Gérard Lancelot

A new NMR solution structure of the SL1 HIV-1Lai loop-loop dimer

Dimerization of genomic RNA is directly related with the event of encapsidation and maturation of the virion. The initiating sequence of the dimerization is a short autocomplementary region in the hairpin loop SL1. We describe here a new solution structure of the RNA dimerization initiation site (DIS) of HIV-1Lai. NMR pulsed field-gradient spin-echo techniques and multidimensional heteronuclear NMR spectroscopy indicate that this structure is formed by two hairpins linked by six Watson–Crick GC base pairs. Hinges between the stems and the loops are stabilized by intra and intermolecular interactions involving the A8, A9 and A16 adenines. The coaxial alignment of the three A-type helices present in the structure is supported by previous crystallography analysis but the A8 and A9 adenines are found in a bulged in position. These data suggest the existence of an equilibrium between bulged in and bulged out conformations in solution.

Synthesis and structural studies of a self-complementary decadeoxynucleotide d(AATTGCAATT) I. - Synthesis and chemical characterization of the decanucleotide

The synthesis of the self-complementary decadeoxynucleotide d(AATTGCAATT) is described. The phosphotriester method has been used with several modifications. Protected nucleotides have been prepared in a one-step reaction involving a new monofunctional phosphorylating agent: p-chlorophenyl-beta-cyanoethyl phosphate. Triethylammonium salts of mononucleoside 3-phosphodiesters were obtained either by decyanoethylation of the triesters or, in the case of thymine, by a one-step reaction starting from 5-0-methoxytritylthymidine and the mixture pyridine-para-chlorophenyl-methyl-phosphorobromidate. The usual coupling reactions were then used to prepare the decadeoxynucleotide in large quantities.

HIV-1(Lai) genomic RNA: combined used of NMR and molecular dynamics simulation for studying the structure and internal dynamics of a mutated SL1 hairpin.

Abstract. The genome of all retroviruses consists of two identical copies of an RNA sequence associated in a non-covalent dimer. A region upstream from the splice donor (SL1) comprising a self-complementary sequence is responsible for the initiation of the dimerization. This region is able to dimerize in two conformations: a loop-loop complex or an extended duplex. Here, we solve by 2D NMR techniques the solution structure of a 23-nucleotide sequence corresponding to HIV-1 SL1Lai in which the mutation G12→A12 is included to prevent dimerization. It is shown that this monomer adopts a stem-loop conformation with a seven base pairs stem and a nine nucleotide loop containing the G10xa0C11xa0A12xa0C13xa0G14xa0C15 sequence. The stem is well structured in an A-form duplex, while the loop is more flexible even though elements of structure are evident. We show that the structure adopted by the stem can be appreciably different from its relaxed structure when the adenines A8, A9 and A16 in the loop are mechanically constrained. This point could be important for the efficiency of the dimerization. This experimental study is complemented with a 10xa0ns molecular dynamics simulation in the presence of counterions and explicit water molecules. This simulation brings about information on the flexibility of the loop, such as a hinge motion between the stem and the loop and a labile lattice of hydrogen bonds in the loop. The bases of the nucleotides G10 to C15 were found outside of the loop during a part of the trajectory, which is certainly necessary to initiate the dimerization process of the genuine SL1Lai sequence.

A new peculiar DNA structure: NMR solution structure of a DNA kissing complex

Abstract The deoxyoligoribonucleotide d(CTTGCTGAAGCGCGCACGGCAAG) (dSL1) corresponding to the reverse transcripted sequence of the dimerization initiation site SL1 of HIV-1Lai RNA was synthesized using phosphoramidite chemistry. Like its oligoribonucleotide counterpart, dSL1 dimerized spontaneously in solution. Here we report the first NMR solution structure of a kissing complex formed with two DNA strands. The melting point of the DNA dimer (35 °C) was found slightly higher than the one of the corresponding RNA dimer (32 °C). Despite this only slight difference in melting point, several structural differences were observed between the ribo- and the deoxyribo- dimers. The solution structure of the deoxy- dimer was a symmetric homodimer with a loop-loop interaction stabilized by four central G-C base-pairs, a head to tail A-A base-pair arrangement between the A8 residues of the two strands and a stacking of A9 with C15. As a consequence, G10 was not paired and occupied a position outside the stem and the loop. Each stem was formed by seven base-pairs whose axis made an angle of about 100° with the plane of the loops. The distortion of the helix at the junction of the stem and of the loop induced a fold up of the A8pA9 step with a phosphate-phosphate distance lowered to 4.5Å. The plane of the non-canonical A-A base-pair was oriented perpendicularly to the axis of the stems. The four central base-pairs formed an open fan-shaped motif with an angle of 20° between the bases and each of them was oriented perpendicularly to the A8-A8 plane. The deviation of the computed chemical shifts and the experimental ones for the aromatic proton was always less than 0.25ppm for each of the 16 converged solution structures and their average less than 0.1ppm.

Modeling the dynamics of a mutated stem-loop in the SL1 domain of HIV-1Lai genomic RNA by 1H-NOESY spectra

The cross-peaks of 1H-NOESY spectra at different time delays are compared to a mode-coupling diffusion (MCD) calculation, including the evaluation of the full 1H relaxation matrix, in the case of a 23 nucleotide fragment of the stem-loop SL1 domain of HIV-1Lai genomic RNA mutated in a single position. The MCD theory gives significant agreement with 1H relaxation experiments enabling a thorough understanding of the differential local dynamics along the sequence and particularly of the dynamics of nucleotides in the stem and in the loop. The differential dynamics of this hairpin structure is important in directing the dimerization of the retroviral genome, a fundamental step in the infectious process. The demonstration of a reliable use of time dependent NOE cross-peaks, largely available from NMR solution structure determination, coupled to MCD analysis, to probe the local dynamics of biological macromolecules, is a result of general interest of this paper.

Synthesis and structural studies of a self-complementary decadeoxynucleotide d(AATTGCAATT) II. - Proton magnetic resonance studies

Proton magnetic resonance spectra of the self-complementary decadeoxynucleotide d(AATTGCAATT) at 90 and 250 MHz have been obtained at different temperatures. The assignment of the different resonance lines to the base protons was obtained by combining the data derived from various methods: hydrogen in equilibrium with deuterium exchange at the H8 position of purines; comparison of NMR spectra obtained at high temperature with those of mononucleotides; comparison of the variations in chemical shifts obtained between 280 K and 360 K with calculated values; determination of half-transition temperatures for each base pair. On the basis of computed chemical shifts for stacked base-pairs it is concluded that the decadeoxynucleotide duplex exists in the B form in solution at 280 K. Propagation of the opening of the mini double helix from terminal to central base pairs if reflected in the variation of half-transition temperatures which vary between 306 K and 327 K.

NMR Studies of lac Operator and lac Repressor

Abstract For many years, and even now, the lac operon of Escherichia coli has been a model for gene regulation. The lac repressor is a tetrameric protein that binds strongly the lac operator and the affinity of which is modulated by an inducer molecule. The lac repressor– lac operator complex was the first nucleic acid–protein complex studied with success by NMR. The NMR studies began in 1973 and gave publications until now. 1 H, 13 C, 15 N, 19 F, 31 P NMR and photo-CIDNP studies increased progressively the level of understanding of the system during these decades. At first, it appeared to be possible to duplicate the basic lac repressor– lac operator interaction by using a smaller protein: the lac repressor N-terminal fragment. In 1987, the Kapteins group reported the first model of interaction where the motif helix-turn-helix plays a central role in the recognition process of the large groove of the operator by the protein. Later, heteronuclear NMR studies refined the NMR solution structure of both free protein and complexed protein. 13 C and 15 N relaxation rate measurements showed dynamic processes in both nucleic acid and protein moieties during the complexation. By studying a larger headpiece (1-62) Kapteins group showed the fundamental role played by the hinge helix, which interacts, in the minor groove of the DNA. The protection factor of a dimeric lac repressor headpiece mutant, which has a high affinity for DNA, was measured and used to qualify the folding of this protein during the complexation. The role played by the asymmetry of the lac operator was studied and explained. NMR studies allowed proposing a scheme to explain the loss in affinity of the lac repressor for its operator when binding to an inducer molecule. Comparison of these data with crystal structures showed that, until now, the more detailed view of the lac repressor– lac operator interaction comes from NMR studies. The NMR solution structures and the dynamic processes revealed by relaxation are in agreement with all the genetic and biochemical data.

THE RECOGNITION OF NUCLEIC ACID STRUCTURES AND BASE SEQUENCES BY PROTEINS. ROLE OF STACKING AND HYDROGEN BONDING INTERACTIONS

Publisher Summary Associations of proteins with nucleic acids are required at every step of DNA replication, repair, and transcription and of messenger RNA translation on ribosomes. The recognition of nucleic acid structures and base sequences by proteins requires a structural complementarity between the interacting regions of the two macromolecules. It also requires the formation of non-covalent bonds between the chemical groups of both partners in the complex. Both nucleic acids and proteins possess a large number of chemical groups that are able to act as hydrogen bond donors or acceptors. Some of the amino acid side chains have the capability of forming a pair of hydrogen bonds with nucleic acid bases.