Main

It is well recognized that the platelets of neonates exhibit a transient functional insufficiency (1–6). Very few, and often controversial, studies describing the expression of membrane GP on neonatal platelets have been published. Rajasekhar et al. (7) reported that platelet surface expression of GP IIb-IIIa on resting platelets was not significantly different in neonates and adults. In contrast, Kühne et al. (8) showed significantly reduced expression of GP IIb-IIIa on neonatal platelets compared with platelets of their healthy mothers. Meher-Hojmi et al. (9) reported no difference in the expression of GP IIIa on platelets of pregnant versus nonpregnant women. Rajasekhar et al. (7) reported lower reactivity of neonatal platelets to a variety of platelet agonists on the basis of the extent of increase in the platelet surface expression of P-selectin and the GP IIb-IIIa complex and the extent of decrease in the platelet surface expression of the GP Ib-IX complex. A more profound platelet hyporeactivity to agonists was recently described by these authors in a small group of very low birth weight neonates (10). In addition, Gatti et al. (11) showed that the expression of a conformational epitope on GP IIb-IIIa recognized by the MAb PAC-1 after platelet activation was significantly lower in neonatal platelets compared with adult controls.

To avoid possible artifacts caused by platelet isolation, we performed a whole blood flow-cytometric study of neonatal platelets with immunostaining within an hour after sampling and before fixation. In the first part of our study, we compared the surface expression of the major platelet membrane glycoproteins GP Ia-IIa, GP Ib, GP IIb-IIIa, and GP IV and the platelet activation markers CD63 and P-selectin on resting platelets of full-term newborns and healthy adult donors. In the second part of the study, we evaluated changes in the surface expression of GP IIb-IIIa, GP IV, and activation markers P-selectin and CD63 after in vitro activation of platelets with different concentrations of a strong and stable agonist, the human TRAP (12).

METHODS

Subjects.

Resting platelets were studied in 27 full-term newborns, sampled at 83 ± 32 h of life for other biochemical blood tests ordered by a physician because of increased values of transcutaneous bilirubin as assessed by an AirShields jaundice meter 101 (Minolta Camera, Osaka, Japan). Newborns were healthy except for a moderate hyperbilirubinemia with serum bilirubin level of 237 ± 49 μM. No blood group incompatibility, infection, or other complications were detected. Jaundice was detected in all studied newborns only after the first 24 h of life and did not persist for more than 1 wk. Direct (conjugated) serum bilirubin was never >20 μM and hypoalbuminemia, defined as serum albumin level <33 g/L, was not detected. The mean gestational age was 39.6 ± 1.5 wk. A group of 12 healthy adult donors with a mean age of 26.8 ± 8.2 y was used for comparison.

Platelet activation experiments were performed in a separate group of 15 full-term newborns, sampled at 71 ± 32 h of life with the same clinical status as the first group. The mean gestational age of this group was 39.3 ± 1.6 wk, and the mean serum bilirubin level was 239 ± 59 μM. For comparison, platelets of 15 healthy adult blood donors of a mean age of 29.3 ± 9.8 y were studied at the same time.

None of the mothers or neonates had been given aspirin or other antiinflammatory drugs. The study was approved by the Ethical Board of the Institute for the Care of Mother and Child in Prague, and all blood samples of neonates were taken after written consent of the parents.

Blood sampling.

Blood was drawn by venipuncture of a cubital vein using a 20-gauge needle, and the first 1 mL was used for non–platelet-related tests. Blood for platelet analysis was collected in 0.38% (final concentration) trisodium citrate in polypropylene tubes. Within 1 h of being drawn, 50 μL of anticoagulated blood was mixed with 450 μL of Tyrode-HEPES-BSA buffer, pH 7.35 (4 mM HEPES, 137 mM NaCl, 2.7 mM KCl, 1 mM MgCl2, 5.5 mM glucose, 3 mM NaH2PO4, 0.1% BSA), and immunolabeling of samples was initiated immediately.

MAb.

FITC- and PE-conjugated antibodies against platelet GP Ia (CD49b FITC, Gi9), GPIb (CD42b FITC/PE, SZ2), GP IIb-IIIa (CD41a FITC, P2), P-selectin (CD62 PE, CLBThromb/6), GP 53 (CD63 PE, CLBGran/12), and isotypic IgG1 (FITC, PE) controls were from Immunotech S.A. (Czech Republic). Anti-GP IV (CD36 FITC, CLB-IVC7) MAb was purchased from the central laboratory of the Netherlands Red Cross Blood Transfusion Service CLB (Amsterdam, The Netherlands).

Flow cytometry.

For each sample, the fluorescence from 5000 platelets, gated by their characteristic forward and side-light scatter, was measured on an Epics XL flow cytometer (Coulter, Fullerton, CA). Intensity of fluorescence was determined in logarithmic mode with the scale divided into 1023 channels, and the value for each sample was expressed as a mean fluorescence channel. The mean fluorescence of the negative control was subtracted from the fluorescence of the sample. For analysis of activated platelets, the discriminator was set to analyze only particles positive for platelet-specific fluorescence. The stability of performance of the flow cytometer was tested daily using DNA-Check beads (Coulter).

Analysis of resting platelets.

Fifty microliters of diluted blood was incubated for 20 min at room temperature with 5 μL of FITC-labeled MAb (saturating concentration). After incubation, samples were diluted with 750 μL of cold 1% solution of paraformaldehyde in PBS and immediately analyzed using flow cytometry. Fixation after immunolabeling did not lead to significant changes in platelet immunofluorescence, compared with analysis of unfixed samples, and the platelet immunofluorescence of fixed samples was stable for at least 6 h.

Analysis of TRAP-activated platelets.

Aliquots of diluted blood (20 μL) were double-stained during activation for 20 min at room temperature with 20 μL of a mixture of saturating concentrations of the following MAb—anti-GP IIb-IIIa FITC/anti-P-selectin PE or anti-GP IV FITC/anti-CD63 PE—and with 0–15 μM human TRAP (SFLLRNP, synthesized by Dr. Bláha at the Institute of Organic Chemistry and Biochemistry, Czech Academy of Science, Prague) in Tyrode-HEPES-BSA buffer, pH 7.35. The incubation was stopped with 0.5 mL of cold 1% solution of paraformaldehyde in PBS. Samples were stored at 4°C and analyzed using flow cytometry within 4 h.

Statistics.

Results are presented either in box plots indicating 25th–75th (vertical boxes) and 10th–90th (vertical lines beyond the boxes) percentiles of the group distribution, or as a mean values of all subjects in the groups ± SD. Statistical analysis was achieved by unpaired t testing or by a Mann-Whitney nonparametric test, when appropriate. p< 0.05 was regarded as significant.

RESULTS

Analysis of resting platelets.

Platelets of newborns showed significantly lower expression of GP Ia-IIa, GP Ib (p< 0.01) (Fig. 1A), and GP IIb-IIIa (p< 0.001) (Fig. 1B). In contrast, the expression of GP IV on neonatal platelets was significantly higher compared with platelets of adults (p< 0.001) (Fig. 1B). The surface expression of platelet activation markers CD63 and P-selectin on resting platelets was very low and did not differ significantly in neonates and adults. The nonspecific binding of IgG1 isotypic control antibody and the size of platelets assessed by mean forward scatter were not significantly different in neonates and adults (data not shown).

Figure 1
figure 1

Surface expression of membrane glycoproteins on resting platelets. Surface expression of membrane glycoproteins GP Ia-IIa and GP Ib (A) and GP IIb-IIIa and GP IV (B) on resting platelets of full-term neonates (n= 27) and healthy adult donors (n= 12) as measured with fluorescein-labeled MAb and flow cytometry. Results are presented in box plots indicating the median (full line), mean (dashed line), and 25th–75th and 10th–90th percentiles of the group distribution as boxes and “error” bars, respectively. Outlying values are plotted as empty circles.

Full size image

Analysis of TRAP-activated platelets.

After activation with TRAP, a significant increase in the surface expression of GP IIb-IIIa was observed in platelets of both adults and newborns (Fig. 2A). At 15 μM TRAP, the relative increase of expression of GP IIb-IIIa on activated platelets of neonates and adults was similar, with 147 ± 22% and 144 ± 23% of the resting values, respectively. However, the expression of GP IIb-IIIa on activated adult platelets was 18% higher (p< 0.01) than on neonatal platelets. An equivalent difference was also observed in resting platelets. Surface expression of GP IV after TRAP activation increased slightly, but significantly, in neonates as well as in adults compared with the expression on resting platelets and platelets activated with 0.5 μM TRAP (p< 0.05) (Fig. 2B). At TRAP concentration of 15 μM, the increase of GP IV surface expression was highly significant (p< 0.001), with the maximum relative increase to 114 ± 14% and 111 ± 9% of the resting value for neonatal and adult platelets, respectively. The expression of GP IV was 42% higher on platelets of neonates when compared with platelets of adults and was also observed in resting platelets (Fig. 1B). The extent of the surface expression of P-selectin (Fig. 2C) and CD63 (Fig. 2D) on platelets of neonates and adults at TRAP concentrations of 0.5–5 μM was not significantly different. At TRAP concentrations of 15 μM, the mean extent of the expression of both activation markers was higher on platelets of adults, with 51% and 47% for P-selectin and CD63, respectively.

Figure 2
figure 2

Surface expression of membrane glycoproteins on human TRAP-activated platelets. Surface expression of membrane glycoproteins GP IIb-IIIa (A) and GP IV (B) and activation markers P-selectin (C) and CD63 (D) on platelets of full-term neonates (n= 15) and healthy adult donors (n= 15) after activation with different concentrations of TRAP. Activation was measured using double staining with a mixture of MAb anti-GP IIb-IIIa FITC/anti-P-selectin PE or of anti-GP IV FITC/anti-CD63 PE and flow cytometry. Results are presented as mean platelet fluorescence (±SD; *p< 0.05, **p< 0.01).

Full size image

DISCUSSION

Platelets of neonates exhibit relative functional insufficiency when compared with platelets of adults (1–6). Functional responses of platelets, such as adhesion, activation, and aggregation, are mediated by their membrane receptors. In the first part of the current study, we addressed the question of whether differences exist in the expression of major receptors on resting neonatal and adult platelets. We studied surface expression of the platelet membrane glycoproteins GP Ia-IIa, GP Ib, GP IIb-IIIa, and GP IV. The GP Ia-IIa complex is a collagen receptor (13, 14), whereas the GP Ib-IX complex is an essential adhesive receptor for von Willebrand factor and also a receptor for thrombin (15, 16). The GP IIb-IIIa complex plays a critical role in platelet aggregation and adhesion as an integrin receptor for fibrinogen, von Willebrand factor, fibronectin, and other Arg-Gly-Asp-proteins (17). The GP IV is a receptor for collagen and thrombospondin (18–22). Because activation of platelets during delivery has been suggested as a possible cause of neonatal platelet dysfunction (5), we also studied the surface expression of platelet activation markers P-selectin (GMP 140) and CD63 (GP 53). P-selectin is a platelet α-granular membrane GP (23, 24), and CD63 is a GP present in lysosomes and dense granules of resting platelets (25, 26).

Our results show a significantly lower expression of the membrane glycoproteins GP Ia-IIa, GP Ib, and GP IIb-IIIa on resting neonatal platelets compared with platelets of adults. Whereas it is unlikely that moderate decreases of any of the individual receptors can cause platelet insufficiency, the cumulative effect of their decrease cannot be excluded. Previous studies on neonatal platelets reported either decreased or normal expression of GP Ib and GP IIb-IIIa when compared with adult platelets (7–10). Similarly, Israels et al. (2) reported no difference in the expression of GP Ia-IIa on isolated, fixed platelets in seven adult-neonate sample pairs. One interesting finding in our study was the significantly higher expression of GP IV on neonatal platelets. Increase of platelet GP IV was previously reported in patients with myeloproliferative syndromes (27). The only available data about GP IV on neonatal platelets by Kühne et al. (8) demonstrated no difference in the expression of GP IV between newborns and their healthy mothers. It is not clear how an increase in GPIV could contribute to a functional defect of neonatal platelets.

All available studies are consistent concerning the expression of activation markers CD63 and P-selectin on neonatal platelets. Analysis of the cord blood at time of delivery, as well as of the venous blood within the first days of life, showed a very low expression of P-selectin and CD63 on resting neonatal platelets (7–9). These data suggest that neonatal platelets do not circulate in a degranulated state.

The second part of our study was focused on the ability of neonatal platelets to up-regulate membrane glycoproteins GP IIb-IIIa, GP IV, P-selectin, and CD63 after in vitro activation. Our platelet activation experiments with TRAP may be compared with the published studies of thrombin-activated neonatal platelets (7, 8, 10). Both agonists act through the seven transmembrane-domain thrombin receptor and are able to facilitate full platelet activation. However, TRAP does not stimulate the release of Ca2+ from internal platelet stores as efficiently as thrombin, which probably also acts through an additional binding site (28). Activation of platelets with thrombin in whole blood in the presence of anti–fibrin-polymerization peptide resulted in a significantly higher expression of GP IIb-IIIa and P-selectin on platelets of adults compared with platelets of neonates (7, 10). This is consistent with our results with TRAP-activated platelets. Several other studies showed an increase in platelet surface expression of GP IV after platelet activation (29–32). We also observed a moderate increase of GP IV in TRAP-activated platelets of neonates as well as of adults.

Hyporeactivity of platelets in healthy full-term newborns has been shown to be transient and to be restored to normal levels by the 10th day after birth (11, 33). Owing to ethical reasons, blood sampling in healthy full-term newborns in our institute was approved only for diagnostic tests, when transcutaneous bilirubin was increased. Previous studies of neonatal platelet GP were mostly conducted on blood collected at time of delivery or during the first 24 h of life. For our study, we selected newborns with a mean age of approximately 72 h and with a moderate hyperbilirubinemia without any complications. The mean serum bilirubin level of newborns in our study is on the borderline of the generally accepted limit for physiologic jaundice (34). Hyperbilirubinemia in neonates is a phenomenon that we think is unlikely to affect constitutive expression of platelet membrane GP. Karim et al. (35) did not find any difference in plasma level of the platelet activation marker β-thromboglobulin between jaundiced and nonjaundiced newborns. However, another study showed alteration of platelet thromboxane B2 production during spontaneous clotting of blood by hyperbilirubinemia (36). We cannot exclude the possible influence of hyperbilirubinemia on the expression of platelet surface receptors and on the impairment of activation response of neonatal platelets.

In conclusion, the lower expression of adhesive receptors GP Ia-IIa, GP Ib, and GP IIb-IIIa on resting neonatal platelets and the lower ability of neonatal platelets to up-regulate CD63 and P-selectin after in vitro activation may collectively play a role in the functional impairment of neonatal platelets.