DNA and Double-Stranded Oligonucleotides - Free-Radical-Induced DNA Damage and Its Repair - page 407

damping effect of the interspersed medium, and R AD the distance between donor
and acceptor.

e - β R DA

k ∝ (76)

value ranging between 1.0 and 1.5 Å −1
are observed (Beratan et al. 1997; Priyadarshy et al. 1996). This is, for example,
found for charge transfer through proteins (Nocek et al. 1996). Such high

When the charge transfer occurs high β

val-
ues have also been reported for DNA under certain conditions, e.g., (Lewis et al.
1999a, 1999b, 2000a; Brun and Harriman 1994; Meade and Kayyem 1995; Cai
and Sevilla 2000; Messer et al. 2000; Fukui and Tanaka 1997; Krider and Meade
1998; Meggers et al. 1998a, b).
Lower

β

0.1 were observed for the charge transfer from
excited 2-aminopurine to G through interlacing adenines (Wan et al. 2000) and
in DNA hairpins (Lewis et al. 1997; Lewis and Letsinger 1998; for a review see
Lewis et al. 2001).
However, very low

β values around 0.6 ±

values (0.2 Å −1 and even lower) have also been reported
(Murphy et al. 1993; Arkin et al. 1996; Kelley et al. 1997; Holmlin et al. 1997,
1998; Henderson et al. 1999; Ly et al. 1999; Kelley and Barton 1999; Kelley et al.
1999; Giese et al. 1999; Schiemann et al. 2000; Stemp et al. 2000; for studies on ET
through solid DNA see, e.g., Eley and Spivey 1962; Okahata et al. 1998; Fink and
Schönenberger 1999; Porath et al. 2001).
A small

β

0.1 Å −1 indicates that the ET rate only weakly depends on
the distance between donor and acceptor. It has been pointed out (Grozema et
al. 2000) that two different electron transport mechanism may give rise to such a
weak distance dependence, the “molecular wire” mechanism (Turro and Barton
1998; Mujica et al. 1999), and the “incoherent hopping” mechanism (Jortner et al.
1998; Bixon et al. 1999). In the “molecular wire” mechanism, donor and accep-
tor are strongly coupled to one another through the intervening bridge, and the
charge can travel through this

β value of ≈

-way. In the "incoherent hopping" mechanism
the charge travels in a multistep process "hopping" between GC pairs that act as
localization sites until it reaches the acceptor. The rate of this process does not
decay exponentially with the distance and hence

π

is not a suitable parameter.
Here, the logarithm of the rate of charge transfer is proportional to the number
N of hopping steps; Eq. (77).

β

ln k ∝ - η ln N (77)

The power parameter equals 2 for unbiased diffusive hopping from the donor

to the acceptor, and between 1 and 2 for a direction-biased random walk process
(Priyadarshy et al. 1996).
Quantum-mechanical calculations (Grozema et al. 1999, 2000; Berlin et
al. 2000) that take the molecular wire and quantum mechanical channels (cf.
Fig. 12.14) into account are capable of explaining the wide range of

η

values
reported. This theory also allows for the observed drastic dependence on the
rate of hole transfer on the number and nature of the nucleotide pairs AT vs GC
between the donor (G • + ) and the acceptor (GGG) as reported by Meggers et al.

β