抄録
Most of the active components of polypeptides have a complex molecular structure, large molecular size. Such components may also be structurally heterogeneous. Therefore, development of a method that can confirm the consistency of polypeptides amino-acid sequences for product characterization is desirable. In general, it is extremely difficult to distinguish differences of a few amino acid residues in the 1H-NMR spectrum of polypeptides with molecular weights greater than several thousand. However, we have been able to distinguish between three insulin species differing in one to three amino acid residues using a combination of multivariate statistics and 1H-NMR spectra. These results demonstrate that this methodology could be useful for characterization of polypeptides.
