Subunits A and S inheritance in four families with congenital factor XIII deficiency

Bleeding risk and reproductive capacity among patients with factor XIII deficiency: a case presentation and review of the literature

Factor XIII deficiency is an uncommon, inherited bleeding disorder that usually manifests in infancy or early childhood, involving both boys and girls. We present the case of a woman who had experienced two previous intracranial bleeding events, and was treated before and during her current pregnancy with factor XIII concentrate. Her pregnancy was successful, and she experienced an uncomplicated vaginal delivery. To better understand the issues surrounding bleeding, reproductive capacity, and management of factor XIII deficiency during pregnancy, we conducted a systematic literature review using MEDLINE from 1966 to December 1998. We also examined the bibliographic references from all articles, and included all cases, case reports, or case series of patients with factor XIII deficiency. We retrieved data on 117 patients from 37 articles, the majority of which had type II deficiency. Among untreated patients with type II factor XIII deficiency, the literature suggests an elevated mortality rate due to uncontrolled bleeding and intracranial hemorrhage. Because of its high degree of efficacy, the evidence supports the use of life long prophylactic therapy with at least monthly infusions of factor XIII concentrate, including during pregnancy. The opinion that women with type II factor XIII deficiency have inevitable recurrent abortions, or that men are sterile, is not well substantiated. No data were found on whether treatment alters male reproductive capacity. A policy of universal factor XIII replacement, starting in childhood, will likely enable more patients to attain reproductive status. The development of an international data registry would optimally address both bleeding risk and reproductive capacity among patients with factor XIII deficiency.

Novel aspects of blood coagulation factor XIII. I. Structure, distribution, activation, and function

Blood coagulation factor XIII (FXIII) is a protransglutaminase that becomes activated by the concerted action of thrombin and Ca2+ in the final stage of the clotting cascade. In addition to plasma, FXIII also occurs in platelets, monocytes, and monocyte-derived macrophages. While the plasma factor is a heterotetramer consisting of paired A and B subunits (A2B2), its cellular counterpart lacks the B subunits and is a homodimer of potentially active A subunits (A2). The gene coding for the A and B subunits has been localized to chromosomes 6p24-25 and 1q31-32.1, respectively. The genomic as well as the primary protein structure of both subunits has been established, and most recently the three-dimensional structure of recombinant cellular FXIII has also been revealed. Monocytes/macrophages synthesize their own FXIII, and very likely FXIII in platelets is synthesized by the megakaryocytes. Cells of bone marrow origin seem to be the primary site for the synthesis of subunit A in plasma FXIII, but hepatocytes might also contribute. The B subunit of plasma FXIII is synthesized in the liver. Plasma FXIII circulates in association with its substrate precursor, fibrinogen. Fibrin(ogen) has an important regulatory role in the activation of plasma FXIII. The most important steps of the activation of plasma FXIII are the proteolytic removal of activation peptide by thrombin, the dissociation of subunits A and B, and the exposure of the originally buried active site on the free A subunits. The end result of this process is the formation of an active transglutaminase, which cross-links peptide chains through epsilon(gamma-glutamyl)lysyl isopeptide bonds. Cellular FXIII in platelets becomes activated through a nonproteolytic process. When intracytoplasmic Ca2+ is raised during platelet activation, the zymogen--in the absence of subunit B--assumes an active configuration. The protein substrates of activated FXIII include components of the clotting-fibrinolytic system, adhesive and contractile proteins. The main physiological function of plasma FXIII is to cross-link fibrin and protect it from the fibrinolytic plasmin. The latter effect is achieved mainly by covalently linking alpha 2 antiplasmin, the most potent physiological inhibitor of plasmin, to fibrin. Plasma FXIII seems to be involved in wound healing and tissue repair, and it is essential to maintaining pregnancy. Cellular FXIII, if exposed to the surface of the cells, might support or perhaps take over the hemostatic functions of plasma FXIII; however, its intracellular role has remained mostly unexplored.

Factor XIII: inherited and acquired deficiency

Factor XIII (XIII), an enzyme found in plasma (present as a pro-enzyme), platelets and monocytes, is essential for normal haemostasis. It may also have a role to play in the processes of wound healing and tissue repair. Inherited XIII deficiency results in a life-long, severe bleeding diathesis which, if untreated, carries a very high risk of death in early life from intracranial bleeding. XIII is a zymogen requiring thrombin and calcium for activation. In plasma, XIII has two subunits: the 'a' subunit, which is the active enzyme, and the 'b' subunit which is a carrier protein. Activated XIII modifies the structure of clot by covalently crosslinking fibrin through an epsilon (gamma-glutamyl)lysine link. It also crosslinks other proteins, including fibronectin and alpha-2-plasmin inhibitor (alpha-2PI), into the clot through the same link. Clot modified by XIII is physically stronger, relatively more resistant to fibrinolysis and may be a more suitable medium for the ingrowth of fibroblasts. Inheritance of factor XIII is autosomal recessive. The majority of patients with the inherited defect show no XIII activity and absence of 'a' subunit protein in plasma, platelets and monocytes. At the molecular level, the defect is not a major gene rearrangement or deletion, but most likely a single point mutation which may be different in each family. Because of the severity of the bleeding diathesis, prophylaxis is desirable and has been shown to be very effective as the in vivo half-life of plasma XIII is long, and low plasma levels are sufficient for haemostasis. Acquired inhibitors have been reported in only two cases with inherited XIII deficiency. Acquired XIII deficiency has been described in a variety of diseases and bleeding has been controlled by therapy with large doses of XIII in such conditions as Henoch-Schönlein purpura, various forms of colitis, erosive gastritis and some forms of leukaemia. Large dose XIII therapy has also been used in an endeavour to promote wound healing after surgery and bone union in non-healing fractures. The use of XIII in these conditions remains controversial. Very rarely a bleeding diathesis results from the development of a specific inhibitor to XIII arising de novo, often as a complication in the course of a disease or in association with long-term drug therapy. The bleeding diathesis in these patients is difficult to treat.

Clinical pharmacokinetics of a placenta-derived factor XIII concentrate in type I and type II factor XIII deficiency

Limited data are available about the pharmacokinetics of placenta-derived factor XIII (FXIII) concentrate in patients with FXIII deficiency. This concentrate contains only the active subunit A but not the carrier subunit B of the factor, and perplexities have been raised about its clinical use. Moreover, no data are available on its use in the rare patients completely lacking both subunit A and subunit B. Therefore, we evaluated the pharmacokinetics of a commercial placenta concentrate in three patients with FXIII deficiency: two lacking subunit A (type II) and one lacking both subunits (type I). The elimination half-life of the infused placenta subunit A in the three patients was very similar (280, 283, and 272 hr) and was also consistent with the previously reported data for plasma-derived FXIII. No thrombin-independent activity was observed in our concentrate batches. The recovery was significantly lower in the type I patient, in whom infusion of subunit A was not able to elicit a monthly increment of subunit B, as usually observed in type II patients. Monthly infusions of placenta concentrate (at higher dosage in type I patient) have been administered to our patients for two to three years and no evidence of inhibitor against factor XIII activity has been observed. We conclude that placenta concentrates may be as effective as plasma derivatives in replacement therapy of factor XIII deficiency, even in patients who lack subunit B.

Human mononuclear phagocyte transglutaminase activity cross-links fibrin

The physiologic function of the monocyte transglutaminases is not known. In this study, we detected Factor XIII A-subunit antigen and "tissue" transglutaminase antigen in human monocytes by polyacrylamide gel electrophoresis and immunoblotting techniques. Flow cytometric analysis demonstrated that 27% and 49% of the total Factor XIII antigen in monocytes and human peritoneal macrophages, respectively, are expressed on the surface of the cells. Monocytes maintained in culture for 8 days had a 4-fold increase in Factor XIIIa activity and a 3.2-fold increase in the amount of Factor XIII antigen/mg cell protein. However, there was no increase in the "tissue" transglutaminase activity or antigen levels in cultured monocytes. In addition, we identified a Factor XIII deficient individual who does not express Factor XIII activity or antigen in plasma, platelets, monocytes, lymphocytes or erythrocytes. Intact monocytes from normal donors were able to cross-link fibrin formed in the plasma from the Factor XIII deficient individual. This suggests that transglutaminase activity expressed by peripheral blood monocytes may play a physiologic role in cross-linking fibrin during blood clotting or inflammation.

Localization of the coagulation factor XIII B subunit gene (F13B) to chromosome bands 1q31-32.1 and restriction fragment length polymorphism at the locus

In situ hybridization of tritiated cDNA probes for the gene for the B subunit of coagulation factor XIII localized the F13B locus to bands q31-q32.1 on human chromosome 1 and perhaps more precisely to sub-bands 1q31.2 or 1q31.3. Restriction fragment length polymorphisms (RFLPs) were detected with BglII, EcoRI and XbaI. Because the RFLPs detected with each of the three enzymes were concordant in every individual studied and since each showed a similar size difference, it was concluded that the RFLPs probably result from an insertion or deletion of length approximately 0.37-0.4 kb.

Successful pregnancy in a woman with congenital factor XIII deficiency treated with substitutive therapy. Report of a second case

A syndrome of marked fetal wastage is associated with congenital factor XIII deficiency in adult women. A previously unreported case of a woman with factor XIII deficiency is described, in which substitutive treatment with normal plasma or placental factor XIII concentrate permitted two normal pregnancies. Factor XIII activity was maintained above 1-2% with intermittent infusion of 300 ml to 450 ml of plasma every 14 days or of 500 units of concentrate every 21 days. This case confirms the only other case so far reported in which factor XIII substitutive therapy was able to permit a normal pregnancy in a woman with factor XIII deficiency and seems to suggest factor XIII to be involved in the process of annidation.

Neonatal factor XIII deficiency

We describe a patient diagnosed in the neonatal period as having factor XIII deficiency who presented with persistent umbilical bleeding. Factor XIII deficiency is the only coagulation factor deficiency that cannot be detected by classical hemostatic tests, and a rapid diagnosis is vital during the first decade of life. A newborn presenting with persistent umbilical stump bleeding should be screened for factor XIII deficiency when routine coagulation tests prove normal.

Coagulation factor XIII: a useful polymorphic genetic marker

The plasmas of two groups of subjects were examined for blood coagulation Factor XIII-A (FXIII-A, F13A) by electrophoresis in agarose using a Tris-EDTA-borate buffer to separate the common variants, F13A*1, F13A*2, and F13A*3. Dimeric subunits were visualized in UV light as monodansyl cadaverine bound to casein at the position of the transglutaminase activity representing F13A. One test group consisted of 307 members of three large Caucasian families. The other consisted of 148 consecutive patients whose plasmas had been sent to the clinical laboratory for determination of prothrombin time. Segregation analysis and father-to-son transmission confirmed that F13A is inherited as an autosomal co-dominant trait. The allelic frequencies in the random sample were F13A*1 = 0.82 and F13A*2 = 0.18. This sample included both blacks and whites, and the gene frequencies were not significantly different in the two races. The gene frequencies among the unrelated spouses of the three white families were A*1 = 0.75, A*2 = 0.24, A*3 = 0.01. Genetic equilibrium was present in both groups. The degree of polymorphism, the availability of blood, the ease of assessment, the absence of selective pressure, and the uniformity of gene frequencies in two major American ethnic groups make F13A a very useful marker for linkage studies and paternity testing. F13A has been provisionally assigned to chromosome 6. Linkage analysis of our family data did not provide evidence of linkage to two chromosome 6 markers, properdin factor B (BF) and glyoxalase 1 (GLO). The highest lod score (Z) was between F13A and the Kidd (Jk) blood group (theta = 0.68 at = 0.24).

Hereditary factor XIII deficiency: report of four families and definition of the carrier state

Definition of the carrier state of hereditary factor XIII deficiency was attempted by a study of four unrelated families with patients affected by the disease. Ten homozygotes, 14 obligatory carriers and 20 suspected carriers were available for determination of plasma factor XIII subunits A and B levels (by electroimmunoassay) and for assay of factor XIII enzymatic activity (by 14C-putrescine incorporation into casein). Evaluation of the ability of seven variables related to these tests to discriminate between healthy controls and obligatory carriers disclosed that two variables related to the enzymatic assay classified correctly 92.9% of the carriers and 73.3% of the controls, whereas the variables related to subunit A had a lower discriminative ability. Entering both enzymatic and immunologic variables into the discriminant function did not improve the result of the former. A classification function was established from which the posterior probability of suspected carriers can be determined by measuring the incorporation of 14C-putrescine into casein at 90 min after adding the labelled amine.

Factor XIII deficiency

A girl and her newborn brother with factor XIII deficiency from a family, which has not previously been reported, as described; two other Israeli families are reviewed. The sexes are equally affected. In two of the three families there was consanguinity among the parents. The families fit autosomal recessive inheritance and rule out X-linked recessive mode of transmission. The proposita was born to unrelated healthy parents from the Bnei Israel Jewish community of Bombay. She was re-admitted to hospital at the age of 15 days after umbilical bleeding. Later in infancy there were repeated haematomata and the diagnosis of factor XIII deficiency was established at the age of 1 1/2 years.

Failure to measure plasma activated factor XIII during disseminated intravascular coagulation

The search for activated products of coagulation factors in blood is of considerable clinical interest because its detection would indicate activation of the clotting system. Factor XIII circulates as inactive zymogen and becomes enzymatically active after thrombin or factor Xa activation. Employing an amine-incorporating system, activated factor XIII was measured in normal and pathological plasma from leukemic patients with overt laboratory signs of disseminated intravascular coagulation (DIC). Only traces of the activated factor were detected in both normal and DIC plasma. The method, sensitive and specific, as shown by the results of measurements on the plasma of 8 patients with congenital deficiency and on normal serum, did not prove to be useful for detecting pathological in vivo thrombin generation.

An electrophoretic and quantitative analysis of coagulation factor XIII in normal and deficient subjects

Previous electrophoretic studies of the A and B subunits of factor XIII have revealed considerably genetic heterogeneity. The present work investigates the electrophoretic forms and quantitates the A and B subunits in a family with inherited factor XIII deficiency. The data indicate that the deficiency in this family is due to a null allele at the locus controlling the A subunit. All family members were found to have decreased levels of B subunit. The data also indicate that there is no difference in thrombin activated transamidase activity between normal individuals with the three commonly occurring electrophoretic phenotypes of the A subunit.

Fibrin cross-linking in congenital factor XIII deficiency

Homozygous patients with factor XIII deficiency are devoid of immunologically identifiable A protein, the active enzymatic component. Quantitative studies of transamidase activity of the factor are available in only a few cases, and the fibrin cross-linking pattern is not well known. The present paper deals with the quantitative estimation of factor XIII transamidase activity (dansylcadaverine system), factor XIII molecular subunits, and the corresponding fibrin cross-linking pattern in seven homozygous patients with factor XIII deficiency. The results indicate that transamidase activity was present in all patients, and the range was 0.5-1.7%. The pattern of fibrin stabiisation showed an absence of cross-linking in two patients, the presence of gamma-gamma-dimers (traces) in four, and gamma-gamma-dimers plus incomplete alpha-polymers (traces) in one patient. In conclusion, the homozygous patients reported here were not completely devoid of functioning factor XIII.

A specific, fluorescent activity staining procedure applied to plasma and red blood cells in congenital factor XIII deficiency

The activity staining procedure introduced by Stenberg & Stenflo (1979) has been applied to studies on human blood transamidases (transglutaminases; endo-gamma-glutamine:epsilon-lysine transferases; e.g. factor XIII). The technique combines agarose gel electrophoresis with activity staining based on the transamidase catalysed incorporation of monodansylthiacadaverine (N-(5-amino-3-thiapentyl)-5-dimethylamino-1-naphtalenesulfonamide) into casein. The method permits detection of plasma factor XIII activity down to 1% of the normal adult standard. The technique was used on plasma from two patients with tentative congenital plasma factor XIII deficiency (based on clot solubility). No activity was found in platelet poor as well as in platelet rich plasma which confirmed the diagnosis. In the erythrocytes studied in genetic determinations of the plasma and red blood cell transamidases. Using immunoelectrophoresis, the plasma factor XIII b subunit was found to be 43% and 44% of the concentration in normal standard plasma.