The Solution Structure of Yeast tRNAPhe as Studied by Nuclear Overhauser Effects in NMR

Abstract

Recently, the imino proton spectrum of yeast tRNA^Phe has been assigned by means of the application of the nuclear Overhauser effect (NOE). In the present paper it will be shown that even for tRNA (MW 28000) connectivities between the imino proton spins can be observed using two-dimensional NOE spectroscopy. In this way the imino proton resonances of the D-stem region are assigned. The results are discussed in relation to those obtained by the classical one-dimensional nuclear Overhauser effect. It turns out that in 2D-NOE experiments connectivities from overlapping resonances can be observed which cannot be determined by one-dimensional Overhauser experiments. Moreover, the total assignment of the imino proton spectrum of yeast tRNA^Phe is used to relate the three-dimensional crystal structure of the tRNA to its solution structure. It is shown that the principle elements of the X-ray structure, i.e. the hydrogen bonding network and the stacking of the stems upon one another, are also found in solution. This is true for the presence as well as for the absence of magnesium ions. However, in absence of magnesium ions the tRNA structure appears to differ in details from that in the presence of magnesium ions. Finally, the influence of the elongation factor Tu from B.stearothermophilus on the tRNA structure is discussed.     

Thermodynamics of transfer ribonucleic acids: the effect of sodium on the thermal unfolding of yeast tRNAPhe.

The thermal unfolding of yeast phenylalanine-specific tRNA (tRNA^Phe) has been calorimetrically investigated at several salt concentrations in the absence of magnesium. Application of the deconvolution theory of macromolecular conformational transitions allows calculation of the thermodynamic parameters of unfolding. It is demonstrated that the unfolding of tRNA^Phe occurs in a sequential fashion and that four separate transitions or five macromolecular thermodynamic states exist in the temperature range 8–72°C under the experimental conditions of these studies (0.067–0.52M Na⁺). The enthalpy and entropy changes between states and the relative population of each state as a function of temperature and salt concentration have been obtained. Sodium stabilizes the low-temperature conformations of tRNA^Phe. The increase in the melting temperatures of each transition is shown to be linearly dependent on the logarithm of sodium concentration. These results allow calculation of the “phase” diagram for the transitions as a function of salt concentration.     

The distance between the anticodon loops of two tRNAs bound to the 70 S Escherichia coli ribosome.

Using singlet-singlet energy transfer, we have measured the distance between the anticodons of two transfer RNAs simultaneously bound to a messengerprogramed Escherichia coli 70 S ribosome. The fluorescent Y base adjacent to the anticodon of yeast tRNA_Y^Phe serves as a donor. A proflavine (Pf) chemically substituted for the Y base in tRNA_Pf^Phe serves as an acceptor. By exploiting the sequential binding properties of 70 S ribosomes for two deacylated tRNAs, we can fill the strong site with either tRNA_Y^Phe or tRNA_Pf^Phe and then the weak site with the other tRNA. In both cases donor quenching and sensitized emission of the acceptor are observed. Analysis of these results leads to an estimate for the Y-proflavine distance of 18 ± 2 Å. This distance is very short and suggests strongly that the two tRNAs are simultaneously in contact with adjacent codons of the message. Separate experiments show that binding of a tRNA to the weak site does not perturb the environment of the hypermodified base of a tRNA bound to the strong site. This supports the assignment of the strong site as the peptidyl site. It also indicates that binding of the second tRNA proceeds without a change in the anticodon structure of a pre-existing tRNA at the peptidyl site.     

Conformational peculiarities of rRNA inff supMet from E. coli as revealed by fluorescent methods

Conformational transitions in several individual tRNAs (tRNA _inff ^supMet , tRNA^Phe from E. coli, tRNA _inf1 ^supVal , tRNA^Ser, tRNA^Phe from yeast) have been studied under various environmental conditions. The binding isotherms studies for dyes-tRNA complexes exhibited similarities in conformational states of all tRNAs investigated at low ionic strength (0.01 M NaCl). By contrast, at high ionic strength (0.4 M NaCl or 2×10^-4 M Mg²⁺) a marked difference is found in structural features of tRNA _inff ^supMet as compared with other tRNAs used. The tRNA _inff ^supMet is the only tRNA species that does not reveal the strong type of complexes with ethidium bromide, acriflavine and acridine orange.     

Conformational change in yeast tRNAAsp

The structure of yeast tRNA^Asp in aqueous solutions has been analyzed in the light of results obtained from Raman spectra recorded at from 5 to 82°C and compared to those of tRNA^Phe. Firm evidence is given of a reversible conformation transition for tRNA^Asp at 20°C. This transition is observed for the first time in the tRNA series. The low-temperature conformation appears to have a more regular ribose–phosphate backbone and a more effective G base-stacking. This conformational change, which occurs essentially in the D loop, could be connected to the existence of two (A and B) crystal forms obtained depending on crystallization conditions. The melting temperatures, which are different for each base stacking in tRNA^Asp, lie in a range of about 70°C, much higher than for tRNA^Phe. This fact is interpreted by a higher ratio of G-C base pairs in tRNA^Asp.     

Isolation and chromatographic behaviour of phenylalanine tRNA from barley embryos

Two fractions of phenylalanine tRNA (tRNA^Phe₁ and tRNA^Phe₂) were purified by BD-cellulose and RPC-5 chromatography of crude tRNA isolated from barley embryos. Successive RPC-5 rechromatography runs of tRNA^Phe₂ showed its conversion into more stable tRNA^Phe₁, suggesting that the two fractions have essentially the same primary structure. Both tRNA^Phe₁ and tRNA^Phe₂ had about the same acceptor activity, but tRNA^Phe₂ was aminoacylated much faster than tRNA^Phe₁. RPC-5 chromatography of crude aminoacylated tRNA showed higher contents of phe-tRNA^Phe₂ than of phe-tRNA^Phe₁ but the ratio of these two fractions estimated by relative fluorescence intensity was about 1. Fluorescence spectra of tRNA^Phe from barley embryos suggest that it contains Y base similar to Y_w from wheat tRNA^Phe.     

NMR relaxation processes of 31P in macromolecules

³¹P-Nmr relaxation parameters (spin-lattice relaxation time, linewidth, and nuclear Overhauser effect) were obtained at three different frequencies for poly(U) and a well-defined (145 ± 3 base-pair) fragment of DNA in solution. Data sets for the two samples were analyzed by theories which included relaxation by the mechanisms of ³¹P chemical shift anisotropy as well as by ¹H-³¹P dipole–dipole interaction. Neither data set could be satisfactorily described by a single correlation time. A model of a rigid rotor most nearly fits the data for the DNA molecule. Parameters obtained from the least-square fit indicate (1) that the DNA undergoes anisotropic reorientation with a correlation time τ₀ = 6.5 × 10^?7 sec for the end-to-end motion, (2) the ratio of diffusion constants D_∥/D_⊥ is 91, and (3) that the linewidth is due to chemical shift dispersion to the extent of 0.5 ppm. Some deviations of the calculated from the observed values suggested that significant torsional and bending motions may also take place for this DNA. Another model which contains isotropic motion but with a broad distribution of correlation times was required to fit the data for poly(U). A log ? χ² distribution function of correlation times [Scheafer, J. (1973) Macromolecules 6 , 881–888] described well the motion of poly(U) with the average correlation time τ = 3.3 × 10^?9 sec and a distribution parameter p = 14.     

Age-dependent changes in the specificity of tRNA methyltransferases in the cerebellum of the icteric and nonicteric Gunn rat

The activity of tRNA methyltransferases present in the cerebellum of 6- and 21-day-old nonicteric and icteric Gunn rats was compared using purifiedE. coli tRNAs as substrates. At 6 days the tRNA methyltransferases of the icteric animals were significantly more effective in methylating tRNA^Glu ₂ and tRNA^Phe than were those of their nonicteric counterparts. This relationship reversed itself at 21 days. The action of the tRNA methyltransferases from the 6-day-old icteric animals led to higher proportions of 1-methyladenine in tRNA^Glu ₂ and tRNA^Phe than were obtained using the corresponding enzymes of the nonicteric animals. The proportion ofN ²-methylguanine was also higher, yet only in tRNA^fMet and not in tRNA^Phe. The study reveals much more extensive fluctuations in the activity and in the substrate recognition specificity among the cerebellar tRNA methyltransferases of the icteric than among those of the nonicteric controls during the crucial 6–21 day period of cerebellar development.     

Ternary complex between elongation factor Tu•GTP and Phe-tRNA

The effect of aminoacylation and ternary complex formation with elongation factor Tu•GTP on the tertiary structure of yeast tRNA^Phe was examined by ¹H-NMR spectroscopy. Esterification of phenylalanine to tRNA^Phe does not lead to changes with respect to the secondary and tertiary base pair interactions of tRNA. Complex formation of Phe-tRNA^Phe with elongation factor Tu•GTP results in a broadening of all imino proton resonances of the tRNA. The chemical shifts of several NH proton resonances are slightly changed as compared to free tRNA, indicating a minor conformational rearrangement of Phe-tRNA^Phe upon binding to elongation factor Tu•GTP. All NH proton resonances corresponding to the secondary and tertiary base pairs of tRNA, except those arising from the first three base pairs in the aminoacyl stem, are detectable in the Phe-tRNA^Phe•elongation factor Tu•GTP ternary complex. Thus, although the interactions between elongation factor Tu and tRNA accelerate the rate of NH proton exchange in the aminoacyl stem-region, the Phe-tRNA^Phe preserves its typical L-shaped tertiary structure in the complex. At high (> 10⁻⁴ M) ligand concentrations a complex between tRNA^Phe and elongation factor Tu•GDP can be detected on the NMR time-scale. Formation of this complex is inhibited by the presence of any RNA not related to the tRNA structure. Using the known tertiary structures of yeast tRNA^Phe and Thermus thermophilus elongation factor Tu in its active, GTP form, a model of the ternary complex was constructed.     

Conformational Change of the L-Shaped tRNAPhe Molecule

Abstract

Fluorophore of proflavine was introduced onto the 3′-terminal ribose moiety of yeast tRNA^Phe. The distance between the fluorophore and the fluorescent Y base in the anticodon of yeast tRNA^Phe was measured by a singlet-singlet energy transfer. Conformational changes of tRNA^Phe with binding of tRNA^Glu ₂, which has the anticodon UUC complementary to the anticodon GAA of tRNA^Phe, were investigated. The distance obtained at the ionic strength of 100 mM K⁺ and 10 mM Mg²⁺ is very close to the distance from x-ray diffraction, while the distance obtained in the presence of tRNA^Glu ₂ is significantly smaller. Further, using a fluorescent probe of 4-bromomethl-7-methoxycoumarin introduced onto pseudouridine residue Ψ₅₅ in the TΨC loop of tRNA^Phe, Stern-Volmer quenching experiments for the probe with or without added tRNA^Glu ₂were carried out. The results showed greater access of the probe to the quencher with added tRNA^Glu ₂. These results suggest that both arms of the L-shaped tRNA structure tend to bend inside with binding of tRNA^Glu ₂ and some structural collapse occurs at the corner of the L-shaped structure.     

Calorimetric investigations of the helix-coil conversion of phenylalaninespecific transfer ribonucleic acid

The enthalpy of the helix-coil conversion of phenylalaninespecific transfer ribonucleic acid from brewer's yeast (tRNA^Phe_{brewer's yeast}) has been measured using both an LKB 10700-2 batch miciocalorimeter and an adiabatic differential scanning calorimeter. In the mixing calorimeter the conversion from coil to helix was induced by mixing a tRNA^Phe solution with a solution containing an excess of MgSO₄. We measured the enthalpy of this reaction stepwise in the temperature range from +9 to +60° C. For the enthalpy of folding of tRNA^Phe from coil to helix this method yielded the remarkably high value of ?310 $\text{kcal}\text{mole}$ of tRNA^Phe. With the differential scanning calorimeter in which the helix-coil conversion is simply induced by raising the temperature we found a value of +240 $\text{kcal}\text{mole}$ of tRNA^Phe at a Tm value of 76° C and a value of +200 $\text{kcal}\text{mole}$ of tRNA^Phe at a T_m value of 50° C. A comparison of the apparent van't Hoff enthalpies with the calorimetrically measured enthalpies shows, that the cooperativity of the system increases continually with rising melting temperatures - which are achieved by increasing Mg²⁺ concentrations - reaching a constant value at about 57° C. Above this temperature value the thermodynamic behaviour of the helix-coil conversion of tRNA^Phe may be approximately described by the model of an all-or-none process.     

31P NMR relaxation measurements of the phosphate backbone of a double stranded hexadeoxynucleotide in solution: determination of the chemical shift anisotropy

The five phosphates of the deoxynucleotide d(CpGpTpApCpG)₂ have been assigned by two-dimensional heteronuclear NMR spectroscopy. The chemical shift anisotropy and correlation time of each phosphate group has been determined from measurements of the spin-lattice, spin-spin relaxation rate constants and the ³¹P-{¹H} nuclear Overhauser enhancement (NOE) at three magnetic field strengths (4.7 T, 9.4 T, and 11.75 T) and two temperatures (288 K and 298 K). As expected, the relaxation data require two mechanisms to account for the observed rate constants, i.e. dipole-dipole and chemical shift anisotropy. At 9.4 T and 11.75 T, the latter mechanism dominates the relaxation, leading to insignificant NOE intensities. The correlation time, chemical shift anisotropy and effective P-H distance were obtained from least-squares fitting to the data. Comparison of the fitted value for the correlation time with that obtained from ¹H measurements shows that the molecule behaves essentially as rigid rotor on the nanosecond timescale. Large amplitude motions observed in long segments of DNA are due to bending motions that do not contribute significantly to relaxation in short oligonucleotides.Abbreviations CSA chemical shift anisotropy - NOE nuclear Overhauser enhancement Offprint requests to: A. N. Lane     

31P nmr spin-lattice relaxation studies of deoxyoligonucleotides

Temperature-dependent conformational transitions of deoxyoligonucleotides have been monitored by measuring ³¹P chemical shifts, spin-lattice relaxation times (T₁), and ³¹P-{H} nuclear Overhauser enhancements (NOEs). The measured NOE ranged from 30 to 80%, compared to the theoretical maximum of 124% for a dipolar relaxation mediated by rapid isotropic rotation. The observed 3′-5′ phosphate diester ³¹P T₁ showed a similar temperature dependence over the range 2–75°C for both double- and single-stranded oligonucleotides, and for dinucleotides. The results show that dipole–dipole interactions dominate the internucleotide phosphate relaxation rate in oligonucleotides. The same is true of terminal phosphate groups at low temperature; but at higher temperature another process, possibly due to contamination by paramagnetic ions, becomes dominant. The rotational correlation time τ_R calculated from the dipole–dipole relaxation rate of the internucleotide phosphate in d(pA)₂ at 16°C is τ_R = 5.0 × 10^?10 sec, implying a Stokes radius for isotropic rotation of 7.6 Å. The T₁ and NOE values for the double-helical octanucleotide d(pA)₃pGpC(pT)₃ are consistent with dominance of dipole–dipole relaxation and isotropic rotation of a sphere of radius 14 Å, a reasonable dimension for the double helix. Activation energies for the rotation of dinucleotides range from 4 to 6 kcal/mol, close to the value of 4 kcal/mol expected for isotropic rotation. In order to test the possible effect of internal motion of correlation time τ_G on the results, we considered a model in which the nucleotide chain rotates about the P-O bonds. Comparison of the calculation with our experimental results shows that internal motion with τ_G ? 10^?9 sec, as found from other studies to be present for large nucleic acids, would not influence out T₁ and NOE values enough to be distinguished from isotropic rotation. However, we can conclude that τ_G cannot be as fast as 10^?10 sec, even for dinucleotides.     

Temperature dependence of the aminoacylation of tRNA by Bacillus stearothermophilus aminoacyl-tRNA synthetases

Valine specific transfer RNA (tRNA^Val) was isolated from Bacillus stearothermophilus and Escherichia coli by chromatography on benzoylated DEAE–cellulose (BD–cellulose). Likewise isoleucine specific transfer RNA (tRNA^Ile) was isolated from B. stearothermophilus and from Mycoplasma sp. Kid. The thermal denaturation profiles (melting curves) of the two tRNA^Val species in the presence of Mg^{+ +} were nearly identical. However, the T_m for the Kid tRNA^Ile was about 10°C lower than that for the B. stearothermophilus tRNA^Ile. A nuclease and tRNA-free aminoacyl-tRNA synthetase (AA-tRNA synthetase) preparation from B. stearothermophilus was able to function efficiently at temperatures up to 80°C in the aminoacylation of all four tRNA species. Determination of the amino acid-acceptor activity of each tRNA species as a function of temperature of the aminoacylation reaction showed in each case a strong correlation between the loss of acceptor activity and the thermal denaturation profile of the tRNA. Evidence is presented that the loss in acceptor activity is most likely due to a change in structure of the tRNA as opposed to denaturation of the enzyme. These results further support the idea that correct secondary and/or tertiary structure must be maintained for tRNA to be active as a substrate for the AA-tRNA synthetase.     

An energy transfer equilibrium between two identical copies of a ribosome-bound fluorescent transfer RNA analogue: implications for the possible structure of codon-anticodon complexes.

When yeast tRNA_Pf^Phe, a derivative of tRNA^Phe in which proflavine replaces the Y base, is bound simultaneously to both the peptidyl and aminoacyl sites of a 70 S Escherichia coli ribosome, there is a rapid mutual energy transfer between the two bound tRNAs. Analysis of this energy transfer yields an upper limit for the proflavine-proflavine distance of 20 Å. It also allows an unequivocal measurement of the emission spectrum of tRNA_Pf^Phe bound at the aminoacyl site. In the presence of message this spectrum is very different from that seen in the peptidyl site, implying that in the two sites the hypermodified bases exist in significantly different environments. The rapid energy transfer leads to some loss of fluorescence anisotropy. This can be analyzed to obtain an estimate of the angle between the two proflavines: 28 ° ± 10 ° or 152 ° ± 10 °. Taken together all of these results place severe constraints on possible models of codon-anticodon complexes. The mutual energy transfer seen and analyzed on the ribosome is a convenient aspect of fluorescence spectroscopy, and it is one that should see broad application where multiple copies of a fluorescent ligand interact on a macromolecular substrate.     

Transfer RNA Identity Change in Anticodon Variants of E. coli tRNAPhe in Vivo

The anticodon sequence is a major recognition element for most aminoacyl-tRNA synthetases. We investigated the in vivo effects of changing the anticodon on the aminoacylation specificity in the example of E. coli tRNA^Phe. Constructing different anticodon mutants of E. coli tRNA^Phe by site-directed mutagenesis, we isolated 22 anticodon mutant tRNA^Phe; the anticodons corresponded to 16 amino acids and an opal stop codon. To examine whether the mutant tRNAs had changed their amino acid acceptor specificity in vivo, we tested the viability of E. coli strains containing these tRNA^Phe genes in a medium which permitted tRNA induction. Fourteen mutant tRNA genes did not affect host viability. However, eight mutant tRNA genes were toxic to the host and prevented growth, presumably because the anticodon mutants led to translational errors. Many mutant tRNAs which did not affect host viability were not aminoacylated in vivo. Three mutant tRNAs containing anticodon sequences corresponding to lysine (UUU), methionine (CAU) and threonine (UGU) were charged with the amino acid corresponding to their anticodon, but not with phenylalanine. These three tRNAs and tRNA^Phe are located in the same cluster in a sequence similarity dendrogram of total E. coli tRNAs. The results support the idea that such tRNAs arising from in vivo evolution are derived by anticodon change from the same ancestor tRNA.     

Electrostatic effects in divalent ion binding to tRNA.

The binding of Mn²⁺ to yeast tRNA^Phe at 25°C is measured by esr, and found to depend strongly on the concentration of monovalent cations, showing the importance of electrostatic effects. In low sodium (<15mM/l.), the affinity is high and the Scatchard plots are distinctly curved. In high sodium (>50mM/l.), the affinity and the curvature are reduced. In a limited range of sodium concentrations (15–30mM/l.), the folding of tRNA which is induced by the divalent ions results in cooperative binding, leading to upwards convexity of the Scatchard plot. An electrostatic model is developed, based on a single type of binding site which we take to be the phosphates, with a binding constant for Mn²⁺ in the range of that found for ApA, 10 l./M. We compute the change in the binding constant due to the electrostatic potential of the distant charges (other phosphates and counterions), using a single set of parameters for all sodium concentrations. The model predicts that the plots in low sodium are curved, and a good fit to the experimental results is obtained: it is therefore not legitimate or necessary to interpret these results in terms of two types of binding sites. In high salt, the model gives plots that are only slightly curved, corresponding to weaker electrostatic effects. This shows that a search for sites with a special binding mode should be done in high salt. The computed plots are in good agreement with the data, except for slight differences concerning the first bound ions, which give a possible indication in favor of special binding. Given the observation of one special site for Mg²⁺ at 4°C in high sodium [Stein, A. & Crothers, D. M. (1976) Biochemistry 15 , 157–160] in E. coli tRNA^fMet, we have measured the binding of Mn²⁺ at lower temperature. At 12°C, in both yeast tRNA^Phe and E. coli tRNA^fMet, the plots clearly indicate special binding. A site found in high sodium is on a very different footing from the four to six so-called strong sites unduly derived from low-salt binding plots.     

Proton and phosphorus magnetic relaxation studies on the dynamic structure of single-stranded polyriboadenylic acid

Proton and phosphorus-31 nuclear spin-lattice relaxation times (T₁) have been measured with the Fourier-transform method at 100 and 40.5 MHz, respectively, on single-stranded polyriboadenylic acid (poly(A)) in a neutral D₂O solution in the temperature range of 14–82°C. T₁ minimum is observed around 35–45°C for H(8), H(1′), and phosphorus resonances. Rotational correlation times have been deduced from the T₁ data, which indicate that the sugar–phosphate backbone as well as the base–sugar segment is undergoing rapid internal motion of 10^?8–10^?10 sec range. The molecular motion of the sugar–phosphate backbone as deduced from the phosphorus relaxation is well-characterized by a single activation enthalpy of 8.1 kal/mole for the whole temperature range of 14–82°C. Activation enthalpies of similar magnitude have been obtained for the motion of the adenine–ribose moiety from H(8) and H(1′) relaxation. The relative magnitude of T₁ for H(8) and H(1′) infers that the poly(A) nucleotide exists on the average as anti in the single-stranded form. The phosphorus T₁ value is consistent with a conformation such that both C(4′)–C(5′) and C(4′)–C(3′) bonds are nearly trans to their connected O–P bonds.