- Email: [email protected]
Selectins (CD62L, CD62P) and megakaryocytic glycoproteins (CD41a, CD42b) mediate megakaryocyte–fibroblast interactions in human bone marrow
Leukemia Research 24 (2000) 1013 – 1021 www.elsevier.com/locate/leukres
Selectins (CD62L, CD62P) and megakaryocytic glycoproteins (CD41a, CD42b) mediate megakaryocyte–fibroblast interactions in human bone marrow Claudia Wickenhauser *,1, Beate Schmitz1, Stephan Ernst Baldus, Franc Henze, Parvis Farahmand, Semra Frimpong, Ju¨rgen Thiele, Robert Fischer Institute of Pathology, Uni6ersity of Cologne, Joseph-Stelzmann-Str. 9, D-50934 Cologne, Germany Received 29 December 1999; accepted 2 May 2000
Abstract Previous in vitro studies are in keeping with the finding that isolated and enriched megakaryocytes attach to bone marrow fibroblasts and generate an increased growth of these cells. This process was assumed to depend on a close spatial relationship between both cell types which supports the paracrine effect of platelet-derived growth factor (PDGF) and transforming growth factor (TGF)-b1. Moreover, adhesion molecules including b1 integrin receptors and fucosylated structures were determined to play an important role in these complex interactions. However, up to now the influence of megakaryocyte expressed glycoproteins CD41a and CD42b in these processes was not investigated. In addition, the role of megakaryocytic CD62P and also of CD62L, both adhesion molecules of the selectin group, could also be of interest. Following isolation and enrichment of bone marrow megakaryocytes and fibroblasts, both cell populations were characterized regarding their expression of these factors by applying immunocytochemical techniques. Additionally, their influence on adhesion of megakaryocytes to fibroblasts as well as fibroblast growth was evaluated by comparative megakaryocyte– fibroblast co-cultures and inhibition studies using specific monoclonal antibodies (mabs). Fibroblast monocultures served as controls. In these experiments, selectin-specific antibodies significantly reduced megakaryocyte attachment to fibroblast feeder layers and fibroblast growth in the co-cultures. The effect of CD41a and CD42b specific antibodies was limited to megakaryocyte-dependent fibroblast growth. These results elucidate the involvement of the selectins CD62P and CD62L in the basal activation of megakaryocytes inducing their attachment to bone marrow fibroblasts. In contrast, the megakaryocyte glycoproteins CD41a and CD42b exert their effect on the megakaryocyte dependent fibroblast growth. Altogether, it is tempting to speculate that the various interactions of these mediators reflect certain steps in the complex pathomechanisms causing the evolution of (reactive) myelofibrosis in hematopoietic neoplasias accompanied by megakaryocytic proliferation. © 2000 Elsevier Science Ltd. All rights reserved. Keywords: Selectins; Adhesion; Human bone marrow; Megakaryocyte; fibroblast interaction
1. Introduction
Abbre6iations: CMPD, chronic myeloproliferative disorder; FCS, fetal calf serum; FGI, fibroblast growth index; LPA, Limulus polyphemus agglutinin; MAA, Maackia amurensis agglutinin; mab, monoclonal antibody; PDGF, platelet derived growth factor; TGFb1, transforming growth factor ß1; UEA-I, Ulex europaeus agglutinin I. * Corresponding author. Tel.: + 49-221-4785008; fax: + 49-2214786360. E-mail address: [email protected] (C. Wickenhauser). 1 Claudia Wickenhauser and Beate Schmitz contributed equally to this work.
A progressive fibroblast growth and collagen synthesis is often complicating the course of chronic myeloproliferative disorders (CMPD) with megakaryocytic proliferation and acute megakaryoblastic leukemia [1– 7]. In this context, the vicinity of megakaryocytes and fibroblasts suggests a direct megakaryocyte –fibroblast interaction functionally associated with myelofibrosis [1,5,7–9]. In the normal bone marrow a complex interaction between megakaryocytes and fibroblasts regulates the production of proplatelets and their migration
0145-2126/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 1 4 5 - 2 1 2 6 ( 0 0 ) 0 0 0 6 3 - 1
1014
C. Wickenhauser et al. / Leukemia Research 24 (2000) 1013–1021
into the sinusoidal lumina [10 – 12]. Whereas the extravasation of hematopoietic cells and their interactions with the endothelium in inflammatory processes may be mediated by the selectins CD62L and CD62P, the migration of hematopoietic cells into the blood vessels is poorly understood. Obviously, adhesion processes represent an important function by enhancing the megakaryocyte-dependent growth of bone marrow fibroblasts [13,14]. The adherence of megakaryocytes to fibroblasts results in a close spatial relationship of these cell types which seems to be mandatory to exceed the threshold concentrations of megakaryocyte-released platelet-derived growth factor (PDGF) and TGFb1 [13,14]. Both mediators are secreted by megakaryocytes in an active manner and have been found to induce proliferation of fibroblasts as well as secretion of matrix products such as collagen and fibronectin [15 – 18]. Recently we demonstrated that fucosylated structures are involved in these pathomechanisms. The application of lectins like Ulex europaeus agglutinin I (UEA-I), which recognizes terminal a1-2-bound fucose-residues as well as a1-2 fucosylated neoglycoproteins caused a remarkable impairment of fibroblast proliferation in the presence of megakaryocytes [19]. In addition, the adhesion of megakaryocytes to a fibroblast layer decreased significantly during administration of UEA-I [19]. In extension to the above mentioned factors we now analyzed the role of megakaryocyte glycoproteins and selectins regarding their impact on megakaryocyte – fibroblast interaction. In detail, we studied (1) the influence of CD62P (P-selectin; GMP-140; PADGEM) and CD62L (L-selectin; Lecam-1) which are known to recognize sialylated and fucosylated molecules and belong to the selectin family [20,21]; (2) the effect of the glycoprotein CD42b (gp Iba), which is part of the gpIb-V-IX complex [22], and suggested to express UEA-I-binding oligosaccharides and to represent an additional fucosylated structure on the megakaryocyte surface [19]; and (3) the function of the glycoprotein CD41a (gpIIb), which interacts with CD61 (gpIIIa) and forms a receptor mediating the binding to fibrinogen and other ligands mainly involved in platelet aggregation [23,24]. In control experiments, we investigated the capacity of the sialic acid recognizing lectins Limulus polyphemus agglutinin (LPA) and Maackia amurensis agglutinin (MAA) to inhibit megakaryocyte-dependent fibroblast growth.
2. Material and methods
2.1. Bone marrow specimens Following informed consent, sternal human bone marrow was obtained from 67 patients (55 males; 12 females; median age: 64 years; range: 43 – 78 years)
undergoing thoracotomy. The donors presented without any hematological pathology, in particular platelet count was within the normal range. Bone marrow was agitated for several hours in a-minimal essential medium (aMEM) supplemented with 12.5% fetal calf serum (FCS), 12.5% horse serum (all Gibco, Paisley, UK), penicillin (100 U/ml), streptomycin (100 mg/ml), and glutamine (2 mM; all Seromed, Berlin, Germany) at 37°C in a humidified atmosphere. Cells were incubated overnight under the same conditions and used for co-culture experiments during the next day.
2.2. Isolation of megakaryocytes Megakaryocytes were isolated from the bone marrow by a combination of Percoll density centrifugation (1.05 g/ml; Seromed) and immunomagnetic enrichment (magnetic activated cell sorting — MACS). This procedure has already been described in detail [25]. Briefly, after density centrifugation additional centrifugation steps with low rotation (g= 200) were performed to remove platelets. During the enrichment procedure the cell fractions were carefully washed with EDTA (0.5 mM in PBS). Cells of the interphase were washed and incubated with a CD61 specific monoclonal antibody (mab) (Y2/51; 10 mg/ml; DAKO, Hamburg, Germany) [26]. In a second step the primary antibody was coupled to anti-IgG1MACS-beads (1:5; Miltenyi Biotec, Bergisch Gladbach, Germany) and anti-IgG(Fab)2-DTAF (3 mg/ ml: Dianova, Hamburg, Germany). The cells were passed twice over the selection columns. The purity of megakaryocytic enrichment was evaluated by FACScan analysis, non-viable cells were excluded by propidium iodide staining (0.1 mg/ml) (Fig. 1A–B) (Becton–Dickinson, Mountain View, CA, USA). In addition, puritiy of the megakaryocytes and platelet contamination was always checked in CD61 immunostained cytospin preparations.
2.3. Culture of bone marrow fibroblasts Bone marrow cells were cultivated in aMEM/10% FCS (supplemented with penicillin (100 U/ml), streptomycin (100 U/ml) and glutamine (2 mM, Gibco), and Hepes (10 mM). Under these conditions fibroblasts were selected (low serum concentration). When the adherent cells became confluent they were trypsinized and transferred. For experiments, fibroblasts were identified by their characteristic morphology (size, spindleshape). In addition, 40% of the cells presented positivity in the alkaline phosphatase reaction. Immuncytochemical analysis showed a strong positivity with smooth muscle actin (mab 1A4; Dako) and collagen (GCR 1; kindly provided by Dr. Garrido, Granada, Spain). Contamination with monocytes and macrophages was assessed by CD14 and CD68 antibody staining (Dako)
C. Wickenhauser et al. / Leukemia Research 24 (2000) 1013–1021
and the non-specific esterase reaction. According to these staining results the fibroblast populations contained very low amounts of these cell lineages (B1%). Fibroblasts of the third or fourth passage from ten patients were taken for the study.
2.4. Double-immunostaining Cytospin preparations of enriched CD61+-fractions were prepared for double-immunostaining procedures. They were first fixed in 90 and 100% acetone each for 5 min and subsequently incubated overnight with monoclonal antibodies (mabs) raised against CD42b (4 mg/ ml; clone: P2, IgG1k), CD41a (4 mg/ml; clone: SZ2, IgG1k; both Immunotech, Hamburg, Germany), CD62P (2 mg/ml; clone: Ak-6, IgG1k; Southern Biotechnology, Birmingham, AL, USA), and CD62L (2 mg/ml; clone: Sk11, IgG2ak; Camfolio, Becton Dickinson, San Jose, CA, USA). The biotinylated lectin Limulus polyphemus agglutinin (LPA; 0.2 mg/ml; recognizing sialic acid) was purchased from Sigma, Deisenhofen, Germany and Maackia amurensis agglutinin (MAA; 0.1 mg/ml; binding to sialic acid a(2,3) galactose) from EY Laboratories, San Mateo, CA, USA. Consequently,
1015
IgG(Fab)2-Cy3 (10 mg/ml; Dianova, for labelling of the primary antibodies) or Vector Red (Vector, Sigma, for lectin staining) was coupled for 1 h at 37°C. Cytospins were incubated with anti-CD61-FITC antibody (1:10; DAKO) for 2 h at 37°C in a wet chamber. Finally, slides were stained with 4%6-diamidin-2%phenyl-indol-dihydrochlorid (DAPI; 0.1 mg/ml; Boehringer Mannheim, Germany). Subconfluent fibroblasts growing on chamber slides (Lab-Tek, Nunc, Napersville, IL, USA) were fixed in 90 and 100% acetone for 5 min, airdried and incubated overnight in a wet chamber at 4°C with the above mentioned mabs and lectins. Subsequently, fibroblasts were labelled with the adequate fluorescent secondary antibodies IgG(Fab)2-DTAF (10 mg/ml) or anti-biotinFITC (1:80; Sigma) as described. Unbound antibody was removed and the nuclei were counterstained with 0.6 mg/ml propidiumiodide (PJ)/antifade (Oncor, Gainsborough, MD, USA). Evaluation was performed with a fluorescence microscope (Leica DMIL) utilizing a double-bandpass filter (BP 490/20, 575/30) or three-bandpass filter (BP 400/20, 495/15, 570/50) to visualize simultaneously the various fluorescence stainings.
Fig. 1. FACScan analysis of human bone marrow megakaryocytes enriched by a combined method using Percoll density centrifugation and MACS. (A) The histogram shows the distribution of the Y2/51-DTAF positive cells in relation to their size. In addition, the isotype control is presented (B).
1016
C. Wickenhauser et al. / Leukemia Research 24 (2000) 1013–1021
2.5. Cell culture Selected fibroblasts were transferred into 96-multiwell cell culture plates (Falcon, Becton Dickinson) at a density of about 50 cells per well. After 24 h the medium was removed and replaced by aMEM-Glutamax/25%FCS supplemented with penicillin (100 U/ml), streptomycin (100 mg/ml), and 20 mM Hepes. The same batch of FCS was used in all experiments. Before CD61+ cells were added fibroblast count was ascertained by to independent observers. To each well 1000 cells of the CD61+-fraction were added. Fibroblast monocultures without application of additional cells served as controls. All experiments were performed in quadruplicate. Fibroblasts were counted at the beginning of the experiment (ns) and after 6 days, at the end of the experiment (ne). The ratio between the fibroblast numbers indicated the fibroblast growth index (FGI=ne/ns) and was used as a measure for the determination of fibroblast proliferation.
2.6. Adhesion experiments Enriched megakaryocytes, indirectly labelled with anti-CD61/anti-IgG DTAF were suspended in aMEMGlutamax/25% FCS and transferred to 96-multiwell culture plates (2×104 cells/well). Each well contained a near-confluent layer of fibroblasts. The various specific antibodies and lectins were added for 2 h (37°C, humidified atmosphere). Subsequently, the culture plates were carefully washed (aMEM-Glutamax/25% FCS). The assessment of remaining labelled megakaryocytes was semiquantitatively assessed by counting the cells in at least ten different high power fields of the various regions of the wells in comparison with the so-called reference wells (controls) which were incubated without antibodies or lectins applying a Leica inverse fluorescence microscope (DMIL). Always at least two persons were evaluating the test.
CD41a, CD62P, and CD62L. In preliminary studies we analyzed the inhibitory effects of different concentrations of these substances (for every experiment megakaryocytes of three patients were studied). Antibodies against CD 42b, CD41a and CD62L had no inhibitory effect in concentrations up to 0.5 mg/ml. Concentrations of 1 mg/ml had a slight inhibitory effect, this effect was maximized between 5–50 mg/ml, 200 mg/ml had a toxic effect on fibroblasts. Concerning CD62P, concentrations lower than 0.05 mg/ml had no effect on fibroblast count in the co-cultures. 0.1 mg/ml CD62P induced a slight reduction in fibroblast number. The strongest impairment was realized with concentrations between 0.25 and 25 mg/ml, and 200 mg/ml had a toxic effect on fibroblasts in the co-cultures. Finally we took the concentrations as follows: CD42b, CD41a and CD62L 5 mg/ml; CD62P 0.25 mg/ml. In control experiments chromPure mouse pan IgG was used instead of them (5 mg/ml; Dianova). Furthermore, we tested LPA and MAA. Concentrations of 25, 2.5 and 0.25 mg/ml had the same effect on fibroblast count in the co-cultures. Each antibody or lectin was applied at the start and on the 4th day of the experiment to the cell cultures. The selected concentrations were based on the recommendations of the suppliers.
3.1. Statistics For statistical evaluation we employed the Wilcoxon U-test with a level of significance PB 0.05.
4. Results
4.1. Purity of the enriched megakaryocytes FACScan analysis and cytospins of enriched CD61+ populations revealed a megakaryocyte purity of ca. 98% (Fig. 1).
3. Inhibition experiments
4.2. Purity of the fibroblast monocultures
Mabs directed against the following adhesion molecules were used in the cell culture experiments to inhibit megakaryocyte – fibroblast interaction: CD42b,
After 10–12 weeks fibroblasts were selected for experiments and were evaluated by their spindle-shape morphology and their immunological as well as en-
Fig. 2. Immunofluorescent stainings of cultured human bone marrow fibroblasts and megakaryocytes derived from bone marrow smears. Fibroblast labelling is visualized by a green fluorescent (DTAF/FITC) and a red nuclei counterstaining (PJ) whereas megakaryocytes are labelled with a red fluorescent (Vector Red/Cy3) for the demonstration of CD62L, MAA and LPA and with a green fluorescent dye (FITC) for CD61-phenotyping. Nuclei are counterstained with DAPI (blue fluorescent). Panels a, c and e demonstrate fibroblasts and panels b, d and f demonstrate megakaryocytes. (Magnification: a, c, e 50 × ; b, d, f 100 ×). (a, b) Corresponding stainings of fibroblasts and megakaryocytes with CD62L. Staining of the megakaryocytes is negative for CD62L, only the CD61-fluorescent (green) staining is visible. (c, d) MAA-staining of fibroblasts and megakaryocytes. Both cell types react positive with this lectin which is indicated by the green fluorescent of the fibroblasts and the double-fluorescent signal (yellow/orange) of the megakaryocyte. (e, f) LPA-staining of fibroblasts and megakaryocytes which demonstrated in both cases a positive staining: green signal in the case of fibroblasts, yellow/orange signal of the megakaryocytes.
C. Wickenhauser et al. / Leukemia Research 24 (2000) 1013–1021
Fig. 2.
1017
C. Wickenhauser et al. / Leukemia Research 24 (2000) 1013–1021
1018
Table 1 Impairment of megakaryocyte adhesion on a fibroblast feeder layer using antibodies against adhesion molecules and lectinsa Applied antibodies/lectins Number of experiments
Reference wells (controls)
Density of remaining fluorescent megakaryocytes
CD62P CD62L CD42b CD41a MAA LPA IgG
+++ +++ +++/++ +++ +++/++ +++ +++/++
+ +(+) +++/++ +++/++ +++ +++/++ +++/++
5 5 5 5 3 3 5
a The amount of remaining fluorescent megakaryocytes was evaluated by semiquantitative parameters in comparison with the controls (w/o antibody application). +++: high megakaryocyte density. ++: medium megakaryocyte density. +(+): low megakaryocyte density. +: very low megakaryocyte density.
zyme-cytochemical staining properties. The isolated cells demonstrated the properties of a highly purified fibroblastic population with a strong positivity for collagen and a-actin [27]. No obvious differences could be observed between the various fibroblastic populations from ten patients.
4.3. Distribution of the selectins and glycoproteins Immunostaining of cytospin preparations confirmed the expression of CD41a and CD42b as well as CD62P by megakaryocytes while no megakaryocytic staining for CD62L was obtained (Fig. 2b). In control experiments, a reactivity of LPA and MAA (Fig. 2d, f) could be demonstrated. In contrast, in fibroblasts no reaction with mabs against CD41a, CD42b, and CD62P was detectable. Conspicuously, a distinct fibrillar staining pattern was achieved when incubating fibroblasts with a L-selectin specific mab (Fig. 2a). In addition, fibroblasts presented a cytoplasmic and membrane staining with LPA and MAA as could be demonstrated by immuncytochemistry and flow cytometry (Fig. 2c, e).
4.4. Effect of selectins and glycoproteins on megakaryocyte adhesion to fibroblasts Addition of anti-CD62L to the adhesion assay induced a significant decrease in megakaryocyte attachment. Moreover, administration of anti-CD62P yielded an obvious loss of CD61 immunphenotyped megakaryocytes adhering to the fibroblast layer (Table 1). On the other hand, CD41a or CD42b as well as administration of the control chromPure mouse pan IgG or MAA and LPA did not exert any alteration in megakaryocyte adhesion (Table 1). Every test was semiquantitatively evaluated by two independent observers.
4.5. Impairment of fibroblast growth Co-cultures of megakaryocytes and fibroblasts demonstrated a significant increase in fibroblast growth
when compared to fibroblast monocultures (Fig. 3). In contrast, in CD61-depleted cell fractions fibroblast count was not altered. Addition of anti-CD62P to the co-cultures resulted in a relevant decrease in fibroblast growth, whereas no effect on fibroblast monocultures could be observed (Fig. 3). Application of mabs against the glycoproteins CD41a and CD42b or against CD62L generated a similar impairment in fibroblast growth. On the other hand, in fibroblast monocultures no significant increase in fibroblast count could be detected following administration of these mabs (Fig. 3). No proliferation-inducing effect was obtained by applying sialic acid-recognizing lectins MAA and LPA to the megakaryocyte –fibroblast co-cultures. Addition of chromPure mouse IgG yielded no alteration in fibroblast proliferation (Fig. 3). The evaluation of fibroblast growth was analyzed by counting the fibroblasts by two independent observers and in all cases the results of both were identical or varied less than 2%.
5. Discussion Generally, adhesive interactions with the bone marrow microenvironment are responsible for the retention of hematopoietic cells in the marrow and modulate growth and migration of mature hematopoietic cells in circulation. There is mounting evidence that not only cytokine modulated actions but also the direct contact alone play an important role in the regulation of the diverse hematopoietic processes [28–30]. Regarding megakaryopoiesis, a very complex, but until now not very well understood interaction with the hematopoietic stroma allows maturation and shedding of platelets into the bone marrow sinuses. In the present investigation, we analyzed the influence of the megakaryocytic glycoproteins CD41a (gp IIb), CD42b (gp Iba), and CD62P with respect to adhesion of megakaryocytes to bone marrow fibroblasts. Moreover, the effect of these agents on the megakaryocyte-dependent fibroblast proliferation was
C. Wickenhauser et al. / Leukemia Research 24 (2000) 1013–1021
studied. CD41a, known to form a complex with CD61 is randomly dispersed on megakaryocyte and platelet surfaces and is capable of recognizing immobilized fibrinogen [31]. In response to platelet stimulation, however, the complex becomes a receptor for several soluble adhesive proteins including fibrinogen, fibronectin, von Willebrand factor and vitronectin, probably caused by an introduction of a conformation change of this complex [32 – 34]. CD42b, a prominent component of the gpIb-V-IX complex, is also known to form one of the major adhesion receptors on platelet and megakaryocyte surfaces and is essential for the binding to the von Willebrand factor [22], to filamentin [35] and thrombin [36]. CD62P, detected in the membranes of megakaryocyte a granules, is an integral membrane glycoprotein on stimulated platelets and megakaryocytes [37]. Discussion and controversy still exists whether sLeX or LeX is sufficient to support CD62P binding [38,39]. The candidate CD62P ligand
1019
bearing cells so far have included neutrophils and monocytes [40]. Finally, the role of CD62L was analyzed as well. This selectin is described to be expressed on neutrophils, monocytes and the majority of B and T lymphocytes and is best known for regulating leukocyte attachment to endothelium [41]. On the other hand, it is not only detectable in follicular dendritic cells [42] but in bone marrow fibroblasts as we could demonstrate in this study. Similar to the other members of the selectin family, L-selectin behaves as a lectin, recognizing carbohydrate ligands. In this study determination of fibroblast growth and adhesion of megakaryocytes to fibroblasts was evaluated by simple techniques. However, counting of fibroblasts by two independent observers presented low intraindividual deviation (less than 2%) and, in addition, concerning the semiquantitative analysis of the adhesion assays, in none of the cases there was a difference in classification of megakaryocyte attach-
Fig. 3. Influence of various antibodies against adhesion molecules on megakaryocyte-dependent fibroblast growth. Application of 1000 cells/well of the CD61+-fraction initially containing 50 fibroblasts/well. Fibroblasts are counted at the start of the cultures and after 6 days, at the end of the experiments. The ratio between the fibroblast numbers indicated the fibroblast growth index (FGI =ne/ns). All experiments are set up in quadruplicate. Points represent means ( 9 S.D). Stars indicate a significant increase of fibroblast growth (PB 0.05) in the co-culture compared to the monoculture. Arrowheads indicate the significant impairment in fibroblast growth compared to basal co-cultures (PB 0.05). The number of experiments was 27 (CD62P, CD62L) and 20 (CD41a, CD42b). MAA and LPA (0.25, 2.5 and 25 mg/ml) did not alter the megakaryocyte induced fibroblast proliferation (data not shown).
1020
C. Wickenhauser et al. / Leukemia Research 24 (2000) 1013–1021
ment by the two independent observers. Our results demonstrate a direct involvement of the selectins CD62P and CD62L in megakaryocyte adherence to a fibroblast feeder layer. Additionally, antibodies against both selectins generated an impairment of megakaryocyte-dependent fibroblast growth. Together with the missing reaction of sialic acid recognizing lectins these observations underline the important part of fucose residues in the mechanism of megakaryocyte triggered bone marrow fibroblast proliferation [43]. The actions of CD41a and CD42b, however, were limited to the influence of megakaryocyte dependent fibroblast growth and therefore demonstrate the importance of these glycoproteins in activated megakaryocytes. Taken together, these results emphasize a hierarchical role of several megakaryocyte expressed adhesion receptors in the attachment on fibroblasts and induction of fibroblast proliferation. A cascade of adhesion events starting with CD62P as well as fibroblast expressed CD62L interaction may generate a specific activation of fibroblasts and, in consequence, megakaryocytic glycoproteins CD41a and CD42b are needed for the induction of fibroblast proliferation. In addition, results of recently published studies of our group indicate that also the secretion of PDGF and TGFb as well as the megakaryocyte expressed integrins a5b1 and a3b1 may enhance this effect. Consequently, further studies have to elucidate to which extent the various factors inducing fibroblast growth in vitro are involved in the complex pathomechanisms generating myelofibrosis as a reactive phenomenon in several neoplastic diseases accompanied by megakaryocytic proliferation [1,2,5,6,44 – 47]. Acknowledgements The authors are indebted to the staff of the Clinic of Cardiac Surgery (director Prof. E.R. de Vivie, M.D.) of the University of Cologne for supplying the bone marrow tissue. This work was supported by a grant from the ‘Dr Mildred Scheel Stiftung fu¨r Krebsforschung’ (10-1088TH3). C. Wickenhauser and B. Schmitz provided the concept, design, analysis of data, drafted the paper and gave final approval. S.E. Baldus provided critical revision. F. Hrnze and P. Farahmand provided study materials, statistical expertise and assembled the data. S. Frimpong provided technical support. J. Thiele contributed to the design of the study, gave critical input to the revision and gave final approval. R. Fischer gave critical input and gave final approval to the article. References [1] Burkhardt R, Bartl R, Jaeger K, Frisch B, Kettner G, Mahl G, Sund M. Chronic myeloproliferative disorders (CMPD). Pathol Res Pract 1984;179:131.
[2] McCarthy D. Fibrosis of the bone marrow: content and causes. Br J Haematol 1985;59:1. [3] Dekmezian R, Kantarjian HM, Keating M, Talpaz M, McCredie KB, Freireich EJ. The relevance of reticulin stain-measured fibrosis at diagnosis in chronic myelogeneous leukemia. Cancer 1987;59:1739. [4] Greenberg BR, Woo L, Veomett IC, Payne CM, Ahmann FR. Cytogenetics of bone marrow fibroblastic cells in idiopathic chronic myelofibrosis. Br J Haematol 1987;66:487. [5] Thiele J, Fischer R. Megakaryocytopoiesis in haematological disorders: diagnostic features of bone marrow biopsies. Virchows Arch A Pathol Anat 1991;418:87. [6] Wang JC, Lang HD, Lichter S, Weinstein M, Benn P. Cytogenetic studies of bone marrow fibroblasts cultured from patients with myelofibrosis and myeloid metaplasia. Br J Haematol 1992;80:184. [7] Thiele J, Kvasnicka HM, Titius BR, Parpert U, Nebel R, Zankovich R, Dienemann D, Stein H, Diehl V, Fischer R. Histological features of prognostic significance in CML — an immunohistological and morphometric study (multivariate regression analysis) on trephine biopsies of the bone marrow. Ann Hematol 1993;66:291. [8] Thiele J, Hoeppner B, Zankovich R, Fischer R. Histomorphometry of bone marrow biopsies in primary osteomyelofibrosis/-sclerosis (agnogenic myeloid metaplasia) — correlations between clinical and morphological features. Virchows Arch A Pathol Anat 1989;415:191. [9] Buhr T, Choritz H, Georgii A. The impact of megakaryocyte proliferation for the evolution of myelofibrosis. Histological follow-up study in 186 patients with chronic myeloid leukemia. Virchows Arch A 1992;420:473. [10] Becker RP, De Bruyn PPH. The transmural passage of blood cells into myeloid sinusoids and the entry of platelets into the sinusoidal circulation: a scanning electron microscopic investigation. Am J Anat 1976;145:183. [11] Muto MA. Scanning and transmission electron microscopic study on rat bone marrow sinuses and transmural migration of blood cells. Arch Histol Jpn 1976;39:51. [12] Petrides PE, Dittman KH. How do normal and leukemic white blood cells egress from the bone marrow? Blut 1990;61:3. [13] Schmitz B, Thiele J, Kaufmann R, Witte O, Wickenhauser C, Kuhn-Regnier F, Fischer R. Megakaryocytes and fibroblasts-interactions determined in normal human bone marrow specimens. Leuk Res 1995;19:629. [14] Schmitz B, Thiele J, Witte O, Kaufmann R, Wickenhauser C, Fischer R. Influence of cytokines (IL-1a, IL-3, IL-11, GM-CSF) on megakaryocyte-fibroblast interactions in normal human bone marrow. Eur J Hematol 1995;55:24. [15] Castro-Malaspina H, Rabellino EM, Yen A, Nachman RL, Moore MAS. Human megakaryocyte stimulation of bone marrow fibroblasts. Blood 1981;57:781. [16] Kimura A, Katoh O, Kuramoto A. Effects of platelet-derived growth factor, epidermal growth factor and transforming growth factor-b on the growth of human marrow fibroblasts. Br J Haematol 1988;69:9. [17] Kimura A, Katoh O, Hyodo H, Kuramoto A. Transforming growth factor-b regulates growth as well as collagen and fibronectin synthesis of human marrow fibroblasts. Br J Haematol 1989;72:486. [18] Wickenhauser C, Hillienhof A, Jungheim C, Lorenzen J, Ruskowski H, Hansmann M-L, Thiele J, Fischer R. Detection and quantification of transforming growth factor b and plateletderived growth factor release by normal human megakaryocytes. Leukemia 1995;9:310. [19] Schmitz B, Thiele J, Otto F, Theile-Ochel S, Heedt T, Zensen U, Baldus SE, Wickenhauser C, Fischer R. Interactions between endogeneous lectins and fucosylated oligosaccharides in
C. Wickenhauser et al. / Leukemia Research 24 (2000) 1013–1021
[20]
[21] [22]
[23]
[24] [25]
[26]
[27]
[28]
[29]
[30]
[31] [32] [33]
megakaryocyte-dependent fibroblast growth of the normal bone marrow. Leukemia 1996;10:1604. Picker LJ, Warnock RA, Burns AR, Doerschuk CM, Berg EL, Butcher EC. The neutrophil selectin LECAM-1 represents carbohydrate ligands to the vascular selectins ELAM-1 and GMP-140. Cell 1991;66:921. Springer TA, Lasky LA. Sticky sugars for selectins. Nature 1991;349:196. Clemetson KJ, Clemetson JN. Platelet GPIb-V-IX complex: structure, function, physiology and pathology. Semin Thromb Hemostas 1995;21:130. Block K, Poncz M. Platelet glycoprotein Ib gene expression as a model of megakaryocyte-specific expression. Stem Cells 1995;13:135. Hynes RO. Integrins: Versatility, modulation, and signalling in cell adhesion. Cell 1992;69:11. Schmitz B, Radbruch A, Ku¨mmel T, Wickenhauser C, Korb H, Hansmann M-L, Thiele J, Fischer R. Magnetic activated cell sorting (MACS) — a new immunological method for megakaryocytic cell isolation: Comparison of different separation techniques. Eur J Hematol 1994;52:267. Gatter KC, Cordell JL, Turley H, Heryet A, Kieffer N, Anstee DJ, Mason DY. The immunohistological detection of platelets, megakaryocytes and thrombi in routinely processed specimens. Histopathology 1988;13:257. Schmitz B, Park IA, Kaufmann R, Thiele J, Fischer R. Influence of cytokine stimulation (granulocyte macrophage-colony stimulating factor, interleukin-3, and transforming growth factor-beta1) on adhesion molecule expression in normal human bone marrow fibroblasts. Acta Haematol 1995;94:173. Eaves CJ, Cashman JD, Sutherland HJ, Otsuka T, Humphries RK, Hogge DE, Lansdorp PL, Eaves AC. Molecular analysis of primitive hematopoietic cell proliferation control mechanisms. Ann NY Acad Sci 1991;628:298. Hurley RW, McCarthy JB, Verfaille CM. Direct adhesion to bone marrow stroma via fibronectin receptors inhibits hematopoietic progenitor proliferation. J Clin Invest 1995;96:511. Verfaillie CM, Catanzaro P. Direct contact with stroma inhibits proliferation of human long-term culture initiating cells. Leukemia 1996;10:498. Kieffer N, Phillips DR. Platelet membrane glycoproteins: functions in cellular interactions. Annu Rev Cell Biol 1990;6:329. Plow EF, Ginsberg MH. Cellular adhesion: GPIIb-IIIa as a prototype adhesion receptor. Prog Hemost Thromb 1989;9:117. Peterson JA, Visentin GP, Newman PJ, Aster RH. A recombinant soluble form of the integrin aIIbb3 (gpIIb-IIIa) assumes an active, ligand-binding conformation and is recognized by gpIIbIIIa specific monoclonal, allo-, auto-, and drug-dependent platelet antibodies. Blood 1998;92:2053.
.
1021
[34] Schick PK, Wojenski CM, He X, Walker J, Marcinkiewicz C, Niewiarowski S. Integrins involved in the adhesion of megakaryocytes to fibronectin and fibrinogen. Blood 1998;92:2650. [35] Andrews RK, Fox JEB. Interaction of purified actin-binding protein with the platelet membrane glycoprotein Ib-IX complex. J Biol Chem 1991;266:7144. [36] De Marco L, Mazzucato M, Masotti A, Ruggeri ZM. Localization and characterization of an a-thrombin-binding site on platelet glycoprotein Iba. J Biol Chem 1994;269:6478. [37] Berman C, Yeo EL, Wemncel-Drake JD, Furie BC, Ginsberg MH, Furie B. A platelet alpha granule membrane protein that is associated with the plasma membrane after activation. J Clin Invest 1986;78:130. [38] Larsen GR, Sako D, Ahern TJ, Shaffer M, Erban J, Sajer SA, Gibson RM, Wagner DD, Furie BC, Furie B. P-selectin and E-selectin. Distinct but overlapping leukocyte ligand specifities. J Biol Chem 1992;267:104. [39] Polley MJ, Phillips ML, Wayner E, Nudelman E, Singhal AK, Hakomori S, Paulson JC. CD62 and endothelial cell-leukocyte adhesion molecule 1 (ELAM-1) recognize the same carbohydrate ligand, sialyl-Lewis X. Proc Natl Acad Sci USA 1991;88:6224. [40] Larsen E, Celi A, Gilbert GE, Furie BC, Erban JK, Bonfanti R, Wagner DD, Furie B. PADGEM protein: A receptor that mediates the interaction of activated platelets with neutrophils and monocytes. Cell 1989;59:305. [41] Springer TA. Traffic signals for lymphocyte recirculation and leukocyte emigration: The multistep paradigm. Cell 1994;76:301. [42] Munro JM, Briscoe DM, Tedder TF. Differential regulation of leukocyte L-selectin (CD62L) expression in normal lymphoid and inflamed extralymphoid tissues. J Clin Pathol 1996;49:721. [43] Schmitz B, Thiele J, Otto F, Farahmand P, Henze F, Frimpong S, Wickenhauser C, Fischer R. Evidence for integrin receptor involvement in megakaryocyte-fibroblast interaction — a possible pathomechanism for the evolution of myelofibrosis. J Cell Physiol 1998;176:445. [44] Martyre´ MC. Platelet PDGF and TGF: levels in myeloproliferative disorders. Leukemia Lymphoma 1991;6:1. [45] Martyre´ MC, Romquin N, Bousse-Kerdiles MC, Chewillard S, Benyahia B, Dupriez B, Demory JL, Bauters F. Transforming growth factor-b and megakaryocytes in the pathogenesis of idiopathic myelofibrosis. Br J Haematol 1994;88:9. [46] Martyre´ MC. TGF-b and megakaryocytes in the pathogenesis of myelofibrosis in myeloproliferative disorders. Leukemia Lymphoma 1995;20:39. [47] Thiele J, Kvasnicka HM, Fischer R, Diehl V. Clinicopathological impact of the interaction between megakaryocytes and myeloid stroma in chronic myeloproliferative disorders: a concise update. Leukemia Lymphoma 1997;24:463.
Selectins (CD62L, CD62P) and megakaryocytic glycoproteins (CD41a, CD42b) mediate megakaryocyte–fibroblast interactions in human bone marrow Claudia Wickenhauser *,1, Beate Schmitz1, Stephan Ernst Baldus, Franc Henze, Parvis Farahmand, Semra Frimpong, Ju¨rgen Thiele, Robert Fischer Institute of Pathology, Uni6ersity of Cologne, Joseph-Stelzmann-Str. 9, D-50934 Cologne, Germany Received 29 December 1999; accepted 2 May 2000
Abstract Previous in vitro studies are in keeping with the finding that isolated and enriched megakaryocytes attach to bone marrow fibroblasts and generate an increased growth of these cells. This process was assumed to depend on a close spatial relationship between both cell types which supports the paracrine effect of platelet-derived growth factor (PDGF) and transforming growth factor (TGF)-b1. Moreover, adhesion molecules including b1 integrin receptors and fucosylated structures were determined to play an important role in these complex interactions. However, up to now the influence of megakaryocyte expressed glycoproteins CD41a and CD42b in these processes was not investigated. In addition, the role of megakaryocytic CD62P and also of CD62L, both adhesion molecules of the selectin group, could also be of interest. Following isolation and enrichment of bone marrow megakaryocytes and fibroblasts, both cell populations were characterized regarding their expression of these factors by applying immunocytochemical techniques. Additionally, their influence on adhesion of megakaryocytes to fibroblasts as well as fibroblast growth was evaluated by comparative megakaryocyte– fibroblast co-cultures and inhibition studies using specific monoclonal antibodies (mabs). Fibroblast monocultures served as controls. In these experiments, selectin-specific antibodies significantly reduced megakaryocyte attachment to fibroblast feeder layers and fibroblast growth in the co-cultures. The effect of CD41a and CD42b specific antibodies was limited to megakaryocyte-dependent fibroblast growth. These results elucidate the involvement of the selectins CD62P and CD62L in the basal activation of megakaryocytes inducing their attachment to bone marrow fibroblasts. In contrast, the megakaryocyte glycoproteins CD41a and CD42b exert their effect on the megakaryocyte dependent fibroblast growth. Altogether, it is tempting to speculate that the various interactions of these mediators reflect certain steps in the complex pathomechanisms causing the evolution of (reactive) myelofibrosis in hematopoietic neoplasias accompanied by megakaryocytic proliferation. © 2000 Elsevier Science Ltd. All rights reserved. Keywords: Selectins; Adhesion; Human bone marrow; Megakaryocyte; fibroblast interaction
1. Introduction
Abbre6iations: CMPD, chronic myeloproliferative disorder; FCS, fetal calf serum; FGI, fibroblast growth index; LPA, Limulus polyphemus agglutinin; MAA, Maackia amurensis agglutinin; mab, monoclonal antibody; PDGF, platelet derived growth factor; TGFb1, transforming growth factor ß1; UEA-I, Ulex europaeus agglutinin I. * Corresponding author. Tel.: + 49-221-4785008; fax: + 49-2214786360. E-mail address: [email protected] (C. Wickenhauser). 1 Claudia Wickenhauser and Beate Schmitz contributed equally to this work.
A progressive fibroblast growth and collagen synthesis is often complicating the course of chronic myeloproliferative disorders (CMPD) with megakaryocytic proliferation and acute megakaryoblastic leukemia [1– 7]. In this context, the vicinity of megakaryocytes and fibroblasts suggests a direct megakaryocyte –fibroblast interaction functionally associated with myelofibrosis [1,5,7–9]. In the normal bone marrow a complex interaction between megakaryocytes and fibroblasts regulates the production of proplatelets and their migration
0145-2126/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 1 4 5 - 2 1 2 6 ( 0 0 ) 0 0 0 6 3 - 1
1014
C. Wickenhauser et al. / Leukemia Research 24 (2000) 1013–1021
into the sinusoidal lumina [10 – 12]. Whereas the extravasation of hematopoietic cells and their interactions with the endothelium in inflammatory processes may be mediated by the selectins CD62L and CD62P, the migration of hematopoietic cells into the blood vessels is poorly understood. Obviously, adhesion processes represent an important function by enhancing the megakaryocyte-dependent growth of bone marrow fibroblasts [13,14]. The adherence of megakaryocytes to fibroblasts results in a close spatial relationship of these cell types which seems to be mandatory to exceed the threshold concentrations of megakaryocyte-released platelet-derived growth factor (PDGF) and TGFb1 [13,14]. Both mediators are secreted by megakaryocytes in an active manner and have been found to induce proliferation of fibroblasts as well as secretion of matrix products such as collagen and fibronectin [15 – 18]. Recently we demonstrated that fucosylated structures are involved in these pathomechanisms. The application of lectins like Ulex europaeus agglutinin I (UEA-I), which recognizes terminal a1-2-bound fucose-residues as well as a1-2 fucosylated neoglycoproteins caused a remarkable impairment of fibroblast proliferation in the presence of megakaryocytes [19]. In addition, the adhesion of megakaryocytes to a fibroblast layer decreased significantly during administration of UEA-I [19]. In extension to the above mentioned factors we now analyzed the role of megakaryocyte glycoproteins and selectins regarding their impact on megakaryocyte – fibroblast interaction. In detail, we studied (1) the influence of CD62P (P-selectin; GMP-140; PADGEM) and CD62L (L-selectin; Lecam-1) which are known to recognize sialylated and fucosylated molecules and belong to the selectin family [20,21]; (2) the effect of the glycoprotein CD42b (gp Iba), which is part of the gpIb-V-IX complex [22], and suggested to express UEA-I-binding oligosaccharides and to represent an additional fucosylated structure on the megakaryocyte surface [19]; and (3) the function of the glycoprotein CD41a (gpIIb), which interacts with CD61 (gpIIIa) and forms a receptor mediating the binding to fibrinogen and other ligands mainly involved in platelet aggregation [23,24]. In control experiments, we investigated the capacity of the sialic acid recognizing lectins Limulus polyphemus agglutinin (LPA) and Maackia amurensis agglutinin (MAA) to inhibit megakaryocyte-dependent fibroblast growth.
2. Material and methods
2.1. Bone marrow specimens Following informed consent, sternal human bone marrow was obtained from 67 patients (55 males; 12 females; median age: 64 years; range: 43 – 78 years)
undergoing thoracotomy. The donors presented without any hematological pathology, in particular platelet count was within the normal range. Bone marrow was agitated for several hours in a-minimal essential medium (aMEM) supplemented with 12.5% fetal calf serum (FCS), 12.5% horse serum (all Gibco, Paisley, UK), penicillin (100 U/ml), streptomycin (100 mg/ml), and glutamine (2 mM; all Seromed, Berlin, Germany) at 37°C in a humidified atmosphere. Cells were incubated overnight under the same conditions and used for co-culture experiments during the next day.
2.2. Isolation of megakaryocytes Megakaryocytes were isolated from the bone marrow by a combination of Percoll density centrifugation (1.05 g/ml; Seromed) and immunomagnetic enrichment (magnetic activated cell sorting — MACS). This procedure has already been described in detail [25]. Briefly, after density centrifugation additional centrifugation steps with low rotation (g= 200) were performed to remove platelets. During the enrichment procedure the cell fractions were carefully washed with EDTA (0.5 mM in PBS). Cells of the interphase were washed and incubated with a CD61 specific monoclonal antibody (mab) (Y2/51; 10 mg/ml; DAKO, Hamburg, Germany) [26]. In a second step the primary antibody was coupled to anti-IgG1MACS-beads (1:5; Miltenyi Biotec, Bergisch Gladbach, Germany) and anti-IgG(Fab)2-DTAF (3 mg/ ml: Dianova, Hamburg, Germany). The cells were passed twice over the selection columns. The purity of megakaryocytic enrichment was evaluated by FACScan analysis, non-viable cells were excluded by propidium iodide staining (0.1 mg/ml) (Fig. 1A–B) (Becton–Dickinson, Mountain View, CA, USA). In addition, puritiy of the megakaryocytes and platelet contamination was always checked in CD61 immunostained cytospin preparations.
2.3. Culture of bone marrow fibroblasts Bone marrow cells were cultivated in aMEM/10% FCS (supplemented with penicillin (100 U/ml), streptomycin (100 U/ml) and glutamine (2 mM, Gibco), and Hepes (10 mM). Under these conditions fibroblasts were selected (low serum concentration). When the adherent cells became confluent they were trypsinized and transferred. For experiments, fibroblasts were identified by their characteristic morphology (size, spindleshape). In addition, 40% of the cells presented positivity in the alkaline phosphatase reaction. Immuncytochemical analysis showed a strong positivity with smooth muscle actin (mab 1A4; Dako) and collagen (GCR 1; kindly provided by Dr. Garrido, Granada, Spain). Contamination with monocytes and macrophages was assessed by CD14 and CD68 antibody staining (Dako)
C. Wickenhauser et al. / Leukemia Research 24 (2000) 1013–1021
and the non-specific esterase reaction. According to these staining results the fibroblast populations contained very low amounts of these cell lineages (B1%). Fibroblasts of the third or fourth passage from ten patients were taken for the study.
2.4. Double-immunostaining Cytospin preparations of enriched CD61+-fractions were prepared for double-immunostaining procedures. They were first fixed in 90 and 100% acetone each for 5 min and subsequently incubated overnight with monoclonal antibodies (mabs) raised against CD42b (4 mg/ ml; clone: P2, IgG1k), CD41a (4 mg/ml; clone: SZ2, IgG1k; both Immunotech, Hamburg, Germany), CD62P (2 mg/ml; clone: Ak-6, IgG1k; Southern Biotechnology, Birmingham, AL, USA), and CD62L (2 mg/ml; clone: Sk11, IgG2ak; Camfolio, Becton Dickinson, San Jose, CA, USA). The biotinylated lectin Limulus polyphemus agglutinin (LPA; 0.2 mg/ml; recognizing sialic acid) was purchased from Sigma, Deisenhofen, Germany and Maackia amurensis agglutinin (MAA; 0.1 mg/ml; binding to sialic acid a(2,3) galactose) from EY Laboratories, San Mateo, CA, USA. Consequently,
1015
IgG(Fab)2-Cy3 (10 mg/ml; Dianova, for labelling of the primary antibodies) or Vector Red (Vector, Sigma, for lectin staining) was coupled for 1 h at 37°C. Cytospins were incubated with anti-CD61-FITC antibody (1:10; DAKO) for 2 h at 37°C in a wet chamber. Finally, slides were stained with 4%6-diamidin-2%phenyl-indol-dihydrochlorid (DAPI; 0.1 mg/ml; Boehringer Mannheim, Germany). Subconfluent fibroblasts growing on chamber slides (Lab-Tek, Nunc, Napersville, IL, USA) were fixed in 90 and 100% acetone for 5 min, airdried and incubated overnight in a wet chamber at 4°C with the above mentioned mabs and lectins. Subsequently, fibroblasts were labelled with the adequate fluorescent secondary antibodies IgG(Fab)2-DTAF (10 mg/ml) or anti-biotinFITC (1:80; Sigma) as described. Unbound antibody was removed and the nuclei were counterstained with 0.6 mg/ml propidiumiodide (PJ)/antifade (Oncor, Gainsborough, MD, USA). Evaluation was performed with a fluorescence microscope (Leica DMIL) utilizing a double-bandpass filter (BP 490/20, 575/30) or three-bandpass filter (BP 400/20, 495/15, 570/50) to visualize simultaneously the various fluorescence stainings.
Fig. 1. FACScan analysis of human bone marrow megakaryocytes enriched by a combined method using Percoll density centrifugation and MACS. (A) The histogram shows the distribution of the Y2/51-DTAF positive cells in relation to their size. In addition, the isotype control is presented (B).
1016
C. Wickenhauser et al. / Leukemia Research 24 (2000) 1013–1021
2.5. Cell culture Selected fibroblasts were transferred into 96-multiwell cell culture plates (Falcon, Becton Dickinson) at a density of about 50 cells per well. After 24 h the medium was removed and replaced by aMEM-Glutamax/25%FCS supplemented with penicillin (100 U/ml), streptomycin (100 mg/ml), and 20 mM Hepes. The same batch of FCS was used in all experiments. Before CD61+ cells were added fibroblast count was ascertained by to independent observers. To each well 1000 cells of the CD61+-fraction were added. Fibroblast monocultures without application of additional cells served as controls. All experiments were performed in quadruplicate. Fibroblasts were counted at the beginning of the experiment (ns) and after 6 days, at the end of the experiment (ne). The ratio between the fibroblast numbers indicated the fibroblast growth index (FGI=ne/ns) and was used as a measure for the determination of fibroblast proliferation.
2.6. Adhesion experiments Enriched megakaryocytes, indirectly labelled with anti-CD61/anti-IgG DTAF were suspended in aMEMGlutamax/25% FCS and transferred to 96-multiwell culture plates (2×104 cells/well). Each well contained a near-confluent layer of fibroblasts. The various specific antibodies and lectins were added for 2 h (37°C, humidified atmosphere). Subsequently, the culture plates were carefully washed (aMEM-Glutamax/25% FCS). The assessment of remaining labelled megakaryocytes was semiquantitatively assessed by counting the cells in at least ten different high power fields of the various regions of the wells in comparison with the so-called reference wells (controls) which were incubated without antibodies or lectins applying a Leica inverse fluorescence microscope (DMIL). Always at least two persons were evaluating the test.
CD41a, CD62P, and CD62L. In preliminary studies we analyzed the inhibitory effects of different concentrations of these substances (for every experiment megakaryocytes of three patients were studied). Antibodies against CD 42b, CD41a and CD62L had no inhibitory effect in concentrations up to 0.5 mg/ml. Concentrations of 1 mg/ml had a slight inhibitory effect, this effect was maximized between 5–50 mg/ml, 200 mg/ml had a toxic effect on fibroblasts. Concerning CD62P, concentrations lower than 0.05 mg/ml had no effect on fibroblast count in the co-cultures. 0.1 mg/ml CD62P induced a slight reduction in fibroblast number. The strongest impairment was realized with concentrations between 0.25 and 25 mg/ml, and 200 mg/ml had a toxic effect on fibroblasts in the co-cultures. Finally we took the concentrations as follows: CD42b, CD41a and CD62L 5 mg/ml; CD62P 0.25 mg/ml. In control experiments chromPure mouse pan IgG was used instead of them (5 mg/ml; Dianova). Furthermore, we tested LPA and MAA. Concentrations of 25, 2.5 and 0.25 mg/ml had the same effect on fibroblast count in the co-cultures. Each antibody or lectin was applied at the start and on the 4th day of the experiment to the cell cultures. The selected concentrations were based on the recommendations of the suppliers.
3.1. Statistics For statistical evaluation we employed the Wilcoxon U-test with a level of significance PB 0.05.
4. Results
4.1. Purity of the enriched megakaryocytes FACScan analysis and cytospins of enriched CD61+ populations revealed a megakaryocyte purity of ca. 98% (Fig. 1).
3. Inhibition experiments
4.2. Purity of the fibroblast monocultures
Mabs directed against the following adhesion molecules were used in the cell culture experiments to inhibit megakaryocyte – fibroblast interaction: CD42b,
After 10–12 weeks fibroblasts were selected for experiments and were evaluated by their spindle-shape morphology and their immunological as well as en-
Fig. 2. Immunofluorescent stainings of cultured human bone marrow fibroblasts and megakaryocytes derived from bone marrow smears. Fibroblast labelling is visualized by a green fluorescent (DTAF/FITC) and a red nuclei counterstaining (PJ) whereas megakaryocytes are labelled with a red fluorescent (Vector Red/Cy3) for the demonstration of CD62L, MAA and LPA and with a green fluorescent dye (FITC) for CD61-phenotyping. Nuclei are counterstained with DAPI (blue fluorescent). Panels a, c and e demonstrate fibroblasts and panels b, d and f demonstrate megakaryocytes. (Magnification: a, c, e 50 × ; b, d, f 100 ×). (a, b) Corresponding stainings of fibroblasts and megakaryocytes with CD62L. Staining of the megakaryocytes is negative for CD62L, only the CD61-fluorescent (green) staining is visible. (c, d) MAA-staining of fibroblasts and megakaryocytes. Both cell types react positive with this lectin which is indicated by the green fluorescent of the fibroblasts and the double-fluorescent signal (yellow/orange) of the megakaryocyte. (e, f) LPA-staining of fibroblasts and megakaryocytes which demonstrated in both cases a positive staining: green signal in the case of fibroblasts, yellow/orange signal of the megakaryocytes.
C. Wickenhauser et al. / Leukemia Research 24 (2000) 1013–1021
Fig. 2.
1017
C. Wickenhauser et al. / Leukemia Research 24 (2000) 1013–1021
1018
Table 1 Impairment of megakaryocyte adhesion on a fibroblast feeder layer using antibodies against adhesion molecules and lectinsa Applied antibodies/lectins Number of experiments
Reference wells (controls)
Density of remaining fluorescent megakaryocytes
CD62P CD62L CD42b CD41a MAA LPA IgG
+++ +++ +++/++ +++ +++/++ +++ +++/++
+ +(+) +++/++ +++/++ +++ +++/++ +++/++
5 5 5 5 3 3 5
a The amount of remaining fluorescent megakaryocytes was evaluated by semiquantitative parameters in comparison with the controls (w/o antibody application). +++: high megakaryocyte density. ++: medium megakaryocyte density. +(+): low megakaryocyte density. +: very low megakaryocyte density.
zyme-cytochemical staining properties. The isolated cells demonstrated the properties of a highly purified fibroblastic population with a strong positivity for collagen and a-actin [27]. No obvious differences could be observed between the various fibroblastic populations from ten patients.
4.3. Distribution of the selectins and glycoproteins Immunostaining of cytospin preparations confirmed the expression of CD41a and CD42b as well as CD62P by megakaryocytes while no megakaryocytic staining for CD62L was obtained (Fig. 2b). In control experiments, a reactivity of LPA and MAA (Fig. 2d, f) could be demonstrated. In contrast, in fibroblasts no reaction with mabs against CD41a, CD42b, and CD62P was detectable. Conspicuously, a distinct fibrillar staining pattern was achieved when incubating fibroblasts with a L-selectin specific mab (Fig. 2a). In addition, fibroblasts presented a cytoplasmic and membrane staining with LPA and MAA as could be demonstrated by immuncytochemistry and flow cytometry (Fig. 2c, e).
4.4. Effect of selectins and glycoproteins on megakaryocyte adhesion to fibroblasts Addition of anti-CD62L to the adhesion assay induced a significant decrease in megakaryocyte attachment. Moreover, administration of anti-CD62P yielded an obvious loss of CD61 immunphenotyped megakaryocytes adhering to the fibroblast layer (Table 1). On the other hand, CD41a or CD42b as well as administration of the control chromPure mouse pan IgG or MAA and LPA did not exert any alteration in megakaryocyte adhesion (Table 1). Every test was semiquantitatively evaluated by two independent observers.
4.5. Impairment of fibroblast growth Co-cultures of megakaryocytes and fibroblasts demonstrated a significant increase in fibroblast growth
when compared to fibroblast monocultures (Fig. 3). In contrast, in CD61-depleted cell fractions fibroblast count was not altered. Addition of anti-CD62P to the co-cultures resulted in a relevant decrease in fibroblast growth, whereas no effect on fibroblast monocultures could be observed (Fig. 3). Application of mabs against the glycoproteins CD41a and CD42b or against CD62L generated a similar impairment in fibroblast growth. On the other hand, in fibroblast monocultures no significant increase in fibroblast count could be detected following administration of these mabs (Fig. 3). No proliferation-inducing effect was obtained by applying sialic acid-recognizing lectins MAA and LPA to the megakaryocyte –fibroblast co-cultures. Addition of chromPure mouse IgG yielded no alteration in fibroblast proliferation (Fig. 3). The evaluation of fibroblast growth was analyzed by counting the fibroblasts by two independent observers and in all cases the results of both were identical or varied less than 2%.
5. Discussion Generally, adhesive interactions with the bone marrow microenvironment are responsible for the retention of hematopoietic cells in the marrow and modulate growth and migration of mature hematopoietic cells in circulation. There is mounting evidence that not only cytokine modulated actions but also the direct contact alone play an important role in the regulation of the diverse hematopoietic processes [28–30]. Regarding megakaryopoiesis, a very complex, but until now not very well understood interaction with the hematopoietic stroma allows maturation and shedding of platelets into the bone marrow sinuses. In the present investigation, we analyzed the influence of the megakaryocytic glycoproteins CD41a (gp IIb), CD42b (gp Iba), and CD62P with respect to adhesion of megakaryocytes to bone marrow fibroblasts. Moreover, the effect of these agents on the megakaryocyte-dependent fibroblast proliferation was
C. Wickenhauser et al. / Leukemia Research 24 (2000) 1013–1021
studied. CD41a, known to form a complex with CD61 is randomly dispersed on megakaryocyte and platelet surfaces and is capable of recognizing immobilized fibrinogen [31]. In response to platelet stimulation, however, the complex becomes a receptor for several soluble adhesive proteins including fibrinogen, fibronectin, von Willebrand factor and vitronectin, probably caused by an introduction of a conformation change of this complex [32 – 34]. CD42b, a prominent component of the gpIb-V-IX complex, is also known to form one of the major adhesion receptors on platelet and megakaryocyte surfaces and is essential for the binding to the von Willebrand factor [22], to filamentin [35] and thrombin [36]. CD62P, detected in the membranes of megakaryocyte a granules, is an integral membrane glycoprotein on stimulated platelets and megakaryocytes [37]. Discussion and controversy still exists whether sLeX or LeX is sufficient to support CD62P binding [38,39]. The candidate CD62P ligand
1019
bearing cells so far have included neutrophils and monocytes [40]. Finally, the role of CD62L was analyzed as well. This selectin is described to be expressed on neutrophils, monocytes and the majority of B and T lymphocytes and is best known for regulating leukocyte attachment to endothelium [41]. On the other hand, it is not only detectable in follicular dendritic cells [42] but in bone marrow fibroblasts as we could demonstrate in this study. Similar to the other members of the selectin family, L-selectin behaves as a lectin, recognizing carbohydrate ligands. In this study determination of fibroblast growth and adhesion of megakaryocytes to fibroblasts was evaluated by simple techniques. However, counting of fibroblasts by two independent observers presented low intraindividual deviation (less than 2%) and, in addition, concerning the semiquantitative analysis of the adhesion assays, in none of the cases there was a difference in classification of megakaryocyte attach-
Fig. 3. Influence of various antibodies against adhesion molecules on megakaryocyte-dependent fibroblast growth. Application of 1000 cells/well of the CD61+-fraction initially containing 50 fibroblasts/well. Fibroblasts are counted at the start of the cultures and after 6 days, at the end of the experiments. The ratio between the fibroblast numbers indicated the fibroblast growth index (FGI =ne/ns). All experiments are set up in quadruplicate. Points represent means ( 9 S.D). Stars indicate a significant increase of fibroblast growth (PB 0.05) in the co-culture compared to the monoculture. Arrowheads indicate the significant impairment in fibroblast growth compared to basal co-cultures (PB 0.05). The number of experiments was 27 (CD62P, CD62L) and 20 (CD41a, CD42b). MAA and LPA (0.25, 2.5 and 25 mg/ml) did not alter the megakaryocyte induced fibroblast proliferation (data not shown).
1020
C. Wickenhauser et al. / Leukemia Research 24 (2000) 1013–1021
ment by the two independent observers. Our results demonstrate a direct involvement of the selectins CD62P and CD62L in megakaryocyte adherence to a fibroblast feeder layer. Additionally, antibodies against both selectins generated an impairment of megakaryocyte-dependent fibroblast growth. Together with the missing reaction of sialic acid recognizing lectins these observations underline the important part of fucose residues in the mechanism of megakaryocyte triggered bone marrow fibroblast proliferation [43]. The actions of CD41a and CD42b, however, were limited to the influence of megakaryocyte dependent fibroblast growth and therefore demonstrate the importance of these glycoproteins in activated megakaryocytes. Taken together, these results emphasize a hierarchical role of several megakaryocyte expressed adhesion receptors in the attachment on fibroblasts and induction of fibroblast proliferation. A cascade of adhesion events starting with CD62P as well as fibroblast expressed CD62L interaction may generate a specific activation of fibroblasts and, in consequence, megakaryocytic glycoproteins CD41a and CD42b are needed for the induction of fibroblast proliferation. In addition, results of recently published studies of our group indicate that also the secretion of PDGF and TGFb as well as the megakaryocyte expressed integrins a5b1 and a3b1 may enhance this effect. Consequently, further studies have to elucidate to which extent the various factors inducing fibroblast growth in vitro are involved in the complex pathomechanisms generating myelofibrosis as a reactive phenomenon in several neoplastic diseases accompanied by megakaryocytic proliferation [1,2,5,6,44 – 47]. Acknowledgements The authors are indebted to the staff of the Clinic of Cardiac Surgery (director Prof. E.R. de Vivie, M.D.) of the University of Cologne for supplying the bone marrow tissue. This work was supported by a grant from the ‘Dr Mildred Scheel Stiftung fu¨r Krebsforschung’ (10-1088TH3). C. Wickenhauser and B. Schmitz provided the concept, design, analysis of data, drafted the paper and gave final approval. S.E. Baldus provided critical revision. F. Hrnze and P. Farahmand provided study materials, statistical expertise and assembled the data. S. Frimpong provided technical support. J. Thiele contributed to the design of the study, gave critical input to the revision and gave final approval. R. Fischer gave critical input and gave final approval to the article. References [1] Burkhardt R, Bartl R, Jaeger K, Frisch B, Kettner G, Mahl G, Sund M. Chronic myeloproliferative disorders (CMPD). Pathol Res Pract 1984;179:131.
[2] McCarthy D. Fibrosis of the bone marrow: content and causes. Br J Haematol 1985;59:1. [3] Dekmezian R, Kantarjian HM, Keating M, Talpaz M, McCredie KB, Freireich EJ. The relevance of reticulin stain-measured fibrosis at diagnosis in chronic myelogeneous leukemia. Cancer 1987;59:1739. [4] Greenberg BR, Woo L, Veomett IC, Payne CM, Ahmann FR. Cytogenetics of bone marrow fibroblastic cells in idiopathic chronic myelofibrosis. Br J Haematol 1987;66:487. [5] Thiele J, Fischer R. Megakaryocytopoiesis in haematological disorders: diagnostic features of bone marrow biopsies. Virchows Arch A Pathol Anat 1991;418:87. [6] Wang JC, Lang HD, Lichter S, Weinstein M, Benn P. Cytogenetic studies of bone marrow fibroblasts cultured from patients with myelofibrosis and myeloid metaplasia. Br J Haematol 1992;80:184. [7] Thiele J, Kvasnicka HM, Titius BR, Parpert U, Nebel R, Zankovich R, Dienemann D, Stein H, Diehl V, Fischer R. Histological features of prognostic significance in CML — an immunohistological and morphometric study (multivariate regression analysis) on trephine biopsies of the bone marrow. Ann Hematol 1993;66:291. [8] Thiele J, Hoeppner B, Zankovich R, Fischer R. Histomorphometry of bone marrow biopsies in primary osteomyelofibrosis/-sclerosis (agnogenic myeloid metaplasia) — correlations between clinical and morphological features. Virchows Arch A Pathol Anat 1989;415:191. [9] Buhr T, Choritz H, Georgii A. The impact of megakaryocyte proliferation for the evolution of myelofibrosis. Histological follow-up study in 186 patients with chronic myeloid leukemia. Virchows Arch A 1992;420:473. [10] Becker RP, De Bruyn PPH. The transmural passage of blood cells into myeloid sinusoids and the entry of platelets into the sinusoidal circulation: a scanning electron microscopic investigation. Am J Anat 1976;145:183. [11] Muto MA. Scanning and transmission electron microscopic study on rat bone marrow sinuses and transmural migration of blood cells. Arch Histol Jpn 1976;39:51. [12] Petrides PE, Dittman KH. How do normal and leukemic white blood cells egress from the bone marrow? Blut 1990;61:3. [13] Schmitz B, Thiele J, Kaufmann R, Witte O, Wickenhauser C, Kuhn-Regnier F, Fischer R. Megakaryocytes and fibroblasts-interactions determined in normal human bone marrow specimens. Leuk Res 1995;19:629. [14] Schmitz B, Thiele J, Witte O, Kaufmann R, Wickenhauser C, Fischer R. Influence of cytokines (IL-1a, IL-3, IL-11, GM-CSF) on megakaryocyte-fibroblast interactions in normal human bone marrow. Eur J Hematol 1995;55:24. [15] Castro-Malaspina H, Rabellino EM, Yen A, Nachman RL, Moore MAS. Human megakaryocyte stimulation of bone marrow fibroblasts. Blood 1981;57:781. [16] Kimura A, Katoh O, Kuramoto A. Effects of platelet-derived growth factor, epidermal growth factor and transforming growth factor-b on the growth of human marrow fibroblasts. Br J Haematol 1988;69:9. [17] Kimura A, Katoh O, Hyodo H, Kuramoto A. Transforming growth factor-b regulates growth as well as collagen and fibronectin synthesis of human marrow fibroblasts. Br J Haematol 1989;72:486. [18] Wickenhauser C, Hillienhof A, Jungheim C, Lorenzen J, Ruskowski H, Hansmann M-L, Thiele J, Fischer R. Detection and quantification of transforming growth factor b and plateletderived growth factor release by normal human megakaryocytes. Leukemia 1995;9:310. [19] Schmitz B, Thiele J, Otto F, Theile-Ochel S, Heedt T, Zensen U, Baldus SE, Wickenhauser C, Fischer R. Interactions between endogeneous lectins and fucosylated oligosaccharides in
C. Wickenhauser et al. / Leukemia Research 24 (2000) 1013–1021
[20]
[21] [22]
[23]
[24] [25]
[26]
[27]
[28]
[29]
[30]
[31] [32] [33]
megakaryocyte-dependent fibroblast growth of the normal bone marrow. Leukemia 1996;10:1604. Picker LJ, Warnock RA, Burns AR, Doerschuk CM, Berg EL, Butcher EC. The neutrophil selectin LECAM-1 represents carbohydrate ligands to the vascular selectins ELAM-1 and GMP-140. Cell 1991;66:921. Springer TA, Lasky LA. Sticky sugars for selectins. Nature 1991;349:196. Clemetson KJ, Clemetson JN. Platelet GPIb-V-IX complex: structure, function, physiology and pathology. Semin Thromb Hemostas 1995;21:130. Block K, Poncz M. Platelet glycoprotein Ib gene expression as a model of megakaryocyte-specific expression. Stem Cells 1995;13:135. Hynes RO. Integrins: Versatility, modulation, and signalling in cell adhesion. Cell 1992;69:11. Schmitz B, Radbruch A, Ku¨mmel T, Wickenhauser C, Korb H, Hansmann M-L, Thiele J, Fischer R. Magnetic activated cell sorting (MACS) — a new immunological method for megakaryocytic cell isolation: Comparison of different separation techniques. Eur J Hematol 1994;52:267. Gatter KC, Cordell JL, Turley H, Heryet A, Kieffer N, Anstee DJ, Mason DY. The immunohistological detection of platelets, megakaryocytes and thrombi in routinely processed specimens. Histopathology 1988;13:257. Schmitz B, Park IA, Kaufmann R, Thiele J, Fischer R. Influence of cytokine stimulation (granulocyte macrophage-colony stimulating factor, interleukin-3, and transforming growth factor-beta1) on adhesion molecule expression in normal human bone marrow fibroblasts. Acta Haematol 1995;94:173. Eaves CJ, Cashman JD, Sutherland HJ, Otsuka T, Humphries RK, Hogge DE, Lansdorp PL, Eaves AC. Molecular analysis of primitive hematopoietic cell proliferation control mechanisms. Ann NY Acad Sci 1991;628:298. Hurley RW, McCarthy JB, Verfaille CM. Direct adhesion to bone marrow stroma via fibronectin receptors inhibits hematopoietic progenitor proliferation. J Clin Invest 1995;96:511. Verfaillie CM, Catanzaro P. Direct contact with stroma inhibits proliferation of human long-term culture initiating cells. Leukemia 1996;10:498. Kieffer N, Phillips DR. Platelet membrane glycoproteins: functions in cellular interactions. Annu Rev Cell Biol 1990;6:329. Plow EF, Ginsberg MH. Cellular adhesion: GPIIb-IIIa as a prototype adhesion receptor. Prog Hemost Thromb 1989;9:117. Peterson JA, Visentin GP, Newman PJ, Aster RH. A recombinant soluble form of the integrin aIIbb3 (gpIIb-IIIa) assumes an active, ligand-binding conformation and is recognized by gpIIbIIIa specific monoclonal, allo-, auto-, and drug-dependent platelet antibodies. Blood 1998;92:2053.
.
1021
[34] Schick PK, Wojenski CM, He X, Walker J, Marcinkiewicz C, Niewiarowski S. Integrins involved in the adhesion of megakaryocytes to fibronectin and fibrinogen. Blood 1998;92:2650. [35] Andrews RK, Fox JEB. Interaction of purified actin-binding protein with the platelet membrane glycoprotein Ib-IX complex. J Biol Chem 1991;266:7144. [36] De Marco L, Mazzucato M, Masotti A, Ruggeri ZM. Localization and characterization of an a-thrombin-binding site on platelet glycoprotein Iba. J Biol Chem 1994;269:6478. [37] Berman C, Yeo EL, Wemncel-Drake JD, Furie BC, Ginsberg MH, Furie B. A platelet alpha granule membrane protein that is associated with the plasma membrane after activation. J Clin Invest 1986;78:130. [38] Larsen GR, Sako D, Ahern TJ, Shaffer M, Erban J, Sajer SA, Gibson RM, Wagner DD, Furie BC, Furie B. P-selectin and E-selectin. Distinct but overlapping leukocyte ligand specifities. J Biol Chem 1992;267:104. [39] Polley MJ, Phillips ML, Wayner E, Nudelman E, Singhal AK, Hakomori S, Paulson JC. CD62 and endothelial cell-leukocyte adhesion molecule 1 (ELAM-1) recognize the same carbohydrate ligand, sialyl-Lewis X. Proc Natl Acad Sci USA 1991;88:6224. [40] Larsen E, Celi A, Gilbert GE, Furie BC, Erban JK, Bonfanti R, Wagner DD, Furie B. PADGEM protein: A receptor that mediates the interaction of activated platelets with neutrophils and monocytes. Cell 1989;59:305. [41] Springer TA. Traffic signals for lymphocyte recirculation and leukocyte emigration: The multistep paradigm. Cell 1994;76:301. [42] Munro JM, Briscoe DM, Tedder TF. Differential regulation of leukocyte L-selectin (CD62L) expression in normal lymphoid and inflamed extralymphoid tissues. J Clin Pathol 1996;49:721. [43] Schmitz B, Thiele J, Otto F, Farahmand P, Henze F, Frimpong S, Wickenhauser C, Fischer R. Evidence for integrin receptor involvement in megakaryocyte-fibroblast interaction — a possible pathomechanism for the evolution of myelofibrosis. J Cell Physiol 1998;176:445. [44] Martyre´ MC. Platelet PDGF and TGF: levels in myeloproliferative disorders. Leukemia Lymphoma 1991;6:1. [45] Martyre´ MC, Romquin N, Bousse-Kerdiles MC, Chewillard S, Benyahia B, Dupriez B, Demory JL, Bauters F. Transforming growth factor-b and megakaryocytes in the pathogenesis of idiopathic myelofibrosis. Br J Haematol 1994;88:9. [46] Martyre´ MC. TGF-b and megakaryocytes in the pathogenesis of myelofibrosis in myeloproliferative disorders. Leukemia Lymphoma 1995;20:39. [47] Thiele J, Kvasnicka HM, Fischer R, Diehl V. Clinicopathological impact of the interaction between megakaryocytes and myeloid stroma in chronic myeloproliferative disorders: a concise update. Leukemia Lymphoma 1997;24:463.