Abstract

This paper reports the first ultrafast studies on the rates of DNA-mediated forward and reverse electron transfer between photoexcited [M(phen)₂dppz]²⁺ (M=Ru or Os, phen =1,10-phenanthroline, dppz = dipyrido[3,2:a-2′,3′:c]-phenazine) and various electron acceptors in order to ultimately determine the distance dependence of electron transfer kinetics with DNA as an environment.¹,² Previously Barton, Turro, and coworkers have presented evidence that electron transfer in DNA can occur rapidly over an extraordinarily large distance³ with a more shallow distance dependence than that for other media, such as liquids and proteins.⁴,⁵ Some theoretical work supports the shallow distance dependence which is attributed to a long range, thermally activated coherent mechanism involving virtual excitation of the DNA "bridge".⁶ Utilizing time-correlated single photon counting (TCSPC), we observe a substantial fraction of photoexcited [M(phen)₂dppz]²⁺ (M=Ru, Os) exhibits fast oxidative quenching (k_q > 3 x 10¹⁰ s⁻¹) in the presence of intercalating Rh(III) acceptors while the remaining excited-state species exhibit a range of quenching constants less than 10⁸ s^_1. Transient-absorption experiments on the picosecond timescale indicate that, for a series of donors bound to mixed sequence DNA, the majority of back electron transfer is also very fast (ca. 10¹⁰ s⁻¹). Importantly, the rate constant for the fast ground-state recovery is independent of loading of Rh(III) intercalators on DNA. As shown in Figure 1, regardless of whether the average loading of metal complexes is 1 in 33 basepairs or 1 in 10 basepairs, the fast bleach recovery exhibited by [Ru(phen)₂dppz]²⁺ is well fit under all conditions by an exponential decay of 9 x 10⁹s⁻¹. Separate ultrafast data suggests that, like the recombination reaction, the fast quenching *M²⁺ is simple first-order at early time. If the early time electron transfer kinetics were not simple first order, and the electron transfer rate decayed with an exponential distance dependence (i.e. ≅ a factor of 30 per base step [3.4 Å]), our TCSPC apparatus should be able to observe some evidence of the slower components with rate constants in the range of 10¹⁰ - 10⁸ s⁻¹. The absence of rates in this range is evidence that the electron transfer is simple first order at time < 10 ns. This result has implications with regards to the donor-acceptor separations on DNA. Throughout most of the titration, intercalators are dilute on the double helix and statistics show that the amount of fast quenching and ground-state recovery observed is too great to be accounted for by random loading of nearest-neighbor pairs. Thus, either the electron transfer reaction must involve clustering of the donor and acceptor on the helix or the DNA-mediated interaction must occur over long distance.

© 1996 Optical Society of America