DNA Recognition by Triple Helix Formation

DAVID A. RUSLING, TOM BROWN AND KEITH R. FOX

1.1 Introduction

Oligonucleotides can bind in the major groove of double-stranded DNA by forming hydrogen bonds with exposed groups on the base pairs, generating a triple-helical structure (Figure 1A). This was first demonstrated nearly 50 years ago by Rich and co-workers by mixing the synthetic polyribonucleotides polyU and polyA in a 2:1 ratio. Further studies showed that polyC and polyG can generate a similar structure under conditions of low pH and a variety of DNA and RNA triple-stranded structures have since been identified.

Since these complexes form in a sequence-specific fashion they can be used to target unique sequences. By knowing the rules that govern triplex formation, it should be possible to design oligonucleotides to interact with any desired DNA sequence. The formation of intermolecular triple-helical DNA therefore has a number of applications including inhibition of gene transcription, site-directed mutagenesis, and various biotechnological applications.

1.1.1 Triplets and Triplex Motifs

Triplex-forming oligonucleotides bind in the DNA major groove and make specific contacts with the purine strand of the duplex. The binding can be either parallel or antiparallel to the target strand, depending on the base composition of the oligonucleotide.

Pyrimidine-rich oligonucleotides bind under low pH conditions in a parallel orientation to the purine strand of the target duplex, with T and protonated C forming Hoogsteen hydrogen bonds with AT and GC base pairs, respectively. This generates the base triplets T.AT and C+.GC as shown in Figure 1B. (The notation X.ZY refers to a triplet in which the third-strand base X interacts with the duplex base pair ZY, forming hydrogen bonds to base Z). These triplets are isomorphic, that is if the C-1' atoms of their Watson-Crick base pairs are superimposed, the positions of the C-1' atoms of the third strand are almost identical. This minimizes backbone distortion of both the third strand and duplex between adjacent triplets. It is also possible to form a G.GC triplet within this motif, though this is not isomorphic with T.AT and C+.GC (Figure 1B (iii)). Most of this chapter will concentrate on the parallel motif.

Purine-rich oligonucleotides bind in an antiparallel orientation to the purine strand of the target duplex, with A and G forming reverse-Hoogsteen hydrogen bonds with AT and GC base pairs respectively, generating A.AT and G.GC triplets (see Figure 1C). In contrast to the parallel triplets, A.AT and G.GC triplets are not isomorphic, leading to structural distortions at junctions between each triplet. As a consequence of this and other factors, the antiparallel motif is often less stable than the parallel motif. T.AT triplets can also adopt an antiparallel orientation.

Since G.GC and T.AT triplets can be formed in both the parallel and antiparallel motifs GT-containing oligonucleotides can be designed to bind in either orientation, parallel or antiparallel. The backbone distortion imposed by the non-isomorphic nature of these two triplets means that the most stable orientation is dependent on the number of GpT and TpG steps.

Triplex formation in the parallel motif suffers from the requirement for low pH, which is necessary for protonation of cytosine at N3. Without protonation the C.GC triplet only contains one hydrogen bond between the exocyclic N4 of cytosine and 6-keto group of guanine. The pKa of cytosine is around 4.5, though this is increased on triplex formation, and is higher in the centre than the termini of a triplex. Runs of contiguous cytosine residues are destabilizing as they decrease the pKa. A large number of cytosine analogues have been prepared to overcome this problem and are described in a later section.

Several reports have suggested that C+.GC is more stable than T.AT. This is attributed to electrostatic interactions between the positive charge of cytosine and the negatively charged phosphodiester backbone and/or favourable stacking interactions between the charged base and the π-stack.

1.2 Strategies to Increase Triplex Stability

1.2.1 Sugar Modifications

The affinity of a third strand for its target is affected by its ability to adopt N- or S-type conformations, since the former require less distortion of the duplex purine strand upon triplex formation. This explains why RNA TFOs have a higher affinity for duplex DNA than those composed of DNA. Oligonucleotide modifications that favour N-type sugars are therefore expected to produce more stable triplexes and some examples are shown in Figure 2. The addition of an electronegative group at the 2'-position of the sugar, as in RNA, strongly favours the N-type sugar pucker due to the gauche effect and the 2'-O-methyl modification (Figure 2a) enhances triplex stability. NMR studies have confirmed that the 2'-methoxy group increases triplex stability by reducing the distortion of the duplex purine strand and enhancing the rigidity of the triplex.

Other modifications can also be used to restrict the sugar pucker and reduce the rotational freedom of the sugar phosphate backbone. The best characterized of these modifications is locked nucleic acid (LNA), which is also known as bridged nucleic acid (BNA), in which a 2'-O,4'-C methylene bridge is used to constrain the sugar to N-type (Figure 2b). This modification was developed independently by the Wengel and Imanishi groups for use in antisense or antigene applications, respectively. TFOs that contain LNA residues every 2-3 nucleotides are markedly more stable than their unmodified counterparts. Two further derivatives have also been developed from this. ENA contains an additional carbon in the bridge and unlike LNA can be fully substituted into TFOs (Figure 2c). In addition, 3'-amino-2',4'-LNA combines the LNA sugar with the N3'-P5' modification (considered below), though triplexes with this analogue are no more stable than those formed with LNA.

The bicyclo and tricyclo...