Flt3/Flk-2-Ligand in Synergy with Thrombopoietin Delays ...

2 downloads 0 Views 270KB Size Report
Megakaryocyte Development and Increases the Numbers of ... senescent megakaryocytes but not megakaryocyte progenitor cells. In this study, a serum-free ...
JOURNAL OF HEMATOTHERAPY & STEM CELL RESEARCH 11:389–400 (2002) © Mary Ann Liebert, Inc.

Research Report Flt3/Flk-2-Ligand in Synergy with Thrombopoietin Delays Megakaryocyte Development and Increases the Numbers of Megakaryocyte Progenitor Cells in Serum-Free Cultures Initiated with CD341 Cells ÓLAFUR EYSTEINN SIGURJÓNSSON,1 KRISTBJÖRN ORRI GUDMUNDSSON,1 VILHELMÍNA HARALDSDÓTTIR,2 THORUNN RAFNAR,3 and SVEINN GUDMUNDSSON1

ABSTRACT Megakaryocytopoiesis involves proliferation and maturation of committed precursors that increase their size by polyploidy, a process that is believed to be critical for the efficient production and release of platelets. Thrombopoietin has been shown to act on proliferation, maturation, and survival pathways in megakaryocytopoiesis. Less is known about the role of Flt3/Flk-2-ligand in this development. Apoptosis has an important role in hematopoiesis in general. It has been shown to have an effect on senescent megakaryocytes but not megakaryocyte progenitor cells. In this study, a serum-free culture model was developed, differentiating bone marrow CD341 hematopoietic stem cells into megakaryocytes, using thrombopoietin and Flt3/Flk-2-ligand. The model was used to study the effect of these growth factors on expansion of megakaryocyte progenitor cells, differentiation of megakaryocytes, and ploidy. Our results demonstrate that bone marrow CD341 cells cultured with thrombopoietin and Flt3/Flk-2-ligand show a lower developmental rate into MK cells compared to cells cultured with thrombopoietin alone. Cells cultured with thrombopoietin and Flt3/Flk-2-ligand expressed less CD41, the ploidy level was lower, and they appeared less mature. On the other hand, the cells showed up to 10-fold increase in cell numbers compared to five-fold increase when cultured with thrombopoietin alone. These results suggest that Flt3/Flk-2-ligand in synergy with thrombopoietin may slow down megakaryocyte development by causing increased proliferation of megakaryocyte progenitor cells. INTRODUCTION

M

by which hematopoietic stem cells (HSC) in the bone marrow (BM) develop into mature megakaryocytes (MKs), which then generate functional platelets (1). Circulating platelets have a vital function in the maintenance of hemostasis in the blood vasculature. This process can be critically impaired in patients undergoing HSC transplantation to reconstitute hematopoiesis following myeEGAKARYOCYTOPOIESIS IS THE PROCESS

loablative chemotherapy (2–4). Thus, chemotherapy can leave patients thrombocytopenic for several weeks, requiring platelet transfusions, with potential increase in the risk of infections and considerable costs (5). This has prompted a search for alternative methods of restoring the platelet population in these patients. A number of cytokines, either independently or in combination, modulate megakaryocytopoiesis. These include thrombopoietin (TPO), the ligand for the c-mpl receptor (6–8), interleukin-3 (IL-3), IL-6, IL-11, and

1 The

Blood Bank, Department of Basic Research, Landspitali-University Hospital, 101 Reykjavík, Iceland. of Hematology and Oncology, Landspitali-University Hospital, 101 Reykjavík, Iceland. 3 Icelandic Genomics Corporation, 105 Reykjavík, Iceland. 2 Department

389

SIGURJÓNSSON ET AL. Flt3/Flk-2-ligand (FL), which is known to be a potent costimulator for the expansion of HSC in synergy with TPO (9,10). Megakaryocytes (MKs) express a number of antigens during their development, including CD34, CD41, and HLA-DR, which can be used to monitor MK development in in vitro cultures (1,11). MKs undergo a peculiar and irreversible process by which they become polyploid through repeated cycles of DNA synthesis without concomitant cell division. During this program, termed endomitosis, MKs acquire a DNA content up to 128N, 16–32N being the prevalent ploidy (12). The level of MK polyploidization is inversely related to the platelet count (13,14) and is accompanied by an increase in plasma volume, nuclear, and cell size (15,16). The rate of platelet recovery following HSC transplantation correlates with the number of CFU-MKs infused and/or the number of CD341 CD411 cells that promote platelet engraftment (17,18). Because it is usually not possible to obtain a sufficient number of these MK progenitors, supplementation of conventional autografts with ex vivo-expanded MK-rich products may enhance in vivo platelet production and shorten the period of thrombocytopenia (4,19,20). Bertolini et al. (19) demonstrated that half of the patients receiving expanded MK progenitor cells along with HSC did not need additional platelet transfusions. TPO, the cytokine that has been studied most for its effect on MK development, has little proliferative effect on human CD341 cells in culture and the addition of early-acting cytokines, i.e., SCF and FL, is believed to be required to increase total cell proliferation and MK expansion (21,22). The aim of this study was to develop an in vitro serumfree culture model to differentiate BM CD341 HSC into MKs and use that model to study the effects of TPO and FL on MK differentiation.

MATERIALS AND METHODS Bone marrow The research protocol was approved by the LandspitaliUniversity Hospital Ethical Committee. Normal femoral bone marrow (BM) samples were collected from patients undergoing hip prosthesis surgery (total hip arthroplasty). The BM was suspended in isotonic NaCl solution (Icelandic Pharmaceuticals Ltd, Reykjavík, Iceland) containing 1000 IU/ml of EDTA (Merck, Darmstadt, Germany).

Isolation of BM mononuclear cells and CD341 cells

(density, 1.077g/ml) (Sigma Chemical Company, St. Louis, MO) and washed three times in RPMI-1640 medium (Life Technologies Ltd, Paisley, UK). Adherent cells were discarded after 12 hr of incubation at 37°C and 5% CO2 in RPMI-1640 containing 10% fetal bovine serum (FBS) (Life Technologies Ltd.) supplemented with 100 U/ml penicillin (Life Technologies Ltd) and 100 U/ml streptomycin (Life Technologies Ltd). The nonadherent MNC were used for further isolation. CD341 cells were isolated from the nonadherent portion of MNCs using the Dynal CD34 Progenitor Cell Selection System (Dynal, Oslo, Norway). MNCs were incubated with Dynabeads (magnetic beads, Dynal) coated with antibody against the CD34 molecule at 4°C for 30 min. Cells were washed and resuspended in DETACHaBEAD (Dynal), and incubated for 45 min at 22°C to release the Dynabeads from the CD341 cells. The CD341 cells were collected, and purity was assessed by flow cytometry after staining with CD34 antibody (HPCA-2-PE, Becton Dickinson, San Diego, CA) using a Becton Dickinson FACSCalibur flow cytometer equipped with an argon ion laser. A minimum of 10,000 events were acquired in list mode and analyzed with CellQuest software 3.1f. (Becton Dickinson).

Viability and cell number Viability and cell number were assessed using Acridin Orange-Ethidium Bromide (Sigma) staining. Live cells appear as green and dead cells as orange using fluorescence microscopy (Leitz-Weitzler, Darmstadt, Germany). Live cells were counted using a hematocytometer glass-slide (American Optical Corporation, Buffalo, NY).

Cell cultures CD341 cells (5 3 104 /ml) were cultured in 24-well plates (Falcon, NJ) for a maximum of 21 days at 37°C and 5% CO2 in Stemspan serum-free medium (Stem Cell Inc, Vancouver, BC, Canada) supplemented with 100 U/ml penicillin and 100 U/ml streptomycin. Cultures were initiated with TPO (50 ng/ml) (R&D Systems Inc, Minneapolis, MN) and/or FL (50 ng/ml) (R&D). Every 5 days, one-half of the culture volume was removed and replaced with fresh medium and growth factors, bringing the total volume to 1 ml. Cells were harvested either at days 1, 3, 5, 7, 14, and 21 or days 1, 5, 10, 15, and 20. Parallel control experiments were performed without the addition of either of these growth factors.

Phenotypic analysis

Low-density mononuclear cells (MNC) were isolated by centrifugation with HISTOPAQUE-1077 solution

Differentiation of CD341 cells toward the MK lineage was monitored by measuring the expression of CD41

390

MEGAKARYOCYTE DEVELOPMENT IN VITRO The cells were washed twice in PBS and resuspended in 400 ml of 1% paraformaldehyde solution (Merck, Darmstadt, Germany), and then analyzed by a flow cytometer. A minimum of 10,000 events were acquired in list mode and analyzed with the CellQuest software 3.1f. (Becton Dickinson).

Ploidy analysis

FIG. 1. Number of viable cells in cultures with thrombopoietin (TPO) (j ) or TPO and Flt3/Flk-2-ligand (FL) ( u ), n 5 5. Data are presented as mean 6 SEM. Significant difference was found between TPO and TPO/FL from day 1 through 21 (p , 0.05).

(HIP8-FITC, Pharmingen, San Diego, CA), CD34 (HPCA-2 PE, Becton Dickinson), and HLA-DR (G46-6PerCP, Becton Dickinson) during the culture period. Cells were harvested and washed twice in a phosphatebuffered saline solution (PBS). The cells were then resuspended in 100 ml of PBS and incubated with either 10 ml of CD41a-FITC, CD34-PE, and HLA-DR-PerCP or isotype-matched controls; mouse immunoglobulin G 1 (IgG1) (X40, FITC/PE Becton Dickinson), and mouse IgG2a (X39, PerCP, Becton Dickinson) for 30 min at 4°C.

Ploidy in cultured cells was analyzed using a flow cytometer. The cells were harvested, washed two times in PBS (200 3 g/5 min), and subsequently fixed in 70% ethanol (vol/vol) and stored at 220°C for up to 1 month. In some experiments, cells were incubated with 10 ml of CD41 FITC and subsequently fixed in 70% ethanol (vol/vol) and stored at 220°C for 2–4 hr. The ethanol was discarded after centrifugation (230 3 g/5 min). To stain the DNA, the cells were incubated for 15 min at 37°C with propidium iodide (PI) (1 mg/ml) (Sigma), in PBS containing RNase A (200 mg/ml) (Sigma) and 0.1% (vol/vol) Tween 20 (Merck). A minimum of 10,000 events were acquired in list mode and analyzed with the CellQuest software 3.1f.

Morphology Morphology of cells was determined using WrightGiemsa staining. In brief, cells were harvested and spun onto glass slides using a Cytospin® 3 Cytocentrifuge (Shandon, Pittsburgh, PA). The cells were air-dried at

FIG. 2. Wright-Giemsa staining of cultured cells. CD341 cells cultured with thrombopoietin (TPO). (A) Day 3 (1003). (B) Day 14 (4003). (C) Day 21 (4003). (D) Day 14 (4003). The arrows point to megakaryocytes.

391

SIGURJÓNSSON ET AL. 37°C for 1 hr and fixed in acetone for 5 min. The slides were then stained with Wright-Giemsa dye (Sigma) for 1 hr and then rinsed and mounted with aqueous mounting medium (Sigma).

Statistical analysis

TPO and FL the number of viable cells increased 10-fold from day 0 to 14 and then decreased after 21 days in culture. Thus, co-culturing with TPO and FL significantly increased the viability of CD341 cells as compared to culturing with TPO alone (p , 0.001).

Megakaryocytopoiesis

Statistical calculations were made using GraphPad Prism version 3.00 for Windows, (GraphPad Software, San Diego, CA). Significance tests were performed using a paired Student’s t-test. The difference was considered significant if p , 0.05. The results are presented as mean 6 standard error of the mean.

RESULTS Cell number and viability When CD341 cells were cultured with TPO alone, the number of viable cells increased five-fold from day 0 to 14 and then decreased after 21 days in culture (Fig. 1). On the other hand, when CD341 cells were cultured with

Differentiation of CD341 cells toward the MK lineage was analyzed using three different criteria: (1) changes in morphology of cells in culture, (2) changes in expression of the cell-surface antigens CD41, CD34 and HLADR, and (3) changes in ploidy. Morphology: Staining of cultured cells with WrightGiemsa demonstrated morphological changes of MK development. The mature cells were large, with big lobulated nuclei and extensive cytoplasm (Fig. 2). Changes in size and granularity were monitored by flow cytometry. During the culture period the cells developed a higher forward scatter profile (FSC) (reflecting an increase in size) and a higher side scatter profile (SCC) (reflecting an increase in granularity) (Fig. 3).On days 14 and 21, the cultured cells showed a notable

FIG. 3. Changes in size (FSC) and granularity (SSC) during culture of CD341 cells from bone marrow in serum-free medium without growth factors (NG) or supplemented with TPO or TPO/FL.

392

MEGAKARYOCYTE DEVELOPMENT IN VITRO

FIG. 4. Proportion of CD341 CD412 cells (A), CD341 CD411 cells (B) and CD342 CD411 cells. (C) CD341 cells were cultured with thrombopoietin (TPO) (j ) or TPO/FL ( u ). Data are presented as mean 6 SEM. No significant difference was found between CD341 CD412 cells cultured with TPO or TPO/FL (p . 0.05), n 5 5. The proportion of CD341 CD411 cells cultured with TPO was significantly higher compared to cells cultured with TPO/FL on day 5 and 7 (p , 0.05), n 5 5. The proportion of CD342 CD411 cells cultured with TPO was significantly higher compared to cells cultured with TPO/FL on days 14 and 21 (p , 0.05) n 5 5.

change in FSC and SSC often connected with apopototic cells (Fig. 3). Cells cultured in a serum-free medium without growth factors gradually lost their viability, and, after 7 days, 90% of the cells were dead (Fig. 3). However, cells cultured with TPO and FL showed more viability compared to cells cultured with TPO alone. Expression of CD34, CD41, and HLA-DR in megakaryocyte development: To monitor the differentiation of CD341 cells toward the MK lineage, we analyzed the expression of the CD34 and CD41 molecules. CD341 cells from bone marrow were cultured for 21 days with TPO or TPO and FL, as described above. CD341 CD412 cells

represent the progenitor cells that have the ability to expand and differentiate into MKs as well as other types of blood cells. The proportion of CD341 CD412 cells decreased throughout the cultures, both when TPO and TPO and FL were present in the cultures (Figs. 4A and 5). Coculturing the cells with TPO and FL had no significant effect on the proportion of CD341 CD412 cells compared to culturing with TPO alone (p . 0.05). CD341 CD411 cells represent MK progenitor cells that are probably not able to differentiate into other cell types than MKs but are able to expand to some extent. When CD341 cells were cultured with TPO alone the propor-

FIG. 5. Representative dot plots showing changes in CD34 and CD41 expression of bone marrow CD341 cells cultured with TPO or TPO/FL on days 1, 7, 14, and 21.

393

SIGURJÓNSSON ET AL. tion of CD341 CD411 cells peaked around day 7, as well as with TPO and FL (Figs. 4B and 5). On days 14 and 21, the proportion of CD341 CD411 cells decreased, both when cultured with TPO or TPO and FL. The proportion of the CD341 CD411 cells was significantly higher, when cultured with TPO, on day 5 (p , 0.0471) and day 7 (p , 0.0056) compared to culturing the cells with TPO and FL (Figs. 4B and 5). CD342 CD411 cells represent cells with a typical MK phenotype that increase their DNA content without cell division. The proportion of CD341 CD411 cells started to increase on day 5 and did so throughout the culture period, both when TPO or TPO and FL were present in the cultures (Figs. 4C and 5). Culturing the cells with TPO alone significantly increased the proportion of CD342 CD411 cells compared to coculturing with TPO and FL, on days 14 (p , 0.001) and 21 (p , 0.001). It has been demonstrated that it is possible to discriminate between early and late MK progenitor cells by examining their HLA-DR expression. This has mainly been done in clonal cultures where it has been shown that late MK progenitor cells express DR and CD41—colonyforming unit–megakaryocyte (CFU-MK), whereas the early progenitors express neither—burst-forming unit–megakaryocyte (BFU-MK) (23,24). CD341 cells from bone marrow were cultured for 21 days with TPO or TPO and FL as described. Expression of HLA-DR (DR), CD34, and CD41 was analyzed by flow cytometry after three-color staining. When CD341 cells were cultured with TPO alone, the proportion of CD411 DR1 cells increased from day 0 to 7 and then decreased (Figs. 6A and 7). On the other hand, when CD341 cells were cultured with TPO and FL, the proportion of CD411 DR1 increased until day 5, when they reached plateau with a slight decrease in days 7 and

21 (Figs. 6A and 7). No significant difference was found in HLA-DR and CD41 expression between cultures with TPO or TPO and FL. When CD341 cells were cultured with TPO alone, the porportion of CD411 DR2 cells increased from day 3 to 21 (Figs. 6A and 7). On the other hand, when CD341 cells were cultured with TPO and FL, the proportion of CD411 DR2 cells increased from day 3 to 14 and decreased on day 21 (Figs. 6B and 7). Culturing the cells with TPO alone significantly increased the proportion of CD411 DR2 cells compared to culturing the cells with TPO alone, on day 14 (p 5 0.016) and 21 (p 5 0.070). Taken together, our results show that DR expression is high in the CD34 population, decreasing with time. However, in the CD41 population, the DR expression increases during the first 7 days of the cultures and then starts to decrease. Polyploidy in megakaryocyte development: The development of polyploidy is probably the single best indicator of MK differentiation. To demonstrate further that our serum-free culture model was differentiating CD341 cells toward MKs, ploidy was analyzed by staining the cells with PI and analyzing them by flow cytometry. The proportion of diploid cells (2N) steadily decreased during the culture period and accordingly the porportion of cells with ploidy greater than 2N increased. When CD341 cells were cultured with TPO alone, the proportion of 2N cells decreased from day 1 to 14 and then more rapidly on day 21 (Fig. 8). On the other hand, when CD341 cells were cultured with TPO and FL, the proportion of 2N cells decreased from day 1 to 14 and then increased on day 21 (Fig. 8). Culturing the cells with TPO and FL significantly increased the proportion of 2N cells, on day 21 (p 5 0.027), compared to culturing the cells with TPO alone.

FIG. 6. Proportion of CD411 HLA-DR 1 cells (A) and CD411 HLA-DR 2 cells. (B) CD341 cells were cultured with thrombopoietin (TPO) (j ) or TPO/FL ( u ). Data are presented as mean 6 SEM. Significant difference was found between CD341 cells cultured with TPO and TPO/FL on day 14 and 21 (p , 0.05), n 5 5.

394

MEGAKARYOCYTE DEVELOPMENT IN VITRO

FIG. 7. Representative dot plots showing changes in CD41 and HLA-DR expression of bone marrow CD341 cells cultured with TPO or TPO/FL on days 1, 7, 14, and 21.

The proportion of polyploid cells (4N1) steadily increased until around day 14 of the culture period (Fig. 8). Cells with 4N ploidy increased from day 0 to 7 and then decreased until day 21. There was a significantly higher proportion of 4N cells when cultured with TPO and FL compared to TPO alone (p 5 0.008) (Fig. 8). Cells with 8N ploidy peaked around day 7 and then decreased on day 14 before plummeting on day 21 (Fig. 8).

It was difficult to discriminate the 32N population from the 16N population, so they were defined as a 16N1 population. The proportion of 16N1 cells increased from day 0 to day 14 and the dropped on day 21 (Fig. 8). Culturing the cells with TPO or TPO and FL showed no significant difference between the 8N and 16N1 populations (p . 0.05). To investigate further the decrease in polyploidy on

FIG. 8. Representative histograms for ploidy analysis in CD341 cells cultured with TPO or TPO/FL on days 1, 7, 14, and 21. Ploidy was analyzed by propidium iodide (PI) staining.

395

SIGURJÓNSSON ET AL. day 21, we specifically analyzed the subdiploid cells, which are considered to be apoptotic. The proportion of subdiploid cells showed a slight increase from day 0 to 3 and then decreased until day 7 (Figs. 8 and 9). On day 14, the proportion of subdiploid cells started to increase, being significantly higher on day 21 in cultures with TPO alone compared to TPO and FL (p 5 0.003) (Figs. 8 and 9). These results support our earlier findings that CD341 cells in the serum-free culture model are differentiating along the MK lineage. It is also interesting to observe that cells cultured with TPO show a trend toward higher ploidy (8N–16N) than cells cultured with TPO and FL (Fig. 8), even though it is not statistically significant. Polyploidy and CD41 expression: To investigate further the effect of FL on polyploidy in megakaryocytopoiesis, we examined the expression of CD41 and ploidy using dual staining and flow cytometry. There was a good correlation between an increase in CD41 expression and the increased level of polyploidy in CD341 cells cultured with either TPO or TPO and FL (Fig. 10). Culturing the cells with TPO alone led to a higher expression of CD41 in the 2N–16N1 fraction than in cultures with TPO and FL. The proportion of CD411 (2N–16N1)1 cells increased from day 1 to 14 and then decreased on day 21 (Fig. 10). CD341 cells cultured with TPO alone showed a significant higher proportion of the CD411 (2N-16N1)1 cells compared to cultures with TPO and FL, on days 5 and 7 (Fig. 10). At the same time, the proportion of CD411 (2N–4N)1 cells decreased, with the decrease being more apparent when cells were cultured with TPO alone. CD341 cells cultured with TPO and FL showed a significantly higher proportion of the CD412 (2N–4N)1 cells compared to cultures with TPO

FIG. 9. Proportion of subdiploid (apoptotic) cells in cultures. CD341 cells were cultured with TPO (j ) or TPO/FL ( u ). Data are presented as mean 6 SEM. Significant difference was found between CD341 cells cultured with TPO and TPO/FL on day 21 (p , 0.05), n 5 3.

alone, on day 3 (p 5 0.034), day 5 (p , 0.001), day 7 (p , 0.001), day 14 (p , 0.001), and day 21 (p 5 0.007). Subdiploid (apoptotic cells) CD411 cells started to increase on days 14–21. Culturing the cells with TPO and FL significantly increased the proportion of CD411 subdiploid1 cells compared to culturing the cells with TPO alone, on day 14 (p 5 0.009) and day 21 (p 5 0.005). Only the CD411 cells became polyploid in the cultures. Culturing the cells with TPO and FL shows a delayed increase in polyploid CD411 cells, but a more apparent increase in CD412 2N or 4N cells. It is also interesting to observe that FL in combination with TPO decreases the number of subdiploid (apoptotic) CD411 cells compared to TPO alone.

DISCUSSION Megakaryocytopoiesis is a unique pathway in hematopoiesis that requires both mitotic cell divisions and endomitotic nuclear replication for the development of mature MKs. As a general rule, when somatic cells become polyploid, apoptosis is initiated unless cell cycle control has been damanged, e.g., by inactivation of p53 in cancer. The fact that MKs become polyploid but do not enter apoptosis until after platelet release is one of the enigmas of MK biology. Among the serious complications of myeloablative chemotherapy, which usually precedes HSC transplantation, are neutropenia and thrombocytopenia. Transplanting CD341 HSC has been shown to reduce neutropenia, while patients can remain thrombocytopenic for several weeks. This increases the need for platelet transfusions that are costly and increase the risk of infections in these patients. The use of ex vivo-expanded HSC has increased in recent years. This has been followed by the use of ex vivo-generated MK progenitor cells along with HSC to reduce the need for platelet transfusions (4,19,20). The aim of our study was to develop an in vitro serumfree culture model, differentiating BM CD341 HSC into MKs. An additional goal was to study the effects of TPO and FL on MK differentiation and ploidy. The use of a serum-free system to monitor the effects of cytokines on MK development has the advantage of reducing the number of variables in the cell culture. In this study, the purity of the CD341 cells after magnetic isolation was around 80%. This is somewhat lower than has been reported in other studies, where purity of cells isolated form bone marrow ranged from 85% to 97% (25–27). However, that kind of purity of CD341 cells from BM usually requires by further sorting the cells on a cell sorter following the magnetic selection (27). CD341 cells were cultured with TPO or TPO and FL for 21 days.

396

MEGAKARYOCYTE DEVELOPMENT IN VITRO

FIG. 10. Representative dot plots showing changes in CD41 expression and ploidy in CD341 cells cultured with TPO or TPO/FL on days 1, 7, 14, and 21. Ploidy was analyzed by propidium iodide (PI) staining.

The number of viable cells increased until day 14, when the cell number started to decrease. This coincided with simultaneous increase in the number of subdiploid (apoptotic) cells. TPO alone increases the number of viable cells five-fold while TPO and FL increase the number of viable cells 10-fold. TPO has previously been shown to be sufficient to differentiate CD341 cells to MKs, whereas FL alone has little or no effect on either expansion or differentiation. However, FL seems to have a synergistic effect with TPO, increasing the number of viable cells significantly more than TPO alone. Similar results have been found by others using different types of growth factors and medium containing serum (26,27). An increase in the number of viable cells has been shown to be greater when CD341 cells from CB were cultured using TPO and/or TPO and FL as compared to CD341 cells from BM (20). Morphology of cells in culture using Giemsa staining demonstrated apparent changes. Cells with large lobulated nuclei and large cytoplasm became more apparent as the cultures progressed, similar morphological changes have been observed by other investigators using different cytokines and cell sources (25,26,28). Using FSC and SSC profiles to discriminate blood cells from viable cells is a widely accepted method (26,29). When CD341 cells differentiate toward the MK lineage they become larger (higher FSC) and more granular (higher SSC). This can be used to monitor their development and to discriminate then from dead cells and debris. CD341 cells cultured without TPO or TPO and FL

do not differentiate toward the MK lineage and lose their viability early in the cultures. Culturing the cells with TPO alone or TPO and FL had similar effects on the proportion of the CD341 CD412 subset. CD341 cells decreased in numbers as the culture progressed, as demonstrated in other studies using the same kind of cytokines or different cytokine combinations, serum-supplemented or serum-free medium, CD341 cells from BM, or a different source (CB, MPB) (11,25,27). TPO alone increased the proportion of the CD341 CD411 subset (MK progenitors) and the CD342 CD411 subset (MKs) at a higher rate as compared to cells cultured with TPO and FL. A possible explanation is that the addition of FL slows down the differentiation toward the MK lineage. An alternative explanation is that FL in synergy with TPO promotes a higher expansion rate of the CD341 CD412 subset than TPO alone. This could lead to a slower MK development. If this were the case, addition of FL to the TPO cultures would increase the proportion of CD341 CD412 cells or cause a delay in the down-regulation of CD34 expression. Although the difference did not reach statistical significance, there was an apparent trend toward a larger proportion of CD341 CD412 cells where FL is present in the culture. The difference in the CD342 CD411 subset between cultures with TPO or TPO and FL also supports our findings that FL in synergy with TPO slows down MK development. The role of FL in MK development has not been well defined, but FL has been shown to be a good expansion factor in synergy with other growth

397

SIGURJÓNSSON ET AL. factors, including TPO (30). Furthermore, FL has been shown to augment the ability of IL-3 and SCF to promote long-term megakaryocytopoiesis, alone or in association with other cytokines (31). This difference in progression of the cells to the MK lineage, based on CD34 and CD41 expression, led us to monitor MK development further by looking at HLA-DR (DR) expression. It has been demonstrated that CD341 DR1 cells are able to develop into blood cells of all lineages and are therefore still considered to be true stem cells (32,33). The mononuclear cells isolated from BM had an apparent high expression of DR that decreased during the culture period, being significantly higher at later stages in the cultures where TPO and FL were used together. This was confirmed by analyzing the CD34 subset, which showed higher degree of DR expression early in the culture that dropped in the later stages, still being higher where TPO and FL were used together. Discriminating between early and late MK progenitor cells has been done by looking at their DR expression. This has mainly been performed in clonal cultures where it has been shown that late MK progenitor cells express both DR and CD41 (CFU-MK) while the early progenitors (BFU-MK) express neither CD41 nor DR (23,24). In our experiments, culturing the cells with TPO and FL maintained the CD411 DR1 population significantly longer in the cultures compared to TPO alone. On the other hand, CD411 DR2 cells only become prominent at the later stages of the culture period or at the same time as the DR expression starts to decrease. The development of polyploidy is probably the single best indicator of MK differentiation. This unique process starts in the pro-megakaryoblasts (PMKB) and continues throughout MK maturation. It is accompanied by an increase in cytoplasmic volume and cell size. In vivo the ploidy can reach 128N, the cells usually being 16N–32N. However, ploidy seldom reaches more than 16N–32N in vitro, with majority of the cells being 2N–8N. In our experiments, we noted that cells cultured with TPO had higher ploidy (8N–16N) than cells cultured with TPO and FL. There was no difference in cells with lower ploidy (2N–4N) until day 21, where cells cultured with TPO and FL had a higher number of the 2N–4N subsets compared to cells cultured with TPO alone. These results further support our hypothesis that FL may work as a delaying factor on megakaryocytopoiesis. When combined with TPO, FL can increase the number of MK progenitor cells and immature MKs at the expense of their developmental progression. To examine further the effect of FL on polyploidy in megakaryocytopoiesis, we examined the expression of CD41 and ploidy using dual staining and flow cytometry analysis. In our experiments, an increase in ploidy was accompanied by an increase in CD41 expression.

This is in contrast with results from Baatout et al. (34), which demonstrated that an increase in ploidy was correlated with a decrease in CD61 expression. CD61 forms a complex with CD41 in MKs and platelets, and its expression would therefore be expected to be directly correlated to CD41 expression. Our results are, however, in concordance with other studies showing an increase in expression of CD41 with an increase in ploidy (27,35).

CONCLUSIONS Megakaryocytopoiesis is a unique pathway that requires both mitotic cell divisions and endomitotic nuclear replication for the development of mature MKs. Regulation of apoptosis has been shown to be an important factor in MK development, being prevented in MK progenitor cells and most likely immature MKs. Subsequently, this mechanism is activated to accomplish the effective removal of MKs after platelet shedding. Our results suggest a role for FL in megakaryocytopoiesis. In this study, we demonstrated that CD341 cells from BM cultured with TPO and FL show a slower development rate (assessed by phenotypical analysis, ploidy, and morphology), higher expansion rate, and less ploidy compared to cells cultured with TPO alone. These results suggest that FL, in synergy with TPO, may slow down MK development and increase the number of MK progenitor cells, possibly due to the early action of FL at the pluripotent stem cell stage.

ACKNOWLEDGMENTS The investigators would like to thank Halldór Jónsson Jr. M.D., Ph.D., RíkarDur Sigfússon, M.D., Svavar Haraldsson, M.D., Ingibjörg Sigurvinsdóttir, and Bjarni A. Agnarsson, M.D., for their valuable assistance.

398

REFERENCES 1. Kaushansky K. (1999). The enigmatic megakaryocyte gradually reveals its secrets. Bioessays 21:353–360. 2. Kessinger A, J Armitage, P Bierman, M Bishop, S Joshi, E Reed, G Sharp, J Talmadge and J Vose. (1994). Clinical outcome of peripheral blood stem cell support. Med Oncol 11:43–46. 3. Williams SF, JD Bitran, JM Richards, PJ DeChristopher and AR Orlina. (1990). Peripheral blood-derived stem cell collections for use in autologous transplantation after high dose chemotherapy: an alternative approach. Prog Clin Biol Res 333:461–469. 4. Lefebvre P, JN Winter, Y Meng and I Cohen. (2000). Ex vivo expansion of early and late megakaryocyte progenitors. J Hematother Stem Cell Res 9:913–921.

MEGAKARYOCYTE DEVELOPMENT IN VITRO 5. Webb IJ and KC Anderson. (1999). Risks, costs, and alternatives to platelet transfusions. Leuk Lymphoma 34:71–84. 6. Bartley TD, J Bogenberger, P Hunt, YS Li, HS Lu, F Martin, MS Chang, B Samal, JL Nichol, S Swift et al. (1994). Identification and cloning of a megakaryocyte growth and development factor that is a ligand for the cytokine receptor Mpl. Cell 77:1117–1124. 7. de Sauvage FJ, PE Hass, SD Spencer, BE Malloy, AL Gurney, SA Spencer, WC Darbonne, WJ Henzel, SC Wong, WJ Kuang et al. (1994). Stimulation of megakaryocytopoiesis and thrombopoiesis by the c-Mpl ligand. Nature 369:533–538. 8. Kaushansky K, S Lik, RD Holly, VC Broudy, N Lin, MC Bailey, JW Forstrom, MM Buddle, PJ Oort, FS Hagen et al. (1994). Promotion of megakaryocyte progenitor expansion and differentiation by the c-Mpl ligand thrombopoietin. Nature 369:568–571. 9. Broxmeyer HE, L Lu, S Cooper, L Ruggieri, ZH Li and SD Lyman. (1995). Flt3 ligand stimulates/costimulates the growth of myeloid stem/progenitor cells. Exp Hematol 23:1121–1129. 10. Li K, M Yang, AC Lam, FW Yau and PM Yuen. (2000). Effects of flt-3 ligand in combination with TPO on the expansion of megakaryocytic progenitors. Cell Transplant 9:125–131. 11. Ayala IA, A Tomer and KL Kellar. (1996). Flow cytometric analysis of megakaryocyte-associated antigens on CD34 cells and their progeny in liquid culture. Stem Cells 14:320–329. 12. Kaushansky K. (1995). Thrombopoietin: the primary regulator of platelet production. Blood 86:419–431. 13. Jackson CW, LK Brown, BC Somerville, SA Lyles and AT Look. (1984). Two-color flow cytometric measurement of DNA distributions of rat megakaryocytes in unfixed, unfractionated marrow cell suspensions. Blood 63:768–778. 14. Kuter DJ and RD Rosenberg. (1994). Appearance of a megakaryocyte growth-promoting activity, megapoietin, during acute thrombocytopenia in the rabbit. Blood 84:1464–1472. 15. Italiano JE, Jr., P Lecine, RA Shivdasani and JH Hartwig. (1999). Blood platelets are assembled principally at the ends of proplatelet processes produced by differentiated megakaryocytes. J Cell Biol 147:1299–1312. 16. Cramer EM, F Norol, J Guichard, J Breton-Gorius, W Vainchenker, JM Masse and N Debili. (1997). Ultrastructure of platelet formation by human megakaryocytes cultured with the Mpl ligand. Blood 89:2336–2346. 17. Takamatsu Y, M Harada, T Teshima, S Makino, S Inaba, K Akashi, T Shibuya and Y Niho. (1995). Relationship of infused CFU-GM and CFU-Mk mobilized by chemotherapy with or without G-CSF to platelet recovery after autologous blood stem cell transplantation. Exp Hematol 23:8–13. 18. Feng R, C Shimazaki, T Inaba, R Takahashi, H Hirai, T Kikuta, T Sumikuma, N Yamagata, E Ashihara, N Fujita and M Nakagawa. (1998). CD341 /CD41a1 cells best predict platelet recovery after autologous peripheral blood stem cell transplantation. Bone Marrow Transplant 21:1217–1222.

19. Bertolini F, M Battaglia, P Pedrazzoli, GA Da Prada, A Lanza, D Soligo, L Caneva, B Sarina, S Murphy, T Thomas and GR della Cuna. (1997). Megakaryocytic progenitors can be generated ex vivo and safely administered to autologous peripheral blood progenitor cell transplant recipients. Blood 89:2679–2688. 20. Lefebvre P, JN Winter, LE Kahn, JG Giri and I Cohen. (1999). Megakaryocyte ex vivo expansion potential of three hematopoietic sources in serum and serum-free medium. J Hematother 8:199–208. 21. Lefebvre P, JN Winter, AW Rademaker, C Goolsby and I Cohen. (1997). In vitro production of megakaryocytes from PIXY321 versus GM-CSF- mobilized peripheral blood progenitor cells. Stem Cells 15:112–118. 22. Williams JL, GG Pipia, NS Datta and MW Long. (1998). Thrombopoietin requires additional megakaryocyte-active cytokines for optimal ex vivo expansion of megakaryocyte precursor cells. Blood 91:4118–4126. 23. Briddell RA, JE Brandt, JE Straneva, EF Srour and R Hoffman. (1989). Characterization of the human burst-forming unit-megakaryocyte. Blood 74:145–151. 24. Briddell RA and R Hoffman. (1990). Cytokine regulation of the human burst-forming unit-megakaryocyte. Blood 76:516–622. 25. Guerriero R, U Testa, M Gabbianelli, G Mattia, E Montesoro, G Macioce, A Pace, B Ziegler, HJ Hassan and C Peschle. (1995). Unilineage megakaryocytic proliferation and differentiation of purified hematopoietic progenitors in serum-free liquid culture. Blood 86:3725–3736. 26. Zauli G, M Vitale, E Falcieri, D Gibellini, A Bassini, C Celeghini, M Columbaro and S Capitani. (1997). In vitro senescence and apoptotic cell death of human megakaryocytes. Blood 90:2234–2243. 27. van den Oudenrijn S, AE von dem Borne and M de Haas. (2000). Differences in megakaryocyte expansion potential between CD34(1) stem cells derived from cord blood, peripheral blood, and bone marrow from adults and children. Exp Hematol 28:1054–1061. 28. Won JH, SD Cho, SK Park, GT Lee, SH Baick, WS Suh, DS Hong and HS Park. (2000). Thrombopoietin is synergistic with other cytokines for expansion of cord blood progenitor cells. J Hematother Stem Cell Res 9:465–473. 29. Clay D, E Rubinstein, Z Mishal, A Anjo, M Prenant, C Jasmin, C Boucheix and MC Le Bousse-Kerdiles. (2001). CD9 and megakaryocyte differentiation. Blood 97:1982–1989. 30. Piacibello W, F Sanavio, L Garetto, A Severino, D Bergandi, J Ferrario, F Fagioli, M Berger and M Aglietta. (1997). Extensive amplification and self-renewal of human primitive hematopoietic stem cells from cord blood. Blood 89:2644–2653. 31. Piacibello W, L Garetto, F Sanavio, A Severino, L Fubini, A Stacchini, G Dragonetti and M Aglietta. (1996). The effects of human FLT3 ligand on in vitro megakaryocytopoiesis. Exp Hematol 24:340–346. 32. Kim DK, Y Fujiki, T Fukushima, H Ema, A Shibuya and H Nakauchi. (1999). Comparison of hematopoietic activities of human bone marrow and umbilical cord blood CD34 positive and negative cells. Stem Cells 17:286–294. 33. Huss R. (2000). Isolation of primary and immortalized

399

SIGURJÓNSSON ET AL. CD34-hematopoietic and mesenchymal stem cells from various sources. Stem Cells 18:1–9. 34. Baatout S, B Chatelain, P Staquet, M Symann and C Chatelain. (1999). Human megakaryocyte polyploidization is associated with a decrease in GPIIA expression. Anticancer Res 19:5187–5189. 35. Miyazaki R, H Ogata, T Iguchi, S Sogo, T Kushida, T Ito, M Inaba, S Ikehara and Y Kobayashi. (2000). Comparative analyses of megakaryocytes derived from cord blood and bone marrow. Br J Haematol 108:602–609.

Address reprint requests to: Ólafur Eysteinn Sigurjónsson The Blood Bank, Landspitali-University Hospital P.O. Box 1408 101 Reykjavík, Iceland E-mail: [email protected] Received October 18, 2001; accepted November 15, 2001.

400