Thermodynamic and kinetic characterization of the dissociation and assembly of quadruplex nucleic acids

The dissociation and assembly of quadruplex DNA structures (and a few quadruplex RNAs) have been characterized at several levels of rigor, ranging from gross descriptions of factors that govern each process, to semiquantitative comparisons of the relative abilities of these factors to induce stabilization or destabilization, to quantitative studies of binding energies (thermodynamics), transformational rates (kinetics), and analysis of their transition-state energies and mechanisms. This survey classifies these factors, describes the trends and focuses on their interdependencies. Quadruplex assembly is induced most efficiently by added K(+) and elevating the strand concentration; however, Na(+), NH(4)(+), Sr(2+), and Pb(2+) are also very effective stabilizers. Quadruplex dissociation is typically accomplished by thermal denaturation, "melting"; however, when the quadruplex and monovalent cation concentrations are low enough, or the temperature is sufficiently high, several divalent cations, e.g., Ca(2+), Co(2+), Mn(2+), Zn(2+), Ni(2+) and Mg(2+) can induce dissociation. Stabilization also depends on the type of structure adopted by the strand (or strands) in question. Variants include intramolecular, two- and four-stranded quadruplexes. Other important variables include strand sequence, the size of intervening loops and pH, especially when cytosines are present, base methylation, and the replacement of backbone phosphates with phosphorothioates. Competitive equilibria can also modulate the formation of quadruplex DNAs. For example, reactions leading to Watson-Crick (WC) duplex and hairpin DNAs, triplex DNAs, and even other types of quadruplexes can compete with quadruplex association reactions for strands. Others include nonprotein catalysts, small molecules such as aromatic dyes, metalloporphyrins, and carbohydrates (osmolytes). Other nucleic acid strands have been found to drive quadruplex formation. To help reinforce the implications of each piece of information, each functional conclusion drawn from each cited piece of thermodynamic or kinetic data has been summarized briefly in a standardized table entry.

Phospholipid/fatty acid-induced secondary structural change in beta-lactoglobulin during heat-induced gelation

Effects of phosphatidylcholine (PC) and the predominant fatty acids (FAs) in milk, butyrate, oleate, and palmitate, on secondary structural changes in beta-lactoglobulin (beta-LG) during heat-induced gelation were analyzed on the basis of circular dichroism (CD) spectra. Small-strain oscillatory measurements were carried out to characterize viscoelastic properties of the heat-induced gels. In the absence of added salt, PC and FAs induced helix formation of beta-LG on heating to 80 degrees C and increased the storage moduli (G') of heat-induced gels. In the presence of 500 mM NaCl, PC did not change the CD spectrum of beta-LG but decreased G'. In contrast, butyrate substantially unfolded beta-LG in 500 mM NaCl on heating, forming very elastic gels with increased G' values. Palmitate and oleate induced beta-LG gel formation at 25 degrees C without heating; heating to 80 degrees C almost completely unfolded beta-LG in 500 mM NaCl.

Folding of pyrimidine-enriched RNA fragments from the vicinity of the internal ribosomal entry site of hepatitis A virus

Two RNA fragments from the region just upstream of the internal ribosome entry site of Hepatitis A virus (HAV) were studied, a 35mer (HAV-35), 5'U4C3U3C3U4C3U3C2UAU2C3U33(4), and a 23mer (HAV-23), 5(4)U4C3U3C3U4C3U33(4). Secondary structural predictions and nuclease digestion patterns obtained with genomic RNAs suggested that they link two stable Watson-Crick (WC) hairpins in the genomic RNA and do not form conventional WC secondary structure, but do fold to form a condensed, stacked 'domain'. To obtain more information, folding of HAV-23 and -35 RNA fragments was characterized using 1H nuclear magnetic resonance, in H2O as a function of pH and temperature, circular dichroism as a function of NaCl concentration, pH and temperature, and square-wave voltammetry as a function of pH. The results indicate that these oligo-nucleotides form intramolecular structures that contain transient U*U base pairs at pH 7 and moderate ionic strength (100 mM NaCl). This folded structure becomes destabilized and loses the U*U base pairs above and below neutral pH, especially at ionic strengths above 0.1. All of the cytidine protons exchange relatively rapidly with solvent protons (exchange lifetimes shorter than 1 ms), so the structure contains few if any C*CH+base pairs at neutral pH, but can apparently form them at pH values below 6. We present a series of possible models in which chain folding draws the strand termini closer together, possibly serving to pull the attached WC hairpin domains together and providing a functional advantage by nucleating reversible formation of a more viable RNA substrate.

Construction and characterization of a quadruplex DNA selective single-chain autoantibody from a viable motheaten mouse hybridoma with homology to telomeric DNA binding proteins

An autoantibody produced by a hybridoma derived from a viable motheaten mouse was isolated and found to have moderately high binding affinity for nucleic acids and specific preference for quadruplex DNAs. Polymerase chain reaction primers were designed to link the cloned parental antibody variable region fragments together in a subcloning vector. This single-chain variable fragment construct was then subcloned into the T7 promoter-driven expression vector pET22b(+). The construct contains (N- to C-terminal) a pelB leader sequence, variable heavy chain, glycine-serine polylinker, variable light chain, and biotin mimic peptide "strep-tag" sequence (pelB-VH-linker-VL-strep-tag). The ca. 29 kDa protein was expressed, exported to the periplasmic space of NovaBlue (DE) Escherichia coli, and purified by streptavidin affinity chromatography by binding the fused strep-tag peptide. The specificity of the purified single-chain variable fragment (scFv) for quadruplex and duplex DNAs was evaluated by a radioimmunofilterbinding assay. It retained about 10-fold higher affinity for quadruplexes relative to duplex DNA, a reduction of ca. 4-fold from the relative preferences of the parent IgG. The complementary-determining regions contain sequences that are homologous to or conservatively divergent from the key DNA-binding helix-turn-helix-forming motifs of Myb/RAP1 family telomeric DNA-binding proteins (1-3). The presence of this antibody in the autoimmune repertoire suggests a possible linkage between autoimmunity, telomeric DNA binding proteins, and aging.

Isolation and characterization of a monoclonal anti-quadruplex DNA antibody from autoimmune "viable motheaten" mice

A cell line that produces an autoantibody specific for DNA quadruplex structures has been isolated and cloned from a hybridoma library derived from 3-month-old nonimmunized autoimmune, immunodeficient "viable motheaten" mice. This antibody has been tested extensively in vitro and found to bind specifically to DNA quadruplex structures formed by two biologically relevant sequence motifs. Scatchard and nonlinear regression analyses using both one- and two-site models were used to derive association constants for the antibody-DNA binding reactions. In both cases, quadruplexes had higher association constants than triplex and duplex molecules. The anti-quadruplex antibody binds to the quadruplex formed by the promoter-region-derived oligonucleotide d(CGCG4GCG) (Ka = 3.3 x 10(6) M-1), and has enhanced affinity for telomere-derived quadruplexes formed by the oligonucleotides d(TG4) and d(T2G4T2G4T2G4T2G4) (Ka = 5.38 x 10(6) and 1.66 x 10(7) M-1, respectively). The antibody binds both types of quadruplexes but has preferential affinity for the parallel four-stranded structure. In vitro radioimmunofilter binding experiments demonstrated that purified anti-DNA quadruplex antibodies from anti-quadruplex antibody-producing tissue culture supernatants have at least 10-fold higher affinity for quadruplexes than for triplex and duplex DNA structures of similar base composition and length. The antibody binds intramolecular DNA triplexes formed by d(G4T3G4T3C4) and d(C4T3G4T3G4), and the duplex d(CGCGCGCGCG)2 with an affinities of 6. 76 x 10(5), 5.59 x 10(5), and 8.26 x 10(5) M-1, respectively. Competition experiments showed that melted quadruplexes are not effective competitors for antibody binding when compared to native structures, confirming that the quadruplex is bound structure-specifically. To our knowledge, this is the first immunological reagent known to specifically recognize quadruplex structures. Subsequent sequence analysis demonstrates homologies between the antibody complementarity determining regions and sequences from Myb family telomere binding proteins, which are hypothesized to control cell aging via telomeric DNA interactions. The presence of this antibody in the autoimmune repertoire suggests a possible linkage between autoimmunity, telomeric DNA binding proteins, and aging.

Allosteric interactions between DNA strands and monovalent cations in DNA quadruplex assembly: thermodynamic evidence for three linked association pathways

The series of cooperative transitions that lead to [d(TG4)4.(K+)m] quadruplex assembly upon rapid addition of KCl to d(TG4) strands were studied. Quadruplex samples were dialyzed against KCl then Li-EDTA and found to retain between three and five strongly bound potassiums with affinities >10(6) M-2. Absorbance thermal denaturation (melt) and circular dichroism (CD) equilibrium binding data were obtained. The latter were analyzed using two classes of binding models to simulate the effects of the assumed intermolecular interactions on the binding curves (isotherms). The melt experiments yielded equilibrium dissociation constants (Kd) ranging from 10(-11) to 10(-12) M3 at the melting temperatures. Extrapolating these values to 23 degrees C predicts Kd values in the 10(-28) M3 range if the heat capacity (Cp) is not strongly dependent upon temperature changes over this range. Assuming Ka is equal to 1/Kd (from melting analyses), very large association free energies stabilize the quadruplex at 23 degrees C in 100 mM KCl (DeltaGa = -43 kcal mol-1). Plots of the differential melt curve peak half-widths, a measure of cooperativity, versus d(TG4) concentration showed that quadruplex dissociation is much more cooperative at 400 mM KCl than at 100 mM KCl. Forty-eight hour quadruplex assembly time courses were monitored by CD at 264 nm. Equilibrium quadruplex accumulation generally required over 10 h, and net reaction extents were in the 10-85% range. Hill plots of the data show that initial steps in the multistep pathway are positively cooperative, presumably due to strong strand-cation and strand-strand binding interactions in duplex and triplex assembly reactions, then negatively cooperative in quadruplex formation. Models were developed to rationalize the experimental observations in terms of consecutive cooperative allosteric transitions from cation-deficient relaxed (R) strand-aggregates to cation-containing tense (T) structures, driven by the allosteric effector K+. Quantitative mappings of positive and then negative cooperativity were obtained by fitting the results as a function of strand number incorporated during quadruplex assembly. Surprisingly, models for reactions involving incorporation of five and six strands fit the data better than models involving only four strands. The 5-step "induced fit" model fits the data as well as or better than 3- and 4-step models and better than all of the strand aggregation models that were devised and investigated. Net association free energies (summation operatori=1,n) ranged from -20 to -26 kcal mol-1, approximately half the magnitude of the apparent stabilities measured by absorbance melts. Likely explanations for this discrepancy involve hysteresis and errors due to inadequate equilibration in the melt experiments. Hysteresis is thought to be produced by irreversibility due to different predominant mechanisms in absorbance (dissociation) and CD (association) experiments. The kinetic block to quadruplex assembly can be unambiguously attributed to quadruplex formation and not intermediate steps in the assembly mechanism. On the basis of these results we propose that, in addition to the more conventional assembly mechanisms involving duplex dimerization and stepwise strand addition, quadruplex formation can also proceed by triplex-triplex disproportionation. Interaction statistics arguments that support the energetic feasibility of the disproportionation pathway are presented. The allosteric quadruplex assembly model provides a mechanism which could be used by the cell to simultaneously modulate DNA structure and activity within telomeres, transcriptional promoters, recombination-prone chromatin, and other G-rich DNAs. As a result of this allosterism, cation and strand availability and strand-pairing capabilities could profoundly influence the functional capacity of a particular strand over a relatively narrow range of effector concentration changes. (ABSTRACT TRUNCATED)

Specific interaction of glyceraldehyde 3-phosphate dehydrogenase with the 5'-nontranslated RNA of hepatitis A virus

Initiation of translation of hepatitis A virus (HAV) RNA occurs by internal entry and is likely to involve the interaction of trans-acting cellular protein factors with cis-acting structural elements of an internal ribosomal entry segment (IRES) within the 5'-nontranslated RNA. To characterize interactions between African green monkey kidney (BS-C-1) cell proteins and the predicted stem-loop IIIa (nucleotides 155-235) located at the 5' border of the HAV IRES, we utilized an electrophoresis mobility shift assay (EMSA) to identify a 39-kDa RNA-binding protein (p39). Amino-terminal amino acid sequencing of highly purified p39 revealed absolute identity with human glyceraldehyde 3-phosphate dehydrogenase (GAPDH). The identity of p39 as simian GAPDH was further confirmed by antigenic and biochemical similarities between p39 and human GAPDH. Analysis of the RNA binding properties of simian GAPDH revealed that this cellular protein interacts with two additional sites in the HAV 5'-nontranslated RNA, one located between nucleotides 1-148 and the other between nucleotides 597-746. Competitive EMSAs also demonstrated that GAPDH and human polypyrimidine tract-binding protein, a putative picornavirus translation initiation factor, compete with each other for binding to stem-loop IIIa, suggesting that the relative cytoplasmic abundance of GAPDH and polypyrimidine tract-binding protein in individual cell-types may be an important determinant of viral translation activity. Human GAPDH was found to destabilize the folded structure of the stem-loop IIIa RNA based upon observed decreases in the circular dichroism spectra of this RNA following binding of the protein. This RNA helix-destabilizing activity of GAPDH could directly influence IRES-dependent translation and/or replication of picornavirus RNA.

Purification and characterization of beta-structural domains of beta-lactoglobulin liberated by limited proteolysis

Incubation of beta-lactoglobulin with immobilized trypsin at 5-10 degrees C results in a time-dependent release of several fragments of the core domain in yields approaching 15%. Digests were fractionated by ion-exchange chromatography with a Mono Q HR 5/5 column and analyzed after disulfide reduction by polyacrylamide gel electrophoresis in sodium dodecylsulfate. Three fragments with approximate molecular weights of 13.8, 9.6, and 6.7 kD were identified. The fraction from ion-exchange chromatography yielding the 6.7 kD fraction after disulfide reduction was further characterized because it was most homogeneous and gave the highest yield. The C-terminal cleavage site of the 6.7 kD core fragment appeared to be Lys100 or Lys101 as determined by C-terminal amino acid analysis. The exact masses, after reduction with dithiothreitol, are 6195 and 6926 as determined by laser desorption mass spectrometry, corresponding to residues 48-101 and 41-100. Prior to reduction, beta-lactoglobulin C-terminal residues 149-162 are connected to these core domain fragments as shown by C-terminal analysis and mass spectrometry. Structural studies indicate that these 7.9 and 8.6 kD core domain fragments released by immobilized trypsin retain much of their native structure. CD spectra indicate the presence of antiparallel beta-sheet structure similar to the native protein but the alpha-helix is lost. Spectra in the aromatic region indicate the existence of tertiary structure. Moreover, structural transitions in urea are completely reversible as measured by CD spectra, although the extrapolated delta GDH20 and the urea concentration at the transition midpoint are lower than for the native protein. The core domain fragments also display a pH-dependent binding to immobilized trans-retinal as does intact protein. A single endotherm is obtained for both core domain fragments and native protein upon differential scanning calorimetry, but again, the domain is less stable as indicated by a transition peak maxima of 56.9 degrees C as compared with 81.1 degrees C for native protein.

Cytosine-cytosine+ base pairing stabilizes DNA quadruplexes and cytosine methylation greatly enhances the effect

Previous spectroscopic studies demonstrated that the oligodeoxynucleotide d(CGC G3 GCG) undergoes a reversible cation-dependent transition between Watson-Crick (WC) hairpin and parallel-stranded "G-DNA" quadruplex structures [Hardin, C.C., Watson, T., Corregan, M., & Bailey, C. (1992) Biochemistry 31, 833-841]. The relative stabilities of the structures were assessed as a function of pH, and it was found that the quadruplex was substantially stabilized (delta Tm = +15 degrees C) when the pH was shifted from 7.5 to 6 (apparent pKa = 6.8). In the present study, the effects of different cations and pH on four specific sequence varients were determined to test the proposal that this stabilization is due to C.C+ base pair formation mediated by N3-protonation of cytosine. Characteristically large differences in stability were observed when structures formed by d(TAT G3 ATA) and d(TAT G4 ATA) were thermally dissociated at pH 7 in the presence of different cations, verifying that Gn tracts bordered by TAT- and -ATA sequences form quadruplex structures. Imino proton NMR results indicate that the d(m5C G m5C G3 G m5C G)4 and d(TAT G4 ATA)4 quadruplex structures are parallel-stranded. It was necessary to increase the K+ concentration from 40 mM to ca. 200 mM to stabilize d(TAT G3 ATA)4, while the d(TAT G4 ATA)4 complex was nearly as stable as the quadruplex formed by d(CGC G3 GCG) under the same conditions. The d(TAT G4 ATA)4 quadruplex was only slightly stabilized at pH 6 relative to pH 7.5 (delta Tm = +3 degrees C), confirming that the unique stabilization that occurs in the pH 6.8 range with [d(CGC Gn GCG)4.ionn] complexes is due to the C residues. The sequence d(m5C G m5C G3 G m5C G) was found to form a very stable quadruplex in K+ or Ca2+. As with the quadruplex formed by the unmethylated analog, the stability is greatly enhanced when the pH is decreased below about 7.2 (pKa,obs = 6.8). Dissociation kinetic constants and activation energies were determined for quadruplexes formed by d(CGC G3 GCG), d(m5C G m5C G3 G m5C G) and d(TAT G4 ATA). Quantitative comparisons showed that methylation produces a complex that is much more stable at pH 7 in 40 mM Na+ than either of the unmodified structures; the rate-limiting activation energy for dissociation of d(CGC G3 GCG)4 was 22 kcal mol-1 less than for the methylated analog.(ABSTRACT TRUNCATED AT 400 WORDS)

Direct observation of a DNA quadruplex by electrospray ionization mass spectrometry

In 10 mM sodium phosphate, pH 7.6, containing 0.1 mM ethylenediaminetetraacetic acid, ions correspondings to the non-calent, four-stranded oligonucleotide, d(CGCG4GCG)4, were detected by negative ion electrospray ionization (ESI) mass spectrometry at a low nozzle-skimmer (delta NS) bias (-150 V), but not at a higher delta NS bias (> -250 V). In contrast, when the sample was desalted and analyzed by ESI mass spectrometry at a low delta NS bias only ions for the single-stranded d(CGCG4GCG) species were observed. These data agree with spectroscopic evidence which showed that oligonucleotides with the sequence motif 5'd(CGCGnGCG)3', where n = 2-5, formed stable four-stranded complexes in the presence of monatomic cations, like K+, Ca2+, Na+ and Li+, but not in their absence.

A magnesium-induced conformational transition in the loop of a DNA analog of the yeast tRNA(Phe) anticodon is dependent on RNA-like modifications of the bases of the stem

Two single-stranded DNA heptadecamers corresponding to the yeast tRNA(Phe) anticodon stem-loop were synthesized, and the solution structures of the oligonucleotides, d(CCAGACTGAAGATCTGG) and d(CCAGACTGAAGAU-m5C-UGG), were investigated using spectroscopic methods. The second, or modified, base sequence differs from that of DNA by RNA-like modifications at three positions; dT residues were replaced at positions 13 and 15 with dU, and the dC at position 14 with d(m5C), corresponding to positions where these nucleosides occur in tRNA(Phe). Both oligonucleotides form intramolecular structures at pH 7 in the absence of Mg2+ and undergo monophasic thermal denaturation transitions (Tm = 47 degrees C). However, in the presence of 10 mM Mg2+, the modified DNa adopted a structure that exhibited a biphasic "melting" transition (Tm values of 23 and 52 degrees C) whereas the unmodified DNA structure exhibited a monophasic denaturation (Tm = 52 degrees C). The low-temperature, Mg(2+)-dependent structural transition of the modified DNA was also detected using circular dichroism (CD) spectroscopy. No such transition was exhibited by the unmodified DNA. This transition, unique to the modified DNA, was dependent on divalent cations and occurred most efficiently with Mg2+; however, Ca2+ also stabilized the alternative conformation at low temperature. NMR studies showed that the predominant structure of the modified DNA in sodium phosphate (pH 7) buffer in the absence of Mg2+ was a hairpin containing a 7-nucleotide loop and a stem composed of 3 stable base pairs. In the Mg(2+)-stabilized conformation, the loop became a two-base turn due to the formation of two additional base pairs across the loop.(ABSTRACT TRUNCATED AT 250 WORDS)

Solution structure of a synthetic peptide corresponding to a receptor binding region of FSH (hFSH-beta 33-53)

The receptor binding surface of human follicle-stimulating hormone (hFSH) is mimicked by synthetic peptides corresponding to the hFSH-beta chain amino acid sequences 33-53 [Santa-Coloma, T. A., Dattatreyamurty, D., and Reichert, L. E., Jr. (1990), Biochemistry 29, 1194-1200], 81-95 [Santa-Coloma, T. A., Reichert, L. E., Jr. (1990), J. Biol. Chem. 265, 5037-5042], and the combined sequence (33-53)-(81-95) [Santa-Coloma, T. A., Crabb, J. W., and Reichert, L. E., Jr. (1991), Mol. Cell. Endocrinol. 78, 197-204]. These peptides have been shown to inhibit binding of hFSH to its receptor. Circular dichroism (CD) and nuclear magnetic resonance (NMR) spectroscopy were used to determine the structure of the first peptide in this series, the 21 amino acid peptide hFSH-beta-(33-53), H2N-YTRDLVYKDPARPKIQKTCTF-COOH. Analysis of CD data indicated the presence of approximately equal amounts of antiparallel beta-pleated sheet, turns including a beta-turn, "other" structures, and a small amount of alpha-helix. The major characteristics of the structure were found to be relatively stable at acidic pH and the predominant effect of increased solvent polarity was a small increase in alpha-helical content. One- and two-dimensional NMR techniques were used to obtain full proton and carbon signal assignments in aqueous solution at pH 3.1. Analysis of NMR results confirmed the presence of the structural features revealed by CD analysis and provided a detailed picture of the secondary structural elements and global folding pattern in hFSH-beta-(33-53). These features included an antiparallel beta-sheet (residues 38-51 and 46-48), turns within residues 41-46, and 50-52 (a beta-turn) and a small N-terminal helical region comprised of amino acids 34-36. One of the turns is facilitated by prolines 42 and 45. Proline-45 was constrained to the trans conformation, whereas proline-42 favored the trans conformer (approximately 70%) over the cis (approximately 30%). Two resonances were observed for the single alanine residue (A-43) sequentially proximal to P-42, but the rest of the structure was minimally affected by the isomerization at proline-42. The major population of molecules, containing trans-42 and trans-45 prolines, presented 120 NOEs. Distance geometry calculations with 140 distance constraints and energy minimization refinements were used to derive a moderately well-defined model of the peptide's structure.(ABSTRACT TRUNCATED AT 400 WORDS)

Cation-dependent transition between the quadruplex and Watson-Crick hairpin forms of d(CGCG3GCG)

The DNA oligonucleotide d(CGCG3GCG) can form either a Watson-Crick (WC) hairpin or a parallel-stranded quadruplex structure containing six G-quartet base pair assemblies. The exchange between these forms and single strands can be monitored using circular dichroism (CD). NMR results verified the assignment of specific CD bands to quadruplex and hairpin species, respectively. Cations stabilize the quadruplex in the order K+ greater than Ca2+ greater than Na+ greater than Mg2+ greater than Li+ and K+ greater than Rb+ greater than Cs+, indicating that K+ has an optimum ionic radius for complex formation and that ionic charge affects the extent of ion-induced stabilization. The quadruplex is stable in the presence of 40 mM K+ at micromolar DNA concentration and can be kinetically trapped as a metastable form when prepared at millimolar DNA concentration and then diluted into buffer containing 40 mM Na+. The concentration of K+ required to reverse the equilibrium from the hairpin to the quadruplex decreases sharply with increased DNA concentration. The quadruplex has an unusual pKa of ca. 6.8, indicating that C.C+ base pairs are probably forming. This system provides insights into some of the detailed structural characteristics of a ["G4-DNA".ion] complex and an experimental model for the recently proposed "sodium-potassium conformational switch" [Sen, D., & Gilbert, W. (1988) Nature 334, 364-366; Sen, D., & Gilbert, W. (1990) Nature 344, 410-414]. These results may help to explain the lack of cytidine residues in G-rich telomeric DNAs and suggest that methylation of GC-rich duplex DNAs in "GpC islands" may induce quadruplex formation within heterochromatin domains, resulting in reversible chromosomal condensation.

Monovalent cation induced structural transitions in telomeric DNAs: G-DNA folding intermediates

Telomeric DNA consists of G- and C-rich strands that are always polarized such that the G-rich strand extends past the 3' end of the duplex to form a 12-16-base overhang. These overhanging strands can self-associate in vitro to form intramolecular structures that have several unusual physical properties and at least one common feature, the presence of non-Watson-Crick G.G base pairs. The term "G-DNA" was coined for this class of structures (Cech, 1988). On the basis of gel electrophoresis, imino proton NMR, and circular dichroism (CD) results, we find that changing the counterions from sodium to potassium (in 20 mM phosphate buffers) specifically induces conformational transitions in the G-rich telomeric DNA from Tetrahymena, d(T2G4)4 (TET4), which results in a change from the intramolecular species to an apparent multistranded structure, accompanied by an increase in the melting temperature of the base pairs of greater than 25 degrees, as monitored by loss of the imino proton NMR signals. NMR semiselective spin-lattice relaxation rate measurements and HPLC size-exclusion chromatography studies show that in 20 mM potassium phosphate (pH 7) buffer (KP) TET4 is approximately twice the length of the form obtained in 20 mM sodium phosphate (pH 7) buffer (NaP) and that mixtures of Na+ and K+ produce mixtures of the two forms whose populations depend on the ratio of the cations. Since K+ and NH4+ are known to stabilize a parallel-stranded quadruplex structure of poly[r(I)4], we infer that the multistranded structure is a quadruplex. Our results indicate that specific differences in ionic interactions can result in a switch in telomeric DNAs between intramolecular hairpin-like or quadruplex-containing species and intermolecular quadruplex structures, all of which involve G.G base pairing interactions. We propose a model in which duplex or hairpin forms of G-DNA are folding intermediates in the formation of either 1-, 2-, or 4-stranded quadruplex structures. In this model monovalent cations stabilize the duplex and quadruplex forms via two distinct mechanisms, counterion condensation and octahedral coordination to the carbonyl groups in stacked planar guanine "quartet" base assemblies. Substituting one of the guanosine residues in each of the repeats of the Tetrahymena sequence to give the human telomeric DNA, d(T2AG3)4, results in less effective K(+)-dependent stabilization. Thus, the ion-dependent stabilization is attenuated by altering the sequence. Upon addition of the Watson-Crick (WC) complementary strand, only the Na(+)-stabilized structure dissociates quickly to form a WC double helix.(ABSTRACT TRUNCATED AT 400 WORDS)

Effects of nucleotide bromination on the stabilities of Z-RNA and Z-DNA: a molecular mechanics/thermodynamic perturbation study

The structures of ZI- and ZII-form RNA and DNA oligonucleotides were energy minimized in vacuum using the AMBER molecular mechanics force field. Alternating C-G sequences were studied containing either unmodified nucleotides, 8-bromoguanosine in place of all guanosine residues, 5-bromocytidine in place of all cytidine residues, or all modified residues. Some molecules were also energy minimized in the presence of H2O and cations. Free energy perturbation calculations were done in which G8 and C5 hydrogen atoms in one or two residues of Z-form RNAs and DNAs were replaced in a stepwise manner by bromines. Bromination had little effect on the structures of the energy-minimized molecules. Both the minimized molecular energies and the results of the perturbation calculations indicate that bromination of guanosine at C8 will stabilize the Z forms of RNA and DNA relative to the nonbrominated Z form, while bromination of cytidine at C5 stabilizes Z-DNA and destabilizes Z-RNA. These results are in agreement with experimental data. The destabilizing effect of br5C in Z-RNAs is apparently due to an unfavorable interaction between the negatively charged C5 bromine atom and the guanosine hydroxyl group. The vacuum-minimized energies of the ZII-form oligonucleotides are lower than those of the corresponding ZI-form molecules for both RNA and DNA. Previous x-ray diffraction, nmr, and molecular mechanics studies indicate that hydration effects may favor the ZI conformation over the ZII form in DNA. Molecular mechanics calculations show that the ZII-ZI energy differences for the RNAs are greater than three times those obtained for the DNAs. This is due to structurally reinforcing hydrogen-bonding interactions involving the hydroxyl groups in the ZII form, especially between the guanosine hydroxyl hydrogen atom and the 3'-adjacent phosphate oxygen. In addition, the cytidine hydroxyl oxygen forms a hydrogen bond with the 5'-adjacent guanosine amino group in the ZII-form molecule. Both of these interactions are less likely in the ZI-form molecule: the former due to the orientation of the GpC phosphate away from the guanosine ribose in the ZI form, and the latter apparently due to competitive hydrogen bonding of the cytidine 2'-hydroxyl hydrogen with the cytosine carbonyl oxygen in the ZI form. The hydrogen-bonding interaction between the cytidine hydroxyl oxygen and the 5'-adjacent guanosine amino group in Z-RNA twists the amino group out of the plane of the base. This may be responsible for differences in the CD and Raman spectra of Z-RNA and Z-DNA.

Conformational analysis of d(C3G3), a B-family duplex in solution

NMR and circular dichroism studies of the duplex formed by the self-complementary DNA hexanucleotide d(C3G3) indicate that it is a B-type structure but differs from standard B-form. An analysis of NMR coupling constants within the deoxyribose moieties yields a 70% or greater contribution from pseudorotation phase angles corresponding to the C3'-exo conformation, a conformation similar to the C2'-endo conformation associated with B-form DNA. Intranucleotide interproton distances are consistent with a B-form structure, but some internucleotide distances are intermediate between A- and B-form structures. Circular dichroism spectra have B-form characteristics but also include an unusual negative band at 282 nm. The solution spectroscopic results are in contrast with X-ray crystallographic studies which find A-form structures for similar sequences.

Comparison of the B- and Z-form hairpin loop structures formed by d(CG)5T4(CG)5

The partially self-complementary synthetic DNA oligonucleotide d(CG)5T4(CG)5 has been studied by using 1H and 31P NMR and circular dichroism. Results show that, under low-salt conditions (120 mM NaCl buffer), an intramolecular hairpin loop exists in which the double-helical stem region is B-form and the thymidine loop residues have predominantly southern (C2'-endo) sugar conformations. The thymidine glycosidic torsion angles are intermediate between syn and anti or exist as an equilibrium mixture of residues in the two extremes. NOESY data indicate that the structure of the loop region is very similar to that found for d(CG)2T4(CG)2 [Hare, D. R., & Reid, B. R. (1986) Biochemistry 25, 5341-5350]. Under high-salt conditions (6 M NaClO4 buffer), the dominant form (approximately equal to 85%) is an intramolecular hairpin structure in which the stem region forms a Z-form double helix. As in the B-form, the loop thymidine residues are intermediate between the syn and anti conformations or exist as an equilibrium mixture of the two, but the thymidine sugar conformations differ in that they are biased toward northern (C3'-endo) conformations.

Characterization of anti-Z-RNA polyclonal antibodies: epitope properties and recognition of Z-DNA

Chemically brominated poly[r(C-G)] [Br-poly[r(C-G)]] containing 32% br8G and 26% br5C was recently shown to contain a 1:1 mixture of A- and Z-form unmodified nucleotides under physiological conditions of temperature, pH, and ionic strength [Hardin, C. C., Zarling, D. A., Puglisi, J. D., Trulson, M. O., Davis, P. W., & Tinoco, I., Jr. (1987) Biochemistry 26, 5191-5199]. Proton NMR results show that more extensive bromination of poly[r(C-G)] (49% br8G, 43% br5C) produces polynucleotides containing greater than 80% unmodified Z-form nucleotides. Using these polynucleotides as antigens, polyclonal antibodies were elicited in rabbits and mice specific for the Z-form of RNA. IgG fractions were purified from rabbit anti-Br-poly[r(C-G)] sera and characterized by immunoprecipitation, nitrocellulose filter binding, and ELISA. Two different anti-Z-RNA IgG specificities were observed. Decreased levels of brominated nucleotides in the immunogen correlated with an increased extent of specific cross-reactivity with Z-DNA. Inoculation of rabbits with polynucleotide immunogens containing 49% br8G and 43% of br5C produced specific anti-Z-RNA IgGs that do not recognize Z-DNA determinants. This suggests that the 2'-OH group is part of the anti-Z-RNA IgG determinant. In contrast, Br-poly[r(C-G)] immunogens containing 32% br8G and 26% br5C produced IgGs that specifically recognize both Z-RNA and Z-DNA. These results show that the bromine atoms are not required for recognition of the Z conformation by the antibodies. The affinity of these anti-Z-RNA IgGs for Z-RNA is about 10-fold higher than for Z-DNA.(ABSTRACT TRUNCATED AT 250 WORDS)

Binding of ethidium ion to left-handed Z-RNA induces a cooperative transition to right-handed RNA at the intercalation site

The equilibrium binding of the ethidium cation (Etd+) to the right-handed A-form of poly-[r(C-G)], the B-form of poly[d(C-G)], and the left-handed Z-forms of Br-poly[r(C-G)] and Br-poly[d(C-G)] was investigated in 0.22 M NaCl by optical methods. Scatchard analysis indicates that Etd+ intercalates into right-handed forms of poly[r(C-G)] and poly[d(C-G)] in a noncooperative manner. Correlation of Etd+ absorbance binding isotherms and polynucleotide circular dichroism data indicates that drug binding to Br-poly[r(C-G) and Br-poly[d(C-G)] results in cooperative conversion from left-handed Z-forms to right-handed intercalated conformations. Approximate stoichiometries necessary to induce the left- to right-handed transitions are 1 Etd+/9 base pairs (bp) for Z-RNA and 1 Etd+/6 bp for Z-DNA. The apparent limiting binding stoichiometries are approximately 1 Etd+/3 bp for RNA and 1 Etd+/2 bp for DNA. The equilibrium binding constants for binding to the right-handed forms decrease in the order Br-poly[d(C-G)], Br-poly[r(C-G)], poly[d(C-G)], and poly[r(C-G)]. Thermodynamic parameters are obtained by van't Hoff analysis of Etd+ absorbance thermal dissociation data. Enthalpy values for all four polynucleotides are negative and of similar magnitude. Negative entropy values indicate that the binding processes are primarily enthalpically driven.(ABSTRACT TRUNCATED AT 250 WORDS)

Partial assignment of resonances in the 19F nuclear magnetic resonance spectra of 5-fluorouracil-substituted transfer RNAs

Features of the 19F nuclear magnetic resonance (NMR) spectra of three purified 5-fluorouracil-(FUra-) substituted Escherichia coli tRNAs, tRNA(1Val), tRNA(mMet), and tRNA(fMet), are compared. Each of the tRNA species can be resolved into two isoaccepting forms, A and B, whose 19F NMR spectra differ in the shift of one peak from the 4.5 to 4.8 parts per million (ppm) range (FUra = O) in the spectrum of isoacceptor B upfield to ca. -15 ppm in that of isoacceptor A. Because the sequences of the two isoacceptors of each tRNA differ only at one position in the D loop, that normally occupied by a dihydrouridine residue, we assign the 4.5 ppm peak in the spectrum of fluorine-labeled tRNA(1Val) to FUra17 and the resonance at 4.6 ppm in the spectrum of fluorouracil-substituted tRNA(mMet) to FUra20. A reciprocal 19F[19F] nuclear Overhauser effect is observed between the downfield peaks A and B in the 19F NMR spectrum of 19F-labeled tRNA(1Val). Assuming that fluorine-labeled tRNA(1Val) has a structure similar to that of yeast tRNA(Phe), only FUra54 and -55 are close enough (4-5 A) to give an appreciable 19F homonuclear Overhauser effect. Peaks A and B have therefore been assigned to FUra54 and -55. As the temperature is raised from 30 to 45 degrees C, the intensity of peak B (6.6 ppm) in the spectrum of 19F-labeled tRNA(1Val) gradually shifts upfield to 6.4 ppm (Tm = 36 degrees C), indicating a temperature-dependent slow exchange of the corresponding 5-fluorouracil residue between two magnetically distinct environments.(ABSTRACT TRUNCATED AT 250 WORDS)

Telomeric DNA oligonucleotides form novel intramolecular structures containing guanine-guanine base pairs

Structural properties of DNA oligonucleotides corresponding to the single-stranded molecular terminus of telomeres from several organisms were analyzed. Based on physical studies including nondenaturing polyacrylamide gel electrophoresis, absorbance thermal denaturation analysis, and 1H and 31P nuclear magnetic resonance spectroscopy, we conclude that these molecules can self-associate by forming non-Watson-Crick, guanine.guanine based-paired, intramolecular structures. These structures form below 40 degrees C at moderate ionic strength and neutral pH and behave like hairpin duplexes in nondenaturing polyacrylamide gels. Detailed analysis of the hairpin structure formed by the telomeric sequence from Tetrahymena, (T2G4)4, shows that it is a unique structure stabilized by hydrogen bonds and contains G residues in the syn conformation. We propose that this novel form of DNA is important for telomere function and sets a precedent for the biological relevance of non-Watson-Crick base-paired DNA structures.

19F nuclear magnetic resonance as a probe of anticodon structure in 5-fluorouracil-substituted Escherichia coli transfer RNA

The use of 19F nuclear magnetic resonance (n.m.r.) spectroscopy as a probe of anticodon structure has been extended by investigating the effects of tetranucleotide binding to 5-fluorouracil-substituted Escherichia coli tRNA(Val)1 (anticodon FAC). 19F n.m.r. spectra were obtained in the absence and presence of different concentrations of oligonucleotides having the sequence GpUpApX (X = A,G,C,U), which contain the valine codon GpUpA. Structural changes in the tRNA were monitored via the 5-fluorouracil residues located at positions 33 and 34 in the anticodon loop, as well as in all other loops and stems of the molecule. Binding of GpUpApA, which is complementary to the anticodon and the 5'-adjacent FUra 33, shifts two resonances in the 19F spectrum. One, peak H (3.90 p.p.m.), is also shifted by GpUpA and was previously assigned to FUra 34 at the wobble position of the anticodon. The effects of GpUpApA differ from those of GpUpA in that the tetranucleotide induces the downfield shift of a second resonance, peak F (4.5 p.p.m.), in the 19F spectrum of 19F-labeled tRNA(Val)1. Evidence that the codon-containing oligonucleotides bind to the anticodon was obtained from shifts in the methyl proton spectrum of the 6-methyladenosine residue adjacent to the anticodon and from cleavage of the tRNA at the anticodon by RNase H after binding dGpTpApA, a deoxy analog of the ribonucleotide codon. The association constant for the binding of GpUpApA to fluorinated tRNA(Val)1, obtained by Scatchard analysis of the n.m.r. results, is in good agreement with values obtained by other methods. On the basis of these results, we assign peak F in the 19F n.m.r. spectrum of 19F-labeled tRNA(Val)1 to FUra 33. This assignment and the previous assignment of peak H to FUra 34 are supported by the observation that the intensities of peaks F and H in the 19F spectrum of fluorinated tRNA(Val)1 are specifically decreased after partial hydrolysis with nucleass S1 under conditions leading to cleavage in the anticodon loop. The downfield shift of peak F occurs only with adenosine in the 3'-position of the tetranucleotide; binding of GpUpApG, GpUpApC, or GpUpApU results only in the upfield shift of peak H. The possibility is discussed that this base-specific interaction between the 3'-terminal adenosine and the 5-fluorouracil residue at position 33 involves a 5'-stacked conformation of the anticodon loop. Evidence also is presented for a temperature-dependent conformational change in the anticodon loop below the melting temperature of the tRNA.

Mobility of individual 5-fluorouridine residues in 5-fluorouracil-substituted Escherichia coli valine transfer RNA. A 19F nuclear magnetic resonance relaxation study

19F nuclear magnetic resonance (n.m.r.) relaxation parameters of 5-fluorouracil-substituted Escherichia coli tRNA(Val)1 were measured and used to characterize the internal mobility of individual 5-fluorouridine (FUrd) residues in terms of several models of molecular motion. Measured relaxation parameters include the spin-lattice (T1) relaxation time at 282 MHz, the 19F(1H) NOE at 282 MHz, and the spin-spin (T2) relaxation time, estimated from linewidth data at 338 MHz, 282 MHz and 84 MHz. Dipolar and chemical shift anisotropy contributions to the 19F relaxation parameters were determined from the field-dependence of T2. The results demonstrate a large chemical shift anisotropy contribution to the 19F linewidths at 282 and 338 MHz. Analysis of chemical shift anisotropy relaxation data shows that, relative to overall tumbling of the macromolecule, negligible torsional motion occurs about the glycosidic bond of FUrd residues in 19F-labeled tRNA(Val)1, consistent with the maintenance of base-base hydrogen-bond and/or stacking interactions at all fluorouracil residues in the molecule. The dipolar relaxation data are analyzed by using the "two-state jump" and "diffusion in a cone" formalisms. Motional amplitudes (theta) are interpreted as being due to pseudorotational fluctuations within the ribose ring of the fluorinated nucleoside. These amplitudes range from approximately 30 degrees to 60 degrees, assuming a correlation time (tau i,2) of 1.6 ns. By using available 19F n.m.r. assignment data for the 14 FUrd residues in 5-fluorouracil-substituted tRNA(Val)1, these motional amplitudes can be correlated directly with the environmental domain of the residue. Residues located in tertiary and helical structural domains show markedly less motion (theta approximately equal to 30 to 35 degrees) than residues located in loops (theta approximately equal to 45 to 60 degrees). A correlation between residue mobility and solvent exposure is also demonstrated. The amplitudes of internal motion for specific residues agree quite well with those derived from X-ray diffraction and molecular dynamics data for yeast tRNA(Phe).

Cytoplasmic Z-RNA

Specific immunochemical probes for Z-RNA were generated and characterized to search for possible Z-RNA-like double helices in cells. Z-RNA was detected in the cytoplasm of fixed protozoan cells by immunofluorescence microscopy using these anti-Z-RNA IgGs. In contrast, autoimmune or experimentally elicited anti-DNA antibodies, specifically reactive with B-DNA or Z-DNA, stained the nuclei. Pre-or nonimmune IgGs did not bind to the cells. RNase A or T1 digestion eliminated anti-Z-RNA IgG binding to cytoplasmic determinants; however, DNase I or mung bean nuclease had no effect. Doxorubicin and ethidium bromide prevented anti-Z-RNA antibody binding; however, actinomycin D, which does not bind double-stranded RNA, did not. Anti-Z-RNA immunofluorescence was specifically blocked in competition assays by synthetic Z-RNA but not Z-DNA, A-RNA, or single-stranded RNAs. Thus, some cytoplasmic sequences in fixed cells exist in the left-handed Z-RNA conformation.

Stabilization of Z-RNA by chemical bromination and its recognition by anti-Z-DNA antibodies

Limited chemical bromination of poly[r(C-G)] (32% br8G, 26% br5C) results in partial modification of guanine C8 and cytosine C5, producing a mixture of A- and Z-RNA forms. The Z conformation in the brominated polynucleotide is stabilized at much lower ionic strength than in the unmodified polynucleotide. More extensive bromination of poly[r(C-G)] (greater than 49% br8G, 43% br5C) results in stabilization of a form of RNA having a Z-DNA-like (ZD) CD spectrum in low-salt, pH 7.0-7.5 buffers. Raising the ionic strength to 6 M NaBr or NaClO4 results in a transition in Br-poly[r(C-G)] to a Z-RNA (ZR) conformation as judged by CD spectroscopy. At lower ionic strength Z-DNA-like (ZD) and A-RNA conformations are also present. 1H NMR data demonstrate a 1/1 mixture of A- and Z-RNAs in 110 mM NaBr buffer at 37 degrees C. Nuclear Overhauser effect (NOE) experiments permit complete assignments of GH8, CH6, CH5, GH1', and CH1' resonances in both the A- and Z-forms. GH8----GH1' NOEs demonstrate the presence of both A- and Z-form GH8 resonances in slow exchange on the NMR time scale. The NMR results indicate that unbrominated guanine residues undergo transition to the syn conformation (Z-form). Raman scattering data are consistent with a mixture of A- and Z-RNAs in 110 mM NaCl buffer at 37 degrees C. Comparison with the spectrum of Z-DNA indicates that there may be different glycosidic torsion angles in Z-RNA and Z-DNA [Tinoco, I., Jr., Cruz, P., Davis, P., Hall, K., Hardin, C. C., Mathies, R. A., Puglisi, J. D., Trulson, M. O., Johnson, W. C., & Neilson, T. (1986) in Structure and Dynamics of RNA, pp 55-68, Plenum, New York].(ABSTRACT TRUNCATED AT 250 WORDS)

Fluorine-19 nuclear magnetic resonance studies of the structure of 5-fluorouracil-substituted Escherichia coli transfer RNA

19F nuclear magnetic resonance has been used to study fully active Escherichia coli tRNA1Val in which 5-fluorouracil has replaced more than 90% of all uracil and uracil-derived modified bases. The 19F spectrum of the native tRNA contains resolved resonances for all 14 incorporated 5-fluorouracils. These are spread over a 6 ppm range, from 1.8 to 7.7 ppm downfield of the standard free 5-fluorouracil. The 19F resonances serve as sensitive monitors of tRNA conformation. Removal of magnesium or addition of NaCl produces major, reversible changes in the 19F spectrum. Most affected is the lowest field resonance (peak A) in the spectrum of the native tRNA. This shifts 2-3 ppm upfield as the Mg2+ concentration is lowered or the NaCl concentration is raised. Thermal denaturation of the tRNA results in a collapse of the spectrum to a single broad peak centered at 4.7 ppm. Study of the pH dependence of the 19F spectrum shows that five incorporated fluorouracils with 19F signals in the central, 4-5.5 ppm, region of the spectrum, peaks C, D, E, F, and H, are accessible to titration in the pH 4.5-9 range. All have pKa's close to that of free 5-fluorouridine (ca. 7.5). Evidence for a conformation change in the tRNA at mildly acidic pHs, ca. 5.5, is also presented. Four of the titratable 5-fluorouracil residues, those corresponding to peaks D, E/F, and H in the 19F spectrum of fluorine-labeled tRNAVal1, are essentially completely exposed to solvent as determined by the solvent isotope shift (SIS) on transfer of the tRNA from H2O to 2H2O. These are also the 5-fluorouracils that readily form adducts with bisulfite, a reagent that reacts preferentially with pyrimidines in single-stranded regions. On the basis of these results, resonances D, E, F, and H in the middle of the 19F spectrum are attributed to 5-fluorouracils in non-base-paired (loop) regions of the tRNA. Evidence from the ionic strength dependence of the 19F spectrum and arguments based on other recent studies with fluorinated tRNAs support earlier suggestions [Horowitz, J., Ofengand, J., Daniel, W. E., & Cohn, M. (1977) J. Biol. Chem. 252, 4418-4420] that the resonances at lowest field correspond to tertiary hydrogen-bonded 5-fluorouracils. Consideration of ring-current effects and the preferential perturbation of upfield 19F resonances by the cyclophotoaddition of 4'-(hydroxymethyl)-4,5',8-trimethylpsoralen, which is known to react most readily with pyrimidines in double-stranded regions, permits initial assignment of upfield resonances to 5-fluorouracils in helical stems.(ABSTRACT TRUNCATED AT 400 WORDS)

Fluorine-19 nuclear magnetic resonance study of codon-anticodon interaction in 5-fluorouracil-substituted E. coli transfer RNAs

Codon-anticodon interaction was investigated in fully active 5-fluorouracil-substituted E. coli tRNAVal1 (anticodon FAC) by 19F NMR spectroscopy. Binding of the codon GpUpA results in the upfield shift of a 19F resonance at 3.9 ppm in the central region of the 19F NMR spectrum, whereas trinucleotides not complementary to the anticodon have no effect. The same 19F resonance shifts upfield upon formation of an anticodon-anticodon dimer between the 19F-labeled tRNA and E. coli tRNATyr2 (anticodon QUA). These results permit assignment of the peak at 3.9 ppm to the 5-fluorouracil at position 34 in the anticodon of fluorouracil-substituted tRNAVal1. The methionine codon ApUpG also causes a sequence-specific upfield shift of a peak in the central part of the 19F NMR spectrum of fluorinated E. coli tRNAMetm. However, ApUpG has no effect on the 19F spectrum of 19F-labeled E. coli tRNAMetf, indicating possible conformational differences between the anticodon loop of initiator and chain-elongating methionine tRNAs. 19F NMR experiments detect no binding of CpGpApA to the complementary FpFpCpG (replaces Tp psi pCpG) in the T-loop of 5-fluorouracil-substituted tRNAVal1, in the presence or absence of codon, suggesting that the tertiary interactions between the T- and D-loops are not disrupted by codon-anticodon interactions.

Characterization of the fluorodihydrouracil substituent in 5-fluorouracil-containing Escherichia coli transfer RNA

The fluorodihydrouridine derivative previously detected in one of two isoaccepting forms of FUra-substituted Escherichia coli tRNAMetf has been further characterized. This substituent is responsible for the 19F resonance observed 15 ppm upfield from free FUra (= 0 ppm) in the high resolution 19F-NMR spectra of FUra-substituted tRNA purified by chromatography on DEAE-cellulose, at pH 8.9, to remove normal tRNA. Similar highfield 19F signals have now been observed in the spectra of two other purified fluorinated E. coli tRNAs, tRNAMetm and tRNAVal1, as well as in unfractionated tRNA, indicating the widespread occurrence of the constituent. Comparison with 19F spectrum of the model compound 5'-deoxy-5-fluoro-5,6-dihydrouridine (dH56FUrd) (delta FUra = -31.4 ppm; JHF = 48 Hz) indicates that the substituent does not contain an intact fluorodihydrouridine ring. dH56FUrd is considerably more alkali labile than 5,6-dihydrouridine (H56Urd). At pH 8.9, where H56Urd is stable, dH56FUrd is degraded to a derivative, presumably a fluoroureidopropionic acid, with a 19F resonance at - 15.7 ppm that nearly coincides with the upfield peak in the spectrum of pH 8.9-treated tRNA. The 19F-NMR spectrum of fluorinated tRNA, not exposed to pH 8.9, exhibits two peaks 31 and 32 ppm upfield of FUra, in place of the 19F signal at - 15 ppm. Hydrolysis of this tRNA with RNAase T2 produces a sharp doublet 33 ppm upfield (JHF = 45 Hz). Similarities of the 19F chemical shift and coupling constant to those of dH56FUrd, allows assignment of the peak at -33 ppm to an intact fluorodihydrouridine residue in the tRNA. Our results demonstrate that FUra residues incorporated into E. coli tRNA at sites normally occupied by dihydrouridine can be recognized by tRNA-modifying enzymes and reduced to fluorodihydrouridine. This substituent is labile at moderately alkaline pH values and undergoes ring-opening during purification of the tRNA.