SummaryThrombomodulin, TM, is an endothelial cell membrane protein acting as a cofactor for the activation of plasma protein C. Soluble TM is present in plasma and urine of normal subjects. Enzyme immunoassay, EIA, for human TM was developed using mouse monoclonal antibodies against human placental TM in this paper. We obtained four types of the monoclonal antibodies against human placental TM. EIA sandwich method using three types of the monoclonal antibodies enabled us to measure almost all of 6 and 7 TM subspecies in plasma and urine, respectively, except 1 subspecies, 31 kDa TM. There was no interference from other components of plasma and urine in the assay conditions. Titration curves of purified TM in buffer or in normal plasma were linear within the range from 0.08 to 10 ng/ml. The coefficient of variation at 0.08 ng/ml TM was 4.7%. TM titer with buffer, assayed by this method, was reduced by the addition of thrombin at the final concentration of 20 U/ml, but the titer with plasma was not reduced even at 100 U/ml. These concentrations of thrombin are far larger than those which would be formed in circulation. TM levels in plasma and urine of normal subjects collected in the morning were 35.2 ± 8.32 ng/ml (n = 346) and 111 ± 31.6 ng/ml (n = 33), respectively. TM level in plasma did not differ from the level in serum. Circadian fluctuation of plasma TM was not significant in 10 normal adults, although a tendency of increase in TM excretion to urine was found rather in the day time than the other times. It was concluded that this EIA is reliable for TM assay in human plasma and urine, which will reflect activation or injury of endothelial cells.

Endothelial cell injury is thought to be one of the causative factors in thrombotic thrombocytopenic purpura (TTP). A novel index of endothelial injury, plasma thrombomodulin, was measured in 13 patients with acute TTP. The mean plasma concentration of thrombomodulin was elevated in patients with TTP (34.23 +/- 19.08 ng/ml) as compared with healthy subjects (16.99 +/- 2.63 ng/ml, P less than 0.001). Eight (61.5%) of 13 patients had high thrombomodulin values. Markedly elevated thrombomodulin levels were observed in TTP patients who had suffered from systemic lupus erythematosus, in whom plasma thrombomodulin was still elevated when they achieved remission. Five of these 13 patients with TTP had normal plasma levels of thrombomodulin. In addition, the plasma thrombomodulin concentrations were correlated well with von Willebrand factor antigen and tissue-type plasminogen activator antigen levels, both of which are released from stimulated or damaged endothelial cells. No difference was found in plasma thrombomodulin levels between patients who achieved remission and who did not. These findings suggest that the magnitude of the endothelial damage in TTP is variable among patients and that plasma thrombomodulin has limited clinical relevance to the severity of TTP.

Thrombosis and llaemostasis-@ E K. SchattauerVerlagsgesellschaft mbH (Stuttgart) 65 (4) 451 (1991) DDAVP Does not lnduce the Release of Thrombomodulin from Endothelial Cells Dear Sir, The endothelial cell surface membrane protein thrombomodulin (TM) binds thrombin with high affinity and acts as a cofactor for thrombin-catalyzed activation of anticoagulant protein C. When complexed with TM, thrombin loses its procoagulant activities (L ,2). A soluble form of TM is present in human plasma (3) and presumably represents a cleaved form of vascular TM with loss of part of the O-glycosylation site-rich region, the transmem

Thrombomodulin is an endothelial cell membrane protein that is cofactor required for rapid activation of plasma protein C. Recently, thrombomodulin was found also in plasma in normal subject. However, the information of the soluble thrombomodulin in plasma is very poor. We now report the preparation of polyclonal and monoclonal anti-human thrombomodulin and their application for study on plasma thrombomodulin.Thrombomodulin was extracted from human placenta by 0.5% of Triton X-100 and purified by DIP-thrombin column chromatography. Polyclonal anti-thrombomodulin antibody was obtained from rabbit serum. The three types of monoclonal anti-human thrombomodulin antibodies were obrained from hybrid cell of BALB/C spleen cell and SP2 mouse myeloma cell. The type 1 monoclonal antibody inhibited the thrombomodulin activity. The type 2 of the antibody did not inhibit the activity. The type 3 of antibody cross reacted with rabbit lung thrombomodulin. The IgGs of these antibodies were separated by protein A Sepharose. The plasma thrombomodulin was separated using polyclonal anti-thrombomodulin IgG Sepharose. The apparent molecular weight of plasma thrombomodulin was estimated by immunoblot analysis using monoclonal anti-thrombomodulin IgG. When run with mercaptoethanol, plasma thrombomodulin migrated mainly at Mr=85,000 against Mr=105,000 of tissue thrombomodulin. The plasma thrombomodulin had an apparent Km for 1 nM compared with 0.3nM for tissue thrombomodulin. The apparent Km for protein C was same between tissue and plasma thrombomodulin.

This two-site immunoassay measures erythrocyte aldose reductase by using monoclonal and polyclonal antibodies to recombinant human enzyme. Total incubation time is 2.5 h, and the limit of detection is < 0.05 microgram/L. Analytical recovery tested with blood samples from healthy and diabetic individuals was 101-106%. Average CVs within and between assays were 3.7% and 4.8%, respectively. The enzyme content determined by this system correlated well with the activity of aldose reductase isolated from the same erythrocyte preparations. The amount of erythrocyte aldose reductase per milligram of hemoglobin was higher in women than in men (P < 0.001), but no significant correlation was observed between the amount of enzyme and the age of the individuals. This assay method should provide useful clinical information to optimize administration of aldose reductase inhibitors for effective prevention and treatment of diabetic complications.