Hoogsteen pairing

Fig. 5. Possible base-pairing interactions for the quadruplex homodimer. Hoogsteen pairings of the parallel stranded PtDNA duplex requires protonation of the GC pair. Adapted from Ref. (28).

The symbol denotes Watson-Crick pairing, while - means Hoogsteen pairing CH+ is protonated C (cytosine). 

For example, Hoogsteen proposed an alternative A-T pairing using the 6-NH2 and N-7 of adenine.34 Here the distance spanned by the base pair, between the C-T sugar carbons, is 0.88 nm, less than the 1.08 ran of the Watson-Crick pairs. Duplexes of certain substituted poly (A) and poly (U) chains contain only Hoogsteen base pairs35 and numerous X-ray structure determinations have established that Hoogsteen pairs... 

Figure 5-7 Two base triplets that form in triple-stranded DNA and involve both Watson-Crick and Hoogsteen base pairing. (A) The triplet T A T, where the T (marked T ) of the third strand is hydrogen bonded as a Hoogsteen pair...

Hoogsteen pairs were first observed in nature in transfer RNA molecules (Fig. 5-31). These molecules contain mostly Watson-Crick base pairs but there are also two reversed Hoogsteen pairs. One of them, between U8 and A14, is invariant in all tRNAs studied. Hoogsteen pairing also occurs in four-stranded DNA, which has important biological functions. A G quartet from a DNA tetraplex held together by Hoogsteen base pairs is shown in Fig. 5-8. 

While tRNAs consist largely of loops and stems containing Watson-Crick base pairs, they also contain Hoogsteen pairs, wobble pairs, and triplets such as the following. 

An example of Hoogsteen pairing between T and an A carrying a Pd(II) at N3 has recently been reported by Houlton and co-workers (122). This structure seems to confirm earlier predictions (110) according to which a metal at N3 of A weakens Watson-Crick pairing. 

In the case of G C Hoogsteen pairing and transplatin, the platinated pair fits almost exactly the requirement of normal DNA as far as interglycosyl distances are concerned. The platinated cytosine residue can be replaced by a platinated guanine residue but the cross-linking reaction requires a switch from the anti- to yvn-conformation for this residue as shown on Fig. 9. 

Fig. 9. Base triplets formed between a Watson-Crick base pair G C and a third base G or C Hoogsteen-pairing) (left) or a transplatin-modified G or C (right)...

Of particular interest is the comparison of the four types of TA dimers that were examined. The calculations " indicate a clear preference for Hoogsteen pairing over Watson-Crick in both cases, it does not matter whether these are standard or reverse form. One cannot attribute the discrepancy to the type of H-bonds present. In all four instances, there is one NH-N and one NH—0 bond. The adenine homodimer and thymine homodimer are both slightly less strongly bound. 

An X-ray crystallography structure of the duplex from d(ATA UAT) and d( ATAT AT) demonstrates that an alternative to the classical B-DNA double helix is possible. This sequence is found not only in TATA boxes, but also in other regulatory regions of DNA. The structure is not related to those found in triplexes or to parallel DNA duplexes, though its conformational parameters are very similar to those of B-form DNA. Bases of the two antiparallel strands form Hoogsteen pairs, with adenines in the syn conformation. 

Frangois et al. (1989) designed a 17-mer homopyrimidine oligomer to recognize, via Hoogsteen pairing, a 17-bp homopurine-homopyrimidine sequence. Inhibition of restriction in this sequence could be obtained in the micromolar range (Fig. 12.4 (III)). 

Both Watson-Crick and other forms of base pairing are found in parts of RNA molecules. One alternative form of base pairing is known as Hoogsteen pairing and other forms are generally known as non-Watson-Crick base pairs. 

---