Atomic force microscopy study of DNA flexibility on short length scales: smooth bending versus kinking.

Alexey K Mazur,Mounir Maaloum

doi:10.1093/nar/gku1192

Abstract

The apparently anomalous flexibility of DNA on short length scales has attracted a lot of attention in recent years. We use atomic force microscopy (AFM) in solution to directly study the DNA bending statistics for small lengths down to one helical turn. The accuracy of experimental estimates could be improved due to a large data volume and a refined algorithm for image processing and measuring bend angles. It is found that, at length scales beyond two helical turns (7 nm), DNA is well described by the harmonic worm-like chain (WLC) model with the bending persistence length of 56 nm. Below this threshold, the AFM data are also described by the WLC model assuming that the accuracy of measured bend angles is limited by the physical width of the double helix. We conclude that the double helical DNA behaves as a uniform elastic rod even at very short length scales. Strong bends due to kinks, melting bubbles and other deviations from the WLC model are statistically negligible.

Highlights

The ability of duplex DNA to loop, fold and wrap around proteins plays important roles in many biological processes
The atomic force microscopy (AFM) data are described by the wormlike chain (WLC) model assuming that the accuracy of measured bend angles is limited by the physical width of the double helix
It was always understood that these observations, regardless of experimental errors [36], could be rationalized without questioning the fundamental physics of the double helix by assuming a certain probability of strong bends due to melting bubbles, for instance [18,22]

Summary

Introduction

The ability of duplex DNA to loop, fold and wrap around proteins plays important roles in many biological processes. It was shown by calculations that strong bends in duplex DNA can be achieved without destacking, by subtle low-energy fluctuations of bond angles and torsions [6] This second picture corresponds to the elastic rod or wormlike chain (WLC) model earlier proposed in polymer physics [7,8,9]. For short and intermediate length scales, deviations from predictions of the WLC model were uncovered using different experimental methods [11,12,13,14,15,16] According to these data, short DNA looks much more flexible than the WLC prediction, which was explained by assuming inherently high ‘kinkability’ of the double helix [17,18,19,20,21,22,23,24]. An insightful review devoted to this problem was published recently [36]

Methods

Results

Conclusion