• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Jan 18, 2005; 102(3): 862–867.
Published online Jan 10, 2005. doi:  10.1073/pnas.0406008102
PMCID: PMC545528

A hematopoietic growth factor, thrombopoietin, has a proapoptotic role in the brain


Central nervous and hematopoietic systems share developmental features. We report that thrombopoietin (TPO), a stimulator of platelet formation, acts in the brain as a counterpart of erythropoietin (EPO), a hematopoietic growth factor with neuroprotective properties. TPO is most prominent in postnatal brain, whereas EPO is abundant in embryonic brain and decreases postnatally. Upon hypoxia, EPO and its receptor are rapidly reexpressed, whereas neuronal TPO and its receptor are down-regulated. Unexpectedly, TPO is strongly proapoptotic in the brain, causing death of newly generated neurons through the Ras-extracellular signal-regulated kinase 1/2 pathway. This effect is not only inhibited by EPO but also by neurotrophins. We suggest that the proapoptotic function of TPO helps to select for neurons that have acquired target-derived neurotrophic support.

Keywords: astrocytes, erythropoietin, neurons, differentiation, development

In the hematopoietic system, survival, proliferation, and differentiation of cells are regulated by a plethora of growth factors (1-4). The effect of erythropoietin (EPO) on the generation of red blood cells is well known. The hematopoietic growth factor thrombopoietin (TPO) stimulates megakaryopoiesis and thrombocyte formation (1-3, 5, 6). During hematopoiesis, EPO and TPO can interact in a synergistic and an antagonistic fashion (1-3, 7).

EPO and TPO exhibit significant homology in their receptor-binding domain (20% identity and 25% similarity). Likewise, they bind to receptors, erythropoietin receptor (EPOR) and thrombopoietin receptor (TPOR), respectively, that belong to the same cytokine receptor superfamily (1-3, 8-10). Previous studies reported a neurotrophin-like motif in the N-terminal receptor binding region of the TPO molecule, with conflicting data about the presence of TPO in the brain (5, 6, 11-13).

For EPO, it is well established that the gene is expressed in the embryonic CNS. EPO has a marked effect as a survival factor for neurons and their progenitors (14, 15), presumably to overcome phases of physiological hypoxia (16, 17). The widespread but rather “unspecific” neuroprotective potential of EPO is regained in the adult CNS upon distress or injury. This finding has been confirmed in rodent models of cerebral ischemia (18-23), brain trauma (18), and neurodegenerative disease (18), as well as in a clinical study with stroke patients (24).

Here we show that TPO plays a previously unrecognized proapoptotic role in the brain.

Materials and Methods

All experiments were approved by and conducted in accordance with the regulations of the local Animal Care and Use Committee. For detailed information on all methods see Supporting Materials and Methods, which is published as supporting information on the PNAS web site.

Cell Culture. Primary hippocampal neuronal cultures were prepared from newborn Wistar-Imamichi rats, cultured under serum-free conditions (25, 26), and used for experiments after five days (purity: >95% neurons). Neuronal cell number and viability was assessed by trypan blue dye exclusion method. Spontaneous cell death rate in neuronal cultures at the time of experiments (5 days plus 15 h) was 17 ± 9% (mean ± SD, n = 40). Neuronal survival on experimental conditions is expressed as percent of spontaneous death rate in each particular experiment. Effects of growth factors on extracellular signal-regulated kinase (ERK)1/2 phosphorylation were tested on day 5 by incubation at 37°C for 10 min. For measuring long-term effects on neuronal growth, TPO (10 pM) was added to the culture medium at the time of plating and supplemented again on day 3.

Primary astrocyte cultures were prepared from the cortices of 1-day-old Wistar-Imamichi rats as described in refs. 25 and 26, yielding 98% positive staining for glial fibrillary acidic protein at 2-3 weeks, i.e., the time of experiments.

Hypoxic conditions were induced by purging an incubator with a mixture of 95% N2/5% CO2 (neurons) or 90% N2/10% CO2 (astrocytes) as described in refs. 25 and 26 and maintained for 15 h. Control experiments were simultaneously performed on the same cell batch under normoxic conditions. For drug treatments, see Supporting Materials and Methods.

Experiments in Vivo. For induction of hypoxic/ischemic brain injury, a standard method for immature rats (P14 Wistar-Imamichi) was used (27), combining common carotid artery ligation with hypoxia for 1 h (“moderate”) or 2 h (“severe”) in a standardized airtight chamber, flushed with 8% oxygen. RhTPO (R & D Systems) (1 nmol/kg), rhEPO (Janssen-Cilag, Neuss, Germany) (1.4 nmol/kg), or vehicle were injected i.p. immediately before, 24 h after, and 48 h after hypoxia exposure. Brains were removed 72 h after hypoxia. Hematoxylin-eosin stained sections were scored for structural damage in cortex and hippocampus ipsilateral to the ligation (two to three coronal sections per brain; 20 250× fields per section) from 0-3 (0, ≤1 apoptotic cell per field; 1, 1-3 apoptotic or dark shrunken neurons per field; 2, 4-10 apoptotic, dark shrunken, or eosinophilic neurons per field; 3, >10 apoptotic or eosinophilic, necrotic cells per field, cortical infarcts).

Expression of TPO and EPO During Brain Development. Forebrain and hindbrain from C57B6 mice fetuses [embryonic days (E)11, 13, 15, and 18], newborn [postnatal day (P) 0], 14-day-old (P14), and adult mice were used for mRNA or protein extraction. For densitometric analysis of TPO and EPO protein, Western blots were analyzed by National Institutes of Health image densitometry with α-tubulin III as an internal standard.

Expression Analysis by Quantitative Real-Time RT-PCR. First-strand cDNA was generated from total RNA by random priming (GIBCO/Pharmacia, Freiburg, Germany). Detailed information on rat and mouse primer pairs is available in Supporting Materials and Methods. PCR reactions were carried out on a LightCycler real-time PCR machine (Roche Molecular Biochemicals). RACE of rat TPOR/Mpl mRNA was performed to allow for the design of rat-specific primers in the 3′ UTR that are not sensitive toward differentially spliced isoforms (28).

Western Blotting. Protein extracts transferred to nitrocellulose membranes were incubated with rabbit/goat anti-TPO (Sigma), EPO, TPOR/Mpl, or EPOR antibodies (Santa Cruz Biotechnology), mouse anti-phospho-p44/42-ERK (Thr-202/Tyr-204) (New England Biolabs) or rabbit pan-ERK polyclonal antibody (New England Biolabs). Immunoreactive bands were visualized by using secondary antibodies coupled to horseradish peroxidase by enhanced chemoluminescence (Amersham Pharmacia).

Immunohistochemistry and in Situ Oligo Ligation (ISOL). For information on this subject, see Supporting Materials and Methods.

Statistical Analysis. Data, expressed as mean ± SEM in figures and text, were compared by ANOVA with post hoc planned comparisons or Duncan test, Kruskal-Wallis ANOVA with Mann-Whitney U test, or the Fisher exact probability test.


Inverted Pattern of TPO/TPOR and EPO/EPOR mRNA Expression in Brain Cells During Development and upon Hypoxia. In studies aimed originally at understanding the role of EPO in neuroprotection (21, 24, 25, 29), we noticed that TPO and TPOR also are widely expressed in the rodent CNS (Fig. 1a). We quantified and compared TPO and EPO mRNA in the mouse brain at various developmental stages by real time RT-PCR. Although TPO mRNA steady-state levels increased between E11 and adult, those of EPO decreased between E11 and E15 and stayed low thereafter (Fig. 1b).

Fig. 1.
Antagonistic gene expression of brain EPO and TPO systems. (a) RT-PCR illustrating presence of mRNA of TPO, EPO, and their receptors in fetal and adult rat tissues. HC, hippocampus; CX, cortex. (b) Quantitative PCR of the developing mouse forebrain demonstrates ...

Unexpectedly, when neuronal or astrocytic cultures were challenged with hypoxia (<1% O2) for 15 h, the steady-state levels of EPO mRNA and TPO mRNA became inversely related. In hippocampal neurons, EPO mRNA increased by ≈400%, whereas TPO mRNA decreased by ≈40% (Fig. 1c). Such a loss is remarkably high because a 40% decrease of TPO mRNA would be expected for the complete shutoff of TPO transcription in combination with a calculated mRNA half-life of 18.75 h. A similar dissociating response after hypoxia was observed for the expression of EPOR (increased) and TPOR (decreased) in cultured neurons. In astrocytes, however, both TPOR and EPOR mRNA were augmented by hypoxia (Fig. 1c).

Antagonistic TPO/TPOR and EPO/EPOR Protein Expression in Neurons and Astrocytes During Brain Development and After Hypoxia. At the protein level, TPO and TPOR expression was weak in the fetal brain and prominent in the juvenile hippocampus, with no obvious difference (in whole-tissue lysates) 6 h and 24 h after moderate hypoxia/ischemia (data not shown). Quantification of protein levels between E11 and adult stages indicated again a regulatory dissociation of the two growth factors: whereas TPO peaked in the adult brain, EPO decreased to a nearly undetectable level (Fig. 2a). For both TPO and TPOR, cultured neurons revealed weak but distinct specific labeling of cell bodies and processes (Fig. 2a, and Fig. 5 a-d, which is published as supporting information on the PNAS web site). In agreement with the mRNA data, immunostaining was reduced in neurons after 15 h of hypoxia, whereas that of EPO and EPOR was enhanced. In hypoxic astrocytes, staining of TPO was weaker and staining of TPOR was unchanged as compared with normoxia (Fig. 2b). Taken together, neuronal EPO and TPO systems are regulated in opposite ways, both during brain development and after hypoxia.

Fig. 2.
Antagonistic TPO/TPOR and EPO/EPOR protein expression in brain tissue and cultured neurons and astrocytes. (a) Densitometric analysis of 70-kDa (TPO) and 38-kDa (EPO) bands corresponding to known sizes of TPO (3) and EPO (22) in Western blots of developing ...

TPO and EPO Exert Opposite Actions on Neuronal Survival. We anticipated that TPO, like EPO, would enhance neuronal survival. Surprisingly, TPO, added to cultured hippocampal neurons for 15 h at concentrations as low as 10 pM, increased the spontaneous cell death rate by >60% (determined under normoxia) (Fig. 3a). Higher concentrations diminished this effect, and TPO lost its death promoting activity at 10 nM (i.e., the “toxic” concentration range of EPO, see below), resulting in a bell-shaped dose-response curve (Fig. 3a). To have a direct comparison of the effect of TPO on neuronal and hematopoietic cell survival, we performed additional experiments on murine bone marrow hematopoietic cell cultures by using the same preparation and concentrations of TPO as in the neuron cultures (Supporting Materials and Methods and Fig. 6, which is published as supporting information on the PNAS web site). In this preparation, TPO was clearly proproliferative and, interestingly, had a dose-response curve almost identical to that of the proapoptotic effect of TPO on neurons.

Fig. 3.
TPO and EPO exert opposite actions on neuronal survival. (a) Dose-response curves of TPO (filled circles) and EPO (open circles) effects on cell death rate in primary hippocampal neurons under normoxic conditions. n = 4; *, P < 0.05 compared with ...

At all concentrations tested, the cell-killing effect of TPO was completely antagonized by 100 pM EPO (Fig. 3b). Interestingly, granulocyte colony-stimulating factor (1 nM), another hematopoietic growth factor with neuroprotective potential (and a receptor of the same cytokine receptor superfamily) (30), prevented TPO-induced neuronal death. In addition, neurotrophins known to play a role in hippocampus (31), i.e., nerve growth factor, neurotrophin-3, and BDNF (each at 1 nM), salvaged neurons from TPO-mediated cell death (Fig. 3c).

The spontaneous death rate of hippocampal neurons is higher under hypoxic compared with normoxic conditions (21, 25). Remarkably, adding TPO to neuronal cultures that were kept under hypoxia (<1% O2; 15 h) did not further increase cell death (100% under normoxia, 148 ± 6% under hypoxia alone, 153 ± 7% under hypoxia plus 10 pM TPO; n = 5). This result may be explained by the demonstrated loss of neuronal TPOR under hypoxia (compare Figs. Figs.1c1c and and2b).2b). Growing neurons in the continuous presence of 10 pM TPO (“chronic condition” of 6 days instead of only 15 h) enhanced apoptotic cell death. Under TPO, the percentage of ISOL-positive apoptotic cells increased to 58 ± 1% compared with 31 ± 3% under continuous placebo treatment (P < 0.01, n = 4), suggesting that the phenomenon is not subject to any tolerance.

When applied to primary hippocampal neurons under normoxic conditions, EPO did not promote cell survival (25), consistent with low neuronal EPOR expression under normoxia (compare Figs. Figs.1c1c and and2b).2b). At higher concentrations (10 nM), however, EPO also had a cell death promoting effect (Fig. 3a). Thus, both EPO and TPO have bell-shaped dose-response curves.

Signaling Pathways Involved in TPO and EPO Action on Neuronal Survival. The deleterious effect of TPO on neuronal survival was unexpected, and the antagonistic response of EPO was unexplained. To determine the underlying mechanisms, we studied candidate second-messenger systems, previously associated with EPOR signaling (21, 23, 32), in primary hippocampal neurons. Both AG-490 (20 μM), an inhibitor of JAK2-transphosphorylation, and PD-98059 (50 μM), an inhibitor of ERK1/2, completely eliminated TPO-induced neuronal death, as did the caspase-3 inhibitor Ac-VAD-CHO (50 μM) (Fig. 3d). In contrast, the phosphatidylinositol 3-kinase (PI3K) inhibitor LY-294002 (100 μM) did not affect TPO-induced neuronal death. This finding indicates that the PI3K-Akt/protein kinase B pathway is not involved in death signaling through TPOR. However, this pathway is activated by EPO (21, 23) and is critical for the antagonistic effect of EPO toward TPO. In the presence of LY-294002, EPO was unable to prevent TPO-induced cell death (Fig. 3d).

Both TPO and EPO administration to hippocampal neurons caused an increased phosphorylation of ERK1/2 (Fig. 3e), which could be blocked by AG-490 (20 μM) (data not shown). We note that the Ras-ERK1/2 pathway has been implicated in both cell death and survival, whereas the PI3K-Akt/protein kinase B pathway acts only in a protective fashion (21, 23, 33). Therefore, the interaction of TPO and EPO on neuronal survival is best explained with a cross talk between two intracellular signaling pathways that associate with EPOR and TPOR. TPOR engagement activates Ras-ERK1/2, leading to neuronal apoptosis that can be prevented downstream by EPO through activation of PI3K-Akt/protein kinase B (summarized in Fig. 3f).

TPO-Induced Apoptotic Cell Death and Neuronal Differentiation Stage. TPO is known to inhibit the differentiation of cultured embryonic stem cells (34). We wondered whether the death-promoting effect of TPO could be linked to the differentiation of neuronal progenitors, e.g., during brain development or repair. We determined in hippocampal neuronal cultures the developmental stage of cells undergoing TPO-induced apoptosis by using a series of well known differentiation markers (nestin, β-tubulin III, and microtubule-associated-protein 2). When primary cultures were continuously exposed to TPO (10 pM until day 4 or day 6), nuclear condensation and expression of apoptosis markers were most frequently seen in early postmitotic neurons (β-tubulin III-positive) but never in precursors (nestin-positive) and rarely in mature neurons (microtubule-associated-protein 2-positive) (Fig. 3 g-i). In fact, TPO administration resulted in a significant decrease of β-tubulin III immunoreactive cells relative to the total number of cells (39 ± 6% control versus 27 ± 6% TPO, P < 0.05, n = 4).

In Vivo Effects of TPO in the Pathological Brain: “Gain-of-Function” Experiments. To investigate the in vivo role of TPO in the brain, we exploited our observation that TPO in neurons is reduced after hypoxia (compare Figs. Figs.1c1c and and2b).2b). We experimentally applied TPO to juvenile rats in combination with a standard model of hypoxic/ischemic brain damage (27). In this setting, unphysiologically high levels of TPO should result in an informative gain-of-function phenotype. The entire procedure involved a one-sided permanent carotid artery ligation, followed 2 h later by exposure to a moderate or severe hypoxia (i.e., 1 or 2 h of 8% oxygen). Immediately before hypoxia (and 24 h and 48 h thereafter), we gave i.p. injections of TPO (1 nmol/kg, or vehicle placebo). After 3 days, the tissue damage of the ipsilateral cortex and hippocampus was scored as described under Materials and Methods.

As expected, carotid artery ligation by itself (i.e., without hypoxia) did not cause obvious brain pathology. However, 50% of artery-ligated animals that additionally received TPO exhibited a damage score of ≥1 (compared with only 8% of placebo-treated rats; n = 10-12 per group; P < 0.05). The number of cleaved caspase-3-positive neurons in TPO versus placebo-treated rats was 114 ± 62 versus 1 ± 1 cells/square unit, respectively (n = 9 per group; P = 0.02). At that time point, there was not yet any difference in the number of apoptotic (ISOL-positive) cells in adjacent sections (9 ± 4 cells/square unit in TPO-treated versus 11 ± 7 cells/square unit in placebo-treated rats; n = 8-10 per group). These data suggest that TPO triggers a cell death program in distressed neurons in vivo. To detect potential systemic effects of TPO on the hematopoietic system or on thrombosis/fibrinolysis, platelet counts as well as plasma D-dimers were determined in juvenile rats (n = 4 per group) after three injections of TPO or placebo (at time points 0, 24, and 48 h) upon killing at 72 h (see Supporting Materials and Methods). In agreement with ref. 35, there was no difference among the groups with respect to these parameters (TPO, 555 ± 100 × 103/μl platelets and 0.04 ± 0.02 ng/μl D-dimers, n = 4; placebo, 559 ± 93 × 103/μl platelets and 0.03 ± 0.01 ng/μl D-dimers, n = 4).

In Vivo Effects of TPO Administration upon Hypoxic/Ischemic Brain Damage. When the carotid ligation protocol was followed by severe (2 h) hypoxia, the damage was extensive but about the same between TPO-treated and placebo-treated rats (data not shown). However, if exposed to only moderate (1 h) hypoxia after ligation, TPO-treated animals had a significantly higher damage score and incidence of apoptosis than placebo-treated rats (Fig. 4 a-c). This finding indicates that TPO worsens the outcome of moderate hypoxic/ischemic brain injury. In marked contrast, administration of EPO (3 × 1.4 nmol/kg i.p.) was neuroprotective, even under severe hypoxia (damage score 1.0 ± 0.3 in hypoxia plus EPO versus 1.8 ± 0.4 in hypoxia alone; n = 12 per group; P < 0.01). Also, apoptotic cells were reduced in EPO-treated compared with placebo-treated rats (62 ± 41 versus 378 ± 172 ISOL-positive cells per square unit in corresponding coronal sections; n = 8 per group; P < 0.05).

Fig. 4.
TPO increases tissue damage upon cerebral hypoxia/ischemia in juvenile rats. (a) Representative low-magnification photomicrograph depicting cortical (CX) and hippocampal (HC) areas of a placebo-treated (Left) and a TPO-treated (Center) rat 72 h after ...

Interestingly, the areas of cortical and hippocampal damage in TPO-treated rats showed a higher density of cells positive for TPOR than in placebo-treated rats and a significantly more intense specific TPOR immunostaining of individual cells (Figs. (Figs.4d4d and 5 e and f). In fact, TPOR immunoreactivity was frequently colocalized with two apoptosis markers, cleaved caspase-3 and ISOL (Fig. 7 a and b, which is published as supporting information on the PNAS web site).

Cells Preferentially Targeted by TPO in the Hypoxic Brain. Double-immunolabeling of TPOR-positive cells within the damaged cortical and hippocampal areas identified nestin-positive progenitors, β-tubulin III-positive (young postmitotic) neurons, and NF-200-positive (mature) neurons, as well as glial fibrillary acidic protein-positive glial cells (Fig. 7 c-f). Nestin-positive cells at the site of injury could be derived from adult stem cells and could be involved in tissue repair (36). In contrast to the widespread cellular TPOR immunoreactivity, TPOR-associated apoptosis was more cell type-specific; most frequently we found signs of apoptosis in β-tubulin III-positive neurons. In fact, TPO administration resulted in a significant decrease in β-tubulin III immunoreactivity, determined as staining density in the cortex (3 ± 1 versus 6 ± 2 arbitrary units/mm2, TPO versus placebo, respectively, P < 0.05, n = 11 per group). In contrast, NF-200-positive (mature) neurons were rarely labeled, and glial fibrillary acidic protein-positive astrocytes and nestin-positive precursors were never labeled with apoptotic markers.


Here we show that TPO/TPOR are expressed in the brain in astrocytes and neurons of various differentiation stages. TPO/TPOR and EPO/EPOR display an inverse pattern, the latter decreasing with increasing brain maturation but up-regulated and protective upon distress, and the former following the opposite rule. Whereas the protective role of the EPO/EPOR system in the brain has extensively been documented (18-22, 24, 29) the “detrimental” role of TPO/TPOR is entirely novel.

In fact, the most surprising finding of this work is that a prominent member of the cytokine type 1 receptor superfamily, TPOR/TPO, induces powerful proapoptotic signaling in cells of the nervous system. In contrast, related growth factors, EPO, granulocyte colony-stimulating factor, and growth hormones have all been shown to act in an antiapoptotic fashion on neurons (4, 21, 23, 25, 30, 37). Similarly, TPO has antiapoptotic effects in the hematopoietic system (1-3, 34). To the best of our knowledge, the previously uncharacterized proapoptotic function of TPO in the nervous system has not yet been described for any other cell type. In the brain, TPO-induced apoptosis appears to be restricted predominantly to maturating neuronal cells, suggesting a role for TPO in the selection of differentiated neurons.

The necessity of developmental apoptosis in the brain has long been recognized (38, 39). Among others, type β TGF has been shown to play a pivotal proapoptotic part in the developing nervous system of the chicken embryo (40). Now, TPO is another factor to be considered in the regulation of neuronal apoptosis. Remarkably, it acts at very low concentrations. We note that in human cerebrospinal fluid the level of TPO (≈1 pM by ELISA) (13) is ≈10% of the maximally effective proapoptotic concentration in vitro (10 pM TPO), and serum levels are up to 10-fold higher (7), emphasizing the physiological relevance of our findings.

Another somewhat puzzling finding is the bell-shaped dose-response curve for both EPO and TPO effects on cultured neurons. Considering the structural similarity of the receptor binding domain in EPO and TPO (3), we cannot exclude that EPO (when present at high concentrations) binds to the neuronal TPOR, and vice versa. This interaction could explain the cell death promoting effect of EPO at very high concentrations, i.e., 1,000 times the antiapoptotically effective dose, or the loss of the cell-killing effect of TPO at a similarly high concentration range. In experiments performed with transfected cell lines, however, such nonspecific binding of TPO to EPOR and EPO to TPOR has not been demonstrated (41, 42). Interestingly, a bell-shaped dose-response curve of EPO, with a gradual loss of its protective effect upon increasing concentrations, has been found in vivo by others (20) and ourselves (43).

Our in vivo gain-of-function model may help to understand why hypoxia physiologically down-regulates the neuronal TPO system. Under conditions of distress, the presumably regulatory proapoptotic action of TPO would not be desirable. TPO injected at that time point is detrimental for the damaged brain: It leads to enhanced expression of its own receptor before inducing cell death. However, despite widespread up-regulation of TPOR on immature precursors and mature neurons, induction of apoptosis is restricted to maturating neurons, whereas progenitors are largely unaffected. A reduction in maturating neurons in turn will compromise the brain's repair capacity, thereby increasing the damage.

We have no direct evidence yet that TPO, when given peripherally, crosses the intact blood-brain barrier, as has been unequivocally demonstrated for EPO (18, 29, 44). However, hypoxia/ischemia, as applied in the present study, is well known to compromise blood-brain barrier function (45). Also, the demonstrated effect of TPO on neurons, together with the absence of systemic changes affecting the hematopoietic or the fibrinolytic system, make it very likely that the effect of TPO is a direct action on the brain. An additional indirect effect on thrombotic capacity due to hyperactive platelets, however, cannot be entirely ruled out.

At a more general level, we found that two classical hematopoietic growth factors, EPO and TPO, are differentially regulated in the mammalian brain, where they influence neuronal survival in an antagonistic way. In contrast to the globally neuroprotective effects of EPO during early brain development (14, 15), the action of TPO is selectively proapoptotic and increasingly prominent during later brain development. The effect of TPO is overcome by neurotrophins such as neurotrophin-3, brain-derived neurotrophic factor and nerve growth factor. Based on these findings, we suggest a model in which decreasing EPO and rising TPO expression in the brain make it increasingly difficult for newly generated neurons to survive. The shifting ratio of proapoptotic TPO over neuroprotective EPO is then likely to make other forms of neuronal support necessary, for example by target-derived neurotrophins. Thus, brain TPO may contribute to the timing of neuronal selection by neurotrophins. We would predict that a complete lack of TPO or TPOR expression in mutant mice does not perturb gross brain development. However, a temporal delay in neurotrophic selection may alter the final size of neuronal populations. We note that a mutation of the human TPOR/Mpl1 gene, causing congenital amegakaryocytic thrombocytopenia, has also been associated with abnormal brain MRI findings (46). In mice, TPO and TPOR mutations are viable (1), but neurodevelopmental defects have not been analyzed. Studies with conditional mouse mutants of the EPO and TPO system would help to better understand the CNS functions of hematopoietic growth factors.

Supplementary Material

Supporting Information:


We thank Prof. Hans Thoenen (Max Planck Institute for Neurobiology, Martinsried, Germany) for carefully reading the manuscript; Sandra Hartung for technical assistance; and all members of the H.E. laboratory for stimulating scientific discussions and helpful criticism. This work was supported in part by the DFG Research Center for Molecular Physiology of the Brain.


Author contributions: A.-L.S. designed research; K.V. and A.-L.S. performed research; K.V. and A.-L.S. contributed new reagents/analytic tools; A.-L.S. analyzed data; and K.-A.N. and A.-L.S. wrote the paper.

This paper was submitted directly (Track II) to the PNAS office.

Abbreviations: En, embryonic day n; EPO, erythropoietin; EPOR, EPO receptor; ERK, extracellular signal-regulated kinase; ISOL, in situ oligo ligation; PI3K, phosphatidylinositol 3-kinase; Pn, postnatal day n; TPO, thrombopoietin; TPOR, TPO receptor.


1. Alexander, W. S. (1999) Int. J. Biochem. Cell Biol. 31, 1027-1035. [PubMed]
2. Geddis, A. E., Linden, H. M. & Kaushansky, K. (2002) Cytokine Growth Factor Rev. 13, 61-73. [PubMed]
3. Wolber, E. M. & Jelkmann, W. (2002) News Physiol. Sci. 17, 6-10. [PubMed]
4. Spivak, J. L., Pham, T., Isaacs, M. & Hankins, W. D. (1991) Blood 77, 1228-1233. [PubMed]
5. de Sauvage, F. J., Hass, P. E., Spencer, S. D., Malloy, B. E., Gurney, A. L., Spencer, S. A., Darbonne, W. C., Henzel, W. J., Wong, S. C., Kuang, W. J., et al. (1994) Nature 369, 533-538. [PubMed]
6. Lok, S., Kaushansky, K., Holly, R. D., Kuijper, J. L., Lofton-Day, C. E., Oort, P. J., Grant, F. J., Heipel, M. D., Burkhead, S. K., Kramer, J. M., et al. (1994) Nature 369, 565-568. [PubMed]
7. Schmitt, M., Gleiter, C. H., Nichol, J. L., Pralle, L., Hasselblatt, M., Poser, W. & Ehrenreich, H. (1999) Thromb. Haemostasis 82, 1422-1427. [PubMed]
8. Vigon, I., Mornon, J. P., Cocault, L., Mitjavila, M. T., Tambourin, P., Gisselbrecht, S. & Souyri, M. (1992) Proc. Natl. Acad. Sci. USA 89, 5640-5644. [PMC free article] [PubMed]
9. Bartley, T. D., Bogenberger, J., Hunt, P., Li, Y. S., Lu, H. S., Martin, F., Chang, M. S., Samal, B., Nichol, J. L., Swift, S., et al. (1994) Cell 77, 1117-1124. [PubMed]
10. Columbyova, L., Loda, M. & Scadden, D. T. (1995) Cancer Res. 55, 3509-3512. [PubMed]
11. Li, B., Pan, H., Winkelmann, J. C. & Dai, W. (1996) Blood 87, 5382-5384. [PubMed]
12. Nakanishi, K., Tajima, F., Osada, H., Kato, T., Miyazaki, H., Kawai, T., Torikata, C., Suga, T., Takishima, K., Aurues, T., et al. (1999) Lab. Invest 79, 679-688. [PubMed]
13. Dame, C., Wolber, E. M., Freitag, P., Hofmann, D., Bartmann, P. & Fandrey, J. (2003) Dev. Brain Res. 143, 217-223. [PubMed]
14. Yu, X., Shacka, J. J., Eells, J. B., Suarez-Quian, C., Przygodzki, R. M., Beleslin-Cokic, B., Lin, C. S., Nikodem, V. M., Hempstead, B., Flanders, K. C., et al. (2002) Development (Cambridge, U.K.) 129, 505-516.
15. Shingo, T., Sorokan, S. T., Shimazaki, T. & Weiss, S. (2001) J. Neurosci. 21, 9733-9743. [PubMed]
16. Knabe, W., Knerlich, F., Washausen, S., T., K., Sirén, A.-L., Brunnett, G., Kuhn, H.-J. & Ehrenreich, H. (2004) Anat. Embryol. 207, 503-512. [PubMed]
17. Chen, E. Y., Fujinaga, M. & Giaccia, A. J. (1999) Teratology 60, 215-225. [PubMed]
18. Brines, M. L., Ghezzi, P., Keenan, S., Agnello, D., de Lanerolle, N. C., Cerami, C., Itri, L. M. & Cerami, A. (2000) Proc. Natl. Acad. Sci. USA 97, 10526-10531. [PMC free article] [PubMed]
19. Erbayraktar, S., Grasso, G., Sfacteria, A., Xie, Q. W., Coleman, T., Kreilgaard, M., Torup, L., Sager, T., Erbayraktar, Z., Gokmen, N., et al. (2003) Proc. Natl. Acad. Sci. USA 100, 6741-6746. [PMC free article] [PubMed]
20. Sakanaka, M., Wen, T. C., Matsuda, S., Masuda, S., Morishita, E., Nagao, M. & Sasaki, R. (1998) Proc. Natl. Acad. Sci. USA 95, 4635-4640. [PMC free article] [PubMed]
21. Sirén, A. L., Fratelli, M., Brines, M., Goemans, C., Casagrande, S., Lewczuk, P., Keenan, S., Gleiter, C., Pasquali, C., Capobianco, A., et al. (2001) Proc. Natl. Acad. Sci. USA 98, 4044-4049. [PMC free article] [PubMed]
22. Bernaudin, M., Marti, H. H., Roussel, S., Divoux, D., Nouvelot, A., MacKenzie, E. T. & Petit, E. (1999) J. Cereb. Blood Flow Metab. 19, 643-651. [PubMed]
23. Ruscher, K., Freyer, D., Karsch, M., Isaev, N., Megow, D., Sawitzki, B., Priller, J., Dirnagl, U. & Meisel, A. (2002) J. Neurosci. 22, 10291-10301. [PubMed]
24. Ehrenreich, H., Hasselblatt, M., Dembowski, C., Cepek, L., Lewczuk, P., Stiefel, M., Rustenbeck, H.-H., Breiter, N., Jacob, S., Knerlich, F., et al. (2002) Mol. Med. 8, 495-505. [PMC free article] [PubMed]
25. Lewczuk, P., Hasselblatt, M., Kamrowski-Kruck, H., Heyer, A., Unzicker, C., Sirén, A. L. & Ehrenreich, H. (2000) NeuroReport 11, 3485-3488. [PubMed]
26. Hasselblatt, M., Lewczuk, P., Loffler, B. M., Kamrowski-Kruck, H., von Ahsen, N., Sirén, A. L. & Ehrenreich, H. (2001) Glia 34, 18-26. [PubMed]
27. Towfighi, J., Mauger, D., Vannucci, R. C. & Vannucci, S. J. (1997) Dev. Brain Res. 100, 149-160. [PubMed]
28. Matz, M., Shagin, D., Bogdanova, E., Britanova, O., Lukyanov, S., Diatchenko, L. & Chenchik, A. (1999) Nucleic Acids Res. 27, 1558-1560. [PMC free article] [PubMed]
29. Ehrenreich, H., Degner, D., Meller, J., Brines, M., Behe, M., Hasselblatt, M., Woldt, H., Falkai, P., Knerlich, F., Jacob, S., et al. (2004) Mol. Psychiatry 9, 42-54. [PubMed]
30. Schabitz, W. R., Kollmar, R., Schwaninger, M., Juettler, E., Bardutzky, J., Scholzke, M. N., Sommer, C. & Schwab, S. (2003) Stroke 34, 745-751. [PubMed]
31. Huang, E. J. & Reichardt, L. F. (2001) Annu. Rev. Neurosci. 24, 677-736. [PMC free article] [PubMed]
32. Digicaylioglu, M. & Lipton, S. A. (2001) Nature 412, 641-647. [PubMed]
33. Downward, J. (1998) Curr. Opin. Genet. Dev. 8, 49-54. [PubMed]
34. Xie, X., Chan, R. J. & Yoder, M. C. (2002) FEBS Lett. 529, 361-364. [PubMed]
35. Mouthon, M. A., Gaugler, M. H., Van der Meeren, A., Vandamme, M., Gourmelon, P. & Wagemaker, G. (2001) Exp. Hematol. 29, 30-40. [PubMed]
36. Arvidsson, A., Collin, T., Kirik, D., Kokaia, Z. & Lindvall, O. (2002) Nat. Med. 8, 963-970. [PubMed]
37. Shin, D. H., Lee, E., Kim, J. W., Kwon, B. S., Jung, M. K., Jee, Y. H., Kim, J., Bae, S. R. & Chang, Y. P. (2004) Neurosci. Lett. 354, 64-68. [PubMed]
38. Roth, K. A. & D'Sa, C. (2001) Ment. Retard. Dev. Disabil. Res. Rev. 7, 261-266. [PubMed]
39. Oppenheim, R. W. (1991) Annu. Rev. Neurosci. 14, 453-501. [PubMed]
40. Krieglstein, K., Richter, S., Farkas, L., Schuster, N., Dunker, N., Oppenheim, R. W. & Unsicker, K. (2000) Nat. Neurosci. 3, 1085-1090. [PubMed]
41. Yamada, M., Komatsu, N., Kirito, K., Kashii, Y., Tomizuka, H., Okada, K., Endo, T., Fukumaki, Y., Shinjo, K., Abe, K., et al. (1998) Cell Growth Differ. 9, 487-496. [PubMed]
42. Broudy, V. C., Lin, N. L., Sabath, D. F., Papayannopoulou, T. & Kaushansky, K. (1997) Blood 89, 1896-1904. [PubMed]
43. Weishaupt, J. H., Rohde, G., Pölking, E., Sirén, A.-L., Ehrenreich, H. & Bähr, M. (2004) Invest. Ophthalmol. Vis. Sci. 45, 1514-1522. [PubMed]
44. Martinez-Estrada, O. M., Rodriguez-Millan, E., Gonzalez-De Vicente, E., Reina, M., Vilaro, S. & Fabre, M. (2003) Eur. J. Neurosci. 18, 2538-2544. [PubMed]
45. Ikeda, T., Xia, X. Y., Xia, Y. X. & Ikenoue, T. (1999) Brain Res. Dev. Brain Res. 117, 53-58. [PubMed]
46. Ihara, K., Ishii, E., Eguchi, M., Takada, H., Suminoe, A., Good, R. A. & Hara, T. (1999) Proc. Natl. Acad. Sci. USA 96, 3132-3136. [PMC free article] [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...