Shape-selective recognition of a model Okazaki fragment by geometrically-constrained bis-distamycins

Okazaki fragments represent interesting targets for the design of anticancer drugs because of their selective occurrence during DNA replication, a process often elevated in aggressive malignancies. Structural studies have indicated a bend occurs in the helical axis at the junction region (JR) that joins the DNA duplex region (DDR) and the RNA-DNA hybrid duplex region (HDR) of model Okazaki fragments. To identify a structural motif that provides a shape complementary to the Okazaki fragment minor groove, we have investigated the binding of geometrically-constrained bis-distamycins to a model Okazaki fragment, [OKA], with a sequence derived from the genome of simian virus 40 (SV40). Both the JR and the DDR of [OKA] contain consecutive A/T base pairs that could accommodate distamycin binding. Of the six bis-distamycins selected for analysis, the two with a para configuration of the distamycins on the benzene or pyridine scaffold bound [OKA] tightly (Kd approximately 10(-6) M from gel-shift assays; Kd approximately 10(-8) M from deltaT(M)) while the four with a meta orientation did not bind. The two mono-distamycins studied also did not bind [OKA]. Molecular modeling of the complex between the para bis-distamycin MT-9 and [OKA] revealed MT-9 adopted an S- shape complementary to the minor groove of the model Okazaki fragment.     

Sequence specificity for the initiation of RNA-primed simian virus 40 DNA synthesis in vivo

Analysis of the nucleotide sequences at the 5' ends of RNA-primed nascent DNA chains (Okazaki fragments) and of their locations in replicating simian virus 40 (SV40) DNA revealed the precise nature of Okazaki fragment initiation sites in vivo. The primary initiation site for mammalian DNA primase was 3'-purine-dT-5' in the DNA template and the secondary site was 3'-purine-dC-5', with the 5' end of the RNA primer complementary to either the dT or dC. The third position of the initiation site was variable with a preference for dT or dA. About 81% of the available 3'-purine-dT-5' sites and 20% of the 3'-purine-dC-5' sites were used. Purine-rich sites, such as PuPuPu and PyPuPu , were excluded. The 5'-terminal ribonucleotide composition of Okazaki fragments corroborated these conclusions. Furthermore, the length of individual RNA primers was not unique, but varied in size from six to ten bases with some appearing as short as three bases and some as long as 12 bases, depending on the initiation site used. This result was consistent with the average size (9 to 11 bases) of RNA primers isolated from specific regions of the genome. Excision of RNA primers did not appear to stop at the RNA-DNA junction, but removed a variable number of deoxyribonucleotides from the 5' end of the nascent DNA chain. Finally, only one-fourth of the replication forks contained an Okazaki fragment, and the distribution of their initiation sites between the two arms revealed that Okazaki fragments were initiated exclusively (99%) on retrograde DNA templates. The data obtained at two genomic sites about 350 and 1780 bases from ori were essentially the same as that reported for the ori region (Hay & DePamphilis , 1982), suggesting that the mechanism used to synthesize the first DNA chain at ori is the same as that used to synthesize Okazaki fragments throughout the genome.     

Structure of the histone mRNA hairpin required for cell cycle regulation of histone gene expression

Expression of replication-dependent histone genes requires a conserved hairpin RNA element in the 3' untranslated regions of poly(A)-less histone mRNAs. The 3' hairpin element is recognized by the hairpin-binding protein or stem-loop-binding protein (HBP/SLBP). This protein-RNA interaction is important for the endonucleolytic cleavage generating the mature mRNA 3' end. The 3' hairpin and presumably HBP/SLBP are also required for nucleocytoplasmic transport, translation, and stability of histone mRNAs. RNA 3' processing and mRNA stability are both regulated during the cell cycle. Here, we have determined the three-dimensional structure of a 24-mer RNA comprising a mammalian histone RNA hairpin using heteronuclear multidimensional NMR spectroscopy. The hairpin adopts a novel UUUC tetraloop conformation that is stabilized by base stacking involving the first and third loop uridines and a closing U-A base pair, and by hydrogen bonding between the first and third uridines in the tetraloop. The HBP interaction of hairpin RNA variants was analyzed in band shift experiments. Particularly important interactions for HBP recognition are mediated by the closing U-A base pair and the first and third loop uridines, whose Watson-Crick functional groups are exposed towards the major groove of the RNA hairpin. The results obtained provide novel structural insight into the interaction of the histone 3' hairpin with HBP, and thus the regulation of histone mRNA metabolism.     

Hydration of [d(CGC)r(aaa)d(TTTGCG)](2)

We have studied the hydration and dynamics of RNA C2'-OH in a DNA. RNA hybrid chimeric duplex [d(CGC)r(aaa)d(TTTGCG)](2). Long-lived water molecules with correlation time tau(c) larger than 0.3 ns were found close to the RNA adenine H2 and H1' protons in the hybrid segment. A possible long-lived water molecule was also detected close to the methyl group of 7T in the RNA-DNA junction but not to the other two thymine bases (8T and 9T). This result correlates with the structural studies that only DNA residue 7T in the RNA-DNA junction adopts an O4'-endo sugar conformation (intermediate between B-form and A-form), while the other DNA residues including 3C in the DNA-RNA junction, adopt C1'-exo or C2'-endo conformations (in the B-form domain). Based on the NOE cross-peak patterns, we have found that RNA C2'-OH tends to orient toward the O3' direction, forming a possible hydrogen bond with the 3'-phosphate group. The exchange rates for RNA C2'-OH were found to be around 5-20 s(-1), compared to 26.7(+/-13.8) s(-1) reported previously for the other DNA.RNA hybrid duplex. This slow exchange rate may be due to the narrow minor groove width of [d(CGC)r(aaa)d(TTTGCG)](2), which may trap the water molecules and restrict the dynamic motion of hydroxyl protons. The distinct hydration patterns of the RNA adenine H2 and H1' protons and the DNA 7T methyl group in the hybrid segment, as well as the orientation and dynamics of the RNA C2'-OH protons, may provide a molecular basis for further understanding the structure and recognition of DNA.RNA hybrid and chimeric duplexes.     

Synthesis and hybridization properties of oligonucleotides containing (2S,3R)-9-(2,3,4-trihydroxybutyl)adenine

The synthesis and properties of oligonucleotides (ONs) containing 9-(2,3,4-trihydroxybutyl)adenine, A(C2) and A(C3), are described. The ON containing A(C2) involves the 3'-->4' and 3-->5' phosphodiester linkages in the strand, whereas that containing A(C3) possesses the 3'-->4' and 2'-->5' phosphodiester linkages. It was found that incorporation of the analogs, A(C2) or A(C3), into ONs significantly reduces the thermal and thermodynamic stabilities of the ON/DNA duplexes, but does not largely decrease the thermal and thermodynamic stabilities of the ON/RNA duplexes as compared with the case of the ON/DNA duplexes. It was revealed that the base recognition ability of A(C2) is greater than that of A(C3) in the ON/RNA duplexes.     

RNA involvement in T4 DNA synthesis in toluene-treated cells

In T4-infected cells made permeable with toluene, pulses with [(alpha-32P deoxyribonucleoside triphosphates demonstrated covalent linkage of RNA to DNA of the Okazaki fragments. Analysis of the transfer of the 32P label to the 2'(3') ribonucleoside monophosphates indicated that the 3'-end of the RNA primer is heterogeneous. The most frequently encountered ribonucleotide was rCMP, but also transfer to rUMP, rAMP and rGMP occurred at different frequencies. In contrast, no heterogeneity was observed for the deoxyribonucleoside at the RNA-DNA junction. Of all the [to-32P] deoxyribonucleoside triphosphates tested, transfer of the 32P label to 2'(3') rNMPs was predominant when [alpha32P] dGTP was the substrate, indicating that the deoxyribonucleoside most frequently encountered at the RNA-DNA linkage is dG. These observations suggest that the starts for the Okazaki fragments may occur at unique sites of the T4 genome.     

Metabolism of Okazaki fragments during simian virus 40 DNA replication.

Essentially all of the Okazaki fragments on replicating Simian virus 40 (SV40)DNA could be grouped into one of three classes. Class I Okazaki fragments (about 20%) were separated from longer nascent DNA chains by a single phosphodiester bond interruption (nick) and were quantitatively identified by treating purified replicating DNA with Escherichia coli DNA ligase and then measuring the fraction of Okazaki fragments joined to longer nascent DNA chains. Similarly, class II Okazaki fragments (about 30%) were separated by a region of single-stranded DNA template (gap) that could be filled and sealed by T4 DNA polymerase plus E. coli DNA ligase, and class III fragments (about 50%) were separated by RNA primers that could be removed with E. coli DNA olymerase I, allowing the fragments to be joined with E. coli DNA ligase. These results were obtained with replicating SV40 DNA that had been briefly labeled with radioactive precursors in either intact cells or isolated nuclei. When isolated nuclei were further incubated in the presence of cytosol, all of the Okazaki fragments were converted into longer DNA strands as expected for intermediates in DNA synthesis. However, when washed nuclei were incubated in the abscence of cytosol, both class I and class II Okazaki fragments accumulated despite the excision of RNA primers: class III Okazaki fragments and RNA-DNA covalent linkages both disappeared at similar rates. These data demonstrate the existence of RNA primers in whole cells as well as in isolated nuclei, and identify a unique gap-filling step that is not simply an extension of the DNA chain elongation process concomitant with the excision of RNA primers. One or more factos found in cytosol, in addition to DNA polymerase alpha, are specifically involved in the gap-filling and ligation steps. The sizes of mature Okazaki fragments (class I) and Okazaki fragments whose synthesis was completed by T4 DNA polymerase were measured by gel electrophoresis and found to be broadly distributed between 40 and 290 nucleotides with an average length of 135 nucleotides. Since 80% and 90% of the Okazaments does not occur at uniformly spaced intervals along the DNA template. During the excision of RNA primers, nascent DNA chains with a single ribonucleotide covalently attached to the 5' terminus were identified as transient intermediates. These intermediates accumulated during excision of RNA primers in the presence of adenine 9-beta-D-arabinoside 5'-triphosphate, and those Okazaki fragments blocked by RNA primers (class III) were found to have originated the farthest from the 5' ends of long nascent DNA strands. Thus, RNA primers appear to be excised in two steps with the second step, removal of the final ribonucleotide, being stimulated by concomitant DNA synthesis. These and other data were used to construct a comprehensive metabolic pathway for the initiation, elongation, and maturation of Okazaki fragments at mammalian DNA replication forks.     

Two-step cleavage of hairpin RNA with 5' overhangs by human DICER

Background  

DICER is an RNase III family endoribonuclease that processes precursor microRNAs (pre-miRNAs) and long double-stranded RNAs, generating microRNA (miRNA) duplexes and short interfering RNA duplexes with 20~23 nucleotides (nts) in length. The typical form of pre-miRNA processed by the Drosha protein is a hairpin RNA with 2-nt 3' overhangs. On the other hand, production of mature miRNA from an endogenous hairpin RNA with 5' overhangs has also been reported, although the mechanism for this process is unknown.     

Effect of DNA local conformation on the affinity and binding specificity of bis-netropsins to DNA

The interaction of short nucleotide duplexes with bis-netropsins, in which netropsin fragments are linked in the tail-to-tail orientation via cis-diammineplatinum group (<--Nt-Pt(NH3)-Nt-->) or aliphatic pentamethylene chain (<--Nt-(CH2)5-Nt-->), has been studied. Both the bis-netropsins have been shown to bind to DNA oligomer 5'-CCTATATCC-3' (I) as a hairpin with parallel orientation of netropsin fragments in 1:1 stoichiometry. Monodentate binding has been detected upon binding of bis-netropsins to other duplexes of sequences 5'-CCXCC-3'--where X = TTATT (II), TTAAT (III), TTTTT (IV), and AATTT (V)--along with the binding of bis-netropsins as a hairpin. The formation of dimeric antiparallel motif between the halves of two bound bis-netropsin molecules has been observed in the complexes of <--Nt-(CH2)5-Nt--> with DNA oligomers IV and V. The ratio of binding constant of bis-netropsin as a hairpin (K2) to monodentate binding constant (K1) has been shown to correlate with the width and/or conformational lability of DNA in the binding site. The share of bis-netropsin bound as a hairpin decreases in the order: TATAT > TTATT > TTAAT > TTTTT > AATTT, whereas the contribution of monodentate binding rises. The minimal strong binding site for <--Nt-Pt(NH3)-Nt--> and <--Nt-(CH2)5-Nt--> binding as a hairpin has been found to be DNA duplex 5'-CGTATACG-3'.     

Junction ribonuclease: a ribonuclease HII orthologue from Thermus thermophilus HB8 prefers the RNA-DNA junction to the RNA/DNA heteroduplex

The genome of an extremely thermophilic bacterium, Thermus thermophilus HB8, contains a single ORF (open reading frame) encoding an RNase-HII-like sequence. Despite the presence of significant amino acid sequence identities with RNase (ribonuclease) HII enzymes, the ORF TTHA0198 could not suppress the temperature-sensitive growth defect of an RNase-H-deficient Escherichia coli mutant and the purified recombinant protein could not cleave an RNA strand of an RNA/DNA heteroduplex, suggesting that the TTHA0198 exhibited no RNase H activity both in vivo and in vitro. When oligomeric RNA-DNA/DNAs were used as a mimic substrate for Okazaki fragments, however, the protein cleaved them only at the 5' side of the last ribonucleotide at the RNA-DNA junction. In fact, the TTHA0198 protein prefers the RNA-DNA junction to the RNA/DNA hybrid. We have referred to this activity as JRNase (junction RNase) activity, which recognizes an RNA-DNA junction of the RNA-DNA/DNA heteroduplex and cleaves it leaving a mono-ribonucleotide at the 5' terminus of the RNA-DNA junction. E. coli and Deinococcus radiodurans RNases HII also cleaved the RNA-DNA/DNA substrates at the same site with a different metal-ion preference from that for RNase H activity, implying that the enzymes have JRNase activity as well as RNase H activity. The specialization in the JRNase activity of the RNase HII orthologue from T. thermophilus HB8 (Tth-JRNase) suggests that the JRNase activity of RNase HII enzymes might be independent of the RNase H activity.     

Cytarabine-induced destabilization of a model Okazaki fragment.

Cytarabine is a potent anticancer drug that interferes with elongation of the lagging strand at the replication fork during DNA synthesis. The effects of cytarabine substitution on the structural and thermodynamic properties of a model Okazaki fragment were investigated using UV hyperchromicity and 1H NMR spectroscopy to determine how cytarabine alters the physicochemical properties of Okazaki fragments that are intermediates during DNA replication. Two model Okazaki fragments were prepared corresponding to a primary initiation site for DNA replication in the SV40 viral genome. One model Okazaki fragment consisted of five ribo- and seven deoxyribonucleotides on the hybrid strand, together with its complementary (DNA) strand. The second model Okazaki fragment was identical to the first with the exception of cytarabine substitution for deoxycytidine at the third DNA nucleotide of the hybrid strand. Thermodynamic parameters for the duplex to single strand transition for each model Okazaki fragment were calculated from the concentration dependence of the T m at 260 nm. Cytarabine significantly decreased the stability of this model Okazaki fragment, decreasing the melting temperature from 46.8 to 42.4 degrees C at a concentration of 1.33 x 10(-5) M. The free energy for the duplex to single strand transition was 1.2 kcal/mol less favorable for the cytarabine-substituted Okazaki fragment relative to the control at 37 degrees C. Analysis of the temperature dependence of the imino1H resonances for the two duplexes demonstrated that cytarabine specifically destabilized the DNA:DNA duplex portion of the model Okazaki fragment. These results are consistent with inhibition of lagging strand DNA synthesis by cytarabine substitution resulting from destabilization of the DNA:DNA duplex portion of Okazaki fragments in vivo .     

The effects of loop size on Sac7d-hairpin DNA interactions

Hairpin structure is a common feature of DNA molecules. They are located near functional loci, such as regulation and promotion sites, as well as in cruciform structures, and they provide potential binding sites for endogenous proteins. The effects of different hairpin loops that are composed of one to five thymidines, designated as L1-L5, and have a common self-complementary stem, CTATATAG, on the interactions with Sac7d were studied. In thermostability studies, Sac7d stabilized a tetra-loop hairpin DNA and hairpin DNA with GTTC tetra-loop regions better than it stabilized tri- and penta-loops. Circular dichroism (CD) spectra showed that hairpins retained primarily a B-type conformation upon Sac7d binding. Intermolecular interactions between hairpins were likely decreased, due to the Sac7d-induced kinks, as shown by an increase at 220nm in the CD spectra. Surface plasmon resonance (SPR) observations suggested that the rates of Sac7d binding to hairpin DNA depend on the loop size of the hairpin duplexes. At a fixed stem length, Sac7d binds to tetra-loop hairpin DNA duplexes with a higher association rate and lower dissociation rate, compared with their tri- and penta-loop counterparts. In addition, the tri-loop and GTC tri-loop hairpin DNA had lower affinity for Sac7d because of the smaller and tighter loop size. Our study indicates that Sac7d binding affinity to hairpin DNA is primarily determined by loop size and stem integrity, and the results presented here provide a model for studies concerning other minor groove DNA-binding proteins that kink hairpin DNA.     

The molecular structure of Okazaki fragment r(CGCA)d(AAAAAGCG):d(CGCTTTTTTGCG) in solution

The hetero duplex molecule, r(CGCA)d(AAAAAGCG):d(CGCTTTTTTGCG) which corresponds to Okazaki fragment was synthesized and its molecular structure has been analyzed by NMR study. The RNA strand of RNA-DNA hybrid region adopts A-form and DNA strand of the same region deviates from the standard B-form. The conformation of DNA-DNA duplex segment belongs to B-form. The hybrid-DNA duplex junction shows a structural discontinuities, A-B junction. The same conformational characteristic of oligo(dA): oligo(dT) tract as that of DNA oligomer which has same base sequence has been observed.     

Solution structure of an arabinonucleic acid (ANA)/RNA duplex in a chimeric hairpin: comparison with 2′-fluoro-ANA/RNA and DNA/RNA hybrids

Hybrids of RNA and arabinonucleic acid (ANA) as well as the 2′-fluoro-ANA analog (2′F-ANA) were recently shown to be substrates of the enzyme RNase H. Although RNase H binds to double-stranded RNA, no cleavage occurs with such duplexes. Therefore, knowledge of the structure of ANA/RNA hybrids may prove helpful in the design of future antisense oligonucleotide analogs. In this study, we have determined the NMR solution structures of ANA/RNA and DNA/RNA hairpin duplexes and compared them to the recently published structure of a 2′F-ANA/RNA hairpin duplex. We demonstrate here that the sugars of RNA nucleotides of the ANA/RNA hairpin stem adopt the C3′-endo (north, A-form) conformation, whereas those of the ANA strand adopt a ‘rigid’ O4′-endo (east) sugar pucker. The DNA strand of the DNA/RNA hairpin stem is flexible, but the average DNA/RNA hairpin structural parameters are close to the ANA/RNA and 2′F-ANA/RNA hairpin parameters. The minor groove width of ANA/RNA, 2′F-ANA/RNA and DNA/RNA helices is 9.0 ± 0.5 Å, a value that is intermediate between that of A- and B-form duplexes. These results rationalize the ability of ANA/RNA and 2′F-ANA/RNA hybrids to elicit RNase H activity.     

Structure and stability of the consecutive stereoregulated chiral phosphorothioate DNA duplex

The duplex structures of the stereoregulated phosphorothioate DNAs, [R(p),R(p)]- and [S(p),S(p)]-[d(GC(ps)T(ps)ACG)] (ps, phosphorothioate; PS-DNA), with their complementary RNA have been investigated by combined use of (1)H NMR and restrained molecular dynamics calculation. Compared to those obtained for the unmodified duplex structures (PO-DNA.RNA), the NOE cross-peak intensities are virtually identical for the PS-DNA.RNA hybrid duplexes. The structural analysis on the basis of the NOE restraints reveals that all of the three DNA.RNA duplexes take a A-form conformation and that there is no significant difference in the base stacking for the DNA.RNA hybrid duplexes. On the other hand, the NOE cross-peak intensities of the protons around the central T(ps)A step of the PS-DNA.DNA duplexes are apparently different from those of PO-DNA. DNA. The chemical shifts of H8/6 and H1' at the T(ps)A step are also largely different among PS-DNA.DNAs and PO-DNA.DNA, suggesting that the DNA.DNA structure is readily changed by the introduction of the phosphorothioate groups to the central T(p)A step. The structure calculations indicate that all of these DNA.DNA duplexes are B-form although there exist some small differences in helical parameters between the [R(p),R(p)]- and [S(p),S(p)]PS-DNA.DNA duplexes. The melting temperatures (T(m)) were determined for all of the duplexes by plotting the chemical shift change of isolated peaks as a function of temperature. For the PS-DNA.RNA hybrid duplexes, the [S(p),S(p)] isomer is less stable than the [R(p),R(p)] isomer while this trend is reversed for the PS-DNA.DNA duplexes. Consequently, although the PS-DNA.RNA duplexes take the similar A-form structure, the duplex stability is different between PS-DNA.RNA duplexes. The stability of the DNA.RNA duplexes may not be governed by the A-form structure itself but by some other factors such as the hydration around the phosphorothioate backbone, although the T(m) difference of the DNA.DNA duplexes could be explained by the structural factor.     

Oligoribonucleotides containing 2',5'-phosphodiester linkages exhibit binding selectivity for 3',5'-RNA over 3',5'-ssDNA.

Oligoribonucleotides containing 2',5'-phosphodiester linkages have been synthesized on a solid support by the 'silyl-phosphoramidite' method. The stability of complexes formed between these oligonucleotides and complementary 3',5'-RNA strands have been studied using oligoadenylates and a variety of oligonucleotides of mixed base sequences including phosphorothioate backbones. In many cases, particularly for 2',5'-linked adenylates, the UV melting profiles are quite sharp and exhibit large hyperchromic changes. Substituting a few 3',5'-linkages with the 2',5'-linkage within an oligomer lowers the Tm of the complex and the degree of destabilization depends on the neighboring residues and neighboring linkages. The 2',5'-linked oligoribonucleotides prepared in this study exhibited remarkable selectivity for complementary single stranded RNA over DNA. For example, in 0.01 M phosphate buffer--0.10 M NaCl (pH 7.0), no association was observed between 2',5'-r(CCC UCU CCC UUC U) and its Watson-Crick DNA complement 3',5'-d(AGAAGGGAGAGGG). However, 2',5'-r(CCC UCU CCC UUC U) with its RNA complement 3',5'-r(AGAAGGGAGAGGG) forms a duplex which melts at 40 degrees C. The decamer 2',5'-r(Ap)9A forms a complex with both poly dT and poly rU but the complex [2',5'-r(Ap)9A]:[poly dT] is unstable (Tm, -1 degree C) and is seen only at high salt concentrations. In view of their unnatural character and remarkable selectivity for single stranded RNA, 2',5'-oligo-RNAs and their derivatives may find use as selective inhibitors of viral mRNA translation, and as affinity ligands for the purification of cellular RNA.     

Flexibility in RNA priming of Okazaki pieces at the E. coli replication fork

We present results which suggest considerable flexibility in the RNA priming of Okazaki pieces at the E. coli replication fork. Using film lysates on cellophane discs, we have identified RNA at the 5' ends of Okazaki pieces. All four ribonucleotides are found to be present at the RNA-DNA junction if all four ribonucleoside triphosphates are used. However, if only ATP, or ATP and GTP are used, then only 2' (3')AMP, or 2' (3')AMP and 2' (3')GMP are found at the RNA-DNA junction. A nearest neighbor analysis of RNA associated with Okazaki pieces using alpha 32P-CTP as a probe shows a similar dependence of nearest neighbor composition on the ribonucleoside triphosphate composition of the incubation mixture. Thus, the nucleotide composition of the RNA primers at the ends of Okazaki pieces varies as a function of the ribonucleoside triphosphates available.     

Discontinuous synthesis of both strands at the growing fork during polyoma DNA replication in vitro.

In discontinuous polyoma DNA replication, the synthesis of Okazaki fragments is primed by RNA. During viral DNA synthesis in nuclei isolated from infected cells, 40% of the nascent short DNA fragments had the polarity of the leading strand which, in theory, could have been synthesized by a continuous mechanism. To rule out that the leading strand fragments were generated by degradation of nascent DNA, they were further characterized. DNA fragments from a segment of the genome which replication forks pass in only one direction were strand separated. The sizes of the fragments from both strands were similar, suggesting that one strand was not specifically degraded. Most important, however, the majority of the Okazaki fragments of both strands were linked to RNA at their 5' ends. For identification, the RNA was labeled at the 5' ends by [beta-32P]GTP, internally by [3H]CTP, [3H]GTP, and [3H]UTP, or at the 3' ends by 32P transfer from adjacent [32P]dTMP residues. All three kinds of labeling indicated that an equal proportion of DNA fragments from the two strands was linked to RNA primers.