Probing Proteins and Nucleic Acids

Proteins, like shoelaces, fold and unfold in a variety of ways. By learning how and why a protein occasionally folds incorrectly, researchers may be able to better treat victims of Alzheimer’s, Mad Cow, and other neurodegenerative diseases.

The ability to gain such an understanding of protein folding and misfolding requires measurements of fluctuating distance distributions occurring over nanoscale distances (0.1–100 nm) and many timescales (picoseconds to minutes). Folded, unfolded, and partially folded species may be simultaneously present and rapidly interconverting, obscuring the properties of individual species. As a result, we needed an experimental method to unravel distance distributions and fast conformational fluctuations.

The efficiencey of FRET based on whether the polymer is rigid, wormlike, or Gaussian.
Figure 1. FRET, or fluorescence resonance energy transfer, reveals distance distributions related to a polymer’s flexibility. The more flexible the polymer is, the higher the distance distributions of energy transfer efficiency, as illustrated in the plot based on computational predictions.

Using a technique called fluorescence resonance energy transfer, or FRET, we are able to measure the distances between two specific points on a protein. We use highly sensitive microscopes, which allow us to detect FRET signals from individual molecules with a time resolution of approximately 100 μs. The time resolution is limited by the strength of single-molecule signals.

Previously, we measured FRET for single proteins, distinguishing folded and unfolded proteins one at a time. We have now extended single-molecule methods to measure distance distributions that fluctuate on timescales longer than 1 ns.

Relevance to PLS Research Themes

In this research, we are using methods from the physical sciences that have been adapted and then applied to areas of high biological importance, such as protein folding. Our goal is to develop a quantitative description of heteropolymers, such as DNA (deoxyribonucleic acid) and proteins with strong charge and/or specific intrachain interactions.

Our work will eventually lead to a deeper understanding of disease mechanisms and contribute to national security through improved human health. Single-molecule methods related to this work may, in the future, provide biowarfare agent detection with the ultimate sensitivity—detecting pathogens one molecule at a time.

Major Accomplishments in 2005

Combining the strengths of singlemolecule as well as ensemble-level approaches, we introduced a new technique that interlaced picosecond pulses from two synchronized pulsed lasers to perform alternating laser excitation (ALEX) experiments on single molecules in conjunction with timeresolved FRET measurements. We sorted the molecules into subpopulations (such as folded and unfolded proteins) based on single-molecule signals detected within 0.1–1 ms. Within these subpopulations, we studied distance distributions fluctuating faster than 100 μs using fluorescence lifetime analysis. Even when folded proteins were present, we successfully studied the nanosecond timescale fluctuations of only unfolded proteins, made possible by excluding signals from the folded proteins.

Distance distributions of energy transfer efficiency depend on the stiffness of the polymer, which depends on the contour length and the persistence length (Figure 1). The contour length is the length of the polymer when it is fully extended or stretched out. The persistence length is the length over which the polymer naturally stays straight. If the contour length is much less than the persistence length, then the polymer is a rigid rod (black line). If the contour length is much greater than the persistence length, then the polymer is a very flexible Gaussian chain (green line) with maximal distribution in distances, denoted as delta FRET efficiency. In between, the polymer is a wormlike chain (red line).

In our studies of unfolded protein behavior in the presence of folded protein, we looked at double-stranded and single-stranded DNA as well as the unfolded states of two well-studied proteins—CI2 (chymotrypsin inhibitor 2) and ACBP (acyl-CoA binding protein). As illustrated in Figure 2, all of the DNA measurements of distance distributions lie between the rigid rod limit (black line) and the Gaussian chain limit (green line), exhibiting varying effects in a generally unstructured complex.

FRET efficiency in measuring DNA FRET efficiency in measuring C12 and ACBP protein measurements.
Figure 2. Using a nanosecond alternating laser excitation technique, we can measure both single- and double-stranded fluorescent-labeled DNA. All DNA data points lie between the rigid rod limit (black line) and the Gaussian chain limit (green line). Single-stranded DNA has a distance distribution closer to the Gaussian chain limit. Data are shown as symbols, and simulations are shown as lines. Figure 3. All of the CI2 and ACBP protein measurements are at or above the Gaussian chain limit (green line). For each data set, lowering the denaturant concentration results in measurements moving toward the upper-right, indicating more compact states with increasing fluctuations.

However, the unfolded CI2 and ACBP proteins exhibit fluctuations above the Gaussian chain limit (Figure 3). Nanosecond ALEX-based single-molecule sorting allows the exclusion of signal from folded CI2 and ACBP. These extra large fluctuations become more pronounced at lower denaturant concentrations (higher mean E), as the solution conditions become closer to natural conditions. Our observations can be explained by transient formation of structures, caused by protein denaturation, which constantly changes the effective contour length and persistence length. Such measurements of unfolded proteins in native-like conditions will improve our understanding of the starting point of the protein-folding process. Questions regarding protein misfolding will be more easily addressed in the context of these experiments.

Scientific Impact

Our understanding of biopolymers in general, and polyelectrolytes and protein folding in particular, will benefit from our ability to sort molecules into subpopulations and probe their distance distributions. We can gain new insights into fast conformational dynamics involved in protein machines, cellsignaling, and other biological processes. The quantitative information obtained from these new biophysical methods promises to revolutionize our basic understanding of life at the molecular level. For example, we can gain a better understanding of how pathogens highjack cellular networks, providing targets for drug development, as well as ways to detect emerging natural and engineered threats.

Related Publications

Laurence, T. A., et al., “Probing Structural Heterogeneities and Conformational Fluctuations of Biopolymers,” Proc. Natl. Acad. Sci. 102(48), 17348–17353 (2005).

Kapanidis, A. N., T. A. Laurence, et al., “Alternating-Laser Excitation of Single Biomolecules,” Acc. Chem. Res. 38(7), 523–533 (2005).

Contact: Ted Laurence [bio], 925-422-1788,

This content requires the Adobe Flash Player Get Flash

Ted Laurence

New Frontiers

The next step in gaining a better understanding of how proteins fold is to extend our techniques so that we can compare how proteins fold into their native and functional structures in vitro and within living cells. We are now testing our hypothesis that the large delta FRET efficiency is due to transient residual structure. By measuring the dependence of delta FRET efficiency on the position along the protein chain and on side chain truncations (removal of sections of the protein that form structure), we hope to elucidate how protein denaturation affects the transient formation of protein structures.