HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints/Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Birkmann, J. Articles by Gallmeier, W. M. Search for Related Content PubMed PubMed Citation Articles by Birkmann, J. Articles by Gallmeier, W. M.

Effects of Recombinant Human Thrombopoietin Alone and in Combination with Erythropoietin and Early-Acting Cytokines on Human Mobilized Purified CD34⁺ Progenitor Cells Cultured in Serum-Depleted Medium

Institute for Med. Oncology and Hematology, Med. Clinic 5, Nürnberg City Hospital, Nürnberg, Germany

Key Words. Thrombopoietin • Human • Progenitor cells • Megakaryocytes • Lineage specificity • Maturation • Serum-free culture

Dr. Josef Birkmann, Inst. f. Med. Onkologie und Hämatologie, Med. Klinik 5, Flurstrasse 17, D-90340 Nürnberg, Germany.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
The effects of recombinant thrombopoietin (TPO) alone and in combination with erythropoietin (EPO) and early-acting cytokines such as interleukin 3 (IL-3), stem cell factor (SCF) and GM-CSF on highly purified mobilized human CD34⁺ progenitor cells were studied in a serum-depleted culture system. Eight leukapheresis samples were cultured for seven days and analyzed; aliquots were replated and re-evaluated on day 12. Three-color flow cytometry was used together with morphologic analysis to determine proliferation and megakaryocytic or erythroid maturation.

TPO alone was sufficient for cell survival and proliferation in serum-depleted medium. In the absence of other growth factors, almost all CD34⁺ cells differentiated along the megakaryocytic pathway within 12 days. Concomitantly, the progenitor cells gradually acquired the morphologic features of mature megakaryocytes. After exposure to TPO for one week, 50% of the cells still expressed CD34; by day 12 the remaining CD34⁺ cells (11%) were all coexpressing CD41. TPO alone did not support proliferation of glycophorin-A-positive cells.

The addition of TPO to early-acting cytokines (EPO, GM-CSF, SCF and/or IL-3) not only increased the overall megakaryocyte expansion, but also generated a different maturation pattern of the CD41⁺ megakaryocyte progenitors. It further doubled the number of erythroid cells and c-kit⁺ cells in the second week of culture. Interestingly, the overall number of CD34⁺ cells was increased about fivefold when TPO was added to the early-acting cytokines, with a marked expansion of the CD34⁺/CD41⁺ and CD34⁺/CD117⁺ subpopulations. TPO can augment the pool of committed progenitors, thereby increasing the number of its own target cells and the number of EPO-responsive cells. These properties make TPO an interesting cytokine for the ex vivo expansion of human progenitor cells.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
In 1992, a novel member of the cytokine receptor family, c-Mpl, was characterized by Vigon and coworkers [1]. The presence of c-Mpl mRNA in both primitive and mature cells committed to the megakaryocytic lineage [2], and the fact that antisense oligonucleotides abrogated megakaryocyte colony formation [3], strongly suggested that the ligand of c-Mpl was a megakaryopoietic cytokine. Two years later, five groups using different strategies reported on the purification and cloning of a new cytokine, c-Mpl ligand: de Sauvage (Genentech; South San Francisco, CA) purified the factor from aplastic porcine plasma and tested the activity on human leukapheresis products [4], Bartley (Amgen; Thousand Oaks, CA) purified it from aplastic canine plasma and tested it on human CD34⁺ cells enriched for megakaryocyte progenitors [5]. The group of Lok (ZymoGenetics; Seattle, WA) and Kaushansky (University of Washington School of Medicine; Seattle, WA) tested the effects on human cell lines that express c-Mpl [6, 7]. Kato's group (Kirin Brewery; Maebashi, Japan) purified the factor directly from the plasma of sublethally irradiated rats [8, 9]. This group used rat gpIIb/IIIa⁺ cells from the bone marrow as target cells, and ¹⁴C-5-hydroxytryptamine incorporation and megakaryocyte colony formation as an indicator system. Kuter's group purified c-Mpl ligand from the plasma of sheep [10]. Meanwhile, the human thrombopoietin (TPO) gene has been cloned and mapped to chromosome 3q27-28 [11, 12]. The terms TPO and c-Mpl ligand refer to the natural human glycoprotein, whereas the recombinant protein is also called megakaryocyte growth and development factor (MGDF). The history of TPO and the potential therapeutic application of this molecule have recently been reviewed by Lok [13], Hunt [14] and Kaushansky [15].

When given to carboplatin-treated [16] and/or sublethally irradiated mice, TPO significantly reduces the depth and duration of thrombocytopenia and markedly augments megakaryopoiesis in the bone marrow [17]. Similar effects have been shown in rhesus monkeys [18]. It is not clear whether the effects observed in vivo are mainly caused by TPO itself or represent a synergistic effect of TPO and other cytokines on various progenitor cell populations. The published in vitro studies using bone marrow or CD34⁺ peripheral blood progenitor cells demonstrated an increase of megakaryocyte colony-forming units (CFU-Meg), in vitro maturation of megakaryocytes, increase in megakaryocyte ploidy [19] and the appearance of functionally active platelets [20]. Moreover, data indicating a role for TPO in direct platelet activation have most recently been published [21]. However, nearly all of the in vitro studies describing the effects of TPO have been performed with serum-containing media, so that it cannot be excluded that the observed effects attributed to TPO are at least partly caused by interactions with other factors present in the serum.

Recently, the possibility of culturing hematopoietic cells in serum-free medium [22] has provided biochemically defined culture conditions. Several formulae for serum-depleted media supporting clonogenic growth in semisolid media [23, 24], liquid cultures of bone marrow [25] or expansion of CD34⁺ cells in liquid culture [26] have been published. In addition, based on various techniques using monoclonal anti-CD34 antibodies, the CD34⁺ progenitor cells collected by leukapheresis can be highly enriched to a purity of more than 95%, thus providing ideal targets to assess the effects of cytokines in early hematopoiesis. When highly enriched CD34⁺ cells are cultured in serum-free medium, both the cell type and the culture conditions are well defined.

We used highly purified human mobilized CD34⁺ progenitor cells from leukapheresis products to study the effects of TPO alone and in combination with other cytokines on early hematopoiesis in a serum-depleted liquid culture system. Using multiparameter flow cytometry, the phenotypic development of the cells with regard to megakaryocytic commitment and maturation versus maintenance of immature markers was monitored over a period of 12 days. Concomitantly, cytospins were prepared at the same intervals to determine the corresponding morphology.

In addition to stimulation with TPO alone, four "early-acting cytokines," each of which is said to promote megakaryopoiesis, were chosen to evaluate additive or synergistic effects on megakaryopoiesis and erythropoiesis: erythropoietin (EPO), stem cell factor (SCF), interleukin 3 (IL-3) and GM-CSF. In previous experiments in our laboratory, the mixture of these four cytokines best expanded the immature CD34⁺ cell population. In colony-forming assays using human bone marrow or cord blood, TPO enhances erythropoiesis in the presence of EPO [27]. Conversely, TPO and EPO are reported to act synergistically on murine megakaryocyte colony formation [28]. GM-CSF strongly enhances megakaryopoiesis in vitro [19, 29]. IL-3 has been shown to promote megakaryocyte proliferation and maturation [29] and to possess some activity on megakaryopoiesis in patients with normal hematopoiesis and bone marrow failure [30]. In the CFU-Meg assay, TPO showed an additive response with IL-3, whereas a synergistic effect was reported for TPO and SCF [28, 31].

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Collection of Peripheral Blood Progenitor Cells
Patients for whom high-dose chemotherapy with autologous stem cell transplantation was planned were prepared for leukapheresis with chemotherapy followed by filgrastim, 300 µg/d. When the proportion of CD34⁺ cells in the peripheral blood exceeded 1.0% of total leukocytes, patients were subjected to leukapheresis using an AS-104 cell separator (Fresenius AG; St. Wendel, Germany). During each apheresis session, 7.2 liters of blood were processed yielding a separation product of 9 to 15 billion low-density mononuclear cells. After removing platelet-rich plasma, 10 ml of the apheresis product were further processed for enrichment of CD34⁺ cells. Eight separation products from eight different patients were used for this study. Three of these patients had acute leukemia in second remission, three suffered from high-grade malignant lymphoma, one patient had Hodgkin's disease and one patient had a germ cell tumor.

Isolation of CD34⁺ Cells
Purification of CD34⁺ cells was performed according to the method described by Nichol [32] with some modifications. An aliquot of the apheresis product calculated to contain about six to nine million CD34⁺ cells was separated over a Ficoll density gradient (Biochrom; Berlin, Germany). The interface was washed and further processed with the CD34 Progenitor Cell Isolation Kit (Miltenyi Biotec GmbH; Bergisch Gladbach, Germany) according to the manufacturer's instructions. Briefly, after blocking nonspecific binding with human polyclonal IgG, the cells were incubated at 4°C with a class II anti-CD34 monoclonal antibody (QBEND/10, mouse IgG1) and colloidal superparamagnetic microbeads and run over a MiniMACS separation column. During this process, the CD34⁺ cells coated with the monoclonal antibody attached to the microbeads are retained through a strong magnetic field, while the negative cells are allowed to pass through the column. After removing the column from the magnet, the CD34⁺ cells are eluted. Reanalysis of the enriched cell fraction, performed with a class I anti-CD34 antibody, (HPCA-2, clone 8G12, Becton Dickinson; Heidelberg, Germany), revealed a purity of 88.2% to 98% CD34⁺ cells (mean 91%). The recovery rate was 50%-65%, and the trypan-blue dye exclusion test showed a viability of 95%-99%.

Cytokines
Recombinant TPO, E. coli-derived (MGDF), was kindly provided by Jakob Bogenberger (Amgen, Inc.). Initially, experiments were carried out to establish a dose-response curve using TPO at final concentrations between 0.30 and 5.00 ng/ml. In all further assays, a final concentration of 2.5 ng/ml was chosen. EPO was provided by CILAG Biotech (Sulzbach, Germany) and added to the cultures at a final concentration of 2 U/ml. GM-CSF was provided by Behringwerke (Marburg, Germany) and given at a concentration of 10 ng/ml. SCF was provided by ICC (Ismaning, Germany) and added at a concentration of 20 ng/ml. IL-3 was obtained from Sandoz AG (Nürnberg, Germany); the final concentration in the culture medium was 2 ng/ml.

Cell Culture
We used Iscove's modified Dulbecco's medium (IMDM) supplemented with 100 U penicillin/100 µg streptomycin, 2 µM L-glutamin, (Biochrom), and 20% v/v BIT (StemCell Technologies Inc.; Vancouver, BC, Canada) as culture medium. This medium had yielded in our hands a two- to threefold higher overall cell proliferation in liquid culture than did the previously used medium supplemented with 10% fetal calf serum. BIT is composed of bovine serum albumin, 50 mg/ml, bovine pancreatic insulin, 50 µg/ml, and human transferrin, 1 mg/ml.

The purified progenitor cells were cultured in aliquots of 300,000 cells suspended in 1 ml culture medium in 24-well microtiter plates (Greiner GmbH; Frickenhausen, Germany). The high cell number was chosen in order to obtain enough cells for detailed reanalysis of subpopulations on day 7 and day 12. For each specimen, eight wells with different cytokine combinations were prepared as follows: A) medium control; B) TPO; C) TPO + EPO; D) TPO + EPO + GM-CSF; E) TPO + EPO + GM-CSF + SCF; F) TPO + EPO + GM-CSF + IL-3; G) TPO + EPO + GM-CSF + SCF + IL-3; and H) EPO + GM-CSF + SCF + IL-3 without TPO.

Cells were incubated at 37°C and 5% CO₂ and 100% moisture for 12 days in a Haereus incubator (Haereus Instruments; Hanau, Germany). Every three days, half of the culture medium was removed and replaced by fresh medium containing BIT and the respective cytokines. On day 7, an aliquot of cells from each well was analyzed for cell number, viability, marker expression and morphology. The concentration of the remaining cells was readjusted to 3 x 10⁵/ml and the cells were resuspended in fresh medium enriched with BIT and restimulated with the respective cytokines. On day 12, the cells from each well were again reanalyzed for cell number, viability, marker expression and morphology.

Evaluation of Growth and Differentiation
Before collecting the cells for reanalysis on days 7 and 12, each well was screened for cell number and gross morphology using an inverted microscope with bright-field and phase-contrast illumination (Diavert, Leitz, Wetzlar, Germany). Cell viability was assessed with the trypan-blue exclusion test. Two cytospins were prepared from each well. For three-color flow cytometry, the cells were washed with phosphate-buffered saline (PBS) and incubated at 4°C with different sets of the following monoclonal antibodies:

For each well, the proliferation rate (PR₁) on day 7 was calculated as the number of viable cells on day 7, divided by the cell number on day 0 (300,000/well). On replating on day 7, the cell numbers were readjusted to 300,000/well; consequently the proliferation rate PR₂ for the interval from day 7 to day 12 was defined as number of viable cells on day 12, divided by 300,000. The overall proliferation index (PR₃) on day 12 was defined as the product of PR₁ and PR₂. The absolute numbers of cells positive for a certain marker in each well on day 0 was calculated as 300,000x (percentage of positive cells); for day 7, this value was multiplied with PR₁, and for day 12, the respective value was multiplied with PR₃.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Marker Expression of Mobilized CD34⁺ Cells before Culturing (Day 0)
The CD34 molecule is a highly glycosylated transmembrane protein which is mainly found on hematopoietic stem cells and precursor cells [33]. At the time of leukapheresis, between 1% and 2.8% of the harvested cells expressed CD34. Following purification by immunomagnetic microbeads between 88.2% and 98% of the cells (mean 91%) were CD34⁺.

CD41, also named glycoprotein IIb, is a protein specific for the megakaryocytic lineage. It is expressed throughout the maturation process from early committed progenitor cells to mature platelets [34, 35]. It functions as a receptor for fibrinogen, fibronectin and von Willebrand factor, and forms a calcium-dependent complex with glycoprotein IIIa, clustered as CD61. Immediately after leukapheresis, CD41 was found on 24% of the CD34⁺ cells (range 11% to 37.5%), while CD61 was only found on 6% (range 2% to 11.8%). All CD61⁺ cells coexpressed CD41. Out of the CD42 family, the CD42b molecule was chosen as a marker for more mature megakaryocytes. CD42b or glycoprotein Ib-a acts mainly as a receptor for von Willebrand factor on megakaryocytes, proplatelets [36], mature platelets and endothelial cells [35]. CD42b was found on 0.2% of the CD34⁺ cells in the fresh leukapheresis product. Although the proportion of CD34⁺/CD41⁺ cells was rather high in some samples, it is most unlikely that this coexpression of CD41 on progenitor cells originates from platelets adhering to the CD34⁺ cells [37]. Mature platelets express CD42b, but we found virtually no CD34⁺/CD41⁺/CD42b⁺ triple-positive cells in our specimens.

The c-kit molecule, clustered as CD117, is the receptor for the SCF [38]. SCF is known to support the cycling of CD34⁺ cells in the absence of other cytokines [39] and to act as a survival factor for progenitor cells [40]. After purification, 6% of the mobilized cells coexpressed CD117, and 2.2% were CD41/CD117 double-positive.

Glycophorin A (GP-A) is considered to be specific for the erythroid lineage. Only very few of the enriched CD34⁺ progenitor cells (0.3%) coexpressed GP-A at the time of harvesting.

Without the addition of any cytokines to the culture medium, almost all cells died within one week. The mere addition of BIT to the medium without growth factors was not sufficient for survival of mobilized cells.

Effects of TPO without the Addition of Other Cytokines
In initial experiments, purified progenitor cells were exposed to TPO for seven days at different concentrations ranging from 0.3 ng/ml to 5.0 ng/ml in order to find out the appropriate dose. In our setting, a nonlinear dose-response curve was obtained (Fig. 1) not only with respect to growth of CD41⁺ and CD42b⁺ cells, but also of CD34⁺ cells. A concentration of 2.5 ng/ml was chosen for subsequent experiments.

View larger version (19K): [in this window] [in a new window]   Figure 1. Dose-response curve for TPO. Samples of purified mobilized human CD34+ progenitor cells (purity >97%) from three different donors were cultured for seven days in IMDM supplemented with BIT, Pen/Strep and glutamine at different concentrations of TPO. For each experiment, 300,000 cells were cultured. On day 7, the expression of CD41 (FITC), CD42b (PE) and CD34 (Cy5) was measured by three-color flow cytometry. For all three markers, a dose-dependency was observed. A double Y axis was used to demonstrate the response of the small CD42b+ subpopulation (right Y axis sale) in comparison to the CD41+ and CD34+ fraction (left Y axis scale).

View larger version (20K): [in this window] [in a new window]   Figure 2. Proliferation rates of purified CD34+ progenitor cells following stimulation with different cytokine combinations. For each assay, 300,000 purified progenitor cells were cultured in serum-free medium as described. On day 7, the number of viable cells was determined and the proliferation rate (PR) was calculated by dividing the cell number by 300,000 (black bars). Cells were repleted on day 7 after readjusting the cell number to 300,000/well. On day 12, the number of viable cells was again measured and the proliferation rate of the second week, PR2, was calculated by dividing each of these figures by 300,000 (white bars). For the whole culture period, an operational overall index PR3 was defined as PR1 x PR2. Results are shown as mean values of eight samples. Error bars indicate SEM.

After one week exposure to TPO, still 50% of the cells expressed CD34, and by day 12, CD34 was found on 11% of the cells. All other cytokine combinations tested yielded a much lower proportion of CD34⁺ cells on day 7. By day 12, all CD34⁺ cells coexpressed CD41. About half of this megakaryocytic committed progenitor cell population (6%) further coexpressed CD61. In terms of absolute counts, TPO alone increased the overall number of CD34⁺ cells in the first week to 1.5 million cells; this value returned to 714,000 counts on day 12. The subpopulation of CD41⁺CD34⁺ cells increased by factor 11 from 68,000 to 734,000 in the first week; nearly the same counts were measured on day 12.

The proportion of cells expressing the SCF receptor, c-kit or CD117, rose initially from 6% to 18% in the first week and remained unchanged on day 12. The CD117⁺CD34⁺ cells transiently rose to 14% on day 7, the highest value of all assays, but fell to 2% on day 12. The absolute counts of CD117⁺ cells increased 29-fold in the first week; only one-fifth of this population coexpressed CD41, while in the second week nearly all coexpressed CD41.

Culturing the CD34⁺ cells with TPO alone did not stimulate GP-A⁺ erythroid progenitors.

Table 1 summarizes the changes in marker expression following exposure to different cytokine combinations.

View larger version (310K): [in this window] [in a new window]   Figure 3. Morphology of purified progenitor cells stimulated with TPO. A) After seven days of culture with TPO alone, many small to medium-sized cells are seen in the inverted microscope, corresponding to immature megakaryocytic cells (cf. Figure 4A). A small amount of very large cells corresponding to mature megakaryocytes can already be determined. (Leitz Diavert, bright field illumination, 125x). B) On day 12, i.e., after replating the cells on day 7 and another five days of culture, the picture is dominated by "giant" cells, which represent mature polyploid megakaryocytes. A certain amount of cell debris is also found in the medium. (Lietz Diavert, bright field illumination, 125x).

View larger version (192K): [in this window] [in a new window]   Figure 4. Cytospins of progenitor cells cultured with TPO. A) After seven days of culture with TPO alone, most of the progenitor cells have developed into immature megakaryoblasts with one or two eccentric nuclei and a prominent perinuclear Gogli apparatus. Occasionally, polynucleated megakaryocytes are seen. (Leitz Orthoplan, 600x). B) On cytospins prepared on day 12 following stimulation with TPO abundant mature polynucleated megakaryocytes are seen, some of which show cytoplasmic protrusions. (Leitz Orthplan, 600x).

View larger version (33K): [in this window] [in a new window]   Figure 5. Expression of CD41 (FITC), CD42b (PE) and CD34 (Cy5) in small and large cells following stimulation with TPO for seven days. Upper row: in the first scattergram, a bitmap (gate A) is drawn around the small cells with little side scatter. Nearly all of these cells are CD41+ at a medium fluorescence intensity (second scattergram), and very few (2%) coexpress CD42b with low fluorescence intensity. About half of the CD41+ cells coexpress CD34 (third scattergram), also with low fluorescence. The CD42b+ cells are negative for CD34 (fourth scattergram). Most of the cells in gate A are immature cells committed to the megakaryocytic lineage. Lower row: this series shows the same measurements for the large cells in gate B (first scattergram). These cells are all strongly CD41+, about 25% coexpress CD42b (second scattergram). Only a few of these large CD41+ cells (6%) coexpress CD34 (third scattergram). Again, the CD42b+ cells are negative for CD34 (fourth scattergram). Most of the cells in gate B are megakaryocytes at different stages of maturation, having lost the CD34 marker.

View larger version (27K): [in this window] [in a new window]   Figure 6. Marker expression on day 7 following stimulation with different combinations of cytokines. An aliquot of the cells was analyzed for expression of CD41, CD34, CD117 and Glycophorin A after seven days of culture. Each diagram represents the mean absolute number of viable cells x1000 (Y axis), which express the marker indicated in the diagram title. Absolute numbers were calculated as described in Materials and Methods. The error bars correspond to SD. The cytokine combinations indicated below the X axis apply to both rows. White bars: cell numbers on day 0; black bars: cell numbers on day 7.

View larger version (24K): [in this window] [in a new window]   Figure 7. Marker expression on day 12 compared to day 7. As in Figure 6, each diagram represents the mean absolute number of viable cells x1000 (Y axis), which are positive for the markers indicated in the diagram title. Absolute numbers were calculated as described in Materials and Methods. The error bars represent SD. The cytokine combinations indicated below the X axis apply to both rows. Gray bars: cell numbers on day 0; dark bars: cell numbers on day 7; light bars: cell numbers on day 12.

View larger version (28K): [in this window] [in a new window]   Figure 8. CD41 subpopulations on day 7 following stimulation with different combinations of cytokines. As in Figure 6, each diagram represents the mean absolute number of viable cells x1000 (Y axis), which are double-positive for the markers indicated in the diagram title. The error bars represent the respective SD. The cytokine combinations indicated below the X axis apply to both rows. White bars: cell number on day 0; gray bars: cell numbers on day 7.

View larger version (25K): [in this window] [in a new window]   Figure 9. CD41 subpopulations on day 12 compared to day 7. As in Figure 8, each diagram represents the mean absolute number of viable cells x1000 (Y axis), which are double-positive for the markers indicated in the diagram title. Absolute numbers were calculated as described in Materials and Methods. The error bars correspond to SD. The cytokine combinations indicated below the X axis apply to both rows. Gray bars: cell numbers on day 0; dark bars: cell numbers on day 7; light bars: cell numbers on day 12.

The proportion of CD41⁺ cells was maintained in the first week (26%), but dropped to one-third of the initial value in the second week. There was only a transient marginal increase of CD41⁺CD42b⁺ cells on day 7 from 0.18% to 0.53%. Compared to TPO alone, the combination of TPO + EPO generated much less cells with megakaryocytic markers, especially in the second week of culture.

In contrast to stimulation with TPO alone, nearly all cells lost CD34 when cultured with TPO + EPO. The proportion of CD41⁺CD34⁺ cells decreased to 10% and <1%, respectively. The CD117⁺ cells had tripled on day 7, but decreased to initial values on day 12, as did the CD41⁺CD117⁺ cells.

On the other hand, the proportion of GP-A⁺ cells rose steeply from 0.3% on day 0 to mean 50% on day 7 and to 92% on day 12, i.e., the combination of TPO + EPO committed about the same proportion of cells to the erythroid lineage as did TPO alone without EPO to the megakaryocytic lineage. The GP-A⁺CD34⁺ cells comprised only 1.5%-2%, suggesting that the acquisition of the erythroid marker GP-A means loss of the CD34 molecule. The absolute counts of GP-A⁺ cells rose to two million in the first week and to 12 million in the second week. With respect to morphology, megakaryocytes at various stages of maturation and small cells forming three-dimensional clusters could be identified, the latter corresponding to normoblasts.

Addition of Early-Acting Cytokines (GM-CSF, SCF, IL-3)
The overall proliferation rate increased when earlyacting cytokines were added, from eightfold (GM-CSF) to 14-fold (GM-CSF + SCF + IL-3).

Compared to TPO alone, the combination of TPO + EPO + GM-CSF yielded about the same counts of total CD41⁺ cells and CD41⁺/CD61⁺ cells, but less than half the amount of immature CD41⁺/CD34⁺ cells (Figs. 8 and 9). With further addition of SCF, the absolute counts of CD41⁺ cells, CD61⁺ cells and double-positive cells exceeded the counts obtained with TPO alone, both on day 7 and on day 12. The amount of more mature CD42b⁺ cells was about the same as with TPO alone.

When SCF was replaced by IL-3, the absolute counts of CD41⁺ cells and CD41⁺CD61⁺ cells were higher on day 7 and lower on day 12. While there was an increase up to 70,000 CD42b⁺ cells on day 7, only 40,000 were recovered on day 12, a very small population compared to the 900,000 CD42b⁺ cells in the SCF-containing wells.

The absolute counts of CD41⁺ cells reached extremely high values when IL-3 was added together with SCF and GM-CSF: there was a 25-fold increase in the first week and another 9.8-fold increase in the second, yielding an average of 1.8 million CD41⁺ cells by day 7 and 17.6 million CD41⁺ cells by day 12 from an initial 71,300 cells. The same high expansion rate was found for the CD61⁺ cells and the CD41⁺CD61⁺ cells, whereas the CD42b⁺ cells were only moderately expanded. Compared to TPO alone, the combination of TPO with EPO plus three early-acting cytokines generated a fourfold amount of CD41⁺ cells, but with different subpopulations (Table 2): there was much less coexpression of CD42b and CD34 and slightly more coexpression of c-kit.

The absolute counts of CD117⁺ cells on day 7 increased with the number of cytokines added to the cultures. On day 12, the amount of CD117⁺ cells was extremely high in the wells containing SCF: 15.6 million when GM-CSF + SCF were added to TPO + EPO, 45 million CD117⁺ cells after further addition of IL-3, confirming the synergism between SCF and IL-3 for the expansion of c-kit⁺ cells. Without SCF, only 5.3 million CD117⁺ cells were found on day 12.

The proliferation of erythroid progenitors was extremely dependent on the culture period: the absolute counts of GP-A⁺ cells reached only 1.7 million in the first week, but 102 million in the second week in the wells containing all five cytokines. In the first seven days, the addition of IL-3 expanded the GP-A⁺ cell pool more than SCF; in the second week the SCF-containing cultures yielded by far the highest counts of erythroid cells (Fig. 8).

Comparison of the Effects of EPO, GM-CSF, SCF and IL-3 with and without TPO
To evaluate the contribution of TPO on the effects obtained with the cytokine combination EPO + GM-CSF + SCF + IL-3, these cytokines were tested without the addition of TPO.

Surprisingly, there was no difference in the proportion of CD41⁺ cells in the wells with or without TPO. Yet the absolute counts of CD41⁺ cells on day 12 were 3.7-fold lower in the wells without TPO, and marker expression was different. Without TPO, only one-fourth of the CD41⁺ cells coexpressed CD42b, only half of them coexpressed CD34 on day 12, but coexpression of c-kit was higher (Table 2).

In terms of absolute counts, the combination of EPO + GM-CSF + SCF + IL-3 showed some thrombopoietic activity; the four cytokines together yielded about the same amount of total CD41⁺ cells as did TPO alone.

The most striking difference between the cultures with and without TPO was the amount of CD34⁺ cells, and, even more pronounced, the number of CD41⁺/CD34⁺ cells. Through the addition of TPO to the early-acting cytokines, the number of these immature subpopulations was six- to eightfold higher (Figs. 6 and 7).

Without TPO, the proportion of GP-A⁺ cells was very high: it reached 79% on day 12. The absolute counts of GP-A⁺ cells in the first week were similar to those obtained with all other EPO-containing combinations. Yet on day 12, without TPO in the culture medium, the number of GP-A⁺ cells was only half of those found in the wells containing the same cytokines plus TPO (49.8 million versus 102.5 million). This demonstrates a substantial enhancement of erythroid growth through the action of TPO after a certain lag phase.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
Effects of TPO Alone on Megakaryocytic Commitment, Maturation, CD34 Expansion and Erythroid Growth
In the absence of EPO and early-acting cytokines like IL-3, SCF and GM-CSF, TPO at first sight acts as a strictly lineage-specific cytokine: most CD34⁺ cells are driven into the megakaryocyte-platelet lineage within 12 days of culture. Guerriero et al. [41] recently reported a unilineage megakaryocytic proliferation and differentiation in a similar assay system. Yet the effects of TPO alone were crucially dependent on the culture period; the molecule acted on different target cell populations during the first and second week of culture. Initially, when TPO was incubated with freshly isolated mobilized CD34⁺ cells, a fivefold expansion of total CD34⁺ cells and an 11-fold increase of CD41⁺/CD34⁺ cells was observed, while the number of the more mature CD41⁺/CD42b⁺ cells remained very low. Thus in the first week of culture, TPO alone predominantly expands an immature CD34⁺/CD41⁺ cell population, either by recruiting CD34⁺CD41^– cells into the double-positive state or by inducing proliferation in the CD34⁺CD41⁺ cells, as it has been demonstrated by Debili et al. [42] with human purified CD34⁺CD41⁺ bone marrow cells. A substantial proportion of these double-positive cells further coexpress the c-kit molecule. The number of CD117⁺ cells and CD34⁺/CD117⁺ cells is expanded 20- to 25-fold in the first culture period.

The more committed progenitor cells which are generated in the first week following stimulation by TPO continue to serve as target cells for this molecule, yielding cell populations with a mature megakaryocytic phenotype after the second week of culture. Apart from a lower proliferation rate, the total number of CD34⁺ cells and of CD41⁺/CD34⁺ cells had declined, whereas the more mature CD42b⁺ cells expanded 25-fold. With respect to c-kit expression, there was a shift from expansion of CD34⁺CD117⁺CD41^– cells in the first week to an enormous increase of CD34^–CD117⁺CD41⁺ cells in the second week. The remaining CD34⁺ cells virtually all coexpressed CD41; moreover, nearly one half of them were CD61⁺.

Despite a vast homology with the EPO molecule, TPO alone was not able to support the growth of erythroid progenitors in serum-free medium. This fits the data of Kobayashi [27], who found no erythroid burst formation after stimulating human bone marrow and cord blood cells with TPO alone.

Thus, when added to freshly isolated human CD34⁺ cells, TPO in the absence of other cytokines behaves more like an early-acting cytokine generating committed CD41⁺ and c-kit⁺ progenitor cells which still bear the CD34 molecule. When incubated with these committed progenitors, TPO shows a highly lineage-specific activity as a MGDF.

Different Proliferation Rates and Commitment through the Addition of Further Cytokines
The expansion of erythroid cells following stimulation with TPO + EPO very closely resembled the increase of CD41⁺ cells after exposure to TPO alone. In the absence of early-acting cytokines, most of the progenitor cells were recruited into the erythroid lineage. Again, this effect was most pronounced in the second week of culture. The combination of TPO + EPO yielded the lowest proportion and number of CD34⁺ cells measured in all experiments. Clearly, the combination of these two cytokines drives progenitor cells very strongly to differentiate and lose the immature phenotype.

The addition of GM-CSF to TPO + EPO slightly increased the proportion of the subpopulations bearing megakaryocytic markers on day 12. Compared to the cultures stimulated with TPO alone, the higher proliferation rate in the wells with the triple cytokine combination outweighed the lower expression of megakaryocytic markers, resulting in a similar amount of megakaryocyte-committed CD41⁺ cells as in the cultures stimulated with TPO alone. Yet, coexpression of the maturation markers CD61 or CD42b was much lower than with TPO alone (Table 2). The CD34⁺ cell pool was less expanded than with TPO alone, and only half of the CD34⁺ cells on day 12 showed commitment to the megakaryocyte lineage. By contrast, overall c-kit expression and the number of CD117⁺CD34⁺ cells were twice as high, suggesting a different commitment of CD34⁺ cells. The increase of CD117⁺CD34⁺ cells probably reflects an increase of EPO-sensitive progenitors, since the addition of GM-CSF to TPO + EPO doubled the yield of GP-A⁺ cells in the second week. De Jong et al., recently identified the CD34⁺CD117^bright subpopulation as a fraction containing predominantly erythroid progenitors [43].

The SCF was found to act synergistically with TPO and EPO [28] on megakaryocyte colony formation in semisolid media. Our results confirmed these data in serum-depleted liquid culture. This effect was more pronounced on day 12 than on day 7, suggesting that a certain "lag time" is needed for SCF priming of the progenitor cells. The addition of SCF dramatically expanded the initially very low subpopulations of c-kit-bearing cells (total CD117⁺ cells, CD34⁺CD117⁺ and CD41⁺CD117⁺ cells), especially during the second week. These findings differ from the results of Uoshima et al., who reported a down-modulation of c-kit expression on erythroid progenitors in the presence of SCF [44]. The addition of SCF further doubled the number of GP-A⁺ erythroid progenitors on day 12.

When SCF was substituted by IL-3, the yield of cells expressing CD41, CD61, CD42b and the respective double-positive phenotype was consistently lower, confirming a merely additive and not synergistic effect of TPO + IL-3 compared to TPO + SCF, as reported by Lok [13]. Costimulation with IL-3 instead of SCF produced a different growth pattern both in the overall CD117⁺ population and in the CD41⁺CD117⁺ and CD34⁺CD117⁺ subpopulations: in the first week they all grew more than with SCF, in the second week they were by far less expanded than with SCF. The fact that GP-A⁺ erythroid cells followed the same pattern is consistent with the findings of Muta et al., who showed that SCF stimulates proliferation of erythroid progenitors but retards differentiation [45].

Effects of EPO + SCF + IL-3 + GM-CSF with and without TPO
There is a clear over-additive effect between the early-acting cytokines and TPO with respect to expansion of the CD41⁺ cell pool. Although the four cytokines EPO + SCF + IL-3 + GM-CSF generated the same overall amount of CD41⁺ cells as did TPO alone, substantial differences could be found with regard to the maturation status of these megakaryocytic committed cells (Table 2): TPO alone yielded more cells coexpressing CD61, much more mature cells bearing the CD42b molecule and CD41⁺CD34⁺ cells, whereas the combination of the four cytokines without TPO yielded by far more CD41/c-kit double-positive cells. While TPO alone was able to increase the committed progenitor cell pool and subsequently to generate mature megakaryocytes, the combination of the four "thrombopoietic" cytokines enhanced predominantly a subpopulation of probably intermediate differentiation. The addition of TPO to these four cytokines changed the phenotype of the CD41⁺ cells towards the pattern obtained with TPO alone (Figs. 8 and 9).

The progressive expansion of cells of the megakaryocytic lineage through the addition of early-acting cytokines to TPO may be due to the fact that these cytokines augment the TPO-responsive cell fraction. On the other hand, TPO might be able to induce release of other cytokines from its target cells. Guerriero et al. demonstrated that human peripheral blood megakaryocytes cultured with TPO are able to secrete low amounts of IL-6 and GM-CSF [41], whereas Banu et al. found no induction of cytokine secretion by human bone marrow megakaryocytes after treatment with TPO [46]. Detailed studies using very immature progenitor cells as targets for TPO are needed to clarify the question of the earliest TPO-responsive cell.

TPO is Both an Early-Acting Cytokine and an MGDF
TPO together with the four cytokines yielded a 2.2-fold higher amount of CD34⁺ cells on day 7 and a 4.8-fold higher number on day 12, compared to the wells without TPO. While TPO alone only transiently supports the growth of CD34⁺ megakaryocyte-committed progenitor cells to give way to more maturation processes along the megakaryocyte-platelet pathway, the combination of TPO and early-acting cytokines can expand the number of CD34⁺ cells for at least two weeks.

Although in the absence of EPO it does not support the growth of GP-A⁺ cells, TPO together with EPO and early-acting cytokines can considerably enhance erythroid growth, provided that the appropriate committed progenitor cells are present in the culture. The expansion of a CD34⁺CD117⁺ progenitor cell pool caused by TPO increases the number of target cells for EPO, thereby enhancing erythroid growth. This process can be considerably enhanced by the action of SCF. Kaushansky et al. reported that TPO in combination with EPO, SCF and IL-3 augmented the formation of early and late erythroid progenitors in a murine model. Moreover, the administration of TPO to myelosuppressed mice enhanced the recovery of all three hematopoietic lineages and shortened the duration of thrombocytopenia, anemia and, to a lesser extent, leukopenia [47, 48]. Our results strongly suggest that also in human hematopoiesis the action of TPO is not restricted to the megakaryocytic lineage.

Since TPO can induce a variety of proliferation and maturation events in purified progenitor cells without the presence of early-acting cytokines in serum-depleted medium, it is probably an early-acting cytokine itself. This issue is supported by the findings of Alexander et al., who found a significant reduction not only in megakaryocytes, but also in the production of committed progenitors of multiple hematopoietic lineages in c-mpl-deficient mice [49]. Moreover, Ku et al. recently proposed that TPO can function as an early-acting cytokine having found a synergism between SCF and TPO with respect to multilineage colony formation in a murine model [50], which could be abrogated by a neutralizing antibody to the kit protein.

The effects of TPO, alone and in combination with the other cytokines described, make this molecule a very interesting candidate for ex vivo expansion of human peripheral blood stem cells.

	Acknowledgments

 
The authors thank Mrs. Elisabeth Holzmann, Mrs. Manuela Stüwe and Mrs. Andrea Weissenberger for their excellent technical assistance. We are grateful to Dr. Jakob Bogenberger (Amgen Inc., Thousand Oaks, CA) and Prof. Karl Welte (Hannover Medical School, Hannover, Germany) for providing the recombinant thrombopoietin. We thank Behringwerke (Marburg, Germany) for the GM-CSF, CILAG Biotech (Sulzbach, Germany) for the EPO, ICC (Ismaning, Germany) for the SCF, and Sandoz AG (Nürnberg, Germany) for the IL-3.

This work was supported by Deutsche Krebshilfe, Bonn, Germany.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Vigon I, Mornon J-P, Cocault L et al. Molecular cloning and characterization of MPL, the human homolog of the v-mpl oncogene: identification of a member of the hematopoietic growth factor receptor superfamily. Proc Natl Acad Sci USA 1992;89:5640-5644.[Abstract/Free Full Text]

2. Debili N, Wendling F, Cosman D et al. The Mpl receptor is expressed in the megakaryocytic lineage from late progenitors to platelets. Blood 1995;85(suppl 2):391-401.[Abstract/Free Full Text]

3. Methia N, Louache F, Vainchenker W et al. Oligodeoxynucleotides antisense to the proto-oncogene c-mpl specifically inhibit in vitro megakaryopoiesis. Blood 1993;82:1395-1401.[Abstract/Free Full Text]

4. de Sauvage FJ, Hass PE, Spencer SD et al. Stimulation of megakaryocytopoiesis and thrombopoiesis by the c-Mpl ligand. Nature 1994;369:533-538.[Medline]

5. Bartley TD, Bogenberger J, Hunt P. Identification and cloning of a megakaryocyte growth and development factor that is a ligand for the cytokine receptor Mpl. Cell 1994;77:1117-1124.[Medline]

6. Lok S, Kaushansky K, Holly RD et al. Cloning and expression of murine thrombopoetin cDNA and stimulation of platelet production in vivo. Nature 1994;369:565-568.[Medline]

7. Kaushansky K, Lok S, Holly RD et al. Promotion of megakaryocyte progenitor expansion and differentiation by the c-Mpl ligand thrombopoetin. Nature 1994;369:568-571.[Medline]

8. Kato T, Iwamatsu A, Shimada Y et al. Purification and characterization of thrombopoietin derived from thrombocytopenic rat plasma. Blood 1994;84(suppl):329a.

9. Kato T, Ogami K, Shimada Y et al. Purification and characterization of thrombopoetin. J Biochem 1995;118:229-236.[Abstract/Free Full Text]

10. Kuter DJ, Beeler DL, Rosenberg RD. The purification of megapoietin: a physiological regulator of megakaryocyte growth and platelet production. Proc Natl Acad Sci USA 1994;91:11104-11108.[Abstract/Free Full Text]

11. Gurney AL, Kuang WJ, Xie MH et al. Genomic structure, chromosomal localization, and conserved alternative splice forms of thrombopoietin. Blood 1995;85:981-988.[Abstract/Free Full Text]

12. Sohma Y, Akahori H, Seki N et al. Molecular cloning and chromosomal localization of the human thrombopoetin gene. FEBS Lett 1994;353:57-61.[Medline]

13. Lok S, Foster DC. The structure, biology and potential therapeutic applications of recombinant thrombopoetin. STEM CELLS 1994;12:586-598.[Abstract]

14. Hunt P. The physiologic role and therapeutic potential of the Mpl-ligand in thrombopoiesis. STEM CELLS 1995;13:579-587.

15. Kaushansky K. Thrombopoietin: the primary regulator of platelet production. Blood 1995;86:419-431.[Free Full Text]

16. Ulich TR, del Castillo J, Yin S et al. Megakaryocyte growth and development factor ameliorates carboplatin-induced thrombocytopenia in mice. Blood 1995;86:971-976.[Abstract/Free Full Text]

17. Hokom MM, Lacey D, Kinstler OB et al. Pegylated megakaryocyte growth and development factor abrogates the lethal thrombocytopenia associated with carboplatin and irradiation in mice. Blood 1995;86:4486-4492.[Abstract/Free Full Text]

18. Farese AM, Hunt P, Boone T et al. Recombinant human megakaryocyte growth and development factor stimulates thrombocytopoiesis in normal nonhuman primates. Blood 1995;86:54-59.[Abstract/Free Full Text]

19. Debili N, Hegyi E, Navarro S. In vitro effects of hematopoietic growth factors on the proliferation, endoreplication, and maturation of human megakaryocytes. Blood 1991;77:23-26.

20. Choi ES, Hokom MM, Bartley TD et al. Megakaryocyte growth and development factor produces functional human platelets in vitro. STEM CELLS 1995;13:317-322.

21. Chen JC, HercegHarjacek L, Groopman JE et al. Regulation of platelet activation in vitro by the c-Mpl ligand, thrombopoietin. Blood 1995;86:4054-4062.[Abstract/Free Full Text]

22. Lebkowski JS, Schain LR, Okarma TB. Serum-free culture of hematopoietic stem cells: a review. STEM CELLS 1995;13:607-612.[Abstract]

23. Eliason JF, Odartchenko N. Colony formation by primitive hemopoietic progenitor cells in serum-free medium. Proc Natl Acad Sci USA 1985;82:775-779.[Abstract/Free Full Text]

24. Iizuka Y, Murphy MJ Jr. Colony formation of granulocyte (CFU-G) and macrophage (CFU-M) precursors in serum- and albumin-free culture: effect of transferrin on clonal growth. Exp Cell Biol 1986;54:275-280.[Medline]

25. Drouet X, Douay L, Giarratana MC et al. Human liquid bone marrow culture in serum-free medium. Br J Haematol 1989;73:143-147.[Medline]

26. Lebkowski JS, Schain LR, Hall MA et al. Rapid isolation and serum-free expansion of human CD34⁺ cells. Blood Cells 1994;20:404-410.[Medline]

27. Kobayashi M, Laver JH, Kato T et al. Recombinant human thrombopoietin (Mpl ligand) enhances proliferation of erythroid progenitors. Blood 1995;86:2494-2499.[Abstract/Free Full Text]

28. Broudy VC, Lin NL, Kaushansky K. Thrombopoietin (c-mpl ligand) acts synergistically with erythropoietin, stem cell factor, and interleukin-11 to enhance murine megakaryocyte colony growth and increases megakaryocyte ploidy in vitro. Blood 1995;85:1719-1726.[Abstract/Free Full Text]

29. Bruno E, Cooper RJ, Briddell RA et al. Further examination of the effects of recombinant cytokines on the proliferation of human megakaryocyte progenitor cells. Blood 1991;77:2339-2346.[Abstract/Free Full Text]

30. Ganser A, Lindemann A, Seipelt G et al. Effect of recombinant human interleukin 3 in patients with normal hematopoiesis and in patients with bone marrow failure. Blood 1990;76:666-676.[Abstract/Free Full Text]

31. Hunt P, Li YS, Nichol JL et al. Purification and biologic characterization of plasma-derived megakaryocyte growth and development factor. Blood 1995;86:540-547.[Abstract/Free Full Text]

32. Nichol JL, Hornkohl AC, Choi ES et al. Enrichment and characterization of peripheral blood-derived megakaryocyte progenitors that mature in short-term liquid culture. STEM CELLS 1994;12:494-505.[Abstract]

33. Krause DS, Fackler MJ, Civin CI et al. CD34: structure, biology, and clinical utility. Blood 1996;87:1-13.[Free Full Text]

34. Block KL, Poncz M. Platelet glycoprotein IIb gene expression as a model of megakaryocyte-specific expression. STEM CELLS 1995;13:135-145.[Abstract]

35. Shaw S, Luce GG, Gilks WR et al. Leukocyte differentiation antigen database in leukocyte typing V. Oxford: Oxford University Press, 1994.

36. Choi ES, Nichol JL, Hokom MM et al. Platelets generated in vitro from proplatelet-displaying human megakaryocytes are functional. Blood 1995;85(suppl 2):402-413.[Abstract/Free Full Text]

37. Dercksen MW, Weimar IS, Richel DJ et al. The value of flow cytometric analysis of platelet glycoprotein expression on CD34(⁺) cells measured under conditions that prevent P-selectin-mediated binding of platelets. Blood 1995;86:3771-3782.[Abstract/Free Full Text]

38. Lerner NB, Nocka KH, Cole SR et al. Monoclonal antibody YB5.B8 identifies the human c-kit protein product. Blood 1991;77:1876-1883.[Abstract/Free Full Text]

39. Gore SD, Amin S, Weng LJ et al. Steel factor supports the cycling of isolated human CD34(⁺) cells in the absence of other growth factors. Exp Hematol 1995;23:413-421.[Medline]

40. Keller JR, Ortiz M, Ruscetti FW. Steel factor (c-kit ligand) promotes the survival of hematopoietic stem progenitor cells in the absence of cell division. Blood 1995;86:1757-1764.[Abstract/Free Full Text]

41. Guerriero R, Testa U, Gabbianelli M et al. Unilineage megakaryocytic proliferation and differentiation of purified hematopoietic progenitors in serum-free liquid culture. Blood 1995;86:3725-3736.[Abstract/Free Full Text]

42. Debili N, Wendling F, Katz A et al. The Mpl-ligand or thrombopoietin or megakaryocyte growth and differentiative factor has both direct proliferative and differentiative activities on human megakaryocyte progenitors. Blood 1995;86:2516-2525.[Abstract/Free Full Text]

43. DeJong MO, Wagemaker G, Wognum AW. Separation of myeloid and erythroid progenitors based on expression of CD34 and c-kit. Blood 1995;86:4076-4085.[Abstract/Free Full Text]

44. Uoshima N, Ozawa M, Kimura S et al. Changes in c-Kit expression and effects of SCF during differentiation of human erythroid progenitor cells. Br J Haematol 1995;91:30-36.[Medline]

45. Muta K, Krantz SB, Bondurant MC et al. Stem cell factor retards differentiation of normal human erythroid progenitor cells while stimulating proliferation. Blood 1995;86:572-580.[Abstract/Free Full Text]

46. Banu N, Wang JF, Deng BJ et al. Modulation of megakaryocytopoiesis by thrombopoietin: the c-Mpl ligand. Blood 1995;86:1331-1338.[Abstract/Free Full Text]

47. Kaushansky K, Broudy VC, Grossmann A et al. Thrombopoietin expands erythroid progenitors, increases red cell production, and enhances erythroid recovery after myelosuppressive therapy. J Clin Invest 1995;96:1683-1687.

48. Kaushansky K, Lin N, Grossmann A et al. Thrombopoietin expands erythroid, granulocyte-macrophage, and megakaryocytic progenitor cells in normal and myelosuppressed mice. Exp Hematol 1996;24:265-269.[Medline]

49. Alexander WS, Roberts AW, Nicola NA et al. Deficiencies in progenitor cells of multiple hematopoietic lineages and defective megakaryocytopoiesis in mice lacking the thrombopoietin receptor c-Mpl. Blood 1996;87:2162-2170.[Abstract/Free Full Text]

50. Ku H, Yonemura Y, Kaushansky K et al. Thrombopoietin, the ligand for the Mpl receptor, synergizes with steel factor and other eary acting cytokines in supporting proliferation of primitive hematopoietic progenitors of mice. Blood 1996;87:4544-4551.[Abstract/Free Full Text]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints/Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Birkmann, J. Articles by Gallmeier, W. M. Search for Related Content PubMed PubMed Citation Articles by Birkmann, J. Articles by Gallmeier, W. M.