The Agency of Industrial Science and Technology (AIST, President: Hiroyuki Yoshikawa) has developed, for the first time in the world, three major research tools for "sugar chains," which are key biomolecules for research of cancer, immunity, infections and regeneration medicine. These three tools are sugar chain genes, a sugar chain synthesis robot, and a rapid microanalysis system of sugar chains. They will allow the rapid advance of research on various life phenomena which were not able to be explained solely by nucleic acids and proteins.

In contrast with nucleic acids and proteins, which have linear structures, sugar chains have complex structures based on branching or stereoisomeric structures. For these reasons, a period from six months to one year is required to produce only one type of sugar chain by organic synthesis. AIST has succeeded in syntheses of more than 160 types of sugar chain synthetic enzymes using various types of sugar chain genes, and a method called the Enzyme-Cue Synthesis Method has been developed using these enzymes. This method allows simultaneous production of several types of sugar chain libraries (mixtures) in a short period of time. A robot for automatic synthesis of the sugar chain library was developed for the first time utilizing this synthesis method. Therefore, the robot has succeeded in synthesis of a library composed of several dozen types of sugar chains in two days. In addition, this method has a low environmental impact, as no harmful solvent has been used.

Furthermore, the AIST, together with Shimadzu Corporation (abbreviated as Shimadzu; President: Shigehiko Hattori) and Mitsui Knowledge Industry Co., Ltd. (abbreviated as MKI; President: Jun Masuda) have developed a rapid microanalysis system of sugar chains based on a mass spectroscopic database. Analysis of sugar chains has been difficult due to their complex structures, even combining several methods to analyze the sequence. In AIST, using the AXIMA-QIT mass spectrometer developed by Koichi Tanaka et al. of Shimadzu, to investigate the multistage tandem mass spectroscopic spectra of sugar chains, it was found that each sugar chain has a specific spectrum pattern. To apply these spectrum patterns to structural analysis, they were incorporated into a database for diverse sugar chains based on the principles of fingerprint recognition. MKI developed an information processing system that allows the matching of sugar chains with similar spectrum patterns in this database. Shimadzu developed corresponding software that links the information processing system and the mass spectrometer through the Internet. Use of this rapid microanalysis system for sugar chains allows the analysis of complex sugar chain structures from samples of only one nano-gram in several minutes.

Developments of sequencers and synthesizers for nucleic acid and proteins have dramatically advanced, but these types of equipment are not available for sugar chains, the third chain molecules present in the living organisms. It has been long desired to develop such equipment. Thus the combination of the genetics with equipment for synthesis and structural analysis is expected to accelerate research on sugar chains. This research is the key for cancer, immunity, infections, and regeneration medicine, and it is expected to further develop tools for rapid diagnostic method of diseases and tools for identification of target molecules for new drugs. The present research has been supported by the New Energy and Industrial Technology Development Organization (NEDO) through two projects, "Glycogene Project" and "Structural Glycoproteomics Project." These results were presented at the Annual Meeting of the Japanese Biochemical Society held in Kobe on October 22.

Japan is the undisputed world leader regarding research on sugar chains, staying ahead in structural analysis, synthesis, and functional research. On the other hand, with the completion of the decoding of the human genome, we are currently entering the "post-genome" era, and exhaustive analyses of proteins called proteomics have been taken up as a crucial issue worldwide. However, it has been established that more than half of the proteins are bonded to sugar chains, which control the functions of these proteins. Large-scale projects related to sugar chains have already started in the USA and EU as post-genome research following proteomics. Unlike nucleic acids or proteins, sugar chains have tree-like structures that cannot be expressed easily, therefore sequencers and automatic synthesizers cannot be used. As the amplification method available for nucleic acid does not applied to sugar chains, there are many unsolved issues.

AIST collaborated with Osaka University, Aichi Medical University, Nagoya University, and Soka University to conduct research on the "Glycogene Project" (Project Leader: Hisashi Narimatsu) entrusted by NEDO from fiscal year 2001 to 2003. As a result, it became possible to open roads for synthesis of diverse sugar chains by biological methods, not by organic synthesis. From fiscal year 2003, also under trust from NEDO, the project "Structural Glycoproteomics Project" to pursue the elucidation of the functions of sugar chains, is being implemented. The present results constitute a part of this project.

Structural analyses of sugar chains has been performed by highly trained specialists who can combine diverse analysis methods. Our AIST group has aimed at developing a quick analysis system using only mass spectrometry, which allows easy operation with minute amount of sample. This method is based on the principle in which, after constructing a database of the multistage tandem mass spectrometry patterns of a large number of sugar chain standards, the same pattern as that of the target sugar chain is searched for in the database. At the beginning of the fragmentation stage, several fragmentary ions are formed from a sugar chain. As only one portion of these ions shows a fragmented pattern with structural characteristics, to carry out a rapid analysis it is necessary to select a suitable fragmented ion after the 2nd stage fragmentation. To this end, we have developed a system which enables automatic selection of fragment ions that will give the information required for the structural analysis called the Intelligent Measurement Method. The rapid analysis system of a minute amount of samples here developed makes it possible to understand the structures of sugar chains including branched structure or stereo-isomeric structure by measuring the tandem spectrometry twice.

A large number of sugar chains with a clearly identified structure is required to establish the database. In order to obtain many sugar chain standards, our AIST group produced more than 160 types of sugar chain synthetic enzymes for synthesis of the standards from the sugar chain genes obtained through the project "Glycogene Project". Furthermore, to produce diverse sugar chains simultaneously in a short time, a library synthesis method was developed, which was named the Enzyme-Cue Synthesis. According to this method, each reaction is stopped midway before starting the following reaction. Therefore, when the reaction is repeated n times, it is possible to obtain theoretically 2 to the n power number of mixtures. If the enzymes for synthesis are selected to ensure that each component in a sugar chain library has a different molecular weight, a one to one correspondence between molecular weight and sugar chain structure can be established. Thus measurements of those molecular weights would be enough to identify the structure of the sugar chain included in the library. Division by molecular weight will lead to only one type of sugar chain.