DNA and protein sequencing methods, the first of which were developed decades ago, are crucial techniques for biologists. But, despite the essentiality of carbohydrates for cell structure and cell communication, the development of a comparable approach for carbohydrate sequencing has lagged far behind.

DNA is a chemically simple polymer, made up of just four nucleotides. In proteins, the 20 amino acid building blocks are strung together in linear chains by a single type of chemical bond. Carbohydrates, however, are another story. Monosaccharide building blocks (of which there are many) come in series of isomers that differ only in the arrangement of their atoms. Monosaccharides are linked together via glycosidic bonds, but such linkages can vary in both stereochemistry and regiochemistry. These variations are relevant to biological function and therefore are important to capture. But this chemical complexity—not to mention the branched structures of many carbohydrates—has made it challenging for researchers to develop a sequencing approach that can resolve carbohydrate structures without ambiguity.

“The development of all kinds of analytical strategies [for carbohydrate science] is slowed down in the absence of a robust, routine [sequencing] technology,” notes Isabelle Compagnon of the Université de Lyon in France. Currently, the only method for complete carbohydrate structural analysis is nuclear magnetic resonance (NMR) spectroscopy, which is slow and requires specialized expertise. Many groups have been working on developing more practical and high-throughput methods for carbohydrate sequencing, several based on mass spectrometry (MS), but none of the methods developed to date have been demonstrated to resolve all possible isomerisms.

Compagnon and her colleagues recently reported a promising new method for oligosaccharide sequence analysis that combines the strengths of MS with infrared (IR) spectroscopy. “On one hand, spectroscopy provides direct and refined structural detail at the atomic level but requires a relatively large amount of thoroughly purified sample,” says Compagnon. “On the other hand, MS readily applies to complex samples in small amounts with limited sample preparation, but has a poor structural resolution for carbohydrates since it does not intrinsically disambiguate all types of isomers.” By combining the two complementary technologies into a single instrument, the group developed a technique that reliably resolves carbohydrate isomerisms and may be developed into a robust sequencing approach.

Monosaccharide building blocks of carbohydrates can be distinguished by their IR signatures. Credit: Image reprinted with permission from Schindler et al. (Springer Nature).

Compagnon and her colleagues systematically analyzed monosaccharide and disaccharide standard compounds and found that these compounds can be distinguished by their unique, gas-phase IR spectral signatures. Starting from an oligosaccharide, tandem MS generates monosaccharide fragments in a sequential manner. By capturing IR spectra for each of these fragments in turn, an oligosaccharide sequence can be reconstructed by matching the fragment IR spectra to reference monosaccharide spectra. As a demonstration, the team applied the MS IR method to sequence a crude mixture of linear chito-oligosaccharides. Importantly, they found that the memory of the glycosidic bond stereochemistry was maintained throughout the MS manipulations. At the moment, it remains to be seen whether the technique could be further developed to resolve longer and more complex branching carbohydrate structures.

Compagnon's team is currently working with a prototype instrument, an ion-trap mass spectrometer modified with an IR laser to capture IR spectra immediately following molecular fragmentation. Using this instrument (currently available to external users at the OptoLYSE facility at Université de Lyon) requires an operator with both MS and laser technology expertise. Their goal for the next prototype is to automate spectroscopy data collection and analysis; such an instrument could be readily operated by a typical MS user.