The synthesis of the alternating copolymer poly(r(A-s4U)) by RNA polymerase of Escherichia coli

Conformation and structural fluctuations of a 218 nucleotides long rRNA fragment: 4-thiouridine as an intrinsic photolabelling probe

The structure in solution of an RNA fragment (218 nucleotides long) containing part of E. coli 16S rRNA domain 2 has been studied using the intrinsic photoaffinity probe 4-thiouridine (s4U). In vitro transcription with T7 polymerase, in the presence of s4U triphosphate yielded complete RNA molecules. An affinity electrophoresis system based on Phenylmercuric substituted polyacrylamide (APM) gels allows separation of the RNA chains as a function of their s4U content. Distribution of s4U within chains follows a binomial law indicating that (i) substitution is close to random, (ii) efficiency of s4U incorporation is 0.22 times that of U. The monothiolated RNA fraction isolated from APM gel was irradiated at 366 nm under native conditions and the intramolecularly crosslinked molecules, (34%), were separated on denaturing polyacrylamide gel according to loop size. The positions of the two partners of bridges were identified by mean of reverse transcription and RNA sequencing. 17 of the 41 possible s4U positions lead to detectable bridges. These crosslinks formed efficiently at the border of bihelical regions or when structural mobility is allowed. The pattern of crosslinks is in agreement with the previously proposed secondary structure but indicates that it is much more flexible than expected.

Substitution of uridine in vivo by the intrinsic photoactivable probe 4-thiouridine in Escherichia coli RNA. Its use for E. coli ribosome structural analysis

In vivo incorporation of the uridine-photoactivable analogue, 4-thiouridine, into the ribosomal RNA of an Escherichia coli pyrD strain has been demonstrated. It is highly dependent on the exogenous uridine and 4-thiouridine concentrations as well as on temperature. We have defined conditions allowing the substitution of 13 +/- 2% of the uridine residues in bulk RNA by 4-thiouridine. On a high-Mg2+ sucrose gradient, 33 +/- 3% of ribonucleic particles sediment as 70S ribosomes, the remaining being in the form of non-associated 50S and 30S particles containing immature rRNA. The thiolated 70S ribosomes tolerate a 4-5% substitution level (40 thiouridine molecules/particle). Surprisingly, 3-4% of ribosomal proteins, about two protein molecules/particle, were spontaneously covalently bound to 4-thiouridine-substituted rRNA. Specific 366-nm photoactivation increased this proportion to 10-12%, i.e. up to six or seven ribosomal protein molecules/particle. The photochemical cross-linking proceeds with apparent first-order kinetics with a quantum yield close to 5 X 10(-3). Although extensive photodynamic breakage of rRNA occurs under aerobic conditions, both the kinetics and yield of ribosomal protein cross-linking were independent of oxygenation conditions. The thiolated (4.5%) 70S ribosomes allowed the poly(U)-directed poly(Phe)synthesis at 48% the control rate. Photoactivation decreased this activity to 28% and 10% when performed under nitrogen and in aerated conditions, respectively.

The effect of nucleotide analogs on cell-free gene expression

The effects of the following pyrimidine nucleoside 5'-triphosphates: f5 UTP, br5 UTP, rTTP, s2 UTP, s4 UTP and s2 CTP on cell-free expression of the beta-galactosidase gene in lambda h80dlac DNA as well as the galactokinase gene in plasmid 01-14 were investigated. Only rTTP could substitute UTP in cell-free gene expression without restriction. Combinations of the other analogs with their respective natural congeners led to inhibition of gene expression. All analogs were found to inhibit transcription. Whereas br5 UTP and s4 UTP did not affect translation, mRNA containing s2 UMP or s2 CMP residues respectively was found to function poorly in translation. Only in the case of f5 UTP could ambiguitive behaviour be demonstrated. Whether mispairing of f5 UMP residues, responsible for this ambiguity takes place in transcription or in translation, could not be decided.

Properties of P3 esters of nucleoside triphosphates as substrates for RNA polymerase from Escherichia coli

P3-[(2,4-Dinitrophenyl)amino]ethyl (DNPNHEt) and P3-methyl phosphate esters of nucleoside 5'-triphosphates have been synthesized. Their properties as substrates in the initiation and elongation steps of transcription have been examined by using RNA polymerase from Escherichia coli and poly[d(A-T)] or T7 DNA as templates. It is shown that transcription can be initiated by ATP-EtNHDNP and that 2,4-dinitrophenyl residues are incorporated at the 5' end of the RNA molecules. Steady-state kinetic experiments of abortive initation on promoters A1 and A3 of T7 DNA revealed that ATP-EtNHDNP, ADP-EtNHDNP, and ATP-OCH3 have lower Km values and markedly reduced Vmax values compared to those of ATP. The two classes of esters, NTR-EtNHDNP and NTP-OCH3, were found to differ regarding their utilization as substrates for elongation. Both ATP-OCH3 and UTP-OCH3 are substrates for transcription. However, only the pyrimidine derivatives of NTP-EtNHDNP are elongation substrates which release DNPNHEt-PP upon utilization. This dramatic difference between the purine and pyrimidine derivatives of NTP-EtNHDNP reflects a selective process in the transcriptional complex for purines and pyrimidines.

The stereochemical basis of template function

The behavior of nucelotides with thioketo-substituted pyrimidine bases (4-thiouracil, 2-thiouracil and 2-thiocytosine) or amino-analogue purine bases (2-aminopurine and 2,6-diaminopurine) in transcription and translation was investigated. The experimental results obtained led to the following conclusions. 1. The stereochemical basis of substrate selection in transcription is the geometry of Watson-Crick base pairs A-U (or A-T) and G-C between substrate and template bases. 2. The topology of the active site of Escherichia coli RNA polymerase is precisely adopted to the geometry of Watson-Crick base pairs. 3. The enzyme active site discriminates between A-U (A-T) and G-C base pairs. An essential feature in this discrimination is the 6-NH2 group of the A-U (A-T) base pair and the 2-keto group of cytosine in the G-C base pair. 4. The codon properties of a nucleic acid base in messenger RNA can be predicted on the basis of its specificity in polynucleotide interactions. There seems to be no evidence for the participation of protein topological sites in the control of the specificity of codon-anticodon interactions in translation.

Induction of interferon by polyribonucleotides containing thiopyrimidines

The influence of thioketo substitution in pyrimidine bases of double-stranded polynucleotides on interferon induction was investigated. The stabilizing effect of 2-thioketo substitution was reflected in the increased interferon inducing activity of poly(A-s2U) over that of poly(A-U). Poly(A-s2U) and poly(I)-poly(s2C) were as effective as poly(I)-Poly(C) in rabbit cells. Poly(I)-poly(C) and poly(I)-poly(s2C) were compared in several animal species. No differences in biological effects were observed in rabbits and dogs. In rodents, poly(I)-poly(s2C) was less effective and less toxic.Poly(I)-poly(s2C) was highly resistant against degradation by human serum. Further investigations seem to be justified to elucidate whether this property offers any advantages for the potential clinical utilization of poly(I)-poly(s2C).

The effect of thioketo substitution on uracil-2-aminopurine and uracil-2, 6-diaminopurine interactions in polynucleotides

The existence of the complexes poly[r(s4U)] . poly[r(n2h6A] and poly-[r(s4U)] . poly[r(n2A)] was demonstrated by means of spectrophotometric titration and sedimentation veolicty analysis. According to the absorption-temperature profiles thioketo substitution of poly[r(U)] . poly[r(n2h6A)] led to stabilisation of the helical structure, thus implying that the 4-thioketo group does not participate in s4U . n2h6A base pairing. In the case of poly[r(s4U)] . poly[r(n2A)] drastic destabilistaion of the helical structure by thioketo substitution was observed. This indicates that the thioketo substituents participate in s4U .n2A base pairing.

The effects of thioketo substitution upon uracil-adenine interactions in polyribonucleotides. Synthesis and properties of the alternating polynucleotides poly(r(A-s2U)) and poly(r(A-s2s4U))

The polynucleotides poly[r(A-s-2U)] and poly]r(A-s2s4U)] have been synthesized and characterized by nearest-neighbour analysis, sedimentation analysis as well as spectroscopic techniques. Absorption-temperature profile and absorption-pH profile of poly[r(A-s-2U)] did not reveal a structural transition between 10 and 95 degrees C even at low ionic strength, although a variety of properties indicated a helical structure of poly[r(A-s-2U)]: remarkable hyperchromicity of the absorption spectrum, circular dichroic spectrum displaying extrema of large amplitudes, resistance against hydrolysis by ribonuclease and interaction with ethidium bromide in a manner which is characteristic of helical polynucleotides. Our results show that interactions of the type A-s-2U and A-s-2s-4U do in fact exist in helical polynucleotides. The properties of poly]r(As-2U)] furthermore demonstrate the general stabilizing effect of 2-thioketopyrimidine bases in helical polynucleotides by virtue of vertical stacking interactions with neighbouring pyrimiding and purine bases.

Properties of ATP and UTP analogues with P-S-C-5' bonds

Analogues of ATP and UTP bearing C-5'-S-P ester bonds were found not to be substrates but weak competitive inhibitors of Escherichia coli RNA polymerase. The K-i values of the analogues obtained in the transcription of poly[d(A-T)] or poly(dT) under various conditions are in the order of millimolar. Evidence was derived from proton magnetic resonance spectra that nucleotides with C-5'-S-P bonds do not exist in gauche-gauche conformation normally adopted by natural occurring nucleotides. This leads us to assume that the gauche-gauche conformation is an essental prerequisite for substrates of RNA polymerase. Ado-5'-S-PPP substituted for ATP as substrate of hexokinase from yeast rather effectively thus indicating that a distinct stereochemical orientation of the alpha-phosphate ester bond is not a stringent requirement for substrates of this enzyme.