| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Medicine, Division of Rheumatology;
Department of Chemistry and Biochemistry; and
Department of Pediatrics, University of California, Los Angeles, CA 90095
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2-glycoprotein-1 (2GPI), prothrombin (PT), protein C (PC), and protein S (7, 8, 9, 10, 11, 12). Hemostasis consists of primary and secondary components. The primary component involves platelet activation and aggregation, and the secondary component activates the plasma coagulation system and results in fibrin formation. The key coagulation cascade consists of 1) the expression of tissue factor (TF), which binds and activates factor VII (generating factor VIIa); 2) TF-VIIa complexes, which activate factors IX and X; 3) factor Xa, which works with factor Va to convert PT to thrombin, whereas factor IXa works with VIIIa to generate more factor Xa; and 4) thrombin, which converts fibrinogen to fibrin clots. The coagulation cascade is subjected to three major pathways of feedback regulation: 1) the TF pathway inhibitor inhibits TF-VIIa complexes from activating factors IX and X; 2) antithrombin (AT) binds to thrombin and inactivates thrombin; and 3) activated PC (which is generated from PC after a cleavage in the H chain of PC) works with protein S to inactivate factors Va and VIIIa (13, 14). Moreover, there is the fibrinolytic system that generates plasmin, which dissolves fibrin in hemostatic plugs and thrombi. As such, fibrin formation and fibrinolysis are in dynamic balance, and abnormalities in either hemostasis or fibrinolysis may promote thrombosis (14, 15).
Recently, during our studies of anti-PT Ab (aPT), we found that most, but not all, aPT-positive patient plasma samples contained anti-thrombin Ab (16). Moreover, the IS6 monoclonal IgG aPT (derived from a primary APS patient) binds to cardiolipin (CL) and thrombin, and five of seven patient-derived monoclonal IgG aCL cross-react with PT and thrombin (Table I) (16, 17, 18). Of these six thrombin-reactive IgG mAb, the CL24 mAb could interfere with inactivation of thrombin by AT (16). When CL24 and two other thrombin-reactive mAb were analyzed for their binding affinities to thrombin and PT by competitive inhibition, the results showed that soluble thrombin was more effective than PT in inhibiting all three mAb binding to either PT or thrombin in the solid phase. Importantly, thrombin could inhibit all tested mAb from binding to thrombin and PT, whereas PT could only inhibit mAb from binding to PT, but not thrombin. These results demonstrated that these three mAb are more specific for thrombin than PT. Based on these inhibition data, the relative Kd values of these anti-thrombin Ab are 
1.7–7.5 x 10–6 M (16).
|
Similarly, it is known that plasmin and thrombin also share homologous trypsin/thrombin-like serine protease domains with a 48% similarity (Fig. 1). Taken together with the discovery of anti-thrombin and anti-activated PC Ab in APS, we hypothesize that some APS patients may have Ab against plasmin and that some of the anti-plasmin Ab may interfere with fibrinolysis and thus promote thrombosis in such patients. In this study we report the detection of Ab against plasmin in APS patients and the inhibitory effects of one patient-derived IgG monoclonal anti-plasmin Ab on the lysis of fibrin clots. These findings define a novel anti-plasmin autoantibody in APS and show that some anti-plasmin autoantibody may interfere with the lysis of fibrin clots in the circulation.
|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The amino acid sequence of thrombin (residues 328–622 of PT, with the accession name of tbhu in the Protein Information Resource database) was compared with that of plasmin (residues 97–810 of the plasminogen precursor, pir:plhu) using the Gap program in the Genetics Computer Group software package (20). The Gap program uses a scoring matrix with matches scored as 1.5, and mismatches scored according to the evolutionary distance between the amino acids. The 3D structures of thrombin and plasmin in the ribbon and the space-filling models were generated based, respectively, on the coordinates of human thrombin (1A5G) (21) and human plasmin (1BML) (22), using RasMol software (version 2.7; Bernstein & Sons, Bellport, NY).
Patients and healthy controls
Plasma samples were obtained from 25 patients with APS and 20 normal controls at University of California Medical Center (Los Angeles, CA) and University of California Medical Center (San Diego, CA). All APS patients in this study satisfied the Sapporo classification criteria for definite APS (2). Medical charts and laboratory test reports for each patient entered in this study were reviewed by a rheumatologist (J. M. Grossman or B. H. Hahn). Patients were then classified as primary APS if they had no associated autoimmune disease or as secondary APS if they also fulfilled criteria for another autoimmune disease.
Of the 25 APS patients, 11 were primary APS (44%) and 14 were secondary APS (56%); the latter group included 12 patients with systemic lupus erythematosus (SLE), one with SLE-like disease, and one with both SLE and Sjögren’s syndrome. Twenty-two APS patients were positive for LAC (88%), and 21 were positive for aCL (88%). LAC were determined by either the dilute Russell’s viper venom time test or the rabbit brain neutralization procedure test (23).
Patient-derived monoclonal aCL and aPT
Seven IgG monoclonal aCL and one IgG monoclonal aPT were analyzed in the present study. The aCL included CL1, CL15, CL24, IS1, IS2, IS3, and IS4 (17), and the single aPT was IS6 (18). Their generation and characterization were reported previously (17, 18).
ELISA for Ab against plasmin and plasminogen
The ELISA for anti-plasmin and anti-plasminogen Ab was performed as follows. Briefly, high binding ELISA plates (Costar, Cambridge, MA) were coated with 5 µg/ml of either human plasmin or human plasminogen (Haematologic Technologies, Essex Junction, VT) in PBS, pH 7.4. After incubating overnight at 4°C, plates were blocked with PBS containing 0.25% gelatin. Then test plasma samples (1/300 dilution) or purified IgG (at the indicated concentrations) in PBS/0.1% gelatin were distributed to wells in duplicate and incubated for 1.5 h at room temperature. The control human IgG was obtained from Jackson ImmunoResearch Laboratories (West Grove, PA). After washing with PBS, bound human IgG was detected with HRP-conjugated goat anti-human IgG (
-chain specific; BioSource International, Camarillo, CA) and the peroxidase substrate tetramethylbenzidine (Kirkegaard & Perry Laboratories, Gaithersburg, MD).
A competitive inhibition assay was used to study the binding properties of selected mAb to plasmin and plasminogen. Briefly, each mAb (1 µg/ml) was preincubated for 1.5 h with various concentrations of either plasmin or plasminogen in PBS/0.1% gelatin; each mAb at 1 µg/ml was in the linear range of its titration curve. Then, the mixture was distributed to the plasmin- or plasminogen-coated wells in duplicate. After incubation, bound IgG was measured. The amount of inhibition for an mAb at a given concentration of soluble plasmin was calculated as follows: % inhibition of mAb binding to plasmin = [(OD from a test mAb alone) – (OD from the same mAb plus plasmin at the given concentration)]/(OD from the same mAb alone) x 100. The inhibition data of each mAb were used to calculate its relative Kd to plasmin and plasminogen (24).
A few studies reported plasmin digestion of some human IgG (25, 26), suggesting that some of the observed plasmin inhibition of mAb binding to plasmin on plates might actually be due to plasmin digestion of mAb. To address this issue, 2 µM plasmin was first incubated with a plasmin inhibitor, 2 µM aprotinin (Sigma-Aldrich, St. Louis, MO), for 30 min. Then, the plasmin-inhibitor mixture was serially diluted 10-fold, and the diluted samples were mixed with a test mAb in the competitive inhibition assay described above.
In addition, plasmin was irreversibly inactivated with N
-tosyl-L-lysine chloromethyl ketone-HCl (TLCK; Sigma-Aldrich), which forms a chemical bond with His622 (one of the three conserved catalytic residues of all proteases in the trypsin family). Briefly, 12 µM plasmin was first incubated with 6 mM TLCK in PBS (pH 7.4) for 100 min at room temperature, which inhibited >99% of the plasmin amidolytic activity using S-2403. After incubation, the excess TLCK was removed by dialyzing the plasmin-TLCK mixture against PBS at 4°C overnight. Thereafter, the TLCK-inactivated plasmin was mixed with a test mAb in the above-mentioned competitive inhibition assay. The percent inhibition of mAb binding to plasmin was calculated as described above.
Affinity purification of IgG from patients’ plasma
IgG was purified from two chosen plasma samples positive for IgG anti-plasmin Ab using a protein G-Sepharose column (Amersham Pharmacia Biotech, Piscataway, NJ) according to the manufacturer’s instructions.
Effects of anti-plasmin mAb on plasmin activity
The effects of mAb on plasmin activity were first studied using the plasmin chromogenic substrate Spectrozyme-PL (H-D-norleucyl-hexahydrotyrosyl-lysine-p-nitroanilide; American Diagnostica, Greenwich, CT) according to the manufacturer’s instructions. Briefly, 12.5 µl of human plasmin (120 nM) was mixed separately with 12.5 µl of a test mAb or a control human IgG (all at 300 µg/ml) for 1 h at room temperature. Then, to each reaction mixture was added 50 µl of Spectrozyme-PL (150 µM) in 10 mM HEPES-NaOH and 150 mM NaCl (pH 7.4). The final concentrations of plasmin, mAb, and Spectrozyme-PL were 20 nM, 50 µg/ml, and 100 µM, respectively. Generation of p-nitroaniline was monitored by measuring OD at 405 nm.
In addition, the effects of plasmin-reactive mAb on plasmin activity were determined according to the method described by Van Nostrand et al. (27) with minor modifications. Briefly, fibrin was prepared in microtiter wells by incubating 5.4 µM fibrinogen and 1.5 nM thrombin (Haematologic Technologies) in a final volume of 100 µl of 50 mM Tris-HCl, 150 mM NaCl, and 5 mM CaCl2 (pH 7.5) for 1 h at room temperature.
At the same time, human plasmin (300 nM) and hirudin (6 U/ml; 1 U is defined as the amount of hirudin that will neutralize 1 National Institutes of Health unit of thrombin at 37°C, pH 7.5; Calbiochem, San Diego, CA) were mixed separately with a test mAb or a control human IgG (all at 300 µg/ml) for 1 h at room temperature. Thereafter, 50 µl of each test mixture was added to fibrin wells in duplicate, and plates were incubated with shaking at 37°C in a humidified environment. The extent of fibrinolysis was determined as a reduction in absorbance at 405 nm with time.
Because all six test mAb react with thrombin, and two of six mAb inhibit thrombin activity (16), hirudin (a specific thrombin inhibitor) was added to the fibrinolysis assay to completely stop thrombin activity. This prevented any confusion from possible mAb-mediated reduction of additional fibrin generation. Of note, hirudin had no effect on plasmin activity and did not interfere with binding of mAb to plasmin (data not shown).
Statistical analysis
The mean OD plus 3 SD of the 20 normal controls was used as the cutoff, and the plasma samples with OD values consistently higher than the cutoff in two separate experiments were considered positive. Differences in APS and control groups were analyzed using the unpaired t test. Differences in the test Ab-induced inhibition of plasmin activity were analyzed using paired ANOVA, followed by the Bonferroni multiple comparison test. A value of p <0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To test the hypothesis that there are anti-plasmin Ab in APS patients, we developed an ELISA for IgG anti-plasmin Ab and used the assay to analyze plasma samples from 25 APS patients and 20 normal controls. All samples were analyzed at 1/300 dilutions. The results showed that APS patients have more IgG anti-plasmin Ab than normal controls (p < 0.01, by unpaired t test; Fig. 2A). When the mean + 3 SD of the normal controls was used as the cutoff, IgG anti-plasmin Ab were positive in seven of 25 (28%) APS patients (Fig. 2A).
|
2GPI that bound to 2GPI, which is bound to plasmin to some extent, as plasmin is responsible for the known cleavage of 2GPI. To address this possibility, IgG was purified from two positive plasma samples and analyzed for binding to plasmin. Fig. 2B shows that IgG from both patients bound to plasmin in a concentration-dependent fashion compared with normal human IgG at the same concentrations. Combined, these data demonstrate the presence of IgG anti-plasmin Ab in some APS patients. Identification of monoclonal anti-plasmin Ab and anti-plasminogen Ab
As polyclonal anti-plasmin Ab in APS patients are likely to be heterogeneous (particularly in their functional activity), it would be important to obtain anti-plasmin mAb and use such mAb to study the functional significance of anti-plasmin Ab in APS. Accordingly, we examined our six patient-derived IgG monoclonal anti-thrombin/aCL Ab for their reactivity with human Lys-plasmin. The results showed that all six anti-thrombin/aCL mAb bound to plasmin, whereas two of two non-thrombin-reactive IgG monoclonal aCL did not react with plasmin (Fig. 3A). The plasmin-reactive mAb included CL1, CL15, CL24, IS3, IS4, and IS6. The first three mAb were from a secondary APS patient with primary SLE, and the latter three mAb were from a primary APS patient (17, 18). Subsequently, we analyzed these mAb for their reactivity with human plasminogen. Fig. 3B shows that all six plasmin-reactive mAb also bound to plasminogen, whereas neither of the two non-thrombin-reactive monoclonal aCL reacted with plasminogen.
|
As a first step to assess the clinical significance of the above patient-derived mAb against plasmin and plasminogen, we studied the binding affinity of all six mAb to plasmin and plasminogen by competitive and cross-inhibition. The results showed that plasmin was much more effective than plasminogen at inhibiting all mAb binding to the corresponding Ags. Significantly, plasmin at 1 nM inhibited 70% of binding to plasmin by CL1 and CL24. Based on these inhibition data, the relative Kd values of the plasmin-reactive mAb to plasmin were 3 x 10–10, 7 x 10–10, and 4 x 10–10 M for CL1, CL15, and CL24, respectively (Fig. 4A). IS3 and IS4 bound to plasmin with lower affinity, with relative Kd values of 3 x 10–8 and 9 x 10–9 M, respectively.
|
55% of CL24 binding to plasminogen. Based on these inhibition data, the relative Kd values of these mAb binding to plasminogen were 3 x 10–7 and 7 x 10–7 M for CL24 and CL1, respectively, and 2 x 10–6 M for CL15, IS3, and IS6. Considering that the plasma concentration of plasminogen is 2 µM (15), these six mAb are unlikely to have a meaningful reactivity with plasminogen in plasma. When these six mAb were analyzed by cross-inhibition, plasmin inhibited six mAb from binding to plasminogen, whereas plasminogen (up to 1 µM) failed to inhibit six mAb from binding to plasmin, except for CL1 (Fig. 4, C and D). Combined, these results demonstrate that these six mAb are more specific for plasmin than plasminogen and can interact meaningfully with plasmin even in the presence of plasminogen in plasma.
A few studies reported plasmin digestion of some human IgG (25, 26), suggesting that the observed high affinity of mAb binding to plasmin may be due in part to plasmin digestion of mAb. To address this issue, we first repeated the competitive inhibition assay in the presence of a plasmin inhibitor, aprotinin. The results showed that aprotinin (at equal molar concentrations to plasmin) substantially reduced the binding affinity of mAb to plasmin. For example, the relative Kd value of CL15 to plasmin was 7 x 10–7 M in the presence of equimolar aprotinin (data not shown), compared with 7 x 10–10 M in the absence of aprotinin. The findings suggest that plasmin cleaved IgG, resulting in reduced binding of mAb to plasmin on wells and false inhibition. However, these data may also suggest that aprotinin and the plasmin-reactive mAb bind to the same active site of plasmin, leading to aprotinin interference of mAb binding to plasmin. This contention is supported by the report that human autoantibodies to poly(adenosine diphosphate-ribose) polymerase recognized the catalytic site of the enzyme (28), and the high binding affinity of aprotinin to plasmin (with a Kd of 2.3 x 10–10 M) (29).
Second, we preincubated each mAb with plasmin under the same conditions as in the competitive inhibition, and then distributed the mixtures to wells precoated with affinity-purified goat anti-human Ig. After incubation, bound human IgG was detected with HRP-conjugated goat anti-human IgG (-chain specific). The results showed that plasmin up to 1 µM (that causes 100% inhibition for all six mAb binding to plasmin in solid phase in Fig. 4A) did not cause any detectable reduction in OD of our mAb bound to the capturing Ab (data not shown). These findings suggest that plasmin in the competitive inhibition assay did not cleave mAb to cause to false inhibition.
Third, to further address the possibility that extremely high binding affinity of mAb to plasmin may, at least in part, be due to plasmin digestion of IgG, we repeated the competitive inhibition for all six plasmin-reactive mAb with the TLCK-inactivated plasmin. TLCK forms a chemical bond with His622, which is one of the three conserved catalytic residues of all proteases in the trypsin family. As shown in Fig. 5, the relative Kd values were 5.6 x 10–8 M for CL1, 1 x 10–7 M for CL15, and 1 x 10–6 M for CL24 (Fig. 5). Compared with relative Kd values of mAb determined with plasmin alone in Fig. 4A, these affinities were lowered by 10- to 2500-fold. For example, the relative Kd value for IS3 was reduced from 3 x 10–8 to 2.7 x 10–7 M, whereas the relative Kd value for CL24 was reduced from 4 x 10–10 to 1 x 10–6 M.
|
Assuming that anti-plasminogen or anti-plasmin IgG accounts for 1% of the total IgG, the plasma concentration of IgG anti-plasmin or anti-plasminogen Ab will be 0.7 µM, based on a plasma IgG concentration of 10 mg/ml (equivalent to 67 µM). In this context, the high affinity anti-plasmin Ab (with relative Kd values ranging from 10–8 to 10–10 M) are likely to bind to plasmin to form stable complexes, whereas the low affinity anti-plasminogen Ab (with relative Kd values ranging from 10–6 to 10–7 M) may not be able to bind to plasminogen to form stable complexes. Therefore, we studied only the effects of these mAb on plasmin activity. We first examined the effects of these six mAb on the amidolytic activity of plasmin using a small chromogenic substrate of plasmin, Spectrozyme-PL. None of the six mAb affected the amidolytic activity of plasmin (data not shown). As Spectrozyme-PL is a small molecule, the assay might not reflect the inhibitory effects of some plasmin-reactive mAb on lysis of fibrin by plasmin. Accordingly, we examined the effects of plasmin-reactive mAb on plasmin in a fibrinolysis assay, using freshly generated fibrin (from fibrinogen by thrombin) as the substrate for plasmin.
As shown in Fig. 6A, when 100 nM plasmin (final concentration) was added to fibrin in microtiter wells, fibrin was dissolved over a period of 220 min, resulting in a reduction of OD from 0.33 at 0 min to 0 at 220 min. The lysis curve appears to be biphasic. It may be speculated that the biphasic appearance represents a faster dissolution of longitudinal fibers and the subsequent slower cleavage of shorter fibers. However, the precise meaning of this biphasic appearance is not clear at this point. Regardless, based on this lysis curve, we elected to analyze the effects of mAb on plasmin fibrinolytic activity at 90 min, when 50% of fibrin was dissolved.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
More importantly, CL15 reduced by 18% the fibrinolysis mediated by 100 nM plasmin compared with fibrinolysis in the presence of a control IgG3 (Fig. 6, B and C). Combined, these data show that some IgG anti-plasmin Ab in APS patients (such as CL15 at 100 µg/ml, 1% of the plasma IgG concentration) could interfere with plasmin-mediated fibrinolysis.
The fibrinolytic system comprises the proenzyme, plasminogen, which is converted to plasmin mainly by tissue plasminogen activator (t-PA). Activation of plasminogen by t-PA is not efficient in solution, but becomes 400-fold more efficient when both proteins are bound to fibrin (15). The fibrinolytic system is tightly regulated; t-PA is inhibited by plasminogen activator inhibitors 1 and 2 (PAI-1 and PAI-2), and plasmin by 2-antiplasmin (2-AP). The plasma concentrations of t-PA, PAI-1, plasminogen, and 2-AP are 80 pM (5 ng/ml), 400 pM (20 ng/ml), 2 µM (0.2 mg/ml), and 1 µM, respectively. Thus, most plasmin in the fluid phase is inhibited by 2-AP, whereas plasmin formed on a fibrin surface is protected and can exert its activity for longer.
Studies have suggested that the fibrinolytic system is an important defense mechanism against thrombosis. The reduction of fibrinolysis may allow the growth and development of thrombi, resulting in a prothrombotic state. Decreased levels of t-PA and/or increased levels of PAI-1 have been reported in patients with deep vein thrombosis (30). Along this line, mice lacking plasminogen were predisposed to severe thrombosis (31). In contrast, venous thrombosis occurs frequently in transgenic mice overexpressing the native form of PAI-1 under control of the metallothionein promoter (32), whereas age-dependent spontaneous coronary arterial thrombosis occurs in mice that express a stable form of human PAI-1 (33). In this context, it is conceivable that CL15 may promote thrombosis in the host patient by inhibiting plasmin-mediated fibrinolysis.
This contention is consistent with some recent studies that suggested defective fibrinolysis in APS. In 1999, Ieko et al. (34, 35) reported that 2GPI exerted a dose-dependent enhancement of t-PA activity in the presence of PAI-1, and that addition of EY1C8 patient-derived monoclonal aCL reduced t-PA activity by 45% in a mixture of t-PA (3.6 U/ml), PAI-1 (7.1 ng/ml), and 2GPI (3.8 µM). In 2000, Cugno et al. (36) showed that of 39 patients with primary APS, three had high titers of IgG Ab against t-PA, and four had high titers of Ab against fibrin-bound t-PA, which is the physiologically active form of t-PA. Very recently, Kolev et al. (37) reported that IgG from APS patients impaired the fibrin dissolution with plasmin. Interestingly, these investigators noticed that although normal human IgG appeared to slightly retard the dissolution of fibrin by plasmin, retardation was much stronger with IgG from APS patients. These findings are similar to the effects we observed of normal human IgG and CL15 on plasmin-mediated fibrinolysis (Fig. 6). The slight inhibitory effect of normal human IgG on plasmin-mediated fibrinolysis may be due to some IgG that is bound and cleaved by plasmin (25, 26), resulting in its competition with fibrin for proteolytic activity of plasmin. In addition, IgG from APS patients did not affect the amidolytic activity of plasmin when analyzed with the substrate Spectrozyme-PL (37), which is also similar to our current finding that CL15 failed to inhibit plasmin from digesting Spectrozyme-PL.
Along this line, Takeuchi et al. (38) showed that the EY2C9 IgM anti-2GPI mAb (from an APS patient) suppressed intrinsic fibrinolysis in the presence of 2GPI, apparently by enhancing the 2GPI-mediated weak suppression of intrinsic fibrinolysis. Similar to intrinsic coagulation, intrinsic fibrinolysis is initiated by the contact activation of coagulation factor XII (FXII). Then, activated FXII (FXIIa) converts prekallikrein to kallikrein, which, in turn, converts prourokinase (single-chain, urokinase-type plasminogen activator) to urokinase-type plasminogen activator. In this context, Takeuchi et al. (38) suggested that 2GPI and anti-2GPI may either inhibit FXIIa from activating prekallikrein or suppress kallikrein from activating urokinase-type plasminogen activator, leading to reduced plasmin generation. Taken together with our present findings, these data suggest that there are two kinds of aCL that may inhibit fibrinolysis, one (like EY2C9) suppresses plasmin generation, and one (like CL15) inhibits plasmin activity.
It is noteworthy that plasminogen belongs to the kringle protein family that also includes PT. Each kringle protein consists of a trypsin/thrombin-like serine protease domain and a variable number of loop-like kringle structures. There are extensive sequence homologies among the protease domains and among the kringle domains of different kringle proteins. This understanding led Puurunen et al. (39) to hypothesize that some aPT in patients developing myocardial infarction might cross-react with plasminogen and thus interfere with the fibrinolytic function of plasminogen. These investigators found that soluble PT inhibited up to 50% of aPT binding to immobilized PT in sera from 17 of 17 patients, whereas plasminogen could inhibit aPT binding by at least 20% in nine of 17 (53%) sera (39). These data indicate frequent cross-reactivity of aPT with plasminogen. These findings are now supported further by the present data showing that six of six patient-derived IgG mAb that bind to thrombin and PT also bind to plasmin and plasminogen (Fig. 3, A and B).
What are the likely epitopes on plasmin and thrombin that are recognized by the thrombin- and plasmin-reactive mAb? As all six mAb react with both plasmin and thrombin, it is conceivable that the potential epitopes most likely reside in and/or around the surface homologous regions that are shared by plasmin and thrombin. Fig. 7 shows the homologous structures shared by human plasmin and thrombin. As is evident from their 3D structures in a ribbon model (Fig. 7, A and B), plasmin and thrombin share two -helixes on the left side, four -strands at the bottom, and a homologous active site at the center that consists of the identical catalytic triad residues (H622, D665, and S760 for plasmin; H406, D462, and S568 for thrombin) (21, 22). Moreover, the space-filling models of both molecules show that many amino acid residues in the homologous regions are on the surface (Fig. 6, C and D). In particular, one-half of the C756-P763 region of plasmin and the C564-P571 region of thrombin (in orange) are on the surface. Therefore, the potential epitope on plasmin and thrombin probably resides in and/or around the C756-P763 region of plasmin and the C564-P571 region of thrombin.
|
Our accumulated analyses of seven mAb generated by screening against CL in the presence of bovine serum reveal that five of seven bind to thrombin, PT, plasmin, and plasminogen. Of these five mAb, IS3, IS4, CL1, and CL24 also react with 2GPI (17), the major autoantigen or cofactor for autoantibodies detected by the conventional aCL ELISA. Among the five reactive autoantigens, all five mAb bind to plasmin with the highest affinity, suggesting that plasmin may be the key autoantigen that drives the aPL B cells in APS patients, and that some induced autoantibodies may act directly by binding to plasmin and inhibiting fibrinolysis, leading to a prothrombotic state. Alternatively, some induced Ab may cross-react with other autoantigens in the coagulation cascade, such as thrombin, and promote thrombosis (like CL24 in reducing inactivation of thrombin by AT) (16). In the future it will be important to test this hypothesis by immunizing mice with plasmin and then studying the induced Ab. If immunization with plasmin leads to prothrombotic aPL, it will be interesting to explore the possibility of inhibiting aPL production by tolerizing mice with plasmin with or without modification.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 C.Y. and K.H. contributed equally to this manuscript and are considered to be co-first authors.
3 Current address: Department of Rheumatology, Ren Ji Hospital, Shanghai Second Medical University, 145 Shan Dong Zhong Road, Shanghai 200001, China.
4 Address correspondence and reprint requests to Dr. Kwan-Ki Hwang, Medicine/Rheumatology, University of California, 32-59 Rehabilitation Center, 1000 Veteran Avenue, Los Angeles, CA 90095-1670. E-mail address: hwangkk{at}ucla.edu
5 Abbreviations used in this paper: APS, antiphospholipid syndrome; aCL, anti-cardiolipin Ab; 2-AP, 2-antiplasmin; aPL, anti-phospholipid Ab; aPT, anti-prothrombin Ab; AT, antithrombin; CL, cardiolipin; 3D, three-dimensional; 2GPI, 2-glycoprotein-1; LAC, lupus anticoagulant; PAI, plasminogen activator inhibitor; PC, protein C; PL, phospholipid; PT, prothrombin; SLE, systemic lupus erythematosus; TF, tissue factor; TLCK, N-tosyl-L-lysine chloromethyl ketone-HCl; t-PA, tissue plasminogen activator.
Received for publication January 27, 2003. Accepted for publication February 18, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2-glycoprotein I (apolipoprotein H). Proc. Natl. Acad. Sci. USA 87:4120.2GP1 antibodies from two patients with the anti-phospholipid syndrome reveals three species of antibodies. Br. J. Haematol. 105:102.[Medline]
-strand mimetic inhibitors of thrombin. J. Med. Chem. 42:1376.[Medline]
-protein precursor. J. Biol. Chem. 270:22827.2-glycoprotein I and monoclonal anticardiolipin antibodies on extrinsic fibrinolysis. Semin. Thromb. Hemost. 26:85.[Medline]
-2 glycoprotein I autoantibodies: possible mechanism for thrombosis in patients with antiphospholipid syndrome. Br. J. Haematol. 119:781.[Medline]
This article has been cited by other articles:
|
|
 |
|
  Y.-H. Yang, D. Chien, M. Wu, J. FitzGerald, J. M. Grossman, B. H. Hahn, K.-K. Hwang, and P. P. Chen Novel Autoantibodies against the Activated Coagulation Factor IX (FIXa) in the Antiphospholipid Syndrome That Interpose the FIXa Regulation by Antithrombin J. Immunol., February 1, 2009; 182(3): 1674 - 1680. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Bautz, G. A. Preston, S. Lionaki, P. Hewins, A. S. Wolberg, J. J. Yang, S. L. Hogan, H. Chin, S. Moll, J. C. Jennette, et al. Antibodies with Dual Reactivity to Plasminogen and Complementary PR3 in PR3-ANCA Vasculitis J. Am. Soc. Nephrol., December 1, 2008; 19(12): 2421 - 2429. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R Forastiero and M Martinuzzo Prothrombotic mechanisms based on the impairment of fibrinolysis in the antiphospholipid syndrome Lupus, October 1, 2008; 17(10): 872 - 877. [Abstract] [PDF]
|
|
|
|
|
|
P. Chen, C. Yang, K Ede, C. Wu, J. FitzGerald, and J. Grossman Some antiphospholipid antibodies bind to hemostasis and fibrinolysis proteases and promote thrombosis Lupus, October 1, 2008; 17(10): 916 - 921. [Abstract] [PDF]
|
|
|
|
|
|
X.-X. Chen, Y.-Y. Gu, S.-J. Li, J. Qian, K.-K. Hwang, P. P. Chen, S.-L. Chen, and C.-D. Yang Some Plasmin-Induced Antibodies Bind to Cardiolipin, Display Lupus Anticoagulant Activity and Induce Fetal Loss in Mice J. Immunol., April 15, 2007; 178(8): 5351 - 5356. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Giannakopoulos, F. Passam, S. Rahgozar, and S. A. Krilis Current concepts on the pathogenesis of the antiphospholipid syndrome Blood, January 15, 2007; 109(2): 422 - 430. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Abo and V. A. DeBari Laboratory Evaluation of the Antiphospholipid Syndrome Ann. Clin. Lab. Sci., January 1, 2007; 37(1): 3 - 14. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-H. Yang, K.-K. Hwang, J. FitzGerald, J. M. Grossman, M. Taylor, B. H. Hahn, and P. P. Chen Antibodies against the Activated Coagulation Factor X (FXa) in the Antiphospholipid Syndrome That Interfere with the FXa Inactivation by Antithrombin J. Immunol., December 1, 2006; 177(11): 8219 - 8225. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
0.1. Three regions of the most homologous amino acid sequences are underlined.
