Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (112K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (13)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Sawai, H.
Right arrow Articles by Ozaki, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sawai, H.
Right arrow Articles by Ozaki, H.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Nucleic Acids Research Pages 2995-3000  

Non-enzymatic, template-directed ligation of 2[prime]-5[prime] oligoribonucleotides. Joining of a template and a ligator strand
Introduction
Materials And Methods
   Materials
   Synthesis of templates [2[prime]-5[prime]]U10 and [3[prime]-5[prime]]U10
   HPLC
   UV spectroscopy and oligoribonucleotide concentration
   Circular dichroism (CD) spectroscopy and melting studies
   Template-directed condensation
   Characterization of the reaction products
Results And Discussion
   CD of helices composed of [2[prime]-5[prime]](pA)2 and [2[prime]-5[prime]]U10 or [3[prime]-5[prime]] U10
   Template-directed ligation
   Characterization of the reaction products
References

Non-enzymatic, template-directed ligation of 2[prime]-5[prime] oligoribonucleotides. Joining of a template and a ligator strand

Non-enzymatic, template-directed ligation of 2[prime]-5[prime] oligoribonucleotides. Joining of a template and a ligator strand

Hiroaki Sawai*, Shuichi Totsuka, Kenji Yamamoto, Hiroaki Ozaki

Department of Chemistry, Faculty of Engineering, Gunma University, Kiryu, Gunma 376-8515, Japan

Received February 9, 1998; Revised and Accepted April 27, 1998

ABSTRACT

Decauridylate containing exclusively a 2[prime]-5[prime] phospho-diester bond ([2[prime]-5[prime]]U10) served as a template for the synthesis of oligoadenylates [oligo(A)s] from the 5[prime]-phosphorimidazolide of 2[prime]-5[prime] diadenylate (ImpA-2[prime]p5[prime]A). Joining of [2[prime]-5[prime]]U10 and ImpA2[prime]p5[prime]A also took place in substantial amounts to yield long-chain oligoribonucleotides in the template-directed reaction. An unusual CD spectrum ascribed to helix formation between [2[prime]-5[prime]]U10 and [2[prime]-5[prime]](pA)2 was observed under the same conditions as that of the template-directed reaction. The 3[prime]-5[prime] linked decauridylate ([3[prime]-5[prime]]U10) also promoted the template-directed synthesis of oligo(A)s from ImpA2[prime]p5[prime]A, but more slowly compared with [2[prime]-5[prime]]U10. The results indicate that short-chain RNA oligomers with a 2[prime]-5[prime] phosphodiester bond could lead to longer oligoribonucleotides by template-directed chain elongation.

INTRODUCTION

It has been proposed that RNA could have played the roles of information carrier and catalyst in the prebiotic era (1,2). Contemporary RNA exclusively has a 3[prime]-5[prime] phosphodiester bond. However, 2[prime]-5[prime] linked oligoribonucleotides are chemically feasible and are formed in model processes of prebiotic synthesis of RNA. For example, 2[prime]-5[prime] linked oligoribonucleotides are mainly formed from activated mononucleotides in the presence of a divalent metal (3,4) or uranyl ion catalyst (5,6). Template-directed oligomerization of the activated nucleotides also produces mainly 2[prime]-5[prime] linked oligoribonucleotides in the non-enzymatic model reaction of RNA replication (7). On the other hand, 3[prime]-5[prime] linked oligoribonucleotides are preferentially formed in oligomerization by a clay mineral catalyst (8-10) or in template-directed oligomerization of 2-methylimidazole-activated guanylate on a 3[prime]-5[prime]-linked poly(C) template (11-14). The Zn2+-catalyzed oligomerization of guanosine 5[prime]-phosphorimidazolide on 3[prime]-5[prime]-linked poly(C) also yields 3[prime]-5[prime]-linked oligoguanylates [oligo(G)s], while the Pb2+ catalyst yields mainly 2[prime]-5[prime]-linked oligo(G)s under the same conditions (15). Lohrmann and Orgel reported that the 2[prime]-OH group of ribonucleotide is more reactive than the 3[prime]-OH group and the 2[prime]-5[prime] phosphodiester bond is formed more easily than the 3[prime]-5[prime] bond in non-enzymatic oligoribonucleotide synthesis (16), although the reactivity of each OH group is affected by the ribose ring conformation and by a steric effect imposed by the activated group or a template. Thus 2[prime]-5[prime]- and 3[prime]-5[prime]-linked oligoribonucleotides could have been formed during chemical evolution and possibly served as a template for complementary oligoribonucleotide synthesis from mono- or short-chain oligoribonucleotides. The 3[prime]-5[prime]-linked poly- or oligoribonucleotides have been proved to serve as a template in the non-enzymatic oligomerization or ligation of oligoribonucleotides (7,15-17). Very recently, Ferris et al. reported the template-directed synthesis of oligo(G)s from activated guanylate on 2[prime]-5[prime]-linked oligo(C)s formed by a clay mineral catalyst (18,19). Prakash et al. have also demonstrated 2[prime]-5[prime]-linked oligo(C)s in a hairpin template-promoted oligomerization of the 2-methylimidazolide of guanosine-5[prime]-phosphate (2-MeImpG) in the presence of Mg2+. They have shown that the 2[prime]-5[prime]-linked oligo(C) template is less efficient than the 3[prime]-5[prime]-linked oligo(C) for oligomerization of 2-MeImpG (20). However, template-directed ligation of 2[prime]-5[prime]-linked short-chain oligoribonucleotides on the complementary 2[prime]-5[prime]- or 3[prime]-5[prime]-linked oligoribonucleotides, which could take place in the prebiotic chemical process, has been a matter of speculation.

In this study, we investigated helix formation and non-enzymatic template-directed ligation of 2[prime]-5[prime]-linked diadenylate on 2[prime]-5[prime]- or 3[prime]-5[prime]-linked decauridylate, [2[prime]-5[prime]]U10 or [3[prime]-5[prime]]U10. The 2[prime]-5[prime]-linked oligonucleotides were formed from the activated nucleotide by a uranyl ion catalyst. We found that the [2[prime]-5[prime]]U10 template promoted ligation of the 2[prime]-5[prime] diadenylate to yield oligo(A)s faster than [3[prime]-5[prime]]U10 (reaction a, Scheme 1). Moreover, the 2[prime]-5[prime] diadenylate substantially condensed to the template [2[prime]-5[prime]]U10, as shown in Scheme 1, reaction b.


Scheme 1. ImpA2[prime]p5[prime]A and template-directed ligation on the [2[prime]-5[prime]]U10 template.


Figure 1. CD spectra of the helix composed of [2[prime]-5[prime]](pA)2 and [2[prime]-5[prime]]U10. (A) The measurement was carried out at 10 mM [2[prime]-5[prime]](pA)2 and 20 mM [2[prime]-5[prime]]U10 concentrations in the presence of 30 mM MgCl2 and 0.2 M NaCl in 0.2 M N-ethylmorpholine buffer (pH 7.0) using a 0.0025 cm path length cuvette at 0 (---), 5 (- - -), 10 (@@@) and 30°C (- @ -).(B) Melting curve constructed from CD at 260 nm.

MATERIALS AND METHODS

Materials

Poly(A), poly(U), adenosine 5[prime]-monophosphate, uridine 5[prime]-monophosphate and nuclease P1 (NP1) were from Yamasa (Tokyo). Snake venom phosphodiesterase (SVPD) and bacterial alkaline phosphatase (BAP) were from Worthington Biochemical (Freehold, NJ). Nuclease SW was a generous gift from Dr J.Mukai (Kyushu University, Japan). All other materials were obtained commercially. The 2[prime]-5[prime]-linked oligo(A)s ([2[prime]-5[prime]](pA)n, n = 2, 4 and 6) were prepared by UO22+ ion-catalyzed polymerization of adenosine 5[prime]-phosphorimidazolide (ImpA) in aqueous solution by the published procedure (5,21). The phosphorimidazolide of 2[prime]-5[prime]-linked diadenylate (ImpA2[prime]p5[prime]A) was prepared from [2[prime]-5[prime]](pA)2 and imidazole by the same procedure as for the synthesis of ImpA (5,16).

Synthesis of templates [2[prime]-5[prime]]U10 and [3[prime]-5[prime]]U10

The 2[prime]-5[prime]-linked decauridylate ([2[prime]-5[prime]]U10) was prepared by UO22+-catalyzed polymerization of uridine 5[prime]-phosphorimidazolide (ImpU) by the literature procedure (5,21), followed by treatment with NP1 and BAP to concomitantly degrade 3[prime]-5[prime] linkages and the 5[prime]-terminal phosphate. The 2[prime]-5[prime]-linked oligo(U)s were purified by DEAE-Sephadex ion exchange column chromatography using linear gradient elution with triethylammonium hydrogen carbonate buffer (0-1.0 M). Fractions of [2[prime]-5[prime]]U10 were collected and dialyzed several times to remove triethylammonium hydrogen carbonate completely. The 3[prime]-5[prime]-linked decauridylate ([3[prime]-5[prime]]U10) was prepared by partial digestion of poly(U) with nuclease SW (21) followed by treatment with BAP to cleave the 5[prime]-terminal phosphate and purified by the same procedure as described for [2[prime]-5[prime]]U10. The purity of the oligoribonucleotides was determined by HPLC.

HPLC

HPLC analyses were performed on an RPC-5 column with linear gradient elution with NaClO4 (0-0.05 M) buffered with 2.5 mM Tris-acetate (pH 7.5) and 0.1 mM EDTA or an ODS-Silicagel column with a linear gradient of acetonitrile (2-30 %) buffered with 0.1 M triethylammonium acetate (pH 7.0).

UV spectroscopy and oligoribonucleotide concentration

UV spectra were taken with a Hitachi 3200 spectrophotometer. Oligoribonucleotide concentrations were determined by UVabsorbance at 260 nm using the residual molar extinction coefficients of A and U, [epsis]260 = 15.3 × 103 and 10.0 × 103/M/cm respectively, after correcting for the hypochromicity of each oligoribonucleotide. Hypochromicity of the oligoribonucleotides was determined by the ratio of absorbance at 260 nm in Tris buffer (pH 7.0) before and after alkaline hydrolysis of the oligoribonucleotide. The hypochromicity value was the mean of five determinations. The hypochromicities of [2[prime]-5[prime]](pA)2, [2[prime]-5[prime]](pA)4, [2[prime]-5[prime]](pA)6, [2[prime]-5[prime]]U10 and [3[prime]-5[prime]]U10 were 18, 25, 28, 15 and 10% respectively. All oligomer concentrations are expressed based on a monomer equivalent.

Circular dichroism (CD) spectroscopy and melting studies

CD spectra were recorded with a JASCO-720 spectropolarimeter at several temperatures with a PTC-343 temperature controller. The CD results were expressed in terms of [[thetas]] (molecular ellipticity). CD versus temperature plots were recorded at 260 or 265 nm measuring at every 0.1°C with a temperature increase rate of 0.25°C/min. The melting temperature was determined from the first differential of the CD melting curve. The CD measurements of the combinations [2[prime]-5[prime]](pA)2:[2[prime]-5[prime]]U10 and [2[prime]-5[prime]](pA)2:[3[prime]-5[prime]]U10 were carried out at high oligoribonucleotide concentrations using a 0.0025 cm path length cuvette to obtain information on helix formation under the conditions of template-directed synthesis.

Template-directed condensation

A reaction mixture (50 µl) containing 0.01 M ImpA2[prime]p5[prime]A, 0.02 M [2[prime]-5[prime]]U10 or [3[prime]-5[prime]]U10, 0.03 M MgCl2, 0.2 M NaCl and 0.2 M N-ethylmorpholine buffer (pH 7.0) was prepared at 0°C and allowed to stand for 28 days at 0°C. Aliquots of the mixture were taken out after 1, 3, 7, 10, 14 and 21 days and analyzed by HPLC. A control reaction of ImpA2[prime]p5[prime]A without template was also done under the same conditions as for the template-directed condensation.

Characterization of the reaction products

Identification of the reaction products, [2[prime]-5[prime]](pA)4, [2[prime]-5[prime]](pA)6, [2[prime]-5[prime]]A2, [2[prime]-5[prime]]A4 and pA2[prime]pA3[prime]pA2[prime]pA, was carried out by comparison of HPLC retention times with those of authentic samples. Characterization of other minor oligo(A)s containing a partial 3[prime]-5[prime] linkage was not attempted. However, digestion of the reaction mixture with NP1 was conducted to degrade only the 3[prime]-5[prime] linkage, leaving only 2[prime]-5[prime] oligoribonucleotides, which were analyzed by HPLC. Long-chain products, which were formed by condensation between ImpA2[prime]p5[prime]A and template [2[prime]-5[prime]]U10, were isolated by preparative HPLC on ODS-Silicagel. The isolated oligoribonucleotides were characterized by two different enzyme digestion systems. Degradation of the oligoribonucleotide with SVPD and BAP was carried out to analyze the ratio of adenosine to uridine in the oligoribonucleotide. SVPD cleaves both the 2[prime]-5[prime] and 3[prime]-5[prime] phosphodiester bonds of the oligonucleotide to yield the component 5[prime]-mononucleotides, whose 5[prime]-phosphates were cleaved by BAP. Enzyme digestion was carried out in a mixture (50 µl) containing 0.5 U SVPD, 0.2 U BAP, 0.1 OD260 substrate oligoribonucleotide and 0.01 M MgCl2 in 0.02 M Tris-HCl buffer (pH 8.0) at 37°C for 8 h. Treatment of the oligoribonucleotides with NP1 was further conducted to analyze the type of linkage formed in the template-directed reaction, because NP1 degrades only the 3[prime]-5[prime] linkage. The NP1 digestion was carried out in a mixture (50 µl) containing 10 µg NP1 and 0.25 OD260 substrate oligoribonucleotide in 0.01 M ammonium acetate buffer (pH 5.3) for 8 h at 37°C. The molecular weight of the isolated long-chain oligoribonucleotides was further analyzed in a PE-Sciex API-100 electrospray mass spectrometer.

RESULTS AND DISCUSSION

CD of helices composed of [2[prime]-5[prime]](pA)2 and [2[prime]-5[prime]]U10 or [3[prime]-5[prime]] U10

We have examined helix formation between [2[prime]-5[prime]](pA)2 and [2[prime]-5[prime]]U10 or [3[prime]-5[prime]]U10 by CD under the conditions of template-directed condensation of ImpA2[prime]p5[prime]A on U10 template using a 0.0025 cm cuvette. Figure 1A shows the CD spectrum of the mixture of 0.01 M [2[prime]-5[prime]](pA)2 and 0.02 M [2[prime]-5[prime]]U10 in the presence of 0.03 M Mg2+ at 0, 5, 10 and 30°C. The spectrum at 0°C, shown by a solid line, has a peak at 255 nm which is ascribed to the triple helix formed from 2[prime]-5[prime]-linked oligo(A) and 2[prime]-5[prime]-linked oligo(U) in the presence of Mg2+ ion (22). This unique CD spectrum suggests that the triplex composed of the 2[prime]-5[prime]-linked oligo(A) and oligo(U) has a different conformational form from that composed of 3[prime]-5[prime]-linked RNAs. The peak at 255 nm disappeared with rising temperature and a CD band with a new peak at 270 nm appeared, which is ascribed to a mixture of the single-stranded [2[prime]-5[prime]](pA)2 and [2[prime]-5[prime]]U10. The CD spectrum of the mixture at 30°C is essentially the same as that of the sum of the component strands under the same conditions at 30°C (data not shown). Figure 1B shows the CD melting curve of the triplex composed of [2[prime]-5[prime]](pA)2 and [2[prime]-5[prime]]U10, which was monitored by CD at 260 nm. The CD amplitude versus temperature profile indicates that helix formation takes place below 20°C. Analysis of the first differential curve allows determination of the melting temperature of the helix as 8.5°C. Figure 2A shows the CD spectrum of a 1:2 mixture of [2[prime]-5[prime]](pA)2 and [3[prime]-5[prime]]U10 under the same conditions as for the mixture of [2[prime]-5[prime]](pA)2 and [2[prime]-5[prime]]U10. Triple helix formation between [2[prime]-5[prime]](pA)2 and [3[prime]-5[prime]]U10 was observed as an increased CD band and shorter wavelength shift of the peak CD around 270 nm (22); the CD band decreased considerably at high temperature, reflecting melting of the helix to the single-stranded components. The melting curve of the helix monitored by CD at 270 nm is shown in Figure 2B. The melting temperature, which was estimated from the first differential of the melting curve, was 5.9°C. The CD studies of helix formation indicate the possibility of template-directed condensation of 2[prime]-5[prime] diadenylate on the [2[prime]-5[prime]]U10 or [3[prime]-5[prime]]U10 template at 0°C.


Figure 2. CD spectra of the helix composed of [2[prime]-5[prime]](pA)2 and [3[prime]-5[prime]]U10. (A) The measurement was carried out at 10 mM [2[prime]-5[prime]](pA)2 and 20 mM [3[prime]-5[prime]]U10 concentrations in the presence of 30 mM MgCl2 and 0.2 M NaCl in 0.2 M N-ethylmorpholine buffer (pH 7.0) using a 0.0025 cm path length cuvette at 0 (---), 5 (- - -), 15 (- - -) and 20 (@@@), 35 (-@-) and 50 °C (-@@-). (B) Melting curve constructed from CD at 270 nm.

Template-directed ligation

Figure 3A illustrates the HPLC profiles of the reaction products from ImpA2[prime]p5[prime]A in the presence of the [2[prime]-5[prime]]U10 template at 0°C. After 28 days the starting ImpA2[prime]p5[prime]A disappeared and the condensation products, tetra- and hexaadenylates, were formed in 21 and 1.5% yields respectively, in addition to large amounts of two long-chain ligation products from [2[prime]-5[prime]]U10 and pA2[prime]p5[prime]A, I and II (Table 1). The main peak in the fraction of tetraadenylate was fully 2[prime]-5[prime]-linked (pA)4 and the subsidiary peak after the main peak was tetraadenylate with an internal 3[prime]-5[prime] linkage, pA2[prime]pA3[prime]pA2[prime]pA. The molar ratio of the resulting linkage isomers of the tetraadenylate, pA2[prime]pA2[prime]pA2[prime]pA to pA2[prime]pA3[prime]pA2[prime]pA, was 81:19%. The template-directed synthesis reaction mixture was further subjected to NP1 digestion. NP1 degrades only the 3[prime]-5[prime] linkage, to yield fully 2[prime]-5[prime]-linked oligonucleotides or a mononucleotide with a 5[prime]-terminal phosphate, although concomitant phosphatase digestion gave a small amount of 5[prime]-dephosphorylated products. The results of the NP1 digestion demonstrates that fully 2[prime]-5[prime]-linked oligonucleotides were the main products of template-directed synthesis. With a [3[prime]-5[prime]]U10 template instead of [2[prime]-5[prime]]U10, a small amount of the starting ImpA2[prime]p5[prime]A survived even after 28 days, and tetra- and hexaadenylates were formed in 8 and 0.5% yields respectively (Fig. 3B). The internal linkage of the tetraadenylate formed in this reaction was mainly the 2[prime]-5[prime] phosphodiester bond, as shown in Table 1. No long-chain oligoribonucleotide corresponding to compound I or II was obtained from ImpA2[prime]p5[prime]A with [3[prime]-5[prime]]U10. Tetraadenylate was formed in very small amounts and the starting ImpA2[prime]p5[prime]A survived in substantial amounts in the control reaction without template (Fig. 3C). The half-life of hydrolysis of the phosphorimidazolide bond of ImpA2[prime]p5[prime]A was 17 days at 0°C and pH 7.0 in the absence of template. Figure 4 exhibits the time course of the condensation reaction of ImpA2[prime]p5[prime]A on [2[prime]-5[prime]] or [3[prime]-5[prime]]U10 template. The reaction was almost complete in 14 days with a [2[prime]-5[prime]]U10 template but required over 30 days with a [3[prime]-5[prime]]U10 template. The [2[prime]-5[prime]]U10 template promoted condensation of ImpA2[prime]p5[prime]A more efficiently than the [3[prime]-5[prime]]U10 template, although helix formation was observed in both cases at 0°C under the same conditions.


Figure 3. HPLC profile of the reaction products from template-directed condensation of ImpA2[prime]p5[prime]A. (A) Reaction on [2[prime]-5[prime]]U10 template. Template-directed synthesis was performed in 50 µl solution containing 10 mM ImpA2[prime]p5[prime]A, 20 mM template, 30 mM MgCl2, 0.2 M NaCl and 0.2 M N-ethylmorpholine buffer (pH 7.0) for 28 days at 0°C. (B) Reaction on [3[prime]-5[prime]]U10 template. The reaction was carried out under the same conditions as described above except for the template. (C) Control reaction without template. The reaction was performed under the same conditions as described above.


Figure 4. Time course of the template-directed synthesis. (A) Reaction on [2[prime]-5[prime]]U10 template. (B) Reaction on [3[prime]-5[prime]]U10 template. Yields were estimated from the peak area of the products on HPLC and are expressed based on the starting ImpA2[prime]pA: m, ImpA2[prime]p5[prime]A; l, pA2[prime]p5[prime]A; q, tetramer; n, hexamer; z, ligation products (I + II).

Characterization of the reaction products

Identification of the ligation products was carried out by comparison of the HPLC with that of an authentic sample or selective enzyme digestion. Structure analysis of the long-chain ligation products I and II was performed on fractions isolated by reverse phased ODS-Silicagel HPLC. We used ODS-Silicagel HPLC for preparative purposes, because the sample charging capacity of the RPC-5 column is too small to obtain sufficient amounts of the samples for characterization. Enzyme digestion of compounds I and II with SVPD and BAP yielded adenosine and uridine in ratios of nearly 10:2 and 10:4, respectively. The result indicates that compound I was formed by condensation of [2[prime]-5[prime]]U10 and ImpA2[prime]p5[prime]A and II from I and another ImpA2[prime]p5[prime]A. The structure of compounds I and II was also confirmed by their molecular weight determination by electrospray mass spectrometry. The observed masses of I and II were 3762.2 and 4425.6 respectively, which correspond to the calculated masses of U10A2-triethylammonium (mol wt 3762.2) and U10A4-triethylammonium (mol wt 4424.6), respectively. Compounds I and II were further treated with NP1 to determine the type of phosphodiester bond between [2[prime]-5[prime]]U10 and [2[prime]-5[prime]]A2. Substantial amounts of 2[prime]-5[prime]U10 and pA2[prime]p5[prime]A were formed in addition to the intact form by NP1 digestion of I. In compound I, [2[prime]-5[prime]]U102[prime]pA2[prime]pA was resistant to NP1, while [2[prime]-5[prime]]U103[prime]pA2[prime]pA was cleaved to [2[prime]-5[prime]]U10 and pA2[prime]p5[prime]A. NP1 digestion of compound II yielded fully 2[prime]-5[prime]-linked U10A2, U10, A4 and A2, in addition to intact U10A4. The result indicates that compound II is a mixture of [2[prime]-5[prime]]U103[prime]pA2[prime]pA2[prime]pA2[prime]pA, [2[prime]-5[prime]]U103[prime]pA2[prime]pA3[prime]pA2[prime]pA, [2[prime]-5[prime]]U102[prime]pA2[prime]pA2[prime]pA2[prime]pA and [2[prime]-5[prime]]U102[prime]pA2[prime]pA3[prime]pA2[prime]pA. The yields and the types of linkage of I and II formed from ImpA2[prime]p5[prime]A and template [2[prime]-5[prime]]U10 are listed in Table 1. [2[prime]-5[prime]]U10 and pA2[prime]pA were joined by 2[prime]-5[prime] and 3[prime]-5[prime] phosphodiester bonds in nearly the same ratio. The possibility of a 5[prime]-5[prime] phosphodiester bond between [2[prime]-5[prime]]U10 and pA2[prime]pA could be excluded by the fact that only the 2[prime]- or 3[prime]-OH group of the ribosyl system takes part in efficient non-enzymatic template-directed condensation (23). Formation of compounds I and II indicates that ImpA2[prime]p5[prime]A reacts with the 2[prime]- or 3[prime]-terminal OH group of [2[prime]-5[prime]]U10 in the helix state. It is noteworthy that template-directed ligation takes place between the template and ligator strands in the 2[prime]-5[prime] oligoribonucleotide system. No such ligation product was formed from ImpA2[prime]p5[prime]A and [3[prime]-5[prime]]U10, although helix formation takes place in this system. The unique conformational geometry of the helix composed of [2[prime]-5[prime]]U10 and pA2[prime]p5[prime]A, as suggested by CD studies, may promote this unusual ligation reaction. The mechanism of the ligation reaction is not clear, however, one could propose a conceivable model of how the reaction might work. In the py-pu-py triplex model, the purine and the third strand are aligned in parallel and the two phosphodiester backbones are close together. This allows close proximity of the 2[prime],3[prime]-OH of the template end with the activated 5[prime]-end of a dimer unit bound to the last two bases of the template.

Previous studies demonstrate that 3[prime]-5[prime]-linked poly- or oligoribo-nucleotides serve as a template for oligomerization of com-plementary activated mononucleotides to form the corresponding oligoribonucleotides with 2[prime]-5[prime] or 3[prime]-5[prime] phosphodiester bonds (7,11-17). For example, polymerization of ImpA on 3[prime]-5[prime]-linked poly(U) or an oligo(U) template yields 2[prime]-5[prime]- and 3[prime]-5[prime]-linked oligo(A)s (14,24,25). The Pb2+- or Zn2+-catalyzed polymerization of ImpG on 3[prime]-5[prime]-linked poly(C) gives mainly 2[prime]-5[prime]- or 3[prime]-5[prime]-linked oligo(G)s respectively (15). 3[prime]-5[prime]-Linked poly- or oligo(C)s also promote template-directed polymerization of guanosine 5[prime]-phosphor-2-methylimidazolide to yield oligo(G)s with mainly a 3[prime]-5[prime] phosphodiester bond (11-14). The poly(U) template assists ligation of short-chain oligo(A)s to yield longer oligo(A)s with 2[prime]-5[prime] and 3[prime]-5[prime] phosphodiester bonds (26,27). Recently, Szostak et al. reported the non-enzymatic ligation of 3[prime]-5[prime]-linked hetero-oligoribonucleotides on the complementary 3[prime]-5[prime]-linked oligoribo-nucleotide template (28-30). Thus, non-enzymatic template-directed synthesis of oligonucleotides has been proved to proceed on 3[prime]-5[prime]-linked RNA templates. Very recently, Ertem and Ferris have reported that 2[prime]-5[prime]-linked oligo(C)s prepared from ImpC by montmorillonite catalysis serve as a template for formation of complementary oligo(G) from ImpG (18,19). The template ability of 2[prime]-5[prime]-linked oligo(C)s was also confirmed in the oligomerization of 2-MeImpG using hairpin templates (20). However, partially 2[prime]-5[prime]-linked oligo(U)s up to 9mer formed from ImpU by a montmorillonite catalyst failed in template-directed formation of oligo(A)s from ImpA, because hydrogen bonding in the A-U system is weaker than that in the G-C system and thus insufficient helix formation takes place between the oligo(U)s and ImpA (31). The current results establish that use of 2[prime]-5[prime]-linked diadenylate or longer oligoadenylates with a 2[prime]-5[prime] or 3[prime]-5[prime]-linked oligo(U) template overcomes the above limitation of the A-U system in the template-directed chain elongation of oligoribonucleotides.

Table 1. Template-directed oligomerization of ImpA2[prime]p5[prime]A and ligation of ImpA2[prime]p5[prime]A to the template strand [2[prime]-5[prime]]U10
Template Product Yield(%)a Type of formed linkage (%)
      2[prime]-5[prime] 3[prime]-5[prime]
[2[prime]-5[prime]]U10 (pA)4 21 81b 19b
  (pA)6 1.5    
  I(U10A2) 22 45c 55c
  II(U10A4) 10 64c 36c
      35d 65d
[3[prime]-5[prime]]U10 (pA)4 8 94b 6b
  (pA)6 0.5    
The reaction of ImpA2[prime]p5[prime]A (0.01 M) on the U10 template was run in the presence of 0.03 M MgCl2 and 0.2 M NaCl in N-ethylmorpholine buffer (pH 7.0) at 0°C for 28 days.
aYield was estimated from the peak area of products in the HPLC of the reaction mixture. The peak area of compounds I or II was allocated to the area due to [2[prime]-5[prime]]U10 and [2[prime]-5[prime]]A2 or A4 after correcting for their extinction coefficients. The yield data are expressed based on the starting ImpA2[prime]p5[prime]A.
bRatio of type of linkage between two [2[prime]-5[prime]](pA)2 units, pA2[prime]pA-pA2[prime]pA.
cRatio of type of linkage between [2[prime]-5[prime]]U10 and [2[prime]-5[prime]](pA)2 or (pA)4.
dRatio of type of linkage between [2[prime]-5[prime]]U10A2 and [2[prime]-5[prime]](pA)2.

In conclusion, the present finding demonstrates that long-chain 2[prime]-5[prime]-linked oligo(U)s formed by UO22+ ion-catalyzed polymerization of ImpU serves as a template for the template-directed synthesis of oligo(A)s from ImpA2[prime]p5[prime]A. [3[prime]-5[prime]]U10 also functions as a template for condensation of ImpA2[prime]p5[prime]A, but at lower efficiency compared with [2[prime]-5[prime]]U10. The observation that 2[prime]-5[prime]-linked oligo(U) is more active than 3[prime]-5[prime]-linked oligo(U) as a template for oligomerization of ImpA2[prime]pA is in contradiction of the finding by Prakash et al. that 3[prime]-5[prime]-linked oligo(C) is a more active template than the corresponding 2[prime]-5[prime]-linked oligo(C) for oligo-merization of 2-MeImpG (20). The discrepancy is probably due to the fact the they used 2-MeImpG monomers rather than 2[prime]-5[prime]-linked dimers, ImpA2[prime]pA, for template-directed oligomerization. Use of 2[prime]-5[prime]-linked RNAs as both the ligator strand and the template strand may promote template-directed oligomerization more efficiently than the chimeric system where 2[prime]-5[prime]-linked short oligoribonucleotides condense on 3[prime]-5[prime]-linked RNA templates. Joining of ImpA2[prime]p5[prime]A with the template [2[prime]-5[prime]]U10 also takes place to yield the longer oligonucleotides, which could also function as a template for other oligonucleotide syntheses. The 2[prime]-5[prime]-linked oligoribonucleotides are formed along with the 3[prime]-5[prime]-linked oligoribonucleotides by montmorillonite or metal ion catalysts or by any other model process of prebiotic synthesis of oligonucleotides. The 2[prime]-5[prime]- and 3[prime]-5[prime]-linked oligoribonucleotides could have formed a helix between complementary strands (22,32,33) and initiated template-directed ligation in the prebiotic era, prior to the advent of exclusively 3[prime]-5[prime]-linked RNA, which was selected due to its advantage of slow hydrolysis when the helix is formed (34) and/or of high helix forming ability (22,32,33) and conformational flexibility (21) compared with 2[prime]-5[prime]-linked RNA.

REFERENCES

1. Gilbert,W. (1986) Nature, 319, 618.

2. Joyce,G.F. and Orgel,L.E. (1993) In Gesteland,R.F. and Atkins,J.F. (eds), The RNA World. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp. 1-25.

3. Sawai,H. (1976) J. Am. Chem. Soc., 98, 7037-7039. MEDLINE Abstract

4. Sawai,H. (1988) J. Mol. Evol., 27, 181-186. MEDLINE Abstract

5. Sawai,H., Kuroda,K. and Hojo,T. (1989) Bull. Chem. Soc. Jpn., 62, 2018-2023.

6. Sawai,H., Higa,K. and Kuroda,K. (1992) J. Chem. Soc. Perkin Trans., I, 505-508.

7. Orgel,L.E. and Lohrmann,R. (1973) Acc. Chem. Res., 7, 368-377.

8. Ferris,J.P. and Ertem,G. (1992) Science, 257,1387-1389. MEDLINE Abstract

9. Ferris,J.P. and Ertem,G. (1993) J. Am. Chem. Soc., 115, 12270-12275.

10. Kawamura,K. and Ferris,J.P. (1994) J. Am. Chem. Soc., 116, 7564-7572.

11. Inoue,T. and Orgel,L.E. (1981) J. Am. Chem. Soc., 103, 7666-7667.

12. Inoue,T. and Orgel,L.E. (1981) Science, 219, 859-862.

13. Joyce,G.F. (1987) Cold Spring Harbor Symp. Quant. Biol., LII, 41-51. MEDLINE Abstract

14. Orgel,L.E. (1995) Acc. Chem. Res., 28, 109-118.

15. Lohrmann,R., Bridson,P.K. and Orgel,L.E. (1980) Science, 208, 1464-1465. MEDLINE Abstract

16. Lohrmann,R. and Orgel,L.E. (1978) Tetrahedron, 34, 853-855.

17. Orgel,L.E. (1992) Nature, 358, 203-209. MEDLINE Abstract

18. Ertem,G. and Ferris,J.P. (1996) Nature, 379, 238-240. MEDLINE Abstract

19. Ertem,G. and Ferris,J.P. (1997) J. Am. Chem. Soc., 119, 7197-7201.

20. Prakash,T.P., Roberts,C. and Switzer,C. (1997) Angew. Chem. Int. Edn English, 36, 1522-1523.

21. Sawai,H., Kuroda,K., Seki,J. and Ozaki,H. (1996) Biopolymers, 39, 173-182. MEDLINE Abstract

22. Sawai,H., Seki,J. and Ozaki,H. (1996) J. Biomol. Struct. Dynam., 13, 1043-1051.

23. Lohrmann,L.E. and Orgel,L.E. 1979) J. Mol. Biol., 113, 193-198.

24. Sleeper,H.L., Lohrmann,R. and Orgel,L.E. (1979) J. Mol. Evol., 13, 203-214. MEDLINE Abstract

25. Sawai,H. (1981) J. Mol. Evol., 17, 48-51.

26. Lohrmann,L.E. and Orgel,L.E. (1979) J. Mol. Evol., 12, 237-257. MEDLINE Abstract

27. Kanaya,E. and Yanagawa,H. (1986) Biochemistry, 25, 7423-7430. MEDLINE Abstract

28. Bartel,D.P. and Szostak,J.W. (1993) Science, 261, 1411-1418. MEDLINE Abstract

29. Rohatgi,R., Bartel,D.P. and Szostak,J.W. (1996) J. Am. Chem. Soc., 118, 3332-3339.

30. Rohatgi,R., Bartel,D.P. and Szostak,J.W. (1996) J. Am. Chem. Soc., 118, 3340-3344.

31. Ding,P.Z., Kawamura,K. and Ferris,J.P. (1996) Origins Life Evol. Biosphere, 26, 151-171.

32. Giannaris,P G. and Damha,M.J. (1993) Nucleic Acid Res., 21, 4742-4749. MEDLINE Abstract

33. Kierzek,R., He,L. and Turner,D.H. (1992) Nucleic Acid Res., 20, 1685-1690. MEDLINE Abstract

34. Usher,D.A. and McHall,A.H. (1976) Proc. Natl. Acad. Sci. USA, 73, 1149-1153. MEDLINE Abstract

*To whom correspondence should be addressed. Tel: +81 277 30 1220; Fax: +81 277 30 1224; Email: sawai@gunma-u.ac.jp

This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 4 Jun 1998
Copyright©Oxford University Press, 1998.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
G. Costanzo, S. Pino, F. Ciciriello, and E. Di Mauro
Generation of Long RNA Chains in Water
J. Biol. Chem., November 27, 2009; 284(48): 33206 - 33216.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
L. F. Landweber and I. D. Pokrovskaya
Emergence of a dual-catalytic RNA with metal-specific cleavage and ligase activities: The spandrels of RNA evolution
PNAS, January 5, 1999; 96(1): 173 - 178.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
I. A. Kozlov, S. Pitsch, and L. E. Orgel
Oligomerization of activated D- and L-guanosine mononucleotides on templates containing D- and L-deoxycytidylate residues
PNAS, November 10, 1998; 95(23): 13448 - 13452.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (112K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (13)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Sawai, H.
Right arrow Articles by Ozaki, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sawai, H.
Right arrow Articles by Ozaki, H.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?