SEARCH

SEARCH BY CITATION

Keywords:

  • platelet activation markers;
  • platelet–leucocyte complexes;
  • antiphospholipid syndrome;
  • systemic lupus erythematosus;
  • rheumatoid arthritis

Abstract

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Discussion
  6. References

It is possible that platelet activation may play a pathogenic role in the increased risk of thrombosis associated with antiphospholipid antibodies (APA). In this study, levels of in vivo platelet activation were measured in 20 patients with primary antiphospholipid syndrome (PAPS) and 30 systemic lupus erythematosus (SLE) patients (14 of whom had secondary APS) using sensitive flow cytometry. Soluble P-selectin levels were also assayed. Platelet CD63 expression was significantly higher in PAPS than normal controls (P = 0·007), as well as SLE patients with and without secondary APS (P = 0·03 and P = 0·002 respectively). PAC-1 binding was significantly higher in PAPS than the control group (P = 0·007) and SLE patients without APS (P = 0·015). Platelet–leucocyte complexes were significantly higher in SLE patients than both PAPS and the control group, and platelet–monocyte complexes were significantly increased in PAPS compared with the control group. (Platelet–leucocyte complexes were also significantly higher than controls in 10 rheumatoid arthritis (RA) patients without APA). Soluble P-selectin levels were significantly higher in PAPS and SLE patients than the control group. Platelet CD62p expression, annexin V binding and platelet microparticle numbers were not increased in PAPS or SLE patients. We conclude that there is evidence of increased platelet activation in PAPS and SLE, and this is important to note as it may have potential therapeutic implications with respect to use of antiplatelet agents in these patients.

The antiphospholipid syndrome (APS) is characterized by the occurrence of thrombosis (arterial or venous), recurrent unexplained fetal loss and/or thrombocytopenia in a patient in whom laboratory tests for antiphospholipid antibody (APA) are positive (Harris, 1987). The presence of APS in patients with no other evidence of auto-immune disease is known as primary APS (PAPS) (Asherson et al, 1989), whereas APS concomitant with other disorders, such as systemic lupus erythematosus (SLE), is called secondary APS. Thrombosis is one of the most common clinical events associated with APA. However, the precise mechanism by which APA may promote thrombosis is unresolved. Many studies highlighting abnormalities of several ‘haemostatic’ factors in APS have been published, and a number of plausible hypotheses reviewed (Triplett, 1993; Santoro, 1994; Esmon et al, 1997; Petri, 1997). One model draws parallels with the thrombosis of heparin-induced thrombocytopenia (HIT) (Arnout, 1996), suggesting that a possible pathogenic interaction between APA and platelets may exist.

Numerous studies investigating platelet activation in APS have been performed, and earlier studies assessed platelet activation by measuring platelet release products such as β-thromboglobulin. The introduction of whole blood flow cytometry has been a major advance as it circumvents many of the problems associated with older methods (Shattil et al, 1987). Platelets can be distinguished easily from other circulating blood cells by the use of platelet-specific antibodies to membrane glycoproteins (GP) such as GPIb or GPIIb–IIIa. Platelet degranulation results in expression on the platelet surface membrane of CD62p (P-selectin) after α-granule release (Stenber et al, 1985; Berman et al, 1986), and CD63, which follows lysosomal and dense granule secretion (Nieuwenhuis et al, 1987; Nishibori et al, 1993). A soluble form of P-selectin can be measured using an immunoassay (Dunlop et al, 1992). The GPIIb–IIIa complex undergoes a conformational change when the platelet is activated, and this can be detected using the IgM monoclonal antibody PAC-1 (Shattil et al, 1985). Platelet-derived ‘microparticles’, which can form as a result of platelet activation, from shear forces or by in vitro freeze–thawing (Owens, 1994), can be enumerated using flow cytometric techniques. Cellular activation, which is accompanied by an increase in intracellular calcium, induces a fast transbilayer movement of all phospholipids known as ‘flip-flop’ (Bevers et al, 1983), which results in phosphatidylserine (PS) exposure on the outer leaflet of the cell. PS can be detected by flow cytometry using the GP annexin V (conjugated to fluorescein). Finally, it is known that thrombin-stimulated platelets bind to leucocytes (particularly monocytes rather than neutrophils (Rinder et al, 1991a,b), hence analysis of platelet–leucocyte complexes provides an additional means of assessing platelet activation. The co-localization of platelets and leucocytes is important in pathophysiological events including inflammation and blood coagulation, and it is well documented that platelet–monocyte interactions can accelerate generation of tissue factor by activated monocytes (Silverstein & Nachman, 1987).

Studies of in vivo platelet activation in APS patients have described increased urinary excretion of thromboxane metabolites (Martinuzzo et al, 1993; Forastiero et al, 1998), accelerated spontaneous platelet aggregation (Wiener et al, 1991), significantly higher levels of platelet CD62p (Fanelli et al, 1997), and a significant increase in the percentage of platelet microparticles (Galli et al, 1993). Previously, we reported a significant increase in platelet CD63 expression in 20 PAPS patients compared with a group of healthy controls (Joseph et al, 1998). In the present study we examined platelet activation in patients with PAPS and SLE using several methods because it is recognized that platelet activation is a complex process and measuring degranulation markers alone may limit the ability to detect platelet activation under all circumstances. This is important to measure as antiplatelet agents form a significant part of the therapeutic regime in APS patients.

Patients and methods

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Discussion
  6. References

Patients attending routine haemostasis or rheumatology clinics were studied, and only patients with stable, ‘chronic’ disease were included.

PAPS patients

Twenty patients (12 women) were studied, and their ages ranged from 23 to 76 years (median 49 years). Sixteen patients had a history of one or more thrombotic events. Of these, 38% were venous thrombosis only, 44% were arterial and 18% both venous and arterial. Seven women had a history of recurrent miscarriage. Significant thrombocytopenia (defined as a platelet count < 100 × 109/l) had been documented in 30% of patients, either at the time of study or before. Eighteen patients were receiving long-term aspirin and/or anticoagulant therapy. Of these patients, 33% were taking aspirin alone, 23% were taking warfarin or Fragmin® and 44% were receiving both therapies. APA testing was performed on the study day, and all patients satisfied the laboratory criteria for APS. Twelve patients had a positive lupus anticoagulant (LA) test, 15 had an elevated anticardiolipin (ACL) IgG level and 13 had an elevated ACL IgM level. Fourteen patients had antibodies to β2GPI.

SLE patients

Thirty women patients were studied, and their ages ranged from 19 to 65 years (median 39 years). Fourteen SLE patients were classified as having secondary APS. A total of nine SLE patients had a history of thrombosis, and eight of these had APA. Nine patients had miscarriages and of these, three had occurred in APS-negative women. Patients with thrombocytopenia as the only manifestation of APS were included in the secondary APS group if they also had positive APA testing. Sixteen patients were receiving long-term aspirin or anticoagulant therapy. The majority of these were in the APS group. A total of 21 patients were receiving immunosuppressive therapy, which usually consisted of prednisolone and/or azathioprine. At the time of the study, 10 patients had a positive LA test, 17 had an elevated ACL IgG level and seven had an elevated ACL IgM level. Antibodies to β2GPI were detected in 12 of the 14 SLE patients with secondary APS.

Rheumatoid arthritis (RA) patients

Ten patients (eight women) with stable RA were also studied as a ‘control’ auto-immune group. Their ages ranged between 56 and 80 years (median 71 years). These patients were negative for LA, aCL and anti β2GPI-antibodies.

Control group

Twenty healthy normal adult subjects (12 women) not taking any medication at the time of the study were used as the control group. Their ages ranged between 21 and 40 years (median 33 years).

Blood collection

Patients and controls were rested for 20 min before venepuncture to minimize artefactual platelet activation. Blood was collected from the antecubital fossa using a 21 G butterfly needle and Vacutainer® system with luer adapter (Becton Dickinson, Meylan, Cedex-France). The first 5 ml of blood collected was drawn directly into a sterile Vacutainer® containing 0·054 ml of EDTA (K3) to determine platelet count; the next 4·5 ml of blood was collected into a sterile Vacutainer® containing 0·5 ml of 0·105 mol/l sodium citrate for platelet activation marker studies. Further tubes of citrated and non-anticoagulated blood were collected for APA measurements.

Flow cytometry

Platelet surface markers CD42b, CD62p and CD63; PAC-1 and annexin V binding The method used to detect CD42b and degranulation markers CD62p and CD63 was based on a whole blood protocol as previously described (Shattil et al, 1987). Briefly, 5 µl of whole blood was incubated with 40 µl of HEPES-buffered saline (HBS) and 5 µl of monoclonal antibody CD42b-PE, CD62p-phycoerythrin (PE) or CD63-fluorescein isothiocyanate (FITC) or corresponding isotypic control (IgG1-PE or IgG1-FITC) (Immunotech, Beckman Coulter, France). PAC-1 binding was detected by incubating blood with PAC-1-FITC (Becton Dickinson, Oxford, UK) or IgM isotype control (Sigma-Aldrich, Dorset, UK), and Dulbecco's phosphate-buffered saline (Life Technologies, Paisley, UK) was substituted as the diluent. After a 20 min incubation at room temperature, the samples were fixed in 0·5 ml of 0·2% formalin saline before flow cytometric analysis. The method used to detect annexin V binding was adapted from recently published studies (Metcalfe et al, 1997; Ruf et al, 1997). Whole blood (5 µl) was diluted initially in 43 µl of either 5 mmol/l K2EDTA-HBS (used as the control) or 2·5 mmol/l CaCl2–HBS, before the addition of 2 µl of Annexin V Fluos (Boehringer Mannheim, East Sussex, UK) to both tubes. After a 20 min incubation at room temperature, the samples were diluted in 0·5 ml of either K2EDTA–HBS or CaCl2–HBS before being analysed using the flow cytometer.

All flow cytometry was performed on a Coulter® Epics® XL-MCL flow cytometer (Beckman Coulter). Platelets were identified by the characteristic ‘platelet cloud’ seen on log-forward versus log-side scatter. Analysis regions were adjusted for each sample to ensure that > 95% of particles within the cloud were positive for anti-CD42b. Antibody binding was expressed as the percentage of platelets staining positive. An analysis marker placed to the right of the negative control fluorescence histogram was set so that 0·5% events were positive, and a total of 10 000 events were analysed.

The normal reference ranges (based on geometric mean ± 2 SD of log-normalized data) were 0·3–3·6% (CD62p), 1·3–8·3% (CD63), 2·1– 6·4% (PAC-1) and 1·3–6·0% (annexin V). Intra-assay coeffiients of variance (CVs) on resting samples were ≤ 8·2%.

Platelet–leucocyte complexes The method used was based on one published previously (Furman et al, 1998). Within 5 min of venepuncture, 25 µl of whole blood was aliquoted into a tube containing 65 µl of HBS and 5 µl of anti-CD45-RPE-Cy5 (Dako, Buckinghamshire, UK), plus 5 µl of either anti-CD42b monoclonal antibody or isotype control (Immunotech, Beckman Coulter). After a 10 min incubation at room temperature, the samples were fixed in 84 µl of fixative (consisting of 0·5 ml of 10% formaldehyde solution, 0·6 ml of Hanks balanced saline solution 10% and 0·9 ml of distilled water). After a further 10 min incubation, 840 µl of distilled water was added to induce erythrocyte lysis. After a further 10 min, samples were ready for flow cytometric analysis, using a protocol in which only CD45-positive events (i.e. leucocytes) were detected. On the initial scatterplot, log fluorescence (FL4–RPE-Cy5) was plotted against side scatter, and a listmode gate used to exclude red cell debris from being analysed. Events in the gated region were analysed further and, on a scatterplot of side scatter versus forward scatter, three distinct populations of leucocytes (i.e. monocytes, granulocytes and lymphocytes) were identified. Platelet-specific events occurring within these three separate regions were measured and counted as platelet–leucocyte complexes. A minimum of 1000 monocyte events were analysed.

The normal reference range for platelet–leucocyte complexes (based on the geometric mean ± 2 SD of log-normalized data) was 2·5–4·8% (platelet–granulocyte complexes), 3·4–9·1% (platelet–monocyte complexes), 1·9–4·6% (platelet–lymphocyte complexes), and 3·0–5·5% (total platelet–leucocyte complexes). The intra assay CVs for measuring platelet–leucocyte complexes in resting samples were 2·1–10·5%.

Platelet-derived microparticles The technique used was adapted from a previously described method, in which 30 µl of platelet-poor plasma was incubated with 5 µl of monoclonal antibody anti-CD41-RPE or isotype control (Dako) for 30 min at room temperature (Combes et al, 1997). Samples were diluted in 0·5 ml of filtered Isoton® II (Beckman Coulter) before being analysed using a flow cytometer. Platelet-derived microparticles were counted using a protocol in which GPIIb–IIIa-positive events of ≤ 0·8-µm-diameter were classified as microparticles. A fixed volume of sample was analysed, and the concentration of microparticles in plasma was then calculated. The normal reference range for number of platelet microparticles (based on the geometric mean ± 2 SD of log-normalized data) was 1·1 × 105−1·2 × 106 microparticles/ml plasma. The intra assay CV for measuring microparticles in resting samples was 2·7%.

Soluble P-selectin

Levels of soluble P-selectin were measured on citrated plasma using a commercial enzyme-linked immunosorbent assay (ELISA) kit (R & D Systems, Abingdon, UK). The kit reference range for levels of soluble P-selectin is 20–44 ng/ml.

Lupus anticoagulant

Citrated blood was centrifuged for 15 min at 2000 g; the plasma was then aspirated into a clean plastic tube and centrifuged again. LA was detected according to standard criteria (British Committee on Standards in Haematology, 1991), using a commercial kit for the dilute Russell's viper venom time (Unicorn Diagnostics Ltd, London, UK), including a platelet neutralization procedure. For patients receiving oral anticoagulants, a Taipan venom time (Diagnostic Reagents Ltd, Thames, UK) was also used (Rooney et al, 1994).

Anticardiolipin antibody

IgG and IgM antibodies to cardiolipin were measured on serum samples by ELISA using microtitre plates (Immunoplate II, Polysorp, Life Technologies) coated with purified cardiolipin (bovine heart, Sigma). Assays were standardized using serum calibrated against appropriate international reference samples (Harris et al, 1987). The normal reference range for both IgG and IgM aCL was < 5 IgG antiphospholipid (GPL) or IgM antiphospholipid (MPL) units respectively.

Anti β2GPI antibody

IgG antibodies to β2GPI were measured by ELISA using a commercial kit (Diastat antiβ2GPI, Shield Diagnostics, Dundee, UK), based on a previously reported method (McNally et al, 1995). The assay was standardized using plasma with a known high concentration of antiβ2GPI. The cut-off point for positivity (> 2·88%) was established as 2 SD above the mean on log-transformed data of 30 control samples.

Statistical analysis

Data was distributed non-parametrically in either one or both of the control and patient groups; hence, non-parametric tests were used. Wilcoxon Mann–Whitney U-testing was performed to determine the significance level for differences between patient and control groups. Correlation testing (using Spearman rank correlation coefficient) was performed to assess the strength of relationships between multiple variables. (Regression analysis was not used as there was no firm belief that one variable was predictive of another). A probability value of < 0·05 was taken to be statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Discussion
  6. References

Expression of platelet degranulation markers

Results for the percentage of platelets expressing CD62p and CD63 are shown in Fig 1(A and B). Median values for CD62p expression were similar among the three groups, although some individual patients had relatively high levels of CD62p expression. Median values for CD63 expression were 3% for the control group, 5·2% for PAPS and 3·1% for SLE patients. There was a significant difference between PAPS and control groups (P = 0·007) but not between SLE and control groups. Median CD63 expression was significantly higher in PAPS patients than both SLE patients with secondary APS (P = 0·03) and SLE patients without APS (P = 0·002). However there was no significant difference in median CD62p or CD63 expression between SLE patients with APS and SLE patients without APS. In both patient groups, there was no correlation between the values obtained for CD62p and CD63 expression, nor was there any correlation between APA titre and expression of platelet degranulation markers.

image

Figure 1.  (A) Percentage of platelets expressing CD62p in the control group, PAPS and SLE patient groups. Horizontal line indicates median; box, the 25th to 75th percentile; and bars, the 5th and 95th percentile. P = ns (all groups). (B) Percentage of platelets expressing CD63 in the control group, PAPS and SLE patient groups. *P = 0·007 (PAPS versus control), †P = 0·002 (PAPS versus SLE), P = ns (SLE versus control). (C) Percentage of platelets binding PAC-1 in the control group, PAPS and SLE patient groups. *P = 0·007 (PAPS versus control), †P = 0·015 (PAPS versus SLE), P = ns (SLE versus control). (D) Percentage of platelets binding annexin V in the control group, PAPS and SLE patient groups. P = ns (all groups).

Download figure to PowerPoint

PAC-1 and annexin V binding to platelets

Results for the percentage of platelets binding PAC-1 and annexin V are shown in Fig 1(C and D). The median values for PAC-1 binding were 3·7% in control subjects, 7·5% in PAPS patients, and 3·6% in SLE patients, and there was a significant difference between PAPS patients and the control group (P = 0·007), as well as PAPS and SLE patients (P = 0·015). The difference between PAPS and SLE patients with secondary APS was not significant for this parameter. There was no significant difference between the median values for annexin V binding between the three groups, although some individual patients had elevated levels. There was no difference in either PAC-1 or annexin V binding between SLE patients with APS and SLE patients without APS, and there was no correlation with APA titre.

Platelet-derived microparticles

There was no significant difference in the median values for microparticles between the three groups (Fig 2), although some SLE patients had elevated numbers. In the PAPS patient group there was a positive correlation between absolute platelet count and numbers of microparticles/ml plasma (rs = 0·69, P = 0·0009).

image

Figure 2.  Numbers of microparticles/ml plasma in control the group, PAPS and SLE patient groups. P = ns (all groups).

Download figure to PowerPoint

Platelet–leucocyte complexes

Results for the percentage of platelet–leucocyte complexes are illustrated in Fig 3(A–C). The median values for platelet–granulocyte complexes were 3·5% in controls, 3·7% in PAPS patients and 4·6% in SLE. These values were significantly higher in SLE patients than both control (P = 0·003) and PAPS groups (P = 0·025). There was no significant difference between SLE patients with APS and SLE patients without APS. Median values for platelet–monocyte complexes were 5·5% in the control group, 7·6% in PAPS patients and 9·2% in SLE patients. There was a significant difference in percentage platelet–monocyte complexes between control and PAPS (P = 0·048) groups, control and SLE (P < 0·0001) groups and PAPS and SLE (P = 0·019) patients. There was no significant difference between SLE patients with APS and SLE patients without APS. The median values for percentage platelet–lymphocyte complexes were 3·0% in the control group, 2·9% in PAPS patients, and 3·7% in SLE patients. The percentage of platelet–lymphocyte complexes was significantly higher in SLE patients than both control (P = 0·006) and PAPS (P = 0·01) groups. Median levels were also significantly higher in SLE patients without APS compared with those with APS (P = 0·003). In PAPS patients, there was a positive correlation between platelet count and percentage platelet–granulocyte complexes (rs = 0·78, P < 0·0001); platelet–monocyte complexes (rs = 0·62, P = 0·004); and platelet–lymphocyte complexes (rs = 0·58, P = 0·007). A positive correlation between platelet–granulocyte and both platelet–monocyte complexes (rs = 0·68, P = 0·001) and platelet–lymphocyte complexes (rs = 0·77, P < 0·0001) was also noted. In SLE patients, there was a positive correlation between platelet count and the percentage of platelet–granulocyte complexes (rs = 0·54, P = 0·002) and platelet–lymphocyte complexes (rs = 0·49, P = 0·006). A positive correlation between platelet–granulocyte and both platelet–monocyte complexes (rs = 0·79, P < 0·0001) and platelet–lymphocyte complexes (rs = 0·79, P < 0·0001) was also noted. There was no significant difference in platelet counts between the control group, PAPS and SLE patients.

image

Figure 3.  (A) Percentage of platelet–granulocyte complexes in the control group, PAPS, SLE and RA patient groups. *P = 0·003 (SLE versus control); †P = 0·025 (SLE versus PAPS); ‡P = 0·001 (RA versus control), P = ns (PAPS versus control). (B) Percentage of platelet–monocyte complexes in the control group, PAPS, SLE and RA patient groups. P = 0·048 (PAPS versus control); †P < 0·0001 (SLE versus control); ‡P = 0·019 (SLE versus PAPS); **P = 0·001 (RA versus control). (C) Percentage of platelet–lymphocyte complexes in the control group, PAPS, SLE and RA patient groups.*P = 0·006 (SLE versus control); †P = 0·012 (SLE versus PAPS); P = ns (PAPS versus control); P = ns (RA versus control).

Download figure to PowerPoint

Because an increase in platelet–leucocyte complexes was found particularly in patients with SLE, this parameter was studied in 10 RA patients (negative for LA, ACL and antiβ2GPI antibodies). The difference in percentage platelet–granulocyte complexes between RA patients and the control group was significant (P = 0·0008), as was the difference in platelet–monocyte complexes (P = 0·002). Although the difference in percentage platelet–lymphocyte complexes failed to reach statistical significance, some RA patients had elevated levels of platelet–lymphocyte complexes. Notably, RA patients had a significantly higher platelet count than the control group (P = 0·0003), PAPS patients (P = 0·012), and SLE patients (P = 0·0002).

Effects of aspirin and warfarin therapy

No significant differences in platelet activation levels were found between patients on aspirin and those not on aspirin. This was also the case when warfarinized patients were compared with non-warfarinized patients.

Platelet activation markers and thrombosis

In both PAPS and SLE patient groups, there was no significant difference in any of the platelet activation markers measured between patients with a history of thrombosis and those without.

Plasma soluble P-selectin

Median levels of plasma soluble P-selectin were significantly higher in PAPS (P = 0·0005) and SLE patients (P = 0·0002) than the control group (Fig 4). Values were 18·8 ng/ml in the control group, 41·4 ng/ml in PAPS patients and 32·1 ng/ml in SLE patients. The difference between soluble P-selectin in PAPS and SLE patients was also significant (P < 0·05). In the PAPS patient group, there were weak positive correlations between levels of soluble P-selectin and both platelet CD63 expression (rs = 0·45, P < 0·05), and percentage platelet–monocyte complexes (rs = 0·48, P = 0·03). However, this was not the case with SLE patients. There was no significant difference in P-selectin levels between SLE patients with APS and those without APS.

image

Figure 4.  Plasma soluble P-selectin levels in the control group, PAPS and SLE patient groups. *P = 0·001 (PAPS versus control); †P = 0·001 (SLE versus control); ‡P = 0·045 (SLE versus PAPS).

Download figure to PowerPoint

Despite the fact that there was no general correlation between all of the platelet activation markers in individual patients, there was one particular patient who had very high levels of multiple markers. This patient had SLE with secondary APS and a history of venous thrombosis, recurrent miscarriage and immune-mediated thrombocytopenia for which a splenectomy had been performed. Levels of CD62p expression, PAC-1 binding, annexin-V binding and percentage platelet–leucocyte complexes (of all three types) were all significantly elevated, and among the highest levels detected in any of the patients studied.

Discussion

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Discussion
  6. References

Platelet activation may play an important role in the thrombosis associated with APS. A recent study found increased CD62p (but not CD63) expression in the platelet-rich plasma of PAPS patients when compared with normal plasma (Fanelli et al, 1997); whereas, previously, we have reported a significant increase in median platelet CD63 expression and plasma soluble P-selectin in 20 PAPS patients (Joseph et al, 1998). Because platelet activation is a complex process, it is likely that measuring degranulation markers alone may limit the ability to detect platelet activation under all circumstances. Therefore, in the current study, several platelet activation-dependent changes were measured in order to study a group of PAPS patients as well as SLE patients with and without secondary APS.

In this investigation, there was no significant difference in median CD62p expression among the three groups studied. However, median CD63 expression was significantly higher in PAPS patients compared with the control group and SLE patient group. There are several possible explanations for these results. First, it has been shown that circulating degranulated platelets rapidly lose their surface P-selectin to the plasma pool (Michelson et al, 1996). Therefore, platelets may circulate in an increased state of activation but express normal levels of CD62p. Second, it is known that platelets expressing higher levels of CD62p bind preferentially to leucocytes (monocytes and neutrophils) (McEver, 1991), and so would have been excluded from this flow cytometric analysis. However, we were unable to conclude that this preferential binding to leucocytes necessarily results in a reduced number of CD62p-positive single platelets. Median CD63 expression was significantly higher in PAPS patients than in SLE patients with secondary APS. One explanation for this result is that the majority of SLE patients were receiving immunosuppressive therapy at the time of the study, which could potentially ameliorate a platelet-activating/degranulating action of APA.

PAC-1 binding was significantly increased in PAPS, although one SLE patient with secondary APS had the highest level measured. In comparison with results for CD63 expression, there was no significant difference in levels of PAC-1 binding between PAPS and SLE patients with secondary APS. This may be because PAC-1 binding is a more ‘sensitive’ index of platelet activation than the expression of degranulation markers (Bihour et al, 1995).

There was no significant difference in median levels of annexin V binding between the control group, PAPS and SLE patients, although some individuals had increased values. This is not unexpected as it is unlikely that significant aminophospholipid exposure occurs under normal resting conditions. In vitro experiments demonstrate that strong agonists such as calcium ionophore A23187 and collagen and thrombin increase the number of annexin V binding sites on platelets (Thiagarajan & Tait, 1990). It is also possible that highly ‘reactive’ PS-expressing platelets are rapidly cleared from the circulation, or that APA may interfere with the binding of annexin V to PS (Rand et al, 1998).

Microparticle numbers were not significantly increased in PAPS or SLE patients, although several patients had elevated levels. Similar to the findings for annexin V binding, it may be that microparticle numbers are not increased in resting conditions, as there is only extensive in vitro formation of platelet microparticles after exposure to calcium ionophore A23187, collagen plus thrombin or complement C5b-9 (Zwaal et al, 1992). Alternatively, microparticles may be cleared rapidly from the circulation. There have been very few reports of microparticle numbers in APS – one group reported a significant increase in mean percentage of microparticles in 11 APA patients (10%) compared with a normal control group (5·5%) (Galli et al, 1993).

Platelet–granulocyte and platelet–lymphocyte complexes were significantly higher in SLE patients compared with PAPS patients and the control group, and platelet–monocyte complexes were significantly elevated in both patient groups. Platelet–leucocyte complexes enable leucocytes to be involved in haemostasis and thrombosis, and platelet–monocyte interactions may accelerate generation of tissue factor by activated monocytes (Silverstein & Nachman, 1987). Previously, one group reported increased platelet–leucocyte aggregates in SLE and APS patients (Specker et al, 1998). Although it is known that activated platelets bind to leucocytes, it is unlikely that this is the only mechanism for their formation. Evidence of platelet activation was found more often in PAPS than SLE patients, yet, generally, SLE patients had higher percentages of circulating platelet–leucocyte complexes. It may also be that platelet–leucocyte complexes are a feature of auto-immune disease (as part of an inflammatory response), or that they are not cleared from the circulation as rapidly because the reticuloendothelial system may be ‘blocked’ as a result of immune complex deposition. For this reason, patients with RA were also studied. Only platelet–leucocyte complexes were analysed in these patients, hence one cannot exclude the presence of platelet degranulation. However, it is unlikely to have been any more significant than that found in SLE patients without secondary APS. The median values of platelet–granulocyte and platelet–monocyte complexes were significantly higher in RA patients than the control group, suggesting that platelet–leucocyte complexes may be increased in auto-immune–inflammatory diseases in general. The role of leucocyte stimulation in the formation of platelet–leucocyte complexes is somewhat unclear. One group have found that selective activation of leucocytes by N-formyl–methionyl–leucyl–phenylalanine (fMLP) resulted in no increase in the percentage of activated leucocyte-resting platelet conjugates (Rinder et al, 1994), whereas another group demonstrated that fMLP alone increases platelet–leucocyte aggregates dose-dependently in unfixed whole blood (Li et al, 1997). In general, there appeared to be a correlation between the platelet count and level of platelet–leucocyte complexes, as well as between the different subtypes of platelet–leucocyte complexes. These variables may be interdependent in a way not fully described by correlations, therefore we cannot make major inferences. However, it is unlikely that an increased platelet count alone is responsible for the increased platelet–leucocyte complexes seen in the patient groups, as there was no significant difference in platelet counts among the control and patient groups.

Plasma levels of soluble P-selectin were significantly elevated in both PAPS and SLE patients compared with the control group, and were significantly higher in PAPS than in SLE patients. Because soluble P-selectin is a marker of platelet (and possibly endothelial cell) activation, this is consistent with the previous findings of increased platelet activation primarily in PAPS.

In summary, the observations made in this study demonstrate an elevated level of platelet activation in PAPS, as seen by increased expression of platelet CD63, PAC-1 binding and platelet–leucocyte complexes. The finding that levels of other platelet activation markers (i.e. annexin V binding and numbers of microparticles) were not significantly elevated, indicates that the type of platelet activation in PAPS is limited, and does not appear to involve increased levels of platelet aminophospholipid exposure (although this cannot be completely excluded). Although we cannot state for certain, it is unlikely that any of these observations occurred as a result of thrombosis because all patients were stable when studied and had not suffered an acute thrombotic event for months or years. Aspirin did not appear to have a significant effect on expression of the platelet activation markers studied, but patient numbers were small. This may be of clinical significance, however, as it may signify that aspirin alone may not necessarily be the ideal antiplatelet drug of choice in APS. Despite having a lesser degree of increased platelet activation, SLE patients had the highest levels of circulating platelet–leucocyte complexes, and this may be related to factors such as leucocyte activation as well as reduced clearance of these complexes from the circulation. An increase in platelet–leucocyte complexes was also found in RA patients without APA. Whether or not the observed increase in platelet activation in this study is caused by APA IgG is still unclear, and remains the topic of further study.

References

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Discussion
  6. References
  • Arnout, J. (1996) The pathogenesis of the antiphospholipid syndrome: a hypothesis based on parallelisms with heparin-induced thrombocytopenia. Thrombosis and Haemostasis, 75, 536541.
  • Asherson, R.A., Khamashta, M., Ordi-Ros, J., Derksen, R.H.W.M., Machin, S.J., Barquinero, J., Outt, H.H., Harris, E.N., Vilardell-Torres, M., Hughes, G.R.V. (1989) The ‘primary’ antiphospholipid syndrome: major clincial and serological features. Medicine (Baltimore), 68, 366374.
  • Berman, C.L., Yeo, E.L., Wencel-Drake, J., Furie, B.C., Ginsberg, M.H., Furie, B. (1986) A platelet alpha-granule protein that is associated with the plasma membrane after activation: characterisation and subcellular localisation of platelet-activation dependent granule-external membrane protein. Journal of Clinical Investigation, 78, 130137.
  • Bevers, E.M., Comfurius, P., Zwaal, R.F.A. (1983) Changes in membrane phospholipid distribution during platelet activation. Biochimistry Biophysica Acta, 736, 5766.
  • Bihour, C., Durrieu-Jais, C., Besse, P., Nurden, P., Nurden, A.T. (1995) Flow cytometry reveals activated GP IIb-IIIa complexes on platelets from patients undergoing thrombolytic therapy after acute myocardial infarction. Blood Coagulation and Fibrinolysis, 6, 395410.
  • British Committee on Standards in Haematology (1991) Guidelines on testing for the lupus anticoagulant. Lupus Anticoagulant Working Party of the BCSH Haemostasis and Thrombosis Task Force. Journal of Clinical Pathology, 44, 885889.
  • Combes, V., Dignat-George, F., Mutin, M., Sampol, J. (1997) A new flow cytometry method of platelet-derived microvesicle quantitation in plasma. Thrombosis and Haemostasis, 77, 220.
  • Dunlop, L.C., Skinner, M.P., Bendall, L.J., Favoloro, E.J., Castaldi, P.A., Gorman, J.J., Gamble, J.R., Vadas, M.A., Berndt, M.C. (1992) Characterization of GMP-140 (P-selectin) as a circulating plasma protein. Journal of Experminetal Medicine, 175, 11471150.
  • Esmon, N.L., Smirnov, M.D., Esmon, C.T. (1997) Thrombogenic mechanisms of antiphospholipid antibodies. Thrombosis and Haemostasis, 78, 7982.
  • Fanelli, A., Bergamini, C., Rapi, S., Caldini, A., Spinelli, A., Buggiani, A., Emmi, L. (1997) Flow cytometric detection of circulating activated platelets in primary antiphospholipid syndrome. Correlation with thrombocytopenia and anticardiolipin antibodies. Lupus, 6, 261267.
  • Forastiero, R., Martinuzzo, M., Carreras, L.O., Maclouf, J. (1998) Anti-beta2 glycoprotein I. antibodies and platelet activation in patients with antiphospholipid antibodies: association with increased excretion of platelet-derived thromboxane urinary metabolites. Thrombosis and Haemostasis, 79, 4245.
  • Furman, M.I., Benoit, S.E., Barnard, M.R., Valeri, C.R., Borbone, M.L., Becker, R.C., Hechtman, H.B., Michelson, A.D. (1998) Increased platelet reactivity and circulating monocyte-platelet aggregates in patients with stable coronary artery disease. Journal of American College of Cardiology, 31, 352358.
  • Galli, M., Grassi, A., Barbui, T. (1993) Platelet-derived microvesicles in patients with antiphospholipid antibodies. Thrombosis and Haemostasis, 69, 541(Abstract).
  • Harris, E.N. (1987) Syndrome of the black swan. British Journal of Rheumatology, 26, 324326.
  • Harris, E.N., Gharavi, A.E., Patel, S.P., Hughes, G.R.V. (1987) Evaluation of the anticardiolipin test: report of an international workshop held 4 April 1986. Clinical and Experimental Immunology, 68, 215222.
  • Joseph, J.E., Donohoe, S., Harrison, P., Mackie, I.J., Machin, S.J. (1998) Platelet activation and turnover in the primary antiphospholipid syndrome. Lupus, 7, 333340.
  • Li, N., Goodall, A.H., Hjemdahl, P. (1997) A sensitive flow cytometric assay for circulating platelet-leucocyte aggregates. British Journal of Haematology, 99, 808816.
  • McEver, R.P. (1991) Leukocyte interactions mediated by selectins. Thrombosis and Haemostasis, 66, 8087.
  • McNally, T., Mackie, I.J., Machin, S.J., Isenberg, D.A. (1995) Increased levels of β2 glycoprotein-I antigen and β2 glycoprotein-I binding antibodies are associated with a history of thromboembolic complications in patients with SLE and primary antiphospholipid syndrome. British Journal of Rheumatology, 34, 10311036.
  • Martinuzzo, M.E., Maclouf, J., Carreras, L.O., Levy-Toledano, S. (1993) Antiphospholipid antibodies enhance thrombin-induced platelet activation and thromboxane formation. Thrombosis and Haemostasis, 70, 667671.
  • Metcalfe, P., Williamson, L.M., Reutelingsperger, C.P., Swann, I., Ouwehand, W.H., Goodall, A.H. (1997) Activation during preparation of therapeutic platelets affects deterioration during storage: a comparative flow cytometric study of different production methods. British Journal of Haematology, 98, 8695.
  • Michelson, A.D., Barnard, M.R., Hechtman, H.B., MacGregor, H., Connolly, R.J., Loscalzo, J., Valeri, C.R. (1996) In vivo tracking of platelets: circulating degranulated platelets rapidly lose surface P-selectin but continue to circulate and function. Proceedings of the National Academy of Sciences of the United States of America., 93, 1187711882.
  • Nieuwenhuis, H.K., Van Oosterhout, J.J., Rozemuller, E., Van Iwaarden, F., Sixma, J.J. (1987) Studies with a monoclonal antibody against activated platelets. evidence that a secreted 53,000-molecular weight lysosome-like granule protein is exposed on the surface of activated platelets in the circulation. Blood, 70, 838845.
  • Nishibori, M., Cham, B., McNicol, A., Shalev, A., Jain, N., Gerrard, J.M. (1993) The protein CD63 is in platelet dense granules, is deficient in a patient with Hermansky–Pudlak syndrome, and appears identical to granulophysin. Journal of Clinical Investigation, 91, 17751782.
  • Owens, M.R. (1994) The role of platelet microparticles in hemostasis. Transfusion Medicine Reviews, 8, 3744.
  • Petri, M. (1997) Pathogenesis and treatment of the antiphospholipid antibody syndrome. Medical Clinics of North America, 81, 151177.
  • Rand, J.H., Wu, X.X., Andree, H.A., Ross, J.B., Rusinova, E., Gascon-Lema, M.G., Calandri, C., Harpel, P.C. (1998) Antiphospholipid antibodies accelerate plasma coagulation by inhibiting annexin-V binding to phospholipids. a ‘lupus procoagulant’ phenomenon. Blood, 92, 16521660.
  • Rinder, H.M., Bonan, J.L., Rinder, C.S., Ault, K.A., Smith, B.R. (1991a) Dynamics of leukocyte-platelet adhesion in whole blood. Blood, 78, 17301737.
  • Rinder, H.M., Bonan, J.L., Rinder, C.S., Ault, K.A., Smith, B.R. (1991b) Activated and unactivated platelet adhesion to monocytes and neutrophils. Blood, 78, 17601769.
  • Rinder, H.M., Tracey, J.L., Rinder, C.S., Leitenberg, D., Smith, B.R. (1994) Neutrophil but not monocyte activation inhibits P-selectin mediated platelet adhesion. Thrombosis and Haemostasis, 71, 750756.
  • Rooney, A.M., McNally, T., Mackie, I.J., Machin, S.J. (1994) The taipan snake venom time: a new test for lupus anticoagulant. Journal of Clinical Pathology, 47, 497501.
  • Ruf, A., Pick, M., Deutsch, V., Patscheke, H., Goldfarb, A., Rachmilewitz, E.A., Guillin, M.C., Eldor, A. (1997) In-vivo platelet activation correlates with red cell anionic phospholipid exposure in patients with beta-thalassaemia major. British Journal of Haematology, 98, 5156.
  • Santoro, S. (1994) Antiphospholipid antibodies and thrombotic predisposition. underlying pathogenetic mechanisms. Blood, 83, 23892391.
  • Shattil, S.J., Hoxie, J.A., Cunningham, M., Brass, L.F. (1985) Changes in the platelet membrane glycoprotein IIb. IIIa complex during platelet activation. Journal of Biological Chemistry, 260, 1110711114.
  • Shattil, S.J., Cunningham, M., Hoxie, J.A. (1987) Detection of activated platelets in whole blood using activation- dependent monoclonal antibodies and flow cytometry. Blood, 70, 307315.
  • Silverstein, R.L. & Nachman, R.L. (1987) Thrombospondin binds to monocytes-macrophages and mediates platelet-monocyte adhesion. Journal of Clinical Investigation, 79, 867874.
  • Specker, C., Perniok, A., Brauckmann, U., Siebler, M., Schneider, M. (1998) Detection of cerebral microemboli in APS – introducing a novel investigation method and implications of analogies with carotid artery disease. Lupus, 7, S75S80.
  • Stenberg, P.E., McEver, R.P., Shuman, M.A., Jacques, Y.V., Bainton, D.F. (1985) A platelet alpha-granule membrane protein (GMP-140) is expressed on the plasma membrane after activation. Journal of Cell Biology, 101, 880886.
  • Thiagarajan, P. & Tait, J.F. (1990) Binding of annexin V/placental anticoagulant protein I to platelets. Evidence for phosphatidylserine exposure in the procoagulant response of activated platelets. Journal of Biological Chemistry, 265, 1742017423.
  • Triplett, D.A. (1993) Antiphospholipid antibodies and thrombosis. A consequence, coincidence, or cause? Archives of Pathology and Laboratory Medicine, 117, 7888.
  • Wiener, H.M., Vardinon, N., Yust, I. (1991) Platelet antibody binding and spontaneous aggregation in 21 lupus anticoagulant patients. Vox Sanguinis, 61, 111121.
  • Zwaal, R.F.A., Comfurius, P., Bevers, E.M. (1992) Platelet procoagulant activity and microvesicle formation. Its putative role in hemostasis and thrombosis. Biochimistry and Biophysical Acta, 1180, 18 .