Nucleic Acids in Chemistry and Biology

The conclusion is that for all drug–DNA interactions, intercalation and groove binding represent two
potential low-energy wells in a continuous free energy surface. The binding mode with lower energy will
depend on DNA sequence and conformation as well as the molecular features of the bound ligand.

9.9 Co-operativity in Ligand–DNA Interactions

Genomic DNA is somewhat different from model duplexes used in structural and biophysical studies,
since it possesses a continuous array of potential binding sites that are incorporated into a large nucleic
acid–protein complex. Thus the biological activity of a DNA binding drug will be affected by co-operativity.
Co-operativityis an important and pervasive concept in biology. The phenomenon of allostery is well
known in protein–ligand interactions. A type of negative co-operativity in drug–DNA interactions is neigh-
bour exclusion (Section 9.6.3). Here each bound drug excludes one or more adjacent base pairs from being
potential drug binding sites. This is often referred to as site-exclusion co-operativity. However, it is also pos-
sible for there to be co-operative effects between adjacent binding sites (composed of more than one base
pair) and in such cases the data can be described by a modified version of the McGhee and von Hippel rela-
tionship.^20 (Section 9.6.3 and Equation 9.13). The modified form of this equation is

(9.17)

where ris the ratio of bound drug to total DNA sites (in base pairs), Cfthe concentration of free drug, K
the binding constant for the interaction of a drug with an isolated site, and nthe neighbour exclusion
parameter. The parameter is the co-operativity function; which in effect is the equilibrium constant for
moving a bound drug molecule from a totally isolated binding site to a site contiguous to another bound
ligand. If is greater than 1, this indicates positive co-operativity. Conversely, if is less than 1, there is
negative co-operativity. When 1 there is no co-operativity and then Equation 9.17 can be simplified
to Equation 9.13. This method of analysis is very useful for interactions involving long polynucleotides or
natural DNA samples. Equation 9.17 can be incorporated into nonlinear least squares fitting routines to fit
equilibrium binding isotherms and the parameter can thus be evaluated readily. However, binding data
should primarily be fitted to the simplest model, therefore Equation 9.17 should only be used if it yields a
significantly better statistical fit to the data compared to Equation 9.13.

9.10 Small Molecule Interactions with Higher-Order DNA

The vast majority of genomic DNA exists as a double-stranded, antiparallel structure in the B-conformation.
However in the past fifteen years, it has become clear that DNA can adopt a number ofhigher-order
conformations. These involve the self-assembly of three or four DNA strands or the intramolecular fold-
ing of a single strand such that it adopts a triplexor quadruplex (tetraplex) conformation (Section 2.3.6 and
2.3.7). Further, it has become apparent that, far from being simple biophysical curiosities, these multi-
stranded DNA structures have considerable biological significance. This has led to triplex and quadruplex
DNA becoming major drug targets. Below are summarised the key features of these higher-order struc-
tures and a discussion of how small molecules have been used to target them and effect their stabilization.

9.10.1 Triplex DNA and its Interactions with Small Molecules

Triplex DNAis formed when a single-stranded oligonucleotide specifically recognizes the major groove
of duplex DNA and forms hydrogen bonds with the Watson–Crick base pairs (Section 2.3.6). Due to the

r

C

Knr

nr r R

nr

nrR

n

f

   

        
   

    



()

()()

()( )

()

1

211

211

11

1















 221

1141

2

2

()

{[ ( ) ] ( )}

   

  

nr

Rnrrnr















372 Chapter 9

http://www.ebook3000.com