| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1
Pediatric Hematology and Oncology and
2
Institute of Clinical Chemistry and Laboratory Medicine and Institute of Arteriosclerosis Research, University of Münster, Albert Schweitzer-Strasse 33, 48129 Münster, Germany;
a address correspondence to this author at: Institute of Clinical Chemistry and Laboratory Medicine, University of Münster, Albert Schweitzer-Strasse 33, 48129 Münster, Germany
A simple, rapid method is needed for collection of platelets for flow cytometric measurement of platelet activation in investigations relating to coronary heart disease, stroke, and peripheral arterial disease (1)(2)(3)(4)(5)(6). Specimen collection and sample preparation must minimize activation of platelets (7)(8)(9).
The most frequently used anticoagulant for platelet analysis is sodium citrate, but it is deficient because of the difficulty in controlling osmolarity in functional assays (10). Other anticoagulants (EDTA and recombinant thrombin inhibitors such hirudin or low-molecular weight heparin) offer no alternative because of possible interactions with other substances used for analysis (10). In addition, platelets stimulated with ADP in citrate blood usually aggregate if stirred. To prevent clotting and cell-to-cell-adhesion, blood must be diluted and stirring reduced (7) when unfixed platelets are used. However, the advantage of unfixed samples is the opportunity to check the influence of substances added in vitro (11).
The use of citrate-theophylline-adenine-dipyridamol (CTAD) buffer rather than citrate decreased platelet activation as indicated by lowering the plasma concentrations of platelet factor 4 (12). To investigate the influence of this buffer on platelet activation, we measured flow cytometrically detectable activation of platelets isolated from citrate-buffered blood and from commercially available CTAD-buffered blood.
Blood was taken from 10 healthy volunteers who had not ingested drugs that affect platelet function for at least 14 days. Venipuncture was performed using a butterfly needle (21G; Sarstedt). Blood was aspirated directly into four Monovette tubes (Sarstedt) containing citrate or CTAD buffer.
Platelets were isolated by a modification of the method described by Faraday et al. (13). The Monovette tubes were centrifuged at 180g for 18 min. Platelet-rich plasma (PRP) was obtained and carefully mixed with H-D-Phe-Pro-Arg-chloromethylketone (Bachem), a highly effective irreversible thrombin inhibitor, in a final concentration of 30 nmol/L. The PRP was then centrifuged for 12 min at 750g, and the supernatant (platelet-poor plasma) was removed. The remaining PRP was centrifuged at 750g for 8 min after the addition of 2 mL of 0.01 mol/L Tris buffer, pH 7.4, containing 138 mmol/L NaCl, 5.5 mmol/L glucose, 1.8 mmol/L CaCl2, 0.49 mmol/L MgCl2, and 3.5 g/L bovine serum albumin (BSA). After the rinse was removed, the platelets were carefully resuspended in fresh 0.01 mol/L Tris buffer, pH 7.4. The platelet concentration was measured on a hematology analyzer (MD II; Coulter) and adjusted with the same Tris buffer to a final concentration of 100 000 platelets/µL.
Before and after activation with ADP (20 µmol/L) for 10 min at room temperature, the platelets were incubated for 20 min at room temperature in the dark with anti-CD62P-fluorescein isothiocyanate (FITC), or anti-CD63-FITC antibodies, in final concentrations of 16.7 mg/L (7)(14). An anti-mouse-IgG1-FITC antibody (Immunotech) of the same concentration and fluorescence/protein ratio was used as a negative control. Additional samples were prepared with fibrinogen coupled to Oregon Green (fibrinogen-OG; Molecular Probes) in a final concentration of 0.3 g/L. Platelets incubated with a conjugate of BSA and Oregon Green (BSA-OG; Molecular Probes) at the same concentration and fluorescence/protein ratio as for the fibrinogen-OG reaction served as a negative control.
The platelets were analyzed with a Coulter Epics Elite flow cytometer after each sample was diluted with Tris buffer to a flow count of 1000/s. The instrument was checked daily by DNAcheck beads (Coulter) and was calibrated by beads of six different fluorescence intensities (Fluorespheres; Dako). For the different fluorescence intensities the CV from day to day (total of 6 days) varied from 1% to 4%. Light scatter and fluorescence data were obtained in the logarithmic mode (range, 0.1–1024; four decades; 1024 channels).
Five thousand cells of each sample were analyzed. The platelet population was identified by its light scatter characteristics. To determine the shape change, a sensitive indicator of platelet activation (15), the length/width (L/W) ratio of an automatically set gate around the platelet population was calculated. For each single measurement, the gate around the platelet population drawn within the forward scatter/side scatter dot plot was set by the "autogate" algorithm, which is a part of the XL2 software (Coulter). This algorithm draws an elliptical region at 1.1% of the processed peak. The quotient of the maximum and minimum diameters of the region was calculated and given as the L/W ratio of the platelet population.
The mean fluorescence of samples containing anti-CD62P-FITC or anti-CD63-FITC was calculated by subtracting the mean fluorescence of samples containing anti-mouse-IgG1-FITC. Fibrinogen binding was calculated by subtracting the mean fluorescence of samples containing BSA-OG from samples containing fibrinogen-OG.
After the mean fluorescence values of the negative controls were subtracted, the means and SDs as well as the medians and 95% confidence intervals of the fluorescence values of the positive samples were calculated. A test for gaussian distribution was performed. Wilcoxon or Mann-Whitney tests were performed for comparison of the mean fluorescence values. All statistical analyses were performed using the MedCalc software (MedCalc).
The results of the shape change measurement are given in Table 1
. A significant difference was found for the L/W ratio of
platelets isolated from CTAD-buffered blood before and after activation
with ADP (P <0.05; Table 1
).
|
For platelets isolated from citrate-buffered blood, the resulting fluorescence values [mean ± SD (median/95% confidence interval)] in arbitrary units (AU) were 0.9 ± 0.4 AU (0.9 AU/0.46–1.41 AU) for anti-CD62P binding and 0.4 ± 0.3 AU (0.4 AU/0.23–0.94 AU) for anti-CD63 binding before activation. After activation, the values were 0.7 ± 0.4 AU (0.8 AU/0.43–1.05 AU) for anti-CD62P and 0.5 ± 0.3 AU (0.4 AU/0.09–0.68 AU) for anti-CD63.
For the platelet population isolated from CTAD-buffered blood, the mean
fluorescence values were 0.5 ± 0.3 AU (0.4 AU/0.24–0.94 AU) for
anti-CD62P binding and 0.4 ± 0.2 AU (0.4 AU/0.19–0.70 AU) for
anti-CD63 binding before activation. After activation, the values
changed to 1.9 ± 0.5 AU (2.0 AU/1.26–2.29 AU) for anti-CD62P and
1.1 ± 0.5 AU (1.0 AU/0.60–1.62 AU) for anti-CD63. For the
differences and significance values, see Fig. 1
.
|
After activation of the platelets isolated from citrate-buffered blood,
the fluorescence was 14.1 ± 15.1 AU (11.1 AU/3.66–30.74 AU).
Before activation, fibrinogen binding was 1.2 ± 6.0 AU (1.2
AU/-5.89 to 6.24 AU). For the platelet population isolated from
CTAD-buffered blood, the fluorescence changed from 2.9 ± 3.5 AU
(2.7 AU/0.90–8.25 AU) before activation to 293.2 ± 328.2 AU (92
AU/33.01–793.81 AU) after activation. For the differences and
significance values, see Fig. 1
.
Platelet shape change is a more sensitive marker for platelet activation in vitro than fibrinogen or anti-CD62P and anti-CD63 binding (15). Platelets activated with ADP change from a discoid into a more spherical form (16).
We found a higher preactivation, indicated by a higher shape change of the platelet population, in platelets isolated from citrate-buffered blood than in platelets from CTAD-buffered blood. In citrate-buffered blood, the shape differed only slightly before and after incubation with ADP. In contrast, platelets from CTAD-buffered blood showed a significant difference toward higher values after activation.
Because shape change is a reversible process (15), CD62P and CD63 binding is better suited to the detection of activated platelets in vivo. Furthermore, Faraday et al. (13) reported that binding of fibrinogen to platelets is a specific and saturable process.
Platelet preactivation was lower when the CTAD buffer was used, as indicated by the lower binding of anti-CD62P on platelets without the addition of ADP. Moreover, in platelets isolated from CTAD-buffered blood, we found a significant increase of anti-CD62P, anti-CD63, and fibrinogen binding after activation, whereas in platelets isolated from citrate-buffered blood the only increase observed was for fibrinogen binding. The binding of anti-CD62P, anti-CD63, and fibrinogen was stronger in platelets when the CTAD buffer was used than when citrate was used.
Thus, in addition to a higher preactivation state, platelet function was negatively affected by citrate buffer. Activated platelets will be detected in vivo when citrate buffer is used as an anticoagulant because of artificial activation. The advantages of using CTAD Monovettes also include their price, which is only slightly higher than that of citrate Monovettes. Therefore, for flow cytometric measurement of platelet activation, the CTAD buffer is preferable to sodium citrate as an anticoagulant for blood collection. Moreover, we surmise that the CTAD buffer is also suitable for other methods for determining platelet function, e.g., whole blood methods and aggregometry in PRP.
Acknowledgments
This study was supported by the Landesversicherungsanstalt Westfalen and the Landesversicherungsanstalt Rheinprovinz. We thank Margit Käse, Pia Becker, and Ruth Bäumer for excellent technical assistance.
Footnotes
fax 49-251-8347227, e-mail junkerr{at}uni-muenster.de
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  R. Zimmermann, J. Koenig, J. Zingsem, V. Weisbach, E. Strasser, J. Ringwald, and R. Eckstein Effect of Specimen Anticoagulation on the Measurement of Circulating Platelet-Derived Growth Factors Clin. Chem., December 1, 2005; 51(12): 2365 - 2368. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
0.05; **, P