Identification of defined sequences in domain V of E. coli 23S rRNA in the 50S subunit accessible for hybridization with complementary oligodeoxyribonucleotides.

The accessibility of specific sequences in domain V of E. coli 23s rRNA in the 50S subunit to complementary oligodeoxyribonucleotides (cDNA) has been investigated. The apparent percentage of subunits engaged in complex formation was determined by incubation of radiolabeled cDNA probe with 50S subunits, followed by nitrocellulose membrane filtration of the reaction mixtures and measurement of the bound radiolabeled cDNA probes by liquid scintillation counting of the filters. The site(s) of hybridization were determined by digestion of the RNA in the RNA/DNA heteroduplex by RNase H. The results of this study indicated that single-stranded sequences, 2058-2062, 2448-2454, 2467-2483, and 2497-2505 were available for hybridization to cDNA probes. Bases 2489-2496, which have been postulated to be base paired with 2455-2461 were also accessible for hybridization.     

Evidence for a tRNA/rRNA interaction site within the peptidyltransferase center of the Escherichia coli ribosome

A nine-base oligodeoxyribonucleotide complementary to bases 2497-2505 of 23S rRNA was hybridized to both 50S subunits and 70S ribosomes. The binding of the probe to the ribosome or ribosomal subunits was assayed by nitrocellulose filtration and by sucrose gradient centrifugation techniques. The location of the hybridization site was determined by digestion of the rRNA/cDNA heteroduplex with ribonuclease H and gel electrophoresis of the digestion products, followed by the isolation and sequencing of the smaller digestion fragment. The cDNA probe was found to interact specifically with its rRNA target site. The effects on probe hybridization to both 50S and 70S ribosomes as a result of binding deacylated tRNA(Phe) were investigated. The binding of deacylated tRNA(Phe), either with or without the addition of poly(uridylic acid), caused attenuation of probe binding to both 50S and 70S ribosomes. Probe hybridization to 23S rRNA was decreased by about 75% in both 50S subunits and 70S ribosomes. These results suggest that bases within the 2497-2505 site may participate in a deacylated tRNA/rRNA interaction.     

A photolabile oligodeoxyribonucleotide probe of the peptidyltransferase center: identification of neighboring ribosomal components

In this work we report the synthesis of a radioactive, photolabile oligodeoxyribonucleotide probe and its exploitation in identifying 50S ribosomal subunit components neighboring its target site in 23S rRNA. The probe is complementary to 23S rRNA nucleotides 2497-2505, a single-stranded sequence that has been shown to fall within the peptidyltransferase center of Escherichia coli ribosomes [Cooperman, B. S., Weitzmann, C. J., & Fernandez, C. L. (1990) in The Ribosome: Structure, Function, & Evolution (Hill, W. E., Dahlberg, A., Garrett, R. A., Moore, P. B., Schlesinger, D., & Warner, J. R., Eds.) pp 491-501, American Society of Microbiology, Washington]. On photolysis in the presence of 50S ribosomes, it site-specifically incorporates into protein L3 (identified by both SDS-PAGE and immunological methods) and into three separate 23S rRNA regions: specifically, nucleotides 2454; 2501, 2502, 2505, 2506; and 2583, 2584. These results provide clear evidence that G-2505 in 23S rRNA is within 24 A (the distance between G-2505 and the photogenerated nitrene) of protein L3 and of each of the nucleotides mentioned above and are of obvious importance in the construction of detailed three-dimensional models of ribosomal structure. The approach we present is general and can be applied to determining ribosomal components neighboring regions of rRNA that are susceptible to binding by complementary oligodeoxyribonucleotides, both in intact 30S and 50S subunits and in subunits at various stages of reconstitution.     

Identification of 50S components neighboring 23S rRNA nucleotides A2448 and U2604 within the peptidyl transferase center of Escherichia coli ribosomes

The 23S rRNA nucleotides 2604-12 and 2448-58 fall within the central loop of domain V, which forms a major part of the peptidyl transferase center of the ribosome. We report the synthesis of radioactive, photolabile 2'-O-methyloligoRNAs, PHONTs 1 and 2, complementary to these nucleotides and their exploitation in identifying 50S ribosomal subunit components neighboring their target sites. Photolysis of the 50S complex with PHONT 1, complementary to nts 2604-12, leads to target site-specific photoincorporation into protein L2 and 23S rRNA nucleotides A886, Alpha1918, A1919, G1922-C1924, U2563, U2586, and C2601. Photolysis of the 50S complex with PHONT 2, complementary to nts 2448-58, leads to target site-specific probe photoincorporation into proteins L2, L3, one or more of proteins L17, L18, L21, and of proteins L9, L15, L16, and 23S rRNA nucleotides C2456 and psi2457. Chemical footprinting studies show that 2'-O-methyloligoRNA binding causes little distortion of the peptidyl transferase center but do provide suggestive evidence for the location of flexible regions within 23S rRNA. The significance of these results for the structure of the peptidyl transferase center is considered.     

Label-free detection of 16S ribosomal RNA hybridization on reusable DNA arrays using surface plasmon resonance imaging

In this paper, we describe the detection of bacterial cell-extracted 16S ribosomal RNA (rRNA) using an emerging technology, surface plasmon resonance (SPR) imaging of DNA arrays. Surface plasmon resonance enables detection of molecular interactions on surfaces in response to changes in the index of refraction, therefore eliminating the need for a fluorescent or radioactive label. A variation of the more common SPR techniques, SPR imaging enables detection from multiple probes in a reusable array format. The arrays developed here contain DNA probes (15-21 bases) designed to be complementary to 16S rRNA gene sequences of Escherichia coli and Bacillus subtilis as well as to a highly conserved sequence found in rRNAs from most members of the domain Bacteria. We report species-specific hybridization of cell-extracted total RNA and in vitro transcribed 16S rRNA to oligonucleotide probes on SPR arrays. We tested multiple probe sequences for each species, and found that success or failure of hybridization was dependent upon probe position in the 16S rRNA molecule. It was also determined that one of the probes intended to bind 16S rRNA also bound an unknown protein. The amount of binding to these probes was quantified with SPR imaging. A detection limit of 2 micro g ml-1 was determined for fragmented E. coli total cellular RNA under the experimental conditions used. These results indicate the feasibility of using SPR imaging for 16S rRNA identification and encourage further development of this method for direct detection of other RNA molecules.     

Identification of 23S rRNA nucleotides neighboring the P-loop in the Escherichia coli 50S subunit.

We report the synthesis of a radioactive, photolabile 2'-O-methyloligoRNA probe, 2258-53/52(SAz)-48, PHONT1, and its exploitation in identifying 23S rRNA nucleotides neighboring the so-called 'P-loop'. The probe is complementary to nt 2248-2258 in Escherichia coli 50S subunits. PHONT1 contains a p-azidophenacyl group attached to a phosphorothioate bridge between the nucleotides complementary to the positions 2252-2253, such that the photogenerated nitrene is maximally 17-19 A from 23S RNA nucleotides G2252 and G2253. PHONT1 binds to the 50S subunit, and photoincorporates within or immediately adjacent to its target site, as well as into several nucleotides falling between G2357 and A2430. The significance of these results for the structure of the peptidyl transferase center is considered. The PHONT approach is generally applicable to studies of complex RNA-containing molecules.     

Polynucleotide probes that target a hypervariable region of 16S rRNA genes to identify bacterial isolates corresponding to bands of community fingerprints

Temperature gradient gel electrophoresis (TGGE) is well suited for fingerprinting bacterial communities by separating PCR-amplified fragments of 16S rRNA genes (16S ribosomal DNA [rDNA]). A strategy was developed and was generally applicable for linking 16S rDNA from community fingerprints to pure culture isolates from the same habitat. For this, digoxigenin-labeled polynucleotide probes were generated by PCR, using bands excised from TGGE community fingerprints as a template, and applied in hybridizations with dot blotted 16S rDNA amplified from bacterial isolates. Within 16S rDNA, the hypervariable V6 region, corresponding to positions 984 to 1047 (Escherichia coli 16S rDNA sequence), which is a subset of the region used for TGGE (positions 968 to 1401), best met the criteria of high phylogenetic variability, required for sufficient probe specificity, and closely flanking conserved priming sites for amplification. Removal of flanking conserved bases was necessary to enable the differentiation of closely related species. This was achieved by 5' exonuclease digestion, terminated by phosphorothioate bonds which were synthesized into the primers. The remaining complementary strand was removed by single-strand-specific digestion. Standard hybridization with truncated probes allowed differentiation of bacteria which differed by only two bases within the probe target site and 1.2% within the complete 16S rDNA. However, a truncated probe, derived from an excised TGGE band of a rhizosphere community, hybridized with three phylogenetically related isolates with identical V6 sequences. Only one of the isolates comigrated with the excised band in TGGE, which was shown to be due to identical sequences, demonstrating the utility of a combined TGGE and V6 probe approach.     

Mapping the ribosomal RNA neighborhood of protein L11 by directed hydroxyl radical probing.

Ribosomal protein L11 is a highly conserved protein that has been implicated in binding of elongation factors to ribosomes and associated GTP hydrolysis. Here, we have analyzed the ribosomal RNA neighborhood of Escherichia coli L11 in 50 S subunits by directed hydroxyl radical probing from Fe(II) tethered to five engineered cysteine residues at positions 19, 84, 85, 92 and 116 via the linker 1-(p -bromoacetamidobenzyl)-EDTA. Correct assembly of the L11 derivatives was analyzed by incorporating the modified proteins into 50 S subunits isolated from an E. coli strain that lacks L11 and testing for previously characterized L11-dependent footprints in domain II of 23 S rRNA. Hydroxyl radicals were generated from Fe(II) tethered to L11 and sites of cleavage in the ribosomal RNA were detected by primer extension. Strong cleavages were detected within the previously described binding site of L11, in the 1100 region of 23 S rRNA. Moreover, Fe(II) tethered to position 19 in L11 targeted the backbone of the sarcin loop in domain VI while probing from position 92 cleaved the backbone around bases 900 and 2470 in domains II and V, respectively. Fe(II) tethered to positions 84, 85 and 92 also generated cleavages in 5 S rRNA around helix II. These data provide new information about the positions of specific features of 23 S rRNA and 5 S rRNA relative to protein L11 in the 50 S subunit and show that L11 is near highly conserved elements of the rRNA that have been implicated in binding of tRNA and elongation factors to the ribosome.     

Puromycin-rRNA interaction sites at the peptidyl transferase center

The binding site of puromycin was probed chemically in the peptidyl-transferase center of ribosomes from Escherichia coli and of puromycin-hypersensitive ribosomes from the archaeon Haloferax gibbonsii. Several nucleotides of the 23S rRNAs showed altered chemical reactivities in the presence of puromycin. They include A2439, G2505, and G2553 for E. coli, and G2058, A2503, G2505, and G2553 for Hf. gibbonsii (using the E. coli numbering system). Reproducible enhanced reactivities were also observed at A508 and A1579 within domains I and III, respectively, of E. coli 23S rRNA. In further experiments, puromycin was shown to produce a major reduction in the UV-induced crosslinking of deacylated-(2N3A76)tRNA to U2506 within the P' site of E. coli ribosomes. Moreover, it strongly stimulated the putative UV-induced crosslink between a streptogramin B drug and m2A2503/psi2504 at an adjacent site in E. coli 23S rRNA. These data strongly support the concept that puromycin, along with other peptidyl-transferase antibiotics, in particular the streptogramin B drugs, bind to an RNA structural motif that contains several conserved and accessible base moieties of the peptidyl transferase loop region. This streptogramin motif is also likely to provide binding sites for the 3' termini of the acceptor and donor tRNAs. In contrast, the effects at A508 and A1579, which are located at the exit site of the peptide channel, are likely to be caused by a structural effect transmitted along the peptide channel.     

Magnesium ion induced proton release as a probe for the polyelectrolytic structure of ribosomal RNAs and subunits

E coli ribosomes and rRNA's released 20 to 50 protons upon jump of magnesium ion concentration from 1 mM to 20 mM. The Mg2+-induced proton release was measured separately for 16S rRNA, 23S rRNA, 30S subunit, and 50S subunit by a new spectrophotometric method that had a much better sensitivity than the pH-stat method. The proton release from the subunits and rRNA's were similar in the number of protons, the pH dependence that had a minimum at neutral pH, and the upward concaveness of the Scatchard plot. From these results, the main source of protons in ribosomal subunits was assigned to nucleotide bases of rRNA's that showed a downward pKa shift upon Mg2+-ion binding. The subunits and rRNA's, however, differed in the proton release. 16S rRNA released protons somewhat more effectively than 23S rRNA, while 30S subunit released protons 2 to 5 times more effectively than 50S subunit. The marked difference between the two subunits suggest that ionizable bases in 16S and 23S rRNA's are covered and their pKa values are shifted by ribosomal proteins to different extents. The association of 30S and 50S subunits induced little proton release, showing that few ionizable groups with pKa near neutral pH are involved in the association. E. coli tRNA and poly U also showed Mg2+-induced proton release. The amounts of protons released from rRNA's, tRNA, and poly U were roughly proportional to the amount of bases not hydrogen bonded. The Mg2+-induced proton release from the natural and synthetic RNA's can be explained by the electrostatic field effect of polyphosphate backbones on bases not hydrogen bonded, as proposed in a previous paper. It also reflects the conformational structure of each RNA molecule.     

Probing dynamic changes in rRNA conformation in the 30S subunit of the Escherichia coli ribosome.

Ribosomal RNA molecules within each ribosomal subunit are folded in a specific three-dimensional form. The accessibility of specific sequences of rRNA of the small ribosomal subunit of Escherichia coli was analyzed using complementary oligodeoxyribonucleotides, 6-15 nucleotides long. The degree of hybridization of these oligomers to their RNA complements within the 30S subunit was assessed using nitrocellulose membrane filter binding assays. Specifically, the binding of short DNA oligomers (hexameric and longer) complementary to nucleotides 919-928, 1384-1417, 1490-1505, and 1530-1542 of 16S rRNA was monitored, and in particular how such binding was affected by the change in the activation state of the subunit. We found that nucleotides 1397-1404 comprise an unusually accessible sequence in both active and inactive subunits. Nucleotides 919-924 are partially available for hybridization in active subunits and somewhat more so in inactive subunits. Nucleotides 1534-1542 are freely accessible in active, but only partially accessible in inactive subunits, while nucleotides 1490-1505 and 1530-1533 are inaccessible in both, under the conditions tested. These results are in general agreement with results obtained using other methods and suggest a significant conformational change upon subunit activation.     

In situ accessibility of Escherichia coli 23S rRNA to fluorescently labeled oligonucleotide probes

One of the main causes of failure of fluorescence in situ hybridization with rRNA-targeted oligonucleotides, besides low cellular ribosome content and impermeability of cell walls, is the inaccessibility of probe target sites due to higher-order structure of the ribosome. Analogous to a study on the 16S rRNA (B. M. Fuchs, G. Wallner, W. Beisker, I. Schwippl, W. Ludwig, and R. Amann, Appl. Environ. Microbiol. 64:4973-4982, 1998), the accessibility of the 23S rRNA of Escherichia coli DSM 30083(T) was studied in detail with a set of 184 CY3-labeled oligonucleotide probes. The probe-conferred fluorescence was quantified flow cytometrically. The brightest signal resulted from probe 23S-2018, complementary to positions 2018 to 2035. The distribution of probe-conferred cell fluorescence in six arbitrarily set brightness classes (classes I to VI, 100 to 81%, 80 to 61%, 60 to 41%, 40 to 21%, 20 to 6%, and 5 to 0% of the brightness of 23S-2018, respectively) was as follows: class I, 3%; class II, 21%; class III, 35%; class IV, 18%; class V, 16%; and class VI, 7%. A fine-resolution analysis of selected areas confirmed steep changes in accessibility on the 23S RNA to oligonucleotide probes. This is similar to the situation for the 16S rRNA. Indeed, no significant differences were found between the hybridization of oligonucleotide probes to 16S and 23S rRNA. Interestingly, indications were obtained of an effect of the type of fluorescent dye coupled to a probe on in situ accessibility. The results were translated into an accessibility map for the 23S rRNA of E. coli, which may be extrapolated to other bacteria. Thereby, it may contribute to a better exploitation of the high potential of the 23S rRNA for identification of bacteria in the future.     

Interaction of the antibiotics clindamycin and lincomycin with Escherichia coli 23S ribosomal RNA.

Interaction of the antibiotics clindamycin and lincomycin with Escherichia coli ribosomes has been compared by chemical footprinting. The protection afforded by both drugs is limited to the peptidyl transferase loop of 23S rRNA. Under conditions of stoichiometric binding at 1 mM drug concentration in vitro, both drugs strongly protect 23S rRNA bases A2058 and A2451 from dimethyl sulphate and G2505 from kethoxal modification; G2061 is also weakly protected from kethoxal. The modification patterns differ in that A2059 is additionally protected by clindamycin but not by lincomycin. The affinity of the two drugs for the ribosome, estimated by footprinting, is approximately the same, giving Kdiss values of 5 microM for lincomycin and 8 microM for clindamycin. The results show that in vitro the drugs are equally potent in blocking their ribosomal target site. Their inhibitory effects on peptide bond formation could, however, be subtly different.     

Lack of complete cooperativity of ribosome assembly in vitro and its possible relevance to in vivo ribosome assembly and the regulation of ribosomal gene expression.

Earlier studies have shown that the reconstitution of Escherichia coli 50S as well as 30S ribosomal subunits from component rRNA and ribosomal protein (r-protein) molecules in vitro is not completely cooperative and binding of more than one r-protein to a single 16S rRNA (or 23S rRNA) molecule is required to initiate a successful 30S (or 50S) ribosome assembly reaction. We first confirmed this conclusion by carrying out 30S subunit reconstitution in the presence of a constant amount of 16S rRNA together with various amounts of total 30S r-proteins (TP30) and by analyzing the physical state of reconstituted particles rather than by assaying protein synthesizing activity of the particles as was done in the earlier studies. As expected, under conditions of excess rRNA, the efficiency of 30S subunit reconstitution per unit amount of TP30 decreased greatly with the decrease in the ratio of TP30 to rRNA, indicating the lack of complete cooperativity in the assembly reaction. We then asked the question whether the cooperativity of ribosome assembly is complete in vivo. We treated exponentially growing E coli cells with low concentrations of chloramphenicol which is known to inhibit protein synthesis without inhibiting rRNA synthesis, creating conditions of excess synthesis of rRNA relative to r-proteins. Several concentrations of chloramphenicol (ranging from 0.4 to 4.0 micrograms/ml) were used so that inhibition of protein synthesis ranged from 40 to 95%. Under these conditions, we examined the synthesis of RNA, ribosomal proteins and 50S ribosomal subunits as well as the synthesis of total protein. We found that the synthesis of 50S subunits was not inhibited as much as the synthesis of total protein at lower concentrations of chloramphenicol, but the degree of inhibition of 50S subunit synthesis increased sharply with increasing concentrations of chloramphenicol and was in fact greater than the degree of inhibition of total protein synthesis at chloramphenicol concentrations of 2 micrograms/ml or higher. The inhibition of 50S subunit synthesis was significantly greater than the inhibition of r-protein synthesis at all chloramphenicol concentrations examined. These data are consistent with the hypothesis that the cooperativity of ribosome assembly in vivo is also not complete as is the case for in vitro ribosome reconstitution, but are difficult, if not impossible, to explain on the basis of the complete cooperativity model.(ABSTRACT TRUNCATED AT 400 WORDS)     

DNA-hybridization electron microscopy. Localization of five regions of 16 S rRNA on the surface of 30 S ribosomal subunits

DNA-hybridization electron microscopy has been used to locate five regions of 16 S rRNA on the surface of 30 S ribosomal subunits. Biotinylated DNA probes that are complementary to selected regions of 16 S rRNA were hybridized to activated 30 S ribosomal subunits. These hybridized probes were reacted with avidin and localized by electron microscopy. The specificity of DNA binding was monitored with RNase H, which recognizes RNA-DNA hybrids and cleaves the RNA. Three of the five sequences examined were mapped on the platform. These sequences are 686-703, 714-733 and 787-803. Region 1492-1505 is mapped in the cleft and region 518-533 is at the neck on the side opposite the platform, respectively.     

Nuclease S1 mapping of 16S ribosomal RNA in ribosomes

Escherichia coli 16S rRNA and 16S-like rRNAs from other species have several universally conserved sequences which are believed to be single-stranded in ribosomes. The quantitative disposition of these sequences within ribosomes is not known. Here we describe experiments designed to explore the availability of universal 16S rRNA sequences for hybridization with DNA probes in 30S particles and 70S ribosomes. Unlike previous investigations, quantitative data on the accessibility of DNA probes to the conserved portions of 16S rRNA within ribosomes was acquired. Uniquely, the experimental design also permitted investigation of cooperative interactions involving portions of conserved 16S rRNA. The basic strategy employed ribosomes, 30S subunits, and 16S rRNAs, which were quantitatively analyzed for hybridization efficiency with synthetic DNA in combination with nuclease S1. In deproteinated E. coli 16S rRNA and 30S subunits, the regions 520-530, 1396-1404, 1493-1504, and 1533-1542 are all single-stranded and unrestricted for hybridization to short synthetic DNAs. However, the quantitative disposition of the sequences in 70S ribosomes varies with each position. In 30S subunits there appear to be no cooperative interactions between the 16S rRNA universal sequences investigated.     

Conformational studies of Escherichia coli ribosomes with the use of acridine orange as a probe

The interaction of E. coli vacant ribosomes with acridine orange (AO) was studied, to obtain conformational information about rRNAs in ribosomes. Acridine orange binds to an RNA in two different modes: cooperative outside binding with stacking of bound AO's and intercalation between nucleotide bases. Free 16S and 23S rRNAs have almost identical affinities to AO. At 1 mM Mg2+, AO can achieve stacking binding on about 40% of rRNA phosphate groups. The number of stacking binding sites falls to about 1/3 in the 30S subunit in comparison with free 16S rRNA. In the 50S subunit, the number of stacking binding sites is only 1/5 in comparison with free 23S rRNA. Mg2+ ions are more inhibitory for the binding of AO to ribosomes than to free rRNAs. The strength of stacking binding appears to be more markedly reduced by Mg2+ in active ribosomes than in rRNAs. "Tight couple" 70S particles are less accessible for stacking binding than free subunits. The 30S subunits that have irreversibly lost the capability for 70S formation under low Mg2+ conditions have an affinity to AO that is very different from that of active 30S but similar to that of free rRNA, though the number of stacking binding sites is little changed by the inactivation. 70S and 30S ribosomes with stacking bound AO's have normal sedimentation constants, but the 50S subunits reversibly form aggregates.     

The importance of highly conserved nucleotides in the binding region of chloramphenicol at the peptidyl transfer centre of Escherichia coli 23S ribosomal RNA.

The peptidyl transfer site has been localized at the centre of domain V of 23S-like ribosomal RNA (rRNA) primarily on the basis of a chloramphenicol binding site. The implicated region constitutes an unstructured circle in the current secondary structural model which contains several universally conserved nucleotides. With a view to investigate the function of this RNA region further, four of these conserved nucleotides, including one indirectly implicated in chloramphenicol binding, were selected for mutation in Escherichia coli 23S rRNA using oligonucleotide primers. Mutant RNAs were expressed in vivo on a plasmid-encoded rRNA (rrnB) operon and each one yielded dramatically altered phenotypes. Cells exhibiting A2060----C or A2450----C transversions were inviable and it was shown by inserting the mutated genes after a temperature-inducible promoter that the mutant RNAs were directly responsible. In addition, a G2502----A transition caused a decreased growth rate, probably due to a partial selection against mutant ribosome incorporation into polysomes, while an A2503----C transversion produced a decreased growth rate and conferred resistance to chloramphenicol. All of the mutant RNAs were incorporated into 50S subunits, but while the two lethal mutant RNAs were strongly selected against in 70S ribosomes, the plasmid-encoded A2503----C RNA was preferred over the chromosome-encoded RNA, contrary to current regulatory theories. The results establish the critical structural and functional importance of highly conserved nucleotides in the chloramphenicol binding region. A mechanistic model is also presented to explain the disruptive effect of chloramphenicol (and other antibiotics) on peptide bond formation at the ribosomal subunit interface.     

Proteins neighboring 18S rRNA conserved sequences 609-618 and 1047-1061 within the 40S human ribosomal subunit

The structure of human 40S ribosomal subunits has been probed by a cross-linking strategy based on the use of oligonucleotide derivatives that modify proteins in the vicinity of specific 18S rRNA sequences. The oligonucleotide derivatives carried a p-azidoperfluorobenzamide group at the 5' ends and were complementary to 18S rRNA sequences 609-618 and 1047-1061, homologous to the highly conserved regions designated as the "530 stem-loop" and "790 stem-loop", respectively, in Escherichia coli 16S rRNA. Ribosomal proteins surrounding these sequences were the main targets of the cross-linking. Proteins S3 and S5 were cross-linked to the derivative complementary to the sequence 609-618, and proteins S2 and S3 were modified by the derivative complementary to the sequence 1047-1061. Cross-linking was not affected by binding of 40S subunits to either poly(U) or poly(U) and Phe-tRNA(Phe).