My research group focuses on the characterization and use of novel metal and semiconductor nanomaterials in the development of new ultrasensitive chemical methods of detection and identification. In the past decade, numerous techniques, such as laser-induced fluorescence (LIF), have demonstrated the capability of detecting single analyte molecules. However, these techniques do not often provide sufficiently detailed structural information necessary for chemical identification. For example, LIF measurements yield little structural information while also requiring a fluorescent label that suffers from rapid photobleaching. Recent research has developed two new methods of detection that can overcome some of these drawbacks: (1) surface-enhanced Raman scattering (SERS) on colloidal metal nanoparticles and (2) luminescent semiconductor nanocrystals (i.e., quantum dots).
This research is interdisciplinary by nature and therefore exposes students to various aspects of laser spectroscopy, microscopy, biochemistry, molecular biology, photophysics, and materials science. Research projects for students include:
I. Metal Nanoparticles for Ultrasensitive Surface-Enhanced Raman Scattering (SERS)
Background. Raman spectroscopy is capable of providing highly resolved vibrational information at room temperature and does not suffer from rapid photobleaching. However, Raman scattering is an extremely inefficient process with scattering cross sections (~10-30 cm2 per molecule) approximately 14 orders of magnitude smaller than the absorption cross sections (~10-16 cm2 per molecule) of fluorescent dye molecules. To achieve single-molecule sensitivity, the normal Raman scattering efficiency must be enhanced 1014 fold or more. Such enormous degrees of enhancement have been achieved using silver and gold nanoparticles. These particles are relatively large (>50 nm in diameter), faceted nanocrystals (Fig. 1) that are able to enhance the Raman scattering cross sections of adsorbed analyte molecules by as much as 1015 fold. This large enhancement allows both the detection and identification of single, nonfluorescent molecules. Both electromagnetic-field and chemical enhancement models are used to explain the SERS phenomenon. Currently, the overall molecular detection efficiency of these studies has been relatively low (<10% analyte molecules detected/analyte molecules in sample) because not all molecules are adsorbed, not all adsorbed molecules are surface-enhanced, and not all nanoparticles are SERS-active. Great strides towards producing efficient and reliable SERS-substrates can thus be made by improving nanoparticle synthesis, separation, and assembly methods. Our focus is on developing novel synthesis and assembly strategies to create highly SERS-active nanostructures to make ultrasensitive analytical measurements.
Template-Directed Assembly of SERS-Active Nanostructures. New strategies are required if highly SERS-active substrates are to be efficiently and reliably constructed. In our approach, tobacco mosaic virus (TMV) capsids (Fig. 2) are used as "scaffolds" to assemble Ag or Ag-cladded Au nanoparticles into discrete nanostructures for ultrasensitive SERS spectroscopy. TMV surface amino acid residues are chemically modified to serve as either electrostatic or chemical "anchors" to assemble closely packed SERS-active nanostructures. The interparticle sites created in these novel nanoassemblies are expected to exhibit the large intrinsic enhancement factors (~1015) observed in single-molecule (SM) SERS.
Improving SERS-active nanostructures by controlling nanometer-scale architecture will dramatically impact the analytical utility of SERS and facilitate SM-SERS studies. Information gained from these studies about the relationship between nanostructure morphology and SERS-activity will also aid in developing new insights into the fundamental mechanisms of SERS. Finally, the assembly methods developed in this study will broadly impact the field of nanotechnology by providing an additional tool to rationally construct novel nanostructures and functional nanomaterials.
Single-Molecule Enzymology. The goal of this project is to observe the stochastic events of single molecule enzymatic reactions and the transient reaction intermediates that are often not observable in conventional measurements using SERS. These SM-SERS enzymology studies will complement current SM-LIF efforts and will extend research to the study of nonfluorescent species. In addition, the information rich SM-SERS spectra will provide sufficient molecular information via a vibrational fingerprint that will aid in identifying molecular species such as reaction intermediates. We are currently investigating is the redox activity single cytochrome c proteins. A compact microscope-mounted flow cell is used to rapidly introduce redox agents to control the oxidation state of the cytochrome iron cation located in porphoryin (i.e., heme). Bulk studies have demonstrated that the oxidation state of the iron can be directly determined from SERS spectra. Our goal is to monitor the molecular dynamics of cytochrome c as it changes from Fe(II) to Fe(III) at the single-molecule level.
II. Semiconductor Quantum Dot Bioconjugates for Fluorescence Correlation Spectroscopy
Background. This project focuses on the development of a new class of quantum dot (QD) based fluorescence correlation spectroscopy based bioassays for the detection of molecular associations including DNA hybridization, protein-nucleic acid interactions, and antibody-antigen recognition. Accurate and sensitive determination of these associations is imperative for the early detection of cancer, infectious diseases, genetic abnormalities, and autoimmune disorders. Traditional bioassays such as enzyme-linked immunoassays (ELISA) or fluorescence in situ hybridization (FISH) suffer from non-specific binding and require multiple rinse steps. Homogenous (i.e., no rinsing) fluorophore-based FCS bioassays have been demonstrated, but broad emission properties of fluorophores and instrument alignment problems has slowed progress. A homogenous QD-based bioassay utilizing two-color fluorescence cross-correlation spectroscopy (TC-FCCS) detection will be developed to address these issues.
Fluorescence Correlation Spectroscopy (FCS). FCS describes a class of methods that monitors and extracts information from the time-dependent fluctuations in fluorescence intensity observed in extremely small probe volumes (~10-15 L). Analyte concentrations, photophysical properties, and diffusion coefficients are accurately determined using one-color fluorescence autocorrelation spectroscopy (OC-FACS). However, OC-FACS is not well suited for analyzing complex systems because it relies only on subtle changes in the diffusion coefficient for identification. TC-FCCS improves selectivity and sensitivity over OC-FACS by employing two excitation lasers to excite two spectrally separable fluorescent probes that are monitored by two separate detection channels. A positive cross-correlation indicates the simultaneous detection of both fluorescent labels, and hence the presence of moieties colabeled with both fluorophores. The amplitude of the cross-correlation function is directly proportional to analyte concentration. However, precise alignment of two excitation lasers is difficult and spectral cross talk between detection channels limits sensitivity.
Quantum Dots (QDs). QDs are inorganic nanometer-sized crystals (~0.1 to 10 nm) that are the subject of much research. The great interest is sparked by their size-tunable optical and electronic proper-ties. Figure 3 illustrates the size-tunable luminescent proper-ties of a series of cadmium selenide (CdSe) QDs. QDs exhibit nar-row emission profiles (full-width half-maximum 25-40 nm), are extremely photostable, show broad absorption spectra extending from the UV to the blue edge of the emission peak, emit
photons intermittently, and can be ~20 times brighter than conventional fluorophores. Recently, QDs have been used as luminescent labels for biological applications. The favorable photophysical properties of QDs and the ability to use a single excitation frequency to excite QDs with widely separated emission spectra make QDs excellent candidates for TC-FCCS bioassays. Finally, the narrow emission spectra and broad excitation spectra of QDs also offer the potential of performing multiple bioassays (i.e., multiple targets) on a single sample simultaneously. This multiplexing advantage will aid in rapid medical diagnostics and drug discovery.
Feasibility of TC-FCCS Bioassays. The initial goal is to demonstrate the feasibility of a QD-based TC-FCCS bioassay (Fig 4). As a model system the bind-ing of two-different colored QDs labeled with bioti-nylated bovine serum albumin (biotin-BSA) and streptavidin will be studied. Aqueous CdSe QDs will be prepared from organic stock solutions using established procedures (e.g., carbodiimide or iminothiolane). These conjugation methods attach biomolecules to QDs via chemisorption of thiol groups to the nanoparticle's surface. Alternative conjugation methods such as electrostatic adsorption will also be explored. Quantum yields, laser excitation saturation intensities, and diffusion coefficients of these QD-bioconjugates will be measured by OC-FACS. Once reliable QD-bioconjugates are prepared, two different, wavelength-separated QD-bioconjugates will be linked together via biotin/streptavidin recognition. TC-FCCS will be used to detect the linked two-color QD pair. Reaction kinetics, spectral cross talk, binding constants, and detection limits will be determined.
Development of Advanced TC-FCCS Bioassays. The biotin-BSA/streptavidin QD-bioconjugate system is an excellent platform to launch efforts to develop advanced QD-based TC-FCCS bioassays. For example, a TC-FCCS sandwich immunoassay for human insulin will be developed. Monoclonal antibodies that recognize nonoverlapping regions of the insulin peptide will be conjugated to QDs. Insulin concentration will be determined by the presence of colabeled insulin species measured by TC-FCCS. Another project we intend to pursue is a TC-FCCS DNA hybridization assay. Biotinylated hybridization probes will be attached to streptavidin labeled QDs via biotin/streptavidin binding. DNA probes will be designed to recognize either two separate genes or sequence regions on a single strand of DNA. Presence of the target genes or sequences will be determined by the detection of colabeled DNA species using TC-FCCS. Detection limits of these bioassays may ultimately extend into the femtomolar (10-15 M) range, which will meet or exceed detection limits of current methods and will be competitive with radiolabel bioassays without the hazards of radioactive labels.