SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Objective

Microparticles are released from endothelial cells in response to a variety of injurious stimuli and recently have been shown to be increased in a number of diseases associated with endothelial dysfunction. This study examined endothelial microparticle (EMP) and platelet microparticle (PMP) profiles in children with systemic vasculitis to test the hypothesis that EMPs may provide a noninvasive means of examining endothelial activation or injury.

Methods

The study cohort comprised 39 children with systemic vasculitis at various stages of disease activity, 24 control children with febrile disease, and a control group of 43 healthy subjects. Plasma was ultracentrifuged at 17,000g for 60 minutes, and the microparticle pellets were examined using flow cytometry.

Results

Plasma from patients with active systemic vasculitis contained significantly higher numbers of E-selectin–positive EMPs compared with that from patients in remission, healthy controls, or febrile disease controls (P = 0.000 for each). A similar result was obtained for the numbers of EMPs expressing the marker CD105. There was also a significant increase in PMPs expressing CD42a in the active vasculitis group as compared with the other groups, but this difference was not significant for PMPs expressing P-selectin. The EMP counts correlated with the Birmingham Vasculitis Activity Score and the acute-phase reactant levels in the patients with systemic vasculitis, but there was a poor correlation overall between EMP counts and the acute-phase reactant levels in the febrile disease controls.

Conclusion

EMPs may provide a window to the activated endothelium and could provide important pathophysiologic insights into the vascular injury associated with vasculitis of the young.

Vasculitis occurs in many different diseases and syndromes in childhood. It is the predominant manifestation of the disorder in some children, but in others it may be one aspect of a more widespread disease (1). The cause of the majority of the childhood vasculitides is unknown, and it is likely that a complex interaction between inherited determinants and environmental factors, such as infections, trigger the disease. Up-regulated expression of adhesion molecules on endothelial cells and infiltrating inflammatory cells occurs in primary systemic vasculitis, and endothelial dysfunction is likely to be central to the mechanism through which adhesion molecules and cytokines contribute to the pathogenesis of vasculitis (2).

A general feature of all activated cells and cells undergoing apoptosis is a loss of asymmetry of normal cell membrane phospholipids, resulting in an increase in phosphatidylserine on the outer leaflet of the bilipid membrane layer, as well as blebbing of the membrane causing microparticle formation and shedding by a process of exocytic budding (3). Although platelet microparticles (PMPs) have been extensively studied, until recently there has been very little interest in microparticles of endothelial cell origin (EMPs). It is now becoming apparent that EMPs may provide a window to the activated endothelium in a number of disease states in which endothelial injury is central to the disease process, including atherosclerosis (4), acute coronary syndromes (5), antiphospholipid syndrome (6), thrombotic thrombocytopenic purpura (7), and multiple sclerosis (MS) (8).

The functional significance of cellular and platelet phosphatidylserine externalization and microparticle formation has not been fully elucidated, but one important pathophysiologic consequence is a prothrombotic tendency mediated by activation of the extrinsic coagulation pathway (i.e., the tissue factor/factor VII–dependent pathway) (4, 5, 9–11). Recently, it has been suggested that EMPs may be useful diagnostically for the detection of relapses of MS (8), although other investigators have suggested that EMPs may increase in a number of autoimmune states other than MS (12). Moreover, circulating endothelial cells have been found to be increased in small-vessel vasculitis in adults (13) and Kawasaki disease in children (14). Such observations may provide important pathophysiologic insights into the mechanisms involved in vascular injury in these syndromes.

Since endothelial cell activation and injury occur in the vasculitides, the hypothesis that we examined in this study was that circulating EMPs are increased during active vasculitis, providing an opportunity to assess endothelial injury. The aims of this study were, therefore, to examine the profiles of circulating EMPs and PMPs in children with active and inactive vasculitis as compared with healthy control subjects and control patients with childhood septic and autoimmune disease.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Patients.

A total of 39 children with primary systemic vasculitis were studied. The classification of the type of vasculitic illness was defined using the Chapel Hill criteria (15). The assessment of vasculitic disease activity was performed using the Birmingham Vasculitis Activity Score (BVAS) (16). Although the BVAS has never been formally validated in children, the criteria included within this score are entirely relevant to the vasculitic syndromes of childhood, and the BVAS is the most widely used and robust clinical score of vasculitic disease activity available.

Twenty-nine of the children (14 male, 15 female, mean age 7.24 years, range 0.7–15.8 years) had active systemic vasculitis as defined by a BVAS greater than zero (mean BVAS score 10.6 [maximum 63], range 3–24) in the absence of intercurrent infection. Ten of the children had polyarteritis nodosa (PAN); the diagnosis of PAN was based on visceral arteriographic findings demonstrating typical aneurysmal changes affecting predominantly medium-sized arteries in the renal and/or mesenteric vascular beds in all 10 patients, as well as biopsy-proven medium-sized arterial vasculitis in 5 of the patients. None of the patients with PAN was positive for antineutrophil cytoplasmic antibodies (ANCA). Three of the children had microscopic polyangiitis (MPA); all 3 had crescentic nephritis and perinuclear ANCA (pANCA), 1 had biopsy-proven Wegener's granulomatosis (WG) (with cytoplasmic ANCA [cANCA]), and 15 had complete Kawasaki disease as defined by the American Heart Association criteria (17). All patients with Kawasaki disease were febrile (>38°C) and in the second week of illness, 2 had coronary arterial aneurysms, and 1 had myocarditis with supraventricular arrhythmia in the absence of coronary arterial abnormalities. In the Kawasaki disease group, all samples were obtained prior to treatment with intravenous immunoglobulin.

Ten children with inactive vasculitis (4 male, 6 female, mean age 9.0 years, range 1.6–15.8 years) were examined. In this group, 4 had PAN (again with diagnostic visceral arteriographic findings documented at initial presentation), 3 had Kawasaki disease (posttreatment, afebrile, and 4–7 weeks after the initial episode), 1 had WG (cANCA positive), and 2 had MPA (both pANCA positive). All had a BVAS of zero.

Control samples were obtained from 43 healthy subjects, comprising 20 healthy children (10 male, 10 female, mean age 10.3 years, range 2–15.1 years) and 23 healthy young adults (10 male, 13 female, mean age 29.3 years, range 23–40 years).

The disease control group comprised 24 age-matched febrile children (11 male, 13 female, mean age 7.16 years, range 0.6–15.2 years). Twelve of these children were under the care of a general practice physician and had either fever and (presumed) viral rash (n = 9) or upper respiratory tract infection (n = 3), while 8 were hospital inpatients with viral or suspected bacterial sepsis (including 2 patients with suspected meningococcal sepsis, subsequently found not to have invasive bacteremia). The remaining 4 children were hospital outpatients with inactive systemic lupus erythematosus (SLE) and fever secondary to upper respiratory tract infection, but in the absence of clinical vasculitis or antiphospholipid syndrome (including negative anticardiolipin antibodies and absence of lupus anticoagulant).

In addition, samples from 5 patients in the active vasculitis group were examined before and after induction of remission of vasculitis. The diagnoses in these 5 children were PAN (n = 3), WG (n = 1), and MPA (n = 1).

Informed consent was obtained from the parents of all children involved in the study. The study was approved by the local research ethics committee.

Validation of microparticle separation and analysis by flow cytometry.

The method for detecting EMPs was established using tumor necrosis factor α (TNFα)–stimulated monolayers of human umbilical vein endothelial cells (HUVECs) in vitro. HUVECs were prepared as previously described (18). Monolayers of HUVECs were stimulated with 100 ng/ml of TNFα for 4 hours at 37°C. Exact volumes of supernatant (400–800 μl) were then centrifuged at 17,000g for 60 minutes and the supernatant decanted to obtain the microparticle pellet. The microparticles were then reconstituted in 350 μl of annexin V buffer (Bender MedSystems, Milan, Italy), and divided into six 35-μl aliquots plated onto the first 6 wells of a 96-well U-bottomed plate. Following TNFα exposure, there was typically a rise in EMP binding to annexin V in the supernatants, from 0 million/ml to 1–2 million/ml (Brogan PA, et al: unpublished data), a findng consistent with that previously observed in other studies (6). Flow cytometric instrument settings for detecting EMPs in vitro informed the gating and analysis that was subsequently used to determine microparticle populations in whole blood. The preparation for flow cytometry and the gating protocol are described below.

Preparation of platelet-poor plasma.

Whole blood (1.4–5 ml) was collected into bottles containing 3.2% trisodium citrate (Becton Dickinson, Mountain View, CA). Platelet-poor plasma was obtained by immediate centrifugation of the whole blood at 5,000g twice for 5 minutes. Plasma was then stored at −70°C until used.

For isolation of microparticles from platelet-poor plasma, the platelet-poor plasma was defrosted in a water bath at 37°C. Exact volumes of plasma (400–800 μl) were then centrifuged at 17,000g for 60 minutes and the supernatant decanted to obtain the microparticle pellet. The microparticles were then reconstituted in 350 μl of annexin V buffer (Bender MedSystems), and divided into ten 35-μl aliquots plated onto the first 10 wells of a 96-well U-bottomed plate.

Labeling of microparticles with annexin V and monoclonal antibodies.

The labeling and quantification of microparticles from TNFα-stimulated HUVEC supernatants or whole blood was achieved as follows: 5 μl of a 1:10 dilution of fluorescein isothiocyanate (FITC)–conjugated annexin V buffer (Bender MedSystems) was added to every well. In addition, antibodies against platelet or endothelial surface markers conjugated to red (phycoerythrin [PE]) or far-red (peridin chlorophyll protein [PerCP] or Cy-Chrome [CYC]) fluorochromes were used to differentiate microparticles of platelet or endothelial origin. Platelet markers examined were the constitutively expressed platelet marker CD42a (mouse IgG1 anti-human CD42a-PerCP; Becton Dickinson) and the platelet activation marker P-selectin (mouse anti-human CD62P-PE; BD PharMingen, San Diego, CA). Endothelial surface markers examined were E-selectin (mouse IgG1 anti-human CD62E-CYC; BD PharMingen), CD105 (endoglin, or mouse anti-human IgG1 CD105-PE; Serotec, Oxford, UK), intercellular adhesion molecule 1 (ICAM-1) (mouse IgG1 anti-human CD54-CYC; BD PharMingen), and vascular cell adhesion molecule 1 (VCAM-1) (mouse IgG1 anti-human CD106-PE; BD PharMingen).

Ten microliters of each antibody (diluted 1:3 with RPMI medium with 5% fetal calf serum) was added to individual wells. The final dilution of the annexin V–FITC was 1:100, and 1:15 for each of the conjugated antibodies. Microparticles were incubated with the labeled antibodies and annexin V for 10 minutes at room temperature with gentle shaking. The incubation was then terminated by adding 200 μl of annexin V buffer to each well, and the samples were transferred to tubes prior to flow cytometry.

Flow cytometric analysis and determination of absolute microparticle numbers.

All analyses were performed on a FACSCalibur flow cytometer (Becton Dickinson). To obtain optimal forward and side scatter instrument settings for microparticles, 0.8-μm and 3-μm latex beads (Sigma, St. Louis, MO) were run and logarithmic forward and side scatter plots obtained. Gates were then set to include particles of <1.5 μm, but to exclude the first forward scatter channel containing maximal noise. Optimal compensation was set for green, red, and far-red fluorescence. Specific binding for each antibody was determined using isotype control antibodies with equal protein:fluorescein ratios, and at the same final dilution as per the manufacturer's recommendation. Since annexin V is a protein and not an antibody (and thus no isotype control antibody exists), the threshold for annexin V binding was determined by using the fluorescence threshold established for microparticles in the absence of labeled annexin V. Particles smaller than ∼1.5 μm in size and binding to annexin V were then gated, and histograms were obtained for this gated population for binding to individual monoclonal antibodies to determine the cell of origin of the microparticle.

The gate was checked (as in previous studies [6]) by examining EMPs derived from supernatants taken from the monolayers of HUVECs stimulated with 100 ng/ml of TNFα (Sigma) at 24 hours as described above. Microparticle samples were run at medium flow rate with a cut-off time of 1 minute. To convert flow cytometer counts to an estimate of the number of microparticles per ml of plasma, a predetermined number of 3-μm latex beads (Sigma) was run concurrently with the microparticle samples. The absolute number of annexin V–binding microparticles per milliliter of plasma was then determined by using the proportion of beads counted and the exact volume of plasma from which the microparticles were analyzed (19). Since samples from each individual were run 10 times, microparticle counts from individual subjects were expressed as the mean number per milliliter of plasma, with the standard error of the mean based on 10 measurements. A representative set of flow cytometric plots and the gating protocol are shown in Figure 1.

thumbnail image

Figure 1. Representative flow cytometric density plot demonstrating the gating protocol for microparticles. In this example, obtained from a patient with Kawasaki disease, 24% of the microparticle population were E-selectin positive. The gate was established using 0.8-μm latex beads (results not shown) and 3-μm latex beads (results shown). Since forward scatter (FSC) equates with size, particles to the left of the bead population on the X-axis are smaller than 3 μm. The gate was checked by examining microparticles derived from human umbilical vein endothelial cells stimulated with tumor necrosis factor α (see Patients and Methods). An V FITC = annexin V–fluorescein isothiocyanate; CYC = Cy-chrome.

Download figure to PowerPoint

Statistical analysis.

Comparison of mean EMP and PMP numbers between the different patient groups (independent analysis) was analyzed using the Kruskal-Wallis test. Paired data obtained from children before and after induction of remission were compared using the Wilcoxon signed-rank test. Spearman's rank correlation coefficients were calculated to investigate the relationship between microparticle counts, the BVAS, and other conventional acute-phase reactants. All P values were adjusted for multiple comparisons using the Bonferroni technique. Receiver operator characteristic (ROC) curves, sensitivity, specificity, positive and negative predictive values, and likelihood ratios were calculated to examine the diagnostic test characteristics of EMPs for active vasculitis (20).

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Comparison of mean microparticle numbers between patient groups.

The absolute numbers of microparticles of total, platelet, or endothelial origin (expressed in millions per milliliter of plasma) from the different patient groups are summarized in Figures 2–4.

thumbnail image

Figure 2. Column scatter plot of the total microparticle (MP) number obtained from patients with active vasculitis (▪; n = 29), patients with inactive vasculitis (▴; n = 10), healthy controls (▾; n = 43), and disease controls (⧫; n = 24).

Download figure to PowerPoint

thumbnail image

Figure 3. Column scatter plots of the number of CD105 (A) and E-selectin (B) endothelial microparticles (MPs) obtained from patients with active vasculitis (▪; n = 29), patients with inactive vasculitis (▴; n = 10), healthy controls (▾; n = 43), and disease controls (⧫; n = 24).

Download figure to PowerPoint

thumbnail image

Figure 4. Bar chart of the complete microparticle profile from patients with active vasculitis (▪; n = 29), patients with inactive vasculitis (□; n = 10), healthy controls (equation image; n = 43), and disease controls (equation image; n = 24). = significant difference between the active vasculitis group compared with all other groups; = significant difference between the disease control group compared with all other groups. Bars show the mean and SEM. Psel = P-selectin; ICAM-1 = intercellular adhesion molecule 1; VCAM-1 = vascular cell adhesion molecule 1; Esel = E-selectin.

Download figure to PowerPoint

The median total number of microparticles was significantly higher in the active vasculitis group than in the children with inactive vasculitis, healthy controls, and disease controls. The median total numbers were 4.42 million/ml in the active vasculitis group, 1.12 million/ml in the inactive vasculitis group, 1.45 million/ml in the healthy controls, and 0.57 million/ml in the disease controls (Figure 2).

The increased microparticles in children with active vasculitis were derived from both endothelial cells and platelets. The median number of CD105-positive microparticles in the different patient groups was 0.41 million/ml in the active vasculitis group, 0.08 million/ml in the inactive vasculitis group, 0.08 million/ml in the healthy controls, and 0.00 million/ml in the disease controls (Figure 3A).

In addition to the increase seen in CD105-positive microparticles in vasculitis patients, there was also evidence that they could have emanated from activated endothelium. The median number of E-selectin–positive microparticles in the active vasculitis group was significantly higher than in all of the other groups. The median numbers of E-selectin–positive microparticles were 1.15 million/ml in the active vasculitis group, 0.22 million/ml in the inactive vasculitis group, 0.20 million/ml in the healthy controls, and 0.03 million/ml in the disease controls (Figure 3B).

There was also a significant increase in PMPs expressing CD42a in the active vasculitis group as compared with the other groups. The median number of CD42a-positive microparticles in the active vasculitis group was 1.49 million/ml, 0.15 million/ml in the inactive vasculitis group, 0.4 million/ml in the healthy controls, and 0.22 million/ml in the disease controls (Figure 4). These differences were statistically significant (P = 0.006 for the comparison between active vasculitis and inactive vasculitis patients; P = 0.04 between active vasculitis patients and healthy controls; and P = 0.000 between active vasculitis patients and disease controls).

In contrast, there was no significant difference in the number of PMPs expressing P-selectin between the active vasculitis, inactive vasculitis, or healthy control groups (Figure 4). Interestingly, the number of P-selectin microparticles was significantly lower in the septic disease controls than in the other patient populations (true for the comparison between septic controls and active vasculitis patients, inactive vasculitis patients, and healthy controls) (Figure 4).

There was no difference in the numbers of microparticles expressing the adhesion molecules ICAM-1 or VCAM-1 between the groups.

Longitudinal analysis of endothelial and platelet microparticle profiles before and after induction of remission of vasculitis.

Paired samples obtained from 5 children before and after induction of remission of vasculitis were analyzed for changes in EMP and PMP profiles. All 5 patients with active vasculitis demonstrated high levels of EMPs, which fell to normal levels following induction of remission (a 10.5-fold decrease in CD105 microparticles [P = 0.04] and an 8.7-fold decrease in E-selectin microparticles [P = 0.04]). There was also a significant fall in PMPs expressing CD42a after induction of remission (a 3.9-fold decrease [P = 0.04]), but no significant change in microparticles expressing P-selectin. There was no significant difference in microparticles expressing ICAM-1 or VCAM-1 before and after induction of remission of vasculitis. These data are summarized in Figures 5a–c.

thumbnail image

Figure 5. Changes in the endothelial (E-selectin and CD105) (a and b) and platelet (CD42a) (c) microparticle profile before and after induction of remission in 5 children with vasculitis (3 with polyarteritis nodosa, 1 with microscopic polyangiitis, 1 with Wegener's granulomatosis).

Download figure to PowerPoint

Correlation of EMPs with the BVAS and conventional acute-phase reactants.

Correlation coefficients between the positive microparticle counts and the BVAS and conventional acute-phase markers are shown in Table 1. There was a significant correlation between the numbers of total microparticles (r = 0.65, P = 0.000), CD105 (r = 0.51, P = 0.001), and E-selectin (r = 0.65, P = 0.000) in the active and inactive vasculitis patients combined (n = 39 measurements in 34 patients) and the BVAS in the vasculitis patients as a whole. There was a weaker (but significant) correlation between CD42a microparticle counts and the BVAS (r = 0.42, P = 0.01).

Table 1. Correlation between endothelial and platelet microparticle (MP) counts, disease activity scores, and conventional acute-phase reactant markers in the vasculitis and disease control patient groups*
 HgbTotal WCCPlateletsESRCRPAlbuminBVAS
 rPrPrPrPrPrPrP
  • *

    Values for vasculitis patients are the correlation between MP counts, laboratory markers, and the Birmingham Vasculitis Activity Score (BVAS); values for disease control patients are the correlation between MP counts and laboratory markers. Hgb = hemoglobin; WCC = white cell count; ESR = erythrocyte sedimentation rate; CRP = C-reactive protein; NS = not significant; ICAM-1 = intercellular adhesion molecule 1; VCAM-1 = vascular cell adhesion molecule 1; NA = not applicable.

Vasculitis patients              
 Total MP count−0.520.0010.23NS0.410.010.500.0070.25NS−0.540.0090.650.000
 E-selectin MPs−0.500.0050.05NS0.30NS0.400.020.06NS−0.40NS0.650.000
 CD105 MPs−0.400.010.10NS0.30NS0.600.0020.30NS−0.600.0070.510.001
 CD42a MPs−0.20NS−0.03NS0.21NS0.30NS−0.04NS−0.20NS0.420.01
 P-selectin MPs−0.13NS−0.26NS0.21NS−0.06NS−0.23NS−0.14NS−0.26NS
 ICAM-1 MPs0.00NS−0.12NS−0.17NS−0.34NS−0.23NS−0.32NS−0.17NS
 VCAM-1 MPs0.29NS0.06NS−0.22NS−0.05NS−0.12NS0.26NS−0.22NS
Disease control patients              
 Total MP count−0.04NS−0.14NS−0.04NS−0.09NS0.16NS0.24NSNA
 E-selectin MPs−0.08NS0.00NS−0.20NS0.20NS0.30NS0.500.04NA
 CD105 MPs−0.05NS0.04NS−0.08NS−0.03NS0.00NS−0.03NSNA
 CD42a MPs−0.15NS−0.05NS0.08NS0.35NS0.23NS−0.41NSNA
 P-selectin MPs0.12NS0.10NS−0.04NS0.50NS0.19NS−0.680.002NA
 ICAM-1 MPs−0.12NS0.08NS−0.22NS−0.12NS−0.19NS0.063NSNA
 VCAM-1 MPs−0.07NS0.00NS0.06NS−0.03NS0.032NS0.090NSNA

As expected, certain conventional acute-phase laboratory markers correlated with the BVAS (for hemoglobin, r = −0.61, P = 0.000; for total white cell count, r = 0.35, P = 0.038; for erythrocyte sedimentation rate, r = 0.72, P = 0.000; for plasma albumin, r = −0.50, P = 0.02). Similarly, in patients with vasculitis (active or inactive) there was a correlation between E-selectin– and CD105-positive microparticles and other conventional acute-phase reactants (Table 1). In contrast, in the disease control group, the strength of the relationship between microparticle counts (total, endothelial, or platelet) and conventional acute-phase reactants was weaker overall (Table 1), with the notable exception of P-selectin platelet microparticles, which were negatively correlated with plasma albumin levels (r = −0.68, P = 0.002).

Notably, there was no significant correlation between the platelet count and CD42a microparticles or P-selectin microparticles in the vasculitis patients, controls, or disease controls (Table 1).

Test characteristics of EMPs for the diagnosis of active vasculitis.

The ROC curves for EMPs expressing E-selectin or CD105 at varying definitions of positivity are shown in Figure 6. For comparison, the ROC curve for renal angiography in childhood PAN, based on a previously published series of 25 children (21), is also shown. Table 2 summarizes the test characteristics of EMPs for the diagnosis of active vasculitis at varying definitions of test positivity. The cut-off values for test positivity in Table 2 correspond to individual points on the ROC curves.

thumbnail image

Figure 6. Receiver operator characteristic (ROC) curves for E-selectin and CD105 endothelial microparticles (EMPs) for the diagnosis of active vasculitis. = for comparison, the ROC curve for the diagnosis of polyarteritis nodosa in a previously published cohort of children (21).

Download figure to PowerPoint

Table 2. Test characteristics of endothelial microparticles for the diagnosis of active vasculitis*
Cut-off for test positivitySensitivity, % (95% CI)Specificity, % (95% CI)PPV, % (95% CI)NPV, % (95% CI)LR+LR−
  • *

    PPV = positive predictive value; NPV = negative predictive value; 95% CI = 95% confidence interval; LR+ = likelihood ratio for a positive test result; LR− = likelihood ratio for a negative test result.

E-selectin microparticles, millions/ml of plasma      
 0.197 (82–100)43 (32–55)39 (30–48)97 (94–100)1.680.08
 0.390 (73–98)65 (53–75)49 (40–59)94 (90–99)2.560.16
 0.590 (73–98)78 (67–87)60 (51–70)95 (91–99)4.060.13
 0.779 (60–92)92 (84–97)79 (72–87)92 (87–97)10.180.22
 0.966 (46–82)92 (84–97)76 (68–84)88 (81–94)8.410.37
 1.062 (42–79)95 (87–99)82 (74–89)87 (80–93)11.950.4
 1.159 (39–76)95 (87–99)81 (73–88)86 (79–93)11.280.44
 1.245 (26–64)95 (87–99)76 (74–89)82 (75–89)11.360.57
 1.345 (26–64)95 (87–99)76 (68–85)82 (75–89)8.630.58
 1.445 (26–64)96 (89–99)81 (74–89)82 (75–90)11.510.57
 1.541 (24–61)96 (89–99)80 (72–88)81 (74–89)10.60.61
 1.641 (24–61)97 (89–99)86 (79–92)82 (74–89)15.930.6
 1.741 (24–61)97 (89–99)86 (79–92)82 (74–89)15.930.6
 1.838 (21–58)99 (93–100)92 (86–97)81 (73–88)29.20.63
 2.034 (18–54)99 (93–100)91 (85–96)80 (73–88)26.90.66
 2.234 (18–54)100 (95–100)100 (97–100)80 (73–88)Infinity0.65
CD105 microparticles, millions/ml of plasma      
 0.186 (76–91)62 (53–72)46 (31–50)92 (81–98)2.290.22
 0.362 (53–7187 (81–93)62 (55–73)86 (76–93)4.780.44
 0.545 (35–54)94 (89–98)72 (64–81)82 (72–89)6.900.59
 0.731 (22–40)96 (89–98)75 (55–73)79 (69–86)7.970.72
 0.921 (13–28)99 (97–100)86 (79–92)77 (67–85)15.930.80
 1.017 (10–24)99 (91–99)83 (36–99)76 (66–84)13.280.84
 1.117 (10–24)99 (97–100)83 (36–99)76 (66–84)13.280.84
 1.217 (10–24)99 (97–100)83 (36–99)76 (66–84)13.280.84
 1.317 (10–24)99 (97–100)83 (36–99)76 (66–84)13.280.84
 1.417 (10–24)99 (97–100)83 (36–99)76 (66–84)13.280.84
 1.517 (10–24)100 (97–100)100 (48–100)76 (66–84)Infinity0.83
 1.610 (5–16)100 (97–100)100 (48–100)75 (66–83)Infinity0.90
 1.77 (2–12)100 (97–100)100 (48–100)74 (66–82)Infinity0.93
 1.87 (2–12)100 (97–100)100 (48–100)74 (66–82)Infinity0.93
 2.07 (2–12)100 (97–100)100 (48–100)74 (66–82)Infinity0.93
 2.27 (2–12)100 (97–100)100 (48–100)74 (66–82)Infinity0.93

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

This study is the first to show that microparticles are elevated in active childhood systemic vasculitis. This result was independent of immunosuppression per se; the overriding correlation was whether or not disease activity was controlled.

We utilized a number of markers to determine the origin of these microparticles and showed that the endothelium was an important source. CD105 (endoglin) is constitutively expressed by endothelial cells, but is also expressed on some activated monocytes and leukemia cells (22). As such, it is relatively specific for the endothelium. E-selectin is exclusively expressed on activated endothelial cells in vivo (23), although very recently it has been demonstrated to be up-regulated on CD4 T cells in vitro (24). Other potential endothelial markers failed to discriminate between the patient groups. This may be because ICAM-1, constitutively expressed on endothelial cells and up-regulated during endothelial cell activation, has a much wider tissue expression, including expression on activated leukocytes (25). Similarly, although VCAM-1 is predominantly expressed on vascular endothelium, it also has been identified on dendritic cells, macrophages, and non–vascular cell populations within several organs including the joints and kidney (26). Our observation that VCAM-1 microparticles were not increased in active vasculitis (predominantly PAN and Kawasaki disease) is consistent with the findings of Nash et al, who demonstrated that soluble VCAM-1 failed to discriminate patients with active Kawasaki disease from febrile controls (27).

Although our results suggest that these microparticles are of endothelial cell origin, one potential confounder is that soluble E-selectin (increased in many vasculitis syndromes [27]) could bind phosphatidylserine-rich microparticles released from platelets and other cells. We suggest that this is unlikely, however, because although it is known that L-selectin can bind to negatively charged phospholipids such as phosphatidylserine and cardiolipin, E-selectin has been shown unequivocally not to bind to phospholipids (28).

We also observed an increased number of microparticles of platelet origin expressing CD42a in the active vasculitis group, although the difference between the groups was less dramatic than that observed for EMPs. This finding is not entirely surprising, since most vasculitis syndromes are characterized by high platelet counts and by micro- and macroscopic thrombus formation; indeed, antiplatelet therapy is recommended for most forms of childhood vasculitic illness (29). Interestingly, however, there was no correlation between PMPs and the absolute platelet count in any of the patient groups.

There is an emerging concept that endothelial cell dysfunction is central to the pathogenesis of many diseases including atherosclerosis (30, 31), MS (8), and autoimmune diseases that are typified by thrombosis and vascular injury (7), and SLE with antiphospholipid syndrome (6). In support of this concept, previous studies have been able to demonstrate an elevated number of microparticles of endothelial origin in these disorders, as a consequence of increased endothelial cell activation or apoptosis. However, these studies were performed in adult patients and as such are prone to perturbation from environmental factors such as smoking or other comorbid states such as advanced atherosclerosis. It could be argued that the childhood vasculitides would be the “prototypic” disease states in which to study EMPs, since endothelial dysfunction occurs in most vasculitis syndromes and children do not have the confounding factors that could influence endothelial injury and microparticle generation.

We observed high numbers of E-selectin and CD105 EMPs in the patients with active small-vessel vasculitis (MPA and WG), as well as in those with active medium-sized vessel disease (PAN and Kawasaki disease), although patient numbers in the former group were limited. We described patients with active vasculitis together, irrespective of the size of the artery predominantly affected, because there is a significant degree of polyangiitis overlap in children with vasculitis, perhaps even more so than in adults (21, 32, 33). When considering basic mechanisms of endothelial injury, we would be surprised if molecular indices of endothelial injury fell neatly into the relatively arbitrary categories defined by the Chapel Hill consensus or the American College of Rheumatology classification criteria. Indeed, this point is emphasized by the recent reports with regard to circulating endothelial cells that have now been shown to be increased in ANCA-associated vasculitides (13) and in Kawasaki disease (14).

The functional significance of increased microparticle numbers (of all types) in active vasculitis was not investigated in this study, although it is possible that microparticles themselves could contribute to the pathogenesis of vasculitis. The prothrombotic potential of cellular and platelet microparticles is well established and has been demonstrated in acute coronary syndromes, atherosclerosis, antiphospholipid syndrome, and meningococcal sepsis. This occurs mainly as a result of the rich phosphatidylserine content of the microparticles, which acts as one of the essential lipid cofactors for clotting and supports thrombin generation via the tissue factor/factor VII–mediated pathway. Since a prothrombotic tendency occurs in vasculitis (34, 35), it is entirely conceivable that the increased circulating microparticles observed during active vasculitis may contribute to this.

The recent work by Jimenez et al (36) may provide some insights regarding pathophysiology that are relevant to the data presented in our study. By studying the response of renal and brain microvascular endothelial cell lines to activation (TNFα stimulation) or apoptosis (growth factor deprivation), those authors demonstrated that endothelial cells release qualitatively and quantitatively distinct microparticles in activation compared with apoptosis. EMP-expressing inducible markers (predominantly E-selectin) were markedly increased in activation; in contrast, EMP-expressing constitutively expressed markers (CD31 and CD105) or annexin V were more typically found in apoptosis. Our results demonstrated quantitatively greater rises in EMP-expressing E-selectin than EMP-expressing constitutively expressed markers (CD105 and ICAM-1) in children with active vasculitis, suggesting that the endothelium was expressing an activated, rather than apoptotic, phenotype. Activation and apoptosis may not necessarily be mutually exclusive in the vasculitides, however. Indeed, we also observed increased numbers of EMP-expressing annexin V, which may lend credence to the hypothesis that both activation and apoptosis (and, eventually, necrosis as a downstream event) contribute to the vascular injury associated with vasculitis syndromes of the young. In fact, there are surprisingly little published in vivo data examining endothelial apoptosis in primary systemic vasculitides, an area worthy of future study.

EMPs may subsequently prove useful for the diagnosis and monitoring of vasculitic disease activity, and to differentiate sepsis from vasculitis in patients in whom the diagnosis of vasculitis has already been established. Based on our preliminary observations, we summarized the diagnostic test characteristics of EMPs in this context (Table 2 and Figure 6). The use of EMPs as a diagnostic tool in routine clinical practice is not at present warranted. However, in the absence of reliable tools for diagnosing active vasculitis, this area of investigation could prove useful in the future.

In conclusion, we have demonstrated increased levels of circulating endothelial microparticles in children with active vasculitis. We propose that this novel finding may provide a window to the activated endothelium. Further studies are required to determine the full functional and pathophysiologic significance of this observation.

Acknowledgements

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

We acknowledge the London Kawasaki Disease Research Group for help in recruiting children with Kawasaki disease. We also thank Ms Gerry Jewell and Ms Carol Hutchinson for help in obtaining and processing blood samples from disease control and healthy control children.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  • 1
    Brogan PA, Dillon MJ. Vasculitis from the pediatric perspective. Curr Rheumatol Rep 2000; 2: 4116.
  • 2
    Sundy JS, Haynes BF. Cytokines and adhesion molecules in the pathogenesis of vasculitis. Curr Rheumatol Rep 2000; 2: 40210.
  • 3
    Zwaal RF, Schroit AJ. Pathophysiologic implications of membrane phospholipid asymmetry in blood cells. Blood 1997; 89: 112132.
  • 4
    Mallat Z, Hugel B, Ohan J, Leseche G, Freyssinet JM, Tedgui A. Shed membrane microparticles with procoagulant potential in human atherosclerotic plaques: a role for apoptosis in plaque thrombogenicity. Circulation 1999; 99: 34853.
  • 5
    Mallat Z, Benamer H, Hugel B, Benessiano J, Steg PG, Freyssinet JM, et al. Elevated levels of shed membrane microparticles with procoagulant potential in the peripheral circulating blood of patients with acute coronary syndromes. Circulation 2000; 101: 8413.
  • 6
    Combes V, Simon AC, Grau GE, Arnoux D, Camoin L, Sabatier F, et al. In vitro generation of endothelial microparticles and possible prothrombotic activity in patients with lupus anticoagulant. J Clin Invest 1999; 104: 93102.
  • 7
    Jimenez JJ, Jy W, Mauro LM, Horstman LL, Ahn YS. Elevated endothelial microparticles in thrombotic thrombocytopenic purpura: findings from brain and renal microvascular cell culture and patients with active disease. Br J Haematol 2001; 112: 8190.
  • 8
    Minagar A, Jy W, Jimenez JJ, Sheremata WA, Mauro LM, Mao WW, et al. Elevated plasma endothelial microparticles in multiple sclerosis. Neurology 2001; 56: 131924.
  • 9
    Berckmans RJ, Neiuwland R, Boing AN, Romijn FP, Hack CE, Sturk A. Cell-derived microparticles circulate in healthy humans and support low grade thrombin generation. Thromb Haemost 2001; 85: 63946.
  • 10
    Joop K, Berckmans RJ, Nieuwland R, Berkhout J, Romijn FP, Hack CE, et al. Microparticles from patients with multiple organ dysfunction syndrome and sepsis support coagulation through multiple mechanisms. Thromb Haemost 2001; 85: 81020.
  • 11
    Nieuwland R, Berckmans RJ, McGregor S, Boing AN, Romijn FP, Westendorp RG, et al. Cellular origin and procoagulant properties of microparticles in meningococcal sepsis. Blood 2000; 95: 9305.
  • 12
    Larkin M. Raised endothelial microparticles: an early marker for multiple sclerosis? Lancet 2001; 357: 1679.
  • 13
    Woywodt A, Streiber F, deGroot K, Regelsberger H, Haller H, Haubitz M. Circulating endothelial cells as markers for ANCA-associated small-vessel vasculitis. Lancet 2003; 361: 20610.
  • 14
    Nakatani K, Takeshita S, Tsujimoto H, Kawamura Y, Tokutomi T, Sekine I. Circulating endothelial cells in Kawasaki disease. Clin Exp Immunol 2003; 131: 53640.
  • 15
    Jennette JC, Falk RJ, Andrassy K, Bacon PA, Churg J, Gross WL, et al. Nomenclature of systemic vasculitides: proposal of an international consensus conference. Arthritis Rheum 1994; 37: 18792.
  • 16
    Luqmani RA, Bacon PA, Moots RJ, Janssen BA, Pall A, Emery P, et al. Birmingham Vasculitis Activity Score in systemic necrotizing vasculitis. QJM 1994; 87: 6718.
  • 17
    Dajani AS, Taubert KA, Gerber MA, Shulman ST, Ferrieri P, Freed M, et al. Diagnosis and therapy of Kawasaki disease in children. Circulation 1993; 87: 177680.
  • 18
    Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical veins: identification by morphologic and immunologic criteria. J Clin Invest 1973; 52: 274556.
  • 19
    Combes V, Dignat-George F, Mutin M, Sampol J. A new flow cytometry method of platelet-derived microvesicle quantitation in plasma. Thromb Haemost 1997; 77: 220.
  • 20
    Gilbert R, Logan S. Assessing diagnostic and screening tests. In: MoyerVA, ElliotEJ, DavisRL, GilbertR, KlassenT, LoganS, et al, editors. Evidence based pediatrics and child health. London: BMJ Books; 2000. p. 2436.
  • 21
    Brogan PA, Davies R, Gordon I, Dillon MJ. Renal angiography in children with polyarteritis nodosa. Pediatr Nephrol 2002; 17: 27783.
  • 22
    Barclay AN, Brown MH, Law SKA, McKnight AJ, Tomlinson MG, van der Merwe PA. CD105. In: BarclayAN, BrownMH, LawSKA, McKnightAJ, TomlinsonMG, van der MerwePA, editors. The leucocyte antigen facts book. San Diego: Academic Press, Harcourt Brace and Company; 1997. p. 3845.
  • 23
    Barclay AN, Brown MH, Law SKA, McKnight AJ, Tomlinson MG, van der Merwe PA. CD62E. In: BarclayAN, BrownMH, LawSKA, McKnightAJ, TomlinsonMG, van der MerwePA, editors. The leucocyte antigen facts book. San Diego: Academic Press, Harcourt Brace and Company; 1997. p. 2957.
  • 24
    Harashima S, Horiuchi T, Hatta N, Morita C, Higuchi M, Sawabe T, et al. Outside-to-inside signal through the membrane TNF-alpha induces E-selectin (CD62E) expression on activated human CD4+ T cells. J Immunol 2001; 166: 1306.
  • 25
    Barclay AN, Brown MH, Law SKA, McKnight AJ, Tomlinson MG, van der Merwe PA. CD54. In: BarclayAN, BrownMH, LawSKA, McKnightAJ, TomlinsonMG, van der MerwePA, editors. The leucocyte antigen facts book. San Diego: Academic Press, Harcourt Brace and Company; 1997. p. 2789.
  • 26
    Barclay AN, Brown MH, Law SKA, McKnight AJ, Tomlinson MG, van der Merwe PA. CD106. In: BarclayAN, BrownMH, LawSKA, McKnightAJ, TomlinsonMG, van der MerwePA, editors. The leucocyte antigen facts book. San Diego: Academic Press, Harcourt Brace and Company; 1997. p. 3867.
  • 27
    Nash MC, Shah V, Dillon MJ. Soluble cell adhesion molecules and von Willebrand factor in children with Kawasaki disease. Clin Exp Immunol 1995; 101: 137.
  • 28
    Malhotra R, Taylor NR, Bird MI. Anionic phospholipids bind to L-selectin (but not E-selectin) at a site distinct from the carbohydrate-binding site. Biochem J 1996; 314: 297303.
  • 29
    Brogan PA, Dillon MJ. The use of immunosuppressive and cytotoxic drugs in non-malignant disease. Arch Dis Child 2000; 83: 25964.
  • 30
    Mallat Z, Tedgui A. Current perspective on the role of apoptosis in atherothrombotic disease. Circ Res 2001; 88: 9981003.
  • 31
    Tedgui A, Mallat Z. Apoptosis as a determinant of atherothrombosis. Thromb Haemost 2001; 86: 4206.
  • 32
    Brogan PA, Shah V, Bagga A, Klein N, Dillon MJ. T cell Vβ repertoires in childhood vasculitides. Clin Exp Immunol 2003; 131: 51727.
  • 33
    Leavitt RY, Fauci AS. Polyangiitis overlap syndrome. Am J Med 1986; 81: 7985.
  • 34
    Kiraz S, Ertenli I, Ozturk MA, Haznedaroglu IC, Celik I, Calguneri M. Pathological haemostasis and “prothrombotic state” in Behcet's disease. Thromb Res 2002; 105: 12533.
  • 35
    Savage CO, Cooke SP. The role of the endothelium in systemic vasculitis. J Autoimmun 1993; 6: 23749.
  • 36
    Jimenez JJ, Wenche J, Mauro LM, Soderland C, Horstman LL, Ahn Y. Endothelial cells release phenotypically and quantitatively distinct microparticles in activation and apoptosis. Thromb Res 2003; 109: 17580.