The degenerative joint disease osteoarthritis affects more than 20 million Americans, attacking and ultimately destroying cartilage. Chemistry professor Michael D. Morris of the University of Michigan, Ann Arbor, noted that cartilage can repair itself but that the process is slow and limited. The ability to detect the disease at an early stage therefore might enable damage to be minimized, thereby avoiding pain and treatment such as knee and other joint replacements. Unfortunately, current technology stands in the way. “At present, there are no reliable tests for early-stage arthritis,” he said. Morris, postdoctoral researcher Gurjit Mandair, graduate student Karen Dehring, and rheumatologist Blake Roessler, a professor of internal medicine at the University of Michigan Medical School, may have found a solution. They’ve used surface-enhanced Raman spectroscopy (SERS) to detect small but significant changes in synovial fluid, the viscous fluid in cartilage. Working with microliters of synovial fluid, they measured a decrease in the hyaluronic acid content of the fluid, which eventually may prove to be a reliable indicator of early-stage arthritis. They performed these measurements by detecting the Raman scattering of gluconate and glucosamine, which make up the polymer in hyaluronic acid. Raman spectroscopy ferrets out molecular vibrational information that acts as a chemical fingerprint because vibrational characteristics are very specific to a given chemical bond. However, Raman scattering, upon which the spectroscopy depends, typically is very weak and therefore is difficult to detect, particularly in small samples. Raman spectra of a 0.5-mg/ml aqueous hyaluronic acid solution deposited onto a gold-coated surface-enhanced Raman spectroscopy substrate (a) and bare gold (b) illustrate how the choice of substrate can enhance the Raman signal. The measurement times were 60 s and 120 s, respectively. Courtesy of Michael D. Morris, University of Michigan. In SERS, Raman signals are magnified by a million to a trillion times compared with the signal from a bulk sample. SERS takes place only when molecules are adsorbed to a conductive surface that isn’t flat on a microscopic scale. The effect is the result of an increase in the local optical field that arises from the sharp points of textured metals such as gold, silver or copper, or from nanometer-size gaps between colloidal particles. When a laser beam of the right wavelength strikes the metal substrate, it generates surface plasmons, which assist in delivering light to the molecule and in getting out the resulting Raman signal. The key to SERS is the substrate, which has proved difficult to consistently fabricate. Traditional substrate manufacturing techniques suffer from large variations in surface roughness and, as a consequence, the degree of enhancement. Morris noted that a reproducible, commercially available substrate is needed for SERS to be clinically useful. The active area for this surface-enhanced Raman spectroscopy (SERS) substrate is the dark square in the golden rectangle. Samples are placed atop the substrate and then analyzed using Raman spectroscopy equipment. Image courtesy of Mesophotonics Ltd. “Clinical laboratories will not make their own SERS substrates,” he said. The researchers turned to Klarite, a relatively new substrate from Mesophotonics Ltd. of Southampton, UK. A spin-off of the University of Southampton, Mesophotonics specializes in the manufacture of photonic crystals, periodic arrays of microscopic holes laid out in dielectrics such as silicon. Unlike processes that rough up a metal surface or deposit gold colloidal particles across a substrate, the company’s proprietary approach uses standard semiconductor manufacturing techniques. These photolithography-based methods result in a regular array of micron-size holes spaced nanometers apart in a 4-mm-square section of silicon, which then is coated with a thin layer of gold. The entire chip is 6 x 10 mm in area and 0.5 mm in thickness. It can be mounted on a 75 x 25-mm glass slide or used as is. The mounted chip can be used in a standard Raman setup. The gold-coated, textured silicon surface created a more than millionfold enhancement of the Raman signal compared with a smooth surface at 633- or 785-nm excitation. Tests conducted by Mesophotonics have shown the enhanced Raman signal to have less than a 10 percent relative standard deviation at 5-mW excitation, with that figure encompassing the variation across and between chips. David Reece, Mesophotonics sales manager, said that, because the substrates are manufactured with semiconductor tools and techniques, they can be produced in volume at what he believes is an affordable price. The company plans to develop applications that will help to establish SERS as a routine analytical technique. To that end, it has created a tool kit for the development of trace-level molecular spectroscopy applications. Using the current Klarite substrate and Raman spectroscopy equipment from Kaiser Optical Systems Inc. of Ann Arbor, the Michigan researchers have shown that they can detect and quantify hyaluronic acid at clinically important levels. In a paper that will be published by the American Chemical Society, they report that the width of the marker bands is a better quantifier than the intensity of the signal. Morris attributed this finding to being close to surface saturation. “As concentration increases, the number of possible conformations of the polymer increases, and so the band broadens, he explained.” Although the current setup works, he noted that there are some areas that need improvement. One is the methodology used to deposit the hyaluronic acid. The result is a ring, which makes it difficult to examine the sample efficiently. The researchers are working to correct this. Another is the cost of the substrate, which Morris hopes will decrease with widespread use. Contact: Michael D. Morris, University of Michigan, Ann Arbor; e-mail: mdmorris@umich.edu. David Reece, Mesophotonics Ltd., Southampton, U.K.; e-mail: david.reece@mesophotonics.com.