Key Points• The K d for the association of FXIII subunits is in the range of 10 210 M, and in plasma approximately 1% of FXIII-A 2 exists in free form.• The binding site for FXIII-A is located within the 2 N-terminal sushi domains of FXIII-B.Coagulation factor XIII (FXIII) is a heterotetramer consisting of 2 catalytic A subunits (FXIII-A 2 ) and 2 protective/inhibitory B subunits (FXIII-B 2 ). FXIII-B, a mosaic protein consisting of 10 sushi domains, significantly prolongs the lifespan of catalytic subunits in the circulation and prevents their slow progressive activation in plasmatic conditions. In this study, the biochemistry of the interaction between the 2 FXIII subunits was investigated. , it can cross-link fibrin. In cerebrospinal fluid and tears with much lower FXIII subunit concentrations, >80% of FXIII-A 2 existed in free form. A monoclonal anti-FXIII-B antibody that prevented the interaction between the 2 subunits reacted with the recombinant combined first and second sushi domains of FXIII-B, and its epitope was localized to the peptide spanning positions 96 to 103 in the second sushi domain. (Blood.

Coagulation factor XIII (FXIII) is converted by thrombin and Ca 2 + into an active transglutaminase (FXIIIa) in the fi nal phase of coagulation cascade. Its main function is the mechanical stabilization of fi brin clot and its protection from fi brinolysis by cross-linking of fi brin chains and α 2 -plasmin inhibitor to fi brin. In non-substituted patients FXIII defi ciency is a severe hemorrhagic diathesis, not infrequently with fatal consequences. The main reason for using FXIII assays is the diagnosis of FXIII defi ciency. The aim of this review is to provide a comprehensive critical evaluation of the methods reported for the determination of FXIII activity in the plasma. Such methods are based on two principles: 1) measurement of labeled amines incorporated by FXIIIa into a glutamine residue of a substrate protein, 2) monitoring ammonia released from a peptide bound glutamine residue by FXIIIa using NAD(P)H dependent glutamate dehydrogenase indicator reaction. The incorporation assays are sensitive, but cumbersome and time-consuming, they are diffi cult to standardize and cannot be automated. The ammonia release assays are less sensitive, but quick, well standardized, and can be automated; this type of assay is recommended for the screening of FXIII defi ciency. The traditional clot solubility assay should not be used for this purpose.

To cite this article: Pé nzes K, Kö vé r KE, Fazakas F, Haramura G, Muszbek L. Molecular mechanism of the interaction between activated factor XIII and its glutamine donor peptide substrate. J Thromb Haemost 2009; 7: 627-33.Summary. Background: Activated factor XIII (FXIII), a dimer of truncated A-subunits (FXIII-A 2 *), is a transglutaminase that crosslinks primary amines to peptide-bound glutamine residues. Because in the few natural substrates of FXIII-A 2 * no consensus sequence could be identified around the reactive glutamine, studying the interaction between individual substrates and FXIII-A 2 * is of primary importance. Most of the a 2 -plasmin inhibitor (a 2 PI) molecules become truncated by a plasma protease, and the truncated isoform (N1-a 2 PI) is an important substrate of FXIII-A 2 *. The crosslinking of N1-a 2 PI to fibrin plays a major role in protecting fibrin from fibrinolysis. Methods: We studied the interaction of FXIII-A 2 * with its dodecapeptide glutamine donor substrate, N1-a 2 PI(1-12), the sequence of which corresponds to the N-terminal sequence of N1-a 2 PI. Kinetic parameters for N1-a 2 PI(1-12) and for its truncated or synthetic mutants were determined by a spectrophotometric assay. The interaction of N1-a 2 PI(1-12) with FXIII-A 2 * was investigated by proton nuclear magnetic resonance (NMR) and saturating transfer difference (STD) NMR. Results and Conclusions: Kinetic experiments with peptides in which the Asn1 residue was either truncated or replaced by alanine and proton NMR analysis of the FXIII-A 2 *-N1-a 2 PI(1-12) complex demonstrated that Asn1 is essential for effective enzyme-substrate interaction. Experiments with C-terminally truncated peptides proved that amino acids 7-12 are essential for the interaction of N1-a 2 PI(1-12) with the enzyme, and suggested the existence of a secondary binding site on FXIII-A 2 *. Hydrophobic residues, particularly Leu10 and the C-terminal Lys12, seemed to be especially important in this respect, and direct interaction between hydrophobic C-terminal residues and FXIII-A 2 * was demonstrated by STD NMR.

The anti-FXIII-A alloantibody caused an unmanageable bleeding complication. The antibody was of combined subtype which accelerated the elimination of FXIII and exerted a multiple inhibitory effect on FXIII activation/activity.

Acquired FXIII deficiencies caused by autoantibodies against FXIII subunits represent rare but very severe bleeding diatheses. Alloantibodies in FXIII-deficient patients also cause life-threatening bleeding complications, but they develop extremely rarely. In this review we provide an overview of the diagnosis and classification of anti-FXIII antibodies and analyze 48 patients with autoimmune FXIII deficiency and four additional FXIII-deficient patients who developed anti-FXIII alloantibody. The patients were collected from peer-reviewed publications from which relevant data could be extracted. With the exception of two cases the antibodies were directed against FXIII-A. The difficulties in the diagnosis of FXIII deficiency in the presence of anti-FXIII antibodies are discussed and a scheme for the functional classification of the anti-FXIII antibodies is recommended. The three main categories are neutralizing and non-neutralizing antibodies and antibodies with combined effect. The methods being used for detecting and quantifying the inhibitory effect on FXIII activation and on the transglutaminase activity of activated FXIII are summarized and techniques for the classification of neutralizing anti-FXIII antibodies are outlined. The importance of clearance studies in these cases is emphasized. Binding assays, useful for the identification of non-neutralizing and combined type antibodies, were collected from the literature and their informative power is demonstrated by examples. The most frequently occurring bleeding symptoms in patients with anti-FXIII antibodies were soft tissue bleeding; intracranial bleedings also occurred, but less frequently than in inherited FXIII deficiency. Treatment of such patients is extremely challenging; the main aim should be eradication of the antibody.