Xiaoliang Sunney Xie

Department of Chemistry and Chemical Biology
Harvard University
Mallinckrodt B-026
12 Oxford Street, Cambridge, MA 02138

tel: (617) 496-9925; fax: (617) 496-8709
email: xie@chemistry.harvard.edu

Research Interests:

Professor Xie and his research group have made advances in room-temperature single-molecule spectroscopy in order to investigate molecular interactions and chemical reactions on a detailed level. Our current research has two objectives: (1) to understand conformational and chemical dynamics of biomolecules such as enzymes by single-molecule spectroscopic studies; (2) to image macromolecules and to probe their chemical activities on biological membranes and in living cells with near-field and nonlinear optical microscopy.

Single-Molecule Enzymatic Dynamics

Chemists are used to visualizing molecular properties and chemical changes on a single-molecule basis. However, information regarding molecular interactions and chemical dynamics has been derived exclusively from experiments conducted on large ensembles of molecules. Although ensemble-averaged results are essential, they often preclude detailed information for heterogeneous systems--biological systems in particular--because of the lack of a priori knowledge of the distributions and fluctuations of molecular properties. A single-molecule experiment directly measures the distribution of a molecular property and its dynamical fluctuation, revealing information hidden in the ensemble average. For example, we made real-time observation of enzymatic turnovers of single molecules of cholesterol oxidase, an enzyme that catalyzes the oxidation of cholesterol by oxygen. The active site of the enzyme involves a flavin adenine dinucleotide (FAD), which is naturally fluorescent in its oxidized form but not in its reduced form. The FAD is first reduced by a cholesterol molecule, and is then oxidized by molecular oxygen. We observed that the FAD emission blinked on and off, each on-off cycle corresponding to an enzymatic turnover. Statistical analyses of the data revealed a significant and slow fluctuation in the rate of cholesterol oxidation. We also independently observed slow conformational fluctuation of the enzyme through the fluctuating emission spectrum. The conformational changes caused the rate fluctuation. The rate fluctuation is not described by the fundamental Michaelis-Menten mechanism of enzymology, which works well for describing only the averaged behavior of turnover events. This illustrates the influence of conformational dynamics on enzyme functions, which the single-molecule approach can reveal. Our goal is to conduct similar single-molecule experiments under natural biological conditions. We have developed the following two new techniques for imaging biomolecules.

Near-field Optical Imaging of Photosynthetic Membranes

Near-field optical microscopy is capable of overcoming the diffraction limit, which limits the resolution of ordinary optical microscopes to about half of the wavelength of light. We have developed a new high-resolution near-field optical microscopy using a sharp metal tip (~10 nm end diameter) illuminated with a laser. The illuminating light induces a strongly enhanced field, which is composed of mainly non-propagating evanescent components and is confined to the tip end. This results in a highly localized excitation source for molecular fluorescence. Excitation by femtosecond pulses via two-photon absorption of the sample was used to provide a good image contrast. This method has resulted in a spatial resolution of 20 nm. Our work is aimed at spectroscopic imaging of photosynthetic membranes. Protein complexes, such as light-harvesting complexes and reaction centers, are closely packed in the membrane and convert solar energy to chemical energy. The distribution of the protein complexes in the membrane is not well understood. This information is needed for a complete understanding of the energy disposal mechanisms operating among the various complexes. Our knowledge of the spatial distribution is limited because it is based on electron microscopy, which destroys samples in preparation and can only provide information regarding size, not spectroscopic identity. The fluorescent protein complexes containing chlorophyll a and b molecules have distinct spectral properties, such as excitation spectra and fluorescence lifetimes, that can be used to identify the various complexes. Our new technique, which has a spatial resolution comparable to the size of the complexes, will allow us to resolve spectroscopically the individual protein complexes in the membrane.

Coherent Anti-Stokes Raman Scattering (CARS) Microscopy of Living Cells

Confocal and multiphoton fluorescence microscopy are powerful tools for three dimensional imaging of chemical and biological samples, especially for living cells. For chemical species or cellular components that either do not fluoresce or cannot tolerate fluorescence labeling, Raman microscopy provides a contrast mechanism based on molecular vibrations. However, the intrinsically weak Raman signal necessitates high laser powers and is often overwhelmed by the fluorescence background of the sample. To overcome the problems, we have demonstrated Coherent Anti-Stokes Raman Microscopy using solid-state femtosecond lasers operating in the near-infrared region (800nm-1200nm). The advantages of CARS microscopy are: (1) CARS microscopy provides contrast based on vibrational characteristics, which are intrinsic to the samples. It does not require natural and artificial fluorescent probes. (2) CARS microscopy is much more sensitive than the spontaneous Raman microscopy. Therefore, CARS microscopy requires only a moderate average power tolerable by most biological samples. This high sensitivity arises from the fact that CARS is a coherent radiation from many constructively interfering vibrational modes. (3) CARS microscopy has the three dimensional sectioning ability, which is particularly useful for imaging through tissues or cells structures. This results from the nonlinear CARS signal generated only at the focus where the excitation intensities peak. (4) There is little scattering of the near-infrared excitation beams, allowing a deep penetration (>0.3mm) through thick tissues or cells. (5) There is little absorption of the near-infrared excitation beams, reducing photodamages to biological samples. We have demonstrated CARS microscopy of living cells when tuning into the C-H stretching vibrational frequency around 3000cm-1. Applications to biological imaging are being explored.

Selected Publications:

A. Zumbusch, GR Holtom, XS Xie (1999). Three-Dimensional Vibrational Imaging by Coherent Anti-Stokes Raman Scattering. Phys. Rev. Letters 82, 4142-4145.

EJ Sanchez, L Novotny, XS Xie (1999). Near-Field Fluorescence Microscopy Based On Two-Photon Excitation With Metal Tips. Phys. Rev. Letters 82, 4014-4017.

HP Lu, L Xun, XS Xie (1998). Single-Molecule Enzymatic Dynamics. Science 282, 1877-1882.

Page created and maintained by Xaq Pitkow