Kinetics of the hydrolysis of guanosine 5′-phospho-2-methylimidazolide

We have studied the hydrolysis of guanosine 5-phospho-2-methylimidazolide, 2-MeImpG, in aqueous buffered solutions of various pH's at 75°C and 37°C. At 75°C and pH1.0, two kinetic processes were observed spectrophotometrically: the first and more rapid one is attributed to the hydrolysis of the phosphoimidazolide P-N bond; the second and much slower one, to the cleavage of the glycosidic bond. At 37°C, pH 2.0, the spectrophotometrically determined rate constant of P–N bond hydrolysis was confirmed by using high pressure liquid chromatography, HPLC. With the latter technique it was possible to separate reactants and products and also to extend the pH-rate profile into the neutral region where rates are slower and, therefore, difficult to measure spectrophotometrically. The pH-rate profiles at both temperatures exhibit similar behavior. At pH<2 the pseudo-first-order rate constant increases with decreasing pH; in the region 27. These data are consistent with a reactivity order zwitterion>anion for P–N bond hydrolysis. It is noteworthy that P–N bond hydrolysis in phosphoimidazolides is very slow compared to other phosphoramidates. This may be one of the reasons why this compound showed extraordinary ability in forming long oligomers under template-directed conditions.     

Limiting concentrations of activated mononucleotides necessary for poly(C)-directed elongation of oligoguanylates

Summary Selected imidazolide-activated nucleotides have been subjected to hydrolysis under conditions similar to those that favor their template-directed oligomerization. Rate constants of hydrolysis of the P–N bond in guanosine 5-monophosphate 2-methylimidazolide (2-MeImpG) and in guanosine 5-monophosphate imidazolide (ImpG), k_h, have been determined in the presence/absence of magnesium ion as a function of temperature and polycytidylate [poly(C)] concentration. Using the rate constant of hydrolysis of 2-MeImpG and the rate constant of elongation, i.e., the reaction of an oligoguanylate with 2-MeImpG in the presence of poly(C) acting as template, the limiting concentration of 2-MeImpG necessary for oligonucleotide elongation to compete with hydrolysis can be calculated. The limiting concentration is defined as the initial concentration of monomer that results in its equal consumption by hydrolysis and by elongation. These limiting concentrations of 2-MeImpG are found to be 1.7 mM at 37°C and 0.36 mM at 1°C. Boundary conditions in the form of limiting concentration of activated nucleotide may be used to evaluate a prebiotic model for chemical synthesis of biopolymers. For instance, the limiting concentration of monomer can be used as a basis of comparison among catalytic, but nonenzymatic, RNA-type systems.We also determined the rate constant of dimerization of 2-MeImpG, k₂=0.45±0.06 M^–1 h^–1 in the absence of poly(C), and 0.45±0.06k₂0.97±0.13 M^–1 h^–1 in its presence at 37°C and pH 7.95. This dimerization, as well as the trimerization of 2-MeImpG, which represent the first steps in the oligomerization reaction, are markedly slower than the elongation of longer oligoguanylates, (pG) _n n>6. This means that in the presence of low concentrations of 2-MeImpG (1.7 mM) the system directs the elongation of longer oligomers more efficiently than the formation of short oligomers such as dimers and trimers. These results will be discussed as a possible example of chemical selection in template-directed reactions of nucleotides.     

Relative Reactivity of Ribosyl 2′-OH vs. 3′-OH in Concentrated Aqueous Solutions of Phosphoimidazolide Activated Nucleotides

Phosphoimidazolide activated ribomononucleotides (*pN, see structure) are useful substrates for the non-enzymatic synthesis of oligonucleotides. In the presence of metal ions, aqueous solutions of *pN yield primarily the two internucleotide-linked (pN^2'pN and pN^3'pN) and the pyrophosphate-linked (N^5'ppN) dimers. Small amounts of cyclic dimers and higher oligomers are also produced. In this study the relative reactivity of 2-OH vs. 3-OH was determined from the ratio of the yields of pN^2'pN vs. pN^3'pN. Experiments were performed at 23 °C in the range 7.2 pH 8.4 with substrates that differ in nucleobase (guanosine (G), cytidine (C), uridine (U), and adenosine (A)) and leaving group (imidazole (Im), 2-methylimidazole (2-MeIm) and 2,4-dimethylimidazole (2,4-diMeIm)). Two metal ions (Mg²⁺ or Mn²⁺) were employed as catalysts. The conditions used here, i.e. a substrate concentration in the range 0.1 M to 1.0 M and metal ion concentration in the range 0.05 M to 0.2 M, favor base-stacking interactions. The ratio pN^2'pN: pN^3'pN = 2-5: 3-5 was found independent of nucleobase and typically varied between 2 to 3 indicating that the 2-OH is about 2 to 3 times more reactive than the 3-OH. *pN with Im, compared to 2-MeIm and 2,4-diMeIm leaving group, produce lower yields of internucleotide linked dimers, and a higher pN^2'pN: pN^3'pN ratio. Trends in the data, observed with all three leaving groups, suggest an increase in pN^2'pN: pN^3'pN ratio with decreasing substrate concentration (up to 5.47 with 0.051 M ImpG). The observations are in accord with earlier studies reporting a relative reactivity 2'-5': 3'-5'= 6 to 9 obtained with Im as the leaving group, in dilute nucleotide solutions and under conditions that disfavor stacking. It is speculated that the concentration induced change in the relative reactivity is the result of self-association via base-stacking that enhances selectively the proximity of the 3-OH of one molecule to the reactive P-N bond of an other molecule. The implication of these conclusions for oligomerization/ligation reactions is discussed.     

Osmium-Based Pyrimidine Contrast Tags for Enhanced Nanopore-Based DNA Base Discrimination

Nanopores are a promising platform in next generation DNA sequencing. In this platform, an individual DNA strand is threaded into nanopore using an electric field, and enzyme-based ratcheting is used to move the strand through the detector. During this process the residual ion current through the pore is measured, which exhibits unique levels for different base combinations inside the pore. While this approach has shown great promise, accuracy is not optimal because the four bases are chemically comparable to one another, leading to small differences in current obstruction. Nucleobase-specific chemical tagging can be a viable approach to enhancing the contrast between different bases in the sequence. Herein we show that covalent modification of one or both of the pyrimidine bases by an osmium bipyridine complex leads to measureable differences in the blockade amplitudes of DNA molecules. We qualitatively determine the degree of osmylation of a DNA strand by passing it through a solid-state nanopore, and are thus able to gauge T and C base content. In addition, we show that osmium bipyridine reacts with dsDNA, leading to substantially different current blockade levels than exhibited for bare dsDNA. This work serves as a proof of principle for nanopore sequencing and mapping via base-specific DNA osmylation.     

Affinity of guanosine derivatives for polycytidylate revisited

Evidence is presented for complexation of guanosine 5-monophosphate 2-methylimidazolide (2-MeImpG) with polycytidylate (poly(C)) at pH 8.0 and 23°C in the presence of 1.0 M NaCl and 0.2 M MgCl₂ in water. The association of 2-McImpG with poly(C) was investigated using UV-vis spectroscopy as well as by monitoring the kinetics of the nucleophilic substitution reaction of the imidazole moiety by amines. The results of both methods are consistent with moderately strong poly(C) · 2-McImpG complexation and the spectrophotometric measurements allowed the construction of a binding isotherm with a concentration of 2-McImpG equal to 5.55 ± 0.15 mM at half occupancy. UV spectroscopy was employed to establish the binding of other guanosine derivatives on poly(C). These derivatives are guanosine 5-monophosphate (5GMP), guanosine 5monophosphate imidazolide (ImpG), and guanosine 5monophosphate morpholidate (morpG). Within experimental error these guanosine derivatives exhibit the same affinity for poly(C) as 2-McImpG.     

DIMERIZATION IN HIGHLY CONCENTRATED SOLUTIONS OF PHOSPHOIMIDAZOLIDE ACTIVATED MONONUCLEOTIDES

Phosphoimidazolide activated ribomononucleotides (*pN) are useful substrates for the non-enzymatic synthesis of polynucleotides. However, dilute neutral aqueous solutions of *pN typically yield small amounts of dimers and traces of polymers; most of *pN hydrolyzes to yield nucleoside 5-monophosphate. Here we report the self-condensation of nucleoside 5-phosphate 2-methylimidazolide (2-MeImpN with N = cytidine, uridine or guanosine) in the presence of Mg2+ in concentrated solutions, such as might have been found in an evaporating lagoon on prebiotic Earth. The product distribution indicates that oligomerization is favored at the expense of hydrolysis. At 1.0 M, 2-MeImpU and 2-MeImpC produce about 65% of oligomers including 4% of the 3,5-linked dimer. Examination of the product distribution of the three isomeric dimers in a self-condensation allows identification of reaction pathways that lead to dimer formation. Condensations in a concentrated mixture of all three nucleotides (U,C,G mixtures) is made possible by the enhanced solubility of 2-MeImpG in such mixtures. Although percent yield of internucleotide linked dimers is enhanced as a function of initial monomer concentration, pyrophosphate dimer yields remain practically unchanged at about 20% for 2-MeImpU, 16% for 2-MeImpC and 25% of the total pyrophosphate in the U,C,G mixtures. The efficiency by which oligomers are produced in these concentrated solutions makes the evaporating lagoon scenario a potentially interesting medium for the prebiotic synthesis of dimers and short RNAs.     

Self-replication of chemical systems based on recognition within a double or a triple helix: a realistic hypothesis.

A scenario is proposed by which non-enzymatic self-replication of short RNA molecules could occur. The hypothesis is illustrated for the self-replication of an oligopyrimidine (Y) strand. The successful replication of Y requires a series of plausible steps. The first, experimentally feasible, step involves the template-directed polynucleotide synthesis, based on Watson-Crick base pairing, of an oligopurine (R) strand using Y as the template, and chemically activated mononucleotides as the building blocks. This step will result in the formation of an oligopyrimidine.oligopurine (YR) double helix. The second step requires the use of the double helix as the template for the synthesis of a second oligopyrimidine (Y') strand from activated pyrimidine monomers. This synthesis could be facilitated by the binding of the monopyrimidines in the major groove of the YR double helix, via Hoogsteen-type base pairing with the R strand, establishing in that sense triple helix recognition. This step, if successful, should result in the formation of a new strand, Y', that runs parallel to the oligopurine strand. Y' differs from Y in that all 3'-5' phosphodiester linkages in Y are replaced by 5'-3' linkages in Y'. The resulting triple helix (YRY') is in dynamic equilibrium with YR and free Y'. In subsequent steps, unassociated Y' directs the synthesis of the complementary oligopurine (R') strand forming a new double helix Y'R' that may direct the synthesis of an oligopyrimidine strand, Y, that is expected to be identical to the first strand that started the whole sequence. An attempt is made to generalize the above hypothesis to mixed oligonucleotides containing all four bases and identify the limitations of this hypothesis.     

Nucleotides as nucleophiles: reactions of nucleotides with phosphoimidazolide activated guanosine.

An earlier study of the reaction of phosphoimidazolide activated nucleosides (ImpN) in aqueous phosphate buffers indicated two modes of reaction of the phosphate monoanion and dianion. The first mode is catalysis of the hydrolysis of the P-N bond in ImpN's which leads to imidazole and nucleoside 5'-monophosphate. The second represents a nucleophilic substitution of the imidazole to yield the nucleoside 5'-diphosphate. This earlier study thus served as a model for the reaction of ImpN with nucleoside monophosphates (pN) because the latter can be regarded as phosphate derivatives. In the present study we investigated the reaction of guanosine 5'-phosphate-2-methylimidazolide, 2-MeImpG, in the presence of pN (N = guanosine, adenosine and uridine) in the range 6.9 less than or equal to pH less than or equal to 7.7. We observed that pN's do act as nucleophiles to form NppG, and as general base to enhance the hydrolysis of the P-N bond in 2-MeImpG, i.e. pN show the same behavior as inorganic phosphate. The kinetic analysis yields the following rate constants for the dianion pN2-: knpN = 0.17 +/- 0.02 M-1 h-1 for nucleophilic attack and khpN = 0.11 +/- 0.07 M-1 h-1 for general base catalysis of the hydrolysis. These rate constants which are independent of the nucleobase compare with kp.2 = 0.415 M-1 h-1 and khp2. = 0.217 M-1 h-1 for the reactions of HPO4(2-). In addition, this study shows that under conditions where pN presumably form stacks, the reaction mechanism remains unchanged although in quantitative terms stacked pN are somewhat less reactive. Attack by the 2'-OH and 3'-OH groups of the ribose moiety in amounts greater than or equal to 1% is not observed; this is attributed to the large difference in nucleophilicity in the neutral pH range between the phosphate group and the ribose hydroxyls. This nucleophilicity rank is not altered by stacking.     

Kinetic analysis of the template effect in ribooligoguanylate elongation

We have undertaken a complete kinetic analysis of the template-directed oligoguanylate synthesis originated in Orgel's laboratory (Inoue and Orgel, 1982). The reaction of guanosine 5′-phospho-2-methylimidazolide, 2-MelmpG, with ribooligoguanylates all 3′–5′ linked, designatedn ³ withn=7−12, was studied in the presence/absence of the complementary template polycytidylic acid, poly(C). Conditions were chosen where poly(C) and 2-MelmpG are in large excess over the oligoguanylate. In the absence of the template at 37 °C the reaction leads to three isomeric oligomers that are elongated by one monomer unit. They are the 3′–5′ linked, (n+1)³, the 2′–5′ linked, (n+1)², and the pyrophosphate product, (n+1) ^p , formed in an approximate ratio 1:2:5. In the presence of the template the reaction is 20-fold faster and yields productsn+1,n+2,n+3 etc. as long as 2-MelmpG is available. Most importantly the formation of the natural, 3′–5′ linked isomer, is enhanced selectively by 140-fold at 37 °C. Qualitative observations allow the conclusion that this enhancement is temperature dependent and increases with decreasing temperature. For example, at 1 °C only the 3′–5′ linked isomers were detected. Initial rates for the disappearance of then ³ oligoguanylate were determined at 1, 23, and 37 °C. It was found that the pseudo-first order rate constant for oligoguanylate elongation was linearly proportional to the 2-MelmpG concentration. This implies that the reaction complex poly(C)·n ³·2-MelmpG does not accumulate under the reaction conditions, a conclusion which is also supported by infrared data (Miles and Frazier, 1982). The implication of the above results with respect to chemical evolution is that lower temperatures, i.e., close to freezing, enhance the regioselectivity of these template-directed reactions and that one way to improve replication models may be sought in finding conditions that favor stable reaction complexes. NASA — National Research Council Research Fellow.