Directly measuring single-molecule heterogeneity using force spectroscopy

Significance The relationship between structure and function is the heart of modern cell biology. Technological innovations in manipulating single molecules of proteins, DNA/RNA, and their complexes, are beginning to reveal the surprising intricacies of this relationship. In certain cases, the same molecule randomly switches between various long-lived structures, each with different functional properties. We present a theory to extract the extent and dynamics of these structural fluctuations from single-molecule experimental data. We find large heterogeneity in DNA and RNA complexes, supporting the notion that energy landscapes involving nucleic acids are rugged. Our work shows that functional heterogeneity is far more common than previously thought and suggests experimental approaches for estimating the timescales of these fluctuations with unprecedented accuracy. One of the most intriguing results of single-molecule experiments on proteins and nucleic acids is the discovery of functional heterogeneity: the observation that complex cellular machines exhibit multiple, biologically active conformations. The structural differences between these conformations may be subtle, but each distinct state can be remarkably long-lived, with interconversions between states occurring only at macroscopic timescales, fractions of a second or longer. Although we now have proof of functional heterogeneity in a handful of systems—enzymes, motors, adhesion complexes—identifying and measuring it remains a formidable challenge. Here, we show that evidence of this phenomenon is more widespread than previously known, encoded in data collected from some of the most well-established single-molecule techniques: atomic force microscopy or optical tweezer pulling experiments. We present a theoretical procedure for analyzing distributions of rupture/unfolding forces recorded at different pulling speeds. This results in a single parameter, quantifying the degree of heterogeneity, and also leads to bounds on the equilibration and conformational interconversion timescales. Surveying 10 published datasets, we find heterogeneity in 5 of them, all with interconversion rates slower than 10 s−1. Moreover, we identify two systems where additional data at realizable pulling velocities is likely to find a theoretically predicted, but so far unobserved crossover regime between heterogeneous and nonheterogeneous behavior. The significance of this regime is that it will allow far more precise estimates of the slow conformational switching times, one of the least understood aspects of functional heterogeneity.