Moving Raman to the UV Range

DO THINGS a little differently from the way they're usually done, and the results might turn out surprisingly well. By modifying familiar recipes, chefs sometimes produce masterpieces. By altering their daily routes, commuters may shave minutes off their travel times. Even spectroscopists stand to gain by trying something different.

A case in point is Raman spectroscopy, an analytical technique in which samples are most often probed with light in the visible and near-infrared regions (about 400 to 700 nm). In the past several years, UV-Raman aficionados have pushed to capitalize on the benefits of carrying out a variant of the technique using ultraviolet light (typically 200 to 300 nm). Exciting samples deep in the UV region can enhance signal intensities by a factor of 1 million or more compared with conventional Raman spectroscopy. UV light also offers a method for probing certain functional groups selectively. In addition, moving to shorter excitation wavelengths can avoid interference from fluorescence, which sometimes obliterates standard Raman spectra.

Attendees of Pittcon earlier this month in Chicago learned about some of the most recent advances in the UV-Raman field. Practitioners and instrument makers gathered at a symposium to report on applications in biochemistry and catalysis and to discuss commercialization of UV-Raman instrumentation.

"It can be argued that the single most important issue in biology today is understanding how proteins fold into their native states." That assertion was made by University of Pittsburgh chemistry professor Sanford A. Asher, a pioneer of UV-Raman methodology. Asher explained that conformational variations (misfoldings) that cause proteins to adopt structures that differ from normal protein geometries have been fingered as key factors in Alzheimer's, Huntington's, and several other diseases.

"If we understood the mechanism through which proteins fold or misfold, we would have an opportunity to intervene chemically," he said. So Asher's group has been probing protein-folding processes with UV-Raman spectroscopy, a method that, he argued, is particularly well-suited to studying dilute solutions.

The nuances of a protein's folding and coiling geometry, which collectively are known as a protein's secondary structure, are dictated in large part by hydrogen-bonding interactions along the molecule's peptide backbone. As Asher pointed out, it's known from X-ray crystallography studies and other types of analyses that proteins often adopt a combination of common secondary structural motifs, including the so-called random coil, which is reminiscent of cooked spaghetti, and the more orderly α-helix and β-sheet arrangements.

The particular conformation adopted by a peptide chain is determined by rotations about C-C and C-N amide bonds and the resulting dihedral angles, known as the Ramachandran Ψ and Φ angles, respectively. These angles are named for the Indian biophysicist G. N. Ramachandran, who began studying them in the 1960s.

To get a spectroscopic handle on the secondary structure, Asher's group analyzed UV-Raman data from a number of proteins and peptides with known secondary structures and determined how amide vibrational bands correlate with the various geometries. Then they calculated basis spectra, which represent the Raman signals that would be measured from proteins that adopt only a single structural motif-for example, pure α-helix. Next, they showed that measured spectra could be modeled by the sum of appropriately weighted basis spectra, thereby providing a way to determine secondary structures of previously uncharacterized peptides and proteins.

THOSE RESULTS set the stage for a number of follow-up studies and key findings. For example, the Pittsburgh group observed that as polyglutamic acid is heated, it undergoes a structural transition (melts) from the α-helix conformation to a random coil. As the conformation evolves, a large change occurs in the Ψ angle, which in turn causes a large change in the frequency of one of the amide vibrational bands known as the amide III₃ band. In contrast, the researchers found that the band depends only weakly on the Φ angle.

Following other UV-Raman studies, Alexander V. Mikhonin, a graduate student in Asher's group, developed a method for determining Ψ angles from spectra measured under various experimental conditions. That development is noteworthy, Asher argued, because the Ψ angle serves as a textbook-type reaction coordinate for evolution of protein and peptide secondary structures. As such, the researchers uncovered a way to experimentally monitor secondary structures as they evolve.

The team then used that methodology to interrogate α-helix melting in a number of samples, including a 21-residue peptide known as AP, which consists mainly of alanine units. Some of the main findings of that investigation are that AP melts at a significantly higher temperature than previously reported and that other types of conformations melt before the α-helices unfold. Furthermore, in the melted state, AP's structure closely resembles that of polyproline II, which consists of left-handed helices with three residues per helical turn. (J. Am. Chem. Soc. 2006, 128, 13789).

MOVING A STEP closer to a detailed understanding of secondary structure, the Pittsburgh group recently showed that the distribution of Ψ angles can be used to calculate energy landscapes associated with peptide conformations. The plots dictate which structures are energetically favored and which are blocked by energy barriers.

Switching gears, Northwestern University chemistry professor Peter C. Stair lauded the ability of UV-Raman spectroscopy to probe reactions occurring at the surfaces of heterogeneous catalysts. "There is virtually no other technique that can see the catalyst and the molecules [reactants and products] in a single measurement and under reaction conditions," Stair declared. He added that before the development of UV-Raman methods, researchers' efforts were often thwarted by fluorescence signals that buried the Raman data.

Drawing on results of an investigation of butane dehydrogenation on supported vanadia catalysts, Stair demonstrated that UV-Raman analysis sheds light on several experimental parameters. For example, the results show that at a catalyst loading (concentration on the support material) of roughly one vanadium atom per square nanometer, which corresponds to an active form of the catalyst, vanadia consists of roughly equal proportions of monomeric and polymeric forms of V=O. At higher loadings, the polymeric form prevails and the catalyst is inactive. The method also indicates that while the catalyst is active, it's coated with a carbonaceous (coke) layer that resembles polystyrene. In contrast, layers composed of coronene, pyrene, and other polycyclic aromatic hydrocarbons with 2-D sheetlike structures poison the catalyst.

Despite the strengths of UV-Raman spectroscopy, the method is not practiced in many labs today. William F. Hug, president of Photon Systems, a Covina, Calif., maker of low-cost UV lasers and other equipment, remarked that widely used analytical instruments such as UV-visible spectrometers are easy to operate, are rugged, require little or no servicing, and are relatively inexpensive (in the $15,000 to $80,000 range). UV-Raman instruments, in sharp contrast, are used strictly by specialists, and systems with high-end deep-UV lasers can cost well over a half-million dollars, he said.

The good news, according to Hug, is that the past several years have seen "tremendous progress" in lowering the costs of UV lasers, detectors, lenses, mirrors, and other components of UV-Raman systems. Hug's presentation touched off an impromptu discussion of analytical-method limitations that culminated with symposium organizer Michael W. Blades of the University of British Columbia declaring that "nothing is really holding back UV-Raman anymore." He added, "We should expect more of these sessions and better attendance in the future."

[Top of Page]