A tetrapeptide, residues 6 to 9 in normal prothrombin, was isolated from the NH2-terminal, Ca2+-binding part of normal prothrombin. The electrophoretic mobility of the peptide was too high to be explained entirely by its amino-acid composition. According to 1H nuclear magnetic resonance spectroscopy and mass spectrometry, the peptide contained two residues of modified glutamic acid, -y-carboxyglutamic acid (3-amino-1,1,3-propanetricarboxylic acid), a hitherto unidentified amino acid. This amino acid gives normal prothrombin the Ca2 +_ binding ability that is necessary for its activation. Observations indicate that abnormal prothrombin, induced by the vitamin K antagonist, dicoumarol, lacks these modified glutamic acid residues and that this is the reason why abnormal prothrombin does not bind Ca2+ and is nonfunctioning in blood coagulation.Prothrombin is a plasma glycoprotein that is activated during the process of blood coagulation to the proteolytic enzyme thrombin. The biosynthesis of prothrombin is vitamin K dependent, and deficiency of this vitamin or administration of the vitamin K antagonist, dicoumarol, gives rise to an abnormal prothrombin which does not function in blood coagulation (1-6). The activation of prothrombin in vivo requires the binding of Ca2+ (7); abnormal prothrombin does not bind Ca2+ (2, 8, 9). During the activation of normal prothrombin an NH2-terminal fragment (molecular weight approximately 25,000) is split off; the difference between abnormal and normal prothrombin has been localized to this part of the molecule. Evidence has been produced that the difference is due to the lack of certain prosthetic groups in abnormal prothrombin (10-13).In an endeavour to define the difference between normal and abnormal prothrombin, the NH2-terminal fragments from both proteins were isolated and degraded further. A heptapeptide from normal prothrombin (residues 4 to 10) and a corresponding heptapeptide from abnormal prothrombin were isolated by BrCN degradation and trypsin digestion. The heptapeptide from normal prothrombin differed from the corresponding peptide in abnormal prothrombin in that it had a higher anodal electrophoretic mobility at pH 6.5 (13).By extensive proteolytic digestion, the heptapeptide from normal prothrombin was degraded to a tetrapeptide. This tetrapeptide, containing residues 6 to 9, still had an abnormally high anodal electrophoretic mobility at pH 6.5. This paper reports evidence that each of the two glutamic acid residues of this peptide are modified by replacement of one hydrogen on the -y carbon atom by a carboxyl group. This work will be described in greater detail elsewhere. MATERIALS AND METHODSIsolation of Tetrapeptide. The heptapeptide from normal prothrombin (residues 4 to 10) (ref. 13) was first thoroughly digested with aminopeptidase M (Sigma) and afterwards with carboxypeptidase B (Sigma). A tetrapeptide was isolated from the digest by gel chromatography on Sephadex G-25 superfine and obtained in pure form as judged by high voltage electrophoresis at...

21] 12 years after diagnosis, P < 0.05), ICA developed after diagnosis in 6 of 105 originally antibody negative mostly overweight patients. In conclusion, multiple islet antibodies or GADA alone at diagnosis of diabetes predict future complete ␤-cell failure. After diagnosis, GADA persisted in most patients, whereas ICA development in patients who were antibody negative at diagnosis indicated decreasing ␤-cell function.

In patients considered type II diabetic with ICA, beta-cell function progressively decreased after diagnosis, and after 3 yr was similar to type I diabetic patients, whereas beta-cell function in type II diabetic patients without ICA was unchanged.

Previous work has shown that the light chain of protein C, an anticoagulant plasma protein, contains an unusual amino acid [Fernlund, P. & Stenflo, J. (1982) J. Biol. Chem. 257, 12170-12179]. To determine the structure of this amino acid a heptapeptide, CMCys-Ile-X-Gly-Leu-Gly-Gly (residues 69-75 in the light chain), was isolated from enzymatic digests of the light chain. According to automatic Edman sequence analysis, 1H NMR spectroscopy, and mass spectrometry the heptapeptide had beta-hydroxyaspartic acid in its third position, which corresponds to position 71 in the light chain of protein C. Analysis of acid and aminopeptidase M hydrolysates of the heptapeptide showed the beta-hydroxyaspartic acid to be the erythro form. Acid hydrolysis of protein C released approximately equal to 1 mol of beta-hydroxyaspartic acid per mol of protein. The function of this amino acid, which, to the best of our knowledge, has not been found previously in proteins, is unknown.