Structure and assembly of the spliceosomal snRNPs. Novartis Medal Lecture

The spliceosome is a macromolecular machine that carries out the excision of introns from eukaryotic pre-mRNAs and splicing together of exons. Four large RNA-protein complexes, called the U1, U2, U4/U6 and U5 small nuclear ribonucleoprotein particles (snRNPs), and some non-snRNP proteins assemble around three short conserved sequences within the intron in an ordered manner to form the active spliceosome. We aim to provide insight into the molecular details of the mechanism of pre-mRNA splicing through crystallographic studies of the snRNPs. We have solved the X-ray crystal structure of some snRNP proteins as part of either protein-protein complexes or RNA-protein complexes. These structures have provided an important insight into the overall architecture of the U1 and U2 snRNPs and the mechanisms of RNA-protein and protein-protein recognition.

Stoichiometry of the Sm proteins in yeast spliceosomal snRNPs supports the heptamer ring model of the core domain

Seven Sm proteins (B/B', D1, D2, D3, E, F and G proteins) containing a common sequence motif form a globular core domain within the U1, U2, U5 and U4/U6 spliceosomal snRNPs. Based on the crystal structure of two Sm protein dimers we have previously proposed a model of the snRNP core domain consisting of a ring of seven Sm proteins. This model postulates that there is only a single copy of each Sm protein in the core domain. In order to test this model we have determined the stoichiometry of the Sm proteins in yeast spliceosomal snRNPs. We have constructed seven different yeast strains each of which produces one of the Sm proteins tagged with a calmodulin-binding peptide (CBP). Further, each of these strains was transformed with one of seven different plasmids coding for one of the seven Sm proteins tagged with protein A. When one Sm protein is expressed as a CBP-tagged protein from the chromosome and a second protein was produced with a protein A-tag from the plasmid, the protein A-tag was detected strongly in the fraction bound to calmodulin beads, demonstrating that two different tagged Sm proteins can be assembled into functional snRNPs. In contrast when the CBP and protein A-tagged forms of the same Sm protein were co-expressed, no protein A-tag was detectable in the fraction bound to calmodulin. These results indicate that there is only a single copy of each Sm protein in the spliceosomal snRNP core domain and therefore strongly support the heptamer ring model of the spliceosomal snRNP core domain.

Crystal structure of the actin-binding region of utrophin reveals a head-to-tail dimer

BACKGROUND

Utrophin is a large multidomain protein that belongs to a superfamily of actin-binding proteins, which includes dystrophin, alpha-actinin, beta-spectrin, fimbrin, filamin and plectin. All the members of this family contain a common actin-binding region at their N termini and perform a wide variety of roles associated with the actin cytoskeleton. Utrophin is the autosomal homologue of dystrophin, the protein defective in the X-linked Duchenne and Becker muscular dystrophies, and upregulation of utrophin has been suggested as a potential therapy for muscular dystrophy patients.

RESULTS

The structure of the actin-binding region of utrophin, consisting of two calponin-homology (CH) domains, has been solved at 3.0 A resolution. It is composed of an antiparallel dimer with each of the monomers being present in an extended dumbell shape and the two CH domains being separated by a long central helix. This extended conformation is in sharp contrast to the compact monomer structure of the N-terminal actin-binding region of fimbrin.

CONCLUSIONS

The crystal structure of the actin-binding region of utrophin suggests that these actin-binding domains may be more flexible than was previously thought and that this flexibility may allow domain reorganisation and play a role in the actin-binding mechanism. Thus utrophin could possibly bind to actin in an extended conformation so that the sites previously identified as being important for actin binding may be directly involved in this interaction.

Hsp26: a temperature-regulated chaperone

Small heat shock proteins (sHsps) are a conserved protein family, with members found in all organisms analysed so far. Several sHsps have been shown to exhibit chaperone activity and protect proteins from irreversible aggregation in vitro. Here we show that Hsp26, an sHsp from Saccharomyces cerevisiae, is a temperature-regulated molecular chaperone. Like other sHsps, Hsp26 forms large oligomeric complexes. At heat shock temperatures, however, the 24mer chaperone complex dissociates. Interestingly, chaperone assays performed at different temperatures show that the dissociation of the Hsp26 complex at heat shock temperatures is a prerequisite for efficient chaperone activity. Binding of non-native proteins to dissociated Hsp26 produces large globular assemblies with a structure that appears to be completely reorganized relative to the original Hsp26 oligomers. In this complex one monomer of substrate is bound per Hsp26 dimer. The temperature-dependent dissociation of the large storage form of Hsp26 into a smaller, active species and the subsequent re-association to a defined large chaperone-substrate complex represents a novel mechanism for the functional activation of a molecular chaperone.

Structure and assembly of the spliceosomal small nuclear ribonucleoprotein particles

The spliceosome is a macromolecular assembly that carries out the excision of introns from nuclear pre-mRNAs. It consists of four large RNA-protein complexes, called the U1, U2, U4/U6 and U5 small nuclear ribonucleoproteins (snRNPs), and many protein factors. Crystal structures of seven protein components and fragments of the U1 and U2 small nuclear RNAs have been determined in the form of RNA-protein and protein-protein complexes. Together with electron microscopy studies of the snRNPs, these structures have begun to provide important insights into the architecture of the snRNPs and the mechanisms of RNA-protein and protein-protein recognition.

Crystal structures of two Sm protein complexes and their implications for the assembly of the spliceosomal snRNPs

The U1, U2, U4/U6, and U5 small nuclear ribonucleoprotein particles (snRNPs) involved in pre-mRNA splicing contain seven Sm proteins (B/B', D1, D2, D3, E, F, and G) in common, which assemble around the Sm site present in four of the major spliceosomal small nuclear RNAs (snRNAs). These proteins share a common sequence motif in two segments, Sm1 and Sm2, separated by a short variable linker. Crystal structures of two Sm protein complexes, D3B and D1D2, show that these proteins have a common fold containing an N-terminal helix followed by a strongly bent five-stranded antiparallel beta sheet, and the D1D2 and D3B dimers superpose closely in their core regions, including the dimer interfaces. The crystal structures suggest that the seven Sm proteins could form a closed ring and the snRNAs may be bound in the positively charged central hole.

S. cerevisiae and sulfur: a unique way to deal with the environment

Saccharomyces cerevisiae is by far the best-studied unicellular eukaryote. Although yeast cells are very similar to higher eukaryotes in many respects, there is striking evidence that S. cerevisiae is not a perfect model for a eukaryotic cell (cf. 1). Here we report that yeast proteins contain a significantly lower amount of cysteine residues compared to other eukaryotes. Explanations for this phenomenon could not be found in the sulfur metabolism of yeast, which showed no major differences from other organisms (2-4). However, previous examinations could link a defect in sulfate uptake of S. cerevisiae to an increased resistance against toxic substances like selenate and chromate in the environment, which share the same permeases (5-7). This environmental problem might have caused S. cerevisiae to down-regulate its sulfate uptake and therefore lead to a lower amount of available sulfur in the cell, making it necessary to replace all dispensable sulfur amino acids in proteins. We show in two examples that S. cerevisiae proteins contain only such cysteine residues that are structurally or functionally needed. Therefore, we conclude that S. cerevisiae has solved a widespread environmental problem in a specific way which might be unique among eukaryotes.

The single-ring Thermoanaerobacter brockii chaperonin 60 (Tbr-EL7) dimerizes to Tbr-EL14.Tbr-ES7 under protein folding conditions

Chaperone proteins assist in the folding of some newly synthesized proteins and inhibit protein aggregation. The Thermoanaerobacter brockii chaperonin proteins (Tbr-EL and Tbr-ES) have recently been purified and characterized [Truscott, W.N., Høj, P. B., & Scopes, R. K. (1994) Eur. J. Biochem. 222, 277-284]; Tbr-EL was a single seven-membered toroid, unlike most GroELs which exist as double toroids. Using high-resolution gel filtration chromatography, we have resolved the purified Tbr-EL into single ringed (Tbr-EL7) and double ringed (Tbr-EL14) species. The latter contained tightly bound Tbr-ES co-chaperonin (Tbr-EL14.Tbr-ES7). In the presence of Mg.ATP and either Escherichia coli GroES (Eco-ES) or Tbr-ES (i.e., under protein folding conditions), the isolated Tbr-EL7 rapidly dimerized to the Tbr-EL14.Eco-ES7 or Tbr-EL14.Tbr-ES7 complexes. The doubly toroidal species thus formed contained > or = 6 molecules tightly bound ADP and one GroES7 and are similar to the asymmetric chaperonin complex isolated from Thermus thermophilus [Taguch, H., Konishi, J., Ishii, N., & Yoshida, M. (1991) J. Biol. Chem. 266, 22411-22418]. The isolated Tbr-EL7 and Tbr-EL14.Tbr-ES7 hydrolyzed ATP at approximate to 2 and 1 min-1, respectively. Addition of a molar excess of Eco-ES7 to the isolated Tbr-EL7 reduced the ATPase activity to 1 min-1, consistent with the formation of Tbr-EL14.Eco-ES7. Eco-ES7 failed to inhibit the Tbr-El14.Tbr-ES7 complex. The isolated Tbr-EL14.Tbr-ES7 complex did not support the folding of Rubisco under nonpermissive conditions. Only when the complex was supplemental with additional GroES was folding of Rubisco observed; i.e., one molar equivalent of GroES was not sufficient for folding. Both Tbr-EL7 and Tbr-EL14.Tbr-ES7 bound on unfolded [35S] Rhodospirillum rubrum Rubisco per mole particle. In contrast, Eco-EL14 bound 2 mol of protein per mole particle, consistent with each toroid having a peptide binding site. Eco-EL14.Eco-ES7 complex only bound one unfolded protein, thus GroES binding blocks one GroEL peptide binding site. Addition of Eco-ES7 to a Eco-EL14.Rubisco2 complex did not result in the displacement of one molecule of Rubisco but in the formation of a ternary Eco-EL14.Rubisco2.Eco-ES7 complex.