Factor XIII

Saturday July 12th, 2003
9.00 to 13.00
Hall 9
Birmingham International Conference Centre


Chair: RAS Ariëns-UK
Co-Chairs: A Ichinose-Japan, P Bishop-USA
Active Members: CS Greenberg-USA, L Muszbek-Hungary

Apologies were received from Dr Greenberg. All other chairs and members were present. The FXIII SSC meeting had a full programme, all speakers and approximately 130 people in the audience attended the meeting. The presentations were of a high standard and ranged from novel methodology including gene knockout, atomic force microscopy, site-directed mutagenesis to study protein interactions and protein chemistry to development of the first international standard for FXIII. The presentations of Dr Lim and Dr Raut were exchanged, as Dr Raut had to be at another SSC meeting in the late morning.

The meeting was opened by Dr Ariëns, University of Leeds, UK, who put forward for discussion the issue of whether the FXIII SSC should continue as a separate committee or whether it could merge with that of fibrinogen to a combined Fibrinogen and FXIII subcommittee. Dr Ariëns pointed out that most subcommittees relate to work in areas of research that involve many proteins and genes such as that of fibrinolysis, coagulation inhibitors and contact activation, and that there are only three subcommittees out of 20 that are dealing with one coagulation factor in isolation: the SSC for FXIII, fibrinogen and von Willebrand factor. Dr Ariëns suggested that it might be a good idea to combine the FXIII and fibrinogen SSC in one subcommittee as the two proteins are functionally closely involved in the final stages of the coagulation cascade. It was also suggested that thrombin could be included in this new SSC as well. Advantages of a possible merger would be that there would be a larger platform to discuss novel ideas regarding fibrinogen, FXIII and thrombin, with people present from various expertise. Work relating to standardization would be easier to perform and to discuss. Issues regarding standardization for FXIII, fibrinogen and thrombin are more often related than not, such as in the case of fibrin glue for example, that contains both FXIII and fibrinogen.

The idea for a merger with the fibrinogen SSC was met by a short but lively discussion. Dr Lord, University of North Carolina, USA, Acting Chair of the fibrinogen SSC, was supportive of a possible merger and mentioned that the idea would also be discussed in the fibrinogen subcommittee to be held the same afternoon. Dr Bishop, Zymogenetics, expressed concern that there would not be enough space on the agenda of a merged subcommittee to discuss all relevant issues. Dr Muszbek, Debrecen University, Hungary, said that the SSC for FXIII and fibrinogen used to be combined several years ago and that it was then decided to separate them. He expressed concern that issues relating to fibrinogen would dominate meetings of a merged SSC. Dr Kohler, University of Bern, Switzerland, argued the opposite, and suggested FXIII issues may dominate those of fibrinogen. Overall, there seemed to be concern regarding dominance of one of the factors over the other and a possible lack of agenda space in meetings of a merged SSC. Dr Ariëns suggested that space issues could possibly be solved with longer agendas if necessary. Dr Muszbek suggested organizing a trial of a combined SSC, by holding a combined meeting for FXIII and fibrinogen at the next SSC. This suggestion was welcomed by Dr Ariëns. The agenda item was closed with a call to everyone in the audience to express their opinion on the matter by filling in a questionnaire form, so that a recommendation on this issue in some form or the other could be presented by the chair at the SSC Business meeting.

The main programme of the FXIII SSC was opened with a presentation from Dr Ichinose, Yamagata University, Japan, on abortion in FXIII gene-knockout mice. Dr Ichinose started his presentation with two minutes silence in honour of the work of Dr Peter Steinert who recently passed away. Dr Steinert has characterised many of the transglutaminase genes. Dr Ichinose presented data on the molecular biology of FXIII deficiency. There are approximately 500 cases of FXIII A-subunit deficiency known. Mutations occur over the whole molecule. There do not appear to be any particular hotspots for mutations in this gene. B-subunit deficiency is rare, currently only 4 cases are known with a characterized mutation in the B-subunit gene. FXIII deficiency is associated with severe bleeding diathesis, in particular umbilical cord, intracraneal and peritoneal bleeding, with poor wound healing and with miscarriage in pregnancy. Dr Ichinose has developed a FXIII A-subunit gene-knockout model in mice for the study of miscarriage in FXIII deficiency. The main symptoms in the A knockout mice were vaginal and intrauterine bleeding. Dr Ichinose concluded from this mouse model that spontaneous abortion in FXIII deficiency is caused mainly by bleeding.

Dr Kohler, University of Bern, Switzerland, presented data on the role of FXIII in vascular disease. Dr Kohler mentioned the importance of distinguishing between chronic and acute vascular disease. The first interest in the role of FXIII in vascular disease was developed by findings of an association between FXIII Val34Leu and vascular disease. Recent studies suggest that this polymorphism affects fibrin structure function through an interaction of FXIII Val34Leu and fibrinogen levels. Other studies have shown an interaction between insulin resistance and FXIII subunit levels, B-subunit levels in particular. Factor XIII levels have been shown to relate to the extent of coronary artery disease. In the Northwich Park Heart Study II, FXIII A-subunit levels were reduced in the acute phase of the disease, with no difference in B-subunit levels. The reduction in FXIII A is possibly due to activation and increased consumption. This is in agreement with a study in acute stroke, where FXIII A-subunit was reduced in patients with poor survival and the level inversely correlated with F1+2. In another study of acute pulmonary embolism, FXIII A-subunit levels were also reduced, with normal B-subunit levels. The relationship was linear with occlusion rate, and Dr Kohler suggested that FXIII consumption is related to the thrombus size. The relationship between FXIII and vascular disease indicate the importance of standardization of the assays to measure FXIII levels (activity, A-subunit and B-subunit).

Results from a pilot study for the standardization of FXIII activity assays were reported by Dr Raut, National Institute for Biological Standards and Control, UK. Phase 4 (out of 7) of the study was performed. Data were obtained on two FXIII concentrates and 1 fibrinogen concentrate. The aims of were to compare the Berichrom FXIII activity assay and the new Pentapharm FXIII activity assay. The two assays use different principles to measure FXIII activity, in the Pentapharm assay incorporation of pentylamine into fibrinogen by FXIII is measured, whereas in the Berichrom assay the amount of ammonia released during the cross-linking reaction is measured. Another aim was to analyse assay differences in the presence or absence of fibrinogen. Collaborating laboratories were that of Dr Muszbek, Dr Seitz and Dr Barrowcliffe. It was found that differences between the two assays were large when buffer was used as diluent, whereas differences were small with FXIII deficient plasma as diluent. There was a high variability of measurements in fibrinogen concentrates. Dr Raut ended with an update on the development of the first international standard for FXIII. Recombinant FXIII was suggested as possible additional standard material, and a resource laboratory for this material would need to come forward. To establish an International standard for FXIII, 12-20 laboratories will have to participate and laboratories willing to participate are asked to come forward. Please contact the chair of this SSC or Dr Raut if you are interested. In the discussion, Dr Muszbek suggested that the difference between variability between buffer and deficient plasma as diluent is due to the presence of fibrinogen in plasma, which greatly enhances activation rates of FXIII. This may lead to differences between assays that employ different thrombin concentrations and have different sensitivity to the activation step.

Dr Weisel, University of Pennsylvania, USA, presented data on the effect of fibrin a- and g-chain cross-linking by activated FXIII on fibrin structure and function. Recombinant Aa251 fibrinogen lacks the aC domain and its connector polypeptide, and is therefore deficient of the a-chain FXIII cross-linking sites. Dr Weisel showed data using this fibrinogen to quantify the relative effects of a- and g-chain cross-linking on fibrin structure and elastic properties. Experiments were performed with purified FXIII from pooled plasma, which was fully activated to avoid any effects of the Val34Leu polymorphism on fibrin structure and stiffness. Recombinant Aa251 fibrinogen produced clots with smaller pores, thinner fibers and increased number of branchpoints when compared with normal fibrinogen. Cross-linking by pooled plasma FXIII had relatively little effect on overall fibrin structure, but there was a dramatic effect on stiffness of the clot. Cross-linked Aa251 fibrin was much less stiff and showed more non-elastic slippage of the protofibrils than cross-linked normal fibrin. This effect was due to the absence of a-chain cross-linking in Aa251 fibrin. Gamma chain cross-linking also contributed to clot stiffness as Aa251 fibrin showed reduced stiffness in the absence of FXIIIa when compared to Aa251 fibrin in the presence of FXIIIa, the latter of which is expected to show normal -gamma chain cross-linking. The effects of alpha and gamma-chain cross-linking were also noted on the lysis speeds of the fibrin clots as analysed by laser scanning confocal microscopy.

After a short coffee break of 15-20 minutes, the meeting was resumed with a presentation on the use of FXIII peptides to investigate thrombin-FXIII interaction by Dr Maurer, University of Louisville, USA. Using a variety of novel methods, including 2D NOESY spectra, MALDI-TOF and pulsed alkylation mass spectrometry, Dr Maurer investigated the relative roles of P1-4 of the FXIII activation peptide in the interaction with thrombin. It was found that the P4 residue, which is the site of a common Val34Leu polymorphism, plays an important role in the interaction with thrombin. A change of Val to Leu change at this position dramatically improved the kcat as well as Km. Substitution of P4 with Ile or Ala improved the Km but the kcat worsened. The future plans are to apply the same methodology to study interaction with the whole FXIII protein, possibly as a recombinant. The aim is to obtain a fuller understanding of the conformational changes that occur in FXIII upon binding to thrombin.

In the next presentation by Dr Philippou, Imperial College, UK, FXIII and thrombin interactions were investigated from the thrombin point of view. Dr Philippou used a library of 58 thrombin mutants encompassing a total of 78 mutated surface exposed residues, to study their interaction with FXIII. The mutants were screened with a pentylamine incorporation assay using casein, fibrinogen or fibrin as substrate, and the results were confirmed by kinetic analysis of the activation peptide release by HPLC. The results showed that residues R78/R180/D183, W50, E229, R233 of thrombin were involved in direct interaction with FXIII during its activation. Additional thrombin residues H66, Y71, N74 were involved in binding to fibrin to cause the fibrin-enhancement effect of FXIII activation. Thrombin has many substrates, and general implications were drawn from these findings for the mechanism by which thrombin chooses its substrate. It was concluded that thrombin’s fate is directed primarily by competition between its cofactors. In the specific case for FXIII for example, activation is directed by competition between fibrin and thrombomodulin for their overlapping binding site on thrombin. A model was proposed to explain the enhancement effect of fibrin on FXIII activation.

Dr Reynolds
from Zymogenetics, USA, presented the first data regarding pharmacokinetics and safety of a recombinant FXIII preparation in healthy human volunteers. Recombinant FXIII A-subunit was produced in yeast, and administered at doses of 2, 5, 10, 25 and 50 u/kg to 8 healthy subjects. Placebo was administered in 2 subjects. The highest dose of Rec FXIII was chosen such that it would increase the baseline level of FXIII by 100%. Levels were monitored using the Berichrom FXIII activity assay. Halflife of FXIII after injection was 9-11 days, which is in good agreement with previous findings using plasma substitution. There was a specific increase of FXIII of 1.77% per 1U/kg. The product was found to be safe as adverse effects such as headache, cramp, pain in limb, prolongation of thrombin time, abdominal pain or cough were all absent. There was an increase in D-dimer and a slight drop in FXIII B-subunit levels at high concentrations of recombinant FXIII.

Dr Lim, University of Leeds, UK, presented data using novel technology of atomic force microscopy (AFM) to investigate the structure of fibrin(ogen) and FXIII. AFM was developed in 1986 for applications in physics, but it only since recently that it is being used more often in lifesciences. AFM allows real-time visualization under physiological aqueous conditions at almost unlimited resolution potential. Topographic images are generated by a piezo-electric scanner that moves or taps a microtip over the sample. The resolution is determined by the size of the tip with which the microscope scans or ‘taps’ the surface. Current tips used are around 10-30 nm in diameter, but future usage of tips of around 1 nm using carbon nanotubes should improve resolution. For comparison, length of a fibrinogen molecule is around 45 nm. Further advantages of AFM are that it can be used to measure attractive and repulsive forces on a single molecule level. Potential limitations of AFM are that the sample can be altered or damaged by the tip movement. AFM images were shown of fibrinogen molecules, showing their characteristic three-nodular appearance. Images were also shown of fibrin polymerization at various time-frames after the addition of thrombin, showing growing protofibrils that aggregate laterally and branch out. The first AFM images of FXIII were produced, showing a tetrameric structure that appears to be in agreement with the models of tetramer association based on transmission electron microscopic images of FXIII published by Carrell et al in 1989.

An update of the ETRO FXIII deficiency registry was presented by Dr Ivaskevicius, Frankfurt Red Cross Blood Center, Germany. A background was given about the structure of the FXIII A- and B-subunit genes and FXIII deficiency, which is an autosomal recessive disorder that affects around 1 subject in every 1-3 million. The registry had 72 entries in 1996, 96 in 2000 and 109 in 2003, with entries from a total of 20 different countries. Around 45% of the entries have also been genetically characterized. A particular mutation in intron E (IVS5-1 G>A) was most common, possibly due to an ancient founder effect, as it was found in several countries. Dr Ivaskevicius continued the presentation with data showing the usage of Denaturing HPLC (Wave technology) to screen DNA for mutations. DHPLC proves to be a useful and rapid methodology for screening, although some mutations are not detected. The average detection rate is reported to be around 95%. Further characterization in healthy subjects is required to exclude the possibility that a certain change found in the DNA of the FXIII deficient patients is a common genetic polymorphism.

The meeting was closed at 12.55.

The questionnaire slip was returned by 48 people, of whom
17 or 35% voted in favor of a merger of the FXIII and fibrinogen SSC,
29 or 61% against and
2 or 4% said they were indifferent.