Mammalian mitochondrial ribosomal proteins (4). Amino acid sequencing, characterization, and identification of corresponding gene sequences.

Mitochondrial ribosomal proteins (MRPs) are required for the translation of all 13 mitochondrial encoded genes in humans. It has been speculated that mutations and polymorphisms in the human MRPs may be a primary cause of some oxidative phosphorylation disorders or modulate the severity and tissue specificity of pathogenic mitochondrial DNA mutations. Although the sequences of most of the yeast MRPs are known, only very few mammalian and nearly no human MRPs have been completely characterized. MRPs differ greatly in sequence, and sometimes biochemical properties, between different species, not allowing easy recognition by sequence homology. Therefore, the Mammalian Mitochondrial Ribosomal Consortium is using a direct approach of purifying individual mammalian (bovine) MRPs, determining their N-terminal and/or internal peptide sequences using different protein sequencing techniques, and using the resulting sequence information for screening expressed sequence tags and genomic data bases to determine human, mouse, and rat homologues of the bovine proteins. Two proteins of the large and three proteins of the small ribosomal subunit have been analyzed in this manner. Three of them represent "new," i.e. formerly unknown mammalian mitochondrial ribosomal protein classes. Only one of these three different MRPs shows significant sequence similarities to known ribosomal proteins. In one case, the corresponding human genomic DNA sequences were found in the data bases, and the exon/intron structure was determined.

Heart-specific splice-variant of a human mitochondrial ribosomal protein (mRNA processing; tissue specific splicing).

It has been proposed that splice-variants of proteins involved in mitochondrial RNA processing and translation may be involved in the tissue specificity of mitochondrial DNA disease mutations (Fischel-Ghodsian, 1998. Mol. Genet. Metab. 65, 97-104). To identify and characterize the structural components of mitochondrial RNA processing and translation, the Mammalian Mitochondrial Ribosomal Consortium has been formed. The 338 amino acid (aa) residues long MRP-L5 was identified (O'Brien et al., 1999. J. Biol. Chem. 274, 36043-36051), and its transcript was screened for tissue specific splice-variants. Screening of the EST databases revealed a single putative splice-variant, due to the insertion of an exon consisting of 89 nucleotides prior to the last exon. Screening of multiple cDNA libraries revealed this inserted exon to be present only in heart tissue, in addition to the predominant MRP-L5 transcript. Sequencing of this region confirmed the EST sequence, and showed in the splice-variant a termination triplet at the beginning of the last exon. Thus the inserted exon replaces the coding sequence of the regular last exon, and creates a new 353 aa long protein (MRP-L5V1). Sequence analysis and 3D modeling reveal similarity between MRP-L5 and threonyl-t-RNA synthetases, and a likely RNA binding site within MRP-L5, with the C-terminus in proximity to the RNA binding site. Sequence analysis of MRP-L5V1 also suggests a likely transmembrane domain at the C-terminus. Thus it is possible that the MRP-L5V1 C-terminus could interfere with RNA binding and may have gained a transmembrane domain. Further studies will be required to elucidate the functional significance of MRP-L5V1.

Mammalian mitochondrial ribosomal proteins (2). Amino acid sequencing, characterization, and identification of corresponding gene sequences.

Four different classes of mammalian mitochondrial ribosomal proteins were identified and characterized. Mature proteins were purified from bovine liver and subjected to N-terminal or matrix-assisted laser-desorption mass spectroscopic amino acid sequencing after tryptic in-gel digestion and high pressure liquid chromatography separation of the resulting peptides. Peptide sequences obtained were used to virtually screen expressed sequence tag data bases from human, mouse, and rat. Consensus cDNAs were assembled in silico from various expressed sequence tag sequences identified. Deduced mammalian protein sequences were characterized and compared with ribosomal protein sequences of Escherichia coli and yeast mitochondria. Significant sequence similarities to ribosomal proteins of other sources were detected for three out of four different mammalian protein classes determined. However, the sequence conservation between mitochondrial ribosomal proteins of mammalian and yeast origin is much less than the sequence conservation between cytoplasmic ribosomal proteins of the same species. In particular, this is shown for the mammalian counterparts of the E. coli EcoL2 ribosomal protein (MRP-L14), that do not conserve the specific and functional highly important His(229) residue of E. coli and the corresponding yeast mitochondrial Rml2p.

Identification of mammalian mitochondrial ribosomal proteins (MRPs) by N-terminal sequencing of purified bovine MRPs and comparison to data bank sequences: the large subribosomal particle.

Bovine mitochondrial ribosomes are presented as a model system for mammalian mitochondrial ribosomes. An alternative system for identifying individual bovine mitochondrial ribosomal proteins (MRPs) by RP-HPLC is described. To identify and to characterize individual MRPs proteins were purified from bovine liver, separated by RP-HPLC, and identified by 2D PAGE techniques and immunoblotting. Molecular masses of individual MRPs were determined. Selected proteins were subjected to N-terminal amino acid sequencing. The peptide sequences obtained were used to screen different databases to identify several corresponding MRP sequences from human, mouse, rat, and yeast. Signal sequences for mitochondrial import were postulated by comparison of the bovine mature N-termini determined by amino acid sequencing with the deduced mammalian MRP sequences. Significant sequence similarities of these new MRPs to known r-proteins from other sources, e.g., E. coli, were detected only for two of the four MRP families presented. This finding suggests that mammalian mitochondrial ribosomes contain several novel proteins. Amino acid sequence information for all of the bovine MRPs will prove invaluable for assigning functions to their genes, which would otherwise remain unknown.

Site-directed hydroxyl radical probing of 30S ribosomal subunits by using Fe(II) tethered to an interruption in the 16S rRNA chain.

Two in vitro transcripts, one corresponding to the 5' and central domains (residues 1-920) of 16S rRNA and the other corresponding to its 3' domain (residues 922-1542), assemble efficiently in trans with 30S ribosomal proteins to form a compact ribonucleoprotein particle that cosediments with natural 30S subunits. Isolated particles are similar in appearance to natural 30S subunits with electron microscopy and contain a full complement of the small subunit ribosomal proteins. The particles have a reduced ability to bind tRNA (attributable to the location of the discontinuity in a conserved region of the rRNA) near features that have been implicated in tRNA binding. Association of these two halves of 16S rRNA in trans must be stabilized by either previously unidentified RNA-RNA contacts or interactions mediated by ribosomal proteins because there are no known direct interactions between them. The trans construct was used to probe the three-dimensional RNA neighborhood around position 922 of 16S rRNA by generating hydroxyl radicals from Fe(II) tethered to the 5' end of the 3' transcript. Hydroxyl radical-induced cuts in the 16S rRNA chain were localized by primer extension to nucleotides 923-929 and 1192-1198, providing evidence for the mutual proximity of the 920 and 1192 regions.

Independent in vitro assembly of a ribonucleoprotein particle containing the 3' domain of 16S rRNA.

Small (30S) subunits of Escherichia coli ribosomes are composed of 21 proteins and a 1542-nucleotide 16S rRNA, whose secondary structure is divided into three domains. An in vitro transcript of the 3' domain of 16S rRNA (residues 923-1542), assembles efficiently with 30S ribosomal proteins to form a compact ribonucleoprotein (RNP) particle. Isolated particles examined under the electron microscope have a globular appearance, similar in size and shape to the head of the 30S ribosomal subunit. Two-dimensional gel analysis of the particles indicates the presence of proteins S3, S7, S9, S10, S13, S14, and S19 and smaller amounts of S2, all of which have been localized to the head of the 30S subunit by immunoelectron microscopy and neutron diffraction and belong to the S7 assembly family. Interestingly, protein S4, which is believed to interact exclusively with the 5' domain, is also reproducibly found associated with the particles in significant amounts. Chemical probing of the RNA in the assembled particle reveals characteristic cleavage protection patterns, showing that the proteins assemble with the 3'-domain RNA similarly to the way in which they assemble with 16S rRNA, although some of the later steps of assembly appear to be incomplete. These results show that the 3' domain of 16S rRNA can indeed assemble independently of the rest of the 30S subunit into a particle that resembles its structure in the ribosome. In addition, the assembled particles are able to bind spectinomycin with an affinity comparable to that of 30S subunits.

Phosphorylation of rabbit reticulocyte guanine nucleotide exchange factor in vivo. Identification of putative casein kinase II phosphorylation sites.

The guanine nucleotide exchange factor (GEF) is a multi-subunit protein which catalyzes the exchange of GDP for GTP in eukaryotic chain initiation factor 2. Phosphorylation of the 82-kDa subunit of GEF in vitro by casein kinase II (CK-II) is associated with a 5-fold increase in nucleotide exchange activity. However, phosphorylation of GEF in vivo has not been studied, and the kinase(s) that phosphorylate GEF have not been identified. The 82-kDa subunit of GEF was partially sequenced, and a synthetic peptide was used to generate polyclonal anti-peptide antibodies that react specifically with this subunit. To examine the phosphorylation of GEF in intact cells, the protein was isolated and purified extensively from metabolically 32P-labeled rabbit reticulocytes. Only the 82-kDa subunit was found to be phosphorylated, and on Western blots the anti-peptide antisera reacted specifically with the labeled subunit. Phosphoamino acid analysis indicated that phosphorylation occurred exclusively on Ser residues. Digestion with cyanogen bromide of in vivo labeled protein and GEF phosphorylated in vitro by CK-II produced comparable phosphopeptide maps. However, additional phosphopeptide bands were also observed with GEF derived from intact cells. Sequence analysis obtained by Edman degradation of the phosphopeptides was compared with the deduced amino acid sequence of a cloned 82-kDa subunit of GEF [Bushman, J. L., Asuru, A. I., Matts, R. L., & Hinnenbusch, A. G. (1993) Mol. Cell. Biol. 13, 1920-1932]. Putative sites of phosphorylation were identified at Ser 703 and/or 704, which contain the sequence S(P)XXD, a CK-II consensus recognition motif.(ABSTRACT TRUNCATED AT 250 WORDS)

Proteins of mammalian mitochondrial ribosomes.

The bovine mitochondrial system is being developed as a model system for studies on mammalian mitochondrial ribosomes. Information is emerging on the structural organization and RNA binding properties of proteins in these mitochondrial ribosomes. Unexpectedly, these ribosomes appear to interact directly with GTP, via a high affinity binding site on the small subunit. Despite major differences in their RNA content and physical properties, mammalian mitochondrial and cytoplasmic ribosomes contain about the same number of proteins. The proteins in each kind of ribosome have a similar size distribution, and both sets are entirely coded by nuclear genes, raising the possibility that these different ribosomes may contain the same set of proteins. Comparison of bovine mitochondrial and cytoplasmic r-proteins by co-electrophoresis in two-dimensional gels reveals that most of the cytoplasmic ribosomal proteins are more basic than the mitochondrial ribosomal proteins, and that none are co-migratory with mitochondrial ribosomal proteins, suggesting that the proteins in the two ribosomes are different. To exclude the possibility that the electrophoretic differences result only from post-translational modification of otherwise identical proteins, antibodies against several proteins from the large subunit of bovine mitochondrial ribosomes were tested against cytoplasmic ribosomes by solid phase radioimmunoassay and against cytoplasmic ribosomal proteins on Western blots. The lack of cross-reaction of these antibodies with cytoplasmic r-proteins suggests that mitochondrial ribosomal proteins have different primary structures and thus are most likely encoded by a separate set of nuclear genes.

Bovine mitochondrial ribosomes possess a high affinity binding site for guanine nucleotides.

Mammalian mitochondrial ribosomes possess a binding site for guanine nucleotides. GTP binds in unit stoichiometry and with high affinity (Kd = 15.3 +/- 2.8 nM) to the small subunit of bovine mitochondrial ribosomes. This binding activity survives high salt washes, indicating that the nucleotide binds to an integral site within this subunit. GDP also binds to the small subunit with high affinity (Kd = 17 +/- 5.8 nm) and in unit stoichiometry. The GTP binding activity can be competed with GDP but not appreciably by other nucleotides, indicating that both GTP and GDP bind specifically and to the same site. The non-hydrolyzable analogs of GTP, guanylyl-5'-imidophosphate, and guanylyl-(beta,gamma-methylene)- diphosphonate also bind to the small subunit, but with reduced affinity. These results indicate that mammalian mitochondrial ribosomes, unlike other ribosomes, are able to interact directly with guanosine triphosphate, suggesting that the bound GTP may be involved in a novel regulatory mechanism in mitochondrial protein synthesis.

The translation system of mammalian mitochondria.

Oligoribonucleotides and mRNA were used to define properties of the bovine mitoribosomal mRNA binding site. The RNA binding domain on the 28 S subunit spans approx. 80 nucleotides of the template, based on ribosome protection experiments, but the major interaction with the ribosome occurs over a 30 nucleotide stretch. The binding site for E. coli IF3 is conserved in bovine mitoribosomes, but mitochondrial factors appear essential for proper interaction of mRNA with mitoribosomes. The small subunit of bovine mitoribosomes contains a high-affinity binding site for guanyl nucleotides, further indication of specialized mechanisms for initiation complex formation and function of mammalian mitochondrial ribosomes.

Mechanism of mRNA binding to bovine mitochondrial ribosomes.

The binding of mRNA to bovine mitochondrial ribosomes was investigated using triplet codons, homopolymers and heteropolymers of various lengths, and human mitochondrial mRNAs. In the absence of initiation factors and initiator tRNA, mitochondrial ribosomes do not bind triplet codons (AUG and UUU) or homopolymers (oligo(U] shorter than about 10 nucleotides. The RNA binding domain on the 28 S mitoribosomal subunit spans approximately 80 nucleotides of the mRNA, judging from the size of the fragments of poly(U,G) and natural mRNAs protected from RNase T1 digestion by this subunit, but the major binding interaction with the ribosome appears to occur over a 30-nucleotide stretch. Human mitochondrial mRNAs coding for subunits II and III of cytochrome c oxidase and subunit 1 of the NADH-ubiquinone oxidoreductase (complex I) were used in studying in detail the binding of mRNA to the small subunit of bovine mitochondrial ribosomes. We have determined that these mRNAs have considerable secondary structure in their 5'-terminal regions and that the initiation codon of each mRNA is sequestered in a stem structure. Little mRNA was bound to ribosomes in a manner conferring protection of the 5' termini from RNase T1 digestion, under standard conditions supporting the binding of artificial templates, but such binding was greatly stimulated by the addition of a mitochondrial extract. Initiation factors and tRNAs from Escherichia coli were unable to stimulate the 5' terminus protected binding of these mRNA molecules, demonstrating a requirement for homologous factors. Our results strongly suggest that mitochondrial initiation factors are required for the proper recognition and melting of the secondary structure in the 5'-terminal region of mitochondrial mRNAs, as a prerequisite for initiation of protein synthesis in mammalian mitochondria.

Interaction of bovine mitochondrial ribosomes with Escherichia coli initiation factor 3 (IF3).

Mammalian mitochondrial ribosomes are distinguished from their bacterial and eukaryotic-cytoplasmic counterparts, as well as from mitochondrial ribosomes of lower eukaryotes, by their physical and chemical properties and their high protein content. However, they do share more functional homologies with bacterial ribosomes than with cytoplasmic ribosomes. To search for possible homologies between mammalian mitochondrial ribosomes and bacterial ribosomes at the level of initiation factor binding sites, we studied the interaction of Escherichia coli initiation factor 3 (IF3) with bovine mitochondrial ribosomes. Bacterial IF3 was found to bind to the small subunit of bovine mitochondrial ribosomes with an affinity of the same order of magnitude as that for bacterial ribosomes, suggesting that most of the functional groups contributing to the IF3 binding site in bacterial ribosomes are conserved in mitochondrial ribosomes. Increasing ionic strength affects binding to both ribosomes similarly and suggests a large electrostatic contribution to the reaction. Furthermore, bacterial IF3 inhibits the Mg2+-dependent association of mitochondrial ribosomal subunits, suggesting that the bacterial IF3 binds to mitochondrial small subunits in a functional way.

RNA binding proteins of the large subunit of bovine mitochondrial ribosomes.

RNA binding properties of proteins from the large subunit of bovine mitochondrial ribosomes were studied using four different approaches: binding of radiolabeled RNA to western blotted proteins; disassembly of the intact 39 S ribosomal subunits with urea; binding of ribosomal proteins to RNA in the presence of urea; and binding of proteins extracted with lithium chloride to RNA. Results from these four approaches allowed us to identify a set of six proteins (L7, L13, L14, L21, L26, and L44) which appear to be strong RNA binding proteins. Seven additional proteins (L8, L11, L28, L35, L40, L49, and L50) were identified as secondary RNA binding proteins. RNA binding properties of the proteins in both of these sets were compared with the topographic disposition and susceptibility towards lithium chloride extraction of the individual proteins. Proteins from the first set are good candidates for early assembly proteins since they have a high affinity for RNA, are generally found in 4M lithium chloride core particles, and are among the most buried proteins in the 39 S subunit.

Production of monospecific antibodies to a low-abundance hepatic membrane protein using nitrocellulose immobilized protein as antigen.

Membrane proteins from primary cultures of rat hepatocytes were separated by two-dimensional polyacrylamide gel electrophoresis. The proteins were transferred to nitrocellulose paper which was then dissolved in dimethyl sulfoxide and this mixture was used as a primary immunogen in rabbits. Subsequent immunizations were performed using nonsolubilized protein immobilized on nitrocellulose paper. A monospecific polyclonal antibody was generated against a specific mitochondrial membrane protein (MP-73) for which de novo synthesis appeared to be induced by amino acid starvation of the hepatocytes. A minimum of 15-20 micrograms of protein antigen was required to elicit significant antibody production. Serum antibody titer was sufficient to allow detection of MP-73 at a serum dilution of 1:2000.

Evolution of proteins in mammalian cytoplasmic and mitochondrial ribosomes.

The proteins of cytoplasmic and mitochondrial ribosomes from the cow and the rat were analyzed by co-electrophoresis in two dimensional polyacrylamide gels to determine their relative evolutionary rates. In a pairwise comparison of individual ribosomal proteins (r-proteins) from the cow and the rat, over 85% of the cytoplasmic r-proteins have conserved electrophoretic properties in this system, while only 15% of the proteins of mitochondrial ribosomes from these animals fell into this category. These values predict that mammalian mitochondrial r-proteins are evolving about 13 times more rapidly than cytoplasmic r-proteins. Based on actual evolutionary rates for representative cytoplasmic r-proteins, this mitochondrial r-protein evolutionary rate corresponds to an amino acid substitution rate of 40 X 10(-10) per site per year, placing mitochondrial r-proteins in the category of rapidly evolving proteins. The mitochondrial r-proteins are apparently evolving at a rate comparable to that of the mitochondrial rRNA, suggesting that functional constraints act more or less equally on both kinds of molecules in the ribosome. It is significant that mammalian mitochondrial r-proteins are evolving more rapidly than cytoplasmic r-proteins in the same cell, since both sets of r-proteins are encoded by nuclear genes. Such a difference in evolutionary rates implies that the functional constraints operating on ribosomes are somewhat relaxed for mitochondrial ribosomes.

Site of synthesis of the proteins of mammalian mitochondrial ribosomes. Evidence from cultured bovine cells.

In order to determine the sites of synthesis of the proteins of the mammalian mitochondrial ribosome (mitoribosome), bovine (MDBK) cells were labeled with [35S]methionine in the presence of inhibitors of mitochondrial and cytoplasmic protein synthesis. Labeling in the absence of cytoplasmic protein synthesis produced a "blank" fluorogram, indicating that there is no mitochondrial product. Additionally, incorporation of [35S]methionine into the enumerated mitoribosomal proteins continued in the absence of mitochondrial protein synthesis. Finally, it was demonstrated that mitoribosomal proteins can be both translated and assembled into complete mitoribosomes in the absence of mitochondrial protein synthesis. These results indicate that in mammals, as opposed to lower eukaryotes, all of the mitoribosomal proteins are products of cytoplasmic protein synthesis.

Organization of proteins in mammalian mitochondrial ribosomes. Accessibility to lactoperoxidase-catalyzed radioiodination.

To assess the relative exposure of individual ribosomal proteins (r-proteins) in the large and small subunits of the bovine mitochondrial ribosome, we used a double label iodination technique. Regions of r-proteins exposed in purified ribosomal subunits were labeled with 131I using the lactoperoxidase-catalyzed iodination system, and additional reactive groups available upon denaturing the r-proteins in urea were labeled with 125I using the chloramine-T mediated reaction. The ratio of 131I to 125I incorporated into individual proteins under these conditions is representative of the degree of exposure for each of the proteins in the subunits. In this manner, the r-proteins have been grouped into 3 classes depending on their degree of exposure: high exposure, intermediate exposure, and essentially buried. While both subunits have a few proteins in the "highly exposed" group, and a large number of proteins in the "intermediate exposure" group, only the large ribosomal subunit has an appreciable number of proteins which appear essentially buried. The more buried proteins may serve mainly structural roles, perhaps acting as "assembly proteins," since many from this group bind to ribosomal RNA. The more superficially disposed proteins may comprise binding sites for macromolecules that interact with ribosomes during protein synthesis, as well as stabilizing the association of the large and small subribosomal particles.

Reconstitution of 50 S ribosomal subunits from Bacillus stearothermophilus with 5 S RNA from spinach chloroplasts and low-Mr RNA from mitochondria of Locusta migratoria and bovine liver.

Reconstitution experiments with 50 S ribosomal subunits from Bacillus stearothermophilus demonstrate that spinach chloroplast 5 S rRNA can be incorporated into the bacterial ribosome and yield biologically active particles, thereby establishing the eubacterial nature of chloroplast 5 S rRNA. In contrast, mitochondria from Locusta migratoria or bovine liver do not appear to contain discrete, low-Mr RNAs, which can replace 5 S rRNA in the functional reconstitution of B. stearothermophilus ribosomes.

Protein composition of the bovine mitochondrial ribosome.

The protein complement of the bovine mitochondrial ribosome has been analyzed by two-dimensional electrophoresis in polyacrylamide gels to determine the number and molecular weights of the ribosomal proteins. Salt-washed ribosomal subunits are found to contain a total of 85 ribosomal proteins, 84 of which are electrophoretically distinct between the two subunits. These proteins are also electrophoretically distinguished from those of cytoplasmic ribosomes. This large number of proteins does not appear to be due to contamination by cytoplasmic ribosomal proteins or by adherent nonribosomal proteins. The molecular weights of these proteins are considerably larger than those of Escherichia coli ribosomal proteins, and are similar to those of bovine cytoplasmic ribosomal proteins. The sum of the molecular weights of the 85 proteins agrees well with that predicted by physical chemical measurements of the total mass of protein in the two subunits. Bovine mitochondrial ribosomes thus contain about twice as much protein as RNA, a highly unusual composition in comparison to the other kinds of ribosomes which have been characterized to date. In addition, it appears that the ribosomal proteins themselves are less basic than the proteins of most other ribosomes.