function of this particle.     

Fractionation and characterization of human small nuclear ribonucleoproteins containing U1 and U2 RNAs

Human small nuclear ribonucleoproteins (snRNPs) containing U1 and U2 snRNAs have been isolated from cultured cells by nonimmunological methods. The U1 snRNP population remained immunoprecipitable by systemic lupus erythematosis anti-RNP and anti-Sm antibodies throughout fractionation and contained polypeptides of molecular weights corresponding to those defined as U1 snRNP polypeptides by immunoprecipitation of crude extracts. The purified assemblies contained U1 RNA and nine snRNP polypeptides of molecular weights 67,000 (P67), 30,000 (P30), 23,000 (P23), 21,500 (P22), 17,500 (P18), 12,300 (P12), 10,200 (P10), 9,100 (P9), and 8,500 (P8). P67, P30, and P18 were unique to U1 snRNPs. The U2 snRNP population remained immunoprecipitable by the systemic lupus erythematosis anti-Sm antibody throughout fractionation. The purified U2 assemblies contained six polypeptides of molecular weights corresponding to those defined by immunoprecipitation to be common to U1 and U2 snRNPs including P23, P22, P12, P10, P9, and P8. In addition, U2 snRNPs contained a unique polypeptide of 27,000 Da.     

In vitro assembly of U1 snRNPs.

An efficient system for the in vitro assembly of U1 snRNPs is described. RNA-protein interactions in a series of U1 snRNA mutants assembled both in vivo and in vitro were studied in order to verify the accuracy of the system. Two discrete protein binding sites are defined by immunoprecipitation with antibodies against different protein components of the U1 snRNP and a newly developed protein sequestering assay. The U1 snRNP-specific proteins 70K and A require only the 5'-most stem-loop structure of U1 snRNA for binding, the common U snRNP proteins require the conserved Sm binding site (AUnG). Interactions between these two groups of proteins are detected. These results are combined to derive a model of the U1 snRNP structure. The potential use of the in vitro system in the functional analysis of U1 snRNP proteins is discussed.     

An abundant U6 snRNP found in germ cells and embryos of Xenopus laevis.

The particle state of U snRNPs was analyzed in oocytes, eggs, embryos and testes from Xenopus laevis. In each case both the relative abundance and the composition of some U snRNPs were found to differ from that of somatic cells. U2 and U6 snRNPs were the most prominent U snRNPs in germ cells and early embryos. In particular, the concentration of U6 snRNA was 10-20 times higher than that of U4 snRNA. Most of the U6 snRNA was not associated with U4 snRNA and migrated on sucrose gradients as a U6 snRNP. The structure of this novel U snRNP was analyzed. A single protein of 50 kd was copurified with U6 snRNPs by a combination of gradient fractionation, immunodepletion with anti-Sm antibodies and immunoprecipitation with anti-6-methyl adenosine antibodies. Although the U6 snRNP did not contain Sm proteins it migrated into the nucleus when U6 snRNA was injected into the cytoplasm of oocytes. Two U6 snRNA elements have been identified. The first is essential for nuclear migration in oocytes, but not for the formation of U4/6 snRNPs in vitro and might be the binding site of a U6-specific protein. The second element was required for interaction with U4 snRNPs but not for nuclear targeting.     

Human U1 and U2 small nuclear ribonucleoproteins contain common and unique polypeptides.

Immunoprecipitation of human small nuclear ribonucleoproteins (snRNPs) containing the small nuclear RNAs U1, U2, U4, U5, and U6 with two antibodies produced in certain patients suffering from systemic lupus erythematosus was used to identify the polypeptides present on human U1 and U2 snRNPs. U1 and U2 snRNPs contain both common and unique polypeptides; visualization of the differences was possible through the use of non-methionine protein labeling and partial fractionation of snRNP populations. To facilitate comparisons with results from other laboratories, we have designated the snRNP polypeptides by their molecular weights. Four small polypeptides, P8, P9, P10, and P12, of 8,000 to 12,000 daltons, are each present in equal amounts on both U1 and U2 snRNPs. U1 snRNPs also contain a unique 30,000-dalton polypeptide, P30, whereas U2 snRNPs contain a unique 27,000-dalton, methionine-deficient polypeptide, P27. A closely migrating pair of polypeptides, P23 and P22, of 23,000 and 21,500 daltons, respectively, is present on both snRNPs; U2 snRNPs are enriched in the former, and U1 snRNPs are enriched in the latter.     

Immunoprecipitation distinguishes non-overlapping groups of snRNPs in Schizosaccharomyces pombe.

The large number of snRNAs in the fission yeast Schizosaccharomyces pombe can be divided into four non-overlapping groups by immunoprecipitation with antibodies directed against mammalian snRNP proteins. 1) Of the abundant snRNAs, anti-Sm sera precipitate only the spliceosomal snRNAs U1, U2, U4, U5 and U6. Surprisingly, three Sm-sera tested distinguish between U2, U4 and U5 and U1 from S.pombe; one precipitating only U1 and two precipitating U2, U4 and U5 but not U1. 2) A group of 11 moderately abundant snRNAs are not detectably precipitated by human anti-Sm sera, but are specifically precipitated by monoclonal antibody H57 specific for the human B/B' polypeptides. From Aspergillus nidulans this antibody also precipitates at least 12 snRNAs. 3) Anti-(U3)RNP sera do not precipitate the above snRNAs, but precipitate at least 6 further snRNAs, including the homologues of U3. Both the anti-(U3)RNP sera and H57 also efficiently precipitate a number of discrete non-capped RNAs. 4) A small number of additional snRNAs are not detectably precipitated by any anti-serum tested to date, further analysis may identify antisera specific for these snRNPs. Western blots of purified snRNP proteins were used to identify the S.pombe proteins responsible for these immunoprecipitations. Several Sm-sera decorate a 16.3kD protein which may be a D protein homologue, monoclonal H57 decorates a further protein of 16kD and an anti-(U3)RNP serum decorates the homologue of the 36kD U3-specific protein, fibrillarin.     

Localization and structure of snRNPs during mitosis. Immunofluorescent and biochemical studies

The distribution of U snRNAs during mitosis was studied by indirect immunofluorescence microscopy with snRNA cap-specific anti-m3G antibodies. Whereas the snRNAs are strictly nuclear at late prophase, they become distributed in the cell plasm at metaphase and anaphase. They re-enter the newly formed nuclei of the two daughter cells at early telophase, producing speckled nuclear fluorescent patterns typical of interphase cells. While the snRNAs become concentrated at the rim of the condensing chromosomes and at interchromosomal regions at late prophase, essentially no association of the snRNAs was observed with the condensed chromosomes during metaphase and anaphase. Independent immunofluorescent studies with anti-(U1)RNP autoantibodies, which react specifically with proteins unique to the U1 snRNP species, showed the same distribution of snRNP antigens during mitosis as was observed with the snRNA-specific anti-m3G antibody. Immunoprecipitation studies with anti-(U1)RNP and anti-Sm autoantibodies, as well as protein analysis of snRNPs isolated from extracts of mitotic cells, demonstrate that the snRNAs remain associated in a specific manner with the same set of proteins during interphase and mitosis. The concept that the overall structure of the snRNPs is maintained during mitosis also applies to the coexistence of the snRNAs U4 and U6 in a single ribonucleoprotein complex. Particle sedimentation studies in sucrose gradients reveal that most of the snRNPs present in sonicates of mitotic cells do not sediment as free RNP particles, but remain associated with high molecular weight (HMW) structures other than chromatin, most probably with hnRNA/RNP.     

Characterisation of human and murine snRNP proteins by two-dimensional gel electrophoresis and phosphopeptide analysis of U1-specific 70K protein variants.

The proteins of the major human snRNPs U1, U2, U4/U6 and U5 were characterised by two-dimensional electrophoresis, with isoelectric focussing in the first dimension and SDS-polyacrylamide gel electrophoresis in the second. With the exception of protein F, which exhibits an acidic pl value (pl = 3.3), the snRNP proteins are basic. Post-translational modification was found among the proteins associated specifically with the U1 and U2 particles. The most complex modification pattern was observed for the U1-specific 70K protein. This was found in at least 13 isoelectric variants, with pl values ranging from 6.7 to 8.7; these variants differed also in molecular weight. All of the 70K variants are phosphorylated in the cell. Thin-layer analysis of their tryptic phosphopeptides revealed that the 70K variants have four major phosphopeptides in common, in addition to which at least four additional serine residues are phosphorylated to different extents. The comparative phosphopeptide analysis shows that differential phosphorylation alone is not sufficient to explain the occurrence of the many isoelectric variants of 70K, so that the final charge of the 70K variants is determined both by phosphorylation and by other, as yet unidentified posttranslational modifications. By two-dimensional separation of snRNP proteins obtained from mouse Ehrlich ascites tumour cells, it was shown that the pattern of pl values of the mouse proteins was almost identical with the corresponding pattern for human proteins. Even the complex modification patterns of the 70K protein are identical in mouse and man, indicating that the presence in the cell of so many variants of this protein may have functional importance. The major difference between murine and human snRNP proteins is the absence of protein B' from mouse snRNPs. This suggests that the homologous protein B may be able to carry out the task of protein B'.     

In vitro reconstitution of mammalian U2 and U5 snRNPs active in splicing: Sm proteins are functionally interchangeable and are essential for the formation of functional U2 and U5 snRNPs.

An in vitro reconstitution/splicing complementation system has been developed which has allowed the investigation of the role of mammalian U2 and U5 snRNP components in splicing. U2 or U5 snRNP cores are first reconstituted from purified native snRNP core proteins and snRNA in the absence of cellular extract and are subsequently added to splicing extracts depleted of either U2 or U5 snRNP. When snRNPs reconstituted with HeLa U2 or U5 snRNA were added to U2- or U5-depleted nuclear extract, splicing was complemented. Addition of naked snRNA, on the other hand, did not restore splicing, demonstrating that the core proteins are essential for both U2 and U5 snRNP functions in splicing. Hybrid U2 or U5 snRNPs, reconstituted with core proteins isolated from U1 or U2 snRNPs, were equally active in splicing complementation, indicating that the snRNP core proteins are functionally interchangeable. U5 snRNPs reconstituted from in vitro transcribed U5 snRNA restored splicing to a level identical to that observed with particles reconstituted from authentic HeLa U5 snRNA. In contrast, splicing could not be restored to U2-depleted extract by the addition of snRNPs reconstituted from synthetic U2 snRNA, suggesting that U2 snRNA base modifications are essential for U2 snRNP function.     

A split luciferase-based reporter for detection of a cellular macromolecular complex

The spliceosome is a highly dynamic macromolecular ribonucleoprotein (RNP) machine that catalyzes pre-mRNA splicing by assembling U1, U2, U4, U5, and U6 small nuclear RNPs (snRNPs). To process large numbers of introns with a limited number of snRNPs, synthesis and recycling of snRNPs must be maintained within an appropriate range to avoid their shortage. However, the mechanism that maintains cellular snRNP levels is unknown. Molecules that modulate cellular snRNP levels may help to define this mechanism but are not available. Therefore, the goal of the current study was to develop a reporter for snRNP levels using split luciferase based on proteomic analysis of snRNPs. We constructed an expression library of a luciferase fragment fused to core components of U5 snRNP and used it to isolate pre-mRNA processing factor 6 (PRPF6) and small nuclear ribonucleoprotein 40 kDa (U5-40K) that specifically reconstitute luciferase activity in the U5 snRNP complex. Here we show that this reporter detects the effects of small molecules on the levels of the U5 snRNP reporter protein complex. Our approach provides an alternative assay to discover small molecules targeting a macromolecular complex when the structure of the complex is not precisely identified.     

A 69-kD protein that associates reversibly with the Sm core domain of several spliceosomal snRNP species

The biogenesis of the spliceosomal small nuclear ribonucleoproteins (snRNPs) U1, U2, U4, and U5 involves: (a) migration of the snRNA molecules from the nucleus to the cytoplasm; (b) assembly of a group of common proteins (Sm proteins) and their binding to a region on the snRNAs called the Sm-binding site; and (c) translocation of the RNP back to the nucleus. A first prerequisite for understanding the assembly pathway and nuclear transport of the snRNPs in more detail is the knowledge of all the snRNP proteins that play essential roles in these processes. We have recently observed a previously undetected 69- kD protein in 12S U1 snRNPs isolated from HeLa nuclear extracts under non-denaturing conditions that is clearly distinct from the U1-70K protein. The following evidence indicates that the 69-kD protein is a common, rather than a U1-specific, protein, possibly associating with the snRNP core particles by protein-protein interaction. (a) Antibodies raised against the 69-kD protein, which did not cross-react with any of the Sm proteins B'-G, precipitated not only U1 snRNPs, but also the other spliceosomal snRNPs U2, U4/U6 and U5, albeit to a lower extent. (b) U1, U2, and U5 core RNP particles reconstituted in vitro contain the 69-kD protein. (c) Xenopus laevis oocytes contain an immunologically related homologue of the human 69-kD protein. When U1 snRNA as well as a mutant U1 snRNA, that can bind the Sm core proteins but lacks the capacity to bind the U1-specific proteins 70K, A, and C, were injected into Xenopus oocytes to allow assembly in vivo, they were recognized by antibodies specific against the 69-kD protein in the ooplasm and in the nucleus. The 69-kD protein is under-represented, if present at all, in purified 17S U2 and in 25S [U4/U6.U5] tri-snRNPs, isolated from HeLa nuclear extracts. Our results are consistent with the working hypothesis that this protein may either play a role in the cytoplasmic assembly of the core domain of the snRNPs and/or in the nuclear transport of the snRNPs. After transport of the snRNPs into the nucleus, it may dissociate from the particles as for example in the case of the 17S U2 or the 25S [U4/U6.U5] tri-snRNP, which bind more than 10 different snRNP specific proteins each in the nucleus.     

Polypeptide components of Drosophila small nuclear ribonucleoprotein particles.

In eukaryotes splicing of pre-mRNAs is mediated by the spliceosome, a dynamic complex of small nuclear ribonucleoprotein particles (snRNPs) that associate transiently during spliceosome assembly and the splicing reaction. We have purified snRNPs from nuclear extracts of Drosophila cells by affinity chromatography with an antibody specific for the trimethylguanosine (m3G) cap structure of snRNAs U1-U5. The polypeptide components of Drosophila snRNPs have been characterized and shown to consist of a number of proteins shared by all the snRNPs, and some proteins which appear to be specific to individual snRNP particles. On the basis of their apparent molecular weight and antigenicity many of these common and particle specific Drosophila snRNP proteins are remarkably conserved between Drosophila and human spliceosomes. By probing western blots of the Drosophila snRNP polypeptides with a number of antisera raised against human snRNP proteins, Drosophila polypeptides equivalent to many of the HeLa snRNP-common proteins have been identified, as well as candidates for a number of U1, U2 and U5-specific proteins.     

The human 18S U11/U12 snRNP contains a set of novel proteins not found in the U2-dependent spliceosome

U11 and U12 snRNPs bind U12-type pre-mRNAs as a preformed di-snRNP complex, simultaneously recognizing the 5' splice site and branchpoint sequence. Thus, within the U12-type prespliceosome, U11/U12 components form a molecular bridge connecting both ends of the intron. We have affinity purified human 18S U11/U12 and 12S U11 snRNPs, and identified their protein components by using mass spectrometry. U11/U12 snRNPs lack all known U1 snRNP proteins but contain seven novel proteins (i.e., 65K, 59K, 48K, 35K, 31K, 25K, 20K) not found in the major spliceosome, four of which (59K, 48K, 35K, and 25K) are U11-associated. Thus, protein-protein and protein-RNA interactions contributing to 5' splice site recognition and/or intron bridging appear to differ significantly in the minor versus major prespliceosome. The majority of U11/U12 proteins are highly conserved in organisms known to contain U12-type introns. However, homologs of those associated with U11 were not detected in Drosophila melanogaster, consistent with the presence of a divergent U11 snRNP in flies. RNAi experiments revealed that several U11/U12 proteins are essential for cell viability, suggesting they play key roles in U12-type splicing. The presence of unique U11/U12 snRNP proteins in the U12-type spliceosome provides insight into potential evolutionary relationships between the major and minor spliceosome.     

Nuclear organization of splicing small nuclear ribonucleoproteins in adenovirus-infected cells.

We have studied the effect of adenovirus infection on the nuclear organization of splicing small nuclear ribonucleoproteins (snRNPs) in HeLa cells. In uninfected HeLa cells, snRNPs are widespread throughout the nucleoplasm but also are concentrated in specific nuclear structures, including coiled bodies, interchromatin granules, and perichromatin fibrils. We have used immunofluorescence microscopy to study the localization of splicing snRNPs relative to centers of viral DNA synthesis and accumulation identified with antiserum against the viral 72,000-molecular-weight single-stranded DNA-binding protein (72K protein). Splicing snRNPs were independently detected with both monoclonal and polyclonal antibodies specific for common snRNP antigens, snRNP-specific proteins, and the snRNA-specific 2,2,7-trimethylguanosine 5' cap structure. We have examined infected cells 2 to 24 h after infection, and, in the majority of these cells, we observed no colocalization of the snRNP and 72K-protein staining patterns. In the late phase, snRNPs were found to markedly concentrate in discrete clusters that were distinct from the centers of viral DNA synthesis and accumulation identified with anti-72K protein. We have treated cells with hydroxyurea at various times after infection to inhibit aspects of the virus infectious program. We have found that the accumulation of snRNP clusters is correlated with late gene expression rather than with DNA synthesis or early gene expression. Finally, we show that the late-phase snRNP clusters colocalize with a monoclonal antibody that primarily stains interchromatin granules. These results suggest that the centers of snRNP concentration in late-phase infected cells are likely to correspond to interchromatin granule clusters.     

RNA-protein organization of U1, U5 and U4-U6 small nuclear ribonucleoproteins in HeLa cells

Small nuclear ribonucleoproteins (snRNPs) containing U1 and U5 snRNAs from HeLa cells have been fractionated using a combination of isopycnic centrifugation in cesium chloride and ion-exchange chromatography on DEAE-Sepharose. The procedure is based on the extreme stability conferred upon snRNPs by Mg2+ enabling them to withstand the very high ionic strength that prevails in cesium chloride. U1 snRNP prepared by this method contains all nine major proteins (68K, A, B, B', C, D, E, F, G) corresponding to those previously identified by immunoprecipitation and is therefore precipitable by anti-RNP and anti-Sm antibodies. U5 snRNP purified in this way contains the common D to G proteins and is also enriched in a 25 X 10(3) Mr protein that may be U5 snRNP-specific. The core-resistant U5 snRNA sequence (nucleotide 84 to 3' OH) covered by D to G proteins is extended by only six nucleotides. A similar situation is seen in U4-U6 snRNP, which we have obtained in a sufficiently pure form to examine protected sequences. However, the core-resistant sequence of U4 (nucleotide 116 to 3' OH) in U4-U6 snRNP is extended by 37 nucleotides, suggesting that the protein composition of this particle could be more complex than that of U5 snRNP. The ribonucleoprotein organization of snRNPs is summarized and discussed in view of our current knowledge on snRNA sequences protected by proteins.     

The 5'' end domain of U2 snRNA is required to establish the interaction of U2 snRNP with U2 auxiliary factor(s) during mammalian spliceosome assembly.

Stable association of U2 snRNP with the branchpoint sequence of mammalian pre-mRNAs requires binding of a non-snRNP protein to the polypyrimidine tract. In order to determine how U2 snRNP contacts this protein, we have used an RNA containing the consensus 5' and the (Py)n-AG 3' splice sites but lacking the branchpoint sequence so as to prevent direct U2 snRNA base pairing to the branchpoint. Different approaches including electrophoretic separation of RNP complexes formed in nuclear extracts, RNase T1 protection immunoprecipitation assays with antibodies against snRNPs and UV cross-linking experiments coupled to immunoprecipitations allowed us to demonstrate that at least three splicing factors contact this RNA at 0 degree C without ATP. As expected, U1 snRNP interacts with the region comprising the 5' splice site. A protein of approximately 65,000 molecular weight recognizes the RNA specifically at the 5' boundary of the polypyrimidine tract. It could be either the U2 auxiliary factor (U2AF) (Zamore and Green (1989) PNAS 86, 9243-9247), the polypyrimidine tract binding protein (pPTB) (Garcia-Blanco et al. (1989) Genes and Dev. 3, 1874-1886) or a mixture of both. U2 snRNP also contacts the RNA in a way depending on p65 binding, thereby further arguing that the latter may correspond to the previously characterized U2AF and pPTB. Cleavage of U2 snRNA sequence by a complementary oligonucleotide and RNase H led us to conclude that the 5' terminus of U2 snRNA is required to ensure the contact between U2 snRNP and p65 bound to the RNA. More importantly, this conclusion can be extended to authentic pre-mRNAs. When we have used a human beta-globin pre-mRNA instead of the above artificial substrate, RNA bound p65 became precipitable by anti-(U2) RNP and anti-Sm antibodies except when the 5' end of U2 snRNA was selectively cleaved.