Spatial arrangement of ribosomal proteins: reaction of the Escherichia coli 30S subunit with bis-imidoesters

A modular cross-linking approach for exploring protein interactions

A method is described for the elucidation of protein-protein interactions using novel cross-linking reagents and mass spectrometry. The method incorporates (1) a modular solid-phase synthetic strategy for generating the cross-linking reagents, (2) enrichment and digestion of cross-linked proteins using microconcentrators, (3) mass spectrometric analysis of cross-linked peptides, and (4) comprehensive computational analysis of the cross-linking data. This integrated approach has been applied to the study of cross-linking between the components of the heterodimeric protein complex negative cofactor 2.

Electron microscopy study of the aminoacyl-tRNA synthetase multienzymatic complex purified from rabbit reticulocytes

The morphology of the high molecular weight complex of aminoacyl-tRNA synthetases purified from rabbit reticulocytes has been investigated by electron microscopy. To stabilize it against dissociation, the complex was also studied after chemical crosslinking. Freeze fracture, drying shadowing and negative staining were used. The reticulocyte complex appears as a moderately elongated object with no simple compact shape. Upon rapid drying, the native complex dissociates and shows the presence of approximately 8 globular components, the individual size of which is 80-100 A. The surface of the cross-linked complex shows several distinct globules which appear to extend out of a central core. The irregularly shaped crosslinked complex has a maximal dimension of 350 +/- 50 A. The morphology of the synthetase complex is discussed with respect to some of the properties of this type of multienzymatic system.

Can multiple low-density immunoadsorbents and 'negative' distance matrices help predict the three-dimensional structure of any globular protein?

Chemical cross-linking of a protein immobilized on multiple, site-specific, monoclonal, chemically modified antibody matrices combined with analysis of data on a distance matrix is likely to yield sufficient positive and 'negative' long-range distance information to help predict the three dimensional structure of any single-chain globular protein.

Conformational analysis of serum apolipoprotein AII in lipoprotein complexes with bifunctional crosslinking reagents

Apolipoprotein AII isolated from human serum high density lipoproteins was recombined with phosphatidylcholine to yield homogeneous particles of 80-120 A diameter. The radioactive bifunctional crosslinkers dimethyl [1,1'-14C]suberimidate and dimethyl 4,4'-dithiobis([1-14C]butyrimidate) were reacted with these particles. The kinetics of the reactions and the localisation of the crosslinked lysines of the polypeptide chains were determined. Thermolysin hydrolysis followed by two-dimensional separation of the peptides and isolation of the mono- and bifunctionally modified peptides allowed the assignment of the crosslinked peptides of the apoprotein AII seqeunce. The crosslinking pattern indicates a close-neighbour relationship (13-15 A) of the peptide chains between amino acid residues 3, 23, 46 and 55 of the symmetrical halves of the apo AII molecule. A reconstruction of the secondary structure of Apo AII in the lipoprotein complex on the basis of theoretical calculations is given and correlated with the chemical data.

Chemical Cross-linking of Neighboring Thylakoid Membrane Polypeptides

Cross-linking between protein components of whole spinach (Spinacia oleracea var. Nobel) thylakoids and of photosystem I- and II-enriched thylakoid fractions has been produced by reaction with the bifunctional imidoester dimethyl-3,3'-dithiobispropionimidate dihydrochloride as well as by the oxidation of intrinsic sulfydryl groups with an orthophenanthrolinecupric ion complex. The mixture of membrane proteins and their cross-linked products has been analyzed by two-dimensional sodium dodecyl sulfate electrophoresis, with a reductive cleavage step of the cross-linkages before the second dimension. Cross-linked aggregates up to a molecular weight of about 130 kilodaltons (kD) were analyzed, and it was inferred that the polypeptides appearing together in the same aggregates were neighbors within the membrane.In thylakoids as well as in isolated photosystem fractions, oligomers were formed by cross-linking polypeptides of the 60 to 90 kD range, among them the polypeptides of the chlorophyll-protein complex I. Polypeptides of 46, 19, and 12 kD were cross-linked to these complexes. Polypeptides of 25 and 22 kD, which are related to the chlorophyll-protein complex II, were cross-linked in thylakoids as well as in photosystem II fractions, suggesting that in the membrane these molecules are close together. In photosystem II fractions an oligomer having a molecular weight of about 60 kD was formed by cross-linking several polypeptides of different molecular weights: 40, 25, and 22 kD.Our cross-linking experiments show that protein interactions in the thylakoid membrane occurred mainly among the polypeptides of the two chlorophyll-protein complexes, thus suggesting an oligomeric nature of these apoproteins.

Synthesis and use of bifunctional chloromethylalkanedione derivatives of variable chain length for cross-linking thiol groups in oligomeric proteins. Specific cross-linking in glyceraldehyde 3-phosphate dehydrogenase

Bischloromethylpentanedione, bischloromethylhexanedione, bischloromethyloctanedione and bischloromethyldecanedione were synthesized from their corresponding dicarboxylic acids via the bis-acyl chloride and the bisdiazomethylketone derivatives. These compounds proved to be highly specific cross-linking reagents for rabbit skeletal-muscle glyceraldehyde 3-phosphate dehydrogenase. Incubation of the enzyme with cross-linking reagents resulted in both a time- and concentration-dependent formation of covalently linked oligomeric structures. The major cross-linked product detected by sodium dodecyl sulphate/polyacrylamide-gel electrophoresis was the dimer (mol. wt. 72000). Sepharose 6B chromatography of the cross-linked enzyme showed that it still existed as the tetramer. Cross-linking was dependent on the native structure of the enzyme, since it was abolished on denaturation of the enzyme. The actual covalently linked product depends on the conditions of modification and the chain length of the reagent. The maximum yield of dimer (70-80%) was obtained with bischloromethylhexanedione, and the yield decreased with either shorter- or longer-chain compounds. The calculated distance between the two reactive points in bischloromethylhexanedione is 1.21-1.45nm. Bischloromethylhexanedione modified at least two thiol groups per monomer. Modification of the active-site thiol, cysteine-149, was not essential for cross-linking, since glyceraldehyde 3-phosphate dehydrogenase carboxymethylated on cysteine-149 still reacted to form the dimer. The rate of chemical cross-linking was markedly decreased by increasing the NAD(+) occupancy of the enzyme active sites. These experiments are discussed in terms of the asymmetry of the enzyme structure in solution.

Studies of glutamate dehydrogenase. Analysis of quaternary structure and contact areas between the polypeptide chains

Cross-linking of the unimer of glutamate dehydrogenase from beef liver (consisting of six polypeptide chains each having a molecular weight of 56000) with dimethyladipimidate and subsequent analysis by sodium dodecylsulfate electrophoresis shows predominantly the trimeric species (molecular weight 168000). Treatment with dimethylimidates of other chain length yields significantly less trimeric species indicating that the amino groups being cross-linked are within a distance of about 0.85 nm. Comparison of the molar amount of incorporated [14C]dimethyladipimidate with the number of modified amino groups (determined with trinitrobenzenesulfonic acid) shows that although 8-9 of the 34 amino groups have reacted, only 2-3 of them are involved in cross-links. Reaction with dimethylimidates inactivates the enzyme. The loss of the activity is partly concomitant to cross-linking to the trimeric species and not simply due to the modification of essential lysine residues. This is supported by the fact that, although more lysine residues react with mono-functional methylimidates, the loss of activity is reduced. Purified chymotryptic and tryptic peptides of the radioactive-labeled trimeric species were subjected to sequence analysis. Six peptides containing 75% of the total label were identified: one involves the amino-terminal residue alanine-1 and the others involve lysine-105, lysine-154, lysine-269, lysine-358 and lysine-399. Quantitative analysis of the specific radioactivity of each peptide/mol lysine leads to the conclusion that only lysine-105, lysine-154, lysine-269 and lysine-358 participate in cross-links, lysine-269 and lysine-358, respectively, being at isologous and lysine-105 cross-linked with lysine-154 at heterologous contact domains of the enzyme. A model for the planar arrangement of the trimeric species in the quaternary structure of glutamate dehydrogenase is discussed. It includes both isologous and heterologous contact areas between the polypeptide chains.

Near-neighbor relationships of the subunits of cytochrome c oxidase

Cytochrome c oxidase in detergent dispersion has been cross-linked with two reversible cross-linking agents, dithiobissuccinimidylpropionate (DSP) and dimethyl-3,3'-dithiobispropionimidate (DTBP), and the cross-linked products formed have been analyzed by two-dimensional gel electrophoresis. Under mild reaction conditions, several subunit pairs were seen including II and V, V and VII, IV and VI. With higher levels of DSP, larger aggregates were seen until a cross-linked product with an apparent molecular weight of 140 000 was the predominant band on gels. This is the smallest molecular weight aggregate to contain all seven subunits of the enzyme and most likely represents the "unit" or two heme and two copper containing complex of cytochrome c oxidase.

Single major polypeptide of a calicivirus: characterization by polyacrylamide gel electrophoresis and stabilization of virions by cross-linking with dimethyl suberimidate

A calicivirus, San Miguel sea lion virus serotype 4, isolate 15FT, externally labelled with 125I, was shown by gel electrophoresis to possess a single major polypeptide. The polypeptide migrated anomalously upon electrophoresis in two sodium dodecyl sulfate (SDS) systems: more slowly than bovine serum albumin in a continuous phosphate-buffered system and more rapidly than bovine serum albumin in a discontinuous system. Estimated molecular weights in the two systems were approximately 71,000 and 64,000, respectively. There was no clear evidence for a minor virion polypeptide. Treatment of purified San Miguel sea lion virions with dimethyl suberimidate, a cross-linking reagent, preserved virion integrity during long-term storage at 4 degrees C. Oligomeric species of the polypeptide were observed upon electrophoresis of products from cross-linked virions. Based upon a preferred polypeptide molecular weight estimate of 71,000 and distribution of oligomeric species, a calicivirion model with 120 monomeric protein units is proposed as an alternative to a 180-unit model.

The use of a cleavable crosslinking reagent to identify neighboring proteins in the 30-S ribosomal subunit of Escherichia coli

A cleavable bifunctional reagent, dimethyl 3,3'-dithiobispropionimidate, has been used to crosslink proteins that occupy neighboring positions in the 30-S ribosomal subunit of Escherichia coli. The crosslinked proteins were identified, fully or partly, by their positions in two two-dimensional gel electrophoretic systems, one diagonal and the other quasi-diagonal, in which the complexes were cleaved after the first-dimensional run. It was found to be necessary to block the protein sulfhydryl groups in order to prevent artifactual disulfide crosslinking after extraction of the protein from ribosome. Eleven crosslinked complexes were detected. Four were fully identified: the triplet S4-S5-S8, and the pairs S2-S3, S4-S5, and S5-S8. In five others one component was identified unambiguously. No additional complexes were seen when the longer homologous butyro and capro reagents were used.

Synthesis of a cleavable protein-crosslinking reagent for the investigation of ribosome structure

This communication describes a simple method for synthesizing cleavable bifunctional imido esters of different chain lengths. These reagents, which form covalent crosslinks between lysine residues of proteins, contain a disulfide bond which is cleaved under mild conditions by reducing agents such as 2-mercaptoethanol. The reagents are synthesized via the dithiobisnitrile which is prepared in high yield by reacting the appropriate omega-activated nitrile with sodium polysulfide and is then converted quantitatively to the diimidate. Three such reagents were prepared: dimethyl 3.3'-dithiobispropionimidate, dimethyl 4,4'-dithiobisbutyrimidate, and dimethyl 6-6'-dithiobiscaproimidate. The first was synthesized from acrylonitrile, and the others from the appropriate omega-bromonitriles. Experiments with the bispropionimidate and a test protein, pancreatic ribonuclease, have shown the reagent to be effective in producing multimeric crosslinked complexes, from which monomeric proteins can recovered after treatment with 2-mercaptoethanol. The reagents are suitable for studies of ribosomal structure.

Physical properties of some ribosomal proteins in solution and evidence for molecular interactions between isolated ribosomal proteins

Many previous studies have been directed toward obtaining a physical visualization of the relationship between the protein and RNA in the ribosomal subunits isolated from Escherichia coli. The current study is the first report where an attempt has been made to directly assess interactions between a pair of isolated ribosomal proteins separate from the intact system by means of sedimentation equilibrium analysis. The molecular weights of the proteins S3, S4, S5, S6, S7, S8, and S20 from the 30S subunit of the E. coli ribosome were determined under conditions of assembly of the subunit by sedimentation equilibrium. All of the proteins exhibited molecular weights consistent with monomeric behavior (i.e., in agreement with the measurement of the ultimate molecular weight in denaturing solvents as reported in other studies as well as in the current study) except S8 which indicates a tendency to self-associate. Hydrodynamic measurements on the proteins indicate that these proteins are not completely disorganized in solution such as a random coil, although not as compact as globular proteins. The frictional coefficient ratios found for these ribosomal proteins range from 1.4 to 1.9. The hydrodynamic data are discussed as containing some evidence that stable interaction sites could exist in the proteins. The molecular weight data are considered pertinent to a sedimentation equilibrium study of protein-protein interactions that may be occurring in the ribosomal subunits. Two proteins, S3 and S5, considered in this investigation were found to exhibit no tendency to self-associate under conditions of reassembly. When the two proteins are mixed under those same conditions, however, a species with a molecular weight greater than that of either S3 or S5 is observed to be formed. The interpretation is presented that a molecular interaction between S3 and S5 is the cause. The system is described as containing S3, S5, and a complex between S3 and S5 with a stoichiometry of 1:1 and an association equilibrium constant of 5.7 times 10-5 l./mol (delta G-o equals minus 7.25 kcal/mol). Since the association appears to be specific and of moderate strength, it is concluded that the interaction could have some pertinence with respect to conferring a structural arrangement in the ribosomal subunit. Moreover, it is concluded that protein-protein interactions, in general, must be considered in addition to the well documented significant RNA-protein relationships when models for ribosome structure and assembly are formulated.

The protein topography of the E. coli 30S ribosomal subunit: a preliminary model E. coli/30S subunit/ribosome/spatial arrangement of proteins

A three dimensional model is presented which shows the apatial arrangement of 20 of the 21 proteins of the 30S ribosomal subunit of Escherichia coli. The model fulfills several purposes: (a) It summarizes currently available structural and functional data on ribosomal proteins; (b) It suggests an interesting correlation between stoichiometry and function. Functional proteins are both clustered and fractional; (c) It can be evaluated in relationships or point out critical experiments by which it can be tested.

Structure of isometric viruses containing nonidentical polypeptide chains

The theory of Caspar and Klug (1962) for the structure of isometric viruses has been generalized to the case in which the identical repeating unit is composed of n nonidentical polypeptide chains. This modified theory accounts for the structure of picornaviruses, the lambda phage head, cowpea mosaic virus, and phiX174, while at the same time conserving the principle of having identical subunits in identical environments. Furthermore, the modified theory suggests amending the triangulation number to T[n] for capsids with n nonidentical polypeptide chains as the repeating unit.

Location of the spike glycoproteins in the Semliki Forest virus membrane

Labeling experiments with formyl-[(35)S]-methionyl sulfone methylphosphate and crosslinking studies with dimethylsuberimidate suggest that the spike glycoproteins of Semliki Forest virus extend through the viral membrane into close contact with the nucleocapsid. Based on this finding, we present a mechanism for the formation of virus-specific patches in the host cell plasma membrane during virus assembly.

Proteins at the tRNA binding sites of Escherichia coli ribosomes

p - Nitrophenylcarbamyl - phenylalanyl-tRNA binds to ribosomes in response to poly(U) and reacts with ribosomal proteins via covalent bond formation. Ribosomal proteins of Escherichia coli were characterized by two-dimensional gel electrophoresis and by reaction with specific antibodies. The affinity label is shown to react with the 50S proteins L27, L15, L2, L16, and L14. We therefore conclude that these particular proteins are located at or near tRNA-binding sites within the 50S ribosomal subunit.

Assembly of bacterial ribosomes

I have not mentioned the remarkable progress made mainly by Fellner and his co-workers (86) in the elucidation of the primary structure of rRNA's and by Wittmann and his co-workers (87) in determining the structure of several ribosomal proteins. Such knowledge of primary structures is certainly the basis of complete understanding of the structure of the ribosome. With the current progress in technology, complete elucidation of the primary structure of all the ribosomal components is probably a matter of time. As indicated in this article, a rough approximation of the three-dimensional structure of ribosomes is likely to emerge soon. Although not mentioned in this article, studies of ribosomes from higher organisms are also progressing. We must, therefore, consider what further studies should be conducted and what kinds of questions we would like to solve. Some groups of investigators aim to elucidate the complete three-dimensional structure of ribosomes and to find out how these complex cell organelles function; they hope to determine the conformational changes of many of the component molecules within the ribosome structure in response to external macromolecules and cofactors engaged in protein synthesis. Such knowledge will also be important in enabling us to understand the regulation of translation of genetic messages. Other groups of investigators aim to elucidate the complex series of events which originate in the transcription of the more than 60 genes and culminate in the formation of the specific structure of the organelle. Complete reproduction in vitro of all the assembly events that occur in vivo should not be difficult to achieve in principle. It should then become possible to study in vitro any factor regulating the biogenesis of the organelle. Although we do not know whether such studies would reveal any new fundamental principle that governs the complex circuits of interconnected macromolecular interactions, the achievement of such a complete in vitro system would represent a necessary step in the comprehensive understanding of biogenesis of organelles, and eventually, of the more complex behavior and genesis of cells (89).

Requirement of proteins S5 and S9 from 30S subunits for the ribosome-dependent GTPase activity of elongation factor G

During CsCl isopycnic centrifugation at 20 mM Mg(++), Escherichia coli 30S ribosomal subunits specifically lose proteins S1, S2, S3, S5, S9, S10, and S14. The resultant 30S core is unable to stimulate the GTPase activity of EF-G in the presence of 50S subunits. Activity could be restored to a small extent by adding back S2, S5, or S9. However, when S5 and S9 were added together, they cooperatively produced 30S particles 1.5 times more active than the original native 30S subunits. The small amount of activity restored by S2 was simply additive to that restored by S5 or S9. None of the other split proteins showed any restoring capability. Ability of the various protein-deficient 30S particles to couple with 50S subunits corresponded closely to their activity in the EF-G GTPase reaction. It is concluded that S5 and S9 together enable the 30S subunit to participate in the formation of a GTPase-active 30S-50S-EF-G complex.