Preparation of active ribosomal proteins by HPLC

Different consequences of incorporating chloroplast ribosomal proteins L12 and S18 into the bacterial ribosomes of Escherichia coli

We have incorporated chloroplast ribosomal proteins (R-proteins) L12 and S18 into Escherichia coli ribosomes and examined the hybrid ribosomes for their ability to form polysomes in vivo and perform poly(U)-dependent poly(Phe) synthesis in vitro. The rye chloroplast S18 used for the experiment is a highly divergent protein (170 amino acid residues; E. coil S18, 74 residues), containing a repeating, chloroplast-specific, heptapeptide motif, and has amino acid sequence identity of only 35% to E. coli S18. When expressed in E. coli, chloroplast S18 was assembled in E. coli ribosomes. The latter formed polysomes in vivo at about the same rate as the host ribosomes, indicating that the replacement of E. coli S18 with its chloroplast homologue has only a minor, if any, effect on function. The L12 protein is much more conserved in sequence and chain length, and is known to have a very important function. The Arabidopsis chloroplast L12 used in the experiment was incorporated into E. coli 50S subunits that associated with the 30S subunits to form ribosomes, but the latter were unable to form polysomes. This result indicates functional inactivation of E. coil ribosomes by a chloroplast R-protein. To further confirm this result, we overproduced chloroplast L12 through the use of a secretion vector and purified the protein to homogeneity. Chloroplast L12 could be efficiently incorporated in vitro into L7/12-lacking E. coli ribosomes, but the hybrid ribosomes were totally inactive in poly(U)-dependent poly(Phe) synthesis. Computer modeling of the spatial structure of all known chloroplast L12 proteins (using E. coli L12 coordinates) indicated a 'chloroplast loop' present only in chloroplast L12. The presence of this loop might have a role in the observed inactivation. Taken together with previously reported results (summarized in this paper), it would appear that the features of chloroplast R-proteins concerned with specific functions are more divergent than their assembly properties. We have previously described methods suitable for overproduction and purification of chloroplast R-proteins that are encoded in organellar DNA (approximately 20), but that gave poor yield for those encoded in the nuclear DNA (approximately 45). Here we describe a method that overcomes this problem and allows the purification of nucleus-encoded chloroplast R-proteins in milligram quantities.

Direct shape determination of ribosomal proteins in solution and within the ribosome by means of neutron scattering

Following the 'strategy of the glassy ribosome' single protonated ribosomal proteins (r-proteins) were reconstituted into deuterated 50S subunits of Escherichia coli. The deuteration of both rRNA and r-proteins were individually adjusted to such a degree that the ribosomal matrix appeared nearly homogeneous with respect to coherent neutron scattering and had a scattering density equivalent to a D2O solution of about 90%. Neutron scattering of ribosomal subunits was recorded in reconstitution buffer containing three different concentrations of D2O around 90% D2O (contrast variation). The signal-to-noise ratio achieved allowed us to make a direct determination of the radii of gyration of r-proteins within the 50S subunit and thus provides the first information relating to the shape of these proteins in situ. We present the radii of gyration of 11 r-proteins incorporated into 50S subunits and of 9 isolated r-proteins in solution. In addition, the data concerning the overall dimensions of the r-proteins we report on indicate that conformational changes of at least two individual r-proteins occur during the assembly process of the ribosome.

Modification of histidine residues on proteins from the 50S subunit of the Escherichia coli ribosome. Effects on subunit assembly and peptidyl transferase centre activity

L2, L3, L4, L16 and L20 are proteins of the 50S ribosomal subunit of Escherichia coli which are essential for the assembly and activity of the peptidyl transferase centre. These proteins have been modified with the histidine-specific reagent, diethylpyrocarbonate, while L17 and L18 were treated as controls. Each modified protein tested was able to participate in the reconstitution of a 50S particle when replacing its normal counterpart, although the particles assembled with modified L2 were heterogeneous. However, although they could support assembly, modified L16 and L20 were not themselves reconstituted stably, and modified L2 and L3 were found in less than stoichiometric amounts. Particles assembled in the presence of modified L16 retained significant peptidyl transferase activity (60-70% at 10 mM diethylpyrocarbonate) whereas those reconstituted with modified L2, L3, L4 or L20 had low activity (10-30% at 10 mM diethylpyrocarbonate). The particles assembled with the modified control protein, L17, retained 80% of their peptidyl transferase activity under the same conditions. The histidine residues within the essential proteins therefore contribute to ribosome structure and function in three significant ways; in the correct assembly of the ribosomal subunit (L2), for the stable assembly of the proteins within the ribosomal particle (L20 and L16 in particular), and directly or indirectly for the subsequent activity of the peptidyl transferase centre (L2, L3, L4 and L20). The essential nature of the unmodified histidines for assembly events precludes the use of the chemical-modification strategy to test the proposal that a histidine on one of the proteins might participate in the catalytic activity of the centre.

The complex between ribosomal proteins and aminoacyl-tRNA: the interactions and hydrolytic activities are not confined to the proteins L2 and L16 of Escherichia coli ribosomes

The capacity of some Escherichia coli (E. coli) ribosomal proteins to bind to tRNA and to hydrolyse their aminoacylated derivatives has been analysed. The following results were obtained: (1) The basic proteins L2, L16 and L33 and S20 bound f[3H]Met-tRNA to a similar extent as the total proteins from 30 S (TP30) or 50 S (TP50) when tested by nitrocellulose filtration, in contrast to the more acidic proteins L7/L12 and S8. (2) The proteins of the peptidyltransferase centre, L2 and L16, showed no distinct specificity, binding various charged tRNAs from E. coli and Saccharomyces cerevisiae (S. cerevisiae). (3) A number of isolated ribosomal proteins hydrolysed aminoacyl-tRNA as assessed by trichloroacetic acid precipitation, in contrast to the TP30 and TP50. (4) The loss of radiolabel from Ac[14C]Phe-tRNA and from [14C]tRNA in the presence of these proteins could not be prevented by RNasin, a ribonuclease inhibitor, whereas that mediated by a sample of non-RNase-free bovine serum albumin was inhibited. (5) When double-labelled, Ac[3H]Phe-[14C]tRNA was incubated with L2 both radiolabels were lost, indicating that this potential candidate for a peptidyltransferase enzyme does not specifically cleave the ester bond between the aminoacyl residue and the tRNA.