Affinities of packaging domain loops in HIV-1 RNA for the nucleocapsid protein

To design anti-nucleocapsid drugs, it is useful to know the affinities the protein has for its natural substrates under physiological conditions. Dissociation equilibrium constants are reported for seven RNA stem-loops bound to the mature HIV-1 nucleocapsid protein, NCp7. The loops include SL1, SL2, SL3, and SL4 from the major packaging domain of genomic RNA. The binding assay is based on quenching the fluorescence of tryptophan-37 in the protein by G residues in the single-stranded loops. Tightly bound RNA molecules quench nearly all the fluorescence of freshly purified NCp7 in 0.2 M NaCl. In contrast, when the GGAG-tetraloop of tight-binding SL3 is replaced with UUCG or GAUA, quenching is almost nil, indicating very low affinity. Interpreting fluorescence titrations in terms of a rapidly equilibrating 1:1 complex explains nearly all of the experimental variance for the loops. Analyzed in this way, the highest affinities are for 20mer SL3 and 19mer SL2 hairpin constructs (K(d) = 28 +/- 3 and 23 +/- 2 nM, respectively). The 20mer stem-UUCG-loop and GAUA-loop constructs have <0.5% of the affinity for NCp7 relative to SL3. Affinities relative to SL3 for the other stem-loops are the following: 10% for a 16mer construct to model SL4, 30% for a 27mer model of the 9-residue apical loop of SL1, and 20% for a 23mer model of a 1 x 3 asymmetric internal loop in SL1. A 154mer construct that includes all four stem-loops binds tightly to NCp7, with the equivalent of three NCp7 molecules bound with high affinity per RNA; it is also possible that two strong sites and several weaker ones combine to give the appearance of three strong sites.     

HIV-1 nucleocapsid protein NCp7 and its RNA stem loop 3 partner: rotational dynamics of spin-labeled RNA stem loop 3

The tumbling dynamics of a 20-mer HIV-1 RNA stem loop 3 spin-labeled at the 5' position were probed in the nanosecond time range. This RNA interacted with the HIV-1 nucleocapsid Zn-finger protein, 1-55 NCp7, and specialized stopped-flow EPR revealed concomitant kinetics of probe immobilization from milliseconds to seconds. RNA stem loop 3 is highly conserved in HIV, while NCp7 is critical to HIV-RNA packaging and annealing. The 5' probe did not perturb RNA melting or the NCp7/RNA interaction monitored by gel shift and fluorescence. The 5'-labeled RNA tumbled with a subnanosecond isotropic correlation time (approximately 0.60 ns at room temperature) reflecting both local viscosity-independent bond rotation of the probe and viscosity-dependent diffusion of 40-60% of the RNA. The binding of NCp7 to spin-labeled RNA stem loop 3 in a 1:1 ratio increased the spin-labeled tumbling time by about 40%. At low ionic strength with a ratio of NCp7 to RNA >or=3 (i.e., an NCp7 to nucleotide ratio or=3:1 complex also required intact Zn fingers. Stopped-flow EPR kinetics with NCP7/RNA mixed at a 4:1 ratio showed the major phase of NCp7 interaction with RNA stem loop 3 occurred within 4 ms, a second phase occurred with a time constant of approximately 30 ms, and a slower immobilization, possibly concomitant with large complex formation, proceeded over seconds. This work points the way for spin-labeling to investigate oligonucleotide-protein complexes, notably those lacking precise stoichiometry, that are requisite for viral packaging and genome fabrication.     

Mechanism of zinc coordination by point-mutated structures of the distal CCHC binding motif of the HIV-1 NCp7 protein

The kinetics of Zn(2+) binding by two point-mutated forms of the HIV-1 NCp7 C-terminal zinc finger, each containing tridentate binding motif HCC [Ser49(35-50)NCp7] or CCC [Ala44(35-50)NCp7], has been studied by stopped-flow spectrofluorimetry. Both the formation and dissociation rate constants of the complexes between Zn(2+) and the two model peptides depend on pH. The results are interpreted on the basis of a multistep reaction model involving three Zn(2+) binding paths due to three deprotonated states of the coordinating motif, acting as monodentate, bidentate, and tridentate ligands. For Ser49(35-50)NCp7 around neutral pH, binding preferentially occurs via the deprotonated Cys36 in the bidentate state also involving His44. The binding rate constants for the monodentate and bidentate states are 1 x 10(6) and 3.9 x 10(7) M(-)(1) s(-)(1), respectively. For Ala44(35-50)NCp7, intermolecular Zn(2+) binding predominantly occurs via the deprotonated Cys36 in the monodentate state with a rate constant of 3.6 x 10(7) M(-)(1) s(-)(1). In both mutants, the final state of the Zn(2+) complex is reached by subsequent stepwise ligand deprotonation and intramolecular substitution of coordinated water molecules. The rate constants for the intermolecular binding paths of the bidentate and tridentate states of Ala44(35-50)NCp7 and of the tridentate state of Ser49(35-50)NCp7 are much smaller than expected according to electrostatic considerations. This is attributed to conformational constraints required to achieve proper metal coordination during folding. The dissociation of Zn(2+) from both peptides is again characterized by a multistep process and takes place fastest via the protonated Zn(2+)-bound bidentate and monodentate states, with rate constants of approximately 0.3 and approximately 10(3) s(-)(1), respectively, for Ser49(35-50)NCp7 and approximately 4 x 10(-)(3) and approximately 500 s(-)(1), respectively, for Ala44(35-50)NCp7.     

Effects of temperature on the dynamic behaviour of the HIV-1 nucleocapsid NCp7 and its DNA complex

The nucleocapsid protein NCp7 of human immunodeficiency virus type 1 (HIV-1) contains two highly conserved CCHC zinc fingers and is involved in many crucial steps of the virus life-cycle. A large number of physiological r?les of NCp7 involve its binding to single-stranded nucleic acid chains. Several solution structures of NCp7 and its complex with single-stranded RNA or DNA have been reported. We have investigated the changes in the dynamic behaviour experienced by the (12-53)NCp7 peptide upon DNA binding using (15)N heteronuclear relaxation measurements at 293 K and 308 K, and fluorescence spectroscopy. The relaxation data were interpreted using the reduced spectral density approach, which allowed the high-frequency motion, overall tumbling rates and the conformational exchange contributions to be characterized for various states of the peptide without using a specific motional model. Analysis of the temperature-dependent correlation times derived from both NMR and fluorescence data indicated a co-operative change of the molecular shape of apo (12-53)NCp7 around 303 K, leading to an increased hydrodynamic radius at higher temperatures. The binding of (12-53)NCp7 to a single-stranded d(ACGCC) pentanucleotide DNA led to a reduction of the conformational flexibility that characterized the apo peptide. Translational diffusion experiments as well as rotational correlation times indicated that the (12-53)NCp7/d(ACGCC) complex tumbles as a rigid object. The amplitudes of high-frequency motions were restrained in the complex and the occurrence of conformational exchange was displaced from the second zinc finger to the linker residue Ala30.     

Interactions of the C-terminus of viral protein R with nucleic acids are modulated by its N-terminus.

The basic viral protein R (Vpr) performs several functions during the human immunodeficiency virus HIV-1 retroviral cycle, including G2 mitosis arrest and nuclear import of the preintegration complex allowing lentivirus to replicate in nondividing cells. Accordingly, this protein was found in the nucleus of infected cells. In the virus, Vpr is incorporated through interaction with both nucleocapsid protein 7 (NCp7) and p6, two small proteins encoded by the C-terminal part of the Gag precursor. NCp7 is also involved in genomic RNA encapsidation during the budding process suggesting a possible interaction of Vpr with nucleic acids, either directly or via the NCp7 intermediate. Gel shift experiments were carried out with RNA and DNA using synthetic Vpr and peptide derivatives. The results show that Vpr binds to nucleic-acid inducing aggregates. This process, which requires the C-terminal basic domain of the protein (in particular the helical 70-80 domain), is regulated by the N-terminal region of Vpr. Moreover, NCp7 was shown to enhance RNA recognition by Vpr, a feature that could be required for Vpr encapsidation and during nuclear import of the preintegration complex.     

Analysis of the nucleic acid annealing activities of nucleocapsid protein from HIV-1.

Retroviral nucleocapsid (NC) protein is an integral part of the virion nucleocapsid where it is in tight association with genomic RNA and the tRNA primer. NC protein is necessary for the dimerization and encapsidation of genomic RNA, the annealing of the tRNA primer to the primer binding site (PBS) and the initial strand transfer event. Due to the general nature of NC protein-promoted annealing, its use to improve nucleic acid interactions in various reactions can be envisioned. Parameters affecting NC-promoted nucleic acid annealing of NCp7 from HIV-1 have been analyzed. The promotion of RNA:RNA and RNA:DNA annealing by NCp7 is more sensitive to the concentration of MgCl2 than the promotion of DNA:DNA hybridization. Stimulation of complex formation for all three complexes was efficient at 0-90 mM NaCl, between 23 and 55 degrees C and at pH values between 6.5 and 9.5, inclusive. Parameters affecting NCp7-promoted hybridization of tRNA(Lys,3) to the PBS, which appears to be specific for NC protein, will be discussed. Results implicate the basic regions of NCp7, but not the zinc fingers, in promoting the annealing of complementary nucleic acid sequences. Finally, NCp7 strand transfer activity aids the formation of the most stable nucleic acid complex.     

Phosphorescence and optically detected magnetic resonance of HIV-1 nucleocapsid protein complexes with stem-loop sequences of the genomic Psi-recognition element

The binding of NCp7, the nucleocapsid protein of human immunodeficiency virus type 1, to oligonucleotide stem--loop (SL) sequences of the genomic Psi-recognition element has been studied using fluorescence, phosphorescence, and optically detected magnetic resonance (ODMR). RNA SL2, SL3, and SL4 constructs bind with higher affinity than the corresponding DNAs. G to I substitutions in the SL3 DNA loop sequence lead to reduced binding affinity and significant changes in the triplet state properties of Trp37 of NCp7, implicating these bases in contacts with aromatic amino acid residues of the zinc finger domains of NCp7, in agreement with the NMR structure of the 1:1 complex of NCp7 and SL3 RNA [DeGuzman, R. N., Wu, Z. R., Stalling, C. C., Pappaladro, L., Borer, P. N., and Summers, M. F. (1998) Science 279, 384-388]. The NCp7 to SL binding stoichiometry is 2:1 for intact SL sequences but is reduced to 1:1 for SL variants with an abasic or hydrocarbon loop. It is proposed that Delta D/Delta E(0,0), where Delta D is the change in the zero-field splitting D parameter and Delta E(0,0) is the shift of the tryptophan phosphorescence origin, provides a measure of aromatic stacking interactions with nucleic acid bases. Values on the order of 10(-5) indicate significant stacking interactions, while values closer to 10(-6) result from interactions not involving aromatic stacking. Binding of NCp7 to oligonucleotide substrates produces shortened Trp37 triplet state lifetimes by enhancement of k(x) and an increase of the relative value of P(x), the intersystem crossing rate to the T(x) sublevel. These effects are attributed to a reduction in the degree of electronic symmetry of Trp37 in the complexes. Guanine and adenine triplet states produced by optical pumping of SL3 DNA are characterized. We find, as with tryptophan, that D < 3E.     

Interactions of ribosomal protein S1 with DsrA and rpoS mRNA

Ribosomal protein S1 is shown to interact with the non-coding RNA DsrA and with rpoS mRNA. DsrA is a non-coding RNA that is important in controlling expression of the rpoS gene product in Escherichia coli. Photochemical crosslinking, quadrupole-time of flight tandem mass spectrometry, and peptide sequencing have identified an interaction between DsrA and S1 in the 30S ribosomal subunit. Purified S1 binds both DsrA (K(obs) approximately 6 x 10(6) M(-1)) and rpoS mRNA (K(obs) approximately 3 x 10(7) M(-1)). Ribonuclease probing experiments indicate that S1 binding has a weak but detectable effect on the secondary structure of DsrA or rpoS mRNA.     

Specific binding of HIV-1 nucleocapsid protein to PSI RNA in vitro requires N-terminal zinc finger and flanking basic amino acid residues.

The nucleocapsid (NC) protein of human immunodeficiency virus HIV-1 (NCp7) is responsible for packaging the viral RNA by recognizing a packaging site (PSI) on the viral RNA genome. NCp7 is a molecule of 55 amino acids containing two zinc fingers, with only the first one being highly conserved among retroviruses. The first zinc finger is flanked by two basic amino acid clusters. Here we demonstrate that chemically synthesized NCp7 specifically binds to viral RNA containing the PSI using competitive filter binding assays. Deletion of the PSI from the RNA abrogates this effect. The 35 N-terminal amino acids of NCp7, comprising the first zinc finger, are sufficient for specific RNA binding. Chemically synthesized mutants of the first zinc finger demonstrate that the amino acid residues C-C-C/H-C/H are required for specific RNA binding and zinc coordination. Amino acid residues F16 and T24, but not K20, E21 and G22, located within this zinc finger, are essential for specific RNA binding as well. The second zinc finger cannot replace the first one. Furthermore, mutations in the basic amino acid residues flanking the first zinc finger demonstrate that R3, 7, 10, 29 and 32 but not K11, 14, 33 and 34 are also essential for specific binding. Specific binding to viral RNA is also observed with recombinant NCp15 and Pr55Gag. The results demonstrate for the first time specific interaction of a retroviral NC protein with its PSI RNA in vitro.     

NMR structure of the complex between the zinc finger protein NCp10 of Moloney murine leukemia virus and the single-stranded pentanucleotide d(ACGCC): comparison with HIV-NCp7 complexes.

The structure of the 56 amino acid nucleocapsid protein NCp10 of retrovirus MoMuLV, which contains a single CX(2)CX(4)HX(4)C-type zinc finger, has been determined previously by NMR. The important role of NCp10 (or NCp7 for HIV-1) in the retroviral life cycle seems mainly related to their preferential binding to single-stranded nucleic acids. We report here the structure of the complex formed between the biologically active (14-53)NCp10 and the oligonucleotide d(ACGCC) in aqueous solution determined by 2D (1)H NMR based methods. The aromatic residue Trp(35) of NCp10 directs nucleic acid complexation as shown by its complete fluorescence quenching upon addition of d(ACGCC). (1)H and (31)P NMR studies support the insertion of Trp(35) between the G(3) and C(4) bases. A total of 577 NOE distance restraints, of which 40 were intermolecular, were used for the structure determination. The zinc finger provides a well-defined surface for the binding of d(ACGCC) through hydrophobic interactions and tryptophan stacking on the guanine. This latter interaction was also observed in the NMR-derived structures of the complexes between NCp7, which contains two successive zinc fingers, and single-stranded DNA and RNA, supporting the proposal for a major role played by aromatic residues of NCp proteins in nucleic acid recognition. Upon binding to the nucleotide a new loop in NCp10 that participates in the intermolecular interaction is formed. Additional interactions provided by positively charged residues surrounding the zinc finger appear necessary for tight binding. The structure of the complex NCp10-d(ACGCC) gives a structural explanation for the loss of virus infectivity following point mutations in the finger domain.     

Ordered aggregation of ribonucleic acids by the human immunodeficiency virus type 1 nucleocapsid protein

The nucleocapsid protein NCp7, which is the major genomic RNA binding protein of human immunodeficiency virus type 1, plays an important role in several key steps of the viral life cycle. Many of the NCp7 activities, notably the nucleic acid annealing and the genomic RNA wrapping ones, are thought to be linked to a nonspecific binding of NCp7 to its nucleic acid targets. The mechanism of these activities is still debated but several clues are in favor of an intermediate aggregation of nucleic acids by NCp7. To check and characterize the nucleic acid aggregating properties of NCp7, we investigated the interaction of NCp7 with the model RNA homopolymer, polyA, by quasielastic light scattering and optical density measurements. The ordered growth of monodisperse large particles independently of the nucleic acid size and the almost complete covering of polyA by NCp7 strongly suggested an ordered aggregation mechanism. The aggregate kinetics of growth in the optimum protein concentration range (≥2 μM) were governed by a so-called Ostwald ripening mechanism limited by transfer of NCp7-covered polyA complexes from small to large aggregates. The aggregation process was strongly dependent on both Na⁺ and Mg²⁺ concentrations, the optimum concentrations being in the physiological range. Similar conclusions held true when polyA was replaced by 16S + 23S ribosomal RNA, suggesting that the NCp7 aggregating properties were only poorly dependent on the nucleic acid sequence and structure. Finally, as in the NCp7 annealing activities, the basic regions of NCp7, but not the zinc fingers, were found critical in nucleic acid aggregation. Taken together, our data indicate that NCp7 is a highly efficient nucleic acid aggregating agent and strengthen the hypothesis that aggregation may constitute a transient step in various NCp7 functions. © 1997 John Wiley & Sons, Inc.     

Solid phase synthesis of the retroviral nucleocapsid protein NCp10 of Moloney murine leukaemia virus and related "zinc-fingers" in free SH forms. Influence of zinc chelation on structural and biochemical properties.

The core of retroviruses contains a highly conserved, low molecular weight, basic protein that binds nucleic acids and is essential for genomic RNA packaging. The 56 amino acid protein, NCp10, of Moloney Murine Leukaemia virus (MoMuLV) has the CysX2 CysX4 HisX4 Cys zinc finger-like motif shared by all retrovirus nucleocapsid proteins. The native protein and five modified peptides containing the zinc binding domain were synthesized by solid phase in order to investigate the structural and biochemical role of Zn2+ chelation in MoMuLV NCp10 activity. The purity of the synthetic molecules was verified by HPLC and their sequences were confirmed by amino acid analysis and sequencing in the case of NCp10. Thiol dosage agreed with the theoretical value of free cysteine for all these molecules. Fluorescence measurements performed on synthetic NCp10 and zinc finger fragments showed that the tryptophan quantum yield was Zn2(+)-dependent, allowing a 1:1 stoichiometry for the complex to be determined. The apparent affinity constant of NCp10 for the metal was estimated to be superior to 10(6) M-1. The synthetic protein, in the presence of Zn2+ ions, possesses all the biological properties of NCp10 isolated from virions. It catalyzes both the MoMuLV RNA dimerization and the annealing of the replication primer tRNA(Pro) onto MoMuLV RNA.