Site-Directed Hydroxyl Radical Probing of the rRNA Neighborhood of Ribosomal Protein S5

Abstract

Cysteine residues were introduced into three different positions distributed on the surface of ribosomal protein S5, to serve as targets for derivatization with an Fe(II)-ethyl-enediaminetetraacetic acid linker. Hydroxyl radicals generated locally from the tethered Fe(II) in intermediate ribonucleoprotein particles or in 30S ribosomal subunits reconstituted from derivatized S5 caused cleavage of the RNA, resulting in characteristically different cleavage patterns for the three different tethering positions. These findings provide constraints for the three-dimensional folding of 16S ribosomal RNA (rRNA) and for the orientation of S5 in the 30S subunit, and they further suggest that antibiotic resistance and accuracy mutations in S5 may involve perturbation of 16S rRNA.

Efficient reconstitution of functional Escherichia coli 30S ribosomal subunits from a complete set of recombinant small subunit ribosomal proteins

“…The genes encoding S5 and S17 had previously been cloned (Heilek & Noller, 1996b; G+M+ Heilek & H+F+ Noller, unpubl+ results)+ The genes encoding ribosomal proteins S2-S4, S6-S16, and S18-S21 were amplified by polymerase chain reaction (PCR) of E. coli MRE600 genomic DNA+ An NdeI restriction enzyme site was included at the 59 end of every clone+ Either a BamHI (S2-S4, S6, S8-S16, S18, S19, and S21) or a HindIII (S7 and S20) restriction enzyme site was included at the 39 end of the genes+ The PCR products were cleaved with the appropriate enzymes (see above) and ligated into pET24b that had been cleaved with the same enzymes and purified+ Wild-type clones were identified by sequence analysis and transformed into BL21(DE3), where the proteins were overexpressed from an inducible T7 promoter on the plasmid (Studier et al+, 1990; Novagen)+ For overexpression, strains harboring the plasmids were grown to an approximate OD 600 of 0+4, IPTG was added to a final concentration of 1 mM, and the cultures were grown for an additional 4 h prior to harvesting+ Cells were washed once with Buffer E and stored at Ϫ20 8C+ For analysis of overexpression, equal volumes of induced cell culture and SDS-PAGE loading dye containing 6 M Urea were mixed, and 30 mL of this mixture was analyzed by SDS-polyacrylamide gel electrophoresis (Laemmli, 1970; 4% stacking gel; 12% resolving gel, both containing 6 M Urea)+…”

“…Sedimentation analysis of in vitro reconstitution of 30S subunits using a complete set of recombinant small subunit proteins+ Reconstitution of 30S subunits using a complete set of small subunit recombinant proteins and (A) standard reconstitution conditions (Traub & Nomura, 1969) or (B) ordered assembly following the assembly map (Table 2; Mizushima & Nomura, 1970;Held et al+, 1974)+ (I) natural 30S subunits; 30S subunits reconstituted with 16S rRNA (30 pmol) and a complete set of recombinant small subunit proteins; (II) 2 molar equivalents protein; (III) 4 molar equivalents protein; (IV) 6 molar equivalents protein; (V) 8 molar equivalents protein+ C: Comparison of 30S subunit reconstitution systems and procedures+ (I) natural 30S subunits; 30S subunits reconstituted with 16S rRNA (30 pmol) and (II) TP30, a mixture of total proteins isolated from 30S subunits, following the standard protocol (Traub & Nomura, 1969); a complete set of recombinant small subunit proteins following (III) the assembly map groupings (Table 2); (IV) the assembly kinetics groupings (Table 2); (V) the standard protocol (Traub & Nomura, 1969)+ D: Comparison of 30S subunit reconstitution with Fe(II)-derivatized S5+ 30S subunits reconstituted with 16S rRNA (60 pmol), Fe(II)-C129-S5 and (I) the remaining recombinant small subunit proteins following the assembly map groupings; the remaining small subunit proteins individually isolated from 30S subunits following (II) the standard protocol (Traub & Nomura, 1969); (III) the assembly map groupings+ Sedimentation is from left to right, and absorbance was monitored at 254 nm+ ies, 30S subunits containing a single Fe(II)-derivatized protein were reconstituted from proteins that had been individually isolated from ribosomes (Heilek et al+, 1995;Heilek & Noller, 1996a, 1996b; therefore, we wished to compare the efficiencies of these two reconstitution systems for preparing such constructs+ Ribosomal protein S5 derivatized with Fe(II) at a unique site (Fe-C129-S5; Heilek & Noller, 1996b;Culver et al+, 1999), was reconstituted with 16S rRNA and a full complement of the remaining recombinant proteins or proteins that had been individually purified from ribosomes (Fig+ 3D)+ Reconstitution of 30S subunits was dramatically more efficient using ordered assembly with the recombinant proteins (Fig+ 3D, I) compared to reconstitution using proteins individually isolated from ribosomes, either under standard conditions (Fig+ 3D, II) or using the ordered assembly protocol (Fig+ 3D, III)+ Clearly, the recombinant protein system is advantageous for obtaining efficient reconstitution of 30S subunits containing a single derivatized protein+…”

“…Genomic DNA from E. coli MRE600 was used as a template for polymerase chain reaction (PCR) amplification of the genes encoding ribosomal proteins S2-S4, S6-S16, and S18-S21+ The genes encoding S1 (Sorensen et al+, 1998), S5 (Heilek & Noller, 1996b), and S17 (G+M+ Heilek & H+F+ Noller, unpubl+ results) have previously been cloned+ Primers for the PCR reactions were designed to facilitate cloning (by the inclusion of restriction enzyme sites) and expression (by optimizing the distance between the start codon and the ribosome binding site within the vector)+ After amplification and restriction enzyme digestion, the PCR products were cloned directly into the pET24b vector (Novagen), which contains an inducible promoter and an f1 origin for production of single-stranded DNA+ The integrity of each individual clone was confirmed by restriction enzyme digestion and DNA sequence analysis (data not shown)+ Induction of protein expression in the E. coli strain BL21(DE3) results in production of differing amounts of the various proteins of which some are soluble and some insoluble (Fig+ 1A; Table 1); the perceived differences in induction levels may in part reflect varying staining efficiencies of the different proteins (see Fig+ 1)+ For some constructs, it appears that multiple species are overexpressed upon induction (Fig+ 1A); this may be an artifact of the expression system, since similarly sized contaminants are observed with more than one construct (see Fig+ 1A; compare, for example, S18 and S21)+ Nevertheless, these contaminants do not appear to interfere with purification (Figs+ 1B and 2)+ The level of overexpression and the highly charged nature of most of the proteins enables single-column FPLC purification (Figs+ 1B and 2; Table 1)+ Representative chromatograms of FPLC cation-exchange purification of two overexpressed proteins are shown in Figure 2+ S2 has a relatively low isoelectric point (Kaltschmidt, 1971), is insoluble, and is purified at lower pH (Fig+ 2A; Table 1), compared to S4, which is quite basic, soluble, and purified at the higher of the two pHs used during chromatography (Fig+ 2B; Table 1; Kaltschmidt, 1971)+ Each of the overexpressed proteins was similarly purified to near homogeneity by FPLC cation-exchange chromatography, except for S6, which was purified by FPLC anionexchange chromatography (Fig+ 1B; …”

Directed hydroxyl radical probing of 16S ribosomal RNA in ribosomes containing Fe(II) tethered to ribosomal protein S20

“…Single cysteine residues were introduced into Escherichia coli S20, which contains no cysteines in its wildtype sequence+ Sites for cysteine substitution were chosen using an amino acid sequence alignment of S20 proteins from five organisms, targeting nonconserved residues likely to be found on the surface of the protein+ Six sites for cysteine substitution, positions S14, S23, K49, I57, A72, and A77, were chosen out of the 86 amino acids of S20+ The wild-type copy of the S20 gene was amplified by polymerase chain reaction (PCR) and cloned into pET24 (see Materials and Methods), which allows both overexpression of the recombinant protein as well as a means for generating single-stranded DNA for use in site-directed mutagenesis (Kunkel, 1985)+ Wild-type S20 and the six cysteinecontaining S20 mutants were overexpressed, purified, and derivatized essentially as described for ribosomal protein S5 (Heilek & Noller, 1996b)+…”

Directed hydroxyl radical probing of 16S rRNA in the ribosome: Spatial proximity of RNA elements of the 3′ and 5′ domains

“…Further evidence for correct folding of the 424 region of 16S rRNA in our Fe(II)-BABE-derivatized 30S subunits comes from the fact that the 420 helix is part of the binding site for protein S4 (Powers & Noller, 1995a, 1995b)+ S4 is a primary RNA-binding protein whose incorporation is essential for the correct assembly of 30S subunits (Held et al, 1974)+ The fact that our Fe(II)-BABE-derivatized particles cosediment with 30S subunits and are active in both tRNA binding and subunit association suggests that S4 is properly assembled, and that the 420-region helical elements are properly folded+ Although we cannot exclude some mobility of the nucleotide itself, this level of uncertainty would not severely compromise the relatively low-resolution information obtained by our method+ Hydroxyl radical probing from Fe(II) tethered to nt 424 results in cleavage of the RNA backbone in both the 59 and 39 major domains of 16S rRNA+ In the 59 domain, cleavage is observed in the highly conserved 530 stem and loop, with the strongest cleavage around position 513 in the stem (Fig+ 3)+ Although the location of the 530 loop in relation to other 16S rRNA elements, especially the 1400 decoding region, has been controversial (see, for example, Powers & Noller, 1994;Brimacombe, 1995), the 530 loop has been well constrained to the body of the 30S subunit (Stern et al+, 1988b;Malhotra & Harvey, 1994;Fink et al+, 1996;)+ The cleavages observed in the 39 major domain of 16S rRNA around positions 999, 1029, 1044, and 1206 are especially interesting+ While the 59 domain of 16S rRNA is localized to the body of the 30S subunit, the 39 major domain constitutes the head of the subunit+ The only covalent connection between the head and the body of the subunit is through the 930 helix, which is believed to be the "neck" of the subunit (Stern et al+, 1988b;Malhotra & Harvey, 1994;Fink et al+, 1996;)+ Cleavage of rRNA in the 39 domain from position 424 provides direct evidence for a second point of RNA-RNA proximity between the head and body of the 30S subunit+ The cleavage intensities suggest that the closest approach of nt 424 to the head is around position 1044 (Fig+ 3)+ Previous studies have also suggested proximity between the 420 helix and 530 stem-loop in the 59 domain of 16S rRNA and the 1200 region in the 39 domain+ Directed hydroxyl radical probing from a single position on ribosomal protein S5 results in cleavage of 16S rRNA in both of these regions (Heilek & Noller, 1996)+ Additionally, UV cross-links have been identified between mRNA analogs and 16S rRNA positions 1052, The RNA-RNA proximity that we observe between the 59 and 39 domains of 16S rRNA may correspond to the close approach between the shoulder of the body and the head of the 30S subunit seen in electron micrograph reconstructions of ribosomes (Stark et al+, 1995(Stark et al+, , 1997Agrawal et al+, 1996;…”