An end to 40 years of mistakes in DNA–protein association kinetics?

Abstract

Proteins that bind to specific sequences in long DNA molecules have to locate their target sites amid myriad alternative sequences, yet they do so at remarkably rapid rates, sometimes approaching 10(10) M(-1) x s(-1). Hence, it has been asserted widely that binding to specific DNA sites can surpass the maximal rate for 3D (three-dimensional) diffusion through solution and that this could only be accounted for by a reduction in the dimensionality of the search for the target in effect by 1D (one-dimensional) diffusion (or 'sliding') along the DNA contour. It will be shown here that there is, in fact, no known example of a protein binding to a specific DNA site at a rate above the diffusion limit, and that the rapidity of these reactions is due primarily to electrostatic interactions between oppositely charged molecules. It will also be shown that, contrary to popular belief, reduced dimensionality does not, in general, increase the rate of target-site location but instead reduces it. Finally, it will be demonstrated that proteins locate their target sites primarily by multiple dissociation/reassociation events to other (nearby or distant) sites within the same DNA molecule, and that 1D diffusion is limited to local searches covering approximately 50 bp around each landing site.

Stochastic models of intracellular transport

“…One alternative hypothesis is that the observed fast association rates are due to electrostatic interactions between oppositely charged molecules, and thus do not violate the 3D diffusion limit (Halford, 2009). This is motivated by the theoretical result that the maximal association rate in Smoluchowski theory when there are long-range interactions between the reacting molecules is 98) where U (r) is the interaction potential.…”

Protein–DNA Electrostatics

“…Indeed, if both 1D and 3D diffusion processes can be observed, the conclusion that facilitated diffusion may greatly enhance DNA-protein association rates is more questionable. The point raised by Halford is that the rapidity of these reactions is instead due primarily to electrostatic interactions between oppositely charged molecules [5]. The large association rates reported in the pioneering work [1] where indeed obtained at very low ionic strength, suggesting a role of the electrostatic attraction that becomes negligible, due to screening effects, in higher salt.…”

“…Among them, one-dimensional sliding has been soon considered as necessary by most authors. The relative weight of sliding with respect to three-dimensional diffusion has then be subject to debate [5]. It is obvious that pure sliding would not be very effective if the starting position of the protein on DNA is far from the target sequence, since the protein will then spend too much time in searching remote regions unnecessarily.…”

“…Nonetheless, the actual role of this searching mechanism is still under discussion. An important element in this discussion has been the S. E. Halford's paper [5], where the author contest the need of any mechanism to facilitate the search and affirms that "no known example of a protein binding to a specific DNA site at a rate above the diffusion limit" exist. Indeed, if both 1D and 3D diffusion processes can be observed, the conclusion that facilitated diffusion may greatly enhance DNA-protein association rates is more questionable.…”

“…electrostatics had been explicitly invoked in the original works [1,3,5], most of the work has been based on a purely kinetic approach. The main results can be summarized by the finding that 3D excursions should be alternated by phases of 1D diffusion, named sliding, during which the protein binds DNA and slides along the double helix by thermal 1D diffusion sliding, during which the protein binds DNA and slides along the double helix by thermal 1D diffusion [6].…”