Characterization of blood borne microparticles as markers of premature coronary calcification in newly menopausal women

Abstract

While the risk for symptomatic atherosclerotic disease increases after menopause, currently recognized risk factors do not identify ongoing disease processes in low-risk women. This study tested the hypothesis that circulating cell-derived microparticles may reflect disease processes in women defined as low risk by the Framingham risk score. The concentration and phenotype of circulating microparticles were evaluated in a cross-sectional study of apparently healthy menopausal women, screened for enrollment into the Kronos Early Estrogen Prevention Study. Microparticles were evaluated by flow cytometry, and coronary artery calcification (CAC) was scored using 64-slice computed tomography scanners. The procoagulant activity of isolated microparticles was determined with a sensitive fluorescent thrombin generation assay. Chronological age, body mass index, serum lipids, systolic blood pressure (Framingham risk score < 10%, range 1-3%), and high-sensitivity C-reactive protein did not differ significantly among women with low (0 < 35; range, 0.3-32 Agatston units) or high (>50; range, 93-315 Agatston units) CAC compared with women without calcification. The total concentration and percentage of microparticles derived from platelets and endothelial cells were greatest in women with high CAC scores. The thrombin-generating capacity of the isolated microparticles correlated with phosphatidylserine expression, which also was greatest in women with high CAC scores. The percentages of microparticles expressing granulocyte and monocyte markers were not significantly different among groups. Therefore, the characterization of platelet and endothelial microparticles may identify early menopausal women with premature CAC who would not otherwise be identified by the usual risk factor analysis.

Fig. 1.

Representative scatter plot obtained by FACSCanto flow cytometry. A: control gates of buffer with fluorescein-conjugated antibodies and calibration (size and TruCount beads) beads in the absence of sample. B: gates derived from adding a sample containing microparticles to the buffer with fluorescein-conjugated antibodies and calibration beads. Note that the gate to define the area of interest (P1) was set above the background noise of the machine. C and D: representative quadrants (Q) derived from the microparticle gates shown in A and B, respectively. Counts are separated by antibody binding with Q3 representing microparticles. FL, fluorescence. E and F: representative scanning (E) and transmission (F) electron microscopy of the isolated microparticles. Arrow heads, membranes. G: microparticles imaged by CytoViva dual fluorescence and optical microscopy at ×150. H: fluorescent beads (1 μm in diameter) imaged by the CytoViva cell imaging system at ×150.

Fig. 2.

Analytical variability in total numbers of microparticles between 2 aliquots of the same sample from individual women. Each point represents data from an individual woman.

Fig. 3.

Representative quadrants derived from flow cytometric scatter plots (top) demonstrating annexin-V-negative (Q3) and -positive (Q4) microparticles. Bottom: cumulative data for numbers of total (annexin-V negative, Q3, white bars, plus annexin-V positive, Q4, black bars) microparticles from women with negative (0 Agatston units, n = 10 women) and positive coronary calcification scores (low, <50 Agatston units, n = 18 women; and high, >50 Agatston units, n = 5 women). Data are presented as means ± SE. *P < 0.05, statistical significance from those with zero calcium and low calcium scores. FL1, green fluorescence; FL2, red fluorescence.

Fig. 4.

Representative quadrants derived from flow cytometric scatter plots of microparticles labeled with platelet-specific antibodies CD61 (integrin-β₃; top, left) and CD42a (glycoprotein IX; top, right) in combination with the antibody for annexin-V. Cumulative numbers of CD61 (bottom, left)- and CD42a (bottom, right)-positive microparticles from women with negative (0 Agatston units, n = 10 women) and positive coronary calcification scores (low, 0 < 50 Agatston units, n = 18 women; and high, >50 Agatston units, n = 5 women). Data are presented as means ± SE. *P < 0.05, statistical significance from those with zero calcium and low calcium scores.

Fig. 5.

Representative quadrants from flow cytometric scatter plot of microparticles labeled to identify those derived from the endothelium by CD62-E (E-selectin) antibody and annexin-V antibody (top). Cumulative data of CD62-E-phycoerythrin-positive microparticles with (black bars) and without (white bars) annexin-V-positive expression from women with negative (0 Agatston units, n = 10 women) and positive coronary calcification scores (low, 0 < 50 Agatston units, n = 17 women; and high, >50 Agatston units, n = 5 women). Inset: correlation of CD62-E-positive microparticles (MPs) with the coronary artery calcification (CAC) score. Data are presented as means ± SE. *P < 0.05, statistical significance from those with zero calcium and low calcium scores.

Fig. 6.

Prothrombinase assay of thrombin generation by microparticles (10,000) derived from women with negative (0 Agatston units, n = 10 women; ⧫) and positive coronary calcification scores (low, 0 < 50 Agatston units, n = 18 women; □, and high, >50 Agatston units; n = 5 women, •). Control for buffer with factors but no microparticles (◊; n = 2) is shown. Data are presented as means ± SE. *P < 0.03, significant variability by Bartlett's test and thrombin-generating capacity by log transformation in the high compared with low and zero CAC groups.