As a graduate student, my goal was to develop methods to study protein dynamics on the picosecond to nanosecond time scale. To achieve this goal, I did time-resolved spectroscopy measurements on two model systems of the heme proteins cytochrome c and myoglobin. There are 2 things that were novel about our approach to these measurements:
- Measure down to the timescale of 100s of femtoseconds using femtosecond laser pulses
- Used site-directed mutagenesis to achieve a probe resolution down to a single amino acid
Below is a bit of background about our experimental techniques as well as some examples of the projects that I did as at Ohio State.
We used site-directed mutagenesis to move our spectroscopic probe tryptophan at different postions throughout each protein. As some samples had intrinsic tryptophans, multiple mutations needed to be performed to ensure that we had only one probe at a time in our proteins.
A bit of background information on the experimental technique we used including schematic of our experimental setups for both femtosecond fluorescence upconversion and for transient absorption.
We used ultrafast transient absorption to study the proteinquake dynamics in cytochrome c. We coupled site-directed mutagensis with femtosecond transient absorption in an attempt to study global proteinquake dynamics in cytochrome c. Additionally, we also studied the the unfolding of cytochrome c with time-resolved fluorescence.
We used femtosecond fluorescence upconversion to study Resonance Energy Transfer in myoglobin. One studied looked at the local conformation changes in the protein. The other studied looked at the ultrafast dynamics of nonequilibrium Resonance Energy Transfer and the globular flexibility of myoglobin.
Femtosecond Fluorescence Upconversion
A schematic of our experimental setups for both femtosecond fluorescence upconversion and for transient absorption are shown in the figure below. To understand the method of our data acquisition, you need to first understand a little bit about upconversion.
You can really think of the time resolved data that we acquired as a "movie." More specifically, we were taking a bunch of snapshots at different points in time then using those "pictures" to reconstruct a "movie."
To better understand the reconstructed movie analogy, let's say the dynamics that we are trying to capture is a ball falling. To catch this motion, all you need to do is record with a traditional video camera and you'll get a pretty good idea of how the ball fell. Another way you could capture the dynamics of the falling ball is to repeat the experiment many times and take one picture for each experiment at different times, then combine all of the snapshots to recreate a movie.
The reconstructed "movie" approach is essentially what we are doing with fluorescence upconversion. In the case of fluorescence upconversion, the "picture" happens to be the emission coming from the fluorescence probe in the protein that we are monitoring over time. The reason that we need to use the reconstructed "movie" approach is because the dynamics that we are trying to capture are faster than the electronics in a traditional detector.
Ultrafast Studies of Proteinquake Dynamics in Cytochrome c
We systematically studied heme dynamics and induced protein conformational relaxations in two redox states of cytochrome c upon femtosecond excitation. With a wide range of probing wavelengths from the visible to UV and a site-directed mutation we unambiguously determined that protein dynamics in the two states are drastically different. Notably, we observed impulsive bond breaking in the ferrous state and late re-binding generate proteinquakes that strongly perturb the local heme site and a previously unreported shaking of global protein conformation.
Ultrafast Studies of Global Proteinquake Dynamics in Cytochrome c
Again following from our previous work of cytochrome c, we systematically mapped heme dynamics and induced protein conformational relaxations in cytochrome c upon femtosecond excitation. We made tryptophan mutations one-at-a-time at 13 points throughout the protein and observed significant proteinquakes in several of the mutations. We further studied the effects of partial thermal and chemical denaturing on each cytochrome c mutant.
Unfolding of Cytochrome c Observed through Time-Resolved Fluorescence
We employed femtosecond fluorescence upconversion to characterize the unfolding curve of the two redox states of cytochrome c through changes of tryptophan fluorescence due to with varying resonance energy transfer rates. Tryptophan fluorescence rates varied as a function of chemical denaturant. Different turning points for each redox state showed the highly sensitive nature of resonance energy transfer pairs as a function of cytochrome c unfolding. This method demonstrates the ability of energy transfer pairs as a probe for further studies of protein folding and unfolding.
Ultrafast Studies of Resonance Energy Transfer: Local Conformation Studies of Myoglobin
We examined resonance energy transfer of the protein myoglobin using an intrinsic tryptophan and prosthetic heme utilizing the femtosecond fluorescence up-conversion method. With site-directed mutagenesis, we placed one-at-a-time a single tryptophan donor at 4 points in the A-helix of myoglobin. We also used molecular dynamics simulations to infer structure and dipole orientation fluctuations for intrinsic tryptophans. Both methodologies were used to characterize the local dynamic nature of myoglobin in solution compared to the static crystal structure.
Ultrafast Dynamics of Nonequilibrium Resonance Energy Transfer and Probing Globular Protein Flexibility of Myoglobin
We extensively characterized global flexibility of myoglobin using resonance energy transfer as a molecular ruler. With site-directed mutagenesis, we use a tryptophan scan to examine local structural fluctuations from B to H helices utilizing 10 tryptophan-heme energy transfer pairs with femtosecond resolution. We observed ultrafast resonance energy transfer dynamics by following a nearly single exponential behavior in 10-100 ps, strongly indicating that the globular structure of myoglobin is relatively rigid, with no observable static or slow dynamic conformational heterogeneity. The observation is against our molecular dynamics simulations, which show large local fluctuations and give multiple exponential energy transfer behaviors, suggesting too flexible of the global structure and thus raising a serious issue of the force fields used in simulations. Finally, these ultrafast energy transfer dynamics all occur on the similar time scales of local environmental relaxations (solvation), leading to nonexponential processes caused by energy relaxations, not structural fluctuations. Our analyses of such processes reveal an intrinsic compressed- and/or stretched-exponential behaviors and elucidate the nature of inherent nonequilibrium of ultrafast resonance energy transfer in proteins. This new concept of compressed nonequilibrium transfer dynamics should be applied to all protein studies by time-resolved Forster resonance energy transfer (FRET).
- C. Saxena, A. Sancar, and D. Zhong. Femtosecond Dynamics of DNA Photolyase: Energy Transfer of Antenna Initiation and Electron Transfer of Cofactor Reduction. The Journal of Physical Chemistry B, 108(46):18026-18033, November 2004.
- L. Zhang, Y.-T. Kao, W. Qiu, L. Wang, and D. Zhong. Femtosecond Studies of Tryptophan Fluorescence Dynamics in Proteins: Local Solvation and Electronic Quenching. The Journal of Physical Chemistry B, 110(37):18097-18103, September 2006.