Main

Thrombocytopenia (platelets <150 × 109/L) is the commonest hematologic abnormality presenting in the newborn(1–3). Nearly 1 in 4 of all newborns admitted to neonatal intensive care units will develop thrombocytopenia(1), whereas in sick preterm babies the incidence of thrombocytopenia can be as high as 72%(2). Thrombocytopenia occurs by 48 h of life in 75% of affected babies(1, 3). Recent evidence demonstrates that the majority of such babies with early thrombocytopenia have vastly reduced numbers of circulating megakaryocyte progenitor and precursor cells at birth(3). This strongly suggests that impaired fetal megakaryocytopoiesis and platelet production is the principal underlying mechanism responsible for their thrombocytopenia. The cause of this is unknown.

Tpo, the recently characterized specific and principal regulator of megakaryocytopoiesis and platelet production, is present in small amounts in healthy adults but is considerably raised in most thrombocytopenic adults(4, 5). Whether Tpo plays an equally pivotal role in megakaryocytopoiesis and platelet production in the healthy fetus and newborn is yet to be confirmed. Indeed nothing is currently known about Tpo levels in healthy term or preterm babies or if Tpo levels change in response to thrombocytopenia in the fetus and newborn. In addition, although megakaryocyte progenitor cells from healthy preterm babies appear to have increased sensitivity to Tpo compared with healthy term babies(6), it is not known if this sensitivity changes in disease states leading to early neonatal thrombocytopenia.

To answer these questions the experiments in this study were designed to:1) measure Tpo levels at birth in healthy term and preterm babies and in preterm babies who develop early thrombocytopenia (≤48 h of age); and 2) assess the in vitro effect of recombinant human Tpo on megakaryocyte precursors from each group.

METHODS

Patients. The study was performed at the Hammersmith Hospital, London. Cord/peripheral blood was obtained at birth from 17 healthy term babies (GA range 37-41 wk) and 29 preterm babies (GA range 24-34 wk). Samples were collected into 10 U/mL preservative-free heparin (CP Pharmaceuticals Ltd). All the term babies were from normal pregnancies and had uncomplicated neonatal courses. All the preterm babies were admitted to the neonatal intensive care unit. 13/29 (45%) developed thrombocytopenia (platelets < 150 × 109/L) by 48 h of age, whereas the remaining 16/29 maintained normal platelet counts until at least d 5. Clinical characteristics and platelet counts of the 29 preterm babies are shown in Table 1. For comparison peripheral blood for Tpo levels was also obtained from three children (age range 7-14 y) with severe nonimmune thrombocytopenia (platelets <50 × 109/L). This study was approved by the local ethical committee.

Table 1 Clinical details of 29 preterm babies assessed at birth for Tpo levels
Full size table

Thrombopoietin ELISA. Plasma for Tpo ELISA was obtained from blood samples by centrifugation at 500 × g for 10 min within 1 h of collection. Plasma samples were stored at -20°C until analyzed. Tpo levels were measured using a commercially available ELISA (R & D Systems Quantikine Human Tpo ELISA kit), with a lower limit of detection of 15 pg/mL.

MKp Assay. Remaining cell samples (after removal of plasma) were resuspended in an equal volume of RPMI 1640 medium (GIBCO). MNC were separated by density centrifugation on Lymphoprep (Nycomed) (density 1.077 g/mL) and depleted of adherent cells as previously described(7). These MNC are a plentiful source of normal megakaryocyte lineage precursor cells(7)-see below. MNC were cultured over a recombinant human Tpo (R & D Systems) dose range 0-100 ng/mL in a modification of the serum-dependent liquid culture system previously described(7). Briefly, MNC were suspended at 2 × 105/mL in 10 mL tissue culture tubes (Nunc) in Iscove's modified Dulbecco's medium (GIBCO), supplemented with 20% FCS (Harlan Sera-Lab), 1% BSA (GIBCO), 200 μg/mL iron saturated transferrin (Sigma Chemical Co.), 2 × 10-5 M α-monothioglycerol (Sigma Chemical Co.), 10 μL/mL minimal essential medium nonessential amino acids(100x) (GIBCO), and 10 μL/mL minimal essential medium vitamins solution(100×) (GIBCO). Cultures were incubated at 37°C in a humidified atmosphere of 5% CO2 in air in a tissue culture incubator for 10 d. After this, MKp were identified by alkaline phosphatase anti-alkaline phosphatase using a anti-human CD61 primary antibody (Dako) by the criteria previously described(7). (In brief, this assay measures the total number of megakaryocyte lineage precursor cells-from burst-forming unit-megakaryocyte (BFU-MK) to mononuclear megakaryoblasts, which can be produced in liquid culture from a given sample of MNC.) CD61 labeling of preculture MNC showed that no more than 1% of these cells (i.e.<2000) belonged to the megakaryocyte lineage (data not shown). This is in keeping with previous estimations of circulating megakaryocyte lineage cells at birth(8, 9). Where blood samples yielded sufficient MNC, response characteristics of MKp at 100 ng/mL recombinant human IL-3 (R & D Systems) were assessed in parallel by identical methods.

Statistics. All results are expressed as median (range) unless otherwise stated. Comparisons within groups were carried out using the Wilcoxon matched pairs test, and when comparing results between groups the Kruskal-Wallis and Mann-Whitney tests (with Bonferroni correction) were used as appropriate.

RESULTS

Thrombopoietin levels. Tpo was detected in plasma from all 46 babies (Table 2). Healthy term and preterm babies and thrombocytopenic preterm babies all had Tpo levels marginally higher than the normal adult range(4, 5): term, 145 pg/mL (52-237 pg/mL); preterm, 132 pg/mL (32-318 pg/mL); preterm thrombocytopenic, 185(46-264 pg/mL). However, despite the fact that the thrombocytopenic preterm babies had significantly lower platelet counts at birth, and during the 1st wk of life (Table 1), when compared with the healthy preterm babies, there was no significant difference between the Tpo levels in any of the groups (p = 0.311). Moreover, Tpo levels in the thrombocytopenic preterm babies were much lower than the levels seen in the three thrombocytopenic children assessed (905, 2138, and 2700 pg/mL) and as previously reported in thrombocytopenic adults(4, 5).

Table 2 Tpo levels and in vitro response of MKp to recombinant human Tpo in healthy term and preterm babies and in thrombocytopenic preterm babies
Full size table

Response of MKp to exogenous Tpo. MKp from all three groups showed dose-dependent proliferation in response to Tpo. In term and healthy preterm babies MKp numbers in culture showed a significant increase over baseline (no Tpo) at both 50 and 100 ng/mL Tpo (p < 0.01 for both comparisons), (Fig 1). MKp from thrombocytopenic preterm babies also showed a similar significant increase (p < 0.01) in response to 100 ng/mL Tpo (Table 2). [As, there are fewer MKp in cord/peripheral blood at birth in preterm babies with early thrombocytopenia(3) (Table 2), we were only able to assess the MKp Tpo response in these babies at 0 and 100 ng/mL Tpo.] MKp from both healthy and thrombocytopenic preterm babies showed a significantly greater response to Tpo than MKp from term babies when the MKp response to Tpo at 100 ng/mL in each group was expressed as a ratio of increase of MKp numbers over cultures containing no Tpo-48.2-, 24.6-, and 9.9-fold, respectively (p < 0.05 for both comparisons,Table 2). MKp from healthy term and preterm babies also showed significant proliferation in response to 100 ng/mL IL-3(Fig. 1); however, this increase was significantly less than that seen with 100 ng/mL Tpo (p < 0.05)(Fig 1).

Figure 1
figure 1

Dose-response characteristics of MKp to Tpo and IL-3 in healthy term and preterm babies. MKp in both groups show significant proliferation over baseline (no Tpo) at 50 and 100 ng/mL Tpo (p < 0.01 for both comparisons). MKp in both groups also show proliferation over baseline (no Tpo) at 100 ng/mL IL-3; however, this is significantly less pronounced when compared with 100 ng/mL Tpo (p < 0.05).

Full size image

DISCUSSION

Although c-Mpl ligand or thrombopoietin (Tpo) was only cloned and sequenced in 1994(10–13), characterization of its in vitro and in vivo activity clearly show it to be the major physiologic regulator of megakaryocytopoiesis and platelet production in adults(14). Early onset thrombocytopenia is very common in sick newborns, especially those born preterm. In the main this appears to be the consequence of impaired fetal megakaryocytopoiesis and platelet production occurring at least as early as the most primitive committed megakaryocyte progenitor cell-BFU-MK(3). This suggests that abnormalities of Tpo production and/or Tpo responsiveness of megakaryocyte progenitor and precursor cells in the fetus and newborn may play a role in the development of thrombocytopenia in these groups. However, until now Tpo levels in healthy or thrombocytopenic newborns of all gestations were unknown, as was the responsiveness of megakaryocyte lineage precursor cells to Tpo in thrombocytopenic preterm babies.

This study shows that healthy term and preterm babies and preterm babies with early onset thrombocytopenia (≤48 h) all have detectable levels of Tpo in their plasma at birth-145 pg/mL (52-237 pg/mL), 132 pg/mL (32-318 pg/mL), and 185 (46-264 pg/mL), respectively. These levels are marginally higher than Tpo levels previously reported in healthy adults(4, 5). Preterm babies with early onset thrombocytopenia have slightly higher Tpo levels than those seen in healthy term and preterm babies, but this difference is not statistically significant. MKp from healthy term and preterm babies and from preterm babies with early onset thrombocytopenia all show in vitro dose-dependent proliferation in response to Tpo. This response is greater than that to IL-3. In both healthy and thrombocytopenic preterm babies MKp show significantly greater proliferation in response to Tpo than term babies. These data demonstrate, for the first time, a number of important points in relation to megakaryocytopoiesis and platelet production in the fetus/newborn. First, Tpo is detectable in the human fetus/newborn baby from as early as 24 wk of gestation. In addition, MKp from newborn babies of all viable gestations show a significant proliferative response to Tpo. Because this response is greater than that to IL-3 (the most potent in vitro hemopoietic growth factor stimulator of fetal/neonatal megakaryocyte precursors previously known)(3, 15), this indicates that Tpo is likely to be the major regulator of megakaryocytopoiesis and platelet production in the fetus/newborn, as it is in adults. Second, MKp from both healthy and thrombocytopenic preterm babies show a greater proliferative response to Tpo than MKp from term babies. This is similar to the enhanced response of other “preterm” hemopoietic progenitors to their specific hemopoietic growth factors(16–19). Third, preterm babies appear to have an impaired ability to significantly increase Tpo levels, even in response to thrombocytopenia (as evidenced by the much higher levels of Tpo seen in the thrombocytopenic children we studied and levels reported in thrombocytopenic adults)(4, 5). Inappropriately low Tpo production may have considerable implications for the etiology of early neonatal thrombocytopenia and might be analogous to the impaired erythropoietin production which characterizes “anemia of prematurity”(20). Megakaryocytopoiesis is already impaired in these babies at birth, as evidenced by the reduced numbers of MKp present in cord blood(3) (Table 2), and their lower platelet counts at birth (Table 1). This suggests that there is a dysregulation of fetal megakaryocytopoiesis and that the fetus may not be able to increase platelet production appropriately as a consequence of an inadequate Tpo response. A caveat to this conclusion is that because we measured Tpo levels in the thrombocytopenic babies at birth, and the platelet nadir for most babies was not reached until d 4, it is possible that Tpo levels measured at the platelet nadir would show higher values. (We are currently conducting a study to answer this question.)

In conclusion these studies are the first to begin to define the role of Tpo production/Tpo action in the clinically important setting of neonatal thrombocytopenia. Although more studies will clearly be required, the data from this and from studies in neonatal rats(21) suggest that rh Tpo may have a role in the treatment of neonatal thrombocytopenia, particularly in preterm babies in the early days of life.