Solution structure of the partially folded high-risk human papilloma virus 45 oncoprotein E7.

The oncoprotein E7 of human papilloma viruses (HPV) is involved in the pathogenesis and maintenance of human cervical cancers. The most prevalent HPV types found in cervix carcinomas are HPV16, 18 and 45. The structure of the E7 dimer from HPV45 (PDB 2F8B) was determined by nuclear magnetic resonance spectroscopy. Each monomer comprises an unfolded N-terminus and a well-structured C-terminal domain with a beta1beta2alpha1beta3alpha2 topology representing a unique zinc-binding fold found only for E7. Dimerization occurs through the alpha1/alpha1' helices and intermolecular beta-sheet formation but excludes the zinc-binding sites. E7 is reported to interact with a number of cellular proteins (e.g. pRb, p21(CIP1)). Binding of a peptide derived from the C-terminus of p21(CIP1) to the C-terminal domain of E7 was characterized by monitoring chemical shift perturbations of the amide groups of E7. This provides direct evidence that a shallow groove situated between alpha1 and beta1 of the E7 C-terminal domain is interacting with the C-terminus of p21(CIP1). Intriguingly, this binding site overlaps with the low-affinity binding site on E7 for the C-domain of pRb.

2D relayed anisotropy correlation NMR: characterization of the 13C' chemical shift tensor orientation in the peptide plane of the dipeptide AibAib.

An approach to the determination of the orientation of the carbonyl chemical shift (CS) tensor in a 13C-15N-1H dipolar coupled spin network is proposed. The method involves the measurement of the Euler angles of the 13C'-15N and 15N-1H dipolar vectors in the 13C' CS tensor principal axes system, respectively, via a 13C-15N REDOR experiment and by a 2D relayed anisotropy correlation of the 13C' CSA (omega2) and 15N-1H dipolar interaction (omega1). Via numerical simulations the sensitivity of the omega1 cross sections of the 2D spectrum to the Euler angles of the 15N-1H bond vector in the 13C' CSA frame is shown. Employing the procedure outlined in this work, we have determined the orientation of the 13C' CS tensor in the peptide plane of the dipeptide AibAib-NH2 (Aib = alpha-aminoisobutyric acid). The Euler angles are found to be (chiCN, psiCN) = (34 degrees +/- 2 degrees, 88 degrees +/- 2 degrees) and (chiNH, psiNH) = (90 degrees +/- 10 degrees, 80 degrees +/- 10 degrees). From the measured Euler angles it is seen that the sigma33 and sigma22 components of the 13C' CS tensor approximately lie in the peptide plane.

The 3D arrangement of the 23 S and 5 S rRNA in the Escherichia coli 50 S ribosomal subunit based on a cryo-electron microscopic reconstruction at 7.5 A resolution.

The Escherichia coli 23 S and 5 S rRNA molecules have been fitted helix by helix to a cryo-electron microscopic (EM) reconstruction of the 50 S ribosomal subunit, using an unfiltered version of the recently published 50 S reconstruction at 7.5 A resolution. At this resolution, the EM density shows a well-defined network of fine structural elements, in which the major and minor grooves of the rRNA helices can be discerned at many locations. The 3D folding of the rRNA molecules within this EM density is constrained by their well-established secondary structures, and further constraints are provided by intra and inter-rRNA crosslinking data, as well as by tertiary interactions and pseudoknots. RNA-protein cross-link and foot-print sites on the 23 S and 5 S rRNA were used to position the rRNA elements concerned in relation to the known arrangement of the ribosomal proteins as determined by immuno-electron microscopy. The published X-ray or NMR structures of seven 50 S ribosomal proteins or RNA-protein complexes were incorporated into the EM density. The 3D locations of cross-link and foot-print sites to the 23 S rRNA from tRNA bound to the ribosomal A, P or E sites were correlated with the positions of the tRNA molecules directly observed in earlier reconstructions of the 70 S ribosome at 13 A or 20 A. Similarly, the positions of cross-link sites within the peptidyl transferase ring of the 23 S rRNA from the aminoacyl residue of tRNA were correlated with the locations of the CCA ends of the A and P site tRNA. Sites on the 23 S rRNA that are cross-linked to the N termini of peptides of different lengths were all found to lie within or close to the internal tunnel connecting the peptidyl transferase region with the presumed peptide exit site on the solvent side of the 50 S subunit. The post-transcriptionally modified bases in the 23 S rRNA form a cluster close to the peptidyl transferase area. The minimum conserved core elements of the secondary structure of the 23 S rRNA form a compact block within the 3D structure and, conversely, the points corresponding to the locations of expansion segments in 28 S rRNA all lie on the outside of the structure.

The NMR structure of the 5S rRNA E-domain-protein L25 complex shows preformed and induced recognition.

The structure of the complex between ribosomal protein L25 and a 37 nucleotide RNA molecule, which contains the E-loop and helix IV regions of the E-domain of Escherichia coli 5S rRNA, has been determined to an overall r.m.s. displacement of 1.08 A (backbone heavy atoms) by heteronuclear NMR spectroscopy (Protein Databank code 1d6k). The interacting molecular surfaces are bipartite for both the RNA and the protein. One side of the six-stranded beta-barrel of L25 recognizes the minor groove of the E-loop with very little change in the conformations of either the protein or the RNA and with the RNA-protein interactions occurring mainly along one strand of the E-loop duplex. This minor groove recognition module includes two parallel beta-strands of L25, a hitherto unknown RNA binding topology. Binding of the RNA also induces conversion of a flexible loop to an alpha-helix in L25, the N-terminal tip of which interacts with the widened major groove at the E-loop/helix IV junction of the RNA. The structure of the complex reveals that the E-domain RNA serves as a preformed docking partner, while the L25 protein has one preformed and one induced recognition module.

Direct identification of NH...N hydrogen bonds in non-canonical base pairs of RNA by NMR spectroscopy.

It is shown that the recently developed quantitative J(NN)HNN-COSY experiment can be used for the direct identification of hydrogen bonds in non-canonical base pairs in RNA. Scalar(2h)J(NN)couplings across NH.N hydrogen bonds are observed in imino hydrogen bonded GA base pairs of the hpGA RNA molecule, which contains a tandem GA mismatch, and in the reverse Hoogsteen AU base pairs of the E-loop of Escherichia coli 5S rRNA. These scalar couplings correlate the imino donor(15)N nucleus of guanine or uridine with the acceptor N1 or N7 nucleus of adenine. The values of the corresponding(2h)J(NN)coupling constants are similar in size to those observed in Watson-Crick base pairs. The reverse Hoogsteen base pairs could be directly detected for the E-loop of E.coli 5S rRNA both in the free form and in a complex with the ribosomal protein L25. This supports the notion that the E-loop is a pre-folded RNA recognition site that is not subject to significant induced conformational changes. Since Watson-Crick GC and AU base pairs are also readily detected the HNN-COSY experiment provides a useful and sensitive tool for the rapid identification of RNA secondary structure elements.

Triple-resonance experiments for correlation of H5 and exchangeable pyrimidine base hydrogens in (13)C,(15)N-labeled RNA.

Triple-resonance two-dimensional H5(C5C4N)H experiments are described that provide through-bond H5 to imino/amino connectivities in uridines and cytidines in (13)C, (15)N-labeled RNAs. The experiments employ selective INEPT steps for transferring magnetization from the H5 hydrogens through the intervening C5, C4, and N3/N4 nuclei to the imino/amino hydrogens. The improved sensitivity of these experiments for assignments in a large 43-nucleotide RNA is demonstrated.

Functional proteomics analysis of signal transduction pathways of the platelet-derived growth factor beta receptor.

We report efficient methods for using functional proteomics to study signal transduction pathways in mouse fibroblasts following stimulation with PDGF. After stimulation, complete cellular proteins were separated using two-dimensional electrophoresis and phosphorylated proteins were detected with anti-phosphotyrosine and anti-phosphoserine antibodies. About 260 and 300 phosphorylated proteins were detected with the anti-phosphotyrosine and anti-phosphoserine antibodies, respectively, at least 100 of which showed prominent changes in phosphorylation as a function of time after stimulation. Proteins showing major time-dependent changes in phosphorylation were subjected to in-gel digestion with trypsin and identified by mass spectroscopy using MALDI-TOF mass fingerprinting and ESI peptide sequencing. We have observed phosphorylated proteins known to be part of the PDGF signal transduction pathway such as ERK 1, serine/threonine protein kinase akt and protein tyrosine phosphatase syp, proteins such as proto-oncogene tyrosine kinase fgr previously known to participate in other signal transduction pathways, and some proteins such as plexin-like protein with no previously known function in signal transduction. Information about the phosphorylation site was obtained for proto-oncogene tyrosine kinase fgr and for cardiac alpha-actin. The methods used here have proven to be suitable for the identification of time-dependent changes in large numbers of proteins involved in signal transduction pathways.

The NMR structure of Escherichia coli ribosomal protein L25 shows homology to general stress proteins and glutaminyl-tRNA synthetases.

The structure of the Escherichia coli ribosomal protein L25 has been determined to an r.m.s. displacement of backbone heavy atoms of 0.62 +/- 0.14 A by multi-dimensional heteronuclear NMR spectroscopy on protein samples uniformly labeled with 15N or 15N/13C. L25 shows a new topology for RNA-binding proteins consisting of a six-stranded beta-barrel and two alpha-helices. A putative RNA-binding surface for L25 has been obtained by comparison of backbone 15N chemical shifts for L25 with and without a bound cognate RNA containing the eubacterial E-loop that is the site for binding of L25 to 5S ribosomal RNA. Sequence comparisons with related proteins, including the general stress protein, CTC, show that the residues involved in RNA binding are highly conserved, thereby providing further confirmation of the binding surface. Tertiary structure comparisons indicate that the six-stranded beta-barrels of L25 and of the tRNA anticodon-binding domain of glutaminyl-tRNA synthetase are similar.

Structure of an RNA hairpin loop with a 5'-CGUUUCG-3' loop motif by heteronuclear NMR spectroscopy and distance geometry.

Structural features of a 19-nucleotide RNA hairpin loop (5'-GGCGUACGUUUCGUACGCC-3'), a loop motif which occurs in eukaryotic 18S rRNA, have been derived using multidimensional heteronuclear NMR spectroscopy in combination with local conformational analysis and torsion angle distance geometry followed by restrained energy minimization. A method to obtain both the 3JC4'P3' and 3JC4'P5' coupling constants from a set of spin-echo difference constant time HSQC spectra is introduced, and it is shown how these couplings can be assigned to the backbone angles beta and epsilon. A total of 280 distance constraints as well as 132 homo- and heteronuclear three-bond scalar coupling constants were derived from the NMR data. The structure which has been determined is a pentaloop rather than a triloop with no base pairing between G8 and C12. G8 is pointed to the minor groove where it forms a base triplet with C7-G13 that is further stabilized by hydrogen bonding to the 2'-hydroxyl group of C7. C12 is directed to the major groove where its conformation is stabilized by hydrogen bonding between O2 and HO2'. The NMR data suggest two possible, interconverting conformations with stacking of bases U10-G8 or U11-C7. Overall, the loop provides a variety of interaction sites for RNA or protein interactions.

An NMR study of the interaction of 15N-labelled bradykinin with an antibody mimic of the bradykinin B2 receptor.

An isotope-edited NMR study of the peptide hormone bradykinin (RPPGFSPFR) bound to the Fab fragment of a monoclonal antibody against bradykinin (MBK3) is reported. MBK3 was previously shown to provide a binding site model of the B2 bradykinin receptor [Haasemann, M., Buschko, J., Faussner, A., Roscher, A. A., Hoebeke, J., Burch, R. M. & Muller-Esterl, W. (1991) Anti-idiotypic antibodies bearing the internal image of a bradykinin epitope, J. Immunol. 147, 3882-3892]. Bradykinin was obtained in a uniformly 15N-labelled form using recombinant expression of a fusion protein consisting of the glutathione-binding domain of glutathione S-transferase fused to residues 354-375 of the high-molecular-mass kininogen from which bradykinin was released by proteolytic digestion with its natural protease plasma kallikrein. Bradykinin forms a complex with the Fab fragment of MBK3 which exchanges slowly on the NMR time scale. The 15N and 1H resonances of the tightly bound residues of bradykinin show appreciable changes in chemical shift with respect to the free form, while the 15N and 1H linewidths indicate that the hydrodynamic behaviour of bound bradykinin is dominated by the high-molecular-mass Fab fragment. The NMR data indicate that essentially the entire nonapeptide is involved in binding. The kinetics of the ligand-exchange process, together with resonance assignments obtained via exchange spectroscopy. indicate that bradykinin binds to MBK3 only in the all-trans conformation at all three Xaa-Pro amide bonds. NH-NH NOE connectivities suggest that bradykinin is bound in an extended conformation. The spectroscopic data obtained from this study are compared to recently proposed computational models of the conformation of bradykinin bound to the B2 receptor.

Initial analysis of 750 MHz NMR spectra of selective 15N-G,U labelled E. coli 5S rRNA.

The overall folding of an RNA molecule is reflected in its base pairing pattern. The identification of that pattern provides a first step towards the determination of the structure of an RNA molecule. We show that the application of heteronuclear NMR methods at 750 MHz to E. coli 5S rRNA (120 nucleotides) selectively labelled with 15N in guanine and uridine allows observation of base pairing patterns for a larger RNA molecule. We also present evidence that the fold of the E-domain of the 5S rRNA (nt 79-97) as a contiguous part of the 5S rRNA and as an isolated molecule is virtually the same.

cAMP-dependent phosphorylation of cytokeratin in hepatic inner mitochondrial membrane.

The phosphorylation pattern in mitochondrial fractions isolated from hepatocytes, preincubated with 32P-phosphate and stimulated with glucagon and calcium mobilizing hormones, was studied. Only in mitochondria from glucagon treated hepatocytes two phosphorylated protein bands were observed, one with a molecular weight (MW) of 54 kDa in the outer membrane fraction which, according to the literature, is suggested to represent protein kinase A; one with a MW of 20 kDa in the inner membrane fraction which has not been described earlier. Electroelution and digestion of the 20 kDa protein band yielded two tryptic peptides which were identified as fragments homologous to human cytokeratin type II (the sequence of rat cytokeratin type II is not known). From the amino acid composition and sequence, and from the known structure of type II cytokeratins, it is concluded that the 20 kDa phosphoprotein is composed of amino- and carboxylterminal proteolytic fragments of rat cytokeratin C8 which are tightly anchored in the inner mitochondrial membrane. The physiological significance of the possible interaction of cytoskeletal proteins with the mitochondrial inner membrane and its hormonal regulation are discussed.

The determinants of RNA-binding specificity of the heterogeneous nuclear ribonucleoprotein C proteins.

The hnRNP C proteins (C1/C2) are tenacious nuclear pre-mRNA-binding proteins that belong to the large RNP motif family of RNA-binding proteins. This motif identifies an RNA-binding domain (RBD) that consists of a four-stranded antiparallel beta-sheet packed against two alpha-helices. Despite considerable information on the structure of the hnRNP C RBD, little is known about its RNA-binding properties. To address this we used in vitro selection/amplification from pools of random sequence RNA to determine the RNA-binding specificity of hnRNP C1. After 8 rounds of selection/amplification nearly all RNAs contained contiguous stretches of at least 5 U residues, and filter-binding assays demonstrated that this sequence constitutes a high-affinity (Kd = 170 nM) binding site for hnRNP C1. The highest affinity we measured for hnRNP C1 was for r(U)14 (Kd = 14 nM). An RBD-containing peptide fragment of hnRNP C1 (amino acids 2-94) bound oligoribonucleotides containing an hnRNP C1 high-affinity binding site with nearly equal affinity to that of hnRNP C1. Unlike hnRNP C1, however, this peptide also bound oligoribonucleotides that do not contain high-affinity hnRNP C1-binding sites. We identified a region of 10 amino acids, immediately COOH-terminal to the RNP motif (amino acids 95-104), that prevents the minimal RBD from binding nonspecific RNA ligands. We propose that the highly conserved beta alpha beta beta alpha beta core structure of the RNP motif RBD confers a general RNA binding activity to RNP motif RBDs and that the determinants of RNA-binding specificity reside in the most variable regions, the loops connecting the beta-strands and/or the contiguous NH2 and COOH termini of the RBD.

The mRNA poly(A)-binding protein: localization, abundance, and RNA-binding specificity.

The poly(A)-binding protein (PABP) binds to the messenger (mRNA) 3'-poly(A) tail found on most eukaryotic mRNAs and together with the poly(A) tail has been implicated in governing the stability and the translation of mRNA. In order to further understand the role of the PABP in these processes, we have undertaken a detailed analysis of the cellular localization, the abundance, and the RNA-binding properties of the human PABP (hPABP). We raised monoclonal antibodies against the 70-kDa hPABP and confocal immunofluorescence microscopy with these antibodies reveals that it is localized exclusively to the cytoplasm. The hPABP exhibits a very low turnover rate in these cells and quantitative immunoblotting experiments demonstrated that growing HeLa cells contain a surprisingly high number of approximately 8 x 10(6) PABP molecules per cell, which corresponds to an intracellular concentration of about 4 microM. In an in vitro selection/amplification assay from random sequence oligonucleotide pools the hPABP selects oligo(rA)-rich sequences and it binds oligo(rA)25 with an apparent Kd of 7 nM. The hPABP binds to unrelated RNA sequences with an about 100-fold lower affinity (Kd > or = 0.5 microM). The abundance of the hPABP indicates that there is an approximately three-fold excess of the protein over binding sites on cytoplasmic poly(A). This excess and the high concentration of the hPABP, which is three orders of magnitude above its Kd for oligo(rA)25, suggest that the hPABP may bind to additional, lower affinity binding sites in vivo.

Interaction of the RNA-binding domain of the hnRNP C proteins with RNA.

The hnRNP C proteins are among the most abundant and avid pre-mRNA-binding proteins and they contain a consensus sequence RNA-binding domain (RBD) that is found in a large number of RNA-binding proteins. The interaction of the RBD of the hnRNP C proteins with an RNA oligonucleotide [r(U)8] was monitored by nuclear magnetic resonance (NMR). 15N and 13C/15N-labelled hnRNP C protein RBD was mixed with r(U)8 and one- and two-dimensional (1D and 2D) NMR spectra were recorded in a titration experiment. NMR studies of the uncomplexed 93 amino acid hnRNP C RBD (Wittekind et al., 1992) have shown that it has a compact folded structure (beta alpha beta beta alpha beta), which is typical for the RBD of this family of proteins and which is comprised of a four-stranded antiparallel beta-sheet, two alpha-helices and relatively unstructured amino- and carboxy-terminal regions. Sequential assignments of the polypeptide main-chain atoms of the hnRNP C RBD-r(U)8 complex revealed that these typical structural features are maintained in the complex, but significant perturbations of the chemical shifts of amide group atoms occur in a large number of residues. Most of these residues are in the beta-sheet region and especially in the terminal regions of the RBD. In contrast; chemical shifts of the residues of the well conserved alpha-helices, with the exception of Lys30, are not significantly perturbed. These observations localize the candidate residues of the RBD that are involved in the interaction with the RNA.(ABSTRACT TRUNCATED AT 250 WORDS)

1H, 13C, and 15N NMR assignments and global folding pattern of the RNA-binding domain of the human hnRNP C proteins.

The hnRNP C1 and C2 proteins are abundant nuclear proteins that bind avidly to heterogeneous nuclear RNAs (hnRNAs) and appear to be involved with pre-mRNA processing. The RNA-binding activity of the hnRNP C proteins is contained in the amino-terminal 94 amino acid RNA-binding domain (RBD) that is identical for these two proteins. We have obtained the 1H, 13C, and 15N NMR assignments for the RBD of the human hnRNP C proteins. The assignment process was facilitated by extensive utilization of three- and four-dimensional heteronuclear-edited spectra. Sequential assignments of the backbone resonances were made using a combination of 15N-edited 3D NOESY-HMQC, 3D TOCSY-HMQC, and 3D TOCSY-NOESY-HSQC as well as 3D HNCA, HNCO, and HCACO spectra. Side-chain resonances were assigned using 3D HCCH-COSY and 3D HCH-TOCSY spectra. Four-dimensional 13C/13C-edited NOESY and 13C/15N-edited NOESY experiments were used to unambigously resolve NOEs. The overall global folding pattern was established by calculating a set of preliminary structures using constraints derived from the sequential NOEs and a small number of long-range NOEs. The beta alpha beta-beta alpha beta domain structure exhibits an antiparallel beta-sheet with the conserved RNP 1 and RNP 2 sequences [Dreyfuss et al. (1988) Trends Biochem. Sci. 13, 86-91] located adjacent to one another as the two inner strands of the beta-sheet.

Primary structures of the heterogeneous nuclear ribonucleoprotein A2, B1, and C2 proteins: a diversity of RNA binding proteins is generated by small peptide inserts.

We have isolated cDNAs for the major heterogeneous nuclear ribonucleoprotein (hnRNP) A2, B1, and C2 proteins and determined their nucleotide and deduced amino acid sequences. The A2 and B1 cDNAs are identical except for a 36-nucleotide in-frame insert in B1. Similarly, the sequence of the C2 protein cDNA is related to that of C1 in that C2 contains an extra 39 in-frame nucleotides. Therefore, the B1 amino acid sequence is identical to A2 except for the insertion of 12 amino acids near its amino terminus, and C1 and C2 are also identical to each other except for an extra 13 amino acids near the middle of C2. All three proteins are members of a large family of RNA binding proteins that contain the consensus sequence-type RNA binding domain (CS-RBD). The A2 and B1 proteins have a modular structure similar to that of the hnRNP protein A1: they contain two CS-RBDs and a glycine-rich auxiliary domain at the carboxyl terminus. The CS-RBDs of A2 and B1 have approximately 80% amino acid identity with those of A1, whereas the glycine-rich auxiliary domain is considerably more divergent with less than 30% of the amino acids being identical. These findings indicate that the addition of small peptides, probably by alternative pre-mRNA splicing, generates some of the diversity apparent among hnRNP proteins.

Binding of globin mRNA, beta-globin mRNA segments and RNA homopolymers by immobilized protein of polysomal globin messenger ribonucleoprotein.

The binding of rabbit globin mRNA, in-vitro-generated beta-globin mRNA segments, and RNA homopolymers by proteins of rabbit reticulocyte polysomal messenger ribonucleoproteins (mRNP) after SDS gel electrophoresis and electroblotting was examined. The polysomal mRNP proteins have a higher affinity for mRNA than for rRNA and tRNA while having a higher affinity for polypurine than polypyrimidine homopolymers. Binding experiments with synthetic poly(A) and with segments of beta-globin mRNA transcribed from a cDNA in vitro revealed a set of polysomal mRNP proteins which preferentially bind the poly(A)-free beta-globin mRNA. A protein of Mr 90,000 binds specifically the 3'-nontranslated trailer of the poly(A)-free beta-globin mRNA and not the poly(A)-containing globin mRNA. Another set of proteins preferentially binds poly(A). The latter group of proteins contains a prominent species of Mr 72,000, which is most likely the rabbit poly(A)-binding protein. Three polysomal mRNP proteins which bound rabbit globin mRNA did not bind preferentially any of the other RNA probes used.