研究課題
Proteins that are large, membrane-bound, or studied in living cells by in-cell NMR can in general not be assigned by the conventional solution NMR method that relies on uniform 13C/15N-labeling because the resonance lines become too broad and overlapping. Sparse labeling, in particular of methyl groups is an approach to restore interpretable spectra by drastically reducing relaxation rates and the number of signals. However, it remains difficult to establish resonance assignments, even if the three-dimensional structure of the protein is already known. In this project we develop computational methods to assign the spectra of sparsely labeled difficult proteins by structure-based automated assignment. On the basis of our existing FLYA algorithm for the assignment of uniformly labeled proteins, a new automated assignment method has been developed that can assign large proteins using methyl group labeling or other sparse labeling techniques, and NOESY spectra in conjunction with a known three-dimensional structure. The FLYA algorithm finds assignments by optimizing a mapping between the expected peaks, which one anticipates to see based on the protein sequence, and the measured peaks, which have been identified by peak picking. The general design of the algorithm makes it possible to exploit peak lists from virtually any type of NMR spectrum. In particular, we could show that FLYA is able to assign proteins using as input exclusively NOESY spectra. The new approach has been applied to proteins large proteins up to 360 kDa size and to proteins in living cells.
1: 当初の計画以上に進展している
On the basis of the existing FLYA algorithm for the assignment of uniformly labeled proteins, we have developed a new automated assignment method that can assign large proteins using methyl group labeling or other sparse labeling techniques, and NOESY spectra in conjunction with a known three-dimensional structure. The new approach was successfully applied to the 36 kDa ATCase-r2 protein dimer from sparse NOESY data. Applications to other proteins up to 360 kDa size are under way. The automated assignment method is also being applied to in-cell NMR in order to support the first structure determinations of proteins directly in living eukaryotic cells.
For the final year of the project we plan to publish a full account of our methyl-FLYA method for structure-based assignment of large, specifically methyl-labeled proteins. We will perform automated assignment for five proteins with sizes between 27 and 360 kDa, for which experimental NOESY spectra are available. The first structure determinations of proteins directly in living eukaryotic cells, which use the automated assignment method, shall be completed and published.
研究成果発表のための学会参加費と論文等の出版にかかると経費が,予定よりもかからなかったため.次年度は,初年度の成果を発表する機会が増えるため,残った分の経費は次年度の成果発表の経費として使用予定.
すべて 2018 2017 その他
すべて 国際共同研究 (2件) 雑誌論文 (5件) (うち国際共著 5件、 査読あり 5件、 オープンアクセス 2件) 学会発表 (4件) (うち国際学会 3件、 招待講演 4件) 図書 (1件) 備考 (1件)
Journal of Physics Conference Series
巻: printing in progress ページ: -
Arch. Biochem. Biophys.
巻: 628 ページ: 24-32
10.1016/j.abb.2017.02.011
Nat. Struct. Mol. Biol.
巻: 24 ページ: 187-193
10.1038/nsmb.3345
Protein Sci.
巻: 26 ページ: 280-291
10.1002/pro.3080
J. Biomol. NMR
巻: 67 ページ: 63-76
10.1007/s10858-016-0084-3
http://www.cyana.org/