Alain Favre

THIONUCLEOBASES AS INTRINSIC PHOTOAFFINITY PROBES OF NUCLEIC ACID STRUCTURE AND NUCLEIC ACID-PROTEIN INTERACTIONS

In the past few years thionucleobases have been extensively used as intrinsic photolabels to probe the structure in solution of folded RNA molecules and to identify contacts within nucleic acids and/or between nucleic acids and proteins, in complex nucleoprotein assemblies. These thio residues such as 4-thiouracil found in E. coli tRNA and its non-natural congeners 4-thiothymine, 6-thioguanine and 6-mercaptopurine absorb light at wavelengths longer than 320 nm and, thus, can be selectively photoactivated. Synthetic or enzymatic procedures have been established, allowing the random or site-specific incorporation of thionucleotide(s) within a RNA (DNA) chain which, in most cases, retains unaltered structural and biological properties. Owing to the high photoreactivity of their triplet state (intersystem yield close to unity), 4-thiouracil and 4-thiothymine derivatives exhibit a high photocrosslinking ability towards pyrimidines (particularly thymine) but also purines. From the nature of the photoproducts obtained in base or nucleotide mixtures and in dinucleotides, the main photochemical pathway was identified as a (2 + 2) photoaddition of the excited C-S bond onto the 5, 6 double bond of pyrimidines yielding thietane intermediates whose structure could be characterized. Depending on the mutual orientation of these bonds in the thietanes, their subsequent dark rearrangement yielded, respectively, either the 5-4 or 6-4 bipyrimidine photoadduct. A similar mechanism appears to be involved in the formation of the unique photoadduct formed between 4-thiothymidine and adenosine. The higher reactivity of thymine derived acceptors can be explained by an additional pathway which involves hydrogen abstraction from the thymine methyl group, followed by radical recombination, leading to methylene linked bipyrimidines. The high photocrosslinking potential of thionucleosides inserted in nucleic acid chains has been used to probe RNA-RNA contacts within the ribosome permitting, in particular, the elucidation of the path of mRNA throughout the small ribosomal subunit. Functional interactions between the mRNA spliced sites and U RNAs could be detected within the spliceosome. Analysis of the photocrosslinks obtained within small endonucleolytic ribozymes in solution led to a tertiary folded pseudo-knot structure for the HDV ribozyme and allowed the construction of a Y form of a hammerhead ribozyme, which revealed to be in close agreement with the structure observed in crystals. Thionucleosides incorporated in nucleic acids crosslink efficiently amino-acid residues of proteins in contact with them. Despite the fact that little is known about the nature of the photoadducts formed, this approach has been extensively used to identify protein components interacting at a defined nucleic acid site and applied to various systems (replisome, spliceosome, transcription complexes and ribosomes).

Photochemistry of 4-thiouridine in Escherichia coli transfer RNA1Val

Abstract Irradiation of pure transfer RNA1Val with monochromatic light (334 nm) produces characteristic changes in the spectral properties of 4-thiouridine, the only base which strongly absorbs light at this wavelength. Variations in absorption and fluorescence of 4-thiouridine during irradiation are interpreted in terms of a specific, quantitative photoreaction which proceeds with a yield of about 5 × 10−3 E/mole. The photoreaction occurs under conditions where tRNA1Val is biologically active but not under conditions that destroy the tertiary structure of the 4-thiouridine region. Sequence analysis of irradiated 32P-labelled tRNA1Val, reported elsewhere, showed the formation of a covalent linkage between 4-thiouridine in the 8th position and cytosine in the 13th position. The 4-thiouridine-cytosine dimer has been isolated in pure form and is quite distinct from known pyrimidine photoproducts. Sodium borohydride reduces the 4-thiouridine-cytosine dimer to a new chromophore, which strongly fluoresces in the tRNA. Physical studies of the native and modified tRNA1Val show that introduction of the cross-link results in a restricted conformational change of the tRNA molecule. The results are discussed with respect to tRNA tertiary structure and biological activity.

Human replication protein A unfolds telomeric G-quadruplexes

G-quadruplex structures inhibit telomerase activity and must be disrupted for telomere elongation during S phase. It has been suggested that the replication protein A (RPA) could unwind and maintain single-stranded DNA in a state amenable to the binding of telomeric components. We show here that under near-physiological in vitro conditions, human RPA is able to bind and unfold G-quadruplex structures formed from a 21mer human telomeric sequence. Analyses by native gel electrophoresis, cross-linking and fluorescence resonance energy transfer indicate the formation of both 1:1 and 2:1 complexes in which G-quadruplexes are unfolded. In addition, quadruplex opening by hRPA is much faster than observed with the complementary DNA, demonstrating that this protein efficiently unfolds G-quartets. A two-step mechanism accounting for the binding of hRPA to G-quadruplexes is proposed. These data point to the involvement of hRPA in regulation of telomere maintenance.

The invariant uridine of stop codons contacts the conserved NIKSR loop of human eRF1 in the ribosome

To unravel the region of human eukaryotic release factor 1 (eRF1) that is close to stop codons within the ribosome, we used mRNAs containing a single photoactivatable 4‐thiouridine (s4U) residue in the first position of stop or control sense codons. Accurate phasing of these mRNAs onto the ribosome was achieved by the addition of tRNAAsp. Under these conditions, eRF1 was shown to crosslink exclusively to mRNAs containing a stop or s4UGG codon. A procedure that yielded 32P‐labeled eRF1 deprived of the mRNA chain was developed; analysis of the labeled peptides generated after specific cleavage of both wild‐type and mutant eRF1s maps the crosslink in the tripeptide KSR (positions 63–65 of human eRF1) and points to K63 located in the conserved NIKS loop as the main crosslinking site. These data directly show the interaction of the N‐terminal (N) domain of eRF1 with stop codons within the 40S ribosomal subunit and provide strong support for the positioning of the eRF1 middle (M) domain on the 60S subunit. Thus, the N and M domains mimic the tRNA anticodon and acceptor arms, respectively.

Mutagenesis and growth delay induced in Escherichia coli by near-ultraviolet radiations

The literature relating to genetic changes induced in Escherichia coli by near-ultraviolet radiations is reviewed and summarized: i) these radiations are much less mutagenic than would be expected from the known level of DNA damage, ii) pre-illumination with near-UV light antagonizes the mutagenic effect of UV (254 nm) light. In agreement with these findings, the SOS functions are not induced by near-UV radiations. Furthermore prior exposure of cells to near-UV light inhibits the subsequent 254 nm induction of the SOS response. Among the several hypothesis considered to explain these observations, one can be clearly favoured. Near-UV light triggers, at sublethal fluences, the growth delay effect. The target molecules, tRNAs, are photocrosslinked and some tRNA species become poor substrates in the acylation reaction. In vivo these tRNA molecules accumulate on the uncharged form, leading to a transient cessation of protein synthesis. The SOS response is inducible and as such requires protein synthesis. We therefore propose that near-ultraviolet radiations have a dual effect: i) they induce, mostly indirectly, DNA lesions which are potentially able to trigger the SOS response, ii) they prevent the expression of the SOS functions through the transient inhibition of protein synthesis (growth delay).

4-thiouridine as the target for near-ultraviolet light induced growth delay in Escherichia coli

Abstract 4-thiouridine present in position 8 of a number of tRNAs is the chromophore to near ultraviolet light that leads to growth delay in E. coli, through 8–13 link formation. 1. i) 4-thiouridine absorption spectra and the action spectra for cross-link formation fit more closely the action spectra for growth delay than any absorption spectra of other known cellular chromophores. 2. ii) 8–13 link formation occurs in vivo. The amount of cross-linked tRNA in the cells is intimately correlated to growth delay. 3. iii) The resumption of growth of irradiated cells occur when cross-linked tRNA are no more detectable in vivo. Essentially, the growth delay is induced through inactivation of tRNAPhe, and possibly some other tRNA species.

TARGETED PHOTOCHEMICAL MODIFICATION OF HIV‐DERIVED OLIGORIBONUCLEOTIDES BY ANTISENSE OLIGODEOXYNUCLEOTIDES LINKED TO PORPHYRINS

Abstract Antisense oligodeoxynucleotides directed against a 24‐mer RNA derived from the long terminal repeat (LTR) region of HIV were linked to proto‐ and methylpyrroporphyrin and their zinc derivatives. The oligonucleotide‐porphyrin conjugates were tested for their ability to induce photodamage on the target RNA. Upon hybridization followed by irradiation at 405 nm, the photochemical reaction led to photocross‐linking of the antisense derivative to the RNA substrate. The protoporphyrin exhibited a much higher cross‐linking yield than the methylpyrroporphyrin while the Zn‐porphyrin derivatives were found to be less efficient than their corresponding nonmetallated congeners. The specificity of the photocross‐linking reaction between the porphyrin‐oligomer and its target RNA was demonstrated by the following evidence: (1) hybrid formation was required for photocross‐linking to occur, (2) the sites of cross‐linking on the target RNA were identified at G residues located in close proximity to the porphyrin photoactive center in the hybrid and (3) addition of bulk calf liver RNA did not affect the photocross‐linking efficiency.

Near ultraviolet DNA damage induces the SOS responses in Escherichia coli

The influence of the growth delay induced by near u.v. radiation on the SOS response was monitored by comparing the level of sfiA expression by means of a sfiA::lacZ fusion in both a nuvA+ cell and an isogenic nuvA mutant. The mutant lacks 4‐thiouridine in its tRNA and does not exhibit the near u.v.‐induced growth delay. Although the two strains exhibit similar sfiA induction levels after 254 nm irradiation, their behaviour is different after illumination with near u.v. light, including solar u.v. Inducibility is 10‐20 times higher in the nuvA mutant than in the parent strain. Furthermore, pre‐illumination with broad band near u.v. light does not affect the 254 nm‐induced sfiA response in the mutant but reduces it by a factor of 3‐4 in the parent strain. The kinetics of sfiA induction in near u.v.‐illuminated nuvA+ cells, whether treated with 254 nm light or not, is unusual and follows the growth curve: only after 50 min is sfiA derepression observed. It can be concluded that (i) near u.v.‐induced DNA lesions are able to trigger the SOS response and (ii) the growth delay effect reduces this response, whether triggered by u.v. or near u.v. light. Hence 4‐thiouridine in tRNA acts as a built‐in antiphotomutagenic ‘device’ protecting Escherichia coli cells against mutagenesis and the induction of the SOS response by near u.v. light and sunlight.

Biological activity of irradiated tRNAVal containing a 4-thiouridine-cytosine dimer☆

Abstract The biological activity of 335 nm irradiated tRNAVal containing a covalent bond between the 4tU in position 8 and the C in position 13 was tested. The irradiated tRNA is fully acylated but its affinity for the synthetase is decreased by a factor of three as revealed by competition experiments. On the contrary, no difference was observed between normal and irradiated tRNA for ribosome binding in the presence of poly (U,G). Furthermore, the valine acylated to irradiated tRNA was incorporated into polypeptide chains with a rate that was twofold slower than with normal tRNA. These results show that the 4tU-C region can influence the synthetase binding and that tRNA function does not require large spatial separation of bases in this region.

Stop Codons and UGG Promote Efficient Binding of the Polypeptide Release Factor eRF1 to the Ribosomal A Site

To investigate the codon dependence of human eRF1 binding to the mRNA-ribosome complex, we examined the formation of photocrosslinks between ribosomal components and mRNAs bearing a photoactivable 4-thiouridine probe in the first position of the codon located in the A site. Addition of eRF1 to the phased mRNA-ribosome complexes triggers a codon-dependent quenching of crosslink formation. The concentration of eRF1 triggering half quenching ranges from low for the three stop codons, to intermediate for s4UGG and high for other near-cognate triplets. A theoretical analysis of the photochemical processes occurring in a two-state bimolecular model raises a number of stringent conditions, fulfilled by the system studied here, and shows that in any case sound KD values can be extracted if the ratio mT/KD<<1 (mT is total concentration of mRNA added). Considering the KD values obtained for the stop, s4UGG and sense codons (approximately 0.06 microM, 0.45 microM and 2.3 microM, respectively) and our previous finding that only the stop and s4UGG codons are able to promote formation of an eRF1-mRNA crosslink, implying a role for the NIKS loop at the tip of the N domain, we propose a two-step model for eRF1 binding to the A site: a codon-independent bimolecular step is followed by an isomerisation step observed solely with stop and s4UGG codons. Full recognition of the stop codons by the N domain of eRF1 triggers a rearrangement of bound eRF1 from an open to a closed conformation, allowing the universally conserved GGQ loop at the tip of the M domain to come into close proximity of the peptidyl transferase center of the ribosome. UGG is expected to behave as a cryptic stop codon, which, owing to imperfect eRF1-codon recognition, does not allow full reorientation of the M domain of eRF1. As far as the physical steps of eRF1 binding to the ribosome are considered, they appear to closely mimic the behaviour of the tRNA/EF-Tu/GTP complex, but clearly eRF1 is endowed with a greater conformational flexibility than tRNA.