SEARCH

SEARCH BY CITATION

Abstract

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

Objective

Complement-activation product C4d is deposited on normal erythrocytes, while abnormal levels have been observed on the surface of erythrocytes of patients with systemic lupus erythematosus (SLE). This study examines whether C4d also deposits on human platelet surfaces, and whether platelet-bound C4d may provide a biomarker for SLE.

Methods

We conducted a cross-sectional study of 105 patients with SLE, 115 patients with other diseases, and 100 healthy controls. Levels of C4d on the surface of platelets were examined by flow cytometry and scanning confocal microscopy. Statistical analyses were performed to determine the clinical variables associated with platelet C4d.

Results

Abnormal levels of platelet C4d were found to be highly specific for SLE. Platelet C4d was detected in 18% of patients with SLE, being 100% specific for a diagnosis of SLE compared with healthy controls and 98% specific for SLE compared with patients with other diseases (P < 0.0001). In addition, platelet C4d was significantly associated with positivity for lupus anticoagulant (P < 0.0001) and anticardiolipin antibodies of the IgG (P = 0.035) or the IgM (P = 0.016) isotype. Platelet C4d was also significantly associated with SLE disease activity according to the SLE Disease Activity Index (P = 0.039), low serum C4 (P = 0.046), an elevated erythrocyte sedimentation rate (P = 0.006), and abnormal levels of C4d on erythrocytes (P < 0.0001).

Conclusion

This observation suggests that platelet-bound C4d may be a useful biomarker for SLE and may be a clue to the pathogenic mechanisms responsible for the myriad thrombotic and vascular complications of lupus associated with antiphospholipid antibodies.

The pathogenesis of fetal loss and thrombosis in patients with systemic lupus erythematosus (SLE) associated with antiphospholipid antibodies (aPL) is not fully understood, but in vivo and in vitro studies have demonstrated that aPL activate endothelial cells and platelets and induce thrombosis and tissue injury (1–3). These antibodies can bind directly to platelet surfaces (4–7). Recent compelling data presented by Holers et al show that in vivo complement activation is required for aPL-induced fetal loss and growth retardation (8). Further evidence of the importance of complement activation in the setting of aPL-associated thrombosis comes from prior studies that have shown an increase in complement-activation products in the serum of patients with aPL who develop stroke and transient ischemic attack (9).

We recently identified abnormal levels of the complement-activation product C4d on erythrocytes from patients with SLE (10). Erythrocyte C4d appears to be a useful lupus biomarker for diagnosis and monitoring of disease activity over time. In this study we examined whether C4d also deposits on platelet surfaces, and whether platelet-bound C4d may be associated with the presence of aPL. Our observation that C4d deposition occurs on platelets in association with the presence of aPL in patients with SLE may provide the first clue to a potential mechanism for the interaction between platelets, complement activation, and aPL in vascular thrombosis.

PATIENTS AND METHODS

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

Study participants and blood specimens.

Patients with SLE (n = 105) were recruited from our ongoing study, Heart Effects and Risk of Thrombosis in SLE (HEARTS). The HEARTS study, involving women with SLE having no history of a cardiovascular event, is designed to examine the risk factors present at baseline that may be predictive of future thrombotic cardiovascular events. Along with demographic variables and cardiovascular risk factors, measurements of aPL, platelet C4d, and erythrocyte C4d were obtained in all participants. The demographic and clinical characteristics of the 105 patients with SLE are listed in Table 1. Clinical manifestations of SLE were defined by the American College of Rheumatology revised classification criteria (11).

Table 1. Demographic and clinical characteristics of the patients with systemic lupus erythematosus (n = 105)*
  • *

    SLAM = Systemic Lupus Activity Measure; SELENA-SLEDAI = Safety of Estrogens in Lupus Erythematosus National Assessment version of the Systemic Lupus Erythematosus Disease Activity Index.

  • Antiphospholipid antibodies were defined by the presence of anticardiolipin (IgG or IgM) or lupus anticoagulant (by partial thromboplastin time with mix, or by Russell's viper venom time) or antibodies to β2-glycoprotein I.

Age, mean ± SD (range) years49.52 ± 10.10 (29–90)
Race, % white83
Sex, % female100
SLE disease duration, mean ± SD (range) years16.53 ± 7.43 (5.00–47.00)
Malar rash, %49
Discoid rash, %5
Photosensitivity, %70
Oral ulcers, %59
Arthritis, %91
Serositis, %53
Renal disease, %27
Neurologic disease, %9
Hematologic manifestations, %60
 Leukopenia22
 Thrombocytopenia0
Antinuclear antibodies, %95
Immunologic disorders, %66
 Antiphospholipid antibodies50 (45 of 90 tested)
 Anticardiolipin antibodies39 (34 of 88 tested)
 Lupus anticoagulant25 (26 of 104 tested)
 Anti–β2-glycoprotein I antibodies33 (32 of 98 tested)
Currently receiving antiplatelet therapy, %11
SLAM, mean ± SD score4.27 ± 2.68
SELENA-SLEDAI, mean ± SD score1.85 ± 2.43
Platelet count at time of visit (×1,000), mean ± SD248.16 ± 71.79
Serum C3, % below normal range32.38
Serum C4, % below normal range52.38
Erythrocyte sedimentation rate, mean ± SD mm/hour (n = 104)14.38 ± 13.64

To examine the specificity of platelet C4d for SLE, 115 subjects with other rheumatic inflammatory/autoimmune or hematologic diseases were recruited for this study; this group comprised patients with systemic scleroderma (n = 19), myositis (n = 21), Sjögren's syndrome (n = 5), rheumatoid arthritis (n = 17), Wegener's granulomatosis (n = 1), hepatitis C (n = 23), urticarial vasculitis (n = 1), sickle cell anemia (n = 8), hematologic malignancies (n = 8), primary Raynaud's phenomenon (n = 5), osteoarthritis (n = 2), hemophilia (n = 3), and psoriatic arthritis (n = 2). The mean ± SD age of these subjects was 50 ± 15 years (range 7–79 years). Eighty-five percent of these subjects were white, and 65% were women. In addition, 100 healthy control subjects, with a mean ± SD age of 40 ± 13 years (range 18–67 years), were recruited for the study, of whom 83% were white and 85% were women. The Institutional Review Board of the University of Pittsburgh provided approval for the study, and all participants provided their written informed consent.

At each study visit, a clinical history was obtained, a physical examination was performed, and routine laboratory tests (e.g., determination of the erythrocyte sedimentation rate [ESR], complete blood cell count, and levels of serum C3, C4, and aPL) were done. Concurrently, an aliquot of blood was collected into 4-cc Vacutainer tubes containing 7.2 mg EDTA as an anticoagulant (Becton Dickinson, Franklin Lakes, NJ), and processed within 2 hours.

Antibodies, flow cytometry, and scanning confocal microscopy.

Whole blood was diluted in phosphate buffered saline and labeled for immunofluorescence for flow cytometry, using phycoerythrin-conjugated anti-CD42b (BD Biosciences, San Jose, CA) and a monoclonal antibody (mAb) conjugated to Alexa Fluor 488 (Molecular Probes, Eugene, OR) or using Alexa Fluor 488 with a Zenon Mouse IgG labeling kit (Molecular Probes). The mAb were anti-C4d (reactive with C4d-containing fragments of all major allotypes of C4) (Quidel, San Diego, CA) or the IgG1-isotype control, MOPC21. Samples were analyzed on a FACSCalibur flow cytometer (BD Immunocytometry Systems). Platelets were electronically gated by forward scatter properties and by expression of CD42b as a platelet-specific marker.

To ensure the day-to-day reliability of platelet C4d measurement, the FACSCalibur flow cytometer was calibrated daily using CaliBRITE 3 beads and FACSComp software (BD Immunocytometry Systems). The instrument settings were also calibrated daily using blood samples stained with isotype control MOPC21 to ensure that the background fluorescence intensity was less than or equal to 5.0. Platelet C4d–specific fluorescence was determined by subtracting the median fluorescence intensity (MFI) of the MOPC21 isotype control from the MFI of platelets stained with anti-C4d. After preliminary experiments in which repeated measures of platelet samples derived from healthy controls were performed, a cutoff value of 2.15 was empirically determined for this assay. This cutoff took into account slight variations in fluorescence labeling between the MOPC21 isotype control and the anti-C4d antibodies, as well as the detection limitations of the flow cytometer. Platelet C4d–specific fluorescence intensity values of greater than or equal to 2.15 were considered to be positive for complement deposition.

For confocal microscopy studies, platelets were separated from whole blood by centrifugation and incubated with anti-CD42b (BD Biosciences) labeled with Alexa Fluor 647 using a Zenon Mouse IgG labeling kit (Molecular Probes) and either MOPC21 or anti-C4d labeled with Alexa Fluor 488 (Molecular Probes). Cells were scanned using an Olympus Fluoview 500 confocal microscope.

Statistical analysis.

Univariate analysis was performed to determine the clinical variables (Table 1) that might be associated with platelet C4d. Chi-square, Fisher's exact, or Wilcoxon rank sum tests were used to determine P values. Logistic regression was used in multivariate analysis to determine the clinical variables that were independent predictors of a positive outcome for detection of platelet C4d.

RESULTS

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

Specificity of platelet C4d for SLE and association with aPL.

Blood specimens were obtained from all participants, and platelet C4d was assessed by 2-color flow cytometry. Platelets were identified and electronically gated by expression of the platelet-specific marker CD42b and by forward scatter properties, which reflect cell size. Initial results showed a wide range of platelet C4d in a significant number of patients with SLE, whereas healthy controls were consistently negative for platelet C4d. C4d-specific fluorescence for the gated platelet population from 6 healthy controls and from 6 patients with SLE is shown in Figure 1A. Platelets from all healthy controls were negative for platelet C4d, whereas those obtained from each of the 6 patients with SLE had remarkably high fluorescence intensity for platelet C4d.

thumbnail image

Figure 1. Deposition of C4d fragments on platelets from patients with systemic lupus erythematosus (SLE). A, Flow cytometry was performed on whole blood from 6 healthy controls and 6 patients with SLE. To identify platelets, specimens were stained with phycoerythrin-conjugated anti-CD42b monoclonal antibodies (mAb) and Alexa Fluor 488–labeled anti-C4d mAb (solid line), or an IgG1 isotype–matched control, MOPC21 (gray shaded area), and analyzed using a FACSCalibur system. Platelet C4d (P-C4d)–specific fluorescence was defined as the C4d-specific median fluorescence intensity (MFI) minus the MFI of the isotype control. B, To analyze the deposition of P-C4d on platelets in the blood specimens from patients with SLE in comparison with patients with other diseases and healthy controls, P-C4d fluorescence of ≥2.15 was considered to be positive for complement deposition, as indicated by the cutoff (solid line). Each patient was tested once.

Download figure to PowerPoint

Detection of platelet C4d in 105 patients with SLE, 115 patients with other diseases, and 100 healthy controls is shown in Figure 1B. The platelet C4d–positive phenotype was, with rare exceptions (in a patient with scleroderma and a patient with rheumatoid arthritis), specific for SLE. When patients with SLE were compared with healthy controls, positivity for platelet C4d was 100% specific for SLE, and when patients with SLE were compared with patients with other diseases, positivity for platelet C4d was 98% specific for SLE (P < 0.0001).

In this cross-sectional analysis, platelet C4d was detected in 18% of patients with SLE. We performed a univariate analysis to determine the clinical variables (as listed in Table 1) that might be associated (by chi-square, Fisher's exact, or Wilcoxon's rank sum test) with expression of C4d on platelets in these patients. Platelet C4d was significantly associated with a positive test result for lupus anticoagulant (LAC) (P < 0.0001) and anticardiolipin antibodies (aCL) of the IgG (P = 0.035) or the IgM (P = 0.016) isotype. Platelet C4d was also significantly associated with SLE disease activity as measured by the SLE Disease Activity Index (SLEDAI) (12) (P = 0.039), low serum C4 (P = 0.046), an elevated ESR (P = 0.006), and abnormal levels of C4d on the surface of erythrocytes (P < 0.0001). These results indicate a correlation between abnormal platelet C4d levels with increased disease activity in patients with SLE.

Logistic regression was used in a multivariate analysis to determine the clinical variables that were independent predictors of a positive outcome for platelet C4d detection. Independent predictors of the presence of platelet C4d included positivity for LAC (odds ratio [OR] 4.20, 95% confidence interval [95% CI] 1.32–13.35) and positivity for erythrocyte C4d (OR 1.07, 95% CI 1.02–1.12). These data suggest a strong and independent association between platelet C4d and the presence of aPL, particularly with LAC.

Deposition of platelet C4d on the surface of the entire platelet population.

Flow cytometric data suggest that when C4d is deposited on platelets, it is deposited on the entire population rather than on a subset of cells, as indicated by the shift in fluorescence intensity that was observed in the entire platelet population (Figure 1A). This finding was true in all patients studied, and indicates that all circulating platelets, including those that are newly synthesized, are C4d positive. This observation suggests that platelet C4d is not generated by periodic deposition during disease flares.

To further investigate the distribution of C4d on platelets, a scanning confocal microscopy study of C4d-positive platelets (MFI of 26.3) from a patient with SLE was performed (Figures 2A–D). In addition to confirming the presence of C4d on all platelets examined, these results also revealed a homogeneous pattern of C4d deposition on the platelet membrane, similar to the distribution of the platelet antigen CD42b (Figure 2C). Presence of platelet C4d on the entire platelet population in certain patients suggests that complement split products generated during complement activation may bind indifferently to multiple moieties on the surface of platelets, or that a diffuse change in platelet membranes may lead to C4d deposition.

thumbnail image

Figure 2. Homogeneous membrane deposition of platelet C4d on confocal images of platelets from a patient with systemic lupus erythematosus. Panels demonstrate green staining for C4d (A,top) and red staining for the platelet marker CD42b (B,top) and their respective MOPC21 controls (A and B,bottom). Images from A and B are shown together in C, and areas where immunofluorescent labels colocalize appear yellow. Differential interference contrast image of cells shown in AC(D). Cells were scanned using an Olympus Fluoview 500 confocal microscope through the midplane of the cells. Images shown in AD are a composite of 3 (top) and 2 (bottom) representative fields.

Download figure to PowerPoint

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 demonstrate that C4d is deposited on human platelets, and that this deposition is highly specific for SLE and correlates with disease activity as measured on the SLEDAI. Moreover, platelet C4d is significantly associated with the presence of aPL. None of our patients had significant thrombocytopenia (<100,000), suggesting that platelet C4d is not associated with immune-mediated thrombocytopenic purpura and antiplatelet antibodies. Furthermore, platelet surfaces were consistently negative for deposited IgM and IgG (results not shown), suggesting that immune complexes and antiplatelet antibodies are not responsible for the production of platelet C4d.

It is reasonable to speculate that deposition of C4d on platelet surfaces may influence platelet aggregation and/or platelet interactions with monocytes and endothelial cells. Based on evidence from previous investigations of the critical role of complement activation in aPL-induced fetal loss (8, 13), we believe that platelet C4d may serve to identify the subset of patients with aPL who are at highest risk of a thrombotic event. Indeed, although 18% of our general cohort of patients with SLE had C4d-positive platelets, a significantly higher prevalence (14 [31%] of 45 patients) and lower prevalence (5 [11%] of 45 patients) of platelet C4d was observed in patients with and without detectable aPL, respectively. Interestingly, several patients with primary antiphospholipid syndrome did not have C4d-positive platelets (results not shown). These results reinforce the notion that abnormal levels of platelet C4d are associated with aPL-related thrombotic and vascular complications in SLE. In addition, P-C4d may be used to identify patients with SLE without aPL who are at an increased risk of thrombosis.

This study was designed to include only female subjects. However, male patients with SLE who were enrolled in a companion study (n = 18) were also tested for platelet C4d and aPL. Preliminary results indicated that, similar to their female counterparts in this study (Table 1), ∼50% and ∼17% of the male patients had detectable aPL (aCL and/or LAC) and C4d-positive platelets, respectively (Navratil JS, et al: unpublished observations). An association between aPL and platelet C4d positivity was also observed in these male patients, indicating that it is not a sex-specific phenomenon.

In summary, there is an urgent need for recognition of biomarkers that can be used to reliably identify and predict important clinical outcomes in patients with SLE. This study identifies platelet C4d as a promising candidate, based on a biologically plausible mechanism and the results of cross-sectional analyses. Of particular interest, platelet C4d may provide clues to the pathogenic mechanisms responsible for the thrombotic and vascular complications in SLE associated with aPL. Longitudinal studies will be required to address the role of platelet C4d as a predictor of future vascular events.

Acknowledgements

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

The following physicians provided the patients' blood samples and clinical information for this study: Drs. Dana Ascherman, Brian Berk, Timothy Carlos, Albert Donnenberg, Thomas Medsger, Chester Oddis, Margaret Ragni, William Ridgway, and Mary Chester Wasko. We thank Jason Brickner and Ron Bardelli for providing graphics assistance, and the University of Pittsburgh Center for Biologic Imaging for providing technical assistance. In addition, we thank Janice Sabatine for providing editorial assistance.

REFERENCES

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